diff --git a/.mailmap b/.mailmap index 0374777cc662..5378f04b2566 100644 --- a/.mailmap +++ b/.mailmap @@ -73,6 +73,8 @@ Andrey Ryabinin Andrzej Hajda André Almeida Andy Adamson +Andy Chiu +Andy Chiu Andy Shevchenko Andy Shevchenko Anilkumar Kolli @@ -197,18 +199,23 @@ Elliot Berman Enric Balletbo i Serra Enric Balletbo i Serra Erik Kaneda -Eugen Hristev +Eugen Hristev +Eugen Hristev Evgeniy Polyakov Ezequiel Garcia Faith Ekstrand Faith Ekstrand Faith Ekstrand +Fangrui Song Felipe W Damasio Felix Kuhling Felix Moeller Fenglin Wu Filipe Lautert Finn Thain +Fiona Behrens +Fiona Behrens +Fiona Behrens Franck Bui-Huu Frank Rowand Frank Rowand @@ -276,7 +283,7 @@ Jan Glauber Jan Kuliga Jarkko Sakkinen Jarkko Sakkinen -Jarkko Sakkinen +Jarkko Sakkinen Jason Gunthorpe Jason Gunthorpe Jason Gunthorpe @@ -300,6 +307,11 @@ Jens Axboe Jens Axboe Jens Osterkamp Jernej Skrabec +Jesper Dangaard Brouer +Jesper Dangaard Brouer +Jesper Dangaard Brouer +Jesper Dangaard Brouer +Jesper Dangaard Brouer Jessica Zhang Jilai Wang Jiri Kosina diff --git a/CREDITS b/CREDITS index d439f5a1bc00..63f53feefa0a 100644 --- a/CREDITS +++ b/CREDITS @@ -1358,10 +1358,6 @@ D: Major kbuild rework during the 2.5 cycle D: ISDN Maintainer S: USA -N: Gerrit Renker -E: gerrit@erg.abdn.ac.uk -D: DCCP protocol support. - N: Philip Gladstone E: philip@gladstonefamily.net D: Kernel / timekeeping stuff @@ -1677,11 +1673,6 @@ W: http://www.carumba.com/ D: bug toaster (A1 sauce makes all the difference) D: Random linux hacker -N: James Hogan -E: jhogan@kernel.org -D: Metag architecture maintainer -D: TZ1090 SoC maintainer - N: Tim Hockin E: thockin@hockin.org W: http://www.hockin.org/~thockin @@ -1697,6 +1688,11 @@ D: hwmon subsystem maintainer D: i2c-sis96x and i2c-stub SMBus drivers S: USA +N: James Hogan +E: jhogan@kernel.org +D: Metag architecture maintainer +D: TZ1090 SoC maintainer + N: Dirk Hohndel E: hohndel@suse.de D: The XFree86[tm] Project @@ -1872,6 +1868,10 @@ S: K osmidomkum 723 S: 160 00 Praha 6 S: Czech Republic +N: Seth Jennings +E: sjenning@redhat.com +D: Creation and maintenance of zswap + N: Jeremy Kerr D: Maintainer of SPU File System @@ -2188,19 +2188,6 @@ N: Mike Kravetz E: mike.kravetz@oracle.com D: Maintenance and development of the hugetlb subsystem -N: Seth Jennings -E: sjenning@redhat.com -D: Creation and maintenance of zswap - -N: Dan Streetman -E: ddstreet@ieee.org -D: Maintenance and development of zswap -D: Creation and maintenance of the zpool API - -N: Vitaly Wool -E: vitaly.wool@konsulko.com -D: Maintenance and development of zswap - N: Andreas S. Krebs E: akrebs@altavista.net D: CYPRESS CY82C693 chipset IDE, Digital's PC-Alpha 164SX boards @@ -3191,6 +3178,11 @@ N: Ken Pizzini E: ken@halcyon.com D: CDROM driver "sonycd535" (Sony CDU-535/531) +N: Mathieu Poirier +E: mathieu.poirier@linaro.org +D: CoreSight kernel subsystem, Maintainer 2014-2022 +D: Perf tool support for CoreSight + N: Stelian Pop E: stelian@popies.net P: 1024D/EDBB6147 7B36 0E07 04BC 11DC A7A0 D3F7 7185 9E7A EDBB 6147 @@ -3300,6 +3292,10 @@ S: Schlossbergring 9 S: 79098 Freiburg S: Germany +N: Gerrit Renker +E: gerrit@erg.abdn.ac.uk +D: DCCP protocol support. + N: Thomas Renninger E: trenn@suse.de D: cpupowerutils @@ -3576,11 +3572,6 @@ D: several improvements to system programs S: Oldenburg S: Germany -N: Mathieu Poirier -E: mathieu.poirier@linaro.org -D: CoreSight kernel subsystem, Maintainer 2014-2022 -D: Perf tool support for CoreSight - N: Robert Schwebel E: robert@schwebel.de W: https://www.schwebel.de @@ -3771,6 +3762,11 @@ S: Chr. Winthersvej 1 B, st.th. S: DK-1860 Frederiksberg C S: Denmark +N: Dan Streetman +E: ddstreet@ieee.org +D: Maintenance and development of zswap +D: Creation and maintenance of the zpool API + N: Drew Sullivan E: drew@ss.org W: http://www.ss.org/ @@ -4286,6 +4282,10 @@ S: Pipers Way S: Swindon. SN3 1RJ S: England +N: Vitaly Wool +E: vitaly.wool@konsulko.com +D: Maintenance and development of zswap + N: Chris Wright E: chrisw@sous-sol.org D: hacking on LSM framework and security modules. diff --git a/Documentation/ABI/testing/sysfs-driver-intel-i915-hwmon b/Documentation/ABI/testing/sysfs-driver-intel-i915-hwmon index be4141a7522f..a885e5316d02 100644 --- a/Documentation/ABI/testing/sysfs-driver-intel-i915-hwmon +++ b/Documentation/ABI/testing/sysfs-driver-intel-i915-hwmon @@ -83,3 +83,11 @@ Contact: intel-gfx@lists.freedesktop.org Description: RO. Fan speed of device in RPM. Only supported for particular Intel i915 graphics platforms. + +What: /sys/bus/pci/drivers/i915/.../hwmon/hwmon/temp1_input +Date: November 2024 +KernelVersion: 6.12 +Contact: intel-gfx@lists.freedesktop.org +Description: RO. GPU package temperature in millidegree Celsius. + + Only supported for particular Intel i915 graphics platforms. diff --git a/Documentation/ABI/testing/sysfs-driver-panthor-profiling b/Documentation/ABI/testing/sysfs-driver-panthor-profiling new file mode 100644 index 000000000000..af05fccedc15 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-driver-panthor-profiling @@ -0,0 +1,10 @@ +What: /sys/bus/platform/drivers/panthor/.../profiling +Date: September 2024 +KernelVersion: 6.11.0 +Contact: Adrian Larumbe +Description: + Bitmask to enable drm fdinfo's job profiling measurements. + Valid values are: + 0: Don't enable fdinfo job profiling sources. + 1: Enable GPU cycle measurements for running jobs. + 2: Enable GPU timestamp sampling for running jobs. diff --git a/Documentation/accel/qaic/aic080.rst b/Documentation/accel/qaic/aic080.rst new file mode 100644 index 000000000000..d563771ea6ce --- /dev/null +++ b/Documentation/accel/qaic/aic080.rst @@ -0,0 +1,14 @@ +.. SPDX-License-Identifier: GPL-2.0-only + +=============================== + Qualcomm Cloud AI 80 (AIC080) +=============================== + +Overview +======== + +The Qualcomm Cloud AI 80/AIC080 family of products are a derivative of AIC100. +The number of NSPs and clock rates are reduced to fit within resource +constrained solutions. The PCIe Product ID is 0xa080. + +As a derivative product, all AIC100 documentation applies. diff --git a/Documentation/accel/qaic/aic100.rst b/Documentation/accel/qaic/aic100.rst index 590dae77ea12..273da6192fb3 100644 --- a/Documentation/accel/qaic/aic100.rst +++ b/Documentation/accel/qaic/aic100.rst @@ -229,6 +229,8 @@ of the defined channels, and their uses. | _PERIODIC | | | timestamps in the device side logs with| | | | | the host time source. | +----------------+---------+----------+----------------------------------------+ +| IPCR | 24 & 25 | AMSS | AF_QIPCRTR clients and servers. | ++----------------+---------+----------+----------------------------------------+ DMA Bridge ========== diff --git a/Documentation/accel/qaic/index.rst b/Documentation/accel/qaic/index.rst index ad19b88d1a66..967b9dd8bace 100644 --- a/Documentation/accel/qaic/index.rst +++ b/Documentation/accel/qaic/index.rst @@ -10,4 +10,5 @@ accelerator cards. .. toctree:: qaic + aic080 aic100 diff --git a/Documentation/admin-guide/LSM/ipe.rst b/Documentation/admin-guide/LSM/ipe.rst index f38e641df0e9..f93a467db628 100644 --- a/Documentation/admin-guide/LSM/ipe.rst +++ b/Documentation/admin-guide/LSM/ipe.rst @@ -223,7 +223,10 @@ are signed through the PKCS#7 message format to enforce some level of authorization of the policies (prohibiting an attacker from gaining unconstrained root, and deploying an "allow all" policy). These policies must be signed by a certificate that chains to the -``SYSTEM_TRUSTED_KEYRING``. With openssl, the policy can be signed by:: +``SYSTEM_TRUSTED_KEYRING``, or to the secondary and/or platform keyrings if +``CONFIG_IPE_POLICY_SIG_SECONDARY_KEYRING`` and/or +``CONFIG_IPE_POLICY_SIG_PLATFORM_KEYRING`` are enabled, respectively. +With openssl, the policy can be signed by:: openssl smime -sign \ -in "$MY_POLICY" \ @@ -266,7 +269,7 @@ in the kernel. This file is write-only and accepts a PKCS#7 signed policy. Two checks will always be performed on this policy: First, the ``policy_names`` must match with the updated version and the existing version. Second the updated policy must have a policy version greater than -or equal to the currently-running version. This is to prevent rollback attacks. +the currently-running version. This is to prevent rollback attacks. The ``delete`` file is used to remove a policy that is no longer needed. This file is write-only and accepts a value of ``1`` to delete the policy. diff --git a/Documentation/admin-guide/pm/cpufreq.rst b/Documentation/admin-guide/pm/cpufreq.rst index fe1be4ad88cb..a21369eba034 100644 --- a/Documentation/admin-guide/pm/cpufreq.rst +++ b/Documentation/admin-guide/pm/cpufreq.rst @@ -425,8 +425,8 @@ This governor exposes only one tunable: ``rate_limit_us`` Minimum time (in microseconds) that has to pass between two consecutive - runs of governor computations (default: 1000 times the scaling driver's - transition latency). + runs of governor computations (default: 1.5 times the scaling driver's + transition latency or the maximum 2ms). The purpose of this tunable is to reduce the scheduler context overhead of the governor which might be excessive without it. @@ -474,17 +474,17 @@ This governor exposes the following tunables: This is how often the governor's worker routine should run, in microseconds. - Typically, it is set to values of the order of 10000 (10 ms). Its - default value is equal to the value of ``cpuinfo_transition_latency`` - for each policy this governor is attached to (but since the unit here - is greater by 1000, this means that the time represented by - ``sampling_rate`` is 1000 times greater than the transition latency by - default). + Typically, it is set to values of the order of 2000 (2 ms). Its + default value is to add a 50% breathing room + to ``cpuinfo_transition_latency`` on each policy this governor is + attached to. The minimum is typically the length of two scheduler + ticks. If this tunable is per-policy, the following shell command sets the time - represented by it to be 750 times as high as the transition latency:: + represented by it to be 1.5 times as high as the transition latency + (the default):: - # echo `$(($(cat cpuinfo_transition_latency) * 750 / 1000)) > ondemand/sampling_rate + # echo `$(($(cat cpuinfo_transition_latency) * 3 / 2)) > ondemand/sampling_rate ``up_threshold`` If the estimated CPU load is above this value (in percent), the governor diff --git a/Documentation/arch/arm/mem_alignment.rst b/Documentation/arch/arm/mem_alignment.rst index aa22893b62bc..64bd77959300 100644 --- a/Documentation/arch/arm/mem_alignment.rst +++ b/Documentation/arch/arm/mem_alignment.rst @@ -12,7 +12,7 @@ ones. Of course this is a bad idea to rely on the alignment trap to perform unaligned memory access in general. If those access are predictable, you -are better to use the macros provided by include/asm/unaligned.h. The +are better to use the macros provided by include/linux/unaligned.h. The alignment trap can fixup misaligned access for the exception cases, but at a high performance cost. It better be rare. diff --git a/Documentation/arch/arm64/silicon-errata.rst b/Documentation/arch/arm64/silicon-errata.rst index 9eb5e70b4888..65bfab1b1861 100644 --- a/Documentation/arch/arm64/silicon-errata.rst +++ b/Documentation/arch/arm64/silicon-errata.rst @@ -146,6 +146,8 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-A715 | #2645198 | ARM64_ERRATUM_2645198 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Cortex-A715 | #3456084 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-A720 | #3456091 | ARM64_ERRATUM_3194386 | +----------------+-----------------+-----------------+-----------------------------+ | ARM | Cortex-A725 | #3456106 | ARM64_ERRATUM_3194386 | @@ -186,6 +188,8 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-N2 | #3324339 | ARM64_ERRATUM_3194386 | +----------------+-----------------+-----------------+-----------------------------+ +| ARM | Neoverse-N3 | #3456111 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-V1 | #1619801 | N/A | +----------------+-----------------+-----------------+-----------------------------+ | ARM | Neoverse-V1 | #3324341 | ARM64_ERRATUM_3194386 | @@ -289,3 +293,5 @@ stable kernels. +----------------+-----------------+-----------------+-----------------------------+ | Microsoft | Azure Cobalt 100| #2253138 | ARM64_ERRATUM_2253138 | +----------------+-----------------+-----------------+-----------------------------+ +| Microsoft | Azure Cobalt 100| #3324339 | ARM64_ERRATUM_3194386 | ++----------------+-----------------+-----------------+-----------------------------+ diff --git a/Documentation/core-api/folio_queue.rst b/Documentation/core-api/folio_queue.rst new file mode 100644 index 000000000000..1fe7a9bc4b8d --- /dev/null +++ b/Documentation/core-api/folio_queue.rst @@ -0,0 +1,212 @@ +.. SPDX-License-Identifier: GPL-2.0+ + +=========== +Folio Queue +=========== + +:Author: David Howells + +.. Contents: + + * Overview + * Initialisation + * Adding and removing folios + * Querying information about a folio + * Querying information about a folio_queue + * Folio queue iteration + * Folio marks + * Lockless simultaneous production/consumption issues + + +Overview +======== + +The folio_queue struct forms a single segment in a segmented list of folios +that can be used to form an I/O buffer. As such, the list can be iterated over +using the ITER_FOLIOQ iov_iter type. + +The publicly accessible members of the structure are:: + + struct folio_queue { + struct folio_queue *next; + struct folio_queue *prev; + ... + }; + +A pair of pointers are provided, ``next`` and ``prev``, that point to the +segments on either side of the segment being accessed. Whilst this is a +doubly-linked list, it is intentionally not a circular list; the outward +sibling pointers in terminal segments should be NULL. + +Each segment in the list also stores: + + * an ordered sequence of folio pointers, + * the size of each folio and + * three 1-bit marks per folio, + +but hese should not be accessed directly as the underlying data structure may +change, but rather the access functions outlined below should be used. + +The facility can be made accessible by:: + + #include + +and to use the iterator:: + + #include + + +Initialisation +============== + +A segment should be initialised by calling:: + + void folioq_init(struct folio_queue *folioq); + +with a pointer to the segment to be initialised. Note that this will not +necessarily initialise all the folio pointers, so care must be taken to check +the number of folios added. + + +Adding and removing folios +========================== + +Folios can be set in the next unused slot in a segment struct by calling one +of:: + + unsigned int folioq_append(struct folio_queue *folioq, + struct folio *folio); + + unsigned int folioq_append_mark(struct folio_queue *folioq, + struct folio *folio); + +Both functions update the stored folio count, store the folio and note its +size. The second function also sets the first mark for the folio added. Both +functions return the number of the slot used. [!] Note that no attempt is made +to check that the capacity wasn't overrun and the list will not be extended +automatically. + +A folio can be excised by calling:: + + void folioq_clear(struct folio_queue *folioq, unsigned int slot); + +This clears the slot in the array and also clears all the marks for that folio, +but doesn't change the folio count - so future accesses of that slot must check +if the slot is occupied. + + +Querying information about a folio +================================== + +Information about the folio in a particular slot may be queried by the +following function:: + + struct folio *folioq_folio(const struct folio_queue *folioq, + unsigned int slot); + +If a folio has not yet been set in that slot, this may yield an undefined +pointer. The size of the folio in a slot may be queried with either of:: + + unsigned int folioq_folio_order(const struct folio_queue *folioq, + unsigned int slot); + + size_t folioq_folio_size(const struct folio_queue *folioq, + unsigned int slot); + +The first function returns the size as an order and the second as a number of +bytes. + + +Querying information about a folio_queue +======================================== + +Information may be retrieved about a particular segment with the following +functions:: + + unsigned int folioq_nr_slots(const struct folio_queue *folioq); + + unsigned int folioq_count(struct folio_queue *folioq); + + bool folioq_full(struct folio_queue *folioq); + +The first function returns the maximum capacity of a segment. It must not be +assumed that this won't vary between segments. The second returns the number +of folios added to a segments and the third is a shorthand to indicate if the +segment has been filled to capacity. + +Not that the count and fullness are not affected by clearing folios from the +segment. These are more about indicating how many slots in the array have been +initialised, and it assumed that slots won't get reused, but rather the segment +will get discarded as the queue is consumed. + + +Folio marks +=========== + +Folios within a queue can also have marks assigned to them. These marks can be +used to note information such as if a folio needs folio_put() calling upon it. +There are three marks available to be set for each folio. + +The marks can be set by:: + + void folioq_mark(struct folio_queue *folioq, unsigned int slot); + void folioq_mark2(struct folio_queue *folioq, unsigned int slot); + void folioq_mark3(struct folio_queue *folioq, unsigned int slot); + +Cleared by:: + + void folioq_unmark(struct folio_queue *folioq, unsigned int slot); + void folioq_unmark2(struct folio_queue *folioq, unsigned int slot); + void folioq_unmark3(struct folio_queue *folioq, unsigned int slot); + +And the marks can be queried by:: + + bool folioq_is_marked(const struct folio_queue *folioq, unsigned int slot); + bool folioq_is_marked2(const struct folio_queue *folioq, unsigned int slot); + bool folioq_is_marked3(const struct folio_queue *folioq, unsigned int slot); + +The marks can be used for any purpose and are not interpreted by this API. + + +Folio queue iteration +===================== + +A list of segments may be iterated over using the I/O iterator facility using +an ``iov_iter`` iterator of ``ITER_FOLIOQ`` type. The iterator may be +initialised with:: + + void iov_iter_folio_queue(struct iov_iter *i, unsigned int direction, + const struct folio_queue *folioq, + unsigned int first_slot, unsigned int offset, + size_t count); + +This may be told to start at a particular segment, slot and offset within a +queue. The iov iterator functions will follow the next pointers when advancing +and prev pointers when reverting when needed. + + +Lockless simultaneous production/consumption issues +=================================================== + +If properly managed, the list can be extended by the producer at the head end +and shortened by the consumer at the tail end simultaneously without the need +to take locks. The ITER_FOLIOQ iterator inserts appropriate barriers to aid +with this. + +Care must be taken when simultaneously producing and consuming a list. If the +last segment is reached and the folios it refers to are entirely consumed by +the IOV iterators, an iov_iter struct will be left pointing to the last segment +with a slot number equal to the capacity of that segment. The iterator will +try to continue on from this if there's another segment available when it is +used again, but care must be taken lest the segment got removed and freed by +the consumer before the iterator was advanced. + +It is recommended that the queue always contain at least one segment, even if +that segment has never been filled or is entirely spent. This prevents the +head and tail pointers from collapsing. + + +API Function Reference +====================== + +.. kernel-doc:: include/linux/folio_queue.h diff --git a/Documentation/core-api/index.rst b/Documentation/core-api/index.rst index a331d2c814f5..6a875743dd4b 100644 --- a/Documentation/core-api/index.rst +++ b/Documentation/core-api/index.rst @@ -37,6 +37,7 @@ Library functionality that is used throughout the kernel. kref cleanup assoc_array + folio_queue xarray maple_tree idr diff --git a/Documentation/core-api/protection-keys.rst b/Documentation/core-api/protection-keys.rst index bf28ac0401f3..7eb7c6023e09 100644 --- a/Documentation/core-api/protection-keys.rst +++ b/Documentation/core-api/protection-keys.rst @@ -12,7 +12,10 @@ Pkeys Userspace (PKU) is a feature which can be found on: * Intel server CPUs, Skylake and later * Intel client CPUs, Tiger Lake (11th Gen Core) and later * Future AMD CPUs + * arm64 CPUs implementing the Permission Overlay Extension (FEAT_S1POE) +x86_64 +====== Pkeys work by dedicating 4 previously Reserved bits in each page table entry to a "protection key", giving 16 possible keys. @@ -28,6 +31,22 @@ register. The feature is only available in 64-bit mode, even though there is theoretically space in the PAE PTEs. These permissions are enforced on data access only and have no effect on instruction fetches. +arm64 +===== + +Pkeys use 3 bits in each page table entry, to encode a "protection key index", +giving 8 possible keys. + +Protections for each key are defined with a per-CPU user-writable system +register (POR_EL0). This is a 64-bit register encoding read, write and execute +overlay permissions for each protection key index. + +Being a CPU register, POR_EL0 is inherently thread-local, potentially giving +each thread a different set of protections from every other thread. + +Unlike x86_64, the protection key permissions also apply to instruction +fetches. + Syscalls ======== @@ -38,11 +57,10 @@ There are 3 system calls which directly interact with pkeys:: int pkey_mprotect(unsigned long start, size_t len, unsigned long prot, int pkey); -Before a pkey can be used, it must first be allocated with -pkey_alloc(). An application calls the WRPKRU instruction -directly in order to change access permissions to memory covered -with a key. In this example WRPKRU is wrapped by a C function -called pkey_set(). +Before a pkey can be used, it must first be allocated with pkey_alloc(). An +application writes to the architecture specific CPU register directly in order +to change access permissions to memory covered with a key. In this example +this is wrapped by a C function called pkey_set(). :: int real_prot = PROT_READ|PROT_WRITE; @@ -64,9 +82,9 @@ is no longer in use:: munmap(ptr, PAGE_SIZE); pkey_free(pkey); -.. note:: pkey_set() is a wrapper for the RDPKRU and WRPKRU instructions. - An example implementation can be found in - tools/testing/selftests/x86/protection_keys.c. +.. note:: pkey_set() is a wrapper around writing to the CPU register. + Example implementations can be found in + tools/testing/selftests/mm/pkey-{arm64,powerpc,x86}.h Behavior ======== @@ -96,3 +114,7 @@ with a read():: The kernel will send a SIGSEGV in both cases, but si_code will be set to SEGV_PKERR when violating protection keys versus SEGV_ACCERR when the plain mprotect() permissions are violated. + +Note that kernel accesses from a kthread (such as io_uring) will use a default +value for the protection key register and so will not be consistent with +userspace's value of the register or mprotect(). diff --git a/Documentation/core-api/unaligned-memory-access.rst b/Documentation/core-api/unaligned-memory-access.rst index 1ee82419d8aa..5ceeb80eb539 100644 --- a/Documentation/core-api/unaligned-memory-access.rst +++ b/Documentation/core-api/unaligned-memory-access.rst @@ -203,7 +203,7 @@ Avoiding unaligned accesses =========================== The easiest way to avoid unaligned access is to use the get_unaligned() and -put_unaligned() macros provided by the header file. +put_unaligned() macros provided by the header file. Going back to an earlier example of code that potentially causes unaligned access:: diff --git a/Documentation/devicetree/bindings/display/bridge/sil,sii9022.yaml b/Documentation/devicetree/bindings/display/bridge/sil,sii9022.yaml index 5a69547ad3d7..1509c4535e53 100644 --- a/Documentation/devicetree/bindings/display/bridge/sil,sii9022.yaml +++ b/Documentation/devicetree/bindings/display/bridge/sil,sii9022.yaml @@ -81,9 +81,22 @@ properties: properties: port@0: - $ref: /schemas/graph.yaml#/properties/port + unevaluatedProperties: false + $ref: /schemas/graph.yaml#/$defs/port-base description: Parallel RGB input port + properties: + endpoint: + $ref: /schemas/graph.yaml#/$defs/endpoint-base + unevaluatedProperties: false + + properties: + bus-width: + description: + Endpoint bus width. + enum: [ 16, 18, 24 ] + default: 24 + port@1: $ref: /schemas/graph.yaml#/properties/port description: HDMI output port diff --git a/Documentation/devicetree/bindings/display/bridge/ti,tdp158.yaml b/Documentation/devicetree/bindings/display/bridge/ti,tdp158.yaml new file mode 100644 index 000000000000..1c522f72c4ba --- /dev/null +++ b/Documentation/devicetree/bindings/display/bridge/ti,tdp158.yaml @@ -0,0 +1,57 @@ +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/bridge/ti,tdp158.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: TI TDP158 HDMI to TMDS Redriver + +maintainers: + - Arnaud Vrac + - Pierre-Hugues Husson + +properties: + compatible: + const: ti,tdp158 + +# The reg property is required if and only if the device is connected +# to an I2C bus. In pin strap mode, reg must not be specified. + reg: + description: I2C address of the device + +# Pin 36 = Operation Enable / Reset Pin +# OE = L: Power Down Mode +# OE = H: Normal Operation +# Internal weak pullup - device resets on H to L transitions + enable-gpios: + description: GPIO controlling bridge enable + + vcc-supply: + description: Power supply 3.3V + + vdd-supply: + description: Power supply 1.1V + + ports: + $ref: /schemas/graph.yaml#/properties/ports + + properties: + port@0: + $ref: /schemas/graph.yaml#/properties/port + description: Bridge input + + port@1: + $ref: /schemas/graph.yaml#/properties/port + description: Bridge output + + required: + - port@0 + - port@1 + +required: + - compatible + - vcc-supply + - vdd-supply + - ports + +additionalProperties: false diff --git a/Documentation/devicetree/bindings/display/bridge/toshiba,tc358768.yaml b/Documentation/devicetree/bindings/display/bridge/toshiba,tc358768.yaml index 779d8c57f854..bb5d3b543800 100644 --- a/Documentation/devicetree/bindings/display/bridge/toshiba,tc358768.yaml +++ b/Documentation/devicetree/bindings/display/bridge/toshiba,tc358768.yaml @@ -60,6 +60,10 @@ properties: data-lines: $ref: /schemas/types.yaml#/definitions/uint32 enum: [ 16, 18, 24 ] + deprecated: true + + bus-width: + enum: [ 16, 18, 24 ] port@1: $ref: /schemas/graph.yaml#/properties/port diff --git a/Documentation/devicetree/bindings/display/elgin,jg10309-01.yaml b/Documentation/devicetree/bindings/display/elgin,jg10309-01.yaml new file mode 100644 index 000000000000..faca0cb3f154 --- /dev/null +++ b/Documentation/devicetree/bindings/display/elgin,jg10309-01.yaml @@ -0,0 +1,54 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/elgin,jg10309-01.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Elgin JG10309-01 SPI-controlled display + +maintainers: + - Fabio Estevam + +description: | + The Elgin JG10309-01 SPI-controlled display is used on the RV1108-Elgin-r1 + board and is a custom display. + +allOf: + - $ref: /schemas/spi/spi-peripheral-props.yaml# + +properties: + compatible: + const: elgin,jg10309-01 + + reg: + maxItems: 1 + + spi-max-frequency: + maximum: 24000000 + + spi-cpha: true + + spi-cpol: true + +required: + - compatible + - reg + - spi-cpha + - spi-cpol + +additionalProperties: false + +examples: + - | + spi { + #address-cells = <1>; + #size-cells = <0>; + + display@0 { + compatible = "elgin,jg10309-01"; + reg = <0>; + spi-max-frequency = <24000000>; + spi-cpha; + spi-cpol; + }; + }; diff --git a/Documentation/devicetree/bindings/display/imx/fsl-imx-drm.txt b/Documentation/devicetree/bindings/display/imx/fsl-imx-drm.txt index 3c35338a2867..269b1ae2fca9 100644 --- a/Documentation/devicetree/bindings/display/imx/fsl-imx-drm.txt +++ b/Documentation/devicetree/bindings/display/imx/fsl-imx-drm.txt @@ -119,7 +119,6 @@ Optional properties: - interface-pix-fmt: How this display is connected to the display interface. Currently supported types: "rgb24", "rgb565", "bgr666" and "lvds666". -- edid: verbatim EDID data block describing attached display. - ddc: phandle describing the i2c bus handling the display data channel - port@[0-1]: Port nodes with endpoint definitions as defined in @@ -131,7 +130,6 @@ example: disp0 { compatible = "fsl,imx-parallel-display"; - edid = [edid-data]; interface-pix-fmt = "rgb24"; port@0 { diff --git a/Documentation/devicetree/bindings/display/imx/ldb.txt b/Documentation/devicetree/bindings/display/imx/ldb.txt index 8e6e7d797943..03653a291b54 100644 --- a/Documentation/devicetree/bindings/display/imx/ldb.txt +++ b/Documentation/devicetree/bindings/display/imx/ldb.txt @@ -62,7 +62,6 @@ Required properties: display-timings are used instead. Optional properties (required if display-timings are used): - - ddc-i2c-bus: phandle of an I2C controller used for DDC EDID probing - display-timings : A node that describes the display timings as defined in Documentation/devicetree/bindings/display/panel/display-timing.txt. - fsl,data-mapping : should be "spwg" or "jeida" diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml index 3a82aec9021c..497c0eb4ed0b 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,dpi.yaml @@ -63,6 +63,16 @@ properties: - const: sleep power-domains: + description: | + The MediaTek DPI module is typically associated with one of the + following multimedia power domains: + POWER_DOMAIN_DISPLAY + POWER_DOMAIN_VDOSYS + POWER_DOMAIN_MM + The specific power domain used varies depending on the SoC design. + + It is recommended to explicitly add the appropriate power domain + property to the DPI node in the device tree. maxItems: 1 port: @@ -79,20 +89,6 @@ required: - clock-names - port -allOf: - - if: - not: - properties: - compatible: - contains: - enum: - - mediatek,mt6795-dpi - - mediatek,mt8173-dpi - - mediatek,mt8186-dpi - then: - properties: - power-domains: false - additionalProperties: false examples: diff --git a/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml b/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml index e4affc854f3d..4b6ff546757e 100644 --- a/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml +++ b/Documentation/devicetree/bindings/display/mediatek/mediatek,split.yaml @@ -38,6 +38,7 @@ properties: description: A phandle and PM domain specifier as defined by bindings of the power controller specified by phandle. See Documentation/devicetree/bindings/power/power-domain.yaml for details. + maxItems: 1 mediatek,gce-client-reg: description: @@ -57,6 +58,9 @@ properties: clocks: items: - description: SPLIT Clock + - description: Used for interfacing with the HDMI RX signal source. + - description: Paired with receiving HDMI RX metadata. + minItems: 1 required: - compatible @@ -72,9 +76,24 @@ allOf: const: mediatek,mt8195-mdp3-split then: + properties: + clocks: + minItems: 3 + required: - mediatek,gce-client-reg + - if: + properties: + compatible: + contains: + const: mediatek,mt8173-disp-split + + then: + properties: + clocks: + maxItems: 1 + additionalProperties: false examples: diff --git a/Documentation/devicetree/bindings/display/panel/panel-common.yaml b/Documentation/devicetree/bindings/display/panel/panel-common.yaml index 0a57a31f4f3d..087415753d60 100644 --- a/Documentation/devicetree/bindings/display/panel/panel-common.yaml +++ b/Documentation/devicetree/bindings/display/panel/panel-common.yaml @@ -51,6 +51,14 @@ properties: $ref: /schemas/types.yaml#/definitions/uint32 enum: [0, 90, 180, 270] + flip-horizontal: + description: boolean to flip image horizontally + type: boolean + + flip-vertical: + description: boolean to flip image vertically + type: boolean + # Display Timings panel-timing: description: diff --git a/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml b/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml index 155d8ffa8f6e..5af2d6930075 100644 --- a/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml +++ b/Documentation/devicetree/bindings/display/panel/panel-lvds.yaml @@ -50,6 +50,8 @@ properties: - hannstar,hsd101pww2 # Hydis Technologies 7" WXGA (800x1280) TFT LCD LVDS panel - hydis,hv070wx2-1e0 + # Jenson Display BL-JT60050-01A 7" WSVGA (1024x600) color TFT LCD LVDS panel + - jenson,bl-jt60050-01a - tbs,a711-panel - const: panel-lvds diff --git a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml index b89e39790579..18b63f356bb4 100644 --- a/Documentation/devicetree/bindings/display/panel/panel-simple.yaml +++ b/Documentation/devicetree/bindings/display/panel/panel-simple.yaml @@ -200,6 +200,8 @@ properties: - logictechno,lttd800480070-l2rt # Logic Technologies LTTD800480070-L6WH-RT 7” 800x480 TFT Resistive Touch Module - logictechno,lttd800480070-l6wh-rt + # Microchip AC69T88A 5" 800X480 LVDS interface TFT LCD Panel + - microchip,ac69t88a # Mitsubishi "AA070MC01 7.0" WVGA TFT LCD panel - mitsubishi,aa070mc01-ca1 # Mitsubishi AA084XE01 8.4" XGA TFT LCD panel diff --git a/Documentation/devicetree/bindings/display/panel/samsung,ams581vf01.yaml b/Documentation/devicetree/bindings/display/panel/samsung,ams581vf01.yaml new file mode 100644 index 000000000000..70dff9c0ef2b --- /dev/null +++ b/Documentation/devicetree/bindings/display/panel/samsung,ams581vf01.yaml @@ -0,0 +1,79 @@ +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/panel/samsung,ams581vf01.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Samsung AMS581VF01 SOFEF01-based 5.81" 1080x2340 MIPI-DSI Panel + +maintainers: + - Danila Tikhonov + +description: + The Samsung AMS581VF01 is a 5.81 inch 1080x2340 MIPI-DSI CMD mode OLED panel. + +allOf: + - $ref: panel-common.yaml# + +properties: + compatible: + const: samsung,ams581vf01 + + reg: + maxItems: 1 + + vdd3p3-supply: + description: 3.3V source voltage rail + + vddio-supply: + description: I/O source voltage rail + + vsn-supply: + description: Negative source voltage rail + + vsp-supply: + description: Positive source voltage rail + + reset-gpios: true + port: true + +required: + - compatible + - reg + - vdd3p3-supply + - vddio-supply + - vsn-supply + - vsp-supply + - reset-gpios + - port + +additionalProperties: false + +examples: + - | + #include + + dsi { + #address-cells = <1>; + #size-cells = <0>; + + panel@0 { + compatible = "samsung,ams581vf01"; + reg = <0>; + + vdd3p3-supply = <&vreg_l7c_3p0>; + vddio-supply = <&vreg_l13a_1p8>; + vsn-supply = <&vreg_ibb>; + vsp-supply = <&vreg_lab>; + + reset-gpios = <&pm6150l_gpios 9 GPIO_ACTIVE_LOW>; + + port { + panel_in: endpoint { + remote-endpoint = <&mdss_dsi0_out>; + }; + }; + }; + }; + +... diff --git a/Documentation/devicetree/bindings/display/panel/samsung,ams639rq08.yaml b/Documentation/devicetree/bindings/display/panel/samsung,ams639rq08.yaml new file mode 100644 index 000000000000..f5b6ecb96f99 --- /dev/null +++ b/Documentation/devicetree/bindings/display/panel/samsung,ams639rq08.yaml @@ -0,0 +1,80 @@ +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/panel/samsung,ams639rq08.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Samsung AMS639RQ08 EA8076-based 6.39" 1080x2340 MIPI-DSI Panel + +maintainers: + - Danila Tikhonov + - Jens Reidel + +description: + The Samsung AMS639RQ08 is a 6.39 inch 1080x2340 MIPI-DSI CMD mode AMOLED panel. + +allOf: + - $ref: panel-common.yaml# + +properties: + compatible: + const: samsung,ams639rq08 + + reg: + maxItems: 1 + + vdd3p3-supply: + description: 3.3V source voltage rail + + vddio-supply: + description: I/O source voltage rail + + vsn-supply: + description: Negative source voltage rail + + vsp-supply: + description: Positive source voltage rail + + reset-gpios: true + port: true + +required: + - compatible + - reg + - vdd3p3-supply + - vddio-supply + - vsn-supply + - vsp-supply + - reset-gpios + - port + +additionalProperties: false + +examples: + - | + #include + + dsi { + #address-cells = <1>; + #size-cells = <0>; + + panel@0 { + compatible = "samsung,ams639rq08"; + reg = <0>; + + vdd3p3-supply = <&vreg_l18a_2p8>; + vddio-supply = <&vreg_l13a_1p8>; + vsn-supply = <&vreg_ibb>; + vsp-supply = <&vreg_lab>; + + reset-gpios = <&pm6150l_gpios 9 GPIO_ACTIVE_LOW>; + + port { + panel_in: endpoint { + remote-endpoint = <&mdss_dsi0_out>; + }; + }; + }; + }; + +... diff --git a/Documentation/devicetree/bindings/display/panel/samsung,s6e3ha8.yaml b/Documentation/devicetree/bindings/display/panel/samsung,s6e3ha8.yaml new file mode 100644 index 000000000000..05a78429aaea --- /dev/null +++ b/Documentation/devicetree/bindings/display/panel/samsung,s6e3ha8.yaml @@ -0,0 +1,75 @@ +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/panel/samsung,s6e3ha8.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Samsung s6e3ha8 AMOLED DSI panel + +description: The s6e3ha8 is a 1440x2960 DPI display panel from Samsung Mobile + Displays (SMD). + +maintainers: + - Dzmitry Sankouski + +allOf: + - $ref: panel-common.yaml# + +properties: + compatible: + const: samsung,s6e3ha8 + + reg: + maxItems: 1 + + reset-gpios: true + + port: true + + vdd3-supply: + description: VDD regulator + + vci-supply: + description: VCI regulator + + vddr-supply: + description: VDDR regulator + +required: + - compatible + - reset-gpios + - vdd3-supply + - vci-supply + - vddr-supply + +unevaluatedProperties: false + +examples: + - | + #include + + dsi { + #address-cells = <1>; + #size-cells = <0>; + + panel@0 { + compatible = "samsung,s6e3ha8"; + reg = <0>; + vci-supply = <&s2dos05_ldo4>; + vddr-supply = <&s2dos05_buck1>; + vdd3-supply = <&s2dos05_ldo1>; + te-gpios = <&tlmm 10 GPIO_ACTIVE_HIGH>; + reset-gpios = <&tlmm 6 GPIO_ACTIVE_HIGH>; + pinctrl-0 = <&sde_dsi_active &sde_te_active_sleep>; + pinctrl-1 = <&sde_dsi_suspend &sde_te_active_sleep>; + pinctrl-names = "default", "sleep"; + + port { + panel_in: endpoint { + remote-endpoint = <&mdss_dsi0_out>; + }; + }; + }; + }; + +... diff --git a/Documentation/devicetree/bindings/display/panel/samsung,s6e88a0-ams427ap24.yaml b/Documentation/devicetree/bindings/display/panel/samsung,s6e88a0-ams427ap24.yaml new file mode 100644 index 000000000000..db284ba5be20 --- /dev/null +++ b/Documentation/devicetree/bindings/display/panel/samsung,s6e88a0-ams427ap24.yaml @@ -0,0 +1,65 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/panel/samsung,s6e88a0-ams427ap24.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Samsung AMS427AP24 panel with S6E88A0 controller + +maintainers: + - Jakob Hauser + +allOf: + - $ref: panel-common.yaml# + +properties: + compatible: + const: samsung,s6e88a0-ams427ap24 + + reg: + maxItems: 1 + + port: true + reset-gpios: true + flip-horizontal: true + + vdd3-supply: + description: core voltage supply + + vci-supply: + description: voltage supply for analog circuits + +required: + - compatible + - reg + - port + - reset-gpios + - vdd3-supply + - vci-supply + +additionalProperties: false + +examples: + - | + #include + + dsi { + #address-cells = <1>; + #size-cells = <0>; + + panel@0 { + compatible = "samsung,s6e88a0-ams427ap24"; + reg = <0>; + + vdd3-supply = <&pm8916_l17>; + vci-supply = <&pm8916_l6>; + reset-gpios = <&tlmm 25 GPIO_ACTIVE_LOW>; + flip-horizontal; + + port { + panel_in: endpoint { + remote-endpoint = <&mdss_dsi0_out>; + }; + }; + }; + }; diff --git a/Documentation/devicetree/bindings/display/panel/samsung,s6e8aa0.yaml b/Documentation/devicetree/bindings/display/panel/samsung,s6e8aa0.yaml index 4601fa460680..19c8cc83db97 100644 --- a/Documentation/devicetree/bindings/display/panel/samsung,s6e8aa0.yaml +++ b/Documentation/devicetree/bindings/display/panel/samsung,s6e8aa0.yaml @@ -21,6 +21,8 @@ properties: reset-gpios: true display-timings: true + flip-horizontal: true + flip-vertical: true vdd3-supply: description: core voltage supply @@ -46,14 +48,6 @@ properties: panel-height-mm: description: physical panel height [mm] - flip-horizontal: - description: boolean to flip image horizontally - type: boolean - - flip-vertical: - description: boolean to flip image vertically - type: boolean - required: - compatible - reg diff --git a/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml b/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml new file mode 100644 index 000000000000..d8e761865f27 --- /dev/null +++ b/Documentation/devicetree/bindings/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml @@ -0,0 +1,188 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/rockchip/rockchip,rk3588-dw-hdmi-qp.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Rockchip DW HDMI QP TX Encoder + +maintainers: + - Cristian Ciocaltea + +description: | + Rockchip RK3588 SoC integrates the Synopsys DesignWare HDMI QP TX controller + IP and a HDMI/eDP TX Combo PHY based on a Samsung IP block, providing the + following features, among others: + + * Fixed Rate Link (FRL) + * Display Stream Compression (DSC) + * 4K@120Hz and 8K@60Hz video modes + * Variable Refresh Rate (VRR) including Quick Media Switching (QMS) + * Fast Vactive (FVA) + * SCDC I2C DDC access + * Multi-stream audio + * Enhanced Audio Return Channel (EARC) + +allOf: + - $ref: /schemas/sound/dai-common.yaml# + +properties: + compatible: + enum: + - rockchip,rk3588-dw-hdmi-qp + + reg: + maxItems: 1 + + clocks: + items: + - description: Peripheral/APB bus clock + - description: EARC RX biphase clock + - description: Reference clock + - description: Audio interface clock + - description: TMDS/FRL link clock + - description: Video datapath clock + + clock-names: + items: + - const: pclk + - const: earc + - const: ref + - const: aud + - const: hdp + - const: hclk_vo1 + + interrupts: + items: + - description: AVP Unit interrupt + - description: CEC interrupt + - description: eARC RX interrupt + - description: Main Unit interrupt + - description: HPD interrupt + + interrupt-names: + items: + - const: avp + - const: cec + - const: earc + - const: main + - const: hpd + + phys: + maxItems: 1 + description: The HDMI/eDP PHY + + ports: + $ref: /schemas/graph.yaml#/properties/ports + + properties: + port@0: + $ref: /schemas/graph.yaml#/properties/port + description: Video port for RGB/YUV input. + + port@1: + $ref: /schemas/graph.yaml#/properties/port + description: Video port for HDMI/eDP output. + + required: + - port@0 + - port@1 + + power-domains: + maxItems: 1 + + resets: + maxItems: 2 + + reset-names: + items: + - const: ref + - const: hdp + + "#sound-dai-cells": + const: 0 + + rockchip,grf: + $ref: /schemas/types.yaml#/definitions/phandle + description: + Some HDMI QP related data is accessed through SYS GRF regs. + + rockchip,vo-grf: + $ref: /schemas/types.yaml#/definitions/phandle + description: + Additional HDMI QP related data is accessed through VO GRF regs. + +required: + - compatible + - reg + - clocks + - clock-names + - interrupts + - interrupt-names + - phys + - ports + - resets + - reset-names + - rockchip,grf + - rockchip,vo-grf + +unevaluatedProperties: false + +examples: + - | + #include + #include + #include + #include + #include + + soc { + #address-cells = <2>; + #size-cells = <2>; + + hdmi@fde80000 { + compatible = "rockchip,rk3588-dw-hdmi-qp"; + reg = <0x0 0xfde80000 0x0 0x20000>; + clocks = <&cru PCLK_HDMITX0>, + <&cru CLK_HDMITX0_EARC>, + <&cru CLK_HDMITX0_REF>, + <&cru MCLK_I2S5_8CH_TX>, + <&cru CLK_HDMIHDP0>, + <&cru HCLK_VO1>; + clock-names = "pclk", "earc", "ref", "aud", "hdp", "hclk_vo1"; + interrupts = , + , + , + , + ; + interrupt-names = "avp", "cec", "earc", "main", "hpd"; + phys = <&hdptxphy_hdmi0>; + power-domains = <&power RK3588_PD_VO1>; + resets = <&cru SRST_HDMITX0_REF>, <&cru SRST_HDMIHDP0>; + reset-names = "ref", "hdp"; + rockchip,grf = <&sys_grf>; + rockchip,vo-grf = <&vo1_grf>; + #sound-dai-cells = <0>; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0>; + + hdmi0_in_vp0: endpoint { + remote-endpoint = <&vp0_out_hdmi0>; + }; + }; + + port@1 { + reg = <1>; + + hdmi0_out_con0: endpoint { + remote-endpoint = <&hdmi_con0_in>; + }; + }; + }; + }; + }; diff --git a/Documentation/devicetree/bindings/display/sharp,ls010b7dh04.yaml b/Documentation/devicetree/bindings/display/sharp,ls010b7dh04.yaml new file mode 100644 index 000000000000..8097f091c2a5 --- /dev/null +++ b/Documentation/devicetree/bindings/display/sharp,ls010b7dh04.yaml @@ -0,0 +1,92 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/display/sharp,ls010b7dh04.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Sharp Memory LCD panels + +maintainers: + - Alex Lanzano + +description: + Sharp Memory LCDs are a series of monochrome displays that operate over + a SPI bus. The displays require a signal (VCOM) to be generated to prevent + DC bias build up resulting in pixels being unable to change. Three modes + can be used to provide the VCOM signal ("software", "external", "pwm"). + +properties: + compatible: + enum: + - sharp,ls010b7dh04 + - sharp,ls011b7dh03 + - sharp,ls012b7dd01 + - sharp,ls013b7dh03 + - sharp,ls013b7dh05 + - sharp,ls018b7dh02 + - sharp,ls027b7dh01 + - sharp,ls027b7dh01a + - sharp,ls032b7dd02 + - sharp,ls044q7dh01 + + reg: + maxItems: 1 + + spi-max-frequency: + maximum: 2000000 + + sharp,vcom-mode: + $ref: /schemas/types.yaml#/definitions/string + description: | + software - This mode relies on a software operation to send a + "maintain display" message to the display, toggling the vcom + bit on and off with each message + + external - This mode relies on an external clock to generate + the signal on the EXTCOMM pin + + pwm - This mode relies on a pwm device to generate the signal + on the EXTCOMM pin + + enum: [software, external, pwm] + + enable-gpios: true + + pwms: + maxItems: 1 + description: External VCOM signal + +required: + - compatible + - reg + - sharp,vcom-mode + +allOf: + - $ref: panel/panel-common.yaml# + - $ref: /schemas/spi/spi-peripheral-props.yaml# + + - if: + properties: + sharp,vcom-mode: + const: pwm + then: + required: + - pwms + +unevaluatedProperties: false + +examples: + - | + spi { + #address-cells = <1>; + #size-cells = <0>; + + display@0 { + compatible = "sharp,ls013b7dh03"; + reg = <0>; + spi-cs-high; + spi-max-frequency = <1000000>; + sharp,vcom-mode = "software"; + }; + }; +... diff --git a/Documentation/devicetree/bindings/gpu/arm,mali-bifrost.yaml b/Documentation/devicetree/bindings/gpu/arm,mali-bifrost.yaml index 278399adc550..735c7f06c24e 100644 --- a/Documentation/devicetree/bindings/gpu/arm,mali-bifrost.yaml +++ b/Documentation/devicetree/bindings/gpu/arm,mali-bifrost.yaml @@ -26,6 +26,7 @@ properties: - renesas,r9a07g054-mali - rockchip,px30-mali - rockchip,rk3568-mali + - rockchip,rk3576-mali - const: arm,mali-bifrost # Mali Bifrost GPU model/revision is fully discoverable - items: - enum: diff --git a/Documentation/devicetree/bindings/iio/adc/adi,ad7380.yaml b/Documentation/devicetree/bindings/iio/adc/adi,ad7380.yaml index bd19abb867d9..0065d6508824 100644 --- a/Documentation/devicetree/bindings/iio/adc/adi,ad7380.yaml +++ b/Documentation/devicetree/bindings/iio/adc/adi,ad7380.yaml @@ -67,6 +67,10 @@ properties: A 2.5V to 3.3V supply for the external reference voltage. When omitted, the internal 2.5V reference is used. + refin-supply: + description: + A 2.5V to 3.3V supply for external reference voltage, for ad7380-4 only. + aina-supply: description: The common mode voltage supply for the AINA- pin on pseudo-differential @@ -135,6 +139,23 @@ allOf: ainc-supply: false aind-supply: false + # ad7380-4 uses refin-supply as external reference. + # All other chips from ad738x family use refio as optional external reference. + # When refio-supply is omitted, internal reference is used. + - if: + properties: + compatible: + enum: + - adi,ad7380-4 + then: + properties: + refio-supply: false + required: + - refin-supply + else: + properties: + refin-supply: false + examples: - | #include diff --git a/Documentation/devicetree/bindings/iio/dac/adi,ad5686.yaml b/Documentation/devicetree/bindings/iio/dac/adi,ad5686.yaml index b4400c52bec3..713f535bb33a 100644 --- a/Documentation/devicetree/bindings/iio/dac/adi,ad5686.yaml +++ b/Documentation/devicetree/bindings/iio/dac/adi,ad5686.yaml @@ -4,7 +4,7 @@ $id: http://devicetree.org/schemas/iio/dac/adi,ad5686.yaml# $schema: http://devicetree.org/meta-schemas/core.yaml# -title: Analog Devices AD5360 and similar DACs +title: Analog Devices AD5360 and similar SPI DACs maintainers: - Michael Hennerich @@ -12,41 +12,22 @@ maintainers: properties: compatible: - oneOf: - - description: SPI devices - enum: - - adi,ad5310r - - adi,ad5672r - - adi,ad5674r - - adi,ad5676 - - adi,ad5676r - - adi,ad5679r - - adi,ad5681r - - adi,ad5682r - - adi,ad5683 - - adi,ad5683r - - adi,ad5684 - - adi,ad5684r - - adi,ad5685r - - adi,ad5686 - - adi,ad5686r - - description: I2C devices - enum: - - adi,ad5311r - - adi,ad5337r - - adi,ad5338r - - adi,ad5671r - - adi,ad5675r - - adi,ad5691r - - adi,ad5692r - - adi,ad5693 - - adi,ad5693r - - adi,ad5694 - - adi,ad5694r - - adi,ad5695r - - adi,ad5696 - - adi,ad5696r - + enum: + - adi,ad5310r + - adi,ad5672r + - adi,ad5674r + - adi,ad5676 + - adi,ad5676r + - adi,ad5679r + - adi,ad5681r + - adi,ad5682r + - adi,ad5683 + - adi,ad5683r + - adi,ad5684 + - adi,ad5684r + - adi,ad5685r + - adi,ad5686 + - adi,ad5686r reg: maxItems: 1 diff --git a/Documentation/devicetree/bindings/iio/dac/adi,ad5696.yaml b/Documentation/devicetree/bindings/iio/dac/adi,ad5696.yaml index 56b0cda0f30a..b5a88b03dc2f 100644 --- a/Documentation/devicetree/bindings/iio/dac/adi,ad5696.yaml +++ b/Documentation/devicetree/bindings/iio/dac/adi,ad5696.yaml @@ -4,7 +4,7 @@ $id: http://devicetree.org/schemas/iio/dac/adi,ad5696.yaml# $schema: http://devicetree.org/meta-schemas/core.yaml# -title: Analog Devices AD5696 and similar multi-channel DACs +title: Analog Devices AD5696 and similar I2C multi-channel DACs maintainers: - Michael Auchter @@ -16,6 +16,7 @@ properties: compatible: enum: - adi,ad5311r + - adi,ad5337r - adi,ad5338r - adi,ad5671r - adi,ad5675r diff --git a/Documentation/devicetree/bindings/interrupt-controller/fsl,ls-extirq.yaml b/Documentation/devicetree/bindings/interrupt-controller/fsl,ls-extirq.yaml index 199b34fdbefc..7ff4efc4758a 100644 --- a/Documentation/devicetree/bindings/interrupt-controller/fsl,ls-extirq.yaml +++ b/Documentation/devicetree/bindings/interrupt-controller/fsl,ls-extirq.yaml @@ -82,9 +82,6 @@ allOf: enum: - fsl,ls1043a-extirq - fsl,ls1046a-extirq - - fsl,ls1088a-extirq - - fsl,ls2080a-extirq - - fsl,lx2160a-extirq then: properties: interrupt-map: @@ -95,6 +92,29 @@ allOf: - const: 0xf - const: 0 + - if: + properties: + compatible: + contains: + enum: + - fsl,ls1088a-extirq + - fsl,ls2080a-extirq + - fsl,lx2160a-extirq +# The driver(drivers/irqchip/irq-ls-extirq.c) have not use standard DT +# function to parser interrupt-map. So it doesn't consider '#address-size' +# in parent interrupt controller, such as GIC. +# +# When dt-binding verify interrupt-map, item data matrix is spitted at +# incorrect position. Remove interrupt-map restriction because it always +# wrong. + + then: + properties: + interrupt-map-mask: + items: + - const: 0xf + - const: 0 + additionalProperties: false examples: diff --git a/Documentation/devicetree/bindings/misc/fsl,qoriq-mc.yaml b/Documentation/devicetree/bindings/misc/fsl,qoriq-mc.yaml index 01b00d89a992..df45ff56d444 100644 --- a/Documentation/devicetree/bindings/misc/fsl,qoriq-mc.yaml +++ b/Documentation/devicetree/bindings/misc/fsl,qoriq-mc.yaml @@ -113,7 +113,7 @@ properties: msi-parent: deprecated: true - $ref: /schemas/types.yaml#/definitions/phandle + maxItems: 1 description: Describes the MSI controller node handling message interrupts for the MC. When there is no translation diff --git a/Documentation/devicetree/bindings/net/brcm,unimac-mdio.yaml b/Documentation/devicetree/bindings/net/brcm,unimac-mdio.yaml index 23dfe0838dca..63bee5b542f5 100644 --- a/Documentation/devicetree/bindings/net/brcm,unimac-mdio.yaml +++ b/Documentation/devicetree/bindings/net/brcm,unimac-mdio.yaml @@ -26,6 +26,7 @@ properties: - brcm,asp-v2.1-mdio - brcm,asp-v2.2-mdio - brcm,unimac-mdio + - brcm,bcm6846-mdio reg: minItems: 1 diff --git a/Documentation/devicetree/bindings/net/xlnx,axi-ethernet.yaml b/Documentation/devicetree/bindings/net/xlnx,axi-ethernet.yaml index bbe89ea9590c..e95c21628281 100644 --- a/Documentation/devicetree/bindings/net/xlnx,axi-ethernet.yaml +++ b/Documentation/devicetree/bindings/net/xlnx,axi-ethernet.yaml @@ -34,6 +34,7 @@ properties: and length of the AXI DMA controller IO space, unless axistream-connected is specified, in which case the reg attribute of the node referenced by it is used. + minItems: 1 maxItems: 2 interrupts: @@ -181,7 +182,7 @@ examples: clock-names = "s_axi_lite_clk", "axis_clk", "ref_clk", "mgt_clk"; clocks = <&axi_clk>, <&axi_clk>, <&pl_enet_ref_clk>, <&mgt_clk>; phy-mode = "mii"; - reg = <0x00 0x40000000 0x00 0x40000>; + reg = <0x40000000 0x40000>; xlnx,rxcsum = <0x2>; xlnx,rxmem = <0x800>; xlnx,txcsum = <0x2>; diff --git a/Documentation/devicetree/bindings/phy/qcom,sc8280xp-qmp-pcie-phy.yaml b/Documentation/devicetree/bindings/phy/qcom,sc8280xp-qmp-pcie-phy.yaml index dcf4fa55fbba..380a9222a51d 100644 --- a/Documentation/devicetree/bindings/phy/qcom,sc8280xp-qmp-pcie-phy.yaml +++ b/Documentation/devicetree/bindings/phy/qcom,sc8280xp-qmp-pcie-phy.yaml @@ -154,8 +154,6 @@ allOf: - qcom,sm8550-qmp-gen4x2-pcie-phy - qcom,sm8650-qmp-gen3x2-pcie-phy - qcom,sm8650-qmp-gen4x2-pcie-phy - - qcom,x1e80100-qmp-gen3x2-pcie-phy - - qcom,x1e80100-qmp-gen4x2-pcie-phy then: properties: clocks: @@ -171,6 +169,8 @@ allOf: - qcom,sc8280xp-qmp-gen3x1-pcie-phy - qcom,sc8280xp-qmp-gen3x2-pcie-phy - qcom,sc8280xp-qmp-gen3x4-pcie-phy + - qcom,x1e80100-qmp-gen3x2-pcie-phy + - qcom,x1e80100-qmp-gen4x2-pcie-phy - qcom,x1e80100-qmp-gen4x4-pcie-phy then: properties: @@ -201,6 +201,7 @@ allOf: - qcom,sm8550-qmp-gen4x2-pcie-phy - qcom,sm8650-qmp-gen4x2-pcie-phy - qcom,x1e80100-qmp-gen4x2-pcie-phy + - qcom,x1e80100-qmp-gen4x4-pcie-phy then: properties: resets: diff --git a/Documentation/devicetree/bindings/sound/davinci-mcasp-audio.yaml b/Documentation/devicetree/bindings/sound/davinci-mcasp-audio.yaml index 7735e08d35ba..beef193aaaeb 100644 --- a/Documentation/devicetree/bindings/sound/davinci-mcasp-audio.yaml +++ b/Documentation/devicetree/bindings/sound/davinci-mcasp-audio.yaml @@ -102,21 +102,21 @@ properties: default: 2 interrupts: - anyOf: - - minItems: 1 - items: - - description: TX interrupt - - description: RX interrupt - - items: - - description: common/combined interrupt + minItems: 1 + maxItems: 2 interrupt-names: oneOf: - - minItems: 1 + - description: TX interrupt + const: tx + - description: RX interrupt + const: rx + - description: TX and RX interrupts items: - const: tx - const: rx - - const: common + - description: Common/combined interrupt + const: common fck_parent: $ref: /schemas/types.yaml#/definitions/string diff --git a/Documentation/devicetree/bindings/sound/qcom,sm8250.yaml b/Documentation/devicetree/bindings/sound/qcom,sm8250.yaml index 1d3acdc0c733..2e2e01493a5f 100644 --- a/Documentation/devicetree/bindings/sound/qcom,sm8250.yaml +++ b/Documentation/devicetree/bindings/sound/qcom,sm8250.yaml @@ -30,6 +30,7 @@ properties: - qcom,apq8096-sndcard - qcom,qcm6490-idp-sndcard - qcom,qcs6490-rb3gen2-sndcard + - qcom,qrb4210-rb2-sndcard - qcom,qrb5165-rb5-sndcard - qcom,sc7180-qdsp6-sndcard - qcom,sc8280xp-sndcard diff --git a/Documentation/devicetree/bindings/sound/renesas,rsnd.yaml b/Documentation/devicetree/bindings/sound/renesas,rsnd.yaml index 3bc93c59535e..6d0d1514cd42 100644 --- a/Documentation/devicetree/bindings/sound/renesas,rsnd.yaml +++ b/Documentation/devicetree/bindings/sound/renesas,rsnd.yaml @@ -302,7 +302,7 @@ allOf: reg-names: items: enum: - - scu + - sru - ssi - adg # for Gen2/Gen3 diff --git a/Documentation/devicetree/bindings/sound/rockchip,rk3308-codec.yaml b/Documentation/devicetree/bindings/sound/rockchip,rk3308-codec.yaml index ecf3d7d968c8..2cf229a076f0 100644 --- a/Documentation/devicetree/bindings/sound/rockchip,rk3308-codec.yaml +++ b/Documentation/devicetree/bindings/sound/rockchip,rk3308-codec.yaml @@ -48,6 +48,10 @@ properties: - const: mclk_rx - const: hclk + port: + $ref: audio-graph-port.yaml# + unevaluatedProperties: false + resets: maxItems: 1 diff --git a/Documentation/devicetree/bindings/trivial-devices.yaml b/Documentation/devicetree/bindings/trivial-devices.yaml index 0108d7507215..9bf0fb17a05e 100644 --- a/Documentation/devicetree/bindings/trivial-devices.yaml +++ b/Documentation/devicetree/bindings/trivial-devices.yaml @@ -101,8 +101,6 @@ properties: - domintech,dmard09 # DMARD10: 3-axis Accelerometer - domintech,dmard10 - # Elgin SPI-controlled LCD - - elgin,jg10309-01 # MMA7660FC: 3-Axis Orientation/Motion Detection Sensor - fsl,mma7660 # MMA8450Q: Xtrinsic Low-power, 3-axis Xtrinsic Accelerometer diff --git a/Documentation/devicetree/bindings/vendor-prefixes.yaml b/Documentation/devicetree/bindings/vendor-prefixes.yaml index b320a39de7fe..46901c8d7eb0 100644 --- a/Documentation/devicetree/bindings/vendor-prefixes.yaml +++ b/Documentation/devicetree/bindings/vendor-prefixes.yaml @@ -752,6 +752,8 @@ patternProperties: description: Japan Display Inc. "^jedec,.*": description: JEDEC Solid State Technology Association + "^jenson,.*": + description: Jenson Display Co. Ltd. "^jesurun,.*": description: Shenzhen Jesurun Electronics Business Dept. "^jethome,.*": diff --git a/Documentation/driver-api/wmi.rst b/Documentation/driver-api/wmi.rst index 6ca58c8249e5..4e8dbdb1fc67 100644 --- a/Documentation/driver-api/wmi.rst +++ b/Documentation/driver-api/wmi.rst @@ -7,12 +7,11 @@ WMI Driver API The WMI driver core supports a more modern bus-based interface for interacting with WMI devices, and an older GUID-based interface. The latter interface is considered to be deprecated, so new WMI drivers should generally avoid it since -it has some issues with multiple WMI devices and events sharing the same GUIDs -and/or notification IDs. The modern bus-based interface instead maps each -WMI device to a :c:type:`struct wmi_device `, so it supports -WMI devices sharing GUIDs and/or notification IDs. Drivers can then register -a :c:type:`struct wmi_driver `, which will be bound to compatible -WMI devices by the driver core. +it has some issues with multiple WMI devices sharing the same GUID. +The modern bus-based interface instead maps each WMI device to a +:c:type:`struct wmi_device `, so it supports WMI devices sharing the +same GUID. Drivers can then register a :c:type:`struct wmi_driver ` +which will be bound to compatible WMI devices by the driver core. .. kernel-doc:: include/linux/wmi.h :internal: diff --git a/Documentation/filesystems/caching/cachefiles.rst b/Documentation/filesystems/caching/cachefiles.rst index e04a27bdbe19..b3ccc782cb3b 100644 --- a/Documentation/filesystems/caching/cachefiles.rst +++ b/Documentation/filesystems/caching/cachefiles.rst @@ -115,7 +115,7 @@ set up cache ready for use. The following script commands are available: This mask can also be set through sysfs, eg:: - echo 5 >/sys/modules/cachefiles/parameters/debug + echo 5 > /sys/module/cachefiles/parameters/debug Starting the Cache diff --git a/Documentation/filesystems/iomap/operations.rst b/Documentation/filesystems/iomap/operations.rst index 8e6c721d2330..b93115ab8748 100644 --- a/Documentation/filesystems/iomap/operations.rst +++ b/Documentation/filesystems/iomap/operations.rst @@ -208,7 +208,7 @@ The filesystem must arrange to `cancel such `reservations `_ because writeback will not consume the reservation. -The ``iomap_file_buffered_write_punch_delalloc`` can be called from a +The ``iomap_write_delalloc_release`` can be called from a ``->iomap_end`` function to find all the clean areas of the folios caching a fresh (``IOMAP_F_NEW``) delalloc mapping. It takes the ``invalidate_lock``. diff --git a/Documentation/filesystems/netfs_library.rst b/Documentation/filesystems/netfs_library.rst index f0d2cb257bb8..73f0bfd7e903 100644 --- a/Documentation/filesystems/netfs_library.rst +++ b/Documentation/filesystems/netfs_library.rst @@ -592,4 +592,3 @@ API Function Reference .. kernel-doc:: include/linux/netfs.h .. kernel-doc:: fs/netfs/buffered_read.c -.. kernel-doc:: fs/netfs/io.c diff --git a/Documentation/gpu/amdgpu/display/dc-arch-overview.svg b/Documentation/gpu/amdgpu/display/dc-arch-overview.svg new file mode 100644 index 000000000000..23394931cf26 --- /dev/null +++ b/Documentation/gpu/amdgpu/display/dc-arch-overview.svg @@ -0,0 +1,731 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + image/svg+xml + + + + + + + + + + Board/Platform + SoC + Component + DRAM + + dc_plane + + + + dc_plane + + + + + DC + + + + + dc_link + + + + + dc_link + + + + + + + dc_link + + + + + + + + + + + DCN + SoC + Board/Platform + Display + Connector + Connector + + diff --git a/Documentation/gpu/amdgpu/display/dc-components.svg b/Documentation/gpu/amdgpu/display/dc-components.svg new file mode 100644 index 000000000000..f84bb2a57c05 --- /dev/null +++ b/Documentation/gpu/amdgpu/display/dc-components.svg @@ -0,0 +1,732 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + image/svg+xml + + + + + + + + + Core + + + + + + + + + + Display Core API (dc/dc.h) + Link + Hardware Sequencer API(dc/inc/hw_sequence.h) + Hardware Sequencer + Block Level API (dc/inc/hw) + + + DCHUB + + + + HUBP + + + + DPP + + + + MPC + + + + ... + + Hardware Registers + DMUBBlock + DMUB Hardware API(dmub/dmub_srv.h) + DMUB Service + DMUB Service API(dc/dc_dmub_srv.h) + + diff --git a/Documentation/gpu/amdgpu/display/dc-debug.rst b/Documentation/gpu/amdgpu/display/dc-debug.rst index 817631b1dbf3..013f63b271f3 100644 --- a/Documentation/gpu/amdgpu/display/dc-debug.rst +++ b/Documentation/gpu/amdgpu/display/dc-debug.rst @@ -2,6 +2,181 @@ Display Core Debug tools ======================== +In this section, you will find helpful information on debugging the amdgpu +driver from the display perspective. This page introduces debug mechanisms and +procedures to help you identify if some issues are related to display code. + +Narrow down display issues +========================== + +Since the display is the driver's visual component, it is common to see users +reporting issues as a display when another component causes the problem. This +section equips users to determine if a specific issue was caused by the display +component or another part of the driver. + +DC dmesg important messages +--------------------------- + +The dmesg log is the first source of information to be checked, and amdgpu +takes advantage of this feature by logging some valuable information. When +looking for the issues associated with amdgpu, remember that each component of +the driver (e.g., smu, PSP, dm, etc.) is loaded one by one, and this +information can be found in the dmesg log. In this sense, look for the part of +the log that looks like the below log snippet:: + + [ 4.254295] [drm] initializing kernel modesetting (IP DISCOVERY 0x1002:0x744C 0x1002:0x0E3B 0xC8). + [ 4.254718] [drm] register mmio base: 0xFCB00000 + [ 4.254918] [drm] register mmio size: 1048576 + [ 4.260095] [drm] add ip block number 0 + [ 4.260318] [drm] add ip block number 1 + [ 4.260510] [drm] add ip block number 2 + [ 4.260696] [drm] add ip block number 3 + [ 4.260878] [drm] add ip block number 4 + [ 4.261057] [drm] add ip block number 5 + [ 4.261231] [drm] add ip block number 6 + [ 4.261402] [drm] add ip block number 7 + [ 4.261568] [drm] add ip block number 8 + [ 4.261729] [drm] add ip block number 9 + [ 4.261887] [drm] add ip block number 10 + +From the above example, you can see the line that reports that ``, +(**Display Manager**), was loaded, which means that display can be part of the +issue. If you do not see that line, something else might have failed before +amdgpu loads the display component, indicating that we don't have a +display issue. + +After you identified that the DM was loaded correctly, you can check for the +display version of the hardware in use, which can be retrieved from the dmesg +log with the command:: + + dmesg | grep -i 'display core' + +This command shows a message that looks like this:: + + [ 4.655828] [drm] Display Core v3.2.285 initialized on DCN 3.2 + +This message has two key pieces of information: + +* **The DC version (e.g., v3.2.285)**: Display developers release a new DC version + every week, and this information can be advantageous in a situation where a + user/developer must find a good point versus a bad point based on a tested + version of the display code. Remember from page :ref:`Display Core `, + that every week the new patches for display are heavily tested with IGT and + manual tests. +* **The DCN version (e.g., DCN 3.2)**: The DCN block is associated with the + hardware generation, and the DCN version conveys the hardware generation that + the driver is currently running. This information helps to narrow down the + code debug area since each DCN version has its files in the DC folder per DCN + component (from the example, the developer might want to focus on + files/folders/functions/structs with the dcn32 label might be executed). + However, keep in mind that DC reuses code across different DCN versions; for + example, it is expected to have some callbacks set in one DCN that are the same + as those from another DCN. In summary, use the DCN version just as a guide. + +From the dmesg file, it is also possible to get the ATOM bios code by using:: + + dmesg | grep -i 'ATOM BIOS' + +Which generates an output that looks like this:: + + [ 4.274534] amdgpu: ATOM BIOS: 113-D7020100-102 + +This type of information is useful to be reported. + +Avoid loading display core +-------------------------- + +Sometimes, it might be hard to figure out which part of the driver is causing +the issue; if you suspect that the display is not part of the problem and your +bug scenario is simple (e.g., some desktop configuration) you can try to remove +the display component from the equation. First, you need to identify `dm` ID +from the dmesg log; for example, search for the following log:: + + [ 4.254295] [drm] initializing kernel modesetting (IP DISCOVERY 0x1002:0x744C 0x1002:0x0E3B 0xC8). + [..] + [ 4.260095] [drm] add ip block number 0 + [ 4.260318] [drm] add ip block number 1 + [..] + [ 4.261057] [drm] add ip block number 5 + +Notice from the above example that the `dm` id is 5 for this specific hardware. +Next, you need to run the following binary operation to identify the IP block +mask:: + + 0xffffffff & ~(1 << [DM ID]) + +From our example the IP mask is:: + + 0xffffffff & ~(1 << 5) = 0xffffffdf + +Finally, to disable DC, you just need to set the below parameter in your +bootloader:: + + amdgpu.ip_block_mask = 0xffffffdf + +If you can boot your system with the DC disabled and still see the issue, it +means you can rule DC out of the equation. However, if the bug disappears, you +still need to consider the DC part of the problem and keep narrowing down the +issue. In some scenarios, disabling DC is impossible since it might be +necessary to use the display component to reproduce the issue (e.g., play a +game). + +**Note: This will probably lead to the absence of a display output.** + +Display flickering +------------------ + +Display flickering might have multiple causes; one is the lack of proper power +to the GPU or problems in the DPM switches. A good first generic verification +is to set the GPU to use high voltage:: + + bash -c "echo high > /sys/class/drm/card0/device/power_dpm_force_performance_level" + +The above command sets the GPU/APU to use the maximum power allowed which +disables DPM switches. If forcing DPM levels high does not fix the issue, it +is less likely that the issue is related to power management. If the issue +disappears, there is a good chance that other components might be involved, and +the display should not be ignored since this could be a DPM issues. From the +display side, if the power increase fixes the issue, it is worth debugging the +clock configuration and the pipe split police used in the specific +configuration. + +Display artifacts +----------------- + +Users may see some screen artifacts that can be categorized into two different +types: localized artifacts and general artifacts. The localized artifacts +happen in some specific areas, such as around the UI window corners; if you see +this type of issue, there is a considerable chance that you have a userspace +problem, likely Mesa or similar. The general artifacts usually happen on the +entire screen. They might be caused by a misconfiguration at the driver level +of the display parameters, but the userspace might also cause this issue. One +way to identify the source of the problem is to take a screenshot or make a +desktop video capture when the problem happens; after checking the +screenshot/video recording, if you don't see any of the artifacts, it means +that the issue is likely on the the driver side. If you can still see the +problem in the data collected, it is an issue that probably happened during +rendering, and the display code just got the framebuffer already corrupted. + +Disabling/Enabling specific features +==================================== + +DC has a struct named `dc_debug_options`, which is statically initialized by +all DCE/DCN components based on the specific hardware characteristic. This +structure usually facilitates the bring-up phase since developers can start +with many disabled features and enable them individually. This is also an +important debug feature since users can change it when debugging specific +issues. + +For example, dGPU users sometimes see a problem where a horizontal fillet of +flickering happens in some specific part of the screen. This could be an +indication of Sub-Viewport issues; after the users identified the target DCN, +they can set the `force_disable_subvp` field to true in the statically +initialized version of `dc_debug_options` to see if the issue gets fixed. Along +the same lines, users/developers can also try to turn off `fams2_config` and +`enable_single_display_2to1_odm_policy`. In summary, the `dc_debug_options` is +an interesting form for identifying the problem. + DC Visual Confirmation ====================== @@ -76,6 +251,18 @@ change in real-time by using something like:: When reporting a bug related to DC, consider attaching this log before and after you reproduce the bug. +Collect Firmware information +============================ + +When reporting issues, it is important to have the firmware information since +it can be helpful for debugging purposes. To get all the firmware information, +use the command:: + + cat /sys/kernel/debug/dri/0/amdgpu_firmware_info + +From the display perspective, pay attention to the firmware of the DMCU and +DMCUB. + DMUB Firmware Debug =================== diff --git a/Documentation/gpu/amdgpu/display/dcn-blocks.rst b/Documentation/gpu/amdgpu/display/dcn-blocks.rst index 5e34366f6dbe..756957128dad 100644 --- a/Documentation/gpu/amdgpu/display/dcn-blocks.rst +++ b/Documentation/gpu/amdgpu/display/dcn-blocks.rst @@ -1,3 +1,5 @@ +.. _dcn_blocks: + ========== DCN Blocks ========== diff --git a/Documentation/gpu/amdgpu/display/dcn-overview.rst b/Documentation/gpu/amdgpu/display/dcn-overview.rst index 9fea6500448b..eb54a6802e04 100644 --- a/Documentation/gpu/amdgpu/display/dcn-overview.rst +++ b/Documentation/gpu/amdgpu/display/dcn-overview.rst @@ -1,3 +1,5 @@ +.. _dcn_overview: + ======================= Display Core Next (DCN) ======================= diff --git a/Documentation/gpu/amdgpu/display/index.rst b/Documentation/gpu/amdgpu/display/index.rst index f0c342e00a39..bd2d797c123e 100644 --- a/Documentation/gpu/amdgpu/display/index.rst +++ b/Documentation/gpu/amdgpu/display/index.rst @@ -90,6 +90,7 @@ table of content: display-manager.rst dcn-overview.rst dcn-blocks.rst + programming-model-dcn.rst mpo-overview.rst dc-debug.rst display-contributing.rst diff --git a/Documentation/gpu/amdgpu/display/programming-model-dcn.rst b/Documentation/gpu/amdgpu/display/programming-model-dcn.rst new file mode 100644 index 000000000000..c1b48d49fb0b --- /dev/null +++ b/Documentation/gpu/amdgpu/display/programming-model-dcn.rst @@ -0,0 +1,162 @@ +==================== +DC Programming Model +==================== + +In the :ref:`Display Core Next (DCN) ` and :ref:`DCN Block +` pages, you learned about the hardware components and how they +interact with each other. On this page, the focus is shifted to the display +code architecture. Hence, it is reasonable to remind the reader that the code +in DC is shared with other OSes; for this reason, DC provides a set of +abstractions and operations to connect different APIs with the hardware +configuration. See DC as a service available for a Display Manager (amdgpu_dm) +to access and configure DCN/DCE hardware (DCE is also part of DC, but for +simplicity's sake, this documentation only examines DCN). + +.. note:: + For this page, we will use the term GPU to refers to dGPU and APU. + +Overview +======== + +From the display hardware perspective, it is plausible to expect that if a +problem is well-defined, it will probably be implemented at the hardware level. +On the other hand, when there are multiple ways of achieving something without +a very well-defined scope, the solution is usually implemented as a policy at +the DC level. In other words, some policies are defined in the DC core +(resource management, power optimization, image quality, etc.), and the others +implemented in hardware are enabled via DC configuration. + +In terms of hardware management, DCN has multiple instances of the same block +(e.g., HUBP, DPP, MPC, etc), and during the driver execution, it might be +necessary to use some of these instances. The core has policies in place for +handling those instances. Regarding resource management, the DC objective is +quite simple: minimize the hardware shuffle when the driver performs some +actions. When the state changes from A to B, the transition is considered +easier to maneuver if the hardware resource is still used for the same set of +driver objects. Usually, adding and removing a resource to a `pipe_ctx` (more +details below) is not a problem; however, moving a resource from one `pipe_ctx` +to another should be avoided. + +Another area of influence for DC is power optimization, which has a myriad of +arrangement possibilities. In some way, just displaying an image via DCN should +be relatively straightforward; however, showing it with the best power +footprint is more desirable, but it has many associated challenges. +Unfortunately, there is no straight-forward analytic way to determine if a +configuration is the best one for the context due to the enormous variety of +variables related to this problem (e.g., many different DCN/DCE hardware +versions, different displays configurations, etc.) for this reason DC +implements a dedicated library for trying some configuration and verify if it +is possible to support it or not. This type of policy is extremely complex to +create and maintain, and amdgpu driver relies on Display Mode Library (DML) to +generate the best decisions. + +In summary, DC must deal with the complexity of handling multiple scenarios and +determine policies to manage them. All of the above information is conveyed to +give the reader some idea about the complexity of driving a display from the +driver's perspective. This page hopes to allow the reader to better navigate +over the amdgpu display code. + +Display Driver Architecture Overview +==================================== + +The diagram below provides an overview of the display driver architecture; +notice it illustrates the software layers adopted by DC: + +.. kernel-figure:: dc-components.svg + +The first layer of the diagram is the high-level DC API represented by the +`dc.h` file; below it are two big blocks represented by Core and Link. Next is +the hardware configuration block; the main file describing it is +the`hw_sequencer.h`, where the implementation of the callbacks can be found in +the hardware sequencer folder. Almost at the end, you can see the block level +API (`dc/inc/hw`), which represents each DCN low-level block, such as HUBP, +DPP, MPC, OPTC, etc. Notice on the left side of the diagram that we have a +different set of layers representing the interaction with the DMUB +microcontroller. + +Basic Objects +------------- + +The below diagram outlines the basic display objects. In particular, pay +attention to the names in the boxes since they represent a data structure in +the driver: + +.. kernel-figure:: dc-arch-overview.svg + +Let's start with the central block in the image, `dc`. The `dc` struct is +initialized per GPU; for example, one GPU has one `dc` instance, two GPUs have +two `dc` instances, and so forth. In other words we have one 'dc' per 'amdgpu' +instance. In some ways, this object behaves like the `Singleton` pattern. + +After the `dc` block in the diagram, you can see the `dc_link` component, which +is a low-level abstraction for the connector. One interesting aspect of the +image is that connectors are not part of the DCN block; they are defined by the +platform/board and not by the SoC. The `dc_link` struct is the high-level data +container with information such as connected sinks, connection status, signal +types, etc. After `dc_link`, there is the `dc_sink`, which is the object that +represents the connected display. + +.. note:: + For historical reasons, we used the name `dc_link`, which gives the + wrong impression that this abstraction only deals with physical connections + that the developer can easily manipulate. However, this also covers + conections like eDP or cases where the output is connected to other devices. + +There are two structs that are not represented in the diagram since they were +elaborated in the DCN overview page (check the DCN block diagram :ref:`Display +Core Next (DCN) `); still, it is worth bringing back for this +overview which is `dc_stream` and `dc_state`. The `dc_stream` stores many +properties associated with the data transmission, but most importantly, it +represents the data flow from the connector to the display. Next we have +`dc_state`, which represents the logic state within the hardware at the moment; +`dc_state` is composed of `dc_stream` and `dc_plane`. The `dc_stream` is the DC +version of `drm_crtc` and represents the post-blending pipeline. + +Speaking of the `dc_plane` data structure (first part of the diagram), you can +think about it as an abstraction similar to `drm_plane` that represents the +pre-blending portion of the pipeline. This image was probably processed by GFX +and is ready to be composited under a `dc_stream`. Normally, the driver may +have one or more `dc_plane` connected to the same `dc_stream`, which defines a +composition at the DC level. + +Basic Operations +---------------- + +Now that we have covered the basic objects, it is time to examine some of the +basic hardware/software operations. Let's start with the `dc_create()` +function, which directly works with the `dc` data struct; this function behaves +like a constructor responsible for the basic software initialization and +preparing for enabling other parts of the API. It is important to highlight +that this operation does not touch any hardware configuration; it is only a +software initialization. + +Next, we have the `dc_hardware_init()`, which also relies on the `dc` data +struct. Its main function is to put the hardware in a valid state. It is worth +highlighting that the hardware might initialize in an unknown state, and it is +a requirement to put it in a valid state; this function has multiple callbacks +for the hardware-specific initialization, whereas `dc_hardware_init` does the +hardware initialization and is the first point where we touch hardware. + +The `dc_get_link_at_index` is an operation that depends on the `dc_link` data +structure. This function retrieves and enumerates all the `dc_links` available +on the device; this is required since this information is not part of the SoC +definition but depends on the board configuration. As soon as the `dc_link` is +initialized, it is useful to figure out if any of them are already connected to +the display by using the `dc_link_detect()` function. After the driver figures +out if any display is connected to the device, the challenging phase starts: +configuring the monitor to show something. Nonetheless, dealing with the ideal +configuration is not a DC task since this is the Display Manager (`amdgpu_dm`) +responsibility which in turn is responsible for dealing with the atomic +commits. The only interface DC provides to the configuration phase is the +function `dc_validate_with_context` that receives the configuration information +and, based on that, validates whether the hardware can support it or not. It is +important to add that even if the display supports some specific configuration, +it does not mean the DCN hardware can support it. + +After the DM and DC agree upon the configuration, the stream configuration +phase starts. This task activates one or more `dc_stream` at this phase, and in +the best-case scenario, you might be able to turn the display on with a black +screen (it does not show anything yet since it does not have any plane +associated with it). The final step would be to call the +`dc_update_planes_and_stream,` which will add or remove planes. + diff --git a/Documentation/gpu/automated_testing.rst b/Documentation/gpu/automated_testing.rst index 2d5a28866afe..6d7c6086034d 100644 --- a/Documentation/gpu/automated_testing.rst +++ b/Documentation/gpu/automated_testing.rst @@ -68,19 +68,25 @@ known to behave unreliably. These tests won't cause a job to fail regardless of the result. They will still be run. Each new flake entry must be associated with a link to the email reporting the -bug to the author of the affected driver, the board name or Device Tree name of -the board, the first kernel version affected, the IGT version used for tests, -and an approximation of the failure rate. +bug to the author of the affected driver or the relevant GitLab issue. The entry +must also include the board name or Device Tree name, the first kernel version +affected, the IGT version used for tests, and an approximation of the failure rate. They should be provided under the following format:: - # Bug Report: $LORE_OR_PATCHWORK_URL + # Bug Report: $LORE_URL_OR_GITLAB_ISSUE # Board Name: broken-board.dtb # Linux Version: 6.6-rc1 # IGT Version: 1.28-gd2af13d9f # Failure Rate: 100 flaky-test +Use the appropriate link below to create a GitLab issue: +amdgpu driver: https://gitlab.freedesktop.org/drm/amd/-/issues +i915 driver: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues +msm driver: https://gitlab.freedesktop.org/drm/msm/-/issues +xe driver: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues + drivers/gpu/drm/ci/${DRIVER_NAME}-${HW_REVISION}-skips.txt ----------------------------------------------------------- diff --git a/Documentation/gpu/drivers.rst b/Documentation/gpu/drivers.rst index b899cbc5c2b4..1f17ad0790d7 100644 --- a/Documentation/gpu/drivers.rst +++ b/Documentation/gpu/drivers.rst @@ -22,6 +22,8 @@ GPU Driver Documentation afbc komeda-kms panfrost + panthor + zynqmp .. only:: subproject and html diff --git a/Documentation/gpu/drm-client.rst b/Documentation/gpu/drm-client.rst index 58b5a1d1219d..cbcfe30de777 100644 --- a/Documentation/gpu/drm-client.rst +++ b/Documentation/gpu/drm-client.rst @@ -13,3 +13,6 @@ Kernel clients .. kernel-doc:: drivers/gpu/drm/drm_client_modeset.c :export: + +.. kernel-doc:: drivers/gpu/drm/drm_client_event.c + :export: diff --git a/Documentation/gpu/drm-internals.rst b/Documentation/gpu/drm-internals.rst index 11d9a5730fb2..cb9ae282771c 100644 --- a/Documentation/gpu/drm-internals.rst +++ b/Documentation/gpu/drm-internals.rst @@ -75,18 +75,6 @@ Module Initialization .. kernel-doc:: include/drm/drm_module.h :doc: overview -Managing Ownership of the Framebuffer Aperture ----------------------------------------------- - -.. kernel-doc:: drivers/gpu/drm/drm_aperture.c - :doc: overview - -.. kernel-doc:: include/drm/drm_aperture.h - :internal: - -.. kernel-doc:: drivers/gpu/drm/drm_aperture.c - :export: - Device Instance and Driver Handling ----------------------------------- diff --git a/Documentation/gpu/drm-kms-helpers.rst b/Documentation/gpu/drm-kms-helpers.rst index 8435e8621cc0..8cf2f041af47 100644 --- a/Documentation/gpu/drm-kms-helpers.rst +++ b/Documentation/gpu/drm-kms-helpers.rst @@ -110,15 +110,6 @@ fbdev Helper Functions Reference .. kernel-doc:: drivers/gpu/drm/drm_fb_helper.c :doc: fbdev helpers -.. kernel-doc:: drivers/gpu/drm/drm_fbdev_dma.c - :export: - -.. kernel-doc:: drivers/gpu/drm/drm_fbdev_shmem.c - :export: - -.. kernel-doc:: drivers/gpu/drm/drm_fbdev_ttm.c - :export: - .. kernel-doc:: include/drm/drm_fb_helper.h :internal: @@ -181,7 +172,7 @@ Bridge Operations Bridge Connector Helper ----------------------- -.. kernel-doc:: drivers/gpu/drm/drm_bridge_connector.c +.. kernel-doc:: drivers/gpu/drm/display/drm_bridge_connector.c :doc: overview @@ -204,7 +195,7 @@ MIPI-DSI bridge operation Bridge Connector Helper Reference --------------------------------- -.. kernel-doc:: drivers/gpu/drm/drm_bridge_connector.c +.. kernel-doc:: drivers/gpu/drm/display/drm_bridge_connector.c :export: Panel-Bridge Helper Reference diff --git a/Documentation/gpu/drm-uapi.rst b/Documentation/gpu/drm-uapi.rst index 370d820be248..b75cc9a70d1f 100644 --- a/Documentation/gpu/drm-uapi.rst +++ b/Documentation/gpu/drm-uapi.rst @@ -305,13 +305,26 @@ Kernel Mode Driver ------------------ The KMD is responsible for checking if the device needs a reset, and to perform -it as needed. Usually a hang is detected when a job gets stuck executing. KMD -should keep track of resets, because userspace can query any time about the -reset status for a specific context. This is needed to propagate to the rest of -the stack that a reset has happened. Currently, this is implemented by each -driver separately, with no common DRM interface. Ideally this should be properly -integrated at DRM scheduler to provide a common ground for all drivers. After a -reset, KMD should reject new command submissions for affected contexts. +it as needed. Usually a hang is detected when a job gets stuck executing. + +Propagation of errors to userspace has proven to be tricky since it goes in +the opposite direction of the usual flow of commands. Because of this vendor +independent error handling was added to the &dma_fence object, this way drivers +can add an error code to their fences before signaling them. See function +dma_fence_set_error() on how to do this and for examples of error codes to use. + +The DRM scheduler also allows setting error codes on all pending fences when +hardware submissions are restarted after an reset. Error codes are also +forwarded from the hardware fence to the scheduler fence to bubble up errors +to the higher levels of the stack and eventually userspace. + +Fence errors can be queried by userspace through the generic SYNC_IOC_FILE_INFO +IOCTL as well as through driver specific interfaces. + +Additional to setting fence errors drivers should also keep track of resets per +context, the DRM scheduler provides the drm_sched_entity_error() function as +helper for this use case. After a reset, KMD should reject new command +submissions for affected contexts. User Mode Driver ---------------- diff --git a/Documentation/gpu/drm-usage-stats.rst b/Documentation/gpu/drm-usage-stats.rst index a80f95ca1b2f..2717cb2a597e 100644 --- a/Documentation/gpu/drm-usage-stats.rst +++ b/Documentation/gpu/drm-usage-stats.rst @@ -73,6 +73,11 @@ scope of each device, in which case `drm-pdev` shall be present as well. Userspace should make sure to not double account any usage statistics by using the above described criteria in order to associate data to individual clients. +- drm-client-name: + +String optionally set by userspace using DRM_IOCTL_SET_CLIENT_NAME. + + Utilization ^^^^^^^^^^^ @@ -144,7 +149,9 @@ Memory Each possible memory type which can be used to store buffer objects by the GPU in question shall be given a stable and unique name to be returned as the -string here. The name "memory" is reserved to refer to normal system memory. +string here. + +The region name "memory" is reserved to refer to normal system memory. Value shall reflect the amount of storage currently consumed by the buffer objects belong to this client, in the respective memory region. @@ -152,6 +159,9 @@ objects belong to this client, in the respective memory region. Default unit shall be bytes with optional unit specifiers of 'KiB' or 'MiB' indicating kibi- or mebi-bytes. +This key is deprecated and is an alias for drm-resident-. Only one of +the two should be present in the output. + - drm-shared-: [KiB|MiB] The total size of buffers that are shared with another file (e.g., have more @@ -159,20 +169,34 @@ than a single handle). - drm-total-: [KiB|MiB] -The total size of buffers that including shared and private memory. +The total size of all created buffers including shared and private memory. The +backing store for the buffers does not have to be currently instantiated to be +counted under this category. - drm-resident-: [KiB|MiB] -The total size of buffers that are resident in the specified region. +The total size of buffers that are resident (have their backing store present or +instantiated) in the specified region. + +This is an alias for drm-memory- and only one of the two should be +present in the output. - drm-purgeable-: [KiB|MiB] The total size of buffers that are purgeable. +For example drivers which implement a form of 'madvise' like functionality can +here count buffers which have instantiated backing store, but have been marked +with an equivalent of MADV_DONTNEED. + - drm-active-: [KiB|MiB] The total size of buffers that are active on one or more engines. +One practical example of this can be presence of unsignaled fences in an GEM +buffer reservation object. Therefore the active category is a subset of +resident. + Implementation Details ====================== @@ -186,4 +210,5 @@ Driver specific implementations * :ref:`i915-usage-stats` * :ref:`panfrost-usage-stats` +* :ref:`panthor-usage-stats` * :ref:`xe-usage-stats` diff --git a/Documentation/gpu/panthor.rst b/Documentation/gpu/panthor.rst new file mode 100644 index 000000000000..3f8979fa2b86 --- /dev/null +++ b/Documentation/gpu/panthor.rst @@ -0,0 +1,46 @@ +.. SPDX-License-Identifier: GPL-2.0+ + +========================= + drm/Panthor CSF driver +========================= + +.. _panthor-usage-stats: + +Panthor DRM client usage stats implementation +============================================== + +The drm/Panthor driver implements the DRM client usage stats specification as +documented in :ref:`drm-client-usage-stats`. + +Example of the output showing the implemented key value pairs and entirety of +the currently possible format options: + +:: + pos: 0 + flags: 02400002 + mnt_id: 29 + ino: 491 + drm-driver: panthor + drm-client-id: 10 + drm-engine-panthor: 111110952750 ns + drm-cycles-panthor: 94439687187 + drm-maxfreq-panthor: 1000000000 Hz + drm-curfreq-panthor: 1000000000 Hz + drm-total-memory: 16480 KiB + drm-shared-memory: 0 + drm-active-memory: 16200 KiB + drm-resident-memory: 16480 KiB + drm-purgeable-memory: 0 + +Possible `drm-engine-` key names are: `panthor`. +`drm-curfreq-` values convey the current operating frequency for that engine. + +Users must bear in mind that engine and cycle sampling are disabled by default, +because of power saving concerns. `fdinfo` users and benchmark applications which +query the fdinfo file must make sure to toggle the job profiling status of the +driver by writing into the appropriate sysfs node:: + + echo > /sys/bus/platform/drivers/panthor/[a-f0-9]*.gpu/profiling + +Where `N` is a bit mask where cycle and timestamp sampling are respectively +enabled by the first and second bits. diff --git a/Documentation/gpu/todo.rst b/Documentation/gpu/todo.rst index 2b281e3c75a4..256d0d1cb216 100644 --- a/Documentation/gpu/todo.rst +++ b/Documentation/gpu/todo.rst @@ -834,6 +834,22 @@ Contact: Javier Martinez Canillas Level: Advanced +Querying errors from drm_syncobj +================================ + +The drm_syncobj container can be used by driver independent code to signal +complection of submission. + +One minor feature still missing is a generic DRM IOCTL to query the error +status of binary and timeline drm_syncobj. + +This should probably be improved by implementing the necessary kernel interface +and adding support for that in the userspace stack. + +Contact: Christian König + +Level: Starter + Outside DRM =========== diff --git a/Documentation/gpu/zynqmp.rst b/Documentation/gpu/zynqmp.rst new file mode 100644 index 000000000000..f57bfa0ad6ec --- /dev/null +++ b/Documentation/gpu/zynqmp.rst @@ -0,0 +1,149 @@ +.. SPDX-License-Identifier: GPL-2.0+ + +=============================================== +Xilinx ZynqMP Ultrascale+ DisplayPort Subsystem +=============================================== + +This subsystem handles DisplayPort video and audio output on the ZynqMP. It +supports in-memory framebuffers with the DisplayPort DMA controller +(xilinx-dpdma), as well as "live" video and audio from the programmable logic +(PL). This subsystem can perform several transformations, including color space +conversion, alpha blending, and audio mixing, although not all features are +currently supported. + +debugfs +------- + +To support debugging and compliance testing, several test modes can be enabled +though debugfs. The following files in /sys/kernel/debug/dri/X/DP-1/test/ +control the DisplayPort test modes: + +active: + Writing a 1 to this file will activate test mode, and writing a 0 will + deactivate test mode. Writing a 1 or 0 when the test mode is already + active/inactive will re-activate/re-deactivate test mode. When test + mode is inactive, changes made to other files will have no (immediate) + effect, although the settings will be saved for when test mode is + activated. When test mode is active, changes made to other files will + apply immediately. + +custom: + Custom test pattern value + +downspread: + Enable/disable clock downspreading (spread-spectrum clocking) by + writing 1/0 + +enhanced: + Enable/disable enhanced framing + +ignore_aux_errors: + Ignore AUX errors when set to 1. Writes to this file take effect + immediately (regardless of whether test mode is active) and affect all + AUX transfers. + +ignore_hpd: + Ignore hotplug events (such as cable removals or monitor link + retraining requests) when set to 1. Writes to this file take effect + immediately (regardless of whether test mode is active). + +laneX_preemphasis: + Preemphasis from 0 (lowest) to 2 (highest) for lane X + +laneX_swing: + Voltage swing from 0 (lowest) to 3 (highest) for lane X + +lanes: + Number of lanes to use (1, 2, or 4) + +pattern: + Test pattern. May be one of: + + video + Use regular video input + + symbol-error + Symbol error measurement pattern + + prbs7 + Output of the PRBS7 (x^7 + x^6 + 1) polynomial + + 80bit-custom + A custom 80-bit pattern + + cp2520 + HBR2 compliance eye pattern + + tps1 + Link training symbol pattern TPS1 (/D10.2/) + + tps2 + Link training symbol pattern TPS2 + + tps3 + Link training symbol pattern TPS3 (for HBR2) + +rate: + Rate in hertz. One of + + * 5400000000 (HBR2) + * 2700000000 (HBR) + * 1620000000 (RBR) + +You can dump the displayport test settings with the following command:: + + for prop in /sys/kernel/debug/dri/1/DP-1/test/*; do + printf '%-17s ' ${prop##*/} + if [ ${prop##*/} = custom ]; then + hexdump -C $prop | head -1 + else + cat $prop + fi + done + +The output could look something like:: + + active 1 + custom 00000000 00 00 00 00 00 00 00 00 00 00 |..........| + downspread 0 + enhanced 1 + ignore_aux_errors 1 + ignore_hpd 1 + lane0_preemphasis 0 + lane0_swing 3 + lane1_preemphasis 0 + lane1_swing 3 + lanes 2 + pattern prbs7 + rate 1620000000 + +The recommended test procedure is to connect the board to a monitor, +configure test mode, activate test mode, and then disconnect the cable +and connect it to your test equipment of choice. For example, one +sequence of commands could be:: + + echo 1 > /sys/kernel/debug/dri/1/DP-1/test/enhanced + echo tps1 > /sys/kernel/debug/dri/1/DP-1/test/pattern + echo 1620000000 > /sys/kernel/debug/dri/1/DP-1/test/rate + echo 1 > /sys/kernel/debug/dri/1/DP-1/test/ignore_aux_errors + echo 1 > /sys/kernel/debug/dri/1/DP-1/test/ignore_hpd + echo 1 > /sys/kernel/debug/dri/1/DP-1/test/active + +at which point the cable could be disconnected from the monitor. + +Internals +--------- + +.. kernel-doc:: drivers/gpu/drm/xlnx/zynqmp_disp.h + +.. kernel-doc:: drivers/gpu/drm/xlnx/zynqmp_dpsub.h + +.. kernel-doc:: drivers/gpu/drm/xlnx/zynqmp_kms.h + +.. kernel-doc:: drivers/gpu/drm/xlnx/zynqmp_disp.c + +.. kernel-doc:: drivers/gpu/drm/xlnx/zynqmp_dp.c + +.. kernel-doc:: drivers/gpu/drm/xlnx/zynqmp_dpsub.c + +.. kernel-doc:: drivers/gpu/drm/xlnx/zynqmp_kms.c diff --git a/Documentation/iio/ad7380.rst b/Documentation/iio/ad7380.rst index 9c784c1e652e..6f70b49b9ef2 100644 --- a/Documentation/iio/ad7380.rst +++ b/Documentation/iio/ad7380.rst @@ -41,13 +41,22 @@ supports only 1 SDO line. Reference voltage ----------------- -2 possible reference voltage sources are supported: +ad7380-4 +~~~~~~~~ + +ad7380-4 supports only an external reference voltage (2.5V to 3.3V). It must be +declared in the device tree as ``refin-supply``. + +All other devices from ad738x family +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +All other devices from ad738x support 2 possible reference voltage sources: - Internal reference (2.5V) - External reference (2.5V to 3.3V) The source is determined by the device tree. If ``refio-supply`` is present, -then the external reference is used, else the internal reference is used. +then it is used as external reference, else the internal reference is used. Oversampling and resolution boost --------------------------------- diff --git a/Documentation/mm/damon/maintainer-profile.rst b/Documentation/mm/damon/maintainer-profile.rst index 2365c9a3c1f0..ce3e98458339 100644 --- a/Documentation/mm/damon/maintainer-profile.rst +++ b/Documentation/mm/damon/maintainer-profile.rst @@ -7,26 +7,26 @@ The DAMON subsystem covers the files that are listed in 'DATA ACCESS MONITOR' section of 'MAINTAINERS' file. The mailing lists for the subsystem are damon@lists.linux.dev and -linux-mm@kvack.org. Patches should be made against the mm-unstable `tree -` whenever possible and posted to -the mailing lists. +linux-mm@kvack.org. Patches should be made against the `mm-unstable tree +`_ whenever possible and posted +to the mailing lists. SCM Trees --------- There are multiple Linux trees for DAMON development. Patches under development or testing are queued in `damon/next -` by the DAMON maintainer. +`_ by the DAMON maintainer. Sufficiently reviewed patches will be queued in `mm-unstable -` by the memory management +`_ by the memory management subsystem maintainer. After more sufficient tests, the patches will be queued -in `mm-stable ` , and finally +in `mm-stable `_, and finally pull-requested to the mainline by the memory management subsystem maintainer. -Note again the patches for mm-unstable `tree -` are queued by the memory +Note again the patches for `mm-unstable tree +`_ are queued by the memory management subsystem maintainer. If the patches requires some patches in -damon/next `tree ` which not yet merged +`damon/next tree `_ which not yet merged in mm-unstable, please make sure the requirement is clearly specified. Submit checklist addendum @@ -37,25 +37,25 @@ When making DAMON changes, you should do below. - Build changes related outputs including kernel and documents. - Ensure the builds introduce no new errors or warnings. - Run and ensure no new failures for DAMON `selftests - ` and + `_ and `kunittests - `. + `_. Further doing below and putting the results will be helpful. - Run `damon-tests/corr - ` for normal + `_ for normal changes. - Run `damon-tests/perf - ` for performance + `_ for performance changes. Key cycle dates --------------- Patches can be sent anytime. Key cycle dates of the `mm-unstable -` and `mm-stable -` trees depend on the memory +`_ and `mm-stable +`_ trees depend on the memory management subsystem maintainer. Review cadence @@ -72,13 +72,13 @@ Mailing tool Like many other Linux kernel subsystems, DAMON uses the mailing lists (damon@lists.linux.dev and linux-mm@kvack.org) as the major communication channel. There is a simple tool called `HacKerMaiL -` (``hkml``), which is for people who +`_ (``hkml``), which is for people who are not very familiar with the mailing lists based communication. The tool could be particularly helpful for DAMON community members since it is developed and maintained by DAMON maintainer. The tool is also officially announced to support DAMON and general Linux kernel development workflow. -In other words, `hkml ` is a mailing +In other words, `hkml `_ is a mailing tool for DAMON community, which DAMON maintainer is committed to support. Please feel free to try and report issues or feature requests for the tool to the maintainer. @@ -98,8 +98,8 @@ slots, and attendees should reserve one of those at least 24 hours before the time slot, by reaching out to the maintainer. Schedules and available reservation time slots are available at the Google `doc -`. +`_. There is also a public Google `calendar -` +`_ that has the events. Anyone can subscribe it. DAMON maintainer will also provide periodic reminder to the mailing list (damon@lists.linux.dev). diff --git a/Documentation/networking/napi.rst b/Documentation/networking/napi.rst index 7bf7b95c4f7a..dfa5d549be9c 100644 --- a/Documentation/networking/napi.rst +++ b/Documentation/networking/napi.rst @@ -144,9 +144,8 @@ IRQ should only be unmasked after a successful call to napi_complete_done(): napi_schedule_irqoff() is a variant of napi_schedule() which takes advantage of guarantees given by being invoked in IRQ context (no need to -mask interrupts). Note that PREEMPT_RT forces all interrupts -to be threaded so the interrupt may need to be marked ``IRQF_NO_THREAD`` -to avoid issues on real-time kernel configurations. +mask interrupts). napi_schedule_irqoff() will fall back to napi_schedule() if +IRQs are threaded (such as if ``PREEMPT_RT`` is enabled). Instance to queue mapping ------------------------- diff --git a/Documentation/networking/packet_mmap.rst b/Documentation/networking/packet_mmap.rst index dca15d15feaf..02370786e77b 100644 --- a/Documentation/networking/packet_mmap.rst +++ b/Documentation/networking/packet_mmap.rst @@ -16,7 +16,7 @@ ii) transmit network traffic, or any other that needs raw Howto can be found at: - https://sites.google.com/site/packetmmap/ + https://web.archive.org/web/20220404160947/https://sites.google.com/site/packetmmap/ Please send your comments to - Ulisses Alonso Camaró @@ -166,7 +166,8 @@ As capture, each frame contains two parts:: /* bind socket to eth0 */ bind(this->socket, (struct sockaddr *)&my_addr, sizeof(struct sockaddr_ll)); - A complete tutorial is available at: https://sites.google.com/site/packetmmap/ + A complete tutorial is available at: + https://web.archive.org/web/20220404160947/https://sites.google.com/site/packetmmap/ By default, the user should put data at:: diff --git a/Documentation/networking/tcp_ao.rst b/Documentation/networking/tcp_ao.rst index e96e62d1dab3..d5b6d0df63c3 100644 --- a/Documentation/networking/tcp_ao.rst +++ b/Documentation/networking/tcp_ao.rst @@ -9,7 +9,7 @@ segments between trusted peers. It adds a new TCP header option with a Message Authentication Code (MAC). MACs are produced from the content of a TCP segment using a hashing function with a password known to both peers. The intent of TCP-AO is to deprecate TCP-MD5 providing better security, -key rotation and support for variety of hashing algorithms. +key rotation and support for a variety of hashing algorithms. 1. Introduction =============== @@ -164,9 +164,9 @@ A: It should not, no action needs to be performed [7.5.2.e]:: is not available, no action is required (RNextKeyID of a received segment needs to match the MKT’s SendID). -Q: How current_key is set and when does it change? It is a user-triggered -change, or is it by a request from the remote peer? Is it set by the user -explicitly, or by a matching rule? +Q: How is current_key set, and when does it change? Is it a user-triggered +change, or is it triggered by a request from the remote peer? Is it set by the +user explicitly, or by a matching rule? A: current_key is set by RNextKeyID [6.1]:: @@ -233,8 +233,8 @@ always have one current_key [3.3]:: Q: Can a non-TCP-AO connection become a TCP-AO-enabled one? -A: No: for already established non-TCP-AO connection it would be impossible -to switch using TCP-AO as the traffic key generation requires the initial +A: No: for an already established non-TCP-AO connection it would be impossible +to switch to using TCP-AO, as the traffic key generation requires the initial sequence numbers. Paraphrasing, starting using TCP-AO would require re-establishing the TCP connection. @@ -292,7 +292,7 @@ no transparency is really needed and modern BGP daemons already have Linux provides a set of ``setsockopt()s`` and ``getsockopt()s`` that let userspace manage TCP-AO on a per-socket basis. In order to add/delete MKTs -``TCP_AO_ADD_KEY`` and ``TCP_AO_DEL_KEY`` TCP socket options must be used +``TCP_AO_ADD_KEY`` and ``TCP_AO_DEL_KEY`` TCP socket options must be used. It is not allowed to add a key on an established non-TCP-AO connection as well as to remove the last key from TCP-AO connection. @@ -361,7 +361,7 @@ not implemented. 4. ``setsockopt()`` vs ``accept()`` race ======================================== -In contrast with TCP-MD5 established connection which has just one key, +In contrast with an established TCP-MD5 connection which has just one key, TCP-AO connections may have many keys, which means that accepted connections on a listen socket may have any amount of keys as well. As copying all those keys on a first properly signed SYN would make the request socket bigger, that @@ -374,7 +374,7 @@ keys from sockets that were already established, but not yet ``accept()``'ed, hanging in the accept queue. The reverse is valid as well: if userspace adds a new key for a peer on -a listener socket, the established sockets in accept queue won't +a listener socket, the established sockets in the accept queue won't have the new keys. At this moment, the resolution for the two races: @@ -382,7 +382,7 @@ At this moment, the resolution for the two races: and ``setsockopt(TCP_AO_DEL_KEY)`` vs ``accept()`` is delegated to userspace. This means that it's expected that userspace would check the MKTs on the socket that was returned by ``accept()`` to verify that any key rotation that -happened on listen socket is reflected on the newly established connection. +happened on the listen socket is reflected on the newly established connection. This is a similar "do-nothing" approach to TCP-MD5 from the kernel side and may be changed later by introducing new flags to ``tcp_ao_add`` diff --git a/Documentation/process/maintainer-netdev.rst b/Documentation/process/maintainer-netdev.rst index c9edf9e7362d..1ae71e31591c 100644 --- a/Documentation/process/maintainer-netdev.rst +++ b/Documentation/process/maintainer-netdev.rst @@ -355,6 +355,8 @@ just do it. As a result, a sequence of smaller series gets merged quicker and with better review coverage. Re-posting large series also increases the mailing list traffic. +.. _rcs: + Local variable ordering ("reverse xmas tree", "RCS") ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ @@ -391,6 +393,21 @@ APIs and helpers, especially scoped iterators. However, direct use of ``__free()`` within networking core and drivers is discouraged. Similar guidance applies to declaring variables mid-function. +Clean-up patches +~~~~~~~~~~~~~~~~ + +Netdev discourages patches which perform simple clean-ups, which are not in +the context of other work. For example: + +* Addressing ``checkpatch.pl`` warnings +* Addressing :ref:`Local variable ordering` issues +* Conversions to device-managed APIs (``devm_`` helpers) + +This is because it is felt that the churn that such changes produce comes +at a greater cost than the value of such clean-ups. + +Conversely, spelling and grammar fixes are not discouraged. + Resending after review ~~~~~~~~~~~~~~~~~~~~~~ diff --git a/Documentation/process/maintainer-soc.rst b/Documentation/process/maintainer-soc.rst index 12637530d68f..fe9d8bcfbd2b 100644 --- a/Documentation/process/maintainer-soc.rst +++ b/Documentation/process/maintainer-soc.rst @@ -30,10 +30,13 @@ tree as a dedicated branch covering multiple subsystems. The main SoC tree is housed on git.kernel.org: https://git.kernel.org/pub/scm/linux/kernel/git/soc/soc.git/ +Maintainers +----------- + Clearly this is quite a wide range of topics, which no one person, or even small group of people are capable of maintaining. Instead, the SoC subsystem -is comprised of many submaintainers, each taking care of individual platforms -and driver subdirectories. +is comprised of many submaintainers (platform maintainers), each taking care of +individual platforms and driver subdirectories. In this regard, "platform" usually refers to a series of SoCs from a given vendor, for example, Nvidia's series of Tegra SoCs. Many submaintainers operate on a vendor level, responsible for multiple product lines. For several reasons, @@ -43,14 +46,43 @@ MAINTAINERS file. Most of these submaintainers have their own trees where they stage patches, sending pull requests to the main SoC tree. These trees are usually, but not -always, listed in MAINTAINERS. The main SoC maintainers can be reached via the -alias soc@kernel.org if there is no platform-specific maintainer, or if they -are unresponsive. +always, listed in MAINTAINERS. What the SoC tree is not, however, is a location for architecture-specific code changes. Each architecture has its own maintainers that are responsible for architectural details, CPU errata and the like. +Submitting Patches for Given SoC +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +All typical platform related patches should be sent via SoC submaintainers +(platform-specific maintainers). This includes also changes to per-platform or +shared defconfigs (scripts/get_maintainer.pl might not provide correct +addresses in such case). + +Submitting Patches to the Main SoC Maintainers +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +The main SoC maintainers can be reached via the alias soc@kernel.org only in +following cases: + +1. There are no platform-specific maintainers. + +2. Platform-specific maintainers are unresponsive. + +3. Introducing a completely new SoC platform. Such new SoC work should be sent + first to common mailing lists, pointed out by scripts/get_maintainer.pl, for + community review. After positive community review, work should be sent to + soc@kernel.org in one patchset containing new arch/foo/Kconfig entry, DTS + files, MAINTAINERS file entry and optionally initial drivers with their + Devicetree bindings. The MAINTAINERS file entry should list new + platform-specific maintainers, who are going to be responsible for handling + patches for the platform from now on. + +Note that the soc@kernel.org is usually not the place to discuss the patches, +thus work sent to this address should be already considered as acceptable by +the community. + Information for (new) Submaintainers ------------------------------------ diff --git a/Documentation/rust/arch-support.rst b/Documentation/rust/arch-support.rst index 750ff371570a..54be7ddf3e57 100644 --- a/Documentation/rust/arch-support.rst +++ b/Documentation/rust/arch-support.rst @@ -17,7 +17,7 @@ Architecture Level of support Constraints ============= ================ ============================================== ``arm64`` Maintained Little Endian only. ``loongarch`` Maintained \- -``riscv`` Maintained ``riscv64`` only. +``riscv`` Maintained ``riscv64`` and LLVM/Clang only. ``um`` Maintained \- ``x86`` Maintained ``x86_64`` only. ============= ================ ============================================== diff --git a/Documentation/scheduler/sched-ext.rst b/Documentation/scheduler/sched-ext.rst index 6c0d70e2e27d..7b59bbd2e564 100644 --- a/Documentation/scheduler/sched-ext.rst +++ b/Documentation/scheduler/sched-ext.rst @@ -66,7 +66,7 @@ BPF scheduler and reverts all tasks back to CFS. .. code-block:: none # make -j16 -C tools/sched_ext - # tools/sched_ext/scx_simple + # tools/sched_ext/build/bin/scx_simple local=0 global=3 local=5 global=24 local=9 global=44 diff --git a/Documentation/translations/zh_CN/core-api/unaligned-memory-access.rst b/Documentation/translations/zh_CN/core-api/unaligned-memory-access.rst index 29c33e7e0855..fbe0989a8ce5 100644 --- a/Documentation/translations/zh_CN/core-api/unaligned-memory-access.rst +++ b/Documentation/translations/zh_CN/core-api/unaligned-memory-access.rst @@ -175,7 +175,7 @@ field2会导致非对齐访问，这并不是不合理的。你会期望field2 避免非对齐访问 ============== -避免非对齐访问的最简单方法是使用头文件提供的get_unaligned()和 +避免非对齐访问的最简单方法是使用头文件提供的get_unaligned()和 put_unaligned()宏。 回到前面的一个可能导致非对齐访问的代码例子:: diff --git a/Documentation/userspace-api/mseal.rst b/Documentation/userspace-api/mseal.rst index 4132eec995a3..41102f74c5e2 100644 --- a/Documentation/userspace-api/mseal.rst +++ b/Documentation/userspace-api/mseal.rst @@ -23,177 +23,166 @@ applications can additionally seal security critical data at runtime. A similar feature already exists in the XNU kernel with the VM_FLAGS_PERMANENT flag [1] and on OpenBSD with the mimmutable syscall [2]. -User API -======== -mseal() ------------ -The mseal() syscall has the following signature: +SYSCALL +======= +mseal syscall signature +----------------------- + ``int mseal(void \* addr, size_t len, unsigned long flags)`` -``int mseal(void addr, size_t len, unsigned long flags)`` + **addr**/**len**: virtual memory address range. + The address range set by **addr**/**len** must meet: + - The start address must be in an allocated VMA. + - The start address must be page aligned. + - The end address (**addr** + **len**) must be in an allocated VMA. + - no gap (unallocated memory) between start and end address. -**addr/len**: virtual memory address range. + The ``len`` will be paged aligned implicitly by the kernel. -The address range set by ``addr``/``len`` must meet: - - The start address must be in an allocated VMA. - - The start address must be page aligned. - - The end address (``addr`` + ``len``) must be in an allocated VMA. - - no gap (unallocated memory) between start and end address. + **flags**: reserved for future use. -The ``len`` will be paged aligned implicitly by the kernel. + **Return values**: + - **0**: Success. + - **-EINVAL**: + * Invalid input ``flags``. + * The start address (``addr``) is not page aligned. + * Address range (``addr`` + ``len``) overflow. + - **-ENOMEM**: + * The start address (``addr``) is not allocated. + * The end address (``addr`` + ``len``) is not allocated. + * A gap (unallocated memory) between start and end address. + - **-EPERM**: + * sealing is supported only on 64-bit CPUs, 32-bit is not supported. -**flags**: reserved for future use. + **Note about error return**: + - For above error cases, users can expect the given memory range is + unmodified, i.e. no partial update. + - There might be other internal errors/cases not listed here, e.g. + error during merging/splitting VMAs, or the process reaching the maximum + number of supported VMAs. In those cases, partial updates to the given + memory range could happen. However, those cases should be rare. -**return values**: + **Architecture support**: + mseal only works on 64-bit CPUs, not 32-bit CPUs. -- ``0``: Success. + **Idempotent**: + users can call mseal multiple times. mseal on an already sealed memory + is a no-action (not error). -- ``-EINVAL``: - - Invalid input ``flags``. - - The start address (``addr``) is not page aligned. - - Address range (``addr`` + ``len``) overflow. + **no munseal** + Once mapping is sealed, it can't be unsealed. The kernel should never + have munseal, this is consistent with other sealing feature, e.g. + F_SEAL_SEAL for file. -- ``-ENOMEM``: - - The start address (``addr``) is not allocated. - - The end address (``addr`` + ``len``) is not allocated. - - A gap (unallocated memory) between start and end address. +Blocked mm syscall for sealed mapping +------------------------------------- + It might be important to note: **once the mapping is sealed, it will + stay in the process's memory until the process terminates**. -- ``-EPERM``: - - sealing is supported only on 64-bit CPUs, 32-bit is not supported. + Example:: -- For above error cases, users can expect the given memory range is - unmodified, i.e. no partial update. + *ptr = mmap(0, 4096, PROT_READ, MAP_ANONYMOUS | MAP_PRIVATE, 0, 0); + rc = mseal(ptr, 4096, 0); + /* munmap will fail */ + rc = munmap(ptr, 4096); + assert(rc < 0); -- There might be other internal errors/cases not listed here, e.g. - error during merging/splitting VMAs, or the process reaching the max - number of supported VMAs. In those cases, partial updates to the given - memory range could happen. However, those cases should be rare. + Blocked mm syscall: + - munmap + - mmap + - mremap + - mprotect and pkey_mprotect + - some destructive madvise behaviors: MADV_DONTNEED, MADV_FREE, + MADV_DONTNEED_LOCKED, MADV_FREE, MADV_DONTFORK, MADV_WIPEONFORK -**Blocked operations after sealing**: - Unmapping, moving to another location, and shrinking the size, - via munmap() and mremap(), can leave an empty space, therefore - can be replaced with a VMA with a new set of attributes. + The first set of syscalls to block is munmap, mremap, mmap. They can + either leave an empty space in the address space, therefore allowing + replacement with a new mapping with new set of attributes, or can + overwrite the existing mapping with another mapping. - Moving or expanding a different VMA into the current location, - via mremap(). + mprotect and pkey_mprotect are blocked because they changes the + protection bits (RWX) of the mapping. - Modifying a VMA via mmap(MAP_FIXED). + Certain destructive madvise behaviors, specifically MADV_DONTNEED, + MADV_FREE, MADV_DONTNEED_LOCKED, and MADV_WIPEONFORK, can introduce + risks when applied to anonymous memory by threads lacking write + permissions. Consequently, these operations are prohibited under such + conditions. The aforementioned behaviors have the potential to modify + region contents by discarding pages, effectively performing a memset(0) + operation on the anonymous memory. - Size expansion, via mremap(), does not appear to pose any - specific risks to sealed VMAs. It is included anyway because - the use case is unclear. In any case, users can rely on - merging to expand a sealed VMA. + Kernel will return -EPERM for blocked syscalls. - mprotect() and pkey_mprotect(). + When blocked syscall return -EPERM due to sealing, the memory regions may + or may not be changed, depends on the syscall being blocked: - Some destructive madvice() behaviors (e.g. MADV_DONTNEED) - for anonymous memory, when users don't have write permission to the - memory. Those behaviors can alter region contents by discarding pages, - effectively a memset(0) for anonymous memory. + - munmap: munmap is atomic. If one of VMAs in the given range is + sealed, none of VMAs are updated. + - mprotect, pkey_mprotect, madvise: partial update might happen, e.g. + when mprotect over multiple VMAs, mprotect might update the beginning + VMAs before reaching the sealed VMA and return -EPERM. + - mmap and mremap: undefined behavior. - Kernel will return -EPERM for blocked operations. - - For blocked operations, one can expect the given address is unmodified, - i.e. no partial update. Note, this is different from existing mm - system call behaviors, where partial updates are made till an error is - found and returned to userspace. To give an example: - - Assume following code sequence: - - - ptr = mmap(null, 8192, PROT_NONE); - - munmap(ptr + 4096, 4096); - - ret1 = mprotect(ptr, 8192, PROT_READ); - - mseal(ptr, 4096); - - ret2 = mprotect(ptr, 8192, PROT_NONE); - - ret1 will be -ENOMEM, the page from ptr is updated to PROT_READ. - - ret2 will be -EPERM, the page remains to be PROT_READ. - -**Note**: - -- mseal() only works on 64-bit CPUs, not 32-bit CPU. - -- users can call mseal() multiple times, mseal() on an already sealed memory - is a no-action (not error). - -- munseal() is not supported. - -Use cases: -========== +Use cases +========= - glibc: The dynamic linker, during loading ELF executables, can apply sealing to - non-writable memory segments. + mapping segments. -- Chrome browser: protect some security sensitive data-structures. +- Chrome browser: protect some security sensitive data structures. -Notes on which memory to seal: -============================== - -It might be important to note that sealing changes the lifetime of a mapping, -i.e. the sealed mapping won’t be unmapped till the process terminates or the -exec system call is invoked. Applications can apply sealing to any virtual -memory region from userspace, but it is crucial to thoroughly analyze the -mapping's lifetime prior to apply the sealing. +When not to use mseal +===================== +Applications can apply sealing to any virtual memory region from userspace, +but it is *crucial to thoroughly analyze the mapping's lifetime* prior to +apply the sealing. This is because the sealed mapping *won’t be unmapped* +until the process terminates or the exec system call is invoked. For example: + - aio/shm + aio/shm can call mmap and munmap on behalf of userspace, e.g. + ksys_shmdt() in shm.c. The lifetimes of those mapping are not tied to + the lifetime of the process. If those memories are sealed from userspace, + then munmap will fail, causing leaks in VMA address space during the + lifetime of the process. -- aio/shm + - ptr allocated by malloc (heap) + Don't use mseal on the memory ptr return from malloc(). + malloc() is implemented by allocator, e.g. by glibc. Heap manager might + allocate a ptr from brk or mapping created by mmap. + If an app calls mseal on a ptr returned from malloc(), this can affect + the heap manager's ability to manage the mappings; the outcome is + non-deterministic. - aio/shm can call mmap()/munmap() on behalf of userspace, e.g. ksys_shmdt() in - shm.c. The lifetime of those mapping are not tied to the lifetime of the - process. If those memories are sealed from userspace, then munmap() will fail, - causing leaks in VMA address space during the lifetime of the process. + Example:: -- Brk (heap) + ptr = malloc(size); + /* don't call mseal on ptr return from malloc. */ + mseal(ptr, size); + /* free will success, allocator can't shrink heap lower than ptr */ + free(ptr); - Currently, userspace applications can seal parts of the heap by calling - malloc() and mseal(). - let's assume following calls from user space: +mseal doesn't block +=================== +In a nutshell, mseal blocks certain mm syscall from modifying some of VMA's +attributes, such as protection bits (RWX). Sealed mappings doesn't mean the +memory is immutable. - - ptr = malloc(size); - - mprotect(ptr, size, RO); - - mseal(ptr, size); - - free(ptr); - - Technically, before mseal() is added, the user can change the protection of - the heap by calling mprotect(RO). As long as the user changes the protection - back to RW before free(), the memory range can be reused. - - Adding mseal() into the picture, however, the heap is then sealed partially, - the user can still free it, but the memory remains to be RO. If the address - is re-used by the heap manager for another malloc, the process might crash - soon after. Therefore, it is important not to apply sealing to any memory - that might get recycled. - - Furthermore, even if the application never calls the free() for the ptr, - the heap manager may invoke the brk system call to shrink the size of the - heap. In the kernel, the brk-shrink will call munmap(). Consequently, - depending on the location of the ptr, the outcome of brk-shrink is - nondeterministic. - - -Additional notes: -================= As Jann Horn pointed out in [3], there are still a few ways to write -to RO memory, which is, in a way, by design. Those cases are not covered -by mseal(). If applications want to block such cases, sandbox tools (such as -seccomp, LSM, etc) might be considered. +to RO memory, which is, in a way, by design. And those could be blocked +by different security measures. Those cases are: -- Write to read-only memory through /proc/self/mem interface. -- Write to read-only memory through ptrace (such as PTRACE_POKETEXT). -- userfaultfd. + - Write to read-only memory through /proc/self/mem interface (FOLL_FORCE). + - Write to read-only memory through ptrace (such as PTRACE_POKETEXT). + - userfaultfd. The idea that inspired this patch comes from Stephen Röttger’s work in V8 CFI [4]. Chrome browser in ChromeOS will be the first user of this API. -Reference: -========== -[1] https://github.com/apple-oss-distributions/xnu/blob/1031c584a5e37aff177559b9f69dbd3c8c3fd30a/osfmk/mach/vm_statistics.h#L274 - -[2] https://man.openbsd.org/mimmutable.2 - -[3] https://lore.kernel.org/lkml/CAG48ez3ShUYey+ZAFsU2i1RpQn0a5eOs2hzQ426FkcgnfUGLvA@mail.gmail.com - -[4] https://docs.google.com/document/d/1O2jwK4dxI3nRcOJuPYkonhTkNQfbmwdvxQMyXgeaRHo/edit#heading=h.bvaojj9fu6hc +Reference +========= +- [1] https://github.com/apple-oss-distributions/xnu/blob/1031c584a5e37aff177559b9f69dbd3c8c3fd30a/osfmk/mach/vm_statistics.h#L274 +- [2] https://man.openbsd.org/mimmutable.2 +- [3] https://lore.kernel.org/lkml/CAG48ez3ShUYey+ZAFsU2i1RpQn0a5eOs2hzQ426FkcgnfUGLvA@mail.gmail.com +- [4] https://docs.google.com/document/d/1O2jwK4dxI3nRcOJuPYkonhTkNQfbmwdvxQMyXgeaRHo/edit#heading=h.bvaojj9fu6hc diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst index e32471977d0a..edc070c6e19b 100644 --- a/Documentation/virt/kvm/api.rst +++ b/Documentation/virt/kvm/api.rst @@ -8098,13 +8098,15 @@ KVM_X86_QUIRK_MWAIT_NEVER_UD_FAULTS By default, KVM emulates MONITOR/MWAIT (if KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT is disabled. -KVM_X86_QUIRK_SLOT_ZAP_ALL By default, KVM invalidates all SPTEs in - fast way for memslot deletion when VM type - is KVM_X86_DEFAULT_VM. - When this quirk is disabled or when VM type - is other than KVM_X86_DEFAULT_VM, KVM zaps - only leaf SPTEs that are within the range of - the memslot being deleted. +KVM_X86_QUIRK_SLOT_ZAP_ALL By default, for KVM_X86_DEFAULT_VM VMs, KVM + invalidates all SPTEs in all memslots and + address spaces when a memslot is deleted or + moved. When this quirk is disabled (or the + VM type isn't KVM_X86_DEFAULT_VM), KVM only + ensures the backing memory of the deleted + or moved memslot isn't reachable, i.e KVM + _may_ invalidate only SPTEs related to the + memslot. =================================== ============================================ 7.32 KVM_CAP_MAX_VCPU_ID diff --git a/Documentation/virt/kvm/locking.rst b/Documentation/virt/kvm/locking.rst index 20a9a37d1cdd..1bedd56e2fe3 100644 --- a/Documentation/virt/kvm/locking.rst +++ b/Documentation/virt/kvm/locking.rst @@ -136,7 +136,7 @@ For direct sp, we can easily avoid it since the spte of direct sp is fixed to gfn. For indirect sp, we disabled fast page fault for simplicity. A solution for indirect sp could be to pin the gfn, for example via -kvm_vcpu_gfn_to_pfn_atomic, before the cmpxchg. After the pinning: +gfn_to_pfn_memslot_atomic, before the cmpxchg. After the pinning: - We have held the refcount of pfn; that means the pfn can not be freed and be reused for another gfn. diff --git a/Documentation/wmi/devices/dell-wmi-ddv.rst b/Documentation/wmi/devices/dell-wmi-ddv.rst index 2fcdfcf03327..e0c20af30948 100644 --- a/Documentation/wmi/devices/dell-wmi-ddv.rst +++ b/Documentation/wmi/devices/dell-wmi-ddv.rst @@ -8,7 +8,7 @@ Introduction ============ Many Dell notebooks made after ~2020 support a WMI-based interface for -retrieving various system data like battery temperature, ePPID, diagostic data +retrieving various system data like battery temperature, ePPID, diagnostic data and fan/thermal sensor data. This interface is likely used by the `Dell Data Vault` software on Windows, @@ -277,7 +277,7 @@ Reverse-Engineering the DDV WMI interface 4. Try to deduce the meaning of a certain WMI method by comparing the control flow with other ACPI methods (_BIX or _BIF for battery related methods for example). -5. Use the built-in UEFI diagostics to view sensor types/values for fan/thermal +5. Use the built-in UEFI diagnostics to view sensor types/values for fan/thermal related methods (sometimes overwriting static ACPI data fields can be used to test different sensor type values, since on some machines this data is not reinitialized upon a warm reset). diff --git a/MAINTAINERS b/MAINTAINERS index c27f3190737f..3779b3cc138e 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -258,12 +258,6 @@ L: linux-acenic@sunsite.dk S: Maintained F: drivers/net/ethernet/alteon/acenic* -ACER ASPIRE 1 EMBEDDED CONTROLLER DRIVER -M: Nikita Travkin -S: Maintained -F: Documentation/devicetree/bindings/platform/acer,aspire1-ec.yaml -F: drivers/platform/arm64/acer-aspire1-ec.c - ACER ASPIRE ONE TEMPERATURE AND FAN DRIVER M: Peter Kaestle L: platform-driver-x86@vger.kernel.org @@ -860,7 +854,7 @@ F: drivers/crypto/allwinner/ ALLWINNER DMIC DRIVERS M: Ban Tao -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/allwinner,sun50i-h6-dmic.yaml F: sound/soc/sunxi/sun50i-dmic.c @@ -888,7 +882,6 @@ F: drivers/staging/media/sunxi/cedrus/ ALPHA PORT M: Richard Henderson -M: Ivan Kokshaysky M: Matt Turner L: linux-alpha@vger.kernel.org S: Odd Fixes @@ -1517,7 +1510,7 @@ F: drivers/iio/gyro/adxrs290.c ANALOG DEVICES INC ASOC CODEC DRIVERS M: Lars-Peter Clausen M: Nuno Sá -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Supported W: http://wiki.analog.com/ W: https://ez.analog.com/linux-software-drivers @@ -1594,7 +1587,7 @@ F: drivers/rtc/rtc-goldfish.c AOA (Apple Onboard Audio) ALSA DRIVER M: Johannes Berg L: linuxppc-dev@lists.ozlabs.org -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: sound/aoa/ @@ -1761,8 +1754,8 @@ F: include/uapi/linux/if_arcnet.h ARM AND ARM64 SoC SUB-ARCHITECTURES (COMMON PARTS) M: Arnd Bergmann M: Olof Johansson -M: soc@kernel.org L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) +L: soc@lists.linux.dev S: Maintained P: Documentation/process/maintainer-soc.rst C: irc://irc.libera.chat/armlinux @@ -2091,7 +2084,7 @@ F: drivers/crypto/amlogic/ ARM/Amlogic Meson SoC Sound Drivers M: Jerome Brunet -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/amlogic* F: sound/soc/meson/ @@ -2129,7 +2122,7 @@ F: drivers/*/*alpine* ARM/APPLE MACHINE SOUND DRIVERS M: Martin Povišer L: asahi@lists.linux.dev -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/adi,ssm3515.yaml F: Documentation/devicetree/bindings/sound/apple,* @@ -2263,12 +2256,6 @@ L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) S: Maintained F: arch/arm/mach-ep93xx/ts72xx.c -ARM/CIRRUS LOGIC CLPS711X ARM ARCHITECTURE -M: Alexander Shiyan -L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) -S: Odd Fixes -N: clps711x - ARM/CIRRUS LOGIC EP93XX ARM ARCHITECTURE M: Hartley Sweeten M: Alexander Sverdlin @@ -3732,7 +3719,7 @@ F: arch/arm/boot/dts/microchip/at91-tse850-3.dts AXENTIA ASOC DRIVERS M: Peter Rosin -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/axentia,* F: sound/soc/atmel/tse850-pcm5142.c @@ -3815,14 +3802,6 @@ F: drivers/video/backlight/ F: include/linux/backlight.h F: include/linux/pwm_backlight.h -BAIKAL-T1 PVT HARDWARE MONITOR DRIVER -M: Serge Semin -L: linux-hwmon@vger.kernel.org -S: Supported -F: Documentation/devicetree/bindings/hwmon/baikal,bt1-pvt.yaml -F: Documentation/hwmon/bt1-pvt.rst -F: drivers/hwmon/bt1-pvt.[ch] - BARCO P50 GPIO DRIVER M: Santosh Kumar Yadav M: Peter Korsgaard @@ -4851,7 +4830,7 @@ F: include/uapi/linux/bsg.h BT87X AUDIO DRIVER M: Clemens Ladisch -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git F: Documentation/sound/cards/bt87x.rst @@ -4913,7 +4892,7 @@ F: drivers/net/can/bxcan.c C-MEDIA CMI8788 DRIVER M: Clemens Ladisch -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git F: sound/pci/oxygen/ @@ -6476,7 +6455,6 @@ F: drivers/mtd/nand/raw/denali* DESIGNWARE EDMA CORE IP DRIVER M: Manivannan Sadhasivam -R: Serge Semin L: dmaengine@vger.kernel.org S: Maintained F: drivers/dma/dw-edma/ @@ -7097,12 +7075,10 @@ M: Javier Martinez Canillas L: dri-devel@lists.freedesktop.org S: Maintained T: git https://gitlab.freedesktop.org/drm/misc/kernel.git -F: drivers/gpu/drm/drm_aperture.c F: drivers/gpu/drm/tiny/ofdrm.c F: drivers/gpu/drm/tiny/simpledrm.c F: drivers/video/aperture.c F: drivers/video/nomodeset.c -F: include/drm/drm_aperture.h F: include/linux/aperture.h F: include/video/nomodeset.h @@ -7383,6 +7359,18 @@ S: Maintained F: Documentation/devicetree/bindings/display/panel/samsung,s6d7aa0.yaml F: drivers/gpu/drm/panel/panel-samsung-s6d7aa0.c +DRM DRIVER FOR SAMSUNG S6E3HA8 PANELS +M: Dzmitry Sankouski +S: Maintained +F: Documentation/devicetree/bindings/display/panel/samsung,s6e3ha8.yaml +F: drivers/gpu/drm/panel/panel-samsung-s6e3ha8.c + +DRM DRIVER FOR SHARP MEMORY LCD +M: Alex Lanzano +S: Maintained +F: Documentation/devicetree/bindings/display/sharp,ls010b7dh04.yaml +F: drivers/gpu/drm/tiny/sharp-memory.c + DRM DRIVER FOR SITRONIX ST7586 PANELS M: David Lechner S: Maintained @@ -7460,8 +7448,8 @@ T: git https://gitlab.freedesktop.org/drm/misc/kernel.git F: drivers/gpu/drm/udl/ DRM DRIVER FOR VIRTUAL KERNEL MODESETTING (VKMS) -M: Rodrigo Siqueira M: Maíra Canal +M: Louis Chauvet R: Haneen Mohammed R: Simona Vetter R: Melissa Wen @@ -7793,6 +7781,7 @@ F: include/uapi/drm/v3d_drm.h DRM DRIVERS FOR VC4 M: Maxime Ripard M: Dave Stevenson +R: Maíra Canal R: Raspberry Pi Kernel Maintenance S: Supported T: git https://gitlab.freedesktop.org/drm/misc/kernel.git @@ -7827,11 +7816,14 @@ L: dri-devel@lists.freedesktop.org S: Maintained T: git https://gitlab.freedesktop.org/drm/misc/kernel.git F: Documentation/devicetree/bindings/display/xlnx/ +F: Documentation/gpu/zynqmp.rst F: drivers/gpu/drm/xlnx/ DRM GPU SCHEDULER M: Luben Tuikov M: Matthew Brost +M: Danilo Krummrich +M: Philipp Stanner L: dri-devel@lists.freedesktop.org S: Maintained T: git https://gitlab.freedesktop.org/drm/misc/kernel.git @@ -8252,7 +8244,7 @@ F: drivers/edac/ti_edac.c EDIROL UA-101/UA-1000 DRIVER M: Clemens Ladisch -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git F: sound/usb/misc/ua101.c @@ -8814,7 +8806,7 @@ F: drivers/net/can/usb/f81604.c FIREWIRE AUDIO DRIVERS and IEC 61883-1/6 PACKET STREAMING ENGINE M: Clemens Ladisch M: Takashi Sakamoto -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git F: include/uapi/sound/firewire.h @@ -8888,7 +8880,7 @@ F: drivers/input/joystick/fsia6b.c FOCUSRITE SCARLETT2 MIXER DRIVER (Scarlett Gen 2+ and Clarett) M: Geoffrey D. Bennett -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained W: https://github.com/geoffreybennett/scarlett-gen2 B: https://github.com/geoffreybennett/scarlett-gen2/issues @@ -8912,6 +8904,7 @@ F: include/linux/fortify-string.h F: lib/fortify_kunit.c F: lib/memcpy_kunit.c F: lib/test_fortify/* +K: \bunsafe_memcpy\b K: \b__NO_FORTIFY\b FPGA DFL DRIVERS @@ -9209,7 +9202,7 @@ M: Shengjiu Wang M: Xiubo Li R: Fabio Estevam R: Nicolin Chen -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org L: linuxppc-dev@lists.ozlabs.org S: Maintained F: sound/soc/fsl/fsl* @@ -9219,7 +9212,7 @@ FREESCALE SOC LPC32XX SOUND DRIVERS M: J.M.B. Downing M: Piotr Wojtaszczyk R: Vladimir Zapolskiy -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org L: linuxppc-dev@lists.ozlabs.org S: Maintained F: Documentation/devicetree/bindings/sound/nxp,lpc3220-i2s.yaml @@ -9227,7 +9220,7 @@ F: sound/soc/fsl/lpc3xxx-* FREESCALE SOC SOUND QMC DRIVER M: Herve Codina -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org L: linuxppc-dev@lists.ozlabs.org S: Maintained F: Documentation/devicetree/bindings/sound/fsl,qmc-audio.yaml @@ -9742,6 +9735,7 @@ F: include/dt-bindings/gpio/ F: include/linux/gpio.h F: include/linux/gpio/ F: include/linux/of_gpio.h +K: (devm_)?gpio_(request|free|direction|get|set) GPIO UAPI M: Bartosz Golaszewski @@ -9756,14 +9750,6 @@ F: drivers/gpio/gpiolib-cdev.c F: include/uapi/linux/gpio.h F: tools/gpio/ -GRE DEMULTIPLEXER DRIVER -M: Dmitry Kozlov -L: netdev@vger.kernel.org -S: Maintained -F: include/net/gre.h -F: net/ipv4/gre_demux.c -F: net/ipv4/gre_offload.c - GRETH 10/100/1G Ethernet MAC device driver M: Andreas Larsson L: netdev@vger.kernel.org @@ -10267,7 +10253,7 @@ F: Documentation/devicetree/bindings/arm/hisilicon/low-pin-count.yaml F: drivers/bus/hisi_lpc.c HISILICON NETWORK SUBSYSTEM 3 DRIVER (HNS3) -M: Yisen Zhuang +M: Jian Shen M: Salil Mehta M: Jijie Shao L: netdev@vger.kernel.org @@ -10276,7 +10262,7 @@ W: http://www.hisilicon.com F: drivers/net/ethernet/hisilicon/hns3/ HISILICON NETWORK SUBSYSTEM DRIVER -M: Yisen Zhuang +M: Jian Shen M: Salil Mehta L: netdev@vger.kernel.org S: Maintained @@ -11154,7 +11140,7 @@ F: drivers/iio/pressure/dps310.c INFINEON PEB2466 ASoC CODEC M: Herve Codina -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/infineon,peb2466.yaml F: sound/soc/codecs/peb2466.c @@ -11280,10 +11266,10 @@ F: security/integrity/ F: security/integrity/ima/ INTEGRITY POLICY ENFORCEMENT (IPE) -M: Fan Wu +M: Fan Wu L: linux-security-module@vger.kernel.org S: Supported -T: git https://github.com/microsoft/ipe.git +T: git git://git.kernel.org/pub/scm/linux/kernel/git/wufan/ipe.git F: Documentation/admin-guide/LSM/ipe.rst F: Documentation/security/ipe.rst F: scripts/ipe/ @@ -11317,7 +11303,7 @@ M: Bard Liao M: Ranjani Sridharan M: Kai Vehmanen R: Pierre-Louis Bossart -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Supported F: sound/soc/intel/ @@ -11496,7 +11482,7 @@ F: include/uapi/linux/idxd.h INTEL IN FIELD SCAN (IFS) DEVICE M: Jithu Joseph -R: Ashok Raj +R: Ashok Raj R: Tony Luck S: Maintained F: drivers/platform/x86/intel/ifs @@ -11601,6 +11587,16 @@ F: drivers/crypto/intel/keembay/keembay-ocs-hcu-core.c F: drivers/crypto/intel/keembay/ocs-hcu.c F: drivers/crypto/intel/keembay/ocs-hcu.h +INTEL LA JOLLA COVE ADAPTER (LJCA) USB I/O EXPANDER DRIVERS +M: Wentong Wu +M: Sakari Ailus +S: Maintained +F: drivers/gpio/gpio-ljca.c +F: drivers/i2c/busses/i2c-ljca.c +F: drivers/spi/spi-ljca.c +F: drivers/usb/misc/usb-ljca.c +F: include/linux/usb/ljca.h + INTEL MANAGEMENT ENGINE (mei) M: Tomas Winkler L: linux-kernel@vger.kernel.org @@ -12001,7 +11997,7 @@ F: drivers/tty/ipwireless/ IRON DEVICE AUDIO CODEC DRIVERS M: Kiseok Jo -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/irondevice,* F: sound/soc/codecs/sma* @@ -12239,6 +12235,7 @@ R: Dmitry Vyukov R: Vincenzo Frascino L: kasan-dev@googlegroups.com S: Maintained +B: https://bugzilla.kernel.org/buglist.cgi?component=Sanitizers&product=Memory%20Management F: Documentation/dev-tools/kasan.rst F: arch/*/include/asm/*kasan.h F: arch/*/mm/kasan_init* @@ -12262,6 +12259,7 @@ R: Dmitry Vyukov R: Andrey Konovalov L: kasan-dev@googlegroups.com S: Maintained +B: https://bugzilla.kernel.org/buglist.cgi?component=Sanitizers&product=Memory%20Management F: Documentation/dev-tools/kcov.rst F: include/linux/kcov.h F: include/uapi/linux/kcov.h @@ -12343,6 +12341,7 @@ F: include/linux/randomize_kstack.h F: kernel/configs/hardening.config F: lib/usercopy_kunit.c F: mm/usercopy.c +F: security/Kconfig.hardening K: \b(add|choose)_random_kstack_offset\b K: \b__check_(object_size|heap_object)\b K: \b__counted_by\b @@ -12459,7 +12458,7 @@ F: virt/kvm/* KERNEL VIRTUAL MACHINE FOR ARM64 (KVM/arm64) M: Marc Zyngier M: Oliver Upton -R: James Morse +R: Joey Gouly R: Suzuki K Poulose R: Zenghui Yu L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) @@ -12940,49 +12939,29 @@ LIBATA PATA ARASAN COMPACT FLASH CONTROLLER M: Viresh Kumar L: linux-ide@vger.kernel.org S: Maintained -T: git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git F: drivers/ata/pata_arasan_cf.c F: include/linux/pata_arasan_cf_data.h -LIBATA PATA DRIVERS -R: Sergey Shtylyov -L: linux-ide@vger.kernel.org -F: drivers/ata/ata_*.c -F: drivers/ata/pata_*.c - LIBATA PATA FARADAY FTIDE010 AND GEMINI SATA BRIDGE DRIVERS M: Linus Walleij L: linux-ide@vger.kernel.org S: Maintained -T: git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git F: drivers/ata/pata_ftide010.c F: drivers/ata/sata_gemini.c F: drivers/ata/sata_gemini.h LIBATA SATA AHCI PLATFORM devices support M: Hans de Goede -M: Jens Axboe L: linux-ide@vger.kernel.org S: Maintained -T: git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git F: drivers/ata/ahci_platform.c F: drivers/ata/libahci_platform.c F: include/linux/ahci_platform.h -LIBATA SATA AHCI SYNOPSYS DWC CONTROLLER DRIVER -M: Serge Semin -L: linux-ide@vger.kernel.org -S: Maintained -T: git git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/libata.git -F: Documentation/devicetree/bindings/ata/baikal,bt1-ahci.yaml -F: Documentation/devicetree/bindings/ata/snps,dwc-ahci.yaml -F: drivers/ata/ahci_dwc.c - LIBATA SATA PROMISE TX2/TX4 CONTROLLER DRIVER M: Mikael Pettersson L: linux-ide@vger.kernel.org S: Maintained -T: git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git F: drivers/ata/sata_promise.* LIBATA SUBSYSTEM (Serial and Parallel ATA drivers) @@ -13952,7 +13931,7 @@ F: drivers/media/i2c/max96717.c MAX9860 MONO AUDIO VOICE CODEC DRIVER M: Peter Rosin -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/max9860.txt F: sound/soc/codecs/max9860.* @@ -14175,8 +14154,7 @@ T: git git://linuxtv.org/media_tree.git F: drivers/media/platform/nxp/imx-pxp.[ch] MEDIA DRIVERS FOR ASCOT2E -M: Sergey Kozlov -M: Abylay Ospan +M: Abylay Ospan L: linux-media@vger.kernel.org S: Supported W: https://linuxtv.org @@ -14193,8 +14171,7 @@ T: git git://linuxtv.org/media_tree.git F: drivers/media/dvb-frontends/cxd2099* MEDIA DRIVERS FOR CXD2841ER -M: Sergey Kozlov -M: Abylay Ospan +M: Abylay Ospan L: linux-media@vger.kernel.org S: Supported W: https://linuxtv.org @@ -14247,7 +14224,7 @@ F: drivers/media/platform/nxp/imx7-media-csi.c F: drivers/media/platform/nxp/imx8mq-mipi-csi2.c MEDIA DRIVERS FOR HELENE -M: Abylay Ospan +M: Abylay Ospan L: linux-media@vger.kernel.org S: Supported W: https://linuxtv.org @@ -14256,8 +14233,7 @@ T: git git://linuxtv.org/media_tree.git F: drivers/media/dvb-frontends/helene* MEDIA DRIVERS FOR HORUS3A -M: Sergey Kozlov -M: Abylay Ospan +M: Abylay Ospan L: linux-media@vger.kernel.org S: Supported W: https://linuxtv.org @@ -14266,8 +14242,7 @@ T: git git://linuxtv.org/media_tree.git F: drivers/media/dvb-frontends/horus3a* MEDIA DRIVERS FOR LNBH25 -M: Sergey Kozlov -M: Abylay Ospan +M: Abylay Ospan L: linux-media@vger.kernel.org S: Supported W: https://linuxtv.org @@ -14283,8 +14258,7 @@ T: git git://linuxtv.org/media_tree.git F: drivers/media/dvb-frontends/mxl5xx* MEDIA DRIVERS FOR NETUP PCI UNIVERSAL DVB devices -M: Sergey Kozlov -M: Abylay Ospan +M: Abylay Ospan L: linux-media@vger.kernel.org S: Supported W: https://linuxtv.org @@ -14909,9 +14883,10 @@ N: include/linux/page[-_]* MEMORY MAPPING M: Andrew Morton -R: Liam R. Howlett +M: Liam R. Howlett +M: Lorenzo Stoakes R: Vlastimil Babka -R: Lorenzo Stoakes +R: Jann Horn L: linux-mm@kvack.org S: Maintained W: http://www.linux-mm.org @@ -14934,13 +14909,6 @@ F: drivers/mtd/ F: include/linux/mtd/ F: include/uapi/mtd/ -MEMSENSING MICROSYSTEMS MSA311 DRIVER -M: Dmitry Rokosov -L: linux-iio@vger.kernel.org -S: Maintained -F: Documentation/devicetree/bindings/iio/accel/memsensing,msa311.yaml -F: drivers/iio/accel/msa311.c - MEN A21 WATCHDOG DRIVER M: Johannes Thumshirn L: linux-watchdog@vger.kernel.org @@ -15085,7 +15053,8 @@ F: drivers/spi/spi-at91-usart.c MICROCHIP AUDIO ASOC DRIVERS M: Claudiu Beznea -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +M: Andrei Simion +L: linux-sound@vger.kernel.org S: Supported F: Documentation/devicetree/bindings/sound/atmel* F: Documentation/devicetree/bindings/sound/axentia,tse850-pcm5142.txt @@ -15193,6 +15162,7 @@ F: include/video/atmel_lcdc.h MICROCHIP MCP16502 PMIC DRIVER M: Claudiu Beznea +M: Andrei Simion L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) S: Supported F: Documentation/devicetree/bindings/regulator/microchip,mcp16502.yaml @@ -15274,7 +15244,6 @@ F: drivers/tty/serial/8250/8250_pci1xxxx.c MICROCHIP POLARFIRE FPGA DRIVERS M: Conor Dooley -R: Vladimir Georgiev L: linux-fpga@vger.kernel.org S: Supported F: Documentation/devicetree/bindings/fpga/microchip,mpf-spi-fpga-mgr.yaml @@ -15324,6 +15293,7 @@ F: drivers/spi/spi-atmel.* MICROCHIP SSC DRIVER M: Claudiu Beznea +M: Andrei Simion L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) S: Supported F: Documentation/devicetree/bindings/misc/atmel-ssc.txt @@ -15529,17 +15499,6 @@ F: arch/mips/ F: drivers/platform/mips/ F: include/dt-bindings/mips/ -MIPS BAIKAL-T1 PLATFORM -M: Serge Semin -L: linux-mips@vger.kernel.org -S: Supported -F: Documentation/devicetree/bindings/bus/baikal,bt1-*.yaml -F: Documentation/devicetree/bindings/clock/baikal,bt1-*.yaml -F: drivers/bus/bt1-*.c -F: drivers/clk/baikal-t1/ -F: drivers/memory/bt1-l2-ctl.c -F: drivers/mtd/maps/physmap-bt1-rom.[ch] - MIPS BOSTON DEVELOPMENT BOARD M: Paul Burton L: linux-mips@vger.kernel.org @@ -15552,7 +15511,6 @@ F: include/dt-bindings/clock/boston-clock.h MIPS CORE DRIVERS M: Thomas Bogendoerfer -M: Serge Semin L: linux-mips@vger.kernel.org S: Supported F: drivers/bus/mips_cdmm.c @@ -15957,7 +15915,7 @@ F: include/linux/mtd/*nand*.h NATIVE INSTRUMENTS USB SOUND INTERFACE DRIVER M: Daniel Mack -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained W: http://www.native-instruments.com F: sound/usb/caiaq/ @@ -16088,6 +16046,7 @@ F: include/uapi/linux/net_dropmon.h F: net/core/drop_monitor.c NETWORKING DRIVERS +M: Andrew Lunn M: "David S. Miller" M: Eric Dumazet M: Jakub Kicinski @@ -16153,6 +16112,7 @@ M: "David S. Miller" M: Eric Dumazet M: Jakub Kicinski M: Paolo Abeni +R: Simon Horman L: netdev@vger.kernel.org S: Maintained P: Documentation/process/maintainer-netdev.rst @@ -16195,10 +16155,22 @@ F: include/uapi/linux/rtnetlink.h F: lib/net_utils.c F: lib/random32.c F: net/ +F: samples/pktgen/ F: tools/net/ F: tools/testing/selftests/net/ +X: Documentation/networking/mac80211-injection.rst +X: Documentation/networking/mac80211_hwsim/ +X: Documentation/networking/regulatory.rst +X: include/net/cfg80211.h +X: include/net/ieee80211_radiotap.h +X: include/net/iw_handler.h +X: include/net/mac80211.h +X: include/net/wext.h X: net/9p/ X: net/bluetooth/ +X: net/mac80211/ +X: net/rfkill/ +X: net/wireless/ NETWORKING [IPSEC] M: Steffen Klassert @@ -16508,12 +16480,6 @@ F: include/linux/ntb.h F: include/linux/ntb_transport.h F: tools/testing/selftests/ntb/ -NTB IDT DRIVER -M: Serge Semin -L: ntb@lists.linux.dev -S: Supported -F: drivers/ntb/hw/idt/ - NTB INTEL DRIVER M: Dave Jiang L: ntb@lists.linux.dev @@ -16728,7 +16694,7 @@ F: drivers/extcon/extcon-ptn5150.c NXP SGTL5000 DRIVER M: Fabio Estevam -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/fsl,sgtl5000.yaml F: sound/soc/codecs/sgtl5000* @@ -16752,7 +16718,7 @@ K: "nxp,tda998x" NXP TFA9879 DRIVER M: Peter Rosin -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/nxp,tfa9879.yaml F: sound/soc/codecs/tfa9879* @@ -16764,7 +16730,7 @@ F: drivers/nfc/nxp-nci NXP/Goodix TFA989X (TFA1) DRIVER M: Stephan Gerhold -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/nxp,tfa989x.yaml F: sound/soc/codecs/tfa989x.c @@ -16850,7 +16816,7 @@ F: include/uapi/misc/ocxl.h OMAP AUDIO SUPPORT M: Peter Ujfalusi M: Jarkko Nikula -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org L: linux-omap@vger.kernel.org S: Maintained F: sound/soc/ti/n810.c @@ -17407,7 +17373,7 @@ F: include/linux/pm_opp.h OPL4 DRIVER M: Clemens Ladisch -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git F: sound/drivers/opl4/ @@ -18534,13 +18500,6 @@ F: drivers/pps/ F: include/linux/pps*.h F: include/uapi/linux/pps.h -PPTP DRIVER -M: Dmitry Kozlov -L: netdev@vger.kernel.org -S: Maintained -W: http://sourceforge.net/projects/accel-pptp -F: drivers/net/ppp/pptp.c - PRESSURE STALL INFORMATION (PSI) M: Johannes Weiner M: Suren Baghdasaryan @@ -18790,7 +18749,7 @@ F: drivers/crypto/intel/qat/ QCOM AUDIO (ASoC) DRIVERS M: Srinivas Kandagatla -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org L: linux-arm-msm@vger.kernel.org S: Supported F: Documentation/devicetree/bindings/soc/qcom/qcom,apr* @@ -19514,6 +19473,14 @@ S: Maintained F: Documentation/tools/rtla/ F: tools/tracing/rtla/ +Real-time Linux (PREEMPT_RT) +M: Sebastian Andrzej Siewior +M: Clark Williams +M: Steven Rostedt +L: linux-rt-devel@lists.linux.dev +S: Supported +K: PREEMPT_RT + REALTEK AUDIO CODECS M: Oder Chiou S: Maintained @@ -19623,15 +19590,6 @@ S: Supported F: Documentation/devicetree/bindings/i2c/renesas,iic-emev2.yaml F: drivers/i2c/busses/i2c-emev2.c -RENESAS ETHERNET AVB DRIVER -R: Sergey Shtylyov -L: netdev@vger.kernel.org -L: linux-renesas-soc@vger.kernel.org -F: Documentation/devicetree/bindings/net/renesas,etheravb.yaml -F: drivers/net/ethernet/renesas/Kconfig -F: drivers/net/ethernet/renesas/Makefile -F: drivers/net/ethernet/renesas/ravb* - RENESAS ETHERNET SWITCH DRIVER R: Yoshihiro Shimoda L: netdev@vger.kernel.org @@ -19652,7 +19610,7 @@ F: drivers/net/ethernet/renesas/rtsn.* RENESAS IDT821034 ASoC CODEC M: Herve Codina -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/renesas,idt821034.yaml F: sound/soc/codecs/idt821034.c @@ -19681,14 +19639,6 @@ F: Documentation/devicetree/bindings/i2c/renesas,rmobile-iic.yaml F: drivers/i2c/busses/i2c-rcar.c F: drivers/i2c/busses/i2c-sh_mobile.c -RENESAS R-CAR SATA DRIVER -R: Sergey Shtylyov -L: linux-ide@vger.kernel.org -L: linux-renesas-soc@vger.kernel.org -S: Supported -F: Documentation/devicetree/bindings/ata/renesas,rcar-sata.yaml -F: drivers/ata/sata_rcar.c - RENESAS R-CAR THERMAL DRIVERS M: Niklas Söderlund L: linux-renesas-soc@vger.kernel.org @@ -19764,16 +19714,6 @@ S: Supported F: Documentation/devicetree/bindings/i2c/renesas,rzv2m.yaml F: drivers/i2c/busses/i2c-rzv2m.c -RENESAS SUPERH ETHERNET DRIVER -R: Sergey Shtylyov -L: netdev@vger.kernel.org -L: linux-renesas-soc@vger.kernel.org -F: Documentation/devicetree/bindings/net/renesas,ether.yaml -F: drivers/net/ethernet/renesas/Kconfig -F: drivers/net/ethernet/renesas/Makefile -F: drivers/net/ethernet/renesas/sh_eth* -F: include/linux/sh_eth.h - RENESAS USB PHY DRIVER M: Yoshihiro Shimoda L: linux-renesas-soc@vger.kernel.org @@ -20403,7 +20343,7 @@ F: security/safesetid/ SAMSUNG AUDIO (ASoC) DRIVERS M: Sylwester Nawrocki -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained B: mailto:linux-samsung-soc@vger.kernel.org F: Documentation/devicetree/bindings/sound/samsung* @@ -20939,7 +20879,7 @@ F: drivers/media/rc/serial_ir.c SERIAL LOW-POWER INTER-CHIP MEDIA BUS (SLIMbus) M: Srinivas Kandagatla -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/slimbus/ F: drivers/slimbus/ @@ -21373,7 +21313,7 @@ F: Documentation/devicetree/bindings/i2c/socionext,synquacer-i2c.yaml F: drivers/i2c/busses/i2c-synquacer.c SOCIONEXT UNIPHIER SOUND DRIVER -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Orphan F: sound/soc/uniphier/ @@ -21632,7 +21572,7 @@ F: tools/testing/selftests/alsa SOUND - COMPRESSED AUDIO M: Vinod Koul -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Supported T: git git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git F: Documentation/sound/designs/compress-offload.rst @@ -21695,7 +21635,7 @@ M: Vinod Koul M: Bard Liao R: Pierre-Louis Bossart R: Sanyog Kale -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Supported T: git git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire.git F: Documentation/driver-api/soundwire/ @@ -21768,8 +21708,8 @@ F: drivers/accessibility/speakup/ SPEAR PLATFORM/CLOCK/PINCTRL SUPPORT M: Viresh Kumar M: Shiraz Hashim -M: soc@kernel.org L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) +L: soc@lists.linux.dev S: Maintained W: http://www.st.com/spear F: arch/arm/boot/dts/st/spear* @@ -22168,7 +22108,7 @@ F: kernel/static_call.c STI AUDIO (ASoC) DRIVERS M: Arnaud Pouliquen -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/st,sti-asoc-card.txt F: sound/soc/sti/ @@ -22189,7 +22129,7 @@ F: drivers/media/usb/stk1160/ STM32 AUDIO (ASoC) DRIVERS M: Olivier Moysan M: Arnaud Pouliquen -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/iio/adc/st,stm32-dfsdm-adc.yaml F: Documentation/devicetree/bindings/sound/st,stm32-*.yaml @@ -22427,19 +22367,11 @@ F: drivers/tty/serial/8250/8250_lpss.c SYNOPSYS DESIGNWARE APB GPIO DRIVER M: Hoan Tran -M: Serge Semin L: linux-gpio@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/gpio/snps,dw-apb-gpio.yaml F: drivers/gpio/gpio-dwapb.c -SYNOPSYS DESIGNWARE APB SSI DRIVER -M: Serge Semin -L: linux-spi@vger.kernel.org -S: Supported -F: Documentation/devicetree/bindings/spi/snps,dw-apb-ssi.yaml -F: drivers/spi/spi-dw* - SYNOPSYS DESIGNWARE AXI DMAC DRIVER M: Eugeniy Paltsev S: Maintained @@ -22892,7 +22824,7 @@ F: drivers/irqchip/irq-xtensa-* TEXAS INSTRUMENTS ASoC DRIVERS M: Peter Ujfalusi -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/davinci-mcasp-audio.yaml F: sound/soc/ti/ @@ -22901,7 +22833,7 @@ TEXAS INSTRUMENTS AUDIO (ASoC/HDA) DRIVERS M: Shenghao Ding M: Kevin Lu M: Baojun Xu -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/tas2552.txt F: Documentation/devicetree/bindings/sound/ti,tas2562.yaml @@ -23269,7 +23201,7 @@ F: drivers/soc/ti/* TI LM49xxx FAMILY ASoC CODEC DRIVERS M: M R Swami Reddy M: Vishwas A Deshpande -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: sound/soc/codecs/isabelle* F: sound/soc/codecs/lm49453* @@ -23283,15 +23215,15 @@ F: Documentation/devicetree/bindings/iio/adc/ti,lmp92064.yaml F: drivers/iio/adc/ti-lmp92064.c TI PCM3060 ASoC CODEC DRIVER -M: Kirill Marinushkin -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +M: Kirill Marinushkin +L: linux-sound@vger.kernel.org S: Maintained F: Documentation/devicetree/bindings/sound/pcm3060.txt F: sound/soc/codecs/pcm3060* TI TAS571X FAMILY ASoC CODEC DRIVER M: Kevin Cernekee -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Odd Fixes F: sound/soc/codecs/tas571x* @@ -23319,7 +23251,7 @@ F: drivers/iio/adc/ti-tsc2046.c TI TWL4030 SERIES SOC CODEC DRIVER M: Peter Ujfalusi -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: sound/soc/codecs/twl4030* @@ -23749,12 +23681,6 @@ L: linux-input@vger.kernel.org S: Maintained F: drivers/hid/hid-udraw-ps3.c -UFS FILESYSTEM -M: Evgeniy Dushistov -S: Maintained -F: Documentation/admin-guide/ufs.rst -F: fs/ufs/ - UHID USERSPACE HID IO DRIVER M: David Rheinsberg L: linux-input@vger.kernel.org @@ -23995,7 +23921,7 @@ F: drivers/usb/storage/ USB MIDI DRIVER M: Clemens Ladisch -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git F: sound/usb/midi.* @@ -24057,6 +23983,7 @@ USB RAW GADGET DRIVER R: Andrey Konovalov L: linux-usb@vger.kernel.org S: Maintained +B: https://github.com/xairy/raw-gadget/issues F: Documentation/usb/raw-gadget.rst F: drivers/usb/gadget/legacy/raw_gadget.c F: include/uapi/linux/usb/raw_gadget.h @@ -24173,8 +24100,12 @@ F: drivers/usb/host/xhci* USER DATAGRAM PROTOCOL (UDP) M: Willem de Bruijn +L: netdev@vger.kernel.org S: Maintained F: include/linux/udp.h +F: include/net/udp.h +F: include/trace/events/udp.h +F: include/uapi/linux/udp.h F: net/ipv4/udp.c F: net/ipv6/udp.c @@ -24201,6 +24132,7 @@ F: lib/iov_iter.c USERSPACE DMA BUFFER DRIVER M: Gerd Hoffmann +M: Vivek Kasireddy L: dri-devel@lists.freedesktop.org S: Maintained T: git https://gitlab.freedesktop.org/drm/misc/kernel.git @@ -24655,7 +24587,7 @@ VIRTIO SOUND DRIVER M: Anton Yakovlev M: "Michael S. Tsirkin" L: virtualization@lists.linux.dev -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Maintained F: include/uapi/linux/virtio_snd.h F: sound/virtio/* @@ -24724,9 +24656,10 @@ F: tools/testing/vsock/ VMA M: Andrew Morton -R: Liam R. Howlett +M: Liam R. Howlett +M: Lorenzo Stoakes R: Vlastimil Babka -R: Lorenzo Stoakes +R: Jann Horn L: linux-mm@kvack.org S: Maintained W: https://www.linux-mm.org @@ -25384,7 +25317,7 @@ F: include/xen/interface/io/usbif.h XEN SOUND FRONTEND DRIVER M: Oleksandr Andrushchenko L: xen-devel@lists.xenproject.org (moderated for non-subscribers) -L: alsa-devel@alsa-project.org (moderated for non-subscribers) +L: linux-sound@vger.kernel.org S: Supported F: sound/xen/* @@ -25400,7 +25333,7 @@ F: include/xen/arm/swiotlb-xen.h F: include/xen/swiotlb-xen.h XFS FILESYSTEM -M: Chandan Babu R +M: Carlos Maiolino R: Darrick J. Wong L: linux-xfs@vger.kernel.org S: Supported diff --git a/Makefile b/Makefile index 187a4ce2728e..b8efbfe9da94 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 6 PATCHLEVEL = 12 SUBLEVEL = 0 -EXTRAVERSION = -rc1 +EXTRAVERSION = -rc6 NAME = Baby Opossum Posse # *DOCUMENTATION* @@ -1645,7 +1645,7 @@ help: echo '* dtbs - Build device tree blobs for enabled boards'; \ echo ' dtbs_install - Install dtbs to $(INSTALL_DTBS_PATH)'; \ echo ' dt_binding_check - Validate device tree binding documents and examples'; \ - echo ' dt_binding_schema - Build processed device tree binding schemas'; \ + echo ' dt_binding_schemas - Build processed device tree binding schemas'; \ echo ' dtbs_check - Validate device tree source files';\ echo '') diff --git a/arch/Kconfig b/arch/Kconfig index 98157b38f5cf..bd9f095d69fa 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -838,7 +838,7 @@ config CFI_CLANG config CFI_ICALL_NORMALIZE_INTEGERS bool "Normalize CFI tags for integers" depends on CFI_CLANG - depends on $(cc-option,-fsanitize=kcfi -fsanitize-cfi-icall-experimental-normalize-integers) + depends on HAVE_CFI_ICALL_NORMALIZE_INTEGERS_CLANG help This option normalizes the CFI tags for integer types so that all integer types of the same size and signedness receive the same CFI @@ -851,6 +851,20 @@ config CFI_ICALL_NORMALIZE_INTEGERS This option is necessary for using CFI with Rust. If unsure, say N. +config HAVE_CFI_ICALL_NORMALIZE_INTEGERS_CLANG + def_bool y + depends on $(cc-option,-fsanitize=kcfi -fsanitize-cfi-icall-experimental-normalize-integers) + # With GCOV/KASAN we need this fix: https://github.com/llvm/llvm-project/pull/104826 + depends on CLANG_VERSION >= 190103 || (!GCOV_KERNEL && !KASAN_GENERIC && !KASAN_SW_TAGS) + +config HAVE_CFI_ICALL_NORMALIZE_INTEGERS_RUSTC + def_bool y + depends on HAVE_CFI_ICALL_NORMALIZE_INTEGERS_CLANG + depends on RUSTC_VERSION >= 107900 + # With GCOV/KASAN we need this fix: https://github.com/rust-lang/rust/pull/129373 + depends on (RUSTC_LLVM_VERSION >= 190103 && RUSTC_VERSION >= 108200) || \ + (!GCOV_KERNEL && !KASAN_GENERIC && !KASAN_SW_TAGS) + config CFI_PERMISSIVE bool "Use CFI in permissive mode" depends on CFI_CLANG diff --git a/arch/alpha/kernel/traps.c b/arch/alpha/kernel/traps.c index 6afae65e9a8b..a9a38c80c4a7 100644 --- a/arch/alpha/kernel/traps.c +++ b/arch/alpha/kernel/traps.c @@ -22,7 +22,7 @@ #include #include -#include +#include #include #include #include diff --git a/arch/arc/include/asm/io.h b/arch/arc/include/asm/io.h index 4fdb7350636c..f57cb5a6b624 100644 --- a/arch/arc/include/asm/io.h +++ b/arch/arc/include/asm/io.h @@ -9,7 +9,7 @@ #include #include #include -#include +#include #ifdef CONFIG_ISA_ARCV2 #include diff --git a/arch/arc/include/asm/mmu.h b/arch/arc/include/asm/mmu.h index 9febf5bc3de6..4ae2db59d494 100644 --- a/arch/arc/include/asm/mmu.h +++ b/arch/arc/include/asm/mmu.h @@ -14,6 +14,7 @@ typedef struct { unsigned long asid[NR_CPUS]; /* 8 bit MMU PID + Generation cycle */ } mm_context_t; +struct pt_regs; extern void do_tlb_overlap_fault(unsigned long, unsigned long, struct pt_regs *); #endif diff --git a/arch/arc/include/asm/unaligned.h b/arch/arc/include/asm/unaligned.h deleted file mode 100644 index cf5a02382e0e..000000000000 --- a/arch/arc/include/asm/unaligned.h +++ /dev/null @@ -1,27 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0-only */ -/* - * Copyright (C) 2004, 2007-2010, 2011-2012 Synopsys, Inc. (www.synopsys.com) - */ - -#ifndef _ASM_ARC_UNALIGNED_H -#define _ASM_ARC_UNALIGNED_H - -/* ARC700 can't handle unaligned Data accesses. */ - -#include -#include - -#ifdef CONFIG_ARC_EMUL_UNALIGNED -int misaligned_fixup(unsigned long address, struct pt_regs *regs, - struct callee_regs *cregs); -#else -static inline int -misaligned_fixup(unsigned long address, struct pt_regs *regs, - struct callee_regs *cregs) -{ - /* Not fixed */ - return 1; -} -#endif - -#endif /* _ASM_ARC_UNALIGNED_H */ diff --git a/arch/arc/kernel/traps.c b/arch/arc/kernel/traps.c index a19751e824fb..8d2ea2cbd98b 100644 --- a/arch/arc/kernel/traps.c +++ b/arch/arc/kernel/traps.c @@ -18,8 +18,9 @@ #include #include #include -#include +#include #include +#include "unaligned.h" void die(const char *str, struct pt_regs *regs, unsigned long address) { diff --git a/arch/arc/kernel/unaligned.c b/arch/arc/kernel/unaligned.c index 99a9b92ed98d..d2f5ceaaed1b 100644 --- a/arch/arc/kernel/unaligned.c +++ b/arch/arc/kernel/unaligned.c @@ -12,6 +12,7 @@ #include #include #include +#include "unaligned.h" #ifdef CONFIG_CPU_BIG_ENDIAN #define BE 1 diff --git a/arch/arc/kernel/unaligned.h b/arch/arc/kernel/unaligned.h new file mode 100644 index 000000000000..5244453bb85f --- /dev/null +++ b/arch/arc/kernel/unaligned.h @@ -0,0 +1,16 @@ +struct pt_regs; +struct callee_regs; + +#ifdef CONFIG_ARC_EMUL_UNALIGNED +int misaligned_fixup(unsigned long address, struct pt_regs *regs, + struct callee_regs *cregs); +#else +static inline int +misaligned_fixup(unsigned long address, struct pt_regs *regs, + struct callee_regs *cregs) +{ + /* Not fixed */ + return 1; +} +#endif + diff --git a/arch/arc/kernel/unwind.c b/arch/arc/kernel/unwind.c index 9270d0a713c3..d8969dab12d4 100644 --- a/arch/arc/kernel/unwind.c +++ b/arch/arc/kernel/unwind.c @@ -19,7 +19,7 @@ #include #include #include -#include +#include #include extern char __start_unwind[], __end_unwind[]; diff --git a/arch/arm/boot/dts/broadcom/bcm2837-rpi-cm3-io3.dts b/arch/arm/boot/dts/broadcom/bcm2837-rpi-cm3-io3.dts index 72d26d130efa..85f54fa595aa 100644 --- a/arch/arm/boot/dts/broadcom/bcm2837-rpi-cm3-io3.dts +++ b/arch/arm/boot/dts/broadcom/bcm2837-rpi-cm3-io3.dts @@ -77,7 +77,7 @@ }; &hdmi { - hpd-gpios = <&expgpio 1 GPIO_ACTIVE_LOW>; + hpd-gpios = <&expgpio 0 GPIO_ACTIVE_LOW>; power-domains = <&power RPI_POWER_DOMAIN_HDMI>; status = "okay"; }; diff --git a/arch/arm/crypto/aes-ce-glue.c b/arch/arm/crypto/aes-ce-glue.c index f5b66f4cf45d..21df5e7f51f9 100644 --- a/arch/arm/crypto/aes-ce-glue.c +++ b/arch/arm/crypto/aes-ce-glue.c @@ -8,7 +8,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/arm/crypto/crc32-ce-glue.c b/arch/arm/crypto/crc32-ce-glue.c index 4ff18044af07..20b4dff13e3a 100644 --- a/arch/arm/crypto/crc32-ce-glue.c +++ b/arch/arm/crypto/crc32-ce-glue.c @@ -18,7 +18,7 @@ #include #include #include -#include +#include #define PMULL_MIN_LEN 64L /* minimum size of buffer * for crc32_pmull_le_16 */ diff --git a/arch/arm/crypto/ghash-ce-glue.c b/arch/arm/crypto/ghash-ce-glue.c index 3ddf05b4234d..3af997082534 100644 --- a/arch/arm/crypto/ghash-ce-glue.c +++ b/arch/arm/crypto/ghash-ce-glue.c @@ -9,7 +9,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/arm/crypto/poly1305-glue.c b/arch/arm/crypto/poly1305-glue.c index 8482e302c45a..4464ffbf8fd1 100644 --- a/arch/arm/crypto/poly1305-glue.c +++ b/arch/arm/crypto/poly1305-glue.c @@ -8,7 +8,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/arm/crypto/sha2-ce-glue.c b/arch/arm/crypto/sha2-ce-glue.c index c62ce89dd3e0..aeac45bfbf9f 100644 --- a/arch/arm/crypto/sha2-ce-glue.c +++ b/arch/arm/crypto/sha2-ce-glue.c @@ -16,7 +16,7 @@ #include #include #include -#include +#include #include "sha256_glue.h" diff --git a/arch/arm/include/asm/uaccess.h b/arch/arm/include/asm/uaccess.h index 6c9c16d767cf..f90be312418e 100644 --- a/arch/arm/include/asm/uaccess.h +++ b/arch/arm/include/asm/uaccess.h @@ -12,7 +12,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/arm/mm/alignment.c b/arch/arm/mm/alignment.c index f8dd0b3cc8e0..3c6ddb1afdc4 100644 --- a/arch/arm/mm/alignment.c +++ b/arch/arm/mm/alignment.c @@ -22,7 +22,7 @@ #include #include -#include +#include #include #include "fault.h" diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 3e29b44d2d7b..fd9df6dcc593 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -200,7 +200,8 @@ config ARM64 select HAVE_DMA_CONTIGUOUS select HAVE_DYNAMIC_FTRACE select HAVE_DYNAMIC_FTRACE_WITH_ARGS \ - if $(cc-option,-fpatchable-function-entry=2) + if (GCC_SUPPORTS_DYNAMIC_FTRACE_WITH_ARGS || \ + CLANG_SUPPORTS_DYNAMIC_FTRACE_WITH_ARGS) select HAVE_DYNAMIC_FTRACE_WITH_DIRECT_CALLS \ if DYNAMIC_FTRACE_WITH_ARGS && DYNAMIC_FTRACE_WITH_CALL_OPS select HAVE_DYNAMIC_FTRACE_WITH_CALL_OPS \ @@ -286,12 +287,10 @@ config CLANG_SUPPORTS_DYNAMIC_FTRACE_WITH_ARGS def_bool CC_IS_CLANG # https://github.com/ClangBuiltLinux/linux/issues/1507 depends on AS_IS_GNU || (AS_IS_LLVM && (LD_IS_LLD || LD_VERSION >= 23600)) - select HAVE_DYNAMIC_FTRACE_WITH_ARGS config GCC_SUPPORTS_DYNAMIC_FTRACE_WITH_ARGS def_bool CC_IS_GCC depends on $(cc-option,-fpatchable-function-entry=2) - select HAVE_DYNAMIC_FTRACE_WITH_ARGS config 64BIT def_bool y @@ -1097,6 +1096,7 @@ config ARM64_ERRATUM_3194386 * ARM Cortex-A78C erratum 3324346 * ARM Cortex-A78C erratum 3324347 * ARM Cortex-A710 erratam 3324338 + * ARM Cortex-A715 errartum 3456084 * ARM Cortex-A720 erratum 3456091 * ARM Cortex-A725 erratum 3456106 * ARM Cortex-X1 erratum 3324344 @@ -1107,6 +1107,7 @@ config ARM64_ERRATUM_3194386 * ARM Cortex-X925 erratum 3324334 * ARM Neoverse-N1 erratum 3324349 * ARM Neoverse N2 erratum 3324339 + * ARM Neoverse-N3 erratum 3456111 * ARM Neoverse-V1 erratum 3324341 * ARM Neoverse V2 erratum 3324336 * ARM Neoverse-V3 erratum 3312417 diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile index b058c4803efb..9efd3f37c2fd 100644 --- a/arch/arm64/Makefile +++ b/arch/arm64/Makefile @@ -10,7 +10,7 @@ # # Copyright (C) 1995-2001 by Russell King -LDFLAGS_vmlinux :=--no-undefined -X +LDFLAGS_vmlinux :=--no-undefined -X --pic-veneer ifeq ($(CONFIG_RELOCATABLE), y) # Pass --no-apply-dynamic-relocs to restore pre-binutils-2.27 behaviour diff --git a/arch/arm64/boot/dts/marvell/cn9130-sr-som.dtsi b/arch/arm64/boot/dts/marvell/cn9130-sr-som.dtsi index 4676e3488f54..cb8d54895a77 100644 --- a/arch/arm64/boot/dts/marvell/cn9130-sr-som.dtsi +++ b/arch/arm64/boot/dts/marvell/cn9130-sr-som.dtsi @@ -136,7 +136,7 @@ }; cp0_mdio_pins: cp0-mdio-pins { - marvell,pins = "mpp40", "mpp41"; + marvell,pins = "mpp0", "mpp1"; marvell,function = "ge"; }; diff --git a/arch/arm64/crypto/aes-ce-ccm-glue.c b/arch/arm64/crypto/aes-ce-ccm-glue.c index ce9b28e3c7d6..a523b519700f 100644 --- a/arch/arm64/crypto/aes-ce-ccm-glue.c +++ b/arch/arm64/crypto/aes-ce-ccm-glue.c @@ -9,7 +9,7 @@ */ #include -#include +#include #include #include #include diff --git a/arch/arm64/crypto/aes-ce-glue.c b/arch/arm64/crypto/aes-ce-glue.c index e921823ca103..00b8749013c5 100644 --- a/arch/arm64/crypto/aes-ce-glue.c +++ b/arch/arm64/crypto/aes-ce-glue.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include #include #include diff --git a/arch/arm64/crypto/ghash-ce-glue.c b/arch/arm64/crypto/ghash-ce-glue.c index 97331b454ea8..da7b7ec1a664 100644 --- a/arch/arm64/crypto/ghash-ce-glue.c +++ b/arch/arm64/crypto/ghash-ce-glue.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include #include #include diff --git a/arch/arm64/crypto/poly1305-glue.c b/arch/arm64/crypto/poly1305-glue.c index 9c4bfd62e789..18883ea438f3 100644 --- a/arch/arm64/crypto/poly1305-glue.c +++ b/arch/arm64/crypto/poly1305-glue.c @@ -8,7 +8,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/arm64/crypto/sha1-ce-glue.c b/arch/arm64/crypto/sha1-ce-glue.c index 1dd93e1fcb39..cbd14f208f83 100644 --- a/arch/arm64/crypto/sha1-ce-glue.c +++ b/arch/arm64/crypto/sha1-ce-glue.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include #include #include diff --git a/arch/arm64/crypto/sha2-ce-glue.c b/arch/arm64/crypto/sha2-ce-glue.c index 0a44d2e7ee1f..6b4866a88ded 100644 --- a/arch/arm64/crypto/sha2-ce-glue.c +++ b/arch/arm64/crypto/sha2-ce-glue.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include #include #include diff --git a/arch/arm64/crypto/sha3-ce-glue.c b/arch/arm64/crypto/sha3-ce-glue.c index 250e1377c481..5662c3ac49e9 100644 --- a/arch/arm64/crypto/sha3-ce-glue.c +++ b/arch/arm64/crypto/sha3-ce-glue.c @@ -12,7 +12,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/arm64/crypto/sha512-ce-glue.c b/arch/arm64/crypto/sha512-ce-glue.c index f3431fc62315..071f64293227 100644 --- a/arch/arm64/crypto/sha512-ce-glue.c +++ b/arch/arm64/crypto/sha512-ce-glue.c @@ -11,7 +11,7 @@ #include #include -#include +#include #include #include #include diff --git a/arch/arm64/crypto/sm3-ce-glue.c b/arch/arm64/crypto/sm3-ce-glue.c index 54bf6ebcfffb..1a71788c4cda 100644 --- a/arch/arm64/crypto/sm3-ce-glue.c +++ b/arch/arm64/crypto/sm3-ce-glue.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include #include #include diff --git a/arch/arm64/crypto/sm3-neon-glue.c b/arch/arm64/crypto/sm3-neon-glue.c index 7182ee683f14..8dd71ce79b69 100644 --- a/arch/arm64/crypto/sm3-neon-glue.c +++ b/arch/arm64/crypto/sm3-neon-glue.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include #include #include diff --git a/arch/arm64/include/asm/cputype.h b/arch/arm64/include/asm/cputype.h index 5a7dfeb8e8eb..488f8e751349 100644 --- a/arch/arm64/include/asm/cputype.h +++ b/arch/arm64/include/asm/cputype.h @@ -94,6 +94,7 @@ #define ARM_CPU_PART_NEOVERSE_V3 0xD84 #define ARM_CPU_PART_CORTEX_X925 0xD85 #define ARM_CPU_PART_CORTEX_A725 0xD87 +#define ARM_CPU_PART_NEOVERSE_N3 0xD8E #define APM_CPU_PART_XGENE 0x000 #define APM_CPU_VAR_POTENZA 0x00 @@ -176,6 +177,7 @@ #define MIDR_NEOVERSE_V3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_V3) #define MIDR_CORTEX_X925 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_X925) #define MIDR_CORTEX_A725 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A725) +#define MIDR_NEOVERSE_N3 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_NEOVERSE_N3) #define MIDR_THUNDERX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX) #define MIDR_THUNDERX_81XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_81XX) #define MIDR_THUNDERX_83XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_83XX) diff --git a/arch/arm64/include/asm/kvm_asm.h b/arch/arm64/include/asm/kvm_asm.h index b36a3b6cc011..67afac659231 100644 --- a/arch/arm64/include/asm/kvm_asm.h +++ b/arch/arm64/include/asm/kvm_asm.h @@ -178,6 +178,7 @@ struct kvm_nvhe_init_params { unsigned long hcr_el2; unsigned long vttbr; unsigned long vtcr; + unsigned long tmp; }; /* diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index 329619c6fa96..bf64fed9820e 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -51,6 +51,7 @@ #define KVM_REQ_RELOAD_PMU KVM_ARCH_REQ(5) #define KVM_REQ_SUSPEND KVM_ARCH_REQ(6) #define KVM_REQ_RESYNC_PMU_EL0 KVM_ARCH_REQ(7) +#define KVM_REQ_NESTED_S2_UNMAP KVM_ARCH_REQ(8) #define KVM_DIRTY_LOG_MANUAL_CAPS (KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE | \ KVM_DIRTY_LOG_INITIALLY_SET) @@ -211,6 +212,12 @@ struct kvm_s2_mmu { */ bool nested_stage2_enabled; + /* + * true when this MMU needs to be unmapped before being used for a new + * purpose. + */ + bool pending_unmap; + /* * 0: Nobody is currently using this, check vttbr for validity * >0: Somebody is actively using this. @@ -1441,11 +1448,6 @@ void kvm_set_vm_id_reg(struct kvm *kvm, u32 reg, u64 val); sign_extend64(__val, id##_##fld##_WIDTH - 1); \ }) -#define expand_field_sign(id, fld, val) \ - (id##_##fld##_SIGNED ? \ - __expand_field_sign_signed(id, fld, val) : \ - __expand_field_sign_unsigned(id, fld, val)) - #define get_idreg_field_unsigned(kvm, id, fld) \ ({ \ u64 __val = kvm_read_vm_id_reg((kvm), SYS_##id); \ @@ -1461,20 +1463,26 @@ void kvm_set_vm_id_reg(struct kvm *kvm, u32 reg, u64 val); #define get_idreg_field_enum(kvm, id, fld) \ get_idreg_field_unsigned(kvm, id, fld) -#define get_idreg_field(kvm, id, fld) \ +#define kvm_cmp_feat_signed(kvm, id, fld, op, limit) \ + (get_idreg_field_signed((kvm), id, fld) op __expand_field_sign_signed(id, fld, limit)) + +#define kvm_cmp_feat_unsigned(kvm, id, fld, op, limit) \ + (get_idreg_field_unsigned((kvm), id, fld) op __expand_field_sign_unsigned(id, fld, limit)) + +#define kvm_cmp_feat(kvm, id, fld, op, limit) \ (id##_##fld##_SIGNED ? \ - get_idreg_field_signed(kvm, id, fld) : \ - get_idreg_field_unsigned(kvm, id, fld)) + kvm_cmp_feat_signed(kvm, id, fld, op, limit) : \ + kvm_cmp_feat_unsigned(kvm, id, fld, op, limit)) #define kvm_has_feat(kvm, id, fld, limit) \ - (get_idreg_field((kvm), id, fld) >= expand_field_sign(id, fld, limit)) + kvm_cmp_feat(kvm, id, fld, >=, limit) #define kvm_has_feat_enum(kvm, id, fld, val) \ - (get_idreg_field_unsigned((kvm), id, fld) == __expand_field_sign_unsigned(id, fld, val)) + kvm_cmp_feat_unsigned(kvm, id, fld, ==, val) #define kvm_has_feat_range(kvm, id, fld, min, max) \ - (get_idreg_field((kvm), id, fld) >= expand_field_sign(id, fld, min) && \ - get_idreg_field((kvm), id, fld) <= expand_field_sign(id, fld, max)) + (kvm_cmp_feat(kvm, id, fld, >=, min) && \ + kvm_cmp_feat(kvm, id, fld, <=, max)) /* Check for a given level of PAuth support */ #define kvm_has_pauth(k, l) \ diff --git a/arch/arm64/include/asm/kvm_mmu.h b/arch/arm64/include/asm/kvm_mmu.h index cd4087fbda9a..66d93e320ec8 100644 --- a/arch/arm64/include/asm/kvm_mmu.h +++ b/arch/arm64/include/asm/kvm_mmu.h @@ -166,7 +166,8 @@ int create_hyp_exec_mappings(phys_addr_t phys_addr, size_t size, int create_hyp_stack(phys_addr_t phys_addr, unsigned long *haddr); void __init free_hyp_pgds(void); -void kvm_stage2_unmap_range(struct kvm_s2_mmu *mmu, phys_addr_t start, u64 size); +void kvm_stage2_unmap_range(struct kvm_s2_mmu *mmu, phys_addr_t start, + u64 size, bool may_block); void kvm_stage2_flush_range(struct kvm_s2_mmu *mmu, phys_addr_t addr, phys_addr_t end); void kvm_stage2_wp_range(struct kvm_s2_mmu *mmu, phys_addr_t addr, phys_addr_t end); diff --git a/arch/arm64/include/asm/kvm_nested.h b/arch/arm64/include/asm/kvm_nested.h index e8bc6d67aba2..233e65522716 100644 --- a/arch/arm64/include/asm/kvm_nested.h +++ b/arch/arm64/include/asm/kvm_nested.h @@ -78,6 +78,8 @@ extern void kvm_s2_mmu_iterate_by_vmid(struct kvm *kvm, u16 vmid, extern void kvm_vcpu_load_hw_mmu(struct kvm_vcpu *vcpu); extern void kvm_vcpu_put_hw_mmu(struct kvm_vcpu *vcpu); +extern void check_nested_vcpu_requests(struct kvm_vcpu *vcpu); + struct kvm_s2_trans { phys_addr_t output; unsigned long block_size; @@ -124,7 +126,7 @@ extern int kvm_s2_handle_perm_fault(struct kvm_vcpu *vcpu, struct kvm_s2_trans *trans); extern int kvm_inject_s2_fault(struct kvm_vcpu *vcpu, u64 esr_el2); extern void kvm_nested_s2_wp(struct kvm *kvm); -extern void kvm_nested_s2_unmap(struct kvm *kvm); +extern void kvm_nested_s2_unmap(struct kvm *kvm, bool may_block); extern void kvm_nested_s2_flush(struct kvm *kvm); unsigned long compute_tlb_inval_range(struct kvm_s2_mmu *mmu, u64 val); diff --git a/arch/arm64/include/asm/uprobes.h b/arch/arm64/include/asm/uprobes.h index 2b09495499c6..014b02897f8e 100644 --- a/arch/arm64/include/asm/uprobes.h +++ b/arch/arm64/include/asm/uprobes.h @@ -10,11 +10,9 @@ #include #include -#define MAX_UINSN_BYTES AARCH64_INSN_SIZE - #define UPROBE_SWBP_INSN cpu_to_le32(BRK64_OPCODE_UPROBES) #define UPROBE_SWBP_INSN_SIZE AARCH64_INSN_SIZE -#define UPROBE_XOL_SLOT_BYTES MAX_UINSN_BYTES +#define UPROBE_XOL_SLOT_BYTES AARCH64_INSN_SIZE typedef __le32 uprobe_opcode_t; @@ -23,8 +21,8 @@ struct arch_uprobe_task { struct arch_uprobe { union { - u8 insn[MAX_UINSN_BYTES]; - u8 ixol[MAX_UINSN_BYTES]; + __le32 insn; + __le32 ixol; }; struct arch_probe_insn api; bool simulate; diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c index 27de1dddb0ab..b21dd24b8efc 100644 --- a/arch/arm64/kernel/asm-offsets.c +++ b/arch/arm64/kernel/asm-offsets.c @@ -146,6 +146,7 @@ int main(void) DEFINE(NVHE_INIT_HCR_EL2, offsetof(struct kvm_nvhe_init_params, hcr_el2)); DEFINE(NVHE_INIT_VTTBR, offsetof(struct kvm_nvhe_init_params, vttbr)); DEFINE(NVHE_INIT_VTCR, offsetof(struct kvm_nvhe_init_params, vtcr)); + DEFINE(NVHE_INIT_TMP, offsetof(struct kvm_nvhe_init_params, tmp)); #endif #ifdef CONFIG_CPU_PM DEFINE(CPU_CTX_SP, offsetof(struct cpu_suspend_ctx, sp)); diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c index dfefbdf4073a..a78f247029ae 100644 --- a/arch/arm64/kernel/cpu_errata.c +++ b/arch/arm64/kernel/cpu_errata.c @@ -439,6 +439,7 @@ static const struct midr_range erratum_spec_ssbs_list[] = { MIDR_ALL_VERSIONS(MIDR_CORTEX_A78), MIDR_ALL_VERSIONS(MIDR_CORTEX_A78C), MIDR_ALL_VERSIONS(MIDR_CORTEX_A710), + MIDR_ALL_VERSIONS(MIDR_CORTEX_A715), MIDR_ALL_VERSIONS(MIDR_CORTEX_A720), MIDR_ALL_VERSIONS(MIDR_CORTEX_A725), MIDR_ALL_VERSIONS(MIDR_CORTEX_X1), @@ -447,8 +448,10 @@ static const struct midr_range erratum_spec_ssbs_list[] = { MIDR_ALL_VERSIONS(MIDR_CORTEX_X3), MIDR_ALL_VERSIONS(MIDR_CORTEX_X4), MIDR_ALL_VERSIONS(MIDR_CORTEX_X925), + MIDR_ALL_VERSIONS(MIDR_MICROSOFT_AZURE_COBALT_100), MIDR_ALL_VERSIONS(MIDR_NEOVERSE_N1), MIDR_ALL_VERSIONS(MIDR_NEOVERSE_N2), + MIDR_ALL_VERSIONS(MIDR_NEOVERSE_N3), MIDR_ALL_VERSIONS(MIDR_NEOVERSE_V1), MIDR_ALL_VERSIONS(MIDR_NEOVERSE_V2), MIDR_ALL_VERSIONS(MIDR_NEOVERSE_V3), diff --git a/arch/arm64/kernel/probes/decode-insn.c b/arch/arm64/kernel/probes/decode-insn.c index 968d5fffe233..3496d6169e59 100644 --- a/arch/arm64/kernel/probes/decode-insn.c +++ b/arch/arm64/kernel/probes/decode-insn.c @@ -99,10 +99,6 @@ arm_probe_decode_insn(probe_opcode_t insn, struct arch_probe_insn *api) aarch64_insn_is_blr(insn) || aarch64_insn_is_ret(insn)) { api->handler = simulate_br_blr_ret; - } else if (aarch64_insn_is_ldr_lit(insn)) { - api->handler = simulate_ldr_literal; - } else if (aarch64_insn_is_ldrsw_lit(insn)) { - api->handler = simulate_ldrsw_literal; } else { /* * Instruction cannot be stepped out-of-line and we don't @@ -140,6 +136,17 @@ arm_kprobe_decode_insn(kprobe_opcode_t *addr, struct arch_specific_insn *asi) probe_opcode_t insn = le32_to_cpu(*addr); probe_opcode_t *scan_end = NULL; unsigned long size = 0, offset = 0; + struct arch_probe_insn *api = &asi->api; + + if (aarch64_insn_is_ldr_lit(insn)) { + api->handler = simulate_ldr_literal; + decoded = INSN_GOOD_NO_SLOT; + } else if (aarch64_insn_is_ldrsw_lit(insn)) { + api->handler = simulate_ldrsw_literal; + decoded = INSN_GOOD_NO_SLOT; + } else { + decoded = arm_probe_decode_insn(insn, &asi->api); + } /* * If there's a symbol defined in front of and near enough to @@ -157,7 +164,6 @@ arm_kprobe_decode_insn(kprobe_opcode_t *addr, struct arch_specific_insn *asi) else scan_end = addr - MAX_ATOMIC_CONTEXT_SIZE; } - decoded = arm_probe_decode_insn(insn, &asi->api); if (decoded != INSN_REJECTED && scan_end) if (is_probed_address_atomic(addr - 1, scan_end)) diff --git a/arch/arm64/kernel/probes/simulate-insn.c b/arch/arm64/kernel/probes/simulate-insn.c index 22d0b3252476..b65334ab79d2 100644 --- a/arch/arm64/kernel/probes/simulate-insn.c +++ b/arch/arm64/kernel/probes/simulate-insn.c @@ -171,17 +171,15 @@ simulate_tbz_tbnz(u32 opcode, long addr, struct pt_regs *regs) void __kprobes simulate_ldr_literal(u32 opcode, long addr, struct pt_regs *regs) { - u64 *load_addr; + unsigned long load_addr; int xn = opcode & 0x1f; - int disp; - disp = ldr_displacement(opcode); - load_addr = (u64 *) (addr + disp); + load_addr = addr + ldr_displacement(opcode); if (opcode & (1 << 30)) /* x0-x30 */ - set_x_reg(regs, xn, *load_addr); + set_x_reg(regs, xn, READ_ONCE(*(u64 *)load_addr)); else /* w0-w30 */ - set_w_reg(regs, xn, *load_addr); + set_w_reg(regs, xn, READ_ONCE(*(u32 *)load_addr)); instruction_pointer_set(regs, instruction_pointer(regs) + 4); } @@ -189,14 +187,12 @@ simulate_ldr_literal(u32 opcode, long addr, struct pt_regs *regs) void __kprobes simulate_ldrsw_literal(u32 opcode, long addr, struct pt_regs *regs) { - s32 *load_addr; + unsigned long load_addr; int xn = opcode & 0x1f; - int disp; - disp = ldr_displacement(opcode); - load_addr = (s32 *) (addr + disp); + load_addr = addr + ldr_displacement(opcode); - set_x_reg(regs, xn, *load_addr); + set_x_reg(regs, xn, READ_ONCE(*(s32 *)load_addr)); instruction_pointer_set(regs, instruction_pointer(regs) + 4); } diff --git a/arch/arm64/kernel/probes/uprobes.c b/arch/arm64/kernel/probes/uprobes.c index d49aef2657cd..a2f137a595fc 100644 --- a/arch/arm64/kernel/probes/uprobes.c +++ b/arch/arm64/kernel/probes/uprobes.c @@ -42,7 +42,7 @@ int arch_uprobe_analyze_insn(struct arch_uprobe *auprobe, struct mm_struct *mm, else if (!IS_ALIGNED(addr, AARCH64_INSN_SIZE)) return -EINVAL; - insn = *(probe_opcode_t *)(&auprobe->insn[0]); + insn = le32_to_cpu(auprobe->insn); switch (arm_probe_decode_insn(insn, &auprobe->api)) { case INSN_REJECTED: @@ -108,7 +108,7 @@ bool arch_uprobe_skip_sstep(struct arch_uprobe *auprobe, struct pt_regs *regs) if (!auprobe->simulate) return false; - insn = *(probe_opcode_t *)(&auprobe->insn[0]); + insn = le32_to_cpu(auprobe->insn); addr = instruction_pointer(regs); if (auprobe->api.handler) diff --git a/arch/arm64/kernel/process.c b/arch/arm64/kernel/process.c index 0540653fbf38..3e7c8c8195c3 100644 --- a/arch/arm64/kernel/process.c +++ b/arch/arm64/kernel/process.c @@ -412,6 +412,9 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args) p->thread.cpu_context.x19 = (unsigned long)args->fn; p->thread.cpu_context.x20 = (unsigned long)args->fn_arg; + + if (system_supports_poe()) + p->thread.por_el0 = POR_EL0_INIT; } p->thread.cpu_context.pc = (unsigned long)ret_from_fork; p->thread.cpu_context.sp = (unsigned long)childregs; diff --git a/arch/arm64/kernel/signal.c b/arch/arm64/kernel/signal.c index 561986947530..c7d311d8b92a 100644 --- a/arch/arm64/kernel/signal.c +++ b/arch/arm64/kernel/signal.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -66,10 +67,63 @@ struct rt_sigframe_user_layout { unsigned long end_offset; }; +/* + * Holds any EL0-controlled state that influences unprivileged memory accesses. + * This includes both accesses done in userspace and uaccess done in the kernel. + * + * This state needs to be carefully managed to ensure that it doesn't cause + * uaccess to fail when setting up the signal frame, and the signal handler + * itself also expects a well-defined state when entered. + */ +struct user_access_state { + u64 por_el0; +}; + #define BASE_SIGFRAME_SIZE round_up(sizeof(struct rt_sigframe), 16) #define TERMINATOR_SIZE round_up(sizeof(struct _aarch64_ctx), 16) #define EXTRA_CONTEXT_SIZE round_up(sizeof(struct extra_context), 16) +/* + * Save the user access state into ua_state and reset it to disable any + * restrictions. + */ +static void save_reset_user_access_state(struct user_access_state *ua_state) +{ + if (system_supports_poe()) { + u64 por_enable_all = 0; + + for (int pkey = 0; pkey < arch_max_pkey(); pkey++) + por_enable_all |= POE_RXW << (pkey * POR_BITS_PER_PKEY); + + ua_state->por_el0 = read_sysreg_s(SYS_POR_EL0); + write_sysreg_s(por_enable_all, SYS_POR_EL0); + /* Ensure that any subsequent uaccess observes the updated value */ + isb(); + } +} + +/* + * Set the user access state for invoking the signal handler. + * + * No uaccess should be done after that function is called. + */ +static void set_handler_user_access_state(void) +{ + if (system_supports_poe()) + write_sysreg_s(POR_EL0_INIT, SYS_POR_EL0); +} + +/* + * Restore the user access state to the values saved in ua_state. + * + * No uaccess should be done after that function is called. + */ +static void restore_user_access_state(const struct user_access_state *ua_state) +{ + if (system_supports_poe()) + write_sysreg_s(ua_state->por_el0, SYS_POR_EL0); +} + static void init_user_layout(struct rt_sigframe_user_layout *user) { const size_t reserved_size = @@ -261,18 +315,20 @@ static int restore_fpmr_context(struct user_ctxs *user) return err; } -static int preserve_poe_context(struct poe_context __user *ctx) +static int preserve_poe_context(struct poe_context __user *ctx, + const struct user_access_state *ua_state) { int err = 0; __put_user_error(POE_MAGIC, &ctx->head.magic, err); __put_user_error(sizeof(*ctx), &ctx->head.size, err); - __put_user_error(read_sysreg_s(SYS_POR_EL0), &ctx->por_el0, err); + __put_user_error(ua_state->por_el0, &ctx->por_el0, err); return err; } -static int restore_poe_context(struct user_ctxs *user) +static int restore_poe_context(struct user_ctxs *user, + struct user_access_state *ua_state) { u64 por_el0; int err = 0; @@ -282,7 +338,7 @@ static int restore_poe_context(struct user_ctxs *user) __get_user_error(por_el0, &(user->poe->por_el0), err); if (!err) - write_sysreg_s(por_el0, SYS_POR_EL0); + ua_state->por_el0 = por_el0; return err; } @@ -850,7 +906,8 @@ invalid: } static int restore_sigframe(struct pt_regs *regs, - struct rt_sigframe __user *sf) + struct rt_sigframe __user *sf, + struct user_access_state *ua_state) { sigset_t set; int i, err; @@ -899,7 +956,7 @@ static int restore_sigframe(struct pt_regs *regs, err = restore_zt_context(&user); if (err == 0 && system_supports_poe() && user.poe) - err = restore_poe_context(&user); + err = restore_poe_context(&user, ua_state); return err; } @@ -908,6 +965,7 @@ SYSCALL_DEFINE0(rt_sigreturn) { struct pt_regs *regs = current_pt_regs(); struct rt_sigframe __user *frame; + struct user_access_state ua_state; /* Always make any pending restarted system calls return -EINTR */ current->restart_block.fn = do_no_restart_syscall; @@ -924,12 +982,14 @@ SYSCALL_DEFINE0(rt_sigreturn) if (!access_ok(frame, sizeof (*frame))) goto badframe; - if (restore_sigframe(regs, frame)) + if (restore_sigframe(regs, frame, &ua_state)) goto badframe; if (restore_altstack(&frame->uc.uc_stack)) goto badframe; + restore_user_access_state(&ua_state); + return regs->regs[0]; badframe: @@ -1035,7 +1095,8 @@ static int setup_sigframe_layout(struct rt_sigframe_user_layout *user, } static int setup_sigframe(struct rt_sigframe_user_layout *user, - struct pt_regs *regs, sigset_t *set) + struct pt_regs *regs, sigset_t *set, + const struct user_access_state *ua_state) { int i, err = 0; struct rt_sigframe __user *sf = user->sigframe; @@ -1097,10 +1158,9 @@ static int setup_sigframe(struct rt_sigframe_user_layout *user, struct poe_context __user *poe_ctx = apply_user_offset(user, user->poe_offset); - err |= preserve_poe_context(poe_ctx); + err |= preserve_poe_context(poe_ctx, ua_state); } - /* ZA state if present */ if (system_supports_sme() && err == 0 && user->za_offset) { struct za_context __user *za_ctx = @@ -1237,9 +1297,6 @@ static void setup_return(struct pt_regs *regs, struct k_sigaction *ka, sme_smstop(); } - if (system_supports_poe()) - write_sysreg_s(POR_EL0_INIT, SYS_POR_EL0); - if (ka->sa.sa_flags & SA_RESTORER) sigtramp = ka->sa.sa_restorer; else @@ -1253,6 +1310,7 @@ static int setup_rt_frame(int usig, struct ksignal *ksig, sigset_t *set, { struct rt_sigframe_user_layout user; struct rt_sigframe __user *frame; + struct user_access_state ua_state; int err = 0; fpsimd_signal_preserve_current_state(); @@ -1260,13 +1318,14 @@ static int setup_rt_frame(int usig, struct ksignal *ksig, sigset_t *set, if (get_sigframe(&user, ksig, regs)) return 1; + save_reset_user_access_state(&ua_state); frame = user.sigframe; __put_user_error(0, &frame->uc.uc_flags, err); __put_user_error(NULL, &frame->uc.uc_link, err); err |= __save_altstack(&frame->uc.uc_stack, regs->sp); - err |= setup_sigframe(&user, regs, set); + err |= setup_sigframe(&user, regs, set, &ua_state); if (err == 0) { setup_return(regs, &ksig->ka, &user, usig); if (ksig->ka.sa.sa_flags & SA_SIGINFO) { @@ -1276,6 +1335,11 @@ static int setup_rt_frame(int usig, struct ksignal *ksig, sigset_t *set, } } + if (err == 0) + set_handler_user_access_state(); + else + restore_user_access_state(&ua_state); + return err; } diff --git a/arch/arm64/kvm/arm.c b/arch/arm64/kvm/arm.c index a0d01c46e408..48cafb65d6ac 100644 --- a/arch/arm64/kvm/arm.c +++ b/arch/arm64/kvm/arm.c @@ -997,6 +997,9 @@ static int kvm_vcpu_suspend(struct kvm_vcpu *vcpu) static int check_vcpu_requests(struct kvm_vcpu *vcpu) { if (kvm_request_pending(vcpu)) { + if (kvm_check_request(KVM_REQ_VM_DEAD, vcpu)) + return -EIO; + if (kvm_check_request(KVM_REQ_SLEEP, vcpu)) kvm_vcpu_sleep(vcpu); @@ -1031,6 +1034,8 @@ static int check_vcpu_requests(struct kvm_vcpu *vcpu) if (kvm_dirty_ring_check_request(vcpu)) return 0; + + check_nested_vcpu_requests(vcpu); } return 1; diff --git a/arch/arm64/kvm/hyp/include/hyp/switch.h b/arch/arm64/kvm/hyp/include/hyp/switch.h index 46d52e8a3df3..5310fe1da616 100644 --- a/arch/arm64/kvm/hyp/include/hyp/switch.h +++ b/arch/arm64/kvm/hyp/include/hyp/switch.h @@ -338,7 +338,7 @@ static inline void __hyp_sve_save_host(void) struct cpu_sve_state *sve_state = *host_data_ptr(sve_state); sve_state->zcr_el1 = read_sysreg_el1(SYS_ZCR); - write_sysreg_s(ZCR_ELx_LEN_MASK, SYS_ZCR_EL2); + write_sysreg_s(sve_vq_from_vl(kvm_host_sve_max_vl) - 1, SYS_ZCR_EL2); __sve_save_state(sve_state->sve_regs + sve_ffr_offset(kvm_host_sve_max_vl), &sve_state->fpsr, true); diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-init.S b/arch/arm64/kvm/hyp/nvhe/hyp-init.S index 401af1835be6..fc1866226067 100644 --- a/arch/arm64/kvm/hyp/nvhe/hyp-init.S +++ b/arch/arm64/kvm/hyp/nvhe/hyp-init.S @@ -24,28 +24,25 @@ .align 11 SYM_CODE_START(__kvm_hyp_init) - ventry __invalid // Synchronous EL2t - ventry __invalid // IRQ EL2t - ventry __invalid // FIQ EL2t - ventry __invalid // Error EL2t + ventry . // Synchronous EL2t + ventry . // IRQ EL2t + ventry . // FIQ EL2t + ventry . // Error EL2t - ventry __invalid // Synchronous EL2h - ventry __invalid // IRQ EL2h - ventry __invalid // FIQ EL2h - ventry __invalid // Error EL2h + ventry . // Synchronous EL2h + ventry . // IRQ EL2h + ventry . // FIQ EL2h + ventry . // Error EL2h ventry __do_hyp_init // Synchronous 64-bit EL1 - ventry __invalid // IRQ 64-bit EL1 - ventry __invalid // FIQ 64-bit EL1 - ventry __invalid // Error 64-bit EL1 + ventry . // IRQ 64-bit EL1 + ventry . // FIQ 64-bit EL1 + ventry . // Error 64-bit EL1 - ventry __invalid // Synchronous 32-bit EL1 - ventry __invalid // IRQ 32-bit EL1 - ventry __invalid // FIQ 32-bit EL1 - ventry __invalid // Error 32-bit EL1 - -__invalid: - b . + ventry . // Synchronous 32-bit EL1 + ventry . // IRQ 32-bit EL1 + ventry . // FIQ 32-bit EL1 + ventry . // Error 32-bit EL1 /* * Only uses x0..x3 so as to not clobber callee-saved SMCCC registers. @@ -76,6 +73,13 @@ __do_hyp_init: eret SYM_CODE_END(__kvm_hyp_init) +SYM_CODE_START_LOCAL(__kvm_init_el2_state) + /* Initialize EL2 CPU state to sane values. */ + init_el2_state // Clobbers x0..x2 + finalise_el2_state + ret +SYM_CODE_END(__kvm_init_el2_state) + /* * Initialize the hypervisor in EL2. * @@ -102,9 +106,12 @@ SYM_CODE_START_LOCAL(___kvm_hyp_init) // TPIDR_EL2 is used to preserve x0 across the macro maze... isb msr tpidr_el2, x0 - init_el2_state - finalise_el2_state + str lr, [x0, #NVHE_INIT_TMP] + + bl __kvm_init_el2_state + mrs x0, tpidr_el2 + ldr lr, [x0, #NVHE_INIT_TMP] 1: ldr x1, [x0, #NVHE_INIT_TPIDR_EL2] @@ -199,9 +206,8 @@ SYM_CODE_START_LOCAL(__kvm_hyp_init_cpu) 2: msr SPsel, #1 // We want to use SP_EL{1,2} - /* Initialize EL2 CPU state to sane values. */ - init_el2_state // Clobbers x0..x2 - finalise_el2_state + bl __kvm_init_el2_state + __init_el2_nvhe_prepare_eret /* Enable MMU, set vectors and stack. */ diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-main.c b/arch/arm64/kvm/hyp/nvhe/hyp-main.c index 87692b566d90..fefc89209f9e 100644 --- a/arch/arm64/kvm/hyp/nvhe/hyp-main.c +++ b/arch/arm64/kvm/hyp/nvhe/hyp-main.c @@ -33,7 +33,7 @@ static void __hyp_sve_save_guest(struct kvm_vcpu *vcpu) */ sve_cond_update_zcr_vq(vcpu_sve_max_vq(vcpu) - 1, SYS_ZCR_EL2); __sve_save_state(vcpu_sve_pffr(vcpu), &vcpu->arch.ctxt.fp_regs.fpsr, true); - write_sysreg_s(ZCR_ELx_LEN_MASK, SYS_ZCR_EL2); + write_sysreg_s(sve_vq_from_vl(kvm_host_sve_max_vl) - 1, SYS_ZCR_EL2); } static void __hyp_sve_restore_host(void) @@ -45,10 +45,11 @@ static void __hyp_sve_restore_host(void) * the host. The layout of the data when saving the sve state depends * on the VL, so use a consistent (i.e., the maximum) host VL. * - * Setting ZCR_EL2 to ZCR_ELx_LEN_MASK sets the effective length - * supported by the system (or limited at EL3). + * Note that this constrains the PE to the maximum shared VL + * that was discovered, if we wish to use larger VLs this will + * need to be revisited. */ - write_sysreg_s(ZCR_ELx_LEN_MASK, SYS_ZCR_EL2); + write_sysreg_s(sve_vq_from_vl(kvm_host_sve_max_vl) - 1, SYS_ZCR_EL2); __sve_restore_state(sve_state->sve_regs + sve_ffr_offset(kvm_host_sve_max_vl), &sve_state->fpsr, true); @@ -488,7 +489,8 @@ void handle_trap(struct kvm_cpu_context *host_ctxt) case ESR_ELx_EC_SVE: cpacr_clear_set(0, CPACR_ELx_ZEN); isb(); - sve_cond_update_zcr_vq(ZCR_ELx_LEN_MASK, SYS_ZCR_EL2); + sve_cond_update_zcr_vq(sve_vq_from_vl(kvm_host_sve_max_vl) - 1, + SYS_ZCR_EL2); break; case ESR_ELx_EC_IABT_LOW: case ESR_ELx_EC_DABT_LOW: diff --git a/arch/arm64/kvm/hyp/nvhe/pkvm.c b/arch/arm64/kvm/hyp/nvhe/pkvm.c index 187a5f4d56c0..077d4098548d 100644 --- a/arch/arm64/kvm/hyp/nvhe/pkvm.c +++ b/arch/arm64/kvm/hyp/nvhe/pkvm.c @@ -574,12 +574,14 @@ int __pkvm_init_vcpu(pkvm_handle_t handle, struct kvm_vcpu *host_vcpu, unlock: hyp_spin_unlock(&vm_table_lock); - if (ret) + if (ret) { unmap_donated_memory(hyp_vcpu, sizeof(*hyp_vcpu)); + return ret; + } hyp_vcpu->vcpu.arch.cptr_el2 = kvm_get_reset_cptr_el2(&hyp_vcpu->vcpu); - return ret; + return 0; } static void diff --git a/arch/arm64/kvm/hypercalls.c b/arch/arm64/kvm/hypercalls.c index 5763d979d8ca..ee6573befb81 100644 --- a/arch/arm64/kvm/hypercalls.c +++ b/arch/arm64/kvm/hypercalls.c @@ -317,7 +317,7 @@ int kvm_smccc_call_handler(struct kvm_vcpu *vcpu) * to the guest, and hide SSBS so that the * guest stays protected. */ - if (cpus_have_final_cap(ARM64_SSBS)) + if (kvm_has_feat(vcpu->kvm, ID_AA64PFR1_EL1, SSBS, IMP)) break; fallthrough; case SPECTRE_UNAFFECTED: @@ -428,7 +428,7 @@ int kvm_arm_copy_fw_reg_indices(struct kvm_vcpu *vcpu, u64 __user *uindices) * Convert the workaround level into an easy-to-compare number, where higher * values mean better protection. */ -static int get_kernel_wa_level(u64 regid) +static int get_kernel_wa_level(struct kvm_vcpu *vcpu, u64 regid) { switch (regid) { case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1: @@ -449,7 +449,7 @@ static int get_kernel_wa_level(u64 regid) * don't have any FW mitigation if SSBS is there at * all times. */ - if (cpus_have_final_cap(ARM64_SSBS)) + if (kvm_has_feat(vcpu->kvm, ID_AA64PFR1_EL1, SSBS, IMP)) return KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2_NOT_AVAIL; fallthrough; case SPECTRE_UNAFFECTED: @@ -486,7 +486,7 @@ int kvm_arm_get_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg) case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_1: case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_2: case KVM_REG_ARM_SMCCC_ARCH_WORKAROUND_3: - val = get_kernel_wa_level(reg->id) & KVM_REG_FEATURE_LEVEL_MASK; + val = get_kernel_wa_level(vcpu, reg->id) & KVM_REG_FEATURE_LEVEL_MASK; break; case KVM_REG_ARM_STD_BMAP: val = READ_ONCE(smccc_feat->std_bmap); @@ -588,7 +588,7 @@ int kvm_arm_set_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg) if (val & ~KVM_REG_FEATURE_LEVEL_MASK) return -EINVAL; - if (get_kernel_wa_level(reg->id) < val) + if (get_kernel_wa_level(vcpu, reg->id) < val) return -EINVAL; return 0; @@ -624,7 +624,7 @@ int kvm_arm_set_fw_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg) * We can deal with NOT_AVAIL on NOT_REQUIRED, but not the * other way around. */ - if (get_kernel_wa_level(reg->id) < wa_level) + if (get_kernel_wa_level(vcpu, reg->id) < wa_level) return -EINVAL; return 0; diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index a509b63bd4dd..0f7658aefa1a 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -328,9 +328,10 @@ static void __unmap_stage2_range(struct kvm_s2_mmu *mmu, phys_addr_t start, u64 may_block)); } -void kvm_stage2_unmap_range(struct kvm_s2_mmu *mmu, phys_addr_t start, u64 size) +void kvm_stage2_unmap_range(struct kvm_s2_mmu *mmu, phys_addr_t start, + u64 size, bool may_block) { - __unmap_stage2_range(mmu, start, size, true); + __unmap_stage2_range(mmu, start, size, may_block); } void kvm_stage2_flush_range(struct kvm_s2_mmu *mmu, phys_addr_t addr, phys_addr_t end) @@ -1015,7 +1016,7 @@ static void stage2_unmap_memslot(struct kvm *kvm, if (!(vma->vm_flags & VM_PFNMAP)) { gpa_t gpa = addr + (vm_start - memslot->userspace_addr); - kvm_stage2_unmap_range(&kvm->arch.mmu, gpa, vm_end - vm_start); + kvm_stage2_unmap_range(&kvm->arch.mmu, gpa, vm_end - vm_start, true); } hva = vm_end; } while (hva < reg_end); @@ -1042,7 +1043,7 @@ void stage2_unmap_vm(struct kvm *kvm) kvm_for_each_memslot(memslot, bkt, slots) stage2_unmap_memslot(kvm, memslot); - kvm_nested_s2_unmap(kvm); + kvm_nested_s2_unmap(kvm, true); write_unlock(&kvm->mmu_lock); mmap_read_unlock(current->mm); @@ -1912,7 +1913,7 @@ bool kvm_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range) (range->end - range->start) << PAGE_SHIFT, range->may_block); - kvm_nested_s2_unmap(kvm); + kvm_nested_s2_unmap(kvm, range->may_block); return false; } @@ -2179,8 +2180,8 @@ void kvm_arch_flush_shadow_memslot(struct kvm *kvm, phys_addr_t size = slot->npages << PAGE_SHIFT; write_lock(&kvm->mmu_lock); - kvm_stage2_unmap_range(&kvm->arch.mmu, gpa, size); - kvm_nested_s2_unmap(kvm); + kvm_stage2_unmap_range(&kvm->arch.mmu, gpa, size, true); + kvm_nested_s2_unmap(kvm, true); write_unlock(&kvm->mmu_lock); } diff --git a/arch/arm64/kvm/nested.c b/arch/arm64/kvm/nested.c index f9e30dd34c7a..c4b17d90fc49 100644 --- a/arch/arm64/kvm/nested.c +++ b/arch/arm64/kvm/nested.c @@ -632,9 +632,9 @@ static struct kvm_s2_mmu *get_s2_mmu_nested(struct kvm_vcpu *vcpu) /* Set the scene for the next search */ kvm->arch.nested_mmus_next = (i + 1) % kvm->arch.nested_mmus_size; - /* Clear the old state */ + /* Make sure we don't forget to do the laundry */ if (kvm_s2_mmu_valid(s2_mmu)) - kvm_stage2_unmap_range(s2_mmu, 0, kvm_phys_size(s2_mmu)); + s2_mmu->pending_unmap = true; /* * The virtual VMID (modulo CnP) will be used as a key when matching @@ -650,6 +650,16 @@ static struct kvm_s2_mmu *get_s2_mmu_nested(struct kvm_vcpu *vcpu) out: atomic_inc(&s2_mmu->refcnt); + + /* + * Set the vCPU request to perform an unmap, even if the pending unmap + * originates from another vCPU. This guarantees that the MMU has been + * completely unmapped before any vCPU actually uses it, and allows + * multiple vCPUs to lend a hand with completing the unmap. + */ + if (s2_mmu->pending_unmap) + kvm_make_request(KVM_REQ_NESTED_S2_UNMAP, vcpu); + return s2_mmu; } @@ -663,6 +673,13 @@ void kvm_init_nested_s2_mmu(struct kvm_s2_mmu *mmu) void kvm_vcpu_load_hw_mmu(struct kvm_vcpu *vcpu) { + /* + * The vCPU kept its reference on the MMU after the last put, keep + * rolling with it. + */ + if (vcpu->arch.hw_mmu) + return; + if (is_hyp_ctxt(vcpu)) { vcpu->arch.hw_mmu = &vcpu->kvm->arch.mmu; } else { @@ -674,10 +691,18 @@ void kvm_vcpu_load_hw_mmu(struct kvm_vcpu *vcpu) void kvm_vcpu_put_hw_mmu(struct kvm_vcpu *vcpu) { - if (kvm_is_nested_s2_mmu(vcpu->kvm, vcpu->arch.hw_mmu)) { + /* + * Keep a reference on the associated stage-2 MMU if the vCPU is + * scheduling out and not in WFI emulation, suggesting it is likely to + * reuse the MMU sometime soon. + */ + if (vcpu->scheduled_out && !vcpu_get_flag(vcpu, IN_WFI)) + return; + + if (kvm_is_nested_s2_mmu(vcpu->kvm, vcpu->arch.hw_mmu)) atomic_dec(&vcpu->arch.hw_mmu->refcnt); - vcpu->arch.hw_mmu = NULL; - } + + vcpu->arch.hw_mmu = NULL; } /* @@ -730,7 +755,7 @@ void kvm_nested_s2_wp(struct kvm *kvm) } } -void kvm_nested_s2_unmap(struct kvm *kvm) +void kvm_nested_s2_unmap(struct kvm *kvm, bool may_block) { int i; @@ -740,7 +765,7 @@ void kvm_nested_s2_unmap(struct kvm *kvm) struct kvm_s2_mmu *mmu = &kvm->arch.nested_mmus[i]; if (kvm_s2_mmu_valid(mmu)) - kvm_stage2_unmap_range(mmu, 0, kvm_phys_size(mmu)); + kvm_stage2_unmap_range(mmu, 0, kvm_phys_size(mmu), may_block); } } @@ -1184,3 +1209,17 @@ int kvm_init_nv_sysregs(struct kvm *kvm) return 0; } + +void check_nested_vcpu_requests(struct kvm_vcpu *vcpu) +{ + if (kvm_check_request(KVM_REQ_NESTED_S2_UNMAP, vcpu)) { + struct kvm_s2_mmu *mmu = vcpu->arch.hw_mmu; + + write_lock(&vcpu->kvm->mmu_lock); + if (mmu->pending_unmap) { + kvm_stage2_unmap_range(mmu, 0, kvm_phys_size(mmu), true); + mmu->pending_unmap = false; + } + write_unlock(&vcpu->kvm->mmu_lock); + } +} diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index dad88e31f953..ff8c4e1b847e 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -1527,6 +1527,14 @@ static u64 __kvm_read_sanitised_id_reg(const struct kvm_vcpu *vcpu, val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_EL1_MTE); val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_EL1_SME); + val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_EL1_RNDR_trap); + val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_EL1_NMI); + val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_EL1_MTE_frac); + val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_EL1_GCS); + val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_EL1_THE); + val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_EL1_MTEX); + val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_EL1_DF2); + val &= ~ARM64_FEATURE_MASK(ID_AA64PFR1_EL1_PFAR); break; case SYS_ID_AA64PFR2_EL1: /* We only expose FPMR */ @@ -1550,7 +1558,8 @@ static u64 __kvm_read_sanitised_id_reg(const struct kvm_vcpu *vcpu, val &= ~ID_AA64MMFR2_EL1_CCIDX_MASK; break; case SYS_ID_AA64MMFR3_EL1: - val &= ID_AA64MMFR3_EL1_TCRX | ID_AA64MMFR3_EL1_S1POE; + val &= ID_AA64MMFR3_EL1_TCRX | ID_AA64MMFR3_EL1_S1POE | + ID_AA64MMFR3_EL1_S1PIE; break; case SYS_ID_MMFR4_EL1: val &= ~ARM64_FEATURE_MASK(ID_MMFR4_EL1_CCIDX); @@ -1985,7 +1994,7 @@ static u64 reset_clidr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r) * one cache line. */ if (kvm_has_mte(vcpu->kvm)) - clidr |= 2 << CLIDR_TTYPE_SHIFT(loc); + clidr |= 2ULL << CLIDR_TTYPE_SHIFT(loc); __vcpu_sys_reg(vcpu, r->reg) = clidr; @@ -2376,7 +2385,19 @@ static const struct sys_reg_desc sys_reg_descs[] = { ID_AA64PFR0_EL1_RAS | ID_AA64PFR0_EL1_AdvSIMD | ID_AA64PFR0_EL1_FP), }, - ID_SANITISED(ID_AA64PFR1_EL1), + ID_WRITABLE(ID_AA64PFR1_EL1, ~(ID_AA64PFR1_EL1_PFAR | + ID_AA64PFR1_EL1_DF2 | + ID_AA64PFR1_EL1_MTEX | + ID_AA64PFR1_EL1_THE | + ID_AA64PFR1_EL1_GCS | + ID_AA64PFR1_EL1_MTE_frac | + ID_AA64PFR1_EL1_NMI | + ID_AA64PFR1_EL1_RNDR_trap | + ID_AA64PFR1_EL1_SME | + ID_AA64PFR1_EL1_RES0 | + ID_AA64PFR1_EL1_MPAM_frac | + ID_AA64PFR1_EL1_RAS_frac | + ID_AA64PFR1_EL1_MTE)), ID_WRITABLE(ID_AA64PFR2_EL1, ID_AA64PFR2_EL1_FPMR), ID_UNALLOCATED(4,3), ID_WRITABLE(ID_AA64ZFR0_EL1, ~ID_AA64ZFR0_EL1_RES0), @@ -2390,7 +2411,21 @@ static const struct sys_reg_desc sys_reg_descs[] = { .get_user = get_id_reg, .set_user = set_id_aa64dfr0_el1, .reset = read_sanitised_id_aa64dfr0_el1, - .val = ID_AA64DFR0_EL1_PMUVer_MASK | + /* + * Prior to FEAT_Debugv8.9, the architecture defines context-aware + * breakpoints (CTX_CMPs) as the highest numbered breakpoints (BRPs). + * KVM does not trap + emulate the breakpoint registers, and as such + * cannot support a layout that misaligns with the underlying hardware. + * While it may be possible to describe a subset that aligns with + * hardware, just prevent changes to BRPs and CTX_CMPs altogether for + * simplicity. + * + * See DDI0487K.a, section D2.8.3 Breakpoint types and linking + * of breakpoints for more details. + */ + .val = ID_AA64DFR0_EL1_DoubleLock_MASK | + ID_AA64DFR0_EL1_WRPs_MASK | + ID_AA64DFR0_EL1_PMUVer_MASK | ID_AA64DFR0_EL1_DebugVer_MASK, }, ID_SANITISED(ID_AA64DFR1_EL1), ID_UNALLOCATED(5,2), @@ -2433,6 +2468,7 @@ static const struct sys_reg_desc sys_reg_descs[] = { ID_AA64MMFR2_EL1_NV | ID_AA64MMFR2_EL1_CCIDX)), ID_WRITABLE(ID_AA64MMFR3_EL1, (ID_AA64MMFR3_EL1_TCRX | + ID_AA64MMFR3_EL1_S1PIE | ID_AA64MMFR3_EL1_S1POE)), ID_SANITISED(ID_AA64MMFR4_EL1), ID_UNALLOCATED(7,5), @@ -2903,7 +2939,7 @@ static bool handle_alle1is(struct kvm_vcpu *vcpu, struct sys_reg_params *p, * Drop all shadow S2s, resulting in S1/S2 TLBIs for each of the * corresponding VMIDs. */ - kvm_nested_s2_unmap(vcpu->kvm); + kvm_nested_s2_unmap(vcpu->kvm, true); write_unlock(&vcpu->kvm->mmu_lock); @@ -2955,7 +2991,30 @@ union tlbi_info { static void s2_mmu_unmap_range(struct kvm_s2_mmu *mmu, const union tlbi_info *info) { - kvm_stage2_unmap_range(mmu, info->range.start, info->range.size); + /* + * The unmap operation is allowed to drop the MMU lock and block, which + * means that @mmu could be used for a different context than the one + * currently being invalidated. + * + * This behavior is still safe, as: + * + * 1) The vCPU(s) that recycled the MMU are responsible for invalidating + * the entire MMU before reusing it, which still honors the intent + * of a TLBI. + * + * 2) Until the guest TLBI instruction is 'retired' (i.e. increment PC + * and ERET to the guest), other vCPUs are allowed to use stale + * translations. + * + * 3) Accidentally unmapping an unrelated MMU context is nonfatal, and + * at worst may cause more aborts for shadow stage-2 fills. + * + * Dropping the MMU lock also implies that shadow stage-2 fills could + * happen behind the back of the TLBI. This is still safe, though, as + * the L1 needs to put its stage-2 in a consistent state before doing + * the TLBI. + */ + kvm_stage2_unmap_range(mmu, info->range.start, info->range.size, true); } static bool handle_vmalls12e1is(struct kvm_vcpu *vcpu, struct sys_reg_params *p, @@ -3050,7 +3109,11 @@ static void s2_mmu_unmap_ipa(struct kvm_s2_mmu *mmu, max_size = compute_tlb_inval_range(mmu, info->ipa.addr); base_addr &= ~(max_size - 1); - kvm_stage2_unmap_range(mmu, base_addr, max_size); + /* + * See comment in s2_mmu_unmap_range() for why this is allowed to + * reschedule. + */ + kvm_stage2_unmap_range(mmu, base_addr, max_size, true); } static bool handle_ipas2e1is(struct kvm_vcpu *vcpu, struct sys_reg_params *p, diff --git a/arch/arm64/kvm/vgic/vgic-init.c b/arch/arm64/kvm/vgic/vgic-init.c index e7c53e8af3d1..48c952563e85 100644 --- a/arch/arm64/kvm/vgic/vgic-init.c +++ b/arch/arm64/kvm/vgic/vgic-init.c @@ -417,8 +417,28 @@ static void __kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu) kfree(vgic_cpu->private_irqs); vgic_cpu->private_irqs = NULL; - if (vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) + if (vcpu->kvm->arch.vgic.vgic_model == KVM_DEV_TYPE_ARM_VGIC_V3) { + /* + * If this vCPU is being destroyed because of a failed creation + * then unregister the redistributor to avoid leaving behind a + * dangling pointer to the vCPU struct. + * + * vCPUs that have been successfully created (i.e. added to + * kvm->vcpu_array) get unregistered in kvm_vgic_destroy(), as + * this function gets called while holding kvm->arch.config_lock + * in the VM teardown path and would otherwise introduce a lock + * inversion w.r.t. kvm->srcu. + * + * vCPUs that failed creation are torn down outside of the + * kvm->arch.config_lock and do not get unregistered in + * kvm_vgic_destroy(), meaning it is both safe and necessary to + * do so here. + */ + if (kvm_get_vcpu_by_id(vcpu->kvm, vcpu->vcpu_id) != vcpu) + vgic_unregister_redist_iodev(vcpu); + vgic_cpu->rd_iodev.base_addr = VGIC_ADDR_UNDEF; + } } void kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu) @@ -524,22 +544,31 @@ int kvm_vgic_map_resources(struct kvm *kvm) if (ret) goto out; - dist->ready = true; dist_base = dist->vgic_dist_base; mutex_unlock(&kvm->arch.config_lock); ret = vgic_register_dist_iodev(kvm, dist_base, type); - if (ret) + if (ret) { kvm_err("Unable to register VGIC dist MMIO regions\n"); + goto out_slots; + } + /* + * kvm_io_bus_register_dev() guarantees all readers see the new MMIO + * registration before returning through synchronize_srcu(), which also + * implies a full memory barrier. As such, marking the distributor as + * 'ready' here is guaranteed to be ordered after all vCPUs having seen + * a completely configured distributor. + */ + dist->ready = true; goto out_slots; out: mutex_unlock(&kvm->arch.config_lock); out_slots: - mutex_unlock(&kvm->slots_lock); - if (ret) - kvm_vgic_destroy(kvm); + kvm_vm_dead(kvm); + + mutex_unlock(&kvm->slots_lock); return ret; } diff --git a/arch/arm64/kvm/vgic/vgic-kvm-device.c b/arch/arm64/kvm/vgic/vgic-kvm-device.c index 1d26bb5b02f4..5f4f57aaa23e 100644 --- a/arch/arm64/kvm/vgic/vgic-kvm-device.c +++ b/arch/arm64/kvm/vgic/vgic-kvm-device.c @@ -236,7 +236,12 @@ static int vgic_set_common_attr(struct kvm_device *dev, mutex_lock(&dev->kvm->arch.config_lock); - if (vgic_ready(dev->kvm) || dev->kvm->arch.vgic.nr_spis) + /* + * Either userspace has already configured NR_IRQS or + * the vgic has already been initialized and vgic_init() + * supplied a default amount of SPIs. + */ + if (dev->kvm->arch.vgic.nr_spis) ret = -EBUSY; else dev->kvm->arch.vgic.nr_spis = diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c index 8bbd0b20136a..5db82bfc9dc1 100644 --- a/arch/arm64/net/bpf_jit_comp.c +++ b/arch/arm64/net/bpf_jit_comp.c @@ -2220,7 +2220,11 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im, emit(A64_STR64I(A64_R(20), A64_SP, regs_off + 8), ctx); if (flags & BPF_TRAMP_F_CALL_ORIG) { - emit_a64_mov_i64(A64_R(0), (const u64)im, ctx); + /* for the first pass, assume the worst case */ + if (!ctx->image) + ctx->idx += 4; + else + emit_a64_mov_i64(A64_R(0), (const u64)im, ctx); emit_call((const u64)__bpf_tramp_enter, ctx); } @@ -2264,7 +2268,11 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im, if (flags & BPF_TRAMP_F_CALL_ORIG) { im->ip_epilogue = ctx->ro_image + ctx->idx; - emit_a64_mov_i64(A64_R(0), (const u64)im, ctx); + /* for the first pass, assume the worst case */ + if (!ctx->image) + ctx->idx += 4; + else + emit_a64_mov_i64(A64_R(0), (const u64)im, ctx); emit_call((const u64)__bpf_tramp_exit, ctx); } diff --git a/arch/loongarch/crypto/crc32-loongarch.c b/arch/loongarch/crypto/crc32-loongarch.c index 3eebea3a7b47..b7d9782827f5 100644 --- a/arch/loongarch/crypto/crc32-loongarch.c +++ b/arch/loongarch/crypto/crc32-loongarch.c @@ -13,7 +13,7 @@ #include #include -#include +#include #define _CRC32(crc, value, size, type) \ do { \ diff --git a/arch/loongarch/include/asm/bootinfo.h b/arch/loongarch/include/asm/bootinfo.h index 6d5846dd075c..7657e016233f 100644 --- a/arch/loongarch/include/asm/bootinfo.h +++ b/arch/loongarch/include/asm/bootinfo.h @@ -26,6 +26,10 @@ struct loongson_board_info { #define NR_WORDS DIV_ROUND_UP(NR_CPUS, BITS_PER_LONG) +/* + * The "core" of cores_per_node and cores_per_package stands for a + * logical core, which means in a SMT system it stands for a thread. + */ struct loongson_system_configuration { int nr_cpus; int nr_nodes; diff --git a/arch/loongarch/include/asm/kasan.h b/arch/loongarch/include/asm/kasan.h index cd6084f4e153..c6bce5fbff57 100644 --- a/arch/loongarch/include/asm/kasan.h +++ b/arch/loongarch/include/asm/kasan.h @@ -16,7 +16,7 @@ #define XRANGE_SHIFT (48) /* Valid address length */ -#define XRANGE_SHADOW_SHIFT (PGDIR_SHIFT + PAGE_SHIFT - 3) +#define XRANGE_SHADOW_SHIFT min(cpu_vabits, VA_BITS) /* Used for taking out the valid address */ #define XRANGE_SHADOW_MASK GENMASK_ULL(XRANGE_SHADOW_SHIFT - 1, 0) /* One segment whole address space size */ diff --git a/arch/loongarch/include/asm/loongarch.h b/arch/loongarch/include/asm/loongarch.h index 26542413a5b0..64ad277e096e 100644 --- a/arch/loongarch/include/asm/loongarch.h +++ b/arch/loongarch/include/asm/loongarch.h @@ -250,7 +250,7 @@ #define CSR_ESTAT_IS_WIDTH 15 #define CSR_ESTAT_IS (_ULCAST_(0x7fff) << CSR_ESTAT_IS_SHIFT) -#define LOONGARCH_CSR_ERA 0x6 /* ERA */ +#define LOONGARCH_CSR_ERA 0x6 /* Exception return address */ #define LOONGARCH_CSR_BADV 0x7 /* Bad virtual address */ diff --git a/arch/loongarch/include/asm/pgalloc.h b/arch/loongarch/include/asm/pgalloc.h index 4e2d6b7ca2ee..a7b9c9e73593 100644 --- a/arch/loongarch/include/asm/pgalloc.h +++ b/arch/loongarch/include/asm/pgalloc.h @@ -10,6 +10,7 @@ #define __HAVE_ARCH_PMD_ALLOC_ONE #define __HAVE_ARCH_PUD_ALLOC_ONE +#define __HAVE_ARCH_PTE_ALLOC_ONE_KERNEL #include static inline void pmd_populate_kernel(struct mm_struct *mm, @@ -44,6 +45,16 @@ extern void pagetable_init(void); extern pgd_t *pgd_alloc(struct mm_struct *mm); +static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm) +{ + pte_t *pte = __pte_alloc_one_kernel(mm); + + if (pte) + kernel_pte_init(pte); + + return pte; +} + #define __pte_free_tlb(tlb, pte, address) \ do { \ pagetable_pte_dtor(page_ptdesc(pte)); \ diff --git a/arch/loongarch/include/asm/pgtable.h b/arch/loongarch/include/asm/pgtable.h index 9965f52ef65b..20714b73f14c 100644 --- a/arch/loongarch/include/asm/pgtable.h +++ b/arch/loongarch/include/asm/pgtable.h @@ -269,6 +269,7 @@ extern void set_pmd_at(struct mm_struct *mm, unsigned long addr, pmd_t *pmdp, pm extern void pgd_init(void *addr); extern void pud_init(void *addr); extern void pmd_init(void *addr); +extern void kernel_pte_init(void *addr); /* * Encode/decode swap entries and swap PTEs. Swap PTEs are all PTEs that @@ -325,39 +326,17 @@ static inline void set_pte(pte_t *ptep, pte_t pteval) { WRITE_ONCE(*ptep, pteval); - if (pte_val(pteval) & _PAGE_GLOBAL) { - pte_t *buddy = ptep_buddy(ptep); - /* - * Make sure the buddy is global too (if it's !none, - * it better already be global) - */ - if (pte_none(ptep_get(buddy))) { #ifdef CONFIG_SMP - /* - * For SMP, multiple CPUs can race, so we need - * to do this atomically. - */ - __asm__ __volatile__( - __AMOR "$zero, %[global], %[buddy] \n" - : [buddy] "+ZB" (buddy->pte) - : [global] "r" (_PAGE_GLOBAL) - : "memory"); - - DBAR(0b11000); /* o_wrw = 0b11000 */ -#else /* !CONFIG_SMP */ - WRITE_ONCE(*buddy, __pte(pte_val(ptep_get(buddy)) | _PAGE_GLOBAL)); -#endif /* CONFIG_SMP */ - } - } + if (pte_val(pteval) & _PAGE_GLOBAL) + DBAR(0b11000); /* o_wrw = 0b11000 */ +#endif } static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { - /* Preserve global status for the pair */ - if (pte_val(ptep_get(ptep_buddy(ptep))) & _PAGE_GLOBAL) - set_pte(ptep, __pte(_PAGE_GLOBAL)); - else - set_pte(ptep, __pte(0)); + pte_t pte = ptep_get(ptep); + pte_val(pte) &= _PAGE_GLOBAL; + set_pte(ptep, pte); } #define PGD_T_LOG2 (__builtin_ffs(sizeof(pgd_t)) - 1) diff --git a/arch/loongarch/kernel/process.c b/arch/loongarch/kernel/process.c index f2ff8b5d591e..6e58f65455c7 100644 --- a/arch/loongarch/kernel/process.c +++ b/arch/loongarch/kernel/process.c @@ -293,13 +293,15 @@ unsigned long stack_top(void) { unsigned long top = TASK_SIZE & PAGE_MASK; - /* Space for the VDSO & data page */ - top -= PAGE_ALIGN(current->thread.vdso->size); - top -= VVAR_SIZE; + if (current->thread.vdso) { + /* Space for the VDSO & data page */ + top -= PAGE_ALIGN(current->thread.vdso->size); + top -= VVAR_SIZE; - /* Space to randomize the VDSO base */ - if (current->flags & PF_RANDOMIZE) - top -= VDSO_RANDOMIZE_SIZE; + /* Space to randomize the VDSO base */ + if (current->flags & PF_RANDOMIZE) + top -= VDSO_RANDOMIZE_SIZE; + } return top; } diff --git a/arch/loongarch/kernel/setup.c b/arch/loongarch/kernel/setup.c index 00e307203ddb..cbd3c09a93c1 100644 --- a/arch/loongarch/kernel/setup.c +++ b/arch/loongarch/kernel/setup.c @@ -55,6 +55,7 @@ #define SMBIOS_FREQHIGH_OFFSET 0x17 #define SMBIOS_FREQLOW_MASK 0xFF #define SMBIOS_CORE_PACKAGE_OFFSET 0x23 +#define SMBIOS_THREAD_PACKAGE_OFFSET 0x25 #define LOONGSON_EFI_ENABLE (1 << 3) unsigned long fw_arg0, fw_arg1, fw_arg2; @@ -125,7 +126,7 @@ static void __init parse_cpu_table(const struct dmi_header *dm) cpu_clock_freq = freq_temp * 1000000; loongson_sysconf.cpuname = (void *)dmi_string_parse(dm, dmi_data[16]); - loongson_sysconf.cores_per_package = *(dmi_data + SMBIOS_CORE_PACKAGE_OFFSET); + loongson_sysconf.cores_per_package = *(dmi_data + SMBIOS_THREAD_PACKAGE_OFFSET); pr_info("CpuClock = %llu\n", cpu_clock_freq); } diff --git a/arch/loongarch/kernel/traps.c b/arch/loongarch/kernel/traps.c index f9f4eb00c92e..c57b4134f3e8 100644 --- a/arch/loongarch/kernel/traps.c +++ b/arch/loongarch/kernel/traps.c @@ -555,6 +555,9 @@ asmlinkage void noinstr do_ale(struct pt_regs *regs) #else unsigned int *pc; + if (regs->csr_prmd & CSR_PRMD_PIE) + local_irq_enable(); + perf_sw_event(PERF_COUNT_SW_ALIGNMENT_FAULTS, 1, regs, regs->csr_badvaddr); /* @@ -579,6 +582,8 @@ sigbus: die_if_kernel("Kernel ale access", regs); force_sig_fault(SIGBUS, BUS_ADRALN, (void __user *)regs->csr_badvaddr); out: + if (regs->csr_prmd & CSR_PRMD_PIE) + local_irq_disable(); #endif irqentry_exit(regs, state); } diff --git a/arch/loongarch/kernel/vdso.c b/arch/loongarch/kernel/vdso.c index f6fcc52aefae..2c0d852ca536 100644 --- a/arch/loongarch/kernel/vdso.c +++ b/arch/loongarch/kernel/vdso.c @@ -34,7 +34,6 @@ static union { struct loongarch_vdso_data vdata; } loongarch_vdso_data __page_aligned_data; -static struct page *vdso_pages[] = { NULL }; struct vdso_data *vdso_data = generic_vdso_data.data; struct vdso_pcpu_data *vdso_pdata = loongarch_vdso_data.vdata.pdata; struct vdso_rng_data *vdso_rng_data = &loongarch_vdso_data.vdata.rng_data; @@ -85,10 +84,8 @@ static vm_fault_t vvar_fault(const struct vm_special_mapping *sm, struct loongarch_vdso_info vdso_info = { .vdso = vdso_start, - .size = PAGE_SIZE, .code_mapping = { .name = "[vdso]", - .pages = vdso_pages, .mremap = vdso_mremap, }, .data_mapping = { @@ -103,11 +100,14 @@ static int __init init_vdso(void) unsigned long i, cpu, pfn; BUG_ON(!PAGE_ALIGNED(vdso_info.vdso)); - BUG_ON(!PAGE_ALIGNED(vdso_info.size)); for_each_possible_cpu(cpu) vdso_pdata[cpu].node = cpu_to_node(cpu); + vdso_info.size = PAGE_ALIGN(vdso_end - vdso_start); + vdso_info.code_mapping.pages = + kcalloc(vdso_info.size / PAGE_SIZE, sizeof(struct page *), GFP_KERNEL); + pfn = __phys_to_pfn(__pa_symbol(vdso_info.vdso)); for (i = 0; i < vdso_info.size / PAGE_SIZE; i++) vdso_info.code_mapping.pages[i] = pfn_to_page(pfn + i); diff --git a/arch/loongarch/kvm/timer.c b/arch/loongarch/kvm/timer.c index 74a4b5c272d6..32dc213374be 100644 --- a/arch/loongarch/kvm/timer.c +++ b/arch/loongarch/kvm/timer.c @@ -161,10 +161,11 @@ static void _kvm_save_timer(struct kvm_vcpu *vcpu) if (kvm_vcpu_is_blocking(vcpu)) { /* - * HRTIMER_MODE_PINNED is suggested since vcpu may run in - * the same physical cpu in next time + * HRTIMER_MODE_PINNED_HARD is suggested since vcpu may run in + * the same physical cpu in next time, and the timer should run + * in hardirq context even in the PREEMPT_RT case. */ - hrtimer_start(&vcpu->arch.swtimer, expire, HRTIMER_MODE_ABS_PINNED); + hrtimer_start(&vcpu->arch.swtimer, expire, HRTIMER_MODE_ABS_PINNED_HARD); } } diff --git a/arch/loongarch/kvm/vcpu.c b/arch/loongarch/kvm/vcpu.c index 0697b1064251..174734a23d0a 100644 --- a/arch/loongarch/kvm/vcpu.c +++ b/arch/loongarch/kvm/vcpu.c @@ -1457,7 +1457,7 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu) vcpu->arch.vpid = 0; vcpu->arch.flush_gpa = INVALID_GPA; - hrtimer_init(&vcpu->arch.swtimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); + hrtimer_init(&vcpu->arch.swtimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED_HARD); vcpu->arch.swtimer.function = kvm_swtimer_wakeup; vcpu->arch.handle_exit = kvm_handle_exit; diff --git a/arch/loongarch/mm/init.c b/arch/loongarch/mm/init.c index 8a87a482c8f4..188b52bbb254 100644 --- a/arch/loongarch/mm/init.c +++ b/arch/loongarch/mm/init.c @@ -201,7 +201,9 @@ pte_t * __init populate_kernel_pte(unsigned long addr) pte = memblock_alloc(PAGE_SIZE, PAGE_SIZE); if (!pte) panic("%s: Failed to allocate memory\n", __func__); + pmd_populate_kernel(&init_mm, pmd, pte); + kernel_pte_init(pte); } return pte_offset_kernel(pmd, addr); diff --git a/arch/loongarch/mm/pgtable.c b/arch/loongarch/mm/pgtable.c index eb6a29b491a7..3fa69b23ff84 100644 --- a/arch/loongarch/mm/pgtable.c +++ b/arch/loongarch/mm/pgtable.c @@ -116,6 +116,26 @@ void pud_init(void *addr) EXPORT_SYMBOL_GPL(pud_init); #endif +void kernel_pte_init(void *addr) +{ + unsigned long *p, *end; + + p = (unsigned long *)addr; + end = p + PTRS_PER_PTE; + + do { + p[0] = _PAGE_GLOBAL; + p[1] = _PAGE_GLOBAL; + p[2] = _PAGE_GLOBAL; + p[3] = _PAGE_GLOBAL; + p[4] = _PAGE_GLOBAL; + p += 8; + p[-3] = _PAGE_GLOBAL; + p[-2] = _PAGE_GLOBAL; + p[-1] = _PAGE_GLOBAL; + } while (p != end); +} + pmd_t mk_pmd(struct page *page, pgprot_t prot) { pmd_t pmd; diff --git a/arch/microblaze/include/asm/flat.h b/arch/microblaze/include/asm/flat.h index 79a749f4ad04..edff4306fa70 100644 --- a/arch/microblaze/include/asm/flat.h +++ b/arch/microblaze/include/asm/flat.h @@ -8,7 +8,7 @@ #ifndef _ASM_MICROBLAZE_FLAT_H #define _ASM_MICROBLAZE_FLAT_H -#include +#include /* * Microblaze works a little differently from other arches, because diff --git a/arch/mips/boot/compressed/decompress.c b/arch/mips/boot/compressed/decompress.c index adb6d5b0e6eb..90021c6a8cab 100644 --- a/arch/mips/boot/compressed/decompress.c +++ b/arch/mips/boot/compressed/decompress.c @@ -16,7 +16,7 @@ #include #include -#include +#include #include #include "decompress.h" diff --git a/arch/mips/crypto/crc32-mips.c b/arch/mips/crypto/crc32-mips.c index 2a59b85f88aa..a7a1d43a1b2c 100644 --- a/arch/mips/crypto/crc32-mips.c +++ b/arch/mips/crypto/crc32-mips.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include #include diff --git a/arch/mips/crypto/poly1305-glue.c b/arch/mips/crypto/poly1305-glue.c index 867728ee535a..c03ad0bbe69c 100644 --- a/arch/mips/crypto/poly1305-glue.c +++ b/arch/mips/crypto/poly1305-glue.c @@ -5,7 +5,7 @@ * Copyright (C) 2019 Linaro Ltd. */ -#include +#include #include #include #include diff --git a/arch/mips/kernel/cmpxchg.c b/arch/mips/kernel/cmpxchg.c index e974a4954df8..c371def2302d 100644 --- a/arch/mips/kernel/cmpxchg.c +++ b/arch/mips/kernel/cmpxchg.c @@ -102,3 +102,4 @@ unsigned long __cmpxchg_small(volatile void *ptr, unsigned long old, return old; } } +EXPORT_SYMBOL(__cmpxchg_small); diff --git a/arch/nios2/kernel/misaligned.c b/arch/nios2/kernel/misaligned.c index 23e0544e117c..2f2862eab3c6 100644 --- a/arch/nios2/kernel/misaligned.c +++ b/arch/nios2/kernel/misaligned.c @@ -23,7 +23,7 @@ #include #include -#include +#include /* instructions we emulate */ #define INST_LDHU 0x0b diff --git a/arch/parisc/boot/compressed/misc.c b/arch/parisc/boot/compressed/misc.c index d389359e22ac..9c83bd06ef15 100644 --- a/arch/parisc/boot/compressed/misc.c +++ b/arch/parisc/boot/compressed/misc.c @@ -6,7 +6,7 @@ #include #include -#include +#include #include #include "sizes.h" diff --git a/arch/parisc/include/asm/unaligned.h b/arch/parisc/include/asm/unaligned.h deleted file mode 100644 index c0621295100d..000000000000 --- a/arch/parisc/include/asm/unaligned.h +++ /dev/null @@ -1,11 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -#ifndef _ASM_PARISC_UNALIGNED_H -#define _ASM_PARISC_UNALIGNED_H - -#include - -struct pt_regs; -void handle_unaligned(struct pt_regs *regs); -int check_unaligned(struct pt_regs *regs); - -#endif /* _ASM_PARISC_UNALIGNED_H */ diff --git a/arch/parisc/kernel/traps.c b/arch/parisc/kernel/traps.c index 294b0e026c9a..b9b3d527bc90 100644 --- a/arch/parisc/kernel/traps.c +++ b/arch/parisc/kernel/traps.c @@ -36,7 +36,7 @@ #include #include #include -#include +#include #include #include #include @@ -47,6 +47,8 @@ #include #include +#include "unaligned.h" + #if defined(CONFIG_LIGHTWEIGHT_SPINLOCK_CHECK) #include #endif diff --git a/arch/parisc/kernel/unaligned.c b/arch/parisc/kernel/unaligned.c index 3e79e40e361d..f4626943633a 100644 --- a/arch/parisc/kernel/unaligned.c +++ b/arch/parisc/kernel/unaligned.c @@ -12,9 +12,10 @@ #include #include #include -#include +#include #include #include +#include "unaligned.h" /* #define DEBUG_UNALIGNED 1 */ diff --git a/arch/parisc/kernel/unaligned.h b/arch/parisc/kernel/unaligned.h new file mode 100644 index 000000000000..c1aa4b12e284 --- /dev/null +++ b/arch/parisc/kernel/unaligned.h @@ -0,0 +1,3 @@ +struct pt_regs; +void handle_unaligned(struct pt_regs *regs); +int check_unaligned(struct pt_regs *regs); diff --git a/arch/powerpc/crypto/aes-gcm-p10-glue.c b/arch/powerpc/crypto/aes-gcm-p10-glue.c index f62ee54076c0..f66ad56e765f 100644 --- a/arch/powerpc/crypto/aes-gcm-p10-glue.c +++ b/arch/powerpc/crypto/aes-gcm-p10-glue.c @@ -5,7 +5,7 @@ * Copyright 2022- IBM Inc. All rights reserved */ -#include +#include #include #include #include diff --git a/arch/powerpc/crypto/poly1305-p10-glue.c b/arch/powerpc/crypto/poly1305-p10-glue.c index 95dd708573ee..369686e9370b 100644 --- a/arch/powerpc/crypto/poly1305-p10-glue.c +++ b/arch/powerpc/crypto/poly1305-p10-glue.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include #include #include diff --git a/arch/powerpc/kernel/head_8xx.S b/arch/powerpc/kernel/head_8xx.S index 811a7130505c..56c5ebe21b99 100644 --- a/arch/powerpc/kernel/head_8xx.S +++ b/arch/powerpc/kernel/head_8xx.S @@ -494,6 +494,7 @@ FixupDAR:/* Entry point for dcbx workaround. */ bctr /* jump into table */ 152: mfdar r11 + mtdar r10 mtctr r11 /* restore ctr reg from DAR */ mfspr r11, SPRN_SPRG_THREAD stw r10, DAR(r11) diff --git a/arch/powerpc/kernel/vdso/Makefile b/arch/powerpc/kernel/vdso/Makefile index 56fb1633529a..31ca5a547004 100644 --- a/arch/powerpc/kernel/vdso/Makefile +++ b/arch/powerpc/kernel/vdso/Makefile @@ -22,7 +22,7 @@ endif ifneq ($(c-getrandom-y),) CFLAGS_vgetrandom-32.o += -include $(c-getrandom-y) - CFLAGS_vgetrandom-64.o += -include $(c-getrandom-y) $(call cc-option, -ffixed-r30) + CFLAGS_vgetrandom-64.o += -include $(c-getrandom-y) endif # Build rules diff --git a/arch/powerpc/platforms/powernv/opal-irqchip.c b/arch/powerpc/platforms/powernv/opal-irqchip.c index 56a1f7ce78d2..d92759c21fae 100644 --- a/arch/powerpc/platforms/powernv/opal-irqchip.c +++ b/arch/powerpc/platforms/powernv/opal-irqchip.c @@ -282,6 +282,7 @@ int __init opal_event_init(void) name, NULL); if (rc) { pr_warn("Error %d requesting OPAL irq %d\n", rc, (int)r->start); + kfree(name); continue; } } diff --git a/arch/powerpc/platforms/pseries/papr_scm.c b/arch/powerpc/platforms/pseries/papr_scm.c index d95e03b3d3e3..9e297f88adc5 100644 --- a/arch/powerpc/platforms/pseries/papr_scm.c +++ b/arch/powerpc/platforms/pseries/papr_scm.c @@ -19,7 +19,7 @@ #include #include #include -#include +#include #include #define BIND_ANY_ADDR (~0ul) diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig index 22dc5ea4196c..f4c570538d55 100644 --- a/arch/riscv/Kconfig +++ b/arch/riscv/Kconfig @@ -177,7 +177,7 @@ config RISCV select HAVE_REGS_AND_STACK_ACCESS_API select HAVE_RETHOOK if !XIP_KERNEL select HAVE_RSEQ - select HAVE_RUST if RUSTC_SUPPORTS_RISCV + select HAVE_RUST if RUSTC_SUPPORTS_RISCV && CC_IS_CLANG select HAVE_SAMPLE_FTRACE_DIRECT select HAVE_SAMPLE_FTRACE_DIRECT_MULTI select HAVE_STACKPROTECTOR @@ -777,8 +777,7 @@ config IRQ_STACKS config THREAD_SIZE_ORDER int "Kernel stack size (in power-of-two numbers of page size)" if VMAP_STACK && EXPERT range 0 4 - default 1 if 32BIT && !KASAN - default 3 if 64BIT && KASAN + default 1 if 32BIT default 2 help Specify the Pages of thread stack size (from 4KB to 64KB), which also diff --git a/arch/riscv/errata/Makefile b/arch/riscv/errata/Makefile index 8a2739485123..f0da9d7b39c3 100644 --- a/arch/riscv/errata/Makefile +++ b/arch/riscv/errata/Makefile @@ -2,6 +2,12 @@ ifdef CONFIG_RELOCATABLE KBUILD_CFLAGS += -fno-pie endif +ifdef CONFIG_RISCV_ALTERNATIVE_EARLY +ifdef CONFIG_FORTIFY_SOURCE +KBUILD_CFLAGS += -D__NO_FORTIFY +endif +endif + obj-$(CONFIG_ERRATA_ANDES) += andes/ obj-$(CONFIG_ERRATA_SIFIVE) += sifive/ obj-$(CONFIG_ERRATA_THEAD) += thead/ diff --git a/arch/riscv/include/asm/thread_info.h b/arch/riscv/include/asm/thread_info.h index ebe52f96da34..9c10fb180f43 100644 --- a/arch/riscv/include/asm/thread_info.h +++ b/arch/riscv/include/asm/thread_info.h @@ -13,7 +13,12 @@ #include /* thread information allocation */ -#define THREAD_SIZE_ORDER CONFIG_THREAD_SIZE_ORDER +#ifdef CONFIG_KASAN +#define KASAN_STACK_ORDER 1 +#else +#define KASAN_STACK_ORDER 0 +#endif +#define THREAD_SIZE_ORDER (CONFIG_THREAD_SIZE_ORDER + KASAN_STACK_ORDER) #define THREAD_SIZE (PAGE_SIZE << THREAD_SIZE_ORDER) /* diff --git a/arch/riscv/kernel/Makefile b/arch/riscv/kernel/Makefile index 7f88cc4931f5..69dc8aaab3fb 100644 --- a/arch/riscv/kernel/Makefile +++ b/arch/riscv/kernel/Makefile @@ -36,6 +36,11 @@ KASAN_SANITIZE_alternative.o := n KASAN_SANITIZE_cpufeature.o := n KASAN_SANITIZE_sbi_ecall.o := n endif +ifdef CONFIG_FORTIFY_SOURCE +CFLAGS_alternative.o += -D__NO_FORTIFY +CFLAGS_cpufeature.o += -D__NO_FORTIFY +CFLAGS_sbi_ecall.o += -D__NO_FORTIFY +endif endif extra-y += vmlinux.lds diff --git a/arch/riscv/kernel/acpi.c b/arch/riscv/kernel/acpi.c index 6e0d333f57e5..2fd29695a788 100644 --- a/arch/riscv/kernel/acpi.c +++ b/arch/riscv/kernel/acpi.c @@ -210,7 +210,7 @@ void __init __iomem *__acpi_map_table(unsigned long phys, unsigned long size) if (!size) return NULL; - return early_ioremap(phys, size); + return early_memremap(phys, size); } void __init __acpi_unmap_table(void __iomem *map, unsigned long size) @@ -218,7 +218,7 @@ void __init __acpi_unmap_table(void __iomem *map, unsigned long size) if (!map || !size) return; - early_iounmap(map, size); + early_memunmap(map, size); } void __iomem *acpi_os_ioremap(acpi_physical_address phys, acpi_size size) diff --git a/arch/riscv/kernel/asm-offsets.c b/arch/riscv/kernel/asm-offsets.c index e94180ba432f..c2f3129a8e5c 100644 --- a/arch/riscv/kernel/asm-offsets.c +++ b/arch/riscv/kernel/asm-offsets.c @@ -4,8 +4,6 @@ * Copyright (C) 2017 SiFive */ -#define GENERATING_ASM_OFFSETS - #include #include #include diff --git a/arch/riscv/kernel/cacheinfo.c b/arch/riscv/kernel/cacheinfo.c index b320b1d9aa01..2d40736fc37c 100644 --- a/arch/riscv/kernel/cacheinfo.c +++ b/arch/riscv/kernel/cacheinfo.c @@ -80,8 +80,7 @@ int populate_cache_leaves(unsigned int cpu) { struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu); struct cacheinfo *this_leaf = this_cpu_ci->info_list; - struct device_node *np = of_cpu_device_node_get(cpu); - struct device_node *prev = NULL; + struct device_node *np, *prev; int levels = 1, level = 1; if (!acpi_disabled) { @@ -105,6 +104,10 @@ int populate_cache_leaves(unsigned int cpu) return 0; } + np = of_cpu_device_node_get(cpu); + if (!np) + return -ENOENT; + if (of_property_read_bool(np, "cache-size")) ci_leaf_init(this_leaf++, CACHE_TYPE_UNIFIED, level); if (of_property_read_bool(np, "i-cache-size")) diff --git a/arch/riscv/kernel/cpu-hotplug.c b/arch/riscv/kernel/cpu-hotplug.c index 28b58fc5ad19..a1e38ecfc8be 100644 --- a/arch/riscv/kernel/cpu-hotplug.c +++ b/arch/riscv/kernel/cpu-hotplug.c @@ -58,7 +58,7 @@ void arch_cpuhp_cleanup_dead_cpu(unsigned int cpu) if (cpu_ops->cpu_is_stopped) ret = cpu_ops->cpu_is_stopped(cpu); if (ret) - pr_warn("CPU%d may not have stopped: %d\n", cpu, ret); + pr_warn("CPU%u may not have stopped: %d\n", cpu, ret); } /* diff --git a/arch/riscv/kernel/efi-header.S b/arch/riscv/kernel/efi-header.S index 515b2dfbca75..c5f17c2710b5 100644 --- a/arch/riscv/kernel/efi-header.S +++ b/arch/riscv/kernel/efi-header.S @@ -64,7 +64,7 @@ extra_header_fields: .long efi_header_end - _start // SizeOfHeaders .long 0 // CheckSum .short IMAGE_SUBSYSTEM_EFI_APPLICATION // Subsystem - .short 0 // DllCharacteristics + .short IMAGE_DLL_CHARACTERISTICS_NX_COMPAT // DllCharacteristics .quad 0 // SizeOfStackReserve .quad 0 // SizeOfStackCommit .quad 0 // SizeOfHeapReserve diff --git a/arch/riscv/kernel/pi/Makefile b/arch/riscv/kernel/pi/Makefile index d5bf1bc7de62..81d69d45c06c 100644 --- a/arch/riscv/kernel/pi/Makefile +++ b/arch/riscv/kernel/pi/Makefile @@ -16,8 +16,12 @@ KBUILD_CFLAGS := $(filter-out $(CC_FLAGS_LTO), $(KBUILD_CFLAGS)) KBUILD_CFLAGS += -mcmodel=medany CFLAGS_cmdline_early.o += -D__NO_FORTIFY -CFLAGS_lib-fdt_ro.o += -D__NO_FORTIFY CFLAGS_fdt_early.o += -D__NO_FORTIFY +# lib/string.c already defines __NO_FORTIFY +CFLAGS_ctype.o += -D__NO_FORTIFY +CFLAGS_lib-fdt.o += -D__NO_FORTIFY +CFLAGS_lib-fdt_ro.o += -D__NO_FORTIFY +CFLAGS_archrandom_early.o += -D__NO_FORTIFY $(obj)/%.pi.o: OBJCOPYFLAGS := --prefix-symbols=__pi_ \ --remove-section=.note.gnu.property \ diff --git a/arch/riscv/kernel/traps_misaligned.c b/arch/riscv/kernel/traps_misaligned.c index d4fd8af7aaf5..1b9867136b61 100644 --- a/arch/riscv/kernel/traps_misaligned.c +++ b/arch/riscv/kernel/traps_misaligned.c @@ -136,8 +136,6 @@ #define REG_PTR(insn, pos, regs) \ (ulong *)((ulong)(regs) + REG_OFFSET(insn, pos)) -#define GET_RM(insn) (((insn) >> 12) & 7) - #define GET_RS1(insn, regs) (*REG_PTR(insn, SH_RS1, regs)) #define GET_RS2(insn, regs) (*REG_PTR(insn, SH_RS2, regs)) #define GET_RS1S(insn, regs) (*REG_PTR(RVC_RS1S(insn), 0, regs)) diff --git a/arch/riscv/kernel/vdso/Makefile b/arch/riscv/kernel/vdso/Makefile index 960feb1526ca..3f1c4b2d0b06 100644 --- a/arch/riscv/kernel/vdso/Makefile +++ b/arch/riscv/kernel/vdso/Makefile @@ -18,6 +18,7 @@ obj-vdso = $(patsubst %, %.o, $(vdso-syms)) note.o ccflags-y := -fno-stack-protector ccflags-y += -DDISABLE_BRANCH_PROFILING +ccflags-y += -fno-builtin ifneq ($(c-gettimeofday-y),) CFLAGS_vgettimeofday.o += -fPIC -include $(c-gettimeofday-y) diff --git a/arch/riscv/kvm/aia_imsic.c b/arch/riscv/kvm/aia_imsic.c index 0a1e859323b4..a8085cd8215e 100644 --- a/arch/riscv/kvm/aia_imsic.c +++ b/arch/riscv/kvm/aia_imsic.c @@ -55,7 +55,7 @@ struct imsic { /* IMSIC SW-file */ struct imsic_mrif *swfile; phys_addr_t swfile_pa; - spinlock_t swfile_extirq_lock; + raw_spinlock_t swfile_extirq_lock; }; #define imsic_vs_csr_read(__c) \ @@ -622,7 +622,7 @@ static void imsic_swfile_extirq_update(struct kvm_vcpu *vcpu) * interruptions between reading topei and updating pending status. */ - spin_lock_irqsave(&imsic->swfile_extirq_lock, flags); + raw_spin_lock_irqsave(&imsic->swfile_extirq_lock, flags); if (imsic_mrif_atomic_read(mrif, &mrif->eidelivery) && imsic_mrif_topei(mrif, imsic->nr_eix, imsic->nr_msis)) @@ -630,7 +630,7 @@ static void imsic_swfile_extirq_update(struct kvm_vcpu *vcpu) else kvm_riscv_vcpu_unset_interrupt(vcpu, IRQ_VS_EXT); - spin_unlock_irqrestore(&imsic->swfile_extirq_lock, flags); + raw_spin_unlock_irqrestore(&imsic->swfile_extirq_lock, flags); } static void imsic_swfile_read(struct kvm_vcpu *vcpu, bool clear, @@ -1051,7 +1051,7 @@ int kvm_riscv_vcpu_aia_imsic_init(struct kvm_vcpu *vcpu) } imsic->swfile = page_to_virt(swfile_page); imsic->swfile_pa = page_to_phys(swfile_page); - spin_lock_init(&imsic->swfile_extirq_lock); + raw_spin_lock_init(&imsic->swfile_extirq_lock); /* Setup IO device */ kvm_iodevice_init(&imsic->iodev, &imsic_iodoev_ops); diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c index 99f34409fb60..4cc631fa7039 100644 --- a/arch/riscv/net/bpf_jit_comp64.c +++ b/arch/riscv/net/bpf_jit_comp64.c @@ -18,6 +18,7 @@ #define RV_MAX_REG_ARGS 8 #define RV_FENTRY_NINSNS 2 #define RV_FENTRY_NBYTES (RV_FENTRY_NINSNS * 4) +#define RV_KCFI_NINSNS (IS_ENABLED(CONFIG_CFI_CLANG) ? 1 : 0) /* imm that allows emit_imm to emit max count insns */ #define RV_MAX_COUNT_IMM 0x7FFF7FF7FF7FF7FF @@ -271,7 +272,8 @@ static void __build_epilogue(bool is_tail_call, struct rv_jit_context *ctx) if (!is_tail_call) emit_addiw(RV_REG_A0, RV_REG_A5, 0, ctx); emit_jalr(RV_REG_ZERO, is_tail_call ? RV_REG_T3 : RV_REG_RA, - is_tail_call ? (RV_FENTRY_NINSNS + 1) * 4 : 0, /* skip reserved nops and TCC init */ + /* kcfi, fentry and TCC init insns will be skipped on tailcall */ + is_tail_call ? (RV_KCFI_NINSNS + RV_FENTRY_NINSNS + 1) * 4 : 0, ctx); } @@ -548,8 +550,8 @@ static void emit_atomic(u8 rd, u8 rs, s16 off, s32 imm, bool is64, rv_lr_w(r0, 0, rd, 0, 0), ctx); jmp_offset = ninsns_rvoff(8); emit(rv_bne(RV_REG_T2, r0, jmp_offset >> 1), ctx); - emit(is64 ? rv_sc_d(RV_REG_T3, rs, rd, 0, 0) : - rv_sc_w(RV_REG_T3, rs, rd, 0, 0), ctx); + emit(is64 ? rv_sc_d(RV_REG_T3, rs, rd, 0, 1) : + rv_sc_w(RV_REG_T3, rs, rd, 0, 1), ctx); jmp_offset = ninsns_rvoff(-6); emit(rv_bne(RV_REG_T3, 0, jmp_offset >> 1), ctx); emit(rv_fence(0x3, 0x3), ctx); diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig index 9b57add02cd5..fb0e9a1d9be2 100644 --- a/arch/s390/configs/debug_defconfig +++ b/arch/s390/configs/debug_defconfig @@ -50,7 +50,6 @@ CONFIG_NUMA=y CONFIG_HZ_100=y CONFIG_CERT_STORE=y CONFIG_EXPOLINE=y -# CONFIG_EXPOLINE_EXTERN is not set CONFIG_EXPOLINE_AUTO=y CONFIG_CHSC_SCH=y CONFIG_VFIO_CCW=m @@ -95,6 +94,7 @@ CONFIG_BINFMT_MISC=m CONFIG_ZSWAP=y CONFIG_ZSWAP_ZPOOL_DEFAULT_ZBUD=y CONFIG_ZSMALLOC_STAT=y +CONFIG_SLAB_BUCKETS=y CONFIG_SLUB_STATS=y # CONFIG_COMPAT_BRK is not set CONFIG_MEMORY_HOTPLUG=y @@ -426,6 +426,13 @@ CONFIG_DEVTMPFS_SAFE=y # CONFIG_FW_LOADER is not set CONFIG_CONNECTOR=y CONFIG_ZRAM=y +CONFIG_ZRAM_BACKEND_LZ4=y +CONFIG_ZRAM_BACKEND_LZ4HC=y +CONFIG_ZRAM_BACKEND_ZSTD=y +CONFIG_ZRAM_BACKEND_DEFLATE=y +CONFIG_ZRAM_BACKEND_842=y +CONFIG_ZRAM_BACKEND_LZO=y +CONFIG_ZRAM_DEF_COMP_DEFLATE=y CONFIG_BLK_DEV_LOOP=m CONFIG_BLK_DEV_DRBD=m CONFIG_BLK_DEV_NBD=m @@ -486,6 +493,7 @@ CONFIG_DM_UEVENT=y CONFIG_DM_FLAKEY=m CONFIG_DM_VERITY=m CONFIG_DM_VERITY_VERIFY_ROOTHASH_SIG=y +CONFIG_DM_VERITY_VERIFY_ROOTHASH_SIG_PLATFORM_KEYRING=y CONFIG_DM_SWITCH=m CONFIG_DM_INTEGRITY=m CONFIG_DM_VDO=m @@ -535,6 +543,7 @@ CONFIG_NLMON=m CONFIG_MLX4_EN=m CONFIG_MLX5_CORE=m CONFIG_MLX5_CORE_EN=y +# CONFIG_NET_VENDOR_META is not set # CONFIG_NET_VENDOR_MICREL is not set # CONFIG_NET_VENDOR_MICROCHIP is not set # CONFIG_NET_VENDOR_MICROSEMI is not set @@ -695,6 +704,7 @@ CONFIG_NFSD=m CONFIG_NFSD_V3_ACL=y CONFIG_NFSD_V4=y CONFIG_NFSD_V4_SECURITY_LABEL=y +# CONFIG_NFSD_LEGACY_CLIENT_TRACKING is not set CONFIG_CIFS=m CONFIG_CIFS_UPCALL=y CONFIG_CIFS_XATTR=y @@ -740,7 +750,6 @@ CONFIG_CRYPTO_DH=m CONFIG_CRYPTO_ECDH=m CONFIG_CRYPTO_ECDSA=m CONFIG_CRYPTO_ECRDSA=m -CONFIG_CRYPTO_SM2=m CONFIG_CRYPTO_CURVE25519=m CONFIG_CRYPTO_AES_TI=m CONFIG_CRYPTO_ANUBIS=m diff --git a/arch/s390/configs/defconfig b/arch/s390/configs/defconfig index df4addd1834a..88be0a734b60 100644 --- a/arch/s390/configs/defconfig +++ b/arch/s390/configs/defconfig @@ -48,7 +48,6 @@ CONFIG_NUMA=y CONFIG_HZ_100=y CONFIG_CERT_STORE=y CONFIG_EXPOLINE=y -# CONFIG_EXPOLINE_EXTERN is not set CONFIG_EXPOLINE_AUTO=y CONFIG_CHSC_SCH=y CONFIG_VFIO_CCW=m @@ -89,6 +88,7 @@ CONFIG_BINFMT_MISC=m CONFIG_ZSWAP=y CONFIG_ZSWAP_ZPOOL_DEFAULT_ZBUD=y CONFIG_ZSMALLOC_STAT=y +CONFIG_SLAB_BUCKETS=y # CONFIG_COMPAT_BRK is not set CONFIG_MEMORY_HOTPLUG=y CONFIG_MEMORY_HOTREMOVE=y @@ -416,6 +416,13 @@ CONFIG_DEVTMPFS_SAFE=y # CONFIG_FW_LOADER is not set CONFIG_CONNECTOR=y CONFIG_ZRAM=y +CONFIG_ZRAM_BACKEND_LZ4=y +CONFIG_ZRAM_BACKEND_LZ4HC=y +CONFIG_ZRAM_BACKEND_ZSTD=y +CONFIG_ZRAM_BACKEND_DEFLATE=y +CONFIG_ZRAM_BACKEND_842=y +CONFIG_ZRAM_BACKEND_LZO=y +CONFIG_ZRAM_DEF_COMP_DEFLATE=y CONFIG_BLK_DEV_LOOP=m CONFIG_BLK_DEV_DRBD=m CONFIG_BLK_DEV_NBD=m @@ -476,6 +483,7 @@ CONFIG_DM_UEVENT=y CONFIG_DM_FLAKEY=m CONFIG_DM_VERITY=m CONFIG_DM_VERITY_VERIFY_ROOTHASH_SIG=y +CONFIG_DM_VERITY_VERIFY_ROOTHASH_SIG_PLATFORM_KEYRING=y CONFIG_DM_SWITCH=m CONFIG_DM_INTEGRITY=m CONFIG_DM_VDO=m @@ -525,6 +533,7 @@ CONFIG_NLMON=m CONFIG_MLX4_EN=m CONFIG_MLX5_CORE=m CONFIG_MLX5_CORE_EN=y +# CONFIG_NET_VENDOR_META is not set # CONFIG_NET_VENDOR_MICREL is not set # CONFIG_NET_VENDOR_MICROCHIP is not set # CONFIG_NET_VENDOR_MICROSEMI is not set @@ -682,6 +691,7 @@ CONFIG_NFSD=m CONFIG_NFSD_V3_ACL=y CONFIG_NFSD_V4=y CONFIG_NFSD_V4_SECURITY_LABEL=y +# CONFIG_NFSD_LEGACY_CLIENT_TRACKING is not set CONFIG_CIFS=m CONFIG_CIFS_UPCALL=y CONFIG_CIFS_XATTR=y @@ -726,7 +736,6 @@ CONFIG_CRYPTO_DH=m CONFIG_CRYPTO_ECDH=m CONFIG_CRYPTO_ECDSA=m CONFIG_CRYPTO_ECRDSA=m -CONFIG_CRYPTO_SM2=m CONFIG_CRYPTO_CURVE25519=m CONFIG_CRYPTO_AES_TI=m CONFIG_CRYPTO_ANUBIS=m @@ -767,6 +776,7 @@ CONFIG_CRYPTO_LZ4=m CONFIG_CRYPTO_LZ4HC=m CONFIG_CRYPTO_ZSTD=m CONFIG_CRYPTO_ANSI_CPRNG=m +CONFIG_CRYPTO_JITTERENTROPY_OSR=1 CONFIG_CRYPTO_USER_API_HASH=m CONFIG_CRYPTO_USER_API_SKCIPHER=m CONFIG_CRYPTO_USER_API_RNG=m diff --git a/arch/s390/configs/zfcpdump_defconfig b/arch/s390/configs/zfcpdump_defconfig index 8c2b61363bab..bcbaa069de96 100644 --- a/arch/s390/configs/zfcpdump_defconfig +++ b/arch/s390/configs/zfcpdump_defconfig @@ -49,6 +49,7 @@ CONFIG_ZFCP=y # CONFIG_HVC_IUCV is not set # CONFIG_HW_RANDOM_S390 is not set # CONFIG_HMC_DRV is not set +# CONFIG_S390_UV_UAPI is not set # CONFIG_S390_TAPE is not set # CONFIG_VMCP is not set # CONFIG_MONWRITER is not set diff --git a/arch/s390/include/asm/io.h b/arch/s390/include/asm/io.h index 0fbc992d7a5e..fc9933a743d6 100644 --- a/arch/s390/include/asm/io.h +++ b/arch/s390/include/asm/io.h @@ -16,8 +16,10 @@ #include #define xlate_dev_mem_ptr xlate_dev_mem_ptr +#define kc_xlate_dev_mem_ptr xlate_dev_mem_ptr void *xlate_dev_mem_ptr(phys_addr_t phys); #define unxlate_dev_mem_ptr unxlate_dev_mem_ptr +#define kc_unxlate_dev_mem_ptr unxlate_dev_mem_ptr void unxlate_dev_mem_ptr(phys_addr_t phys, void *addr); #define IO_SPACE_LIMIT 0 diff --git a/arch/s390/include/asm/perf_event.h b/arch/s390/include/asm/perf_event.h index 66200d4a2134..29ee289108c5 100644 --- a/arch/s390/include/asm/perf_event.h +++ b/arch/s390/include/asm/perf_event.h @@ -49,6 +49,7 @@ struct perf_sf_sde_regs { }; #define perf_arch_fetch_caller_regs(regs, __ip) do { \ + (regs)->psw.mask = 0; \ (regs)->psw.addr = (__ip); \ (regs)->gprs[15] = (unsigned long)__builtin_frame_address(0) - \ offsetof(struct stack_frame, back_chain); \ diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c index 2a32438e09ce..74f73141f9b9 100644 --- a/arch/s390/kvm/diag.c +++ b/arch/s390/kvm/diag.c @@ -77,7 +77,7 @@ static int __diag_page_ref_service(struct kvm_vcpu *vcpu) vcpu->stat.instruction_diagnose_258++; if (vcpu->run->s.regs.gprs[rx] & 7) return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); - rc = read_guest(vcpu, vcpu->run->s.regs.gprs[rx], rx, &parm, sizeof(parm)); + rc = read_guest_real(vcpu, vcpu->run->s.regs.gprs[rx], &parm, sizeof(parm)); if (rc) return kvm_s390_inject_prog_cond(vcpu, rc); if (parm.parm_version != 2 || parm.parm_len < 5 || parm.code != 0x258) diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c index e65f597e3044..a688351f4ab5 100644 --- a/arch/s390/kvm/gaccess.c +++ b/arch/s390/kvm/gaccess.c @@ -828,6 +828,8 @@ static int access_guest_page(struct kvm *kvm, enum gacc_mode mode, gpa_t gpa, const gfn_t gfn = gpa_to_gfn(gpa); int rc; + if (!gfn_to_memslot(kvm, gfn)) + return PGM_ADDRESSING; if (mode == GACC_STORE) rc = kvm_write_guest_page(kvm, gfn, data, offset, len); else @@ -985,6 +987,8 @@ int access_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, gra += fragment_len; data += fragment_len; } + if (rc > 0) + vcpu->arch.pgm.code = rc; return rc; } diff --git a/arch/s390/kvm/gaccess.h b/arch/s390/kvm/gaccess.h index b320d12aa049..3fde45a151f2 100644 --- a/arch/s390/kvm/gaccess.h +++ b/arch/s390/kvm/gaccess.h @@ -405,11 +405,12 @@ int read_guest_abs(struct kvm_vcpu *vcpu, unsigned long gpa, void *data, * @len: number of bytes to copy * * Copy @len bytes from @data (kernel space) to @gra (guest real address). - * It is up to the caller to ensure that the entire guest memory range is - * valid memory before calling this function. * Guest low address and key protection are not checked. * - * Returns zero on success or -EFAULT on error. + * Returns zero on success, -EFAULT when copying from @data failed, or + * PGM_ADRESSING in case @gra is outside a memslot. In this case, pgm check info + * is also stored to allow injecting into the guest (if applicable) using + * kvm_s390_inject_prog_cond(). * * If an error occurs data may have been copied partially to guest memory. */ @@ -428,11 +429,12 @@ int write_guest_real(struct kvm_vcpu *vcpu, unsigned long gra, void *data, * @len: number of bytes to copy * * Copy @len bytes from @gra (guest real address) to @data (kernel space). - * It is up to the caller to ensure that the entire guest memory range is - * valid memory before calling this function. * Guest key protection is not checked. * - * Returns zero on success or -EFAULT on error. + * Returns zero on success, -EFAULT when copying to @data failed, or + * PGM_ADRESSING in case @gra is outside a memslot. In this case, pgm check info + * is also stored to allow injecting into the guest (if applicable) using + * kvm_s390_inject_prog_cond(). * * If an error occurs data may have been copied partially to kernel space. */ diff --git a/arch/s390/pci/pci_event.c b/arch/s390/pci/pci_event.c index dbe95ec5917e..d4f19d33914c 100644 --- a/arch/s390/pci/pci_event.c +++ b/arch/s390/pci/pci_event.c @@ -280,18 +280,19 @@ static void __zpci_event_error(struct zpci_ccdf_err *ccdf) goto no_pdev; switch (ccdf->pec) { - case 0x003a: /* Service Action or Error Recovery Successful */ + case 0x002a: /* Error event concerns FMB */ + case 0x002b: + case 0x002c: + break; + case 0x0040: /* Service Action or Error Recovery Failed */ + case 0x003b: + zpci_event_io_failure(pdev, pci_channel_io_perm_failure); + break; + default: /* PCI function left in the error state attempt to recover */ ers_res = zpci_event_attempt_error_recovery(pdev); if (ers_res != PCI_ERS_RESULT_RECOVERED) zpci_event_io_failure(pdev, pci_channel_io_perm_failure); break; - default: - /* - * Mark as frozen not permanently failed because the device - * could be subsequently recovered by the platform. - */ - zpci_event_io_failure(pdev, pci_channel_io_frozen); - break; } pci_dev_put(pdev); no_pdev: diff --git a/arch/sh/include/asm/flat.h b/arch/sh/include/asm/flat.h index fee4f25555cb..70752c7bc55f 100644 --- a/arch/sh/include/asm/flat.h +++ b/arch/sh/include/asm/flat.h @@ -9,7 +9,7 @@ #ifndef __ASM_SH_FLAT_H #define __ASM_SH_FLAT_H -#include +#include static inline int flat_get_addr_from_rp(u32 __user *rp, u32 relval, u32 flags, u32 *addr) diff --git a/arch/sh/kernel/dwarf.c b/arch/sh/kernel/dwarf.c index 45c8ae20d109..a1b54bedc929 100644 --- a/arch/sh/kernel/dwarf.c +++ b/arch/sh/kernel/dwarf.c @@ -24,7 +24,7 @@ #include #include #include -#include +#include #include /* Reserve enough memory for two stack frames */ diff --git a/arch/sh/kernel/module.c b/arch/sh/kernel/module.c index b9cee98a754e..a469a80840d3 100644 --- a/arch/sh/kernel/module.c +++ b/arch/sh/kernel/module.c @@ -18,7 +18,7 @@ #include #include #include -#include +#include #include int apply_relocate_add(Elf32_Shdr *sechdrs, diff --git a/arch/sparc/crypto/crc32c_glue.c b/arch/sparc/crypto/crc32c_glue.c index 688db0dcb97d..913b9a09e885 100644 --- a/arch/sparc/crypto/crc32c_glue.c +++ b/arch/sparc/crypto/crc32c_glue.c @@ -20,7 +20,7 @@ #include #include -#include +#include #include "opcodes.h" diff --git a/arch/um/drivers/virt-pci.c b/arch/um/drivers/virt-pci.c index 6100819681b5..744e7f31e8ef 100644 --- a/arch/um/drivers/virt-pci.c +++ b/arch/um/drivers/virt-pci.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include #include #define MAX_DEVICES 8 diff --git a/arch/um/include/asm/uaccess.h b/arch/um/include/asm/uaccess.h index 7d9d60e41e4e..1d4b6bbc1b65 100644 --- a/arch/um/include/asm/uaccess.h +++ b/arch/um/include/asm/uaccess.h @@ -8,7 +8,7 @@ #define __UM_UACCESS_H #include -#include +#include #define __under_task_size(addr, size) \ (((unsigned long) (addr) < TASK_SIZE) && \ diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 2852fcd82cbd..16354dfa6d96 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -2257,6 +2257,7 @@ config RANDOMIZE_MEMORY_PHYSICAL_PADDING config ADDRESS_MASKING bool "Linear Address Masking support" depends on X86_64 + depends on COMPILE_TEST || !CPU_MITIGATIONS # wait for LASS help Linear Address Masking (LAM) modifies the checking that is applied to 64-bit linear addresses, allowing software to use of the diff --git a/arch/x86/crypto/camellia_glue.c b/arch/x86/crypto/camellia_glue.c index d45e9c0c42ac..f110708c8038 100644 --- a/arch/x86/crypto/camellia_glue.c +++ b/arch/x86/crypto/camellia_glue.c @@ -8,7 +8,7 @@ * Copyright (C) 2006 NTT (Nippon Telegraph and Telephone Corporation) */ -#include +#include #include #include #include diff --git a/arch/x86/crypto/ghash-clmulni-intel_glue.c b/arch/x86/crypto/ghash-clmulni-intel_glue.c index 700ecaee9a08..41bc02e48916 100644 --- a/arch/x86/crypto/ghash-clmulni-intel_glue.c +++ b/arch/x86/crypto/ghash-clmulni-intel_glue.c @@ -19,7 +19,7 @@ #include #include #include -#include +#include #define GHASH_BLOCK_SIZE 16 #define GHASH_DIGEST_SIZE 16 diff --git a/arch/x86/entry/entry.S b/arch/x86/entry/entry.S index d9feadffa972..324686bca368 100644 --- a/arch/x86/entry/entry.S +++ b/arch/x86/entry/entry.S @@ -9,6 +9,8 @@ #include #include #include +#include +#include #include "calling.h" @@ -19,6 +21,9 @@ SYM_FUNC_START(entry_ibpb) movl $PRED_CMD_IBPB, %eax xorl %edx, %edx wrmsr + + /* Make sure IBPB clears return stack preductions too. */ + FILL_RETURN_BUFFER %rax, RSB_CLEAR_LOOPS, X86_BUG_IBPB_NO_RET RET SYM_FUNC_END(entry_ibpb) /* For KVM */ diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S index d3a814efbff6..20be5758c2d2 100644 --- a/arch/x86/entry/entry_32.S +++ b/arch/x86/entry/entry_32.S @@ -871,6 +871,8 @@ SYM_FUNC_START(entry_SYSENTER_32) /* Now ready to switch the cr3 */ SWITCH_TO_USER_CR3 scratch_reg=%eax + /* Clobbers ZF */ + CLEAR_CPU_BUFFERS /* * Restore all flags except IF. (We restore IF separately because @@ -881,7 +883,6 @@ SYM_FUNC_START(entry_SYSENTER_32) BUG_IF_WRONG_CR3 no_user_check=1 popfl popl %eax - CLEAR_CPU_BUFFERS /* * Return back to the vDSO, which will pop ecx and edx. @@ -1144,7 +1145,6 @@ SYM_CODE_START(asm_exc_nmi) /* Not on SYSENTER stack. */ call exc_nmi - CLEAR_CPU_BUFFERS jmp .Lnmi_return .Lnmi_from_sysenter_stack: @@ -1165,6 +1165,7 @@ SYM_CODE_START(asm_exc_nmi) CHECK_AND_APPLY_ESPFIX RESTORE_ALL_NMI cr3_reg=%edi pop=4 + CLEAR_CPU_BUFFERS jmp .Lirq_return #ifdef CONFIG_X86_ESPFIX32 @@ -1206,6 +1207,7 @@ SYM_CODE_START(asm_exc_nmi) * 1 - orig_ax */ lss (1+5+6)*4(%esp), %esp # back to espfix stack + CLEAR_CPU_BUFFERS jmp .Lirq_return #endif SYM_CODE_END(asm_exc_nmi) diff --git a/arch/x86/include/asm/amd_nb.h b/arch/x86/include/asm/amd_nb.h index 6f3b6aef47ba..d0caac26533f 100644 --- a/arch/x86/include/asm/amd_nb.h +++ b/arch/x86/include/asm/amd_nb.h @@ -116,7 +116,10 @@ static inline bool amd_gart_present(void) #define amd_nb_num(x) 0 #define amd_nb_has_feature(x) false -#define node_to_amd_nb(x) NULL +static inline struct amd_northbridge *node_to_amd_nb(int node) +{ + return NULL; +} #define amd_gart_present(x) false #endif diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index dd4682857c12..913fd3a7bac6 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -215,7 +215,7 @@ #define X86_FEATURE_SPEC_STORE_BYPASS_DISABLE ( 7*32+23) /* Disable Speculative Store Bypass. */ #define X86_FEATURE_LS_CFG_SSBD ( 7*32+24) /* AMD SSBD implementation via LS_CFG MSR */ #define X86_FEATURE_IBRS ( 7*32+25) /* "ibrs" Indirect Branch Restricted Speculation */ -#define X86_FEATURE_IBPB ( 7*32+26) /* "ibpb" Indirect Branch Prediction Barrier */ +#define X86_FEATURE_IBPB ( 7*32+26) /* "ibpb" Indirect Branch Prediction Barrier without a guaranteed RSB flush */ #define X86_FEATURE_STIBP ( 7*32+27) /* "stibp" Single Thread Indirect Branch Predictors */ #define X86_FEATURE_ZEN ( 7*32+28) /* Generic flag for all Zen and newer */ #define X86_FEATURE_L1TF_PTEINV ( 7*32+29) /* L1TF workaround PTE inversion */ @@ -348,6 +348,7 @@ #define X86_FEATURE_CPPC (13*32+27) /* "cppc" Collaborative Processor Performance Control */ #define X86_FEATURE_AMD_PSFD (13*32+28) /* Predictive Store Forwarding Disable */ #define X86_FEATURE_BTC_NO (13*32+29) /* Not vulnerable to Branch Type Confusion */ +#define X86_FEATURE_AMD_IBPB_RET (13*32+30) /* IBPB clears return address predictor */ #define X86_FEATURE_BRS (13*32+31) /* "brs" Branch Sampling available */ /* Thermal and Power Management Leaf, CPUID level 0x00000006 (EAX), word 14 */ @@ -523,4 +524,5 @@ #define X86_BUG_DIV0 X86_BUG(1*32 + 1) /* "div0" AMD DIV0 speculation bug */ #define X86_BUG_RFDS X86_BUG(1*32 + 2) /* "rfds" CPU is vulnerable to Register File Data Sampling */ #define X86_BUG_BHI X86_BUG(1*32 + 3) /* "bhi" CPU is affected by Branch History Injection */ +#define X86_BUG_IBPB_NO_RET X86_BUG(1*32 + 4) /* "ibpb_no_ret" IBPB omits return target predictions */ #endif /* _ASM_X86_CPUFEATURES_H */ diff --git a/arch/x86/include/asm/ftrace.h b/arch/x86/include/asm/ftrace.h index 0152a81d9b4a..b4d719de2c84 100644 --- a/arch/x86/include/asm/ftrace.h +++ b/arch/x86/include/asm/ftrace.h @@ -2,6 +2,8 @@ #ifndef _ASM_X86_FTRACE_H #define _ASM_X86_FTRACE_H +#include + #ifdef CONFIG_FUNCTION_TRACER #ifndef CC_USING_FENTRY # error Compiler does not support fentry? diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h index ff5f1ecc7d1e..96b410b1d4e8 100644 --- a/arch/x86/include/asm/nospec-branch.h +++ b/arch/x86/include/asm/nospec-branch.h @@ -323,7 +323,16 @@ * Note: Only the memory operand variant of VERW clears the CPU buffers. */ .macro CLEAR_CPU_BUFFERS - ALTERNATIVE "", __stringify(verw _ASM_RIP(mds_verw_sel)), X86_FEATURE_CLEAR_CPU_BUF +#ifdef CONFIG_X86_64 + ALTERNATIVE "", "verw mds_verw_sel(%rip)", X86_FEATURE_CLEAR_CPU_BUF +#else + /* + * In 32bit mode, the memory operand must be a %cs reference. The data + * segments may not be usable (vm86 mode), and the stack segment may not + * be flat (ESPFIX32). + */ + ALTERNATIVE "", "verw %cs:mds_verw_sel", X86_FEATURE_CLEAR_CPU_BUF +#endif .endm #ifdef CONFIG_X86_64 diff --git a/arch/x86/include/asm/reboot.h b/arch/x86/include/asm/reboot.h index c02183d3cdd7..ecd58ea9a837 100644 --- a/arch/x86/include/asm/reboot.h +++ b/arch/x86/include/asm/reboot.h @@ -26,7 +26,7 @@ void __noreturn machine_real_restart(unsigned int type); #define MRR_APM 1 typedef void (cpu_emergency_virt_cb)(void); -#if IS_ENABLED(CONFIG_KVM_INTEL) || IS_ENABLED(CONFIG_KVM_AMD) +#if IS_ENABLED(CONFIG_KVM_X86) void cpu_emergency_register_virt_callback(cpu_emergency_virt_cb *callback); void cpu_emergency_unregister_virt_callback(cpu_emergency_virt_cb *callback); void cpu_emergency_disable_virtualization(void); @@ -34,7 +34,7 @@ void cpu_emergency_disable_virtualization(void); static inline void cpu_emergency_register_virt_callback(cpu_emergency_virt_cb *callback) {} static inline void cpu_emergency_unregister_virt_callback(cpu_emergency_virt_cb *callback) {} static inline void cpu_emergency_disable_virtualization(void) {} -#endif /* CONFIG_KVM_INTEL || CONFIG_KVM_AMD */ +#endif /* CONFIG_KVM_X86 */ typedef void (*nmi_shootdown_cb)(int, struct pt_regs*); void nmi_shootdown_cpus(nmi_shootdown_cb callback); diff --git a/arch/x86/include/asm/runtime-const.h b/arch/x86/include/asm/runtime-const.h index 24e3a53ca255..6652ebddfd02 100644 --- a/arch/x86/include/asm/runtime-const.h +++ b/arch/x86/include/asm/runtime-const.h @@ -6,7 +6,7 @@ typeof(sym) __ret; \ asm_inline("mov %1,%0\n1:\n" \ ".pushsection runtime_ptr_" #sym ",\"a\"\n\t" \ - ".long 1b - %c2 - .\n\t" \ + ".long 1b - %c2 - .\n" \ ".popsection" \ :"=r" (__ret) \ :"i" ((unsigned long)0x0123456789abcdefull), \ @@ -20,7 +20,7 @@ typeof(0u+(val)) __ret = (val); \ asm_inline("shrl $12,%k0\n1:\n" \ ".pushsection runtime_shift_" #sym ",\"a\"\n\t" \ - ".long 1b - 1 - .\n\t" \ + ".long 1b - 1 - .\n" \ ".popsection" \ :"+r" (__ret)); \ __ret; }) diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h index afce8ee5d7b7..b0a887209400 100644 --- a/arch/x86/include/asm/uaccess_64.h +++ b/arch/x86/include/asm/uaccess_64.h @@ -12,6 +12,13 @@ #include #include #include +#include + +/* + * Virtual variable: there's no actual backing store for this, + * it can purely be used as 'runtime_const_ptr(USER_PTR_MAX)' + */ +extern unsigned long USER_PTR_MAX; #ifdef CONFIG_ADDRESS_MASKING /* @@ -46,19 +53,24 @@ static inline unsigned long __untagged_addr_remote(struct mm_struct *mm, #endif -/* - * The virtual address space space is logically divided into a kernel - * half and a user half. When cast to a signed type, user pointers - * are positive and kernel pointers are negative. - */ -#define valid_user_address(x) ((__force long)(x) >= 0) +#define valid_user_address(x) \ + ((__force unsigned long)(x) <= runtime_const_ptr(USER_PTR_MAX)) /* * Masking the user address is an alternative to a conditional * user_access_begin that can avoid the fencing. This only works * for dense accesses starting at the address. */ -#define mask_user_address(x) ((typeof(x))((long)(x)|((long)(x)>>63))) +static inline void __user *mask_user_address(const void __user *ptr) +{ + unsigned long mask; + asm("cmp %1,%0\n\t" + "sbb %0,%0" + :"=r" (mask) + :"r" (ptr), + "0" (runtime_const_ptr(USER_PTR_MAX))); + return (__force void __user *)(mask | (__force unsigned long)ptr); +} #define masked_user_access_begin(x) ({ \ __auto_type __masked_ptr = (x); \ __masked_ptr = mask_user_address(__masked_ptr); \ @@ -69,23 +81,16 @@ static inline unsigned long __untagged_addr_remote(struct mm_struct *mm, * arbitrary values in those bits rather then masking them off. * * Enforce two rules: - * 1. 'ptr' must be in the user half of the address space + * 1. 'ptr' must be in the user part of the address space * 2. 'ptr+size' must not overflow into kernel addresses * - * Note that addresses around the sign change are not valid addresses, - * and will GP-fault even with LAM enabled if the sign bit is set (see - * "CR3.LAM_SUP" that can narrow the canonicality check if we ever - * enable it, but not remove it entirely). - * - * So the "overflow into kernel addresses" does not imply some sudden - * exact boundary at the sign bit, and we can allow a lot of slop on the - * size check. + * Note that we always have at least one guard page between the + * max user address and the non-canonical gap, allowing us to + * ignore small sizes entirely. * * In fact, we could probably remove the size check entirely, since * any kernel accesses will be in increasing address order starting - * at 'ptr', and even if the end might be in kernel space, we'll - * hit the GP faults for non-canonical accesses before we ever get - * there. + * at 'ptr'. * * That's a separate optimization, for now just handle the small * constant case. diff --git a/arch/x86/kernel/amd_nb.c b/arch/x86/kernel/amd_nb.c index dc5d3216af24..9fe9972d2071 100644 --- a/arch/x86/kernel/amd_nb.c +++ b/arch/x86/kernel/amd_nb.c @@ -44,6 +44,7 @@ #define PCI_DEVICE_ID_AMD_19H_M70H_DF_F4 0x14f4 #define PCI_DEVICE_ID_AMD_19H_M78H_DF_F4 0x12fc #define PCI_DEVICE_ID_AMD_1AH_M00H_DF_F4 0x12c4 +#define PCI_DEVICE_ID_AMD_1AH_M20H_DF_F4 0x16fc #define PCI_DEVICE_ID_AMD_1AH_M60H_DF_F4 0x124c #define PCI_DEVICE_ID_AMD_1AH_M70H_DF_F4 0x12bc #define PCI_DEVICE_ID_AMD_MI200_DF_F4 0x14d4 @@ -127,6 +128,7 @@ static const struct pci_device_id amd_nb_link_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_19H_M78H_DF_F4) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CNB17H_F4) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_1AH_M00H_DF_F4) }, + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_1AH_M20H_DF_F4) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_1AH_M60H_DF_F4) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_1AH_M70H_DF_F4) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_MI200_DF_F4) }, diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 6513c53c9459..c5fb28e6451a 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -440,7 +440,19 @@ static int lapic_timer_shutdown(struct clock_event_device *evt) v = apic_read(APIC_LVTT); v |= (APIC_LVT_MASKED | LOCAL_TIMER_VECTOR); apic_write(APIC_LVTT, v); - apic_write(APIC_TMICT, 0); + + /* + * Setting APIC_LVT_MASKED (above) should be enough to tell + * the hardware that this timer will never fire. But AMD + * erratum 411 and some Intel CPU behavior circa 2024 say + * otherwise. Time for belt and suspenders programming: mask + * the timer _and_ zero the counter registers: + */ + if (v & APIC_LVT_TIMER_TSCDEADLINE) + wrmsrl(MSR_IA32_TSC_DEADLINE, 0); + else + apic_write(APIC_TMICT, 0); + return 0; } diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 015971adadfc..fab5caec0b72 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -1202,5 +1202,6 @@ void amd_check_microcode(void) if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) return; - on_each_cpu(zenbleed_check_cpu, NULL, 1); + if (cpu_feature_enabled(X86_FEATURE_ZEN2)) + on_each_cpu(zenbleed_check_cpu, NULL, 1); } diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index d1915427b4ff..47a01d4028f6 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -1115,8 +1115,25 @@ do_cmd_auto: case RETBLEED_MITIGATION_IBPB: setup_force_cpu_cap(X86_FEATURE_ENTRY_IBPB); + + /* + * IBPB on entry already obviates the need for + * software-based untraining so clear those in case some + * other mitigation like SRSO has selected them. + */ + setup_clear_cpu_cap(X86_FEATURE_UNRET); + setup_clear_cpu_cap(X86_FEATURE_RETHUNK); + setup_force_cpu_cap(X86_FEATURE_IBPB_ON_VMEXIT); mitigate_smt = true; + + /* + * There is no need for RSB filling: entry_ibpb() ensures + * all predictions, including the RSB, are invalidated, + * regardless of IBPB implementation. + */ + setup_clear_cpu_cap(X86_FEATURE_RSB_VMEXIT); + break; case RETBLEED_MITIGATION_STUFF: @@ -2627,6 +2644,14 @@ static void __init srso_select_mitigation(void) if (has_microcode) { setup_force_cpu_cap(X86_FEATURE_ENTRY_IBPB); srso_mitigation = SRSO_MITIGATION_IBPB; + + /* + * IBPB on entry already obviates the need for + * software-based untraining so clear those in case some + * other mitigation like Retbleed has selected them. + */ + setup_clear_cpu_cap(X86_FEATURE_UNRET); + setup_clear_cpu_cap(X86_FEATURE_RETHUNK); } } else { pr_err("WARNING: kernel not compiled with MITIGATION_IBPB_ENTRY.\n"); @@ -2638,6 +2663,13 @@ static void __init srso_select_mitigation(void) if (!boot_cpu_has(X86_FEATURE_ENTRY_IBPB) && has_microcode) { setup_force_cpu_cap(X86_FEATURE_IBPB_ON_VMEXIT); srso_mitigation = SRSO_MITIGATION_IBPB_ON_VMEXIT; + + /* + * There is no need for RSB filling: entry_ibpb() ensures + * all predictions, including the RSB, are invalidated, + * regardless of IBPB implementation. + */ + setup_clear_cpu_cap(X86_FEATURE_RSB_VMEXIT); } } else { pr_err("WARNING: kernel not compiled with MITIGATION_SRSO.\n"); diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 07a34d723505..a5f221ea5688 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -69,6 +69,7 @@ #include #include #include +#include #include "cpu.h" @@ -1443,6 +1444,9 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) boot_cpu_has(X86_FEATURE_HYPERVISOR))) setup_force_cpu_bug(X86_BUG_BHI); + if (cpu_has(c, X86_FEATURE_AMD_IBPB) && !cpu_has(c, X86_FEATURE_AMD_IBPB_RET)) + setup_force_cpu_bug(X86_BUG_IBPB_NO_RET); + if (cpu_matches(cpu_vuln_whitelist, NO_MELTDOWN)) return; @@ -2386,6 +2390,15 @@ void __init arch_cpu_finalize_init(void) alternative_instructions(); if (IS_ENABLED(CONFIG_X86_64)) { + unsigned long USER_PTR_MAX = TASK_SIZE_MAX-1; + + /* + * Enable this when LAM is gated on LASS support + if (cpu_feature_enabled(X86_FEATURE_LAM)) + USER_PTR_MAX = (1ul << 63) - PAGE_SIZE - 1; + */ + runtime_const_init(ptr, USER_PTR_MAX); + /* * Make sure the first 2MB area is not mapped by huge pages * There are typically fixed size MTRRs in there and overlapping diff --git a/arch/x86/kernel/cpu/microcode/amd.c b/arch/x86/kernel/cpu/microcode/amd.c index f63b051f25a0..31a73715d755 100644 --- a/arch/x86/kernel/cpu/microcode/amd.c +++ b/arch/x86/kernel/cpu/microcode/amd.c @@ -584,7 +584,7 @@ void __init load_ucode_amd_bsp(struct early_load_data *ed, unsigned int cpuid_1_ native_rdmsr(MSR_AMD64_PATCH_LEVEL, ed->new_rev, dummy); } -static enum ucode_state load_microcode_amd(u8 family, const u8 *data, size_t size); +static enum ucode_state _load_microcode_amd(u8 family, const u8 *data, size_t size); static int __init save_microcode_in_initrd(void) { @@ -605,7 +605,7 @@ static int __init save_microcode_in_initrd(void) if (!desc.mc) return -EINVAL; - ret = load_microcode_amd(x86_family(cpuid_1_eax), desc.data, desc.size); + ret = _load_microcode_amd(x86_family(cpuid_1_eax), desc.data, desc.size); if (ret > UCODE_UPDATED) return -EINVAL; @@ -613,16 +613,19 @@ static int __init save_microcode_in_initrd(void) } early_initcall(save_microcode_in_initrd); -static inline bool patch_cpus_equivalent(struct ucode_patch *p, struct ucode_patch *n) +static inline bool patch_cpus_equivalent(struct ucode_patch *p, + struct ucode_patch *n, + bool ignore_stepping) { /* Zen and newer hardcode the f/m/s in the patch ID */ if (x86_family(bsp_cpuid_1_eax) >= 0x17) { union cpuid_1_eax p_cid = ucode_rev_to_cpuid(p->patch_id); union cpuid_1_eax n_cid = ucode_rev_to_cpuid(n->patch_id); - /* Zap stepping */ - p_cid.stepping = 0; - n_cid.stepping = 0; + if (ignore_stepping) { + p_cid.stepping = 0; + n_cid.stepping = 0; + } return p_cid.full == n_cid.full; } else { @@ -644,13 +647,13 @@ static struct ucode_patch *cache_find_patch(struct ucode_cpu_info *uci, u16 equi WARN_ON_ONCE(!n.patch_id); list_for_each_entry(p, µcode_cache, plist) - if (patch_cpus_equivalent(p, &n)) + if (patch_cpus_equivalent(p, &n, false)) return p; return NULL; } -static inline bool patch_newer(struct ucode_patch *p, struct ucode_patch *n) +static inline int patch_newer(struct ucode_patch *p, struct ucode_patch *n) { /* Zen and newer hardcode the f/m/s in the patch ID */ if (x86_family(bsp_cpuid_1_eax) >= 0x17) { @@ -659,6 +662,9 @@ static inline bool patch_newer(struct ucode_patch *p, struct ucode_patch *n) zp.ucode_rev = p->patch_id; zn.ucode_rev = n->patch_id; + if (zn.stepping != zp.stepping) + return -1; + return zn.rev > zp.rev; } else { return n->patch_id > p->patch_id; @@ -668,10 +674,14 @@ static inline bool patch_newer(struct ucode_patch *p, struct ucode_patch *n) static void update_cache(struct ucode_patch *new_patch) { struct ucode_patch *p; + int ret; list_for_each_entry(p, µcode_cache, plist) { - if (patch_cpus_equivalent(p, new_patch)) { - if (!patch_newer(p, new_patch)) { + if (patch_cpus_equivalent(p, new_patch, true)) { + ret = patch_newer(p, new_patch); + if (ret < 0) + continue; + else if (!ret) { /* we already have the latest patch */ kfree(new_patch->data); kfree(new_patch); @@ -944,6 +954,20 @@ static enum ucode_state __load_microcode_amd(u8 family, const u8 *data, return UCODE_OK; } +static enum ucode_state _load_microcode_amd(u8 family, const u8 *data, size_t size) +{ + enum ucode_state ret; + + /* free old equiv table */ + free_equiv_cpu_table(); + + ret = __load_microcode_amd(family, data, size); + if (ret != UCODE_OK) + cleanup(); + + return ret; +} + static enum ucode_state load_microcode_amd(u8 family, const u8 *data, size_t size) { struct cpuinfo_x86 *c; @@ -951,14 +975,9 @@ static enum ucode_state load_microcode_amd(u8 family, const u8 *data, size_t siz struct ucode_patch *p; enum ucode_state ret; - /* free old equiv table */ - free_equiv_cpu_table(); - - ret = __load_microcode_amd(family, data, size); - if (ret != UCODE_OK) { - cleanup(); + ret = _load_microcode_amd(family, data, size); + if (ret != UCODE_OK) return ret; - } for_each_node(nid) { cpu = cpumask_first(cpumask_of_node(nid)); diff --git a/arch/x86/kernel/cpu/resctrl/core.c b/arch/x86/kernel/cpu/resctrl/core.c index 8591d53c144b..b681c2e07dbf 100644 --- a/arch/x86/kernel/cpu/resctrl/core.c +++ b/arch/x86/kernel/cpu/resctrl/core.c @@ -207,7 +207,7 @@ static inline bool rdt_get_mb_table(struct rdt_resource *r) return false; } -static bool __get_mem_config_intel(struct rdt_resource *r) +static __init bool __get_mem_config_intel(struct rdt_resource *r) { struct rdt_hw_resource *hw_res = resctrl_to_arch_res(r); union cpuid_0x10_3_eax eax; @@ -241,7 +241,7 @@ static bool __get_mem_config_intel(struct rdt_resource *r) return true; } -static bool __rdt_get_mem_config_amd(struct rdt_resource *r) +static __init bool __rdt_get_mem_config_amd(struct rdt_resource *r) { struct rdt_hw_resource *hw_res = resctrl_to_arch_res(r); u32 eax, ebx, ecx, edx, subleaf; diff --git a/arch/x86/kernel/cpu/resctrl/ctrlmondata.c b/arch/x86/kernel/cpu/resctrl/ctrlmondata.c index 50fa1fe9a073..200d89a64027 100644 --- a/arch/x86/kernel/cpu/resctrl/ctrlmondata.c +++ b/arch/x86/kernel/cpu/resctrl/ctrlmondata.c @@ -29,10 +29,10 @@ * hardware. The allocated bandwidth percentage is rounded to the next * control step available on the hardware. */ -static bool bw_validate(char *buf, unsigned long *data, struct rdt_resource *r) +static bool bw_validate(char *buf, u32 *data, struct rdt_resource *r) { - unsigned long bw; int ret; + u32 bw; /* * Only linear delay values is supported for current Intel SKUs. @@ -42,16 +42,21 @@ static bool bw_validate(char *buf, unsigned long *data, struct rdt_resource *r) return false; } - ret = kstrtoul(buf, 10, &bw); + ret = kstrtou32(buf, 10, &bw); if (ret) { - rdt_last_cmd_printf("Non-decimal digit in MB value %s\n", buf); + rdt_last_cmd_printf("Invalid MB value %s\n", buf); return false; } - if ((bw < r->membw.min_bw || bw > r->default_ctrl) && - !is_mba_sc(r)) { - rdt_last_cmd_printf("MB value %ld out of range [%d,%d]\n", bw, - r->membw.min_bw, r->default_ctrl); + /* Nothing else to do if software controller is enabled. */ + if (is_mba_sc(r)) { + *data = bw; + return true; + } + + if (bw < r->membw.min_bw || bw > r->default_ctrl) { + rdt_last_cmd_printf("MB value %u out of range [%d,%d]\n", + bw, r->membw.min_bw, r->default_ctrl); return false; } @@ -65,7 +70,7 @@ int parse_bw(struct rdt_parse_data *data, struct resctrl_schema *s, struct resctrl_staged_config *cfg; u32 closid = data->rdtgrp->closid; struct rdt_resource *r = s->res; - unsigned long bw_val; + u32 bw_val; cfg = &d->staged_config[s->conf_type]; if (cfg->have_new_ctrl) { diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c index 263f8aed4e2c..21e9e4845354 100644 --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -37,6 +37,7 @@ #include #include #include +#include #include #include #include @@ -980,6 +981,9 @@ static void __init kvm_init_platform(void) } kvmclock_init(); x86_platform.apic_post_init = kvm_apic_init; + + /* Set WB as the default cache mode for SEV-SNP and TDX */ + mtrr_overwrite_state(NULL, 0, MTRR_TYPE_WRBACK); } #if defined(CONFIG_AMD_MEM_ENCRYPT) diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 0e0a4cf6b5eb..615922838c51 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -530,7 +530,7 @@ static inline void kb_wait(void) static inline void nmi_shootdown_cpus_on_restart(void); -#if IS_ENABLED(CONFIG_KVM_INTEL) || IS_ENABLED(CONFIG_KVM_AMD) +#if IS_ENABLED(CONFIG_KVM_X86) /* RCU-protected callback to disable virtualization prior to reboot. */ static cpu_emergency_virt_cb __rcu *cpu_emergency_virt_callback; @@ -600,7 +600,7 @@ static void emergency_reboot_disable_virtualization(void) } #else static void emergency_reboot_disable_virtualization(void) { } -#endif /* CONFIG_KVM_INTEL || CONFIG_KVM_AMD */ +#endif /* CONFIG_KVM_X86 */ void __attribute__((weak)) mach_reboot_fixups(void) { diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index d05392db5d0f..2dbadf347b5f 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -261,12 +261,6 @@ static noinstr bool handle_bug(struct pt_regs *regs) int ud_type; u32 imm; - /* - * Normally @regs are unpoisoned by irqentry_enter(), but handle_bug() - * is a rare case that uses @regs without passing them to - * irqentry_enter(). - */ - kmsan_unpoison_entry_regs(regs); ud_type = decode_bug(regs->ip, &imm); if (ud_type == BUG_NONE) return handled; @@ -275,6 +269,12 @@ static noinstr bool handle_bug(struct pt_regs *regs) * All lies, just get the WARN/BUG out. */ instrumentation_begin(); + /* + * Normally @regs are unpoisoned by irqentry_enter(), but handle_bug() + * is a rare case that uses @regs without passing them to + * irqentry_enter(). + */ + kmsan_unpoison_entry_regs(regs); /* * Since we're emulating a CALL with exceptions, restore the interrupt * state to what it was at the exception site. diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index 6726be89b7a6..b8c5741d2fb4 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -358,6 +358,7 @@ SECTIONS #endif RUNTIME_CONST_VARIABLES + RUNTIME_CONST(ptr, USER_PTR_MAX) . = ALIGN(PAGE_SIZE); diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig index 730c2f34d347..f09f13c01c6b 100644 --- a/arch/x86/kvm/Kconfig +++ b/arch/x86/kvm/Kconfig @@ -17,8 +17,8 @@ menuconfig VIRTUALIZATION if VIRTUALIZATION -config KVM - tristate "Kernel-based Virtual Machine (KVM) support" +config KVM_X86 + def_tristate KVM if KVM_INTEL || KVM_AMD depends on X86_LOCAL_APIC select KVM_COMMON select KVM_GENERIC_MMU_NOTIFIER @@ -44,7 +44,11 @@ config KVM select HAVE_KVM_PM_NOTIFIER if PM select KVM_GENERIC_HARDWARE_ENABLING select KVM_GENERIC_PRE_FAULT_MEMORY + select KVM_GENERIC_PRIVATE_MEM if KVM_SW_PROTECTED_VM select KVM_WERROR if WERROR + +config KVM + tristate "Kernel-based Virtual Machine (KVM) support" help Support hosting fully virtualized guest machines using hardware virtualization extensions. You will need a fairly recent @@ -77,7 +81,6 @@ config KVM_SW_PROTECTED_VM bool "Enable support for KVM software-protected VMs" depends on EXPERT depends on KVM && X86_64 - select KVM_GENERIC_PRIVATE_MEM help Enable support for KVM software-protected VMs. Currently, software- protected VMs are purely a development and testing vehicle for diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile index 5494669a055a..f9dddb8cb466 100644 --- a/arch/x86/kvm/Makefile +++ b/arch/x86/kvm/Makefile @@ -32,7 +32,7 @@ kvm-intel-y += vmx/vmx_onhyperv.o vmx/hyperv_evmcs.o kvm-amd-y += svm/svm_onhyperv.o endif -obj-$(CONFIG_KVM) += kvm.o +obj-$(CONFIG_KVM_X86) += kvm.o obj-$(CONFIG_KVM_INTEL) += kvm-intel.o obj-$(CONFIG_KVM_AMD) += kvm-amd.o diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index e52f990548df..8e853a5fc867 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -1556,6 +1556,17 @@ bool kvm_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range) { bool flush = false; + /* + * To prevent races with vCPUs faulting in a gfn using stale data, + * zapping a gfn range must be protected by mmu_invalidate_in_progress + * (and mmu_invalidate_seq). The only exception is memslot deletion; + * in that case, SRCU synchronization ensures that SPTEs are zapped + * after all vCPUs have unlocked SRCU, guaranteeing that vCPUs see the + * invalid slot. + */ + lockdep_assert_once(kvm->mmu_invalidate_in_progress || + lockdep_is_held(&kvm->slots_lock)); + if (kvm_memslots_have_rmaps(kvm)) flush = __kvm_rmap_zap_gfn_range(kvm, range->slot, range->start, range->end, @@ -7047,14 +7058,42 @@ void kvm_arch_flush_shadow_all(struct kvm *kvm) kvm_mmu_zap_all(kvm); } -/* - * Zapping leaf SPTEs with memslot range when a memslot is moved/deleted. - * - * Zapping non-leaf SPTEs, a.k.a. not-last SPTEs, isn't required, worst - * case scenario we'll have unused shadow pages lying around until they - * are recycled due to age or when the VM is destroyed. - */ -static void kvm_mmu_zap_memslot_leafs(struct kvm *kvm, struct kvm_memory_slot *slot) +static void kvm_mmu_zap_memslot_pages_and_flush(struct kvm *kvm, + struct kvm_memory_slot *slot, + bool flush) +{ + LIST_HEAD(invalid_list); + unsigned long i; + + if (list_empty(&kvm->arch.active_mmu_pages)) + goto out_flush; + + /* + * Since accounting information is stored in struct kvm_arch_memory_slot, + * all MMU pages that are shadowing guest PTEs must be zapped before the + * memslot is deleted, as freeing such pages after the memslot is freed + * will result in use-after-free, e.g. in unaccount_shadowed(). + */ + for (i = 0; i < slot->npages; i++) { + struct kvm_mmu_page *sp; + gfn_t gfn = slot->base_gfn + i; + + for_each_gfn_valid_sp_with_gptes(kvm, sp, gfn) + kvm_mmu_prepare_zap_page(kvm, sp, &invalid_list); + + if (need_resched() || rwlock_needbreak(&kvm->mmu_lock)) { + kvm_mmu_remote_flush_or_zap(kvm, &invalid_list, flush); + flush = false; + cond_resched_rwlock_write(&kvm->mmu_lock); + } + } + +out_flush: + kvm_mmu_remote_flush_or_zap(kvm, &invalid_list, flush); +} + +static void kvm_mmu_zap_memslot(struct kvm *kvm, + struct kvm_memory_slot *slot) { struct kvm_gfn_range range = { .slot = slot, @@ -7062,11 +7101,11 @@ static void kvm_mmu_zap_memslot_leafs(struct kvm *kvm, struct kvm_memory_slot *s .end = slot->base_gfn + slot->npages, .may_block = true, }; + bool flush; write_lock(&kvm->mmu_lock); - if (kvm_unmap_gfn_range(kvm, &range)) - kvm_flush_remote_tlbs_memslot(kvm, slot); - + flush = kvm_unmap_gfn_range(kvm, &range); + kvm_mmu_zap_memslot_pages_and_flush(kvm, slot, flush); write_unlock(&kvm->mmu_lock); } @@ -7082,7 +7121,7 @@ void kvm_arch_flush_shadow_memslot(struct kvm *kvm, if (kvm_memslot_flush_zap_all(kvm)) kvm_mmu_zap_all_fast(kvm); else - kvm_mmu_zap_memslot_leafs(kvm, slot); + kvm_mmu_zap_memslot(kvm, slot); } void kvm_mmu_invalidate_mmio_sptes(struct kvm *kvm, u64 gen) diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c index d5314cb7dff4..cf84103ce38b 100644 --- a/arch/x86/kvm/svm/nested.c +++ b/arch/x86/kvm/svm/nested.c @@ -63,8 +63,12 @@ static u64 nested_svm_get_tdp_pdptr(struct kvm_vcpu *vcpu, int index) u64 pdpte; int ret; + /* + * Note, nCR3 is "assumed" to be 32-byte aligned, i.e. the CPU ignores + * nCR3[4:0] when loading PDPTEs from memory. + */ ret = kvm_vcpu_read_guest_page(vcpu, gpa_to_gfn(cr3), &pdpte, - offset_in_page(cr3) + index * 8, 8); + (cr3 & GENMASK(11, 5)) + index * 8, 8); if (ret) return 0; return pdpte; diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index 1a4438358c5e..81ed596e4454 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -4888,9 +4888,6 @@ void vmx_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event) vmx->hv_deadline_tsc = -1; kvm_set_cr8(vcpu, 0); - vmx_segment_cache_clear(vmx); - kvm_register_mark_available(vcpu, VCPU_EXREG_SEGMENTS); - seg_setup(VCPU_SREG_CS); vmcs_write16(GUEST_CS_SELECTOR, 0xf000); vmcs_writel(GUEST_CS_BASE, 0xffff0000ul); @@ -4917,6 +4914,9 @@ void vmx_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event) vmcs_writel(GUEST_IDTR_BASE, 0); vmcs_write32(GUEST_IDTR_LIMIT, 0xffff); + vmx_segment_cache_clear(vmx); + kvm_register_mark_available(vcpu, VCPU_EXREG_SEGMENTS); + vmcs_write32(GUEST_ACTIVITY_STATE, GUEST_ACTIVITY_ACTIVE); vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, 0); vmcs_writel(GUEST_PENDING_DBG_EXCEPTIONS, 0); diff --git a/arch/x86/lib/getuser.S b/arch/x86/lib/getuser.S index d066aecf8aeb..4357ec2a0bfc 100644 --- a/arch/x86/lib/getuser.S +++ b/arch/x86/lib/getuser.S @@ -39,8 +39,13 @@ .macro check_range size:req .if IS_ENABLED(CONFIG_X86_64) - mov %rax, %rdx - sar $63, %rdx + movq $0x0123456789abcdef,%rdx + 1: + .pushsection runtime_ptr_USER_PTR_MAX,"a" + .long 1b - 8 - . + .popsection + cmp %rax, %rdx + sbb %rdx, %rdx or %rdx, %rax .else cmp $TASK_SIZE_MAX-\size+1, %eax diff --git a/arch/x86/lib/insn.c b/arch/x86/lib/insn.c index 5952ab41c60f..6ffb931b9fb1 100644 --- a/arch/x86/lib/insn.c +++ b/arch/x86/lib/insn.c @@ -13,7 +13,7 @@ #endif #include /*__ignore_sync_check__ */ #include /* __ignore_sync_check__ */ -#include /* __ignore_sync_check__ */ +#include /* __ignore_sync_check__ */ #include #include diff --git a/arch/x86/virt/svm/sev.c b/arch/x86/virt/svm/sev.c index 0ce17766c0e5..9a6a943d8e41 100644 --- a/arch/x86/virt/svm/sev.c +++ b/arch/x86/virt/svm/sev.c @@ -173,6 +173,8 @@ static void __init __snp_fixup_e820_tables(u64 pa) e820__range_update(pa, PMD_SIZE, E820_TYPE_RAM, E820_TYPE_RESERVED); e820__range_update_table(e820_table_kexec, pa, PMD_SIZE, E820_TYPE_RAM, E820_TYPE_RESERVED); e820__range_update_table(e820_table_firmware, pa, PMD_SIZE, E820_TYPE_RAM, E820_TYPE_RESERVED); + if (!memblock_is_region_reserved(pa, PMD_SIZE)) + memblock_reserve(pa, PMD_SIZE); } } diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c index 2c12ae42dc8b..d6818c6cafda 100644 --- a/arch/x86/xen/enlighten_pv.c +++ b/arch/x86/xen/enlighten_pv.c @@ -1032,6 +1032,10 @@ static u64 xen_do_read_msr(unsigned int msr, int *err) switch (msr) { case MSR_IA32_APICBASE: val &= ~X2APIC_ENABLE; + if (smp_processor_id() == 0) + val |= MSR_IA32_APICBASE_BSP; + else + val &= ~MSR_IA32_APICBASE_BSP; break; } return val; diff --git a/arch/xtensa/include/asm/flat.h b/arch/xtensa/include/asm/flat.h index ed5870c779f9..4854419dcd86 100644 --- a/arch/xtensa/include/asm/flat.h +++ b/arch/xtensa/include/asm/flat.h @@ -2,7 +2,7 @@ #ifndef __ASM_XTENSA_FLAT_H #define __ASM_XTENSA_FLAT_H -#include +#include static inline int flat_get_addr_from_rp(u32 __user *rp, u32 relval, u32 flags, u32 *addr) diff --git a/block/blk-integrity.c b/block/blk-integrity.c index 0a2b1c5d0ebf..83b696ba0cac 100644 --- a/block/blk-integrity.c +++ b/block/blk-integrity.c @@ -56,8 +56,7 @@ new_segment: /** * blk_rq_map_integrity_sg - Map integrity metadata into a scatterlist - * @q: request queue - * @bio: bio with integrity metadata attached + * @rq: request to map * @sglist: target scatterlist * * Description: Map the integrity vectors in request into a diff --git a/block/blk-iocost.c b/block/blk-iocost.c index 9dc9323f84ac..384aa15e8260 100644 --- a/block/blk-iocost.c +++ b/block/blk-iocost.c @@ -3166,7 +3166,7 @@ static u64 ioc_qos_prfill(struct seq_file *sf, struct blkg_policy_data *pd, if (!dname) return 0; - spin_lock_irq(&ioc->lock); + spin_lock(&ioc->lock); seq_printf(sf, "%s enable=%d ctrl=%s rpct=%u.%02u rlat=%u wpct=%u.%02u wlat=%u min=%u.%02u max=%u.%02u\n", dname, ioc->enabled, ioc->user_qos_params ? "user" : "auto", ioc->params.qos[QOS_RPPM] / 10000, @@ -3179,7 +3179,7 @@ static u64 ioc_qos_prfill(struct seq_file *sf, struct blkg_policy_data *pd, ioc->params.qos[QOS_MIN] % 10000 / 100, ioc->params.qos[QOS_MAX] / 10000, ioc->params.qos[QOS_MAX] % 10000 / 100); - spin_unlock_irq(&ioc->lock); + spin_unlock(&ioc->lock); return 0; } @@ -3366,14 +3366,14 @@ static u64 ioc_cost_model_prfill(struct seq_file *sf, if (!dname) return 0; - spin_lock_irq(&ioc->lock); + spin_lock(&ioc->lock); seq_printf(sf, "%s ctrl=%s model=linear " "rbps=%llu rseqiops=%llu rrandiops=%llu " "wbps=%llu wseqiops=%llu wrandiops=%llu\n", dname, ioc->user_cost_model ? "user" : "auto", u[I_LCOEF_RBPS], u[I_LCOEF_RSEQIOPS], u[I_LCOEF_RRANDIOPS], u[I_LCOEF_WBPS], u[I_LCOEF_WSEQIOPS], u[I_LCOEF_WRANDIOPS]); - spin_unlock_irq(&ioc->lock); + spin_unlock(&ioc->lock); return 0; } diff --git a/block/blk-map.c b/block/blk-map.c index 0e1167b23934..b5fd1d857461 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -561,57 +561,33 @@ EXPORT_SYMBOL(blk_rq_append_bio); /* Prepare bio for passthrough IO given ITER_BVEC iter */ static int blk_rq_map_user_bvec(struct request *rq, const struct iov_iter *iter) { - struct request_queue *q = rq->q; - size_t nr_iter = iov_iter_count(iter); - size_t nr_segs = iter->nr_segs; - struct bio_vec *bvecs, *bvprvp = NULL; - const struct queue_limits *lim = &q->limits; - unsigned int nsegs = 0, bytes = 0; + const struct queue_limits *lim = &rq->q->limits; + unsigned int max_bytes = lim->max_hw_sectors << SECTOR_SHIFT; + unsigned int nsegs; struct bio *bio; - size_t i; + int ret; - if (!nr_iter || (nr_iter >> SECTOR_SHIFT) > queue_max_hw_sectors(q)) - return -EINVAL; - if (nr_segs > queue_max_segments(q)) + if (!iov_iter_count(iter) || iov_iter_count(iter) > max_bytes) return -EINVAL; - /* no iovecs to alloc, as we already have a BVEC iterator */ + /* reuse the bvecs from the iterator instead of allocating new ones */ bio = blk_rq_map_bio_alloc(rq, 0, GFP_KERNEL); - if (bio == NULL) + if (!bio) return -ENOMEM; - bio_iov_bvec_set(bio, (struct iov_iter *)iter); - blk_rq_bio_prep(rq, bio, nr_segs); - /* loop to perform a bunch of sanity checks */ - bvecs = (struct bio_vec *)iter->bvec; - for (i = 0; i < nr_segs; i++) { - struct bio_vec *bv = &bvecs[i]; - - /* - * If the queue doesn't support SG gaps and adding this - * offset would create a gap, fallback to copy. - */ - if (bvprvp && bvec_gap_to_prev(lim, bvprvp, bv->bv_offset)) { - blk_mq_map_bio_put(bio); - return -EREMOTEIO; - } - /* check full condition */ - if (nsegs >= nr_segs || bytes > UINT_MAX - bv->bv_len) - goto put_bio; - if (bytes + bv->bv_len > nr_iter) - goto put_bio; - if (bv->bv_offset + bv->bv_len > PAGE_SIZE) - goto put_bio; - - nsegs++; - bytes += bv->bv_len; - bvprvp = bv; + /* check that the data layout matches the hardware restrictions */ + ret = bio_split_rw_at(bio, lim, &nsegs, max_bytes); + if (ret) { + /* if we would have to split the bio, copy instead */ + if (ret > 0) + ret = -EREMOTEIO; + blk_mq_map_bio_put(bio); + return ret; } + + blk_rq_bio_prep(rq, bio, nsegs); return 0; -put_bio: - blk_mq_map_bio_put(bio); - return -EINVAL; } /** diff --git a/block/blk-mq.c b/block/blk-mq.c index 4b2c8e940f59..cf626e061dd7 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -4310,6 +4310,12 @@ int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, /* mark the queue as mq asap */ q->mq_ops = set->ops; + /* + * ->tag_set has to be setup before initialize hctx, which cpuphp + * handler needs it for checking queue mapping + */ + q->tag_set = set; + if (blk_mq_alloc_ctxs(q)) goto err_exit; @@ -4328,8 +4334,6 @@ int blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, INIT_WORK(&q->timeout_work, blk_mq_timeout_work); blk_queue_rq_timeout(q, set->timeout ? set->timeout : 30 * HZ); - q->tag_set = set; - q->queue_flags |= QUEUE_FLAG_MQ_DEFAULT; INIT_DELAYED_WORK(&q->requeue_work, blk_mq_requeue_work); diff --git a/block/blk-rq-qos.c b/block/blk-rq-qos.c index 2cfb297d9a62..058f92c4f9d5 100644 --- a/block/blk-rq-qos.c +++ b/block/blk-rq-qos.c @@ -219,8 +219,8 @@ static int rq_qos_wake_function(struct wait_queue_entry *curr, data->got_token = true; smp_wmb(); - list_del_init(&curr->entry); wake_up_process(data->task); + list_del_init_careful(&curr->entry); return 1; } diff --git a/block/elevator.c b/block/elevator.c index 4122026b11f1..9430cde13d1a 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -106,8 +106,7 @@ static struct elevator_type *__elevator_find(const char *name) return NULL; } -static struct elevator_type *elevator_find_get(struct request_queue *q, - const char *name) +static struct elevator_type *elevator_find_get(const char *name) { struct elevator_type *e; @@ -551,7 +550,7 @@ EXPORT_SYMBOL_GPL(elv_unregister); static inline bool elv_support_iosched(struct request_queue *q) { if (!queue_is_mq(q) || - (q->tag_set && (q->tag_set->flags & BLK_MQ_F_NO_SCHED))) + (q->tag_set->flags & BLK_MQ_F_NO_SCHED)) return false; return true; } @@ -562,14 +561,14 @@ static inline bool elv_support_iosched(struct request_queue *q) */ static struct elevator_type *elevator_get_default(struct request_queue *q) { - if (q->tag_set && q->tag_set->flags & BLK_MQ_F_NO_SCHED_BY_DEFAULT) + if (q->tag_set->flags & BLK_MQ_F_NO_SCHED_BY_DEFAULT) return NULL; if (q->nr_hw_queues != 1 && !blk_mq_is_shared_tags(q->tag_set->flags)) return NULL; - return elevator_find_get(q, "mq-deadline"); + return elevator_find_get("mq-deadline"); } /* @@ -697,7 +696,7 @@ static int elevator_change(struct request_queue *q, const char *elevator_name) if (q->elevator && elevator_match(q->elevator->type, elevator_name)) return 0; - e = elevator_find_get(q, elevator_name); + e = elevator_find_get(elevator_name); if (!e) return -EINVAL; ret = elevator_switch(q, e); @@ -709,13 +708,21 @@ int elv_iosched_load_module(struct gendisk *disk, const char *buf, size_t count) { char elevator_name[ELV_NAME_MAX]; + struct elevator_type *found; + const char *name; if (!elv_support_iosched(disk->queue)) return -EOPNOTSUPP; strscpy(elevator_name, buf, sizeof(elevator_name)); + name = strstrip(elevator_name); - request_module("%s-iosched", strstrip(elevator_name)); + spin_lock(&elv_list_lock); + found = __elevator_find(name); + spin_unlock(&elv_list_lock); + + if (!found) + request_module("%s-iosched", name); return 0; } diff --git a/block/partitions/ldm.h b/block/partitions/ldm.h index 0a747a0c782d..e259180c8914 100644 --- a/block/partitions/ldm.h +++ b/block/partitions/ldm.h @@ -15,7 +15,7 @@ #include #include #include -#include +#include #include struct parsed_partitions; diff --git a/block/partitions/msdos.c b/block/partitions/msdos.c index b5d5c229cc3b..073be78ba0b0 100644 --- a/block/partitions/msdos.c +++ b/block/partitions/msdos.c @@ -36,7 +36,7 @@ * the nr_sects and start_sect partition table entries are * at a 2 (mod 4) address. */ -#include +#include static inline sector_t nr_sects(struct msdos_partition *p) { diff --git a/block/t10-pi.c b/block/t10-pi.c index e7052a728966..2d05421f0fa5 100644 --- a/block/t10-pi.c +++ b/block/t10-pi.c @@ -9,7 +9,7 @@ #include #include #include -#include +#include #include "blk.h" struct blk_integrity_iter { diff --git a/crypto/aes_generic.c b/crypto/aes_generic.c index 666474b81c6a..3c66d425c97b 100644 --- a/crypto/aes_generic.c +++ b/crypto/aes_generic.c @@ -54,7 +54,7 @@ #include #include #include -#include +#include static inline u8 byte(const u32 x, const unsigned n) { diff --git a/crypto/algapi.c b/crypto/algapi.c index 74e2261c184c..004d27e41315 100644 --- a/crypto/algapi.c +++ b/crypto/algapi.c @@ -373,7 +373,7 @@ found: q->cra_flags |= CRYPTO_ALG_DEAD; alg = test->adult; - if (list_empty(&alg->cra_list)) + if (crypto_is_dead(alg)) goto complete; if (err == -ECANCELED) diff --git a/crypto/blake2b_generic.c b/crypto/blake2b_generic.c index 32e380b714b6..04a712ddfb43 100644 --- a/crypto/blake2b_generic.c +++ b/crypto/blake2b_generic.c @@ -15,7 +15,7 @@ * More information about BLAKE2 can be found at https://blake2.net. */ -#include +#include #include #include #include diff --git a/crypto/blowfish_generic.c b/crypto/blowfish_generic.c index 0e74c7242e77..0146bc762c09 100644 --- a/crypto/blowfish_generic.c +++ b/crypto/blowfish_generic.c @@ -16,7 +16,7 @@ #include #include #include -#include +#include #include #include diff --git a/crypto/camellia_generic.c b/crypto/camellia_generic.c index c04670cf51ac..197fcf3abc89 100644 --- a/crypto/camellia_generic.c +++ b/crypto/camellia_generic.c @@ -15,7 +15,7 @@ #include #include #include -#include +#include static const u32 camellia_sp1110[256] = { 0x70707000, 0x82828200, 0x2c2c2c00, 0xececec00, diff --git a/crypto/cast5_generic.c b/crypto/cast5_generic.c index 085a1eedae03..f3e57775fa02 100644 --- a/crypto/cast5_generic.c +++ b/crypto/cast5_generic.c @@ -13,7 +13,7 @@ */ -#include +#include #include #include #include diff --git a/crypto/cast6_generic.c b/crypto/cast6_generic.c index 34f1ab53e3a7..11b725b12f27 100644 --- a/crypto/cast6_generic.c +++ b/crypto/cast6_generic.c @@ -10,7 +10,7 @@ */ -#include +#include #include #include #include diff --git a/crypto/chacha_generic.c b/crypto/chacha_generic.c index 8beea79ab117..ba7fcb47f9aa 100644 --- a/crypto/chacha_generic.c +++ b/crypto/chacha_generic.c @@ -6,7 +6,7 @@ * Copyright (C) 2018 Google LLC */ -#include +#include #include #include #include diff --git a/crypto/crc32_generic.c b/crypto/crc32_generic.c index a989cb44fd16..d1251663ed66 100644 --- a/crypto/crc32_generic.c +++ b/crypto/crc32_generic.c @@ -7,7 +7,7 @@ * This is crypto api shash wrappers to crc32_le. */ -#include +#include #include #include #include diff --git a/crypto/crc32c_generic.c b/crypto/crc32c_generic.c index 768614738541..a8c90b3f4c6c 100644 --- a/crypto/crc32c_generic.c +++ b/crypto/crc32c_generic.c @@ -30,7 +30,7 @@ * Copyright (c) 2008 Herbert Xu */ -#include +#include #include #include #include diff --git a/crypto/crc64_rocksoft_generic.c b/crypto/crc64_rocksoft_generic.c index 9e812bb26dba..ce0f3059b912 100644 --- a/crypto/crc64_rocksoft_generic.c +++ b/crypto/crc64_rocksoft_generic.c @@ -3,7 +3,7 @@ #include #include #include -#include +#include static int chksum_init(struct shash_desc *desc) { diff --git a/crypto/ecc.c b/crypto/ecc.c index 420decdad7d9..50ad2d4ed672 100644 --- a/crypto/ecc.c +++ b/crypto/ecc.c @@ -33,7 +33,7 @@ #include #include #include -#include +#include #include #include "ecc_curve_defs.h" diff --git a/crypto/michael_mic.c b/crypto/michael_mic.c index f4c31049601c..0d14e980d4d6 100644 --- a/crypto/michael_mic.c +++ b/crypto/michael_mic.c @@ -7,7 +7,7 @@ * Copyright (c) 2004 Jouni Malinen */ #include -#include +#include #include #include #include diff --git a/crypto/nhpoly1305.c b/crypto/nhpoly1305.c index 8a3006c3b51b..a661d4f667cd 100644 --- a/crypto/nhpoly1305.c +++ b/crypto/nhpoly1305.c @@ -30,7 +30,7 @@ * (https://cr.yp.to/mac/poly1305-20050329.pdf) */ -#include +#include #include #include #include diff --git a/crypto/poly1305_generic.c b/crypto/poly1305_generic.c index 94af47eb6fa6..e6f29a98725a 100644 --- a/crypto/poly1305_generic.c +++ b/crypto/poly1305_generic.c @@ -17,7 +17,7 @@ #include #include #include -#include +#include static int crypto_poly1305_init(struct shash_desc *desc) { diff --git a/crypto/polyval-generic.c b/crypto/polyval-generic.c index 16bfa6925b31..4f98910bcdb5 100644 --- a/crypto/polyval-generic.c +++ b/crypto/polyval-generic.c @@ -44,7 +44,7 @@ * */ -#include +#include #include #include #include diff --git a/crypto/serpent_generic.c b/crypto/serpent_generic.c index c6bca47931e2..f6ef187be6fe 100644 --- a/crypto/serpent_generic.c +++ b/crypto/serpent_generic.c @@ -11,7 +11,7 @@ #include #include #include -#include +#include #include #include diff --git a/crypto/sha256_generic.c b/crypto/sha256_generic.c index bf147b01e313..b00521f1a6d4 100644 --- a/crypto/sha256_generic.c +++ b/crypto/sha256_generic.c @@ -15,7 +15,7 @@ #include #include #include -#include +#include const u8 sha224_zero_message_hash[SHA224_DIGEST_SIZE] = { 0xd1, 0x4a, 0x02, 0x8c, 0x2a, 0x3a, 0x2b, 0xc9, 0x47, diff --git a/crypto/sha3_generic.c b/crypto/sha3_generic.c index 3e4069935b53..b103642b56ea 100644 --- a/crypto/sha3_generic.c +++ b/crypto/sha3_generic.c @@ -13,7 +13,7 @@ #include #include #include -#include +#include /* * On some 32-bit architectures (h8300), GCC ends up using diff --git a/crypto/sha512_generic.c b/crypto/sha512_generic.c index be70e76d6d86..ed81813bd420 100644 --- a/crypto/sha512_generic.c +++ b/crypto/sha512_generic.c @@ -16,7 +16,7 @@ #include #include #include -#include +#include const u8 sha384_zero_message_hash[SHA384_DIGEST_SIZE] = { 0x38, 0xb0, 0x60, 0xa7, 0x51, 0xac, 0x96, 0x38, diff --git a/crypto/sm3.c b/crypto/sm3.c index d473e358a873..18c2fb73ba16 100644 --- a/crypto/sm3.c +++ b/crypto/sm3.c @@ -9,7 +9,7 @@ */ #include -#include +#include #include static const u32 ____cacheline_aligned K[64] = { diff --git a/crypto/sm3_generic.c b/crypto/sm3_generic.c index a215c1c37e73..a2d23a46924e 100644 --- a/crypto/sm3_generic.c +++ b/crypto/sm3_generic.c @@ -17,7 +17,7 @@ #include #include #include -#include +#include const u8 sm3_zero_message_hash[SM3_DIGEST_SIZE] = { 0x1A, 0xB2, 0x1D, 0x83, 0x55, 0xCF, 0xA1, 0x7F, diff --git a/crypto/sm4.c b/crypto/sm4.c index 2c44193bc27e..f4cd7edc11f0 100644 --- a/crypto/sm4.c +++ b/crypto/sm4.c @@ -8,7 +8,7 @@ */ #include -#include +#include #include static const u32 ____cacheline_aligned fk[4] = { diff --git a/crypto/sm4_generic.c b/crypto/sm4_generic.c index 560eba37dc55..7df86369ac00 100644 --- a/crypto/sm4_generic.c +++ b/crypto/sm4_generic.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include /** * sm4_setkey - Set the SM4 key. diff --git a/crypto/testmgr.c b/crypto/testmgr.c index ee8da628e9da..2f5f6b52b2d4 100644 --- a/crypto/testmgr.c +++ b/crypto/testmgr.c @@ -1940,7 +1940,7 @@ static int __alg_test_hash(const struct hash_testvec *vecs, atfm = crypto_alloc_ahash(driver, type, mask); if (IS_ERR(atfm)) { if (PTR_ERR(atfm) == -ENOENT) - return -ENOENT; + return 0; pr_err("alg: hash: failed to allocate transform for %s: %ld\n", driver, PTR_ERR(atfm)); return PTR_ERR(atfm); @@ -2706,7 +2706,7 @@ static int alg_test_aead(const struct alg_test_desc *desc, const char *driver, tfm = crypto_alloc_aead(driver, type, mask); if (IS_ERR(tfm)) { if (PTR_ERR(tfm) == -ENOENT) - return -ENOENT; + return 0; pr_err("alg: aead: failed to allocate transform for %s: %ld\n", driver, PTR_ERR(tfm)); return PTR_ERR(tfm); @@ -3285,7 +3285,7 @@ static int alg_test_skcipher(const struct alg_test_desc *desc, tfm = crypto_alloc_skcipher(driver, type, mask); if (IS_ERR(tfm)) { if (PTR_ERR(tfm) == -ENOENT) - return -ENOENT; + return 0; pr_err("alg: skcipher: failed to allocate transform for %s: %ld\n", driver, PTR_ERR(tfm)); return PTR_ERR(tfm); @@ -3700,7 +3700,7 @@ static int alg_test_cipher(const struct alg_test_desc *desc, tfm = crypto_alloc_cipher(driver, type, mask); if (IS_ERR(tfm)) { if (PTR_ERR(tfm) == -ENOENT) - return -ENOENT; + return 0; printk(KERN_ERR "alg: cipher: Failed to load transform for " "%s: %ld\n", driver, PTR_ERR(tfm)); return PTR_ERR(tfm); @@ -3726,7 +3726,7 @@ static int alg_test_comp(const struct alg_test_desc *desc, const char *driver, acomp = crypto_alloc_acomp(driver, type, mask); if (IS_ERR(acomp)) { if (PTR_ERR(acomp) == -ENOENT) - return -ENOENT; + return 0; pr_err("alg: acomp: Failed to load transform for %s: %ld\n", driver, PTR_ERR(acomp)); return PTR_ERR(acomp); @@ -3740,7 +3740,7 @@ static int alg_test_comp(const struct alg_test_desc *desc, const char *driver, comp = crypto_alloc_comp(driver, type, mask); if (IS_ERR(comp)) { if (PTR_ERR(comp) == -ENOENT) - return -ENOENT; + return 0; pr_err("alg: comp: Failed to load transform for %s: %ld\n", driver, PTR_ERR(comp)); return PTR_ERR(comp); @@ -3818,7 +3818,7 @@ static int alg_test_cprng(const struct alg_test_desc *desc, const char *driver, rng = crypto_alloc_rng(driver, type, mask); if (IS_ERR(rng)) { if (PTR_ERR(rng) == -ENOENT) - return -ENOENT; + return 0; printk(KERN_ERR "alg: cprng: Failed to load transform for %s: " "%ld\n", driver, PTR_ERR(rng)); return PTR_ERR(rng); @@ -3846,12 +3846,11 @@ static int drbg_cavs_test(const struct drbg_testvec *test, int pr, drng = crypto_alloc_rng(driver, type, mask); if (IS_ERR(drng)) { + kfree_sensitive(buf); if (PTR_ERR(drng) == -ENOENT) - goto out_no_rng; + return 0; printk(KERN_ERR "alg: drbg: could not allocate DRNG handle for " "%s\n", driver); -out_no_rng: - kfree_sensitive(buf); return PTR_ERR(drng); } @@ -4095,7 +4094,7 @@ static int alg_test_kpp(const struct alg_test_desc *desc, const char *driver, tfm = crypto_alloc_kpp(driver, type, mask); if (IS_ERR(tfm)) { if (PTR_ERR(tfm) == -ENOENT) - return -ENOENT; + return 0; pr_err("alg: kpp: Failed to load tfm for %s: %ld\n", driver, PTR_ERR(tfm)); return PTR_ERR(tfm); @@ -4325,7 +4324,7 @@ static int alg_test_akcipher(const struct alg_test_desc *desc, tfm = crypto_alloc_akcipher(driver, type, mask); if (IS_ERR(tfm)) { if (PTR_ERR(tfm) == -ENOENT) - return -ENOENT; + return 0; pr_err("alg: akcipher: Failed to load tfm for %s: %ld\n", driver, PTR_ERR(tfm)); return PTR_ERR(tfm); diff --git a/crypto/twofish_generic.c b/crypto/twofish_generic.c index 557915e4062d..19f2b365e140 100644 --- a/crypto/twofish_generic.c +++ b/crypto/twofish_generic.c @@ -24,7 +24,7 @@ * Third Edition. */ -#include +#include #include #include #include diff --git a/crypto/vmac.c b/crypto/vmac.c index 0a1d8efa6c1a..bd9d70eac22e 100644 --- a/crypto/vmac.c +++ b/crypto/vmac.c @@ -28,7 +28,7 @@ * Last modified: 17 APR 08, 1700 PDT */ -#include +#include #include #include #include diff --git a/crypto/xxhash_generic.c b/crypto/xxhash_generic.c index 55d1c8a76127..ac206ad4184d 100644 --- a/crypto/xxhash_generic.c +++ b/crypto/xxhash_generic.c @@ -4,7 +4,7 @@ #include #include #include -#include +#include #define XXHASH64_BLOCK_SIZE 32 #define XXHASH64_DIGEST_SIZE 8 diff --git a/drivers/accel/ivpu/Kconfig b/drivers/accel/ivpu/Kconfig index 682c53245286..9e055b5ce03d 100644 --- a/drivers/accel/ivpu/Kconfig +++ b/drivers/accel/ivpu/Kconfig @@ -8,6 +8,7 @@ config DRM_ACCEL_IVPU select FW_LOADER select DRM_GEM_SHMEM_HELPER select GENERIC_ALLOCATOR + select WANT_DEV_COREDUMP help Choose this option if you have a system with an 14th generation Intel CPU (Meteor Lake) or newer. Intel NPU (formerly called Intel VPU) @@ -15,3 +16,12 @@ config DRM_ACCEL_IVPU and Deep Learning applications. If "M" is selected, the module will be called intel_vpu. + +config DRM_ACCEL_IVPU_DEBUG + bool "Intel NPU debug mode" + depends on DRM_ACCEL_IVPU + help + Choose this option to enable additional + debug features for the Intel NPU driver: + - Always print debug messages regardless of dyndbg config, + - Enable unsafe module params. diff --git a/drivers/accel/ivpu/Makefile b/drivers/accel/ivpu/Makefile index ebd682a42eb1..1029e0bab061 100644 --- a/drivers/accel/ivpu/Makefile +++ b/drivers/accel/ivpu/Makefile @@ -16,8 +16,14 @@ intel_vpu-y := \ ivpu_mmu_context.o \ ivpu_ms.o \ ivpu_pm.o \ - ivpu_sysfs.o + ivpu_sysfs.o \ + ivpu_trace_points.o intel_vpu-$(CONFIG_DEBUG_FS) += ivpu_debugfs.o +intel_vpu-$(CONFIG_DEV_COREDUMP) += ivpu_coredump.o obj-$(CONFIG_DRM_ACCEL_IVPU) += intel_vpu.o + +subdir-ccflags-$(CONFIG_DRM_ACCEL_IVPU_DEBUG) += -DDEBUG + +CFLAGS_ivpu_trace_points.o = -I$(src) diff --git a/drivers/accel/ivpu/ivpu_coredump.c b/drivers/accel/ivpu/ivpu_coredump.c new file mode 100644 index 000000000000..16ad0c30818c --- /dev/null +++ b/drivers/accel/ivpu/ivpu_coredump.c @@ -0,0 +1,39 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2020-2024 Intel Corporation + */ + +#include +#include + +#include "ivpu_coredump.h" +#include "ivpu_fw.h" +#include "ivpu_gem.h" +#include "vpu_boot_api.h" + +#define CRASH_DUMP_HEADER "Intel NPU crash dump" +#define CRASH_DUMP_HEADERS_SIZE SZ_4K + +void ivpu_dev_coredump(struct ivpu_device *vdev) +{ + struct drm_print_iterator pi = {}; + struct drm_printer p; + size_t coredump_size; + char *coredump; + + coredump_size = CRASH_DUMP_HEADERS_SIZE + FW_VERSION_HEADER_SIZE + + ivpu_bo_size(vdev->fw->mem_log_crit) + ivpu_bo_size(vdev->fw->mem_log_verb); + coredump = vmalloc(coredump_size); + if (!coredump) + return; + + pi.data = coredump; + pi.remain = coredump_size; + p = drm_coredump_printer(&pi); + + drm_printf(&p, "%s\n", CRASH_DUMP_HEADER); + drm_printf(&p, "FW version: %s\n", vdev->fw->version); + ivpu_fw_log_print(vdev, false, &p); + + dev_coredumpv(vdev->drm.dev, coredump, pi.offset, GFP_KERNEL); +} diff --git a/drivers/accel/ivpu/ivpu_coredump.h b/drivers/accel/ivpu/ivpu_coredump.h new file mode 100644 index 000000000000..8efb09d02441 --- /dev/null +++ b/drivers/accel/ivpu/ivpu_coredump.h @@ -0,0 +1,25 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2020-2024 Intel Corporation + */ + +#ifndef __IVPU_COREDUMP_H__ +#define __IVPU_COREDUMP_H__ + +#include + +#include "ivpu_drv.h" +#include "ivpu_fw_log.h" + +#ifdef CONFIG_DEV_COREDUMP +void ivpu_dev_coredump(struct ivpu_device *vdev); +#else +static inline void ivpu_dev_coredump(struct ivpu_device *vdev) +{ + struct drm_printer p = drm_info_printer(vdev->drm.dev); + + ivpu_fw_log_print(vdev, false, &p); +} +#endif + +#endif /* __IVPU_COREDUMP_H__ */ diff --git a/drivers/accel/ivpu/ivpu_debugfs.c b/drivers/accel/ivpu/ivpu_debugfs.c index 6f86f8df30db..8180b95ed69d 100644 --- a/drivers/accel/ivpu/ivpu_debugfs.c +++ b/drivers/accel/ivpu/ivpu_debugfs.c @@ -45,6 +45,14 @@ static int fw_name_show(struct seq_file *s, void *v) return 0; } +static int fw_version_show(struct seq_file *s, void *v) +{ + struct ivpu_device *vdev = seq_to_ivpu(s); + + seq_printf(s, "%s\n", vdev->fw->version); + return 0; +} + static int fw_trace_capability_show(struct seq_file *s, void *v) { struct ivpu_device *vdev = seq_to_ivpu(s); @@ -108,42 +116,43 @@ static int reset_pending_show(struct seq_file *s, void *v) return 0; } +static int firewall_irq_counter_show(struct seq_file *s, void *v) +{ + struct ivpu_device *vdev = seq_to_ivpu(s); + + seq_printf(s, "%d\n", atomic_read(&vdev->hw->firewall_irq_counter)); + return 0; +} + static const struct drm_debugfs_info vdev_debugfs_list[] = { {"bo_list", bo_list_show, 0}, {"fw_name", fw_name_show, 0}, + {"fw_version", fw_version_show, 0}, {"fw_trace_capability", fw_trace_capability_show, 0}, {"fw_trace_config", fw_trace_config_show, 0}, {"last_bootmode", last_bootmode_show, 0}, {"reset_counter", reset_counter_show, 0}, {"reset_pending", reset_pending_show, 0}, + {"firewall_irq_counter", firewall_irq_counter_show, 0}, }; -static ssize_t -dvfs_mode_fops_write(struct file *file, const char __user *user_buf, size_t size, loff_t *pos) +static int dvfs_mode_get(void *data, u64 *dvfs_mode) { - struct ivpu_device *vdev = file->private_data; - struct ivpu_fw_info *fw = vdev->fw; - u32 dvfs_mode; - int ret; + struct ivpu_device *vdev = (struct ivpu_device *)data; - ret = kstrtou32_from_user(user_buf, size, 0, &dvfs_mode); - if (ret < 0) - return ret; - - fw->dvfs_mode = dvfs_mode; - - ret = pci_try_reset_function(to_pci_dev(vdev->drm.dev)); - if (ret) - return ret; - - return size; + *dvfs_mode = vdev->fw->dvfs_mode; + return 0; } -static const struct file_operations dvfs_mode_fops = { - .owner = THIS_MODULE, - .open = simple_open, - .write = dvfs_mode_fops_write, -}; +static int dvfs_mode_set(void *data, u64 dvfs_mode) +{ + struct ivpu_device *vdev = (struct ivpu_device *)data; + + vdev->fw->dvfs_mode = (u32)dvfs_mode; + return pci_try_reset_function(to_pci_dev(vdev->drm.dev)); +} + +DEFINE_DEBUGFS_ATTRIBUTE(dvfs_mode_fops, dvfs_mode_get, dvfs_mode_set, "%llu\n"); static ssize_t fw_dyndbg_fops_write(struct file *file, const char __user *user_buf, size_t size, loff_t *pos) @@ -192,7 +201,7 @@ fw_log_fops_write(struct file *file, const char __user *user_buf, size_t size, l if (!size) return -EINVAL; - ivpu_fw_log_clear(vdev); + ivpu_fw_log_mark_read(vdev); return size; } @@ -337,49 +346,23 @@ static const struct file_operations ivpu_force_recovery_fops = { .write = ivpu_force_recovery_fn, }; -static ssize_t -ivpu_reset_engine_fn(struct file *file, const char __user *user_buf, size_t size, loff_t *pos) +static int ivpu_reset_engine_fn(void *data, u64 val) { - struct ivpu_device *vdev = file->private_data; + struct ivpu_device *vdev = (struct ivpu_device *)data; - if (!size) - return -EINVAL; - - if (ivpu_jsm_reset_engine(vdev, DRM_IVPU_ENGINE_COMPUTE)) - return -ENODEV; - if (ivpu_jsm_reset_engine(vdev, DRM_IVPU_ENGINE_COPY)) - return -ENODEV; - - return size; + return ivpu_jsm_reset_engine(vdev, (u32)val); } -static const struct file_operations ivpu_reset_engine_fops = { - .owner = THIS_MODULE, - .open = simple_open, - .write = ivpu_reset_engine_fn, -}; +DEFINE_DEBUGFS_ATTRIBUTE(ivpu_reset_engine_fops, NULL, ivpu_reset_engine_fn, "0x%02llx\n"); -static ssize_t -ivpu_resume_engine_fn(struct file *file, const char __user *user_buf, size_t size, loff_t *pos) +static int ivpu_resume_engine_fn(void *data, u64 val) { - struct ivpu_device *vdev = file->private_data; + struct ivpu_device *vdev = (struct ivpu_device *)data; - if (!size) - return -EINVAL; - - if (ivpu_jsm_hws_resume_engine(vdev, DRM_IVPU_ENGINE_COMPUTE)) - return -ENODEV; - if (ivpu_jsm_hws_resume_engine(vdev, DRM_IVPU_ENGINE_COPY)) - return -ENODEV; - - return size; + return ivpu_jsm_hws_resume_engine(vdev, (u32)val); } -static const struct file_operations ivpu_resume_engine_fops = { - .owner = THIS_MODULE, - .open = simple_open, - .write = ivpu_resume_engine_fn, -}; +DEFINE_DEBUGFS_ATTRIBUTE(ivpu_resume_engine_fops, NULL, ivpu_resume_engine_fn, "0x%02llx\n"); static int dct_active_get(void *data, u64 *active_percent) { @@ -423,7 +406,7 @@ void ivpu_debugfs_init(struct ivpu_device *vdev) debugfs_create_file("force_recovery", 0200, debugfs_root, vdev, &ivpu_force_recovery_fops); - debugfs_create_file("dvfs_mode", 0200, debugfs_root, vdev, + debugfs_create_file("dvfs_mode", 0644, debugfs_root, vdev, &dvfs_mode_fops); debugfs_create_file("fw_dyndbg", 0200, debugfs_root, vdev, diff --git a/drivers/accel/ivpu/ivpu_drv.c b/drivers/accel/ivpu/ivpu_drv.c index c91400ecf926..ca2bf47ce248 100644 --- a/drivers/accel/ivpu/ivpu_drv.c +++ b/drivers/accel/ivpu/ivpu_drv.c @@ -7,6 +7,7 @@ #include #include #include +#include #include #include @@ -14,7 +15,7 @@ #include #include -#include "vpu_boot_api.h" +#include "ivpu_coredump.h" #include "ivpu_debugfs.h" #include "ivpu_drv.h" #include "ivpu_fw.h" @@ -29,10 +30,10 @@ #include "ivpu_ms.h" #include "ivpu_pm.h" #include "ivpu_sysfs.h" +#include "vpu_boot_api.h" #ifndef DRIVER_VERSION_STR -#define DRIVER_VERSION_STR __stringify(DRM_IVPU_DRIVER_MAJOR) "." \ - __stringify(DRM_IVPU_DRIVER_MINOR) "." +#define DRIVER_VERSION_STR "1.0.0 " UTS_RELEASE #endif static struct lock_class_key submitted_jobs_xa_lock_class_key; @@ -42,8 +43,10 @@ module_param_named(dbg_mask, ivpu_dbg_mask, int, 0644); MODULE_PARM_DESC(dbg_mask, "Driver debug mask. See IVPU_DBG_* macros."); int ivpu_test_mode; +#if IS_ENABLED(CONFIG_DRM_ACCEL_IVPU_DEBUG) module_param_named_unsafe(test_mode, ivpu_test_mode, int, 0644); MODULE_PARM_DESC(test_mode, "Test mode mask. See IVPU_TEST_MODE_* macros."); +#endif u8 ivpu_pll_min_ratio; module_param_named(pll_min_ratio, ivpu_pll_min_ratio, byte, 0644); @@ -53,9 +56,9 @@ u8 ivpu_pll_max_ratio = U8_MAX; module_param_named(pll_max_ratio, ivpu_pll_max_ratio, byte, 0644); MODULE_PARM_DESC(pll_max_ratio, "Maximum PLL ratio used to set NPU frequency"); -int ivpu_sched_mode; +int ivpu_sched_mode = IVPU_SCHED_MODE_AUTO; module_param_named(sched_mode, ivpu_sched_mode, int, 0444); -MODULE_PARM_DESC(sched_mode, "Scheduler mode: 0 - Default scheduler, 1 - Force HW scheduler"); +MODULE_PARM_DESC(sched_mode, "Scheduler mode: -1 - Use default scheduler, 0 - Use OS scheduler, 1 - Use HW scheduler"); bool ivpu_disable_mmu_cont_pages; module_param_named(disable_mmu_cont_pages, ivpu_disable_mmu_cont_pages, bool, 0444); @@ -85,7 +88,7 @@ static void file_priv_unbind(struct ivpu_device *vdev, struct ivpu_file_priv *fi ivpu_cmdq_release_all_locked(file_priv); ivpu_bo_unbind_all_bos_from_context(vdev, &file_priv->ctx); - ivpu_mmu_user_context_fini(vdev, &file_priv->ctx); + ivpu_mmu_context_fini(vdev, &file_priv->ctx); file_priv->bound = false; drm_WARN_ON(&vdev->drm, !xa_erase_irq(&vdev->context_xa, file_priv->ctx.id)); } @@ -103,6 +106,8 @@ static void file_priv_release(struct kref *ref) pm_runtime_get_sync(vdev->drm.dev); mutex_lock(&vdev->context_list_lock); file_priv_unbind(vdev, file_priv); + drm_WARN_ON(&vdev->drm, !xa_empty(&file_priv->cmdq_xa)); + xa_destroy(&file_priv->cmdq_xa); mutex_unlock(&vdev->context_list_lock); pm_runtime_put_autosuspend(vdev->drm.dev); @@ -116,8 +121,6 @@ void ivpu_file_priv_put(struct ivpu_file_priv **link) struct ivpu_file_priv *file_priv = *link; struct ivpu_device *vdev = file_priv->vdev; - drm_WARN_ON(&vdev->drm, !file_priv); - ivpu_dbg(vdev, KREF, "file_priv put: ctx %u refcount %u\n", file_priv->ctx.id, kref_read(&file_priv->ref)); @@ -255,9 +258,14 @@ static int ivpu_open(struct drm_device *dev, struct drm_file *file) goto err_unlock; } - ret = ivpu_mmu_user_context_init(vdev, &file_priv->ctx, ctx_id); - if (ret) - goto err_xa_erase; + ivpu_mmu_context_init(vdev, &file_priv->ctx, ctx_id); + + file_priv->job_limit.min = FIELD_PREP(IVPU_JOB_ID_CONTEXT_MASK, (file_priv->ctx.id - 1)); + file_priv->job_limit.max = file_priv->job_limit.min | IVPU_JOB_ID_JOB_MASK; + + xa_init_flags(&file_priv->cmdq_xa, XA_FLAGS_ALLOC1); + file_priv->cmdq_limit.min = IVPU_CMDQ_MIN_ID; + file_priv->cmdq_limit.max = IVPU_CMDQ_MAX_ID; mutex_unlock(&vdev->context_list_lock); drm_dev_exit(idx); @@ -269,8 +277,6 @@ static int ivpu_open(struct drm_device *dev, struct drm_file *file) return 0; -err_xa_erase: - xa_erase_irq(&vdev->context_xa, ctx_id); err_unlock: mutex_unlock(&vdev->context_list_lock); mutex_destroy(&file_priv->ms_lock); @@ -346,7 +352,7 @@ static int ivpu_hw_sched_init(struct ivpu_device *vdev) { int ret = 0; - if (vdev->hw->sched_mode == VPU_SCHEDULING_MODE_HW) { + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) { ret = ivpu_jsm_hws_setup_priority_bands(vdev); if (ret) { ivpu_err(vdev, "Failed to enable hw scheduler: %d", ret); @@ -380,10 +386,7 @@ int ivpu_boot(struct ivpu_device *vdev) ret = ivpu_wait_for_ready(vdev); if (ret) { ivpu_err(vdev, "Failed to boot the firmware: %d\n", ret); - ivpu_hw_diagnose_failure(vdev); - ivpu_mmu_evtq_dump(vdev); - ivpu_fw_log_dump(vdev); - return ret; + goto err_diagnose_failure; } ivpu_hw_irq_clear(vdev); @@ -394,12 +397,20 @@ int ivpu_boot(struct ivpu_device *vdev) if (ivpu_fw_is_cold_boot(vdev)) { ret = ivpu_pm_dct_init(vdev); if (ret) - return ret; + goto err_diagnose_failure; - return ivpu_hw_sched_init(vdev); + ret = ivpu_hw_sched_init(vdev); + if (ret) + goto err_diagnose_failure; } return 0; + +err_diagnose_failure: + ivpu_hw_diagnose_failure(vdev); + ivpu_mmu_evtq_dump(vdev); + ivpu_dev_coredump(vdev); + return ret; } void ivpu_prepare_for_reset(struct ivpu_device *vdev) @@ -446,9 +457,16 @@ static const struct drm_driver driver = { .name = DRIVER_NAME, .desc = DRIVER_DESC, + +#ifdef DRIVER_DATE .date = DRIVER_DATE, - .major = DRM_IVPU_DRIVER_MAJOR, - .minor = DRM_IVPU_DRIVER_MINOR, + .major = DRIVER_MAJOR, + .minor = DRIVER_MINOR, + .patchlevel = DRIVER_PATCHLEVEL, +#else + .date = UTS_RELEASE, + .major = 1, +#endif }; static void ivpu_context_abort_invalid(struct ivpu_device *vdev) @@ -606,6 +624,9 @@ static int ivpu_dev_init(struct ivpu_device *vdev) lockdep_set_class(&vdev->submitted_jobs_xa.xa_lock, &submitted_jobs_xa_lock_class_key); INIT_LIST_HEAD(&vdev->bo_list); + vdev->db_limit.min = IVPU_MIN_DB; + vdev->db_limit.max = IVPU_MAX_DB; + ret = drmm_mutex_init(&vdev->drm, &vdev->context_list_lock); if (ret) goto err_xa_destroy; @@ -632,9 +653,7 @@ static int ivpu_dev_init(struct ivpu_device *vdev) if (ret) goto err_shutdown; - ret = ivpu_mmu_global_context_init(vdev); - if (ret) - goto err_shutdown; + ivpu_mmu_global_context_init(vdev); ret = ivpu_mmu_init(vdev); if (ret) @@ -722,6 +741,7 @@ static struct pci_device_id ivpu_pci_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_MTL) }, { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_ARL) }, { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_LNL) }, + { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_PTL_P) }, { } }; MODULE_DEVICE_TABLE(pci, ivpu_pci_ids); diff --git a/drivers/accel/ivpu/ivpu_drv.h b/drivers/accel/ivpu/ivpu_drv.h index 63f13b697eed..3fdff3f6cffd 100644 --- a/drivers/accel/ivpu/ivpu_drv.h +++ b/drivers/accel/ivpu/ivpu_drv.h @@ -21,11 +21,11 @@ #define DRIVER_NAME "intel_vpu" #define DRIVER_DESC "Driver for Intel NPU (Neural Processing Unit)" -#define DRIVER_DATE "20230117" -#define PCI_DEVICE_ID_MTL 0x7d1d -#define PCI_DEVICE_ID_ARL 0xad1d -#define PCI_DEVICE_ID_LNL 0x643e +#define PCI_DEVICE_ID_MTL 0x7d1d +#define PCI_DEVICE_ID_ARL 0xad1d +#define PCI_DEVICE_ID_LNL 0x643e +#define PCI_DEVICE_ID_PTL_P 0xb03e #define IVPU_HW_IP_37XX 37 #define IVPU_HW_IP_40XX 40 @@ -46,17 +46,22 @@ #define IVPU_MIN_DB 1 #define IVPU_MAX_DB 255 -#define IVPU_NUM_ENGINES 2 -#define IVPU_NUM_PRIORITIES 4 -#define IVPU_NUM_CMDQS_PER_CTX (IVPU_NUM_ENGINES * IVPU_NUM_PRIORITIES) +#define IVPU_JOB_ID_JOB_MASK GENMASK(7, 0) +#define IVPU_JOB_ID_CONTEXT_MASK GENMASK(31, 8) -#define IVPU_CMDQ_INDEX(engine, priority) ((engine) * IVPU_NUM_PRIORITIES + (priority)) +#define IVPU_NUM_PRIORITIES 4 +#define IVPU_NUM_CMDQS_PER_CTX (IVPU_NUM_PRIORITIES) + +#define IVPU_CMDQ_MIN_ID 1 +#define IVPU_CMDQ_MAX_ID 255 #define IVPU_PLATFORM_SILICON 0 #define IVPU_PLATFORM_SIMICS 2 #define IVPU_PLATFORM_FPGA 3 #define IVPU_PLATFORM_INVALID 8 +#define IVPU_SCHED_MODE_AUTO -1 + #define IVPU_DBG_REG BIT(0) #define IVPU_DBG_IRQ BIT(1) #define IVPU_DBG_MMU BIT(2) @@ -134,6 +139,8 @@ struct ivpu_device { struct xa_limit context_xa_limit; struct xarray db_xa; + struct xa_limit db_limit; + u32 db_next; struct mutex bo_list_lock; /* Protects bo_list */ struct list_head bo_list; @@ -152,6 +159,7 @@ struct ivpu_device { int tdr; int autosuspend; int d0i3_entry_msg; + int state_dump_msg; } timeout; }; @@ -163,11 +171,15 @@ struct ivpu_file_priv { struct kref ref; struct ivpu_device *vdev; struct mutex lock; /* Protects cmdq */ - struct ivpu_cmdq *cmdq[IVPU_NUM_CMDQS_PER_CTX]; + struct xarray cmdq_xa; struct ivpu_mmu_context ctx; struct mutex ms_lock; /* Protects ms_instance_list, ms_info_bo */ struct list_head ms_instance_list; struct ivpu_bo *ms_info_bo; + struct xa_limit job_limit; + u32 job_id_next; + struct xa_limit cmdq_limit; + u32 cmdq_id_next; bool has_mmu_faults; bool bound; bool aborted; @@ -185,9 +197,9 @@ extern bool ivpu_force_snoop; #define IVPU_TEST_MODE_NULL_SUBMISSION BIT(2) #define IVPU_TEST_MODE_D0I3_MSG_DISABLE BIT(4) #define IVPU_TEST_MODE_D0I3_MSG_ENABLE BIT(5) -#define IVPU_TEST_MODE_PREEMPTION_DISABLE BIT(6) -#define IVPU_TEST_MODE_HWS_EXTRA_EVENTS BIT(7) +#define IVPU_TEST_MODE_MIP_DISABLE BIT(6) #define IVPU_TEST_MODE_DISABLE_TIMEOUTS BIT(8) +#define IVPU_TEST_MODE_TURBO BIT(9) extern int ivpu_test_mode; struct ivpu_file_priv *ivpu_file_priv_get(struct ivpu_file_priv *file_priv); @@ -215,6 +227,8 @@ static inline int ivpu_hw_ip_gen(struct ivpu_device *vdev) return IVPU_HW_IP_37XX; case PCI_DEVICE_ID_LNL: return IVPU_HW_IP_40XX; + case PCI_DEVICE_ID_PTL_P: + return IVPU_HW_IP_50XX; default: dump_stack(); ivpu_err(vdev, "Unknown NPU IP generation\n"); @@ -229,6 +243,7 @@ static inline int ivpu_hw_btrs_gen(struct ivpu_device *vdev) case PCI_DEVICE_ID_ARL: return IVPU_HW_BTRS_MTL; case PCI_DEVICE_ID_LNL: + case PCI_DEVICE_ID_PTL_P: return IVPU_HW_BTRS_LNL; default: dump_stack(); diff --git a/drivers/accel/ivpu/ivpu_fw.c b/drivers/accel/ivpu/ivpu_fw.c index ede6165e09d9..6037ec0b3096 100644 --- a/drivers/accel/ivpu/ivpu_fw.c +++ b/drivers/accel/ivpu/ivpu_fw.c @@ -25,7 +25,6 @@ #define FW_SHAVE_NN_MAX_SIZE SZ_2M #define FW_RUNTIME_MIN_ADDR (FW_GLOBAL_MEM_START) #define FW_RUNTIME_MAX_ADDR (FW_GLOBAL_MEM_END - FW_SHARED_MEM_SIZE) -#define FW_VERSION_HEADER_SIZE SZ_4K #define FW_FILE_IMAGE_OFFSET (VPU_FW_HEADER_SIZE + FW_VERSION_HEADER_SIZE) #define WATCHDOG_MSS_REDIRECT 32 @@ -47,8 +46,10 @@ #define IVPU_FOCUS_PRESENT_TIMER_MS 1000 static char *ivpu_firmware; +#if IS_ENABLED(CONFIG_DRM_ACCEL_IVPU_DEBUG) module_param_named_unsafe(firmware, ivpu_firmware, charp, 0644); MODULE_PARM_DESC(firmware, "NPU firmware binary in /lib/firmware/.."); +#endif static struct { int gen; @@ -58,11 +59,14 @@ static struct { { IVPU_HW_IP_37XX, "intel/vpu/vpu_37xx_v0.0.bin" }, { IVPU_HW_IP_40XX, "vpu_40xx.bin" }, { IVPU_HW_IP_40XX, "intel/vpu/vpu_40xx_v0.0.bin" }, + { IVPU_HW_IP_50XX, "vpu_50xx.bin" }, + { IVPU_HW_IP_50XX, "intel/vpu/vpu_50xx_v0.0.bin" }, }; /* Production fw_names from the table above */ MODULE_FIRMWARE("intel/vpu/vpu_37xx_v0.0.bin"); MODULE_FIRMWARE("intel/vpu/vpu_40xx_v0.0.bin"); +MODULE_FIRMWARE("intel/vpu/vpu_50xx_v0.0.bin"); static int ivpu_fw_request(struct ivpu_device *vdev) { @@ -135,6 +139,15 @@ static bool is_within_range(u64 addr, size_t size, u64 range_start, size_t range return true; } +static u32 +ivpu_fw_sched_mode_select(struct ivpu_device *vdev, const struct vpu_firmware_header *fw_hdr) +{ + if (ivpu_sched_mode != IVPU_SCHED_MODE_AUTO) + return ivpu_sched_mode; + + return VPU_SCHEDULING_MODE_OS; +} + static int ivpu_fw_parse(struct ivpu_device *vdev) { struct ivpu_fw_info *fw = vdev->fw; @@ -191,8 +204,10 @@ static int ivpu_fw_parse(struct ivpu_device *vdev) ivpu_dbg(vdev, FW_BOOT, "Header version: 0x%x, format 0x%x\n", fw_hdr->header_version, fw_hdr->image_format); - ivpu_info(vdev, "Firmware: %s, version: %s", fw->name, - (const char *)fw_hdr + VPU_FW_HEADER_SIZE); + if (!scnprintf(fw->version, sizeof(fw->version), "%s", fw->file->data + VPU_FW_HEADER_SIZE)) + ivpu_warn(vdev, "Missing firmware version\n"); + + ivpu_info(vdev, "Firmware: %s, version: %s\n", fw->name, fw->version); if (IVPU_FW_CHECK_API_COMPAT(vdev, fw_hdr, BOOT, 3)) return -EINVAL; @@ -208,14 +223,16 @@ static int ivpu_fw_parse(struct ivpu_device *vdev) fw->cold_boot_entry_point = fw_hdr->entry_point; fw->entry_point = fw->cold_boot_entry_point; - fw->trace_level = min_t(u32, ivpu_log_level, IVPU_FW_LOG_FATAL); + fw->trace_level = min_t(u32, ivpu_fw_log_level, IVPU_FW_LOG_FATAL); fw->trace_destination_mask = VPU_TRACE_DESTINATION_VERBOSE_TRACING; fw->trace_hw_component_mask = -1; fw->dvfs_mode = 0; + fw->sched_mode = ivpu_fw_sched_mode_select(vdev, fw_hdr); fw->primary_preempt_buf_size = fw_hdr->preemption_buffer_1_size; fw->secondary_preempt_buf_size = fw_hdr->preemption_buffer_2_size; + ivpu_info(vdev, "Scheduler mode: %s\n", fw->sched_mode ? "HW" : "OS"); if (fw_hdr->ro_section_start_address && !is_within_range(fw_hdr->ro_section_start_address, fw_hdr->ro_section_size, @@ -311,7 +328,7 @@ static int ivpu_fw_mem_init(struct ivpu_device *vdev) goto err_free_fw_mem; } - if (ivpu_log_level <= IVPU_FW_LOG_INFO) + if (ivpu_fw_log_level <= IVPU_FW_LOG_INFO) log_verb_size = IVPU_FW_VERBOSE_BUFFER_LARGE_SIZE; else log_verb_size = IVPU_FW_VERBOSE_BUFFER_SMALL_SIZE; @@ -567,8 +584,10 @@ void ivpu_fw_boot_params_setup(struct ivpu_device *vdev, struct vpu_boot_params boot_params->ipc_payload_area_start = ipc_mem_rx->vpu_addr + ivpu_bo_size(ipc_mem_rx) / 2; boot_params->ipc_payload_area_size = ivpu_bo_size(ipc_mem_rx) / 2; - boot_params->global_aliased_pio_base = vdev->hw->ranges.user.start; - boot_params->global_aliased_pio_size = ivpu_hw_range_size(&vdev->hw->ranges.user); + if (ivpu_hw_ip_gen(vdev) == IVPU_HW_IP_37XX) { + boot_params->global_aliased_pio_base = vdev->hw->ranges.user.start; + boot_params->global_aliased_pio_size = ivpu_hw_range_size(&vdev->hw->ranges.user); + } /* Allow configuration for L2C_PAGE_TABLE with boot param value */ boot_params->autoconfig = 1; @@ -604,8 +623,8 @@ void ivpu_fw_boot_params_setup(struct ivpu_device *vdev, struct vpu_boot_params boot_params->punit_telemetry_sram_base = ivpu_hw_telemetry_offset_get(vdev); boot_params->punit_telemetry_sram_size = ivpu_hw_telemetry_size_get(vdev); boot_params->vpu_telemetry_enable = ivpu_hw_telemetry_enable_get(vdev); - boot_params->vpu_scheduling_mode = vdev->hw->sched_mode; - if (vdev->hw->sched_mode == VPU_SCHEDULING_MODE_HW) + boot_params->vpu_scheduling_mode = vdev->fw->sched_mode; + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) boot_params->vpu_focus_present_timer_ms = IVPU_FOCUS_PRESENT_TIMER_MS; boot_params->dvfs_mode = vdev->fw->dvfs_mode; if (!IVPU_WA(disable_d0i3_msg)) diff --git a/drivers/accel/ivpu/ivpu_fw.h b/drivers/accel/ivpu/ivpu_fw.h index 40d9d17be3f5..1d0b2bd9d65c 100644 --- a/drivers/accel/ivpu/ivpu_fw.h +++ b/drivers/accel/ivpu/ivpu_fw.h @@ -1,11 +1,16 @@ /* SPDX-License-Identifier: GPL-2.0-only */ /* - * Copyright (C) 2020-2023 Intel Corporation + * Copyright (C) 2020-2024 Intel Corporation */ #ifndef __IVPU_FW_H__ #define __IVPU_FW_H__ +#include "vpu_jsm_api.h" + +#define FW_VERSION_HEADER_SIZE SZ_4K +#define FW_VERSION_STR_SIZE SZ_256 + struct ivpu_device; struct ivpu_bo; struct vpu_boot_params; @@ -13,6 +18,7 @@ struct vpu_boot_params; struct ivpu_fw_info { const struct firmware *file; const char *name; + char version[FW_VERSION_STR_SIZE]; struct ivpu_bo *mem; struct ivpu_bo *mem_shave_nn; struct ivpu_bo *mem_log_crit; @@ -32,6 +38,7 @@ struct ivpu_fw_info { u32 secondary_preempt_buf_size; u64 read_only_addr; u32 read_only_size; + u32 sched_mode; }; int ivpu_fw_init(struct ivpu_device *vdev); diff --git a/drivers/accel/ivpu/ivpu_fw_log.c b/drivers/accel/ivpu/ivpu_fw_log.c index ef0adb5e0fbe..337c906b0210 100644 --- a/drivers/accel/ivpu/ivpu_fw_log.c +++ b/drivers/accel/ivpu/ivpu_fw_log.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-only /* - * Copyright (C) 2020-2023 Intel Corporation + * Copyright (C) 2020-2024 Intel Corporation */ #include @@ -15,19 +15,19 @@ #include "ivpu_fw_log.h" #include "ivpu_gem.h" -#define IVPU_FW_LOG_LINE_LENGTH 256 +#define IVPU_FW_LOG_LINE_LENGTH 256 -unsigned int ivpu_log_level = IVPU_FW_LOG_ERROR; -module_param(ivpu_log_level, uint, 0444); -MODULE_PARM_DESC(ivpu_log_level, - "NPU firmware default trace level: debug=" __stringify(IVPU_FW_LOG_DEBUG) +unsigned int ivpu_fw_log_level = IVPU_FW_LOG_ERROR; +module_param_named(fw_log_level, ivpu_fw_log_level, uint, 0444); +MODULE_PARM_DESC(fw_log_level, + "NPU firmware default log level: debug=" __stringify(IVPU_FW_LOG_DEBUG) " info=" __stringify(IVPU_FW_LOG_INFO) " warn=" __stringify(IVPU_FW_LOG_WARN) " error=" __stringify(IVPU_FW_LOG_ERROR) " fatal=" __stringify(IVPU_FW_LOG_FATAL)); -static int fw_log_ptr(struct ivpu_device *vdev, struct ivpu_bo *bo, u32 *offset, - struct vpu_tracing_buffer_header **log_header) +static int fw_log_from_bo(struct ivpu_device *vdev, struct ivpu_bo *bo, u32 *offset, + struct vpu_tracing_buffer_header **out_log) { struct vpu_tracing_buffer_header *log; @@ -48,7 +48,7 @@ static int fw_log_ptr(struct ivpu_device *vdev, struct ivpu_bo *bo, u32 *offset, return -EINVAL; } - *log_header = log; + *out_log = log; *offset += log->size; ivpu_dbg(vdev, FW_BOOT, @@ -59,7 +59,7 @@ static int fw_log_ptr(struct ivpu_device *vdev, struct ivpu_bo *bo, u32 *offset, return 0; } -static void buffer_print(char *buffer, u32 size, struct drm_printer *p) +static void fw_log_print_lines(char *buffer, u32 size, struct drm_printer *p) { char line[IVPU_FW_LOG_LINE_LENGTH]; u32 index = 0; @@ -87,56 +87,89 @@ static void buffer_print(char *buffer, u32 size, struct drm_printer *p) } line[index] = 0; if (index != 0) - drm_printf(p, "%s\n", line); + drm_printf(p, "%s", line); } -static void fw_log_print_buffer(struct ivpu_device *vdev, struct vpu_tracing_buffer_header *log, - const char *prefix, bool only_new_msgs, struct drm_printer *p) +static void fw_log_print_buffer(struct vpu_tracing_buffer_header *log, const char *prefix, + bool only_new_msgs, struct drm_printer *p) { - char *log_buffer = (void *)log + log->header_size; - u32 log_size = log->size - log->header_size; - u32 log_start = log->read_index; - u32 log_end = log->write_index; + char *log_data = (void *)log + log->header_size; + u32 data_size = log->size - log->header_size; + u32 log_start = only_new_msgs ? READ_ONCE(log->read_index) : 0; + u32 log_end = READ_ONCE(log->write_index); - if (!(log->write_index || log->wrap_count) || - (log->write_index == log->read_index && only_new_msgs)) { - drm_printf(p, "==== %s \"%s\" log empty ====\n", prefix, log->name); - return; + if (log->wrap_count == log->read_wrap_count) { + if (log_end <= log_start) { + drm_printf(p, "==== %s \"%s\" log empty ====\n", prefix, log->name); + return; + } + } else if (log->wrap_count == log->read_wrap_count + 1) { + if (log_end > log_start) + log_start = log_end; + } else { + log_start = log_end; } drm_printf(p, "==== %s \"%s\" log start ====\n", prefix, log->name); - if (log->write_index > log->read_index) { - buffer_print(log_buffer + log_start, log_end - log_start, p); + if (log_end > log_start) { + fw_log_print_lines(log_data + log_start, log_end - log_start, p); } else { - buffer_print(log_buffer + log_end, log_size - log_end, p); - buffer_print(log_buffer, log_end, p); + fw_log_print_lines(log_data + log_start, data_size - log_start, p); + fw_log_print_lines(log_data, log_end, p); } - drm_printf(p, "\x1b[0m"); + drm_printf(p, "\n\x1b[0m"); /* add new line and clear formatting */ drm_printf(p, "==== %s \"%s\" log end ====\n", prefix, log->name); } +static void +fw_log_print_all_in_bo(struct ivpu_device *vdev, const char *name, + struct ivpu_bo *bo, bool only_new_msgs, struct drm_printer *p) +{ + struct vpu_tracing_buffer_header *log; + u32 next = 0; + + while (fw_log_from_bo(vdev, bo, &next, &log) == 0) + fw_log_print_buffer(log, name, only_new_msgs, p); +} + void ivpu_fw_log_print(struct ivpu_device *vdev, bool only_new_msgs, struct drm_printer *p) { - struct vpu_tracing_buffer_header *log_header; - u32 next = 0; - - while (fw_log_ptr(vdev, vdev->fw->mem_log_crit, &next, &log_header) == 0) - fw_log_print_buffer(vdev, log_header, "NPU critical", only_new_msgs, p); - - next = 0; - while (fw_log_ptr(vdev, vdev->fw->mem_log_verb, &next, &log_header) == 0) - fw_log_print_buffer(vdev, log_header, "NPU verbose", only_new_msgs, p); + fw_log_print_all_in_bo(vdev, "NPU critical", vdev->fw->mem_log_crit, only_new_msgs, p); + fw_log_print_all_in_bo(vdev, "NPU verbose", vdev->fw->mem_log_verb, only_new_msgs, p); } -void ivpu_fw_log_clear(struct ivpu_device *vdev) +void ivpu_fw_log_mark_read(struct ivpu_device *vdev) { - struct vpu_tracing_buffer_header *log_header; - u32 next = 0; - - while (fw_log_ptr(vdev, vdev->fw->mem_log_crit, &next, &log_header) == 0) - log_header->read_index = log_header->write_index; + struct vpu_tracing_buffer_header *log; + u32 next; next = 0; - while (fw_log_ptr(vdev, vdev->fw->mem_log_verb, &next, &log_header) == 0) - log_header->read_index = log_header->write_index; + while (fw_log_from_bo(vdev, vdev->fw->mem_log_crit, &next, &log) == 0) { + log->read_index = READ_ONCE(log->write_index); + log->read_wrap_count = READ_ONCE(log->wrap_count); + } + + next = 0; + while (fw_log_from_bo(vdev, vdev->fw->mem_log_verb, &next, &log) == 0) { + log->read_index = READ_ONCE(log->write_index); + log->read_wrap_count = READ_ONCE(log->wrap_count); + } +} + +void ivpu_fw_log_reset(struct ivpu_device *vdev) +{ + struct vpu_tracing_buffer_header *log; + u32 next; + + next = 0; + while (fw_log_from_bo(vdev, vdev->fw->mem_log_crit, &next, &log) == 0) { + log->read_index = 0; + log->read_wrap_count = 0; + } + + next = 0; + while (fw_log_from_bo(vdev, vdev->fw->mem_log_verb, &next, &log) == 0) { + log->read_index = 0; + log->read_wrap_count = 0; + } } diff --git a/drivers/accel/ivpu/ivpu_fw_log.h b/drivers/accel/ivpu/ivpu_fw_log.h index 0b2573f6f315..8bb528a73cb7 100644 --- a/drivers/accel/ivpu/ivpu_fw_log.h +++ b/drivers/accel/ivpu/ivpu_fw_log.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0-only */ /* - * Copyright (C) 2020-2023 Intel Corporation + * Copyright (C) 2020-2024 Intel Corporation */ #ifndef __IVPU_FW_LOG_H__ @@ -8,8 +8,6 @@ #include -#include - #include "ivpu_drv.h" #define IVPU_FW_LOG_DEFAULT 0 @@ -19,20 +17,15 @@ #define IVPU_FW_LOG_ERROR 4 #define IVPU_FW_LOG_FATAL 5 -extern unsigned int ivpu_log_level; - #define IVPU_FW_VERBOSE_BUFFER_SMALL_SIZE SZ_1M #define IVPU_FW_VERBOSE_BUFFER_LARGE_SIZE SZ_8M #define IVPU_FW_CRITICAL_BUFFER_SIZE SZ_512K +extern unsigned int ivpu_fw_log_level; + void ivpu_fw_log_print(struct ivpu_device *vdev, bool only_new_msgs, struct drm_printer *p); -void ivpu_fw_log_clear(struct ivpu_device *vdev); +void ivpu_fw_log_mark_read(struct ivpu_device *vdev); +void ivpu_fw_log_reset(struct ivpu_device *vdev); -static inline void ivpu_fw_log_dump(struct ivpu_device *vdev) -{ - struct drm_printer p = drm_info_printer(vdev->drm.dev); - - ivpu_fw_log_print(vdev, false, &p); -} #endif /* __IVPU_FW_LOG_H__ */ diff --git a/drivers/accel/ivpu/ivpu_gem.c b/drivers/accel/ivpu/ivpu_gem.c index 1b409dbd332d..d8e97a760fbc 100644 --- a/drivers/accel/ivpu/ivpu_gem.c +++ b/drivers/accel/ivpu/ivpu_gem.c @@ -384,6 +384,9 @@ int ivpu_bo_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file timeout = drm_timeout_abs_to_jiffies(args->timeout_ns); + /* Add 1 jiffy to ensure the wait function never times out before intended timeout_ns */ + timeout += 1; + obj = drm_gem_object_lookup(file, args->handle); if (!obj) return -EINVAL; diff --git a/drivers/accel/ivpu/ivpu_hw.c b/drivers/accel/ivpu/ivpu_hw.c index 27f0fe4d54e0..4e1054f3466e 100644 --- a/drivers/accel/ivpu/ivpu_hw.c +++ b/drivers/accel/ivpu/ivpu_hw.c @@ -89,12 +89,14 @@ static void timeouts_init(struct ivpu_device *vdev) vdev->timeout.tdr = 2000000; vdev->timeout.autosuspend = -1; vdev->timeout.d0i3_entry_msg = 500; + vdev->timeout.state_dump_msg = 10; } else if (ivpu_is_simics(vdev)) { vdev->timeout.boot = 50; vdev->timeout.jsm = 500; vdev->timeout.tdr = 10000; - vdev->timeout.autosuspend = -1; + vdev->timeout.autosuspend = 100; vdev->timeout.d0i3_entry_msg = 100; + vdev->timeout.state_dump_msg = 10; } else { vdev->timeout.boot = 1000; vdev->timeout.jsm = 500; @@ -104,6 +106,7 @@ static void timeouts_init(struct ivpu_device *vdev) else vdev->timeout.autosuspend = 100; vdev->timeout.d0i3_entry_msg = 5; + vdev->timeout.state_dump_msg = 10; } } @@ -111,14 +114,14 @@ static void memory_ranges_init(struct ivpu_device *vdev) { if (ivpu_hw_ip_gen(vdev) == IVPU_HW_IP_37XX) { ivpu_hw_range_init(&vdev->hw->ranges.global, 0x80000000, SZ_512M); - ivpu_hw_range_init(&vdev->hw->ranges.user, 0xc0000000, 255 * SZ_1M); + ivpu_hw_range_init(&vdev->hw->ranges.user, 0x88000000, 511 * SZ_1M); ivpu_hw_range_init(&vdev->hw->ranges.shave, 0x180000000, SZ_2G); - ivpu_hw_range_init(&vdev->hw->ranges.dma, 0x200000000, SZ_8G); + ivpu_hw_range_init(&vdev->hw->ranges.dma, 0x200000000, SZ_128G); } else { ivpu_hw_range_init(&vdev->hw->ranges.global, 0x80000000, SZ_512M); - ivpu_hw_range_init(&vdev->hw->ranges.user, 0x80000000, SZ_256M); - ivpu_hw_range_init(&vdev->hw->ranges.shave, 0x80000000 + SZ_256M, SZ_2G - SZ_256M); - ivpu_hw_range_init(&vdev->hw->ranges.dma, 0x200000000, SZ_8G); + ivpu_hw_range_init(&vdev->hw->ranges.shave, 0x80000000, SZ_2G); + ivpu_hw_range_init(&vdev->hw->ranges.user, 0x100000000, SZ_256G); + vdev->hw->ranges.dma = vdev->hw->ranges.user; } } @@ -249,6 +252,7 @@ int ivpu_hw_init(struct ivpu_device *vdev) platform_init(vdev); wa_init(vdev); timeouts_init(vdev); + atomic_set(&vdev->hw->firewall_irq_counter, 0); return 0; } diff --git a/drivers/accel/ivpu/ivpu_hw.h b/drivers/accel/ivpu/ivpu_hw.h index 1c0c98e3afb8..fc4dbfc980c8 100644 --- a/drivers/accel/ivpu/ivpu_hw.h +++ b/drivers/accel/ivpu/ivpu_hw.h @@ -46,12 +46,12 @@ struct ivpu_hw_info { u32 profiling_freq; } pll; u32 tile_fuse; - u32 sched_mode; u32 sku; u16 config; int dma_bits; ktime_t d0i3_entry_host_ts; u64 d0i3_entry_vpu_ts; + atomic_t firewall_irq_counter; }; int ivpu_hw_init(struct ivpu_device *vdev); diff --git a/drivers/accel/ivpu/ivpu_hw_40xx_reg.h b/drivers/accel/ivpu/ivpu_hw_40xx_reg.h index d0b795b344c7..fc0ee8d637f9 100644 --- a/drivers/accel/ivpu/ivpu_hw_40xx_reg.h +++ b/drivers/accel/ivpu/ivpu_hw_40xx_reg.h @@ -115,6 +115,8 @@ #define VPU_50XX_HOST_SS_AON_PWR_ISLAND_EN_POST_DLY 0x00030068u #define VPU_50XX_HOST_SS_AON_PWR_ISLAND_EN_POST_DLY_POST_DLY_MASK GENMASK(7, 0) +#define VPU_50XX_HOST_SS_AON_PWR_ISLAND_EN_POST_DLY_POST1_DLY_MASK GENMASK(15, 8) +#define VPU_50XX_HOST_SS_AON_PWR_ISLAND_EN_POST_DLY_POST2_DLY_MASK GENMASK(23, 16) #define VPU_50XX_HOST_SS_AON_PWR_ISLAND_STATUS_DLY 0x0003006cu #define VPU_50XX_HOST_SS_AON_PWR_ISLAND_STATUS_DLY_STATUS_DLY_MASK GENMASK(7, 0) diff --git a/drivers/accel/ivpu/ivpu_hw_btrs.c b/drivers/accel/ivpu/ivpu_hw_btrs.c index 745e5248803d..3212c99f3682 100644 --- a/drivers/accel/ivpu/ivpu_hw_btrs.c +++ b/drivers/accel/ivpu/ivpu_hw_btrs.c @@ -141,16 +141,10 @@ static int read_tile_config_fuse(struct ivpu_device *vdev, u32 *tile_fuse_config } config = REG_GET_FLD(VPU_HW_BTRS_LNL_TILE_FUSE, CONFIG, fuse); - if (!tile_disable_check(config)) { - ivpu_err(vdev, "Fuse: Invalid tile disable config (0x%x)\n", config); - return -EIO; - } + if (!tile_disable_check(config)) + ivpu_warn(vdev, "More than 1 tile disabled, tile fuse config mask: 0x%x\n", config); - if (config) - ivpu_dbg(vdev, MISC, "Fuse: %d tiles enabled. Tile number %d disabled\n", - BTRS_LNL_TILE_MAX_NUM - 1, ffs(config) - 1); - else - ivpu_dbg(vdev, MISC, "Fuse: All %d tiles enabled\n", BTRS_LNL_TILE_MAX_NUM); + ivpu_dbg(vdev, MISC, "Tile disable config mask: 0x%x\n", config); *tile_fuse_config = config; return 0; @@ -163,7 +157,6 @@ static int info_init_mtl(struct ivpu_device *vdev) hw->tile_fuse = BTRS_MTL_TILE_FUSE_ENABLE_BOTH; hw->sku = BTRS_MTL_TILE_SKU_BOTH; hw->config = BTRS_MTL_WP_CONFIG_2_TILE_4_3_RATIO; - hw->sched_mode = ivpu_sched_mode; return 0; } @@ -178,7 +171,6 @@ static int info_init_lnl(struct ivpu_device *vdev) if (ret) return ret; - hw->sched_mode = ivpu_sched_mode; hw->tile_fuse = tile_fuse_config; hw->pll.profiling_freq = PLL_PROFILING_FREQ_DEFAULT; @@ -315,10 +307,6 @@ static void prepare_wp_request(struct ivpu_device *vdev, struct wp_request *wp, wp->cdyn = enable ? PLL_CDYN_DEFAULT : 0; wp->epp = enable ? PLL_EPP_DEFAULT : 0; } - - /* Simics cannot start without at least one tile */ - if (enable && ivpu_is_simics(vdev)) - wp->cfg = 1; } static int wait_for_pll_lock(struct ivpu_device *vdev, bool enable) @@ -465,9 +453,6 @@ int ivpu_hw_btrs_wait_for_clock_res_own_ack(struct ivpu_device *vdev) if (ivpu_hw_btrs_gen(vdev) == IVPU_HW_BTRS_MTL) return 0; - if (ivpu_is_simics(vdev)) - return 0; - return REGB_POLL_FLD(VPU_HW_BTRS_LNL_VPU_STATUS, CLOCK_RESOURCE_OWN_ACK, 1, TIMEOUT_US); } diff --git a/drivers/accel/ivpu/ivpu_hw_ip.c b/drivers/accel/ivpu/ivpu_hw_ip.c index dfd2f4a5b526..029dd065614b 100644 --- a/drivers/accel/ivpu/ivpu_hw_ip.c +++ b/drivers/accel/ivpu/ivpu_hw_ip.c @@ -8,15 +8,12 @@ #include "ivpu_hw.h" #include "ivpu_hw_37xx_reg.h" #include "ivpu_hw_40xx_reg.h" +#include "ivpu_hw_btrs.h" #include "ivpu_hw_ip.h" #include "ivpu_hw_reg_io.h" #include "ivpu_mmu.h" #include "ivpu_pm.h" -#define PWR_ISLAND_EN_POST_DLY_FREQ_DEFAULT 0 -#define PWR_ISLAND_EN_POST_DLY_FREQ_HIGH 18 -#define PWR_ISLAND_STATUS_DLY_FREQ_DEFAULT 3 -#define PWR_ISLAND_STATUS_DLY_FREQ_HIGH 46 #define PWR_ISLAND_STATUS_TIMEOUT_US (5 * USEC_PER_MSEC) #define TIM_SAFE_ENABLE 0xf1d0dead @@ -268,20 +265,15 @@ void ivpu_hw_ip_idle_gen_disable(struct ivpu_device *vdev) idle_gen_drive_40xx(vdev, false); } -static void pwr_island_delay_set_50xx(struct ivpu_device *vdev) +static void +pwr_island_delay_set_50xx(struct ivpu_device *vdev, u32 post, u32 post1, u32 post2, u32 status) { - u32 val, post, status; - - if (vdev->hw->pll.profiling_freq == PLL_PROFILING_FREQ_DEFAULT) { - post = PWR_ISLAND_EN_POST_DLY_FREQ_DEFAULT; - status = PWR_ISLAND_STATUS_DLY_FREQ_DEFAULT; - } else { - post = PWR_ISLAND_EN_POST_DLY_FREQ_HIGH; - status = PWR_ISLAND_STATUS_DLY_FREQ_HIGH; - } + u32 val; val = REGV_RD32(VPU_50XX_HOST_SS_AON_PWR_ISLAND_EN_POST_DLY); val = REG_SET_FLD_NUM(VPU_50XX_HOST_SS_AON_PWR_ISLAND_EN_POST_DLY, POST_DLY, post, val); + val = REG_SET_FLD_NUM(VPU_50XX_HOST_SS_AON_PWR_ISLAND_EN_POST_DLY, POST1_DLY, post1, val); + val = REG_SET_FLD_NUM(VPU_50XX_HOST_SS_AON_PWR_ISLAND_EN_POST_DLY, POST2_DLY, post2, val); REGV_WR32(VPU_50XX_HOST_SS_AON_PWR_ISLAND_EN_POST_DLY, val); val = REGV_RD32(VPU_50XX_HOST_SS_AON_PWR_ISLAND_STATUS_DLY); @@ -311,9 +303,6 @@ static void pwr_island_trickle_drive_40xx(struct ivpu_device *vdev, bool enable) val = REG_CLR_FLD(VPU_40XX_HOST_SS_AON_PWR_ISLAND_TRICKLE_EN0, CSS_CPU, val); REGV_WR32(VPU_40XX_HOST_SS_AON_PWR_ISLAND_TRICKLE_EN0, val); - - if (enable) - ndelay(500); } static void pwr_island_drive_37xx(struct ivpu_device *vdev, bool enable) @@ -326,9 +315,6 @@ static void pwr_island_drive_37xx(struct ivpu_device *vdev, bool enable) val = REG_CLR_FLD(VPU_40XX_HOST_SS_AON_PWR_ISLAND_EN0, CSS_CPU, val); REGV_WR32(VPU_40XX_HOST_SS_AON_PWR_ISLAND_EN0, val); - - if (!enable) - ndelay(500); } static void pwr_island_drive_40xx(struct ivpu_device *vdev, bool enable) @@ -347,9 +333,11 @@ static void pwr_island_enable(struct ivpu_device *vdev) { if (ivpu_hw_ip_gen(vdev) == IVPU_HW_IP_37XX) { pwr_island_trickle_drive_37xx(vdev, true); + ndelay(500); pwr_island_drive_37xx(vdev, true); } else { pwr_island_trickle_drive_40xx(vdev, true); + ndelay(500); pwr_island_drive_40xx(vdev, true); } } @@ -686,13 +674,36 @@ static void dpu_active_drive_37xx(struct ivpu_device *vdev, bool enable) REGV_WR32(VPU_37XX_HOST_SS_AON_DPU_ACTIVE, val); } +static void pwr_island_delay_set(struct ivpu_device *vdev) +{ + bool high = vdev->hw->pll.profiling_freq == PLL_PROFILING_FREQ_HIGH; + u32 post, post1, post2, status; + + if (ivpu_hw_ip_gen(vdev) < IVPU_HW_IP_50XX) + return; + + switch (ivpu_device_id(vdev)) { + case PCI_DEVICE_ID_PTL_P: + post = high ? 18 : 0; + post1 = 0; + post2 = 0; + status = high ? 46 : 3; + break; + + default: + dump_stack(); + ivpu_err(vdev, "Unknown device ID\n"); + return; + } + + pwr_island_delay_set_50xx(vdev, post, post1, post2, status); +} + int ivpu_hw_ip_pwr_domain_enable(struct ivpu_device *vdev) { int ret; - if (ivpu_hw_ip_gen(vdev) == IVPU_HW_IP_50XX) - pwr_island_delay_set_50xx(vdev); - + pwr_island_delay_set(vdev); pwr_island_enable(vdev); ret = wait_for_pwr_island_status(vdev, 0x1); @@ -1062,7 +1073,10 @@ static void irq_wdt_mss_handler(struct ivpu_device *vdev) static void irq_noc_firewall_handler(struct ivpu_device *vdev) { - ivpu_pm_trigger_recovery(vdev, "NOC Firewall IRQ"); + atomic_inc(&vdev->hw->firewall_irq_counter); + + ivpu_dbg(vdev, IRQ, "NOC Firewall interrupt detected, counter %d\n", + atomic_read(&vdev->hw->firewall_irq_counter)); } /* Handler for IRQs from NPU core */ diff --git a/drivers/accel/ivpu/ivpu_ipc.c b/drivers/accel/ivpu/ivpu_ipc.c index 78b32a823241..01ebf88fe6ef 100644 --- a/drivers/accel/ivpu/ivpu_ipc.c +++ b/drivers/accel/ivpu/ivpu_ipc.c @@ -15,6 +15,7 @@ #include "ivpu_ipc.h" #include "ivpu_jsm_msg.h" #include "ivpu_pm.h" +#include "ivpu_trace.h" #define IPC_MAX_RX_MSG 128 @@ -227,6 +228,7 @@ int ivpu_ipc_send(struct ivpu_device *vdev, struct ivpu_ipc_consumer *cons, stru goto unlock; ivpu_ipc_tx(vdev, cons->tx_vpu_addr); + trace_jsm("[tx]", req); unlock: mutex_unlock(&ipc->lock); @@ -278,12 +280,13 @@ int ivpu_ipc_receive(struct ivpu_device *vdev, struct ivpu_ipc_consumer *cons, u32 size = min_t(int, rx_msg->ipc_hdr->data_size, sizeof(*jsm_msg)); if (rx_msg->jsm_msg->result != VPU_JSM_STATUS_SUCCESS) { - ivpu_dbg(vdev, IPC, "IPC resp result error: %d\n", rx_msg->jsm_msg->result); + ivpu_err(vdev, "IPC resp result error: %d\n", rx_msg->jsm_msg->result); ret = -EBADMSG; } if (jsm_msg) memcpy(jsm_msg, rx_msg->jsm_msg, size); + trace_jsm("[rx]", rx_msg->jsm_msg); } ivpu_ipc_rx_msg_del(vdev, rx_msg); @@ -291,15 +294,16 @@ int ivpu_ipc_receive(struct ivpu_device *vdev, struct ivpu_ipc_consumer *cons, return ret; } -static int +int ivpu_ipc_send_receive_internal(struct ivpu_device *vdev, struct vpu_jsm_msg *req, enum vpu_ipc_msg_type expected_resp_type, - struct vpu_jsm_msg *resp, u32 channel, - unsigned long timeout_ms) + struct vpu_jsm_msg *resp, u32 channel, unsigned long timeout_ms) { struct ivpu_ipc_consumer cons; int ret; + drm_WARN_ON(&vdev->drm, pm_runtime_status_suspended(vdev->drm.dev)); + ivpu_ipc_consumer_add(vdev, &cons, channel, NULL); ret = ivpu_ipc_send(vdev, &cons, req); @@ -325,19 +329,21 @@ consumer_del: return ret; } -int ivpu_ipc_send_receive_active(struct ivpu_device *vdev, struct vpu_jsm_msg *req, - enum vpu_ipc_msg_type expected_resp, struct vpu_jsm_msg *resp, - u32 channel, unsigned long timeout_ms) +int ivpu_ipc_send_receive(struct ivpu_device *vdev, struct vpu_jsm_msg *req, + enum vpu_ipc_msg_type expected_resp, struct vpu_jsm_msg *resp, + u32 channel, unsigned long timeout_ms) { struct vpu_jsm_msg hb_req = { .type = VPU_JSM_MSG_QUERY_ENGINE_HB }; struct vpu_jsm_msg hb_resp; int ret, hb_ret; - drm_WARN_ON(&vdev->drm, pm_runtime_status_suspended(vdev->drm.dev)); + ret = ivpu_rpm_get(vdev); + if (ret < 0) + return ret; ret = ivpu_ipc_send_receive_internal(vdev, req, expected_resp, resp, channel, timeout_ms); if (ret != -ETIMEDOUT) - return ret; + goto rpm_put; hb_ret = ivpu_ipc_send_receive_internal(vdev, &hb_req, VPU_JSM_MSG_QUERY_ENGINE_HB_DONE, &hb_resp, VPU_IPC_CHAN_ASYNC_CMD, @@ -345,21 +351,33 @@ int ivpu_ipc_send_receive_active(struct ivpu_device *vdev, struct vpu_jsm_msg *r if (hb_ret == -ETIMEDOUT) ivpu_pm_trigger_recovery(vdev, "IPC timeout"); +rpm_put: + ivpu_rpm_put(vdev); return ret; } -int ivpu_ipc_send_receive(struct ivpu_device *vdev, struct vpu_jsm_msg *req, - enum vpu_ipc_msg_type expected_resp, struct vpu_jsm_msg *resp, - u32 channel, unsigned long timeout_ms) +int ivpu_ipc_send_and_wait(struct ivpu_device *vdev, struct vpu_jsm_msg *req, + u32 channel, unsigned long timeout_ms) { + struct ivpu_ipc_consumer cons; int ret; ret = ivpu_rpm_get(vdev); if (ret < 0) return ret; - ret = ivpu_ipc_send_receive_active(vdev, req, expected_resp, resp, channel, timeout_ms); + ivpu_ipc_consumer_add(vdev, &cons, channel, NULL); + ret = ivpu_ipc_send(vdev, &cons, req); + if (ret) { + ivpu_warn_ratelimited(vdev, "IPC send failed: %d\n", ret); + goto consumer_del; + } + + msleep(timeout_ms); + +consumer_del: + ivpu_ipc_consumer_del(vdev, &cons); ivpu_rpm_put(vdev); return ret; } @@ -518,7 +536,6 @@ void ivpu_ipc_fini(struct ivpu_device *vdev) { struct ivpu_ipc_info *ipc = vdev->ipc; - drm_WARN_ON(&vdev->drm, ipc->on); drm_WARN_ON(&vdev->drm, !list_empty(&ipc->cons_list)); drm_WARN_ON(&vdev->drm, !list_empty(&ipc->cb_msg_list)); drm_WARN_ON(&vdev->drm, atomic_read(&ipc->rx_msg_count) > 0); diff --git a/drivers/accel/ivpu/ivpu_ipc.h b/drivers/accel/ivpu/ivpu_ipc.h index 4fe38141045e..b4dfb504679b 100644 --- a/drivers/accel/ivpu/ivpu_ipc.h +++ b/drivers/accel/ivpu/ivpu_ipc.h @@ -101,12 +101,13 @@ int ivpu_ipc_send(struct ivpu_device *vdev, struct ivpu_ipc_consumer *cons, int ivpu_ipc_receive(struct ivpu_device *vdev, struct ivpu_ipc_consumer *cons, struct ivpu_ipc_hdr *ipc_buf, struct vpu_jsm_msg *jsm_msg, unsigned long timeout_ms); - -int ivpu_ipc_send_receive_active(struct ivpu_device *vdev, struct vpu_jsm_msg *req, - enum vpu_ipc_msg_type expected_resp, struct vpu_jsm_msg *resp, - u32 channel, unsigned long timeout_ms); +int ivpu_ipc_send_receive_internal(struct ivpu_device *vdev, struct vpu_jsm_msg *req, + enum vpu_ipc_msg_type expected_resp_type, + struct vpu_jsm_msg *resp, u32 channel, unsigned long timeout_ms); int ivpu_ipc_send_receive(struct ivpu_device *vdev, struct vpu_jsm_msg *req, enum vpu_ipc_msg_type expected_resp, struct vpu_jsm_msg *resp, u32 channel, unsigned long timeout_ms); +int ivpu_ipc_send_and_wait(struct ivpu_device *vdev, struct vpu_jsm_msg *req, + u32 channel, unsigned long timeout_ms); #endif /* __IVPU_IPC_H__ */ diff --git a/drivers/accel/ivpu/ivpu_job.c b/drivers/accel/ivpu/ivpu_job.c index be2e2bf0f43f..7149312f16e1 100644 --- a/drivers/accel/ivpu/ivpu_job.c +++ b/drivers/accel/ivpu/ivpu_job.c @@ -18,11 +18,10 @@ #include "ivpu_job.h" #include "ivpu_jsm_msg.h" #include "ivpu_pm.h" +#include "ivpu_trace.h" #include "vpu_boot_api.h" #define CMD_BUF_IDX 0 -#define JOB_ID_JOB_MASK GENMASK(7, 0) -#define JOB_ID_CONTEXT_MASK GENMASK(31, 8) #define JOB_MAX_BUFFER_COUNT 65535 static void ivpu_cmdq_ring_db(struct ivpu_device *vdev, struct ivpu_cmdq *cmdq) @@ -35,24 +34,20 @@ static int ivpu_preemption_buffers_create(struct ivpu_device *vdev, { u64 primary_size = ALIGN(vdev->fw->primary_preempt_buf_size, PAGE_SIZE); u64 secondary_size = ALIGN(vdev->fw->secondary_preempt_buf_size, PAGE_SIZE); - struct ivpu_addr_range range; - if (vdev->hw->sched_mode != VPU_SCHEDULING_MODE_HW) + if (vdev->fw->sched_mode != VPU_SCHEDULING_MODE_HW || + ivpu_test_mode & IVPU_TEST_MODE_MIP_DISABLE) return 0; - range.start = vdev->hw->ranges.user.end - (primary_size * IVPU_NUM_CMDQS_PER_CTX); - range.end = vdev->hw->ranges.user.end; - cmdq->primary_preempt_buf = ivpu_bo_create(vdev, &file_priv->ctx, &range, primary_size, - DRM_IVPU_BO_WC); + cmdq->primary_preempt_buf = ivpu_bo_create(vdev, &file_priv->ctx, &vdev->hw->ranges.user, + primary_size, DRM_IVPU_BO_WC); if (!cmdq->primary_preempt_buf) { ivpu_err(vdev, "Failed to create primary preemption buffer\n"); return -ENOMEM; } - range.start = vdev->hw->ranges.shave.end - (secondary_size * IVPU_NUM_CMDQS_PER_CTX); - range.end = vdev->hw->ranges.shave.end; - cmdq->secondary_preempt_buf = ivpu_bo_create(vdev, &file_priv->ctx, &range, secondary_size, - DRM_IVPU_BO_WC); + cmdq->secondary_preempt_buf = ivpu_bo_create(vdev, &file_priv->ctx, &vdev->hw->ranges.dma, + secondary_size, DRM_IVPU_BO_WC); if (!cmdq->secondary_preempt_buf) { ivpu_err(vdev, "Failed to create secondary preemption buffer\n"); goto err_free_primary; @@ -62,24 +57,24 @@ static int ivpu_preemption_buffers_create(struct ivpu_device *vdev, err_free_primary: ivpu_bo_free(cmdq->primary_preempt_buf); + cmdq->primary_preempt_buf = NULL; return -ENOMEM; } static void ivpu_preemption_buffers_free(struct ivpu_device *vdev, struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq) { - if (vdev->hw->sched_mode != VPU_SCHEDULING_MODE_HW) + if (vdev->fw->sched_mode != VPU_SCHEDULING_MODE_HW) return; - drm_WARN_ON(&vdev->drm, !cmdq->primary_preempt_buf); - drm_WARN_ON(&vdev->drm, !cmdq->secondary_preempt_buf); - ivpu_bo_free(cmdq->primary_preempt_buf); - ivpu_bo_free(cmdq->secondary_preempt_buf); + if (cmdq->primary_preempt_buf) + ivpu_bo_free(cmdq->primary_preempt_buf); + if (cmdq->secondary_preempt_buf) + ivpu_bo_free(cmdq->secondary_preempt_buf); } static struct ivpu_cmdq *ivpu_cmdq_alloc(struct ivpu_file_priv *file_priv) { - struct xa_limit db_xa_limit = {.max = IVPU_MAX_DB, .min = IVPU_MIN_DB}; struct ivpu_device *vdev = file_priv->vdev; struct ivpu_cmdq *cmdq; int ret; @@ -88,25 +83,33 @@ static struct ivpu_cmdq *ivpu_cmdq_alloc(struct ivpu_file_priv *file_priv) if (!cmdq) return NULL; - ret = xa_alloc(&vdev->db_xa, &cmdq->db_id, NULL, db_xa_limit, GFP_KERNEL); - if (ret) { + ret = xa_alloc_cyclic(&vdev->db_xa, &cmdq->db_id, NULL, vdev->db_limit, &vdev->db_next, + GFP_KERNEL); + if (ret < 0) { ivpu_err(vdev, "Failed to allocate doorbell id: %d\n", ret); goto err_free_cmdq; } + ret = xa_alloc_cyclic(&file_priv->cmdq_xa, &cmdq->id, cmdq, file_priv->cmdq_limit, + &file_priv->cmdq_id_next, GFP_KERNEL); + if (ret < 0) { + ivpu_err(vdev, "Failed to allocate command queue id: %d\n", ret); + goto err_erase_db_xa; + } + cmdq->mem = ivpu_bo_create_global(vdev, SZ_4K, DRM_IVPU_BO_WC | DRM_IVPU_BO_MAPPABLE); if (!cmdq->mem) - goto err_erase_xa; + goto err_erase_cmdq_xa; ret = ivpu_preemption_buffers_create(vdev, file_priv, cmdq); if (ret) - goto err_free_cmdq_mem; + ivpu_warn(vdev, "Failed to allocate preemption buffers, preemption limited\n"); return cmdq; -err_free_cmdq_mem: - ivpu_bo_free(cmdq->mem); -err_erase_xa: +err_erase_cmdq_xa: + xa_erase(&file_priv->cmdq_xa, cmdq->id); +err_erase_db_xa: xa_erase(&vdev->db_xa, cmdq->db_id); err_free_cmdq: kfree(cmdq); @@ -130,13 +133,13 @@ static int ivpu_hws_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq struct ivpu_device *vdev = file_priv->vdev; int ret; - ret = ivpu_jsm_hws_create_cmdq(vdev, file_priv->ctx.id, file_priv->ctx.id, cmdq->db_id, + ret = ivpu_jsm_hws_create_cmdq(vdev, file_priv->ctx.id, file_priv->ctx.id, cmdq->id, task_pid_nr(current), engine, cmdq->mem->vpu_addr, ivpu_bo_size(cmdq->mem)); if (ret) return ret; - ret = ivpu_jsm_hws_set_context_sched_properties(vdev, file_priv->ctx.id, cmdq->db_id, + ret = ivpu_jsm_hws_set_context_sched_properties(vdev, file_priv->ctx.id, cmdq->id, priority); if (ret) return ret; @@ -149,21 +152,22 @@ static int ivpu_register_db(struct ivpu_file_priv *file_priv, struct ivpu_cmdq * struct ivpu_device *vdev = file_priv->vdev; int ret; - if (vdev->hw->sched_mode == VPU_SCHEDULING_MODE_HW) - ret = ivpu_jsm_hws_register_db(vdev, file_priv->ctx.id, cmdq->db_id, cmdq->db_id, + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) + ret = ivpu_jsm_hws_register_db(vdev, file_priv->ctx.id, cmdq->id, cmdq->db_id, cmdq->mem->vpu_addr, ivpu_bo_size(cmdq->mem)); else ret = ivpu_jsm_register_db(vdev, file_priv->ctx.id, cmdq->db_id, cmdq->mem->vpu_addr, ivpu_bo_size(cmdq->mem)); if (!ret) - ivpu_dbg(vdev, JOB, "DB %d registered to ctx %d\n", cmdq->db_id, file_priv->ctx.id); + ivpu_dbg(vdev, JOB, "DB %d registered to cmdq %d ctx %d\n", + cmdq->db_id, cmdq->id, file_priv->ctx.id); return ret; } static int -ivpu_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq, u16 engine, u8 priority) +ivpu_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq, u8 priority) { struct ivpu_device *vdev = file_priv->vdev; struct vpu_job_queue_header *jobq_header; @@ -179,13 +183,18 @@ ivpu_cmdq_init(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cmdq, u16 eng cmdq->jobq = (struct vpu_job_queue *)ivpu_bo_vaddr(cmdq->mem); jobq_header = &cmdq->jobq->header; - jobq_header->engine_idx = engine; + jobq_header->engine_idx = VPU_ENGINE_COMPUTE; jobq_header->head = 0; jobq_header->tail = 0; + if (ivpu_test_mode & IVPU_TEST_MODE_TURBO) { + ivpu_dbg(vdev, JOB, "Turbo mode enabled"); + jobq_header->flags = VPU_JOB_QUEUE_FLAGS_TURBO_MODE; + } + wmb(); /* Flush WC buffer for jobq->header */ - if (vdev->hw->sched_mode == VPU_SCHEDULING_MODE_HW) { - ret = ivpu_hws_cmdq_init(file_priv, cmdq, engine, priority); + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) { + ret = ivpu_hws_cmdq_init(file_priv, cmdq, VPU_ENGINE_COMPUTE, priority); if (ret) return ret; } @@ -211,10 +220,10 @@ static int ivpu_cmdq_fini(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cm cmdq->db_registered = false; - if (vdev->hw->sched_mode == VPU_SCHEDULING_MODE_HW) { - ret = ivpu_jsm_hws_destroy_cmdq(vdev, file_priv->ctx.id, cmdq->db_id); + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) { + ret = ivpu_jsm_hws_destroy_cmdq(vdev, file_priv->ctx.id, cmdq->id); if (!ret) - ivpu_dbg(vdev, JOB, "Command queue %d destroyed\n", cmdq->db_id); + ivpu_dbg(vdev, JOB, "Command queue %d destroyed\n", cmdq->id); } ret = ivpu_jsm_unregister_db(vdev, cmdq->db_id); @@ -224,55 +233,46 @@ static int ivpu_cmdq_fini(struct ivpu_file_priv *file_priv, struct ivpu_cmdq *cm return 0; } -static struct ivpu_cmdq *ivpu_cmdq_acquire(struct ivpu_file_priv *file_priv, u16 engine, - u8 priority) +static struct ivpu_cmdq *ivpu_cmdq_acquire(struct ivpu_file_priv *file_priv, u8 priority) { - int cmdq_idx = IVPU_CMDQ_INDEX(engine, priority); - struct ivpu_cmdq *cmdq = file_priv->cmdq[cmdq_idx]; + struct ivpu_cmdq *cmdq; + unsigned long cmdq_id; int ret; lockdep_assert_held(&file_priv->lock); + xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) + if (cmdq->priority == priority) + break; + if (!cmdq) { cmdq = ivpu_cmdq_alloc(file_priv); if (!cmdq) return NULL; - file_priv->cmdq[cmdq_idx] = cmdq; + cmdq->priority = priority; } - ret = ivpu_cmdq_init(file_priv, cmdq, engine, priority); + ret = ivpu_cmdq_init(file_priv, cmdq, priority); if (ret) return NULL; return cmdq; } -static void ivpu_cmdq_release_locked(struct ivpu_file_priv *file_priv, u16 engine, u8 priority) +void ivpu_cmdq_release_all_locked(struct ivpu_file_priv *file_priv) { - int cmdq_idx = IVPU_CMDQ_INDEX(engine, priority); - struct ivpu_cmdq *cmdq = file_priv->cmdq[cmdq_idx]; + struct ivpu_cmdq *cmdq; + unsigned long cmdq_id; lockdep_assert_held(&file_priv->lock); - if (cmdq) { - file_priv->cmdq[cmdq_idx] = NULL; + xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) { + xa_erase(&file_priv->cmdq_xa, cmdq_id); ivpu_cmdq_fini(file_priv, cmdq); ivpu_cmdq_free(file_priv, cmdq); } } -void ivpu_cmdq_release_all_locked(struct ivpu_file_priv *file_priv) -{ - u16 engine; - u8 priority; - - lockdep_assert_held(&file_priv->lock); - - for (engine = 0; engine < IVPU_NUM_ENGINES; engine++) - for (priority = 0; priority < IVPU_NUM_PRIORITIES; priority++) - ivpu_cmdq_release_locked(file_priv, engine, priority); -} - /* * Mark the doorbell as unregistered * This function needs to be called when the VPU hardware is restarted @@ -281,20 +281,13 @@ void ivpu_cmdq_release_all_locked(struct ivpu_file_priv *file_priv) */ static void ivpu_cmdq_reset(struct ivpu_file_priv *file_priv) { - u16 engine; - u8 priority; + struct ivpu_cmdq *cmdq; + unsigned long cmdq_id; mutex_lock(&file_priv->lock); - for (engine = 0; engine < IVPU_NUM_ENGINES; engine++) { - for (priority = 0; priority < IVPU_NUM_PRIORITIES; priority++) { - int cmdq_idx = IVPU_CMDQ_INDEX(engine, priority); - struct ivpu_cmdq *cmdq = file_priv->cmdq[cmdq_idx]; - - if (cmdq) - cmdq->db_registered = false; - } - } + xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) + cmdq->db_registered = false; mutex_unlock(&file_priv->lock); } @@ -314,17 +307,11 @@ void ivpu_cmdq_reset_all_contexts(struct ivpu_device *vdev) static void ivpu_cmdq_fini_all(struct ivpu_file_priv *file_priv) { - u16 engine; - u8 priority; + struct ivpu_cmdq *cmdq; + unsigned long cmdq_id; - for (engine = 0; engine < IVPU_NUM_ENGINES; engine++) { - for (priority = 0; priority < IVPU_NUM_PRIORITIES; priority++) { - int cmdq_idx = IVPU_CMDQ_INDEX(engine, priority); - - if (file_priv->cmdq[cmdq_idx]) - ivpu_cmdq_fini(file_priv, file_priv->cmdq[cmdq_idx]); - } - } + xa_for_each(&file_priv->cmdq_xa, cmdq_id, cmdq) + ivpu_cmdq_fini(file_priv, cmdq); } void ivpu_context_abort_locked(struct ivpu_file_priv *file_priv) @@ -335,7 +322,7 @@ void ivpu_context_abort_locked(struct ivpu_file_priv *file_priv) ivpu_cmdq_fini_all(file_priv); - if (vdev->hw->sched_mode == VPU_SCHEDULING_MODE_OS) + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_OS) ivpu_jsm_context_release(vdev, file_priv->ctx.id); } @@ -349,24 +336,29 @@ static int ivpu_cmdq_push_job(struct ivpu_cmdq *cmdq, struct ivpu_job *job) /* Check if there is space left in job queue */ if (next_entry == header->head) { - ivpu_dbg(vdev, JOB, "Job queue full: ctx %d engine %d db %d head %d tail %d\n", - job->file_priv->ctx.id, job->engine_idx, cmdq->db_id, header->head, tail); + ivpu_dbg(vdev, JOB, "Job queue full: ctx %d cmdq %d db %d head %d tail %d\n", + job->file_priv->ctx.id, cmdq->id, cmdq->db_id, header->head, tail); return -EBUSY; } - entry = &cmdq->jobq->job[tail]; + entry = &cmdq->jobq->slot[tail].job; entry->batch_buf_addr = job->cmd_buf_vpu_addr; entry->job_id = job->job_id; entry->flags = 0; if (unlikely(ivpu_test_mode & IVPU_TEST_MODE_NULL_SUBMISSION)) entry->flags = VPU_JOB_FLAGS_NULL_SUBMISSION_MASK; - if (vdev->hw->sched_mode == VPU_SCHEDULING_MODE_HW && - (unlikely(!(ivpu_test_mode & IVPU_TEST_MODE_PREEMPTION_DISABLE)))) { - entry->primary_preempt_buf_addr = cmdq->primary_preempt_buf->vpu_addr; - entry->primary_preempt_buf_size = ivpu_bo_size(cmdq->primary_preempt_buf); - entry->secondary_preempt_buf_addr = cmdq->secondary_preempt_buf->vpu_addr; - entry->secondary_preempt_buf_size = ivpu_bo_size(cmdq->secondary_preempt_buf); + if (vdev->fw->sched_mode == VPU_SCHEDULING_MODE_HW) { + if (cmdq->primary_preempt_buf) { + entry->primary_preempt_buf_addr = cmdq->primary_preempt_buf->vpu_addr; + entry->primary_preempt_buf_size = ivpu_bo_size(cmdq->primary_preempt_buf); + } + + if (cmdq->secondary_preempt_buf) { + entry->secondary_preempt_buf_addr = cmdq->secondary_preempt_buf->vpu_addr; + entry->secondary_preempt_buf_size = + ivpu_bo_size(cmdq->secondary_preempt_buf); + } } wmb(); /* Ensure that tail is updated after filling entry */ @@ -457,6 +449,7 @@ ivpu_job_create(struct ivpu_file_priv *file_priv, u32 engine_idx, u32 bo_count) job->file_priv = ivpu_file_priv_get(file_priv); + trace_job("create", job); ivpu_dbg(vdev, JOB, "Job created: ctx %2d engine %d", file_priv->ctx.id, job->engine_idx); return job; @@ -496,6 +489,7 @@ static int ivpu_job_signal_and_destroy(struct ivpu_device *vdev, u32 job_id, u32 job->bos[CMD_BUF_IDX]->job_status = job_status; dma_fence_signal(job->done_fence); + trace_job("done", job); ivpu_dbg(vdev, JOB, "Job complete: id %3u ctx %2d engine %d status 0x%x\n", job->job_id, job->file_priv->ctx.id, job->engine_idx, job_status); @@ -519,7 +513,6 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority) { struct ivpu_file_priv *file_priv = job->file_priv; struct ivpu_device *vdev = job->vdev; - struct xa_limit job_id_range; struct ivpu_cmdq *cmdq; bool is_first_job; int ret; @@ -530,7 +523,7 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority) mutex_lock(&file_priv->lock); - cmdq = ivpu_cmdq_acquire(job->file_priv, job->engine_idx, priority); + cmdq = ivpu_cmdq_acquire(file_priv, priority); if (!cmdq) { ivpu_warn_ratelimited(vdev, "Failed to get job queue, ctx %d engine %d prio %d\n", file_priv->ctx.id, job->engine_idx, priority); @@ -538,13 +531,11 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority) goto err_unlock_file_priv; } - job_id_range.min = FIELD_PREP(JOB_ID_CONTEXT_MASK, (file_priv->ctx.id - 1)); - job_id_range.max = job_id_range.min | JOB_ID_JOB_MASK; - xa_lock(&vdev->submitted_jobs_xa); is_first_job = xa_empty(&vdev->submitted_jobs_xa); - ret = __xa_alloc(&vdev->submitted_jobs_xa, &job->job_id, job, job_id_range, GFP_KERNEL); - if (ret) { + ret = __xa_alloc_cyclic(&vdev->submitted_jobs_xa, &job->job_id, job, file_priv->job_limit, + &file_priv->job_id_next, GFP_KERNEL); + if (ret < 0) { ivpu_dbg(vdev, JOB, "Too many active jobs in ctx %d\n", file_priv->ctx.id); ret = -EBUSY; @@ -566,6 +557,7 @@ static int ivpu_job_submit(struct ivpu_job *job, u8 priority) vdev->busy_start_ts = ktime_get(); } + trace_job("submit", job); ivpu_dbg(vdev, JOB, "Job submitted: id %3u ctx %2d engine %d prio %d addr 0x%llx next %d\n", job->job_id, file_priv->ctx.id, job->engine_idx, priority, job->cmd_buf_vpu_addr, cmdq->jobq->header.tail); @@ -673,7 +665,7 @@ int ivpu_submit_ioctl(struct drm_device *dev, void *data, struct drm_file *file) int idx, ret; u8 priority; - if (params->engine > DRM_IVPU_ENGINE_COPY) + if (params->engine != DRM_IVPU_ENGINE_COMPUTE) return -EINVAL; if (params->priority > DRM_IVPU_JOB_PRIORITY_REALTIME) diff --git a/drivers/accel/ivpu/ivpu_job.h b/drivers/accel/ivpu/ivpu_job.h index 6accb94028c7..8b19e3f8b4cf 100644 --- a/drivers/accel/ivpu/ivpu_job.h +++ b/drivers/accel/ivpu/ivpu_job.h @@ -28,8 +28,10 @@ struct ivpu_cmdq { struct ivpu_bo *secondary_preempt_buf; struct ivpu_bo *mem; u32 entry_count; + u32 id; u32 db_id; bool db_registered; + u8 priority; }; /** diff --git a/drivers/accel/ivpu/ivpu_jsm_msg.c b/drivers/accel/ivpu/ivpu_jsm_msg.c index 46ef16c3c069..30a40be76930 100644 --- a/drivers/accel/ivpu/ivpu_jsm_msg.c +++ b/drivers/accel/ivpu/ivpu_jsm_msg.c @@ -48,9 +48,10 @@ const char *ivpu_jsm_msg_type_to_str(enum vpu_ipc_msg_type type) IVPU_CASE_TO_STR(VPU_JSM_MSG_HWS_RESUME_ENGINE_DONE); IVPU_CASE_TO_STR(VPU_JSM_MSG_STATE_DUMP); IVPU_CASE_TO_STR(VPU_JSM_MSG_STATE_DUMP_RSP); - IVPU_CASE_TO_STR(VPU_JSM_MSG_BLOB_DEINIT); + IVPU_CASE_TO_STR(VPU_JSM_MSG_BLOB_DEINIT_DEPRECATED); IVPU_CASE_TO_STR(VPU_JSM_MSG_DYNDBG_CONTROL); IVPU_CASE_TO_STR(VPU_JSM_MSG_JOB_DONE); + IVPU_CASE_TO_STR(VPU_JSM_MSG_NATIVE_FENCE_SIGNALLED); IVPU_CASE_TO_STR(VPU_JSM_MSG_ENGINE_RESET_DONE); IVPU_CASE_TO_STR(VPU_JSM_MSG_ENGINE_PREEMPT_DONE); IVPU_CASE_TO_STR(VPU_JSM_MSG_REGISTER_DB_DONE); @@ -131,7 +132,7 @@ int ivpu_jsm_get_heartbeat(struct ivpu_device *vdev, u32 engine, u64 *heartbeat) struct vpu_jsm_msg resp; int ret; - if (engine > VPU_ENGINE_COPY) + if (engine != VPU_ENGINE_COMPUTE) return -EINVAL; req.payload.query_engine_hb.engine_idx = engine; @@ -154,7 +155,7 @@ int ivpu_jsm_reset_engine(struct ivpu_device *vdev, u32 engine) struct vpu_jsm_msg resp; int ret; - if (engine > VPU_ENGINE_COPY) + if (engine != VPU_ENGINE_COMPUTE) return -EINVAL; req.payload.engine_reset.engine_idx = engine; @@ -173,7 +174,7 @@ int ivpu_jsm_preempt_engine(struct ivpu_device *vdev, u32 engine, u32 preempt_id struct vpu_jsm_msg resp; int ret; - if (engine > VPU_ENGINE_COPY) + if (engine != VPU_ENGINE_COMPUTE) return -EINVAL; req.payload.engine_preempt.engine_idx = engine; @@ -196,7 +197,7 @@ int ivpu_jsm_dyndbg_control(struct ivpu_device *vdev, char *command, size_t size strscpy(req.payload.dyndbg_control.dyndbg_cmd, command, VPU_DYNDBG_CMD_MAX_LEN); ret = ivpu_ipc_send_receive(vdev, &req, VPU_JSM_MSG_DYNDBG_CONTROL_RSP, &resp, - VPU_IPC_CHAN_ASYNC_CMD, vdev->timeout.jsm); + VPU_IPC_CHAN_GEN_CMD, vdev->timeout.jsm); if (ret) ivpu_warn_ratelimited(vdev, "Failed to send command \"%s\": ret %d\n", command, ret); @@ -270,9 +271,8 @@ int ivpu_jsm_pwr_d0i3_enter(struct ivpu_device *vdev) req.payload.pwr_d0i3_enter.send_response = 1; - ret = ivpu_ipc_send_receive_active(vdev, &req, VPU_JSM_MSG_PWR_D0I3_ENTER_DONE, - &resp, VPU_IPC_CHAN_GEN_CMD, - vdev->timeout.d0i3_entry_msg); + ret = ivpu_ipc_send_receive_internal(vdev, &req, VPU_JSM_MSG_PWR_D0I3_ENTER_DONE, &resp, + VPU_IPC_CHAN_GEN_CMD, vdev->timeout.d0i3_entry_msg); if (ret) return ret; @@ -346,7 +346,7 @@ int ivpu_jsm_hws_resume_engine(struct ivpu_device *vdev, u32 engine) struct vpu_jsm_msg resp; int ret; - if (engine >= VPU_ENGINE_NB) + if (engine != VPU_ENGINE_COMPUTE) return -EINVAL; req.payload.hws_resume_engine.engine_idx = engine; @@ -394,8 +394,6 @@ int ivpu_jsm_hws_set_scheduling_log(struct ivpu_device *vdev, u32 engine_idx, u3 req.payload.hws_set_scheduling_log.host_ssid = host_ssid; req.payload.hws_set_scheduling_log.vpu_log_buffer_va = vpu_log_buffer_va; req.payload.hws_set_scheduling_log.notify_index = 0; - req.payload.hws_set_scheduling_log.enable_extra_events = - ivpu_test_mode & IVPU_TEST_MODE_HWS_EXTRA_EVENTS; ret = ivpu_ipc_send_receive(vdev, &req, VPU_JSM_MSG_HWS_SET_SCHEDULING_LOG_RSP, &resp, VPU_IPC_CHAN_ASYNC_CMD, vdev->timeout.jsm); @@ -430,8 +428,8 @@ int ivpu_jsm_hws_setup_priority_bands(struct ivpu_device *vdev) req.payload.hws_priority_band_setup.normal_band_percentage = 10; - ret = ivpu_ipc_send_receive_active(vdev, &req, VPU_JSM_MSG_SET_PRIORITY_BAND_SETUP_RSP, - &resp, VPU_IPC_CHAN_ASYNC_CMD, vdev->timeout.jsm); + ret = ivpu_ipc_send_receive_internal(vdev, &req, VPU_JSM_MSG_SET_PRIORITY_BAND_SETUP_RSP, + &resp, VPU_IPC_CHAN_ASYNC_CMD, vdev->timeout.jsm); if (ret) ivpu_warn_ratelimited(vdev, "Failed to set priority bands: %d\n", ret); @@ -544,9 +542,8 @@ int ivpu_jsm_dct_enable(struct ivpu_device *vdev, u32 active_us, u32 inactive_us req.payload.pwr_dct_control.dct_active_us = active_us; req.payload.pwr_dct_control.dct_inactive_us = inactive_us; - return ivpu_ipc_send_receive_active(vdev, &req, VPU_JSM_MSG_DCT_ENABLE_DONE, - &resp, VPU_IPC_CHAN_ASYNC_CMD, - vdev->timeout.jsm); + return ivpu_ipc_send_receive_internal(vdev, &req, VPU_JSM_MSG_DCT_ENABLE_DONE, &resp, + VPU_IPC_CHAN_ASYNC_CMD, vdev->timeout.jsm); } int ivpu_jsm_dct_disable(struct ivpu_device *vdev) @@ -554,7 +551,14 @@ int ivpu_jsm_dct_disable(struct ivpu_device *vdev) struct vpu_jsm_msg req = { .type = VPU_JSM_MSG_DCT_DISABLE }; struct vpu_jsm_msg resp; - return ivpu_ipc_send_receive_active(vdev, &req, VPU_JSM_MSG_DCT_DISABLE_DONE, - &resp, VPU_IPC_CHAN_ASYNC_CMD, - vdev->timeout.jsm); + return ivpu_ipc_send_receive_internal(vdev, &req, VPU_JSM_MSG_DCT_DISABLE_DONE, &resp, + VPU_IPC_CHAN_ASYNC_CMD, vdev->timeout.jsm); +} + +int ivpu_jsm_state_dump(struct ivpu_device *vdev) +{ + struct vpu_jsm_msg req = { .type = VPU_JSM_MSG_STATE_DUMP }; + + return ivpu_ipc_send_and_wait(vdev, &req, VPU_IPC_CHAN_ASYNC_CMD, + vdev->timeout.state_dump_msg); } diff --git a/drivers/accel/ivpu/ivpu_jsm_msg.h b/drivers/accel/ivpu/ivpu_jsm_msg.h index e4e42c0ff6e6..9e84d3526a14 100644 --- a/drivers/accel/ivpu/ivpu_jsm_msg.h +++ b/drivers/accel/ivpu/ivpu_jsm_msg.h @@ -43,4 +43,6 @@ int ivpu_jsm_metric_streamer_info(struct ivpu_device *vdev, u64 metric_group_mas u64 buffer_size, u32 *sample_size, u64 *info_size); int ivpu_jsm_dct_enable(struct ivpu_device *vdev, u32 active_us, u32 inactive_us); int ivpu_jsm_dct_disable(struct ivpu_device *vdev); +int ivpu_jsm_state_dump(struct ivpu_device *vdev); + #endif diff --git a/drivers/accel/ivpu/ivpu_mmu.c b/drivers/accel/ivpu/ivpu_mmu.c index c078e214b221..26ef52fbb93e 100644 --- a/drivers/accel/ivpu/ivpu_mmu.c +++ b/drivers/accel/ivpu/ivpu_mmu.c @@ -696,7 +696,7 @@ unlock: return ret; } -static int ivpu_mmu_cd_add(struct ivpu_device *vdev, u32 ssid, u64 cd_dma) +static int ivpu_mmu_cdtab_entry_set(struct ivpu_device *vdev, u32 ssid, u64 cd_dma, bool valid) { struct ivpu_mmu_info *mmu = vdev->mmu; struct ivpu_mmu_cdtab *cdtab = &mmu->cdtab; @@ -708,30 +708,29 @@ static int ivpu_mmu_cd_add(struct ivpu_device *vdev, u32 ssid, u64 cd_dma) return -EINVAL; entry = cdtab->base + (ssid * IVPU_MMU_CDTAB_ENT_SIZE); + drm_WARN_ON(&vdev->drm, (entry[0] & IVPU_MMU_CD_0_V) == valid); - if (cd_dma != 0) { - cd[0] = FIELD_PREP(IVPU_MMU_CD_0_TCR_T0SZ, IVPU_MMU_T0SZ_48BIT) | - FIELD_PREP(IVPU_MMU_CD_0_TCR_TG0, 0) | - FIELD_PREP(IVPU_MMU_CD_0_TCR_IRGN0, 0) | - FIELD_PREP(IVPU_MMU_CD_0_TCR_ORGN0, 0) | - FIELD_PREP(IVPU_MMU_CD_0_TCR_SH0, 0) | - FIELD_PREP(IVPU_MMU_CD_0_TCR_IPS, IVPU_MMU_IPS_48BIT) | - FIELD_PREP(IVPU_MMU_CD_0_ASID, ssid) | - IVPU_MMU_CD_0_TCR_EPD1 | - IVPU_MMU_CD_0_AA64 | - IVPU_MMU_CD_0_R | - IVPU_MMU_CD_0_ASET | - IVPU_MMU_CD_0_V; - cd[1] = cd_dma & IVPU_MMU_CD_1_TTB0_MASK; - cd[2] = 0; - cd[3] = 0x0000000000007444; + cd[0] = FIELD_PREP(IVPU_MMU_CD_0_TCR_T0SZ, IVPU_MMU_T0SZ_48BIT) | + FIELD_PREP(IVPU_MMU_CD_0_TCR_TG0, 0) | + FIELD_PREP(IVPU_MMU_CD_0_TCR_IRGN0, 0) | + FIELD_PREP(IVPU_MMU_CD_0_TCR_ORGN0, 0) | + FIELD_PREP(IVPU_MMU_CD_0_TCR_SH0, 0) | + FIELD_PREP(IVPU_MMU_CD_0_TCR_IPS, IVPU_MMU_IPS_48BIT) | + FIELD_PREP(IVPU_MMU_CD_0_ASID, ssid) | + IVPU_MMU_CD_0_TCR_EPD1 | + IVPU_MMU_CD_0_AA64 | + IVPU_MMU_CD_0_R | + IVPU_MMU_CD_0_ASET; + cd[1] = cd_dma & IVPU_MMU_CD_1_TTB0_MASK; + cd[2] = 0; + cd[3] = 0x0000000000007444; - /* For global context generate memory fault on VPU */ - if (ssid == IVPU_GLOBAL_CONTEXT_MMU_SSID) - cd[0] |= IVPU_MMU_CD_0_A; - } else { - memset(cd, 0, sizeof(cd)); - } + /* For global context generate memory fault on VPU */ + if (ssid == IVPU_GLOBAL_CONTEXT_MMU_SSID) + cd[0] |= IVPU_MMU_CD_0_A; + + if (valid) + cd[0] |= IVPU_MMU_CD_0_V; WRITE_ONCE(entry[1], cd[1]); WRITE_ONCE(entry[2], cd[2]); @@ -741,8 +740,8 @@ static int ivpu_mmu_cd_add(struct ivpu_device *vdev, u32 ssid, u64 cd_dma) if (!ivpu_is_force_snoop_enabled(vdev)) clflush_cache_range(entry, IVPU_MMU_CDTAB_ENT_SIZE); - ivpu_dbg(vdev, MMU, "CDTAB %s entry (SSID=%u, dma=%pad): 0x%llx, 0x%llx, 0x%llx, 0x%llx\n", - cd_dma ? "write" : "clear", ssid, &cd_dma, cd[0], cd[1], cd[2], cd[3]); + ivpu_dbg(vdev, MMU, "CDTAB set %s entry (SSID=%u, dma=%pad): 0x%llx, 0x%llx, 0x%llx, 0x%llx\n", + valid ? "valid" : "invalid", ssid, &cd_dma, cd[0], cd[1], cd[2], cd[3]); mutex_lock(&mmu->lock); if (!mmu->on) @@ -750,38 +749,18 @@ static int ivpu_mmu_cd_add(struct ivpu_device *vdev, u32 ssid, u64 cd_dma) ret = ivpu_mmu_cmdq_write_cfgi_all(vdev); if (ret) - goto unlock; + goto err_invalidate; ret = ivpu_mmu_cmdq_sync(vdev); + if (ret) + goto err_invalidate; unlock: mutex_unlock(&mmu->lock); - return ret; -} - -static int ivpu_mmu_cd_add_gbl(struct ivpu_device *vdev) -{ - int ret; - - ret = ivpu_mmu_cd_add(vdev, 0, vdev->gctx.pgtable.pgd_dma); - if (ret) - ivpu_err(vdev, "Failed to add global CD entry: %d\n", ret); - - return ret; -} - -static int ivpu_mmu_cd_add_user(struct ivpu_device *vdev, u32 ssid, dma_addr_t cd_dma) -{ - int ret; - - if (ssid == 0) { - ivpu_err(vdev, "Invalid SSID: %u\n", ssid); - return -EINVAL; - } - - ret = ivpu_mmu_cd_add(vdev, ssid, cd_dma); - if (ret) - ivpu_err(vdev, "Failed to add CD entry SSID=%u: %d\n", ssid, ret); + return 0; +err_invalidate: + WRITE_ONCE(entry[0], 0); + mutex_unlock(&mmu->lock); return ret; } @@ -808,12 +787,6 @@ int ivpu_mmu_init(struct ivpu_device *vdev) return ret; } - ret = ivpu_mmu_cd_add_gbl(vdev); - if (ret) { - ivpu_err(vdev, "Failed to initialize strtab: %d\n", ret); - return ret; - } - ret = ivpu_mmu_enable(vdev); if (ret) { ivpu_err(vdev, "Failed to resume MMU: %d\n", ret); @@ -966,12 +939,12 @@ void ivpu_mmu_irq_gerr_handler(struct ivpu_device *vdev) REGV_WR32(IVPU_MMU_REG_GERRORN, gerror_val); } -int ivpu_mmu_set_pgtable(struct ivpu_device *vdev, int ssid, struct ivpu_mmu_pgtable *pgtable) +int ivpu_mmu_cd_set(struct ivpu_device *vdev, int ssid, struct ivpu_mmu_pgtable *pgtable) { - return ivpu_mmu_cd_add_user(vdev, ssid, pgtable->pgd_dma); + return ivpu_mmu_cdtab_entry_set(vdev, ssid, pgtable->pgd_dma, true); } -void ivpu_mmu_clear_pgtable(struct ivpu_device *vdev, int ssid) +void ivpu_mmu_cd_clear(struct ivpu_device *vdev, int ssid) { - ivpu_mmu_cd_add_user(vdev, ssid, 0); /* 0 will clear CD entry */ + ivpu_mmu_cdtab_entry_set(vdev, ssid, 0, false); } diff --git a/drivers/accel/ivpu/ivpu_mmu.h b/drivers/accel/ivpu/ivpu_mmu.h index 6fa35c240710..7afea9cd8731 100644 --- a/drivers/accel/ivpu/ivpu_mmu.h +++ b/drivers/accel/ivpu/ivpu_mmu.h @@ -40,8 +40,8 @@ struct ivpu_mmu_info { int ivpu_mmu_init(struct ivpu_device *vdev); void ivpu_mmu_disable(struct ivpu_device *vdev); int ivpu_mmu_enable(struct ivpu_device *vdev); -int ivpu_mmu_set_pgtable(struct ivpu_device *vdev, int ssid, struct ivpu_mmu_pgtable *pgtable); -void ivpu_mmu_clear_pgtable(struct ivpu_device *vdev, int ssid); +int ivpu_mmu_cd_set(struct ivpu_device *vdev, int ssid, struct ivpu_mmu_pgtable *pgtable); +void ivpu_mmu_cd_clear(struct ivpu_device *vdev, int ssid); int ivpu_mmu_invalidate_tlb(struct ivpu_device *vdev, u16 ssid); void ivpu_mmu_irq_evtq_handler(struct ivpu_device *vdev); diff --git a/drivers/accel/ivpu/ivpu_mmu_context.c b/drivers/accel/ivpu/ivpu_mmu_context.c index bbe652a7019d..891967a95bc3 100644 --- a/drivers/accel/ivpu/ivpu_mmu_context.c +++ b/drivers/accel/ivpu/ivpu_mmu_context.c @@ -90,19 +90,6 @@ static void ivpu_pgtable_free_page(struct ivpu_device *vdev, u64 *cpu_addr, dma_ } } -static int ivpu_mmu_pgtable_init(struct ivpu_device *vdev, struct ivpu_mmu_pgtable *pgtable) -{ - dma_addr_t pgd_dma; - - pgtable->pgd_dma_ptr = ivpu_pgtable_alloc_page(vdev, &pgd_dma); - if (!pgtable->pgd_dma_ptr) - return -ENOMEM; - - pgtable->pgd_dma = pgd_dma; - - return 0; -} - static void ivpu_mmu_pgtables_free(struct ivpu_device *vdev, struct ivpu_mmu_pgtable *pgtable) { int pgd_idx, pud_idx, pmd_idx; @@ -140,6 +127,27 @@ static void ivpu_mmu_pgtables_free(struct ivpu_device *vdev, struct ivpu_mmu_pgt } ivpu_pgtable_free_page(vdev, pgtable->pgd_dma_ptr, pgtable->pgd_dma); + pgtable->pgd_dma_ptr = NULL; + pgtable->pgd_dma = 0; +} + +static u64* +ivpu_mmu_ensure_pgd(struct ivpu_device *vdev, struct ivpu_mmu_pgtable *pgtable) +{ + u64 *pgd_dma_ptr = pgtable->pgd_dma_ptr; + dma_addr_t pgd_dma; + + if (pgd_dma_ptr) + return pgd_dma_ptr; + + pgd_dma_ptr = ivpu_pgtable_alloc_page(vdev, &pgd_dma); + if (!pgd_dma_ptr) + return NULL; + + pgtable->pgd_dma_ptr = pgd_dma_ptr; + pgtable->pgd_dma = pgd_dma; + + return pgd_dma_ptr; } static u64* @@ -237,6 +245,12 @@ ivpu_mmu_context_map_page(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx int pmd_idx = FIELD_GET(IVPU_MMU_PMD_INDEX_MASK, vpu_addr); int pte_idx = FIELD_GET(IVPU_MMU_PTE_INDEX_MASK, vpu_addr); + drm_WARN_ON(&vdev->drm, ctx->id == IVPU_RESERVED_CONTEXT_MMU_SSID); + + /* Allocate PGD - first level page table if needed */ + if (!ivpu_mmu_ensure_pgd(vdev, &ctx->pgtable)) + return -ENOMEM; + /* Allocate PUD - second level page table if needed */ if (!ivpu_mmu_ensure_pud(vdev, &ctx->pgtable, pgd_idx)) return -ENOMEM; @@ -418,6 +432,7 @@ int ivpu_mmu_context_map_sgt(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u64 vpu_addr, struct sg_table *sgt, bool llc_coherent) { + size_t start_vpu_addr = vpu_addr; struct scatterlist *sg; int ret; u64 prot; @@ -448,20 +463,36 @@ ivpu_mmu_context_map_sgt(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, ret = ivpu_mmu_context_map_pages(vdev, ctx, vpu_addr, dma_addr, size, prot); if (ret) { ivpu_err(vdev, "Failed to map context pages\n"); - mutex_unlock(&ctx->lock); - return ret; + goto err_unmap_pages; } vpu_addr += size; } + if (!ctx->is_cd_valid) { + ret = ivpu_mmu_cd_set(vdev, ctx->id, &ctx->pgtable); + if (ret) { + ivpu_err(vdev, "Failed to set context descriptor for context %u: %d\n", + ctx->id, ret); + goto err_unmap_pages; + } + ctx->is_cd_valid = true; + } + /* Ensure page table modifications are flushed from wc buffers to memory */ wmb(); - mutex_unlock(&ctx->lock); - ret = ivpu_mmu_invalidate_tlb(vdev, ctx->id); - if (ret) + if (ret) { ivpu_err(vdev, "Failed to invalidate TLB for ctx %u: %d\n", ctx->id, ret); + goto err_unmap_pages; + } + + mutex_unlock(&ctx->lock); + return 0; + +err_unmap_pages: + ivpu_mmu_context_unmap_pages(ctx, start_vpu_addr, vpu_addr - start_vpu_addr); + mutex_unlock(&ctx->lock); return ret; } @@ -530,65 +561,75 @@ ivpu_mmu_context_remove_node(struct ivpu_mmu_context *ctx, struct drm_mm_node *n mutex_unlock(&ctx->lock); } -static int -ivpu_mmu_context_init(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u32 context_id) +void ivpu_mmu_context_init(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u32 context_id) { u64 start, end; - int ret; mutex_init(&ctx->lock); - ret = ivpu_mmu_pgtable_init(vdev, &ctx->pgtable); - if (ret) { - ivpu_err(vdev, "Failed to initialize pgtable for ctx %u: %d\n", context_id, ret); - return ret; - } - if (!context_id) { start = vdev->hw->ranges.global.start; end = vdev->hw->ranges.shave.end; } else { - start = vdev->hw->ranges.user.start; - end = vdev->hw->ranges.dma.end; + start = min_t(u64, vdev->hw->ranges.user.start, vdev->hw->ranges.shave.start); + end = max_t(u64, vdev->hw->ranges.user.end, vdev->hw->ranges.dma.end); } drm_mm_init(&ctx->mm, start, end - start); ctx->id = context_id; - - return 0; } -static void ivpu_mmu_context_fini(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx) +void ivpu_mmu_context_fini(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx) { - if (drm_WARN_ON(&vdev->drm, !ctx->pgtable.pgd_dma_ptr)) - return; + if (ctx->is_cd_valid) { + ivpu_mmu_cd_clear(vdev, ctx->id); + ctx->is_cd_valid = false; + } mutex_destroy(&ctx->lock); ivpu_mmu_pgtables_free(vdev, &ctx->pgtable); drm_mm_takedown(&ctx->mm); - - ctx->pgtable.pgd_dma_ptr = NULL; - ctx->pgtable.pgd_dma = 0; } -int ivpu_mmu_global_context_init(struct ivpu_device *vdev) +void ivpu_mmu_global_context_init(struct ivpu_device *vdev) { - return ivpu_mmu_context_init(vdev, &vdev->gctx, IVPU_GLOBAL_CONTEXT_MMU_SSID); + ivpu_mmu_context_init(vdev, &vdev->gctx, IVPU_GLOBAL_CONTEXT_MMU_SSID); } void ivpu_mmu_global_context_fini(struct ivpu_device *vdev) { - return ivpu_mmu_context_fini(vdev, &vdev->gctx); + ivpu_mmu_context_fini(vdev, &vdev->gctx); } int ivpu_mmu_reserved_context_init(struct ivpu_device *vdev) { - return ivpu_mmu_user_context_init(vdev, &vdev->rctx, IVPU_RESERVED_CONTEXT_MMU_SSID); + int ret; + + ivpu_mmu_context_init(vdev, &vdev->rctx, IVPU_RESERVED_CONTEXT_MMU_SSID); + + mutex_lock(&vdev->rctx.lock); + + if (!ivpu_mmu_ensure_pgd(vdev, &vdev->rctx.pgtable)) { + ivpu_err(vdev, "Failed to allocate root page table for reserved context\n"); + ret = -ENOMEM; + goto unlock; + } + + ret = ivpu_mmu_cd_set(vdev, vdev->rctx.id, &vdev->rctx.pgtable); + if (ret) { + ivpu_err(vdev, "Failed to set context descriptor for reserved context\n"); + goto unlock; + } + +unlock: + mutex_unlock(&vdev->rctx.lock); + return ret; } void ivpu_mmu_reserved_context_fini(struct ivpu_device *vdev) { - return ivpu_mmu_user_context_fini(vdev, &vdev->rctx); + ivpu_mmu_cd_clear(vdev, vdev->rctx.id); + ivpu_mmu_context_fini(vdev, &vdev->rctx); } void ivpu_mmu_user_context_mark_invalid(struct ivpu_device *vdev, u32 ssid) @@ -603,36 +644,3 @@ void ivpu_mmu_user_context_mark_invalid(struct ivpu_device *vdev, u32 ssid) xa_unlock(&vdev->context_xa); } - -int ivpu_mmu_user_context_init(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u32 ctx_id) -{ - int ret; - - drm_WARN_ON(&vdev->drm, !ctx_id); - - ret = ivpu_mmu_context_init(vdev, ctx, ctx_id); - if (ret) { - ivpu_err(vdev, "Failed to initialize context %u: %d\n", ctx_id, ret); - return ret; - } - - ret = ivpu_mmu_set_pgtable(vdev, ctx_id, &ctx->pgtable); - if (ret) { - ivpu_err(vdev, "Failed to set page table for context %u: %d\n", ctx_id, ret); - goto err_context_fini; - } - - return 0; - -err_context_fini: - ivpu_mmu_context_fini(vdev, ctx); - return ret; -} - -void ivpu_mmu_user_context_fini(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx) -{ - drm_WARN_ON(&vdev->drm, !ctx->id); - - ivpu_mmu_clear_pgtable(vdev, ctx->id); - ivpu_mmu_context_fini(vdev, ctx); -} diff --git a/drivers/accel/ivpu/ivpu_mmu_context.h b/drivers/accel/ivpu/ivpu_mmu_context.h index 7f9aaf3d10c2..8042fc067062 100644 --- a/drivers/accel/ivpu/ivpu_mmu_context.h +++ b/drivers/accel/ivpu/ivpu_mmu_context.h @@ -23,19 +23,20 @@ struct ivpu_mmu_pgtable { }; struct ivpu_mmu_context { - struct mutex lock; /* Protects: mm, pgtable */ + struct mutex lock; /* Protects: mm, pgtable, is_cd_valid */ struct drm_mm mm; struct ivpu_mmu_pgtable pgtable; + bool is_cd_valid; u32 id; }; -int ivpu_mmu_global_context_init(struct ivpu_device *vdev); +void ivpu_mmu_context_init(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u32 context_id); +void ivpu_mmu_context_fini(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx); +void ivpu_mmu_global_context_init(struct ivpu_device *vdev); void ivpu_mmu_global_context_fini(struct ivpu_device *vdev); int ivpu_mmu_reserved_context_init(struct ivpu_device *vdev); void ivpu_mmu_reserved_context_fini(struct ivpu_device *vdev); -int ivpu_mmu_user_context_init(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx, u32 ctx_id); -void ivpu_mmu_user_context_fini(struct ivpu_device *vdev, struct ivpu_mmu_context *ctx); void ivpu_mmu_user_context_mark_invalid(struct ivpu_device *vdev, u32 ssid); int ivpu_mmu_context_insert_node(struct ivpu_mmu_context *ctx, const struct ivpu_addr_range *range, diff --git a/drivers/accel/ivpu/ivpu_ms.c b/drivers/accel/ivpu/ivpu_ms.c index 2f9d37f5c208..ffe7b10f8a76 100644 --- a/drivers/accel/ivpu/ivpu_ms.c +++ b/drivers/accel/ivpu/ivpu_ms.c @@ -11,7 +11,7 @@ #include "ivpu_ms.h" #include "ivpu_pm.h" -#define MS_INFO_BUFFER_SIZE SZ_16K +#define MS_INFO_BUFFER_SIZE SZ_64K #define MS_NUM_BUFFERS 2 #define MS_READ_PERIOD_MULTIPLIER 2 #define MS_MIN_SAMPLE_PERIOD_NS 1000000 diff --git a/drivers/accel/ivpu/ivpu_pm.c b/drivers/accel/ivpu/ivpu_pm.c index 59d3170f5e35..dbc0711e28d1 100644 --- a/drivers/accel/ivpu/ivpu_pm.c +++ b/drivers/accel/ivpu/ivpu_pm.c @@ -9,21 +9,25 @@ #include #include -#include "vpu_boot_api.h" +#include "ivpu_coredump.h" #include "ivpu_drv.h" -#include "ivpu_hw.h" #include "ivpu_fw.h" #include "ivpu_fw_log.h" +#include "ivpu_hw.h" #include "ivpu_ipc.h" #include "ivpu_job.h" #include "ivpu_jsm_msg.h" #include "ivpu_mmu.h" #include "ivpu_ms.h" #include "ivpu_pm.h" +#include "ivpu_trace.h" +#include "vpu_boot_api.h" static bool ivpu_disable_recovery; +#if IS_ENABLED(CONFIG_DRM_ACCEL_IVPU_DEBUG) module_param_named_unsafe(disable_recovery, ivpu_disable_recovery, bool, 0644); MODULE_PARM_DESC(disable_recovery, "Disables recovery when NPU hang is detected"); +#endif static unsigned long ivpu_tdr_timeout_ms; module_param_named(tdr_timeout_ms, ivpu_tdr_timeout_ms, ulong, 0644); @@ -37,6 +41,7 @@ static void ivpu_pm_prepare_cold_boot(struct ivpu_device *vdev) ivpu_cmdq_reset_all_contexts(vdev); ivpu_ipc_reset(vdev); + ivpu_fw_log_reset(vdev); ivpu_fw_load(vdev); fw->entry_point = fw->cold_boot_entry_point; } @@ -123,7 +128,8 @@ static void ivpu_pm_recovery_work(struct work_struct *work) if (ret) ivpu_err(vdev, "Failed to resume NPU: %d\n", ret); - ivpu_fw_log_dump(vdev); + ivpu_jsm_state_dump(vdev); + ivpu_dev_coredump(vdev); atomic_inc(&vdev->pm->reset_counter); atomic_set(&vdev->pm->reset_pending, 1); @@ -195,6 +201,7 @@ int ivpu_pm_suspend_cb(struct device *dev) struct ivpu_device *vdev = to_ivpu_device(drm); unsigned long timeout; + trace_pm("suspend"); ivpu_dbg(vdev, PM, "Suspend..\n"); timeout = jiffies + msecs_to_jiffies(vdev->timeout.tdr); @@ -212,6 +219,7 @@ int ivpu_pm_suspend_cb(struct device *dev) ivpu_pm_prepare_warm_boot(vdev); ivpu_dbg(vdev, PM, "Suspend done.\n"); + trace_pm("suspend done"); return 0; } @@ -222,6 +230,7 @@ int ivpu_pm_resume_cb(struct device *dev) struct ivpu_device *vdev = to_ivpu_device(drm); int ret; + trace_pm("resume"); ivpu_dbg(vdev, PM, "Resume..\n"); ret = ivpu_resume(vdev); @@ -229,6 +238,7 @@ int ivpu_pm_resume_cb(struct device *dev) ivpu_err(vdev, "Failed to resume: %d\n", ret); ivpu_dbg(vdev, PM, "Resume done.\n"); + trace_pm("resume done"); return ret; } @@ -243,6 +253,7 @@ int ivpu_pm_runtime_suspend_cb(struct device *dev) drm_WARN_ON(&vdev->drm, !xa_empty(&vdev->submitted_jobs_xa)); drm_WARN_ON(&vdev->drm, work_pending(&vdev->pm->recovery_work)); + trace_pm("runtime suspend"); ivpu_dbg(vdev, PM, "Runtime suspend..\n"); ivpu_mmu_disable(vdev); @@ -262,13 +273,14 @@ int ivpu_pm_runtime_suspend_cb(struct device *dev) if (!is_idle || ret_d0i3) { ivpu_err(vdev, "Forcing cold boot due to previous errors\n"); atomic_inc(&vdev->pm->reset_counter); - ivpu_fw_log_dump(vdev); + ivpu_dev_coredump(vdev); ivpu_pm_prepare_cold_boot(vdev); } else { ivpu_pm_prepare_warm_boot(vdev); } ivpu_dbg(vdev, PM, "Runtime suspend done.\n"); + trace_pm("runtime suspend done"); return 0; } @@ -279,6 +291,7 @@ int ivpu_pm_runtime_resume_cb(struct device *dev) struct ivpu_device *vdev = to_ivpu_device(drm); int ret; + trace_pm("runtime resume"); ivpu_dbg(vdev, PM, "Runtime resume..\n"); ret = ivpu_resume(vdev); @@ -286,6 +299,7 @@ int ivpu_pm_runtime_resume_cb(struct device *dev) ivpu_err(vdev, "Failed to set RESUME state: %d\n", ret); ivpu_dbg(vdev, PM, "Runtime resume done.\n"); + trace_pm("runtime resume done"); return ret; } @@ -411,7 +425,7 @@ int ivpu_pm_dct_enable(struct ivpu_device *vdev, u8 active_percent) ret = ivpu_jsm_dct_enable(vdev, active_us, inactive_us); if (ret) { - ivpu_err_ratelimited(vdev, "Filed to enable DCT: %d\n", ret); + ivpu_err_ratelimited(vdev, "Failed to enable DCT: %d\n", ret); return ret; } @@ -428,7 +442,7 @@ int ivpu_pm_dct_disable(struct ivpu_device *vdev) ret = ivpu_jsm_dct_disable(vdev); if (ret) { - ivpu_err_ratelimited(vdev, "Filed to disable DCT: %d\n", ret); + ivpu_err_ratelimited(vdev, "Failed to disable DCT: %d\n", ret); return ret; } diff --git a/drivers/accel/ivpu/ivpu_sysfs.c b/drivers/accel/ivpu/ivpu_sysfs.c index 913669f1786e..616477fc17fa 100644 --- a/drivers/accel/ivpu/ivpu_sysfs.c +++ b/drivers/accel/ivpu/ivpu_sysfs.c @@ -6,6 +6,8 @@ #include #include +#include "ivpu_drv.h" +#include "ivpu_fw.h" #include "ivpu_hw.h" #include "ivpu_sysfs.h" @@ -39,8 +41,30 @@ npu_busy_time_us_show(struct device *dev, struct device_attribute *attr, char *b static DEVICE_ATTR_RO(npu_busy_time_us); +/** + * DOC: sched_mode + * + * The sched_mode is used to report current NPU scheduling mode. + * + * It returns following strings: + * - "HW" - Hardware Scheduler mode + * - "OS" - Operating System Scheduler mode + * + */ +static ssize_t +sched_mode_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct drm_device *drm = dev_get_drvdata(dev); + struct ivpu_device *vdev = to_ivpu_device(drm); + + return sysfs_emit(buf, "%s\n", vdev->fw->sched_mode ? "HW" : "OS"); +} + +static DEVICE_ATTR_RO(sched_mode); + static struct attribute *ivpu_dev_attrs[] = { &dev_attr_npu_busy_time_us.attr, + &dev_attr_sched_mode.attr, NULL, }; diff --git a/drivers/accel/ivpu/ivpu_trace.h b/drivers/accel/ivpu/ivpu_trace.h new file mode 100644 index 000000000000..eb792038e701 --- /dev/null +++ b/drivers/accel/ivpu/ivpu_trace.h @@ -0,0 +1,73 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2020-2024 Intel Corporation + */ + +#if !defined(__IVPU_TRACE_H__) || defined(TRACE_HEADER_MULTI_READ) +#define __IVPU_TRACE_H__ + +#include +#include "ivpu_drv.h" +#include "ivpu_job.h" +#include "vpu_jsm_api.h" +#include "ivpu_jsm_msg.h" +#include "ivpu_ipc.h" + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM vpu +#define TRACE_INCLUDE_FILE ivpu_trace + +TRACE_EVENT(pm, + TP_PROTO(const char *event), + TP_ARGS(event), + TP_STRUCT__entry(__field(const char *, event)), + TP_fast_assign(__entry->event = event;), + TP_printk("%s", __entry->event) +); + +TRACE_EVENT(job, + TP_PROTO(const char *event, struct ivpu_job *job), + TP_ARGS(event, job), + TP_STRUCT__entry(__field(const char *, event) + __field(u32, ctx_id) + __field(u32, engine_id) + __field(u32, job_id) + ), + TP_fast_assign(__entry->event = event; + __entry->ctx_id = job->file_priv->ctx.id; + __entry->engine_id = job->engine_idx; + __entry->job_id = job->job_id;), + TP_printk("%s context:%d engine:%d job:%d", + __entry->event, + __entry->ctx_id, + __entry->engine_id, + __entry->job_id) +); + +TRACE_EVENT(jsm, + TP_PROTO(const char *event, struct vpu_jsm_msg *msg), + TP_ARGS(event, msg), + TP_STRUCT__entry(__field(const char *, event) + __field(const char *, type) + __field(enum vpu_ipc_msg_status, status) + __field(u32, request_id) + __field(u32, result) + ), + TP_fast_assign(__entry->event = event; + __entry->type = ivpu_jsm_msg_type_to_str(msg->type); + __entry->status = msg->status; + __entry->request_id = msg->request_id; + __entry->result = msg->result;), + TP_printk("%s type:%s, status:%#x, id:%#x, result:%#x", + __entry->event, + __entry->type, + __entry->status, + __entry->request_id, + __entry->result) +); + +#endif /* __IVPU_TRACE_H__ */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#include diff --git a/drivers/accel/ivpu/ivpu_trace_points.c b/drivers/accel/ivpu/ivpu_trace_points.c new file mode 100644 index 000000000000..f8fb99de0de3 --- /dev/null +++ b/drivers/accel/ivpu/ivpu_trace_points.c @@ -0,0 +1,9 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2020-2024 Intel Corporation + */ + +#ifndef __CHECKER__ +#define CREATE_TRACE_POINTS +#include "ivpu_trace.h" +#endif diff --git a/drivers/accel/ivpu/vpu_boot_api.h b/drivers/accel/ivpu/vpu_boot_api.h index 82954b91b748..908e68ea1c39 100644 --- a/drivers/accel/ivpu/vpu_boot_api.h +++ b/drivers/accel/ivpu/vpu_boot_api.h @@ -1,14 +1,13 @@ /* SPDX-License-Identifier: MIT */ /* - * Copyright (c) 2020-2023, Intel Corporation. + * Copyright (c) 2020-2024, Intel Corporation. */ #ifndef VPU_BOOT_API_H #define VPU_BOOT_API_H /* - * =========== FW API version information beginning ================ - * The bellow values will be used to construct the version info this way: + * The below values will be used to construct the version info this way: * fw_bin_header->api_version[VPU_BOOT_API_VER_ID] = (VPU_BOOT_API_VER_MAJOR << 16) | * VPU_BOOT_API_VER_MINOR; * VPU_BOOT_API_VER_PATCH will be ignored. KMD and compatibility is not affected if this changes @@ -27,19 +26,18 @@ * Minor version changes when API backward compatibility is preserved. * Resets to 0 if Major version is incremented. */ -#define VPU_BOOT_API_VER_MINOR 24 +#define VPU_BOOT_API_VER_MINOR 26 /* * API header changed (field names, documentation, formatting) but API itself has not been changed */ -#define VPU_BOOT_API_VER_PATCH 0 +#define VPU_BOOT_API_VER_PATCH 3 /* * Index in the API version table * Must be unique for each API */ #define VPU_BOOT_API_VER_INDEX 0 -/* ------------ FW API version information end ---------------------*/ #pragma pack(push, 4) @@ -164,8 +162,6 @@ enum vpu_trace_destination { /* VPU 30xx HW component IDs are sequential, so define first and last IDs. */ #define VPU_TRACE_PROC_BIT_30XX_FIRST VPU_TRACE_PROC_BIT_LRT #define VPU_TRACE_PROC_BIT_30XX_LAST VPU_TRACE_PROC_BIT_SHV_15 -#define VPU_TRACE_PROC_BIT_KMB_FIRST VPU_TRACE_PROC_BIT_30XX_FIRST -#define VPU_TRACE_PROC_BIT_KMB_LAST VPU_TRACE_PROC_BIT_30XX_LAST struct vpu_boot_l2_cache_config { u8 use; @@ -199,6 +195,17 @@ struct vpu_warm_boot_section { */ #define POWER_PROFILE_SURVIVABILITY 0x1 +/** + * Enum for dvfs_mode boot param. + */ +enum vpu_governor { + VPU_GOV_DEFAULT = 0, /* Default Governor for the system */ + VPU_GOV_MAX_PERFORMANCE = 1, /* Maximum performance governor */ + VPU_GOV_ON_DEMAND = 2, /* On Demand frequency control governor */ + VPU_GOV_POWER_SAVE = 3, /* Power save governor */ + VPU_GOV_ON_DEMAND_PRIORITY_AWARE = 4 /* On Demand priority based governor */ +}; + struct vpu_boot_params { u32 magic; u32 vpu_id; @@ -301,7 +308,14 @@ struct vpu_boot_params { u32 temp_sensor_period_ms; /** PLL ratio for efficient clock frequency */ u32 pn_freq_pll_ratio; - /** DVFS Mode: Default: 0, Max Performance: 1, On Demand: 2, Power Save: 3 */ + /** + * DVFS Mode: + * 0 - Default, DVFS mode selected by the firmware + * 1 - Max Performance + * 2 - On Demand + * 3 - Power Save + * 4 - On Demand Priority Aware + */ u32 dvfs_mode; /** * Depending on DVFS Mode: @@ -332,8 +346,8 @@ struct vpu_boot_params { u64 d0i3_entry_vpu_ts; /* * The system time of the host operating system in microseconds. - * E.g the number of microseconds since 1st of January 1970, or whatever date the - * host operating system uses to maintain system time. + * E.g the number of microseconds since 1st of January 1970, or whatever + * date the host operating system uses to maintain system time. * This value will be used to track system time on the VPU. * The KMD is required to update this value on every VPU reset. */ @@ -382,10 +396,7 @@ struct vpu_boot_params { u32 pad6[734]; }; -/* - * Magic numbers set between host and vpu to detect corruptio of tracing init - */ - +/* Magic numbers set between host and vpu to detect corruption of tracing init */ #define VPU_TRACING_BUFFER_CANARY (0xCAFECAFE) /* Tracing buffer message format definitions */ @@ -405,7 +416,9 @@ struct vpu_tracing_buffer_header { u32 host_canary_start; /* offset from start of buffer for trace entries */ u32 read_index; - u32 pad_to_cache_line_size_0[14]; + /* keeps track of wrapping on the reader side */ + u32 read_wrap_count; + u32 pad_to_cache_line_size_0[13]; /* End of first cache line */ /** diff --git a/drivers/accel/ivpu/vpu_jsm_api.h b/drivers/accel/ivpu/vpu_jsm_api.h index 33f462b1a25d..7215c144158c 100644 --- a/drivers/accel/ivpu/vpu_jsm_api.h +++ b/drivers/accel/ivpu/vpu_jsm_api.h @@ -22,7 +22,7 @@ /* * Minor version changes when API backward compatibility is preserved. */ -#define VPU_JSM_API_VER_MINOR 16 +#define VPU_JSM_API_VER_MINOR 25 /* * API header changed (field names, documentation, formatting) but API itself has not been changed @@ -36,7 +36,7 @@ /* * Number of Priority Bands for Hardware Scheduling - * Bands: RealTime, Focus, Normal, Idle + * Bands: Idle(0), Normal(1), Focus(2), RealTime(3) */ #define VPU_HWS_NUM_PRIORITY_BANDS 4 @@ -74,6 +74,7 @@ #define VPU_JSM_STATUS_MVNCI_INTERNAL_ERROR 0xCU /* Job status returned when the job was preempted mid-inference */ #define VPU_JSM_STATUS_PREEMPTED_MID_INFERENCE 0xDU +#define VPU_JSM_STATUS_MVNCI_CONTEXT_VIOLATION_HW 0xEU /* * Host <-> VPU IPC channels. @@ -86,18 +87,58 @@ /* * Job flags bit masks. */ -#define VPU_JOB_FLAGS_NULL_SUBMISSION_MASK 0x00000001 -#define VPU_JOB_FLAGS_PRIVATE_DATA_MASK 0xFF000000 +enum { + /* + * Null submission mask. + * When set, batch buffer's commands are not processed but returned as + * successful immediately, except fences and timestamps. + * When cleared, batch buffer's commands are processed normally. + * Used for testing and profiling purposes. + */ + VPU_JOB_FLAGS_NULL_SUBMISSION_MASK = (1 << 0U), + /* + * Inline command mask. + * When set, the object in job queue is an inline command (see struct vpu_inline_cmd below). + * When cleared, the object in job queue is a job (see struct vpu_job_queue_entry below). + */ + VPU_JOB_FLAGS_INLINE_CMD_MASK = (1 << 1U), + /* + * VPU private data mask. + * Reserved for the VPU to store private data about the job (or inline command) + * while being processed. + */ + VPU_JOB_FLAGS_PRIVATE_DATA_MASK = 0xFFFF0000U +}; /* - * Sizes of the reserved areas in jobs, in bytes. + * Job queue flags bit masks. */ -#define VPU_JOB_RESERVED_BYTES 8 +enum { + /* + * No job done notification mask. + * When set, indicates that no job done notification should be sent for any + * job from this queue. When cleared, indicates that job done notification + * should be sent for every job completed from this queue. + */ + VPU_JOB_QUEUE_FLAGS_NO_JOB_DONE_MASK = (1 << 0U), + /* + * Native fence usage mask. + * When set, indicates that job queue uses native fences (as inline commands + * in job queue). Such queues may also use legacy fences (as commands in batch buffers). + * When cleared, indicates the job queue only uses legacy fences. + * NOTE: For queues using native fences, VPU expects that all jobs in the queue + * are immediately followed by an inline command object. This object is expected + * to be a fence signal command in most cases, but can also be a NOP in case the host + * does not need per-job fence signalling. Other inline commands objects can be + * inserted between "job and inline command" pairs. + */ + VPU_JOB_QUEUE_FLAGS_USE_NATIVE_FENCE_MASK = (1 << 1U), -/* - * Sizes of the reserved areas in job queues, in bytes. - */ -#define VPU_JOB_QUEUE_RESERVED_BYTES 52 + /* + * Enable turbo mode for testing NPU performance; not recommended for regular usage. + */ + VPU_JOB_QUEUE_FLAGS_TURBO_MODE = (1 << 2U) +}; /* * Max length (including trailing NULL char) of trace entity name (e.g., the @@ -140,24 +181,113 @@ */ #define VPU_HWS_INVALID_CMDQ_HANDLE 0ULL +/* + * Inline commands types. + */ +/* + * NOP. + * VPU does nothing other than consuming the inline command object. + */ +#define VPU_INLINE_CMD_TYPE_NOP 0x0 +/* + * Fence wait. + * VPU waits for the fence current value to reach monitored value. + * Fence wait operations are executed upon job dispatching. While waiting for + * the fence to be satisfied, VPU blocks fetching of the next objects in the queue. + * Jobs present in the queue prior to the fence wait object may be processed + * concurrently. + */ +#define VPU_INLINE_CMD_TYPE_FENCE_WAIT 0x1 +/* + * Fence signal. + * VPU sets the fence current value to the provided value. If new current value + * is equal to or higher than monitored value, VPU sends fence signalled notification + * to the host. Fence signal operations are executed upon completion of all the jobs + * present in the queue prior to them, and in-order relative to each other in the queue. + * But jobs in-between them may be processed concurrently and may complete out-of-order. + */ +#define VPU_INLINE_CMD_TYPE_FENCE_SIGNAL 0x2 + +/* + * Job scheduling priority bands for both hardware scheduling and OS scheduling. + */ +enum vpu_job_scheduling_priority_band { + VPU_JOB_SCHEDULING_PRIORITY_BAND_IDLE = 0, + VPU_JOB_SCHEDULING_PRIORITY_BAND_NORMAL = 1, + VPU_JOB_SCHEDULING_PRIORITY_BAND_FOCUS = 2, + VPU_JOB_SCHEDULING_PRIORITY_BAND_REALTIME = 3, + VPU_JOB_SCHEDULING_PRIORITY_BAND_COUNT = 4, +}; + /* * Job format. + * Jobs defines the actual workloads to be executed by a given engine. */ struct vpu_job_queue_entry { - u64 batch_buf_addr; /**< Address of VPU commands batch buffer */ - u32 job_id; /**< Job ID */ - u32 flags; /**< Flags bit field, see VPU_JOB_FLAGS_* above */ - u64 root_page_table_addr; /**< Address of root page table to use for this job */ - u64 root_page_table_update_counter; /**< Page tables update events counter */ - u64 primary_preempt_buf_addr; + /**< Address of VPU commands batch buffer */ + u64 batch_buf_addr; + /**< Job ID */ + u32 job_id; + /**< Flags bit field, see VPU_JOB_FLAGS_* above */ + u32 flags; + /** + * Doorbell ring timestamp taken by KMD from SoC's global system clock, in + * microseconds. NPU can convert this value to its own fixed clock's timebase, + * to match other profiling timestamps. + */ + u64 doorbell_timestamp; + /**< Extra id for job tracking, used only in the firmware perf traces */ + u64 host_tracking_id; /**< Address of the primary preemption buffer to use for this job */ - u32 primary_preempt_buf_size; + u64 primary_preempt_buf_addr; /**< Size of the primary preemption buffer to use for this job */ - u32 secondary_preempt_buf_size; + u32 primary_preempt_buf_size; /**< Size of secondary preemption buffer to use for this job */ - u64 secondary_preempt_buf_addr; + u32 secondary_preempt_buf_size; /**< Address of secondary preemption buffer to use for this job */ - u8 reserved_0[VPU_JOB_RESERVED_BYTES]; + u64 secondary_preempt_buf_addr; + u64 reserved_0; +}; + +/* + * Inline command format. + * Inline commands are the commands executed at scheduler level (typically, + * synchronization directives). Inline command and job objects must be of + * the same size and have flags field at same offset. + */ +struct vpu_inline_cmd { + u64 reserved_0; + /* Inline command type, see VPU_INLINE_CMD_TYPE_* defines. */ + u32 type; + /* Flags bit field, see VPU_JOB_FLAGS_* above. */ + u32 flags; + /* Inline command payload. Depends on inline command type. */ + union { + /* Fence (wait and signal) commands' payload. */ + struct { + /* Fence object handle. */ + u64 fence_handle; + /* User VA of the current fence value. */ + u64 current_value_va; + /* User VA of the monitored fence value (read-only). */ + u64 monitored_value_va; + /* Value to wait for or write in fence location. */ + u64 value; + /* User VA of the log buffer in which to add log entry on completion. */ + u64 log_buffer_va; + } fence; + /* Other commands do not have a payload. */ + /* Payload definition for future inline commands can be inserted here. */ + u64 reserved_1[6]; + } payload; +}; + +/* + * Job queue slots can be populated either with job objects or inline command objects. + */ +union vpu_jobq_slot { + struct vpu_job_queue_entry job; + struct vpu_inline_cmd inline_cmd; }; /* @@ -167,7 +297,21 @@ struct vpu_job_queue_header { u32 engine_idx; u32 head; u32 tail; - u8 reserved_0[VPU_JOB_QUEUE_RESERVED_BYTES]; + u32 flags; + /* Set to 1 to indicate priority_band field is valid */ + u32 priority_band_valid; + /* + * Priority for the work of this job queue, valid only if the HWS is NOT used + * and the `priority_band_valid` is set to 1. It is applied only during + * the VPU_JSM_MSG_REGISTER_DB message processing. + * The device firmware might use the `priority_band` to optimize the power + * management logic, but it will not affect the order of jobs. + * Available priority bands: @see enum vpu_job_scheduling_priority_band + */ + u32 priority_band; + /* Inside realtime band assigns a further priority, limited to 0..31 range */ + u32 realtime_priority_level; + u32 reserved_0[9]; }; /* @@ -175,7 +319,7 @@ struct vpu_job_queue_header { */ struct vpu_job_queue { struct vpu_job_queue_header header; - struct vpu_job_queue_entry job[]; + union vpu_jobq_slot slot[]; }; /** @@ -197,9 +341,7 @@ enum vpu_trace_entity_type { struct vpu_hws_log_buffer_header { /* Written by VPU after adding a log entry. Initialised by host to 0. */ u32 first_free_entry_index; - /* Incremented by VPU every time the VPU overwrites the 0th entry; - * initialised by host to 0. - */ + /* Incremented by VPU every time the VPU writes the 0th entry; initialised by host to 0. */ u32 wraparound_count; /* * This is the number of buffers that can be stored in the log buffer provided by the host. @@ -230,14 +372,80 @@ struct vpu_hws_log_buffer_entry { u64 operation_data[2]; }; +/* Native fence log buffer types. */ +enum vpu_hws_native_fence_log_type { + VPU_HWS_NATIVE_FENCE_LOG_TYPE_WAITS = 1, + VPU_HWS_NATIVE_FENCE_LOG_TYPE_SIGNALS = 2 +}; + +/* HWS native fence log buffer header. */ +struct vpu_hws_native_fence_log_header { + union { + struct { + /* Index of the first free entry in buffer. */ + u32 first_free_entry_idx; + /* Incremented each time NPU wraps around the buffer to write next entry. */ + u32 wraparound_count; + }; + /* Field allowing atomic update of both fields above. */ + u64 atomic_wraparound_and_entry_idx; + }; + /* Log buffer type, see enum vpu_hws_native_fence_log_type. */ + u64 type; + /* Allocated number of entries in the log buffer. */ + u64 entry_nb; + u64 reserved[2]; +}; + +/* Native fence log operation types. */ +enum vpu_hws_native_fence_log_op { + VPU_HWS_NATIVE_FENCE_LOG_OP_SIGNAL_EXECUTED = 0, + VPU_HWS_NATIVE_FENCE_LOG_OP_WAIT_UNBLOCKED = 1 +}; + +/* HWS native fence log entry. */ +struct vpu_hws_native_fence_log_entry { + /* Newly signaled/unblocked fence value. */ + u64 fence_value; + /* Native fence object handle to which this operation belongs. */ + u64 fence_handle; + /* Operation type, see enum vpu_hws_native_fence_log_op. */ + u64 op_type; + u64 reserved_0; + /* + * VPU_HWS_NATIVE_FENCE_LOG_OP_WAIT_UNBLOCKED only: Timestamp at which fence + * wait was started (in NPU SysTime). + */ + u64 fence_wait_start_ts; + u64 reserved_1; + /* Timestamp at which fence operation was completed (in NPU SysTime). */ + u64 fence_end_ts; +}; + +/* Native fence log buffer. */ +struct vpu_hws_native_fence_log_buffer { + struct vpu_hws_native_fence_log_header header; + struct vpu_hws_native_fence_log_entry entry[]; +}; + /* * Host <-> VPU IPC messages types. */ enum vpu_ipc_msg_type { VPU_JSM_MSG_UNKNOWN = 0xFFFFFFFF, + /* IPC Host -> Device, Async commands */ VPU_JSM_MSG_ASYNC_CMD = 0x1100, VPU_JSM_MSG_ENGINE_RESET = VPU_JSM_MSG_ASYNC_CMD, + /** + * Preempt engine. The NPU stops (preempts) all the jobs currently + * executing on the target engine making the engine become idle and ready to + * execute new jobs. + * NOTE: The NPU does not remove unstarted jobs (if any) from job queues of + * the target engine, but it stops processing them (until the queue doorbell + * is rung again); the host is responsible to reset the job queue, either + * after preemption or when resubmitting jobs to the queue. + */ VPU_JSM_MSG_ENGINE_PREEMPT = 0x1101, VPU_JSM_MSG_REGISTER_DB = 0x1102, VPU_JSM_MSG_UNREGISTER_DB = 0x1103, @@ -323,9 +531,10 @@ enum vpu_ipc_msg_type { * NOTE: Please introduce new ASYNC commands before this one. * */ VPU_JSM_MSG_STATE_DUMP = 0x11FF, + /* IPC Host -> Device, General commands */ VPU_JSM_MSG_GENERAL_CMD = 0x1200, - VPU_JSM_MSG_BLOB_DEINIT = VPU_JSM_MSG_GENERAL_CMD, + VPU_JSM_MSG_BLOB_DEINIT_DEPRECATED = VPU_JSM_MSG_GENERAL_CMD, /** * Control dyndbg behavior by executing a dyndbg command; equivalent to * Linux command: `echo '' > /dynamic_debug/control`. @@ -335,8 +544,12 @@ enum vpu_ipc_msg_type { * Perform the save procedure for the D0i3 entry */ VPU_JSM_MSG_PWR_D0I3_ENTER = 0x1202, + /* IPC Device -> Host, Job completion */ VPU_JSM_MSG_JOB_DONE = 0x2100, + /* IPC Device -> Host, Fence signalled */ + VPU_JSM_MSG_NATIVE_FENCE_SIGNALLED = 0x2101, + /* IPC Device -> Host, Async command completion */ VPU_JSM_MSG_ASYNC_CMD_DONE = 0x2200, VPU_JSM_MSG_ENGINE_RESET_DONE = VPU_JSM_MSG_ASYNC_CMD_DONE, @@ -422,6 +635,7 @@ enum vpu_ipc_msg_type { * NOTE: Please introduce new ASYNC responses before this one. * */ VPU_JSM_MSG_STATE_DUMP_RSP = 0x22FF, + /* IPC Device -> Host, General command completion */ VPU_JSM_MSG_GENERAL_CMD_DONE = 0x2300, VPU_JSM_MSG_BLOB_DEINIT_DONE = VPU_JSM_MSG_GENERAL_CMD_DONE, @@ -600,11 +814,6 @@ struct vpu_jsm_metric_streamer_update { u64 next_buffer_size; }; -struct vpu_ipc_msg_payload_blob_deinit { - /* 64-bit unique ID for the blob to be de-initialized. */ - u64 blob_id; -}; - struct vpu_ipc_msg_payload_job_done { /* Engine to which the job was submitted. */ u32 engine_idx; @@ -622,6 +831,21 @@ struct vpu_ipc_msg_payload_job_done { u64 cmdq_id; }; +/* + * Notification message upon native fence signalling. + * @see VPU_JSM_MSG_NATIVE_FENCE_SIGNALLED + */ +struct vpu_ipc_msg_payload_native_fence_signalled { + /* Engine ID. */ + u32 engine_idx; + /* Host SSID. */ + u32 host_ssid; + /* CMDQ ID */ + u64 cmdq_id; + /* Fence object handle. */ + u64 fence_handle; +}; + struct vpu_jsm_engine_reset_context { /* Host SSID */ u32 host_ssid; @@ -700,11 +924,6 @@ struct vpu_ipc_msg_payload_get_power_level_count_done { u8 power_limit[16]; }; -struct vpu_ipc_msg_payload_blob_deinit_done { - /* 64-bit unique ID for the blob de-initialized. */ - u64 blob_id; -}; - /* HWS priority band setup request / response */ struct vpu_ipc_msg_payload_hws_priority_band_setup { /* @@ -794,7 +1013,10 @@ struct vpu_ipc_msg_payload_hws_set_context_sched_properties { u32 reserved_0; /* Command queue id */ u64 cmdq_id; - /* Priority band to assign to work of this context */ + /* + * Priority band to assign to work of this context. + * Available priority bands: @see enum vpu_job_scheduling_priority_band + */ u32 priority_band; /* Inside realtime band assigns a further priority */ u32 realtime_priority_level; @@ -869,9 +1091,7 @@ struct vpu_ipc_msg_payload_hws_set_scheduling_log { */ u64 notify_index; /* - * Enable extra events to be output to log for debug of scheduling algorithm. - * Interpreted by VPU as a boolean to enable or disable, expected values are - * 0 and 1. + * Field is now deprecated, will be removed when KMD is updated to support removal */ u32 enable_extra_events; /* Zero Padding */ @@ -1243,10 +1463,10 @@ union vpu_ipc_msg_payload { struct vpu_jsm_metric_streamer_start metric_streamer_start; struct vpu_jsm_metric_streamer_stop metric_streamer_stop; struct vpu_jsm_metric_streamer_update metric_streamer_update; - struct vpu_ipc_msg_payload_blob_deinit blob_deinit; struct vpu_ipc_msg_payload_ssid_release ssid_release; struct vpu_jsm_hws_register_db hws_register_db; struct vpu_ipc_msg_payload_job_done job_done; + struct vpu_ipc_msg_payload_native_fence_signalled native_fence_signalled; struct vpu_ipc_msg_payload_engine_reset_done engine_reset_done; struct vpu_ipc_msg_payload_engine_preempt_done engine_preempt_done; struct vpu_ipc_msg_payload_register_db_done register_db_done; @@ -1254,7 +1474,6 @@ union vpu_ipc_msg_payload { struct vpu_ipc_msg_payload_query_engine_hb_done query_engine_hb_done; struct vpu_ipc_msg_payload_get_power_level_count_done get_power_level_count_done; struct vpu_jsm_metric_streamer_done metric_streamer_done; - struct vpu_ipc_msg_payload_blob_deinit_done blob_deinit_done; struct vpu_ipc_msg_payload_trace_config trace_config; struct vpu_ipc_msg_payload_trace_capability_rsp trace_capability; struct vpu_ipc_msg_payload_trace_get_name trace_get_name; diff --git a/drivers/accel/qaic/mhi_controller.c b/drivers/accel/qaic/mhi_controller.c index ada9b1eb0787..8ab82e78dd94 100644 --- a/drivers/accel/qaic/mhi_controller.c +++ b/drivers/accel/qaic/mhi_controller.c @@ -405,6 +405,38 @@ static const struct mhi_channel_config aic100_channels[] = { .auto_queue = false, .wake_capable = false, }, + { + .name = "IPCR", + .num = 24, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_TO_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = false, + .wake_capable = false, + }, + { + .name = "IPCR", + .num = 25, + .num_elements = 32, + .local_elements = 0, + .event_ring = 0, + .dir = DMA_FROM_DEVICE, + .ee_mask = MHI_CH_EE_AMSS, + .pollcfg = 0, + .doorbell = MHI_DB_BRST_DISABLE, + .lpm_notify = false, + .offload_channel = false, + .doorbell_mode_switch = false, + .auto_queue = true, + .wake_capable = false, + }, }; static struct mhi_event_config aic100_events[] = { diff --git a/drivers/accel/qaic/qaic_control.c b/drivers/accel/qaic/qaic_control.c index 9e8a8cbadf6b..d8bdab69f800 100644 --- a/drivers/accel/qaic/qaic_control.c +++ b/drivers/accel/qaic/qaic_control.c @@ -496,7 +496,7 @@ static int encode_addr_size_pairs(struct dma_xfer *xfer, struct wrapper_list *wr nents = sgt->nents; nents_dma = nents; *size = QAIC_MANAGE_EXT_MSG_LENGTH - msg_hdr_len - sizeof(**out_trans); - for_each_sgtable_sg(sgt, sg, i) { + for_each_sgtable_dma_sg(sgt, sg, i) { *size -= sizeof(*asp); /* Save 1K for possible follow-up transactions. */ if (*size < SZ_1K) { diff --git a/drivers/accel/qaic/qaic_data.c b/drivers/accel/qaic/qaic_data.c index e86e71c1cdd8..c20eb63750f5 100644 --- a/drivers/accel/qaic/qaic_data.c +++ b/drivers/accel/qaic/qaic_data.c @@ -184,7 +184,7 @@ static int clone_range_of_sgt_for_slice(struct qaic_device *qdev, struct sg_tabl nents = 0; size = size ? size : PAGE_SIZE; - for (sg = sgt_in->sgl; sg; sg = sg_next(sg)) { + for_each_sgtable_dma_sg(sgt_in, sg, j) { len = sg_dma_len(sg); if (!len) @@ -221,7 +221,7 @@ static int clone_range_of_sgt_for_slice(struct qaic_device *qdev, struct sg_tabl /* copy relevant sg node and fix page and length */ sgn = sgf; - for_each_sgtable_sg(sgt, sg, j) { + for_each_sgtable_dma_sg(sgt, sg, j) { memcpy(sg, sgn, sizeof(*sg)); if (sgn == sgf) { sg_dma_address(sg) += offf; @@ -301,7 +301,7 @@ static int encode_reqs(struct qaic_device *qdev, struct bo_slice *slice, * fence. */ dev_addr = req->dev_addr; - for_each_sgtable_sg(slice->sgt, sg, i) { + for_each_sgtable_dma_sg(slice->sgt, sg, i) { slice->reqs[i].cmd = cmd; slice->reqs[i].src_addr = cpu_to_le64(slice->dir == DMA_TO_DEVICE ? sg_dma_address(sg) : dev_addr); diff --git a/drivers/accel/qaic/qaic_debugfs.c b/drivers/accel/qaic/qaic_debugfs.c index 20b653d99e52..ba0cf2f94732 100644 --- a/drivers/accel/qaic/qaic_debugfs.c +++ b/drivers/accel/qaic/qaic_debugfs.c @@ -64,20 +64,9 @@ static int bootlog_show(struct seq_file *s, void *unused) return 0; } -static int bootlog_fops_open(struct inode *inode, struct file *file) -{ - return single_open(file, bootlog_show, inode->i_private); -} +DEFINE_SHOW_ATTRIBUTE(bootlog); -static const struct file_operations bootlog_fops = { - .owner = THIS_MODULE, - .open = bootlog_fops_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static int read_dbc_fifo_size(struct seq_file *s, void *unused) +static int fifo_size_show(struct seq_file *s, void *unused) { struct dma_bridge_chan *dbc = s->private; @@ -85,20 +74,9 @@ static int read_dbc_fifo_size(struct seq_file *s, void *unused) return 0; } -static int fifo_size_open(struct inode *inode, struct file *file) -{ - return single_open(file, read_dbc_fifo_size, inode->i_private); -} +DEFINE_SHOW_ATTRIBUTE(fifo_size); -static const struct file_operations fifo_size_fops = { - .owner = THIS_MODULE, - .open = fifo_size_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -static int read_dbc_queued(struct seq_file *s, void *unused) +static int queued_show(struct seq_file *s, void *unused) { struct dma_bridge_chan *dbc = s->private; u32 tail = 0, head = 0; @@ -115,18 +93,7 @@ static int read_dbc_queued(struct seq_file *s, void *unused) return 0; } -static int queued_open(struct inode *inode, struct file *file) -{ - return single_open(file, read_dbc_queued, inode->i_private); -} - -static const struct file_operations queued_fops = { - .owner = THIS_MODULE, - .open = queued_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; +DEFINE_SHOW_ATTRIBUTE(queued); void qaic_debugfs_init(struct qaic_drm_device *qddev) { diff --git a/drivers/accel/qaic/qaic_drv.c b/drivers/accel/qaic/qaic_drv.c index bf10156c334e..3575e0c984d6 100644 --- a/drivers/accel/qaic/qaic_drv.c +++ b/drivers/accel/qaic/qaic_drv.c @@ -34,6 +34,7 @@ MODULE_IMPORT_NS(DMA_BUF); +#define PCI_DEV_AIC080 0xa080 #define PCI_DEV_AIC100 0xa100 #define QAIC_NAME "qaic" #define QAIC_DESC "Qualcomm Cloud AI Accelerators" @@ -53,12 +54,12 @@ static void qaicm_wq_release(struct drm_device *dev, void *res) destroy_workqueue(wq); } -static struct workqueue_struct *qaicm_wq_init(struct drm_device *dev, const char *fmt) +static struct workqueue_struct *qaicm_wq_init(struct drm_device *dev, const char *name) { struct workqueue_struct *wq; int ret; - wq = alloc_workqueue(fmt, WQ_UNBOUND, 0); + wq = alloc_workqueue("%s", WQ_UNBOUND, 0, name); if (!wq) return ERR_PTR(-ENOMEM); ret = drmm_add_action_or_reset(dev, qaicm_wq_release, wq); @@ -365,7 +366,7 @@ static struct qaic_device *create_qdev(struct pci_dev *pdev, const struct pci_de return NULL; qdev->dev_state = QAIC_OFFLINE; - if (id->device == PCI_DEV_AIC100) { + if (id->device == PCI_DEV_AIC080 || id->device == PCI_DEV_AIC100) { qdev->num_dbc = 16; qdev->dbc = devm_kcalloc(dev, qdev->num_dbc, sizeof(*qdev->dbc), GFP_KERNEL); if (!qdev->dbc) @@ -607,6 +608,7 @@ static struct mhi_driver qaic_mhi_driver = { }; static const struct pci_device_id qaic_ids[] = { + { PCI_DEVICE(PCI_VENDOR_ID_QCOM, PCI_DEV_AIC080), }, { PCI_DEVICE(PCI_VENDOR_ID_QCOM, PCI_DEV_AIC100), }, { } }; diff --git a/drivers/accel/qaic/sahara.c b/drivers/accel/qaic/sahara.c index bf94bbab6be5..6d772143d612 100644 --- a/drivers/accel/qaic/sahara.c +++ b/drivers/accel/qaic/sahara.c @@ -2,6 +2,7 @@ /* Copyright (c) 2024 Qualcomm Innovation Center, Inc. All rights reserved. */ +#include #include #include #include @@ -9,6 +10,7 @@ #include #include #include +#include #include #include "sahara.h" @@ -36,12 +38,14 @@ #define SAHARA_PACKET_MAX_SIZE 0xffffU /* MHI_MAX_MTU */ #define SAHARA_TRANSFER_MAX_SIZE 0x80000 +#define SAHARA_READ_MAX_SIZE 0xfff0U /* Avoid unaligned requests */ #define SAHARA_NUM_TX_BUF DIV_ROUND_UP(SAHARA_TRANSFER_MAX_SIZE,\ SAHARA_PACKET_MAX_SIZE) #define SAHARA_IMAGE_ID_NONE U32_MAX #define SAHARA_VERSION 2 #define SAHARA_SUCCESS 0 +#define SAHARA_TABLE_ENTRY_STR_LEN 20 #define SAHARA_MODE_IMAGE_TX_PENDING 0x0 #define SAHARA_MODE_IMAGE_TX_COMPLETE 0x1 @@ -53,6 +57,8 @@ #define SAHARA_END_OF_IMAGE_LENGTH 0x10 #define SAHARA_DONE_LENGTH 0x8 #define SAHARA_RESET_LENGTH 0x8 +#define SAHARA_MEM_DEBUG64_LENGTH 0x18 +#define SAHARA_MEM_READ64_LENGTH 0x18 struct sahara_packet { __le32 cmd; @@ -80,18 +86,95 @@ struct sahara_packet { __le32 image; __le32 status; } end_of_image; + struct { + __le64 table_address; + __le64 table_length; + } memory_debug64; + struct { + __le64 memory_address; + __le64 memory_length; + } memory_read64; }; }; +struct sahara_debug_table_entry64 { + __le64 type; + __le64 address; + __le64 length; + char description[SAHARA_TABLE_ENTRY_STR_LEN]; + char filename[SAHARA_TABLE_ENTRY_STR_LEN]; +}; + +struct sahara_dump_table_entry { + u64 type; + u64 address; + u64 length; + char description[SAHARA_TABLE_ENTRY_STR_LEN]; + char filename[SAHARA_TABLE_ENTRY_STR_LEN]; +}; + +#define SAHARA_DUMP_V1_MAGIC 0x1234567890abcdef +#define SAHARA_DUMP_V1_VER 1 +struct sahara_memory_dump_meta_v1 { + u64 magic; + u64 version; + u64 dump_size; + u64 table_size; +}; + +/* + * Layout of crashdump provided to user via devcoredump + * +------------------------------------------+ + * | Crashdump Meta structure | + * | type: struct sahara_memory_dump_meta_v1 | + * +------------------------------------------+ + * | Crashdump Table | + * | type: array of struct | + * | sahara_dump_table_entry | + * | | + * | | + * +------------------------------------------+ + * | Crashdump | + * | | + * | | + * | | + * | | + * | | + * +------------------------------------------+ + * + * First is the metadata header. Userspace can use the magic number to verify + * the content type, and then check the version for the rest of the format. + * New versions should keep the magic number location/value, and version + * location, but increment the version value. + * + * For v1, the metadata lists the size of the entire dump (header + table + + * dump) and the size of the table. Then the dump image table, which describes + * the contents of the dump. Finally all the images are listed in order, with + * no deadspace in between. Userspace can use the sizes listed in the image + * table to reconstruct the individual images. + */ + struct sahara_context { struct sahara_packet *tx[SAHARA_NUM_TX_BUF]; struct sahara_packet *rx; - struct work_struct work; + struct work_struct fw_work; + struct work_struct dump_work; struct mhi_device *mhi_dev; const char **image_table; u32 table_size; u32 active_image_id; const struct firmware *firmware; + u64 dump_table_address; + u64 dump_table_length; + size_t rx_size; + size_t rx_size_requested; + void *mem_dump; + size_t mem_dump_sz; + struct sahara_dump_table_entry *dump_image; + u64 dump_image_offset; + void *mem_dump_freespace; + u64 dump_images_left; + bool is_mem_dump_mode; }; static const char *aic100_image_table[] = { @@ -153,6 +236,8 @@ static void sahara_send_reset(struct sahara_context *context) { int ret; + context->is_mem_dump_mode = false; + context->tx[0]->cmd = cpu_to_le32(SAHARA_RESET_CMD); context->tx[0]->length = cpu_to_le32(SAHARA_RESET_LENGTH); @@ -186,7 +271,8 @@ static void sahara_hello(struct sahara_context *context) } if (le32_to_cpu(context->rx->hello.mode) != SAHARA_MODE_IMAGE_TX_PENDING && - le32_to_cpu(context->rx->hello.mode) != SAHARA_MODE_IMAGE_TX_COMPLETE) { + le32_to_cpu(context->rx->hello.mode) != SAHARA_MODE_IMAGE_TX_COMPLETE && + le32_to_cpu(context->rx->hello.mode) != SAHARA_MODE_MEMORY_DEBUG) { dev_err(&context->mhi_dev->dev, "Unsupported hello packet - mode %d\n", le32_to_cpu(context->rx->hello.mode)); return; @@ -320,9 +406,70 @@ static void sahara_end_of_image(struct sahara_context *context) dev_dbg(&context->mhi_dev->dev, "Unable to send done response %d\n", ret); } +static void sahara_memory_debug64(struct sahara_context *context) +{ + int ret; + + dev_dbg(&context->mhi_dev->dev, + "MEMORY DEBUG64 cmd received. length:%d table_address:%#llx table_length:%#llx\n", + le32_to_cpu(context->rx->length), + le64_to_cpu(context->rx->memory_debug64.table_address), + le64_to_cpu(context->rx->memory_debug64.table_length)); + + if (le32_to_cpu(context->rx->length) != SAHARA_MEM_DEBUG64_LENGTH) { + dev_err(&context->mhi_dev->dev, "Malformed memory debug64 packet - length %d\n", + le32_to_cpu(context->rx->length)); + return; + } + + context->dump_table_address = le64_to_cpu(context->rx->memory_debug64.table_address); + context->dump_table_length = le64_to_cpu(context->rx->memory_debug64.table_length); + + if (context->dump_table_length % sizeof(struct sahara_debug_table_entry64) != 0 || + !context->dump_table_length) { + dev_err(&context->mhi_dev->dev, "Malformed memory debug64 packet - table length %lld\n", + context->dump_table_length); + return; + } + + /* + * From this point, the protocol flips. We make memory_read requests to + * the device, and the device responds with the raw data. If the device + * has an error, it will send an End of Image command. First we need to + * request the memory dump table so that we know where all the pieces + * of the dump are that we can consume. + */ + + context->is_mem_dump_mode = true; + + /* + * Assume that the table is smaller than our MTU so that we can read it + * in one shot. The spec does not put an upper limit on the table, but + * no known device will exceed this. + */ + if (context->dump_table_length > SAHARA_PACKET_MAX_SIZE) { + dev_err(&context->mhi_dev->dev, "Memory dump table length %lld exceeds supported size. Discarding dump\n", + context->dump_table_length); + sahara_send_reset(context); + return; + } + + context->tx[0]->cmd = cpu_to_le32(SAHARA_MEM_READ64_CMD); + context->tx[0]->length = cpu_to_le32(SAHARA_MEM_READ64_LENGTH); + context->tx[0]->memory_read64.memory_address = cpu_to_le64(context->dump_table_address); + context->tx[0]->memory_read64.memory_length = cpu_to_le64(context->dump_table_length); + + context->rx_size_requested = context->dump_table_length; + + ret = mhi_queue_buf(context->mhi_dev, DMA_TO_DEVICE, context->tx[0], + SAHARA_MEM_READ64_LENGTH, MHI_EOT); + if (ret) + dev_err(&context->mhi_dev->dev, "Unable to send read for dump table %d\n", ret); +} + static void sahara_processing(struct work_struct *work) { - struct sahara_context *context = container_of(work, struct sahara_context, work); + struct sahara_context *context = container_of(work, struct sahara_context, fw_work); int ret; switch (le32_to_cpu(context->rx->cmd)) { @@ -338,6 +485,12 @@ static void sahara_processing(struct work_struct *work) case SAHARA_DONE_RESP_CMD: /* Intentional do nothing as we don't need to exit an app */ break; + case SAHARA_RESET_RESP_CMD: + /* Intentional do nothing as we don't need to exit an app */ + break; + case SAHARA_MEM_DEBUG64_CMD: + sahara_memory_debug64(context); + break; default: dev_err(&context->mhi_dev->dev, "Unknown command %d\n", le32_to_cpu(context->rx->cmd)); @@ -350,6 +503,217 @@ static void sahara_processing(struct work_struct *work) dev_err(&context->mhi_dev->dev, "Unable to requeue rx buf %d\n", ret); } +static void sahara_parse_dump_table(struct sahara_context *context) +{ + struct sahara_dump_table_entry *image_out_table; + struct sahara_debug_table_entry64 *dev_table; + struct sahara_memory_dump_meta_v1 *dump_meta; + u64 table_nents; + u64 dump_length; + int ret; + u64 i; + + table_nents = context->dump_table_length / sizeof(*dev_table); + context->dump_images_left = table_nents; + dump_length = 0; + + dev_table = (struct sahara_debug_table_entry64 *)(context->rx); + for (i = 0; i < table_nents; ++i) { + /* Do not trust the device, ensure the strings are terminated */ + dev_table[i].description[SAHARA_TABLE_ENTRY_STR_LEN - 1] = 0; + dev_table[i].filename[SAHARA_TABLE_ENTRY_STR_LEN - 1] = 0; + + dump_length = size_add(dump_length, le64_to_cpu(dev_table[i].length)); + if (dump_length == SIZE_MAX) { + /* Discard the dump */ + sahara_send_reset(context); + return; + } + + dev_dbg(&context->mhi_dev->dev, + "Memory dump table entry %lld type: %lld address: %#llx length: %#llx description: \"%s\" filename \"%s\"\n", + i, + le64_to_cpu(dev_table[i].type), + le64_to_cpu(dev_table[i].address), + le64_to_cpu(dev_table[i].length), + dev_table[i].description, + dev_table[i].filename); + } + + dump_length = size_add(dump_length, sizeof(*dump_meta)); + if (dump_length == SIZE_MAX) { + /* Discard the dump */ + sahara_send_reset(context); + return; + } + dump_length = size_add(dump_length, size_mul(sizeof(*image_out_table), table_nents)); + if (dump_length == SIZE_MAX) { + /* Discard the dump */ + sahara_send_reset(context); + return; + } + + context->mem_dump_sz = dump_length; + context->mem_dump = vzalloc(dump_length); + if (!context->mem_dump) { + /* Discard the dump */ + sahara_send_reset(context); + return; + } + + /* Populate the dump metadata and table for userspace */ + dump_meta = context->mem_dump; + dump_meta->magic = SAHARA_DUMP_V1_MAGIC; + dump_meta->version = SAHARA_DUMP_V1_VER; + dump_meta->dump_size = dump_length; + dump_meta->table_size = context->dump_table_length; + + image_out_table = context->mem_dump + sizeof(*dump_meta); + for (i = 0; i < table_nents; ++i) { + image_out_table[i].type = le64_to_cpu(dev_table[i].type); + image_out_table[i].address = le64_to_cpu(dev_table[i].address); + image_out_table[i].length = le64_to_cpu(dev_table[i].length); + strscpy(image_out_table[i].description, dev_table[i].description, + SAHARA_TABLE_ENTRY_STR_LEN); + strscpy(image_out_table[i].filename, + dev_table[i].filename, + SAHARA_TABLE_ENTRY_STR_LEN); + } + + context->mem_dump_freespace = &image_out_table[i]; + + /* Done parsing the table, switch to image dump mode */ + context->dump_table_length = 0; + + /* Request the first chunk of the first image */ + context->dump_image = &image_out_table[0]; + dump_length = min(context->dump_image->length, SAHARA_READ_MAX_SIZE); + /* Avoid requesting EOI sized data so that we can identify errors */ + if (dump_length == SAHARA_END_OF_IMAGE_LENGTH) + dump_length = SAHARA_END_OF_IMAGE_LENGTH / 2; + + context->dump_image_offset = dump_length; + + context->tx[0]->cmd = cpu_to_le32(SAHARA_MEM_READ64_CMD); + context->tx[0]->length = cpu_to_le32(SAHARA_MEM_READ64_LENGTH); + context->tx[0]->memory_read64.memory_address = cpu_to_le64(context->dump_image->address); + context->tx[0]->memory_read64.memory_length = cpu_to_le64(dump_length); + + context->rx_size_requested = dump_length; + + ret = mhi_queue_buf(context->mhi_dev, DMA_TO_DEVICE, context->tx[0], + SAHARA_MEM_READ64_LENGTH, MHI_EOT); + if (ret) + dev_err(&context->mhi_dev->dev, "Unable to send read for dump content %d\n", ret); +} + +static void sahara_parse_dump_image(struct sahara_context *context) +{ + u64 dump_length; + int ret; + + memcpy(context->mem_dump_freespace, context->rx, context->rx_size); + context->mem_dump_freespace += context->rx_size; + + if (context->dump_image_offset >= context->dump_image->length) { + /* Need to move to next image */ + context->dump_image++; + context->dump_images_left--; + context->dump_image_offset = 0; + + if (!context->dump_images_left) { + /* Dump done */ + dev_coredumpv(context->mhi_dev->mhi_cntrl->cntrl_dev, + context->mem_dump, + context->mem_dump_sz, + GFP_KERNEL); + context->mem_dump = NULL; + sahara_send_reset(context); + return; + } + } + + /* Get next image chunk */ + dump_length = context->dump_image->length - context->dump_image_offset; + dump_length = min(dump_length, SAHARA_READ_MAX_SIZE); + /* Avoid requesting EOI sized data so that we can identify errors */ + if (dump_length == SAHARA_END_OF_IMAGE_LENGTH) + dump_length = SAHARA_END_OF_IMAGE_LENGTH / 2; + + context->tx[0]->cmd = cpu_to_le32(SAHARA_MEM_READ64_CMD); + context->tx[0]->length = cpu_to_le32(SAHARA_MEM_READ64_LENGTH); + context->tx[0]->memory_read64.memory_address = + cpu_to_le64(context->dump_image->address + context->dump_image_offset); + context->tx[0]->memory_read64.memory_length = cpu_to_le64(dump_length); + + context->dump_image_offset += dump_length; + context->rx_size_requested = dump_length; + + ret = mhi_queue_buf(context->mhi_dev, DMA_TO_DEVICE, context->tx[0], + SAHARA_MEM_READ64_LENGTH, MHI_EOT); + if (ret) + dev_err(&context->mhi_dev->dev, + "Unable to send read for dump content %d\n", ret); +} + +static void sahara_dump_processing(struct work_struct *work) +{ + struct sahara_context *context = container_of(work, struct sahara_context, dump_work); + int ret; + + /* + * We should get the expected raw data, but if the device has an error + * it is supposed to send EOI with an error code. + */ + if (context->rx_size != context->rx_size_requested && + context->rx_size != SAHARA_END_OF_IMAGE_LENGTH) { + dev_err(&context->mhi_dev->dev, + "Unexpected response to read_data. Expected size: %#zx got: %#zx\n", + context->rx_size_requested, + context->rx_size); + goto error; + } + + if (context->rx_size == SAHARA_END_OF_IMAGE_LENGTH && + le32_to_cpu(context->rx->cmd) == SAHARA_END_OF_IMAGE_CMD) { + dev_err(&context->mhi_dev->dev, + "Unexpected EOI response to read_data. Status: %d\n", + le32_to_cpu(context->rx->end_of_image.status)); + goto error; + } + + if (context->rx_size == SAHARA_END_OF_IMAGE_LENGTH && + le32_to_cpu(context->rx->cmd) != SAHARA_END_OF_IMAGE_CMD) { + dev_err(&context->mhi_dev->dev, + "Invalid EOI response to read_data. CMD: %d\n", + le32_to_cpu(context->rx->cmd)); + goto error; + } + + /* + * Need to know if we received the dump table, or part of a dump image. + * Since we get raw data, we cannot tell from the data itself. Instead, + * we use the stored dump_table_length, which we zero after we read and + * process the entire table. + */ + if (context->dump_table_length) + sahara_parse_dump_table(context); + else + sahara_parse_dump_image(context); + + ret = mhi_queue_buf(context->mhi_dev, DMA_FROM_DEVICE, context->rx, + SAHARA_PACKET_MAX_SIZE, MHI_EOT); + if (ret) + dev_err(&context->mhi_dev->dev, "Unable to requeue rx buf %d\n", ret); + + return; + +error: + vfree(context->mem_dump); + context->mem_dump = NULL; + sahara_send_reset(context); +} + static int sahara_mhi_probe(struct mhi_device *mhi_dev, const struct mhi_device_id *id) { struct sahara_context *context; @@ -382,7 +746,8 @@ static int sahara_mhi_probe(struct mhi_device *mhi_dev, const struct mhi_device_ } context->mhi_dev = mhi_dev; - INIT_WORK(&context->work, sahara_processing); + INIT_WORK(&context->fw_work, sahara_processing); + INIT_WORK(&context->dump_work, sahara_dump_processing); context->image_table = aic100_image_table; context->table_size = ARRAY_SIZE(aic100_image_table); context->active_image_id = SAHARA_IMAGE_ID_NONE; @@ -405,7 +770,10 @@ static void sahara_mhi_remove(struct mhi_device *mhi_dev) { struct sahara_context *context = dev_get_drvdata(&mhi_dev->dev); - cancel_work_sync(&context->work); + cancel_work_sync(&context->fw_work); + cancel_work_sync(&context->dump_work); + if (context->mem_dump) + vfree(context->mem_dump); sahara_release_image(context); mhi_unprepare_from_transfer(mhi_dev); } @@ -418,8 +786,14 @@ static void sahara_mhi_dl_xfer_cb(struct mhi_device *mhi_dev, struct mhi_result { struct sahara_context *context = dev_get_drvdata(&mhi_dev->dev); - if (!mhi_result->transaction_status) - schedule_work(&context->work); + if (!mhi_result->transaction_status) { + context->rx_size = mhi_result->bytes_xferd; + if (context->is_mem_dump_mode) + schedule_work(&context->dump_work); + else + schedule_work(&context->fw_work); + } + } static const struct mhi_device_id sahara_mhi_match_table[] = { diff --git a/drivers/acpi/apei/apei-base.c b/drivers/acpi/apei/apei-base.c index c7c26872f4ce..9c84f3da7c09 100644 --- a/drivers/acpi/apei/apei-base.c +++ b/drivers/acpi/apei/apei-base.c @@ -28,7 +28,7 @@ #include #include #include -#include +#include #include "apei-internal.h" diff --git a/drivers/acpi/apei/einj-core.c b/drivers/acpi/apei/einj-core.c index 73903a497d73..5c22720f43cc 100644 --- a/drivers/acpi/apei/einj-core.c +++ b/drivers/acpi/apei/einj-core.c @@ -22,7 +22,7 @@ #include #include #include -#include +#include #include "apei-internal.h" diff --git a/drivers/acpi/apei/einj-cxl.c b/drivers/acpi/apei/einj-cxl.c index 4f81a119ec08..a4e709937236 100644 --- a/drivers/acpi/apei/einj-cxl.c +++ b/drivers/acpi/apei/einj-cxl.c @@ -63,7 +63,7 @@ static int cxl_dport_get_sbdf(struct pci_dev *dport_dev, u64 *sbdf) seg = bridge->domain_nr; bus = pbus->number; - *sbdf = (seg << 24) | (bus << 16) | dport_dev->devfn; + *sbdf = (seg << 24) | (bus << 16) | (dport_dev->devfn << 8); return 0; } diff --git a/drivers/acpi/battery.c b/drivers/acpi/battery.c index f4599261cfc3..65fa3444367a 100644 --- a/drivers/acpi/battery.c +++ b/drivers/acpi/battery.c @@ -21,7 +21,7 @@ #include #include -#include +#include #include #include @@ -703,28 +703,35 @@ static LIST_HEAD(acpi_battery_list); static LIST_HEAD(battery_hook_list); static DEFINE_MUTEX(hook_mutex); -static void __battery_hook_unregister(struct acpi_battery_hook *hook, int lock) +static void battery_hook_unregister_unlocked(struct acpi_battery_hook *hook) { struct acpi_battery *battery; + /* * In order to remove a hook, we first need to * de-register all the batteries that are registered. */ - if (lock) - mutex_lock(&hook_mutex); list_for_each_entry(battery, &acpi_battery_list, list) { if (!hook->remove_battery(battery->bat, hook)) power_supply_changed(battery->bat); } - list_del(&hook->list); - if (lock) - mutex_unlock(&hook_mutex); + list_del_init(&hook->list); + pr_info("extension unregistered: %s\n", hook->name); } void battery_hook_unregister(struct acpi_battery_hook *hook) { - __battery_hook_unregister(hook, 1); + mutex_lock(&hook_mutex); + /* + * Ignore already unregistered battery hooks. This might happen + * if a battery hook was previously unloaded due to an error when + * adding a new battery. + */ + if (!list_empty(&hook->list)) + battery_hook_unregister_unlocked(hook); + + mutex_unlock(&hook_mutex); } EXPORT_SYMBOL_GPL(battery_hook_unregister); @@ -733,7 +740,6 @@ void battery_hook_register(struct acpi_battery_hook *hook) struct acpi_battery *battery; mutex_lock(&hook_mutex); - INIT_LIST_HEAD(&hook->list); list_add(&hook->list, &battery_hook_list); /* * Now that the driver is registered, we need @@ -750,7 +756,7 @@ void battery_hook_register(struct acpi_battery_hook *hook) * hooks. */ pr_err("extension failed to load: %s", hook->name); - __battery_hook_unregister(hook, 0); + battery_hook_unregister_unlocked(hook); goto end; } @@ -804,7 +810,7 @@ static void battery_hook_add_battery(struct acpi_battery *battery) */ pr_err("error in extension, unloading: %s", hook_node->name); - __battery_hook_unregister(hook_node, 0); + battery_hook_unregister_unlocked(hook_node); } } mutex_unlock(&hook_mutex); @@ -837,7 +843,7 @@ static void __exit battery_hook_exit(void) * need to remove the hooks. */ list_for_each_entry_safe(hook, ptr, &battery_hook_list, list) { - __battery_hook_unregister(hook, 1); + battery_hook_unregister(hook); } mutex_destroy(&hook_mutex); } diff --git a/drivers/acpi/button.c b/drivers/acpi/button.c index 51470208e6da..7773e6b860e7 100644 --- a/drivers/acpi/button.c +++ b/drivers/acpi/button.c @@ -130,6 +130,17 @@ static const struct dmi_system_id dmi_lid_quirks[] = { }, .driver_data = (void *)(long)ACPI_BUTTON_LID_INIT_OPEN, }, + { + /* + * Samsung galaxybook2 ,initial _LID device notification returns + * lid closed. + */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "SAMSUNG ELECTRONICS CO., LTD."), + DMI_MATCH(DMI_PRODUCT_NAME, "750XED"), + }, + .driver_data = (void *)(long)ACPI_BUTTON_LID_INIT_OPEN, + }, {} }; diff --git a/drivers/acpi/cppc_acpi.c b/drivers/acpi/cppc_acpi.c index 5b06e236aabe..1a40f0514eaa 100644 --- a/drivers/acpi/cppc_acpi.c +++ b/drivers/acpi/cppc_acpi.c @@ -41,7 +41,7 @@ #include #include #include -#include +#include #include @@ -867,7 +867,7 @@ int acpi_cppc_processor_probe(struct acpi_processor *pr) /* Store CPU Logical ID */ cpc_ptr->cpu_id = pr->id; - spin_lock_init(&cpc_ptr->rmw_lock); + raw_spin_lock_init(&cpc_ptr->rmw_lock); /* Parse PSD data for this CPU */ ret = acpi_get_psd(cpc_ptr, handle); @@ -1087,6 +1087,7 @@ static int cpc_write(int cpu, struct cpc_register_resource *reg_res, u64 val) int pcc_ss_id = per_cpu(cpu_pcc_subspace_idx, cpu); struct cpc_reg *reg = ®_res->cpc_entry.reg; struct cpc_desc *cpc_desc; + unsigned long flags; size = GET_BIT_WIDTH(reg); @@ -1126,7 +1127,7 @@ static int cpc_write(int cpu, struct cpc_register_resource *reg_res, u64 val) return -ENODEV; } - spin_lock(&cpc_desc->rmw_lock); + raw_spin_lock_irqsave(&cpc_desc->rmw_lock, flags); switch (size) { case 8: prev_val = readb_relaxed(vaddr); @@ -1141,7 +1142,7 @@ static int cpc_write(int cpu, struct cpc_register_resource *reg_res, u64 val) prev_val = readq_relaxed(vaddr); break; default: - spin_unlock(&cpc_desc->rmw_lock); + raw_spin_unlock_irqrestore(&cpc_desc->rmw_lock, flags); return -EFAULT; } val = MASK_VAL_WRITE(reg, prev_val, val); @@ -1174,7 +1175,7 @@ static int cpc_write(int cpu, struct cpc_register_resource *reg_res, u64 val) } if (reg->space_id == ACPI_ADR_SPACE_SYSTEM_MEMORY) - spin_unlock(&cpc_desc->rmw_lock); + raw_spin_unlock_irqrestore(&cpc_desc->rmw_lock, flags); return ret_val; } @@ -1916,9 +1917,15 @@ unsigned int cppc_perf_to_khz(struct cppc_perf_caps *caps, unsigned int perf) u64 mul, div; if (caps->lowest_freq && caps->nominal_freq) { - mul = caps->nominal_freq - caps->lowest_freq; + /* Avoid special case when nominal_freq is equal to lowest_freq */ + if (caps->lowest_freq == caps->nominal_freq) { + mul = caps->nominal_freq; + div = caps->nominal_perf; + } else { + mul = caps->nominal_freq - caps->lowest_freq; + div = caps->nominal_perf - caps->lowest_perf; + } mul *= KHZ_PER_MHZ; - div = caps->nominal_perf - caps->lowest_perf; offset = caps->nominal_freq * KHZ_PER_MHZ - div64_u64(caps->nominal_perf * mul, div); } else { @@ -1939,11 +1946,17 @@ unsigned int cppc_khz_to_perf(struct cppc_perf_caps *caps, unsigned int freq) { s64 retval, offset = 0; static u64 max_khz; - u64 mul, div; + u64 mul, div; if (caps->lowest_freq && caps->nominal_freq) { - mul = caps->nominal_perf - caps->lowest_perf; - div = caps->nominal_freq - caps->lowest_freq; + /* Avoid special case when nominal_freq is equal to lowest_freq */ + if (caps->lowest_freq == caps->nominal_freq) { + mul = caps->nominal_perf; + div = caps->nominal_freq; + } else { + mul = caps->nominal_perf - caps->lowest_perf; + div = caps->nominal_freq - caps->lowest_freq; + } /* * We don't need to convert to kHz for computing offset and can * directly use nominal_freq and lowest_freq as the div64_u64 diff --git a/drivers/acpi/prmt.c b/drivers/acpi/prmt.c index 1cfaa5957ac4..747f83f7114d 100644 --- a/drivers/acpi/prmt.c +++ b/drivers/acpi/prmt.c @@ -52,7 +52,7 @@ struct prm_context_buffer { static LIST_HEAD(prm_module_list); struct prm_handler_info { - guid_t guid; + efi_guid_t guid; efi_status_t (__efiapi *handler_addr)(u64, void *); u64 static_data_buffer_addr; u64 acpi_param_buffer_addr; @@ -72,17 +72,21 @@ struct prm_module_info { struct prm_handler_info handlers[] __counted_by(handler_count); }; -static u64 efi_pa_va_lookup(u64 pa) +static u64 efi_pa_va_lookup(efi_guid_t *guid, u64 pa) { efi_memory_desc_t *md; u64 pa_offset = pa & ~PAGE_MASK; u64 page = pa & PAGE_MASK; for_each_efi_memory_desc(md) { - if (md->phys_addr < pa && pa < md->phys_addr + PAGE_SIZE * md->num_pages) + if ((md->attribute & EFI_MEMORY_RUNTIME) && + (md->phys_addr < pa && pa < md->phys_addr + PAGE_SIZE * md->num_pages)) { return pa_offset + md->virt_addr + page - md->phys_addr; + } } + pr_warn("Failed to find VA for GUID: %pUL, PA: 0x%llx", guid, pa); + return 0; } @@ -148,9 +152,15 @@ acpi_parse_prmt(union acpi_subtable_headers *header, const unsigned long end) th = &tm->handlers[cur_handler]; guid_copy(&th->guid, (guid_t *)handler_info->handler_guid); - th->handler_addr = (void *)efi_pa_va_lookup(handler_info->handler_address); - th->static_data_buffer_addr = efi_pa_va_lookup(handler_info->static_data_buffer_address); - th->acpi_param_buffer_addr = efi_pa_va_lookup(handler_info->acpi_param_buffer_address); + th->handler_addr = + (void *)efi_pa_va_lookup(&th->guid, handler_info->handler_address); + + th->static_data_buffer_addr = + efi_pa_va_lookup(&th->guid, handler_info->static_data_buffer_address); + + th->acpi_param_buffer_addr = + efi_pa_va_lookup(&th->guid, handler_info->acpi_param_buffer_address); + } while (++cur_handler < tm->handler_count && (handler_info = get_next_handler(handler_info))); return 0; @@ -277,6 +287,13 @@ static acpi_status acpi_platformrt_space_handler(u32 function, if (!handler || !module) goto invalid_guid; + if (!handler->handler_addr || + !handler->static_data_buffer_addr || + !handler->acpi_param_buffer_addr) { + buffer->prm_status = PRM_HANDLER_ERROR; + return AE_OK; + } + ACPI_COPY_NAMESEG(context.signature, "PRMC"); context.revision = 0x0; context.reserved = 0x0; diff --git a/drivers/acpi/resource.c b/drivers/acpi/resource.c index 8a4726e2eb69..7fe842dae1ec 100644 --- a/drivers/acpi/resource.c +++ b/drivers/acpi/resource.c @@ -441,108 +441,59 @@ static const struct dmi_system_id irq1_level_low_skip_override[] = { }, }, { - /* Asus ExpertBook B1402CBA */ + /* Asus Vivobook X1704VAP */ .matches = { DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "B1402CBA"), + DMI_MATCH(DMI_BOARD_NAME, "X1704VAP"), }, }, { - /* Asus ExpertBook B1402CVA */ + /* Asus ExpertBook B1402C* */ .matches = { DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "B1402CVA"), + DMI_MATCH(DMI_BOARD_NAME, "B1402C"), }, }, { - /* Asus ExpertBook B1502CBA */ + /* Asus ExpertBook B1502C* */ .matches = { DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "B1502CBA"), + DMI_MATCH(DMI_BOARD_NAME, "B1502C"), }, }, { - /* Asus ExpertBook B1502CGA */ + /* Asus ExpertBook B2402 (B2402CBA / B2402FBA / B2402CVA / B2402FVA) */ .matches = { DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "B1502CGA"), - }, - }, - { - /* Asus ExpertBook B1502CVA */ - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "B1502CVA"), - }, - }, - { - /* Asus ExpertBook B2402CBA */ - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "B2402CBA"), + DMI_MATCH(DMI_BOARD_NAME, "B2402"), }, }, { - /* Asus ExpertBook B2402FBA */ + /* Asus ExpertBook B2502 (B2502CBA / B2502FBA / B2502CVA / B2502FVA) */ .matches = { DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "B2402FBA"), + DMI_MATCH(DMI_BOARD_NAME, "B2502"), }, }, { - /* Asus ExpertBook B2502 */ + /* Asus Vivobook Go E1404GA* */ .matches = { DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "B2502CBA"), + DMI_MATCH(DMI_BOARD_NAME, "E1404GA"), }, }, { - /* Asus ExpertBook B2502FBA */ - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "B2502FBA"), - }, - }, - { - /* Asus Vivobook Go E1404GAB */ - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "E1404GAB"), - }, - }, - { - /* Asus Vivobook E1504GA */ + /* Asus Vivobook E1504GA* */ .matches = { DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), DMI_MATCH(DMI_BOARD_NAME, "E1504GA"), }, }, { - /* Asus Vivobook E1504GAB */ + /* Asus Vivobook Pro N6506M* */ .matches = { DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "E1504GAB"), - }, - }, - { - /* Asus Vivobook Pro N6506MV */ - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "N6506MV"), - }, - }, - { - /* Asus Vivobook Pro N6506MU */ - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "N6506MU"), - }, - }, - { - /* Asus Vivobook Pro N6506MJ */ - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), - DMI_MATCH(DMI_BOARD_NAME, "N6506MJ"), + DMI_MATCH(DMI_BOARD_NAME, "N6506M"), }, }, { @@ -552,6 +503,13 @@ static const struct dmi_system_id irq1_level_low_skip_override[] = { DMI_MATCH(DMI_BOARD_NAME, "17U70P"), }, }, + { + /* LG Electronics 16T90SP */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LG Electronics"), + DMI_MATCH(DMI_BOARD_NAME, "16T90SP"), + }, + }, { } }; diff --git a/drivers/acpi/video_detect.c b/drivers/acpi/video_detect.c index b70e84e8049a..015bd8e66c1c 100644 --- a/drivers/acpi/video_detect.c +++ b/drivers/acpi/video_detect.c @@ -844,6 +844,15 @@ static const struct dmi_system_id video_detect_dmi_table[] = { * controller board in their ACPI tables (and may even have one), but * which need native backlight control nevertheless. */ + { + /* https://github.com/zabbly/linux/issues/26 */ + .callback = video_detect_force_native, + /* Dell OptiPlex 5480 AIO */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 5480 AIO"), + }, + }, { /* https://bugzilla.redhat.com/show_bug.cgi?id=2303936 */ .callback = video_detect_force_native, diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index cdb20a700b55..c085dd81ebe7 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -50,7 +50,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index 3f0144e7dc80..3b303d4ae37a 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -651,6 +651,7 @@ void ata_scsi_cmd_error_handler(struct Scsi_Host *host, struct ata_port *ap, /* the scmd has an associated qc */ if (!(qc->flags & ATA_QCFLAG_EH)) { /* which hasn't failed yet, timeout */ + set_host_byte(scmd, DID_TIME_OUT); qc->err_mask |= AC_ERR_TIMEOUT; qc->flags |= ATA_QCFLAG_EH; nr_timedout++; @@ -4099,10 +4100,20 @@ static void ata_eh_handle_port_suspend(struct ata_port *ap) WARN_ON(ap->pflags & ATA_PFLAG_SUSPENDED); - /* Set all devices attached to the port in standby mode */ - ata_for_each_link(link, ap, HOST_FIRST) { - ata_for_each_dev(dev, link, ENABLED) - ata_dev_power_set_standby(dev); + /* + * We will reach this point for all of the PM events: + * PM_EVENT_SUSPEND (if runtime pm, PM_EVENT_AUTO will also be set) + * PM_EVENT_FREEZE, and PM_EVENT_HIBERNATE. + * + * We do not want to perform disk spin down for PM_EVENT_FREEZE. + * (Spin down will be performed by the subsequent PM_EVENT_HIBERNATE.) + */ + if (!(ap->pm_mesg.event & PM_EVENT_FREEZE)) { + /* Set all devices attached to the port in standby mode */ + ata_for_each_link(link, ap, HOST_FIRST) { + ata_for_each_dev(dev, link, ENABLED) + ata_dev_power_set_standby(dev); + } } /* diff --git a/drivers/ata/libata-sata.c b/drivers/ata/libata-sata.c index c8b119a06bb2..9c76fb1ad2ec 100644 --- a/drivers/ata/libata-sata.c +++ b/drivers/ata/libata-sata.c @@ -13,7 +13,7 @@ #include #include #include -#include +#include #include "libata.h" #include "libata-transport.h" diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index a4aedf7e1775..f915e3df57a9 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -30,7 +30,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/auxdisplay/ht16k33.c b/drivers/auxdisplay/ht16k33.c index 8a7034b41d50..a816f9e10255 100644 --- a/drivers/auxdisplay/ht16k33.c +++ b/drivers/auxdisplay/ht16k33.c @@ -25,7 +25,7 @@ #include #include -#include +#include #include "line-display.h" diff --git a/drivers/base/core.c b/drivers/base/core.c index a4c853411a6b..048ff98dbdfd 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -26,7 +26,6 @@ #include #include #include -#include #include #include #include @@ -2634,7 +2633,6 @@ static const char *dev_uevent_name(const struct kobject *kobj) static int dev_uevent(const struct kobject *kobj, struct kobj_uevent_env *env) { const struct device *dev = kobj_to_dev(kobj); - struct device_driver *driver; int retval = 0; /* add device node properties if present */ @@ -2663,12 +2661,8 @@ static int dev_uevent(const struct kobject *kobj, struct kobj_uevent_env *env) if (dev->type && dev->type->name) add_uevent_var(env, "DEVTYPE=%s", dev->type->name); - /* Synchronize with module_remove_driver() */ - rcu_read_lock(); - driver = READ_ONCE(dev->driver); - if (driver) - add_uevent_var(env, "DRIVER=%s", driver->name); - rcu_read_unlock(); + if (dev->driver) + add_uevent_var(env, "DRIVER=%s", dev->driver->name); /* Add common DT information about the device */ of_device_uevent(dev, env); @@ -2738,8 +2732,11 @@ static ssize_t uevent_show(struct device *dev, struct device_attribute *attr, if (!env) return -ENOMEM; + /* Synchronize with really_probe() */ + device_lock(dev); /* let the kset specific function add its keys */ retval = kset->uevent_ops->uevent(&dev->kobj, env); + device_unlock(dev); if (retval) goto out; @@ -4037,6 +4034,41 @@ int device_for_each_child_reverse(struct device *parent, void *data, } EXPORT_SYMBOL_GPL(device_for_each_child_reverse); +/** + * device_for_each_child_reverse_from - device child iterator in reversed order. + * @parent: parent struct device. + * @from: optional starting point in child list + * @fn: function to be called for each device. + * @data: data for the callback. + * + * Iterate over @parent's child devices, starting at @from, and call @fn + * for each, passing it @data. This helper is identical to + * device_for_each_child_reverse() when @from is NULL. + * + * @fn is checked each iteration. If it returns anything other than 0, + * iteration stop and that value is returned to the caller of + * device_for_each_child_reverse_from(); + */ +int device_for_each_child_reverse_from(struct device *parent, + struct device *from, const void *data, + int (*fn)(struct device *, const void *)) +{ + struct klist_iter i; + struct device *child; + int error = 0; + + if (!parent->p) + return 0; + + klist_iter_init_node(&parent->p->klist_children, &i, + (from ? &from->p->knode_parent : NULL)); + while ((child = prev_device(&i)) && !error) + error = fn(child, data); + klist_iter_exit(&i); + return error; +} +EXPORT_SYMBOL_GPL(device_for_each_child_reverse_from); + /** * device_find_child - device iterator for locating a particular device. * @parent: parent struct device diff --git a/drivers/base/module.c b/drivers/base/module.c index c4eaa1158d54..5bc71bea883a 100644 --- a/drivers/base/module.c +++ b/drivers/base/module.c @@ -7,7 +7,6 @@ #include #include #include -#include #include "base.h" static char *make_driver_name(const struct device_driver *drv) @@ -102,9 +101,6 @@ void module_remove_driver(const struct device_driver *drv) if (!drv) return; - /* Synchronize with dev_uevent() */ - synchronize_rcu(); - sysfs_remove_link(&drv->p->kobj, "module"); if (drv->owner) diff --git a/drivers/base/power/common.c b/drivers/base/power/common.c index 8c34ae1cd8d5..cca2fd0a1aed 100644 --- a/drivers/base/power/common.c +++ b/drivers/base/power/common.c @@ -195,6 +195,7 @@ int dev_pm_domain_attach_list(struct device *dev, struct device *pd_dev = NULL; int ret, i, num_pds = 0; bool by_id = true; + size_t size; u32 pd_flags = data ? data->pd_flags : 0; u32 link_flags = pd_flags & PD_FLAG_NO_DEV_LINK ? 0 : DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME; @@ -217,19 +218,17 @@ int dev_pm_domain_attach_list(struct device *dev, if (num_pds <= 0) return 0; - pds = devm_kzalloc(dev, sizeof(*pds), GFP_KERNEL); + pds = kzalloc(sizeof(*pds), GFP_KERNEL); if (!pds) return -ENOMEM; - pds->pd_devs = devm_kcalloc(dev, num_pds, sizeof(*pds->pd_devs), - GFP_KERNEL); - if (!pds->pd_devs) - return -ENOMEM; - - pds->pd_links = devm_kcalloc(dev, num_pds, sizeof(*pds->pd_links), - GFP_KERNEL); - if (!pds->pd_links) - return -ENOMEM; + size = sizeof(*pds->pd_devs) + sizeof(*pds->pd_links); + pds->pd_devs = kcalloc(num_pds, size, GFP_KERNEL); + if (!pds->pd_devs) { + ret = -ENOMEM; + goto free_pds; + } + pds->pd_links = (void *)(pds->pd_devs + num_pds); if (link_flags && pd_flags & PD_FLAG_DEV_LINK_ON) link_flags |= DL_FLAG_RPM_ACTIVE; @@ -272,6 +271,9 @@ err_attach: device_link_del(pds->pd_links[i]); dev_pm_domain_detach(pds->pd_devs[i], true); } + kfree(pds->pd_devs); +free_pds: + kfree(pds); return ret; } EXPORT_SYMBOL_GPL(dev_pm_domain_attach_list); @@ -363,6 +365,9 @@ void dev_pm_domain_detach_list(struct dev_pm_domain_list *list) device_link_del(list->pd_links[i]); dev_pm_domain_detach(list->pd_devs[i], true); } + + kfree(list->pd_devs); + kfree(list); } EXPORT_SYMBOL_GPL(dev_pm_domain_detach_list); diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index 9ed842d17642..4ded93687c1f 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -17,7 +17,7 @@ #include #include #include -#include +#include #define CREATE_TRACE_POINTS #include "trace.h" diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index cc9077b588d7..92b06d1de4cc 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -14,7 +14,7 @@ #include #include #include -#include +#include #include #include "aoe.h" @@ -361,6 +361,7 @@ ata_rw_frameinit(struct frame *f) } ah->cmdstat = ATA_CMD_PIO_READ | writebit | extbit; + dev_hold(t->ifp->nd); skb->dev = t->ifp->nd; } @@ -401,6 +402,8 @@ aoecmd_ata_rw(struct aoedev *d) __skb_queue_head_init(&queue); __skb_queue_tail(&queue, skb); aoenet_xmit(&queue); + } else { + dev_put(f->t->ifp->nd); } return 1; } @@ -483,10 +486,13 @@ resend(struct aoedev *d, struct frame *f) memcpy(h->dst, t->addr, sizeof h->dst); memcpy(h->src, t->ifp->nd->dev_addr, sizeof h->src); + dev_hold(t->ifp->nd); skb->dev = t->ifp->nd; skb = skb_clone(skb, GFP_ATOMIC); - if (skb == NULL) + if (skb == NULL) { + dev_put(t->ifp->nd); return; + } f->sent = ktime_get(); __skb_queue_head_init(&queue); __skb_queue_tail(&queue, skb); @@ -617,6 +623,8 @@ probe(struct aoetgt *t) __skb_queue_head_init(&queue); __skb_queue_tail(&queue, skb); aoenet_xmit(&queue); + } else { + dev_put(f->t->ifp->nd); } } @@ -1395,6 +1403,7 @@ aoecmd_ata_id(struct aoedev *d) ah->cmdstat = ATA_CMD_ID_ATA; ah->lba3 = 0xa0; + dev_hold(t->ifp->nd); skb->dev = t->ifp->nd; d->rttavg = RTTAVG_INIT; @@ -1404,6 +1413,8 @@ aoecmd_ata_id(struct aoedev *d) skb = skb_clone(skb, GFP_ATOMIC); if (skb) f->sent = ktime_get(); + else + dev_put(t->ifp->nd); return skb; } diff --git a/drivers/block/aoe/aoenet.c b/drivers/block/aoe/aoenet.c index 923a134fd766..66e617664c14 100644 --- a/drivers/block/aoe/aoenet.c +++ b/drivers/block/aoe/aoenet.c @@ -10,7 +10,7 @@ #include #include #include -#include +#include #include "aoe.h" #define NECODES 5 diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h index 2a05d955e30b..e21492981f7d 100644 --- a/drivers/block/drbd/drbd_int.h +++ b/drivers/block/drbd/drbd_int.h @@ -1364,7 +1364,6 @@ extern struct bio_set drbd_io_bio_set; extern struct mutex resources_mutex; -extern int conn_lowest_minor(struct drbd_connection *connection); extern enum drbd_ret_code drbd_create_device(struct drbd_config_context *adm_ctx, unsigned int minor); extern void drbd_destroy_device(struct kref *kref); extern void drbd_delete_device(struct drbd_device *device); diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c index 0d74d75260ef..5bbd312c3e14 100644 --- a/drivers/block/drbd/drbd_main.c +++ b/drivers/block/drbd/drbd_main.c @@ -471,20 +471,6 @@ void _drbd_thread_stop(struct drbd_thread *thi, int restart, int wait) wait_for_completion(&thi->stop); } -int conn_lowest_minor(struct drbd_connection *connection) -{ - struct drbd_peer_device *peer_device; - int vnr = 0, minor = -1; - - rcu_read_lock(); - peer_device = idr_get_next(&connection->peer_devices, &vnr); - if (peer_device) - minor = device_to_minor(peer_device->device); - rcu_read_unlock(); - - return minor; -} - #ifdef CONFIG_SMP /* * drbd_calc_cpu_mask() - Generate CPU masks, spread over all CPUs diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c index 5d65c9754d83..720fc30e2ecc 100644 --- a/drivers/block/drbd/drbd_nl.c +++ b/drivers/block/drbd/drbd_nl.c @@ -25,7 +25,7 @@ #include "drbd_protocol.h" #include "drbd_req.h" #include "drbd_state_change.h" -#include +#include #include #include diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c index 499c110465e3..65b96c083b3c 100644 --- a/drivers/block/pktcdvd.c +++ b/drivers/block/pktcdvd.c @@ -71,7 +71,7 @@ #include #include -#include +#include #define DRIVER_NAME "pktcdvd" diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c index a6c8e5cc6051..6ba2c1dd1d87 100644 --- a/drivers/block/ublk_drv.c +++ b/drivers/block/ublk_drv.c @@ -2380,10 +2380,19 @@ static int ublk_ctrl_add_dev(struct io_uring_cmd *cmd) * TODO: provide forward progress for RECOVERY handler, so that * unprivileged device can benefit from it */ - if (info.flags & UBLK_F_UNPRIVILEGED_DEV) + if (info.flags & UBLK_F_UNPRIVILEGED_DEV) { info.flags &= ~(UBLK_F_USER_RECOVERY_REISSUE | UBLK_F_USER_RECOVERY); + /* + * For USER_COPY, we depends on userspace to fill request + * buffer by pwrite() to ublk char device, which can't be + * used for unprivileged device + */ + if (info.flags & UBLK_F_USER_COPY) + return -EINVAL; + } + /* the created device is always owned by current user */ ublk_store_owner_uid_gid(&info.owner_uid, &info.owner_gid); diff --git a/drivers/bluetooth/ath3k.c b/drivers/bluetooth/ath3k.c index ce97b336fbfb..fc796f1dbda9 100644 --- a/drivers/bluetooth/ath3k.c +++ b/drivers/bluetooth/ath3k.c @@ -11,7 +11,7 @@ #include #include #include -#include +#include #include #define VERSION "1.0" diff --git a/drivers/bluetooth/btbcm.c b/drivers/bluetooth/btbcm.c index f9a7c790d7e2..eef00467905e 100644 --- a/drivers/bluetooth/btbcm.c +++ b/drivers/bluetooth/btbcm.c @@ -12,7 +12,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c index 1ccbb5157515..438b92967bc3 100644 --- a/drivers/bluetooth/btintel.c +++ b/drivers/bluetooth/btintel.c @@ -11,7 +11,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/bluetooth/btintel_pcie.c b/drivers/bluetooth/btintel_pcie.c index fda47948c35d..5252125b003f 100644 --- a/drivers/bluetooth/btintel_pcie.c +++ b/drivers/bluetooth/btintel_pcie.c @@ -14,7 +14,7 @@ #include #include -#include +#include #include #include diff --git a/drivers/bluetooth/btmrvl_sdio.c b/drivers/bluetooth/btmrvl_sdio.c index 85b7f2bb4259..07cd308f7abf 100644 --- a/drivers/bluetooth/btmrvl_sdio.c +++ b/drivers/bluetooth/btmrvl_sdio.c @@ -92,7 +92,7 @@ static int btmrvl_sdio_probe_of(struct device *dev, } else { ret = devm_request_irq(dev, cfg->irq_bt, btmrvl_wake_irq_bt, - 0, "bt_wake", card); + IRQF_NO_AUTOEN, "bt_wake", card); if (ret) { dev_err(dev, "Failed to request irq_bt %d (%d)\n", @@ -101,7 +101,6 @@ static int btmrvl_sdio_probe_of(struct device *dev, /* Configure wakeup (enabled by default) */ device_init_wakeup(dev, true); - disable_irq(cfg->irq_bt); } } diff --git a/drivers/bluetooth/btmtk.c b/drivers/bluetooth/btmtk.c index 2b7c80043aa2..9bbf20502163 100644 --- a/drivers/bluetooth/btmtk.c +++ b/drivers/bluetooth/btmtk.c @@ -6,7 +6,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/bluetooth/btmtksdio.c b/drivers/bluetooth/btmtksdio.c index 497e4c87f5be..11d33cd7b08f 100644 --- a/drivers/bluetooth/btmtksdio.c +++ b/drivers/bluetooth/btmtksdio.c @@ -10,7 +10,7 @@ * */ -#include +#include #include #include #include diff --git a/drivers/bluetooth/btmtkuart.c b/drivers/bluetooth/btmtkuart.c index aa87c3e78871..64e4d835af52 100644 --- a/drivers/bluetooth/btmtkuart.c +++ b/drivers/bluetooth/btmtkuart.c @@ -8,7 +8,7 @@ * */ -#include +#include #include #include #include diff --git a/drivers/bluetooth/btnxpuart.c b/drivers/bluetooth/btnxpuart.c index 7c2030cec10e..5ea0d23e88c0 100644 --- a/drivers/bluetooth/btnxpuart.c +++ b/drivers/bluetooth/btnxpuart.c @@ -10,7 +10,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/bluetooth/btrsi.c b/drivers/bluetooth/btrsi.c index 0c91d7635ac3..6c1f584c8a33 100644 --- a/drivers/bluetooth/btrsi.c +++ b/drivers/bluetooth/btrsi.c @@ -17,7 +17,7 @@ #include #include #include -#include +#include #include #define RSI_DMA_ALIGN 8 diff --git a/drivers/bluetooth/btrtl.c b/drivers/bluetooth/btrtl.c index 2d95b3ea046d..0bcb44cf7b31 100644 --- a/drivers/bluetooth/btrtl.c +++ b/drivers/bluetooth/btrtl.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include #include diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 6c9c761d5b93..e9534fbc92e3 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -17,7 +17,7 @@ #include #include #include -#include +#include #include #include @@ -1345,10 +1345,15 @@ static int btusb_submit_intr_urb(struct hci_dev *hdev, gfp_t mem_flags) if (!urb) return -ENOMEM; - /* Use maximum HCI Event size so the USB stack handles - * ZPL/short-transfer automatically. - */ - size = HCI_MAX_EVENT_SIZE; + if (le16_to_cpu(data->udev->descriptor.idVendor) == 0x0a12 && + le16_to_cpu(data->udev->descriptor.idProduct) == 0x0001) + /* Fake CSR devices don't seem to support sort-transter */ + size = le16_to_cpu(data->intr_ep->wMaxPacketSize); + else + /* Use maximum HCI Event size so the USB stack handles + * ZPL/short-transfer automatically. + */ + size = HCI_MAX_EVENT_SIZE; buf = kmalloc(size, mem_flags); if (!buf) { @@ -4041,8 +4046,10 @@ static int btusb_suspend(struct usb_interface *intf, pm_message_t message) BT_DBG("intf %p", intf); - /* Don't suspend if there are connections */ - if (hci_conn_count(data->hdev)) + /* Don't auto-suspend if there are connections; external suspend calls + * shall never fail. + */ + if (PMSG_IS_AUTO(message) && hci_conn_count(data->hdev)) return -EBUSY; if (data->suspend_count++) diff --git a/drivers/bluetooth/h4_recv.h b/drivers/bluetooth/h4_recv.h index 647d37ca4cdd..28cf2d8c2d48 100644 --- a/drivers/bluetooth/h4_recv.h +++ b/drivers/bluetooth/h4_recv.h @@ -6,7 +6,7 @@ * Copyright (C) 2015-2018 Intel Corporation */ -#include +#include struct h4_recv_pkt { u8 type; /* Packet type */ diff --git a/drivers/bluetooth/hci_bcm4377.c b/drivers/bluetooth/hci_bcm4377.c index 77a5454a8721..9bce53e49cfa 100644 --- a/drivers/bluetooth/hci_bcm4377.c +++ b/drivers/bluetooth/hci_bcm4377.c @@ -17,7 +17,7 @@ #include #include -#include +#include #include #include diff --git a/drivers/bluetooth/hci_bcsp.c b/drivers/bluetooth/hci_bcsp.c index 2a5a27d713f8..76878119d910 100644 --- a/drivers/bluetooth/hci_bcsp.c +++ b/drivers/bluetooth/hci_bcsp.c @@ -25,7 +25,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/bluetooth/hci_h4.c b/drivers/bluetooth/hci_h4.c index 1d0cdf023243..9070e31a68bf 100644 --- a/drivers/bluetooth/hci_h4.c +++ b/drivers/bluetooth/hci_h4.c @@ -25,7 +25,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/bluetooth/hci_nokia.c b/drivers/bluetooth/hci_nokia.c index 62633d9ba7c4..49bbe4975be4 100644 --- a/drivers/bluetooth/hci_nokia.c +++ b/drivers/bluetooth/hci_nokia.c @@ -20,7 +20,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c index 678f150229e7..37fddf6055be 100644 --- a/drivers/bluetooth/hci_qca.c +++ b/drivers/bluetooth/hci_qca.c @@ -32,7 +32,7 @@ #include #include #include -#include +#include #include #include diff --git a/drivers/bluetooth/hci_vhci.c b/drivers/bluetooth/hci_vhci.c index aa6af351d02d..7651321d351c 100644 --- a/drivers/bluetooth/hci_vhci.c +++ b/drivers/bluetooth/hci_vhci.c @@ -9,7 +9,7 @@ */ #include -#include +#include #include #include diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c index 9b0f37d4b9d4..6a99a459b80b 100644 --- a/drivers/cdrom/cdrom.c +++ b/drivers/cdrom/cdrom.c @@ -2313,7 +2313,7 @@ static int cdrom_ioctl_media_changed(struct cdrom_device_info *cdi, return -EINVAL; /* Prevent arg from speculatively bypassing the length check */ - barrier_nospec(); + arg = array_index_nospec(arg, cdi->capacity); info = kmalloc(sizeof(*info), GFP_KERNEL); if (!info) diff --git a/drivers/char/tpm/tpm-chip.c b/drivers/char/tpm/tpm-chip.c index 854546000c92..1ff99a7091bb 100644 --- a/drivers/char/tpm/tpm-chip.c +++ b/drivers/char/tpm/tpm-chip.c @@ -674,6 +674,16 @@ EXPORT_SYMBOL_GPL(tpm_chip_register); */ void tpm_chip_unregister(struct tpm_chip *chip) { +#ifdef CONFIG_TCG_TPM2_HMAC + int rc; + + rc = tpm_try_get_ops(chip); + if (!rc) { + tpm2_end_auth_session(chip); + tpm_put_ops(chip); + } +#endif + tpm_del_legacy_sysfs(chip); if (tpm_is_hwrng_enabled(chip)) hwrng_unregister(&chip->hwrng); diff --git a/drivers/char/tpm/tpm-dev-common.c b/drivers/char/tpm/tpm-dev-common.c index c3fbbf4d3db7..48ff87444f85 100644 --- a/drivers/char/tpm/tpm-dev-common.c +++ b/drivers/char/tpm/tpm-dev-common.c @@ -27,6 +27,9 @@ static ssize_t tpm_dev_transmit(struct tpm_chip *chip, struct tpm_space *space, struct tpm_header *header = (void *)buf; ssize_t ret, len; + if (chip->flags & TPM_CHIP_FLAG_TPM2) + tpm2_end_auth_session(chip); + ret = tpm2_prepare_space(chip, space, buf, bufsiz); /* If the command is not implemented by the TPM, synthesize a * response with a TPM2_RC_COMMAND_CODE return for user-space. diff --git a/drivers/char/tpm/tpm-interface.c b/drivers/char/tpm/tpm-interface.c index 5da134f12c9a..8134f002b121 100644 --- a/drivers/char/tpm/tpm-interface.c +++ b/drivers/char/tpm/tpm-interface.c @@ -379,10 +379,12 @@ int tpm_pm_suspend(struct device *dev) rc = tpm_try_get_ops(chip); if (!rc) { - if (chip->flags & TPM_CHIP_FLAG_TPM2) + if (chip->flags & TPM_CHIP_FLAG_TPM2) { + tpm2_end_auth_session(chip); tpm2_shutdown(chip, TPM2_SU_STATE); - else + } else { rc = tpm1_pm_suspend(chip, tpm_suspend_pcr); + } tpm_put_ops(chip); } diff --git a/drivers/char/tpm/tpm2-sessions.c b/drivers/char/tpm/tpm2-sessions.c index 44f60730cff4..0739830904b2 100644 --- a/drivers/char/tpm/tpm2-sessions.c +++ b/drivers/char/tpm/tpm2-sessions.c @@ -71,7 +71,7 @@ #include "tpm.h" #include #include -#include +#include #include #include #include @@ -333,6 +333,9 @@ void tpm_buf_append_hmac_session(struct tpm_chip *chip, struct tpm_buf *buf, } #ifdef CONFIG_TCG_TPM2_HMAC + /* The first write to /dev/tpm{rm0} will flush the session. */ + attributes |= TPM2_SA_CONTINUE_SESSION; + /* * The Architecture Guide requires us to strip trailing zeros * before computing the HMAC @@ -484,7 +487,8 @@ static void tpm2_KDFe(u8 z[EC_PT_SZ], const char *str, u8 *pt_u, u8 *pt_v, sha256_final(&sctx, out); } -static void tpm_buf_append_salt(struct tpm_buf *buf, struct tpm_chip *chip) +static void tpm_buf_append_salt(struct tpm_buf *buf, struct tpm_chip *chip, + struct tpm2_auth *auth) { struct crypto_kpp *kpp; struct kpp_request *req; @@ -543,7 +547,7 @@ static void tpm_buf_append_salt(struct tpm_buf *buf, struct tpm_chip *chip) sg_set_buf(&s[0], chip->null_ec_key_x, EC_PT_SZ); sg_set_buf(&s[1], chip->null_ec_key_y, EC_PT_SZ); kpp_request_set_input(req, s, EC_PT_SZ*2); - sg_init_one(d, chip->auth->salt, EC_PT_SZ); + sg_init_one(d, auth->salt, EC_PT_SZ); kpp_request_set_output(req, d, EC_PT_SZ); crypto_kpp_compute_shared_secret(req); kpp_request_free(req); @@ -554,8 +558,7 @@ static void tpm_buf_append_salt(struct tpm_buf *buf, struct tpm_chip *chip) * This works because KDFe fully consumes the secret before it * writes the salt */ - tpm2_KDFe(chip->auth->salt, "SECRET", x, chip->null_ec_key_x, - chip->auth->salt); + tpm2_KDFe(auth->salt, "SECRET", x, chip->null_ec_key_x, auth->salt); out: crypto_free_kpp(kpp); @@ -853,7 +856,9 @@ int tpm_buf_check_hmac_response(struct tpm_chip *chip, struct tpm_buf *buf, if (rc) /* manually close the session if it wasn't consumed */ tpm2_flush_context(chip, auth->handle); - memzero_explicit(auth, sizeof(*auth)); + + kfree_sensitive(auth); + chip->auth = NULL; } else { /* reset for next use */ auth->session = TPM_HEADER_SIZE; @@ -881,7 +886,8 @@ void tpm2_end_auth_session(struct tpm_chip *chip) return; tpm2_flush_context(chip, auth->handle); - memzero_explicit(auth, sizeof(*auth)); + kfree_sensitive(auth); + chip->auth = NULL; } EXPORT_SYMBOL(tpm2_end_auth_session); @@ -915,33 +921,37 @@ static int tpm2_parse_start_auth_session(struct tpm2_auth *auth, static int tpm2_load_null(struct tpm_chip *chip, u32 *null_key) { - int rc; unsigned int offset = 0; /* dummy offset for null seed context */ u8 name[SHA256_DIGEST_SIZE + 2]; + u32 tmp_null_key; + int rc; rc = tpm2_load_context(chip, chip->null_key_context, &offset, - null_key); - if (rc != -EINVAL) - return rc; + &tmp_null_key); + if (rc != -EINVAL) { + if (!rc) + *null_key = tmp_null_key; + goto err; + } - /* an integrity failure may mean the TPM has been reset */ - dev_err(&chip->dev, "NULL key integrity failure!\n"); - /* check the null name against what we know */ - tpm2_create_primary(chip, TPM2_RH_NULL, NULL, name); - if (memcmp(name, chip->null_key_name, sizeof(name)) == 0) - /* name unchanged, assume transient integrity failure */ - return rc; - /* - * Fatal TPM failure: the NULL seed has actually changed, so - * the TPM must have been illegally reset. All in-kernel TPM - * operations will fail because the NULL primary can't be - * loaded to salt the sessions, but disable the TPM anyway so - * userspace programmes can't be compromised by it. - */ - dev_err(&chip->dev, "NULL name has changed, disabling TPM due to interference\n"); + /* Try to re-create null key, given the integrity failure: */ + rc = tpm2_create_primary(chip, TPM2_RH_NULL, &tmp_null_key, name); + if (rc) + goto err; + + /* Return null key if the name has not been changed: */ + if (!memcmp(name, chip->null_key_name, sizeof(name))) { + *null_key = tmp_null_key; + return 0; + } + + /* Deduce from the name change TPM interference: */ + dev_err(&chip->dev, "null key integrity check failed\n"); + tpm2_flush_context(chip, tmp_null_key); chip->flags |= TPM_CHIP_FLAG_DISABLE; - return rc; +err: + return rc ? -ENODEV : 0; } /** @@ -958,16 +968,20 @@ static int tpm2_load_null(struct tpm_chip *chip, u32 *null_key) */ int tpm2_start_auth_session(struct tpm_chip *chip) { + struct tpm2_auth *auth; struct tpm_buf buf; - struct tpm2_auth *auth = chip->auth; - int rc; u32 null_key; + int rc; - if (!auth) { - dev_warn_once(&chip->dev, "auth session is not active\n"); + if (chip->auth) { + dev_warn_once(&chip->dev, "auth session is active\n"); return 0; } + auth = kzalloc(sizeof(*auth), GFP_KERNEL); + if (!auth) + return -ENOMEM; + rc = tpm2_load_null(chip, &null_key); if (rc) goto out; @@ -988,7 +1002,7 @@ int tpm2_start_auth_session(struct tpm_chip *chip) tpm_buf_append(&buf, auth->our_nonce, sizeof(auth->our_nonce)); /* append encrypted salt and squirrel away unencrypted in auth */ - tpm_buf_append_salt(&buf, chip); + tpm_buf_append_salt(&buf, chip, auth); /* session type (HMAC, audit or policy) */ tpm_buf_append_u8(&buf, TPM2_SE_HMAC); @@ -1010,10 +1024,13 @@ int tpm2_start_auth_session(struct tpm_chip *chip) tpm_buf_destroy(&buf); - if (rc) - goto out; + if (rc == TPM2_RC_SUCCESS) { + chip->auth = auth; + return 0; + } - out: +out: + kfree_sensitive(auth); return rc; } EXPORT_SYMBOL(tpm2_start_auth_session); @@ -1347,18 +1364,21 @@ static int tpm2_create_null_primary(struct tpm_chip *chip) * * Derive and context save the null primary and allocate memory in the * struct tpm_chip for the authorizations. + * + * Return: + * * 0 - OK + * * -errno - A system error + * * TPM_RC - A TPM error */ int tpm2_sessions_init(struct tpm_chip *chip) { int rc; rc = tpm2_create_null_primary(chip); - if (rc) - dev_err(&chip->dev, "TPM: security failed (NULL seed derivation): %d\n", rc); - - chip->auth = kmalloc(sizeof(*chip->auth), GFP_KERNEL); - if (!chip->auth) - return -ENOMEM; + if (rc) { + dev_err(&chip->dev, "null key creation failed with %d\n", rc); + return rc; + } return rc; } diff --git a/drivers/char/tpm/tpm2-space.c b/drivers/char/tpm/tpm2-space.c index 25a66870c165..60354cd53b5c 100644 --- a/drivers/char/tpm/tpm2-space.c +++ b/drivers/char/tpm/tpm2-space.c @@ -12,7 +12,7 @@ */ #include -#include +#include #include "tpm.h" enum tpm2_handle_types { diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c index 99a7f2441e70..c62b208b42f1 100644 --- a/drivers/char/virtio_console.c +++ b/drivers/char/virtio_console.c @@ -2006,25 +2006,27 @@ static int virtcons_probe(struct virtio_device *vdev) multiport = true; } - err = init_vqs(portdev); - if (err < 0) { - dev_err(&vdev->dev, "Error %d initializing vqs\n", err); - goto free_chrdev; - } - spin_lock_init(&portdev->ports_lock); INIT_LIST_HEAD(&portdev->ports); INIT_LIST_HEAD(&portdev->list); - virtio_device_ready(portdev->vdev); - INIT_WORK(&portdev->config_work, &config_work_handler); INIT_WORK(&portdev->control_work, &control_work_handler); if (multiport) { spin_lock_init(&portdev->c_ivq_lock); spin_lock_init(&portdev->c_ovq_lock); + } + err = init_vqs(portdev); + if (err < 0) { + dev_err(&vdev->dev, "Error %d initializing vqs\n", err); + goto free_chrdev; + } + + virtio_device_ready(portdev->vdev); + + if (multiport) { err = fill_queue(portdev->c_ivq, &portdev->c_ivq_lock); if (err < 0) { dev_err(&vdev->dev, diff --git a/drivers/clk/clk-si5341.c b/drivers/clk/clk-si5341.c index 6e8dd7387cfd..5004888c7eca 100644 --- a/drivers/clk/clk-si5341.c +++ b/drivers/clk/clk-si5341.c @@ -21,7 +21,7 @@ #include #include #include -#include +#include #define SI5341_NUM_INPUTS 4 diff --git a/drivers/clk/clk_test.c b/drivers/clk/clk_test.c index 41fc8eba3418..aa3ddcfc00eb 100644 --- a/drivers/clk/clk_test.c +++ b/drivers/clk/clk_test.c @@ -473,7 +473,7 @@ clk_multiple_parents_mux_test_init(struct kunit *test) &clk_dummy_rate_ops, 0); ctx->parents_ctx[0].rate = DUMMY_CLOCK_RATE_1; - ret = clk_hw_register(NULL, &ctx->parents_ctx[0].hw); + ret = clk_hw_register_kunit(test, NULL, &ctx->parents_ctx[0].hw); if (ret) return ret; @@ -481,7 +481,7 @@ clk_multiple_parents_mux_test_init(struct kunit *test) &clk_dummy_rate_ops, 0); ctx->parents_ctx[1].rate = DUMMY_CLOCK_RATE_2; - ret = clk_hw_register(NULL, &ctx->parents_ctx[1].hw); + ret = clk_hw_register_kunit(test, NULL, &ctx->parents_ctx[1].hw); if (ret) return ret; @@ -489,23 +489,13 @@ clk_multiple_parents_mux_test_init(struct kunit *test) ctx->hw.init = CLK_HW_INIT_PARENTS("test-mux", parents, &clk_multiple_parents_mux_ops, CLK_SET_RATE_PARENT); - ret = clk_hw_register(NULL, &ctx->hw); + ret = clk_hw_register_kunit(test, NULL, &ctx->hw); if (ret) return ret; return 0; } -static void -clk_multiple_parents_mux_test_exit(struct kunit *test) -{ - struct clk_multiple_parent_ctx *ctx = test->priv; - - clk_hw_unregister(&ctx->hw); - clk_hw_unregister(&ctx->parents_ctx[0].hw); - clk_hw_unregister(&ctx->parents_ctx[1].hw); -} - /* * Test that for a clock with multiple parents, clk_get_parent() * actually returns the current one. @@ -561,18 +551,18 @@ clk_test_multiple_parents_mux_set_range_set_parent_get_rate(struct kunit *test) { struct clk_multiple_parent_ctx *ctx = test->priv; struct clk_hw *hw = &ctx->hw; - struct clk *clk = clk_hw_get_clk(hw, NULL); + struct clk *clk = clk_hw_get_clk_kunit(test, hw, NULL); struct clk *parent1, *parent2; unsigned long rate; int ret; kunit_skip(test, "This needs to be fixed in the core."); - parent1 = clk_hw_get_clk(&ctx->parents_ctx[0].hw, NULL); + parent1 = clk_hw_get_clk_kunit(test, &ctx->parents_ctx[0].hw, NULL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, parent1); KUNIT_ASSERT_TRUE(test, clk_is_match(clk_get_parent(clk), parent1)); - parent2 = clk_hw_get_clk(&ctx->parents_ctx[1].hw, NULL); + parent2 = clk_hw_get_clk_kunit(test, &ctx->parents_ctx[1].hw, NULL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, parent2); ret = clk_set_rate(parent1, DUMMY_CLOCK_RATE_1); @@ -593,10 +583,6 @@ clk_test_multiple_parents_mux_set_range_set_parent_get_rate(struct kunit *test) KUNIT_ASSERT_GT(test, rate, 0); KUNIT_EXPECT_GE(test, rate, DUMMY_CLOCK_RATE_1 - 1000); KUNIT_EXPECT_LE(test, rate, DUMMY_CLOCK_RATE_1 + 1000); - - clk_put(parent2); - clk_put(parent1); - clk_put(clk); } static struct kunit_case clk_multiple_parents_mux_test_cases[] = { @@ -617,7 +603,6 @@ static struct kunit_suite clk_multiple_parents_mux_test_suite = { .name = "clk-multiple-parents-mux-test", .init = clk_multiple_parents_mux_test_init, - .exit = clk_multiple_parents_mux_test_exit, .test_cases = clk_multiple_parents_mux_test_cases, }; @@ -637,29 +622,20 @@ clk_orphan_transparent_multiple_parent_mux_test_init(struct kunit *test) &clk_dummy_rate_ops, 0); ctx->parents_ctx[1].rate = DUMMY_CLOCK_INIT_RATE; - ret = clk_hw_register(NULL, &ctx->parents_ctx[1].hw); + ret = clk_hw_register_kunit(test, NULL, &ctx->parents_ctx[1].hw); if (ret) return ret; ctx->hw.init = CLK_HW_INIT_PARENTS("test-orphan-mux", parents, &clk_multiple_parents_mux_ops, CLK_SET_RATE_PARENT); - ret = clk_hw_register(NULL, &ctx->hw); + ret = clk_hw_register_kunit(test, NULL, &ctx->hw); if (ret) return ret; return 0; } -static void -clk_orphan_transparent_multiple_parent_mux_test_exit(struct kunit *test) -{ - struct clk_multiple_parent_ctx *ctx = test->priv; - - clk_hw_unregister(&ctx->hw); - clk_hw_unregister(&ctx->parents_ctx[1].hw); -} - /* * Test that, for a mux whose current parent hasn't been registered yet and is * thus orphan, clk_get_parent() will return NULL. @@ -912,7 +888,7 @@ clk_test_orphan_transparent_multiple_parent_mux_set_range_set_parent_get_rate(st { struct clk_multiple_parent_ctx *ctx = test->priv; struct clk_hw *hw = &ctx->hw; - struct clk *clk = clk_hw_get_clk(hw, NULL); + struct clk *clk = clk_hw_get_clk_kunit(test, hw, NULL); struct clk *parent; unsigned long rate; int ret; @@ -921,7 +897,7 @@ clk_test_orphan_transparent_multiple_parent_mux_set_range_set_parent_get_rate(st clk_hw_set_rate_range(hw, DUMMY_CLOCK_RATE_1, DUMMY_CLOCK_RATE_2); - parent = clk_hw_get_clk(&ctx->parents_ctx[1].hw, NULL); + parent = clk_hw_get_clk_kunit(test, &ctx->parents_ctx[1].hw, NULL); KUNIT_ASSERT_NOT_ERR_OR_NULL(test, parent); ret = clk_set_parent(clk, parent); @@ -931,9 +907,6 @@ clk_test_orphan_transparent_multiple_parent_mux_set_range_set_parent_get_rate(st KUNIT_ASSERT_GT(test, rate, 0); KUNIT_EXPECT_GE(test, rate, DUMMY_CLOCK_RATE_1); KUNIT_EXPECT_LE(test, rate, DUMMY_CLOCK_RATE_2); - - clk_put(parent); - clk_put(clk); } static struct kunit_case clk_orphan_transparent_multiple_parent_mux_test_cases[] = { @@ -961,7 +934,6 @@ static struct kunit_case clk_orphan_transparent_multiple_parent_mux_test_cases[] static struct kunit_suite clk_orphan_transparent_multiple_parent_mux_test_suite = { .name = "clk-orphan-transparent-multiple-parent-mux-test", .init = clk_orphan_transparent_multiple_parent_mux_test_init, - .exit = clk_orphan_transparent_multiple_parent_mux_test_exit, .test_cases = clk_orphan_transparent_multiple_parent_mux_test_cases, }; @@ -986,7 +958,7 @@ static int clk_single_parent_mux_test_init(struct kunit *test) &clk_dummy_rate_ops, 0); - ret = clk_hw_register(NULL, &ctx->parent_ctx.hw); + ret = clk_hw_register_kunit(test, NULL, &ctx->parent_ctx.hw); if (ret) return ret; @@ -994,7 +966,7 @@ static int clk_single_parent_mux_test_init(struct kunit *test) &clk_dummy_single_parent_ops, CLK_SET_RATE_PARENT); - ret = clk_hw_register(NULL, &ctx->hw); + ret = clk_hw_register_kunit(test, NULL, &ctx->hw); if (ret) return ret; @@ -1060,7 +1032,7 @@ clk_test_single_parent_mux_set_range_disjoint_child_last(struct kunit *test) { struct clk_single_parent_ctx *ctx = test->priv; struct clk_hw *hw = &ctx->hw; - struct clk *clk = clk_hw_get_clk(hw, NULL); + struct clk *clk = clk_hw_get_clk_kunit(test, hw, NULL); struct clk *parent; int ret; @@ -1074,8 +1046,6 @@ clk_test_single_parent_mux_set_range_disjoint_child_last(struct kunit *test) ret = clk_set_rate_range(clk, 3000, 4000); KUNIT_EXPECT_LT(test, ret, 0); - - clk_put(clk); } /* @@ -1092,7 +1062,7 @@ clk_test_single_parent_mux_set_range_disjoint_parent_last(struct kunit *test) { struct clk_single_parent_ctx *ctx = test->priv; struct clk_hw *hw = &ctx->hw; - struct clk *clk = clk_hw_get_clk(hw, NULL); + struct clk *clk = clk_hw_get_clk_kunit(test, hw, NULL); struct clk *parent; int ret; @@ -1106,8 +1076,6 @@ clk_test_single_parent_mux_set_range_disjoint_parent_last(struct kunit *test) ret = clk_set_rate_range(parent, 3000, 4000); KUNIT_EXPECT_LT(test, ret, 0); - - clk_put(clk); } /* @@ -1238,7 +1206,6 @@ static struct kunit_suite clk_single_parent_mux_test_suite = { .name = "clk-single-parent-mux-test", .init = clk_single_parent_mux_test_init, - .exit = clk_single_parent_mux_test_exit, .test_cases = clk_single_parent_mux_test_cases, }; diff --git a/drivers/clk/rockchip/clk.c b/drivers/clk/rockchip/clk.c index 2fa7253c73b2..88629a9abc9c 100644 --- a/drivers/clk/rockchip/clk.c +++ b/drivers/clk/rockchip/clk.c @@ -439,7 +439,7 @@ unsigned long rockchip_clk_find_max_clk_id(struct rockchip_clk_branch *list, if (list->id > max) max = list->id; if (list->child && list->child->id > max) - max = list->id; + max = list->child->id; } return max; diff --git a/drivers/clk/samsung/clk-exynosautov920.c b/drivers/clk/samsung/clk-exynosautov920.c index 7ba9748c0526..f60f0a0c598d 100644 --- a/drivers/clk/samsung/clk-exynosautov920.c +++ b/drivers/clk/samsung/clk-exynosautov920.c @@ -1155,6 +1155,7 @@ static const struct of_device_id exynosautov920_cmu_of_match[] = { .compatible = "samsung,exynosautov920-cmu-peric0", .data = &peric0_cmu_info, }, + { } }; static struct platform_driver exynosautov920_cmu_driver __refdata = { diff --git a/drivers/comedi/drivers/usbduxsigma.c b/drivers/comedi/drivers/usbduxsigma.c index 2aaeaf44fbe5..3f215ae228b2 100644 --- a/drivers/comedi/drivers/usbduxsigma.c +++ b/drivers/comedi/drivers/usbduxsigma.c @@ -39,7 +39,7 @@ #include #include #include -#include +#include #include /* timeout for the USB-transfer in ms*/ diff --git a/drivers/counter/104-quad-8.c b/drivers/counter/104-quad-8.c index ed1f57511955..4a6868b8f58b 100644 --- a/drivers/counter/104-quad-8.c +++ b/drivers/counter/104-quad-8.c @@ -22,7 +22,7 @@ #include #include -#include +#include #define QUAD8_EXTENT 32 diff --git a/drivers/counter/i8254.c b/drivers/counter/i8254.c index c41e4fdc9601..6d74e8ef92f0 100644 --- a/drivers/counter/i8254.c +++ b/drivers/counter/i8254.c @@ -15,7 +15,7 @@ #include #include -#include +#include #define I8254_COUNTER_REG(_counter) (_counter) #define I8254_CONTROL_REG 0x3 diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index 15e201d5e911..b63863f77c67 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -536,11 +536,16 @@ static int amd_pstate_verify(struct cpufreq_policy_data *policy) static int amd_pstate_update_min_max_limit(struct cpufreq_policy *policy) { - u32 max_limit_perf, min_limit_perf, lowest_perf; + u32 max_limit_perf, min_limit_perf, lowest_perf, max_perf; struct amd_cpudata *cpudata = policy->driver_data; - max_limit_perf = div_u64(policy->max * cpudata->highest_perf, cpudata->max_freq); - min_limit_perf = div_u64(policy->min * cpudata->highest_perf, cpudata->max_freq); + if (cpudata->boost_supported && !policy->boost_enabled) + max_perf = READ_ONCE(cpudata->nominal_perf); + else + max_perf = READ_ONCE(cpudata->highest_perf); + + max_limit_perf = div_u64(policy->max * max_perf, policy->cpuinfo.max_freq); + min_limit_perf = div_u64(policy->min * max_perf, policy->cpuinfo.max_freq); lowest_perf = READ_ONCE(cpudata->lowest_perf); if (min_limit_perf < lowest_perf) @@ -1201,11 +1206,21 @@ static int amd_pstate_register_driver(int mode) return -EINVAL; cppc_state = mode; + + ret = amd_pstate_enable(true); + if (ret) { + pr_err("failed to enable cppc during amd-pstate driver registration, return %d\n", + ret); + amd_pstate_driver_cleanup(); + return ret; + } + ret = cpufreq_register_driver(current_pstate_driver); if (ret) { amd_pstate_driver_cleanup(); return ret; } + return 0; } @@ -1496,10 +1511,13 @@ static int amd_pstate_epp_update_limit(struct cpufreq_policy *policy) u64 value; s16 epp; - max_perf = READ_ONCE(cpudata->highest_perf); + if (cpudata->boost_supported && !policy->boost_enabled) + max_perf = READ_ONCE(cpudata->nominal_perf); + else + max_perf = READ_ONCE(cpudata->highest_perf); min_perf = READ_ONCE(cpudata->lowest_perf); - max_limit_perf = div_u64(policy->max * cpudata->highest_perf, cpudata->max_freq); - min_limit_perf = div_u64(policy->min * cpudata->highest_perf, cpudata->max_freq); + max_limit_perf = div_u64(policy->max * max_perf, policy->cpuinfo.max_freq); + min_limit_perf = div_u64(policy->min * max_perf, policy->cpuinfo.max_freq); if (min_limit_perf < min_perf) min_limit_perf = min_perf; diff --git a/drivers/cpufreq/cppc_cpufreq.c b/drivers/cpufreq/cppc_cpufreq.c index 1a5ad184d28f..2b8708475ac7 100644 --- a/drivers/cpufreq/cppc_cpufreq.c +++ b/drivers/cpufreq/cppc_cpufreq.c @@ -22,7 +22,7 @@ #include #include -#include +#include #include diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index aaea9a39eced..b0018f371ea3 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -1845,7 +1845,7 @@ static void intel_pstate_notify_work(struct work_struct *work) wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_STATUS, 0); } -static DEFINE_SPINLOCK(hwp_notify_lock); +static DEFINE_RAW_SPINLOCK(hwp_notify_lock); static cpumask_t hwp_intr_enable_mask; #define HWP_GUARANTEED_PERF_CHANGE_STATUS BIT(0) @@ -1868,7 +1868,7 @@ void notify_hwp_interrupt(void) if (!(value & status_mask)) return; - spin_lock_irqsave(&hwp_notify_lock, flags); + raw_spin_lock_irqsave(&hwp_notify_lock, flags); if (!cpumask_test_cpu(this_cpu, &hwp_intr_enable_mask)) goto ack_intr; @@ -1876,13 +1876,13 @@ void notify_hwp_interrupt(void) schedule_delayed_work(&all_cpu_data[this_cpu]->hwp_notify_work, msecs_to_jiffies(10)); - spin_unlock_irqrestore(&hwp_notify_lock, flags); + raw_spin_unlock_irqrestore(&hwp_notify_lock, flags); return; ack_intr: wrmsrl_safe(MSR_HWP_STATUS, 0); - spin_unlock_irqrestore(&hwp_notify_lock, flags); + raw_spin_unlock_irqrestore(&hwp_notify_lock, flags); } static void intel_pstate_disable_hwp_interrupt(struct cpudata *cpudata) @@ -1895,9 +1895,9 @@ static void intel_pstate_disable_hwp_interrupt(struct cpudata *cpudata) /* wrmsrl_on_cpu has to be outside spinlock as this can result in IPC */ wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_INTERRUPT, 0x00); - spin_lock_irq(&hwp_notify_lock); + raw_spin_lock_irq(&hwp_notify_lock); cancel_work = cpumask_test_and_clear_cpu(cpudata->cpu, &hwp_intr_enable_mask); - spin_unlock_irq(&hwp_notify_lock); + raw_spin_unlock_irq(&hwp_notify_lock); if (cancel_work) cancel_delayed_work_sync(&cpudata->hwp_notify_work); @@ -1912,10 +1912,10 @@ static void intel_pstate_enable_hwp_interrupt(struct cpudata *cpudata) if (boot_cpu_has(X86_FEATURE_HWP_NOTIFY)) { u64 interrupt_mask = HWP_GUARANTEED_PERF_CHANGE_REQ; - spin_lock_irq(&hwp_notify_lock); + raw_spin_lock_irq(&hwp_notify_lock); INIT_DELAYED_WORK(&cpudata->hwp_notify_work, intel_pstate_notify_work); cpumask_set_cpu(cpudata->cpu, &hwp_intr_enable_mask); - spin_unlock_irq(&hwp_notify_lock); + raw_spin_unlock_irq(&hwp_notify_lock); if (cpu_feature_enabled(X86_FEATURE_HWP_HIGHEST_PERF_CHANGE)) interrupt_mask |= HWP_HIGHEST_PERF_CHANGE_REQ; diff --git a/drivers/crypto/allwinner/sun4i-ss/sun4i-ss-hash.c b/drivers/crypto/allwinner/sun4i-ss/sun4i-ss-hash.c index f7893e4ac59d..434f2b271012 100644 --- a/drivers/crypto/allwinner/sun4i-ss/sun4i-ss-hash.c +++ b/drivers/crypto/allwinner/sun4i-ss/sun4i-ss-hash.c @@ -9,7 +9,7 @@ * You could find the datasheet in Documentation/arch/arm/sunxi.rst */ #include "sun4i-ss.h" -#include +#include #include /* This is a totally arbitrary value */ diff --git a/drivers/crypto/caam/caamalg.c b/drivers/crypto/caam/caamalg.c index 066f08a3a040..2cfb1b8d8c7c 100644 --- a/drivers/crypto/caam/caamalg.c +++ b/drivers/crypto/caam/caamalg.c @@ -56,7 +56,7 @@ #include "sg_sw_sec4.h" #include "key_gen.h" #include "caamalg_desc.h" -#include +#include #include #include #include diff --git a/drivers/crypto/caam/caamalg_qi.c b/drivers/crypto/caam/caamalg_qi.c index 13347dfecf7a..65f6adb6c673 100644 --- a/drivers/crypto/caam/caamalg_qi.c +++ b/drivers/crypto/caam/caamalg_qi.c @@ -19,7 +19,7 @@ #include "jr.h" #include "caamalg_desc.h" #include -#include +#include #include #include #include diff --git a/drivers/crypto/caam/caamalg_qi2.c b/drivers/crypto/caam/caamalg_qi2.c index 44e1f8f46967..e809d030ab11 100644 --- a/drivers/crypto/caam/caamalg_qi2.c +++ b/drivers/crypto/caam/caamalg_qi2.c @@ -22,7 +22,7 @@ #include #include #include -#include +#include #define CAAM_CRA_PRIORITY 2000 diff --git a/drivers/crypto/inside-secure/safexcel_cipher.c b/drivers/crypto/inside-secure/safexcel_cipher.c index 42677f7458b7..919e5a2cab95 100644 --- a/drivers/crypto/inside-secure/safexcel_cipher.c +++ b/drivers/crypto/inside-secure/safexcel_cipher.c @@ -5,7 +5,7 @@ * Antoine Tenart */ -#include +#include #include #include #include diff --git a/drivers/crypto/marvell/cesa/hash.c b/drivers/crypto/marvell/cesa/hash.c index 8d84ad45571c..f150861ceaf6 100644 --- a/drivers/crypto/marvell/cesa/hash.c +++ b/drivers/crypto/marvell/cesa/hash.c @@ -947,7 +947,7 @@ struct ahash_alg mv_md5_alg = { .base = { .cra_name = "md5", .cra_driver_name = "mv-md5", - .cra_priority = 300, + .cra_priority = 0, .cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY | CRYPTO_ALG_KERN_DRIVER_ONLY, @@ -1018,7 +1018,7 @@ struct ahash_alg mv_sha1_alg = { .base = { .cra_name = "sha1", .cra_driver_name = "mv-sha1", - .cra_priority = 300, + .cra_priority = 0, .cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY | CRYPTO_ALG_KERN_DRIVER_ONLY, @@ -1092,7 +1092,7 @@ struct ahash_alg mv_sha256_alg = { .base = { .cra_name = "sha256", .cra_driver_name = "mv-sha256", - .cra_priority = 300, + .cra_priority = 0, .cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY | CRYPTO_ALG_KERN_DRIVER_ONLY, @@ -1302,7 +1302,7 @@ struct ahash_alg mv_ahmac_md5_alg = { .base = { .cra_name = "hmac(md5)", .cra_driver_name = "mv-hmac-md5", - .cra_priority = 300, + .cra_priority = 0, .cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY | CRYPTO_ALG_KERN_DRIVER_ONLY, @@ -1373,7 +1373,7 @@ struct ahash_alg mv_ahmac_sha1_alg = { .base = { .cra_name = "hmac(sha1)", .cra_driver_name = "mv-hmac-sha1", - .cra_priority = 300, + .cra_priority = 0, .cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY | CRYPTO_ALG_KERN_DRIVER_ONLY, @@ -1444,7 +1444,7 @@ struct ahash_alg mv_ahmac_sha256_alg = { .base = { .cra_name = "hmac(sha256)", .cra_driver_name = "mv-hmac-sha256", - .cra_priority = 300, + .cra_priority = 0, .cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY | CRYPTO_ALG_KERN_DRIVER_ONLY, diff --git a/drivers/crypto/rockchip/rk3288_crypto_ahash.c b/drivers/crypto/rockchip/rk3288_crypto_ahash.c index a235e6c300f1..69d6019d8abc 100644 --- a/drivers/crypto/rockchip/rk3288_crypto_ahash.c +++ b/drivers/crypto/rockchip/rk3288_crypto_ahash.c @@ -9,7 +9,7 @@ * Some ideas are from marvell/cesa.c and s5p-sss.c driver. */ -#include +#include #include #include #include diff --git a/drivers/crypto/stm32/stm32-crc32.c b/drivers/crypto/stm32/stm32-crc32.c index b0cf6d2fd352..e0faddbf8990 100644 --- a/drivers/crypto/stm32/stm32-crc32.c +++ b/drivers/crypto/stm32/stm32-crc32.c @@ -17,7 +17,7 @@ #include -#include +#include #define DRIVER_NAME "stm32-crc32" #define CHKSUM_DIGEST_SIZE 4 diff --git a/drivers/cxl/Kconfig b/drivers/cxl/Kconfig index 29c192f20082..876469e23f7a 100644 --- a/drivers/cxl/Kconfig +++ b/drivers/cxl/Kconfig @@ -60,6 +60,7 @@ config CXL_ACPI default CXL_BUS select ACPI_TABLE_LIB select ACPI_HMAT + select CXL_PORT help Enable support for host managed device memory (HDM) resources published by a platform's ACPI CXL memory layout description. See diff --git a/drivers/cxl/Makefile b/drivers/cxl/Makefile index db321f48ba52..2caa90fa4bf2 100644 --- a/drivers/cxl/Makefile +++ b/drivers/cxl/Makefile @@ -1,13 +1,21 @@ # SPDX-License-Identifier: GPL-2.0 + +# Order is important here for the built-in case: +# - 'core' first for fundamental init +# - 'port' before platform root drivers like 'acpi' so that CXL-root ports +# are immediately enabled +# - 'mem' and 'pmem' before endpoint drivers so that memdevs are +# immediately enabled +# - 'pci' last, also mirrors the hardware enumeration hierarchy obj-y += core/ -obj-$(CONFIG_CXL_PCI) += cxl_pci.o -obj-$(CONFIG_CXL_MEM) += cxl_mem.o +obj-$(CONFIG_CXL_PORT) += cxl_port.o obj-$(CONFIG_CXL_ACPI) += cxl_acpi.o obj-$(CONFIG_CXL_PMEM) += cxl_pmem.o -obj-$(CONFIG_CXL_PORT) += cxl_port.o +obj-$(CONFIG_CXL_MEM) += cxl_mem.o +obj-$(CONFIG_CXL_PCI) += cxl_pci.o -cxl_mem-y := mem.o -cxl_pci-y := pci.o +cxl_port-y := port.o cxl_acpi-y := acpi.o cxl_pmem-y := pmem.o security.o -cxl_port-y := port.o +cxl_mem-y := mem.o +cxl_pci-y := pci.o diff --git a/drivers/cxl/acpi.c b/drivers/cxl/acpi.c index 82b78e331d8e..432b7cfd12a8 100644 --- a/drivers/cxl/acpi.c +++ b/drivers/cxl/acpi.c @@ -924,6 +924,13 @@ static void __exit cxl_acpi_exit(void) /* load before dax_hmem sees 'Soft Reserved' CXL ranges */ subsys_initcall(cxl_acpi_init); + +/* + * Arrange for host-bridge ports to be active synchronous with + * cxl_acpi_probe() exit. + */ +MODULE_SOFTDEP("pre: cxl_port"); + module_exit(cxl_acpi_exit); MODULE_DESCRIPTION("CXL ACPI: Platform Support"); MODULE_LICENSE("GPL v2"); diff --git a/drivers/cxl/core/cdat.c b/drivers/cxl/core/cdat.c index ef1621d40f05..e9cd7939c407 100644 --- a/drivers/cxl/core/cdat.c +++ b/drivers/cxl/core/cdat.c @@ -641,6 +641,9 @@ static int cxl_endpoint_gather_bandwidth(struct cxl_region *cxlr, void *ptr; int rc; + if (!dev_is_pci(cxlds->dev)) + return -ENODEV; + if (cxlds->rcd) return -ENODEV; diff --git a/drivers/cxl/core/hdm.c b/drivers/cxl/core/hdm.c index 3df10517a327..223c273c0cd1 100644 --- a/drivers/cxl/core/hdm.c +++ b/drivers/cxl/core/hdm.c @@ -712,7 +712,44 @@ static int cxl_decoder_commit(struct cxl_decoder *cxld) return 0; } -static int cxl_decoder_reset(struct cxl_decoder *cxld) +static int commit_reap(struct device *dev, const void *data) +{ + struct cxl_port *port = to_cxl_port(dev->parent); + struct cxl_decoder *cxld; + + if (!is_switch_decoder(dev) && !is_endpoint_decoder(dev)) + return 0; + + cxld = to_cxl_decoder(dev); + if (port->commit_end == cxld->id && + ((cxld->flags & CXL_DECODER_F_ENABLE) == 0)) { + port->commit_end--; + dev_dbg(&port->dev, "reap: %s commit_end: %d\n", + dev_name(&cxld->dev), port->commit_end); + } + + return 0; +} + +void cxl_port_commit_reap(struct cxl_decoder *cxld) +{ + struct cxl_port *port = to_cxl_port(cxld->dev.parent); + + lockdep_assert_held_write(&cxl_region_rwsem); + + /* + * Once the highest committed decoder is disabled, free any other + * decoders that were pinned allocated by out-of-order release. + */ + port->commit_end--; + dev_dbg(&port->dev, "reap: %s commit_end: %d\n", dev_name(&cxld->dev), + port->commit_end); + device_for_each_child_reverse_from(&port->dev, &cxld->dev, NULL, + commit_reap); +} +EXPORT_SYMBOL_NS_GPL(cxl_port_commit_reap, CXL); + +static void cxl_decoder_reset(struct cxl_decoder *cxld) { struct cxl_port *port = to_cxl_port(cxld->dev.parent); struct cxl_hdm *cxlhdm = dev_get_drvdata(&port->dev); @@ -721,14 +758,14 @@ static int cxl_decoder_reset(struct cxl_decoder *cxld) u32 ctrl; if ((cxld->flags & CXL_DECODER_F_ENABLE) == 0) - return 0; + return; - if (port->commit_end != id) { + if (port->commit_end == id) + cxl_port_commit_reap(cxld); + else dev_dbg(&port->dev, "%s: out of order reset, expected decoder%d.%d\n", dev_name(&cxld->dev), port->id, port->commit_end); - return -EBUSY; - } down_read(&cxl_dpa_rwsem); ctrl = readl(hdm + CXL_HDM_DECODER0_CTRL_OFFSET(id)); @@ -741,7 +778,6 @@ static int cxl_decoder_reset(struct cxl_decoder *cxld) writel(0, hdm + CXL_HDM_DECODER0_BASE_LOW_OFFSET(id)); up_read(&cxl_dpa_rwsem); - port->commit_end--; cxld->flags &= ~CXL_DECODER_F_ENABLE; /* Userspace is now responsible for reconfiguring this decoder */ @@ -751,8 +787,6 @@ static int cxl_decoder_reset(struct cxl_decoder *cxld) cxled = to_cxl_endpoint_decoder(&cxld->dev); cxled->state = CXL_DECODER_STATE_MANUAL; } - - return 0; } static int cxl_setup_hdm_decoder_from_dvsec( diff --git a/drivers/cxl/core/mbox.c b/drivers/cxl/core/mbox.c index 946f8e44455f..5175138c4fb7 100644 --- a/drivers/cxl/core/mbox.c +++ b/drivers/cxl/core/mbox.c @@ -4,7 +4,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/drivers/cxl/core/port.c b/drivers/cxl/core/port.c index e666ec6a9085..af92c67bc954 100644 --- a/drivers/cxl/core/port.c +++ b/drivers/cxl/core/port.c @@ -2084,11 +2084,18 @@ static void cxl_bus_remove(struct device *dev) static struct workqueue_struct *cxl_bus_wq; +static int cxl_rescan_attach(struct device *dev, void *data) +{ + int rc = device_attach(dev); + + dev_vdbg(dev, "rescan: %s\n", rc ? "attach" : "detached"); + + return 0; +} + static void cxl_bus_rescan_queue(struct work_struct *w) { - int rc = bus_rescan_devices(&cxl_bus_type); - - pr_debug("CXL bus rescan result: %d\n", rc); + bus_for_each_dev(&cxl_bus_type, NULL, NULL, cxl_rescan_attach); } void cxl_bus_rescan(void) diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c index e701e4b04032..dff618c708dc 100644 --- a/drivers/cxl/core/region.c +++ b/drivers/cxl/core/region.c @@ -232,8 +232,8 @@ static int cxl_region_invalidate_memregion(struct cxl_region *cxlr) "Bypassing cpu_cache_invalidate_memregion() for testing!\n"); return 0; } else { - dev_err(&cxlr->dev, - "Failed to synchronize CPU cache state\n"); + dev_WARN(&cxlr->dev, + "Failed to synchronize CPU cache state\n"); return -ENXIO; } } @@ -242,19 +242,17 @@ static int cxl_region_invalidate_memregion(struct cxl_region *cxlr) return 0; } -static int cxl_region_decode_reset(struct cxl_region *cxlr, int count) +static void cxl_region_decode_reset(struct cxl_region *cxlr, int count) { struct cxl_region_params *p = &cxlr->params; - int i, rc = 0; + int i; /* - * Before region teardown attempt to flush, and if the flush - * fails cancel the region teardown for data consistency - * concerns + * Before region teardown attempt to flush, evict any data cached for + * this region, or scream loudly about missing arch / platform support + * for CXL teardown. */ - rc = cxl_region_invalidate_memregion(cxlr); - if (rc) - return rc; + cxl_region_invalidate_memregion(cxlr); for (i = count - 1; i >= 0; i--) { struct cxl_endpoint_decoder *cxled = p->targets[i]; @@ -277,23 +275,17 @@ static int cxl_region_decode_reset(struct cxl_region *cxlr, int count) cxl_rr = cxl_rr_load(iter, cxlr); cxld = cxl_rr->decoder; if (cxld->reset) - rc = cxld->reset(cxld); - if (rc) - return rc; + cxld->reset(cxld); set_bit(CXL_REGION_F_NEEDS_RESET, &cxlr->flags); } endpoint_reset: - rc = cxled->cxld.reset(&cxled->cxld); - if (rc) - return rc; + cxled->cxld.reset(&cxled->cxld); set_bit(CXL_REGION_F_NEEDS_RESET, &cxlr->flags); } /* all decoders associated with this region have been torn down */ clear_bit(CXL_REGION_F_NEEDS_RESET, &cxlr->flags); - - return 0; } static int commit_decoder(struct cxl_decoder *cxld) @@ -409,16 +401,8 @@ static ssize_t commit_store(struct device *dev, struct device_attribute *attr, * still pending. */ if (p->state == CXL_CONFIG_RESET_PENDING) { - rc = cxl_region_decode_reset(cxlr, p->interleave_ways); - /* - * Revert to committed since there may still be active - * decoders associated with this region, or move forward - * to active to mark the reset successful - */ - if (rc) - p->state = CXL_CONFIG_COMMIT; - else - p->state = CXL_CONFIG_ACTIVE; + cxl_region_decode_reset(cxlr, p->interleave_ways); + p->state = CXL_CONFIG_ACTIVE; } } @@ -794,26 +778,50 @@ out: return rc; } +static int check_commit_order(struct device *dev, const void *data) +{ + struct cxl_decoder *cxld = to_cxl_decoder(dev); + + /* + * if port->commit_end is not the only free decoder, then out of + * order shutdown has occurred, block further allocations until + * that is resolved + */ + if (((cxld->flags & CXL_DECODER_F_ENABLE) == 0)) + return -EBUSY; + return 0; +} + static int match_free_decoder(struct device *dev, void *data) { + struct cxl_port *port = to_cxl_port(dev->parent); struct cxl_decoder *cxld; - int *id = data; + int rc; if (!is_switch_decoder(dev)) return 0; cxld = to_cxl_decoder(dev); - /* enforce ordered allocation */ - if (cxld->id != *id) + if (cxld->id != port->commit_end + 1) return 0; - if (!cxld->region) - return 1; + if (cxld->region) { + dev_dbg(dev->parent, + "next decoder to commit (%s) is already reserved (%s)\n", + dev_name(dev), dev_name(&cxld->region->dev)); + return 0; + } - (*id)++; - - return 0; + rc = device_for_each_child_reverse_from(dev->parent, dev, NULL, + check_commit_order); + if (rc) { + dev_dbg(dev->parent, + "unable to allocate %s due to out of order shutdown\n", + dev_name(dev)); + return 0; + } + return 1; } static int match_auto_decoder(struct device *dev, void *data) @@ -840,7 +848,6 @@ cxl_region_find_decoder(struct cxl_port *port, struct cxl_region *cxlr) { struct device *dev; - int id = 0; if (port == cxled_to_port(cxled)) return &cxled->cxld; @@ -849,7 +856,7 @@ cxl_region_find_decoder(struct cxl_port *port, dev = device_find_child(&port->dev, &cxlr->params, match_auto_decoder); else - dev = device_find_child(&port->dev, &id, match_free_decoder); + dev = device_find_child(&port->dev, NULL, match_free_decoder); if (!dev) return NULL; /* @@ -2054,13 +2061,7 @@ static int cxl_region_detach(struct cxl_endpoint_decoder *cxled) get_device(&cxlr->dev); if (p->state > CXL_CONFIG_ACTIVE) { - /* - * TODO: tear down all impacted regions if a device is - * removed out of order - */ - rc = cxl_region_decode_reset(cxlr, p->interleave_ways); - if (rc) - goto out; + cxl_region_decode_reset(cxlr, p->interleave_ways); p->state = CXL_CONFIG_ACTIVE; } diff --git a/drivers/cxl/core/trace.h b/drivers/cxl/core/trace.h index 9167cfba7f59..8389a94adb1a 100644 --- a/drivers/cxl/core/trace.h +++ b/drivers/cxl/core/trace.h @@ -8,7 +8,7 @@ #include #include -#include +#include #include #include @@ -279,7 +279,7 @@ TRACE_EVENT(cxl_generic_event, #define CXL_GMER_MEM_EVT_TYPE_ECC_ERROR 0x00 #define CXL_GMER_MEM_EVT_TYPE_INV_ADDR 0x01 #define CXL_GMER_MEM_EVT_TYPE_DATA_PATH_ERROR 0x02 -#define show_mem_event_type(type) __print_symbolic(type, \ +#define show_gmer_mem_event_type(type) __print_symbolic(type, \ { CXL_GMER_MEM_EVT_TYPE_ECC_ERROR, "ECC Error" }, \ { CXL_GMER_MEM_EVT_TYPE_INV_ADDR, "Invalid Address" }, \ { CXL_GMER_MEM_EVT_TYPE_DATA_PATH_ERROR, "Data Path Error" } \ @@ -373,7 +373,7 @@ TRACE_EVENT(cxl_general_media, "hpa=%llx region=%s region_uuid=%pUb", __entry->dpa, show_dpa_flags(__entry->dpa_flags), show_event_desc_flags(__entry->descriptor), - show_mem_event_type(__entry->type), + show_gmer_mem_event_type(__entry->type), show_trans_type(__entry->transaction_type), __entry->channel, __entry->rank, __entry->device, __print_hex(__entry->comp_id, CXL_EVENT_GEN_MED_COMP_ID_SIZE), @@ -391,6 +391,17 @@ TRACE_EVENT(cxl_general_media, * DRAM Event Record defines many fields the same as the General Media Event * Record. Reuse those definitions as appropriate. */ +#define CXL_DER_MEM_EVT_TYPE_ECC_ERROR 0x00 +#define CXL_DER_MEM_EVT_TYPE_SCRUB_MEDIA_ECC_ERROR 0x01 +#define CXL_DER_MEM_EVT_TYPE_INV_ADDR 0x02 +#define CXL_DER_MEM_EVT_TYPE_DATA_PATH_ERROR 0x03 +#define show_dram_mem_event_type(type) __print_symbolic(type, \ + { CXL_DER_MEM_EVT_TYPE_ECC_ERROR, "ECC Error" }, \ + { CXL_DER_MEM_EVT_TYPE_SCRUB_MEDIA_ECC_ERROR, "Scrub Media ECC Error" }, \ + { CXL_DER_MEM_EVT_TYPE_INV_ADDR, "Invalid Address" }, \ + { CXL_DER_MEM_EVT_TYPE_DATA_PATH_ERROR, "Data Path Error" } \ +) + #define CXL_DER_VALID_CHANNEL BIT(0) #define CXL_DER_VALID_RANK BIT(1) #define CXL_DER_VALID_NIBBLE BIT(2) @@ -477,7 +488,7 @@ TRACE_EVENT(cxl_dram, "hpa=%llx region=%s region_uuid=%pUb", __entry->dpa, show_dpa_flags(__entry->dpa_flags), show_event_desc_flags(__entry->descriptor), - show_mem_event_type(__entry->type), + show_dram_mem_event_type(__entry->type), show_trans_type(__entry->transaction_type), __entry->channel, __entry->rank, __entry->nibble_mask, __entry->bank_group, __entry->bank, diff --git a/drivers/cxl/cxl.h b/drivers/cxl/cxl.h index 0d8b810a51f0..5406e3ab3d4a 100644 --- a/drivers/cxl/cxl.h +++ b/drivers/cxl/cxl.h @@ -359,7 +359,7 @@ struct cxl_decoder { struct cxl_region *region; unsigned long flags; int (*commit)(struct cxl_decoder *cxld); - int (*reset)(struct cxl_decoder *cxld); + void (*reset)(struct cxl_decoder *cxld); }; /* @@ -730,6 +730,7 @@ static inline bool is_cxl_root(struct cxl_port *port) int cxl_num_decoders_committed(struct cxl_port *port); bool is_cxl_port(const struct device *dev); struct cxl_port *to_cxl_port(const struct device *dev); +void cxl_port_commit_reap(struct cxl_decoder *cxld); struct pci_bus; int devm_cxl_register_pci_bus(struct device *host, struct device *uport_dev, struct pci_bus *bus); diff --git a/drivers/cxl/pci.c b/drivers/cxl/pci.c index 37164174b5fb..188412d45e0d 100644 --- a/drivers/cxl/pci.c +++ b/drivers/cxl/pci.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0-only /* Copyright(c) 2020 Intel Corporation. All rights reserved. */ -#include +#include #include #include #include diff --git a/drivers/cxl/pmem.c b/drivers/cxl/pmem.c index a6538a5f5c9f..d2d43a4fc053 100644 --- a/drivers/cxl/pmem.c +++ b/drivers/cxl/pmem.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0-only /* Copyright(c) 2021 Intel Corporation. All rights reserved. */ #include -#include +#include #include #include #include diff --git a/drivers/cxl/port.c b/drivers/cxl/port.c index 861dde65768f..9dc394295e1f 100644 --- a/drivers/cxl/port.c +++ b/drivers/cxl/port.c @@ -208,7 +208,22 @@ static struct cxl_driver cxl_port_driver = { }, }; -module_cxl_driver(cxl_port_driver); +static int __init cxl_port_init(void) +{ + return cxl_driver_register(&cxl_port_driver); +} +/* + * Be ready to immediately enable ports emitted by the platform CXL root + * (e.g. cxl_acpi) when CONFIG_CXL_PORT=y. + */ +subsys_initcall(cxl_port_init); + +static void __exit cxl_port_exit(void) +{ + cxl_driver_unregister(&cxl_port_driver); +} +module_exit(cxl_port_exit); + MODULE_DESCRIPTION("CXL: Port enumeration and services"); MODULE_LICENSE("GPL v2"); MODULE_IMPORT_NS(CXL); diff --git a/drivers/cxl/security.c b/drivers/cxl/security.c index 452d1a9b9148..ab793e8577c7 100644 --- a/drivers/cxl/security.c +++ b/drivers/cxl/security.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0-only /* Copyright(c) 2022 Intel Corporation. All rights reserved. */ #include -#include +#include #include #include #include diff --git a/drivers/dax/device.c b/drivers/dax/device.c index 9c1a729cd77e..6d74e62bbee0 100644 --- a/drivers/dax/device.c +++ b/drivers/dax/device.c @@ -86,7 +86,7 @@ static void dax_set_mapping(struct vm_fault *vmf, pfn_t pfn, nr_pages = 1; pgoff = linear_page_index(vmf->vma, - ALIGN(vmf->address, fault_size)); + ALIGN_DOWN(vmf->address, fault_size)); for (i = 0; i < nr_pages; i++) { struct page *page = pfn_to_page(pfn_t_to_pfn(pfn) + i); diff --git a/drivers/dma-buf/Kconfig b/drivers/dma-buf/Kconfig index b46eb8a552d7..fee04fdb0822 100644 --- a/drivers/dma-buf/Kconfig +++ b/drivers/dma-buf/Kconfig @@ -36,6 +36,7 @@ config UDMABUF depends on DMA_SHARED_BUFFER depends on MEMFD_CREATE || COMPILE_TEST depends on MMU + select VMAP_PFN help A driver to let userspace turn memfd regions into dma-bufs. Qemu can use this to create host dmabufs for guest framebuffers. diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c index 8892bc701a66..5ad0e9e2e1b9 100644 --- a/drivers/dma-buf/dma-buf.c +++ b/drivers/dma-buf/dma-buf.c @@ -176,8 +176,9 @@ static loff_t dma_buf_llseek(struct file *file, loff_t offset, int whence) dmabuf = file->private_data; /* only support discovering the end of the buffer, - but also allow SEEK_SET to maintain the idiomatic - SEEK_END(0), SEEK_CUR(0) pattern */ + * but also allow SEEK_SET to maintain the idiomatic + * SEEK_END(0), SEEK_CUR(0) pattern. + */ if (whence == SEEK_END) base = dmabuf->size; else if (whence == SEEK_SET) @@ -558,7 +559,7 @@ static struct file *dma_buf_getfile(size_t size, int flags) * Override ->i_ino with the unique and dmabuffs specific * value. */ - inode->i_ino = atomic64_add_return(1, &dmabuf_inode); + inode->i_ino = atomic64_inc_return(&dmabuf_inode); flags &= O_ACCMODE | O_NONBLOCK; file = alloc_file_pseudo(inode, dma_buf_mnt, "dmabuf", flags, &dma_buf_fops); @@ -782,13 +783,14 @@ static void mangle_sg_table(struct sg_table *sg_table) /* To catch abuse of the underlying struct page by importers mix * up the bits, but take care to preserve the low SG_ bits to * not corrupt the sgt. The mixing is undone in __unmap_dma_buf - * before passing the sgt back to the exporter. */ + * before passing the sgt back to the exporter. + */ for_each_sgtable_sg(sg_table, sg, i) sg->page_link ^= ~0xffUL; #endif } -static struct sg_table * __map_dma_buf(struct dma_buf_attachment *attach, +static struct sg_table *__map_dma_buf(struct dma_buf_attachment *attach, enum dma_data_direction direction) { struct sg_table *sg_table; @@ -1296,10 +1298,12 @@ EXPORT_SYMBOL_NS_GPL(dma_buf_move_notify, DMA_BUF); * vmap interface is introduced. Note that on very old 32-bit architectures * vmalloc space might be limited and result in vmap calls failing. * - * Interfaces:: + * Interfaces: * - * void \*dma_buf_vmap(struct dma_buf \*dmabuf, struct iosys_map \*map) - * void dma_buf_vunmap(struct dma_buf \*dmabuf, struct iosys_map \*map) + * .. code-block:: c + * + * void *dma_buf_vmap(struct dma_buf *dmabuf, struct iosys_map *map) + * void dma_buf_vunmap(struct dma_buf *dmabuf, struct iosys_map *map) * * The vmap call can fail if there is no vmap support in the exporter, or if * it runs out of vmalloc space. Note that the dma-buf layer keeps a reference @@ -1356,10 +1360,11 @@ EXPORT_SYMBOL_NS_GPL(dma_buf_move_notify, DMA_BUF); * enough, since adding interfaces to intercept pagefaults and allow pte * shootdowns would increase the complexity quite a bit. * - * Interface:: + * Interface: * - * int dma_buf_mmap(struct dma_buf \*, struct vm_area_struct \*, - * unsigned long); + * .. code-block:: c + * + * int dma_buf_mmap(struct dma_buf *, struct vm_area_struct *, unsigned long); * * If the importing subsystem simply provides a special-purpose mmap call to * set up a mapping in userspace, calling do_mmap with &dma_buf.file will @@ -1694,7 +1699,7 @@ static int dma_buf_init_debugfs(void) dma_buf_debugfs_dir = d; - d = debugfs_create_file("bufinfo", S_IRUGO, dma_buf_debugfs_dir, + d = debugfs_create_file("bufinfo", 0444, dma_buf_debugfs_dir, NULL, &dma_buf_debug_fops); if (IS_ERR(d)) { pr_debug("dma_buf: debugfs: failed to create node bufinfo\n"); diff --git a/drivers/dma-buf/dma-fence.c b/drivers/dma-buf/dma-fence.c index 0393a9bba3a8..f0cdd3e99d36 100644 --- a/drivers/dma-buf/dma-fence.c +++ b/drivers/dma-buf/dma-fence.c @@ -309,8 +309,8 @@ bool dma_fence_begin_signalling(void) if (in_atomic()) return true; - /* ... and non-recursive readlock */ - lock_acquire(&dma_fence_lockdep_map, 0, 0, 1, 1, NULL, _RET_IP_); + /* ... and non-recursive successful read_trylock */ + lock_acquire(&dma_fence_lockdep_map, 0, 1, 1, 1, NULL, _RET_IP_); return false; } @@ -341,7 +341,7 @@ void __dma_fence_might_wait(void) lock_map_acquire(&dma_fence_lockdep_map); lock_map_release(&dma_fence_lockdep_map); if (tmp) - lock_acquire(&dma_fence_lockdep_map, 0, 0, 1, 1, NULL, _THIS_IP_); + lock_acquire(&dma_fence_lockdep_map, 0, 1, 1, 1, NULL, _THIS_IP_); } #endif @@ -412,7 +412,7 @@ int dma_fence_signal_timestamp(struct dma_fence *fence, ktime_t timestamp) unsigned long flags; int ret; - if (!fence) + if (WARN_ON(!fence)) return -EINVAL; spin_lock_irqsave(fence->lock, flags); @@ -464,7 +464,7 @@ int dma_fence_signal(struct dma_fence *fence) int ret; bool tmp; - if (!fence) + if (WARN_ON(!fence)) return -EINVAL; tmp = dma_fence_begin_signalling(); diff --git a/drivers/dma-buf/heaps/cma_heap.c b/drivers/dma-buf/heaps/cma_heap.c index 93be88b805fe..9512d050563a 100644 --- a/drivers/dma-buf/heaps/cma_heap.c +++ b/drivers/dma-buf/heaps/cma_heap.c @@ -309,13 +309,13 @@ static struct dma_buf *cma_heap_allocate(struct dma_heap *heap, struct page *page = cma_pages; while (nr_clear_pages > 0) { - void *vaddr = kmap_atomic(page); + void *vaddr = kmap_local_page(page); memset(vaddr, 0, PAGE_SIZE); - kunmap_atomic(vaddr); + kunmap_local(vaddr); /* * Avoid wasting time zeroing memory if the process - * has been killed by by SIGKILL + * has been killed by SIGKILL. */ if (fatal_signal_pending(current)) goto free_cma; @@ -366,7 +366,7 @@ static const struct dma_heap_ops cma_heap_ops = { .allocate = cma_heap_allocate, }; -static int __add_cma_heap(struct cma *cma, void *data) +static int __init __add_cma_heap(struct cma *cma, void *data) { struct cma_heap *cma_heap; struct dma_heap_export_info exp_info; @@ -391,7 +391,7 @@ static int __add_cma_heap(struct cma *cma, void *data) return 0; } -static int add_default_cma_heap(void) +static int __init add_default_cma_heap(void) { struct cma *default_cma = dev_get_cma_area(NULL); int ret = 0; diff --git a/drivers/dma-buf/heaps/system_heap.c b/drivers/dma-buf/heaps/system_heap.c index d78cdb9d01e5..26d5dc89ea16 100644 --- a/drivers/dma-buf/heaps/system_heap.c +++ b/drivers/dma-buf/heaps/system_heap.c @@ -421,7 +421,7 @@ static const struct dma_heap_ops system_heap_ops = { .allocate = system_heap_allocate, }; -static int system_heap_create(void) +static int __init system_heap_create(void) { struct dma_heap_export_info exp_info; diff --git a/drivers/dma-buf/sw_sync.c b/drivers/dma-buf/sw_sync.c index c353029789cf..f5905d67dedb 100644 --- a/drivers/dma-buf/sw_sync.c +++ b/drivers/dma-buf/sw_sync.c @@ -173,11 +173,6 @@ static bool timeline_fence_signaled(struct dma_fence *fence) return !__dma_fence_is_later(fence->seqno, parent->value, fence->ops); } -static bool timeline_fence_enable_signaling(struct dma_fence *fence) -{ - return true; -} - static void timeline_fence_value_str(struct dma_fence *fence, char *str, int size) { @@ -211,7 +206,6 @@ static void timeline_fence_set_deadline(struct dma_fence *fence, ktime_t deadlin static const struct dma_fence_ops timeline_fence_ops = { .get_driver_name = timeline_fence_get_driver_name, .get_timeline_name = timeline_fence_get_timeline_name, - .enable_signaling = timeline_fence_enable_signaling, .signaled = timeline_fence_signaled, .release = timeline_fence_release, .fence_value_str = timeline_fence_value_str, diff --git a/drivers/dma-buf/udmabuf.c b/drivers/dma-buf/udmabuf.c index 047c3cd2ceff..8ce1f074c2d3 100644 --- a/drivers/dma-buf/udmabuf.c +++ b/drivers/dma-buf/udmabuf.c @@ -27,15 +27,21 @@ MODULE_PARM_DESC(size_limit_mb, "Max size of a dmabuf, in megabytes. Default is struct udmabuf { pgoff_t pagecount; struct folio **folios; + + /** + * Unlike folios, pinned_folios is only used for unpin. + * So, nr_pinned is not the same to pagecount, the pinned_folios + * only set each folio which already pinned when udmabuf_create. + * Note that, since a folio may be pinned multiple times, each folio + * can be added to pinned_folios multiple times, depending on how many + * times the folio has been pinned when create. + */ + pgoff_t nr_pinned; + struct folio **pinned_folios; + struct sg_table *sg; struct miscdevice *device; pgoff_t *offsets; - struct list_head unpin_list; -}; - -struct udmabuf_folio { - struct folio *folio; - struct list_head list; }; static vm_fault_t udmabuf_vm_fault(struct vm_fault *vmf) @@ -43,7 +49,8 @@ static vm_fault_t udmabuf_vm_fault(struct vm_fault *vmf) struct vm_area_struct *vma = vmf->vma; struct udmabuf *ubuf = vma->vm_private_data; pgoff_t pgoff = vmf->pgoff; - unsigned long pfn; + unsigned long addr, pfn; + vm_fault_t ret; if (pgoff >= ubuf->pagecount) return VM_FAULT_SIGBUS; @@ -51,7 +58,35 @@ static vm_fault_t udmabuf_vm_fault(struct vm_fault *vmf) pfn = folio_pfn(ubuf->folios[pgoff]); pfn += ubuf->offsets[pgoff] >> PAGE_SHIFT; - return vmf_insert_pfn(vma, vmf->address, pfn); + ret = vmf_insert_pfn(vma, vmf->address, pfn); + if (ret & VM_FAULT_ERROR) + return ret; + + /* pre fault */ + pgoff = vma->vm_pgoff; + addr = vma->vm_start; + + for (; addr < vma->vm_end; pgoff++, addr += PAGE_SIZE) { + if (addr == vmf->address) + continue; + + if (WARN_ON(pgoff >= ubuf->pagecount)) + break; + + pfn = folio_pfn(ubuf->folios[pgoff]); + pfn += ubuf->offsets[pgoff] >> PAGE_SHIFT; + + /** + * If the below vmf_insert_pfn() fails, we do not return an + * error here during this pre-fault step. However, an error + * will be returned if the failure occurs when the addr is + * truly accessed. + */ + if (vmf_insert_pfn(vma, addr, pfn) & VM_FAULT_ERROR) + break; + } + + return ret; } static const struct vm_operations_struct udmabuf_vm_ops = { @@ -74,21 +109,29 @@ static int mmap_udmabuf(struct dma_buf *buf, struct vm_area_struct *vma) static int vmap_udmabuf(struct dma_buf *buf, struct iosys_map *map) { struct udmabuf *ubuf = buf->priv; - struct page **pages; + unsigned long *pfns; void *vaddr; pgoff_t pg; dma_resv_assert_held(buf->resv); - pages = kmalloc_array(ubuf->pagecount, sizeof(*pages), GFP_KERNEL); - if (!pages) + /** + * HVO may free tail pages, so just use pfn to map each folio + * into vmalloc area. + */ + pfns = kvmalloc_array(ubuf->pagecount, sizeof(*pfns), GFP_KERNEL); + if (!pfns) return -ENOMEM; - for (pg = 0; pg < ubuf->pagecount; pg++) - pages[pg] = &ubuf->folios[pg]->page; + for (pg = 0; pg < ubuf->pagecount; pg++) { + unsigned long pfn = folio_pfn(ubuf->folios[pg]); - vaddr = vm_map_ram(pages, ubuf->pagecount, -1); - kfree(pages); + pfn += ubuf->offsets[pg] >> PAGE_SHIFT; + pfns[pg] = pfn; + } + + vaddr = vmap_pfn(pfns, ubuf->pagecount, PAGE_KERNEL); + kvfree(pfns); if (!vaddr) return -EINVAL; @@ -159,34 +202,42 @@ static void unmap_udmabuf(struct dma_buf_attachment *at, return put_sg_table(at->dev, sg, direction); } -static void unpin_all_folios(struct list_head *unpin_list) +static void unpin_all_folios(struct udmabuf *ubuf) { - struct udmabuf_folio *ubuf_folio; + pgoff_t i; - while (!list_empty(unpin_list)) { - ubuf_folio = list_first_entry(unpin_list, - struct udmabuf_folio, list); - unpin_folio(ubuf_folio->folio); + for (i = 0; i < ubuf->nr_pinned; ++i) + unpin_folio(ubuf->pinned_folios[i]); - list_del(&ubuf_folio->list); - kfree(ubuf_folio); - } + kvfree(ubuf->pinned_folios); } -static int add_to_unpin_list(struct list_head *unpin_list, - struct folio *folio) +static __always_inline int init_udmabuf(struct udmabuf *ubuf, pgoff_t pgcnt) { - struct udmabuf_folio *ubuf_folio; - - ubuf_folio = kzalloc(sizeof(*ubuf_folio), GFP_KERNEL); - if (!ubuf_folio) + ubuf->folios = kvmalloc_array(pgcnt, sizeof(*ubuf->folios), GFP_KERNEL); + if (!ubuf->folios) + return -ENOMEM; + + ubuf->offsets = kvcalloc(pgcnt, sizeof(*ubuf->offsets), GFP_KERNEL); + if (!ubuf->offsets) + return -ENOMEM; + + ubuf->pinned_folios = kvmalloc_array(pgcnt, + sizeof(*ubuf->pinned_folios), + GFP_KERNEL); + if (!ubuf->pinned_folios) return -ENOMEM; - ubuf_folio->folio = folio; - list_add_tail(&ubuf_folio->list, unpin_list); return 0; } +static __always_inline void deinit_udmabuf(struct udmabuf *ubuf) +{ + unpin_all_folios(ubuf); + kvfree(ubuf->offsets); + kvfree(ubuf->folios); +} + static void release_udmabuf(struct dma_buf *buf) { struct udmabuf *ubuf = buf->priv; @@ -195,9 +246,7 @@ static void release_udmabuf(struct dma_buf *buf) if (ubuf->sg) put_sg_table(dev, ubuf->sg, DMA_BIDIRECTIONAL); - unpin_all_folios(&ubuf->unpin_list); - kfree(ubuf->offsets); - kfree(ubuf->folios); + deinit_udmabuf(ubuf); kfree(ubuf); } @@ -254,9 +303,6 @@ static int check_memfd_seals(struct file *memfd) { int seals; - if (!memfd) - return -EBADFD; - if (!shmem_file(memfd) && !is_file_hugepages(memfd)) return -EBADFD; @@ -291,100 +337,116 @@ static int export_udmabuf(struct udmabuf *ubuf, return dma_buf_fd(buf, flags); } +static long udmabuf_pin_folios(struct udmabuf *ubuf, struct file *memfd, + loff_t start, loff_t size, struct folio **folios) +{ + pgoff_t nr_pinned = ubuf->nr_pinned; + pgoff_t upgcnt = ubuf->pagecount; + u32 cur_folio, cur_pgcnt; + pgoff_t pgoff, pgcnt; + long nr_folios; + loff_t end; + + pgcnt = size >> PAGE_SHIFT; + end = start + (pgcnt << PAGE_SHIFT) - 1; + nr_folios = memfd_pin_folios(memfd, start, end, folios, pgcnt, &pgoff); + if (nr_folios <= 0) + return nr_folios ? nr_folios : -EINVAL; + + cur_pgcnt = 0; + for (cur_folio = 0; cur_folio < nr_folios; ++cur_folio) { + pgoff_t subpgoff = pgoff; + size_t fsize = folio_size(folios[cur_folio]); + + ubuf->pinned_folios[nr_pinned++] = folios[cur_folio]; + + for (; subpgoff < fsize; subpgoff += PAGE_SIZE) { + ubuf->folios[upgcnt] = folios[cur_folio]; + ubuf->offsets[upgcnt] = subpgoff; + ++upgcnt; + + if (++cur_pgcnt >= pgcnt) + goto end; + } + + /** + * In a given range, only the first subpage of the first folio + * has an offset, that is returned by memfd_pin_folios(). + * The first subpages of other folios (in the range) have an + * offset of 0. + */ + pgoff = 0; + } +end: + ubuf->pagecount = upgcnt; + ubuf->nr_pinned = nr_pinned; + return 0; +} + static long udmabuf_create(struct miscdevice *device, struct udmabuf_create_list *head, struct udmabuf_create_item *list) { - pgoff_t pgoff, pgcnt, pglimit, pgbuf = 0; - long nr_folios, ret = -EINVAL; - struct file *memfd = NULL; - struct folio **folios; + unsigned long max_nr_folios = 0; + struct folio **folios = NULL; + pgoff_t pgcnt = 0, pglimit; struct udmabuf *ubuf; - u32 i, j, k, flags; - loff_t end; + long ret = -EINVAL; + u32 i, flags; ubuf = kzalloc(sizeof(*ubuf), GFP_KERNEL); if (!ubuf) return -ENOMEM; - INIT_LIST_HEAD(&ubuf->unpin_list); pglimit = (size_limit_mb * 1024 * 1024) >> PAGE_SHIFT; for (i = 0; i < head->count; i++) { - if (!IS_ALIGNED(list[i].offset, PAGE_SIZE)) - goto err; - if (!IS_ALIGNED(list[i].size, PAGE_SIZE)) - goto err; - ubuf->pagecount += list[i].size >> PAGE_SHIFT; - if (ubuf->pagecount > pglimit) - goto err; + pgoff_t subpgcnt; + + if (!PAGE_ALIGNED(list[i].offset)) + goto err_noinit; + if (!PAGE_ALIGNED(list[i].size)) + goto err_noinit; + + subpgcnt = list[i].size >> PAGE_SHIFT; + pgcnt += subpgcnt; + if (pgcnt > pglimit) + goto err_noinit; + + max_nr_folios = max_t(unsigned long, subpgcnt, max_nr_folios); } - if (!ubuf->pagecount) + if (!pgcnt) + goto err_noinit; + + ret = init_udmabuf(ubuf, pgcnt); + if (ret) goto err; - ubuf->folios = kmalloc_array(ubuf->pagecount, sizeof(*ubuf->folios), - GFP_KERNEL); - if (!ubuf->folios) { - ret = -ENOMEM; - goto err; - } - ubuf->offsets = kcalloc(ubuf->pagecount, sizeof(*ubuf->offsets), - GFP_KERNEL); - if (!ubuf->offsets) { + folios = kvmalloc_array(max_nr_folios, sizeof(*folios), GFP_KERNEL); + if (!folios) { ret = -ENOMEM; goto err; } - pgbuf = 0; for (i = 0; i < head->count; i++) { - memfd = fget(list[i].memfd); + struct file *memfd = fget(list[i].memfd); + + if (!memfd) { + ret = -EBADFD; + goto err; + } + ret = check_memfd_seals(memfd); - if (ret < 0) - goto err; - - pgcnt = list[i].size >> PAGE_SHIFT; - folios = kmalloc_array(pgcnt, sizeof(*folios), GFP_KERNEL); - if (!folios) { - ret = -ENOMEM; + if (ret < 0) { + fput(memfd); goto err; } - end = list[i].offset + (pgcnt << PAGE_SHIFT) - 1; - ret = memfd_pin_folios(memfd, list[i].offset, end, - folios, pgcnt, &pgoff); - if (ret <= 0) { - kfree(folios); - if (!ret) - ret = -EINVAL; - goto err; - } - - nr_folios = ret; - pgoff >>= PAGE_SHIFT; - for (j = 0, k = 0; j < pgcnt; j++) { - ubuf->folios[pgbuf] = folios[k]; - ubuf->offsets[pgbuf] = pgoff << PAGE_SHIFT; - - if (j == 0 || ubuf->folios[pgbuf-1] != folios[k]) { - ret = add_to_unpin_list(&ubuf->unpin_list, - folios[k]); - if (ret < 0) { - kfree(folios); - goto err; - } - } - - pgbuf++; - if (++pgoff == folio_nr_pages(folios[k])) { - pgoff = 0; - if (++k == nr_folios) - break; - } - } - - kfree(folios); + ret = udmabuf_pin_folios(ubuf, memfd, list[i].offset, + list[i].size, folios); fput(memfd); - memfd = NULL; + if (ret) + goto err; } flags = head->flags & UDMABUF_FLAGS_CLOEXEC ? O_CLOEXEC : 0; @@ -392,15 +454,14 @@ static long udmabuf_create(struct miscdevice *device, if (ret < 0) goto err; + kvfree(folios); return ret; err: - if (memfd) - fput(memfd); - unpin_all_folios(&ubuf->unpin_list); - kfree(ubuf->offsets); - kfree(ubuf->folios); + deinit_udmabuf(ubuf); +err_noinit: kfree(ubuf); + kvfree(folios); return ret; } diff --git a/drivers/dma/ep93xx_dma.c b/drivers/dma/ep93xx_dma.c index 995427afe077..6b98a23e3332 100644 --- a/drivers/dma/ep93xx_dma.c +++ b/drivers/dma/ep93xx_dma.c @@ -1391,11 +1391,12 @@ static struct ep93xx_dma_engine *ep93xx_dma_of_probe(struct platform_device *pde INIT_LIST_HEAD(&dma_dev->channels); for (i = 0; i < edma->num_channels; i++) { struct ep93xx_dma_chan *edmac = &edma->channels[i]; + int len; edmac->chan.device = dma_dev; edmac->regs = devm_platform_ioremap_resource(pdev, i); if (IS_ERR(edmac->regs)) - return edmac->regs; + return ERR_CAST(edmac->regs); edmac->irq = fwnode_irq_get(dev_fwnode(dev), i); if (edmac->irq < 0) @@ -1404,9 +1405,11 @@ static struct ep93xx_dma_engine *ep93xx_dma_of_probe(struct platform_device *pde edmac->edma = edma; if (edma->m2m) - snprintf(dma_clk_name, sizeof(dma_clk_name), "m2m%u", i); + len = snprintf(dma_clk_name, sizeof(dma_clk_name), "m2m%u", i); else - snprintf(dma_clk_name, sizeof(dma_clk_name), "m2p%u", i); + len = snprintf(dma_clk_name, sizeof(dma_clk_name), "m2p%u", i); + if (len >= sizeof(dma_clk_name)) + return ERR_PTR(-ENOBUFS); edmac->clk = devm_clk_get(dev, dma_clk_name); if (IS_ERR(edmac->clk)) { diff --git a/drivers/dma/sh/rz-dmac.c b/drivers/dma/sh/rz-dmac.c index 65a27c5a7bce..811389fc9cb8 100644 --- a/drivers/dma/sh/rz-dmac.c +++ b/drivers/dma/sh/rz-dmac.c @@ -601,22 +601,25 @@ static int rz_dmac_config(struct dma_chan *chan, struct rz_dmac_chan *channel = to_rz_dmac_chan(chan); u32 val; - channel->src_per_address = config->src_addr; channel->dst_per_address = config->dst_addr; - - val = rz_dmac_ds_to_val_mapping(config->dst_addr_width); - if (val == CHCFG_DS_INVALID) - return -EINVAL; - channel->chcfg &= ~CHCFG_FILL_DDS_MASK; - channel->chcfg |= FIELD_PREP(CHCFG_FILL_DDS_MASK, val); + if (channel->dst_per_address) { + val = rz_dmac_ds_to_val_mapping(config->dst_addr_width); + if (val == CHCFG_DS_INVALID) + return -EINVAL; - val = rz_dmac_ds_to_val_mapping(config->src_addr_width); - if (val == CHCFG_DS_INVALID) - return -EINVAL; + channel->chcfg |= FIELD_PREP(CHCFG_FILL_DDS_MASK, val); + } + channel->src_per_address = config->src_addr; channel->chcfg &= ~CHCFG_FILL_SDS_MASK; - channel->chcfg |= FIELD_PREP(CHCFG_FILL_SDS_MASK, val); + if (channel->src_per_address) { + val = rz_dmac_ds_to_val_mapping(config->src_addr_width); + if (val == CHCFG_DS_INVALID) + return -EINVAL; + + channel->chcfg |= FIELD_PREP(CHCFG_FILL_SDS_MASK, val); + } return 0; } diff --git a/drivers/dma/ti/k3-udma.c b/drivers/dma/ti/k3-udma.c index 406ee199c2ac..b3f27b3f9209 100644 --- a/drivers/dma/ti/k3-udma.c +++ b/drivers/dma/ti/k3-udma.c @@ -3185,27 +3185,40 @@ static int udma_configure_statictr(struct udma_chan *uc, struct udma_desc *d, d->static_tr.elcnt = elcnt; - /* - * PDMA must to close the packet when the channel is in packet mode. - * For TR mode when the channel is not cyclic we also need PDMA to close - * the packet otherwise the transfer will stall because PDMA holds on - * the data it has received from the peripheral. - */ if (uc->config.pkt_mode || !uc->cyclic) { + /* + * PDMA must close the packet when the channel is in packet mode. + * For TR mode when the channel is not cyclic we also need PDMA + * to close the packet otherwise the transfer will stall because + * PDMA holds on the data it has received from the peripheral. + */ unsigned int div = dev_width * elcnt; if (uc->cyclic) d->static_tr.bstcnt = d->residue / d->sglen / div; else d->static_tr.bstcnt = d->residue / div; + } else if (uc->ud->match_data->type == DMA_TYPE_BCDMA && + uc->config.dir == DMA_DEV_TO_MEM && + uc->cyclic) { + /* + * For cyclic mode with BCDMA we have to set EOP in each TR to + * prevent short packet errors seen on channel teardown. So the + * PDMA must close the packet after every TR transfer by setting + * burst count equal to the number of bytes transferred. + */ + struct cppi5_tr_type1_t *tr_req = d->hwdesc[0].tr_req_base; - if (uc->config.dir == DMA_DEV_TO_MEM && - d->static_tr.bstcnt > uc->ud->match_data->statictr_z_mask) - return -EINVAL; + d->static_tr.bstcnt = + (tr_req->icnt0 * tr_req->icnt1) / dev_width; } else { d->static_tr.bstcnt = 0; } + if (uc->config.dir == DMA_DEV_TO_MEM && + d->static_tr.bstcnt > uc->ud->match_data->statictr_z_mask) + return -EINVAL; + return 0; } @@ -3450,8 +3463,9 @@ udma_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl, /* static TR for remote PDMA */ if (udma_configure_statictr(uc, d, dev_width, burst)) { dev_err(uc->ud->dev, - "%s: StaticTR Z is limited to maximum 4095 (%u)\n", - __func__, d->static_tr.bstcnt); + "%s: StaticTR Z is limited to maximum %u (%u)\n", + __func__, uc->ud->match_data->statictr_z_mask, + d->static_tr.bstcnt); udma_free_hwdesc(uc, d); kfree(d); @@ -3476,6 +3490,7 @@ udma_prep_dma_cyclic_tr(struct udma_chan *uc, dma_addr_t buf_addr, u16 tr0_cnt0, tr0_cnt1, tr1_cnt0; unsigned int i; int num_tr; + u32 period_csf = 0; num_tr = udma_get_tr_counters(period_len, __ffs(buf_addr), &tr0_cnt0, &tr0_cnt1, &tr1_cnt0); @@ -3498,6 +3513,20 @@ udma_prep_dma_cyclic_tr(struct udma_chan *uc, dma_addr_t buf_addr, period_addr = buf_addr | ((u64)uc->config.asel << K3_ADDRESS_ASEL_SHIFT); + /* + * For BCDMA <-> PDMA transfers, the EOP flag needs to be set on the + * last TR of a descriptor, to mark the packet as complete. + * This is required for getting the teardown completion message in case + * of TX, and to avoid short-packet error in case of RX. + * + * As we are in cyclic mode, we do not know which period might be the + * last one, so set the flag for each period. + */ + if (uc->config.ep_type == PSIL_EP_PDMA_XY && + uc->ud->match_data->type == DMA_TYPE_BCDMA) { + period_csf = CPPI5_TR_CSF_EOP; + } + for (i = 0; i < periods; i++) { int tr_idx = i * num_tr; @@ -3525,8 +3554,10 @@ udma_prep_dma_cyclic_tr(struct udma_chan *uc, dma_addr_t buf_addr, } if (!(flags & DMA_PREP_INTERRUPT)) - cppi5_tr_csf_set(&tr_req[tr_idx].flags, - CPPI5_TR_CSF_SUPR_EVT); + period_csf |= CPPI5_TR_CSF_SUPR_EVT; + + if (period_csf) + cppi5_tr_csf_set(&tr_req[tr_idx].flags, period_csf); period_addr += period_len; } @@ -3655,8 +3686,9 @@ udma_prep_dma_cyclic(struct dma_chan *chan, dma_addr_t buf_addr, size_t buf_len, /* static TR for remote PDMA */ if (udma_configure_statictr(uc, d, dev_width, burst)) { dev_err(uc->ud->dev, - "%s: StaticTR Z is limited to maximum 4095 (%u)\n", - __func__, d->static_tr.bstcnt); + "%s: StaticTR Z is limited to maximum %u (%u)\n", + __func__, uc->ud->match_data->statictr_z_mask, + d->static_tr.bstcnt); udma_free_hwdesc(uc, d); kfree(d); diff --git a/drivers/firewire/core-topology.c b/drivers/firewire/core-topology.c index 6adadb11962e..892b94cfd626 100644 --- a/drivers/firewire/core-topology.c +++ b/drivers/firewire/core-topology.c @@ -204,7 +204,7 @@ static struct fw_node *build_tree(struct fw_card *card, const u32 *sid, int self // the node->ports array where the parent node should be. Later, // when we handle the parent node, we fix up the reference. ++parent_count; - node->color = i; + node->color = port_index; break; case PHY_PACKET_SELF_ID_PORT_STATUS_CHILD: diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c index 7a4d1a478e33..1bf0e15c1540 100644 --- a/drivers/firewire/net.c +++ b/drivers/firewire/net.c @@ -28,7 +28,7 @@ #include #include -#include +#include #include #include diff --git a/drivers/firmware/arm_ffa/driver.c b/drivers/firmware/arm_ffa/driver.c index 4d231bc375e0..b14cbdae94e8 100644 --- a/drivers/firmware/arm_ffa/driver.c +++ b/drivers/firmware/arm_ffa/driver.c @@ -481,11 +481,16 @@ static int ffa_msg_send_direct_req2(u16 src_id, u16 dst_id, const uuid_t *uuid, struct ffa_send_direct_data2 *data) { u32 src_dst_ids = PACK_TARGET_INFO(src_id, dst_id); + union { + uuid_t uuid; + __le64 regs[2]; + } uuid_regs = { .uuid = *uuid }; ffa_value_t ret, args = { - .a0 = FFA_MSG_SEND_DIRECT_REQ2, .a1 = src_dst_ids, + .a0 = FFA_MSG_SEND_DIRECT_REQ2, + .a1 = src_dst_ids, + .a2 = le64_to_cpu(uuid_regs.regs[0]), + .a3 = le64_to_cpu(uuid_regs.regs[1]), }; - - export_uuid((u8 *)&args.a2, uuid); memcpy((void *)&args + offsetof(ffa_value_t, a4), data, sizeof(*data)); invoke_ffa_fn(args, &ret); @@ -496,7 +501,7 @@ static int ffa_msg_send_direct_req2(u16 src_id, u16 dst_id, const uuid_t *uuid, return ffa_to_linux_errno((int)ret.a2); if (ret.a0 == FFA_MSG_SEND_DIRECT_RESP2) { - memcpy(data, &ret.a4, sizeof(*data)); + memcpy(data, (void *)&ret + offsetof(ffa_value_t, a4), sizeof(*data)); return 0; } diff --git a/drivers/firmware/arm_scmi/common.h b/drivers/firmware/arm_scmi/common.h index 6d9227db473f..c4b8e7ff88aa 100644 --- a/drivers/firmware/arm_scmi/common.h +++ b/drivers/firmware/arm_scmi/common.h @@ -22,7 +22,7 @@ #include #include -#include +#include #include "protocols.h" #include "notify.h" diff --git a/drivers/firmware/arm_scmi/driver.c b/drivers/firmware/arm_scmi/driver.c index 88c5c4ff4bb6..a477b5ade38d 100644 --- a/drivers/firmware/arm_scmi/driver.c +++ b/drivers/firmware/arm_scmi/driver.c @@ -2976,10 +2976,8 @@ static struct scmi_debug_info *scmi_debugfs_common_setup(struct scmi_info *info) dbg->top_dentry = top_dentry; if (devm_add_action_or_reset(info->dev, - scmi_debugfs_common_cleanup, dbg)) { - scmi_debugfs_common_cleanup(dbg); + scmi_debugfs_common_cleanup, dbg)) return NULL; - } return dbg; } diff --git a/drivers/firmware/arm_scmi/protocols.h b/drivers/firmware/arm_scmi/protocols.h index 8e95f53bd7b7..aaee57cdcd55 100644 --- a/drivers/firmware/arm_scmi/protocols.h +++ b/drivers/firmware/arm_scmi/protocols.h @@ -22,7 +22,7 @@ #include #include -#include +#include #define PROTOCOL_REV_MINOR_MASK GENMASK(15, 0) #define PROTOCOL_REV_MAJOR_MASK GENMASK(31, 16) diff --git a/drivers/firmware/arm_scmi/transports/Makefile b/drivers/firmware/arm_scmi/transports/Makefile index 362a406f08e6..3ba3d3bee151 100644 --- a/drivers/firmware/arm_scmi/transports/Makefile +++ b/drivers/firmware/arm_scmi/transports/Makefile @@ -1,8 +1,10 @@ # SPDX-License-Identifier: GPL-2.0-only -scmi_transport_mailbox-objs := mailbox.o -obj-$(CONFIG_ARM_SCMI_TRANSPORT_MAILBOX) += scmi_transport_mailbox.o +# Keep before scmi_transport_mailbox.o to allow precedence +# while matching the compatible. scmi_transport_smc-objs := smc.o obj-$(CONFIG_ARM_SCMI_TRANSPORT_SMC) += scmi_transport_smc.o +scmi_transport_mailbox-objs := mailbox.o +obj-$(CONFIG_ARM_SCMI_TRANSPORT_MAILBOX) += scmi_transport_mailbox.o scmi_transport_optee-objs := optee.o obj-$(CONFIG_ARM_SCMI_TRANSPORT_OPTEE) += scmi_transport_optee.o scmi_transport_virtio-objs := virtio.o diff --git a/drivers/firmware/arm_scmi/transports/mailbox.c b/drivers/firmware/arm_scmi/transports/mailbox.c index 1a754dee24f7..e3d5f7560990 100644 --- a/drivers/firmware/arm_scmi/transports/mailbox.c +++ b/drivers/firmware/arm_scmi/transports/mailbox.c @@ -25,6 +25,7 @@ * @chan_platform_receiver: Optional Platform Receiver mailbox unidirectional channel * @cinfo: SCMI channel info * @shmem: Transmit/Receive shared memory area + * @chan_lock: Lock that prevents multiple xfers from being queued */ struct scmi_mailbox { struct mbox_client cl; @@ -33,6 +34,7 @@ struct scmi_mailbox { struct mbox_chan *chan_platform_receiver; struct scmi_chan_info *cinfo; struct scmi_shared_mem __iomem *shmem; + struct mutex chan_lock; }; #define client_to_scmi_mailbox(c) container_of(c, struct scmi_mailbox, cl) @@ -238,6 +240,7 @@ static int mailbox_chan_setup(struct scmi_chan_info *cinfo, struct device *dev, cinfo->transport_info = smbox; smbox->cinfo = cinfo; + mutex_init(&smbox->chan_lock); return 0; } @@ -267,13 +270,23 @@ static int mailbox_send_message(struct scmi_chan_info *cinfo, struct scmi_mailbox *smbox = cinfo->transport_info; int ret; + /* + * The mailbox layer has its own queue. However the mailbox queue + * confuses the per message SCMI timeouts since the clock starts when + * the message is submitted into the mailbox queue. So when multiple + * messages are queued up the clock starts on all messages instead of + * only the one inflight. + */ + mutex_lock(&smbox->chan_lock); + ret = mbox_send_message(smbox->chan, xfer); + /* mbox_send_message returns non-negative value on success */ + if (ret < 0) { + mutex_unlock(&smbox->chan_lock); + return ret; + } - /* mbox_send_message returns non-negative value on success, so reset */ - if (ret > 0) - ret = 0; - - return ret; + return 0; } static void mailbox_mark_txdone(struct scmi_chan_info *cinfo, int ret, @@ -281,13 +294,10 @@ static void mailbox_mark_txdone(struct scmi_chan_info *cinfo, int ret, { struct scmi_mailbox *smbox = cinfo->transport_info; - /* - * NOTE: we might prefer not to need the mailbox ticker to manage the - * transfer queueing since the protocol layer queues things by itself. - * Unfortunately, we have to kick the mailbox framework after we have - * received our message. - */ mbox_client_txdone(smbox->chan, ret); + + /* Release channel */ + mutex_unlock(&smbox->chan_lock); } static void mailbox_fetch_response(struct scmi_chan_info *cinfo, diff --git a/drivers/firmware/arm_sdei.c b/drivers/firmware/arm_sdei.c index 285fe7ad490d..3e8051fe8296 100644 --- a/drivers/firmware/arm_sdei.c +++ b/drivers/firmware/arm_sdei.c @@ -763,7 +763,7 @@ static int sdei_device_freeze(struct device *dev) int err; /* unregister private events */ - cpuhp_remove_state(sdei_entry_point); + cpuhp_remove_state(sdei_hp_state); err = sdei_unregister_shared(); if (err) diff --git a/drivers/firmware/dmi_scan.c b/drivers/firmware/dmi_scan.c index 23b002e4d4a0..fde0656481cc 100644 --- a/drivers/firmware/dmi_scan.c +++ b/drivers/firmware/dmi_scan.c @@ -9,7 +9,7 @@ #include #include #include -#include +#include #ifndef SMBIOS_ENTRY_POINT_SCAN_START #define SMBIOS_ENTRY_POINT_SCAN_START 0xF0000 diff --git a/drivers/firmware/efi/fdtparams.c b/drivers/firmware/efi/fdtparams.c index 0ec83ba58097..b815d2a754ee 100644 --- a/drivers/firmware/efi/fdtparams.c +++ b/drivers/firmware/efi/fdtparams.c @@ -8,7 +8,7 @@ #include #include -#include +#include enum { SYSTAB, diff --git a/drivers/firmware/efi/libstub/riscv-stub.c b/drivers/firmware/efi/libstub/riscv-stub.c index c96d6dcee86c..e7d9204baee3 100644 --- a/drivers/firmware/efi/libstub/riscv-stub.c +++ b/drivers/firmware/efi/libstub/riscv-stub.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include "efistub.h" diff --git a/drivers/firmware/efi/libstub/riscv.c b/drivers/firmware/efi/libstub/riscv.c index 8022b104c3e6..f66f33ceb99e 100644 --- a/drivers/firmware/efi/libstub/riscv.c +++ b/drivers/firmware/efi/libstub/riscv.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include "efistub.h" diff --git a/drivers/firmware/efi/libstub/zboot.c b/drivers/firmware/efi/libstub/zboot.c index 1ceace956758..af23b3c50228 100644 --- a/drivers/firmware/efi/libstub/zboot.c +++ b/drivers/firmware/efi/libstub/zboot.c @@ -3,7 +3,7 @@ #include #include #include -#include +#include #include "efistub.h" diff --git a/drivers/firmware/sysfb.c b/drivers/firmware/sysfb.c index 02a07d3d0d40..a3df782fa687 100644 --- a/drivers/firmware/sysfb.c +++ b/drivers/firmware/sysfb.c @@ -67,9 +67,11 @@ static bool sysfb_unregister(void) void sysfb_disable(struct device *dev) { struct screen_info *si = &screen_info; + struct device *parent; mutex_lock(&disable_lock); - if (!dev || dev == sysfb_parent_dev(si)) { + parent = sysfb_parent_dev(si); + if (!dev || !parent || dev == parent) { sysfb_unregister(); disabled = true; } diff --git a/drivers/fpga/microchip-spi.c b/drivers/fpga/microchip-spi.c index 2a82c726d6e5..6134cea86ac8 100644 --- a/drivers/fpga/microchip-spi.c +++ b/drivers/fpga/microchip-spi.c @@ -3,7 +3,7 @@ * Microchip Polarfire FPGA programming over slave SPI interface. */ -#include +#include #include #include #include diff --git a/drivers/fsi/fsi-occ.c b/drivers/fsi/fsi-occ.c index f58b158d097c..a6d4c8f123a5 100644 --- a/drivers/fsi/fsi-occ.c +++ b/drivers/fsi/fsi-occ.c @@ -20,7 +20,7 @@ #include #include #include -#include +#include #define OCC_SRAM_BYTES 4096 #define OCC_CMD_DATA_BYTES 4090 diff --git a/drivers/gpio/gpio-aspeed.c b/drivers/gpio/gpio-aspeed.c index 04c03402db6d..ea40ad43a79b 100644 --- a/drivers/gpio/gpio-aspeed.c +++ b/drivers/gpio/gpio-aspeed.c @@ -406,6 +406,8 @@ static void __aspeed_gpio_set(struct gpio_chip *gc, unsigned int offset, gpio->dcache[GPIO_BANK(offset)] = reg; iowrite32(reg, addr); + /* Flush write */ + ioread32(addr); } static void aspeed_gpio_set(struct gpio_chip *gc, unsigned int offset, @@ -1191,7 +1193,7 @@ static int __init aspeed_gpio_probe(struct platform_device *pdev) if (!gpio_id) return -EINVAL; - gpio->clk = of_clk_get(pdev->dev.of_node, 0); + gpio->clk = devm_clk_get_enabled(&pdev->dev, NULL); if (IS_ERR(gpio->clk)) { dev_warn(&pdev->dev, "Failed to get clock from devicetree, debouncing disabled\n"); diff --git a/drivers/gpio/gpio-davinci.c b/drivers/gpio/gpio-davinci.c index b54fef6b1e12..76b58c70b257 100644 --- a/drivers/gpio/gpio-davinci.c +++ b/drivers/gpio/gpio-davinci.c @@ -253,7 +253,7 @@ static int davinci_gpio_probe(struct platform_device *pdev) * serve as EDMA event triggers. */ -static void gpio_irq_disable(struct irq_data *d) +static void gpio_irq_mask(struct irq_data *d) { struct davinci_gpio_regs __iomem *g = irq2regs(d); uintptr_t mask = (uintptr_t)irq_data_get_irq_handler_data(d); @@ -262,7 +262,7 @@ static void gpio_irq_disable(struct irq_data *d) writel_relaxed(mask, &g->clr_rising); } -static void gpio_irq_enable(struct irq_data *d) +static void gpio_irq_unmask(struct irq_data *d) { struct davinci_gpio_regs __iomem *g = irq2regs(d); uintptr_t mask = (uintptr_t)irq_data_get_irq_handler_data(d); @@ -288,8 +288,8 @@ static int gpio_irq_type(struct irq_data *d, unsigned trigger) static struct irq_chip gpio_irqchip = { .name = "GPIO", - .irq_enable = gpio_irq_enable, - .irq_disable = gpio_irq_disable, + .irq_unmask = gpio_irq_unmask, + .irq_mask = gpio_irq_mask, .irq_set_type = gpio_irq_type, .flags = IRQCHIP_SET_TYPE_MASKED | IRQCHIP_SKIP_SET_WAKE, }; @@ -472,7 +472,7 @@ static int davinci_gpio_irq_setup(struct platform_device *pdev) return PTR_ERR(clk); } - if (chips->gpio_unbanked) { + if (!chips->gpio_unbanked) { irq = devm_irq_alloc_descs(dev, -1, 0, ngpio, 0); if (irq < 0) { dev_err(dev, "Couldn't allocate IRQ numbers\n"); diff --git a/drivers/gpio/gpio-sloppy-logic-analyzer.c b/drivers/gpio/gpio-sloppy-logic-analyzer.c index 07e0d7180579..59a8f3a5c4e4 100644 --- a/drivers/gpio/gpio-sloppy-logic-analyzer.c +++ b/drivers/gpio/gpio-sloppy-logic-analyzer.c @@ -234,7 +234,9 @@ static int gpio_la_poll_probe(struct platform_device *pdev) if (!priv) return -ENOMEM; - devm_mutex_init(dev, &priv->blob_lock); + ret = devm_mutex_init(dev, &priv->blob_lock); + if (ret) + return ret; fops_buf_size_set(priv, GPIO_LA_DEFAULT_BUF_SIZE); diff --git a/drivers/gpio/gpiolib-swnode.c b/drivers/gpio/gpiolib-swnode.c index 2b2dd7e92211..51d2475c05c5 100644 --- a/drivers/gpio/gpiolib-swnode.c +++ b/drivers/gpio/gpiolib-swnode.c @@ -64,7 +64,7 @@ struct gpio_desc *swnode_find_gpio(struct fwnode_handle *fwnode, struct fwnode_reference_args args; struct gpio_desc *desc; char propname[32]; /* 32 is max size of property name */ - int ret; + int ret = 0; swnode = to_software_node(fwnode); if (!swnode) diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c index c6afbf434366..2b02655abb56 100644 --- a/drivers/gpio/gpiolib.c +++ b/drivers/gpio/gpiolib.c @@ -114,12 +114,12 @@ const char *gpiod_get_label(struct gpio_desc *desc) srcu_read_lock_held(&desc->gdev->desc_srcu)); if (test_bit(FLAG_USED_AS_IRQ, &flags)) - return label->str ?: "interrupt"; + return label ? label->str : "interrupt"; if (!test_bit(FLAG_REQUESTED, &flags)) return NULL; - return label->str; + return label ? label->str : NULL; } static void desc_free_label(struct rcu_head *rh) @@ -4926,6 +4926,8 @@ static void *gpiolib_seq_start(struct seq_file *s, loff_t *pos) return NULL; s->private = priv; + if (*pos > 0) + priv->newline = true; priv->idx = srcu_read_lock(&gpio_devices_srcu); list_for_each_entry_srcu(gdev, &gpio_devices, list, @@ -4969,7 +4971,7 @@ static int gpiolib_seq_show(struct seq_file *s, void *v) gc = srcu_dereference(gdev->chip, &gdev->srcu); if (!gc) { - seq_printf(s, "%s%s: (dangling chip)", + seq_printf(s, "%s%s: (dangling chip)\n", priv->newline ? "\n" : "", dev_name(&gdev->dev)); return 0; diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig index 1cb5a4f19293..5504721007cc 100644 --- a/drivers/gpu/drm/Kconfig +++ b/drivers/gpu/drm/Kconfig @@ -9,9 +9,6 @@ menuconfig DRM tristate "Direct Rendering Manager (XFree86 4.1.0 and higher DRI support)" depends on (AGP || AGP=n) && !EMULATED_CMPXCHG && HAS_DMA select DRM_PANEL_ORIENTATION_QUIRKS - select DRM_KMS_HELPER if DRM_FBDEV_EMULATION - select FB_CORE if DRM_FBDEV_EMULATION - select FB_SYSMEM_HELPERS_DEFERRED if DRM_FBDEV_EMULATION select HDMI select I2C select DMA_SHARED_BUFFER @@ -152,6 +149,7 @@ config DRM_PANIC_SCREEN config DRM_PANIC_SCREEN_QR_CODE bool "Add a panic screen with a QR code" depends on DRM_PANIC && RUST + select ZLIB_DEFLATE help This option adds a QR code generator, and a panic screen with a QR code. The QR code will contain the last lines of kmsg and other debug @@ -210,9 +208,47 @@ config DRM_DEBUG_MODESET_LOCK If in doubt, say "N". +config DRM_CLIENT + bool + depends on DRM + help + Enables support for DRM clients. DRM drivers that need + struct drm_client_dev and its interfaces should select this + option. Drivers that support the default clients should + select DRM_CLIENT_SELECTION instead. + +config DRM_CLIENT_LIB + tristate + depends on DRM + select DRM_KMS_HELPER if DRM_FBDEV_EMULATION + select FB_CORE if DRM_FBDEV_EMULATION + help + This option enables the DRM client library and selects all + modules and components according to the enabled clients. + +config DRM_CLIENT_SELECTION + tristate + depends on DRM + select DRM_CLIENT_LIB if DRM_FBDEV_EMULATION + help + Drivers that support in-kernel DRM clients have to select this + option. + +config DRM_CLIENT_SETUP + bool + depends on DRM_CLIENT_SELECTION + help + Enables the DRM client selection. DRM drivers that support the + default clients should select DRM_CLIENT_SELECTION instead. + +menu "Supported DRM clients" + depends on DRM_CLIENT_SELECTION + config DRM_FBDEV_EMULATION bool "Enable legacy fbdev support for your modesetting driver" - depends on DRM + depends on DRM_CLIENT_SELECTION + select DRM_CLIENT + select DRM_CLIENT_SETUP select FRAMEBUFFER_CONSOLE_DETECT_PRIMARY if FRAMEBUFFER_CONSOLE default FB help @@ -251,6 +287,8 @@ config DRM_FBDEV_LEAK_PHYS_SMEM If in doubt, say "N" or spread the word to your closed source library vendor. +endmenu + config DRM_LOAD_EDID_FIRMWARE bool "Allow to specify an EDID data set instead of probing for it" depends on DRM @@ -320,19 +358,21 @@ config DRM_TTM_HELPER tristate depends on DRM select DRM_TTM + select FB_SYSMEM_HELPERS_DEFERRED if DRM_FBDEV_EMULATION help Helpers for ttm-based gem objects config DRM_GEM_DMA_HELPER tristate depends on DRM - select FB_DMAMEM_HELPERS if DRM_FBDEV_EMULATION + select FB_DMAMEM_HELPERS_DEFERRED if DRM_FBDEV_EMULATION help Choose this if you need the GEM DMA helper functions config DRM_GEM_SHMEM_HELPER tristate depends on DRM && MMU + select FB_SYSMEM_HELPERS_DEFERRED if DRM_FBDEV_EMULATION help Choose this if you need the GEM shmem helper functions @@ -472,6 +512,7 @@ source "drivers/gpu/drm/imagination/Kconfig" config DRM_HYPERV tristate "DRM Support for Hyper-V synthetic video device" depends on DRM && PCI && MMU && HYPERV + select DRM_CLIENT_SELECTION select DRM_KMS_HELPER select DRM_GEM_SHMEM_HELPER help diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile index 784229d4504d..463afad1b5ca 100644 --- a/drivers/gpu/drm/Makefile +++ b/drivers/gpu/drm/Makefile @@ -34,15 +34,12 @@ endif subdir-ccflags-$(CONFIG_DRM_WERROR) += -Werror drm-y := \ - drm_aperture.o \ drm_atomic.o \ drm_atomic_uapi.o \ drm_auth.o \ drm_blend.o \ drm_bridge.o \ drm_cache.o \ - drm_client.o \ - drm_client_modeset.o \ drm_color_mgmt.o \ drm_connector.o \ drm_crtc.o \ @@ -68,6 +65,7 @@ drm-y := \ drm_prime.o \ drm_print.o \ drm_property.o \ + drm_rect.o \ drm_syncobj.o \ drm_sysfs.o \ drm_trace_points.o \ @@ -75,6 +73,10 @@ drm-y := \ drm_vblank_work.o \ drm_vma_manager.o \ drm_writeback.o +drm-$(CONFIG_DRM_CLIENT) += \ + drm_client.o \ + drm_client_event.o \ + drm_client_modeset.o drm-$(CONFIG_DRM_LIB_RANDOM) += lib/drm_random.o drm-$(CONFIG_COMPAT) += drm_ioc32.o drm-$(CONFIG_DRM_PANEL) += drm_panel.o @@ -140,13 +142,20 @@ drm_kms_helper-y := \ drm_modeset_helper.o \ drm_plane_helper.o \ drm_probe_helper.o \ - drm_rect.o \ drm_self_refresh_helper.o \ drm_simple_kms_helper.o drm_kms_helper-$(CONFIG_DRM_PANEL_BRIDGE) += bridge/panel.o drm_kms_helper-$(CONFIG_DRM_FBDEV_EMULATION) += drm_fb_helper.o obj-$(CONFIG_DRM_KMS_HELPER) += drm_kms_helper.o +# +# DRM clients +# + +drm_client_lib-y := drm_client_setup.o +drm_client_lib-$(CONFIG_DRM_FBDEV_EMULATION) += drm_fbdev_client.o +obj-$(CONFIG_DRM_CLIENT_LIB) += drm_client_lib.o + # # Drivers and the rest # diff --git a/drivers/gpu/drm/amd/amdgpu/Kconfig b/drivers/gpu/drm/amd/amdgpu/Kconfig index 0051fb1b437f..41fa3377d9cf 100644 --- a/drivers/gpu/drm/amd/amdgpu/Kconfig +++ b/drivers/gpu/drm/amd/amdgpu/Kconfig @@ -5,7 +5,10 @@ config DRM_AMDGPU depends on DRM && PCI && MMU depends on !UML select FW_LOADER + select DRM_CLIENT + select DRM_CLIENT_SELECTION select DRM_DISPLAY_DP_HELPER + select DRM_DISPLAY_DSC_HELPER select DRM_DISPLAY_HDMI_HELPER select DRM_DISPLAY_HDCP_HELPER select DRM_DISPLAY_HELPER diff --git a/drivers/gpu/drm/amd/amdgpu/aldebaran.c b/drivers/gpu/drm/amd/amdgpu/aldebaran.c index b0f95a7649bf..3a588fecb0c5 100644 --- a/drivers/gpu/drm/amd/amdgpu/aldebaran.c +++ b/drivers/gpu/drm/amd/amdgpu/aldebaran.c @@ -85,16 +85,9 @@ static int aldebaran_mode2_suspend_ip(struct amdgpu_device *adev) AMD_IP_BLOCK_TYPE_SDMA)) continue; - r = adev->ip_blocks[i].version->funcs->suspend(adev); - - if (r) { - dev_err(adev->dev, - "suspend of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + r = amdgpu_ip_block_suspend(&adev->ip_blocks[i]); + if (r) return r; - } - - adev->ip_blocks[i].status.hw = false; } return 0; @@ -246,7 +239,7 @@ static int aldebaran_mode2_restore_ip(struct amdgpu_device *adev) dev_err(adev->dev, "Failed to get BIF handle\n"); return -EINVAL; } - r = cmn_block->version->funcs->resume(adev); + r = amdgpu_ip_block_resume(cmn_block); if (r) return r; @@ -282,15 +275,10 @@ static int aldebaran_mode2_restore_ip(struct amdgpu_device *adev) adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SDMA)) continue; - r = adev->ip_blocks[i].version->funcs->resume(adev); - if (r) { - dev_err(adev->dev, - "resume of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); - return r; - } - adev->ip_blocks[i].status.hw = true; + r = amdgpu_ip_block_resume(&adev->ip_blocks[i]); + if (r) + return r; } for (i = 0; i < adev->num_ip_blocks; i++) { @@ -304,7 +292,7 @@ static int aldebaran_mode2_restore_ip(struct amdgpu_device *adev) if (adev->ip_blocks[i].version->funcs->late_init) { r = adev->ip_blocks[i].version->funcs->late_init( - (void *)adev); + &adev->ip_blocks[i]); if (r) { dev_err(adev->dev, "late_init of IP block <%s> failed %d after reset\n", @@ -417,6 +405,7 @@ static struct amdgpu_reset_handler aldebaran_mode2_handler = { static struct amdgpu_reset_handler *aldebaran_rst_handlers[AMDGPU_RESET_MAX_HANDLERS] = { &aldebaran_mode2_handler, + &xgmi_reset_on_init_handler, }; int aldebaran_reset_init(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 9b1e0ede05a4..3af5acff8518 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -131,10 +131,6 @@ struct amdgpu_mgpu_info { uint32_t num_gpu; uint32_t num_dgpu; uint32_t num_apu; - - /* delayed reset_func for XGMI configuration if necessary */ - struct delayed_work delayed_reset_work; - bool pending_reset; }; enum amdgpu_ss { @@ -365,8 +361,11 @@ void amdgpu_device_ip_get_clockgating_state(struct amdgpu_device *adev, u64 *flags); int amdgpu_device_ip_wait_for_idle(struct amdgpu_device *adev, enum amd_ip_block_type block_type); -bool amdgpu_device_ip_is_idle(struct amdgpu_device *adev, +bool amdgpu_device_ip_is_valid(struct amdgpu_device *adev, enum amd_ip_block_type block_type); +int amdgpu_ip_block_suspend(struct amdgpu_ip_block *ip_block); + +int amdgpu_ip_block_resume(struct amdgpu_ip_block *ip_block); #define AMDGPU_MAX_IP_NUM 16 @@ -389,6 +388,7 @@ struct amdgpu_ip_block_version { struct amdgpu_ip_block { struct amdgpu_ip_block_status status; const struct amdgpu_ip_block_version *version; + struct amdgpu_device *adev; }; int amdgpu_device_ip_block_version_cmp(struct amdgpu_device *adev, @@ -563,6 +563,7 @@ enum amd_reset_method { AMD_RESET_METHOD_MODE2, AMD_RESET_METHOD_BACO, AMD_RESET_METHOD_PCI, + AMD_RESET_METHOD_ON_INIT, }; struct amdgpu_video_codec_info { @@ -821,6 +822,24 @@ struct amdgpu_mqd { struct amdgpu_mqd_prop *p); }; +/* + * Custom Init levels could be defined for different situations where a full + * initialization of all hardware blocks are not expected. Sample cases are + * custom init sequences after resume after S0i3/S3, reset on initialization, + * partial reset of blocks etc. Presently, this defines only two levels. Levels + * are described in corresponding struct definitions - amdgpu_init_default, + * amdgpu_init_minimal_xgmi. + */ +enum amdgpu_init_lvl_id { + AMDGPU_INIT_LEVEL_DEFAULT, + AMDGPU_INIT_LEVEL_MINIMAL_XGMI, +}; + +struct amdgpu_init_level { + enum amdgpu_init_lvl_id level; + uint32_t hwini_ip_block_mask; +}; + #define AMDGPU_RESET_MAGIC_NUM 64 #define AMDGPU_MAX_DF_PERFMONS 4 struct amdgpu_reset_domain; @@ -1166,6 +1185,8 @@ struct amdgpu_device { bool enforce_isolation[MAX_XCP]; /* Added this mutex for cleaner shader isolation between GFX and compute processes */ struct mutex enforce_isolation_mutex; + + struct amdgpu_init_level *init_lvl; }; static inline uint32_t amdgpu_ip_version(const struct amdgpu_device *adev, @@ -1261,6 +1282,8 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, int amdgpu_do_asic_reset(struct list_head *device_list_handle, struct amdgpu_reset_context *reset_context); +int amdgpu_device_reinit_after_reset(struct amdgpu_reset_context *reset_context); + int emu_soc_asic_init(struct amdgpu_device *adev); /* @@ -1450,23 +1473,15 @@ void amdgpu_register_atpx_handler(void); void amdgpu_unregister_atpx_handler(void); bool amdgpu_has_atpx_dgpu_power_cntl(void); bool amdgpu_is_atpx_hybrid(void); -bool amdgpu_atpx_dgpu_req_power_for_displays(void); bool amdgpu_has_atpx(void); #else static inline void amdgpu_register_atpx_handler(void) {} static inline void amdgpu_unregister_atpx_handler(void) {} static inline bool amdgpu_has_atpx_dgpu_power_cntl(void) { return false; } static inline bool amdgpu_is_atpx_hybrid(void) { return false; } -static inline bool amdgpu_atpx_dgpu_req_power_for_displays(void) { return false; } static inline bool amdgpu_has_atpx(void) { return false; } #endif -#if defined(CONFIG_VGA_SWITCHEROO) && defined(CONFIG_ACPI) -void *amdgpu_atpx_get_dhandle(void); -#else -static inline void *amdgpu_atpx_get_dhandle(void) { return NULL; } -#endif - /* * KMS */ @@ -1619,4 +1634,6 @@ extern const struct attribute_group amdgpu_vram_mgr_attr_group; extern const struct attribute_group amdgpu_gtt_mgr_attr_group; extern const struct attribute_group amdgpu_flash_attr_group; +void amdgpu_set_init_level(struct amdgpu_device *adev, + enum amdgpu_init_lvl_id lvl); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c index bf6c4a0d0525..ec5e0dcf8613 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c @@ -98,9 +98,9 @@ enum { ACP_TILE_DSP2, }; -static int acp_sw_init(void *handle) +static int acp_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->acp.parent = adev->dev; @@ -112,9 +112,9 @@ static int acp_sw_init(void *handle) return 0; } -static int acp_sw_fini(void *handle) +static int acp_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->acp.cgs_device) amdgpu_cgs_destroy_device(adev->acp.cgs_device); @@ -219,10 +219,10 @@ static const struct dmi_system_id acp_quirk_table[] = { /** * acp_hw_init - start and test ACP block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -static int acp_hw_init(void *handle) +static int acp_hw_init(struct amdgpu_ip_block *ip_block) { int r; u64 acp_base; @@ -230,13 +230,7 @@ static int acp_hw_init(void *handle) u32 count = 0; struct i2s_platform_data *i2s_pdata = NULL; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - const struct amdgpu_ip_block *ip_block = - amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_ACP); - - if (!ip_block) - return -EINVAL; + struct amdgpu_device *adev = ip_block->adev; r = amd_acp_hw_init(adev->acp.cgs_device, ip_block->version->major, ip_block->version->minor); @@ -503,14 +497,14 @@ failure: /** * acp_hw_fini - stop the hardware block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -static int acp_hw_fini(void *handle) +static int acp_hw_fini(struct amdgpu_ip_block *ip_block) { u32 val = 0; u32 count = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* return early if no ACP */ if (!adev->acp.acp_genpd) { @@ -565,9 +559,9 @@ static int acp_hw_fini(void *handle) return 0; } -static int acp_suspend(void *handle) +static int acp_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* power up on suspend */ if (!adev->acp.acp_cell) @@ -575,9 +569,9 @@ static int acp_suspend(void *handle) return 0; } -static int acp_resume(void *handle) +static int acp_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* power down again on resume */ if (!adev->acp.acp_cell) @@ -585,26 +579,11 @@ static int acp_resume(void *handle) return 0; } -static int acp_early_init(void *handle) -{ - return 0; -} - static bool acp_is_idle(void *handle) { return true; } -static int acp_wait_for_idle(void *handle) -{ - return 0; -} - -static int acp_soft_reset(void *handle) -{ - return 0; -} - static int acp_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -624,8 +603,6 @@ static int acp_set_powergating_state(void *handle, static const struct amd_ip_funcs acp_ip_funcs = { .name = "acp_ip", - .early_init = acp_early_init, - .late_init = NULL, .sw_init = acp_sw_init, .sw_fini = acp_sw_fini, .hw_init = acp_hw_init, @@ -633,12 +610,8 @@ static const struct amd_ip_funcs acp_ip_funcs = { .suspend = acp_suspend, .resume = acp_resume, .is_idle = acp_is_idle, - .wait_for_idle = acp_wait_for_idle, - .soft_reset = acp_soft_reset, .set_clockgating_state = acp_set_clockgating_state, .set_powergating_state = acp_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; const struct amdgpu_ip_block_version acp_ip_block = { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index f85ace0384d2..cce85389427f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -147,6 +147,7 @@ static union acpi_object *amdgpu_atif_call(struct amdgpu_atif *atif, struct acpi_buffer *params) { acpi_status status; + union acpi_object *obj; union acpi_object atif_arg_elements[2]; struct acpi_object_list atif_arg; struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; @@ -169,16 +170,24 @@ static union acpi_object *amdgpu_atif_call(struct amdgpu_atif *atif, status = acpi_evaluate_object(atif->handle, NULL, &atif_arg, &buffer); + obj = (union acpi_object *)buffer.pointer; - /* Fail only if calling the method fails and ATIF is supported */ + /* Fail if calling the method fails and ATIF is supported */ if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) { DRM_DEBUG_DRIVER("failed to evaluate ATIF got %s\n", acpi_format_exception(status)); - kfree(buffer.pointer); + kfree(obj); return NULL; } - return buffer.pointer; + if (obj->type != ACPI_TYPE_BUFFER) { + DRM_DEBUG_DRIVER("bad object returned from ATIF: %d\n", + obj->type); + kfree(obj); + return NULL; + } + + return obj; } /** @@ -791,6 +800,7 @@ int amdgpu_acpi_power_shift_control(struct amdgpu_device *adev, return -EIO; } + kfree(info); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index 4f08b153cb66..b545940e512b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -889,3 +889,18 @@ int amdgpu_amdkfd_start_sched(struct amdgpu_device *adev, uint32_t node_id) return kgd2kfd_start_sched(adev->kfd.dev, node_id); } + +/* Config CGTT_SQ_CLK_CTRL */ +int amdgpu_amdkfd_config_sq_perfmon(struct amdgpu_device *adev, uint32_t xcp_id, + bool core_override_enable, bool reg_override_enable, bool perfmon_override_enable) +{ + int r; + + if (!adev->kfd.init_complete) + return 0; + + r = psp_config_sq_perfmon(&adev->psp, xcp_id, core_override_enable, + reg_override_enable, perfmon_override_enable); + + return r; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index f9d119448442..7e0a22072536 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -266,6 +266,9 @@ int amdgpu_amdkfd_unmap_hiq(struct amdgpu_device *adev, u32 doorbell_off, u32 inst); int amdgpu_amdkfd_start_sched(struct amdgpu_device *adev, uint32_t node_id); int amdgpu_amdkfd_stop_sched(struct amdgpu_device *adev, uint32_t node_id); +int amdgpu_amdkfd_config_sq_perfmon(struct amdgpu_device *adev, uint32_t xcp_id, + bool core_override_enable, bool reg_override_enable, bool perfmon_override_enable); + /* Read user wptr from a specified user address space with page fault * disabled. The memory must be pinned and mapped to the hardware when diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c index 9435af2e6bdc..9abf29b58ac7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_arcturus.c @@ -299,7 +299,7 @@ static int suspend_resume_compute_scheduler(struct amdgpu_device *adev, bool sus if (r) goto out; } else { - drm_sched_start(&ring->sched); + drm_sched_start(&ring->sched, 0); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index 3bc0cbf45bc5..cc66ebb7bae1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -944,9 +944,7 @@ static void unlock_spi_csq_mutexes(struct amdgpu_device *adev) * * @adev: Handle of device whose registers are to be read * @queue_idx: Index of queue in the queue-map bit-field - * @wave_cnt: Output parameter updated with number of waves in flight - * @vmid: Output parameter updated with VMID of queue whose wave count - * is being collected + * @queue_cnt: Stores the wave count and doorbell offset for an active queue * @inst: xcc's instance number on a multi-XCC setup */ static void get_wave_count(struct amdgpu_device *adev, int queue_idx, @@ -1133,10 +1131,6 @@ uint64_t kgd_gfx_v9_hqd_get_pq_addr(struct amdgpu_device *adev, uint32_t low, high; uint64_t queue_addr = 0; - if (!adev->debug_exp_resets && - !adev->gfx.num_gfx_rings) - return 0; - kgd_gfx_v9_acquire_queue(adev, pipe_id, queue_id, inst); amdgpu_gfx_rlc_enter_safe_mode(adev, inst); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index ce5ca304dba9..f30548f4c3b3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -1439,8 +1439,8 @@ static int init_kfd_vm(struct amdgpu_vm *vm, void **process_info, list_add_tail(&vm->vm_list_node, &(vm->process_info->vm_list_head)); vm->process_info->n_vms++; - - *ef = dma_fence_get(&vm->process_info->eviction_fence->base); + if (ef) + *ef = dma_fence_get(&vm->process_info->eviction_fence->base); mutex_unlock(&vm->process_info->lock); return 0; @@ -2524,11 +2524,14 @@ int amdgpu_amdkfd_evict_userptr(struct mmu_interval_notifier *mni, /* First eviction, stop the queues */ r = kgd2kfd_quiesce_mm(mni->mm, KFD_QUEUE_EVICTION_TRIGGER_USERPTR); - if (r) + + if (r && r != -ESRCH) pr_err("Failed to quiesce KFD\n"); - queue_delayed_work(system_freezable_wq, - &process_info->restore_userptr_work, - msecs_to_jiffies(AMDGPU_USERPTR_RESTORE_DELAY_MS)); + + if (r != -ESRCH) + queue_delayed_work(system_freezable_wq, + &process_info->restore_userptr_work, + msecs_to_jiffies(AMDGPU_USERPTR_RESTORE_DELAY_MS)); } mutex_unlock(&process_info->notifier_lock); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c index 0c8975ac5af9..093141ad6ed0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c @@ -1145,8 +1145,8 @@ int amdgpu_atombios_get_memory_pll_dividers(struct amdgpu_device *adev, return 0; } -void amdgpu_atombios_set_engine_dram_timings(struct amdgpu_device *adev, - u32 eng_clock, u32 mem_clock) +int amdgpu_atombios_set_engine_dram_timings(struct amdgpu_device *adev, + u32 eng_clock, u32 mem_clock) { SET_ENGINE_CLOCK_PS_ALLOCATION args; int index = GetIndexIntoMasterTable(COMMAND, DynamicMemorySettings); @@ -1161,8 +1161,8 @@ void amdgpu_atombios_set_engine_dram_timings(struct amdgpu_device *adev, if (mem_clock) args.sReserved.ulClock = cpu_to_le32(mem_clock & SET_CLOCK_FREQ_MASK); - amdgpu_atom_execute_table(adev->mode_info.atom_context, index, (uint32_t *)&args, - sizeof(args)); + return amdgpu_atom_execute_table(adev->mode_info.atom_context, index, + (uint32_t *)&args, sizeof(args)); } void amdgpu_atombios_get_default_voltages(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h index 0811474e8fd3..0e16432d9a72 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.h @@ -163,8 +163,8 @@ int amdgpu_atombios_get_memory_pll_dividers(struct amdgpu_device *adev, bool strobe_mode, struct atom_mpll_param *mpll_param); -void amdgpu_atombios_set_engine_dram_timings(struct amdgpu_device *adev, - u32 eng_clock, u32 mem_clock); +int amdgpu_atombios_set_engine_dram_timings(struct amdgpu_device *adev, + u32 eng_clock, u32 mem_clock); bool amdgpu_atombios_is_voltage_gpio(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c index 375f02002579..3893e6fc2f03 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c @@ -89,18 +89,6 @@ bool amdgpu_is_atpx_hybrid(void) return amdgpu_atpx_priv.atpx.is_hybrid; } -bool amdgpu_atpx_dgpu_req_power_for_displays(void) -{ - return amdgpu_atpx_priv.atpx.dgpu_req_power_for_displays; -} - -#if defined(CONFIG_ACPI) -void *amdgpu_atpx_get_dhandle(void) -{ - return amdgpu_atpx_priv.dhandle; -} -#endif - /** * amdgpu_atpx_call - call an ATPX method * diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 1e475eb01417..d891ab779ca7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -265,7 +265,7 @@ static int amdgpu_cs_pass1(struct amdgpu_cs_parser *p, /* Only a single BO list is allowed to simplify handling. */ if (p->bo_list) - ret = -EINVAL; + goto free_partial_kdata; ret = amdgpu_cs_p1_bo_handles(p, p->chunks[i].kdata); if (ret) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index cbef720de779..37d8657f0776 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -2095,6 +2095,8 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev) if (amdgpu_umsch_mm & amdgpu_umsch_mm_fwlog) amdgpu_debugfs_umsch_fwlog_init(adev, &adev->umsch_mm); + amdgpu_debugfs_jpeg_sched_mask_init(adev); + amdgpu_ras_debugfs_create_all(adev); amdgpu_rap_debugfs_init(adev); amdgpu_securedisplay_debugfs_init(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_dev_coredump.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_dev_coredump.c index 5ac59b62020c..946c48829f19 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_dev_coredump.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_dev_coredump.c @@ -203,6 +203,7 @@ amdgpu_devcoredump_read(char *buffer, loff_t offset, size_t count, struct amdgpu_coredump_info *coredump = data; struct drm_print_iterator iter; struct amdgpu_vm_fault_info *fault_info; + struct amdgpu_ip_block *ip_block; int ver; iter.data = buffer; @@ -282,13 +283,10 @@ amdgpu_devcoredump_read(char *buffer, loff_t offset, size_t count, /* dump the ip state for each ip */ drm_printf(&p, "IP Dump\n"); for (int i = 0; i < coredump->adev->num_ip_blocks; i++) { - if (coredump->adev->ip_blocks[i].version->funcs->print_ip_state) { - drm_printf(&p, "IP: %s\n", - coredump->adev->ip_blocks[i] - .version->funcs->name); - coredump->adev->ip_blocks[i] - .version->funcs->print_ip_state( - (void *)coredump->adev, &p); + ip_block = &coredump->adev->ip_blocks[i]; + if (ip_block->version->funcs->print_ip_state) { + drm_printf(&p, "IP: %s\n", ip_block->version->funcs->name); + ip_block->version->funcs->print_ip_state(ip_block, &p); drm_printf(&p, "\n"); } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index c2394c8b4d6b..f12fab13386a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -25,6 +25,8 @@ * Alex Deucher * Jerome Glisse */ + +#include #include #include #include @@ -35,10 +37,9 @@ #include #include -#include #include +#include #include -#include #include #include #include @@ -144,6 +145,51 @@ const char *amdgpu_asic_name[] = { "LAST", }; +#define AMDGPU_IP_BLK_MASK_ALL GENMASK(AMDGPU_MAX_IP_NUM - 1, 0) +/* + * Default init level where all blocks are expected to be initialized. This is + * the level of initialization expected by default and also after a full reset + * of the device. + */ +struct amdgpu_init_level amdgpu_init_default = { + .level = AMDGPU_INIT_LEVEL_DEFAULT, + .hwini_ip_block_mask = AMDGPU_IP_BLK_MASK_ALL, +}; + +/* + * Minimal blocks needed to be initialized before a XGMI hive can be reset. This + * is used for cases like reset on initialization where the entire hive needs to + * be reset before first use. + */ +struct amdgpu_init_level amdgpu_init_minimal_xgmi = { + .level = AMDGPU_INIT_LEVEL_MINIMAL_XGMI, + .hwini_ip_block_mask = + BIT(AMD_IP_BLOCK_TYPE_GMC) | BIT(AMD_IP_BLOCK_TYPE_SMC) | + BIT(AMD_IP_BLOCK_TYPE_COMMON) | BIT(AMD_IP_BLOCK_TYPE_IH) | + BIT(AMD_IP_BLOCK_TYPE_PSP) +}; + +static inline bool amdgpu_ip_member_of_hwini(struct amdgpu_device *adev, + enum amd_ip_block_type block) +{ + return (adev->init_lvl->hwini_ip_block_mask & (1U << block)) != 0; +} + +void amdgpu_set_init_level(struct amdgpu_device *adev, + enum amdgpu_init_lvl_id lvl) +{ + switch (lvl) { + case AMDGPU_INIT_LEVEL_MINIMAL_XGMI: + adev->init_lvl = &amdgpu_init_minimal_xgmi; + break; + case AMDGPU_INIT_LEVEL_DEFAULT: + fallthrough; + default: + adev->init_lvl = &amdgpu_init_default; + break; + } +} + static inline void amdgpu_device_stop_pending_resets(struct amdgpu_device *adev); /** @@ -227,6 +273,42 @@ void amdgpu_reg_state_sysfs_fini(struct amdgpu_device *adev) sysfs_remove_bin_file(&adev->dev->kobj, &bin_attr_reg_state); } +int amdgpu_ip_block_suspend(struct amdgpu_ip_block *ip_block) +{ + int r; + + if (ip_block->version->funcs->suspend) { + r = ip_block->version->funcs->suspend(ip_block); + if (r) { + dev_err(ip_block->adev->dev, + "suspend of IP block <%s> failed %d\n", + ip_block->version->funcs->name, r); + return r; + } + } + + ip_block->status.hw = false; + return 0; +} + +int amdgpu_ip_block_resume(struct amdgpu_ip_block *ip_block) +{ + int r; + + if (ip_block->version->funcs->resume) { + r = ip_block->version->funcs->resume(ip_block); + if (r) { + dev_err(ip_block->adev->dev, + "resume of IP block <%s> failed %d\n", + ip_block->version->funcs->name, r); + return r; + } + } + + ip_block->status.hw = true; + return 0; +} + /** * DOC: board_info * @@ -1655,7 +1737,7 @@ bool amdgpu_device_need_post(struct amdgpu_device *adev) } /* Don't post if we need to reset whole hive on init */ - if (adev->gmc.xgmi.pending_reset) + if (adev->init_lvl->level == AMDGPU_INIT_LEVEL_MINIMAL_XGMI) return false; if (adev->has_hw_reset) { @@ -2159,9 +2241,12 @@ int amdgpu_device_ip_wait_for_idle(struct amdgpu_device *adev, if (!adev->ip_blocks[i].status.valid) continue; if (adev->ip_blocks[i].version->type == block_type) { - r = adev->ip_blocks[i].version->funcs->wait_for_idle((void *)adev); - if (r) - return r; + if (adev->ip_blocks[i].version->funcs->wait_for_idle) { + r = adev->ip_blocks[i].version->funcs->wait_for_idle( + &adev->ip_blocks[i]); + if (r) + return r; + } break; } } @@ -2170,26 +2255,24 @@ int amdgpu_device_ip_wait_for_idle(struct amdgpu_device *adev, } /** - * amdgpu_device_ip_is_idle - is the hardware IP idle + * amdgpu_device_ip_is_valid - is the hardware IP enabled * * @adev: amdgpu_device pointer * @block_type: Type of hardware IP (SMU, GFX, UVD, etc.) * - * Check if the hardware IP is idle or not. - * Returns true if it the IP is idle, false if not. + * Check if the hardware IP is enable or not. + * Returns true if it the IP is enable, false if not. */ -bool amdgpu_device_ip_is_idle(struct amdgpu_device *adev, - enum amd_ip_block_type block_type) +bool amdgpu_device_ip_is_valid(struct amdgpu_device *adev, + enum amd_ip_block_type block_type) { int i; for (i = 0; i < adev->num_ip_blocks; i++) { - if (!adev->ip_blocks[i].status.valid) - continue; if (adev->ip_blocks[i].version->type == block_type) - return adev->ip_blocks[i].version->funcs->is_idle((void *)adev); + return adev->ip_blocks[i].status.valid; } - return true; + return false; } @@ -2271,6 +2354,8 @@ int amdgpu_device_ip_block_add(struct amdgpu_device *adev, DRM_INFO("add ip block number %d <%s>\n", adev->num_ip_blocks, ip_block_version->funcs->name); + adev->ip_blocks[adev->num_ip_blocks].adev = adev; + adev->ip_blocks[adev->num_ip_blocks++].version = ip_block_version; return 0; @@ -2566,25 +2651,25 @@ static int amdgpu_device_ip_early_init(struct amdgpu_device *adev) total = true; for (i = 0; i < adev->num_ip_blocks; i++) { + ip_block = &adev->ip_blocks[i]; + if ((amdgpu_ip_block_mask & (1 << i)) == 0) { DRM_WARN("disabled ip block: %d <%s>\n", i, adev->ip_blocks[i].version->funcs->name); adev->ip_blocks[i].status.valid = false; - } else { - if (adev->ip_blocks[i].version->funcs->early_init) { - r = adev->ip_blocks[i].version->funcs->early_init((void *)adev); - if (r == -ENOENT) { - adev->ip_blocks[i].status.valid = false; - } else if (r) { - DRM_ERROR("early_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); - total = false; - } else { - adev->ip_blocks[i].status.valid = true; - } + } else if (ip_block->version->funcs->early_init) { + r = ip_block->version->funcs->early_init(ip_block); + if (r == -ENOENT) { + adev->ip_blocks[i].status.valid = false; + } else if (r) { + DRM_ERROR("early_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); + total = false; } else { adev->ip_blocks[i].status.valid = true; } + } else { + adev->ip_blocks[i].status.valid = true; } /* get the vbios after the asic_funcs are set up */ if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_COMMON) { @@ -2633,10 +2718,13 @@ static int amdgpu_device_ip_hw_init_phase1(struct amdgpu_device *adev) continue; if (adev->ip_blocks[i].status.hw) continue; + if (!amdgpu_ip_member_of_hwini( + adev, adev->ip_blocks[i].version->type)) + continue; if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_COMMON || (amdgpu_sriov_vf(adev) && (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_PSP)) || adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH) { - r = adev->ip_blocks[i].version->funcs->hw_init(adev); + r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { DRM_ERROR("hw_init of IP block <%s> failed %d\n", adev->ip_blocks[i].version->funcs->name, r); @@ -2658,7 +2746,10 @@ static int amdgpu_device_ip_hw_init_phase2(struct amdgpu_device *adev) continue; if (adev->ip_blocks[i].status.hw) continue; - r = adev->ip_blocks[i].version->funcs->hw_init(adev); + if (!amdgpu_ip_member_of_hwini( + adev, adev->ip_blocks[i].version->type)) + continue; + r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { DRM_ERROR("hw_init of IP block <%s> failed %d\n", adev->ip_blocks[i].version->funcs->name, r); @@ -2681,6 +2772,10 @@ static int amdgpu_device_fw_loading(struct amdgpu_device *adev) if (adev->ip_blocks[i].version->type != AMD_IP_BLOCK_TYPE_PSP) continue; + if (!amdgpu_ip_member_of_hwini(adev, + AMD_IP_BLOCK_TYPE_PSP)) + break; + if (!adev->ip_blocks[i].status.sw) continue; @@ -2689,22 +2784,18 @@ static int amdgpu_device_fw_loading(struct amdgpu_device *adev) break; if (amdgpu_in_reset(adev) || adev->in_suspend) { - r = adev->ip_blocks[i].version->funcs->resume(adev); - if (r) { - DRM_ERROR("resume of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + r = amdgpu_ip_block_resume(&adev->ip_blocks[i]); + if (r) return r; - } } else { - r = adev->ip_blocks[i].version->funcs->hw_init(adev); + r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { DRM_ERROR("hw_init of IP block <%s> failed %d\n", adev->ip_blocks[i].version->funcs->name, r); return r; } + adev->ip_blocks[i].status.hw = true; } - - adev->ip_blocks[i].status.hw = true; break; } } @@ -2786,6 +2877,7 @@ static int amdgpu_device_init_schedulers(struct amdgpu_device *adev) */ static int amdgpu_device_ip_init(struct amdgpu_device *adev) { + bool init_badpage; int i, r; r = amdgpu_ras_init(adev); @@ -2795,17 +2887,23 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) for (i = 0; i < adev->num_ip_blocks; i++) { if (!adev->ip_blocks[i].status.valid) continue; - r = adev->ip_blocks[i].version->funcs->sw_init((void *)adev); - if (r) { - DRM_ERROR("sw_init of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); - goto init_failed; + if (adev->ip_blocks[i].version->funcs->sw_init) { + r = adev->ip_blocks[i].version->funcs->sw_init(&adev->ip_blocks[i]); + if (r) { + DRM_ERROR("sw_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); + goto init_failed; + } } adev->ip_blocks[i].status.sw = true; + if (!amdgpu_ip_member_of_hwini( + adev, adev->ip_blocks[i].version->type)) + continue; + if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_COMMON) { /* need to do common hw init early so everything is set up for gmc */ - r = adev->ip_blocks[i].version->funcs->hw_init((void *)adev); + r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { DRM_ERROR("hw_init %d failed %d\n", i, r); goto init_failed; @@ -2822,7 +2920,7 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) DRM_ERROR("amdgpu_mem_scratch_init failed %d\n", r); goto init_failed; } - r = adev->ip_blocks[i].version->funcs->hw_init((void *)adev); + r = adev->ip_blocks[i].version->funcs->hw_init(&adev->ip_blocks[i]); if (r) { DRM_ERROR("hw_init %d failed %d\n", i, r); goto init_failed; @@ -2895,7 +2993,8 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) * Note: theoretically, this should be called before all vram allocations * to protect retired page from abusing */ - r = amdgpu_ras_recovery_init(adev); + init_badpage = (adev->init_lvl->level != AMDGPU_INIT_LEVEL_MINIMAL_XGMI); + r = amdgpu_ras_recovery_init(adev, init_badpage); if (r) goto init_failed; @@ -2935,7 +3034,7 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev) amdgpu_ttm_set_buffer_funcs_status(adev, true); /* Don't init kfd if whole hive need to be reset during init */ - if (!adev->gmc.xgmi.pending_reset) { + if (adev->init_lvl->level != AMDGPU_INIT_LEVEL_MINIMAL_XGMI) { kgd2kfd_init_zone_device(adev); amdgpu_amdkfd_device_init(adev); } @@ -3135,7 +3234,7 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) if (!adev->ip_blocks[i].status.hw) continue; if (adev->ip_blocks[i].version->funcs->late_init) { - r = adev->ip_blocks[i].version->funcs->late_init((void *)adev); + r = adev->ip_blocks[i].version->funcs->late_init(&adev->ip_blocks[i]); if (r) { DRM_ERROR("late_init of IP block <%s> failed %d\n", adev->ip_blocks[i].version->funcs->name, r); @@ -3206,6 +3305,25 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) return 0; } +static void amdgpu_ip_block_hw_fini(struct amdgpu_ip_block *ip_block) +{ + int r; + + if (!ip_block->version->funcs->hw_fini) { + DRM_ERROR("hw_fini of IP block <%s> not defined\n", + ip_block->version->funcs->name); + } else { + r = ip_block->version->funcs->hw_fini(ip_block); + /* XXX handle errors */ + if (r) { + DRM_DEBUG("hw_fini of IP block <%s> failed %d\n", + ip_block->version->funcs->name, r); + } + } + + ip_block->status.hw = false; +} + /** * amdgpu_device_smu_fini_early - smu hw_fini wrapper * @@ -3215,7 +3333,7 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) */ static void amdgpu_device_smu_fini_early(struct amdgpu_device *adev) { - int i, r; + int i; if (amdgpu_ip_version(adev, GC_HWIP, 0) > IP_VERSION(9, 0, 0)) return; @@ -3224,13 +3342,7 @@ static void amdgpu_device_smu_fini_early(struct amdgpu_device *adev) if (!adev->ip_blocks[i].status.hw) continue; if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) { - r = adev->ip_blocks[i].version->funcs->hw_fini((void *)adev); - /* XXX handle errors */ - if (r) { - DRM_DEBUG("hw_fini of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); - } - adev->ip_blocks[i].status.hw = false; + amdgpu_ip_block_hw_fini(&adev->ip_blocks[i]); break; } } @@ -3244,7 +3356,7 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev) if (!adev->ip_blocks[i].version->funcs->early_fini) continue; - r = adev->ip_blocks[i].version->funcs->early_fini((void *)adev); + r = adev->ip_blocks[i].version->funcs->early_fini(&adev->ip_blocks[i]); if (r) { DRM_DEBUG("early_fini of IP block <%s> failed %d\n", adev->ip_blocks[i].version->funcs->name, r); @@ -3263,14 +3375,7 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev) if (!adev->ip_blocks[i].status.hw) continue; - r = adev->ip_blocks[i].version->funcs->hw_fini((void *)adev); - /* XXX handle errors */ - if (r) { - DRM_DEBUG("hw_fini of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); - } - - adev->ip_blocks[i].status.hw = false; + amdgpu_ip_block_hw_fini(&adev->ip_blocks[i]); } if (amdgpu_sriov_vf(adev)) { @@ -3316,12 +3421,13 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev) amdgpu_ib_pool_fini(adev); amdgpu_seq64_fini(adev); } - - r = adev->ip_blocks[i].version->funcs->sw_fini((void *)adev); - /* XXX handle errors */ - if (r) { - DRM_DEBUG("sw_fini of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + if (adev->ip_blocks[i].version->funcs->sw_fini) { + r = adev->ip_blocks[i].version->funcs->sw_fini(&adev->ip_blocks[i]); + /* XXX handle errors */ + if (r) { + DRM_DEBUG("sw_fini of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); + } } adev->ip_blocks[i].status.sw = false; adev->ip_blocks[i].status.valid = false; @@ -3331,7 +3437,7 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev) if (!adev->ip_blocks[i].status.late_initialized) continue; if (adev->ip_blocks[i].version->funcs->late_fini) - adev->ip_blocks[i].version->funcs->late_fini((void *)adev); + adev->ip_blocks[i].version->funcs->late_fini(&adev->ip_blocks[i]); adev->ip_blocks[i].status.late_initialized = false; } @@ -3403,15 +3509,9 @@ static int amdgpu_device_ip_suspend_phase1(struct amdgpu_device *adev) continue; /* XXX handle errors */ - r = adev->ip_blocks[i].version->funcs->suspend(adev); - /* XXX handle errors */ - if (r) { - DRM_ERROR("suspend of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + r = amdgpu_ip_block_suspend(&adev->ip_blocks[i]); + if (r) return r; - } - - adev->ip_blocks[i].status.hw = false; } return 0; @@ -3449,14 +3549,9 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev) } /* skip unnecessary suspend if we do not initialize them yet */ - if (adev->gmc.xgmi.pending_reset && - !(adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GMC || - adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC || - adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_COMMON || - adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH)) { - adev->ip_blocks[i].status.hw = false; + if (!amdgpu_ip_member_of_hwini( + adev, adev->ip_blocks[i].version->type)) continue; - } /* skip suspend of gfx/mes and psp for S0ix * gfx is in gfxoff state, so on resume it will exit gfxoff just @@ -3490,13 +3585,9 @@ static int amdgpu_device_ip_suspend_phase2(struct amdgpu_device *adev) continue; /* XXX handle errors */ - r = adev->ip_blocks[i].version->funcs->suspend(adev); - /* XXX handle errors */ - if (r) { - DRM_ERROR("suspend of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); - } + r = amdgpu_ip_block_suspend(&adev->ip_blocks[i]); adev->ip_blocks[i].status.hw = false; + /* handle putting the SMC in the appropriate state */ if (!amdgpu_sriov_vf(adev)) { if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) { @@ -3570,7 +3661,7 @@ static int amdgpu_device_ip_reinit_early_sriov(struct amdgpu_device *adev) !block->status.valid) continue; - r = block->version->funcs->hw_init(adev); + r = block->version->funcs->hw_init(&adev->ip_blocks[i]); DRM_INFO("RE-INIT-early: %s %s\n", block->version->funcs->name, r?"failed":"succeeded"); if (r) return r; @@ -3609,15 +3700,19 @@ static int amdgpu_device_ip_reinit_late_sriov(struct amdgpu_device *adev) block->status.hw) continue; - if (block->version->type == AMD_IP_BLOCK_TYPE_SMC) - r = block->version->funcs->resume(adev); - else - r = block->version->funcs->hw_init(adev); - - DRM_INFO("RE-INIT-late: %s %s\n", block->version->funcs->name, r?"failed":"succeeded"); - if (r) - return r; - block->status.hw = true; + if (block->version->type == AMD_IP_BLOCK_TYPE_SMC) { + r = amdgpu_ip_block_resume(&adev->ip_blocks[i]); + if (r) + return r; + } else { + r = block->version->funcs->hw_init(&adev->ip_blocks[i]); + if (r) { + DRM_ERROR("hw_init of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); + return r; + } + block->status.hw = true; + } } } @@ -3648,13 +3743,9 @@ static int amdgpu_device_ip_resume_phase1(struct amdgpu_device *adev) adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH || (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_PSP && amdgpu_sriov_vf(adev))) { - r = adev->ip_blocks[i].version->funcs->resume(adev); - if (r) { - DRM_ERROR("resume of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + r = amdgpu_ip_block_resume(&adev->ip_blocks[i]); + if (r) return r; - } - adev->ip_blocks[i].status.hw = true; } } @@ -3686,13 +3777,9 @@ static int amdgpu_device_ip_resume_phase2(struct amdgpu_device *adev) adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH || adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_PSP) continue; - r = adev->ip_blocks[i].version->funcs->resume(adev); - if (r) { - DRM_ERROR("resume of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + r = amdgpu_ip_block_resume(&adev->ip_blocks[i]); + if (r) return r; - } - adev->ip_blocks[i].status.hw = true; } return 0; @@ -4193,13 +4280,19 @@ int amdgpu_device_init(struct amdgpu_device *adev, amdgpu_device_set_mcbp(adev); + /* + * By default, use default mode where all blocks are expected to be + * initialized. At present a 'swinit' of blocks is required to be + * completed before the need for a different level is detected. + */ + amdgpu_set_init_level(adev, AMDGPU_INIT_LEVEL_DEFAULT); /* early init functions */ r = amdgpu_device_ip_early_init(adev); if (r) return r; /* Get rid of things like offb */ - r = drm_aperture_remove_conflicting_pci_framebuffers(adev->pdev, &amdgpu_kms_driver); + r = aperture_remove_conflicting_pci_devices(adev->pdev, amdgpu_kms_driver.name); if (r) return r; @@ -4265,20 +4358,8 @@ int amdgpu_device_init(struct amdgpu_device *adev, if (!amdgpu_sriov_vf(adev) && amdgpu_asic_need_reset_on_init(adev)) { if (adev->gmc.xgmi.num_physical_nodes) { dev_info(adev->dev, "Pending hive reset.\n"); - adev->gmc.xgmi.pending_reset = true; - /* Only need to init necessary block for SMU to handle the reset */ - for (i = 0; i < adev->num_ip_blocks; i++) { - if (!adev->ip_blocks[i].status.valid) - continue; - if (!(adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GMC || - adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_COMMON || - adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH || - adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC)) { - DRM_DEBUG("IP %s disabled for hw_init.\n", - adev->ip_blocks[i].version->funcs->name); - adev->ip_blocks[i].status.hw = true; - } - } + amdgpu_set_init_level(adev, + AMDGPU_INIT_LEVEL_MINIMAL_XGMI); } else if (amdgpu_ip_version(adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 10) && !amdgpu_device_has_display_hardware(adev)) { r = psp_gpu_reset(adev); @@ -4386,7 +4467,7 @@ fence_driver_init: /* enable clockgating, etc. after ib tests, etc. since some blocks require * explicit gating rather than handling it automatically. */ - if (!adev->gmc.xgmi.pending_reset) { + if (adev->init_lvl->level != AMDGPU_INIT_LEVEL_MINIMAL_XGMI) { r = amdgpu_device_ip_late_init(adev); if (r) { dev_err(adev->dev, "amdgpu_device_ip_late_init failed\n"); @@ -4436,6 +4517,7 @@ fence_driver_init: amdgpu_fru_sysfs_init(adev); amdgpu_reg_state_sysfs_init(adev); + amdgpu_xcp_cfg_sysfs_init(adev); if (IS_ENABLED(CONFIG_PERF_EVENTS)) r = amdgpu_pmu_init(adev); @@ -4463,9 +4545,8 @@ fence_driver_init: if (px) vga_switcheroo_init_domain_pm_ops(adev->dev, &adev->vga_pm_domain); - if (adev->gmc.xgmi.pending_reset) - queue_delayed_work(system_wq, &mgpu_info.delayed_reset_work, - msecs_to_jiffies(AMDGPU_RESUME_MS)); + if (adev->init_lvl->level == AMDGPU_INIT_LEVEL_MINIMAL_XGMI) + amdgpu_xgmi_reset_on_init(adev); amdgpu_device_check_iommu_direct_map(adev); @@ -4559,6 +4640,7 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev) amdgpu_fru_sysfs_fini(adev); amdgpu_reg_state_sysfs_fini(adev); + amdgpu_xcp_cfg_sysfs_fini(adev); /* disable ras feature must before hw fini */ amdgpu_ras_pre_fini(adev); @@ -4694,7 +4776,7 @@ int amdgpu_device_prepare(struct drm_device *dev) continue; if (!adev->ip_blocks[i].version->funcs->prepare_suspend) continue; - r = adev->ip_blocks[i].version->funcs->prepare_suspend((void *)adev); + r = adev->ip_blocks[i].version->funcs->prepare_suspend(&adev->ip_blocks[i]); if (r) goto unprepare; } @@ -4711,13 +4793,13 @@ unprepare: * amdgpu_device_suspend - initiate device suspend * * @dev: drm dev pointer - * @fbcon : notify the fbdev of suspend + * @notify_clients: notify in-kernel DRM clients * * Puts the hw in the suspend state (all asics). * Returns 0 for success or an error on failure. * Called at driver suspend. */ -int amdgpu_device_suspend(struct drm_device *dev, bool fbcon) +int amdgpu_device_suspend(struct drm_device *dev, bool notify_clients) { struct amdgpu_device *adev = drm_to_adev(dev); int r = 0; @@ -4737,8 +4819,8 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon) if (amdgpu_acpi_smart_shift_update(dev, AMDGPU_SS_DEV_D3)) DRM_WARN("smart shift update failed\n"); - if (fbcon) - drm_fb_helper_set_suspend_unlocked(adev_to_drm(adev)->fb_helper, true); + if (notify_clients) + drm_client_dev_suspend(adev_to_drm(adev), false); cancel_delayed_work_sync(&adev->delayed_init_work); @@ -4773,13 +4855,13 @@ int amdgpu_device_suspend(struct drm_device *dev, bool fbcon) * amdgpu_device_resume - initiate device resume * * @dev: drm dev pointer - * @fbcon : notify the fbdev of resume + * @notify_clients: notify in-kernel DRM clients * * Bring the hw back to operating state (all asics). * Returns 0 for success or an error on failure. * Called at driver resume. */ -int amdgpu_device_resume(struct drm_device *dev, bool fbcon) +int amdgpu_device_resume(struct drm_device *dev, bool notify_clients) { struct amdgpu_device *adev = drm_to_adev(dev); int r = 0; @@ -4835,8 +4917,8 @@ exit: /* Make sure IB tests flushed */ flush_delayed_work(&adev->delayed_init_work); - if (fbcon) - drm_fb_helper_set_suspend_unlocked(adev_to_drm(adev)->fb_helper, false); + if (notify_clients) + drm_client_dev_resume(adev_to_drm(adev), false); amdgpu_ras_resume(adev); @@ -4898,7 +4980,8 @@ static bool amdgpu_device_ip_check_soft_reset(struct amdgpu_device *adev) continue; if (adev->ip_blocks[i].version->funcs->check_soft_reset) adev->ip_blocks[i].status.hang = - adev->ip_blocks[i].version->funcs->check_soft_reset(adev); + adev->ip_blocks[i].version->funcs->check_soft_reset( + &adev->ip_blocks[i]); if (adev->ip_blocks[i].status.hang) { dev_info(adev->dev, "IP block:%s is hung!\n", adev->ip_blocks[i].version->funcs->name); asic_hang = true; @@ -4927,7 +5010,7 @@ static int amdgpu_device_ip_pre_soft_reset(struct amdgpu_device *adev) continue; if (adev->ip_blocks[i].status.hang && adev->ip_blocks[i].version->funcs->pre_soft_reset) { - r = adev->ip_blocks[i].version->funcs->pre_soft_reset(adev); + r = adev->ip_blocks[i].version->funcs->pre_soft_reset(&adev->ip_blocks[i]); if (r) return r; } @@ -4989,7 +5072,7 @@ static int amdgpu_device_ip_soft_reset(struct amdgpu_device *adev) continue; if (adev->ip_blocks[i].status.hang && adev->ip_blocks[i].version->funcs->soft_reset) { - r = adev->ip_blocks[i].version->funcs->soft_reset(adev); + r = adev->ip_blocks[i].version->funcs->soft_reset(&adev->ip_blocks[i]); if (r) return r; } @@ -5018,7 +5101,7 @@ static int amdgpu_device_ip_post_soft_reset(struct amdgpu_device *adev) continue; if (adev->ip_blocks[i].status.hang && adev->ip_blocks[i].version->funcs->post_soft_reset) - r = adev->ip_blocks[i].version->funcs->post_soft_reset(adev); + r = adev->ip_blocks[i].version->funcs->post_soft_reset(&adev->ip_blocks[i]); if (r) return r; } @@ -5309,7 +5392,7 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, for (i = 0; i < tmp_adev->num_ip_blocks; i++) if (tmp_adev->ip_blocks[i].version->funcs->dump_ip_state) tmp_adev->ip_blocks[i].version->funcs - ->dump_ip_state((void *)tmp_adev); + ->dump_ip_state((void *)&tmp_adev->ip_blocks[i]); dev_info(tmp_adev->dev, "Dumping IP State Completed\n"); } @@ -5325,74 +5408,25 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, return r; } -int amdgpu_do_asic_reset(struct list_head *device_list_handle, - struct amdgpu_reset_context *reset_context) +int amdgpu_device_reinit_after_reset(struct amdgpu_reset_context *reset_context) { - struct amdgpu_device *tmp_adev = NULL; - bool need_full_reset, skip_hw_reset, vram_lost = false; - int r = 0; + struct list_head *device_list_handle; + bool full_reset, vram_lost = false; + struct amdgpu_device *tmp_adev; + int r; - /* Try reset handler method first */ - tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device, - reset_list); + device_list_handle = reset_context->reset_device_list; - reset_context->reset_device_list = device_list_handle; - r = amdgpu_reset_perform_reset(tmp_adev, reset_context); - /* If reset handler not implemented, continue; otherwise return */ - if (r == -EOPNOTSUPP) - r = 0; - else - return r; + if (!device_list_handle) + return -EINVAL; - /* Reset handler not implemented, use the default method */ - need_full_reset = - test_bit(AMDGPU_NEED_FULL_RESET, &reset_context->flags); - skip_hw_reset = test_bit(AMDGPU_SKIP_HW_RESET, &reset_context->flags); - - /* - * ASIC reset has to be done on all XGMI hive nodes ASAP - * to allow proper links negotiation in FW (within 1 sec) - */ - if (!skip_hw_reset && need_full_reset) { - list_for_each_entry(tmp_adev, device_list_handle, reset_list) { - /* For XGMI run all resets in parallel to speed up the process */ - if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) { - tmp_adev->gmc.xgmi.pending_reset = false; - if (!queue_work(system_unbound_wq, &tmp_adev->xgmi_reset_work)) - r = -EALREADY; - } else - r = amdgpu_asic_reset(tmp_adev); - - if (r) { - dev_err(tmp_adev->dev, "ASIC reset failed with error, %d for drm dev, %s", - r, adev_to_drm(tmp_adev)->unique); - goto out; - } - } - - /* For XGMI wait for all resets to complete before proceed */ - if (!r) { - list_for_each_entry(tmp_adev, device_list_handle, reset_list) { - if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) { - flush_work(&tmp_adev->xgmi_reset_work); - r = tmp_adev->asic_reset_res; - if (r) - break; - } - } - } - } - - if (!r && amdgpu_ras_intr_triggered()) { - list_for_each_entry(tmp_adev, device_list_handle, reset_list) { - amdgpu_ras_reset_error_count(tmp_adev, AMDGPU_RAS_BLOCK__MMHUB); - } - - amdgpu_ras_intr_cleared(); - } + full_reset = test_bit(AMDGPU_NEED_FULL_RESET, &reset_context->flags); + r = 0; list_for_each_entry(tmp_adev, device_list_handle, reset_list) { - if (need_full_reset) { + /* After reset, it's default init level */ + amdgpu_set_init_level(tmp_adev, AMDGPU_INIT_LEVEL_DEFAULT); + if (full_reset) { /* post card */ amdgpu_ras_set_fed(tmp_adev, false); r = amdgpu_device_asic_init(tmp_adev); @@ -5448,7 +5482,7 @@ int amdgpu_do_asic_reset(struct list_head *device_list_handle, if (r) goto out; - drm_fb_helper_set_suspend_unlocked(adev_to_drm(tmp_adev)->fb_helper, false); + drm_client_dev_resume(adev_to_drm(tmp_adev), false); /* * The GPU enters bad state once faulty pages @@ -5482,7 +5516,6 @@ out: r = amdgpu_ib_ring_tests(tmp_adev); if (r) { dev_err(tmp_adev->dev, "ib ring test failed (%d).\n", r); - need_full_reset = true; r = -EAGAIN; goto end; } @@ -5493,10 +5526,85 @@ out: } end: - if (need_full_reset) + return r; +} + +int amdgpu_do_asic_reset(struct list_head *device_list_handle, + struct amdgpu_reset_context *reset_context) +{ + struct amdgpu_device *tmp_adev = NULL; + bool need_full_reset, skip_hw_reset; + int r = 0; + + /* Try reset handler method first */ + tmp_adev = list_first_entry(device_list_handle, struct amdgpu_device, + reset_list); + + reset_context->reset_device_list = device_list_handle; + r = amdgpu_reset_perform_reset(tmp_adev, reset_context); + /* If reset handler not implemented, continue; otherwise return */ + if (r == -EOPNOTSUPP) + r = 0; + else + return r; + + /* Reset handler not implemented, use the default method */ + need_full_reset = + test_bit(AMDGPU_NEED_FULL_RESET, &reset_context->flags); + skip_hw_reset = test_bit(AMDGPU_SKIP_HW_RESET, &reset_context->flags); + + /* + * ASIC reset has to be done on all XGMI hive nodes ASAP + * to allow proper links negotiation in FW (within 1 sec) + */ + if (!skip_hw_reset && need_full_reset) { + list_for_each_entry(tmp_adev, device_list_handle, reset_list) { + /* For XGMI run all resets in parallel to speed up the process */ + if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) { + if (!queue_work(system_unbound_wq, + &tmp_adev->xgmi_reset_work)) + r = -EALREADY; + } else + r = amdgpu_asic_reset(tmp_adev); + + if (r) { + dev_err(tmp_adev->dev, + "ASIC reset failed with error, %d for drm dev, %s", + r, adev_to_drm(tmp_adev)->unique); + goto out; + } + } + + /* For XGMI wait for all resets to complete before proceed */ + if (!r) { + list_for_each_entry(tmp_adev, device_list_handle, + reset_list) { + if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) { + flush_work(&tmp_adev->xgmi_reset_work); + r = tmp_adev->asic_reset_res; + if (r) + break; + } + } + } + } + + if (!r && amdgpu_ras_intr_triggered()) { + list_for_each_entry(tmp_adev, device_list_handle, reset_list) { + amdgpu_ras_reset_error_count(tmp_adev, + AMDGPU_RAS_BLOCK__MMHUB); + } + + amdgpu_ras_intr_cleared(); + } + + r = amdgpu_device_reinit_after_reset(reset_context); + if (r == -EAGAIN) set_bit(AMDGPU_NEED_FULL_RESET, &reset_context->flags); else clear_bit(AMDGPU_NEED_FULL_RESET, &reset_context->flags); + +out: return r; } @@ -5734,7 +5842,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, */ amdgpu_unregister_gpu_instance(tmp_adev); - drm_fb_helper_set_suspend_unlocked(adev_to_drm(tmp_adev)->fb_helper, true); + drm_client_dev_suspend(adev_to_drm(tmp_adev), false); /* disable ras on ALL IPs */ if (!need_emergency_restart && @@ -5824,7 +5932,7 @@ skip_hw_reset: if (!amdgpu_ring_sched_ready(ring)) continue; - drm_sched_start(&ring->sched); + drm_sched_start(&ring->sched, 0); } if (!drm_drv_uses_atomic_modeset(adev_to_drm(tmp_adev)) && !job_signaled) @@ -6331,7 +6439,7 @@ void amdgpu_pci_resume(struct pci_dev *pdev) if (!amdgpu_ring_sched_ready(ring)) continue; - drm_sched_start(&ring->sched); + drm_sched_start(&ring->sched, 0); } amdgpu_device_unset_mp1_state(adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index 4bd61c169ca8..73f4d56c5de4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -1723,38 +1723,76 @@ union nps_info { struct nps_info_v1_0 v1; }; +static int amdgpu_discovery_refresh_nps_info(struct amdgpu_device *adev, + union nps_info *nps_data) +{ + uint64_t vram_size, pos, offset; + struct nps_info_header *nhdr; + struct binary_header bhdr; + uint16_t checksum; + + vram_size = (uint64_t)RREG32(mmRCC_CONFIG_MEMSIZE) << 20; + pos = vram_size - DISCOVERY_TMR_OFFSET; + amdgpu_device_vram_access(adev, pos, &bhdr, sizeof(bhdr), false); + + offset = le16_to_cpu(bhdr.table_list[NPS_INFO].offset); + checksum = le16_to_cpu(bhdr.table_list[NPS_INFO].checksum); + + amdgpu_device_vram_access(adev, (pos + offset), nps_data, + sizeof(*nps_data), false); + + nhdr = (struct nps_info_header *)(nps_data); + if (!amdgpu_discovery_verify_checksum((uint8_t *)nps_data, + le32_to_cpu(nhdr->size_bytes), + checksum)) { + dev_err(adev->dev, "nps data refresh, checksum mismatch\n"); + return -EINVAL; + } + + return 0; +} + int amdgpu_discovery_get_nps_info(struct amdgpu_device *adev, uint32_t *nps_type, struct amdgpu_gmc_memrange **ranges, - int *range_cnt) + int *range_cnt, bool refresh) { struct amdgpu_gmc_memrange *mem_ranges; struct binary_header *bhdr; union nps_info *nps_info; + union nps_info nps_data; u16 offset; - int i; + int i, r; if (!nps_type || !range_cnt || !ranges) return -EINVAL; - if (!adev->mman.discovery_bin) { - dev_err(adev->dev, - "fetch mem range failed, ip discovery uninitialized\n"); - return -EINVAL; + if (refresh) { + r = amdgpu_discovery_refresh_nps_info(adev, &nps_data); + if (r) + return r; + nps_info = &nps_data; + } else { + if (!adev->mman.discovery_bin) { + dev_err(adev->dev, + "fetch mem range failed, ip discovery uninitialized\n"); + return -EINVAL; + } + + bhdr = (struct binary_header *)adev->mman.discovery_bin; + offset = le16_to_cpu(bhdr->table_list[NPS_INFO].offset); + + if (!offset) + return -ENOENT; + + /* If verification fails, return as if NPS table doesn't exist */ + if (amdgpu_discovery_verify_npsinfo(adev, bhdr)) + return -ENOENT; + + nps_info = + (union nps_info *)(adev->mman.discovery_bin + offset); } - bhdr = (struct binary_header *)adev->mman.discovery_bin; - offset = le16_to_cpu(bhdr->table_list[NPS_INFO].offset); - - if (!offset) - return -ENOENT; - - /* If verification fails, return as if NPS table doesn't exist */ - if (amdgpu_discovery_verify_npsinfo(adev, bhdr)) - return -ENOENT; - - nps_info = (union nps_info *)(adev->mman.discovery_bin + offset); - switch (le16_to_cpu(nps_info->v1.header.version_major)) { case 1: *nps_type = nps_info->v1.nps_type; @@ -2492,6 +2530,7 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) adev->ip_versions[GC_HWIP][0] = IP_VERSION(9, 2, 2); adev->ip_versions[UVD_HWIP][0] = IP_VERSION(1, 0, 1); adev->ip_versions[DCE_HWIP][0] = IP_VERSION(1, 0, 1); + adev->ip_versions[ISP_HWIP][0] = IP_VERSION(2, 0, 0); } else { adev->ip_versions[MMHUB_HWIP][0] = IP_VERSION(9, 1, 0); adev->ip_versions[ATHUB_HWIP][0] = IP_VERSION(9, 1, 0); @@ -2508,6 +2547,7 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev) adev->ip_versions[GC_HWIP][0] = IP_VERSION(9, 1, 0); adev->ip_versions[UVD_HWIP][0] = IP_VERSION(1, 0, 0); adev->ip_versions[DCE_HWIP][0] = IP_VERSION(1, 0, 0); + adev->ip_versions[ISP_HWIP][0] = IP_VERSION(2, 0, 0); } break; case CHIP_VEGA20: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h index f5d36525ec3e..b44d56465c5b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h @@ -33,6 +33,6 @@ int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev); int amdgpu_discovery_get_nps_info(struct amdgpu_device *adev, uint32_t *nps_type, struct amdgpu_gmc_memrange **ranges, - int *range_cnt); + int *range_cnt, bool refresh); #endif /* __AMDGPU_DISCOVERY__ */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 81d9877c8735..6ac7d335e28e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -23,6 +23,7 @@ */ #include +#include #include #include #include @@ -231,8 +232,6 @@ int amdgpu_wbrf = -1; int amdgpu_damage_clips = -1; /* auto */ int amdgpu_umsch_mm_fwlog; -static void amdgpu_drv_delayed_reset_work_handler(struct work_struct *work); - DECLARE_DYNDBG_CLASSMAP(drm_debug_classes, DD_CLASS_TYPE_DISJOINT_BITS, 0, "DRM_UT_CORE", "DRM_UT_DRIVER", @@ -247,9 +246,6 @@ DECLARE_DYNDBG_CLASSMAP(drm_debug_classes, DD_CLASS_TYPE_DISJOINT_BITS, 0, struct amdgpu_mgpu_info mgpu_info = { .mutex = __MUTEX_INITIALIZER(mgpu_info.mutex), - .delayed_reset_work = __DELAYED_WORK_INITIALIZER( - mgpu_info.delayed_reset_work, - amdgpu_drv_delayed_reset_work_handler, 0), }; int amdgpu_ras_enable = -1; uint amdgpu_ras_mask = 0xffffffff; @@ -2365,11 +2361,15 @@ retry_init: */ if (adev->mode_info.mode_config_initialized && !list_empty(&adev_to_drm(adev)->mode_config.connector_list)) { + const struct drm_format_info *format; + /* select 8 bpp console on low vram cards */ if (adev->gmc.real_vram_size <= (32*1024*1024)) - drm_fbdev_ttm_setup(adev_to_drm(adev), 8); + format = drm_format_info(DRM_FORMAT_C8); else - drm_fbdev_ttm_setup(adev_to_drm(adev), 32); + format = NULL; + + drm_client_setup(adev_to_drm(adev), format); } ret = amdgpu_debugfs_init(adev); @@ -2434,6 +2434,7 @@ amdgpu_pci_remove(struct pci_dev *pdev) struct amdgpu_device *adev = drm_to_adev(dev); amdgpu_xcp_dev_unplug(adev); + amdgpu_gmc_prepare_nps_mode_change(adev); drm_dev_unplug(dev); if (adev->pm.rpm_mode != AMDGPU_RUNPM_NONE) { @@ -2472,82 +2473,6 @@ amdgpu_pci_shutdown(struct pci_dev *pdev) adev->mp1_state = PP_MP1_STATE_NONE; } -/** - * amdgpu_drv_delayed_reset_work_handler - work handler for reset - * - * @work: work_struct. - */ -static void amdgpu_drv_delayed_reset_work_handler(struct work_struct *work) -{ - struct list_head device_list; - struct amdgpu_device *adev; - int i, r; - struct amdgpu_reset_context reset_context; - - memset(&reset_context, 0, sizeof(reset_context)); - - mutex_lock(&mgpu_info.mutex); - if (mgpu_info.pending_reset == true) { - mutex_unlock(&mgpu_info.mutex); - return; - } - mgpu_info.pending_reset = true; - mutex_unlock(&mgpu_info.mutex); - - /* Use a common context, just need to make sure full reset is done */ - reset_context.method = AMD_RESET_METHOD_NONE; - set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); - - for (i = 0; i < mgpu_info.num_dgpu; i++) { - adev = mgpu_info.gpu_ins[i].adev; - reset_context.reset_req_dev = adev; - r = amdgpu_device_pre_asic_reset(adev, &reset_context); - if (r) { - dev_err(adev->dev, "GPU pre asic reset failed with err, %d for drm dev, %s ", - r, adev_to_drm(adev)->unique); - } - if (!queue_work(system_unbound_wq, &adev->xgmi_reset_work)) - r = -EALREADY; - } - for (i = 0; i < mgpu_info.num_dgpu; i++) { - adev = mgpu_info.gpu_ins[i].adev; - flush_work(&adev->xgmi_reset_work); - adev->gmc.xgmi.pending_reset = false; - } - - /* reset function will rebuild the xgmi hive info , clear it now */ - for (i = 0; i < mgpu_info.num_dgpu; i++) - amdgpu_xgmi_remove_device(mgpu_info.gpu_ins[i].adev); - - INIT_LIST_HEAD(&device_list); - - for (i = 0; i < mgpu_info.num_dgpu; i++) - list_add_tail(&mgpu_info.gpu_ins[i].adev->reset_list, &device_list); - - /* unregister the GPU first, reset function will add them back */ - list_for_each_entry(adev, &device_list, reset_list) - amdgpu_unregister_gpu_instance(adev); - - /* Use a common context, just need to make sure full reset is done */ - set_bit(AMDGPU_SKIP_HW_RESET, &reset_context.flags); - set_bit(AMDGPU_SKIP_COREDUMP, &reset_context.flags); - r = amdgpu_do_asic_reset(&device_list, &reset_context); - - if (r) { - DRM_ERROR("reinit gpus failure"); - return; - } - for (i = 0; i < mgpu_info.num_dgpu; i++) { - adev = mgpu_info.gpu_ins[i].adev; - if (!adev->kfd.init_complete) { - kgd2kfd_init_zone_device(adev); - amdgpu_amdkfd_device_init(adev); - amdgpu_amdkfd_drm_client_create(adev); - } - amdgpu_ttm_set_buffer_funcs_status(adev, true); - } -} - static int amdgpu_pmops_prepare(struct device *dev) { struct drm_device *drm_dev = dev_get_drvdata(dev); @@ -2982,6 +2907,7 @@ static const struct drm_driver amdgpu_kms_driver = { .num_ioctls = ARRAY_SIZE(amdgpu_ioctls_kms), .dumb_create = amdgpu_mode_dumb_create, .dumb_map_offset = amdgpu_mode_dumb_mmap, + DRM_FBDEV_TTM_DRIVER_OPS, .fops = &amdgpu_driver_kms_fops, .release = &amdgpu_driver_release_kms, #ifdef CONFIG_PROC_FS @@ -3008,6 +2934,7 @@ const struct drm_driver amdgpu_partition_driver = { .num_ioctls = ARRAY_SIZE(amdgpu_ioctls_kms), .dumb_create = amdgpu_mode_dumb_create, .dumb_map_offset = amdgpu_mode_dumb_mmap, + DRM_FBDEV_TTM_DRIVER_OPS, .fops = &amdgpu_driver_kms_fops, .release = &amdgpu_driver_release_kms, @@ -3068,6 +2995,12 @@ static int __init amdgpu_init(void) /* Ignore KFD init failures. Normal when CONFIG_HSA_AMD is not set. */ amdgpu_amdkfd_init(); + if (amdgpu_pp_feature_mask & PP_OVERDRIVE_MASK) { + add_taint(TAINT_CPU_OUT_OF_SPEC, LOCKDEP_STILL_OK); + pr_crit("Overdrive is enabled, please disable it before " + "reporting any bugs unrelated to overdrive.\n"); + } + /* let modprobe override vga console setting */ return pci_register_driver(&amdgpu_kms_pci_driver); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c index c7df7fa3459f..00a4ab082459 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c @@ -59,18 +59,21 @@ void amdgpu_show_fdinfo(struct drm_printer *p, struct drm_file *file) struct amdgpu_fpriv *fpriv = file->driver_priv; struct amdgpu_vm *vm = &fpriv->vm; - struct amdgpu_mem_stats stats; + struct amdgpu_mem_stats stats[__AMDGPU_PL_LAST + 1] = { }; ktime_t usage[AMDGPU_HW_IP_NUM]; - unsigned int hw_ip; + const char *pl_name[] = { + [TTM_PL_VRAM] = "vram", + [TTM_PL_TT] = "gtt", + [TTM_PL_SYSTEM] = "cpu", + }; + unsigned int hw_ip, i; int ret; - memset(&stats, 0, sizeof(stats)); - ret = amdgpu_bo_reserve(vm->root.bo, false); if (ret) return; - amdgpu_vm_get_memory(vm, &stats); + amdgpu_vm_get_memory(vm, stats, ARRAY_SIZE(stats)); amdgpu_bo_unreserve(vm->root.bo); amdgpu_ctx_mgr_usage(&fpriv->ctx_mgr, usage); @@ -82,24 +85,35 @@ void amdgpu_show_fdinfo(struct drm_printer *p, struct drm_file *file) */ drm_printf(p, "pasid:\t%u\n", fpriv->vm.pasid); - drm_printf(p, "drm-memory-vram:\t%llu KiB\n", stats.vram/1024UL); - drm_printf(p, "drm-memory-gtt: \t%llu KiB\n", stats.gtt/1024UL); - drm_printf(p, "drm-memory-cpu: \t%llu KiB\n", stats.cpu/1024UL); + + for (i = 0; i < TTM_PL_PRIV; i++) + drm_print_memory_stats(p, + &stats[i].drm, + DRM_GEM_OBJECT_RESIDENT | + DRM_GEM_OBJECT_PURGEABLE, + pl_name[i]); + + /* Legacy amdgpu keys, alias to drm-resident-memory-: */ + drm_printf(p, "drm-memory-vram:\t%llu KiB\n", + stats[TTM_PL_VRAM].total/1024UL); + drm_printf(p, "drm-memory-gtt: \t%llu KiB\n", + stats[TTM_PL_TT].total/1024UL); + drm_printf(p, "drm-memory-cpu: \t%llu KiB\n", + stats[TTM_PL_SYSTEM].total/1024UL); + + /* Amdgpu specific memory accounting keys: */ drm_printf(p, "amd-memory-visible-vram:\t%llu KiB\n", - stats.visible_vram/1024UL); + stats[TTM_PL_VRAM].visible/1024UL); drm_printf(p, "amd-evicted-vram:\t%llu KiB\n", - stats.evicted_vram/1024UL); + stats[TTM_PL_VRAM].evicted/1024UL); drm_printf(p, "amd-evicted-visible-vram:\t%llu KiB\n", - stats.evicted_visible_vram/1024UL); + stats[TTM_PL_VRAM].evicted_visible/1024UL); drm_printf(p, "amd-requested-vram:\t%llu KiB\n", - stats.requested_vram/1024UL); + stats[TTM_PL_VRAM].requested/1024UL); drm_printf(p, "amd-requested-visible-vram:\t%llu KiB\n", - stats.requested_visible_vram/1024UL); + stats[TTM_PL_VRAM].requested_visible/1024UL); drm_printf(p, "amd-requested-gtt:\t%llu KiB\n", - stats.requested_gtt/1024UL); - drm_printf(p, "drm-shared-vram:\t%llu KiB\n", stats.vram_shared/1024UL); - drm_printf(p, "drm-shared-gtt:\t%llu KiB\n", stats.gtt_shared/1024UL); - drm_printf(p, "drm-shared-cpu:\t%llu KiB\n", stats.cpu_shared/1024UL); + stats[TTM_PL_TT].requested/1024UL); for (hw_ip = 0; hw_ip < AMDGPU_HW_IP_NUM; ++hw_ip) { if (!usage[hw_ip]) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 83e54697f0ee..e96984c53e72 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -87,16 +87,6 @@ int amdgpu_gfx_me_queue_to_bit(struct amdgpu_device *adev, return bit; } -void amdgpu_gfx_bit_to_me_queue(struct amdgpu_device *adev, int bit, - int *me, int *pipe, int *queue) -{ - *queue = bit % adev->gfx.me.num_queue_per_pipe; - *pipe = (bit / adev->gfx.me.num_queue_per_pipe) - % adev->gfx.me.num_pipe_per_me; - *me = (bit / adev->gfx.me.num_queue_per_pipe) - / adev->gfx.me.num_pipe_per_me; -} - bool amdgpu_gfx_is_me_queue_enabled(struct amdgpu_device *adev, int me, int pipe, int queue) { @@ -415,7 +405,7 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, } /* prepare MQD backup */ - kiq->mqd_backup = kmalloc(mqd_size, GFP_KERNEL); + kiq->mqd_backup = kzalloc(mqd_size, GFP_KERNEL); if (!kiq->mqd_backup) { dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); @@ -438,7 +428,7 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, ring->mqd_size = mqd_size; /* prepare MQD backup */ - adev->gfx.me.mqd_backup[i] = kmalloc(mqd_size, GFP_KERNEL); + adev->gfx.me.mqd_backup[i] = kzalloc(mqd_size, GFP_KERNEL); if (!adev->gfx.me.mqd_backup[i]) { dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); return -ENOMEM; @@ -462,7 +452,7 @@ int amdgpu_gfx_mqd_sw_init(struct amdgpu_device *adev, ring->mqd_size = mqd_size; /* prepare MQD backup */ - adev->gfx.mec.mqd_backup[j] = kmalloc(mqd_size, GFP_KERNEL); + adev->gfx.mec.mqd_backup[j] = kzalloc(mqd_size, GFP_KERNEL); if (!adev->gfx.mec.mqd_backup[j]) { dev_warn(adev->dev, "no memory to create MQD backup for ring %s\n", ring->name); return -ENOMEM; @@ -1363,35 +1353,35 @@ static ssize_t amdgpu_gfx_set_compute_partition(struct device *dev, return count; } +static const char *xcp_desc[] = { + [AMDGPU_SPX_PARTITION_MODE] = "SPX", + [AMDGPU_DPX_PARTITION_MODE] = "DPX", + [AMDGPU_TPX_PARTITION_MODE] = "TPX", + [AMDGPU_QPX_PARTITION_MODE] = "QPX", + [AMDGPU_CPX_PARTITION_MODE] = "CPX", +}; + static ssize_t amdgpu_gfx_get_available_compute_partition(struct device *dev, struct device_attribute *addr, char *buf) { struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); - char *supported_partition; + struct amdgpu_xcp_mgr *xcp_mgr = adev->xcp_mgr; + int size = 0, mode; + char *sep = ""; - /* TBD */ - switch (NUM_XCC(adev->gfx.xcc_mask)) { - case 8: - supported_partition = "SPX, DPX, QPX, CPX"; - break; - case 6: - supported_partition = "SPX, TPX, CPX"; - break; - case 4: - supported_partition = "SPX, DPX, CPX"; - break; - /* this seems only existing in emulation phase */ - case 2: - supported_partition = "SPX, CPX"; - break; - default: - supported_partition = "Not supported"; - break; + if (!xcp_mgr || !xcp_mgr->avail_xcp_modes) + return sysfs_emit(buf, "Not supported\n"); + + for_each_inst(mode, xcp_mgr->avail_xcp_modes) { + size += sysfs_emit_at(buf, size, "%s%s", sep, xcp_desc[mode]); + sep = ", "; } - return sysfs_emit(buf, "%s\n", supported_partition); + size += sysfs_emit_at(buf, size, "\n"); + + return size; } static int amdgpu_gfx_run_cleaner_shader_job(struct amdgpu_ring *ring) @@ -1614,32 +1604,55 @@ static DEVICE_ATTR(available_compute_partition, 0444, int amdgpu_gfx_sysfs_init(struct amdgpu_device *adev) { + struct amdgpu_xcp_mgr *xcp_mgr = adev->xcp_mgr; + bool xcp_switch_supported; int r; + if (!xcp_mgr) + return 0; + + xcp_switch_supported = + (xcp_mgr->funcs && xcp_mgr->funcs->switch_partition_mode); + + if (!xcp_switch_supported) + dev_attr_current_compute_partition.attr.mode &= + ~(S_IWUSR | S_IWGRP | S_IWOTH); + r = device_create_file(adev->dev, &dev_attr_current_compute_partition); if (r) return r; - r = device_create_file(adev->dev, &dev_attr_available_compute_partition); + if (xcp_switch_supported) + r = device_create_file(adev->dev, + &dev_attr_available_compute_partition); return r; } void amdgpu_gfx_sysfs_fini(struct amdgpu_device *adev) { + struct amdgpu_xcp_mgr *xcp_mgr = adev->xcp_mgr; + bool xcp_switch_supported; + + if (!xcp_mgr) + return; + + xcp_switch_supported = + (xcp_mgr->funcs && xcp_mgr->funcs->switch_partition_mode); device_remove_file(adev->dev, &dev_attr_current_compute_partition); - device_remove_file(adev->dev, &dev_attr_available_compute_partition); + + if (xcp_switch_supported) + device_remove_file(adev->dev, + &dev_attr_available_compute_partition); } int amdgpu_gfx_sysfs_isolation_shader_init(struct amdgpu_device *adev) { int r; - if (!amdgpu_sriov_vf(adev)) { - r = device_create_file(adev->dev, &dev_attr_enforce_isolation); - if (r) - return r; - } + r = device_create_file(adev->dev, &dev_attr_enforce_isolation); + if (r) + return r; r = device_create_file(adev->dev, &dev_attr_run_cleaner_shader); if (r) @@ -1650,8 +1663,7 @@ int amdgpu_gfx_sysfs_isolation_shader_init(struct amdgpu_device *adev) void amdgpu_gfx_sysfs_isolation_shader_fini(struct amdgpu_device *adev) { - if (!amdgpu_sriov_vf(adev)) - device_remove_file(adev->dev, &dev_attr_enforce_isolation); + device_remove_file(adev->dev, &dev_attr_enforce_isolation); device_remove_file(adev->dev, &dev_attr_run_cleaner_shader); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h index 5644e10a86a9..f710178a21bc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h @@ -540,8 +540,6 @@ bool amdgpu_gfx_is_high_priority_graphics_queue(struct amdgpu_device *adev, struct amdgpu_ring *ring); int amdgpu_gfx_me_queue_to_bit(struct amdgpu_device *adev, int me, int pipe, int queue); -void amdgpu_gfx_bit_to_me_queue(struct amdgpu_device *adev, int bit, - int *me, int *pipe, int *queue); bool amdgpu_gfx_is_me_queue_enabled(struct amdgpu_device *adev, int me, int pipe, int queue); void amdgpu_gfx_off_ctrl(struct amdgpu_device *adev, bool enable); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 17a19d49d30a..1c19a65e6553 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -1065,18 +1065,6 @@ uint64_t amdgpu_gmc_vram_pa(struct amdgpu_device *adev, struct amdgpu_bo *bo) return amdgpu_gmc_vram_mc2pa(adev, amdgpu_bo_gpu_offset(bo)); } -/** - * amdgpu_gmc_vram_cpu_pa - calculate vram buffer object's physical address - * from CPU's view - * - * @adev: amdgpu_device pointer - * @bo: amdgpu buffer object - */ -uint64_t amdgpu_gmc_vram_cpu_pa(struct amdgpu_device *adev, struct amdgpu_bo *bo) -{ - return amdgpu_bo_gpu_offset(bo) - adev->gmc.vram_start + adev->gmc.aper_base; -} - int amdgpu_gmc_vram_checking(struct amdgpu_device *adev) { struct amdgpu_bo *vram_bo = NULL; @@ -1130,6 +1118,79 @@ release_buffer: return ret; } +static const char *nps_desc[] = { + [AMDGPU_NPS1_PARTITION_MODE] = "NPS1", + [AMDGPU_NPS2_PARTITION_MODE] = "NPS2", + [AMDGPU_NPS3_PARTITION_MODE] = "NPS3", + [AMDGPU_NPS4_PARTITION_MODE] = "NPS4", + [AMDGPU_NPS6_PARTITION_MODE] = "NPS6", + [AMDGPU_NPS8_PARTITION_MODE] = "NPS8", +}; + +static ssize_t available_memory_partition_show(struct device *dev, + struct device_attribute *addr, + char *buf) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + int size = 0, mode; + char *sep = ""; + + for_each_inst(mode, adev->gmc.supported_nps_modes) { + size += sysfs_emit_at(buf, size, "%s%s", sep, nps_desc[mode]); + sep = ", "; + } + size += sysfs_emit_at(buf, size, "\n"); + + return size; +} + +static ssize_t current_memory_partition_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct drm_device *ddev = dev_get_drvdata(dev); + struct amdgpu_device *adev = drm_to_adev(ddev); + enum amdgpu_memory_partition mode; + struct amdgpu_hive_info *hive; + int i; + + mode = UNKNOWN_MEMORY_PARTITION_MODE; + for_each_inst(i, adev->gmc.supported_nps_modes) { + if (!strncasecmp(nps_desc[i], buf, strlen(nps_desc[i]))) { + mode = i; + break; + } + } + + if (mode == UNKNOWN_MEMORY_PARTITION_MODE) + return -EINVAL; + + if (mode == adev->gmc.gmc_funcs->query_mem_partition_mode(adev)) { + dev_info( + adev->dev, + "requested NPS mode is same as current NPS mode, skipping\n"); + return count; + } + + /* If device is part of hive, all devices in the hive should request the + * same mode. Hence store the requested mode in hive. + */ + hive = amdgpu_get_xgmi_hive(adev); + if (hive) { + atomic_set(&hive->requested_nps_mode, mode); + amdgpu_put_xgmi_hive(hive); + } else { + adev->gmc.requested_nps_mode = mode; + } + + dev_info( + adev->dev, + "NPS mode change requested, please remove and reload the driver\n"); + + return count; +} + static ssize_t current_memory_partition_show( struct device *dev, struct device_attribute *addr, char *buf) { @@ -1138,53 +1199,65 @@ static ssize_t current_memory_partition_show( enum amdgpu_memory_partition mode; mode = adev->gmc.gmc_funcs->query_mem_partition_mode(adev); - switch (mode) { - case AMDGPU_NPS1_PARTITION_MODE: - return sysfs_emit(buf, "NPS1\n"); - case AMDGPU_NPS2_PARTITION_MODE: - return sysfs_emit(buf, "NPS2\n"); - case AMDGPU_NPS3_PARTITION_MODE: - return sysfs_emit(buf, "NPS3\n"); - case AMDGPU_NPS4_PARTITION_MODE: - return sysfs_emit(buf, "NPS4\n"); - case AMDGPU_NPS6_PARTITION_MODE: - return sysfs_emit(buf, "NPS6\n"); - case AMDGPU_NPS8_PARTITION_MODE: - return sysfs_emit(buf, "NPS8\n"); - default: + if ((mode >= ARRAY_SIZE(nps_desc)) || + (BIT(mode) & AMDGPU_ALL_NPS_MASK) != BIT(mode)) return sysfs_emit(buf, "UNKNOWN\n"); - } + + return sysfs_emit(buf, "%s\n", nps_desc[mode]); } -static DEVICE_ATTR_RO(current_memory_partition); +static DEVICE_ATTR_RW(current_memory_partition); +static DEVICE_ATTR_RO(available_memory_partition); int amdgpu_gmc_sysfs_init(struct amdgpu_device *adev) { + bool nps_switch_support; + int r = 0; + if (!adev->gmc.gmc_funcs->query_mem_partition_mode) return 0; + nps_switch_support = (hweight32(adev->gmc.supported_nps_modes & + AMDGPU_ALL_NPS_MASK) > 1); + if (!nps_switch_support) + dev_attr_current_memory_partition.attr.mode &= + ~(S_IWUSR | S_IWGRP | S_IWOTH); + else + r = device_create_file(adev->dev, + &dev_attr_available_memory_partition); + + if (r) + return r; + return device_create_file(adev->dev, &dev_attr_current_memory_partition); } void amdgpu_gmc_sysfs_fini(struct amdgpu_device *adev) { + if (!adev->gmc.gmc_funcs->query_mem_partition_mode) + return; + device_remove_file(adev->dev, &dev_attr_current_memory_partition); + device_remove_file(adev->dev, &dev_attr_available_memory_partition); } int amdgpu_gmc_get_nps_memranges(struct amdgpu_device *adev, struct amdgpu_mem_partition_info *mem_ranges, - int exp_ranges) + uint8_t *exp_ranges) { struct amdgpu_gmc_memrange *ranges; int range_cnt, ret, i, j; uint32_t nps_type; + bool refresh; - if (!mem_ranges) + if (!mem_ranges || !exp_ranges) return -EINVAL; + refresh = (adev->init_lvl->level != AMDGPU_INIT_LEVEL_MINIMAL_XGMI) && + (adev->gmc.reset_flags & AMDGPU_GMC_INIT_RESET_NPS); ret = amdgpu_discovery_get_nps_info(adev, &nps_type, &ranges, - &range_cnt); + &range_cnt, refresh); if (ret) return ret; @@ -1192,16 +1265,16 @@ int amdgpu_gmc_get_nps_memranges(struct amdgpu_device *adev, /* TODO: For now, expect ranges and partition count to be the same. * Adjust if there are holes expected in any NPS domain. */ - if (range_cnt != exp_ranges) { + if (*exp_ranges && (range_cnt != *exp_ranges)) { dev_warn( adev->dev, "NPS config mismatch - expected ranges: %d discovery - nps mode: %d, nps ranges: %d", - exp_ranges, nps_type, range_cnt); + *exp_ranges, nps_type, range_cnt); ret = -EINVAL; goto err; } - for (i = 0; i < exp_ranges; ++i) { + for (i = 0; i < range_cnt; ++i) { if (ranges[i].base_address >= ranges[i].limit_address) { dev_warn( adev->dev, @@ -1242,8 +1315,81 @@ int amdgpu_gmc_get_nps_memranges(struct amdgpu_device *adev, ranges[i].limit_address - ranges[i].base_address + 1; } + if (!*exp_ranges) + *exp_ranges = range_cnt; err: kfree(ranges); return ret; } + +int amdgpu_gmc_request_memory_partition(struct amdgpu_device *adev, + int nps_mode) +{ + /* Not supported on VF devices and APUs */ + if (amdgpu_sriov_vf(adev) || (adev->flags & AMD_IS_APU)) + return -EOPNOTSUPP; + + if (!adev->psp.funcs) { + dev_err(adev->dev, + "PSP interface not available for nps mode change request"); + return -EINVAL; + } + + return psp_memory_partition(&adev->psp, nps_mode); +} + +static inline bool amdgpu_gmc_need_nps_switch_req(struct amdgpu_device *adev, + int req_nps_mode, + int cur_nps_mode) +{ + return (((BIT(req_nps_mode) & adev->gmc.supported_nps_modes) == + BIT(req_nps_mode)) && + req_nps_mode != cur_nps_mode); +} + +void amdgpu_gmc_prepare_nps_mode_change(struct amdgpu_device *adev) +{ + int req_nps_mode, cur_nps_mode, r; + struct amdgpu_hive_info *hive; + + if (amdgpu_sriov_vf(adev) || !adev->gmc.supported_nps_modes || + !adev->gmc.gmc_funcs->request_mem_partition_mode) + return; + + cur_nps_mode = adev->gmc.gmc_funcs->query_mem_partition_mode(adev); + hive = amdgpu_get_xgmi_hive(adev); + if (hive) { + req_nps_mode = atomic_read(&hive->requested_nps_mode); + if (!amdgpu_gmc_need_nps_switch_req(adev, req_nps_mode, + cur_nps_mode)) { + amdgpu_put_xgmi_hive(hive); + return; + } + r = amdgpu_xgmi_request_nps_change(adev, hive, req_nps_mode); + amdgpu_put_xgmi_hive(hive); + goto out; + } + + req_nps_mode = adev->gmc.requested_nps_mode; + if (!amdgpu_gmc_need_nps_switch_req(adev, req_nps_mode, cur_nps_mode)) + return; + + /* even if this fails, we should let driver unload w/o blocking */ + r = adev->gmc.gmc_funcs->request_mem_partition_mode(adev, req_nps_mode); +out: + if (r) + dev_err(adev->dev, "NPS mode change request failed\n"); + else + dev_info( + adev->dev, + "NPS mode change request done, reload driver to complete the change\n"); +} + +bool amdgpu_gmc_need_reset_on_init(struct amdgpu_device *adev) +{ + if (adev->gmc.gmc_funcs->need_reset_on_init) + return adev->gmc.gmc_funcs->need_reset_on_init(adev); + + return false; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 4d951a1baefa..459a30fe239f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -73,6 +73,13 @@ enum amdgpu_memory_partition { AMDGPU_NPS8_PARTITION_MODE = 8, }; +#define AMDGPU_ALL_NPS_MASK \ + (BIT(AMDGPU_NPS1_PARTITION_MODE) | BIT(AMDGPU_NPS2_PARTITION_MODE) | \ + BIT(AMDGPU_NPS3_PARTITION_MODE) | BIT(AMDGPU_NPS4_PARTITION_MODE) | \ + BIT(AMDGPU_NPS6_PARTITION_MODE) | BIT(AMDGPU_NPS8_PARTITION_MODE)) + +#define AMDGPU_GMC_INIT_RESET_NPS BIT(0) + /* * GMC page fault information */ @@ -161,6 +168,10 @@ struct amdgpu_gmc_funcs { enum amdgpu_memory_partition (*query_mem_partition_mode)( struct amdgpu_device *adev); + /* Request NPS mode */ + int (*request_mem_partition_mode)(struct amdgpu_device *adev, + int nps_mode); + bool (*need_reset_on_init)(struct amdgpu_device *adev); }; struct amdgpu_xgmi_ras { @@ -182,7 +193,6 @@ struct amdgpu_xgmi { bool supported; struct ras_common_if *ras_if; bool connected_to_cpu; - bool pending_reset; struct amdgpu_xgmi_ras *ras; }; @@ -305,6 +315,9 @@ struct amdgpu_gmc { struct amdgpu_mem_partition_info *mem_partitions; uint8_t num_mem_partitions; const struct amdgpu_gmc_funcs *gmc_funcs; + enum amdgpu_memory_partition requested_nps_mode; + uint32_t supported_nps_modes; + uint32_t reset_flags; struct amdgpu_xgmi xgmi; struct amdgpu_irq_src ecc_irq; @@ -447,13 +460,17 @@ void amdgpu_gmc_get_vbios_allocations(struct amdgpu_device *adev); void amdgpu_gmc_init_pdb0(struct amdgpu_device *adev); uint64_t amdgpu_gmc_vram_mc2pa(struct amdgpu_device *adev, uint64_t mc_addr); uint64_t amdgpu_gmc_vram_pa(struct amdgpu_device *adev, struct amdgpu_bo *bo); -uint64_t amdgpu_gmc_vram_cpu_pa(struct amdgpu_device *adev, struct amdgpu_bo *bo); int amdgpu_gmc_vram_checking(struct amdgpu_device *adev); int amdgpu_gmc_sysfs_init(struct amdgpu_device *adev); void amdgpu_gmc_sysfs_fini(struct amdgpu_device *adev); int amdgpu_gmc_get_nps_memranges(struct amdgpu_device *adev, struct amdgpu_mem_partition_info *mem_ranges, - int exp_ranges); + uint8_t *exp_ranges); + +int amdgpu_gmc_request_memory_partition(struct amdgpu_device *adev, + int nps_mode); +void amdgpu_gmc_prepare_nps_mode_change(struct amdgpu_device *adev); +bool amdgpu_gmc_need_reset_on_init(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c index 00d6211e0fbf..f0765ccde668 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.c @@ -225,15 +225,6 @@ void amdgpu_i2c_destroy(struct amdgpu_i2c_chan *i2c) kfree(i2c); } -/* Add the default buses */ -void amdgpu_i2c_init(struct amdgpu_device *adev) -{ - if (amdgpu_hw_i2c) - DRM_INFO("hw_i2c forced on, you may experience display detection problems!\n"); - - amdgpu_atombios_i2c_init(adev); -} - /* remove all the buses */ void amdgpu_i2c_fini(struct amdgpu_device *adev) { @@ -247,22 +238,6 @@ void amdgpu_i2c_fini(struct amdgpu_device *adev) } } -/* Add additional buses */ -void amdgpu_i2c_add(struct amdgpu_device *adev, - const struct amdgpu_i2c_bus_rec *rec, - const char *name) -{ - struct drm_device *dev = adev_to_drm(adev); - int i; - - for (i = 0; i < AMDGPU_MAX_I2C_BUS; i++) { - if (!adev->i2c_bus[i]) { - adev->i2c_bus[i] = amdgpu_i2c_create(dev, rec, name); - return; - } - } -} - /* looks up bus based on id */ struct amdgpu_i2c_chan * amdgpu_i2c_lookup(struct amdgpu_device *adev, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.h index 63c2ff7499e1..21e3d1dad0a1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_i2c.h @@ -28,11 +28,7 @@ struct amdgpu_i2c_chan *amdgpu_i2c_create(struct drm_device *dev, const struct amdgpu_i2c_bus_rec *rec, const char *name); void amdgpu_i2c_destroy(struct amdgpu_i2c_chan *i2c); -void amdgpu_i2c_init(struct amdgpu_device *adev); void amdgpu_i2c_fini(struct amdgpu_device *adev); -void amdgpu_i2c_add(struct amdgpu_device *adev, - const struct amdgpu_i2c_bus_rec *rec, - const char *name); struct amdgpu_i2c_chan * amdgpu_i2c_lookup(struct amdgpu_device *adev, const struct amdgpu_i2c_bus_rec *i2c_bus); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c index 92d27d32de41..8e712a11aba5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ids.c @@ -342,15 +342,13 @@ static int amdgpu_vmid_grab_reserved(struct amdgpu_vm *vm, * @ring: ring we want to submit job to * @job: job who wants to use the VMID * @id: resulting VMID - * @fence: fence to wait for if no id could be grabbed * * Try to reuse a VMID for this submission. */ static int amdgpu_vmid_grab_used(struct amdgpu_vm *vm, struct amdgpu_ring *ring, struct amdgpu_job *job, - struct amdgpu_vmid **id, - struct dma_fence **fence) + struct amdgpu_vmid **id) { struct amdgpu_device *adev = ring->adev; unsigned vmhub = ring->vm_hub; @@ -429,7 +427,7 @@ int amdgpu_vmid_grab(struct amdgpu_vm *vm, struct amdgpu_ring *ring, if (r || !id) goto error; } else { - r = amdgpu_vmid_grab_used(vm, ring, job, &id, fence); + r = amdgpu_vmid_grab_used(vm, ring, job, &id); if (r) goto error; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.c index 4766e99dd98f..263ce1811cc8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_isp.c @@ -33,33 +33,17 @@ #include "isp_v4_1_0.h" #include "isp_v4_1_1.h" -static int isp_sw_init(void *handle) -{ - return 0; -} - -static int isp_sw_fini(void *handle) -{ - return 0; -} - /** * isp_hw_init - start and test isp block * - * @handle: handle for amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -static int isp_hw_init(void *handle) +static int isp_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_isp *isp = &adev->isp; - const struct amdgpu_ip_block *ip_block = - amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_ISP); - - if (!ip_block) - return -EINVAL; - if (isp->funcs->hw_init != NULL) return isp->funcs->hw_init(isp); @@ -69,13 +53,12 @@ static int isp_hw_init(void *handle) /** * isp_hw_fini - stop the hardware block * - * @handle: handle for amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -static int isp_hw_fini(void *handle) +static int isp_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - struct amdgpu_isp *isp = &adev->isp; + struct amdgpu_isp *isp = &ip_block->adev->isp; if (isp->funcs->hw_fini != NULL) return isp->funcs->hw_fini(isp); @@ -83,16 +66,6 @@ static int isp_hw_fini(void *handle) return -ENODEV; } -static int isp_suspend(void *handle) -{ - return 0; -} - -static int isp_resume(void *handle) -{ - return 0; -} - static int isp_load_fw_by_psp(struct amdgpu_device *adev) { const struct common_firmware_header *hdr; @@ -122,9 +95,10 @@ static int isp_load_fw_by_psp(struct amdgpu_device *adev) return r; } -static int isp_early_init(void *handle) +static int isp_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_isp *isp = &adev->isp; switch (amdgpu_ip_version(adev, ISP_HWIP, 0)) { @@ -154,16 +128,6 @@ static bool isp_is_idle(void *handle) return true; } -static int isp_wait_for_idle(void *handle) -{ - return 0; -} - -static int isp_soft_reset(void *handle) -{ - return 0; -} - static int isp_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -179,16 +143,9 @@ static int isp_set_powergating_state(void *handle, static const struct amd_ip_funcs isp_ip_funcs = { .name = "isp_ip", .early_init = isp_early_init, - .late_init = NULL, - .sw_init = isp_sw_init, - .sw_fini = isp_sw_fini, .hw_init = isp_hw_init, .hw_fini = isp_hw_fini, - .suspend = isp_suspend, - .resume = isp_resume, .is_idle = isp_is_idle, - .wait_for_idle = isp_wait_for_idle, - .soft_reset = isp_soft_reset, .set_clockgating_state = isp_set_clockgating_state, .set_powergating_state = isp_set_powergating_state, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index 16f2605ac50b..ea2663169bf3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -42,7 +42,7 @@ static void amdgpu_job_do_core_dump(struct amdgpu_device *adev, for (i = 0; i < adev->num_ip_blocks; i++) if (adev->ip_blocks[i].version->funcs->dump_ip_state) adev->ip_blocks[i].version->funcs - ->dump_ip_state((void *)adev); + ->dump_ip_state((void *)&adev->ip_blocks[i]); dev_info(adev->dev, "Dumping IP State Completed\n"); amdgpu_coredump(adev, true, false, job); @@ -149,7 +149,7 @@ static enum drm_gpu_sched_stat amdgpu_job_timedout(struct drm_sched_job *s_job) atomic_inc(&ring->adev->gpu_reset_counter); amdgpu_fence_driver_force_completion(ring); if (amdgpu_ring_sched_ready(ring)) - drm_sched_start(&ring->sched); + drm_sched_start(&ring->sched, 0); goto exit; } } @@ -356,10 +356,10 @@ amdgpu_job_prepare_job(struct drm_sched_job *sched_job, if (r) goto error; - if (!fence && job->gang_submit) + if (job->gang_submit) fence = amdgpu_device_switch_gang(ring->adev, job->gang_submit); - while (!fence && job->vm && !job->vmid) { + if (!fence && job->vm && !job->vmid) { r = amdgpu_vmid_grab(job->vm, ring, job, &fence); if (r) { dev_err(ring->adev->dev, "Error getting VM ID (%d)\n", r); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c index 6df99cb00d9a..95e2796919fc 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.c @@ -342,3 +342,76 @@ int amdgpu_jpeg_psp_update_sram(struct amdgpu_device *adev, int inst_idx, return psp_execute_ip_fw_load(&adev->psp, &ucode); } + +/* + * debugfs for to enable/disable jpeg job submission to specific core. + */ +#if defined(CONFIG_DEBUG_FS) +static int amdgpu_debugfs_jpeg_sched_mask_set(void *data, u64 val) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)data; + u32 i, j; + u64 mask = 0; + struct amdgpu_ring *ring; + + if (!adev) + return -ENODEV; + + mask = (1 << (adev->jpeg.num_jpeg_inst * adev->jpeg.num_jpeg_rings)) - 1; + if ((val & mask) == 0) + return -EINVAL; + + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + ring = &adev->jpeg.inst[i].ring_dec[j]; + if (val & (1 << ((i * adev->jpeg.num_jpeg_rings) + j))) + ring->sched.ready = true; + else + ring->sched.ready = false; + } + } + /* publish sched.ready flag update effective immediately across smp */ + smp_rmb(); + return 0; +} + +static int amdgpu_debugfs_jpeg_sched_mask_get(void *data, u64 *val) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)data; + u32 i, j; + u64 mask = 0; + struct amdgpu_ring *ring; + + if (!adev) + return -ENODEV; + for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + ring = &adev->jpeg.inst[i].ring_dec[j]; + if (ring->sched.ready) + mask |= 1 << ((i * adev->jpeg.num_jpeg_rings) + j); + } + } + *val = mask; + return 0; +} + +DEFINE_DEBUGFS_ATTRIBUTE(amdgpu_debugfs_jpeg_sched_mask_fops, + amdgpu_debugfs_jpeg_sched_mask_get, + amdgpu_debugfs_jpeg_sched_mask_set, "%llx\n"); + +#endif + +void amdgpu_debugfs_jpeg_sched_mask_init(struct amdgpu_device *adev) +{ +#if defined(CONFIG_DEBUG_FS) + struct drm_minor *minor = adev_to_drm(adev)->primary; + struct dentry *root = minor->debugfs_root; + char name[32]; + + if (!(adev->jpeg.num_jpeg_inst > 1) && !(adev->jpeg.num_jpeg_rings > 1)) + return; + sprintf(name, "amdgpu_jpeg_sched_mask"); + debugfs_create_file(name, 0600, root, adev, + &amdgpu_debugfs_jpeg_sched_mask_fops); +#endif +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h index f9cdd873ac9b..819dc7a0af99 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_jpeg.h @@ -149,5 +149,6 @@ int amdgpu_jpeg_ras_late_init(struct amdgpu_device *adev, int amdgpu_jpeg_ras_sw_init(struct amdgpu_device *adev); int amdgpu_jpeg_psp_update_sram(struct amdgpu_device *adev, int inst_idx, enum AMDGPU_UCODE_ID ucode_id); +void amdgpu_debugfs_jpeg_sched_mask_init(struct amdgpu_device *adev); #endif /*__AMDGPU_JPEG_H__*/ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c index 10b61ff63802..6909af56fcad 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c @@ -905,7 +905,7 @@ int amdgpu_mes_reset_legacy_queue(struct amdgpu_device *adev, queue_input.me_id = ring->me; queue_input.pipe_id = ring->pipe; queue_input.queue_id = ring->queue; - queue_input.mqd_addr = amdgpu_bo_gpu_offset(ring->mqd_obj); + queue_input.mqd_addr = ring->mqd_obj ? amdgpu_bo_gpu_offset(ring->mqd_obj) : 0; queue_input.wptr_addr = ring->wptr_gpu_addr; queue_input.vmid = vmid; queue_input.use_mmio = use_mmio; @@ -1203,8 +1203,10 @@ int amdgpu_mes_add_ring(struct amdgpu_device *adev, int gang_id, r = amdgpu_ring_init(adev, ring, 1024, NULL, 0, AMDGPU_RING_PRIO_DEFAULT, NULL); - if (r) + if (r) { + amdgpu_mes_unlock(&adev->mes); goto clean_up_memory; + } amdgpu_mes_ring_to_queue_props(adev, ring, &qprops); @@ -1237,7 +1239,6 @@ clean_up_ring: amdgpu_ring_fini(ring); clean_up_memory: kfree(ring); - amdgpu_mes_unlock(&adev->mes); return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h index f61d117b0caf..79c2f807b9fe 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_nbio.h @@ -101,6 +101,7 @@ struct amdgpu_nbio_funcs { int (*get_compute_partition_mode)(struct amdgpu_device *adev); u32 (*get_memory_partition_mode)(struct amdgpu_device *adev, u32 *supp_modes); + bool (*is_nps_switch_requested)(struct amdgpu_device *adev); u64 (*get_pcie_replay_count)(struct amdgpu_device *adev); void (*set_reg_remap)(struct amdgpu_device *adev); }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index 44819cdba7fb..1e6a044e3143 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -1170,54 +1170,92 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, } void amdgpu_bo_get_memory(struct amdgpu_bo *bo, - struct amdgpu_mem_stats *stats) + struct amdgpu_mem_stats *stats, + unsigned int sz) { + const unsigned int domain_to_pl[] = { + [ilog2(AMDGPU_GEM_DOMAIN_CPU)] = TTM_PL_SYSTEM, + [ilog2(AMDGPU_GEM_DOMAIN_GTT)] = TTM_PL_TT, + [ilog2(AMDGPU_GEM_DOMAIN_VRAM)] = TTM_PL_VRAM, + [ilog2(AMDGPU_GEM_DOMAIN_GDS)] = AMDGPU_PL_GDS, + [ilog2(AMDGPU_GEM_DOMAIN_GWS)] = AMDGPU_PL_GWS, + [ilog2(AMDGPU_GEM_DOMAIN_OA)] = AMDGPU_PL_OA, + [ilog2(AMDGPU_GEM_DOMAIN_DOORBELL)] = AMDGPU_PL_DOORBELL, + }; struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); struct ttm_resource *res = bo->tbo.resource; + struct drm_gem_object *obj = &bo->tbo.base; uint64_t size = amdgpu_bo_size(bo); - struct drm_gem_object *obj; - bool shared; + unsigned int type; - /* Abort if the BO doesn't currently have a backing store */ - if (!res) - return; + if (!res) { + /* + * If no backing store use one of the preferred domain for basic + * stats. We take the MSB since that should give a reasonable + * view. + */ + BUILD_BUG_ON(TTM_PL_VRAM < TTM_PL_TT || + TTM_PL_VRAM < TTM_PL_SYSTEM); + type = fls(bo->preferred_domains & AMDGPU_GEM_DOMAIN_MASK); + if (!type) + return; + type--; + if (drm_WARN_ON_ONCE(&adev->ddev, + type >= ARRAY_SIZE(domain_to_pl))) + return; + type = domain_to_pl[type]; + } else { + type = res->mem_type; + } - obj = &bo->tbo.base; - shared = drm_gem_object_is_shared_for_memory_stats(obj); - - switch (res->mem_type) { + /* Squash some into 'cpu' to keep the legacy userspace view. */ + switch (type) { case TTM_PL_VRAM: - stats->vram += size; - if (amdgpu_res_cpu_visible(adev, res)) - stats->visible_vram += size; - if (shared) - stats->vram_shared += size; - break; case TTM_PL_TT: - stats->gtt += size; - if (shared) - stats->gtt_shared += size; - break; case TTM_PL_SYSTEM: + break; default: - stats->cpu += size; - if (shared) - stats->cpu_shared += size; + type = TTM_PL_SYSTEM; break; } - if (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) { - stats->requested_vram += size; - if (bo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) - stats->requested_visible_vram += size; + if (drm_WARN_ON_ONCE(&adev->ddev, type >= sz)) + return; - if (res->mem_type != TTM_PL_VRAM) { - stats->evicted_vram += size; + /* DRM stats common fields: */ + + stats[type].total += size; + if (drm_gem_object_is_shared_for_memory_stats(obj)) + stats[type].drm.shared += size; + else + stats[type].drm.private += size; + + if (res) { + stats[type].drm.resident += size; + + if (!dma_resv_test_signaled(obj->resv, DMA_RESV_USAGE_BOOKKEEP)) + stats[type].drm.active += size; + else if (bo->flags & AMDGPU_GEM_CREATE_DISCARDABLE) + stats[type].drm.purgeable += size; + + if (type == TTM_PL_VRAM && amdgpu_res_cpu_visible(adev, res)) + stats[type].visible += size; + } + + /* amdgpu specific stats: */ + + if (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) { + stats[TTM_PL_VRAM].requested += size; + if (bo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) + stats[TTM_PL_VRAM].requested_visible += size; + + if (type != TTM_PL_VRAM) { + stats[TTM_PL_VRAM].evicted += size; if (bo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) - stats->evicted_visible_vram += size; + stats[TTM_PL_VRAM].evicted_visible += size; } } else if (bo->preferred_domains & AMDGPU_GEM_DOMAIN_GTT) { - stats->requested_gtt += size; + stats[TTM_PL_TT].requested += size; } } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index 717e47b46167..7260349917ef 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -140,30 +140,14 @@ struct amdgpu_bo_vm { }; struct amdgpu_mem_stats { - /* current VRAM usage, includes visible VRAM */ - uint64_t vram; - /* current shared VRAM usage, includes visible VRAM */ - uint64_t vram_shared; - /* current visible VRAM usage */ - uint64_t visible_vram; - /* current GTT usage */ - uint64_t gtt; - /* current shared GTT usage */ - uint64_t gtt_shared; - /* current system memory usage */ - uint64_t cpu; - /* current shared system memory usage */ - uint64_t cpu_shared; - /* sum of evicted buffers, includes visible VRAM */ - uint64_t evicted_vram; - /* sum of evicted buffers due to CPU access */ - uint64_t evicted_visible_vram; - /* how much userspace asked for, includes vis.VRAM */ - uint64_t requested_vram; - /* how much userspace asked for */ - uint64_t requested_visible_vram; - /* how much userspace asked for */ - uint64_t requested_gtt; + struct drm_memory_stats drm; + + uint64_t total; + uint64_t visible; + uint64_t evicted; + uint64_t evicted_visible; + uint64_t requested; + uint64_t requested_visible; }; static inline struct amdgpu_bo *ttm_to_amdgpu_bo(struct ttm_buffer_object *tbo) @@ -328,7 +312,8 @@ int amdgpu_bo_sync_wait(struct amdgpu_bo *bo, void *owner, bool intr); u64 amdgpu_bo_gpu_offset(struct amdgpu_bo *bo); u64 amdgpu_bo_gpu_offset_no_check(struct amdgpu_bo *bo); void amdgpu_bo_get_memory(struct amdgpu_bo *bo, - struct amdgpu_mem_stats *stats); + struct amdgpu_mem_stats *stats, + unsigned int size); uint32_t amdgpu_bo_get_preferred_domain(struct amdgpu_device *adev, uint32_t domain); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index 0b28b2cf1517..abd5e980c9c7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -159,9 +159,9 @@ static int psp_init_sriov_microcode(struct psp_context *psp) return ret; } -static int psp_early_init(void *handle) +static int psp_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct psp_context *psp = &adev->psp; psp->autoload_supported = true; @@ -421,9 +421,9 @@ static bool psp_get_runtime_db_entry(struct amdgpu_device *adev, return ret; } -static int psp_sw_init(void *handle) +static int psp_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct psp_context *psp = &adev->psp; int ret; struct psp_runtime_boot_cfg_entry boot_cfg_entry; @@ -527,9 +527,9 @@ failed1: return ret; } -static int psp_sw_fini(void *handle) +static int psp_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct psp_context *psp = &adev->psp; struct psp_gfx_cmd_resp *cmd = psp->cmd; @@ -639,6 +639,8 @@ static const char *psp_gfx_cmd_name(enum psp_gfx_cmd_id cmd_id) return "AUTOLOAD_RLC"; case GFX_CMD_ID_BOOT_CFG: return "BOOT_CFG"; + case GFX_CMD_ID_CONFIG_SQ_PERFMON: + return "CONFIG_SQ_PERFMON"; default: return "UNKNOWN CMD"; } @@ -1043,6 +1045,31 @@ static int psp_rl_load(struct amdgpu_device *adev) return ret; } +int psp_memory_partition(struct psp_context *psp, int mode) +{ + struct psp_gfx_cmd_resp *cmd; + int ret; + + if (amdgpu_sriov_vf(psp->adev)) + return 0; + + cmd = acquire_psp_cmd_buf(psp); + + cmd->cmd_id = GFX_CMD_ID_FB_NPS_MODE; + cmd->cmd.cmd_memory_part.mode = mode; + + dev_info(psp->adev->dev, + "Requesting %d memory partition change through PSP", mode); + ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr); + if (ret) + dev_err(psp->adev->dev, + "PSP request failed to change to NPS%d mode\n", mode); + + release_psp_cmd_buf(psp); + + return ret; +} + int psp_spatial_partition(struct psp_context *psp, int mode) { struct psp_gfx_cmd_resp *cmd; @@ -2264,6 +2291,19 @@ bool amdgpu_psp_get_ras_capability(struct psp_context *psp) } } +bool amdgpu_psp_tos_reload_needed(struct amdgpu_device *adev) +{ + struct psp_context *psp = &adev->psp; + + if (amdgpu_sriov_vf(adev) || (adev->flags & AMD_IS_APU)) + return false; + + if (psp->funcs && psp->funcs->is_reload_needed) + return psp->funcs->is_reload_needed(psp); + + return false; +} + static int psp_hw_start(struct psp_context *psp) { struct amdgpu_device *adev = psp->adev; @@ -2958,10 +2998,10 @@ failed: return ret; } -static int psp_hw_init(void *handle) +static int psp_hw_init(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; mutex_lock(&adev->firmware.mutex); /* @@ -2987,9 +3027,9 @@ failed: return -EINVAL; } -static int psp_hw_fini(void *handle) +static int psp_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct psp_context *psp = &adev->psp; if (psp->ta_fw) { @@ -3011,10 +3051,10 @@ static int psp_hw_fini(void *handle) return 0; } -static int psp_suspend(void *handle) +static int psp_suspend(struct amdgpu_ip_block *ip_block) { int ret = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct psp_context *psp = &adev->psp; if (adev->gmc.xgmi.num_physical_nodes > 1 && @@ -3074,10 +3114,10 @@ out: return ret; } -static int psp_resume(void *handle) +static int psp_resume(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct psp_context *psp = &adev->psp; dev_info(adev->dev, "PSP is resuming...\n"); @@ -3736,8 +3776,44 @@ out: return err; } +int psp_config_sq_perfmon(struct psp_context *psp, + uint32_t xcp_id, bool core_override_enable, + bool reg_override_enable, bool perfmon_override_enable) +{ + int ret; + + if (amdgpu_sriov_vf(psp->adev)) + return 0; + + if (xcp_id > MAX_XCP) { + dev_err(psp->adev->dev, "invalid xcp_id %d\n", xcp_id); + return -EINVAL; + } + + if (amdgpu_ip_version(psp->adev, MP0_HWIP, 0) != IP_VERSION(13, 0, 6)) { + dev_err(psp->adev->dev, "Unsupported MP0 version 0x%x for CONFIG_SQ_PERFMON command\n", + amdgpu_ip_version(psp->adev, MP0_HWIP, 0)); + return -EINVAL; + } + struct psp_gfx_cmd_resp *cmd = acquire_psp_cmd_buf(psp); + + cmd->cmd_id = GFX_CMD_ID_CONFIG_SQ_PERFMON; + cmd->cmd.config_sq_perfmon.gfx_xcp_mask = BIT_MASK(xcp_id); + cmd->cmd.config_sq_perfmon.core_override = core_override_enable; + cmd->cmd.config_sq_perfmon.reg_override = reg_override_enable; + cmd->cmd.config_sq_perfmon.perfmon_override = perfmon_override_enable; + + ret = psp_cmd_submit_buf(psp, NULL, cmd, psp->fence_buf_mc_addr); + if (ret) + dev_warn(psp->adev->dev, "PSP failed to config sq: xcp%d core%d reg%d perfmon%d\n", + xcp_id, core_override_enable, reg_override_enable, perfmon_override_enable); + + release_psp_cmd_buf(psp); + return ret; +} + static int psp_set_clockgating_state(void *handle, - enum amd_clockgating_state state) + enum amd_clockgating_state state) { return 0; } @@ -4019,17 +4095,12 @@ const struct attribute_group amdgpu_flash_attr_group = { const struct amd_ip_funcs psp_ip_funcs = { .name = "psp", .early_init = psp_early_init, - .late_init = NULL, .sw_init = psp_sw_init, .sw_fini = psp_sw_fini, .hw_init = psp_hw_init, .hw_fini = psp_hw_fini, .suspend = psp_suspend, .resume = psp_resume, - .is_idle = NULL, - .check_soft_reset = NULL, - .wait_for_idle = NULL, - .soft_reset = NULL, .set_clockgating_state = psp_set_clockgating_state, .set_powergating_state = psp_set_powergating_state, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index e8abbbcb4326..567cb1f924ca 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -139,6 +139,7 @@ struct psp_funcs { int (*fatal_error_recovery_quirk)(struct psp_context *psp); bool (*get_ras_capability)(struct psp_context *psp); bool (*is_aux_sos_load_required)(struct psp_context *psp); + bool (*is_reload_needed)(struct psp_context *psp); }; struct ta_funcs { @@ -552,9 +553,15 @@ int psp_load_fw_list(struct psp_context *psp, void psp_copy_fw(struct psp_context *psp, uint8_t *start_addr, uint32_t bin_size); int psp_spatial_partition(struct psp_context *psp, int mode); +int psp_memory_partition(struct psp_context *psp, int mode); int is_psp_fw_valid(struct psp_bin_desc bin); int amdgpu_psp_wait_for_bootloader(struct amdgpu_device *adev); bool amdgpu_psp_get_ras_capability(struct psp_context *psp); + +int psp_config_sq_perfmon(struct psp_context *psp, uint32_t xcp_id, + bool core_override_enable, bool reg_override_enable, bool perfmon_override_enable); +bool amdgpu_psp_tos_reload_needed(struct amdgpu_device *adev); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 1a1395c5fff1..1d9eda883bb8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -3146,7 +3146,42 @@ static int amdgpu_ras_page_retirement_thread(void *param) return 0; } -int amdgpu_ras_recovery_init(struct amdgpu_device *adev) +int amdgpu_ras_init_badpage_info(struct amdgpu_device *adev) +{ + struct amdgpu_ras *con = amdgpu_ras_get_context(adev); + int ret; + + if (!con || amdgpu_sriov_vf(adev)) + return 0; + + ret = amdgpu_ras_eeprom_init(&con->eeprom_control); + + if (ret) + return ret; + + /* HW not usable */ + if (amdgpu_ras_is_rma(adev)) + return -EHWPOISON; + + if (con->eeprom_control.ras_num_recs) { + ret = amdgpu_ras_load_bad_pages(adev); + if (ret) + return ret; + + amdgpu_dpm_send_hbm_bad_pages_num( + adev, con->eeprom_control.ras_num_recs); + + if (con->update_channel_flag == true) { + amdgpu_dpm_send_hbm_bad_channel_flag( + adev, con->eeprom_control.bad_channel_bitmap); + con->update_channel_flag = false; + } + } + + return ret; +} + +int amdgpu_ras_recovery_init(struct amdgpu_device *adev, bool init_bp_info) { struct amdgpu_ras *con = amdgpu_ras_get_context(adev); struct ras_err_handler_data **data; @@ -3181,31 +3216,10 @@ int amdgpu_ras_recovery_init(struct amdgpu_device *adev) max_eeprom_records_count = amdgpu_ras_eeprom_max_record_count(&con->eeprom_control); amdgpu_ras_validate_threshold(adev, max_eeprom_records_count); - /* Todo: During test the SMU might fail to read the eeprom through I2C - * when the GPU is pending on XGMI reset during probe time - * (Mostly after second bus reset), skip it now - */ - if (adev->gmc.xgmi.pending_reset) - return 0; - ret = amdgpu_ras_eeprom_init(&con->eeprom_control); - /* - * This calling fails when is_rma is true or - * ret != 0. - */ - if (amdgpu_ras_is_rma(adev) || ret) - goto free; - - if (con->eeprom_control.ras_num_recs) { - ret = amdgpu_ras_load_bad_pages(adev); + if (init_bp_info) { + ret = amdgpu_ras_init_badpage_info(adev); if (ret) goto free; - - amdgpu_dpm_send_hbm_bad_pages_num(adev, con->eeprom_control.ras_num_recs); - - if (con->update_channel_flag == true) { - amdgpu_dpm_send_hbm_bad_channel_flag(adev, con->eeprom_control.bad_channel_bitmap); - con->update_channel_flag = false; - } } mutex_init(&con->page_rsv_lock); @@ -4294,8 +4308,27 @@ int amdgpu_ras_reset_gpu(struct amdgpu_device *adev) ras->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE1_RESET; } - if (atomic_cmpxchg(&ras->in_recovery, 0, 1) == 0) + if (atomic_cmpxchg(&ras->in_recovery, 0, 1) == 0) { + struct amdgpu_hive_info *hive = amdgpu_get_xgmi_hive(adev); + int hive_ras_recovery = 0; + + if (hive) { + hive_ras_recovery = atomic_read(&hive->ras_recovery); + amdgpu_put_xgmi_hive(hive); + } + /* In the case of multiple GPUs, after a GPU has started + * resetting all GPUs on hive, other GPUs do not need to + * trigger GPU reset again. + */ + if (!hive_ras_recovery) + amdgpu_reset_domain_schedule(ras->adev->reset_domain, &ras->recovery_work); + else + atomic_set(&ras->in_recovery, 0); + } else { + flush_work(&ras->recovery_work); amdgpu_reset_domain_schedule(ras->adev->reset_domain, &ras->recovery_work); + } + return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h index 669720a9c60a..871b2d6278e0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h @@ -736,8 +736,8 @@ struct amdgpu_ras_block_hw_ops { * 8: feature disable */ - -int amdgpu_ras_recovery_init(struct amdgpu_device *adev); +int amdgpu_ras_init_badpage_info(struct amdgpu_device *adev); +int amdgpu_ras_recovery_init(struct amdgpu_device *adev, bool init_bp_info); void amdgpu_ras_resume(struct amdgpu_device *adev); void amdgpu_ras_suspend(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c index 66c1a868c0e1..24dae7cdbe95 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.c @@ -26,6 +26,156 @@ #include "sienna_cichlid.h" #include "smu_v13_0_10.h" +static int amdgpu_reset_xgmi_reset_on_init_suspend(struct amdgpu_device *adev) +{ + int i; + + for (i = adev->num_ip_blocks - 1; i >= 0; i--) { + if (!adev->ip_blocks[i].status.valid) + continue; + if (!adev->ip_blocks[i].status.hw) + continue; + /* displays are handled in phase1 */ + if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_DCE) + continue; + + /* XXX handle errors */ + amdgpu_ip_block_suspend(&adev->ip_blocks[i]); + adev->ip_blocks[i].status.hw = false; + } + + /* VCN FW shared region is in frambuffer, there are some flags + * initialized in that region during sw_init. Make sure the region is + * backed up. + */ + amdgpu_vcn_save_vcpu_bo(adev); + + return 0; +} + +static int amdgpu_reset_xgmi_reset_on_init_prep_hwctxt( + struct amdgpu_reset_control *reset_ctl, + struct amdgpu_reset_context *reset_context) +{ + struct list_head *reset_device_list = reset_context->reset_device_list; + struct amdgpu_device *tmp_adev; + int r; + + list_for_each_entry(tmp_adev, reset_device_list, reset_list) { + amdgpu_unregister_gpu_instance(tmp_adev); + r = amdgpu_reset_xgmi_reset_on_init_suspend(tmp_adev); + if (r) { + dev_err(tmp_adev->dev, + "xgmi reset on init: prepare for reset failed"); + return r; + } + } + + return r; +} + +static int amdgpu_reset_xgmi_reset_on_init_restore_hwctxt( + struct amdgpu_reset_control *reset_ctl, + struct amdgpu_reset_context *reset_context) +{ + struct list_head *reset_device_list = reset_context->reset_device_list; + struct amdgpu_device *tmp_adev = NULL; + int r; + + r = amdgpu_device_reinit_after_reset(reset_context); + if (r) + return r; + list_for_each_entry(tmp_adev, reset_device_list, reset_list) { + if (!tmp_adev->kfd.init_complete) { + kgd2kfd_init_zone_device(tmp_adev); + amdgpu_amdkfd_device_init(tmp_adev); + amdgpu_amdkfd_drm_client_create(tmp_adev); + } + } + + return r; +} + +static int amdgpu_reset_xgmi_reset_on_init_perform_reset( + struct amdgpu_reset_control *reset_ctl, + struct amdgpu_reset_context *reset_context) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)reset_ctl->handle; + struct list_head *reset_device_list = reset_context->reset_device_list; + struct amdgpu_device *tmp_adev = NULL; + int r; + + dev_dbg(adev->dev, "xgmi roi - hw reset\n"); + + list_for_each_entry(tmp_adev, reset_device_list, reset_list) { + mutex_lock(&tmp_adev->reset_cntl->reset_lock); + tmp_adev->reset_cntl->active_reset = + amdgpu_asic_reset_method(adev); + } + r = 0; + /* Mode1 reset needs to be triggered on all devices together */ + list_for_each_entry(tmp_adev, reset_device_list, reset_list) { + /* For XGMI run all resets in parallel to speed up the process */ + if (!queue_work(system_unbound_wq, &tmp_adev->xgmi_reset_work)) + r = -EALREADY; + if (r) { + dev_err(tmp_adev->dev, + "xgmi reset on init: reset failed with error, %d", + r); + break; + } + } + + /* For XGMI wait for all resets to complete before proceed */ + if (!r) { + list_for_each_entry(tmp_adev, reset_device_list, reset_list) { + flush_work(&tmp_adev->xgmi_reset_work); + r = tmp_adev->asic_reset_res; + if (r) + break; + } + } + + list_for_each_entry(tmp_adev, reset_device_list, reset_list) { + mutex_unlock(&tmp_adev->reset_cntl->reset_lock); + tmp_adev->reset_cntl->active_reset = AMD_RESET_METHOD_NONE; + } + + return r; +} + +int amdgpu_reset_do_xgmi_reset_on_init( + struct amdgpu_reset_context *reset_context) +{ + struct list_head *reset_device_list = reset_context->reset_device_list; + struct amdgpu_device *adev; + int r; + + if (!reset_device_list || list_empty(reset_device_list) || + list_is_singular(reset_device_list)) + return -EINVAL; + + adev = list_first_entry(reset_device_list, struct amdgpu_device, + reset_list); + r = amdgpu_reset_prepare_hwcontext(adev, reset_context); + if (r) + return r; + + r = amdgpu_reset_perform_reset(adev, reset_context); + + return r; +} + +struct amdgpu_reset_handler xgmi_reset_on_init_handler = { + .reset_method = AMD_RESET_METHOD_ON_INIT, + .prepare_env = NULL, + .prepare_hwcontext = amdgpu_reset_xgmi_reset_on_init_prep_hwctxt, + .perform_reset = amdgpu_reset_xgmi_reset_on_init_perform_reset, + .restore_hwcontext = amdgpu_reset_xgmi_reset_on_init_restore_hwctxt, + .restore_env = NULL, + .do_reset = NULL, +}; + int amdgpu_reset_init(struct amdgpu_device *adev) { int ret = 0; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h index 1cb920abc2fe..f8628bc898df 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_reset.h @@ -153,4 +153,9 @@ void amdgpu_reset_get_desc(struct amdgpu_reset_context *rst_ctxt, char *buf, for (i = 0; (i < AMDGPU_RESET_MAX_HANDLERS) && \ (handler = (*reset_ctl->reset_handlers)[i]); \ ++i) + +extern struct amdgpu_reset_handler xgmi_reset_on_init_handler; +int amdgpu_reset_do_xgmi_reset_on_init( + struct amdgpu_reset_context *reset_context); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 690976665cf6..42f616c05f50 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -108,10 +108,26 @@ int amdgpu_ring_alloc(struct amdgpu_ring *ring, unsigned int ndw) */ void amdgpu_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) { - int i; + uint32_t occupied, chunk1, chunk2; + uint32_t *dst; - for (i = 0; i < count; i++) - amdgpu_ring_write(ring, ring->funcs->nop); + occupied = ring->wptr & ring->buf_mask; + dst = (void *)&ring->ring[occupied]; + chunk1 = ring->buf_mask + 1 - occupied; + chunk1 = (chunk1 >= count) ? count : chunk1; + chunk2 = count - chunk1; + + if (chunk1) + memset32(dst, ring->funcs->nop, chunk1); + + if (chunk2) { + dst = (void *)ring->ring; + memset32(dst, ring->funcs->nop, chunk2); + } + + ring->wptr += count; + ring->wptr &= ring->ptr_mask; + ring->count_dw -= count; } /** @@ -141,6 +157,9 @@ void amdgpu_ring_commit(struct amdgpu_ring *ring) { uint32_t count; + if (ring->count_dw < 0) + DRM_ERROR("amdgpu: writing more dwords to the ring than expected!\n"); + /* We pad to match fetch size */ count = ring->funcs->align_mask + 1 - (ring->wptr & ring->funcs->align_mask); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index f93f51002201..574336d6714a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -377,8 +377,6 @@ static inline void amdgpu_ring_clear_ring(struct amdgpu_ring *ring) static inline void amdgpu_ring_write(struct amdgpu_ring *ring, uint32_t v) { - if (ring->count_dw <= 0) - DRM_ERROR("amdgpu: writing more dwords to the ring than expected!\n"); ring->ring[ring->wptr++ & ring->buf_mask] = v; ring->wptr &= ring->ptr_mask; ring->count_dw--; @@ -390,9 +388,6 @@ static inline void amdgpu_ring_write_multiple(struct amdgpu_ring *ring, unsigned occupied, chunk1, chunk2; void *dst; - if (unlikely(ring->count_dw < count_dw)) - DRM_ERROR("amdgpu: writing more dwords to the ring than expected!\n"); - occupied = ring->wptr & ring->buf_mask; dst = (void *)&ring->ring[occupied]; chunk1 = ring->buf_mask + 1 - occupied; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c index 74adb983ab03..0637414fc70e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c @@ -812,7 +812,7 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_device *bdev, /* Map SG to device */ r = dma_map_sgtable(adev->dev, ttm->sg, direction, 0); if (r) - goto release_sg; + goto release_sg_table; /* convert SG to linear array of pages and dma addresses */ drm_prime_sg_to_dma_addr_array(ttm->sg, gtt->ttm.dma_address, @@ -820,6 +820,8 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_device *bdev, return 0; +release_sg_table: + sg_free_table(ttm->sg); release_sg: kfree(ttm->sg); ttm->sg = NULL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h index 138d80017f35..2852a6064c9a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h @@ -34,6 +34,7 @@ #define AMDGPU_PL_OA (TTM_PL_PRIV + 2) #define AMDGPU_PL_PREEMPT (TTM_PL_PRIV + 3) #define AMDGPU_PL_DOORBELL (TTM_PL_PRIV + 4) +#define __AMDGPU_PL_LAST (TTM_PL_PRIV + 4) #define AMDGPU_GTT_MAX_TRANSFER_SIZE 512 #define AMDGPU_GTT_NUM_TRANSFER_WINDOWS 2 diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_umsch_mm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_umsch_mm.c index 6162582d0aa2..bd2d3863c3ed 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_umsch_mm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_umsch_mm.c @@ -765,9 +765,9 @@ static int umsch_mm_init(struct amdgpu_device *adev) } -static int umsch_mm_early_init(void *handle) +static int umsch_mm_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; switch (amdgpu_ip_version(adev, VCN_HWIP, 0)) { case IP_VERSION(4, 0, 5): @@ -784,9 +784,9 @@ static int umsch_mm_early_init(void *handle) return 0; } -static int umsch_mm_late_init(void *handle) +static int umsch_mm_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_in_reset(adev) || adev->in_s0ix || adev->in_suspend) return 0; @@ -794,9 +794,9 @@ static int umsch_mm_late_init(void *handle) return umsch_mm_test(adev); } -static int umsch_mm_sw_init(void *handle) +static int umsch_mm_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = umsch_mm_init(adev); @@ -815,9 +815,9 @@ static int umsch_mm_sw_init(void *handle) return 0; } -static int umsch_mm_sw_fini(void *handle) +static int umsch_mm_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; release_firmware(adev->umsch_mm.fw); adev->umsch_mm.fw = NULL; @@ -839,9 +839,9 @@ static int umsch_mm_sw_fini(void *handle) return 0; } -static int umsch_mm_hw_init(void *handle) +static int umsch_mm_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = umsch_mm_load_microcode(&adev->umsch_mm); @@ -857,9 +857,9 @@ static int umsch_mm_hw_init(void *handle) return 0; } -static int umsch_mm_hw_fini(void *handle) +static int umsch_mm_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; umsch_mm_ring_stop(&adev->umsch_mm); @@ -873,18 +873,14 @@ static int umsch_mm_hw_fini(void *handle) return 0; } -static int umsch_mm_suspend(void *handle) +static int umsch_mm_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return umsch_mm_hw_fini(adev); + return umsch_mm_hw_fini(ip_block); } -static int umsch_mm_resume(void *handle) +static int umsch_mm_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return umsch_mm_hw_init(adev); + return umsch_mm_hw_init(ip_block); } void amdgpu_umsch_fwlog_init(struct amdgpu_umsch_mm *umsch_mm) @@ -997,8 +993,6 @@ static const struct amd_ip_funcs umsch_mm_v4_0_ip_funcs = { .hw_fini = umsch_mm_hw_fini, .suspend = umsch_mm_suspend, .resume = umsch_mm_resume, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; const struct amdgpu_ip_block_version umsch_mm_v4_0_ip_block = { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index 43f44cc201cb..aecb78e0519f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -294,21 +294,12 @@ bool amdgpu_vcn_is_disabled_vcn(struct amdgpu_device *adev, enum vcn_ring_type t return ret; } -int amdgpu_vcn_suspend(struct amdgpu_device *adev) +int amdgpu_vcn_save_vcpu_bo(struct amdgpu_device *adev) { unsigned int size; void *ptr; int i, idx; - bool in_ras_intr = amdgpu_ras_intr_triggered(); - - cancel_delayed_work_sync(&adev->vcn.idle_work); - - /* err_event_athub will corrupt VCPU buffer, so we need to - * restore fw data and clear buffer in amdgpu_vcn_resume() */ - if (in_ras_intr) - return 0; - for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { if (adev->vcn.harvest_config & (1 << i)) continue; @@ -327,9 +318,24 @@ int amdgpu_vcn_suspend(struct amdgpu_device *adev) drm_dev_exit(idx); } } + return 0; } +int amdgpu_vcn_suspend(struct amdgpu_device *adev) +{ + bool in_ras_intr = amdgpu_ras_intr_triggered(); + + cancel_delayed_work_sync(&adev->vcn.idle_work); + + /* err_event_athub will corrupt VCPU buffer, so we need to + * restore fw data and clear buffer in amdgpu_vcn_resume() */ + if (in_ras_intr) + return 0; + + return amdgpu_vcn_save_vcpu_bo(adev); +} + int amdgpu_vcn_resume(struct amdgpu_device *adev) { unsigned int size; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h index 2a1f3dbb14d3..765b809d48a2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h @@ -518,5 +518,6 @@ int amdgpu_vcn_ras_sw_init(struct amdgpu_device *adev); int amdgpu_vcn_psp_update_sram(struct amdgpu_device *adev, int inst_idx, enum AMDGPU_UCODE_ID ucode_id); +int amdgpu_vcn_save_vcpu_bo(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c index d4c2afafbb73..8bf28d336807 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c @@ -493,10 +493,10 @@ const struct drm_mode_config_funcs amdgpu_vkms_mode_funcs = { .atomic_commit = drm_atomic_helper_commit, }; -static int amdgpu_vkms_sw_init(void *handle) +static int amdgpu_vkms_sw_init(struct amdgpu_ip_block *ip_block) { int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->amdgpu_vkms_output = kcalloc(adev->mode_info.num_crtc, sizeof(struct amdgpu_vkms_output), GFP_KERNEL); @@ -536,9 +536,9 @@ static int amdgpu_vkms_sw_init(void *handle) return 0; } -static int amdgpu_vkms_sw_fini(void *handle) +static int amdgpu_vkms_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i = 0; for (i = 0; i < adev->mode_info.num_crtc; i++) @@ -555,9 +555,9 @@ static int amdgpu_vkms_sw_fini(void *handle) return 0; } -static int amdgpu_vkms_hw_init(void *handle) +static int amdgpu_vkms_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; switch (adev->asic_type) { #ifdef CONFIG_DRM_AMDGPU_SI @@ -600,31 +600,31 @@ static int amdgpu_vkms_hw_init(void *handle) return 0; } -static int amdgpu_vkms_hw_fini(void *handle) +static int amdgpu_vkms_hw_fini(struct amdgpu_ip_block *ip_block) { return 0; } -static int amdgpu_vkms_suspend(void *handle) +static int amdgpu_vkms_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = drm_mode_config_helper_suspend(adev_to_drm(adev)); if (r) return r; - return amdgpu_vkms_hw_fini(handle); + + return 0; } -static int amdgpu_vkms_resume(void *handle) +static int amdgpu_vkms_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = amdgpu_vkms_hw_init(handle); + r = amdgpu_vkms_hw_init(ip_block); if (r) return r; - return drm_mode_config_helper_resume(adev_to_drm(adev)); + return drm_mode_config_helper_resume(adev_to_drm(ip_block->adev)); } static bool amdgpu_vkms_is_idle(void *handle) @@ -632,16 +632,6 @@ static bool amdgpu_vkms_is_idle(void *handle) return true; } -static int amdgpu_vkms_wait_for_idle(void *handle) -{ - return 0; -} - -static int amdgpu_vkms_soft_reset(void *handle) -{ - return 0; -} - static int amdgpu_vkms_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -656,8 +646,6 @@ static int amdgpu_vkms_set_powergating_state(void *handle, static const struct amd_ip_funcs amdgpu_vkms_ip_funcs = { .name = "amdgpu_vkms", - .early_init = NULL, - .late_init = NULL, .sw_init = amdgpu_vkms_sw_init, .sw_fini = amdgpu_vkms_sw_fini, .hw_init = amdgpu_vkms_hw_init, @@ -665,12 +653,8 @@ static const struct amd_ip_funcs amdgpu_vkms_ip_funcs = { .suspend = amdgpu_vkms_suspend, .resume = amdgpu_vkms_resume, .is_idle = amdgpu_vkms_is_idle, - .wait_for_idle = amdgpu_vkms_wait_for_idle, - .soft_reset = amdgpu_vkms_soft_reset, .set_clockgating_state = amdgpu_vkms_set_clockgating_state, .set_powergating_state = amdgpu_vkms_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; const struct amdgpu_ip_block_version amdgpu_vkms_ip_block = { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 6005280f5f38..6b855810ee86 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -1083,7 +1083,8 @@ error_free: } static void amdgpu_vm_bo_get_memory(struct amdgpu_bo_va *bo_va, - struct amdgpu_mem_stats *stats) + struct amdgpu_mem_stats *stats, + unsigned int size) { struct amdgpu_vm *vm = bo_va->base.vm; struct amdgpu_bo *bo = bo_va->base.bo; @@ -1099,34 +1100,35 @@ static void amdgpu_vm_bo_get_memory(struct amdgpu_bo_va *bo_va, !dma_resv_trylock(bo->tbo.base.resv)) return; - amdgpu_bo_get_memory(bo, stats); + amdgpu_bo_get_memory(bo, stats, size); if (!amdgpu_vm_is_bo_always_valid(vm, bo)) dma_resv_unlock(bo->tbo.base.resv); } void amdgpu_vm_get_memory(struct amdgpu_vm *vm, - struct amdgpu_mem_stats *stats) + struct amdgpu_mem_stats *stats, + unsigned int size) { struct amdgpu_bo_va *bo_va, *tmp; spin_lock(&vm->status_lock); list_for_each_entry_safe(bo_va, tmp, &vm->idle, base.vm_status) - amdgpu_vm_bo_get_memory(bo_va, stats); + amdgpu_vm_bo_get_memory(bo_va, stats, size); list_for_each_entry_safe(bo_va, tmp, &vm->evicted, base.vm_status) - amdgpu_vm_bo_get_memory(bo_va, stats); + amdgpu_vm_bo_get_memory(bo_va, stats, size); list_for_each_entry_safe(bo_va, tmp, &vm->relocated, base.vm_status) - amdgpu_vm_bo_get_memory(bo_va, stats); + amdgpu_vm_bo_get_memory(bo_va, stats, size); list_for_each_entry_safe(bo_va, tmp, &vm->moved, base.vm_status) - amdgpu_vm_bo_get_memory(bo_va, stats); + amdgpu_vm_bo_get_memory(bo_va, stats, size); list_for_each_entry_safe(bo_va, tmp, &vm->invalidated, base.vm_status) - amdgpu_vm_bo_get_memory(bo_va, stats); + amdgpu_vm_bo_get_memory(bo_va, stats, size); list_for_each_entry_safe(bo_va, tmp, &vm->done, base.vm_status) - amdgpu_vm_bo_get_memory(bo_va, stats); + amdgpu_vm_bo_get_memory(bo_va, stats, size); spin_unlock(&vm->status_lock); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index 52dd7cdfdc81..c5b41e3ed14f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -567,7 +567,8 @@ void amdgpu_vm_set_task_info(struct amdgpu_vm *vm); void amdgpu_vm_move_to_lru_tail(struct amdgpu_device *adev, struct amdgpu_vm *vm); void amdgpu_vm_get_memory(struct amdgpu_vm *vm, - struct amdgpu_mem_stats *stats); + struct amdgpu_mem_stats *stats, + unsigned int size); int amdgpu_vm_pt_clear(struct amdgpu_device *adev, struct amdgpu_vm *vm, struct amdgpu_bo_vm *vmbo, bool immediate); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c index 5acd20ff5979..46713a158d90 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vpe.c @@ -295,9 +295,9 @@ int amdgpu_vpe_ring_fini(struct amdgpu_vpe *vpe) return 0; } -static int vpe_early_init(void *handle) +static int vpe_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_vpe *vpe = &adev->vpe; switch (amdgpu_ip_version(adev, VPE_HWIP, 0)) { @@ -356,9 +356,9 @@ static int vpe_common_init(struct amdgpu_vpe *vpe) return 0; } -static int vpe_sw_init(void *handle) +static int vpe_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_vpe *vpe = &adev->vpe; int ret; @@ -381,9 +381,9 @@ out: return ret; } -static int vpe_sw_fini(void *handle) +static int vpe_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_vpe *vpe = &adev->vpe; release_firmware(vpe->fw); @@ -398,9 +398,9 @@ static int vpe_sw_fini(void *handle) return 0; } -static int vpe_hw_init(void *handle) +static int vpe_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_vpe *vpe = &adev->vpe; int ret; @@ -421,9 +421,9 @@ static int vpe_hw_init(void *handle) return 0; } -static int vpe_hw_fini(void *handle) +static int vpe_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_vpe *vpe = &adev->vpe; vpe_ring_stop(vpe); @@ -434,20 +434,18 @@ static int vpe_hw_fini(void *handle) return 0; } -static int vpe_suspend(void *handle) +static int vpe_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->vpe.idle_work); - return vpe_hw_fini(adev); + return vpe_hw_fini(ip_block); } -static int vpe_resume(void *handle) +static int vpe_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return vpe_hw_init(adev); + return vpe_hw_init(ip_block); } static void vpe_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count) @@ -908,14 +906,12 @@ static void vpe_set_ring_funcs(struct amdgpu_device *adev) const struct amd_ip_funcs vpe_ip_funcs = { .name = "vpe_v6_1", .early_init = vpe_early_init, - .late_init = NULL, .sw_init = vpe_sw_init, .sw_fini = vpe_sw_fini, .hw_init = vpe_hw_init, .hw_fini = vpe_hw_fini, .suspend = vpe_suspend, .resume = vpe_resume, - .soft_reset = NULL, .set_clockgating_state = vpe_set_clockgating_state, .set_powergating_state = vpe_set_powergating_state, }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c index a6d456ec6aeb..83a16918ea76 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.c @@ -433,3 +433,246 @@ void amdgpu_xcp_release_sched(struct amdgpu_device *adev, } } +#define XCP_CFG_SYSFS_RES_ATTR_SHOW(_name) \ + static ssize_t amdgpu_xcp_res_sysfs_##_name##_show( \ + struct amdgpu_xcp_res_details *xcp_res, char *buf) \ + { \ + return sysfs_emit(buf, "%d\n", xcp_res->_name); \ + } + +struct amdgpu_xcp_res_sysfs_attribute { + struct attribute attr; + ssize_t (*show)(struct amdgpu_xcp_res_details *xcp_res, char *buf); +}; + +#define XCP_CFG_SYSFS_RES_ATTR(_name) \ + struct amdgpu_xcp_res_sysfs_attribute xcp_res_sysfs_attr_##_name = { \ + .attr = { .name = __stringify(_name), .mode = 0400 }, \ + .show = amdgpu_xcp_res_sysfs_##_name##_show, \ + } + +XCP_CFG_SYSFS_RES_ATTR_SHOW(num_inst) +XCP_CFG_SYSFS_RES_ATTR(num_inst); +XCP_CFG_SYSFS_RES_ATTR_SHOW(num_shared) +XCP_CFG_SYSFS_RES_ATTR(num_shared); + +#define XCP_CFG_SYSFS_RES_ATTR_PTR(_name) xcp_res_sysfs_attr_##_name.attr + +static struct attribute *xcp_cfg_res_sysfs_attrs[] = { + &XCP_CFG_SYSFS_RES_ATTR_PTR(num_inst), + &XCP_CFG_SYSFS_RES_ATTR_PTR(num_shared), NULL +}; + +static const char *xcp_desc[] = { + [AMDGPU_SPX_PARTITION_MODE] = "SPX", + [AMDGPU_DPX_PARTITION_MODE] = "DPX", + [AMDGPU_TPX_PARTITION_MODE] = "TPX", + [AMDGPU_QPX_PARTITION_MODE] = "QPX", + [AMDGPU_CPX_PARTITION_MODE] = "CPX", +}; + +ATTRIBUTE_GROUPS(xcp_cfg_res_sysfs); + +#define to_xcp_attr(x) \ + container_of(x, struct amdgpu_xcp_res_sysfs_attribute, attr) +#define to_xcp_res(x) container_of(x, struct amdgpu_xcp_res_details, kobj) + +static ssize_t xcp_cfg_res_sysfs_attr_show(struct kobject *kobj, + struct attribute *attr, char *buf) +{ + struct amdgpu_xcp_res_sysfs_attribute *attribute; + struct amdgpu_xcp_res_details *xcp_res; + + attribute = to_xcp_attr(attr); + xcp_res = to_xcp_res(kobj); + + if (!attribute->show) + return -EIO; + + return attribute->show(xcp_res, buf); +} + +static const struct sysfs_ops xcp_cfg_res_sysfs_ops = { + .show = xcp_cfg_res_sysfs_attr_show, +}; + +static const struct kobj_type xcp_cfg_res_sysfs_ktype = { + .sysfs_ops = &xcp_cfg_res_sysfs_ops, + .default_groups = xcp_cfg_res_sysfs_groups, +}; + +const char *xcp_res_names[] = { + [AMDGPU_XCP_RES_XCC] = "xcc", + [AMDGPU_XCP_RES_DMA] = "dma", + [AMDGPU_XCP_RES_DEC] = "dec", + [AMDGPU_XCP_RES_JPEG] = "jpeg", +}; + +static int amdgpu_xcp_get_res_info(struct amdgpu_xcp_mgr *xcp_mgr, + int mode, + struct amdgpu_xcp_cfg *xcp_cfg) +{ + if (xcp_mgr->funcs && xcp_mgr->funcs->get_xcp_res_info) + return xcp_mgr->funcs->get_xcp_res_info(xcp_mgr, mode, xcp_cfg); + + return -EOPNOTSUPP; +} + +#define to_xcp_cfg(x) container_of(x, struct amdgpu_xcp_cfg, kobj) +static ssize_t supported_xcp_configs_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + struct amdgpu_xcp_cfg *xcp_cfg = to_xcp_cfg(kobj); + struct amdgpu_xcp_mgr *xcp_mgr = xcp_cfg->xcp_mgr; + int size = 0, mode; + char *sep = ""; + + if (!xcp_mgr || !xcp_mgr->supp_xcp_modes) + return sysfs_emit(buf, "Not supported\n"); + + for_each_inst(mode, xcp_mgr->supp_xcp_modes) { + size += sysfs_emit_at(buf, size, "%s%s", sep, xcp_desc[mode]); + sep = ", "; + } + + size += sysfs_emit_at(buf, size, "\n"); + + return size; +} + +static ssize_t xcp_config_show(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + struct amdgpu_xcp_cfg *xcp_cfg = to_xcp_cfg(kobj); + + return sysfs_emit(buf, "%s\n", + amdgpu_gfx_compute_mode_desc(xcp_cfg->mode)); +} + +static ssize_t xcp_config_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t size) +{ + struct amdgpu_xcp_cfg *xcp_cfg = to_xcp_cfg(kobj); + int mode, r; + + if (!strncasecmp("SPX", buf, strlen("SPX"))) + mode = AMDGPU_SPX_PARTITION_MODE; + else if (!strncasecmp("DPX", buf, strlen("DPX"))) + mode = AMDGPU_DPX_PARTITION_MODE; + else if (!strncasecmp("TPX", buf, strlen("TPX"))) + mode = AMDGPU_TPX_PARTITION_MODE; + else if (!strncasecmp("QPX", buf, strlen("QPX"))) + mode = AMDGPU_QPX_PARTITION_MODE; + else if (!strncasecmp("CPX", buf, strlen("CPX"))) + mode = AMDGPU_CPX_PARTITION_MODE; + else + return -EINVAL; + + r = amdgpu_xcp_get_res_info(xcp_cfg->xcp_mgr, mode, xcp_cfg); + + if (r) + return r; + + xcp_cfg->mode = mode; + return size; +} + +static struct kobj_attribute xcp_cfg_sysfs_mode = + __ATTR_RW_MODE(xcp_config, 0644); + +static void xcp_cfg_sysfs_release(struct kobject *kobj) +{ + struct amdgpu_xcp_cfg *xcp_cfg = to_xcp_cfg(kobj); + + kfree(xcp_cfg); +} + +static const struct kobj_type xcp_cfg_sysfs_ktype = { + .release = xcp_cfg_sysfs_release, + .sysfs_ops = &kobj_sysfs_ops, +}; + +static struct kobj_attribute supp_part_sysfs_mode = + __ATTR_RO(supported_xcp_configs); + +static const struct attribute *xcp_attrs[] = { + &supp_part_sysfs_mode.attr, + &xcp_cfg_sysfs_mode.attr, + NULL, +}; + +void amdgpu_xcp_cfg_sysfs_init(struct amdgpu_device *adev) +{ + struct amdgpu_xcp_res_details *xcp_res; + struct amdgpu_xcp_cfg *xcp_cfg; + int i, r, j, rid, mode; + + if (!adev->xcp_mgr) + return; + + xcp_cfg = kzalloc(sizeof(*xcp_cfg), GFP_KERNEL); + if (!xcp_cfg) + return; + xcp_cfg->xcp_mgr = adev->xcp_mgr; + + r = kobject_init_and_add(&xcp_cfg->kobj, &xcp_cfg_sysfs_ktype, + &adev->dev->kobj, "compute_partition_config"); + if (r) + goto err1; + + r = sysfs_create_files(&xcp_cfg->kobj, xcp_attrs); + if (r) + goto err1; + + mode = (xcp_cfg->xcp_mgr->mode == + AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE) ? + AMDGPU_SPX_PARTITION_MODE : + xcp_cfg->xcp_mgr->mode; + r = amdgpu_xcp_get_res_info(xcp_cfg->xcp_mgr, mode, xcp_cfg); + if (r) + goto err1; + + xcp_cfg->mode = mode; + for (i = 0; i < xcp_cfg->num_res; i++) { + xcp_res = &xcp_cfg->xcp_res[i]; + rid = xcp_res->id; + r = kobject_init_and_add(&xcp_res->kobj, + &xcp_cfg_res_sysfs_ktype, + &xcp_cfg->kobj, "%s", + xcp_res_names[rid]); + if (r) + goto err; + } + + adev->xcp_mgr->xcp_cfg = xcp_cfg; + return; +err: + for (j = 0; j < i; j++) { + xcp_res = &xcp_cfg->xcp_res[i]; + kobject_put(&xcp_res->kobj); + } + + sysfs_remove_files(&xcp_cfg->kobj, xcp_attrs); +err1: + kobject_put(&xcp_cfg->kobj); +} + +void amdgpu_xcp_cfg_sysfs_fini(struct amdgpu_device *adev) +{ + struct amdgpu_xcp_res_details *xcp_res; + struct amdgpu_xcp_cfg *xcp_cfg; + int i; + + if (!adev->xcp_mgr) + return; + + xcp_cfg = adev->xcp_mgr->xcp_cfg; + for (i = 0; i < xcp_cfg->num_res; i++) { + xcp_res = &xcp_cfg->xcp_res[i]; + kobject_put(&xcp_res->kobj); + } + + sysfs_remove_files(&xcp_cfg->kobj, xcp_attrs); + kobject_put(&xcp_cfg->kobj); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h index 32775260556f..7ac89d78a5bf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xcp.h @@ -56,6 +56,29 @@ enum AMDGPU_XCP_STATE { AMDGPU_XCP_RESUME, }; +enum amdgpu_xcp_res_id { + AMDGPU_XCP_RES_XCC, + AMDGPU_XCP_RES_DMA, + AMDGPU_XCP_RES_DEC, + AMDGPU_XCP_RES_JPEG, + AMDGPU_XCP_RES_MAX, +}; + +struct amdgpu_xcp_res_details { + enum amdgpu_xcp_res_id id; + u8 num_inst; + u8 num_shared; + struct kobject kobj; +}; + +struct amdgpu_xcp_cfg { + u8 mode; + struct amdgpu_xcp_res_details xcp_res[AMDGPU_XCP_RES_MAX]; + u8 num_res; + struct amdgpu_xcp_mgr *xcp_mgr; + struct kobject kobj; +}; + struct amdgpu_xcp_ip_funcs { int (*prepare_suspend)(void *handle, uint32_t inst_mask); int (*suspend)(void *handle, uint32_t inst_mask); @@ -97,6 +120,9 @@ struct amdgpu_xcp_mgr { /* Used to determine KFD memory size limits per XCP */ unsigned int num_xcp_per_mem_partition; + struct amdgpu_xcp_cfg *xcp_cfg; + uint32_t supp_xcp_modes; + uint32_t avail_xcp_modes; }; struct amdgpu_xcp_mgr_funcs { @@ -108,7 +134,9 @@ struct amdgpu_xcp_mgr_funcs { struct amdgpu_xcp_ip *ip); int (*get_xcp_mem_id)(struct amdgpu_xcp_mgr *xcp_mgr, struct amdgpu_xcp *xcp, uint8_t *mem_id); - + int (*get_xcp_res_info)(struct amdgpu_xcp_mgr *xcp_mgr, + int mode, + struct amdgpu_xcp_cfg *xcp_cfg); int (*prepare_suspend)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); int (*suspend)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); int (*prepare_resume)(struct amdgpu_xcp_mgr *xcp_mgr, int xcp_id); @@ -146,6 +174,9 @@ int amdgpu_xcp_open_device(struct amdgpu_device *adev, void amdgpu_xcp_release_sched(struct amdgpu_device *adev, struct amdgpu_ctx_entity *entity); +void amdgpu_xcp_cfg_sysfs_init(struct amdgpu_device *adev); +void amdgpu_xcp_cfg_sysfs_fini(struct amdgpu_device *adev); + #define amdgpu_xcp_select_scheds(adev, e, c, d, x, y) \ ((adev)->xcp_mgr && (adev)->xcp_mgr->funcs && \ (adev)->xcp_mgr->funcs->select_scheds ? \ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 7de449fae1e3..b47422b0b5b1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -667,6 +667,7 @@ struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev) task_barrier_init(&hive->tb); hive->pstate = AMDGPU_XGMI_PSTATE_UNKNOWN; hive->hi_req_gpu = NULL; + atomic_set(&hive->requested_nps_mode, UNKNOWN_MEMORY_PARTITION_MODE); /* * hive pstate on boot is high in vega20 so we have to go to low @@ -800,6 +801,23 @@ int amdgpu_xgmi_get_num_links(struct amdgpu_device *adev, return -EINVAL; } +bool amdgpu_xgmi_get_is_sharing_enabled(struct amdgpu_device *adev, + struct amdgpu_device *peer_adev) +{ + struct psp_xgmi_topology_info *top = &adev->psp.xgmi_context.top_info; + int i; + + /* Sharing should always be enabled for non-SRIOV. */ + if (!amdgpu_sriov_vf(adev)) + return true; + + for (i = 0 ; i < top->num_nodes; ++i) + if (top->nodes[i].node_id == peer_adev->gmc.xgmi.node_id) + return !!top->nodes[i].is_sharing_enabled; + + return false; +} + /* * Devices that support extended data require the entire hive to initialize with * the shared memory buffer flag set. @@ -860,8 +878,7 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev) if (!adev->gmc.xgmi.supported) return 0; - if (!adev->gmc.xgmi.pending_reset && - amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_PSP)) { + if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_PSP)) { ret = psp_xgmi_initialize(&adev->psp, false, true); if (ret) { dev_err(adev->dev, @@ -907,8 +924,7 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev) task_barrier_add_task(&hive->tb); - if (!adev->gmc.xgmi.pending_reset && - amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_PSP)) { + if (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_PSP)) { list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) { /* update node list for other device in the hive */ if (tmp_adev != adev) { @@ -985,7 +1001,7 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev) } } - if (!ret && !adev->gmc.xgmi.pending_reset) + if (!ret) ret = amdgpu_xgmi_sysfs_add_dev_info(adev, hive); exit_unlock: @@ -1500,3 +1516,117 @@ int amdgpu_xgmi_ras_sw_init(struct amdgpu_device *adev) return 0; } + +static void amdgpu_xgmi_reset_on_init_work(struct work_struct *work) +{ + struct amdgpu_hive_info *hive = + container_of(work, struct amdgpu_hive_info, reset_on_init_work); + struct amdgpu_reset_context reset_context; + struct amdgpu_device *tmp_adev; + struct list_head device_list; + int r; + + mutex_lock(&hive->hive_lock); + + INIT_LIST_HEAD(&device_list); + list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) + list_add_tail(&tmp_adev->reset_list, &device_list); + + tmp_adev = list_first_entry(&device_list, struct amdgpu_device, + reset_list); + amdgpu_device_lock_reset_domain(tmp_adev->reset_domain); + + reset_context.method = AMD_RESET_METHOD_ON_INIT; + reset_context.reset_req_dev = tmp_adev; + reset_context.hive = hive; + reset_context.reset_device_list = &device_list; + set_bit(AMDGPU_NEED_FULL_RESET, &reset_context.flags); + set_bit(AMDGPU_SKIP_COREDUMP, &reset_context.flags); + + amdgpu_reset_do_xgmi_reset_on_init(&reset_context); + mutex_unlock(&hive->hive_lock); + amdgpu_device_unlock_reset_domain(tmp_adev->reset_domain); + + list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) { + r = amdgpu_ras_init_badpage_info(tmp_adev); + if (r && r != -EHWPOISON) + dev_err(tmp_adev->dev, + "error during bad page data initialization"); + } +} + +static void amdgpu_xgmi_schedule_reset_on_init(struct amdgpu_hive_info *hive) +{ + INIT_WORK(&hive->reset_on_init_work, amdgpu_xgmi_reset_on_init_work); + amdgpu_reset_domain_schedule(hive->reset_domain, + &hive->reset_on_init_work); +} + +int amdgpu_xgmi_reset_on_init(struct amdgpu_device *adev) +{ + struct amdgpu_hive_info *hive; + bool reset_scheduled; + int num_devs; + + hive = amdgpu_get_xgmi_hive(adev); + if (!hive) + return -EINVAL; + + mutex_lock(&hive->hive_lock); + num_devs = atomic_read(&hive->number_devices); + reset_scheduled = false; + if (num_devs == adev->gmc.xgmi.num_physical_nodes) { + amdgpu_xgmi_schedule_reset_on_init(hive); + reset_scheduled = true; + } + + mutex_unlock(&hive->hive_lock); + amdgpu_put_xgmi_hive(hive); + + if (reset_scheduled) + flush_work(&hive->reset_on_init_work); + + return 0; +} + +int amdgpu_xgmi_request_nps_change(struct amdgpu_device *adev, + struct amdgpu_hive_info *hive, + int req_nps_mode) +{ + struct amdgpu_device *tmp_adev; + int cur_nps_mode, r; + + /* This is expected to be called only during unload of driver. The + * request needs to be placed only once for all devices in the hive. If + * one of them fail, revert the request for previous successful devices. + * After placing the request, make hive mode as UNKNOWN so that other + * devices don't request anymore. + */ + mutex_lock(&hive->hive_lock); + if (atomic_read(&hive->requested_nps_mode) == + UNKNOWN_MEMORY_PARTITION_MODE) { + dev_dbg(adev->dev, "Unexpected entry for hive NPS change"); + mutex_unlock(&hive->hive_lock); + return 0; + } + list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) { + r = adev->gmc.gmc_funcs->request_mem_partition_mode( + tmp_adev, req_nps_mode); + if (r) + break; + } + if (r) { + /* Request back current mode if one of the requests failed */ + cur_nps_mode = + adev->gmc.gmc_funcs->query_mem_partition_mode(tmp_adev); + list_for_each_entry_continue_reverse( + tmp_adev, &hive->device_list, gmc.xgmi.head) + adev->gmc.gmc_funcs->request_mem_partition_mode( + tmp_adev, cur_nps_mode); + } + /* Set to UNKNOWN so that other devices don't request anymore */ + atomic_set(&hive->requested_nps_mode, UNKNOWN_MEMORY_PARTITION_MODE); + mutex_unlock(&hive->hive_lock); + + return r; +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h index a3bfc16de6d4..8cc7ab38db7c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h @@ -45,6 +45,8 @@ struct amdgpu_hive_info { struct amdgpu_reset_domain *reset_domain; atomic_t ras_recovery; struct ras_event_manager event_mgr; + struct work_struct reset_on_init_work; + atomic_t requested_nps_mode; }; struct amdgpu_pcs_ras_field { @@ -64,6 +66,8 @@ int amdgpu_xgmi_get_hops_count(struct amdgpu_device *adev, struct amdgpu_device *peer_adev); int amdgpu_xgmi_get_num_links(struct amdgpu_device *adev, struct amdgpu_device *peer_adev); +bool amdgpu_xgmi_get_is_sharing_enabled(struct amdgpu_device *adev, + struct amdgpu_device *peer_adev); uint64_t amdgpu_xgmi_get_relative_phy_addr(struct amdgpu_device *adev, uint64_t addr); static inline bool amdgpu_xgmi_same_hive(struct amdgpu_device *adev, @@ -75,5 +79,10 @@ static inline bool amdgpu_xgmi_same_hive(struct amdgpu_device *adev, adev->gmc.xgmi.hive_id == bo_adev->gmc.xgmi.hive_id); } int amdgpu_xgmi_ras_sw_init(struct amdgpu_device *adev); +int amdgpu_xgmi_reset_on_init(struct amdgpu_device *adev); + +int amdgpu_xgmi_request_nps_change(struct amdgpu_device *adev, + struct amdgpu_hive_info *hive, + int req_nps_mode); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c index 5e8833e4fed2..890976b7ce77 100644 --- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c +++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c @@ -447,6 +447,61 @@ static int __aqua_vanjaram_get_xcp_ip_info(struct amdgpu_xcp_mgr *xcp_mgr, int x return 0; } +static int aqua_vanjaram_get_xcp_res_info(struct amdgpu_xcp_mgr *xcp_mgr, + int mode, + struct amdgpu_xcp_cfg *xcp_cfg) +{ + struct amdgpu_device *adev = xcp_mgr->adev; + int max_res[AMDGPU_XCP_RES_MAX] = {}; + bool res_lt_xcp; + int num_xcp, i; + + if (!(xcp_mgr->supp_xcp_modes & BIT(mode))) + return -EINVAL; + + max_res[AMDGPU_XCP_RES_XCC] = NUM_XCC(adev->gfx.xcc_mask); + max_res[AMDGPU_XCP_RES_DMA] = adev->sdma.num_instances; + max_res[AMDGPU_XCP_RES_DEC] = adev->vcn.num_vcn_inst; + max_res[AMDGPU_XCP_RES_JPEG] = adev->jpeg.num_jpeg_inst; + + switch (mode) { + case AMDGPU_SPX_PARTITION_MODE: + num_xcp = 1; + break; + case AMDGPU_DPX_PARTITION_MODE: + num_xcp = 2; + break; + case AMDGPU_TPX_PARTITION_MODE: + num_xcp = 3; + break; + case AMDGPU_QPX_PARTITION_MODE: + num_xcp = 4; + break; + case AMDGPU_CPX_PARTITION_MODE: + num_xcp = NUM_XCC(adev->gfx.xcc_mask); + break; + default: + return -EINVAL; + } + + xcp_cfg->num_res = ARRAY_SIZE(max_res); + + for (i = 0; i < xcp_cfg->num_res; i++) { + res_lt_xcp = max_res[i] < num_xcp; + xcp_cfg->xcp_res[i].id = i; + xcp_cfg->xcp_res[i].num_inst = + res_lt_xcp ? 1 : max_res[i] / num_xcp; + xcp_cfg->xcp_res[i].num_inst = + i == AMDGPU_XCP_RES_JPEG ? + xcp_cfg->xcp_res[i].num_inst * + adev->jpeg.num_jpeg_rings : xcp_cfg->xcp_res[i].num_inst; + xcp_cfg->xcp_res[i].num_shared = + res_lt_xcp ? num_xcp / max_res[i] : 1; + } + + return 0; +} + static enum amdgpu_gfx_partition __aqua_vanjaram_get_auto_mode(struct amdgpu_xcp_mgr *xcp_mgr) { @@ -530,6 +585,57 @@ static int __aqua_vanjaram_post_partition_switch(struct amdgpu_xcp_mgr *xcp_mgr, return ret; } +static void +__aqua_vanjaram_update_supported_modes(struct amdgpu_xcp_mgr *xcp_mgr) +{ + struct amdgpu_device *adev = xcp_mgr->adev; + + xcp_mgr->supp_xcp_modes = 0; + + switch (NUM_XCC(adev->gfx.xcc_mask)) { + case 8: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_DPX_PARTITION_MODE) | + BIT(AMDGPU_QPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + case 6: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_TPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + case 4: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_DPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + /* this seems only existing in emulation phase */ + case 2: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + case 1: + xcp_mgr->supp_xcp_modes = BIT(AMDGPU_SPX_PARTITION_MODE) | + BIT(AMDGPU_CPX_PARTITION_MODE); + break; + + default: + break; + } +} + +static void __aqua_vanjaram_update_available_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr) +{ + int mode; + + xcp_mgr->avail_xcp_modes = 0; + + for_each_inst(mode, xcp_mgr->supp_xcp_modes) { + if (__aqua_vanjaram_is_valid_mode(xcp_mgr, mode)) + xcp_mgr->avail_xcp_modes |= BIT(mode); + } +} + static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, int mode, int *num_xcps) { @@ -578,6 +684,8 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr, amdgpu_xcp_init(xcp_mgr, *num_xcps, mode); ret = __aqua_vanjaram_post_partition_switch(xcp_mgr, flags); + if (!ret) + __aqua_vanjaram_update_available_partition_mode(xcp_mgr); unlock: if (flags & AMDGPU_XCP_OPS_KFD) amdgpu_amdkfd_unlock_kfd(adev); @@ -656,9 +764,11 @@ struct amdgpu_xcp_mgr_funcs aqua_vanjaram_xcp_funcs = { .switch_partition_mode = &aqua_vanjaram_switch_partition_mode, .query_partition_mode = &aqua_vanjaram_query_partition_mode, .get_ip_details = &aqua_vanjaram_get_xcp_ip_details, + .get_xcp_res_info = &aqua_vanjaram_get_xcp_res_info, .get_xcp_mem_id = &aqua_vanjaram_get_xcp_mem_id, .select_scheds = &aqua_vanjaram_select_scheds, - .update_partition_sched_list = &aqua_vanjaram_update_partition_sched_list + .update_partition_sched_list = + &aqua_vanjaram_update_partition_sched_list }; static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev) @@ -673,6 +783,7 @@ static int aqua_vanjaram_xcp_mgr_init(struct amdgpu_device *adev) if (ret) return ret; + __aqua_vanjaram_update_supported_modes(adev->xcp_mgr); /* TODO: Default memory node affinity init */ return ret; diff --git a/drivers/gpu/drm/amd/amdgpu/atom.c b/drivers/gpu/drm/amd/amdgpu/atom.c index 09715b506468..81d195d366ce 100644 --- a/drivers/gpu/drm/amd/amdgpu/atom.c +++ b/drivers/gpu/drm/amd/amdgpu/atom.c @@ -27,7 +27,7 @@ #include #include -#include +#include #include diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c index cf1d5d462b67..e2cb1f080e88 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik.c +++ b/drivers/gpu/drm/amd/amdgpu/cik.c @@ -1985,9 +1985,9 @@ static const struct amdgpu_asic_funcs cik_asic_funcs = .query_video_codecs = &cik_query_video_codecs, }; -static int cik_common_early_init(void *handle) +static int cik_common_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->smc_rreg = &cik_smc_rreg; adev->smc_wreg = &cik_smc_wreg; @@ -2124,19 +2124,9 @@ static int cik_common_early_init(void *handle) return 0; } -static int cik_common_sw_init(void *handle) +static int cik_common_hw_init(struct amdgpu_ip_block *ip_block) { - return 0; -} - -static int cik_common_sw_fini(void *handle) -{ - return 0; -} - -static int cik_common_hw_init(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* move the golden regs per IP block */ cik_init_golden_registers(adev); @@ -2148,23 +2138,14 @@ static int cik_common_hw_init(void *handle) return 0; } -static int cik_common_hw_fini(void *handle) +static int cik_common_hw_fini(struct amdgpu_ip_block *ip_block) { return 0; } -static int cik_common_suspend(void *handle) +static int cik_common_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return cik_common_hw_fini(adev); -} - -static int cik_common_resume(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return cik_common_hw_init(adev); + return cik_common_hw_init(ip_block); } static bool cik_common_is_idle(void *handle) @@ -2172,12 +2153,9 @@ static bool cik_common_is_idle(void *handle) return true; } -static int cik_common_wait_for_idle(void *handle) -{ - return 0; -} -static int cik_common_soft_reset(void *handle) + +static int cik_common_soft_reset(struct amdgpu_ip_block *ip_block) { /* XXX hard reset?? */ return 0; @@ -2198,20 +2176,13 @@ static int cik_common_set_powergating_state(void *handle, static const struct amd_ip_funcs cik_common_ip_funcs = { .name = "cik_common", .early_init = cik_common_early_init, - .late_init = NULL, - .sw_init = cik_common_sw_init, - .sw_fini = cik_common_sw_fini, .hw_init = cik_common_hw_init, .hw_fini = cik_common_hw_fini, - .suspend = cik_common_suspend, .resume = cik_common_resume, .is_idle = cik_common_is_idle, - .wait_for_idle = cik_common_wait_for_idle, .soft_reset = cik_common_soft_reset, .set_clockgating_state = cik_common_set_clockgating_state, .set_powergating_state = cik_common_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ip_block_version cik_common_ip_block = diff --git a/drivers/gpu/drm/amd/amdgpu/cik_ih.c b/drivers/gpu/drm/amd/amdgpu/cik_ih.c index 576baa9dbb0e..1da17755ad53 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/cik_ih.c @@ -283,9 +283,9 @@ static void cik_ih_set_rptr(struct amdgpu_device *adev, WREG32(mmIH_RB_RPTR, ih->rptr); } -static int cik_ih_early_init(void *handle) +static int cik_ih_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; ret = amdgpu_irq_add_domain(adev); @@ -297,10 +297,10 @@ static int cik_ih_early_init(void *handle) return 0; } -static int cik_ih_sw_init(void *handle) +static int cik_ih_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 64 * 1024, false); if (r) @@ -311,9 +311,9 @@ static int cik_ih_sw_init(void *handle) return r; } -static int cik_ih_sw_fini(void *handle) +static int cik_ih_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_fini_sw(adev); amdgpu_irq_remove_domain(adev); @@ -321,34 +321,28 @@ static int cik_ih_sw_fini(void *handle) return 0; } -static int cik_ih_hw_init(void *handle) +static int cik_ih_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return cik_ih_irq_init(adev); } -static int cik_ih_hw_fini(void *handle) +static int cik_ih_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - cik_ih_irq_disable(adev); + cik_ih_irq_disable(ip_block->adev); return 0; } -static int cik_ih_suspend(void *handle) +static int cik_ih_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return cik_ih_hw_fini(adev); + return cik_ih_hw_fini(ip_block); } -static int cik_ih_resume(void *handle) +static int cik_ih_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return cik_ih_hw_init(adev); + return cik_ih_hw_init(ip_block); } static bool cik_ih_is_idle(void *handle) @@ -362,11 +356,11 @@ static bool cik_ih_is_idle(void *handle) return true; } -static int cik_ih_wait_for_idle(void *handle) +static int cik_ih_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { /* read MC_STATUS */ @@ -378,9 +372,9 @@ static int cik_ih_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int cik_ih_soft_reset(void *handle) +static int cik_ih_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset = 0; u32 tmp = RREG32(mmSRBM_STATUS); @@ -423,7 +417,6 @@ static int cik_ih_set_powergating_state(void *handle, static const struct amd_ip_funcs cik_ih_ip_funcs = { .name = "cik_ih", .early_init = cik_ih_early_init, - .late_init = NULL, .sw_init = cik_ih_sw_init, .sw_fini = cik_ih_sw_fini, .hw_init = cik_ih_hw_init, @@ -435,8 +428,6 @@ static const struct amd_ip_funcs cik_ih_ip_funcs = { .soft_reset = cik_ih_soft_reset, .set_clockgating_state = cik_ih_set_clockgating_state, .set_powergating_state = cik_ih_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ih_funcs cik_ih_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c index 952737de9411..ede1a028d48d 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c +++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c @@ -54,7 +54,7 @@ static void cik_sdma_set_ring_funcs(struct amdgpu_device *adev); static void cik_sdma_set_irq_funcs(struct amdgpu_device *adev); static void cik_sdma_set_buffer_funcs(struct amdgpu_device *adev); static void cik_sdma_set_vm_pte_funcs(struct amdgpu_device *adev); -static int cik_sdma_soft_reset(void *handle); +static int cik_sdma_soft_reset(struct amdgpu_ip_block *ip_block); MODULE_FIRMWARE("amdgpu/bonaire_sdma.bin"); MODULE_FIRMWARE("amdgpu/bonaire_sdma1.bin"); @@ -918,9 +918,9 @@ static void cik_enable_sdma_mgls(struct amdgpu_device *adev, } } -static int cik_sdma_early_init(void *handle) +static int cik_sdma_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; adev->sdma.num_instances = SDMA_MAX_INSTANCE; @@ -937,10 +937,10 @@ static int cik_sdma_early_init(void *handle) return 0; } -static int cik_sdma_sw_init(void *handle) +static int cik_sdma_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r, i; /* SDMA trap event */ @@ -977,9 +977,9 @@ static int cik_sdma_sw_init(void *handle) return r; } -static int cik_sdma_sw_fini(void *handle) +static int cik_sdma_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->sdma.num_instances; i++) @@ -989,10 +989,10 @@ static int cik_sdma_sw_fini(void *handle) return 0; } -static int cik_sdma_hw_init(void *handle) +static int cik_sdma_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = cik_sdma_start(adev); if (r) @@ -1001,9 +1001,9 @@ static int cik_sdma_hw_init(void *handle) return r; } -static int cik_sdma_hw_fini(void *handle) +static int cik_sdma_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cik_ctx_switch_enable(adev, false); cik_sdma_enable(adev, false); @@ -1011,20 +1011,16 @@ static int cik_sdma_hw_fini(void *handle) return 0; } -static int cik_sdma_suspend(void *handle) +static int cik_sdma_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return cik_sdma_hw_fini(adev); + return cik_sdma_hw_fini(ip_block); } -static int cik_sdma_resume(void *handle) +static int cik_sdma_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + cik_sdma_soft_reset(ip_block); - cik_sdma_soft_reset(handle); - - return cik_sdma_hw_init(adev); + return cik_sdma_hw_init(ip_block); } static bool cik_sdma_is_idle(void *handle) @@ -1039,11 +1035,11 @@ static bool cik_sdma_is_idle(void *handle) return true; } -static int cik_sdma_wait_for_idle(void *handle) +static int cik_sdma_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { tmp = RREG32(mmSRBM_STATUS2) & (SRBM_STATUS2__SDMA_BUSY_MASK | @@ -1056,10 +1052,10 @@ static int cik_sdma_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int cik_sdma_soft_reset(void *handle) +static int cik_sdma_soft_reset(struct amdgpu_ip_block *ip_block) { u32 srbm_soft_reset = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 tmp; /* sdma0 */ @@ -1217,7 +1213,6 @@ static int cik_sdma_set_powergating_state(void *handle, static const struct amd_ip_funcs cik_sdma_ip_funcs = { .name = "cik_sdma", .early_init = cik_sdma_early_init, - .late_init = NULL, .sw_init = cik_sdma_sw_init, .sw_fini = cik_sdma_sw_fini, .hw_init = cik_sdma_hw_init, @@ -1229,8 +1224,6 @@ static const struct amd_ip_funcs cik_sdma_ip_funcs = { .soft_reset = cik_sdma_soft_reset, .set_clockgating_state = cik_sdma_set_clockgating_state, .set_powergating_state = cik_sdma_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs cik_sdma_ring_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/cz_ih.c b/drivers/gpu/drm/amd/amdgpu/cz_ih.c index 072643787384..d72973bd570d 100644 --- a/drivers/gpu/drm/amd/amdgpu/cz_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/cz_ih.c @@ -274,9 +274,9 @@ static void cz_ih_set_rptr(struct amdgpu_device *adev, WREG32(mmIH_RB_RPTR, ih->rptr); } -static int cz_ih_early_init(void *handle) +static int cz_ih_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; ret = amdgpu_irq_add_domain(adev); @@ -288,10 +288,10 @@ static int cz_ih_early_init(void *handle) return 0; } -static int cz_ih_sw_init(void *handle) +static int cz_ih_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 64 * 1024, false); if (r) @@ -302,9 +302,9 @@ static int cz_ih_sw_init(void *handle) return r; } -static int cz_ih_sw_fini(void *handle) +static int cz_ih_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_fini_sw(adev); amdgpu_irq_remove_domain(adev); @@ -312,10 +312,10 @@ static int cz_ih_sw_fini(void *handle) return 0; } -static int cz_ih_hw_init(void *handle) +static int cz_ih_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = cz_ih_irq_init(adev); if (r) @@ -324,27 +324,21 @@ static int cz_ih_hw_init(void *handle) return 0; } -static int cz_ih_hw_fini(void *handle) +static int cz_ih_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - cz_ih_irq_disable(adev); + cz_ih_irq_disable(ip_block->adev); return 0; } -static int cz_ih_suspend(void *handle) +static int cz_ih_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return cz_ih_hw_fini(adev); + return cz_ih_hw_fini(ip_block); } -static int cz_ih_resume(void *handle) +static int cz_ih_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return cz_ih_hw_init(adev); + return cz_ih_hw_init(ip_block); } static bool cz_ih_is_idle(void *handle) @@ -358,11 +352,11 @@ static bool cz_ih_is_idle(void *handle) return true; } -static int cz_ih_wait_for_idle(void *handle) +static int cz_ih_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { /* read MC_STATUS */ @@ -374,10 +368,10 @@ static int cz_ih_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int cz_ih_soft_reset(void *handle) +static int cz_ih_soft_reset(struct amdgpu_ip_block *ip_block) { u32 srbm_soft_reset = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 tmp = RREG32(mmSRBM_STATUS); if (tmp & SRBM_STATUS__IH_BUSY_MASK) @@ -421,7 +415,6 @@ static int cz_ih_set_powergating_state(void *handle, static const struct amd_ip_funcs cz_ih_ip_funcs = { .name = "cz_ih", .early_init = cz_ih_early_init, - .late_init = NULL, .sw_init = cz_ih_sw_init, .sw_fini = cz_ih_sw_fini, .hw_init = cz_ih_hw_init, @@ -433,8 +426,6 @@ static const struct amd_ip_funcs cz_ih_ip_funcs = { .soft_reset = cz_ih_soft_reset, .set_clockgating_state = cz_ih_set_clockgating_state, .set_powergating_state = cz_ih_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ih_funcs cz_ih_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c index 70c1399f738d..5098c50d54c8 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c @@ -2738,9 +2738,9 @@ static int dce_v10_0_crtc_init(struct amdgpu_device *adev, int index) return 0; } -static int dce_v10_0_early_init(void *handle) +static int dce_v10_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->audio_endpt_rreg = &dce_v10_0_audio_endpt_rreg; adev->audio_endpt_wreg = &dce_v10_0_audio_endpt_wreg; @@ -2765,10 +2765,10 @@ static int dce_v10_0_early_init(void *handle) return 0; } -static int dce_v10_0_sw_init(void *handle) +static int dce_v10_0_sw_init(struct amdgpu_ip_block *ip_block) { int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->mode_info.num_crtc; i++) { r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, i + 1, &adev->crtc_irq); @@ -2844,9 +2844,9 @@ static int dce_v10_0_sw_init(void *handle) return 0; } -static int dce_v10_0_sw_fini(void *handle) +static int dce_v10_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; drm_edid_free(adev->mode_info.bios_hardcoded_edid); @@ -2862,10 +2862,10 @@ static int dce_v10_0_sw_fini(void *handle) return 0; } -static int dce_v10_0_hw_init(void *handle) +static int dce_v10_0_hw_init(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; dce_v10_0_init_golden_registers(adev); @@ -2887,10 +2887,10 @@ static int dce_v10_0_hw_init(void *handle) return 0; } -static int dce_v10_0_hw_fini(void *handle) +static int dce_v10_0_hw_fini(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; dce_v10_0_hpd_fini(adev); @@ -2905,9 +2905,9 @@ static int dce_v10_0_hw_fini(void *handle) return 0; } -static int dce_v10_0_suspend(void *handle) +static int dce_v10_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_display_suspend_helper(adev); @@ -2917,18 +2917,18 @@ static int dce_v10_0_suspend(void *handle) adev->mode_info.bl_level = amdgpu_atombios_encoder_get_backlight_level_from_reg(adev); - return dce_v10_0_hw_fini(handle); + return dce_v10_0_hw_fini(ip_block); } -static int dce_v10_0_resume(void *handle) +static int dce_v10_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; amdgpu_atombios_encoder_set_backlight_level_to_reg(adev, adev->mode_info.bl_level); - ret = dce_v10_0_hw_init(handle); + ret = dce_v10_0_hw_init(ip_block); /* turn on the BL */ if (adev->mode_info.bl_encoder) { @@ -2948,22 +2948,17 @@ static bool dce_v10_0_is_idle(void *handle) return true; } -static int dce_v10_0_wait_for_idle(void *handle) +static bool dce_v10_0_check_soft_reset(struct amdgpu_ip_block *ip_block) { - return 0; -} - -static bool dce_v10_0_check_soft_reset(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return dce_v10_0_is_display_hung(adev); } -static int dce_v10_0_soft_reset(void *handle) +static int dce_v10_0_soft_reset(struct amdgpu_ip_block *ip_block) { u32 srbm_soft_reset = 0, tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (dce_v10_0_is_display_hung(adev)) srbm_soft_reset |= SRBM_SOFT_RESET__SOFT_RESET_DC_MASK; @@ -3322,7 +3317,6 @@ static int dce_v10_0_set_powergating_state(void *handle, static const struct amd_ip_funcs dce_v10_0_ip_funcs = { .name = "dce_v10_0", .early_init = dce_v10_0_early_init, - .late_init = NULL, .sw_init = dce_v10_0_sw_init, .sw_fini = dce_v10_0_sw_fini, .hw_init = dce_v10_0_hw_init, @@ -3330,13 +3324,10 @@ static const struct amd_ip_funcs dce_v10_0_ip_funcs = { .suspend = dce_v10_0_suspend, .resume = dce_v10_0_resume, .is_idle = dce_v10_0_is_idle, - .wait_for_idle = dce_v10_0_wait_for_idle, .check_soft_reset = dce_v10_0_check_soft_reset, .soft_reset = dce_v10_0_soft_reset, .set_clockgating_state = dce_v10_0_set_clockgating_state, .set_powergating_state = dce_v10_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static void diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c index f154c24499c8..c5680ff4ab9f 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c @@ -2851,9 +2851,9 @@ static int dce_v11_0_crtc_init(struct amdgpu_device *adev, int index) return 0; } -static int dce_v11_0_early_init(void *handle) +static int dce_v11_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->audio_endpt_rreg = &dce_v11_0_audio_endpt_rreg; adev->audio_endpt_wreg = &dce_v11_0_audio_endpt_wreg; @@ -2891,10 +2891,10 @@ static int dce_v11_0_early_init(void *handle) return 0; } -static int dce_v11_0_sw_init(void *handle) +static int dce_v11_0_sw_init(struct amdgpu_ip_block *ip_block) { int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->mode_info.num_crtc; i++) { r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, i + 1, &adev->crtc_irq); @@ -2971,9 +2971,9 @@ static int dce_v11_0_sw_init(void *handle) return 0; } -static int dce_v11_0_sw_fini(void *handle) +static int dce_v11_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; drm_edid_free(adev->mode_info.bios_hardcoded_edid); @@ -2989,10 +2989,10 @@ static int dce_v11_0_sw_fini(void *handle) return 0; } -static int dce_v11_0_hw_init(void *handle) +static int dce_v11_0_hw_init(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; dce_v11_0_init_golden_registers(adev); @@ -3025,10 +3025,10 @@ static int dce_v11_0_hw_init(void *handle) return 0; } -static int dce_v11_0_hw_fini(void *handle) +static int dce_v11_0_hw_fini(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; dce_v11_0_hpd_fini(adev); @@ -3043,9 +3043,9 @@ static int dce_v11_0_hw_fini(void *handle) return 0; } -static int dce_v11_0_suspend(void *handle) +static int dce_v11_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_display_suspend_helper(adev); @@ -3055,18 +3055,18 @@ static int dce_v11_0_suspend(void *handle) adev->mode_info.bl_level = amdgpu_atombios_encoder_get_backlight_level_from_reg(adev); - return dce_v11_0_hw_fini(handle); + return dce_v11_0_hw_fini(ip_block); } -static int dce_v11_0_resume(void *handle) +static int dce_v11_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; amdgpu_atombios_encoder_set_backlight_level_to_reg(adev, adev->mode_info.bl_level); - ret = dce_v11_0_hw_init(handle); + ret = dce_v11_0_hw_init(ip_block); /* turn on the BL */ if (adev->mode_info.bl_encoder) { @@ -3086,15 +3086,10 @@ static bool dce_v11_0_is_idle(void *handle) return true; } -static int dce_v11_0_wait_for_idle(void *handle) -{ - return 0; -} - -static int dce_v11_0_soft_reset(void *handle) +static int dce_v11_0_soft_reset(struct amdgpu_ip_block *ip_block) { u32 srbm_soft_reset = 0, tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (dce_v11_0_is_display_hung(adev)) srbm_soft_reset |= SRBM_SOFT_RESET__SOFT_RESET_DC_MASK; @@ -3454,7 +3449,6 @@ static int dce_v11_0_set_powergating_state(void *handle, static const struct amd_ip_funcs dce_v11_0_ip_funcs = { .name = "dce_v11_0", .early_init = dce_v11_0_early_init, - .late_init = NULL, .sw_init = dce_v11_0_sw_init, .sw_fini = dce_v11_0_sw_fini, .hw_init = dce_v11_0_hw_init, @@ -3462,12 +3456,9 @@ static const struct amd_ip_funcs dce_v11_0_ip_funcs = { .suspend = dce_v11_0_suspend, .resume = dce_v11_0_resume, .is_idle = dce_v11_0_is_idle, - .wait_for_idle = dce_v11_0_wait_for_idle, .soft_reset = dce_v11_0_soft_reset, .set_clockgating_state = dce_v11_0_set_clockgating_state, .set_powergating_state = dce_v11_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static void diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c index a7fcb135827f..eb7de9122d99 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v6_0.c @@ -2633,9 +2633,9 @@ static int dce_v6_0_crtc_init(struct amdgpu_device *adev, int index) return 0; } -static int dce_v6_0_early_init(void *handle) +static int dce_v6_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->audio_endpt_rreg = &dce_v6_0_audio_endpt_rreg; adev->audio_endpt_wreg = &dce_v6_0_audio_endpt_wreg; @@ -2664,11 +2664,11 @@ static int dce_v6_0_early_init(void *handle) return 0; } -static int dce_v6_0_sw_init(void *handle) +static int dce_v6_0_sw_init(struct amdgpu_ip_block *ip_block) { int r, i; bool ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->mode_info.num_crtc; i++) { r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, i + 1, &adev->crtc_irq); @@ -2743,9 +2743,9 @@ static int dce_v6_0_sw_init(void *handle) return r; } -static int dce_v6_0_sw_fini(void *handle) +static int dce_v6_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; drm_edid_free(adev->mode_info.bios_hardcoded_edid); @@ -2760,10 +2760,10 @@ static int dce_v6_0_sw_fini(void *handle) return 0; } -static int dce_v6_0_hw_init(void *handle) +static int dce_v6_0_hw_init(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* disable vga render */ dce_v6_0_set_vga_render_state(adev, false); @@ -2783,10 +2783,10 @@ static int dce_v6_0_hw_init(void *handle) return 0; } -static int dce_v6_0_hw_fini(void *handle) +static int dce_v6_0_hw_fini(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; dce_v6_0_hpd_fini(adev); @@ -2801,9 +2801,9 @@ static int dce_v6_0_hw_fini(void *handle) return 0; } -static int dce_v6_0_suspend(void *handle) +static int dce_v6_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_display_suspend_helper(adev); @@ -2812,18 +2812,18 @@ static int dce_v6_0_suspend(void *handle) adev->mode_info.bl_level = amdgpu_atombios_encoder_get_backlight_level_from_reg(adev); - return dce_v6_0_hw_fini(handle); + return dce_v6_0_hw_fini(ip_block); } -static int dce_v6_0_resume(void *handle) +static int dce_v6_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; amdgpu_atombios_encoder_set_backlight_level_to_reg(adev, adev->mode_info.bl_level); - ret = dce_v6_0_hw_init(handle); + ret = dce_v6_0_hw_init(ip_block); /* turn on the BL */ if (adev->mode_info.bl_encoder) { @@ -2843,12 +2843,7 @@ static bool dce_v6_0_is_idle(void *handle) return true; } -static int dce_v6_0_wait_for_idle(void *handle) -{ - return 0; -} - -static int dce_v6_0_soft_reset(void *handle) +static int dce_v6_0_soft_reset(struct amdgpu_ip_block *ip_block) { DRM_INFO("xxxx: dce_v6_0_soft_reset --- no impl!!\n"); return 0; @@ -3144,7 +3139,6 @@ static int dce_v6_0_set_powergating_state(void *handle, static const struct amd_ip_funcs dce_v6_0_ip_funcs = { .name = "dce_v6_0", .early_init = dce_v6_0_early_init, - .late_init = NULL, .sw_init = dce_v6_0_sw_init, .sw_fini = dce_v6_0_sw_fini, .hw_init = dce_v6_0_hw_init, @@ -3152,12 +3146,9 @@ static const struct amd_ip_funcs dce_v6_0_ip_funcs = { .suspend = dce_v6_0_suspend, .resume = dce_v6_0_resume, .is_idle = dce_v6_0_is_idle, - .wait_for_idle = dce_v6_0_wait_for_idle, .soft_reset = dce_v6_0_soft_reset, .set_clockgating_state = dce_v6_0_set_clockgating_state, .set_powergating_state = dce_v6_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static void diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c index 77ac3f114d24..04b79ff87f75 100644 --- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c @@ -2644,9 +2644,9 @@ static int dce_v8_0_crtc_init(struct amdgpu_device *adev, int index) return 0; } -static int dce_v8_0_early_init(void *handle) +static int dce_v8_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->audio_endpt_rreg = &dce_v8_0_audio_endpt_rreg; adev->audio_endpt_wreg = &dce_v8_0_audio_endpt_wreg; @@ -2680,10 +2680,10 @@ static int dce_v8_0_early_init(void *handle) return 0; } -static int dce_v8_0_sw_init(void *handle) +static int dce_v8_0_sw_init(struct amdgpu_ip_block *ip_block) { int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->mode_info.num_crtc; i++) { r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, i + 1, &adev->crtc_irq); @@ -2764,9 +2764,9 @@ static int dce_v8_0_sw_init(void *handle) return 0; } -static int dce_v8_0_sw_fini(void *handle) +static int dce_v8_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; drm_edid_free(adev->mode_info.bios_hardcoded_edid); @@ -2782,10 +2782,10 @@ static int dce_v8_0_sw_fini(void *handle) return 0; } -static int dce_v8_0_hw_init(void *handle) +static int dce_v8_0_hw_init(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* disable vga render */ dce_v8_0_set_vga_render_state(adev, false); @@ -2805,10 +2805,10 @@ static int dce_v8_0_hw_init(void *handle) return 0; } -static int dce_v8_0_hw_fini(void *handle) +static int dce_v8_0_hw_fini(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; dce_v8_0_hpd_fini(adev); @@ -2823,9 +2823,9 @@ static int dce_v8_0_hw_fini(void *handle) return 0; } -static int dce_v8_0_suspend(void *handle) +static int dce_v8_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_display_suspend_helper(adev); @@ -2835,18 +2835,18 @@ static int dce_v8_0_suspend(void *handle) adev->mode_info.bl_level = amdgpu_atombios_encoder_get_backlight_level_from_reg(adev); - return dce_v8_0_hw_fini(handle); + return dce_v8_0_hw_fini(ip_block); } -static int dce_v8_0_resume(void *handle) +static int dce_v8_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; amdgpu_atombios_encoder_set_backlight_level_to_reg(adev, adev->mode_info.bl_level); - ret = dce_v8_0_hw_init(handle); + ret = dce_v8_0_hw_init(ip_block); /* turn on the BL */ if (adev->mode_info.bl_encoder) { @@ -2866,15 +2866,10 @@ static bool dce_v8_0_is_idle(void *handle) return true; } -static int dce_v8_0_wait_for_idle(void *handle) -{ - return 0; -} - -static int dce_v8_0_soft_reset(void *handle) +static int dce_v8_0_soft_reset(struct amdgpu_ip_block *ip_block) { u32 srbm_soft_reset = 0, tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (dce_v8_0_is_display_hung(adev)) srbm_soft_reset |= SRBM_SOFT_RESET__SOFT_RESET_DC_MASK; @@ -3232,7 +3227,6 @@ static int dce_v8_0_set_powergating_state(void *handle, static const struct amd_ip_funcs dce_v8_0_ip_funcs = { .name = "dce_v8_0", .early_init = dce_v8_0_early_init, - .late_init = NULL, .sw_init = dce_v8_0_sw_init, .sw_fini = dce_v8_0_sw_fini, .hw_init = dce_v8_0_hw_init, @@ -3240,12 +3234,9 @@ static const struct amd_ip_funcs dce_v8_0_ip_funcs = { .suspend = dce_v8_0_suspend, .resume = dce_v8_0_resume, .is_idle = dce_v8_0_is_idle, - .wait_for_idle = dce_v8_0_wait_for_idle, .soft_reset = dce_v8_0_soft_reset, .set_clockgating_state = dce_v8_0_set_clockgating_state, .set_powergating_state = dce_v8_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static void diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c index 45ed97038df0..9da95b25e158 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c @@ -3677,13 +3677,19 @@ static int gfx_v10_0_set_powergating_state(void *handle, enum amd_powergating_state state); static void gfx10_kiq_set_resources(struct amdgpu_ring *kiq_ring, uint64_t queue_mask) { + struct amdgpu_device *adev = kiq_ring->adev; + u64 shader_mc_addr; + + /* Cleaner shader MC address */ + shader_mc_addr = adev->gfx.cleaner_shader_gpu_addr >> 8; + amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_SET_RESOURCES, 6)); amdgpu_ring_write(kiq_ring, PACKET3_SET_RESOURCES_VMID_MASK(0) | PACKET3_SET_RESOURCES_QUEUE_TYPE(0)); /* vmid_mask:0 queue_type:0 (KIQ) */ amdgpu_ring_write(kiq_ring, lower_32_bits(queue_mask)); /* queue mask lo */ amdgpu_ring_write(kiq_ring, upper_32_bits(queue_mask)); /* queue mask hi */ - amdgpu_ring_write(kiq_ring, 0); /* gws mask lo */ - amdgpu_ring_write(kiq_ring, 0); /* gws mask hi */ + amdgpu_ring_write(kiq_ring, lower_32_bits(shader_mc_addr)); /* cleaner shader addr lo */ + amdgpu_ring_write(kiq_ring, upper_32_bits(shader_mc_addr)); /* cleaner shader addr hi */ amdgpu_ring_write(kiq_ring, 0); /* oac mask */ amdgpu_ring_write(kiq_ring, 0); /* gds heap base:0, gds heap size:0 */ } @@ -4683,11 +4689,11 @@ static void gfx_v10_0_alloc_ip_dump(struct amdgpu_device *adev) } } -static int gfx_v10_0_sw_init(void *handle) +static int gfx_v10_0_sw_init(struct amdgpu_ip_block *ip_block) { int i, j, k, r, ring_id = 0; int xcc_id = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { case IP_VERSION(10, 1, 10): @@ -4726,6 +4732,11 @@ static int gfx_v10_0_sw_init(void *handle) adev->gfx.mec.num_queue_per_pipe = 8; break; } + switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { + default: + adev->gfx.enable_cleaner_shader = false; + break; + } /* KIQ event */ r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_GRBM_CP, @@ -4842,6 +4853,9 @@ static int gfx_v10_0_sw_init(void *handle) gfx_v10_0_alloc_ip_dump(adev); + r = amdgpu_gfx_sysfs_isolation_shader_init(adev); + if (r) + return r; return 0; } @@ -4866,10 +4880,10 @@ static void gfx_v10_0_me_fini(struct amdgpu_device *adev) (void **)&adev->gfx.me.me_fw_ptr); } -static int gfx_v10_0_sw_fini(void *handle) +static int gfx_v10_0_sw_fini(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->gfx.num_gfx_rings; i++) amdgpu_ring_fini(&adev->gfx.gfx_ring[i]); @@ -4881,6 +4895,8 @@ static int gfx_v10_0_sw_fini(void *handle) amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring); amdgpu_gfx_kiq_fini(adev, 0); + amdgpu_gfx_cleaner_shader_sw_fini(adev); + gfx_v10_0_pfp_fini(adev); gfx_v10_0_ce_fini(adev); gfx_v10_0_me_fini(adev); @@ -4891,6 +4907,7 @@ static int gfx_v10_0_sw_fini(void *handle) gfx_v10_0_rlc_backdoor_autoload_buffer_fini(adev); gfx_v10_0_free_microcode(adev); + amdgpu_gfx_sysfs_isolation_shader_fini(adev); kfree(adev->gfx.ip_dump_core); kfree(adev->gfx.ip_dump_compute_queues); @@ -6374,7 +6391,7 @@ static int gfx_v10_0_cp_gfx_resume(struct amdgpu_device *adev) WREG32_SOC15(GC, 0, mmCP_RB0_WPTR, lower_32_bits(ring->wptr)); WREG32_SOC15(GC, 0, mmCP_RB0_WPTR_HI, upper_32_bits(ring->wptr)); - /* set the wb address wether it's enabled or not */ + /* set the wb address whether it's enabled or not */ rptr_addr = ring->rptr_gpu_addr; WREG32_SOC15(GC, 0, mmCP_RB0_RPTR_ADDR, lower_32_bits(rptr_addr)); WREG32_SOC15(GC, 0, mmCP_RB0_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & @@ -6412,7 +6429,7 @@ static int gfx_v10_0_cp_gfx_resume(struct amdgpu_device *adev) ring->wptr = 0; WREG32_SOC15(GC, 0, mmCP_RB1_WPTR, lower_32_bits(ring->wptr)); WREG32_SOC15(GC, 0, mmCP_RB1_WPTR_HI, upper_32_bits(ring->wptr)); - /* Set the wb address wether it's enabled or not */ + /* Set the wb address whether it's enabled or not */ rptr_addr = ring->rptr_gpu_addr; WREG32_SOC15(GC, 0, mmCP_RB1_RPTR_ADDR, lower_32_bits(rptr_addr)); WREG32_SOC15(GC, 0, mmCP_RB1_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & @@ -7366,14 +7383,17 @@ static void gfx_v10_0_disable_gpa_mode(struct amdgpu_device *adev) WREG32_SOC15(GC, 0, mmCPG_PSP_DEBUG, data); } -static int gfx_v10_0_hw_init(void *handle) +static int gfx_v10_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (!amdgpu_emu_mode) gfx_v10_0_init_golden_registers(adev); + amdgpu_gfx_cleaner_shader_init(adev, adev->gfx.cleaner_shader_size, + adev->gfx.cleaner_shader_ptr); + if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) { /** * For gfx 10, rlc firmware loading relies on smu firmware is @@ -7418,9 +7438,9 @@ static int gfx_v10_0_hw_init(void *handle) return r; } -static int gfx_v10_0_hw_fini(void *handle) +static int gfx_v10_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0); amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0); @@ -7431,7 +7451,7 @@ static int gfx_v10_0_hw_fini(void *handle) * otherwise the gfxoff disallowing will be failed to set. */ if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(10, 3, 1)) - gfx_v10_0_set_powergating_state(handle, AMD_PG_STATE_UNGATE); + gfx_v10_0_set_powergating_state(ip_block->adev, AMD_PG_STATE_UNGATE); if (!adev->no_hw_access) { if (amdgpu_async_gfx_ring) { @@ -7456,14 +7476,14 @@ static int gfx_v10_0_hw_fini(void *handle) return 0; } -static int gfx_v10_0_suspend(void *handle) +static int gfx_v10_0_suspend(struct amdgpu_ip_block *ip_block) { - return gfx_v10_0_hw_fini(handle); + return gfx_v10_0_hw_fini(ip_block); } -static int gfx_v10_0_resume(void *handle) +static int gfx_v10_0_resume(struct amdgpu_ip_block *ip_block) { - return gfx_v10_0_hw_init(handle); + return gfx_v10_0_hw_init(ip_block); } static bool gfx_v10_0_is_idle(void *handle) @@ -7477,11 +7497,11 @@ static bool gfx_v10_0_is_idle(void *handle) return true; } -static int gfx_v10_0_wait_for_idle(void *handle) +static int gfx_v10_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned int i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { /* read MC_STATUS */ @@ -7495,11 +7515,11 @@ static int gfx_v10_0_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int gfx_v10_0_soft_reset(void *handle) +static int gfx_v10_0_soft_reset(struct amdgpu_ip_block *ip_block) { u32 grbm_soft_reset = 0; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* GRBM_STATUS */ tmp = RREG32_SOC15(GC, 0, mmGRBM_STATUS); @@ -7678,9 +7698,9 @@ static void gfx_v10_0_ring_emit_gds_switch(struct amdgpu_ring *ring, (1 << (oa_size + oa_base)) - (1 << oa_base)); } -static int gfx_v10_0_early_init(void *handle) +static int gfx_v10_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->gfx.funcs = &gfx_v10_0_gfx_funcs; @@ -7722,9 +7742,9 @@ static int gfx_v10_0_early_init(void *handle) return gfx_v10_0_init_microcode(adev); } -static int gfx_v10_0_late_init(void *handle) +static int gfx_v10_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_irq_get(adev, &adev->gfx.priv_reg_irq, 0); @@ -9402,8 +9422,6 @@ static void gfx_v10_0_emit_mem_sync(struct amdgpu_ring *ring) static void gfx_v10_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) { - int i; - /* Header itself is a NOP packet */ if (num_nop == 1) { amdgpu_ring_write(ring, ring->funcs->nop); @@ -9414,8 +9432,7 @@ static void gfx_v10_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) amdgpu_ring_write(ring, PACKET3(PACKET3_NOP, min(num_nop - 2, 0x3ffe))); /* Header is at index 0, followed by num_nops - 1 NOP packet's */ - for (i = 1; i < num_nop; i++) - amdgpu_ring_write(ring, ring->funcs->nop); + amdgpu_ring_insert_nop(ring, num_nop - 1); } static int gfx_v10_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) @@ -9568,9 +9585,9 @@ static int gfx_v10_0_reset_kcq(struct amdgpu_ring *ring, return amdgpu_ring_test_ring(ring); } -static void gfx_v10_ip_print(void *handle, struct drm_printer *p) +static void gfx_v10_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t i, j, k, reg, index = 0; uint32_t reg_count = ARRAY_SIZE(gc_reg_list_10_1); @@ -9632,9 +9649,9 @@ static void gfx_v10_ip_print(void *handle, struct drm_printer *p) } } -static void gfx_v10_ip_dump(void *handle) +static void gfx_v10_ip_dump(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t i, j, k, reg, index = 0; uint32_t reg_count = ARRAY_SIZE(gc_reg_list_10_1); @@ -9699,6 +9716,13 @@ static void gfx_v10_ip_dump(void *handle) amdgpu_gfx_off_ctrl(adev, true); } +static void gfx_v10_0_ring_emit_cleaner_shader(struct amdgpu_ring *ring) +{ + /* Emit the cleaner shader */ + amdgpu_ring_write(ring, PACKET3(PACKET3_RUN_CLEANER_SHADER, 0)); + amdgpu_ring_write(ring, 0); /* RESERVED field, programmed to zero */ +} + static const struct amd_ip_funcs gfx_v10_0_ip_funcs = { .name = "gfx_v10_0", .early_init = gfx_v10_0_early_init, @@ -9749,7 +9773,8 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = { 5 + /* HDP_INVL */ 8 + 8 + /* FENCE x2 */ 2 + /* SWITCH_BUFFER */ - 8, /* gfx_v10_0_emit_mem_sync */ + 8 + /* gfx_v10_0_emit_mem_sync */ + 2, /* gfx_v10_0_ring_emit_cleaner_shader */ .emit_ib_size = 4, /* gfx_v10_0_ring_emit_ib_gfx */ .emit_ib = gfx_v10_0_ring_emit_ib_gfx, .emit_fence = gfx_v10_0_ring_emit_fence, @@ -9772,6 +9797,9 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_gfx = { .soft_recovery = gfx_v10_0_ring_soft_recovery, .emit_mem_sync = gfx_v10_0_emit_mem_sync, .reset = gfx_v10_0_reset_kgq, + .emit_cleaner_shader = gfx_v10_0_ring_emit_cleaner_shader, + .begin_use = amdgpu_gfx_enforce_isolation_ring_begin_use, + .end_use = amdgpu_gfx_enforce_isolation_ring_end_use, }; static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_compute = { @@ -9791,7 +9819,8 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_compute = { SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + 2 + /* gfx_v10_0_ring_emit_vm_flush */ 8 + 8 + 8 + /* gfx_v10_0_ring_emit_fence x3 for user fence, vm fence */ - 8, /* gfx_v10_0_emit_mem_sync */ + 8 + /* gfx_v10_0_emit_mem_sync */ + 2, /* gfx_v10_0_ring_emit_cleaner_shader */ .emit_ib_size = 7, /* gfx_v10_0_ring_emit_ib_compute */ .emit_ib = gfx_v10_0_ring_emit_ib_compute, .emit_fence = gfx_v10_0_ring_emit_fence, @@ -9809,6 +9838,9 @@ static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_compute = { .soft_recovery = gfx_v10_0_ring_soft_recovery, .emit_mem_sync = gfx_v10_0_emit_mem_sync, .reset = gfx_v10_0_reset_kcq, + .emit_cleaner_shader = gfx_v10_0_ring_emit_cleaner_shader, + .begin_use = amdgpu_gfx_enforce_isolation_ring_begin_use, + .end_use = amdgpu_gfx_enforce_isolation_ring_end_use, }; static const struct amdgpu_ring_funcs gfx_v10_0_ring_funcs_kiq = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c index d3e8be82a172..894fc04201c3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c @@ -293,14 +293,20 @@ static void gfx_v11_0_update_perf_clk(struct amdgpu_device *adev, static void gfx11_kiq_set_resources(struct amdgpu_ring *kiq_ring, uint64_t queue_mask) { + struct amdgpu_device *adev = kiq_ring->adev; + u64 shader_mc_addr; + + /* Cleaner shader MC address */ + shader_mc_addr = adev->gfx.cleaner_shader_gpu_addr >> 8; + amdgpu_ring_write(kiq_ring, PACKET3(PACKET3_SET_RESOURCES, 6)); amdgpu_ring_write(kiq_ring, PACKET3_SET_RESOURCES_VMID_MASK(0) | PACKET3_SET_RESOURCES_UNMAP_LATENTY(0xa) | /* unmap_latency: 0xa (~ 1s) */ PACKET3_SET_RESOURCES_QUEUE_TYPE(0)); /* vmid_mask:0 queue_type:0 (KIQ) */ amdgpu_ring_write(kiq_ring, lower_32_bits(queue_mask)); /* queue mask lo */ amdgpu_ring_write(kiq_ring, upper_32_bits(queue_mask)); /* queue mask hi */ - amdgpu_ring_write(kiq_ring, 0); /* gws mask lo */ - amdgpu_ring_write(kiq_ring, 0); /* gws mask hi */ + amdgpu_ring_write(kiq_ring, lower_32_bits(shader_mc_addr)); /* cleaner shader addr lo */ + amdgpu_ring_write(kiq_ring, upper_32_bits(shader_mc_addr)); /* cleaner shader addr hi */ amdgpu_ring_write(kiq_ring, 0); /* oac mask */ amdgpu_ring_write(kiq_ring, 0); /* gds heap base:0, gds heap size:0 */ } @@ -483,8 +489,6 @@ static void gfx_v11_0_wait_reg_mem(struct amdgpu_ring *ring, int eng_sel, static void gfx_v11_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) { - int i; - /* Header itself is a NOP packet */ if (num_nop == 1) { amdgpu_ring_write(ring, ring->funcs->nop); @@ -495,8 +499,7 @@ static void gfx_v11_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) amdgpu_ring_write(ring, PACKET3(PACKET3_NOP, min(num_nop - 2, 0x3ffe))); /* Header is at index 0, followed by num_nops - 1 NOP packet's */ - for (i = 1; i < num_nop; i++) - amdgpu_ring_write(ring, ring->funcs->nop); + amdgpu_ring_insert_nop(ring, num_nop - 1); } static int gfx_v11_0_ring_test_ring(struct amdgpu_ring *ring) @@ -1536,11 +1539,11 @@ static void gfx_v11_0_alloc_ip_dump(struct amdgpu_device *adev) } } -static int gfx_v11_0_sw_init(void *handle) +static int gfx_v11_0_sw_init(struct amdgpu_ip_block *ip_block) { int i, j, k, r, ring_id = 0; int xcc_id = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { case IP_VERSION(11, 0, 0): @@ -1575,6 +1578,11 @@ static int gfx_v11_0_sw_init(void *handle) break; } + switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { + default: + adev->gfx.enable_cleaner_shader = false; + } + /* Enable CG flag in one VF mode for enabling RLC safe mode enter/exit */ if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(11, 0, 3) && amdgpu_sriov_is_pp_one_vf(adev)) @@ -1700,6 +1708,10 @@ static int gfx_v11_0_sw_init(void *handle) gfx_v11_0_alloc_ip_dump(adev); + r = amdgpu_gfx_sysfs_isolation_shader_init(adev); + if (r) + return r; + return 0; } @@ -1732,10 +1744,10 @@ static void gfx_v11_0_rlc_autoload_buffer_fini(struct amdgpu_device *adev) (void **)&adev->gfx.rlc.rlc_autoload_ptr); } -static int gfx_v11_0_sw_fini(void *handle) +static int gfx_v11_0_sw_fini(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->gfx.num_gfx_rings; i++) amdgpu_ring_fini(&adev->gfx.gfx_ring[i]); @@ -1749,6 +1761,8 @@ static int gfx_v11_0_sw_fini(void *handle) amdgpu_gfx_kiq_fini(adev, 0); } + amdgpu_gfx_cleaner_shader_sw_fini(adev); + gfx_v11_0_pfp_fini(adev); gfx_v11_0_me_fini(adev); gfx_v11_0_rlc_fini(adev); @@ -1759,6 +1773,8 @@ static int gfx_v11_0_sw_fini(void *handle) gfx_v11_0_free_microcode(adev); + amdgpu_gfx_sysfs_isolation_shader_fini(adev); + kfree(adev->gfx.ip_dump_core); kfree(adev->gfx.ip_dump_compute_queues); kfree(adev->gfx.ip_dump_gfx_queues); @@ -1893,8 +1909,10 @@ static void gfx_v11_0_init_compute_vmid(struct amdgpu_device *adev) soc21_grbm_select(adev, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); - /* Initialize all compute VMIDs to have no GDS, GWS, or OA - acccess. These should be enabled by FW for target VMIDs. */ + /* + * Initialize all compute VMIDs to have no GDS, GWS, or OA + * access. These should be enabled by FW for target VMIDs. + */ for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) { WREG32_SOC15_OFFSET(GC, 0, regGDS_VMID0_BASE, 2 * i, 0); WREG32_SOC15_OFFSET(GC, 0, regGDS_VMID0_SIZE, 2 * i, 0); @@ -3555,7 +3573,7 @@ static int gfx_v11_0_cp_gfx_resume(struct amdgpu_device *adev) WREG32_SOC15(GC, 0, regCP_RB0_WPTR, lower_32_bits(ring->wptr)); WREG32_SOC15(GC, 0, regCP_RB0_WPTR_HI, upper_32_bits(ring->wptr)); - /* set the wb address wether it's enabled or not */ + /* set the wb address whether it's enabled or not */ rptr_addr = ring->rptr_gpu_addr; WREG32_SOC15(GC, 0, regCP_RB0_RPTR_ADDR, lower_32_bits(rptr_addr)); WREG32_SOC15(GC, 0, regCP_RB0_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & @@ -3593,7 +3611,7 @@ static int gfx_v11_0_cp_gfx_resume(struct amdgpu_device *adev) ring->wptr = 0; WREG32_SOC15(GC, 0, regCP_RB1_WPTR, lower_32_bits(ring->wptr)); WREG32_SOC15(GC, 0, regCP_RB1_WPTR_HI, upper_32_bits(ring->wptr)); - /* Set the wb address wether it's enabled or not */ + /* Set the wb address whether it's enabled or not */ rptr_addr = ring->rptr_gpu_addr; WREG32_SOC15(GC, 0, regCP_RB1_RPTR_ADDR, lower_32_bits(rptr_addr)); WREG32_SOC15(GC, 0, regCP_RB1_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & @@ -4568,10 +4586,13 @@ static void gfx_v11_0_disable_gpa_mode(struct amdgpu_device *adev) WREG32_SOC15(GC, 0, regCPG_PSP_DEBUG, data); } -static int gfx_v11_0_hw_init(void *handle) +static int gfx_v11_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; + + amdgpu_gfx_cleaner_shader_init(adev, adev->gfx.cleaner_shader_size, + adev->gfx.cleaner_shader_ptr); if (adev->firmware.load_type == AMDGPU_FW_LOAD_RLC_BACKDOOR_AUTO) { if (adev->gfx.imu.funcs) { @@ -4665,9 +4686,9 @@ static int gfx_v11_0_hw_init(void *handle) return r; } -static int gfx_v11_0_hw_fini(void *handle) +static int gfx_v11_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0); amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0); @@ -4703,14 +4724,14 @@ static int gfx_v11_0_hw_fini(void *handle) return 0; } -static int gfx_v11_0_suspend(void *handle) +static int gfx_v11_0_suspend(struct amdgpu_ip_block *ip_block) { - return gfx_v11_0_hw_fini(handle); + return gfx_v11_0_hw_fini(ip_block); } -static int gfx_v11_0_resume(void *handle) +static int gfx_v11_0_resume(struct amdgpu_ip_block *ip_block) { - return gfx_v11_0_hw_init(handle); + return gfx_v11_0_hw_init(ip_block); } static bool gfx_v11_0_is_idle(void *handle) @@ -4724,11 +4745,11 @@ static bool gfx_v11_0_is_idle(void *handle) return true; } -static int gfx_v11_0_wait_for_idle(void *handle) +static int gfx_v11_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { /* read MC_STATUS */ @@ -4774,12 +4795,12 @@ int gfx_v11_0_request_gfx_index_mutex(struct amdgpu_device *adev, return 0; } -static int gfx_v11_0_soft_reset(void *handle) +static int gfx_v11_0_soft_reset(struct amdgpu_ip_block *ip_block) { u32 grbm_soft_reset = 0; u32 tmp; int r, i, j, k; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_gfx_rlc_enter_safe_mode(adev, 0); @@ -4905,10 +4926,10 @@ static int gfx_v11_0_soft_reset(void *handle) return gfx_v11_0_cp_resume(adev); } -static bool gfx_v11_0_check_soft_reset(void *handle) +static bool gfx_v11_0_check_soft_reset(struct amdgpu_ip_block *ip_block) { int i, r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; long tmo = msecs_to_jiffies(1000); @@ -4929,12 +4950,13 @@ static bool gfx_v11_0_check_soft_reset(void *handle) return false; } -static int gfx_v11_0_post_soft_reset(void *handle) +static int gfx_v11_0_post_soft_reset(struct amdgpu_ip_block *ip_block) { + struct amdgpu_device *adev = ip_block->adev; /** * GFX soft reset will impact MES, need resume MES when do GFX soft reset */ - return amdgpu_mes_resume((struct amdgpu_device *)handle); + return amdgpu_mes_resume(adev); } static uint64_t gfx_v11_0_get_gpu_clock_counter(struct amdgpu_device *adev) @@ -4995,9 +5017,9 @@ static void gfx_v11_0_ring_emit_gds_switch(struct amdgpu_ring *ring, (1 << (oa_size + oa_base)) - (1 << oa_base)); } -static int gfx_v11_0_early_init(void *handle) +static int gfx_v11_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->gfx.funcs = &gfx_v11_0_gfx_funcs; @@ -5018,9 +5040,9 @@ static int gfx_v11_0_early_init(void *handle) return gfx_v11_0_init_microcode(adev); } -static int gfx_v11_0_late_init(void *handle) +static int gfx_v11_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_irq_get(adev, &adev->gfx.priv_reg_irq, 0); @@ -6639,9 +6661,9 @@ static int gfx_v11_0_reset_kcq(struct amdgpu_ring *ring, unsigned int vmid) return amdgpu_ring_test_ring(ring); } -static void gfx_v11_ip_print(void *handle, struct drm_printer *p) +static void gfx_v11_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t i, j, k, reg, index = 0; uint32_t reg_count = ARRAY_SIZE(gc_reg_list_11_0); @@ -6703,9 +6725,9 @@ static void gfx_v11_ip_print(void *handle, struct drm_printer *p) } } -static void gfx_v11_ip_dump(void *handle) +static void gfx_v11_ip_dump(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t i, j, k, reg, index = 0; uint32_t reg_count = ARRAY_SIZE(gc_reg_list_11_0); @@ -6769,6 +6791,13 @@ static void gfx_v11_ip_dump(void *handle) amdgpu_gfx_off_ctrl(adev, true); } +static void gfx_v11_0_ring_emit_cleaner_shader(struct amdgpu_ring *ring) +{ + /* Emit the cleaner shader */ + amdgpu_ring_write(ring, PACKET3(PACKET3_RUN_CLEANER_SHADER, 0)); + amdgpu_ring_write(ring, 0); /* RESERVED field, programmed to zero */ +} + static const struct amd_ip_funcs gfx_v11_0_ip_funcs = { .name = "gfx_v11_0", .early_init = gfx_v11_0_early_init, @@ -6818,7 +6847,8 @@ static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_gfx = { 5 + /* HDP_INVL */ 22 + /* SET_Q_PREEMPTION_MODE */ 8 + 8 + /* FENCE x2 */ - 8, /* gfx_v11_0_emit_mem_sync */ + 8 + /* gfx_v11_0_emit_mem_sync */ + 2, /* gfx_v11_0_ring_emit_cleaner_shader */ .emit_ib_size = 4, /* gfx_v11_0_ring_emit_ib_gfx */ .emit_ib = gfx_v11_0_ring_emit_ib_gfx, .emit_fence = gfx_v11_0_ring_emit_fence, @@ -6841,6 +6871,9 @@ static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_gfx = { .soft_recovery = gfx_v11_0_ring_soft_recovery, .emit_mem_sync = gfx_v11_0_emit_mem_sync, .reset = gfx_v11_0_reset_kgq, + .emit_cleaner_shader = gfx_v11_0_ring_emit_cleaner_shader, + .begin_use = amdgpu_gfx_enforce_isolation_ring_begin_use, + .end_use = amdgpu_gfx_enforce_isolation_ring_end_use, }; static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_compute = { @@ -6861,7 +6894,8 @@ static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_compute = { SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + 2 + /* gfx_v11_0_ring_emit_vm_flush */ 8 + 8 + 8 + /* gfx_v11_0_ring_emit_fence x3 for user fence, vm fence */ - 8, /* gfx_v11_0_emit_mem_sync */ + 8 + /* gfx_v11_0_emit_mem_sync */ + 2, /* gfx_v11_0_ring_emit_cleaner_shader */ .emit_ib_size = 7, /* gfx_v11_0_ring_emit_ib_compute */ .emit_ib = gfx_v11_0_ring_emit_ib_compute, .emit_fence = gfx_v11_0_ring_emit_fence, @@ -6879,6 +6913,9 @@ static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_compute = { .soft_recovery = gfx_v11_0_ring_soft_recovery, .emit_mem_sync = gfx_v11_0_emit_mem_sync, .reset = gfx_v11_0_reset_kcq, + .emit_cleaner_shader = gfx_v11_0_ring_emit_cleaner_shader, + .begin_use = amdgpu_gfx_enforce_isolation_ring_begin_use, + .end_use = amdgpu_gfx_enforce_isolation_ring_end_use, }; static const struct amdgpu_ring_funcs gfx_v11_0_ring_funcs_kiq = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c index 47b47d21f464..9fec28d8a5fc 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v12_0.c @@ -1319,12 +1319,12 @@ static void gfx_v12_0_alloc_ip_dump(struct amdgpu_device *adev) } } -static int gfx_v12_0_sw_init(void *handle) +static int gfx_v12_0_sw_init(struct amdgpu_ip_block *ip_block) { int i, j, k, r, ring_id = 0; unsigned num_compute_rings; int xcc_id = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { case IP_VERSION(12, 0, 0): @@ -1346,6 +1346,12 @@ static int gfx_v12_0_sw_init(void *handle) break; } + switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { + default: + adev->gfx.enable_cleaner_shader = false; + break; + } + /* recalculate compute rings to use based on hardware configuration */ num_compute_rings = (adev->gfx.mec.num_pipe_per_mec * adev->gfx.mec.num_queue_per_pipe) / 2; @@ -1460,6 +1466,10 @@ static int gfx_v12_0_sw_init(void *handle) gfx_v12_0_alloc_ip_dump(adev); + r = amdgpu_gfx_sysfs_isolation_shader_init(adev); + if (r) + return r; + return 0; } @@ -1492,10 +1502,10 @@ static void gfx_v12_0_rlc_autoload_buffer_fini(struct amdgpu_device *adev) (void **)&adev->gfx.rlc.rlc_autoload_ptr); } -static int gfx_v12_0_sw_fini(void *handle) +static int gfx_v12_0_sw_fini(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->gfx.num_gfx_rings; i++) amdgpu_ring_fini(&adev->gfx.gfx_ring[i]); @@ -1519,6 +1529,8 @@ static int gfx_v12_0_sw_fini(void *handle) gfx_v12_0_free_microcode(adev); + amdgpu_gfx_sysfs_isolation_shader_fini(adev); + kfree(adev->gfx.ip_dump_core); kfree(adev->gfx.ip_dump_compute_queues); kfree(adev->gfx.ip_dump_gfx_queues); @@ -2601,7 +2613,7 @@ static int gfx_v12_0_cp_gfx_resume(struct amdgpu_device *adev) WREG32_SOC15(GC, 0, regCP_RB0_WPTR, lower_32_bits(ring->wptr)); WREG32_SOC15(GC, 0, regCP_RB0_WPTR_HI, upper_32_bits(ring->wptr)); - /* set the wb address wether it's enabled or not */ + /* set the wb address whether it's enabled or not */ rptr_addr = ring->rptr_gpu_addr; WREG32_SOC15(GC, 0, regCP_RB0_RPTR_ADDR, lower_32_bits(rptr_addr)); WREG32_SOC15(GC, 0, regCP_RB0_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & @@ -3513,10 +3525,10 @@ static void gfx_v12_0_init_golden_registers(struct amdgpu_device *adev) } } -static int gfx_v12_0_hw_init(void *handle) +static int gfx_v12_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->firmware.load_type == AMDGPU_FW_LOAD_RLC_BACKDOOR_AUTO) { if (adev->gfx.imu.funcs && (amdgpu_dpm > 0)) { @@ -3603,9 +3615,9 @@ static int gfx_v12_0_hw_init(void *handle) return r; } -static int gfx_v12_0_hw_fini(void *handle) +static int gfx_v12_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t tmp; amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0); @@ -3643,14 +3655,14 @@ static int gfx_v12_0_hw_fini(void *handle) return 0; } -static int gfx_v12_0_suspend(void *handle) +static int gfx_v12_0_suspend(struct amdgpu_ip_block *ip_block) { - return gfx_v12_0_hw_fini(handle); + return gfx_v12_0_hw_fini(ip_block); } -static int gfx_v12_0_resume(void *handle) +static int gfx_v12_0_resume(struct amdgpu_ip_block *ip_block) { - return gfx_v12_0_hw_init(handle); + return gfx_v12_0_hw_init(ip_block); } static bool gfx_v12_0_is_idle(void *handle) @@ -3664,11 +3676,11 @@ static bool gfx_v12_0_is_idle(void *handle) return true; } -static int gfx_v12_0_wait_for_idle(void *handle) +static int gfx_v12_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { /* read MC_STATUS */ @@ -3695,9 +3707,9 @@ static uint64_t gfx_v12_0_get_gpu_clock_counter(struct amdgpu_device *adev) return clock; } -static int gfx_v12_0_early_init(void *handle) +static int gfx_v12_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->gfx.funcs = &gfx_v12_0_gfx_funcs; @@ -3717,9 +3729,9 @@ static int gfx_v12_0_early_init(void *handle) return gfx_v12_0_init_microcode(adev); } -static int gfx_v12_0_late_init(void *handle) +static int gfx_v12_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_irq_get(adev, &adev->gfx.priv_reg_irq, 0); @@ -5022,8 +5034,6 @@ static void gfx_v12_0_emit_mem_sync(struct amdgpu_ring *ring) static void gfx_v12_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) { - int i; - /* Header itself is a NOP packet */ if (num_nop == 1) { amdgpu_ring_write(ring, ring->funcs->nop); @@ -5034,13 +5044,19 @@ static void gfx_v12_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) amdgpu_ring_write(ring, PACKET3(PACKET3_NOP, min(num_nop - 2, 0x3ffe))); /* Header is at index 0, followed by num_nops - 1 NOP packet's */ - for (i = 1; i < num_nop; i++) - amdgpu_ring_write(ring, ring->funcs->nop); + amdgpu_ring_insert_nop(ring, num_nop - 1); } -static void gfx_v12_ip_print(void *handle, struct drm_printer *p) +static void gfx_v12_0_ring_emit_cleaner_shader(struct amdgpu_ring *ring) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + /* Emit the cleaner shader */ + amdgpu_ring_write(ring, PACKET3(PACKET3_RUN_CLEANER_SHADER, 0)); + amdgpu_ring_write(ring, 0); /* RESERVED field, programmed to zero */ +} + +static void gfx_v12_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer *p) +{ + struct amdgpu_device *adev = ip_block->adev; uint32_t i, j, k, reg, index = 0; uint32_t reg_count = ARRAY_SIZE(gc_reg_list_12_0); @@ -5102,9 +5118,9 @@ static void gfx_v12_ip_print(void *handle, struct drm_printer *p) } } -static void gfx_v12_ip_dump(void *handle) +static void gfx_v12_ip_dump(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t i, j, k, reg, index = 0; uint32_t reg_count = ARRAY_SIZE(gc_reg_list_12_0); @@ -5297,7 +5313,8 @@ static const struct amdgpu_ring_funcs gfx_v12_0_ring_funcs_gfx = { 3 + /* CNTX_CTRL */ 5 + /* HDP_INVL */ 8 + 8 + /* FENCE x2 */ - 8, /* gfx_v12_0_emit_mem_sync */ + 8 + /* gfx_v12_0_emit_mem_sync */ + 2, /* gfx_v12_0_ring_emit_cleaner_shader */ .emit_ib_size = 4, /* gfx_v12_0_ring_emit_ib_gfx */ .emit_ib = gfx_v12_0_ring_emit_ib_gfx, .emit_fence = gfx_v12_0_ring_emit_fence, @@ -5318,6 +5335,9 @@ static const struct amdgpu_ring_funcs gfx_v12_0_ring_funcs_gfx = { .soft_recovery = gfx_v12_0_ring_soft_recovery, .emit_mem_sync = gfx_v12_0_emit_mem_sync, .reset = gfx_v12_0_reset_kgq, + .emit_cleaner_shader = gfx_v12_0_ring_emit_cleaner_shader, + .begin_use = amdgpu_gfx_enforce_isolation_ring_begin_use, + .end_use = amdgpu_gfx_enforce_isolation_ring_end_use, }; static const struct amdgpu_ring_funcs gfx_v12_0_ring_funcs_compute = { @@ -5336,7 +5356,8 @@ static const struct amdgpu_ring_funcs gfx_v12_0_ring_funcs_compute = { SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 7 + 2 + /* gfx_v12_0_ring_emit_vm_flush */ 8 + 8 + 8 + /* gfx_v12_0_ring_emit_fence x3 for user fence, vm fence */ - 8, /* gfx_v12_0_emit_mem_sync */ + 8 + /* gfx_v12_0_emit_mem_sync */ + 2, /* gfx_v12_0_ring_emit_cleaner_shader */ .emit_ib_size = 7, /* gfx_v12_0_ring_emit_ib_compute */ .emit_ib = gfx_v12_0_ring_emit_ib_compute, .emit_fence = gfx_v12_0_ring_emit_fence, @@ -5353,6 +5374,9 @@ static const struct amdgpu_ring_funcs gfx_v12_0_ring_funcs_compute = { .soft_recovery = gfx_v12_0_ring_soft_recovery, .emit_mem_sync = gfx_v12_0_emit_mem_sync, .reset = gfx_v12_0_reset_kcq, + .emit_cleaner_shader = gfx_v12_0_ring_emit_cleaner_shader, + .begin_use = amdgpu_gfx_enforce_isolation_ring_begin_use, + .end_use = amdgpu_gfx_enforce_isolation_ring_end_use, }; static const struct amdgpu_ring_funcs gfx_v12_0_ring_funcs_kiq = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c index 564f0b9336b6..41f50bf380c4 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c @@ -3023,9 +3023,9 @@ static const struct amdgpu_rlc_funcs gfx_v6_0_rlc_funcs = { .start = gfx_v6_0_rlc_start }; -static int gfx_v6_0_early_init(void *handle) +static int gfx_v6_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->gfx.xcc_mask = 1; adev->gfx.num_gfx_rings = GFX6_NUM_GFX_RINGS; @@ -3039,10 +3039,10 @@ static int gfx_v6_0_early_init(void *handle) return 0; } -static int gfx_v6_0_sw_init(void *handle) +static int gfx_v6_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, r; r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, 181, &adev->gfx.eop_irq); @@ -3107,10 +3107,10 @@ static int gfx_v6_0_sw_init(void *handle) return r; } -static int gfx_v6_0_sw_fini(void *handle) +static int gfx_v6_0_sw_fini(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->gfx.num_gfx_rings; i++) amdgpu_ring_fini(&adev->gfx.gfx_ring[i]); @@ -3122,10 +3122,10 @@ static int gfx_v6_0_sw_fini(void *handle) return 0; } -static int gfx_v6_0_hw_init(void *handle) +static int gfx_v6_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gfx_v6_0_constants_init(adev); @@ -3142,9 +3142,9 @@ static int gfx_v6_0_hw_init(void *handle) return r; } -static int gfx_v6_0_hw_fini(void *handle) +static int gfx_v6_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gfx_v6_0_cp_enable(adev, false); adev->gfx.rlc.funcs->stop(adev); @@ -3153,18 +3153,14 @@ static int gfx_v6_0_hw_fini(void *handle) return 0; } -static int gfx_v6_0_suspend(void *handle) +static int gfx_v6_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return gfx_v6_0_hw_fini(adev); + return gfx_v6_0_hw_fini(ip_block); } -static int gfx_v6_0_resume(void *handle) +static int gfx_v6_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return gfx_v6_0_hw_init(adev); + return gfx_v6_0_hw_init(ip_block); } static bool gfx_v6_0_is_idle(void *handle) @@ -3177,24 +3173,19 @@ static bool gfx_v6_0_is_idle(void *handle) return true; } -static int gfx_v6_0_wait_for_idle(void *handle) +static int gfx_v6_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { - if (gfx_v6_0_is_idle(handle)) + if (gfx_v6_0_is_idle(adev)) return 0; udelay(1); } return -ETIMEDOUT; } -static int gfx_v6_0_soft_reset(void *handle) -{ - return 0; -} - static void gfx_v6_0_set_gfx_eop_interrupt_state(struct amdgpu_device *adev, enum amdgpu_interrupt_state state) { @@ -3444,7 +3435,6 @@ static void gfx_v6_0_emit_mem_sync(struct amdgpu_ring *ring) static const struct amd_ip_funcs gfx_v6_0_ip_funcs = { .name = "gfx_v6_0", .early_init = gfx_v6_0_early_init, - .late_init = NULL, .sw_init = gfx_v6_0_sw_init, .sw_fini = gfx_v6_0_sw_fini, .hw_init = gfx_v6_0_hw_init, @@ -3453,11 +3443,8 @@ static const struct amd_ip_funcs gfx_v6_0_ip_funcs = { .resume = gfx_v6_0_resume, .is_idle = gfx_v6_0_is_idle, .wait_for_idle = gfx_v6_0_wait_for_idle, - .soft_reset = gfx_v6_0_soft_reset, .set_clockgating_state = gfx_v6_0_set_clockgating_state, .set_powergating_state = gfx_v6_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs gfx_v6_0_ring_funcs_gfx = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index f146806c4633..824d5913103b 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -2559,7 +2559,7 @@ static int gfx_v7_0_cp_gfx_resume(struct amdgpu_device *adev) ring->wptr = 0; WREG32(mmCP_RB0_WPTR, lower_32_bits(ring->wptr)); - /* set the wb address wether it's enabled or not */ + /* set the wb address whether it's enabled or not */ rptr_addr = ring->rptr_gpu_addr; WREG32(mmCP_RB0_RPTR_ADDR, lower_32_bits(rptr_addr)); WREG32(mmCP_RB0_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & 0xFF); @@ -2876,7 +2876,7 @@ static void gfx_v7_0_mqd_init(struct amdgpu_device *adev, mqd->cp_hqd_pq_wptr_poll_addr_lo = wb_gpu_addr & 0xfffffffc; mqd->cp_hqd_pq_wptr_poll_addr_hi = upper_32_bits(wb_gpu_addr) & 0xffff; - /* set the wb address wether it's enabled or not */ + /* set the wb address whether it's enabled or not */ wb_gpu_addr = ring->rptr_gpu_addr; mqd->cp_hqd_pq_rptr_report_addr_lo = wb_gpu_addr & 0xfffffffc; mqd->cp_hqd_pq_rptr_report_addr_hi = @@ -4134,9 +4134,9 @@ static const struct amdgpu_rlc_funcs gfx_v7_0_rlc_funcs = { .update_spm_vmid = gfx_v7_0_update_spm_vmid }; -static int gfx_v7_0_early_init(void *handle) +static int gfx_v7_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->gfx.xcc_mask = 1; adev->gfx.num_gfx_rings = GFX7_NUM_GFX_RINGS; @@ -4151,9 +4151,9 @@ static int gfx_v7_0_early_init(void *handle) return 0; } -static int gfx_v7_0_late_init(void *handle) +static int gfx_v7_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_irq_get(adev, &adev->gfx.priv_reg_irq, 0); @@ -4343,10 +4343,10 @@ static int gfx_v7_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, return 0; } -static int gfx_v7_0_sw_init(void *handle) +static int gfx_v7_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j, k, r, ring_id; switch (adev->asic_type) { @@ -4439,9 +4439,9 @@ static int gfx_v7_0_sw_init(void *handle) return r; } -static int gfx_v7_0_sw_fini(void *handle) +static int gfx_v7_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->gfx.num_gfx_rings; i++) @@ -4465,10 +4465,10 @@ static int gfx_v7_0_sw_fini(void *handle) return 0; } -static int gfx_v7_0_hw_init(void *handle) +static int gfx_v7_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gfx_v7_0_constants_init(adev); @@ -4486,9 +4486,9 @@ static int gfx_v7_0_hw_init(void *handle) return r; } -static int gfx_v7_0_hw_fini(void *handle) +static int gfx_v7_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0); amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0); @@ -4499,18 +4499,14 @@ static int gfx_v7_0_hw_fini(void *handle) return 0; } -static int gfx_v7_0_suspend(void *handle) +static int gfx_v7_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return gfx_v7_0_hw_fini(adev); + return gfx_v7_0_hw_fini(ip_block); } -static int gfx_v7_0_resume(void *handle) +static int gfx_v7_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return gfx_v7_0_hw_init(adev); + return gfx_v7_0_hw_init(ip_block); } static bool gfx_v7_0_is_idle(void *handle) @@ -4523,11 +4519,11 @@ static bool gfx_v7_0_is_idle(void *handle) return true; } -static int gfx_v7_0_wait_for_idle(void *handle) +static int gfx_v7_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { /* read MC_STATUS */ @@ -4540,11 +4536,11 @@ static int gfx_v7_0_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int gfx_v7_0_soft_reset(void *handle) +static int gfx_v7_0_soft_reset(struct amdgpu_ip_block *ip_block) { u32 grbm_soft_reset = 0, srbm_soft_reset = 0; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* GRBM_STATUS */ tmp = RREG32(mmGRBM_STATUS); @@ -5009,8 +5005,6 @@ static const struct amd_ip_funcs gfx_v7_0_ip_funcs = { .soft_reset = gfx_v7_0_soft_reset, .set_clockgating_state = gfx_v7_0_set_clockgating_state, .set_powergating_state = gfx_v7_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index bc8295812cc8..480c41ee947e 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -1894,12 +1894,12 @@ static int gfx_v8_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, static void gfx_v8_0_sq_irq_work_func(struct work_struct *work); -static int gfx_v8_0_sw_init(void *handle) +static int gfx_v8_0_sw_init(struct amdgpu_ip_block *ip_block) { int i, j, k, r, ring_id; int xcc_id = 0; struct amdgpu_ring *ring; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; switch (adev->asic_type) { case CHIP_TONGA: @@ -2037,9 +2037,9 @@ static int gfx_v8_0_sw_init(void *handle) return 0; } -static int gfx_v8_0_sw_fini(void *handle) +static int gfx_v8_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->gfx.num_gfx_rings; i++) @@ -4260,7 +4260,7 @@ static int gfx_v8_0_cp_gfx_resume(struct amdgpu_device *adev) ring->wptr = 0; WREG32(mmCP_RB0_WPTR, lower_32_bits(ring->wptr)); - /* set the wb address wether it's enabled or not */ + /* set the wb address whether it's enabled or not */ rptr_addr = ring->rptr_gpu_addr; WREG32(mmCP_RB0_RPTR_ADDR, lower_32_bits(rptr_addr)); WREG32(mmCP_RB0_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & 0xFF); @@ -4783,10 +4783,10 @@ static void gfx_v8_0_cp_enable(struct amdgpu_device *adev, bool enable) gfx_v8_0_cp_compute_enable(adev, enable); } -static int gfx_v8_0_hw_init(void *handle) +static int gfx_v8_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gfx_v8_0_init_golden_registers(adev); gfx_v8_0_constants_init(adev); @@ -4865,13 +4865,13 @@ static int gfx_v8_0_wait_for_rlc_idle(void *handle) return -ETIMEDOUT; } -static int gfx_v8_0_wait_for_idle(void *handle) +static int gfx_v8_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { - if (gfx_v8_0_is_idle(handle)) + if (gfx_v8_0_is_idle(adev)) return 0; udelay(1); @@ -4879,9 +4879,9 @@ static int gfx_v8_0_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int gfx_v8_0_hw_fini(void *handle) +static int gfx_v8_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0); amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0); @@ -4897,8 +4897,9 @@ static int gfx_v8_0_hw_fini(void *handle) pr_debug("For SRIOV client, shouldn't do anything.\n"); return 0; } + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); - if (!gfx_v8_0_wait_for_idle(adev)) + if (!gfx_v8_0_wait_for_idle(ip_block)) gfx_v8_0_cp_enable(adev, false); else pr_err("cp is busy, skip halt cp\n"); @@ -4911,19 +4912,19 @@ static int gfx_v8_0_hw_fini(void *handle) return 0; } -static int gfx_v8_0_suspend(void *handle) +static int gfx_v8_0_suspend(struct amdgpu_ip_block *ip_block) { - return gfx_v8_0_hw_fini(handle); + return gfx_v8_0_hw_fini(ip_block); } -static int gfx_v8_0_resume(void *handle) +static int gfx_v8_0_resume(struct amdgpu_ip_block *ip_block) { - return gfx_v8_0_hw_init(handle); + return gfx_v8_0_hw_init(ip_block); } -static bool gfx_v8_0_check_soft_reset(void *handle) +static bool gfx_v8_0_check_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 grbm_soft_reset = 0, srbm_soft_reset = 0; u32 tmp; @@ -4983,9 +4984,9 @@ static bool gfx_v8_0_check_soft_reset(void *handle) } } -static int gfx_v8_0_pre_soft_reset(void *handle) +static int gfx_v8_0_pre_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 grbm_soft_reset = 0; if ((!adev->gfx.grbm_soft_reset) && @@ -5024,9 +5025,9 @@ static int gfx_v8_0_pre_soft_reset(void *handle) return 0; } -static int gfx_v8_0_soft_reset(void *handle) +static int gfx_v8_0_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 grbm_soft_reset = 0, srbm_soft_reset = 0; u32 tmp; @@ -5086,9 +5087,9 @@ static int gfx_v8_0_soft_reset(void *handle) return 0; } -static int gfx_v8_0_post_soft_reset(void *handle) +static int gfx_v8_0_post_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 grbm_soft_reset = 0; if ((!adev->gfx.grbm_soft_reset) && @@ -5254,9 +5255,9 @@ static const struct amdgpu_gfx_funcs gfx_v8_0_gfx_funcs = { .select_me_pipe_q = &gfx_v8_0_select_me_pipe_q }; -static int gfx_v8_0_early_init(void *handle) +static int gfx_v8_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->gfx.xcc_mask = 1; adev->gfx.num_gfx_rings = GFX8_NUM_GFX_RINGS; @@ -5271,9 +5272,9 @@ static int gfx_v8_0_early_init(void *handle) return 0; } -static int gfx_v8_0_late_init(void *handle) +static int gfx_v8_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_irq_get(adev, &adev->gfx.priv_reg_irq, 0); @@ -6947,8 +6948,6 @@ static const struct amd_ip_funcs gfx_v8_0_ip_funcs = { .set_clockgating_state = gfx_v8_0_set_clockgating_state, .set_powergating_state = gfx_v8_0_set_powergating_state, .get_clockgating_state = gfx_v8_0_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = { diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index 23f0573ae47b..66947850d7e4 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -2198,12 +2198,12 @@ static void gfx_v9_0_alloc_ip_dump(struct amdgpu_device *adev) } } -static int gfx_v9_0_sw_init(void *handle) +static int gfx_v9_0_sw_init(struct amdgpu_ip_block *ip_block) { int i, j, k, r, ring_id; int xcc_id = 0; struct amdgpu_ring *ring; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; unsigned int hw_prio; switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { @@ -2223,6 +2223,18 @@ static int gfx_v9_0_sw_init(void *handle) } switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { + case IP_VERSION(9, 4, 2): + adev->gfx.cleaner_shader_ptr = gfx_9_4_2_cleaner_shader_hex; + adev->gfx.cleaner_shader_size = sizeof(gfx_9_4_2_cleaner_shader_hex); + if (adev->gfx.mec_fw_version >= 88) { + adev->gfx.enable_cleaner_shader = true; + r = amdgpu_gfx_cleaner_shader_sw_init(adev, adev->gfx.cleaner_shader_size); + if (r) { + adev->gfx.enable_cleaner_shader = false; + dev_err(adev->dev, "Failed to initialize cleaner shader\n"); + } + } + break; default: adev->gfx.enable_cleaner_shader = false; break; @@ -2398,10 +2410,10 @@ static int gfx_v9_0_sw_init(void *handle) } -static int gfx_v9_0_sw_fini(void *handle) +static int gfx_v9_0_sw_fini(struct amdgpu_ip_block *ip_block) { int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->gfx.mcbp && adev->gfx.num_gfx_rings) { for (i = 0; i < GFX9_NUM_SW_GFX_RINGS; i++) @@ -2418,6 +2430,8 @@ static int gfx_v9_0_sw_fini(void *handle) amdgpu_gfx_kiq_free_ring(&adev->gfx.kiq[0].ring); amdgpu_gfx_kiq_fini(adev, 0); + amdgpu_gfx_cleaner_shader_sw_fini(adev); + gfx_v9_0_mec_fini(adev); amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, &adev->gfx.rlc.clear_state_gpu_addr, @@ -3184,6 +3198,15 @@ static void gfx_v9_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable) { u32 tmp = RREG32_SOC15(GC, 0, mmCP_ME_CNTL); + tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, CE_INVALIDATE_ICACHE, enable ? 0 : 1); + tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, PFP_INVALIDATE_ICACHE, enable ? 0 : 1); + tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, ME_INVALIDATE_ICACHE, enable ? 0 : 1); + tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, CE_PIPE0_RESET, enable ? 0 : 1); + tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, CE_PIPE1_RESET, enable ? 0 : 1); + tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, PFP_PIPE0_RESET, enable ? 0 : 1); + tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, PFP_PIPE1_RESET, enable ? 0 : 1); + tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, ME_PIPE0_RESET, enable ? 0 : 1); + tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, ME_PIPE1_RESET, enable ? 0 : 1); tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, ME_HALT, enable ? 0 : 1); tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, PFP_HALT, enable ? 0 : 1); tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, CE_HALT, enable ? 0 : 1); @@ -3346,7 +3369,7 @@ static int gfx_v9_0_cp_gfx_resume(struct amdgpu_device *adev) WREG32_SOC15(GC, 0, mmCP_RB0_WPTR, lower_32_bits(ring->wptr)); WREG32_SOC15(GC, 0, mmCP_RB0_WPTR_HI, upper_32_bits(ring->wptr)); - /* set the wb address wether it's enabled or not */ + /* set the wb address whether it's enabled or not */ rptr_addr = ring->rptr_gpu_addr; WREG32_SOC15(GC, 0, mmCP_RB0_RPTR_ADDR, lower_32_bits(rptr_addr)); WREG32_SOC15(GC, 0, mmCP_RB0_RPTR_ADDR_HI, upper_32_bits(rptr_addr) & CP_RB_RPTR_ADDR_HI__RB_RPTR_ADDR_HI_MASK); @@ -3393,7 +3416,15 @@ static void gfx_v9_0_cp_compute_enable(struct amdgpu_device *adev, bool enable) WREG32_SOC15_RLC(GC, 0, mmCP_MEC_CNTL, 0); } else { WREG32_SOC15_RLC(GC, 0, mmCP_MEC_CNTL, - (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK)); + (CP_MEC_CNTL__MEC_INVALIDATE_ICACHE_MASK | + CP_MEC_CNTL__MEC_ME1_PIPE0_RESET_MASK | + CP_MEC_CNTL__MEC_ME1_PIPE1_RESET_MASK | + CP_MEC_CNTL__MEC_ME1_PIPE2_RESET_MASK | + CP_MEC_CNTL__MEC_ME1_PIPE3_RESET_MASK | + CP_MEC_CNTL__MEC_ME2_PIPE0_RESET_MASK | + CP_MEC_CNTL__MEC_ME2_PIPE1_RESET_MASK | + CP_MEC_CNTL__MEC_ME1_HALT_MASK | + CP_MEC_CNTL__MEC_ME2_HALT_MASK)); adev->gfx.kiq[0].ring.sched.ready = false; } udelay(50); @@ -3914,6 +3945,10 @@ static int gfx_v9_0_cp_resume(struct amdgpu_device *adev) return r; } + if (adev->gfx.num_gfx_rings) + gfx_v9_0_cp_gfx_enable(adev, false); + gfx_v9_0_cp_compute_enable(adev, false); + r = gfx_v9_0_kiq_resume(adev); if (r) return r; @@ -3970,10 +4005,10 @@ static void gfx_v9_0_cp_enable(struct amdgpu_device *adev, bool enable) gfx_v9_0_cp_compute_enable(adev, enable); } -static int gfx_v9_0_hw_init(void *handle) +static int gfx_v9_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_gfx_cleaner_shader_init(adev, adev->gfx.cleaner_shader_size, adev->gfx.cleaner_shader_ptr); @@ -3999,9 +4034,9 @@ static int gfx_v9_0_hw_init(void *handle) return r; } -static int gfx_v9_0_hw_fini(void *handle) +static int gfx_v9_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__GFX)) amdgpu_irq_put(adev, &adev->gfx.cp_ecc_error_irq, 0); @@ -4051,14 +4086,14 @@ static int gfx_v9_0_hw_fini(void *handle) return 0; } -static int gfx_v9_0_suspend(void *handle) +static int gfx_v9_0_suspend(struct amdgpu_ip_block *ip_block) { - return gfx_v9_0_hw_fini(handle); + return gfx_v9_0_hw_fini(ip_block); } -static int gfx_v9_0_resume(void *handle) +static int gfx_v9_0_resume(struct amdgpu_ip_block *ip_block) { - return gfx_v9_0_hw_init(handle); + return gfx_v9_0_hw_init(ip_block); } static bool gfx_v9_0_is_idle(void *handle) @@ -4072,24 +4107,24 @@ static bool gfx_v9_0_is_idle(void *handle) return true; } -static int gfx_v9_0_wait_for_idle(void *handle) +static int gfx_v9_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { - if (gfx_v9_0_is_idle(handle)) + if (gfx_v9_0_is_idle(adev)) return 0; udelay(1); } return -ETIMEDOUT; } -static int gfx_v9_0_soft_reset(void *handle) +static int gfx_v9_0_soft_reset(struct amdgpu_ip_block *ip_block) { u32 grbm_soft_reset = 0; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* GRBM_STATUS */ tmp = RREG32_SOC15(GC, 0, mmGRBM_STATUS); @@ -4745,9 +4780,9 @@ fail: return r; } -static int gfx_v9_0_early_init(void *handle) +static int gfx_v9_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->gfx.funcs = &gfx_v9_0_gfx_funcs; @@ -4771,9 +4806,9 @@ static int gfx_v9_0_early_init(void *handle) return gfx_v9_0_init_microcode(adev); } -static int gfx_v9_0_ecc_late_init(void *handle) +static int gfx_v9_0_ecc_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; /* @@ -4805,9 +4840,9 @@ static int gfx_v9_0_ecc_late_init(void *handle) return 0; } -static int gfx_v9_0_late_init(void *handle) +static int gfx_v9_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_irq_get(adev, &adev->gfx.priv_reg_irq, 0); @@ -4822,7 +4857,7 @@ static int gfx_v9_0_late_init(void *handle) if (r) return r; - r = gfx_v9_0_ecc_late_init(handle); + r = gfx_v9_0_ecc_late_init(ip_block); if (r) return r; @@ -7167,8 +7202,6 @@ static void gfx_v9_0_emit_wave_limit(struct amdgpu_ring *ring, bool enable) static void gfx_v9_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) { - int i; - /* Header itself is a NOP packet */ if (num_nop == 1) { amdgpu_ring_write(ring, ring->funcs->nop); @@ -7179,8 +7212,7 @@ static void gfx_v9_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) amdgpu_ring_write(ring, PACKET3(PACKET3_NOP, min(num_nop - 2, 0x3ffe))); /* Header is at index 0, followed by num_nops - 1 NOP packet's */ - for (i = 1; i < num_nop; i++) - amdgpu_ring_write(ring, ring->funcs->nop); + amdgpu_ring_insert_nop(ring, num_nop - 1); } static int gfx_v9_0_reset_kgq(struct amdgpu_ring *ring, unsigned int vmid) @@ -7237,10 +7269,6 @@ static int gfx_v9_0_reset_kcq(struct amdgpu_ring *ring, unsigned long flags; int i, r; - if (!adev->debug_exp_resets && - !adev->gfx.num_gfx_rings) - return -EINVAL; - if (amdgpu_sriov_vf(adev)) return -EINVAL; @@ -7316,9 +7344,9 @@ static int gfx_v9_0_reset_kcq(struct amdgpu_ring *ring, return amdgpu_ring_test_ring(ring); } -static void gfx_v9_ip_print(void *handle, struct drm_printer *p) +static void gfx_v9_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t i, j, k, reg, index = 0; uint32_t reg_count = ARRAY_SIZE(gc_reg_list_9); @@ -7356,9 +7384,9 @@ static void gfx_v9_ip_print(void *handle, struct drm_printer *p) } -static void gfx_v9_ip_dump(void *handle) +static void gfx_v9_ip_dump(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t i, j, k, reg, index = 0; uint32_t reg_count = ARRAY_SIZE(gc_reg_list_9); diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0_cleaner_shader.h b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0_cleaner_shader.h index 36c0292b5110..0b6bd09b7529 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0_cleaner_shader.h +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0_cleaner_shader.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: MIT */ /* - * Copyright 2018 Advanced Micro Devices, Inc. + * Copyright 2024 Advanced Micro Devices, Inc. * * Permission is hereby granted, free of charge, to any person obtaining a * copy of this software and associated documentation files (the "Software"), @@ -24,3 +24,45 @@ static const u32 __maybe_unused gfx_9_0_cleaner_shader_hex[] = { /* Add the cleaner shader code here */ }; + +/* Define the cleaner shader gfx_9_4_2 */ +static const u32 gfx_9_4_2_cleaner_shader_hex[] = { + 0xbf068100, 0xbf84003b, + 0xbf8a0000, 0xb07c0000, + 0xbe8200ff, 0x00000078, + 0xbf110802, 0x7e000280, + 0x7e020280, 0x7e040280, + 0x7e060280, 0x7e080280, + 0x7e0a0280, 0x7e0c0280, + 0x7e0e0280, 0x80828802, + 0xbe803202, 0xbf84fff5, + 0xbf9c0000, 0xbe8200ff, + 0x80000000, 0x86020102, + 0xbf840011, 0xbefe00c1, + 0xbeff00c1, 0xd28c0001, + 0x0001007f, 0xd28d0001, + 0x0002027e, 0x10020288, + 0xbe8200bf, 0xbefc00c1, + 0xd89c2000, 0x00020201, + 0xd89c6040, 0x00040401, + 0x320202ff, 0x00000400, + 0x80828102, 0xbf84fff8, + 0xbefc00ff, 0x0000005c, + 0xbf800000, 0xbe802c80, + 0xbe812c80, 0xbe822c80, + 0xbe832c80, 0x80fc847c, + 0xbf84fffa, 0xbee60080, + 0xbee70080, 0xbeea0180, + 0xbeec0180, 0xbeee0180, + 0xbef00180, 0xbef20180, + 0xbef40180, 0xbef60180, + 0xbef80180, 0xbefa0180, + 0xbf810000, 0xbf8d0001, + 0xbefc00ff, 0x0000005c, + 0xbf800000, 0xbe802c80, + 0xbe812c80, 0xbe822c80, + 0xbe832c80, 0x80fc847c, + 0xbf84fffa, 0xbee60080, + 0xbee70080, 0xbeea01ff, + 0x000000ee, 0xbf810000, +}; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2_cleaner_shader.asm b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2_cleaner_shader.asm new file mode 100644 index 000000000000..35b8cf9070bd --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_2_cleaner_shader.asm @@ -0,0 +1,153 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright 2024 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ + +// This shader is to clean LDS, SGPRs and VGPRs. It is first 64 Dwords or 256 bytes of 192 Dwords cleaner shader. +//To turn this shader program on for complitaion change this to main and lower shader main to main_1 + +// MI200 : Clear SGPRs, VGPRs and LDS +// Uses two kernels launched separately: +// 1. Clean VGPRs, LDS, and lower SGPRs +// Launches one workgroup per CU, each workgroup with 4x wave64 per SIMD in the CU +// Waves are "wave64" and have 128 VGPRs each, which uses all 512 VGPRs per SIMD +// Waves in the workgroup share the 64KB of LDS +// Each wave clears SGPRs 0 - 95. Because there are 4 waves/SIMD, this is physical SGPRs 0-383 +// Each wave clears 128 VGPRs, so all 512 in the SIMD +// The first wave of the workgroup clears its 64KB of LDS +// The shader starts with "S_BARRIER" to ensure SPI has launched all waves of the workgroup +// before any wave in the workgroup could end. Without this, it is possible not all SGPRs get cleared. +// 2. Clean remaining SGPRs +// Launches a workgroup with 24 waves per workgroup, yielding 6 waves per SIMD in each CU +// Waves are allocating 96 SGPRs +// CP sets up SPI_RESOURCE_RESERVE_* registers to prevent these waves from allocating SGPRs 0-223. +// As such, these 6 waves per SIMD are allocated physical SGPRs 224-799 +// Barriers do not work for >16 waves per workgroup, so we cannot start with S_BARRIER +// Instead, the shader starts with an S_SETHALT 1. Once all waves are launched CP will send unhalt command +// The shader then clears all SGPRs allocated to it, cleaning out physical SGPRs 224-799 + +shader main + asic(MI200) + type(CS) + wave_size(64) +// Note: original source code from SQ team + +// (theorhetical fastest = ~512clks vgpr + 1536 lds + ~128 sgpr = 2176 clks) + + s_cmp_eq_u32 s0, 1 // Bit0 is set, sgpr0 is set then clear VGPRS and LDS as FW set COMPUTE_USER_DATA_3 + s_cbranch_scc0 label_0023 // Clean VGPRs and LDS if sgpr0 of wave is set, scc = (s3 == 1) + S_BARRIER + + s_movk_i32 m0, 0x0000 + s_mov_b32 s2, 0x00000078 // Loop 128/8=16 times (loop unrolled for performance) + // + // CLEAR VGPRs + // + s_set_gpr_idx_on s2, 0x8 // enable Dest VGPR indexing +label_0005: + v_mov_b32 v0, 0 + v_mov_b32 v1, 0 + v_mov_b32 v2, 0 + v_mov_b32 v3, 0 + v_mov_b32 v4, 0 + v_mov_b32 v5, 0 + v_mov_b32 v6, 0 + v_mov_b32 v7, 0 + s_sub_u32 s2, s2, 8 + s_set_gpr_idx_idx s2 + s_cbranch_scc0 label_0005 + s_set_gpr_idx_off + + // + // + + s_mov_b32 s2, 0x80000000 // Bit31 is first_wave + s_and_b32 s2, s2, s1 // sgpr0 has tg_size (first_wave) term as in ucode only COMPUTE_PGM_RSRC2.tg_size_en is set + s_cbranch_scc0 label_clean_sgpr_1 // Clean LDS if its first wave of ThreadGroup/WorkGroup + // CLEAR LDS + // + s_mov_b32 exec_lo, 0xffffffff + s_mov_b32 exec_hi, 0xffffffff + v_mbcnt_lo_u32_b32 v1, exec_hi, 0 // Set V1 to thread-ID (0..63) + v_mbcnt_hi_u32_b32 v1, exec_lo, v1 // Set V1 to thread-ID (0..63) + v_mul_u32_u24 v1, 0x00000008, v1 // * 8, so each thread is a double-dword address (8byte) + s_mov_b32 s2, 0x00000003f // 64 loop iterations + s_mov_b32 m0, 0xffffffff + // Clear all of LDS space + // Each FirstWave of WorkGroup clears 64kbyte block + +label_001F: + ds_write2_b64 v1, v[2:3], v[2:3] offset1:32 + ds_write2_b64 v1, v[4:5], v[4:5] offset0:64 offset1:96 + v_add_co_u32 v1, vcc, 0x00000400, v1 + s_sub_u32 s2, s2, 1 + s_cbranch_scc0 label_001F + // + // CLEAR SGPRs + // +label_clean_sgpr_1: + s_mov_b32 m0, 0x0000005c // Loop 96/4=24 times (loop unrolled for performance) + s_nop 0 +label_sgpr_loop: + s_movreld_b32 s0, 0 + s_movreld_b32 s1, 0 + s_movreld_b32 s2, 0 + s_movreld_b32 s3, 0 + s_sub_u32 m0, m0, 4 + s_cbranch_scc0 label_sgpr_loop + + //clear vcc, flat scratch + s_mov_b32 flat_scratch_lo, 0 //clear flat scratch lo SGPR + s_mov_b32 flat_scratch_hi, 0 //clear flat scratch hi SGPR + s_mov_b64 vcc, 0 //clear vcc + s_mov_b64 ttmp0, 0 //Clear ttmp0 and ttmp1 + s_mov_b64 ttmp2, 0 //Clear ttmp2 and ttmp3 + s_mov_b64 ttmp4, 0 //Clear ttmp4 and ttmp5 + s_mov_b64 ttmp6, 0 //Clear ttmp6 and ttmp7 + s_mov_b64 ttmp8, 0 //Clear ttmp8 and ttmp9 + s_mov_b64 ttmp10, 0 //Clear ttmp10 and ttmp11 + s_mov_b64 ttmp12, 0 //Clear ttmp12 and ttmp13 + s_mov_b64 ttmp14, 0 //Clear ttmp14 and ttmp15 +s_endpgm + +label_0023: + + s_sethalt 1 + + s_mov_b32 m0, 0x0000005c // Loop 96/4=24 times (loop unrolled for performance) + s_nop 0 +label_sgpr_loop1: + + s_movreld_b32 s0, 0 + s_movreld_b32 s1, 0 + s_movreld_b32 s2, 0 + s_movreld_b32 s3, 0 + s_sub_u32 m0, m0, 4 + s_cbranch_scc0 label_sgpr_loop1 + + //clear vcc, flat scratch + s_mov_b32 flat_scratch_lo, 0 //clear flat scratch lo SGPR + s_mov_b32 flat_scratch_hi, 0 //clear flat scratch hi SGPR + s_mov_b64 vcc, 0xee //clear vcc + +s_endpgm +end + diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c index c100845409f7..016290f00592 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c @@ -1049,10 +1049,10 @@ static void gfx_v9_4_3_alloc_ip_dump(struct amdgpu_device *adev) } } -static int gfx_v9_4_3_sw_init(void *handle) +static int gfx_v9_4_3_sw_init(struct amdgpu_ip_block *ip_block) { int i, j, k, r, ring_id, xcc_id, num_xcc; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { case IP_VERSION(9, 4, 3): @@ -1165,12 +1165,9 @@ static int gfx_v9_4_3_sw_init(void *handle) if (r) return r; - - if (!amdgpu_sriov_vf(adev)) { - r = amdgpu_gfx_sysfs_init(adev); - if (r) - return r; - } + r = amdgpu_gfx_sysfs_init(adev); + if (r) + return r; gfx_v9_4_3_alloc_ip_dump(adev); @@ -1181,10 +1178,10 @@ static int gfx_v9_4_3_sw_init(void *handle) return 0; } -static int gfx_v9_4_3_sw_fini(void *handle) +static int gfx_v9_4_3_sw_fini(struct amdgpu_ip_block *ip_block) { int i, num_xcc; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; num_xcc = NUM_XCC(adev->gfx.xcc_mask); for (i = 0; i < adev->gfx.num_compute_rings * num_xcc; i++) @@ -1201,8 +1198,7 @@ static int gfx_v9_4_3_sw_fini(void *handle) gfx_v9_4_3_mec_fini(adev); amdgpu_bo_unref(&adev->gfx.rlc.clear_state_obj); gfx_v9_4_3_free_microcode(adev); - if (!amdgpu_sriov_vf(adev)) - amdgpu_gfx_sysfs_fini(adev); + amdgpu_gfx_sysfs_fini(adev); amdgpu_gfx_sysfs_isolation_shader_fini(adev); kfree(adev->gfx.ip_dump_core); @@ -1247,8 +1243,10 @@ static void gfx_v9_4_3_xcc_init_compute_vmid(struct amdgpu_device *adev, soc15_grbm_select(adev, 0, 0, 0, 0, GET_INST(GC, xcc_id)); mutex_unlock(&adev->srbm_mutex); - /* Initialize all compute VMIDs to have no GDS, GWS, or OA - acccess. These should be enabled by FW for target VMIDs. */ + /* + * Initialize all compute VMIDs to have no GDS, GWS, or OA + * access. These should be enabled by FW for target VMIDs. + */ for (i = adev->vm_manager.first_kfd_vmid; i < AMDGPU_NUM_VMID; i++) { WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_VMID0_BASE, 2 * i, 0); WREG32_SOC15_OFFSET(GC, GET_INST(GC, xcc_id), regGDS_VMID0_SIZE, 2 * i, 0); @@ -2343,10 +2341,10 @@ static void gfx_v9_4_3_xcc_fini(struct amdgpu_device *adev, int xcc_id) gfx_v9_4_3_xcc_cp_compute_enable(adev, false, xcc_id); } -static int gfx_v9_4_3_hw_init(void *handle) +static int gfx_v9_4_3_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_gfx_cleaner_shader_init(adev, adev->gfx.cleaner_shader_size, adev->gfx.cleaner_shader_ptr); @@ -2367,9 +2365,9 @@ static int gfx_v9_4_3_hw_init(void *handle) return r; } -static int gfx_v9_4_3_hw_fini(void *handle) +static int gfx_v9_4_3_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, num_xcc; amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0); @@ -2384,14 +2382,14 @@ static int gfx_v9_4_3_hw_fini(void *handle) return 0; } -static int gfx_v9_4_3_suspend(void *handle) +static int gfx_v9_4_3_suspend(struct amdgpu_ip_block *ip_block) { - return gfx_v9_4_3_hw_fini(handle); + return gfx_v9_4_3_hw_fini(ip_block); } -static int gfx_v9_4_3_resume(void *handle) +static int gfx_v9_4_3_resume(struct amdgpu_ip_block *ip_block) { - return gfx_v9_4_3_hw_init(handle); + return gfx_v9_4_3_hw_init(ip_block); } static bool gfx_v9_4_3_is_idle(void *handle) @@ -2408,24 +2406,24 @@ static bool gfx_v9_4_3_is_idle(void *handle) return true; } -static int gfx_v9_4_3_wait_for_idle(void *handle) +static int gfx_v9_4_3_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { - if (gfx_v9_4_3_is_idle(handle)) + if (gfx_v9_4_3_is_idle(adev)) return 0; udelay(1); } return -ETIMEDOUT; } -static int gfx_v9_4_3_soft_reset(void *handle) +static int gfx_v9_4_3_soft_reset(struct amdgpu_ip_block *ip_block) { u32 grbm_soft_reset = 0; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* GRBM_STATUS */ tmp = RREG32_SOC15(GC, GET_INST(GC, 0), regGRBM_STATUS); @@ -2509,9 +2507,9 @@ static void gfx_v9_4_3_ring_emit_gds_switch(struct amdgpu_ring *ring, (1 << (oa_size + oa_base)) - (1 << oa_base)); } -static int gfx_v9_4_3_early_init(void *handle) +static int gfx_v9_4_3_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev), AMDGPU_MAX_COMPUTE_RINGS); @@ -2527,9 +2525,9 @@ static int gfx_v9_4_3_early_init(void *handle) return gfx_v9_4_3_init_microcode(adev); } -static int gfx_v9_4_3_late_init(void *handle) +static int gfx_v9_4_3_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_irq_get(adev, &adev->gfx.priv_reg_irq, 0); @@ -3056,9 +3054,6 @@ static void gfx_v9_4_3_ring_soft_recovery(struct amdgpu_ring *ring, struct amdgpu_device *adev = ring->adev; uint32_t value = 0; - if (!adev->debug_exp_resets) - return; - value = REG_SET_FIELD(value, SQ_CMD, CMD, 0x03); value = REG_SET_FIELD(value, SQ_CMD, MODE, 0x01); value = REG_SET_FIELD(value, SQ_CMD, CHECK_VMID, 1); @@ -3574,9 +3569,6 @@ static int gfx_v9_4_3_reset_kcq(struct amdgpu_ring *ring, unsigned long flags; int r; - if (!adev->debug_exp_resets) - return -EINVAL; - if (amdgpu_sriov_vf(adev)) return -EINVAL; @@ -4567,8 +4559,6 @@ static void gfx_v9_4_3_enable_watchdog_timer(struct amdgpu_device *adev) static void gfx_v9_4_3_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) { - int i; - /* Header itself is a NOP packet */ if (num_nop == 1) { amdgpu_ring_write(ring, ring->funcs->nop); @@ -4579,13 +4569,12 @@ static void gfx_v9_4_3_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_no amdgpu_ring_write(ring, PACKET3(PACKET3_NOP, min(num_nop - 2, 0x3ffe))); /* Header is at index 0, followed by num_nops - 1 NOP packet's */ - for (i = 1; i < num_nop; i++) - amdgpu_ring_write(ring, ring->funcs->nop); + amdgpu_ring_insert_nop(ring, num_nop - 1); } -static void gfx_v9_4_3_ip_print(void *handle, struct drm_printer *p) +static void gfx_v9_4_3_ip_print(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t i, j, k; uint32_t xcc_id, xcc_offset, inst_offset; uint32_t num_xcc, reg, num_inst; @@ -4643,9 +4632,9 @@ static void gfx_v9_4_3_ip_print(void *handle, struct drm_printer *p) } } -static void gfx_v9_4_3_ip_dump(void *handle) +static void gfx_v9_4_3_ip_dump(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t i, j, k; uint32_t num_xcc, reg, num_inst; uint32_t xcc_id, xcc_offset, inst_offset; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index 9784a2892185..697599c46240 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -175,7 +175,10 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev, addr, entry->client_id, soc15_ih_clientid_name[entry->client_id]); - if (!amdgpu_sriov_vf(adev)) + /* Only print L2 fault status if the status register could be read and + * contains useful information + */ + if (status != 0) hub->vmhub_funcs->print_l2_protection_fault_status(adev, status); @@ -630,9 +633,9 @@ static void gmc_v10_0_set_gfxhub_funcs(struct amdgpu_device *adev) } -static int gmc_v10_0_early_init(void *handle) +static int gmc_v10_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gmc_v10_0_set_mmhub_funcs(adev); gmc_v10_0_set_gfxhub_funcs(adev); @@ -651,9 +654,9 @@ static int gmc_v10_0_early_init(void *handle) return 0; } -static int gmc_v10_0_late_init(void *handle) +static int gmc_v10_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_gmc_allocate_vm_inv_eng(adev); @@ -769,10 +772,10 @@ static int gmc_v10_0_gart_init(struct amdgpu_device *adev) return amdgpu_gart_table_vram_alloc(adev); } -static int gmc_v10_0_sw_init(void *handle) +static int gmc_v10_0_sw_init(struct amdgpu_ip_block *ip_block) { int r, vram_width = 0, vram_type = 0, vram_vendor = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->gfxhub.funcs->init(adev); @@ -920,9 +923,9 @@ static void gmc_v10_0_gart_fini(struct amdgpu_device *adev) amdgpu_gart_table_vram_free(adev); } -static int gmc_v10_0_sw_fini(void *handle) +static int gmc_v10_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_vm_manager_fini(adev); gmc_v10_0_gart_fini(adev); @@ -985,9 +988,9 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev) return 0; } -static int gmc_v10_0_hw_init(void *handle) +static int gmc_v10_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; adev->gmc.flush_pasid_uses_kiq = !amdgpu_emu_mode; @@ -1032,9 +1035,9 @@ static void gmc_v10_0_gart_disable(struct amdgpu_device *adev) adev->mmhub.funcs->gart_disable(adev); } -static int gmc_v10_0_hw_fini(void *handle) +static int gmc_v10_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gmc_v10_0_gart_disable(adev); @@ -1053,25 +1056,22 @@ static int gmc_v10_0_hw_fini(void *handle) return 0; } -static int gmc_v10_0_suspend(void *handle) +static int gmc_v10_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - gmc_v10_0_hw_fini(adev); + gmc_v10_0_hw_fini(ip_block); return 0; } -static int gmc_v10_0_resume(void *handle) +static int gmc_v10_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = gmc_v10_0_hw_init(adev); + r = gmc_v10_0_hw_init(ip_block); if (r) return r; - amdgpu_vmid_reset_all(adev); + amdgpu_vmid_reset_all(ip_block->adev); return 0; } @@ -1082,17 +1082,12 @@ static bool gmc_v10_0_is_idle(void *handle) return true; } -static int gmc_v10_0_wait_for_idle(void *handle) +static int gmc_v10_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* There is no need to wait for MC idle in GMC v10.*/ return 0; } -static int gmc_v10_0_soft_reset(void *handle) -{ - return 0; -} - static int gmc_v10_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -1154,7 +1149,6 @@ const struct amd_ip_funcs gmc_v10_0_ip_funcs = { .resume = gmc_v10_0_resume, .is_idle = gmc_v10_0_is_idle, .wait_for_idle = gmc_v10_0_wait_for_idle, - .soft_reset = gmc_v10_0_soft_reset, .set_clockgating_state = gmc_v10_0_set_clockgating_state, .set_powergating_state = gmc_v10_0_set_powergating_state, .get_clockgating_state = gmc_v10_0_get_clockgating_state, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index 2797fd84432b..f893ab4c14df 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -144,7 +144,10 @@ static int gmc_v11_0_process_interrupt(struct amdgpu_device *adev, dev_err(adev->dev, " in page starting at address 0x%016llx from client %d\n", addr, entry->client_id); - if (!amdgpu_sriov_vf(adev)) + /* Only print L2 fault status if the status register could be read and + * contains useful information + */ + if (status != 0) hub->vmhub_funcs->print_l2_protection_fault_status(adev, status); } @@ -601,9 +604,9 @@ static void gmc_v11_0_set_gfxhub_funcs(struct amdgpu_device *adev) } } -static int gmc_v11_0_early_init(void *handle) +static int gmc_v11_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gmc_v11_0_set_gfxhub_funcs(adev); gmc_v11_0_set_mmhub_funcs(adev); @@ -622,9 +625,9 @@ static int gmc_v11_0_early_init(void *handle) return 0; } -static int gmc_v11_0_late_init(void *handle) +static int gmc_v11_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_gmc_allocate_vm_inv_eng(adev); @@ -729,10 +732,10 @@ static int gmc_v11_0_gart_init(struct amdgpu_device *adev) return amdgpu_gart_table_vram_alloc(adev); } -static int gmc_v11_0_sw_init(void *handle) +static int gmc_v11_0_sw_init(struct amdgpu_ip_block *ip_block) { int r, vram_width = 0, vram_type = 0, vram_vendor = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->mmhub.funcs->init(adev); @@ -849,9 +852,9 @@ static void gmc_v11_0_gart_fini(struct amdgpu_device *adev) amdgpu_gart_table_vram_free(adev); } -static int gmc_v11_0_sw_fini(void *handle) +static int gmc_v11_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_vm_manager_fini(adev); gmc_v11_0_gart_fini(adev); @@ -908,9 +911,9 @@ static int gmc_v11_0_gart_enable(struct amdgpu_device *adev) return 0; } -static int gmc_v11_0_hw_init(void *handle) +static int gmc_v11_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; adev->gmc.flush_pasid_uses_kiq = !amdgpu_emu_mode; @@ -940,9 +943,9 @@ static void gmc_v11_0_gart_disable(struct amdgpu_device *adev) adev->mmhub.funcs->gart_disable(adev); } -static int gmc_v11_0_hw_fini(void *handle) +static int gmc_v11_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) { /* full access mode, so don't touch any GMC register */ @@ -961,25 +964,22 @@ static int gmc_v11_0_hw_fini(void *handle) return 0; } -static int gmc_v11_0_suspend(void *handle) +static int gmc_v11_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - gmc_v11_0_hw_fini(adev); + gmc_v11_0_hw_fini(ip_block); return 0; } -static int gmc_v11_0_resume(void *handle) +static int gmc_v11_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = gmc_v11_0_hw_init(adev); + r = gmc_v11_0_hw_init(ip_block); if (r) return r; - amdgpu_vmid_reset_all(adev); + amdgpu_vmid_reset_all(ip_block->adev); return 0; } @@ -990,17 +990,12 @@ static bool gmc_v11_0_is_idle(void *handle) return true; } -static int gmc_v11_0_wait_for_idle(void *handle) +static int gmc_v11_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* There is no need to wait for MC idle in GMC v11.*/ return 0; } -static int gmc_v11_0_soft_reset(void *handle) -{ - return 0; -} - static int gmc_v11_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -1041,7 +1036,6 @@ const struct amd_ip_funcs gmc_v11_0_ip_funcs = { .resume = gmc_v11_0_resume, .is_idle = gmc_v11_0_is_idle, .wait_for_idle = gmc_v11_0_wait_for_idle, - .soft_reset = gmc_v11_0_soft_reset, .set_clockgating_state = gmc_v11_0_set_clockgating_state, .set_powergating_state = gmc_v11_0_set_powergating_state, .get_clockgating_state = gmc_v11_0_get_clockgating_state, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c index edcb5351f8cc..d22b027fd0bb 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c @@ -137,7 +137,10 @@ static int gmc_v12_0_process_interrupt(struct amdgpu_device *adev, dev_err(adev->dev, " in page starting at address 0x%016llx from client %d\n", addr, entry->client_id); - if (!amdgpu_sriov_vf(adev)) + /* Only print L2 fault status if the status register could be read and + * contains useful information + */ + if (status != 0) hub->vmhub_funcs->print_l2_protection_fault_status(adev, status); } @@ -604,9 +607,9 @@ static void gmc_v12_0_set_gfxhub_funcs(struct amdgpu_device *adev) } } -static int gmc_v12_0_early_init(void *handle) +static int gmc_v12_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gmc_v12_0_set_gfxhub_funcs(adev); gmc_v12_0_set_mmhub_funcs(adev); @@ -624,9 +627,9 @@ static int gmc_v12_0_early_init(void *handle) return 0; } -static int gmc_v12_0_late_init(void *handle) +static int gmc_v12_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_gmc_allocate_vm_inv_eng(adev); @@ -731,10 +734,10 @@ static int gmc_v12_0_gart_init(struct amdgpu_device *adev) return amdgpu_gart_table_vram_alloc(adev); } -static int gmc_v12_0_sw_init(void *handle) +static int gmc_v12_0_sw_init(struct amdgpu_ip_block *ip_block) { int r, vram_width = 0, vram_type = 0, vram_vendor = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->mmhub.funcs->init(adev); @@ -841,9 +844,9 @@ static void gmc_v12_0_gart_fini(struct amdgpu_device *adev) amdgpu_gart_table_vram_free(adev); } -static int gmc_v12_0_sw_fini(void *handle) +static int gmc_v12_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_vm_manager_fini(adev); gmc_v12_0_gart_fini(adev); @@ -894,10 +897,10 @@ static int gmc_v12_0_gart_enable(struct amdgpu_device *adev) return 0; } -static int gmc_v12_0_hw_init(void *handle) +static int gmc_v12_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* The sequence of these two function calls matters.*/ gmc_v12_0_init_golden_registers(adev); @@ -924,9 +927,9 @@ static void gmc_v12_0_gart_disable(struct amdgpu_device *adev) adev->mmhub.funcs->gart_disable(adev); } -static int gmc_v12_0_hw_fini(void *handle) +static int gmc_v12_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) { /* full access mode, so don't touch any GMC register */ @@ -945,25 +948,22 @@ static int gmc_v12_0_hw_fini(void *handle) return 0; } -static int gmc_v12_0_suspend(void *handle) +static int gmc_v12_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - gmc_v12_0_hw_fini(adev); + gmc_v12_0_hw_fini(ip_block); return 0; } -static int gmc_v12_0_resume(void *handle) +static int gmc_v12_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = gmc_v12_0_hw_init(adev); + r = gmc_v12_0_hw_init(ip_block); if (r) return r; - amdgpu_vmid_reset_all(adev); + amdgpu_vmid_reset_all(ip_block->adev); return 0; } @@ -974,17 +974,12 @@ static bool gmc_v12_0_is_idle(void *handle) return true; } -static int gmc_v12_0_wait_for_idle(void *handle) +static int gmc_v12_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* There is no need to wait for MC idle in GMC v11.*/ return 0; } -static int gmc_v12_0_soft_reset(void *handle) -{ - return 0; -} - static int gmc_v12_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -1025,7 +1020,6 @@ const struct amd_ip_funcs gmc_v12_0_ip_funcs = { .resume = gmc_v12_0_resume, .is_idle = gmc_v12_0_is_idle, .wait_for_idle = gmc_v12_0_wait_for_idle, - .soft_reset = gmc_v12_0_soft_reset, .set_clockgating_state = gmc_v12_0_set_clockgating_state, .set_powergating_state = gmc_v12_0_set_powergating_state, .get_clockgating_state = gmc_v12_0_get_clockgating_state, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index d36725666b54..ca000b3d1afc 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -43,7 +43,7 @@ static void gmc_v6_0_set_gmc_funcs(struct amdgpu_device *adev); static void gmc_v6_0_set_irq_funcs(struct amdgpu_device *adev); -static int gmc_v6_0_wait_for_idle(void *handle); +static int gmc_v6_0_wait_for_idle(struct amdgpu_ip_block *ip_block); MODULE_FIRMWARE("amdgpu/tahiti_mc.bin"); MODULE_FIRMWARE("amdgpu/pitcairn_mc.bin"); @@ -64,8 +64,13 @@ MODULE_FIRMWARE("amdgpu/si58_mc.bin"); static void gmc_v6_0_mc_stop(struct amdgpu_device *adev) { u32 blackout; + struct amdgpu_ip_block *ip_block; - gmc_v6_0_wait_for_idle((void *)adev); + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_GMC); + if (!ip_block) + return; + + gmc_v6_0_wait_for_idle(ip_block); blackout = RREG32(mmMC_SHARED_BLACKOUT_CNTL); if (REG_GET_FIELD(blackout, MC_SHARED_BLACKOUT_CNTL, BLACKOUT_MODE) != 1) { @@ -213,6 +218,8 @@ static void gmc_v6_0_vram_gtt_location(struct amdgpu_device *adev, static void gmc_v6_0_mc_program(struct amdgpu_device *adev) { int i, j; + struct amdgpu_ip_block *ip_block; + /* Initialize HDP */ for (i = 0, j = 0; i < 32; i++, j += 0x6) { @@ -224,7 +231,11 @@ static void gmc_v6_0_mc_program(struct amdgpu_device *adev) } WREG32(mmHDP_REG_COHERENCY_FLUSH_CNTL, 0); - if (gmc_v6_0_wait_for_idle((void *)adev)) + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_GMC); + if (!ip_block) + return; + + if (gmc_v6_0_wait_for_idle(ip_block)) dev_warn(adev->dev, "Wait for MC idle timedout !\n"); if (adev->mode_info.num_crtc) { @@ -251,7 +262,7 @@ static void gmc_v6_0_mc_program(struct amdgpu_device *adev) WREG32(mmMC_VM_AGP_TOP, adev->gmc.agp_end >> 22); WREG32(mmMC_VM_AGP_BOT, adev->gmc.agp_start >> 22); - if (gmc_v6_0_wait_for_idle((void *)adev)) + if (gmc_v6_0_wait_for_idle(ip_block)) dev_warn(adev->dev, "Wait for MC idle timedout !\n"); } @@ -762,9 +773,9 @@ static int gmc_v6_0_convert_vram_type(int mc_seq_vram_type) } } -static int gmc_v6_0_early_init(void *handle) +static int gmc_v6_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gmc_v6_0_set_gmc_funcs(adev); gmc_v6_0_set_irq_funcs(adev); @@ -772,9 +783,9 @@ static int gmc_v6_0_early_init(void *handle) return 0; } -static int gmc_v6_0_late_init(void *handle) +static int gmc_v6_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_vm_fault_stop != AMDGPU_VM_FAULT_STOP_ALWAYS) return amdgpu_irq_get(adev, &adev->gmc.vm_fault, 0); @@ -799,10 +810,10 @@ static unsigned int gmc_v6_0_get_vbios_fb_size(struct amdgpu_device *adev) return size; } -static int gmc_v6_0_sw_init(void *handle) +static int gmc_v6_0_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask); @@ -876,9 +887,9 @@ static int gmc_v6_0_sw_init(void *handle) return 0; } -static int gmc_v6_0_sw_fini(void *handle) +static int gmc_v6_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_gem_force_release(adev); amdgpu_vm_manager_fini(adev); @@ -889,10 +900,10 @@ static int gmc_v6_0_sw_fini(void *handle) return 0; } -static int gmc_v6_0_hw_init(void *handle) +static int gmc_v6_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gmc_v6_0_mc_program(adev); @@ -914,9 +925,9 @@ static int gmc_v6_0_hw_init(void *handle) return 0; } -static int gmc_v6_0_hw_fini(void *handle) +static int gmc_v6_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0); gmc_v6_0_gart_disable(adev); @@ -924,21 +935,19 @@ static int gmc_v6_0_hw_fini(void *handle) return 0; } -static int gmc_v6_0_suspend(void *handle) +static int gmc_v6_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - gmc_v6_0_hw_fini(adev); + gmc_v6_0_hw_fini(ip_block); return 0; } -static int gmc_v6_0_resume(void *handle) +static int gmc_v6_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; - r = gmc_v6_0_hw_init(adev); + r = gmc_v6_0_hw_init(ip_block); if (r) return r; @@ -950,6 +959,7 @@ static int gmc_v6_0_resume(void *handle) static bool gmc_v6_0_is_idle(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + u32 tmp = RREG32(mmSRBM_STATUS); if (tmp & (SRBM_STATUS__MCB_BUSY_MASK | SRBM_STATUS__MCB_NON_DISPLAY_BUSY_MASK | @@ -959,13 +969,13 @@ static bool gmc_v6_0_is_idle(void *handle) return true; } -static int gmc_v6_0_wait_for_idle(void *handle) +static int gmc_v6_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned int i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { - if (gmc_v6_0_is_idle(handle)) + if (gmc_v6_0_is_idle(adev)) return 0; udelay(1); } @@ -973,9 +983,10 @@ static int gmc_v6_0_wait_for_idle(void *handle) } -static int gmc_v6_0_soft_reset(void *handle) +static int gmc_v6_0_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; + u32 srbm_soft_reset = 0; u32 tmp = RREG32(mmSRBM_STATUS); @@ -992,7 +1003,8 @@ static int gmc_v6_0_soft_reset(void *handle) if (srbm_soft_reset) { gmc_v6_0_mc_stop(adev); - if (gmc_v6_0_wait_for_idle(adev)) + + if (gmc_v6_0_wait_for_idle(ip_block)) dev_warn(adev->dev, "Wait for GMC idle timed out !\n"); tmp = RREG32(mmSRBM_SOFT_RESET); @@ -1109,8 +1121,6 @@ static const struct amd_ip_funcs gmc_v6_0_ip_funcs = { .soft_reset = gmc_v6_0_soft_reset, .set_clockgating_state = gmc_v6_0_set_clockgating_state, .set_powergating_state = gmc_v6_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_gmc_funcs gmc_v6_0_gmc_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index 994432fb57ea..07f45f1a503a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -52,7 +52,7 @@ static void gmc_v7_0_set_gmc_funcs(struct amdgpu_device *adev); static void gmc_v7_0_set_irq_funcs(struct amdgpu_device *adev); -static int gmc_v7_0_wait_for_idle(void *handle); +static int gmc_v7_0_wait_for_idle(struct amdgpu_ip_block *ip_block); MODULE_FIRMWARE("amdgpu/bonaire_mc.bin"); MODULE_FIRMWARE("amdgpu/hawaii_mc.bin"); @@ -921,9 +921,9 @@ static int gmc_v7_0_convert_vram_type(int mc_seq_vram_type) } } -static int gmc_v7_0_early_init(void *handle) +static int gmc_v7_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gmc_v7_0_set_gmc_funcs(adev); gmc_v7_0_set_irq_funcs(adev); @@ -940,9 +940,9 @@ static int gmc_v7_0_early_init(void *handle) return 0; } -static int gmc_v7_0_late_init(void *handle) +static int gmc_v7_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_vm_fault_stop != AMDGPU_VM_FAULT_STOP_ALWAYS) return amdgpu_irq_get(adev, &adev->gmc.vm_fault, 0); @@ -968,10 +968,10 @@ static unsigned int gmc_v7_0_get_vbios_fb_size(struct amdgpu_device *adev) return size; } -static int gmc_v7_0_sw_init(void *handle) +static int gmc_v7_0_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask); @@ -1060,9 +1060,9 @@ static int gmc_v7_0_sw_init(void *handle) return 0; } -static int gmc_v7_0_sw_fini(void *handle) +static int gmc_v7_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_gem_force_release(adev); amdgpu_vm_manager_fini(adev); @@ -1074,10 +1074,10 @@ static int gmc_v7_0_sw_fini(void *handle) return 0; } -static int gmc_v7_0_hw_init(void *handle) +static int gmc_v7_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gmc_v7_0_init_golden_registers(adev); @@ -1101,9 +1101,9 @@ static int gmc_v7_0_hw_init(void *handle) return 0; } -static int gmc_v7_0_hw_fini(void *handle) +static int gmc_v7_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0); gmc_v7_0_gart_disable(adev); @@ -1111,25 +1111,22 @@ static int gmc_v7_0_hw_fini(void *handle) return 0; } -static int gmc_v7_0_suspend(void *handle) +static int gmc_v7_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - gmc_v7_0_hw_fini(adev); + gmc_v7_0_hw_fini(ip_block); return 0; } -static int gmc_v7_0_resume(void *handle) +static int gmc_v7_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = gmc_v7_0_hw_init(adev); + r = gmc_v7_0_hw_init(ip_block); if (r) return r; - amdgpu_vmid_reset_all(adev); + amdgpu_vmid_reset_all(ip_block->adev); return 0; } @@ -1146,11 +1143,11 @@ static bool gmc_v7_0_is_idle(void *handle) return true; } -static int gmc_v7_0_wait_for_idle(void *handle) +static int gmc_v7_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned int i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { /* read MC_STATUS */ @@ -1167,9 +1164,9 @@ static int gmc_v7_0_wait_for_idle(void *handle) } -static int gmc_v7_0_soft_reset(void *handle) +static int gmc_v7_0_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset = 0; u32 tmp = RREG32(mmSRBM_STATUS); @@ -1351,8 +1348,6 @@ static const struct amd_ip_funcs gmc_v7_0_ip_funcs = { .soft_reset = gmc_v7_0_soft_reset, .set_clockgating_state = gmc_v7_0_set_clockgating_state, .set_powergating_state = gmc_v7_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_gmc_funcs gmc_v7_0_gmc_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 86488c052f82..12d5967ecd45 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -53,7 +53,7 @@ static void gmc_v8_0_set_gmc_funcs(struct amdgpu_device *adev); static void gmc_v8_0_set_irq_funcs(struct amdgpu_device *adev); -static int gmc_v8_0_wait_for_idle(void *handle); +static int gmc_v8_0_wait_for_idle(struct amdgpu_ip_block *ip_block); MODULE_FIRMWARE("amdgpu/tonga_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris11_mc.bin"); @@ -170,8 +170,13 @@ static void gmc_v8_0_init_golden_registers(struct amdgpu_device *adev) static void gmc_v8_0_mc_stop(struct amdgpu_device *adev) { u32 blackout; + struct amdgpu_ip_block *ip_block; - gmc_v8_0_wait_for_idle(adev); + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_GMC); + if (!ip_block) + return; + + gmc_v8_0_wait_for_idle(ip_block); blackout = RREG32(mmMC_SHARED_BLACKOUT_CNTL); if (REG_GET_FIELD(blackout, MC_SHARED_BLACKOUT_CNTL, BLACKOUT_MODE) != 1) { @@ -426,6 +431,7 @@ static void gmc_v8_0_vram_gtt_location(struct amdgpu_device *adev, */ static void gmc_v8_0_mc_program(struct amdgpu_device *adev) { + struct amdgpu_ip_block *ip_block; u32 tmp; int i, j; @@ -439,7 +445,11 @@ static void gmc_v8_0_mc_program(struct amdgpu_device *adev) } WREG32(mmHDP_REG_COHERENCY_FLUSH_CNTL, 0); - if (gmc_v8_0_wait_for_idle((void *)adev)) + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_GMC); + if (!ip_block) + return; + + if (gmc_v8_0_wait_for_idle(ip_block)) dev_warn(adev->dev, "Wait for MC idle timedout !\n"); if (adev->mode_info.num_crtc) { @@ -474,7 +484,7 @@ static void gmc_v8_0_mc_program(struct amdgpu_device *adev) WREG32(mmMC_VM_AGP_BASE, 0); WREG32(mmMC_VM_AGP_TOP, adev->gmc.agp_end >> 22); WREG32(mmMC_VM_AGP_BOT, adev->gmc.agp_start >> 22); - if (gmc_v8_0_wait_for_idle((void *)adev)) + if (gmc_v8_0_wait_for_idle(ip_block)) dev_warn(adev->dev, "Wait for MC idle timedout !\n"); WREG32(mmBIF_FB_EN, BIF_FB_EN__FB_READ_EN_MASK | BIF_FB_EN__FB_WRITE_EN_MASK); @@ -1027,9 +1037,9 @@ static int gmc_v8_0_convert_vram_type(int mc_seq_vram_type) } } -static int gmc_v8_0_early_init(void *handle) +static int gmc_v8_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gmc_v8_0_set_gmc_funcs(adev); gmc_v8_0_set_irq_funcs(adev); @@ -1046,9 +1056,9 @@ static int gmc_v8_0_early_init(void *handle) return 0; } -static int gmc_v8_0_late_init(void *handle) +static int gmc_v8_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_vm_fault_stop != AMDGPU_VM_FAULT_STOP_ALWAYS) return amdgpu_irq_get(adev, &adev->gmc.vm_fault, 0); @@ -1076,10 +1086,10 @@ static unsigned int gmc_v8_0_get_vbios_fb_size(struct amdgpu_device *adev) #define mmMC_SEQ_MISC0_FIJI 0xA71 -static int gmc_v8_0_sw_init(void *handle) +static int gmc_v8_0_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; set_bit(AMDGPU_GFXHUB(0), adev->vmhubs_mask); @@ -1173,9 +1183,9 @@ static int gmc_v8_0_sw_init(void *handle) return 0; } -static int gmc_v8_0_sw_fini(void *handle) +static int gmc_v8_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_gem_force_release(adev); amdgpu_vm_manager_fini(adev); @@ -1187,10 +1197,10 @@ static int gmc_v8_0_sw_fini(void *handle) return 0; } -static int gmc_v8_0_hw_init(void *handle) +static int gmc_v8_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gmc_v8_0_init_golden_registers(adev); @@ -1222,9 +1232,9 @@ static int gmc_v8_0_hw_init(void *handle) return 0; } -static int gmc_v8_0_hw_fini(void *handle) +static int gmc_v8_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0); gmc_v8_0_gart_disable(adev); @@ -1232,25 +1242,22 @@ static int gmc_v8_0_hw_fini(void *handle) return 0; } -static int gmc_v8_0_suspend(void *handle) +static int gmc_v8_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - gmc_v8_0_hw_fini(adev); + gmc_v8_0_hw_fini(ip_block); return 0; } -static int gmc_v8_0_resume(void *handle) +static int gmc_v8_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = gmc_v8_0_hw_init(adev); + r = gmc_v8_0_hw_init(ip_block); if (r) return r; - amdgpu_vmid_reset_all(adev); + amdgpu_vmid_reset_all(ip_block->adev); return 0; } @@ -1267,11 +1274,11 @@ static bool gmc_v8_0_is_idle(void *handle) return true; } -static int gmc_v8_0_wait_for_idle(void *handle) +static int gmc_v8_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned int i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { /* read MC_STATUS */ @@ -1289,10 +1296,10 @@ static int gmc_v8_0_wait_for_idle(void *handle) } -static bool gmc_v8_0_check_soft_reset(void *handle) +static bool gmc_v8_0_check_soft_reset(struct amdgpu_ip_block *ip_block) { u32 srbm_soft_reset = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 tmp = RREG32(mmSRBM_STATUS); if (tmp & SRBM_STATUS__VMC_BUSY_MASK) @@ -1316,23 +1323,23 @@ static bool gmc_v8_0_check_soft_reset(void *handle) return false; } -static int gmc_v8_0_pre_soft_reset(void *handle) +static int gmc_v8_0_pre_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (!adev->gmc.srbm_soft_reset) return 0; gmc_v8_0_mc_stop(adev); - if (gmc_v8_0_wait_for_idle(adev)) + if (gmc_v8_0_wait_for_idle(ip_block)) dev_warn(adev->dev, "Wait for GMC idle timed out !\n"); return 0; } -static int gmc_v8_0_soft_reset(void *handle) +static int gmc_v8_0_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset; if (!adev->gmc.srbm_soft_reset) @@ -1361,9 +1368,9 @@ static int gmc_v8_0_soft_reset(void *handle) return 0; } -static int gmc_v8_0_post_soft_reset(void *handle) +static int gmc_v8_0_post_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (!adev->gmc.srbm_soft_reset) return 0; @@ -1715,8 +1722,6 @@ static const struct amd_ip_funcs gmc_v8_0_ip_funcs = { .set_clockgating_state = gmc_v8_0_set_clockgating_state, .set_powergating_state = gmc_v8_0_set_powergating_state, .get_clockgating_state = gmc_v8_0_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_gmc_funcs gmc_v8_0_gmc_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index c76ac0dfe572..f43ded8a0aab 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -672,6 +672,12 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, (amdgpu_ip_version(adev, GC_HWIP, 0) >= IP_VERSION(9, 4, 2))) return 0; + /* Only print L2 fault status if the status register could be read and + * contains useful information + */ + if (!status) + return 0; + if (!amdgpu_sriov_vf(adev)) WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1); @@ -1386,15 +1392,45 @@ gmc_v9_0_get_memory_partition(struct amdgpu_device *adev, u32 *supp_modes) return mode; } +static enum amdgpu_memory_partition +gmc_v9_0_query_vf_memory_partition(struct amdgpu_device *adev) +{ + switch (adev->gmc.num_mem_partitions) { + case 0: + return UNKNOWN_MEMORY_PARTITION_MODE; + case 1: + return AMDGPU_NPS1_PARTITION_MODE; + case 2: + return AMDGPU_NPS2_PARTITION_MODE; + case 4: + return AMDGPU_NPS4_PARTITION_MODE; + default: + return AMDGPU_NPS1_PARTITION_MODE; + } + + return AMDGPU_NPS1_PARTITION_MODE; +} + static enum amdgpu_memory_partition gmc_v9_0_query_memory_partition(struct amdgpu_device *adev) { if (amdgpu_sriov_vf(adev)) - return AMDGPU_NPS1_PARTITION_MODE; + return gmc_v9_0_query_vf_memory_partition(adev); return gmc_v9_0_get_memory_partition(adev, NULL); } +static bool gmc_v9_0_need_reset_on_init(struct amdgpu_device *adev) +{ + if (adev->nbio.funcs && adev->nbio.funcs->is_nps_switch_requested && + adev->nbio.funcs->is_nps_switch_requested(adev)) { + adev->gmc.reset_flags |= AMDGPU_GMC_INIT_RESET_NPS; + return true; + } + + return false; +} + static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = { .flush_gpu_tlb = gmc_v9_0_flush_gpu_tlb, .flush_gpu_tlb_pasid = gmc_v9_0_flush_gpu_tlb_pasid, @@ -1406,6 +1442,8 @@ static const struct amdgpu_gmc_funcs gmc_v9_0_gmc_funcs = { .override_vm_pte_flags = gmc_v9_0_override_vm_pte_flags, .get_vbios_fb_size = gmc_v9_0_get_vbios_fb_size, .query_mem_partition_mode = &gmc_v9_0_query_memory_partition, + .request_mem_partition_mode = &amdgpu_gmc_request_memory_partition, + .need_reset_on_init = &gmc_v9_0_need_reset_on_init, }; static void gmc_v9_0_set_gmc_funcs(struct amdgpu_device *adev) @@ -1545,9 +1583,31 @@ static void gmc_v9_0_set_xgmi_ras_funcs(struct amdgpu_device *adev) adev->gmc.xgmi.ras = &xgmi_ras; } -static int gmc_v9_0_early_init(void *handle) +static void gmc_v9_0_init_nps_details(struct amdgpu_device *adev) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->gmc.supported_nps_modes = 0; + + if (amdgpu_sriov_vf(adev) || (adev->flags & AMD_IS_APU)) + return; + + /*TODO: Check PSP version also which supports NPS switch. Otherwise keep + * supported modes as 0. + */ + switch (amdgpu_ip_version(adev, GC_HWIP, 0)) { + case IP_VERSION(9, 4, 3): + case IP_VERSION(9, 4, 4): + adev->gmc.supported_nps_modes = + BIT(AMDGPU_NPS1_PARTITION_MODE) | + BIT(AMDGPU_NPS4_PARTITION_MODE); + break; + default: + break; + } +} + +static int gmc_v9_0_early_init(struct amdgpu_ip_block *ip_block) +{ + struct amdgpu_device *adev = ip_block->adev; /* * 9.4.0, 9.4.1 and 9.4.3 don't have XGMI defined @@ -1601,9 +1661,9 @@ static int gmc_v9_0_early_init(void *handle) return 0; } -static int gmc_v9_0_late_init(void *handle) +static int gmc_v9_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_gmc_allocate_vm_inv_eng(adev); @@ -1900,6 +1960,8 @@ gmc_v9_0_init_sw_mem_ranges(struct amdgpu_device *adev, switch (mode) { case UNKNOWN_MEMORY_PARTITION_MODE: + adev->gmc.num_mem_partitions = 0; + break; case AMDGPU_NPS1_PARTITION_MODE: adev->gmc.num_mem_partitions = 1; break; @@ -1919,7 +1981,7 @@ gmc_v9_0_init_sw_mem_ranges(struct amdgpu_device *adev, /* Use NPS range info, if populated */ r = amdgpu_gmc_get_nps_memranges(adev, mem_ranges, - adev->gmc.num_mem_partitions); + &adev->gmc.num_mem_partitions); if (!r) { l = 0; for (i = 1; i < adev->gmc.num_mem_partitions; ++i) { @@ -1929,6 +1991,11 @@ gmc_v9_0_init_sw_mem_ranges(struct amdgpu_device *adev, } } else { + if (!adev->gmc.num_mem_partitions) { + dev_err(adev->dev, + "Not able to detect NPS mode, fall back to NPS1"); + adev->gmc.num_mem_partitions = 1; + } /* Fallback to sw based calculation */ size = (adev->gmc.real_vram_size + SZ_16M) >> AMDGPU_GPU_PAGE_SHIFT; size /= adev->gmc.num_mem_partitions; @@ -1987,10 +2054,10 @@ static void gmc_v9_4_3_init_vram_info(struct amdgpu_device *adev) adev->gmc.vram_width = 128 * 64; } -static int gmc_v9_0_sw_init(void *handle) +static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block) { int r, vram_width = 0, vram_type = 0, vram_vendor = 0, dma_addr_bits; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; unsigned long inst_mask = adev->aid_mask; adev->gfxhub.funcs->init(adev); @@ -2165,6 +2232,7 @@ static int gmc_v9_0_sw_init(void *handle) if (r) return r; + gmc_v9_0_init_nps_details(adev); /* * number of VMs * VMID 0 is reserved for System @@ -2198,9 +2266,9 @@ static int gmc_v9_0_sw_init(void *handle) return 0; } -static int gmc_v9_0_sw_fini(void *handle) +static int gmc_v9_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 3) || amdgpu_ip_version(adev, GC_HWIP, 0) == IP_VERSION(9, 4, 4)) @@ -2308,9 +2376,9 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev) return 0; } -static int gmc_v9_0_hw_init(void *handle) +static int gmc_v9_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; bool value; int i, r; @@ -2393,9 +2461,9 @@ static void gmc_v9_0_gart_disable(struct amdgpu_device *adev) adev->mmhub.funcs->gart_disable(adev); } -static int gmc_v9_0_hw_fini(void *handle) +static int gmc_v9_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; gmc_v9_0_gart_disable(adev); @@ -2413,32 +2481,44 @@ static int gmc_v9_0_hw_fini(void *handle) if (adev->mmhub.funcs->update_power_gating) adev->mmhub.funcs->update_power_gating(adev, false); - amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0); + /* + * For minimal init, late_init is not called, hence VM fault/RAS irqs + * are not enabled. + */ + if (adev->init_lvl->level != AMDGPU_INIT_LEVEL_MINIMAL_XGMI) { + amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0); - if (adev->gmc.ecc_irq.funcs && - amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC)) - amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0); + if (adev->gmc.ecc_irq.funcs && + amdgpu_ras_is_supported(adev, AMDGPU_RAS_BLOCK__UMC)) + amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0); + } return 0; } -static int gmc_v9_0_suspend(void *handle) +static int gmc_v9_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return gmc_v9_0_hw_fini(adev); + return gmc_v9_0_hw_fini(ip_block); } -static int gmc_v9_0_resume(void *handle) +static int gmc_v9_0_resume(struct amdgpu_ip_block *ip_block) { + struct amdgpu_device *adev = ip_block->adev; int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = gmc_v9_0_hw_init(adev); + /* If a reset is done for NPS mode switch, read the memory range + * information again. + */ + if (adev->gmc.reset_flags & AMDGPU_GMC_INIT_RESET_NPS) { + gmc_v9_0_init_sw_mem_ranges(adev, adev->gmc.mem_partitions); + adev->gmc.reset_flags &= ~AMDGPU_GMC_INIT_RESET_NPS; + } + + r = gmc_v9_0_hw_init(ip_block); if (r) return r; - amdgpu_vmid_reset_all(adev); + amdgpu_vmid_reset_all(ip_block->adev); return 0; } @@ -2449,13 +2529,13 @@ static bool gmc_v9_0_is_idle(void *handle) return true; } -static int gmc_v9_0_wait_for_idle(void *handle) +static int gmc_v9_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* There is no need to wait for MC idle in GMC v9.*/ return 0; } -static int gmc_v9_0_soft_reset(void *handle) +static int gmc_v9_0_soft_reset(struct amdgpu_ip_block *ip_block) { /* XXX for emulation.*/ return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/iceland_ih.c b/drivers/gpu/drm/amd/amdgpu/iceland_ih.c index 07984f7c3ae7..7f45e93c0397 100644 --- a/drivers/gpu/drm/amd/amdgpu/iceland_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/iceland_ih.c @@ -273,9 +273,9 @@ static void iceland_ih_set_rptr(struct amdgpu_device *adev, WREG32(mmIH_RB_RPTR, ih->rptr); } -static int iceland_ih_early_init(void *handle) +static int iceland_ih_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; ret = amdgpu_irq_add_domain(adev); @@ -287,10 +287,10 @@ static int iceland_ih_early_init(void *handle) return 0; } -static int iceland_ih_sw_init(void *handle) +static int iceland_ih_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 64 * 1024, false); if (r) @@ -301,9 +301,9 @@ static int iceland_ih_sw_init(void *handle) return r; } -static int iceland_ih_sw_fini(void *handle) +static int iceland_ih_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_fini_sw(adev); amdgpu_irq_remove_domain(adev); @@ -311,34 +311,28 @@ static int iceland_ih_sw_fini(void *handle) return 0; } -static int iceland_ih_hw_init(void *handle) +static int iceland_ih_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return iceland_ih_irq_init(adev); } -static int iceland_ih_hw_fini(void *handle) +static int iceland_ih_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - iceland_ih_irq_disable(adev); + iceland_ih_irq_disable(ip_block->adev); return 0; } -static int iceland_ih_suspend(void *handle) +static int iceland_ih_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return iceland_ih_hw_fini(adev); + return iceland_ih_hw_fini(ip_block); } -static int iceland_ih_resume(void *handle) +static int iceland_ih_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return iceland_ih_hw_init(adev); + return iceland_ih_hw_init(ip_block); } static bool iceland_ih_is_idle(void *handle) @@ -352,11 +346,11 @@ static bool iceland_ih_is_idle(void *handle) return true; } -static int iceland_ih_wait_for_idle(void *handle) +static int iceland_ih_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { /* read MC_STATUS */ @@ -368,10 +362,10 @@ static int iceland_ih_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int iceland_ih_soft_reset(void *handle) +static int iceland_ih_soft_reset(struct amdgpu_ip_block *ip_block) { u32 srbm_soft_reset = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 tmp = RREG32(mmSRBM_STATUS); if (tmp & SRBM_STATUS__IH_BUSY_MASK) @@ -413,7 +407,6 @@ static int iceland_ih_set_powergating_state(void *handle, static const struct amd_ip_funcs iceland_ih_ip_funcs = { .name = "iceland_ih", .early_init = iceland_ih_early_init, - .late_init = NULL, .sw_init = iceland_ih_sw_init, .sw_fini = iceland_ih_sw_fini, .hw_init = iceland_ih_hw_init, @@ -425,8 +418,6 @@ static const struct amd_ip_funcs iceland_ih_ip_funcs = { .soft_reset = iceland_ih_soft_reset, .set_clockgating_state = iceland_ih_set_clockgating_state, .set_powergating_state = iceland_ih_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ih_funcs iceland_ih_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c index 18a761d6ef33..38f953fd65d9 100644 --- a/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_0.c @@ -559,19 +559,19 @@ static void ih_v6_0_set_self_irq_funcs(struct amdgpu_device *adev) adev->irq.self_irq.funcs = &ih_v6_0_self_irq_funcs; } -static int ih_v6_0_early_init(void *handle) +static int ih_v6_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; ih_v6_0_set_interrupt_funcs(adev); ih_v6_0_set_self_irq_funcs(adev); return 0; } -static int ih_v6_0_sw_init(void *handle) +static int ih_v6_0_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; bool use_bus_addr; r = amdgpu_irq_add_id(adev, SOC21_IH_CLIENTID_IH, 0, @@ -614,19 +614,19 @@ static int ih_v6_0_sw_init(void *handle) return r; } -static int ih_v6_0_sw_fini(void *handle) +static int ih_v6_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_fini_sw(adev); return 0; } -static int ih_v6_0_hw_init(void *handle) +static int ih_v6_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = ih_v6_0_irq_init(adev); if (r) @@ -635,27 +635,21 @@ static int ih_v6_0_hw_init(void *handle) return 0; } -static int ih_v6_0_hw_fini(void *handle) +static int ih_v6_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - ih_v6_0_irq_disable(adev); + ih_v6_0_irq_disable(ip_block->adev); return 0; } -static int ih_v6_0_suspend(void *handle) +static int ih_v6_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return ih_v6_0_hw_fini(adev); + return ih_v6_0_hw_fini(ip_block); } -static int ih_v6_0_resume(void *handle) +static int ih_v6_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return ih_v6_0_hw_init(adev); + return ih_v6_0_hw_init(ip_block); } static bool ih_v6_0_is_idle(void *handle) @@ -664,13 +658,13 @@ static bool ih_v6_0_is_idle(void *handle) return true; } -static int ih_v6_0_wait_for_idle(void *handle) +static int ih_v6_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* todo */ return -ETIMEDOUT; } -static int ih_v6_0_soft_reset(void *handle) +static int ih_v6_0_soft_reset(struct amdgpu_ip_block *ip_block) { /* todo */ return 0; @@ -785,7 +779,6 @@ static void ih_v6_0_get_clockgating_state(void *handle, u64 *flags) static const struct amd_ip_funcs ih_v6_0_ip_funcs = { .name = "ih_v6_0", .early_init = ih_v6_0_early_init, - .late_init = NULL, .sw_init = ih_v6_0_sw_init, .sw_fini = ih_v6_0_sw_fini, .hw_init = ih_v6_0_hw_init, @@ -798,8 +791,6 @@ static const struct amd_ip_funcs ih_v6_0_ip_funcs = { .set_clockgating_state = ih_v6_0_set_clockgating_state, .set_powergating_state = ih_v6_0_set_powergating_state, .get_clockgating_state = ih_v6_0_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ih_funcs ih_v6_0_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v6_1.c b/drivers/gpu/drm/amd/amdgpu/ih_v6_1.c index 2e0469feca1e..61381e0c3795 100644 --- a/drivers/gpu/drm/amd/amdgpu/ih_v6_1.c +++ b/drivers/gpu/drm/amd/amdgpu/ih_v6_1.c @@ -532,9 +532,9 @@ static void ih_v6_1_set_self_irq_funcs(struct amdgpu_device *adev) adev->irq.self_irq.funcs = &ih_v6_1_self_irq_funcs; } -static int ih_v6_1_early_init(void *handle) +static int ih_v6_1_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; ret = amdgpu_irq_add_domain(adev); @@ -547,10 +547,10 @@ static int ih_v6_1_early_init(void *handle) return 0; } -static int ih_v6_1_sw_init(void *handle) +static int ih_v6_1_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; bool use_bus_addr; r = amdgpu_irq_add_id(adev, SOC21_IH_CLIENTID_IH, 0, @@ -593,19 +593,19 @@ static int ih_v6_1_sw_init(void *handle) return r; } -static int ih_v6_1_sw_fini(void *handle) +static int ih_v6_1_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_fini_sw(adev); return 0; } -static int ih_v6_1_hw_init(void *handle) +static int ih_v6_1_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = ih_v6_1_irq_init(adev); if (r) @@ -614,27 +614,21 @@ static int ih_v6_1_hw_init(void *handle) return 0; } -static int ih_v6_1_hw_fini(void *handle) +static int ih_v6_1_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - ih_v6_1_irq_disable(adev); + ih_v6_1_irq_disable(ip_block->adev); return 0; } -static int ih_v6_1_suspend(void *handle) +static int ih_v6_1_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return ih_v6_1_hw_fini(adev); + return ih_v6_1_hw_fini(ip_block); } -static int ih_v6_1_resume(void *handle) +static int ih_v6_1_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return ih_v6_1_hw_init(adev); + return ih_v6_1_hw_init(ip_block); } static bool ih_v6_1_is_idle(void *handle) @@ -643,13 +637,13 @@ static bool ih_v6_1_is_idle(void *handle) return true; } -static int ih_v6_1_wait_for_idle(void *handle) +static int ih_v6_1_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* todo */ return -ETIMEDOUT; } -static int ih_v6_1_soft_reset(void *handle) +static int ih_v6_1_soft_reset(struct amdgpu_ip_block *ip_block) { /* todo */ return 0; @@ -768,7 +762,6 @@ static void ih_v6_1_get_clockgating_state(void *handle, u64 *flags) static const struct amd_ip_funcs ih_v6_1_ip_funcs = { .name = "ih_v6_1", .early_init = ih_v6_1_early_init, - .late_init = NULL, .sw_init = ih_v6_1_sw_init, .sw_fini = ih_v6_1_sw_fini, .hw_init = ih_v6_1_hw_init, @@ -781,8 +774,6 @@ static const struct amd_ip_funcs ih_v6_1_ip_funcs = { .set_clockgating_state = ih_v6_1_set_clockgating_state, .set_powergating_state = ih_v6_1_set_powergating_state, .get_clockgating_state = ih_v6_1_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ih_funcs ih_v6_1_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/ih_v7_0.c b/drivers/gpu/drm/amd/amdgpu/ih_v7_0.c index 6852081fcff2..d2428cf5d385 100644 --- a/drivers/gpu/drm/amd/amdgpu/ih_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/ih_v7_0.c @@ -528,19 +528,19 @@ static void ih_v7_0_set_self_irq_funcs(struct amdgpu_device *adev) adev->irq.self_irq.funcs = &ih_v7_0_self_irq_funcs; } -static int ih_v7_0_early_init(void *handle) +static int ih_v7_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; ih_v7_0_set_interrupt_funcs(adev); ih_v7_0_set_self_irq_funcs(adev); return 0; } -static int ih_v7_0_sw_init(void *handle) +static int ih_v7_0_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; bool use_bus_addr; r = amdgpu_irq_add_id(adev, SOC21_IH_CLIENTID_IH, 0, @@ -583,19 +583,19 @@ static int ih_v7_0_sw_init(void *handle) return r; } -static int ih_v7_0_sw_fini(void *handle) +static int ih_v7_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_fini_sw(adev); return 0; } -static int ih_v7_0_hw_init(void *handle) +static int ih_v7_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = ih_v7_0_irq_init(adev); if (r) @@ -604,27 +604,21 @@ static int ih_v7_0_hw_init(void *handle) return 0; } -static int ih_v7_0_hw_fini(void *handle) +static int ih_v7_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - ih_v7_0_irq_disable(adev); + ih_v7_0_irq_disable(ip_block->adev); return 0; } -static int ih_v7_0_suspend(void *handle) +static int ih_v7_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return ih_v7_0_hw_fini(adev); + return ih_v7_0_hw_fini(ip_block); } -static int ih_v7_0_resume(void *handle) +static int ih_v7_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return ih_v7_0_hw_init(adev); + return ih_v7_0_hw_init(ip_block); } static bool ih_v7_0_is_idle(void *handle) @@ -633,13 +627,13 @@ static bool ih_v7_0_is_idle(void *handle) return true; } -static int ih_v7_0_wait_for_idle(void *handle) +static int ih_v7_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* todo */ return -ETIMEDOUT; } -static int ih_v7_0_soft_reset(void *handle) +static int ih_v7_0_soft_reset(struct amdgpu_ip_block *ip_block) { /* todo */ return 0; @@ -758,7 +752,6 @@ static void ih_v7_0_get_clockgating_state(void *handle, u64 *flags) static const struct amd_ip_funcs ih_v7_0_ip_funcs = { .name = "ih_v7_0", .early_init = ih_v7_0_early_init, - .late_init = NULL, .sw_init = ih_v7_0_sw_init, .sw_fini = ih_v7_0_sw_fini, .hw_init = ih_v7_0_hw_init, @@ -771,8 +764,6 @@ static const struct amd_ip_funcs ih_v7_0_ip_funcs = { .set_clockgating_state = ih_v7_0_set_clockgating_state, .set_powergating_state = ih_v7_0_set_powergating_state, .get_clockgating_state = ih_v7_0_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ih_funcs ih_v7_0_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c index 6e0e88076224..03b8b7cd5229 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.c @@ -458,13 +458,13 @@ static int jpeg_v1_0_process_interrupt(struct amdgpu_device *adev, /** * jpeg_v1_0_early_init - set function pointers * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers */ -int jpeg_v1_0_early_init(void *handle) +int jpeg_v1_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->jpeg.num_jpeg_inst = 1; adev->jpeg.num_jpeg_rings = 1; @@ -478,12 +478,12 @@ int jpeg_v1_0_early_init(void *handle) /** * jpeg_v1_0_sw_init - sw init for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -int jpeg_v1_0_sw_init(void *handle) +int jpeg_v1_0_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int r; @@ -509,13 +509,13 @@ int jpeg_v1_0_sw_init(void *handle) /** * jpeg_v1_0_sw_fini - sw fini for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * JPEG free up sw allocation */ -void jpeg_v1_0_sw_fini(void *handle) +void jpeg_v1_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_ring_fini(adev->jpeg.inst->ring_dec); } diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.h b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.h index 9654d22e0376..097328635083 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.h +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v1_0.h @@ -24,9 +24,9 @@ #ifndef __JPEG_V1_0_H__ #define __JPEG_V1_0_H__ -int jpeg_v1_0_early_init(void *handle); -int jpeg_v1_0_sw_init(void *handle); -void jpeg_v1_0_sw_fini(void *handle); +int jpeg_v1_0_early_init(struct amdgpu_ip_block *ip_block); +int jpeg_v1_0_sw_init(struct amdgpu_ip_block *ip_block); +void jpeg_v1_0_sw_fini(struct amdgpu_ip_block *ip_block); void jpeg_v1_0_start(struct amdgpu_device *adev, int mode); #define JPEG_V1_REG_RANGE_START 0x8000 diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c index 41c0f8750dc1..d6823fb45d32 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_0.c @@ -41,13 +41,13 @@ static int jpeg_v2_0_set_powergating_state(void *handle, /** * jpeg_v2_0_early_init - set function pointers * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers */ -static int jpeg_v2_0_early_init(void *handle) +static int jpeg_v2_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->jpeg.num_jpeg_inst = 1; adev->jpeg.num_jpeg_rings = 1; @@ -61,13 +61,13 @@ static int jpeg_v2_0_early_init(void *handle) /** * jpeg_v2_0_sw_init - sw init for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int jpeg_v2_0_sw_init(void *handle) +static int jpeg_v2_0_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int r; @@ -104,14 +104,14 @@ static int jpeg_v2_0_sw_init(void *handle) /** * jpeg_v2_0_sw_fini - sw fini for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * JPEG suspend and free up sw allocation */ -static int jpeg_v2_0_sw_fini(void *handle) +static int jpeg_v2_0_sw_fini(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_jpeg_suspend(adev); if (r) @@ -125,12 +125,12 @@ static int jpeg_v2_0_sw_fini(void *handle) /** * jpeg_v2_0_hw_init - start and test JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -static int jpeg_v2_0_hw_init(void *handle) +static int jpeg_v2_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell, @@ -142,13 +142,13 @@ static int jpeg_v2_0_hw_init(void *handle) /** * jpeg_v2_0_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the JPEG block, mark ring as not ready any more */ -static int jpeg_v2_0_hw_fini(void *handle) +static int jpeg_v2_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -162,20 +162,19 @@ static int jpeg_v2_0_hw_fini(void *handle) /** * jpeg_v2_0_suspend - suspend JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend JPEG block */ -static int jpeg_v2_0_suspend(void *handle) +static int jpeg_v2_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = jpeg_v2_0_hw_fini(adev); + r = jpeg_v2_0_hw_fini(ip_block); if (r) return r; - r = amdgpu_jpeg_suspend(adev); + r = amdgpu_jpeg_suspend(ip_block->adev); return r; } @@ -183,20 +182,19 @@ static int jpeg_v2_0_suspend(void *handle) /** * jpeg_v2_0_resume - resume JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init JPEG block */ -static int jpeg_v2_0_resume(void *handle) +static int jpeg_v2_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_jpeg_resume(adev); + r = amdgpu_jpeg_resume(ip_block->adev); if (r) return r; - r = jpeg_v2_0_hw_init(adev); + r = jpeg_v2_0_hw_init(ip_block); return r; } @@ -666,9 +664,9 @@ static bool jpeg_v2_0_is_idle(void *handle) UVD_JRBC_STATUS__RB_JOB_DONE_MASK); } -static int jpeg_v2_0_wait_for_idle(void *handle) +static int jpeg_v2_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; ret = SOC15_WAIT_ON_RREG(JPEG, 0, mmUVD_JRBC_STATUS, UVD_JRBC_STATUS__RB_JOB_DONE_MASK, @@ -744,7 +742,6 @@ static int jpeg_v2_0_process_interrupt(struct amdgpu_device *adev, static const struct amd_ip_funcs jpeg_v2_0_ip_funcs = { .name = "jpeg_v2_0", .early_init = jpeg_v2_0_early_init, - .late_init = NULL, .sw_init = jpeg_v2_0_sw_init, .sw_fini = jpeg_v2_0_sw_fini, .hw_init = jpeg_v2_0_hw_init, @@ -753,14 +750,8 @@ static const struct amd_ip_funcs jpeg_v2_0_ip_funcs = { .resume = jpeg_v2_0_resume, .is_idle = jpeg_v2_0_is_idle, .wait_for_idle = jpeg_v2_0_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = jpeg_v2_0_set_clockgating_state, .set_powergating_state = jpeg_v2_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs jpeg_v2_0_dec_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c index eedb9a829d95..5063a38801d6 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c @@ -50,13 +50,13 @@ static int amdgpu_ih_clientid_jpeg[] = { /** * jpeg_v2_5_early_init - set function pointers * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers */ -static int jpeg_v2_5_early_init(void *handle) +static int jpeg_v2_5_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 harvest; int i; @@ -81,15 +81,15 @@ static int jpeg_v2_5_early_init(void *handle) /** * jpeg_v2_5_sw_init - sw init for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int jpeg_v2_5_sw_init(void *handle) +static int jpeg_v2_5_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int i, r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { if (adev->jpeg.harvest_config & (1 << i)) @@ -153,14 +153,14 @@ static int jpeg_v2_5_sw_init(void *handle) /** * jpeg_v2_5_sw_fini - sw fini for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * JPEG suspend and free up sw allocation */ -static int jpeg_v2_5_sw_fini(void *handle) +static int jpeg_v2_5_sw_fini(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_jpeg_suspend(adev); if (r) @@ -174,12 +174,12 @@ static int jpeg_v2_5_sw_fini(void *handle) /** * jpeg_v2_5_hw_init - start and test JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -static int jpeg_v2_5_hw_init(void *handle) +static int jpeg_v2_5_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, r; @@ -202,13 +202,13 @@ static int jpeg_v2_5_hw_init(void *handle) /** * jpeg_v2_5_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the JPEG block, mark ring as not ready any more */ -static int jpeg_v2_5_hw_fini(void *handle) +static int jpeg_v2_5_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -231,20 +231,19 @@ static int jpeg_v2_5_hw_fini(void *handle) /** * jpeg_v2_5_suspend - suspend JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend JPEG block */ -static int jpeg_v2_5_suspend(void *handle) +static int jpeg_v2_5_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = jpeg_v2_5_hw_fini(adev); + r = jpeg_v2_5_hw_fini(ip_block); if (r) return r; - r = amdgpu_jpeg_suspend(adev); + r = amdgpu_jpeg_suspend(ip_block->adev); return r; } @@ -252,20 +251,19 @@ static int jpeg_v2_5_suspend(void *handle) /** * jpeg_v2_5_resume - resume JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init JPEG block */ -static int jpeg_v2_5_resume(void *handle) +static int jpeg_v2_5_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = amdgpu_jpeg_resume(adev); + r = amdgpu_jpeg_resume(ip_block->adev); if (r) return r; - r = jpeg_v2_5_hw_init(adev); + r = jpeg_v2_5_hw_init(ip_block); return r; } @@ -501,9 +499,9 @@ static bool jpeg_v2_5_is_idle(void *handle) return ret; } -static int jpeg_v2_5_wait_for_idle(void *handle) +static int jpeg_v2_5_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, ret; for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { @@ -615,7 +613,6 @@ static int jpeg_v2_5_process_interrupt(struct amdgpu_device *adev, static const struct amd_ip_funcs jpeg_v2_5_ip_funcs = { .name = "jpeg_v2_5", .early_init = jpeg_v2_5_early_init, - .late_init = NULL, .sw_init = jpeg_v2_5_sw_init, .sw_fini = jpeg_v2_5_sw_fini, .hw_init = jpeg_v2_5_hw_init, @@ -624,20 +621,13 @@ static const struct amd_ip_funcs jpeg_v2_5_ip_funcs = { .resume = jpeg_v2_5_resume, .is_idle = jpeg_v2_5_is_idle, .wait_for_idle = jpeg_v2_5_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = jpeg_v2_5_set_clockgating_state, .set_powergating_state = jpeg_v2_5_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amd_ip_funcs jpeg_v2_6_ip_funcs = { .name = "jpeg_v2_6", .early_init = jpeg_v2_5_early_init, - .late_init = NULL, .sw_init = jpeg_v2_5_sw_init, .sw_fini = jpeg_v2_5_sw_fini, .hw_init = jpeg_v2_5_hw_init, @@ -646,14 +636,8 @@ static const struct amd_ip_funcs jpeg_v2_6_ip_funcs = { .resume = jpeg_v2_5_resume, .is_idle = jpeg_v2_5_is_idle, .wait_for_idle = jpeg_v2_5_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = jpeg_v2_5_set_clockgating_state, .set_powergating_state = jpeg_v2_5_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs jpeg_v2_5_dec_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c index b1e7fd25afbc..10adbb7cbf53 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c @@ -42,13 +42,13 @@ static int jpeg_v3_0_set_powergating_state(void *handle, /** * jpeg_v3_0_early_init - set function pointers * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers */ -static int jpeg_v3_0_early_init(void *handle) +static int jpeg_v3_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 harvest; @@ -75,13 +75,13 @@ static int jpeg_v3_0_early_init(void *handle) /** * jpeg_v3_0_sw_init - sw init for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int jpeg_v3_0_sw_init(void *handle) +static int jpeg_v3_0_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int r; @@ -118,13 +118,13 @@ static int jpeg_v3_0_sw_init(void *handle) /** * jpeg_v3_0_sw_fini - sw fini for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * JPEG suspend and free up sw allocation */ -static int jpeg_v3_0_sw_fini(void *handle) +static int jpeg_v3_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_jpeg_suspend(adev); @@ -139,12 +139,12 @@ static int jpeg_v3_0_sw_fini(void *handle) /** * jpeg_v3_0_hw_init - start and test JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -static int jpeg_v3_0_hw_init(void *handle) +static int jpeg_v3_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; adev->nbio.funcs->vcn_doorbell_range(adev, ring->use_doorbell, @@ -156,13 +156,13 @@ static int jpeg_v3_0_hw_init(void *handle) /** * jpeg_v3_0_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the JPEG block, mark ring as not ready any more */ -static int jpeg_v3_0_hw_fini(void *handle) +static int jpeg_v3_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -176,20 +176,19 @@ static int jpeg_v3_0_hw_fini(void *handle) /** * jpeg_v3_0_suspend - suspend JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend JPEG block */ -static int jpeg_v3_0_suspend(void *handle) +static int jpeg_v3_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = jpeg_v3_0_hw_fini(adev); + r = jpeg_v3_0_hw_fini(ip_block); if (r) return r; - r = amdgpu_jpeg_suspend(adev); + r = amdgpu_jpeg_suspend(ip_block->adev); return r; } @@ -197,20 +196,19 @@ static int jpeg_v3_0_suspend(void *handle) /** * jpeg_v3_0_resume - resume JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init JPEG block */ -static int jpeg_v3_0_resume(void *handle) +static int jpeg_v3_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = amdgpu_jpeg_resume(adev); + r = amdgpu_jpeg_resume(ip_block->adev); if (r) return r; - r = jpeg_v3_0_hw_init(adev); + r = jpeg_v3_0_hw_init(ip_block); return r; } @@ -459,9 +457,9 @@ static bool jpeg_v3_0_is_idle(void *handle) return ret; } -static int jpeg_v3_0_wait_for_idle(void *handle) +static int jpeg_v3_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return SOC15_WAIT_ON_RREG(JPEG, 0, mmUVD_JRBC_STATUS, UVD_JRBC_STATUS__RB_JOB_DONE_MASK, @@ -535,7 +533,6 @@ static int jpeg_v3_0_process_interrupt(struct amdgpu_device *adev, static const struct amd_ip_funcs jpeg_v3_0_ip_funcs = { .name = "jpeg_v3_0", .early_init = jpeg_v3_0_early_init, - .late_init = NULL, .sw_init = jpeg_v3_0_sw_init, .sw_fini = jpeg_v3_0_sw_fini, .hw_init = jpeg_v3_0_hw_init, @@ -544,14 +541,8 @@ static const struct amd_ip_funcs jpeg_v3_0_ip_funcs = { .resume = jpeg_v3_0_resume, .is_idle = jpeg_v3_0_is_idle, .wait_for_idle = jpeg_v3_0_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = jpeg_v3_0_set_clockgating_state, .set_powergating_state = jpeg_v3_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs jpeg_v3_0_dec_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c index 6c5c1a68a9b7..89953c0f5f1f 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c @@ -48,13 +48,13 @@ static void jpeg_v4_0_dec_ring_set_wptr(struct amdgpu_ring *ring); /** * jpeg_v4_0_early_init - set function pointers * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers */ -static int jpeg_v4_0_early_init(void *handle) +static int jpeg_v4_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->jpeg.num_jpeg_inst = 1; @@ -70,13 +70,13 @@ static int jpeg_v4_0_early_init(void *handle) /** * jpeg_v4_0_sw_init - sw init for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int jpeg_v4_0_sw_init(void *handle) +static int jpeg_v4_0_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int r; @@ -130,13 +130,13 @@ static int jpeg_v4_0_sw_init(void *handle) /** * jpeg_v4_0_sw_fini - sw fini for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * JPEG suspend and free up sw allocation */ -static int jpeg_v4_0_sw_fini(void *handle) +static int jpeg_v4_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_jpeg_suspend(adev); @@ -151,12 +151,12 @@ static int jpeg_v4_0_sw_fini(void *handle) /** * jpeg_v4_0_hw_init - start and test JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -static int jpeg_v4_0_hw_init(void *handle) +static int jpeg_v4_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; int r; @@ -187,13 +187,13 @@ static int jpeg_v4_0_hw_init(void *handle) /** * jpeg_v4_0_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the JPEG block, mark ring as not ready any more */ -static int jpeg_v4_0_hw_fini(void *handle) +static int jpeg_v4_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->vcn.idle_work); if (!amdgpu_sriov_vf(adev)) { @@ -210,20 +210,19 @@ static int jpeg_v4_0_hw_fini(void *handle) /** * jpeg_v4_0_suspend - suspend JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend JPEG block */ -static int jpeg_v4_0_suspend(void *handle) +static int jpeg_v4_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = jpeg_v4_0_hw_fini(adev); + r = jpeg_v4_0_hw_fini(ip_block); if (r) return r; - r = amdgpu_jpeg_suspend(adev); + r = amdgpu_jpeg_suspend(ip_block->adev); return r; } @@ -231,20 +230,19 @@ static int jpeg_v4_0_suspend(void *handle) /** * jpeg_v4_0_resume - resume JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init JPEG block */ -static int jpeg_v4_0_resume(void *handle) +static int jpeg_v4_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = amdgpu_jpeg_resume(adev); + r = amdgpu_jpeg_resume(ip_block->adev); if (r) return r; - r = jpeg_v4_0_hw_init(adev); + r = jpeg_v4_0_hw_init(ip_block); return r; } @@ -621,9 +619,9 @@ static bool jpeg_v4_0_is_idle(void *handle) return ret; } -static int jpeg_v4_0_wait_for_idle(void *handle) +static int jpeg_v4_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return SOC15_WAIT_ON_RREG(JPEG, 0, regUVD_JRBC_STATUS, UVD_JRBC_STATUS__RB_JOB_DONE_MASK, @@ -702,7 +700,6 @@ static int jpeg_v4_0_process_interrupt(struct amdgpu_device *adev, static const struct amd_ip_funcs jpeg_v4_0_ip_funcs = { .name = "jpeg_v4_0", .early_init = jpeg_v4_0_early_init, - .late_init = NULL, .sw_init = jpeg_v4_0_sw_init, .sw_fini = jpeg_v4_0_sw_fini, .hw_init = jpeg_v4_0_hw_init, @@ -711,14 +708,8 @@ static const struct amd_ip_funcs jpeg_v4_0_ip_funcs = { .resume = jpeg_v4_0_resume, .is_idle = jpeg_v4_0_is_idle, .wait_for_idle = jpeg_v4_0_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = jpeg_v4_0_set_clockgating_state, .set_powergating_state = jpeg_v4_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs jpeg_v4_0_dec_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c index 86958cb2c2ab..6917e4a8e96a 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_3.c @@ -68,13 +68,13 @@ static inline bool jpeg_v4_0_3_normalizn_reqd(struct amdgpu_device *adev) /** * jpeg_v4_0_3_early_init - set function pointers * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers */ -static int jpeg_v4_0_3_early_init(void *handle) +static int jpeg_v4_0_3_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->jpeg.num_jpeg_rings = AMDGPU_MAX_JPEG_RINGS; @@ -88,13 +88,13 @@ static int jpeg_v4_0_3_early_init(void *handle) /** * jpeg_v4_0_3_sw_init - sw init for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int jpeg_v4_0_3_sw_init(void *handle) +static int jpeg_v4_0_3_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, j, r, jpeg_inst; @@ -165,13 +165,13 @@ static int jpeg_v4_0_3_sw_init(void *handle) /** * jpeg_v4_0_3_sw_fini - sw fini for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * JPEG suspend and free up sw allocation */ -static int jpeg_v4_0_3_sw_fini(void *handle) +static int jpeg_v4_0_3_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_jpeg_suspend(adev); @@ -299,12 +299,12 @@ static int jpeg_v4_0_3_start_sriov(struct amdgpu_device *adev) /** * jpeg_v4_0_3_hw_init - start and test JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -static int jpeg_v4_0_3_hw_init(void *handle) +static int jpeg_v4_0_3_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, j, r, jpeg_inst; @@ -358,13 +358,13 @@ static int jpeg_v4_0_3_hw_init(void *handle) /** * jpeg_v4_0_3_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the JPEG block, mark ring as not ready any more */ -static int jpeg_v4_0_3_hw_fini(void *handle) +static int jpeg_v4_0_3_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret = 0; cancel_delayed_work_sync(&adev->jpeg.idle_work); @@ -380,20 +380,19 @@ static int jpeg_v4_0_3_hw_fini(void *handle) /** * jpeg_v4_0_3_suspend - suspend JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend JPEG block */ -static int jpeg_v4_0_3_suspend(void *handle) +static int jpeg_v4_0_3_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = jpeg_v4_0_3_hw_fini(adev); + r = jpeg_v4_0_3_hw_fini(ip_block); if (r) return r; - r = amdgpu_jpeg_suspend(adev); + r = amdgpu_jpeg_suspend(ip_block->adev); return r; } @@ -401,20 +400,19 @@ static int jpeg_v4_0_3_suspend(void *handle) /** * jpeg_v4_0_3_resume - resume JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init JPEG block */ -static int jpeg_v4_0_3_resume(void *handle) +static int jpeg_v4_0_3_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = amdgpu_jpeg_resume(adev); + r = amdgpu_jpeg_resume(ip_block->adev); if (r) return r; - r = jpeg_v4_0_3_hw_init(adev); + r = jpeg_v4_0_3_hw_init(ip_block); return r; } @@ -674,11 +672,12 @@ void jpeg_v4_0_3_dec_ring_insert_start(struct amdgpu_ring *ring) amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET, 0, 0, PACKETJ_TYPE0)); amdgpu_ring_write(ring, 0x62a04); /* PCTL0_MMHUB_DEEPSLEEP_IB */ - } - amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR, - 0, 0, PACKETJ_TYPE0)); - amdgpu_ring_write(ring, 0x80004000); + amdgpu_ring_write(ring, + PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR, 0, + 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x80004000); + } } /** @@ -694,11 +693,12 @@ void jpeg_v4_0_3_dec_ring_insert_end(struct amdgpu_ring *ring) amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET, 0, 0, PACKETJ_TYPE0)); amdgpu_ring_write(ring, 0x62a04); - } - amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR, - 0, 0, PACKETJ_TYPE0)); - amdgpu_ring_write(ring, 0x00004000); + amdgpu_ring_write(ring, + PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR, 0, + 0, PACKETJ_TYPE0)); + amdgpu_ring_write(ring, 0x00004000); + } } /** @@ -743,14 +743,6 @@ void jpeg_v4_0_3_dec_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6)); amdgpu_ring_write(ring, 0); - amdgpu_ring_write(ring, PACKETJ(regUVD_JRBC_EXTERNAL_REG_INTERNAL_OFFSET, - 0, 0, PACKETJ_TYPE0)); - amdgpu_ring_write(ring, 0x3fbc); - - amdgpu_ring_write(ring, PACKETJ(JRBC_DEC_EXTERNAL_REG_WRITE_ADDR, - 0, 0, PACKETJ_TYPE0)); - amdgpu_ring_write(ring, 0x1); - amdgpu_ring_write(ring, PACKETJ(0, 0, 0, PACKETJ_TYPE6)); amdgpu_ring_write(ring, 0); @@ -929,9 +921,9 @@ static bool jpeg_v4_0_3_is_idle(void *handle) return ret; } -static int jpeg_v4_0_3_wait_for_idle(void *handle) +static int jpeg_v4_0_3_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret = 0; int i, j; @@ -1058,7 +1050,6 @@ static int jpeg_v4_0_3_process_interrupt(struct amdgpu_device *adev, static const struct amd_ip_funcs jpeg_v4_0_3_ip_funcs = { .name = "jpeg_v4_0_3", .early_init = jpeg_v4_0_3_early_init, - .late_init = NULL, .sw_init = jpeg_v4_0_3_sw_init, .sw_fini = jpeg_v4_0_3_sw_fini, .hw_init = jpeg_v4_0_3_hw_init, @@ -1067,14 +1058,8 @@ static const struct amd_ip_funcs jpeg_v4_0_3_ip_funcs = { .resume = jpeg_v4_0_3_resume, .is_idle = jpeg_v4_0_3_is_idle, .wait_for_idle = jpeg_v4_0_3_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = jpeg_v4_0_3_set_clockgating_state, .set_powergating_state = jpeg_v4_0_3_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = { @@ -1088,7 +1073,7 @@ static const struct amdgpu_ring_funcs jpeg_v4_0_3_dec_ring_vm_funcs = { SOC15_FLUSH_GPU_TLB_NUM_WREG * 6 + SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 8 + 8 + /* jpeg_v4_0_3_dec_ring_emit_vm_flush */ - 22 + 22 + /* jpeg_v4_0_3_dec_ring_emit_fence x2 vm fence */ + 18 + 18 + /* jpeg_v4_0_3_dec_ring_emit_fence x2 vm fence */ 8 + 16, .emit_ib_size = 22, /* jpeg_v4_0_3_dec_ring_emit_ib */ .emit_ib = jpeg_v4_0_3_dec_ring_emit_ib, diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c index 44eeed445ea9..f3cce523f3cb 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v4_0_5.c @@ -61,13 +61,13 @@ static int amdgpu_ih_clientid_jpeg[] = { /** * jpeg_v4_0_5_early_init - set function pointers * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers */ -static int jpeg_v4_0_5_early_init(void *handle) +static int jpeg_v4_0_5_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; switch (amdgpu_ip_version(adev, UVD_HWIP, 0)) { case IP_VERSION(4, 0, 5): @@ -94,13 +94,13 @@ static int jpeg_v4_0_5_early_init(void *handle) /** * jpeg_v4_0_5_sw_init - sw init for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int jpeg_v4_0_5_sw_init(void *handle) +static int jpeg_v4_0_5_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int r, i; @@ -159,13 +159,13 @@ static int jpeg_v4_0_5_sw_init(void *handle) /** * jpeg_v4_0_5_sw_fini - sw fini for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * JPEG suspend and free up sw allocation */ -static int jpeg_v4_0_5_sw_fini(void *handle) +static int jpeg_v4_0_5_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_jpeg_suspend(adev); @@ -180,12 +180,12 @@ static int jpeg_v4_0_5_sw_fini(void *handle) /** * jpeg_v4_0_5_hw_init - start and test JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -static int jpeg_v4_0_5_hw_init(void *handle) +static int jpeg_v4_0_5_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, r = 0; @@ -210,13 +210,13 @@ static int jpeg_v4_0_5_hw_init(void *handle) /** * jpeg_v4_0_5_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the JPEG block, mark ring as not ready any more */ -static int jpeg_v4_0_5_hw_fini(void *handle) +static int jpeg_v4_0_5_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -237,20 +237,19 @@ static int jpeg_v4_0_5_hw_fini(void *handle) /** * jpeg_v4_0_5_suspend - suspend JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend JPEG block */ -static int jpeg_v4_0_5_suspend(void *handle) +static int jpeg_v4_0_5_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = jpeg_v4_0_5_hw_fini(adev); + r = jpeg_v4_0_5_hw_fini(ip_block); if (r) return r; - r = amdgpu_jpeg_suspend(adev); + r = amdgpu_jpeg_suspend(ip_block->adev); return r; } @@ -258,20 +257,19 @@ static int jpeg_v4_0_5_suspend(void *handle) /** * jpeg_v4_0_5_resume - resume JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init JPEG block */ -static int jpeg_v4_0_5_resume(void *handle) +static int jpeg_v4_0_5_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = amdgpu_jpeg_resume(adev); + r = amdgpu_jpeg_resume(ip_block->adev); if (r) return r; - r = jpeg_v4_0_5_hw_init(adev); + r = jpeg_v4_0_5_hw_init(ip_block); return r; } @@ -637,9 +635,9 @@ static bool jpeg_v4_0_5_is_idle(void *handle) return ret; } -static int jpeg_v4_0_5_wait_for_idle(void *handle) +static int jpeg_v4_0_5_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { @@ -743,7 +741,6 @@ static int jpeg_v4_0_5_process_interrupt(struct amdgpu_device *adev, static const struct amd_ip_funcs jpeg_v4_0_5_ip_funcs = { .name = "jpeg_v4_0_5", .early_init = jpeg_v4_0_5_early_init, - .late_init = NULL, .sw_init = jpeg_v4_0_5_sw_init, .sw_fini = jpeg_v4_0_5_sw_fini, .hw_init = jpeg_v4_0_5_hw_init, @@ -752,14 +749,8 @@ static const struct amd_ip_funcs jpeg_v4_0_5_ip_funcs = { .resume = jpeg_v4_0_5_resume, .is_idle = jpeg_v4_0_5_is_idle, .wait_for_idle = jpeg_v4_0_5_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = jpeg_v4_0_5_set_clockgating_state, .set_powergating_state = jpeg_v4_0_5_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs jpeg_v4_0_5_dec_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c index d662aa841f97..06840d1dae79 100644 --- a/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c +++ b/drivers/gpu/drm/amd/amdgpu/jpeg_v5_0_0.c @@ -42,13 +42,13 @@ static int jpeg_v5_0_0_set_powergating_state(void *handle, /** * jpeg_v5_0_0_early_init - set function pointers * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers */ -static int jpeg_v5_0_0_early_init(void *handle) +static int jpeg_v5_0_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->jpeg.num_jpeg_inst = 1; adev->jpeg.num_jpeg_rings = 1; @@ -62,13 +62,13 @@ static int jpeg_v5_0_0_early_init(void *handle) /** * jpeg_v5_0_0_sw_init - sw init for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int jpeg_v5_0_0_sw_init(void *handle) +static int jpeg_v5_0_0_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int r; @@ -106,13 +106,13 @@ static int jpeg_v5_0_0_sw_init(void *handle) /** * jpeg_v5_0_0_sw_fini - sw fini for JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * JPEG suspend and free up sw allocation */ -static int jpeg_v5_0_0_sw_fini(void *handle) +static int jpeg_v5_0_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_jpeg_suspend(adev); @@ -127,12 +127,12 @@ static int jpeg_v5_0_0_sw_fini(void *handle) /** * jpeg_v5_0_0_hw_init - start and test JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * */ -static int jpeg_v5_0_0_hw_init(void *handle) +static int jpeg_v5_0_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring = adev->jpeg.inst->ring_dec; int r; @@ -153,13 +153,13 @@ static int jpeg_v5_0_0_hw_init(void *handle) /** * jpeg_v5_0_0_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the JPEG block, mark ring as not ready any more */ -static int jpeg_v5_0_0_hw_fini(void *handle) +static int jpeg_v5_0_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -173,20 +173,19 @@ static int jpeg_v5_0_0_hw_fini(void *handle) /** * jpeg_v5_0_0_suspend - suspend JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend JPEG block */ -static int jpeg_v5_0_0_suspend(void *handle) +static int jpeg_v5_0_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = jpeg_v5_0_0_hw_fini(adev); + r = jpeg_v5_0_0_hw_fini(ip_block); if (r) return r; - r = amdgpu_jpeg_suspend(adev); + r = amdgpu_jpeg_suspend(ip_block->adev); return r; } @@ -194,20 +193,19 @@ static int jpeg_v5_0_0_suspend(void *handle) /** * jpeg_v5_0_0_resume - resume JPEG block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init JPEG block */ -static int jpeg_v5_0_0_resume(void *handle) +static int jpeg_v5_0_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = amdgpu_jpeg_resume(adev); + r = amdgpu_jpeg_resume(ip_block->adev); if (r) return r; - r = jpeg_v5_0_0_hw_init(adev); + r = jpeg_v5_0_0_hw_init(ip_block); return r; } @@ -546,9 +544,9 @@ static bool jpeg_v5_0_0_is_idle(void *handle) return ret; } -static int jpeg_v5_0_0_wait_for_idle(void *handle) +static int jpeg_v5_0_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return SOC15_WAIT_ON_RREG(JPEG, 0, regUVD_JRBC_STATUS, UVD_JRBC_STATUS__RB_JOB_DONE_MASK, @@ -622,7 +620,6 @@ static int jpeg_v5_0_0_process_interrupt(struct amdgpu_device *adev, static const struct amd_ip_funcs jpeg_v5_0_0_ip_funcs = { .name = "jpeg_v5_0_0", .early_init = jpeg_v5_0_0_early_init, - .late_init = NULL, .sw_init = jpeg_v5_0_0_sw_init, .sw_fini = jpeg_v5_0_0_sw_fini, .hw_init = jpeg_v5_0_0_hw_init, @@ -631,14 +628,8 @@ static const struct amd_ip_funcs jpeg_v5_0_0_ip_funcs = { .resume = jpeg_v5_0_0_resume, .is_idle = jpeg_v5_0_0_is_idle, .wait_for_idle = jpeg_v5_0_0_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = jpeg_v5_0_0_set_clockgating_state, .set_powergating_state = jpeg_v5_0_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs jpeg_v5_0_0_dec_ring_vm_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index 231a3d490ea8..0e758ebf2372 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -55,8 +55,8 @@ MODULE_FIRMWARE("amdgpu/gc_11_5_1_mes1.bin"); MODULE_FIRMWARE("amdgpu/gc_11_5_2_mes_2.bin"); MODULE_FIRMWARE("amdgpu/gc_11_5_2_mes1.bin"); -static int mes_v11_0_hw_init(void *handle); -static int mes_v11_0_hw_fini(void *handle); +static int mes_v11_0_hw_init(struct amdgpu_ip_block *ip_block); +static int mes_v11_0_hw_fini(struct amdgpu_ip_block *ip_block); static int mes_v11_0_kiq_hw_init(struct amdgpu_device *adev); static int mes_v11_0_kiq_hw_fini(struct amdgpu_device *adev); @@ -366,7 +366,7 @@ static int mes_v11_0_reset_queue_mmio(struct amdgpu_mes *mes, uint32_t queue_typ uint32_t queue_id, uint32_t vmid) { struct amdgpu_device *adev = mes->adev; - uint32_t value; + uint32_t value, reg; int i, r = 0; amdgpu_gfx_rlc_enter_safe_mode(adev, 0); @@ -424,6 +424,31 @@ static int mes_v11_0_reset_queue_mmio(struct amdgpu_mes *mes, uint32_t queue_typ } soc21_grbm_select(adev, 0, 0, 0, 0); mutex_unlock(&adev->srbm_mutex); + } else if (queue_type == AMDGPU_RING_TYPE_SDMA) { + dev_info(adev->dev, "reset sdma queue (%d:%d:%d)\n", + me_id, pipe_id, queue_id); + switch (me_id) { + case 1: + reg = SOC15_REG_OFFSET(GC, 0, regSDMA1_QUEUE_RESET_REQ); + break; + case 0: + default: + reg = SOC15_REG_OFFSET(GC, 0, regSDMA0_QUEUE_RESET_REQ); + break; + } + + value = 1 << queue_id; + WREG32(reg, value); + /* wait for queue reset done */ + for (i = 0; i < adev->usec_timeout; i++) { + if (!(RREG32(reg) & value)) + break; + udelay(1); + } + if (i >= adev->usec_timeout) { + dev_err(adev->dev, "failed to wait on sdma queue reset done\n"); + r = -ETIMEDOUT; + } } amdgpu_gfx_rlc_exit_safe_mode(adev, 0); @@ -1336,9 +1361,9 @@ static int mes_v11_0_mqd_sw_init(struct amdgpu_device *adev, return 0; } -static int mes_v11_0_sw_init(void *handle) +static int mes_v11_0_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int pipe, r; adev->mes.funcs = &mes_v11_0_funcs; @@ -1377,9 +1402,9 @@ static int mes_v11_0_sw_init(void *handle) return 0; } -static int mes_v11_0_sw_fini(void *handle) +static int mes_v11_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int pipe; for (pipe = 0; pipe < AMDGPU_MAX_MES_PIPES; pipe++) { @@ -1473,6 +1498,7 @@ static void mes_v11_0_kiq_clear(struct amdgpu_device *adev) static int mes_v11_0_kiq_hw_init(struct amdgpu_device *adev) { int r = 0; + struct amdgpu_ip_block *ip_block; if (adev->firmware.load_type == AMDGPU_FW_LOAD_DIRECT) { @@ -1496,6 +1522,12 @@ static int mes_v11_0_kiq_hw_init(struct amdgpu_device *adev) mes_v11_0_kiq_setting(&adev->gfx.kiq[0].ring); + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_MES); + if (unlikely(!ip_block)) { + dev_err(adev->dev, "Failed to get MES handle\n"); + return -EINVAL; + } + r = mes_v11_0_queue_init(adev, AMDGPU_MES_KIQ_PIPE); if (r) goto failure; @@ -1506,7 +1538,7 @@ static int mes_v11_0_kiq_hw_init(struct amdgpu_device *adev) adev->mes.enable_legacy_queue_map = false; if (adev->mes.enable_legacy_queue_map) { - r = mes_v11_0_hw_init(adev); + r = mes_v11_0_hw_init(ip_block); if (r) goto failure; } @@ -1514,7 +1546,7 @@ static int mes_v11_0_kiq_hw_init(struct amdgpu_device *adev) return r; failure: - mes_v11_0_hw_fini(adev); + mes_v11_0_hw_fini(ip_block); return r; } @@ -1535,10 +1567,10 @@ static int mes_v11_0_kiq_hw_fini(struct amdgpu_device *adev) return 0; } -static int mes_v11_0_hw_init(void *handle) +static int mes_v11_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->mes.ring[0].sched.ready) goto out; @@ -1590,13 +1622,13 @@ out: return 0; failure: - mes_v11_0_hw_fini(adev); + mes_v11_0_hw_fini(ip_block); return r; } -static int mes_v11_0_hw_fini(void *handle) +static int mes_v11_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_is_mes_info_enable(adev)) { amdgpu_bo_free_kernel(&adev->mes.resource_1, &adev->mes.resource_1_gpu_addr, &adev->mes.resource_1_addr); @@ -1604,33 +1636,31 @@ static int mes_v11_0_hw_fini(void *handle) return 0; } -static int mes_v11_0_suspend(void *handle) +static int mes_v11_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_mes_suspend(adev); + r = amdgpu_mes_suspend(ip_block->adev); if (r) return r; - return mes_v11_0_hw_fini(adev); + return mes_v11_0_hw_fini(ip_block); } -static int mes_v11_0_resume(void *handle) +static int mes_v11_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = mes_v11_0_hw_init(adev); + r = mes_v11_0_hw_init(ip_block); if (r) return r; - return amdgpu_mes_resume(adev); + return amdgpu_mes_resume(ip_block->adev); } -static int mes_v11_0_early_init(void *handle) +static int mes_v11_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int pipe, r; for (pipe = 0; pipe < AMDGPU_MAX_MES_PIPES; pipe++) { @@ -1644,9 +1674,9 @@ static int mes_v11_0_early_init(void *handle) return 0; } -static int mes_v11_0_late_init(void *handle) +static int mes_v11_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* it's only intended for use in mes_self_test case, not for s0ix and reset */ if (!amdgpu_in_reset(adev) && !adev->in_s0ix && !adev->in_suspend && @@ -1666,8 +1696,6 @@ static const struct amd_ip_funcs mes_v11_0_ip_funcs = { .hw_fini = mes_v11_0_hw_fini, .suspend = mes_v11_0_suspend, .resume = mes_v11_0_resume, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; const struct amdgpu_ip_block_version mes_v11_0_ip_block = { diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c index 8d27421689c9..3daa8862e622 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c @@ -39,8 +39,8 @@ MODULE_FIRMWARE("amdgpu/gc_12_0_1_mes.bin"); MODULE_FIRMWARE("amdgpu/gc_12_0_1_mes1.bin"); MODULE_FIRMWARE("amdgpu/gc_12_0_1_uni_mes.bin"); -static int mes_v12_0_hw_init(void *handle); -static int mes_v12_0_hw_fini(void *handle); +static int mes_v12_0_hw_init(struct amdgpu_ip_block *ip_block); +static int mes_v12_0_hw_fini(struct amdgpu_ip_block *ip_block); static int mes_v12_0_kiq_hw_init(struct amdgpu_device *adev); static int mes_v12_0_kiq_hw_fini(struct amdgpu_device *adev); @@ -621,7 +621,7 @@ static int mes_v12_0_set_hw_resources(struct amdgpu_mes *mes, int pipe) if (amdgpu_mes_log_enable) { mes_set_hw_res_pkt.enable_mes_event_int_logging = 1; - mes_set_hw_res_pkt.event_intr_history_gpu_mc_ptr = mes->event_log_gpu_addr; + mes_set_hw_res_pkt.event_intr_history_gpu_mc_ptr = mes->event_log_gpu_addr + pipe * AMDGPU_MES_LOG_BUFFER_SIZE; } return mes_v12_0_submit_pkt_and_poll_completion(mes, pipe, @@ -1326,9 +1326,9 @@ static int mes_v12_0_mqd_sw_init(struct amdgpu_device *adev, return 0; } -static int mes_v12_0_sw_init(void *handle) +static int mes_v12_0_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int pipe, r; adev->mes.funcs = &mes_v12_0_funcs; @@ -1336,7 +1336,7 @@ static int mes_v12_0_sw_init(void *handle) adev->mes.kiq_hw_fini = &mes_v12_0_kiq_hw_fini; adev->mes.enable_legacy_queue_map = true; - adev->mes.event_log_size = AMDGPU_MES_LOG_BUFFER_SIZE; + adev->mes.event_log_size = adev->enable_uni_mes ? (AMDGPU_MAX_MES_PIPES * AMDGPU_MES_LOG_BUFFER_SIZE) : AMDGPU_MES_LOG_BUFFER_SIZE; r = amdgpu_mes_init(adev); if (r) @@ -1362,9 +1362,9 @@ static int mes_v12_0_sw_init(void *handle) return 0; } -static int mes_v12_0_sw_fini(void *handle) +static int mes_v12_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int pipe; for (pipe = 0; pipe < AMDGPU_MAX_MES_PIPES; pipe++) { @@ -1452,6 +1452,7 @@ static void mes_v12_0_kiq_setting(struct amdgpu_ring *ring) static int mes_v12_0_kiq_hw_init(struct amdgpu_device *adev) { int r = 0; + struct amdgpu_ip_block *ip_block; if (adev->enable_uni_mes) mes_v12_0_kiq_setting(&adev->mes.ring[AMDGPU_MES_KIQ_PIPE]); @@ -1479,6 +1480,12 @@ static int mes_v12_0_kiq_hw_init(struct amdgpu_device *adev) mes_v12_0_enable(adev, true); + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_MES); + if (unlikely(!ip_block)) { + dev_err(adev->dev, "Failed to get MES handle\n"); + return -EINVAL; + } + r = mes_v12_0_queue_init(adev, AMDGPU_MES_KIQ_PIPE); if (r) goto failure; @@ -1492,7 +1499,7 @@ static int mes_v12_0_kiq_hw_init(struct amdgpu_device *adev) } if (adev->mes.enable_legacy_queue_map) { - r = mes_v12_0_hw_init(adev); + r = mes_v12_0_hw_init(ip_block); if (r) goto failure; } @@ -1500,7 +1507,7 @@ static int mes_v12_0_kiq_hw_init(struct amdgpu_device *adev) return r; failure: - mes_v12_0_hw_fini(adev); + mes_v12_0_hw_fini(ip_block); return r; } @@ -1522,10 +1529,10 @@ static int mes_v12_0_kiq_hw_fini(struct amdgpu_device *adev) return 0; } -static int mes_v12_0_hw_init(void *handle) +static int mes_v12_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->mes.ring[0].sched.ready) goto out; @@ -1584,42 +1591,40 @@ out: return 0; failure: - mes_v12_0_hw_fini(adev); + mes_v12_0_hw_fini(ip_block); return r; } -static int mes_v12_0_hw_fini(void *handle) +static int mes_v12_0_hw_fini(struct amdgpu_ip_block *ip_block) { return 0; } -static int mes_v12_0_suspend(void *handle) +static int mes_v12_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_mes_suspend(adev); + r = amdgpu_mes_suspend(ip_block->adev); if (r) return r; - return mes_v12_0_hw_fini(adev); + return mes_v12_0_hw_fini(ip_block); } -static int mes_v12_0_resume(void *handle) +static int mes_v12_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = mes_v12_0_hw_init(adev); + r = mes_v12_0_hw_init(ip_block); if (r) return r; - return amdgpu_mes_resume(adev); + return amdgpu_mes_resume(ip_block->adev); } -static int mes_v12_0_early_init(void *handle) +static int mes_v12_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int pipe, r; for (pipe = 0; pipe < AMDGPU_MAX_MES_PIPES; pipe++) { @@ -1631,9 +1636,9 @@ static int mes_v12_0_early_init(void *handle) return 0; } -static int mes_v12_0_late_init(void *handle) +static int mes_v12_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* it's only intended for use in mes_self_test case, not for s0ix and reset */ if (!amdgpu_in_reset(adev) && !adev->in_s0ix && !adev->in_suspend) diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index e3ddd22aa172..e9a6f33ca710 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -229,6 +229,52 @@ static void mmhub_v1_0_disable_identity_aperture(struct amdgpu_device *adev) 0); } +static void mmhub_v1_0_init_saw(struct amdgpu_device *adev) +{ + uint64_t pt_base = amdgpu_gmc_pd_addr(adev->gart.bo); + uint32_t tmp; + + /* VM_9_X_REGISTER_VM_L2_SAW_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32 */ + WREG32_SOC15(MMHUB, 0, mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32, + lower_32_bits(pt_base >> 12)); + + /* VM_9_X_REGISTER_VM_L2_SAW_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32 */ + WREG32_SOC15(MMHUB, 0, mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32, + upper_32_bits(pt_base >> 12)); + + /* VM_9_X_REGISTER_VM_L2_SAW_CONTEXT0_PAGE_TABLE_START_ADDR_LO32 */ + WREG32_SOC15(MMHUB, 0, mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_START_ADDR_LO32, + (u32)(adev->gmc.gart_start >> 12)); + + /* VM_9_X_REGISTER_VM_L2_SAW_CONTEXT0_PAGE_TABLE_START_ADDR_HI32 */ + WREG32_SOC15(MMHUB, 0, mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_START_ADDR_HI32, + (u32)(adev->gmc.gart_start >> 44)); + + /* VM_9_X_REGISTER_VM_L2_SAW_CONTEXT0_PAGE_TABLE_END_ADDR_LO32 */ + WREG32_SOC15(MMHUB, 0, mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_END_ADDR_LO32, + (u32)(adev->gmc.gart_end >> 12)); + + /* VM_9_X_REGISTER_VM_L2_SAW_CONTEXT0_PAGE_TABLE_END_ADDR_HI32 */ + WREG32_SOC15(MMHUB, 0, mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_END_ADDR_HI32, + (u32)(adev->gmc.gart_end >> 44)); + + /* Program SAW CONTEXT0 CNTL */ + tmp = RREG32_SOC15(MMHUB, 0, mmVM_L2_SAW_CONTEXT0_CNTL); + tmp |= 1 << CONTEXT0_CNTL_ENABLE_OFFSET; + tmp &= ~(3 << CONTEXT0_CNTL_PAGE_TABLE_DEPTH_OFFSET); + WREG32_SOC15(MMHUB, 0, mmVM_L2_SAW_CONTEXT0_CNTL, tmp); + + /* Disable all Contexts except Context0 */ + tmp = 0xfffe; + WREG32_SOC15(MMHUB, 0, mmVM_L2_SAW_CONTEXTS_DISABLE, tmp); + + /* Program SAW CNTL4 */ + tmp = RREG32_SOC15(MMHUB, 0, mmVM_L2_SAW_CNTL4); + tmp |= 1 << VMC_TAP_PDE_REQUEST_SNOOP_OFFSET; + tmp |= 1 << VMC_TAP_PTE_REQUEST_SNOOP_OFFSET; + WREG32_SOC15(MMHUB, 0, mmVM_L2_SAW_CNTL4, tmp); +} + static void mmhub_v1_0_setup_vmid_config(struct amdgpu_device *adev) { struct amdgpu_vmhub *hub = &adev->vmhub[AMDGPU_MMHUB0(0)]; @@ -283,6 +329,9 @@ static void mmhub_v1_0_setup_vmid_config(struct amdgpu_device *adev) i * hub->ctx_addr_distance, upper_32_bits(adev->vm_manager.max_pfn - 1)); } + + if (amdgpu_ip_version(adev, ISP_HWIP, 0)) + mmhub_v1_0_init_saw(adev); } static void mmhub_v1_0_program_invalidation(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c index b281462093f1..0820ed62e2e8 100644 --- a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c @@ -542,19 +542,19 @@ static void navi10_ih_set_self_irq_funcs(struct amdgpu_device *adev) adev->irq.self_irq.funcs = &navi10_ih_self_irq_funcs; } -static int navi10_ih_early_init(void *handle) +static int navi10_ih_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; navi10_ih_set_interrupt_funcs(adev); navi10_ih_set_self_irq_funcs(adev); return 0; } -static int navi10_ih_sw_init(void *handle) +static int navi10_ih_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; bool use_bus_addr; r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_IH, 0, @@ -593,43 +593,37 @@ static int navi10_ih_sw_init(void *handle) return r; } -static int navi10_ih_sw_fini(void *handle) +static int navi10_ih_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_fini_sw(adev); return 0; } -static int navi10_ih_hw_init(void *handle) +static int navi10_ih_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return navi10_ih_irq_init(adev); } -static int navi10_ih_hw_fini(void *handle) +static int navi10_ih_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - navi10_ih_irq_disable(adev); + navi10_ih_irq_disable(ip_block->adev); return 0; } -static int navi10_ih_suspend(void *handle) +static int navi10_ih_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return navi10_ih_hw_fini(adev); + return navi10_ih_hw_fini(ip_block); } -static int navi10_ih_resume(void *handle) +static int navi10_ih_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return navi10_ih_hw_init(adev); + return navi10_ih_hw_init(ip_block); } static bool navi10_ih_is_idle(void *handle) @@ -638,13 +632,13 @@ static bool navi10_ih_is_idle(void *handle) return true; } -static int navi10_ih_wait_for_idle(void *handle) +static int navi10_ih_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* todo */ return -ETIMEDOUT; } -static int navi10_ih_soft_reset(void *handle) +static int navi10_ih_soft_reset(struct amdgpu_ip_block *ip_block) { /* todo */ return 0; @@ -700,7 +694,6 @@ static void navi10_ih_get_clockgating_state(void *handle, u64 *flags) static const struct amd_ip_funcs navi10_ih_ip_funcs = { .name = "navi10_ih", .early_init = navi10_ih_early_init, - .late_init = NULL, .sw_init = navi10_ih_sw_init, .sw_fini = navi10_ih_sw_fini, .hw_init = navi10_ih_hw_init, @@ -713,8 +706,6 @@ static const struct amd_ip_funcs navi10_ih_ip_funcs = { .set_clockgating_state = navi10_ih_set_clockgating_state, .set_powergating_state = navi10_ih_set_powergating_state, .get_clockgating_state = navi10_ih_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ih_funcs navi10_ih_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_sdma_pkt_open.h b/drivers/gpu/drm/amd/amdgpu/navi10_sdma_pkt_open.h index a5b60c9a2418..c88284ff92d8 100644 --- a/drivers/gpu/drm/amd/amdgpu/navi10_sdma_pkt_open.h +++ b/drivers/gpu/drm/amd/amdgpu/navi10_sdma_pkt_open.h @@ -68,6 +68,7 @@ #define SDMA_SUBOP_POLL_REG_WRITE_MEM 1 #define SDMA_SUBOP_POLL_DBIT_WRITE_MEM 2 #define SDMA_SUBOP_POLL_MEM_VERIFY 3 +#define SDMA_SUBOP_VM_INVALIDATION 4 #define HEADER_AGENT_DISPATCH 4 #define HEADER_BARRIER 5 #define SDMA_OP_AQL_COPY 0 @@ -4040,6 +4041,69 @@ #define SDMA_PKT_POLL_MEM_VERIFY_RESERVED_RESERVED(x) (((x) & SDMA_PKT_POLL_MEM_VERIFY_RESERVED_reserved_mask) << SDMA_PKT_POLL_MEM_VERIFY_RESERVED_reserved_shift) +/* +** Definitions for SDMA_PKT_VM_INVALIDATION packet +*/ + +/*define for HEADER word*/ +/*define for op field*/ +#define SDMA_PKT_VM_INVALIDATION_HEADER_op_offset 0 +#define SDMA_PKT_VM_INVALIDATION_HEADER_op_mask 0x000000FF +#define SDMA_PKT_VM_INVALIDATION_HEADER_op_shift 0 +#define SDMA_PKT_VM_INVALIDATION_HEADER_OP(x) (((x) & SDMA_PKT_VM_INVALIDATION_HEADER_op_mask) << SDMA_PKT_VM_INVALIDATION_HEADER_op_shift) + +/*define for sub_op field*/ +#define SDMA_PKT_VM_INVALIDATION_HEADER_sub_op_offset 0 +#define SDMA_PKT_VM_INVALIDATION_HEADER_sub_op_mask 0x000000FF +#define SDMA_PKT_VM_INVALIDATION_HEADER_sub_op_shift 8 +#define SDMA_PKT_VM_INVALIDATION_HEADER_SUB_OP(x) (((x) & SDMA_PKT_VM_INVALIDATION_HEADER_sub_op_mask) << SDMA_PKT_VM_INVALIDATION_HEADER_sub_op_shift) + +/*define for gfx_eng_id field*/ +#define SDMA_PKT_VM_INVALIDATION_HEADER_gfx_eng_id_offset 0 +#define SDMA_PKT_VM_INVALIDATION_HEADER_gfx_eng_id_mask 0x0000001F +#define SDMA_PKT_VM_INVALIDATION_HEADER_gfx_eng_id_shift 16 +#define SDMA_PKT_VM_INVALIDATION_HEADER_GFX_ENG_ID(x) (((x) & SDMA_PKT_VM_INVALIDATION_HEADER_gfx_eng_id_mask) << SDMA_PKT_VM_INVALIDATION_HEADER_gfx_eng_id_shift) + +/*define for mm_eng_id field*/ +#define SDMA_PKT_VM_INVALIDATION_HEADER_mm_eng_id_offset 0 +#define SDMA_PKT_VM_INVALIDATION_HEADER_mm_eng_id_mask 0x0000001F +#define SDMA_PKT_VM_INVALIDATION_HEADER_mm_eng_id_shift 24 +#define SDMA_PKT_VM_INVALIDATION_HEADER_MM_ENG_ID(x) (((x) & SDMA_PKT_VM_INVALIDATION_HEADER_mm_eng_id_mask) << SDMA_PKT_VM_INVALIDATION_HEADER_mm_eng_id_shift) + +/*define for INVALIDATEREQ word*/ +/*define for invalidatereq field*/ +#define SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_invalidatereq_offset 1 +#define SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_invalidatereq_mask 0xFFFFFFFF +#define SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_invalidatereq_shift 0 +#define SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_INVALIDATEREQ(x) (((x) & SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_invalidatereq_mask) << SDMA_PKT_VM_INVALIDATION_INVALIDATEREQ_invalidatereq_shift) + +/*define for ADDRESSRANGELO word*/ +/*define for addressrangelo field*/ +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_addressrangelo_offset 2 +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_addressrangelo_mask 0xFFFFFFFF +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_addressrangelo_shift 0 +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_ADDRESSRANGELO(x) (((x) & SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_addressrangelo_mask) << SDMA_PKT_VM_INVALIDATION_ADDRESSRANGELO_addressrangelo_shift) + +/*define for ADDRESSRANGEHI word*/ +/*define for invalidateack field*/ +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_invalidateack_offset 3 +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_invalidateack_mask 0x0000FFFF +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_invalidateack_shift 0 +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_INVALIDATEACK(x) (((x) & SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_invalidateack_mask) << SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_invalidateack_shift) + +/*define for addressrangehi field*/ +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_addressrangehi_offset 3 +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_addressrangehi_mask 0x0000001F +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_addressrangehi_shift 16 +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_ADDRESSRANGEHI(x) (((x) & SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_addressrangehi_mask) << SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_addressrangehi_shift) + +/*define for reserved field*/ +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_reserved_offset 3 +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_reserved_mask 0x000001FF +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_reserved_shift 23 +#define SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_RESERVED(x) (((x) & SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_reserved_mask) << SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_reserved_shift) + + /* ** Definitions for SDMA_PKT_ATOMIC packet */ diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c index 8d80df94bd8b..a26a9be58eac 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c @@ -414,8 +414,7 @@ static void nbio_v7_4_handle_ras_controller_intr_no_bifring(struct amdgpu_device /* ras_controller_int is dedicated for nbif ras error, * not the global interrupt for sync flood */ - amdgpu_ras_set_fed(adev, true); - amdgpu_ras_reset_gpu(adev); + amdgpu_ras_global_ras_isr(adev); } amdgpu_ras_error_data_fini(&err_data); diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c index d1bd79bbae53..8a0a63ac88d2 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_9.c @@ -401,6 +401,17 @@ static int nbio_v7_9_get_compute_partition_mode(struct amdgpu_device *adev) return px; } +static bool nbio_v7_9_is_nps_switch_requested(struct amdgpu_device *adev) +{ + u32 tmp; + + tmp = RREG32_SOC15(NBIO, 0, regBIF_BX_PF0_PARTITION_MEM_STATUS); + tmp = REG_GET_FIELD(tmp, BIF_BX_PF0_PARTITION_MEM_STATUS, + CHANGE_STATUE); + + /* 0x8 - NPS switch requested */ + return (tmp == 0x8); +} static u32 nbio_v7_9_get_memory_partition_mode(struct amdgpu_device *adev, u32 *supp_modes) { @@ -508,6 +519,7 @@ const struct amdgpu_nbio_funcs nbio_v7_9_funcs = { .remap_hdp_registers = nbio_v7_9_remap_hdp_registers, .get_compute_partition_mode = nbio_v7_9_get_compute_partition_mode, .get_memory_partition_mode = nbio_v7_9_get_memory_partition_mode, + .is_nps_switch_requested = nbio_v7_9_is_nps_switch_requested, .init_registers = nbio_v7_9_init_registers, .get_pcie_replay_count = nbio_v7_9_get_pcie_replay_count, .set_reg_remap = nbio_v7_9_set_reg_remap, diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c index 4938e6b340e9..6b72169be8f8 100644 --- a/drivers/gpu/drm/amd/amdgpu/nv.c +++ b/drivers/gpu/drm/amd/amdgpu/nv.c @@ -634,9 +634,9 @@ static const struct amdgpu_asic_funcs nv_asic_funcs = { .query_video_codecs = &nv_query_video_codecs, }; -static int nv_common_early_init(void *handle) +static int nv_common_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->nbio.funcs->set_reg_remap(adev); adev->smc_rreg = NULL; @@ -944,9 +944,9 @@ static int nv_common_early_init(void *handle) return 0; } -static int nv_common_late_init(void *handle) +static int nv_common_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) { xgpu_nv_mailbox_get_irq(adev); @@ -973,9 +973,9 @@ static int nv_common_late_init(void *handle) return 0; } -static int nv_common_sw_init(void *handle) +static int nv_common_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) xgpu_nv_mailbox_add_irq_id(adev); @@ -983,14 +983,9 @@ static int nv_common_sw_init(void *handle) return 0; } -static int nv_common_sw_fini(void *handle) +static int nv_common_hw_init(struct amdgpu_ip_block *ip_block) { - return 0; -} - -static int nv_common_hw_init(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->nbio.funcs->apply_lc_spc_mode_wa) adev->nbio.funcs->apply_lc_spc_mode_wa(adev); @@ -1014,9 +1009,9 @@ static int nv_common_hw_init(void *handle) return 0; } -static int nv_common_hw_fini(void *handle) +static int nv_common_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* Disable the doorbell aperture and selfring doorbell aperture * separately in hw_fini because nv_enable_doorbell_aperture @@ -1029,18 +1024,14 @@ static int nv_common_hw_fini(void *handle) return 0; } -static int nv_common_suspend(void *handle) +static int nv_common_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return nv_common_hw_fini(adev); + return nv_common_hw_fini(ip_block); } -static int nv_common_resume(void *handle) +static int nv_common_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return nv_common_hw_init(adev); + return nv_common_hw_init(ip_block); } static bool nv_common_is_idle(void *handle) @@ -1048,16 +1039,6 @@ static bool nv_common_is_idle(void *handle) return true; } -static int nv_common_wait_for_idle(void *handle) -{ - return 0; -} - -static int nv_common_soft_reset(void *handle) -{ - return 0; -} - static int nv_common_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -1115,17 +1096,12 @@ static const struct amd_ip_funcs nv_common_ip_funcs = { .early_init = nv_common_early_init, .late_init = nv_common_late_init, .sw_init = nv_common_sw_init, - .sw_fini = nv_common_sw_fini, .hw_init = nv_common_hw_init, .hw_fini = nv_common_hw_fini, .suspend = nv_common_suspend, .resume = nv_common_resume, .is_idle = nv_common_is_idle, - .wait_for_idle = nv_common_wait_for_idle, - .soft_reset = nv_common_soft_reset, .set_clockgating_state = nv_common_set_clockgating_state, .set_powergating_state = nv_common_set_powergating_state, .get_clockgating_state = nv_common_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h index 37b5ddd6f13b..f4a91b126c73 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h +++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h @@ -103,6 +103,10 @@ enum psp_gfx_cmd_id GFX_CMD_ID_AUTOLOAD_RLC = 0x00000021, /* Indicates all graphics fw loaded, start RLC autoload */ GFX_CMD_ID_BOOT_CFG = 0x00000022, /* Boot Config */ GFX_CMD_ID_SRIOV_SPATIAL_PART = 0x00000027, /* Configure spatial partitioning mode */ + /*IDs of performance monitoring/profiling*/ + GFX_CMD_ID_CONFIG_SQ_PERFMON = 0x00000046, /* Config CGTT_SQ_CLK_CTRL */ + /* Dynamic memory partitioninig (NPS mode change)*/ + GFX_CMD_ID_FB_NPS_MODE = 0x00000048, /* Configure memory partitioning mode */ }; /* PSP boot config sub-commands */ @@ -351,6 +355,20 @@ struct psp_gfx_cmd_sriov_spatial_part { uint32_t override_this_aid; }; +/*Structure for sq performance monitoring/profiling enable/disable*/ +struct psp_gfx_cmd_config_sq_perfmon { + uint32_t gfx_xcp_mask; + uint8_t core_override; + uint8_t reg_override; + uint8_t perfmon_override; + uint8_t reserved[5]; +}; + +struct psp_gfx_cmd_fb_memory_part { + uint32_t mode; /* requested NPS mode */ + uint32_t resvd; +}; + /* All GFX ring buffer commands. */ union psp_gfx_commands { @@ -365,6 +383,8 @@ union psp_gfx_commands struct psp_gfx_cmd_load_toc cmd_load_toc; struct psp_gfx_cmd_boot_cfg boot_cfg; struct psp_gfx_cmd_sriov_spatial_part cmd_spatial_part; + struct psp_gfx_cmd_config_sq_perfmon config_sq_perfmon; + struct psp_gfx_cmd_fb_memory_part cmd_memory_part; }; struct psp_gfx_uresp_reserved diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c index 51e470e8d67d..c4b775aaee9f 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v13_0.c @@ -823,6 +823,30 @@ static bool psp_v13_0_is_aux_sos_load_required(struct psp_context *psp) return (pmfw_ver < 0x557300); } +static bool psp_v13_0_is_reload_needed(struct psp_context *psp) +{ + uint32_t ucode_ver; + + if (!psp_v13_0_is_sos_alive(psp)) + return false; + + /* Restrict reload support only to specific IP versions */ + switch (amdgpu_ip_version(psp->adev, MP0_HWIP, 0)) { + case IP_VERSION(13, 0, 2): + case IP_VERSION(13, 0, 6): + case IP_VERSION(13, 0, 14): + /* TOS version read from microcode header */ + ucode_ver = psp->sos.fw_version; + /* Read TOS version from hardware */ + psp_v13_0_init_sos_version(psp); + return (ucode_ver != psp->sos.fw_version); + default: + return false; + } + + return false; +} + static const struct psp_funcs psp_v13_0_funcs = { .init_microcode = psp_v13_0_init_microcode, .wait_for_bootloader = psp_v13_0_wait_for_bootloader_steady_state, @@ -847,6 +871,7 @@ static const struct psp_funcs psp_v13_0_funcs = { .fatal_error_recovery_quirk = psp_v13_0_fatal_error_recovery_quirk, .get_ras_capability = psp_v13_0_get_ras_capability, .is_aux_sos_load_required = psp_v13_0_is_aux_sos_load_required, + .is_reload_needed = psp_v13_0_is_reload_needed, }; void psp_v13_0_set_psp_funcs(struct psp_context *psp) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c index 725392522267..7948d74f8722 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c @@ -807,9 +807,9 @@ static void sdma_v2_4_ring_emit_wreg(struct amdgpu_ring *ring, amdgpu_ring_write(ring, val); } -static int sdma_v2_4_early_init(void *handle) +static int sdma_v2_4_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; adev->sdma.num_instances = SDMA_MAX_INSTANCE; @@ -826,11 +826,11 @@ static int sdma_v2_4_early_init(void *handle) return 0; } -static int sdma_v2_4_sw_init(void *handle) +static int sdma_v2_4_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* SDMA trap event */ r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, VISLANDS30_IV_SRCID_SDMA_TRAP, @@ -866,9 +866,9 @@ static int sdma_v2_4_sw_init(void *handle) return r; } -static int sdma_v2_4_sw_fini(void *handle) +static int sdma_v2_4_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->sdma.num_instances; i++) @@ -878,10 +878,10 @@ static int sdma_v2_4_sw_fini(void *handle) return 0; } -static int sdma_v2_4_hw_init(void *handle) +static int sdma_v2_4_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; sdma_v2_4_init_golden_registers(adev); @@ -892,27 +892,21 @@ static int sdma_v2_4_hw_init(void *handle) return r; } -static int sdma_v2_4_hw_fini(void *handle) +static int sdma_v2_4_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - sdma_v2_4_enable(adev, false); + sdma_v2_4_enable(ip_block->adev, false); return 0; } -static int sdma_v2_4_suspend(void *handle) +static int sdma_v2_4_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v2_4_hw_fini(adev); + return sdma_v2_4_hw_fini(ip_block); } -static int sdma_v2_4_resume(void *handle) +static int sdma_v2_4_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v2_4_hw_init(adev); + return sdma_v2_4_hw_init(ip_block); } static bool sdma_v2_4_is_idle(void *handle) @@ -927,11 +921,11 @@ static bool sdma_v2_4_is_idle(void *handle) return true; } -static int sdma_v2_4_wait_for_idle(void *handle) +static int sdma_v2_4_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { tmp = RREG32(mmSRBM_STATUS2) & (SRBM_STATUS2__SDMA_BUSY_MASK | @@ -944,10 +938,10 @@ static int sdma_v2_4_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int sdma_v2_4_soft_reset(void *handle) +static int sdma_v2_4_soft_reset(struct amdgpu_ip_block *ip_block) { u32 srbm_soft_reset = 0; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 tmp = RREG32(mmSRBM_STATUS2); if (tmp & SRBM_STATUS2__SDMA_BUSY_MASK) { @@ -1102,7 +1096,6 @@ static int sdma_v2_4_set_powergating_state(void *handle, static const struct amd_ip_funcs sdma_v2_4_ip_funcs = { .name = "sdma_v2_4", .early_init = sdma_v2_4_early_init, - .late_init = NULL, .sw_init = sdma_v2_4_sw_init, .sw_fini = sdma_v2_4_sw_fini, .hw_init = sdma_v2_4_hw_init, @@ -1114,8 +1107,6 @@ static const struct amd_ip_funcs sdma_v2_4_ip_funcs = { .soft_reset = sdma_v2_4_soft_reset, .set_clockgating_state = sdma_v2_4_set_clockgating_state, .set_powergating_state = sdma_v2_4_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs sdma_v2_4_ring_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c index e65194fe94af..9a3d729545a7 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c @@ -1080,9 +1080,9 @@ static void sdma_v3_0_ring_emit_wreg(struct amdgpu_ring *ring, amdgpu_ring_write(ring, val); } -static int sdma_v3_0_early_init(void *handle) +static int sdma_v3_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; switch (adev->asic_type) { @@ -1106,11 +1106,11 @@ static int sdma_v3_0_early_init(void *handle) return 0; } -static int sdma_v3_0_sw_init(void *handle) +static int sdma_v3_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* SDMA trap event */ r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, VISLANDS30_IV_SRCID_SDMA_TRAP, @@ -1152,9 +1152,9 @@ static int sdma_v3_0_sw_init(void *handle) return r; } -static int sdma_v3_0_sw_fini(void *handle) +static int sdma_v3_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->sdma.num_instances; i++) @@ -1164,10 +1164,10 @@ static int sdma_v3_0_sw_fini(void *handle) return 0; } -static int sdma_v3_0_hw_init(void *handle) +static int sdma_v3_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; sdma_v3_0_init_golden_registers(adev); @@ -1178,9 +1178,9 @@ static int sdma_v3_0_hw_init(void *handle) return r; } -static int sdma_v3_0_hw_fini(void *handle) +static int sdma_v3_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; sdma_v3_0_ctx_switch_enable(adev, false); sdma_v3_0_enable(adev, false); @@ -1188,18 +1188,14 @@ static int sdma_v3_0_hw_fini(void *handle) return 0; } -static int sdma_v3_0_suspend(void *handle) +static int sdma_v3_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v3_0_hw_fini(adev); + return sdma_v3_0_hw_fini(ip_block); } -static int sdma_v3_0_resume(void *handle) +static int sdma_v3_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v3_0_hw_init(adev); + return sdma_v3_0_hw_init(ip_block); } static bool sdma_v3_0_is_idle(void *handle) @@ -1214,11 +1210,11 @@ static bool sdma_v3_0_is_idle(void *handle) return true; } -static int sdma_v3_0_wait_for_idle(void *handle) +static int sdma_v3_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { tmp = RREG32(mmSRBM_STATUS2) & (SRBM_STATUS2__SDMA_BUSY_MASK | @@ -1231,9 +1227,9 @@ static int sdma_v3_0_wait_for_idle(void *handle) return -ETIMEDOUT; } -static bool sdma_v3_0_check_soft_reset(void *handle) +static bool sdma_v3_0_check_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset = 0; u32 tmp = RREG32(mmSRBM_STATUS2); @@ -1252,9 +1248,9 @@ static bool sdma_v3_0_check_soft_reset(void *handle) } } -static int sdma_v3_0_pre_soft_reset(void *handle) +static int sdma_v3_0_pre_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset = 0; if (!adev->sdma.srbm_soft_reset) @@ -1271,9 +1267,9 @@ static int sdma_v3_0_pre_soft_reset(void *handle) return 0; } -static int sdma_v3_0_post_soft_reset(void *handle) +static int sdma_v3_0_post_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset = 0; if (!adev->sdma.srbm_soft_reset) @@ -1290,9 +1286,9 @@ static int sdma_v3_0_post_soft_reset(void *handle) return 0; } -static int sdma_v3_0_soft_reset(void *handle) +static int sdma_v3_0_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset = 0; u32 tmp; @@ -1538,7 +1534,6 @@ static void sdma_v3_0_get_clockgating_state(void *handle, u64 *flags) static const struct amd_ip_funcs sdma_v3_0_ip_funcs = { .name = "sdma_v3_0", .early_init = sdma_v3_0_early_init, - .late_init = NULL, .sw_init = sdma_v3_0_sw_init, .sw_fini = sdma_v3_0_sw_fini, .hw_init = sdma_v3_0_hw_init, @@ -1554,8 +1549,6 @@ static const struct amd_ip_funcs sdma_v3_0_ip_funcs = { .set_clockgating_state = sdma_v3_0_set_clockgating_state, .set_powergating_state = sdma_v3_0_set_powergating_state, .get_clockgating_state = sdma_v3_0_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs sdma_v3_0_ring_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index 23ef4eb36b40..c1f98f6cf20d 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -1751,9 +1751,9 @@ static bool sdma_v4_0_fw_support_paging_queue(struct amdgpu_device *adev) } } -static int sdma_v4_0_early_init(void *handle) +static int sdma_v4_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = sdma_v4_0_init_microcode(adev); @@ -1780,9 +1780,9 @@ static int sdma_v4_0_process_ras_data_cb(struct amdgpu_device *adev, void *err_data, struct amdgpu_iv_entry *entry); -static int sdma_v4_0_late_init(void *handle) +static int sdma_v4_0_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; sdma_v4_0_setup_ulv(adev); @@ -1792,11 +1792,11 @@ static int sdma_v4_0_late_init(void *handle) return 0; } -static int sdma_v4_0_sw_init(void *handle) +static int sdma_v4_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_4_0); uint32_t *ptr; @@ -1929,9 +1929,9 @@ static int sdma_v4_0_sw_init(void *handle) return r; } -static int sdma_v4_0_sw_fini(void *handle) +static int sdma_v4_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->sdma.num_instances; i++) { @@ -1951,9 +1951,9 @@ static int sdma_v4_0_sw_fini(void *handle) return 0; } -static int sdma_v4_0_hw_init(void *handle) +static int sdma_v4_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->flags & AMD_IS_APU) amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_SDMA, false); @@ -1964,9 +1964,9 @@ static int sdma_v4_0_hw_init(void *handle) return sdma_v4_0_start(adev); } -static int sdma_v4_0_hw_fini(void *handle) +static int sdma_v4_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; if (amdgpu_sriov_vf(adev)) @@ -1988,9 +1988,9 @@ static int sdma_v4_0_hw_fini(void *handle) return 0; } -static int sdma_v4_0_suspend(void *handle) +static int sdma_v4_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* SMU saves SDMA state for us */ if (adev->in_s0ix) { @@ -1998,12 +1998,12 @@ static int sdma_v4_0_suspend(void *handle) return 0; } - return sdma_v4_0_hw_fini(adev); + return sdma_v4_0_hw_fini(ip_block); } -static int sdma_v4_0_resume(void *handle) +static int sdma_v4_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* SMU restores SDMA state for us */ if (adev->in_s0ix) { @@ -2012,7 +2012,7 @@ static int sdma_v4_0_resume(void *handle) return 0; } - return sdma_v4_0_hw_init(adev); + return sdma_v4_0_hw_init(ip_block); } static bool sdma_v4_0_is_idle(void *handle) @@ -2030,11 +2030,11 @@ static bool sdma_v4_0_is_idle(void *handle) return true; } -static int sdma_v4_0_wait_for_idle(void *handle) +static int sdma_v4_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i, j; u32 sdma[AMDGPU_MAX_SDMA_INSTANCES]; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { for (j = 0; j < adev->sdma.num_instances; j++) { @@ -2049,7 +2049,7 @@ static int sdma_v4_0_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int sdma_v4_0_soft_reset(void *handle) +static int sdma_v4_0_soft_reset(struct amdgpu_ip_block *ip_block) { /* todo */ @@ -2350,9 +2350,9 @@ static void sdma_v4_0_get_clockgating_state(void *handle, u64 *flags) *flags |= AMD_CG_SUPPORT_SDMA_LS; } -static void sdma_v4_0_print_ip_state(void *handle, struct drm_printer *p) +static void sdma_v4_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_4_0); uint32_t instance_offset; @@ -2371,9 +2371,9 @@ static void sdma_v4_0_print_ip_state(void *handle, struct drm_printer *p) } } -static void sdma_v4_0_dump_ip_state(void *handle) +static void sdma_v4_0_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t instance_offset; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_4_0); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c index c77889040760..9c7cea0890c9 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c @@ -1290,9 +1290,9 @@ static bool sdma_v4_4_2_fw_support_paging_queue(struct amdgpu_device *adev) } } -static int sdma_v4_4_2_early_init(void *handle) +static int sdma_v4_4_2_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = sdma_v4_4_2_init_microcode(adev); @@ -1318,9 +1318,9 @@ static int sdma_v4_4_2_process_ras_data_cb(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry); #endif -static int sdma_v4_4_2_late_init(void *handle) +static int sdma_v4_4_2_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; #if 0 struct ras_ih_if ih_info = { .cb = sdma_v4_4_2_process_ras_data_cb, @@ -1332,11 +1332,11 @@ static int sdma_v4_4_2_late_init(void *handle) return 0; } -static int sdma_v4_4_2_sw_init(void *handle) +static int sdma_v4_4_2_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 aid_id; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_4_4_2); uint32_t *ptr; @@ -1445,9 +1445,9 @@ static int sdma_v4_4_2_sw_init(void *handle) return r; } -static int sdma_v4_4_2_sw_fini(void *handle) +static int sdma_v4_4_2_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->sdma.num_instances; i++) { @@ -1467,10 +1467,10 @@ static int sdma_v4_4_2_sw_fini(void *handle) return 0; } -static int sdma_v4_4_2_hw_init(void *handle) +static int sdma_v4_4_2_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t inst_mask; inst_mask = GENMASK(adev->sdma.num_instances - 1, 0); @@ -1482,9 +1482,9 @@ static int sdma_v4_4_2_hw_init(void *handle) return r; } -static int sdma_v4_4_2_hw_fini(void *handle) +static int sdma_v4_4_2_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t inst_mask; int i; @@ -1508,21 +1508,19 @@ static int sdma_v4_4_2_hw_fini(void *handle) static int sdma_v4_4_2_set_clockgating_state(void *handle, enum amd_clockgating_state state); -static int sdma_v4_4_2_suspend(void *handle) +static int sdma_v4_4_2_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_in_reset(adev)) sdma_v4_4_2_set_clockgating_state(adev, AMD_CG_STATE_UNGATE); - return sdma_v4_4_2_hw_fini(adev); + return sdma_v4_4_2_hw_fini(ip_block); } -static int sdma_v4_4_2_resume(void *handle) +static int sdma_v4_4_2_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v4_4_2_hw_init(adev); + return sdma_v4_4_2_hw_init(ip_block); } static bool sdma_v4_4_2_is_idle(void *handle) @@ -1540,11 +1538,11 @@ static bool sdma_v4_4_2_is_idle(void *handle) return true; } -static int sdma_v4_4_2_wait_for_idle(void *handle) +static int sdma_v4_4_2_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i, j; u32 sdma[AMDGPU_MAX_SDMA_INSTANCES]; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { for (j = 0; j < adev->sdma.num_instances; j++) { @@ -1559,7 +1557,7 @@ static int sdma_v4_4_2_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int sdma_v4_4_2_soft_reset(void *handle) +static int sdma_v4_4_2_soft_reset(struct amdgpu_ip_block *ip_block) { /* todo */ @@ -1857,9 +1855,9 @@ static void sdma_v4_4_2_get_clockgating_state(void *handle, u64 *flags) *flags |= AMD_CG_SUPPORT_SDMA_LS; } -static void sdma_v4_4_2_print_ip_state(void *handle, struct drm_printer *p) +static void sdma_v4_4_2_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_4_4_2); uint32_t instance_offset; @@ -1878,9 +1876,9 @@ static void sdma_v4_4_2_print_ip_state(void *handle, struct drm_printer *p) } } -static void sdma_v4_4_2_dump_ip_state(void *handle) +static void sdma_v4_4_2_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t instance_offset; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_4_4_2); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 3e48ea38385d..d31c4860933f 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -705,14 +705,16 @@ static void sdma_v5_0_enable(struct amdgpu_device *adev, bool enable) } /** - * sdma_v5_0_gfx_resume - setup and start the async dma engines + * sdma_v5_0_gfx_resume_instance - start/restart a certain sdma engine * * @adev: amdgpu_device pointer + * @i: instance + * @restore: used to restore wptr when restart * - * Set up the gfx DMA ring buffers and enable them (NAVI10). - * Returns 0 for success, error for failure. + * Set up the gfx DMA ring buffers and enable them. On restart, we will restore wptr and rptr. + * Return 0 for success. */ -static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) +static int sdma_v5_0_gfx_resume_instance(struct amdgpu_device *adev, int i, bool restore) { struct amdgpu_ring *ring; u32 rb_cntl, ib_cntl; @@ -722,142 +724,163 @@ static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) u32 temp; u32 wptr_poll_cntl; u64 wptr_gpu_addr; - int i, r; - for (i = 0; i < adev->sdma.num_instances; i++) { - ring = &adev->sdma.instance[i].ring; + ring = &adev->sdma.instance[i].ring; - if (!amdgpu_sriov_vf(adev)) - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); + if (!amdgpu_sriov_vf(adev)) + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); - /* Set ring buffer size in dwords */ - rb_bufsz = order_base_2(ring->ring_size / 4); - rb_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL)); - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SIZE, rb_bufsz); + /* Set ring buffer size in dwords */ + rb_bufsz = order_base_2(ring->ring_size / 4); + rb_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL)); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SIZE, rb_bufsz); #ifdef __BIG_ENDIAN - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SWAP_ENABLE, 1); - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, - RPTR_WRITEBACK_SWAP_ENABLE, 1); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SWAP_ENABLE, 1); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, + RPTR_WRITEBACK_SWAP_ENABLE, 1); #endif - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); - /* Initialize the ring buffer's read and write pointers */ + /* Initialize the ring buffer's read and write pointers */ + if (restore) { + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR), lower_32_bits(ring->wptr << 2)); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_HI), upper_32_bits(ring->wptr << 2)); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), lower_32_bits(ring->wptr << 2)); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), upper_32_bits(ring->wptr << 2)); + } else { WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR), 0); WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_HI), 0); WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), 0); WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), 0); + } + /* setup the wptr shadow polling */ + wptr_gpu_addr = ring->wptr_gpu_addr; + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_LO), + lower_32_bits(wptr_gpu_addr)); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_HI), + upper_32_bits(wptr_gpu_addr)); + wptr_poll_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, + mmSDMA0_GFX_RB_WPTR_POLL_CNTL)); + wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl, + SDMA0_GFX_RB_WPTR_POLL_CNTL, + F32_POLL_ENABLE, 1); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL), + wptr_poll_cntl); - /* setup the wptr shadow polling */ - wptr_gpu_addr = ring->wptr_gpu_addr; - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_LO), - lower_32_bits(wptr_gpu_addr)); - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_HI), - upper_32_bits(wptr_gpu_addr)); - wptr_poll_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, - mmSDMA0_GFX_RB_WPTR_POLL_CNTL)); - wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl, - SDMA0_GFX_RB_WPTR_POLL_CNTL, - F32_POLL_ENABLE, 1); - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL), - wptr_poll_cntl); + /* set the wb address whether it's enabled or not */ + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_HI), + upper_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFF); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_LO), + lower_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFC); - /* set the wb address whether it's enabled or not */ - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_HI), - upper_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFF); - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_LO), - lower_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFC); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1); - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1); - - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE), - ring->gpu_addr >> 8); - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE_HI), - ring->gpu_addr >> 40); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE), + ring->gpu_addr >> 8); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE_HI), + ring->gpu_addr >> 40); + if (!restore) ring->wptr = 0; - /* before programing wptr to a less value, need set minor_ptr_update first */ - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 1); + /* before programing wptr to a less value, need set minor_ptr_update first */ + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 1); - if (!amdgpu_sriov_vf(adev)) { /* only bare-metal use register write for wptr */ - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), - lower_32_bits(ring->wptr << 2)); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), - upper_32_bits(ring->wptr << 2)); - } + if (!amdgpu_sriov_vf(adev)) { /* only bare-metal use register write for wptr */ + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), + lower_32_bits(ring->wptr << 2)); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), + upper_32_bits(ring->wptr << 2)); + } - doorbell = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL)); - doorbell_offset = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, - mmSDMA0_GFX_DOORBELL_OFFSET)); + doorbell = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL)); + doorbell_offset = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, + mmSDMA0_GFX_DOORBELL_OFFSET)); - if (ring->use_doorbell) { - doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 1); - doorbell_offset = REG_SET_FIELD(doorbell_offset, SDMA0_GFX_DOORBELL_OFFSET, - OFFSET, ring->doorbell_index); - } else { - doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 0); - } - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL), doorbell); - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET), - doorbell_offset); + if (ring->use_doorbell) { + doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 1); + doorbell_offset = REG_SET_FIELD(doorbell_offset, SDMA0_GFX_DOORBELL_OFFSET, + OFFSET, ring->doorbell_index); + } else { + doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 0); + } + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL), doorbell); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET), + doorbell_offset); - adev->nbio.funcs->sdma_doorbell_range(adev, i, ring->use_doorbell, - ring->doorbell_index, 20); + adev->nbio.funcs->sdma_doorbell_range(adev, i, ring->use_doorbell, + ring->doorbell_index, 20); - if (amdgpu_sriov_vf(adev)) - sdma_v5_0_ring_set_wptr(ring); + if (amdgpu_sriov_vf(adev)) + sdma_v5_0_ring_set_wptr(ring); - /* set minor_ptr_update to 0 after wptr programed */ - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 0); + /* set minor_ptr_update to 0 after wptr programed */ + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 0); - if (!amdgpu_sriov_vf(adev)) { - /* set utc l1 enable flag always to 1 */ - temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1); + if (!amdgpu_sriov_vf(adev)) { + /* set utc l1 enable flag always to 1 */ + temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1); - /* enable MCBP */ - temp = REG_SET_FIELD(temp, SDMA0_CNTL, MIDCMD_PREEMPT_ENABLE, 1); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), temp); + /* enable MCBP */ + temp = REG_SET_FIELD(temp, SDMA0_CNTL, MIDCMD_PREEMPT_ENABLE, 1); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), temp); - /* Set up RESP_MODE to non-copy addresses */ - temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL), temp); + /* Set up RESP_MODE to non-copy addresses */ + temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL), temp); - /* program default cache read and write policy */ - temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE)); - /* clean read policy and write policy bits */ - temp &= 0xFF0FFF; - temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | (CACHE_WRITE_POLICY_L2__DEFAULT << 14)); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE), temp); - } + /* program default cache read and write policy */ + temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE)); + /* clean read policy and write policy bits */ + temp &= 0xFF0FFF; + temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | (CACHE_WRITE_POLICY_L2__DEFAULT << 14)); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE), temp); + } - if (!amdgpu_sriov_vf(adev)) { - /* unhalt engine */ - temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0); - WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), temp); - } + if (!amdgpu_sriov_vf(adev)) { + /* unhalt engine */ + temp = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), temp); + } - /* enable DMA RB */ - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 1); - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); + /* enable DMA RB */ + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 1); + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); - ib_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL)); - ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 1); + ib_cntl = RREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL)); + ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 1); #ifdef __BIG_ENDIAN - ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_SWAP_ENABLE, 1); + ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_SWAP_ENABLE, 1); #endif - /* enable DMA IBs */ - WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); + /* enable DMA IBs */ + WREG32_SOC15_IP(GC, sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); - if (amdgpu_sriov_vf(adev)) { /* bare-metal sequence doesn't need below to lines */ - sdma_v5_0_ctx_switch_enable(adev, true); - sdma_v5_0_enable(adev, true); - } + if (amdgpu_sriov_vf(adev)) { /* bare-metal sequence doesn't need below to lines */ + sdma_v5_0_ctx_switch_enable(adev, true); + sdma_v5_0_enable(adev, true); + } - r = amdgpu_ring_test_helper(ring); + return amdgpu_ring_test_helper(ring); +} + +/** + * sdma_v5_0_gfx_resume - setup and start the async dma engines + * + * @adev: amdgpu_device pointer + * + * Set up the gfx DMA ring buffers and enable them (NAVI10). + * Returns 0 for success, error for failure. + */ +static int sdma_v5_0_gfx_resume(struct amdgpu_device *adev) +{ + int i, r; + + for (i = 0; i < adev->sdma.num_instances; i++) { + r = sdma_v5_0_gfx_resume_instance(adev, i, false); if (r) return r; } @@ -1366,9 +1389,9 @@ static void sdma_v5_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring, amdgpu_ring_emit_reg_wait(ring, reg1, mask, mask); } -static int sdma_v5_0_early_init(void *handle) +static int sdma_v5_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = sdma_v5_0_init_microcode(adev); @@ -1385,11 +1408,11 @@ static int sdma_v5_0_early_init(void *handle) } -static int sdma_v5_0_sw_init(void *handle) +static int sdma_v5_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_5_0); uint32_t *ptr; @@ -1439,9 +1462,9 @@ static int sdma_v5_0_sw_init(void *handle) return r; } -static int sdma_v5_0_sw_fini(void *handle) +static int sdma_v5_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->sdma.num_instances; i++) @@ -1454,10 +1477,10 @@ static int sdma_v5_0_sw_fini(void *handle) return 0; } -static int sdma_v5_0_hw_init(void *handle) +static int sdma_v5_0_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; sdma_v5_0_init_golden_registers(adev); @@ -1466,9 +1489,9 @@ static int sdma_v5_0_hw_init(void *handle) return r; } -static int sdma_v5_0_hw_fini(void *handle) +static int sdma_v5_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) return 0; @@ -1479,18 +1502,14 @@ static int sdma_v5_0_hw_fini(void *handle) return 0; } -static int sdma_v5_0_suspend(void *handle) +static int sdma_v5_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v5_0_hw_fini(adev); + return sdma_v5_0_hw_fini(ip_block); } -static int sdma_v5_0_resume(void *handle) +static int sdma_v5_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v5_0_hw_init(adev); + return sdma_v5_0_hw_init(ip_block); } static bool sdma_v5_0_is_idle(void *handle) @@ -1508,11 +1527,11 @@ static bool sdma_v5_0_is_idle(void *handle) return true; } -static int sdma_v5_0_wait_for_idle(void *handle) +static int sdma_v5_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 sdma0, sdma1; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { sdma0 = RREG32(sdma_v5_0_get_reg_offset(adev, 0, mmSDMA0_STATUS_REG)); @@ -1525,13 +1544,100 @@ static int sdma_v5_0_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int sdma_v5_0_soft_reset(void *handle) +static int sdma_v5_0_soft_reset(struct amdgpu_ip_block *ip_block) { /* todo */ return 0; } +static int sdma_v5_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +{ + struct amdgpu_device *adev = ring->adev; + int i, j, r; + u32 rb_cntl, ib_cntl, f32_cntl, freeze, cntl, preempt, soft_reset, stat1_reg; + + if (amdgpu_sriov_vf(adev)) + return -EINVAL; + + for (i = 0; i < adev->sdma.num_instances; i++) { + if (ring == &adev->sdma.instance[i].ring) + break; + } + + if (i == adev->sdma.num_instances) { + DRM_ERROR("sdma instance not found\n"); + return -EINVAL; + } + + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); + + /* stop queue */ + ib_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL)); + ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); + + rb_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL)); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); + + /* engine stop SDMA1_F32_CNTL.HALT to 1 and SDMAx_FREEZE freeze bit to 1 */ + freeze = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_FREEZE)); + freeze = REG_SET_FIELD(freeze, SDMA0_FREEZE, FREEZE, 1); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_FREEZE), freeze); + + for (j = 0; j < adev->usec_timeout; j++) { + freeze = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_FREEZE)); + if (REG_GET_FIELD(freeze, SDMA0_FREEZE, FROZEN) & 1) + break; + udelay(1); + } + + /* check sdma copy engine all idle if frozen not received*/ + if (j == adev->usec_timeout) { + stat1_reg = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_STATUS1_REG)); + if ((stat1_reg & 0x3FF) != 0x3FF) { + DRM_ERROR("cannot soft reset as sdma not idle\n"); + r = -ETIMEDOUT; + goto err0; + } + } + + f32_cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL)); + f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_F32_CNTL, HALT, 1); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), f32_cntl); + + cntl = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL)); + cntl = REG_SET_FIELD(cntl, SDMA0_CNTL, UTC_L1_ENABLE, 0); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_CNTL), cntl); + + /* soft reset SDMA_GFX_PREEMPT.IB_PREEMPT = 0 mmGRBM_SOFT_RESET.SOFT_RESET_SDMA0/1 = 1 */ + preempt = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_PREEMPT)); + preempt = REG_SET_FIELD(preempt, SDMA0_GFX_PREEMPT, IB_PREEMPT, 0); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_GFX_PREEMPT), preempt); + + soft_reset = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); + soft_reset |= 1 << GRBM_SOFT_RESET__SOFT_RESET_SDMA0__SHIFT << i; + + WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, soft_reset); + + udelay(50); + + soft_reset &= ~(1 << GRBM_SOFT_RESET__SOFT_RESET_SDMA0__SHIFT << i); + WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, soft_reset); + + /* unfreeze*/ + freeze = RREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_FREEZE)); + freeze = REG_SET_FIELD(freeze, SDMA0_FREEZE, FREEZE, 0); + WREG32(sdma_v5_0_get_reg_offset(adev, i, mmSDMA0_FREEZE), freeze); + + r = sdma_v5_0_gfx_resume_instance(adev, i, true); + +err0: + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); + return r; +} + static int sdma_v5_0_ring_preempt_ib(struct amdgpu_ring *ring) { int i, r = 0; @@ -1778,9 +1884,9 @@ static void sdma_v5_0_get_clockgating_state(void *handle, u64 *flags) *flags |= AMD_CG_SUPPORT_SDMA_LS; } -static void sdma_v5_0_print_ip_state(void *handle, struct drm_printer *p) +static void sdma_v5_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_5_0); uint32_t instance_offset; @@ -1799,9 +1905,9 @@ static void sdma_v5_0_print_ip_state(void *handle, struct drm_printer *p) } } -static void sdma_v5_0_dump_ip_state(void *handle) +static void sdma_v5_0_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t instance_offset; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_5_0); @@ -1823,7 +1929,6 @@ static void sdma_v5_0_dump_ip_state(void *handle) static const struct amd_ip_funcs sdma_v5_0_ip_funcs = { .name = "sdma_v5_0", .early_init = sdma_v5_0_early_init, - .late_init = NULL, .sw_init = sdma_v5_0_sw_init, .sw_fini = sdma_v5_0_sw_fini, .hw_init = sdma_v5_0_hw_init, @@ -1874,6 +1979,7 @@ static const struct amdgpu_ring_funcs sdma_v5_0_ring_funcs = { .emit_reg_write_reg_wait = sdma_v5_0_ring_emit_reg_write_reg_wait, .init_cond_exec = sdma_v5_0_ring_init_cond_exec, .preempt_ib = sdma_v5_0_ring_preempt_ib, + .reset = sdma_v5_0_reset_queue, }; static void sdma_v5_0_set_ring_funcs(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index bc9b240a3488..ffa8c62ac101 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -522,14 +522,17 @@ static void sdma_v5_2_enable(struct amdgpu_device *adev, bool enable) } /** - * sdma_v5_2_gfx_resume - setup and start the async dma engines + * sdma_v5_2_gfx_resume_instance - start/restart a certain sdma engine * * @adev: amdgpu_device pointer + * @i: instance + * @restore: used to restore wptr when restart * - * Set up the gfx DMA ring buffers and enable them. - * Returns 0 for success, error for failure. + * Set up the gfx DMA ring buffers and enable them. On restart, we will restore wptr and rptr. + * Return 0 for success. */ -static int sdma_v5_2_gfx_resume(struct amdgpu_device *adev) + +static int sdma_v5_2_gfx_resume_instance(struct amdgpu_device *adev, int i, bool restore) { struct amdgpu_ring *ring; u32 rb_cntl, ib_cntl; @@ -539,139 +542,161 @@ static int sdma_v5_2_gfx_resume(struct amdgpu_device *adev) u32 temp; u32 wptr_poll_cntl; u64 wptr_gpu_addr; - int i, r; - for (i = 0; i < adev->sdma.num_instances; i++) { - ring = &adev->sdma.instance[i].ring; + ring = &adev->sdma.instance[i].ring; - if (!amdgpu_sriov_vf(adev)) - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); + if (!amdgpu_sriov_vf(adev)) + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); - /* Set ring buffer size in dwords */ - rb_bufsz = order_base_2(ring->ring_size / 4); - rb_cntl = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL)); - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SIZE, rb_bufsz); + /* Set ring buffer size in dwords */ + rb_bufsz = order_base_2(ring->ring_size / 4); + rb_cntl = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL)); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SIZE, rb_bufsz); #ifdef __BIG_ENDIAN - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SWAP_ENABLE, 1); - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, - RPTR_WRITEBACK_SWAP_ENABLE, 1); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SWAP_ENABLE, 1); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, + RPTR_WRITEBACK_SWAP_ENABLE, 1); #endif - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); - /* Initialize the ring buffer's read and write pointers */ + /* Initialize the ring buffer's read and write pointers */ + if (restore) { + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR), lower_32_bits(ring->wptr << 2)); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_HI), upper_32_bits(ring->wptr << 2)); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), lower_32_bits(ring->wptr << 2)); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), upper_32_bits(ring->wptr << 2)); + } else { WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR), 0); WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_HI), 0); WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), 0); WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), 0); + } - /* setup the wptr shadow polling */ - wptr_gpu_addr = ring->wptr_gpu_addr; - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_LO), - lower_32_bits(wptr_gpu_addr)); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_HI), - upper_32_bits(wptr_gpu_addr)); - wptr_poll_cntl = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, - mmSDMA0_GFX_RB_WPTR_POLL_CNTL)); - wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl, - SDMA0_GFX_RB_WPTR_POLL_CNTL, - F32_POLL_ENABLE, 1); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL), - wptr_poll_cntl); + /* setup the wptr shadow polling */ + wptr_gpu_addr = ring->wptr_gpu_addr; + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_LO), + lower_32_bits(wptr_gpu_addr)); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_HI), + upper_32_bits(wptr_gpu_addr)); + wptr_poll_cntl = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, + mmSDMA0_GFX_RB_WPTR_POLL_CNTL)); + wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl, + SDMA0_GFX_RB_WPTR_POLL_CNTL, + F32_POLL_ENABLE, 1); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL), + wptr_poll_cntl); - /* set the wb address whether it's enabled or not */ - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_HI), - upper_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFF); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_LO), - lower_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFC); + /* set the wb address whether it's enabled or not */ + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_HI), + upper_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFF); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_LO), + lower_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFC); - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE), ring->gpu_addr >> 8); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE_HI), ring->gpu_addr >> 40); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE), ring->gpu_addr >> 8); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE_HI), ring->gpu_addr >> 40); + if (!restore) ring->wptr = 0; - /* before programing wptr to a less value, need set minor_ptr_update first */ - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 1); + /* before programing wptr to a less value, need set minor_ptr_update first */ + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 1); - if (!amdgpu_sriov_vf(adev)) { /* only bare-metal use register write for wptr */ - WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), lower_32_bits(ring->wptr << 2)); - WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), upper_32_bits(ring->wptr << 2)); - } + if (!amdgpu_sriov_vf(adev)) { /* only bare-metal use register write for wptr */ + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), lower_32_bits(ring->wptr << 2)); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), upper_32_bits(ring->wptr << 2)); + } - doorbell = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL)); - doorbell_offset = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET)); + doorbell = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL)); + doorbell_offset = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET)); - if (ring->use_doorbell) { - doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 1); - doorbell_offset = REG_SET_FIELD(doorbell_offset, SDMA0_GFX_DOORBELL_OFFSET, - OFFSET, ring->doorbell_index); - } else { - doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 0); - } - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL), doorbell); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET), doorbell_offset); + if (ring->use_doorbell) { + doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 1); + doorbell_offset = REG_SET_FIELD(doorbell_offset, SDMA0_GFX_DOORBELL_OFFSET, + OFFSET, ring->doorbell_index); + } else { + doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 0); + } + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL), doorbell); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET), doorbell_offset); - adev->nbio.funcs->sdma_doorbell_range(adev, i, ring->use_doorbell, - ring->doorbell_index, - adev->doorbell_index.sdma_doorbell_range); + adev->nbio.funcs->sdma_doorbell_range(adev, i, ring->use_doorbell, + ring->doorbell_index, + adev->doorbell_index.sdma_doorbell_range); - if (amdgpu_sriov_vf(adev)) - sdma_v5_2_ring_set_wptr(ring); + if (amdgpu_sriov_vf(adev)) + sdma_v5_2_ring_set_wptr(ring); - /* set minor_ptr_update to 0 after wptr programed */ + /* set minor_ptr_update to 0 after wptr programed */ - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 0); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 0); - /* SRIOV VF has no control of any of registers below */ - if (!amdgpu_sriov_vf(adev)) { - /* set utc l1 enable flag always to 1 */ - temp = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1); + /* SRIOV VF has no control of any of registers below */ + if (!amdgpu_sriov_vf(adev)) { + /* set utc l1 enable flag always to 1 */ + temp = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1); - /* enable MCBP */ - temp = REG_SET_FIELD(temp, SDMA0_CNTL, MIDCMD_PREEMPT_ENABLE, 1); - WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL), temp); + /* enable MCBP */ + temp = REG_SET_FIELD(temp, SDMA0_CNTL, MIDCMD_PREEMPT_ENABLE, 1); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL), temp); - /* Set up RESP_MODE to non-copy addresses */ - temp = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL), temp); + /* Set up RESP_MODE to non-copy addresses */ + temp = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_CNTL), temp); - /* program default cache read and write policy */ - temp = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE)); - /* clean read policy and write policy bits */ - temp &= 0xFF0FFF; - temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | - (CACHE_WRITE_POLICY_L2__DEFAULT << 14) | - SDMA0_UTCL1_PAGE__LLC_NOALLOC_MASK); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE), temp); + /* program default cache read and write policy */ + temp = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE)); + /* clean read policy and write policy bits */ + temp &= 0xFF0FFF; + temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | + (CACHE_WRITE_POLICY_L2__DEFAULT << 14) | + SDMA0_UTCL1_PAGE__LLC_NOALLOC_MASK); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_UTCL1_PAGE), temp); - /* unhalt engine */ - temp = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_F32_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0); - WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), temp); - } + /* unhalt engine */ + temp = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_F32_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), temp); + } - /* enable DMA RB */ - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 1); - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); + /* enable DMA RB */ + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 1); + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); - ib_cntl = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL)); - ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 1); + ib_cntl = RREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL)); + ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 1); #ifdef __BIG_ENDIAN - ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_SWAP_ENABLE, 1); + ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_SWAP_ENABLE, 1); #endif - /* enable DMA IBs */ - WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); + /* enable DMA IBs */ + WREG32_SOC15_IP(GC, sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); - if (amdgpu_sriov_vf(adev)) { /* bare-metal sequence doesn't need below to lines */ - sdma_v5_2_ctx_switch_enable(adev, true); - sdma_v5_2_enable(adev, true); - } + if (amdgpu_sriov_vf(adev)) { /* bare-metal sequence doesn't need below to lines */ + sdma_v5_2_ctx_switch_enable(adev, true); + sdma_v5_2_enable(adev, true); + } - r = amdgpu_ring_test_helper(ring); + return amdgpu_ring_test_helper(ring); +} + +/** + * sdma_v5_2_gfx_resume - setup and start the async dma engines + * + * @adev: amdgpu_device pointer + * + * Set up the gfx DMA ring buffers and enable them. + * Returns 0 for success, error for failure. + */ +static int sdma_v5_2_gfx_resume(struct amdgpu_device *adev) +{ + int i, r; + + for (i = 0; i < adev->sdma.num_instances; i++) { + r = sdma_v5_2_gfx_resume_instance(adev, i, false); if (r) return r; } @@ -736,9 +761,9 @@ static int sdma_v5_2_load_microcode(struct amdgpu_device *adev) return 0; } -static int sdma_v5_2_soft_reset(void *handle) +static int sdma_v5_2_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 grbm_soft_reset; u32 tmp; int i; @@ -778,6 +803,7 @@ static int sdma_v5_2_soft_reset(void *handle) static int sdma_v5_2_start(struct amdgpu_device *adev) { int r = 0; + struct amdgpu_ip_block *ip_block; if (amdgpu_sriov_vf(adev)) { sdma_v5_2_ctx_switch_enable(adev, false); @@ -798,7 +824,11 @@ static int sdma_v5_2_start(struct amdgpu_device *adev) msleep(1000); } - sdma_v5_2_soft_reset(adev); + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_SDMA); + if (!ip_block) + return -EINVAL; + + sdma_v5_2_soft_reset(ip_block); /* unhalt the MEs */ sdma_v5_2_enable(adev, true); /* enable sdma ring preemption */ @@ -1180,7 +1210,28 @@ static void sdma_v5_2_ring_emit_pipeline_sync(struct amdgpu_ring *ring) static void sdma_v5_2_ring_emit_vm_flush(struct amdgpu_ring *ring, unsigned vmid, uint64_t pd_addr) { - amdgpu_gmc_emit_flush_gpu_tlb(ring, vmid, pd_addr); + struct amdgpu_vmhub *hub = &ring->adev->vmhub[ring->vm_hub]; + uint32_t req = hub->vmhub_funcs->get_invalidate_req(vmid, 0); + + /* Update the PD address for this VMID. */ + amdgpu_ring_emit_wreg(ring, hub->ctx0_ptb_addr_lo32 + + (hub->ctx_addr_distance * vmid), + lower_32_bits(pd_addr)); + amdgpu_ring_emit_wreg(ring, hub->ctx0_ptb_addr_hi32 + + (hub->ctx_addr_distance * vmid), + upper_32_bits(pd_addr)); + + /* Trigger invalidation. */ + amdgpu_ring_write(ring, + SDMA_PKT_VM_INVALIDATION_HEADER_OP(SDMA_OP_POLL_REGMEM) | + SDMA_PKT_VM_INVALIDATION_HEADER_SUB_OP(SDMA_SUBOP_VM_INVALIDATION) | + SDMA_PKT_VM_INVALIDATION_HEADER_GFX_ENG_ID(ring->vm_inv_eng) | + SDMA_PKT_VM_INVALIDATION_HEADER_MM_ENG_ID(0x1f)); + amdgpu_ring_write(ring, req); + amdgpu_ring_write(ring, 0xFFFFFFFF); + amdgpu_ring_write(ring, + SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_INVALIDATEACK(1 << vmid) | + SDMA_PKT_VM_INVALIDATION_ADDRESSRANGEHI_ADDRESSRANGEHI(0x1F)); } static void sdma_v5_2_ring_emit_wreg(struct amdgpu_ring *ring, @@ -1216,9 +1267,9 @@ static void sdma_v5_2_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring, amdgpu_ring_emit_reg_wait(ring, reg1, mask, mask); } -static int sdma_v5_2_early_init(void *handle) +static int sdma_v5_2_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_sdma_init_microcode(adev, 0, true); @@ -1268,11 +1319,11 @@ static unsigned sdma_v5_2_seq_to_trap_id(int seq_num) return -EINVAL; } -static int sdma_v5_2_sw_init(void *handle) +static int sdma_v5_2_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_5_2); uint32_t *ptr; @@ -1316,9 +1367,9 @@ static int sdma_v5_2_sw_init(void *handle) return r; } -static int sdma_v5_2_sw_fini(void *handle) +static int sdma_v5_2_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->sdma.num_instances; i++) @@ -1331,16 +1382,16 @@ static int sdma_v5_2_sw_fini(void *handle) return 0; } -static int sdma_v5_2_hw_init(void *handle) +static int sdma_v5_2_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return sdma_v5_2_start(adev); } -static int sdma_v5_2_hw_fini(void *handle) +static int sdma_v5_2_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) return 0; @@ -1351,18 +1402,14 @@ static int sdma_v5_2_hw_fini(void *handle) return 0; } -static int sdma_v5_2_suspend(void *handle) +static int sdma_v5_2_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v5_2_hw_fini(adev); + return sdma_v5_2_hw_fini(ip_block); } -static int sdma_v5_2_resume(void *handle) +static int sdma_v5_2_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v5_2_hw_init(adev); + return sdma_v5_2_hw_init(ip_block); } static bool sdma_v5_2_is_idle(void *handle) @@ -1380,11 +1427,11 @@ static bool sdma_v5_2_is_idle(void *handle) return true; } -static int sdma_v5_2_wait_for_idle(void *handle) +static int sdma_v5_2_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 sdma0, sdma1, sdma2, sdma3; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { sdma0 = RREG32(sdma_v5_2_get_reg_offset(adev, 0, mmSDMA0_STATUS_REG)); @@ -1399,6 +1446,96 @@ static int sdma_v5_2_wait_for_idle(void *handle) return -ETIMEDOUT; } +static int sdma_v5_2_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +{ + struct amdgpu_device *adev = ring->adev; + int i, j, r; + u32 rb_cntl, ib_cntl, f32_cntl, freeze, cntl, preempt, soft_reset, stat1_reg; + + if (amdgpu_sriov_vf(adev)) + return -EINVAL; + + for (i = 0; i < adev->sdma.num_instances; i++) { + if (ring == &adev->sdma.instance[i].ring) + break; + } + + if (i == adev->sdma.num_instances) { + DRM_ERROR("sdma instance not found\n"); + return -EINVAL; + } + + amdgpu_gfx_rlc_enter_safe_mode(adev, 0); + + /* stop queue */ + ib_cntl = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL)); + ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl); + + rb_cntl = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL)); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl); + + /*engine stop SDMA1_F32_CNTL.HALT to 1 and SDMAx_FREEZE freeze bit to 1 */ + freeze = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_FREEZE)); + freeze = REG_SET_FIELD(freeze, SDMA0_FREEZE, FREEZE, 1); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_FREEZE), freeze); + + for (j = 0; j < adev->usec_timeout; j++) { + freeze = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_FREEZE)); + + if (REG_GET_FIELD(freeze, SDMA0_FREEZE, FROZEN) & 1) + break; + udelay(1); + } + + + if (j == adev->usec_timeout) { + stat1_reg = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_STATUS1_REG)); + if ((stat1_reg & 0x3FF) != 0x3FF) { + DRM_ERROR("cannot soft reset as sdma not idle\n"); + r = -ETIMEDOUT; + goto err0; + } + } + + f32_cntl = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_F32_CNTL)); + f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_F32_CNTL, HALT, 1); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), f32_cntl); + + cntl = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL)); + cntl = REG_SET_FIELD(cntl, SDMA0_CNTL, UTC_L1_ENABLE, 0); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_CNTL), cntl); + + /* soft reset SDMA_GFX_PREEMPT.IB_PREEMPT = 0 mmGRBM_SOFT_RESET.SOFT_RESET_SDMA0/1 = 1 */ + preempt = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_PREEMPT)); + preempt = REG_SET_FIELD(preempt, SDMA0_GFX_PREEMPT, IB_PREEMPT, 0); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_GFX_PREEMPT), preempt); + + soft_reset = RREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET); + soft_reset |= 1 << GRBM_SOFT_RESET__SOFT_RESET_SDMA0__SHIFT << i; + + + WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, soft_reset); + + udelay(50); + + soft_reset &= ~(1 << GRBM_SOFT_RESET__SOFT_RESET_SDMA0__SHIFT << i); + + WREG32_SOC15(GC, 0, mmGRBM_SOFT_RESET, soft_reset); + + /* unfreeze and unhalt */ + freeze = RREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_FREEZE)); + freeze = REG_SET_FIELD(freeze, SDMA0_FREEZE, FREEZE, 0); + WREG32(sdma_v5_2_get_reg_offset(adev, i, mmSDMA0_FREEZE), freeze); + + r = sdma_v5_2_gfx_resume_instance(adev, i, true); + +err0: + amdgpu_gfx_rlc_exit_safe_mode(adev, 0); + return r; +} + static int sdma_v5_2_ring_preempt_ib(struct amdgpu_ring *ring) { int i, r = 0; @@ -1736,9 +1873,9 @@ static void sdma_v5_2_ring_end_use(struct amdgpu_ring *ring) amdgpu_gfx_off_ctrl(adev, true); } -static void sdma_v5_2_print_ip_state(void *handle, struct drm_printer *p) +static void sdma_v5_2_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_5_2); uint32_t instance_offset; @@ -1757,9 +1894,9 @@ static void sdma_v5_2_print_ip_state(void *handle, struct drm_printer *p) } } -static void sdma_v5_2_dump_ip_state(void *handle) +static void sdma_v5_2_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t instance_offset; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_5_2); @@ -1781,7 +1918,6 @@ static void sdma_v5_2_dump_ip_state(void *handle) static const struct amd_ip_funcs sdma_v5_2_ip_funcs = { .name = "sdma_v5_2", .early_init = sdma_v5_2_early_init, - .late_init = NULL, .sw_init = sdma_v5_2_sw_init, .sw_fini = sdma_v5_2_sw_fini, .hw_init = sdma_v5_2_hw_init, @@ -1834,6 +1970,7 @@ static const struct amdgpu_ring_funcs sdma_v5_2_ring_funcs = { .emit_reg_write_reg_wait = sdma_v5_2_ring_emit_reg_write_reg_wait, .init_cond_exec = sdma_v5_2_ring_init_cond_exec, .preempt_ib = sdma_v5_2_ring_preempt_ib, + .reset = sdma_v5_2_reset_queue, }; static void sdma_v5_2_set_ring_funcs(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c index 208a1fa9d4e7..5635f2d84090 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c @@ -468,6 +468,157 @@ static void sdma_v6_0_enable(struct amdgpu_device *adev, bool enable) } } +/** + * sdma_v6_0_gfx_resume_instance - start/restart a certain sdma engine + * + * @adev: amdgpu_device pointer + * @i: instance + * @restore: used to restore wptr when restart + * + * Set up the gfx DMA ring buffers and enable them. On restart, we will restore wptr and rptr. + * Return 0 for success. + */ +static int sdma_v6_0_gfx_resume_instance(struct amdgpu_device *adev, int i, bool restore) +{ + struct amdgpu_ring *ring; + u32 rb_cntl, ib_cntl; + u32 rb_bufsz; + u32 doorbell; + u32 doorbell_offset; + u32 temp; + u64 wptr_gpu_addr; + + ring = &adev->sdma.instance[i].ring; + if (!amdgpu_sriov_vf(adev)) + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); + + /* Set ring buffer size in dwords */ + rb_bufsz = order_base_2(ring->ring_size / 4); + rb_cntl = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_CNTL)); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RB_SIZE, rb_bufsz); +#ifdef __BIG_ENDIAN + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RB_SWAP_ENABLE, 1); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, + RPTR_WRITEBACK_SWAP_ENABLE, 1); +#endif + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RB_PRIV, 1); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_CNTL), rb_cntl); + + /* Initialize the ring buffer's read and write pointers */ + if (restore) { + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_RPTR), lower_32_bits(ring->wptr << 2)); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_RPTR_HI), upper_32_bits(ring->wptr << 2)); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR), lower_32_bits(ring->wptr << 2)); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR_HI), upper_32_bits(ring->wptr << 2)); + } else { + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_RPTR), 0); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_RPTR_HI), 0); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR), 0); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR_HI), 0); + } + /* setup the wptr shadow polling */ + wptr_gpu_addr = ring->wptr_gpu_addr; + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR_POLL_ADDR_LO), + lower_32_bits(wptr_gpu_addr)); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR_POLL_ADDR_HI), + upper_32_bits(wptr_gpu_addr)); + + /* set the wb address whether it's enabled or not */ + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_RPTR_ADDR_HI), + upper_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFF); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_RPTR_ADDR_LO), + lower_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFC); + + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, WPTR_POLL_ENABLE, 0); + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, F32_WPTR_POLL_ENABLE, 1); + + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_BASE), ring->gpu_addr >> 8); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_BASE_HI), ring->gpu_addr >> 40); + + if (!restore) + ring->wptr = 0; + + /* before programing wptr to a less value, need set minor_ptr_update first */ + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_MINOR_PTR_UPDATE), 1); + + if (!amdgpu_sriov_vf(adev)) { /* only bare-metal use register write for wptr */ + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR), lower_32_bits(ring->wptr) << 2); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR_HI), upper_32_bits(ring->wptr) << 2); + } + + doorbell = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_DOORBELL)); + doorbell_offset = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_DOORBELL_OFFSET)); + + if (ring->use_doorbell) { + doorbell = REG_SET_FIELD(doorbell, SDMA0_QUEUE0_DOORBELL, ENABLE, 1); + doorbell_offset = REG_SET_FIELD(doorbell_offset, SDMA0_QUEUE0_DOORBELL_OFFSET, + OFFSET, ring->doorbell_index); + } else { + doorbell = REG_SET_FIELD(doorbell, SDMA0_QUEUE0_DOORBELL, ENABLE, 0); + } + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_DOORBELL), doorbell); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_DOORBELL_OFFSET), doorbell_offset); + + if (i == 0) + adev->nbio.funcs->sdma_doorbell_range(adev, i, ring->use_doorbell, + ring->doorbell_index, + adev->doorbell_index.sdma_doorbell_range * adev->sdma.num_instances); + + if (amdgpu_sriov_vf(adev)) + sdma_v6_0_ring_set_wptr(ring); + + /* set minor_ptr_update to 0 after wptr programed */ + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_MINOR_PTR_UPDATE), 0); + + /* Set up sdma hang watchdog */ + temp = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_WATCHDOG_CNTL)); + /* 100ms per unit */ + temp = REG_SET_FIELD(temp, SDMA0_WATCHDOG_CNTL, QUEUE_HANG_COUNT, + max(adev->usec_timeout/100000, 1)); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_WATCHDOG_CNTL), temp); + + /* Set up RESP_MODE to non-copy addresses */ + temp = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_UTCL1_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); + temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_UTCL1_CNTL), temp); + + /* program default cache read and write policy */ + temp = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_UTCL1_PAGE)); + /* clean read policy and write policy bits */ + temp &= 0xFF0FFF; + temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | + (CACHE_WRITE_POLICY_L2__DEFAULT << 14) | + SDMA0_UTCL1_PAGE__LLC_NOALLOC_MASK); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_UTCL1_PAGE), temp); + + if (!amdgpu_sriov_vf(adev)) { + /* unhalt engine */ + temp = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_F32_CNTL)); + temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0); + temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, TH1_RESET, 0); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_F32_CNTL), temp); + } + + /* enable DMA RB */ + rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RB_ENABLE, 1); + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_CNTL), rb_cntl); + + ib_cntl = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_IB_CNTL)); + ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_QUEUE0_IB_CNTL, IB_ENABLE, 1); +#ifdef __BIG_ENDIAN + ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_QUEUE0_IB_CNTL, IB_SWAP_ENABLE, 1); +#endif + /* enable DMA IBs */ + WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_IB_CNTL), ib_cntl); + + if (amdgpu_sriov_vf(adev)) + sdma_v6_0_enable(adev, true); + + return amdgpu_ring_test_helper(ring); +} + /** * sdma_v6_0_gfx_resume - setup and start the async dma engines * @@ -478,139 +629,10 @@ static void sdma_v6_0_enable(struct amdgpu_device *adev, bool enable) */ static int sdma_v6_0_gfx_resume(struct amdgpu_device *adev) { - struct amdgpu_ring *ring; - u32 rb_cntl, ib_cntl; - u32 rb_bufsz; - u32 doorbell; - u32 doorbell_offset; - u32 temp; - u64 wptr_gpu_addr; int i, r; for (i = 0; i < adev->sdma.num_instances; i++) { - ring = &adev->sdma.instance[i].ring; - - if (!amdgpu_sriov_vf(adev)) - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0); - - /* Set ring buffer size in dwords */ - rb_bufsz = order_base_2(ring->ring_size / 4); - rb_cntl = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_CNTL)); - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RB_SIZE, rb_bufsz); -#ifdef __BIG_ENDIAN - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RB_SWAP_ENABLE, 1); - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, - RPTR_WRITEBACK_SWAP_ENABLE, 1); -#endif - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RB_PRIV, 1); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_CNTL), rb_cntl); - - /* Initialize the ring buffer's read and write pointers */ - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_RPTR), 0); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_RPTR_HI), 0); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR), 0); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR_HI), 0); - - /* setup the wptr shadow polling */ - wptr_gpu_addr = ring->wptr_gpu_addr; - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR_POLL_ADDR_LO), - lower_32_bits(wptr_gpu_addr)); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR_POLL_ADDR_HI), - upper_32_bits(wptr_gpu_addr)); - - /* set the wb address whether it's enabled or not */ - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_RPTR_ADDR_HI), - upper_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFF); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_RPTR_ADDR_LO), - lower_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFC); - - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1); - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, WPTR_POLL_ENABLE, 0); - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, F32_WPTR_POLL_ENABLE, 1); - - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_BASE), ring->gpu_addr >> 8); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_BASE_HI), ring->gpu_addr >> 40); - - ring->wptr = 0; - - /* before programing wptr to a less value, need set minor_ptr_update first */ - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_MINOR_PTR_UPDATE), 1); - - if (!amdgpu_sriov_vf(adev)) { /* only bare-metal use register write for wptr */ - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR), lower_32_bits(ring->wptr) << 2); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_WPTR_HI), upper_32_bits(ring->wptr) << 2); - } - - doorbell = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_DOORBELL)); - doorbell_offset = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_DOORBELL_OFFSET)); - - if (ring->use_doorbell) { - doorbell = REG_SET_FIELD(doorbell, SDMA0_QUEUE0_DOORBELL, ENABLE, 1); - doorbell_offset = REG_SET_FIELD(doorbell_offset, SDMA0_QUEUE0_DOORBELL_OFFSET, - OFFSET, ring->doorbell_index); - } else { - doorbell = REG_SET_FIELD(doorbell, SDMA0_QUEUE0_DOORBELL, ENABLE, 0); - } - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_DOORBELL), doorbell); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_DOORBELL_OFFSET), doorbell_offset); - - if (i == 0) - adev->nbio.funcs->sdma_doorbell_range(adev, i, ring->use_doorbell, - ring->doorbell_index, - adev->doorbell_index.sdma_doorbell_range * adev->sdma.num_instances); - - if (amdgpu_sriov_vf(adev)) - sdma_v6_0_ring_set_wptr(ring); - - /* set minor_ptr_update to 0 after wptr programed */ - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_MINOR_PTR_UPDATE), 0); - - /* Set up sdma hang watchdog */ - temp = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_WATCHDOG_CNTL)); - /* 100ms per unit */ - temp = REG_SET_FIELD(temp, SDMA0_WATCHDOG_CNTL, QUEUE_HANG_COUNT, - max(adev->usec_timeout/100000, 1)); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_WATCHDOG_CNTL), temp); - - /* Set up RESP_MODE to non-copy addresses */ - temp = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_UTCL1_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, RESP_MODE, 3); - temp = REG_SET_FIELD(temp, SDMA0_UTCL1_CNTL, REDO_DELAY, 9); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_UTCL1_CNTL), temp); - - /* program default cache read and write policy */ - temp = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_UTCL1_PAGE)); - /* clean read policy and write policy bits */ - temp &= 0xFF0FFF; - temp |= ((CACHE_READ_POLICY_L2__DEFAULT << 12) | - (CACHE_WRITE_POLICY_L2__DEFAULT << 14) | - SDMA0_UTCL1_PAGE__LLC_NOALLOC_MASK); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_UTCL1_PAGE), temp); - - if (!amdgpu_sriov_vf(adev)) { - /* unhalt engine */ - temp = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_F32_CNTL)); - temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0); - temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, TH1_RESET, 0); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_F32_CNTL), temp); - } - - /* enable DMA RB */ - rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RB_ENABLE, 1); - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_CNTL), rb_cntl); - - ib_cntl = RREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_IB_CNTL)); - ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_QUEUE0_IB_CNTL, IB_ENABLE, 1); -#ifdef __BIG_ENDIAN - ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_QUEUE0_IB_CNTL, IB_SWAP_ENABLE, 1); -#endif - /* enable DMA IBs */ - WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_IB_CNTL), ib_cntl); - - if (amdgpu_sriov_vf(adev)) - sdma_v6_0_enable(adev, true); - - r = amdgpu_ring_test_helper(ring); + r = sdma_v6_0_gfx_resume_instance(adev, i, false); if (r) return r; } @@ -733,9 +755,9 @@ static int sdma_v6_0_load_microcode(struct amdgpu_device *adev) return 0; } -static int sdma_v6_0_soft_reset(void *handle) +static int sdma_v6_0_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 tmp; int i; @@ -769,9 +791,9 @@ static int sdma_v6_0_soft_reset(void *handle) return sdma_v6_0_start(adev); } -static bool sdma_v6_0_check_soft_reset(void *handle) +static bool sdma_v6_0_check_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, r; long tmo = msecs_to_jiffies(1000); @@ -1272,9 +1294,9 @@ static void sdma_v6_0_set_ras_funcs(struct amdgpu_device *adev) } } -static int sdma_v6_0_early_init(void *handle) +static int sdma_v6_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_sdma_init_microcode(adev, 0, true); @@ -1291,11 +1313,11 @@ static int sdma_v6_0_early_init(void *handle) return 0; } -static int sdma_v6_0_sw_init(void *handle) +static int sdma_v6_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_6_0); uint32_t *ptr; @@ -1343,9 +1365,9 @@ static int sdma_v6_0_sw_init(void *handle) return r; } -static int sdma_v6_0_sw_fini(void *handle) +static int sdma_v6_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->sdma.num_instances; i++) @@ -1358,16 +1380,16 @@ static int sdma_v6_0_sw_fini(void *handle) return 0; } -static int sdma_v6_0_hw_init(void *handle) +static int sdma_v6_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return sdma_v6_0_start(adev); } -static int sdma_v6_0_hw_fini(void *handle) +static int sdma_v6_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) return 0; @@ -1378,18 +1400,14 @@ static int sdma_v6_0_hw_fini(void *handle) return 0; } -static int sdma_v6_0_suspend(void *handle) +static int sdma_v6_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v6_0_hw_fini(adev); + return sdma_v6_0_hw_fini(ip_block); } -static int sdma_v6_0_resume(void *handle) +static int sdma_v6_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v6_0_hw_init(adev); + return sdma_v6_0_hw_init(ip_block); } static bool sdma_v6_0_is_idle(void *handle) @@ -1407,11 +1425,11 @@ static bool sdma_v6_0_is_idle(void *handle) return true; } -static int sdma_v6_0_wait_for_idle(void *handle) +static int sdma_v6_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 sdma0, sdma1; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { sdma0 = RREG32(sdma_v6_0_get_reg_offset(adev, 0, regSDMA0_STATUS_REG)); @@ -1469,6 +1487,31 @@ static int sdma_v6_0_ring_preempt_ib(struct amdgpu_ring *ring) return r; } +static int sdma_v6_0_reset_queue(struct amdgpu_ring *ring, unsigned int vmid) +{ + struct amdgpu_device *adev = ring->adev; + int i, r; + + if (amdgpu_sriov_vf(adev)) + return -EINVAL; + + for (i = 0; i < adev->sdma.num_instances; i++) { + if (ring == &adev->sdma.instance[i].ring) + break; + } + + if (i == adev->sdma.num_instances) { + DRM_ERROR("sdma instance not found\n"); + return -EINVAL; + } + + r = amdgpu_mes_reset_legacy_queue(adev, ring, vmid, true); + if (r) + return r; + + return sdma_v6_0_gfx_resume_instance(adev, i, true); +} + static int sdma_v6_0_set_trap_irq_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, unsigned type, @@ -1556,9 +1599,9 @@ static void sdma_v6_0_get_clockgating_state(void *handle, u64 *flags) { } -static void sdma_v6_0_print_ip_state(void *handle, struct drm_printer *p) +static void sdma_v6_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_6_0); uint32_t instance_offset; @@ -1577,9 +1620,9 @@ static void sdma_v6_0_print_ip_state(void *handle, struct drm_printer *p) } } -static void sdma_v6_0_dump_ip_state(void *handle) +static void sdma_v6_0_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t instance_offset; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_6_0); @@ -1601,7 +1644,6 @@ static void sdma_v6_0_dump_ip_state(void *handle) const struct amd_ip_funcs sdma_v6_0_ip_funcs = { .name = "sdma_v6_0", .early_init = sdma_v6_0_early_init, - .late_init = NULL, .sw_init = sdma_v6_0_sw_init, .sw_fini = sdma_v6_0_sw_fini, .hw_init = sdma_v6_0_hw_init, @@ -1652,6 +1694,7 @@ static const struct amdgpu_ring_funcs sdma_v6_0_ring_funcs = { .emit_reg_write_reg_wait = sdma_v6_0_ring_emit_reg_write_reg_wait, .init_cond_exec = sdma_v6_0_ring_init_cond_exec, .preempt_ib = sdma_v6_0_ring_preempt_ib, + .reset = sdma_v6_0_reset_queue, }; static void sdma_v6_0_set_ring_funcs(struct amdgpu_device *adev) @@ -1726,7 +1769,7 @@ static void sdma_v6_0_emit_fill_buffer(struct amdgpu_ib *ib, uint64_t dst_offset, uint32_t byte_count) { - ib->ptr[ib->length_dw++] = SDMA_PKT_COPY_LINEAR_HEADER_OP(SDMA_OP_CONST_FILL); + ib->ptr[ib->length_dw++] = SDMA_PKT_CONSTANT_FILL_HEADER_OP(SDMA_OP_CONST_FILL); ib->ptr[ib->length_dw++] = lower_32_bits(dst_offset); ib->ptr[ib->length_dw++] = upper_32_bits(dst_offset); ib->ptr[ib->length_dw++] = src_data; diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c index a8763496aed3..d2ce6b6a7ff6 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v7_0.c @@ -51,6 +51,12 @@ MODULE_FIRMWARE("amdgpu/sdma_7_0_1.bin"); #define SDMA0_HYP_DEC_REG_END 0x589a #define SDMA1_HYP_DEC_REG_OFFSET 0x20 +/*define for compression field for sdma7*/ +#define SDMA_PKT_CONSTANT_FILL_HEADER_compress_offset 0 +#define SDMA_PKT_CONSTANT_FILL_HEADER_compress_mask 0x00000001 +#define SDMA_PKT_CONSTANT_FILL_HEADER_compress_shift 16 +#define SDMA_PKT_CONSTANT_FILL_HEADER_COMPRESS(x) (((x) & SDMA_PKT_CONSTANT_FILL_HEADER_compress_mask) << SDMA_PKT_CONSTANT_FILL_HEADER_compress_shift) + static const struct amdgpu_hwip_reg_entry sdma_reg_list_7_0[] = { SOC15_REG_ENTRY_STR(GC, 0, regSDMA0_STATUS_REG), SOC15_REG_ENTRY_STR(GC, 0, regSDMA0_STATUS1_REG), @@ -747,9 +753,9 @@ static int sdma_v7_0_load_microcode(struct amdgpu_device *adev) return 0; } -static int sdma_v7_0_soft_reset(void *handle) +static int sdma_v7_0_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 tmp; int i; @@ -783,9 +789,9 @@ static int sdma_v7_0_soft_reset(void *handle) return sdma_v7_0_start(adev); } -static bool sdma_v7_0_check_soft_reset(void *handle) +static bool sdma_v7_0_check_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, r; long tmo = msecs_to_jiffies(1000); @@ -1253,9 +1259,9 @@ static void sdma_v7_0_ring_emit_reg_write_reg_wait(struct amdgpu_ring *ring, amdgpu_ring_emit_reg_wait(ring, reg1, mask, mask); } -static int sdma_v7_0_early_init(void *handle) +static int sdma_v7_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_sdma_init_microcode(adev, 0, true); @@ -1273,11 +1279,11 @@ static int sdma_v7_0_early_init(void *handle) return 0; } -static int sdma_v7_0_sw_init(void *handle) +static int sdma_v7_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_7_0); uint32_t *ptr; @@ -1320,9 +1326,9 @@ static int sdma_v7_0_sw_init(void *handle) return r; } -static int sdma_v7_0_sw_fini(void *handle) +static int sdma_v7_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->sdma.num_instances; i++) @@ -1338,16 +1344,16 @@ static int sdma_v7_0_sw_fini(void *handle) return 0; } -static int sdma_v7_0_hw_init(void *handle) +static int sdma_v7_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return sdma_v7_0_start(adev); } -static int sdma_v7_0_hw_fini(void *handle) +static int sdma_v7_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) return 0; @@ -1358,18 +1364,14 @@ static int sdma_v7_0_hw_fini(void *handle) return 0; } -static int sdma_v7_0_suspend(void *handle) +static int sdma_v7_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v7_0_hw_fini(adev); + return sdma_v7_0_hw_fini(ip_block); } -static int sdma_v7_0_resume(void *handle) +static int sdma_v7_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return sdma_v7_0_hw_init(adev); + return sdma_v7_0_hw_init(ip_block); } static bool sdma_v7_0_is_idle(void *handle) @@ -1387,11 +1389,11 @@ static bool sdma_v7_0_is_idle(void *handle) return true; } -static int sdma_v7_0_wait_for_idle(void *handle) +static int sdma_v7_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 sdma0, sdma1; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { sdma0 = RREG32(sdma_v7_0_get_reg_offset(adev, 0, regSDMA0_STATUS_REG)); @@ -1538,9 +1540,9 @@ static void sdma_v7_0_get_clockgating_state(void *handle, u64 *flags) { } -static void sdma_v7_0_print_ip_state(void *handle, struct drm_printer *p) +static void sdma_v7_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_7_0); uint32_t instance_offset; @@ -1559,9 +1561,9 @@ static void sdma_v7_0_print_ip_state(void *handle, struct drm_printer *p) } } -static void sdma_v7_0_dump_ip_state(void *handle) +static void sdma_v7_0_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t instance_offset; uint32_t reg_count = ARRAY_SIZE(sdma_reg_list_7_0); @@ -1724,7 +1726,8 @@ static void sdma_v7_0_emit_fill_buffer(struct amdgpu_ib *ib, uint64_t dst_offset, uint32_t byte_count) { - ib->ptr[ib->length_dw++] = SDMA_PKT_COPY_LINEAR_HEADER_OP(SDMA_OP_CONST_FILL); + ib->ptr[ib->length_dw++] = SDMA_PKT_CONSTANT_FILL_HEADER_OP(SDMA_OP_CONST_FILL) | + SDMA_PKT_CONSTANT_FILL_HEADER_COMPRESS(1); ib->ptr[ib->length_dw++] = lower_32_bits(dst_offset); ib->ptr[ib->length_dw++] = upper_32_bits(dst_offset); ib->ptr[ib->length_dw++] = src_data; diff --git a/drivers/gpu/drm/amd/amdgpu/si.c b/drivers/gpu/drm/amd/amdgpu/si.c index 85235470e872..00f63d3fbea7 100644 --- a/drivers/gpu/drm/amd/amdgpu/si.c +++ b/drivers/gpu/drm/amd/amdgpu/si.c @@ -2022,9 +2022,9 @@ static uint32_t si_get_rev_id(struct amdgpu_device *adev) >> CC_DRM_ID_STRAPS__ATI_REV_ID__SHIFT; } -static int si_common_early_init(void *handle) +static int si_common_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->smc_rreg = &si_smc_rreg; adev->smc_wreg = &si_smc_wreg; @@ -2148,17 +2148,6 @@ static int si_common_early_init(void *handle) return 0; } -static int si_common_sw_init(void *handle) -{ - return 0; -} - -static int si_common_sw_fini(void *handle) -{ - return 0; -} - - static void si_init_golden_registers(struct amdgpu_device *adev) { switch (adev->asic_type) { @@ -2633,9 +2622,9 @@ static void si_fix_pci_max_read_req_size(struct amdgpu_device *adev) pcie_set_readrq(adev->pdev, 512); } -static int si_common_hw_init(void *handle) +static int si_common_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; si_fix_pci_max_read_req_size(adev); si_init_golden_registers(adev); @@ -2645,23 +2634,14 @@ static int si_common_hw_init(void *handle) return 0; } -static int si_common_hw_fini(void *handle) +static int si_common_hw_fini(struct amdgpu_ip_block *ip_block) { return 0; } -static int si_common_suspend(void *handle) +static int si_common_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return si_common_hw_fini(adev); -} - -static int si_common_resume(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return si_common_hw_init(adev); + return si_common_hw_init(ip_block); } static bool si_common_is_idle(void *handle) @@ -2669,16 +2649,6 @@ static bool si_common_is_idle(void *handle) return true; } -static int si_common_wait_for_idle(void *handle) -{ - return 0; -} - -static int si_common_soft_reset(void *handle) -{ - return 0; -} - static int si_common_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -2694,20 +2664,12 @@ static int si_common_set_powergating_state(void *handle, static const struct amd_ip_funcs si_common_ip_funcs = { .name = "si_common", .early_init = si_common_early_init, - .late_init = NULL, - .sw_init = si_common_sw_init, - .sw_fini = si_common_sw_fini, .hw_init = si_common_hw_init, .hw_fini = si_common_hw_fini, - .suspend = si_common_suspend, .resume = si_common_resume, .is_idle = si_common_is_idle, - .wait_for_idle = si_common_wait_for_idle, - .soft_reset = si_common_soft_reset, .set_clockgating_state = si_common_set_clockgating_state, .set_powergating_state = si_common_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ip_block_version si_common_ip_block = diff --git a/drivers/gpu/drm/amd/amdgpu/si_dma.c b/drivers/gpu/drm/amd/amdgpu/si_dma.c index 11db5b755832..47647a6083e8 100644 --- a/drivers/gpu/drm/amd/amdgpu/si_dma.c +++ b/drivers/gpu/drm/amd/amdgpu/si_dma.c @@ -457,9 +457,9 @@ static void si_dma_ring_emit_wreg(struct amdgpu_ring *ring, amdgpu_ring_write(ring, val); } -static int si_dma_early_init(void *handle) +static int si_dma_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->sdma.num_instances = 2; @@ -471,11 +471,11 @@ static int si_dma_early_init(void *handle) return 0; } -static int si_dma_sw_init(void *handle) +static int si_dma_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* DMA0 trap event */ r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, 224, @@ -506,9 +506,9 @@ static int si_dma_sw_init(void *handle) return r; } -static int si_dma_sw_fini(void *handle) +static int si_dma_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; for (i = 0; i < adev->sdma.num_instances; i++) @@ -517,39 +517,34 @@ static int si_dma_sw_fini(void *handle) return 0; } -static int si_dma_hw_init(void *handle) +static int si_dma_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return si_dma_start(adev); } -static int si_dma_hw_fini(void *handle) +static int si_dma_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - si_dma_stop(adev); + si_dma_stop(ip_block->adev); return 0; } -static int si_dma_suspend(void *handle) +static int si_dma_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return si_dma_hw_fini(adev); + return si_dma_hw_fini(ip_block); } -static int si_dma_resume(void *handle) +static int si_dma_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return si_dma_hw_init(adev); + return si_dma_hw_init(ip_block); } static bool si_dma_is_idle(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + u32 tmp = RREG32(SRBM_STATUS2); if (tmp & (DMA_BUSY_MASK | DMA1_BUSY_MASK)) @@ -558,20 +553,20 @@ static bool si_dma_is_idle(void *handle) return true; } -static int si_dma_wait_for_idle(void *handle) +static int si_dma_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { - if (si_dma_is_idle(handle)) + if (si_dma_is_idle(adev)) return 0; udelay(1); } return -ETIMEDOUT; } -static int si_dma_soft_reset(void *handle) +static int si_dma_soft_reset(struct amdgpu_ip_block *ip_block) { DRM_INFO("si_dma_soft_reset --- not implemented !!!!!!!\n"); return 0; @@ -696,7 +691,6 @@ static int si_dma_set_powergating_state(void *handle, static const struct amd_ip_funcs si_dma_ip_funcs = { .name = "si_dma", .early_init = si_dma_early_init, - .late_init = NULL, .sw_init = si_dma_sw_init, .sw_fini = si_dma_sw_fini, .hw_init = si_dma_hw_init, @@ -708,8 +702,6 @@ static const struct amd_ip_funcs si_dma_ip_funcs = { .soft_reset = si_dma_soft_reset, .set_clockgating_state = si_dma_set_clockgating_state, .set_powergating_state = si_dma_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs si_dma_ring_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/si_ih.c b/drivers/gpu/drm/amd/amdgpu/si_ih.c index 5237395e4fab..2ec1ebe4db11 100644 --- a/drivers/gpu/drm/amd/amdgpu/si_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/si_ih.c @@ -156,19 +156,19 @@ static void si_ih_set_rptr(struct amdgpu_device *adev, WREG32(IH_RB_RPTR, ih->rptr); } -static int si_ih_early_init(void *handle) +static int si_ih_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; si_ih_set_interrupt_funcs(adev); return 0; } -static int si_ih_sw_init(void *handle) +static int si_ih_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 64 * 1024, false); if (r) @@ -177,43 +177,37 @@ static int si_ih_sw_init(void *handle) return amdgpu_irq_init(adev); } -static int si_ih_sw_fini(void *handle) +static int si_ih_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_fini_sw(adev); return 0; } -static int si_ih_hw_init(void *handle) +static int si_ih_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return si_ih_irq_init(adev); } -static int si_ih_hw_fini(void *handle) +static int si_ih_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - si_ih_irq_disable(adev); + si_ih_irq_disable(ip_block->adev); return 0; } -static int si_ih_suspend(void *handle) +static int si_ih_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return si_ih_hw_fini(adev); + return si_ih_hw_fini(ip_block); } -static int si_ih_resume(void *handle) +static int si_ih_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return si_ih_hw_init(adev); + return si_ih_hw_init(ip_block); } static bool si_ih_is_idle(void *handle) @@ -227,22 +221,22 @@ static bool si_ih_is_idle(void *handle) return true; } -static int si_ih_wait_for_idle(void *handle) +static int si_ih_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { - if (si_ih_is_idle(handle)) + if (si_ih_is_idle(adev)) return 0; udelay(1); } return -ETIMEDOUT; } -static int si_ih_soft_reset(void *handle) +static int si_ih_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset = 0; u32 tmp = RREG32(SRBM_STATUS); @@ -284,7 +278,6 @@ static int si_ih_set_powergating_state(void *handle, static const struct amd_ip_funcs si_ih_ip_funcs = { .name = "si_ih", .early_init = si_ih_early_init, - .late_init = NULL, .sw_init = si_ih_sw_init, .sw_fini = si_ih_sw_fini, .hw_init = si_ih_hw_init, @@ -296,8 +289,6 @@ static const struct amd_ip_funcs si_ih_ip_funcs = { .soft_reset = si_ih_soft_reset, .set_clockgating_state = si_ih_set_clockgating_state, .set_powergating_state = si_ih_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ih_funcs si_ih_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c b/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c index 481217c32d85..9b01e074af47 100644 --- a/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c +++ b/drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c @@ -81,15 +81,9 @@ static int sienna_cichlid_mode2_suspend_ip(struct amdgpu_device *adev) AMD_IP_BLOCK_TYPE_SDMA)) continue; - r = adev->ip_blocks[i].version->funcs->suspend(adev); - - if (r) { - dev_err(adev->dev, - "suspend of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + r = amdgpu_ip_block_suspend(&adev->ip_blocks[i]); + if (r) return r; - } - adev->ip_blocks[i].status.hw = false; } return 0; @@ -175,15 +169,9 @@ static int sienna_cichlid_mode2_restore_ip(struct amdgpu_device *adev) for (i = 0; i < adev->num_ip_blocks; i++) { if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_IH) { - r = adev->ip_blocks[i].version->funcs->resume(adev); - if (r) { - dev_err(adev->dev, - "resume of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + r = amdgpu_ip_block_resume(&adev->ip_blocks[i]); + if (r) return r; - } - - adev->ip_blocks[i].status.hw = true; } } @@ -193,15 +181,9 @@ static int sienna_cichlid_mode2_restore_ip(struct amdgpu_device *adev) adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SDMA)) continue; - r = adev->ip_blocks[i].version->funcs->resume(adev); - if (r) { - dev_err(adev->dev, - "resume of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + r = amdgpu_ip_block_resume(&adev->ip_blocks[i]); + if (r) return r; - } - - adev->ip_blocks[i].status.hw = true; } for (i = 0; i < adev->num_ip_blocks; i++) { @@ -213,7 +195,7 @@ static int sienna_cichlid_mode2_restore_ip(struct amdgpu_device *adev) if (adev->ip_blocks[i].version->funcs->late_init) { r = adev->ip_blocks[i].version->funcs->late_init( - (void *)adev); + &adev->ip_blocks[i]); if (r) { dev_err(adev->dev, "late_init of IP block <%s> failed %d after reset\n", diff --git a/drivers/gpu/drm/amd/amdgpu/smu_v13_0_10.c b/drivers/gpu/drm/amd/amdgpu/smu_v13_0_10.c index 0af648931df5..e70ebad3f9fa 100644 --- a/drivers/gpu/drm/amd/amdgpu/smu_v13_0_10.c +++ b/drivers/gpu/drm/amd/amdgpu/smu_v13_0_10.c @@ -80,15 +80,9 @@ static int smu_v13_0_10_mode2_suspend_ip(struct amdgpu_device *adev) AMD_IP_BLOCK_TYPE_MES)) continue; - r = adev->ip_blocks[i].version->funcs->suspend(adev); - - if (r) { - dev_err(adev->dev, - "suspend of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + r = amdgpu_ip_block_suspend(&adev->ip_blocks[i]); + if (r) return r; - } - adev->ip_blocks[i].status.hw = false; } return 0; @@ -186,15 +180,9 @@ static int smu_v13_0_10_mode2_restore_ip(struct amdgpu_device *adev) adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SDMA)) continue; - r = adev->ip_blocks[i].version->funcs->resume(adev); - if (r) { - dev_err(adev->dev, - "resume of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); + r = amdgpu_ip_block_resume(&adev->ip_blocks[i]); + if (r) return r; - } - - adev->ip_blocks[i].status.hw = true; } for (i = 0; i < adev->num_ip_blocks; i++) { @@ -208,7 +196,7 @@ static int smu_v13_0_10_mode2_restore_ip(struct amdgpu_device *adev) if (adev->ip_blocks[i].version->funcs->late_init) { r = adev->ip_blocks[i].version->funcs->late_init( - (void *)adev); + &adev->ip_blocks[i]); if (r) { dev_err(adev->dev, "late_init of IP block <%s> failed %d after reset\n", diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 8d16dacdc172..93e44e7ee3fa 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -587,11 +587,13 @@ static bool soc15_need_reset_on_resume(struct amdgpu_device *adev) * 2) S3 suspend abort and TOS already launched. */ if (adev->flags & AMD_IS_APU && adev->in_s3 && - !adev->suspend_complete && - sol_reg) + sol_reg) { + adev->suspend_complete = false; return true; - - return false; + } else { + adev->suspend_complete = true; + return false; + } } static int soc15_asic_reset(struct amdgpu_device *adev) @@ -829,6 +831,10 @@ static bool soc15_need_reset_on_init(struct amdgpu_device *adev) if (adev->asic_type == CHIP_RENOIR) return true; + if (amdgpu_gmc_need_reset_on_init(adev)) + return true; + if (amdgpu_psp_tos_reload_needed(adev)) + return true; /* Just return false for soc15 GPUs. Reset does not seem to * be necessary. */ @@ -929,9 +935,9 @@ static const struct amdgpu_asic_funcs aqua_vanjaram_asic_funcs = .get_reg_state = &aqua_vanjaram_get_reg_state, }; -static int soc15_common_early_init(void *handle) +static int soc15_common_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->nbio.funcs->set_reg_remap(adev); adev->smc_rreg = NULL; @@ -1198,9 +1204,9 @@ static int soc15_common_early_init(void *handle) return 0; } -static int soc15_common_late_init(void *handle) +static int soc15_common_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) xgpu_ai_mailbox_get_irq(adev); @@ -1213,9 +1219,9 @@ static int soc15_common_late_init(void *handle) return 0; } -static int soc15_common_sw_init(void *handle) +static int soc15_common_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) xgpu_ai_mailbox_add_irq_id(adev); @@ -1227,9 +1233,9 @@ static int soc15_common_sw_init(void *handle) return 0; } -static int soc15_common_sw_fini(void *handle) +static int soc15_common_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->df.funcs && adev->df.funcs->sw_fini) @@ -1251,9 +1257,9 @@ static void soc15_sdma_doorbell_range_init(struct amdgpu_device *adev) } } -static int soc15_common_hw_init(void *handle) +static int soc15_common_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* enable aspm */ soc15_program_aspm(adev); @@ -1280,9 +1286,9 @@ static int soc15_common_hw_init(void *handle) return 0; } -static int soc15_common_hw_fini(void *handle) +static int soc15_common_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* Disable the doorbell aperture and selfring doorbell aperture * separately in hw_fini because soc15_enable_doorbell_aperture @@ -1295,7 +1301,12 @@ static int soc15_common_hw_fini(void *handle) if (amdgpu_sriov_vf(adev)) xgpu_ai_mailbox_put_irq(adev); + /* + * For minimal init, late_init is not called, hence RAS irqs are not + * enabled. + */ if ((!amdgpu_sriov_vf(adev)) && + (adev->init_lvl->level != AMDGPU_INIT_LEVEL_MINIMAL_XGMI) && adev->nbio.ras_if && amdgpu_ras_is_supported(adev, adev->nbio.ras_if->block)) { if (adev->nbio.ras && @@ -1309,22 +1320,20 @@ static int soc15_common_hw_fini(void *handle) return 0; } -static int soc15_common_suspend(void *handle) +static int soc15_common_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return soc15_common_hw_fini(adev); + return soc15_common_hw_fini(ip_block); } -static int soc15_common_resume(void *handle) +static int soc15_common_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (soc15_need_reset_on_resume(adev)) { dev_info(adev->dev, "S3 suspend abort case, let's reset ASIC.\n"); soc15_asic_reset(adev); } - return soc15_common_hw_init(adev); + return soc15_common_hw_init(ip_block); } static bool soc15_common_is_idle(void *handle) @@ -1332,16 +1341,6 @@ static bool soc15_common_is_idle(void *handle) return true; } -static int soc15_common_wait_for_idle(void *handle) -{ - return 0; -} - -static int soc15_common_soft_reset(void *handle) -{ - return 0; -} - static void soc15_update_drm_clock_gating(struct amdgpu_device *adev, bool enable) { uint32_t def, data; @@ -1492,11 +1491,7 @@ static const struct amd_ip_funcs soc15_common_ip_funcs = { .suspend = soc15_common_suspend, .resume = soc15_common_resume, .is_idle = soc15_common_is_idle, - .wait_for_idle = soc15_common_wait_for_idle, - .soft_reset = soc15_common_soft_reset, .set_clockgating_state = soc15_common_set_clockgating_state, .set_powergating_state = soc15_common_set_powergating_state, .get_clockgating_state= soc15_common_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; diff --git a/drivers/gpu/drm/amd/amdgpu/soc21.c b/drivers/gpu/drm/amd/amdgpu/soc21.c index d30ad7d56def..1c07ebdc0d1f 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc21.c +++ b/drivers/gpu/drm/amd/amdgpu/soc21.c @@ -556,9 +556,9 @@ static const struct amdgpu_asic_funcs soc21_asic_funcs = { .update_umd_stable_pstate = &soc21_update_umd_stable_pstate, }; -static int soc21_common_early_init(void *handle) +static int soc21_common_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->nbio.funcs->set_reg_remap(adev); adev->smc_rreg = NULL; @@ -794,9 +794,9 @@ static int soc21_common_early_init(void *handle) return 0; } -static int soc21_common_late_init(void *handle) +static int soc21_common_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) { xgpu_nv_mailbox_get_irq(adev); @@ -832,9 +832,9 @@ static int soc21_common_late_init(void *handle) return 0; } -static int soc21_common_sw_init(void *handle) +static int soc21_common_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) xgpu_nv_mailbox_add_irq_id(adev); @@ -842,14 +842,9 @@ static int soc21_common_sw_init(void *handle) return 0; } -static int soc21_common_sw_fini(void *handle) +static int soc21_common_hw_init(struct amdgpu_ip_block *ip_block) { - return 0; -} - -static int soc21_common_hw_init(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* enable aspm */ soc21_program_aspm(adev); @@ -867,9 +862,9 @@ static int soc21_common_hw_init(void *handle) return 0; } -static int soc21_common_hw_fini(void *handle) +static int soc21_common_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* Disable the doorbell aperture and selfring doorbell aperture * separately in hw_fini because soc21_enable_doorbell_aperture @@ -890,11 +885,9 @@ static int soc21_common_hw_fini(void *handle) return 0; } -static int soc21_common_suspend(void *handle) +static int soc21_common_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return soc21_common_hw_fini(adev); + return soc21_common_hw_fini(ip_block); } static bool soc21_need_reset_on_resume(struct amdgpu_device *adev) @@ -917,16 +910,16 @@ static bool soc21_need_reset_on_resume(struct amdgpu_device *adev) return false; } -static int soc21_common_resume(void *handle) +static int soc21_common_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (soc21_need_reset_on_resume(adev)) { dev_info(adev->dev, "S3 suspend aborted, resetting..."); soc21_asic_reset(adev); } - return soc21_common_hw_init(adev); + return soc21_common_hw_init(ip_block); } static bool soc21_common_is_idle(void *handle) @@ -934,16 +927,6 @@ static bool soc21_common_is_idle(void *handle) return true; } -static int soc21_common_wait_for_idle(void *handle) -{ - return 0; -} - -static int soc21_common_soft_reset(void *handle) -{ - return 0; -} - static int soc21_common_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -1002,17 +985,12 @@ static const struct amd_ip_funcs soc21_common_ip_funcs = { .early_init = soc21_common_early_init, .late_init = soc21_common_late_init, .sw_init = soc21_common_sw_init, - .sw_fini = soc21_common_sw_fini, .hw_init = soc21_common_hw_init, .hw_fini = soc21_common_hw_fini, .suspend = soc21_common_suspend, .resume = soc21_common_resume, .is_idle = soc21_common_is_idle, - .wait_for_idle = soc21_common_wait_for_idle, - .soft_reset = soc21_common_soft_reset, .set_clockgating_state = soc21_common_set_clockgating_state, .set_powergating_state = soc21_common_set_powergating_state, .get_clockgating_state = soc21_common_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; diff --git a/drivers/gpu/drm/amd/amdgpu/soc24.c b/drivers/gpu/drm/amd/amdgpu/soc24.c index fd4c3d4f8387..3af10ef4b793 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc24.c +++ b/drivers/gpu/drm/amd/amdgpu/soc24.c @@ -363,9 +363,9 @@ static const struct amdgpu_asic_funcs soc24_asic_funcs = { .update_umd_stable_pstate = &soc24_update_umd_stable_pstate, }; -static int soc24_common_early_init(void *handle) +static int soc24_common_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->nbio.funcs->set_reg_remap(adev); adev->smc_rreg = NULL; @@ -440,9 +440,9 @@ static int soc24_common_early_init(void *handle) return 0; } -static int soc24_common_late_init(void *handle) +static int soc24_common_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) xgpu_nv_mailbox_get_irq(adev); @@ -455,9 +455,9 @@ static int soc24_common_late_init(void *handle) return 0; } -static int soc24_common_sw_init(void *handle) +static int soc24_common_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) xgpu_nv_mailbox_add_irq_id(adev); @@ -465,14 +465,9 @@ static int soc24_common_sw_init(void *handle) return 0; } -static int soc24_common_sw_fini(void *handle) +static int soc24_common_hw_init(struct amdgpu_ip_block *ip_block) { - return 0; -} - -static int soc24_common_hw_init(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* enable aspm */ soc24_program_aspm(adev); @@ -494,9 +489,9 @@ static int soc24_common_hw_init(void *handle) return 0; } -static int soc24_common_hw_fini(void *handle) +static int soc24_common_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* Disable the doorbell aperture and selfring doorbell aperture * separately in hw_fini because soc21_enable_doorbell_aperture @@ -512,18 +507,14 @@ static int soc24_common_hw_fini(void *handle) return 0; } -static int soc24_common_suspend(void *handle) +static int soc24_common_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return soc24_common_hw_fini(adev); + return soc24_common_hw_fini(ip_block); } -static int soc24_common_resume(void *handle) +static int soc24_common_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return soc24_common_hw_init(adev); + return soc24_common_hw_init(ip_block); } static bool soc24_common_is_idle(void *handle) @@ -531,16 +522,6 @@ static bool soc24_common_is_idle(void *handle) return true; } -static int soc24_common_wait_for_idle(void *handle) -{ - return 0; -} - -static int soc24_common_soft_reset(void *handle) -{ - return 0; -} - static int soc24_common_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -595,14 +576,11 @@ static const struct amd_ip_funcs soc24_common_ip_funcs = { .early_init = soc24_common_early_init, .late_init = soc24_common_late_init, .sw_init = soc24_common_sw_init, - .sw_fini = soc24_common_sw_fini, .hw_init = soc24_common_hw_init, .hw_fini = soc24_common_hw_fini, .suspend = soc24_common_suspend, .resume = soc24_common_resume, .is_idle = soc24_common_is_idle, - .wait_for_idle = soc24_common_wait_for_idle, - .soft_reset = soc24_common_soft_reset, .set_clockgating_state = soc24_common_set_clockgating_state, .set_powergating_state = soc24_common_set_powergating_state, .get_clockgating_state = soc24_common_get_clockgating_state, diff --git a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c index 24d49d813607..5a04a6770138 100644 --- a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c @@ -283,9 +283,9 @@ static void tonga_ih_set_rptr(struct amdgpu_device *adev, } } -static int tonga_ih_early_init(void *handle) +static int tonga_ih_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; ret = amdgpu_irq_add_domain(adev); @@ -297,10 +297,10 @@ static int tonga_ih_early_init(void *handle) return 0; } -static int tonga_ih_sw_init(void *handle) +static int tonga_ih_sw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_ih_ring_init(adev, &adev->irq.ih, 64 * 1024, true); if (r) @@ -314,9 +314,9 @@ static int tonga_ih_sw_init(void *handle) return r; } -static int tonga_ih_sw_fini(void *handle) +static int tonga_ih_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_fini_sw(adev); amdgpu_irq_remove_domain(adev); @@ -324,10 +324,10 @@ static int tonga_ih_sw_fini(void *handle) return 0; } -static int tonga_ih_hw_init(void *handle) +static int tonga_ih_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = tonga_ih_irq_init(adev); if (r) @@ -336,27 +336,21 @@ static int tonga_ih_hw_init(void *handle) return 0; } -static int tonga_ih_hw_fini(void *handle) +static int tonga_ih_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - tonga_ih_irq_disable(adev); + tonga_ih_irq_disable(ip_block->adev); return 0; } -static int tonga_ih_suspend(void *handle) +static int tonga_ih_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return tonga_ih_hw_fini(adev); + return tonga_ih_hw_fini(ip_block); } -static int tonga_ih_resume(void *handle) +static int tonga_ih_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return tonga_ih_hw_init(adev); + return tonga_ih_hw_init(ip_block); } static bool tonga_ih_is_idle(void *handle) @@ -370,11 +364,11 @@ static bool tonga_ih_is_idle(void *handle) return true; } -static int tonga_ih_wait_for_idle(void *handle) +static int tonga_ih_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; u32 tmp; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { /* read MC_STATUS */ @@ -386,9 +380,9 @@ static int tonga_ih_wait_for_idle(void *handle) return -ETIMEDOUT; } -static bool tonga_ih_check_soft_reset(void *handle) +static bool tonga_ih_check_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset = 0; u32 tmp = RREG32(mmSRBM_STATUS); @@ -405,29 +399,27 @@ static bool tonga_ih_check_soft_reset(void *handle) } } -static int tonga_ih_pre_soft_reset(void *handle) +static int tonga_ih_pre_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (!ip_block->adev->irq.srbm_soft_reset) + return 0; + + return tonga_ih_hw_fini(ip_block); +} + +static int tonga_ih_post_soft_reset(struct amdgpu_ip_block *ip_block) +{ + struct amdgpu_device *adev = ip_block->adev; if (!adev->irq.srbm_soft_reset) return 0; - return tonga_ih_hw_fini(adev); + return tonga_ih_hw_init(ip_block); } -static int tonga_ih_post_soft_reset(void *handle) +static int tonga_ih_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - if (!adev->irq.srbm_soft_reset) - return 0; - - return tonga_ih_hw_init(adev); -} - -static int tonga_ih_soft_reset(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset; if (!adev->irq.srbm_soft_reset) @@ -471,7 +463,6 @@ static int tonga_ih_set_powergating_state(void *handle, static const struct amd_ip_funcs tonga_ih_ip_funcs = { .name = "tonga_ih", .early_init = tonga_ih_early_init, - .late_init = NULL, .sw_init = tonga_ih_sw_init, .sw_fini = tonga_ih_sw_fini, .hw_init = tonga_ih_hw_init, @@ -486,8 +477,6 @@ static const struct amd_ip_funcs tonga_ih_ip_funcs = { .post_soft_reset = tonga_ih_post_soft_reset, .set_clockgating_state = tonga_ih_set_clockgating_state, .set_powergating_state = tonga_ih_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ih_funcs tonga_ih_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c b/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c index 805d6662c88b..bdbca25d80c4 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v3_1.c @@ -531,9 +531,9 @@ static void uvd_v3_1_set_irq_funcs(struct amdgpu_device *adev) } -static int uvd_v3_1_early_init(void *handle) +static int uvd_v3_1_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->uvd.num_uvd_inst = 1; uvd_v3_1_set_ring_funcs(adev); @@ -542,10 +542,10 @@ static int uvd_v3_1_early_init(void *handle) return 0; } -static int uvd_v3_1_sw_init(void *handle) +static int uvd_v3_1_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; void *ptr; uint32_t ucode_len; @@ -580,10 +580,10 @@ static int uvd_v3_1_sw_init(void *handle) return r; } -static int uvd_v3_1_sw_fini(void *handle) +static int uvd_v3_1_sw_fini(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_uvd_suspend(adev); if (r) @@ -621,13 +621,13 @@ static void uvd_v3_1_enable_mgcg(struct amdgpu_device *adev, /** * uvd_v3_1_hw_init - start and test UVD block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int uvd_v3_1_hw_init(void *handle) +static int uvd_v3_1_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring = &adev->uvd.inst->ring; uint32_t tmp; int r; @@ -688,13 +688,13 @@ done: /** * uvd_v3_1_hw_fini - stop the hardware block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the UVD block, mark ring as not ready any more */ -static int uvd_v3_1_hw_fini(void *handle) +static int uvd_v3_1_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->uvd.idle_work); @@ -704,17 +704,17 @@ static int uvd_v3_1_hw_fini(void *handle) return 0; } -static int uvd_v3_1_prepare_suspend(void *handle) +static int uvd_v3_1_prepare_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return amdgpu_uvd_prepare_suspend(adev); } -static int uvd_v3_1_suspend(void *handle) +static int uvd_v3_1_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* * Proper cleanups before halting the HW engine: @@ -740,23 +740,22 @@ static int uvd_v3_1_suspend(void *handle) AMD_CG_STATE_GATE); } - r = uvd_v3_1_hw_fini(adev); + r = uvd_v3_1_hw_fini(ip_block); if (r) return r; return amdgpu_uvd_suspend(adev); } -static int uvd_v3_1_resume(void *handle) +static int uvd_v3_1_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_uvd_resume(adev); + r = amdgpu_uvd_resume(ip_block->adev); if (r) return r; - return uvd_v3_1_hw_init(adev); + return uvd_v3_1_hw_init(ip_block); } static bool uvd_v3_1_is_idle(void *handle) @@ -766,10 +765,10 @@ static bool uvd_v3_1_is_idle(void *handle) return !(RREG32(mmSRBM_STATUS) & SRBM_STATUS__UVD_BUSY_MASK); } -static int uvd_v3_1_wait_for_idle(void *handle) +static int uvd_v3_1_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { if (!(RREG32(mmSRBM_STATUS) & SRBM_STATUS__UVD_BUSY_MASK)) @@ -778,9 +777,9 @@ static int uvd_v3_1_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int uvd_v3_1_soft_reset(void *handle) +static int uvd_v3_1_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uvd_v3_1_stop(adev); @@ -806,7 +805,6 @@ static int uvd_v3_1_set_powergating_state(void *handle, static const struct amd_ip_funcs uvd_v3_1_ip_funcs = { .name = "uvd_v3_1", .early_init = uvd_v3_1_early_init, - .late_init = NULL, .sw_init = uvd_v3_1_sw_init, .sw_fini = uvd_v3_1_sw_fini, .hw_init = uvd_v3_1_hw_init, @@ -819,8 +817,6 @@ static const struct amd_ip_funcs uvd_v3_1_ip_funcs = { .soft_reset = uvd_v3_1_soft_reset, .set_clockgating_state = uvd_v3_1_set_clockgating_state, .set_powergating_state = uvd_v3_1_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; const struct amdgpu_ip_block_version uvd_v3_1_ip_block = { diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c index 3f19c606f4de..a836dc9cfcad 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c @@ -90,9 +90,9 @@ static void uvd_v4_2_ring_set_wptr(struct amdgpu_ring *ring) WREG32(mmUVD_RBC_RB_WPTR, lower_32_bits(ring->wptr)); } -static int uvd_v4_2_early_init(void *handle) +static int uvd_v4_2_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->uvd.num_uvd_inst = 1; uvd_v4_2_set_ring_funcs(adev); @@ -101,10 +101,10 @@ static int uvd_v4_2_early_init(void *handle) return 0; } -static int uvd_v4_2_sw_init(void *handle) +static int uvd_v4_2_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; /* UVD TRAP */ @@ -130,10 +130,10 @@ static int uvd_v4_2_sw_init(void *handle) return r; } -static int uvd_v4_2_sw_fini(void *handle) +static int uvd_v4_2_sw_fini(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_uvd_suspend(adev); if (r) @@ -147,13 +147,13 @@ static void uvd_v4_2_enable_mgcg(struct amdgpu_device *adev, /** * uvd_v4_2_hw_init - start and test UVD block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int uvd_v4_2_hw_init(void *handle) +static int uvd_v4_2_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring = &adev->uvd.inst->ring; uint32_t tmp; int r; @@ -202,13 +202,13 @@ done: /** * uvd_v4_2_hw_fini - stop the hardware block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the UVD block, mark ring as not ready any more */ -static int uvd_v4_2_hw_fini(void *handle) +static int uvd_v4_2_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->uvd.idle_work); @@ -218,17 +218,17 @@ static int uvd_v4_2_hw_fini(void *handle) return 0; } -static int uvd_v4_2_prepare_suspend(void *handle) +static int uvd_v4_2_prepare_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return amdgpu_uvd_prepare_suspend(adev); } -static int uvd_v4_2_suspend(void *handle) +static int uvd_v4_2_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* * Proper cleanups before halting the HW engine: @@ -254,23 +254,22 @@ static int uvd_v4_2_suspend(void *handle) AMD_CG_STATE_GATE); } - r = uvd_v4_2_hw_fini(adev); + r = uvd_v4_2_hw_fini(ip_block); if (r) return r; return amdgpu_uvd_suspend(adev); } -static int uvd_v4_2_resume(void *handle) +static int uvd_v4_2_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_uvd_resume(adev); + r = amdgpu_uvd_resume(ip_block->adev); if (r) return r; - return uvd_v4_2_hw_init(adev); + return uvd_v4_2_hw_init(ip_block); } /** @@ -666,10 +665,10 @@ static bool uvd_v4_2_is_idle(void *handle) return !(RREG32(mmSRBM_STATUS) & SRBM_STATUS__UVD_BUSY_MASK); } -static int uvd_v4_2_wait_for_idle(void *handle) +static int uvd_v4_2_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { if (!(RREG32(mmSRBM_STATUS) & SRBM_STATUS__UVD_BUSY_MASK)) @@ -678,9 +677,9 @@ static int uvd_v4_2_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int uvd_v4_2_soft_reset(void *handle) +static int uvd_v4_2_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uvd_v4_2_stop(adev); @@ -756,7 +755,6 @@ static int uvd_v4_2_set_powergating_state(void *handle, static const struct amd_ip_funcs uvd_v4_2_ip_funcs = { .name = "uvd_v4_2", .early_init = uvd_v4_2_early_init, - .late_init = NULL, .sw_init = uvd_v4_2_sw_init, .sw_fini = uvd_v4_2_sw_fini, .hw_init = uvd_v4_2_hw_init, @@ -769,8 +767,6 @@ static const struct amd_ip_funcs uvd_v4_2_ip_funcs = { .soft_reset = uvd_v4_2_soft_reset, .set_clockgating_state = uvd_v4_2_set_clockgating_state, .set_powergating_state = uvd_v4_2_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs uvd_v4_2_ring_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c index efd903c21d48..ab55fae3569e 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c @@ -88,9 +88,9 @@ static void uvd_v5_0_ring_set_wptr(struct amdgpu_ring *ring) WREG32(mmUVD_RBC_RB_WPTR, lower_32_bits(ring->wptr)); } -static int uvd_v5_0_early_init(void *handle) +static int uvd_v5_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->uvd.num_uvd_inst = 1; uvd_v5_0_set_ring_funcs(adev); @@ -99,10 +99,10 @@ static int uvd_v5_0_early_init(void *handle) return 0; } -static int uvd_v5_0_sw_init(void *handle) +static int uvd_v5_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; /* UVD TRAP */ @@ -128,10 +128,10 @@ static int uvd_v5_0_sw_init(void *handle) return r; } -static int uvd_v5_0_sw_fini(void *handle) +static int uvd_v5_0_sw_fini(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_uvd_suspend(adev); if (r) @@ -143,13 +143,13 @@ static int uvd_v5_0_sw_fini(void *handle) /** * uvd_v5_0_hw_init - start and test UVD block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int uvd_v5_0_hw_init(void *handle) +static int uvd_v5_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring = &adev->uvd.inst->ring; uint32_t tmp; int r; @@ -200,13 +200,13 @@ done: /** * uvd_v5_0_hw_fini - stop the hardware block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the UVD block, mark ring as not ready any more */ -static int uvd_v5_0_hw_fini(void *handle) +static int uvd_v5_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->uvd.idle_work); @@ -216,17 +216,17 @@ static int uvd_v5_0_hw_fini(void *handle) return 0; } -static int uvd_v5_0_prepare_suspend(void *handle) +static int uvd_v5_0_prepare_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return amdgpu_uvd_prepare_suspend(adev); } -static int uvd_v5_0_suspend(void *handle) +static int uvd_v5_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* * Proper cleanups before halting the HW engine: @@ -252,23 +252,22 @@ static int uvd_v5_0_suspend(void *handle) AMD_CG_STATE_GATE); } - r = uvd_v5_0_hw_fini(adev); + r = uvd_v5_0_hw_fini(ip_block); if (r) return r; return amdgpu_uvd_suspend(adev); } -static int uvd_v5_0_resume(void *handle) +static int uvd_v5_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_uvd_resume(adev); + r = amdgpu_uvd_resume(ip_block->adev); if (r) return r; - return uvd_v5_0_hw_init(adev); + return uvd_v5_0_hw_init(ip_block); } /** @@ -588,10 +587,10 @@ static bool uvd_v5_0_is_idle(void *handle) return !(RREG32(mmSRBM_STATUS) & SRBM_STATUS__UVD_BUSY_MASK); } -static int uvd_v5_0_wait_for_idle(void *handle) +static int uvd_v5_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { if (!(RREG32(mmSRBM_STATUS) & SRBM_STATUS__UVD_BUSY_MASK)) @@ -600,9 +599,9 @@ static int uvd_v5_0_wait_for_idle(void *handle) return -ETIMEDOUT; } -static int uvd_v5_0_soft_reset(void *handle) +static int uvd_v5_0_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; uvd_v5_0_stop(adev); @@ -796,10 +795,15 @@ static int uvd_v5_0_set_clockgating_state(void *handle, { struct amdgpu_device *adev = (struct amdgpu_device *)handle; bool enable = (state == AMD_CG_STATE_GATE); + struct amdgpu_ip_block *ip_block; + + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_UVD); + if (!ip_block) + return -EINVAL; if (enable) { /* wait for STATUS to clear */ - if (uvd_v5_0_wait_for_idle(handle)) + if (uvd_v5_0_wait_for_idle(ip_block)) return -EBUSY; uvd_v5_0_enable_clock_gating(adev, true); @@ -863,7 +867,6 @@ out: static const struct amd_ip_funcs uvd_v5_0_ip_funcs = { .name = "uvd_v5_0", .early_init = uvd_v5_0_early_init, - .late_init = NULL, .sw_init = uvd_v5_0_sw_init, .sw_fini = uvd_v5_0_sw_fini, .hw_init = uvd_v5_0_hw_init, @@ -877,8 +880,6 @@ static const struct amd_ip_funcs uvd_v5_0_ip_funcs = { .set_clockgating_state = uvd_v5_0_set_clockgating_state, .set_powergating_state = uvd_v5_0_set_powergating_state, .get_clockgating_state = uvd_v5_0_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs uvd_v5_0_ring_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c index 495de5068455..39f8c3d3a135 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c @@ -354,9 +354,9 @@ error: return r; } -static int uvd_v6_0_early_init(void *handle) +static int uvd_v6_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->uvd.num_uvd_inst = 1; if (!(adev->flags & AMD_IS_APU) && @@ -375,11 +375,11 @@ static int uvd_v6_0_early_init(void *handle) return 0; } -static int uvd_v6_0_sw_init(void *handle) +static int uvd_v6_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int i, r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* UVD TRAP */ r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, VISLANDS30_IV_SRCID_UVD_SYSTEM_MESSAGE, &adev->uvd.inst->irq); @@ -435,10 +435,10 @@ static int uvd_v6_0_sw_init(void *handle) return r; } -static int uvd_v6_0_sw_fini(void *handle) +static int uvd_v6_0_sw_fini(struct amdgpu_ip_block *ip_block) { int i, r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_uvd_suspend(adev); if (r) @@ -455,13 +455,13 @@ static int uvd_v6_0_sw_fini(void *handle) /** * uvd_v6_0_hw_init - start and test UVD block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int uvd_v6_0_hw_init(void *handle) +static int uvd_v6_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring = &adev->uvd.inst->ring; uint32_t tmp; int i, r; @@ -524,13 +524,13 @@ done: /** * uvd_v6_0_hw_fini - stop the hardware block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the UVD block, mark ring as not ready any more */ -static int uvd_v6_0_hw_fini(void *handle) +static int uvd_v6_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->uvd.idle_work); @@ -540,17 +540,17 @@ static int uvd_v6_0_hw_fini(void *handle) return 0; } -static int uvd_v6_0_prepare_suspend(void *handle) +static int uvd_v6_0_prepare_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return amdgpu_uvd_prepare_suspend(adev); } -static int uvd_v6_0_suspend(void *handle) +static int uvd_v6_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* * Proper cleanups before halting the HW engine: @@ -576,23 +576,22 @@ static int uvd_v6_0_suspend(void *handle) AMD_CG_STATE_GATE); } - r = uvd_v6_0_hw_fini(adev); + r = uvd_v6_0_hw_fini(ip_block); if (r) return r; return amdgpu_uvd_suspend(adev); } -static int uvd_v6_0_resume(void *handle) +static int uvd_v6_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_uvd_resume(adev); + r = amdgpu_uvd_resume(ip_block->adev); if (r) return r; - return uvd_v6_0_hw_init(adev); + return uvd_v6_0_hw_init(ip_block); } /** @@ -1151,22 +1150,22 @@ static bool uvd_v6_0_is_idle(void *handle) return !(RREG32(mmSRBM_STATUS) & SRBM_STATUS__UVD_BUSY_MASK); } -static int uvd_v6_0_wait_for_idle(void *handle) +static int uvd_v6_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) { - if (uvd_v6_0_is_idle(handle)) + if (uvd_v6_0_is_idle(adev)) return 0; } return -ETIMEDOUT; } #define AMDGPU_UVD_STATUS_BUSY_MASK 0xfd -static bool uvd_v6_0_check_soft_reset(void *handle) +static bool uvd_v6_0_check_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset = 0; u32 tmp = RREG32(mmSRBM_STATUS); @@ -1184,9 +1183,9 @@ static bool uvd_v6_0_check_soft_reset(void *handle) } } -static int uvd_v6_0_pre_soft_reset(void *handle) +static int uvd_v6_0_pre_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (!adev->uvd.inst->srbm_soft_reset) return 0; @@ -1195,9 +1194,9 @@ static int uvd_v6_0_pre_soft_reset(void *handle) return 0; } -static int uvd_v6_0_soft_reset(void *handle) +static int uvd_v6_0_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset; if (!adev->uvd.inst->srbm_soft_reset) @@ -1226,9 +1225,9 @@ static int uvd_v6_0_soft_reset(void *handle) return 0; } -static int uvd_v6_0_post_soft_reset(void *handle) +static int uvd_v6_0_post_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (!adev->uvd.inst->srbm_soft_reset) return 0; @@ -1455,11 +1454,16 @@ static int uvd_v6_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_ip_block *ip_block; bool enable = (state == AMD_CG_STATE_GATE); + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_UVD); + if (!ip_block) + return -EINVAL; + if (enable) { /* wait for STATUS to clear */ - if (uvd_v6_0_wait_for_idle(handle)) + if (uvd_v6_0_wait_for_idle(ip_block)) return -EBUSY; uvd_v6_0_enable_clock_gating(adev, true); /* enable HW gates because UVD is idle */ @@ -1528,7 +1532,6 @@ out: static const struct amd_ip_funcs uvd_v6_0_ip_funcs = { .name = "uvd_v6_0", .early_init = uvd_v6_0_early_init, - .late_init = NULL, .sw_init = uvd_v6_0_sw_init, .sw_fini = uvd_v6_0_sw_fini, .hw_init = uvd_v6_0_hw_init, @@ -1545,8 +1548,6 @@ static const struct amd_ip_funcs uvd_v6_0_ip_funcs = { .set_clockgating_state = uvd_v6_0_set_clockgating_state, .set_powergating_state = uvd_v6_0_set_powergating_state, .get_clockgating_state = uvd_v6_0_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs uvd_v6_0_ring_phys_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c index 6068b784dc69..079131aeb2f7 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c @@ -361,9 +361,9 @@ error: return r; } -static int uvd_v7_0_early_init(void *handle) +static int uvd_v7_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->asic_type == CHIP_VEGA20) { u32 harvest; @@ -395,12 +395,12 @@ static int uvd_v7_0_early_init(void *handle) return 0; } -static int uvd_v7_0_sw_init(void *handle) +static int uvd_v7_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int i, j, r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (j = 0; j < adev->uvd.num_uvd_inst; j++) { if (adev->uvd.harvest_config & (1 << j)) @@ -487,10 +487,10 @@ static int uvd_v7_0_sw_init(void *handle) return r; } -static int uvd_v7_0_sw_fini(void *handle) +static int uvd_v7_0_sw_fini(struct amdgpu_ip_block *ip_block) { int i, j, r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_virt_free_mm_table(adev); @@ -510,13 +510,13 @@ static int uvd_v7_0_sw_fini(void *handle) /** * uvd_v7_0_hw_init - start and test UVD block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int uvd_v7_0_hw_init(void *handle) +static int uvd_v7_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; uint32_t tmp; int i, j, r; @@ -588,13 +588,13 @@ done: /** * uvd_v7_0_hw_fini - stop the hardware block * - * @handle: handle used to pass amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the UVD block, mark ring as not ready any more */ -static int uvd_v7_0_hw_fini(void *handle) +static int uvd_v7_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->uvd.idle_work); @@ -608,17 +608,17 @@ static int uvd_v7_0_hw_fini(void *handle) return 0; } -static int uvd_v7_0_prepare_suspend(void *handle) +static int uvd_v7_0_prepare_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; return amdgpu_uvd_prepare_suspend(adev); } -static int uvd_v7_0_suspend(void *handle) +static int uvd_v7_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* * Proper cleanups before halting the HW engine: @@ -644,23 +644,22 @@ static int uvd_v7_0_suspend(void *handle) AMD_CG_STATE_GATE); } - r = uvd_v7_0_hw_fini(adev); + r = uvd_v7_0_hw_fini(ip_block); if (r) return r; return amdgpu_uvd_suspend(adev); } -static int uvd_v7_0_resume(void *handle) +static int uvd_v7_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_uvd_resume(adev); + r = amdgpu_uvd_resume(ip_block->adev); if (r) return r; - return uvd_v7_0_hw_init(adev); + return uvd_v7_0_hw_init(ip_block); } /** @@ -1463,104 +1462,6 @@ static void uvd_v7_0_enc_ring_emit_wreg(struct amdgpu_ring *ring, amdgpu_ring_write(ring, val); } -#if 0 -static bool uvd_v7_0_is_idle(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return !(RREG32(mmSRBM_STATUS) & SRBM_STATUS__UVD_BUSY_MASK); -} - -static int uvd_v7_0_wait_for_idle(void *handle) -{ - unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - for (i = 0; i < adev->usec_timeout; i++) { - if (uvd_v7_0_is_idle(handle)) - return 0; - } - return -ETIMEDOUT; -} - -#define AMDGPU_UVD_STATUS_BUSY_MASK 0xfd -static bool uvd_v7_0_check_soft_reset(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - u32 srbm_soft_reset = 0; - u32 tmp = RREG32(mmSRBM_STATUS); - - if (REG_GET_FIELD(tmp, SRBM_STATUS, UVD_RQ_PENDING) || - REG_GET_FIELD(tmp, SRBM_STATUS, UVD_BUSY) || - (RREG32_SOC15(UVD, ring->me, mmUVD_STATUS) & - AMDGPU_UVD_STATUS_BUSY_MASK)) - srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, - SRBM_SOFT_RESET, SOFT_RESET_UVD, 1); - - if (srbm_soft_reset) { - adev->uvd.inst[ring->me].srbm_soft_reset = srbm_soft_reset; - return true; - } else { - adev->uvd.inst[ring->me].srbm_soft_reset = 0; - return false; - } -} - -static int uvd_v7_0_pre_soft_reset(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - if (!adev->uvd.inst[ring->me].srbm_soft_reset) - return 0; - - uvd_v7_0_stop(adev); - return 0; -} - -static int uvd_v7_0_soft_reset(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - u32 srbm_soft_reset; - - if (!adev->uvd.inst[ring->me].srbm_soft_reset) - return 0; - srbm_soft_reset = adev->uvd.inst[ring->me].srbm_soft_reset; - - if (srbm_soft_reset) { - u32 tmp; - - tmp = RREG32(mmSRBM_SOFT_RESET); - tmp |= srbm_soft_reset; - dev_info(adev->dev, "SRBM_SOFT_RESET=0x%08X\n", tmp); - WREG32(mmSRBM_SOFT_RESET, tmp); - tmp = RREG32(mmSRBM_SOFT_RESET); - - udelay(50); - - tmp &= ~srbm_soft_reset; - WREG32(mmSRBM_SOFT_RESET, tmp); - tmp = RREG32(mmSRBM_SOFT_RESET); - - /* Wait a little for things to settle down */ - udelay(50); - } - - return 0; -} - -static int uvd_v7_0_post_soft_reset(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - if (!adev->uvd.inst[ring->me].srbm_soft_reset) - return 0; - - mdelay(5); - - return uvd_v7_0_start(adev); -} -#endif - static int uvd_v7_0_set_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *source, unsigned type, @@ -1610,171 +1511,6 @@ static int uvd_v7_0_process_interrupt(struct amdgpu_device *adev, return 0; } -#if 0 -static void uvd_v7_0_set_sw_clock_gating(struct amdgpu_device *adev) -{ - uint32_t data, data1, data2, suvd_flags; - - data = RREG32_SOC15(UVD, ring->me, mmUVD_CGC_CTRL); - data1 = RREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_GATE); - data2 = RREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_CTRL); - - data &= ~(UVD_CGC_CTRL__CLK_OFF_DELAY_MASK | - UVD_CGC_CTRL__CLK_GATE_DLY_TIMER_MASK); - - suvd_flags = UVD_SUVD_CGC_GATE__SRE_MASK | - UVD_SUVD_CGC_GATE__SIT_MASK | - UVD_SUVD_CGC_GATE__SMP_MASK | - UVD_SUVD_CGC_GATE__SCM_MASK | - UVD_SUVD_CGC_GATE__SDB_MASK; - - data |= UVD_CGC_CTRL__DYN_CLOCK_MODE_MASK | - (1 << REG_FIELD_SHIFT(UVD_CGC_CTRL, CLK_GATE_DLY_TIMER)) | - (4 << REG_FIELD_SHIFT(UVD_CGC_CTRL, CLK_OFF_DELAY)); - - data &= ~(UVD_CGC_CTRL__UDEC_RE_MODE_MASK | - UVD_CGC_CTRL__UDEC_CM_MODE_MASK | - UVD_CGC_CTRL__UDEC_IT_MODE_MASK | - UVD_CGC_CTRL__UDEC_DB_MODE_MASK | - UVD_CGC_CTRL__UDEC_MP_MODE_MASK | - UVD_CGC_CTRL__SYS_MODE_MASK | - UVD_CGC_CTRL__UDEC_MODE_MASK | - UVD_CGC_CTRL__MPEG2_MODE_MASK | - UVD_CGC_CTRL__REGS_MODE_MASK | - UVD_CGC_CTRL__RBC_MODE_MASK | - UVD_CGC_CTRL__LMI_MC_MODE_MASK | - UVD_CGC_CTRL__LMI_UMC_MODE_MASK | - UVD_CGC_CTRL__IDCT_MODE_MASK | - UVD_CGC_CTRL__MPRD_MODE_MASK | - UVD_CGC_CTRL__MPC_MODE_MASK | - UVD_CGC_CTRL__LBSI_MODE_MASK | - UVD_CGC_CTRL__LRBBM_MODE_MASK | - UVD_CGC_CTRL__WCB_MODE_MASK | - UVD_CGC_CTRL__VCPU_MODE_MASK | - UVD_CGC_CTRL__JPEG_MODE_MASK | - UVD_CGC_CTRL__JPEG2_MODE_MASK | - UVD_CGC_CTRL__SCPU_MODE_MASK); - data2 &= ~(UVD_SUVD_CGC_CTRL__SRE_MODE_MASK | - UVD_SUVD_CGC_CTRL__SIT_MODE_MASK | - UVD_SUVD_CGC_CTRL__SMP_MODE_MASK | - UVD_SUVD_CGC_CTRL__SCM_MODE_MASK | - UVD_SUVD_CGC_CTRL__SDB_MODE_MASK); - data1 |= suvd_flags; - - WREG32_SOC15(UVD, ring->me, mmUVD_CGC_CTRL, data); - WREG32_SOC15(UVD, ring->me, mmUVD_CGC_GATE, 0); - WREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_GATE, data1); - WREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_CTRL, data2); -} - -static void uvd_v7_0_set_hw_clock_gating(struct amdgpu_device *adev) -{ - uint32_t data, data1, cgc_flags, suvd_flags; - - data = RREG32_SOC15(UVD, ring->me, mmUVD_CGC_GATE); - data1 = RREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_GATE); - - cgc_flags = UVD_CGC_GATE__SYS_MASK | - UVD_CGC_GATE__UDEC_MASK | - UVD_CGC_GATE__MPEG2_MASK | - UVD_CGC_GATE__RBC_MASK | - UVD_CGC_GATE__LMI_MC_MASK | - UVD_CGC_GATE__IDCT_MASK | - UVD_CGC_GATE__MPRD_MASK | - UVD_CGC_GATE__MPC_MASK | - UVD_CGC_GATE__LBSI_MASK | - UVD_CGC_GATE__LRBBM_MASK | - UVD_CGC_GATE__UDEC_RE_MASK | - UVD_CGC_GATE__UDEC_CM_MASK | - UVD_CGC_GATE__UDEC_IT_MASK | - UVD_CGC_GATE__UDEC_DB_MASK | - UVD_CGC_GATE__UDEC_MP_MASK | - UVD_CGC_GATE__WCB_MASK | - UVD_CGC_GATE__VCPU_MASK | - UVD_CGC_GATE__SCPU_MASK | - UVD_CGC_GATE__JPEG_MASK | - UVD_CGC_GATE__JPEG2_MASK; - - suvd_flags = UVD_SUVD_CGC_GATE__SRE_MASK | - UVD_SUVD_CGC_GATE__SIT_MASK | - UVD_SUVD_CGC_GATE__SMP_MASK | - UVD_SUVD_CGC_GATE__SCM_MASK | - UVD_SUVD_CGC_GATE__SDB_MASK; - - data |= cgc_flags; - data1 |= suvd_flags; - - WREG32_SOC15(UVD, ring->me, mmUVD_CGC_GATE, data); - WREG32_SOC15(UVD, ring->me, mmUVD_SUVD_CGC_GATE, data1); -} - -static void uvd_v7_0_set_bypass_mode(struct amdgpu_device *adev, bool enable) -{ - u32 tmp = RREG32_SMC(ixGCK_DFS_BYPASS_CNTL); - - if (enable) - tmp |= (GCK_DFS_BYPASS_CNTL__BYPASSDCLK_MASK | - GCK_DFS_BYPASS_CNTL__BYPASSVCLK_MASK); - else - tmp &= ~(GCK_DFS_BYPASS_CNTL__BYPASSDCLK_MASK | - GCK_DFS_BYPASS_CNTL__BYPASSVCLK_MASK); - - WREG32_SMC(ixGCK_DFS_BYPASS_CNTL, tmp); -} - - -static int uvd_v7_0_set_clockgating_state(void *handle, - enum amd_clockgating_state state) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - bool enable = (state == AMD_CG_STATE_GATE); - - uvd_v7_0_set_bypass_mode(adev, enable); - - if (!(adev->cg_flags & AMD_CG_SUPPORT_UVD_MGCG)) - return 0; - - if (enable) { - /* disable HW gating and enable Sw gating */ - uvd_v7_0_set_sw_clock_gating(adev); - } else { - /* wait for STATUS to clear */ - if (uvd_v7_0_wait_for_idle(handle)) - return -EBUSY; - - /* enable HW gates because UVD is idle */ - /* uvd_v7_0_set_hw_clock_gating(adev); */ - } - - return 0; -} - -static int uvd_v7_0_set_powergating_state(void *handle, - enum amd_powergating_state state) -{ - /* This doesn't actually powergate the UVD block. - * That's done in the dpm code via the SMC. This - * just re-inits the block as necessary. The actual - * gating still happens in the dpm code. We should - * revisit this when there is a cleaner line between - * the smc and the hw blocks - */ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - if (!(adev->pg_flags & AMD_PG_SUPPORT_UVD)) - return 0; - - WREG32_SOC15(UVD, ring->me, mmUVD_POWER_STATUS, UVD_POWER_STATUS__UVD_PG_EN_MASK); - - if (state == AMD_PG_STATE_GATE) { - uvd_v7_0_stop(adev); - return 0; - } else { - return uvd_v7_0_start(adev); - } -} -#endif - static int uvd_v7_0_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -1785,7 +1521,6 @@ static int uvd_v7_0_set_clockgating_state(void *handle, const struct amd_ip_funcs uvd_v7_0_ip_funcs = { .name = "uvd_v7_0", .early_init = uvd_v7_0_early_init, - .late_init = NULL, .sw_init = uvd_v7_0_sw_init, .sw_fini = uvd_v7_0_sw_fini, .hw_init = uvd_v7_0_hw_init, @@ -1793,12 +1528,6 @@ const struct amd_ip_funcs uvd_v7_0_ip_funcs = { .prepare_suspend = uvd_v7_0_prepare_suspend, .suspend = uvd_v7_0_suspend, .resume = uvd_v7_0_resume, - .is_idle = NULL /* uvd_v7_0_is_idle */, - .wait_for_idle = NULL /* uvd_v7_0_wait_for_idle */, - .check_soft_reset = NULL /* uvd_v7_0_check_soft_reset */, - .pre_soft_reset = NULL /* uvd_v7_0_pre_soft_reset */, - .soft_reset = NULL /* uvd_v7_0_soft_reset */, - .post_soft_reset = NULL /* uvd_v7_0_post_soft_reset */, .set_clockgating_state = uvd_v7_0_set_clockgating_state, .set_powergating_state = NULL /* uvd_v7_0_set_powergating_state */, }; diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c index 66fada199bda..c1ed91b39415 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c @@ -208,13 +208,13 @@ static bool vce_v2_0_is_idle(void *handle) return !(RREG32(mmSRBM_STATUS2) & SRBM_STATUS2__VCE_BUSY_MASK); } -static int vce_v2_0_wait_for_idle(void *handle) +static int vce_v2_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; unsigned i; for (i = 0; i < adev->usec_timeout; i++) { - if (vce_v2_0_is_idle(handle)) + if (vce_v2_0_is_idle(adev)) return 0; } return -ETIMEDOUT; @@ -274,15 +274,21 @@ static int vce_v2_0_start(struct amdgpu_device *adev) static int vce_v2_0_stop(struct amdgpu_device *adev) { + struct amdgpu_ip_block *ip_block; int i; int status; + if (vce_v2_0_lmi_clean(adev)) { DRM_INFO("vce is not idle \n"); return 0; } - if (vce_v2_0_wait_for_idle(adev)) { + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_VCN); + if (!ip_block) + return -EINVAL; + + if (vce_v2_0_wait_for_idle(ip_block)) { DRM_INFO("VCE is busy, Can't set clock gating"); return 0; } @@ -398,9 +404,9 @@ static void vce_v2_0_enable_mgcg(struct amdgpu_device *adev, bool enable, } } -static int vce_v2_0_early_init(void *handle) +static int vce_v2_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->vce.num_rings = 2; @@ -410,11 +416,11 @@ static int vce_v2_0_early_init(void *handle) return 0; } -static int vce_v2_0_sw_init(void *handle) +static int vce_v2_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* VCE */ r = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, 167, &adev->vce.irq); @@ -444,10 +450,10 @@ static int vce_v2_0_sw_init(void *handle) return r; } -static int vce_v2_0_sw_fini(void *handle) +static int vce_v2_0_sw_fini(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_vce_suspend(adev); if (r) @@ -456,10 +462,10 @@ static int vce_v2_0_sw_fini(void *handle) return amdgpu_vce_sw_fini(adev); } -static int vce_v2_0_hw_init(void *handle) +static int vce_v2_0_hw_init(struct amdgpu_ip_block *ip_block) { int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_asic_set_vce_clocks(adev, 10000, 10000); vce_v2_0_enable_mgcg(adev, true, false); @@ -475,19 +481,17 @@ static int vce_v2_0_hw_init(void *handle) return 0; } -static int vce_v2_0_hw_fini(void *handle) +static int vce_v2_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - cancel_delayed_work_sync(&adev->vce.idle_work); + cancel_delayed_work_sync(&ip_block->adev->vce.idle_work); return 0; } -static int vce_v2_0_suspend(void *handle) +static int vce_v2_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* @@ -513,28 +517,27 @@ static int vce_v2_0_suspend(void *handle) AMD_CG_STATE_GATE); } - r = vce_v2_0_hw_fini(adev); + r = vce_v2_0_hw_fini(ip_block); if (r) return r; return amdgpu_vce_suspend(adev); } -static int vce_v2_0_resume(void *handle) +static int vce_v2_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_vce_resume(adev); + r = amdgpu_vce_resume(ip_block->adev); if (r) return r; - return vce_v2_0_hw_init(adev); + return vce_v2_0_hw_init(ip_block); } -static int vce_v2_0_soft_reset(void *handle) +static int vce_v2_0_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; WREG32_FIELD(SRBM_SOFT_RESET, SOFT_RESET_VCE, 1); mdelay(5); @@ -614,7 +617,6 @@ static int vce_v2_0_set_powergating_state(void *handle, static const struct amd_ip_funcs vce_v2_0_ip_funcs = { .name = "vce_v2_0", .early_init = vce_v2_0_early_init, - .late_init = NULL, .sw_init = vce_v2_0_sw_init, .sw_fini = vce_v2_0_sw_fini, .hw_init = vce_v2_0_hw_init, @@ -626,8 +628,6 @@ static const struct amd_ip_funcs vce_v2_0_ip_funcs = { .soft_reset = vce_v2_0_soft_reset, .set_clockgating_state = vce_v2_0_set_clockgating_state, .set_powergating_state = vce_v2_0_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs vce_v2_0_ring_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c index 4bfba2931b08..6bb318a06f19 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c @@ -64,7 +64,7 @@ static void vce_v3_0_mc_resume(struct amdgpu_device *adev, int idx); static void vce_v3_0_set_ring_funcs(struct amdgpu_device *adev); static void vce_v3_0_set_irq_funcs(struct amdgpu_device *adev); -static int vce_v3_0_wait_for_idle(void *handle); +static int vce_v3_0_wait_for_idle(struct amdgpu_ip_block *ip_block); static int vce_v3_0_set_clockgating_state(void *handle, enum amd_clockgating_state state); /** @@ -396,9 +396,9 @@ static unsigned vce_v3_0_get_harvest_config(struct amdgpu_device *adev) } } -static int vce_v3_0_early_init(void *handle) +static int vce_v3_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->vce.harvest_config = vce_v3_0_get_harvest_config(adev); @@ -415,9 +415,9 @@ static int vce_v3_0_early_init(void *handle) return 0; } -static int vce_v3_0_sw_init(void *handle) +static int vce_v3_0_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int r, i; @@ -453,10 +453,10 @@ static int vce_v3_0_sw_init(void *handle) return r; } -static int vce_v3_0_sw_fini(void *handle) +static int vce_v3_0_sw_fini(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_vce_suspend(adev); if (r) @@ -465,10 +465,10 @@ static int vce_v3_0_sw_fini(void *handle) return amdgpu_vce_sw_fini(adev); } -static int vce_v3_0_hw_init(void *handle) +static int vce_v3_0_hw_init(struct amdgpu_ip_block *ip_block) { int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; vce_v3_0_override_vce_clock_gating(adev, true); @@ -485,14 +485,14 @@ static int vce_v3_0_hw_init(void *handle) return 0; } -static int vce_v3_0_hw_fini(void *handle) +static int vce_v3_0_hw_fini(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->vce.idle_work); - r = vce_v3_0_wait_for_idle(handle); + r = vce_v3_0_wait_for_idle(ip_block); if (r) return r; @@ -500,10 +500,10 @@ static int vce_v3_0_hw_fini(void *handle) return vce_v3_0_set_clockgating_state(adev, AMD_CG_STATE_GATE); } -static int vce_v3_0_suspend(void *handle) +static int vce_v3_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* * Proper cleanups before halting the HW engine: @@ -528,23 +528,22 @@ static int vce_v3_0_suspend(void *handle) AMD_CG_STATE_GATE); } - r = vce_v3_0_hw_fini(adev); + r = vce_v3_0_hw_fini(ip_block); if (r) return r; return amdgpu_vce_suspend(adev); } -static int vce_v3_0_resume(void *handle) +static int vce_v3_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_vce_resume(adev); + r = amdgpu_vce_resume(ip_block->adev); if (r) return r; - return vce_v3_0_hw_init(adev); + return vce_v3_0_hw_init(ip_block); } static void vce_v3_0_mc_resume(struct amdgpu_device *adev, int idx) @@ -609,13 +608,13 @@ static bool vce_v3_0_is_idle(void *handle) return !(RREG32(mmSRBM_STATUS2) & mask); } -static int vce_v3_0_wait_for_idle(void *handle) +static int vce_v3_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (i = 0; i < adev->usec_timeout; i++) - if (vce_v3_0_is_idle(handle)) + if (vce_v3_0_is_idle(adev)) return 0; return -ETIMEDOUT; @@ -627,9 +626,9 @@ static int vce_v3_0_wait_for_idle(void *handle) #define AMDGPU_VCE_STATUS_BUSY_MASK (VCE_STATUS_VCPU_REPORT_AUTO_BUSY_MASK | \ VCE_STATUS_VCPU_REPORT_RB0_BUSY_MASK) -static bool vce_v3_0_check_soft_reset(void *handle) +static bool vce_v3_0_check_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset = 0; /* According to VCE team , we should use VCE_STATUS instead @@ -668,9 +667,9 @@ static bool vce_v3_0_check_soft_reset(void *handle) } } -static int vce_v3_0_soft_reset(void *handle) +static int vce_v3_0_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; u32 srbm_soft_reset; if (!adev->vce.srbm_soft_reset) @@ -699,29 +698,29 @@ static int vce_v3_0_soft_reset(void *handle) return 0; } -static int vce_v3_0_pre_soft_reset(void *handle) +static int vce_v3_0_pre_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (!adev->vce.srbm_soft_reset) return 0; mdelay(5); - return vce_v3_0_suspend(adev); + return vce_v3_0_suspend(ip_block); } -static int vce_v3_0_post_soft_reset(void *handle) +static int vce_v3_0_post_soft_reset(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (!adev->vce.srbm_soft_reset) return 0; mdelay(5); - return vce_v3_0_resume(adev); + return vce_v3_0_resume(ip_block); } static int vce_v3_0_set_interrupt_state(struct amdgpu_device *adev, @@ -897,7 +896,6 @@ static void vce_v3_0_emit_pipeline_sync(struct amdgpu_ring *ring) static const struct amd_ip_funcs vce_v3_0_ip_funcs = { .name = "vce_v3_0", .early_init = vce_v3_0_early_init, - .late_init = NULL, .sw_init = vce_v3_0_sw_init, .sw_fini = vce_v3_0_sw_fini, .hw_init = vce_v3_0_hw_init, @@ -913,8 +911,6 @@ static const struct amd_ip_funcs vce_v3_0_ip_funcs = { .set_clockgating_state = vce_v3_0_set_clockgating_state, .set_powergating_state = vce_v3_0_set_powergating_state, .get_clockgating_state = vce_v3_0_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ring_funcs vce_v3_0_ring_phys_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c index 0748bf44c880..79ee555768a5 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c @@ -407,9 +407,9 @@ static int vce_v4_0_stop(struct amdgpu_device *adev) return 0; } -static int vce_v4_0_early_init(void *handle) +static int vce_v4_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) /* currently only VCN0 support SRIOV */ adev->vce.num_rings = 1; @@ -422,9 +422,9 @@ static int vce_v4_0_early_init(void *handle) return 0; } -static int vce_v4_0_sw_init(void *handle) +static int vce_v4_0_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; unsigned size; @@ -493,10 +493,10 @@ static int vce_v4_0_sw_init(void *handle) return r; } -static int vce_v4_0_sw_fini(void *handle) +static int vce_v4_0_sw_fini(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* free MM table */ amdgpu_virt_free_mm_table(adev); @@ -513,10 +513,10 @@ static int vce_v4_0_sw_fini(void *handle) return amdgpu_vce_sw_fini(adev); } -static int vce_v4_0_hw_init(void *handle) +static int vce_v4_0_hw_init(struct amdgpu_ip_block *ip_block) { int r, i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) r = vce_v4_0_sriov_start(adev); @@ -536,14 +536,14 @@ static int vce_v4_0_hw_init(void *handle) return 0; } -static int vce_v4_0_hw_fini(void *handle) +static int vce_v4_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->vce.idle_work); if (!amdgpu_sriov_vf(adev)) { - /* vce_v4_0_wait_for_idle(handle); */ + /* vce_v4_0_wait_for_idle(ip_block); */ vce_v4_0_stop(adev); } else { /* full access mode, so don't touch any VCE register */ @@ -553,9 +553,9 @@ static int vce_v4_0_hw_fini(void *handle) return 0; } -static int vce_v4_0_suspend(void *handle) +static int vce_v4_0_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r, idx; if (adev->vce.vcpu_bo == NULL) @@ -594,16 +594,16 @@ static int vce_v4_0_suspend(void *handle) AMD_CG_STATE_GATE); } - r = vce_v4_0_hw_fini(adev); + r = vce_v4_0_hw_fini(ip_block); if (r) return r; return amdgpu_vce_suspend(adev); } -static int vce_v4_0_resume(void *handle) +static int vce_v4_0_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r, idx; if (adev->vce.vcpu_bo == NULL) @@ -624,7 +624,7 @@ static int vce_v4_0_resume(void *handle) return r; } - return vce_v4_0_hw_init(adev); + return vce_v4_0_hw_init(ip_block); } static void vce_v4_0_mc_resume(struct amdgpu_device *adev) @@ -691,273 +691,6 @@ static int vce_v4_0_set_clockgating_state(void *handle, return 0; } -#if 0 -static bool vce_v4_0_is_idle(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - u32 mask = 0; - - mask |= (adev->vce.harvest_config & AMDGPU_VCE_HARVEST_VCE0) ? 0 : SRBM_STATUS2__VCE0_BUSY_MASK; - mask |= (adev->vce.harvest_config & AMDGPU_VCE_HARVEST_VCE1) ? 0 : SRBM_STATUS2__VCE1_BUSY_MASK; - - return !(RREG32(mmSRBM_STATUS2) & mask); -} - -static int vce_v4_0_wait_for_idle(void *handle) -{ - unsigned i; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - for (i = 0; i < adev->usec_timeout; i++) - if (vce_v4_0_is_idle(handle)) - return 0; - - return -ETIMEDOUT; -} - -#define VCE_STATUS_VCPU_REPORT_AUTO_BUSY_MASK 0x00000008L /* AUTO_BUSY */ -#define VCE_STATUS_VCPU_REPORT_RB0_BUSY_MASK 0x00000010L /* RB0_BUSY */ -#define VCE_STATUS_VCPU_REPORT_RB1_BUSY_MASK 0x00000020L /* RB1_BUSY */ -#define AMDGPU_VCE_STATUS_BUSY_MASK (VCE_STATUS_VCPU_REPORT_AUTO_BUSY_MASK | \ - VCE_STATUS_VCPU_REPORT_RB0_BUSY_MASK) - -static bool vce_v4_0_check_soft_reset(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - u32 srbm_soft_reset = 0; - - /* According to VCE team , we should use VCE_STATUS instead - * SRBM_STATUS.VCE_BUSY bit for busy status checking. - * GRBM_GFX_INDEX.INSTANCE_INDEX is used to specify which VCE - * instance's registers are accessed - * (0 for 1st instance, 10 for 2nd instance). - * - *VCE_STATUS - *|UENC|ACPI|AUTO ACTIVE|RB1 |RB0 |RB2 | |FW_LOADED|JOB | - *|----+----+-----------+----+----+----+----------+---------+----| - *|bit8|bit7| bit6 |bit5|bit4|bit3| bit2 | bit1 |bit0| - * - * VCE team suggest use bit 3--bit 6 for busy status check - */ - mutex_lock(&adev->grbm_idx_mutex); - WREG32_FIELD(GRBM_GFX_INDEX, INSTANCE_INDEX, 0); - if (RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_STATUS) & AMDGPU_VCE_STATUS_BUSY_MASK) { - srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_VCE0, 1); - srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_VCE1, 1); - } - WREG32_FIELD(GRBM_GFX_INDEX, INSTANCE_INDEX, 0x10); - if (RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_STATUS) & AMDGPU_VCE_STATUS_BUSY_MASK) { - srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_VCE0, 1); - srbm_soft_reset = REG_SET_FIELD(srbm_soft_reset, SRBM_SOFT_RESET, SOFT_RESET_VCE1, 1); - } - WREG32_FIELD(GRBM_GFX_INDEX, INSTANCE_INDEX, 0); - mutex_unlock(&adev->grbm_idx_mutex); - - if (srbm_soft_reset) { - adev->vce.srbm_soft_reset = srbm_soft_reset; - return true; - } else { - adev->vce.srbm_soft_reset = 0; - return false; - } -} - -static int vce_v4_0_soft_reset(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - u32 srbm_soft_reset; - - if (!adev->vce.srbm_soft_reset) - return 0; - srbm_soft_reset = adev->vce.srbm_soft_reset; - - if (srbm_soft_reset) { - u32 tmp; - - tmp = RREG32(mmSRBM_SOFT_RESET); - tmp |= srbm_soft_reset; - dev_info(adev->dev, "SRBM_SOFT_RESET=0x%08X\n", tmp); - WREG32(mmSRBM_SOFT_RESET, tmp); - tmp = RREG32(mmSRBM_SOFT_RESET); - - udelay(50); - - tmp &= ~srbm_soft_reset; - WREG32(mmSRBM_SOFT_RESET, tmp); - tmp = RREG32(mmSRBM_SOFT_RESET); - - /* Wait a little for things to settle down */ - udelay(50); - } - - return 0; -} - -static int vce_v4_0_pre_soft_reset(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - if (!adev->vce.srbm_soft_reset) - return 0; - - mdelay(5); - - return vce_v4_0_suspend(adev); -} - - -static int vce_v4_0_post_soft_reset(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - if (!adev->vce.srbm_soft_reset) - return 0; - - mdelay(5); - - return vce_v4_0_resume(adev); -} - -static void vce_v4_0_override_vce_clock_gating(struct amdgpu_device *adev, bool override) -{ - u32 tmp, data; - - tmp = data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_RB_ARB_CTRL)); - if (override) - data |= VCE_RB_ARB_CTRL__VCE_CGTT_OVERRIDE_MASK; - else - data &= ~VCE_RB_ARB_CTRL__VCE_CGTT_OVERRIDE_MASK; - - if (tmp != data) - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_RB_ARB_CTRL), data); -} - -static void vce_v4_0_set_vce_sw_clock_gating(struct amdgpu_device *adev, - bool gated) -{ - u32 data; - - /* Set Override to disable Clock Gating */ - vce_v4_0_override_vce_clock_gating(adev, true); - - /* This function enables MGCG which is controlled by firmware. - With the clocks in the gated state the core is still - accessible but the firmware will throttle the clocks on the - fly as necessary. - */ - if (gated) { - data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_CLOCK_GATING_B)); - data |= 0x1ff; - data &= ~0xef0000; - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_CLOCK_GATING_B), data); - - data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_CLOCK_GATING)); - data |= 0x3ff000; - data &= ~0xffc00000; - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_CLOCK_GATING), data); - - data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_CLOCK_GATING_2)); - data |= 0x2; - data &= ~0x00010000; - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_CLOCK_GATING_2), data); - - data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_REG_CLOCK_GATING)); - data |= 0x37f; - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_REG_CLOCK_GATING), data); - - data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_DMA_DCLK_CTRL)); - data |= VCE_UENC_DMA_DCLK_CTRL__WRDMCLK_FORCEON_MASK | - VCE_UENC_DMA_DCLK_CTRL__RDDMCLK_FORCEON_MASK | - VCE_UENC_DMA_DCLK_CTRL__REGCLK_FORCEON_MASK | - 0x8; - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_DMA_DCLK_CTRL), data); - } else { - data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_CLOCK_GATING_B)); - data &= ~0x80010; - data |= 0xe70008; - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_CLOCK_GATING_B), data); - - data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_CLOCK_GATING)); - data |= 0xffc00000; - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_CLOCK_GATING), data); - - data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_CLOCK_GATING_2)); - data |= 0x10000; - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_CLOCK_GATING_2), data); - - data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_REG_CLOCK_GATING)); - data &= ~0xffc00000; - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_REG_CLOCK_GATING), data); - - data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_DMA_DCLK_CTRL)); - data &= ~(VCE_UENC_DMA_DCLK_CTRL__WRDMCLK_FORCEON_MASK | - VCE_UENC_DMA_DCLK_CTRL__RDDMCLK_FORCEON_MASK | - VCE_UENC_DMA_DCLK_CTRL__REGCLK_FORCEON_MASK | - 0x8); - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_DMA_DCLK_CTRL), data); - } - vce_v4_0_override_vce_clock_gating(adev, false); -} - -static void vce_v4_0_set_bypass_mode(struct amdgpu_device *adev, bool enable) -{ - u32 tmp = RREG32_SMC(ixGCK_DFS_BYPASS_CNTL); - - if (enable) - tmp |= GCK_DFS_BYPASS_CNTL__BYPASSECLK_MASK; - else - tmp &= ~GCK_DFS_BYPASS_CNTL__BYPASSECLK_MASK; - - WREG32_SMC(ixGCK_DFS_BYPASS_CNTL, tmp); -} - -static int vce_v4_0_set_clockgating_state(void *handle, - enum amd_clockgating_state state) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - bool enable = (state == AMD_CG_STATE_GATE); - int i; - - if ((adev->asic_type == CHIP_POLARIS10) || - (adev->asic_type == CHIP_TONGA) || - (adev->asic_type == CHIP_FIJI)) - vce_v4_0_set_bypass_mode(adev, enable); - - if (!(adev->cg_flags & AMD_CG_SUPPORT_VCE_MGCG)) - return 0; - - mutex_lock(&adev->grbm_idx_mutex); - for (i = 0; i < 2; i++) { - /* Program VCE Instance 0 or 1 if not harvested */ - if (adev->vce.harvest_config & (1 << i)) - continue; - - WREG32_FIELD(GRBM_GFX_INDEX, VCE_INSTANCE, i); - - if (enable) { - /* initialize VCE_CLOCK_GATING_A: Clock ON/OFF delay */ - uint32_t data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_CLOCK_GATING_A); - data &= ~(0xf | 0xff0); - data |= ((0x0 << 0) | (0x04 << 4)); - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_CLOCK_GATING_A, data); - - /* initialize VCE_UENC_CLOCK_GATING: Clock ON/OFF delay */ - data = RREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_CLOCK_GATING); - data &= ~(0xf | 0xff0); - data |= ((0x0 << 0) | (0x04 << 4)); - WREG32(SOC15_REG_OFFSET(VCE, 0, mmVCE_UENC_CLOCK_GATING, data); - } - - vce_v4_0_set_vce_sw_clock_gating(adev, enable); - } - - WREG32_FIELD(GRBM_GFX_INDEX, VCE_INSTANCE, 0); - mutex_unlock(&adev->grbm_idx_mutex); - - return 0; -} -#endif - static int vce_v4_0_set_powergating_state(void *handle, enum amd_powergating_state state) { @@ -1076,19 +809,12 @@ static int vce_v4_0_process_interrupt(struct amdgpu_device *adev, const struct amd_ip_funcs vce_v4_0_ip_funcs = { .name = "vce_v4_0", .early_init = vce_v4_0_early_init, - .late_init = NULL, .sw_init = vce_v4_0_sw_init, .sw_fini = vce_v4_0_sw_fini, .hw_init = vce_v4_0_hw_init, .hw_fini = vce_v4_0_hw_fini, .suspend = vce_v4_0_suspend, .resume = vce_v4_0_resume, - .is_idle = NULL /* vce_v4_0_is_idle */, - .wait_for_idle = NULL /* vce_v4_0_wait_for_idle */, - .check_soft_reset = NULL /* vce_v4_0_check_soft_reset */, - .pre_soft_reset = NULL /* vce_v4_0_pre_soft_reset */, - .soft_reset = NULL /* vce_v4_0_soft_reset */, - .post_soft_reset = NULL /* vce_v4_0_post_soft_reset */, .set_clockgating_state = vce_v4_0_set_clockgating_state, .set_powergating_state = vce_v4_0_set_powergating_state, }; diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index ecdfbfefd66a..10e99c926fb8 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -95,14 +95,14 @@ static void vcn_v1_0_ring_begin_use(struct amdgpu_ring *ring); /** * vcn_v1_0_early_init - set function pointers and load microcode * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers * Load microcode from filesystem */ -static int vcn_v1_0_early_init(void *handle) +static int vcn_v1_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->vcn.num_enc_rings = 2; @@ -110,7 +110,7 @@ static int vcn_v1_0_early_init(void *handle) vcn_v1_0_set_enc_ring_funcs(adev); vcn_v1_0_set_irq_funcs(adev); - jpeg_v1_0_early_init(handle); + jpeg_v1_0_early_init(ip_block); return amdgpu_vcn_early_init(adev); } @@ -118,17 +118,17 @@ static int vcn_v1_0_early_init(void *handle) /** * vcn_v1_0_sw_init - sw init for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int vcn_v1_0_sw_init(void *handle) +static int vcn_v1_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int i, r; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_1_0); uint32_t *ptr; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* VCN DEC TRAP */ r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VCN, @@ -197,7 +197,7 @@ static int vcn_v1_0_sw_init(void *handle) amdgpu_vcn_fwlog_init(adev->vcn.inst); } - r = jpeg_v1_0_sw_init(handle); + r = jpeg_v1_0_sw_init(ip_block); /* Allocate memory for VCN IP Dump buffer */ ptr = kcalloc(adev->vcn.num_vcn_inst * reg_count, sizeof(uint32_t), GFP_KERNEL); @@ -213,20 +213,20 @@ static int vcn_v1_0_sw_init(void *handle) /** * vcn_v1_0_sw_fini - sw fini for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * VCN suspend and free up sw allocation */ -static int vcn_v1_0_sw_fini(void *handle) +static int vcn_v1_0_sw_fini(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_vcn_suspend(adev); if (r) return r; - jpeg_v1_0_sw_fini(handle); + jpeg_v1_0_sw_fini(ip_block); r = amdgpu_vcn_sw_fini(adev); @@ -238,13 +238,13 @@ static int vcn_v1_0_sw_fini(void *handle) /** * vcn_v1_0_hw_init - start and test VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int vcn_v1_0_hw_init(void *handle) +static int vcn_v1_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec; int i, r; @@ -268,13 +268,13 @@ static int vcn_v1_0_hw_init(void *handle) /** * vcn_v1_0_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the VCN block, mark ring as not ready any more */ -static int vcn_v1_0_hw_fini(void *handle) +static int vcn_v1_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -290,14 +290,14 @@ static int vcn_v1_0_hw_fini(void *handle) /** * vcn_v1_0_suspend - suspend VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend VCN block */ -static int vcn_v1_0_suspend(void *handle) +static int vcn_v1_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; bool idle_work_unexecuted; idle_work_unexecuted = cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -306,7 +306,7 @@ static int vcn_v1_0_suspend(void *handle) amdgpu_dpm_enable_uvd(adev, false); } - r = vcn_v1_0_hw_fini(adev); + r = vcn_v1_0_hw_fini(ip_block); if (r) return r; @@ -318,20 +318,19 @@ static int vcn_v1_0_suspend(void *handle) /** * vcn_v1_0_resume - resume VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init VCN block */ -static int vcn_v1_0_resume(void *handle) +static int vcn_v1_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_vcn_resume(adev); + r = amdgpu_vcn_resume(ip_block->adev); if (r) return r; - r = vcn_v1_0_hw_init(adev); + r = vcn_v1_0_hw_init(ip_block); return r; } @@ -1384,9 +1383,9 @@ static bool vcn_v1_0_is_idle(void *handle) return (RREG32_SOC15(VCN, 0, mmUVD_STATUS) == UVD_STATUS__IDLE); } -static int vcn_v1_0_wait_for_idle(void *handle) +static int vcn_v1_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; ret = SOC15_WAIT_ON_RREG(VCN, 0, mmUVD_STATUS, UVD_STATUS__IDLE, @@ -1925,9 +1924,9 @@ void vcn_v1_0_ring_end_use(struct amdgpu_ring *ring) mutex_unlock(&ring->adev->vcn.vcn1_jpeg1_workaround); } -static void vcn_v1_0_print_ip_state(void *handle, struct drm_printer *p) +static void vcn_v1_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_1_0); uint32_t inst_off, is_powered; @@ -1957,9 +1956,9 @@ static void vcn_v1_0_print_ip_state(void *handle, struct drm_printer *p) } } -static void vcn_v1_0_dump_ip_state(void *handle) +static void vcn_v1_0_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; bool is_powered; uint32_t inst_off; @@ -1988,7 +1987,6 @@ static void vcn_v1_0_dump_ip_state(void *handle) static const struct amd_ip_funcs vcn_v1_0_ip_funcs = { .name = "vcn_v1_0", .early_init = vcn_v1_0_early_init, - .late_init = NULL, .sw_init = vcn_v1_0_sw_init, .sw_fini = vcn_v1_0_sw_fini, .hw_init = vcn_v1_0_hw_init, @@ -1997,10 +1995,6 @@ static const struct amd_ip_funcs vcn_v1_0_ip_funcs = { .resume = vcn_v1_0_resume, .is_idle = vcn_v1_0_is_idle, .wait_for_idle = vcn_v1_0_wait_for_idle, - .check_soft_reset = NULL /* vcn_v1_0_check_soft_reset */, - .pre_soft_reset = NULL /* vcn_v1_0_pre_soft_reset */, - .soft_reset = NULL /* vcn_v1_0_soft_reset */, - .post_soft_reset = NULL /* vcn_v1_0_post_soft_reset */, .set_clockgating_state = vcn_v1_0_set_clockgating_state, .set_powergating_state = vcn_v1_0_set_powergating_state, .dump_ip_state = vcn_v1_0_dump_ip_state, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c index bfd067e2d2f1..e0322cbca3ec 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_0.c @@ -100,14 +100,14 @@ static int vcn_v2_0_start_sriov(struct amdgpu_device *adev); /** * vcn_v2_0_early_init - set function pointers and load microcode * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers * Load microcode from filesystem */ -static int vcn_v2_0_early_init(void *handle) +static int vcn_v2_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) adev->vcn.num_enc_rings = 1; @@ -124,17 +124,17 @@ static int vcn_v2_0_early_init(void *handle) /** * vcn_v2_0_sw_init - sw init for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int vcn_v2_0_sw_init(void *handle) +static int vcn_v2_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int i, r; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_0); uint32_t *ptr; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; volatile struct amdgpu_fw_shared *fw_shared; /* VCN DEC TRAP */ @@ -237,14 +237,14 @@ static int vcn_v2_0_sw_init(void *handle) /** * vcn_v2_0_sw_fini - sw fini for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * VCN suspend and free up sw allocation */ -static int vcn_v2_0_sw_fini(void *handle) +static int vcn_v2_0_sw_fini(struct amdgpu_ip_block *ip_block) { int r, idx; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; volatile struct amdgpu_fw_shared *fw_shared = adev->vcn.inst->fw_shared.cpu_addr; if (drm_dev_enter(adev_to_drm(adev), &idx)) { @@ -268,13 +268,13 @@ static int vcn_v2_0_sw_fini(void *handle) /** * vcn_v2_0_hw_init - start and test VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int vcn_v2_0_hw_init(void *handle) +static int vcn_v2_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring = &adev->vcn.inst->ring_dec; int i, r; @@ -305,13 +305,13 @@ static int vcn_v2_0_hw_init(void *handle) /** * vcn_v2_0_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the VCN block, mark ring as not ready any more */ -static int vcn_v2_0_hw_fini(void *handle) +static int vcn_v2_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -326,20 +326,19 @@ static int vcn_v2_0_hw_fini(void *handle) /** * vcn_v2_0_suspend - suspend VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend VCN block */ -static int vcn_v2_0_suspend(void *handle) +static int vcn_v2_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = vcn_v2_0_hw_fini(adev); + r = vcn_v2_0_hw_fini(ip_block); if (r) return r; - r = amdgpu_vcn_suspend(adev); + r = amdgpu_vcn_suspend(ip_block->adev); return r; } @@ -347,20 +346,19 @@ static int vcn_v2_0_suspend(void *handle) /** * vcn_v2_0_resume - resume VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init VCN block */ -static int vcn_v2_0_resume(void *handle) +static int vcn_v2_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_vcn_resume(adev); + r = amdgpu_vcn_resume(ip_block->adev); if (r) return r; - r = vcn_v2_0_hw_init(adev); + r = vcn_v2_0_hw_init(ip_block); return r; } @@ -1326,9 +1324,9 @@ static bool vcn_v2_0_is_idle(void *handle) return (RREG32_SOC15(VCN, 0, mmUVD_STATUS) == UVD_STATUS__IDLE); } -static int vcn_v2_0_wait_for_idle(void *handle) +static int vcn_v2_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int ret; ret = SOC15_WAIT_ON_RREG(VCN, 0, mmUVD_STATUS, UVD_STATUS__IDLE, @@ -2034,9 +2032,9 @@ static int vcn_v2_0_start_sriov(struct amdgpu_device *adev) return vcn_v2_0_start_mmsch(adev, &adev->virt.mm_table); } -static void vcn_v2_0_print_ip_state(void *handle, struct drm_printer *p) +static void vcn_v2_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_0); uint32_t inst_off, is_powered; @@ -2066,9 +2064,9 @@ static void vcn_v2_0_print_ip_state(void *handle, struct drm_printer *p) } } -static void vcn_v2_0_dump_ip_state(void *handle) +static void vcn_v2_0_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; bool is_powered; uint32_t inst_off; @@ -2097,7 +2095,6 @@ static void vcn_v2_0_dump_ip_state(void *handle) static const struct amd_ip_funcs vcn_v2_0_ip_funcs = { .name = "vcn_v2_0", .early_init = vcn_v2_0_early_init, - .late_init = NULL, .sw_init = vcn_v2_0_sw_init, .sw_fini = vcn_v2_0_sw_fini, .hw_init = vcn_v2_0_hw_init, @@ -2106,10 +2103,6 @@ static const struct amd_ip_funcs vcn_v2_0_ip_funcs = { .resume = vcn_v2_0_resume, .is_idle = vcn_v2_0_is_idle, .wait_for_idle = vcn_v2_0_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = vcn_v2_0_set_clockgating_state, .set_powergating_state = vcn_v2_0_set_powergating_state, .dump_ip_state = vcn_v2_0_dump_ip_state, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c index 04e9e806e318..6aa08281d094 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c @@ -110,14 +110,14 @@ static int amdgpu_ih_clientid_vcns[] = { /** * vcn_v2_5_early_init - set function pointers and load microcode * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers * Load microcode from filesystem */ -static int vcn_v2_5_early_init(void *handle) +static int vcn_v2_5_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) { adev->vcn.num_vcn_inst = 2; @@ -151,17 +151,17 @@ static int vcn_v2_5_early_init(void *handle) /** * vcn_v2_5_sw_init - sw init for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int vcn_v2_5_sw_init(void *handle) +static int vcn_v2_5_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int i, j, r; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_5); uint32_t *ptr; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; for (j = 0; j < adev->vcn.num_vcn_inst; j++) { if (adev->vcn.harvest_config & (1 << j)) @@ -295,14 +295,14 @@ static int vcn_v2_5_sw_init(void *handle) /** * vcn_v2_5_sw_fini - sw fini for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * VCN suspend and free up sw allocation */ -static int vcn_v2_5_sw_fini(void *handle) +static int vcn_v2_5_sw_fini(struct amdgpu_ip_block *ip_block) { int i, r, idx; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; volatile struct amdgpu_fw_shared *fw_shared; if (drm_dev_enter(adev_to_drm(adev), &idx)) { @@ -333,13 +333,13 @@ static int vcn_v2_5_sw_fini(void *handle) /** * vcn_v2_5_hw_init - start and test VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int vcn_v2_5_hw_init(void *handle) +static int vcn_v2_5_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, j, r = 0; @@ -381,13 +381,13 @@ static int vcn_v2_5_hw_init(void *handle) /** * vcn_v2_5_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the VCN block, mark ring as not ready any more */ -static int vcn_v2_5_hw_fini(void *handle) +static int vcn_v2_5_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -411,20 +411,19 @@ static int vcn_v2_5_hw_fini(void *handle) /** * vcn_v2_5_suspend - suspend VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend VCN block */ -static int vcn_v2_5_suspend(void *handle) +static int vcn_v2_5_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = vcn_v2_5_hw_fini(adev); + r = vcn_v2_5_hw_fini(ip_block); if (r) return r; - r = amdgpu_vcn_suspend(adev); + r = amdgpu_vcn_suspend(ip_block->adev); return r; } @@ -432,20 +431,19 @@ static int vcn_v2_5_suspend(void *handle) /** * vcn_v2_5_resume - resume VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init VCN block */ -static int vcn_v2_5_resume(void *handle) +static int vcn_v2_5_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_vcn_resume(adev); + r = amdgpu_vcn_resume(ip_block->adev); if (r) return r; - r = vcn_v2_5_hw_init(adev); + r = vcn_v2_5_hw_init(ip_block); return r; } @@ -1786,9 +1784,9 @@ static bool vcn_v2_5_is_idle(void *handle) return ret; } -static int vcn_v2_5_wait_for_idle(void *handle) +static int vcn_v2_5_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, ret = 0; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { @@ -1926,9 +1924,9 @@ static void vcn_v2_5_set_irq_funcs(struct amdgpu_device *adev) } } -static void vcn_v2_5_print_ip_state(void *handle, struct drm_printer *p) +static void vcn_v2_5_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_2_5); uint32_t inst_off, is_powered; @@ -1958,9 +1956,9 @@ static void vcn_v2_5_print_ip_state(void *handle, struct drm_printer *p) } } -static void vcn_v2_5_dump_ip_state(void *handle) +static void vcn_v2_5_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; bool is_powered; uint32_t inst_off; @@ -1989,7 +1987,6 @@ static void vcn_v2_5_dump_ip_state(void *handle) static const struct amd_ip_funcs vcn_v2_5_ip_funcs = { .name = "vcn_v2_5", .early_init = vcn_v2_5_early_init, - .late_init = NULL, .sw_init = vcn_v2_5_sw_init, .sw_fini = vcn_v2_5_sw_fini, .hw_init = vcn_v2_5_hw_init, @@ -1998,10 +1995,6 @@ static const struct amd_ip_funcs vcn_v2_5_ip_funcs = { .resume = vcn_v2_5_resume, .is_idle = vcn_v2_5_is_idle, .wait_for_idle = vcn_v2_5_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = vcn_v2_5_set_clockgating_state, .set_powergating_state = vcn_v2_5_set_powergating_state, .dump_ip_state = vcn_v2_5_dump_ip_state, @@ -2011,7 +2004,6 @@ static const struct amd_ip_funcs vcn_v2_5_ip_funcs = { static const struct amd_ip_funcs vcn_v2_6_ip_funcs = { .name = "vcn_v2_6", .early_init = vcn_v2_5_early_init, - .late_init = NULL, .sw_init = vcn_v2_5_sw_init, .sw_fini = vcn_v2_5_sw_fini, .hw_init = vcn_v2_5_hw_init, @@ -2020,10 +2012,6 @@ static const struct amd_ip_funcs vcn_v2_6_ip_funcs = { .resume = vcn_v2_5_resume, .is_idle = vcn_v2_5_is_idle, .wait_for_idle = vcn_v2_5_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = vcn_v2_5_set_clockgating_state, .set_powergating_state = vcn_v2_5_set_powergating_state, .dump_ip_state = vcn_v2_5_dump_ip_state, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c index 65dd68b32280..6732ad7f16f5 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c @@ -116,14 +116,14 @@ static void vcn_v3_0_enc_ring_set_wptr(struct amdgpu_ring *ring); /** * vcn_v3_0_early_init - set function pointers and load microcode * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers * Load microcode from filesystem */ -static int vcn_v3_0_early_init(void *handle) +static int vcn_v3_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) { adev->vcn.num_vcn_inst = VCN_INSTANCES_SIENNA_CICHLID; @@ -153,18 +153,18 @@ static int vcn_v3_0_early_init(void *handle) /** * vcn_v3_0_sw_init - sw init for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int vcn_v3_0_sw_init(void *handle) +static int vcn_v3_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; int i, j, r; int vcn_doorbell_index = 0; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_3_0); uint32_t *ptr; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = amdgpu_vcn_sw_init(adev); if (r) @@ -299,13 +299,13 @@ static int vcn_v3_0_sw_init(void *handle) /** * vcn_v3_0_sw_fini - sw fini for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * VCN suspend and free up sw allocation */ -static int vcn_v3_0_sw_fini(void *handle) +static int vcn_v3_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, r, idx; if (drm_dev_enter(adev_to_drm(adev), &idx)) { @@ -338,13 +338,13 @@ static int vcn_v3_0_sw_fini(void *handle) /** * vcn_v3_0_hw_init - start and test VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int vcn_v3_0_hw_init(void *handle) +static int vcn_v3_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, j, r; @@ -413,13 +413,13 @@ static int vcn_v3_0_hw_init(void *handle) /** * vcn_v3_0_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the VCN block, mark ring as not ready any more */ -static int vcn_v3_0_hw_fini(void *handle) +static int vcn_v3_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -443,20 +443,19 @@ static int vcn_v3_0_hw_fini(void *handle) /** * vcn_v3_0_suspend - suspend VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend VCN block */ -static int vcn_v3_0_suspend(void *handle) +static int vcn_v3_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = vcn_v3_0_hw_fini(adev); + r = vcn_v3_0_hw_fini(ip_block); if (r) return r; - r = amdgpu_vcn_suspend(adev); + r = amdgpu_vcn_suspend(ip_block->adev); return r; } @@ -464,20 +463,19 @@ static int vcn_v3_0_suspend(void *handle) /** * vcn_v3_0_resume - resume VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init VCN block */ -static int vcn_v3_0_resume(void *handle) +static int vcn_v3_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_vcn_resume(adev); + r = amdgpu_vcn_resume(ip_block->adev); if (r) return r; - r = vcn_v3_0_hw_init(adev); + r = vcn_v3_0_hw_init(ip_block); return r; } @@ -2116,9 +2114,9 @@ static bool vcn_v3_0_is_idle(void *handle) return ret; } -static int vcn_v3_0_wait_for_idle(void *handle) +static int vcn_v3_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, ret = 0; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { @@ -2251,9 +2249,9 @@ static void vcn_v3_0_set_irq_funcs(struct amdgpu_device *adev) } } -static void vcn_v3_0_print_ip_state(void *handle, struct drm_printer *p) +static void vcn_v3_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_3_0); uint32_t inst_off; @@ -2284,9 +2282,9 @@ static void vcn_v3_0_print_ip_state(void *handle, struct drm_printer *p) } } -static void vcn_v3_0_dump_ip_state(void *handle) +static void vcn_v3_0_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; bool is_powered; uint32_t inst_off; @@ -2315,7 +2313,6 @@ static void vcn_v3_0_dump_ip_state(void *handle) static const struct amd_ip_funcs vcn_v3_0_ip_funcs = { .name = "vcn_v3_0", .early_init = vcn_v3_0_early_init, - .late_init = NULL, .sw_init = vcn_v3_0_sw_init, .sw_fini = vcn_v3_0_sw_fini, .hw_init = vcn_v3_0_hw_init, @@ -2324,10 +2321,6 @@ static const struct amd_ip_funcs vcn_v3_0_ip_funcs = { .resume = vcn_v3_0_resume, .is_idle = vcn_v3_0_is_idle, .wait_for_idle = vcn_v3_0_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = vcn_v3_0_set_clockgating_state, .set_powergating_state = vcn_v3_0_set_powergating_state, .dump_ip_state = vcn_v3_0_dump_ip_state, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c index 26c6f10a8c8f..5512259cac79 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0.c @@ -106,14 +106,14 @@ static void vcn_v4_0_set_ras_funcs(struct amdgpu_device *adev); /** * vcn_v4_0_early_init - set function pointers and load microcode * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers * Load microcode from filesystem */ -static int vcn_v4_0_early_init(void *handle) +static int vcn_v4_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; if (amdgpu_sriov_vf(adev)) { @@ -164,14 +164,14 @@ static int vcn_v4_0_fw_shared_init(struct amdgpu_device *adev, int inst_idx) /** * vcn_v4_0_sw_init - sw init for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int vcn_v4_0_sw_init(void *handle) +static int vcn_v4_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, r; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0); uint32_t *ptr; @@ -253,13 +253,13 @@ static int vcn_v4_0_sw_init(void *handle) /** * vcn_v4_0_sw_fini - sw fini for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * VCN suspend and free up sw allocation */ -static int vcn_v4_0_sw_fini(void *handle) +static int vcn_v4_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, r, idx; if (drm_dev_enter(adev_to_drm(adev), &idx)) { @@ -294,13 +294,13 @@ static int vcn_v4_0_sw_fini(void *handle) /** * vcn_v4_0_hw_init - start and test VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int vcn_v4_0_hw_init(void *handle) +static int vcn_v4_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, r; @@ -341,13 +341,13 @@ static int vcn_v4_0_hw_init(void *handle) /** * vcn_v4_0_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the VCN block, mark ring as not ready any more */ -static int vcn_v4_0_hw_fini(void *handle) +static int vcn_v4_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -372,20 +372,19 @@ static int vcn_v4_0_hw_fini(void *handle) /** * vcn_v4_0_suspend - suspend VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend VCN block */ -static int vcn_v4_0_suspend(void *handle) +static int vcn_v4_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = vcn_v4_0_hw_fini(adev); + r = vcn_v4_0_hw_fini(ip_block); if (r) return r; - r = amdgpu_vcn_suspend(adev); + r = amdgpu_vcn_suspend(ip_block->adev); return r; } @@ -393,20 +392,19 @@ static int vcn_v4_0_suspend(void *handle) /** * vcn_v4_0_resume - resume VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init VCN block */ -static int vcn_v4_0_resume(void *handle) +static int vcn_v4_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_vcn_resume(adev); + r = amdgpu_vcn_resume(ip_block->adev); if (r) return r; - r = vcn_v4_0_hw_init(adev); + r = vcn_v4_0_hw_init(ip_block); return r; } @@ -1975,13 +1973,13 @@ static bool vcn_v4_0_is_idle(void *handle) /** * vcn_v4_0_wait_for_idle - wait for VCN block idle * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Wait for VCN block idle */ -static int vcn_v4_0_wait_for_idle(void *handle) +static int vcn_v4_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, ret = 0; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { @@ -2158,9 +2156,9 @@ static void vcn_v4_0_set_irq_funcs(struct amdgpu_device *adev) } } -static void vcn_v4_0_print_ip_state(void *handle, struct drm_printer *p) +static void vcn_v4_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0); uint32_t inst_off, is_powered; @@ -2190,9 +2188,9 @@ static void vcn_v4_0_print_ip_state(void *handle, struct drm_printer *p) } } -static void vcn_v4_0_dump_ip_state(void *handle) +static void vcn_v4_0_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; bool is_powered; uint32_t inst_off; @@ -2222,7 +2220,6 @@ static void vcn_v4_0_dump_ip_state(void *handle) static const struct amd_ip_funcs vcn_v4_0_ip_funcs = { .name = "vcn_v4_0", .early_init = vcn_v4_0_early_init, - .late_init = NULL, .sw_init = vcn_v4_0_sw_init, .sw_fini = vcn_v4_0_sw_fini, .hw_init = vcn_v4_0_hw_init, @@ -2231,10 +2228,6 @@ static const struct amd_ip_funcs vcn_v4_0_ip_funcs = { .resume = vcn_v4_0_resume, .is_idle = vcn_v4_0_is_idle, .wait_for_idle = vcn_v4_0_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = vcn_v4_0_set_clockgating_state, .set_powergating_state = vcn_v4_0_set_powergating_state, .dump_ip_state = vcn_v4_0_dump_ip_state, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c index 0fda70336300..0d5c94bfc0ef 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c @@ -98,13 +98,13 @@ static void vcn_v4_0_3_enable_ras(struct amdgpu_device *adev, /** * vcn_v4_0_3_early_init - set function pointers * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers */ -static int vcn_v4_0_3_early_init(void *handle) +static int vcn_v4_0_3_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* re-use enc ring as unified ring */ adev->vcn.num_enc_rings = 1; @@ -119,13 +119,13 @@ static int vcn_v4_0_3_early_init(void *handle) /** * vcn_v4_0_3_sw_init - sw init for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int vcn_v4_0_3_sw_init(void *handle) +static int vcn_v4_0_3_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, r, vcn_inst; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_3); @@ -212,13 +212,13 @@ static int vcn_v4_0_3_sw_init(void *handle) /** * vcn_v4_0_3_sw_fini - sw fini for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * VCN suspend and free up sw allocation */ -static int vcn_v4_0_3_sw_fini(void *handle) +static int vcn_v4_0_3_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, r, idx; if (drm_dev_enter(&adev->ddev, &idx)) { @@ -249,13 +249,13 @@ static int vcn_v4_0_3_sw_fini(void *handle) /** * vcn_v4_0_3_hw_init - start and test VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int vcn_v4_0_3_hw_init(void *handle) +static int vcn_v4_0_3_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, r, vcn_inst; @@ -308,13 +308,13 @@ static int vcn_v4_0_3_hw_init(void *handle) /** * vcn_v4_0_3_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the VCN block, mark ring as not ready any more */ -static int vcn_v4_0_3_hw_fini(void *handle) +static int vcn_v4_0_3_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -327,20 +327,19 @@ static int vcn_v4_0_3_hw_fini(void *handle) /** * vcn_v4_0_3_suspend - suspend VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend VCN block */ -static int vcn_v4_0_3_suspend(void *handle) +static int vcn_v4_0_3_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = vcn_v4_0_3_hw_fini(adev); + r = vcn_v4_0_3_hw_fini(ip_block); if (r) return r; - r = amdgpu_vcn_suspend(adev); + r = amdgpu_vcn_suspend(ip_block->adev); return r; } @@ -348,20 +347,19 @@ static int vcn_v4_0_3_suspend(void *handle) /** * vcn_v4_0_3_resume - resume VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init VCN block */ -static int vcn_v4_0_3_resume(void *handle) +static int vcn_v4_0_3_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; int r; - r = amdgpu_vcn_resume(adev); + r = amdgpu_vcn_resume(ip_block->adev); if (r) return r; - r = vcn_v4_0_3_hw_init(adev); + r = vcn_v4_0_3_hw_init(ip_block); return r; } @@ -1567,13 +1565,13 @@ static bool vcn_v4_0_3_is_idle(void *handle) /** * vcn_v4_0_3_wait_for_idle - wait for VCN block idle * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Wait for VCN block idle */ -static int vcn_v4_0_3_wait_for_idle(void *handle) +static int vcn_v4_0_3_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, ret = 0; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { @@ -1733,9 +1731,9 @@ static void vcn_v4_0_3_set_irq_funcs(struct amdgpu_device *adev) adev->vcn.inst->irq.funcs = &vcn_v4_0_3_irq_funcs; } -static void vcn_v4_0_3_print_ip_state(void *handle, struct drm_printer *p) +static void vcn_v4_0_3_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_3); uint32_t inst_off, is_powered; @@ -1765,9 +1763,9 @@ static void vcn_v4_0_3_print_ip_state(void *handle, struct drm_printer *p) } } -static void vcn_v4_0_3_dump_ip_state(void *handle) +static void vcn_v4_0_3_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; bool is_powered; uint32_t inst_off, inst_id; @@ -1798,7 +1796,6 @@ static void vcn_v4_0_3_dump_ip_state(void *handle) static const struct amd_ip_funcs vcn_v4_0_3_ip_funcs = { .name = "vcn_v4_0_3", .early_init = vcn_v4_0_3_early_init, - .late_init = NULL, .sw_init = vcn_v4_0_3_sw_init, .sw_fini = vcn_v4_0_3_sw_fini, .hw_init = vcn_v4_0_3_hw_init, @@ -1807,10 +1804,6 @@ static const struct amd_ip_funcs vcn_v4_0_3_ip_funcs = { .resume = vcn_v4_0_3_resume, .is_idle = vcn_v4_0_3_is_idle, .wait_for_idle = vcn_v4_0_3_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = vcn_v4_0_3_set_clockgating_state, .set_powergating_state = vcn_v4_0_3_set_powergating_state, .dump_ip_state = vcn_v4_0_3_dump_ip_state, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c index 9d4f5352a62c..71961fb3f7ff 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v4_0_5.c @@ -104,14 +104,14 @@ static void vcn_v4_0_5_unified_ring_set_wptr(struct amdgpu_ring *ring); /** * vcn_v4_0_5_early_init - set function pointers and load microcode * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers * Load microcode from filesystem */ -static int vcn_v4_0_5_early_init(void *handle) +static int vcn_v4_0_5_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* re-use enc ring as unified ring */ adev->vcn.num_enc_rings = 1; @@ -124,14 +124,14 @@ static int vcn_v4_0_5_early_init(void *handle) /** * vcn_v4_0_5_sw_init - sw init for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int vcn_v4_0_5_sw_init(void *handle) +static int vcn_v4_0_5_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, r; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_5); uint32_t *ptr; @@ -220,13 +220,13 @@ static int vcn_v4_0_5_sw_init(void *handle) /** * vcn_v4_0_5_sw_fini - sw fini for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * VCN suspend and free up sw allocation */ -static int vcn_v4_0_5_sw_fini(void *handle) +static int vcn_v4_0_5_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, r, idx; if (drm_dev_enter(adev_to_drm(adev), &idx)) { @@ -261,13 +261,13 @@ static int vcn_v4_0_5_sw_fini(void *handle) /** * vcn_v4_0_5_hw_init - start and test VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int vcn_v4_0_5_hw_init(void *handle) +static int vcn_v4_0_5_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, r; @@ -291,13 +291,13 @@ static int vcn_v4_0_5_hw_init(void *handle) /** * vcn_v4_0_5_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the VCN block, mark ring as not ready any more */ -static int vcn_v4_0_5_hw_fini(void *handle) +static int vcn_v4_0_5_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -320,20 +320,19 @@ static int vcn_v4_0_5_hw_fini(void *handle) /** * vcn_v4_0_5_suspend - suspend VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend VCN block */ -static int vcn_v4_0_5_suspend(void *handle) +static int vcn_v4_0_5_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = vcn_v4_0_5_hw_fini(adev); + r = vcn_v4_0_5_hw_fini(ip_block); if (r) return r; - r = amdgpu_vcn_suspend(adev); + r = amdgpu_vcn_suspend(ip_block->adev); return r; } @@ -341,20 +340,19 @@ static int vcn_v4_0_5_suspend(void *handle) /** * vcn_v4_0_5_resume - resume VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init VCN block */ -static int vcn_v4_0_5_resume(void *handle) +static int vcn_v4_0_5_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_vcn_resume(adev); + r = amdgpu_vcn_resume(ip_block->adev); if (r) return r; - r = vcn_v4_0_5_hw_init(adev); + r = vcn_v4_0_5_hw_init(ip_block); return r; } @@ -1469,13 +1467,13 @@ static bool vcn_v4_0_5_is_idle(void *handle) /** * vcn_v4_0_5_wait_for_idle - wait for VCN block idle * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Wait for VCN block idle */ -static int vcn_v4_0_5_wait_for_idle(void *handle) +static int vcn_v4_0_5_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, ret = 0; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { @@ -1616,9 +1614,9 @@ static void vcn_v4_0_5_set_irq_funcs(struct amdgpu_device *adev) } } -static void vcn_v4_0_5_print_ip_state(void *handle, struct drm_printer *p) +static void vcn_v4_0_5_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_4_0_5); uint32_t inst_off, is_powered; @@ -1648,9 +1646,9 @@ static void vcn_v4_0_5_print_ip_state(void *handle, struct drm_printer *p) } } -static void vcn_v4_0_5_dump_ip_state(void *handle) +static void vcn_v4_0_5_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; bool is_powered; uint32_t inst_off; @@ -1680,7 +1678,6 @@ static void vcn_v4_0_5_dump_ip_state(void *handle) static const struct amd_ip_funcs vcn_v4_0_5_ip_funcs = { .name = "vcn_v4_0_5", .early_init = vcn_v4_0_5_early_init, - .late_init = NULL, .sw_init = vcn_v4_0_5_sw_init, .sw_fini = vcn_v4_0_5_sw_fini, .hw_init = vcn_v4_0_5_hw_init, @@ -1689,10 +1686,6 @@ static const struct amd_ip_funcs vcn_v4_0_5_ip_funcs = { .resume = vcn_v4_0_5_resume, .is_idle = vcn_v4_0_5_is_idle, .wait_for_idle = vcn_v4_0_5_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = vcn_v4_0_5_set_clockgating_state, .set_powergating_state = vcn_v4_0_5_set_powergating_state, .dump_ip_state = vcn_v4_0_5_dump_ip_state, diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c index c305386358b4..fe2cc1a80c13 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v5_0_0.c @@ -87,14 +87,14 @@ static void vcn_v5_0_0_unified_ring_set_wptr(struct amdgpu_ring *ring); /** * vcn_v5_0_0_early_init - set function pointers and load microcode * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Set ring and irq function pointers * Load microcode from filesystem */ -static int vcn_v5_0_0_early_init(void *handle) +static int vcn_v5_0_0_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* re-use enc ring as unified ring */ adev->vcn.num_enc_rings = 1; @@ -108,14 +108,14 @@ static int vcn_v5_0_0_early_init(void *handle) /** * vcn_v5_0_0_sw_init - sw init for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Load firmware and sw initialization */ -static int vcn_v5_0_0_sw_init(void *handle) +static int vcn_v5_0_0_sw_init(struct amdgpu_ip_block *ip_block) { struct amdgpu_ring *ring; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, r; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_5_0); uint32_t *ptr; @@ -187,13 +187,13 @@ static int vcn_v5_0_0_sw_init(void *handle) /** * vcn_v5_0_0_sw_fini - sw fini for VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * VCN suspend and free up sw allocation */ -static int vcn_v5_0_0_sw_fini(void *handle) +static int vcn_v5_0_0_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, r, idx; if (drm_dev_enter(adev_to_drm(adev), &idx)) { @@ -225,13 +225,13 @@ static int vcn_v5_0_0_sw_fini(void *handle) /** * vcn_v5_0_0_hw_init - start and test VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the hardware, boot up the VCPU and do some testing */ -static int vcn_v5_0_0_hw_init(void *handle) +static int vcn_v5_0_0_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_ring *ring; int i, r; @@ -255,13 +255,13 @@ static int vcn_v5_0_0_hw_init(void *handle) /** * vcn_v5_0_0_hw_fini - stop the hardware block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Stop the VCN block, mark ring as not ready any more */ -static int vcn_v5_0_0_hw_fini(void *handle) +static int vcn_v5_0_0_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i; cancel_delayed_work_sync(&adev->vcn.idle_work); @@ -284,20 +284,19 @@ static int vcn_v5_0_0_hw_fini(void *handle) /** * vcn_v5_0_0_suspend - suspend VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * HW fini and suspend VCN block */ -static int vcn_v5_0_0_suspend(void *handle) +static int vcn_v5_0_0_suspend(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = vcn_v5_0_0_hw_fini(adev); + r = vcn_v5_0_0_hw_fini(ip_block); if (r) return r; - r = amdgpu_vcn_suspend(adev); + r = amdgpu_vcn_suspend(ip_block->adev); return r; } @@ -305,20 +304,19 @@ static int vcn_v5_0_0_suspend(void *handle) /** * vcn_v5_0_0_resume - resume VCN block * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Resume firmware and hw init VCN block */ -static int vcn_v5_0_0_resume(void *handle) +static int vcn_v5_0_0_resume(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - r = amdgpu_vcn_resume(adev); + r = amdgpu_vcn_resume(ip_block->adev); if (r) return r; - r = vcn_v5_0_0_hw_init(adev); + r = vcn_v5_0_0_hw_init(ip_block); return r; } @@ -1196,13 +1194,13 @@ static bool vcn_v5_0_0_is_idle(void *handle) /** * vcn_v5_0_0_wait_for_idle - wait for VCN block idle * - * @handle: amdgpu_device pointer + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Wait for VCN block idle */ -static int vcn_v5_0_0_wait_for_idle(void *handle) +static int vcn_v5_0_0_wait_for_idle(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, ret = 0; for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { @@ -1343,9 +1341,9 @@ static void vcn_v5_0_0_set_irq_funcs(struct amdgpu_device *adev) } } -static void vcn_v5_0_print_ip_state(void *handle, struct drm_printer *p) +static void vcn_v5_0_print_ip_state(struct amdgpu_ip_block *ip_block, struct drm_printer *p) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; uint32_t reg_count = ARRAY_SIZE(vcn_reg_list_5_0); uint32_t inst_off, is_powered; @@ -1375,9 +1373,9 @@ static void vcn_v5_0_print_ip_state(void *handle, struct drm_printer *p) } } -static void vcn_v5_0_dump_ip_state(void *handle) +static void vcn_v5_0_dump_ip_state(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int i, j; bool is_powered; uint32_t inst_off; @@ -1406,7 +1404,6 @@ static void vcn_v5_0_dump_ip_state(void *handle) static const struct amd_ip_funcs vcn_v5_0_0_ip_funcs = { .name = "vcn_v5_0_0", .early_init = vcn_v5_0_0_early_init, - .late_init = NULL, .sw_init = vcn_v5_0_0_sw_init, .sw_fini = vcn_v5_0_0_sw_fini, .hw_init = vcn_v5_0_0_hw_init, @@ -1415,10 +1412,6 @@ static const struct amd_ip_funcs vcn_v5_0_0_ip_funcs = { .resume = vcn_v5_0_0_resume, .is_idle = vcn_v5_0_0_is_idle, .wait_for_idle = vcn_v5_0_0_wait_for_idle, - .check_soft_reset = NULL, - .pre_soft_reset = NULL, - .soft_reset = NULL, - .post_soft_reset = NULL, .set_clockgating_state = vcn_v5_0_0_set_clockgating_state, .set_powergating_state = vcn_v5_0_0_set_powergating_state, .dump_ip_state = vcn_v5_0_dump_ip_state, diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c index bf68e18e3824..0fedadd0a6a4 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c @@ -472,18 +472,18 @@ static void vega10_ih_set_self_irq_funcs(struct amdgpu_device *adev) adev->irq.self_irq.funcs = &vega10_ih_self_irq_funcs; } -static int vega10_ih_early_init(void *handle) +static int vega10_ih_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; vega10_ih_set_interrupt_funcs(adev); vega10_ih_set_self_irq_funcs(adev); return 0; } -static int vega10_ih_sw_init(void *handle) +static int vega10_ih_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_IH, 0, @@ -525,43 +525,35 @@ static int vega10_ih_sw_init(void *handle) return r; } -static int vega10_ih_sw_fini(void *handle) +static int vega10_ih_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_fini_sw(adev); return 0; } -static int vega10_ih_hw_init(void *handle) +static int vega10_ih_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return vega10_ih_irq_init(adev); + return vega10_ih_irq_init(ip_block->adev); } -static int vega10_ih_hw_fini(void *handle) +static int vega10_ih_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - vega10_ih_irq_disable(adev); + vega10_ih_irq_disable(ip_block->adev); return 0; } -static int vega10_ih_suspend(void *handle) +static int vega10_ih_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return vega10_ih_hw_fini(adev); + return vega10_ih_hw_fini(ip_block); } -static int vega10_ih_resume(void *handle) +static int vega10_ih_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return vega10_ih_hw_init(adev); + return vega10_ih_hw_init(ip_block); } static bool vega10_ih_is_idle(void *handle) @@ -570,13 +562,13 @@ static bool vega10_ih_is_idle(void *handle) return true; } -static int vega10_ih_wait_for_idle(void *handle) +static int vega10_ih_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* todo */ return -ETIMEDOUT; } -static int vega10_ih_soft_reset(void *handle) +static int vega10_ih_soft_reset(struct amdgpu_ip_block *ip_block) { /* todo */ @@ -633,7 +625,6 @@ static int vega10_ih_set_powergating_state(void *handle, const struct amd_ip_funcs vega10_ih_ip_funcs = { .name = "vega10_ih", .early_init = vega10_ih_early_init, - .late_init = NULL, .sw_init = vega10_ih_sw_init, .sw_fini = vega10_ih_sw_fini, .hw_init = vega10_ih_hw_init, diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c index ac439f0565e3..1c9aff742e43 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c @@ -114,6 +114,33 @@ static int vega20_ih_toggle_ring_interrupts(struct amdgpu_device *adev, tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, RB_ENABLE, (enable ? 1 : 0)); tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, RB_GPU_TS_ENABLE, 1); + if (enable) { + /* Unset the CLEAR_OVERFLOW bit to make sure the next step + * is switching the bit from 0 to 1 + */ + tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 0); + if (amdgpu_sriov_vf(adev) && amdgpu_sriov_reg_indirect_ih(adev)) { + if (psp_reg_program(&adev->psp, ih_regs->psp_reg_id, tmp)) + return -ETIMEDOUT; + } else { + WREG32_NO_KIQ(ih_regs->ih_rb_cntl, tmp); + } + + /* Clear RB_OVERFLOW bit */ + tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 1); + if (amdgpu_sriov_vf(adev) && amdgpu_sriov_reg_indirect_ih(adev)) { + if (psp_reg_program(&adev->psp, ih_regs->psp_reg_id, tmp)) + return -ETIMEDOUT; + } else { + WREG32_NO_KIQ(ih_regs->ih_rb_cntl, tmp); + } + + /* Unset the CLEAR_OVERFLOW bit immediately so new overflows + * can be detected. + */ + tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, WPTR_OVERFLOW_CLEAR, 0); + } + /* enable_intr field is only valid in ring0 */ if (ih == &adev->irq.ih) tmp = REG_SET_FIELD(tmp, IH_RB_CNTL, ENABLE_INTR, (enable ? 1 : 0)); @@ -526,18 +553,18 @@ static void vega20_ih_set_self_irq_funcs(struct amdgpu_device *adev) adev->irq.self_irq.funcs = &vega20_ih_self_irq_funcs; } -static int vega20_ih_early_init(void *handle) +static int vega20_ih_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; vega20_ih_set_interrupt_funcs(adev); vega20_ih_set_self_irq_funcs(adev); return 0; } -static int vega20_ih_sw_init(void *handle) +static int vega20_ih_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; bool use_bus_addr = true; int r; @@ -586,19 +613,19 @@ static int vega20_ih_sw_init(void *handle) return r; } -static int vega20_ih_sw_fini(void *handle) +static int vega20_ih_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_irq_fini_sw(adev); return 0; } -static int vega20_ih_hw_init(void *handle) +static int vega20_ih_hw_init(struct amdgpu_ip_block *ip_block) { int r; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; r = vega20_ih_irq_init(adev); if (r) @@ -607,27 +634,21 @@ static int vega20_ih_hw_init(void *handle) return 0; } -static int vega20_ih_hw_fini(void *handle) +static int vega20_ih_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - vega20_ih_irq_disable(adev); + vega20_ih_irq_disable(ip_block->adev); return 0; } -static int vega20_ih_suspend(void *handle) +static int vega20_ih_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return vega20_ih_hw_fini(adev); + return vega20_ih_hw_fini(ip_block); } -static int vega20_ih_resume(void *handle) +static int vega20_ih_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return vega20_ih_hw_init(adev); + return vega20_ih_hw_init(ip_block); } static bool vega20_ih_is_idle(void *handle) @@ -636,13 +657,13 @@ static bool vega20_ih_is_idle(void *handle) return true; } -static int vega20_ih_wait_for_idle(void *handle) +static int vega20_ih_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* todo */ return -ETIMEDOUT; } -static int vega20_ih_soft_reset(void *handle) +static int vega20_ih_soft_reset(struct amdgpu_ip_block *ip_block) { /* todo */ @@ -696,7 +717,6 @@ static int vega20_ih_set_powergating_state(void *handle, const struct amd_ip_funcs vega20_ih_ip_funcs = { .name = "vega20_ih", .early_init = vega20_ih_early_init, - .late_init = NULL, .sw_init = vega20_ih_sw_init, .sw_fini = vega20_ih_sw_fini, .hw_init = vega20_ih_hw_init, diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c index d39c670f6220..b3fa54c0514e 100644 --- a/drivers/gpu/drm/amd/amdgpu/vi.c +++ b/drivers/gpu/drm/amd/amdgpu/vi.c @@ -1455,9 +1455,9 @@ static const struct amdgpu_asic_funcs vi_asic_funcs = #define CZ_REV_BRISTOL(rev) \ ((rev >= 0xC8 && rev <= 0xCE) || (rev >= 0xE1 && rev <= 0xE6)) -static int vi_common_early_init(void *handle) +static int vi_common_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->flags & AMD_IS_APU) { adev->smc_rreg = &cz_smc_rreg; @@ -1679,9 +1679,9 @@ static int vi_common_early_init(void *handle) return 0; } -static int vi_common_late_init(void *handle) +static int vi_common_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) xgpu_vi_mailbox_get_irq(adev); @@ -1689,9 +1689,9 @@ static int vi_common_late_init(void *handle) return 0; } -static int vi_common_sw_init(void *handle) +static int vi_common_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (amdgpu_sriov_vf(adev)) xgpu_vi_mailbox_add_irq_id(adev); @@ -1699,14 +1699,9 @@ static int vi_common_sw_init(void *handle) return 0; } -static int vi_common_sw_fini(void *handle) +static int vi_common_hw_init(struct amdgpu_ip_block *ip_block) { - return 0; -} - -static int vi_common_hw_init(void *handle) -{ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* move the golden regs per IP block */ vi_init_golden_registers(adev); @@ -1718,9 +1713,9 @@ static int vi_common_hw_init(void *handle) return 0; } -static int vi_common_hw_fini(void *handle) +static int vi_common_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; /* enable the doorbell aperture */ vi_enable_doorbell_aperture(adev, false); @@ -1731,18 +1726,14 @@ static int vi_common_hw_fini(void *handle) return 0; } -static int vi_common_suspend(void *handle) +static int vi_common_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return vi_common_hw_fini(adev); + return vi_common_hw_fini(ip_block); } -static int vi_common_resume(void *handle) +static int vi_common_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - - return vi_common_hw_init(adev); + return vi_common_hw_init(ip_block); } static bool vi_common_is_idle(void *handle) @@ -1750,16 +1741,6 @@ static bool vi_common_is_idle(void *handle) return true; } -static int vi_common_wait_for_idle(void *handle) -{ - return 0; -} - -static int vi_common_soft_reset(void *handle) -{ - return 0; -} - static void vi_update_bif_medium_grain_light_sleep(struct amdgpu_device *adev, bool enable) { @@ -2047,19 +2028,14 @@ static const struct amd_ip_funcs vi_common_ip_funcs = { .early_init = vi_common_early_init, .late_init = vi_common_late_init, .sw_init = vi_common_sw_init, - .sw_fini = vi_common_sw_fini, .hw_init = vi_common_hw_init, .hw_fini = vi_common_hw_fini, .suspend = vi_common_suspend, .resume = vi_common_resume, .is_idle = vi_common_is_idle, - .wait_for_idle = vi_common_wait_for_idle, - .soft_reset = vi_common_soft_reset, .set_clockgating_state = vi_common_set_clockgating_state, .set_powergating_state = vi_common_set_powergating_state, .get_clockgating_state = vi_common_get_clockgating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; static const struct amdgpu_ip_block_version vi_common_ip_block = diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 9044bdb38cf4..065d87841459 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -365,7 +365,7 @@ static int kfd_ioctl_create_queue(struct file *filep, struct kfd_process *p, p->pasid, dev->id); - err = pqm_create_queue(&p->pqm, dev, filep, &q_properties, &queue_id, + err = pqm_create_queue(&p->pqm, dev, &q_properties, &queue_id, NULL, NULL, NULL, &doorbell_offset_in_process); if (err != 0) goto err_create_queue; @@ -1148,7 +1148,7 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep, if (flags & KFD_IOC_ALLOC_MEM_FLAGS_AQL_QUEUE_MEM) size >>= 1; - WRITE_ONCE(pdd->vram_usage, pdd->vram_usage + PAGE_ALIGN(size)); + atomic64_add(PAGE_ALIGN(size), &pdd->vram_usage); } mutex_unlock(&p->mutex); @@ -1219,7 +1219,7 @@ static int kfd_ioctl_free_memory_of_gpu(struct file *filep, kfd_process_device_remove_obj_handle( pdd, GET_IDR_HANDLE(args->handle)); - WRITE_ONCE(pdd->vram_usage, pdd->vram_usage - size); + atomic64_sub(size, &pdd->vram_usage); err_unlock: err_pdd: @@ -2347,7 +2347,7 @@ static int criu_restore_memory_of_gpu(struct kfd_process_device *pdd, } else if (bo_bucket->alloc_flags & KFD_IOC_ALLOC_MEM_FLAGS_VRAM) { bo_bucket->restored_offset = offset; /* Update the VRAM usage count */ - WRITE_ONCE(pdd->vram_usage, pdd->vram_usage + bo_bucket->size); + atomic64_add(bo_bucket->size, &pdd->vram_usage); } return 0; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c index 48caecf7e72e..723f1220e1cc 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c @@ -28,6 +28,7 @@ #include "kfd_topology.h" #include "amdgpu.h" #include "amdgpu_amdkfd.h" +#include "amdgpu_xgmi.h" /* GPU Processor ID base for dGPUs for which VCRAT needs to be created. * GPU processor ID are expressed with Bit[31]=1. @@ -2329,6 +2330,8 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image, continue; if (peer_dev->gpu->kfd->hive_id != kdev->kfd->hive_id) continue; + if (!amdgpu_xgmi_get_is_sharing_enabled(kdev->adev, peer_dev->gpu->adev)) + continue; sub_type_hdr = (typeof(sub_type_hdr))( (char *)sub_type_hdr + sizeof(struct crat_subtype_iolink)); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 648f40091aa3..38c19dc8311d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -2048,7 +2048,7 @@ int amdkfd_fence_wait_timeout(struct device_queue_manager *dqm, { unsigned long end_jiffies = msecs_to_jiffies(timeout_ms) + jiffies; struct device *dev = dqm->dev->adev->dev; - uint64_t *fence_addr = dqm->fence_addr; + uint64_t *fence_addr = dqm->fence_addr; while (*fence_addr != fence_value) { /* Fatal err detected, this response won't come */ @@ -2254,6 +2254,7 @@ static int unmap_queues_cpsch(struct device_queue_manager *dqm, goto out; *dqm->fence_addr = KFD_FENCE_INIT; + mb(); pm_send_query_status(&dqm->packet_mgr, dqm->fence_gpu_addr, KFD_FENCE_COMPLETED); /* should be timed out */ @@ -3173,7 +3174,7 @@ struct copy_context_work_handler_workarea { struct kfd_process *p; }; -static void copy_context_work_handler (struct work_struct *work) +static void copy_context_work_handler(struct work_struct *work) { struct copy_context_work_handler_workarea *workarea; struct mqd_manager *mqd_mgr; @@ -3200,6 +3201,9 @@ static void copy_context_work_handler (struct work_struct *work) struct qcm_process_device *qpd = &pdd->qpd; list_for_each_entry(q, &qpd->queues_list, list) { + if (q->properties.type != KFD_QUEUE_TYPE_COMPUTE) + continue; + mqd_mgr = dqm->mqd_mgrs[KFD_MQD_TYPE_CP]; /* We ignore the return value from get_wave_state diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c index 8ee3d07ffbdf..eacfeb32f35d 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c @@ -445,14 +445,13 @@ svm_migrate_vma_to_vram(struct kfd_node *node, struct svm_range *prange, pr_debug("successful/cpages/npages 0x%lx/0x%lx/0x%lx\n", mpages, cpages, migrate.npages); - kfd_smi_event_migration_end(node, p->lead_thread->pid, - start >> PAGE_SHIFT, end >> PAGE_SHIFT, - 0, node->id, trigger); - svm_range_dma_unmap_dev(adev->dev, scratch, 0, npages); out_free: kvfree(buf); + kfd_smi_event_migration_end(node, p->lead_thread->pid, + start >> PAGE_SHIFT, end >> PAGE_SHIFT, + 0, node->id, trigger, r); out: if (!r && mpages) { pdd = svm_range_get_pdd_by_node(prange, node); @@ -751,14 +750,13 @@ svm_migrate_vma_to_ram(struct kfd_node *node, struct svm_range *prange, svm_migrate_copy_done(adev, mfence); migrate_vma_finalize(&migrate); - kfd_smi_event_migration_end(node, p->lead_thread->pid, - start >> PAGE_SHIFT, end >> PAGE_SHIFT, - node->id, 0, trigger); - svm_range_dma_unmap_dev(adev->dev, scratch, 0, npages); out_free: kvfree(buf); + kfd_smi_event_migration_end(node, p->lead_thread->pid, + start >> PAGE_SHIFT, end >> PAGE_SHIFT, + node->id, 0, trigger, r); out: if (!r && cpages) { mpages = cpages - upages; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index d6530febabad..9e5ca0b93b2a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -775,7 +775,7 @@ struct kfd_process_device { enum kfd_pdd_bound bound; /* VRAM usage */ - uint64_t vram_usage; + atomic64_t vram_usage; struct attribute attr_vram; char vram_filename[MAX_SYSFS_FILENAME_LEN]; @@ -1347,7 +1347,6 @@ int pqm_init(struct process_queue_manager *pqm, struct kfd_process *p); void pqm_uninit(struct process_queue_manager *pqm); int pqm_create_queue(struct process_queue_manager *pqm, struct kfd_node *dev, - struct file *f, struct queue_properties *properties, unsigned int *qid, const struct kfd_criu_queue_priv_data *q_data, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index d07acf1b2f93..d4aa843aacfd 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -332,7 +332,7 @@ static ssize_t kfd_procfs_show(struct kobject *kobj, struct attribute *attr, } else if (strncmp(attr->name, "vram_", 5) == 0) { struct kfd_process_device *pdd = container_of(attr, struct kfd_process_device, attr_vram); - return snprintf(buffer, PAGE_SIZE, "%llu\n", READ_ONCE(pdd->vram_usage)); + return snprintf(buffer, PAGE_SIZE, "%llu\n", atomic64_read(&pdd->vram_usage)); } else if (strncmp(attr->name, "sdma_", 5) == 0) { struct kfd_process_device *pdd = container_of(attr, struct kfd_process_device, attr_sdma); @@ -1625,7 +1625,7 @@ struct kfd_process_device *kfd_create_process_device_data(struct kfd_node *dev, pdd->bound = PDD_UNBOUND; pdd->already_dequeued = false; pdd->runtime_inuse = false; - pdd->vram_usage = 0; + atomic64_set(&pdd->vram_usage, 0); pdd->sdma_past_activity_counter = 0; pdd->user_gpu_id = dev->id; atomic64_set(&pdd->evict_duration_counter, 0); @@ -1702,12 +1702,15 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd, ret = amdgpu_amdkfd_gpuvm_acquire_process_vm(dev->adev, avm, &p->kgd_process_info, - &ef); + p->ef ? NULL : &ef); if (ret) { dev_err(dev->adev->dev, "Failed to create process VM object\n"); return ret; } - RCU_INIT_POINTER(p->ef, ef); + + if (!p->ef) + RCU_INIT_POINTER(p->ef, ef); + pdd->drm_priv = drm_file->private_data; ret = kfd_process_device_reserve_ib_mem(pdd); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c index 01b960b15274..c76db22a1000 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c @@ -235,7 +235,7 @@ void pqm_uninit(struct process_queue_manager *pqm) static int init_user_queue(struct process_queue_manager *pqm, struct kfd_node *dev, struct queue **q, struct queue_properties *q_properties, - struct file *f, unsigned int qid) + unsigned int qid) { int retval; @@ -300,7 +300,6 @@ cleanup: int pqm_create_queue(struct process_queue_manager *pqm, struct kfd_node *dev, - struct file *f, struct queue_properties *properties, unsigned int *qid, const struct kfd_criu_queue_priv_data *q_data, @@ -374,7 +373,7 @@ int pqm_create_queue(struct process_queue_manager *pqm, * allocate_sdma_queue() in create_queue() has the * corresponding check logic. */ - retval = init_user_queue(pqm, dev, &q, properties, f, *qid); + retval = init_user_queue(pqm, dev, &q, properties, *qid); if (retval != 0) goto err_create_queue; pqn->q = q; @@ -395,7 +394,7 @@ int pqm_create_queue(struct process_queue_manager *pqm, goto err_create_queue; } - retval = init_user_queue(pqm, dev, &q, properties, f, *qid); + retval = init_user_queue(pqm, dev, &q, properties, *qid); if (retval != 0) goto err_create_queue; pqn->q = q; @@ -1029,8 +1028,7 @@ int kfd_criu_restore_queue(struct kfd_process *p, print_queue_properties(&qp); - ret = pqm_create_queue(&p->pqm, pdd->dev, NULL, &qp, &queue_id, q_data, mqd, ctl_stack, - NULL); + ret = pqm_create_queue(&p->pqm, pdd->dev, &qp, &queue_id, q_data, mqd, ctl_stack, NULL); if (ret) { pr_err("Failed to create new queue err:%d\n", ret); goto exit; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c index de8b9abf7afc..9b8169761ec5 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.c @@ -44,7 +44,7 @@ struct kfd_smi_client { bool suser; }; -#define MAX_KFIFO_SIZE 1024 +#define KFD_MAX_KFIFO_SIZE 8192 static __poll_t kfd_smi_ev_poll(struct file *, struct poll_table_struct *); static ssize_t kfd_smi_ev_read(struct file *, char __user *, size_t, loff_t *); @@ -86,7 +86,7 @@ static ssize_t kfd_smi_ev_read(struct file *filep, char __user *user, struct kfd_smi_client *client = filep->private_data; unsigned char *buf; - size = min_t(size_t, size, MAX_KFIFO_SIZE); + size = min_t(size_t, size, KFD_MAX_KFIFO_SIZE); buf = kmalloc(size, GFP_KERNEL); if (!buf) return -ENOMEM; @@ -292,12 +292,13 @@ void kfd_smi_event_migration_start(struct kfd_node *node, pid_t pid, void kfd_smi_event_migration_end(struct kfd_node *node, pid_t pid, unsigned long start, unsigned long end, - uint32_t from, uint32_t to, uint32_t trigger) + uint32_t from, uint32_t to, uint32_t trigger, + int error_code) { kfd_smi_event_add(pid, node, KFD_SMI_EVENT_MIGRATE_END, KFD_EVENT_FMT_MIGRATE_END( ktime_get_boottime_ns(), pid, start, end - start, - from, to, trigger)); + from, to, trigger, error_code)); } void kfd_smi_event_queue_eviction(struct kfd_node *node, pid_t pid, @@ -354,7 +355,7 @@ int kfd_smi_event_open(struct kfd_node *dev, uint32_t *fd) return -ENOMEM; INIT_LIST_HEAD(&client->list); - ret = kfifo_alloc(&client->fifo, MAX_KFIFO_SIZE, GFP_KERNEL); + ret = kfifo_alloc(&client->fifo, KFD_MAX_KFIFO_SIZE, GFP_KERNEL); if (ret) { kfree(client); return ret; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h index 85010b8307f8..503bff13d815 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_smi_events.h @@ -44,7 +44,8 @@ void kfd_smi_event_migration_start(struct kfd_node *node, pid_t pid, uint32_t trigger); void kfd_smi_event_migration_end(struct kfd_node *node, pid_t pid, unsigned long start, unsigned long end, - uint32_t from, uint32_t to, uint32_t trigger); + uint32_t from, uint32_t to, uint32_t trigger, + int error_code); void kfd_smi_event_queue_eviction(struct kfd_node *node, pid_t pid, uint32_t trigger); void kfd_smi_event_queue_restore(struct kfd_node *node, pid_t pid); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index 04e746923697..3e2911895c74 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -405,6 +405,27 @@ static void svm_range_bo_release(struct kref *kref) spin_lock(&svm_bo->list_lock); } spin_unlock(&svm_bo->list_lock); + + if (mmget_not_zero(svm_bo->eviction_fence->mm)) { + struct kfd_process_device *pdd; + struct kfd_process *p; + struct mm_struct *mm; + + mm = svm_bo->eviction_fence->mm; + /* + * The forked child process takes svm_bo device pages ref, svm_bo could be + * released after parent process is gone. + */ + p = kfd_lookup_process_by_mm(mm); + if (p) { + pdd = kfd_get_process_device_data(svm_bo->node, p); + if (pdd) + atomic64_sub(amdgpu_bo_size(svm_bo->bo), &pdd->vram_usage); + kfd_unref_process(p); + } + mmput(mm); + } + if (!dma_fence_is_signaled(&svm_bo->eviction_fence->base)) /* We're not in the eviction worker. Signal the fence. */ dma_fence_signal(&svm_bo->eviction_fence->base); @@ -532,6 +553,7 @@ int svm_range_vram_node_new(struct kfd_node *node, struct svm_range *prange, bool clear) { + struct kfd_process_device *pdd; struct amdgpu_bo_param bp; struct svm_range_bo *svm_bo; struct amdgpu_bo_user *ubo; @@ -623,6 +645,10 @@ svm_range_vram_node_new(struct kfd_node *node, struct svm_range *prange, list_add(&prange->svm_bo_list, &svm_bo->range_list); spin_unlock(&svm_bo->list_lock); + pdd = svm_range_get_pdd_by_node(prange, node); + if (pdd) + atomic64_add(amdgpu_bo_size(bo), &pdd->vram_usage); + return 0; reserve_bo_failed: @@ -3085,8 +3111,6 @@ retry_write_locked: start = max_t(unsigned long, ALIGN_DOWN(addr, size), prange->start); last = min_t(unsigned long, ALIGN(addr + 1, size) - 1, prange->last); if (prange->actual_loc != 0 || best_loc != 0) { - migration = true; - if (best_loc) { r = svm_migrate_to_vram(prange, best_loc, start, last, mm, KFD_MIGRATE_TRIGGER_PAGEFAULT_GPU); @@ -3109,7 +3133,9 @@ retry_write_locked: if (r) { pr_debug("failed %d to migrate svms %p [0x%lx 0x%lx]\n", r, svms, start, last); - goto out_unlock_range; + goto out_migrate_fail; + } else { + migration = true; } } @@ -3119,6 +3145,7 @@ retry_write_locked: pr_debug("failed %d to map svms 0x%p [0x%lx 0x%lx] to gpus\n", r, svms, start, last); +out_migrate_fail: kfd_smi_event_page_fault_end(node, p->lead_thread->pid, addr, migration); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 6e79028c5d78..75d6b90104f8 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -320,18 +320,18 @@ static bool dm_is_idle(void *handle) return true; } -static int dm_wait_for_idle(void *handle) +static int dm_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* XXX todo */ return 0; } -static bool dm_check_soft_reset(void *handle) +static bool dm_check_soft_reset(struct amdgpu_ip_block *ip_block) { return false; } -static int dm_soft_reset(void *handle) +static int dm_soft_reset(struct amdgpu_ip_block *ip_block) { /* XXX todo */ return 0; @@ -770,6 +770,12 @@ static void dmub_hpd_callback(struct amdgpu_device *adev, return; } + /* Skip DMUB HPD IRQ in suspend/resume. We will probe them later. */ + if (notify->type == DMUB_NOTIFICATION_HPD && adev->in_suspend) { + DRM_INFO("Skip DMUB HPD IRQ callback in suspend/resume\n"); + return; + } + link_index = notify->link_index; link = adev->dm.dc->links[link_index]; dev = adev->dm.ddev; @@ -962,7 +968,7 @@ static int dm_set_powergating_state(void *handle, } /* Prototypes of private functions */ -static int dm_early_init(void *handle); +static int dm_early_init(struct amdgpu_ip_block *ip_block); /* Allocate memory for FBC compressed data */ static void amdgpu_dm_fbc_init(struct drm_connector *connector) @@ -1690,6 +1696,26 @@ dm_allocate_gpu_mem( return da->cpu_ptr; } +void +dm_free_gpu_mem( + struct amdgpu_device *adev, + enum dc_gpu_mem_alloc_type type, + void *pvMem) +{ + struct dal_allocation *da; + + /* walk the da list in DM */ + list_for_each_entry(da, &adev->dm.da_list, list) { + if (pvMem == da->cpu_ptr) { + amdgpu_bo_free_kernel(&da->bo, &da->gpu_addr, &da->cpu_ptr); + list_del(&da->list); + kfree(da); + break; + } + } + +} + static enum dmub_status dm_dmub_send_vbios_gpint_command(struct amdgpu_device *adev, enum dmub_gpint_command command_code, @@ -1756,16 +1782,20 @@ static struct dml2_soc_bb *dm_dmub_get_vbios_bounding_box(struct amdgpu_device * /* Send the chunk */ ret = dm_dmub_send_vbios_gpint_command(adev, send_addrs[i], chunk, 30000); if (ret != DMUB_STATUS_OK) - /* No need to free bb here since it shall be done in dm_sw_fini() */ - return NULL; + goto free_bb; } /* Now ask DMUB to copy the bb */ ret = dm_dmub_send_vbios_gpint_command(adev, DMUB_GPINT__BB_COPY, 1, 200000); if (ret != DMUB_STATUS_OK) - return NULL; + goto free_bb; return bb; + +free_bb: + dm_free_gpu_mem(adev, DC_MEM_ALLOC_TYPE_GART, (void *) bb); + return NULL; + } static enum dmub_ips_disable_type dm_get_default_ips_mode( @@ -1880,7 +1910,11 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) else init_data.flags.gpu_vm_support = (amdgpu_sg_display != 0); } else { - init_data.flags.gpu_vm_support = (amdgpu_sg_display != 0) && (adev->flags & AMD_IS_APU); + if (amdgpu_ip_version(adev, DCE_HWIP, 0) == IP_VERSION(2, 0, 3)) + init_data.flags.gpu_vm_support = (amdgpu_sg_display == 1); + else + init_data.flags.gpu_vm_support = + (amdgpu_sg_display != 0) && (adev->flags & AMD_IS_APU); } adev->mode_info.gpu_vm_support = init_data.flags.gpu_vm_support; @@ -2026,7 +2060,8 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) DRM_ERROR("amdgpu: failed to initialize vblank_workqueue.\n"); } - if (adev->dm.dc->caps.ips_support && adev->dm.dc->config.disable_ips == DMUB_IPS_ENABLE) + if (adev->dm.dc->caps.ips_support && + adev->dm.dc->config.disable_ips != DMUB_IPS_DISABLE_ALL) adev->dm.idle_workqueue = idle_create_workqueue(adev); if (adev->dm.dc->caps.max_links > 0 && adev->family >= AMDGPU_FAMILY_RV) { @@ -2108,9 +2143,9 @@ error: return -EINVAL; } -static int amdgpu_dm_early_fini(void *handle) +static int amdgpu_dm_early_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_dm_audio_fini(adev); @@ -2502,9 +2537,9 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev) return 0; } -static int dm_sw_init(void *handle) +static int dm_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; adev->dm.cgs_device = amdgpu_cgs_create_device(adev); @@ -2524,9 +2559,9 @@ static int dm_sw_init(void *handle) return load_dmcu_fw(adev); } -static int dm_sw_fini(void *handle) +static int dm_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct dal_allocation *da; list_for_each_entry(da, &adev->dm.da_list, list) { @@ -2534,11 +2569,11 @@ static int dm_sw_fini(void *handle) amdgpu_bo_free_kernel(&da->bo, &da->gpu_addr, &da->cpu_ptr); list_del(&da->list); kfree(da); + adev->dm.bb_from_dmub = NULL; break; } } - adev->dm.bb_from_dmub = NULL; kfree(adev->dm.dmub_fb_info); adev->dm.dmub_fb_info = NULL; @@ -2591,9 +2626,9 @@ static int detect_mst_link_for_all_connectors(struct drm_device *dev) return ret; } -static int dm_late_init(void *handle) +static int dm_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct dmcu_iram_parameters params; unsigned int linear_lut[16]; @@ -2783,7 +2818,7 @@ static int amdgpu_dm_smu_write_watermarks_table(struct amdgpu_device *adev) /** * dm_hw_init() - Initialize DC device - * @handle: The base driver device containing the amdgpu_dm device. + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Initialize the &struct amdgpu_display_manager device. This involves calling * the initializers of each DM component, then populating the struct with them. @@ -2801,9 +2836,9 @@ static int amdgpu_dm_smu_write_watermarks_table(struct amdgpu_device *adev) * - Vblank support * - Debug FS entries, if enabled */ -static int dm_hw_init(void *handle) +static int dm_hw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; int r; /* Create DAL display manager */ @@ -2817,15 +2852,15 @@ static int dm_hw_init(void *handle) /** * dm_hw_fini() - Teardown DC device - * @handle: The base driver device containing the amdgpu_dm device. + * @ip_block: Pointer to the amdgpu_ip_block for this hw instance. * * Teardown components within &struct amdgpu_display_manager that require * cleanup. This involves cleaning up the DRM device, DC, and any modules that * were loaded. Also flush IRQ workqueues and disable them. */ -static int dm_hw_fini(void *handle) +static int dm_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; amdgpu_dm_hpd_fini(adev); @@ -2929,9 +2964,9 @@ static void hpd_rx_irq_work_suspend(struct amdgpu_display_manager *dm) } } -static int dm_suspend(void *handle) +static int dm_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_display_manager *dm = &adev->dm; int ret = 0; @@ -2965,10 +3000,11 @@ static int dm_suspend(void *handle) hpd_rx_irq_work_suspend(dm); - if (adev->dm.dc->caps.ips_support) - dc_allow_idle_optimizations(adev->dm.dc, true); - dc_set_power_state(dm->dc, DC_ACPI_CM_POWER_STATE_D3); + + if (dm->dc->caps.ips_support && adev->in_s0ix) + dc_allow_idle_optimizations(dm->dc, true); + dc_dmub_srv_set_power_state(dm->dc->ctx->dmub_srv, DC_ACPI_CM_POWER_STATE_D3); return 0; @@ -3117,9 +3153,9 @@ cleanup: kfree(bundle); } -static int dm_resume(void *handle) +static int dm_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = handle; + struct amdgpu_device *adev = ip_block->adev; struct drm_device *ddev = adev_to_drm(adev); struct amdgpu_display_manager *dm = &adev->dm; struct amdgpu_dm_connector *aconnector; @@ -3371,8 +3407,6 @@ static const struct amd_ip_funcs amdgpu_dm_funcs = { .soft_reset = dm_soft_reset, .set_clockgating_state = dm_set_clockgating_state, .set_powergating_state = dm_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; const struct amdgpu_ip_block_version dm_ip_block = { @@ -3487,7 +3521,7 @@ void amdgpu_dm_update_connector_after_detect( aconnector->dc_sink = sink; dc_sink_retain(aconnector->dc_sink); amdgpu_dm_update_freesync_caps(connector, - aconnector->edid); + aconnector->drm_edid); } else { amdgpu_dm_update_freesync_caps(connector, NULL); if (!aconnector->dc_sink) { @@ -3546,18 +3580,19 @@ void amdgpu_dm_update_connector_after_detect( aconnector->dc_sink = sink; dc_sink_retain(aconnector->dc_sink); if (sink->dc_edid.length == 0) { - aconnector->edid = NULL; + aconnector->drm_edid = NULL; if (aconnector->dc_link->aux_mode) { - drm_dp_cec_unset_edid( - &aconnector->dm_dp_aux.aux); + drm_dp_cec_unset_edid(&aconnector->dm_dp_aux.aux); } } else { - aconnector->edid = - (struct edid *)sink->dc_edid.raw_edid; + const struct edid *edid = (const struct edid *)sink->dc_edid.raw_edid; + + aconnector->drm_edid = drm_edid_alloc(edid, sink->dc_edid.length); + drm_edid_connector_update(connector, aconnector->drm_edid); if (aconnector->dc_link->aux_mode) - drm_dp_cec_set_edid(&aconnector->dm_dp_aux.aux, - aconnector->edid); + drm_dp_cec_attach(&aconnector->dm_dp_aux.aux, + connector->display_info.source_physical_address); } if (!aconnector->timing_requested) { @@ -3568,17 +3603,16 @@ void amdgpu_dm_update_connector_after_detect( "failed to create aconnector->requested_timing\n"); } - drm_connector_update_edid_property(connector, aconnector->edid); - amdgpu_dm_update_freesync_caps(connector, aconnector->edid); + amdgpu_dm_update_freesync_caps(connector, aconnector->drm_edid); update_connector_ext_caps(aconnector); } else { drm_dp_cec_unset_edid(&aconnector->dm_dp_aux.aux); amdgpu_dm_update_freesync_caps(connector, NULL); - drm_connector_update_edid_property(connector, NULL); aconnector->num_modes = 0; dc_sink_release(aconnector->dc_sink); aconnector->dc_sink = NULL; - aconnector->edid = NULL; + drm_edid_free(aconnector->drm_edid); + aconnector->drm_edid = NULL; kfree(aconnector->timing_requested); aconnector->timing_requested = NULL; /* Set CP to DESIRED if it was ENABLED, so we can re-enable it again on hotplug */ @@ -5169,15 +5203,20 @@ static ssize_t s3_debug_store(struct device *device, int s3_state; struct drm_device *drm_dev = dev_get_drvdata(device); struct amdgpu_device *adev = drm_to_adev(drm_dev); + struct amdgpu_ip_block *ip_block; + + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_DCE); + if (!ip_block) + return -EINVAL; ret = kstrtoint(buf, 0, &s3_state); if (ret == 0) { if (s3_state) { - dm_resume(adev); + dm_resume(ip_block); drm_kms_helper_hotplug_event(adev_to_drm(adev)); } else - dm_suspend(adev); + dm_suspend(ip_block); } return ret == 0 ? count : 0; @@ -5249,9 +5288,9 @@ static int dm_init_microcode(struct amdgpu_device *adev) return r; } -static int dm_early_init(void *handle) +static int dm_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct amdgpu_mode_info *mode_info = &adev->mode_info; struct atom_context *ctx = mode_info->atom_context; int index = GetIndexIntoMasterTable(DATA, Object_Header); @@ -6735,12 +6774,21 @@ create_stream_for_sink(struct drm_connector *connector, if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT || stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST || stream->signal == SIGNAL_TYPE_EDP) { + const struct dc_edid_caps *edid_caps; + unsigned int disable_colorimetry = 0; + + if (aconnector->dc_sink) { + edid_caps = &aconnector->dc_sink->edid_caps; + disable_colorimetry = edid_caps->panel_patch.disable_colorimetry; + } + // // should decide stream support vsc sdp colorimetry capability // before building vsc info packet // stream->use_vsc_sdp_for_colorimetry = stream->link->dpcd_caps.dpcd_rev.raw >= 0x14 && - stream->link->dpcd_caps.dprx_feature.bits.VSC_SDP_COLORIMETRY_SUPPORTED; + stream->link->dpcd_caps.dprx_feature.bits.VSC_SDP_COLORIMETRY_SUPPORTED && + !disable_colorimetry; if (stream->out_transfer_func.tf == TRANSFER_FUNCTION_GAMMA22) tf = TRANSFER_FUNC_GAMMA_22; @@ -7105,32 +7153,24 @@ static void amdgpu_dm_connector_funcs_force(struct drm_connector *connector) struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); struct dc_link *dc_link = aconnector->dc_link; struct dc_sink *dc_em_sink = aconnector->dc_em_sink; - struct edid *edid; - struct i2c_adapter *ddc; + const struct drm_edid *drm_edid; - if (dc_link && dc_link->aux_mode) - ddc = &aconnector->dm_dp_aux.aux.ddc; - else - ddc = &aconnector->i2c->base; - - /* - * Note: drm_get_edid gets edid in the following order: - * 1) override EDID if set via edid_override debugfs, - * 2) firmware EDID if set via edid_firmware module parameter - * 3) regular DDC read. - */ - edid = drm_get_edid(connector, ddc); - if (!edid) { + drm_edid = drm_edid_read(connector); + drm_edid_connector_update(connector, drm_edid); + if (!drm_edid) { DRM_ERROR("No EDID found on connector: %s.\n", connector->name); return; } - aconnector->edid = edid; - + aconnector->drm_edid = drm_edid; /* Update emulated (virtual) sink's EDID */ if (dc_em_sink && dc_link) { + // FIXME: Get rid of drm_edid_raw() + const struct edid *edid = drm_edid_raw(drm_edid); + memset(&dc_em_sink->edid_caps, 0, sizeof(struct dc_edid_caps)); - memmove(dc_em_sink->dc_edid.raw_edid, edid, (edid->extensions + 1) * EDID_LENGTH); + memmove(dc_em_sink->dc_edid.raw_edid, edid, + (edid->extensions + 1) * EDID_LENGTH); dm_helpers_parse_edid_caps( dc_link, &dc_em_sink->dc_edid, @@ -7160,36 +7200,26 @@ static int get_modes(struct drm_connector *connector) static void create_eml_sink(struct amdgpu_dm_connector *aconnector) { struct drm_connector *connector = &aconnector->base; - struct dc_link *dc_link = aconnector->dc_link; struct dc_sink_init_data init_params = { .link = aconnector->dc_link, .sink_signal = SIGNAL_TYPE_VIRTUAL }; - struct edid *edid; - struct i2c_adapter *ddc; + const struct drm_edid *drm_edid; + const struct edid *edid; - if (dc_link->aux_mode) - ddc = &aconnector->dm_dp_aux.aux.ddc; - else - ddc = &aconnector->i2c->base; - - /* - * Note: drm_get_edid gets edid in the following order: - * 1) override EDID if set via edid_override debugfs, - * 2) firmware EDID if set via edid_firmware module parameter - * 3) regular DDC read. - */ - edid = drm_get_edid(connector, ddc); - if (!edid) { + drm_edid = drm_edid_read(connector); + drm_edid_connector_update(connector, drm_edid); + if (!drm_edid) { DRM_ERROR("No EDID found on connector: %s.\n", connector->name); return; } - if (drm_detect_hdmi_monitor(edid)) + if (connector->display_info.is_hdmi) init_params.sink_signal = SIGNAL_TYPE_HDMI_TYPE_A; - aconnector->edid = edid; + aconnector->drm_edid = drm_edid; + edid = drm_edid_raw(drm_edid); // FIXME: Get rid of drm_edid_raw() aconnector->dc_em_sink = dc_link_add_remote_sink( aconnector->dc_link, (uint8_t *)edid, @@ -7876,16 +7906,16 @@ static void amdgpu_set_panel_orientation(struct drm_connector *connector) } static void amdgpu_dm_connector_ddc_get_modes(struct drm_connector *connector, - struct edid *edid) + const struct drm_edid *drm_edid) { struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); - if (edid) { + if (drm_edid) { /* empty probed_modes */ INIT_LIST_HEAD(&connector->probed_modes); amdgpu_dm_connector->num_modes = - drm_add_edid_modes(connector, edid); + drm_edid_connector_add_modes(connector); /* sorting the probed modes before calling function * amdgpu_dm_get_native_mode() since EDID can have @@ -7899,10 +7929,10 @@ static void amdgpu_dm_connector_ddc_get_modes(struct drm_connector *connector, amdgpu_dm_get_native_mode(connector); /* Freesync capabilities are reset by calling - * drm_add_edid_modes() and need to be + * drm_edid_connector_add_modes() and need to be * restored here. */ - amdgpu_dm_update_freesync_caps(connector, edid); + amdgpu_dm_update_freesync_caps(connector, drm_edid); } else { amdgpu_dm_connector->num_modes = 0; } @@ -7998,12 +8028,12 @@ static uint add_fs_modes(struct amdgpu_dm_connector *aconnector) } static void amdgpu_dm_connector_add_freesync_modes(struct drm_connector *connector, - struct edid *edid) + const struct drm_edid *drm_edid) { struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); - if (!(amdgpu_freesync_vid_mode && edid)) + if (!(amdgpu_freesync_vid_mode && drm_edid)) return; if (amdgpu_dm_connector->max_vfreq - amdgpu_dm_connector->min_vfreq > 10) @@ -8016,24 +8046,24 @@ static int amdgpu_dm_connector_get_modes(struct drm_connector *connector) struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); struct drm_encoder *encoder; - struct edid *edid = amdgpu_dm_connector->edid; + const struct drm_edid *drm_edid = amdgpu_dm_connector->drm_edid; struct dc_link_settings *verified_link_cap = &amdgpu_dm_connector->dc_link->verified_link_cap; const struct dc *dc = amdgpu_dm_connector->dc_link->dc; encoder = amdgpu_dm_connector_to_encoder(connector); - if (!drm_edid_is_valid(edid)) { + if (!drm_edid) { amdgpu_dm_connector->num_modes = drm_add_modes_noedid(connector, 640, 480); if (dc->link_srv->dp_get_encoding_format(verified_link_cap) == DP_128b_132b_ENCODING) amdgpu_dm_connector->num_modes += drm_add_modes_noedid(connector, 1920, 1080); } else { - amdgpu_dm_connector_ddc_get_modes(connector, edid); + amdgpu_dm_connector_ddc_get_modes(connector, drm_edid); if (encoder) amdgpu_dm_connector_add_common_modes(encoder, connector); - amdgpu_dm_connector_add_freesync_modes(connector, edid); + amdgpu_dm_connector_add_freesync_modes(connector, drm_edid); } amdgpu_dm_fbc_init(connector); @@ -8357,7 +8387,8 @@ static void manage_dm_interrupts(struct amdgpu_device *adev, if (amdgpu_ip_version(adev, DCE_HWIP, 0) < IP_VERSION(3, 5, 0) || acrtc_state->stream->link->psr_settings.psr_version < - DC_PSR_VERSION_UNSUPPORTED) { + DC_PSR_VERSION_UNSUPPORTED || + !(adev->flags & AMD_IS_APU)) { timing = &acrtc_state->stream->timing; /* at least 2 frames */ @@ -9557,7 +9588,7 @@ static void amdgpu_dm_commit_streams(struct drm_atomic_state *state, WARN_ON(!dc_commit_streams(dm->dc, ¶ms)); /* Allow idle optimization when vblank count is 0 for display off */ - if (dm->active_vblank_irq_count == 0) + if ((dm->active_vblank_irq_count == 0) && amdgpu_dm_is_headless(dm->adev)) dc_allow_idle_optimizations(dm->dc, true); mutex_unlock(&dm->dc_lock); @@ -10088,6 +10119,8 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) for (i = 0; i < crtc_disable_count; i++) pm_runtime_put_autosuspend(dev->dev); pm_runtime_mark_last_busy(dev->dev); + + trace_amdgpu_dm_atomic_commit_tail_finish(state); } static int dm_force_atomic_commit(struct drm_connector *connector) @@ -11988,7 +12021,7 @@ static bool parse_edid_cea(struct amdgpu_dm_connector *aconnector, } static void parse_edid_displayid_vrr(struct drm_connector *connector, - struct edid *edid) + const struct edid *edid) { u8 *edid_ext = NULL; int i; @@ -12031,7 +12064,7 @@ static void parse_edid_displayid_vrr(struct drm_connector *connector, } static int parse_amd_vsdb(struct amdgpu_dm_connector *aconnector, - struct edid *edid, struct amdgpu_hdmi_vsdb_info *vsdb_info) + const struct edid *edid, struct amdgpu_hdmi_vsdb_info *vsdb_info) { u8 *edid_ext = NULL; int i; @@ -12066,7 +12099,8 @@ static int parse_amd_vsdb(struct amdgpu_dm_connector *aconnector, } static int parse_hdmi_amd_vsdb(struct amdgpu_dm_connector *aconnector, - struct edid *edid, struct amdgpu_hdmi_vsdb_info *vsdb_info) + const struct edid *edid, + struct amdgpu_hdmi_vsdb_info *vsdb_info) { u8 *edid_ext = NULL; int i; @@ -12100,7 +12134,7 @@ static int parse_hdmi_amd_vsdb(struct amdgpu_dm_connector *aconnector, * amdgpu_dm_update_freesync_caps - Update Freesync capabilities * * @connector: Connector to query. - * @edid: EDID from monitor + * @drm_edid: DRM EDID from monitor * * Amdgpu supports Freesync in DP and HDMI displays, and it is required to keep * track of some of the display information in the internal data struct used by @@ -12108,19 +12142,16 @@ static int parse_hdmi_amd_vsdb(struct amdgpu_dm_connector *aconnector, * FreeSync parameters. */ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, - struct edid *edid) + const struct drm_edid *drm_edid) { int i = 0; - struct detailed_timing *timing; - struct detailed_non_pixel *data; - struct detailed_data_monitor_range *range; struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); struct dm_connector_state *dm_con_state = NULL; struct dc_sink *sink; - struct amdgpu_device *adev = drm_to_adev(connector->dev); struct amdgpu_hdmi_vsdb_info vsdb_info = {0}; + const struct edid *edid; bool freesync_capable = false; enum adaptive_sync_type as_type = ADAPTIVE_SYNC_TYPE_NONE; @@ -12133,13 +12164,13 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, amdgpu_dm_connector->dc_sink : amdgpu_dm_connector->dc_em_sink; - if (!edid || !sink) { + drm_edid_connector_update(connector, drm_edid); + + if (!drm_edid || !sink) { dm_con_state = to_dm_connector_state(connector->state); amdgpu_dm_connector->min_vfreq = 0; amdgpu_dm_connector->max_vfreq = 0; - connector->display_info.monitor_range.min_vfreq = 0; - connector->display_info.monitor_range.max_vfreq = 0; freesync_capable = false; goto update; @@ -12150,6 +12181,8 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, if (!adev->dm.freesync_module) goto update; + edid = drm_edid_raw(drm_edid); // FIXME: Get rid of drm_edid_raw() + /* Some eDP panels only have the refresh rate range info in DisplayID */ if ((connector->display_info.monitor_range.min_vfreq == 0 || connector->display_info.monitor_range.max_vfreq == 0)) @@ -12157,67 +12190,10 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, if (edid && (sink->sink_signal == SIGNAL_TYPE_DISPLAY_PORT || sink->sink_signal == SIGNAL_TYPE_EDP)) { - bool edid_check_required = false; - - if (amdgpu_dm_connector->dc_link && - amdgpu_dm_connector->dc_link->dpcd_caps.allow_invalid_MSA_timing_param) { - if (edid->features & DRM_EDID_FEATURE_CONTINUOUS_FREQ) { - amdgpu_dm_connector->min_vfreq = connector->display_info.monitor_range.min_vfreq; - amdgpu_dm_connector->max_vfreq = connector->display_info.monitor_range.max_vfreq; - if (amdgpu_dm_connector->max_vfreq - - amdgpu_dm_connector->min_vfreq > 10) - freesync_capable = true; - } else { - edid_check_required = edid->version > 1 || - (edid->version == 1 && - edid->revision > 1); - } - } - - if (edid_check_required) { - for (i = 0; i < 4; i++) { - - timing = &edid->detailed_timings[i]; - data = &timing->data.other_data; - range = &data->data.range; - /* - * Check if monitor has continuous frequency mode - */ - if (data->type != EDID_DETAIL_MONITOR_RANGE) - continue; - /* - * Check for flag range limits only. If flag == 1 then - * no additional timing information provided. - * Default GTF, GTF Secondary curve and CVT are not - * supported - */ - if (range->flags != 1) - continue; - - connector->display_info.monitor_range.min_vfreq = range->min_vfreq; - connector->display_info.monitor_range.max_vfreq = range->max_vfreq; - - if (edid->revision >= 4) { - if (data->pad2 & DRM_EDID_RANGE_OFFSET_MIN_VFREQ) - connector->display_info.monitor_range.min_vfreq += 255; - if (data->pad2 & DRM_EDID_RANGE_OFFSET_MAX_VFREQ) - connector->display_info.monitor_range.max_vfreq += 255; - } - - amdgpu_dm_connector->min_vfreq = - connector->display_info.monitor_range.min_vfreq; - amdgpu_dm_connector->max_vfreq = - connector->display_info.monitor_range.max_vfreq; - - break; - } - - if (amdgpu_dm_connector->max_vfreq - - amdgpu_dm_connector->min_vfreq > 10) { - - freesync_capable = true; - } - } + amdgpu_dm_connector->min_vfreq = connector->display_info.monitor_range.min_vfreq; + amdgpu_dm_connector->max_vfreq = connector->display_info.monitor_range.max_vfreq; + if (amdgpu_dm_connector->max_vfreq - amdgpu_dm_connector->min_vfreq > 10) + freesync_capable = true; parse_amd_vsdb(amdgpu_dm_connector, edid, &vsdb_info); if (vsdb_info.replay_mode) { @@ -12226,12 +12202,9 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, amdgpu_dm_connector->as_type = ADAPTIVE_SYNC_TYPE_EDP; } - } else if (edid && sink->sink_signal == SIGNAL_TYPE_HDMI_TYPE_A) { + } else if (drm_edid && sink->sink_signal == SIGNAL_TYPE_HDMI_TYPE_A) { i = parse_hdmi_amd_vsdb(amdgpu_dm_connector, edid, &vsdb_info); if (i >= 0 && vsdb_info.freesync_supported) { - timing = &edid->detailed_timings[i]; - data = &timing->data.other_data; - amdgpu_dm_connector->min_vfreq = vsdb_info.min_refresh_rate_hz; amdgpu_dm_connector->max_vfreq = vsdb_info.max_refresh_rate_hz; if (amdgpu_dm_connector->max_vfreq - amdgpu_dm_connector->min_vfreq > 10) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 15d4690c74d6..25e95775c45c 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -673,7 +673,7 @@ struct amdgpu_dm_connector { /* we need to mind the EDID between detect and get modes due to analog/digital/tvencoder */ - struct edid *edid; + const struct drm_edid *drm_edid; /* shared with amdgpu */ struct amdgpu_hpd hpd; @@ -951,7 +951,7 @@ void dm_restore_drm_connector_state(struct drm_device *dev, struct drm_connector *connector); void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, - struct edid *edid); + const struct drm_edid *drm_edid); void amdgpu_dm_trigger_timing_sync(struct drm_device *dev); @@ -1004,6 +1004,9 @@ void *dm_allocate_gpu_mem(struct amdgpu_device *adev, enum dc_gpu_mem_alloc_type type, size_t size, long long *addr); +void dm_free_gpu_mem(struct amdgpu_device *adev, + enum dc_gpu_mem_alloc_type type, + void *addr); bool amdgpu_dm_is_headless(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c index a2cf2c066a76..ffa4d3965b4b 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crtc.c @@ -36,7 +36,7 @@ #include "amdgpu_dm_debugfs.h" #define HPD_DETECTION_PERIOD_uS 5000000 -#define HPD_DETECTION_TIME_uS 1000 +#define HPD_DETECTION_TIME_uS 100000 void amdgpu_dm_crtc_handle_vblank(struct amdgpu_crtc *acrtc) { @@ -154,6 +154,7 @@ static void amdgpu_dm_crtc_set_panel_sr_feature( amdgpu_dm_psr_enable(vblank_work->stream); if (dm->idle_workqueue && + (dm->dc->config.disable_ips == DMUB_IPS_ENABLE) && dm->dc->idle_optimizations_allowed && dm->idle_workqueue->enable && !dm->idle_workqueue->running) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index db56b0aa5454..6a97bb2d9160 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -1529,7 +1529,6 @@ static ssize_t dp_dsc_clock_en_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { char *rd_buf = NULL; - char *rd_buf_ptr = NULL; struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private; struct display_stream_compressor *dsc; struct dcn_dsc_state dsc_state = {0}; @@ -1543,8 +1542,6 @@ static ssize_t dp_dsc_clock_en_read(struct file *f, char __user *buf, if (!rd_buf) return -ENOMEM; - rd_buf_ptr = rd_buf; - for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream && @@ -1558,10 +1555,9 @@ static ssize_t dp_dsc_clock_en_read(struct file *f, char __user *buf, if (dsc) dsc->funcs->dsc_read_state(dsc, &dsc_state); - snprintf(rd_buf_ptr, str_len, + snprintf(rd_buf, str_len, "%d\n", dsc_state.dsc_clock_en); - rd_buf_ptr += str_len; while (size) { if (*pos >= rd_buf_size) @@ -1719,7 +1715,6 @@ static ssize_t dp_dsc_slice_width_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { char *rd_buf = NULL; - char *rd_buf_ptr = NULL; struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private; struct display_stream_compressor *dsc; struct dcn_dsc_state dsc_state = {0}; @@ -1733,8 +1728,6 @@ static ssize_t dp_dsc_slice_width_read(struct file *f, char __user *buf, if (!rd_buf) return -ENOMEM; - rd_buf_ptr = rd_buf; - for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream && @@ -1748,10 +1741,9 @@ static ssize_t dp_dsc_slice_width_read(struct file *f, char __user *buf, if (dsc) dsc->funcs->dsc_read_state(dsc, &dsc_state); - snprintf(rd_buf_ptr, str_len, + snprintf(rd_buf, str_len, "%d\n", dsc_state.dsc_slice_width); - rd_buf_ptr += str_len; while (size) { if (*pos >= rd_buf_size) @@ -1907,7 +1899,6 @@ static ssize_t dp_dsc_slice_height_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { char *rd_buf = NULL; - char *rd_buf_ptr = NULL; struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private; struct display_stream_compressor *dsc; struct dcn_dsc_state dsc_state = {0}; @@ -1921,8 +1912,6 @@ static ssize_t dp_dsc_slice_height_read(struct file *f, char __user *buf, if (!rd_buf) return -ENOMEM; - rd_buf_ptr = rd_buf; - for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream && @@ -1936,10 +1925,9 @@ static ssize_t dp_dsc_slice_height_read(struct file *f, char __user *buf, if (dsc) dsc->funcs->dsc_read_state(dsc, &dsc_state); - snprintf(rd_buf_ptr, str_len, + snprintf(rd_buf, str_len, "%d\n", dsc_state.dsc_slice_height); - rd_buf_ptr += str_len; while (size) { if (*pos >= rd_buf_size) @@ -2091,7 +2079,6 @@ static ssize_t dp_dsc_bits_per_pixel_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { char *rd_buf = NULL; - char *rd_buf_ptr = NULL; struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private; struct display_stream_compressor *dsc; struct dcn_dsc_state dsc_state = {0}; @@ -2105,8 +2092,6 @@ static ssize_t dp_dsc_bits_per_pixel_read(struct file *f, char __user *buf, if (!rd_buf) return -ENOMEM; - rd_buf_ptr = rd_buf; - for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream && @@ -2120,10 +2105,9 @@ static ssize_t dp_dsc_bits_per_pixel_read(struct file *f, char __user *buf, if (dsc) dsc->funcs->dsc_read_state(dsc, &dsc_state); - snprintf(rd_buf_ptr, str_len, + snprintf(rd_buf, str_len, "%d\n", dsc_state.dsc_bits_per_pixel); - rd_buf_ptr += str_len; while (size) { if (*pos >= rd_buf_size) @@ -2270,7 +2254,6 @@ static ssize_t dp_dsc_pic_width_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { char *rd_buf = NULL; - char *rd_buf_ptr = NULL; struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private; struct display_stream_compressor *dsc; struct dcn_dsc_state dsc_state = {0}; @@ -2284,8 +2267,6 @@ static ssize_t dp_dsc_pic_width_read(struct file *f, char __user *buf, if (!rd_buf) return -ENOMEM; - rd_buf_ptr = rd_buf; - for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream && @@ -2299,10 +2280,9 @@ static ssize_t dp_dsc_pic_width_read(struct file *f, char __user *buf, if (dsc) dsc->funcs->dsc_read_state(dsc, &dsc_state); - snprintf(rd_buf_ptr, str_len, + snprintf(rd_buf, str_len, "%d\n", dsc_state.dsc_pic_width); - rd_buf_ptr += str_len; while (size) { if (*pos >= rd_buf_size) @@ -2328,7 +2308,6 @@ static ssize_t dp_dsc_pic_height_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { char *rd_buf = NULL; - char *rd_buf_ptr = NULL; struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private; struct display_stream_compressor *dsc; struct dcn_dsc_state dsc_state = {0}; @@ -2342,8 +2321,6 @@ static ssize_t dp_dsc_pic_height_read(struct file *f, char __user *buf, if (!rd_buf) return -ENOMEM; - rd_buf_ptr = rd_buf; - for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream && @@ -2357,10 +2334,9 @@ static ssize_t dp_dsc_pic_height_read(struct file *f, char __user *buf, if (dsc) dsc->funcs->dsc_read_state(dsc, &dsc_state); - snprintf(rd_buf_ptr, str_len, + snprintf(rd_buf, str_len, "%d\n", dsc_state.dsc_pic_height); - rd_buf_ptr += str_len; while (size) { if (*pos >= rd_buf_size) @@ -2401,7 +2377,6 @@ static ssize_t dp_dsc_chunk_size_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { char *rd_buf = NULL; - char *rd_buf_ptr = NULL; struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private; struct display_stream_compressor *dsc; struct dcn_dsc_state dsc_state = {0}; @@ -2415,8 +2390,6 @@ static ssize_t dp_dsc_chunk_size_read(struct file *f, char __user *buf, if (!rd_buf) return -ENOMEM; - rd_buf_ptr = rd_buf; - for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream && @@ -2430,10 +2403,9 @@ static ssize_t dp_dsc_chunk_size_read(struct file *f, char __user *buf, if (dsc) dsc->funcs->dsc_read_state(dsc, &dsc_state); - snprintf(rd_buf_ptr, str_len, + snprintf(rd_buf, str_len, "%d\n", dsc_state.dsc_chunk_size); - rd_buf_ptr += str_len; while (size) { if (*pos >= rd_buf_size) @@ -2474,7 +2446,6 @@ static ssize_t dp_dsc_slice_bpg_offset_read(struct file *f, char __user *buf, size_t size, loff_t *pos) { char *rd_buf = NULL; - char *rd_buf_ptr = NULL; struct amdgpu_dm_connector *aconnector = file_inode(f)->i_private; struct display_stream_compressor *dsc; struct dcn_dsc_state dsc_state = {0}; @@ -2488,8 +2459,6 @@ static ssize_t dp_dsc_slice_bpg_offset_read(struct file *f, char __user *buf, if (!rd_buf) return -ENOMEM; - rd_buf_ptr = rd_buf; - for (i = 0; i < MAX_PIPES; i++) { pipe_ctx = &aconnector->dc_link->dc->current_state->res_ctx.pipe_ctx[i]; if (pipe_ctx->stream && @@ -2503,10 +2472,9 @@ static ssize_t dp_dsc_slice_bpg_offset_read(struct file *f, char __user *buf, if (dsc) dsc->funcs->dsc_read_state(dsc, &dsc_state); - snprintf(rd_buf_ptr, str_len, + snprintf(rd_buf, str_len, "%d\n", dsc_state.dsc_slice_bpg_offset); - rd_buf_ptr += str_len; while (size) { if (*pos >= rd_buf_size) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index 50109d13d967..b0fea0856866 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -23,6 +23,8 @@ * */ +#include + #include #include #include @@ -44,6 +46,7 @@ #include "dm_helpers.h" #include "ddc_service_types.h" +#include "clk_mgr.h" static u32 edid_extract_panel_id(struct edid *edid) { @@ -73,6 +76,10 @@ static void apply_edid_quirks(struct edid *edid, struct dc_edid_caps *edid_caps) DRM_DEBUG_DRIVER("Clearing DPCD 0x317 on monitor with panel id %X\n", panel_id); edid_caps->panel_patch.remove_sink_ext_caps = true; break; + case drm_edid_encode_panel_id('S', 'D', 'C', 0x4154): + DRM_DEBUG_DRIVER("Disabling VSC on monitor with panel id %X\n", panel_id); + edid_caps->panel_patch.disable_colorimetry = true; + break; default: return; } @@ -637,6 +644,8 @@ static bool execute_synaptics_rc_command(struct drm_dp_aux *aux, // write rc data memmove(rc_data, data, length); ret = drm_dp_dpcd_write(aux, SYNAPTICS_RC_DATA, rc_data, sizeof(rc_data)); + if (ret < 0) + goto err; } // write rc offset @@ -645,20 +654,21 @@ static bool execute_synaptics_rc_command(struct drm_dp_aux *aux, rc_offset[2] = (unsigned char) (offset >> 16) & 0xFF; rc_offset[3] = (unsigned char) (offset >> 24) & 0xFF; ret = drm_dp_dpcd_write(aux, SYNAPTICS_RC_OFFSET, rc_offset, sizeof(rc_offset)); + if (ret < 0) + goto err; // write rc length rc_length[0] = (unsigned char) length & 0xFF; rc_length[1] = (unsigned char) (length >> 8) & 0xFF; ret = drm_dp_dpcd_write(aux, SYNAPTICS_RC_LENGTH, rc_length, sizeof(rc_length)); + if (ret < 0) + goto err; // write rc cmd rc_cmd = cmd | 0x80; ret = drm_dp_dpcd_write(aux, SYNAPTICS_RC_COMMAND, &rc_cmd, sizeof(rc_cmd)); - - if (ret < 0) { - DRM_ERROR("%s: write cmd ..., err = %d\n", __func__, ret); - return false; - } + if (ret < 0) + goto err; // poll until active is 0 for (i = 0; i < 10; i++) { @@ -681,6 +691,10 @@ static bool execute_synaptics_rc_command(struct drm_dp_aux *aux, drm_dbg_dp(aux->drm_dev, "success = %d\n", success); return success; + +err: + DRM_ERROR("%s: write cmd ..., err = %d\n", __func__, ret); + return false; } static void apply_synaptics_fifo_reset_wa(struct drm_dp_aux *aux) @@ -887,6 +901,60 @@ bool dm_helpers_is_dp_sink_present(struct dc_link *link) return dp_sink_present; } +static int +dm_helpers_probe_acpi_edid(void *data, u8 *buf, unsigned int block, size_t len) +{ + struct drm_connector *connector = data; + struct acpi_device *acpidev = ACPI_COMPANION(connector->dev->dev); + unsigned char start = block * EDID_LENGTH; + void *edid; + int r; + + if (!acpidev) + return -ENODEV; + + /* fetch the entire edid from BIOS */ + r = acpi_video_get_edid(acpidev, ACPI_VIDEO_DISPLAY_LCD, -1, &edid); + if (r < 0) { + drm_dbg(connector->dev, "Failed to get EDID from ACPI: %d\n", r); + return r; + } + if (len > r || start > r || start + len > r) { + r = -EINVAL; + goto cleanup; + } + + memcpy(buf, edid + start, len); + r = 0; + +cleanup: + kfree(edid); + + return r; +} + +static const struct drm_edid * +dm_helpers_read_acpi_edid(struct amdgpu_dm_connector *aconnector) +{ + struct drm_connector *connector = &aconnector->base; + + if (amdgpu_dc_debug_mask & DC_DISABLE_ACPI_EDID) + return NULL; + + switch (connector->connector_type) { + case DRM_MODE_CONNECTOR_LVDS: + case DRM_MODE_CONNECTOR_eDP: + break; + default: + return NULL; + } + + if (connector->force == DRM_FORCE_OFF) + return NULL; + + return drm_edid_read_custom(connector, dm_helpers_probe_acpi_edid, connector); +} + enum dc_edid_status dm_helpers_read_local_edid( struct dc_context *ctx, struct dc_link *link, @@ -897,7 +965,8 @@ enum dc_edid_status dm_helpers_read_local_edid( struct i2c_adapter *ddc; int retry = 3; enum dc_edid_status edid_status; - struct edid *edid; + const struct drm_edid *drm_edid; + const struct edid *edid; if (link->aux_mode) ddc = &aconnector->dm_dp_aux.aux.ddc; @@ -908,26 +977,31 @@ enum dc_edid_status dm_helpers_read_local_edid( * do check sum and retry to make sure read correct edid. */ do { - - edid = drm_get_edid(&aconnector->base, ddc); + drm_edid = dm_helpers_read_acpi_edid(aconnector); + if (drm_edid) + drm_info(connector->dev, "Using ACPI provided EDID for %s\n", connector->name); + else + drm_edid = drm_edid_read_ddc(connector, ddc); + drm_edid_connector_update(connector, drm_edid); /* DP Compliance Test 4.2.2.6 */ if (link->aux_mode && connector->edid_corrupt) drm_dp_send_real_edid_checksum(&aconnector->dm_dp_aux.aux, connector->real_edid_checksum); - if (!edid && connector->edid_corrupt) { + if (!drm_edid && connector->edid_corrupt) { connector->edid_corrupt = false; return EDID_BAD_CHECKSUM; } - if (!edid) + if (!drm_edid) return EDID_NO_RESPONSE; + edid = drm_edid_raw(drm_edid); // FIXME: Get rid of drm_edid_raw() sink->dc_edid.length = EDID_LENGTH * (edid->extensions + 1); memmove(sink->dc_edid.raw_edid, (uint8_t *)edid, sink->dc_edid.length); /* We don't need the original edid anymore */ - kfree(edid); + drm_edid_free(drm_edid); edid_status = dm_helpers_parse_edid_caps( link, @@ -1050,17 +1124,8 @@ void dm_helpers_free_gpu_mem( void *pvMem) { struct amdgpu_device *adev = ctx->driver_context; - struct dal_allocation *da; - /* walk the da list in DM */ - list_for_each_entry(da, &adev->dm.da_list, list) { - if (pvMem == da->cpu_ptr) { - amdgpu_bo_free_kernel(&da->bo, &da->gpu_addr, &da->cpu_ptr); - list_del(&da->list); - kfree(da); - break; - } - } + dm_free_gpu_mem(adev, type, pvMem); } bool dm_helpers_dmub_outbox_interrupt_control(struct dc_context *ctx, bool enable) @@ -1117,6 +1182,8 @@ bool dm_helpers_dp_handle_test_pattern_request( struct pipe_ctx *pipe_ctx = NULL; struct amdgpu_dm_connector *aconnector = link->priv; struct drm_device *dev = aconnector->base.dev; + struct dc_state *dc_state = ctx->dc->current_state; + struct clk_mgr *clk_mgr = ctx->dc->clk_mgr; int i; for (i = 0; i < MAX_PIPES; i++) { @@ -1217,6 +1284,16 @@ bool dm_helpers_dp_handle_test_pattern_request( pipe_ctx->stream->test_pattern.type = test_pattern; pipe_ctx->stream->test_pattern.color_space = test_pattern_color_space; + /* Temp W/A for compliance test failure */ + dc_state->bw_ctx.bw.dcn.clk.p_state_change_support = false; + dc_state->bw_ctx.bw.dcn.clk.dramclk_khz = clk_mgr->dc_mode_softmax_enabled ? + clk_mgr->bw_params->dc_mode_softmax_memclk : clk_mgr->bw_params->max_memclk_mhz; + dc_state->bw_ctx.bw.dcn.clk.idle_dramclk_khz = dc_state->bw_ctx.bw.dcn.clk.dramclk_khz; + ctx->dc->clk_mgr->funcs->update_clocks( + ctx->dc->clk_mgr, + dc_state, + false); + dc_link_dp_set_test_pattern( (struct dc_link *) link, test_pattern, @@ -1297,4 +1374,4 @@ bool dm_helpers_is_hdr_on(struct dc_context *ctx, struct dc_stream_state *stream { // TODO return false; -} \ No newline at end of file +} diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 83a31b97e96b..6e4359490613 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -129,7 +129,7 @@ dm_dp_mst_connector_destroy(struct drm_connector *connector) dc_sink_release(aconnector->dc_sink); } - kfree(aconnector->edid); + drm_edid_free(aconnector->drm_edid); drm_connector_cleanup(connector); drm_dp_mst_put_port_malloc(aconnector->mst_output_port); @@ -182,7 +182,7 @@ amdgpu_dm_mst_connector_early_unregister(struct drm_connector *connector) dc_sink_release(dc_sink); aconnector->dc_sink = NULL; - aconnector->edid = NULL; + aconnector->drm_edid = NULL; aconnector->dsc_aux = NULL; port->passthrough_aux = NULL; } @@ -302,16 +302,18 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector) if (!aconnector) return drm_add_edid_modes(connector, NULL); - if (!aconnector->edid) { - struct edid *edid; + if (!aconnector->drm_edid) { + const struct drm_edid *drm_edid; - edid = drm_dp_mst_get_edid(connector, &aconnector->mst_root->mst_mgr, aconnector->mst_output_port); + drm_edid = drm_dp_mst_edid_read(connector, + &aconnector->mst_root->mst_mgr, + aconnector->mst_output_port); - if (!edid) { + if (!drm_edid) { amdgpu_dm_set_mst_status(&aconnector->mst_status, MST_REMOTE_EDID, false); - drm_connector_update_edid_property( + drm_edid_connector_update( &aconnector->base, NULL); @@ -345,7 +347,7 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector) return ret; } - aconnector->edid = edid; + aconnector->drm_edid = drm_edid; amdgpu_dm_set_mst_status(&aconnector->mst_status, MST_REMOTE_EDID, true); } @@ -360,10 +362,13 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector) struct dc_sink_init_data init_params = { .link = aconnector->dc_link, .sink_signal = SIGNAL_TYPE_DISPLAY_PORT_MST }; + const struct edid *edid; + + edid = drm_edid_raw(aconnector->drm_edid); // FIXME: Get rid of drm_edid_raw() dc_sink = dc_link_add_remote_sink( aconnector->dc_link, - (uint8_t *)aconnector->edid, - (aconnector->edid->extensions + 1) * EDID_LENGTH, + (uint8_t *)edid, + (edid->extensions + 1) * EDID_LENGTH, &init_params); if (!dc_sink) { @@ -405,7 +410,7 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector) if (aconnector->dc_sink) { amdgpu_dm_update_freesync_caps( - connector, aconnector->edid); + connector, aconnector->drm_edid); #if defined(CONFIG_DRM_AMD_DC_FP) if (!validate_dsc_caps_on_connector(aconnector)) @@ -419,10 +424,9 @@ static int dm_dp_mst_get_modes(struct drm_connector *connector) } } - drm_connector_update_edid_property( - &aconnector->base, aconnector->edid); + drm_edid_connector_update(&aconnector->base, aconnector->drm_edid); - ret = drm_add_edid_modes(connector, aconnector->edid); + ret = drm_edid_connector_add_modes(connector); return ret; } @@ -500,7 +504,7 @@ dm_dp_mst_detect(struct drm_connector *connector, dc_sink_release(aconnector->dc_sink); aconnector->dc_sink = NULL; - aconnector->edid = NULL; + aconnector->drm_edid = NULL; aconnector->dsc_aux = NULL; port->passthrough_aux = NULL; @@ -1027,6 +1031,7 @@ static int try_disable_dsc(struct drm_atomic_state *state, int remaining_to_try = 0; int ret; uint16_t fec_overhead_multiplier_x1000 = get_fec_overhead_multiplier(dc_link); + int var_pbn; for (i = 0; i < count; i++) { if (vars[i + k].dsc_enabled @@ -1057,13 +1062,18 @@ static int try_disable_dsc(struct drm_atomic_state *state, break; DRM_DEBUG_DRIVER("MST_DSC index #%d, try no compression\n", next_index); + var_pbn = vars[next_index].pbn; vars[next_index].pbn = kbps_to_peak_pbn(params[next_index].bw_range.stream_kbps, fec_overhead_multiplier_x1000); ret = drm_dp_atomic_find_time_slots(state, params[next_index].port->mgr, params[next_index].port, vars[next_index].pbn); - if (ret < 0) + if (ret < 0) { + DRM_DEBUG_DRIVER("%s:%d MST_DSC index #%d, failed to set pbn to the state, %d\n", + __func__, __LINE__, next_index, ret); + vars[next_index].pbn = var_pbn; return ret; + } ret = drm_dp_mst_atomic_check(state); if (ret == 0) { @@ -1071,14 +1081,17 @@ static int try_disable_dsc(struct drm_atomic_state *state, vars[next_index].dsc_enabled = false; vars[next_index].bpp_x16 = 0; } else { - DRM_DEBUG_DRIVER("MST_DSC index #%d, restore minimum compression\n", next_index); - vars[next_index].pbn = kbps_to_peak_pbn(params[next_index].bw_range.max_kbps, fec_overhead_multiplier_x1000); + DRM_DEBUG_DRIVER("MST_DSC index #%d, restore optimized pbn value\n", next_index); + vars[next_index].pbn = var_pbn; ret = drm_dp_atomic_find_time_slots(state, params[next_index].port->mgr, params[next_index].port, vars[next_index].pbn); - if (ret < 0) + if (ret < 0) { + DRM_DEBUG_DRIVER("%s:%d MST_DSC index #%d, failed to set pbn to the state, %d\n", + __func__, __LINE__, next_index, ret); return ret; + } } tried[next_index] = true; @@ -1111,6 +1124,7 @@ static int compute_mst_dsc_configs_for_link(struct drm_atomic_state *state, int i, k, ret; bool debugfs_overwrite = false; uint16_t fec_overhead_multiplier_x1000 = get_fec_overhead_multiplier(dc_link); + struct drm_connector_state *new_conn_state; memset(params, 0, sizeof(params)); @@ -1118,7 +1132,7 @@ static int compute_mst_dsc_configs_for_link(struct drm_atomic_state *state, return PTR_ERR(mst_state); /* Set up params */ - DRM_DEBUG_DRIVER("%s: MST_DSC Set up params for %d streams\n", __func__, dc_state->stream_count); + DRM_DEBUG_DRIVER("%s: MST_DSC Try to set up params from %d streams\n", __func__, dc_state->stream_count); for (i = 0; i < dc_state->stream_count; i++) { struct dc_dsc_policy dsc_policy = {0}; @@ -1134,6 +1148,14 @@ static int compute_mst_dsc_configs_for_link(struct drm_atomic_state *state, if (!aconnector->mst_output_port) continue; + new_conn_state = drm_atomic_get_new_connector_state(state, &aconnector->base); + + if (!new_conn_state) { + DRM_DEBUG_DRIVER("%s:%d MST_DSC Skip the stream 0x%p with invalid new_conn_state\n", + __func__, __LINE__, stream); + continue; + } + stream->timing.flags.DSC = 0; params[count].timing = &stream->timing; @@ -1166,6 +1188,8 @@ static int compute_mst_dsc_configs_for_link(struct drm_atomic_state *state, count++; } + DRM_DEBUG_DRIVER("%s: MST_DSC Params set up for %d streams\n", __func__, count); + if (count == 0) { ASSERT(0); return 0; @@ -1293,7 +1317,7 @@ static bool is_dsc_need_re_compute( continue; aconnector = (struct amdgpu_dm_connector *) stream->dm_stream_context; - if (!aconnector || !aconnector->dsc_aux) + if (!aconnector) continue; stream_on_link[new_stream_on_link_num] = aconnector; diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c index adc710fe4a45..8d2cf95ae739 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c @@ -78,10 +78,3 @@ void bios_set_scratch_critical_state( uint32_t critial_state = state ? 1 : 0; REG_UPDATE(BIOS_SCRATCH_6, S6_CRITICAL_STATE, critial_state); } - -uint32_t bios_get_vga_enabled_displays( - struct dc_bios *bios) -{ - return REG_READ(BIOS_SCRATCH_3) & 0XFFFF; -} - diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h index e1b4a40a353d..ab162f2fe577 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h @@ -34,7 +34,6 @@ uint8_t *bios_get_image(struct dc_bios *bp, uint32_t offset, bool bios_is_accelerated_mode(struct dc_bios *bios); void bios_set_scratch_acc_mode_change(struct dc_bios *bios, uint32_t state); void bios_set_scratch_critical_state(struct dc_bios *bios, bool state); -uint32_t bios_get_vga_enabled_displays(struct dc_bios *bios); #define GET_IMAGE(type, offset) ((type *) bios_get_image(&bp->base, offset, sizeof(type))) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c index b46a3afe48ca..7d68006137a9 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn35/dcn35_clk_mgr.c @@ -257,11 +257,11 @@ static void dcn35_notify_host_router_bw(struct clk_mgr *clk_mgr_base, struct dc_ struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); uint32_t host_router_bw_kbps[MAX_HOST_ROUTERS_NUM] = { 0 }; int i; - for (i = 0; i < context->stream_count; ++i) { const struct dc_stream_state *stream = context->streams[i]; const struct dc_link *link = stream->link; - uint8_t lowest_dpia_index = 0, hr_index = 0; + uint8_t lowest_dpia_index = 0; + unsigned int hr_index = 0; if (!link) continue; @@ -271,6 +271,8 @@ static void dcn35_notify_host_router_bw(struct clk_mgr *clk_mgr_base, struct dc_ continue; hr_index = (link->link_index - lowest_dpia_index) / 2; + if (hr_index >= MAX_HOST_ROUTERS_NUM) + continue; host_router_bw_kbps[hr_index] += dc_bandwidth_in_kbps_from_timing( &stream->timing, dc_link_get_highest_encoding_format(link)); } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 5c39390ecbd5..5a12fc75f97f 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -621,8 +621,8 @@ dc_stream_forward_crc_window(struct dc_stream_state *stream, * dc_stream_configure_crc() - Configure CRC capture for the given stream. * @dc: DC Object * @stream: The stream to configure CRC on. - * @enable: Enable CRC if true, disable otherwise. * @crc_window: CRC window (x/y start/end) information + * @enable: Enable CRC if true, disable otherwise. * @continuous: Capture CRC on every frame if true. Otherwise, only capture * once. * @@ -1157,6 +1157,8 @@ static void dc_update_visual_confirm_color(struct dc *dc, struct dc_state *conte get_surface_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color)); else if (dc->debug.visual_confirm == VISUAL_CONFIRM_SWIZZLE) get_surface_tile_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color)); + else if (dc->debug.visual_confirm == VISUAL_CONFIRM_HW_CURSOR) + get_cursor_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color)); else { if (dc->ctx->dce_version < DCN_VERSION_2_0) color_space_to_black_color( @@ -1233,16 +1235,8 @@ static void disable_dangling_plane(struct dc *dc, struct dc_state *context) */ if (is_phantom) { if (tg->funcs->enable_crtc) { - int main_pipe_width = 0, main_pipe_height = 0; - struct dc_stream_state *old_paired_stream = dc_state_get_paired_subvp_stream(dc->current_state, old_stream); - - if (old_paired_stream) { - main_pipe_width = old_paired_stream->dst.width; - main_pipe_height = old_paired_stream->dst.height; - } - - if (dc->hwss.blank_phantom) - dc->hwss.blank_phantom(dc, tg, main_pipe_width, main_pipe_height); + if (dc->hwseq->funcs.blank_pixel_data) + dc->hwseq->funcs.blank_pixel_data(dc, pipe, true); tg->funcs->enable_crtc(tg); } } @@ -1437,6 +1431,7 @@ void dc_hardware_init(struct dc *dc) detect_edp_presence(dc); if (dc->ctx->dce_environment != DCE_ENV_VIRTUAL_HW) dc->hwss.init_hw(dc); + dc_dmub_srv_notify_fw_dc_power_state(dc->ctx->dmub_srv, DC_ACPI_CM_POWER_STATE_D0); } void dc_init_callbacks(struct dc *dc, @@ -1876,6 +1871,41 @@ void dc_z10_save_init(struct dc *dc) dc->hwss.z10_save_init(dc); } +/* Set a pipe unlock order based on the change in DET allocation and stores it in dc scratch memory + * Prevents over allocation of DET during unlock process + * e.g. 2 pipe config with different streams with a max of 20 DET segments + * Before: After: + * - Pipe0: 10 DET segments - Pipe0: 12 DET segments + * - Pipe1: 10 DET segments - Pipe1: 8 DET segments + * If Pipe0 gets updated first, 22 DET segments will be allocated + */ +static void determine_pipe_unlock_order(struct dc *dc, struct dc_state *context) +{ + unsigned int i = 0; + struct pipe_ctx *pipe = NULL; + struct timing_generator *tg = NULL; + + if (!dc->config.set_pipe_unlock_order) + return; + + memset(dc->scratch.pipes_to_unlock_first, 0, sizeof(dc->scratch.pipes_to_unlock_first)); + for (i = 0; i < dc->res_pool->pipe_count; i++) { + pipe = &context->res_ctx.pipe_ctx[i]; + tg = pipe->stream_res.tg; + + if (!resource_is_pipe_type(pipe, OTG_MASTER) || + !tg->funcs->is_tg_enabled(tg) || + dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM) { + continue; + } + + if (resource_calculate_det_for_stream(context, pipe) < + resource_calculate_det_for_stream(dc->current_state, &dc->current_state->res_ctx.pipe_ctx[i])) { + dc->scratch.pipes_to_unlock_first[i] = true; + } + } +} + /** * dc_commit_state_no_check - Apply context to the hardware * @@ -1974,6 +2004,7 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c context->streams[i]->update_flags.bits.dsc_changed = prev_dsc_changed; } + determine_pipe_unlock_order(dc, context); /* Program all planes within new context*/ if (dc->res_pool->funcs->prepare_mcache_programming) dc->res_pool->funcs->prepare_mcache_programming(dc, context); @@ -2156,6 +2187,14 @@ enum dc_status dc_commit_streams(struct dc *dc, struct dc_commit_streams_params context->power_source = params->power_source; res = dc_validate_with_context(dc, set, params->stream_count, context, false); + + /* + * Only update link encoder to stream assignment after bandwidth validation passed. + */ + if (res == DC_OK && dc->res_pool->funcs->link_encs_assign) + dc->res_pool->funcs->link_encs_assign( + dc, context, context->streams, context->stream_count); + if (res != DC_OK) { BREAK_TO_DEBUGGER(); goto fail; @@ -2477,41 +2516,35 @@ static enum surface_update_type get_scaling_info_update_type( if (!u->scaling_info) return UPDATE_TYPE_FAST; - if (u->scaling_info->dst_rect.width != u->surface->dst_rect.width + if (u->scaling_info->src_rect.width != u->surface->src_rect.width + || u->scaling_info->src_rect.height != u->surface->src_rect.height + || u->scaling_info->dst_rect.width != u->surface->dst_rect.width || u->scaling_info->dst_rect.height != u->surface->dst_rect.height + || u->scaling_info->clip_rect.width != u->surface->clip_rect.width + || u->scaling_info->clip_rect.height != u->surface->clip_rect.height || u->scaling_info->scaling_quality.integer_scaling != - u->surface->scaling_quality.integer_scaling - ) { + u->surface->scaling_quality.integer_scaling) { update_flags->bits.scaling_change = 1; + if (u->scaling_info->src_rect.width > u->surface->src_rect.width + || u->scaling_info->src_rect.height > u->surface->src_rect.height) + /* Making src rect bigger requires a bandwidth change */ + update_flags->bits.clock_change = 1; + if ((u->scaling_info->dst_rect.width < u->surface->dst_rect.width || u->scaling_info->dst_rect.height < u->surface->dst_rect.height) && (u->scaling_info->dst_rect.width < u->surface->src_rect.width || u->scaling_info->dst_rect.height < u->surface->src_rect.height)) /* Making dst rect smaller requires a bandwidth change */ update_flags->bits.bandwidth_change = 1; + + if (u->scaling_info->src_rect.width > dc->caps.max_optimizable_video_width && + (u->scaling_info->clip_rect.width > u->surface->clip_rect.width || + u->scaling_info->clip_rect.height > u->surface->clip_rect.height)) + /* Changing clip size of a large surface may result in MPC slice count change */ + update_flags->bits.bandwidth_change = 1; } - if (u->scaling_info->src_rect.width != u->surface->src_rect.width - || u->scaling_info->src_rect.height != u->surface->src_rect.height) { - - update_flags->bits.scaling_change = 1; - if (u->scaling_info->src_rect.width > u->surface->src_rect.width - || u->scaling_info->src_rect.height > u->surface->src_rect.height) - /* Making src rect bigger requires a bandwidth change */ - update_flags->bits.clock_change = 1; - } - - if (u->scaling_info->src_rect.width > dc->caps.max_optimizable_video_width && - (u->scaling_info->clip_rect.width > u->surface->clip_rect.width || - u->scaling_info->clip_rect.height > u->surface->clip_rect.height)) - /* Changing clip size of a large surface may result in MPC slice count change */ - update_flags->bits.bandwidth_change = 1; - - if (u->scaling_info->clip_rect.width != u->surface->clip_rect.width || - u->scaling_info->clip_rect.height != u->surface->clip_rect.height) - update_flags->bits.clip_size_change = 1; - if (u->scaling_info->src_rect.x != u->surface->src_rect.x || u->scaling_info->src_rect.y != u->surface->src_rect.y || u->scaling_info->clip_rect.x != u->surface->clip_rect.x @@ -2520,13 +2553,13 @@ static enum surface_update_type get_scaling_info_update_type( || u->scaling_info->dst_rect.y != u->surface->dst_rect.y) update_flags->bits.position_change = 1; + /* process every update flag before returning */ if (update_flags->bits.clock_change || update_flags->bits.bandwidth_change || update_flags->bits.scaling_change) return UPDATE_TYPE_FULL; - if (update_flags->bits.position_change || - update_flags->bits.clip_size_change) + if (update_flags->bits.position_change) return UPDATE_TYPE_MED; return UPDATE_TYPE_FAST; @@ -2617,7 +2650,8 @@ static enum surface_update_type det_surface_update(const struct dc *dc, elevate_update_type(&overall_type, type); } - if (update_flags->bits.lut_3d) { + if (update_flags->bits.lut_3d && + u->surface->mcm_luts.lut3d_data.lut3d_src != DC_CM2_TRANSFER_FUNC_SOURCE_VIDMEM) { type = UPDATE_TYPE_FULL; elevate_update_type(&overall_type, type); } @@ -2637,6 +2671,29 @@ static enum surface_update_type det_surface_update(const struct dc *dc, return overall_type; } +/* May need to flip the desktop plane in cases where MPO plane receives a flip but desktop plane doesn't + * while both planes are flip_immediate + */ +static void force_immediate_gsl_plane_flip(struct dc *dc, struct dc_surface_update *updates, int surface_count) +{ + bool has_flip_immediate_plane = false; + int i; + + for (i = 0; i < surface_count; i++) { + if (updates[i].surface->flip_immediate) { + has_flip_immediate_plane = true; + break; + } + } + + if (has_flip_immediate_plane && surface_count > 1) { + for (i = 0; i < surface_count; i++) { + if (updates[i].surface->flip_immediate) + updates[i].surface->update_flags.bits.addr_update = 1; + } + } +} + static enum surface_update_type check_update_surfaces_for_stream( struct dc *dc, struct dc_surface_update *updates, @@ -2699,6 +2756,9 @@ static enum surface_update_type check_update_surfaces_for_stream( if (stream_update->scaler_sharpener_update) su_flags->bits.scaler_sharpener = 1; + if (stream_update->sharpening_required) + su_flags->bits.sharpening_required = 1; + if (su_flags->raw != 0) overall_type = UPDATE_TYPE_FULL; @@ -2870,10 +2930,20 @@ static void copy_surface_update_to_plane( sizeof(struct dc_transfer_func_distributed_points)); } - if (srf_update->func_shaper) + if (srf_update->cm2_params) { + surface->mcm_shaper_3dlut_setting = srf_update->cm2_params->component_settings.shaper_3dlut_setting; + surface->mcm_lut1d_enable = srf_update->cm2_params->component_settings.lut1d_enable; + surface->mcm_luts = srf_update->cm2_params->cm2_luts; + } + + if (srf_update->func_shaper) { memcpy(&surface->in_shaper_func, srf_update->func_shaper, sizeof(surface->in_shaper_func)); + if (surface->mcm_shaper_3dlut_setting >= DC_CM2_SHAPER_3DLUT_SETTING_ENABLE_SHAPER) + surface->mcm_luts.shaper = &surface->in_shaper_func; + } + if (srf_update->lut3d_func) memcpy(&surface->lut3d_func, srf_update->lut3d_func, sizeof(surface->lut3d_func)); @@ -2886,10 +2956,17 @@ static void copy_surface_update_to_plane( surface->sdr_white_level_nits = srf_update->sdr_white_level_nits; - if (srf_update->blend_tf) + if (srf_update->blend_tf) { memcpy(&surface->blend_tf, srf_update->blend_tf, sizeof(surface->blend_tf)); + if (surface->mcm_lut1d_enable) + surface->mcm_luts.lut1d_func = &surface->blend_tf; + } + + if (srf_update->cm2_params || srf_update->blend_tf) + surface->lut_bank_a = !surface->lut_bank_a; + if (srf_update->input_csc_color_matrix) surface->input_csc_color_matrix = *srf_update->input_csc_color_matrix; @@ -2901,11 +2978,7 @@ static void copy_surface_update_to_plane( if (srf_update->gamut_remap_matrix) surface->gamut_remap_matrix = *srf_update->gamut_remap_matrix; - if (srf_update->cm2_params) { - surface->mcm_shaper_3dlut_setting = srf_update->cm2_params->component_settings.shaper_3dlut_setting; - surface->mcm_lut1d_enable = srf_update->cm2_params->component_settings.lut1d_enable; - surface->mcm_luts = srf_update->cm2_params->cm2_luts; - } + if (srf_update->cursor_csc_color_matrix) surface->cursor_csc_color_matrix = *srf_update->cursor_csc_color_matrix; @@ -3037,6 +3110,8 @@ static void copy_stream_update_to_stream(struct dc *dc, } if (update->scaler_sharpener_update) stream->scaler_sharpener_update = *update->scaler_sharpener_update; + if (update->sharpening_required) + stream->sharpening_required = *update->sharpening_required; } static void backup_planes_and_stream_state( @@ -3153,6 +3228,11 @@ static bool update_planes_and_stream_state(struct dc *dc, context = dc->current_state; update_type = dc_check_update_surfaces_for_stream( dc, srf_updates, surface_count, stream_update, stream_status); + /* It is possible to receive a flip for one plane while there are multiple flip_immediate planes in the same stream. + * E.g. Desktop and MPO plane are flip_immediate but only the MPO plane received a flip + * Force the other flip_immediate planes to flip so GSL doesn't wait for a flip that won't come. + */ + force_immediate_gsl_plane_flip(dc, srf_updates, surface_count); if (update_type == UPDATE_TYPE_FULL) backup_planes_and_stream_state(&dc->scratch.current_state, stream); @@ -3225,8 +3305,7 @@ static bool update_planes_and_stream_state(struct dc *dc, if (update_type != UPDATE_TYPE_MED) continue; - if (surface->update_flags.bits.clip_size_change || - surface->update_flags.bits.position_change) { + if (surface->update_flags.bits.position_change) { for (j = 0; j < dc->res_pool->pipe_count; j++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; @@ -3625,6 +3704,10 @@ static void commit_planes_for_stream_fast(struct dc *dc, struct pipe_ctx *top_pipe_to_program = NULL; struct dc_stream_status *stream_status = NULL; bool should_offload_fams2_flip = false; + bool should_lock_all_pipes = (update_type != UPDATE_TYPE_FAST); + + if (should_lock_all_pipes) + determine_pipe_unlock_order(dc, context); if (dc->debug.fams2_config.bits.enable && dc->debug.fams2_config.bits.enable_offload_flip && @@ -3677,13 +3760,14 @@ static void commit_planes_for_stream_fast(struct dc *dc, if (!pipe_ctx->plane_state) continue; - if (should_update_pipe_for_plane(context, pipe_ctx, plane_state)) + if (!should_update_pipe_for_plane(context, pipe_ctx, plane_state)) continue; + pipe_ctx->plane_state->triplebuffer_flips = false; if (update_type == UPDATE_TYPE_FAST && - dc->hwss.program_triplebuffer != NULL && - !pipe_ctx->plane_state->flip_immediate && dc->debug.enable_tri_buf) { - /*triple buffer for VUpdate only*/ + dc->hwss.program_triplebuffer != NULL && + !pipe_ctx->plane_state->flip_immediate && dc->debug.enable_tri_buf) { + /*triple buffer for VUpdate only*/ pipe_ctx->plane_state->triplebuffer_flips = true; } } @@ -3742,6 +3826,8 @@ static void commit_planes_for_stream(struct dc *dc, bool subvp_curr_use = false; uint8_t current_stream_mask = 0; + if (should_lock_all_pipes) + determine_pipe_unlock_order(dc, context); // Once we apply the new subvp context to hardware it won't be in the // dc->current_state anymore, so we have to cache it before we apply // the new SubVP context @@ -3920,19 +4006,20 @@ static void commit_planes_for_stream(struct dc *dc, struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[j]; if (!pipe_ctx->plane_state) continue; - if (should_update_pipe_for_plane(context, pipe_ctx, plane_state)) + if (!should_update_pipe_for_plane(context, pipe_ctx, plane_state)) continue; pipe_ctx->plane_state->triplebuffer_flips = false; if (update_type == UPDATE_TYPE_FAST && - dc->hwss.program_triplebuffer != NULL && - !pipe_ctx->plane_state->flip_immediate && dc->debug.enable_tri_buf) { - /*triple buffer for VUpdate only*/ - pipe_ctx->plane_state->triplebuffer_flips = true; + dc->hwss.program_triplebuffer != NULL && + !pipe_ctx->plane_state->flip_immediate && dc->debug.enable_tri_buf) { + /*triple buffer for VUpdate only*/ + pipe_ctx->plane_state->triplebuffer_flips = true; } } if (update_type == UPDATE_TYPE_FULL) { /* force vsync flip when reconfiguring pipes to prevent underflow */ plane_state->flip_immediate = false; + plane_state->triplebuffer_flips = false; } } @@ -3953,7 +4040,6 @@ static void commit_planes_for_stream(struct dc *dc, continue; ASSERT(!pipe_ctx->plane_state->triplebuffer_flips); - if (dc->hwss.program_triplebuffer != NULL && dc->debug.enable_tri_buf) { /*turn off triple buffer for full update*/ dc->hwss.program_triplebuffer( @@ -4028,7 +4114,7 @@ static void commit_planes_for_stream(struct dc *dc, /*program triple buffer after lock based on flip type*/ if (dc->hwss.program_triplebuffer != NULL && dc->debug.enable_tri_buf) { - /*only enable triplebuffer for fast_update*/ + /*only enable triplebuffer for fast_update*/ dc->hwss.program_triplebuffer( dc, pipe_ctx, pipe_ctx->plane_state->triplebuffer_flips); } @@ -4777,6 +4863,11 @@ static bool update_planes_and_stream_v1(struct dc *dc, update_type = dc_check_update_surfaces_for_stream( dc, srf_updates, surface_count, stream_update, stream_status); + /* It is possible to receive a flip for one plane while there are multiple flip_immediate planes in the same stream. + * E.g. Desktop and MPO plane are flip_immediate but only the MPO plane received a flip + * Force the other flip_immediate planes to flip so GSL doesn't wait for a flip that won't come. + */ + force_immediate_gsl_plane_flip(dc, srf_updates, surface_count); if (update_type >= UPDATE_TYPE_FULL) { @@ -5065,11 +5156,26 @@ static bool update_planes_and_stream_v3(struct dc *dc, return true; } +static void clear_update_flags(struct dc_surface_update *srf_updates, + int surface_count, struct dc_stream_state *stream) +{ + int i; + + if (stream) + stream->update_flags.raw = 0; + + for (i = 0; i < surface_count; i++) + if (srf_updates[i].surface) + srf_updates[i].surface->update_flags.raw = 0; +} + bool dc_update_planes_and_stream(struct dc *dc, struct dc_surface_update *srf_updates, int surface_count, struct dc_stream_state *stream, struct dc_stream_update *stream_update) { + bool ret = false; + dc_exit_ips_for_hw_access(dc); /* * update planes and stream version 3 separates FULL and FAST updates @@ -5086,10 +5192,16 @@ bool dc_update_planes_and_stream(struct dc *dc, * features as they are now transparent to the new sequence. */ if (dc->ctx->dce_version >= DCN_VERSION_4_01) - return update_planes_and_stream_v3(dc, srf_updates, + ret = update_planes_and_stream_v3(dc, srf_updates, surface_count, stream, stream_update); - return update_planes_and_stream_v2(dc, srf_updates, + else + ret = update_planes_and_stream_v2(dc, srf_updates, surface_count, stream, stream_update); + + if (ret) + clear_update_flags(srf_updates, surface_count, stream); + + return ret; } void dc_commit_updates_for_stream(struct dc *dc, @@ -5099,6 +5211,8 @@ void dc_commit_updates_for_stream(struct dc *dc, struct dc_stream_update *stream_update, struct dc_state *state) { + bool ret = false; + dc_exit_ips_for_hw_access(dc); /* TODO: Since change commit sequence can have a huge impact, * we decided to only enable it for DCN3x. However, as soon as @@ -5106,17 +5220,17 @@ void dc_commit_updates_for_stream(struct dc *dc, * the new sequence for all ASICs. */ if (dc->ctx->dce_version >= DCN_VERSION_4_01) { - update_planes_and_stream_v3(dc, srf_updates, surface_count, + ret = update_planes_and_stream_v3(dc, srf_updates, surface_count, stream, stream_update); - return; - } - if (dc->ctx->dce_version >= DCN_VERSION_3_2) { - update_planes_and_stream_v2(dc, srf_updates, surface_count, + } else if (dc->ctx->dce_version >= DCN_VERSION_3_2) { + ret = update_planes_and_stream_v2(dc, srf_updates, surface_count, stream, stream_update); - return; - } - update_planes_and_stream_v1(dc, srf_updates, surface_count, stream, - stream_update, state); + } else + ret = update_planes_and_stream_v1(dc, srf_updates, surface_count, stream, + stream_update, state); + + if (ret) + clear_update_flags(srf_updates, surface_count, stream); } uint8_t dc_get_current_stream_count(struct dc *dc) @@ -5976,7 +6090,12 @@ struct dc_power_profile dc_get_power_profile_for_dc_state(const struct dc_state { struct dc_power_profile profile = { 0 }; - profile.power_level += !context->bw_ctx.bw.dcn.clk.p_state_change_support; + if (!context || !context->clk_mgr || !context->clk_mgr->ctx || !context->clk_mgr->ctx->dc) + return profile; + struct dc *dc = context->clk_mgr->ctx->dc; + + if (dc->res_pool->funcs->get_power_profile) + profile.power_level = dc->res_pool->funcs->get_power_profile(context); return profile; } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c index 7ee2be8f82c4..2fdcf8d59b9f 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_hw_sequencer.c @@ -497,6 +497,23 @@ void get_mclk_switch_visual_confirm_color( } } +void get_cursor_visual_confirm_color( + struct pipe_ctx *pipe_ctx, + struct tg_color *color) +{ + uint32_t color_value = MAX_TG_COLOR_VALUE; + + if (pipe_ctx->stream && pipe_ctx->stream->cursor_position.enable) { + color->color_r_cr = color_value; + color->color_g_y = 0; + color->color_b_cb = 0; + } else { + color->color_r_cr = 0; + color->color_g_y = 0; + color->color_b_cb = color_value; + } +} + void set_p_state_switch_method( struct dc *dc, struct dc_state *context, @@ -1071,8 +1088,13 @@ void hwss_wait_for_outstanding_hw_updates(struct dc *dc, struct dc_state *dc_con if (!pipe_ctx->stream) continue; - if (pipe_ctx->stream_res.tg->funcs->wait_drr_doublebuffer_pending_clear) - pipe_ctx->stream_res.tg->funcs->wait_drr_doublebuffer_pending_clear(pipe_ctx->stream_res.tg); + /* For full update we must wait for all double buffer updates, not just DRR updates. This + * is particularly important for minimal transitions. Only check for OTG_MASTER pipes, + * as non-OTG Master pipes share the same OTG as + */ + if (resource_is_pipe_type(pipe_ctx, OTG_MASTER) && dc->hwss.wait_for_all_pending_updates) { + dc->hwss.wait_for_all_pending_updates(pipe_ctx); + } hubp = pipe_ctx->plane_res.hubp; if (!hubp) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index c7599c40d4be..33125b95c3a1 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -765,25 +765,6 @@ static inline void get_vp_scan_direction( *flip_horz_scan_dir = !*flip_horz_scan_dir; } -/* - * This is a preliminary vp size calculation to allow us to check taps support. - * The result is completely overridden afterwards. - */ -static void calculate_viewport_size(struct pipe_ctx *pipe_ctx) -{ - struct scaler_data *data = &pipe_ctx->plane_res.scl_data; - - data->viewport.width = dc_fixpt_ceil(dc_fixpt_mul_int(data->ratios.horz, data->recout.width)); - data->viewport.height = dc_fixpt_ceil(dc_fixpt_mul_int(data->ratios.vert, data->recout.height)); - data->viewport_c.width = dc_fixpt_ceil(dc_fixpt_mul_int(data->ratios.horz_c, data->recout.width)); - data->viewport_c.height = dc_fixpt_ceil(dc_fixpt_mul_int(data->ratios.vert_c, data->recout.height)); - if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_90 || - pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270) { - swap(data->viewport.width, data->viewport.height); - swap(data->viewport_c.width, data->viewport_c.height); - } -} - static struct rect intersect_rec(const struct rect *r0, const struct rect *r1) { struct rect rec; @@ -1468,6 +1449,7 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) const struct dc_plane_state *plane_state = pipe_ctx->plane_state; struct dc_crtc_timing *timing = &pipe_ctx->stream->timing; const struct rect odm_slice_src = resource_get_odm_slice_src_rect(pipe_ctx); + struct scaling_taps temp = {0}; bool res = false; DC_LOGGER_INIT(pipe_ctx->stream->ctx->logger); @@ -1525,8 +1507,6 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) calculate_recout(pipe_ctx); /* depends on pixel format */ calculate_scaling_ratios(pipe_ctx); - /* depends on scaling ratios and recout, does not calculate offset yet */ - calculate_viewport_size(pipe_ctx); /* * LB calculations depend on vp size, h/v_active and scaling ratios @@ -1547,6 +1527,24 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.scl_data.lb_params.alpha_en = plane_state->per_pixel_alpha; + // get TAP value with 100x100 dummy data for max scaling qualify, override + // if a new scaling quality required + pipe_ctx->plane_res.scl_data.viewport.width = 100; + pipe_ctx->plane_res.scl_data.viewport.height = 100; + pipe_ctx->plane_res.scl_data.viewport_c.width = 100; + pipe_ctx->plane_res.scl_data.viewport_c.height = 100; + if (pipe_ctx->plane_res.xfm != NULL) + res = pipe_ctx->plane_res.xfm->funcs->transform_get_optimal_number_of_taps( + pipe_ctx->plane_res.xfm, &pipe_ctx->plane_res.scl_data, &plane_state->scaling_quality); + + if (pipe_ctx->plane_res.dpp != NULL) + res = pipe_ctx->plane_res.dpp->funcs->dpp_get_optimal_number_of_taps( + pipe_ctx->plane_res.dpp, &pipe_ctx->plane_res.scl_data, &plane_state->scaling_quality); + + temp = pipe_ctx->plane_res.scl_data.taps; + + calculate_inits_and_viewports(pipe_ctx); + if (pipe_ctx->plane_res.xfm != NULL) res = pipe_ctx->plane_res.xfm->funcs->transform_get_optimal_number_of_taps( pipe_ctx->plane_res.xfm, &pipe_ctx->plane_res.scl_data, &plane_state->scaling_quality); @@ -1573,11 +1571,10 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) &plane_state->scaling_quality); } - /* - * Depends on recout, scaling ratios, h_active and taps - * May need to re-check lb size after this in some obscure scenario - */ - if (res) + if (res && (pipe_ctx->plane_res.scl_data.taps.v_taps != temp.v_taps || + pipe_ctx->plane_res.scl_data.taps.h_taps != temp.h_taps || + pipe_ctx->plane_res.scl_data.taps.v_taps_c != temp.v_taps_c || + pipe_ctx->plane_res.scl_data.taps.h_taps_c != temp.h_taps_c)) calculate_inits_and_viewports(pipe_ctx); /* @@ -4094,14 +4091,6 @@ enum dc_status dc_validate_global_state( if (!dc->res_pool->funcs->validate_bandwidth(dc, new_ctx, fast_validate)) result = DC_FAIL_BANDWIDTH_VALIDATE; - /* - * Only update link encoder to stream assignment after bandwidth validation passed. - * TODO: Split out assignment and validation. - */ - if (result == DC_OK && dc->res_pool->funcs->link_encs_assign && fast_validate == false) - dc->res_pool->funcs->link_encs_assign( - dc, new_ctx, new_ctx->streams, new_ctx->stream_count); - return result; } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index 9a406d74c0dd..5d233c09d239 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -421,7 +421,6 @@ bool dc_stream_program_cursor_position( /* apply/update visual confirm */ if (dc->debug.visual_confirm == VISUAL_CONFIRM_HW_CURSOR) { /* update software state */ - uint32_t color_value = MAX_TG_COLOR_VALUE; int i; for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -429,15 +428,7 @@ bool dc_stream_program_cursor_position( /* adjust visual confirm color for all pipes with current stream */ if (stream == pipe_ctx->stream) { - if (stream->cursor_position.enable) { - pipe_ctx->visual_confirm_color.color_r_cr = color_value; - pipe_ctx->visual_confirm_color.color_g_y = 0; - pipe_ctx->visual_confirm_color.color_b_cb = 0; - } else { - pipe_ctx->visual_confirm_color.color_r_cr = 0; - pipe_ctx->visual_confirm_color.color_g_y = 0; - pipe_ctx->visual_confirm_color.color_b_cb = color_value; - } + get_cursor_visual_confirm_color(pipe_ctx, &(pipe_ctx->visual_confirm_color)); /* programming hardware */ if (pipe_ctx->plane_state) diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index 3992ad73165b..6d76dc110d38 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -55,7 +55,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.301" +#define DC_VER "3.2.306" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -227,6 +227,10 @@ struct dc_dmub_caps { uint8_t fams_ver; }; +struct dc_scl_caps { + bool sharpener_support; +}; + struct dc_caps { uint32_t max_streams; uint32_t max_links; @@ -292,6 +296,7 @@ struct dc_caps { bool sequential_ono; /* Conservative limit for DCC cases which require ODM4:1 to support*/ uint32_t dcc_plane_width_limit; + struct dc_scl_caps scl_caps; }; struct dc_bug_wa { @@ -463,6 +468,7 @@ struct dc_config { unsigned int enable_fpo_flicker_detection; bool disable_hbr_audio_dp2; bool consolidated_dpia_dp_lt; + bool set_pipe_unlock_order; }; enum visual_confirm { @@ -1253,7 +1259,6 @@ union surface_update_flags { uint32_t rotation_change:1; uint32_t swizzle_change:1; uint32_t scaling_change:1; - uint32_t clip_size_change: 1; uint32_t position_change:1; uint32_t in_transfer_func_change:1; uint32_t input_csc_change:1; @@ -1355,6 +1360,7 @@ struct dc_plane_state { enum mpcc_movable_cm_location mcm_location; struct dc_csc_transform cursor_csc_color_matrix; bool adaptive_sharpness_en; + int adaptive_sharpness_policy; int sharpness_level; enum linear_light_scaling linear_light_scaling; unsigned int sdr_white_level_nits; @@ -1461,6 +1467,7 @@ struct dc { struct dc_scratch_space current_state; struct dc_scratch_space new_state; struct dc_stream_state temp_stream; // Used so we don't need to allocate stream on the stack + bool pipes_to_unlock_first[MAX_PIPES]; /* Any of the pipes indicated here should be unlocked first */ } scratch; struct dml2_configuration_options dml2_options; @@ -1513,7 +1520,7 @@ struct dc_surface_update { * change cm2_params.component_settings: Full update * change cm2_params.cm2_luts: Fast update */ - struct dc_cm2_parameters *cm2_params; + const struct dc_cm2_parameters *cm2_params; const struct dc_csc_transform *cursor_csc_color_matrix; unsigned int sdr_white_level_nits; }; @@ -1770,7 +1777,6 @@ struct dc_link { bool dongle_mode_timing_override; bool blank_stream_on_ocs_change; bool read_dpcd204h_on_irq_hpd; - bool disable_assr_for_uhbr; } wa_flags; struct link_mst_stream_allocation_table mst_stream_alloc_table; @@ -1786,6 +1792,7 @@ struct dc_link { // BW ALLOCATON USB4 ONLY struct dc_dpia_bw_alloc dpia_bw_alloc_config; bool skip_implict_edp_power_control; + enum backlight_control_type backlight_control_type; }; /* Return an enumerated dc_link. diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c index 1e7de0f03290..f0417ee6fcf8 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.c @@ -519,7 +519,8 @@ void dc_dmub_srv_get_visual_confirm_color_cmd(struct dc *dc, struct pipe_ctx *pi union dmub_rb_cmd cmd = { 0 }; unsigned int panel_inst = 0; - if (!dc_get_edp_link_panel_inst(dc, pipe_ctx->stream->link, &panel_inst)) + if (!dc_get_edp_link_panel_inst(dc, pipe_ctx->stream->link, &panel_inst) && + dc->debug.visual_confirm == VISUAL_CONFIRM_DISABLE) return; memset(&cmd, 0, sizeof(cmd)); @@ -1294,6 +1295,8 @@ static void dc_dmub_srv_notify_idle(const struct dc *dc, bool allow_idle) memset(&new_signals, 0, sizeof(new_signals)); + new_signals.bits.allow_idle = 1; /* always set */ + if (dc->config.disable_ips == DMUB_IPS_ENABLE || dc->config.disable_ips == DMUB_IPS_DISABLE_DYNAMIC) { new_signals.bits.allow_pg = 1; @@ -1389,7 +1392,7 @@ static void dc_dmub_srv_exit_low_power_state(const struct dc *dc) */ dc_dmub_srv->needs_idle_wake = false; - if (prev_driver_signals.bits.allow_ips2 && + if ((prev_driver_signals.bits.allow_ips2 || prev_driver_signals.all == 0) && (!dc->debug.optimize_ips_handshake || ips_fw->signals.bits.ips2_commit || !ips_fw->signals.bits.in_idle)) { DC_LOG_IPS( @@ -1450,7 +1453,7 @@ static void dc_dmub_srv_exit_low_power_state(const struct dc *dc) } dc_dmub_srv_notify_idle(dc, false); - if (prev_driver_signals.bits.allow_ips1) { + if (prev_driver_signals.bits.allow_ips1 || prev_driver_signals.all == 0) { DC_LOG_IPS( "wait for IPS1 commit clear (ips1_commit=%u ips2_commit=%u)", ips_fw->signals.bits.ips1_commit, @@ -1862,3 +1865,81 @@ void dc_dmub_srv_fams2_passthrough_flip( dm_execute_dmub_cmd_list(dc->ctx, num_cmds, cmds, DM_DMUB_WAIT_TYPE_WAIT); } } + +bool dc_dmub_srv_ips_residency_cntl(struct dc_dmub_srv *dc_dmub_srv, bool start_measurement) +{ + bool result; + + if (!dc_dmub_srv || !dc_dmub_srv->dmub) + return false; + + result = dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__IPS_RESIDENCY, + start_measurement, NULL, DM_DMUB_WAIT_TYPE_WAIT); + + return result; +} + +void dc_dmub_srv_ips_query_residency_info(struct dc_dmub_srv *dc_dmub_srv, struct ips_residency_info *output) +{ + uint32_t i; + enum dmub_gpint_command command_code; + + if (!dc_dmub_srv || !dc_dmub_srv->dmub) + return; + + switch (output->ips_mode) { + case DMUB_IPS_MODE_IPS1_MAX: + command_code = DMUB_GPINT__GET_IPS1_HISTOGRAM_COUNTER; + break; + case DMUB_IPS_MODE_IPS2: + command_code = DMUB_GPINT__GET_IPS2_HISTOGRAM_COUNTER; + break; + case DMUB_IPS_MODE_IPS1_RCG: + command_code = DMUB_GPINT__GET_IPS1_RCG_HISTOGRAM_COUNTER; + break; + case DMUB_IPS_MODE_IPS1_ONO2_ON: + command_code = DMUB_GPINT__GET_IPS1_ONO2_ON_HISTOGRAM_COUNTER; + break; + default: + command_code = DMUB_GPINT__INVALID_COMMAND; + break; + } + + if (command_code == DMUB_GPINT__INVALID_COMMAND) + return; + + // send gpint commands and wait for ack + if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_PERCENT, + (uint16_t)(output->ips_mode), + &output->residency_percent, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) + output->residency_percent = 0; + + if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_ENTRY_COUNTER, + (uint16_t)(output->ips_mode), + &output->entry_counter, DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) + output->entry_counter = 0; + + if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_DURATION_US_LO, + (uint16_t)(output->ips_mode), + &output->total_active_time_us[0], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) + output->total_active_time_us[0] = 0; + if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_RESIDENCY_DURATION_US_HI, + (uint16_t)(output->ips_mode), + &output->total_active_time_us[1], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) + output->total_active_time_us[1] = 0; + + if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_INACTIVE_RESIDENCY_DURATION_US_LO, + (uint16_t)(output->ips_mode), + &output->total_inactive_time_us[0], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) + output->total_inactive_time_us[0] = 0; + if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, DMUB_GPINT__GET_IPS_INACTIVE_RESIDENCY_DURATION_US_HI, + (uint16_t)(output->ips_mode), + &output->total_inactive_time_us[1], DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) + output->total_inactive_time_us[1] = 0; + + // NUM_IPS_HISTOGRAM_BUCKETS = 16 + for (i = 0; i < 16; i++) + if (!dc_wake_and_execute_gpint(dc_dmub_srv->ctx, command_code, i, &output->histogram[i], + DM_DMUB_WAIT_TYPE_WAIT_WITH_REPLY)) + output->histogram[i] = 0; +} diff --git a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h index 42f0cb672d8b..10b48198b7a6 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dmub_srv.h @@ -209,4 +209,43 @@ void dc_dmub_srv_fams2_passthrough_flip( struct dc_stream_state *stream, struct dc_surface_update *srf_updates, int surface_count); + +/** + * struct ips_residency_info - struct containing info from dmub_ips_residency_stats + * + * @ips_mode: The mode of IPS that the follow stats appertain to + * @residency_percent: The percentage of time spent in given IPS mode in millipercent + * @entry_counter: The number of entries made in to this IPS state + * @total_active_time_us: uint32_t array of length 2 representing time in the given IPS mode + * in microseconds. Index 0 is lower 32 bits, index 1 is upper 32 bits. + * @total_inactive_time_us: uint32_t array of length 2 representing time outside the given IPS mode + * in microseconds. Index 0 is lower 32 bits, index 1 is upper 32 bits. + * @histogram: Histogram of given IPS state durations - bucket definitions in dmub_ips.c + */ +struct ips_residency_info { + enum dmub_ips_mode ips_mode; + unsigned int residency_percent; + unsigned int entry_counter; + unsigned int total_active_time_us[2]; + unsigned int total_inactive_time_us[2]; + unsigned int histogram[16]; +}; + +/** + * bool dc_dmub_srv_ips_residency_cntl() - Controls IPS residency measurement status + * + * @dc_dmub_srv: The DC DMUB service pointer + * @start_measurement: Describes whether to start or stop measurement + * + * Return: true if GPINT was sent successfully, false otherwise + */ +bool dc_dmub_srv_ips_residency_cntl(struct dc_dmub_srv *dc_dmub_srv, bool start_measurement); + +/** + * bool dc_dmub_srv_ips_query_residency_info() - Queries DMCUB for residency info + * + * @dc_dmub_srv: The DC DMUB service pointer + * @output: Output struct to copy the the residency info to + */ +void dc_dmub_srv_ips_query_residency_info(struct dc_dmub_srv *dc_dmub_srv, struct ips_residency_info *output); #endif /* _DMUB_DC_SRV_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index 41bd95e9177a..8dd6eb044829 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -1166,6 +1166,7 @@ struct dpcd_caps { int8_t branch_dev_name[6]; int8_t branch_hw_revision; int8_t branch_fw_revision[2]; + int8_t branch_vendor_specific_data[4]; bool allow_invalid_MSA_timing_param; bool panel_mode_edp; @@ -1191,6 +1192,7 @@ struct dpcd_caps { struct edp_psr_info psr_info; struct replay_info pr_info; + uint16_t edp_oled_emission_rate; }; union dpcd_sink_ext_caps { @@ -1204,7 +1206,7 @@ union dpcd_sink_ext_caps { uint8_t oled : 1; uint8_t reserved_2 : 1; uint8_t miniled : 1; - uint8_t reserved : 1; + uint8_t emission_output : 1; } bits; uint8_t raw; }; @@ -1358,6 +1360,9 @@ struct dp_trace { #ifndef DP_TUNNELING_IRQ #define DP_TUNNELING_IRQ (1 << 5) #endif +#ifndef DP_BRANCH_VENDOR_SPECIFIC_START +#define DP_BRANCH_VENDOR_SPECIFIC_START 0x50C +#endif /** USB4 DPCD BW Allocation Registers Chapter 10.7 **/ #ifndef DP_TUNNELING_CAPABILITIES #define DP_TUNNELING_CAPABILITIES 0xE000D /* 1.4a */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_plane.h b/drivers/gpu/drm/amd/display/dc/dc_plane.h index 44afcd989224..bd37ec82b42d 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_plane.h +++ b/drivers/gpu/drm/amd/display/dc/dc_plane.h @@ -26,7 +26,6 @@ #ifndef _DC_PLANE_H_ #define _DC_PLANE_H_ -#include "dc.h" #include "dc_hw_types.h" struct dc_plane_state *dc_create_plane_state(const struct dc *dc); diff --git a/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c b/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c index 603552dbd771..24aa9df892f3 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c +++ b/drivers/gpu/drm/amd/display/dc/dc_spl_translate.c @@ -38,6 +38,7 @@ static void populate_spltaps_from_taps(struct spl_taps *spl_scaling_quality, spl_scaling_quality->h_taps = scaling_quality->h_taps; spl_scaling_quality->v_taps_c = scaling_quality->v_taps_c; spl_scaling_quality->v_taps = scaling_quality->v_taps; + spl_scaling_quality->integer_scaling = scaling_quality->integer_scaling; } static void populate_taps_from_spltaps(struct scaling_taps *scaling_quality, const struct spl_taps *spl_scaling_quality) @@ -187,14 +188,14 @@ void translate_SPL_in_params_from_pipe_ctx(struct pipe_ctx *pipe_ctx, struct spl spl_in->h_active = pipe_ctx->plane_res.scl_data.h_active; spl_in->v_active = pipe_ctx->plane_res.scl_data.v_active; - spl_in->debug.sharpen_policy = (enum sharpen_policy)pipe_ctx->stream->ctx->dc->debug.sharpen_policy; + spl_in->sharpen_policy = (enum sharpen_policy)plane_state->adaptive_sharpness_policy; spl_in->debug.scale_to_sharpness_policy = (enum scale_to_sharpness_policy)pipe_ctx->stream->ctx->dc->debug.scale_to_sharpness_policy; /* Check if it is stream is in fullscreen and if its HDR. * Use this to determine sharpness levels */ - spl_in->is_fullscreen = dm_helpers_is_fullscreen(pipe_ctx->stream->ctx, pipe_ctx->stream); + spl_in->is_fullscreen = pipe_ctx->stream->sharpening_required; spl_in->is_hdr_on = dm_helpers_is_hdr_on(pipe_ctx->stream->ctx, pipe_ctx->stream); spl_in->sdr_white_level_nits = plane_state->sdr_white_level_nits; } diff --git a/drivers/gpu/drm/amd/display/dc/dc_state.h b/drivers/gpu/drm/amd/display/dc/dc_state.h index caa45db50232..db1e63a7d460 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_state.h +++ b/drivers/gpu/drm/amd/display/dc/dc_state.h @@ -26,7 +26,6 @@ #ifndef _DC_STATE_H_ #define _DC_STATE_H_ -#include "dc.h" #include "inc/core_status.h" struct dc_state *dc_state_create(struct dc *dc, struct dc_state_create_params *params); diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index 14ea47eda0c8..413970588a26 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -143,6 +143,7 @@ union stream_update_flags { uint32_t crtc_timing_adjust : 1; uint32_t fams_changed : 1; uint32_t scaler_sharpener : 1; + uint32_t sharpening_required : 1; } bits; uint32_t raw; @@ -310,6 +311,7 @@ struct dc_stream_state { struct luminance_data lumin_data; bool scaler_sharpener_update; + bool sharpening_required; }; #define ABM_LEVEL_IMMEDIATE_DISABLE 255 @@ -356,6 +358,7 @@ struct dc_stream_update { struct dc_cursor_position *cursor_position; bool *hw_cursor_req; bool *scaler_sharpener_update; + bool *sharpening_required; }; bool dc_is_stream_unchanged( diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index fd6dca735714..3401f4c9fb10 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -178,6 +178,8 @@ struct dc_panel_patch { unsigned int skip_avmute; unsigned int mst_start_top_delay; unsigned int remove_sink_ext_caps; + unsigned int disable_colorimetry; + uint8_t blankstream_before_otg_off; }; struct dc_edid_caps { @@ -921,6 +923,12 @@ struct display_endpoint_id { enum display_endpoint_type ep_type; }; +enum backlight_control_type { + BACKLIGHT_CONTROL_PWM = 0, + BACKLIGHT_CONTROL_VESA_AUX = 1, + BACKLIGHT_CONTROL_AMD_AUX = 2, +}; + #if defined(CONFIG_DRM_AMD_SECURE_DISPLAY) struct otg_phy_mux { uint8_t phy_output_num; diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c index 0b889004509a..d3e46c3cfa57 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn401/dcn401_dccg.c @@ -580,9 +580,6 @@ static void dccg401_set_dpstreamclk( int otg_inst, int dp_hpo_inst) { - /* set the dtbclk_p source */ - dccg401_set_dtbclk_p_src(dccg, src, otg_inst); - /* enabled to select one of the DTBCLKs for pipe */ if (src == REFCLK) dccg401_disable_dpstreamclk(dccg, dp_hpo_inst); @@ -805,33 +802,6 @@ static void dccg401_enable_symclk_se(struct dccg *dccg, uint32_t stream_enc_inst { struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); - switch (link_enc_inst) { - case 0: - REG_UPDATE(SYMCLKA_CLOCK_ENABLE, - SYMCLKA_CLOCK_ENABLE, 1); - if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL5, SYMCLKA_ROOT_GATE_DISABLE, 1); - break; - case 1: - REG_UPDATE(SYMCLKB_CLOCK_ENABLE, - SYMCLKB_CLOCK_ENABLE, 1); - if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL5, SYMCLKB_ROOT_GATE_DISABLE, 1); - break; - case 2: - REG_UPDATE(SYMCLKC_CLOCK_ENABLE, - SYMCLKC_CLOCK_ENABLE, 1); - if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL5, SYMCLKC_ROOT_GATE_DISABLE, 1); - break; - case 3: - REG_UPDATE(SYMCLKD_CLOCK_ENABLE, - SYMCLKD_CLOCK_ENABLE, 1); - if (dccg->ctx->dc->debug.root_clock_optimization.bits.symclk32_se) - REG_UPDATE(DCCG_GATE_DISABLE_CNTL5, SYMCLKD_ROOT_GATE_DISABLE, 1); - break; - } - switch (stream_enc_inst) { case 0: REG_UPDATE_2(SYMCLKA_CLOCK_ENABLE, @@ -864,37 +834,8 @@ static void dccg401_enable_symclk_se(struct dccg *dccg, uint32_t stream_enc_inst } } -/*get other front end connected to this backend*/ -static uint8_t dccg401_get_number_enabled_symclk_fe_connected_to_be(struct dccg *dccg, uint32_t link_enc_inst) -{ - uint8_t num_enabled_symclk_fe = 0; - uint32_t fe_clk_en[4] = {0}, be_clk_sel[4] = {0}; - struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); - uint8_t i; - - REG_GET_2(SYMCLKA_CLOCK_ENABLE, SYMCLKA_FE_EN, &fe_clk_en[0], - SYMCLKA_FE_SRC_SEL, &be_clk_sel[0]); - - REG_GET_2(SYMCLKB_CLOCK_ENABLE, SYMCLKB_FE_EN, &fe_clk_en[1], - SYMCLKB_FE_SRC_SEL, &be_clk_sel[1]); - - REG_GET_2(SYMCLKC_CLOCK_ENABLE, SYMCLKC_FE_EN, &fe_clk_en[2], - SYMCLKC_FE_SRC_SEL, &be_clk_sel[2]); - - REG_GET_2(SYMCLKD_CLOCK_ENABLE, SYMCLKD_FE_EN, &fe_clk_en[3], - SYMCLKD_FE_SRC_SEL, &be_clk_sel[3]); - - for (i = 0; i < ARRAY_SIZE(fe_clk_en); i++) { - if (fe_clk_en[i] && be_clk_sel[i] == link_enc_inst) - num_enabled_symclk_fe++; - } - - return num_enabled_symclk_fe; -} - static void dccg401_disable_symclk_se(struct dccg *dccg, uint32_t stream_enc_inst, uint32_t link_enc_inst) { - uint8_t num_enabled_symclk_fe = 0; struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); switch (stream_enc_inst) { @@ -919,31 +860,6 @@ static void dccg401_disable_symclk_se(struct dccg *dccg, uint32_t stream_enc_ins SYMCLKD_FE_SRC_SEL, 0); break; } - - /*check other enabled symclk fe connected to this be */ - num_enabled_symclk_fe = dccg401_get_number_enabled_symclk_fe_connected_to_be(dccg, link_enc_inst); - /*only turn off backend clk if other front ends attached to this backend are all off, - for mst, only turn off the backend if this is the last front end*/ - if (num_enabled_symclk_fe == 0) { - switch (link_enc_inst) { - case 0: - REG_UPDATE(SYMCLKA_CLOCK_ENABLE, - SYMCLKA_CLOCK_ENABLE, 0); - break; - case 1: - REG_UPDATE(SYMCLKB_CLOCK_ENABLE, - SYMCLKB_CLOCK_ENABLE, 0); - break; - case 2: - REG_UPDATE(SYMCLKC_CLOCK_ENABLE, - SYMCLKC_CLOCK_ENABLE, 0); - break; - case 3: - REG_UPDATE(SYMCLKD_CLOCK_ENABLE, - SYMCLKD_CLOCK_ENABLE, 0); - break; - } - } } static const struct dccg_funcs dccg401_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.c b/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.c index 8db9f7514466..889f314cac65 100644 --- a/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce60/dce60_resource.c @@ -717,7 +717,7 @@ static struct link_encoder *dce60_link_encoder_create( kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); int link_regs_id; - if (!enc110) + if (!enc110 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; link_regs_id = diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c index eaed5d1c398a..dcd2cdfe91eb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c @@ -365,23 +365,18 @@ bool cm_helper_translate_curve_to_hw_format(struct dc_context *ctx, region_start = -MAX_LOW_POINT; region_end = NUMBER_REGIONS - MAX_LOW_POINT; } else { - /* 11 segments - * segment is from 2^-10 to 2^1 + /* 13 segments + * segment is from 2^-12 to 2^0 * There are less than 256 points, for optimization */ - seg_distr[0] = 3; - seg_distr[1] = 4; - seg_distr[2] = 4; - seg_distr[3] = 4; - seg_distr[4] = 4; - seg_distr[5] = 4; - seg_distr[6] = 4; - seg_distr[7] = 4; - seg_distr[8] = 4; - seg_distr[9] = 4; - seg_distr[10] = 1; + const uint8_t SEG_COUNT = 12; - region_start = -10; + for (i = 0; i < SEG_COUNT; i++) + seg_distr[i] = 4; + + seg_distr[SEG_COUNT] = 1; + + region_start = -SEG_COUNT; region_end = 1; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_cm_common.c index f31f0e3abfc0..1e1038fb04e8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_cm_common.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_cm_common.c @@ -140,23 +140,18 @@ bool cm3_helper_translate_curve_to_hw_format( region_start = -MAX_LOW_POINT; region_end = NUMBER_REGIONS - MAX_LOW_POINT; } else { - /* 11 segments - * segment is from 2^-10 to 2^0 + /* 13 segments + * segment is from 2^-12 to 2^0 * There are less than 256 points, for optimization */ - seg_distr[0] = 3; - seg_distr[1] = 4; - seg_distr[2] = 4; - seg_distr[3] = 4; - seg_distr[4] = 4; - seg_distr[5] = 4; - seg_distr[6] = 4; - seg_distr[7] = 4; - seg_distr[8] = 4; - seg_distr[9] = 4; - seg_distr[10] = 1; + const uint8_t SEG_COUNT = 12; - region_start = -10; + for (i = 0; i < SEG_COUNT; i++) + seg_distr[i] = 4; + + seg_distr[SEG_COUNT] = 1; + + region_start = -SEG_COUNT; region_end = 1; } diff --git a/drivers/gpu/drm/amd/display/dc/dio/dcn314/dcn314_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dio/dcn314/dcn314_dio_stream_encoder.c index 5b343f745cf3..ae81451a3a72 100644 --- a/drivers/gpu/drm/amd/display/dc/dio/dcn314/dcn314_dio_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dio/dcn314/dcn314_dio_stream_encoder.c @@ -83,6 +83,15 @@ void enc314_disable_fifo(struct stream_encoder *enc) REG_UPDATE(DIG_FIFO_CTRL0, DIG_FIFO_ENABLE, 0); } +static bool enc314_is_fifo_enabled(struct stream_encoder *enc) +{ + struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); + uint32_t reset_val; + + REG_GET(DIG_FIFO_CTRL0, DIG_FIFO_ENABLE, &reset_val); + return (reset_val != 0); +} + void enc314_dp_set_odm_combine( struct stream_encoder *enc, bool odm_combine) @@ -468,6 +477,7 @@ static const struct stream_encoder_funcs dcn314_str_enc_funcs = { .enable_fifo = enc314_enable_fifo, .disable_fifo = enc314_disable_fifo, + .is_fifo_enabled = enc314_is_fifo_enabled, .set_input_mode = enc314_set_dig_input_mode, }; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20.c index 565f3c492477..0c8c4a080c50 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20.c @@ -785,12 +785,9 @@ static bool CalculatePrefetchSchedule( if (MyError) { *PrefetchBandwidth = 0; - TimeForFetchingMetaPTE = 0; - TimeForFetchingRowInVBlank = 0; *DestinationLinesToRequestVMInVBlank = 0; *DestinationLinesToRequestRowInVBlank = 0; *DestinationLinesForPrefetch = 0; - LinesToRequestPrefetchPixelData = 0; *VRatioPrefetchY = 0; *VRatioPrefetchC = 0; *RequiredPrefetchPixDataBW = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c index 9d6675ecc5f1..c935903b68e1 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_mode_vba_20v2.c @@ -845,12 +845,9 @@ static bool CalculatePrefetchSchedule( if (MyError) { *PrefetchBandwidth = 0; - TimeForFetchingMetaPTE = 0; - TimeForFetchingRowInVBlank = 0; *DestinationLinesToRequestVMInVBlank = 0; *DestinationLinesToRequestRowInVBlank = 0; *DestinationLinesForPrefetch = 0; - LinesToRequestPrefetchPixelData = 0; *VRatioPrefetchY = 0; *VRatioPrefetchC = 0; *RequiredPrefetchPixDataBW = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c index 4fce64a030b6..390c1a77fda6 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20.c @@ -443,8 +443,6 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib, blk_bytes = surf_linear ? 256 : get_blk_size_bytes((enum source_macro_tile_size) macro_tile_size); log2_blk_bytes = dml_log2((double) blk_bytes); - log2_blk_height = 0; - log2_blk_width = 0; // remember log rule // "+" in log is multiply @@ -491,8 +489,6 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib, - log2_meta_req_height; meta_req_width = 1 << log2_meta_req_width; meta_req_height = 1 << log2_meta_req_height; - log2_meta_row_height = 0; - meta_row_width_ub = 0; // the dimensions of a meta row are meta_row_width x meta_row_height in elements. // calculate upper bound of the meta_row_width diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c index 3fa9a5da02f6..843d6004258c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn20/display_rq_dlg_calc_20v2.c @@ -443,8 +443,6 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib, blk_bytes = surf_linear ? 256 : get_blk_size_bytes((enum source_macro_tile_size) macro_tile_size); log2_blk_bytes = dml_log2((double) blk_bytes); - log2_blk_height = 0; - log2_blk_width = 0; // remember log rule // "+" in log is multiply @@ -491,8 +489,6 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib, - log2_meta_req_height; meta_req_width = 1 << log2_meta_req_width; meta_req_height = 1 << log2_meta_req_height; - log2_meta_row_height = 0; - meta_row_width_ub = 0; // the dimensions of a meta row are meta_row_width x meta_row_height in elements. // calculate upper bound of the meta_row_width diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c index eb3ed965e48b..cd8cca651419 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_mode_vba_21.c @@ -1049,12 +1049,9 @@ static bool CalculatePrefetchSchedule( if (MyError) { *PrefetchBandwidth = 0; - TimeForFetchingMetaPTE = 0; - TimeForFetchingRowInVBlank = 0; *DestinationLinesToRequestVMInVBlank = 0; *DestinationLinesToRequestRowInVBlank = 0; *DestinationLinesForPrefetch = 0; - LinesToRequestPrefetchPixelData = 0; *VRatioPrefetchY = 0; *VRatioPrefetchC = 0; *RequiredPrefetchPixDataBWLuma = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c index 9e1c18b90805..5718000627b0 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c @@ -435,8 +435,6 @@ static void get_meta_and_pte_attr( blk_bytes = surf_linear ? 256 : get_blk_size_bytes((enum source_macro_tile_size) macro_tile_size); log2_blk_bytes = dml_log2((double) blk_bytes); - log2_blk_height = 0; - log2_blk_width = 0; // remember log rule // "+" in log is multiply @@ -485,8 +483,6 @@ static void get_meta_and_pte_attr( - log2_meta_req_height; meta_req_width = 1 << log2_meta_req_width; meta_req_height = 1 << log2_meta_req_height; - log2_meta_row_height = 0; - meta_row_width_ub = 0; // the dimensions of a meta row are meta_row_width x meta_row_height in elements. // calculate upper bound of the meta_row_width diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c index 1c10ba4dcdde..cee1b351e105 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_mode_vba_30.c @@ -1280,12 +1280,9 @@ static bool CalculatePrefetchSchedule( if (MyError) { *PrefetchBandwidth = 0; - TimeForFetchingMetaPTE = 0; - TimeForFetchingRowInVBlank = 0; *DestinationLinesToRequestVMInVBlank = 0; *DestinationLinesToRequestRowInVBlank = 0; *DestinationLinesForPrefetch = 0; - LinesToRequestPrefetchPixelData = 0; *VRatioPrefetchY = 0; *VRatioPrefetchC = 0; *RequiredPrefetchPixDataBWLuma = 0; @@ -1775,15 +1772,6 @@ static unsigned int CalculateVMAndRowBytes( *PixelPTEReqWidth = 32768.0 / BytePerPixel; *PTERequestSize = 64; FractionOfPTEReturnDrop = 0; - } else if (MacroTileSizeBytes == 4096) { - PixelPTEReqHeightPTEs = 1; - *PixelPTEReqHeight = MacroTileHeight; - *PixelPTEReqWidth = 8 * *MacroTileWidth; - *PTERequestSize = 64; - if (ScanDirection != dm_vert) - FractionOfPTEReturnDrop = 0; - else - FractionOfPTEReturnDrop = 7.0 / 8; } else if (GPUVMMinPageSize == 4 && MacroTileSizeBytes > 4096) { PixelPTEReqHeightPTEs = 16; *PixelPTEReqHeight = 16 * BlockHeight256Bytes; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c index b28fcc8608ff..76d3bb3c9155 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn30/display_rq_dlg_calc_30.c @@ -392,8 +392,6 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib, blk_bytes = surf_linear ? 256 : get_blk_size_bytes((enum source_macro_tile_size) macro_tile_size); log2_blk_bytes = dml_log2((double)blk_bytes); - log2_blk_height = 0; - log2_blk_width = 0; // remember log rule // "+" in log is multiply @@ -464,8 +462,6 @@ static void get_meta_and_pte_attr(struct display_mode_lib *mode_lib, - log2_meta_req_height; meta_req_width = 1 << log2_meta_req_width; meta_req_height = 1 << log2_meta_req_height; - log2_meta_row_height = 0; - meta_row_width_ub = 0; // the dimensions of a meta row are meta_row_width x meta_row_height in elements. // calculate upper bound of the meta_row_width diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c index 2b275e680379..f567a9023682 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c @@ -1444,12 +1444,9 @@ static bool CalculatePrefetchSchedule( if (MyError) { *PrefetchBandwidth = 0; - TimeForFetchingMetaPTE = 0; - TimeForFetchingRowInVBlank = 0; *DestinationLinesToRequestVMInVBlank = 0; *DestinationLinesToRequestRowInVBlank = 0; *DestinationLinesForPrefetch = 0; - LinesToRequestPrefetchPixelData = 0; *VRatioPrefetchY = 0; *VRatioPrefetchC = 0; *RequiredPrefetchPixDataBWLuma = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c index b57b095cd4a8..c46bda2141ac 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_rq_dlg_calc_31.c @@ -413,8 +413,6 @@ static void get_meta_and_pte_attr( log2_blk256_height = dml_log2((double) blk256_height); blk_bytes = surf_linear ? 256 : get_blk_size_bytes((enum source_macro_tile_size) macro_tile_size); log2_blk_bytes = dml_log2((double) blk_bytes); - log2_blk_height = 0; - log2_blk_width = 0; // remember log rule // "+" in log is multiply @@ -481,8 +479,6 @@ static void get_meta_and_pte_attr( log2_meta_req_width = log2_meta_req_bytes + 8 - log2_bytes_per_element - log2_meta_req_height; meta_req_width = 1 << log2_meta_req_width; meta_req_height = 1 << log2_meta_req_height; - log2_meta_row_height = 0; - meta_row_width_ub = 0; // the dimensions of a meta row are meta_row_width x meta_row_height in elements. // calculate upper bound of the meta_row_width diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c index debfa31583a6..5865e8fa2d8e 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_mode_vba_314.c @@ -1461,12 +1461,9 @@ static bool CalculatePrefetchSchedule( if (MyError) { *PrefetchBandwidth = 0; - TimeForFetchingMetaPTE = 0; - TimeForFetchingRowInVBlank = 0; *DestinationLinesToRequestVMInVBlank = 0; *DestinationLinesToRequestRowInVBlank = 0; *DestinationLinesForPrefetch = 0; - LinesToRequestPrefetchPixelData = 0; *VRatioPrefetchY = 0; *VRatioPrefetchC = 0; *RequiredPrefetchPixDataBWLuma = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c index 61b3bebf24c9..b7d2a0caec11 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn314/display_rq_dlg_calc_314.c @@ -501,8 +501,6 @@ static void get_meta_and_pte_attr( log2_blk256_height = dml_log2((double) blk256_height); blk_bytes = surf_linear ? 256 : get_blk_size_bytes((enum source_macro_tile_size) macro_tile_size); log2_blk_bytes = dml_log2((double) blk_bytes); - log2_blk_height = 0; - log2_blk_width = 0; // remember log rule // "+" in log is multiply @@ -569,8 +567,6 @@ static void get_meta_and_pte_attr( log2_meta_req_width = log2_meta_req_bytes + 8 - log2_bytes_per_element - log2_meta_req_height; meta_req_width = 1 << log2_meta_req_width; meta_req_height = 1 << log2_meta_req_height; - log2_meta_row_height = 0; - meta_row_width_ub = 0; // the dimensions of a meta row are meta_row_width x meta_row_height in elements. // calculate upper bound of the meta_row_width diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c index d92fb428ee96..86ac7d59fd32 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/display_mode_vba_util_32.c @@ -4097,12 +4097,9 @@ bool dml32_CalculatePrefetchSchedule( if (MyError) { *PrefetchBandwidth = 0; - TimeForFetchingMetaPTE = 0; - TimeForFetchingRowInVBlank = 0; *DestinationLinesToRequestVMInVBlank = 0; *DestinationLinesToRequestRowInVBlank = 0; *DestinationLinesForPrefetch = 0; - LinesToRequestPrefetchPixelData = 0; *VRatioPrefetchY = 0; *VRatioPrefetchC = 0; *RequiredPrefetchPixDataBWLuma = 0; diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c index a201dbb743d7..d9e63c4fdd95 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn351/dcn351_fpu.c @@ -204,8 +204,8 @@ struct _vcs_dpi_soc_bounding_box_st dcn3_51_soc = { .num_states = 8, .sr_exit_time_us = 28.0, .sr_enter_plus_exit_time_us = 30.0, - .sr_exit_z8_time_us = 250.0, - .sr_enter_plus_exit_z8_time_us = 350.0, + .sr_exit_z8_time_us = 263.0, + .sr_enter_plus_exit_z8_time_us = 363.0, .fclk_change_latency_us = 24.0, .usr_retraining_latency_us = 2, .writeback_latency_us = 12.0, diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c b/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c index d8bfc85e5dcd..88dc2b97e7bf 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dml1_display_rq_dlg_calc.c @@ -559,12 +559,11 @@ static void get_surf_rq_param( const struct _vcs_dpi_display_pipe_source_params_st *pipe_src_param, bool is_chroma) { - bool mode_422 = 0; unsigned int vp_width = 0; unsigned int vp_height = 0; unsigned int data_pitch = 0; unsigned int meta_pitch = 0; - unsigned int ppe = mode_422 ? 2 : 1; + unsigned int ppe = 1; bool surf_linear; bool surf_vert; unsigned int bytes_per_element; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c index 8697eac1e1f7..7a01a956e4bb 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_translation_helper.c @@ -1036,6 +1036,7 @@ void dml21_copy_clocks_to_dc_state(struct dml2_context *in_ctx, struct dc_state context->bw_ctx.bw.dcn.clk.p_state_change_support = in_ctx->v21.mode_programming.programming->uclk_pstate_supported; context->bw_ctx.bw.dcn.clk.dtbclk_en = in_ctx->v21.mode_programming.programming->min_clocks.dcn4x.dtbrefclk_khz > 0; context->bw_ctx.bw.dcn.clk.ref_dtbclk_khz = in_ctx->v21.mode_programming.programming->min_clocks.dcn4x.dtbrefclk_khz; + context->bw_ctx.bw.dcn.clk.socclk_khz = in_ctx->v21.mode_programming.programming->min_clocks.dcn4x.socclk_khz; } void dml21_extract_legacy_watermark_set(const struct dc *in_dc, struct dcn_watermarks *watermark, enum dml2_dchub_watermark_reg_set_index reg_set_idx, struct dml2_context *in_ctx) diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c index d35dd507cb9f..bbc28b9a15a3 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/dml21_wrapper.c @@ -13,11 +13,11 @@ static bool dml21_allocate_memory(struct dml2_context **dml_ctx) { - *dml_ctx = (struct dml2_context *)kzalloc(sizeof(struct dml2_context), GFP_KERNEL); + *dml_ctx = kzalloc(sizeof(struct dml2_context), GFP_KERNEL); if (!(*dml_ctx)) return false; - (*dml_ctx)->v21.dml_init.dml2_instance = (struct dml2_instance *)kzalloc(sizeof(struct dml2_instance), GFP_KERNEL); + (*dml_ctx)->v21.dml_init.dml2_instance = kzalloc(sizeof(struct dml2_instance), GFP_KERNEL); if (!((*dml_ctx)->v21.dml_init.dml2_instance)) return false; @@ -27,7 +27,7 @@ static bool dml21_allocate_memory(struct dml2_context **dml_ctx) (*dml_ctx)->v21.mode_support.display_config = &(*dml_ctx)->v21.display_config; (*dml_ctx)->v21.mode_programming.display_config = (*dml_ctx)->v21.mode_support.display_config; - (*dml_ctx)->v21.mode_programming.programming = (struct dml2_display_cfg_programming *)kzalloc(sizeof(struct dml2_display_cfg_programming), GFP_KERNEL); + (*dml_ctx)->v21.mode_programming.programming = kzalloc(sizeof(struct dml2_display_cfg_programming), GFP_KERNEL); if (!((*dml_ctx)->v21.mode_programming.programming)) return false; diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.c index 0aa4e4d343b0..3d41ffde91c1 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_dcn4.c @@ -159,6 +159,7 @@ static void create_phantom_stream_from_main_stream(struct dml2_stream_parameters phantom->timing.v_total = meta->v_total; phantom->timing.v_active = meta->v_active; phantom->timing.v_front_porch = meta->v_front_porch; + phantom->timing.v_blank_end = phantom->timing.v_total - phantom->timing.v_front_porch - phantom->timing.v_active; phantom->timing.vblank_nom = phantom->timing.v_total - phantom->timing.v_active; phantom->timing.drr_config.enabled = false; } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c index ab229e1598ae..714b5c39b7e6 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_core/dml2_core_utils.c @@ -425,6 +425,7 @@ static void create_phantom_stream_from_main_stream(struct dml2_stream_parameters phantom->timing.v_total = meta->v_total; phantom->timing.v_active = meta->v_active; phantom->timing.v_front_porch = meta->v_front_porch; + phantom->timing.v_blank_end = phantom->timing.v_total - phantom->timing.v_front_porch - phantom->timing.v_active; phantom->timing.vblank_nom = phantom->timing.v_total - phantom->timing.v_active; phantom->timing.drr_config.enabled = false; } diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.c b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.c index 1cf9015e854a..5a09dd298e6f 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml21/src/dml2_pmo/dml2_pmo_dcn4_fams2.c @@ -1798,6 +1798,7 @@ bool pmo_dcn4_fams2_init_for_pstate_support(struct dml2_pmo_init_for_pstate_supp } if (s->pmo_dcn4.num_pstate_candidates > 0) { + s->pmo_dcn4.pstate_strategy_candidates[s->pmo_dcn4.num_pstate_candidates - 1].allow_state_increase = true; s->pmo_dcn4.cur_pstate_candidate = -1; return true; } else { diff --git a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c index 866b0abcff1b..9190c1328d5b 100644 --- a/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml2/dml2_wrapper.c @@ -209,8 +209,6 @@ static bool optimize_configuration(struct dml2_context *dml2, struct dml2_wrappe p->cur_display_config->output.OutputEncoder[0], p->cur_mode_support_info->DSCEnabled[0]) - 1; if (odms_needed <= unused_dpps) { - unused_dpps -= odms_needed; - if (odms_needed == 1) { p->new_policy->ODMUse[0] = dml_odm_use_policy_combine_2to1; optimization_done = true; diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn20/dcn20_dpp.h b/drivers/gpu/drm/amd/display/dc/dpp/dcn20/dcn20_dpp.h index cd1706d301e7..f09cba8e29cc 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn20/dcn20_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn20/dcn20_dpp.h @@ -690,6 +690,7 @@ struct dcn20_dpp { int lb_memory_size; int lb_bits_per_entry; bool is_write_to_ram_a_safe; + bool dispclk_r_gate_disable; struct scaler_data scl_data; struct pwl_params pwl_data; }; diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.h b/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.h index b110f35ef66b..f236824126e9 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.h +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn30/dcn30_dpp.h @@ -572,6 +572,7 @@ struct dcn3_dpp { int lb_memory_size; int lb_bits_per_entry; bool is_write_to_ram_a_safe; + bool dispclk_r_gate_disable; struct scaler_data scl_data; struct pwl_params pwl_data; }; diff --git a/drivers/gpu/drm/amd/display/dc/dpp/dcn35/dcn35_dpp.c b/drivers/gpu/drm/amd/display/dc/dpp/dcn35/dcn35_dpp.c index 8473c694bfdc..62b7012cda43 100644 --- a/drivers/gpu/drm/amd/display/dc/dpp/dcn35/dcn35_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dpp/dcn35/dcn35_dpp.c @@ -50,13 +50,21 @@ void dpp35_dppclk_control( DPPCLK_RATE_CONTROL, dppclk_div, DPP_CLOCK_ENABLE, 1); else - REG_UPDATE_2(DPP_CONTROL, + if (dpp->dispclk_r_gate_disable) + REG_UPDATE_2(DPP_CONTROL, DPP_CLOCK_ENABLE, 1, DISPCLK_R_GATE_DISABLE, 1); + else + REG_UPDATE(DPP_CONTROL, + DPP_CLOCK_ENABLE, 1); } else - REG_UPDATE_2(DPP_CONTROL, + if (dpp->dispclk_r_gate_disable) + REG_UPDATE_2(DPP_CONTROL, DPP_CLOCK_ENABLE, 0, DISPCLK_R_GATE_DISABLE, 0); + else + REG_UPDATE(DPP_CONTROL, + DPP_CLOCK_ENABLE, 0); } void dpp35_program_bias_and_scale_fcnv( @@ -128,6 +136,10 @@ bool dpp35_construct( (const struct dcn3_dpp_mask *)(tf_mask)); dpp->base.funcs = &dcn35_dpp_funcs; + + // w/a for cursor memory stuck in LS by programming DISPCLK_R_GATE_DISABLE, limit w/a to some ASIC revs + if (dpp->base.ctx->asic_id.hw_internal_rev <= 0x10) + dpp->dispclk_r_gate_disable = true; return ret; } diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c index f344478e9bd4..b099989d9364 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c @@ -443,7 +443,6 @@ struct gpio *dal_gpio_create_irq( case GPIO_ID_GPIO_PAD: break; default: - id = GPIO_ID_HPD; ASSERT_CRITICAL(false); return NULL; } diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c index 4fbed0298adf..c31ec44ccd8c 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.c @@ -3142,9 +3142,10 @@ static void dce110_set_cursor_attribute(struct pipe_ctx *pipe_ctx) } bool dce110_set_backlight_level(struct pipe_ctx *pipe_ctx, - uint32_t backlight_pwm_u16_16, - uint32_t frame_ramp) + struct set_backlight_level_params *params) { + uint32_t backlight_pwm_u16_16 = params->backlight_pwm_u16_16; + uint32_t frame_ramp = params->frame_ramp; struct dc_link *link = pipe_ctx->stream->link; struct dc *dc = link->ctx->dc; struct abm *abm = pipe_ctx->stream_res.abm; @@ -3315,7 +3316,7 @@ void dce110_disable_link_output(struct dc_link *link, * from enable/disable link output and only call edp panel control * in enable_link_dp and disable_link_dp once. */ - if (dmcu != NULL && dmcu->funcs->lock_phy) + if (dmcu != NULL && dmcu->funcs->unlock_phy) dmcu->funcs->unlock_phy(dmcu); dc->link_srv->dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_DISABLE_LINK_PHY); } diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h index ed3cc3648e8e..06789ac3a224 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dce110/dce110_hwseq.h @@ -88,8 +88,7 @@ void dce110_edp_wait_for_hpd_ready( bool power_up); bool dce110_set_backlight_level(struct pipe_ctx *pipe_ctx, - uint32_t backlight_pwm_u16_16, - uint32_t frame_ramp); + struct set_backlight_level_params *params); void dce110_set_abm_immediate_disable(struct pipe_ctx *pipe_ctx); void dce110_set_pipe(struct pipe_ctx *pipe_ctx); void dce110_disable_link_output(struct dc_link *link, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c index a6a1db5ba8ba..681bb92c6069 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn10/dcn10_hwseq.c @@ -3453,7 +3453,6 @@ static bool dcn10_can_pipe_disable_cursor(struct pipe_ctx *pipe_ctx) r2 = test_pipe->plane_res.scl_data.recout; r2_r = r2.x + r2.width; r2_b = r2.y + r2.height; - split_pipe = test_pipe; /** * There is another half plane on same layer because of diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c index a80c08582932..05424a9af58b 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn20/dcn20_hwseq.c @@ -1458,8 +1458,12 @@ void dcn20_pipe_control_lock( } else { if (lock) pipe->stream_res.tg->funcs->lock(pipe->stream_res.tg); - else - pipe->stream_res.tg->funcs->unlock(pipe->stream_res.tg); + else { + if (dc->hwseq->funcs.perform_3dlut_wa_unlock) + dc->hwseq->funcs.perform_3dlut_wa_unlock(pipe); + else + pipe->stream_res.tg->funcs->unlock(pipe->stream_res.tg); + } } } @@ -1732,7 +1736,6 @@ static void dcn20_update_dchubp_dpp( if (pipe_ctx->update_flags.bits.scaler || plane_state->update_flags.bits.scaling_change || plane_state->update_flags.bits.position_change || - plane_state->update_flags.bits.clip_size_change || plane_state->update_flags.bits.per_pixel_alpha_change || pipe_ctx->stream->update_flags.bits.scaling) { pipe_ctx->plane_res.scl_data.lb_params.alpha_en = pipe_ctx->plane_state->per_pixel_alpha; @@ -1745,7 +1748,6 @@ static void dcn20_update_dchubp_dpp( if (pipe_ctx->update_flags.bits.viewport || (context == dc->current_state && plane_state->update_flags.bits.position_change) || (context == dc->current_state && plane_state->update_flags.bits.scaling_change) || - (context == dc->current_state && plane_state->update_flags.bits.clip_size_change) || (context == dc->current_state && pipe_ctx->stream->update_flags.bits.scaling)) { hubp->funcs->mem_program_viewport( @@ -2056,22 +2058,15 @@ void dcn20_program_front_end_for_ctx( */ for (i = 0; i < dc->res_pool->pipe_count; i++) { struct dc_stream_state *stream = dc->current_state->res_ctx.pipe_ctx[i].stream; + pipe = &dc->current_state->res_ctx.pipe_ctx[i]; if (context->res_ctx.pipe_ctx[i].update_flags.bits.disable && stream && - dc_state_get_pipe_subvp_type(dc->current_state, &dc->current_state->res_ctx.pipe_ctx[i]) == SUBVP_PHANTOM) { + dc_state_get_pipe_subvp_type(dc->current_state, pipe) == SUBVP_PHANTOM) { struct timing_generator *tg = dc->current_state->res_ctx.pipe_ctx[i].stream_res.tg; if (tg->funcs->enable_crtc) { - if (dc->hwss.blank_phantom) { - int main_pipe_width = 0, main_pipe_height = 0; - struct dc_stream_state *phantom_stream = dc_state_get_paired_subvp_stream(dc->current_state, dc->current_state->res_ctx.pipe_ctx[i].stream); - - if (phantom_stream) { - main_pipe_width = phantom_stream->dst.width; - main_pipe_height = phantom_stream->dst.height; - } - - dc->hwss.blank_phantom(dc, tg, main_pipe_width, main_pipe_height); + if (dc->hwseq->funcs.blank_pixel_data) { + dc->hwseq->funcs.blank_pixel_data(dc, pipe, true); } tg->funcs->enable_crtc(tg); } @@ -2255,9 +2250,9 @@ void dcn20_post_unlock_program_front_end( struct timing_generator *tg = pipe->stream_res.tg; - if (tg->funcs->get_double_buffer_pending) { + if (tg->funcs->get_optc_double_buffer_pending) { for (j = 0; j < TIMEOUT_FOR_PIPE_ENABLE_US / polling_interval_us - && tg->funcs->get_double_buffer_pending(tg); j++) + && tg->funcs->get_optc_double_buffer_pending(tg); j++) udelay(polling_interval_us); } } @@ -2771,7 +2766,6 @@ void dcn20_reset_back_end_for_pipe( struct pipe_ctx *pipe_ctx, struct dc_state *context) { - int i; struct dc_link *link = pipe_ctx->stream->link; const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); @@ -2838,19 +2832,16 @@ void dcn20_reset_back_end_for_pipe( } } - for (i = 0; i < dc->res_pool->pipe_count; i++) - if (&dc->current_state->res_ctx.pipe_ctx[i] == pipe_ctx) - break; - - if (i == dc->res_pool->pipe_count) - return; - /* * In case of a dangling plane, setting this to NULL unconditionally * causes failures during reset hw ctx where, if stream is NULL, * it is expected that the pipe_ctx pointers to pipes and plane are NULL. */ pipe_ctx->stream = NULL; + pipe_ctx->top_pipe = NULL; + pipe_ctx->bottom_pipe = NULL; + pipe_ctx->next_odm_pipe = NULL; + pipe_ctx->prev_odm_pipe = NULL; DC_LOG_DEBUG("Reset back end for pipe %d, tg:%d\n", pipe_ctx->pipe_idx, pipe_ctx->stream_res.tg->inst); } diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c index 1ea95f8d4cbc..630e05f32c80 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.c @@ -242,14 +242,15 @@ void dcn21_set_pipe(struct pipe_ctx *pipe_ctx) } bool dcn21_set_backlight_level(struct pipe_ctx *pipe_ctx, - uint32_t backlight_pwm_u16_16, - uint32_t frame_ramp) + struct set_backlight_level_params *params) { struct dc_context *dc = pipe_ctx->stream->ctx; struct abm *abm = pipe_ctx->stream_res.abm; struct timing_generator *tg = pipe_ctx->stream_res.tg; struct panel_cntl *panel_cntl = pipe_ctx->stream->link->panel_cntl; uint32_t otg_inst; + uint32_t backlight_pwm_u16_16 = params->backlight_pwm_u16_16; + uint32_t frame_ramp = params->frame_ramp; if (!abm || !tg || !panel_cntl) return false; @@ -257,7 +258,7 @@ bool dcn21_set_backlight_level(struct pipe_ctx *pipe_ctx, otg_inst = tg->inst; if (dc->dc->res_pool->dmcu) { - dce110_set_backlight_level(pipe_ctx, backlight_pwm_u16_16, frame_ramp); + dce110_set_backlight_level(pipe_ctx, params); return true; } diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.h index 9cee9bdb8de9..a7eaaa4596be 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn21/dcn21_hwseq.h @@ -50,8 +50,7 @@ void dcn21_PLAT_58856_wa(struct dc_state *context, void dcn21_set_pipe(struct pipe_ctx *pipe_ctx); void dcn21_set_abm_immediate_disable(struct pipe_ctx *pipe_ctx); bool dcn21_set_backlight_level(struct pipe_ctx *pipe_ctx, - uint32_t backlight_pwm_u16_16, - uint32_t frame_ramp); + struct set_backlight_level_params *params); bool dcn21_is_abm_supported(struct dc *dc, struct dc_state *context, struct dc_stream_state *stream); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c index bded33575493..e89ebfda4873 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.c @@ -245,6 +245,7 @@ static bool dcn30_set_mpc_shaper_3dlut(struct pipe_ctx *pipe_ctx, { struct dpp *dpp_base = pipe_ctx->plane_res.dpp; int mpcc_id = pipe_ctx->plane_res.hubp->inst; + struct dc *dc = pipe_ctx->stream->ctx->dc; struct mpc *mpc = pipe_ctx->stream_res.opp->ctx->dc->res_pool->mpc; bool result = false; int acquired_rmu = 0; @@ -283,8 +284,14 @@ static bool dcn30_set_mpc_shaper_3dlut(struct pipe_ctx *pipe_ctx, result = mpc->funcs->program_3dlut(mpc, &stream->lut3d_func->lut_3d, stream->lut3d_func->state.bits.rmu_mux_num); + if (!result) + DC_LOG_ERROR("%s: program_3dlut failed\n", __func__); + result = mpc->funcs->program_shaper(mpc, shaper_lut, stream->lut3d_func->state.bits.rmu_mux_num); + if (!result) + DC_LOG_ERROR("%s: program_shaper failed\n", __func__); + } else { // loop through the available mux and release the requested mpcc_id mpc->funcs->release_rmu(mpc, mpcc_id); @@ -486,7 +493,6 @@ bool dcn30_mmhubbub_warmup( } /*following is the original: warmup each DWB's mcif buffer*/ for (i = 0; i < num_dwb; i++) { - dwb = dc->res_pool->dwbc[wb_info[i].dwb_pipe_inst]; mcif_wb = dc->res_pool->mcif_wb[wb_info[i].dwb_pipe_inst]; /*warmup is for VM mode only*/ if (wb_info[i].mcif_buf_params.p_vmid == 0) @@ -1185,3 +1191,30 @@ void dcn30_prepare_bandwidth(struct dc *dc, if (!dc->clk_mgr->clks.fw_based_mclk_switching) dc_dmub_srv_p_state_delegate(dc, false, context); } + +void dcn30_wait_for_all_pending_updates(const struct pipe_ctx *pipe_ctx) +{ + struct timing_generator *tg = pipe_ctx->stream_res.tg; + bool pending_updates = false; + unsigned int i; + + if (tg && tg->funcs->is_tg_enabled(tg)) { + // Poll for 100ms maximum + for (i = 0; i < 100000; i++) { + pending_updates = false; + if (tg->funcs->get_optc_double_buffer_pending) + pending_updates |= tg->funcs->get_optc_double_buffer_pending(tg); + + if (tg->funcs->get_otg_double_buffer_pending) + pending_updates |= tg->funcs->get_otg_double_buffer_pending(tg); + + if (tg->funcs->get_pipe_update_pending && pipe_ctx->plane_state) + pending_updates |= tg->funcs->get_pipe_update_pending(tg); + + if (!pending_updates) + break; + + udelay(1); + } + } +} diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h index 6a153e7ce910..4b90b781c4f2 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_hwseq.h @@ -96,4 +96,6 @@ void dcn30_set_hubp_blank(const struct dc *dc, void dcn30_prepare_bandwidth(struct dc *dc, struct dc_state *context); +void dcn30_wait_for_all_pending_updates(const struct pipe_ctx *pipe_ctx); + #endif /* __DC_HWSS_DCN30_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c index 2a8dc40d2847..0e8d32e3dbae 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn30/dcn30_init.c @@ -108,7 +108,8 @@ static const struct hw_sequencer_funcs dcn30_funcs = { .set_disp_pattern_generator = dcn30_set_disp_pattern_generator, .get_dcc_en_bits = dcn10_get_dcc_en_bits, .update_visual_confirm_color = dcn10_update_visual_confirm_color, - .is_abm_supported = dcn21_is_abm_supported + .is_abm_supported = dcn21_is_abm_supported, + .wait_for_all_pending_updates = dcn30_wait_for_all_pending_updates, }; static const struct hwseq_private_funcs dcn30_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.c index 93e49d87a67c..780ce4c064aa 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.c @@ -107,6 +107,7 @@ static const struct hw_sequencer_funcs dcn301_funcs = { .optimize_pwr_state = dcn21_optimize_pwr_state, .exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state, .update_visual_confirm_color = dcn10_update_visual_confirm_color, + .wait_for_all_pending_updates = dcn30_wait_for_all_pending_updates, }; static const struct hwseq_private_funcs dcn301_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.h index 0bca48ccbfa2..a6e0115a53ee 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn301/dcn301_init.h @@ -23,8 +23,8 @@ * */ -#ifndef __DC_DCN30_INIT_H__ -#define __DC_DCN30_INIT_H__ +#ifndef __DC_DCN301_INIT_H__ +#define __DC_DCN301_INIT_H__ struct dc; diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c index 3d4b31bd9946..bfc78a42bc2a 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn31/dcn31_hwseq.c @@ -517,6 +517,11 @@ static void dcn31_reset_back_end_for_pipe( dc->hwss.set_abm_immediate_disable(pipe_ctx); + if ((!pipe_ctx->stream->dpms_off || pipe_ctx->stream->link->link_status.link_active) + && pipe_ctx->stream->sink && pipe_ctx->stream->sink->edid_caps.panel_patch.blankstream_before_otg_off) { + dc->hwss.blank_stream(pipe_ctx); + } + pipe_ctx->stream_res.tg->funcs->set_dsc_config( pipe_ctx->stream_res.tg, OPTC_DSC_DISABLED, 0, 0); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c index 4e93eeedfc1b..9b88eb72086d 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn314/dcn314_hwseq.c @@ -355,6 +355,20 @@ void dcn314_calculate_pix_rate_divider( } } +static bool dcn314_is_pipe_dig_fifo_on(struct pipe_ctx *pipe) +{ + return pipe && pipe->stream + // Check dig's otg instance. + && pipe->stream_res.stream_enc + && pipe->stream_res.stream_enc->funcs->dig_source_otg + && pipe->stream_res.tg->inst == pipe->stream_res.stream_enc->funcs->dig_source_otg(pipe->stream_res.stream_enc) + && pipe->stream->link && pipe->stream->link->link_enc + && pipe->stream->link->link_enc->funcs->is_dig_enabled + && pipe->stream->link->link_enc->funcs->is_dig_enabled(pipe->stream->link->link_enc) + && pipe->stream_res.stream_enc->funcs->is_fifo_enabled + && pipe->stream_res.stream_enc->funcs->is_fifo_enabled(pipe->stream_res.stream_enc); +} + void dcn314_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc_state *context, unsigned int current_pipe_idx) { unsigned int i; @@ -371,7 +385,11 @@ void dcn314_resync_fifo_dccg_dio(struct dce_hwseq *hws, struct dc *dc, struct dc if (pipe->top_pipe || pipe->prev_odm_pipe) continue; - if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal))) { + if (pipe->stream && (pipe->stream->dpms_off || dc_is_virtual_signal(pipe->stream->signal)) && + !pipe->stream->apply_seamless_boot_optimization && + !pipe->stream->apply_edp_fast_boot_optimization) { + if (dcn314_is_pipe_dig_fifo_on(pipe)) + continue; pipe->stream_res.tg->funcs->disable_crtc(pipe->stream_res.tg); reset_sync_context_for_pipe(dc, context, i); otg_disabled[i] = true; @@ -478,7 +496,7 @@ void dcn314_disable_link_output(struct dc_link *link, * from enable/disable link output and only call edp panel control * in enable_link_dp and disable_link_dp once. */ - if (dmcu != NULL && dmcu->funcs->lock_phy) + if (dmcu != NULL && dmcu->funcs->unlock_phy) dmcu->funcs->unlock_phy(dmcu); dc->link_srv->dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_DISABLE_LINK_PHY); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c index 2e8c9f738259..d7f8b2dcaa6b 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.c @@ -439,6 +439,7 @@ bool dcn32_set_mpc_shaper_3dlut( { struct dpp *dpp_base = pipe_ctx->plane_res.dpp; int mpcc_id = pipe_ctx->plane_res.hubp->inst; + struct dc *dc = pipe_ctx->stream->ctx->dc; struct mpc *mpc = pipe_ctx->stream_res.opp->ctx->dc->res_pool->mpc; bool result = false; @@ -458,13 +459,13 @@ bool dcn32_set_mpc_shaper_3dlut( if (stream->lut3d_func && stream->lut3d_func->state.bits.initialized == 1) { - result = mpc->funcs->program_3dlut(mpc, - &stream->lut3d_func->lut_3d, - mpcc_id); + result = mpc->funcs->program_3dlut(mpc, &stream->lut3d_func->lut_3d, mpcc_id); + if (!result) + DC_LOG_ERROR("%s: program_3dlut failed\n", __func__); - result = mpc->funcs->program_shaper(mpc, - shaper_lut, - mpcc_id); + result = mpc->funcs->program_shaper(mpc, shaper_lut, mpcc_id); + if (!result) + DC_LOG_ERROR("%s: program_shaper failed\n", __func__); } return result; @@ -1398,10 +1399,10 @@ void dcn32_disable_link_output(struct dc_link *link, link->phy_state.symclk_state = SYMCLK_OFF_TX_OFF; if (signal == SIGNAL_TYPE_EDP && - link->dc->hwss.edp_backlight_control && + link->dc->hwss.edp_power_control && !link->skip_implict_edp_power_control) link->dc->hwss.edp_power_control(link, false); - else if (dmcu != NULL && dmcu->funcs->lock_phy) + else if (dmcu != NULL && dmcu->funcs->unlock_phy) dmcu->funcs->unlock_phy(dmcu); dc->link_srv->dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_DISABLE_LINK_PHY); @@ -1698,52 +1699,6 @@ void dcn32_init_blank( hws->funcs.wait_for_blank_complete(opp); } -void dcn32_blank_phantom(struct dc *dc, - struct timing_generator *tg, - int width, - int height) -{ - struct dce_hwseq *hws = dc->hwseq; - enum dc_color_space color_space; - struct tg_color black_color = {0}; - struct output_pixel_processor *opp = NULL; - uint32_t num_opps, opp_id_src0, opp_id_src1; - uint32_t otg_active_width, otg_active_height; - uint32_t i; - - /* program opp dpg blank color */ - color_space = COLOR_SPACE_SRGB; - color_space_to_black_color(dc, color_space, &black_color); - - otg_active_width = width; - otg_active_height = height; - - /* get the OPTC source */ - tg->funcs->get_optc_source(tg, &num_opps, &opp_id_src0, &opp_id_src1); - ASSERT(opp_id_src0 < dc->res_pool->res_cap->num_opp); - - for (i = 0; i < dc->res_pool->res_cap->num_opp; i++) { - if (dc->res_pool->opps[i] != NULL && dc->res_pool->opps[i]->inst == opp_id_src0) { - opp = dc->res_pool->opps[i]; - break; - } - } - - if (opp && opp->funcs->opp_set_disp_pattern_generator) - opp->funcs->opp_set_disp_pattern_generator( - opp, - CONTROLLER_DP_TEST_PATTERN_SOLID_COLOR, - CONTROLLER_DP_COLOR_SPACE_UDEFINED, - COLOR_DEPTH_UNDEFINED, - &black_color, - otg_active_width, - otg_active_height, - 0); - - if (tg->funcs->is_tg_enabled(tg)) - hws->funcs.wait_for_blank_complete(opp); -} - /* phantom stream id's can change often, but can be identical between contexts. * This function checks for the condition the streams are identical to avoid * redundant pipe transitions. diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h index cac4a08b92a4..0303a5953673 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_hwseq.h @@ -119,11 +119,6 @@ void dcn32_init_blank( struct dc *dc, struct timing_generator *tg); -void dcn32_blank_phantom(struct dc *dc, - struct timing_generator *tg, - int width, - int height); - bool dcn32_is_pipe_topology_transition_seamless(struct dc *dc, const struct dc_state *cur_ctx, const struct dc_state *new_ctx); diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c index 3422b564ae98..dbcd2dfb19c1 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn32/dcn32_init.c @@ -117,10 +117,10 @@ static const struct hw_sequencer_funcs dcn32_funcs = { .update_phantom_vp_position = dcn32_update_phantom_vp_position, .update_dsc_pg = dcn32_update_dsc_pg, .apply_update_flags_for_phantom = dcn32_apply_update_flags_for_phantom, - .blank_phantom = dcn32_blank_phantom, .is_pipe_topology_transition_seamless = dcn32_is_pipe_topology_transition_seamless, .calculate_pix_rate_divider = dcn32_calculate_pix_rate_divider, .program_outstanding_updates = dcn32_program_outstanding_updates, + .wait_for_all_pending_updates = dcn30_wait_for_all_pending_updates, }; static const struct hwseq_private_funcs dcn32_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c index 2bbf1fef94fd..55dc5799e725 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn35/dcn35_init.c @@ -123,7 +123,6 @@ static const struct hw_sequencer_funcs dcn35_funcs = { .root_clock_control = dcn35_root_clock_control, .set_long_vtotal = dcn35_set_long_vblank, .calculate_pix_rate_divider = dcn32_calculate_pix_rate_divider, - .program_outstanding_updates = dcn32_program_outstanding_updates, }; static const struct hwseq_private_funcs dcn35_private_funcs = { diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c index d00822e8daa5..a93864b63d48 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn351/dcn351_init.c @@ -122,7 +122,6 @@ static const struct hw_sequencer_funcs dcn351_funcs = { .root_clock_control = dcn35_root_clock_control, .set_long_vtotal = dcn35_set_long_vblank, .calculate_pix_rate_divider = dcn32_calculate_pix_rate_divider, - .program_outstanding_updates = dcn32_program_outstanding_updates, .setup_hpo_hw_control = dcn35_setup_hpo_hw_control, }; diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c index 0b743669f23b..3c70f40bf047 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.c @@ -506,7 +506,7 @@ void dcn401_populate_mcm_luts(struct dc *dc, dcn401_get_mcm_lut_xable_from_pipe_ctx(dc, pipe_ctx, &shaper_xable, &lut3d_xable, &lut1d_xable); /* 1D LUT */ - if (mcm_luts.lut1d_func && lut3d_xable != MCM_LUT_DISABLE) { + if (mcm_luts.lut1d_func) { memset(&m_lut_params, 0, sizeof(m_lut_params)); if (mcm_luts.lut1d_func->type == TF_TYPE_HWPWL) m_lut_params.pwl = &mcm_luts.lut1d_func->pwl; @@ -521,7 +521,7 @@ void dcn401_populate_mcm_luts(struct dc *dc, mpc->funcs->populate_lut(mpc, MCM_LUT_1DLUT, m_lut_params, lut_bank_a, mpcc_id); } if (mpc->funcs->program_lut_mode) - mpc->funcs->program_lut_mode(mpc, MCM_LUT_1DLUT, lut1d_xable, lut_bank_a, mpcc_id); + mpc->funcs->program_lut_mode(mpc, MCM_LUT_1DLUT, lut1d_xable && m_lut_params.pwl, lut_bank_a, mpcc_id); } /* Shaper */ @@ -669,11 +669,17 @@ bool dcn401_set_mcm_luts(struct pipe_ctx *pipe_ctx, { struct dpp *dpp_base = pipe_ctx->plane_res.dpp; int mpcc_id = pipe_ctx->plane_res.hubp->inst; - struct mpc *mpc = pipe_ctx->stream_res.opp->ctx->dc->res_pool->mpc; + struct dc *dc = pipe_ctx->stream_res.opp->ctx->dc; + struct mpc *mpc = dc->res_pool->mpc; bool result; const struct pwl_params *lut_params = NULL; bool rval; + if (plane_state->mcm_luts.lut3d_data.lut3d_src == DC_CM2_TRANSFER_FUNC_SOURCE_VIDMEM) { + dcn401_populate_mcm_luts(dc, pipe_ctx, plane_state->mcm_luts, plane_state->lut_bank_a); + return true; + } + mpc->funcs->set_movable_cm_location(mpc, MPCC_MOVABLE_CM_LOCATION_BEFORE, mpcc_id); pipe_ctx->plane_state->mcm_location = MPCC_MOVABLE_CM_LOCATION_BEFORE; // 1D LUT @@ -844,6 +850,13 @@ enum dc_status dcn401_enable_stream_timing( odm_slice_width, last_odm_slice_width); } + /* set DTBCLK_P */ + if (dc->res_pool->dccg->funcs->set_dtbclk_p_src) { + if (dc_is_dp_signal(stream->signal) || dc_is_virtual_signal(stream->signal)) { + dc->res_pool->dccg->funcs->set_dtbclk_p_src(dc->res_pool->dccg, DPREFCLK, pipe_ctx->stream_res.tg->inst); + } + } + /* HW program guide assume display already disable * by unplug sequence. OTG assume stop. */ @@ -1004,8 +1017,6 @@ void dcn401_enable_stream(struct pipe_ctx *pipe_ctx) dccg->funcs->enable_symclk32_se(dccg, dp_hpo_inst, phyd32clk); } else { - /* need to set DTBCLK_P source to DPREFCLK for DP8B10B */ - dccg->funcs->set_dtbclk_p_src(dccg, DPREFCLK, tg->inst); dccg->funcs->enable_symclk_se(dccg, stream_enc->stream_enc_inst, link_enc->transmitter - TRANSMITTER_UNIPHY_A); } @@ -1063,7 +1074,6 @@ static bool dcn401_can_pipe_disable_cursor(struct pipe_ctx *pipe_ctx) r2 = test_pipe->plane_res.scl_data.recout; r2_r = r2.x + r2.width; r2_b = r2.y + r2.height; - split_pipe = test_pipe; /** * There is another half plane on same layer because of @@ -1097,6 +1107,58 @@ void adjust_hotspot_between_slices_for_2x_magnify(uint32_t cursor_width, struct } } +static void disable_link_output_symclk_on_tx_off(struct dc_link *link, enum dp_link_encoding link_encoding) +{ + struct dc *dc = link->ctx->dc; + struct pipe_ctx *pipe_ctx = NULL; + uint8_t i; + + for (i = 0; i < MAX_PIPES; i++) { + pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; + if (pipe_ctx->stream && pipe_ctx->stream->link == link && pipe_ctx->top_pipe == NULL) { + pipe_ctx->clock_source->funcs->program_pix_clk( + pipe_ctx->clock_source, + &pipe_ctx->stream_res.pix_clk_params, + link_encoding, + &pipe_ctx->pll_settings); + break; + } + } +} + +void dcn401_disable_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal) +{ + struct dc *dc = link->ctx->dc; + const struct link_hwss *link_hwss = get_link_hwss(link, link_res); + struct dmcu *dmcu = dc->res_pool->dmcu; + + if (signal == SIGNAL_TYPE_EDP && + link->dc->hwss.edp_backlight_control && + !link->skip_implict_edp_power_control) + link->dc->hwss.edp_backlight_control(link, false); + else if (dmcu != NULL && dmcu->funcs->lock_phy) + dmcu->funcs->lock_phy(dmcu); + + if (dc_is_tmds_signal(signal) && link->phy_state.symclk_ref_cnts.otg > 0) { + disable_link_output_symclk_on_tx_off(link, DP_UNKNOWN_ENCODING); + link->phy_state.symclk_state = SYMCLK_ON_TX_OFF; + } else { + link_hwss->disable_link_output(link, link_res, signal); + link->phy_state.symclk_state = SYMCLK_OFF_TX_OFF; + } + + if (signal == SIGNAL_TYPE_EDP && + link->dc->hwss.edp_backlight_control && + !link->skip_implict_edp_power_control) + link->dc->hwss.edp_power_control(link, false); + else if (dmcu != NULL && dmcu->funcs->lock_phy) + dmcu->funcs->unlock_phy(dmcu); + + dc->link_srv->dp_trace_source_sequence(link, DPCD_SOURCE_SEQ_AFTER_DISABLE_LINK_PHY); +} + void dcn401_set_cursor_position(struct pipe_ctx *pipe_ctx) { struct dc_cursor_position pos_cpy = pipe_ctx->stream->cursor_position; @@ -1669,7 +1731,7 @@ void dcn401_hardware_release(struct dc *dc) } } -void dcn401_wait_for_det_buffer_update(struct dc *dc, struct dc_state *context, struct pipe_ctx *otg_master) +void dcn401_wait_for_det_buffer_update_under_otg_master(struct dc *dc, struct dc_state *context, struct pipe_ctx *otg_master) { struct pipe_ctx *opp_heads[MAX_PIPES]; struct pipe_ctx *dpp_pipes[MAX_PIPES]; @@ -1695,6 +1757,9 @@ void dcn401_wait_for_det_buffer_update(struct dc *dc, struct dc_state *context, hubbub->funcs->wait_for_det_update) hubbub->funcs->wait_for_det_update(hubbub, dpp_pipe->plane_res.hubp->inst); } + } else { + if (hubbub && opp_heads[slice_idx]->plane_res.hubp && hubbub->funcs->wait_for_det_update) + hubbub->funcs->wait_for_det_update(hubbub, opp_heads[slice_idx]->plane_res.hubp->inst); } } } @@ -1705,7 +1770,6 @@ void dcn401_interdependent_update_lock(struct dc *dc, unsigned int i = 0; struct pipe_ctx *pipe = NULL; struct timing_generator *tg = NULL; - bool pipe_unlocked[MAX_PIPES] = {0}; if (lock) { for (i = 0; i < dc->res_pool->pipe_count; i++) { @@ -1719,48 +1783,91 @@ void dcn401_interdependent_update_lock(struct dc *dc, dc->hwss.pipe_control_lock(dc, pipe, true); } } else { - /* Unlock pipes based on the change in DET allocation instead of pipe index - * Prevents over allocation of DET during unlock process - * e.g. 2 pipe config with different streams with a max of 20 DET segments - * Before: After: - * - Pipe0: 10 DET segments - Pipe0: 12 DET segments - * - Pipe1: 10 DET segments - Pipe1: 8 DET segments - * If Pipe0 gets updated first, 22 DET segments will be allocated - */ + /* Need to free DET being used first and have pipe update, then unlock the remaining pipes*/ for (i = 0; i < dc->res_pool->pipe_count; i++) { pipe = &context->res_ctx.pipe_ctx[i]; tg = pipe->stream_res.tg; - int current_pipe_idx = i; if (!resource_is_pipe_type(pipe, OTG_MASTER) || !tg->funcs->is_tg_enabled(tg) || dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM) { - pipe_unlocked[i] = true; continue; } - // If the same stream exists in old context, ensure the OTG_MASTER pipes for the same stream get compared - struct pipe_ctx *old_otg_master = resource_get_otg_master_for_stream(&dc->current_state->res_ctx, pipe->stream); - - if (old_otg_master) - current_pipe_idx = old_otg_master->pipe_idx; - if (resource_calculate_det_for_stream(context, pipe) < - resource_calculate_det_for_stream(dc->current_state, &dc->current_state->res_ctx.pipe_ctx[current_pipe_idx])) { + if (dc->scratch.pipes_to_unlock_first[i]) { + struct pipe_ctx *old_pipe = &dc->current_state->res_ctx.pipe_ctx[i]; dc->hwss.pipe_control_lock(dc, pipe, false); - pipe_unlocked[i] = true; - dcn401_wait_for_det_buffer_update(dc, context, pipe); + /* Assumes pipe of the same index in current_state is also an OTG_MASTER pipe*/ + dcn401_wait_for_det_buffer_update_under_otg_master(dc, dc->current_state, old_pipe); } } + /* Unlocking the rest of the pipes */ for (i = 0; i < dc->res_pool->pipe_count; i++) { - if (pipe_unlocked[i]) + if (dc->scratch.pipes_to_unlock_first[i]) continue; + pipe = &context->res_ctx.pipe_ctx[i]; + tg = pipe->stream_res.tg; + if (!resource_is_pipe_type(pipe, OTG_MASTER) || + !tg->funcs->is_tg_enabled(tg) || + dc_state_get_pipe_subvp_type(context, pipe) == SUBVP_PHANTOM) { + continue; + } + dc->hwss.pipe_control_lock(dc, pipe, false); } } } +void dcn401_perform_3dlut_wa_unlock(struct pipe_ctx *pipe_ctx) +{ + /* If 3DLUT FL is enabled and 3DLUT is in use, follow the workaround sequence for pipe unlock to make sure that + * HUBP will properly fetch 3DLUT contents after unlock. + * + * This is meant to work around a known HW issue where VREADY will cancel the pending 3DLUT_ENABLE signal regardless + * of whether OTG lock is currently being held or not. + */ + struct pipe_ctx *wa_pipes[MAX_PIPES] = { NULL }; + struct pipe_ctx *odm_pipe, *mpc_pipe; + int i, wa_pipe_ct = 0; + + for (odm_pipe = pipe_ctx; odm_pipe != NULL; odm_pipe = odm_pipe->next_odm_pipe) { + for (mpc_pipe = odm_pipe; mpc_pipe != NULL; mpc_pipe = mpc_pipe->bottom_pipe) { + if (mpc_pipe->plane_state && mpc_pipe->plane_state->mcm_luts.lut3d_data.lut3d_src + == DC_CM2_TRANSFER_FUNC_SOURCE_VIDMEM + && mpc_pipe->plane_state->mcm_shaper_3dlut_setting + == DC_CM2_SHAPER_3DLUT_SETTING_ENABLE_SHAPER_3DLUT) { + wa_pipes[wa_pipe_ct++] = mpc_pipe; + } + } + } + + if (wa_pipe_ct > 0) { + if (pipe_ctx->stream_res.tg->funcs->set_vupdate_keepout) + pipe_ctx->stream_res.tg->funcs->set_vupdate_keepout(pipe_ctx->stream_res.tg, true); + + for (i = 0; i < wa_pipe_ct; ++i) { + if (wa_pipes[i]->plane_res.hubp->funcs->hubp_enable_3dlut_fl) + wa_pipes[i]->plane_res.hubp->funcs->hubp_enable_3dlut_fl(wa_pipes[i]->plane_res.hubp, true); + } + + pipe_ctx->stream_res.tg->funcs->unlock(pipe_ctx->stream_res.tg); + if (pipe_ctx->stream_res.tg->funcs->wait_update_lock_status) + pipe_ctx->stream_res.tg->funcs->wait_update_lock_status(pipe_ctx->stream_res.tg, false); + + for (i = 0; i < wa_pipe_ct; ++i) { + if (wa_pipes[i]->plane_res.hubp->funcs->hubp_enable_3dlut_fl) + wa_pipes[i]->plane_res.hubp->funcs->hubp_enable_3dlut_fl(wa_pipes[i]->plane_res.hubp, true); + } + + if (pipe_ctx->stream_res.tg->funcs->set_vupdate_keepout) + pipe_ctx->stream_res.tg->funcs->set_vupdate_keepout(pipe_ctx->stream_res.tg, false); + } else { + pipe_ctx->stream_res.tg->funcs->unlock(pipe_ctx->stream_res.tg); + } +} + void dcn401_program_outstanding_updates(struct dc *dc, struct dc_state *context) { @@ -1770,3 +1877,125 @@ void dcn401_program_outstanding_updates(struct dc *dc, if (hubbub->funcs->program_compbuf_segments) hubbub->funcs->program_compbuf_segments(hubbub, context->bw_ctx.bw.dcn.arb_regs.compbuf_size, true); } + +void dcn401_reset_back_end_for_pipe( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + struct dc_state *context) +{ + struct dc_link *link = pipe_ctx->stream->link; + const struct link_hwss *link_hwss = get_link_hwss(link, &pipe_ctx->link_res); + + DC_LOGGER_INIT(dc->ctx->logger); + if (pipe_ctx->stream_res.stream_enc == NULL) { + pipe_ctx->stream = NULL; + return; + } + + /* DPMS may already disable or */ + /* dpms_off status is incorrect due to fastboot + * feature. When system resume from S4 with second + * screen only, the dpms_off would be true but + * VBIOS lit up eDP, so check link status too. + */ + if (!pipe_ctx->stream->dpms_off || link->link_status.link_active) + dc->link_srv->set_dpms_off(pipe_ctx); + else if (pipe_ctx->stream_res.audio) + dc->hwss.disable_audio_stream(pipe_ctx); + + /* free acquired resources */ + if (pipe_ctx->stream_res.audio) { + /*disable az_endpoint*/ + pipe_ctx->stream_res.audio->funcs->az_disable(pipe_ctx->stream_res.audio); + + /*free audio*/ + if (dc->caps.dynamic_audio == true) { + /*we have to dynamic arbitrate the audio endpoints*/ + /*we free the resource, need reset is_audio_acquired*/ + update_audio_usage(&dc->current_state->res_ctx, dc->res_pool, + pipe_ctx->stream_res.audio, false); + pipe_ctx->stream_res.audio = NULL; + } + } + + /* by upper caller loop, parent pipe: pipe0, will be reset last. + * back end share by all pipes and will be disable only when disable + * parent pipe. + */ + if (pipe_ctx->top_pipe == NULL) { + + dc->hwss.set_abm_immediate_disable(pipe_ctx); + + pipe_ctx->stream_res.tg->funcs->disable_crtc(pipe_ctx->stream_res.tg); + + pipe_ctx->stream_res.tg->funcs->enable_optc_clock(pipe_ctx->stream_res.tg, false); + if (pipe_ctx->stream_res.tg->funcs->set_odm_bypass) + pipe_ctx->stream_res.tg->funcs->set_odm_bypass( + pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing); + + if (pipe_ctx->stream_res.tg->funcs->set_drr) + pipe_ctx->stream_res.tg->funcs->set_drr( + pipe_ctx->stream_res.tg, NULL); + /* TODO - convert symclk_ref_cnts for otg to a bit map to solve + * the case where the same symclk is shared across multiple otg + * instances + */ + if (dc_is_hdmi_tmds_signal(pipe_ctx->stream->signal)) + link->phy_state.symclk_ref_cnts.otg = 0; + if (link->phy_state.symclk_state == SYMCLK_ON_TX_OFF) { + link_hwss->disable_link_output(link, + &pipe_ctx->link_res, pipe_ctx->stream->signal); + link->phy_state.symclk_state = SYMCLK_OFF_TX_OFF; + } + + /* reset DTBCLK_P */ + if (dc->res_pool->dccg->funcs->set_dtbclk_p_src) + dc->res_pool->dccg->funcs->set_dtbclk_p_src(dc->res_pool->dccg, REFCLK, pipe_ctx->stream_res.tg->inst); + } + +/* + * In case of a dangling plane, setting this to NULL unconditionally + * causes failures during reset hw ctx where, if stream is NULL, + * it is expected that the pipe_ctx pointers to pipes and plane are NULL. + */ + pipe_ctx->stream = NULL; + pipe_ctx->top_pipe = NULL; + pipe_ctx->bottom_pipe = NULL; + pipe_ctx->next_odm_pipe = NULL; + pipe_ctx->prev_odm_pipe = NULL; + DC_LOG_DEBUG("Reset back end for pipe %d, tg:%d\n", + pipe_ctx->pipe_idx, pipe_ctx->stream_res.tg->inst); +} + +void dcn401_reset_hw_ctx_wrap( + struct dc *dc, + struct dc_state *context) +{ + int i; + struct dce_hwseq *hws = dc->hwseq; + + /* Reset Back End*/ + for (i = dc->res_pool->pipe_count - 1; i >= 0 ; i--) { + struct pipe_ctx *pipe_ctx_old = + &dc->current_state->res_ctx.pipe_ctx[i]; + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (!pipe_ctx_old->stream) + continue; + + if (pipe_ctx_old->top_pipe || pipe_ctx_old->prev_odm_pipe) + continue; + + if (!pipe_ctx->stream || + pipe_need_reprogram(pipe_ctx_old, pipe_ctx)) { + struct clock_source *old_clk = pipe_ctx_old->clock_source; + + if (hws->funcs.reset_back_end_for_pipe) + hws->funcs.reset_back_end_for_pipe(dc, pipe_ctx_old, dc->current_state); + if (hws->funcs.enable_stream_gating) + hws->funcs.enable_stream_gating(dc, pipe_ctx_old); + if (old_clk) + old_clk->funcs->cs_power_down(old_clk); + } + } +} diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h index a27e62081685..28a513dfc005 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_hwseq.h @@ -55,6 +55,10 @@ void dcn401_populate_mcm_luts(struct dc *dc, bool lut_bank_a); void dcn401_setup_hpo_hw_control(const struct dce_hwseq *hws, bool enable); +void dcn401_disable_link_output(struct dc_link *link, + const struct link_resource *link_res, + enum signal_type signal); + void dcn401_set_cursor_position(struct pipe_ctx *pipe_ctx); bool dcn401_apply_idle_power_optimizations(struct dc *dc, bool enable); @@ -81,7 +85,16 @@ void dcn401_hardware_release(struct dc *dc); void dcn401_update_odm(struct dc *dc, struct dc_state *context, struct pipe_ctx *otg_master); void adjust_hotspot_between_slices_for_2x_magnify(uint32_t cursor_width, struct dc_cursor_position *pos_cpy); -void dcn401_wait_for_det_buffer_update(struct dc *dc, struct dc_state *context, struct pipe_ctx *otg_master); +void dcn401_wait_for_det_buffer_update_under_otg_master(struct dc *dc, struct dc_state *context, struct pipe_ctx *otg_master); void dcn401_interdependent_update_lock(struct dc *dc, struct dc_state *context, bool lock); void dcn401_program_outstanding_updates(struct dc *dc, struct dc_state *context); +void dcn401_reset_back_end_for_pipe( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + struct dc_state *context); +void dcn401_reset_hw_ctx_wrap( + struct dc *dc, + struct dc_state *context); +void dcn401_perform_3dlut_wa_unlock(struct pipe_ctx *pipe_ctx); + #endif /* __DC_HWSS_DCN401_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c index a2ca07235c83..c73305e57d39 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c +++ b/drivers/gpu/drm/amd/display/dc/hwss/dcn401/dcn401_init.c @@ -84,7 +84,7 @@ static const struct hw_sequencer_funcs dcn401_funcs = { .enable_lvds_link_output = dce110_enable_lvds_link_output, .enable_tmds_link_output = dce110_enable_tmds_link_output, .enable_dp_link_output = dce110_enable_dp_link_output, - .disable_link_output = dcn32_disable_link_output, + .disable_link_output = dcn401_disable_link_output, .set_disp_pattern_generator = dcn30_set_disp_pattern_generator, .get_dcc_en_bits = dcn10_get_dcc_en_bits, .enable_phantom_streams = dcn32_enable_phantom_streams, @@ -93,13 +93,13 @@ static const struct hw_sequencer_funcs dcn401_funcs = { .update_phantom_vp_position = dcn32_update_phantom_vp_position, .update_dsc_pg = dcn32_update_dsc_pg, .apply_update_flags_for_phantom = dcn32_apply_update_flags_for_phantom, - .blank_phantom = dcn32_blank_phantom, .wait_for_dcc_meta_propagation = dcn401_wait_for_dcc_meta_propagation, .is_pipe_topology_transition_seamless = dcn32_is_pipe_topology_transition_seamless, .fams2_global_control_lock = dcn401_fams2_global_control_lock, .fams2_update_config = dcn401_fams2_update_config, .fams2_global_control_lock_fast = dcn401_fams2_global_control_lock_fast, .program_outstanding_updates = dcn401_program_outstanding_updates, + .wait_for_all_pending_updates = dcn30_wait_for_all_pending_updates, }; static const struct hwseq_private_funcs dcn401_private_funcs = { @@ -111,7 +111,7 @@ static const struct hwseq_private_funcs dcn401_private_funcs = { .power_down = dce110_power_down, .enable_display_power_gating = dcn10_dummy_display_power_gating, .blank_pixel_data = dcn20_blank_pixel_data, - .reset_hw_ctx_wrap = dcn20_reset_hw_ctx_wrap, + .reset_hw_ctx_wrap = dcn401_reset_hw_ctx_wrap, .enable_stream_timing = dcn401_enable_stream_timing, .edp_backlight_control = dce110_edp_backlight_control, .setup_vupdate_interrupt = dcn20_setup_vupdate_interrupt, @@ -136,8 +136,9 @@ static const struct hwseq_private_funcs dcn401_private_funcs = { .update_mall_sel = dcn32_update_mall_sel, .calculate_dccg_k1_k2_values = NULL, .apply_single_controller_ctx_to_hw = dce110_apply_single_controller_ctx_to_hw, - .reset_back_end_for_pipe = dcn20_reset_back_end_for_pipe, + .reset_back_end_for_pipe = dcn401_reset_back_end_for_pipe, .populate_mcm_luts = NULL, + .perform_3dlut_wa_unlock = dcn401_perform_3dlut_wa_unlock, }; void dcn401_hw_sequencer_init_functions(struct dc *dc) diff --git a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h index ac9205625623..1df17c54f3a9 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer.h @@ -174,6 +174,11 @@ union block_sequence_params { struct fams2_global_control_lock_fast_params fams2_global_control_lock_fast_params; }; +struct set_backlight_level_params { + uint32_t backlight_pwm_u16_16; + uint32_t frame_ramp; +}; + enum block_sequence_func { DMUB_SUBVP_PIPE_CONTROL_LOCK_FAST = 0, OPTC_PIPE_CONTROL_LOCK, @@ -365,8 +370,7 @@ struct hw_sequencer_funcs { void (*clear_status_bits)(struct dc *dc, unsigned int mask); bool (*set_backlight_level)(struct pipe_ctx *pipe_ctx, - uint32_t backlight_pwm_u16_16, - uint32_t frame_ramp); + struct set_backlight_level_params *params); void (*set_abm_immediate_disable)(struct pipe_ctx *pipe_ctx); @@ -462,6 +466,7 @@ struct hw_sequencer_funcs { void (*program_outstanding_updates)(struct dc *dc, struct dc_state *context); void (*setup_hpo_hw_control)(const struct dce_hwseq *hws, bool enable); + void (*wait_for_all_pending_updates)(const struct pipe_ctx *pipe_ctx); }; void color_space_to_black_color( @@ -504,6 +509,10 @@ void get_mclk_switch_visual_confirm_color( struct pipe_ctx *pipe_ctx, struct tg_color *color); +void get_cursor_visual_confirm_color( + struct pipe_ctx *pipe_ctx, + struct tg_color *color); + void set_p_state_switch_method( struct dc *dc, struct dc_state *context, diff --git a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h index 0ac675456979..22a5d4a03c98 100644 --- a/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h +++ b/drivers/gpu/drm/amd/display/dc/hwss/hw_sequencer_private.h @@ -182,6 +182,7 @@ struct hwseq_private_funcs { struct pipe_ctx *pipe_ctx, struct dc_cm2_func_luts mcm_luts, bool lut_bank_a); + void (*perform_3dlut_wa_unlock)(struct pipe_ctx *pipe_ctx); }; struct dce_hwseq { diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index bfb8b8502d20..8597e866bfe6 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -215,6 +215,10 @@ struct resource_funcs { void (*get_panel_config_defaults)(struct dc_panel_config *panel_config); void (*build_pipe_pix_clk_params)(struct pipe_ctx *pipe_ctx); + /* + * Get indicator of power from a context that went through full validation + */ + int (*get_power_profile)(const struct dc_state *context); }; struct audio_support{ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h index 3d4c8bd42b49..b74e18cc1e66 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/timing_generator.h @@ -342,7 +342,11 @@ struct timing_generator_funcs { void (*wait_drr_doublebuffer_pending_clear)(struct timing_generator *tg); void (*set_long_vtotal)(struct timing_generator *optc, const struct long_vtotal_params *params); void (*wait_odm_doublebuffer_pending_clear)(struct timing_generator *tg); - bool (*get_double_buffer_pending)(struct timing_generator *tg); + bool (*get_optc_double_buffer_pending)(struct timing_generator *tg); + bool (*get_otg_double_buffer_pending)(struct timing_generator *tg); + bool (*get_pipe_update_pending)(struct timing_generator *tg); + void (*set_vupdate_keepout)(struct timing_generator *tg, bool enable); + bool (*wait_update_lock_status)(struct timing_generator *tg, bool locked); }; #endif diff --git a/drivers/gpu/drm/amd/display/dc/link/link_detection.c b/drivers/gpu/drm/amd/display/dc/link/link_detection.c index d21ee9d12d26..e026c728042a 100644 --- a/drivers/gpu/drm/amd/display/dc/link/link_detection.c +++ b/drivers/gpu/drm/amd/display/dc/link/link_detection.c @@ -48,6 +48,9 @@ #include "dm_helpers.h" #include "clk_mgr.h" + // Offset DPCD 050Eh == 0x5A +#define MST_HUB_ID_0x5A 0x5A + #define DC_LOGGER \ link->ctx->logger #define DC_LOGGER_INIT(logger) @@ -692,6 +695,15 @@ static void apply_dpia_mst_dsc_always_on_wa(struct dc_link *link) link->dpcd_caps.dsc_caps.dsc_basic_caps.fields.dsc_support.DSC_SUPPORT && !link->dc->debug.dpia_debug.bits.disable_mst_dsc_work_around) link->wa_flags.dpia_mst_dsc_always_on = true; + + if (link->ep_type == DISPLAY_ENDPOINT_USB4_DPIA && + link->type == dc_connection_mst_branch && + link->dpcd_caps.branch_dev_id == DP_BRANCH_DEVICE_ID_90CC24 && + link->dpcd_caps.branch_vendor_specific_data[2] == MST_HUB_ID_0x5A && + link->dpcd_caps.dsc_caps.dsc_basic_caps.fields.dsc_support.DSC_SUPPORT && + !link->dc->debug.dpia_debug.bits.disable_mst_dsc_work_around) { + link->wa_flags.dpia_mst_dsc_always_on = true; + } } static void revert_dpia_mst_dsc_always_on_wa(struct dc_link *link) diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c index d78c8ec4de79..e05b8fddf2af 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_capability.c @@ -1207,6 +1207,13 @@ static void get_active_converter_info( dp_hw_fw_revision.ieee_fw_rev, sizeof(dp_hw_fw_revision.ieee_fw_rev)); } + + core_link_read_dpcd( + link, + DP_BRANCH_VENDOR_SPECIFIC_START, + (uint8_t *)link->dpcd_caps.branch_vendor_specific_data, + sizeof(link->dpcd_caps.branch_vendor_specific_data)); + if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_14 && link->dpcd_caps.dongle_type != DISPLAY_DONGLE_NONE) { union dp_dfp_cap_ext dfp_cap_ext; @@ -1409,7 +1416,8 @@ static bool get_usbc_cable_id(struct dc_link *link, union dp_cable_id *cable_id) if (!link->ctx->dmub_srv || link->ep_type != DISPLAY_ENDPOINT_PHY || - link->link_enc->features.flags.bits.DP_IS_USB_C == 0) + link->link_enc->features.flags.bits.DP_IS_USB_C == 0 || + link->link_enc->features.flags.bits.IS_DP2_CAPABLE == 0) return false; memset(&cmd, 0, sizeof(cmd)); @@ -1422,7 +1430,9 @@ static bool get_usbc_cable_id(struct dc_link *link, union dp_cable_id *cable_id) cable_id->raw = cmd.cable_id.data.output_raw; DC_LOG_DC("usbc_cable_id = %d.\n", cable_id->raw); } - return cmd.cable_id.header.ret_status == 1; + + ASSERT(cmd.cable_id.header.ret_status); + return true; } static void retrieve_cable_id(struct dc_link *link) @@ -1626,6 +1636,8 @@ static bool retrieve_link_cap(struct dc_link *link) /* Read DP tunneling information. */ status = dpcd_get_tunneling_device_data(link); + if (status != DC_OK) + dm_error("%s: Read tunneling device data failed.\n", __func__); dpcd_set_source_specific_data(link); /* Sink may need to configure internals based on vendor, so allow some @@ -1842,6 +1854,9 @@ static bool retrieve_link_cap(struct dc_link *link) DP_FEC_CAPABILITY, &link->dpcd_caps.fec_cap.raw, sizeof(link->dpcd_caps.fec_cap.raw)); + if (status != DC_OK) + DC_LOG_ERROR("%s:%d: core_link_read_dpcd (DP_FEC_CAPABILITY) failed\n", __func__, __LINE__); + status = core_link_read_dpcd( link, DP_DSC_SUPPORT, @@ -1864,6 +1879,9 @@ static bool retrieve_link_cap(struct dc_link *link) DP_DSC_BRANCH_OVERALL_THROUGHPUT_0, link->dpcd_caps.dsc_caps.dsc_branch_decoder_caps.raw, sizeof(link->dpcd_caps.dsc_caps.dsc_branch_decoder_caps.raw)); + if (status != DC_OK) + DC_LOG_ERROR("%s:%d: core_link_read_dpcd (DP_DSC_BRANCH_OVERALL_THROUGHPUT_0) failed\n", __func__, __LINE__); + DC_LOG_DSC("DSC branch decoder capability is read at link %d", link->link_index); DC_LOG_DSC("\tBRANCH_OVERALL_THROUGHPUT_0 = 0x%02x", link->dpcd_caps.dsc_caps.dsc_branch_decoder_caps.fields.BRANCH_OVERALL_THROUGHPUT_0); @@ -2055,6 +2073,14 @@ void detect_edp_sink_caps(struct dc_link *link) core_link_read_dpcd(link, DP_SINK_PR_MAX_NUMBER_OF_DEVIATION_LINE, &link->dpcd_caps.pr_info.max_deviation_line, sizeof(link->dpcd_caps.pr_info.max_deviation_line)); + + /* + * OLED Emission Rate info + */ + if (link->dpcd_sink_ext_caps.bits.emission_output) + core_link_read_dpcd(link, DP_SINK_EMISSION_RATE, + (uint8_t *)&link->dpcd_caps.edp_oled_emission_rate, + sizeof(link->dpcd_caps.edp_oled_emission_rate)); } bool dp_get_max_link_enc_cap(const struct dc_link *link, struct dc_link_settings *max_link_enc_cap) @@ -2103,6 +2129,8 @@ struct dc_link_settings dp_get_max_link_cap(struct dc_link *link) /* get max link encoder capability */ if (link_enc) link_enc->funcs->get_max_link_cap(link_enc, &max_link_cap); + else + return max_link_cap; /* Lower link settings based on sink's link cap */ if (link->reported_link_cap.lane_count < max_link_cap.lane_count) @@ -2136,10 +2164,15 @@ struct dc_link_settings dp_get_max_link_cap(struct dc_link *link) */ cable_max_link_rate = get_cable_max_link_rate(link); - if (!link->dc->debug.ignore_cable_id && - cable_max_link_rate != LINK_RATE_UNKNOWN) { - if (cable_max_link_rate < max_link_cap.link_rate) - max_link_cap.link_rate = cable_max_link_rate; + if (!link->dc->debug.ignore_cable_id) { + if (cable_max_link_rate != LINK_RATE_UNKNOWN) + // cable max link rate known + max_link_cap.link_rate = MIN(max_link_cap.link_rate, cable_max_link_rate); + else if (link_enc->funcs->is_in_alt_mode && link_enc->funcs->is_in_alt_mode(link_enc)) + // cable max link rate ambiguous, DP alt mode, limit to HBR3 + max_link_cap.link_rate = MIN(max_link_cap.link_rate, LINK_RATE_HIGH3); + //else {} + // cable max link rate ambiguous, DP, do nothing if (!link->dpcd_caps.cable_id.bits.UHBR13_5_CAPABILITY && link->dpcd_caps.cable_id.bits.CABLE_TYPE >= 2) diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c index 6af42ba9885c..0d123e647652 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_dpia.c @@ -59,12 +59,18 @@ enum dc_status dpcd_get_tunneling_device_data(struct dc_link *link) dpcd_dp_tun_data, sizeof(dpcd_dp_tun_data)); + if (status != DC_OK) + goto err; + status = core_link_read_dpcd( link, DP_USB4_ROUTER_TOPOLOGY_ID, dpcd_topology_data, sizeof(dpcd_topology_data)); + if (status != DC_OK) + goto err; + link->dpcd_caps.usb4_dp_tun_info.dp_tun_cap.raw = dpcd_dp_tun_data[DP_TUNNELING_CAPABILITIES_SUPPORT - DP_TUNNELING_CAPABILITIES_SUPPORT]; link->dpcd_caps.usb4_dp_tun_info.dpia_info.raw = @@ -75,6 +81,7 @@ enum dc_status dpcd_get_tunneling_device_data(struct dc_link *link) for (i = 0; i < DPCD_USB4_TOPOLOGY_ID_LEN; i++) link->dpcd_caps.usb4_dp_tun_info.usb4_topology_id[i] = dpcd_topology_data[i]; +err: return status; } diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c index 27b881f947e8..754c895e1bfb 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training.c @@ -272,7 +272,7 @@ void dp_wait_for_training_aux_rd_interval( struct dc_link *link, uint32_t wait_in_micro_secs) { - fsleep(wait_in_micro_secs); + usleep_range_state(wait_in_micro_secs, wait_in_micro_secs, TASK_UNINTERRUPTIBLE); DC_LOG_HW_LINK_TRAINING("%s:\n wait = %d\n", __func__, @@ -1107,9 +1107,13 @@ enum dc_status dpcd_set_link_settings( status = core_link_write_dpcd(link, DP_DOWNSPREAD_CTRL, &downspread.raw, sizeof(downspread)); + if (status != DC_OK) + DC_LOG_ERROR("%s:%d: core_link_write_dpcd (DP_DOWNSPREAD_CTRL) failed\n", __func__, __LINE__); status = core_link_write_dpcd(link, DP_LANE_COUNT_SET, &lane_count_set.raw, 1); + if (status != DC_OK) + DC_LOG_ERROR("%s:%d: core_link_write_dpcd (DP_LANE_COUNT_SET) failed\n", __func__, __LINE__); if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_13 && lt_settings->link_settings.use_link_rate_set == true) { @@ -1125,12 +1129,19 @@ enum dc_status dpcd_set_link_settings( supported_link_rates, sizeof(supported_link_rates)); } status = core_link_write_dpcd(link, DP_LINK_BW_SET, &rate, 1); + if (status != DC_OK) + DC_LOG_ERROR("%s:%d: core_link_write_dpcd (DP_LINK_BW_SET) failed\n", __func__, __LINE__); + status = core_link_write_dpcd(link, DP_LINK_RATE_SET, <_settings->link_settings.link_rate_set, 1); + if (status != DC_OK) + DC_LOG_ERROR("%s:%d: core_link_write_dpcd (DP_LINK_RATE_SET) failed\n", __func__, __LINE__); } else { rate = get_dpcd_link_rate(<_settings->link_settings); status = core_link_write_dpcd(link, DP_LINK_BW_SET, &rate, 1); + if (status != DC_OK) + DC_LOG_ERROR("%s:%d: core_link_write_dpcd (DP_LINK_BW_SET) failed\n", __func__, __LINE__); } if (rate) { diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c index b5cf75975fff..ccf8096dde29 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_dp_training_fixed_vs_pe_retimer.c @@ -412,7 +412,6 @@ enum link_training_result dp_perform_fixed_vs_pe_training_sequence( /* 5. check CR done*/ if (dp_is_cr_done(lane_count, dpcd_lane_status)) { - status = LINK_TRAINING_SUCCESS; break; } diff --git a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c index 3aa05a2be6c0..43a467f6ce7b 100644 --- a/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c +++ b/drivers/gpu/drm/amd/display/dc/link/protocols/link_edp_panel_control.c @@ -157,31 +157,11 @@ bool edp_set_backlight_level_nits(struct dc_link *link, uint32_t backlight_millinits, uint32_t transition_time_in_ms) { - struct dpcd_source_backlight_set dpcd_backlight_set; - uint8_t backlight_control = isHDR ? 1 : 0; - if (!link || (link->connector_signal != SIGNAL_TYPE_EDP && link->connector_signal != SIGNAL_TYPE_DISPLAY_PORT)) return false; - // OLEDs have no PWM, they can only use AUX - if (link->dpcd_sink_ext_caps.bits.oled == 1) - backlight_control = 1; - - *(uint32_t *)&dpcd_backlight_set.backlight_level_millinits = backlight_millinits; - *(uint16_t *)&dpcd_backlight_set.backlight_transition_time_ms = (uint16_t)transition_time_in_ms; - - - if (!link->dpcd_caps.panel_luminance_control) { - if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL, - (uint8_t *)(&dpcd_backlight_set), - sizeof(dpcd_backlight_set)) != DC_OK) - return false; - - if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_CONTROL, - &backlight_control, 1) != DC_OK) - return false; - } else { + if (link->backlight_control_type == BACKLIGHT_CONTROL_VESA_AUX) { uint8_t backlight_enable = 0; struct target_luminance_value *target_luminance = NULL; @@ -205,6 +185,24 @@ bool edp_set_backlight_level_nits(struct dc_link *link, (uint8_t *)(target_luminance), sizeof(struct target_luminance_value)) != DC_OK) return false; + } else { + struct dpcd_source_backlight_set dpcd_backlight_set; + *(uint32_t *)&dpcd_backlight_set.backlight_level_millinits = backlight_millinits; + *(uint16_t *)&dpcd_backlight_set.backlight_transition_time_ms = (uint16_t)transition_time_in_ms; + + uint8_t backlight_control = isHDR ? 1 : 0; + // OLEDs have no PWM, they can only use AUX + if (link->dpcd_sink_ext_caps.bits.oled == 1) + backlight_control = 1; + + if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_LEVEL, + (uint8_t *)(&dpcd_backlight_set), + sizeof(dpcd_backlight_set)) != DC_OK) + return false; + + if (core_link_write_dpcd(link, DP_SOURCE_BACKLIGHT_CONTROL, + &backlight_control, 1) != DC_OK) + return false; } return true; @@ -523,13 +521,13 @@ bool edp_set_backlight_level(const struct dc_link *link, uint32_t frame_ramp) { struct dc *dc = link->ctx->dc; - DC_LOGGER_INIT(link->ctx->logger); DC_LOG_BACKLIGHT("New Backlight level: %d (0x%X)\n", backlight_pwm_u16_16, backlight_pwm_u16_16); if (dc_is_embedded_signal(link->connector_signal)) { struct pipe_ctx *pipe_ctx = get_pipe_from_link(link); + struct set_backlight_level_params backlight_level_param = { 0 }; if (link->panel_cntl) link->panel_cntl->stored_backlight_registers.USER_LEVEL = backlight_pwm_u16_16; @@ -544,10 +542,12 @@ bool edp_set_backlight_level(const struct dc_link *link, return false; } + backlight_level_param.backlight_pwm_u16_16 = backlight_pwm_u16_16; + backlight_level_param.frame_ramp = frame_ramp; + dc->hwss.set_backlight_level( pipe_ctx, - backlight_pwm_u16_16, - frame_ramp); + &backlight_level_param); } return true; } @@ -940,8 +940,7 @@ bool edp_setup_replay(struct dc_link *link, const struct dc_stream_state *stream struct replay_context replay_context = { 0 }; unsigned int lineTimeInNs = 0; - - union replay_enable_and_configuration replay_config; + union replay_enable_and_configuration replay_config = { 0 }; union dpcd_alpm_configuration alpm_config; @@ -1168,9 +1167,6 @@ static void edp_set_assr_enable(const struct dc *pDC, struct dc_link *link, link_enc_index = link->link_enc->transmitter - TRANSMITTER_UNIPHY_A; if (link_res->hpo_dp_link_enc) { - if (link->wa_flags.disable_assr_for_uhbr) - return; - link_enc_index = link_res->hpo_dp_link_enc->inst; use_hpo_dp_link_enc = true; } diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn10/dcn10_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn10/dcn10_optc.h index b7a57f98553d..40757f20d73f 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn10/dcn10_optc.h +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn10/dcn10_optc.h @@ -202,6 +202,7 @@ struct dcn_optc_registers { uint32_t OPTC_CLOCK_CONTROL; uint32_t OPTC_WIDTH_CONTROL2; uint32_t OTG_PSTATE_REGISTER; + uint32_t OTG_PIPE_UPDATE_STATUS; }; #define TG_COMMON_MASK_SH_LIST_DCN(mask_sh)\ @@ -566,6 +567,12 @@ struct dcn_optc_registers { type OTG_H_TIMING_DIV_MODE_DB_UPDATE_PENDING;\ type OPTC_DOUBLE_BUFFER_PENDING;\ +#define TG_REG_FIELD_LIST_DCN2_0(type) \ + type OTG_FLIP_PENDING;\ + type OTG_DC_REG_UPDATE_PENDING;\ + type OTG_CURSOR_UPDATE_PENDING;\ + type OTG_VUPDATE_KEEPOUT_STATUS;\ + #define TG_REG_FIELD_LIST_DCN3_2(type) \ type OTG_H_TIMING_DIV_MODE_MANUAL; @@ -600,6 +607,7 @@ struct dcn_optc_registers { struct dcn_optc_shift { TG_REG_FIELD_LIST(uint8_t) + TG_REG_FIELD_LIST_DCN2_0(uint8_t) TG_REG_FIELD_LIST_DCN3_2(uint8_t) TG_REG_FIELD_LIST_DCN3_5(uint8_t) TG_REG_FIELD_LIST_DCN401(uint8_t) @@ -607,6 +615,7 @@ struct dcn_optc_shift { struct dcn_optc_mask { TG_REG_FIELD_LIST(uint32_t) + TG_REG_FIELD_LIST_DCN2_0(uint32_t) TG_REG_FIELD_LIST_DCN3_2(uint32_t) TG_REG_FIELD_LIST_DCN3_5(uint32_t) TG_REG_FIELD_LIST_DCN401(uint32_t) diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn20/dcn20_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn20/dcn20_optc.h index 364034b19028..928e110b95fb 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn20/dcn20_optc.h +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn20/dcn20_optc.h @@ -43,7 +43,8 @@ SRI(OPTC_MEMORY_CONFIG, ODM, inst),\ SR(DWB_SOURCE_SELECT),\ SRI(OTG_MANUAL_FLOW_CONTROL, OTG, inst), \ - SRI(OTG_DRR_CONTROL, OTG, inst) + SRI(OTG_DRR_CONTROL, OTG, inst),\ + SRI(OTG_PIPE_UPDATE_STATUS, OTG, inst) #define TG_COMMON_MASK_SH_LIST_DCN2_0(mask_sh)\ TG_COMMON_MASK_SH_LIST_DCN(mask_sh),\ @@ -53,6 +54,10 @@ SF(OTG0_OTG_GLOBAL_CONTROL2, GLOBAL_UPDATE_LOCK_EN, mask_sh),\ SF(OTG0_OTG_GLOBAL_CONTROL2, DIG_UPDATE_LOCATION, mask_sh),\ SF(OTG0_OTG_DOUBLE_BUFFER_CONTROL, OTG_RANGE_TIMING_DBUF_UPDATE_MODE, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_FLIP_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_DC_REG_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_CURSOR_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_VUPDATE_KEEPOUT_STATUS, mask_sh),\ SF(OTG0_OTG_GSL_WINDOW_X, OTG_GSL_WINDOW_START_X, mask_sh),\ SF(OTG0_OTG_GSL_WINDOW_X, OTG_GSL_WINDOW_END_X, mask_sh), \ SF(OTG0_OTG_GSL_WINDOW_Y, OTG_GSL_WINDOW_START_Y, mask_sh),\ diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn30/dcn30_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn30/dcn30_optc.c index abcd03d78668..4c95c0958612 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn30/dcn30_optc.c +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn30/dcn30_optc.c @@ -271,6 +271,48 @@ void optc3_set_odm_combine(struct timing_generator *optc, int *opp_id, int opp_c optc1->opp_count = opp_cnt; } +/* OTG status register that indicates OPTC update is pending */ +bool optc3_get_optc_double_buffer_pending(struct timing_generator *optc) +{ + struct optc *optc1 = DCN10TG_FROM_TG(optc); + uint32_t update_pending = 0; + + REG_GET(OPTC_INPUT_GLOBAL_CONTROL, + OPTC_DOUBLE_BUFFER_PENDING, + &update_pending); + + return (update_pending == 1); +} + +/* OTG status register that indicates OTG update is pending */ +bool optc3_get_otg_update_pending(struct timing_generator *optc) +{ + struct optc *optc1 = DCN10TG_FROM_TG(optc); + uint32_t update_pending = 0; + + REG_GET(OTG_DOUBLE_BUFFER_CONTROL, + OTG_UPDATE_PENDING, + &update_pending); + + return (update_pending == 1); +} + +/* OTG status register that indicates surface update is pending */ +bool optc3_get_pipe_update_pending(struct timing_generator *optc) +{ + struct optc *optc1 = DCN10TG_FROM_TG(optc); + uint32_t flip_pending = 0; + uint32_t dc_update_pending = 0; + + REG_GET_2(OTG_PIPE_UPDATE_STATUS, + OTG_FLIP_PENDING, + &flip_pending, + OTG_DC_REG_UPDATE_PENDING, + &dc_update_pending); + + return (flip_pending == 1 || dc_update_pending == 1); +} + /** * optc3_set_timing_double_buffer() - DRR double buffering control * @@ -375,6 +417,9 @@ static struct timing_generator_funcs dcn30_tg_funcs = { .get_hw_timing = optc1_get_hw_timing, .wait_drr_doublebuffer_pending_clear = optc3_wait_drr_doublebuffer_pending_clear, .is_two_pixels_per_container = optc1_is_two_pixels_per_container, + .get_optc_double_buffer_pending = optc3_get_optc_double_buffer_pending, + .get_otg_double_buffer_pending = optc3_get_otg_update_pending, + .get_pipe_update_pending = optc3_get_pipe_update_pending, }; void dcn30_timing_generator_init(struct optc *optc1) diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn30/dcn30_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn30/dcn30_optc.h index bda974d432ea..e2303f9eaf13 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn30/dcn30_optc.h +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn30/dcn30_optc.h @@ -109,7 +109,8 @@ SRI(OPTC_BYTES_PER_PIXEL, ODM, inst),\ SRI(OPTC_WIDTH_CONTROL, ODM, inst),\ SRI(OPTC_MEMORY_CONFIG, ODM, inst),\ - SR(DWB_SOURCE_SELECT) + SR(DWB_SOURCE_SELECT),\ + SRI(OTG_PIPE_UPDATE_STATUS, OTG, inst) #define DCN30_VTOTAL_REGS_SF(mask_sh) @@ -209,6 +210,7 @@ SF(ODM0_OPTC_INPUT_CLOCK_CONTROL, OPTC_INPUT_CLK_GATE_DIS, mask_sh),\ SF(ODM0_OPTC_INPUT_GLOBAL_CONTROL, OPTC_UNDERFLOW_OCCURRED_STATUS, mask_sh),\ SF(ODM0_OPTC_INPUT_GLOBAL_CONTROL, OPTC_UNDERFLOW_CLEAR, mask_sh),\ + SF(ODM0_OPTC_INPUT_GLOBAL_CONTROL, OPTC_DOUBLE_BUFFER_PENDING, mask_sh),\ SF(VTG0_CONTROL, VTG0_ENABLE, mask_sh),\ SF(VTG0_CONTROL, VTG0_FP2, mask_sh),\ SF(VTG0_CONTROL, VTG0_VCOUNT_INIT, mask_sh),\ @@ -319,7 +321,11 @@ SF(OTG0_OTG_DRR_V_TOTAL_CHANGE, OTG_DRR_V_TOTAL_CHANGE_LIMIT, mask_sh),\ SF(OTG0_OTG_H_TIMING_CNTL, OTG_H_TIMING_DIV_MODE, mask_sh),\ SF(OTG0_OTG_DOUBLE_BUFFER_CONTROL, OTG_DRR_TIMING_DBUF_UPDATE_PENDING, mask_sh),\ - SF(OTG0_OTG_DOUBLE_BUFFER_CONTROL, OTG_DRR_TIMING_DBUF_UPDATE_MODE, mask_sh) + SF(OTG0_OTG_DOUBLE_BUFFER_CONTROL, OTG_DRR_TIMING_DBUF_UPDATE_MODE, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_FLIP_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_DC_REG_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_CURSOR_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_VUPDATE_KEEPOUT_STATUS, mask_sh),\ void dcn30_timing_generator_init(struct optc *optc1); @@ -356,4 +362,7 @@ void optc3_set_odm_combine(struct timing_generator *optc, int *opp_id, int opp_c void optc3_wait_drr_doublebuffer_pending_clear(struct timing_generator *optc); void optc3_tg_init(struct timing_generator *optc); void optc3_set_vtotal_min_max(struct timing_generator *optc, int vtotal_min, int vtotal_max); +bool optc3_get_optc_double_buffer_pending(struct timing_generator *optc); +bool optc3_get_otg_update_pending(struct timing_generator *optc); +bool optc3_get_pipe_update_pending(struct timing_generator *optc); #endif /* __DC_OPTC_DCN30_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn301/dcn301_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn301/dcn301_optc.c index 1a22ae89fb55..d7a45ef2d01b 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn301/dcn301_optc.c +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn301/dcn301_optc.c @@ -169,6 +169,9 @@ static struct timing_generator_funcs dcn30_tg_funcs = { .get_hw_timing = optc1_get_hw_timing, .wait_drr_doublebuffer_pending_clear = optc3_wait_drr_doublebuffer_pending_clear, .is_two_pixels_per_container = optc1_is_two_pixels_per_container, + .get_optc_double_buffer_pending = optc3_get_optc_double_buffer_pending, + .get_otg_double_buffer_pending = optc3_get_otg_update_pending, + .get_pipe_update_pending = optc3_get_pipe_update_pending, }; void dcn301_timing_generator_init(struct optc *optc1) diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn31/dcn31_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn31/dcn31_optc.h index 30b81a448ce2..fbbe86d00c2e 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn31/dcn31_optc.h +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn31/dcn31_optc.h @@ -99,7 +99,8 @@ SRI(OPTC_MEMORY_CONFIG, ODM, inst),\ SRI(OTG_CRC_CNTL2, OTG, inst),\ SR(DWB_SOURCE_SELECT),\ - SRI(OTG_DRR_CONTROL, OTG, inst) + SRI(OTG_DRR_CONTROL, OTG, inst),\ + SRI(OTG_PIPE_UPDATE_STATUS, OTG, inst) #define OPTC_COMMON_MASK_SH_LIST_DCN3_1(mask_sh)\ SF(OTG0_OTG_VSTARTUP_PARAM, VSTARTUP_START, mask_sh),\ @@ -254,7 +255,11 @@ SF(OTG0_OTG_CRC_CNTL2, OTG_CRC_DATA_STREAM_COMBINE_MODE, mask_sh),\ SF(OTG0_OTG_CRC_CNTL2, OTG_CRC_DATA_STREAM_SPLIT_MODE, mask_sh),\ SF(OTG0_OTG_CRC_CNTL2, OTG_CRC_DATA_FORMAT, mask_sh),\ - SF(OTG0_OTG_DRR_CONTROL, OTG_V_TOTAL_LAST_USED_BY_DRR, mask_sh) + SF(OTG0_OTG_DRR_CONTROL, OTG_V_TOTAL_LAST_USED_BY_DRR, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_FLIP_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_DC_REG_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_CURSOR_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_VUPDATE_KEEPOUT_STATUS, mask_sh),\ void dcn31_timing_generator_init(struct optc *optc1); diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn314/dcn314_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn314/dcn314_optc.h index 99c098e76116..0ff72b97b465 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn314/dcn314_optc.h +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn314/dcn314_optc.h @@ -98,7 +98,8 @@ SRI(OPTC_BYTES_PER_PIXEL, ODM, inst),\ SRI(OPTC_WIDTH_CONTROL, ODM, inst),\ SRI(OPTC_MEMORY_CONFIG, ODM, inst),\ - SRI(OTG_DRR_CONTROL, OTG, inst) + SRI(OTG_DRR_CONTROL, OTG, inst),\ + SRI(OTG_PIPE_UPDATE_STATUS, OTG, inst) #define OPTC_COMMON_MASK_SH_LIST_DCN3_14(mask_sh)\ SF(OTG0_OTG_VSTARTUP_PARAM, VSTARTUP_START, mask_sh),\ @@ -248,7 +249,11 @@ SF(OTG0_OTG_H_TIMING_CNTL, OTG_H_TIMING_DIV_MODE, mask_sh),\ SF(OTG0_OTG_H_TIMING_CNTL, OTG_H_TIMING_DIV_MODE_MANUAL, mask_sh),\ SF(OTG0_OTG_DOUBLE_BUFFER_CONTROL, OTG_DRR_TIMING_DBUF_UPDATE_MODE, mask_sh),\ - SF(OTG0_OTG_DRR_CONTROL, OTG_V_TOTAL_LAST_USED_BY_DRR, mask_sh) + SF(OTG0_OTG_DRR_CONTROL, OTG_V_TOTAL_LAST_USED_BY_DRR, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_FLIP_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_DC_REG_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_CURSOR_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_VUPDATE_KEEPOUT_STATUS, mask_sh),\ void dcn314_timing_generator_init(struct optc *optc1); diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c index 00094f0e8470..c217f653b3c8 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.c @@ -297,18 +297,6 @@ static void optc32_set_drr( optc32_setup_manual_trigger(optc); } -bool optc32_get_double_buffer_pending(struct timing_generator *optc) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - uint32_t update_pending = 0; - - REG_GET(OPTC_INPUT_GLOBAL_CONTROL, - OPTC_DOUBLE_BUFFER_PENDING, - &update_pending); - - return (update_pending == 1); -} - static struct timing_generator_funcs dcn32_tg_funcs = { .validate_timing = optc1_validate_timing, .program_timing = optc1_program_timing, @@ -373,7 +361,9 @@ static struct timing_generator_funcs dcn32_tg_funcs = { .setup_manual_trigger = optc2_setup_manual_trigger, .get_hw_timing = optc1_get_hw_timing, .is_two_pixels_per_container = optc1_is_two_pixels_per_container, - .get_double_buffer_pending = optc32_get_double_buffer_pending, + .get_optc_double_buffer_pending = optc3_get_optc_double_buffer_pending, + .get_otg_double_buffer_pending = optc3_get_otg_update_pending, + .get_pipe_update_pending = optc3_get_pipe_update_pending, }; void dcn32_timing_generator_init(struct optc *optc1) diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h index 665d7c52f67c..0b0964a9da74 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn32/dcn32_optc.h @@ -177,7 +177,11 @@ SF(OTG0_OTG_H_TIMING_CNTL, OTG_H_TIMING_DIV_MODE, mask_sh),\ SF(OTG0_OTG_H_TIMING_CNTL, OTG_H_TIMING_DIV_MODE_MANUAL, mask_sh),\ SF(OTG0_OTG_DOUBLE_BUFFER_CONTROL, OTG_DRR_TIMING_DBUF_UPDATE_MODE, mask_sh),\ - SF(OTG0_OTG_DRR_CONTROL, OTG_V_TOTAL_LAST_USED_BY_DRR, mask_sh) + SF(OTG0_OTG_DRR_CONTROL, OTG_V_TOTAL_LAST_USED_BY_DRR, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_FLIP_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_DC_REG_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_CURSOR_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_VUPDATE_KEEPOUT_STATUS, mask_sh) void dcn32_timing_generator_init(struct optc *optc1); void optc32_set_h_timing_div_manual_mode(struct timing_generator *optc, bool manual_mode); @@ -185,6 +189,5 @@ void optc32_get_odm_combine_segments(struct timing_generator *tg, int *odm_combi void optc32_set_odm_bypass(struct timing_generator *optc, const struct dc_crtc_timing *dc_crtc_timing); void optc32_wait_odm_doublebuffer_pending_clear(struct timing_generator *tg); -bool optc32_get_double_buffer_pending(struct timing_generator *optc); #endif /* __DC_OPTC_DCN32_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.h index d077e2392379..be749ab41dce 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.h +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn35/dcn35_optc.h @@ -67,7 +67,11 @@ SF(OTG0_OTG_CRC1_WINDOWB_Y_CONTROL_READBACK, OTG_CRC1_WINDOWB_Y_END_READBACK, mask_sh),\ SF(OPTC_CLOCK_CONTROL, OPTC_FGCG_REP_DIS, mask_sh),\ SF(OTG0_OTG_V_COUNT_STOP_CONTROL, OTG_V_COUNT_STOP, mask_sh),\ - SF(OTG0_OTG_V_COUNT_STOP_CONTROL2, OTG_V_COUNT_STOP_TIMER, mask_sh) + SF(OTG0_OTG_V_COUNT_STOP_CONTROL2, OTG_V_COUNT_STOP_TIMER, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_FLIP_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_DC_REG_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_CURSOR_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_VUPDATE_KEEPOUT_STATUS, mask_sh) void dcn35_timing_generator_init(struct optc *optc1); diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c b/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c index a5d6a7dca554..783ca9acc762 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.c @@ -430,6 +430,35 @@ static void optc401_program_global_sync( REG_UPDATE(OTG_PSTATE_REGISTER, OTG_PSTATE_KEEPOUT_START, pstate_keepout); } +static void optc401_set_vupdate_keepout(struct timing_generator *tg, bool enable) +{ + struct optc *optc1 = DCN10TG_FROM_TG(tg); + + REG_SET_3(OTG_VUPDATE_KEEPOUT, 0, + MASTER_UPDATE_LOCK_VUPDATE_KEEPOUT_START_OFFSET, 0, + MASTER_UPDATE_LOCK_VUPDATE_KEEPOUT_END_OFFSET, optc1->vready_offset + 10, + OTG_MASTER_UPDATE_LOCK_VUPDATE_KEEPOUT_EN, enable); + + return; +} + +static bool optc401_wait_update_lock_status(struct timing_generator *tg, bool locked) +{ + struct optc *optc1 = DCN10TG_FROM_TG(tg); + uint32_t lock_status = 0; + + REG_WAIT(OTG_MASTER_UPDATE_LOCK, + UPDATE_LOCK_STATUS, locked, + 1, 150000); + + REG_GET(OTG_MASTER_UPDATE_LOCK, UPDATE_LOCK_STATUS, &lock_status); + + if (lock_status != locked) + return false; + + return true; +} + static struct timing_generator_funcs dcn401_tg_funcs = { .validate_timing = optc1_validate_timing, .program_timing = optc1_program_timing, @@ -493,7 +522,11 @@ static struct timing_generator_funcs dcn401_tg_funcs = { .setup_manual_trigger = optc2_setup_manual_trigger, .get_hw_timing = optc1_get_hw_timing, .is_two_pixels_per_container = optc1_is_two_pixels_per_container, - .get_double_buffer_pending = optc32_get_double_buffer_pending, + .get_optc_double_buffer_pending = optc3_get_optc_double_buffer_pending, + .get_otg_double_buffer_pending = optc3_get_otg_update_pending, + .get_pipe_update_pending = optc3_get_pipe_update_pending, + .set_vupdate_keepout = optc401_set_vupdate_keepout, + .wait_update_lock_status = optc401_wait_update_lock_status, }; void dcn401_timing_generator_init(struct optc *optc1) diff --git a/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.h b/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.h index bb13a645802d..1be89571986f 100644 --- a/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.h +++ b/drivers/gpu/drm/amd/display/dc/optc/dcn401/dcn401_optc.h @@ -159,7 +159,11 @@ SF(OTG0_OTG_PSTATE_REGISTER, OTG_PSTATE_KEEPOUT_START, mask_sh),\ SF(OTG0_OTG_PSTATE_REGISTER, OTG_PSTATE_EXTEND, mask_sh),\ SF(OTG0_OTG_PSTATE_REGISTER, OTG_UNBLANK, mask_sh),\ - SF(OTG0_OTG_PSTATE_REGISTER, OTG_PSTATE_ALLOW_WIDTH_MIN, mask_sh) + SF(OTG0_OTG_PSTATE_REGISTER, OTG_PSTATE_ALLOW_WIDTH_MIN, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_FLIP_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_DC_REG_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_CURSOR_UPDATE_PENDING, mask_sh),\ + SF(OTG0_OTG_PIPE_UPDATE_STATUS, OTG_VUPDATE_KEEPOUT_STATUS, mask_sh) void dcn401_timing_generator_init(struct optc *optc1); diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c index 53a5f4cb648c..e698543ec937 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce100/dce100_resource.c @@ -623,7 +623,7 @@ static struct link_encoder *dce100_link_encoder_create( kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); int link_regs_id; - if (!enc110) + if (!enc110 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; link_regs_id = diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c index 91da5cf85b69..035c6cfdaee5 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce110/dce110_resource.c @@ -668,7 +668,7 @@ static struct link_encoder *dce110_link_encoder_create( kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); int link_regs_id; - if (!enc110) + if (!enc110 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; link_regs_id = diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c index 162856c523e4..480a50967385 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce112/dce112_resource.c @@ -629,7 +629,7 @@ static struct link_encoder *dce112_link_encoder_create( kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); int link_regs_id; - if (!enc110) + if (!enc110 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; link_regs_id = diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c index 621825a51f46..c63c59623433 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce120/dce120_resource.c @@ -706,7 +706,7 @@ static struct link_encoder *dce120_link_encoder_create( kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); int link_regs_id; - if (!enc110) + if (!enc110 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; link_regs_id = diff --git a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c index a73d3c6ef425..3d5113f010bb 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dce80/dce80_resource.c @@ -723,7 +723,7 @@ static struct link_encoder *dce80_link_encoder_create( kzalloc(sizeof(struct dce110_link_encoder), GFP_KERNEL); int link_regs_id; - if (!enc110) + if (!enc110 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; link_regs_id = diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c index 563c5eec83ff..05d6d41ef9d3 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn10/dcn10_resource.c @@ -751,7 +751,7 @@ static struct link_encoder *dcn10_link_encoder_create( kzalloc(sizeof(struct dcn10_link_encoder), GFP_KERNEL); int link_regs_id; - if (!enc10) + if (!enc10 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; link_regs_id = diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c index eea2b3b307cd..288189913e1e 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn20/dcn20_resource.c @@ -920,7 +920,7 @@ struct link_encoder *dcn20_link_encoder_create( kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); int link_regs_id; - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; link_regs_id = diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn201/dcn201_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn201/dcn201_resource.c index fc54483b9104..15180ad71513 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn201/dcn201_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn201/dcn201_resource.c @@ -797,7 +797,7 @@ static struct link_encoder *dcn201_link_encoder_create( kzalloc(sizeof(struct dcn20_link_encoder), GFP_ATOMIC); struct dcn10_link_encoder *enc10; - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; enc10 = &enc20->enc10; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c index 347e6aaea582..14b28841657d 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn21/dcn21_resource.c @@ -1298,7 +1298,7 @@ static struct link_encoder *dcn21_link_encoder_create( kzalloc(sizeof(struct dcn21_link_encoder), GFP_KERNEL); int link_regs_id; - if (!enc21) + if (!enc21 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; link_regs_id = diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c index 5040a4c6ed18..baa4e2647dad 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn30/dcn30_resource.c @@ -927,7 +927,7 @@ static struct link_encoder *dcn30_link_encoder_create( struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; dcn30_link_encoder_construct(enc20, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn301/dcn301_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn301/dcn301_resource.c index 7d04739c3ba1..d8a7c2cf05de 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn301/dcn301_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn301/dcn301_resource.c @@ -883,7 +883,7 @@ static struct link_encoder *dcn301_link_encoder_create( struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; dcn301_link_encoder_construct(enc20, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn302/dcn302_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn302/dcn302_resource.c index 5791b5cc2875..40c20b04635a 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn302/dcn302_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn302/dcn302_resource.c @@ -893,7 +893,7 @@ static struct link_encoder *dcn302_link_encoder_create( { struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; dcn30_link_encoder_construct(enc20, enc_init_data, &link_enc_feature, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c index 63f0f882c861..daf1b65fd088 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn303/dcn303_resource.c @@ -839,7 +839,7 @@ static struct link_encoder *dcn303_link_encoder_create( { struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; dcn30_link_encoder_construct(enc20, enc_init_data, &link_enc_feature, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c index ac8cb20e2e3b..36bb26182e11 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn31/dcn31_resource.c @@ -1093,7 +1093,7 @@ static struct link_encoder *dcn31_link_encoder_create( struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; dcn31_link_encoder_construct(enc20, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c index 169924d0a839..58a5fbcf22bf 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn314/dcn314_resource.c @@ -1149,7 +1149,7 @@ static struct link_encoder *dcn31_link_encoder_create( struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; dcn31_link_encoder_construct(enc20, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c index 3f4b9dba4112..3acad708c31b 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn315/dcn315_resource.c @@ -1091,7 +1091,7 @@ static struct link_encoder *dcn31_link_encoder_create( struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; dcn31_link_encoder_construct(enc20, @@ -1812,6 +1812,11 @@ static void dcn315_get_panel_config_defaults(struct dc_panel_config *panel_confi *panel_config = panel_config_defaults; } +static int dcn315_get_power_profile(const struct dc_state *context) +{ + return !context->bw_ctx.bw.dcn.clk.p_state_change_support; +} + static struct dc_cap_funcs cap_funcs = { .get_dcc_compression_cap = dcn20_get_dcc_compression_cap }; @@ -1840,6 +1845,7 @@ static struct resource_funcs dcn315_res_pool_funcs = { .update_bw_bounding_box = dcn315_update_bw_bounding_box, .patch_unknown_plane_state = dcn20_patch_unknown_plane_state, .get_panel_config_defaults = dcn315_get_panel_config_defaults, + .get_power_profile = dcn315_get_power_profile, }; static bool dcn315_resource_construct( diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c index 5fd52c5fcee4..ce56f5d162c0 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn316/dcn316_resource.c @@ -1085,7 +1085,7 @@ static struct link_encoder *dcn31_link_encoder_create( struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; dcn31_link_encoder_construct(enc20, diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c index a124ad9bd108..aaaa888d112d 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c @@ -1039,7 +1039,7 @@ static struct link_encoder *dcn32_link_encoder_create( struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; #undef REG_STRUCT @@ -1990,6 +1990,10 @@ unsigned int dcn32_calculate_mall_ways_from_bytes(const struct dc *dc, unsigned return 0; } + if (dc->caps.max_cab_allocation_bytes == 0) { + return 0xffffffff; + } + /* add 2 lines for worst case alignment */ cache_lines_used = total_size_in_mall_bytes / dc->caps.cache_line_size + 2; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h index 7901792afb7b..86c6e5e8c42e 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.h @@ -1054,7 +1054,8 @@ unsigned int dcn32_calculate_mall_ways_from_bytes(const struct dc *dc, unsigned SRI_ARR(OPTC_BYTES_PER_PIXEL, ODM, inst), \ SRI_ARR(OPTC_WIDTH_CONTROL, ODM, inst), \ SRI_ARR(OPTC_MEMORY_CONFIG, ODM, inst), \ - SRI_ARR(OTG_DRR_CONTROL, OTG, inst) + SRI_ARR(OTG_DRR_CONTROL, OTG, inst), \ + SRI_ARR(OTG_PIPE_UPDATE_STATUS, OTG, inst) /* HUBP */ diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c index 827a94f84f10..35acc13cb5a9 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn321/dcn321_resource.c @@ -1035,7 +1035,7 @@ static struct link_encoder *dcn321_link_encoder_create( struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; #undef REG_STRUCT diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c index 893a9d9ee870..795f2c71c70f 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn35/dcn35_resource.c @@ -1074,7 +1074,7 @@ static struct link_encoder *dcn35_link_encoder_create( struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; #undef REG_STRUCT diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c index da9101b83e8c..0b8dc2eff596 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn351/dcn351_resource.c @@ -766,6 +766,7 @@ static const struct dc_debug_options debug_defaults_drv = { .disable_dmub_reallow_idle = false, .static_screen_wait_frames = 2, .notify_dpia_hr_bw = true, + .min_disp_clk_khz = 50000, }; static const struct dc_panel_config panel_config_defaults = { @@ -1053,7 +1054,7 @@ static struct link_encoder *dcn35_link_encoder_create( struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; #undef REG_STRUCT diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c index 9d56fbdcd06a..306b4117e219 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.c @@ -1032,7 +1032,7 @@ static struct link_encoder *dcn401_link_encoder_create( struct dcn20_link_encoder *enc20 = kzalloc(sizeof(struct dcn20_link_encoder), GFP_KERNEL); - if (!enc20) + if (!enc20 || enc_init_data->hpd_source >= ARRAY_SIZE(link_enc_hpd_regs)) return NULL; #undef REG_STRUCT @@ -1579,7 +1579,8 @@ static void dcn401_destroy_resource_pool(struct resource_pool **pool) } static struct dc_cap_funcs cap_funcs = { - .get_dcc_compression_cap = dcn20_get_dcc_compression_cap + .get_dcc_compression_cap = dcn20_get_dcc_compression_cap, + .get_subvp_en = dcn32_subvp_in_use, }; static void dcn401_update_bw_bounding_box(struct dc *dc, struct clk_bw_params *bw_params) @@ -1688,6 +1689,45 @@ static void dcn401_build_pipe_pix_clk_params(struct pipe_ctx *pipe_ctx) } } +static int dcn401_get_power_profile(const struct dc_state *context) +{ + int uclk_mhz = context->bw_ctx.bw.dcn.clk.dramclk_khz / 1000; + int dpm_level = 0; + + for (int i = 0; i < context->clk_mgr->bw_params->clk_table.num_entries_per_clk.num_memclk_levels; i++) { + if (context->clk_mgr->bw_params->clk_table.entries[i].memclk_mhz == 0 || + uclk_mhz < context->clk_mgr->bw_params->clk_table.entries[i].memclk_mhz) + break; + if (uclk_mhz > context->clk_mgr->bw_params->clk_table.entries[i].memclk_mhz) + dpm_level++; + } + + return dpm_level; +} + +static unsigned int dcn401_calc_num_avail_chans_for_mall(struct dc *dc, unsigned int num_chans) +{ + unsigned int num_available_chans = 1; + + /* channels for MALL must be a power of 2 */ + while (num_chans > 1) { + num_available_chans = (num_available_chans << 1); + num_chans = (num_chans >> 1); + } + + /* cannot be odd */ + num_available_chans &= ~1; + + /* clamp to max available channels for MALL per ASIC */ + if (ASICREV_IS_GC_12_0_0_A0(dc->ctx->asic_id.hw_internal_rev)) { + num_available_chans = num_available_chans > 16 ? 16 : num_available_chans; + } else if (ASICREV_IS_GC_12_0_1_A0(dc->ctx->asic_id.hw_internal_rev)) { + num_available_chans = num_available_chans > 8 ? 8 : num_available_chans; + } + + return num_available_chans; +} + static struct resource_funcs dcn401_res_pool_funcs = { .destroy = dcn401_destroy_resource_pool, .link_enc_create = dcn401_link_encoder_create, @@ -1714,6 +1754,7 @@ static struct resource_funcs dcn401_res_pool_funcs = { .prepare_mcache_programming = dcn401_prepare_mcache_programming, .build_pipe_pix_clk_params = dcn401_build_pipe_pix_clk_params, .calculate_mall_ways_from_bytes = dcn32_calculate_mall_ways_from_bytes, + .get_power_profile = dcn401_get_power_profile, }; static uint32_t read_pipe_fuses(struct dc_context *ctx) @@ -1795,14 +1836,12 @@ static bool dcn401_resource_construct( dc->caps.min_horizontal_blanking_period = 80; dc->caps.dmdata_alloc_size = 2048; dc->caps.mall_size_per_mem_channel = 4; - /* total size = mall per channel * num channels * 1024 * 1024 */ - dc->caps.mall_size_total = dc->caps.mall_size_per_mem_channel * dc->ctx->dc_bios->vram_info.num_chans * 1048576; dc->caps.cursor_cache_size = dc->caps.max_cursor_size * dc->caps.max_cursor_size * 8; dc->caps.cache_line_size = 64; dc->caps.cache_num_ways = 16; /* Calculate the available MALL space */ - dc->caps.max_cab_allocation_bytes = dcn32_calc_num_avail_chans_for_mall( + dc->caps.max_cab_allocation_bytes = dcn401_calc_num_avail_chans_for_mall( dc, dc->ctx->dc_bios->vram_info.num_chans) * dc->caps.mall_size_per_mem_channel * 1024 * 1024; dc->caps.mall_size_total = dc->caps.max_cab_allocation_bytes; @@ -1867,6 +1906,7 @@ static bool dcn401_resource_construct( dc->config.prefer_easf = true; dc->config.dc_mode_clk_limit_support = true; dc->config.enable_windowed_mpo_odm = true; + dc->config.set_pipe_unlock_order = true; /* Need to ensure DET gets freed before allocating */ /* read VBIOS LTTPR caps */ { if (ctx->dc_bios->funcs->get_lttpr_caps) { @@ -2132,6 +2172,7 @@ static bool dcn401_resource_construct( /* SPL */ spl_init_easf_filter_coeffs(); spl_init_blur_scale_coeffs(); + dc->caps.scl_caps.sharpener_support = true; return true; diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h index 514d1ce20df9..bdafa7496cea 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn401/dcn401_resource.h @@ -536,8 +536,9 @@ void dcn401_prepare_mcache_programming(struct dc *dc, struct dc_state *context); SRI_ARR(OPTC_WIDTH_CONTROL, ODM, inst), \ SRI_ARR(OPTC_WIDTH_CONTROL2, ODM, inst), \ SRI_ARR(OPTC_MEMORY_CONFIG, ODM, inst), \ - SRI_ARR(OTG_DRR_CONTROL, OTG, inst), \ - SRI_ARR(OTG_PSTATE_REGISTER, OTG, inst) + SRI_ARR(OTG_DRR_CONTROL, OTG, inst), \ + SRI_ARR(OTG_PSTATE_REGISTER, OTG, inst), \ + SRI_ARR(OTG_PIPE_UPDATE_STATUS, OTG, inst) /* HUBBUB */ #define HUBBUB_REG_LIST_DCN4_01_RI(id) \ diff --git a/drivers/gpu/drm/amd/display/dc/spl/dc_spl.c b/drivers/gpu/drm/amd/display/dc/spl/dc_spl.c index 014e8a296f0c..f043c7e32e16 100644 --- a/drivers/gpu/drm/amd/display/dc/spl/dc_spl.c +++ b/drivers/gpu/drm/amd/display/dc/spl/dc_spl.c @@ -848,13 +848,13 @@ static bool spl_get_isharp_en(struct spl_in *spl_in, * surfaces based on policy setting */ if (!spl_is_yuv420(spl_in->basic_in.format) && - (spl_in->debug.sharpen_policy == SHARPEN_YUV)) + (spl_in->sharpen_policy == SHARPEN_YUV)) return enable_isharp; else if ((spl_is_yuv420(spl_in->basic_in.format) && !fullscreen) && - (spl_in->debug.sharpen_policy == SHARPEN_RGB_FULLSCREEN_YUV)) + (spl_in->sharpen_policy == SHARPEN_RGB_FULLSCREEN_YUV)) return enable_isharp; else if (!spl_in->is_fullscreen && - spl_in->debug.sharpen_policy == SHARPEN_FULLSCREEN_ALL) + spl_in->sharpen_policy == SHARPEN_FULLSCREEN_ALL) return enable_isharp; /* @@ -885,6 +885,18 @@ static bool spl_get_optimal_number_of_taps( spl_scratch->scl_data.viewport.width > max_downscale_src_width) return false; + /* Disable adaptive scaler and sharpener when integer scaling is enabled */ + if (spl_in->scaling_quality.integer_scaling) { + spl_scratch->scl_data.taps.h_taps = 1; + spl_scratch->scl_data.taps.v_taps = 1; + spl_scratch->scl_data.taps.v_taps_c = 1; + spl_scratch->scl_data.taps.h_taps_c = 1; + *enable_easf_v = false; + *enable_easf_h = false; + *enable_isharp = false; + return true; + } + /* Check if we are using EASF or not */ skip_easf = enable_easf(spl_in, spl_scratch); diff --git a/drivers/gpu/drm/amd/display/dc/spl/dc_spl_types.h b/drivers/gpu/drm/amd/display/dc/spl/dc_spl_types.h index 2a74ff5fdfdb..fcb5d389592b 100644 --- a/drivers/gpu/drm/amd/display/dc/spl/dc_spl_types.h +++ b/drivers/gpu/drm/amd/display/dc/spl/dc_spl_types.h @@ -5,10 +5,8 @@ #ifndef __DC_SPL_TYPES_H__ #define __DC_SPL_TYPES_H__ +#include "spl_debug.h" #include "spl_os_types.h" // swap -#ifndef SPL_ASSERT -#define SPL_ASSERT(_bool) ((void *)0) -#endif #include "spl_fixpt31_32.h" // fixed31_32 and related functions #include "spl_custom_float.h" // custom float and related functions @@ -510,7 +508,6 @@ struct spl_funcs { struct spl_debug { int visual_confirm_base_offset; int visual_confirm_dpp_offset; - enum sharpen_policy sharpen_policy; enum scale_to_sharpness_policy scale_to_sharpness_policy; }; @@ -532,6 +529,7 @@ struct spl_in { int h_active; int v_active; int sdr_white_level_nits; + enum sharpen_policy sharpen_policy; }; // end of SPL inputs diff --git a/drivers/gpu/drm/amd/display/dc/spl/spl_debug.h b/drivers/gpu/drm/amd/display/dc/spl/spl_debug.h index 5696dafd0894..a6f6132df241 100644 --- a/drivers/gpu/drm/amd/display/dc/spl/spl_debug.h +++ b/drivers/gpu/drm/amd/display/dc/spl/spl_debug.h @@ -5,21 +5,26 @@ #ifndef SPL_DEBUG_H #define SPL_DEBUG_H -#ifdef SPL_ASSERT -#undef SPL_ASSERT -#endif -#define SPL_ASSERT(b) +#if defined(CONFIG_HAVE_KGDB) || defined(CONFIG_KGDB) +#define SPL_ASSERT_CRITICAL(expr) do { \ + if (WARN_ON(!(expr))) { \ + kgdb_breakpoint(); \ + } \ +} while (0) +#else +#define SPL_ASSERT_CRITICAL(expr) do { \ + if (WARN_ON(!(expr))) { \ + ; \ + } \ +} while (0) +#endif /* CONFIG_HAVE_KGDB || CONFIG_KGDB */ -#define SPL_ASSERT_CRITICAL(expr) do {if (expr)/* Do nothing */; } while (0) +#if defined(CONFIG_DEBUG_KERNEL_DC) +#define SPL_ASSERT(expr) SPL_ASSERT_CRITICAL(expr) +#else +#define SPL_ASSERT(expr) WARN_ON(!(expr)) +#endif /* CONFIG_DEBUG_KERNEL_DC */ -#ifdef SPL_DALMSG -#undef SPL_DALMSG -#endif -#define SPL_DALMSG(b) - -#ifdef SPL_DAL_ASSERT_MSG -#undef SPL_DAL_ASSERT_MSG -#endif -#define SPL_DAL_ASSERT_MSG(b, m) +#define SPL_BREAK_TO_DEBUGGER() SPL_ASSERT(0) #endif // SPL_DEBUG_H diff --git a/drivers/gpu/drm/amd/display/dc/spl/spl_fixpt31_32.c b/drivers/gpu/drm/amd/display/dc/spl/spl_fixpt31_32.c index a95565df5487..5fd79d9c67e2 100644 --- a/drivers/gpu/drm/amd/display/dc/spl/spl_fixpt31_32.c +++ b/drivers/gpu/drm/amd/display/dc/spl/spl_fixpt31_32.c @@ -29,7 +29,7 @@ static inline unsigned long long complete_integer_division_u64( { unsigned long long result; - ASSERT(divisor); + SPL_ASSERT(divisor); result = spl_div64_u64_rem(dividend, divisor, remainder); @@ -63,7 +63,7 @@ struct spl_fixed31_32 spl_fixpt_from_fraction(long long numerator, long long den unsigned long long res_value = complete_integer_division_u64( arg1_value, arg2_value, &remainder); - ASSERT(res_value <= LONG_MAX); + SPL_ASSERT(res_value <= (unsigned long long)LONG_MAX); /* determine fractional part */ { @@ -85,7 +85,7 @@ struct spl_fixed31_32 spl_fixpt_from_fraction(long long numerator, long long den { unsigned long long summand = (remainder << 1) >= arg2_value; - ASSERT(res_value <= LLONG_MAX - summand); + SPL_ASSERT(res_value <= (unsigned long long)LLONG_MAX - summand); res_value += summand; } @@ -118,19 +118,19 @@ struct spl_fixed31_32 spl_fixpt_mul(struct spl_fixed31_32 arg1, struct spl_fixed res.value = arg1_int * arg2_int; - ASSERT(res.value <= (long long)LONG_MAX); + SPL_ASSERT(res.value <= (long long)LONG_MAX); res.value <<= FIXED31_32_BITS_PER_FRACTIONAL_PART; tmp = arg1_int * arg2_fra; - ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); + SPL_ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); res.value += tmp; tmp = arg2_int * arg1_fra; - ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); + SPL_ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); res.value += tmp; @@ -139,7 +139,7 @@ struct spl_fixed31_32 spl_fixpt_mul(struct spl_fixed31_32 arg1, struct spl_fixed tmp = (tmp >> FIXED31_32_BITS_PER_FRACTIONAL_PART) + (tmp >= (unsigned long long)spl_fixpt_half.value); - ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); + SPL_ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); res.value += tmp; @@ -163,17 +163,17 @@ struct spl_fixed31_32 spl_fixpt_sqr(struct spl_fixed31_32 arg) res.value = arg_int * arg_int; - ASSERT(res.value <= (long long)LONG_MAX); + SPL_ASSERT(res.value <= (long long)LONG_MAX); res.value <<= FIXED31_32_BITS_PER_FRACTIONAL_PART; tmp = arg_int * arg_fra; - ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); + SPL_ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); res.value += tmp; - ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); + SPL_ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); res.value += tmp; @@ -182,7 +182,7 @@ struct spl_fixed31_32 spl_fixpt_sqr(struct spl_fixed31_32 arg) tmp = (tmp >> FIXED31_32_BITS_PER_FRACTIONAL_PART) + (tmp >= (unsigned long long)spl_fixpt_half.value); - ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); + SPL_ASSERT(tmp <= (unsigned long long)(LLONG_MAX - res.value)); res.value += tmp; @@ -196,7 +196,7 @@ struct spl_fixed31_32 spl_fixpt_recip(struct spl_fixed31_32 arg) * Good idea to use Newton's method */ - ASSERT(arg.value); + SPL_ASSERT(arg.value); return spl_fixpt_from_fraction( spl_fixpt_one.value, @@ -295,7 +295,7 @@ static struct spl_fixed31_32 fixed31_32_exp_from_taylor_series(struct spl_fixed3 n + 1); /* TODO find correct res */ - ASSERT(spl_fixpt_lt(arg, spl_fixpt_one)); + SPL_ASSERT(spl_fixpt_lt(arg, spl_fixpt_one)); do res = spl_fixpt_add( @@ -337,9 +337,9 @@ struct spl_fixed31_32 spl_fixpt_exp(struct spl_fixed31_32 arg) spl_fixpt_ln2, m)); - ASSERT(m != 0); + SPL_ASSERT(m != 0); - ASSERT(spl_fixpt_lt( + SPL_ASSERT(spl_fixpt_lt( spl_fixpt_abs(r), spl_fixpt_one)); @@ -364,7 +364,7 @@ struct spl_fixed31_32 spl_fixpt_log(struct spl_fixed31_32 arg) struct spl_fixed31_32 error; - ASSERT(arg.value > 0); + SPL_ASSERT(arg.value > 0); /* TODO if arg is negative, return NaN */ /* TODO if arg is zero, return -INF */ diff --git a/drivers/gpu/drm/amd/display/dc/spl/spl_fixpt31_32.h b/drivers/gpu/drm/amd/display/dc/spl/spl_fixpt31_32.h index 8a045e2f8699..ed2647f9a099 100644 --- a/drivers/gpu/drm/amd/display/dc/spl/spl_fixpt31_32.h +++ b/drivers/gpu/drm/amd/display/dc/spl/spl_fixpt31_32.h @@ -5,11 +5,8 @@ #ifndef __SPL_FIXED31_32_H__ #define __SPL_FIXED31_32_H__ -#include "os_types.h" +#include "spl_debug.h" #include "spl_os_types.h" // swap -#ifndef ASSERT -#define ASSERT(_bool) ((void *)0) -#endif #ifndef LLONG_MAX #define LLONG_MAX 9223372036854775807ll @@ -194,7 +191,7 @@ static inline struct spl_fixed31_32 spl_fixpt_clamp( */ static inline struct spl_fixed31_32 spl_fixpt_shl(struct spl_fixed31_32 arg, unsigned char shift) { - ASSERT(((arg.value >= 0) && (arg.value <= LLONG_MAX >> shift)) || + SPL_ASSERT(((arg.value >= 0) && (arg.value <= LLONG_MAX >> shift)) || ((arg.value < 0) && (arg.value >= ~(LLONG_MAX >> shift)))); arg.value = arg.value << shift; @@ -231,7 +228,7 @@ static inline struct spl_fixed31_32 spl_fixpt_add(struct spl_fixed31_32 arg1, st { struct spl_fixed31_32 res; - ASSERT(((arg1.value >= 0) && (LLONG_MAX - arg1.value >= arg2.value)) || + SPL_ASSERT(((arg1.value >= 0) && (LLONG_MAX - arg1.value >= arg2.value)) || ((arg1.value < 0) && (LLONG_MIN - arg1.value <= arg2.value))); res.value = arg1.value + arg2.value; @@ -256,7 +253,7 @@ static inline struct spl_fixed31_32 spl_fixpt_sub(struct spl_fixed31_32 arg1, st { struct spl_fixed31_32 res; - ASSERT(((arg2.value >= 0) && (LLONG_MIN + arg2.value <= arg1.value)) || + SPL_ASSERT(((arg2.value >= 0) && (LLONG_MIN + arg2.value <= arg1.value)) || ((arg2.value < 0) && (LLONG_MAX + arg2.value >= arg1.value))); res.value = arg1.value - arg2.value; @@ -448,7 +445,7 @@ static inline int spl_fixpt_round(struct spl_fixed31_32 arg) const long long summand = spl_fixpt_half.value; - ASSERT(LLONG_MAX - (long long)arg_value >= summand); + SPL_ASSERT(LLONG_MAX - (long long)arg_value >= summand); arg_value += summand; @@ -469,7 +466,7 @@ static inline int spl_fixpt_ceil(struct spl_fixed31_32 arg) const long long summand = spl_fixpt_one.value - spl_fixpt_epsilon.value; - ASSERT(LLONG_MAX - (long long)arg_value >= summand); + SPL_ASSERT(LLONG_MAX - (long long)arg_value >= summand); arg_value += summand; @@ -504,7 +501,7 @@ static inline struct spl_fixed31_32 spl_fixpt_truncate(struct spl_fixed31_32 arg bool negative = arg.value < 0; if (frac_bits >= FIXED31_32_BITS_PER_FRACTIONAL_PART) { - ASSERT(frac_bits == FIXED31_32_BITS_PER_FRACTIONAL_PART); + SPL_ASSERT(frac_bits == FIXED31_32_BITS_PER_FRACTIONAL_PART); return arg; } diff --git a/drivers/gpu/drm/amd/display/dc/spl/spl_os_types.h b/drivers/gpu/drm/amd/display/dc/spl/spl_os_types.h index 709706ed4f2c..2e6ba71960ac 100644 --- a/drivers/gpu/drm/amd/display/dc/spl/spl_os_types.h +++ b/drivers/gpu/drm/amd/display/dc/spl/spl_os_types.h @@ -6,6 +6,8 @@ #ifndef _SPL_OS_TYPES_H_ #define _SPL_OS_TYPES_H_ +#include "spl_debug.h" + #include #include #include @@ -18,7 +20,6 @@ * general debug capabilities * */ -#define SPL_BREAK_TO_DEBUGGER() ASSERT(0) static inline uint64_t spl_div_u64_rem(uint64_t dividend, uint32_t divisor, uint32_t *remainder) { diff --git a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h index fe5b6f7a3eb1..ff27229cc3a4 100644 --- a/drivers/gpu/drm/amd/display/dmub/dmub_srv.h +++ b/drivers/gpu/drm/amd/display/dmub/dmub_srv.h @@ -570,6 +570,14 @@ struct dmub_notification { }; }; +/* enum dmub_ips_mode - IPS mode identifier */ +enum dmub_ips_mode { + DMUB_IPS_MODE_IPS1_MAX = 0, + DMUB_IPS_MODE_IPS2, + DMUB_IPS_MODE_IPS1_RCG, + DMUB_IPS_MODE_IPS1_ONO2_ON +}; + /** * DMUB firmware version helper macro - useful for checking if the version * of a firmware to know if feature or functionality is supported or present. diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index ebcf68bfae2b..6edd3d34c7b5 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -170,6 +170,11 @@ #pragma pack(push, 1) #define ABM_NUM_OF_ACE_SEGMENTS 5 +/** + * Debug FW state offset + */ +#define DMUB_DEBUG_FW_STATE_OFFSET 0x300 + union abm_flags { struct { /** @@ -747,7 +752,8 @@ union dmub_shared_state_ips_driver_signals { uint32_t allow_ips1 : 1; /**< 1 is IPS1 is allowed */ uint32_t allow_ips2 : 1; /**< 1 is IPS1 is allowed */ uint32_t allow_z10 : 1; /**< 1 if Z10 is allowed */ - uint32_t reserved_bits : 28; /**< Reversed bits */ + uint32_t allow_idle : 1; /**< 1 if driver is allowing idle */ + uint32_t reserved_bits : 27; /**< Reversed bits */ } bits; uint32_t all; }; @@ -1050,12 +1056,107 @@ enum dmub_gpint_command { */ DMUB_GPINT__GET_TRACE_BUFFER_MASK_WORD3 = 119, + /** + * DESC: Set IPS residency measurement + * ARGS: 0 - Disable ips measurement + * 1 - Enable ips measurement + */ + DMUB_GPINT__IPS_RESIDENCY = 121, + /** * DESC: Enable measurements for various task duration * ARGS: 0 - Disable measurement * 1 - Enable measurement */ DMUB_GPINT__TRACE_DMUB_WAKE_ACTIVITY = 123, + + /** + * DESC: Gets IPS residency in microseconds + * ARGS: 0 - Return IPS1 residency + * 1 - Return IPS2 residency + * 2 - Return IPS1_RCG residency + * 3 - Return IPS1_ONO2_ON residency + * RETURN: Total residency in microseconds - lower 32 bits + */ + DMUB_GPINT__GET_IPS_RESIDENCY_DURATION_US_LO = 124, + + /** + * DESC: Gets IPS1 histogram counts + * ARGS: Bucket index + * RETURN: Total count for the bucket + */ + DMUB_GPINT__GET_IPS1_HISTOGRAM_COUNTER = 125, + + /** + * DESC: Gets IPS2 histogram counts + * ARGS: Bucket index + * RETURN: Total count for the bucket + */ + DMUB_GPINT__GET_IPS2_HISTOGRAM_COUNTER = 126, + + /** + * DESC: Gets IPS residency + * ARGS: 0 - Return IPS1 residency + * 1 - Return IPS2 residency + * 2 - Return IPS1_RCG residency + * 3 - Return IPS1_ONO2_ON residency + * RETURN: Total residency in milli-percent. + */ + DMUB_GPINT__GET_IPS_RESIDENCY_PERCENT = 127, + + /** + * DESC: Gets IPS1_RCG histogram counts + * ARGS: Bucket index + * RETURN: Total count for the bucket + */ + DMUB_GPINT__GET_IPS1_RCG_HISTOGRAM_COUNTER = 128, + + /** + * DESC: Gets IPS1_ONO2_ON histogram counts + * ARGS: Bucket index + * RETURN: Total count for the bucket + */ + DMUB_GPINT__GET_IPS1_ONO2_ON_HISTOGRAM_COUNTER = 129, + + /** + * DESC: Gets IPS entry counter during residency measurement + * ARGS: 0 - Return IPS1 entry counts + * 1 - Return IPS2 entry counts + * 2 - Return IPS1_RCG entry counts + * 3 - Return IPS2_ONO2_ON entry counts + * RETURN: Entry counter for selected IPS mode + */ + DMUB_GPINT__GET_IPS_RESIDENCY_ENTRY_COUNTER = 130, + + /** + * DESC: Gets IPS inactive residency in microseconds + * ARGS: 0 - Return IPS1_MAX residency + * 1 - Return IPS2 residency + * 2 - Return IPS1_RCG residency + * 3 - Return IPS1_ONO2_ON residency + * RETURN: Total inactive residency in microseconds - lower 32 bits + */ + DMUB_GPINT__GET_IPS_INACTIVE_RESIDENCY_DURATION_US_LO = 131, + + /** + * DESC: Gets IPS inactive residency in microseconds + * ARGS: 0 - Return IPS1_MAX residency + * 1 - Return IPS2 residency + * 2 - Return IPS1_RCG residency + * 3 - Return IPS1_ONO2_ON residency + * RETURN: Total inactive residency in microseconds - upper 32 bits + */ + DMUB_GPINT__GET_IPS_INACTIVE_RESIDENCY_DURATION_US_HI = 132, + + /** + * DESC: Gets IPS residency in microseconds + * ARGS: 0 - Return IPS1 residency + * 1 - Return IPS2 residency + * 2 - Return IPS1_RCG residency + * 3 - Return IPS1_ONO2_ON residency + * RETURN: Total residency in microseconds - upper 32 bits + */ + DMUB_GPINT__GET_IPS_RESIDENCY_DURATION_US_HI = 133, }; /** @@ -1306,9 +1407,10 @@ enum dmub_out_cmd_type { /* DMUB_CMD__DPIA command sub-types. */ enum dmub_cmd_dpia_type { DMUB_CMD__DPIA_DIG1_DPIA_CONTROL = 0, - DMUB_CMD__DPIA_SET_CONFIG_ACCESS = 1, + DMUB_CMD__DPIA_SET_CONFIG_ACCESS = 1, // will be replaced by DPIA_SET_CONFIG_REQUEST DMUB_CMD__DPIA_MST_ALLOC_SLOTS = 2, DMUB_CMD__DPIA_SET_TPS_NOTIFICATION = 3, + DMUB_CMD__DPIA_SET_CONFIG_REQUEST = 4, }; /* DMUB_OUT_CMD__DPIA_NOTIFICATION command types. */ @@ -2097,7 +2199,7 @@ struct dmub_rb_cmd_dig1_dpia_control { }; /** - * SET_CONFIG Command Payload + * SET_CONFIG Command Payload (deprecated) */ struct set_config_cmd_payload { uint8_t msg_type; /* set config message type */ @@ -2105,7 +2207,7 @@ struct set_config_cmd_payload { }; /** - * Data passed from driver to FW in a DMUB_CMD__DPIA_SET_CONFIG_ACCESS command. + * Data passed from driver to FW in a DMUB_CMD__DPIA_SET_CONFIG_ACCESS command. (deprecated) */ struct dmub_cmd_set_config_control_data { struct set_config_cmd_payload cmd_pkt; @@ -2113,6 +2215,17 @@ struct dmub_cmd_set_config_control_data { uint8_t immed_status; /* Immediate status returned in case of error */ }; +/** + * SET_CONFIG Request Command Payload + */ +struct set_config_request_cmd_payload { + uint8_t instance; /* DPIA instance */ + uint8_t immed_status; /* Immediate status returned in case of error */ + uint8_t msg_type; /* set config message type */ + uint8_t reserved; + uint32_t msg_data; /* set config message data */ +}; + /** * DMUB command structure for SET_CONFIG command. */ @@ -2121,6 +2234,14 @@ struct dmub_rb_cmd_set_config_access { struct dmub_cmd_set_config_control_data set_config_control; /* set config data */ }; +/** + * DMUB command structure for SET_CONFIG request command. + */ +struct dmub_rb_cmd_set_config_request { + struct dmub_cmd_header header; /* header */ + struct set_config_request_cmd_payload payload; /* set config request payload */ +}; + /** * Data passed from driver to FW in a DMUB_CMD__DPIA_MST_ALLOC_SLOTS command. */ @@ -4315,10 +4436,38 @@ struct dmub_cmd_abm_set_backlight_data { */ uint8_t panel_mask; + /** + * Backlight control type. + * Value 0 is PWM backlight control. + * Value 1 is VAUX backlight control. + * Value 2 is AMD DPCD AUX backlight control. + */ + uint8_t backlight_control_type; + /** * Explicit padding to 4 byte boundary. */ - uint8_t pad[2]; + uint8_t pad[1]; + + /** + * Minimum luminance in nits. + */ + uint32_t min_luminance; + + /** + * Maximum luminance in nits. + */ + uint32_t max_luminance; + + /** + * Minimum backlight in pwm. + */ + uint32_t min_backlight_pwm; + + /** + * Maximum backlight in pwm. + */ + uint32_t max_backlight_pwm; }; /** @@ -5318,7 +5467,11 @@ union dmub_rb_cmd { /** * Definition of a DMUB_CMD__DPIA_SET_CONFIG_ACCESS command. */ - struct dmub_rb_cmd_set_config_access set_config_access; + struct dmub_rb_cmd_set_config_access set_config_access; // (deprecated) + /** + * Definition of a DMUB_CMD__DPIA_SET_CONFIG_ACCESS command. + */ + struct dmub_rb_cmd_set_config_request set_config_request; /** * Definition of a DMUB_CMD__DPIA_MST_ALLOC_SLOTS command. */ diff --git a/drivers/gpu/drm/amd/display/include/dpcd_defs.h b/drivers/gpu/drm/amd/display/include/dpcd_defs.h index aee5170f5fb2..de8f3cfed6c8 100644 --- a/drivers/gpu/drm/amd/display/include/dpcd_defs.h +++ b/drivers/gpu/drm/amd/display/include/dpcd_defs.h @@ -164,18 +164,19 @@ enum dpcd_psr_sink_states { PSR_SINK_STATE_SINK_INTERNAL_ERROR = 7, }; -#define DP_SOURCE_SEQUENCE 0x30c -#define DP_SOURCE_TABLE_REVISION 0x310 -#define DP_SOURCE_PAYLOAD_SIZE 0x311 -#define DP_SOURCE_SINK_CAP 0x317 -#define DP_SOURCE_BACKLIGHT_LEVEL 0x320 -#define DP_SOURCE_BACKLIGHT_CURRENT_PEAK 0x326 -#define DP_SOURCE_BACKLIGHT_CONTROL 0x32E -#define DP_SOURCE_BACKLIGHT_ENABLE 0x32F -#define DP_SOURCE_MINIMUM_HBLANK_SUPPORTED 0x340 +#define DP_SOURCE_SEQUENCE 0x30C +#define DP_SOURCE_TABLE_REVISION 0x310 +#define DP_SOURCE_PAYLOAD_SIZE 0x311 +#define DP_SOURCE_SINK_CAP 0x317 +#define DP_SOURCE_BACKLIGHT_LEVEL 0x320 +#define DP_SOURCE_BACKLIGHT_CURRENT_PEAK 0x326 +#define DP_SOURCE_BACKLIGHT_CONTROL 0x32E +#define DP_SOURCE_BACKLIGHT_ENABLE 0x32F +#define DP_SOURCE_MINIMUM_HBLANK_SUPPORTED 0x340 #define DP_SINK_PR_REPLAY_STATUS 0x378 #define DP_SINK_PR_PIXEL_DEVIATION_PER_LINE 0x379 #define DP_SINK_PR_MAX_NUMBER_OF_DEVIATION_LINE 0x37A +#define DP_SINK_EMISSION_RATE 0x37E /* Remove once drm_dp_helper.h is updated upstream */ #ifndef DP_TOTAL_LTTPR_CNT diff --git a/drivers/gpu/drm/amd/display/include/logger_types.h b/drivers/gpu/drm/amd/display/include/logger_types.h index a48d564d1660..4d68c1c6e210 100644 --- a/drivers/gpu/drm/amd/display/include/logger_types.h +++ b/drivers/gpu/drm/amd/display/include/logger_types.h @@ -61,11 +61,13 @@ #define DC_LOG_ALL_TF_CHANNELS(...) pr_debug("[GAMMA]:"__VA_ARGS__) #define DC_LOG_DSC(...) drm_dbg_dp((DC_LOGGER)->dev, __VA_ARGS__) #define DC_LOG_SMU(...) pr_debug("[SMU_MSG]:"__VA_ARGS__) -#define DC_LOG_MALL(...) pr_debug("[MALL]:"__VA_ARGS__) #define DC_LOG_DWB(...) drm_dbg((DC_LOGGER)->dev, __VA_ARGS__) #define DC_LOG_DP2(...) drm_dbg_dp((DC_LOGGER)->dev, __VA_ARGS__) #define DC_LOG_AUTO_DPM_TEST(...) pr_debug("[AutoDPMTest]: "__VA_ARGS__) #define DC_LOG_IPS(...) pr_debug("[IPS]: "__VA_ARGS__) +#define DC_LOG_MALL(...) pr_debug("[MALL]:"__VA_ARGS__) +#define DC_LOG_REGISTER_READ(...) pr_debug("[REGISTER_READ]: "__VA_ARGS__) +#define DC_LOG_REGISTER_WRITE(...) pr_debug("[REGISTER_WRITE]: "__VA_ARGS__) struct dc_log_buffer_ctx { char *buf; diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index bbd259cea4f4..fc4268729017 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -48,6 +48,7 @@ #define VSYNCS_BETWEEN_FLIP_THRESHOLD 2 #define FREESYNC_CONSEC_FLIP_AFTER_VSYNC 5 #define FREESYNC_VSYNC_TO_FLIP_DELTA_IN_US 500 +#define MICRO_HZ_TO_HZ(x) (x / 1000000) struct core_freesync { struct mod_freesync public; @@ -132,9 +133,19 @@ unsigned int mod_freesync_calc_v_total_from_refresh( ((unsigned int)(div64_u64((1000000000ULL * 1000000), refresh_in_uhz))); - v_total = div64_u64(div64_u64(((unsigned long long)( - frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)), - stream->timing.h_total) + 500000, 1000000); + if (MICRO_HZ_TO_HZ(refresh_in_uhz) <= stream->timing.min_refresh_in_uhz) { + /* When the target refresh rate is the minimum panel refresh rate, + * round down the vtotal value to avoid stretching vblank over + * panel's vtotal boundary. + */ + v_total = div64_u64(div64_u64(((unsigned long long)( + frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)), + stream->timing.h_total), 1000000); + } else { + v_total = div64_u64(div64_u64(((unsigned long long)( + frame_duration_in_ns) * (stream->timing.pix_clk_100hz / 10)), + stream->timing.h_total) + 500000, 1000000); + } /* v_total cannot be less than nominal */ if (v_total < stream->timing.v_total) { diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c index 3cd52e7a9c77..95838c7ab054 100644 --- a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c @@ -841,6 +841,8 @@ bool is_psr_su_specific_panel(struct dc_link *link) isPSRSUSupported = false; else if (dpcd_caps->sink_dev_id_str[1] == 0x08 && dpcd_caps->sink_dev_id_str[0] == 0x03) isPSRSUSupported = false; + else if (dpcd_caps->sink_dev_id_str[1] == 0x08 && dpcd_caps->sink_dev_id_str[0] == 0x01) + isPSRSUSupported = false; else if (dpcd_caps->psr_info.force_psrsu_cap == 0x1) isPSRSUSupported = true; } diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index 3f91926a50e9..7eefcb0f5070 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -28,6 +28,8 @@ #define AMD_MAX_USEC_TIMEOUT 1000000 /* 1000 ms */ +struct amdgpu_ip_block; + /* * Chip flags @@ -337,6 +339,11 @@ enum DC_DEBUG_MASK { * @DC_FORCE_IPS_ENABLE: If set, force enable all IPS, all the time. */ DC_FORCE_IPS_ENABLE = 0x4000, + /** + * @DC_DISABLE_ACPI_EDID: If set, don't attempt to fetch EDID for + * eDP display from ACPI _DDC method. + */ + DC_DISABLE_ACPI_EDID = 0x8000, }; enum amd_dpm_forced_level; @@ -377,30 +384,30 @@ enum amd_dpm_forced_level; */ struct amd_ip_funcs { char *name; - int (*early_init)(void *handle); - int (*late_init)(void *handle); - int (*sw_init)(void *handle); - int (*sw_fini)(void *handle); - int (*early_fini)(void *handle); - int (*hw_init)(void *handle); - int (*hw_fini)(void *handle); - void (*late_fini)(void *handle); - int (*prepare_suspend)(void *handle); - int (*suspend)(void *handle); - int (*resume)(void *handle); + int (*early_init)(struct amdgpu_ip_block *ip_block); + int (*late_init)(struct amdgpu_ip_block *ip_block); + int (*sw_init)(struct amdgpu_ip_block *ip_block); + int (*sw_fini)(struct amdgpu_ip_block *ip_block); + int (*early_fini)(struct amdgpu_ip_block *ip_block); + int (*hw_init)(struct amdgpu_ip_block *ip_block); + int (*hw_fini)(struct amdgpu_ip_block *ip_block); + void (*late_fini)(struct amdgpu_ip_block *ip_block); + int (*prepare_suspend)(struct amdgpu_ip_block *ip_block); + int (*suspend)(struct amdgpu_ip_block *ip_block); + int (*resume)(struct amdgpu_ip_block *ip_block); bool (*is_idle)(void *handle); - int (*wait_for_idle)(void *handle); - bool (*check_soft_reset)(void *handle); - int (*pre_soft_reset)(void *handle); - int (*soft_reset)(void *handle); - int (*post_soft_reset)(void *handle); + int (*wait_for_idle)(struct amdgpu_ip_block *ip_block); + bool (*check_soft_reset)(struct amdgpu_ip_block *ip_block); + int (*pre_soft_reset)(struct amdgpu_ip_block *ip_block); + int (*soft_reset)(struct amdgpu_ip_block *ip_block); + int (*post_soft_reset)(struct amdgpu_ip_block *ip_block); int (*set_clockgating_state)(void *handle, enum amd_clockgating_state state); int (*set_powergating_state)(void *handle, enum amd_powergating_state state); void (*get_clockgating_state)(void *handle, u64 *flags); - void (*dump_ip_state)(void *handle); - void (*print_ip_state)(void *handle, struct drm_printer *p); + void (*dump_ip_state)(struct amdgpu_ip_block *ip_block); + void (*print_ip_state)(struct amdgpu_ip_block *ip_block, struct drm_printer *p); }; diff --git a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_0_offset.h index 2c3ce243861a..380e44230bda 100644 --- a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_0_offset.h +++ b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_1_0_offset.h @@ -1232,6 +1232,29 @@ #define mmMC_VM_MX_L1_PERFCOUNTER_HI 0x059d #define mmMC_VM_MX_L1_PERFCOUNTER_HI_BASE_IDX 0 +// Stand Alone Walker Registers +#define VMC_TAP_PDE_REQUEST_SNOOP_OFFSET 8 +#define VMC_TAP_PTE_REQUEST_SNOOP_OFFSET 11 +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32 0x0606 +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32_BASE_IDX 0 +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32 0x0607 +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32_BASE_IDX 0 +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_START_ADDR_LO32 0x0608 +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_START_ADDR_LO32_BASE_IDX 0 +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_START_ADDR_HI32 0x0609 +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_START_ADDR_HI32_BASE_IDX 0 +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_END_ADDR_LO32 0x060a +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_END_ADDR_LO32_BASE_IDX 0 +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_END_ADDR_HI32 0x060b +#define mmVM_L2_SAW_CONTEXT0_PAGE_TABLE_END_ADDR_HI32_BASE_IDX 0 +#define mmVM_L2_SAW_CONTEXT0_CNTL 0x0604 +#define mmVM_L2_SAW_CONTEXT0_CNTL_BASE_IDX 0 +#define CONTEXT0_CNTL_ENABLE_OFFSET 0 +#define CONTEXT0_CNTL_PAGE_TABLE_DEPTH_OFFSET 1 +#define mmVM_L2_SAW_CONTEXTS_DISABLE 0x060c +#define mmVM_L2_SAW_CONTEXTS_DISABLE_BASE_IDX 0 +#define mmVM_L2_SAW_CNTL4 0x0603 +#define mmVM_L2_SAW_CNTL4_BASE_IDX 0 // addressBlock: mmhub_utcl2_atcl2dec // base address: 0x69900 diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index 19a48d98830a..2fa71f68205e 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -336,7 +336,8 @@ enum pp_policy_soc_pstate { #define MAX_CLKS 4 #define NUM_VCN 4 #define NUM_JPEG_ENG 32 - +#define MAX_XCC 8 +#define NUM_XCP 8 struct seq_file; enum amd_pp_clock_type; struct amd_pp_simple_clock_info; @@ -350,6 +351,15 @@ struct pp_smu_wm_range_sets; struct pp_smu_nv_clock_table; struct dpm_clocks; +struct amdgpu_xcp_metrics { + /* Utilization Instantaneous (%) */ + u32 gfx_busy_inst[MAX_XCC]; + u16 jpeg_busy[NUM_JPEG_ENG]; + u16 vcn_busy[NUM_VCN]; + /* Utilization Accumulated (%) */ + u64 gfx_busy_acc[MAX_XCC]; +}; + struct amd_pm_funcs { /* export for dpm on ci and si */ int (*pre_set_power_state)(void *handle); @@ -872,6 +882,97 @@ struct gpu_metrics_v1_5 { uint16_t padding; }; +struct gpu_metrics_v1_6 { + struct metrics_table_header common_header; + + /* Temperature (Celsius) */ + uint16_t temperature_hotspot; + uint16_t temperature_mem; + uint16_t temperature_vrsoc; + + /* Power (Watts) */ + uint16_t curr_socket_power; + + /* Utilization (%) */ + uint16_t average_gfx_activity; + uint16_t average_umc_activity; // memory controller + + /* Energy (15.259uJ (2^-16) units) */ + uint64_t energy_accumulator; + + /* Driver attached timestamp (in ns) */ + uint64_t system_clock_counter; + + /* Accumulation cycle counter */ + uint32_t accumulation_counter; + + /* Accumulated throttler residencies */ + uint32_t prochot_residency_acc; + uint32_t ppt_residency_acc; + uint32_t socket_thm_residency_acc; + uint32_t vr_thm_residency_acc; + uint32_t hbm_thm_residency_acc; + + /* Clock Lock Status. Each bit corresponds to clock instance */ + uint32_t gfxclk_lock_status; + + /* Link width (number of lanes) and speed (in 0.1 GT/s) */ + uint16_t pcie_link_width; + uint16_t pcie_link_speed; + + /* XGMI bus width and bitrate (in Gbps) */ + uint16_t xgmi_link_width; + uint16_t xgmi_link_speed; + + /* Utilization Accumulated (%) */ + uint32_t gfx_activity_acc; + uint32_t mem_activity_acc; + + /*PCIE accumulated bandwidth (GB/sec) */ + uint64_t pcie_bandwidth_acc; + + /*PCIE instantaneous bandwidth (GB/sec) */ + uint64_t pcie_bandwidth_inst; + + /* PCIE L0 to recovery state transition accumulated count */ + uint64_t pcie_l0_to_recov_count_acc; + + /* PCIE replay accumulated count */ + uint64_t pcie_replay_count_acc; + + /* PCIE replay rollover accumulated count */ + uint64_t pcie_replay_rover_count_acc; + + /* PCIE NAK sent accumulated count */ + uint32_t pcie_nak_sent_count_acc; + + /* PCIE NAK received accumulated count */ + uint32_t pcie_nak_rcvd_count_acc; + + /* XGMI accumulated data transfer size(KiloBytes) */ + uint64_t xgmi_read_data_acc[NUM_XGMI_LINKS]; + uint64_t xgmi_write_data_acc[NUM_XGMI_LINKS]; + + /* PMFW attached timestamp (10ns resolution) */ + uint64_t firmware_timestamp; + + /* Current clocks (Mhz) */ + uint16_t current_gfxclk[MAX_GFX_CLKS]; + uint16_t current_socclk[MAX_CLKS]; + uint16_t current_vclk0[MAX_CLKS]; + uint16_t current_dclk0[MAX_CLKS]; + uint16_t current_uclk; + + /* Number of current partition */ + uint16_t num_partition; + + /* XCP metrics stats */ + struct amdgpu_xcp_metrics xcp_stats[NUM_XCP]; + + /* PCIE other end recovery counter */ + uint32_t pcie_lc_perf_other_end_recovery; +}; + /* * gpu_metrics_v2_0 is not recommended as it's not naturally aligned. * Use gpu_metrics_v2_1 or later instead. diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c index d5d6ab484e5a..ea940773353c 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c @@ -145,15 +145,12 @@ static ssize_t amdgpu_get_power_dpm_state(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; amdgpu_dpm_get_current_power_state(adev, &pm); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return sysfs_emit(buf, "%s\n", @@ -185,11 +182,9 @@ static ssize_t amdgpu_set_power_dpm_state(struct device *dev, else return -EINVAL; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } amdgpu_dpm_set_power_state(adev, state); @@ -273,15 +268,12 @@ static ssize_t amdgpu_get_power_dpm_force_performance_level(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; level = amdgpu_dpm_get_performance_level(adev); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return sysfs_emit(buf, "%s\n", @@ -336,11 +328,9 @@ static ssize_t amdgpu_set_power_dpm_force_performance_level(struct device *dev, return -EINVAL; } - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } mutex_lock(&adev->pm.stable_pstate_ctx_lock); if (amdgpu_dpm_force_performance_level(adev, level)) { @@ -374,16 +364,13 @@ static ssize_t amdgpu_get_pp_num_states(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; if (amdgpu_dpm_get_pp_num_states(adev, &data)) memset(&data, 0, sizeof(data)); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); buf_len = sysfs_emit(buf, "states: %d\n", data.nums); @@ -412,17 +399,14 @@ static ssize_t amdgpu_get_pp_cur_state(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; amdgpu_dpm_get_current_power_state(adev, &pm); ret = amdgpu_dpm_get_pp_num_states(adev, &data); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); if (ret) @@ -485,11 +469,9 @@ static ssize_t amdgpu_set_pp_force_state(struct device *dev, idx = array_index_nospec(idx, ARRAY_SIZE(data.states)); - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } ret = amdgpu_dpm_get_pp_num_states(adev, &data); if (ret) @@ -544,15 +526,12 @@ static ssize_t amdgpu_get_pp_table(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; size = amdgpu_dpm_get_pp_table(adev, &table); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); if (size <= 0) @@ -580,11 +559,9 @@ static ssize_t amdgpu_set_pp_table(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } ret = amdgpu_dpm_set_pp_table(adev, buf, count); @@ -808,11 +785,9 @@ static ssize_t amdgpu_set_pp_od_clk_voltage(struct device *dev, tmp_str++; } - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } if (amdgpu_dpm_set_fine_grain_clk_vol(adev, type, @@ -865,11 +840,9 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; for (clk_index = 0 ; clk_index < 6 ; clk_index++) { ret = amdgpu_dpm_emit_clock_levels(adev, od_clocks[clk_index], buf, &size); @@ -888,7 +861,6 @@ static ssize_t amdgpu_get_pp_od_clk_voltage(struct device *dev, if (size == 0) size = sysfs_emit(buf, "\n"); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return size; @@ -929,11 +901,9 @@ static ssize_t amdgpu_set_pp_features(struct device *dev, if (ret) return -EINVAL; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } ret = amdgpu_dpm_set_ppfeature_status(adev, featuremask); @@ -960,17 +930,14 @@ static ssize_t amdgpu_get_pp_features(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; size = amdgpu_dpm_get_ppfeature_status(adev, buf); if (size <= 0) size = sysfs_emit(buf, "\n"); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return size; @@ -1029,11 +996,9 @@ static ssize_t amdgpu_get_pp_dpm_clock(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; ret = amdgpu_dpm_emit_clock_levels(adev, type, buf, &size); if (ret == -ENOENT) @@ -1042,7 +1007,6 @@ static ssize_t amdgpu_get_pp_dpm_clock(struct device *dev, if (size == 0) size = sysfs_emit(buf, "\n"); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return size; @@ -1102,11 +1066,9 @@ static ssize_t amdgpu_set_pp_dpm_clock(struct device *dev, if (ret) return ret; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } ret = amdgpu_dpm_force_clock_level(adev, type, mask); @@ -1283,15 +1245,12 @@ static ssize_t amdgpu_get_pp_sclk_od(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; value = amdgpu_dpm_get_sclk_od(adev); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return sysfs_emit(buf, "%d\n", value); @@ -1317,11 +1276,9 @@ static ssize_t amdgpu_set_pp_sclk_od(struct device *dev, if (ret) return -EINVAL; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } amdgpu_dpm_set_sclk_od(adev, (uint32_t)value); @@ -1345,15 +1302,12 @@ static ssize_t amdgpu_get_pp_mclk_od(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; value = amdgpu_dpm_get_mclk_od(adev); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return sysfs_emit(buf, "%d\n", value); @@ -1379,11 +1333,9 @@ static ssize_t amdgpu_set_pp_mclk_od(struct device *dev, if (ret) return -EINVAL; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } amdgpu_dpm_set_mclk_od(adev, (uint32_t)value); @@ -1427,17 +1379,14 @@ static ssize_t amdgpu_get_pp_power_profile_mode(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; size = amdgpu_dpm_get_power_profile_mode(adev, buf); if (size <= 0) size = sysfs_emit(buf, "\n"); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return size; @@ -1492,11 +1441,9 @@ static ssize_t amdgpu_set_pp_power_profile_mode(struct device *dev, } parameter[parameter_size] = profile_mode; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } ret = amdgpu_dpm_set_power_profile_mode(adev, parameter, parameter_size); @@ -1520,16 +1467,13 @@ static int amdgpu_hwmon_get_sensor_generic(struct amdgpu_device *adev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (r < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return r; - } + r = pm_runtime_get_if_active(adev->dev); + if (r <= 0) + return r ?: -EPERM; /* get the sensor value */ r = amdgpu_dpm_read_sensor(adev, sensor, query, &size); - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); return r; @@ -1639,15 +1583,12 @@ static ssize_t amdgpu_get_pcie_bw(struct device *dev, if (!adev->asic_funcs->get_pcie_usage) return -ENODATA; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; amdgpu_asic_get_pcie_usage(adev, &count0, &count1); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return sysfs_emit(buf, "%llu %llu %i\n", @@ -1770,11 +1711,9 @@ static ssize_t amdgpu_get_apu_thermal_cap(struct device *dev, struct drm_device *ddev = dev_get_drvdata(dev); struct amdgpu_device *adev = drm_to_adev(ddev); - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; ret = amdgpu_dpm_get_apu_thermal_limit(adev, &limit); if (!ret) @@ -1782,7 +1721,6 @@ static ssize_t amdgpu_get_apu_thermal_cap(struct device *dev, else size = sysfs_emit(buf, "failed to get thermal limit\n"); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return size; @@ -1807,14 +1745,14 @@ static ssize_t amdgpu_set_apu_thermal_cap(struct device *dev, return -EINVAL; } - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } ret = amdgpu_dpm_set_apu_thermal_limit(adev, value); if (ret) { + pm_runtime_mark_last_busy(ddev->dev); + pm_runtime_put_autosuspend(ddev->dev); dev_err(dev, "failed to update thermal limit\n"); return ret; } @@ -1849,15 +1787,12 @@ static ssize_t amdgpu_get_pm_metrics(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; size = amdgpu_dpm_get_pm_metrics(adev, buf, PAGE_SIZE); - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return size; @@ -1890,11 +1825,9 @@ static ssize_t amdgpu_get_gpu_metrics(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); - return ret; - } + ret = pm_runtime_get_if_active(ddev->dev); + if (ret <= 0) + return ret ?: -EPERM; size = amdgpu_dpm_get_gpu_metrics(adev, &gpu_metrics); if (size <= 0) @@ -1906,7 +1839,6 @@ static ssize_t amdgpu_get_gpu_metrics(struct device *dev, memcpy(buf, gpu_metrics, size); out: - pm_runtime_mark_last_busy(ddev->dev); pm_runtime_put_autosuspend(ddev->dev); return size; @@ -2008,11 +1940,9 @@ static ssize_t amdgpu_set_smartshift_bias(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - r = pm_runtime_get_sync(ddev->dev); - if (r < 0) { - pm_runtime_put_autosuspend(ddev->dev); + r = pm_runtime_resume_and_get(ddev->dev); + if (r < 0) return r; - } r = kstrtoint(buf, 10, &bias); if (r) @@ -2335,11 +2265,9 @@ static ssize_t amdgpu_set_pm_policy_attr(struct device *dev, policy_attr = container_of(attr, struct amdgpu_pm_policy_attr, dev_attr); - ret = pm_runtime_get_sync(ddev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(ddev->dev); + ret = pm_runtime_resume_and_get(ddev->dev); + if (ret < 0) return ret; - } ret = amdgpu_dpm_set_pm_policy(adev, policy_attr->id, val); @@ -2772,15 +2700,12 @@ static ssize_t amdgpu_hwmon_get_pwm1_enable(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return ret; - } + ret = pm_runtime_get_if_active(adev->dev); + if (ret <= 0) + return ret ?: -EPERM; ret = amdgpu_dpm_get_fan_control_mode(adev, &pwm_mode); - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); if (ret) @@ -2817,11 +2742,9 @@ static ssize_t amdgpu_hwmon_set_pwm1_enable(struct device *dev, else return -EINVAL; - ret = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + ret = pm_runtime_resume_and_get(adev->dev); + if (ret < 0) return ret; - } ret = amdgpu_dpm_set_fan_control_mode(adev, pwm_mode); @@ -2866,11 +2789,9 @@ static ssize_t amdgpu_hwmon_set_pwm1(struct device *dev, if (err) return err; - err = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (err < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + err = pm_runtime_resume_and_get(adev->dev); + if (err < 0) return err; - } err = amdgpu_dpm_get_fan_control_mode(adev, &pwm_mode); if (err) @@ -2907,15 +2828,12 @@ static ssize_t amdgpu_hwmon_get_pwm1(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - err = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (err < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return err; - } + err = pm_runtime_get_if_active(adev->dev); + if (err <= 0) + return err ?: -EPERM; err = amdgpu_dpm_get_fan_speed_pwm(adev, &speed); - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); if (err) @@ -2937,15 +2855,12 @@ static ssize_t amdgpu_hwmon_get_fan1_input(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - err = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (err < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return err; - } + err = pm_runtime_get_if_active(adev->dev); + if (err <= 0) + return err ?: -EPERM; err = amdgpu_dpm_get_fan_speed_rpm(adev, &speed); - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); if (err) @@ -3001,15 +2916,12 @@ static ssize_t amdgpu_hwmon_get_fan1_target(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - err = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (err < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return err; - } + err = pm_runtime_get_if_active(adev->dev); + if (err <= 0) + return err ?: -EPERM; err = amdgpu_dpm_get_fan_speed_rpm(adev, &rpm); - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); if (err) @@ -3036,11 +2948,9 @@ static ssize_t amdgpu_hwmon_set_fan1_target(struct device *dev, if (err) return err; - err = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (err < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + err = pm_runtime_resume_and_get(adev->dev); + if (err < 0) return err; - } err = amdgpu_dpm_get_fan_control_mode(adev, &pwm_mode); if (err) @@ -3076,15 +2986,12 @@ static ssize_t amdgpu_hwmon_get_fan1_enable(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return ret; - } + ret = pm_runtime_get_if_active(adev->dev); + if (ret <= 0) + return ret ?: -EPERM; ret = amdgpu_dpm_get_fan_control_mode(adev, &pwm_mode); - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); if (ret) @@ -3119,11 +3026,9 @@ static ssize_t amdgpu_hwmon_set_fan1_enable(struct device *dev, else return -EINVAL; - err = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (err < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + err = pm_runtime_resume_and_get(adev->dev); + if (err < 0) return err; - } err = amdgpu_dpm_set_fan_control_mode(adev, pwm_mode); @@ -3248,11 +3153,9 @@ static ssize_t amdgpu_hwmon_show_power_cap_generic(struct device *dev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - r = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (r < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); - return r; - } + r = pm_runtime_get_if_active(adev->dev); + if (r <= 0) + return r ?: -EPERM; r = amdgpu_dpm_get_power_limit(adev, &limit, pp_limit_level, power_type); @@ -3262,7 +3165,6 @@ static ssize_t amdgpu_hwmon_show_power_cap_generic(struct device *dev, else size = sysfs_emit(buf, "\n"); - pm_runtime_mark_last_busy(adev_to_drm(adev)->dev); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); return size; @@ -3339,11 +3241,9 @@ static ssize_t amdgpu_hwmon_set_power_cap(struct device *dev, value = value / 1000000; /* convert to Watt */ value |= limit_type << 24; - err = pm_runtime_get_sync(adev_to_drm(adev)->dev); - if (err < 0) { - pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); + err = pm_runtime_resume_and_get(adev->dev); + if (err < 0) return err; - } err = amdgpu_dpm_set_power_limit(adev, value); @@ -3787,17 +3687,14 @@ static int amdgpu_retrieve_od_settings(struct amdgpu_device *adev, if (adev->in_suspend && !adev->in_runpm) return -EPERM; - ret = pm_runtime_get_sync(adev->dev); - if (ret < 0) { - pm_runtime_put_autosuspend(adev->dev); - return ret; - } + ret = pm_runtime_get_if_active(adev->dev); + if (ret <= 0) + return ret ?: -EPERM; size = amdgpu_dpm_print_clock_levels(adev, od_type, buf); if (size == 0) size = sysfs_emit(buf, "\n"); - pm_runtime_mark_last_busy(adev->dev); pm_runtime_put_autosuspend(adev->dev); return size; @@ -3879,23 +3776,23 @@ amdgpu_distribute_custom_od_settings(struct amdgpu_device *adev, if (ret) return ret; - ret = pm_runtime_get_sync(adev->dev); + ret = pm_runtime_resume_and_get(adev->dev); if (ret < 0) - goto err_out0; + return ret; ret = amdgpu_dpm_odn_edit_dpm_table(adev, cmd_type, parameter, parameter_size); if (ret) - goto err_out1; + goto err_out; if (cmd_type == PP_OD_COMMIT_DPM_TABLE) { ret = amdgpu_dpm_dispatch_task(adev, AMD_PP_TASK_READJUST_POWER_STATE, NULL); if (ret) - goto err_out1; + goto err_out; } pm_runtime_mark_last_busy(adev->dev); @@ -3903,9 +3800,8 @@ amdgpu_distribute_custom_od_settings(struct amdgpu_device *adev, return count; -err_out1: +err_out: pm_runtime_mark_last_busy(adev->dev); -err_out0: pm_runtime_put_autosuspend(adev->dev); return ret; @@ -4758,11 +4654,9 @@ static int amdgpu_debugfs_pm_info_show(struct seq_file *m, void *unused) if (adev->in_suspend && !adev->in_runpm) return -EPERM; - r = pm_runtime_get_sync(dev->dev); - if (r < 0) { - pm_runtime_put_autosuspend(dev->dev); + r = pm_runtime_resume_and_get(dev->dev); + if (r < 0) return r; - } if (amdgpu_dpm_debugfs_print_current_performance_level(adev, m)) { r = amdgpu_debugfs_pm_info_pp(m, adev); @@ -4777,7 +4671,6 @@ static int amdgpu_debugfs_pm_info_show(struct seq_file *m, void *unused) seq_printf(m, "\n"); out: - pm_runtime_mark_last_busy(dev->dev); pm_runtime_put_autosuspend(dev->dev); return r; diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c index e8b6989a40f3..8908646ad620 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/kv_dpm.c @@ -2954,9 +2954,9 @@ static int kv_dpm_get_temp(void *handle) return actual_temp; } -static int kv_dpm_early_init(void *handle) +static int kv_dpm_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->powerplay.pp_funcs = &kv_dpm_funcs; adev->powerplay.pp_handle = adev; @@ -2965,10 +2965,10 @@ static int kv_dpm_early_init(void *handle) return 0; } -static int kv_dpm_late_init(void *handle) +static int kv_dpm_late_init(struct amdgpu_ip_block *ip_block) { /* powerdown unused blocks for now */ - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (!adev->pm.dpm_enabled) return 0; @@ -2979,11 +2979,10 @@ static int kv_dpm_late_init(void *handle) return 0; } -static int kv_dpm_sw_init(void *handle) +static int kv_dpm_sw_init(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - + struct amdgpu_device *adev = ip_block->adev; ret = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, 230, &adev->pm.dpm.thermal.irq); if (ret) @@ -3024,9 +3023,9 @@ dpm_failed: return ret; } -static int kv_dpm_sw_fini(void *handle) +static int kv_dpm_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; flush_work(&adev->pm.dpm.thermal.work); @@ -3035,10 +3034,10 @@ static int kv_dpm_sw_fini(void *handle) return 0; } -static int kv_dpm_hw_init(void *handle) +static int kv_dpm_hw_init(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (!amdgpu_dpm) return 0; @@ -3053,9 +3052,9 @@ static int kv_dpm_hw_init(void *handle) return ret; } -static int kv_dpm_hw_fini(void *handle) +static int kv_dpm_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->pm.dpm_enabled) kv_dpm_disable(adev); @@ -3063,9 +3062,9 @@ static int kv_dpm_hw_fini(void *handle) return 0; } -static int kv_dpm_suspend(void *handle) +static int kv_dpm_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->pm.dpm_enabled) { /* disable dpm */ @@ -3076,10 +3075,10 @@ static int kv_dpm_suspend(void *handle) return 0; } -static int kv_dpm_resume(void *handle) +static int kv_dpm_resume(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->pm.dpm_enabled) { /* asic init will reset to the boot state */ @@ -3100,17 +3099,6 @@ static bool kv_dpm_is_idle(void *handle) return true; } -static int kv_dpm_wait_for_idle(void *handle) -{ - return 0; -} - - -static int kv_dpm_soft_reset(void *handle) -{ - return 0; -} - static int kv_dpm_set_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, unsigned type, @@ -3314,12 +3302,8 @@ static const struct amd_ip_funcs kv_dpm_ip_funcs = { .suspend = kv_dpm_suspend, .resume = kv_dpm_resume, .is_idle = kv_dpm_is_idle, - .wait_for_idle = kv_dpm_wait_for_idle, - .soft_reset = kv_dpm_soft_reset, .set_clockgating_state = kv_dpm_set_clockgating_state, .set_powergating_state = kv_dpm_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; const struct amdgpu_ip_block_version kv_smu_ip_block = { diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c index a1baa13ab2c2..ee23a0f897c5 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c @@ -4755,13 +4755,15 @@ static int si_populate_memory_timing_parameters(struct amdgpu_device *adev, u32 dram_timing; u32 dram_timing2; u32 burst_time; + int ret; arb_regs->mc_arb_rfsh_rate = (u8)si_calculate_memory_refresh_rate(adev, pl->sclk); - amdgpu_atombios_set_engine_dram_timings(adev, - pl->sclk, - pl->mclk); + ret = amdgpu_atombios_set_engine_dram_timings(adev, pl->sclk, + pl->mclk); + if (ret) + return ret; dram_timing = RREG32(MC_ARB_DRAM_TIMING); dram_timing2 = RREG32(MC_ARB_DRAM_TIMING2); @@ -7619,10 +7621,10 @@ static int si_dpm_process_interrupt(struct amdgpu_device *adev, return 0; } -static int si_dpm_late_init(void *handle) +static int si_dpm_late_init(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (!adev->pm.dpm_enabled) return 0; @@ -7716,10 +7718,10 @@ static int si_dpm_init_microcode(struct amdgpu_device *adev) return err; } -static int si_dpm_sw_init(void *handle) +static int si_dpm_sw_init(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; ret = amdgpu_irq_add_id(adev, AMDGPU_IRQ_CLIENTID_LEGACY, 230, &adev->pm.dpm.thermal.irq); if (ret) @@ -7763,9 +7765,9 @@ dpm_failed: return ret; } -static int si_dpm_sw_fini(void *handle) +static int si_dpm_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; flush_work(&adev->pm.dpm.thermal.work); @@ -7774,11 +7776,11 @@ static int si_dpm_sw_fini(void *handle) return 0; } -static int si_dpm_hw_init(void *handle) +static int si_dpm_hw_init(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (!amdgpu_dpm) return 0; @@ -7793,9 +7795,9 @@ static int si_dpm_hw_init(void *handle) return ret; } -static int si_dpm_hw_fini(void *handle) +static int si_dpm_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->pm.dpm_enabled) si_dpm_disable(adev); @@ -7803,9 +7805,9 @@ static int si_dpm_hw_fini(void *handle) return 0; } -static int si_dpm_suspend(void *handle) +static int si_dpm_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->pm.dpm_enabled) { /* disable dpm */ @@ -7816,10 +7818,10 @@ static int si_dpm_suspend(void *handle) return 0; } -static int si_dpm_resume(void *handle) +static int si_dpm_resume(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->pm.dpm_enabled) { /* asic init will reset to the boot state */ @@ -7841,17 +7843,12 @@ static bool si_dpm_is_idle(void *handle) return true; } -static int si_dpm_wait_for_idle(void *handle) +static int si_dpm_wait_for_idle(struct amdgpu_ip_block *ip_block) { /* XXX */ return 0; } -static int si_dpm_soft_reset(void *handle) -{ - return 0; -} - static int si_dpm_set_clockgating_state(void *handle, enum amd_clockgating_state state) { @@ -7928,10 +7925,10 @@ static void si_dpm_print_power_state(void *handle, amdgpu_dpm_print_ps_status(adev, rps); } -static int si_dpm_early_init(void *handle) +static int si_dpm_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; adev->powerplay.pp_funcs = &si_dpm_funcs; adev->powerplay.pp_handle = adev; @@ -8047,11 +8044,8 @@ static const struct amd_ip_funcs si_dpm_ip_funcs = { .resume = si_dpm_resume, .is_idle = si_dpm_is_idle, .wait_for_idle = si_dpm_wait_for_idle, - .soft_reset = si_dpm_soft_reset, .set_clockgating_state = si_dpm_set_clockgating_state, .set_powergating_state = si_dpm_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; const struct amdgpu_ip_block_version si_smu_ip_block = diff --git a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c index a71c6117d7e5..26624a716fc6 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c +++ b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c @@ -75,11 +75,10 @@ static void amd_powerplay_destroy(struct amdgpu_device *adev) hwmgr = NULL; } -static int pp_early_init(void *handle) +static int pp_early_init(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = handle; - + struct amdgpu_device *adev = ip_block->adev; ret = amd_powerplay_create(adev); if (ret != 0) @@ -131,9 +130,9 @@ static void pp_swctf_delayed_work_handler(struct work_struct *work) orderly_poweroff(true); } -static int pp_sw_init(void *handle) +static int pp_sw_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = handle; + struct amdgpu_device *adev = ip_block->adev; struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; int ret = 0; @@ -148,9 +147,9 @@ static int pp_sw_init(void *handle) return ret; } -static int pp_sw_fini(void *handle) +static int pp_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = handle; + struct amdgpu_device *adev = ip_block->adev; struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; hwmgr_sw_fini(hwmgr); @@ -160,10 +159,10 @@ static int pp_sw_fini(void *handle) return 0; } -static int pp_hw_init(void *handle) +static int pp_hw_init(struct amdgpu_ip_block *ip_block) { int ret = 0; - struct amdgpu_device *adev = handle; + struct amdgpu_device *adev = ip_block->adev; struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; ret = hwmgr_hw_init(hwmgr); @@ -174,10 +173,9 @@ static int pp_hw_init(void *handle) return ret; } -static int pp_hw_fini(void *handle) +static int pp_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = handle; - struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; + struct pp_hwmgr *hwmgr = ip_block->adev->powerplay.pp_handle; cancel_delayed_work_sync(&hwmgr->swctf_delayed_work); @@ -217,9 +215,9 @@ static void pp_reserve_vram_for_smu(struct amdgpu_device *adev) } } -static int pp_late_init(void *handle) +static int pp_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = handle; + struct amdgpu_device *adev = ip_block->adev; struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; if (hwmgr && hwmgr->pm_en) @@ -231,9 +229,9 @@ static int pp_late_init(void *handle) return 0; } -static void pp_late_fini(void *handle) +static void pp_late_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = handle; + struct amdgpu_device *adev = ip_block->adev; if (adev->pm.smu_prv_buffer) amdgpu_bo_free_kernel(&adev->pm.smu_prv_buffer, NULL, NULL); @@ -246,25 +244,15 @@ static bool pp_is_idle(void *handle) return false; } -static int pp_wait_for_idle(void *handle) -{ - return 0; -} - -static int pp_sw_reset(void *handle) -{ - return 0; -} - static int pp_set_powergating_state(void *handle, enum amd_powergating_state state) { return 0; } -static int pp_suspend(void *handle) +static int pp_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = handle; + struct amdgpu_device *adev = ip_block->adev; struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; cancel_delayed_work_sync(&hwmgr->swctf_delayed_work); @@ -272,10 +260,9 @@ static int pp_suspend(void *handle) return hwmgr_suspend(hwmgr); } -static int pp_resume(void *handle) +static int pp_resume(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = handle; - struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; + struct pp_hwmgr *hwmgr = ip_block->adev->powerplay.pp_handle; return hwmgr_resume(hwmgr); } @@ -298,12 +285,8 @@ static const struct amd_ip_funcs pp_ip_funcs = { .suspend = pp_suspend, .resume = pp_resume, .is_idle = pp_is_idle, - .wait_for_idle = pp_wait_for_idle, - .soft_reset = pp_sw_reset, .set_clockgating_state = pp_set_clockgating_state, .set_powergating_state = pp_set_powergating_state, - .dump_ip_state = NULL, - .print_ip_state = NULL, }; const struct amdgpu_ip_block_version pp_smu_ip_block = diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c index b56298d9da98..fe24219c3bf4 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.c @@ -28,7 +28,6 @@ #include "ppatomctrl.h" #include "atombios.h" #include "cgs_common.h" -#include "ppevvmath.h" #define MEM_ID_MASK 0xff000000 #define MEM_ID_SHIFT 24 @@ -677,433 +676,6 @@ bool atomctrl_get_pp_assign_pin( return bRet; } -int atomctrl_calculate_voltage_evv_on_sclk( - struct pp_hwmgr *hwmgr, - uint8_t voltage_type, - uint32_t sclk, - uint16_t virtual_voltage_Id, - uint16_t *voltage, - uint16_t dpm_level, - bool debug) -{ - ATOM_ASIC_PROFILING_INFO_V3_4 *getASICProfilingInfo; - struct amdgpu_device *adev = hwmgr->adev; - EFUSE_LINEAR_FUNC_PARAM sRO_fuse; - EFUSE_LINEAR_FUNC_PARAM sCACm_fuse; - EFUSE_LINEAR_FUNC_PARAM sCACb_fuse; - EFUSE_LOGISTIC_FUNC_PARAM sKt_Beta_fuse; - EFUSE_LOGISTIC_FUNC_PARAM sKv_m_fuse; - EFUSE_LOGISTIC_FUNC_PARAM sKv_b_fuse; - EFUSE_INPUT_PARAMETER sInput_FuseValues; - READ_EFUSE_VALUE_PARAMETER sOutput_FuseValues; - - uint32_t ul_RO_fused, ul_CACb_fused, ul_CACm_fused, ul_Kt_Beta_fused, ul_Kv_m_fused, ul_Kv_b_fused; - fInt fSM_A0, fSM_A1, fSM_A2, fSM_A3, fSM_A4, fSM_A5, fSM_A6, fSM_A7; - fInt fMargin_RO_a, fMargin_RO_b, fMargin_RO_c, fMargin_fixed, fMargin_FMAX_mean, fMargin_Plat_mean, fMargin_FMAX_sigma, fMargin_Plat_sigma, fMargin_DC_sigma; - fInt fLkg_FT, repeat; - fInt fMicro_FMAX, fMicro_CR, fSigma_FMAX, fSigma_CR, fSigma_DC, fDC_SCLK, fSquared_Sigma_DC, fSquared_Sigma_CR, fSquared_Sigma_FMAX; - fInt fRLL_LoadLine, fDerateTDP, fVDDC_base, fA_Term, fC_Term, fB_Term, fRO_DC_margin; - fInt fRO_fused, fCACm_fused, fCACb_fused, fKv_m_fused, fKv_b_fused, fKt_Beta_fused, fFT_Lkg_V0NORM; - fInt fSclk_margin, fSclk, fEVV_V; - fInt fV_min, fV_max, fT_prod, fLKG_Factor, fT_FT, fV_FT, fV_x, fTDP_Power, fTDP_Power_right, fTDP_Power_left, fTDP_Current, fV_NL; - uint32_t ul_FT_Lkg_V0NORM; - fInt fLn_MaxDivMin, fMin, fAverage, fRange; - fInt fRoots[2]; - fInt fStepSize = GetScaledFraction(625, 100000); - - int result; - - getASICProfilingInfo = (ATOM_ASIC_PROFILING_INFO_V3_4 *) - smu_atom_get_data_table(hwmgr->adev, - GetIndexIntoMasterTable(DATA, ASIC_ProfilingInfo), - NULL, NULL, NULL); - - if (!getASICProfilingInfo) - return -1; - - if (getASICProfilingInfo->asHeader.ucTableFormatRevision < 3 || - (getASICProfilingInfo->asHeader.ucTableFormatRevision == 3 && - getASICProfilingInfo->asHeader.ucTableContentRevision < 4)) - return -1; - - /*----------------------------------------------------------- - *GETTING MULTI-STEP PARAMETERS RELATED TO CURRENT DPM LEVEL - *----------------------------------------------------------- - */ - fRLL_LoadLine = Divide(getASICProfilingInfo->ulLoadLineSlop, 1000); - - switch (dpm_level) { - case 1: - fDerateTDP = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulTdpDerateDPM1), 1000); - break; - case 2: - fDerateTDP = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulTdpDerateDPM2), 1000); - break; - case 3: - fDerateTDP = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulTdpDerateDPM3), 1000); - break; - case 4: - fDerateTDP = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulTdpDerateDPM4), 1000); - break; - case 5: - fDerateTDP = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulTdpDerateDPM5), 1000); - break; - case 6: - fDerateTDP = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulTdpDerateDPM6), 1000); - break; - case 7: - fDerateTDP = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulTdpDerateDPM7), 1000); - break; - default: - pr_err("DPM Level not supported\n"); - fDerateTDP = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulTdpDerateDPM0), 1000); - } - - /*------------------------- - * DECODING FUSE VALUES - * ------------------------ - */ - /*Decode RO_Fused*/ - sRO_fuse = getASICProfilingInfo->sRoFuse; - - sInput_FuseValues.usEfuseIndex = sRO_fuse.usEfuseIndex; - sInput_FuseValues.ucBitShift = sRO_fuse.ucEfuseBitLSB; - sInput_FuseValues.ucBitLength = sRO_fuse.ucEfuseLength; - - sOutput_FuseValues.sEfuse = sInput_FuseValues; - - result = amdgpu_atom_execute_table(adev->mode_info.atom_context, - GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); - - if (result) - return result; - - /* Finally, the actual fuse value */ - ul_RO_fused = le32_to_cpu(sOutput_FuseValues.ulEfuseValue); - fMin = GetScaledFraction(le32_to_cpu(sRO_fuse.ulEfuseMin), 1); - fRange = GetScaledFraction(le32_to_cpu(sRO_fuse.ulEfuseEncodeRange), 1); - fRO_fused = fDecodeLinearFuse(ul_RO_fused, fMin, fRange, sRO_fuse.ucEfuseLength); - - sCACm_fuse = getASICProfilingInfo->sCACm; - - sInput_FuseValues.usEfuseIndex = sCACm_fuse.usEfuseIndex; - sInput_FuseValues.ucBitShift = sCACm_fuse.ucEfuseBitLSB; - sInput_FuseValues.ucBitLength = sCACm_fuse.ucEfuseLength; - - sOutput_FuseValues.sEfuse = sInput_FuseValues; - - result = amdgpu_atom_execute_table(adev->mode_info.atom_context, - GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); - - if (result) - return result; - - ul_CACm_fused = le32_to_cpu(sOutput_FuseValues.ulEfuseValue); - fMin = GetScaledFraction(le32_to_cpu(sCACm_fuse.ulEfuseMin), 1000); - fRange = GetScaledFraction(le32_to_cpu(sCACm_fuse.ulEfuseEncodeRange), 1000); - - fCACm_fused = fDecodeLinearFuse(ul_CACm_fused, fMin, fRange, sCACm_fuse.ucEfuseLength); - - sCACb_fuse = getASICProfilingInfo->sCACb; - - sInput_FuseValues.usEfuseIndex = sCACb_fuse.usEfuseIndex; - sInput_FuseValues.ucBitShift = sCACb_fuse.ucEfuseBitLSB; - sInput_FuseValues.ucBitLength = sCACb_fuse.ucEfuseLength; - sOutput_FuseValues.sEfuse = sInput_FuseValues; - - result = amdgpu_atom_execute_table(adev->mode_info.atom_context, - GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); - - if (result) - return result; - - ul_CACb_fused = le32_to_cpu(sOutput_FuseValues.ulEfuseValue); - fMin = GetScaledFraction(le32_to_cpu(sCACb_fuse.ulEfuseMin), 1000); - fRange = GetScaledFraction(le32_to_cpu(sCACb_fuse.ulEfuseEncodeRange), 1000); - - fCACb_fused = fDecodeLinearFuse(ul_CACb_fused, fMin, fRange, sCACb_fuse.ucEfuseLength); - - sKt_Beta_fuse = getASICProfilingInfo->sKt_b; - - sInput_FuseValues.usEfuseIndex = sKt_Beta_fuse.usEfuseIndex; - sInput_FuseValues.ucBitShift = sKt_Beta_fuse.ucEfuseBitLSB; - sInput_FuseValues.ucBitLength = sKt_Beta_fuse.ucEfuseLength; - - sOutput_FuseValues.sEfuse = sInput_FuseValues; - - result = amdgpu_atom_execute_table(adev->mode_info.atom_context, - GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); - - if (result) - return result; - - ul_Kt_Beta_fused = le32_to_cpu(sOutput_FuseValues.ulEfuseValue); - fAverage = GetScaledFraction(le32_to_cpu(sKt_Beta_fuse.ulEfuseEncodeAverage), 1000); - fRange = GetScaledFraction(le32_to_cpu(sKt_Beta_fuse.ulEfuseEncodeRange), 1000); - - fKt_Beta_fused = fDecodeLogisticFuse(ul_Kt_Beta_fused, - fAverage, fRange, sKt_Beta_fuse.ucEfuseLength); - - sKv_m_fuse = getASICProfilingInfo->sKv_m; - - sInput_FuseValues.usEfuseIndex = sKv_m_fuse.usEfuseIndex; - sInput_FuseValues.ucBitShift = sKv_m_fuse.ucEfuseBitLSB; - sInput_FuseValues.ucBitLength = sKv_m_fuse.ucEfuseLength; - - sOutput_FuseValues.sEfuse = sInput_FuseValues; - - result = amdgpu_atom_execute_table(adev->mode_info.atom_context, - GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); - if (result) - return result; - - ul_Kv_m_fused = le32_to_cpu(sOutput_FuseValues.ulEfuseValue); - fAverage = GetScaledFraction(le32_to_cpu(sKv_m_fuse.ulEfuseEncodeAverage), 1000); - fRange = GetScaledFraction((le32_to_cpu(sKv_m_fuse.ulEfuseEncodeRange) & 0x7fffffff), 1000); - fRange = fMultiply(fRange, ConvertToFraction(-1)); - - fKv_m_fused = fDecodeLogisticFuse(ul_Kv_m_fused, - fAverage, fRange, sKv_m_fuse.ucEfuseLength); - - sKv_b_fuse = getASICProfilingInfo->sKv_b; - - sInput_FuseValues.usEfuseIndex = sKv_b_fuse.usEfuseIndex; - sInput_FuseValues.ucBitShift = sKv_b_fuse.ucEfuseBitLSB; - sInput_FuseValues.ucBitLength = sKv_b_fuse.ucEfuseLength; - sOutput_FuseValues.sEfuse = sInput_FuseValues; - - result = amdgpu_atom_execute_table(adev->mode_info.atom_context, - GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); - - if (result) - return result; - - ul_Kv_b_fused = le32_to_cpu(sOutput_FuseValues.ulEfuseValue); - fAverage = GetScaledFraction(le32_to_cpu(sKv_b_fuse.ulEfuseEncodeAverage), 1000); - fRange = GetScaledFraction(le32_to_cpu(sKv_b_fuse.ulEfuseEncodeRange), 1000); - - fKv_b_fused = fDecodeLogisticFuse(ul_Kv_b_fused, - fAverage, fRange, sKv_b_fuse.ucEfuseLength); - - /* Decoding the Leakage - No special struct container */ - /* - * usLkgEuseIndex=56 - * ucLkgEfuseBitLSB=6 - * ucLkgEfuseLength=10 - * ulLkgEncodeLn_MaxDivMin=69077 - * ulLkgEncodeMax=1000000 - * ulLkgEncodeMin=1000 - * ulEfuseLogisticAlpha=13 - */ - - sInput_FuseValues.usEfuseIndex = getASICProfilingInfo->usLkgEuseIndex; - sInput_FuseValues.ucBitShift = getASICProfilingInfo->ucLkgEfuseBitLSB; - sInput_FuseValues.ucBitLength = getASICProfilingInfo->ucLkgEfuseLength; - - sOutput_FuseValues.sEfuse = sInput_FuseValues; - - result = amdgpu_atom_execute_table(adev->mode_info.atom_context, - GetIndexIntoMasterTable(COMMAND, ReadEfuseValue), - (uint32_t *)&sOutput_FuseValues, sizeof(sOutput_FuseValues)); - - if (result) - return result; - - ul_FT_Lkg_V0NORM = le32_to_cpu(sOutput_FuseValues.ulEfuseValue); - fLn_MaxDivMin = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulLkgEncodeLn_MaxDivMin), 10000); - fMin = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulLkgEncodeMin), 10000); - - fFT_Lkg_V0NORM = fDecodeLeakageID(ul_FT_Lkg_V0NORM, - fLn_MaxDivMin, fMin, getASICProfilingInfo->ucLkgEfuseLength); - fLkg_FT = fFT_Lkg_V0NORM; - - /*------------------------------------------- - * PART 2 - Grabbing all required values - *------------------------------------------- - */ - fSM_A0 = fMultiply(GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulSM_A0), 1000000), - ConvertToFraction(uPow(-1, getASICProfilingInfo->ucSM_A0_sign))); - fSM_A1 = fMultiply(GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulSM_A1), 1000000), - ConvertToFraction(uPow(-1, getASICProfilingInfo->ucSM_A1_sign))); - fSM_A2 = fMultiply(GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulSM_A2), 100000), - ConvertToFraction(uPow(-1, getASICProfilingInfo->ucSM_A2_sign))); - fSM_A3 = fMultiply(GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulSM_A3), 1000000), - ConvertToFraction(uPow(-1, getASICProfilingInfo->ucSM_A3_sign))); - fSM_A4 = fMultiply(GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulSM_A4), 1000000), - ConvertToFraction(uPow(-1, getASICProfilingInfo->ucSM_A4_sign))); - fSM_A5 = fMultiply(GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulSM_A5), 1000), - ConvertToFraction(uPow(-1, getASICProfilingInfo->ucSM_A5_sign))); - fSM_A6 = fMultiply(GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulSM_A6), 1000), - ConvertToFraction(uPow(-1, getASICProfilingInfo->ucSM_A6_sign))); - fSM_A7 = fMultiply(GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulSM_A7), 1000), - ConvertToFraction(uPow(-1, getASICProfilingInfo->ucSM_A7_sign))); - - fMargin_RO_a = ConvertToFraction(le32_to_cpu(getASICProfilingInfo->ulMargin_RO_a)); - fMargin_RO_b = ConvertToFraction(le32_to_cpu(getASICProfilingInfo->ulMargin_RO_b)); - fMargin_RO_c = ConvertToFraction(le32_to_cpu(getASICProfilingInfo->ulMargin_RO_c)); - - fMargin_fixed = ConvertToFraction(le32_to_cpu(getASICProfilingInfo->ulMargin_fixed)); - - fMargin_FMAX_mean = GetScaledFraction( - le32_to_cpu(getASICProfilingInfo->ulMargin_Fmax_mean), 10000); - fMargin_Plat_mean = GetScaledFraction( - le32_to_cpu(getASICProfilingInfo->ulMargin_plat_mean), 10000); - fMargin_FMAX_sigma = GetScaledFraction( - le32_to_cpu(getASICProfilingInfo->ulMargin_Fmax_sigma), 10000); - fMargin_Plat_sigma = GetScaledFraction( - le32_to_cpu(getASICProfilingInfo->ulMargin_plat_sigma), 10000); - - fMargin_DC_sigma = GetScaledFraction( - le32_to_cpu(getASICProfilingInfo->ulMargin_DC_sigma), 100); - fMargin_DC_sigma = fDivide(fMargin_DC_sigma, ConvertToFraction(1000)); - - fCACm_fused = fDivide(fCACm_fused, ConvertToFraction(100)); - fCACb_fused = fDivide(fCACb_fused, ConvertToFraction(100)); - fKt_Beta_fused = fDivide(fKt_Beta_fused, ConvertToFraction(100)); - fKv_m_fused = fNegate(fDivide(fKv_m_fused, ConvertToFraction(100))); - fKv_b_fused = fDivide(fKv_b_fused, ConvertToFraction(10)); - - fSclk = GetScaledFraction(sclk, 100); - - fV_max = fDivide(GetScaledFraction( - le32_to_cpu(getASICProfilingInfo->ulMaxVddc), 1000), ConvertToFraction(4)); - fT_prod = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulBoardCoreTemp), 10); - fLKG_Factor = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulEvvLkgFactor), 100); - fT_FT = GetScaledFraction(le32_to_cpu(getASICProfilingInfo->ulLeakageTemp), 10); - fV_FT = fDivide(GetScaledFraction( - le32_to_cpu(getASICProfilingInfo->ulLeakageVoltage), 1000), ConvertToFraction(4)); - fV_min = fDivide(GetScaledFraction( - le32_to_cpu(getASICProfilingInfo->ulMinVddc), 1000), ConvertToFraction(4)); - - /*----------------------- - * PART 3 - *----------------------- - */ - - fA_Term = fAdd(fMargin_RO_a, fAdd(fMultiply(fSM_A4, fSclk), fSM_A5)); - fB_Term = fAdd(fAdd(fMultiply(fSM_A2, fSclk), fSM_A6), fMargin_RO_b); - fC_Term = fAdd(fMargin_RO_c, - fAdd(fMultiply(fSM_A0, fLkg_FT), - fAdd(fMultiply(fSM_A1, fMultiply(fLkg_FT, fSclk)), - fAdd(fMultiply(fSM_A3, fSclk), - fSubtract(fSM_A7, fRO_fused))))); - - fVDDC_base = fSubtract(fRO_fused, - fSubtract(fMargin_RO_c, - fSubtract(fSM_A3, fMultiply(fSM_A1, fSclk)))); - fVDDC_base = fDivide(fVDDC_base, fAdd(fMultiply(fSM_A0, fSclk), fSM_A2)); - - repeat = fSubtract(fVDDC_base, - fDivide(fMargin_DC_sigma, ConvertToFraction(1000))); - - fRO_DC_margin = fAdd(fMultiply(fMargin_RO_a, - fGetSquare(repeat)), - fAdd(fMultiply(fMargin_RO_b, repeat), - fMargin_RO_c)); - - fDC_SCLK = fSubtract(fRO_fused, - fSubtract(fRO_DC_margin, - fSubtract(fSM_A3, - fMultiply(fSM_A2, repeat)))); - fDC_SCLK = fDivide(fDC_SCLK, fAdd(fMultiply(fSM_A0, repeat), fSM_A1)); - - fSigma_DC = fSubtract(fSclk, fDC_SCLK); - - fMicro_FMAX = fMultiply(fSclk, fMargin_FMAX_mean); - fMicro_CR = fMultiply(fSclk, fMargin_Plat_mean); - fSigma_FMAX = fMultiply(fSclk, fMargin_FMAX_sigma); - fSigma_CR = fMultiply(fSclk, fMargin_Plat_sigma); - - fSquared_Sigma_DC = fGetSquare(fSigma_DC); - fSquared_Sigma_CR = fGetSquare(fSigma_CR); - fSquared_Sigma_FMAX = fGetSquare(fSigma_FMAX); - - fSclk_margin = fAdd(fMicro_FMAX, - fAdd(fMicro_CR, - fAdd(fMargin_fixed, - fSqrt(fAdd(fSquared_Sigma_FMAX, - fAdd(fSquared_Sigma_DC, fSquared_Sigma_CR)))))); - /* - fA_Term = fSM_A4 * (fSclk + fSclk_margin) + fSM_A5; - fB_Term = fSM_A2 * (fSclk + fSclk_margin) + fSM_A6; - fC_Term = fRO_DC_margin + fSM_A0 * fLkg_FT + fSM_A1 * fLkg_FT * (fSclk + fSclk_margin) + fSM_A3 * (fSclk + fSclk_margin) + fSM_A7 - fRO_fused; - */ - - fA_Term = fAdd(fMultiply(fSM_A4, fAdd(fSclk, fSclk_margin)), fSM_A5); - fB_Term = fAdd(fMultiply(fSM_A2, fAdd(fSclk, fSclk_margin)), fSM_A6); - fC_Term = fAdd(fRO_DC_margin, - fAdd(fMultiply(fSM_A0, fLkg_FT), - fAdd(fMultiply(fMultiply(fSM_A1, fLkg_FT), - fAdd(fSclk, fSclk_margin)), - fAdd(fMultiply(fSM_A3, - fAdd(fSclk, fSclk_margin)), - fSubtract(fSM_A7, fRO_fused))))); - - SolveQuadracticEqn(fA_Term, fB_Term, fC_Term, fRoots); - - if (GreaterThan(fRoots[0], fRoots[1])) - fEVV_V = fRoots[1]; - else - fEVV_V = fRoots[0]; - - if (GreaterThan(fV_min, fEVV_V)) - fEVV_V = fV_min; - else if (GreaterThan(fEVV_V, fV_max)) - fEVV_V = fSubtract(fV_max, fStepSize); - - fEVV_V = fRoundUpByStepSize(fEVV_V, fStepSize, 0); - - /*----------------- - * PART 4 - *----------------- - */ - - fV_x = fV_min; - - while (GreaterThan(fAdd(fV_max, fStepSize), fV_x)) { - fTDP_Power_left = fMultiply(fMultiply(fMultiply(fAdd( - fMultiply(fCACm_fused, fV_x), fCACb_fused), fSclk), - fGetSquare(fV_x)), fDerateTDP); - - fTDP_Power_right = fMultiply(fFT_Lkg_V0NORM, fMultiply(fLKG_Factor, - fMultiply(fExponential(fMultiply(fAdd(fMultiply(fKv_m_fused, - fT_prod), fKv_b_fused), fV_x)), fV_x))); - fTDP_Power_right = fMultiply(fTDP_Power_right, fExponential(fMultiply( - fKt_Beta_fused, fT_prod))); - fTDP_Power_right = fDivide(fTDP_Power_right, fExponential(fMultiply( - fAdd(fMultiply(fKv_m_fused, fT_prod), fKv_b_fused), fV_FT))); - fTDP_Power_right = fDivide(fTDP_Power_right, fExponential(fMultiply( - fKt_Beta_fused, fT_FT))); - - fTDP_Power = fAdd(fTDP_Power_left, fTDP_Power_right); - - fTDP_Current = fDivide(fTDP_Power, fV_x); - - fV_NL = fAdd(fV_x, fDivide(fMultiply(fTDP_Current, fRLL_LoadLine), - ConvertToFraction(10))); - - fV_NL = fRoundUpByStepSize(fV_NL, fStepSize, 0); - - if (GreaterThan(fV_max, fV_NL) && - (GreaterThan(fV_NL, fEVV_V) || - Equal(fV_NL, fEVV_V))) { - fV_NL = fMultiply(fV_NL, ConvertToFraction(1000)); - - *voltage = (uint16_t)fV_NL.partial.real; - break; - } else - fV_x = fAdd(fV_x, fStepSize); - } - - return result; -} - /** * atomctrl_get_voltage_evv_on_sclk: gets voltage via call to ATOM COMMAND table. * @hwmgr: input: pointer to hwManager diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.h index 1f987e846628..22b0ac12df97 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.h +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppatomctrl.h @@ -316,8 +316,6 @@ extern int atomctrl_get_engine_pll_dividers_kong(struct pp_hwmgr *hwmgr, pp_atomctrl_clock_dividers_kong *dividers); extern int atomctrl_read_efuse(struct pp_hwmgr *hwmgr, uint16_t start_index, uint16_t end_index, uint32_t *efuse); -extern int atomctrl_calculate_voltage_evv_on_sclk(struct pp_hwmgr *hwmgr, uint8_t voltage_type, - uint32_t sclk, uint16_t virtual_voltage_Id, uint16_t *voltage, uint16_t dpm_level, bool debug); extern int atomctrl_get_engine_pll_dividers_ai(struct pp_hwmgr *hwmgr, uint32_t clock_value, pp_atomctrl_clock_dividers_ai *dividers); extern int atomctrl_set_ac_timing_ai(struct pp_hwmgr *hwmgr, uint32_t memory_clock, uint8_t level); diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppevvmath.h b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppevvmath.h deleted file mode 100644 index 409aeec6baa9..000000000000 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/ppevvmath.h +++ /dev/null @@ -1,561 +0,0 @@ -/* - * Copyright 2015 Advanced Micro Devices, Inc. - * - * Permission is hereby granted, free of charge, to any person obtaining a - * copy of this software and associated documentation files (the "Software"), - * to deal in the Software without restriction, including without limitation - * the rights to use, copy, modify, merge, publish, distribute, sublicense, - * and/or sell copies of the Software, and to permit persons to whom the - * Software is furnished to do so, subject to the following conditions: - * - * The above copyright notice and this permission notice shall be included in - * all copies or substantial portions of the Software. - * - * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR - * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, - * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL - * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR - * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, - * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR - * OTHER DEALINGS IN THE SOFTWARE. - * - */ -#include - -enum ppevvmath_constants { - /* We multiply all original integers with 2^SHIFT_AMOUNT to get the fInt representation */ - SHIFT_AMOUNT = 16, - - /* Change this value to change the number of decimal places in the final output - 5 is a good default */ - PRECISION = 5, - - SHIFTED_2 = (2 << SHIFT_AMOUNT), - - /* 32767 - Might change in the future */ - MAX = (1 << (SHIFT_AMOUNT - 1)) - 1, -}; - -/* ------------------------------------------------------------------------------- - * NEW TYPE - fINT - * ------------------------------------------------------------------------------- - * A variable of type fInt can be accessed in 3 ways using the dot (.) operator - * fInt A; - * A.full => The full number as it is. Generally not easy to read - * A.partial.real => Only the integer portion - * A.partial.decimal => Only the fractional portion - */ -typedef union _fInt { - int full; - struct _partial { - unsigned int decimal: SHIFT_AMOUNT; /*Needs to always be unsigned*/ - int real: 32 - SHIFT_AMOUNT; - } partial; -} fInt; - -/* ------------------------------------------------------------------------------- - * Function Declarations - * ------------------------------------------------------------------------------- - */ -static fInt ConvertToFraction(int); /* Use this to convert an INT to a FINT */ -static fInt Convert_ULONG_ToFraction(uint32_t); /* Use this to convert an uint32_t to a FINT */ -static fInt GetScaledFraction(int, int); /* Use this to convert an INT to a FINT after scaling it by a factor */ -static int ConvertBackToInteger(fInt); /* Convert a FINT back to an INT that is scaled by 1000 (i.e. last 3 digits are the decimal digits) */ - -static fInt fNegate(fInt); /* Returns -1 * input fInt value */ -static fInt fAdd (fInt, fInt); /* Returns the sum of two fInt numbers */ -static fInt fSubtract (fInt A, fInt B); /* Returns A-B - Sometimes easier than Adding negative numbers */ -static fInt fMultiply (fInt, fInt); /* Returns the product of two fInt numbers */ -static fInt fDivide (fInt A, fInt B); /* Returns A/B */ -static fInt fGetSquare(fInt); /* Returns the square of a fInt number */ -static fInt fSqrt(fInt); /* Returns the Square Root of a fInt number */ - -static int uAbs(int); /* Returns the Absolute value of the Int */ -static int uPow(int base, int exponent); /* Returns base^exponent an INT */ - -static void SolveQuadracticEqn(fInt, fInt, fInt, fInt[]); /* Returns the 2 roots via the array */ -static bool Equal(fInt, fInt); /* Returns true if two fInts are equal to each other */ -static bool GreaterThan(fInt A, fInt B); /* Returns true if A > B */ - -static fInt fExponential(fInt exponent); /* Can be used to calculate e^exponent */ -static fInt fNaturalLog(fInt value); /* Can be used to calculate ln(value) */ - -/* Fuse decoding functions - * ------------------------------------------------------------------------------------- - */ -static fInt fDecodeLinearFuse(uint32_t fuse_value, fInt f_min, fInt f_range, uint32_t bitlength); -static fInt fDecodeLogisticFuse(uint32_t fuse_value, fInt f_average, fInt f_range, uint32_t bitlength); -static fInt fDecodeLeakageID (uint32_t leakageID_fuse, fInt ln_max_div_min, fInt f_min, uint32_t bitlength); - -/* Internal Support Functions - Use these ONLY for testing or adding to internal functions - * ------------------------------------------------------------------------------------- - * Some of the following functions take two INTs as their input - This is unsafe for a variety of reasons. - */ -static fInt Divide (int, int); /* Divide two INTs and return result as FINT */ -static fInt fNegate(fInt); - -static int uGetScaledDecimal (fInt); /* Internal function */ -static int GetReal (fInt A); /* Internal function */ - -/* ------------------------------------------------------------------------------------- - * TROUBLESHOOTING INFORMATION - * ------------------------------------------------------------------------------------- - * 1) ConvertToFraction - InputOutOfRangeException: Only accepts numbers smaller than MAX (default: 32767) - * 2) fAdd - OutputOutOfRangeException: Output bigger than MAX (default: 32767) - * 3) fMultiply - OutputOutOfRangeException: - * 4) fGetSquare - OutputOutOfRangeException: - * 5) fDivide - DivideByZeroException - * 6) fSqrt - NegativeSquareRootException: Input cannot be a negative number - */ - -/* ------------------------------------------------------------------------------------- - * START OF CODE - * ------------------------------------------------------------------------------------- - */ -static fInt fExponential(fInt exponent) /*Can be used to calculate e^exponent*/ -{ - uint32_t i; - bool bNegated = false; - - fInt fPositiveOne = ConvertToFraction(1); - fInt fZERO = ConvertToFraction(0); - - fInt lower_bound = Divide(78, 10000); - fInt solution = fPositiveOne; /*Starting off with baseline of 1 */ - fInt error_term; - - static const uint32_t k_array[11] = {55452, 27726, 13863, 6931, 4055, 2231, 1178, 606, 308, 155, 78}; - static const uint32_t expk_array[11] = {2560000, 160000, 40000, 20000, 15000, 12500, 11250, 10625, 10313, 10156, 10078}; - - if (GreaterThan(fZERO, exponent)) { - exponent = fNegate(exponent); - bNegated = true; - } - - while (GreaterThan(exponent, lower_bound)) { - for (i = 0; i < 11; i++) { - if (GreaterThan(exponent, GetScaledFraction(k_array[i], 10000))) { - exponent = fSubtract(exponent, GetScaledFraction(k_array[i], 10000)); - solution = fMultiply(solution, GetScaledFraction(expk_array[i], 10000)); - } - } - } - - error_term = fAdd(fPositiveOne, exponent); - - solution = fMultiply(solution, error_term); - - if (bNegated) - solution = fDivide(fPositiveOne, solution); - - return solution; -} - -static fInt fNaturalLog(fInt value) -{ - uint32_t i; - fInt upper_bound = Divide(8, 1000); - fInt fNegativeOne = ConvertToFraction(-1); - fInt solution = ConvertToFraction(0); /*Starting off with baseline of 0 */ - fInt error_term; - - static const uint32_t k_array[10] = {160000, 40000, 20000, 15000, 12500, 11250, 10625, 10313, 10156, 10078}; - static const uint32_t logk_array[10] = {27726, 13863, 6931, 4055, 2231, 1178, 606, 308, 155, 78}; - - while (GreaterThan(fAdd(value, fNegativeOne), upper_bound)) { - for (i = 0; i < 10; i++) { - if (GreaterThan(value, GetScaledFraction(k_array[i], 10000))) { - value = fDivide(value, GetScaledFraction(k_array[i], 10000)); - solution = fAdd(solution, GetScaledFraction(logk_array[i], 10000)); - } - } - } - - error_term = fAdd(fNegativeOne, value); - - return fAdd(solution, error_term); -} - -static fInt fDecodeLinearFuse(uint32_t fuse_value, fInt f_min, fInt f_range, uint32_t bitlength) -{ - fInt f_fuse_value = Convert_ULONG_ToFraction(fuse_value); - fInt f_bit_max_value = Convert_ULONG_ToFraction((uPow(2, bitlength)) - 1); - - fInt f_decoded_value; - - f_decoded_value = fDivide(f_fuse_value, f_bit_max_value); - f_decoded_value = fMultiply(f_decoded_value, f_range); - f_decoded_value = fAdd(f_decoded_value, f_min); - - return f_decoded_value; -} - - -static fInt fDecodeLogisticFuse(uint32_t fuse_value, fInt f_average, fInt f_range, uint32_t bitlength) -{ - fInt f_fuse_value = Convert_ULONG_ToFraction(fuse_value); - fInt f_bit_max_value = Convert_ULONG_ToFraction((uPow(2, bitlength)) - 1); - - fInt f_CONSTANT_NEG13 = ConvertToFraction(-13); - fInt f_CONSTANT1 = ConvertToFraction(1); - - fInt f_decoded_value; - - f_decoded_value = fSubtract(fDivide(f_bit_max_value, f_fuse_value), f_CONSTANT1); - f_decoded_value = fNaturalLog(f_decoded_value); - f_decoded_value = fMultiply(f_decoded_value, fDivide(f_range, f_CONSTANT_NEG13)); - f_decoded_value = fAdd(f_decoded_value, f_average); - - return f_decoded_value; -} - -static fInt fDecodeLeakageID (uint32_t leakageID_fuse, fInt ln_max_div_min, fInt f_min, uint32_t bitlength) -{ - fInt fLeakage; - fInt f_bit_max_value = Convert_ULONG_ToFraction((uPow(2, bitlength)) - 1); - - fLeakage = fMultiply(ln_max_div_min, Convert_ULONG_ToFraction(leakageID_fuse)); - fLeakage = fDivide(fLeakage, f_bit_max_value); - fLeakage = fExponential(fLeakage); - fLeakage = fMultiply(fLeakage, f_min); - - return fLeakage; -} - -static fInt ConvertToFraction(int X) /*Add all range checking here. Is it possible to make fInt a private declaration? */ -{ - fInt temp; - - if (X <= MAX) - temp.full = (X << SHIFT_AMOUNT); - else - temp.full = 0; - - return temp; -} - -static fInt fNegate(fInt X) -{ - fInt CONSTANT_NEGONE = ConvertToFraction(-1); - return fMultiply(X, CONSTANT_NEGONE); -} - -static fInt Convert_ULONG_ToFraction(uint32_t X) -{ - fInt temp; - - if (X <= MAX) - temp.full = (X << SHIFT_AMOUNT); - else - temp.full = 0; - - return temp; -} - -static fInt GetScaledFraction(int X, int factor) -{ - int times_shifted, factor_shifted; - bool bNEGATED; - fInt fValue; - - times_shifted = 0; - factor_shifted = 0; - bNEGATED = false; - - if (X < 0) { - X = -1*X; - bNEGATED = true; - } - - if (factor < 0) { - factor = -1*factor; - bNEGATED = !bNEGATED; /*If bNEGATED = true due to X < 0, this will cover the case of negative cancelling negative */ - } - - if ((X > MAX) || factor > MAX) { - if ((X/factor) <= MAX) { - while (X > MAX) { - X = X >> 1; - times_shifted++; - } - - while (factor > MAX) { - factor = factor >> 1; - factor_shifted++; - } - } else { - fValue.full = 0; - return fValue; - } - } - - if (factor == 1) - return ConvertToFraction(X); - - fValue = fDivide(ConvertToFraction(X * uPow(-1, bNEGATED)), ConvertToFraction(factor)); - - fValue.full = fValue.full << times_shifted; - fValue.full = fValue.full >> factor_shifted; - - return fValue; -} - -/* Addition using two fInts */ -static fInt fAdd (fInt X, fInt Y) -{ - fInt Sum; - - Sum.full = X.full + Y.full; - - return Sum; -} - -/* Addition using two fInts */ -static fInt fSubtract (fInt X, fInt Y) -{ - fInt Difference; - - Difference.full = X.full - Y.full; - - return Difference; -} - -static bool Equal(fInt A, fInt B) -{ - if (A.full == B.full) - return true; - else - return false; -} - -static bool GreaterThan(fInt A, fInt B) -{ - if (A.full > B.full) - return true; - else - return false; -} - -static fInt fMultiply (fInt X, fInt Y) /* Uses 64-bit integers (int64_t) */ -{ - fInt Product; - int64_t tempProduct; - - /*The following is for a very specific common case: Non-zero number with ONLY fractional portion*/ - /* TEMPORARILY DISABLED - CAN BE USED TO IMPROVE PRECISION - bool X_LessThanOne, Y_LessThanOne; - - X_LessThanOne = (X.partial.real == 0 && X.partial.decimal != 0 && X.full >= 0); - Y_LessThanOne = (Y.partial.real == 0 && Y.partial.decimal != 0 && Y.full >= 0); - - if (X_LessThanOne && Y_LessThanOne) { - Product.full = X.full * Y.full; - return Product - }*/ - - tempProduct = ((int64_t)X.full) * ((int64_t)Y.full); /*Q(16,16)*Q(16,16) = Q(32, 32) - Might become a negative number! */ - tempProduct = tempProduct >> 16; /*Remove lagging 16 bits - Will lose some precision from decimal; */ - Product.full = (int)tempProduct; /*The int64_t will lose the leading 16 bits that were part of the integer portion */ - - return Product; -} - -static fInt fDivide (fInt X, fInt Y) -{ - fInt fZERO, fQuotient; - int64_t longlongX, longlongY; - - fZERO = ConvertToFraction(0); - - if (Equal(Y, fZERO)) - return fZERO; - - longlongX = (int64_t)X.full; - longlongY = (int64_t)Y.full; - - longlongX = longlongX << 16; /*Q(16,16) -> Q(32,32) */ - - div64_s64(longlongX, longlongY); /*Q(32,32) divided by Q(16,16) = Q(16,16) Back to original format */ - - fQuotient.full = (int)longlongX; - return fQuotient; -} - -static int ConvertBackToInteger (fInt A) /*THIS is the function that will be used to check with the Golden settings table*/ -{ - fInt fullNumber, scaledDecimal, scaledReal; - - scaledReal.full = GetReal(A) * uPow(10, PRECISION-1); /* DOUBLE CHECK THISSSS!!! */ - - scaledDecimal.full = uGetScaledDecimal(A); - - fullNumber = fAdd(scaledDecimal, scaledReal); - - return fullNumber.full; -} - -static fInt fGetSquare(fInt A) -{ - return fMultiply(A, A); -} - -/* x_new = x_old - (x_old^2 - C) / (2 * x_old) */ -static fInt fSqrt(fInt num) -{ - fInt F_divide_Fprime, Fprime; - fInt test; - fInt twoShifted; - int seed, counter, error; - fInt x_new, x_old, C, y; - - fInt fZERO = ConvertToFraction(0); - - /* (0 > num) is the same as (num < 0), i.e., num is negative */ - - if (GreaterThan(fZERO, num) || Equal(fZERO, num)) - return fZERO; - - C = num; - - if (num.partial.real > 3000) - seed = 60; - else if (num.partial.real > 1000) - seed = 30; - else if (num.partial.real > 100) - seed = 10; - else - seed = 2; - - counter = 0; - - if (Equal(num, fZERO)) /*Square Root of Zero is zero */ - return fZERO; - - twoShifted = ConvertToFraction(2); - x_new = ConvertToFraction(seed); - - do { - counter++; - - x_old.full = x_new.full; - - test = fGetSquare(x_old); /*1.75*1.75 is reverting back to 1 when shifted down */ - y = fSubtract(test, C); /*y = f(x) = x^2 - C; */ - - Fprime = fMultiply(twoShifted, x_old); - F_divide_Fprime = fDivide(y, Fprime); - - x_new = fSubtract(x_old, F_divide_Fprime); - - error = ConvertBackToInteger(x_new) - ConvertBackToInteger(x_old); - - if (counter > 20) /*20 is already way too many iterations. If we dont have an answer by then, we never will*/ - return x_new; - - } while (uAbs(error) > 0); - - return x_new; -} - -static void SolveQuadracticEqn(fInt A, fInt B, fInt C, fInt Roots[]) -{ - fInt *pRoots = &Roots[0]; - fInt temp, root_first, root_second; - fInt f_CONSTANT10, f_CONSTANT100; - - f_CONSTANT100 = ConvertToFraction(100); - f_CONSTANT10 = ConvertToFraction(10); - - while (GreaterThan(A, f_CONSTANT100) || GreaterThan(B, f_CONSTANT100) || GreaterThan(C, f_CONSTANT100)) { - A = fDivide(A, f_CONSTANT10); - B = fDivide(B, f_CONSTANT10); - C = fDivide(C, f_CONSTANT10); - } - - temp = fMultiply(ConvertToFraction(4), A); /* root = 4*A */ - temp = fMultiply(temp, C); /* root = 4*A*C */ - temp = fSubtract(fGetSquare(B), temp); /* root = b^2 - 4AC */ - temp = fSqrt(temp); /*root = Sqrt (b^2 - 4AC); */ - - root_first = fSubtract(fNegate(B), temp); /* b - Sqrt(b^2 - 4AC) */ - root_second = fAdd(fNegate(B), temp); /* b + Sqrt(b^2 - 4AC) */ - - root_first = fDivide(root_first, ConvertToFraction(2)); /* [b +- Sqrt(b^2 - 4AC)]/[2] */ - root_first = fDivide(root_first, A); /*[b +- Sqrt(b^2 - 4AC)]/[2*A] */ - - root_second = fDivide(root_second, ConvertToFraction(2)); /* [b +- Sqrt(b^2 - 4AC)]/[2] */ - root_second = fDivide(root_second, A); /*[b +- Sqrt(b^2 - 4AC)]/[2*A] */ - - *(pRoots + 0) = root_first; - *(pRoots + 1) = root_second; -} - -/* ----------------------------------------------------------------------------- - * SUPPORT FUNCTIONS - * ----------------------------------------------------------------------------- - */ - -/* Conversion Functions */ -static int GetReal (fInt A) -{ - return (A.full >> SHIFT_AMOUNT); -} - -static fInt Divide (int X, int Y) -{ - fInt A, B, Quotient; - - A.full = X << SHIFT_AMOUNT; - B.full = Y << SHIFT_AMOUNT; - - Quotient = fDivide(A, B); - - return Quotient; -} - -static int uGetScaledDecimal (fInt A) /*Converts the fractional portion to whole integers - Costly function */ -{ - int dec[PRECISION]; - int i, scaledDecimal = 0, tmp = A.partial.decimal; - - for (i = 0; i < PRECISION; i++) { - dec[i] = tmp / (1 << SHIFT_AMOUNT); - tmp = tmp - ((1 << SHIFT_AMOUNT)*dec[i]); - tmp *= 10; - scaledDecimal = scaledDecimal + dec[i]*uPow(10, PRECISION - 1 - i); - } - - return scaledDecimal; -} - -static int uPow(int base, int power) -{ - if (power == 0) - return 1; - else - return (base)*uPow(base, power - 1); -} - -static int uAbs(int X) -{ - if (X < 0) - return (X * -1); - else - return X; -} - -static fInt fRoundUpByStepSize(fInt A, fInt fStepSize, bool error_term) -{ - fInt solution; - - solution = fDivide(A, fStepSize); - solution.partial.decimal = 0; /*All fractional digits changes to 0 */ - - if (error_term) - solution.partial.real += 1; /*Error term of 1 added */ - - solution = fMultiply(solution, fStepSize); - solution = fAdd(solution, fStepSize); - - return solution; -} - diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_processpptables.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_processpptables.c index 79c817752a33..2b446f8866ba 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_processpptables.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_processpptables.c @@ -62,578 +62,6 @@ static const void *get_powerplay_table(struct pp_hwmgr *hwmgr) return table_address; } -#if 0 -static void dump_pptable(PPTable_t *pptable) -{ - int i; - - pr_info("Version = 0x%08x\n", pptable->Version); - - pr_info("FeaturesToRun[0] = 0x%08x\n", pptable->FeaturesToRun[0]); - pr_info("FeaturesToRun[1] = 0x%08x\n", pptable->FeaturesToRun[1]); - - pr_info("SocketPowerLimitAc0 = %d\n", pptable->SocketPowerLimitAc0); - pr_info("SocketPowerLimitAc0Tau = %d\n", pptable->SocketPowerLimitAc0Tau); - pr_info("SocketPowerLimitAc1 = %d\n", pptable->SocketPowerLimitAc1); - pr_info("SocketPowerLimitAc1Tau = %d\n", pptable->SocketPowerLimitAc1Tau); - pr_info("SocketPowerLimitAc2 = %d\n", pptable->SocketPowerLimitAc2); - pr_info("SocketPowerLimitAc2Tau = %d\n", pptable->SocketPowerLimitAc2Tau); - pr_info("SocketPowerLimitAc3 = %d\n", pptable->SocketPowerLimitAc3); - pr_info("SocketPowerLimitAc3Tau = %d\n", pptable->SocketPowerLimitAc3Tau); - pr_info("SocketPowerLimitDc = %d\n", pptable->SocketPowerLimitDc); - pr_info("SocketPowerLimitDcTau = %d\n", pptable->SocketPowerLimitDcTau); - pr_info("TdcLimitSoc = %d\n", pptable->TdcLimitSoc); - pr_info("TdcLimitSocTau = %d\n", pptable->TdcLimitSocTau); - pr_info("TdcLimitGfx = %d\n", pptable->TdcLimitGfx); - pr_info("TdcLimitGfxTau = %d\n", pptable->TdcLimitGfxTau); - - pr_info("TedgeLimit = %d\n", pptable->TedgeLimit); - pr_info("ThotspotLimit = %d\n", pptable->ThotspotLimit); - pr_info("ThbmLimit = %d\n", pptable->ThbmLimit); - pr_info("Tvr_gfxLimit = %d\n", pptable->Tvr_gfxLimit); - pr_info("Tvr_memLimit = %d\n", pptable->Tvr_memLimit); - pr_info("Tliquid1Limit = %d\n", pptable->Tliquid1Limit); - pr_info("Tliquid2Limit = %d\n", pptable->Tliquid2Limit); - pr_info("TplxLimit = %d\n", pptable->TplxLimit); - pr_info("FitLimit = %d\n", pptable->FitLimit); - - pr_info("PpmPowerLimit = %d\n", pptable->PpmPowerLimit); - pr_info("PpmTemperatureThreshold = %d\n", pptable->PpmTemperatureThreshold); - - pr_info("MemoryOnPackage = 0x%02x\n", pptable->MemoryOnPackage); - pr_info("padding8_limits = 0x%02x\n", pptable->padding8_limits); - pr_info("Tvr_SocLimit = %d\n", pptable->Tvr_SocLimit); - - pr_info("UlvVoltageOffsetSoc = %d\n", pptable->UlvVoltageOffsetSoc); - pr_info("UlvVoltageOffsetGfx = %d\n", pptable->UlvVoltageOffsetGfx); - - pr_info("UlvSmnclkDid = %d\n", pptable->UlvSmnclkDid); - pr_info("UlvMp1clkDid = %d\n", pptable->UlvMp1clkDid); - pr_info("UlvGfxclkBypass = %d\n", pptable->UlvGfxclkBypass); - pr_info("Padding234 = 0x%02x\n", pptable->Padding234); - - pr_info("MinVoltageGfx = %d\n", pptable->MinVoltageGfx); - pr_info("MinVoltageSoc = %d\n", pptable->MinVoltageSoc); - pr_info("MaxVoltageGfx = %d\n", pptable->MaxVoltageGfx); - pr_info("MaxVoltageSoc = %d\n", pptable->MaxVoltageSoc); - - pr_info("LoadLineResistanceGfx = %d\n", pptable->LoadLineResistanceGfx); - pr_info("LoadLineResistanceSoc = %d\n", pptable->LoadLineResistanceSoc); - - pr_info("[PPCLK_GFXCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n", - pptable->DpmDescriptor[PPCLK_GFXCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_GFXCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_GFXCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_GFXCLK].padding, - pptable->DpmDescriptor[PPCLK_GFXCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_GFXCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.c); - - pr_info("[PPCLK_VCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n", - pptable->DpmDescriptor[PPCLK_VCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_VCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_VCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_VCLK].padding, - pptable->DpmDescriptor[PPCLK_VCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_VCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_VCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_VCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_VCLK].SsCurve.c); - - pr_info("[PPCLK_DCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n", - pptable->DpmDescriptor[PPCLK_DCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_DCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_DCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_DCLK].padding, - pptable->DpmDescriptor[PPCLK_DCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_DCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_DCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_DCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_DCLK].SsCurve.c); - - pr_info("[PPCLK_ECLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n", - pptable->DpmDescriptor[PPCLK_ECLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_ECLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_ECLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_ECLK].padding, - pptable->DpmDescriptor[PPCLK_ECLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_ECLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_ECLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_ECLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_ECLK].SsCurve.c); - - pr_info("[PPCLK_SOCCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n", - pptable->DpmDescriptor[PPCLK_SOCCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_SOCCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_SOCCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_SOCCLK].padding, - pptable->DpmDescriptor[PPCLK_SOCCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_SOCCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.c); - - pr_info("[PPCLK_UCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n", - pptable->DpmDescriptor[PPCLK_UCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_UCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_UCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_UCLK].padding, - pptable->DpmDescriptor[PPCLK_UCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_UCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.c); - - pr_info("[PPCLK_DCEFCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n", - pptable->DpmDescriptor[PPCLK_DCEFCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_DCEFCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_DCEFCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_DCEFCLK].padding, - pptable->DpmDescriptor[PPCLK_DCEFCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_DCEFCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_DCEFCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_DCEFCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_DCEFCLK].SsCurve.c); - - pr_info("[PPCLK_DISPCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n", - pptable->DpmDescriptor[PPCLK_DISPCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_DISPCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_DISPCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_DISPCLK].padding, - pptable->DpmDescriptor[PPCLK_DISPCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_DISPCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_DISPCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_DISPCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_DISPCLK].SsCurve.c); - - pr_info("[PPCLK_PIXCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n", - pptable->DpmDescriptor[PPCLK_PIXCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_PIXCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_PIXCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_PIXCLK].padding, - pptable->DpmDescriptor[PPCLK_PIXCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_PIXCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_PIXCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_PIXCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_PIXCLK].SsCurve.c); - - pr_info("[PPCLK_PHYCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n", - pptable->DpmDescriptor[PPCLK_PHYCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_PHYCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_PHYCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_PHYCLK].padding, - pptable->DpmDescriptor[PPCLK_PHYCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_PHYCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_PHYCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_PHYCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_PHYCLK].SsCurve.c); - - pr_info("[PPCLK_FCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n", - pptable->DpmDescriptor[PPCLK_FCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_FCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_FCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_FCLK].padding, - pptable->DpmDescriptor[PPCLK_FCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_FCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.c); - - - pr_info("FreqTableGfx\n"); - for (i = 0; i < NUM_GFXCLK_DPM_LEVELS; i++) - pr_info(" .[%02d] = %d\n", i, pptable->FreqTableGfx[i]); - - pr_info("FreqTableVclk\n"); - for (i = 0; i < NUM_VCLK_DPM_LEVELS; i++) - pr_info(" .[%02d] = %d\n", i, pptable->FreqTableVclk[i]); - - pr_info("FreqTableDclk\n"); - for (i = 0; i < NUM_DCLK_DPM_LEVELS; i++) - pr_info(" .[%02d] = %d\n", i, pptable->FreqTableDclk[i]); - - pr_info("FreqTableEclk\n"); - for (i = 0; i < NUM_ECLK_DPM_LEVELS; i++) - pr_info(" .[%02d] = %d\n", i, pptable->FreqTableEclk[i]); - - pr_info("FreqTableSocclk\n"); - for (i = 0; i < NUM_SOCCLK_DPM_LEVELS; i++) - pr_info(" .[%02d] = %d\n", i, pptable->FreqTableSocclk[i]); - - pr_info("FreqTableUclk\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - pr_info(" .[%02d] = %d\n", i, pptable->FreqTableUclk[i]); - - pr_info("FreqTableFclk\n"); - for (i = 0; i < NUM_FCLK_DPM_LEVELS; i++) - pr_info(" .[%02d] = %d\n", i, pptable->FreqTableFclk[i]); - - pr_info("FreqTableDcefclk\n"); - for (i = 0; i < NUM_DCEFCLK_DPM_LEVELS; i++) - pr_info(" .[%02d] = %d\n", i, pptable->FreqTableDcefclk[i]); - - pr_info("FreqTableDispclk\n"); - for (i = 0; i < NUM_DISPCLK_DPM_LEVELS; i++) - pr_info(" .[%02d] = %d\n", i, pptable->FreqTableDispclk[i]); - - pr_info("FreqTablePixclk\n"); - for (i = 0; i < NUM_PIXCLK_DPM_LEVELS; i++) - pr_info(" .[%02d] = %d\n", i, pptable->FreqTablePixclk[i]); - - pr_info("FreqTablePhyclk\n"); - for (i = 0; i < NUM_PHYCLK_DPM_LEVELS; i++) - pr_info(" .[%02d] = %d\n", i, pptable->FreqTablePhyclk[i]); - - pr_info("DcModeMaxFreq[PPCLK_GFXCLK] = %d\n", pptable->DcModeMaxFreq[PPCLK_GFXCLK]); - pr_info("DcModeMaxFreq[PPCLK_VCLK] = %d\n", pptable->DcModeMaxFreq[PPCLK_VCLK]); - pr_info("DcModeMaxFreq[PPCLK_DCLK] = %d\n", pptable->DcModeMaxFreq[PPCLK_DCLK]); - pr_info("DcModeMaxFreq[PPCLK_ECLK] = %d\n", pptable->DcModeMaxFreq[PPCLK_ECLK]); - pr_info("DcModeMaxFreq[PPCLK_SOCCLK] = %d\n", pptable->DcModeMaxFreq[PPCLK_SOCCLK]); - pr_info("DcModeMaxFreq[PPCLK_UCLK] = %d\n", pptable->DcModeMaxFreq[PPCLK_UCLK]); - pr_info("DcModeMaxFreq[PPCLK_DCEFCLK] = %d\n", pptable->DcModeMaxFreq[PPCLK_DCEFCLK]); - pr_info("DcModeMaxFreq[PPCLK_DISPCLK] = %d\n", pptable->DcModeMaxFreq[PPCLK_DISPCLK]); - pr_info("DcModeMaxFreq[PPCLK_PIXCLK] = %d\n", pptable->DcModeMaxFreq[PPCLK_PIXCLK]); - pr_info("DcModeMaxFreq[PPCLK_PHYCLK] = %d\n", pptable->DcModeMaxFreq[PPCLK_PHYCLK]); - pr_info("DcModeMaxFreq[PPCLK_FCLK] = %d\n", pptable->DcModeMaxFreq[PPCLK_FCLK]); - pr_info("Padding8_Clks = %d\n", pptable->Padding8_Clks); - - pr_info("Mp0clkFreq\n"); - for (i = 0; i < NUM_MP0CLK_DPM_LEVELS; i++) - pr_info(" .[%d] = %d\n", i, pptable->Mp0clkFreq[i]); - - pr_info("Mp0DpmVoltage\n"); - for (i = 0; i < NUM_MP0CLK_DPM_LEVELS; i++) - pr_info(" .[%d] = %d\n", i, pptable->Mp0DpmVoltage[i]); - - pr_info("GfxclkFidle = 0x%x\n", pptable->GfxclkFidle); - pr_info("GfxclkSlewRate = 0x%x\n", pptable->GfxclkSlewRate); - pr_info("CksEnableFreq = 0x%x\n", pptable->CksEnableFreq); - pr_info("Padding789 = 0x%x\n", pptable->Padding789); - pr_info("CksVoltageOffset[a = 0x%08x b = 0x%08x c = 0x%08x]\n", - pptable->CksVoltageOffset.a, - pptable->CksVoltageOffset.b, - pptable->CksVoltageOffset.c); - pr_info("Padding567[0] = 0x%x\n", pptable->Padding567[0]); - pr_info("Padding567[1] = 0x%x\n", pptable->Padding567[1]); - pr_info("Padding567[2] = 0x%x\n", pptable->Padding567[2]); - pr_info("Padding567[3] = 0x%x\n", pptable->Padding567[3]); - pr_info("GfxclkDsMaxFreq = %d\n", pptable->GfxclkDsMaxFreq); - pr_info("GfxclkSource = 0x%x\n", pptable->GfxclkSource); - pr_info("Padding456 = 0x%x\n", pptable->Padding456); - - pr_info("LowestUclkReservedForUlv = %d\n", pptable->LowestUclkReservedForUlv); - pr_info("Padding8_Uclk[0] = 0x%x\n", pptable->Padding8_Uclk[0]); - pr_info("Padding8_Uclk[1] = 0x%x\n", pptable->Padding8_Uclk[1]); - pr_info("Padding8_Uclk[2] = 0x%x\n", pptable->Padding8_Uclk[2]); - - pr_info("PcieGenSpeed\n"); - for (i = 0; i < NUM_LINK_LEVELS; i++) - pr_info(" .[%d] = %d\n", i, pptable->PcieGenSpeed[i]); - - pr_info("PcieLaneCount\n"); - for (i = 0; i < NUM_LINK_LEVELS; i++) - pr_info(" .[%d] = %d\n", i, pptable->PcieLaneCount[i]); - - pr_info("LclkFreq\n"); - for (i = 0; i < NUM_LINK_LEVELS; i++) - pr_info(" .[%d] = %d\n", i, pptable->LclkFreq[i]); - - pr_info("EnableTdpm = %d\n", pptable->EnableTdpm); - pr_info("TdpmHighHystTemperature = %d\n", pptable->TdpmHighHystTemperature); - pr_info("TdpmLowHystTemperature = %d\n", pptable->TdpmLowHystTemperature); - pr_info("GfxclkFreqHighTempLimit = %d\n", pptable->GfxclkFreqHighTempLimit); - - pr_info("FanStopTemp = %d\n", pptable->FanStopTemp); - pr_info("FanStartTemp = %d\n", pptable->FanStartTemp); - - pr_info("FanGainEdge = %d\n", pptable->FanGainEdge); - pr_info("FanGainHotspot = %d\n", pptable->FanGainHotspot); - pr_info("FanGainLiquid = %d\n", pptable->FanGainLiquid); - pr_info("FanGainVrGfx = %d\n", pptable->FanGainVrGfx); - pr_info("FanGainVrSoc = %d\n", pptable->FanGainVrSoc); - pr_info("FanGainPlx = %d\n", pptable->FanGainPlx); - pr_info("FanGainHbm = %d\n", pptable->FanGainHbm); - pr_info("FanPwmMin = %d\n", pptable->FanPwmMin); - pr_info("FanAcousticLimitRpm = %d\n", pptable->FanAcousticLimitRpm); - pr_info("FanThrottlingRpm = %d\n", pptable->FanThrottlingRpm); - pr_info("FanMaximumRpm = %d\n", pptable->FanMaximumRpm); - pr_info("FanTargetTemperature = %d\n", pptable->FanTargetTemperature); - pr_info("FanTargetGfxclk = %d\n", pptable->FanTargetGfxclk); - pr_info("FanZeroRpmEnable = %d\n", pptable->FanZeroRpmEnable); - pr_info("FanTachEdgePerRev = %d\n", pptable->FanTachEdgePerRev); - - pr_info("FuzzyFan_ErrorSetDelta = %d\n", pptable->FuzzyFan_ErrorSetDelta); - pr_info("FuzzyFan_ErrorRateSetDelta = %d\n", pptable->FuzzyFan_ErrorRateSetDelta); - pr_info("FuzzyFan_PwmSetDelta = %d\n", pptable->FuzzyFan_PwmSetDelta); - pr_info("FuzzyFan_Reserved = %d\n", pptable->FuzzyFan_Reserved); - - pr_info("OverrideAvfsGb[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->OverrideAvfsGb[AVFS_VOLTAGE_GFX]); - pr_info("OverrideAvfsGb[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->OverrideAvfsGb[AVFS_VOLTAGE_SOC]); - pr_info("Padding8_Avfs[0] = %d\n", pptable->Padding8_Avfs[0]); - pr_info("Padding8_Avfs[1] = %d\n", pptable->Padding8_Avfs[1]); - - pr_info("qAvfsGb[AVFS_VOLTAGE_GFX]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qAvfsGb[AVFS_VOLTAGE_GFX].a, - pptable->qAvfsGb[AVFS_VOLTAGE_GFX].b, - pptable->qAvfsGb[AVFS_VOLTAGE_GFX].c); - pr_info("qAvfsGb[AVFS_VOLTAGE_SOC]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qAvfsGb[AVFS_VOLTAGE_SOC].a, - pptable->qAvfsGb[AVFS_VOLTAGE_SOC].b, - pptable->qAvfsGb[AVFS_VOLTAGE_SOC].c); - pr_info("dBtcGbGfxCksOn{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbGfxCksOn.a, - pptable->dBtcGbGfxCksOn.b, - pptable->dBtcGbGfxCksOn.c); - pr_info("dBtcGbGfxCksOff{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbGfxCksOff.a, - pptable->dBtcGbGfxCksOff.b, - pptable->dBtcGbGfxCksOff.c); - pr_info("dBtcGbGfxAfll{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbGfxAfll.a, - pptable->dBtcGbGfxAfll.b, - pptable->dBtcGbGfxAfll.c); - pr_info("dBtcGbSoc{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbSoc.a, - pptable->dBtcGbSoc.b, - pptable->dBtcGbSoc.c); - pr_info("qAgingGb[AVFS_VOLTAGE_GFX]{m = 0x%x b = 0x%x}\n", - pptable->qAgingGb[AVFS_VOLTAGE_GFX].m, - pptable->qAgingGb[AVFS_VOLTAGE_GFX].b); - pr_info("qAgingGb[AVFS_VOLTAGE_SOC]{m = 0x%x b = 0x%x}\n", - pptable->qAgingGb[AVFS_VOLTAGE_SOC].m, - pptable->qAgingGb[AVFS_VOLTAGE_SOC].b); - - pr_info("qStaticVoltageOffset[AVFS_VOLTAGE_GFX]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].a, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].b, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].c); - pr_info("qStaticVoltageOffset[AVFS_VOLTAGE_SOC]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].a, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].b, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].c); - - pr_info("DcTol[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcTol[AVFS_VOLTAGE_GFX]); - pr_info("DcTol[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcTol[AVFS_VOLTAGE_SOC]); - - pr_info("DcBtcEnabled[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcEnabled[AVFS_VOLTAGE_GFX]); - pr_info("DcBtcEnabled[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcEnabled[AVFS_VOLTAGE_SOC]); - pr_info("Padding8_GfxBtc[0] = 0x%x\n", pptable->Padding8_GfxBtc[0]); - pr_info("Padding8_GfxBtc[1] = 0x%x\n", pptable->Padding8_GfxBtc[1]); - - pr_info("DcBtcMin[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcMin[AVFS_VOLTAGE_GFX]); - pr_info("DcBtcMin[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcMin[AVFS_VOLTAGE_SOC]); - pr_info("DcBtcMax[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcMax[AVFS_VOLTAGE_GFX]); - pr_info("DcBtcMax[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcMax[AVFS_VOLTAGE_SOC]); - - pr_info("XgmiLinkSpeed\n"); - for (i = 0; i < NUM_XGMI_LEVELS; i++) - pr_info(" .[%d] = %d\n", i, pptable->XgmiLinkSpeed[i]); - pr_info("XgmiLinkWidth\n"); - for (i = 0; i < NUM_XGMI_LEVELS; i++) - pr_info(" .[%d] = %d\n", i, pptable->XgmiLinkWidth[i]); - pr_info("XgmiFclkFreq\n"); - for (i = 0; i < NUM_XGMI_LEVELS; i++) - pr_info(" .[%d] = %d\n", i, pptable->XgmiFclkFreq[i]); - pr_info("XgmiUclkFreq\n"); - for (i = 0; i < NUM_XGMI_LEVELS; i++) - pr_info(" .[%d] = %d\n", i, pptable->XgmiUclkFreq[i]); - pr_info("XgmiSocclkFreq\n"); - for (i = 0; i < NUM_XGMI_LEVELS; i++) - pr_info(" .[%d] = %d\n", i, pptable->XgmiSocclkFreq[i]); - pr_info("XgmiSocVoltage\n"); - for (i = 0; i < NUM_XGMI_LEVELS; i++) - pr_info(" .[%d] = %d\n", i, pptable->XgmiSocVoltage[i]); - - pr_info("DebugOverrides = 0x%x\n", pptable->DebugOverrides); - pr_info("ReservedEquation0{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation0.a, - pptable->ReservedEquation0.b, - pptable->ReservedEquation0.c); - pr_info("ReservedEquation1{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation1.a, - pptable->ReservedEquation1.b, - pptable->ReservedEquation1.c); - pr_info("ReservedEquation2{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation2.a, - pptable->ReservedEquation2.b, - pptable->ReservedEquation2.c); - pr_info("ReservedEquation3{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation3.a, - pptable->ReservedEquation3.b, - pptable->ReservedEquation3.c); - - pr_info("MinVoltageUlvGfx = %d\n", pptable->MinVoltageUlvGfx); - pr_info("MinVoltageUlvSoc = %d\n", pptable->MinVoltageUlvSoc); - - pr_info("MGpuFanBoostLimitRpm = %d\n", pptable->MGpuFanBoostLimitRpm); - pr_info("padding16_Fan = %d\n", pptable->padding16_Fan); - - pr_info("FanGainVrMem0 = %d\n", pptable->FanGainVrMem0); - pr_info("FanGainVrMem0 = %d\n", pptable->FanGainVrMem0); - - pr_info("DcBtcGb[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcGb[AVFS_VOLTAGE_GFX]); - pr_info("DcBtcGb[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcGb[AVFS_VOLTAGE_SOC]); - - for (i = 0; i < 11; i++) - pr_info("Reserved[%d] = 0x%x\n", i, pptable->Reserved[i]); - - for (i = 0; i < 3; i++) - pr_info("Padding32[%d] = 0x%x\n", i, pptable->Padding32[i]); - - pr_info("MaxVoltageStepGfx = 0x%x\n", pptable->MaxVoltageStepGfx); - pr_info("MaxVoltageStepSoc = 0x%x\n", pptable->MaxVoltageStepSoc); - - pr_info("VddGfxVrMapping = 0x%x\n", pptable->VddGfxVrMapping); - pr_info("VddSocVrMapping = 0x%x\n", pptable->VddSocVrMapping); - pr_info("VddMem0VrMapping = 0x%x\n", pptable->VddMem0VrMapping); - pr_info("VddMem1VrMapping = 0x%x\n", pptable->VddMem1VrMapping); - - pr_info("GfxUlvPhaseSheddingMask = 0x%x\n", pptable->GfxUlvPhaseSheddingMask); - pr_info("SocUlvPhaseSheddingMask = 0x%x\n", pptable->SocUlvPhaseSheddingMask); - pr_info("ExternalSensorPresent = 0x%x\n", pptable->ExternalSensorPresent); - pr_info("Padding8_V = 0x%x\n", pptable->Padding8_V); - - pr_info("GfxMaxCurrent = 0x%x\n", pptable->GfxMaxCurrent); - pr_info("GfxOffset = 0x%x\n", pptable->GfxOffset); - pr_info("Padding_TelemetryGfx = 0x%x\n", pptable->Padding_TelemetryGfx); - - pr_info("SocMaxCurrent = 0x%x\n", pptable->SocMaxCurrent); - pr_info("SocOffset = 0x%x\n", pptable->SocOffset); - pr_info("Padding_TelemetrySoc = 0x%x\n", pptable->Padding_TelemetrySoc); - - pr_info("Mem0MaxCurrent = 0x%x\n", pptable->Mem0MaxCurrent); - pr_info("Mem0Offset = 0x%x\n", pptable->Mem0Offset); - pr_info("Padding_TelemetryMem0 = 0x%x\n", pptable->Padding_TelemetryMem0); - - pr_info("Mem1MaxCurrent = 0x%x\n", pptable->Mem1MaxCurrent); - pr_info("Mem1Offset = 0x%x\n", pptable->Mem1Offset); - pr_info("Padding_TelemetryMem1 = 0x%x\n", pptable->Padding_TelemetryMem1); - - pr_info("AcDcGpio = %d\n", pptable->AcDcGpio); - pr_info("AcDcPolarity = %d\n", pptable->AcDcPolarity); - pr_info("VR0HotGpio = %d\n", pptable->VR0HotGpio); - pr_info("VR0HotPolarity = %d\n", pptable->VR0HotPolarity); - - pr_info("VR1HotGpio = %d\n", pptable->VR1HotGpio); - pr_info("VR1HotPolarity = %d\n", pptable->VR1HotPolarity); - pr_info("Padding1 = 0x%x\n", pptable->Padding1); - pr_info("Padding2 = 0x%x\n", pptable->Padding2); - - pr_info("LedPin0 = %d\n", pptable->LedPin0); - pr_info("LedPin1 = %d\n", pptable->LedPin1); - pr_info("LedPin2 = %d\n", pptable->LedPin2); - pr_info("padding8_4 = 0x%x\n", pptable->padding8_4); - - pr_info("PllGfxclkSpreadEnabled = %d\n", pptable->PllGfxclkSpreadEnabled); - pr_info("PllGfxclkSpreadPercent = %d\n", pptable->PllGfxclkSpreadPercent); - pr_info("PllGfxclkSpreadFreq = %d\n", pptable->PllGfxclkSpreadFreq); - - pr_info("UclkSpreadEnabled = %d\n", pptable->UclkSpreadEnabled); - pr_info("UclkSpreadPercent = %d\n", pptable->UclkSpreadPercent); - pr_info("UclkSpreadFreq = %d\n", pptable->UclkSpreadFreq); - - pr_info("FclkSpreadEnabled = %d\n", pptable->FclkSpreadEnabled); - pr_info("FclkSpreadPercent = %d\n", pptable->FclkSpreadPercent); - pr_info("FclkSpreadFreq = %d\n", pptable->FclkSpreadFreq); - - pr_info("FllGfxclkSpreadEnabled = %d\n", pptable->FllGfxclkSpreadEnabled); - pr_info("FllGfxclkSpreadPercent = %d\n", pptable->FllGfxclkSpreadPercent); - pr_info("FllGfxclkSpreadFreq = %d\n", pptable->FllGfxclkSpreadFreq); - - for (i = 0; i < I2C_CONTROLLER_NAME_COUNT; i++) { - pr_info("I2cControllers[%d]:\n", i); - pr_info(" .Enabled = %d\n", - pptable->I2cControllers[i].Enabled); - pr_info(" .SlaveAddress = 0x%x\n", - pptable->I2cControllers[i].SlaveAddress); - pr_info(" .ControllerPort = %d\n", - pptable->I2cControllers[i].ControllerPort); - pr_info(" .ControllerName = %d\n", - pptable->I2cControllers[i].ControllerName); - pr_info(" .ThermalThrottler = %d\n", - pptable->I2cControllers[i].ThermalThrottler); - pr_info(" .I2cProtocol = %d\n", - pptable->I2cControllers[i].I2cProtocol); - pr_info(" .I2cSpeed = %d\n", - pptable->I2cControllers[i].I2cSpeed); - } - - for (i = 0; i < 10; i++) - pr_info("BoardReserved[%d] = 0x%x\n", i, pptable->BoardReserved[i]); - - for (i = 0; i < 8; i++) - pr_info("MmHubPadding[%d] = 0x%x\n", i, pptable->MmHubPadding[i]); -} -#endif - static int check_powerplay_tables( struct pp_hwmgr *hwmgr, const ATOM_Vega20_POWERPLAYTABLE *powerplay_table) @@ -652,8 +80,6 @@ static int check_powerplay_tables( return -EINVAL; } - //dump_pptable(&powerplay_table->smcPPTable); - return 0; } diff --git a/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h index 9118fcddbf11..227bf0e84a13 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h +++ b/drivers/gpu/drm/amd/pm/powerplay/inc/hwmgr.h @@ -60,7 +60,7 @@ struct vi_dpm_level { struct vi_dpm_table { uint32_t count; - struct vi_dpm_level dpm_level[] __counted_by(count); + struct vi_dpm_level dpm_level[]; }; #define PCIE_PERF_REQ_REMOVE_REGISTRY 0 @@ -91,7 +91,7 @@ struct phm_set_power_state_input { struct phm_clock_array { uint32_t count; - uint32_t values[] __counted_by(count); + uint32_t values[]; }; struct phm_clock_voltage_dependency_record { @@ -123,7 +123,7 @@ struct phm_acpclock_voltage_dependency_record { struct phm_clock_voltage_dependency_table { uint32_t count; - struct phm_clock_voltage_dependency_record entries[] __counted_by(count); + struct phm_clock_voltage_dependency_record entries[]; }; struct phm_phase_shedding_limits_record { @@ -140,7 +140,7 @@ struct phm_uvd_clock_voltage_dependency_record { struct phm_uvd_clock_voltage_dependency_table { uint8_t count; - struct phm_uvd_clock_voltage_dependency_record entries[] __counted_by(count); + struct phm_uvd_clock_voltage_dependency_record entries[]; }; struct phm_acp_clock_voltage_dependency_record { @@ -150,7 +150,7 @@ struct phm_acp_clock_voltage_dependency_record { struct phm_acp_clock_voltage_dependency_table { uint32_t count; - struct phm_acp_clock_voltage_dependency_record entries[] __counted_by(count); + struct phm_acp_clock_voltage_dependency_record entries[]; }; struct phm_vce_clock_voltage_dependency_record { @@ -161,32 +161,32 @@ struct phm_vce_clock_voltage_dependency_record { struct phm_phase_shedding_limits_table { uint32_t count; - struct phm_phase_shedding_limits_record entries[] __counted_by(count); + struct phm_phase_shedding_limits_record entries[]; }; struct phm_vceclock_voltage_dependency_table { uint8_t count; - struct phm_vceclock_voltage_dependency_record entries[] __counted_by(count); + struct phm_vceclock_voltage_dependency_record entries[]; }; struct phm_uvdclock_voltage_dependency_table { uint8_t count; - struct phm_uvdclock_voltage_dependency_record entries[] __counted_by(count); + struct phm_uvdclock_voltage_dependency_record entries[]; }; struct phm_samuclock_voltage_dependency_table { uint8_t count; - struct phm_samuclock_voltage_dependency_record entries[] __counted_by(count); + struct phm_samuclock_voltage_dependency_record entries[]; }; struct phm_acpclock_voltage_dependency_table { uint32_t count; - struct phm_acpclock_voltage_dependency_record entries[] __counted_by(count); + struct phm_acpclock_voltage_dependency_record entries[]; }; struct phm_vce_clock_voltage_dependency_table { uint8_t count; - struct phm_vce_clock_voltage_dependency_record entries[] __counted_by(count); + struct phm_vce_clock_voltage_dependency_record entries[]; }; @@ -393,7 +393,7 @@ union phm_cac_leakage_record { struct phm_cac_leakage_table { uint32_t count; - union phm_cac_leakage_record entries[] __counted_by(count); + union phm_cac_leakage_record entries[]; }; struct phm_samu_clock_voltage_dependency_record { @@ -404,7 +404,7 @@ struct phm_samu_clock_voltage_dependency_record { struct phm_samu_clock_voltage_dependency_table { uint8_t count; - struct phm_samu_clock_voltage_dependency_record entries[] __counted_by(count); + struct phm_samu_clock_voltage_dependency_record entries[]; }; struct phm_cac_tdp_table { diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index bb3bc68dfc39..67d5a8123416 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -140,7 +140,8 @@ int smu_set_soft_freq_range(struct smu_context *smu, ret = smu->ppt_funcs->set_soft_freq_limited_range(smu, clk_type, min, - max); + max, + false); return ret; } @@ -549,7 +550,8 @@ bool is_support_sw_smu(struct amdgpu_device *adev) if (adev->asic_type == CHIP_VEGA20) return false; - if (amdgpu_ip_version(adev, MP1_HWIP, 0) >= IP_VERSION(11, 0, 0)) + if ((amdgpu_ip_version(adev, MP1_HWIP, 0) >= IP_VERSION(11, 0, 0)) && + amdgpu_device_ip_is_valid(adev, AMD_IP_BLOCK_TYPE_SMC)) return true; return false; @@ -741,9 +743,9 @@ static int smu_set_funcs(struct amdgpu_device *adev) return 0; } -static int smu_early_init(void *handle) +static int smu_early_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct smu_context *smu; int r; @@ -825,9 +827,9 @@ static int smu_apply_default_config_table_settings(struct smu_context *smu) return smu_set_config_table(smu, &adev->pm.config_table); } -static int smu_late_init(void *handle) +static int smu_late_init(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct smu_context *smu = adev->powerplay.pp_handle; int ret = 0; @@ -1234,9 +1236,17 @@ static void smu_init_xgmi_plpd_mode(struct smu_context *smu) } } -static int smu_sw_init(void *handle) +static bool smu_is_workload_profile_available(struct smu_context *smu, + u32 profile) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + if (profile >= PP_SMC_POWER_PROFILE_COUNT) + return false; + return smu->workload_map && smu->workload_map[profile].valid_mapping; +} + +static int smu_sw_init(struct amdgpu_ip_block *ip_block) +{ + struct amdgpu_device *adev = ip_block->adev; struct smu_context *smu = adev->powerplay.pp_handle; int ret; @@ -1264,7 +1274,12 @@ static int smu_sw_init(void *handle) smu->workload_prority[PP_SMC_POWER_PROFILE_VR] = 4; smu->workload_prority[PP_SMC_POWER_PROFILE_COMPUTE] = 5; smu->workload_prority[PP_SMC_POWER_PROFILE_CUSTOM] = 6; - smu->workload_mask = 1 << smu->workload_prority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT]; + + if (smu->is_apu || + !smu_is_workload_profile_available(smu, PP_SMC_POWER_PROFILE_FULLSCREEN3D)) + smu->workload_mask = 1 << smu->workload_prority[PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT]; + else + smu->workload_mask = 1 << smu->workload_prority[PP_SMC_POWER_PROFILE_FULLSCREEN3D]; smu->workload_setting[0] = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT; smu->workload_setting[1] = PP_SMC_POWER_PROFILE_FULLSCREEN3D; @@ -1313,9 +1328,9 @@ static int smu_sw_init(void *handle) return 0; } -static int smu_sw_fini(void *handle) +static int smu_sw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct smu_context *smu = adev->powerplay.pp_handle; int ret; @@ -1786,10 +1801,10 @@ static int smu_start_smc_engine(struct smu_context *smu) return ret; } -static int smu_hw_init(void *handle) +static int smu_hw_init(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct smu_context *smu = adev->powerplay.pp_handle; if (amdgpu_sriov_vf(adev) && !amdgpu_sriov_is_pp_one_vf(adev)) { @@ -2008,9 +2023,9 @@ static int smu_reset_mp1_state(struct smu_context *smu) return ret; } -static int smu_hw_fini(void *handle) +static int smu_hw_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct smu_context *smu = adev->powerplay.pp_handle; int ret; @@ -2041,9 +2056,9 @@ static int smu_hw_fini(void *handle) return 0; } -static void smu_late_fini(void *handle) +static void smu_late_fini(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = handle; + struct amdgpu_device *adev = ip_block->adev; struct smu_context *smu = adev->powerplay.pp_handle; kfree(smu); @@ -2052,26 +2067,31 @@ static void smu_late_fini(void *handle) static int smu_reset(struct smu_context *smu) { struct amdgpu_device *adev = smu->adev; + struct amdgpu_ip_block *ip_block; int ret; - ret = smu_hw_fini(adev); + ip_block = amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_SMC); + if (!ip_block) + return -EINVAL; + + ret = smu_hw_fini(ip_block); if (ret) return ret; - ret = smu_hw_init(adev); + ret = smu_hw_init(ip_block); if (ret) return ret; - ret = smu_late_init(adev); + ret = smu_late_init(ip_block); if (ret) return ret; return 0; } -static int smu_suspend(void *handle) +static int smu_suspend(struct amdgpu_ip_block *ip_block) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct smu_context *smu = adev->powerplay.pp_handle; int ret; uint64_t count; @@ -2103,10 +2123,10 @@ static int smu_suspend(void *handle) return 0; } -static int smu_resume(void *handle) +static int smu_resume(struct amdgpu_ip_block *ip_block) { int ret; - struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct amdgpu_device *adev = ip_block->adev; struct smu_context *smu = adev->powerplay.pp_handle; if (amdgpu_sriov_vf(adev)&& !amdgpu_sriov_is_pp_one_vf(adev)) @@ -2226,7 +2246,7 @@ static int smu_bump_power_profile_mode(struct smu_context *smu, static int smu_adjust_power_state_dynamic(struct smu_context *smu, enum amd_dpm_forced_level level, bool skip_display_settings, - bool force_update) + bool init) { int ret = 0; int index = 0; @@ -2255,7 +2275,7 @@ static int smu_adjust_power_state_dynamic(struct smu_context *smu, } } - if (force_update || smu_dpm_ctx->dpm_level != level) { + if (smu_dpm_ctx->dpm_level != level) { ret = smu_asic_set_performance_level(smu, level); if (ret) { dev_err(smu->adev->dev, "Failed to set performance level!"); @@ -2272,7 +2292,7 @@ static int smu_adjust_power_state_dynamic(struct smu_context *smu, index = index > 0 && index <= WORKLOAD_POLICY_MAX ? index - 1 : 0; workload[0] = smu->workload_setting[index]; - if (force_update || smu->power_profile_mode != workload[0]) + if (init || smu->power_profile_mode != workload[0]) smu_bump_power_profile_mode(smu, workload, 0); } diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h index b44a185d07e8..8bb32b3f0d9c 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/amdgpu_smu.h @@ -858,11 +858,6 @@ struct pptable_funcs { */ int (*display_disable_memory_clock_switch)(struct smu_context *smu, bool disable_memory_clock_switch); - /** - * @dump_pptable: Print the power play table to the system log. - */ - void (*dump_pptable)(struct smu_context *smu); - /** * @get_power_limit: Get the device's power limits. */ @@ -1260,7 +1255,8 @@ struct pptable_funcs { * @set_soft_freq_limited_range: Set the soft frequency range of a clock * domain in MHz. */ - int (*set_soft_freq_limited_range)(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t min, uint32_t max); + int (*set_soft_freq_limited_range)(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t min, uint32_t max, + bool automatic); /** * @set_power_source: Notify the SMU of the current power source. diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0.h index ee457a6f0813..c2fd0a4a13e5 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu14_driver_if_v14_0.h @@ -25,7 +25,7 @@ #define SMU14_DRIVER_IF_V14_0_H //Increment this version if SkuTable_t or BoardTable_t change -#define PPTABLE_VERSION 0x18 +#define PPTABLE_VERSION 0x1B #define NUM_GFXCLK_DPM_LEVELS 16 #define NUM_SOCCLK_DPM_LEVELS 8 @@ -145,7 +145,7 @@ typedef enum { } FEATURE_BTC_e; // Debug Overrides Bitmask -#define DEBUG_OVERRIDE_DISABLE_VOLT_LINK_VCN_FCLK 0x00000001 +#define DEBUG_OVERRIDE_NOT_USE 0x00000001 #define DEBUG_OVERRIDE_DISABLE_VOLT_LINK_DCN_FCLK 0x00000002 #define DEBUG_OVERRIDE_DISABLE_VOLT_LINK_MP0_FCLK 0x00000004 #define DEBUG_OVERRIDE_DISABLE_VOLT_LINK_VCN_DCFCLK 0x00000008 @@ -161,6 +161,7 @@ typedef enum { #define DEBUG_OVERRIDE_ENABLE_SOC_VF_BRINGUP_MODE 0x00002000 #define DEBUG_OVERRIDE_ENABLE_PER_WGP_RESIENCY 0x00004000 #define DEBUG_OVERRIDE_DISABLE_MEMORY_VOLTAGE_SCALING 0x00008000 +#define DEBUG_OVERRIDE_DFLL_BTC_FCW_LOG 0x00010000 // VR Mapping Bit Defines #define VR_MAPPING_VR_SELECT_MASK 0x01 @@ -391,6 +392,21 @@ typedef struct { EccInfo_t EccInfo[24]; } EccInfoTable_t; +#define EPCS_HIGH_POWER 600 +#define EPCS_NORMAL_POWER 450 +#define EPCS_LOW_POWER 300 +#define EPCS_SHORTED_POWER 150 +#define EPCS_NO_BOOTUP 0 + +typedef enum{ + EPCS_SHORTED_LIMIT, + EPCS_LOW_POWER_LIMIT, + EPCS_NORMAL_POWER_LIMIT, + EPCS_HIGH_POWER_LIMIT, + EPCS_NOT_CONFIGURED, + EPCS_STATUS_COUNT, +} EPCS_STATUS_e; + //D3HOT sequences typedef enum { BACO_SEQUENCE, @@ -662,7 +678,7 @@ typedef enum { } PP_GRTAVFS_FW_SEP_FUSE_e; #define PP_NUM_RTAVFS_PWL_ZONES 5 - +#define PP_NUM_PSM_DIDT_PWL_ZONES 3 // VBIOS or PPLIB configures telemetry slope and offset. Only slope expected to be set for SVI3 // Slope Q1.7, Offset Q1.2 @@ -746,10 +762,10 @@ typedef struct { uint16_t Padding; //Frequency changes - int16_t GfxclkFmin; // MHz - int16_t GfxclkFmax; // MHz - uint16_t UclkFmin; // MHz - uint16_t UclkFmax; // MHz + int16_t GfxclkFoffset; + uint16_t Padding1; + uint16_t UclkFmin; + uint16_t UclkFmax; uint16_t FclkFmin; uint16_t FclkFmax; @@ -770,19 +786,23 @@ typedef struct { uint8_t MaxOpTemp; uint8_t AdvancedOdModeEnabled; - uint8_t Padding1[3]; + uint8_t Padding2[3]; uint16_t GfxVoltageFullCtrlMode; uint16_t SocVoltageFullCtrlMode; uint16_t GfxclkFullCtrlMode; uint16_t UclkFullCtrlMode; uint16_t FclkFullCtrlMode; - uint16_t Padding2; + uint16_t Padding3; int16_t GfxEdc; int16_t GfxPccLimitControl; - uint32_t Spare[10]; + uint16_t GfxclkFmaxVmax; + uint8_t GfxclkFmaxVmaxTemperature; + uint8_t Padding4[1]; + + uint32_t Spare[9]; uint32_t MmHubPadding[8]; // SMU internal use. Adding here instead of external as a workaround } OverDriveTable_t; @@ -802,8 +822,8 @@ typedef struct { uint16_t VddSocVmax; //gfxclk - int16_t GfxclkFmin; // MHz - int16_t GfxclkFmax; // MHz + int16_t GfxclkFoffset; + uint16_t Padding; //uclk uint16_t UclkFmin; // MHz uint16_t UclkFmax; // MHz @@ -828,7 +848,7 @@ typedef struct { uint8_t FanZeroRpmEnable; //temperature uint8_t MaxOpTemp; - uint8_t Padding[2]; + uint8_t Padding1[2]; //Full Ctrl uint16_t GfxVoltageFullCtrlMode; @@ -839,7 +859,7 @@ typedef struct { //EDC int16_t GfxEdc; int16_t GfxPccLimitControl; - int16_t Padding1; + int16_t Padding2; uint32_t Spare[5]; } OverDriveLimits_t; @@ -987,8 +1007,9 @@ typedef struct { uint16_t BaseClockDc; uint16_t GameClockDc; uint16_t BoostClockDc; - - uint32_t Reserved[4]; + uint16_t MaxReportedClock; + uint16_t Padding; + uint32_t Reserved[3]; } DriverReportedClocks_t; typedef struct { @@ -1132,7 +1153,7 @@ typedef struct { uint32_t DcModeMaxFreq [PPCLK_COUNT ]; // In MHz uint16_t GfxclkAibFmax; - uint16_t GfxclkFreqCap; + uint16_t GfxDpmPadding; //GFX Idle Power Settings uint16_t GfxclkFgfxoffEntry; // Entry in RLC stage (PLL), in Mhz @@ -1172,8 +1193,7 @@ typedef struct { uint32_t DvoFmaxLowScaler; //Unitless float // GFX DCS - uint16_t DcsGfxOffVoltage; //Voltage in mV(Q2) applied to VDDGFX when entering DCS GFXOFF phase - uint16_t PaddingDcs; + uint32_t PaddingDcs; uint16_t DcsMinGfxOffTime; //Minimum amount of time PMFW shuts GFX OFF as part of GFX DCS phase uint16_t DcsMaxGfxOffTime; //Maximum amount of time PMFW can shut GFX OFF as part of GFX DCS phase at a stretch. @@ -1205,8 +1225,7 @@ typedef struct { uint16_t DalDcModeMaxUclkFreq; uint8_t PaddingsMem[2]; //FCLK Section - uint16_t FclkDpmDisallowPstateFreq; //Frequency which FW will target when indicated that display config cannot support P-state. Set to 0 use FW calculated value - uint16_t PaddingFclk; + uint32_t PaddingFclk; // Link DPM Settings uint8_t PcieGenSpeed[NUM_LINK_LEVELS]; ///< 0:PciE-gen1 1:PciE-gen2 2:PciE-gen3 3:PciE-gen4 4:PciE-gen5 @@ -1215,12 +1234,19 @@ typedef struct { // SECTION: VDD_GFX AVFS uint8_t OverrideGfxAvfsFuses; - uint8_t GfxAvfsPadding[3]; + uint8_t GfxAvfsPadding[1]; + uint16_t DroopGBStDev; uint32_t SocHwRtAvfsFuses[PP_GRTAVFS_HW_FUSE_COUNT]; //new added for Soc domain uint32_t GfxL2HwRtAvfsFuses[PP_GRTAVFS_HW_FUSE_COUNT]; //see fusedoc for encoding //uint32_t GfxSeHwRtAvfsFuses[PP_GRTAVFS_HW_FUSE_COUNT]; - uint32_t spare_HwRtAvfsFuses[PP_GRTAVFS_HW_FUSE_COUNT]; + + uint16_t PsmDidt_Vcross[PP_NUM_PSM_DIDT_PWL_ZONES-1]; + uint32_t PsmDidt_StaticDroop_A[PP_NUM_PSM_DIDT_PWL_ZONES]; + uint32_t PsmDidt_StaticDroop_B[PP_NUM_PSM_DIDT_PWL_ZONES]; + uint32_t PsmDidt_DynDroop_A[PP_NUM_PSM_DIDT_PWL_ZONES]; + uint32_t PsmDidt_DynDroop_B[PP_NUM_PSM_DIDT_PWL_ZONES]; + uint32_t spare_HwRtAvfsFuses[19]; uint32_t SocCommonRtAvfs[PP_GRTAVFS_FW_COMMON_FUSE_COUNT]; uint32_t GfxCommonRtAvfs[PP_GRTAVFS_FW_COMMON_FUSE_COUNT]; @@ -1246,11 +1272,7 @@ typedef struct { uint32_t dGbV_dT_vmin; uint32_t dGbV_dT_vmax; - //Unused: PMFW-9370 - uint32_t V2F_vmin_range_low; - uint32_t V2F_vmin_range_high; - uint32_t V2F_vmax_range_low; - uint32_t V2F_vmax_range_high; + uint32_t PaddingV2F[4]; AvfsDcBtcParams_t DcBtcGfxParams; QuadraticInt_t SSCurve_GFX; @@ -1327,18 +1349,18 @@ typedef struct { uint16_t PsmDidtReleaseTimer; uint32_t PsmDidtStallPattern; //Will be written to both pattern 1 and didt_static_level_prog // CAC EDC - uint32_t Leakage_C0; // in IEEE float - uint32_t Leakage_C1; // in IEEE float - uint32_t Leakage_C2; // in IEEE float - uint32_t Leakage_C3; // in IEEE float - uint32_t Leakage_C4; // in IEEE float - uint32_t Leakage_C5; // in IEEE float - uint32_t GFX_CLK_SCALAR; // in IEEE float - uint32_t GFX_CLK_INTERCEPT; // in IEEE float - uint32_t GFX_CAC_M; // in IEEE float - uint32_t GFX_CAC_B; // in IEEE float - uint32_t VDD_GFX_CurrentLimitGuardband; // in IEEE float - uint32_t DynToTotalCacScalar; // in IEEE + uint32_t CacEdcCacLeakageC0; + uint32_t CacEdcCacLeakageC1; + uint32_t CacEdcCacLeakageC2; + uint32_t CacEdcCacLeakageC3; + uint32_t CacEdcCacLeakageC4; + uint32_t CacEdcCacLeakageC5; + uint32_t CacEdcGfxClkScalar; + uint32_t CacEdcGfxClkIntercept; + uint32_t CacEdcCac_m; + uint32_t CacEdcCac_b; + uint32_t CacEdcCurrLimitGuardband; + uint32_t CacEdcDynToTotalCacRatio; // GFX EDC XVMIN uint32_t XVmin_Gfx_EdcThreshScalar; uint32_t XVmin_Gfx_EdcEnableFreq; @@ -1467,7 +1489,7 @@ typedef struct { uint8_t VddqOffEnabled; uint8_t PaddingUmcFlags[2]; - uint32_t PostVoltageSetBacoDelay; // in microseconds. Amount of time FW will wait after power good is established or PSI0 command is issued + uint32_t Paddign1; uint32_t BacoEntryDelay; // in milliseconds. Amount of time FW will wait to trigger BACO entry after receiving entry notification from OS uint8_t FuseWritePowerMuxPresent; @@ -1530,7 +1552,7 @@ typedef struct { int16_t FuzzyFan_ErrorSetDelta; int16_t FuzzyFan_ErrorRateSetDelta; int16_t FuzzyFan_PwmSetDelta; - uint16_t FuzzyFan_Reserved; + uint16_t FanPadding2; uint16_t FwCtfLimit[TEMP_COUNT]; @@ -1547,9 +1569,10 @@ typedef struct { uint16_t FanSpare[1]; uint8_t FanIntakeSensorSupport; uint8_t FanIntakePadding; - uint32_t FanAmbientPerfBoostThreshold; uint32_t FanSpare2[12]; + uint32_t ODFeatureCtrlMask; + uint16_t TemperatureLimit_Hynix; // In degrees Celsius. Memory temperature limit associated with Hynix uint16_t TemperatureLimit_Micron; // In degrees Celsius. Memory temperature limit associated with Micron uint16_t TemperatureFwCtfLimit_Hynix; @@ -1637,7 +1660,7 @@ typedef struct { uint16_t AverageDclk0Frequency ; uint16_t AverageVclk1Frequency ; uint16_t AverageDclk1Frequency ; - uint16_t PCIeBusy ; + uint16_t AveragePCIeBusy ; uint16_t dGPU_W_MAX ; uint16_t padding ; @@ -1665,12 +1688,12 @@ typedef struct { uint16_t AverageGfxActivity ; uint16_t AverageUclkActivity ; - uint16_t Vcn0ActivityPercentage ; + uint16_t AverageVcn0ActivityPercentage; uint16_t Vcn1ActivityPercentage ; uint32_t EnergyAccumulator; uint16_t AverageSocketPower; - uint16_t MovingAverageTotalBoardPower; + uint16_t AverageTotalBoardPower; uint16_t AvgTemperature[TEMP_COUNT]; uint16_t AvgTemperatureFanIntake; @@ -1684,7 +1707,8 @@ typedef struct { uint8_t ThrottlingPercentage[THROTTLER_COUNT]; - uint8_t padding1[3]; + uint8_t VmaxThrottlingPercentage; + uint8_t padding1[2]; //metrics for D3hot entry/exit and driver ARM msgs uint32_t D3HotEntryCountPerMode[D3HOT_SEQUENCE_COUNT]; @@ -1693,7 +1717,7 @@ typedef struct { uint16_t ApuSTAPMSmartShiftLimit; uint16_t ApuSTAPMLimit; - uint16_t MovingAvgApuSocketPower; + uint16_t AvgApuSocketPower; uint16_t AverageUclkActivity_MAX; @@ -1823,6 +1847,17 @@ typedef struct { #define TABLE_TRANSFER_FAILED 0xFF #define TABLE_TRANSFER_PENDING 0xAB +#define TABLE_PPT_FAILED 0x100 +#define TABLE_TDC_FAILED 0x200 +#define TABLE_TEMP_FAILED 0x400 +#define TABLE_FAN_TARGET_TEMP_FAILED 0x800 +#define TABLE_FAN_STOP_TEMP_FAILED 0x1000 +#define TABLE_FAN_START_TEMP_FAILED 0x2000 +#define TABLE_FAN_PWM_MIN_FAILED 0x4000 +#define TABLE_ACOUSTIC_TARGET_RPM_FAILED 0x8000 +#define TABLE_ACOUSTIC_LIMIT_RPM_FAILED 0x10000 +#define TABLE_MGPU_ACOUSTIC_TARGET_RPM_FAILED 0x20000 + // Table types #define TABLE_PPTABLE 0 #define TABLE_COMBO_PPTABLE 1 @@ -1849,5 +1884,6 @@ typedef struct { #define IH_INTERRUPT_CONTEXT_ID_THERMAL_THROTTLING 0x7 #define IH_INTERRUPT_CONTEXT_ID_FAN_ABNORMAL 0x8 #define IH_INTERRUPT_CONTEXT_ID_FAN_RECOVERY 0x9 +#define IH_INTERRUPT_CONTEXT_ID_DYNAMIC_TABLE 0xA #endif diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h index 822c6425d90e..0f96b8c59a0e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/pmfw_if/smu_v13_0_6_pmfw.h @@ -123,7 +123,7 @@ typedef enum { VOLTAGE_GUARDBAND_COUNT } GFX_GUARDBAND_e; -#define SMU_METRICS_TABLE_VERSION 0xD +#define SMU_METRICS_TABLE_VERSION 0xE typedef struct __attribute__((packed, aligned(4))) { uint32_t AccumulationCounter; @@ -231,6 +231,9 @@ typedef struct __attribute__((packed, aligned(4))) { // PER XCD ACTIVITY uint32_t GfxBusy[8]; uint64_t GfxBusyAcc[8]; + + //PCIE BW Data and error count + uint32_t PCIeOtherEndRecoveryAcc; // The Pcie counter itself is accumulated } MetricsTableX_t; typedef struct __attribute__((packed, aligned(4))) { diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h index c2ab336bb530..ed8304d82831 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v11_0.h @@ -255,7 +255,7 @@ int smu_v11_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk_type c uint32_t *min, uint32_t *max); int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, uint32_t max); + uint32_t min, uint32_t max, bool automatic); int smu_v11_0_set_hard_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v12_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v12_0.h index 1ad2dff71090..0886d8cffbd0 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v12_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v12_0.h @@ -56,7 +56,7 @@ int smu_v12_0_set_default_dpm_tables(struct smu_context *smu); int smu_v12_0_mode2_reset(struct smu_context *smu); int smu_v12_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, uint32_t max); + uint32_t min, uint32_t max, bool automatic); int smu_v12_0_set_driver_table_location(struct smu_context *smu); diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h index e58220a7ee2f..044d6893b43e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v13_0.h @@ -219,7 +219,7 @@ int smu_v13_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk_type c uint32_t *min, uint32_t *max); int smu_v13_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, uint32_t max); + uint32_t min, uint32_t max, bool automatic); int smu_v13_0_set_hard_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, diff --git a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v14_0.h b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v14_0.h index 46b456590a08..07c220102c1d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v14_0.h +++ b/drivers/gpu/drm/amd/pm/swsmu/inc/smu_v14_0.h @@ -28,7 +28,7 @@ #define SMU14_DRIVER_IF_VERSION_INV 0xFFFFFFFF #define SMU14_DRIVER_IF_VERSION_SMU_V14_0_0 0x7 #define SMU14_DRIVER_IF_VERSION_SMU_V14_0_1 0x6 -#define SMU14_DRIVER_IF_VERSION_SMU_V14_0_2 0x26 +#define SMU14_DRIVER_IF_VERSION_SMU_V14_0_2 0x2E #define FEATURE_MASK(feature) (1ULL << feature) @@ -186,7 +186,7 @@ int smu_v14_0_get_dpm_ultimate_freq(struct smu_context *smu, enum smu_clk_type c uint32_t *min, uint32_t *max); int smu_v14_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, uint32_t max); + uint32_t min, uint32_t max, bool automatic); int smu_v14_0_set_hard_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index c0f6b59369b7..5ad09323a29d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -1559,437 +1559,6 @@ static int arcturus_set_performance_level(struct smu_context *smu, return smu_v11_0_set_performance_level(smu, level); } -static void arcturus_dump_pptable(struct smu_context *smu) -{ - struct smu_table_context *table_context = &smu->smu_table; - PPTable_t *pptable = table_context->driver_pptable; - int i; - - dev_info(smu->adev->dev, "Dumped PPTable:\n"); - - dev_info(smu->adev->dev, "Version = 0x%08x\n", pptable->Version); - - dev_info(smu->adev->dev, "FeaturesToRun[0] = 0x%08x\n", pptable->FeaturesToRun[0]); - dev_info(smu->adev->dev, "FeaturesToRun[1] = 0x%08x\n", pptable->FeaturesToRun[1]); - - for (i = 0; i < PPT_THROTTLER_COUNT; i++) { - dev_info(smu->adev->dev, "SocketPowerLimitAc[%d] = %d\n", i, pptable->SocketPowerLimitAc[i]); - dev_info(smu->adev->dev, "SocketPowerLimitAcTau[%d] = %d\n", i, pptable->SocketPowerLimitAcTau[i]); - } - - dev_info(smu->adev->dev, "TdcLimitSoc = %d\n", pptable->TdcLimitSoc); - dev_info(smu->adev->dev, "TdcLimitSocTau = %d\n", pptable->TdcLimitSocTau); - dev_info(smu->adev->dev, "TdcLimitGfx = %d\n", pptable->TdcLimitGfx); - dev_info(smu->adev->dev, "TdcLimitGfxTau = %d\n", pptable->TdcLimitGfxTau); - - dev_info(smu->adev->dev, "TedgeLimit = %d\n", pptable->TedgeLimit); - dev_info(smu->adev->dev, "ThotspotLimit = %d\n", pptable->ThotspotLimit); - dev_info(smu->adev->dev, "TmemLimit = %d\n", pptable->TmemLimit); - dev_info(smu->adev->dev, "Tvr_gfxLimit = %d\n", pptable->Tvr_gfxLimit); - dev_info(smu->adev->dev, "Tvr_memLimit = %d\n", pptable->Tvr_memLimit); - dev_info(smu->adev->dev, "Tvr_socLimit = %d\n", pptable->Tvr_socLimit); - dev_info(smu->adev->dev, "FitLimit = %d\n", pptable->FitLimit); - - dev_info(smu->adev->dev, "PpmPowerLimit = %d\n", pptable->PpmPowerLimit); - dev_info(smu->adev->dev, "PpmTemperatureThreshold = %d\n", pptable->PpmTemperatureThreshold); - - dev_info(smu->adev->dev, "ThrottlerControlMask = %d\n", pptable->ThrottlerControlMask); - - dev_info(smu->adev->dev, "UlvVoltageOffsetGfx = %d\n", pptable->UlvVoltageOffsetGfx); - dev_info(smu->adev->dev, "UlvPadding = 0x%08x\n", pptable->UlvPadding); - - dev_info(smu->adev->dev, "UlvGfxclkBypass = %d\n", pptable->UlvGfxclkBypass); - dev_info(smu->adev->dev, "Padding234[0] = 0x%02x\n", pptable->Padding234[0]); - dev_info(smu->adev->dev, "Padding234[1] = 0x%02x\n", pptable->Padding234[1]); - dev_info(smu->adev->dev, "Padding234[2] = 0x%02x\n", pptable->Padding234[2]); - - dev_info(smu->adev->dev, "MinVoltageGfx = %d\n", pptable->MinVoltageGfx); - dev_info(smu->adev->dev, "MinVoltageSoc = %d\n", pptable->MinVoltageSoc); - dev_info(smu->adev->dev, "MaxVoltageGfx = %d\n", pptable->MaxVoltageGfx); - dev_info(smu->adev->dev, "MaxVoltageSoc = %d\n", pptable->MaxVoltageSoc); - - dev_info(smu->adev->dev, "LoadLineResistanceGfx = %d\n", pptable->LoadLineResistanceGfx); - dev_info(smu->adev->dev, "LoadLineResistanceSoc = %d\n", pptable->LoadLineResistanceSoc); - - dev_info(smu->adev->dev, "[PPCLK_GFXCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_GFXCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_GFXCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_GFXCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_GFXCLK].padding, - pptable->DpmDescriptor[PPCLK_GFXCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_GFXCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_GFXCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_VCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_VCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_VCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_VCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_VCLK].padding, - pptable->DpmDescriptor[PPCLK_VCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_VCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_VCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_VCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_VCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_VCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_VCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_DCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_DCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_DCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_DCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_DCLK].padding, - pptable->DpmDescriptor[PPCLK_DCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_DCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_DCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_DCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_DCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_DCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_DCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_SOCCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_SOCCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_SOCCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_SOCCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_SOCCLK].padding, - pptable->DpmDescriptor[PPCLK_SOCCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_SOCCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_SOCCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_UCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_UCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_UCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_UCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_UCLK].padding, - pptable->DpmDescriptor[PPCLK_UCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_UCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_UCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_UCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_FCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_FCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_FCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_FCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_FCLK].padding, - pptable->DpmDescriptor[PPCLK_FCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_FCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_FCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_FCLK].Padding16); - - - dev_info(smu->adev->dev, "FreqTableGfx\n"); - for (i = 0; i < NUM_GFXCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = %d\n", i, pptable->FreqTableGfx[i]); - - dev_info(smu->adev->dev, "FreqTableVclk\n"); - for (i = 0; i < NUM_VCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = %d\n", i, pptable->FreqTableVclk[i]); - - dev_info(smu->adev->dev, "FreqTableDclk\n"); - for (i = 0; i < NUM_DCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = %d\n", i, pptable->FreqTableDclk[i]); - - dev_info(smu->adev->dev, "FreqTableSocclk\n"); - for (i = 0; i < NUM_SOCCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = %d\n", i, pptable->FreqTableSocclk[i]); - - dev_info(smu->adev->dev, "FreqTableUclk\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = %d\n", i, pptable->FreqTableUclk[i]); - - dev_info(smu->adev->dev, "FreqTableFclk\n"); - for (i = 0; i < NUM_FCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = %d\n", i, pptable->FreqTableFclk[i]); - - dev_info(smu->adev->dev, "Mp0clkFreq\n"); - for (i = 0; i < NUM_MP0CLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = %d\n", i, pptable->Mp0clkFreq[i]); - - dev_info(smu->adev->dev, "Mp0DpmVoltage\n"); - for (i = 0; i < NUM_MP0CLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = %d\n", i, pptable->Mp0DpmVoltage[i]); - - dev_info(smu->adev->dev, "GfxclkFidle = 0x%x\n", pptable->GfxclkFidle); - dev_info(smu->adev->dev, "GfxclkSlewRate = 0x%x\n", pptable->GfxclkSlewRate); - dev_info(smu->adev->dev, "Padding567[0] = 0x%x\n", pptable->Padding567[0]); - dev_info(smu->adev->dev, "Padding567[1] = 0x%x\n", pptable->Padding567[1]); - dev_info(smu->adev->dev, "Padding567[2] = 0x%x\n", pptable->Padding567[2]); - dev_info(smu->adev->dev, "Padding567[3] = 0x%x\n", pptable->Padding567[3]); - dev_info(smu->adev->dev, "GfxclkDsMaxFreq = %d\n", pptable->GfxclkDsMaxFreq); - dev_info(smu->adev->dev, "GfxclkSource = 0x%x\n", pptable->GfxclkSource); - dev_info(smu->adev->dev, "Padding456 = 0x%x\n", pptable->Padding456); - - dev_info(smu->adev->dev, "EnableTdpm = %d\n", pptable->EnableTdpm); - dev_info(smu->adev->dev, "TdpmHighHystTemperature = %d\n", pptable->TdpmHighHystTemperature); - dev_info(smu->adev->dev, "TdpmLowHystTemperature = %d\n", pptable->TdpmLowHystTemperature); - dev_info(smu->adev->dev, "GfxclkFreqHighTempLimit = %d\n", pptable->GfxclkFreqHighTempLimit); - - dev_info(smu->adev->dev, "FanStopTemp = %d\n", pptable->FanStopTemp); - dev_info(smu->adev->dev, "FanStartTemp = %d\n", pptable->FanStartTemp); - - dev_info(smu->adev->dev, "FanGainEdge = %d\n", pptable->FanGainEdge); - dev_info(smu->adev->dev, "FanGainHotspot = %d\n", pptable->FanGainHotspot); - dev_info(smu->adev->dev, "FanGainVrGfx = %d\n", pptable->FanGainVrGfx); - dev_info(smu->adev->dev, "FanGainVrSoc = %d\n", pptable->FanGainVrSoc); - dev_info(smu->adev->dev, "FanGainVrMem = %d\n", pptable->FanGainVrMem); - dev_info(smu->adev->dev, "FanGainHbm = %d\n", pptable->FanGainHbm); - - dev_info(smu->adev->dev, "FanPwmMin = %d\n", pptable->FanPwmMin); - dev_info(smu->adev->dev, "FanAcousticLimitRpm = %d\n", pptable->FanAcousticLimitRpm); - dev_info(smu->adev->dev, "FanThrottlingRpm = %d\n", pptable->FanThrottlingRpm); - dev_info(smu->adev->dev, "FanMaximumRpm = %d\n", pptable->FanMaximumRpm); - dev_info(smu->adev->dev, "FanTargetTemperature = %d\n", pptable->FanTargetTemperature); - dev_info(smu->adev->dev, "FanTargetGfxclk = %d\n", pptable->FanTargetGfxclk); - dev_info(smu->adev->dev, "FanZeroRpmEnable = %d\n", pptable->FanZeroRpmEnable); - dev_info(smu->adev->dev, "FanTachEdgePerRev = %d\n", pptable->FanTachEdgePerRev); - dev_info(smu->adev->dev, "FanTempInputSelect = %d\n", pptable->FanTempInputSelect); - - dev_info(smu->adev->dev, "FuzzyFan_ErrorSetDelta = %d\n", pptable->FuzzyFan_ErrorSetDelta); - dev_info(smu->adev->dev, "FuzzyFan_ErrorRateSetDelta = %d\n", pptable->FuzzyFan_ErrorRateSetDelta); - dev_info(smu->adev->dev, "FuzzyFan_PwmSetDelta = %d\n", pptable->FuzzyFan_PwmSetDelta); - dev_info(smu->adev->dev, "FuzzyFan_Reserved = %d\n", pptable->FuzzyFan_Reserved); - - dev_info(smu->adev->dev, "OverrideAvfsGb[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->OverrideAvfsGb[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "OverrideAvfsGb[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->OverrideAvfsGb[AVFS_VOLTAGE_SOC]); - dev_info(smu->adev->dev, "Padding8_Avfs[0] = %d\n", pptable->Padding8_Avfs[0]); - dev_info(smu->adev->dev, "Padding8_Avfs[1] = %d\n", pptable->Padding8_Avfs[1]); - - dev_info(smu->adev->dev, "dBtcGbGfxPll{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbGfxPll.a, - pptable->dBtcGbGfxPll.b, - pptable->dBtcGbGfxPll.c); - dev_info(smu->adev->dev, "dBtcGbGfxAfll{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbGfxAfll.a, - pptable->dBtcGbGfxAfll.b, - pptable->dBtcGbGfxAfll.c); - dev_info(smu->adev->dev, "dBtcGbSoc{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbSoc.a, - pptable->dBtcGbSoc.b, - pptable->dBtcGbSoc.c); - - dev_info(smu->adev->dev, "qAgingGb[AVFS_VOLTAGE_GFX]{m = 0x%x b = 0x%x}\n", - pptable->qAgingGb[AVFS_VOLTAGE_GFX].m, - pptable->qAgingGb[AVFS_VOLTAGE_GFX].b); - dev_info(smu->adev->dev, "qAgingGb[AVFS_VOLTAGE_SOC]{m = 0x%x b = 0x%x}\n", - pptable->qAgingGb[AVFS_VOLTAGE_SOC].m, - pptable->qAgingGb[AVFS_VOLTAGE_SOC].b); - - dev_info(smu->adev->dev, "qStaticVoltageOffset[AVFS_VOLTAGE_GFX]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].a, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].b, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].c); - dev_info(smu->adev->dev, "qStaticVoltageOffset[AVFS_VOLTAGE_SOC]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].a, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].b, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].c); - - dev_info(smu->adev->dev, "DcTol[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcTol[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcTol[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcTol[AVFS_VOLTAGE_SOC]); - - dev_info(smu->adev->dev, "DcBtcEnabled[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcEnabled[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcEnabled[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcEnabled[AVFS_VOLTAGE_SOC]); - dev_info(smu->adev->dev, "Padding8_GfxBtc[0] = 0x%x\n", pptable->Padding8_GfxBtc[0]); - dev_info(smu->adev->dev, "Padding8_GfxBtc[1] = 0x%x\n", pptable->Padding8_GfxBtc[1]); - - dev_info(smu->adev->dev, "DcBtcMin[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcMin[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcMin[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcMin[AVFS_VOLTAGE_SOC]); - dev_info(smu->adev->dev, "DcBtcMax[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcMax[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcMax[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcMax[AVFS_VOLTAGE_SOC]); - - dev_info(smu->adev->dev, "DcBtcGb[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcGb[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcGb[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcGb[AVFS_VOLTAGE_SOC]); - - dev_info(smu->adev->dev, "XgmiDpmPstates\n"); - for (i = 0; i < NUM_XGMI_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = %d\n", i, pptable->XgmiDpmPstates[i]); - dev_info(smu->adev->dev, "XgmiDpmSpare[0] = 0x%02x\n", pptable->XgmiDpmSpare[0]); - dev_info(smu->adev->dev, "XgmiDpmSpare[1] = 0x%02x\n", pptable->XgmiDpmSpare[1]); - - dev_info(smu->adev->dev, "VDDGFX_TVmin = %d\n", pptable->VDDGFX_TVmin); - dev_info(smu->adev->dev, "VDDSOC_TVmin = %d\n", pptable->VDDSOC_TVmin); - dev_info(smu->adev->dev, "VDDGFX_Vmin_HiTemp = %d\n", pptable->VDDGFX_Vmin_HiTemp); - dev_info(smu->adev->dev, "VDDGFX_Vmin_LoTemp = %d\n", pptable->VDDGFX_Vmin_LoTemp); - dev_info(smu->adev->dev, "VDDSOC_Vmin_HiTemp = %d\n", pptable->VDDSOC_Vmin_HiTemp); - dev_info(smu->adev->dev, "VDDSOC_Vmin_LoTemp = %d\n", pptable->VDDSOC_Vmin_LoTemp); - dev_info(smu->adev->dev, "VDDGFX_TVminHystersis = %d\n", pptable->VDDGFX_TVminHystersis); - dev_info(smu->adev->dev, "VDDSOC_TVminHystersis = %d\n", pptable->VDDSOC_TVminHystersis); - - dev_info(smu->adev->dev, "DebugOverrides = 0x%x\n", pptable->DebugOverrides); - dev_info(smu->adev->dev, "ReservedEquation0{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation0.a, - pptable->ReservedEquation0.b, - pptable->ReservedEquation0.c); - dev_info(smu->adev->dev, "ReservedEquation1{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation1.a, - pptable->ReservedEquation1.b, - pptable->ReservedEquation1.c); - dev_info(smu->adev->dev, "ReservedEquation2{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation2.a, - pptable->ReservedEquation2.b, - pptable->ReservedEquation2.c); - dev_info(smu->adev->dev, "ReservedEquation3{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation3.a, - pptable->ReservedEquation3.b, - pptable->ReservedEquation3.c); - - dev_info(smu->adev->dev, "MinVoltageUlvGfx = %d\n", pptable->MinVoltageUlvGfx); - dev_info(smu->adev->dev, "PaddingUlv = %d\n", pptable->PaddingUlv); - - dev_info(smu->adev->dev, "TotalPowerConfig = %d\n", pptable->TotalPowerConfig); - dev_info(smu->adev->dev, "TotalPowerSpare1 = %d\n", pptable->TotalPowerSpare1); - dev_info(smu->adev->dev, "TotalPowerSpare2 = %d\n", pptable->TotalPowerSpare2); - - dev_info(smu->adev->dev, "PccThresholdLow = %d\n", pptable->PccThresholdLow); - dev_info(smu->adev->dev, "PccThresholdHigh = %d\n", pptable->PccThresholdHigh); - - dev_info(smu->adev->dev, "Board Parameters:\n"); - dev_info(smu->adev->dev, "MaxVoltageStepGfx = 0x%x\n", pptable->MaxVoltageStepGfx); - dev_info(smu->adev->dev, "MaxVoltageStepSoc = 0x%x\n", pptable->MaxVoltageStepSoc); - - dev_info(smu->adev->dev, "VddGfxVrMapping = 0x%x\n", pptable->VddGfxVrMapping); - dev_info(smu->adev->dev, "VddSocVrMapping = 0x%x\n", pptable->VddSocVrMapping); - dev_info(smu->adev->dev, "VddMemVrMapping = 0x%x\n", pptable->VddMemVrMapping); - dev_info(smu->adev->dev, "BoardVrMapping = 0x%x\n", pptable->BoardVrMapping); - - dev_info(smu->adev->dev, "GfxUlvPhaseSheddingMask = 0x%x\n", pptable->GfxUlvPhaseSheddingMask); - dev_info(smu->adev->dev, "ExternalSensorPresent = 0x%x\n", pptable->ExternalSensorPresent); - - dev_info(smu->adev->dev, "GfxMaxCurrent = 0x%x\n", pptable->GfxMaxCurrent); - dev_info(smu->adev->dev, "GfxOffset = 0x%x\n", pptable->GfxOffset); - dev_info(smu->adev->dev, "Padding_TelemetryGfx = 0x%x\n", pptable->Padding_TelemetryGfx); - - dev_info(smu->adev->dev, "SocMaxCurrent = 0x%x\n", pptable->SocMaxCurrent); - dev_info(smu->adev->dev, "SocOffset = 0x%x\n", pptable->SocOffset); - dev_info(smu->adev->dev, "Padding_TelemetrySoc = 0x%x\n", pptable->Padding_TelemetrySoc); - - dev_info(smu->adev->dev, "MemMaxCurrent = 0x%x\n", pptable->MemMaxCurrent); - dev_info(smu->adev->dev, "MemOffset = 0x%x\n", pptable->MemOffset); - dev_info(smu->adev->dev, "Padding_TelemetryMem = 0x%x\n", pptable->Padding_TelemetryMem); - - dev_info(smu->adev->dev, "BoardMaxCurrent = 0x%x\n", pptable->BoardMaxCurrent); - dev_info(smu->adev->dev, "BoardOffset = 0x%x\n", pptable->BoardOffset); - dev_info(smu->adev->dev, "Padding_TelemetryBoardInput = 0x%x\n", pptable->Padding_TelemetryBoardInput); - - dev_info(smu->adev->dev, "VR0HotGpio = %d\n", pptable->VR0HotGpio); - dev_info(smu->adev->dev, "VR0HotPolarity = %d\n", pptable->VR0HotPolarity); - dev_info(smu->adev->dev, "VR1HotGpio = %d\n", pptable->VR1HotGpio); - dev_info(smu->adev->dev, "VR1HotPolarity = %d\n", pptable->VR1HotPolarity); - - dev_info(smu->adev->dev, "PllGfxclkSpreadEnabled = %d\n", pptable->PllGfxclkSpreadEnabled); - dev_info(smu->adev->dev, "PllGfxclkSpreadPercent = %d\n", pptable->PllGfxclkSpreadPercent); - dev_info(smu->adev->dev, "PllGfxclkSpreadFreq = %d\n", pptable->PllGfxclkSpreadFreq); - - dev_info(smu->adev->dev, "UclkSpreadEnabled = %d\n", pptable->UclkSpreadEnabled); - dev_info(smu->adev->dev, "UclkSpreadPercent = %d\n", pptable->UclkSpreadPercent); - dev_info(smu->adev->dev, "UclkSpreadFreq = %d\n", pptable->UclkSpreadFreq); - - dev_info(smu->adev->dev, "FclkSpreadEnabled = %d\n", pptable->FclkSpreadEnabled); - dev_info(smu->adev->dev, "FclkSpreadPercent = %d\n", pptable->FclkSpreadPercent); - dev_info(smu->adev->dev, "FclkSpreadFreq = %d\n", pptable->FclkSpreadFreq); - - dev_info(smu->adev->dev, "FllGfxclkSpreadEnabled = %d\n", pptable->FllGfxclkSpreadEnabled); - dev_info(smu->adev->dev, "FllGfxclkSpreadPercent = %d\n", pptable->FllGfxclkSpreadPercent); - dev_info(smu->adev->dev, "FllGfxclkSpreadFreq = %d\n", pptable->FllGfxclkSpreadFreq); - - for (i = 0; i < NUM_I2C_CONTROLLERS; i++) { - dev_info(smu->adev->dev, "I2cControllers[%d]:\n", i); - dev_info(smu->adev->dev, " .Enabled = %d\n", - pptable->I2cControllers[i].Enabled); - dev_info(smu->adev->dev, " .SlaveAddress = 0x%x\n", - pptable->I2cControllers[i].SlaveAddress); - dev_info(smu->adev->dev, " .ControllerPort = %d\n", - pptable->I2cControllers[i].ControllerPort); - dev_info(smu->adev->dev, " .ControllerName = %d\n", - pptable->I2cControllers[i].ControllerName); - dev_info(smu->adev->dev, " .ThermalThrottler = %d\n", - pptable->I2cControllers[i].ThermalThrotter); - dev_info(smu->adev->dev, " .I2cProtocol = %d\n", - pptable->I2cControllers[i].I2cProtocol); - dev_info(smu->adev->dev, " .Speed = %d\n", - pptable->I2cControllers[i].Speed); - } - - dev_info(smu->adev->dev, "MemoryChannelEnabled = %d\n", pptable->MemoryChannelEnabled); - dev_info(smu->adev->dev, "DramBitWidth = %d\n", pptable->DramBitWidth); - - dev_info(smu->adev->dev, "TotalBoardPower = %d\n", pptable->TotalBoardPower); - - dev_info(smu->adev->dev, "XgmiLinkSpeed\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = %d\n", i, pptable->XgmiLinkSpeed[i]); - dev_info(smu->adev->dev, "XgmiLinkWidth\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = %d\n", i, pptable->XgmiLinkWidth[i]); - dev_info(smu->adev->dev, "XgmiFclkFreq\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = %d\n", i, pptable->XgmiFclkFreq[i]); - dev_info(smu->adev->dev, "XgmiSocVoltage\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = %d\n", i, pptable->XgmiSocVoltage[i]); - -} - static bool arcturus_is_dpm_running(struct smu_context *smu) { int ret = 0; @@ -2365,8 +1934,6 @@ static const struct pptable_funcs arcturus_ppt_funcs = { .get_power_profile_mode = arcturus_get_power_profile_mode, .set_power_profile_mode = arcturus_set_power_profile_mode, .set_performance_level = arcturus_set_performance_level, - /* debug (internal used) */ - .dump_pptable = arcturus_dump_pptable, .get_power_limit = arcturus_get_power_limit, .is_dpm_running = arcturus_is_dpm_running, .dpm_set_vcn_enable = arcturus_dpm_set_vcn_enable, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 16af1a329621..9fa305ba6422 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -1689,7 +1689,7 @@ static int navi10_force_clk_levels(struct smu_context *smu, if (ret) return 0; - ret = smu_v11_0_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq); + ret = smu_v11_0_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq, false); if (ret) return 0; break; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c index 9c3c48297cba..77e58eb46328 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c @@ -1469,7 +1469,7 @@ static int sienna_cichlid_force_clk_levels(struct smu_context *smu, if (ret) goto forec_level_out; - ret = smu_v11_0_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq); + ret = smu_v11_0_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq, false); if (ret) goto forec_level_out; break; @@ -2493,1274 +2493,6 @@ static bool sienna_cichlid_is_mode1_reset_supported(struct smu_context *smu) return val != 0x0; } -static void beige_goby_dump_pptable(struct smu_context *smu) -{ - struct smu_table_context *table_context = &smu->smu_table; - PPTable_beige_goby_t *pptable = table_context->driver_pptable; - int i; - - dev_info(smu->adev->dev, "Dumped PPTable:\n"); - - dev_info(smu->adev->dev, "Version = 0x%08x\n", pptable->Version); - dev_info(smu->adev->dev, "FeaturesToRun[0] = 0x%08x\n", pptable->FeaturesToRun[0]); - dev_info(smu->adev->dev, "FeaturesToRun[1] = 0x%08x\n", pptable->FeaturesToRun[1]); - - for (i = 0; i < PPT_THROTTLER_COUNT; i++) { - dev_info(smu->adev->dev, "SocketPowerLimitAc[%d] = 0x%x\n", i, pptable->SocketPowerLimitAc[i]); - dev_info(smu->adev->dev, "SocketPowerLimitAcTau[%d] = 0x%x\n", i, pptable->SocketPowerLimitAcTau[i]); - dev_info(smu->adev->dev, "SocketPowerLimitDc[%d] = 0x%x\n", i, pptable->SocketPowerLimitDc[i]); - dev_info(smu->adev->dev, "SocketPowerLimitDcTau[%d] = 0x%x\n", i, pptable->SocketPowerLimitDcTau[i]); - } - - for (i = 0; i < TDC_THROTTLER_COUNT; i++) { - dev_info(smu->adev->dev, "TdcLimit[%d] = 0x%x\n", i, pptable->TdcLimit[i]); - dev_info(smu->adev->dev, "TdcLimitTau[%d] = 0x%x\n", i, pptable->TdcLimitTau[i]); - } - - for (i = 0; i < TEMP_COUNT; i++) { - dev_info(smu->adev->dev, "TemperatureLimit[%d] = 0x%x\n", i, pptable->TemperatureLimit[i]); - } - - dev_info(smu->adev->dev, "FitLimit = 0x%x\n", pptable->FitLimit); - dev_info(smu->adev->dev, "TotalPowerConfig = 0x%x\n", pptable->TotalPowerConfig); - dev_info(smu->adev->dev, "TotalPowerPadding[0] = 0x%x\n", pptable->TotalPowerPadding[0]); - dev_info(smu->adev->dev, "TotalPowerPadding[1] = 0x%x\n", pptable->TotalPowerPadding[1]); - dev_info(smu->adev->dev, "TotalPowerPadding[2] = 0x%x\n", pptable->TotalPowerPadding[2]); - - dev_info(smu->adev->dev, "ApccPlusResidencyLimit = 0x%x\n", pptable->ApccPlusResidencyLimit); - for (i = 0; i < NUM_SMNCLK_DPM_LEVELS; i++) { - dev_info(smu->adev->dev, "SmnclkDpmFreq[%d] = 0x%x\n", i, pptable->SmnclkDpmFreq[i]); - dev_info(smu->adev->dev, "SmnclkDpmVoltage[%d] = 0x%x\n", i, pptable->SmnclkDpmVoltage[i]); - } - dev_info(smu->adev->dev, "ThrottlerControlMask = 0x%x\n", pptable->ThrottlerControlMask); - - dev_info(smu->adev->dev, "FwDStateMask = 0x%x\n", pptable->FwDStateMask); - - dev_info(smu->adev->dev, "UlvVoltageOffsetSoc = 0x%x\n", pptable->UlvVoltageOffsetSoc); - dev_info(smu->adev->dev, "UlvVoltageOffsetGfx = 0x%x\n", pptable->UlvVoltageOffsetGfx); - dev_info(smu->adev->dev, "MinVoltageUlvGfx = 0x%x\n", pptable->MinVoltageUlvGfx); - dev_info(smu->adev->dev, "MinVoltageUlvSoc = 0x%x\n", pptable->MinVoltageUlvSoc); - - dev_info(smu->adev->dev, "SocLIVmin = 0x%x\n", pptable->SocLIVmin); - - dev_info(smu->adev->dev, "GceaLinkMgrIdleThreshold = 0x%x\n", pptable->GceaLinkMgrIdleThreshold); - - dev_info(smu->adev->dev, "MinVoltageGfx = 0x%x\n", pptable->MinVoltageGfx); - dev_info(smu->adev->dev, "MinVoltageSoc = 0x%x\n", pptable->MinVoltageSoc); - dev_info(smu->adev->dev, "MaxVoltageGfx = 0x%x\n", pptable->MaxVoltageGfx); - dev_info(smu->adev->dev, "MaxVoltageSoc = 0x%x\n", pptable->MaxVoltageSoc); - - dev_info(smu->adev->dev, "LoadLineResistanceGfx = 0x%x\n", pptable->LoadLineResistanceGfx); - dev_info(smu->adev->dev, "LoadLineResistanceSoc = 0x%x\n", pptable->LoadLineResistanceSoc); - - dev_info(smu->adev->dev, "VDDGFX_TVmin = 0x%x\n", pptable->VDDGFX_TVmin); - dev_info(smu->adev->dev, "VDDSOC_TVmin = 0x%x\n", pptable->VDDSOC_TVmin); - dev_info(smu->adev->dev, "VDDGFX_Vmin_HiTemp = 0x%x\n", pptable->VDDGFX_Vmin_HiTemp); - dev_info(smu->adev->dev, "VDDGFX_Vmin_LoTemp = 0x%x\n", pptable->VDDGFX_Vmin_LoTemp); - dev_info(smu->adev->dev, "VDDSOC_Vmin_HiTemp = 0x%x\n", pptable->VDDSOC_Vmin_HiTemp); - dev_info(smu->adev->dev, "VDDSOC_Vmin_LoTemp = 0x%x\n", pptable->VDDSOC_Vmin_LoTemp); - dev_info(smu->adev->dev, "VDDGFX_TVminHystersis = 0x%x\n", pptable->VDDGFX_TVminHystersis); - dev_info(smu->adev->dev, "VDDSOC_TVminHystersis = 0x%x\n", pptable->VDDSOC_TVminHystersis); - - dev_info(smu->adev->dev, "[PPCLK_GFXCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_GFXCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_GFXCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_GFXCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_GFXCLK].Padding, - pptable->DpmDescriptor[PPCLK_GFXCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_GFXCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_GFXCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_SOCCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_SOCCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_SOCCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_SOCCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_SOCCLK].Padding, - pptable->DpmDescriptor[PPCLK_SOCCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_SOCCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_SOCCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_UCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_UCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_UCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_UCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_UCLK].Padding, - pptable->DpmDescriptor[PPCLK_UCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_UCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_UCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_UCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_FCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_FCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_FCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_FCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_FCLK].Padding, - pptable->DpmDescriptor[PPCLK_FCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_FCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_FCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_FCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_DCLK_0]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_DCLK_0].VoltageMode, - pptable->DpmDescriptor[PPCLK_DCLK_0].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_DCLK_0].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_DCLK_0].Padding, - pptable->DpmDescriptor[PPCLK_DCLK_0].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_DCLK_0].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_DCLK_0].SsCurve.a, - pptable->DpmDescriptor[PPCLK_DCLK_0].SsCurve.b, - pptable->DpmDescriptor[PPCLK_DCLK_0].SsCurve.c, - pptable->DpmDescriptor[PPCLK_DCLK_0].SsFmin, - pptable->DpmDescriptor[PPCLK_DCLK_0].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_VCLK_0]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_VCLK_0].VoltageMode, - pptable->DpmDescriptor[PPCLK_VCLK_0].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_VCLK_0].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_VCLK_0].Padding, - pptable->DpmDescriptor[PPCLK_VCLK_0].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_VCLK_0].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_VCLK_0].SsCurve.a, - pptable->DpmDescriptor[PPCLK_VCLK_0].SsCurve.b, - pptable->DpmDescriptor[PPCLK_VCLK_0].SsCurve.c, - pptable->DpmDescriptor[PPCLK_VCLK_0].SsFmin, - pptable->DpmDescriptor[PPCLK_VCLK_0].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_DCLK_1]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_DCLK_1].VoltageMode, - pptable->DpmDescriptor[PPCLK_DCLK_1].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_DCLK_1].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_DCLK_1].Padding, - pptable->DpmDescriptor[PPCLK_DCLK_1].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_DCLK_1].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_DCLK_1].SsCurve.a, - pptable->DpmDescriptor[PPCLK_DCLK_1].SsCurve.b, - pptable->DpmDescriptor[PPCLK_DCLK_1].SsCurve.c, - pptable->DpmDescriptor[PPCLK_DCLK_1].SsFmin, - pptable->DpmDescriptor[PPCLK_DCLK_1].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_VCLK_1]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_VCLK_1].VoltageMode, - pptable->DpmDescriptor[PPCLK_VCLK_1].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_VCLK_1].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_VCLK_1].Padding, - pptable->DpmDescriptor[PPCLK_VCLK_1].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_VCLK_1].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_VCLK_1].SsCurve.a, - pptable->DpmDescriptor[PPCLK_VCLK_1].SsCurve.b, - pptable->DpmDescriptor[PPCLK_VCLK_1].SsCurve.c, - pptable->DpmDescriptor[PPCLK_VCLK_1].SsFmin, - pptable->DpmDescriptor[PPCLK_VCLK_1].Padding16); - - dev_info(smu->adev->dev, "FreqTableGfx\n"); - for (i = 0; i < NUM_GFXCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableGfx[i]); - - dev_info(smu->adev->dev, "FreqTableVclk\n"); - for (i = 0; i < NUM_VCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableVclk[i]); - - dev_info(smu->adev->dev, "FreqTableDclk\n"); - for (i = 0; i < NUM_DCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableDclk[i]); - - dev_info(smu->adev->dev, "FreqTableSocclk\n"); - for (i = 0; i < NUM_SOCCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableSocclk[i]); - - dev_info(smu->adev->dev, "FreqTableUclk\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableUclk[i]); - - dev_info(smu->adev->dev, "FreqTableFclk\n"); - for (i = 0; i < NUM_FCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableFclk[i]); - - dev_info(smu->adev->dev, "DcModeMaxFreq\n"); - dev_info(smu->adev->dev, " .PPCLK_GFXCLK = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_GFXCLK]); - dev_info(smu->adev->dev, " .PPCLK_SOCCLK = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_SOCCLK]); - dev_info(smu->adev->dev, " .PPCLK_UCLK = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_UCLK]); - dev_info(smu->adev->dev, " .PPCLK_FCLK = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_FCLK]); - dev_info(smu->adev->dev, " .PPCLK_DCLK_0 = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_DCLK_0]); - dev_info(smu->adev->dev, " .PPCLK_VCLK_0 = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_VCLK_0]); - dev_info(smu->adev->dev, " .PPCLK_DCLK_1 = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_DCLK_1]); - dev_info(smu->adev->dev, " .PPCLK_VCLK_1 = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_VCLK_1]); - - dev_info(smu->adev->dev, "FreqTableUclkDiv\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->FreqTableUclkDiv[i]); - - dev_info(smu->adev->dev, "FclkBoostFreq = 0x%x\n", pptable->FclkBoostFreq); - dev_info(smu->adev->dev, "FclkParamPadding = 0x%x\n", pptable->FclkParamPadding); - - dev_info(smu->adev->dev, "Mp0clkFreq\n"); - for (i = 0; i < NUM_MP0CLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->Mp0clkFreq[i]); - - dev_info(smu->adev->dev, "Mp0DpmVoltage\n"); - for (i = 0; i < NUM_MP0CLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->Mp0DpmVoltage[i]); - - dev_info(smu->adev->dev, "MemVddciVoltage\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->MemVddciVoltage[i]); - - dev_info(smu->adev->dev, "MemMvddVoltage\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->MemMvddVoltage[i]); - - dev_info(smu->adev->dev, "GfxclkFgfxoffEntry = 0x%x\n", pptable->GfxclkFgfxoffEntry); - dev_info(smu->adev->dev, "GfxclkFinit = 0x%x\n", pptable->GfxclkFinit); - dev_info(smu->adev->dev, "GfxclkFidle = 0x%x\n", pptable->GfxclkFidle); - dev_info(smu->adev->dev, "GfxclkSource = 0x%x\n", pptable->GfxclkSource); - dev_info(smu->adev->dev, "GfxclkPadding = 0x%x\n", pptable->GfxclkPadding); - - dev_info(smu->adev->dev, "GfxGpoSubFeatureMask = 0x%x\n", pptable->GfxGpoSubFeatureMask); - - dev_info(smu->adev->dev, "GfxGpoEnabledWorkPolicyMask = 0x%x\n", pptable->GfxGpoEnabledWorkPolicyMask); - dev_info(smu->adev->dev, "GfxGpoDisabledWorkPolicyMask = 0x%x\n", pptable->GfxGpoDisabledWorkPolicyMask); - dev_info(smu->adev->dev, "GfxGpoPadding[0] = 0x%x\n", pptable->GfxGpoPadding[0]); - dev_info(smu->adev->dev, "GfxGpoVotingAllow = 0x%x\n", pptable->GfxGpoVotingAllow); - dev_info(smu->adev->dev, "GfxGpoPadding32[0] = 0x%x\n", pptable->GfxGpoPadding32[0]); - dev_info(smu->adev->dev, "GfxGpoPadding32[1] = 0x%x\n", pptable->GfxGpoPadding32[1]); - dev_info(smu->adev->dev, "GfxGpoPadding32[2] = 0x%x\n", pptable->GfxGpoPadding32[2]); - dev_info(smu->adev->dev, "GfxGpoPadding32[3] = 0x%x\n", pptable->GfxGpoPadding32[3]); - dev_info(smu->adev->dev, "GfxDcsFopt = 0x%x\n", pptable->GfxDcsFopt); - dev_info(smu->adev->dev, "GfxDcsFclkFopt = 0x%x\n", pptable->GfxDcsFclkFopt); - dev_info(smu->adev->dev, "GfxDcsUclkFopt = 0x%x\n", pptable->GfxDcsUclkFopt); - - dev_info(smu->adev->dev, "DcsGfxOffVoltage = 0x%x\n", pptable->DcsGfxOffVoltage); - dev_info(smu->adev->dev, "DcsMinGfxOffTime = 0x%x\n", pptable->DcsMinGfxOffTime); - dev_info(smu->adev->dev, "DcsMaxGfxOffTime = 0x%x\n", pptable->DcsMaxGfxOffTime); - dev_info(smu->adev->dev, "DcsMinCreditAccum = 0x%x\n", pptable->DcsMinCreditAccum); - dev_info(smu->adev->dev, "DcsExitHysteresis = 0x%x\n", pptable->DcsExitHysteresis); - dev_info(smu->adev->dev, "DcsTimeout = 0x%x\n", pptable->DcsTimeout); - - dev_info(smu->adev->dev, "DcsParamPadding[0] = 0x%x\n", pptable->DcsParamPadding[0]); - dev_info(smu->adev->dev, "DcsParamPadding[1] = 0x%x\n", pptable->DcsParamPadding[1]); - dev_info(smu->adev->dev, "DcsParamPadding[2] = 0x%x\n", pptable->DcsParamPadding[2]); - dev_info(smu->adev->dev, "DcsParamPadding[3] = 0x%x\n", pptable->DcsParamPadding[3]); - dev_info(smu->adev->dev, "DcsParamPadding[4] = 0x%x\n", pptable->DcsParamPadding[4]); - - dev_info(smu->adev->dev, "FlopsPerByteTable\n"); - for (i = 0; i < RLC_PACE_TABLE_NUM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->FlopsPerByteTable[i]); - - dev_info(smu->adev->dev, "LowestUclkReservedForUlv = 0x%x\n", pptable->LowestUclkReservedForUlv); - dev_info(smu->adev->dev, "vddingMem[0] = 0x%x\n", pptable->PaddingMem[0]); - dev_info(smu->adev->dev, "vddingMem[1] = 0x%x\n", pptable->PaddingMem[1]); - dev_info(smu->adev->dev, "vddingMem[2] = 0x%x\n", pptable->PaddingMem[2]); - - dev_info(smu->adev->dev, "UclkDpmPstates\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->UclkDpmPstates[i]); - - dev_info(smu->adev->dev, "UclkDpmSrcFreqRange\n"); - dev_info(smu->adev->dev, " .Fmin = 0x%x\n", - pptable->UclkDpmSrcFreqRange.Fmin); - dev_info(smu->adev->dev, " .Fmax = 0x%x\n", - pptable->UclkDpmSrcFreqRange.Fmax); - dev_info(smu->adev->dev, "UclkDpmTargFreqRange\n"); - dev_info(smu->adev->dev, " .Fmin = 0x%x\n", - pptable->UclkDpmTargFreqRange.Fmin); - dev_info(smu->adev->dev, " .Fmax = 0x%x\n", - pptable->UclkDpmTargFreqRange.Fmax); - dev_info(smu->adev->dev, "UclkDpmMidstepFreq = 0x%x\n", pptable->UclkDpmMidstepFreq); - dev_info(smu->adev->dev, "UclkMidstepPadding = 0x%x\n", pptable->UclkMidstepPadding); - - dev_info(smu->adev->dev, "PcieGenSpeed\n"); - for (i = 0; i < NUM_LINK_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->PcieGenSpeed[i]); - - dev_info(smu->adev->dev, "PcieLaneCount\n"); - for (i = 0; i < NUM_LINK_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->PcieLaneCount[i]); - - dev_info(smu->adev->dev, "LclkFreq\n"); - for (i = 0; i < NUM_LINK_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->LclkFreq[i]); - - dev_info(smu->adev->dev, "FanStopTemp = 0x%x\n", pptable->FanStopTemp); - dev_info(smu->adev->dev, "FanStartTemp = 0x%x\n", pptable->FanStartTemp); - - dev_info(smu->adev->dev, "FanGain\n"); - for (i = 0; i < TEMP_COUNT; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->FanGain[i]); - - dev_info(smu->adev->dev, "FanPwmMin = 0x%x\n", pptable->FanPwmMin); - dev_info(smu->adev->dev, "FanAcousticLimitRpm = 0x%x\n", pptable->FanAcousticLimitRpm); - dev_info(smu->adev->dev, "FanThrottlingRpm = 0x%x\n", pptable->FanThrottlingRpm); - dev_info(smu->adev->dev, "FanMaximumRpm = 0x%x\n", pptable->FanMaximumRpm); - dev_info(smu->adev->dev, "MGpuFanBoostLimitRpm = 0x%x\n", pptable->MGpuFanBoostLimitRpm); - dev_info(smu->adev->dev, "FanTargetTemperature = 0x%x\n", pptable->FanTargetTemperature); - dev_info(smu->adev->dev, "FanTargetGfxclk = 0x%x\n", pptable->FanTargetGfxclk); - dev_info(smu->adev->dev, "FanPadding16 = 0x%x\n", pptable->FanPadding16); - dev_info(smu->adev->dev, "FanTempInputSelect = 0x%x\n", pptable->FanTempInputSelect); - dev_info(smu->adev->dev, "FanPadding = 0x%x\n", pptable->FanPadding); - dev_info(smu->adev->dev, "FanZeroRpmEnable = 0x%x\n", pptable->FanZeroRpmEnable); - dev_info(smu->adev->dev, "FanTachEdgePerRev = 0x%x\n", pptable->FanTachEdgePerRev); - - dev_info(smu->adev->dev, "FuzzyFan_ErrorSetDelta = 0x%x\n", pptable->FuzzyFan_ErrorSetDelta); - dev_info(smu->adev->dev, "FuzzyFan_ErrorRateSetDelta = 0x%x\n", pptable->FuzzyFan_ErrorRateSetDelta); - dev_info(smu->adev->dev, "FuzzyFan_PwmSetDelta = 0x%x\n", pptable->FuzzyFan_PwmSetDelta); - dev_info(smu->adev->dev, "FuzzyFan_Reserved = 0x%x\n", pptable->FuzzyFan_Reserved); - - dev_info(smu->adev->dev, "OverrideAvfsGb[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->OverrideAvfsGb[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "OverrideAvfsGb[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->OverrideAvfsGb[AVFS_VOLTAGE_SOC]); - dev_info(smu->adev->dev, "dBtcGbGfxDfllModelSelect = 0x%x\n", pptable->dBtcGbGfxDfllModelSelect); - dev_info(smu->adev->dev, "Padding8_Avfs = 0x%x\n", pptable->Padding8_Avfs); - - dev_info(smu->adev->dev, "qAvfsGb[AVFS_VOLTAGE_GFX]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qAvfsGb[AVFS_VOLTAGE_GFX].a, - pptable->qAvfsGb[AVFS_VOLTAGE_GFX].b, - pptable->qAvfsGb[AVFS_VOLTAGE_GFX].c); - dev_info(smu->adev->dev, "qAvfsGb[AVFS_VOLTAGE_SOC]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qAvfsGb[AVFS_VOLTAGE_SOC].a, - pptable->qAvfsGb[AVFS_VOLTAGE_SOC].b, - pptable->qAvfsGb[AVFS_VOLTAGE_SOC].c); - dev_info(smu->adev->dev, "dBtcGbGfxPll{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbGfxPll.a, - pptable->dBtcGbGfxPll.b, - pptable->dBtcGbGfxPll.c); - dev_info(smu->adev->dev, "dBtcGbGfxAfll{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbGfxDfll.a, - pptable->dBtcGbGfxDfll.b, - pptable->dBtcGbGfxDfll.c); - dev_info(smu->adev->dev, "dBtcGbSoc{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbSoc.a, - pptable->dBtcGbSoc.b, - pptable->dBtcGbSoc.c); - dev_info(smu->adev->dev, "qAgingGb[AVFS_VOLTAGE_GFX]{m = 0x%x b = 0x%x}\n", - pptable->qAgingGb[AVFS_VOLTAGE_GFX].m, - pptable->qAgingGb[AVFS_VOLTAGE_GFX].b); - dev_info(smu->adev->dev, "qAgingGb[AVFS_VOLTAGE_SOC]{m = 0x%x b = 0x%x}\n", - pptable->qAgingGb[AVFS_VOLTAGE_SOC].m, - pptable->qAgingGb[AVFS_VOLTAGE_SOC].b); - - dev_info(smu->adev->dev, "PiecewiseLinearDroopIntGfxDfll\n"); - for (i = 0; i < NUM_PIECE_WISE_LINEAR_DROOP_MODEL_VF_POINTS; i++) { - dev_info(smu->adev->dev, " Fset[%d] = 0x%x\n", - i, pptable->PiecewiseLinearDroopIntGfxDfll.Fset[i]); - dev_info(smu->adev->dev, " Vdroop[%d] = 0x%x\n", - i, pptable->PiecewiseLinearDroopIntGfxDfll.Vdroop[i]); - } - - dev_info(smu->adev->dev, "qStaticVoltageOffset[AVFS_VOLTAGE_GFX]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].a, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].b, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].c); - dev_info(smu->adev->dev, "qStaticVoltageOffset[AVFS_VOLTAGE_SOC]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].a, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].b, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].c); - - dev_info(smu->adev->dev, "DcTol[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcTol[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcTol[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcTol[AVFS_VOLTAGE_SOC]); - - dev_info(smu->adev->dev, "DcBtcEnabled[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcEnabled[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcEnabled[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcEnabled[AVFS_VOLTAGE_SOC]); - dev_info(smu->adev->dev, "Padding8_GfxBtc[0] = 0x%x\n", pptable->Padding8_GfxBtc[0]); - dev_info(smu->adev->dev, "Padding8_GfxBtc[1] = 0x%x\n", pptable->Padding8_GfxBtc[1]); - - dev_info(smu->adev->dev, "DcBtcMin[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcMin[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcMin[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcMin[AVFS_VOLTAGE_SOC]); - dev_info(smu->adev->dev, "DcBtcMax[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcMax[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcMax[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcMax[AVFS_VOLTAGE_SOC]); - - dev_info(smu->adev->dev, "DcBtcGb[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcGb[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcGb[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcGb[AVFS_VOLTAGE_SOC]); - - dev_info(smu->adev->dev, "XgmiDpmPstates\n"); - for (i = 0; i < NUM_XGMI_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->XgmiDpmPstates[i]); - dev_info(smu->adev->dev, "XgmiDpmSpare[0] = 0x%02x\n", pptable->XgmiDpmSpare[0]); - dev_info(smu->adev->dev, "XgmiDpmSpare[1] = 0x%02x\n", pptable->XgmiDpmSpare[1]); - - dev_info(smu->adev->dev, "DebugOverrides = 0x%x\n", pptable->DebugOverrides); - dev_info(smu->adev->dev, "ReservedEquation0{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation0.a, - pptable->ReservedEquation0.b, - pptable->ReservedEquation0.c); - dev_info(smu->adev->dev, "ReservedEquation1{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation1.a, - pptable->ReservedEquation1.b, - pptable->ReservedEquation1.c); - dev_info(smu->adev->dev, "ReservedEquation2{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation2.a, - pptable->ReservedEquation2.b, - pptable->ReservedEquation2.c); - dev_info(smu->adev->dev, "ReservedEquation3{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation3.a, - pptable->ReservedEquation3.b, - pptable->ReservedEquation3.c); - - dev_info(smu->adev->dev, "SkuReserved[0] = 0x%x\n", pptable->SkuReserved[0]); - dev_info(smu->adev->dev, "SkuReserved[1] = 0x%x\n", pptable->SkuReserved[1]); - dev_info(smu->adev->dev, "SkuReserved[2] = 0x%x\n", pptable->SkuReserved[2]); - dev_info(smu->adev->dev, "SkuReserved[3] = 0x%x\n", pptable->SkuReserved[3]); - dev_info(smu->adev->dev, "SkuReserved[4] = 0x%x\n", pptable->SkuReserved[4]); - dev_info(smu->adev->dev, "SkuReserved[5] = 0x%x\n", pptable->SkuReserved[5]); - dev_info(smu->adev->dev, "SkuReserved[6] = 0x%x\n", pptable->SkuReserved[6]); - dev_info(smu->adev->dev, "SkuReserved[7] = 0x%x\n", pptable->SkuReserved[7]); - - dev_info(smu->adev->dev, "GamingClk[0] = 0x%x\n", pptable->GamingClk[0]); - dev_info(smu->adev->dev, "GamingClk[1] = 0x%x\n", pptable->GamingClk[1]); - dev_info(smu->adev->dev, "GamingClk[2] = 0x%x\n", pptable->GamingClk[2]); - dev_info(smu->adev->dev, "GamingClk[3] = 0x%x\n", pptable->GamingClk[3]); - dev_info(smu->adev->dev, "GamingClk[4] = 0x%x\n", pptable->GamingClk[4]); - dev_info(smu->adev->dev, "GamingClk[5] = 0x%x\n", pptable->GamingClk[5]); - - for (i = 0; i < NUM_I2C_CONTROLLERS; i++) { - dev_info(smu->adev->dev, "I2cControllers[%d]:\n", i); - dev_info(smu->adev->dev, " .Enabled = 0x%x\n", - pptable->I2cControllers[i].Enabled); - dev_info(smu->adev->dev, " .Speed = 0x%x\n", - pptable->I2cControllers[i].Speed); - dev_info(smu->adev->dev, " .SlaveAddress = 0x%x\n", - pptable->I2cControllers[i].SlaveAddress); - dev_info(smu->adev->dev, " .ControllerPort = 0x%x\n", - pptable->I2cControllers[i].ControllerPort); - dev_info(smu->adev->dev, " .ControllerName = 0x%x\n", - pptable->I2cControllers[i].ControllerName); - dev_info(smu->adev->dev, " .ThermalThrottler = 0x%x\n", - pptable->I2cControllers[i].ThermalThrotter); - dev_info(smu->adev->dev, " .I2cProtocol = 0x%x\n", - pptable->I2cControllers[i].I2cProtocol); - dev_info(smu->adev->dev, " .PaddingConfig = 0x%x\n", - pptable->I2cControllers[i].PaddingConfig); - } - - dev_info(smu->adev->dev, "GpioScl = 0x%x\n", pptable->GpioScl); - dev_info(smu->adev->dev, "GpioSda = 0x%x\n", pptable->GpioSda); - dev_info(smu->adev->dev, "FchUsbPdSlaveAddr = 0x%x\n", pptable->FchUsbPdSlaveAddr); - dev_info(smu->adev->dev, "I2cSpare[0] = 0x%x\n", pptable->I2cSpare[0]); - - dev_info(smu->adev->dev, "Board Parameters:\n"); - dev_info(smu->adev->dev, "VddGfxVrMapping = 0x%x\n", pptable->VddGfxVrMapping); - dev_info(smu->adev->dev, "VddSocVrMapping = 0x%x\n", pptable->VddSocVrMapping); - dev_info(smu->adev->dev, "VddMem0VrMapping = 0x%x\n", pptable->VddMem0VrMapping); - dev_info(smu->adev->dev, "VddMem1VrMapping = 0x%x\n", pptable->VddMem1VrMapping); - dev_info(smu->adev->dev, "GfxUlvPhaseSheddingMask = 0x%x\n", pptable->GfxUlvPhaseSheddingMask); - dev_info(smu->adev->dev, "SocUlvPhaseSheddingMask = 0x%x\n", pptable->SocUlvPhaseSheddingMask); - dev_info(smu->adev->dev, "VddciUlvPhaseSheddingMask = 0x%x\n", pptable->VddciUlvPhaseSheddingMask); - dev_info(smu->adev->dev, "MvddUlvPhaseSheddingMask = 0x%x\n", pptable->MvddUlvPhaseSheddingMask); - - dev_info(smu->adev->dev, "GfxMaxCurrent = 0x%x\n", pptable->GfxMaxCurrent); - dev_info(smu->adev->dev, "GfxOffset = 0x%x\n", pptable->GfxOffset); - dev_info(smu->adev->dev, "Padding_TelemetryGfx = 0x%x\n", pptable->Padding_TelemetryGfx); - - dev_info(smu->adev->dev, "SocMaxCurrent = 0x%x\n", pptable->SocMaxCurrent); - dev_info(smu->adev->dev, "SocOffset = 0x%x\n", pptable->SocOffset); - dev_info(smu->adev->dev, "Padding_TelemetrySoc = 0x%x\n", pptable->Padding_TelemetrySoc); - - dev_info(smu->adev->dev, "Mem0MaxCurrent = 0x%x\n", pptable->Mem0MaxCurrent); - dev_info(smu->adev->dev, "Mem0Offset = 0x%x\n", pptable->Mem0Offset); - dev_info(smu->adev->dev, "Padding_TelemetryMem0 = 0x%x\n", pptable->Padding_TelemetryMem0); - - dev_info(smu->adev->dev, "Mem1MaxCurrent = 0x%x\n", pptable->Mem1MaxCurrent); - dev_info(smu->adev->dev, "Mem1Offset = 0x%x\n", pptable->Mem1Offset); - dev_info(smu->adev->dev, "Padding_TelemetryMem1 = 0x%x\n", pptable->Padding_TelemetryMem1); - - dev_info(smu->adev->dev, "MvddRatio = 0x%x\n", pptable->MvddRatio); - - dev_info(smu->adev->dev, "AcDcGpio = 0x%x\n", pptable->AcDcGpio); - dev_info(smu->adev->dev, "AcDcPolarity = 0x%x\n", pptable->AcDcPolarity); - dev_info(smu->adev->dev, "VR0HotGpio = 0x%x\n", pptable->VR0HotGpio); - dev_info(smu->adev->dev, "VR0HotPolarity = 0x%x\n", pptable->VR0HotPolarity); - dev_info(smu->adev->dev, "VR1HotGpio = 0x%x\n", pptable->VR1HotGpio); - dev_info(smu->adev->dev, "VR1HotPolarity = 0x%x\n", pptable->VR1HotPolarity); - dev_info(smu->adev->dev, "GthrGpio = 0x%x\n", pptable->GthrGpio); - dev_info(smu->adev->dev, "GthrPolarity = 0x%x\n", pptable->GthrPolarity); - dev_info(smu->adev->dev, "LedPin0 = 0x%x\n", pptable->LedPin0); - dev_info(smu->adev->dev, "LedPin1 = 0x%x\n", pptable->LedPin1); - dev_info(smu->adev->dev, "LedPin2 = 0x%x\n", pptable->LedPin2); - dev_info(smu->adev->dev, "LedEnableMask = 0x%x\n", pptable->LedEnableMask); - dev_info(smu->adev->dev, "LedPcie = 0x%x\n", pptable->LedPcie); - dev_info(smu->adev->dev, "LedError = 0x%x\n", pptable->LedError); - dev_info(smu->adev->dev, "LedSpare1[0] = 0x%x\n", pptable->LedSpare1[0]); - dev_info(smu->adev->dev, "LedSpare1[1] = 0x%x\n", pptable->LedSpare1[1]); - - dev_info(smu->adev->dev, "PllGfxclkSpreadEnabled = 0x%x\n", pptable->PllGfxclkSpreadEnabled); - dev_info(smu->adev->dev, "PllGfxclkSpreadPercent = 0x%x\n", pptable->PllGfxclkSpreadPercent); - dev_info(smu->adev->dev, "PllGfxclkSpreadFreq = 0x%x\n", pptable->PllGfxclkSpreadFreq); - - dev_info(smu->adev->dev, "DfllGfxclkSpreadEnabled = 0x%x\n", pptable->DfllGfxclkSpreadEnabled); - dev_info(smu->adev->dev, "DfllGfxclkSpreadPercent = 0x%x\n", pptable->DfllGfxclkSpreadPercent); - dev_info(smu->adev->dev, "DfllGfxclkSpreadFreq = 0x%x\n", pptable->DfllGfxclkSpreadFreq); - - dev_info(smu->adev->dev, "UclkSpreadPadding = 0x%x\n", pptable->UclkSpreadPadding); - dev_info(smu->adev->dev, "UclkSpreadFreq = 0x%x\n", pptable->UclkSpreadFreq); - - dev_info(smu->adev->dev, "FclkSpreadEnabled = 0x%x\n", pptable->FclkSpreadEnabled); - dev_info(smu->adev->dev, "FclkSpreadPercent = 0x%x\n", pptable->FclkSpreadPercent); - dev_info(smu->adev->dev, "FclkSpreadFreq = 0x%x\n", pptable->FclkSpreadFreq); - - dev_info(smu->adev->dev, "MemoryChannelEnabled = 0x%x\n", pptable->MemoryChannelEnabled); - dev_info(smu->adev->dev, "DramBitWidth = 0x%x\n", pptable->DramBitWidth); - dev_info(smu->adev->dev, "PaddingMem1[0] = 0x%x\n", pptable->PaddingMem1[0]); - dev_info(smu->adev->dev, "PaddingMem1[1] = 0x%x\n", pptable->PaddingMem1[1]); - dev_info(smu->adev->dev, "PaddingMem1[2] = 0x%x\n", pptable->PaddingMem1[2]); - - dev_info(smu->adev->dev, "TotalBoardPower = 0x%x\n", pptable->TotalBoardPower); - dev_info(smu->adev->dev, "BoardPowerPadding = 0x%x\n", pptable->BoardPowerPadding); - - dev_info(smu->adev->dev, "XgmiLinkSpeed\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->XgmiLinkSpeed[i]); - dev_info(smu->adev->dev, "XgmiLinkWidth\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->XgmiLinkWidth[i]); - dev_info(smu->adev->dev, "XgmiFclkFreq\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->XgmiFclkFreq[i]); - dev_info(smu->adev->dev, "XgmiSocVoltage\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->XgmiSocVoltage[i]); - - dev_info(smu->adev->dev, "HsrEnabled = 0x%x\n", pptable->HsrEnabled); - dev_info(smu->adev->dev, "VddqOffEnabled = 0x%x\n", pptable->VddqOffEnabled); - dev_info(smu->adev->dev, "PaddingUmcFlags[0] = 0x%x\n", pptable->PaddingUmcFlags[0]); - dev_info(smu->adev->dev, "PaddingUmcFlags[1] = 0x%x\n", pptable->PaddingUmcFlags[1]); - - dev_info(smu->adev->dev, "BoardReserved[0] = 0x%x\n", pptable->BoardReserved[0]); - dev_info(smu->adev->dev, "BoardReserved[1] = 0x%x\n", pptable->BoardReserved[1]); - dev_info(smu->adev->dev, "BoardReserved[2] = 0x%x\n", pptable->BoardReserved[2]); - dev_info(smu->adev->dev, "BoardReserved[3] = 0x%x\n", pptable->BoardReserved[3]); - dev_info(smu->adev->dev, "BoardReserved[4] = 0x%x\n", pptable->BoardReserved[4]); - dev_info(smu->adev->dev, "BoardReserved[5] = 0x%x\n", pptable->BoardReserved[5]); - dev_info(smu->adev->dev, "BoardReserved[6] = 0x%x\n", pptable->BoardReserved[6]); - dev_info(smu->adev->dev, "BoardReserved[7] = 0x%x\n", pptable->BoardReserved[7]); - dev_info(smu->adev->dev, "BoardReserved[8] = 0x%x\n", pptable->BoardReserved[8]); - dev_info(smu->adev->dev, "BoardReserved[9] = 0x%x\n", pptable->BoardReserved[9]); - dev_info(smu->adev->dev, "BoardReserved[10] = 0x%x\n", pptable->BoardReserved[10]); - - dev_info(smu->adev->dev, "MmHubPadding[0] = 0x%x\n", pptable->MmHubPadding[0]); - dev_info(smu->adev->dev, "MmHubPadding[1] = 0x%x\n", pptable->MmHubPadding[1]); - dev_info(smu->adev->dev, "MmHubPadding[2] = 0x%x\n", pptable->MmHubPadding[2]); - dev_info(smu->adev->dev, "MmHubPadding[3] = 0x%x\n", pptable->MmHubPadding[3]); - dev_info(smu->adev->dev, "MmHubPadding[4] = 0x%x\n", pptable->MmHubPadding[4]); - dev_info(smu->adev->dev, "MmHubPadding[5] = 0x%x\n", pptable->MmHubPadding[5]); - dev_info(smu->adev->dev, "MmHubPadding[6] = 0x%x\n", pptable->MmHubPadding[6]); - dev_info(smu->adev->dev, "MmHubPadding[7] = 0x%x\n", pptable->MmHubPadding[7]); -} - -static void sienna_cichlid_dump_pptable(struct smu_context *smu) -{ - struct smu_table_context *table_context = &smu->smu_table; - PPTable_t *pptable = table_context->driver_pptable; - int i; - - if (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == - IP_VERSION(11, 0, 13)) { - beige_goby_dump_pptable(smu); - return; - } - - dev_info(smu->adev->dev, "Dumped PPTable:\n"); - - dev_info(smu->adev->dev, "Version = 0x%08x\n", pptable->Version); - dev_info(smu->adev->dev, "FeaturesToRun[0] = 0x%08x\n", pptable->FeaturesToRun[0]); - dev_info(smu->adev->dev, "FeaturesToRun[1] = 0x%08x\n", pptable->FeaturesToRun[1]); - - for (i = 0; i < PPT_THROTTLER_COUNT; i++) { - dev_info(smu->adev->dev, "SocketPowerLimitAc[%d] = 0x%x\n", i, pptable->SocketPowerLimitAc[i]); - dev_info(smu->adev->dev, "SocketPowerLimitAcTau[%d] = 0x%x\n", i, pptable->SocketPowerLimitAcTau[i]); - dev_info(smu->adev->dev, "SocketPowerLimitDc[%d] = 0x%x\n", i, pptable->SocketPowerLimitDc[i]); - dev_info(smu->adev->dev, "SocketPowerLimitDcTau[%d] = 0x%x\n", i, pptable->SocketPowerLimitDcTau[i]); - } - - for (i = 0; i < TDC_THROTTLER_COUNT; i++) { - dev_info(smu->adev->dev, "TdcLimit[%d] = 0x%x\n", i, pptable->TdcLimit[i]); - dev_info(smu->adev->dev, "TdcLimitTau[%d] = 0x%x\n", i, pptable->TdcLimitTau[i]); - } - - for (i = 0; i < TEMP_COUNT; i++) { - dev_info(smu->adev->dev, "TemperatureLimit[%d] = 0x%x\n", i, pptable->TemperatureLimit[i]); - } - - dev_info(smu->adev->dev, "FitLimit = 0x%x\n", pptable->FitLimit); - dev_info(smu->adev->dev, "TotalPowerConfig = 0x%x\n", pptable->TotalPowerConfig); - dev_info(smu->adev->dev, "TotalPowerPadding[0] = 0x%x\n", pptable->TotalPowerPadding[0]); - dev_info(smu->adev->dev, "TotalPowerPadding[1] = 0x%x\n", pptable->TotalPowerPadding[1]); - dev_info(smu->adev->dev, "TotalPowerPadding[2] = 0x%x\n", pptable->TotalPowerPadding[2]); - - dev_info(smu->adev->dev, "ApccPlusResidencyLimit = 0x%x\n", pptable->ApccPlusResidencyLimit); - for (i = 0; i < NUM_SMNCLK_DPM_LEVELS; i++) { - dev_info(smu->adev->dev, "SmnclkDpmFreq[%d] = 0x%x\n", i, pptable->SmnclkDpmFreq[i]); - dev_info(smu->adev->dev, "SmnclkDpmVoltage[%d] = 0x%x\n", i, pptable->SmnclkDpmVoltage[i]); - } - dev_info(smu->adev->dev, "ThrottlerControlMask = 0x%x\n", pptable->ThrottlerControlMask); - - dev_info(smu->adev->dev, "FwDStateMask = 0x%x\n", pptable->FwDStateMask); - - dev_info(smu->adev->dev, "UlvVoltageOffsetSoc = 0x%x\n", pptable->UlvVoltageOffsetSoc); - dev_info(smu->adev->dev, "UlvVoltageOffsetGfx = 0x%x\n", pptable->UlvVoltageOffsetGfx); - dev_info(smu->adev->dev, "MinVoltageUlvGfx = 0x%x\n", pptable->MinVoltageUlvGfx); - dev_info(smu->adev->dev, "MinVoltageUlvSoc = 0x%x\n", pptable->MinVoltageUlvSoc); - - dev_info(smu->adev->dev, "SocLIVmin = 0x%x\n", pptable->SocLIVmin); - dev_info(smu->adev->dev, "PaddingLIVmin = 0x%x\n", pptable->PaddingLIVmin); - - dev_info(smu->adev->dev, "GceaLinkMgrIdleThreshold = 0x%x\n", pptable->GceaLinkMgrIdleThreshold); - dev_info(smu->adev->dev, "paddingRlcUlvParams[0] = 0x%x\n", pptable->paddingRlcUlvParams[0]); - dev_info(smu->adev->dev, "paddingRlcUlvParams[1] = 0x%x\n", pptable->paddingRlcUlvParams[1]); - dev_info(smu->adev->dev, "paddingRlcUlvParams[2] = 0x%x\n", pptable->paddingRlcUlvParams[2]); - - dev_info(smu->adev->dev, "MinVoltageGfx = 0x%x\n", pptable->MinVoltageGfx); - dev_info(smu->adev->dev, "MinVoltageSoc = 0x%x\n", pptable->MinVoltageSoc); - dev_info(smu->adev->dev, "MaxVoltageGfx = 0x%x\n", pptable->MaxVoltageGfx); - dev_info(smu->adev->dev, "MaxVoltageSoc = 0x%x\n", pptable->MaxVoltageSoc); - - dev_info(smu->adev->dev, "LoadLineResistanceGfx = 0x%x\n", pptable->LoadLineResistanceGfx); - dev_info(smu->adev->dev, "LoadLineResistanceSoc = 0x%x\n", pptable->LoadLineResistanceSoc); - - dev_info(smu->adev->dev, "VDDGFX_TVmin = 0x%x\n", pptable->VDDGFX_TVmin); - dev_info(smu->adev->dev, "VDDSOC_TVmin = 0x%x\n", pptable->VDDSOC_TVmin); - dev_info(smu->adev->dev, "VDDGFX_Vmin_HiTemp = 0x%x\n", pptable->VDDGFX_Vmin_HiTemp); - dev_info(smu->adev->dev, "VDDGFX_Vmin_LoTemp = 0x%x\n", pptable->VDDGFX_Vmin_LoTemp); - dev_info(smu->adev->dev, "VDDSOC_Vmin_HiTemp = 0x%x\n", pptable->VDDSOC_Vmin_HiTemp); - dev_info(smu->adev->dev, "VDDSOC_Vmin_LoTemp = 0x%x\n", pptable->VDDSOC_Vmin_LoTemp); - dev_info(smu->adev->dev, "VDDGFX_TVminHystersis = 0x%x\n", pptable->VDDGFX_TVminHystersis); - dev_info(smu->adev->dev, "VDDSOC_TVminHystersis = 0x%x\n", pptable->VDDSOC_TVminHystersis); - - dev_info(smu->adev->dev, "[PPCLK_GFXCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_GFXCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_GFXCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_GFXCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_GFXCLK].Padding, - pptable->DpmDescriptor[PPCLK_GFXCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_GFXCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_GFXCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_GFXCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_SOCCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_SOCCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_SOCCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_SOCCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_SOCCLK].Padding, - pptable->DpmDescriptor[PPCLK_SOCCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_SOCCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_SOCCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_SOCCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_UCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_UCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_UCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_UCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_UCLK].Padding, - pptable->DpmDescriptor[PPCLK_UCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_UCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_UCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_UCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_UCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_FCLK]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_FCLK].VoltageMode, - pptable->DpmDescriptor[PPCLK_FCLK].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_FCLK].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_FCLK].Padding, - pptable->DpmDescriptor[PPCLK_FCLK].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_FCLK].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.a, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.b, - pptable->DpmDescriptor[PPCLK_FCLK].SsCurve.c, - pptable->DpmDescriptor[PPCLK_FCLK].SsFmin, - pptable->DpmDescriptor[PPCLK_FCLK].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_DCLK_0]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_DCLK_0].VoltageMode, - pptable->DpmDescriptor[PPCLK_DCLK_0].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_DCLK_0].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_DCLK_0].Padding, - pptable->DpmDescriptor[PPCLK_DCLK_0].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_DCLK_0].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_DCLK_0].SsCurve.a, - pptable->DpmDescriptor[PPCLK_DCLK_0].SsCurve.b, - pptable->DpmDescriptor[PPCLK_DCLK_0].SsCurve.c, - pptable->DpmDescriptor[PPCLK_DCLK_0].SsFmin, - pptable->DpmDescriptor[PPCLK_DCLK_0].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_VCLK_0]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_VCLK_0].VoltageMode, - pptable->DpmDescriptor[PPCLK_VCLK_0].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_VCLK_0].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_VCLK_0].Padding, - pptable->DpmDescriptor[PPCLK_VCLK_0].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_VCLK_0].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_VCLK_0].SsCurve.a, - pptable->DpmDescriptor[PPCLK_VCLK_0].SsCurve.b, - pptable->DpmDescriptor[PPCLK_VCLK_0].SsCurve.c, - pptable->DpmDescriptor[PPCLK_VCLK_0].SsFmin, - pptable->DpmDescriptor[PPCLK_VCLK_0].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_DCLK_1]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_DCLK_1].VoltageMode, - pptable->DpmDescriptor[PPCLK_DCLK_1].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_DCLK_1].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_DCLK_1].Padding, - pptable->DpmDescriptor[PPCLK_DCLK_1].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_DCLK_1].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_DCLK_1].SsCurve.a, - pptable->DpmDescriptor[PPCLK_DCLK_1].SsCurve.b, - pptable->DpmDescriptor[PPCLK_DCLK_1].SsCurve.c, - pptable->DpmDescriptor[PPCLK_DCLK_1].SsFmin, - pptable->DpmDescriptor[PPCLK_DCLK_1].Padding16); - - dev_info(smu->adev->dev, "[PPCLK_VCLK_1]\n" - " .VoltageMode = 0x%02x\n" - " .SnapToDiscrete = 0x%02x\n" - " .NumDiscreteLevels = 0x%02x\n" - " .padding = 0x%02x\n" - " .ConversionToAvfsClk{m = 0x%08x b = 0x%08x}\n" - " .SsCurve {a = 0x%08x b = 0x%08x c = 0x%08x}\n" - " .SsFmin = 0x%04x\n" - " .Padding_16 = 0x%04x\n", - pptable->DpmDescriptor[PPCLK_VCLK_1].VoltageMode, - pptable->DpmDescriptor[PPCLK_VCLK_1].SnapToDiscrete, - pptable->DpmDescriptor[PPCLK_VCLK_1].NumDiscreteLevels, - pptable->DpmDescriptor[PPCLK_VCLK_1].Padding, - pptable->DpmDescriptor[PPCLK_VCLK_1].ConversionToAvfsClk.m, - pptable->DpmDescriptor[PPCLK_VCLK_1].ConversionToAvfsClk.b, - pptable->DpmDescriptor[PPCLK_VCLK_1].SsCurve.a, - pptable->DpmDescriptor[PPCLK_VCLK_1].SsCurve.b, - pptable->DpmDescriptor[PPCLK_VCLK_1].SsCurve.c, - pptable->DpmDescriptor[PPCLK_VCLK_1].SsFmin, - pptable->DpmDescriptor[PPCLK_VCLK_1].Padding16); - - dev_info(smu->adev->dev, "FreqTableGfx\n"); - for (i = 0; i < NUM_GFXCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableGfx[i]); - - dev_info(smu->adev->dev, "FreqTableVclk\n"); - for (i = 0; i < NUM_VCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableVclk[i]); - - dev_info(smu->adev->dev, "FreqTableDclk\n"); - for (i = 0; i < NUM_DCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableDclk[i]); - - dev_info(smu->adev->dev, "FreqTableSocclk\n"); - for (i = 0; i < NUM_SOCCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableSocclk[i]); - - dev_info(smu->adev->dev, "FreqTableUclk\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableUclk[i]); - - dev_info(smu->adev->dev, "FreqTableFclk\n"); - for (i = 0; i < NUM_FCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%02d] = 0x%x\n", i, pptable->FreqTableFclk[i]); - - dev_info(smu->adev->dev, "DcModeMaxFreq\n"); - dev_info(smu->adev->dev, " .PPCLK_GFXCLK = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_GFXCLK]); - dev_info(smu->adev->dev, " .PPCLK_SOCCLK = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_SOCCLK]); - dev_info(smu->adev->dev, " .PPCLK_UCLK = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_UCLK]); - dev_info(smu->adev->dev, " .PPCLK_FCLK = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_FCLK]); - dev_info(smu->adev->dev, " .PPCLK_DCLK_0 = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_DCLK_0]); - dev_info(smu->adev->dev, " .PPCLK_VCLK_0 = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_VCLK_0]); - dev_info(smu->adev->dev, " .PPCLK_DCLK_1 = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_DCLK_1]); - dev_info(smu->adev->dev, " .PPCLK_VCLK_1 = 0x%x\n", pptable->DcModeMaxFreq[PPCLK_VCLK_1]); - - dev_info(smu->adev->dev, "FreqTableUclkDiv\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->FreqTableUclkDiv[i]); - - dev_info(smu->adev->dev, "FclkBoostFreq = 0x%x\n", pptable->FclkBoostFreq); - dev_info(smu->adev->dev, "FclkParamPadding = 0x%x\n", pptable->FclkParamPadding); - - dev_info(smu->adev->dev, "Mp0clkFreq\n"); - for (i = 0; i < NUM_MP0CLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->Mp0clkFreq[i]); - - dev_info(smu->adev->dev, "Mp0DpmVoltage\n"); - for (i = 0; i < NUM_MP0CLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->Mp0DpmVoltage[i]); - - dev_info(smu->adev->dev, "MemVddciVoltage\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->MemVddciVoltage[i]); - - dev_info(smu->adev->dev, "MemMvddVoltage\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->MemMvddVoltage[i]); - - dev_info(smu->adev->dev, "GfxclkFgfxoffEntry = 0x%x\n", pptable->GfxclkFgfxoffEntry); - dev_info(smu->adev->dev, "GfxclkFinit = 0x%x\n", pptable->GfxclkFinit); - dev_info(smu->adev->dev, "GfxclkFidle = 0x%x\n", pptable->GfxclkFidle); - dev_info(smu->adev->dev, "GfxclkSource = 0x%x\n", pptable->GfxclkSource); - dev_info(smu->adev->dev, "GfxclkPadding = 0x%x\n", pptable->GfxclkPadding); - - dev_info(smu->adev->dev, "GfxGpoSubFeatureMask = 0x%x\n", pptable->GfxGpoSubFeatureMask); - - dev_info(smu->adev->dev, "GfxGpoEnabledWorkPolicyMask = 0x%x\n", pptable->GfxGpoEnabledWorkPolicyMask); - dev_info(smu->adev->dev, "GfxGpoDisabledWorkPolicyMask = 0x%x\n", pptable->GfxGpoDisabledWorkPolicyMask); - dev_info(smu->adev->dev, "GfxGpoPadding[0] = 0x%x\n", pptable->GfxGpoPadding[0]); - dev_info(smu->adev->dev, "GfxGpoVotingAllow = 0x%x\n", pptable->GfxGpoVotingAllow); - dev_info(smu->adev->dev, "GfxGpoPadding32[0] = 0x%x\n", pptable->GfxGpoPadding32[0]); - dev_info(smu->adev->dev, "GfxGpoPadding32[1] = 0x%x\n", pptable->GfxGpoPadding32[1]); - dev_info(smu->adev->dev, "GfxGpoPadding32[2] = 0x%x\n", pptable->GfxGpoPadding32[2]); - dev_info(smu->adev->dev, "GfxGpoPadding32[3] = 0x%x\n", pptable->GfxGpoPadding32[3]); - dev_info(smu->adev->dev, "GfxDcsFopt = 0x%x\n", pptable->GfxDcsFopt); - dev_info(smu->adev->dev, "GfxDcsFclkFopt = 0x%x\n", pptable->GfxDcsFclkFopt); - dev_info(smu->adev->dev, "GfxDcsUclkFopt = 0x%x\n", pptable->GfxDcsUclkFopt); - - dev_info(smu->adev->dev, "DcsGfxOffVoltage = 0x%x\n", pptable->DcsGfxOffVoltage); - dev_info(smu->adev->dev, "DcsMinGfxOffTime = 0x%x\n", pptable->DcsMinGfxOffTime); - dev_info(smu->adev->dev, "DcsMaxGfxOffTime = 0x%x\n", pptable->DcsMaxGfxOffTime); - dev_info(smu->adev->dev, "DcsMinCreditAccum = 0x%x\n", pptable->DcsMinCreditAccum); - dev_info(smu->adev->dev, "DcsExitHysteresis = 0x%x\n", pptable->DcsExitHysteresis); - dev_info(smu->adev->dev, "DcsTimeout = 0x%x\n", pptable->DcsTimeout); - - dev_info(smu->adev->dev, "DcsParamPadding[0] = 0x%x\n", pptable->DcsParamPadding[0]); - dev_info(smu->adev->dev, "DcsParamPadding[1] = 0x%x\n", pptable->DcsParamPadding[1]); - dev_info(smu->adev->dev, "DcsParamPadding[2] = 0x%x\n", pptable->DcsParamPadding[2]); - dev_info(smu->adev->dev, "DcsParamPadding[3] = 0x%x\n", pptable->DcsParamPadding[3]); - dev_info(smu->adev->dev, "DcsParamPadding[4] = 0x%x\n", pptable->DcsParamPadding[4]); - - dev_info(smu->adev->dev, "FlopsPerByteTable\n"); - for (i = 0; i < RLC_PACE_TABLE_NUM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->FlopsPerByteTable[i]); - - dev_info(smu->adev->dev, "LowestUclkReservedForUlv = 0x%x\n", pptable->LowestUclkReservedForUlv); - dev_info(smu->adev->dev, "vddingMem[0] = 0x%x\n", pptable->PaddingMem[0]); - dev_info(smu->adev->dev, "vddingMem[1] = 0x%x\n", pptable->PaddingMem[1]); - dev_info(smu->adev->dev, "vddingMem[2] = 0x%x\n", pptable->PaddingMem[2]); - - dev_info(smu->adev->dev, "UclkDpmPstates\n"); - for (i = 0; i < NUM_UCLK_DPM_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->UclkDpmPstates[i]); - - dev_info(smu->adev->dev, "UclkDpmSrcFreqRange\n"); - dev_info(smu->adev->dev, " .Fmin = 0x%x\n", - pptable->UclkDpmSrcFreqRange.Fmin); - dev_info(smu->adev->dev, " .Fmax = 0x%x\n", - pptable->UclkDpmSrcFreqRange.Fmax); - dev_info(smu->adev->dev, "UclkDpmTargFreqRange\n"); - dev_info(smu->adev->dev, " .Fmin = 0x%x\n", - pptable->UclkDpmTargFreqRange.Fmin); - dev_info(smu->adev->dev, " .Fmax = 0x%x\n", - pptable->UclkDpmTargFreqRange.Fmax); - dev_info(smu->adev->dev, "UclkDpmMidstepFreq = 0x%x\n", pptable->UclkDpmMidstepFreq); - dev_info(smu->adev->dev, "UclkMidstepPadding = 0x%x\n", pptable->UclkMidstepPadding); - - dev_info(smu->adev->dev, "PcieGenSpeed\n"); - for (i = 0; i < NUM_LINK_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->PcieGenSpeed[i]); - - dev_info(smu->adev->dev, "PcieLaneCount\n"); - for (i = 0; i < NUM_LINK_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->PcieLaneCount[i]); - - dev_info(smu->adev->dev, "LclkFreq\n"); - for (i = 0; i < NUM_LINK_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->LclkFreq[i]); - - dev_info(smu->adev->dev, "FanStopTemp = 0x%x\n", pptable->FanStopTemp); - dev_info(smu->adev->dev, "FanStartTemp = 0x%x\n", pptable->FanStartTemp); - - dev_info(smu->adev->dev, "FanGain\n"); - for (i = 0; i < TEMP_COUNT; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->FanGain[i]); - - dev_info(smu->adev->dev, "FanPwmMin = 0x%x\n", pptable->FanPwmMin); - dev_info(smu->adev->dev, "FanAcousticLimitRpm = 0x%x\n", pptable->FanAcousticLimitRpm); - dev_info(smu->adev->dev, "FanThrottlingRpm = 0x%x\n", pptable->FanThrottlingRpm); - dev_info(smu->adev->dev, "FanMaximumRpm = 0x%x\n", pptable->FanMaximumRpm); - dev_info(smu->adev->dev, "MGpuFanBoostLimitRpm = 0x%x\n", pptable->MGpuFanBoostLimitRpm); - dev_info(smu->adev->dev, "FanTargetTemperature = 0x%x\n", pptable->FanTargetTemperature); - dev_info(smu->adev->dev, "FanTargetGfxclk = 0x%x\n", pptable->FanTargetGfxclk); - dev_info(smu->adev->dev, "FanPadding16 = 0x%x\n", pptable->FanPadding16); - dev_info(smu->adev->dev, "FanTempInputSelect = 0x%x\n", pptable->FanTempInputSelect); - dev_info(smu->adev->dev, "FanPadding = 0x%x\n", pptable->FanPadding); - dev_info(smu->adev->dev, "FanZeroRpmEnable = 0x%x\n", pptable->FanZeroRpmEnable); - dev_info(smu->adev->dev, "FanTachEdgePerRev = 0x%x\n", pptable->FanTachEdgePerRev); - - dev_info(smu->adev->dev, "FuzzyFan_ErrorSetDelta = 0x%x\n", pptable->FuzzyFan_ErrorSetDelta); - dev_info(smu->adev->dev, "FuzzyFan_ErrorRateSetDelta = 0x%x\n", pptable->FuzzyFan_ErrorRateSetDelta); - dev_info(smu->adev->dev, "FuzzyFan_PwmSetDelta = 0x%x\n", pptable->FuzzyFan_PwmSetDelta); - dev_info(smu->adev->dev, "FuzzyFan_Reserved = 0x%x\n", pptable->FuzzyFan_Reserved); - - dev_info(smu->adev->dev, "OverrideAvfsGb[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->OverrideAvfsGb[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "OverrideAvfsGb[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->OverrideAvfsGb[AVFS_VOLTAGE_SOC]); - dev_info(smu->adev->dev, "dBtcGbGfxDfllModelSelect = 0x%x\n", pptable->dBtcGbGfxDfllModelSelect); - dev_info(smu->adev->dev, "Padding8_Avfs = 0x%x\n", pptable->Padding8_Avfs); - - dev_info(smu->adev->dev, "qAvfsGb[AVFS_VOLTAGE_GFX]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qAvfsGb[AVFS_VOLTAGE_GFX].a, - pptable->qAvfsGb[AVFS_VOLTAGE_GFX].b, - pptable->qAvfsGb[AVFS_VOLTAGE_GFX].c); - dev_info(smu->adev->dev, "qAvfsGb[AVFS_VOLTAGE_SOC]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qAvfsGb[AVFS_VOLTAGE_SOC].a, - pptable->qAvfsGb[AVFS_VOLTAGE_SOC].b, - pptable->qAvfsGb[AVFS_VOLTAGE_SOC].c); - dev_info(smu->adev->dev, "dBtcGbGfxPll{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbGfxPll.a, - pptable->dBtcGbGfxPll.b, - pptable->dBtcGbGfxPll.c); - dev_info(smu->adev->dev, "dBtcGbGfxAfll{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbGfxDfll.a, - pptable->dBtcGbGfxDfll.b, - pptable->dBtcGbGfxDfll.c); - dev_info(smu->adev->dev, "dBtcGbSoc{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->dBtcGbSoc.a, - pptable->dBtcGbSoc.b, - pptable->dBtcGbSoc.c); - dev_info(smu->adev->dev, "qAgingGb[AVFS_VOLTAGE_GFX]{m = 0x%x b = 0x%x}\n", - pptable->qAgingGb[AVFS_VOLTAGE_GFX].m, - pptable->qAgingGb[AVFS_VOLTAGE_GFX].b); - dev_info(smu->adev->dev, "qAgingGb[AVFS_VOLTAGE_SOC]{m = 0x%x b = 0x%x}\n", - pptable->qAgingGb[AVFS_VOLTAGE_SOC].m, - pptable->qAgingGb[AVFS_VOLTAGE_SOC].b); - - dev_info(smu->adev->dev, "PiecewiseLinearDroopIntGfxDfll\n"); - for (i = 0; i < NUM_PIECE_WISE_LINEAR_DROOP_MODEL_VF_POINTS; i++) { - dev_info(smu->adev->dev, " Fset[%d] = 0x%x\n", - i, pptable->PiecewiseLinearDroopIntGfxDfll.Fset[i]); - dev_info(smu->adev->dev, " Vdroop[%d] = 0x%x\n", - i, pptable->PiecewiseLinearDroopIntGfxDfll.Vdroop[i]); - } - - dev_info(smu->adev->dev, "qStaticVoltageOffset[AVFS_VOLTAGE_GFX]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].a, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].b, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_GFX].c); - dev_info(smu->adev->dev, "qStaticVoltageOffset[AVFS_VOLTAGE_SOC]{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].a, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].b, - pptable->qStaticVoltageOffset[AVFS_VOLTAGE_SOC].c); - - dev_info(smu->adev->dev, "DcTol[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcTol[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcTol[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcTol[AVFS_VOLTAGE_SOC]); - - dev_info(smu->adev->dev, "DcBtcEnabled[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcEnabled[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcEnabled[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcEnabled[AVFS_VOLTAGE_SOC]); - dev_info(smu->adev->dev, "Padding8_GfxBtc[0] = 0x%x\n", pptable->Padding8_GfxBtc[0]); - dev_info(smu->adev->dev, "Padding8_GfxBtc[1] = 0x%x\n", pptable->Padding8_GfxBtc[1]); - - dev_info(smu->adev->dev, "DcBtcMin[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcMin[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcMin[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcMin[AVFS_VOLTAGE_SOC]); - dev_info(smu->adev->dev, "DcBtcMax[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcMax[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcMax[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcMax[AVFS_VOLTAGE_SOC]); - - dev_info(smu->adev->dev, "DcBtcGb[AVFS_VOLTAGE_GFX] = 0x%x\n", pptable->DcBtcGb[AVFS_VOLTAGE_GFX]); - dev_info(smu->adev->dev, "DcBtcGb[AVFS_VOLTAGE_SOC] = 0x%x\n", pptable->DcBtcGb[AVFS_VOLTAGE_SOC]); - - dev_info(smu->adev->dev, "XgmiDpmPstates\n"); - for (i = 0; i < NUM_XGMI_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->XgmiDpmPstates[i]); - dev_info(smu->adev->dev, "XgmiDpmSpare[0] = 0x%02x\n", pptable->XgmiDpmSpare[0]); - dev_info(smu->adev->dev, "XgmiDpmSpare[1] = 0x%02x\n", pptable->XgmiDpmSpare[1]); - - dev_info(smu->adev->dev, "DebugOverrides = 0x%x\n", pptable->DebugOverrides); - dev_info(smu->adev->dev, "ReservedEquation0{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation0.a, - pptable->ReservedEquation0.b, - pptable->ReservedEquation0.c); - dev_info(smu->adev->dev, "ReservedEquation1{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation1.a, - pptable->ReservedEquation1.b, - pptable->ReservedEquation1.c); - dev_info(smu->adev->dev, "ReservedEquation2{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation2.a, - pptable->ReservedEquation2.b, - pptable->ReservedEquation2.c); - dev_info(smu->adev->dev, "ReservedEquation3{a = 0x%x b = 0x%x c = 0x%x}\n", - pptable->ReservedEquation3.a, - pptable->ReservedEquation3.b, - pptable->ReservedEquation3.c); - - dev_info(smu->adev->dev, "SkuReserved[0] = 0x%x\n", pptable->SkuReserved[0]); - dev_info(smu->adev->dev, "SkuReserved[1] = 0x%x\n", pptable->SkuReserved[1]); - dev_info(smu->adev->dev, "SkuReserved[2] = 0x%x\n", pptable->SkuReserved[2]); - dev_info(smu->adev->dev, "SkuReserved[3] = 0x%x\n", pptable->SkuReserved[3]); - dev_info(smu->adev->dev, "SkuReserved[4] = 0x%x\n", pptable->SkuReserved[4]); - dev_info(smu->adev->dev, "SkuReserved[5] = 0x%x\n", pptable->SkuReserved[5]); - dev_info(smu->adev->dev, "SkuReserved[6] = 0x%x\n", pptable->SkuReserved[6]); - dev_info(smu->adev->dev, "SkuReserved[7] = 0x%x\n", pptable->SkuReserved[7]); - - dev_info(smu->adev->dev, "GamingClk[0] = 0x%x\n", pptable->GamingClk[0]); - dev_info(smu->adev->dev, "GamingClk[1] = 0x%x\n", pptable->GamingClk[1]); - dev_info(smu->adev->dev, "GamingClk[2] = 0x%x\n", pptable->GamingClk[2]); - dev_info(smu->adev->dev, "GamingClk[3] = 0x%x\n", pptable->GamingClk[3]); - dev_info(smu->adev->dev, "GamingClk[4] = 0x%x\n", pptable->GamingClk[4]); - dev_info(smu->adev->dev, "GamingClk[5] = 0x%x\n", pptable->GamingClk[5]); - - for (i = 0; i < NUM_I2C_CONTROLLERS; i++) { - dev_info(smu->adev->dev, "I2cControllers[%d]:\n", i); - dev_info(smu->adev->dev, " .Enabled = 0x%x\n", - pptable->I2cControllers[i].Enabled); - dev_info(smu->adev->dev, " .Speed = 0x%x\n", - pptable->I2cControllers[i].Speed); - dev_info(smu->adev->dev, " .SlaveAddress = 0x%x\n", - pptable->I2cControllers[i].SlaveAddress); - dev_info(smu->adev->dev, " .ControllerPort = 0x%x\n", - pptable->I2cControllers[i].ControllerPort); - dev_info(smu->adev->dev, " .ControllerName = 0x%x\n", - pptable->I2cControllers[i].ControllerName); - dev_info(smu->adev->dev, " .ThermalThrottler = 0x%x\n", - pptable->I2cControllers[i].ThermalThrotter); - dev_info(smu->adev->dev, " .I2cProtocol = 0x%x\n", - pptable->I2cControllers[i].I2cProtocol); - dev_info(smu->adev->dev, " .PaddingConfig = 0x%x\n", - pptable->I2cControllers[i].PaddingConfig); - } - - dev_info(smu->adev->dev, "GpioScl = 0x%x\n", pptable->GpioScl); - dev_info(smu->adev->dev, "GpioSda = 0x%x\n", pptable->GpioSda); - dev_info(smu->adev->dev, "FchUsbPdSlaveAddr = 0x%x\n", pptable->FchUsbPdSlaveAddr); - dev_info(smu->adev->dev, "I2cSpare[0] = 0x%x\n", pptable->I2cSpare[0]); - - dev_info(smu->adev->dev, "Board Parameters:\n"); - dev_info(smu->adev->dev, "VddGfxVrMapping = 0x%x\n", pptable->VddGfxVrMapping); - dev_info(smu->adev->dev, "VddSocVrMapping = 0x%x\n", pptable->VddSocVrMapping); - dev_info(smu->adev->dev, "VddMem0VrMapping = 0x%x\n", pptable->VddMem0VrMapping); - dev_info(smu->adev->dev, "VddMem1VrMapping = 0x%x\n", pptable->VddMem1VrMapping); - dev_info(smu->adev->dev, "GfxUlvPhaseSheddingMask = 0x%x\n", pptable->GfxUlvPhaseSheddingMask); - dev_info(smu->adev->dev, "SocUlvPhaseSheddingMask = 0x%x\n", pptable->SocUlvPhaseSheddingMask); - dev_info(smu->adev->dev, "VddciUlvPhaseSheddingMask = 0x%x\n", pptable->VddciUlvPhaseSheddingMask); - dev_info(smu->adev->dev, "MvddUlvPhaseSheddingMask = 0x%x\n", pptable->MvddUlvPhaseSheddingMask); - - dev_info(smu->adev->dev, "GfxMaxCurrent = 0x%x\n", pptable->GfxMaxCurrent); - dev_info(smu->adev->dev, "GfxOffset = 0x%x\n", pptable->GfxOffset); - dev_info(smu->adev->dev, "Padding_TelemetryGfx = 0x%x\n", pptable->Padding_TelemetryGfx); - - dev_info(smu->adev->dev, "SocMaxCurrent = 0x%x\n", pptable->SocMaxCurrent); - dev_info(smu->adev->dev, "SocOffset = 0x%x\n", pptable->SocOffset); - dev_info(smu->adev->dev, "Padding_TelemetrySoc = 0x%x\n", pptable->Padding_TelemetrySoc); - - dev_info(smu->adev->dev, "Mem0MaxCurrent = 0x%x\n", pptable->Mem0MaxCurrent); - dev_info(smu->adev->dev, "Mem0Offset = 0x%x\n", pptable->Mem0Offset); - dev_info(smu->adev->dev, "Padding_TelemetryMem0 = 0x%x\n", pptable->Padding_TelemetryMem0); - - dev_info(smu->adev->dev, "Mem1MaxCurrent = 0x%x\n", pptable->Mem1MaxCurrent); - dev_info(smu->adev->dev, "Mem1Offset = 0x%x\n", pptable->Mem1Offset); - dev_info(smu->adev->dev, "Padding_TelemetryMem1 = 0x%x\n", pptable->Padding_TelemetryMem1); - - dev_info(smu->adev->dev, "MvddRatio = 0x%x\n", pptable->MvddRatio); - - dev_info(smu->adev->dev, "AcDcGpio = 0x%x\n", pptable->AcDcGpio); - dev_info(smu->adev->dev, "AcDcPolarity = 0x%x\n", pptable->AcDcPolarity); - dev_info(smu->adev->dev, "VR0HotGpio = 0x%x\n", pptable->VR0HotGpio); - dev_info(smu->adev->dev, "VR0HotPolarity = 0x%x\n", pptable->VR0HotPolarity); - dev_info(smu->adev->dev, "VR1HotGpio = 0x%x\n", pptable->VR1HotGpio); - dev_info(smu->adev->dev, "VR1HotPolarity = 0x%x\n", pptable->VR1HotPolarity); - dev_info(smu->adev->dev, "GthrGpio = 0x%x\n", pptable->GthrGpio); - dev_info(smu->adev->dev, "GthrPolarity = 0x%x\n", pptable->GthrPolarity); - dev_info(smu->adev->dev, "LedPin0 = 0x%x\n", pptable->LedPin0); - dev_info(smu->adev->dev, "LedPin1 = 0x%x\n", pptable->LedPin1); - dev_info(smu->adev->dev, "LedPin2 = 0x%x\n", pptable->LedPin2); - dev_info(smu->adev->dev, "LedEnableMask = 0x%x\n", pptable->LedEnableMask); - dev_info(smu->adev->dev, "LedPcie = 0x%x\n", pptable->LedPcie); - dev_info(smu->adev->dev, "LedError = 0x%x\n", pptable->LedError); - dev_info(smu->adev->dev, "LedSpare1[0] = 0x%x\n", pptable->LedSpare1[0]); - dev_info(smu->adev->dev, "LedSpare1[1] = 0x%x\n", pptable->LedSpare1[1]); - - dev_info(smu->adev->dev, "PllGfxclkSpreadEnabled = 0x%x\n", pptable->PllGfxclkSpreadEnabled); - dev_info(smu->adev->dev, "PllGfxclkSpreadPercent = 0x%x\n", pptable->PllGfxclkSpreadPercent); - dev_info(smu->adev->dev, "PllGfxclkSpreadFreq = 0x%x\n", pptable->PllGfxclkSpreadFreq); - - dev_info(smu->adev->dev, "DfllGfxclkSpreadEnabled = 0x%x\n", pptable->DfllGfxclkSpreadEnabled); - dev_info(smu->adev->dev, "DfllGfxclkSpreadPercent = 0x%x\n", pptable->DfllGfxclkSpreadPercent); - dev_info(smu->adev->dev, "DfllGfxclkSpreadFreq = 0x%x\n", pptable->DfllGfxclkSpreadFreq); - - dev_info(smu->adev->dev, "UclkSpreadPadding = 0x%x\n", pptable->UclkSpreadPadding); - dev_info(smu->adev->dev, "UclkSpreadFreq = 0x%x\n", pptable->UclkSpreadFreq); - - dev_info(smu->adev->dev, "FclkSpreadEnabled = 0x%x\n", pptable->FclkSpreadEnabled); - dev_info(smu->adev->dev, "FclkSpreadPercent = 0x%x\n", pptable->FclkSpreadPercent); - dev_info(smu->adev->dev, "FclkSpreadFreq = 0x%x\n", pptable->FclkSpreadFreq); - - dev_info(smu->adev->dev, "MemoryChannelEnabled = 0x%x\n", pptable->MemoryChannelEnabled); - dev_info(smu->adev->dev, "DramBitWidth = 0x%x\n", pptable->DramBitWidth); - dev_info(smu->adev->dev, "PaddingMem1[0] = 0x%x\n", pptable->PaddingMem1[0]); - dev_info(smu->adev->dev, "PaddingMem1[1] = 0x%x\n", pptable->PaddingMem1[1]); - dev_info(smu->adev->dev, "PaddingMem1[2] = 0x%x\n", pptable->PaddingMem1[2]); - - dev_info(smu->adev->dev, "TotalBoardPower = 0x%x\n", pptable->TotalBoardPower); - dev_info(smu->adev->dev, "BoardPowerPadding = 0x%x\n", pptable->BoardPowerPadding); - - dev_info(smu->adev->dev, "XgmiLinkSpeed\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->XgmiLinkSpeed[i]); - dev_info(smu->adev->dev, "XgmiLinkWidth\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->XgmiLinkWidth[i]); - dev_info(smu->adev->dev, "XgmiFclkFreq\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->XgmiFclkFreq[i]); - dev_info(smu->adev->dev, "XgmiSocVoltage\n"); - for (i = 0; i < NUM_XGMI_PSTATE_LEVELS; i++) - dev_info(smu->adev->dev, " .[%d] = 0x%x\n", i, pptable->XgmiSocVoltage[i]); - - dev_info(smu->adev->dev, "HsrEnabled = 0x%x\n", pptable->HsrEnabled); - dev_info(smu->adev->dev, "VddqOffEnabled = 0x%x\n", pptable->VddqOffEnabled); - dev_info(smu->adev->dev, "PaddingUmcFlags[0] = 0x%x\n", pptable->PaddingUmcFlags[0]); - dev_info(smu->adev->dev, "PaddingUmcFlags[1] = 0x%x\n", pptable->PaddingUmcFlags[1]); - - dev_info(smu->adev->dev, "BoardReserved[0] = 0x%x\n", pptable->BoardReserved[0]); - dev_info(smu->adev->dev, "BoardReserved[1] = 0x%x\n", pptable->BoardReserved[1]); - dev_info(smu->adev->dev, "BoardReserved[2] = 0x%x\n", pptable->BoardReserved[2]); - dev_info(smu->adev->dev, "BoardReserved[3] = 0x%x\n", pptable->BoardReserved[3]); - dev_info(smu->adev->dev, "BoardReserved[4] = 0x%x\n", pptable->BoardReserved[4]); - dev_info(smu->adev->dev, "BoardReserved[5] = 0x%x\n", pptable->BoardReserved[5]); - dev_info(smu->adev->dev, "BoardReserved[6] = 0x%x\n", pptable->BoardReserved[6]); - dev_info(smu->adev->dev, "BoardReserved[7] = 0x%x\n", pptable->BoardReserved[7]); - dev_info(smu->adev->dev, "BoardReserved[8] = 0x%x\n", pptable->BoardReserved[8]); - dev_info(smu->adev->dev, "BoardReserved[9] = 0x%x\n", pptable->BoardReserved[9]); - dev_info(smu->adev->dev, "BoardReserved[10] = 0x%x\n", pptable->BoardReserved[10]); - - dev_info(smu->adev->dev, "MmHubPadding[0] = 0x%x\n", pptable->MmHubPadding[0]); - dev_info(smu->adev->dev, "MmHubPadding[1] = 0x%x\n", pptable->MmHubPadding[1]); - dev_info(smu->adev->dev, "MmHubPadding[2] = 0x%x\n", pptable->MmHubPadding[2]); - dev_info(smu->adev->dev, "MmHubPadding[3] = 0x%x\n", pptable->MmHubPadding[3]); - dev_info(smu->adev->dev, "MmHubPadding[4] = 0x%x\n", pptable->MmHubPadding[4]); - dev_info(smu->adev->dev, "MmHubPadding[5] = 0x%x\n", pptable->MmHubPadding[5]); - dev_info(smu->adev->dev, "MmHubPadding[6] = 0x%x\n", pptable->MmHubPadding[6]); - dev_info(smu->adev->dev, "MmHubPadding[7] = 0x%x\n", pptable->MmHubPadding[7]); -} - static int sienna_cichlid_i2c_xfer(struct i2c_adapter *i2c_adap, struct i2c_msg *msg, int num_msgs) { @@ -4397,7 +3129,6 @@ static const struct pptable_funcs sienna_cichlid_ppt_funcs = { .display_disable_memory_clock_switch = sienna_cichlid_display_disable_memory_clock_switch, .get_power_limit = sienna_cichlid_get_power_limit, .update_pcie_parameters = sienna_cichlid_update_pcie_parameters, - .dump_pptable = sienna_cichlid_dump_pptable, .init_microcode = smu_v11_0_init_microcode, .load_microcode = smu_v11_0_load_microcode, .fini_microcode = smu_v11_0_fini_microcode, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c index 16fcd9dcd202..480cf3cb204d 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/smu_v11_0.c @@ -1616,7 +1616,8 @@ int smu_v11_0_baco_set_state(struct smu_context *smu, enum smu_baco_state state) break; default: if (!ras || !adev->ras_enabled || - adev->gmc.xgmi.pending_reset) { + (adev->init_lvl->level == + AMDGPU_INIT_LEVEL_MINIMAL_XGMI)) { if (amdgpu_ip_version(adev, MP1_HWIP, 0) == IP_VERSION(11, 0, 2)) { data = RREG32_SOC15(THM, 0, mmTHM_BACO_CNTL_ARCT); @@ -1763,7 +1764,8 @@ failed: int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t min, - uint32_t max) + uint32_t max, + bool automatic) { int ret = 0, clk_id = 0; uint32_t param; @@ -1778,7 +1780,10 @@ int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, return clk_id; if (max > 0) { - param = (uint32_t)((clk_id << 16) | (max & 0xffff)); + if (automatic) + param = (uint32_t)((clk_id << 16) | 0xffff); + else + param = (uint32_t)((clk_id << 16) | (max & 0xffff)); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxByFreq, param, NULL); if (ret) @@ -1786,7 +1791,10 @@ int smu_v11_0_set_soft_freq_limited_range(struct smu_context *smu, } if (min > 0) { - param = (uint32_t)((clk_id << 16) | (min & 0xffff)); + if (automatic) + param = (uint32_t)((clk_id << 16) | 0); + else + param = (uint32_t)((clk_id << 16) | (min & 0xffff)); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMinByFreq, param, NULL); if (ret) @@ -1854,6 +1862,7 @@ int smu_v11_0_set_performance_level(struct smu_context *smu, uint32_t mclk_min = 0, mclk_max = 0; uint32_t socclk_min = 0, socclk_max = 0; int ret = 0; + bool auto_level = false; switch (level) { case AMD_DPM_FORCED_LEVEL_HIGH: @@ -1873,6 +1882,7 @@ int smu_v11_0_set_performance_level(struct smu_context *smu, mclk_max = mem_table->max; socclk_min = soc_table->min; socclk_max = soc_table->max; + auto_level = true; break; case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD: sclk_min = sclk_max = pstate_table->gfxclk_pstate.standard; @@ -1905,13 +1915,15 @@ int smu_v11_0_set_performance_level(struct smu_context *smu, if (amdgpu_ip_version(adev, MP1_HWIP, 0) == IP_VERSION(11, 0, 2)) { mclk_min = mclk_max = 0; socclk_min = socclk_max = 0; + auto_level = false; } if (sclk_min && sclk_max) { ret = smu_v11_0_set_soft_freq_limited_range(smu, SMU_GFXCLK, sclk_min, - sclk_max); + sclk_max, + auto_level); if (ret) return ret; } @@ -1920,7 +1932,8 @@ int smu_v11_0_set_performance_level(struct smu_context *smu, ret = smu_v11_0_set_soft_freq_limited_range(smu, SMU_MCLK, mclk_min, - mclk_max); + mclk_max, + auto_level); if (ret) return ret; } @@ -1929,7 +1942,8 @@ int smu_v11_0_set_performance_level(struct smu_context *smu, ret = smu_v11_0_set_soft_freq_limited_range(smu, SMU_SOCCLK, socclk_min, - socclk_max); + socclk_max, + auto_level); if (ret) return ret; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c index 22737b11b1bf..6c43724c01dd 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c @@ -242,7 +242,9 @@ static int vangogh_tables_init(struct smu_context *smu) goto err0_out; smu_table->metrics_time = 0; - smu_table->gpu_metrics_table_size = max(sizeof(struct gpu_metrics_v2_3), sizeof(struct gpu_metrics_v2_2)); + smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v2_2); + smu_table->gpu_metrics_table_size = max(smu_table->gpu_metrics_table_size, sizeof(struct gpu_metrics_v2_3)); + smu_table->gpu_metrics_table_size = max(smu_table->gpu_metrics_table_size, sizeof(struct gpu_metrics_v2_4)); smu_table->gpu_metrics_table = kzalloc(smu_table->gpu_metrics_table_size, GFP_KERNEL); if (!smu_table->gpu_metrics_table) goto err1_out; @@ -1091,9 +1093,10 @@ static int vangogh_set_power_profile_mode(struct smu_context *smu, long *input, } static int vangogh_set_soft_freq_limited_range(struct smu_context *smu, - enum smu_clk_type clk_type, - uint32_t min, - uint32_t max) + enum smu_clk_type clk_type, + uint32_t min, + uint32_t max, + bool automatic) { int ret = 0; @@ -1299,7 +1302,7 @@ static int vangogh_force_dpm_limit_value(struct smu_context *smu, bool highest) return ret; force_freq = highest ? max_freq : min_freq; - ret = vangogh_set_soft_freq_limited_range(smu, clk_type, force_freq, force_freq); + ret = vangogh_set_soft_freq_limited_range(smu, clk_type, force_freq, force_freq, false); if (ret) return ret; } @@ -1335,7 +1338,7 @@ static int vangogh_unforce_dpm_levels(struct smu_context *smu) if (ret) return ret; - ret = vangogh_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq); + ret = vangogh_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq, false); if (ret) return ret; @@ -1354,7 +1357,7 @@ static int vangogh_set_peak_clock_by_device(struct smu_context *smu) if (ret) return ret; - ret = vangogh_set_soft_freq_limited_range(smu, SMU_FCLK, fclk_freq, fclk_freq); + ret = vangogh_set_soft_freq_limited_range(smu, SMU_FCLK, fclk_freq, fclk_freq, false); if (ret) return ret; @@ -1362,7 +1365,7 @@ static int vangogh_set_peak_clock_by_device(struct smu_context *smu) if (ret) return ret; - ret = vangogh_set_soft_freq_limited_range(smu, SMU_SOCCLK, socclk_freq, socclk_freq); + ret = vangogh_set_soft_freq_limited_range(smu, SMU_SOCCLK, socclk_freq, socclk_freq, false); if (ret) return ret; @@ -1370,7 +1373,7 @@ static int vangogh_set_peak_clock_by_device(struct smu_context *smu) if (ret) return ret; - ret = vangogh_set_soft_freq_limited_range(smu, SMU_VCLK, vclk_freq, vclk_freq); + ret = vangogh_set_soft_freq_limited_range(smu, SMU_VCLK, vclk_freq, vclk_freq, false); if (ret) return ret; @@ -1378,7 +1381,7 @@ static int vangogh_set_peak_clock_by_device(struct smu_context *smu) if (ret) return ret; - ret = vangogh_set_soft_freq_limited_range(smu, SMU_DCLK, dclk_freq, dclk_freq); + ret = vangogh_set_soft_freq_limited_range(smu, SMU_DCLK, dclk_freq, dclk_freq, false); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c index cc0504b063fa..0b210b1f2628 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c @@ -707,7 +707,7 @@ static int renoir_force_dpm_limit_value(struct smu_context *smu, bool highest) return ret; force_freq = highest ? max_freq : min_freq; - ret = smu_v12_0_set_soft_freq_limited_range(smu, clk_type, force_freq, force_freq); + ret = smu_v12_0_set_soft_freq_limited_range(smu, clk_type, force_freq, force_freq, false); if (ret) return ret; } @@ -740,7 +740,7 @@ static int renoir_unforce_dpm_levels(struct smu_context *smu) { if (ret) return ret; - ret = smu_v12_0_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq); + ret = smu_v12_0_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq, false); if (ret) return ret; } @@ -911,7 +911,7 @@ static int renoir_set_peak_clock_by_device(struct smu_context *smu) if (ret) return ret; - ret = smu_v12_0_set_soft_freq_limited_range(smu, SMU_SCLK, sclk_freq, sclk_freq); + ret = smu_v12_0_set_soft_freq_limited_range(smu, SMU_SCLK, sclk_freq, sclk_freq, false); if (ret) return ret; @@ -919,7 +919,7 @@ static int renoir_set_peak_clock_by_device(struct smu_context *smu) if (ret) return ret; - ret = smu_v12_0_set_soft_freq_limited_range(smu, SMU_UCLK, uclk_freq, uclk_freq); + ret = smu_v12_0_set_soft_freq_limited_range(smu, SMU_UCLK, uclk_freq, uclk_freq, false); if (ret) return ret; @@ -961,13 +961,13 @@ static int renior_set_dpm_profile_freq(struct smu_context *smu, } if (sclk) - ret = smu_v12_0_set_soft_freq_limited_range(smu, SMU_SCLK, sclk, sclk); + ret = smu_v12_0_set_soft_freq_limited_range(smu, SMU_SCLK, sclk, sclk, false); if (socclk) - ret = smu_v12_0_set_soft_freq_limited_range(smu, SMU_SOCCLK, socclk, socclk); + ret = smu_v12_0_set_soft_freq_limited_range(smu, SMU_SOCCLK, socclk, socclk, false); if (fclk) - ret = smu_v12_0_set_soft_freq_limited_range(smu, SMU_FCLK, fclk, fclk); + ret = smu_v12_0_set_soft_freq_limited_range(smu, SMU_FCLK, fclk, fclk, false); return ret; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c index ed15f5a0fd11..3d3cd546f0ad 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c @@ -211,7 +211,7 @@ int smu_v12_0_mode2_reset(struct smu_context *smu) } int smu_v12_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, uint32_t max) + uint32_t min, uint32_t max, bool automatic) { int ret = 0; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index 2c35eb31475a..f6b029354327 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -1297,9 +1297,10 @@ static int aldebaran_set_performance_level(struct smu_context *smu, } static int aldebaran_set_soft_freq_limited_range(struct smu_context *smu, - enum smu_clk_type clk_type, - uint32_t min, - uint32_t max) + enum smu_clk_type clk_type, + uint32_t min, + uint32_t max, + bool automatic) { struct smu_dpm_context *smu_dpm = &(smu->smu_dpm); struct smu_13_0_dpm_context *dpm_context = smu_dpm->dpm_context; @@ -1328,7 +1329,7 @@ static int aldebaran_set_soft_freq_limited_range(struct smu_context *smu, return 0; ret = smu_v13_0_set_soft_freq_limited_range(smu, SMU_GFXCLK, - min, max); + min, max, false); if (!ret) { pstate_table->gfxclk_pstate.curr.min = min; pstate_table->gfxclk_pstate.curr.max = max; @@ -1348,7 +1349,7 @@ static int aldebaran_set_soft_freq_limited_range(struct smu_context *smu, /* Restore default min/max clocks and enable determinism */ min_clk = dpm_context->dpm_tables.gfx_table.min; max_clk = dpm_context->dpm_tables.gfx_table.max; - ret = smu_v13_0_set_soft_freq_limited_range(smu, SMU_GFXCLK, min_clk, max_clk); + ret = smu_v13_0_set_soft_freq_limited_range(smu, SMU_GFXCLK, min_clk, max_clk, false); if (!ret) { usleep_range(500, 1000); ret = smu_cmn_send_smc_msg_with_param(smu, @@ -1422,7 +1423,7 @@ static int aldebaran_usr_edit_dpm_table(struct smu_context *smu, enum PP_OD_DPM_ min_clk = dpm_context->dpm_tables.gfx_table.min; max_clk = dpm_context->dpm_tables.gfx_table.max; - return aldebaran_set_soft_freq_limited_range(smu, SMU_GFXCLK, min_clk, max_clk); + return aldebaran_set_soft_freq_limited_range(smu, SMU_GFXCLK, min_clk, max_clk, false); } break; case PP_OD_COMMIT_DPM_TABLE: @@ -1441,7 +1442,7 @@ static int aldebaran_usr_edit_dpm_table(struct smu_context *smu, enum PP_OD_DPM_ min_clk = pstate_table->gfxclk_pstate.custom.min; max_clk = pstate_table->gfxclk_pstate.custom.max; - return aldebaran_set_soft_freq_limited_range(smu, SMU_GFXCLK, min_clk, max_clk); + return aldebaran_set_soft_freq_limited_range(smu, SMU_GFXCLK, min_clk, max_clk, false); } break; default: diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index e17466cc1952..6cfd66363915 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -1608,7 +1608,8 @@ failed: int smu_v13_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t min, - uint32_t max) + uint32_t max, + bool automatic) { int ret = 0, clk_id = 0; uint32_t param; @@ -1623,7 +1624,10 @@ int smu_v13_0_set_soft_freq_limited_range(struct smu_context *smu, return clk_id; if (max > 0) { - param = (uint32_t)((clk_id << 16) | (max & 0xffff)); + if (automatic) + param = (uint32_t)((clk_id << 16) | 0xffff); + else + param = (uint32_t)((clk_id << 16) | (max & 0xffff)); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxByFreq, param, NULL); if (ret) @@ -1631,7 +1635,10 @@ int smu_v13_0_set_soft_freq_limited_range(struct smu_context *smu, } if (min > 0) { - param = (uint32_t)((clk_id << 16) | (min & 0xffff)); + if (automatic) + param = (uint32_t)((clk_id << 16) | 0); + else + param = (uint32_t)((clk_id << 16) | (min & 0xffff)); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMinByFreq, param, NULL); if (ret) @@ -1708,6 +1715,7 @@ int smu_v13_0_set_performance_level(struct smu_context *smu, uint32_t dclk_min = 0, dclk_max = 0; uint32_t fclk_min = 0, fclk_max = 0; int ret = 0, i; + bool auto_level = false; switch (level) { case AMD_DPM_FORCED_LEVEL_HIGH: @@ -1739,6 +1747,7 @@ int smu_v13_0_set_performance_level(struct smu_context *smu, dclk_max = dclk_table->max; fclk_min = fclk_table->min; fclk_max = fclk_table->max; + auto_level = true; break; case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD: sclk_min = sclk_max = pstate_table->gfxclk_pstate.standard; @@ -1780,13 +1789,15 @@ int smu_v13_0_set_performance_level(struct smu_context *smu, vclk_min = vclk_max = 0; dclk_min = dclk_max = 0; fclk_min = fclk_max = 0; + auto_level = false; } if (sclk_min && sclk_max) { ret = smu_v13_0_set_soft_freq_limited_range(smu, SMU_GFXCLK, sclk_min, - sclk_max); + sclk_max, + auto_level); if (ret) return ret; @@ -1798,7 +1809,8 @@ int smu_v13_0_set_performance_level(struct smu_context *smu, ret = smu_v13_0_set_soft_freq_limited_range(smu, SMU_MCLK, mclk_min, - mclk_max); + mclk_max, + auto_level); if (ret) return ret; @@ -1810,7 +1822,8 @@ int smu_v13_0_set_performance_level(struct smu_context *smu, ret = smu_v13_0_set_soft_freq_limited_range(smu, SMU_SOCCLK, socclk_min, - socclk_max); + socclk_max, + auto_level); if (ret) return ret; @@ -1825,7 +1838,8 @@ int smu_v13_0_set_performance_level(struct smu_context *smu, ret = smu_v13_0_set_soft_freq_limited_range(smu, i ? SMU_VCLK1 : SMU_VCLK, vclk_min, - vclk_max); + vclk_max, + auto_level); if (ret) return ret; } @@ -1840,7 +1854,8 @@ int smu_v13_0_set_performance_level(struct smu_context *smu, ret = smu_v13_0_set_soft_freq_limited_range(smu, i ? SMU_DCLK1 : SMU_DCLK, dclk_min, - dclk_max); + dclk_max, + auto_level); if (ret) return ret; } @@ -1852,7 +1867,8 @@ int smu_v13_0_set_performance_level(struct smu_context *smu, ret = smu_v13_0_set_soft_freq_limited_range(smu, SMU_FCLK, fclk_min, - fclk_max); + fclk_max, + auto_level); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 1d024b122b0c..8d25cc1f218f 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -736,19 +736,6 @@ static bool smu_v13_0_0_is_dpm_running(struct smu_context *smu) return !!(feature_enabled & SMC_DPM_FEATURE); } -static void smu_v13_0_0_dump_pptable(struct smu_context *smu) -{ - struct smu_table_context *table_context = &smu->smu_table; - PPTable_t *pptable = table_context->driver_pptable; - SkuTable_t *skutable = &pptable->SkuTable; - - dev_info(smu->adev->dev, "Dumped PPTable:\n"); - - dev_info(smu->adev->dev, "Version = 0x%08x\n", skutable->Version); - dev_info(smu->adev->dev, "FeaturesToRun[0] = 0x%08x\n", skutable->FeaturesToRun[0]); - dev_info(smu->adev->dev, "FeaturesToRun[1] = 0x%08x\n", skutable->FeaturesToRun[1]); -} - static int smu_v13_0_0_system_features_control(struct smu_context *smu, bool en) { @@ -1975,7 +1962,8 @@ static int smu_v13_0_0_force_clk_levels(struct smu_context *smu, ret = smu_v13_0_set_soft_freq_limited_range(smu, clk_type, min_freq, - max_freq); + max_freq, + false); break; case SMU_DCEFCLK: case SMU_PCIE: @@ -2485,7 +2473,7 @@ static int smu_v13_0_0_set_power_profile_mode(struct smu_context *smu, DpmActivityMonitorCoeffInt_t *activity_monitor = &(activity_monitor_external.DpmActivityMonitorCoeffInt); int workload_type, ret = 0; - u32 workload_mask; + u32 workload_mask, selected_workload_mask; smu->power_profile_mode = input[size]; @@ -2552,21 +2540,19 @@ static int smu_v13_0_0_set_power_profile_mode(struct smu_context *smu, if (workload_type < 0) return -EINVAL; - workload_mask = 1 << workload_type; + selected_workload_mask = workload_mask = 1 << workload_type; /* Add optimizations for SMU13.0.0/10. Reuse the power saving profile */ - if (smu->power_profile_mode == PP_SMC_POWER_PROFILE_COMPUTE) { - if ((amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 0) && - ((smu->adev->pm.fw_version == 0x004e6601) || - (smu->adev->pm.fw_version >= 0x004e7300))) || - (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 10) && - smu->adev->pm.fw_version >= 0x00504500)) { - workload_type = smu_cmn_to_asic_specific_index(smu, - CMN2ASIC_MAPPING_WORKLOAD, - PP_SMC_POWER_PROFILE_POWERSAVING); - if (workload_type >= 0) - workload_mask |= 1 << workload_type; - } + if ((amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 0) && + ((smu->adev->pm.fw_version == 0x004e6601) || + (smu->adev->pm.fw_version >= 0x004e7300))) || + (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 10) && + smu->adev->pm.fw_version >= 0x00504500)) { + workload_type = smu_cmn_to_asic_specific_index(smu, + CMN2ASIC_MAPPING_WORKLOAD, + PP_SMC_POWER_PROFILE_POWERSAVING); + if (workload_type >= 0) + workload_mask |= 1 << workload_type; } ret = smu_cmn_send_smc_msg_with_param(smu, @@ -2574,7 +2560,7 @@ static int smu_v13_0_0_set_power_profile_mode(struct smu_context *smu, workload_mask, NULL); if (!ret) - smu->workload_mask = workload_mask; + smu->workload_mask = selected_workload_mask; return ret; } @@ -3026,7 +3012,6 @@ static const struct pptable_funcs smu_v13_0_0_ppt_funcs = { .i2c_init = smu_v13_0_0_i2c_control_init, .i2c_fini = smu_v13_0_0_i2c_control_fini, .is_dpm_running = smu_v13_0_0_is_dpm_running, - .dump_pptable = smu_v13_0_0_dump_pptable, .init_microcode = smu_v13_0_init_microcode, .load_microcode = smu_v13_0_load_microcode, .fini_microcode = smu_v13_0_fini_microcode, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c index 9c2c43bfed0b..a71b7c0803f1 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_5_ppt.c @@ -811,9 +811,10 @@ failed: } static int smu_v13_0_5_set_soft_freq_limited_range(struct smu_context *smu, - enum smu_clk_type clk_type, - uint32_t min, - uint32_t max) + enum smu_clk_type clk_type, + uint32_t min, + uint32_t max, + bool automatic) { enum smu_message_type msg_set_min, msg_set_max; uint32_t min_clk = min; @@ -950,7 +951,7 @@ static int smu_v13_0_5_force_clk_levels(struct smu_context *smu, if (ret) goto force_level_out; - ret = smu_v13_0_5_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq); + ret = smu_v13_0_5_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq, false); if (ret) goto force_level_out; break; @@ -1046,9 +1047,10 @@ static int smu_v13_0_5_set_performance_level(struct smu_context *smu, if (sclk_min && sclk_max) { ret = smu_v13_0_5_set_soft_freq_limited_range(smu, - SMU_SCLK, - sclk_min, - sclk_max); + SMU_SCLK, + sclk_min, + sclk_max, + false); if (ret) return ret; @@ -1060,7 +1062,8 @@ static int smu_v13_0_5_set_performance_level(struct smu_context *smu, ret = smu_v13_0_5_set_soft_freq_limited_range(smu, SMU_VCLK, vclk_min, - vclk_max); + vclk_max, + false); if (ret) return ret; } @@ -1069,7 +1072,8 @@ static int smu_v13_0_5_set_performance_level(struct smu_context *smu, ret = smu_v13_0_5_set_soft_freq_limited_range(smu, SMU_DCLK, dclk_min, - dclk_max); + dclk_max, + false); if (ret) return ret; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c index 55ed6247eb61..7ebb675c5786 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c @@ -102,6 +102,24 @@ MODULE_FIRMWARE("amdgpu/smu_13_0_14.bin"); #define MCA_BANK_IPID(_ip, _hwid, _type) \ [AMDGPU_MCA_IP_##_ip] = { .hwid = _hwid, .mcatype = _type, } +static inline bool smu_v13_0_6_is_unified_metrics(struct smu_context *smu) +{ + return (smu->adev->flags & AMD_IS_APU) && + smu->smc_fw_version <= 0x4556900; +} + +static inline bool smu_v13_0_6_is_other_end_count_available(struct smu_context *smu) +{ + switch (amdgpu_ip_version(smu->adev, MP1_HWIP, 0)) { + case IP_VERSION(13, 0, 6): + return smu->smc_fw_version >= 0x557600; + case IP_VERSION(13, 0, 14): + return smu->smc_fw_version >= 0x05550E00; + default: + return false; + } +} + struct mca_bank_ipid { enum amdgpu_mca_ip ip; uint16_t hwid; @@ -253,7 +271,7 @@ struct PPTable_t { #define SMUQ10_TO_UINT(x) ((x) >> 10) #define SMUQ10_FRAC(x) ((x) & 0x3ff) #define SMUQ10_ROUND(x) ((SMUQ10_TO_UINT(x)) + ((SMUQ10_FRAC(x)) >= 0x200)) -#define GET_METRIC_FIELD(field) ((adev->flags & AMD_IS_APU) ?\ +#define GET_METRIC_FIELD(field, flag) ((flag) ?\ (metrics_a->field) : (metrics_x->field)) struct smu_v13_0_6_dpm_map { @@ -352,7 +370,7 @@ static int smu_v13_0_6_tables_init(struct smu_context *smu) return -ENOMEM; smu_table->metrics_time = 0; - smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v1_5); + smu_table->gpu_metrics_table_size = sizeof(struct gpu_metrics_v1_6); smu_table->gpu_metrics_table = kzalloc(smu_table->gpu_metrics_table_size, GFP_KERNEL); if (!smu_table->gpu_metrics_table) { @@ -583,7 +601,7 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) MetricsTableA_t *metrics_a = (MetricsTableA_t *)smu_table->metrics_table; struct PPTable_t *pptable = (struct PPTable_t *)smu_table->driver_pptable; - struct amdgpu_device *adev = smu->adev; + bool flag = smu_v13_0_6_is_unified_metrics(smu); int ret, i, retry = 100; uint32_t table_version; @@ -595,7 +613,7 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) return ret; /* Ensure that metrics have been updated */ - if (GET_METRIC_FIELD(AccumulationCounter)) + if (GET_METRIC_FIELD(AccumulationCounter, flag)) break; usleep_range(1000, 1100); @@ -612,29 +630,29 @@ static int smu_v13_0_6_setup_driver_pptable(struct smu_context *smu) table_version; pptable->MaxSocketPowerLimit = - SMUQ10_ROUND(GET_METRIC_FIELD(MaxSocketPowerLimit)); + SMUQ10_ROUND(GET_METRIC_FIELD(MaxSocketPowerLimit, flag)); pptable->MaxGfxclkFrequency = - SMUQ10_ROUND(GET_METRIC_FIELD(MaxGfxclkFrequency)); + SMUQ10_ROUND(GET_METRIC_FIELD(MaxGfxclkFrequency, flag)); pptable->MinGfxclkFrequency = - SMUQ10_ROUND(GET_METRIC_FIELD(MinGfxclkFrequency)); + SMUQ10_ROUND(GET_METRIC_FIELD(MinGfxclkFrequency, flag)); for (i = 0; i < 4; ++i) { pptable->FclkFrequencyTable[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(FclkFrequencyTable)[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(FclkFrequencyTable, flag)[i]); pptable->UclkFrequencyTable[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(UclkFrequencyTable)[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(UclkFrequencyTable, flag)[i]); pptable->SocclkFrequencyTable[i] = SMUQ10_ROUND( - GET_METRIC_FIELD(SocclkFrequencyTable)[i]); + GET_METRIC_FIELD(SocclkFrequencyTable, flag)[i]); pptable->VclkFrequencyTable[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(VclkFrequencyTable)[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(VclkFrequencyTable, flag)[i]); pptable->DclkFrequencyTable[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(DclkFrequencyTable)[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(DclkFrequencyTable, flag)[i]); pptable->LclkFrequencyTable[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(LclkFrequencyTable)[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(LclkFrequencyTable, flag)[i]); } /* use AID0 serial number by default */ - pptable->PublicSerialNumber_AID = GET_METRIC_FIELD(PublicSerialNumber_AID)[0]; + pptable->PublicSerialNumber_AID = GET_METRIC_FIELD(PublicSerialNumber_AID, flag)[0]; pptable->Init = true; } @@ -957,6 +975,7 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu, struct smu_table_context *smu_table = &smu->smu_table; MetricsTableX_t *metrics_x = (MetricsTableX_t *)smu_table->metrics_table; MetricsTableA_t *metrics_a = (MetricsTableA_t *)smu_table->metrics_table; + bool flag = smu_v13_0_6_is_unified_metrics(smu); struct amdgpu_device *adev = smu->adev; int ret = 0; int xcc_id; @@ -971,50 +990,50 @@ static int smu_v13_0_6_get_smu_metrics_data(struct smu_context *smu, case METRICS_AVERAGE_GFXCLK: if (smu->smc_fw_version >= 0x552F00) { xcc_id = GET_INST(GC, 0); - *value = SMUQ10_ROUND(GET_METRIC_FIELD(GfxclkFrequency)[xcc_id]); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(GfxclkFrequency, flag)[xcc_id]); } else { *value = 0; } break; case METRICS_CURR_SOCCLK: case METRICS_AVERAGE_SOCCLK: - *value = SMUQ10_ROUND(GET_METRIC_FIELD(SocclkFrequency)[0]); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(SocclkFrequency, flag)[0]); break; case METRICS_CURR_UCLK: case METRICS_AVERAGE_UCLK: - *value = SMUQ10_ROUND(GET_METRIC_FIELD(UclkFrequency)); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(UclkFrequency, flag)); break; case METRICS_CURR_VCLK: - *value = SMUQ10_ROUND(GET_METRIC_FIELD(VclkFrequency)[0]); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(VclkFrequency, flag)[0]); break; case METRICS_CURR_DCLK: - *value = SMUQ10_ROUND(GET_METRIC_FIELD(DclkFrequency)[0]); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(DclkFrequency, flag)[0]); break; case METRICS_CURR_FCLK: - *value = SMUQ10_ROUND(GET_METRIC_FIELD(FclkFrequency)); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(FclkFrequency, flag)); break; case METRICS_AVERAGE_GFXACTIVITY: - *value = SMUQ10_ROUND(GET_METRIC_FIELD(SocketGfxBusy)); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(SocketGfxBusy, flag)); break; case METRICS_AVERAGE_MEMACTIVITY: - *value = SMUQ10_ROUND(GET_METRIC_FIELD(DramBandwidthUtilization)); + *value = SMUQ10_ROUND(GET_METRIC_FIELD(DramBandwidthUtilization, flag)); break; case METRICS_CURR_SOCKETPOWER: - *value = SMUQ10_ROUND(GET_METRIC_FIELD(SocketPower)) << 8; + *value = SMUQ10_ROUND(GET_METRIC_FIELD(SocketPower, flag)) << 8; break; case METRICS_TEMPERATURE_HOTSPOT: - *value = SMUQ10_ROUND(GET_METRIC_FIELD(MaxSocketTemperature)) * + *value = SMUQ10_ROUND(GET_METRIC_FIELD(MaxSocketTemperature, flag)) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; case METRICS_TEMPERATURE_MEM: - *value = SMUQ10_ROUND(GET_METRIC_FIELD(MaxHbmTemperature)) * + *value = SMUQ10_ROUND(GET_METRIC_FIELD(MaxHbmTemperature, flag)) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; /* This is the max of all VRs and not just SOC VR. * No need to define another data type for the same. */ case METRICS_TEMPERATURE_VRSOC: - *value = SMUQ10_ROUND(GET_METRIC_FIELD(MaxVrTemperature)) * + *value = SMUQ10_ROUND(GET_METRIC_FIELD(MaxVrTemperature, flag)) * SMU_TEMPERATURE_UNITS_PER_CENTIGRADES; break; default: @@ -1739,7 +1758,7 @@ static int smu_v13_0_6_set_performance_level(struct smu_context *smu, if (uclk_table->max != pstate_table->uclk_pstate.curr.max) { /* Min UCLK is not expected to be changed */ ret = smu_v13_0_set_soft_freq_limited_range( - smu, SMU_UCLK, 0, uclk_table->max); + smu, SMU_UCLK, 0, uclk_table->max, false); if (ret) return ret; pstate_table->uclk_pstate.curr.max = uclk_table->max; @@ -1758,7 +1777,8 @@ static int smu_v13_0_6_set_performance_level(struct smu_context *smu, static int smu_v13_0_6_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, - uint32_t min, uint32_t max) + uint32_t min, uint32_t max, + bool automatic) { struct smu_dpm_context *smu_dpm = &(smu->smu_dpm); struct smu_13_0_dpm_context *dpm_context = smu_dpm->dpm_context; @@ -1806,7 +1826,7 @@ static int smu_v13_0_6_set_soft_freq_limited_range(struct smu_context *smu, return -EOPNOTSUPP; /* Only max clock limiting is allowed for UCLK */ ret = smu_v13_0_set_soft_freq_limited_range( - smu, SMU_UCLK, 0, max); + smu, SMU_UCLK, 0, max, false); if (!ret) pstate_table->uclk_pstate.curr.max = max; } @@ -1946,7 +1966,7 @@ static int smu_v13_0_6_usr_edit_dpm_table(struct smu_context *smu, max_clk = dpm_context->dpm_tables.gfx_table.max; ret = smu_v13_0_6_set_soft_freq_limited_range( - smu, SMU_GFXCLK, min_clk, max_clk); + smu, SMU_GFXCLK, min_clk, max_clk, false); if (ret) return ret; @@ -1954,7 +1974,7 @@ static int smu_v13_0_6_usr_edit_dpm_table(struct smu_context *smu, min_clk = dpm_context->dpm_tables.uclk_table.min; max_clk = dpm_context->dpm_tables.uclk_table.max; ret = smu_v13_0_6_set_soft_freq_limited_range( - smu, SMU_UCLK, min_clk, max_clk); + smu, SMU_UCLK, min_clk, max_clk, false); if (ret) return ret; pstate_table->uclk_pstate.custom.max = 0; @@ -1978,7 +1998,7 @@ static int smu_v13_0_6_usr_edit_dpm_table(struct smu_context *smu, max_clk = pstate_table->gfxclk_pstate.custom.max; ret = smu_v13_0_6_set_soft_freq_limited_range( - smu, SMU_GFXCLK, min_clk, max_clk); + smu, SMU_GFXCLK, min_clk, max_clk, false); if (ret) return ret; @@ -1989,7 +2009,7 @@ static int smu_v13_0_6_usr_edit_dpm_table(struct smu_context *smu, min_clk = pstate_table->uclk_pstate.curr.min; max_clk = pstate_table->uclk_pstate.custom.max; return smu_v13_0_6_set_soft_freq_limited_range( - smu, SMU_UCLK, min_clk, max_clk); + smu, SMU_UCLK, min_clk, max_clk, false); } break; default: @@ -2299,14 +2319,18 @@ static int smu_v13_0_6_get_current_pcie_link_speed(struct smu_context *smu) static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table) { + bool per_inst, smu_13_0_6_per_inst, smu_13_0_14_per_inst, apu_per_inst; struct smu_table_context *smu_table = &smu->smu_table; - struct gpu_metrics_v1_5 *gpu_metrics = - (struct gpu_metrics_v1_5 *)smu_table->gpu_metrics_table; + struct gpu_metrics_v1_6 *gpu_metrics = + (struct gpu_metrics_v1_6 *)smu_table->gpu_metrics_table; + bool flag = smu_v13_0_6_is_unified_metrics(smu); + int ret = 0, xcc_id, inst, i, j, k, idx; struct amdgpu_device *adev = smu->adev; - int ret = 0, xcc_id, inst, i, j; MetricsTableX_t *metrics_x; MetricsTableA_t *metrics_a; + struct amdgpu_xcp *xcp; u16 link_width_level; + u32 inst_mask; metrics_x = kzalloc(max(sizeof(MetricsTableX_t), sizeof(MetricsTableA_t)), GFP_KERNEL); ret = smu_v13_0_6_get_metrics_table(smu, metrics_x, true); @@ -2317,53 +2341,60 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table metrics_a = (MetricsTableA_t *)metrics_x; - smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 5); + smu_cmn_init_soft_gpu_metrics(gpu_metrics, 1, 6); gpu_metrics->temperature_hotspot = - SMUQ10_ROUND(GET_METRIC_FIELD(MaxSocketTemperature)); + SMUQ10_ROUND(GET_METRIC_FIELD(MaxSocketTemperature, flag)); /* Individual HBM stack temperature is not reported */ gpu_metrics->temperature_mem = - SMUQ10_ROUND(GET_METRIC_FIELD(MaxHbmTemperature)); + SMUQ10_ROUND(GET_METRIC_FIELD(MaxHbmTemperature, flag)); /* Reports max temperature of all voltage rails */ gpu_metrics->temperature_vrsoc = - SMUQ10_ROUND(GET_METRIC_FIELD(MaxVrTemperature)); + SMUQ10_ROUND(GET_METRIC_FIELD(MaxVrTemperature, flag)); gpu_metrics->average_gfx_activity = - SMUQ10_ROUND(GET_METRIC_FIELD(SocketGfxBusy)); + SMUQ10_ROUND(GET_METRIC_FIELD(SocketGfxBusy, flag)); gpu_metrics->average_umc_activity = - SMUQ10_ROUND(GET_METRIC_FIELD(DramBandwidthUtilization)); + SMUQ10_ROUND(GET_METRIC_FIELD(DramBandwidthUtilization, flag)); gpu_metrics->curr_socket_power = - SMUQ10_ROUND(GET_METRIC_FIELD(SocketPower)); + SMUQ10_ROUND(GET_METRIC_FIELD(SocketPower, flag)); /* Energy counter reported in 15.259uJ (2^-16) units */ - gpu_metrics->energy_accumulator = GET_METRIC_FIELD(SocketEnergyAcc); + gpu_metrics->energy_accumulator = GET_METRIC_FIELD(SocketEnergyAcc, flag); for (i = 0; i < MAX_GFX_CLKS; i++) { xcc_id = GET_INST(GC, i); if (xcc_id >= 0) gpu_metrics->current_gfxclk[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(GfxclkFrequency)[xcc_id]); + SMUQ10_ROUND(GET_METRIC_FIELD(GfxclkFrequency, flag)[xcc_id]); if (i < MAX_CLKS) { gpu_metrics->current_socclk[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(SocclkFrequency)[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(SocclkFrequency, flag)[i]); inst = GET_INST(VCN, i); if (inst >= 0) { gpu_metrics->current_vclk0[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(VclkFrequency)[inst]); + SMUQ10_ROUND(GET_METRIC_FIELD(VclkFrequency, flag)[inst]); gpu_metrics->current_dclk0[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(DclkFrequency)[inst]); + SMUQ10_ROUND(GET_METRIC_FIELD(DclkFrequency, flag)[inst]); } } } - gpu_metrics->current_uclk = SMUQ10_ROUND(GET_METRIC_FIELD(UclkFrequency)); + gpu_metrics->current_uclk = SMUQ10_ROUND(GET_METRIC_FIELD(UclkFrequency, flag)); - /* Throttle status is not reported through metrics now */ - gpu_metrics->throttle_status = 0; + /* Total accumulated cycle counter */ + gpu_metrics->accumulation_counter = GET_METRIC_FIELD(AccumulationCounter, flag); + + /* Accumulated throttler residencies */ + gpu_metrics->prochot_residency_acc = GET_METRIC_FIELD(ProchotResidencyAcc, flag); + gpu_metrics->ppt_residency_acc = GET_METRIC_FIELD(PptResidencyAcc, flag); + gpu_metrics->socket_thm_residency_acc = GET_METRIC_FIELD(SocketThmResidencyAcc, flag); + gpu_metrics->vr_thm_residency_acc = GET_METRIC_FIELD(VrThmResidencyAcc, flag); + gpu_metrics->hbm_thm_residency_acc = GET_METRIC_FIELD(HbmThmResidencyAcc, flag); /* Clock Lock Status. Each bit corresponds to each GFXCLK instance */ - gpu_metrics->gfxclk_lock_status = GET_METRIC_FIELD(GfxLockXCDMak) >> GET_INST(GC, 0); + gpu_metrics->gfxclk_lock_status = GET_METRIC_FIELD(GfxLockXCDMak, flag) >> GET_INST(GC, 0); if (!(adev->flags & AMD_IS_APU)) { /*Check smu version, PCIE link speed and width will be reported from pmfw metric @@ -2399,41 +2430,77 @@ static ssize_t smu_v13_0_6_get_gpu_metrics(struct smu_context *smu, void **table metrics_x->PCIeNAKSentCountAcc; gpu_metrics->pcie_nak_rcvd_count_acc = metrics_x->PCIeNAKReceivedCountAcc; + if (smu_v13_0_6_is_other_end_count_available(smu)) + gpu_metrics->pcie_lc_perf_other_end_recovery = + metrics_x->PCIeOtherEndRecoveryAcc; + } gpu_metrics->system_clock_counter = ktime_get_boottime_ns(); gpu_metrics->gfx_activity_acc = - SMUQ10_ROUND(GET_METRIC_FIELD(SocketGfxBusyAcc)); + SMUQ10_ROUND(GET_METRIC_FIELD(SocketGfxBusyAcc, flag)); gpu_metrics->mem_activity_acc = - SMUQ10_ROUND(GET_METRIC_FIELD(DramBandwidthUtilizationAcc)); + SMUQ10_ROUND(GET_METRIC_FIELD(DramBandwidthUtilizationAcc, flag)); for (i = 0; i < NUM_XGMI_LINKS; i++) { gpu_metrics->xgmi_read_data_acc[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(XgmiReadDataSizeAcc)[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(XgmiReadDataSizeAcc, flag)[i]); gpu_metrics->xgmi_write_data_acc[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(XgmiWriteDataSizeAcc)[i]); + SMUQ10_ROUND(GET_METRIC_FIELD(XgmiWriteDataSizeAcc, flag)[i]); } - for (i = 0; i < adev->jpeg.num_jpeg_inst; ++i) { - inst = GET_INST(JPEG, i); - for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { - gpu_metrics->jpeg_activity[(i * adev->jpeg.num_jpeg_rings) + j] = - SMUQ10_ROUND(GET_METRIC_FIELD(JpegBusy) - [(inst * adev->jpeg.num_jpeg_rings) + j]); + gpu_metrics->num_partition = adev->xcp_mgr->num_xcps; + + apu_per_inst = (adev->flags & AMD_IS_APU) && (smu->smc_fw_version >= 0x04556A00); + smu_13_0_6_per_inst = !(adev->flags & AMD_IS_APU) && + (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) + == IP_VERSION(13, 0, 6)) && + (smu->smc_fw_version >= 0x556F00); + smu_13_0_14_per_inst = !(adev->flags & AMD_IS_APU) && + (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) + == IP_VERSION(13, 0, 14)) && + (smu->smc_fw_version >= 0x05550B00); + + per_inst = apu_per_inst || smu_13_0_6_per_inst || smu_13_0_14_per_inst; + + for_each_xcp(adev->xcp_mgr, xcp, i) { + amdgpu_xcp_get_inst_details(xcp, AMDGPU_XCP_VCN, &inst_mask); + idx = 0; + for_each_inst(k, inst_mask) { + /* Both JPEG and VCN has same instances */ + inst = GET_INST(VCN, k); + + for (j = 0; j < adev->jpeg.num_jpeg_rings; ++j) { + gpu_metrics->xcp_stats[i].jpeg_busy + [(idx * adev->jpeg.num_jpeg_rings) + j] = + SMUQ10_ROUND(GET_METRIC_FIELD(JpegBusy, flag) + [(inst * adev->jpeg.num_jpeg_rings) + j]); + } + gpu_metrics->xcp_stats[i].vcn_busy[idx] = + SMUQ10_ROUND(GET_METRIC_FIELD(VcnBusy, flag)[inst]); + idx++; + + } + + if (per_inst) { + amdgpu_xcp_get_inst_details(xcp, AMDGPU_XCP_GFX, &inst_mask); + idx = 0; + for_each_inst(k, inst_mask) { + inst = GET_INST(GC, k); + gpu_metrics->xcp_stats[i].gfx_busy_inst[idx] = + SMUQ10_ROUND(metrics_x->GfxBusy[inst]); + gpu_metrics->xcp_stats[i].gfx_busy_acc[idx] = + SMUQ10_ROUND(metrics_x->GfxBusyAcc[inst]); + idx++; + } } } - for (i = 0; i < adev->vcn.num_vcn_inst; ++i) { - inst = GET_INST(VCN, i); - gpu_metrics->vcn_activity[i] = - SMUQ10_ROUND(GET_METRIC_FIELD(VcnBusy)[inst]); - } + gpu_metrics->xgmi_link_width = SMUQ10_ROUND(GET_METRIC_FIELD(XgmiWidth, flag)); + gpu_metrics->xgmi_link_speed = SMUQ10_ROUND(GET_METRIC_FIELD(XgmiBitrate, flag)); - gpu_metrics->xgmi_link_width = SMUQ10_ROUND(GET_METRIC_FIELD(XgmiWidth)); - gpu_metrics->xgmi_link_speed = SMUQ10_ROUND(GET_METRIC_FIELD(XgmiBitrate)); - - gpu_metrics->firmware_timestamp = GET_METRIC_FIELD(Timestamp); + gpu_metrics->firmware_timestamp = GET_METRIC_FIELD(Timestamp, flag); *table = (void *)gpu_metrics; kfree(metrics_x); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index b891a5e0a396..23f13388455f 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -734,19 +734,6 @@ static bool smu_v13_0_7_is_dpm_running(struct smu_context *smu) return !!(feature_enabled & SMC_DPM_FEATURE); } -static void smu_v13_0_7_dump_pptable(struct smu_context *smu) -{ - struct smu_table_context *table_context = &smu->smu_table; - PPTable_t *pptable = table_context->driver_pptable; - SkuTable_t *skutable = &pptable->SkuTable; - - dev_info(smu->adev->dev, "Dumped PPTable:\n"); - - dev_info(smu->adev->dev, "Version = 0x%08x\n", skutable->Version); - dev_info(smu->adev->dev, "FeaturesToRun[0] = 0x%08x\n", skutable->FeaturesToRun[0]); - dev_info(smu->adev->dev, "FeaturesToRun[1] = 0x%08x\n", skutable->FeaturesToRun[1]); -} - static uint32_t smu_v13_0_7_get_throttler_status(SmuMetrics_t *metrics) { uint32_t throttler_status = 0; @@ -1964,7 +1951,8 @@ static int smu_v13_0_7_force_clk_levels(struct smu_context *smu, ret = smu_v13_0_set_soft_freq_limited_range(smu, clk_type, min_freq, - max_freq); + max_freq, + false); break; case SMU_DCEFCLK: case SMU_PCIE: @@ -2605,7 +2593,6 @@ static const struct pptable_funcs smu_v13_0_7_ppt_funcs = { .get_allowed_feature_mask = smu_v13_0_7_get_allowed_feature_mask, .set_default_dpm_table = smu_v13_0_7_set_default_dpm_table, .is_dpm_running = smu_v13_0_7_is_dpm_running, - .dump_pptable = smu_v13_0_7_dump_pptable, .init_microcode = smu_v13_0_init_microcode, .load_microcode = smu_v13_0_load_microcode, .fini_microcode = smu_v13_0_fini_microcode, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c index 260c339f89c5..71d58c8c8cc0 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/yellow_carp_ppt.c @@ -945,9 +945,10 @@ failed: } static int yellow_carp_set_soft_freq_limited_range(struct smu_context *smu, - enum smu_clk_type clk_type, - uint32_t min, - uint32_t max) + enum smu_clk_type clk_type, + uint32_t min, + uint32_t max, + bool automatic) { enum smu_message_type msg_set_min, msg_set_max; uint32_t min_clk = min; @@ -1134,7 +1135,7 @@ static int yellow_carp_force_clk_levels(struct smu_context *smu, if (ret) goto force_level_out; - ret = yellow_carp_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq); + ret = yellow_carp_set_soft_freq_limited_range(smu, clk_type, min_freq, max_freq, false); if (ret) goto force_level_out; break; @@ -1254,9 +1255,10 @@ static int yellow_carp_set_performance_level(struct smu_context *smu, if (sclk_min && sclk_max) { ret = yellow_carp_set_soft_freq_limited_range(smu, - SMU_SCLK, - sclk_min, - sclk_max); + SMU_SCLK, + sclk_min, + sclk_max, + false); if (ret) return ret; @@ -1266,18 +1268,20 @@ static int yellow_carp_set_performance_level(struct smu_context *smu, if (fclk_min && fclk_max) { ret = yellow_carp_set_soft_freq_limited_range(smu, - SMU_FCLK, - fclk_min, - fclk_max); + SMU_FCLK, + fclk_min, + fclk_max, + false); if (ret) return ret; } if (socclk_min && socclk_max) { ret = yellow_carp_set_soft_freq_limited_range(smu, - SMU_SOCCLK, - socclk_min, - socclk_max); + SMU_SOCCLK, + socclk_min, + socclk_max, + false); if (ret) return ret; } @@ -1286,7 +1290,8 @@ static int yellow_carp_set_performance_level(struct smu_context *smu, ret = yellow_carp_set_soft_freq_limited_range(smu, SMU_VCLK, vclk_min, - vclk_max); + vclk_max, + false); if (ret) return ret; } @@ -1295,7 +1300,8 @@ static int yellow_carp_set_performance_level(struct smu_context *smu, ret = yellow_carp_set_soft_freq_limited_range(smu, SMU_DCLK, dclk_min, - dclk_max); + dclk_max, + false); if (ret) return ret; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c index 865e916fc425..f7745eaf118e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0.c @@ -1102,7 +1102,8 @@ failed: int smu_v14_0_set_soft_freq_limited_range(struct smu_context *smu, enum smu_clk_type clk_type, uint32_t min, - uint32_t max) + uint32_t max, + bool automatic) { int ret = 0, clk_id = 0; uint32_t param; @@ -1117,7 +1118,10 @@ int smu_v14_0_set_soft_freq_limited_range(struct smu_context *smu, return clk_id; if (max > 0) { - param = (uint32_t)((clk_id << 16) | (max & 0xffff)); + if (automatic) + param = (uint32_t)((clk_id << 16) | 0xffff); + else + param = (uint32_t)((clk_id << 16) | (max & 0xffff)); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMaxByFreq, param, NULL); if (ret) @@ -1125,7 +1129,10 @@ int smu_v14_0_set_soft_freq_limited_range(struct smu_context *smu, } if (min > 0) { - param = (uint32_t)((clk_id << 16) | (min & 0xffff)); + if (automatic) + param = (uint32_t)((clk_id << 16) | 0); + else + param = (uint32_t)((clk_id << 16) | (min & 0xffff)); ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_SetSoftMinByFreq, param, NULL); if (ret) @@ -1202,6 +1209,7 @@ int smu_v14_0_set_performance_level(struct smu_context *smu, uint32_t dclk_min = 0, dclk_max = 0; uint32_t fclk_min = 0, fclk_max = 0; int ret = 0, i; + bool auto_level = false; switch (level) { case AMD_DPM_FORCED_LEVEL_HIGH: @@ -1233,6 +1241,7 @@ int smu_v14_0_set_performance_level(struct smu_context *smu, dclk_max = dclk_table->max; fclk_min = fclk_table->min; fclk_max = fclk_table->max; + auto_level = true; break; case AMD_DPM_FORCED_LEVEL_PROFILE_STANDARD: sclk_min = sclk_max = pstate_table->gfxclk_pstate.standard; @@ -1268,7 +1277,8 @@ int smu_v14_0_set_performance_level(struct smu_context *smu, ret = smu_v14_0_set_soft_freq_limited_range(smu, SMU_GFXCLK, sclk_min, - sclk_max); + sclk_max, + auto_level); if (ret) return ret; @@ -1280,7 +1290,8 @@ int smu_v14_0_set_performance_level(struct smu_context *smu, ret = smu_v14_0_set_soft_freq_limited_range(smu, SMU_MCLK, mclk_min, - mclk_max); + mclk_max, + auto_level); if (ret) return ret; @@ -1292,7 +1303,8 @@ int smu_v14_0_set_performance_level(struct smu_context *smu, ret = smu_v14_0_set_soft_freq_limited_range(smu, SMU_SOCCLK, socclk_min, - socclk_max); + socclk_max, + auto_level); if (ret) return ret; @@ -1307,7 +1319,8 @@ int smu_v14_0_set_performance_level(struct smu_context *smu, ret = smu_v14_0_set_soft_freq_limited_range(smu, i ? SMU_VCLK1 : SMU_VCLK, vclk_min, - vclk_max); + vclk_max, + auto_level); if (ret) return ret; } @@ -1322,7 +1335,8 @@ int smu_v14_0_set_performance_level(struct smu_context *smu, ret = smu_v14_0_set_soft_freq_limited_range(smu, i ? SMU_DCLK1 : SMU_DCLK, dclk_min, - dclk_max); + dclk_max, + auto_level); if (ret) return ret; } @@ -1334,7 +1348,8 @@ int smu_v14_0_set_performance_level(struct smu_context *smu, ret = smu_v14_0_set_soft_freq_limited_range(smu, SMU_FCLK, fclk_min, - fclk_max); + fclk_max, + auto_level); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c index 5899d01fa73d..cefe10b95d8e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c @@ -732,19 +732,6 @@ static bool smu_v14_0_2_is_dpm_running(struct smu_context *smu) return !!(feature_enabled & SMC_DPM_FEATURE); } -static void smu_v14_0_2_dump_pptable(struct smu_context *smu) -{ - struct smu_table_context *table_context = &smu->smu_table; - PPTable_t *pptable = table_context->driver_pptable; - PFE_Settings_t *PFEsettings = &pptable->PFE_Settings; - - dev_info(smu->adev->dev, "Dumped PPTable:\n"); - - dev_info(smu->adev->dev, "Version = 0x%08x\n", PFEsettings->Version); - dev_info(smu->adev->dev, "FeaturesToRun[0] = 0x%08x\n", PFEsettings->FeaturesToRun[0]); - dev_info(smu->adev->dev, "FeaturesToRun[1] = 0x%08x\n", PFEsettings->FeaturesToRun[1]); -} - static uint32_t smu_v14_0_2_get_throttler_status(SmuMetrics_t *metrics) { uint32_t throttler_status = 0; @@ -1077,12 +1064,9 @@ static void smu_v14_0_2_get_od_setting_limits(struct smu_context *smu, switch (od_feature_bit) { case PP_OD_FEATURE_GFXCLK_FMIN: - od_min_setting = overdrive_lowerlimits->GfxclkFmin; - od_max_setting = overdrive_upperlimits->GfxclkFmin; - break; case PP_OD_FEATURE_GFXCLK_FMAX: - od_min_setting = overdrive_lowerlimits->GfxclkFmax; - od_max_setting = overdrive_upperlimits->GfxclkFmax; + od_min_setting = overdrive_lowerlimits->GfxclkFoffset; + od_max_setting = overdrive_upperlimits->GfxclkFoffset; break; case PP_OD_FEATURE_UCLK_FMIN: od_min_setting = overdrive_lowerlimits->UclkFmin; @@ -1269,10 +1253,16 @@ static int smu_v14_0_2_print_clk_levels(struct smu_context *smu, PP_OD_FEATURE_GFXCLK_BIT)) break; - size += sysfs_emit_at(buf, size, "OD_SCLK:\n"); - size += sysfs_emit_at(buf, size, "0: %uMhz\n1: %uMhz\n", - od_table->OverDriveTable.GfxclkFmin, - od_table->OverDriveTable.GfxclkFmax); + PPTable_t *pptable = smu->smu_table.driver_pptable; + const OverDriveLimits_t * const overdrive_upperlimits = + &pptable->SkuTable.OverDriveLimitsBasicMax; + const OverDriveLimits_t * const overdrive_lowerlimits = + &pptable->SkuTable.OverDriveLimitsBasicMin; + + size += sysfs_emit_at(buf, size, "OD_SCLK_OFFSET:\n"); + size += sysfs_emit_at(buf, size, "0: %dMhz\n1: %uMhz\n", + overdrive_lowerlimits->GfxclkFoffset, + overdrive_upperlimits->GfxclkFoffset); break; case SMU_OD_MCLK: @@ -1414,7 +1404,7 @@ static int smu_v14_0_2_print_clk_levels(struct smu_context *smu, PP_OD_FEATURE_GFXCLK_FMAX, NULL, &max_value); - size += sysfs_emit_at(buf, size, "SCLK: %7uMhz %10uMhz\n", + size += sysfs_emit_at(buf, size, "SCLK_OFFSET: %7dMhz %10uMhz\n", min_value, max_value); } @@ -1516,7 +1506,8 @@ static int smu_v14_0_2_force_clk_levels(struct smu_context *smu, ret = smu_v14_0_set_soft_freq_limited_range(smu, clk_type, min_freq, - max_freq); + max_freq, + false); break; case SMU_DCEFCLK: case SMU_PCIE: @@ -1796,7 +1787,7 @@ static int smu_v14_0_2_set_power_profile_mode(struct smu_context *smu, DpmActivityMonitorCoeffInt_t *activity_monitor = &(activity_monitor_external.DpmActivityMonitorCoeffInt); int workload_type, ret = 0; - + uint32_t current_profile_mode = smu->power_profile_mode; smu->power_profile_mode = input[size]; if (smu->power_profile_mode >= PP_SMC_POWER_PROFILE_COUNT) { @@ -1854,6 +1845,11 @@ static int smu_v14_0_2_set_power_profile_mode(struct smu_context *smu, } } + if (smu->power_profile_mode == PP_SMC_POWER_PROFILE_COMPUTE) + smu_v14_0_deep_sleep_control(smu, false); + else if (current_profile_mode == PP_SMC_POWER_PROFILE_COMPUTE) + smu_v14_0_deep_sleep_control(smu, true); + /* conv PP_SMC_POWER_PROFILE* to WORKLOAD_PPLIB_*_BIT */ workload_type = smu_cmn_to_asic_specific_index(smu, CMN2ASIC_MAPPING_WORKLOAD, @@ -2158,7 +2154,7 @@ static ssize_t smu_v14_0_2_get_gpu_metrics(struct smu_context *smu, gpu_metrics->average_gfx_activity = metrics->AverageGfxActivity; gpu_metrics->average_umc_activity = metrics->AverageUclkActivity; - gpu_metrics->average_mm_activity = max(metrics->Vcn0ActivityPercentage, + gpu_metrics->average_mm_activity = max(metrics->AverageVcn0ActivityPercentage, metrics->Vcn1ActivityPercentage); gpu_metrics->average_socket_power = metrics->AverageSocketPower; @@ -2217,8 +2213,7 @@ static void smu_v14_0_2_dump_od_table(struct smu_context *smu, { struct amdgpu_device *adev = smu->adev; - dev_dbg(adev->dev, "OD: Gfxclk: (%d, %d)\n", od_table->OverDriveTable.GfxclkFmin, - od_table->OverDriveTable.GfxclkFmax); + dev_dbg(adev->dev, "OD: Gfxclk offset: (%d)\n", od_table->OverDriveTable.GfxclkFoffset); dev_dbg(adev->dev, "OD: Uclk: (%d, %d)\n", od_table->OverDriveTable.UclkFmin, od_table->OverDriveTable.UclkFmax); } @@ -2309,10 +2304,8 @@ static int smu_v14_0_2_set_default_od_settings(struct smu_context *smu) memcpy(user_od_table, boot_od_table, sizeof(OverDriveTableExternal_t)); - user_od_table->OverDriveTable.GfxclkFmin = - user_od_table_bak.OverDriveTable.GfxclkFmin; - user_od_table->OverDriveTable.GfxclkFmax = - user_od_table_bak.OverDriveTable.GfxclkFmax; + user_od_table->OverDriveTable.GfxclkFoffset = + user_od_table_bak.OverDriveTable.GfxclkFoffset; user_od_table->OverDriveTable.UclkFmin = user_od_table_bak.OverDriveTable.UclkFmin; user_od_table->OverDriveTable.UclkFmax = @@ -2441,22 +2434,6 @@ static int smu_v14_0_2_od_edit_dpm_table(struct smu_context *smu, } switch (input[i]) { - case 0: - smu_v14_0_2_get_od_setting_limits(smu, - PP_OD_FEATURE_GFXCLK_FMIN, - &minimum, - &maximum); - if (input[i + 1] < minimum || - input[i + 1] > maximum) { - dev_info(adev->dev, "GfxclkFmin (%ld) must be within [%u, %u]!\n", - input[i + 1], minimum, maximum); - return -EINVAL; - } - - od_table->OverDriveTable.GfxclkFmin = input[i + 1]; - od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFXCLK_BIT; - break; - case 1: smu_v14_0_2_get_od_setting_limits(smu, PP_OD_FEATURE_GFXCLK_FMAX, @@ -2469,7 +2446,7 @@ static int smu_v14_0_2_od_edit_dpm_table(struct smu_context *smu, return -EINVAL; } - od_table->OverDriveTable.GfxclkFmax = input[i + 1]; + od_table->OverDriveTable.GfxclkFoffset = input[i + 1]; od_table->OverDriveTable.FeatureCtrlMask |= 1U << PP_OD_FEATURE_GFXCLK_BIT; break; @@ -2480,13 +2457,6 @@ static int smu_v14_0_2_od_edit_dpm_table(struct smu_context *smu, } } - if (od_table->OverDriveTable.GfxclkFmin > od_table->OverDriveTable.GfxclkFmax) { - dev_err(adev->dev, - "Invalid setting: GfxclkFmin(%u) is bigger than GfxclkFmax(%u)\n", - (uint32_t)od_table->OverDriveTable.GfxclkFmin, - (uint32_t)od_table->OverDriveTable.GfxclkFmax); - return -EINVAL; - } break; case PP_OD_EDIT_MCLK_VDDC_TABLE: @@ -2806,7 +2776,6 @@ static const struct pptable_funcs smu_v14_0_2_ppt_funcs = { .i2c_init = smu_v14_0_2_i2c_control_init, .i2c_fini = smu_v14_0_2_i2c_control_fini, .is_dpm_running = smu_v14_0_2_is_dpm_running, - .dump_pptable = smu_v14_0_2_dump_pptable, .init_microcode = smu_v14_0_init_microcode, .load_microcode = smu_v14_0_load_microcode, .fini_microcode = smu_v14_0_fini_microcode, diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index 91ad434bcdae..843f00c9e407 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -1078,6 +1078,9 @@ void smu_cmn_init_soft_gpu_metrics(void *table, uint8_t frev, uint8_t crev) case METRICS_VERSION(1, 5): structure_size = sizeof(struct gpu_metrics_v1_5); break; + case METRICS_VERSION(1, 6): + structure_size = sizeof(struct gpu_metrics_v1_6); + break; case METRICS_VERSION(2, 0): structure_size = sizeof(struct gpu_metrics_v2_0); break; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h b/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h index 6f4d212607d7..c09ecf1a68a0 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_internal.h @@ -78,7 +78,6 @@ #define smu_register_irq_handler(smu) smu_ppt_funcs(register_irq_handler, 0, smu) #define smu_get_dpm_ultimate_freq(smu, param, min, max) smu_ppt_funcs(get_dpm_ultimate_freq, 0, smu, param, min, max) #define smu_asic_set_performance_level(smu, level) smu_ppt_funcs(set_performance_level, -EINVAL, smu, level) -#define smu_dump_pptable(smu) smu_ppt_funcs(dump_pptable, 0, smu) #define smu_update_pcie_parameters(smu, pcie_gen_cap, pcie_width_cap) smu_ppt_funcs(update_pcie_parameters, 0, smu, pcie_gen_cap, pcie_width_cap) #define smu_set_power_source(smu, power_src) smu_ppt_funcs(set_power_source, 0, smu, power_src) #define smu_i2c_init(smu) smu_ppt_funcs(i2c_init, 0, smu) diff --git a/drivers/gpu/drm/arm/Kconfig b/drivers/gpu/drm/arm/Kconfig index ddf20708370f..c901ac00c0c3 100644 --- a/drivers/gpu/drm/arm/Kconfig +++ b/drivers/gpu/drm/arm/Kconfig @@ -6,6 +6,7 @@ config DRM_HDLCD tristate "ARM HDLCD" depends on DRM && OF && (ARM || ARM64 || COMPILE_TEST) depends on COMMON_CLK + select DRM_CLIENT_SELECTION select DRM_KMS_HELPER select DRM_GEM_DMA_HELPER help @@ -27,6 +28,7 @@ config DRM_MALI_DISPLAY tristate "ARM Mali Display Processor" depends on DRM && OF && (ARM || ARM64 || COMPILE_TEST) depends on COMMON_CLK + select DRM_CLIENT_SELECTION select DRM_KMS_HELPER select DRM_GEM_DMA_HELPER select VIDEOMODE_HELPERS diff --git a/drivers/gpu/drm/arm/display/Kconfig b/drivers/gpu/drm/arm/display/Kconfig index 4acc4285a4eb..415c10a6374b 100644 --- a/drivers/gpu/drm/arm/display/Kconfig +++ b/drivers/gpu/drm/arm/display/Kconfig @@ -3,6 +3,7 @@ config DRM_KOMEDA tristate "ARM Komeda display driver" depends on DRM && OF depends on COMMON_CLK + select DRM_CLIENT_SELECTION select DRM_KMS_HELPER select DRM_GEM_DMA_HELPER select VIDEOMODE_HELPERS diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c index 55c3773befde..6d475bb34002 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_drv.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_drv.c @@ -9,7 +9,7 @@ #include #include #include -#include +#include #include #include #include "komeda_dev.h" @@ -84,7 +84,7 @@ static int komeda_platform_probe(struct platform_device *pdev) } dev_set_drvdata(dev, mdrv); - drm_fbdev_dma_setup(&mdrv->kms->base, 32); + drm_client_setup(&mdrv->kms->base, NULL); return 0; diff --git a/drivers/gpu/drm/arm/display/komeda/komeda_kms.c b/drivers/gpu/drm/arm/display/komeda/komeda_kms.c index e5eb5d672bcd..1e7b1fcb2848 100644 --- a/drivers/gpu/drm/arm/display/komeda/komeda_kms.c +++ b/drivers/gpu/drm/arm/display/komeda/komeda_kms.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -58,6 +59,7 @@ static irqreturn_t komeda_kms_irq_handler(int irq, void *data) static const struct drm_driver komeda_kms_driver = { .driver_features = DRIVER_GEM | DRIVER_MODESET | DRIVER_ATOMIC, DRM_GEM_DMA_DRIVER_OPS_WITH_DUMB_CREATE(komeda_gem_dma_dumb_create), + DRM_FBDEV_DMA_DRIVER_OPS, .fops = &komeda_cma_fops, .name = "komeda", .desc = "Arm Komeda Display Processor driver", diff --git a/drivers/gpu/drm/arm/hdlcd_drv.c b/drivers/gpu/drm/arm/hdlcd_drv.c index 32be9e370049..cd4389809d42 100644 --- a/drivers/gpu/drm/arm/hdlcd_drv.c +++ b/drivers/gpu/drm/arm/hdlcd_drv.c @@ -9,6 +9,7 @@ * ARM HDLCD Driver */ +#include #include #include #include @@ -21,8 +22,8 @@ #include #include -#include #include +#include #include #include #include @@ -228,6 +229,7 @@ DEFINE_DRM_GEM_DMA_FOPS(fops); static const struct drm_driver hdlcd_driver = { .driver_features = DRIVER_GEM | DRIVER_MODESET | DRIVER_ATOMIC, DRM_GEM_DMA_DRIVER_OPS, + DRM_FBDEV_DMA_DRIVER_OPS, .fops = &fops, .name = "hdlcd", .desc = "ARM HDLCD Controller DRM", @@ -285,7 +287,7 @@ static int hdlcd_drm_bind(struct device *dev) */ if (hdlcd_read(hdlcd, HDLCD_REG_COMMAND)) { hdlcd_write(hdlcd, HDLCD_REG_COMMAND, 0); - drm_aperture_remove_framebuffers(&hdlcd_driver); + aperture_remove_all_conflicting_devices(hdlcd_driver.name); } drm_mode_config_reset(drm); @@ -299,7 +301,7 @@ static int hdlcd_drm_bind(struct device *dev) if (ret) goto err_register; - drm_fbdev_dma_setup(drm, 32); + drm_client_setup(drm, NULL); return 0; diff --git a/drivers/gpu/drm/arm/malidp_drv.c b/drivers/gpu/drm/arm/malidp_drv.c index 6682131d2910..4cb25004b84f 100644 --- a/drivers/gpu/drm/arm/malidp_drv.c +++ b/drivers/gpu/drm/arm/malidp_drv.c @@ -18,6 +18,7 @@ #include #include +#include #include #include #include @@ -562,6 +563,7 @@ static void malidp_debugfs_init(struct drm_minor *minor) static const struct drm_driver malidp_driver = { .driver_features = DRIVER_GEM | DRIVER_MODESET | DRIVER_ATOMIC, DRM_GEM_DMA_DRIVER_OPS_WITH_DUMB_CREATE(malidp_dumb_create), + DRM_FBDEV_DMA_DRIVER_OPS, #ifdef CONFIG_DEBUG_FS .debugfs_init = malidp_debugfs_init, #endif @@ -852,7 +854,7 @@ static int malidp_bind(struct device *dev) if (ret) goto register_fail; - drm_fbdev_dma_setup(drm, 32); + drm_client_setup(drm, NULL); return 0; diff --git a/drivers/gpu/drm/armada/Kconfig b/drivers/gpu/drm/armada/Kconfig index e5597d7c9ae1..b22c891a670b 100644 --- a/drivers/gpu/drm/armada/Kconfig +++ b/drivers/gpu/drm/armada/Kconfig @@ -2,6 +2,7 @@ config DRM_ARMADA tristate "DRM support for Marvell Armada SoCs" depends on DRM && HAVE_CLK && ARM && MMU + select DRM_CLIENT_SELECTION select DRM_KMS_HELPER select FB_IOMEM_HELPERS if DRM_FBDEV_EMULATION help diff --git a/drivers/gpu/drm/armada/armada_drm.h b/drivers/gpu/drm/armada/armada_drm.h index c303e8c7ff6c..3c0ff221a43b 100644 --- a/drivers/gpu/drm/armada/armada_drm.h +++ b/drivers/gpu/drm/armada/armada_drm.h @@ -16,6 +16,8 @@ struct armada_crtc; struct armada_gem_object; struct clk; struct drm_display_mode; +struct drm_fb_helper; +struct drm_fb_helper_surface_size; static inline void armada_updatel(uint32_t val, uint32_t mask, void __iomem *ptr) @@ -74,10 +76,13 @@ struct armada_private { #define drm_to_armada_dev(dev) container_of(dev, struct armada_private, drm) #if defined(CONFIG_DRM_FBDEV_EMULATION) -void armada_fbdev_setup(struct drm_device *dev); +int armada_fbdev_driver_fbdev_probe(struct drm_fb_helper *fbh, + struct drm_fb_helper_surface_size *sizes); +#define ARMADA_FBDEV_DRIVER_OPS \ + .fbdev_probe = armada_fbdev_driver_fbdev_probe #else -static inline void armada_fbdev_setup(struct drm_device *dev) -{ } +#define ARMADA_FBDEV_DRIVER_OPS \ + .fbdev_probe = NULL #endif int armada_overlay_plane_create(struct drm_device *, unsigned long); diff --git a/drivers/gpu/drm/armada/armada_drv.c b/drivers/gpu/drm/armada/armada_drv.c index e51ecc4f7ef4..5c26f0409478 100644 --- a/drivers/gpu/drm/armada/armada_drv.c +++ b/drivers/gpu/drm/armada/armada_drv.c @@ -3,6 +3,7 @@ * Copyright (C) 2012 Russell King */ +#include #include #include #include @@ -10,8 +11,8 @@ #include #include -#include #include +#include #include #include #include @@ -39,6 +40,7 @@ DEFINE_DRM_GEM_FOPS(armada_drm_fops); static const struct drm_driver armada_drm_driver = { .gem_prime_import = armada_gem_prime_import, .dumb_create = armada_gem_dumb_create, + ARMADA_FBDEV_DRIVER_OPS, .major = 1, .minor = 0, .name = "armada-drm", @@ -91,7 +93,7 @@ static int armada_drm_bind(struct device *dev) } /* Remove early framebuffers */ - ret = drm_aperture_remove_framebuffers(&armada_drm_driver); + ret = aperture_remove_all_conflicting_devices(armada_drm_driver.name); if (ret) { dev_err(dev, "[" DRM_NAME ":%s] can't kick out simple-fb: %d\n", __func__, ret); @@ -137,7 +139,7 @@ static int armada_drm_bind(struct device *dev) armada_drm_debugfs_init(priv->drm.primary); #endif - armada_fbdev_setup(&priv->drm); + drm_client_setup(&priv->drm, NULL); return 0; diff --git a/drivers/gpu/drm/armada/armada_fbdev.c b/drivers/gpu/drm/armada/armada_fbdev.c index d223176912b6..6ee7ce04ee71 100644 --- a/drivers/gpu/drm/armada/armada_fbdev.c +++ b/drivers/gpu/drm/armada/armada_fbdev.c @@ -39,8 +39,10 @@ static const struct fb_ops armada_fb_ops = { .fb_destroy = armada_fbdev_fb_destroy, }; -static int armada_fbdev_create(struct drm_fb_helper *fbh, - struct drm_fb_helper_surface_size *sizes) +static const struct drm_fb_helper_funcs armada_fbdev_helper_funcs; + +int armada_fbdev_driver_fbdev_probe(struct drm_fb_helper *fbh, + struct drm_fb_helper_surface_size *sizes) { struct drm_device *dev = fbh->dev; struct drm_mode_fb_cmd2 mode; @@ -98,6 +100,7 @@ static int armada_fbdev_create(struct drm_fb_helper *fbh, info->fix.smem_len = obj->obj.size; info->screen_size = obj->obj.size; info->screen_base = ptr; + fbh->funcs = &armada_fbdev_helper_funcs; fbh->fb = &dfb->fb; drm_fb_helper_fill_info(info, fbh, sizes); @@ -112,109 +115,3 @@ static int armada_fbdev_create(struct drm_fb_helper *fbh, dfb->fb.funcs->destroy(&dfb->fb); return ret; } - -static int armada_fb_probe(struct drm_fb_helper *fbh, - struct drm_fb_helper_surface_size *sizes) -{ - int ret = 0; - - if (!fbh->fb) { - ret = armada_fbdev_create(fbh, sizes); - if (ret == 0) - ret = 1; - } - return ret; -} - -static const struct drm_fb_helper_funcs armada_fb_helper_funcs = { - .fb_probe = armada_fb_probe, -}; - -/* - * Fbdev client and struct drm_client_funcs - */ - -static void armada_fbdev_client_unregister(struct drm_client_dev *client) -{ - struct drm_fb_helper *fbh = drm_fb_helper_from_client(client); - - if (fbh->info) { - drm_fb_helper_unregister_info(fbh); - } else { - drm_client_release(&fbh->client); - drm_fb_helper_unprepare(fbh); - kfree(fbh); - } -} - -static int armada_fbdev_client_restore(struct drm_client_dev *client) -{ - drm_fb_helper_lastclose(client->dev); - - return 0; -} - -static int armada_fbdev_client_hotplug(struct drm_client_dev *client) -{ - struct drm_fb_helper *fbh = drm_fb_helper_from_client(client); - struct drm_device *dev = client->dev; - int ret; - - if (dev->fb_helper) - return drm_fb_helper_hotplug_event(dev->fb_helper); - - ret = drm_fb_helper_init(dev, fbh); - if (ret) - goto err_drm_err; - - if (!drm_drv_uses_atomic_modeset(dev)) - drm_helper_disable_unused_functions(dev); - - ret = drm_fb_helper_initial_config(fbh); - if (ret) - goto err_drm_fb_helper_fini; - - return 0; - -err_drm_fb_helper_fini: - drm_fb_helper_fini(fbh); -err_drm_err: - drm_err(dev, "armada: Failed to setup fbdev emulation (ret=%d)\n", ret); - return ret; -} - -static const struct drm_client_funcs armada_fbdev_client_funcs = { - .owner = THIS_MODULE, - .unregister = armada_fbdev_client_unregister, - .restore = armada_fbdev_client_restore, - .hotplug = armada_fbdev_client_hotplug, -}; - -void armada_fbdev_setup(struct drm_device *dev) -{ - struct drm_fb_helper *fbh; - int ret; - - drm_WARN(dev, !dev->registered, "Device has not been registered.\n"); - drm_WARN(dev, dev->fb_helper, "fb_helper is already set!\n"); - - fbh = kzalloc(sizeof(*fbh), GFP_KERNEL); - if (!fbh) - return; - drm_fb_helper_prepare(dev, fbh, 32, &armada_fb_helper_funcs); - - ret = drm_client_init(dev, &fbh->client, "fbdev", &armada_fbdev_client_funcs); - if (ret) { - drm_err(dev, "Failed to register client: %d\n", ret); - goto err_drm_client_init; - } - - drm_client_register(&fbh->client); - - return; - -err_drm_client_init: - drm_fb_helper_unprepare(fbh); - kfree(fbh); - return; -} diff --git a/drivers/gpu/drm/aspeed/Kconfig b/drivers/gpu/drm/aspeed/Kconfig index 8137c39b057b..6e68f20aac21 100644 --- a/drivers/gpu/drm/aspeed/Kconfig +++ b/drivers/gpu/drm/aspeed/Kconfig @@ -4,6 +4,7 @@ config DRM_ASPEED_GFX depends on DRM && OF depends on (COMPILE_TEST || ARCH_ASPEED) depends on MMU + select DRM_CLIENT_SELECTION select DRM_KMS_HELPER select DRM_GEM_DMA_HELPER select DMA_CMA if HAVE_DMA_CONTIGUOUS diff --git a/drivers/gpu/drm/aspeed/aspeed_gfx_drv.c b/drivers/gpu/drm/aspeed/aspeed_gfx_drv.c index a7a6b70220eb..109023815fa2 100644 --- a/drivers/gpu/drm/aspeed/aspeed_gfx_drv.c +++ b/drivers/gpu/drm/aspeed/aspeed_gfx_drv.c @@ -14,6 +14,7 @@ #include #include +#include #include #include #include @@ -247,6 +248,7 @@ DEFINE_DRM_GEM_DMA_FOPS(fops); static const struct drm_driver aspeed_gfx_driver = { .driver_features = DRIVER_GEM | DRIVER_MODESET | DRIVER_ATOMIC, DRM_GEM_DMA_DRIVER_OPS, + DRM_FBDEV_DMA_DRIVER_OPS, .fops = &fops, .name = "aspeed-gfx-drm", .desc = "ASPEED GFX DRM", @@ -339,7 +341,7 @@ static int aspeed_gfx_probe(struct platform_device *pdev) if (ret) goto err_unload; - drm_fbdev_dma_setup(&priv->drm, 32); + drm_client_setup(&priv->drm, NULL); return 0; err_unload: diff --git a/drivers/gpu/drm/ast/Kconfig b/drivers/gpu/drm/ast/Kconfig index 563fa7a3b546..da0663542e8a 100644 --- a/drivers/gpu/drm/ast/Kconfig +++ b/drivers/gpu/drm/ast/Kconfig @@ -2,6 +2,7 @@ config DRM_AST tristate "AST server chips" depends on DRM && PCI && MMU + select DRM_CLIENT_SELECTION select DRM_GEM_SHMEM_HELPER select DRM_KMS_HELPER select I2C diff --git a/drivers/gpu/drm/ast/ast_dp.c b/drivers/gpu/drm/ast/ast_dp.c index 00b364f9a71e..0e282b7b167c 100644 --- a/drivers/gpu/drm/ast/ast_dp.c +++ b/drivers/gpu/drm/ast/ast_dp.c @@ -149,28 +149,22 @@ int ast_dp_launch(struct ast_device *ast) return 0; } -static bool ast_dp_power_is_on(struct ast_device *ast) +static bool ast_dp_get_phy_sleep(struct ast_device *ast) { - u8 vgacre3; + u8 vgacre3 = ast_get_index_reg(ast, AST_IO_VGACRI, 0xe3); - vgacre3 = ast_get_index_reg(ast, AST_IO_VGACRI, 0xe3); - - return !(vgacre3 & AST_DP_PHY_SLEEP); + return (vgacre3 & AST_IO_VGACRE3_DP_PHY_SLEEP); } -static void ast_dp_power_on_off(struct drm_device *dev, bool on) +static void ast_dp_set_phy_sleep(struct ast_device *ast, bool sleep) { - struct ast_device *ast = to_ast_device(dev); - // Read and Turn off DP PHY sleep - u8 bE3 = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xE3, AST_DP_VIDEO_ENABLE); + u8 vgacre3 = 0x00; - // Turn on DP PHY sleep - if (!on) - bE3 |= AST_DP_PHY_SLEEP; - - // DP Power on/off - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xE3, (u8) ~AST_DP_PHY_SLEEP, bE3); + if (sleep) + vgacre3 |= AST_IO_VGACRE3_DP_PHY_SLEEP; + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xe3, (u8)~AST_IO_VGACRE3_DP_PHY_SLEEP, + vgacre3); msleep(50); } @@ -192,23 +186,39 @@ static void ast_dp_link_training(struct ast_device *ast) drm_err(dev, "Link training failed\n"); } -static void ast_dp_set_on_off(struct drm_device *dev, bool on) +static bool __ast_dp_wait_enable(struct ast_device *ast, bool enabled) { - struct ast_device *ast = to_ast_device(dev); - u8 video_on_off = on; - u32 i = 0; + u8 vgacrdf_test = 0x00; + u8 vgacrdf; + unsigned int i; - // Video On/Off - ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xE3, (u8) ~AST_DP_VIDEO_ENABLE, on); + if (enabled) + vgacrdf_test |= AST_IO_VGACRDF_DP_VIDEO_ENABLE; - video_on_off <<= 4; - while (ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xDF, - ASTDP_MIRROR_VIDEO_ENABLE) != video_on_off) { - // wait 1 ms - mdelay(1); - if (++i > 200) - break; + for (i = 0; i < 200; ++i) { + if (i) + mdelay(1); + vgacrdf = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xdf, + AST_IO_VGACRDF_DP_VIDEO_ENABLE); + if (vgacrdf == vgacrdf_test) + return true; } + + return false; +} + +static void ast_dp_set_enable(struct ast_device *ast, bool enabled) +{ + struct drm_device *dev = &ast->base; + u8 vgacre3 = 0x00; + + if (enabled) + vgacre3 |= AST_IO_VGACRE3_DP_VIDEO_ENABLE; + + ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xe3, (u8)~AST_IO_VGACRE3_DP_VIDEO_ENABLE, + vgacre3); + + drm_WARN_ON(dev, !__ast_dp_wait_enable(ast, enabled)); } static void ast_dp_set_mode(struct drm_crtc *crtc, struct ast_vbios_mode_info *vbios_mode) @@ -317,26 +327,25 @@ static void ast_astdp_encoder_helper_atomic_mode_set(struct drm_encoder *encoder static void ast_astdp_encoder_helper_atomic_enable(struct drm_encoder *encoder, struct drm_atomic_state *state) { - struct drm_device *dev = encoder->dev; - struct ast_device *ast = to_ast_device(dev); + struct ast_device *ast = to_ast_device(encoder->dev); struct ast_connector *ast_connector = &ast->output.astdp.connector; if (ast_connector->physical_status == connector_status_connected) { - ast_dp_power_on_off(dev, AST_DP_POWER_ON); + ast_dp_set_phy_sleep(ast, false); ast_dp_link_training(ast); ast_wait_for_vretrace(ast); - ast_dp_set_on_off(dev, 1); + ast_dp_set_enable(ast, true); } } static void ast_astdp_encoder_helper_atomic_disable(struct drm_encoder *encoder, struct drm_atomic_state *state) { - struct drm_device *dev = encoder->dev; + struct ast_device *ast = to_ast_device(encoder->dev); - ast_dp_set_on_off(dev, 0); - ast_dp_power_on_off(dev, AST_DP_POWER_OFF); + ast_dp_set_enable(ast, false); + ast_dp_set_phy_sleep(ast, true); } static const struct drm_encoder_helper_funcs ast_astdp_encoder_helper_funcs = { @@ -383,22 +392,21 @@ static int ast_astdp_connector_helper_detect_ctx(struct drm_connector *connector bool force) { struct ast_connector *ast_connector = to_ast_connector(connector); - struct drm_device *dev = connector->dev; struct ast_device *ast = to_ast_device(connector->dev); enum drm_connector_status status = connector_status_disconnected; - bool power_is_on; + bool phy_sleep; mutex_lock(&ast->modeset_lock); - power_is_on = ast_dp_power_is_on(ast); - if (!power_is_on) - ast_dp_power_on_off(dev, true); + phy_sleep = ast_dp_get_phy_sleep(ast); + if (phy_sleep) + ast_dp_set_phy_sleep(ast, false); if (ast_astdp_is_connected(ast)) status = connector_status_connected; - if (!power_is_on && status == connector_status_disconnected) - ast_dp_power_on_off(dev, false); + if (phy_sleep && status == connector_status_disconnected) + ast_dp_set_phy_sleep(ast, true); mutex_unlock(&ast->modeset_lock); @@ -414,6 +422,10 @@ static const struct drm_connector_helper_funcs ast_astdp_connector_helper_funcs .detect_ctx = ast_astdp_connector_helper_detect_ctx, }; +/* + * Output + */ + static const struct drm_connector_funcs ast_astdp_connector_funcs = { .reset = drm_atomic_helper_connector_reset, .fill_modes = drm_helper_probe_single_connector_modes, @@ -422,34 +434,18 @@ static const struct drm_connector_funcs ast_astdp_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int ast_astdp_connector_init(struct drm_device *dev, struct drm_connector *connector) -{ - int ret; - - ret = drm_connector_init(dev, connector, &ast_astdp_connector_funcs, - DRM_MODE_CONNECTOR_DisplayPort); - if (ret) - return ret; - - drm_connector_helper_add(connector, &ast_astdp_connector_helper_funcs); - - connector->interlace_allowed = 0; - connector->doublescan_allowed = 0; - - connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; - - return 0; -} - int ast_astdp_output_init(struct ast_device *ast) { struct drm_device *dev = &ast->base; struct drm_crtc *crtc = &ast->crtc; - struct drm_encoder *encoder = &ast->output.astdp.encoder; - struct ast_connector *ast_connector = &ast->output.astdp.connector; - struct drm_connector *connector = &ast_connector->base; + struct drm_encoder *encoder; + struct ast_connector *ast_connector; + struct drm_connector *connector; int ret; + /* encoder */ + + encoder = &ast->output.astdp.encoder; ret = drm_encoder_init(dev, encoder, &ast_astdp_encoder_funcs, DRM_MODE_ENCODER_TMDS, NULL); if (ret) @@ -458,9 +454,20 @@ int ast_astdp_output_init(struct ast_device *ast) encoder->possible_crtcs = drm_crtc_mask(crtc); - ret = ast_astdp_connector_init(dev, connector); + /* connector */ + + ast_connector = &ast->output.astdp.connector; + connector = &ast_connector->base; + ret = drm_connector_init(dev, connector, &ast_astdp_connector_funcs, + DRM_MODE_CONNECTOR_DisplayPort); if (ret) return ret; + drm_connector_helper_add(connector, &ast_astdp_connector_helper_funcs); + + connector->interlace_allowed = 0; + connector->doublescan_allowed = 0; + connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; + ast_connector->physical_status = connector->status; ret = drm_connector_attach_encoder(connector, encoder); diff --git a/drivers/gpu/drm/ast/ast_dp501.c b/drivers/gpu/drm/ast/ast_dp501.c index e4c636f45082..9e19d8c17730 100644 --- a/drivers/gpu/drm/ast/ast_dp501.c +++ b/drivers/gpu/drm/ast/ast_dp501.c @@ -21,9 +21,9 @@ static void ast_release_firmware(void *data) ast->dp501_fw = NULL; } -static int ast_load_dp501_microcode(struct drm_device *dev) +static int ast_load_dp501_microcode(struct ast_device *ast) { - struct ast_device *ast = to_ast_device(dev); + struct drm_device *dev = &ast->base; int ret; ret = request_firmware(&ast->dp501_fw, "ast_dp501_fw.bin", dev->dev); @@ -109,10 +109,10 @@ static bool wait_fw_ready(struct ast_device *ast) } #endif -static bool ast_write_cmd(struct drm_device *dev, u8 data) +static bool ast_write_cmd(struct ast_device *ast, u8 data) { - struct ast_device *ast = to_ast_device(dev); int retry = 0; + if (wait_nack(ast)) { send_nack(ast); ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0x9a, 0x00, data); @@ -131,10 +131,8 @@ static bool ast_write_cmd(struct drm_device *dev, u8 data) return false; } -static bool ast_write_data(struct drm_device *dev, u8 data) +static bool ast_write_data(struct ast_device *ast, u8 data) { - struct ast_device *ast = to_ast_device(dev); - if (wait_nack(ast)) { send_nack(ast); ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0x9a, 0x00, data); @@ -175,10 +173,10 @@ static void clear_cmd(struct ast_device *ast) } #endif -static void ast_set_dp501_video_output(struct drm_device *dev, u8 mode) +static void ast_set_dp501_video_output(struct ast_device *ast, u8 mode) { - ast_write_cmd(dev, 0x40); - ast_write_data(dev, mode); + ast_write_cmd(ast, 0x40); + ast_write_data(ast, mode); msleep(10); } @@ -188,9 +186,8 @@ static u32 get_fw_base(struct ast_device *ast) return ast_mindwm(ast, 0x1e6e2104) & 0x7fffffff; } -bool ast_backup_fw(struct drm_device *dev, u8 *addr, u32 size) +bool ast_backup_fw(struct ast_device *ast, u8 *addr, u32 size) { - struct ast_device *ast = to_ast_device(dev); u32 i, data; u32 boot_address; @@ -207,9 +204,8 @@ bool ast_backup_fw(struct drm_device *dev, u8 *addr, u32 size) return false; } -static bool ast_launch_m68k(struct drm_device *dev) +static bool ast_launch_m68k(struct ast_device *ast) { - struct ast_device *ast = to_ast_device(dev); u32 i, data, len = 0; u32 boot_address; u8 *fw_addr = NULL; @@ -226,7 +222,7 @@ static bool ast_launch_m68k(struct drm_device *dev) len = 32*1024; } else { if (!ast->dp501_fw && - ast_load_dp501_microcode(dev) < 0) + ast_load_dp501_microcode(ast) < 0) return false; fw_addr = (u8 *)ast->dp501_fw->data; @@ -348,9 +344,8 @@ static int ast_dp512_read_edid_block(void *data, u8 *buf, unsigned int block, si return true; } -static bool ast_init_dvo(struct drm_device *dev) +static bool ast_init_dvo(struct ast_device *ast) { - struct ast_device *ast = to_ast_device(dev); u8 jreg; u32 data; ast_write32(ast, 0xf004, 0x1e6e0000); @@ -421,9 +416,8 @@ static bool ast_init_dvo(struct drm_device *dev) } -static void ast_init_analog(struct drm_device *dev) +static void ast_init_analog(struct ast_device *ast) { - struct ast_device *ast = to_ast_device(dev); u32 data; /* @@ -448,28 +442,28 @@ static void ast_init_analog(struct drm_device *dev) ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x00); } -void ast_init_3rdtx(struct drm_device *dev) +void ast_init_3rdtx(struct ast_device *ast) { - struct ast_device *ast = to_ast_device(dev); - u8 jreg; + u8 vgacrd1; if (IS_AST_GEN4(ast) || IS_AST_GEN5(ast)) { - jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, 0xff); - switch (jreg & 0x0e) { - case 0x04: - ast_init_dvo(dev); + vgacrd1 = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, + AST_IO_VGACRD1_TX_TYPE_MASK); + switch (vgacrd1) { + case AST_IO_VGACRD1_TX_SIL164_VBIOS: + ast_init_dvo(ast); break; - case 0x08: - ast_launch_m68k(dev); + case AST_IO_VGACRD1_TX_DP501_VBIOS: + ast_launch_m68k(ast); break; - case 0x0c: - ast_init_dvo(dev); + case AST_IO_VGACRD1_TX_FW_EMBEDDED_FW: + ast_init_dvo(ast); break; default: - if (ast->tx_chip_types & BIT(AST_TX_SIL164)) - ast_init_dvo(dev); + if (ast->tx_chip == AST_TX_SIL164) + ast_init_dvo(ast); else - ast_init_analog(dev); + ast_init_analog(ast); } } } @@ -485,17 +479,17 @@ static const struct drm_encoder_funcs ast_dp501_encoder_funcs = { static void ast_dp501_encoder_helper_atomic_enable(struct drm_encoder *encoder, struct drm_atomic_state *state) { - struct drm_device *dev = encoder->dev; + struct ast_device *ast = to_ast_device(encoder->dev); - ast_set_dp501_video_output(dev, 1); + ast_set_dp501_video_output(ast, 1); } static void ast_dp501_encoder_helper_atomic_disable(struct drm_encoder *encoder, struct drm_atomic_state *state) { - struct drm_device *dev = encoder->dev; + struct ast_device *ast = to_ast_device(encoder->dev); - ast_set_dp501_video_output(dev, 0); + ast_set_dp501_video_output(ast, 0); } static const struct drm_encoder_helper_funcs ast_dp501_encoder_helper_funcs = { @@ -567,34 +561,22 @@ static const struct drm_connector_funcs ast_dp501_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int ast_dp501_connector_init(struct drm_device *dev, struct drm_connector *connector) -{ - int ret; - - ret = drm_connector_init(dev, connector, &ast_dp501_connector_funcs, - DRM_MODE_CONNECTOR_DisplayPort); - if (ret) - return ret; - - drm_connector_helper_add(connector, &ast_dp501_connector_helper_funcs); - - connector->interlace_allowed = 0; - connector->doublescan_allowed = 0; - - connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; - - return 0; -} +/* + * Output + */ int ast_dp501_output_init(struct ast_device *ast) { struct drm_device *dev = &ast->base; struct drm_crtc *crtc = &ast->crtc; - struct drm_encoder *encoder = &ast->output.dp501.encoder; - struct ast_connector *ast_connector = &ast->output.dp501.connector; - struct drm_connector *connector = &ast_connector->base; + struct drm_encoder *encoder; + struct ast_connector *ast_connector; + struct drm_connector *connector; int ret; + /* encoder */ + + encoder = &ast->output.dp501.encoder; ret = drm_encoder_init(dev, encoder, &ast_dp501_encoder_funcs, DRM_MODE_ENCODER_TMDS, NULL); if (ret) @@ -603,9 +585,20 @@ int ast_dp501_output_init(struct ast_device *ast) encoder->possible_crtcs = drm_crtc_mask(crtc); - ret = ast_dp501_connector_init(dev, connector); + /* connector */ + + ast_connector = &ast->output.dp501.connector; + connector = &ast_connector->base; + ret = drm_connector_init(dev, connector, &ast_dp501_connector_funcs, + DRM_MODE_CONNECTOR_DisplayPort); if (ret) return ret; + drm_connector_helper_add(connector, &ast_dp501_connector_helper_funcs); + + connector->interlace_allowed = 0; + connector->doublescan_allowed = 0; + connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; + ast_connector->physical_status = connector->status; ret = drm_connector_attach_encoder(connector, encoder); diff --git a/drivers/gpu/drm/ast/ast_drv.c b/drivers/gpu/drm/ast/ast_drv.c index 3a908bb015fe..4afe4be072ef 100644 --- a/drivers/gpu/drm/ast/ast_drv.c +++ b/drivers/gpu/drm/ast/ast_drv.c @@ -26,12 +26,13 @@ * Authors: Dave Airlie */ +#include #include #include #include -#include #include +#include #include #include #include @@ -64,7 +65,8 @@ static const struct drm_driver ast_driver = { .minor = DRIVER_MINOR, .patchlevel = DRIVER_PATCHLEVEL, - DRM_GEM_SHMEM_DRIVER_OPS + DRM_GEM_SHMEM_DRIVER_OPS, + DRM_FBDEV_SHMEM_DRIVER_OPS, }; /* @@ -279,7 +281,7 @@ static int ast_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) struct drm_device *drm; bool need_post = false; - ret = drm_aperture_remove_conflicting_pci_framebuffers(pdev, &ast_driver); + ret = aperture_remove_conflicting_pci_devices(pdev, ast_driver.name); if (ret) return ret; @@ -360,7 +362,7 @@ static int ast_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent) if (ret) return ret; - drm_fbdev_shmem_setup(drm, 32); + drm_client_setup(drm, NULL); return 0; } @@ -396,7 +398,7 @@ static int ast_drm_thaw(struct drm_device *dev) ast_enable_vga(ast->ioregs); ast_open_key(ast->ioregs); ast_enable_mmio(dev->dev, ast->ioregs); - ast_post_gpu(dev); + ast_post_gpu(ast); return drm_mode_config_helper_resume(dev); } diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index 91fe07cf7b07..21ce3769bf0d 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -91,11 +91,6 @@ enum ast_tx_chip { AST_TX_ASTDP, }; -#define AST_TX_NONE_BIT BIT(AST_TX_NONE) -#define AST_TX_SIL164_BIT BIT(AST_TX_SIL164) -#define AST_TX_DP501_BIT BIT(AST_TX_DP501) -#define AST_TX_ASTDP_BIT BIT(AST_TX_ASTDP) - enum ast_config_mode { ast_use_p2a, ast_use_dt, @@ -187,10 +182,12 @@ struct ast_device { struct mutex modeset_lock; /* Protects access to modeset I/O registers in ioregs */ + enum ast_tx_chip tx_chip; + struct ast_plane primary_plane; struct ast_plane cursor_plane; struct drm_crtc crtc; - struct { + union { struct { struct drm_encoder encoder; struct ast_connector connector; @@ -211,7 +208,6 @@ struct ast_device { bool support_wide_screen; - unsigned long tx_chip_types; /* bitfield of enum ast_chip_type */ u8 *dp501_fw_addr; const struct firmware *dp501_fw; /* dp501 fw */ }; @@ -407,9 +403,6 @@ int ast_mode_config_init(struct ast_device *ast); #define AST_DP501_LINKRATE 0xf014 #define AST_DP501_EDID_DATA 0xf020 -#define AST_DP_POWER_ON true -#define AST_DP_POWER_OFF false - /* * ASTDP resoultion table: * EX: ASTDP_A_B_C: @@ -453,7 +446,7 @@ int ast_mode_config_init(struct ast_device *ast); int ast_mm_init(struct ast_device *ast); /* ast post */ -void ast_post_gpu(struct drm_device *dev); +void ast_post_gpu(struct ast_device *ast); u32 ast_mindwm(struct ast_device *ast, u32 r); void ast_moutdwm(struct ast_device *ast, u32 r, u32 v); void ast_patch_ahb_2500(void __iomem *regs); @@ -462,8 +455,8 @@ int ast_vga_output_init(struct ast_device *ast); int ast_sil164_output_init(struct ast_device *ast); /* ast dp501 */ -bool ast_backup_fw(struct drm_device *dev, u8 *addr, u32 size); -void ast_init_3rdtx(struct drm_device *dev); +bool ast_backup_fw(struct ast_device *ast, u8 *addr, u32 size); +void ast_init_3rdtx(struct ast_device *ast); int ast_dp501_output_init(struct ast_device *ast); /* aspeed DP */ diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index d836f2a4f9f3..bc37c65305d4 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -68,11 +68,33 @@ static void ast_detect_widescreen(struct ast_device *ast) static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) { + static const char * const info_str[] = { + "analog VGA", + "Sil164 TMDS transmitter", + "DP501 DisplayPort transmitter", + "ASPEED DisplayPort transmitter", + }; + struct drm_device *dev = &ast->base; - u8 jreg; + u8 jreg, vgacrd1; + + /* + * Several of the listed TX chips are not explicitly supported + * by the ast driver. If these exist in real-world devices, they + * are most likely reported as VGA or SIL164 outputs. We warn here + * to get bug reports for these devices. If none come in for some + * time, we can begin to fail device probing on these values. + */ + vgacrd1 = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, AST_IO_VGACRD1_TX_TYPE_MASK); + drm_WARN(dev, vgacrd1 == AST_IO_VGACRD1_TX_ITE66121_VBIOS, + "ITE IT66121 detected, 0x%x, Gen%lu\n", vgacrd1, AST_GEN(ast)); + drm_WARN(dev, vgacrd1 == AST_IO_VGACRD1_TX_CH7003_VBIOS, + "Chrontel CH7003 detected, 0x%x, Gen%lu\n", vgacrd1, AST_GEN(ast)); + drm_WARN(dev, vgacrd1 == AST_IO_VGACRD1_TX_ANX9807_VBIOS, + "Analogix ANX9807 detected, 0x%x, Gen%lu\n", vgacrd1, AST_GEN(ast)); /* Check 3rd Tx option (digital output afaik) */ - ast->tx_chip_types |= AST_TX_NONE_BIT; + ast->tx_chip = AST_TX_NONE; /* * VGACRA3 Enhanced Color Mode Register, check if DVO is already @@ -85,7 +107,7 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) if (!need_post) { jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xff); if (jreg & 0x80) - ast->tx_chip_types = AST_TX_SIL164_BIT; + ast->tx_chip = AST_TX_SIL164; } if (IS_AST_GEN4(ast) || IS_AST_GEN5(ast) || IS_AST_GEN6(ast)) { @@ -94,49 +116,42 @@ static void ast_detect_tx_chip(struct ast_device *ast, bool need_post) * the SOC scratch register #1 bits 11:8 (interestingly marked * as "reserved" in the spec) */ - jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, 0xff); + jreg = ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, + AST_IO_VGACRD1_TX_TYPE_MASK); switch (jreg) { - case 0x04: - ast->tx_chip_types = AST_TX_SIL164_BIT; + case AST_IO_VGACRD1_TX_SIL164_VBIOS: + ast->tx_chip = AST_TX_SIL164; break; - case 0x08: + case AST_IO_VGACRD1_TX_DP501_VBIOS: ast->dp501_fw_addr = drmm_kzalloc(dev, 32*1024, GFP_KERNEL); if (ast->dp501_fw_addr) { /* backup firmware */ - if (ast_backup_fw(dev, ast->dp501_fw_addr, 32*1024)) { + if (ast_backup_fw(ast, ast->dp501_fw_addr, 32*1024)) { drmm_kfree(dev, ast->dp501_fw_addr); ast->dp501_fw_addr = NULL; } } fallthrough; - case 0x0c: - ast->tx_chip_types = AST_TX_DP501_BIT; + case AST_IO_VGACRD1_TX_FW_EMBEDDED_FW: + ast->tx_chip = AST_TX_DP501; } } else if (IS_AST_GEN7(ast)) { - if (ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xD1, TX_TYPE_MASK) == - ASTDP_DPMCU_TX) { + if (ast_get_index_reg_mask(ast, AST_IO_VGACRI, 0xd1, AST_IO_VGACRD1_TX_TYPE_MASK) == + AST_IO_VGACRD1_TX_ASTDP) { int ret = ast_dp_launch(ast); if (!ret) - ast->tx_chip_types = AST_TX_ASTDP_BIT; + ast->tx_chip = AST_TX_ASTDP; } } - /* Print stuff for diagnostic purposes */ - if (ast->tx_chip_types & AST_TX_NONE_BIT) - drm_info(dev, "Using analog VGA\n"); - if (ast->tx_chip_types & AST_TX_SIL164_BIT) - drm_info(dev, "Using Sil164 TMDS transmitter\n"); - if (ast->tx_chip_types & AST_TX_DP501_BIT) - drm_info(dev, "Using DP501 DisplayPort transmitter\n"); - if (ast->tx_chip_types & AST_TX_ASTDP_BIT) - drm_info(dev, "Using ASPEED DisplayPort transmitter\n"); + drm_info(dev, "Using %s\n", info_str[ast->tx_chip]); } -static int ast_get_dram_info(struct drm_device *dev) +static int ast_get_dram_info(struct ast_device *ast) { + struct drm_device *dev = &ast->base; struct device_node *np = dev->dev->of_node; - struct ast_device *ast = to_ast_device(dev); uint32_t mcr_cfg, mcr_scu_mpll, mcr_scu_strap; uint32_t denum, num, div, ref_pll, dsel; @@ -278,7 +293,7 @@ struct drm_device *ast_device_create(struct pci_dev *pdev, ast_detect_widescreen(ast); ast_detect_tx_chip(ast, need_post); - ret = ast_get_dram_info(dev); + ret = ast_get_dram_info(ast); if (ret) return ERR_PTR(ret); @@ -286,7 +301,7 @@ struct drm_device *ast_device_create(struct pci_dev *pdev, ast->mclk, ast->dram_type, ast->dram_bus_width); if (need_post) - ast_post_gpu(dev); + ast_post_gpu(ast); ret = ast_mm_init(ast); if (ret) diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index ed496fb32bf3..9d5321c81e68 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -1287,9 +1287,9 @@ static const struct drm_crtc_funcs ast_crtc_funcs = { .atomic_destroy_state = ast_crtc_atomic_destroy_state, }; -static int ast_crtc_init(struct drm_device *dev) +static int ast_crtc_init(struct ast_device *ast) { - struct ast_device *ast = to_ast_device(dev); + struct drm_device *dev = &ast->base; struct drm_crtc *crtc = &ast->crtc; int ret; @@ -1396,28 +1396,26 @@ int ast_mode_config_init(struct ast_device *ast) if (ret) return ret; - ast_crtc_init(dev); + ret = ast_crtc_init(ast); + if (ret) + return ret; - if (ast->tx_chip_types & AST_TX_NONE_BIT) { + switch (ast->tx_chip) { + case AST_TX_NONE: ret = ast_vga_output_init(ast); - if (ret) - return ret; - } - if (ast->tx_chip_types & AST_TX_SIL164_BIT) { + break; + case AST_TX_SIL164: ret = ast_sil164_output_init(ast); - if (ret) - return ret; - } - if (ast->tx_chip_types & AST_TX_DP501_BIT) { + break; + case AST_TX_DP501: ret = ast_dp501_output_init(ast); - if (ret) - return ret; - } - if (ast->tx_chip_types & AST_TX_ASTDP_BIT) { + break; + case AST_TX_ASTDP: ret = ast_astdp_output_init(ast); - if (ret) - return ret; + break; } + if (ret) + return ret; drm_mode_config_reset(dev); diff --git a/drivers/gpu/drm/ast/ast_post.c b/drivers/gpu/drm/ast/ast_post.c index 65755798ab94..364030f97571 100644 --- a/drivers/gpu/drm/ast/ast_post.c +++ b/drivers/gpu/drm/ast/ast_post.c @@ -34,16 +34,14 @@ #include "ast_dram_tables.h" #include "ast_drv.h" -static void ast_post_chip_2300(struct drm_device *dev); -static void ast_post_chip_2500(struct drm_device *dev); +static void ast_post_chip_2300(struct ast_device *ast); +static void ast_post_chip_2500(struct ast_device *ast); static const u8 extreginfo[] = { 0x0f, 0x04, 0x1c, 0xff }; static const u8 extreginfo_ast2300[] = { 0x0f, 0x04, 0x1f, 0xff }; -static void -ast_set_def_ext_reg(struct drm_device *dev) +static void ast_set_def_ext_reg(struct ast_device *ast) { - struct ast_device *ast = to_ast_device(dev); u8 i, index, reg; const u8 *ext_reg_info; @@ -252,9 +250,8 @@ cbr_start: -static void ast_init_dram_reg(struct drm_device *dev) +static void ast_init_dram_reg(struct ast_device *ast) { - struct ast_device *ast = to_ast_device(dev); u8 j; u32 data, temp, i; const struct ast_dramstruct *dram_reg_info; @@ -343,26 +340,24 @@ static void ast_init_dram_reg(struct drm_device *dev) } while ((j & 0x40) == 0); } -void ast_post_gpu(struct drm_device *dev) +void ast_post_gpu(struct ast_device *ast) { - struct ast_device *ast = to_ast_device(dev); - - ast_set_def_ext_reg(dev); + ast_set_def_ext_reg(ast); if (IS_AST_GEN7(ast)) { - if (ast->tx_chip_types & AST_TX_ASTDP_BIT) + if (ast->tx_chip == AST_TX_ASTDP) ast_dp_launch(ast); } else if (ast->config_mode == ast_use_p2a) { if (IS_AST_GEN6(ast)) - ast_post_chip_2500(dev); + ast_post_chip_2500(ast); else if (IS_AST_GEN5(ast) || IS_AST_GEN4(ast)) - ast_post_chip_2300(dev); + ast_post_chip_2300(ast); else - ast_init_dram_reg(dev); + ast_init_dram_reg(ast); - ast_init_3rdtx(dev); + ast_init_3rdtx(ast); } else { - if (ast->tx_chip_types & AST_TX_SIL164_BIT) + if (ast->tx_chip == AST_TX_SIL164) ast_set_index_reg_mask(ast, AST_IO_VGACRI, 0xa3, 0xcf, 0x80); /* Enable DVO */ } } @@ -1569,9 +1564,8 @@ ddr2_init_start: } -static void ast_post_chip_2300(struct drm_device *dev) +static void ast_post_chip_2300(struct ast_device *ast) { - struct ast_device *ast = to_ast_device(dev); struct ast2300_dram_param param; u32 temp; u8 reg; @@ -2038,9 +2032,9 @@ void ast_patch_ahb_2500(void __iomem *regs) __ast_moutdwm(regs, 0x1e6e207c, 0x08000000); /* clear fast reset */ } -void ast_post_chip_2500(struct drm_device *dev) +void ast_post_chip_2500(struct ast_device *ast) { - struct ast_device *ast = to_ast_device(dev); + struct drm_device *dev = &ast->base; u32 temp; u8 reg; diff --git a/drivers/gpu/drm/ast/ast_reg.h b/drivers/gpu/drm/ast/ast_reg.h index 040961cc1a19..2aadf07d135a 100644 --- a/drivers/gpu/drm/ast/ast_reg.h +++ b/drivers/gpu/drm/ast/ast_reg.h @@ -37,28 +37,29 @@ #define AST_IO_VGACRCB_HWC_16BPP BIT(0) /* set: ARGB4444, cleared: 2bpp palette */ #define AST_IO_VGACRCB_HWC_ENABLED BIT(1) -#define AST_IO_VGACRD1_MCU_FW_EXECUTING BIT(5) +#define AST_IO_VGACRD1_MCU_FW_EXECUTING BIT(5) +/* Display Transmitter Type */ +#define AST_IO_VGACRD1_TX_TYPE_MASK GENMASK(3, 1) +#define AST_IO_VGACRD1_NO_TX 0x00 +#define AST_IO_VGACRD1_TX_ITE66121_VBIOS 0x02 +#define AST_IO_VGACRD1_TX_SIL164_VBIOS 0x04 +#define AST_IO_VGACRD1_TX_CH7003_VBIOS 0x06 +#define AST_IO_VGACRD1_TX_DP501_VBIOS 0x08 +#define AST_IO_VGACRD1_TX_ANX9807_VBIOS 0x0a +#define AST_IO_VGACRD1_TX_FW_EMBEDDED_FW 0x0c /* special case of DP501 */ +#define AST_IO_VGACRD1_TX_ASTDP 0x0e + #define AST_IO_VGACRD7_EDID_VALID_FLAG BIT(0) #define AST_IO_VGACRDC_LINK_SUCCESS BIT(0) #define AST_IO_VGACRDF_HPD BIT(0) +#define AST_IO_VGACRDF_DP_VIDEO_ENABLE BIT(4) /* mirrors AST_IO_VGACRE3_DP_VIDEO_ENABLE */ +#define AST_IO_VGACRE3_DP_VIDEO_ENABLE BIT(0) +#define AST_IO_VGACRE3_DP_PHY_SLEEP BIT(4) #define AST_IO_VGACRE5_EDID_READ_DONE BIT(0) #define AST_IO_VGAIR1_R (0x5A) #define AST_IO_VGAIR1_VREFRESH BIT(3) -/* - * Display Transmitter Type - */ - -#define TX_TYPE_MASK GENMASK(3, 1) -#define NO_TX (0 << 1) -#define ITE66121_VBIOS_TX (1 << 1) -#define SI164_VBIOS_TX (2 << 1) -#define CH7003_VBIOS_TX (3 << 1) -#define DP501_VBIOS_TX (4 << 1) -#define ANX9807_VBIOS_TX (5 << 1) -#define TX_FW_EMBEDDED_FW_TX (6 << 1) -#define ASTDP_DPMCU_TX (7 << 1) #define AST_VRAM_INIT_STATUS_MASK GENMASK(7, 6) //#define AST_VRAM_INIT_BY_BMC BIT(7) @@ -68,18 +69,6 @@ * AST DisplayPort */ -/* Define for Soc scratched reg used on ASTDP */ -#define AST_DP_PHY_SLEEP BIT(4) -#define AST_DP_VIDEO_ENABLE BIT(0) - -/* - * CRDF[b4]: Mirror of AST_DP_VIDEO_ENABLE - * Precondition: A. ~AST_DP_PHY_SLEEP && - * B. DP_HPD && - * C. DP_LINK_SUCCESS - */ -#define ASTDP_MIRROR_VIDEO_ENABLE BIT(4) - /* * ASTDP setmode registers: * CRE0[7:0]: MISC0 ((0x00: 18-bpp) or (0x20: 24-bpp) diff --git a/drivers/gpu/drm/ast/ast_sil164.c b/drivers/gpu/drm/ast/ast_sil164.c index 496c7120e515..be01254dd48a 100644 --- a/drivers/gpu/drm/ast/ast_sil164.c +++ b/drivers/gpu/drm/ast/ast_sil164.c @@ -29,6 +29,8 @@ static int ast_sil164_connector_helper_get_modes(struct drm_connector *connector if (ast_connector->physical_status == connector_status_connected) { count = drm_connector_helper_get_modes(connector); } else { + drm_edid_connector_update(connector, NULL); + /* * There's no EDID data without a connected monitor. Set BMC- * compatible modes in this case. The XGA default resolution @@ -71,52 +73,49 @@ static const struct drm_connector_funcs ast_sil164_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int ast_sil164_connector_init(struct drm_device *dev, struct drm_connector *connector) -{ - struct ast_device *ast = to_ast_device(dev); - struct i2c_adapter *ddc; - int ret; - - ddc = ast_ddc_create(ast); - if (IS_ERR(ddc)) { - ret = PTR_ERR(ddc); - drm_err(dev, "failed to add DDC bus for connector; ret=%d\n", ret); - return ret; - } - - ret = drm_connector_init_with_ddc(dev, connector, &ast_sil164_connector_funcs, - DRM_MODE_CONNECTOR_DVII, ddc); - if (ret) - return ret; - - drm_connector_helper_add(connector, &ast_sil164_connector_helper_funcs); - - connector->interlace_allowed = 0; - connector->doublescan_allowed = 0; - - connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; - - return 0; -} +/* + * Output + */ int ast_sil164_output_init(struct ast_device *ast) { struct drm_device *dev = &ast->base; struct drm_crtc *crtc = &ast->crtc; - struct drm_encoder *encoder = &ast->output.sil164.encoder; - struct ast_connector *ast_connector = &ast->output.sil164.connector; - struct drm_connector *connector = &ast_connector->base; + struct i2c_adapter *ddc; + struct drm_encoder *encoder; + struct ast_connector *ast_connector; + struct drm_connector *connector; int ret; + /* DDC */ + + ddc = ast_ddc_create(ast); + if (IS_ERR(ddc)) + return PTR_ERR(ddc); + + /* encoder */ + + encoder = &ast->output.sil164.encoder; ret = drm_encoder_init(dev, encoder, &ast_sil164_encoder_funcs, DRM_MODE_ENCODER_TMDS, NULL); if (ret) return ret; encoder->possible_crtcs = drm_crtc_mask(crtc); - ret = ast_sil164_connector_init(dev, connector); + /* connector */ + + ast_connector = &ast->output.sil164.connector; + connector = &ast_connector->base; + ret = drm_connector_init_with_ddc(dev, connector, &ast_sil164_connector_funcs, + DRM_MODE_CONNECTOR_DVII, ddc); if (ret) return ret; + drm_connector_helper_add(connector, &ast_sil164_connector_helper_funcs); + + connector->interlace_allowed = 0; + connector->doublescan_allowed = 0; + connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; + ast_connector->physical_status = connector->status; ret = drm_connector_attach_encoder(connector, encoder); diff --git a/drivers/gpu/drm/ast/ast_vga.c b/drivers/gpu/drm/ast/ast_vga.c index 3e815da43fbd..abe0fff8485c 100644 --- a/drivers/gpu/drm/ast/ast_vga.c +++ b/drivers/gpu/drm/ast/ast_vga.c @@ -29,6 +29,8 @@ static int ast_vga_connector_helper_get_modes(struct drm_connector *connector) if (ast_connector->physical_status == connector_status_connected) { count = drm_connector_helper_get_modes(connector); } else { + drm_edid_connector_update(connector, NULL); + /* * There's no EDID data without a connected monitor. Set BMC- * compatible modes in this case. The XGA default resolution @@ -71,52 +73,49 @@ static const struct drm_connector_funcs ast_vga_connector_funcs = { .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, }; -static int ast_vga_connector_init(struct drm_device *dev, struct drm_connector *connector) -{ - struct ast_device *ast = to_ast_device(dev); - struct i2c_adapter *ddc; - int ret; - - ddc = ast_ddc_create(ast); - if (IS_ERR(ddc)) { - ret = PTR_ERR(ddc); - drm_err(dev, "failed to add DDC bus for connector; ret=%d\n", ret); - return ret; - } - - ret = drm_connector_init_with_ddc(dev, connector, &ast_vga_connector_funcs, - DRM_MODE_CONNECTOR_VGA, ddc); - if (ret) - return ret; - - drm_connector_helper_add(connector, &ast_vga_connector_helper_funcs); - - connector->interlace_allowed = 0; - connector->doublescan_allowed = 0; - - connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; - - return 0; -} +/* + * Output + */ int ast_vga_output_init(struct ast_device *ast) { struct drm_device *dev = &ast->base; struct drm_crtc *crtc = &ast->crtc; - struct drm_encoder *encoder = &ast->output.vga.encoder; - struct ast_connector *ast_connector = &ast->output.vga.connector; - struct drm_connector *connector = &ast_connector->base; + struct i2c_adapter *ddc; + struct drm_encoder *encoder; + struct ast_connector *ast_connector; + struct drm_connector *connector; int ret; + /* DDC */ + + ddc = ast_ddc_create(ast); + if (IS_ERR(ddc)) + return PTR_ERR(ddc); + + /* encoder */ + + encoder = &ast->output.vga.encoder; ret = drm_encoder_init(dev, encoder, &ast_vga_encoder_funcs, DRM_MODE_ENCODER_DAC, NULL); if (ret) return ret; encoder->possible_crtcs = drm_crtc_mask(crtc); - ret = ast_vga_connector_init(dev, connector); + /* connector */ + + ast_connector = &ast->output.vga.connector; + connector = &ast_connector->base; + ret = drm_connector_init_with_ddc(dev, connector, &ast_vga_connector_funcs, + DRM_MODE_CONNECTOR_VGA, ddc); if (ret) return ret; + drm_connector_helper_add(connector, &ast_vga_connector_helper_funcs); + + connector->interlace_allowed = 0; + connector->doublescan_allowed = 0; + connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; + ast_connector->physical_status = connector->status; ret = drm_connector_attach_encoder(connector, encoder); diff --git a/drivers/gpu/drm/atmel-hlcdc/Kconfig b/drivers/gpu/drm/atmel-hlcdc/Kconfig index 945f3aa7bb24..f8b9c91907d8 100644 --- a/drivers/gpu/drm/atmel-hlcdc/Kconfig +++ b/drivers/gpu/drm/atmel-hlcdc/Kconfig @@ -2,6 +2,7 @@ config DRM_ATMEL_HLCDC tristate "DRM Support for ATMEL HLCDC Display Controller" depends on DRM && OF && COMMON_CLK && ((MFD_ATMEL_HLCDC && ARM) || COMPILE_TEST) + select DRM_CLIENT_SELECTION select DRM_GEM_DMA_HELPER select DRM_KMS_HELPER select DRM_PANEL diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c index 9ce429f889ca..792dcc19e8e7 100644 --- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c +++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c @@ -18,8 +18,10 @@ #include #include +#include #include #include +#include #include #include #include @@ -840,6 +842,7 @@ DEFINE_DRM_GEM_DMA_FOPS(fops); static const struct drm_driver atmel_hlcdc_dc_driver = { .driver_features = DRIVER_GEM | DRIVER_MODESET | DRIVER_ATOMIC, DRM_GEM_DMA_DRIVER_OPS, + DRM_FBDEV_DMA_DRIVER_OPS, .fops = &fops, .name = "atmel-hlcdc", .desc = "Atmel HLCD Controller DRM", @@ -865,7 +868,7 @@ static int atmel_hlcdc_dc_drm_probe(struct platform_device *pdev) if (ret) goto err_unload; - drm_fbdev_dma_setup(ddev, 24); + drm_client_setup_with_fourcc(ddev, DRM_FORMAT_RGB888); return 0; diff --git a/drivers/gpu/drm/bridge/Kconfig b/drivers/gpu/drm/bridge/Kconfig index 3eb955333c80..609f4d0ac93d 100644 --- a/drivers/gpu/drm/bridge/Kconfig +++ b/drivers/gpu/drm/bridge/Kconfig @@ -140,6 +140,8 @@ config DRM_LONTIUM_LT9611 select DRM_PANEL_BRIDGE select DRM_KMS_HELPER select DRM_MIPI_DSI + select DRM_DISPLAY_HELPER + select DRM_DISPLAY_HDMI_STATE_HELPER select REGMAP_I2C help Driver for Lontium LT9611 DSI to HDMI bridge @@ -368,6 +370,13 @@ config DRM_TI_DLPC3433 It supports up to 720p resolution with 60 and 120 Hz refresh rates. +config DRM_TI_TDP158 + tristate "TI TDP158 HDMI/TMDS bridge" + depends on OF + select DRM_PANEL_BRIDGE + help + Texas Instruments TDP158 HDMI/TMDS Bridge driver + config DRM_TI_TFP410 tristate "TI TFP410 DVI/HDMI bridge" depends on OF diff --git a/drivers/gpu/drm/bridge/Makefile b/drivers/gpu/drm/bridge/Makefile index 7df87b582dca..3daf803ce80b 100644 --- a/drivers/gpu/drm/bridge/Makefile +++ b/drivers/gpu/drm/bridge/Makefile @@ -32,6 +32,7 @@ obj-$(CONFIG_DRM_I2C_ADV7511) += adv7511/ obj-$(CONFIG_DRM_TI_DLPC3433) += ti-dlpc3433.o obj-$(CONFIG_DRM_TI_SN65DSI83) += ti-sn65dsi83.o obj-$(CONFIG_DRM_TI_SN65DSI86) += ti-sn65dsi86.o +obj-$(CONFIG_DRM_TI_TDP158) += ti-tdp158.o obj-$(CONFIG_DRM_TI_TFP410) += ti-tfp410.o obj-$(CONFIG_DRM_TI_TPD12S015) += ti-tpd12s015.o obj-$(CONFIG_DRM_NWL_MIPI_DSI) += nwl-dsi.o diff --git a/drivers/gpu/drm/bridge/analogix/anx7625.c b/drivers/gpu/drm/bridge/analogix/anx7625.c index a2e9bb485c36..a2675b121fe4 100644 --- a/drivers/gpu/drm/bridge/analogix/anx7625.c +++ b/drivers/gpu/drm/bridge/analogix/anx7625.c @@ -2551,6 +2551,8 @@ static int __maybe_unused anx7625_runtime_pm_suspend(struct device *dev) mutex_lock(&ctx->lock); anx7625_stop_dp_work(ctx); + if (!ctx->pdata.panel_bridge) + anx7625_remove_edid(ctx); anx7625_power_standby(ctx); mutex_unlock(&ctx->lock); diff --git a/drivers/gpu/drm/bridge/aux-bridge.c b/drivers/gpu/drm/bridge/aux-bridge.c index b29980f95379..015983c015e5 100644 --- a/drivers/gpu/drm/bridge/aux-bridge.c +++ b/drivers/gpu/drm/bridge/aux-bridge.c @@ -58,9 +58,10 @@ int drm_aux_bridge_register(struct device *parent) adev->id = ret; adev->name = "aux_bridge"; adev->dev.parent = parent; - adev->dev.of_node = of_node_get(parent->of_node); adev->dev.release = drm_aux_bridge_release; + device_set_of_node_from_dev(&adev->dev, parent); + ret = auxiliary_device_init(adev); if (ret) { ida_free(&drm_aux_bridge_ida, adev->id); @@ -120,6 +121,10 @@ static int drm_aux_bridge_probe(struct auxiliary_device *auxdev, data->bridge.funcs = &drm_aux_bridge_funcs; data->bridge.of_node = data->dev->of_node; + /* passthrough data, allow everything */ + data->bridge.interlace_allowed = true; + data->bridge.ycbcr_420_allowed = true; + return devm_drm_bridge_add(data->dev, &data->bridge); } diff --git a/drivers/gpu/drm/bridge/aux-hpd-bridge.c b/drivers/gpu/drm/bridge/aux-hpd-bridge.c index 6886db2d9e00..48f297c78ee6 100644 --- a/drivers/gpu/drm/bridge/aux-hpd-bridge.c +++ b/drivers/gpu/drm/bridge/aux-hpd-bridge.c @@ -180,6 +180,10 @@ static int drm_aux_hpd_bridge_probe(struct auxiliary_device *auxdev, data->bridge.ops = DRM_BRIDGE_OP_HPD; data->bridge.type = id->driver_data; + /* passthrough data, allow everything */ + data->bridge.interlace_allowed = true; + data->bridge.ycbcr_420_allowed = true; + auxiliary_set_drvdata(auxdev, data); return devm_drm_bridge_add(data->dev, &data->bridge); diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c index dee640ab1d3a..41f72d458487 100644 --- a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c +++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-core.c @@ -47,7 +47,7 @@ #include #include -#include +#include #include "cdns-mhdp8546-core.h" #include "cdns-mhdp8546-hdcp.h" diff --git a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-hdcp.c b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-hdcp.c index 5e3b8edcf794..31832ba4017f 100644 --- a/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-hdcp.c +++ b/drivers/gpu/drm/bridge/cadence/cdns-mhdp8546-hdcp.c @@ -9,7 +9,7 @@ #include #include -#include +#include #include diff --git a/drivers/gpu/drm/bridge/display-connector.c b/drivers/gpu/drm/bridge/display-connector.c index ab8e00baf3f1..aab9ce7be94c 100644 --- a/drivers/gpu/drm/bridge/display-connector.c +++ b/drivers/gpu/drm/bridge/display-connector.c @@ -270,6 +270,10 @@ static int display_connector_probe(struct platform_device *pdev) /* All the supported connector types support interlaced modes. */ conn->bridge.interlace_allowed = true; + if (type == DRM_MODE_CONNECTOR_HDMIA || + type == DRM_MODE_CONNECTOR_DisplayPort) + conn->bridge.ycbcr_420_allowed = true; + /* Get the optional connector label. */ of_property_read_string(pdev->dev.of_node, "label", &label); diff --git a/drivers/gpu/drm/bridge/imx/Kconfig b/drivers/gpu/drm/bridge/imx/Kconfig index 8dd89efa8ea7..9a480c6abb85 100644 --- a/drivers/gpu/drm/bridge/imx/Kconfig +++ b/drivers/gpu/drm/bridge/imx/Kconfig @@ -3,6 +3,16 @@ if ARCH_MXC || COMPILE_TEST config DRM_IMX_LDB_HELPER tristate +config DRM_IMX_LEGACY_BRIDGE + tristate + depends on DRM_IMX + help + This is a DRM bridge implementation for the DRM i.MX IPUv3 driver, + that uses of_get_drm_display_mode to acquire display mode. + + Newer designs should not use this bridge and should use proper panel + driver instead. + config DRM_IMX8MP_DW_HDMI_BRIDGE tristate "Freescale i.MX8MP HDMI-TX bridge support" depends on OF diff --git a/drivers/gpu/drm/bridge/imx/Makefile b/drivers/gpu/drm/bridge/imx/Makefile index edb0a7b71b30..dd5d48584806 100644 --- a/drivers/gpu/drm/bridge/imx/Makefile +++ b/drivers/gpu/drm/bridge/imx/Makefile @@ -1,4 +1,5 @@ obj-$(CONFIG_DRM_IMX_LDB_HELPER) += imx-ldb-helper.o +obj-$(CONFIG_DRM_IMX_LEGACY_BRIDGE) += imx-legacy-bridge.o obj-$(CONFIG_DRM_IMX8MP_DW_HDMI_BRIDGE) += imx8mp-hdmi-tx.o obj-$(CONFIG_DRM_IMX8MP_HDMI_PVI) += imx8mp-hdmi-pvi.o obj-$(CONFIG_DRM_IMX8QM_LDB) += imx8qm-ldb.o diff --git a/drivers/gpu/drm/bridge/imx/imx-legacy-bridge.c b/drivers/gpu/drm/bridge/imx/imx-legacy-bridge.c new file mode 100644 index 000000000000..3ebf0b9866de --- /dev/null +++ b/drivers/gpu/drm/bridge/imx/imx-legacy-bridge.c @@ -0,0 +1,88 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* + * Freescale i.MX drm driver + * + * bridge driver for legacy DT bindings, utilizing display-timings node + */ + +#include +#include +#include +#include + +#include