Race condition occurs when CS fence completes and multi CS did not
completed yet, while waiting for multi CS ends and returns indication
to user that the CS completed. Next wait for multi CS may be triggered
by previous multi CS completion without any current CS completed,
causing an error.
Example scenario :
1. User do multi CS wait for CSs 1 and 2 on master QID 0
2. CS 1 and 2 reached the "cs release" code. The thread of CS 1
completed both the CS and multi CS handling but the completion
thread of CS 2 completed the CS but still did not executed
complete_multi_cs (note that in CS completion the sequence is to
first do complete all for the CS and then another complete all to
signal the multi_cs)
3. User received indication that CS 1 and 2 completed (since we check
the CS fence and both indicated as completed) and immediately waits
on CS 3 and 4, also on master QID 0.
4. Completion thread of CS2 executed complete_multi_cs before
completion of CS 3 and 4 and so will trigger the multi CS wait of
CSs 3 and 4 as they wait on master QID 0.
This will trigger multi CS completion although none of its
current CS has been completed.
Fixed by adding multi CS complete handling indication for each CS.
CS will be marked to the user as completed only if its fence completed
and multi CS handling is done.
Signed-off-by: Dani Liberman <dliberman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
There may be a situation where drivers receives continuous fatal H/W
error events from FW immediately post reset cycle.
This may be due to some fault on the silicon itself.
In such case its better to bypass reset cycle so we won't be stuck in
endless loop of resets.
This commit bypasses reset request in case driver received two back to
back FW fatal error before first occurrence of heartbeat event.
Signed-off-by: Bharat Jauhari <bjauhari@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Taking an accurate timestamp in a close proximity of the interrupt is
required for user side statistics management.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
The driver allows only a single process to open a device's FD at any
single time. This is done by checking "hdev->compute_ctx" under mutex.
Therefore, to prevent a race between the moment a user closes it's FD
and when another user tries to open the device, we need to make sure
that clearing this variable is the very last thing that is done in the
code of the FD's release.
I'm moving the idle check before clearing this variable and the
"reset on device release". btw, if the reset happens it will prevent
any other user from opening the device until the reset is finished.
An important thing to note is that we need to remove the user process
that is closing the device from the process list BEFORE calling the
reset function. That is to prevent a case where the reset code will
try to kill that user process and it is unnecessary as the process
doesn't hold any device/driver resources anymore.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reset to the device is not necessarily due to an error, so print it
as info instead of error.
In addition, print the type of reset we are doing:
- reset of the entire device (aka hard reset)
- reset of the device after user have released it (less than hard reset)
- lighter reset of an inference device (aka soft reset)
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Soft-reset is the procedure where we reset only the compute/DMA engines
of the device, without requiring the current user-space process to
release the device.
This type of reset can happen if TDR event occurred (a workload got
stuck) or by a root request through sysfs.
This is only relevant for inference ASICs, as there is no real-world
use-case to do that in training, because training runs on multiple
devices.
In addition, we also do (in certain ASICs) a reset upon device release.
That reset uses the same code as the soft-reset.
Therefore, to better differentiate between the two resets, it is better
to rename the soft-reset support as "inference soft-reset", to make
the code more self-explanatory.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
The translation in debugfs of device memory MMU virtual addresses was
wrong as it did not take into consideration the fact that the page
sizes there can be not power of 2.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In order to avoid user target value wraparound, we modify the
current interface so user will be able to wait for an 8-byte
target value rather than a 4-byte value.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
During TDR handling, we check multiple times if CS is valid.
No need to perform this check as CS must be valid at all time
during the TDR handling.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Add support to retrieve following power info via HWMON:
- instantaneous power value
- highest value since last reset
- reset the highest place holder
Signed-off-by: Rajaravi Krishna Katta <rkatta@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Instead of having dedicated function per message that we want to send
to the firmware in COMMS protocol, have a generic function that we can
call to from other parts of the driver
Signed-off-by: Alon Mizrahi <amizrahi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Instead of using the Linux kernel HWMON enums definition when
communicating with the firmware, use proprietary HWMON based enums
i.e. map hwmon.h header enum to cpucp_if.h based enum while.
This is needed because the HWMON enums are not forcing backward
compatibility and therefore changes can break compatibility between
newer driver and older firmware.
The driver will check for CPU_BOOT_DEV_STS0_MAP_HWMON_EN bit to
validate if f/w supports cpucp->hwmon enum mapping to support older
firmware where this mapping won't be available.
Signed-off-by: Rajaravi Krishna Katta <rkatta@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Command submission timeout is currently determined during driver
loading time. As some environments requires this timeout to be
modified in runtime, we introduce a new debugfs node that controls
the timeout value without the need to reload the driver.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Pull libata fixes from Damien Le Moal:
"Two fixes for this cycle:
- Fix a null pointer dereference in ahci-platform driver (from Hai)
- Fix uninitialized variables in pata_legacy driver (from Dan)"
* tag 'libata-5.15-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/libata:
ata: ahci_platform: fix null-ptr-deref in ahci_platform_enable_regulators()
pata_legacy: fix a couple uninitialized variable bugs
Pull block fixes from Jens Axboe:
"Bigger than usual for this point in time, the majority is fixing some
issues around BDI lifetimes with the move from the request_queue to
the disk in this release. In detail:
- Series on draining fs IO for del_gendisk() (Christoph)
- NVMe pull request via Christoph:
- fix the abort command id (Keith Busch)
- nvme: fix per-namespace chardev deletion (Adam Manzanares)
- brd locking scope fix (Tetsuo)
- BFQ fix (Paolo)"
* tag 'block-5.15-2021-10-17' of git://git.kernel.dk/linux-block:
block, bfq: reset last_bfqq_created on group change
block: warn when putting the final reference on a registered disk
brd: reduce the brd_devices_mutex scope
kyber: avoid q->disk dereferences in trace points
block: keep q_usage_counter in atomic mode after del_gendisk
block: drain file system I/O on del_gendisk
block: split bio_queue_enter from blk_queue_enter
block: factor out a blk_try_enter_queue helper
block: call submit_bio_checks under q_usage_counter
nvme: fix per-namespace chardev deletion
block/rnbd-clt-sysfs: fix a couple uninitialized variable bugs
nvme-pci: Fix abort command id
Pull io_uring fix from Jens Axboe:
"Just a single fix for a wrong condition for grabbing a lock, a
regression in this merge window"
* tag 'io_uring-5.15-2021-10-17' of git://git.kernel.dk/linux-block:
io_uring: fix wrong condition to grab uring lock
Pull virtio fixes from Michael Tsirkin:
"Fixes up some issues in rc5"
* tag 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost:
vhost-vdpa: Fix the wrong input in config_cb
VDUSE: fix documentation underline warning
Revert "virtio-blk: Add validation for block size in config space"
vhost_vdpa: unset vq irq before freeing irq
virtio: write back F_VERSION_1 before validate
Pull powerpc fixes from Michael Ellerman:
- Fix a bug where guests on P9 with interrupts passed through could get
stuck in synchronize_irq().
- Fix a bug in KVM on P8 where secondary threads entering a guest would
write outside their allocated stack.
- Fix a bug in KVM on P8 where secondary threads could confuse the host
offline code and cause the guest or host to crash.
Thanks to Cédric Le Goater.
* tag 'powerpc-5.15-4' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
KVM: PPC: Book3S HV: Make idle_kvm_start_guest() return 0 if it went to guest
KVM: PPC: Book3S HV: Fix stack handling in idle_kvm_start_guest()
powerpc/xive: Discard disabled interrupts in get_irqchip_state()
Pull objtool fixes from Borislav Petkov:
- Update section headers before the respective relocations to not
trigger a safety check in elftoolchain's implementation of libelf
- Do not add garbage data to the .rela.orc_unwind_ip section
* tag 'objtool_urgent_for_v5.15_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
objtool: Update section header before relocations
objtool: Check for gelf_update_rel[a] failures
Pull EDAC fix from Borislav Petkov:
- Log the "correct" uncorrectable error count in the armada_xp driver
* tag 'edac_urgent_for_v5.15_rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
EDAC/armada-xp: Fix output of uncorrectable error counter
Dani Liberman
Oded Gabbay
Bharat Jauhari
Yuri Nudelman
Ofir Bitton
Rajaravi Krishna Katta
Alon Mizrahi
Greg Kroah-Hartman
Linus Torvalds