Adds base i915 perf infrastructure for Gen performance metrics.
This adds a DRM_IOCTL_I915_PERF_OPEN ioctl that takes an array of uint64
properties to configure a stream of metrics and returns a new fd usable
with standard VFS system calls including read() to read typed and sized
records; ioctl() to enable or disable capture and poll() to wait for
data.
A stream is opened something like:
uint64_t properties[] = {
/* Single context sampling */
DRM_I915_PERF_PROP_CTX_HANDLE, ctx_handle,
/* Include OA reports in samples */
DRM_I915_PERF_PROP_SAMPLE_OA, true,
/* OA unit configuration */
DRM_I915_PERF_PROP_OA_METRICS_SET, metrics_set_id,
DRM_I915_PERF_PROP_OA_FORMAT, report_format,
DRM_I915_PERF_PROP_OA_EXPONENT, period_exponent,
};
struct drm_i915_perf_open_param parm = {
.flags = I915_PERF_FLAG_FD_CLOEXEC |
I915_PERF_FLAG_FD_NONBLOCK |
I915_PERF_FLAG_DISABLED,
.properties_ptr = (uint64_t)properties,
.num_properties = sizeof(properties) / 16,
};
int fd = drmIoctl(drm_fd, DRM_IOCTL_I915_PERF_OPEN, ¶m);
Records read all start with a common { type, size } header with
DRM_I915_PERF_RECORD_SAMPLE being of most interest. Sample records
contain an extensible number of fields and it's the
DRM_I915_PERF_PROP_SAMPLE_xyz properties given when opening that
determine what's included in every sample.
No specific streams are supported yet so any attempt to open a stream
will return an error.
v2:
use i915_gem_context_get() - Chris Wilson
v3:
update read() interface to avoid passing state struct - Chris Wilson
fix some rebase fallout, with i915-perf init/deinit
v4:
s/DRM_IORW/DRM_IOW/ - Emil Velikov
Signed-off-by: Robert Bragg <robert@sixbynine.org>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Sourab Gupta <sourab.gupta@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/20161107194957.3385-2-robert@sixbynine.org
Our timelines are more than just a seqno. They also provide an ordered
list of requests to be executed. Due to the restriction of handling
individual address spaces, we are limited to a timeline per address
space but we use a fence context per engine within.
Our first step to introducing independent timelines per context (i.e. to
allow each context to have a queue of requests to execute that have a
defined set of dependencies on other requests) is to provide a timeline
abstraction for the global execution queue.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161028125858.23563-23-chris@chris-wilson.co.uk
Quite a few of our objects used for internal hardware programming do not
benefit from being swappable or from being zero initialised. As such
they do not benefit from using a shmemfs backing storage and since they
are internal and never directly exposed to the user, we do not need to
worry about providing a filp. For these we can use an
drm_i915_gem_object wrapper around a sg_table of plain struct page. They
are not swap backed and not automatically pinned. If they are reaped
by the shrinker, the pages are released and the contents discarded. For
the internal use case, this is fine as for example, ringbuffers are
pinned from being written by a request to be read by the hardware. Once
they are idle, they can be discarded entirely. As such they are a good
match for execlist ringbuffers and a small variety of other internal
objects.
In the first iteration, this is limited to the scratch batch buffers we
use (for command parsing and state initialisation).
v2: Allocate physically contiguous pages, where possible.
v3: Reduce maximum order on subsequent requests following an allocation
failure.
v4: Fix up mismatch between swiotlb segment size and page count (it
counts in 2k units, not 4k pages)
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20161028125858.23563-7-chris@chris-wilson.co.uk
This patch adds a new file, to accommodate lspcon support
for I915 driver. These functions probe, detect, initialize
and configure an on-board lspcon device during the driver
init time.
Also, this patch adds a small structure for lspcon device,
which will provide the runtime status of the device.
V2: addressed ville's review comments
- Clean the leftover macros from previous patch set
V3: Rebase
V4: addressed ville's review comments
- make internal functions static
- remove lspcon_detect_identifier, make it inline with lspcon_probe
- remove is_lspcon_active function
- remove force check while setting a lspcon mode
V5: Rebase
V6: Pass dev_priv to IS_GEN9 check
Signed-off-by: Shashank Sharma <shashank.sharma@intel.com>
Signed-off-by: Akashdeep Sharma <akashdeep.sharma@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Acked-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1476455212-27893-3-git-send-email-shashank.sharma@intel.com
We currently capture the GPU state after we detect a hang. This is vital
for us to both triage and debug hangs in the wild (post-mortem
debugging). However, it comes at the cost of running some potentially
dangerous code (since it has to make very few assumption about the state
of the driver) that is quite resource intensive.
This patch introduces both a method to disable error capture at runtime
(for users who hit bugs at runtime and need a workaround) and to disable
error capture at compiletime (for realtime users who want to minimise
any possible latency, and never require error capture, saving ~30k of
code). The cost is that we now have to be wary of (and test!) a kconfig
flag and a module parameter. The effect of the module parameter is easy
to verify through code inspection and runtime testing, but a kconfig flag
needs regular compile checking.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Acked-by: Jani Nikula <jani.nikula@linux.intel.com>
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch
Link: http://patchwork.freedesktop.org/patch/msgid/20161012090522.367-2-chris@chris-wilson.co.uk
This is really a core kernel struct in disguise until we can finally
place it in kernel/. There is an immediate need for a fence collection
mechanism that is more flexible than fence-array, in particular being
able to easily drive request submission via events (and not just
interrupt driven). The same mechanism would be useful for handling
nonblocking and asynchronous atomic modesets, parallel execution and
more, but for the time being just create a local sw fence for execbuf.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20160909131201.16673-1-chris@chris-wilson.co.uk
This patch provides the infrastructure for performing a 16-byte aligned
read from WC memory using non-temporal instructions introduced with sse4.1.
Using movntdqa we can bypass the CPU caches and read directly from memory
and ignoring the page attributes set on the CPU PTE i.e. negating the
impact of an otherwise UC access. Copying using movntdqa from WC is almost
as fast as reading from WB memory, modulo the possibility of both hitting
the CPU cache or leaving the data in the CPU cache for the next consumer.
(The CPU cache itself my be flushed for the region of the movntdqa and on
later access the movntdqa reads from a separate internal buffer for the
cacheline.) The write back to the memory is however cached.
This will be used in later patches to accelerate accessing WC memory.
v2: Report whether the accelerated copy is successful/possible.
v3: Function alignment override was only necessary when using the
function target("sse4.1") - which is not necessary for emitting movntdqa
from __asm__.
v4: Improve notes on CPU cache behaviour vs non-temporal stores.
v5: Fix byte offsets for unrolled moves.
v6: Find all remaining typos of "movntqda", use kernel_fpu_begin.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Akash Goel <akash.goel@intel.com>
Cc: Damien Lespiau <damien.lespiau@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1471001999-17787-2-git-send-email-chris@chris-wilson.co.uk
With the introduction of requests, we amplified the number of atomic
refcounted objects we use and update every execbuffer; from none to
several references, and a set of references that need to be changed. We
also introduced interesting side-effects in the order of retiring
requests and objects.
Instead of independently tracking the last request for an object, track
the active objects for each request. The object will reside in the
buffer list of its most recent active request and so we reduce the kref
interchange to a list_move. Now retirements are entirely driven by the
request, dramatically simplifying activity tracking on the object
themselves, and removing the ambiguity between retiring objects and
retiring requests.
Furthermore with the consolidation of managing the activity tracking
centrally, we can look forward to using RCU to enable lockless lookup of
the current active requests for an object. In the future, we will be
able to query the status or wait upon rendering to an object without
even touching the struct_mutex BKL.
All told, less code, simpler and faster, and more extensible.
v2: Add a typedef for the function pointer for convenience later.
v3: Make the noop retirement callback explicit. Allow passing NULL to
the init_request_active() which is expanded to a common noop function.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1470293567-10811-16-git-send-email-chris@chris-wilson.co.uk
Common code deserves to be put in a separate file from legacy and
execlists implementation for clarity and ease of maintenance.
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Chris Wilson <chris-wilson.co.uk>
One particularly stressful scenario consists of many independent tasks
all competing for GPU time and waiting upon the results (e.g. realtime
transcoding of many, many streams). One bottleneck in particular is that
each client waits on its own results, but every client is woken up after
every batchbuffer - hence the thunder of hooves as then every client must
do its heavyweight dance to read a coherent seqno to see if it is the
lucky one.
Ideally, we only want one client to wake up after the interrupt and
check its request for completion. Since the requests must retire in
order, we can select the first client on the oldest request to be woken.
Once that client has completed his wait, we can then wake up the
next client and so on. However, all clients then incur latency as every
process in the chain may be delayed for scheduling - this may also then
cause some priority inversion. To reduce the latency, when a client
is added or removed from the list, we scan the tree for completed
seqno and wake up all the completed waiters in parallel.
Using igt/benchmarks/gem_latency, we can demonstrate this effect. The
benchmark measures the number of GPU cycles between completion of a
batch and the client waking up from a call to wait-ioctl. With many
concurrent waiters, with each on a different request, we observe that
the wakeup latency before the patch scales nearly linearly with the
number of waiters (before external factors kick in making the scaling much
worse). After applying the patch, we can see that only the single waiter
for the request is being woken up, providing a constant wakeup latency
for every operation. However, the situation is not quite as rosy for
many waiters on the same request, though to the best of my knowledge this
is much less likely in practice. Here, we can observe that the
concurrent waiters incur extra latency from being woken up by the
solitary bottom-half, rather than directly by the interrupt. This
appears to be scheduler induced (having discounted adverse effects from
having a rbtree walk/erase in the wakeup path), each additional
wake_up_process() costs approximately 1us on big core. Another effect of
performing the secondary wakeups from the first bottom-half is the
incurred delay this imposes on high priority threads - rather than
immediately returning to userspace and leaving the interrupt handler to
wake the others.
To offset the delay incurred with additional waiters on a request, we
could use a hybrid scheme that did a quick read in the interrupt handler
and dequeued all the completed waiters (incurring the overhead in the
interrupt handler, not the best plan either as we then incur GPU
submission latency) but we would still have to wake up the bottom-half
every time to do the heavyweight slow read. Or we could only kick the
waiters on the seqno with the same priority as the current task (i.e. in
the realtime waiter scenario, only it is woken up immediately by the
interrupt and simply queues the next waiter before returning to userspace,
minimising its delay at the expense of the chain, and also reducing
contention on its scheduler runqueue). This is effective at avoid long
pauses in the interrupt handler and at avoiding the extra latency in
realtime/high-priority waiters.
v2: Convert from a kworker per engine into a dedicated kthread for the
bottom-half.
v3: Rename request members and tweak comments.
v4: Use a per-engine spinlock in the breadcrumbs bottom-half.
v5: Fix race in locklessly checking waiter status and kicking the task on
adding a new waiter.
v6: Fix deciding when to force the timer to hide missing interrupts.
v7: Move the bottom-half from the kthread to the first client process.
v8: Reword a few comments
v9: Break the busy loop when the interrupt is unmasked or has fired.
v10: Comments, unnecessary churn, better debugging from Tvrtko
v11: Wake all completed waiters on removing the current bottom-half to
reduce the latency of waking up a herd of clients all waiting on the
same request.
v12: Rearrange missed-interrupt fault injection so that it works with
igt/drv_missed_irq_hang
v13: Rename intel_breadcrumb and friends to intel_wait in preparation
for signal handling.
v14: RCU commentary, assert_spin_locked
v15: Hide BUG_ON behind the compiler; report on gem_latency findings.
v16: Sort seqno-groups by priority so that first-waiter has the highest
task priority (and so avoid priority inversion).
v17: Add waiters to post-mortem GPU hang state.
v18: Return early for a completed wait after acquiring the spinlock.
Avoids adding ourselves to the tree if the is already complete, and
skips the awkward question of why we don't do completion wakeups for
waits earlier than or equal to ourselves.
v19: Prepare for init_breadcrumbs to fail. Later patches may want to
allocate during init, so be prepared to propagate back the error code.
Testcase: igt/gem_concurrent_blit
Testcase: igt/benchmarks/gem_latency
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: "Rogozhkin, Dmitry V" <dmitry.v.rogozhkin@intel.com>
Cc: "Gong, Zhipeng" <zhipeng.gong@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Dave Gordon <david.s.gordon@intel.com>
Cc: "Goel, Akash" <akash.goel@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com> #v18
Link: http://patchwork.freedesktop.org/patch/msgid/1467390209-3576-6-git-send-email-chris@chris-wilson.co.uk
This patch introduces the very basic framework of GVT-g device model,
includes basic prototypes, definitions, initialization.
v12:
- Call intel_gvt_init() in driver early initialization stage. (Chris)
v8:
- Remove the GVT idr and mutex in intel_gvt_host. (Joonas)
v7:
- Refine the URL link in Kconfig. (Joonas)
- Refine the introduction of GVT-g host support in Kconfig. (Joonas)
- Remove the macro GVT_ALIGN(), use round_down() instead. (Joonas)
- Make "struct intel_gvt" a data member in struct drm_i915_private.(Joonas)
- Remove {alloc, free}_gvt_device()
- Rename intel_gvt_{create, destroy}_gvt_device()
- Expost intel_gvt_init_host()
- Remove the dummy "struct intel_gvt" declaration in intel_gvt.h (Joonas)
v6:
- Refine introduction in Kconfig. (Chris)
- The exposed API functions will take struct intel_gvt * instead of
void *. (Chris/Tvrtko)
- Remove most memebers of strct intel_gvt_device_info. Will add them
in the device model patches.(Chris)
- Remove gvt_info() and gvt_err() in debug.h. (Chris)
- Move GVT kernel parameter into i915_params. (Chris)
- Remove include/drm/i915_gvt.h, as GVT-g will be built within i915.
- Remove the redundant struct i915_gvt *, as the functions in i915
will directly take struct intel_gvt *.
- Add more comments for reviewer.
v5:
Take Tvrtko's comments:
- Fix the misspelled words in Kconfig
- Let functions take drm_i915_private * instead of struct drm_device *
- Remove redundant prints/local varible initialization
v3:
Take Joonas' comments:
- Change file name i915_gvt.* to intel_gvt.*
- Move GVT kernel parameter into intel_gvt.c
- Remove redundant debug macros
- Change error handling style
- Add introductions for some stub functions
- Introduce drm/i915_gvt.h.
Take Kevin's comments:
- Move GVT-g host/guest check into intel_vgt_balloon in i915_gem_gtt.c
v2:
- Introduce i915_gvt.c.
It's necessary to introduce the stubs between i915 driver and GVT-g host,
as GVT-g components is configurable in kernel config. When disabled, the
stubs here do nothing.
Take Joonas' comments:
- Replace boolean return value with int.
- Replace customized info/warn/debug macros with DRM macros.
- Document all non-static functions like i915.
- Remove empty and unused functions.
- Replace magic number with marcos.
- Set GVT-g in kernel config to "n" by default.
Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Cc: Kevin Tian <kevin.tian@intel.com>
Signed-off-by: Zhi Wang <zhi.a.wang@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1466078825-6662-5-git-send-email-zhi.a.wang@intel.com
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
If the source of the backlight PWM is from the
panel then the PWM can be controlled by DCS
command, this patch adds the support to
enable/disbale panel PWM, control backlight level
etc...
v2: Moving the CABC bkl functions to new file.(Jani)
v3: Rebase
v4: Rebase
v5: Use mipi_dsi_dcs_write() instead of mipi_dsi_dcs_write_buffer() (Jani)
Move DCS macro`s to include/video/mipi_display.h (Jani)
v6: Rename the file to intel_dsi_panel_pwm.c
Removing the CABC operations
v7 by Jani: renames, rebases, etc.
v8 by Jani: s/INTEL_BACKLIGHT_CABC/INTEL_BACKLIGHT_DSI_DCS/
v9 by Jani: rename init function to intel_dsi_dcs_init_backlight_funcs
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Daniel Vetter <daniel.vetter@intel.com>
Cc: Yetunde Adebisi <yetundex.adebisi@intel.com>
Signed-off-by: Deepak M <m.deepak@intel.com>
Reviewed-by: Yetunde Adebisi <yetundex.adebisi@intel.com>
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/71238a4b14b8c3a6c04070c789f09f1b4bc00a15.1461676337.git.jani.nikula@intel.com
This patch adds support for eDP backlight control using DPCD registers to
backlight hooks in intel_panel.
It checks for backlight control over AUX channel capability and sets up
function pointers to get and set the backlight brightness level if
supported.
v2: Moved backlight functions from intel_dp.c into a new file
intel_dp_aux_backlight.c. Also moved reading of eDP display control
registers to intel_dp_get_dpcd
v3: Correct some formatting mistakes
v4: Updated to use AUX backlight control if PWM control is not possible
(Jani)
v5: Moved call to initialize backlight registers to dp_aux_setup_backlight
v6: Check DP_EDP_BACKLIGHT_PIN_ENABLE_CAP is disabled before setting up AUX
backlight control. To fix BLM_PWM_ENABLE igt test warnings on bdw_ultra
v7: Add enable_dpcd_backlight module parameter.
v8: Rebase onto latest drm-intel-nightly branch
v9: Remove changes to intel_dp_dpcd_read_wake
Split addition edp_dpcd variable into a separate patch
Cc: Bob Paauwe <bob.j.paauwe@intel.com>
Cc: Jani Nikula <jani.nikula@intel.com>
Signed-off-by: Yetunde Adebisi <yetundex.adebisi@intel.com>
[Jani: whitepace changes to appease checkpatch]
Signed-off-by: Jani Nikula <jani.nikula@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/1459865452-9138-4-git-send-email-yetundex.adebisi@intel.com
Our driver compiles clean (nowadays thanks to 0day) but for me, at least,
it would be beneficial if the compiler threw an error rather than a
warning when it found a piece of suspect code. (I use this to
compile-check patch series and want to break on the first compiler error
in order to fix the patch.)
v2: Kick off a new "Debugging" submenu for i915.ko
At this point, we applied it to the kernel and promptly kicked it out
again as it broke buildbots (due to a compiler warning on 32bits):
commit 908d759b21
Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date: Tue May 26 07:46:21 2015 +0200
Revert "drm/i915: Force clean compilation with -Werror"
v3: Avoid enabling -Werror for allyesconfig/allmodconfig builds, using
COMPILE_TEST as a suitable proxy suggested by Andrew Morton. (Damien)
Only make the option available for EXPERT to reinforce that the option
should not be casually enabled.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Jani Nikula <jani.nikula@intel.com>
Cc: Damien Lespiau <damien.lespiau@intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: http://patchwork.freedesktop.org/patch/msgid/1460565315-7748-1-git-send-email-chris@chris-wilson.co.uk
The moves a couple of functions programming the gamma LUT and CSC
units into their own file.
On generations prior to Haswell there is only a gamma LUT. From
haswell on there is also a new enhanced color correction unit that
isn't used yet. This is why we need to set the GAMMA_MODE register,
either we're using the legacy 8bits LUT or enhanced LUTs (of 10 or
12bits).
The CSC unit is only available from Haswell on.
We also need to make a special case for CherryView which is recognized
as a gen 8 but doesn't have the same enhanced color correction unit
from Haswell on.
v2: Fix access to GAMMA_MODE register on older generations than
Haswell (from Matt Roper's comments)
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Matt Roper <matthew.d.roper@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1458125837-2576-2-git-send-email-lionel.g.landwerlin@intel.com
Robert Bragg
Joonas Lahtinen
Mika Kuoppala
Chris Wilson
Shashank Sharma
Tvrtko Ursulin
Zhi Wang
Jani Nikula
Ander Conselvan de Oliveira
Yetunde Adebisi
Lionel Landwerlin