On 21/07/2020 05:00, Umesh Nerlige Ramappa wrote:
From: Piotr Maciejewski
It is useful to have markers in the OA reports to identify triggered
reports. Whitelist some OA counters that can be used as markers.
A triggered report can be found faster if we can sample the HW tail and
head registers
On 21/07/2020 05:00, Umesh Nerlige Ramappa wrote:
From: Piotr Maciejewski
A clock gating switch can control if the performance monitoring and
observation logic is enaled or not. Ensure that we enable the clocks.
v2: Separate code from other patches (Lionel)
v3: Reset PMON enable when disabling
== Series Details ==
Series: Allow privileged user to map the OA buffer (rev3)
URL : https://patchwork.freedesktop.org/series/79460/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8768 -> Patchwork_18217
Summary
---
*
== Series Details ==
Series: Allow privileged user to map the OA buffer (rev3)
URL : https://patchwork.freedesktop.org/series/79460/
State : warning
== Summary ==
$ dim sparse --fast origin/drm-tip
Sparse version: v0.6.0
Fast mode used, each commit won't be checked separately.
__
== Series Details ==
Series: Allow privileged user to map the OA buffer (rev3)
URL : https://patchwork.freedesktop.org/series/79460/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
60886d35a086 drm/i915/perf: Ensure observation logic is not clock gated
d11b49577d7e drm/i915/perf:
Hi Chris,
I have added your comments, but I have a few questions below:
Thanks for your help,
Umesh
On Mon, Jul 20, 2020 at 07:00:12PM -0700, Umesh Nerlige Ramappa wrote:
From: Piotr Maciejewski
i915 used to support time based sampling mode which is good for overall
system monitoring, but is
From: Piotr Maciejewski
It is useful to have markers in the OA reports to identify triggered
reports. Whitelist some OA counters that can be used as markers.
A triggered report can be found faster if we can sample the HW tail and
head registers when the report was triggered. Whitelist OA buffer
From: Piotr Maciejewski
OA reports can be triggered into the OA buffer by writing into the
OAREPORTTRIG registers. Whitelist the registers to allow user to trigger
reports.
v2:
- Move related change to this patch (Lionel)
- Bump up perf revision (Lionel)
v3: Pardon whitelisted registers for sel
This cover letter is included to trigger "Test-with" an IGT patch.
Tests - https://patchwork.freedesktop.org/series/79695/
Signed-off-by: Umesh Nerlige Ramappa
Test-with: 20200721015717.46279-1-umesh.nerlige.rama...@intel.com
Piotr Maciejewski (4):
drm/i915/perf: Ensure observation logic is n
From: Piotr Maciejewski
A clock gating switch can control if the performance monitoring and
observation logic is enaled or not. Ensure that we enable the clocks.
v2: Separate code from other patches (Lionel)
v3: Reset PMON enable when disabling perf to save power (Lionel)
Fixes: 00a7f0d7155c ("
From: Piotr Maciejewski
i915 used to support time based sampling mode which is good for overall
system monitoring, but is not enough for query mode used to measure a
single draw call or dispatch. Gen9-Gen11 are using current i915 perf
implementation for query, but Gen12+ requires a new approach f
== Series Details ==
Series: drm/i915/tgl: Make sure TC-cold is blocked before enabling TC AUX power
wells
URL : https://patchwork.freedesktop.org/series/79691/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8767_full -> Patchwork_18215_full
===
== Series Details ==
Series: series starting with [1/2] drm/dp: Add PHY_TEST_PATTERN CP2520 Pattern
2 and 3
URL : https://patchwork.freedesktop.org/series/79693/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8768 -> Patchwork_18216
On Mon, 2020-07-20 at 17:07 -0700, Manasi Navare wrote:
> On Mon, Jul 20, 2020 at 04:41:25PM -0700, Khaled Almahallawy wrote:
> > Add the missing CP2520 pattern 2 and 3 phy compliance patterns
> >
> > Signed-off-by: Khaled Almahallawy
> > ---
> > drivers/gpu/drm/drm_dp_helper.c | 2 +-
> > inclu
== Series Details ==
Series: series starting with [1/2] drm/dp: Add PHY_TEST_PATTERN CP2520 Pattern
2 and 3
URL : https://patchwork.freedesktop.org/series/79693/
State : warning
== Summary ==
$ dim sparse --fast origin/drm-tip
Sparse version: v0.6.0
Fast mode used, each commit won't be checke
On Mon, 2020-07-20 at 17:11 -0700, Manasi Navare wrote:
> On Mon, Jul 20, 2020 at 04:41:26PM -0700, Khaled Almahallawy wrote:
> > Adding support for TPS4 (CP2520 Pattern 3) PHY pattern source
> > tests.
> >
> > Signed-off-by: Khaled Almahallawy
> > ---
> > drivers/gpu/drm/i915/display/intel_dp.c
On Mon, Jul 20, 2020 at 04:41:26PM -0700, Khaled Almahallawy wrote:
> Adding support for TPS4 (CP2520 Pattern 3) PHY pattern source tests.
>
> Signed-off-by: Khaled Almahallawy
> ---
> drivers/gpu/drm/i915/display/intel_dp.c | 14 --
> drivers/gpu/drm/i915/i915_reg.h | 4 +++
On Mon, Jul 20, 2020 at 04:41:25PM -0700, Khaled Almahallawy wrote:
> Add the missing CP2520 pattern 2 and 3 phy compliance patterns
>
> Signed-off-by: Khaled Almahallawy
> ---
> drivers/gpu/drm/drm_dp_helper.c | 2 +-
> include/drm/drm_dp_helper.h | 4 +++-
> 2 files changed, 4 insertions(+
== Series Details ==
Series: drm/i915/tgl: Make sure TC-cold is blocked before enabling TC AUX power
wells
URL : https://patchwork.freedesktop.org/series/79691/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8767 -> Patchwork_18215
=
On Mon, 2020-07-20 at 10:09 -0700, José Roberto de Souza wrote:
> This new HBR2 table for TGL-U and TGL-Y is required to pass
> DisplayPort compliance.
>
> BSpec: 49291
> Cc: Khaled Almahallawy
> Signed-off-by: José Roberto de Souza
> ---
> drivers/gpu/drm/i915/display/intel_ddi.c | 19
Adding support for TPS4 (CP2520 Pattern 3) PHY pattern source tests.
Signed-off-by: Khaled Almahallawy
---
drivers/gpu/drm/i915/display/intel_dp.c | 14 --
drivers/gpu/drm/i915/i915_reg.h | 4
2 files changed, 16 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/d
Add the missing CP2520 pattern 2 and 3 phy compliance patterns
Signed-off-by: Khaled Almahallawy
---
drivers/gpu/drm/drm_dp_helper.c | 2 +-
include/drm/drm_dp_helper.h | 4 +++-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/
== Series Details ==
Series: drm/i915/tgl: Make sure TC-cold is blocked before enabling TC AUX power
wells
URL : https://patchwork.freedesktop.org/series/79691/
State : warning
== Summary ==
$ dim sparse --fast origin/drm-tip
Sparse version: v0.6.0
Fast mode used, each commit won't be checked
On Tue, 2020-07-21 at 02:29 +0300, Imre Deak wrote:
> The dependency between power wells is determined by the ordering of the
> power well list: when enabling the power wells for a domain, this
> happens walking the power well list forward, while disabling them
> happens in the reverse direction. A
The dependency between power wells is determined by the ordering of the
power well list: when enabling the power wells for a domain, this
happens walking the power well list forward, while disabling them
happens in the reverse direction. Accordingly a power well on the list
must follow any other po
== Series Details ==
Series: HAX sched/core: Paper over the ttwu() race
URL : https://patchwork.freedesktop.org/series/79682/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8766_full -> Patchwork_18214_full
Summary
---
== Series Details ==
Series: series starting with [1/2] drm/i915/tgl: Set subplatforms
URL : https://patchwork.freedesktop.org/series/79681/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8766_full -> Patchwork_18213_full
Su
== Series Details ==
Series: HAX sched/core: Paper over the ttwu() race
URL : https://patchwork.freedesktop.org/series/79682/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8766 -> Patchwork_18214
Summary
---
**SUCCES
== Series Details ==
Series: HAX sched/core: Paper over the ttwu() race
URL : https://patchwork.freedesktop.org/series/79682/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
61aff133b9cc HAX sched/core: Paper over the ttwu() race
-:8: WARNING:COMMIT_LOG_LONG_LINE: Possible unwrap
== Series Details ==
Series: series starting with [1/2] drm/i915/tgl: Set subplatforms
URL : https://patchwork.freedesktop.org/series/79681/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8766 -> Patchwork_18213
Summary
On all Core platforms we see
<4> [181.766629] WARNING: CPU: 0 PID: 1524 at kernel/sched/core.c:2388
ttwu_queue_wakelist+0xbc/0xd0
<4> [181.766637] Modules linked in: i915(+) vgem snd_hda_codec_realtek
snd_hda_codec_generic ledtrig_audio amdgpu mei_hdcp x86_pkg_temp_thermal
coretemp crct10dif_pc
== Series Details ==
Series: series starting with [1/2] drm/i915/tgl: Set subplatforms
URL : https://patchwork.freedesktop.org/series/79681/
State : warning
== Summary ==
$ dim sparse --fast origin/drm-tip
Sparse version: v0.6.0
Fast mode used, each commit won't be checked separately.
__
This new HBR2 table for TGL-U and TGL-Y is required to pass
DisplayPort compliance.
BSpec: 49291
Cc: Khaled Almahallawy
Signed-off-by: José Roberto de Souza
---
drivers/gpu/drm/i915/display/intel_ddi.c | 19 +++
1 file changed, 19 insertions(+)
diff --git a/drivers/gpu/drm/i915
There is no way to differentiate TGL-U from TGL-Y by the PCI ids as
some ids are available in both SKUs.
So here using the root device id in the PCI bus that iGPU is in
to differentiate between U and Y.
BSpec: 44455
Signed-off-by: José Roberto de Souza
---
drivers/gpu/drm/i915/i915_drv.h
== Series Details ==
Series: Asynchronous flip implementation for i915 (rev5)
URL : https://patchwork.freedesktop.org/series/74386/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8766_full -> Patchwork_18212_full
Summary
---
== Series Details ==
Series: Expose crtc dither state and connector max bpc via debugfs (rev2)
URL : https://patchwork.freedesktop.org/series/79664/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8766_full -> Patchwork_18211_full
Quoting Tvrtko Ursulin (2020-07-20 12:23:35)
>
> On 20/07/2020 10:23, Chris Wilson wrote:
> > -void intel_engine_init_breadcrumbs(struct intel_engine_cs *engine)
> > +struct intel_breadcrumbs *
> > +intel_breadcrumbs_create(struct intel_engine_cs *irq_engine)
> > {
> > - struct intel_breadcr
== Series Details ==
Series: series starting with [01/10] drm/i915/gem: Remove disordered per-file
request list for throttling
URL : https://patchwork.freedesktop.org/series/79663/
State : failure
== Summary ==
CI Bug Log - changes from CI_DRM_8766_full -> Patchwork_18210_full
===
== Series Details ==
Series: Asynchronous flip implementation for i915 (rev5)
URL : https://patchwork.freedesktop.org/series/74386/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8766 -> Patchwork_18212
Summary
---
**
== Series Details ==
Series: Asynchronous flip implementation for i915 (rev5)
URL : https://patchwork.freedesktop.org/series/74386/
State : warning
== Summary ==
$ dim sparse --fast origin/drm-tip
Sparse version: v0.6.0
Fast mode used, each commit won't be checked separately.
___
== Series Details ==
Series: Expose crtc dither state and connector max bpc via debugfs (rev2)
URL : https://patchwork.freedesktop.org/series/79664/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8766 -> Patchwork_18211
Summ
Quoting Umesh Nerlige Ramappa (2020-07-18 01:04:37)
> +static vm_fault_t vm_fault_oa(struct vm_fault *vmf)
> +{
> + struct vm_area_struct *vma = vmf->vma;
> + struct i915_perf_stream *stream = vma->vm_private_data;
> + struct drm_i915_gem_object *obj = stream->oa_buffer.vma->obj;
Don't rely on successful write to sysfs control files, assert existence
/ non-existence of a respective device sysfs node as well.
Signed-off-by: Janusz Krzysztofik
---
tests/core_hotunplug.c | 14 ++
1 file changed, 14 insertions(+)
diff --git a/tests/core_hotunplug.c b/tests/core_
Return values of driver bind/unbind / device remove/recover sysfs
operations are now ignored. Assert their correctness.
Signed-off-by: Janusz Krzysztofik
---
tests/core_hotunplug.c | 14 ++
1 file changed, 10 insertions(+), 4 deletions(-)
diff --git a/tests/core_hotunplug.c b/tests
Device bus address structure field is always initialized with a pointer
to a substring of the device sysfs path and never used for its
modification. Declare it as a constant string.
Signed-off-by: Janusz Krzysztofik
---
tests/core_hotunplug.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-
Since the test is still blocklisted due to driver issues and won't be
executed on CI, I'm providing a link where results obtained from a
rybot run with the test removed from the blocklist can be found:
https://patchwork.freedesktop.org/series/79662/
Failures reported there come from driver and not
There is a new library helper that asserts validity of open file
descriptors. Use it instead of open coding.
Signed-off-by: Janusz Krzysztofik
---
tests/core_hotunplug.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/tests/core_hotunplug.c b/tests/core_hotunplug.
Some return values are not useful and can be ignored. Wrap those cases
inside igt_ignore_warn(), not only to make sure compilers are happy but
also to clearly document our decisions.
Signed-off-by: Janusz Krzysztofik
---
tests/core_hotunplug.c | 6 +++---
1 file changed, 3 insertions(+), 3 dele
A pointer to fatal error messages can be passed around via hotunplug
structure, no need to declare it as global.
Signed-off-by: Janusz Krzysztofik
---
tests/core_hotunplug.c | 92 +-
1 file changed, 45 insertions(+), 47 deletions(-)
diff --git a/tests/cor
Some debug messages which designate specific test operations, or their
greater parts at least, sound always the same, no matter which subtest
they are called from. Emit them, possibly updated with subtest
specified modifiers, from inside respective helpers instead of
duplicating them in subtest bo
Each subtest now calls a prepare() helper which opens a couple of files
required by that subtest. Those files are then closed after use,
either directly from the subtest body, or indirectly from inside one of
helper functions called during the subtest execution. That approach
not only makes lifec
Since health checks are now run from follow-up fixture sections, it is
safe to fail subtests without the need to abort the test execution. Do
that on device close errors instead of emitting warnings.
Signed-off-by: Janusz Krzysztofik
---
tests/core_hotunplug.c | 8
1 file changed, 4 in
Subtests now forcibly call or request igt_abort on failures in order to
avoid silently leaving an exercised device in an unusable state.
However, a failure inside a subtest doesn't always mean the device is
no longer working correctly and reboot is needed. On the other hand,
if a subtest just fail
Since we no longer open a device DRM sysfs node, only a PCI one, driver
unbind operations are no longer affected by missed or unsuccessful
sysfs file close attempts. Skip only affected subtests if that
happens.
Signed-off-by: Janusz Krzysztofik
---
tests/core_hotunplug.c | 9 +
1 file c
The following changes to the test are planned:
- avoid global variables,
- skip subtest after device close errors,
- prepare invariant data only once per test run,
- move device health checks to igt_fixture sections,
- try to recover from subtest failures instead of aborting.
For that to be possibl
The test now ignores device close errors. Those errors are believed to
have no influence on device health so there is no need to process them
the same way as we mostly do on errors, i.e., notify CI about a problem
via igt_abort. However, those errors may indicate issues with the test
itself. Mor
A trailing newline is missing from one of fatal error messages, fix it.
Signed-off-by: Janusz Krzysztofik
---
tests/core_hotunplug.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/core_hotunplug.c b/tests/core_hotunplug.c
index 6ceb325ad..cac88c2f3 100644
--- a/tests/c
Return value of igt_device_filter_add() representing a number of
successfully installed device filters is now ignored. Fail if not 1.
Signed-off-by: Janusz Krzysztofik
---
tests/core_hotunplug.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tests/core_hotunplug.c b/tests/c
== Series Details ==
Series: Expose crtc dither state and connector max bpc via debugfs (rev2)
URL : https://patchwork.freedesktop.org/series/79664/
State : warning
== Summary ==
$ dim sparse --fast origin/drm-tip
Sparse version: v0.6.0
Fast mode used, each commit won't be checked separately.
== Series Details ==
Series: Expose crtc dither state and connector max bpc via debugfs (rev2)
URL : https://patchwork.freedesktop.org/series/79664/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
525ee1f3c796 i915/debug: Expose crtc dither state via debugfs
-:45: WARNING:SYMBOLI
Without async flip support in the kernel, fullscreen apps where game
resolution is equal to the screen resolution, must perform an extra blit
per frame prior to flipping.
Asynchronous page flips will also boost the FPS of Mesa benchmarks.
v2: -Few patches have been squashed and patches have been
Since the flip done event will be sent in the flip_done_handler,
no need to add the event to the list and delay it for later.
v2: -Moved the async check above vblank_get as it
was causing issues for PSR.
v3: -No need to wait for vblank to pass, as this wait was causing a
16ms delay once
Add enable/disable flip done functions and the flip done handler
function which handles the flip done interrupt.
Enable the flip done interrupt in IER.
Enable flip done function is called before writing the
surface address register as the write to this register triggers
the flip done interrupt
F
Set the Async Address Update Enable bit in plane ctl
when async flip is requested.
v2: -Move the Async flip enablement to individual patch (Paulo)
v3: -Rebased.
v4: -Add separate plane hook for async flip case (Ville)
v5: -Rebased.
Signed-off-by: Karthik B S
Signed-off-by: Vandita Kulkarni
-
Support added only for async flips on primary plane.
If flip is requested on any other plane, reject it.
Make sure there is no change in fbc, offset and framebuffer modifiers
when async flip is requested.
If any of these are modified, reject async flip.
v2: -Replace DRM_ERROR (Paulo)
-Add ch
Enable asynchronous flips in i915 for gen9+ platforms.
v2: -Async flip enablement should be a stand alone patch (Paulo)
v3: -Move the patch to the end of the serires (Paulo)
v4: -Rebased.
v5: -Rebased.
Signed-off-by: Karthik B S
Signed-off-by: Vandita Kulkarni
---
drivers/gpu/drm/i915/displ
[Why]
It's useful to know the max supported panel BPC for IGT testing.
[How]
Expose the max supported BPC for the panel via a debugfs file on the
connector, "output_bpc".
Example usage: cat /sys/kernel/debug/dri/0/DP-1/output_bpc
Signed-off-by: Bhanuprakash Modem
---
.../drm/i915/display/intel
[Why]
It's useful to know the dithering state for IGT testing.
[How]
Expose the dithering state for the crtc via a debugfs file "dither".
Example usage: cat /sys/kernel/debug/dri/0/crtc-0/dither
Signed-off-by: Bhanuprakash Modem
---
drivers/gpu/drm/i915/i915_debugfs.c | 17 +
1
[why]
It's useful to know the max supported panel BPC and PIPE dither state
for IGT testing.
[how]
* Expose the connector max supported BPC for the panel via a debugfs file on the
connector, "output_bpc".
Example: cat /sys/kernel/debug/dri/0/DP-1/output_bpc
* Expose the dithering state for th
[why]
It's useful to know the max supported panel BPC and PIPE dither state
for IGT testing.
[how]
* Expose the connector max supported BPC for the panel via a debugfs file on the
connector, "output_bpc".
Example: cat /sys/kernel/debug/dri/0/DP-1/output_bpc
* Expose the dithering state for th
[Why]
It's useful to know the max supported panel BPC for IGT testing.
[How]
Expose the max supported BPC for the panel via a debugfs file on the
connector, "output_bpc".
Example usage: cat /sys/kernel/debug/dri/0/DP-1/output_bpc
Signed-off-by: Bhanuprakash Modem
---
.../drm/i915/display/intel
[Why]
It's useful to know the dithering state for IGT testing.
[How]
Expose the dithering state for the crtc via a debugfs file "dither".
Example usage: cat /sys/kernel/debug/dri/0/crtc-0/dither
Signed-off-by: Bhanuprakash Modem
---
drivers/gpu/drm/i915/i915_debugfs.c | 17 +
1
On 20/07/2020 10:23, Chris Wilson wrote:
On the virtual engines, we only use the intel_breadcrumbs for tracking
signaling of stale breadcrumbs from the irq_workers. They do not have
any associated interrupt handling, active requests are passed to a
physical engine and associated breadcrumb inter
Hi,
On 7/9/20 2:33 PM, Daniel Vetter wrote:
Comes up every few years, gets somewhat tedious to discuss, let's
write this down once and for all.
What I'm not sure about is whether the text should be more explicit in
flat out mandating the amdkfd eviction fences for long running compute
workloads
== Series Details ==
Series: series starting with [01/10] drm/i915/gem: Remove disordered per-file
request list for throttling
URL : https://patchwork.freedesktop.org/series/79663/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_8766 -> Patchwork_18210
=
Quoting Matthew Auld (2020-07-20 11:34:10)
> On 15/07/2020 12:50, Chris Wilson wrote:
> > The GEM object is grossly overweight for the practicality of tracking
> > large numbers of individual pages, yet it is currently our only
> > abstraction for tracking DMA allocations. Since those allocations n
On 15/07/2020 12:50, Chris Wilson wrote:
We need to make the DMA allocations used for page directories to be
performed up front so that we can include those allocations in our
memory reservation pass. The downside is that we have to assume the
worst case, even before we know the final layout, and
On 15/07/2020 12:50, Chris Wilson wrote:
The GEM object is grossly overweight for the practicality of tracking
large numbers of individual pages, yet it is currently our only
abstraction for tracking DMA allocations. Since those allocations need
to be reserved upfront before an operation, and tha
== Series Details ==
Series: series starting with [01/10] drm/i915/gem: Remove disordered per-file
request list for throttling
URL : https://patchwork.freedesktop.org/series/79663/
State : warning
== Summary ==
$ dim sparse --fast origin/drm-tip
Sparse version: v0.6.0
Fast mode used, each com
== Series Details ==
Series: series starting with [01/10] drm/i915/gem: Remove disordered per-file
request list for throttling
URL : https://patchwork.freedesktop.org/series/79663/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
5658c6f4f5ec drm/i915/gem: Remove disordered per-f
On 20/07/2020 10:23, Chris Wilson wrote:
After staring at the breadcrumb enabling/cancellation and coming to the
conclusion that the cause of the mysterious stale breadcrumbs must the
act of submitting a completed requests, we can then redirect those
completed requests onto a dedicated signaled
Originally, we used the signal->lock as a means of following the
previous link in its timeline and peeking at the previous fence.
However, we have replaced the explicit serialisation with a series of
very careful probes that anticipate the links being deleted and the
fences recycled before we are a
Move the __intel_breadcrumbs_arm_irq earlier, next to the disarm_irq, so
that we can make use of it in the following patch.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/gt/intel_breadcrumbs.c | 88 ++---
1 file changed, 44 insertions(+), 44 deletions(-)
diff --git a/driv
As we funnel more and more contexts into the breadcrumbs on an engine,
the hold time of b->irq_lock grows. As we may then contend with the
b->irq_lock during request submission, this increases the burden upon
the engine->active.lock and so directly impacts both our execution
latency and client late
After staring at the breadcrumb enabling/cancellation and coming to the
conclusion that the cause of the mysterious stale breadcrumbs must the
act of submitting a completed requests, we can then redirect those
completed requests onto a dedicated signaled_list at the time of
construction and so elim
One more complication of preempt-to-busy with respect to the virtual
engine is that we may have retired the last request along the virtual
engine at the same time as preparing to submit the completed request to
a new engine. That submit will be shortcircuited, but not before we have
updated the con
On the virtual engines, we only use the intel_breadcrumbs for tracking
signaling of stale breadcrumbs from the irq_workers. They do not have
any associated interrupt handling, active requests are passed to a
physical engine and associated breadcrumb interrupt handler. This causes
issues for us as w
I915_GEM_THROTTLE dates back to the time before contexts where there was
just a single engine, and therefore a single timeline and request list
globally. That request list was in execution/retirement order, and so
walking it to find a particular aged request made sense and could be
split per file.
Currently we hold no actual reference to the request nor context while
they are attached to a breadcrumb. To avoid freeing the request/context
too early, we serialise with cancel-breadcrumbs by taking the irq
spinlock in i915_request_retire(). The alternative is to take a
reference for a new breadc
Make b->signaled_requests a lockless-list so that we can manipulate it
outside of the b->irq_lock.
Signed-off-by: Chris Wilson
---
drivers/gpu/drm/i915/gt/intel_breadcrumbs.c | 42 +--
.../gpu/drm/i915/gt/intel_breadcrumbs_types.h | 2 +-
drivers/gpu/drm/i915/i915_request.h
Since the breadcrumb enabling/cancelling itself is serialised by the
breadcrumbs.irq_lock, with a bit of care we can remove the outer
serialisation with i915_request.lock for concurrent
dma_fence_enable_signaling(). This has the important side-effect of
eliminating the nested i915_request.lock with
Hi,
Sorry that this might be a bit late as last week our QA people were
busy on something else..So this is gvt changes queued for 5.9 which is
to improve guest suspend/resume with proper PCI PM state tracking for
resource handling, e.g ppgtt. Hopefully this could still be in queue
for 5.9.
Thank
92 matches
Mail list logo
