[PATCH] drm/i915: Flush tasklet submission before sleeping on i915_request_wait

[PATCH] drm/i915: Flush tasklet submission before sleeping on i915_request_wait

[Chris Wilson]

If the system is being slow and userspace is racing ahead of the GPU and
finds itself waiting for the GPU to catch up, before the process sleeps
give the tasklet a kick, bypassing ksoftirqd. If the system is
overloaded, then ksoftirqd may be delayed incurring additional latency
to our user.

This should not be a frequent problem, but in the past we have observed
several hundred millisecond delays before ksoftirqd services an
interrupt, so burn a few cycles to lend a helping hand.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
---
 drivers/gpu/drm/i915/i915_request.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
index b8a54572a4f8..f1cadad4e81c 100644
--- a/drivers/gpu/drm/i915/i915_request.c
+++ b/drivers/gpu/drm/i915/i915_request.c
@@ -1496,6 +1496,7 @@ long i915_request_wait(struct i915_request *rq,
 			break;
 		}
 
+		intel_engine_flush_submission(rq->engine);
 		timeout = io_schedule_timeout(timeout);
 	}
 	__set_current_state(TASK_RUNNING);
-- 
2.23.0

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915: Flush tasklet submission before sleeping on i915_request_wait

[Tvrtko Ursulin]

On 15/10/2019 14:26, Chris Wilson wrote:
> If the system is being slow and userspace is racing ahead of the GPU and
> finds itself waiting for the GPU to catch up, before the process sleeps
> give the tasklet a kick, bypassing ksoftirqd. If the system is
> overloaded, then ksoftirqd may be delayed incurring additional latency
> to our user.
> 
> This should not be a frequent problem, but in the past we have observed
> several hundred millisecond delays before ksoftirqd services an
> interrupt, so burn a few cycles to lend a helping hand.
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> ---
>   drivers/gpu/drm/i915/i915_request.c | 1 +
>   1 file changed, 1 insertion(+)
> 
> diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c
> index b8a54572a4f8..f1cadad4e81c 100644
> --- a/drivers/gpu/drm/i915/i915_request.c
> +++ b/drivers/gpu/drm/i915/i915_request.c
> @@ -1496,6 +1496,7 @@ long i915_request_wait(struct i915_request *rq,
>   			break;
>   		}
>   
> +		intel_engine_flush_submission(rq->engine);
>   		timeout = io_schedule_timeout(timeout);
>   	}
>   	__set_current_state(TASK_RUNNING);
> 

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Regards,

Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

✗ Fi.CI.BAT: failure for drm/i915: Flush tasklet submission before sleeping on i915_request_wait

[Patchwork]

== Series Details ==

Series: drm/i915: Flush tasklet submission before sleeping on i915_request_wait
URL   : https://patchwork.freedesktop.org/series/68024/
State : failure

== Summary ==

CI Bug Log - changes from CI_DRM_7096 -> Patchwork_14811
====================================================

Summary
-------

  **FAILURE**

  Serious unknown changes coming with Patchwork_14811 absolutely need to be
  verified manually.
  
  If you think the reported changes have nothing to do with the changes
  introduced in Patchwork_14811, please notify your bug team to allow them
  to document this new failure mode, which will reduce false positives in CI.

  External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14811/index.html

Possible new issues
-------------------

  Here are the unknown changes that may have been introduced in Patchwork_14811:

### IGT changes ###

#### Possible regressions ####

  * igt@i915_selftest@live_gem_contexts:
    - fi-cfl-8109u:       [PASS][1] -> [DMESG-FAIL][2]
   [1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7096/fi-cfl-8109u/igt@i915_selftest@live_gem_contexts.html
   [2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14811/fi-cfl-8109u/igt@i915_selftest@live_gem_contexts.html

  
Known issues
------------

  Here are the changes found in Patchwork_14811 that come from known issues:

### IGT changes ###

#### Issues hit ####

  * igt@gem_mmap_gtt@basic-read-write-distinct:
    - fi-icl-u3:          [PASS][3] -> [DMESG-WARN][4] ([fdo#107724]) +1 similar issue
   [3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7096/fi-icl-u3/igt@gem_mmap_gtt@basic-read-write-distinct.html
   [4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14811/fi-icl-u3/igt@gem_mmap_gtt@basic-read-write-distinct.html

  * igt@i915_selftest@live_gem_contexts:
    - fi-kbl-guc:         [PASS][5] -> [INCOMPLETE][6] ([fdo#112002])
   [5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7096/fi-kbl-guc/igt@i915_selftest@live_gem_contexts.html
   [6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14811/fi-kbl-guc/igt@i915_selftest@live_gem_contexts.html

  * igt@kms_chamelium@hdmi-edid-read:
    - fi-icl-u2:          [PASS][7] -> [FAIL][8] ([fdo#109483])
   [7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7096/fi-icl-u2/igt@kms_chamelium@hdmi-edid-read.html
   [8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14811/fi-icl-u2/igt@kms_chamelium@hdmi-edid-read.html

  
#### Possible fixes ####

  * igt@i915_getparams_basic@basic-eu-total:
    - fi-icl-u3:          [DMESG-WARN][9] ([fdo#107724]) -> [PASS][10]
   [9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7096/fi-icl-u3/igt@i915_getparams_basic@basic-eu-total.html
   [10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14811/fi-icl-u3/igt@i915_getparams_basic@basic-eu-total.html

  * igt@i915_selftest@live_hangcheck:
    - {fi-icl-guc}:       [DMESG-FAIL][11] ([fdo#111144] / [fdo#111678]) -> [PASS][12]
   [11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7096/fi-icl-guc/igt@i915_selftest@live_hangcheck.html
   [12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14811/fi-icl-guc/igt@i915_selftest@live_hangcheck.html

  * igt@kms_busy@basic-flip-a:
    - {fi-tgl-u2}:        [DMESG-WARN][13] ([fdo#111600]) -> [PASS][14]
   [13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7096/fi-tgl-u2/igt@kms_busy@basic-flip-a.html
   [14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14811/fi-tgl-u2/igt@kms_busy@basic-flip-a.html

  * igt@kms_flip@basic-flip-vs-wf_vblank:
    - fi-snb-2520m:       [FAIL][15] ([fdo#100368]) -> [PASS][16]
   [15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7096/fi-snb-2520m/igt@kms_flip@basic-flip-vs-wf_vblank.html
   [16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14811/fi-snb-2520m/igt@kms_flip@basic-flip-vs-wf_vblank.html

  
#### Warnings ####

  * igt@kms_chamelium@hdmi-hpd-fast:
    - fi-kbl-7500u:       [FAIL][17] ([fdo#111045] / [fdo#111096]) -> [FAIL][18] ([fdo#111407])
   [17]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_7096/fi-kbl-7500u/igt@kms_chamelium@hdmi-hpd-fast.html
   [18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14811/fi-kbl-7500u/igt@kms_chamelium@hdmi-hpd-fast.html

  
  {name}: This element is suppressed. This means it is ignored when computing
          the status of the difference (SUCCESS, WARNING, or FAILURE).

  [fdo#100368]: https://bugs.freedesktop.org/show_bug.cgi?id=100368
  [fdo#107724]: https://bugs.freedesktop.org/show_bug.cgi?id=107724
  [fdo#109483]: https://bugs.freedesktop.org/show_bug.cgi?id=109483
  [fdo#111045]: https://bugs.freedesktop.org/show_bug.cgi?id=111045
  [fdo#111096]: https://bugs.freedesktop.org/show_bug.cgi?id=111096
  [fdo#111144]: https://bugs.freedesktop.org/show_bug.cgi?id=111144
  [fdo#111407]: https://bugs.freedesktop.org/show_bug.cgi?id=111407
  [fdo#111600]: https://bugs.freedesktop.org/show_bug.cgi?id=111600
  [fdo#111678]: https://bugs.freedesktop.org/show_bug.cgi?id=111678
  [fdo#112002]: https://bugs.freedesktop.org/show_bug.cgi?id=112002


Participating hosts (53 -> 46)
------------------------------

  Missing    (7): fi-ilk-m540 fi-hsw-4200u fi-byt-squawks fi-bsw-cyan fi-icl-y fi-byt-clapper fi-bdw-samus 


Build changes
-------------

  * CI: CI-20190529 -> None
  * Linux: CI_DRM_7096 -> Patchwork_14811

  CI-20190529: 20190529
  CI_DRM_7096: 2ed00991bec362d1e5fce84b6f0863ba44b79eee @ git://anongit.freedesktop.org/gfx-ci/linux
  IGT_5231: e293051f8f99c72cb01d21e4b73a5928ea351eb3 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
  Patchwork_14811: c216e7bdd85454b1769fe6d9d72600fe2df9e92b @ git://anongit.freedesktop.org/gfx-ci/linux


== Linux commits ==

c216e7bdd854 drm/i915: Flush tasklet submission before sleeping on i915_request_wait

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14811/index.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx