New issue
Advanced search Search tips

Issue 727318 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Jun 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Windows
Pri: 2
Type: Bug-Regression

Blocking:
issue angleproject:2034



Sign in to add a comment

25.9% regression in angle_perftests at 474734:474865

Project Member Reported by jmad...@chromium.org, May 29 2017

Issue description

See the link to graphs below.
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=727318

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgIDg-sWS7AgM


Bot(s) for this bug's original alert(s):

chromium-rel-win7-gpu-nvidia
Project Member

Comment 4 by 42576172...@developer.gserviceaccount.com, May 30 2017

Owner: jmad...@chromium.org

=== Auto-CCing suspected CL author jmadill@chromium.org ===

Hi jmadill@chromium.org, the bisect results pointed to your CL, please take a look at the
results.


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Jamie Madill
  Commit : e0e009feecf1e74264c32a244f741cdb4e038357
  Date   : Wed May 24 15:28:55 2017
  Subject: D3D11: Consolidate SRV allocation.

Bisect Details
  Configuration: winx64nvidia_perf_bisect
  Benchmark    : angle_perftests
  Metric       : Textures_d3d11_8_textures_5_rebind_3_state_8_mips/score
  Change       : 28.81% | 1654.5 -> 1177.83333333

Revision                              Result                  N
chromium@474733                       1654.5 +- 27.45         6      good
chromium@474799                       1648.17 +- 96.8341      6      good
chromium@474832                       1676.0 +- 27.2397       6      good
chromium@474849                       1677.17 +- 19.3089      6      good
chromium@474851                       1682.17 +- 27.5832      6      good
chromium@474851,angle@22717ce238      1645.5 +- 115.834       6      good
chromium@474851,angle@39b826fc8a      1661.33 +- 9.86577      6      good
chromium@474851,angle@e0e009feec      1193.67 +- 15.7903      6      bad       <--
chromium@474851,angle@caafd890bf      1191.33 +- 7.70281      6      bad
chromium@474852                       1183.83 +- 66.4743      6      bad
chromium@474853                       1194.5 +- 26.0288       6      bad
chromium@474857                       1177.5 +- 134.326       6      bad
chromium@474865                       1177.83 +- 82.6004      6      bad

To Run This Test
  .\src\out\Release_x64\angle_perftests.exe --test-launcher-print-test-stdio=always --test-launcher-jobs=1

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8978252812993864704

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5548812413698048


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!
Project Member

Comment 5 by 42576172...@developer.gserviceaccount.com, May 30 2017


=== BISECT JOB RESULTS ===
Perf regression found with culprit

Suspected Commit
  Author : Jamie Madill
  Commit : e0e009feecf1e74264c32a244f741cdb4e038357
  Date   : Wed May 24 15:28:55 2017
  Subject: D3D11: Consolidate SRV allocation.

Bisect Details
  Configuration: winx64nvidia_perf_bisect
  Benchmark    : angle_perftests
  Metric       : Textures_d3d11_8_textures_5_rebind_3_state_8_mips/score
  Change       : 27.07% | 1655.5 -> 1207.33333333

Revision                              Result                  N
chromium@474733                       1655.5 +- 27.377        6      good
chromium@474799                       1676.67 +- 20.1825      6      good
chromium@474832                       1688.33 +- 9.55685      6      good
chromium@474849                       1694.17 +- 9.31844      6      good
chromium@474851                       1687.17 +- 9.53065      6      good
chromium@474851,angle@22717ce238      1674.83 +- 5.90198      6      good
chromium@474851,angle@39b826fc8a      1666.33 +- 3.91578      6      good
chromium@474851,angle@e0e009feec      1190.17 +- 40.2099      6      bad       <--
chromium@474851,angle@caafd890bf      1181.83 +- 54.2847      6      bad
chromium@474852                       1201.5 +- 23.696        6      bad
chromium@474853                       1203.5 +- 15.2807       6      bad
chromium@474857                       1210.67 +- 12.3828      6      bad
chromium@474865                       1207.33 +- 13.1656      6      bad

To Run This Test
  .\src\out\Release_x64\angle_perftests.exe --test-launcher-print-test-stdio=always --test-launcher-jobs=1

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8978252805309912656

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6738101742862336


| O O | Visit http://www.chromium.org/developers/speed-infra/perf-bug-faq
|  X  | for more information addressing perf regression bugs. For feedback,
| / \ | file a bug with component Speed>Bisection.  Thank you!
Components: Internals>GPU>ANGLE
Labels: OS-Windows
Status: Assigned (was: Untriaged)
Project Member

Comment 7 by bugdroid1@chromium.org, May 30 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/angle/angle/+/0a82f2fab246e2f4c6cd936fd9c9937dcea381ab

commit 0a82f2fab246e2f4c6cd936fd9c9937dcea381ab
Author: Jamie Madill <jmadill@chromium.org>
Date: Tue May 30 21:18:08 2017

D3D11: Don't copy SharedSRVs around.

This was causing a very large performance regression in Texture
re-binding. Instead pass by const &, which should be fine, since
values in std::map (and unordered_map) are not reallocated when the
map changes.

Also make the SharedResource type non-copyable, and add an explicit
makeCopy method for when we need to clone the shared pointer.

BUG= angleproject:2034 
BUG= chromium:727318 

Change-Id: I39508a6ca4b41e4da31fe68899caa4464138cada
Reviewed-on: https://chromium-review.googlesource.com/517670
Commit-Queue: Jamie Madill <jmadill@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>

[modify] https://crrev.com/0a82f2fab246e2f4c6cd936fd9c9937dcea381ab/src/libANGLE/renderer/d3d/d3d11/RenderTarget11.cpp
[modify] https://crrev.com/0a82f2fab246e2f4c6cd936fd9c9937dcea381ab/src/libANGLE/renderer/d3d/d3d11/TextureStorage11.cpp
[modify] https://crrev.com/0a82f2fab246e2f4c6cd936fd9c9937dcea381ab/src/libANGLE/renderer/d3d/d3d11/Renderer11.cpp
[modify] https://crrev.com/0a82f2fab246e2f4c6cd936fd9c9937dcea381ab/src/libANGLE/renderer/d3d/d3d11/TextureStorage11.h
[modify] https://crrev.com/0a82f2fab246e2f4c6cd936fd9c9937dcea381ab/src/libANGLE/renderer/d3d/d3d11/ResourceManager11.h

Project Member

Comment 8 by bugdroid1@chromium.org, Jun 1 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/47445933f3d85930bde4877c39328ecaffc2618d

commit 47445933f3d85930bde4877c39328ecaffc2618d
Author: Corentin Wallez <cwallez@chromium.org>
Date: Thu Jun 01 15:49:28 2017

Roll ANGLE 63d8c26..9216a6e

https://chromium.googlesource.com/angle/angle.git/+log/63d8c26..9216a6e

BUG= chromium:727318 

TBR=geofflang@chromium.org

TEST=bots

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel

Change-Id: Iea289d65c9904d1b956e1e39a2e2c4e100778ad7
Reviewed-on: https://chromium-review.googlesource.com/519643
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Geoff Lang <geofflang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#476295}
[modify] https://crrev.com/47445933f3d85930bde4877c39328ecaffc2618d/DEPS

Blocking: angleproject:2034
Status: Fixed (was: Assigned)
Graph has recovered.

Sign in to add a comment