New issue
Advanced search Search tips

Issue 614468 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Jun 2016
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

11.5%-44.4% regression in page_cycler.typical_25 at 395431:395455

Project Member Reported by m...@chromium.org, May 24 2016

Issue description

See the link to graphs below.
 
Project Member

Comment 2 by 42576172...@developer.gserviceaccount.com, May 24 2016

Cc: ericrk@chromium.org
Owner: ericrk@chromium.org

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

Hi ericrk@chromium.org, the bisect results pointed to your CL below as possibly
causing a regression. Please have a look at this info and see whether
your CL be related.


===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : Enable GPU Raster for Mac waterfall
Author  : ericrk
Commit description:
  
This causes waterfall bots to run with GPU raster enabled while we
prepare to launch via Finch.

BUG=612275, 542959 

Review-Url: https://codereview.chromium.org/1992603002
Cr-Commit-Position: refs/heads/master@{#395445}
Commit  : f91789afe1568b37a4e93dfe0580c96df24137f6
Date    : Mon May 23 22:58:28 2016


===== TESTED REVISIONS =====
Revision         Mean    Std Dev  N  Good?
chromium@395433  269931  1435.35  5  good
chromium@395441  269779  777.94   5  good
chromium@395443  269319  1432.79  5  good
chromium@395444  269230  1401.96  5  good
chromium@395445  324057  1307.09  5  bad    <--
chromium@395448  325732  2119.76  5  bad

Bisect job ran on: mac_retina_perf_bisect
Bug ID: 614468

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests page_cycler.intl_ar_fa_he
Test Metric: vm_resident_set_size_final_size_total/vm_resident_set_size_final_size_total
Relative Change: 20.67%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_retina_perf_bisect/builds/1310
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9011752091635705104


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5209073130143744

| 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 Tests>AutoBisect.  Thank you!
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, May 24 2016


===== BISECT JOB RESULTS =====
Status: completed


===== SUSPECTED CL(s) =====
Subject : Enable GPU Raster for Mac waterfall
Author  : ericrk
Commit description:
  
This causes waterfall bots to run with GPU raster enabled while we
prepare to launch via Finch.

BUG=612275, 542959 

Review-Url: https://codereview.chromium.org/1992603002
Cr-Commit-Position: refs/heads/master@{#395445}
Commit  : f91789afe1568b37a4e93dfe0580c96df24137f6
Date    : Mon May 23 22:58:28 2016


===== TESTED REVISIONS =====
Revision         Mean    Std Dev  N  Good?
chromium@395433  312832  1477.3   5  good
chromium@395441  312425  761.365  5  good
chromium@395443  311877  1166.68  5  good
chromium@395444  313586  835.114  5  good
chromium@395445  451218  3800.33  5  bad    <--
chromium@395448  450204  2180.59  5  bad

Bisect job ran on: mac_retina_perf_bisect
Bug ID: 614468

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests page_cycler.typical_25
Test Metric: vm_resident_set_size_final_size_total/vm_resident_set_size_final_size_total
Relative Change: 43.91%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_retina_perf_bisect/builds/1309
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9011752148796845360


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5888260937613312

| 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 Tests>AutoBisect.  Thank you!

Comment 4 by ericrk@chromium.org, May 26 2016

 Issue 609517  has been merged into this issue.

Comment 5 by ericrk@chromium.org, May 26 2016

Can investigate a bit - but I wonder how relevant the regression really is - while it is up to a 50% regression, we're talking about going from 300k to 450k of memory - very small numbers.
Most of the concerning regressions for this issue were in the smoothness.top_25_smooth category. These were largely fixed by crrev.com/2028303002. Will likely add an additional change to blacklist the low-end Geforce 320M, which should address the remaining issues.
Labels: -performance-sheriff Merge-Approved-52 Performance-Sheriff
Merge of blacklist in https://codereview.chromium.org/2066733003/ is approved after pinging with ericrk@.
Project Member

Comment 8 by bugdroid1@chromium.org, Jun 24 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/5033c4f6e97a466f03132ed2cbb655a8807e415c

commit 5033c4f6e97a466f03132ed2cbb655a8807e415c
Author: ericrk <ericrk@chromium.org>
Date: Fri Jun 24 19:36:11 2016

This change blacklists GPU rasterization for all GPUs on MacOS that have not had test coverage.

The GPUs which are not blacklisted are those from the following GPU series:
Intel: 6th and 7th Generation GPUs
NVidia: Geforce 7XX GPUs
NVidia: Geforce 6XX GPUs
ATI: R300 Series GPUs
ATI: R200 Series GPUs
ATI: HD 7XXX series GPUs

This gives us roughly 70% GPU Raster enablement on Mac.

BUG= 614468 , 613272 
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2066733003
Cr-Commit-Position: refs/heads/master@{#401933}

[modify] https://crrev.com/5033c4f6e97a466f03132ed2cbb655a8807e415c/gpu/config/software_rendering_list_json.cc

Project Member

Comment 9 by sheriffbot@chromium.org, Jun 25 2016

This issue has been approved for a merge. Please merge the fix to any appropriate branches as soon as possible!

If all merges have been completed, please remove any remaining Merge-Approved labels from this issue.

Thanks for your time! To disable nags, add the Disable-Nags label.

For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot
Project Member

Comment 10 by bugdroid1@chromium.org, Jun 27 2016

Labels: -merge-approved-52 merge-merged-2743
The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/aded6cdaef1f360957d54578bcfffbbc7bbb00b7

commit aded6cdaef1f360957d54578bcfffbbc7bbb00b7
Author: Eric Karl <ericrk@chromium.org>
Date: Mon Jun 27 23:47:36 2016

This change blacklists GPU rasterization for all GPUs on MacOS that have not had test coverage.

The GPUs which are not blacklisted are those from the following GPU series:
Intel: 6th and 7th Generation GPUs
NVidia: Geforce 7XX GPUs
NVidia: Geforce 6XX GPUs
ATI: R300 Series GPUs
ATI: R200 Series GPUs
ATI: HD 7XXX series GPUs

This gives us roughly 70% GPU Raster enablement on Mac.

BUG= 614468 , 613272 
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2066733003
Cr-Commit-Position: refs/heads/master@{#401933}
(cherry picked from commit 5033c4f6e97a466f03132ed2cbb655a8807e415c)

Review URL: https://codereview.chromium.org/2103593002 .

Cr-Commit-Position: refs/branch-heads/2743@{#496}
Cr-Branched-From: 2b3ae3b8090361f8af5a611712fc1a5ab2de53cb-refs/heads/master@{#394939}

[modify] https://crrev.com/aded6cdaef1f360957d54578bcfffbbc7bbb00b7/gpu/config/software_rendering_list_json.cc

Status: Fixed (was: Assigned)
All issues in the original regression have now been regressed either by fixing them, or by blacklisting GPU raster on systems with problematic GPUs.
Nit: chrome://gpu still references this bug, even though it's closed, and referred to a GPU -> GPU regression (where GPU was still much faster than raster). We should probably log a new bug for either widening the whitelist, or explaining why we don't plan to do so, and point chrome://gpu at that.

Sign in to add a comment