New issue
Advanced search Search tips

Issue 609771 link

Starred by 0 users

Issue metadata

Status: Duplicate
Merged: issue 609517
Owner:
Closed: May 2016
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

9.3% regression in scheduler.tough_scheduling_cases at 391631:391676

Project Member Reported by alexclarke@chromium.org, May 6 2016

Issue description

See the link to graphs below.
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=609771

Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?keys=agxzfmNocm9tZXBlcmZyFAsSB0Fub21hbHkYgICg1MeWowoM


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

chromium-rel-mac-hdd
Cc: vmp...@chromium.org
Owner: vmp...@chromium.org

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

Hi vmpstr@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 : cc: Move prepaint region calculations to the tiling set.
Author  : vmpstr
Commit description:
  
This patch moves prepaint region calculations from the tiling to the
tiling set. This ensures that new tilings get proper skewport
information. It also allows us to retrieve prepaint regions
independently of tilings present, which is useful for iterating images.

R=enne, danakj
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review-Url: https://codereview.chromium.org/1939963002
Cr-Commit-Position: refs/heads/master@{#391645}
Commit  : cf1113301b3c251c09ac7f7576b4f98bebf64f25
Date    : Wed May 04 21:23:29 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev    N  Good?
chromium@391588  4.54725  0.0333599  5  good
chromium@391617  4.56101  0.0375449  5  good
chromium@391632  4.60443  0.0170179  5  good
chromium@391639  4.57992  0.0620176  5  good
chromium@391643  4.59143  0.0439686  5  good
chromium@391644  4.54661  0.105345   5  good
chromium@391645  4.78397  0.0413551  5  bad    <--
chromium@391646  4.80713  0.0619678  5  bad

Bisect job ran on: android_nexus6_perf_bisect
Bug ID: 609771

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests thread_times.tough_scrolling_cases
Test Metric: thread_raster_cpu_time_per_frame/thread_raster_cpu_time_per_frame
Relative Change: 5.72%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus6_perf_bisect/builds/2165
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9013419693716696960


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5875456650248192

| 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!

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


===== SUSPECTED CL(s) =====
Subject : cc: Move prepaint region calculations to the tiling set.
Author  : vmpstr
Commit description:
  
This patch moves prepaint region calculations from the tiling to the
tiling set. This ensures that new tilings get proper skewport
information. It also allows us to retrieve prepaint regions
independently of tilings present, which is useful for iterating images.

R=enne, danakj
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review-Url: https://codereview.chromium.org/1939963002
Cr-Commit-Position: refs/heads/master@{#391645}
Commit  : cf1113301b3c251c09ac7f7576b4f98bebf64f25
Date    : Wed May 04 21:23:29 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev    N  Good?
chromium@391626  3.42673  0.0296794  5  good
chromium@391638  3.4508   0.0130002  5  good
chromium@391644  3.43665  0.0200728  5  good
chromium@391645  3.53908  0.0122075  5  bad    <--
chromium@391646  3.51496  0.0105809  5  bad
chromium@391647  3.54767  0.0114232  5  bad
chromium@391649  3.53584  0.0141471  5  bad
chromium@391671  3.54827  0.0233583  5  bad

Bisect job ran on: android_nexus5_perf_bisect
Bug ID: 609771

Test Command: src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --also-run-disabled-tests thread_times.key_silk_cases
Test Metric: thread_renderer_compositor_cpu_time_per_frame/thread_renderer_compositor_cpu_time_per_frame
Relative Change: 3.55%
Score: 99.9

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/android_nexus5_perf_bisect/builds/3670
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9013418092283348352


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6390498927837184

| 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!
Mergedinto: 609517
Status: Duplicate (was: Assigned)

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


===== SUSPECTED CL(s) =====
Subject : This change enables GPU Raster on OSX.
Author  : ericrk
Commit description:
  
I've compiled a bit of benchmark data to support turning this on -
https://docs.google.com/a/google.com/document/d/1RPEnillJOaOKW47t9DYuttQkvWDo7ilx9KDB_nJnwpc/edit?usp=sharing
Still filling in more data, but I'm pretty happy with the results so far. In
a number of cases we have a performance win, and in most cases we have a power win.

Checking this in now will give us a long canary period before the next branch
point to catch any issues.

BUG= 542959 
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;tryserver.blink:mac_blink_rel

Review-Url: https://codereview.chromium.org/1881053003
Cr-Commit-Position: refs/heads/master@{#391655}
Commit  : 8325892c2efdef8cff86ba16e7569f8b424daeda
Date    : Wed May 04 22:14:40 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N  Good?
chromium@391630  57.8679  1.94335   5  good
chromium@391653  55.9964  0.412265  5  good
chromium@391654  55.7151  0.912965  5  good
chromium@391655  61.4547  1.57307   5  bad    <--
chromium@391656  59.871   0.920765  5  bad
chromium@391659  60.6896  1.397     5  bad
chromium@391665  61.1668  1.27132   5  bad
chromium@391676  61.3284  0.429369  5  bad

Bisect job ran on: mac_hdd_perf_bisect
Bug ID: 609771

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests scheduler.tough_scheduling_cases
Test Metric: mean_input_event_latency/mean_input_event_latency
Relative Change: 5.98%
Score: 99.8

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_hdd_perf_bisect/builds/527
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9013419793539648016


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5210800260644864

| 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!

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


===== SUSPECTED CL(s) =====
Subject : This change enables GPU Raster on OSX.
Author  : ericrk
Commit description:
  
I've compiled a bit of benchmark data to support turning this on -
https://docs.google.com/a/google.com/document/d/1RPEnillJOaOKW47t9DYuttQkvWDo7ilx9KDB_nJnwpc/edit?usp=sharing
Still filling in more data, but I'm pretty happy with the results so far. In
a number of cases we have a performance win, and in most cases we have a power win.

Checking this in now will give us a long canary period before the next branch
point to catch any issues.

BUG= 542959 
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;tryserver.blink:mac_blink_rel

Review-Url: https://codereview.chromium.org/1881053003
Cr-Commit-Position: refs/heads/master@{#391655}
Commit  : 8325892c2efdef8cff86ba16e7569f8b424daeda
Date    : Wed May 04 22:14:40 2016


===== TESTED REVISIONS =====
Revision         Mean     Std Dev   N  Good?
chromium@391630  56.2121  0.5516    5  good
chromium@391653  56.4829  0.921786  5  good
chromium@391654  55.9102  0.559729  5  good
chromium@391655  60.9058  1.43209   5  bad    <--
chromium@391656  61.5142  0.803694  5  bad
chromium@391659  60.6398  1.17756   5  bad
chromium@391665  59.8081  0.921713  5  bad
chromium@391676  61.4524  1.04426   5  bad

Bisect job ran on: mac_hdd_perf_bisect
Bug ID: 609771

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --also-run-disabled-tests scheduler.tough_scheduling_cases
Test Metric: mean_input_event_latency/mean_input_event_latency
Relative Change: 9.32%
Score: 99.8

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/mac_hdd_perf_bisect/builds/526
Job details: https://chromeperf.appspot.com/buildbucket_job_status/9013419802032934432


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5319950076477440

| 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!

Sign in to add a comment