New issue
Advanced search Search tips

Issue 660232 link

Starred by 2 users

Issue metadata

Status: WontFix
Owner: ----
Closed: Jan 2017
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression

Blocked on:
issue 676416



Sign in to add a comment

1.7%-5.6% regression in media.tough_video_cases_extra at 427431:427540

Project Member Reported by chcunningham@google.com, Oct 27 2016

Issue description

Increase in seek time on several win7 tests. 
 
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Oct 28 2016


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


=== Bisection aborted ===
The bisect was aborted because The metric values for the initial "good" and "bad" revisions do not represent a clear regression.
Please contact the the team (see below) if you believe this is in error.

=== Warnings ===
The following warnings were raised by the bisect job:

 * Bisect failed to reproduce the regression with enough confidence.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@427457  581.639  10.9093  18  good
chromium@427512  586.388  11.9548  18  bad

Bisect job ran on: winx64ati_perf_bisect
Bug ID: 660232

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra
Test Metric: seek/garden2_10s.webm_seek_cold
Relative Change: 3.04%
Score: 0

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64ati_perf_bisect/builds/1629
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8997604683281150528


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5238738303582208

| 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 failed: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64ati_perf_bisect/builds/1634
Failure reason: the build has failed.
Additional errors:
The metric was not found in the test output.
Either of the initial "good" or "bad" revisions failed to be tested or built.

Bisect failed: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_perf_bisect/builds/7031
Failure reason: the build has failed.
Additional errors:
The metric was not found in the test output.
Either of the initial "good" or "bad" revisions failed to be tested or built.

Bisect failed: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_perf_bisect/builds/7030
Failure reason: the build has failed.
Additional errors:
The metric was not found in the test output.
Either of the initial "good" or "bad" revisions failed to be tested or built.

Bisect failed: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_perf_bisect/builds/7032
Failure reason: the build has failed.
Additional errors:
The metric was not found in the test output.
Either of the initial "good" or "bad" revisions failed to be tested or built.

Labels: Performance
Project Member

Comment 14 by 42576172...@developer.gserviceaccount.com, Dec 20 2016


===== BISECT JOB RESULTS =====
Status: failed


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: No values were found while testing the reference range.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean  Std Dev  N  Good?
chromium@427440  N/A   N/A      0  good
chromium@427524  N/A   N/A      0  bad

Bisect job ran on: winx64intel_perf_bisect
Bug ID: 660232

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=crowd720.vp9.webm.seek.cold media.tough_video_cases_extra
Test Metric: seek/crowd720_vp9.webm_seek_cold
Relative Change: 0.00%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64intel_perf_bisect/builds/1260
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8992796582593918784


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5840479969607680

| 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!
The failure in #c14 was from https://github.com/catapult-project/catapult/issues/3103 which is fixed, retried bisect.
Project Member

Comment 17 by 42576172...@developer.gserviceaccount.com, Dec 22 2016

Bisect failed: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_x64_perf_bisect/builds/1558
Failure reason: the build has failed.

Blockedon: 676416

===== BISECT JOB RESULTS =====
Status: failed


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@427456  422.969  17.1735  21  good
chromium@427539  422.64   9.06038  21  bad

Bisect job ran on: winx64ati_perf_bisect
Bug ID: 660232

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra
Test Metric: seek/tulip2.webm_seek_cold
Relative Change: 0.08%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64ati_perf_bisect/builds/1678
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8991437933632737904


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5289194107174912

| 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: failed


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@427430  319.511  6.25071  21  good
chromium@427512  320.205  5.1059   21  bad

Bisect job ran on: win_x64_perf_bisect
Bug ID: 660232

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra
Test Metric: seek/garden2_10s.webm_seek_warm
Relative Change: 0.22%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_x64_perf_bisect/builds/1562
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8991437891466094256


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5775042820964352

| 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: failed


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@427430  979.528  22.3084  21  good
chromium@427512  979.998  12.2671  21  bad

Bisect job ran on: win_x64_perf_bisect
Bug ID: 660232

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra
Test Metric: seek/garden2_10s.mp4_seek_cold
Relative Change: 0.05%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_x64_perf_bisect/builds/1561
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8991437899241540800


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5805402871037952

| 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: failed


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@427491  1092.72  56.7192  21  good
chromium@427564  1101.76  56.9737  21  bad

Bisect job ran on: win_perf_bisect
Bug ID: 660232

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra
Test Metric: seek/garden2_10s.mp4_seek_cold
Relative Change: 0.83%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_perf_bisect/builds/7088
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8991437922466843056


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=6415094014017536

| 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: failed


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@427491  1002.07  69.9097  21  good
chromium@427564  1010.29  102.784  21  bad

Bisect job ran on: win_perf_bisect
Bug ID: 660232

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra
Test Metric: seek/tulip2.vp9.webm_seek_cold
Relative Change: 0.82%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_perf_bisect/builds/7090
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8991437881449213600


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5848042903699456

| 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: failed


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@427491  1021.62  47.2616  21  good
chromium@427564  1028.52  115.045  21  bad

Bisect job ran on: win_perf_bisect
Bug ID: 660232

Test Command: src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra
Test Metric: seek/video.html?src_garden2_10s.mp4
Relative Change: 0.67%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/win_perf_bisect/builds/7089
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8991437909943165872


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5870139302477824

| 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!
With the various bisect infra bugs fixed, I kicked off a bunch of bisects for the different graphs in this bug. Seems they're all "not confident". 

I'll focus on crowd720_vp9.webm_seek_cold (4.6% regression) and pull the starting range back.

===== BISECT JOB RESULTS =====
Status: failed


=== Bisection aborted ===
The bisect was aborted because Bisect cannot identify a culprit: Bisect failed to reproduce the regression with enough confidence.
Please contact the the team (see below) if you believe this is in error.

===== TESTED REVISIONS =====
Revision         Mean     Std Dev  N   Good?
chromium@427004  175.687  28.4183  21  good
chromium@427669  174.836  12.0083  21  bad

Bisect job ran on: winx64intel_perf_bisect
Bug ID: 660232

Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra
Test Metric: seek/crowd720_vp9.webm_seek_cold
Relative Change: 0.48%

Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64intel_perf_bisect/builds/1267
Job details: https://chromeperf.appspot.com/buildbucket_job_status/8991379283617826880


Not what you expected? We'll investigate and get back to you!
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5310191262760960

| 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!
Cc: -chcunningham@chromium.org sullivan@chromium.org
Mean value is still in the post-regression range even after moving the start revision way back into the "good" range. I notice that all of the tests in this alert are on win7 bots. Perhaps an OS update across all bots hurt seeking performance irrespective of chrome. I've filed a bad-bisect bug in case there is some other explanation. +annie - anything else for us to do here? Does the explanation sound plausible? Do we have a log of OS updates?
Same story for  Issue 660245 
I've found that 9/14 graphs show bumps in the ref metric as well, so I'll "ignore" these 9. Still digging into the other 5.

chromium-rel-win7-x64-dual seek/video.html?src_garden2_10s.mp4 2.8%
https://chromeperf.appspot.com/report?sid=6777b598dbfa7f13eacf1020cdd2826fc4c246cb89acc606cebb665421191c9f&start_rev=425500&end_rev=437766

chromium-rel-win7-x64-dual seek/video.html?src_garden2_10s.webm 2.8%
https://chromeperf.appspot.com/report?sid=fcad808609c8922178050c6b53c6e25eb207755f83b57b87c77e5c4c218fca06&start_rev=425500&end_rev=437766

chromium-rel-win7-dual seek/tulip2.vp9.webm_seek_cold 4.1%
https://chromeperf.appspot.com/report?sid=fd9ac14032789891a8f8b8651bdd3c8627c5e8b9af1ae36378a662d61aa9b183&start_rev=425531&end_rev=435724

chromium-rel-win7-x64-dual seek/garden2_10s.mp4_seek_warm 2.8%
https://chromeperf.appspot.com/report?sid=031956257efae8dff061ede6ebaf93291ae993ef53ae3568d65345db49b62625&start_rev=422347&end_rev=437766

chromium-rel-win7-x64-dual seek/garden2_10s.webm_seek_warm 2.8%
https://chromeperf.appspot.com/report?sid=1a4d1fe483151557b155d5fe973455743e7b5010ef5dc02e4b7fbd716fcf5a28&start_rev=425500&end_rev=437766

chromium-rel-win7-x64-dual seek/garden2_10s.mp4_seek_cold 1.9%
https://chromeperf.appspot.com/report?sid=a0e349ed188bd9d488a4c30d1449e567239538c4346175355231d5070a08b31e&start_rev=425500&end_rev=437766

chromium-rel-win7-dual seek/garden2_10s.mp4_seek_warm 2.4%
https://chromeperf.appspot.com/report?sid=fa620fae918e2a2abc893e50120881df357ae121554d5345969eaa69d8b89037&start_rev=425500&end_rev=437766

chromium-rel-win7-dual seek/video.html?src_garden2_10s.mp4 2.4%
https://chromeperf.appspot.com/report?sid=3df10701f9bbc441dd37ecb3d4b80ae1d8d771ccb79bc298bdbae808e7e90dee&start_rev=425500&end_rev=437766

chromium-rel-win7-dual seek/garden2_10s.mp4_seek_cold 1.9%
https://chromeperf.appspot.com/report?sid=c2a57576bf019d7bc2edab6dc4a83c3cb6dd06f83af6150cbb73d139f46ee82b&start_rev=425500&end_rev=437766
Also ignoring this one. Ref shows a bump (somewhat smaller/less abrupt, but def there)

chromium-rel-win7-gpu-intel seek/crowd720_vp9.webm_seek_cold 4.6%
https://chromeperf.appspot.com/report?sid=f5f35a83d15df644e5f3c92f191b74bdfee349beccb8a158389d8ca251c11dfc&start_rev=425500&end_rev=437766
Here's the breakdown for the final 4 alerts....


For these the ref shows no bump. Even stranger, about 10k revisions later the ref value makes a sudden drop (amount varries) while ref reamins fixed at higher mean. Will kick off a bisect for all of these.

chromium-rel-win7-x64-dual seek/tulip2.vp9.webm_seek_cold 4.3%
https://chromeperf.appspot.com/report?sid=537de83ad81737003507f9d3d3ec8fd0a02d85489491f33691673aa9071e3f89&start_rev=425500&end_rev=437766

chromium-rel-win7-gpu-nvidia seek/tulip2.webm_seek_cold 1.7%
https://chromeperf.appspot.com/report?sid=47c53d2d9f138a9f6b33d9ce6f6624673b482ae3771e5aae17eedb67eb70fdcc&start_rev=425500&end_rev=437766


chromium-rel-win7-gpu-ati seek/tulip2.webm_seek_cold 2.1%
https://chromeperf.appspot.com/report?sid=f6274f360b5cf0e369ac34d21db1b7d47599c5626a8a2401d10ef82511f66379&start_rev=425500&end_rev=437766



This one shows a bump in ref too (so can ignore) but about 10k revisions later the NON-REF drops by over 50%. Will bisect that improvement out of curiousity. 

chromium-rel-win7-gpu-ati seek/garden2_10s.webm_seek_cold 5.6%
https://chromeperf.appspot.com/report?sid=e4911f4d3a649460a2bd93d7fe1704f185ecdd42a6cd6ba0faace6ab9f89fc4c&start_rev=425500&end_rev=437766

=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : media.tough_video_cases_extra
  Metric       : seek/tulip2.webm_seek_cold

Revision             Result                  N
chromium@426963      421.503 +- 9.24532      21      good
chromium@428614      422.513 +- 12.8836      21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8991278167471798064

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5844211121782784


| 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 ===
NO Perf regression found

Bisect Details
  Configuration: winx64nvidia_perf_bisect
  Benchmark    : media.tough_video_cases_extra
  Metric       : seek/tulip2.webm_seek_cold

Revision             Result                  N
chromium@426962      422.383 +- 7.93319      21      good
chromium@428614      422.702 +- 6.226        21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8991278191337455536

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5850086905479168


| 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!
Cc: xiaoche...@chromium.org
Owner: xiaoche...@chromium.org

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

Hi xiaochengh@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 : xiaochengh
  Commit : bc320afb51a662a92f69369d818701dd1898d1cc
  Date   : Tue Oct 18 08:36:24 2016
  Subject: Audit use of updateStyleAndLayoutIgnorePendingStylesheets in FrameSelection::updateIfNeeded

Bisect Details
  Configuration: win_x64_perf_bisect
  Benchmark    : media.tough_video_cases_extra
  Metric       : seek/tulip2.vp9.webm_seek_cold
  Change       : 2.20% | 994.9125 -> 973.026666667

Revision             Result                  N
chromium@425500      994.912 +- 26.004       6       good
chromium@425891      986.437 +- 12.9936      9       good
chromium@425916      1000.42 +- 9.97813      6       good
chromium@425919      989.961 +- 21.1164      9       good
chromium@425921      1013.07 +- 59.1914      6       good
chromium@425922      974.575 +- 12.9101      6       bad       <--
chromium@425928      969.563 +- 18.3044      6       bad
chromium@425940      966.738 +- 10.4628      6       bad
chromium@425989      969.667 +- 7.27364      6       bad
chromium@426086      985.006 +- 123.324      14      bad
chromium@426281      974.93 +- 15.2047       9       bad
chromium@427061      972.742 +- 11.1953      6       bad
chromium@428622      973.027 +- 15.064       6       bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8991278229898075392

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=4931136239697920


| 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!
Owner: ----
Status: Untriaged (was: Assigned)
My CL only hoists the call site of a function in the call stack without changing any behavior. It shouldn't affect the performance.
Status: WontFix (was: Untriaged)
Giving up. None of the bumps are reproduceable. The one found culprit is bogus. Filed bad bisect for that one (even just looking at the list of means in comment 44, the determination of "bad" from "good" seems random). 
Project Member

Comment 50 by 42576172...@developer.gserviceaccount.com, Apr 11 2017


=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: winx64ati_perf_bisect
  Benchmark    : media.tough_video_cases_extra
  Metric       : seek/tulip2.webm_seek_cold

Revision             Result                  N
chromium@427456      422.728 +- 9.74508      21      good
chromium@427539      423.594 +- 11.206       21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8982624342681102400

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5289194107174912


| 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 52 by 42576172...@developer.gserviceaccount.com, Apr 11 2017


=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: win_x64_perf_bisect
  Benchmark    : media.tough_video_cases_extra
  Metric       : seek/garden2_10s.mp4_seek_warm

Revision             Result                  N
chromium@427430      979.368 +- 22.0665      21      good
chromium@427512      981.47 +- 20.5414       21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8982624133816276608

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5302073741017088


| 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 54 by 42576172...@developer.gserviceaccount.com, Apr 11 2017


=== BISECT JOB RESULTS ===
NO Perf regression found

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : media.tough_video_cases_extra
  Metric       : seek/crowd720_vp9.webm_seek_cold

Revision             Result                  N
chromium@427004      178.022 +- 11.8002      21      good
chromium@427669      180.115 +- 25.0646      21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests media.tough_video_cases_extra

Debug Info
  https://chromeperf.appspot.com/buildbucket_job_status/8982620258140164096

Is this bisect wrong?
  https://chromeperf.appspot.com/bad_bisect?try_job_id=5310191262760960


| 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!

Sign in to add a comment