New issue
Advanced search Search tips
Note: Color blocks (like █ or █) mean that a user may not be available. Tooltip shows the reason.

Issue 741733 link

Starred by 2 users
Status: Fixed
Owner:
kouhei@chromium.org
Closed: Aug 2017
Cc:
█ pmeenan@chromium.org
kouhei@chromium.org
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

19.2%-48.7% regression in loading.desktop at 484844:484953

Project Member Reported by pmeenan@chromium.org, Jul 12 2017 
See the link to graphs below.
Project Member

Comment 1 by 42576172...@developer.gserviceaccount.com, Jul 12 2017 

All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=741733

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=48d5d31b6002e720845b24bd244dae1fbc207449321d2531cfc267db19258c64


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

chromium-rel-win7-gpu-intel
chromium-rel-win7-gpu-nvidia
chromium-rel-win8-dual
win-high-dpi
Project Member

Comment 3 by 42576172...@developer.gserviceaccount.com, Jul 12 2017 


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

Bisect Details
  Configuration: win_8_perf_bisect
  Benchmark    : page_cycler_v2_site_isolation.basic_oopif
  Metric       : timeToFirstMeaningfulPaint_avg/pcv1-warm/http___www.nationalgeographic.com_

Revision             Result                  N
chromium@484843      392.843 +- 431.298      21      good
chromium@484904      429.423 +- 29.1487      20      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...www.nationalgeographic.com. page_cycler_v2_site_isolation.basic_oopif

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8974256312266082256


For feedback, file a bug with component Speed>Bisection
Project Member

Comment 5 by 42576172...@developer.gserviceaccount.com, Jul 12 2017 


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

Bisect Details
  Configuration: win_8_perf_bisect
  Benchmark    : page_cycler_v2_site_isolation.basic_oopif
  Metric       : timeToFirstMeaningfulPaint_avg/pcv1-warm/http___www.nationalgeographic.com_

Revision             Result                  N
chromium@484843      417.315 +- 395.115      21      good
chromium@484904      430.061 +- 32.4756      21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...www.nationalgeographic.com. page_cycler_v2_site_isolation.basic_oopif

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8974244664764085488


For feedback, file a bug with component Speed>Bisection
Project Member

Comment 7 by 42576172...@developer.gserviceaccount.com, Jul 13 2017 


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

Bisect Details
  Configuration: win_8_perf_bisect
  Benchmark    : page_cycler_v2_site_isolation.basic_oopif
  Metric       : timeToFirstMeaningfulPaint_avg/pcv1-warm/http___www.nationalgeographic.com_

Revision             Result                  N
chromium@484843      381.329 +- 395.365      20      good
chromium@484904      432.019 +- 29.1558      21      bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...www.nationalgeographic.com. page_cycler_v2_site_isolation.basic_oopif

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8974178717345457824


For feedback, file a bug with component Speed>Bisection
Project Member

Comment 9 by 42576172...@developer.gserviceaccount.com, Jul 13 2017 


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

Bisect Details
  Configuration: winx64intel_perf_bisect
  Benchmark    : loading.desktop
  Metric       : timeToFirstMeaningfulPaint_avg/pcv1-warm/MLB

Revision             Result                  N
chromium@484887      1432.44 +- 591.815      21      good
chromium@484953      1484.41 +- 776.838      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 --story-filter=MLB loading.desktop

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8974169576099865232


For feedback, file a bug with component Speed>Bisection
Project Member

Comment 11 by 42576172...@developer.gserviceaccount.com, Jul 14 2017 


=== BISECT JOB RESULTS ===
Perf regression found but unable to narrow commit range

Build failures prevented the bisect from narrowing the range further.


Bisect Details
  Configuration: win_8_perf_bisect
  Benchmark    : page_cycler_v2_site_isolation.basic_oopif
  Metric       : timeToFirstMeaningfulPaint_avg/pcv1-warm/http___www.nationalgeographic.com_
  Change       : 27.79% | 338.704523809 -> 432.822142857

Suspected Commit Range
  10 commits in range
  https://chromium.googlesource.com/chromium/src/+log/18cea352aa57b1bd0e57e7e60c04cba80728542d..ecd6201eef445e450f2c32aec9841c37e8f3afbd


Revision             Result                  N
chromium@484843      338.705 +- 351.2        21       good
chromium@484859      383.257 +- 316.745      13       good
chromium@484862      327.46 +- 162.087       9        good
chromium@484863      ---                     ---      build failure
chromium@484864      ---                     ---      build failure
chromium@484865      ---                     ---      build failure
chromium@484866      ---                     ---      build failure
chromium@484867      ---                     ---      build failure
chromium@484868      ---                     ---      build failure
chromium@484869      ---                     ---      build failure
chromium@484870      ---                     ---      build failure
chromium@484871      ---                     ---      build failure
chromium@484872      431.305 +- 14.1133      6        bad
chromium@484873      422.985 +- 114.666      9        bad
chromium@484874      425.384 +- 22.9903      14       bad
chromium@484904      432.822 +- 18.9935      14       bad

To Run This Test
  src/tools/perf/run_benchmark -v --browser=release --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...www.nationalgeographic.com. page_cycler_v2_site_isolation.basic_oopif

More information on addressing performance regressions:
  http://g.co/ChromePerformanceRegressions

Debug information about this bisect:
  https://chromeperf.appspot.com/buildbucket_job_status/8974161090022758976


For feedback, file a bug with component Speed>Bisection

Comment 12 by sullivan@chromium.org, Aug 21 2017

Cc: kouhei@chromium.org
Looks like the test is bimodal? The graphs seems to go from noisy to less noisy, and the bisections in #5 and #11 were on the same bot with pretty different results. +kouhei, owner of loading.desktop, any thoughts on what to do here?

Comment 13 by kouhei@chromium.org, Aug 22 2017

Owner: delph...@chromium.org
Status: Assigned (was: Untriaged)
A scheduler change is included in #c11 range.
https://chromium.googlesource.com/chromium/src/+/67fac36a35318d03fe141887898821e42dfaecb4

delphick: Would you mind looking into traces before/after the change?

Comment 14 by delph...@chromium.org, Aug 22 2017

Owner: kouhei@chromium.org
That change was rolled back: https://chromium-review.googlesource.com/c/chromium/src/+/573902

Comment 15 by kouhei@chromium.org, Aug 23 2017

Status: Fixed (was: Assigned)
►

Sign in to add a comment