Issue metadata
Sign in to add a comment
|
blink_perf.paint failing on 2 builders |
||||||||||||||||||||
Issue descriptionblink_perf.paint failing on 2 builders It appears to be frequently timing out [ FAILED ] appending-text.html [ FAILED ] large-table-repaint.html TimeoutException: Timed out while waiting 60s for IsJavaScriptExpressionTrue. Builders failed on: - Android Nexus5 Perf: https://build.chromium.org/p/chromium.perf/builders/Android%20Nexus5%20Perf - Android Nexus7v2 Perf: https://build.chromium.org/p/chromium.perf/builders/Android%20Nexus7v2%20Perf
,
Jul 31 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8972531149134544976
,
Aug 1 2017
=== Auto-CCing suspected CL author nya@chromium.org === Hi nya@chromium.org, the bisect results pointed to your CL, please take a look at the results. === BISECT JOB RESULTS === Test failure found with culprit Suspected Commit Author : Shuhei Takahashi Commit : af68cffd17ef07ce6756e4aacb7c88a2f877f155 Date : Mon Jul 24 05:17:12 2017 Subject: Add myself to watchlists for chrome/browser/chromeos/arc/fileapi. Bisect Details Configuration: android_nexus7_perf_bisect Benchmark : blink_perf.paint Metric : benchmark_duration/benchmark_duration Revision Exit Code N chromium@488897 0 +- N/A 3 good chromium@488908 0 +- N/A 3 good chromium@488913 0 +- N/A 3 good chromium@488914 1 +- N/A 3 bad <-- chromium@488915 1 +- N/A 3 bad chromium@488916 1 +- N/A 3 bad chromium@488918 1 +- N/A 3 bad chromium@488938 1 +- N/A 3 bad chromium@488978 1 +- N/A 3 bad To Run This Test src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.paint More information on addressing performance regressions: http://g.co/ChromePerformanceRegressions Debug information about this bisect: https://chromeperf.appspot.com/buildbucket_job_status/8972531236055789200 For feedback, file a bug with component Speed>Bisection
,
Aug 1 2017
I would be surprised if Blink performance is affected by WATCHLIST.
,
Aug 1 2017
=== BISECT JOB RESULTS === Perf regression found with culprit Suspected Commit Author : Naina Raisinghani Commit : fd7f99c9125b7f9b054b3a32c0100b526792f74c Date : Mon Jul 24 05:21:35 2017 Subject: Remove code and tests using StyleSharing Bisect Details Configuration: android_nexus5_perf_bisect Benchmark : blink_perf.paint Metric : benchmark_duration/benchmark_duration Change : 11.21% | 15.1031682167 -> 16.7967120753 Revision Result N chromium@488562 15.1032 +- 1.73026 6 good chromium@488819 14.7747 +- 0.207264 6 good chromium@488884 14.625 +- 0.265887 6 good chromium@488900 14.7855 +- 0.0868204 6 good chromium@488908 14.6974 +- 0.172426 6 good chromium@488912 14.7252 +- 0.181111 6 good chromium@488914 14.7224 +- 0.214658 6 good chromium@488915 16.8601 +- 0.266832 6 bad <-- chromium@488916 16.906 +- 0.135851 6 bad chromium@488948 16.8576 +- 0.314437 6 bad chromium@489076 16.7967 +- 0.351286 6 bad To Run This Test src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.paint More information on addressing performance regressions: http://g.co/ChromePerformanceRegressions Debug information about this bisect: https://chromeperf.appspot.com/buildbucket_job_status/8972531149134544976 For feedback, file a bug with component Speed>Bisection
,
Aug 1 2017
This patch removes dead code that was disabled in an earlier patch. Don't think its me.
,
Aug 2 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8972366510031531040
,
Aug 2 2017
Kicking off another benchmark_duration bisect. It is really strange that the benchmark takes ~2 minutes longer starting at that CL. nednguyen, any ideas?
,
Aug 2 2017
If test run is slower, the benchmark will surely be slower. So benchmark_duration sometimes do catch chrome binary regression.
,
Aug 2 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8972365080213411424
,
Aug 2 2017
This bug is about a benchmark failure on some bots, which happens in the same CL range as benchmark duration goes up by 2 minutes. I started another return_code bisect, but it seems really curious that the specific CL makes the benchmark take longer when the author believes it just removes dead code. Will see what the other bisects say.
,
Aug 2 2017
It could just be coincidental. I am not sure why you care about benchmark duration in a bug about benchmark failure.
,
Aug 2 2017
Someone else kicked off the bisect, just following up.
,
Aug 2 2017
2 minutes is a very small regression in term of benchmark cycle time, so this could all just be coincidental. Feel free to add me back if you find something deeper behind this.
,
Aug 3 2017
=== BISECT JOB RESULTS === Perf regression found with culprit Suspected Commit Author : Naina Raisinghani Commit : fd7f99c9125b7f9b054b3a32c0100b526792f74c Date : Mon Jul 24 05:21:35 2017 Subject: Remove code and tests using StyleSharing Bisect Details Configuration: android_nexus5_perf_bisect Benchmark : blink_perf.paint Metric : benchmark_duration/benchmark_duration Change : 12.92% | 15.0311802884 -> 16.9727400151 Revision Result N chromium@488899 15.0312 +- 1.62833 6 good chromium@488909 14.774 +- 0.180612 6 good chromium@488914 14.7064 +- 0.24107 6 good chromium@488915 16.9196 +- 0.235475 6 bad <-- chromium@488916 16.9389 +- 0.116163 6 bad chromium@488918 16.9245 +- 0.199072 6 bad chromium@488937 16.8888 +- 0.0985762 6 bad chromium@488975 16.9727 +- 0.190639 6 bad To Run This Test src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.paint More information on addressing performance regressions: http://g.co/ChromePerformanceRegressions Debug information about this bisect: https://chromeperf.appspot.com/buildbucket_job_status/8972366510031531040 For feedback, file a bug with component Speed>Bisection
,
Aug 3 2017
Still waiting for the test failure bisect to come back in.
,
Aug 3 2017
=== BISECT JOB RESULTS === Bisect was unable to run to completion Error: INFRA_FAILURE The bisect was able to narrow the range, you can try running with: good_revision: 32fc87a3e1ba26f1a4038c9583401e14d88e1b49 bad_revision : 6d4d89ab4573b0e513369c1d089c0ea9aada1239 If failures persist contact the team (see below) and report the error. Bisect Details Configuration: android_nexus5_perf_bisect Benchmark : blink_perf.paint Metric : appending-text/appending-text Revision Exit Code N chromium@488886 0 +- N/A 20 good chromium@488909 0 +- N/A 20 good chromium@488931 1 +- N/A 20 bad chromium@488975 1 +- N/A 20 bad To Run This Test src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.paint More information on addressing performance regressions: http://g.co/ChromePerformanceRegressions Debug information about this bisect: https://chromeperf.appspot.com/buildbucket_job_status/8972365080213411424 For feedback, file a bug with component Speed>Bisection
,
Aug 3 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8972272250987003760
,
Aug 3 2017
Re-running return code bisect with a wider range. It does seem to be reproducing the failure, but timed out.
,
Aug 4 2017
=== BISECT JOB RESULTS === Bisect was unable to run to completion Error: INFRA_FAILURE The bisect was able to narrow the range, you can try running with: good_revision: 32fc87a3e1ba26f1a4038c9583401e14d88e1b49 bad_revision : fd7f99c9125b7f9b054b3a32c0100b526792f74c If failures persist contact the team (see below) and report the error. Bisect Details Configuration: android_nexus5_perf_bisect Benchmark : blink_perf.paint Metric : appending-text/appending-text Revision Exit Code N chromium@488909 0 +- N/A 20 good chromium@488915 1 +- N/A 20 bad chromium@488920 1 +- N/A 20 bad chromium@488931 1 +- N/A 20 bad To Run This Test src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests blink_perf.paint More information on addressing performance regressions: http://g.co/ChromePerformanceRegressions Debug information about this bisect: https://chromeperf.appspot.com/buildbucket_job_status/8972272250987003760 For feedback, file a bug with component Speed>Bisection
,
Aug 6
This issue has been Available for over a year. If it's no longer important or seems unlikely to be fixed, please consider closing it out. If it is important, please re-triage the issue. Sorry for the inconvenience if the bug really should have been left as Available. For more details visit https://www.chromium.org/issue-tracking/autotriage - Your friendly Sheriffbot |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by 42576172...@developer.gserviceaccount.com
, Jul 31 2017