Issue metadata
Sign in to add a comment
|
6.2%-4924.4% regression in loading.desktop at 504587:504697 |
||||||||||||||||||||
Issue description500+ alerts in loading.desktop. The only CL that jumps out to me was a catapult roll which includes changes in loading cache. But hopefully bisect will point in the right direction. https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/0ce28daf74eb..d08152f8a5da
,
Oct 2 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8966831023599651104
,
Oct 2 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: winx64intel_perf_bisect Benchmark : loading.desktop Metric : timeToFirstContentfulPaint_avg/warm/FlipKart Change : 4750.55% | 23.0548333327 -> 1118.2865 Suspected Commit Range 8 commits in range https://chromium.googlesource.com/chromium/src/+log/2e8a148e6c387088004f188c1e5f66169d51a340..b64a18def16cc3dd0a034e39adedcdd092e718f4 Revision Result N chromium@504625 23.0548 +- 8.90928 6 good chromium@504657 22.1998 +- 5.73203 6 good chromium@504667 22.9755 +- 5.6022 6 good chromium@504669 24.8462 +- 5.52138 6 good chromium@504670 --- --- build failure chromium@504671 --- --- build failure chromium@504672 --- --- build failure chromium@504673 --- --- build failure chromium@504674 --- --- build failure chromium@504675 --- --- build failure chromium@504676 --- --- build failure chromium@504677 1126.91 +- 105.307 6 bad chromium@504689 1118.29 +- 49.0269 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 --story-filter=FlipKart loading.desktop More information on addressing performance regressions: http://g.co/ChromePerformanceRegressions Debug information about this bisect: https://chromeperf.appspot.com/buildbucket_job_status/8966831023599651104 For feedback, file a bug with component Speed>Bisection
,
Oct 3 2017
,
Oct 3 2017
MarkTelemetryInternal() is not working; loading metrics for cache warming page loads are not ignored after the catapult roll.
,
Oct 3 2017
Issue 770730 has been merged into this issue.
,
Oct 3 2017
The following revision refers to this bug: https://chromium.googlesource.com/catapult/+/b3111d89694181773b830dfebcbc868594cd4be0 commit b3111d89694181773b830dfebcbc868594cd4be0 Author: Kunihiko Sakamoto <ksakamoto@chromium.org> Date: Tue Oct 03 12:18:52 2017 Fix loading metrics after 7c166a82 Catapult 7c166a82 changed the marker string of cache warming periods to "telemetry.internal.warm_cache.{warm,hot}.{start,end}", but it does not match the regexp in ChromeProcessHelper.findTelemetryInternalRanges_. As a result, loading metrics did not ignore cache warming page loads. To fix this, This patch relaxes the regexp to allow string with dots between "telemetry.internal." and ".{start,end}". Bug: chromium:770726 Change-Id: I0049b5b9c4787bfeba1928ee7c9242929682e9ca Reviewed-on: https://chromium-review.googlesource.com/696827 Reviewed-by: Ned Nguyen <nednguyen@google.com> Commit-Queue: Ned Nguyen <nednguyen@google.com> [modify] https://crrev.com/b3111d89694181773b830dfebcbc868594cd4be0/tracing/tracing/metrics/system_health/loading_metric_test.html [modify] https://crrev.com/b3111d89694181773b830dfebcbc868594cd4be0/tracing/tracing/model/helpers/chrome_process_helper.html [modify] https://crrev.com/b3111d89694181773b830dfebcbc868594cd4be0/telemetry/telemetry/page/cache_temperature.py
,
Jan 5 2018
Marking fixed after CL land. |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by 42576172...@developer.gserviceaccount.com
, Oct 2 2017