Issue metadata
Sign in to add a comment
|
timeToFirstContentfulPaint_avg / pcv1-warm dropped down to zero |
||||||||||||||||||||||
Issue descriptionIt seems that timeToFirstContentfulPaint_avg / pcv1-warm dropped down to zero in a lot of PCV2 benchmarks. https://codereview.chromium.org/2467763002 shows in the range, so I suspect this might be related. This is causing all sort of alerts to be triggered
,
Nov 3 2016
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8997022614293043888
,
Nov 4 2016
===== BISECT JOB RESULTS ===== Status: completed ===== SUSPECTED CL(s) ===== Subject : [loading metric] filter out events during cache warming Author : ksakamoto Commit description: The regexp in prepareTelemetryInternalEventPredicate did not match the titles of events logged by cache_temperature.py. Note for perf sheriffs: Improvements are expected in page_cycler_v2 benchmarks. BUG=catapult:#2962 Review-Url: https://codereview.chromium.org/2467763002 Commit : 0d3575d73a471a53e1d1d363a349ad4536f293fc Date : Tue Nov 01 12:17:02 2016 ===== TESTED REVISIONS ===== Revision Mean Std Dev N Good? chromium@428987 1473.49 195.829 5 good chromium@428992 1404.57 134.056 5 good chromium@428992,catapult@0d3575d73a 63.1824 1.50858 5 bad <-- chromium@428993 63.5284 1.72905 5 bad chromium@428994 62.2726 1.6931 5 bad chromium@428995 62.5732 0.619138 5 bad chromium@428997 62.7216 1.041 5 bad chromium@429007 62.4398 0.619536 5 bad Bisect job ran on: winx64ati_perf_bisect Bug ID: 661925 Test Command: src/tools/perf/run_benchmark -v --browser=release_x64 --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests page_cycler_v2.typical_25 Test Metric: timeToFirstContentfulPaint_avg/pcv1-warm/http___colorado.edu Relative Change: 95.76% Score: 99.9 Buildbot stdio: http://build.chromium.org/p/tryserver.chromium.perf/builders/winx64ati_perf_bisect/builds/1637 Job details: https://chromeperf.appspot.com/buildbucket_job_status/8997022614293043888 Not what you expected? We'll investigate and get back to you! https://chromeperf.appspot.com/bad_bisect?try_job_id=5489650171379712 | 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!
,
Nov 4 2016
This is expected, because now the page is very quickly loaded from cache (the value is close to zero, but not really zero). There was a bug that PCv2 recorded values from cache-warming page loads too. |
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by primiano@chromium.org
, Nov 3 2016