Issue metadata
Sign in to add a comment
|
1.7%-2% regression in jetstream at 536349:536534 |
||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Feb 21 2018
📍 Pinpoint job started. https://pinpoint-dot-chromeperf.appspot.com/job/16b98af7840000
,
Feb 21 2018
📍 Found significant differences after each of 2 commits. https://pinpoint-dot-chromeperf.appspot.com/job/16b98af7840000 [turbofan] Masking/poisoning in codegen (optimized code, x64) by mvstanton@chromium.org https://chromium.googlesource.com/v8/v8/+/8f489e73b23f764801a35abaaf10772406fd4209 telemetry: Remove unused LoadTimesTimelineMetric metric. by sadrul@chromium.org https://chromium.googlesource.com/chromium/src/+/dcbee62f9f44f438edadef4bc08a8351518ee83b Understanding performance regressions: http://g.co/ChromePerformanceRegressions
,
Feb 22 2018
https://chromium.googlesource.com/chromium/src/+/dcbee62f9f44f438edadef4bc08a8351518ee83b removes dead code. Assigning to author of the other CL.
,
Mar 23 2018
,
Mar 23 2018
Issue 814168 has been merged into this issue.
,
Mar 23 2018
This is fixed by the following commit: [turbofan] Only lower to PoisonedLoads when --branch-load-poisoning is on. https://chromium.googlesource.com/v8/v8/+/a4353d1464f5c546774f1ae08d8eebfa31fde963 The problem was we were using the infrastructure from my CL even when branch poisoning wasn't turned on.
,
Mar 23 2018
Issue 814054 has been merged into this issue. |
|||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||
Comment 1 by 42576172...@developer.gserviceaccount.com
, Feb 21 2018