blink_perf.svg/SierpinskiCarpet.html in blink_perf.svg crashing on multiple builders |
|||
Issue descriptionFiled by sheriff-o-matic@appspot.gserviceaccount.com on behalf of sullivan@google.com blink_perf.svg/SierpinskiCarpet.html in blink_perf.svg failing on multiple builders Builders failed on: - Mac 10.12 Perf: https://ci.chromium.org/buildbot/chromium.perf/Mac%2010.12%20Perf - Mac Air 10.11 Perf: https://ci.chromium.org/buildbot/chromium.perf/Mac%20Air%2010.11%20Perf - Mac Pro 10.11 Perf: https://ci.chromium.org/buildbot/chromium.perf/Mac%20Pro%2010.11%20Perf - Win 10 High-DPI Perf: https://ci.chromium.org/buildbot/chromium.perf/Win%2010%20High-DPI%20Perf - Win 10 Perf: https://ci.chromium.org/buildbot/chromium.perf/Win%2010%20Perf - Win 7 Nvidia GPU Perf: https://ci.chromium.org/buildbot/chromium.perf/Win%207%20Nvidia%20GPU%20Perf Sample log: https://logs.chromium.org/v/?s=chrome%2Fbb%2Fchromium.perf%2FMac_10.12_Perf%2F2088%2F%2B%2Frecipes%2Fsteps%2Fblink_perf.svg_on_Intel_GPU_on_Mac_on_Mac-10.12%2F0%2Fstdout Output is too long to paste, but it has a minidump at https://console.developers.google.com/m/cloudstorage/b/chrome-telemetry-output/o/7550bcc0-5e88-11e8-ba19-985aebcd7ca1 and you can read the output to see stacks. Disabling test and kicking off a bisect now.
,
May 23 2018
,
May 23 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/4fffb41bc1da1bdafadb20c0f90fc7eb615a1a2d commit 4fffb41bc1da1bdafadb20c0f90fc7eb615a1a2d Author: Annie Sullivan <sullivan@chromium.org> Date: Wed May 23 20:36:58 2018 Disable crashing SierpinskiCarpet story on Mac/Win. Bug: 846061 Change-Id: Ibcfd0cda00575ee5d6717bbc0fc0bda6cfb4895e TBR: charliea@chromium.org NOTRY: true Reviewed-on: https://chromium-review.googlesource.com/1070561 Reviewed-by: Annie Sullivan <sullivan@chromium.org> Commit-Queue: Annie Sullivan <sullivan@chromium.org> Cr-Commit-Position: refs/heads/master@{#561237} [modify] https://crrev.com/4fffb41bc1da1bdafadb20c0f90fc7eb615a1a2d/tools/perf/expectations.config
,
May 24 2018
📍 Found a significant difference after 1 commit. https://pinpoint-dot-chromeperf.appspot.com/job/14e30802240000 [oilpan] Move CompleteSweep() to ScheduleGCIfNeeded() during allocation to play safe. by hpayer@chromium.org https://chromium.googlesource.com/chromium/src/+/5fa0b1a7de7547cb02efdb970f85b9ff25000bc8 Understanding performance regressions: http://g.co/ChromePerformanceRegressions
,
May 24 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/5c523fa4278f8fb9bf2ce4730851ab118e846c6d commit 5c523fa4278f8fb9bf2ce4730851ab118e846c6d Author: Hannes Payer <hpayer@chromium.org> Date: Thu May 24 11:51:32 2018 Revert "[oilpan] Move CompleteSweep() to ScheduleGCIfNeeded() during allocation to play safe." This reverts commit 5fa0b1a7de7547cb02efdb970f85b9ff25000bc8. Reason for revert: This is not correct since external memory reporting may 1) start an incremental V8 gc which 2) invokes wrapper tracing 3) which may fail if GC is forbidden, c.f. crbug/846093 I will look into this offline. Bug: chromium:846093 , chromium:846061 Original change's description: > [oilpan] Move CompleteSweep() to ScheduleGCIfNeeded() during allocation to play safe. > > CompleteSweep() needs to be called before we call ReportMemoryToV8(). This change > makes this dependency more explicit. > > Change-Id: I044af54bcfdc8a7457edbb492fba6bfe4ff210aa > Reviewed-on: https://chromium-review.googlesource.com/1069271 > Reviewed-by: Kentaro Hara <haraken@chromium.org> > Reviewed-by: Michael Lippautz <mlippautz@chromium.org> > Commit-Queue: Hannes Payer <hpayer@chromium.org> > Cr-Commit-Position: refs/heads/master@{#560987} TBR=haraken@chromium.org,hpayer@chromium.org,mlippautz@chromium.org # Not skipping CQ checks because original CL landed > 1 day ago. Change-Id: I06899e2e5bf0c4303afdd946fd3e6a99f2954530 Reviewed-on: https://chromium-review.googlesource.com/1071288 Reviewed-by: Kentaro Hara <haraken@chromium.org> Commit-Queue: Hannes Payer <hpayer@chromium.org> Cr-Commit-Position: refs/heads/master@{#561467} [modify] https://crrev.com/5c523fa4278f8fb9bf2ce4730851ab118e846c6d/third_party/blink/renderer/platform/heap/thread_state.cc
,
May 24 2018
hpayer: should I re-enable the test now that you've reverted? |
|||
►
Sign in to add a comment |
|||
Comment 1 by 42576172...@developer.gserviceaccount.com
, May 23 2018