New issue
Advanced search Search tips

Issue 859885 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Jul 3
Cc:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug-Regression



Sign in to add a comment

80.8% regression in mean_main_thread_scroll_latency at 571800:571819

Project Member Reported by 42576172...@developer.gserviceaccount.com, Jul 3

Issue description

See the link to graphs below.
 
All graphs for this bug:
  https://chromeperf.appspot.com/group_report?bug_id=859885

(For debugging:) Original alerts at time of bug-filing:
  https://chromeperf.appspot.com/group_report?sid=68363ad95354707a7919cb101a312650034a14d442a43ff565d8e46c112df8af


Bot(s) for this bug's original alert(s):

android-nexus5
Cc: wangxianzhu@chromium.org
Owner: wangxianzhu@chromium.org
Status: Assigned (was: Untriaged)
📍 Found a significant difference after 1 commit.
https://pinpoint-dot-chromeperf.appspot.com/job/16a5a688a40000

[PE] Don't call PaintController::FinishCycle() if not repainted by wangxianzhu@chromium.org
https://chromium.googlesource.com/chromium/src/+/e71a3846080ff6cc758f39c2fea5c5a903572e61
39.99 → 79.15 (+39.16)

Understanding performance regressions:
  http://g.co/ChromePerformanceRegressions
Project Member

Comment 4 by bugdroid1@chromium.org, Jul 3

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/74780299ea1b72a9f8b6415943213c0cf0f7ee23

commit 74780299ea1b72a9f8b6415943213c0cf0f7ee23
Author: Xianzhu Wang <wangxianzhu@chromium.org>
Date: Tue Jul 03 18:40:12 2018

Revert "[PE] Don't call PaintController::FinishCycle() if not repainted"

This reverts commit e71a3846080ff6cc758f39c2fea5c5a903572e61.

Reason for revert: Caused performance regressions because changed
paint properties are not cleared changed flags.

Original change's description:
> [PE] Don't call PaintController::FinishCycle() if not repainted
> 
> This is mainly a performance optimization because we don't need
> FinishCycle() if PaintController has nothing changed.
> 
> By the way it can also avoid the immediate bad effect (but not the
> root cause) of  bug 859294 .
> 
> Bug:  859294 
> Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_slimming_paint_v2;master.tryserver.blink:linux_trusty_blink_rel
> Change-Id: Iad3e14ee303c08185922d3d6e495c3b25ead5434
> Reviewed-on: https://chromium-review.googlesource.com/1121783
> Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#571803}

TBR=wangxianzhu@chromium.org,chrishtr@chromium.org

# Not skipping CQ checks because original CL landed > 1 day ago.

Bug:  859294 ,  859885 
Change-Id: I2a4d2a7d09cc3bf5edfbf0c6dee275581030c2c9
Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_slimming_paint_v2;master.tryserver.blink:linux_trusty_blink_rel
Reviewed-on: https://chromium-review.googlesource.com/1124701
Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org>
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#572302}
[modify] https://crrev.com/74780299ea1b72a9f8b6415943213c0cf0f7ee23/third_party/blink/renderer/core/frame/local_frame_view.cc
[modify] https://crrev.com/74780299ea1b72a9f8b6415943213c0cf0f7ee23/third_party/blink/renderer/core/frame/local_frame_view.h
[modify] https://crrev.com/74780299ea1b72a9f8b6415943213c0cf0f7ee23/third_party/blink/renderer/platform/graphics/graphics_layer.cc

Status: Fixed (was: Assigned)

Sign in to add a comment