Issue metadata
Sign in to add a comment
|
2.1%-45% regression in smoothness.sync_scroll.key_mobile_sites_smooth at 457406:458752 |
||||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
Mar 23 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8984291445411915616
,
Mar 24 2017
=== Auto-CCing suspected CL author wangxianzhu@chromium.org === Hi wangxianzhu@chromium.org, the bisect results pointed to your CL, please take a look at the results. === BISECT JOB RESULTS === Perf regression found with culprit Suspected Commit Author : wangxianzhu Commit : 4e0d0f9ddd68e59dccda0a8c9a096c1813a1bd44 Date : Tue Mar 21 05:00:51 2017 Subject: Add StyleDifference::needsVisualRectUpdate Bisect Details Configuration: android_nexus5_perf_bisect Benchmark : smoothness.sync_scroll.key_mobile_sites_smooth Metric : mean_input_event_latency/http___mobile-news.sandbox.google.com_news_pt0 Change : 37.94% | 23.8551666667 -> 32.9051666667 Revision Result N chromium@457405 23.8552 +- 3.6566 6 good chromium@458079 25.8462 +- 9.31702 9 good chromium@458248 24.5262 +- 4.33493 6 good chromium@458290 25.0933 +- 5.43105 6 good chromium@458311 23.5018 +- 0.972929 6 good chromium@458312 24.0252 +- 2.76043 6 good chromium@458313 33.6768 +- 1.78101 6 bad <-- chromium@458314 33.8847 +- 1.65937 6 bad chromium@458317 34.2155 +- 1.31045 6 bad chromium@458322 34.4253 +- 1.28576 6 bad chromium@458332 33.4378 +- 0.839519 6 bad chromium@458416 34.2807 +- 1.55254 6 bad chromium@458752 32.9052 +- 4.30205 6 bad To Run This Test src/tools/perf/run_benchmark -v --browser=android-chromium --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...mobile.news.sandbox.google.com.news.pt0 smoothness.sync_scroll.key_mobile_sites_smooth Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8984291445411915616 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=5245705913892864 | 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 Speed>Bisection. Thank you!
,
Mar 27 2017
,
Mar 27 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/bedba279ba5dc64257efc7bc8a13d697c049d3de commit bedba279ba5dc64257efc7bc8a13d697c049d3de Author: wangxianzhu <wangxianzhu@chromium.org> Date: Mon Mar 27 18:11:13 2017 Revert of Add StyleDifference::needsVisualRectUpdate (patchset #5 id:80001 of https://codereview.chromium.org/2761863002/ ) Reason for revert: BUG= 704728 Original issue's description: > Add StyleDifference::needsVisualRectUpdate > > It's set when a style change will change visual rect (without changing > layout, as visual rect update is implied by layout change). > > BTW optimized paint invalidation of box with resizer. We no longer > force full paint invalidation for box with resizer when resized. > The resizer will be invlaidated in PaintInvalidationCapableScrollableArea > if needed. > CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2 > > Review-Url: https://codereview.chromium.org/2761863002 > Cr-Commit-Position: refs/heads/master@{#458313} > Committed: https://chromium.googlesource.com/chromium/src/+/4e0d0f9ddd68e59dccda0a8c9a096c1813a1bd44 TBR=pdr@chromium.org # Not skipping CQ checks because original CL landed more than 1 days ago. Review-Url: https://codereview.chromium.org/2772353002 Cr-Commit-Position: refs/heads/master@{#459827} [modify] https://crrev.com/bedba279ba5dc64257efc7bc8a13d697c049d3de/third_party/WebKit/LayoutTests/paint/invalidation/textarea-appearance-none-resize-handle-expected.txt [modify] https://crrev.com/bedba279ba5dc64257efc7bc8a13d697c049d3de/third_party/WebKit/Source/core/layout/LayoutObject.cpp [modify] https://crrev.com/bedba279ba5dc64257efc7bc8a13d697c049d3de/third_party/WebKit/Source/core/paint/BoxPaintInvalidator.cpp [modify] https://crrev.com/bedba279ba5dc64257efc7bc8a13d697c049d3de/third_party/WebKit/Source/core/style/ComputedStyle.cpp [modify] https://crrev.com/bedba279ba5dc64257efc7bc8a13d697c049d3de/third_party/WebKit/Source/core/style/ComputedStyle.h [modify] https://crrev.com/bedba279ba5dc64257efc7bc8a13d697c049d3de/third_party/WebKit/Source/core/style/StyleDifference.h
,
Mar 28 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983908399062352544
,
Mar 28 2017
Started bisect job https://chromeperf.appspot.com/buildbucket_job_status/8983908378446818384
,
Mar 28 2017
=== BISECT JOB RESULTS === NO Perf regression found Bisect Details Configuration: android_webview_nexus6_aosp_perf_bisect Benchmark : smoothness.sync_scroll.key_mobile_sites_smooth Metric : frame_times/http___mobile-news.sandbox.google.com_news_pt0 Revision Result N chromium@459831 16.8619 +- 0.384957 21 good chromium@459846 16.8509 +- 0.407055 21 bad To Run This Test src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...mobile.news.sandbox.google.com.news.pt0 smoothness.sync_scroll.key_mobile_sites_smooth Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8983908399062352544 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=5886588183117824 | 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 Speed>Bisection. Thank you!
,
Mar 28 2017
=== BISECT JOB RESULTS === Perf regression found but unable to continue Bisect was stopped because a commit couldn't be classified as either good or bad. Bisect Details Configuration: android_webview_nexus6_aosp_perf_bisect Benchmark : smoothness.sync_scroll.key_mobile_sites_smooth Metric : frame_times/http___mobile-news.sandbox.google.com_news_pt0 Revision Result N chromium@459831 16.8438 +- 0.27427 21 good chromium@459839 16.8369 +- 0.304222 21 unknown chromium@459846 16.8847 +- 0.249712 21 bad To Run This Test src/tools/perf/run_benchmark -v --browser=android-webview --output-format=chartjson --upload-results --pageset-repeat=1 --also-run-disabled-tests --story-filter=http...mobile.news.sandbox.google.com.news.pt0 smoothness.sync_scroll.key_mobile_sites_smooth Debug Info https://chromeperf.appspot.com/buildbucket_job_status/8983908378446818384 Is this bisect wrong? https://chromeperf.appspot.com/bad_bisect?try_job_id=5272704480968704 | 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 Speed>Bisection. Thank you!
,
Mar 29 2017
,
Mar 31 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/36f76920b08dcd81e97f74afd04474d88f0441e2 commit 36f76920b08dcd81e97f74afd04474d88f0441e2 Author: wangxianzhu <wangxianzhu@chromium.org> Date: Fri Mar 31 20:43:08 2017 Reland of Add StyleDifference::needsVisualRectUpdate (patchset #1 id:1 of https://codereview.chromium.org/2772353002/ ) Abandon the changes about StyleDifference:: m_scrollAnchorDisablingPropertyChanged in the original patch to avoid performance regressions. Original issue's description: > Revert of Add StyleDifference::needsVisualRectUpdate (patchset #5 id:80001 of https://codereview.chromium.org/2761863002/ ) > > Reason for revert: > BUG= 704728 > > Original issue's description: > > Add StyleDifference::needsVisualRectUpdate > > > > It's set when a style change will change visual rect (without changing > > layout, as visual rect update is implied by layout change). > > > > BTW optimized paint invalidation of box with resizer. We no longer > > force full paint invalidation for box with resizer when resized. > > The resizer will be invlaidated in PaintInvalidationCapableScrollableArea > > if needed. > > CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2 > > > > Review-Url: https://codereview.chromium.org/2761863002 > > Cr-Commit-Position: refs/heads/master@{#458313} > > Committed: https://chromium.googlesource.com/chromium/src/+/4e0d0f9ddd68e59dccda0a8c9a096c1813a1bd44 > > TBR=pdr@chromium.org > # Not skipping CQ checks because original CL landed more than 1 days ago. > > Review-Url: https://codereview.chromium.org/2772353002 > Cr-Commit-Position: refs/heads/master@{#459827} > Committed: https://chromium.googlesource.com/chromium/src/+/bedba279ba5dc64257efc7bc8a13d697c049d3de TBR=pdr@chromium.org BUG= 704728 CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2 Review-Url: https://codereview.chromium.org/2776413002 Cr-Commit-Position: refs/heads/master@{#461214} [modify] https://crrev.com/36f76920b08dcd81e97f74afd04474d88f0441e2/third_party/WebKit/LayoutTests/paint/invalidation/textarea-appearance-none-resize-handle-expected.txt [modify] https://crrev.com/36f76920b08dcd81e97f74afd04474d88f0441e2/third_party/WebKit/Source/core/layout/LayoutObject.cpp [modify] https://crrev.com/36f76920b08dcd81e97f74afd04474d88f0441e2/third_party/WebKit/Source/core/paint/BoxPaintInvalidator.cpp [modify] https://crrev.com/36f76920b08dcd81e97f74afd04474d88f0441e2/third_party/WebKit/Source/core/style/ComputedStyle.cpp [modify] https://crrev.com/36f76920b08dcd81e97f74afd04474d88f0441e2/third_party/WebKit/Source/core/style/ComputedStyle.h [modify] https://crrev.com/36f76920b08dcd81e97f74afd04474d88f0441e2/third_party/WebKit/Source/core/style/StyleDifference.cpp [modify] https://crrev.com/36f76920b08dcd81e97f74afd04474d88f0441e2/third_party/WebKit/Source/core/style/StyleDifference.h [modify] https://crrev.com/36f76920b08dcd81e97f74afd04474d88f0441e2/third_party/WebKit/Source/core/style/StyleDifferenceTest.cpp
,
Apr 2 2017
|
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by jasontiller@chromium.org
, Mar 23 2017