Issue metadata
Sign in to add a comment
|
2117.6%-2393.4% regression in blink_perf.shadow_dom at 562217:562285 |
||||||||||||||||||||||
Issue descriptionSee the link to graphs below.
,
May 30 2018
📍 Pinpoint job started. https://pinpoint-dot-chromeperf.appspot.com/job/14aa967a240000
,
May 31 2018
📍 Found significant differences after each of 7 commits. https://pinpoint-dot-chromeperf.appspot.com/job/14aa967a240000 Reland "Adding absl includes and defines to rtc_* templates." by mbonadei@webrtc.org https://webrtc.googlesource.com/src/+/fae51e4c95780f59f4bedf6e01403ff86152e168 1847 → 1858 (+10.87) Roll src/third_party/webrtc/ b3085b288..dacec71b1 (5 commits) by webrtc-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com https://chromium.googlesource.com/chromium/src/+/0257046a7d6014205c898f17079f091eccae8f27 1858 → No values Remove important sites on Desktop by dullweber@chromium.org https://chromium.googlesource.com/chromium/src/+/eff032d48fd7361504192c53d7a88ed88f11de12 No values → 1848 [TaskScheduler] Slower latency heartbeat by gab@chromium.org https://chromium.googlesource.com/chromium/src/+/7e7dd55cce56f643d5945f5a2dbd42d475c73e1b 1848 → 1845 (-2.618) Move NavigationRequest creation to NavigationController: 2/3 by clamy@chromium.org https://chromium.googlesource.com/chromium/src/+/ea99ea1dc01c5bbf058c18abe50008147e5b7ae9 1844 → 1837 (-7.418) Add WebUSB settings deletion by dullweber@chromium.org https://chromium.googlesource.com/chromium/src/+/9ccc4fae6f48397e737e3075224688adf133746b 1840 → 1825 (-15.07) [Squad] Recalc style for re-attach for assigned nodes. by futhark@chromium.org https://chromium.googlesource.com/chromium/src/+/329876ebb0709a460e96f540721ec8f8cf68845b 1824 → 4.48e+04 (+4.297e+04) Understanding performance regressions: http://g.co/ChromePerformanceRegressions
,
May 31 2018
,
May 31 2018
,
May 31 2018
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/4ae86dbc5688caa90fda5e56c0437e8573bea32b commit 4ae86dbc5688caa90fda5e56c0437e8573bea32b Author: Rune Lillesveen <futhark@chromium.org> Date: Thu May 31 10:51:27 2018 [Squad] Fix perf regression in reattach recalc. Make sure we don't recalc style for reattach more than once per element. In [1], we introduced recalculating style for reattach of assigned nodes from the slots, but we missed a check to not traverse into the same nodes when traversing light tree children. Added a DCHECK on GetNonAttachedStyle as a sanity check. [1] http://crrev.com/329876ebb0709a460e96f540721ec8f8cf68845b TEST=third_party/blink/perf_tests/shadow_dom/v1-small-deep-layout.html Bug: 847990 Change-Id: I0596d92089b054dafdc3e25b21b6180ca71ec812 Reviewed-on: https://chromium-review.googlesource.com/1079549 Reviewed-by: Morten Stenshorne <mstensho@chromium.org> Commit-Queue: Rune Lillesveen <futhark@chromium.org> Cr-Commit-Position: refs/heads/master@{#563188} [modify] https://crrev.com/4ae86dbc5688caa90fda5e56c0437e8573bea32b/third_party/blink/renderer/core/dom/container_node.cc [modify] https://crrev.com/4ae86dbc5688caa90fda5e56c0437e8573bea32b/third_party/blink/renderer/core/dom/element.cc
,
May 31 2018
|
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by 42576172...@developer.gserviceaccount.com
, May 30 2018