Issue metadata
Sign in to add a comment
|
Hangouts chats in Gmail not redrawing correctly in latest Canary |
||||||||||||||||||||||
Issue descriptionVersion: 52.0.2722.0 (Official Build) canary (64-bit) OS: Mac OS X 10.11.4 What steps will reproduce the problem? (1) Open Gmail (2) Look at the Hangouts chat area in the lower left corner of the Gmail window (3) Make sure the "Recent conversations" button is selected (3) Mouse in and out of the area What is the expected output? Expect to see the Hangouts window draw and redraw correctly, with popups as needed when hovering over your contacts. What do you see instead? The recent conversations tab was blank for me, except for the most recent conversation the first couple of times I used it today. At this point all the conversations are visible, but mousing in or out of the Hangouts area causes the entire area to flash to white before redrawing. The previous Canary was crashing in blink::FramePainter pretty often -- see crash/26c5c86c00000000 which happened while viewing http://www.theverge.com/2016/4/28/11526150/crytek-the-climb-vr-oculus-rift-review -- so I suspect recent changes in that area are responsible.
,
May 2 2016
Raising this to P1 because I've noticed other rendering bugs that seem related and could be affecting real-world web sites. This needs to be triaged. Another example: the manager chain div on teams/ renders as solid white unless you hover over it. If no scrolling is possible, then it's solid white all the time.
,
May 2 2016
CC'ing a few others who work on paint and invalidation who might be able to help triage. Please take this bug if you can help. Thanks.
,
May 2 2016
I can repro on Mac / Retina.
,
May 4 2016
Verified the issue on Latest Chrome Canary# 52.0.2724.0 on Retina Macbook (Mac OS X 10.11.4 and is able to reproduce the issue. This is a regression issue in M52, below are the details: Chrome Good Build -- 52.0.2718.0. Chrome Bad Build -- 52.0.2719.0. You are probably looking for a change made after 390193 (known good), but no later than 390210 (first known bad). CHANGELOG URL: https://chromium.googlesource.com/chromium/src/+log/b19a768741e01a7d8be807cbdbcee0c164b1d2aa..ed1eee6bf1253dcce9eb0d8a0ef0bee86c70f3ac Suspected Commit# 861d774435b04448856e99462d4f4e3d734f72f4 Suspected Review URL# https://codereview.chromium.org/1913843004 @bokan -- Could you please look into the issue, pardon me if it has nothing to do with your changes and if possible please assign it to the concern Dev. Also adding Devs in Cc who might be related. Thank You.
,
May 4 2016
I can't repro this on today's Canary. I tried on a lo-DPI monitor but bumped the settings up to hi-DPI which should faithfully emulate hi-DPI mode. That or it's been fixed since (or I'm doing the repro wrong). kbr@/chrishtr@, can you still repro in today's Canary? My CL is very unlikely to cause a change like this and is behind a runtime flag. There's a skia roll in the range but it looks fairly benign. If I had to guess, vollick@'s CL is at least touching related code.
,
May 4 2016
Yes, the repaint issues are still present in 52.0.2723.0 (Official Build) canary (64-bit) on a Retina MBP with NVIDIA GPU. Hovering over the hangout chat area and labels area in Gmail cause various repaint issues, such as the first few labels disappearing, and the contacts in hangouts chat flashing to white when the mouse enters or leaves that area. See attached screenshot of the former problem.
,
May 4 2016
,
May 5 2016
,
May 5 2016
I'm still seeing this in today's Canary, this bug is pretty bad, it makes Gmail chat unusable for me.
,
May 5 2016
Tried but didn't reproduce. It often crashes instead. Debugging with a local build.
,
May 5 2016
An Autofill bug was fixed yesterday which was causing a lot of crashes in Gmail. I'm seeing other rendering breakage of real-world apps, like https://www.figma.com/ . I don't know whether it's directly related, but this app renders correctly in Stable but not in Canary -- the top part of their app's drawing area doesn't redraw. If anyone would like to access that app for debugging please let me know.
,
May 5 2016
I can repro easily on Mac Retina / 52.0.2725.0.
,
May 5 2016
,
May 5 2016
Reproduced with ToT and aba9829 (the revision after bokan@ and vollick@'s changes, but before the Skia roll). Trying the nearby revisions.
,
May 5 2016
Bad: 94d5fad Use correct clip container when computing acl for scroll children by vollick · 8 days ago Good: 865f2de bluetooth: Move Web Bluetooth from OriginTrialEnabled to RuntimeEnabled. by scheib · 8 days ago
,
May 5 2016
Revert in the CQ:
,
May 5 2016
,
May 5 2016
Auto revert failed to apply. Ian will do it manually now.
,
May 5 2016
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/efadb1362a06b05bc877a13f50010ddf3ce50ccd commit efadb1362a06b05bc877a13f50010ddf3ce50ccd Author: vollick <vollick@chromium.org> Date: Thu May 05 21:40:02 2016 Revert of Use correct clip container when computing acl for scroll children Reason for revert: Caused issue 608410 Original issue's description: > Use correct clip container when computing acl for scroll children > > Kudos to erick@ for the reduced test case. > > BUG= 603997 > > Committed: https://crrev.com/94d5fad34f0351e61fd658ca4ed93648c415620d > Cr-Commit-Position: refs/heads/master@{#390199} TBR=wangxianzhu@chromium.org BUG= 603997 , 608410 Review-Url: https://codereview.chromium.org/1951853006 Cr-Commit-Position: refs/heads/master@{#391913} [modify] https://crrev.com/efadb1362a06b05bc877a13f50010ddf3ce50ccd/third_party/WebKit/LayoutTests/TestExpectations [modify] https://crrev.com/efadb1362a06b05bc877a13f50010ddf3ce50ccd/third_party/WebKit/Source/core/layout/compositing/CompositedLayerMapping.cpp [modify] https://crrev.com/efadb1362a06b05bc877a13f50010ddf3ce50ccd/third_party/WebKit/Source/core/layout/compositing/CompositedLayerMappingTest.cpp
,
May 5 2016
,
May 6 2016
|
|||||||||||||||||||||||
►
Sign in to add a comment |
|||||||||||||||||||||||
Comment 1 by kbr@chromium.org
, May 2 2016Status: Assigned (was: Unconfirmed)