Chrome 58 critical rendering issue
Reported by
ourai...@gmail.com,
May 15 2017
|
|||||||||
Issue descriptionUserAgent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36 Example URL: https://jsfiddle.net/fx4srqy1/4/ Steps to reproduce the problem: 1. Go to the URL 2. Click on "Suivant >" button in the result page of the JSFiddle then click "< Précédent" button, the page is completely broken. What is the expected behavior? A proper rendering. What went wrong? The rendering is completely broken. It seems like it has something to do with the height computing from tag before the vue stepper component, if you add more "<br>" at the start of the html, the page will be even more messed up. I tested the same code on windows and macOS and there is the same issue in Chrome 58 on both OS too. However i tested the same code with chrome 56, chrome 57 and chrome dev 60, and other browser and it's working fine. I think this is an issue occuring only in chrome 58 (maybe 59 too ? havnt tested). Does it occur on multiple sites: N/A Is it a problem with a plugin? N/A Did this work before? Yes Chrome 57 and Chrome dev 60 too Does this work in other browsers? Yes Chrome version: 58.0.3029.110 Channel: stable OS Version: Flash Version: Shockwave Flash 25.0 r0
,
May 16 2017
Tested the issue on windows 7 , Mac OS 10.12.3 and ubuntu 14.04 using chrome M58 #58.0.3029.110 and issue is reproduced. Issue got fixed in latest beta , dev and canary channels and is a regression issue broken in M59. Providing the reverse bisect using per-revision bisect script, Good build:59.0.3050.0 (Revision:459323). Bad build:59.0.3048.0 (Revision:458590). You are probably looking for a change made after 458906 (known good), but no later than 458907 (first known bad). CHANGELOG URL: The script might not always return single CL as suspectas some perf builds might get missing due to failure. https://chromium.googlesource.com/chromium/src/+log/a0ca178033ce1a45919921fa67a050b808d706de..463b2dfe64df899fda4359f2d360ee1cba46e26b From the CL above, assigning the issue to the concern owner @wkorman- Could you please merge the fix into M58 . Review-Url: https://codereview.chromium.org/2767783003 Thanks!
,
May 16 2017
,
May 16 2017
Working on this now.
,
May 16 2017
I think c#2 is a typo and the intent is to merge this to M59 per label. Requesting merge so that I can figure out the branch number which is not apparent anywhere I can find.
,
May 16 2017
Ah, no, it is for M58? It is already in M59 per: Commit 463b2dfe... initially landed in 59.0.3049.0
,
May 16 2017
FTR the change https://codereview.chromium.org/2767783003/ for http://crbug.com/695125 was for a regression seen since M54 whereas this bug states the issue was first seen in M58.
,
May 16 2017
This was merged as https://chromium.googlesource.com/chromium/src.git/+/2c693969ab57ca41b67582265c08c1094a71a0bc but was not tagged with this bug.
,
May 16 2017
wkorman@, change listed at #8 https://chromium.googlesource.com/chromium/src.git/+/2c693969ab57ca41b67582265c08c1094a71a0bc (58.0.3029.118) got merged to M58 (branch 3029) without approval. Please note we are not planning any further M58 stable releases for Desktop.
,
May 16 2017
Apologies, somehow I inferred approval from above (incorrectly). Do I need to revert or is it ok to just leave as is?
,
May 16 2017
No worries, please revert it and leave the "Merge-Request-58" label as it is. Thank you.
,
May 16 2017
So the fixed wont be deploy in the 58 release ?
,
May 16 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/759d0c5a0e1be18687a426932685302a80256b07 commit 759d0c5a0e1be18687a426932685302a80256b07 Author: Walter Korman <wkorman@chromium.org> Date: Tue May 16 19:44:31 2017 Revert "Incorporate ComputedStyle::hasTransform when diffing transform styles." This reverts commit 2c693969ab57ca41b67582265c08c1094a71a0bc. Change was not approved for merge. Further M58 releases not planned. BUG= 722368 CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2 Review-Url: https://codereview.chromium.org/2888703002 . Cr-Commit-Position: refs/branch-heads/3029@{#849} Cr-Branched-From: 939b32ee5ba05c396eef3fd992822fcca9a2e262-refs/heads/master@{#454471} [modify] https://crrev.com/759d0c5a0e1be18687a426932685302a80256b07/third_party/WebKit/Source/core/paint/PaintLayer.cpp [modify] https://crrev.com/759d0c5a0e1be18687a426932685302a80256b07/third_party/WebKit/Source/core/paint/PaintLayer.h [modify] https://crrev.com/759d0c5a0e1be18687a426932685302a80256b07/third_party/WebKit/Source/core/paint/PaintLayerTest.cpp [modify] https://crrev.com/759d0c5a0e1be18687a426932685302a80256b07/third_party/WebKit/Source/core/style/ComputedStyle.cpp [modify] https://crrev.com/759d0c5a0e1be18687a426932685302a80256b07/third_party/WebKit/Source/core/style/ComputedStyle.h [modify] https://crrev.com/759d0c5a0e1be18687a426932685302a80256b07/third_party/WebKit/Source/core/style/ComputedStyleTest.cpp
,
May 16 2017
Re: fix deploy -- correct, the fix is currently planned to ship in M59. You can verify with Beta, Dev or Canary builds. Apologies for the regression and delay.
,
May 18 2017
We're are not planning any further M58 stable releases. Rejecting merge to M58.
,
May 18 2017
,
May 18 2017
Issue 721097 has been merged into this issue. |
|||||||||
►
Sign in to add a comment |
|||||||||
Comment 1 by ligim...@chromium.org
, May 15 2017Components: Blink
Labels: Needs-Triage-M58 Needs-Bisect