Chrome 59 does not render elements position fixed
Reported by
jheren...@mrmilu.com,
Jun 21 2017
|
|||||||
Issue descriptionUserAgent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_5) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/59.0.3071.86 Safari/537.36 Example URL: https://www.pepecar.com/en-GB/buscador/results/?country_code=ES&city=Madrid&pickup_location=530&pickup_date=2017-10-27&pickup_time=10%3A00&return_location=530&return_date=2017-10-29&return_time=10%3A00&mode=cars¤cy_code=EUR&return_country_code=ES&drivers_age=30 Steps to reproduce the problem: 1. Go to the URL given 2. Press the BOOK button in any car 3. Scroll down Here is a video where you can see the problem http://take.ms/5Cerm6 What is the expected behavior? What went wrong? Chrome is working wrong rendering fixed positioned elements: http://take.ms/5Cerm6 as you can see. Does it occur on multiple sites: Yes Is it a problem with a plugin? No Did this work before? N/A Does this work in other browsers? Yes Chrome version: 59.0.3071.86 Channel: n/a OS Version: OS X 10.12.5 Flash Version: This does not happen in Chrome 58 and not happening either in any other browser (I've checked via BrowserStack) I've found that this behavior is happening too in other websites: https://www.diseloporsevillanas.com/es-es/custom http://take.ms/YjTJT
,
Jun 22 2017
There does seem to be something weird going on here. The same page looks fine in Firefox. Over to layout team - a bisect in the meantime would be helpful.
,
Jun 23 2017
Tested the issue on Windows 7 , Mac OS 10.12.5 and ubuntu 14.04 using chrome M59 #59.0.3071.109 and issue is reproduced . But issue got fixed in latest beta , dev and canary channels. Hence providing the reverse bisect results using old script, Good build: 60.0.3098.0(Revision: 471507). Bad build:60.0.3096.0 (Revision: 470759). You are probably looking for a change made after 471157 (known good), but no later than 471170 (first known bad). CHANGELOG URL: https://chromium.googlesource.com/chromium/src/+log/ed59882df38c70c6801b67478454fa00f432d62d..4772a01d5433374cd8dd6f4a9fedaff7eafbd5a7 Unable to find the actual suspect from the above CL's , could someone help us in assiging it to the concerned owner. Thanks!
,
Jun 26 2017
The layout tree looks correct, this is likely a paint invalidation issue.
,
Jun 26 2017
My guess is that this was fixed by https://codereview.chromium.org/2869813002. Checking that now.
,
Jun 26 2017
,
Jun 26 2017
Confirmed that https://codereview.chromium.org/2869813002 fixed it. Great work that we fixed this even before a bug was filed. :) AFAICS it has been a bug for a while now. Yi, even though it's fixed could you please add a layout test with a position:fixed element, to improve test coverage?
,
Jun 27 2017
The following revision refers to this bug: https://chromium.googlesource.com/chromium/src.git/+/0ecc4446c722663734e7a6a4fee324175de0d104 commit 0ecc4446c722663734e7a6a4fee324175de0d104 Author: Yi Gu <yigu@chromium.org> Date: Tue Jun 27 17:16:19 2017 Add layout test for position:fixed under compositing and scroll. Patch http://crrev.com/2869813002 fixed the bug that absolute positioned element may not be painted correctly. The patch also applies to fix positioned element. Adding a layout test to improve the test coverage. Bug: 735547 Change-Id: I864375aa50dd12ebcc2254fa7181bfb8e87026ce Reviewed-on: https://chromium-review.googlesource.com/550015 Reviewed-by: Chris Harrelson <chrishtr@chromium.org> Commit-Queue: Yi Gu <yigu@chromium.org> Cr-Commit-Position: refs/heads/master@{#482658} [add] https://crrev.com/0ecc4446c722663734e7a6a4fee324175de0d104/third_party/WebKit/LayoutTests/compositing/overflow/fixed-element-in-isolated-composited-ancestor-expected.html [add] https://crrev.com/0ecc4446c722663734e7a6a4fee324175de0d104/third_party/WebKit/LayoutTests/compositing/overflow/fixed-element-in-isolated-composited-ancestor.html
,
Jun 27 2017
|
|||||||
►
Sign in to add a comment |
|||||||
Comment 1 by nyerramilli@chromium.org
, Jun 22 2017