New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 797456 link

Starred by 4 users

Issue metadata

Status: WontFix
Owner:
Not on Chrome anymore
Closed: Jan 2018
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux , Windows , Mac
Pri: 1
Type: Bug-Regression



Sign in to add a comment

Display flex in combination with an input field seems to bug the background color of a page

Reported by carnie....@gmail.com, Dec 22 2017

Issue description

UserAgent: Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.108 Safari/537.36

Example URL:
http://www.mitchellvdhut.com/bgbug/index.html

Steps to reproduce the problem:
1. i put a display: flex; on the body element
2. made a form with an input field in it
3. doesn't seem to happen when the css is in the html file

What is the expected behavior?
background should always be the specified color

What went wrong?
It seems that using display: flex; on the body, and having the input field there causes the background-color to not work propperly.

Does it occur on multiple sites: N/A

Is it a problem with a plugin? No 

Did this work before? Yes i do not know

Does this work in other browsers? Yes

Chrome version: 63.0.3239.108  Channel: stable
OS Version: 10.0
Flash Version: 

Sometimes it does work on the first load, but it shows the bug like 90% of the time when you refresh.
Tested this both on my server and locally. Having the css inside the HTML file seems to work 100% of the time.
 
bgbug.PNG
46.2 KB View Download

Comment 1 by woxxom@gmail.com, Dec 23 2017

Bisect info: 493347 (good) - 493359 (bad)
https://chromium.googlesource.com/chromium/src/+log/6f8907e8..161200eb?pretty=fuller
Suspecting r493348 = 9f25abddd6a5e0d25d5082ec0f29727a984f9e03 = https://crrev.com/c/608028 by rune@opera.com
"Propagate style to viewport after style recalc."
Landed in 62.0.3182.0
Labels: Needs-Triage-M63 Needs-Bisect
Able to reproduce the issue on reported version 63.0.3239.108 and the same is not seen on latest canary 64.0.3304.0 using Windows 10, Ubuntu 14.04, Mac 10.12.6 hence providing reverse bisect info

Bisect Info:
================
Last Bad build: 65.0.3284.0
First Good build: 65.0.3285.0

You are probably looking for a change made after 521239 (known good), but no later than 521240 (first known bad).

https://chromium.googlesource.com/chromium/src/+log/3a13f63d2ad6f26c9aefdd2d2a7f8fa34571f23e..b71f5b7df345f1a8603b7cf7b59edf48288e7d96

Reviewed on: https://chromium-review.googlesource.com/765097
Suspecting same from changelog

@Naina Raisinghani: Please confirm the issue and help in re-assigning if it is not related to your change, help in mergign this issue to M-64 if applicable

Thanks!
Cc: viswatej...@techmahindra.com ranjitkan@chromium.org pbomm...@chromium.org sc00335...@techmahindra.com
Components: Blink>CSS Blink>Layout
Labels: -Pri-2 -Type-Compat -Needs-Bisect hasbisect-per-revision ReleaseBlock-Stable Triaged-ET M-64 OS-Linux OS-Mac Pri-1 Type-Bug-Regression
Owner: nainar@chromium.org
Status: Assigned (was: Unconfirmed)
I'd say that both merging Naina's change and writing an alternative patch for M-64 on their own are too risky for gain vs risk for M-64.

The bug happens when background and display property of body happens simultaneously after the layout tree has been attached once, which is probably not very common.

We discussed the effect of Naina's change related to this bug in https://chromium-review.googlesource.com/c/chromium/src/+/765097#message-58cafa6edca614a98aab6ca4bc25901714d246de

I've added tests for background propagation, the 003 test triggering this issue:

https://chromium-review.googlesource.com/c/chromium/src/+/844597

futhark@, 

Thanks for adding the tests!  

To confirm I am understanding the issue correctly. 

The testcase http://www.mitchellvdhut.com/bgbug/index.html doesn't work in M63 (with the external stylesheet), but does in M65 (with my change). I agree merging this change into 64 is too risky. Specially since M65 branches on Jan 18. 

Given that the tests have been added should I mark this issue as WontFix?

Status: WontFix (was: Assigned)
nainar@ That's my conclusion. I just wanted you to get the chance to chime in first.
Cc: vamshi.k...@techmahindra.com brajkumar@chromium.org nainar@chromium.org xiaoche...@chromium.org
 Issue 807914  has been merged into this issue.

Sign in to add a comment