Bug 121054 - ASSERTION FAILED: mainAxisExtent - mainAxisBorderAndPaddingExtentForChild(child) >= 0
Summary: ASSERTION FAILED: mainAxisExtent - mainAxisBorderAndPaddingExtentForChild(chi...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: Layout and Rendering (show other bugs)
Version: 528+ (Nightly build)
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Nobody
URL:
Keywords: InRadar
Depends on: 153991
Blocks:
  Show dependency tree / graph
 
Reported: 2013-09-09 15:12 PDT by Vicki Pfau
Modified: 2016-07-07 15:14 PDT (History)
7 users (show)

See Also:

Attachments
Repro (175 bytes, text/html)
2013-09-09 15:12 PDT, Vicki Pfau 		no flags Details
Repro 2 (305 bytes, text/html)
2015-11-06 02:03 PST, Renata Hodovan 		no flags Details
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Vicki Pfau 2013-09-09 15:12:36 PDT 
Created attachment 211100 [details]
Repro

The assertion mainAxisExtent - mainAxisBorderAndPaddingExtentForChild(child) >= 0 in RenderFlexibleBox::preferredMainAxisContentExtentForChild(WebCore::RenderBox *, bool) can fail in certain cases. A simple repro case is attached.

<rdar://problem/14933505>
Comment 1 Tony Chang 2013-09-26 14:27:18 PDT 
I tried this in a debug build of blink, but blink doesn't hit this assert.  This is probably because blink doesn't support MathML.
Comment 2 Renata Hodovan 2014-03-03 02:20:02 PST 
(In reply to comment #1)
> I tried this in a debug build of blink, but blink doesn't hit this assert.  This is probably because blink doesn't support MathML.

I can reproduce this issue (or at least the same assertion fail) on the TotT Blink with a different test case. The issue with the test case is reported here: https://code.google.com/p/chromium/issues/detail?id=335121.
Comment 3 Renata Hodovan 2015-11-06 02:03:16 PST 
Created attachment 264921 [details]
Repro 2

An even shorter repro is:

<math><menclose>
Comment 4 Frédéric Wang (:fredw) 2016-03-14 03:54:45 PDT 
I didn't not try the testcase, but I'm making this depends on bug 153991 where MathML will have its own layout, independent from flexbox.
Comment 5 Frédéric Wang (:fredw) 2016-06-27 03:54:09 PDT 
I no longer have this assert after the MathML refactoring (probably fixed by bug 155019 and 153918). I'm still waiting bug 153991 before closing this bug so that we are sure that code will never be executed with MathML.
Comment 6 Frédéric Wang (:fredw) 2016-07-07 15:14:07 PDT 
Fixed by http://trac.webkit.org/changeset/202934, since MathML no longer execute RenderFlexibleBox code.