1342871 - stylo: Assertion failure: IsFrameListSorted<IsLessThanOrEqual>(aFrameList) (After we sort a frame list, it should be in sorted order...), at nsIFrame.h:4061

Status: RESOLVED WORKSFORME

(Core :: Layout, defect)

Description

[Carsten Book [:Tomcat]]

Attachment: bughunter stack

Assertion failure: IsFrameListSorted<IsLessThanOrEqual>(aFrameList) (After we sort a frame list, it should be in sorted order...), at /home/worker/workspace/build/src/layout/generic/nsIFrame.h:4061

found by bughunter on topsite tests and reproduced with latest stylo debug build from tinderbox on fedora

Steps to reproduce:
-> Load http://www.eloqua.com
---> Assertion failure

Comment 1

[Bobby Holley (:bholley)]

bz might have some thoughts on this one, given that he's looking at stylo's interactions with the frame tree.

Comment 2

[Boris Zbarsky [:bzbarsky]]

I can't reproduce this particular assertion failure with a stylo build from rev a08ec245fa24 (on Mac).

I go get this, though:

  thread '<unnamed>' panicked at 'Caller should apply sibling hints', servo/components/style/data.rs:235

I guess I'll try updating to m-c tip and reproducing.  Might be useful to know what changeset was tested...

Comment 3

[Bobby Holley (:bholley)]

(In reply to Boris Zbarsky [:bz] (still a bit busy) from comment #2)
> I can't reproduce this particular assertion failure with a stylo build from
> rev a08ec245fa24 (on Mac).
> 
> I go get this, though:
> 
>   thread '<unnamed>' panicked at 'Caller should apply sibling hints',
> servo/components/style/data.rs:235

That's bug 1340334, which I haven't had time to look at yet. That might suggest that there's enough broken here that it's not worth spending too much time on.

Comment 4

[Boris Zbarsky [:bzbarsky]]

Yeah, on tip I get the same panic about sibling hints, after some:

###!!! ASSERTION: Disagreement about whether it's a block or not: 'fromLine->IsBlock() == fromLine->mFirstChild->IsBlockOutside()', file /Users/bzbarsky/mozilla/stylo/mozilla/layout/generic/nsBlockFrame.cpp, line 2877

Comment 5

[Emilio Cobos Álvarez [:emilio]]

I cannot reproduce the panic nor the crash with my patch in bug 1340334. I do hit the disagreement about whether it's a block assertions.

I don't think this is fixed by that bug (this is more likely to be fixed by bug 1322317 or similar), but I think we can call this RESOLVED WORKSFORME with bug 1340334 fixed.