371566 - "ASSERTION: used padding property missing (out of memory?)" with -moz-inline-stack, float, position: absolute

Status: RESOLVED WORKSFORME

(Core :: Layout, defect)

Description

[Jesse Ruderman]

Attachment: testcase

###!!! ASSERTION: used padding property missing (out of memory?): 'p', file /Users/jruderman/trunk/mozilla/layout/generic/nsFrame.cpp, line 764

Comment 1

[Boris Zbarsky [:bzbarsky]]

I think the real issue is:

###!!! ASSERTION: cannot call on a dirty frame not currently being reflowed: '!(GetStateBits() & (NS_FRAME_IS_DIRTY | NS_FRAME_HAS_DIRTY_CHILDREN)) || (GetStateBits() & NS_FRAME_IN_REFLOW)', file ../../../mozilla/layout/generic/nsFrame.cpp, line 740

0  0x056c1774 in nsIFrame::GetUsedPadding (this=0x959ab18)
    at ../../../mozilla/layout/generic/nsFrame.cpp:737
#1  0x05674a80 in nsIFrame::GetUsedBorderAndPadding (this=0x959ab18)
    at ../../../../../mozilla/layout/generic/nsIFrame.h:698
#2  0x056f1807 in nsHTMLReflowState::GetNearestContainingBlock (this=0xbffe56b0, 
    aFrame=0x959ab18, aCBLeftEdge=@0xbffe5328, aCBWidth=@0xbffe5324)
    at ../../../mozilla/layout/generic/nsHTMLReflowState.cpp:593
#3  0x056f21ab in nsHTMLReflowState::InitAbsoluteConstraints (this=0xbffe56b0, 
    aPresContext=0x955cf28, cbrs=0xbffe5d20, containingBlockWidth=60420, 
    containingBlockHeight=36600)
    at ../../../mozilla/layout/generic/nsHTMLReflowState.cpp:940

(gdb) p this
$1 = (nsAreaFrame *) 0x959ab18
(gdb) p this->mContent
$2 = (nsHTMLDivElement *) 0x959a328
(gdb) p this->mContent->GetParent()
[Thread -1243767888 (LWP 1388) exited]
$3 = (nsHTMLDivElement *) 0x959a1b8
(gdb) p this->mContent->GetParent()->GetParent()
$4 = (nsHTMLBodyElement *) 0x959a088

So this is the left-floating div.  Now why it never got reflown, I'm not sure...

Comment 2

[Boris Zbarsky [:bzbarsky]]

So we never reflow the float because it's on the float-list of the body, while the placeholder is inside the stack, so the body never really finds the float...

I know we have bugs on that (and this should depend on them), but I can't find them right now...

Comment 3

[Jesse Ruderman]

WFM with the testcase in comment 0.

bz, should anything from your analysis (comment 1 and comment 2) become a separate bug report?

Comment 4

[Boris Zbarsky [:bzbarsky]]

I think roc's changing to disable floating in XUL probably fixed this.

Comment 5

[Jesse Ruderman]

Crashtest checked in.