4336 - MLK: 5632 bytes leaked- www.visa.com - [top100]leaking StyleContextImpl

Status: VERIFIED WORKSFORME

(Core :: DOM: Core & HTML, defect, P3)

Description

[Bruce Mitchener]

This leak happens when visiting the Visa website.  Build from March 27, 1999.

MLK: 5632 bytes leaked in 11 blocks
  * This memory was allocated from:
        malloc         [rtlib.o]
        __bUiLtIn_nEw  [libgcc.a]
        __builtin_new  [rtlib.o]
        StyleContextImpl::operator  new(unsigned int) [nsStyleContext.cpp:839]

NS_NewStyleContext(nsIStyleContext**,nsIStyleContext*,nsIAtom*,nsISupportsArray*
,nsIPresContext*) [nsStyleContext.cpp:1514]

StyleSetImpl::GetContext(nsIPresContext*,nsIStyleContext*,nsIAtom*,nsISupportsAr
ray*,int,int&) [nsStyleSet.cpp:453]

StyleSetImpl::ResolveStyleFor(nsIPresContext*,nsIContent*,nsIStyleContext*,int)
[nsStyleSet.cpp:546]

nsPresContext::ResolveStyleContextFor(nsIContent*,nsIStyleContext*,int,nsIStyleC
ontext**) [nsPresContext.cpp:384]

nsCSSFrameConstructor::ResolveStyleContext(nsIPresContext*,nsIFrame*,nsIContent*
,nsIAtom*,nsIStyleContext**) [nsCSSFrameConstructor.cpp:2838]

nsCSSFrameConstructor::ConstructFrame(nsIPresContext*,nsIContent*,nsIFrame*,nsAb
soluteItems&,nsFrameItems&,nsAbsoluteItems&,nsAbsoluteItems&)
[nsCSSFrameConstructor.cpp:2861]

nsCSSFrameConstructor::ContentInserted(nsIPresContext*,nsIContent*,nsIContent*,i
nt) [nsCSSFrameConstructor.cpp:3418]

nsCSSFrameConstructor::RecreateFramesForContent(nsIPresContext*,nsIContent*)
[nsCSSFrameConstructor.cpp:4571]

nsCSSFrameConstructor::AttributeChanged(nsIPresContext*,nsIContent*,nsIAtom*,int
) [nsCSSFrameConstructor.cpp:3953]
        StyleSetImpl::AttributeChanged(nsIPresContext*,nsIContent*,nsIAtom*,int)
[nsStyleSet.cpp:793]
        PresShell::AttributeChanged(nsIDocument*,nsIContent*,nsIAtom*,int)
[nsPresShell.cpp:1626]
        nsDocument::AttributeChanged(nsIContent*,nsIAtom*,int)
[nsDocument.cpp:1356]
        nsGenericHTMLElement::SetHTMLAttribute(nsIAtom*,const nsHTMLValue&,int)
[nsGenericHTMLElement.cpp:665]
        nsHTMLOptionElement::SetIndex(int) [nsHTMLOptionElement.cpp:219]
        nsSelectControlFrame::Reset() [nsSelectControlFrame.cpp:709]
        nsSelectControlFrame::PostCreateWidget(nsIPresContext*,int&,int&)
[nsSelectControlFrame.cpp:594]
        nsFormControlFrame::Reflow(nsIPresContext&,nsHTMLReflowMetrics&,const
nsHTMLReflowState&,unsigned int&) [nsFormControlFrame.cpp:311]
        nsLineLayout::ReflowFrame(nsIFrame*,nsIFrame**,unsigned int&)
[nsLineLayout.cpp:841]

nsBlockFrame::ReflowInlineFrame(nsBlockReflowState&,nsLineBox*,nsIFrame*,unsigne
d char*) [nsBlockFrame.cpp:2728]
        nsBlockFrame::ReflowInlineFrames(nsBlockReflowState&,nsLineBox*,int*)
[nsBlockFrame.cpp:2610]
        nsBlockFrame::ReflowLine(nsBlockReflowState&,nsLineBox*,int*)
[nsBlockFrame.cpp:1717]
        nsBlockFrame::ReflowDirtyLines(nsBlockReflowState&)
[nsBlockFrame.cpp:1522]
        nsBlockFrame::Reflow(nsIPresContext&,nsHTMLReflowMetrics&,const
nsHTMLReflowState&,unsigned int&) [nsBlockFrame.cpp:892]
        nsAreaFrame::Reflow(nsIPresContext&,nsHTMLReflowMetrics&,const
nsHTMLReflowState&,unsigned int&) [nsAreaFrame.cpp:509]

nsContainerFrame::ReflowChild(nsIFrame*,nsIPresContext&,nsHTMLReflowMetrics&,con
st nsHTMLReflowState&,unsigned int&) [nsContainerFrame.cpp:388]
        nsTableCellFrame::Reflow(nsIPresContext&,nsHTMLReflowMetrics&,const
nsHTMLReflowState&,unsigned int&) [nsTableCellFrame.cpp:501]
  * Block of 512 bytes (11 times); last block at 0x817270

Comment 1

[Bruce Mitchener]

Changing URL to sample HTML that minimally demonstrates the leak and to form
handling code since it seems to only happen so far in display of select menus
for forms.

Comment 2

[Eric Pollmann]

I'm arbitrarily declaring M8 my UMR/MLK milestone.  :)

Comment 3

[Eric Pollmann]

I still haven't got Purify working on Solaris, and won't receive Purify for NT
for a few weeks.  Marking these M10

Comment 4

[Eric Pollmann]

After careful consideration, I've decided that I probably won't get this bug in
for M12.  Currently I have nearly 50 bugs scheduled for M13, so there is a
possibility that this bug may need to be moved out farther still.

Comment 5

[gerardok]

QA Contact update.

Comment 6

[Eric Pollmann]

Triaged to M14

Comment 7

[leger]

adding top100 keyword.

Comment 8

[Eric Pollmann]

Moving off to M16 - please speak up of you need this for M14, thanks!

Comment 9

[Eric Pollmann]

Rescheduling (*sigh*)  Some of these are from M4.  I wonder if they are all
still valid?

Comment 10

[Eric Pollmann]

We are no longer leaking StyleContextImpl's.  In fact, I just did a walk of 
about half of the top100 (Chofmann's browser buster test).  We leaked 5 
webshells and just over 1Mb of other stuff, but no StyleContextImpl's!

Comment 11

[Vladimir Ermakov]

Updating QA contact.

Comment 12

[Vladimir Ermakov]

Verifying worksforme