65162 - Crash in nsPluginInstanceOwner::GetMode

Status: VERIFIED FIXED

(Core Graveyard :: Plug-ins, defect)

Description

[Vikas Bhargava]

I get the following stack trace when going from a page that has a plugin to 
another page:

nsPluginInstanceOwner::GetMode(nsPluginInstanceOwner * const 0x04ecc9f0, 
nsPluginMode * 0x0000000a) line 1620 + 3 bytes
nsPluginStreamListenerPeer::OnStartRequest(nsPluginStreamListenerPeer * const 
0x04ecc380, nsIChannel * 0x04ecc160, nsISupports * 0x00000000) line 1148
nsHTTPFinalListener::OnStartRequest(nsHTTPFinalListener * const 0x04ecdf20, 
nsIChannel * 0x04ecc160, nsISupports * 0x00000000) line 1131
InterceptStreamListener::OnStartRequest(InterceptStreamListener * const 
0x040af990, nsIChannel * 0x04ecc160, nsISupports * 0x00000000) line 1191
nsHTTPServerListener::FinishedResponseHeaders() line 1056 + 48 bytes
nsHTTPServerListener::OnDataAvailable(nsHTTPServerListener * const 0x040b3780, 
nsIChannel * 0x04ecd8b4, nsISupports * 0x04ecc160, nsIInputStream * 0x040b2220, 
unsigned int 0x00000000, unsigned int 0x000011e7) line 427 + 8 bytes
nsOnDataAvailableEvent::HandleEvent(nsOnDataAvailableEvent * const 0x040a90a0) 
line 407 + 47 bytes
nsStreamListenerEvent::HandlePLEvent(PLEvent * 0x040a90b0) line 92 + 12 bytes
PL_HandleEvent(PLEvent * 0x040a90b0) line 576 + 10 bytes
PL_ProcessPendingEvents(PLEventQueue * 0x00b9bd80) line 509 + 9 bytes
_md_EventReceiverProc(HWND__ * 0x00000af4, unsigned int 0x0000d59a, unsigned int 
0x00000000, long 0x00b9bd80) line 1054 + 9 bytes
KERNEL32! bff7363b()
KERNEL32! bff942e7()
007b8a2e()

Comment 1

[Vikas Bhargava]

Attachment: Proposed fix for crash (diff -wu)

Comment 2

[Peter Lubczynski]

I'm stealing this one....

Comment 3

[Peter Lubczynski]

Reporter: can you provide a testcase so I can verify your patch fixes the crash?

thanks.

Comment 4

[Peter Lubczynski]

*** Bug 65181 has been marked as a duplicate of this bug. ***

Comment 5

[Peter Lubczynski]

Here's a testcase from the dup.

Comment 6

[shrirang khanzode]

confirming bug ;)

Comment 7

[Peter Lubczynski]

The attached patch fixes the bug. It was on oversight on my part of my checkin 
from the other day. 

Andrei, can you approve?

Comment 8

[av (gone)]

r=a=av

Comment 9

[vidur (gone)]

sr=vidur

Comment 10

[Peter Lubczynski]

Patch checked into the trunk. This should be checked into the branch as 
well as it is a regression caused by bug 54437.

Comment 11

[Angela Butler]

Added branch accept to status whiteboard.  This patch is needed to fix 
54437 regression.  Peter, please check into the branch ASAP. Thx
.Angela...

Comment 12

[Peter Lubczynski]

Fix checked into the BRANCH.

Comment 13

[Vikas Bhargava]

Cool, my very first patch to Mozilla and fixed within 24 hours of me submitting 
it!!

Thanks to everyone involved,

Vikas

Comment 14

[shrirang khanzode]

Verified on windows (branch and trunk 0117), mac (branch and trunk 0117), linux 
(branch and trunk 0117). No crash occurs.