563285 - Crash on quit [@ objc_msgSend | __CFRunLoopRun ] related to searchResultsAvailable

Status: RESOLVED FIXED

(Camino Graveyard :: General, defect)

Description

[David Munch]

User-Agent:       Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en; rv:1.9.2.5pre) Gecko/20100503 Camino/2.1a1pre (like Firefox/3.6.5pre)
Build Identifier: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.6; en; rv:1.9.2.5pre) Gecko/20100503 Camino/2.1a1pre (like Firefox/3.6.5pre)

Crash while quitting. No open tabs/windows.

Reproducible: Always

Steps to Reproduce:
1. Quit Camino
2. Random crash
3. ???
4. Profit!
Actual Results:  
Crash was totally random.

Expected Results:  
No crash.

Comment 1

[David Munch]

Attachment: Crash log and console output

Comment 2

[David Munch]

Erh, not always reproducible, first time i got it.. Overlooked that variable.

Comment 3

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

I had David file this for investigation because of the following info in the Apple crash report (yay selector name!):

Application Specific Information:
objc_msgSend() selector name: searchResultsAvailable

Thread 0 Crashed:  Dispatch queue: com.apple.main-thread
0   libobjc.A.dylib               	0x99649eec objc_msgSend + 44

searchResultsAvailable is related to our autocomplete impl; do we possibly have lifetime issues with it?

Comment 4

[Stuart Morgan]

We sure do; AutoCompleteTextField.mm should be calling cancelSearch in its dealloc. I should have caught that in review :(

Comment 5

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Attachment: Fix?

Like so?

Comment 6

[Stuart Morgan]

Comment on attachment 444289 [details] [diff] [review]
Fix?

Exactly like so!
sr=smorgan

Comment 7

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

http://hg.mozilla.org/camino/rev/3b275ed2bae0 and cvs trunk.

At least we caught it before it was widely available (have I mentioned how much I love the objc_msgSend() selector info in Mac OS X crash logs?  I don't suppose Chromium is interested enough in http://code.google.com/p/google-breakpad/issues/detail?id=332 to task someone fix it? ;) ).