Closed Bug 558489 Opened 15 years ago Closed 15 years ago

[OOPP] Minefield autoreleasing NSThread objects with no pool in place

Categories

(Core :: IPC, defect)

x86
macOS
defect
Not set
normal

Tracking

()

RESOLVED FIXED

People

(Reporter: smichaud, Assigned: jfkthame)

References

(Blocks 1 open bug)

Details

(Keywords: memory-leak)

Attachments

(2 files)

Recent Minefield nightlies always display the following error in the console on startup: *** _NSAutoreleaseNoPool(): Object 0x6148a0 of class NSThread autoreleased with no pool in place - just leaking Here's the regression range: firefox-2010-04-02-03-mozilla-central http://hg.mozilla.org/mozilla-central/rev/c8f0660250e0 firefox-2010-04-03-03-mozilla-central http://hg.mozilla.org/mozilla-central/rev/53b8f71cd21e http://hg.mozilla.org/mozilla-central/pushloghtml?startdate=2010-04-02&enddate=2010-04-03+06%3A00%3A00 This probably implicates the patch for bug 555234. In my next comment I'll post a gdb trace of the error.
Forgot to mention that this error happens on both OS X 10.6.3 and 10.5.8.
Blocks: 559073
base::InitThreading() is called very early in startup, before we have any pools in place. The chromium code provides a stack-based autorelease pool helper, so we can simply use that here.
Assignee: nobody → jfkthame
Attachment #439104 - Flags: review?(joshmoz)
Attachment #439104 - Flags: review?(joshmoz) → review+
Status: NEW → RESOLVED
Closed: 15 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: