Comment #14 on issue 25857 by [email protected]: Crash :
-[ChromeEventProcessingWindow shortcircuitEvent:]
http://code.google.com/p/chromium/issues/detail?id=25857
I'm concerned with that solution because I think this is a symptom of a
potentially bigger problem.
Inside of a "close" command we are creating our own autorelease pool and
then actually performing the close,
and then releasing the pool before exiting the close command. We want some
(if not all) of the objects being
autoreleased to be autoreleased in the close commands pool, not a subpool.
We have a similar problem in the
browser window (brower_window_controller.mm #269) where we have the
following comment:
// We can't actually use |-autorelease| here because there's an embedded
// run loop in the |-performClose:| which contains its own autorelease
pool.
// Instead we call it after a zero-length delay, which gets us back
// to the main event loop.
[self performSelector:@selector(autorelease)
withObject:nil
afterDelay:0];
I'm concerned what other latent bugs may be hiding in the dual release pool
world. I have a discussion with
mark and pink scheduled for today to talk about it.
--
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings
--~--~---------~--~----~------------~-------~--~----~
Automated mail from issue updates at http://crbug.com/
Subscription options: http://groups.google.com/group/chromium-bugs
-~----------~----~----~----~------~----~------~--~---
