It appears that stack traces logged on the tryserver for 64-bit Linux
debug builds can be completely bogus. (Look for "ASSERTION: stack" in
https://tbpl.mozilla.org/php/getParsedLog.php?id=13663470&tree=Try&full=1#error0
; I'm trying to use NS_ASSERTION(false, "stack") to dump the stack for
a probl
Hi!
As many of you are no doubt aware, Try Server load has been very high recently
[1] (sometimes taking up to 24 hours to get a full set of results). Bug 772458
has some dependants that will help, but won't make a significant difference in
the short term.
In the meantime please can everyone
On Thu, Jul 19, 2012 at 5:44 AM, Henri Sivonen wrote:
> Is this a known problem and are there known ways to get non-bogus
> stacks logged on 64-bit Linux on try?
Bug 755781.
There is no work around that I'm aware of.
- Kyle
___
dev-platform mailing
nsITimer.idl says nothing about thread safety. (Can we add some?)
From a glance at the code, it appears that it's safe to Cancel and
null-out an nsITimer from any given thread, so long as the timer
callback that's pointed to has thread-safe refcounting. Correct?
Jason
__
On 7/19/12 4:04 PM, Jason Duell wrote:
nsITimer.idl says nothing about thread safety. (Can we add some?)
From a glance at the code, it appears that it's safe to Cancel and
null-out an nsITimer from any given thread, so long as the timer
callback that's pointed to has thread-safe refcounting. Co
On 07/19/2012 01:16 PM, Boris Zbarsky wrote:
On 7/19/12 4:04 PM, Jason Duell wrote:
nsITimer.idl says nothing about thread safety. (Can we add some?)
From a glance at the code, it appears that it's safe to Cancel and
null-out an nsITimer from any given thread, so long as the timer
callback tha
Thanks to everyone who joined this discussion. We've now moved forward with
disabling OS X 10.5 support for FF17 in [1]. Per discussion here, we will
strive to leave code in place through the 17 cycle, but if it becomes difficult
to accomplish necessary work, we may end up breaking things to adv
Until recently, using TryChooser with -p macosx would give you working OS X
10.5 builds & test runs.
However, now that the OS X requirement on trunk has been raised to 10.6 [1],
any Try runs based on trunk need to use -p macosx64 (ie 10.6/10.7) rather than
-p macosx or your test runs will be:
On Thursday, 19 July 2012 23:28:39 UTC+1, Ed Morley wrote:
> However, now that the OS X requirement on trunk has been raised to 10.6 [1]
[1] https://bugzilla.mozilla.org/show_bug.cgi?id=772682
___
dev-platform mailing list
dev-platform@lists.mozilla.org
On Thu, Jul 19, 2012 at 11:04 PM, Jason Duell wrote:
> From a glance at the code, it appears that it's safe to Cancel and null-out
> an nsITimer from any given thread, so long as the timer callback that's
> pointed to has thread-safe refcounting. Correct?
At least the HTML parser relies on it be
10 matches
Mail list logo
