Bug 1059572 - Part 2: Make absolutely sure a timer is removed before reinitting it. r=nfroyd

--HG--
extra : rebase_source : 9a952241e046321a25adf52167bd7a538d25f37a
This commit is contained in:
Byron Campen [:bwc] 2015-07-28 10:10:54 -05:00
Родитель e145322168
Коммит 309a57122b
1 изменённых файлов: 1 добавлений и 17 удалений

Просмотреть файл

@ -205,23 +205,7 @@ nsTimerImpl::InitCommon(uint32_t aType, uint32_t aDelay)
return rv;
}
/**
* In case of re-Init, both with and without a preceding Cancel, clear the
* mCanceled flag and assign a new mGeneration. But first, remove any armed
* timer from the timer thread's list.
*
* If we are racing with the timer thread to remove this timer and we lose,
* the RemoveTimer call made here will fail to find this timer in the timer
* thread's list, and will return false harmlessly. We test mArmed here to
* avoid the small overhead in RemoveTimer of locking the timer thread and
* checking its list for this timer. It's safe to test mArmed even though
* it might be cleared on another thread in the next cycle (or even already
* be cleared by another CPU whose store hasn't reached our CPU's cache),
* because RemoveTimer is idempotent.
*/
if (mArmed) {
gThread->RemoveTimer(this);
}
mCanceled = false;
mTimeout = TimeStamp();
mGeneration = gGenerator++;