19 Jan
2015
19 Jan
'15
10:59 a.m.
On 01/19/2015, 10:52 AM, Takashi Iwai wrote:
At Mon, 19 Jan 2015 10:42:55 +0100, Jiri Slaby wrote:
Wait for the timer to really quit, not only by ensuring it is not in the critical section.
BTW the spin_lock in here does not disable BH, so that it could deadlock.
The stop function is called from the PCM trigger callback which is irq-disabled. So, you can't use del_timer_sync() here.
The same as 3/3 I was looking at compr* :(. Ignore this one, then.
thanks,
--
js
suse labs