Hi Stefan,
On Wed, 2019-01-09 at 09:33 +0100, Stefan Wahren wrote:
>
> But i noticed a regression introduced in 5.0-rc1, which might be
> related. I
> tried to run "vchiq_test -f 1" and it never completes. After
> reverting commit
> 852b2876a8a8 ("staging: vchiq: rework remove_event handling")
> vchiq_test [1]
> works as expected again.
I had a look at that one, it seems that it's due to the fact that vchiq
relies on completions only sleeping whenever their internal count
reaches 0 (it's incremented on every complete()). As opposed to
wait_queues, which will wait until an event is triggered regardless of
previous state.
Regards,
Nicolas
Attachment:
signature.asc
Description: This is a digitally signed message part
_______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
