On Mon, Nov 23, 2015 at 07:13:12PM +0100, Lars-Peter Clausen wrote: > On 11/23/2015 06:28 PM, Vinod Koul wrote: > > On Mon, Nov 23, 2015 at 05:06:41PM +0100, Lars-Peter Clausen wrote: > >> On 11/23/2015 04:52 PM, Vinod Koul wrote: > >>> From: Jeeja KP <jeeja.kp@xxxxxxxxx> > >>> > >>> For DAPM resume, we should first change the power state of the > >>> card and then recheck the endpoints. This ensures the dapm is > >>> resumed first and then userspace can resume the streams. > >> > >> The problem is that DAPM uses the userspace state to decide whether to power > >> up or not. So this change won't work, it will keep the endpoints suspended. > > > > Well our observation are bit different... > > > > When testing with aplay and suspending the system the DAPM was suspending > > fine and then alsa suspends PCMs, but on resume the DAPM is resumed last and > > the PCMs first, due to which the devices start underun! > > The DAPM should be resumed first and then we should allow userspace access > > (PCM resume from aplay) > > > > The change of order here helps in that > > Sorry, overlooked something. We go to D2 earlier on and DAPM only is > suspended when the state is D3 and not only when it is not in D0. No issues > Patch looks good. > > Reviewed-by: Lars-Peter Clausen <lars@xxxxxxxxxx> Thanks for the review -- ~Vinod _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxx http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
