[alsa-devel] [PATCH] hiface: Fix sample rate changes

[Takashi Iwai]

On Mon, 28 Nov 2016 15:11:25 +0100,
Takashi Iwai wrote:
> 
> On Mon, 28 Nov 2016 11:09:25 +0100,
> Michael Trimarchi wrote:
> > 
> > Hi
> > 
> > On Mon, Nov 28, 2016 at 10:55 AM, Jussi Laako <jussi at sonarnerd.net> wrote:
> > > Hi,
> > >
> > >> Can you please sent it inline? Not having access at the hardware now
> > >> but Antonio should have it
> > >
> > >
> > > Sure! I have tested this with my hiFace...
> > >
> > >
> > >         - Jussi
> > >
> > >
> > > From f4eec5602d86b5f938abed48a5725f59141d32cd Mon Sep 17 00:00:00 2001
> > > From: Jussi Laako <jussi at sonarnerd.net>
> > > Date: Mon, 28 Nov 2016 01:11:46 +0200
> > > Subject: [PATCH] Fix M2Tech hiFace driver sampling rate change
> > >
> > > Sampling rate changes after first set one are not reflected to the
> > > hardware, while driver and ALSA think the rate has been changed.
> > >
> > > Fix the problem by properly stopping the interface at the beginning of
> > > prepare call, allowing new rate to be set to the hardware. This keeps
> > > the hardware in sync with the driver.
> > >
> > > Signed-off-by: Jussi Laako <jussi at sonarnerd.net>
> > > ---
> > >  sound/usb/hiface/pcm.c | 2 ++
> > >  1 file changed, 2 insertions(+)
> > >
> > > diff --git a/sound/usb/hiface/pcm.c b/sound/usb/hiface/pcm.c
> > > index 2c44139..33db205 100644
> > > --- a/sound/usb/hiface/pcm.c
> > > +++ b/sound/usb/hiface/pcm.c
> > > @@ -445,6 +445,8 @@ static int hiface_pcm_prepare(struct snd_pcm_substream
> > > *alsa_sub)
> > >
> > >         mutex_lock(&rt->stream_mutex);
> > >
> > > +       hiface_pcm_stream_stop(rt);
> > > +
> > 
> > Ok, I was expecting stream already paused on STOP or SUSPEND call. I
> > have the impression that they enqueue 0 urb still and they don't
> > really
> > stop it.
> 
> The trigger callback is atomic, so it can't kill urb by itself, and
> hiface_pcm_trigger() just flips the sub->active flag.
> 
> So, yes, calling hiface_pcm_stream_stop() should be correct.
> One question is whether it should be unconditionally called, or it
> should be done only when something got changed.
> 
> I guess it's OK to call it always (it's safer), but it may lead to
> some glitches or such that wasn't present beforehand, when the player
> restarts the stream.

And I forgot this patch until now, sorry.  I applied it now for
4.10-rc1.


thanks,

Takashi