[alsa-devel] [PATCH 04/12] ALSA: hda: check stream decoupled register state

[Jeeja KP]

On Fri, Dec 02, 2016 at 08:03:18PM +0100, Takashi Iwai wrote:
> On Fri, 02 Dec 2016 18:41:41 +0100,
> jeeja.kp at intel.com wrote:
> > 
> > From: Jeeja KP <jeeja.kp at intel.com>
> > 
> > Check stream decoupled register value with requested value
> > before decoupling/coupling the stream.
> > 
> > Signed-off-by: Jeeja KP <jeeja.kp at intel.com>
> > ---
> >  sound/hda/ext/hdac_ext_stream.c | 15 +++++++++------
> >  1 file changed, 9 insertions(+), 6 deletions(-)
> > 
> > diff --git a/sound/hda/ext/hdac_ext_stream.c b/sound/hda/ext/hdac_ext_stream.c
> > index 3be051a..bd8187b 100644
> > --- a/sound/hda/ext/hdac_ext_stream.c
> > +++ b/sound/hda/ext/hdac_ext_stream.c
> > @@ -128,14 +128,17 @@ void snd_hdac_ext_stream_decouple(struct hdac_ext_bus *ebus,
> >  {
> >  	struct hdac_stream *hstream = &stream->hstream;
> >  	struct hdac_bus *bus = &ebus->bus;
> > +	u32 val;
> > +	int mask = AZX_PPCTL_PROCEN(hstream->index);
> >  
> >  	spin_lock_irq(&bus->reg_lock);
> > -	if (decouple)
> > -		snd_hdac_updatel(bus->ppcap, AZX_REG_PP_PPCTL, 0,
> > -				AZX_PPCTL_PROCEN(hstream->index));
> > -	else
> > -		snd_hdac_updatel(bus->ppcap, AZX_REG_PP_PPCTL,
> > -					AZX_PPCTL_PROCEN(hstream->index), 0);
> > +	val = ((readw(bus->ppcap + AZX_REG_PP_PPCTL) & mask) >> mask);
> > +
> > +	if (decouple && (val == 0))
> > +		snd_hdac_updatel(bus->ppcap, AZX_REG_PP_PPCTL, 0, mask);
> > +	else if (!decouple && (val > 0))
> > +		snd_hdac_updatel(bus->ppcap, AZX_REG_PP_PPCTL, mask, 0);
> > +
> 
> The usage of snd_hdac_updatel() looks strange.
> The third argument must be the mask bits and the fourth be the value
> bits.  So, usually for clearing a bit
> 
> 	snd_hdac_update(pcap, REG, mask, 0);
> 
> and for setting a bit
> 
> 	snd_hdac_update(pcap, REG, mask, mask);
> 
> Passing 0 to the mask bits means to replace the whole bits.
> 
> This usage pattern is found already in the old code, so it's not new,
> and I'm not sure whether this behavior is intentional...

Agreed, the correct way to set a bit is to pass third and fourth argument
as mask. Will fix this overall usage of this macro and send it in another
patch series.

> 
> 
> thanks,
> 
> Takashi

--