Re: [PATCH] ASoC: fsl_sai: Handle slave mode per TX/RX direction

[Daniel Baluta]

On Wed, Aug 14, 2019 at 4:01 AM Nicolin Chen <nicoleotsuka@xxxxxxxxx> wrote:
>
> On Sun, Aug 11, 2019 at 10:45:17PM +0300, Daniel Baluta wrote:
> > From: Viorel Suman <viorel.suman@xxxxxxx>
> >
> > The SAI interface can be a clock supplier or consumer
> > as a function of stream direction. e.g SAI can be master
> > for Tx and slave for Rx.
> >
> > Signed-off-by: Viorel Suman <viorel.suman@xxxxxxx>
> > Signed-off-by: Daniel Baluta <daniel.baluta@xxxxxxx>
> > ---
> >  sound/soc/fsl/fsl_sai.c | 18 +++++++++---------
> >  sound/soc/fsl/fsl_sai.h |  2 +-
> >  2 files changed, 10 insertions(+), 10 deletions(-)
> >
> > diff --git a/sound/soc/fsl/fsl_sai.c b/sound/soc/fsl/fsl_sai.c
> > index 4a346fcb5630..69cf3678c859 100644
> > --- a/sound/soc/fsl/fsl_sai.c
> > +++ b/sound/soc/fsl/fsl_sai.c
> > @@ -273,18 +273,18 @@ static int fsl_sai_set_dai_fmt_tr(struct snd_soc_dai *cpu_dai,
>
> This function is called for both TX and RX at the same time from
> fsl_sai_set_dai_fmt() so I don't actually see how it can operate
> in two opposite directions from this change alone. Anything that
> I have missed?

Good catch. I'm missing a patch that updates fmt after the first call
of fsl_sai_set_dai_fmt_tr.
Let me update the patch and resend.