[alsa-devel] [PATCHv2 5/5] ASoC: fsl-sai: rename big_endian_data to is_msb_first.

[Nicolin Chen]

On Mon, Aug 25, 2014 at 02:04:46AM +0000, Li.Xiubo at freescale.com wrote:
> > -----Original Message-----
> > From: Nicolin Chen [mailto:nicoleotsuka at gmail.com]
> > Sent: Monday, August 25, 2014 9:51 AM
> > To: Xiubo Li-B47053
> > Cc: broonie at kernel.org; alsa-devel at alsa-project.org; mark.rutland at arm.com;
> > timur at tabi.org; devicetree at vger.kernel.org
> > Subject: Re: [PATCHv2 5/5] ASoC: fsl-sai: rename big_endian_data to
> > is_msb_first.
> > 
> > On Tue, Aug 19, 2014 at 12:14:53PM +0800, Xiubo Li wrote:
> > > Signed-off-by: Xiubo Li <Li.Xiubo at freescale.com>
> > > ---
> > >  Documentation/devicetree/bindings/sound/fsl-sai.txt | 8 ++++----
> > >  sound/soc/fsl/fsl_sai.c                             | 6 +++---
> > >  sound/soc/fsl/fsl_sai.h                             | 2 +-
> > >  3 files changed, 8 insertions(+), 8 deletions(-)
> > >
> > > diff --git a/Documentation/devicetree/bindings/sound/fsl-sai.txt
> > b/Documentation/devicetree/bindings/sound/fsl-sai.txt
> > > index aed1f21..e25ef38 100644
> > > --- a/Documentation/devicetree/bindings/sound/fsl-sai.txt
> > > +++ b/Documentation/devicetree/bindings/sound/fsl-sai.txt
> > > @@ -20,9 +20,9 @@ Required properties:
> > >    See ../pinctrl/pinctrl-bindings.txt for details of the property values.
> > >  - big-endian: Boolean property, required if all the SAI device registers
> > >    are big-endian rather than little-endian.
> > > -- big-endian-data: If this property is absent, the little endian mode will
> > > -  be in use as default, or the big endian mode will be in use for all the
> > > -  fifo data.
> > > +- msb-first: Configures whether the LSB or the MSB is transmitted first for
> > > +  the fifo data. If this property is absent, the LSB is transmitted first
> > as
> > > +  default, or the MSB is transmitted first.
> > >  - fsl,sai-synchronous-rx: This is a boolean property. If present,
> > indicating
> > >    that SAI will work in the synchronous mode (sync Tx with Rx) which means
> > >    both the transimitter and receiver will send and receive data by
> > following
> > > @@ -53,5 +53,5 @@ sai2: sai at 40031000 {
> > >  	      dmas = <&edma0 0 VF610_EDMA_MUXID0_SAI2_TX>,
> > >  		   <&edma0 0 VF610_EDMA_MUXID0_SAI2_RX>;
> > >  	      big-endian;
> > > -	      big-endian-data;
> > > +	      msb-first;
> > >  };
> > > diff --git a/sound/soc/fsl/fsl_sai.c b/sound/soc/fsl/fsl_sai.c
> > > index a6eb784..4e48431 100644
> > > --- a/sound/soc/fsl/fsl_sai.c
> > > +++ b/sound/soc/fsl/fsl_sai.c
> > > @@ -175,7 +175,7 @@ static int fsl_sai_set_dai_fmt_tr(struct snd_soc_dai
> > *cpu_dai,
> > >  	bool tx = fsl_dir == FSL_FMT_TRANSMITTER;
> > >  	u32 val_cr2 = 0, val_cr4 = 0;
> > >
> > > -	if (!sai->big_endian_data)
> > > +	if (!sai->is_msb_first)
> > >  		val_cr4 |= FSL_SAI_CR4_MF;
> > 
> > IIRC, MF stands for 'MSB First' but the condition is !is_msb_first..
> > 
> 
> Yes, I will fix this.
> 
> > And also we can't not simply inverse the condition here since those
> > platforms without the original 'big_endian_data' property will be
> > broken unless they add the new 'is_msb_first' into the DT bindings,
> > which is, however, a violation due to breaking the old bindings.
> > 
> > So I guess is_lsb_first might be better here?
> >
> > And actually, Xiubo, what's your purpose to add this patch? I can't
> > see any commit comments to explain the reason. So could you please
> > say something about it?
> 
> Months ago, the hardware team needed to do some IP test based LS1 needing
> this to be added to the driver. And for now there is hasn't any other reason
> for this yet.
 
Hmm...the patch only renames big_endian_data to is_msb_first.

So whatever the problem of LS1 is, this change doesn't matter to LS1
at all unless LS1 has a specific property in its DT binding: If using
is_msb_first works for LS1, using big_endian_data without applying
this patch does work as well, doesn't it?

I think this patch is more likely making the binding explicit so that
any platform like LS1 will not miss the essential property. Then the
commit comments could have mentioned it if so.

I'm looking forward to your new version. And I give you an extra info
about SAI on i.MX. It should be configured as Little Endian and MSB
first. So with the current DT binding it doesn't have big_endian*
properties. Any replacement for big_endian_data shall be compatible
to this case as well as I mentioned in the previous reply.

Thank you,
Nicolin