Re: [PATCH 2/2] ASoC: sunxi: compatibility for sun6i to SPDIF

[Maxime Ripard <maxime.ripard-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org>]

[-- Attachment #1: Type: text/plain, Size: 1970 bytes --]

On Sun, Jul 31, 2016 at 12:40:48AM +1000, Julian Calaby wrote:
> Hi Marcus,
> 
> On Sun, Jul 31, 2016 at 12:27 AM,  <codekipper-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> > From: Marcus Cooper <codekipper-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
> >
> > The A31 SoC uses the same SPDIF block as found in earlier SoCs, but its
> > reset is controlled via a separate reset controller.
> >
> > The DMA also complains when the maxburst is set to 4 so it's been adjusted
> > to 8 which suites both the older and newer SoCs.
> >
> > Signed-off-by: Marcus Cooper <codekipper-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
> > ---
> >  sound/soc/sunxi/sun4i-spdif.c | 17 ++++++++++++++++-
> >  1 file changed, 16 insertions(+), 1 deletion(-)
> >
> > diff --git a/sound/soc/sunxi/sun4i-spdif.c b/sound/soc/sunxi/sun4i-spdif.c
> > index 0b04fb0..88fbb3a 100644
> > --- a/sound/soc/sunxi/sun4i-spdif.c
> > +++ b/sound/soc/sunxi/sun4i-spdif.c
> > @@ -482,11 +485,23 @@ static int sun4i_spdif_probe(struct platform_device *pdev)
> >         }
> >
> >         host->dma_params_tx.addr = res->start + SUN4I_SPDIF_TXFIFO;
> > -       host->dma_params_tx.maxburst = 4;
> > +       host->dma_params_tx.maxburst = 8;
> >         host->dma_params_tx.addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
> >
> >         platform_set_drvdata(pdev, host);
> >
> > +       if (of_device_is_compatible(pdev->dev.of_node,
> > +                                   "allwinner,sun6i-a31-spdif")) {
> 
> Given how much Allwinner likes to shuffle stuff around with each SoC
> generation, would it make sense to add a flag for this in some
> compatible specific config structure instead of checking against the
> compatible?

It really depends on the size of the quirks you have to maintain. If
it's just a single one like here, I'd say it's less of a hassle (and
actually easier and conciser to implement).

Maxime

-- 
Maxime Ripard, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com