Re: [alsa-devel] [PATCH V6 3/4] ASoC: pcm_dmaengine: Extract snd_dmaengine_pcm_refine_runtime_hwparams

[John Stultz <john.stultz@linaro.org>]

On Thu, Jan 16, 2020 at 11:11 PM Shengjiu Wang <shengjiu.wang@gmail.com> wrote:
>
> Hi
>
> On Thu, Jan 16, 2020 at 1:56 PM John Stultz <john.stultz@linaro.org> wrote:
> >
> > On Wed, Jan 8, 2020 at 8:58 PM John Stultz <john.stultz@linaro.org> wrote:
> > > On Thu, Sep 26, 2019 at 6:50 PM Shengjiu Wang <shengjiu.wang@nxp.com> wrote:
> > > >
> > > > When set the runtime hardware parameters, we may need to query
> > > > the capability of DMA to complete the parameters.
> > > >
> > > > This patch is to Extract this operation from
> > > > dmaengine_pcm_set_runtime_hwparams function to a separate function
> > > > snd_dmaengine_pcm_refine_runtime_hwparams, that other components
> > > > which need this feature can call this function.
> > > >
> > > > Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
> > > > Reviewed-by: Nicolin Chen <nicoleotsuka@gmail.com>
> > >
> > > As a heads up, this patch seems to be causing a regression on the HiKey board.
> > >
> > > On boot up I'm seeing:
> > > [   17.721424] hi6210_i2s f7118000.i2s: ASoC: can't open component
> > > f7118000.i2s: -6
> > >
> > > And HDMI audio isn't working. With this patch reverted, audio works again.
> > >
> > >
> > > > diff --git a/sound/core/pcm_dmaengine.c b/sound/core/pcm_dmaengine.c
> > > > index 89a05926ac73..5749a8a49784 100644
> > > > --- a/sound/core/pcm_dmaengine.c
> > > > +++ b/sound/core/pcm_dmaengine.c
> > > > @@ -369,4 +369,87 @@ int snd_dmaengine_pcm_close_release_chan(struct snd_pcm_substream *substream)
> > > ...
> > > > +       ret = dma_get_slave_caps(chan, &dma_caps);
> > > > +       if (ret == 0) {
> > > > +               if (dma_caps.cmd_pause && dma_caps.cmd_resume)
> > > > +                       hw->info |= SNDRV_PCM_INFO_PAUSE | SNDRV_PCM_INFO_RESUME;
> > > > +               if (dma_caps.residue_granularity <= DMA_RESIDUE_GRANULARITY_SEGMENT)
> > > > +                       hw->info |= SNDRV_PCM_INFO_BATCH;
> > > > +
> > > > +               if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
> > > > +                       addr_widths = dma_caps.dst_addr_widths;
> > > > +               else
> > > > +                       addr_widths = dma_caps.src_addr_widths;
> > > > +       }
> > >
> > > It seems a failing ret from dma_get_slave_caps() here is being returned...
> > >
> > > > +
> > > > +       /*
> > > > +        * If SND_DMAENGINE_PCM_DAI_FLAG_PACK is set keep
> > > > +        * hw.formats set to 0, meaning no restrictions are in place.
> > > > +        * In this case it's the responsibility of the DAI driver to
> > > > +        * provide the supported format information.
> > > > +        */
> > > > +       if (!(dma_data->flags & SND_DMAENGINE_PCM_DAI_FLAG_PACK))
> > > > +               /*
> > > > +                * Prepare formats mask for valid/allowed sample types. If the
> > > > +                * dma does not have support for the given physical word size,
> > > > +                * it needs to be masked out so user space can not use the
> > > > +                * format which produces corrupted audio.
> > > > +                * In case the dma driver does not implement the slave_caps the
> > > > +                * default assumption is that it supports 1, 2 and 4 bytes
> > > > +                * widths.
> > > > +                */
> > > > +               for (i = SNDRV_PCM_FORMAT_FIRST; i <= SNDRV_PCM_FORMAT_LAST; i++) {
> > > > +                       int bits = snd_pcm_format_physical_width(i);
> > > > +
> > > > +                       /*
> > > > +                        * Enable only samples with DMA supported physical
> > > > +                        * widths
> > > > +                        */
> > > > +                       switch (bits) {
> > > > +                       case 8:
> > > > +                       case 16:
> > > > +                       case 24:
> > > > +                       case 32:
> > > > +                       case 64:
> > > > +                               if (addr_widths & (1 << (bits / 8)))
> > > > +                                       hw->formats |= pcm_format_to_bits(i);
> > > > +                               break;
> > > > +                       default:
> > > > +                               /* Unsupported types */
> > > > +                               break;
> > > > +                       }
> > > > +               }
> > > > +
> > > > +       return ret;
> > >
> > > ... down here.
> > >
> > > Where as in the old code...
> > >
> > > > diff --git a/sound/soc/soc-generic-dmaengine-pcm.c b/sound/soc/soc-generic-dmaengine-pcm.c
> > > > index 748f5f641002..b9f147eaf7c4 100644
> > > > --- a/sound/soc/soc-generic-dmaengine-pcm.c
> > > > +++ b/sound/soc/soc-generic-dmaengine-pcm.c
> > >
> > > > @@ -145,56 +140,12 @@ static int dmaengine_pcm_set_runtime_hwparams(struct snd_pcm_substream *substrea
> > > >         if (pcm->flags & SND_DMAENGINE_PCM_FLAG_NO_RESIDUE)
> > > >                 hw.info |= SNDRV_PCM_INFO_BATCH;
> > > >
> > > > -       ret = dma_get_slave_caps(chan, &dma_caps);
> > > > -       if (ret == 0) {
> > > > -               if (dma_caps.cmd_pause && dma_caps.cmd_resume)
> > > > -                       hw.info |= SNDRV_PCM_INFO_PAUSE | SNDRV_PCM_INFO_RESUME;
> > > > -               if (dma_caps.residue_granularity <= DMA_RESIDUE_GRANULARITY_SEGMENT)
> > > > -                       hw.info |= SNDRV_PCM_INFO_BATCH;
> > > > -
> > > > -               if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
> > > > -                       addr_widths = dma_caps.dst_addr_widths;
> > > > -               else
> > > > -                       addr_widths = dma_caps.src_addr_widths;
> > > > -       }
> > >
> > > ...the ret from dma_get_slave_caps()  checked above, but is not
> > > actually returned.
> > >
> > > Suggestions on how to sort this out?
> >
> > Just wanted to check in on this, as I'm still seeing this regression with -rc6.
> >
> Compare with the old code. it seems that we shouldn't check the return value.
>
> Could you help to test below changes?
>
> --- a/sound/soc/soc-generic-dmaengine-pcm.c
> +++ b/sound/soc/soc-generic-dmaengine-pcm.c
> @@ -138,12 +138,10 @@ dmaengine_pcm_set_runtime_hwparams(struct
> snd_soc_component *component,
>         if (pcm->flags & SND_DMAENGINE_PCM_FLAG_NO_RESIDUE)
>                 hw.info |= SNDRV_PCM_INFO_BATCH;
>
> -       ret = snd_dmaengine_pcm_refine_runtime_hwparams(substream,
> +       snd_dmaengine_pcm_refine_runtime_hwparams(substream,
>                                                         dma_data,
>                                                         &hw,
>                                                         chan);
> -       if (ret)
> -               return ret;
>
>         return snd_soc_set_runtime_hwparams(substream, &hw);
>  }

Yes, thanks for taking a look at this! Your patch does appear to avoid
the regression.
(Though you'll want to drop the ret declaration to avoid "warning:
unused variable 'ret'" compiler warnings.)

thanks
-john