On 23.11.2017 09:08, Takashi Iwai wrote:
> On Wed, 22 Nov 2017 20:17:34 +0100,
> Maciej S. Szmigiero wrote:
>>
>> This format is similar to existing SNDRV_PCM_FORMAT_{S,U}20_3 that keep
>> 20-bit PCM samples in 3 bytes, however i.MX6 platform SSI FIFO does not
>> allow 3-byte accesses (including DMA) so a 4-byte format is needed for it.
>>
>> Signed-off-by: Maciej S. Szmigiero <mail@xxxxxxxxxxxxxxxxxxxxx>
>> ---
>> include/sound/pcm.h | 8 ++++++++
>> include/sound/soc-dai.h | 2 ++
>> include/uapi/sound/asound.h | 10 +++++++++-
>> sound/core/pcm_misc.c | 16 ++++++++++++++++
>> 4 files changed, 35 insertions(+), 1 deletion(-)
>>
>> diff --git a/include/sound/pcm.h b/include/sound/pcm.h
>> index 24febf9e177c..7ad2d3f0934f 100644
>> --- a/include/sound/pcm.h
>> +++ b/include/sound/pcm.h
>> @@ -191,6 +191,10 @@ struct snd_pcm_ops {
>> #define SNDRV_PCM_FMTBIT_DSD_U32_LE _SNDRV_PCM_FMTBIT(DSD_U32_LE)
>> #define SNDRV_PCM_FMTBIT_DSD_U16_BE _SNDRV_PCM_FMTBIT(DSD_U16_BE)
>> #define SNDRV_PCM_FMTBIT_DSD_U32_BE _SNDRV_PCM_FMTBIT(DSD_U32_BE)
>> +#define SNDRV_PCM_FMTBIT_S20_4LE _SNDRV_PCM_FMTBIT(S20_4LE)
>> +#define SNDRV_PCM_FMTBIT_U20_4LE _SNDRV_PCM_FMTBIT(U20_4LE)
>> +#define SNDRV_PCM_FMTBIT_S20_4BE _SNDRV_PCM_FMTBIT(S20_4BE)
>> +#define SNDRV_PCM_FMTBIT_U20_4BE _SNDRV_PCM_FMTBIT(U20_4BE)
>
> The conventional names aren't with "4" suffix,
> e.g. SNDRV_PCM_FMTBIT_S20_LE.
Will rename it in a respin.
> Also, there are still empty slots under 32, e.g. start from 25.
> The formats over 31 were used for 3 bytes or other unusual formats
> (although nowadays it makes little sense), and the slots < 32 would
> fit for 4 bytes linear format.
Will renumber these formats to start from 25 in a respin.
> It's still an open question whether we should increase the protocol
> number when we add a new PCM format definition. I guess it's not, as
> the ABI behavior itself doesn't change, but I might have overlooked
> some possible breakage.
This isn't an incompatible, application-breaking, ABI change (it is
rather kind of an additional "ABI"), so I think an ALSA protocol number
should not be increased for it.
> thanks,
>
> Takashi
>
Thanks,
Maciej
_______________________________________________
Alsa-devel mailing list
Alsa-devel@xxxxxxxxxxxxxxxx
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
