Query on get_time_info

[Pierre-Louis Bossart]

On 1/27/23 08:03, Raghu Ballappa Bankapur wrote:
> HI All,
> 
> 
> We see get_time_info is not enabled from ALSA core. ay reason behind?

I think you meant ASoC core?

> we enabled like below with some more code, we are able to get_time_info
> to lower driver.
> 
> static int dpcm_be_connect(struct snd_soc_pcm_runtime *fe,
>                 struct snd_soc_pcm_runtime *be, int stream)
> @@ -2907,6 +2927,7 @@ int soc_new_pcm(struct snd_soc_pcm_runtime *rtd,
> int num)
>                 rtd->ops.hw_free        = dpcm_fe_dai_hw_free;
>                 rtd->ops.close          = dpcm_fe_dai_close;
>                 rtd->ops.pointer        = soc_pcm_pointer;
> +               rtd->ops.get_time_info  = soc_pcm_get_time_info;
>         } else {
>                 rtd->ops.open           = soc_pcm_open;
>                 rtd->ops.hw_params      = soc_pcm_hw_params;
> @@ -2915,6 +2936,7 @@ int soc_new_pcm(struct snd_soc_pcm_runtime *rtd,
> int num)
>                 rtd->ops.hw_free        = soc_pcm_hw_free;
>                 rtd->ops.close          = soc_pcm_close;
>                 rtd->ops.pointer        = soc_pcm_pointer;
> +               rtd->ops.get_time_info  = soc_pcm_get_time_info;
>         }
> 
> Is this ALSA limitation or any other approach is followed to enable this
> implementation.

No, it's a miss.

I had a patch to fix this in
https://github.com/thesofproject/linux/pull/3792, but somehow it never
made its path upstream. I don't fully recall the details, only vaguely
that we first added support for the .delay(), but the get_time_info part
was tested back in August 2022.

Would that patch work for you?