[PATCH] pcm: set the snd_pcm_ioplug_status() tstamp field

[PATCH] pcm: set the snd_pcm_ioplug_status() tstamp field

[Jonas Holmberg]

Signed-off-by: Jonas Holmberg <jonashg@axis.com>
---
 src/pcm/pcm_ioplug.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/pcm/pcm_ioplug.c b/src/pcm/pcm_ioplug.c
index a437ca32..9b1b8ac3 100644
--- a/src/pcm/pcm_ioplug.c
+++ b/src/pcm/pcm_ioplug.c
@@ -115,6 +115,7 @@ static int snd_pcm_ioplug_status(snd_pcm_t *pcm, snd_pcm_status_t * status)
 	snd_pcm_ioplug_hw_ptr_update(pcm);
 	status->state = io->data->state;
 	status->trigger_tstamp = io->trigger_tstamp;
+	gettimestamp(&status->tstamp, pcm->tstamp_type);
 	status->avail = snd_pcm_mmap_avail(pcm);
 	status->avail_max = io->avail_max;
 	return 0;
-- 
2.26.2

[PATCH] pcm: set the snd_pcm_ioplug_status() tstamp field

[Jonas Holmberg]

Set the status tstamp field so that it can be accessed with
snd_pcm_status_get_htstamp().

Signed-off-by: Jonas Holmberg <jonashg@axis.com>
---
 src/pcm/pcm_ioplug.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/pcm/pcm_ioplug.c b/src/pcm/pcm_ioplug.c
index a437ca32..9b1b8ac3 100644
--- a/src/pcm/pcm_ioplug.c
+++ b/src/pcm/pcm_ioplug.c
@@ -115,6 +115,7 @@ static int snd_pcm_ioplug_status(snd_pcm_t *pcm, snd_pcm_status_t * status)
 	snd_pcm_ioplug_hw_ptr_update(pcm);
 	status->state = io->data->state;
 	status->trigger_tstamp = io->trigger_tstamp;
+	gettimestamp(&status->tstamp, pcm->tstamp_type);
 	status->avail = snd_pcm_mmap_avail(pcm);
 	status->avail_max = io->avail_max;
 	return 0;
-- 
2.26.2

Re: [PATCH] pcm: set the snd_pcm_ioplug_status() tstamp field

[Jonas Holmberg]

On Wed, Nov 04, 2020 at 02:49:56PM +0100, Jonas Holmberg wrote:
> Set the status tstamp field so that it can be accessed with
> snd_pcm_status_get_htstamp().
> 
> Signed-off-by: Jonas Holmberg <jonashg@axis.com>
> ---
>  src/pcm/pcm_ioplug.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/src/pcm/pcm_ioplug.c b/src/pcm/pcm_ioplug.c
> index a437ca32..9b1b8ac3 100644
> --- a/src/pcm/pcm_ioplug.c
> +++ b/src/pcm/pcm_ioplug.c
> @@ -115,6 +115,7 @@ static int snd_pcm_ioplug_status(snd_pcm_t *pcm, snd_pcm_status_t * status)
>  	snd_pcm_ioplug_hw_ptr_update(pcm);
>  	status->state = io->data->state;
>  	status->trigger_tstamp = io->trigger_tstamp;
> +	gettimestamp(&status->tstamp, pcm->tstamp_type);
>  	status->avail = snd_pcm_mmap_avail(pcm);
>  	status->avail_max = io->avail_max;
>  	return 0;
> -- 
> 2.26.2
> 

Is there a reason that the tstamp field of snd_pcm_ioplug_status() is 
always 0? I assumed that it was an oversight/bug so this patch sets it 
so that applications can read it with snd_pcm_status_get_htstamp(). I 
have tested it and it seems to work well.

BR
Jonas

Re: [PATCH] pcm: set the snd_pcm_ioplug_status() tstamp field

[Takashi Iwai]

On Thu, 05 Nov 2020 15:24:37 +0100,
Jonas Holmberg wrote:
> 
> On Wed, Nov 04, 2020 at 02:49:56PM +0100, Jonas Holmberg wrote:
> > Set the status tstamp field so that it can be accessed with
> > snd_pcm_status_get_htstamp().
> > 
> > Signed-off-by: Jonas Holmberg <jonashg@axis.com>
> > ---
> >  src/pcm/pcm_ioplug.c | 1 +
> >  1 file changed, 1 insertion(+)
> > 
> > diff --git a/src/pcm/pcm_ioplug.c b/src/pcm/pcm_ioplug.c
> > index a437ca32..9b1b8ac3 100644
> > --- a/src/pcm/pcm_ioplug.c
> > +++ b/src/pcm/pcm_ioplug.c
> > @@ -115,6 +115,7 @@ static int snd_pcm_ioplug_status(snd_pcm_t *pcm, snd_pcm_status_t * status)
> >  	snd_pcm_ioplug_hw_ptr_update(pcm);
> >  	status->state = io->data->state;
> >  	status->trigger_tstamp = io->trigger_tstamp;
> > +	gettimestamp(&status->tstamp, pcm->tstamp_type);
> >  	status->avail = snd_pcm_mmap_avail(pcm);
> >  	status->avail_max = io->avail_max;
> >  	return 0;
> > -- 
> > 2.26.2
> > 
> 
> Is there a reason that the tstamp field of snd_pcm_ioplug_status() is 
> always 0? I assumed that it was an oversight/bug so this patch sets it 
> so that applications can read it with snd_pcm_status_get_htstamp(). I 
> have tested it and it seems to work well.

I guess it's an oversight.  It's one of few places where the no real
hardware slave PCM is present, and the other one (null plugin) already
set it.

Also maybe cause snd_pcm_htimestamp() worked without this change; it's
updating the timstamp with a different code path.

In anyway, I applied your patch now.  Thanks!


Takashi