Re: [PATCH v3 5/6] ALSA: xen-front: Implement ALSA virtual sound driver

[Oleksandr Andrushchenko <andr2000@gmail.com>]

On 05/23/2018 07:00 AM, Takashi Sakamoto wrote:
> Hi,
>
> On May 14 2018 15:27, Oleksandr Andrushchenko wrote:
> > diff --git a/sound/xen/xen_snd_front_alsa.c 
> b/sound/xen/xen_snd_front_alsa.c
> > new file mode 100644
> > index 000000000000..5041f83e98d2
> > --- /dev/null
> > +++ b/sound/xen/xen_snd_front_alsa.c
> > ...
> > +/*
> > + * FIXME: The mmaped data transfer is asynchronous and there is no
> > + * ack signal from user-space when it is done. This is the
> > + * reason it is not implemented in the PV driver as we do need
> > + * to know when the buffer can be transferred to the backend.
> > + */
> > ...
>
> In ALSA PCM interface v2.0.14 or later, SNDRV_PCM_INFO_SYNC_APPLPTR is
> available. This flag express that userspace applications are expected to
> call ioctl(2) with 'SNDRV_PCM_IOCTL_SYNC_PTR' even if they oprate to
> memory mapped page frames for PCM samples. For detail, please refer to a
> commit 42f945970af9 ('ALSA: pcm: Add the explicit appl_ptr sync
> support')[1].
>
> As a supplement, please refer to below commits:
>  * 4b671f577474 ('ALSA: pcm: Add an ioctl to specify the supported
>    protocol version')[2]
>  * b602aa8eb1a0 ('ALSA: pcm: Disable only control mmap for explicit
>    appl_ptr sync')[3]
>
> Alsa-lib v1.1.5 or later supports this flag and existent applications
> can run transparently.
>
> For developers, a tracepoint may be useful:
>  * fccf53881e9b ('ALSA: pcm: add 'applptr' event of tracepoint')[4]
>
> For your information.
>
> [1] 
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=42f945970af9
> [2] 
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=4b671f577474
> [3] 
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=b602aa8eb1a0
> [4] 
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=fccf53881e9b
>
Oh, this is very useful, thank you so much for such a detailed
description. I'll take a close look at the above and see if
Xen sound driver can benefit from this.
>
> Regards
>
> Takashi Sakamoto
Thank you very much,
Oleksandr