mbox series

[v1,0/5] ALSA: hda - Add DP-MST support for NVIDIA codecs

Message ID 20191114033704.18171-1-nmahale@nvidia.com (mailing list archive)
Headers show
Series ALSA: hda - Add DP-MST support for NVIDIA codecs | expand

Message

Nikhil Mahale Nov. 14, 2019, 3:36 a.m. UTC 
Today, DisplayPort multi-stream transport (DP-MST) audio is not
supported on codec drivers which don't use the audio component
(acomp) binding.

The commit ade49db337a9 (ALSA: hda/hdmi - Allow audio component for
AMD/ATI and Nvidia HDMI) added the DRM audio component binding for
the NVIDIA codec drivers, but the traditional HD-audio unsolicited
event code path is still in use by the proprietary NVIDIA graphics
driver.

This patch set adds DP-MST audio support for non-acomp codec drivers
which use the traditional HD-audio unsolicited event code path. This
patch set adds DP-MST support for NVIDIA codecs.

The patch set has been tested for HDMI/DP-SST/DP-MST(4 dp-mst audio
streams) configurations on NVIDIA Kepler and Maxwell GPUs, using
both the nouveau driver and the proprietary NVIDIA graphics driver.

Nikhil Mahale (5):
  ALSA: hda - Rename snd_hda_pin_sense to snd_hda_jack_pin_sense
  ALSA: hda - Add DP-MST jack support
  ALSA: hda - Add DP-MST conn list support
  ALSA: hda - Add DP-MST support for non-acomp codecs
  ALSA: hda - Add DP-MST support for NVIDIA codecs

 sound/pci/hda/hda_codec.c      |   5 +-
 sound/pci/hda/hda_generic.c    |  16 +--
 sound/pci/hda/hda_jack.c       | 111 ++++++++++++------
 sound/pci/hda/hda_jack.h       |  26 +++--
 sound/pci/hda/patch_ca0132.c   |  24 ++--
 sound/pci/hda/patch_cirrus.c   |   4 +-
 sound/pci/hda/patch_conexant.c |   2 +-
 sound/pci/hda/patch_hdmi.c     | 255 +++++++++++++++++++++++++++++------------
 sound/pci/hda/patch_realtek.c  |  46 ++++----
 sound/pci/hda/patch_sigmatel.c |  12 +-
 10 files changed, 333 insertions(+), 168 deletions(-)

Comments

Takashi Iwai Nov. 14, 2019, 10:38 a.m. UTC | #1 
On Thu, 14 Nov 2019 04:36:59 +0100,
Nikhil Mahale wrote:
> 
> Today, DisplayPort multi-stream transport (DP-MST) audio is not
> supported on codec drivers which don't use the audio component
> (acomp) binding.
> 
> The commit ade49db337a9 (ALSA: hda/hdmi - Allow audio component for
> AMD/ATI and Nvidia HDMI) added the DRM audio component binding for
> the NVIDIA codec drivers, but the traditional HD-audio unsolicited
> event code path is still in use by the proprietary NVIDIA graphics
> driver.
> 
> This patch set adds DP-MST audio support for non-acomp codec drivers
> which use the traditional HD-audio unsolicited event code path. This
> patch set adds DP-MST support for NVIDIA codecs.
> 
> The patch set has been tested for HDMI/DP-SST/DP-MST(4 dp-mst audio
> streams) configurations on NVIDIA Kepler and Maxwell GPUs, using
> both the nouveau driver and the proprietary NVIDIA graphics driver.

Thanks for the patch.  The logic looks good, but there are a few
concerns in implementation details.  Will comment in the reply of each
patch.

The timing is a bit late for merging into 5.5, though, for this kind
of non-trivial changes.  It'll be likely applied after 5.5 merge
window, so for 5.6.


Takashi
Nikhil Mahale Nov. 15, 2019, 9:37 a.m. UTC | #2 
On 11/14/19 4:08 PM, Takashi Iwai wrote:
> On Thu, 14 Nov 2019 04:36:59 +0100,
> Nikhil Mahale wrote:
>>
>> Today, DisplayPort multi-stream transport (DP-MST) audio is not
>> supported on codec drivers which don't use the audio component
>> (acomp) binding.
>>
>> The commit ade49db337a9 (ALSA: hda/hdmi - Allow audio component for
>> AMD/ATI and Nvidia HDMI) added the DRM audio component binding for
>> the NVIDIA codec drivers, but the traditional HD-audio unsolicited
>> event code path is still in use by the proprietary NVIDIA graphics
>> driver.
>>
>> This patch set adds DP-MST audio support for non-acomp codec drivers
>> which use the traditional HD-audio unsolicited event code path. This
>> patch set adds DP-MST support for NVIDIA codecs.
>>
>> The patch set has been tested for HDMI/DP-SST/DP-MST(4 dp-mst audio
>> streams) configurations on NVIDIA Kepler and Maxwell GPUs, using
>> both the nouveau driver and the proprietary NVIDIA graphics driver.
> 
> Thanks for the patch.  The logic looks good, but there are a few
> concerns in implementation details.  Will comment in the reply of each
> patch.

Thanks Takashi, I am sending out 2nd version of patch set in which
I tried to address you concerned.

Thanks,
Nikhil Mahale

> The timing is a bit late for merging into 5.5, though, for this kind
> of non-trivial changes.  It'll be likely applied after 5.5 merge
> window, so for 5.6.
> 
> 
> Takashi
> 

-----------------------------------------------------------------------------------
This email message is for the sole use of the intended recipient(s) and may contain
confidential information.  Any unauthorized review, use, disclosure or distribution
is prohibited.  If you are not the intended recipient, please contact the sender by
reply email and destroy all copies of the original message.
-----------------------------------------------------------------------------------