ACK/cmt: [SRU][Focal][OEM-5.6][PATCH 0/1] ASoC:amd:renoir: the dmic can't record sound after suspend and resume

[Colin Ian King]

On 04/08/2020 03:03, Hui Wang wrote:
> BugLink: https://bugs.launchpad.net/bugs/1890220
> 
> Because our OEM project is waiting for this fix, let me submit it to
> OEM-5.6 and focal kernel. for groovy kernel, the patch will be merged
> to G kernel with stable patches.
> 
> [Impact]
> We have backported the amd renoir audio driver to oem-5.6 and focal
> kernel, recently we found if the dmic is working and we suspend and
> resume, the dmic can't record the sound after resume.
> 
> [Fix]
> After resuming, the driver needs to restore 2 more registers, then the
> dmic could record the sound.
> 
> [Test Case]
> Boot the kernel with this patch, open the sound-setting and run suspend
> and resume, then use arecord to record the sound, verify the sound is
> successfully recorded by aplaying it.
> 
> [Regression Risk]
> Low, this patch is verified on the amd renoir machine, and this patch
> got ack from AMD engineer.
> 
> 
> 
> Hui Wang (1):
>   UBUNTU: SAUCE: ASoC: amd: renoir: restore two more registers during
>     resume
> 
>  sound/soc/amd/renoir/acp3x-pdm-dma.c | 29 +++++++++-------------------
>  1 file changed, 9 insertions(+), 20 deletions(-)
> 

This is not even in linux-next, so much like Stefan's comment, I'm a bit
concerned that it's not yet trickled down further into kernel repos that
have had more testing exposure.  However, it is limited to specific H/W
so potential regression impact is small. Not sure if the AMD engineer
tested this, but given they ACK'd it I have to trust their word that it
is OK.

Acked-by: Colin Ian King <colin.king at canonical.com>