Repository navigation
Conversation
Fix out_valid_bit_depth for 16-bit and 24-bit audio formats. Signed-off-by: Yong Zhi <yong.zhi@intel.com>
lgirdwood
left a comment
There was a problem hiding this comment.
@RanderWang @ranj063 pls review
RanderWang
left a comment
There was a problem hiding this comment.
why use 32bit valid sample size ? Internal pipeline uses 24bit valid bits to do calculation.
but this change is needed to make the format matching to work in the kernel isnt it? |
|
why use 32bit valid sample size ? Internal pipeline uses 24bit valid bits to do calculation.
yes. @yongzhi1 What's the problem now ? |
|
Hi @RanderWang , we observed no sound coming from "Deepbuffer Amps" when playing 16-bit material. |
@yongzhi1 it would be helpful if you can file a bug that this PR fixes so we can see why this is needed |
|
The issue seems to be deepbuffer-playback pipeline only supports $HDA_HOST_OUTPUT_CLASS node_type, which works fine for HDA and SDW but not I2S. |
That seems very odd @yongzhi1, @RanderWang just added the deep buffer for the nocodec topology which does use I2S/SSP. |
@plbossart yong is doing experiment and pick my change to his private branch |
|
Learned a clever way to do it from @RanderWang , updated top level conf instead yongzhi1@09966aa, so no need to hard-code the changes here, thanks all for reviewing the changes. |
Fix out_valid_bit_depth for 16-bit and 24-bit audio formats.
Signed-off-by: Yong Zhi yong.zhi@intel.com