Skip to content

apl must support 8-channel, if not, the capture can not work on this platform.#33

Merged
lgirdwood merged 1 commit intomasterfrom
unknown repository
Jul 2, 2018
Merged

apl must support 8-channel, if not, the capture can not work on this platform.#33
lgirdwood merged 1 commit intomasterfrom
unknown repository

Conversation

@ghost
Copy link

@ghost ghost commented Jun 29, 2018

the gpmrb must support 8-channel,
if not, the capture feature can not work.

Signed-off-by: Wu Zhigang zhigang.wu@linux.intel.com

Copy link
Member

@lgirdwood lgirdwood left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This should be 2 pull requests. One for each feature.

Please also be more careful with the commit message descriptions. I think you mean "error" or "bug" instead of "chaos". An English thesaurus can help here.

src/audio/dai.c Outdated
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The pointers must be updated where they are copied, this delays the update for DAI meaning the DAI R/W can become out of sync with the HW before the pipeline is scheduled. This fix should probably go in buffer APIs to better cope with DMA position updates that are async with pipeline scheduling.

Please log an issues for this and include logs and any other relevant details.

Copy link
Author

@ghost ghost Jun 29, 2018

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I did the test with this patch on the UP^2 platform. it can fix the issue:
pause capture when playback is runing will fail #22.
But this patch can ONLY fix this on UP^2 platform. I tried it on GP. It still has this bug.
I think there still other factor will cause this DSP panic.
Maybe UP^2 and GP has something different.
During this test, I have to check whether there is an kernel XRUN happen in GP or not when I am back to office.
During this test, I did not find kernel XRUN On UP^2.

About the pointer update sync, from other side, the HW and buffer pointer are updated at the same time. because during the pipeline is running, the dma is still running.
Actually the pipeline is first triggered start in pipeline_trigger_sched_comp() function, it is same time with dma triggered start. from this point, after move the pointer update, the HW and buffer pointer is synced.

Copy link
Member

@plbossart plbossart left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

please fix the commit message, this is an APL change, not just a GP-MRB issue.

the apl must support 8-channel,
if not, the capture on GP can not work.

Signed-off-by: Wu Zhigang <zhigang.wu@linux.intel.com>
@ghost ghost changed the title gpmrb must 8-channel, if not, the capture can not work on this platform. apl must support 8-channel, if not, the capture can not work on this platform. Jul 2, 2018
@lgirdwood lgirdwood merged commit 1abc668 into thesofproject:master Jul 2, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants