On 1/23/18 11:41 PM, Xiuli Pan wrote:

From: Pan Xiuli xiuli.pan@linux.intel.com

We may have different buffer size around volume. Change the size due to the endpoint setting.

Signed-off-by: Pan Xiuli xiuli.pan@linux.intel.com

topology/sof/pipe-volume-capture.m4 | 2 +- topology/sof/pipe-volume-playback.m4 | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/topology/sof/pipe-volume-capture.m4 b/topology/sof/pipe-volume-capture.m4 index 258f8fa..8532aa3 100644 --- a/topology/sof/pipe-volume-capture.m4 +++ b/topology/sof/pipe-volume-capture.m4 @@ -53,7 +53,7 @@ W_PGA(0, Master Capture Volume, PIPELINE_FORMAT, 2, 2, 2) W_BUFFER(0, COMP_BUFFER_SIZE(2, COMP_SAMPLE_SIZE(PIPELINE_FORMAT), PIPELINE_CHANNELS, SCHEDULE_FRAMES)) W_BUFFER(1, COMP_BUFFER_SIZE(2,

• COMP_SAMPLE_SIZE(PIPELINE_FORMAT), PIPELINE_CHANNELS, SCHEDULE_FRAMES))

• COMP_SAMPLE_SIZE(DAI_FORMAT), PIPELINE_CHANNELS, SCHEDULE_FRAMES))

I understand this fixes a problem but I don't think the way we deal with pipelines makes sense. Somehow we've got to avoid tying pipelines and dais in such a rigid way to scale. There will be configurations with 'dai-to-dai' routing, and conversely 'host-to-host' (e.g. for echo reference).

# # DAI definitions diff --git a/topology/sof/pipe-volume-playback.m4 b/topology/sof/pipe-volume-playback.m4 index b532161..54bc13d 100644 --- a/topology/sof/pipe-volume-playback.m4 +++ b/topology/sof/pipe-volume-playback.m4 @@ -53,7 +53,7 @@ W_PGA(0, Master Playback Volume, PIPELINE_FORMAT, 2, 2, 2) W_BUFFER(0, COMP_BUFFER_SIZE(2, COMP_SAMPLE_SIZE(PIPELINE_FORMAT), PIPELINE_CHANNELS, SCHEDULE_FRAMES)) W_BUFFER(1, COMP_BUFFER_SIZE(2,

• COMP_SAMPLE_SIZE(PIPELINE_FORMAT), PIPELINE_CHANNELS, SCHEDULE_FRAMES))

• COMP_SAMPLE_SIZE(DAI_FORMAT), PIPELINE_CHANNELS, SCHEDULE_FRAMES))

  # # DAI definitions