Pierre-Louis Bossart

@dbaluta the premise of the ioctl was to standardize on common parameters. If we abuse this ioctl and make it dependent on the fimware implementation, it's a very large deviation...

> Module interface cannot make decision by itself. e.g use 32bit width PCM or 24bit width PCM. It needs someone externally to instruct it. Why not? don't we have information...

Minor mistake with COMP_SCHED, now fixed but IPC error on TRIGGER_START with XTOS. Let's see if that happens with Zephyr as well .... ```` [ 65.907869] snd_sof:ipc3_log_header: sof-audio-pci-intel-cnl 0000:00:1f.3: ipc...

With Zephyr I only get an IPC timeout, not DSP panic. ```` 29.457822] snd_sof:ipc3_log_header: sof-audio-pci-intel-cnl 0000:00:1f.3: ipc tx: 0x60040000: GLB_STREAM_MSG: TRIG_START [ 29.960132] sof-audio-pci-intel-cnl 0000:00:1f.3: ipc tx timed out for...

I think I found the problem, it's a conceptual mistake IMHO. For capture, the 'start' component for hw_params and prepare is the host and the propagation works upstream to the...

https://sof-ci.01.org/sofpr/PR6131/build1189/devicetest/ is all good except for a TIMEOUT for APL-Up2-NOCODEC in suspend-resume tests. This is likely not introduced by this PR.

SOFCI TEST

Last update to include a demux on capture. The new topology is as follows:  Local tests show clean audio capture on PCM 0 and PCM4, with an issue however...

Updates for stop seem to work in local tests...

all tests ok so far, except for alsa-bat which fails due to a bad mixer combination. I don't know how many times I've asked that we stop setting mixer values...