@abonislawski @lgirdwood There is a IPC timeout in PTL HDA config in pr-device-test that I've not seen in earlier PRs (granted HDA results not available for all PRs).

@abonislawski @lgirdwood There is a IPC timeout in PTL HDA config in pr-device-test that I've not seen in earlier PRs (granted HDA results not available for all PRs).

@kv2019i is there a link or log for this ?

@abonislawski @lgirdwood There is a IPC timeout in PTL HDA config in pr-device-test that I've not seen in earlier PRs (granted HDA results not available for all PRs).

@kv2019i is there a link or log for this ?

t=1039 ms: cmd3 arecord DMIC Raw: (2/50) Found volume ### | __%, active for 47 ms
t=1040 ms: cmd3 arecord DMIC Raw: Pressing SPACE
t=1184 ms: cmd0 aplay HDA Analog: (3/50) Found   === PAUSE ===  ,  pausing for 22 ms
t=1185 ms: cmd0 aplay HDA Analog: Pressing SPACE
t=1163 ms: cmd3 arecord DMIC Raw: (2/50) Found   === PAUSE ===  ,  pausing for 34 ms
t=1165 ms: cmd3 arecord DMIC Raw: Pressing SPACE
t=2174 ms: cmd3 arecord DMIC Raw: ERROR: arecord: do_pause:1597: pause release error: Connection timed out
t=2400 ms: cmd0 aplay HDA Analog: (4/50) Found volume ### | __%, active for 28 ms
t=2401 ms: cmd0 aplay HDA Analog: Pressing SPACE
t=2907 ms: cmd0 aplay HDA Analog: ERROR: #+                                                 | 00% aplay: pcm_write:2146: write error: Input/output error

2025-02-18 08:40:44 UTC [REMOTE_INFO] pipeline: HDA Analog with aplay
2025-02-18 08:40:44 UTC [REMOTE_INFO] pipeline: DMIC Raw with arecord
2025-02-18 08:40:44 UTC [REMOTE_INFO] Check expect exit status
declare -- cmd="journalctl_cmd --since=@1739867989"
journalctl_cmd is a function
journalctl_cmd () 

I'll rerun CI to be sure

SOFCI TEST

@abonislawski still seeing a failure on PTL HDA. Can you break this Kconfig down into smaller "=y" changes, it just could be one of the options breaking something. Thanks!

@abonislawski looks like just enabling the telemetry by itself is enough to show red on what looks like to be the multicore PTL topologies.
Just checking, does telemetry use VMH in anyway as @marcinszkudlinski is looking at a multicore issue there ?

@lgirdwood Yes, upon checking, I found that we have increased CONFIG_MEMORY_WIN_2_SIZE only for MTL and LNL. This seems to be a remaining issue with telemetry. It helps in a single debug test, waiting for the full CI now.