Jaroslav Kysela

It's a third-party module ? This flag is set by configure (libtool) and it is not required for the standard alsa-lib apps (except the modules). I would suggest to add...

It may not be an audio issue at all. Some other task / kernel driver may eat CPU time preventing the low-latency operation. From the result: ``` avail : 216...

I did some debugging using ftrace (trace-cmd) and it really seems that the system scheduler blocks the latency task by a kworker task: ``` latency-7395 [002] 22719.778930: hwptr: pcmC0D0c/sub0: POS:...

I think that I found the upstream change between 5.9 and 5.10 kernels in the scheduler which causes this behavior: https://lore.kernel.org/lkml/c596a06773658d976fb839e02843a459ed4c2edf.1479204252.git.bristot@redhat.com/ See the patch description. You can set the old...

For the archiving purposes - `trace-cmd` commands: ``` LATENCY=128 ; USLEEP=200 trace-cmd start -b 10000 -e 'sched_wakeup*' -e sched_switch -e gpio_value -e irq_handler_entry -e irq_handler_exit -e snd_pcm \ /home/redhat/git/alsa-lib/test/.libs/latency -C...

> Would a different RT setup in latency program change the behavior? It sounds like a kernel regression if it shows such a latency even with the highest RT priority....

Wow, I did not expect that. It seems that SCHED_OTHER is better than FIFO/RR with the current kernels. I added the SCHED_OTHER policy support to latency.c - `-O other` (75f8e2e73e03f628a4f2ba55ca8aa3e9f50cdbd9)....

I think that it would be bad to change the application defaults according the kernel version. Also, it's clearly the kernel scheduler problem which should be reported to LKML (linux...

It looks like a driver (kernel) issue. The poll() should return with an error. You can also check the stream status in `/proc/asound/card0/pcm0p/sub0/*` files.

For `snd_pcm_hw_params_set_drain_silence(0)`, the silencing the rest of period (and piece of next one for most hw) is not active. The application is responsible to fill this (using standard I/O and...