5 Jul
2018
5 Jul
'18
3:02 p.m.
On Thu, 05 Jul 2018 14:34:02 +0200, Raphaël Dingé wrote:
Hi,
Looking more in depth into the code I understood my problem was between the chair and the computer screen *sigh*.
So the code was calling snd_pcm_sw_params_set_stop_threshold which apart from stopping the stream on underruns apparently "eats" the xrun from the kernel tracing perspective.
If I comment out this portion of code, I get the EPIPE in snd_pcm_avail_update as well as the xrun events in /sys/kernel/tracing/trace.
Thanks a million again to put me on the right track. I've learnt a lot from this exchange.
Good to hear that it's no driver problem :)
thanks,
Takashi