On Wed, Aug 21, 2024 at 05:15:00PM +0300, Mathias Nyman wrote: > My best guess is that these are caused by a full xhci event ring. > > If event ring is full it would stop transfer rings, and we would see > missed service events. This matches the logs. > > The events get handled and cleared in the interrupt handler, but > not every event generates an interrupt. for isoc transfers we intentionally > block a lot of event triggered interrupts. This has piqued my interest. Can you point me to the code in the kernel source where this filtering happens? > The event ring size has doubled since 6.6.45 kernel, using a newer kernel > could help. This changed in 6.7 kernel: > 28084d3fcc3c ("xhci: Use more than one Event Ring segment") I tried with a 6.10.6 kernel from archlinux and the symptoms didnt' change. Right now I'm running 6.7.0.6.realtime1-7-rt from archlinux and the symptoms are also the same. > If possible I'd like to see xhci traces of this issue. Here is a curl'able trace (from running timeout -s KILL 0.1 jackd -R -P 70 -d alsa -d hw:USB -p 48 -n 2 (thanks to Michal for the suggestion)): https://uni-bielefeld.sciebo.de/s/1ixBodEpmSg8M2k/download Or should I include it inline (it's ca. 1M.) Kind regards, FPS