On Sat, Aug 3, 2013 at 11:30 PM, James Stone <jamesmstone@xxxxxxxxx> wrote: > > On Aug 3, 2013 9:28 PM, "Alan Stern" <stern@xxxxxxxxxxxxxxxxxxx> wrote: >> >> On Fri, 2 Aug 2013, James Stone wrote: >> >> > > OK - here's a trace with all patches applied. It didn't occur when >> > > the >> > > webcam was re-plugged this time.. This one happened when opening the >> > > trace file while it was being written to. I hope this is OK? It seems >> > > to be a fairly consistent way to get the audio to hang up, but I'm >> > > not >> > > sure it is such a "natural" way to induce the bug tho? >> > > >> > >> > Another trace - this time occurring when plugging in a usb game >> > controller.. so I guess the webcam isn't the culprit! >> >> Traces that show a delay of less than 1 ms aren't very interesting, in >> general. However, they do show that the scheduling code in ehci-hcd >> takes longer than it should (although it probably runs more quickly >> when tracing isn't turned on). >> >> I've known for a long time that the scheduler in ehci-hcd needs to be >> rewritten completely. It's a big job, so I've been avoiding it... >> >> Some event that causes a whole bunch of those new "underrun" messages >> in the system log would be worth seeing. >> > > All of these traces were associated with underrun messages but it seems > they > are not really capturing anything useful. Don't know if these syslog errors are also related? Aug 3 18:34:06 blueberry kernel: [35122.375869] retire_capture_urb: 3189 callbacks suppressed James -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
