On Thu, 21 Jun 2012, Ming Lei wrote: > >> Also from the usbmon trace, all packets are completed with length of 12, > >> which indicates that there aren't capture data at all. > > > > Maybe there wasn't supposed to be any. �Anyway, that's a separate > > problem. > > It might be a result of mismatch between packet size and altsetting set > into device. No, that's not possible. These are isochronous-IN transfers, which means the device is not aware of the requested packet size. > >> > Now the question is: Why were interrupts not delivered for such a long > >> > period? �I'm not at all sure how to find out. > >> > >> Maybe the irqsoff tracer can help troubleshoot it. > > > > What is the irqsoff tracer? > > Please see the section of "The Tracers" in Documentation/trace/ftrace.txt. Thank you. Yes, that would be a very good thing for the bug reporter to try. > Also considered that io_watchdog is disabled in intel ehci controller, is it > possible there are no interrupts coming from ehci hardware? It's possible but very unlikely. Prior to the 52-ms gap, interrupts arrived like clockwork every 4 ms. Even if one got lost, the next should have arrived with a gap of only 8 ms. Alan Stern -- 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
