On Mon, 5 Jun 2006, David Brownell wrote: > On Saturday 03 June 2006 9:33 am, Alan Stern wrote: > > On Fri, 2 Jun 2006, Vivek Dharmadhikari wrote: > > > > > > Host send OUT transcation with 31 Bytes > > > Device reponds with NYET. > > > 30 seconds IDLE time on bus. > > Where IDLE == only SOFs are seen? > > Or the hardware idle state, by which you mean the port was suspended?
Good question... > If EHCI had sane timer handling, it would be easier to add > code doing things like forcing endpoints out of NYET mode > every few seconds. > > For paranoia's sake, it would be good to verify more of the > interactions with the host. The previous request and response > especially ... did they trigger an error of some kind? > If so, how was it recovered from? > That said, we could more easily change the host software. At least, if > someone with a hardware analyser could isolate the problem ... :) Forunately Vivek does have an analyzer! Vivek, can you please run the same test over again? This time also collect the usb-storage verbose debugging log so we can see exactly what happens from the host's point of view and compare it with the Catalyst's record. We'd like to get as much detailed information as possible leading up to the point where the problem occurs. You can stop the test as soon as one of those 30-second timeouts occurs. Alan Stern _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
