On Wed, 1 Jun 2011, Soho Soho123 wrote: > Dear Alan, > Below is my comments. > > 2011/6/1 Alan Stern <stern@xxxxxxxxxxxxxxxxxxx>: > > On Wed, 1 Jun 2011, Soho Soho123 wrote: > > > >> Dear Alan, > >> > >> > >> We have questions: > >> We still do not get unstanding about :why we get -FEBIG error when uvc > >> driver re-submit urb yet. > >> Because we try to use another platform that CPU is faster. The error > >> is still occur. > >> And,the amount of URB is 5 totally. > >> the another platform : CPU 620MHz, RAM 64MB > > > > Do you still see "now" > stream->next_uframe + 160? > > Soho: Yes, you could check the picture that I attached in last e-mail. This means there is still too much interrupt latency. You should try to find out how much time is spent whenever itd_complete() calls ehci_urb_done(). Most of that time will be in the uvcvideo driver. > >> About the precision of timestamp that I post from usbmon. > >> the precision is in microsecond, do you mean nano is needed, right? > > > > No. Look at the timestamps; you'll see that even though the times > > printed are in microseconds, the last four digits are always 2448. > > (That's also true in the log messages above.) This means the actual > > precision is only 10 ms. > > > > Soho: in our platform the time precision is 10ms per tick . In a PC, the time precision is between 1 and 10 ms per tick. But PCs also have high-precision timers available; apparently your platform does not. Without high-precision timestamps, there's no way to tell what's causing your latencies to be so large. Have you tried running this test on a regular PC instead of your embedded platform? Does that work okay? One other thing you can try: Increase UVC_MAX_PACKETS to 100. 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
