On Mon, 11 May 2009, Dwayne Fontenot <dwayne.fontenot@xxxxxxx> wrote: > I will try your above suggestions tomorrow, however I wanted to update now with > a bit of new information: > > If I self test a single RU which is *not* one of the first 24 to have > an interrupt URB submitted, I never receive an interrupt message response from it. > > It's as if the interrupt URB submissions after the first 24 are "lost" > yet no error is returned from the usb_submit_urb() call. That's basically what I said. The URBs are submitted and sent to the hardware correctly, but the hardware then ignores some of them. > I should also mention that for any specific kernel version / configuration / arch > the number of returned interrupt messages remains constant, but with different > kernel versions / configurations (I have tested back to 2.6.23 on the 32-bit PPC > embedded platform which brought this problem to light) the number of returned > interrupt messages can vary; I have seen 24, 25 and 26. Sometimes it's possible to work around hardware limitations like this, but sometimes it isn't. Without knowing more about the exact nature of the hardware bug, I don't see what we can do about it. Except for the obvious: Never have more than 24 interrupt endpoints active at any time. 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
