On Friday 06 October 2006 8:05 pm, Alan Stern wrote: > On Fri, 6 Oct 2006, Christopher "Monty" Montgomery wrote:
> > In any case, setting a status flag on the next completion is not good > > enough; that tells you an xrun happend a long time ago. Notification > > has to happen upon next submission. > > I don't know the details of ehci-hcd, but this may well be unfeasible. It > would require checking the status of previous submissions every time a new > submission is made, which might be a considerable extra load. It's feasible; it happens now! And the tests are just modular arithmetic to make sure that the recorded "next" schedule slot (microframe number) hasn't already been passed. The "next" slot is easily known, it's the next period after the last queued URB's start frame as well as the next index in the periodic schedule array. (I still wouldn't swear that the math there is 100% correct, but that hasn't been raised as an issue so it seems to _act_ correct enough.) - Dave ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ linux-usb-devel@lists.sourceforge.net To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
