Oops, I only replied to Greg first time. On 10/7/06, Greg KH <[EMAIL PROTECTED]> wrote: > On Sat, Oct 07, 2006 at 01:26:13AM -0400, Christopher Monty Montgomery wrote: > Monty, how does other operating systems report this back to the driver? > I'm guessing that Windows uses yet-another-message, right?
I don't actually know about windows. I have no windows coding experience, I just use a few apps. > How about OS-X? the OSX EHCI driver immediately returns an error status, eg kIOReturnIsoTooOld, when the next URB is submitted, similar to what I was proposing with EL2NSYNC. The low-level OSX driver, though, does not alter internal state when it returns an XRUN; the same submission again would error again. It appears to be up to the higher level driver to decide how to catch up (I'm only reading the code, I do not have a document describing the correct coding convention). Like in my code, OSX does not distinguish between the different cases that delayed the URB (loss of sync or xrun) All it knows is that the transaction showed up too late to go in the schedule where it should have. [BTW, this is assuming what I was told is true and that OSX and Darwin 10.4 share their EHCI code. That is all coming from the Darwin EHCI driver source.] Monty ------------------------------------------------------------------------- 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
