On Saturday 16 February 2008, Karsten Wiese wrote: > Am Mittwoch, 13. Februar 2008 schrieb Karsten Wiese: > > > > ITDs can be detached from urbs, before the active frame elapses. Now those > > ITDs are immediately recycled. > > If the ITD is reused then, transactions based on its new usage can already > > happen while the frame is still active, too early. > > Hi Dave, > > your patch that moves ITD-recycling past URB-completion solves the > "common case": completion callback resubmits urb.
Right, and if you're pinging me on the $SUBJECT patch, the answer is that it's in my review queue. - Dave > It misses > a) client driver delays urb resubmission past completion callback. > I only know my us122l driver doing this. > b) stream stops midframe: ITDs are put back into the dma_pool. > Things go wrong, if another urb_submit() picks up the ITD that has just > been recycled from a still active frame. > This will happen once an ehci client driver starts using uframe precise > scheduling. > Hmm? > > Karsten > - To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
