Thanks Martin, > Finally, it turned out the decision when to append a length=0 usb-packet > is partly done just the wrong way. > > The patch below does > * append length=0 packet in the correct situation only, > i.e. basically when (length%maxpacketsize)==0
Good, that's basically a "should be ==, not !=" patch, yes? Whoops! > * make USB_ZERO_PACKET only effective on OUT transaction, because > it's useless for IN and uhci is restricted to out as well. A patch recently sent by updating flags handling (removing the QUEUE_BULK, including from irda-usb, and renaming DISABLE_SPD) should remove any need for this change. If you use that with your patch, the patches should work fine together. (Do they?) The deal is that the URB submission code should now verify that ZERO_PACKET is only set (from drivers) for writes. One check for all the HCDs, enabled by CONFIG_USB_DEBUG (at least for now). No need to add extra instructions to every driver ... ones that might not act quite the same ... ;) > Testing with 2.5.24 and irda-usb demonstrates things are working again. Good, I'll want 2.5.next to have a slightly simpler patch, since I seem to be on a kick for smaller code ... just change to "==". - Dave ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Stuff, things, and much much more. http://thinkgeek.com/sf _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
