random snippage... > > I as well have noticed the only chipsets I have problems with are the VIA > > uhci chipsets. As I said previously, I know a "special" UHCI driver was > > released by Microsoft/VIA that corrected a lot of problems that were being > > had with VIA's chipset. I would persue a fix for it myself, but I don't > > know enough about the host controller code to be able to talk one-to-one > > to a VIA representative. If the maintainer of one of the *uhci.c drivers > > were to look into it, I think it definitely be worth doing. > > The main problem is the babble. I think there's only so many things that > we can do in that situation. > > Maybe we can retry the failed TD? I think that data toggling will allow > that to be safe. > > That is unless there's hidden proprietary bit twiddling on the HC, but > I'd be surprised if they release that kind of information to us.
I'm not so sure that Microsoft and VIA have a solution for this problem - sure, VIA has a filter driver which may hide some of the weirdness, but it still happens. From my own experience with VIA boards in all revisions of Windows, there's always a few of them which *don't work* (in my particular case, Interrupt out, but it seems that "it" happens with all kinds of transfers). If you have some time to waste, take a look at the forums on http://www.viaarena.com/ - there's one dedicated to USB, and it has enough posts to make you scared. ..tom _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
