On Thu, Mar 05, 2009 at 11:01:11AM -0500, Alan Stern wrote: > On Wed, 4 Mar 2009, Greg KH wrote: > > > Oh horrors. > > > > How about we just leave actual_length as int to allow this to work > > properly? Putting a negative value in an unsigned variable, while > > workable, is just a mess. > > > > How about the patch below instead for the whole thing? > > I'm not so sure. It makes sense for actual_length to be unsigned, so > it ought to be changed. > > The code in uhci-q.c doesn't have to use pseudo-negative values. I did > it that way because it was easy and because it would give the expected > output during debugging. But it doesn't have to work that way. Here's > another approach. That makes sense to me. I've taken your patch and applied it, and also added another one for the actual_length change. I'll repost all 3 patches now. If you like it, can you give your signed-off-by: to it? thanks, greg k-h -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
