On Fri, 20 Feb 2004 16:31:09 -0500 (EST) Alan Stern <[EMAIL PROTECTED]> wrote:
> On Fri, 20 Feb 2004, Stephen Hemminger wrote: > > > The last problem was a red herring, since the UHCI changes hadn't made it into > > 2.6.3. This new problem occurs on 2.6.3-bk plus two of Alan's patches: > > 1) Subject: [linux-usb-devel] PATCH: (as186) Remove unneeded and error-provoking > > variable in UHCI > > 2) Subject: [linux-usb-devel] PATCH: (as194) Simplify locking in UHCI > > > > Now seeing the following OOPS after 15min to 30min of leaving the irda > > driver running. In idle state it just keeps exchanging short packets with > > it's neighbor and each transmit involves aborting the outstanding receive > > urb's first. > > Ah, a new different problem, good. > > This oops occurred because the URB was dequeued before it had been fully > queued. There's already a check on the other side -- the urb_enqueue > function will see that the URB has already been dequeued. But the > urb_dequeue routine did not check for this. > > Here is a patch that I believe will fix the problem. Great, the kernel with this patch ran successfully all weekend. Looks like no more races in the unlink path. ------------------------------------------------------- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
