On Thu, 13 Oct 2011, Pitt, Jason N wrote:
> hey Alan-
>
> That camera bug happened again to two of my cameras....I'm not sure
> if it's related but it never happened over several days when I had
> your other patch. Could there be something different about how this
> patch works that might effect how the uvcvideo driver handles the
> error? I'll try testing both kernel versions and see if it only
> happens in one or the other.
The official patch is functionally the same as the preceding test
version.
It's possible that occasionally the udelay() really is needed after
all. You can try adding it (as in the patch below) to see if it makes
any difference.
Alan Stern
Index: usb-3.1/drivers/usb/host/ehci-sched.c
===================================================================
--- usb-3.1.orig/drivers/usb/host/ehci-sched.c
+++ usb-3.1/drivers/usb/host/ehci-sched.c
@@ -50,8 +50,10 @@ static unsigned ehci_read_frame_index(st
* re-read the register to get the correct value.
*/
uf = ehci_readl(ehci, &ehci->regs->frame_index);
- if (unlikely(ehci->frame_index_bug && ((uf & 7) == 0)))
+ if (unlikely(ehci->frame_index_bug && ((uf & 7) == 0))) {
+ udelay(1);
uf = ehci_readl(ehci, &ehci->regs->frame_index);
+ }
return uf;
}
--
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
