Alright, I see what you're getting at. We get a frame with a RIF from
a host on the local ring, but the RII is not restored before the
frame is passed up the stack. Good catch.
rii_p is overloaded. We use it both to determine "localness" and to
determine whether or not the RII should be restored. Patch to follow
(unless you've got one coded.)
> Also, there are two exit points to the function, one on a alloc failure
> and the other off the end. If the alloc fails the byte is not unclobbered.
> OK, I admit this is picky!
"Picky" is good. :-) This has already been identified and a patch
submitted.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/