> skb_set_owner() code doesn't check for a prior owner -- perhaps we should add a
> test for this?
Yep
It should though be classified as a "until 2.2" debugging test, the
thing to check when looking at the code is that people orphan SKB's
before setting ownership when appropriate.
I really don't want a silly check like this in the final 2.2 code,
especially since set_owner is such a often used operation in every
networking hot path. For code reviews, debugging, and verification
for the remaining 2.1.x stuff, yeah it's fine, but after that toss
it.
Later,
David S. Miller
davem@dm.cobaltmicro.com
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu