On Mon, Jun 09, 2014 at 03:06:22PM -0700, Jarno Rajahalme wrote:
> When reaching the end of a prefix trie, we checked one bit off the end
> to the intended data. However, since the trie node in that case has
> NULLs for both edge links, this did not result in incorrect
> functionality.
>
> Found via check-valgrind.
>
> Reported-by: Ben Pfaff <b...@nicira.com>
> Signed-off-by: Jarno Rajahalme <jrajaha...@nicira.com>
> ---
> v2: Clarify trie_lookup_value() by returning from the case when ofs
> >= n_bits.
I think that the only behavioral change here is:
+ if (ofs >= n_bits) {
+ *checkbits = n_bits; /* Full prefix. */
+ return match_len;
+ }
and updating the callers, which seems good.
Otherwise, I see two stylistic changes:
1. s/plen/ofs/. OK.
2. Change "for" to "while". This one is a little puzzling
because I don't see a need for it.
I'd be tempted to put the behavioral changes in a patch by themselves
(to make it really obvious what's changing), and the stylistic changes
in another, but I'll leave it up to you.
Acked-by: Ben Pfaff <b...@nicira.com>
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
