[NETFILTER]: nf_conntrack: don't inline early_drop()

early_drop() is only called *very* rarely, unfortunately gcc inlines it
into the hotpath because there is only a single caller. Explicitly mark
it noinline.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Patrick McHardy 2008-01-31 04:41:44 -08:00 committed by David S. Miller
parent ffaa9c100b
commit 76eb946040
1 changed files with 3 additions and 3 deletions

View File

@ -420,7 +420,7 @@ EXPORT_SYMBOL_GPL(nf_conntrack_tuple_taken);
/* There's a small race here where we may free a just-assured
connection. Too bad: we're in trouble anyway. */
static int early_drop(unsigned int hash)
static noinline int early_drop(unsigned int hash)
{
/* Use oldest entry, which is roughly LRU */
struct nf_conntrack_tuple_hash *h;
@ -472,8 +472,8 @@ struct nf_conn *nf_conntrack_alloc(const struct nf_conntrack_tuple *orig,
/* We don't want any race condition at early drop stage */
atomic_inc(&nf_conntrack_count);
if (nf_conntrack_max
&& atomic_read(&nf_conntrack_count) > nf_conntrack_max) {
if (nf_conntrack_max &&
unlikely(atomic_read(&nf_conntrack_count) > nf_conntrack_max)) {
unsigned int hash = hash_conntrack(orig);
if (!early_drop(hash)) {
atomic_dec(&nf_conntrack_count);