On Thu, Jun 04, 2009 at 03:16:17PM +0200, Patrick McHardy (kaber@xxxxxxxxx) wrote:
> Exactly. But please verify that by reordering, you're not moving
> the more commonly used members out of the first cacheline.
That's what I ended up with. arptable does not use match entries. Did I
miss others?
As of cache line alignment - I moved the only writable member at the end
of the structure, so it should even improve things a bit, although I
really do not think this will ever be noticeble.
--
Added hook number into match extension parameter structure.
Signed-off-by: Evgeniy Polyakov <zbr@xxxxxxxxxxx>
diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
index 7b1a652..65a08ae 100644
--- a/include/linux/netfilter/x_tables.h
+++ b/include/linux/netfilter/x_tables.h
@@ -184,9 +184,10 @@ struct xt_counters_info
* @matchinfo: per-match data
* @fragoff: packet is a fragment, this is the data offset
* @thoff: position of transport header relative to skb->data
- * @hotdrop: drop packet if we had inspection problems
+ * @hook: hook number given packet came from
* @family: Actual NFPROTO_* through which the function is invoked
* (helpful when match->family == NFPROTO_UNSPEC)
+ * @hotdrop: drop packet if we had inspection problems
*/
struct xt_match_param {
const struct net_device *in, *out;
@@ -194,8 +195,9 @@ struct xt_match_param {
const void *matchinfo;
int fragoff;
unsigned int thoff;
- bool *hotdrop;
+ unsigned int hooknum;
u_int8_t family;
+ bool *hotdrop;
};
/**
diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c
index 820252a..db2540e 100644
--- a/net/bridge/netfilter/ebtables.c
+++ b/net/bridge/netfilter/ebtables.c
@@ -164,7 +164,7 @@ unsigned int ebt_do_table (unsigned int hook, struct sk_buff *skb,
mtpar.in = tgpar.in = in;
mtpar.out = tgpar.out = out;
mtpar.hotdrop = &hotdrop;
- tgpar.hooknum = hook;
+ mtpar.hooknum = tgpar.hooknum = hook;
read_lock_bh(&table->lock);
private = table->private;
diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c
index 810c0b6..385e4a0 100644
--- a/net/ipv4/netfilter/ip_tables.c
+++ b/net/ipv4/netfilter/ip_tables.c
@@ -335,7 +335,7 @@ ipt_do_table(struct sk_buff *skb,
mtpar.in = tgpar.in = in;
mtpar.out = tgpar.out = out;
mtpar.family = tgpar.family = NFPROTO_IPV4;
- tgpar.hooknum = hook;
+ mtpar.hooknum = tgpar.hooknum = hook;
IP_NF_ASSERT(table->valid_hooks & (1 << hook));
diff --git a/net/ipv6/netfilter/ip6_tables.c b/net/ipv6/netfilter/ip6_tables.c
index 800ae85..d603ae4 100644
--- a/net/ipv6/netfilter/ip6_tables.c
+++ b/net/ipv6/netfilter/ip6_tables.c
@@ -361,7 +361,7 @@ ip6t_do_table(struct sk_buff *skb,
mtpar.in = tgpar.in = in;
mtpar.out = tgpar.out = out;
mtpar.family = tgpar.family = NFPROTO_IPV6;
- tgpar.hooknum = hook;
+ mtpar.hooknum = tgpar.hooknum = hook;
IP_NF_ASSERT(table->valid_hooks & (1 << hook));
--
Evgeniy Polyakov
--
To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
