[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <E1M9B60-0001TO-4D@gondolin.me.apana.org.au>
Date: Wed, 27 May 2009 14:50:32 +1000
From: Herbert Xu <herbert@...dor.apana.org.au>
To: "David S. Miller" <davem@...emloft.net>, netdev@...r.kernel.org
Subject: [PATCH 12/14] gro: Nasty optimisations for page frags in skb_gro_receive
gro: Nasty optimisations for page frags in skb_gro_receive
This patch reverses the direction of the frags array copy in
skb_gro_receive in order simplify the loop conditional. It
also avoids touching the first element of the original frags
array.
Signed-off-by: Herbert Xu <herbert@...dor.apana.org.au>
---
net/core/skbuff.c | 25 +++++++++++++++----------
1 file changed, 15 insertions(+), 10 deletions(-)
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 168e949..19afb18 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -2676,21 +2676,26 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb)
goto merge;
else if (headlen <= offset) {
skb_frag_t *frag;
- int i;
+ skb_frag_t *frag2;
+ int i = skb_shinfo(skb)->nr_frags;
+ int nr_frags = skb_shinfo(p)->nr_frags + i;
+
+ offset -= headlen;
- if (skb_shinfo(p)->nr_frags + skb_shinfo(skb)->nr_frags >
- MAX_SKB_FRAGS)
+ if (nr_frags > MAX_SKB_FRAGS)
return -E2BIG;
- skb_shinfo(skb)->frags[0].page_offset += offset - headlen;
- skb_shinfo(skb)->frags[0].size -= offset - headlen;
+ skb_shinfo(p)->nr_frags = nr_frags;
+ skb_shinfo(skb)->nr_frags = 0;
- frag = skb_shinfo(p)->frags + skb_shinfo(p)->nr_frags;
- for (i = 0; i < skb_shinfo(skb)->nr_frags; i++)
- *frag++ = skb_shinfo(skb)->frags[i];
+ frag = skb_shinfo(p)->frags + nr_frags;
+ frag2 = skb_shinfo(skb)->frags + i;
+ do {
+ *--frag = *--frag2;
+ } while (--i);
- skb_shinfo(p)->nr_frags += skb_shinfo(skb)->nr_frags;
- skb_shinfo(skb)->nr_frags = 0;
+ frag->page_offset += offset;
+ frag->size -= offset;
skb->truesize -= skb->data_len;
skb->len -= skb->data_len;
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists