[SK_BUFF]: Convert skb->tail to sk_buff_data_t

So that it is also an offset from skb->head, reduces its size from 8 to 4 bytes
on 64bit architectures, allowing us to combine the 4 bytes hole left by the
layer headers conversion, reducing struct sk_buff size to 256 bytes, i.e. 4
64byte cachelines, and since the sk_buff slab cache is SLAB_HWCACHE_ALIGN...
:-)

Many calculations that previously required that skb->{transport,network,
mac}_header be first converted to a pointer now can be done directly, being
meaningful as offsets or pointers.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Arnaldo Carvalho de Melo
2007-04-19 20:29:13 -07:00
committed by David S. Miller
parent be8bd86321
commit 27a884dc3c
110 changed files with 396 additions and 329 deletions

View File

@@ -1423,7 +1423,7 @@ static struct sk_buff *mld_newpack(struct net_device *dev, int size)
memcpy(skb_put(skb, sizeof(ra)), ra, sizeof(ra));
skb_set_transport_header(skb, skb->tail - skb->data);
skb_set_transport_header(skb, skb_tail_pointer(skb) - skb->data);
skb_put(skb, sizeof(*pmr));
pmr = (struct mld2_report *)skb_transport_header(skb);
pmr->type = ICMPV6_MLD2_REPORT;
@@ -1468,8 +1468,8 @@ static void mld_sendpack(struct sk_buff *skb)
int err;
IP6_INC_STATS(idev, IPSTATS_MIB_OUTREQUESTS);
payload_len = skb->tail - skb_network_header(skb) - sizeof(*pip6);
mldlen = skb->tail - skb_transport_header(skb);
payload_len = (skb->tail - skb->network_header) - sizeof(*pip6);
mldlen = skb->tail - skb->transport_header;
pip6->payload_len = htons(payload_len);
pmr->csum = csum_ipv6_magic(&pip6->saddr, &pip6->daddr, mldlen,