We can reduce pressure on dst entry refcount that slowdown UDP transmit
path on SMP machines. This pressure is visible on RTP servers when
delivering content to mediagateways, especially big ones, handling
thousand of streams. Several cpus send UDP frames to the same
destination, hence use the same dst entry.
This patch makes ip_append_data() eventually steal the refcount its
callers had to take on the dst entry.
This doesnt avoid all refcounting, but still gives speedups on SMP,
on UDP/RAW transmit path
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
int odd, struct sk_buff *skb),
void *from, int len, int protolen,
struct ipcm_cookie *ipc,
int odd, struct sk_buff *skb),
void *from, int len, int protolen,
struct ipcm_cookie *ipc,
unsigned int flags);
extern int ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb);
extern ssize_t ip_append_page(struct sock *sk, struct page *page,
unsigned int flags);
extern int ip_generic_getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb);
extern ssize_t ip_append_page(struct sock *sk, struct page *page,
}
static void icmp_push_reply(struct icmp_bxm *icmp_param,
}
static void icmp_push_reply(struct icmp_bxm *icmp_param,
- struct ipcm_cookie *ipc, struct rtable *rt)
+ struct ipcm_cookie *ipc, struct rtable **rt)
{
struct sock *sk;
struct sk_buff *skb;
{
struct sock *sk;
struct sk_buff *skb;
- sk = icmp_sk(dev_net(rt->u.dst.dev));
+ sk = icmp_sk(dev_net((*rt)->u.dst.dev));
if (ip_append_data(sk, icmp_glue_bits, icmp_param,
icmp_param->data_len+icmp_param->head_len,
icmp_param->head_len,
if (ip_append_data(sk, icmp_glue_bits, icmp_param,
icmp_param->data_len+icmp_param->head_len,
icmp_param->head_len,
}
if (icmpv4_xrlim_allow(net, rt, icmp_param->data.icmph.type,
icmp_param->data.icmph.code))
}
if (icmpv4_xrlim_allow(net, rt, icmp_param->data.icmph.type,
icmp_param->data.icmph.code))
- icmp_push_reply(icmp_param, &ipc, rt);
+ icmp_push_reply(icmp_param, &ipc, &rt);
ip_rt_put(rt);
out_unlock:
icmp_xmit_unlock(sk);
ip_rt_put(rt);
out_unlock:
icmp_xmit_unlock(sk);
icmp_param.data_len = room;
icmp_param.head_len = sizeof(struct icmphdr);
icmp_param.data_len = room;
icmp_param.head_len = sizeof(struct icmphdr);
- icmp_push_reply(&icmp_param, &ipc, rt);
+ icmp_push_reply(&icmp_param, &ipc, &rt);
ende:
ip_rt_put(rt);
out_unlock:
ende:
ip_rt_put(rt);
out_unlock:
int getfrag(void *from, char *to, int offset, int len,
int odd, struct sk_buff *skb),
void *from, int length, int transhdrlen,
int getfrag(void *from, char *to, int offset, int len,
int odd, struct sk_buff *skb),
void *from, int length, int transhdrlen,
- struct ipcm_cookie *ipc, struct rtable *rt,
+ struct ipcm_cookie *ipc, struct rtable **rtp,
unsigned int flags)
{
struct inet_sock *inet = inet_sk(sk);
unsigned int flags)
{
struct inet_sock *inet = inet_sk(sk);
int offset = 0;
unsigned int maxfraglen, fragheaderlen;
int csummode = CHECKSUM_NONE;
int offset = 0;
unsigned int maxfraglen, fragheaderlen;
int csummode = CHECKSUM_NONE;
if (flags&MSG_PROBE)
return 0;
if (flags&MSG_PROBE)
return 0;
inet->cork.flags |= IPCORK_OPT;
inet->cork.addr = ipc->addr;
}
inet->cork.flags |= IPCORK_OPT;
inet->cork.addr = ipc->addr;
}
+ rt = *rtp;
+ /*
+ * We steal reference to this route, caller should not release it
+ */
+ *rtp = NULL;
inet->cork.fragsize = mtu = inet->pmtudisc == IP_PMTUDISC_PROBE ?
rt->u.dst.dev->mtu :
dst_mtu(rt->u.dst.path);
inet->cork.fragsize = mtu = inet->pmtudisc == IP_PMTUDISC_PROBE ?
rt->u.dst.dev->mtu :
dst_mtu(rt->u.dst.path);
sk->sk_protocol = ip_hdr(skb)->protocol;
sk->sk_bound_dev_if = arg->bound_dev_if;
ip_append_data(sk, ip_reply_glue_bits, arg->iov->iov_base, len, 0,
sk->sk_protocol = ip_hdr(skb)->protocol;
sk->sk_bound_dev_if = arg->bound_dev_if;
ip_append_data(sk, ip_reply_glue_bits, arg->iov->iov_base, len, 0,
- &ipc, rt, MSG_DONTWAIT);
+ &ipc, &rt, MSG_DONTWAIT);
if ((skb = skb_peek(&sk->sk_write_queue)) != NULL) {
if (arg->csumoffset >= 0)
*((__sum16 *)skb_transport_header(skb) +
if ((skb = skb_peek(&sk->sk_write_queue)) != NULL) {
if (arg->csumoffset >= 0)
*((__sum16 *)skb_transport_header(skb) +
ipc.addr = rt->rt_dst;
lock_sock(sk);
err = ip_append_data(sk, ip_generic_getfrag, msg->msg_iov, len, 0,
ipc.addr = rt->rt_dst;
lock_sock(sk);
err = ip_append_data(sk, ip_generic_getfrag, msg->msg_iov, len, 0,
- &ipc, rt, msg->msg_flags);
+ &ipc, &rt, msg->msg_flags);
if (err)
ip_flush_pending_frames(sk);
else if (!(msg->msg_flags & MSG_MORE))
if (err)
ip_flush_pending_frames(sk);
else if (!(msg->msg_flags & MSG_MORE))
up->len += ulen;
getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag;
err = ip_append_data(sk, getfrag, msg->msg_iov, ulen,
up->len += ulen;
getfrag = is_udplite ? udplite_getfrag : ip_generic_getfrag;
err = ip_append_data(sk, getfrag, msg->msg_iov, ulen,
- sizeof(struct udphdr), &ipc, rt,
+ sizeof(struct udphdr), &ipc, &rt,
corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
if (err)
udp_flush_pending_frames(sk);
corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
if (err)
udp_flush_pending_frames(sk);