>16:09:06.947418 localhost.1048 > localhost.telnet: S 2378452431:2378452431(0) win 31072 <mss 3884,sackOK,timestamp 567981 0,nop,wscale 0> (DF)
> ^^^^^ ^^^^
>16:09:06.947478 localhost.telnet > localhost.1048: S 2374855488:2374855488(0) ack 2378452432 win 31072 <mss 3884,sackOK,timestamp 567981 567981,nop,wscale 0> (DF)
>16:09:06.947509 localhost.1048 > localhost.telnet: . ack 1 win 31072 <nop,nop,timestamp 567981 567981> (DF)
> ^^^^^
>16:09:06.993209 localhost.telnet > localhost.1048: P 1:4(3) ack 1 win 31072 <nop,nop,timestamp 567985 567981> (DF)
I was wrong, until here it's fine because at this time the rcvbuf is still
the default. But _here_ I set the rcvbuf to 1008.
>16:09:07.005029 localhost.1048 > localhost.telnet: . ack 4 win 31069 <nop,nop,timestamp 567987 567985> (DF)
^^^^^
And here the window get not updated correctly taking care of the rcvbuf
change. I think it's due this code in tcp.h:
extern __inline__ u16 tcp_select_window(struct sock *sk)
{
struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);
u32 cur_win = tcp_receive_window(tp);
u32 new_win = __tcp_select_window(sk);
/* Never shrink the offered window */
if(new_win < cur_win) {
^^^^^^^^^^^^^^^^^
/* Danger Will Robinson!
* Don't update rcv_wup/rcv_wnd here or else
* we will not be able to advertise a zero
* window in time. --DaveM
*/
new_win = cur_win;
^^^^^^^^^^^^^^^^^ so the next advertised window become the
current window
} else {
tp->rcv_wnd = new_win;
tp->rcv_wup = tp->rcv_nxt;
}
/* RFC1323 scaling applied */
return new_win >> tp->rcv_wscale;
}
So what's happening here? It seems that it's impossible to decrease the
receiver window under the current value.
Andi, are you going to develop your `avoid to set a rcvbuf lower than the
minimal truesize' patch? You must also include in it the `check that the
rcvbuf is not too low at MTU size increase', I really don't like a lazy
not clean approch that could cause a TCP connection to deadlock upon MTU
change.
And are we checking that the incoming skb data size is not > of our
advertised mss before queueing it in the receive queue? If so my fix looks
still safe to me. I also changed tcp_queue to wakeup the waiter only if we
received in order data. Note I didn't had the time to try out this last
change, it's just something I thought about and I implemented.
Here my latest snapshot against 2.2.2.
Index: tcp_input.c
===================================================================
RCS file: /var/cvs/linux/net/ipv4/tcp_input.c,v
retrieving revision 1.1.1.6
retrieving revision 1.1.2.20
diff -u -r1.1.1.6 -r1.1.2.20
--- tcp_input.c 1999/02/23 16:48:26 1.1.1.6
+++ tcp_input.c 1999/03/06 18:47:13 1.1.2.20
@@ -55,6 +55,15 @@
* work without delayed acks.
* Andi Kleen: Process packets with PSH set in the
* fast path.
+ * Andrea Arcangeli: TCP_NODELAY disable delacks.
+ * Andrea Arcangeli: Delay also the first ack to be able to
+ * join the ack in the first data packet.
+ * Andrea Arcangeli: Always queue a packet from the other
+ * end if all receive queues are empty,
+ * this avoid the receiver to deadlock
+ * if rcvbuf < packet truesize.
+ * Andrea Arcangeli: Don't wakeup waiters if we get out of
+ * order packets.
*/
#include <linux/config.h>
@@ -99,12 +108,11 @@
if(tp->ato == 0) {
tp->lrcvtime = jiffies;
- /* Help sender leave slow start quickly,
- * and also makes sure we do not take this
- * branch ever again for this connection.
+ /*
+ * Don't enter quickack mode to be able to join the first
+ * ack with a data packet. -arca
*/
- tp->ato = 1;
- tcp_enter_quickack_mode(tp);
+ tp->ato = (HZ+49)/50;
} else {
int m = jiffies - tp->lrcvtime;
@@ -135,9 +143,9 @@
*/
if(th->psh && (skb->len < (tp->mss_cache >> 1))) {
/* Preserve the quickack state. */
- if((tp->ato & 0x7fffffff) > HZ/50)
+ if((tp->ato & 0x7fffffff) > (HZ+49)/50)
tp->ato = ((tp->ato & 0x80000000) |
- (HZ/50));
+ ((HZ+49)/50));
}
}
@@ -241,7 +249,7 @@
extern __inline__ int tcp_paws_discard(struct tcp_opt *tp, struct tcphdr *th, unsigned len)
{
/* ts_recent must be younger than 24 days */
- return (((jiffies - tp->ts_recent_stamp) >= PAWS_24DAYS) ||
+ return (((s32)(jiffies - tp->ts_recent_stamp) >= PAWS_24DAYS) ||
(((s32)(tp->rcv_tsval-tp->ts_recent) < 0) &&
/* Sorry, PAWS as specified is broken wrt. pure-ACKs -DaveM */
(len != (th->doff * 4))));
@@ -1354,7 +1362,7 @@
}
}
-static void tcp_data_queue(struct sock *sk, struct sk_buff *skb)
+static int tcp_data_queue(struct sock *sk, struct sk_buff *skb)
{
struct sk_buff *skb1;
struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);
@@ -1384,7 +1392,7 @@
tp->pred_flags = htonl(((tp->tcp_header_len >> 2) << 28) |
(0x10 << 16) |
tp->snd_wnd);
- return;
+ return 0;
}
/* An old packet, either a retransmit or some packet got lost. */
@@ -1393,7 +1401,7 @@
SOCK_DEBUG(sk, "retransmit received: seq %X\n", TCP_SKB_CB(skb)->seq);
tcp_enter_quickack_mode(tp);
kfree_skb(skb);
- return;
+ return 1;
}
if (before(TCP_SKB_CB(skb)->seq, tp->rcv_nxt)) {
@@ -1415,7 +1423,7 @@
SOCK_DEBUG(sk, "out of order segment: rcv_next %X seq %X - %X\n",
tp->rcv_nxt, TCP_SKB_CB(skb)->seq, TCP_SKB_CB(skb)->end_seq);
- if (skb_peek(&tp->out_of_order_queue) == NULL) {
+ if (!skb_queue_len(&tp->out_of_order_queue)) {
/* Initial out of order segment, build 1 SACK. */
if(tp->sack_ok) {
tp->num_sacks = 1;
@@ -1458,6 +1466,7 @@
}
}
}
+ return 1;
}
@@ -1471,6 +1480,7 @@
{
struct tcphdr *th;
struct tcp_opt *tp = &(sk->tp_pinfo.af_tcp);
+ int ret;
th = skb->h.th;
skb_pull(skb, th->doff*4);
@@ -1484,8 +1494,11 @@
* Make sure to do this before moving snd_nxt, otherwise
* data might be acked for that we don't have enough room.
*/
- if (atomic_read(&sk->rmem_alloc) > sk->rcvbuf) {
- if (prune_queue(sk) < 0) {
+ if (atomic_read(&sk->rmem_alloc) >= sk->rcvbuf &&
+ (skb_queue_len(&tp->out_of_order_queue) ||
+ skb_queue_len(&tp->receive_queue))) {
+ if (prune_queue(sk) < 0 &&
+ skb_queue_len(&tp->receive_queue)) {
/* Still not enough room. That can happen when
* skb->true_size differs significantly from skb->len.
*/
@@ -1493,7 +1506,7 @@
}
}
- tcp_data_queue(sk, skb);
+ ret = tcp_data_queue(sk, skb);
if (before(tp->rcv_nxt, tp->copied_seq)) {
printk(KERN_DEBUG "*** tcp.c:tcp_data bug acked < copied\n");
@@ -1503,7 +1516,7 @@
/* Above, tcp_data_queue() increments delayed_acks appropriately.
* Now tell the user we may have some data.
*/
- if (!sk->dead) {
+ if (!sk->dead && !ret) {
SOCK_DEBUG(sk, "Data wakeup.\n");
sk->data_ready(sk,0);
}
@@ -1580,7 +1593,9 @@
/* We entered "quick ACK" mode or... */
tcp_in_quickack_mode(tp) ||
/* We have out of order data */
- (skb_peek(&tp->out_of_order_queue) != NULL)) {
+ skb_queue_len(&tp->out_of_order_queue) ||
+ /* TCP_NODELAY is set */
+ sk->nonagle == 1) {
/* Then ack it now */
tcp_send_ack(sk);
} else {
@@ -1821,7 +1836,8 @@
goto discard;
}
} else if (TCP_SKB_CB(skb)->ack_seq == tp->snd_una &&
- atomic_read(&sk->rmem_alloc) <= sk->rcvbuf) {
+ (atomic_read(&sk->rmem_alloc) <= sk->rcvbuf ||
+ !skb_queue_len(&sk->receive_queue))) {
/* Bulk data transfer: receiver */
__skb_pull(skb,th->doff*4);
Index: tcp_output.c
===================================================================
RCS file: /var/cvs/linux/net/ipv4/tcp_output.c,v
retrieving revision 1.1.1.5
diff -u -r1.1.1.5 tcp_output.c
--- tcp_output.c 1999/02/23 16:48:26 1.1.1.5
+++ tcp_output.c 1999/03/02 16:53:50
@@ -580,6 +580,19 @@
if(tp->af_specific->rebuild_header(sk))
return 1; /* Routing failure or similar. */
+ /* Solaris sucks. */
+ if(skb->len > 0 &&
+ (TCP_SKB_CB(skb)->flags & TCPCB_FLAG_FIN) &&
+ tp->snd_una == (TCP_SKB_CB(skb)->end_seq - 1)) {
+#if 1
+ printk("TCP: Doing Solaris hack for [%p:%08x:%04x:%08x]\n",
+ skb, sk->daddr, sk->dport, tp->snd_una);
+#endif
+ TCP_SKB_CB(skb)->seq = TCP_SKB_CB(skb)->end_seq - 1;
+ skb_trim(skb, 0);
+ skb->csum = 0;
+ }
+
/* Ok, we're gonna send it out, update state. */
TCP_SKB_CB(skb)->sacked |= TCPCB_SACKED_RETRANS;
tp->retrans_out++;
@@ -973,6 +986,8 @@
timeout = tp->ato;
if (timeout > max_timeout)
timeout = max_timeout;
+ if (timeout < (HZ+49)/50)
+ timeout = (HZ+49)/50;
timeout += jiffies;
/* Use new timeout only if there wasn't a older one earlier. */
@@ -980,7 +995,7 @@
tp->delack_timer.expires = timeout;
add_timer(&tp->delack_timer);
} else {
- if (timeout < tp->delack_timer.expires)
+ if (time_before(timeout, tp->delack_timer.expires))
mod_timer(&tp->delack_timer, timeout);
}
}
Comments?
Andrea Arcangeli
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/