summaryrefslogtreecommitdiff
path: root/net/ipv4/tcp_output.c
diff options
context:
space:
mode:
Diffstat (limited to 'net/ipv4/tcp_output.c')
-rw-r--r--net/ipv4/tcp_output.c69
1 files changed, 54 insertions, 15 deletions
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index c915dd81c927..2506960189fe 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -708,8 +708,9 @@ static unsigned int tcp_established_options(struct sock *sk, struct sk_buff *skb
min_t(unsigned int, eff_sacks,
(remaining - TCPOLEN_SACK_BASE_ALIGNED) /
TCPOLEN_SACK_PERBLOCK);
- size += TCPOLEN_SACK_BASE_ALIGNED +
- opts->num_sack_blocks * TCPOLEN_SACK_PERBLOCK;
+ if (likely(opts->num_sack_blocks))
+ size += TCPOLEN_SACK_BASE_ALIGNED +
+ opts->num_sack_blocks * TCPOLEN_SACK_PERBLOCK;
}
return size;
@@ -1264,6 +1265,7 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len,
struct tcp_sock *tp = tcp_sk(sk);
struct sk_buff *buff;
int nsize, old_factor;
+ long limit;
int nlen;
u8 flags;
@@ -1274,6 +1276,19 @@ int tcp_fragment(struct sock *sk, struct sk_buff *skb, u32 len,
if (nsize < 0)
nsize = 0;
+ /* tcp_sendmsg() can overshoot sk_wmem_queued by one full size skb.
+ * We need some allowance to not penalize applications setting small
+ * SO_SNDBUF values.
+ * Also allow first and last skb in retransmit queue to be split.
+ */
+ limit = sk->sk_sndbuf + 2 * SKB_TRUESIZE(GSO_MAX_SIZE);
+ if (unlikely((sk->sk_wmem_queued >> 1) > limit &&
+ skb != tcp_rtx_queue_head(sk) &&
+ skb != tcp_rtx_queue_tail(sk))) {
+ NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPWQUEUETOOBIG);
+ return -ENOMEM;
+ }
+
if (skb_unclone(skb, gfp))
return -ENOMEM;
@@ -1442,8 +1457,7 @@ static inline int __tcp_mtu_to_mss(struct sock *sk, int pmtu)
mss_now -= icsk->icsk_ext_hdr_len;
/* Then reserve room for full set of TCP options and 8 bytes of data */
- if (mss_now < 48)
- mss_now = 48;
+ mss_now = max(mss_now, sock_net(sk)->ipv4.sysctl_tcp_min_snd_mss);
return mss_now;
}
@@ -1885,7 +1899,9 @@ static int tso_fragment(struct sock *sk, struct sk_buff *skb, unsigned int len,
* This algorithm is from John Heffner.
*/
static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb,
- bool *is_cwnd_limited, u32 max_segs)
+ bool *is_cwnd_limited,
+ bool *is_rwnd_limited,
+ u32 max_segs)
{
const struct inet_connection_sock *icsk = inet_csk(sk);
u32 age, send_win, cong_win, limit, in_flight;
@@ -1893,9 +1909,6 @@ static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb,
struct sk_buff *head;
int win_divisor;
- if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)
- goto send_now;
-
if (icsk->icsk_ca_state >= TCP_CA_Recovery)
goto send_now;
@@ -1951,10 +1964,27 @@ static bool tcp_tso_should_defer(struct sock *sk, struct sk_buff *skb,
if (age < (tp->srtt_us >> 4))
goto send_now;
- /* Ok, it looks like it is advisable to defer. */
+ /* Ok, it looks like it is advisable to defer.
+ * Three cases are tracked :
+ * 1) We are cwnd-limited
+ * 2) We are rwnd-limited
+ * 3) We are application limited.
+ */
+ if (cong_win < send_win) {
+ if (cong_win <= skb->len) {
+ *is_cwnd_limited = true;
+ return true;
+ }
+ } else {
+ if (send_win <= skb->len) {
+ *is_rwnd_limited = true;
+ return true;
+ }
+ }
- if (cong_win < send_win && cong_win <= skb->len)
- *is_cwnd_limited = true;
+ /* If this packet won't get more data, do not wait. */
+ if (TCP_SKB_CB(skb)->tcp_flags & TCPHDR_FIN)
+ goto send_now;
return true;
@@ -1996,7 +2026,7 @@ static bool tcp_can_coalesce_send_queue_head(struct sock *sk, int len)
if (len <= skb->len)
break;
- if (unlikely(TCP_SKB_CB(skb)->eor))
+ if (unlikely(TCP_SKB_CB(skb)->eor) || tcp_has_tx_tstamp(skb))
return false;
len -= skb->len;
@@ -2119,6 +2149,7 @@ static int tcp_mtu_probe(struct sock *sk)
* we need to propagate it to the new skb.
*/
TCP_SKB_CB(nskb)->eor = TCP_SKB_CB(skb)->eor;
+ tcp_skb_collapse_tstamp(nskb, skb);
tcp_unlink_write_queue(skb, sk);
sk_wmem_free_skb(sk, skb);
} else {
@@ -2328,7 +2359,7 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle,
} else {
if (!push_one &&
tcp_tso_should_defer(sk, skb, &is_cwnd_limited,
- max_segs))
+ &is_rwnd_limited, max_segs))
break;
}
@@ -2349,6 +2380,14 @@ static bool tcp_write_xmit(struct sock *sk, unsigned int mss_now, int nonagle,
if (tcp_small_queue_check(sk, skb, 0))
break;
+ /* Argh, we hit an empty skb(), presumably a thread
+ * is sleeping in sendmsg()/sk_stream_wait_memory().
+ * We do not want to send a pure-ack packet and have
+ * a strange looking rtx queue with empty packet(s).
+ */
+ if (TCP_SKB_CB(skb)->end_seq == TCP_SKB_CB(skb)->seq)
+ break;
+
if (unlikely(tcp_transmit_skb(sk, skb, 1, gfp)))
break;
@@ -2708,7 +2747,7 @@ static bool tcp_collapse_retrans(struct sock *sk, struct sk_buff *skb)
if (next_skb_size <= skb_availroom(skb))
skb_copy_bits(next_skb, 0, skb_put(skb, next_skb_size),
next_skb_size);
- else if (!skb_shift(skb, next_skb, next_skb_size))
+ else if (!tcp_skb_shift(skb, next_skb, 1, next_skb_size))
return false;
}
tcp_highest_sack_replace(sk, next_skb, skb);
@@ -2902,7 +2941,7 @@ int __tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb, int segs)
if (likely(!err)) {
TCP_SKB_CB(skb)->sacked |= TCPCB_EVER_RETRANS;
} else if (err != -EBUSY) {
- NET_INC_STATS(sock_net(sk), LINUX_MIB_TCPRETRANSFAIL);
+ NET_ADD_STATS(sock_net(sk), LINUX_MIB_TCPRETRANSFAIL, segs);
}
return err;
}
generated by cgit v1.2.3 (git 2.25.1) at 2025-12-20 02:52:56 +0000