All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH net-next] tcp: suppress too verbose messages in tcp_send_ack()
@ 2015-11-25 21:50 Eric Dumazet
  2015-11-25 22:08 ` Aaron Conole
  2015-11-30 16:06 ` David Miller
  0 siblings, 2 replies; 10+ messages in thread
From: Eric Dumazet @ 2015-11-25 21:50 UTC (permalink / raw)
  To: David Miller; +Cc: netdev

From: Eric Dumazet <edumazet@google.com>

If tcp_send_ack() can not allocate skb, we properly handle this
and setup a timer to try later.

Use __GFP_NOWARN to avoid polluting syslog in the case host is
under memory pressure, so that pertinent messages are not lost under
a flood of useless information.

sk_gfp_atomic() can use its gfp_mask argument (all callers currently
were using GFP_ATOMIC before this patch)

Note that when tcp_transmit_skb() is called with clone_it set to false,
we do not attempt memory allocations, so can pass a 0 gfp_mask, which
most compilers can emit faster than a non zero value.

Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 include/net/sock.h    |    2 +-
 net/ipv4/tcp_output.c |   12 +++++++-----
 2 files changed, 8 insertions(+), 6 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index 7f89e4ba18d1..ead514332ae8 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -776,7 +776,7 @@ static inline int sk_memalloc_socks(void)
 
 static inline gfp_t sk_gfp_atomic(const struct sock *sk, gfp_t gfp_mask)
 {
-	return GFP_ATOMIC | (sk->sk_allocation & __GFP_MEMALLOC);
+	return gfp_mask | (sk->sk_allocation & __GFP_MEMALLOC);
 }
 
 static inline void sk_acceptq_removed(struct sock *sk)
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index cb7ca569052c..0a1d4f6ab52f 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -3352,8 +3352,9 @@ void tcp_send_ack(struct sock *sk)
 	 * tcp_transmit_skb() will set the ownership to this
 	 * sock.
 	 */
-	buff = alloc_skb(MAX_TCP_HEADER, sk_gfp_atomic(sk, GFP_ATOMIC));
-	if (!buff) {
+	buff = alloc_skb(MAX_TCP_HEADER,
+			 sk_gfp_atomic(sk, GFP_ATOMIC | __GFP_NOWARN));
+	if (unlikely(!buff)) {
 		inet_csk_schedule_ack(sk);
 		inet_csk(sk)->icsk_ack.ato = TCP_ATO_MIN;
 		inet_csk_reset_xmit_timer(sk, ICSK_TIME_DACK,
@@ -3375,7 +3376,7 @@ void tcp_send_ack(struct sock *sk)
 
 	/* Send it off, this clears delayed acks for us. */
 	skb_mstamp_get(&buff->skb_mstamp);
-	tcp_transmit_skb(sk, buff, 0, sk_gfp_atomic(sk, GFP_ATOMIC));
+	tcp_transmit_skb(sk, buff, 0, (__force gfp_t)0);
 }
 EXPORT_SYMBOL_GPL(tcp_send_ack);
 
@@ -3396,7 +3397,8 @@ static int tcp_xmit_probe_skb(struct sock *sk, int urgent, int mib)
 	struct sk_buff *skb;
 
 	/* We don't queue it, tcp_transmit_skb() sets ownership. */
-	skb = alloc_skb(MAX_TCP_HEADER, sk_gfp_atomic(sk, GFP_ATOMIC));
+	skb = alloc_skb(MAX_TCP_HEADER,
+			sk_gfp_atomic(sk, GFP_ATOMIC | __GFP_NOWARN));
 	if (!skb)
 		return -1;
 
@@ -3409,7 +3411,7 @@ static int tcp_xmit_probe_skb(struct sock *sk, int urgent, int mib)
 	tcp_init_nondata_skb(skb, tp->snd_una - !urgent, TCPHDR_ACK);
 	skb_mstamp_get(&skb->skb_mstamp);
 	NET_INC_STATS(sock_net(sk), mib);
-	return tcp_transmit_skb(sk, skb, 0, GFP_ATOMIC);
+	return tcp_transmit_skb(sk, skb, 0, (__force gfp_t)0);
 }
 
 void tcp_send_window_probe(struct sock *sk)

^ permalink raw reply related	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2015-12-03  4:44 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-11-25 21:50 [PATCH net-next] tcp: suppress too verbose messages in tcp_send_ack() Eric Dumazet
2015-11-25 22:08 ` Aaron Conole
2015-11-25 22:32   ` Eric Dumazet
2015-11-25 22:35     ` Eric Dumazet
2015-11-26  3:17       ` Aaron Conole
2015-11-26  3:29         ` Eric Dumazet
2015-11-28  2:21           ` Aaron Conole
2015-11-30 16:06 ` David Miller
2015-11-30 16:57   ` [PATCH v2 " Eric Dumazet
2015-12-03  4:44     ` David Miller

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.