diff options
author | Eric Dumazet <edumazet@google.com> | 2019-03-22 18:56:39 +0300 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2019-03-24 04:57:38 +0300 |
commit | 472c2e07eef045145bc1493cc94a01c87140780a (patch) | |
tree | e1980ba54472857684241be387d8f25e561be901 /include | |
parent | dc05360fee660a9dbe59824b3f7896534210432b (diff) | |
download | linux-472c2e07eef045145bc1493cc94a01c87140780a.tar.xz |
tcp: add one skb cache for tx
On hosts with a lot of cores, RPC workloads suffer from heavy contention on slab spinlocks.
20.69% [kernel] [k] queued_spin_lock_slowpath
5.64% [kernel] [k] _raw_spin_lock
3.83% [kernel] [k] syscall_return_via_sysret
3.48% [kernel] [k] __entry_text_start
1.76% [kernel] [k] __netif_receive_skb_core
1.64% [kernel] [k] __fget
For each sendmsg(), we allocate one skb, and free it at the time ACK packet comes.
In many cases, ACK packets are handled by another cpus, and this unfortunately
incurs heavy costs for slab layer.
This patch uses an extra pointer in socket structure, so that we try to reuse
the same skb and avoid these expensive costs.
We cache at most one skb per socket so this should be safe as far as
memory pressure is concerned.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'include')
-rw-r--r-- | include/net/sock.h | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/include/net/sock.h b/include/net/sock.h index fecdf639225c..314c47a8f5d1 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -414,6 +414,7 @@ struct sock { struct sk_buff *sk_send_head; struct rb_root tcp_rtx_queue; }; + struct sk_buff *sk_tx_skb_cache; struct sk_buff_head sk_write_queue; __s32 sk_peek_off; int sk_write_pending; @@ -1463,6 +1464,10 @@ static inline void sk_mem_uncharge(struct sock *sk, int size) static inline void sk_wmem_free_skb(struct sock *sk, struct sk_buff *skb) { + if (!sk->sk_tx_skb_cache) { + sk->sk_tx_skb_cache = skb; + return; + } sock_set_flag(sk, SOCK_QUEUE_SHRUNK); sk->sk_wmem_queued -= skb->truesize; sk_mem_uncharge(sk, skb->truesize); |