diff options
| author | Florian Westphal <fw@strlen.de> | 2017-08-30 19:24:58 +0200 |
|---|---|---|
| committer | David S. Miller <davem@davemloft.net> | 2017-08-30 11:20:09 -0700 |
| commit | 31770e34e43d6c8dee129bfee77e56c34e61f0e5 (patch) | |
| tree | a4635a632732b39ef560a8f72eb3a81d04f01605 /include | |
| parent | c1d2b4c3e204e602c97680335d082b8d012d08cd (diff) | |
| download | linux-31770e34e43d6c8dee129bfee77e56c34e61f0e5.tar.bz2 | |
tcp: Revert "tcp: remove header prediction"
This reverts commit 45f119bf936b1f9f546a0b139c5b56f9bb2bdc78.
Eric Dumazet says:
We found at Google a significant regression caused by
45f119bf936b1f9f546a0b139c5b56f9bb2bdc78 tcp: remove header prediction
In typical RPC (TCP_RR), when a TCP socket receives data, we now call
tcp_ack() while we used to not call it.
This touches enough cache lines to cause a slowdown.
so problem does not seem to be HP removal itself but the tcp_ack()
call. Therefore, it might be possible to remove HP after all, provided
one finds a way to elide tcp_ack for most cases.
Reported-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'include')
| -rw-r--r-- | include/linux/tcp.h | 6 | ||||
| -rw-r--r-- | include/net/tcp.h | 23 | ||||
| -rw-r--r-- | include/uapi/linux/snmp.h | 2 |
3 files changed, 31 insertions, 0 deletions
diff --git a/include/linux/tcp.h b/include/linux/tcp.h index 267164a1d559..4aa40ef02d32 100644 --- a/include/linux/tcp.h +++ b/include/linux/tcp.h @@ -148,6 +148,12 @@ struct tcp_sock { u16 gso_segs; /* Max number of segs per GSO packet */ /* + * Header prediction flags + * 0x5?10 << 16 + snd_wnd in net byte order + */ + __be32 pred_flags; + +/* * RFC793 variables by their proper names. This means you can * read the code and the spec side by side (and laugh ...) * See RFC793 and RFC1122. The RFC writes these in capitals. diff --git a/include/net/tcp.h b/include/net/tcp.h index c546d13ffbca..9c3db054e47f 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -634,6 +634,29 @@ static inline u32 __tcp_set_rto(const struct tcp_sock *tp) return usecs_to_jiffies((tp->srtt_us >> 3) + tp->rttvar_us); } +static inline void __tcp_fast_path_on(struct tcp_sock *tp, u32 snd_wnd) +{ + tp->pred_flags = htonl((tp->tcp_header_len << 26) | + ntohl(TCP_FLAG_ACK) | + snd_wnd); +} + +static inline void tcp_fast_path_on(struct tcp_sock *tp) +{ + __tcp_fast_path_on(tp, tp->snd_wnd >> tp->rx_opt.snd_wscale); +} + +static inline void tcp_fast_path_check(struct sock *sk) +{ + struct tcp_sock *tp = tcp_sk(sk); + + if (RB_EMPTY_ROOT(&tp->out_of_order_queue) && + tp->rcv_wnd && + atomic_read(&sk->sk_rmem_alloc) < sk->sk_rcvbuf && + !tp->urg_data) + tcp_fast_path_on(tp); +} + /* Compute the actual rto_min value */ static inline u32 tcp_rto_min(struct sock *sk) { diff --git a/include/uapi/linux/snmp.h b/include/uapi/linux/snmp.h index b3f346fb9fe3..758f12b58541 100644 --- a/include/uapi/linux/snmp.h +++ b/include/uapi/linux/snmp.h @@ -184,7 +184,9 @@ enum LINUX_MIB_DELAYEDACKLOST, /* DelayedACKLost */ LINUX_MIB_LISTENOVERFLOWS, /* ListenOverflows */ LINUX_MIB_LISTENDROPS, /* ListenDrops */ + LINUX_MIB_TCPHPHITS, /* TCPHPHits */ LINUX_MIB_TCPPUREACKS, /* TCPPureAcks */ + LINUX_MIB_TCPHPACKS, /* TCPHPAcks */ LINUX_MIB_TCPRENORECOVERY, /* TCPRenoRecovery */ LINUX_MIB_TCPSACKRECOVERY, /* TCPSackRecovery */ LINUX_MIB_TCPSACKRENEGING, /* TCPSACKReneging */ |