From mboxrd@z Thu Jan 1 00:00:00 1970 From: Willem de Bruijn Subject: [PATCH net-next 6/7] net-timestamp: ENQ timestamp on enqueue to traffic shaping layer Date: Tue, 24 Jun 2014 11:43:51 -0400 Message-ID: <1403624632-17327-7-git-send-email-willemb@google.com> References: <1403624632-17327-1-git-send-email-willemb@google.com> Cc: eric.dumazet@gmail.com, richardcochran@gmail.com, davem@davemloft.net, Willem de Bruijn To: netdev@vger.kernel.org Return-path: Received: from mail-oa0-f74.google.com ([209.85.219.74]:38935 "EHLO mail-oa0-f74.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754488AbaFXPoA (ORCPT ); Tue, 24 Jun 2014 11:44:00 -0400 Received: by mail-oa0-f74.google.com with SMTP id i7so102031oag.1 for ; Tue, 24 Jun 2014 08:44:00 -0700 (PDT) In-Reply-To: <1403624632-17327-1-git-send-email-willemb@google.com> Sender: netdev-owner@vger.kernel.org List-ID: Kernel transmit latency is often incurred in the traffic shaping layer. This patch adds a new timestamp on transmission just before entering traffic shaping. When data travels through multiple devices (bonding, tunneling, ...) each device will export an individual timestamp. Signed-off-by: Willem de Bruijn --- include/linux/skbuff.h | 8 +++++++- include/linux/socket.h | 3 ++- net/core/dev.c | 3 +++ net/socket.c | 3 +++ 4 files changed, 15 insertions(+), 2 deletions(-) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index ee86654..b6c2926 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -260,9 +260,13 @@ enum { SKBTX_SHARED_FRAG = 1 << 5, SKBTX_ACK_TSTAMP = 1 << 6, + + SKBTX_ENQ_TSTAMP = 1 << 7, }; -#define SKBTX_ANY_SW_TSTAMP (SKBTX_SW_TSTAMP | SKBTX_ACK_TSTAMP) +#define SKBTX_ANY_SW_TSTAMP (SKBTX_SW_TSTAMP | \ + SKBTX_ACK_TSTAMP | \ + SKBTX_ENQ_TSTAMP) /* * The callback notifies userspace to release buffers when skb DMA is done in @@ -2757,6 +2761,8 @@ static inline u8 skbflags_tx_tstamp(int flags) tx_flags |= SKBTX_SW_TSTAMP; if (flags & MSG_TSTAMP_ACK) tx_flags |= SKBTX_ACK_TSTAMP; + if (flags & MSG_TSTAMP_ENQ) + tx_flags |= SKBTX_ENQ_TSTAMP; } return tx_flags; diff --git a/include/linux/socket.h b/include/linux/socket.h index 68d5f48..6d21582 100644 --- a/include/linux/socket.h +++ b/include/linux/socket.h @@ -255,7 +255,8 @@ struct ucred { #define MSG_SENDPAGE_NOTLAST 0x20000 /* sendpage() internal : not the last page */ #define MSG_TSTAMP 0x100000 #define MSG_TSTAMP_ACK 0x200000 -#define MSG_TSTAMP_MASK (MSG_TSTAMP | MSG_TSTAMP_ACK) +#define MSG_TSTAMP_ENQ 0x400000 +#define MSG_TSTAMP_MASK (MSG_TSTAMP | MSG_TSTAMP_ACK | MSG_TSTAMP_ENQ) #define MSG_EOF MSG_FIN #define MSG_FASTOPEN 0x20000000 /* Send data in TCP SYN */ diff --git a/net/core/dev.c b/net/core/dev.c index a04b12f..8df522b 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2859,6 +2859,9 @@ static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv) skb_reset_mac_header(skb); + if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_ENQ_TSTAMP)) + __skb_tstamp_tx(skb, NULL, skb->sk, SKBTX_ENQ_TSTAMP); + /* Disable soft irqs for various locks below. Also * stops preemption for RCU. */ diff --git a/net/socket.c b/net/socket.c index b71001b..9b8deaf 100644 --- a/net/socket.c +++ b/net/socket.c @@ -736,6 +736,9 @@ void __sock_recv_timestamp(struct msghdr *msg, struct sock *sk, case SKBTX_ACK_TSTAMP: tss.ts_type = SCM_TSTAMP_ACK; break; + case SKBTX_ENQ_TSTAMP: + tss.ts_type = SCM_TSTAMP_ENQ; + break; } } if (shhwtstamps) { -- 2.0.0.526.g5318336