diff options
author | Soheil Hassas Yeganeh <soheil@google.com> | 2016-04-03 06:08:12 +0300 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2016-04-04 22:50:30 +0300 |
commit | c14ac9451c34832554db33386a4393be8bba3a7b (patch) | |
tree | 99f4d7c46d01732fcbfdf8de89e1d9846d56c3b3 /net | |
parent | ad1e46a837163a3e7160a1250825bcfafd2e714b (diff) | |
download | linux-c14ac9451c34832554db33386a4393be8bba3a7b.tar.xz |
sock: enable timestamping using control messages
Currently, SOL_TIMESTAMPING can only be enabled using setsockopt.
This is very costly when users want to sample writes to gather
tx timestamps.
Add support for enabling SO_TIMESTAMPING via control messages by
using tsflags added in `struct sockcm_cookie` (added in the previous
patches in this series) to set the tx_flags of the last skb created in
a sendmsg. With this patch, the timestamp recording bits in tx_flags
of the skbuff is overridden if SO_TIMESTAMPING is passed in a cmsg.
Please note that this is only effective for overriding the recording
timestamps flags. Users should enable timestamp reporting (e.g.,
SOF_TIMESTAMPING_SOFTWARE | SOF_TIMESTAMPING_OPT_ID) using
socket options and then should ask for SOF_TIMESTAMPING_TX_*
using control messages per sendmsg to sample timestamps for each
write.
Signed-off-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 'net')
-rw-r--r-- | net/can/raw.c | 2 | ||||
-rw-r--r-- | net/ipv4/ping.c | 5 | ||||
-rw-r--r-- | net/ipv4/raw.c | 11 | ||||
-rw-r--r-- | net/ipv4/tcp.c | 20 | ||||
-rw-r--r-- | net/ipv4/udp.c | 7 | ||||
-rw-r--r-- | net/ipv6/icmp.c | 6 | ||||
-rw-r--r-- | net/ipv6/ip6_output.c | 15 | ||||
-rw-r--r-- | net/ipv6/ping.c | 3 | ||||
-rw-r--r-- | net/ipv6/raw.c | 5 | ||||
-rw-r--r-- | net/ipv6/udp.c | 7 | ||||
-rw-r--r-- | net/l2tp/l2tp_ip6.c | 2 | ||||
-rw-r--r-- | net/packet/af_packet.c | 30 | ||||
-rw-r--r-- | net/socket.c | 10 |
13 files changed, 81 insertions, 42 deletions
diff --git a/net/can/raw.c b/net/can/raw.c index 2e67b1423cd3..972c187d40ab 100644 --- a/net/can/raw.c +++ b/net/can/raw.c @@ -755,7 +755,7 @@ static int raw_sendmsg(struct socket *sock, struct msghdr *msg, size_t size) if (err < 0) goto free_skb; - sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags); + sock_tx_timestamp(sk, sk->sk_tsflags, &skb_shinfo(skb)->tx_flags); skb->dev = dev; skb->sk = sk; diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c index 670639bf9f7e..66ddcb60519a 100644 --- a/net/ipv4/ping.c +++ b/net/ipv4/ping.c @@ -737,6 +737,7 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) /* no remote port */ } + ipc.sockc.tsflags = sk->sk_tsflags; ipc.addr = inet->inet_saddr; ipc.opt = NULL; ipc.oif = sk->sk_bound_dev_if; @@ -744,8 +745,6 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) ipc.ttl = 0; ipc.tos = -1; - sock_tx_timestamp(sk, &ipc.tx_flags); - if (msg->msg_controllen) { err = ip_cmsg_send(sk, msg, &ipc, false); if (unlikely(err)) { @@ -768,6 +767,8 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) rcu_read_unlock(); } + sock_tx_timestamp(sk, ipc.sockc.tsflags, &ipc.tx_flags); + saddr = ipc.addr; ipc.addr = faddr = daddr; diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 088ce665fc7b..438f50c1a676 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -339,8 +339,8 @@ int raw_rcv(struct sock *sk, struct sk_buff *skb) static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4, struct msghdr *msg, size_t length, - struct rtable **rtp, - unsigned int flags) + struct rtable **rtp, unsigned int flags, + const struct sockcm_cookie *sockc) { struct inet_sock *inet = inet_sk(sk); struct net *net = sock_net(sk); @@ -379,7 +379,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4, skb->ip_summed = CHECKSUM_NONE; - sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags); + sock_tx_timestamp(sk, sockc->tsflags, &skb_shinfo(skb)->tx_flags); skb->transport_header = skb->network_header; err = -EFAULT; @@ -540,6 +540,7 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) daddr = inet->inet_daddr; } + ipc.sockc.tsflags = sk->sk_tsflags; ipc.addr = inet->inet_saddr; ipc.opt = NULL; ipc.tx_flags = 0; @@ -638,10 +639,10 @@ back_from_confirm: if (inet->hdrincl) err = raw_send_hdrinc(sk, &fl4, msg, len, - &rt, msg->msg_flags); + &rt, msg->msg_flags, &ipc.sockc); else { - sock_tx_timestamp(sk, &ipc.tx_flags); + sock_tx_timestamp(sk, ipc.sockc.tsflags, &ipc.tx_flags); if (!ipc.addr) ipc.addr = fl4.daddr; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index ce3c9eb901a0..4d73858991af 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -428,13 +428,13 @@ void tcp_init_sock(struct sock *sk) } EXPORT_SYMBOL(tcp_init_sock); -static void tcp_tx_timestamp(struct sock *sk, struct sk_buff *skb) +static void tcp_tx_timestamp(struct sock *sk, u16 tsflags, struct sk_buff *skb) { - if (sk->sk_tsflags) { + if (sk->sk_tsflags || tsflags) { struct skb_shared_info *shinfo = skb_shinfo(skb); struct tcp_skb_cb *tcb = TCP_SKB_CB(skb); - sock_tx_timestamp(sk, &shinfo->tx_flags); + sock_tx_timestamp(sk, tsflags, &shinfo->tx_flags); if (shinfo->tx_flags & SKBTX_ANY_TSTAMP) shinfo->tskey = TCP_SKB_CB(skb)->seq + skb->len - 1; tcb->txstamp_ack = !!(shinfo->tx_flags & SKBTX_ACK_TSTAMP); @@ -959,7 +959,7 @@ new_segment: offset += copy; size -= copy; if (!size) { - tcp_tx_timestamp(sk, skb); + tcp_tx_timestamp(sk, sk->sk_tsflags, skb); goto out; } @@ -1079,6 +1079,7 @@ int tcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) { struct tcp_sock *tp = tcp_sk(sk); struct sk_buff *skb; + struct sockcm_cookie sockc; int flags, err, copied = 0; int mss_now = 0, size_goal, copied_syn = 0; bool sg; @@ -1121,6 +1122,15 @@ int tcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size) /* 'common' sending to sendq */ } + sockc.tsflags = sk->sk_tsflags; + if (msg->msg_controllen) { + err = sock_cmsg_send(sk, msg, &sockc); + if (unlikely(err)) { + err = -EINVAL; + goto out_err; + } + } + /* This should be in poll */ sk_clear_bit(SOCKWQ_ASYNC_NOSPACE, sk); @@ -1239,7 +1249,7 @@ new_segment: copied += copy; if (!msg_data_left(msg)) { - tcp_tx_timestamp(sk, skb); + tcp_tx_timestamp(sk, sockc.tsflags, skb); goto out; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index bccb4e11047a..45ff590661f4 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1027,12 +1027,11 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) */ connected = 1; } - ipc.addr = inet->inet_saddr; + ipc.sockc.tsflags = sk->sk_tsflags; + ipc.addr = inet->inet_saddr; ipc.oif = sk->sk_bound_dev_if; - sock_tx_timestamp(sk, &ipc.tx_flags); - if (msg->msg_controllen) { err = ip_cmsg_send(sk, msg, &ipc, sk->sk_family == AF_INET6); if (unlikely(err)) { @@ -1059,6 +1058,8 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) saddr = ipc.addr; ipc.addr = faddr = daddr; + sock_tx_timestamp(sk, ipc.sockc.tsflags, &ipc.tx_flags); + if (ipc.opt && ipc.opt->opt.srr) { if (!daddr) return -EINVAL; diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 0a37ddc7af51..6b573ebe49de 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -400,6 +400,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) struct icmp6hdr tmp_hdr; struct flowi6 fl6; struct icmpv6_msg msg; + struct sockcm_cookie sockc_unused = {0}; int iif = 0; int addr_type = 0; int len; @@ -527,7 +528,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) len + sizeof(struct icmp6hdr), sizeof(struct icmp6hdr), hlimit, np->tclass, NULL, &fl6, (struct rt6_info *)dst, - MSG_DONTWAIT, np->dontfrag); + MSG_DONTWAIT, np->dontfrag, &sockc_unused); if (err) { ICMP6_INC_STATS(net, idev, ICMP6_MIB_OUTERRORS); ip6_flush_pending_frames(sk); @@ -566,6 +567,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb) int hlimit; u8 tclass; u32 mark = IP6_REPLY_MARK(net, skb->mark); + struct sockcm_cookie sockc_unused = {0}; saddr = &ipv6_hdr(skb)->daddr; @@ -617,7 +619,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb) err = ip6_append_data(sk, icmpv6_getfrag, &msg, skb->len + sizeof(struct icmp6hdr), sizeof(struct icmp6hdr), hlimit, tclass, NULL, &fl6, (struct rt6_info *)dst, MSG_DONTWAIT, - np->dontfrag); + np->dontfrag, &sockc_unused); if (err) { ICMP6_INC_STATS_BH(net, idev, ICMP6_MIB_OUTERRORS); diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index 9428345d3a07..612f3d138bf0 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -1258,7 +1258,8 @@ static int __ip6_append_data(struct sock *sk, int getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb), void *from, int length, int transhdrlen, - unsigned int flags, int dontfrag) + unsigned int flags, int dontfrag, + const struct sockcm_cookie *sockc) { struct sk_buff *skb, *skb_prev = NULL; unsigned int maxfraglen, fragheaderlen, mtu, orig_mtu; @@ -1329,7 +1330,7 @@ emsgsize: csummode = CHECKSUM_PARTIAL; if (sk->sk_type == SOCK_DGRAM || sk->sk_type == SOCK_RAW) { - sock_tx_timestamp(sk, &tx_flags); + sock_tx_timestamp(sk, sockc->tsflags, &tx_flags); if (tx_flags & SKBTX_ANY_SW_TSTAMP && sk->sk_tsflags & SOF_TIMESTAMPING_OPT_ID) tskey = sk->sk_tskey++; @@ -1565,7 +1566,8 @@ int ip6_append_data(struct sock *sk, int odd, struct sk_buff *skb), void *from, int length, int transhdrlen, int hlimit, int tclass, struct ipv6_txoptions *opt, struct flowi6 *fl6, - struct rt6_info *rt, unsigned int flags, int dontfrag) + struct rt6_info *rt, unsigned int flags, int dontfrag, + const struct sockcm_cookie *sockc) { struct inet_sock *inet = inet_sk(sk); struct ipv6_pinfo *np = inet6_sk(sk); @@ -1593,7 +1595,8 @@ int ip6_append_data(struct sock *sk, return __ip6_append_data(sk, fl6, &sk->sk_write_queue, &inet->cork.base, &np->cork, sk_page_frag(sk), getfrag, - from, length, transhdrlen, flags, dontfrag); + from, length, transhdrlen, flags, dontfrag, + sockc); } EXPORT_SYMBOL_GPL(ip6_append_data); @@ -1752,7 +1755,7 @@ struct sk_buff *ip6_make_skb(struct sock *sk, int hlimit, int tclass, struct ipv6_txoptions *opt, struct flowi6 *fl6, struct rt6_info *rt, unsigned int flags, - int dontfrag) + int dontfrag, const struct sockcm_cookie *sockc) { struct inet_cork_full cork; struct inet6_cork v6_cork; @@ -1779,7 +1782,7 @@ struct sk_buff *ip6_make_skb(struct sock *sk, err = __ip6_append_data(sk, fl6, &queue, &cork.base, &v6_cork, ¤t->task_frag, getfrag, from, length + exthdrlen, transhdrlen + exthdrlen, - flags, dontfrag); + flags, dontfrag, sockc); if (err) { __ip6_flush_pending_frames(sk, &queue, &cork, &v6_cork); return ERR_PTR(err); diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c index c382db7a2e73..da1cff79e447 100644 --- a/net/ipv6/ping.c +++ b/net/ipv6/ping.c @@ -62,6 +62,7 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) struct dst_entry *dst; struct rt6_info *rt; struct pingfakehdr pfh; + struct sockcm_cookie junk = {0}; pr_debug("ping_v6_sendmsg(sk=%p,sk->num=%u)\n", inet, inet->inet_num); @@ -144,7 +145,7 @@ static int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) err = ip6_append_data(sk, ping_getfrag, &pfh, len, 0, hlimit, np->tclass, NULL, &fl6, rt, - MSG_DONTWAIT, np->dontfrag); + MSG_DONTWAIT, np->dontfrag, &junk); if (err) { ICMP6_INC_STATS(sock_net(sk), rt->rt6i_idev, diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index f175ec0a97ce..b07ce21983aa 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -822,8 +822,7 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) if (fl6.flowi6_oif == 0) fl6.flowi6_oif = sk->sk_bound_dev_if; - sockc.tsflags = 0; - + sockc.tsflags = sk->sk_tsflags; if (msg->msg_controllen) { opt = &opt_space; memset(opt, 0, sizeof(struct ipv6_txoptions)); @@ -901,7 +900,7 @@ back_from_confirm: lock_sock(sk); err = ip6_append_data(sk, raw6_getfrag, &rfv, len, 0, hlimit, tclass, opt, &fl6, (struct rt6_info *)dst, - msg->msg_flags, dontfrag); + msg->msg_flags, dontfrag, &sockc); if (err) ip6_flush_pending_frames(sk); diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 2a787af42163..b772a7641fbd 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -1248,7 +1248,7 @@ do_udp_sendmsg: fl6.flowi6_oif = np->sticky_pktinfo.ipi6_ifindex; fl6.flowi6_mark = sk->sk_mark; - sockc.tsflags = 0; + sockc.tsflags = sk->sk_tsflags; if (msg->msg_controllen) { opt = &opt_space; @@ -1324,7 +1324,7 @@ back_from_confirm: skb = ip6_make_skb(sk, getfrag, msg, ulen, sizeof(struct udphdr), hlimit, tclass, opt, &fl6, (struct rt6_info *)dst, - msg->msg_flags, dontfrag); + msg->msg_flags, dontfrag, &sockc); err = PTR_ERR(skb); if (!IS_ERR_OR_NULL(skb)) err = udp_v6_send_skb(skb, &fl6); @@ -1351,7 +1351,8 @@ do_append_data: err = ip6_append_data(sk, getfrag, msg, ulen, sizeof(struct udphdr), hlimit, tclass, opt, &fl6, (struct rt6_info *)dst, - corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags, dontfrag); + corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags, dontfrag, + &sockc); if (err) udp_v6_flush_pending_frames(sk); else if (!corkreq) diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c index 4f29a4a0f360..1a38f20b1ca6 100644 --- a/net/l2tp/l2tp_ip6.c +++ b/net/l2tp/l2tp_ip6.c @@ -627,7 +627,7 @@ back_from_confirm: err = ip6_append_data(sk, ip_generic_getfrag, msg, ulen, transhdrlen, hlimit, tclass, opt, &fl6, (struct rt6_info *)dst, - msg->msg_flags, dontfrag); + msg->msg_flags, dontfrag, &sockc_unused); if (err) ip6_flush_pending_frames(sk); else if (!(msg->msg_flags & MSG_MORE)) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 1ecfa710ca98..0007e23202e4 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -1837,6 +1837,7 @@ static int packet_sendmsg_spkt(struct socket *sock, struct msghdr *msg, DECLARE_SOCKADDR(struct sockaddr_pkt *, saddr, msg->msg_name); struct sk_buff *skb = NULL; struct net_device *dev; + struct sockcm_cookie sockc; __be16 proto = 0; int err; int extra_len = 0; @@ -1925,12 +1926,21 @@ retry: goto out_unlock; } + sockc.tsflags = 0; + if (msg->msg_controllen) { + err = sock_cmsg_send(sk, msg, &sockc); + if (unlikely(err)) { + err = -EINVAL; + goto out_unlock; + } + } + skb->protocol = proto; skb->dev = dev; skb->priority = sk->sk_priority; skb->mark = sk->sk_mark; - sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags); + sock_tx_timestamp(sk, sockc.tsflags, &skb_shinfo(skb)->tx_flags); if (unlikely(extra_len == 4)) skb->no_fcs = 1; @@ -2486,7 +2496,8 @@ static int packet_snd_vnet_gso(struct sk_buff *skb, static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb, void *frame, struct net_device *dev, void *data, int tp_len, - __be16 proto, unsigned char *addr, int hlen, int copylen) + __be16 proto, unsigned char *addr, int hlen, int copylen, + const struct sockcm_cookie *sockc) { union tpacket_uhdr ph; int to_write, offset, len, nr_frags, len_max; @@ -2500,7 +2511,7 @@ static int tpacket_fill_skb(struct packet_sock *po, struct sk_buff *skb, skb->dev = dev; skb->priority = po->sk.sk_priority; skb->mark = po->sk.sk_mark; - sock_tx_timestamp(&po->sk, &skb_shinfo(skb)->tx_flags); + sock_tx_timestamp(&po->sk, sockc->tsflags, &skb_shinfo(skb)->tx_flags); skb_shinfo(skb)->destructor_arg = ph.raw; skb_reserve(skb, hlen); @@ -2624,6 +2635,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg) struct sk_buff *skb; struct net_device *dev; struct virtio_net_hdr *vnet_hdr = NULL; + struct sockcm_cookie sockc; __be16 proto; int err, reserve = 0; void *ph; @@ -2655,6 +2667,13 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg) dev = dev_get_by_index(sock_net(&po->sk), saddr->sll_ifindex); } + sockc.tsflags = 0; + if (msg->msg_controllen) { + err = sock_cmsg_send(&po->sk, msg, &sockc); + if (unlikely(err)) + goto out; + } + err = -ENXIO; if (unlikely(dev == NULL)) goto out; @@ -2712,7 +2731,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg) goto out_status; } tp_len = tpacket_fill_skb(po, skb, ph, dev, data, tp_len, proto, - addr, hlen, copylen); + addr, hlen, copylen, &sockc); if (likely(tp_len >= 0) && tp_len > dev->mtu + reserve && !po->has_vnet_hdr && @@ -2851,6 +2870,7 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) if (unlikely(!(dev->flags & IFF_UP))) goto out_unlock; + sockc.tsflags = 0; sockc.mark = sk->sk_mark; if (msg->msg_controllen) { err = sock_cmsg_send(sk, msg, &sockc); @@ -2908,7 +2928,7 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len) goto out_free; } - sock_tx_timestamp(sk, &skb_shinfo(skb)->tx_flags); + sock_tx_timestamp(sk, sockc.tsflags, &skb_shinfo(skb)->tx_flags); if (!vnet_hdr.gso_type && (len > dev->mtu + reserve + extra_len) && !packet_extra_vlan_len_allowed(dev, skb)) { diff --git a/net/socket.c b/net/socket.c index 5f77a8e93830..979d3146b081 100644 --- a/net/socket.c +++ b/net/socket.c @@ -587,20 +587,20 @@ void sock_release(struct socket *sock) } EXPORT_SYMBOL(sock_release); -void __sock_tx_timestamp(const struct sock *sk, __u8 *tx_flags) +void __sock_tx_timestamp(__u16 tsflags, __u8 *tx_flags) { u8 flags = *tx_flags; - if (sk->sk_tsflags & SOF_TIMESTAMPING_TX_HARDWARE) + if (tsflags & SOF_TIMESTAMPING_TX_HARDWARE) flags |= SKBTX_HW_TSTAMP; - if (sk->sk_tsflags & SOF_TIMESTAMPING_TX_SOFTWARE) + if (tsflags & SOF_TIMESTAMPING_TX_SOFTWARE) flags |= SKBTX_SW_TSTAMP; - if (sk->sk_tsflags & SOF_TIMESTAMPING_TX_SCHED) + if (tsflags & SOF_TIMESTAMPING_TX_SCHED) flags |= SKBTX_SCHED_TSTAMP; - if (sk->sk_tsflags & SOF_TIMESTAMPING_TX_ACK) + if (tsflags & SOF_TIMESTAMPING_TX_ACK) flags |= SKBTX_ACK_TSTAMP; *tx_flags = flags; |