[net-next,2/2] tcp-zerocopy: Return sk_err (if set) along with tcp receive zerocopy.
diff mbox series

Message ID 20200214233050.19429-2-arjunroy.kdev@gmail.com
State Accepted
Delegated to: David Miller
Headers show
Series
  • [net-next,1/2] tcp-zerocopy: Return inq along with tcp receive zerocopy.
Related show

Commit Message

Arjun Roy Feb. 14, 2020, 11:30 p.m. UTC
From: Arjun Roy <arjunroy@google.com>

This patchset is intended to reduce the number of extra system calls
imposed by TCP receive zerocopy. For ping-pong RPC style workloads,
this patchset has demonstrated a system call reduction of about 30%
when coupled with userspace changes.

For applications using epoll, returning sk_err along with the result
of tcp receive zerocopy could remove the need to call
recvmsg()=-EAGAIN after a spurious wakeup.

Consider a multi-threaded application using epoll. A thread may awaken
with EPOLLIN but another thread may already be reading. The
spuriously-awoken thread does not necessarily know that another thread
'won'; rather, it may be possible that it was woken up due to the
presence of an error if there is no data. A zerocopy read receiving 0
bytes thus would need to be followed up by recvmsg to be sure.

Instead, we return sk_err directly with zerocopy, so the application
can avoid this extra system call.

Signed-off-by: Arjun Roy <arjunroy@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>

---
 include/uapi/linux/tcp.h | 1 +
 net/ipv4/tcp.c           | 8 +++++++-
 2 files changed, 8 insertions(+), 1 deletion(-)

Comments

David Miller Feb. 17, 2020, 3:25 a.m. UTC | #1
From: Arjun Roy <arjunroy.kdev@gmail.com>
Date: Fri, 14 Feb 2020 15:30:50 -0800

> From: Arjun Roy <arjunroy@google.com>
> 
> This patchset is intended to reduce the number of extra system calls
> imposed by TCP receive zerocopy. For ping-pong RPC style workloads,
> this patchset has demonstrated a system call reduction of about 30%
> when coupled with userspace changes.
> 
> For applications using epoll, returning sk_err along with the result
> of tcp receive zerocopy could remove the need to call
> recvmsg()=-EAGAIN after a spurious wakeup.
> 
> Consider a multi-threaded application using epoll. A thread may awaken
> with EPOLLIN but another thread may already be reading. The
> spuriously-awoken thread does not necessarily know that another thread
> 'won'; rather, it may be possible that it was woken up due to the
> presence of an error if there is no data. A zerocopy read receiving 0
> bytes thus would need to be followed up by recvmsg to be sure.
> 
> Instead, we return sk_err directly with zerocopy, so the application
> can avoid this extra system call.
> 
> Signed-off-by: Arjun Roy <arjunroy@google.com>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>

Applied.

Patch
diff mbox series

diff --git a/include/uapi/linux/tcp.h b/include/uapi/linux/tcp.h
index 19700101cbba..e1706a7c9d88 100644
--- a/include/uapi/linux/tcp.h
+++ b/include/uapi/linux/tcp.h
@@ -344,5 +344,6 @@  struct tcp_zerocopy_receive {
 	__u32 length;		/* in/out: number of bytes to map/mapped */
 	__u32 recv_skip_hint;	/* out: amount of bytes to skip */
 	__u32 inq; /* out: amount of bytes in read queue */
+	__s32 err; /* out: socket error */
 };
 #endif /* _UAPI_LINUX_TCP_H */
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 947be81b35c5..0efac228bbdb 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3667,14 +3667,20 @@  static int do_tcp_getsockopt(struct sock *sk, int level,
 		lock_sock(sk);
 		err = tcp_zerocopy_receive(sk, &zc);
 		release_sock(sk);
+		if (len == sizeof(zc))
+			goto zerocopy_rcv_sk_err;
 		switch (len) {
-		case sizeof(zc):
+		case offsetofend(struct tcp_zerocopy_receive, err):
+			goto zerocopy_rcv_sk_err;
 		case offsetofend(struct tcp_zerocopy_receive, inq):
 			goto zerocopy_rcv_inq;
 		case offsetofend(struct tcp_zerocopy_receive, length):
 		default:
 			goto zerocopy_rcv_out;
 		}
+zerocopy_rcv_sk_err:
+		if (!err)
+			zc.err = sock_error(sk);
 zerocopy_rcv_inq:
 		zc.inq = tcp_inq_hint(sk);
 zerocopy_rcv_out: