[PATCH 1/2] SUNRPC: Use TCP_CORK to optimise send performance on the server

[PATCH 1/2] SUNRPC: Use TCP_CORK to optimise send performance on the server

[trondmy]

From: Trond Myklebust <trond.myklebust@hammerspace.com>

Use a counter to keep track of how many requests are queued behind the
xprt->xpt_mutex, and keep TCP_CORK set until the queue is empty.

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 include/linux/sunrpc/svcsock.h | 2 ++
 net/sunrpc/svcsock.c           | 8 +++++++-
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/include/linux/sunrpc/svcsock.h b/include/linux/sunrpc/svcsock.h
index b7ac7fe68306..bcc555c7ae9c 100644
--- a/include/linux/sunrpc/svcsock.h
+++ b/include/linux/sunrpc/svcsock.h
@@ -35,6 +35,8 @@ struct svc_sock {
 	/* Total length of the data (not including fragment headers)
 	 * received so far in the fragments making up this rpc: */
 	u32			sk_datalen;
+	/* Number of queued send requests */
+	atomic_t		sk_sendqlen;
 
 	struct page *		sk_pages[RPCSVC_MAXPAGES];	/* received data */
 };
diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 5a809c64dc7b..231f510a4830 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -1171,18 +1171,23 @@ static int svc_tcp_sendto(struct svc_rqst *rqstp)
 
 	svc_tcp_release_rqst(rqstp);
 
+	atomic_inc(&svsk->sk_sendqlen);
 	mutex_lock(&xprt->xpt_mutex);
 	if (svc_xprt_is_dead(xprt))
 		goto out_notconn;
+	tcp_sock_set_cork(svsk->sk_sk, true);
 	err = svc_tcp_sendmsg(svsk->sk_sock, &msg, xdr, marker, &sent);
 	xdr_free_bvec(xdr);
 	trace_svcsock_tcp_send(xprt, err < 0 ? err : sent);
 	if (err < 0 || sent != (xdr->len + sizeof(marker)))
 		goto out_close;
+	if (atomic_dec_and_test(&svsk->sk_sendqlen))
+		tcp_sock_set_cork(svsk->sk_sk, false);
 	mutex_unlock(&xprt->xpt_mutex);
 	return sent;
 
 out_notconn:
+	atomic_dec(&svsk->sk_sendqlen);
 	mutex_unlock(&xprt->xpt_mutex);
 	return -ENOTCONN;
 out_close:
@@ -1192,6 +1197,7 @@ static int svc_tcp_sendto(struct svc_rqst *rqstp)
 		  (err < 0) ? err : sent, xdr->len);
 	set_bit(XPT_CLOSE, &xprt->xpt_flags);
 	svc_xprt_enqueue(xprt);
+	atomic_dec(&svsk->sk_sendqlen);
 	mutex_unlock(&xprt->xpt_mutex);
 	return -EAGAIN;
 }
@@ -1261,7 +1267,7 @@ static void svc_tcp_init(struct svc_sock *svsk, struct svc_serv *serv)
 		svsk->sk_datalen = 0;
 		memset(&svsk->sk_pages[0], 0, sizeof(svsk->sk_pages));
 
-		tcp_sk(sk)->nonagle |= TCP_NAGLE_OFF;
+		tcp_sock_set_nodelay(sk);
 
 		set_bit(XPT_DATA, &svsk->sk_xprt.xpt_flags);
 		switch (sk->sk_state) {
-- 
2.29.2

[PATCH 2/2] SUNRPC: Remove redundant socket flags from svc_tcp_sendmsg()

[trondmy]

From: Trond Myklebust <trond.myklebust@hammerspace.com>

Now that the caller controls the TCP_CORK socket option, it is redundant
to set MSG_MORE and MSG_SENDPAGE_NOTLAST in the calls to
kernel_sendpage().

Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
---
 net/sunrpc/svcsock.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/net/sunrpc/svcsock.c b/net/sunrpc/svcsock.c
index 231f510a4830..8c19732e425d 100644
--- a/net/sunrpc/svcsock.c
+++ b/net/sunrpc/svcsock.c
@@ -1088,12 +1088,11 @@ static int svc_tcp_sendmsg(struct socket *sock, struct msghdr *msg,
 		.iov_base	= &marker,
 		.iov_len	= sizeof(marker),
 	};
-	int flags, ret;
+	int ret;
 
 	*sentp = 0;
 	xdr_alloc_bvec(xdr, GFP_KERNEL);
 
-	msg->msg_flags = MSG_MORE;
 	ret = kernel_sendmsg(sock, msg, &rm, 1, rm.iov_len);
 	if (ret < 0)
 		return ret;
@@ -1101,8 +1100,7 @@ static int svc_tcp_sendmsg(struct socket *sock, struct msghdr *msg,
 	if (ret != rm.iov_len)
 		return -EAGAIN;
 
-	flags = head->iov_len < xdr->len ? MSG_MORE | MSG_SENDPAGE_NOTLAST : 0;
-	ret = svc_tcp_send_kvec(sock, head, flags);
+	ret = svc_tcp_send_kvec(sock, head, 0);
 	if (ret < 0)
 		return ret;
 	*sentp += ret;
@@ -1116,15 +1114,11 @@ static int svc_tcp_sendmsg(struct socket *sock, struct msghdr *msg,
 		bvec = xdr->bvec + (xdr->page_base >> PAGE_SHIFT);
 		offset = offset_in_page(xdr->page_base);
 		remaining = xdr->page_len;
-		flags = MSG_MORE | MSG_SENDPAGE_NOTLAST;
 		while (remaining > 0) {
-			if (remaining <= PAGE_SIZE && tail->iov_len == 0)
-				flags = 0;
-
 			len = min(remaining, bvec->bv_len - offset);
 			ret = kernel_sendpage(sock, bvec->bv_page,
 					      bvec->bv_offset + offset,
-					      len, flags);
+					      len, 0);
 			if (ret < 0)
 				return ret;
 			*sentp += ret;
-- 
2.29.2