netdev - [PATCH 08/28] net: add sock_set

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20200528051236.620353-9-hch@lst.de>
Date:   Thu, 28 May 2020 07:12:16 +0200
From:   Christoph Hellwig <hch@....de>
To:     "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>
Cc:     Eric Dumazet <edumazet@...gle.com>,
        Alexey Kuznetsov <kuznet@....inr.ac.ru>,
        Hideaki YOSHIFUJI <yoshfuji@...ux-ipv6.org>,
        Jon Maloy <jmaloy@...hat.com>,
        Ying Xue <ying.xue@...driver.com>, drbd-dev@...ts.linbit.com,
        linux-kernel@...r.kernel.org, linux-rdma@...r.kernel.org,
        linux-nvme@...ts.infradead.org, target-devel@...r.kernel.org,
        linux-afs@...ts.infradead.org, linux-cifs@...r.kernel.org,
        cluster-devel@...hat.com, ocfs2-devel@....oracle.com,
        netdev@...r.kernel.org, ceph-devel@...r.kernel.org,
        rds-devel@....oracle.com, linux-nfs@...r.kernel.org,
        tipc-discussion@...ts.sourceforge.net
Subject: [PATCH 08/28] net: add sock_set_rcvbuf

Add a helper to directly set the SO_RCVBUFFORCE sockopt from kernel space
without going through a fake uaccess.

Signed-off-by: Christoph Hellwig <hch@....de>
---
 fs/dlm/lowcomms.c  |  7 +-----
 include/net/sock.h |  1 +
 net/core/sock.c    | 59 +++++++++++++++++++++++++---------------------
 3 files changed, 34 insertions(+), 33 deletions(-)

diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c
index 138009c6a2ee1..45c37f572c9d2 100644
--- a/fs/dlm/lowcomms.c
+++ b/fs/dlm/lowcomms.c
@@ -1297,7 +1297,6 @@ static int sctp_listen_for_all(void)
 	struct socket *sock = NULL;
 	int result = -EINVAL;
 	struct connection *con = nodeid2con(0, GFP_NOFS);
-	int bufsize = NEEDED_RMEM;
 	int one = 1;
 
 	if (!con)
@@ -1312,11 +1311,7 @@ static int sctp_listen_for_all(void)
 		goto out;
 	}
 
-	result = kernel_setsockopt(sock, SOL_SOCKET, SO_RCVBUFFORCE,
-				 (char *)&bufsize, sizeof(bufsize));
-	if (result)
-		log_print("Error increasing buffer space on socket %d", result);
-
+	sock_set_rcvbuf(sock->sk, NEEDED_RMEM);
 	result = kernel_setsockopt(sock, SOL_SCTP, SCTP_NODELAY, (char *)&one,
 				   sizeof(one));
 	if (result < 0)
diff --git a/include/net/sock.h b/include/net/sock.h
index dc08c176238fd..c997289aabbf9 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -2693,6 +2693,7 @@ void sock_enable_timestamps(struct sock *sk);
 void sock_no_linger(struct sock *sk);
 void sock_set_keepalive(struct sock *sk);
 void sock_set_priority(struct sock *sk, u32 priority);
+void sock_set_rcvbuf(struct sock *sk, int val);
 void sock_set_reuseaddr(struct sock *sk);
 void sock_set_sndtimeo(struct sock *sk, s64 secs);
 
diff --git a/net/core/sock.c b/net/core/sock.c
index 728f5fb156a0c..3c6ebf952e9ad 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -789,6 +789,35 @@ void sock_set_keepalive(struct sock *sk)
 }
 EXPORT_SYMBOL(sock_set_keepalive);
 
+static void __sock_set_rcvbuf(struct sock *sk, int val)
+{
+	/* Ensure val * 2 fits into an int, to prevent max_t() from treating it
+	 * as a negative value.
+	 */
+	val = min_t(int, val, INT_MAX / 2);
+	sk->sk_userlocks |= SOCK_RCVBUF_LOCK;
+
+	/* We double it on the way in to account for "struct sk_buff" etc.
+	 * overhead.   Applications assume that the SO_RCVBUF setting they make
+	 * will allow that much actual data to be received on that socket.
+	 *
+	 * Applications are unaware that "struct sk_buff" and other overheads
+	 * allocate from the receive buffer during socket buffer allocation.
+	 *
+	 * And after considering the possible alternatives, returning the value
+	 * we actually used in getsockopt is the most desirable behavior.
+	 */
+	WRITE_ONCE(sk->sk_rcvbuf, max_t(int, val * 2, SOCK_MIN_RCVBUF));
+}
+
+void sock_set_rcvbuf(struct sock *sk, int val)
+{
+	lock_sock(sk);
+	__sock_set_rcvbuf(sk, val);
+	release_sock(sk);
+}
+EXPORT_SYMBOL(sock_set_rcvbuf);
+
 /*
  *	This is meant for all protocols to use and covers goings on
  *	at the socket level. Everything here is generic.
@@ -885,30 +914,7 @@ int sock_setsockopt(struct socket *sock, int level, int optname,
 		 * play 'guess the biggest size' games. RCVBUF/SNDBUF
 		 * are treated in BSD as hints
 		 */
-		val = min_t(u32, val, sysctl_rmem_max);
-set_rcvbuf:
-		/* Ensure val * 2 fits into an int, to prevent max_t()
-		 * from treating it as a negative value.
-		 */
-		val = min_t(int, val, INT_MAX / 2);
-		sk->sk_userlocks |= SOCK_RCVBUF_LOCK;
-		/*
-		 * We double it on the way in to account for
-		 * "struct sk_buff" etc. overhead.   Applications
-		 * assume that the SO_RCVBUF setting they make will
-		 * allow that much actual data to be received on that
-		 * socket.
-		 *
-		 * Applications are unaware that "struct sk_buff" and
-		 * other overheads allocate from the receive buffer
-		 * during socket buffer allocation.
-		 *
-		 * And after considering the possible alternatives,
-		 * returning the value we actually used in getsockopt
-		 * is the most desirable behavior.
-		 */
-		WRITE_ONCE(sk->sk_rcvbuf,
-			   max_t(int, val * 2, SOCK_MIN_RCVBUF));
+		__sock_set_rcvbuf(sk, min_t(u32, val, sysctl_rmem_max));
 		break;
 
 	case SO_RCVBUFFORCE:
@@ -920,9 +926,8 @@ int sock_setsockopt(struct socket *sock, int level, int optname,
 		/* No negative values (to prevent underflow, as val will be
 		 * multiplied by 2).
 		 */
-		if (val < 0)
-			val = 0;
-		goto set_rcvbuf;
+		__sock_set_rcvbuf(sk, max(val, 0));
+		break;
 
 	case SO_KEEPALIVE:
 		if (sk->sk_prot->keepalive)
-- 
2.26.2