netdev - [PATCH net-next] sock: remove zerocopy sockopt restriction on closed tcp state

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-next>] [day] [month] [year] [list]

Message-Id: <20180314164919.179364-1-willemdebruijn.kernel@gmail.com>
Date:   Wed, 14 Mar 2018 12:49:19 -0400
From:   Willem de Bruijn <willemdebruijn.kernel@...il.com>
To:     netdev@...r.kernel.org
Cc:     soheil@...gle.com, kou@...ich.ibm.com,
        Willem de Bruijn <willemb@...gle.com>
Subject: [PATCH net-next] sock: remove zerocopy sockopt restriction on closed tcp state

From: Willem de Bruijn <willemb@...gle.com>

Socket option SO_ZEROCOPY determines whether the kernel ignores or
processes flag MSG_ZEROCOPY on subsequent send calls. This to avoid
changing behavior for legacy processes.

Limiting the state change to closed sockets is annoying with passive
sockets and not necessary for correctness. Once created, zerocopy skbs
are processed based on their private state, not this socket flag.

Remove the constraint.

Signed-off-by: Willem de Bruijn <willemb@...gle.com>
---
 Documentation/networking/msg_zerocopy.rst | 5 -----
 net/core/sock.c                           | 2 --
 2 files changed, 7 deletions(-)

diff --git a/Documentation/networking/msg_zerocopy.rst b/Documentation/networking/msg_zerocopy.rst
index 291a01264967..fe46d4867e2d 100644
--- a/Documentation/networking/msg_zerocopy.rst
+++ b/Documentation/networking/msg_zerocopy.rst
@@ -72,11 +72,6 @@ this flag, a process must first signal intent by setting a socket option:
 	if (setsockopt(fd, SOL_SOCKET, SO_ZEROCOPY, &one, sizeof(one)))
 		error(1, errno, "setsockopt zerocopy");
 
-Setting the socket option only works when the socket is in its initial
-(TCP_CLOSED) state.  Trying to set the option for a socket returned by accept(),
-for example, will lead to an EBUSY error. In this case, the option should be set
-to the listening socket and it will be inherited by the accepted sockets.
-
 Transmission
 ------------
 
diff --git a/net/core/sock.c b/net/core/sock.c
index 27f218bba43f..a8962d912895 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -1052,8 +1052,6 @@ int sock_setsockopt(struct socket *sock, int level, int optname,
 		if (sk->sk_family == PF_INET || sk->sk_family == PF_INET6) {
 			if (sk->sk_protocol != IPPROTO_TCP)
 				ret = -ENOTSUPP;
-			else if (sk->sk_state != TCP_CLOSE)
-				ret = -EBUSY;
 		} else if (sk->sk_family != PF_RDS) {
 			ret = -ENOTSUPP;
 		}
-- 
2.16.2.804.g6dcf76e118-goog