sock: allow reading and changing sk_userlocks with setsockopt

SOCK_SNDBUF_LOCK and SOCK_RCVBUF_LOCK flags disable automatic socket
buffers adjustment done by kernel (see tcp_fixup_rcvbuf() and
tcp_sndbuf_expand()). If we've just created a new socket this adjustment
is enabled on it, but if one changes the socket buffer size by
setsockopt(SO_{SND,RCV}BUF*) it becomes disabled.

CRIU needs to call setsockopt(SO_{SND,RCV}BUF*) on each socket on
restore as it first needs to increase buffer sizes for packet queues
restore and second it needs to restore back original buffer sizes. So
after CRIU restore all sockets become non-auto-adjustable, which can
decrease network performance of restored applications significantly.

CRIU need to be able to restore sockets with enabled/disabled adjustment
to the same state it was before dump, so let's add special setsockopt
for it.

Let's also export SOCK_SNDBUF_LOCK and SOCK_RCVBUF_LOCK flags to uAPI so
that using these interface one can reenable automatic socket buffer
adjustment on their sockets.

Signed-off-by: Pavel Tikhomirov <ptikhomirov@virtuozzo.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Pavel Tikhomirov 2021-08-04 10:55:56 +03:00 committed by David S. Miller
parent 625af9f029
commit 04190bf894
8 changed files with 29 additions and 2 deletions

View file

@ -129,6 +129,8 @@
#define SO_NETNS_COOKIE 71
#define SO_BUF_LOCK 72
#if !defined(__KERNEL__)
#if __BITS_PER_LONG == 64

View file

@ -140,6 +140,8 @@
#define SO_NETNS_COOKIE 71
#define SO_BUF_LOCK 72
#if !defined(__KERNEL__)
#if __BITS_PER_LONG == 64

View file

@ -121,6 +121,8 @@
#define SO_NETNS_COOKIE 0x4045
#define SO_BUF_LOCK 0x4046
#if !defined(__KERNEL__)
#if __BITS_PER_LONG == 64

View file

@ -122,6 +122,8 @@
#define SO_NETNS_COOKIE 0x0050
#define SO_BUF_LOCK 0x0051
#if !defined(__KERNEL__)

View file

@ -68,6 +68,7 @@
#include <net/tcp_states.h>
#include <linux/net_tstamp.h>
#include <net/l3mdev.h>
#include <uapi/linux/socket.h>
/*
* This structure really needs to be cleaned up.
@ -1438,8 +1439,6 @@ static inline int __sk_prot_rehash(struct sock *sk)
#define RCV_SHUTDOWN 1
#define SEND_SHUTDOWN 2
#define SOCK_SNDBUF_LOCK 1
#define SOCK_RCVBUF_LOCK 2
#define SOCK_BINDADDR_LOCK 4
#define SOCK_BINDPORT_LOCK 8

View file

@ -124,6 +124,8 @@
#define SO_NETNS_COOKIE 71
#define SO_BUF_LOCK 72
#if !defined(__KERNEL__)
#if __BITS_PER_LONG == 64 || (defined(__x86_64__) && defined(__ILP32__))

View file

@ -26,4 +26,9 @@ struct __kernel_sockaddr_storage {
};
};
#define SOCK_SNDBUF_LOCK 1
#define SOCK_RCVBUF_LOCK 2
#define SOCK_BUF_LOCK_MASK (SOCK_SNDBUF_LOCK | SOCK_RCVBUF_LOCK)
#endif /* _UAPI_LINUX_SOCKET_H */

View file

@ -1358,6 +1358,15 @@ int sock_setsockopt(struct socket *sock, int level, int optname,
ret = sock_bindtoindex_locked(sk, val);
break;
case SO_BUF_LOCK:
if (val & ~SOCK_BUF_LOCK_MASK) {
ret = -EINVAL;
break;
}
sk->sk_userlocks = val | (sk->sk_userlocks &
~SOCK_BUF_LOCK_MASK);
break;
default:
ret = -ENOPROTOOPT;
break;
@ -1720,6 +1729,10 @@ int sock_getsockopt(struct socket *sock, int level, int optname,
v.val64 = sock_net(sk)->net_cookie;
break;
case SO_BUF_LOCK:
v.val = sk->sk_userlocks & SOCK_BUF_LOCK_MASK;
break;
default:
/* We implement the SO_SNDLOWAT etc to not be settable
* (1003.1g 7).