mirrors/linux
1
0
Fork 0
mirror of https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git synced 2026-04-29 07:02:29 -04:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'inet-data-races'

Browse Source 
Eric Dumazet says:

====================
inet: socket lock and data-races avoidance

In this series, I converted 20 bits in "struct inet_sock" and made
them truly atomic.

This allows to implement many IP_ socket options in a lockless
fashion (no need to acquire socket lock), and fixes data-races
that were showing up in various KCSAN reports.

I also took care of IP_TTL/IP_MINTTL, but left few other options
for another series.

v4: Rebased after recent mptcp changes.
  Added Reviewed-by: tags from Simon (thanks !)

v3: fixed patch 7, feedback from build bot about ipvs set_mcast_loop()

v2: addressed a feedback from a build bot in patch 9 by removing
 unused issk variable in mptcp_setsockopt_sol_ip_set_transparent()
 Added Acked-by: tags from Soheil (thanks !)
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
David S. Miller
2023-08-16 11:09:18 +01:00
parent 936db833c2 12af73269f
commit 569dce3f8e
39 changed files with 362 additions and 370 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
include/net/inet_connection_sock.h
View File

@@ -342,9 +342,9 @@ static inline bool inet_csk_in_pingpong_mode(struct sock *sk)
return inet_csk(sk)->icsk_ack.pingpong >= TCP_PINGPONG_THRESH;
}
static inline bool inet_csk_has_ulp(struct sock *sk)
static inline bool inet_csk_has_ulp(const struct sock *sk)
{
return inet_sk(sk)->is_icsk && !!inet_csk(sk)->icsk_ulp_ops;
return inet_test_bit(IS_ICSK, sk) && !!inet_csk(sk)->icsk_ulp_ops;
}
#endif /* _INET_CONNECTION_SOCK_H */

92
include/net/inet_sock.h
View File

@@ -194,13 +194,13 @@ struct rtable;
* @inet_rcv_saddr - Bound local IPv4 addr
* @inet_dport - Destination port
* @inet_num - Local port
* @inet_flags - various atomic flags
* @inet_saddr - Sending source
* @uc_ttl - Unicast TTL
* @inet_sport - Source port
* @inet_id - ID counter for DF pkts
* @tos - TOS
* @mc_ttl - Multicasting TTL
* @is_icsk - is this an inet_connection_sock?
* @uc_index - Unicast outgoing device index
* @mc_index - Multicast device index
* @mc_list - Group array
@@ -218,57 +218,88 @@ struct inet_sock {
#define inet_dport sk.__sk_common.skc_dport
#define inet_num sk.__sk_common.skc_num
unsigned long inet_flags;
__be32 inet_saddr;
__s16 uc_ttl;
__u16 cmsg_flags;
struct ip_options_rcu __rcu *inet_opt;
__be16 inet_sport;
struct ip_options_rcu __rcu *inet_opt;
__u16 inet_id;
__u8 tos;
__u8 min_ttl;
__u8 mc_ttl;
__u8 pmtudisc;
__u8 recverr:1,
is_icsk:1,
freebind:1,
hdrincl:1,
mc_loop:1,
transparent:1,
mc_all:1,
nodefrag:1;
__u8 bind_address_no_port:1,
recverr_rfc4884:1,
defer_connect:1; /* Indicates that fastopen_connect is set
* and cookie exists so we defer connect
* until first data frame is written
*/
__u8 rcv_tos;
__u8 convert_csum;
int uc_index;
int mc_index;
__be32 mc_addr;
struct ip_mc_socklist __rcu *mc_list;
struct inet_cork_full cork;
struct {
__u16 lo;
__u16 hi;
} local_port_range;
struct ip_mc_socklist __rcu *mc_list;
struct inet_cork_full cork;
};
#define IPCORK_OPT 1 /* ip-options has been held in ipcork.opt */
#define IPCORK_ALLFRAG 2 /* always fragment (for ipv6 for now) */
enum {
INET_FLAGS_PKTINFO = 0,
INET_FLAGS_TTL = 1,
INET_FLAGS_TOS = 2,
INET_FLAGS_RECVOPTS = 3,
INET_FLAGS_RETOPTS = 4,
INET_FLAGS_PASSSEC = 5,
INET_FLAGS_ORIGDSTADDR = 6,
INET_FLAGS_CHECKSUM = 7,
INET_FLAGS_RECVFRAGSIZE = 8,
INET_FLAGS_RECVERR = 9,
INET_FLAGS_RECVERR_RFC4884 = 10,
INET_FLAGS_FREEBIND = 11,
INET_FLAGS_HDRINCL = 12,
INET_FLAGS_MC_LOOP = 13,
INET_FLAGS_MC_ALL = 14,
INET_FLAGS_TRANSPARENT = 15,
INET_FLAGS_IS_ICSK = 16,
INET_FLAGS_NODEFRAG = 17,
INET_FLAGS_BIND_ADDRESS_NO_PORT = 18,
INET_FLAGS_DEFER_CONNECT = 19,
};
/* cmsg flags for inet */
#define IP_CMSG_PKTINFO BIT(0)
#define IP_CMSG_TTL BIT(1)
#define IP_CMSG_TOS BIT(2)
#define IP_CMSG_RECVOPTS BIT(3)
#define IP_CMSG_RETOPTS BIT(4)
#define IP_CMSG_PASSSEC BIT(5)
#define IP_CMSG_ORIGDSTADDR BIT(6)
#define IP_CMSG_CHECKSUM BIT(7)
#define IP_CMSG_RECVFRAGSIZE BIT(8)
#define IP_CMSG_PKTINFO BIT(INET_FLAGS_PKTINFO)
#define IP_CMSG_TTL BIT(INET_FLAGS_TTL)
#define IP_CMSG_TOS BIT(INET_FLAGS_TOS)
#define IP_CMSG_RECVOPTS BIT(INET_FLAGS_RECVOPTS)
#define IP_CMSG_RETOPTS BIT(INET_FLAGS_RETOPTS)
#define IP_CMSG_PASSSEC BIT(INET_FLAGS_PASSSEC)
#define IP_CMSG_ORIGDSTADDR BIT(INET_FLAGS_ORIGDSTADDR)
#define IP_CMSG_CHECKSUM BIT(INET_FLAGS_CHECKSUM)
#define IP_CMSG_RECVFRAGSIZE BIT(INET_FLAGS_RECVFRAGSIZE)
#define IP_CMSG_ALL (IP_CMSG_PKTINFO | IP_CMSG_TTL | \
IP_CMSG_TOS | IP_CMSG_RECVOPTS | \
IP_CMSG_RETOPTS | IP_CMSG_PASSSEC | \
IP_CMSG_ORIGDSTADDR | IP_CMSG_CHECKSUM | \
IP_CMSG_RECVFRAGSIZE)
static inline unsigned long inet_cmsg_flags(const struct inet_sock *inet)
{
return READ_ONCE(inet->inet_flags) & IP_CMSG_ALL;
}
#define inet_test_bit(nr, sk) \
test_bit(INET_FLAGS_##nr, &inet_sk(sk)->inet_flags)
#define inet_set_bit(nr, sk) \
set_bit(INET_FLAGS_##nr, &inet_sk(sk)->inet_flags)
#define inet_clear_bit(nr, sk) \
clear_bit(INET_FLAGS_##nr, &inet_sk(sk)->inet_flags)
#define inet_assign_bit(nr, sk, val) \
assign_bit(INET_FLAGS_##nr, &inet_sk(sk)->inet_flags, val)
static inline bool sk_is_inet(struct sock *sk)
{
@@ -363,7 +394,7 @@ static inline __u8 inet_sk_flowi_flags(const struct sock *sk)
{
__u8 flags = 0;
if (inet_sk(sk)->transparent || inet_sk(sk)->hdrincl)
if (inet_test_bit(TRANSPARENT, sk) || inet_test_bit(HDRINCL, sk))
flags |= FLOWI_FLAG_ANYSRC;
return flags;
}
@@ -389,7 +420,8 @@ static inline bool inet_can_nonlocal_bind(struct net *net,
struct inet_sock *inet)
{
return READ_ONCE(net->ipv4.sysctl_ip_nonlocal_bind) ||
inet->freebind || inet->transparent;
test_bit(INET_FLAGS_FREEBIND, &inet->inet_flags) ||
test_bit(INET_FLAGS_TRANSPARENT, &inet->inet_flags);
}
static inline bool inet_addr_valid_or_nonlocal(struct net *net,

3
include/net/ipv6.h
View File

@@ -937,7 +937,8 @@ static inline bool ipv6_can_nonlocal_bind(struct net *net,
struct inet_sock *inet)
{
return net->ipv6.sysctl.ip_nonlocal_bind ||
inet->freebind || inet->transparent;
test_bit(INET_FLAGS_FREEBIND, &inet->inet_flags) ||
test_bit(INET_FLAGS_TRANSPARENT, &inet->inet_flags);
}
/* Sysctl settings for net ipv6.auto_flowlabels */

2
include/net/route.h
View File

@@ -298,7 +298,7 @@ static inline void ip_route_connect_init(struct flowi4 *fl4, __be32 dst,
{
__u8 flow_flags = 0;
if (inet_sk(sk)->transparent)
if (inet_test_bit(TRANSPARENT, sk))
flow_flags |= FLOWI_FLAG_ANYSRC;
flowi4_init_output(fl4, oif, READ_ONCE(sk->sk_mark), ip_sock_rt_tos(sk),

2
include/net/tcp.h
View File

@@ -2031,7 +2031,7 @@ static inline bool inet_sk_transparent(const struct sock *sk)
case TCP_NEW_SYN_RECV:
return inet_rsk(inet_reqsk(sk))->no_srccheck;
}
return inet_sk(sk)->transparent;
return inet_test_bit(TRANSPARENT, sk);
}
/* Determines whether this is a thin stream (which may suffer from

2
net/core/sock.c
View File

@@ -767,7 +767,7 @@ bool sk_mc_loop(struct sock *sk)
return true;
switch (sk->sk_family) {
case AF_INET:
return inet_sk(sk)->mc_loop;
return inet_test_bit(MC_LOOP, sk);
#if IS_ENABLED(CONFIG_IPV6)
case AF_INET6:
return inet6_sk(sk)->mc_loop;

4
net/dccp/ipv4.c
View File

@@ -247,7 +247,6 @@ static int dccp_v4_err(struct sk_buff *skb, u32 info)
const u8 offset = iph->ihl << 2;
const struct dccp_hdr *dh;
struct dccp_sock *dp;
struct inet_sock *inet;
const int type = icmp_hdr(skb)->type;
const int code = icmp_hdr(skb)->code;
struct sock *sk;
@@ -361,8 +360,7 @@ static int dccp_v4_err(struct sk_buff *skb, u32 info)
* --ANK (980905)
*/
inet = inet_sk(sk);
if (!sock_owned_by_user(sk) && inet->recverr) {
if (!sock_owned_by_user(sk) && inet_test_bit(RECVERR, sk)) {
sk->sk_err = err;
sk_error_report(sk);
} else { /* Only an error on timeout */

16
net/ipv4/af_inet.c
View File

@@ -331,14 +331,14 @@ static int inet_create(struct net *net, struct socket *sock, int protocol,
sk->sk_reuse = SK_CAN_REUSE;
inet = inet_sk(sk);
inet->is_icsk = (INET_PROTOSW_ICSK & answer_flags) != 0;
inet_assign_bit(IS_ICSK, sk, INET_PROTOSW_ICSK & answer_flags);
inet->nodefrag = 0;
inet_clear_bit(NODEFRAG, sk);
if (SOCK_RAW == sock->type) {
inet->inet_num = protocol;
if (IPPROTO_RAW == protocol)
inet->hdrincl = 1;
inet_set_bit(HDRINCL, sk);
}
if (READ_ONCE(net->ipv4.sysctl_ip_no_pmtu_disc))
@@ -356,9 +356,9 @@ static int inet_create(struct net *net, struct socket *sock, int protocol,
sk->sk_txrehash = READ_ONCE(net->core.sysctl_txrehash);
inet->uc_ttl = -1;
inet->mc_loop = 1;
inet_set_bit(MC_LOOP, sk);
inet->mc_ttl = 1;
inet->mc_all = 1;
inet_set_bit(MC_ALL, sk);
inet->mc_index = 0;
inet->mc_list = NULL;
inet->rcv_tos = 0;
@@ -529,7 +529,7 @@ int __inet_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len,
inet->inet_saddr = 0; /* Use device */
/* Make sure we are allowed to bind here. */
if (snum || !(inet->bind_address_no_port ||
if (snum || !(inet_test_bit(BIND_ADDRESS_NO_PORT, sk) ||
(flags & BIND_FORCE_ADDRESS_NO_PORT))) {
err = sk->sk_prot->get_port(sk, snum);
if (err) {
@@ -656,7 +656,7 @@ int __inet_stream_connect(struct socket *sock, struct sockaddr *uaddr,
err = -EISCONN;
goto out;
case SS_CONNECTING:
if (inet_sk(sk)->defer_connect)
if (inet_test_bit(DEFER_CONNECT, sk))
err = is_sendmsg ? -EINPROGRESS : -EISCONN;
else
err = -EALREADY;
@@ -679,7 +679,7 @@ int __inet_stream_connect(struct socket *sock, struct sockaddr *uaddr,
sock->state = SS_CONNECTING;
if (!err && inet_sk(sk)->defer_connect)
if (!err && inet_test_bit(DEFER_CONNECT, sk))
goto out;
/* Just entered SS_CONNECTING state; the only

4
net/ipv4/cipso_ipv4.c
View File

@@ -1881,7 +1881,7 @@ int cipso_v4_sock_setattr(struct sock *sk,
old = rcu_dereference_protected(sk_inet->inet_opt,
lockdep_sock_is_held(sk));
if (sk_inet->is_icsk) {
if (inet_test_bit(IS_ICSK, sk)) {
sk_conn = inet_csk(sk);
if (old)
sk_conn->icsk_ext_hdr_len -= old->opt.optlen;
@@ -2051,7 +2051,7 @@ void cipso_v4_sock_delattr(struct sock *sk)
sk_inet = inet_sk(sk);
hdr_delta = cipso_v4_delopt(&sk_inet->inet_opt);
if (sk_inet->is_icsk && hdr_delta > 0) {
if (inet_test_bit(IS_ICSK, sk) && hdr_delta > 0) {
struct inet_connection_sock *sk_conn = inet_csk(sk);
sk_conn->icsk_ext_hdr_len -= hdr_delta;
sk_conn->icsk_sync_mss(sk, sk_conn->icsk_pmtu_cookie);

2
net/ipv4/igmp.c
View File

@@ -2658,7 +2658,7 @@ int ip_mc_sf_allow(const struct sock *sk, __be32 loc_addr, __be32 rmt_addr,
(sdif && pmc->multi.imr_ifindex == sdif)))
break;
}
ret = inet->mc_all;
ret = inet_test_bit(MC_ALL, sk);
if (!pmc)
goto unlock;
psl = rcu_dereference(pmc->sflist);

22
net/ipv4/inet_diag.c
View File

@@ -182,17 +182,17 @@ int inet_diag_msg_attrs_fill(struct sock *sk, struct sk_buff *skb,
r->idiag_inode = sock_i_ino(sk);
memset(&inet_sockopt, 0, sizeof(inet_sockopt));
inet_sockopt.recverr = inet->recverr;
inet_sockopt.is_icsk = inet->is_icsk;
inet_sockopt.freebind = inet->freebind;
inet_sockopt.hdrincl = inet->hdrincl;
inet_sockopt.mc_loop = inet->mc_loop;
inet_sockopt.transparent = inet->transparent;
inet_sockopt.mc_all = inet->mc_all;
inet_sockopt.nodefrag = inet->nodefrag;
inet_sockopt.bind_address_no_port = inet->bind_address_no_port;
inet_sockopt.recverr_rfc4884 = inet->recverr_rfc4884;
inet_sockopt.defer_connect = inet->defer_connect;
inet_sockopt.recverr = inet_test_bit(RECVERR, sk);
inet_sockopt.is_icsk = inet_test_bit(IS_ICSK, sk);
inet_sockopt.freebind = inet_test_bit(FREEBIND, sk);
inet_sockopt.hdrincl = inet_test_bit(HDRINCL, sk);
inet_sockopt.mc_loop = inet_test_bit(MC_LOOP, sk);
inet_sockopt.transparent = inet_test_bit(TRANSPARENT, sk);
inet_sockopt.mc_all = inet_test_bit(MC_ALL, sk);
inet_sockopt.nodefrag = inet_test_bit(NODEFRAG, sk);
inet_sockopt.bind_address_no_port = inet_test_bit(BIND_ADDRESS_NO_PORT, sk);
inet_sockopt.recverr_rfc4884 = inet_test_bit(RECVERR_RFC4884, sk);
inet_sockopt.defer_connect = inet_test_bit(DEFER_CONNECT, sk);
if (nla_put(skb, INET_DIAG_SOCKOPT, sizeof(inet_sockopt),
&inet_sockopt))
goto errout;

2
net/ipv4/inet_timewait_sock.c
View File

@@ -203,7 +203,7 @@ struct inet_timewait_sock *inet_twsk_alloc(const struct sock *sk,
tw->tw_reuseport = sk->sk_reuseport;
tw->tw_hash = sk->sk_hash;
tw->tw_ipv6only = 0;
tw->tw_transparent = inet->transparent;
tw->tw_transparent = inet_test_bit(TRANSPARENT, sk);
tw->tw_prot = sk->sk_prot_creator;
atomic64_set(&tw->tw_cookie, atomic64_read(&sk->sk_cookie));
twsk_net_set(tw, sock_net(sk));

7
net/ipv4/ip_output.c
View File

@@ -133,7 +133,7 @@ EXPORT_SYMBOL_GPL(ip_local_out);
static inline int ip_select_ttl(const struct inet_sock *inet,
const struct dst_entry *dst)
{
int ttl = inet->uc_ttl;
int ttl = READ_ONCE(inet->uc_ttl);
if (ttl < 0)
ttl = ip4_dst_hoplimit(dst);
@@ -1039,7 +1039,7 @@ static int __ip_append_data(struct sock *sk,
}
}
} else if ((flags & MSG_SPLICE_PAGES) && length) {
if (inet->hdrincl)
if (inet_test_bit(HDRINCL, sk))
return -EPERM;
if (rt->dst.dev->features & NETIF_F_SG &&
getfrag == ip_generic_getfrag)
@@ -1467,7 +1467,8 @@ struct sk_buff *__ip_make_skb(struct sock *sk,
* so icmphdr does not in skb linear region and can not get icmp_type
* by icmp_hdr(skb)->type.
*/
if (sk->sk_type == SOCK_RAW && !inet_sk(sk)->hdrincl)
if (sk->sk_type == SOCK_RAW &&
!inet_test_bit(HDRINCL, sk))
icmp_type = fl4->fl4_icmp_type;
else
icmp_type = icmp_hdr(skb)->type;

405
net/ipv4/ip_sockglue.c
View File

@@ -171,8 +171,10 @@ static void ip_cmsg_recv_dstaddr(struct msghdr *msg, struct sk_buff *skb)
void ip_cmsg_recv_offset(struct msghdr *msg, struct sock *sk,
struct sk_buff *skb, int tlen, int offset)
{
struct inet_sock *inet = inet_sk(sk);
unsigned int flags = inet->cmsg_flags;
unsigned long flags = inet_cmsg_flags(inet_sk(sk));
if (!flags)
return;
/* Ordered by supposed usage frequency */
if (flags & IP_CMSG_PKTINFO) {
@@ -431,7 +433,7 @@ void ip_icmp_error(struct sock *sk, struct sk_buff *skb, int err,
serr->port = port;
if (skb_pull(skb, payload - skb->data)) {
if (inet_sk(sk)->recverr_rfc4884)
if (inet_test_bit(RECVERR_RFC4884, sk))
ipv4_icmp_error_rfc4884(skb, &serr->ee.ee_rfc4884);
skb_reset_transport_header(skb);
@@ -444,12 +446,11 @@ EXPORT_SYMBOL_GPL(ip_icmp_error);
void ip_local_error(struct sock *sk, int err, __be32 daddr, __be16 port, u32 info)
{
struct inet_sock *inet = inet_sk(sk);
struct sock_exterr_skb *serr;
struct iphdr *iph;
struct sk_buff *skb;
if (!inet->recverr)
if (!inet_test_bit(RECVERR, sk))
return;
skb = alloc_skb(sizeof(struct iphdr), GFP_ATOMIC);
@@ -568,7 +569,7 @@ int ip_recv_error(struct sock *sk, struct msghdr *msg, int len, int *addr_len)
if (ipv4_datagram_support_cmsg(sk, skb, serr->ee.ee_origin)) {
sin->sin_family = AF_INET;
sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
if (inet_sk(sk)->cmsg_flags)
if (inet_cmsg_flags(inet_sk(sk)))
ip_cmsg_recv(msg, skb);
}
@@ -607,17 +608,13 @@ EXPORT_SYMBOL(ip_sock_set_tos);
void ip_sock_set_freebind(struct sock *sk)
{
lock_sock(sk);
inet_sk(sk)->freebind = true;
release_sock(sk);
inet_set_bit(FREEBIND, sk);
}
EXPORT_SYMBOL(ip_sock_set_freebind);
void ip_sock_set_recverr(struct sock *sk)
{
lock_sock(sk);
inet_sk(sk)->recverr = true;
release_sock(sk);
inet_set_bit(RECVERR, sk);
}
EXPORT_SYMBOL(ip_sock_set_recverr);
@@ -634,9 +631,7 @@ EXPORT_SYMBOL(ip_sock_set_mtu_discover);
void ip_sock_set_pktinfo(struct sock *sk)
{
lock_sock(sk);
inet_sk(sk)->cmsg_flags |= IP_CMSG_PKTINFO;
release_sock(sk);
inet_set_bit(PKTINFO, sk);
}
EXPORT_SYMBOL(ip_sock_set_pktinfo);
@@ -950,6 +945,104 @@ int do_ip_setsockopt(struct sock *sk, int level, int optname,
if (ip_mroute_opt(optname))
return ip_mroute_setsockopt(sk, optname, optval, optlen);
/* Handle options that can be set without locking the socket. */
switch (optname) {
case IP_PKTINFO:
inet_assign_bit(PKTINFO, sk, val);
return 0;
case IP_RECVTTL:
inet_assign_bit(TTL, sk, val);
return 0;
case IP_RECVTOS:
inet_assign_bit(TOS, sk, val);
return 0;
case IP_RECVOPTS:
inet_assign_bit(RECVOPTS, sk, val);
return 0;
case IP_RETOPTS:
inet_assign_bit(RETOPTS, sk, val);
return 0;
case IP_PASSSEC:
inet_assign_bit(PASSSEC, sk, val);
return 0;
case IP_RECVORIGDSTADDR:
inet_assign_bit(ORIGDSTADDR, sk, val);
return 0;
case IP_RECVFRAGSIZE:
if (sk->sk_type != SOCK_RAW && sk->sk_type != SOCK_DGRAM)
return -EINVAL;
inet_assign_bit(RECVFRAGSIZE, sk, val);
return 0;
case IP_RECVERR:
inet_assign_bit(RECVERR, sk, val);
if (!val)
skb_queue_purge(&sk->sk_error_queue);
return 0;
case IP_RECVERR_RFC4884:
if (val < 0 || val > 1)
return -EINVAL;
inet_assign_bit(RECVERR_RFC4884, sk, val);
return 0;
case IP_FREEBIND:
if (optlen < 1)
return -EINVAL;
inet_assign_bit(FREEBIND, sk, val);
return 0;
case IP_HDRINCL:
if (sk->sk_type != SOCK_RAW)
return -ENOPROTOOPT;
inet_assign_bit(HDRINCL, sk, val);
return 0;
case IP_MULTICAST_LOOP:
if (optlen < 1)
return -EINVAL;
inet_assign_bit(MC_LOOP, sk, val);
return 0;
case IP_MULTICAST_ALL:
if (optlen < 1)
return -EINVAL;
if (val != 0 && val != 1)
return -EINVAL;
inet_assign_bit(MC_ALL, sk, val);
return 0;
case IP_TRANSPARENT:
if (!!val && !sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_RAW) &&
!sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) {
err = -EPERM;
break;
}
if (optlen < 1)
goto e_inval;
inet_assign_bit(TRANSPARENT, sk, val);
return 0;
case IP_NODEFRAG:
if (sk->sk_type != SOCK_RAW)
return -ENOPROTOOPT;
inet_assign_bit(NODEFRAG, sk, val);
return 0;
case IP_BIND_ADDRESS_NO_PORT:
inet_assign_bit(BIND_ADDRESS_NO_PORT, sk, val);
return 0;
case IP_TTL:
if (optlen < 1)
return -EINVAL;
if (val != -1 && (val < 1 || val > 255))
return -EINVAL;
WRITE_ONCE(inet->uc_ttl, val);
return 0;
case IP_MINTTL:
if (optlen < 1)
return -EINVAL;
if (val < 0 || val > 255)
return -EINVAL;
if (val)
static_branch_enable(&ip4_min_ttl);
WRITE_ONCE(inet->min_ttl, val);
return 0;
}
err = 0;
if (needs_rtnl)
rtnl_lock();
@@ -967,7 +1060,7 @@ int do_ip_setsockopt(struct sock *sk, int level, int optname,
break;
old = rcu_dereference_protected(inet->inet_opt,
lockdep_sock_is_held(sk));
if (inet->is_icsk) {
if (inet_test_bit(IS_ICSK, sk)) {
struct inet_connection_sock *icsk = inet_csk(sk);
#if IS_ENABLED(CONFIG_IPV6)
if (sk->sk_family == PF_INET ||
@@ -989,111 +1082,27 @@ int do_ip_setsockopt(struct sock *sk, int level, int optname,
kfree_rcu(old, rcu);
break;
}
case IP_PKTINFO:
if (val)
inet->cmsg_flags |= IP_CMSG_PKTINFO;
else
inet->cmsg_flags &= ~IP_CMSG_PKTINFO;
break;
case IP_RECVTTL:
if (val)
inet->cmsg_flags |= IP_CMSG_TTL;
else
inet->cmsg_flags &= ~IP_CMSG_TTL;
break;
case IP_RECVTOS:
if (val)
inet->cmsg_flags |= IP_CMSG_TOS;
else
inet->cmsg_flags &= ~IP_CMSG_TOS;
break;
case IP_RECVOPTS:
if (val)
inet->cmsg_flags |= IP_CMSG_RECVOPTS;
else
inet->cmsg_flags &= ~IP_CMSG_RECVOPTS;
break;
case IP_RETOPTS:
if (val)
inet->cmsg_flags |= IP_CMSG_RETOPTS;
else
inet->cmsg_flags &= ~IP_CMSG_RETOPTS;
break;
case IP_PASSSEC:
if (val)
inet->cmsg_flags |= IP_CMSG_PASSSEC;
else
inet->cmsg_flags &= ~IP_CMSG_PASSSEC;
break;
case IP_RECVORIGDSTADDR:
if (val)
inet->cmsg_flags |= IP_CMSG_ORIGDSTADDR;
else
inet->cmsg_flags &= ~IP_CMSG_ORIGDSTADDR;
break;
case IP_CHECKSUM:
if (val) {
if (!(inet->cmsg_flags & IP_CMSG_CHECKSUM)) {
if (!(inet_test_bit(CHECKSUM, sk))) {
inet_inc_convert_csum(sk);
inet->cmsg_flags |= IP_CMSG_CHECKSUM;
inet_set_bit(CHECKSUM, sk);
}
} else {
if (inet->cmsg_flags & IP_CMSG_CHECKSUM) {
if (inet_test_bit(CHECKSUM, sk)) {
inet_dec_convert_csum(sk);
inet->cmsg_flags &= ~IP_CMSG_CHECKSUM;
inet_clear_bit(CHECKSUM, sk);
}
}
break;
case IP_RECVFRAGSIZE:
if (sk->sk_type != SOCK_RAW && sk->sk_type != SOCK_DGRAM)
goto e_inval;
if (val)
inet->cmsg_flags |= IP_CMSG_RECVFRAGSIZE;
else
inet->cmsg_flags &= ~IP_CMSG_RECVFRAGSIZE;
break;
case IP_TOS: /* This sets both TOS and Precedence */
__ip_sock_set_tos(sk, val);
break;
case IP_TTL:
if (optlen < 1)
goto e_inval;
if (val != -1 && (val < 1 || val > 255))
goto e_inval;
inet->uc_ttl = val;
break;
case IP_HDRINCL:
if (sk->sk_type != SOCK_RAW) {
err = -ENOPROTOOPT;
break;
}
inet->hdrincl = val ? 1 : 0;
break;
case IP_NODEFRAG:
if (sk->sk_type != SOCK_RAW) {
err = -ENOPROTOOPT;
break;
}
inet->nodefrag = val ? 1 : 0;
break;
case IP_BIND_ADDRESS_NO_PORT:
inet->bind_address_no_port = val ? 1 : 0;
break;
case IP_MTU_DISCOVER:
if (val < IP_PMTUDISC_DONT || val > IP_PMTUDISC_OMIT)
goto e_inval;
inet->pmtudisc = val;
break;
case IP_RECVERR:
inet->recverr = !!val;
if (!val)
skb_queue_purge(&sk->sk_error_queue);
break;
case IP_RECVERR_RFC4884:
if (val < 0 || val > 1)
goto e_inval;
inet->recverr_rfc4884 = !!val;
break;
case IP_MULTICAST_TTL:
if (sk->sk_type == SOCK_STREAM)
goto e_inval;
@@ -1105,11 +1114,6 @@ int do_ip_setsockopt(struct sock *sk, int level, int optname,
goto e_inval;
inet->mc_ttl = val;
break;
case IP_MULTICAST_LOOP:
if (optlen < 1)
goto e_inval;
inet->mc_loop = !!val;
break;
case IP_UNICAST_IF:
{
struct net_device *dev = NULL;
@@ -1214,7 +1218,7 @@ int do_ip_setsockopt(struct sock *sk, int level, int optname,
struct ip_mreqn mreq;
err = -EPROTO;
if (inet_sk(sk)->is_icsk)
if (inet_test_bit(IS_ICSK, sk))
break;
if (optlen < sizeof(struct ip_mreq))
@@ -1325,20 +1329,6 @@ int do_ip_setsockopt(struct sock *sk, int level, int optname,
else
err = ip_set_mcast_msfilter(sk, optval, optlen);
break;
case IP_MULTICAST_ALL:
if (optlen < 1)
goto e_inval;
if (val != 0 && val != 1)
goto e_inval;
inet->mc_all = val;
break;
case IP_FREEBIND:
if (optlen < 1)
goto e_inval;
inet->freebind = !!val;
break;
case IP_IPSEC_POLICY:
case IP_XFRM_POLICY:
err = -EPERM;
@@ -1347,32 +1337,6 @@ int do_ip_setsockopt(struct sock *sk, int level, int optname,
err = xfrm_user_policy(sk, optname, optval, optlen);
break;
case IP_TRANSPARENT:
if (!!val && !sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_RAW) &&
!sockopt_ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN)) {
err = -EPERM;
break;
}
if (optlen < 1)
goto e_inval;
inet->transparent = !!val;
break;
case IP_MINTTL:
if (optlen < 1)
goto e_inval;
if (val < 0 || val > 255)
goto e_inval;
if (val)
static_branch_enable(&ip4_min_ttl);
/* tcp_v4_err() and tcp_v4_rcv() might read min_ttl
* while we are changint it.
*/
WRITE_ONCE(inet->min_ttl, val);
break;
case IP_LOCAL_PORT_RANGE:
{
const __u16 lo = val;
@@ -1415,7 +1379,7 @@ int do_ip_setsockopt(struct sock *sk, int level, int optname,
void ipv4_pktinfo_prepare(const struct sock *sk, struct sk_buff *skb)
{
struct in_pktinfo *pktinfo = PKTINFO_SKB_CB(skb);
bool prepare = (inet_sk(sk)->cmsg_flags & IP_CMSG_PKTINFO) ||
bool prepare = inet_test_bit(PKTINFO, sk) ||
ipv6_sk_rxinfo(sk);
if (prepare && skb_rtable(skb)) {
@@ -1566,6 +1530,72 @@ int do_ip_getsockopt(struct sock *sk, int level, int optname,
if (len < 0)
return -EINVAL;
/* Handle options that can be read without locking the socket. */
switch (optname) {
case IP_PKTINFO:
val = inet_test_bit(PKTINFO, sk);
goto copyval;
case IP_RECVTTL:
val = inet_test_bit(TTL, sk);
goto copyval;
case IP_RECVTOS:
val = inet_test_bit(TOS, sk);
goto copyval;
case IP_RECVOPTS:
val = inet_test_bit(RECVOPTS, sk);
goto copyval;
case IP_RETOPTS:
val = inet_test_bit(RETOPTS, sk);
goto copyval;
case IP_PASSSEC:
val = inet_test_bit(PASSSEC, sk);
goto copyval;
case IP_RECVORIGDSTADDR:
val = inet_test_bit(ORIGDSTADDR, sk);
goto copyval;
case IP_CHECKSUM:
val = inet_test_bit(CHECKSUM, sk);
goto copyval;
case IP_RECVFRAGSIZE:
val = inet_test_bit(RECVFRAGSIZE, sk);
goto copyval;
case IP_RECVERR:
val = inet_test_bit(RECVERR, sk);
goto copyval;
case IP_RECVERR_RFC4884:
val = inet_test_bit(RECVERR_RFC4884, sk);
goto copyval;
case IP_FREEBIND:
val = inet_test_bit(FREEBIND, sk);
goto copyval;
case IP_HDRINCL:
val = inet_test_bit(HDRINCL, sk);
goto copyval;
case IP_MULTICAST_LOOP:
val = inet_test_bit(MC_LOOP, sk);
goto copyval;
case IP_MULTICAST_ALL:
val = inet_test_bit(MC_ALL, sk);
goto copyval;
case IP_TRANSPARENT:
val = inet_test_bit(TRANSPARENT, sk);
goto copyval;
case IP_NODEFRAG:
val = inet_test_bit(NODEFRAG, sk);
goto copyval;
case IP_BIND_ADDRESS_NO_PORT:
val = inet_test_bit(BIND_ADDRESS_NO_PORT, sk);
goto copyval;
case IP_TTL:
val = READ_ONCE(inet->uc_ttl);
if (val < 0)
val = READ_ONCE(sock_net(sk)->ipv4.sysctl_ip_default_ttl);
goto copyval;
case IP_MINTTL:
val = READ_ONCE(inet->min_ttl);
goto copyval;
}
if (needs_rtnl)
rtnl_lock();
sockopt_lock_sock(sk);
@@ -1600,53 +1630,9 @@ int do_ip_getsockopt(struct sock *sk, int level, int optname,
return -EFAULT;
return 0;
}
case IP_PKTINFO:
val = (inet->cmsg_flags & IP_CMSG_PKTINFO) != 0;
break;
case IP_RECVTTL:
val = (inet->cmsg_flags & IP_CMSG_TTL) != 0;
break;
case IP_RECVTOS:
val = (inet->cmsg_flags & IP_CMSG_TOS) != 0;
break;
case IP_RECVOPTS:
val = (inet->cmsg_flags & IP_CMSG_RECVOPTS) != 0;
break;
case IP_RETOPTS:
val = (inet->cmsg_flags & IP_CMSG_RETOPTS) != 0;
break;
case IP_PASSSEC:
val = (inet->cmsg_flags & IP_CMSG_PASSSEC) != 0;
break;
case IP_RECVORIGDSTADDR:
val = (inet->cmsg_flags & IP_CMSG_ORIGDSTADDR) != 0;
break;
case IP_CHECKSUM:
val = (inet->cmsg_flags & IP_CMSG_CHECKSUM) != 0;
break;
case IP_RECVFRAGSIZE:
val = (inet->cmsg_flags & IP_CMSG_RECVFRAGSIZE) != 0;
break;
case IP_TOS:
val = inet->tos;
break;
case IP_TTL:
{
struct net *net = sock_net(sk);
val = (inet->uc_ttl == -1 ?
READ_ONCE(net->ipv4.sysctl_ip_default_ttl) :
inet->uc_ttl);
break;
}
case IP_HDRINCL:
val = inet->hdrincl;
break;
case IP_NODEFRAG:
val = inet->nodefrag;
break;
case IP_BIND_ADDRESS_NO_PORT:
val = inet->bind_address_no_port;
break;
case IP_MTU_DISCOVER:
val = inet->pmtudisc;
break;
@@ -1665,18 +1651,9 @@ int do_ip_getsockopt(struct sock *sk, int level, int optname,
}
break;
}
case IP_RECVERR:
val = inet->recverr;
break;
case IP_RECVERR_RFC4884:
val = inet->recverr_rfc4884;
break;
case IP_MULTICAST_TTL:
val = inet->mc_ttl;
break;
case IP_MULTICAST_LOOP:
val = inet->mc_loop;
break;
case IP_UNICAST_IF:
val = (__force int)htonl((__u32) inet->uc_index);
break;
@@ -1715,9 +1692,6 @@ int do_ip_getsockopt(struct sock *sk, int level, int optname,
else
err = ip_get_mcast_msfilter(sk, optval, optlen, len);
goto out;
case IP_MULTICAST_ALL:
val = inet->mc_all;
break;
case IP_PKTOPTIONS:
{
struct msghdr msg;
@@ -1737,7 +1711,7 @@ int do_ip_getsockopt(struct sock *sk, int level, int optname,
msg.msg_controllen = len;
msg.msg_flags = in_compat_syscall() ? MSG_CMSG_COMPAT : 0;
if (inet->cmsg_flags & IP_CMSG_PKTINFO) {
if (inet_test_bit(PKTINFO, sk)) {
struct in_pktinfo info;
info.ipi_addr.s_addr = inet->inet_rcv_saddr;
@@ -1745,26 +1719,17 @@ int do_ip_getsockopt(struct sock *sk, int level, int optname,
info.ipi_ifindex = inet->mc_index;
put_cmsg(&msg, SOL_IP, IP_PKTINFO, sizeof(info), &info);
}
if (inet->cmsg_flags & IP_CMSG_TTL) {
if (inet_test_bit(TTL, sk)) {
int hlim = inet->mc_ttl;
put_cmsg(&msg, SOL_IP, IP_TTL, sizeof(hlim), &hlim);
}
if (inet->cmsg_flags & IP_CMSG_TOS) {
if (inet_test_bit(TOS, sk)) {
int tos = inet->rcv_tos;
put_cmsg(&msg, SOL_IP, IP_TOS, sizeof(tos), &tos);
}
len -= msg.msg_controllen;
return copy_to_sockptr(optlen, &len, sizeof(int));
}
case IP_FREEBIND:
val = inet->freebind;
break;
case IP_TRANSPARENT:
val = inet->transparent;
break;
case IP_MINTTL:
val = inet->min_ttl;
break;
case IP_LOCAL_PORT_RANGE:
val = inet->local_port_range.hi << 16 | inet->local_port_range.lo;
break;
@@ -1776,7 +1741,7 @@ int do_ip_getsockopt(struct sock *sk, int level, int optname,
return -ENOPROTOOPT;
}
sockopt_release_sock(sk);
copyval:
if (len < sizeof(int) && len > 0 && val >= 0 && val <= 255) {
unsigned char ucval = (unsigned char)val;
len = 1;

2
net/ipv4/netfilter/nf_defrag_ipv4.c
View File

@@ -66,7 +66,7 @@ static unsigned int ipv4_conntrack_defrag(void *priv,
struct sock *sk = skb->sk;
if (sk && sk_fullsock(sk) && (sk->sk_family == PF_INET) &&
inet_sk(sk)->nodefrag)
inet_test_bit(NODEFRAG, sk))
return NF_ACCEPT;
#if IS_ENABLED(CONFIG_NF_CONNTRACK)

7
net/ipv4/ping.c
View File

@@ -580,7 +580,7 @@ void ping_err(struct sk_buff *skb, int offset, u32 info)
* RFC1122: OK. Passes ICMP errors back to application, as per
* 4.1.3.3.
*/
if ((family == AF_INET && !inet_sock->recverr) ||
if ((family == AF_INET && !inet_test_bit(RECVERR, sk)) ||
(family == AF_INET6 && !inet6_sk(sk)->recverr)) {
if (!harderr || sk->sk_state != TCP_ESTABLISHED)
goto out;
@@ -894,7 +894,7 @@ int ping_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags,
*addr_len = sizeof(*sin);
}
if (isk->cmsg_flags)
if (inet_cmsg_flags(isk))
ip_cmsg_recv(msg, skb);
#if IS_ENABLED(CONFIG_IPV6)
@@ -921,7 +921,8 @@ int ping_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags,
if (skb->protocol == htons(ETH_P_IPV6) &&
inet6_sk(sk)->rxopt.all)
pingv6_ops.ip6_datagram_recv_specific_ctl(sk, msg, skb);
else if (skb->protocol == htons(ETH_P_IP) && isk->cmsg_flags)
else if (skb->protocol == htons(ETH_P_IP) &&
inet_cmsg_flags(isk))
ip_cmsg_recv(msg, skb);
#endif
} else {

26
net/ipv4/raw.c
View File

@@ -203,8 +203,9 @@ static void raw_err(struct sock *sk, struct sk_buff *skb, u32 info)
struct inet_sock *inet = inet_sk(sk);
const int type = icmp_hdr(skb)->type;
const int code = icmp_hdr(skb)->code;
int err = 0;
int harderr = 0;
bool recverr;
int err = 0;
if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED)
ipv4_sk_update_pmtu(skb, sk, info);
@@ -218,7 +219,8 @@ static void raw_err(struct sock *sk, struct sk_buff *skb, u32 info)
2. Socket is connected (otherwise the error indication
is useless without ip_recverr and error is hard.
*/
if (!inet->recverr && sk->sk_state != TCP_ESTABLISHED)
recverr = inet_test_bit(RECVERR, sk);
if (!recverr && sk->sk_state != TCP_ESTABLISHED)
return;
switch (type) {
@@ -245,16 +247,16 @@ static void raw_err(struct sock *sk, struct sk_buff *skb, u32 info)
}
}
if (inet->recverr) {
if (recverr) {
const struct iphdr *iph = (const struct iphdr *)skb->data;
u8 *payload = skb->data + (iph->ihl << 2);
if (inet->hdrincl)
if (inet_test_bit(HDRINCL, sk))
payload = skb->data;
ip_icmp_error(sk, skb, err, 0, info, payload);
}
if (inet->recverr || harderr) {
if (recverr || harderr) {
sk->sk_err = err;
sk_error_report(sk);
}
@@ -413,7 +415,7 @@ static int raw_send_hdrinc(struct sock *sk, struct flowi4 *fl4,
kfree_skb(skb);
error:
IP_INC_STATS(net, IPSTATS_MIB_OUTDISCARDS);
if (err == -ENOBUFS && !inet->recverr)
if (err == -ENOBUFS && !inet_test_bit(RECVERR, sk))
err = 0;
return err;
}
@@ -489,12 +491,8 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
if (len > 0xFFFF)
goto out;
/* hdrincl should be READ_ONCE(inet->hdrincl)
* but READ_ONCE() doesn't work with bit fields.
* Doing this indirectly yields the same result.
*/
hdrincl = inet->hdrincl;
hdrincl = READ_ONCE(hdrincl);
hdrincl = inet_test_bit(HDRINCL, sk);
/*
* Check the flags.
*/
@@ -645,7 +643,7 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
ip_flush_pending_frames(sk);
else if (!(msg->msg_flags & MSG_MORE)) {
err = ip_push_pending_frames(sk, &fl4);
if (err == -ENOBUFS && !inet->recverr)
if (err == -ENOBUFS && !inet_test_bit(RECVERR, sk))
err = 0;
}
release_sock(sk);
@@ -767,7 +765,7 @@ static int raw_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
memset(&sin->sin_zero, 0, sizeof(sin->sin_zero));
*addr_len = sizeof(*sin);
}
if (inet->cmsg_flags)
if (inet_cmsg_flags(inet))
ip_cmsg_recv(msg, skb);
if (flags & MSG_TRUNC)
copied = skb->len;

8
net/ipv4/route.c
View File

@@ -515,13 +515,12 @@ static void __build_flow_key(const struct net *net, struct flowi4 *fl4,
__u8 scope = RT_SCOPE_UNIVERSE;
if (sk) {
const struct inet_sock *inet = inet_sk(sk);
oif = sk->sk_bound_dev_if;
mark = READ_ONCE(sk->sk_mark);
tos = ip_sock_rt_tos(sk);
scope = ip_sock_rt_scope(sk);
prot = inet->hdrincl ? IPPROTO_RAW : sk->sk_protocol;
prot = inet_test_bit(HDRINCL, sk) ? IPPROTO_RAW :
sk->sk_protocol;
}
flowi4_init_output(fl4, oif, mark, tos & IPTOS_RT_MASK, scope,
@@ -555,7 +554,8 @@ static void build_sk_flow_key(struct flowi4 *fl4, const struct sock *sk)
flowi4_init_output(fl4, sk->sk_bound_dev_if, READ_ONCE(sk->sk_mark),
ip_sock_rt_tos(sk) & IPTOS_RT_MASK,
ip_sock_rt_scope(sk),
inet->hdrincl ? IPPROTO_RAW : sk->sk_protocol,
inet_test_bit(HDRINCL, sk) ?
IPPROTO_RAW : sk->sk_protocol,
inet_sk_flowi_flags(sk),
daddr, inet->inet_saddr, 0, 0, sk->sk_uid);
rcu_read_unlock();

12
net/ipv4/tcp.c
View File

@@ -583,7 +583,8 @@ __poll_t tcp_poll(struct file *file, struct socket *sock, poll_table *wait)
if (urg_data & TCP_URG_VALID)
mask |= EPOLLPRI;
} else if (state == TCP_SYN_SENT && inet_sk(sk)->defer_connect) {
} else if (state == TCP_SYN_SENT &&
inet_test_bit(DEFER_CONNECT, sk)) {
/* Active TCP fastopen socket with defer_connect
* Return EPOLLOUT so application can call write()
* in order for kernel to generate SYN+data
@@ -1007,7 +1008,7 @@ int tcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg, int *copied,
tp->fastopen_req->size = size;
tp->fastopen_req->uarg = uarg;
if (inet->defer_connect) {
if (inet_test_bit(DEFER_CONNECT, sk)) {
err = tcp_connect(sk);
/* Same failure procedure as in tcp_v4/6_connect */
if (err) {
@@ -1025,7 +1026,7 @@ int tcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg, int *copied,
if (tp->fastopen_req) {
*copied = tp->fastopen_req->copied;
tcp_free_fastopen_req(tp);
inet->defer_connect = 0;
inet_clear_bit(DEFER_CONNECT, sk);
}
return err;
}
@@ -1066,7 +1067,8 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size)
zc = MSG_SPLICE_PAGES;
}
if (unlikely(flags & MSG_FASTOPEN || inet_sk(sk)->defer_connect) &&
if (unlikely(flags & MSG_FASTOPEN ||
inet_test_bit(DEFER_CONNECT, sk)) &&
!tp->repair) {
err = tcp_sendmsg_fastopen(sk, msg, &copied_syn, size, uarg);
if (err == -EINPROGRESS && copied_syn > 0)
@@ -3088,7 +3090,7 @@ int tcp_disconnect(struct sock *sk, int flags)
/* Clean up fastopen related fields */
tcp_free_fastopen_req(tp);
inet->defer_connect = 0;
inet_clear_bit(DEFER_CONNECT, sk);
tp->fastopen_client_fail = 0;
WARN_ON(inet->inet_num && !icsk->icsk_bind_hash);

2
net/ipv4/tcp_fastopen.c
View File

@@ -451,7 +451,7 @@ bool tcp_fastopen_defer_connect(struct sock *sk, int *err)
if (tp->fastopen_connect && !tp->fastopen_req) {
if (tcp_fastopen_cookie_check(sk, &mss, &cookie)) {
inet_sk(sk)->defer_connect = 1;
inet_set_bit(DEFER_CONNECT, sk);
return true;
}

2
net/ipv4/tcp_input.c
View File

@@ -7000,7 +7000,7 @@ int tcp_conn_request(struct request_sock_ops *rsk_ops,
tmp_opt.tstamp_ok = tmp_opt.saw_tstamp;
tcp_openreq_init(req, &tmp_opt, skb, sk);
inet_rsk(req)->no_srccheck = inet_sk(sk)->transparent;
inet_rsk(req)->no_srccheck = inet_test_bit(TRANSPARENT, sk);
/* Note: tcp_v6_init_req() might override ir_iif for link locals */
inet_rsk(req)->ir_iif = inet_request_bound_dev_if(sk, skb);

5
net/ipv4/tcp_ipv4.c
View File

@@ -477,7 +477,6 @@ int tcp_v4_err(struct sk_buff *skb, u32 info)
const struct iphdr *iph = (const struct iphdr *)skb->data;
struct tcphdr *th = (struct tcphdr *)(skb->data + (iph->ihl << 2));
struct tcp_sock *tp;
struct inet_sock *inet;
const int type = icmp_hdr(skb)->type;
const int code = icmp_hdr(skb)->code;
struct sock *sk;
@@ -625,8 +624,8 @@ int tcp_v4_err(struct sk_buff *skb, u32 info)
* --ANK (980905)
*/
inet = inet_sk(sk);
if (!sock_owned_by_user(sk) && inet->recverr) {
if (!sock_owned_by_user(sk) &&
inet_test_bit(RECVERR, sk)) {
WRITE_ONCE(sk->sk_err, err);
sk_error_report(sk);
} else { /* Only an error on timeout */

3
net/ipv4/tcp_minisocks.c
View File

@@ -289,9 +289,8 @@ void tcp_time_wait(struct sock *sk, int state, int timeo)
if (tw) {
struct tcp_timewait_sock *tcptw = tcp_twsk((struct sock *)tw);
const int rto = (icsk->icsk_rto << 2) - (icsk->icsk_rto >> 1);
struct inet_sock *inet = inet_sk(sk);
tw->tw_transparent = inet->transparent;
tw->tw_transparent = inet_test_bit(TRANSPARENT, sk);
tw->tw_mark = sk->sk_mark;
tw->tw_priority = sk->sk_priority;
tw->tw_rcv_wscale = tp->rx_opt.rcv_wscale;

7
net/ipv4/udp.c
View File

@@ -779,7 +779,7 @@ int __udp4_lib_err(struct sk_buff *skb, u32 info, struct udp_table *udptable)
(u8 *)(uh+1));
goto out;
}
if (!inet->recverr) {
if (!inet_test_bit(RECVERR, sk)) {
if (!harderr || sk->sk_state != TCP_ESTABLISHED)
goto out;
} else
@@ -962,7 +962,8 @@ static int udp_send_skb(struct sk_buff *skb, struct flowi4 *fl4,
send:
err = ip_send_skb(sock_net(sk), skb);
if (err) {
if (err == -ENOBUFS && !inet->recverr) {
if (err == -ENOBUFS &&
!inet_test_bit(RECVERR, sk)) {
UDP_INC_STATS(sock_net(sk),
UDP_MIB_SNDBUFERRORS, is_udplite);
err = 0;
@@ -1870,7 +1871,7 @@ int udp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len, int flags,
if (udp_sk(sk)->gro_enabled)
udp_cmsg_recv(msg, sk, skb);
if (inet->cmsg_flags)
if (inet_cmsg_flags(inet))
ip_cmsg_recv_offset(msg, sk, skb, sizeof(struct udphdr), off);
err = copied;

2
net/ipv4/udp_tunnel_core.c
View File

@@ -63,7 +63,7 @@ void setup_udp_tunnel_sock(struct net *net, struct socket *sock,
struct sock *sk = sock->sk;
/* Disable multicast loopback */
inet_sk(sk)->mc_loop = 0;
inet_clear_bit(MC_LOOP, sk);
/* Enable CHECKSUM_UNNECESSARY to CHECKSUM_COMPLETE conversion */
inet_inc_convert_csum(sk);

8
net/ipv6/af_inet6.c
View File

@@ -200,12 +200,12 @@ static int inet6_create(struct net *net, struct socket *sock, int protocol,
sk->sk_reuse = SK_CAN_REUSE;
inet = inet_sk(sk);
inet->is_icsk = (INET_PROTOSW_ICSK & answer_flags) != 0;
inet_assign_bit(IS_ICSK, sk, INET_PROTOSW_ICSK & answer_flags);
if (SOCK_RAW == sock->type) {
inet->inet_num = protocol;
if (IPPROTO_RAW == protocol)
inet->hdrincl = 1;
inet_set_bit(HDRINCL, sk);
}
sk->sk_destruct = inet6_sock_destruct;
@@ -229,7 +229,7 @@ static int inet6_create(struct net *net, struct socket *sock, int protocol,
*/
inet->uc_ttl = -1;
inet->mc_loop = 1;
inet_set_bit(MC_LOOP, sk);
inet->mc_ttl = 1;
inet->mc_index = 0;
RCU_INIT_POINTER(inet->mc_list, NULL);
@@ -399,7 +399,7 @@ static int __inet6_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len,
sk->sk_ipv6only = 1;
/* Make sure we are allowed to bind here. */
if (snum || !(inet->bind_address_no_port ||
if (snum || !(inet_test_bit(BIND_ADDRESS_NO_PORT, sk) ||
(flags & BIND_FORCE_ADDRESS_NO_PORT))) {
err = sk->sk_prot->get_port(sk, snum);
if (err) {

2
net/ipv6/datagram.c
View File

@@ -524,7 +524,7 @@ int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len, int *addr_len)
} else {
ipv6_addr_set_v4mapped(ip_hdr(skb)->saddr,
&sin->sin6_addr);
if (inet_sk(sk)->cmsg_flags)
if (inet_cmsg_flags(inet_sk(sk)))
ip_cmsg_recv(msg, skb);
}
}

5
net/ipv6/ip6_output.c
View File

@@ -1591,7 +1591,7 @@ static int __ip6_append_data(struct sock *sk,
}
}
} else if ((flags & MSG_SPLICE_PAGES) && length) {
if (inet_sk(sk)->hdrincl)
if (inet_test_bit(HDRINCL, sk))
return -EPERM;
if (rt->dst.dev->features & NETIF_F_SG &&
getfrag == ip_generic_getfrag)
@@ -1995,7 +1995,8 @@ struct sk_buff *__ip6_make_skb(struct sock *sk,
struct inet6_dev *idev = ip6_dst_idev(skb_dst(skb));
u8 icmp6_type;
if (sk->sk_socket->type == SOCK_RAW && !inet_sk(sk)->hdrincl)
if (sk->sk_socket->type == SOCK_RAW &&
!inet_test_bit(HDRINCL, sk))
icmp6_type = fl6->fl6_icmp_type;
else
icmp6_type = icmp6_hdr(skb)->icmp6_type;

12
net/ipv6/ipv6_sockglue.c
View File

@@ -102,7 +102,7 @@ int ip6_ra_control(struct sock *sk, int sel)
struct ipv6_txoptions *ipv6_update_options(struct sock *sk,
struct ipv6_txoptions *opt)
{
if (inet_sk(sk)->is_icsk) {
if (inet_test_bit(IS_ICSK, sk)) {
if (opt &&
!((1 << sk->sk_state) & (TCPF_LISTEN | TCPF_CLOSE)) &&
inet_sk(sk)->inet_daddr != LOOPBACK4_IPV6) {
@@ -633,7 +633,7 @@ int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
if (optlen < sizeof(int))
goto e_inval;
/* we don't have a separate transparent bit for IPV6 we use the one in the IPv4 socket */
inet_sk(sk)->transparent = valbool;
inet_assign_bit(TRANSPARENT, sk, valbool);
retv = 0;
break;
@@ -641,7 +641,7 @@ int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
if (optlen < sizeof(int))
goto e_inval;
/* we also don't have a separate freebind bit for IPV6 */
inet_sk(sk)->freebind = valbool;
inet_assign_bit(FREEBIND, sk, valbool);
retv = 0;
break;
@@ -831,7 +831,7 @@ int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
goto e_inval;
retv = -EPROTO;
if (inet_sk(sk)->is_icsk)
if (inet_test_bit(IS_ICSK, sk))
break;
retv = -EFAULT;
@@ -1330,11 +1330,11 @@ int do_ipv6_getsockopt(struct sock *sk, int level, int optname,
}
case IPV6_TRANSPARENT:
val = inet_sk(sk)->transparent;
val = inet_test_bit(TRANSPARENT, sk);
break;
case IPV6_FREEBIND:
val = inet_sk(sk)->freebind;
val = inet_test_bit(FREEBIND, sk);
break;
case IPV6_RECVORIGDSTADDR:

16
net/ipv6/raw.c
View File

@@ -291,7 +291,6 @@ static void rawv6_err(struct sock *sk, struct sk_buff *skb,
struct inet6_skb_parm *opt,
u8 type, u8 code, int offset, __be32 info)
{
struct inet_sock *inet = inet_sk(sk);
struct ipv6_pinfo *np = inet6_sk(sk);
int err;
int harderr;
@@ -315,7 +314,7 @@ static void rawv6_err(struct sock *sk, struct sk_buff *skb,
}
if (np->recverr) {
u8 *payload = skb->data;
if (!inet->hdrincl)
if (!inet_test_bit(HDRINCL, sk))
payload += offset;
ipv6_icmp_error(sk, skb, err, 0, ntohl(info), payload);
}
@@ -406,7 +405,7 @@ int rawv6_rcv(struct sock *sk, struct sk_buff *skb)
skb->len,
inet->inet_num, 0));
if (inet->hdrincl) {
if (inet_test_bit(HDRINCL, sk)) {
if (skb_checksum_complete(skb)) {
atomic_inc(&sk->sk_drops);
kfree_skb_reason(skb, SKB_DROP_REASON_SKB_CSUM);
@@ -762,12 +761,7 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
if (msg->msg_flags & MSG_OOB)
return -EOPNOTSUPP;
/* hdrincl should be READ_ONCE(inet->hdrincl)
* but READ_ONCE() doesn't work with bit fields.
* Doing this indirectly yields the same result.
*/
hdrincl = inet->hdrincl;
hdrincl = READ_ONCE(hdrincl);
hdrincl = inet_test_bit(HDRINCL, sk);
/*
* Get and verify the address.
@@ -1000,7 +994,7 @@ static int do_rawv6_setsockopt(struct sock *sk, int level, int optname,
case IPV6_HDRINCL:
if (sk->sk_type != SOCK_RAW)
return -EINVAL;
inet_sk(sk)->hdrincl = !!val;
inet_assign_bit(HDRINCL, sk, val);
return 0;
case IPV6_CHECKSUM:
if (inet_sk(sk)->inet_num == IPPROTO_ICMPV6 &&
@@ -1068,7 +1062,7 @@ static int do_rawv6_getsockopt(struct sock *sk, int level, int optname,
switch (optname) {
case IPV6_HDRINCL:
val = inet_sk(sk)->hdrincl;
val = inet_test_bit(HDRINCL, sk);
break;
case IPV6_CHECKSUM:
/*

2
net/ipv6/udp.c
View File

@@ -420,7 +420,7 @@ int udpv6_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
ip6_datagram_recv_common_ctl(sk, msg, skb);
if (is_udp4) {
if (inet->cmsg_flags)
if (inet_cmsg_flags(inet))
ip_cmsg_recv_offset(msg, sk, skb,
sizeof(struct udphdr), off);
} else {

2
net/l2tp/l2tp_ip.c
View File

@@ -552,7 +552,7 @@ static int l2tp_ip_recvmsg(struct sock *sk, struct msghdr *msg,
memset(&sin->sin_zero, 0, sizeof(sin->sin_zero));
*addr_len = sizeof(*sin);
}
if (inet->cmsg_flags)
if (inet_cmsg_flags(inet))
ip_cmsg_recv(msg, skb);
if (flags & MSG_TRUNC)
copied = skb->len;

10
net/mptcp/protocol.c
View File

@@ -1686,7 +1686,7 @@ static int mptcp_sendmsg_fastopen(struct sock *sk, struct msghdr *msg,
if (!mptcp_disconnect(sk, 0))
sk->sk_socket->state = SS_UNCONNECTED;
}
inet_sk(sk)->defer_connect = 0;
inet_clear_bit(DEFER_CONNECT, sk);
return ret;
}
@@ -1704,7 +1704,8 @@ static int mptcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
lock_sock(sk);
if (unlikely(inet_sk(sk)->defer_connect || msg->msg_flags & MSG_FASTOPEN)) {
if (unlikely(inet_test_bit(DEFER_CONNECT, sk) ||
msg->msg_flags & MSG_FASTOPEN)) {
int copied_syn = 0;
ret = mptcp_sendmsg_fastopen(sk, msg, len, &copied_syn);
@@ -3601,7 +3602,7 @@ static int mptcp_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len)
if (err < 0)
goto out;
inet_sk(sk)->defer_connect = inet_sk(ssk)->defer_connect;
inet_assign_bit(DEFER_CONNECT, sk, inet_test_bit(DEFER_CONNECT, ssk));
out:
if (!msk->fastopening)
@@ -3827,7 +3828,8 @@ static __poll_t mptcp_poll(struct file *file, struct socket *sock,
mask |= EPOLLOUT | EPOLLWRNORM;
else
mask |= mptcp_check_writeable(msk);
} else if (state == TCP_SYN_SENT && inet_sk(sk)->defer_connect) {
} else if (state == TCP_SYN_SENT &&
inet_test_bit(DEFER_CONNECT, sk)) {
/* cf tcp_poll() note about TFO */
mask |= EPOLLOUT | EPOLLWRNORM;
}

18
net/mptcp/sockopt.c
View File

@@ -416,10 +416,12 @@ static int mptcp_setsockopt_v6(struct mptcp_sock *msk, int optname,
sk->sk_ipv6only = ssk->sk_ipv6only;
break;
case IPV6_TRANSPARENT:
inet_sk(sk)->transparent = inet_sk(ssk)->transparent;
inet_assign_bit(TRANSPARENT, sk,
inet_test_bit(TRANSPARENT, ssk));
break;
case IPV6_FREEBIND:
inet_sk(sk)->freebind = inet_sk(ssk)->freebind;
inet_assign_bit(FREEBIND, sk,
inet_test_bit(FREEBIND, ssk));
break;
}
@@ -684,7 +686,6 @@ static int mptcp_setsockopt_sol_ip_set_transparent(struct mptcp_sock *msk, int o
sockptr_t optval, unsigned int optlen)
{
struct sock *sk = (struct sock *)msk;
struct inet_sock *issk;
struct sock *ssk;
int err;
@@ -700,14 +701,13 @@ static int mptcp_setsockopt_sol_ip_set_transparent(struct mptcp_sock *msk, int o
return PTR_ERR(ssk);
}
issk = inet_sk(ssk);
switch (optname) {
case IP_FREEBIND:
issk->freebind = inet_sk(sk)->freebind;
inet_assign_bit(FREEBIND, ssk, inet_test_bit(FREEBIND, sk));
break;
case IP_TRANSPARENT:
issk->transparent = inet_sk(sk)->transparent;
inet_assign_bit(TRANSPARENT, ssk,
inet_test_bit(TRANSPARENT, sk));
break;
default:
release_sock(sk);
@@ -1440,8 +1440,8 @@ static void sync_socket_options(struct mptcp_sock *msk, struct sock *ssk)
__tcp_sock_set_cork(ssk, !!msk->cork);
__tcp_sock_set_nodelay(ssk, !!msk->nodelay);
inet_sk(ssk)->transparent = inet_sk(sk)->transparent;
inet_sk(ssk)->freebind = inet_sk(sk)->freebind;
inet_assign_bit(TRANSPARENT, ssk, inet_test_bit(TRANSPARENT, sk));
inet_assign_bit(FREEBIND, ssk, inet_test_bit(FREEBIND, sk));
}
static void __mptcp_sockopt_sync(struct mptcp_sock *msk, struct sock *ssk)

4
net/netfilter/ipvs/ip_vs_core.c
View File

@@ -1346,7 +1346,7 @@ ip_vs_out_hook(void *priv, struct sk_buff *skb, const struct nf_hook_state *stat
if (unlikely(sk && hooknum == NF_INET_LOCAL_OUT &&
af == AF_INET)) {
if (sk->sk_family == PF_INET && inet_sk(sk)->nodefrag)
if (sk->sk_family == PF_INET && inet_test_bit(NODEFRAG, sk))
return NF_ACCEPT;
}
@@ -1946,7 +1946,7 @@ ip_vs_in_hook(void *priv, struct sk_buff *skb, const struct nf_hook_state *state
if (unlikely(sk && hooknum == NF_INET_LOCAL_OUT &&
af == AF_INET)) {
if (sk->sk_family == PF_INET && inet_sk(sk)->nodefrag)
if (sk->sk_family == PF_INET && inet_test_bit(NODEFRAG, sk))
return NF_ACCEPT;
}

4
net/netfilter/ipvs/ip_vs_sync.c
View File

@@ -1297,11 +1297,9 @@ static void set_sock_size(struct sock *sk, int mode, int val)
*/
static void set_mcast_loop(struct sock *sk, u_char loop)
{
struct inet_sock *inet = inet_sk(sk);
/* setsockopt(sock, SOL_IP, IP_MULTICAST_LOOP, &loop, sizeof(loop)); */
lock_sock(sk);
inet->mc_loop = loop ? 1 : 0;
inet_assign_bit(MC_LOOP, sk, loop);
#ifdef CONFIG_IP_VS_IPV6
if (sk->sk_family == AF_INET6) {
struct ipv6_pinfo *np = inet6_sk(sk);

2
net/sctp/input.c
View File

@@ -581,7 +581,7 @@ static void sctp_v4_err_handle(struct sctp_transport *t, struct sk_buff *skb,
default:
return;
}
if (!sock_owned_by_user(sk) && inet_sk(sk)->recverr) {
if (!sock_owned_by_user(sk) && inet_test_bit(RECVERR, sk)) {
sk->sk_err = err;
sk_error_report(sk);
} else { /* Only an error on timeout */

2
net/sctp/protocol.c
View File

@@ -360,7 +360,7 @@ static int sctp_v4_available(union sctp_addr *addr, struct sctp_sock *sp)
ret = inet_addr_type_table(net, addr->v4.sin_addr.s_addr, tb_id);
if (addr->v4.sin_addr.s_addr != htonl(INADDR_ANY) &&
ret != RTN_LOCAL &&
!sp->inet.freebind &&
!inet_test_bit(FREEBIND, sk) &&
!READ_ONCE(net->ipv4.sysctl_ip_nonlocal_bind))
return 0;

2
net/sctp/socket.c
View File

@@ -9482,7 +9482,7 @@ void sctp_copy_sock(struct sock *newsk, struct sock *sk,
newinet->inet_id = get_random_u16();
newinet->uc_ttl = inet->uc_ttl;
newinet->mc_loop = 1;
inet_set_bit(MC_LOOP, newsk);
newinet->mc_ttl = 1;
newinet->mc_index = 0;
newinet->mc_list = NULL;