From a1cdec57e03a1352e92fbbe7974039dda4efcec0 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 17 Feb 2022 09:05:02 -0800 Subject: net-timestamp: convert sk->sk_tskey to atomic_t UDP sendmsg() can be lockless, this is causing all kinds of data races. This patch converts sk->sk_tskey to remove one of these races. BUG: KCSAN: data-race in __ip_append_data / __ip_append_data read to 0xffff8881035d4b6c of 4 bytes by task 8877 on cpu 1: __ip_append_data+0x1c1/0x1de0 net/ipv4/ip_output.c:994 ip_make_skb+0x13f/0x2d0 net/ipv4/ip_output.c:1636 udp_sendmsg+0x12bd/0x14c0 net/ipv4/udp.c:1249 inet_sendmsg+0x5f/0x80 net/ipv4/af_inet.c:819 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmmsg+0x267/0x4c0 net/socket.c:2553 __do_sys_sendmmsg net/socket.c:2582 [inline] __se_sys_sendmmsg net/socket.c:2579 [inline] __x64_sys_sendmmsg+0x53/0x60 net/socket.c:2579 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae write to 0xffff8881035d4b6c of 4 bytes by task 8880 on cpu 0: __ip_append_data+0x1d8/0x1de0 net/ipv4/ip_output.c:994 ip_make_skb+0x13f/0x2d0 net/ipv4/ip_output.c:1636 udp_sendmsg+0x12bd/0x14c0 net/ipv4/udp.c:1249 inet_sendmsg+0x5f/0x80 net/ipv4/af_inet.c:819 sock_sendmsg_nosec net/socket.c:705 [inline] sock_sendmsg net/socket.c:725 [inline] ____sys_sendmsg+0x39a/0x510 net/socket.c:2413 ___sys_sendmsg net/socket.c:2467 [inline] __sys_sendmmsg+0x267/0x4c0 net/socket.c:2553 __do_sys_sendmmsg net/socket.c:2582 [inline] __se_sys_sendmmsg net/socket.c:2579 [inline] __x64_sys_sendmmsg+0x53/0x60 net/socket.c:2579 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x44/0xd0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae value changed: 0x0000054d -> 0x0000054e Reported by Kernel Concurrency Sanitizer on: CPU: 0 PID: 8880 Comm: syz-executor.5 Not tainted 5.17.0-rc2-syzkaller-00167-gdcb85f85fa6f-dirty #0 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 Fixes: 09c2d251b707 ("net-timestamp: add key to disambiguate concurrent datagrams") Signed-off-by: Eric Dumazet Cc: Willem de Bruijn Reported-by: syzbot Signed-off-by: David S. Miller --- include/net/sock.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/net') diff --git a/include/net/sock.h b/include/net/sock.h index ff9b508d9c5f..50aecd28b355 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -507,7 +507,7 @@ struct sock { #endif u16 sk_tsflags; u8 sk_shutdown; - u32 sk_tskey; + atomic_t sk_tskey; atomic_t sk_zckey; u8 sk_clockid; @@ -2667,7 +2667,7 @@ static inline void _sock_tx_timestamp(struct sock *sk, __u16 tsflags, __sock_tx_timestamp(tsflags, tx_flags); if (tsflags & SOF_TIMESTAMPING_OPT_ID && tskey && tsflags & SOF_TIMESTAMPING_TX_RECORD_MASK) - *tskey = sk->sk_tskey++; + *tskey = atomic_inc_return(&sk->sk_tskey) - 1; } if (unlikely(sock_flag(sk, SOCK_WIFI_STATUS))) *tx_flags |= SKBTX_WIFI_STATUS; -- cgit v1.2.3 From 5486f5bf790b5c664913076c3194b8f916a5c7ad Mon Sep 17 00:00:00 2001 From: Christophe Leroy Date: Thu, 17 Feb 2022 14:35:49 +0100 Subject: net: Force inlining of checksum functions in net/checksum.h All functions defined as static inline in net/checksum.h are meant to be inlined for performance reason. But since commit ac7c3e4ff401 ("compiler: enable CONFIG_OPTIMIZE_INLINING forcibly") the compiler is allowed to uninline functions when it wants. Fair enough in the general case, but for tiny performance critical checksum helpers that's counter-productive. The problem mainly arises when selecting CONFIG_CC_OPTIMISE_FOR_SIZE, Those helpers being 'static inline' in header files you suddenly find them duplicated many times in the resulting vmlinux. Here is a typical exemple when building powerpc pmac32_defconfig with CONFIG_CC_OPTIMISE_FOR_SIZE. csum_sub() appears 4 times: c04a23cc : c04a23cc: 7c 84 20 f8 not r4,r4 c04a23d0: 7c 63 20 14 addc r3,r3,r4 c04a23d4: 7c 63 01 94 addze r3,r3 c04a23d8: 4e 80 00 20 blr ... c04a2ce8: 4b ff f6 e5 bl c04a23cc ... c04a2d2c: 4b ff f6 a1 bl c04a23cc ... c04a2d54: 4b ff f6 79 bl c04a23cc ... c04a754c : c04a754c: 7c 84 20 f8 not r4,r4 c04a7550: 7c 63 20 14 addc r3,r3,r4 c04a7554: 7c 63 01 94 addze r3,r3 c04a7558: 4e 80 00 20 blr ... c04ac930: 4b ff ac 1d bl c04a754c ... c04ad264: 4b ff a2 e9 bl c04a754c ... c04e3b08 : c04e3b08: 7c 84 20 f8 not r4,r4 c04e3b0c: 7c 63 20 14 addc r3,r3,r4 c04e3b10: 7c 63 01 94 addze r3,r3 c04e3b14: 4e 80 00 20 blr ... c04e5788: 4b ff e3 81 bl c04e3b08 ... c04e65c8: 4b ff d5 41 bl c04e3b08 ... c0512d34 : c0512d34: 7c 84 20 f8 not r4,r4 c0512d38: 7c 63 20 14 addc r3,r3,r4 c0512d3c: 7c 63 01 94 addze r3,r3 c0512d40: 4e 80 00 20 blr ... c0512dfc: 4b ff ff 39 bl c0512d34 ... c05138bc: 4b ff f4 79 bl c0512d34 ... Restore the expected behaviour by using __always_inline for all functions defined in net/checksum.h vmlinux size is even reduced by 256 bytes with this patch: text data bss dec hex filename 6980022 2515362 194384 9689768 93daa8 vmlinux.before 6979862 2515266 194384 9689512 93d9a8 vmlinux.now Fixes: ac7c3e4ff401 ("compiler: enable CONFIG_OPTIMIZE_INLINING forcibly") Cc: Masahiro Yamada Cc: Nick Desaulniers Cc: Andrew Morton Signed-off-by: Christophe Leroy Signed-off-by: David S. Miller --- include/net/checksum.h | 47 ++++++++++++++++++++++++----------------------- 1 file changed, 24 insertions(+), 23 deletions(-) (limited to 'include/net') diff --git a/include/net/checksum.h b/include/net/checksum.h index 5218041e5c8f..02d0c2d01014 100644 --- a/include/net/checksum.h +++ b/include/net/checksum.h @@ -22,7 +22,7 @@ #include #ifndef _HAVE_ARCH_COPY_AND_CSUM_FROM_USER -static inline +static __always_inline __wsum csum_and_copy_from_user (const void __user *src, void *dst, int len) { @@ -33,7 +33,7 @@ __wsum csum_and_copy_from_user (const void __user *src, void *dst, #endif #ifndef HAVE_CSUM_COPY_USER -static __inline__ __wsum csum_and_copy_to_user +static __always_inline __wsum csum_and_copy_to_user (const void *src, void __user *dst, int len) { __wsum sum = csum_partial(src, len, ~0U); @@ -45,7 +45,7 @@ static __inline__ __wsum csum_and_copy_to_user #endif #ifndef _HAVE_ARCH_CSUM_AND_COPY -static inline __wsum +static __always_inline __wsum csum_partial_copy_nocheck(const void *src, void *dst, int len) { memcpy(dst, src, len); @@ -54,7 +54,7 @@ csum_partial_copy_nocheck(const void *src, void *dst, int len) #endif #ifndef HAVE_ARCH_CSUM_ADD -static inline __wsum csum_add(__wsum csum, __wsum addend) +static __always_inline __wsum csum_add(__wsum csum, __wsum addend) { u32 res = (__force u32)csum; res += (__force u32)addend; @@ -62,12 +62,12 @@ static inline __wsum csum_add(__wsum csum, __wsum addend) } #endif -static inline __wsum csum_sub(__wsum csum, __wsum addend) +static __always_inline __wsum csum_sub(__wsum csum, __wsum addend) { return csum_add(csum, ~addend); } -static inline __sum16 csum16_add(__sum16 csum, __be16 addend) +static __always_inline __sum16 csum16_add(__sum16 csum, __be16 addend) { u16 res = (__force u16)csum; @@ -75,12 +75,12 @@ static inline __sum16 csum16_add(__sum16 csum, __be16 addend) return (__force __sum16)(res + (res < (__force u16)addend)); } -static inline __sum16 csum16_sub(__sum16 csum, __be16 addend) +static __always_inline __sum16 csum16_sub(__sum16 csum, __be16 addend) { return csum16_add(csum, ~addend); } -static inline __wsum csum_shift(__wsum sum, int offset) +static __always_inline __wsum csum_shift(__wsum sum, int offset) { /* rotate sum to align it with a 16b boundary */ if (offset & 1) @@ -88,42 +88,43 @@ static inline __wsum csum_shift(__wsum sum, int offset) return sum; } -static inline __wsum +static __always_inline __wsum csum_block_add(__wsum csum, __wsum csum2, int offset) { return csum_add(csum, csum_shift(csum2, offset)); } -static inline __wsum +static __always_inline __wsum csum_block_add_ext(__wsum csum, __wsum csum2, int offset, int len) { return csum_block_add(csum, csum2, offset); } -static inline __wsum +static __always_inline __wsum csum_block_sub(__wsum csum, __wsum csum2, int offset) { return csum_block_add(csum, ~csum2, offset); } -static inline __wsum csum_unfold(__sum16 n) +static __always_inline __wsum csum_unfold(__sum16 n) { return (__force __wsum)n; } -static inline __wsum csum_partial_ext(const void *buff, int len, __wsum sum) +static __always_inline +__wsum csum_partial_ext(const void *buff, int len, __wsum sum) { return csum_partial(buff, len, sum); } #define CSUM_MANGLED_0 ((__force __sum16)0xffff) -static inline void csum_replace_by_diff(__sum16 *sum, __wsum diff) +static __always_inline void csum_replace_by_diff(__sum16 *sum, __wsum diff) { *sum = csum_fold(csum_add(diff, ~csum_unfold(*sum))); } -static inline void csum_replace4(__sum16 *sum, __be32 from, __be32 to) +static __always_inline void csum_replace4(__sum16 *sum, __be32 from, __be32 to) { __wsum tmp = csum_sub(~csum_unfold(*sum), (__force __wsum)from); @@ -136,7 +137,7 @@ static inline void csum_replace4(__sum16 *sum, __be32 from, __be32 to) * m : old value of a 16bit field * m' : new value of a 16bit field */ -static inline void csum_replace2(__sum16 *sum, __be16 old, __be16 new) +static __always_inline void csum_replace2(__sum16 *sum, __be16 old, __be16 new) { *sum = ~csum16_add(csum16_sub(~(*sum), old), new); } @@ -150,16 +151,16 @@ void inet_proto_csum_replace16(__sum16 *sum, struct sk_buff *skb, void inet_proto_csum_replace_by_diff(__sum16 *sum, struct sk_buff *skb, __wsum diff, bool pseudohdr); -static inline void inet_proto_csum_replace2(__sum16 *sum, struct sk_buff *skb, - __be16 from, __be16 to, - bool pseudohdr) +static __always_inline +void inet_proto_csum_replace2(__sum16 *sum, struct sk_buff *skb, + __be16 from, __be16 to, bool pseudohdr) { inet_proto_csum_replace4(sum, skb, (__force __be32)from, (__force __be32)to, pseudohdr); } -static inline __wsum remcsum_adjust(void *ptr, __wsum csum, - int start, int offset) +static __always_inline __wsum remcsum_adjust(void *ptr, __wsum csum, + int start, int offset) { __sum16 *psum = (__sum16 *)(ptr + offset); __wsum delta; @@ -175,12 +176,12 @@ static inline __wsum remcsum_adjust(void *ptr, __wsum csum, return delta; } -static inline void remcsum_unadjust(__sum16 *psum, __wsum delta) +static __always_inline void remcsum_unadjust(__sum16 *psum, __wsum delta) { *psum = csum_fold(csum_sub(delta, (__force __wsum)*psum)); } -static inline __wsum wsum_negate(__wsum val) +static __always_inline __wsum wsum_negate(__wsum val) { return (__force __wsum)-((__force u32)val); } -- cgit v1.2.3 From b1a5983f56e371046dcf164f90bfaf704d2b89f6 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Thu, 17 Feb 2022 23:41:20 +0100 Subject: netfilter: nf_tables_offload: incorrect flow offload action array size immediate verdict expression needs to allocate one slot in the flow offload action array, however, immediate data expression does not need to do so. fwd and dup expression need to allocate one slot, this is missing. Add a new offload_action interface to report if this expression needs to allocate one slot in the flow offload action array. Fixes: be2861dc36d7 ("netfilter: nft_{fwd,dup}_netdev: add offload support") Reported-and-tested-by: Nick Gregory Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_tables.h | 2 +- include/net/netfilter/nf_tables_offload.h | 2 -- 2 files changed, 1 insertion(+), 3 deletions(-) (limited to 'include/net') diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index eaf55da9a205..c4c0861deac1 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -905,9 +905,9 @@ struct nft_expr_ops { int (*offload)(struct nft_offload_ctx *ctx, struct nft_flow_rule *flow, const struct nft_expr *expr); + bool (*offload_action)(const struct nft_expr *expr); void (*offload_stats)(struct nft_expr *expr, const struct flow_stats *stats); - u32 offload_flags; const struct nft_expr_type *type; void *data; }; diff --git a/include/net/netfilter/nf_tables_offload.h b/include/net/netfilter/nf_tables_offload.h index f9d95ff82df8..797147843958 100644 --- a/include/net/netfilter/nf_tables_offload.h +++ b/include/net/netfilter/nf_tables_offload.h @@ -67,8 +67,6 @@ struct nft_flow_rule { struct flow_rule *rule; }; -#define NFT_OFFLOAD_F_ACTION (1 << 0) - void nft_flow_rule_set_addr_type(struct nft_flow_rule *flow, enum flow_dissector_key_id addr_type); -- cgit v1.2.3 From d9b5ae5c1b241b91480aa30408be12fe91af834a Mon Sep 17 00:00:00 2001 From: Paul Blakey Date: Wed, 23 Feb 2022 18:34:16 +0200 Subject: openvswitch: Fix setting ipv6 fields causing hw csum failure Ipv6 ttl, label and tos fields are modified without first pulling/pushing the ipv6 header, which would have updated the hw csum (if available). This might cause csum validation when sending the packet to the stack, as can be seen in the trace below. Fix this by updating skb->csum if available. Trace resulted by ipv6 ttl dec and then sending packet to conntrack [actions: set(ipv6(hlimit=63)),ct(zone=99)]: [295241.900063] s_pf0vf2: hw csum failure [295241.923191] Call Trace: [295241.925728] [295241.927836] dump_stack+0x5c/0x80 [295241.931240] __skb_checksum_complete+0xac/0xc0 [295241.935778] nf_conntrack_tcp_packet+0x398/0xba0 [nf_conntrack] [295241.953030] nf_conntrack_in+0x498/0x5e0 [nf_conntrack] [295241.958344] __ovs_ct_lookup+0xac/0x860 [openvswitch] [295241.968532] ovs_ct_execute+0x4a7/0x7c0 [openvswitch] [295241.979167] do_execute_actions+0x54a/0xaa0 [openvswitch] [295242.001482] ovs_execute_actions+0x48/0x100 [openvswitch] [295242.006966] ovs_dp_process_packet+0x96/0x1d0 [openvswitch] [295242.012626] ovs_vport_receive+0x6c/0xc0 [openvswitch] [295242.028763] netdev_frame_hook+0xc0/0x180 [openvswitch] [295242.034074] __netif_receive_skb_core+0x2ca/0xcb0 [295242.047498] netif_receive_skb_internal+0x3e/0xc0 [295242.052291] napi_gro_receive+0xba/0xe0 [295242.056231] mlx5e_handle_rx_cqe_mpwrq_rep+0x12b/0x250 [mlx5_core] [295242.062513] mlx5e_poll_rx_cq+0xa0f/0xa30 [mlx5_core] [295242.067669] mlx5e_napi_poll+0xe1/0x6b0 [mlx5_core] [295242.077958] net_rx_action+0x149/0x3b0 [295242.086762] __do_softirq+0xd7/0x2d6 [295242.090427] irq_exit+0xf7/0x100 [295242.093748] do_IRQ+0x7f/0xd0 [295242.096806] common_interrupt+0xf/0xf [295242.100559] [295242.102750] RIP: 0033:0x7f9022e88cbd [295242.125246] RSP: 002b:00007f9022282b20 EFLAGS: 00000246 ORIG_RAX: ffffffffffffffda [295242.132900] RAX: 0000000000000005 RBX: 0000000000000010 RCX: 0000000000000000 [295242.140120] RDX: 00007f9022282ba8 RSI: 00007f9022282a30 RDI: 00007f9014005c30 [295242.147337] RBP: 00007f9014014d60 R08: 0000000000000020 R09: 00007f90254a8340 [295242.154557] R10: 00007f9022282a28 R11: 0000000000000246 R12: 0000000000000000 [295242.161775] R13: 00007f902308c000 R14: 000000000000002b R15: 00007f9022b71f40 Fixes: 3fdbd1ce11e5 ("openvswitch: add ipv6 'set' action") Signed-off-by: Paul Blakey Link: https://lore.kernel.org/r/20220223163416.24096-1-paulb@nvidia.com Signed-off-by: Jakub Kicinski --- include/net/checksum.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include/net') diff --git a/include/net/checksum.h b/include/net/checksum.h index 02d0c2d01014..79c67f14c448 100644 --- a/include/net/checksum.h +++ b/include/net/checksum.h @@ -142,6 +142,11 @@ static __always_inline void csum_replace2(__sum16 *sum, __be16 old, __be16 new) *sum = ~csum16_add(csum16_sub(~(*sum), old), new); } +static inline void csum_replace(__wsum *csum, __wsum old, __wsum new) +{ + *csum = csum_add(csum_sub(*csum, old), new); +} + struct sk_buff; void inet_proto_csum_replace4(__sum16 *sum, struct sk_buff *skb, __be32 from, __be32 to, bool pseudohdr); -- cgit v1.2.3