tcp: Follow up for DCCP removal.
Subsystems (10)
Kuniyuki Iwashima
reviewerv2 cover tcp: Follow up for DCCP removal.
As I mentioned in [0], TCP still has code for DCCP.
This series cleans up such leftovers.
[0]: https://patch.msgid.link/[email protected]
Changes:
v2:
* Fix up my email address in patch 1~3
* Drop patch 2
v1: https://lore.kernel.org/netdev/[email protected]/
Kuniyuki Iwashima (6):
tcp: Remove sk_protocol test for tcp_twsk_unique().
tcp: Remove timewait_sock_ops.twsk_destructor().
tcp: Remove hashinfo test for inet6?_lookup_run_sk_lookup().
tcp: Don't pass hashinfo to socket lookup helpers.
tcp: Don't pass hashinfo to inet_diag helpers.
tcp: Move TCP-specific diag functions to tcp_diag.c.
.../mellanox/mlx5/core/en_accel/ktls_rx.c | 9 +- .../net/ethernet/netronome/nfp/crypto/tls.c | 9 +- include/linux/inet_diag.h | 13 - include/net/inet6_hashtables.h | 18 +- include/net/inet_hashtables.h | 37 +- include/net/timewait_sock.h | 7 - net/core/filter.c | 5 +- net/ipv4/esp4.c | 4 +- net/ipv4/inet_diag.c | 479 ------------------ net/ipv4/inet_hashtables.c | 34 +- net/ipv4/inet_timewait_sock.c | 5 +- net/ipv4/netfilter/nf_socket_ipv4.c | 3 +- net/ipv4/netfilter/nf_tproxy_ipv4.c | 5 +- net/ipv4/tcp_diag.c | 463 ++++++++++++++++- net/ipv4/tcp_ipv4.c | 17 +- net/ipv4/tcp_minisocks.c | 1 - net/ipv4/tcp_offload.c | 3 +- net/ipv6/esp6.c | 4 +- net/ipv6/inet6_hashtables.c | 51 +- net/ipv6/netfilter/nf_socket_ipv6.c | 3 +- net/ipv6/netfilter/nf_tproxy_ipv6.c | 5 +- net/ipv6/tcp_ipv6.c | 15 +- net/ipv6/tcpv6_offload.c | 3 +- 23 files changed, 549 insertions(+), 644 deletions(-)
Signature · 1 line
2.51.0.rc2.233.g662b1ed5c5-goog
Kuniyuki Iwashima
reviewerPATCH v2 1/6 tcp: Remove sk_protocol test for tcp_twsk_unique().
Commit 383eed2de529 ("tcp: get rid of twsk_unique()") added sk->sk_protocol test in __inet_check_established() and __inet6_check_established() to remove twsk_unique() and call tcp_twsk_unique() directly.
DCCP has gone, and the condition is always true.
Let's remove the sk_protocol test.
--- net/ipv4/inet_hashtables.c | 3 +-- net/ipv6/inet6_hashtables.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-)
net/ipv4/inet_hashtables.c +1 -2
@@ -579,8 +579,7 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row,if (likely (inet_match (net, sk2, acookie, ports, dif, sdif))) {if (sk2->sk_state ==TCP_TIME_WAIT ) { tw =inet_twsk (sk2); -if (sk->sk_protocol ==IPPROTO_TCP && -tcp_twsk_unique (sk, sk2, twp)) +if (tcp_twsk_unique (sk, sk2, twp))break ; } goto not_unique;
net/ipv6/inet6_hashtables.c +1 -2
@@ -305,8 +305,7 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row, dif, sdif))) {if (sk2->sk_state ==TCP_TIME_WAIT ) { tw =inet_twsk (sk2); -if (sk->sk_protocol ==IPPROTO_TCP && -tcp_twsk_unique (sk, sk2, twp)) +if (tcp_twsk_unique (sk, sk2, twp))break ; } goto not_unique;
Signature · 1 line
2.51.0.rc2.233.g662b1ed5c5-goog
Kuniyuki Iwashima
reviewerPATCH v2 2/6 tcp: Remove timewait_sock_ops.twsk_destructor().
In reply to Kuniyuki IwashimaSince DCCP has been removed, sk->sk_prot->twsk_prot->twsk_destructor is always tcp_twsk_destructor().
Let's call tcp_twsk_destructor() directly in inet_twsk_free() and remove ->twsk_destructor().
While at it, tcp_twsk_destructor() is un-exported.
--- include/net/timewait_sock.h | 7 ------- net/ipv4/inet_timewait_sock.c | 5 +++-- net/ipv4/tcp_ipv4.c | 1 - net/ipv4/tcp_minisocks.c | 1 - net/ipv6/tcp_ipv6.c | 1 - 5 files changed, 3 insertions(+), 12 deletions(-)
include/net/timewait_sock.h +0 -7
@@ -15,13 +15,6 @@ struct timewait_sock_ops {struct kmem_cache *twsk_slab;char *twsk_slab_name; unsigned int twsk_obj_size; - void (*twsk_destructor)(struct sock *sk); }; -static inline void twsk_destructor (struct sock *sk) -{ -if (sk->sk_prot->twsk_prot->twsk_destructor !=NULL ) - sk->sk_prot->twsk_prot->twsk_destructor (sk); -} -#endif /* _TIMEWAIT_SOCK_H */
net/ipv4/inet_timewait_sock.c +3 -2
@@ -15,7 +15,7 @@#include <net/inet_hashtables.h> #include <net/inet_timewait_sock.h> #include <net/ip.h> - +#include <net/tcp.h> /** * inet_twsk_bind_unhash - unhash a timewait socket from bind hash
@@ -74,7 +74,8 @@ static void inet_twsk_kill(struct inet_timewait_sock *tw)void inet_twsk_free (struct inet_timewait_sock *tw) {struct module *owner = tw->tw_prot->owner; -twsk_destructor ((struct sock *)tw); + +tcp_twsk_destructor ((struct sock *)tw);kmem_cache_free (tw->tw_prot->twsk_prot->twsk_slab, tw);module_put (owner); }
net/ipv4/tcp_ipv4.c +0 -1
@@ -2459,7 +2459,6 @@ int tcp_v4_rcv(struct sk_buff *skb)static struct timewait_sock_ops tcp_timewait_sock_ops = { .twsk_obj_size =sizeof (struct tcp_timewait_sock ), - .twsk_destructor= tcp_twsk_destructor, };void inet_sk_rx_dst_set (struct sock *sk,const struct sk_buff *skb)
net/ipv4/tcp_minisocks.c +0 -1
@@ -401,7 +401,6 @@ void tcp_twsk_destructor(struct sock *sk)#endif tcp_ao_destroy_sock (sk, true); } -EXPORT_IPV6_MOD_GPL (tcp_twsk_destructor);void tcp_twsk_purge (struct list_head *net_exit_list) {
net/ipv6/tcp_ipv6.c +0 -1
@@ -2050,7 +2050,6 @@ void tcp_v6_early_demux(struct sk_buff *skb)static struct timewait_sock_ops tcp6_timewait_sock_ops = { .twsk_obj_size =sizeof (struct tcp6_timewait_sock ), - .twsk_destructor = tcp_twsk_destructor, }; INDIRECT_CALLABLE_SCOPE void tcp_v6_send_check(struct sock *sk, struct sk_buff *skb)
Signature · 1 line
2.51.0.rc2.233.g662b1ed5c5-goog
Kuniyuki Iwashima
reviewerPATCH v2 3/6 tcp: Remove hashinfo test for inet6?_lookup_run_sk_lookup().
In reply to Kuniyuki IwashimaCommit 6c886db2e78c ("net: remove duplicate sk_lookup helpers") started to check if hashinfo == net->ipv4.tcp_death_row.hashinfo in __inet_lookup_listener() and inet6_lookup_listener() and stopped invoking BPF sk_lookup prog for DCCP.
DCCP has gone and the condition is always true.
Let's remove the hashinfo test.
--- net/ipv4/inet_hashtables.c | 3 +-- net/ipv6/inet6_hashtables.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-)
net/ipv4/inet_hashtables.c +1 -2
@@ -436,8 +436,7 @@ struct sock *__inet_lookup_listener(const struct net *net,unsigned int hash2;/* Lookup redirect from BPF */ -if (static_branch_unlikely (&bpf_sk_lookup_enabled) && - hashinfo == net->ipv4.tcp_death_row.hashinfo) { +if (static_branch_unlikely (&bpf_sk_lookup_enabled)) { result =inet_lookup_run_sk_lookup (net,IPPROTO_TCP , skb, doff, saddr, sport, daddr, hnum, dif, inet_ehashfn);
net/ipv6/inet6_hashtables.c +1 -2
@@ -211,8 +211,7 @@ struct sock *inet6_lookup_listener(const struct net *net,unsigned int hash2;/* Lookup redirect from BPF */ -if (static_branch_unlikely (&bpf_sk_lookup_enabled) && - hashinfo == net->ipv4.tcp_death_row.hashinfo) { +if (static_branch_unlikely (&bpf_sk_lookup_enabled)) { result =inet6_lookup_run_sk_lookup (net,IPPROTO_TCP , skb, doff, saddr, sport, daddr, hnum, dif, inet6_ehashfn);
Signature · 1 line
2.51.0.rc2.233.g662b1ed5c5-goog
Kuniyuki Iwashima
reviewerPATCH v2 4/6 tcp: Don't pass hashinfo to socket lookup helpers.
In reply to Kuniyuki IwashimaThese socket lookup functions required struct inet_hashinfo because they are shared by TCP and DCCP.
* __inet_lookup_established()
* __inet_lookup_listener()
* __inet6_lookup_established()
* inet6_lookup_listener()
DCCP has gone, and we don't need to pass hashinfo down to them.
Let's fetch net->ipv4.tcp_death_row.hashinfo directly in the above 4 functions.
--- .../mellanox/mlx5/core/en_accel/ktls_rx.c | 9 ++-- .../net/ethernet/netronome/nfp/crypto/tls.c | 9 ++-- include/net/inet6_hashtables.h | 18 +++----- include/net/inet_hashtables.h | 37 +++++++-------- net/core/filter.c | 5 +-- net/ipv4/esp4.c | 4 +- net/ipv4/inet_diag.c | 6 +-- net/ipv4/inet_hashtables.c | 28 ++++++------ net/ipv4/netfilter/nf_socket_ipv4.c | 3 +- net/ipv4/netfilter/nf_tproxy_ipv4.c | 5 +-- net/ipv4/tcp_ipv4.c | 16 +++---- net/ipv4/tcp_offload.c | 3 +- net/ipv6/esp6.c | 4 +- net/ipv6/inet6_hashtables.c | 45 ++++++++++--------- net/ipv6/netfilter/nf_socket_ipv6.c | 3 +- net/ipv6/netfilter/nf_tproxy_ipv6.c | 5 +-- net/ipv6/tcp_ipv6.c | 14 +++--- net/ipv6/tcpv6_offload.c | 3 +- 18 files changed, 94 insertions(+), 123 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c +4 -5
@@ -498,9 +498,9 @@ static void resync_update_sn(struct mlx5e_rq *rq, struct sk_buff *skb) depth +=sizeof (struct iphdr ); th = (void *)iph +sizeof (struct iphdr ); - sk =inet_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, - iph->saddr, th->source, iph->daddr, - th->dest, netdev->ifindex); + sk =inet_lookup_established (net, iph->saddr, th->source, + iph->daddr, th->dest, + netdev->ifindex);#if IS_ENABLED (CONFIG_IPV6 ) } else {struct ipv6hdr *ipv6h = (struct ipv6hdr *)iph;
@@ -508,8 +508,7 @@ static void resync_update_sn(struct mlx5e_rq *rq, struct sk_buff *skb) depth +=sizeof (struct ipv6hdr ); th = (void *)ipv6h +sizeof (struct ipv6hdr ); - sk =__inet6_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, - &ipv6h->saddr, th->source, + sk =__inet6_lookup_established (net, &ipv6h->saddr, th->source, &ipv6h->daddr,ntohs (th->dest), netdev->ifindex,0 );#endif
drivers/net/ethernet/netronome/nfp/crypto/tls.c +4 -5
@@ -495,14 +495,13 @@ int nfp_net_tls_rx_resync_req(struct net_device *netdev,switch (ipv6h->version) {case 4 : - sk =inet_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, - iph->saddr, th->source, iph->daddr, - th->dest, netdev->ifindex); + sk =inet_lookup_established (net, iph->saddr, th->source, + iph->daddr, th->dest, + netdev->ifindex);break ;#if IS_ENABLED (CONFIG_IPV6 )case 6 : - sk =__inet6_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, - &ipv6h->saddr, th->source, + sk =__inet6_lookup_established (net, &ipv6h->saddr, th->source, &ipv6h->daddr,ntohs (th->dest), netdev->ifindex,0 );break ;
include/net/inet6_hashtables.h +6 -12
@@ -41,7 +41,6 @@ static inline unsigned int __inet6_ehashfn(const u32 lhash,* The sockhash lock must be held as a reader here. */ struct sock *__inet6_lookup_established(const struct net *net, - struct inet_hashinfo *hashinfo, const struct in6_addr *saddr, const __be16 sport, const struct in6_addr *daddr,
@@ -65,7 +64,6 @@ struct sock *inet6_lookup_reuseport(const struct net *net, struct sock *sk, inet6_ehashfn_t *ehashfn); struct sock *inet6_lookup_listener(const struct net *net, - struct inet_hashinfo *hashinfo, struct sk_buff *skb, int doff, const struct in6_addr *saddr, const __be16 sport,
@@ -83,7 +81,6 @@ struct sock *inet6_lookup_run_sk_lookup(const struct net *net, inet6_ehashfn_t *ehashfn); static inline struct sock *__inet6_lookup(const struct net *net, - struct inet_hashinfo *hashinfo, struct sk_buff *skb, int doff, const struct in6_addr *saddr, const __be16 sport,
@@ -92,14 +89,14 @@ static inline struct sock *__inet6_lookup(const struct net *net, const int dif, const int sdif, bool *refcounted) { - struct sock *sk = __inet6_lookup_established(net, hashinfo, saddr, - sport, daddr, hnum, + struct sock *sk = __inet6_lookup_established(net, saddr, sport, + daddr, hnum, dif, sdif); *refcounted = true;if (sk)return sk; *refcounted = false; -return inet6_lookup_listener (net, hashinfo, skb, doff, saddr, sport, +return inet6_lookup_listener (net, skb, doff, saddr, sport, daddr, hnum, dif, sdif); }
@@ -143,8 +140,7 @@ struct sock *inet6_steal_sock(struct net *net, struct sk_buff *skb, int doff,return reuse_sk; } -static inline struct sock *__inet6_lookup_skb(struct inet_hashinfo *hashinfo, - struct sk_buff *skb, int doff, +static inline struct sock *__inet6_lookup_skb(struct sk_buff *skb, int doff, const __be16 sport, const __be16 dport, int iif, int sdif,
@@ -161,14 +157,12 @@ static inline struct sock *__inet6_lookup_skb(struct inet_hashinfo *hashinfo,if (sk)return sk; -return __inet6_lookup (net, hashinfo, skb, - doff, &ip6h->saddr, sport, +return __inet6_lookup (net, skb, doff, &ip6h->saddr, sport, &ip6h->daddr,ntohs (dport), iif, sdif, refcounted); } -struct sock *inet6_lookup (const struct net *net,struct inet_hashinfo *hashinfo, -struct sk_buff *skb,int doff, +struct sock *inet6_lookup (const struct net *net,struct sk_buff *skb,int doff,const struct in6_addr *saddr,const __be16 sport,const struct in6_addr *daddr,const __be16 dport,const int dif);
include/net/inet_hashtables.h +15 -22
@@ -294,7 +294,6 @@ int inet_hash(struct sock *sk);void inet_unhash (struct sock *sk); struct sock *__inet_lookup_listener(const struct net *net, - struct inet_hashinfo *hashinfo, struct sk_buff *skb, int doff, const __be32 saddr, const __be16 sport, const __be32 daddr,
@@ -302,12 +301,12 @@ struct sock *__inet_lookup_listener(const struct net *net, const int dif, const int sdif);static inline struct sock *inet_lookup_listener (struct net *net, -struct inet_hashinfo *hashinfo, -struct sk_buff *skb,int doff, -__be32 saddr,__be16 sport, -__be32 daddr,__be16 dport,int dif,int sdif) +struct sk_buff *skb,int doff, +__be32 saddr,__be16 sport, +__be32 daddr,__be16 dport, +int dif,int sdif) { -return __inet_lookup_listener (net, hashinfo, skb, doff, saddr, sport, +return __inet_lookup_listener (net, skb, doff, saddr, sport, daddr,ntohs (dport), dif, sdif); }
@@ -358,7 +357,6 @@ static inline bool inet_match(const struct net *net, const struct sock *sk,* not check it for lookups anymore, thanks Alexey. -DaveM */ struct sock *__inet_lookup_established (const struct net *net, -struct inet_hashinfo *hashinfo,const __be32 saddr,const __be16 sport,const __be32 daddr,const u16 hnum,const int dif,const int sdif);
@@ -384,18 +382,16 @@ struct sock *inet_lookup_run_sk_lookup(const struct net *net, __be32 daddr, u16 hnum, const int dif, inet_ehashfn_t *ehashfn); -static inline struct sock * -inet_lookup_established (struct net *net,struct inet_hashinfo *hashinfo, -const __be32 saddr,const __be16 sport, -const __be32 daddr,const __be16 dport, -const int dif) +static inline struct sock *inet_lookup_established (struct net *net, +const __be32 saddr,const __be16 sport, +const __be32 daddr,const __be16 dport, +const int dif) { -return __inet_lookup_established (net, hashinfo, saddr, sport, daddr, +return __inet_lookup_established (net, saddr, sport, daddr,ntohs (dport), dif,0 ); } static inline struct sock *__inet_lookup(struct net *net, - struct inet_hashinfo *hashinfo, struct sk_buff *skb, int doff, const __be32 saddr, const __be16 sport, const __be32 daddr, const __be16 dport,
@@ -405,18 +401,17 @@ static inline struct sock *__inet_lookup(struct net *net,u16 hnum =ntohs (dport);struct sock *sk; - sk =__inet_lookup_established (net, hashinfo, saddr, sport, + sk =__inet_lookup_established (net, saddr, sport, daddr, hnum, dif, sdif); *refcounted = true;if (sk)return sk; *refcounted = false; -return __inet_lookup_listener (net, hashinfo, skb, doff, saddr, +return __inet_lookup_listener (net, skb, doff, saddr, sport, daddr, hnum, dif, sdif); } static inline struct sock *inet_lookup(struct net *net, - struct inet_hashinfo *hashinfo, struct sk_buff *skb, int doff, const __be32 saddr, const __be16 sport, const __be32 daddr, const __be16 dport,
@@ -425,7 +420,7 @@ static inline struct sock *inet_lookup(struct net *net,struct sock *sk;bool refcounted; - sk =__inet_lookup (net, hashinfo, skb, doff, saddr, sport, daddr, + sk =__inet_lookup (net, skb, doff, saddr, sport, daddr, dport, dif,0 , &refcounted);if (sk && !refcounted && !refcount_inc_not_zero (&sk->sk_refcnt))
@@ -473,8 +468,7 @@ struct sock *inet_steal_sock(struct net *net, struct sk_buff *skb, int doff,return reuse_sk; } -static inline struct sock *__inet_lookup_skb(struct inet_hashinfo *hashinfo, - struct sk_buff *skb, +static inline struct sock *__inet_lookup_skb(struct sk_buff *skb, int doff, const __be16 sport, const __be16 dport,
@@ -492,8 +486,7 @@ static inline struct sock *__inet_lookup_skb(struct inet_hashinfo *hashinfo,if (sk)return sk; -return __inet_lookup (net, hashinfo, skb, - doff, iph->saddr, sport, +return __inet_lookup (net, skb, doff, iph->saddr, sport, iph->daddr, dport,inet_iif (skb), sdif, refcounted); }
net/core/filter.c +2 -3
@@ -6767,7 +6767,6 @@ static const struct bpf_func_proto bpf_lwt_seg6_adjust_srh_proto = {static struct sock *sk_lookup (struct net *net,struct bpf_sock_tuple *tuple,int dif,int sdif,u8 family,u8 proto) { -struct inet_hashinfo *hinfo = net->ipv4.tcp_death_row.hashinfo;bool refcounted = false;struct sock *sk =NULL ;
@@ -6776,7 +6775,7 @@ static struct sock *sk_lookup(struct net *net, struct bpf_sock_tuple *tuple,__be32 dst4 = tuple->ipv4.daddr;if (proto ==IPPROTO_TCP ) - sk =__inet_lookup (net, hinfo,NULL ,0 , + sk =__inet_lookup (net,NULL ,0 , src4, tuple->ipv4.sport, dst4, tuple->ipv4.dport, dif, sdif, &refcounted);
@@ -6790,7 +6789,7 @@ static struct sock *sk_lookup(struct net *net, struct bpf_sock_tuple *tuple,struct in6_addr *dst6 = (struct in6_addr *)&tuple->ipv6.daddr;if (proto ==IPPROTO_TCP ) - sk =__inet6_lookup (net, hinfo,NULL ,0 , + sk =__inet6_lookup (net,NULL ,0 , src6, tuple->ipv6.sport, dst6,ntohs (tuple->ipv6.dport), dif, sdif, &refcounted);
net/ipv4/esp4.c +2 -2
@@ -132,8 +132,8 @@ static struct sock *esp_find_tcp_sk(struct xfrm_state *x) dport = encap->encap_dport;spin_unlock_bh (&x->lock); - sk =inet_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, x->id.daddr.a4, - dport, x->props.saddr.a4, sport,0 ); + sk =inet_lookup_established (net, x->id.daddr.a4, dport, + x->props.saddr.a4, sport,0 );if (!sk)return ERR_PTR (-ENOENT );
net/ipv4/inet_diag.c +3 -3
@@ -526,18 +526,18 @@ struct sock *inet_diag_find_one_icsk(struct net *net,rcu_read_lock ();if (req->sdiag_family ==AF_INET ) - sk =inet_lookup (net, hashinfo,NULL ,0 , req->id.idiag_dst[0 ], + sk =inet_lookup (net,NULL ,0 , req->id.idiag_dst[0 ], req->id.idiag_dport, req->id.idiag_src[0 ], req->id.idiag_sport, req->id.idiag_if);#if IS_ENABLED (CONFIG_IPV6 ) else if (req->sdiag_family == AF_INET6) {if (ipv6_addr_v4mapped ((struct in6_addr *)req->id.idiag_dst) &&ipv6_addr_v4mapped ((struct in6_addr *)req->id.idiag_src)) - sk =inet_lookup (net, hashinfo,NULL ,0 , req->id.idiag_dst[3 ], + sk =inet_lookup (net,NULL ,0 , req->id.idiag_dst[3 ], req->id.idiag_dport, req->id.idiag_src[3 ], req->id.idiag_sport, req->id.idiag_if);else - sk = inet6_lookup(net, hashinfo, NULL, 0, + sk = inet6_lookup(net, NULL, 0, (struct in6_addr *)req->id.idiag_dst, req->id.idiag_dport, (struct in6_addr *)req->id.idiag_src,
net/ipv4/inet_hashtables.c +15 -13
@@ -425,13 +425,13 @@ struct sock *inet_lookup_run_sk_lookup(const struct net *net, }struct sock *__inet_lookup_listener (const struct net *net, -struct inet_hashinfo *hashinfo,struct sk_buff *skb,int doff,const __be32 saddr,__be16 sport,const __be32 daddr,const unsigned short hnum,const int dif,const int sdif) {struct inet_listen_hashbucket *ilb2; +struct inet_hashinfo *hashinfo;struct sock *result =NULL ;unsigned int hash2;
@@ -444,6 +444,7 @@ struct sock *__inet_lookup_listener(const struct net *net, goto done; } + hashinfo = net->ipv4.tcp_death_row.hashinfo; hash2 =ipv4_portaddr_hash (net, daddr, hnum); ilb2 =inet_lhash2_bucket (hashinfo, hash2);
@@ -489,21 +490,22 @@ void sock_edemux(struct sk_buff *skb)EXPORT_SYMBOL (sock_edemux);struct sock *__inet_lookup_established (const struct net *net, -struct inet_hashinfo *hashinfo, -const __be32 saddr,const __be16 sport, -const __be32 daddr,const u16 hnum, -const int dif,const int sdif) +const __be32 saddr,const __be16 sport, +const __be32 daddr,const u16 hnum, +const int dif,const int sdif) { -INET_ADDR_COOKIE (acookie, saddr, daddr);const __portpair ports =INET_COMBINED_PORTS (sport, hnum); -struct sock *sk; +INET_ADDR_COOKIE (acookie, saddr, daddr);const struct hlist_nulls_node *node; -/* Optimize here for direct hit, only listening connections can -* have wildcards anyways. -*/ -unsigned int hash =inet_ehashfn (net, daddr, hnum, saddr, sport); -unsigned int slot = hash & hashinfo->ehash_mask; -struct inet_ehash_bucket *head = &hashinfo->ehash[slot]; +struct inet_ehash_bucket *head; +struct inet_hashinfo *hashinfo; +unsigned int hash, slot; +struct sock *sk; + + hashinfo = net->ipv4.tcp_death_row.hashinfo; + hash =inet_ehashfn (net, daddr, hnum, saddr, sport); + slot = hash & hashinfo->ehash_mask; + head = &hashinfo->ehash[slot]; begin:sk_nulls_for_each_rcu (sk, node, &head->chain) {
net/ipv4/netfilter/nf_socket_ipv4.c +1 -2
@@ -71,8 +71,7 @@ nf_socket_get_sock_v4(struct net *net, struct sk_buff *skb, const int doff, {switch (protocol) {case IPPROTO_TCP : -return inet_lookup (net, net->ipv4.tcp_death_row.hashinfo, - skb, doff, saddr, sport, daddr, dport, +return inet_lookup (net, skb, doff, saddr, sport, daddr, dport, in->ifindex);case IPPROTO_UDP : return udp4_lib_lookup(net, saddr, sport, daddr, dport,
net/ipv4/netfilter/nf_tproxy_ipv4.c +2 -3
@@ -81,7 +81,6 @@ nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb,const struct net_device *in, const enum nf_tproxy_lookup_t lookup_type) { - struct inet_hashinfo *hinfo = net->ipv4.tcp_death_row.hashinfo; struct sock *sk;switch (protocol) {
@@ -95,7 +94,7 @@ nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb,switch (lookup_type) {case NF_TPROXY_LOOKUP_LISTENER : - sk =inet_lookup_listener (net, hinfo, skb, + sk =inet_lookup_listener (net, skb,ip_hdrlen (skb) +__tcp_hdrlen (hp), saddr, sport, daddr, dport, in->ifindex,0 );
@@ -109,7 +108,7 @@ nf_tproxy_get_sock_v4(struct net *net, struct sk_buff *skb,*/ break ;case NF_TPROXY_LOOKUP_ESTABLISHED : - sk =inet_lookup_established (net, hinfo, saddr, sport, + sk =inet_lookup_established (net, saddr, sport, daddr, dport, in->ifindex);break ;default :
net/ipv4/tcp_ipv4.c +5 -11
@@ -506,8 +506,7 @@ int tcp_v4_err(struct sk_buff *skb, u32 info)struct sock *sk;int err; - sk =__inet_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, - iph->daddr, th->dest, iph->saddr, + sk =__inet_lookup_established (net, iph->daddr, th->dest, iph->saddr,ntohs (th->source),inet_iif (skb),0 );if (!sk) {__ICMP_INC_STATS (net,ICMP_MIB_INERRORS );
@@ -823,8 +822,7 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb,* Incoming packet is checked with md5 hash with finding key, * no RST generated if md5 hash doesn't match. */ - sk1 =__inet_lookup_listener (net, net->ipv4.tcp_death_row.hashinfo, -NULL ,0 ,ip_hdr (skb)->saddr, + sk1 =__inet_lookup_listener (net,NULL ,0 ,ip_hdr (skb)->saddr, th->source,ip_hdr (skb)->daddr,ntohs (th->source), dif, sdif);/* don't send rst if it can't find key */
@@ -1992,8 +1990,7 @@ int tcp_v4_early_demux(struct sk_buff *skb)if (th->doff <sizeof (struct tcphdr ) /4 )return 0 ; - sk =__inet_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, - iph->saddr, th->source, + sk =__inet_lookup_established (net, iph->saddr, th->source, iph->daddr,ntohs (th->dest), skb->skb_iif,inet_sdif (skb));if (sk) {
@@ -2236,8 +2233,7 @@ int tcp_v4_rcv(struct sk_buff *skb) th = (const struct tcphdr *)skb->data; iph =ip_hdr (skb); lookup: - sk =__inet_lookup_skb (net->ipv4.tcp_death_row.hashinfo, - skb,__tcp_hdrlen (th), th->source, + sk =__inet_lookup_skb (skb,__tcp_hdrlen (th), th->source, th->dest, sdif, &refcounted);if (!sk) goto no_tcp_socket;
@@ -2426,9 +2422,7 @@ int tcp_v4_rcv(struct sk_buff *skb) &drop_reason);switch (tw_status) {case TCP_TW_SYN : { - struct sock *sk2 = inet_lookup_listener(net, - net->ipv4.tcp_death_row.hashinfo, - skb, __tcp_hdrlen(th), + struct sock *sk2 = inet_lookup_listener(net, skb, __tcp_hdrlen(th), iph->saddr, th->source, iph->daddr, th->dest, inet_iif(skb),
net/ipv4/tcp_offload.c +1 -2
@@ -434,8 +434,7 @@ static void tcp4_check_fraglist_gro(struct list_head *head, struct sk_buff *skb,inet_get_iif_sdif (skb, &iif, &sdif); iph =skb_gro_network_header (skb); net =dev_net_rcu (skb->dev); - sk =__inet_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, - iph->saddr, th->source, + sk =__inet_lookup_established (net, iph->saddr, th->source, iph->daddr,ntohs (th->dest), iif, sdif);NAPI_GRO_CB (skb)->is_flist = !sk;
net/ipv6/esp6.c +2 -2
@@ -149,8 +149,8 @@ static struct sock *esp6_find_tcp_sk(struct xfrm_state *x) dport = encap->encap_dport;spin_unlock_bh (&x->lock); - sk =__inet6_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, &x->id.daddr.in6, - dport, &x->props.saddr.in6,ntohs (sport),0 ,0 ); + sk =__inet6_lookup_established (net, &x->id.daddr.in6, dport, + &x->props.saddr.in6,ntohs (sport),0 ,0 );if (!sk)return ERR_PTR (-ENOENT );
net/ipv6/inet6_hashtables.c +23 -22
80 lines not on this page · raw message
net/ipv6/netfilter/nf_socket_ipv6.c +1 -2
@@ -83,8 +83,7 @@ nf_socket_get_sock_v6(struct net *net, struct sk_buff *skb, int doff, {switch (protocol) {case IPPROTO_TCP : -return inet6_lookup (net, net->ipv4.tcp_death_row.hashinfo, - skb, doff, saddr, sport, daddr, dport, +return inet6_lookup (net, skb, doff, saddr, sport, daddr, dport, in->ifindex);case IPPROTO_UDP : return udp6_lib_lookup(net, saddr, sport, daddr, dport,
net/ipv6/netfilter/nf_tproxy_ipv6.c +2 -3
@@ -80,7 +80,6 @@ nf_tproxy_get_sock_v6(struct net *net, struct sk_buff *skb, int thoff,const struct net_device *in, const enum nf_tproxy_lookup_t lookup_type) { - struct inet_hashinfo *hinfo = net->ipv4.tcp_death_row.hashinfo; struct sock *sk;switch (protocol) {
@@ -94,7 +93,7 @@ nf_tproxy_get_sock_v6(struct net *net, struct sk_buff *skb, int thoff,switch (lookup_type) {case NF_TPROXY_LOOKUP_LISTENER : - sk = inet6_lookup_listener(net, hinfo, skb, + sk = inet6_lookup_listener(net, skb, thoff + __tcp_hdrlen(hp), saddr, sport, daddr, ntohs(dport),
@@ -109,7 +108,7 @@ nf_tproxy_get_sock_v6(struct net *net, struct sk_buff *skb, int thoff,*/ break ;case NF_TPROXY_LOOKUP_ESTABLISHED : - sk =__inet6_lookup_established (net, hinfo, saddr, sport, daddr, + sk =__inet6_lookup_established (net, saddr, sport, daddr,ntohs (dport), in->ifindex,0 );break ;default :
net/ipv6/tcp_ipv6.c +5 -9
@@ -388,8 +388,7 @@ static int tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,bool fatal;int err; - sk =__inet6_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, - &hdr->daddr, th->dest, + sk =__inet6_lookup_established (net, &hdr->daddr, th->dest, &hdr->saddr,ntohs (th->source), skb->dev->ifindex,inet6_sdif (skb));
@@ -1073,8 +1072,7 @@ static void tcp_v6_send_reset(const struct sock *sk, struct sk_buff *skb,* Incoming packet is checked with md5 hash with finding key, * no RST generated if md5 hash doesn't match. */ - sk1 =inet6_lookup_listener (net, net->ipv4.tcp_death_row.hashinfo, -NULL ,0 , &ipv6h->saddr, th->source, + sk1 =inet6_lookup_listener (net,NULL ,0 , &ipv6h->saddr, th->source, &ipv6h->daddr,ntohs (th->source), dif, sdif);if (!sk1)
@@ -1789,7 +1787,7 @@ INDIRECT_CALLABLE_SCOPE int tcp_v6_rcv(struct sk_buff *skb) hdr =ipv6_hdr (skb); lookup: - sk =__inet6_lookup_skb (net->ipv4.tcp_death_row.hashinfo, skb,__tcp_hdrlen (th), + sk =__inet6_lookup_skb (skb,__tcp_hdrlen (th), th->source, th->dest,inet6_iif (skb), sdif, &refcounted);if (!sk)
@@ -1976,8 +1974,7 @@ INDIRECT_CALLABLE_SCOPE int tcp_v6_rcv(struct sk_buff *skb) {struct sock *sk2; - sk2 = inet6_lookup_listener(net, net->ipv4.tcp_death_row.hashinfo, - skb, __tcp_hdrlen(th), + sk2 = inet6_lookup_listener(net, skb, __tcp_hdrlen(th), &ipv6_hdr(skb)->saddr, th->source, &ipv6_hdr(skb)->daddr, ntohs(th->dest),
@@ -2029,8 +2026,7 @@ void tcp_v6_early_demux(struct sk_buff *skb)return ;/* Note : We use inet6_iif() here, not tcp_v6_iif() */ - sk =__inet6_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, - &hdr->saddr, th->source, + sk =__inet6_lookup_established (net, &hdr->saddr, th->source, &hdr->daddr,ntohs (th->dest),inet6_iif (skb),inet6_sdif (skb));if (sk) {
net/ipv6/tcpv6_offload.c +1 -2
@@ -36,8 +36,7 @@ static void tcp6_check_fraglist_gro(struct list_head *head, struct sk_buff *skb,inet6_get_iif_sdif (skb, &iif, &sdif); hdr =skb_gro_network_header (skb); net =dev_net_rcu (skb->dev); - sk =__inet6_lookup_established (net, net->ipv4.tcp_death_row.hashinfo, - &hdr->saddr, th->source, + sk =__inet6_lookup_established (net, &hdr->saddr, th->source, &hdr->daddr,ntohs (th->dest), iif, sdif);NAPI_GRO_CB (skb)->is_flist = !sk;
Signature · 1 line
2.51.0.rc2.233.g662b1ed5c5-goog
Re: [PATCH v2 net-next 4/6] tcp: Don't pass hashinfo to socket lookup helpers.
Quoted 9 lines · Kuniyuki Iwashima
These socket lookup functions required struct
inet_hashinfobecause they are shared by TCP and DCCP.*
__inet_lookup_established()
*__inet_lookup_listener()
*__inet6_lookup_established()
*inet6_lookup_listener()
Quoted 7 lines
DCCP has gone, and we don't need to pass hashinfo down to them.
Let's fetch
net->ipv4.tcp_death_row.hashinfodirectly in the above 4 functions.Signed-off-by Kuniyuki Iwashima <[email protected]>---
Okay, although this is a lot of code churn.
Kuniyuki Iwashima
reviewerPATCH v2 5/6 tcp: Don't pass hashinfo to inet_diag helpers.
In reply to Kuniyuki IwashimaThese inet_diag functions required struct inet_hashinfo because they are shared by TCP and DCCP:
* inet_diag_dump_icsk()
* inet_diag_dump_one_icsk()
* inet_diag_find_one_icsk()
DCCP has gone, and we don't need to pass hashinfo down to them.
Let's fetch net->ipv4.tcp_death_row.hashinfo directly in the first 2 functions.
Note that inet_diag_find_one_icsk() don't need hashinfo since the previous patch.
We will move TCP-specific functions to tcp_diag.c in the next patch.
--- include/linux/inet_diag.h | 6 ++---- net/ipv4/inet_diag.c | 10 +++++----- net/ipv4/tcp_diag.c | 17 +++-------------- 3 files changed, 10 insertions(+), 23 deletions(-)
include/linux/inet_diag.h +2 -4
@@ -48,15 +48,13 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, struct sk_buff *skb, struct netlink_callback *cb, const struct inet_diag_req_v2 *req, u16 nlmsg_flags, bool net_admin); -void inet_diag_dump_icsk (struct inet_hashinfo *h,struct sk_buff *skb, +void inet_diag_dump_icsk (struct sk_buff *skb,struct netlink_callback *cb,const struct inet_diag_req_v2 *r); -int inet_diag_dump_one_icsk (struct inet_hashinfo *hashinfo, -struct netlink_callback *cb, +int inet_diag_dump_one_icsk (struct netlink_callback *cb,const struct inet_diag_req_v2 *req);struct sock *inet_diag_find_one_icsk (struct net *net, -struct inet_hashinfo *hashinfo,const struct inet_diag_req_v2 *req);int inet_diag_bc_sk (const struct nlattr *_bc,struct sock *sk);
net/ipv4/inet_diag.c +5 -5
@@ -519,7 +519,6 @@ static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, }struct sock *inet_diag_find_one_icsk (struct net *net, -struct inet_hashinfo *hashinfo,const struct inet_diag_req_v2 *req) {struct sock *sk;
@@ -562,8 +561,7 @@ struct sock *inet_diag_find_one_icsk(struct net *net, }EXPORT_SYMBOL_GPL (inet_diag_find_one_icsk); -int inet_diag_dump_one_icsk (struct inet_hashinfo *hashinfo, -struct netlink_callback *cb, +int inet_diag_dump_one_icsk (struct netlink_callback *cb,const struct inet_diag_req_v2 *req) {struct sk_buff *in_skb = cb->skb;
@@ -573,7 +571,7 @@ int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo,struct sock *sk;int err; - sk =inet_diag_find_one_icsk (net, hashinfo, req); + sk =inet_diag_find_one_icsk (net, req);if (IS_ERR (sk))return PTR_ERR (sk);
@@ -1018,7 +1016,7 @@ static void twsk_build_assert(void)#endif } -void inet_diag_dump_icsk (struct inet_hashinfo *hashinfo,struct sk_buff *skb, +void inet_diag_dump_icsk (struct sk_buff *skb,struct netlink_callback *cb,const struct inet_diag_req_v2 *r) {
@@ -1026,10 +1024,12 @@ void inet_diag_dump_icsk(struct inet_hashinfo *hashinfo, struct sk_buff *skb,struct inet_diag_dump_data *cb_data = cb->data;struct net *net =sock_net (skb->sk);u32 idiag_states = r->idiag_states; +struct inet_hashinfo *hashinfo;int i, num, s_i, s_num;struct nlattr *bc;struct sock *sk; + hashinfo = net->ipv4.tcp_death_row.hashinfo; bc = cb_data->inet_diag_nla_bc;if (idiag_states &TCPF_SYN_RECV ) idiag_states |=TCPF_NEW_SYN_RECV ;
net/ipv4/tcp_diag.c +3 -14
@@ -180,21 +180,13 @@ static size_t tcp_diag_get_aux_size(struct sock *sk, bool net_admin)static void tcp_diag_dump (struct sk_buff *skb,struct netlink_callback *cb,const struct inet_diag_req_v2 *r) { -struct inet_hashinfo *hinfo; - - hinfo =sock_net (cb->skb->sk)->ipv4.tcp_death_row.hashinfo; - -inet_diag_dump_icsk (hinfo, skb, cb, r); +inet_diag_dump_icsk (skb, cb, r); }static int tcp_diag_dump_one (struct netlink_callback *cb,const struct inet_diag_req_v2 *req) { -struct inet_hashinfo *hinfo; - - hinfo =sock_net (cb->skb->sk)->ipv4.tcp_death_row.hashinfo; - -return inet_diag_dump_one_icsk (hinfo, cb, req); +return inet_diag_dump_one_icsk (cb, req); }#ifdef CONFIG_INET_DIAG_DESTROY
@@ -202,13 +194,10 @@ static int tcp_diag_destroy(struct sk_buff *in_skb, const struct inet_diag_req_v2 *req) {struct net *net =sock_net (in_skb->sk); -struct inet_hashinfo *hinfo;struct sock *sk;int err; - hinfo = net->ipv4.tcp_death_row.hashinfo; - sk =inet_diag_find_one_icsk (net, hinfo, req); - + sk =inet_diag_find_one_icsk (net, req);if (IS_ERR (sk))return PTR_ERR (sk);
Signature · 1 line
2.51.0.rc2.233.g662b1ed5c5-goog
Re: [PATCH v2 net-next 5/6] tcp: Don't pass hashinfo to inet_diag helpers.
Quoted 13 lines · Kuniyuki Iwashima
These
inet_diagfunctions required structinet_hashinfobecause they are shared by TCP and DCCP:*
inet_diag_dump_icsk()
*inet_diag_dump_one_icsk()
*inet_diag_find_one_icsk()DCCP has gone, and we don't need to pass hashinfo down to them.
Let's fetch
net->ipv4.tcp_death_row.hashinfodirectly in the first 2 functions.
Quoted 6 lines
Note that
inet_diag_find_one_icsk()don't need hashinfo since the previous patch.We will move TCP-specific functions to
tcp_diag.cin the next patch.Signed-off-by Kuniyuki Iwashima <[email protected]>
Kuniyuki Iwashima
reviewerPATCH v2 6/6 tcp: Move TCP-specific diag functions to tcp_diag.c.
In reply to Kuniyuki Iwashimatcp_diag_dump() / tcp_diag_dump_one() is just a wrapper of inet_diag_dump_icsk() / inet_diag_dump_one_icsk(), respectively.
Let's inline them in tcp_diag.c and move static callees as well.
Note that inet_sk_attr_size() is merged into tcp_diag_get_aux_size(), and we remove inet_diag_handler.idiag_get_aux_size() accordingly.
While at it, BUG_ON() is replaced with DEBUG_NET_WARN_ON_ONCE().
--- include/linux/inet_diag.h | 11 - net/ipv4/inet_diag.c | 479 -------------------------------------- net/ipv4/tcp_diag.c | 460 +++++++++++++++++++++++++++++++++++- 3 files changed, 455 insertions(+), 495 deletions(-)
include/linux/inet_diag.h +0 -11
@@ -24,9 +24,6 @@ struct inet_diag_handler {bool net_admin, struct sk_buff *skb); -size_t (*idiag_get_aux_size)(struct sock *sk, -bool net_admin); -int (*destroy)(struct sk_buff *in_skb,const struct inet_diag_req_v2 *req);
@@ -48,14 +45,6 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk, struct sk_buff *skb, struct netlink_callback *cb, const struct inet_diag_req_v2 *req, u16 nlmsg_flags, bool net_admin); -void inet_diag_dump_icsk (struct sk_buff *skb, -struct netlink_callback *cb, -const struct inet_diag_req_v2 *r); -int inet_diag_dump_one_icsk (struct netlink_callback *cb, -const struct inet_diag_req_v2 *req); - -struct sock *inet_diag_find_one_icsk (struct net *net, -const struct inet_diag_req_v2 *req);int inet_diag_bc_sk (const struct nlattr *_bc,struct sock *sk);
net/ipv4/inet_diag.c +0 -479
503 lines not on this page · raw message
net/ipv4/tcp_diag.c +455 -5
494 lines not on this page · raw message
Signature · 1 line
2.51.0.rc2.233.g662b1ed5c5-goog
This series was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:
Hello:
This series was applied to netdev/net-next.git (main) by Jakub Kicinski <[email protected]>:
Quoted 8 lines · On Fri, 22 Aug 2025 19:06:55 +0000 you wrote:
As I mentioned in [0], TCP still has code for DCCP.
This series cleans up such leftovers.
[0]: https://patch.msgid.link/[email protected]
[...]
Here is the summary with links:
- [v2,net-next,1/6] tcp: Remove sk_protocol test for tcp_twsk_unique().
https://git.kernel.org/netdev/net-next/c/9db0163e3cad - [v2,net-next,2/6] tcp: Remove timewait_sock_ops.twsk_destructor(). https://git.kernel.org/netdev/net-next/c/2d842b6c670b - [v2,net-next,3/6] tcp: Remove hashinfo test for inet6?_lookup_run_sk_lookup(). https://git.kernel.org/netdev/net-next/c/8150f3a44b17 - [v2,net-next,4/6] tcp: Don't pass hashinfo to socket lookup helpers. https://git.kernel.org/netdev/net-next/c/cb16f4b6c73d - [v2,net-next,5/6] tcp: Don't pass hashinfo to inet_diag helpers. https://git.kernel.org/netdev/net-next/c/f1241200cd66 - [v2,net-next,6/6] tcp: Move TCP-specific diag functions to tcp_diag.c. https://git.kernel.org/netdev/net-next/c/382a4d9cb6dc
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html