[PATCH OLK-6.6 0/8] net: tcp: scale connect() under pressure
scale connect() under pressure. Eric Dumazet (6): tcp: use RCU in __inet{6}_check_established() tcp: optimize inet_use_bhash2_on_bind() tcp: add RCU management to inet_bind_bucket tcp: use RCU lookup in __inet_hash_connect() inet: change lport contribution to inet_ehashfn() and inet6_ehashfn() inet: call inet6_ehashfn() once from inet6_hash_connect() Kuniyuki Iwashima (2): tcp: Use bhash2 for v4-mapped-v6 non-wildcard address. tcp: Fix bind() regression for v6-only wildcard and v4-mapped-v6 non-wildcard addresses. include/net/inet_hashtables.h | 9 ++-- include/net/ip.h | 2 +- net/ipv4/inet_connection_sock.c | 20 +++++--- net/ipv4/inet_hashtables.c | 88 ++++++++++++++++++++++++--------- net/ipv4/inet_timewait_sock.c | 2 +- net/ipv6/inet6_hashtables.c | 40 +++++++++++---- 6 files changed, 118 insertions(+), 43 deletions(-) -- 2.25.1
From: Eric Dumazet <edumazet@google.com> mainline inclusion from mainline-v6.15-rc1 commit ae9d5b19b322d4b557efda684da2f4df21670ef8 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ICA4Z5 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- When __inet_hash_connect() has to try many 4-tuples before finding an available one, we see a high spinlock cost from __inet_check_established() and/or __inet6_check_established(). This patch adds an RCU lookup to avoid the spinlock acquisition when the 4-tuple is found in the hash table. Note that there are still spin_lock_bh() calls in __inet_hash_connect() to protect inet_bind_hashbucket, this will be fixed later in this series. Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Jason Xing <kerneljasonxing@gmail.com> Tested-by: Jason Xing <kerneljasonxing@gmail.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Link: https://patch.msgid.link/20250302124237.3913746-2-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Dong Chenchen <dongchenchen2@huawei.com> --- net/ipv4/inet_hashtables.c | 19 ++++++++++++++++--- net/ipv6/inet6_hashtables.c | 19 ++++++++++++++++--- 2 files changed, 32 insertions(+), 6 deletions(-) diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 9ecabf87b474..7a1092529a8e 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -559,11 +559,24 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row, unsigned int hash = inet_ehashfn(net, daddr, lport, saddr, inet->inet_dport); struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, hash); - spinlock_t *lock = inet_ehash_lockp(hinfo, hash); - struct sock *sk2; - const struct hlist_nulls_node *node; struct inet_timewait_sock *tw = NULL; + const struct hlist_nulls_node *node; + struct sock *sk2; + spinlock_t *lock; + + rcu_read_lock(); + sk_nulls_for_each(sk2, node, &head->chain) { + if (sk2->sk_hash != hash || + !inet_match(net, sk2, acookie, ports, dif, sdif)) + continue; + if (sk2->sk_state == TCP_TIME_WAIT) + break; + rcu_read_unlock(); + return -EADDRNOTAVAIL; + } + rcu_read_unlock(); + lock = inet_ehash_lockp(hinfo, hash); spin_lock(lock); sk_nulls_for_each(sk2, node, &head->chain) { diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index b0e8d278e8a9..8c133f6f21b9 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -276,11 +276,24 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row, const unsigned int hash = inet6_ehashfn(net, daddr, lport, saddr, inet->inet_dport); struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, hash); - spinlock_t *lock = inet_ehash_lockp(hinfo, hash); - struct sock *sk2; - const struct hlist_nulls_node *node; struct inet_timewait_sock *tw = NULL; + const struct hlist_nulls_node *node; + struct sock *sk2; + spinlock_t *lock; + + rcu_read_lock(); + sk_nulls_for_each(sk2, node, &head->chain) { + if (sk2->sk_hash != hash || + !inet6_match(net, sk2, saddr, daddr, ports, dif, sdif)) + continue; + if (sk2->sk_state == TCP_TIME_WAIT) + break; + rcu_read_unlock(); + return -EADDRNOTAVAIL; + } + rcu_read_unlock(); + lock = inet_ehash_lockp(hinfo, hash); spin_lock(lock); sk_nulls_for_each(sk2, node, &head->chain) { -- 2.25.1
From: Kuniyuki Iwashima <kuniyu@amazon.com> mainline inclusion from mainline-v6.8-rc1 commit 5e07e672412bed473122813ab35d4f7d42fd9635 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ICA4Z5 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- While checking port availability in bind() or listen(), we used only bhash for all v4-mapped-v6 addresses. But there is no good reason not to use bhash2 for v4-mapped-v6 non-wildcard addresses. Let's do it by returning true in inet_use_bhash2_on_bind(). Then, we also need to add a test in inet_bind2_bucket_match_addr_any() so that ::ffff:X.X.X.X will match with 0.0.0.0. Note that sk->sk_rcv_saddr is initialised for v4-mapped-v6 sk in __inet6_bind(). Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Dong Chenchen <dongchenchen2@huawei.com> --- net/ipv4/inet_connection_sock.c | 7 +++++-- net/ipv4/inet_hashtables.c | 3 ++- 2 files changed, 7 insertions(+), 3 deletions(-) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index bd032ac2376e..4a53c538dcaa 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -157,8 +157,11 @@ static bool inet_use_bhash2_on_bind(const struct sock *sk) if (sk->sk_family == AF_INET6) { int addr_type = ipv6_addr_type(&sk->sk_v6_rcv_saddr); - return addr_type != IPV6_ADDR_ANY && - addr_type != IPV6_ADDR_MAPPED; + if (addr_type == IPV6_ADDR_ANY) + return false; + + if (addr_type != IPV6_ADDR_MAPPED) + return true; } #endif return sk->sk_rcv_saddr != htonl(INADDR_ANY); diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 7a1092529a8e..7d274faeb884 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -856,7 +856,8 @@ bool inet_bind2_bucket_match_addr_any(const struct inet_bind2_bucket *tb, const return ipv6_addr_any(&tb->v6_rcv_saddr) || ipv6_addr_v4mapped_any(&tb->v6_rcv_saddr); - return false; + return ipv6_addr_v4mapped(&sk->sk_v6_rcv_saddr) && + tb->rcv_saddr == 0; } if (sk->sk_family == AF_INET6) -- 2.25.1
From: Eric Dumazet <edumazet@google.com> mainline inclusion from mainline-v6.15-rc1 commit ca79d80b0b9f42362a893f06413a9fe91811158a category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ICA4Z5 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- There is no reason to call ipv6_addr_type(). Instead, use highly optimized ipv6_addr_any() and ipv6_addr_v4mapped(). Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Jason Xing <kerneljasonxing@gmail.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Link: https://patch.msgid.link/20250302124237.3913746-3-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Dong Chenchen <dongchenchen2@huawei.com> --- net/ipv4/inet_connection_sock.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 4a53c538dcaa..d94db664e118 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -155,12 +155,10 @@ static bool inet_use_bhash2_on_bind(const struct sock *sk) { #if IS_ENABLED(CONFIG_IPV6) if (sk->sk_family == AF_INET6) { - int addr_type = ipv6_addr_type(&sk->sk_v6_rcv_saddr); - - if (addr_type == IPV6_ADDR_ANY) + if (ipv6_addr_any(&sk->sk_v6_rcv_saddr)) return false; - if (addr_type != IPV6_ADDR_MAPPED) + if (!ipv6_addr_v4mapped(&sk->sk_v6_rcv_saddr)) return true; } #endif -- 2.25.1
From: Eric Dumazet <edumazet@google.com> mainline inclusion from mainline-v6.15-rc1 commit d186f405fdf4229d0e9a52ba71662404b06cc002 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ICA4Z5 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- Add RCU protection to inet_bind_bucket structure. - Add rcu_head field to the structure definition. - Use kfree_rcu() at destroy time, and remove inet_bind_bucket_destroy() first argument. - Use hlist_del_rcu() and hlist_add_head_rcu() methods. Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Jason Xing <kerneljasonxing@gmail.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Link: https://patch.msgid.link/20250302124237.3913746-4-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Conflicts: include/net/inet_hashtables.h net/ipv4/inet_connection_sock.c net/ipv4/inet_hashtables.c net/ipv4/inet_timewait_sock.c [patchset "tcp: Refactor bhash2 and remove sk_bind2_node." is not backport.] Signed-off-by: Dong Chenchen <dongchenchen2@huawei.com> --- include/net/inet_hashtables.h | 4 ++-- net/ipv4/inet_connection_sock.c | 2 +- net/ipv4/inet_hashtables.c | 14 +++++++------- net/ipv4/inet_timewait_sock.c | 2 +- 4 files changed, 11 insertions(+), 11 deletions(-) diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index 3ecfeadbfa06..467c087411b9 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -89,6 +89,7 @@ struct inet_bind_bucket { bool fast_ipv6_only; struct hlist_node node; struct hlist_head owners; + struct rcu_head rcu; }; struct inet_bind2_bucket { @@ -231,8 +232,7 @@ struct inet_bind_bucket * inet_bind_bucket_create(struct kmem_cache *cachep, struct net *net, struct inet_bind_hashbucket *head, const unsigned short snum, int l3mdev); -void inet_bind_bucket_destroy(struct kmem_cache *cachep, - struct inet_bind_bucket *tb); +void inet_bind_bucket_destroy(struct inet_bind_bucket *tb); bool inet_bind_bucket_match(const struct inet_bind_bucket *tb, const struct net *net, unsigned short port, diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index d94db664e118..2e253df378ae 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -594,7 +594,7 @@ int inet_csk_get_port(struct sock *sk, unsigned short snum) fail_unlock: if (ret) { if (bhash_created) - inet_bind_bucket_destroy(hinfo->bind_bucket_cachep, tb); + inet_bind_bucket_destroy(tb); if (bhash2_created) inet_bind2_bucket_destroy(hinfo->bind2_bucket_cachep, tb2); diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 7d274faeb884..4958d700880f 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -79,7 +79,7 @@ struct inet_bind_bucket *inet_bind_bucket_create(struct kmem_cache *cachep, tb->fastreuse = 0; tb->fastreuseport = 0; INIT_HLIST_HEAD(&tb->owners); - hlist_add_head(&tb->node, &head->chain); + hlist_add_head_rcu(&tb->node, &head->chain); } return tb; } @@ -87,11 +87,11 @@ struct inet_bind_bucket *inet_bind_bucket_create(struct kmem_cache *cachep, /* * Caller must hold hashbucket lock for this tb with local BH disabled */ -void inet_bind_bucket_destroy(struct kmem_cache *cachep, struct inet_bind_bucket *tb) +void inet_bind_bucket_destroy(struct inet_bind_bucket *tb) { if (hlist_empty(&tb->owners)) { - __hlist_del(&tb->node); - kmem_cache_free(cachep, tb); + hlist_del_rcu(&tb->node); + kfree_rcu(tb, rcu); } } @@ -197,7 +197,7 @@ static void __inet_put_port(struct sock *sk) __sk_del_bind_node(sk); inet_csk(sk)->icsk_bind_hash = NULL; inet_sk(sk)->inet_num = 0; - inet_bind_bucket_destroy(hashinfo->bind_bucket_cachep, tb); + inet_bind_bucket_destroy(tb); spin_lock(&head2->lock); if (inet_csk(sk)->icsk_bind2_hash) { @@ -293,7 +293,7 @@ int __inet_inherit_port(const struct sock *sk, struct sock *child) error: if (created_inet_bind_bucket) - inet_bind_bucket_destroy(table->bind_bucket_cachep, tb); + inet_bind_bucket_destroy(tb); spin_unlock(&head2->lock); spin_unlock(&head->lock); return -ENOMEM; @@ -1171,7 +1171,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, spin_unlock(&head2->lock); if (tb_created) - inet_bind_bucket_destroy(hinfo->bind_bucket_cachep, tb); + inet_bind_bucket_destroy(tb); spin_unlock(&head->lock); if (tw) diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c index fff53144250c..f26745fa3d63 100644 --- a/net/ipv4/inet_timewait_sock.c +++ b/net/ipv4/inet_timewait_sock.c @@ -37,7 +37,7 @@ void inet_twsk_bind_unhash(struct inet_timewait_sock *tw, __hlist_del(&tw->tw_bind_node); tw->tw_tb = NULL; - inet_bind_bucket_destroy(hashinfo->bind_bucket_cachep, tb); + inet_bind_bucket_destroy(tb); __hlist_del(&tw->tw_bind2_node); tw->tw_tb2 = NULL; -- 2.25.1
From: Eric Dumazet <edumazet@google.com> mainline inclusion from mainline-v6.15-rc1 commit 86c2bc293b8130aec9fa504e953531a84a6eb9a6 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ICA4Z5 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- When __inet_hash_connect() has to try many 4-tuples before finding an available one, we see a high spinlock cost from the many spin_lock_bh(&head->lock) performed in its loop. This patch adds an RCU lookup to avoid the spinlock cost. check_established() gets a new @rcu_lookup argument. First reason is to not make any changes while head->lock is not held. Second reason is to not make this RCU lookup a second time after the spinlock has been acquired. Tested: Server: ulimit -n 40000; neper/tcp_crr -T 200 -F 30000 -6 --nolog Client: ulimit -n 40000; neper/tcp_crr -T 200 -F 30000 -6 --nolog -c -H server Before series: utime_start=0.288582 utime_end=1.548707 stime_start=20.637138 stime_end=2002.489845 num_transactions=484453 latency_min=0.156279245 latency_max=20.922042756 latency_mean=1.546521274 latency_stddev=3.936005194 num_samples=312537 throughput=47426.00 perf top on the client: 49.54% [kernel] [k] _raw_spin_lock 25.87% [kernel] [k] _raw_spin_lock_bh 5.97% [kernel] [k] queued_spin_lock_slowpath 5.67% [kernel] [k] __inet_hash_connect 3.53% [kernel] [k] __inet6_check_established 3.48% [kernel] [k] inet6_ehashfn 0.64% [kernel] [k] rcu_all_qs After this series: utime_start=0.271607 utime_end=3.847111 stime_start=18.407684 stime_end=1997.485557 num_transactions=1350742 latency_min=0.014131929 latency_max=17.895073144 latency_mean=0.505675853 # Nice reduction of latency metrics latency_stddev=2.125164772 num_samples=307884 throughput=139866.80 # 190 % increase perf top on client: 56.86% [kernel] [k] __inet6_check_established 17.96% [kernel] [k] __inet_hash_connect 13.88% [kernel] [k] inet6_ehashfn 2.52% [kernel] [k] rcu_all_qs 2.01% [kernel] [k] __cond_resched 0.41% [kernel] [k] _raw_spin_lock Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Jason Xing <kerneljasonxing@gmail.com> Tested-by: Jason Xing <kerneljasonxing@gmail.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Link: https://patch.msgid.link/20250302124237.3913746-5-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Conflicts: net/ipv4/inet_hashtables.c net/ipv6/inet6_hashtables.c [commit d4433e8b405a and commit 207184853dbd are not backport] Signed-off-by: Dong Chenchen <dongchenchen2@huawei.com> --- include/net/inet_hashtables.h | 3 +- net/ipv4/inet_hashtables.c | 52 +++++++++++++++++++++++------------ net/ipv6/inet6_hashtables.c | 24 ++++++++-------- 3 files changed, 50 insertions(+), 29 deletions(-) diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index 467c087411b9..d052d75202cc 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -534,7 +534,8 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, struct sock *sk, u64 port_offset, int (*check_established)(struct inet_timewait_death_row *, struct sock *, __u16, - struct inet_timewait_sock **)); + struct inet_timewait_sock **, + bool rcu_lookup)); int inet_hash_connect(struct inet_timewait_death_row *death_row, struct sock *sk); diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 4958d700880f..ca0f9097a53b 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -545,7 +545,8 @@ EXPORT_SYMBOL_GPL(__inet_lookup_established); /* called with local bh disabled */ static int __inet_check_established(struct inet_timewait_death_row *death_row, struct sock *sk, __u16 lport, - struct inet_timewait_sock **twp) + struct inet_timewait_sock **twp, + bool rcu_lookup) { struct inet_hashinfo *hinfo = death_row->hashinfo; struct inet_sock *inet = inet_sk(sk); @@ -564,17 +565,17 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row, struct sock *sk2; spinlock_t *lock; - rcu_read_lock(); - sk_nulls_for_each(sk2, node, &head->chain) { - if (sk2->sk_hash != hash || - !inet_match(net, sk2, acookie, ports, dif, sdif)) - continue; - if (sk2->sk_state == TCP_TIME_WAIT) - break; - rcu_read_unlock(); - return -EADDRNOTAVAIL; + if (rcu_lookup) { + sk_nulls_for_each(sk2, node, &head->chain) { + if (sk2->sk_hash != hash || + !inet_match(net, sk2, acookie, ports, dif, sdif)) + continue; + if (sk2->sk_state == TCP_TIME_WAIT) + break; + return -EADDRNOTAVAIL; + } + return 0; } - rcu_read_unlock(); lock = inet_ehash_lockp(hinfo, hash); spin_lock(lock); @@ -1016,7 +1017,8 @@ static u32 *table_perturb; int __inet_hash_connect(struct inet_timewait_death_row *death_row, struct sock *sk, u64 port_offset, int (*check_established)(struct inet_timewait_death_row *, - struct sock *, __u16, struct inet_timewait_sock **)) + struct sock *, __u16, struct inet_timewait_sock **, + bool rcu_lookup)) { struct inet_hashinfo *hinfo = death_row->hashinfo; struct inet_bind_hashbucket *head, *head2; @@ -1033,7 +1035,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, if (port) { local_bh_disable(); - ret = check_established(death_row, sk, port, NULL); + ret = check_established(death_row, sk, port, NULL, false); local_bh_enable(); return ret; } @@ -1071,6 +1073,21 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, continue; head = &hinfo->bhash[inet_bhashfn(net, port, hinfo->bhash_size)]; + rcu_read_lock(); + hlist_for_each_entry_rcu(tb, &head->chain, node) { + if (!inet_bind_bucket_match(tb, net, port, l3mdev)) + continue; + if (tb->fastreuse >= 0 || tb->fastreuseport >= 0) { + rcu_read_unlock(); + goto next_port; + } + if (!check_established(death_row, sk, port, &tw, true)) + break; + rcu_read_unlock(); + goto next_port; + } + rcu_read_unlock(); + spin_lock_bh(&head->lock); /* Does not bother with rcv_saddr checks, because @@ -1080,12 +1097,12 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, if (inet_bind_bucket_match(tb, net, port, l3mdev)) { if (tb->fastreuse >= 0 || tb->fastreuseport >= 0) - goto next_port; + goto next_port_unlock; WARN_ON(hlist_empty(&tb->owners)); if (!check_established(death_row, sk, - port, &tw)) + port, &tw, false)) goto ok; - goto next_port; + goto next_port_unlock; } } @@ -1099,8 +1116,9 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, tb->fastreuse = -1; tb->fastreuseport = -1; goto ok; -next_port: +next_port_unlock: spin_unlock_bh(&head->lock); +next_port: cond_resched(); } diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index 8c133f6f21b9..b72ebbd65f8d 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -263,7 +263,8 @@ EXPORT_SYMBOL_GPL(inet6_lookup); static int __inet6_check_established(struct inet_timewait_death_row *death_row, struct sock *sk, const __u16 lport, - struct inet_timewait_sock **twp) + struct inet_timewait_sock **twp, + bool rcu_lookup) { struct inet_hashinfo *hinfo = death_row->hashinfo; struct inet_sock *inet = inet_sk(sk); @@ -281,17 +282,18 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row, struct sock *sk2; spinlock_t *lock; - rcu_read_lock(); - sk_nulls_for_each(sk2, node, &head->chain) { - if (sk2->sk_hash != hash || - !inet6_match(net, sk2, saddr, daddr, ports, dif, sdif)) - continue; - if (sk2->sk_state == TCP_TIME_WAIT) - break; - rcu_read_unlock(); - return -EADDRNOTAVAIL; + if (rcu_lookup) { + sk_nulls_for_each(sk2, node, &head->chain) { + if (sk2->sk_hash != hash || + !inet6_match(net, sk2, saddr, daddr, + ports, dif, sdif)) + continue; + if (sk2->sk_state == TCP_TIME_WAIT) + break; + return -EADDRNOTAVAIL; + } + return 0; } - rcu_read_unlock(); lock = inet_ehash_lockp(hinfo, hash); spin_lock(lock); -- 2.25.1
From: Eric Dumazet <edumazet@google.com> mainline inclusion from mainline-v6.15-rc1 commit 9544d60a2605d1500cf5c3e331a76b9eaf4538c9 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ICA4Z5 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- In order to speedup __inet_hash_connect(), we want to ensure hash values for <source address, port X, destination address, destination port> are not randomly spread, but monotonically increasing. Goal is to allow __inet_hash_connect() to derive the hash value of a candidate 4-tuple with a single addition in the following patch in the series. Given : hash_0 = inet_ehashfn(saddr, 0, daddr, dport) hash_sport = inet_ehashfn(saddr, sport, daddr, dport) Then (hash_sport == hash_0 + sport) for all sport values. As far as I know, there is no security implication with this change. After this patch, when __inet_hash_connect() has to try XXXX candidates, the hash table buckets are contiguous and packed, allowing a better use of cpu caches and hardware prefetchers. Tested: Server: ulimit -n 40000; neper/tcp_crr -T 200 -F 30000 -6 --nolog Client: ulimit -n 40000; neper/tcp_crr -T 200 -F 30000 -6 --nolog -c -H server Before this patch: utime_start=0.271607 utime_end=3.847111 stime_start=18.407684 stime_end=1997.485557 num_transactions=1350742 latency_min=0.014131929 latency_max=17.895073144 latency_mean=0.505675853 latency_stddev=2.125164772 num_samples=307884 throughput=139866.80 perf top on client: 56.86% [kernel] [k] __inet6_check_established 17.96% [kernel] [k] __inet_hash_connect 13.88% [kernel] [k] inet6_ehashfn 2.52% [kernel] [k] rcu_all_qs 2.01% [kernel] [k] __cond_resched 0.41% [kernel] [k] _raw_spin_lock After this patch: utime_start=0.286131 utime_end=4.378886 stime_start=11.952556 stime_end=1991.655533 num_transactions=1446830 latency_min=0.001061085 latency_max=12.075275028 latency_mean=0.376375302 latency_stddev=1.361969596 num_samples=306383 throughput=151866.56 perf top: 50.01% [kernel] [k] __inet6_check_established 20.65% [kernel] [k] __inet_hash_connect 15.81% [kernel] [k] inet6_ehashfn 2.92% [kernel] [k] rcu_all_qs 2.34% [kernel] [k] __cond_resched 0.50% [kernel] [k] _raw_spin_lock 0.34% [kernel] [k] sched_balance_trigger 0.24% [kernel] [k] queued_spin_lock_slowpath There is indeed an increase of throughput and reduction of latency. Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Tested-by: Jason Xing <kerneljasonxing@gmail.com> Reviewed-by: Jason Xing <kerneljasonxing@gmail.com> Link: https://patch.msgid.link/20250305034550.879255-2-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Conflicts: net/ipv4/inet_hashtables.c net/ipv6/inet6_hashtables.c [commit 6e0735723ab4 and df51b8456415 are not backport] Signed-off-by: Dong Chenchen <dongchenchen2@huawei.com> --- net/ipv4/inet_hashtables.c | 4 ++-- net/ipv6/inet6_hashtables.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index ca0f9097a53b..a3621a056f31 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -38,8 +38,8 @@ u32 inet_ehashfn(const struct net *net, const __be32 laddr, net_get_random_once(&inet_ehash_secret, sizeof(inet_ehash_secret)); - return __inet_ehashfn(laddr, lport, faddr, fport, - inet_ehash_secret + net_hash_mix(net)); + return lport + __inet_ehashfn(laddr, 0, faddr, fport, + inet_ehash_secret + net_hash_mix(net)); } EXPORT_SYMBOL_GPL(inet_ehashfn); diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index b72ebbd65f8d..e5e6cb26a432 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -36,8 +36,8 @@ u32 inet6_ehashfn(const struct net *net, lhash = (__force u32)laddr->s6_addr32[3]; fhash = __ipv6_addr_jhash(faddr, ipv6_hash_secret); - return __inet6_ehashfn(lhash, lport, fhash, fport, - inet6_ehash_secret + net_hash_mix(net)); + return lport + __inet6_ehashfn(lhash, 0, fhash, fport, + inet6_ehash_secret + net_hash_mix(net)); } EXPORT_SYMBOL_GPL(inet6_ehashfn); -- 2.25.1
From: Eric Dumazet <edumazet@google.com> mainline inclusion from mainline-v6.15-rc1 commit d4438ce68bf145aa1d7ed03ebf3b8ece337e3f64 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ICA4Z5 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- inet6_ehashfn() being called from __inet6_check_established() has a big impact on performance, as shown in the Tested section. After prior patch, we can compute the hash for port 0 from inet6_hash_connect(), and derive each hash in __inet_hash_connect() from this initial hash: hash(saddr, lport, daddr, dport) == hash(saddr, 0, daddr, dport) + lport Apply the same principle for __inet_check_established(), although inet_ehashfn() has a smaller cost. Tested: Server: ulimit -n 40000; neper/tcp_crr -T 200 -F 30000 -6 --nolog Client: ulimit -n 40000; neper/tcp_crr -T 200 -F 30000 -6 --nolog -c -H server Before this patch: utime_start=0.286131 utime_end=4.378886 stime_start=11.952556 stime_end=1991.655533 num_transactions=1446830 latency_min=0.001061085 latency_max=12.075275028 latency_mean=0.376375302 latency_stddev=1.361969596 num_samples=306383 throughput=151866.56 perf top: 50.01% [kernel] [k] __inet6_check_established 20.65% [kernel] [k] __inet_hash_connect 15.81% [kernel] [k] inet6_ehashfn 2.92% [kernel] [k] rcu_all_qs 2.34% [kernel] [k] __cond_resched 0.50% [kernel] [k] _raw_spin_lock 0.34% [kernel] [k] sched_balance_trigger 0.24% [kernel] [k] queued_spin_lock_slowpath After this patch: utime_start=0.315047 utime_end=9.257617 stime_start=7.041489 stime_end=1923.688387 num_transactions=3057968 latency_min=0.003041375 latency_max=7.056589232 latency_mean=0.141075048 # Better latency metrics latency_stddev=0.526900516 num_samples=312996 throughput=320677.21 # 111 % increase, and 229 % for the series perf top: inet6_ehashfn is no longer seen. 39.67% [kernel] [k] __inet_hash_connect 37.06% [kernel] [k] __inet6_check_established 4.79% [kernel] [k] rcu_all_qs 3.82% [kernel] [k] __cond_resched 1.76% [kernel] [k] sched_balance_domains 0.82% [kernel] [k] _raw_spin_lock 0.81% [kernel] [k] sched_balance_rq 0.81% [kernel] [k] sched_balance_trigger 0.76% [kernel] [k] queued_spin_lock_slowpath Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com> Tested-by: Jason Xing <kerneljasonxing@gmail.com> Reviewed-by: Jason Xing <kerneljasonxing@gmail.com> Link: https://patch.msgid.link/20250305034550.879255-3-edumazet@google.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Conflicts: include/net/ip.h net/ipv4/inet_hashtables.c net/ipv6/inet6_hashtables.c [patchset "tcp: Refactor bhash2 and remove sk_bind2_node." was not backport] Signed-off-by: Dong Chenchen <dongchenchen2@huawei.com> --- include/net/inet_hashtables.h | 4 +++- include/net/ip.h | 2 +- net/ipv4/inet_hashtables.c | 26 ++++++++++++++++++-------- net/ipv6/inet6_hashtables.c | 15 +++++++++++---- 4 files changed, 33 insertions(+), 14 deletions(-) diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index d052d75202cc..bdc4c38baa9f 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -532,10 +532,12 @@ static inline void sk_rcv_saddr_set(struct sock *sk, __be32 addr) int __inet_hash_connect(struct inet_timewait_death_row *death_row, struct sock *sk, u64 port_offset, + u32 hash_port0, int (*check_established)(struct inet_timewait_death_row *, struct sock *, __u16, struct inet_timewait_sock **, - bool rcu_lookup)); + bool rcu_lookup, + u32 hash)); int inet_hash_connect(struct inet_timewait_death_row *death_row, struct sock *sk); diff --git a/include/net/ip.h b/include/net/ip.h index ae68d3f157a8..a21f2af40be0 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -353,7 +353,7 @@ void inet_get_local_port_range(const struct net *net, int *low, int *high); void inet_sk_get_local_port_range(const struct sock *sk, int *low, int *high); #ifdef CONFIG_SYSCTL -static inline bool inet_is_local_reserved_port(struct net *net, unsigned short port) +static inline bool inet_is_local_reserved_port(const struct net *net, unsigned short port) { if (!net->ipv4.sysctl_local_reserved_ports) return false; diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index a3621a056f31..5374ef7a7bf6 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -546,7 +546,8 @@ EXPORT_SYMBOL_GPL(__inet_lookup_established); static int __inet_check_established(struct inet_timewait_death_row *death_row, struct sock *sk, __u16 lport, struct inet_timewait_sock **twp, - bool rcu_lookup) + bool rcu_lookup, + u32 hash) { struct inet_hashinfo *hinfo = death_row->hashinfo; struct inet_sock *inet = inet_sk(sk); @@ -557,8 +558,6 @@ static int __inet_check_established(struct inet_timewait_death_row *death_row, int sdif = l3mdev_master_ifindex_by_index(net, dif); INET_ADDR_COOKIE(acookie, saddr, daddr); const __portpair ports = INET_COMBINED_PORTS(inet->inet_dport, lport); - unsigned int hash = inet_ehashfn(net, daddr, lport, - saddr, inet->inet_dport); struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, hash); struct inet_timewait_sock *tw = NULL; const struct hlist_nulls_node *node; @@ -1016,9 +1015,10 @@ static u32 *table_perturb; int __inet_hash_connect(struct inet_timewait_death_row *death_row, struct sock *sk, u64 port_offset, + u32 hash_port0, int (*check_established)(struct inet_timewait_death_row *, struct sock *, __u16, struct inet_timewait_sock **, - bool rcu_lookup)) + bool rcu_lookup, u32 hash)) { struct inet_hashinfo *hinfo = death_row->hashinfo; struct inet_bind_hashbucket *head, *head2; @@ -1035,7 +1035,8 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, if (port) { local_bh_disable(); - ret = check_established(death_row, sk, port, NULL, false); + ret = check_established(death_row, sk, port, NULL, false, + hash_port0 + port); local_bh_enable(); return ret; } @@ -1081,7 +1082,8 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, rcu_read_unlock(); goto next_port; } - if (!check_established(death_row, sk, port, &tw, true)) + if (!check_established(death_row, sk, port, &tw, true, + hash_port0 + port)) break; rcu_read_unlock(); goto next_port; @@ -1100,7 +1102,8 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, goto next_port_unlock; WARN_ON(hlist_empty(&tb->owners)); if (!check_established(death_row, sk, - port, &tw, false)) + port, &tw, false, + hash_port0 + port)) goto ok; goto next_port_unlock; } @@ -1206,11 +1209,18 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row, int inet_hash_connect(struct inet_timewait_death_row *death_row, struct sock *sk) { + const struct inet_sock *inet = inet_sk(sk); + const struct net *net = sock_net(sk); u64 port_offset = 0; + u32 hash_port0; if (!inet_sk(sk)->inet_num) port_offset = inet_sk_port_offset(sk); - return __inet_hash_connect(death_row, sk, port_offset, + + hash_port0 = inet_ehashfn(net, inet->inet_rcv_saddr, 0, + inet->inet_daddr, inet->inet_dport); + + return __inet_hash_connect(death_row, sk, port_offset, hash_port0, __inet_check_established); } EXPORT_SYMBOL_GPL(inet_hash_connect); diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c index e5e6cb26a432..6a838ce959ed 100644 --- a/net/ipv6/inet6_hashtables.c +++ b/net/ipv6/inet6_hashtables.c @@ -264,7 +264,8 @@ EXPORT_SYMBOL_GPL(inet6_lookup); static int __inet6_check_established(struct inet_timewait_death_row *death_row, struct sock *sk, const __u16 lport, struct inet_timewait_sock **twp, - bool rcu_lookup) + bool rcu_lookup, + u32 hash) { struct inet_hashinfo *hinfo = death_row->hashinfo; struct inet_sock *inet = inet_sk(sk); @@ -274,8 +275,6 @@ static int __inet6_check_established(struct inet_timewait_death_row *death_row, struct net *net = sock_net(sk); const int sdif = l3mdev_master_ifindex_by_index(net, dif); const __portpair ports = INET_COMBINED_PORTS(inet->inet_dport, lport); - const unsigned int hash = inet6_ehashfn(net, daddr, lport, saddr, - inet->inet_dport); struct inet_ehash_bucket *head = inet_ehash_bucket(hinfo, hash); struct inet_timewait_sock *tw = NULL; const struct hlist_nulls_node *node; @@ -353,11 +352,19 @@ static u64 inet6_sk_port_offset(const struct sock *sk) int inet6_hash_connect(struct inet_timewait_death_row *death_row, struct sock *sk) { + const struct in6_addr *daddr = &sk->sk_v6_rcv_saddr; + const struct in6_addr *saddr = &sk->sk_v6_daddr; + const struct inet_sock *inet = inet_sk(sk); + const struct net *net = sock_net(sk); u64 port_offset = 0; + u32 hash_port0; if (!inet_sk(sk)->inet_num) port_offset = inet6_sk_port_offset(sk); - return __inet_hash_connect(death_row, sk, port_offset, + + hash_port0 = inet6_ehashfn(net, daddr, 0, saddr, inet->inet_dport); + + return __inet_hash_connect(death_row, sk, port_offset, hash_port0, __inet6_check_established); } EXPORT_SYMBOL_GPL(inet6_hash_connect); -- 2.25.1
From: Kuniyuki Iwashima <kuniyu@amazon.com> mainline inclusion from mainline-v6.9-rc3 commit ea111449501ea32bf6da82750de860243691efc7 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ICA4Z5 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- Commit 5e07e672412b ("tcp: Use bhash2 for v4-mapped-v6 non-wildcard address.") introduced bind() regression for v4-mapped-v6 address. When we bind() the following two addresses on the same port, the 2nd bind() should succeed but fails now. 1. [::] w/ IPV6_ONLY 2. ::ffff:127.0.0.1 After the chagne, v4-mapped-v6 uses bhash2 instead of bhash to detect conflict faster, but I forgot to add a necessary change. During the 2nd bind(), inet_bind2_bucket_match_addr_any() returns the tb2 bucket of [::], and inet_bhash2_conflict() finally calls inet_bind_conflict(), which returns true, meaning conflict. inet_bhash2_addr_any_conflict |- inet_bind2_bucket_match_addr_any <-- return [::] bucket `- inet_bhash2_conflict `- __inet_bhash2_conflict <-- checks IPV6_ONLY for AF_INET | but not for v4-mapped-v6 address `- inet_bind_conflict <-- does not check address inet_bind_conflict() does not check socket addresses because __inet_bhash2_conflict() is expected to do so. However, it checks IPV6_V6ONLY attribute only against AF_INET socket, and not for v4-mapped-v6 address. As a result, v4-mapped-v6 address conflicts with v6-only wildcard address. To avoid that, let's add the missing test to use bhash2 for v4-mapped-v6 address. Fixes: 5e07e672412b ("tcp: Use bhash2 for v4-mapped-v6 non-wildcard address.") Signed-off-by: Kuniyuki Iwashima <kuniyu@amazon.com> Link: https://lore.kernel.org/r/20240326204251.51301-2-kuniyu@amazon.com Signed-off-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Dong Chenchen <dongchenchen2@huawei.com> --- net/ipv4/inet_connection_sock.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 2e253df378ae..177dac731fcd 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -199,8 +199,15 @@ static bool __inet_bhash2_conflict(const struct sock *sk, struct sock *sk2, kuid_t sk_uid, bool relax, bool reuseport_cb_ok, bool reuseport_ok) { - if (sk->sk_family == AF_INET && ipv6_only_sock(sk2)) - return false; + if (ipv6_only_sock(sk2)) { + if (sk->sk_family == AF_INET) + return false; + +#if IS_ENABLED(CONFIG_IPV6) + if (ipv6_addr_v4mapped(&sk->sk_v6_rcv_saddr)) + return false; +#endif + } return inet_bind_conflict(sk, sk2, sk_uid, relax, reuseport_cb_ok, reuseport_ok); -- 2.25.1
反馈： 您发送到kernel@openeuler.org的补丁/补丁集，已成功转换为PR！ PR链接地址： https://gitee.com/openeuler/kernel/pulls/16481 邮件列表地址：https://mailweb.openeuler.org/archives/list/kernel@openeuler.org/message/IGR... FeedBack: The patch(es) which you have sent to kernel@openeuler.org mailing list has been converted to a pull request successfully! Pull request link: https://gitee.com/openeuler/kernel/pulls/16481 Mailing list address: https://mailweb.openeuler.org/archives/list/kernel@openeuler.org/message/IGR...
participants (2)
-
Dong Chenchen -
patchwork bot