summaryrefslogtreecommitdiff
path: root/net/ipv4/inet_hashtables.c
diff options
context:
space:
mode:
authorEric Dumazet <edumazet@google.com>2021-10-14 16:41:26 +0300
committerDavid S. Miller <davem@davemloft.net>2021-10-15 13:28:34 +0300
commit19757cebf0c5016a1f36f7fe9810a9f0b33c0832 (patch)
tree9d2349b63f328c9cb79fe6580215f823f8765b78 /net/ipv4/inet_hashtables.c
parent0b93aed2842d950e8d2625e975e5a57febeff33d (diff)
downloadlinux-19757cebf0c5016a1f36f7fe9810a9f0b33c0832.tar.xz
tcp: switch orphan_count to bare per-cpu counters
Use of percpu_counter structure to track count of orphaned sockets is causing problems on modern hosts with 256 cpus or more. Stefan Bach reported a serious spinlock contention in real workloads, that I was able to reproduce with a netfilter rule dropping incoming FIN packets. 53.56% server [kernel.kallsyms] [k] queued_spin_lock_slowpath | ---queued_spin_lock_slowpath | --53.51%--_raw_spin_lock_irqsave | --53.51%--__percpu_counter_sum tcp_check_oom | |--39.03%--__tcp_close | tcp_close | inet_release | inet6_release | sock_close | __fput | ____fput | task_work_run | exit_to_usermode_loop | do_syscall_64 | entry_SYSCALL_64_after_hwframe | __GI___libc_close | --14.48%--tcp_out_of_resources tcp_write_timeout tcp_retransmit_timer tcp_write_timer_handler tcp_write_timer call_timer_fn expire_timers __run_timers run_timer_softirq __softirqentry_text_start As explained in commit cf86a086a180 ("net/dst: use a smaller percpu_counter batch for dst entries accounting"), default batch size is too big for the default value of tcp_max_orphans (262144). But even if we reduce batch sizes, there would still be cases where the estimated count of orphans is beyond the limit, and where tcp_too_many_orphans() has to call the expensive percpu_counter_sum_positive(). One solution is to use plain per-cpu counters, and have a timer to periodically refresh this cache. Updating this cache every 100ms seems about right, tcp pressure state is not radically changing over shorter periods. percpu_counter was nice 15 years ago while hosts had less than 16 cpus, not anymore by current standards. v2: Fix the build issue for CONFIG_CRYPTO_DEV_CHELSIO_TLS=m, reported by kernel test robot <lkp@intel.com> Remove unused socket argument from tcp_too_many_orphans() Fixes: dd24c00191d5 ("net: Use a percpu_counter for orphan_count") Signed-off-by: Eric Dumazet <edumazet@google.com> Reported-by: Stefan Bach <sfb@google.com> Cc: Neal Cardwell <ncardwell@google.com> Acked-by: Neal Cardwell <ncardwell@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4/inet_hashtables.c')
-rw-r--r--net/ipv4/inet_hashtables.c2
1 files changed, 1 insertions, 1 deletions
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index bfb522e51346..75737267746f 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -598,7 +598,7 @@ bool inet_ehash_nolisten(struct sock *sk, struct sock *osk, bool *found_dup_sk)
if (ok) {
sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
} else {
- percpu_counter_inc(sk->sk_prot->orphan_count);
+ this_cpu_inc(*sk->sk_prot->orphan_count);
inet_sk_set_state(sk, TCP_CLOSE);
sock_set_flag(sk, SOCK_DEAD);
inet_csk_destroy_sock(sk);