diff options
| author | Eric Dumazet <edumazet@google.com> | 2024-03-28 17:40:32 +0300 |
|---|---|---|
| committer | Jakub Kicinski <kuba@kernel.org> | 2024-03-30 01:03:11 +0300 |
| commit | 1abe267f173eae7ae76cf56232292e9641eb652f (patch) | |
| tree | b8f7952bdc3936448e9ee4da907745b2f9c1a7f8 /include | |
| parent | 612b1c0dec5bc7367f90fc508448b8d0d7c05414 (diff) | |
| download | linux-1abe267f173eae7ae76cf56232292e9641eb652f.tar.xz | |
net: add sk_wake_async_rcu() helper
While looking at UDP receive performance, I saw sk_wake_async()
was no longer inlined.
This matters at least on AMD Zen1-4 platforms (see SRSO)
This might be because rcu_read_lock() and rcu_read_unlock()
are no longer nops in recent kernels ?
Add sk_wake_async_rcu() variant, which must be called from
contexts already holding rcu lock.
As SOCK_FASYNC is deprecated in modern days, use unlikely()
to give a hint to the compiler.
sk_wake_async_rcu() is properly inlined from
__udp_enqueue_schedule_skb() and sock_def_readable().
Signed-off-by: Eric Dumazet <edumazet@google.com>
Link: https://lore.kernel.org/r/20240328144032.1864988-5-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Diffstat (limited to 'include')
| -rw-r--r-- | include/net/sock.h | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/include/net/sock.h b/include/net/sock.h index f57bfd8a2ad2..2253eefe2848 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2513,6 +2513,12 @@ static inline void sk_wake_async(const struct sock *sk, int how, int band) } } +static inline void sk_wake_async_rcu(const struct sock *sk, int how, int band) +{ + if (unlikely(sock_flag(sk, SOCK_FASYNC))) + sock_wake_async(rcu_dereference(sk->sk_wq), how, band); +} + /* Since sk_{r,w}mem_alloc sums skb->truesize, even a small frame might * need sizeof(sk_buff) + MTU + padding, unless net driver perform copybreak. * Note: for send buffers, TCP works better if we can build two skbs at |