Message ID | 20240328144032.1864988-1-edumazet@google.com (mailing list archive) |
---|---|
Headers | show |
Series | udp: small changes on receive path | expand |
Hello: This series was applied to netdev/net-next.git (main) by Jakub Kicinski <kuba@kernel.org>: On Thu, 28 Mar 2024 14:40:28 +0000 you wrote: > This series is based on an observation I made in UDP receive path. > > The sock_def_readable() costs are pretty high, especially when > epoll is used to generate EPOLLIN events. > > First patch annotates races on sk->sk_rcvbuf reads. > > [...] Here is the summary with links: - [net-next,1/4] udp: annotate data-race in __udp_enqueue_schedule_skb() https://git.kernel.org/netdev/net-next/c/605579699513 - [net-next,2/4] udp: relax atomic operation on sk->sk_rmem_alloc https://git.kernel.org/netdev/net-next/c/6a1f12dd85a8 - [net-next,3/4] udp: avoid calling sock_def_readable() if possible https://git.kernel.org/netdev/net-next/c/612b1c0dec5b - [net-next,4/4] net: add sk_wake_async_rcu() helper https://git.kernel.org/netdev/net-next/c/1abe267f173e You are awesome, thank you!