[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id:
<171317942715.31565.13734913046471806260.git-patchwork-notify@kernel.org>
Date: Mon, 15 Apr 2024 11:10:27 +0000
From: patchwork-bot+netdevbpf@...nel.org
To: Gabriel Krisman Bertazi <krisman@...e.de>
Cc: willemdebruijn.kernel@...il.com, davem@...emloft.net, kuniyu@...zon.com,
netdev@...r.kernel.org, martin.lau@...nel.org, lmb@...valent.com
Subject: Re: [PATCH v3] udp: Avoid call to compute_score on multiple sites
Hello:
This patch was applied to netdev/net-next.git (main)
by David S. Miller <davem@...emloft.net>:
On Fri, 12 Apr 2024 17:20:04 -0400 you wrote:
> We've observed a 7-12% performance regression in iperf3 UDP ipv4 and
> ipv6 tests with multiple sockets on Zen3 cpus, which we traced back to
> commit f0ea27e7bfe1 ("udp: re-score reuseport groups when connected
> sockets are present"). The failing tests were those that would spawn
> UDP sockets per-cpu on systems that have a high number of cpus.
>
> Unsurprisingly, it is not caused by the extra re-scoring of the reused
> socket, but due to the compiler no longer inlining compute_score, once
> it has the extra call site in udp4_lib_lookup2. This is augmented by
> the "Safe RET" mitigation for SRSO, needed in our Zen3 cpus.
>
> [...]
Here is the summary with links:
- [v3] udp: Avoid call to compute_score on multiple sites
https://git.kernel.org/netdev/net-next/c/50aee97d1511
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
Powered by blists - more mailing lists