[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id:
<174248223302.1795325.5848574485105137769.git-patchwork-notify@kernel.org>
Date: Thu, 20 Mar 2025 14:50:33 +0000
From: patchwork-bot+netdevbpf@...nel.org
To: Mark Bloch <mbloch@...dia.com>
Cc: steffen.klassert@...unet.com, herbert@...dor.apana.org.au,
davem@...emloft.net, edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com,
horms@...nel.org, chopps@...n.net, cratiu@...dia.com, ychemla@...dia.com,
wangfe@...gle.com, lucien.xin@...il.com, dtatulea@...dia.com,
netdev@...r.kernel.org
Subject: Re: [PATCH net] xfrm: Force software GSO only in tunnel mode
Hello:
This patch was applied to netdev/net.git (main)
by Steffen Klassert <steffen.klassert@...unet.com>:
On Mon, 17 Mar 2025 12:32:05 +0200 you wrote:
> From: Cosmin Ratiu <cratiu@...dia.com>
>
> The cited commit fixed a software GSO bug with VXLAN + IPSec in tunnel
> mode. Unfortunately, it is slightly broader than necessary, as it also
> severely affects performance for Geneve + IPSec transport mode over a
> device capable of both HW GSO and IPSec crypto offload. In this case,
> xfrm_output unnecessarily triggers software GSO instead of letting the
> HW do it. In simple iperf3 tests over Geneve + IPSec transport mode over
> a back-2-back pair of NICs with MTU 1500, the performance was observed
> to be up to 6x worse when doing software GSO compared to leaving it to
> the hardware.
>
> [...]
Here is the summary with links:
- [net] xfrm: Force software GSO only in tunnel mode
https://git.kernel.org/netdev/net/c/0aae2867aa60
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