Hello:

This series was applied to netdev/net-next.git (main)
by Paolo Abeni [off-list ref]:

On Sun, 28 Sep 2025 08:49:31 +0000 you wrote:

Platforms with many cpus and relatively slow inter connect show
a significant spinlock contention in skb_attempt_defer_free().

This series refactors this infrastructure to be NUMA aware,
and lockless.

Tested on various platforms, including AMD Zen 2/3/4
and Intel Granite Rapids, showing significant cost reductions
under network stress (more than 20 Mpps).

[...]

Here is the summary with links:
  - [v2,net-next,1/3] net: make softnet_data.defer_count an atomic
    https://git.kernel.org/netdev/net-next/c/9c94ae6bb0b2
  - [v2,net-next,2/3] net: use llist for sd->defer_list
    https://git.kernel.org/netdev/net-next/c/844c9db7f7f5
  - [v2,net-next,3/3] net: add NUMA awareness to skb_attempt_defer_free()
    https://git.kernel.org/netdev/net-next/c/5628f3fe3b16

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html