[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87ms9mn7sg.fsf@cloudflare.com>
Date: Wed, 02 Jul 2025 13:36:47 +0200
From: Jakub Sitnicki <jakub@...udflare.com>
To: Cong Wang <xiyou.wangcong@...il.com>
Cc: netdev@...r.kernel.org, bpf@...r.kernel.org, john.fastabend@...il.com,
zijianzhang@...edance.com, zhoufeng.zf@...edance.com, Cong Wang
<cong.wang@...edance.com>
Subject: Re: [Patch bpf-next v4 2/4] skmsg: implement slab allocator cache
for sk_msg
On Mon, Jun 30, 2025 at 06:11 PM -07, Cong Wang wrote:
> From: Zijian Zhang <zijianzhang@...edance.com>
>
> Optimizing redirect ingress performance requires frequent allocation and
> deallocation of sk_msg structures. Introduce a dedicated kmem_cache for
> sk_msg to reduce memory allocation overhead and improve performance.
>
> Acked-by: John Fastabend <john.fastabend@...il.com>
> Reviewed-by: Cong Wang <cong.wang@...edance.com>
> Signed-off-by: Zijian Zhang <zijianzhang@...edance.com>
> ---
skb ref management looks all right. Meaning I see no double release.
sk_msg_free clears the msg->skb field.
While sk_psock_skb_ingress_enqueue sets msg->skb only when it succeeds.
Reviewed-by: Jakub Sitnicki <jakub@...udflare.com>
Powered by blists - more mailing lists