[PATCH bpf-next] bpf, sockmap: Call skb_linearize only when required in sk_psock_skb_ingress_enqueue

[PATCH bpf-next] bpf, sockmap: Call skb_linearize only when required in sk_psock_skb_ingress_enqueue

[Liu Jian]

The skb_to_sgvec fails only when the number of frag_list and frags exceeds
MAX_MSG_FRAGS. Therefore, we can call skb_linearize only when the
conversion fails.

Signed-off-by: Liu Jian <liujian56@huawei.com>
---
 net/core/skmsg.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

diff --git a/net/core/skmsg.c b/net/core/skmsg.c
index cc381165ea08..22b983ade0e7 100644
--- a/net/core/skmsg.c
+++ b/net/core/skmsg.c
@@ -524,16 +524,20 @@ static int sk_psock_skb_ingress_enqueue(struct sk_buff *skb,
 {
 	int num_sge, copied;
 
-	/* skb linearize may fail with ENOMEM, but lets simply try again
-	 * later if this happens. Under memory pressure we don't want to
-	 * drop the skb. We need to linearize the skb so that the mapping
-	 * in skb_to_sgvec can not error.
-	 */
-	if (skb_linearize(skb))
-		return -EAGAIN;
 	num_sge = skb_to_sgvec(skb, msg->sg.data, off, len);
-	if (unlikely(num_sge < 0))
-		return num_sge;
+	if (num_sge < 0) {
+		/* skb linearize may fail with ENOMEM, but lets simply try again
+		 * later if this happens. Under memory pressure we don't want to
+		 * drop the skb. We need to linearize the skb so that the mapping
+		 * in skb_to_sgvec can not error.
+		 */
+		if (skb_linearize(skb))
+			return -EAGAIN;
+
+		num_sge = skb_to_sgvec(skb, msg->sg.data, off, len);
+		if (unlikely(num_sge < 0))
+			return num_sge;
+	}
 
 	copied = len;
 	msg->sg.start = 0;
-- 
2.17.1

RE: [PATCH bpf-next] bpf, sockmap: Call skb_linearize only when required in sk_psock_skb_ingress_enqueue

[John Fastabend]

Liu Jian wrote:
> The skb_to_sgvec fails only when the number of frag_list and frags exceeds
> MAX_MSG_FRAGS. Therefore, we can call skb_linearize only when the
> conversion fails.
> 
> Signed-off-by: Liu Jian <liujian56@huawei.com>
> ---

Looks good.  I guess performance improvement for some use cases with lots of packets that
no longer need linearizing must be fairly good.

Acked-by: John Fastabend <john.fastabend@gmail.com>

Re: [PATCH bpf-next] bpf, sockmap: Call skb_linearize only when required in sk_psock_skb_ingress_enqueue

[patchwork-bot+netdevbpf]

Hello:

This patch was applied to bpf/bpf-next.git (master)
by Daniel Borkmann <daniel@iogearbox.net>:

On Wed, 27 Apr 2022 19:51:50 +0800 you wrote:
> The skb_to_sgvec fails only when the number of frag_list and frags exceeds
> MAX_MSG_FRAGS. Therefore, we can call skb_linearize only when the
> conversion fails.
> 
> Signed-off-by: Liu Jian <liujian56@huawei.com>
> ---
>  net/core/skmsg.c | 22 +++++++++++++---------
>  1 file changed, 13 insertions(+), 9 deletions(-)

Here is the summary with links:
  - [bpf-next] bpf, sockmap: Call skb_linearize only when required in sk_psock_skb_ingress_enqueue
    https://git.kernel.org/bpf/bpf-next/c/3527bfe6a92d

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