* [PATCH net-next] mpls: Do not orphan the skb
@ 2024-03-06 18:11 Christoph Paasch
2024-03-06 18:40 ` Eric Dumazet
2024-03-08 5:00 ` patchwork-bot+netdevbpf
0 siblings, 2 replies; 3+ messages in thread
From: Christoph Paasch @ 2024-03-06 18:11 UTC (permalink / raw)
To: David Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni
Cc: netdev, Roopa Prabhu, Craig Taylor
We observed that TCP-pacing was falling back to the TCP-layer pacing
instead of utilizing sch_fq for the pacing. This causes significant
CPU-usage due to the hrtimer running on a per-TCP-connection basis.
The issue is that mpls_xmit() calls skb_orphan() and thus sets
skb->sk to NULL. Which implies that many of the goodies of TCP won't
work. Pacing falls back to TCP-layer pacing. TCP Small Queues does not
work, ...
It is safe to remove this call to skb_orphan() in mpls_xmit() as there
really is not reason for it to be there. It appears that this call to
skb_orphan comes from the very initial implementation of MPLS.
Cc: Roopa Prabhu <roopa@nvidia•com>
Reported-by: Craig Taylor <cmtaylor@apple•com>
Signed-off-by: Christoph Paasch <cpaasch@apple•com>
---
net/mpls/mpls_iptunnel.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/net/mpls/mpls_iptunnel.c b/net/mpls/mpls_iptunnel.c
index ef59e25dc482..8fc790f2a01b 100644
--- a/net/mpls/mpls_iptunnel.c
+++ b/net/mpls/mpls_iptunnel.c
@@ -55,8 +55,6 @@ static int mpls_xmit(struct sk_buff *skb)
out_dev = dst->dev;
net = dev_net(out_dev);
- skb_orphan(skb);
-
if (!mpls_output_possible(out_dev) ||
!dst->lwtstate || skb_warn_if_lro(skb))
goto drop;
--
2.39.3 (Apple Git-146)
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH net-next] mpls: Do not orphan the skb
2024-03-06 18:11 [PATCH net-next] mpls: Do not orphan the skb Christoph Paasch
@ 2024-03-06 18:40 ` Eric Dumazet
2024-03-08 5:00 ` patchwork-bot+netdevbpf
1 sibling, 0 replies; 3+ messages in thread
From: Eric Dumazet @ 2024-03-06 18:40 UTC (permalink / raw)
To: Christoph Paasch
Cc: David Miller, Jakub Kicinski, Paolo Abeni, netdev, Roopa Prabhu,
Craig Taylor
On Wed, Mar 6, 2024 at 7:11 PM Christoph Paasch <cpaasch@apple•com> wrote:
>
> We observed that TCP-pacing was falling back to the TCP-layer pacing
> instead of utilizing sch_fq for the pacing. This causes significant
> CPU-usage due to the hrtimer running on a per-TCP-connection basis.
>
> The issue is that mpls_xmit() calls skb_orphan() and thus sets
> skb->sk to NULL. Which implies that many of the goodies of TCP won't
> work. Pacing falls back to TCP-layer pacing. TCP Small Queues does not
> work, ...
>
> It is safe to remove this call to skb_orphan() in mpls_xmit() as there
> really is not reason for it to be there. It appears that this call to
> skb_orphan comes from the very initial implementation of MPLS.
>
> Cc: Roopa Prabhu <roopa@nvidia•com>
> Reported-by: Craig Taylor <cmtaylor@apple•com>
> Signed-off-by: Christoph Paasch <cpaasch@apple•com>
> ---
This seems reasonable to me, thanks.
Packets are immediately sent, there is no queue.
Reviewed-by: Eric Dumazet <edumazet@google•com>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH net-next] mpls: Do not orphan the skb
2024-03-06 18:11 [PATCH net-next] mpls: Do not orphan the skb Christoph Paasch
2024-03-06 18:40 ` Eric Dumazet
@ 2024-03-08 5:00 ` patchwork-bot+netdevbpf
1 sibling, 0 replies; 3+ messages in thread
From: patchwork-bot+netdevbpf @ 2024-03-08 5:00 UTC (permalink / raw)
To: Christoph Paasch; +Cc: davem, edumazet, kuba, pabeni, netdev, roopa, cmtaylor
Hello:
This patch was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel•org>:
On Wed, 06 Mar 2024 10:11:17 -0800 you wrote:
> We observed that TCP-pacing was falling back to the TCP-layer pacing
> instead of utilizing sch_fq for the pacing. This causes significant
> CPU-usage due to the hrtimer running on a per-TCP-connection basis.
>
> The issue is that mpls_xmit() calls skb_orphan() and thus sets
> skb->sk to NULL. Which implies that many of the goodies of TCP won't
> work. Pacing falls back to TCP-layer pacing. TCP Small Queues does not
> work, ...
>
> [...]
Here is the summary with links:
- [net-next] mpls: Do not orphan the skb
https://git.kernel.org/netdev/net-next/c/8edbd3960150
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2024-03-08 5:00 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-03-06 18:11 [PATCH net-next] mpls: Do not orphan the skb Christoph Paasch
2024-03-06 18:40 ` Eric Dumazet
2024-03-08 5:00 ` patchwork-bot+netdevbpf
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox