Re: [PATCH net-next 05/10] udp: refine __udp_enqueue_schedule_skb() test

[Willem de Bruijn <willemdebruijn.kernel@gmail.com>]

Kuniyuki Iwashima wrote:
> On Wed, Sep 17, 2025 at 8:57 AM Eric Dumazet <edumazet@google.com> wrote:
> >
> > On Wed, Sep 17, 2025 at 8:00 AM Willem de Bruijn
> > <willemdebruijn.kernel@gmail.com> wrote:
> > >
> > > Eric Dumazet wrote:
> > > > Commit 5a465a0da13e ("udp: Fix multiple wraparounds
> > > > of sk->sk_rmem_alloc.") allowed to slightly overshoot
> > > > sk->sk_rmem_alloc, when many cpus are trying
> > > > to feed packets to a common UDP socket.
> > > >
> > > > This patch, combined with the following one reduces
> > > > false sharing on the victim socket under DDOS.
> > >
> > > It also changes the behavior. There was likely a reason to allow
> > > at least one packet if the buffer is small. Kuniyuki?
> >
> > It should not change the behavior.
> >
> > rmem would be zero if there is no packet in the queue : We still
> > accept the incoming skb, regardless of its truesize.
> >
> > If there is any packet, rmem > 0
> 
> Agreed, this change should be fine.
> 
> The rule comes from 0fd7bac6b6157, and later 850cbaddb52d
> tried to be more strict but caused regression, and the condition
> was converted to the current form in 363dc73acacbb.
> 
> Reviewed-by: Kuniyuki Iwashima <kuniyu@google.com>

Interesting. Thanks both!

Reviewed-by: Willem de Bruijn <willemb@google.com>