| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <CANn89iLsA1N7NDz71GRaoMWn2xWmJvMP4oPs=yMCbLvE--R8kQ@mail.gmail.com>
Date: Mon, 25 Aug 2025 22:46:22 -0700
From: Eric Dumazet <edumazet@...gle.com>
To: Kuniyuki Iwashima <kuniyu@...gle.com>
Cc: "David S . Miller" <davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>,
Paolo Abeni <pabeni@...hat.com>, Simon Horman <horms@...nel.org>, netdev@...r.kernel.org,
eric.dumazet@...il.com, Willem de Bruijn <willemb@...gle.com>
Subject: Re: [PATCH net-next 3/3] net: add new sk->sk_drops1 field
On Mon, Aug 25, 2025 at 1:19 PM Kuniyuki Iwashima <kuniyu@...gle.com> wrote:
>
> On Mon, Aug 25, 2025 at 12:59 PM Eric Dumazet <edumazet@...gle.com> wrote:
> >
> > sk->sk_drops can be heavily contended when
> > changed from many cpus.
> >
> > Instead using too expensive per-cpu data structure,
> > add a second sk->sk_drops1 field and change
> > sk_drops_inc() to be NUMA aware.
> >
> > This patch adds 64 bytes per socket.
> >
> > For hosts having more than two memory nodes, sk_drops_inc()
> > might not be optimal and can be refined later.
> >
> > Tested with the following stress test, sending about 11 Mpps
> > to a dual socket AMD EPYC 7B13 64-Core.
> >
> > super_netperf 20 -t UDP_STREAM -H DUT -l10 -- -n -P,1000 -m 120
> > Note: due to socket lookup, only one UDP socket will receive
> > packets on DUT.
> >
> > Then measure receiver (DUT) behavior. We can see
> > consumer and BH handlers can process more packets per second.
> >
> > Before:
> >
> > nstat -n ; sleep 1 ; nstat | grep Udp
> > Udp6InDatagrams 855592 0.0
> > Udp6InErrors 5621467 0.0
> > Udp6RcvbufErrors 5621467 0.0
> >
> > After:
> > nstat -n ; sleep 1 ; nstat | grep Udp
> > Udp6InDatagrams 914537 0.0
> > Udp6InErrors 6888487 0.0
> > Udp6RcvbufErrors 6888487 0.0
> >
> > Signed-off-by: Eric Dumazet <edumazet@...gle.com>
> > ---
> > include/net/sock.h | 20 +++++++++++++++++++
> > .../selftests/bpf/progs/bpf_iter_netlink.c | 3 ++-
> > .../selftests/bpf/progs/bpf_iter_udp4.c | 2 +-
> > .../selftests/bpf/progs/bpf_iter_udp6.c | 2 +-
> > 4 files changed, 24 insertions(+), 3 deletions(-)
> >
> > diff --git a/include/net/sock.h b/include/net/sock.h
> > index f40e3c4883be32c8282694ab215bcf79eb87cbd7..318169eb1a3d40eefac50147012551614abc6f7a 100644
> > --- a/include/net/sock.h
> > +++ b/include/net/sock.h
> > @@ -282,6 +282,7 @@ struct sk_filter;
> > * @sk_err_soft: errors that don't cause failure but are the cause of a
> > * persistent failure not just 'timed out'
> > * @sk_drops: raw/udp drops counter
> > + * @sk_drops1: second drops counter
> > * @sk_ack_backlog: current listen backlog
> > * @sk_max_ack_backlog: listen backlog set in listen()
> > * @sk_uid: user id of owner
> > @@ -571,6 +572,11 @@ struct sock {
> > atomic_t sk_drops ____cacheline_aligned_in_smp;
> > struct rcu_head sk_rcu;
> > netns_tracker ns_tracker;
> > +#if defined(CONFIG_NUMA)
> > + atomic_t sk_drops1 ____cacheline_aligned_in_smp;
> > +#else
> > + atomic_t sk_drops1;
>
> Is this #else to be friendly to bpf selftests ?
Sure !
What problem are you seeing exactly ?
>
>
> > +#endif
> > };
> >
> > struct sock_bh_locked {
> > @@ -2684,17 +2690,31 @@ struct sock_skb_cb {
> >
> > static inline void sk_drops_inc(struct sock *sk)
> > {
> > +#if defined(CONFIG_NUMA)
> > + int n = numa_node_id() % 2;
> > +
> > + if (n)
> > + atomic_inc(&sk->sk_drops1);
> > + else
> > + atomic_inc(&sk->sk_drops);
> > +#else
> > atomic_inc(&sk->sk_drops);
> > +#endif
> > }
> >
> > static inline int sk_drops_read(const struct sock *sk)
> > {
> > +#if defined(CONFIG_NUMA)
> > + return atomic_read(&sk->sk_drops) + atomic_read(&sk->sk_drops1);
> > +#else
> > return atomic_read(&sk->sk_drops);
> > +#endif
> > }
> >
> > static inline void sk_drops_reset(struct sock *sk)
> > {
> > atomic_set(&sk->sk_drops, 0);
> > + atomic_set(&sk->sk_drops1, 0);
> > }
> >
> > static inline void
> > diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_netlink.c b/tools/testing/selftests/bpf/progs/bpf_iter_netlink.c
> > index 00b2ceae81fb0914f2de3634eb342004e8bc3c5b..31ad9fcc6022d5d31b9c6a35daacaad7c887a51f 100644
> > --- a/tools/testing/selftests/bpf/progs/bpf_iter_netlink.c
> > +++ b/tools/testing/selftests/bpf/progs/bpf_iter_netlink.c
> > @@ -57,7 +57,8 @@ int dump_netlink(struct bpf_iter__netlink *ctx)
> > inode = SOCK_INODE(sk);
> > bpf_probe_read_kernel(&ino, sizeof(ino), &inode->i_ino);
> > }
> > - BPF_SEQ_PRINTF(seq, "%-8u %-8lu\n", s->sk_drops.counter, ino);
> > + BPF_SEQ_PRINTF(seq, "%-8u %-8lu\n",
> > + s->sk_drops.counter + s->sk_drops1.counter, ino);
> >
> > return 0;
> > }
> > diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_udp4.c b/tools/testing/selftests/bpf/progs/bpf_iter_udp4.c
> > index ffbd4b116d17ffbb9f14440c788e50490fb0f4e0..192ab5693a7131c1ec5879e539651c21f6f3c9ae 100644
> > --- a/tools/testing/selftests/bpf/progs/bpf_iter_udp4.c
> > +++ b/tools/testing/selftests/bpf/progs/bpf_iter_udp4.c
> > @@ -64,7 +64,7 @@ int dump_udp4(struct bpf_iter__udp *ctx)
> > 0, 0L, 0, ctx->uid, 0,
> > sock_i_ino(&inet->sk),
> > inet->sk.sk_refcnt.refs.counter, udp_sk,
> > - inet->sk.sk_drops.counter);
> > + inet->sk.sk_drops.counter + inet->sk.sk_drops1.counter);
> >
> > return 0;
> > }
> > diff --git a/tools/testing/selftests/bpf/progs/bpf_iter_udp6.c b/tools/testing/selftests/bpf/progs/bpf_iter_udp6.c
> > index 47ff7754f4fda4c9db92fbf1dc2e6a68f044174e..5170bdf458fa1b9a4eea9240fbaa5934182a7776 100644
> > --- a/tools/testing/selftests/bpf/progs/bpf_iter_udp6.c
> > +++ b/tools/testing/selftests/bpf/progs/bpf_iter_udp6.c
> > @@ -72,7 +72,7 @@ int dump_udp6(struct bpf_iter__udp *ctx)
> > 0, 0L, 0, ctx->uid, 0,
> > sock_i_ino(&inet->sk),
> > inet->sk.sk_refcnt.refs.counter, udp_sk,
> > - inet->sk.sk_drops.counter);
> > + inet->sk.sk_drops.counter + inet->sk.sk_drops1.counter);
> >
> > return 0;
> > }
> > --
> > 2.51.0.261.g7ce5a0a67e-goog
> >
Powered by blists - more mailing lists