Re: [PATCH net] tcp_metrics: hash the destination address with SipHash
From: Juan Agustín Nicco
Date: Mon Nov 16 2026 - 22:48:18 EST
On Mon, Nov 16, 2026 at 7:36 PM Kuniyuki Iwashima <kuniyu@xxxxxxxxxx> wrote:
>
> On Thu, 12 Nov 2026 01:46:24 -0500, Juan Agustin Niccolini wrote:
> > The tcp_metrics bucket is derived from a fold of the peer address.
> > For IPv4 the fold is ipv4_addr_hash(), which returns the address
> > unchanged. For IPv6 it is ipv6_addr_hash(), which XORs the four
> > 32-bit words of the address. net_hash_mix() is only XORed in after
> > that fold, so two addresses of either family with the same fold share
> > a bucket in every netns.
> >
> > A host owning an IPv6 /64 can thus compute, without knowing
> > net_hash_mix(), addresses that collide with a chosen IPv4 or IPv6 peer:
> >
> > w3 = w2 ^ p0 ^ p1 ^ target /* p0, p1: prefix, w2: free */
> >
> > A bucket holds six entries and the seventh insert replaces the oldest.
> > An off-path host can fill a peer's bucket with its own entries and
> > later tell whether one was evicted, because tcp_init_metrics() seeds
> > the initial RTO from the cached RTT. This reveals whether that peer,
> > or another address in the same bucket, connected in the meantime.
> > XOR and multiply is not a PRF, so net_hash_mix() can be recovered the
> > same way, after which evictions leak about 14 bits of any client's
> > address, depending on the number of buckets.
> >
> > Hash the address with SipHash and a random key instead, as commit
> > 2ee66e948717 ("inetpeer: randomize RB-tree node comparison using SipHash")
> > did for inet_peer. net_hash_mix() and hash_32() are kept, to separate
> > namespaces and to reduce the result to the table size.
> >
> > Fixes: 51c5d0c4b169 ("tcp: Maintain dynamic metrics in local cache.")
> > Cc: stable@xxxxxxxxxxxxxxx
> > Signed-off-by: Juan Agustin Niccolini <juanagusnicco@xxxxxxxxx>
> > ---
> > This can be exploited remotely to recover a client's IPv4 address.
> > A single host narrows the address to about 262 thousand candidates,
> > and because each netns keys the hash differently, intersecting buckets
> > extracted from multiple hosts converges quickly into one candidate;
> > 3-5 are enough. The net_hash_mix() value only needs to be extracted
> > once per boot per host.
> >
> > Many common cases facilitate this. For instance, a browser loading
> > one page connects to several hosts within the same second. That
> > supplies the buckets needed to converge, and the shared timing shows
> > they belong to the same client.
> >
> > I reproduced the bug on a QEMU guest running mainline (2a500a04d6f9) and
> > against Ubuntu 24.04 over the internet. With the patch applied, the
> > reproducer no longer finds addresses that share a bucket. The
> > reproducer is available on request.
> >
> > For stable: 5.10 and 5.15 lack siphash_aligned_key_t;
> > "static siphash_key_t tcpm_hash_key __read_mostly __aligned(16);" can
> > be used there.
> >
> > This bug was discovered during an LLM-based code review.
> >
> > net/ipv4/tcp_metrics.c | 32 +++++++++++++++++++++++++-------
> > 1 file changed, 25 insertions(+), 7 deletions(-)
> >
> > --- a/net/ipv4/tcp_metrics.c
> > +++ b/net/ipv4/tcp_metrics.c
> > @@ -8,6 +8,7 @@
> > #include <linux/init.h>
> > #include <linux/tcp.h>
> > #include <linux/hash.h>
> > +#include <linux/siphash.h>
> > #include <linux/tcp_metrics.h>
> > #include <linux/vmalloc.h>
> >
> > @@ -94,6 +95,23 @@
> >
> > static DEFINE_SPINLOCK(tcp_metrics_lock);
> > static DEFINE_SEQLOCK(fastopen_seqlock);
> > +
> > +static siphash_aligned_key_t tcpm_hash_key __read_mostly;
> > +
> > +static u32 tcpm_addr_hash(const struct inetpeer_addr *a)
> > +{
> > + net_get_random_once(&tcpm_hash_key, sizeof(tcpm_hash_key));
> > +
> > + if (a->family == AF_INET)
> > + return siphash_2u32((__force u32)a->a4.addr, a->a4.vif,
>
> Isn't a->a4.vif always zero ?
Yes. tcp_metrics only builds its keys with inetpeer_set_addr_v4(),
which sets vif to 0. I carried it over from inetpeer's hash, where the
vif can be set.
I will hash the address alone with siphash_1u32() in v2.
Thanks for the review.
pw-bot: cr