netdev - Re: [PATCH v2 net] tcp: Reduce chance of collisions in inet6

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

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CANn89iKo9h6fFg4BbYBZeWH1qdZKS4SDn1ufcKbSmRiviXvfTg@mail.gmail.com>
Date: Mon, 24 Jul 2023 11:41:12 +0200
From: Eric Dumazet <edumazet@...gle.com>
To: Kuniyuki Iwashima <kuniyu@...zon.com>
Cc: "David S. Miller" <davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>, 
	Paolo Abeni <pabeni@...hat.com>, David Ahern <dsahern@...nel.org>, 
	Amit Klein <aksecurity@...il.com>, Benjamin Herrenschmidt <benh@...zon.com>, 
	Kuniyuki Iwashima <kuni1840@...il.com>, netdev@...r.kernel.org, 
	Stewart Smith <trawets@...zon.com>, Samuel Mendoza-Jonas <samjonas@...zon.com>
Subject: Re: [PATCH v2 net] tcp: Reduce chance of collisions in inet6_hashfn().

On Sat, Jul 22, 2023 at 12:24 AM Kuniyuki Iwashima <kuniyu@...zon.com> wrote:
>
> From: Stewart Smith <trawets@...zon.com>
>
> For both IPv4 and IPv6 incoming TCP connections are tracked in a hash
> table with a hash over the source & destination addresses and ports.
> However, the IPv6 hash is insufficient and can lead to a high rate of
> collisions.
>
> The IPv6 hash used an XOR to fit everything into the 96 bits for the
> fast jenkins hash, meaning it is possible for an external entity to
> ensure the hash collides, thus falling back to a linear search in the
> bucket, which is slow.
>
> We take the approach of hash the full length of IPv6 address in
> __ipv6_addr_jhash() so that all users can benefit from a more secure
> version.
>
> While this may look like it adds overhead, the reality of modern CPUs
> means that this is unmeasurable in real world scenarios.
>
> In simulating with llvm-mca, the increase in cycles for the hashing
> code was ~16 cycles on Skylake (from a base of ~155), and an extra ~9
> on Nehalem (base of ~173).
>
> In commit dd6d2910c5e0 ("netfilter: conntrack: switch to siphash")
> netfilter switched from a jenkins hash to a siphash, but even the faster
> hsiphash is a more significant overhead (~20-30%) in some preliminary
> testing.  So, in this patch, we keep to the more conservative approach to
> ensure we don't add much overhead per SYN.
>
> In testing, this results in a consistently even spread across the
> connection buckets.  In both testing and real-world scenarios, we have
> not found any measurable performance impact.
>
> Fixes: 08dcdbf6a7b9 ("ipv6: use a stronger hash for tcp")
> Signed-off-by: Stewart Smith <trawets@...zon.com>
> Signed-off-by: Samuel Mendoza-Jonas <samjonas@...zon.com>
> Suggested-by: Eric Dumazet <edumazet@...gle.com>
> Signed-off-by: Kuniyuki Iwashima <kuniyu@...zon.com>
> ---
> v2:
>   * Hash all IPv6 bytes once in __ipv6_addr_jhash() instead of reusing
>     some bytes twice.

Thanks, I think we should merge this patch as is, and think about
further improvements later.

Reviewed-by: Eric Dumazet <edumazet@...gle.com>