netdev - Re: [PATCH] net: change maximum number of UDP segments to 128

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: <66116432b55ce_16bd4c2943e@willemb.c.googlers.com.notmuch>
Date: Sat, 06 Apr 2024 11:03:14 -0400
From: Willem de Bruijn <willemdebruijn.kernel@...il.com>
To: Yuri Benditovich <yuri.benditovich@...nix.com>, 
 willemdebruijn.kernel@...il.com, 
 davem@...emloft.net, 
 edumazet@...gle.com, 
 kuba@...nel.org, 
 pabeni@...hat.com, 
 shuah@...nel.org, 
 linux-kernel@...r.kernel.org, 
 netdev@...r.kernel.org, 
 linux-kselftest@...r.kernel.org
Cc: yan@...nix.com, 
 andrew@...nix.com
Subject: Re: [PATCH] net: change maximum number of UDP segments to 128

Yuri Benditovich wrote:
> Earlier commit fc8b2a619469378 ("net: more strict VIRTIO_NET_HDR_GSO_UDP_L4 validation")
> added check of potential number of UDP segment vs UDP_MAX_SEGMENTS
> in linux/virtio_net.h.
> After this change certification test of USO guest-to-guest
> transmit on Windows driver for virtio-net device fails,
> for example with packet size of ~64K and mss of 536 bytes.
> In general the USO should not be more restrictive than TSO.

Ack

> Indeed, in case of unreasonably small mss a lot of segments
> can cause queue overflow and packet loss on the destination.
> Limit of 128 segments is good for any practical purpose,
> with minimal meaningful mss of 536 the maximal UDP packet will
> be divided to ~120 segments.
> 
> Signed-off-by: Yuri Benditovich <yuri.benditovich@...nix.com>

Please mark fixes as [PATCH net] and include a fixes tag:

Fixes: fc8b2a619469 ("net: more strict VIRTIO_NET_HDR_GSO_UDP_L4 validation")

Otherwise this looks fine to me. UDP_MAX_SEGMENTS exists to block
egregious examples, such as 64K 1B segments. Doubling to 128 to handle
536B MSS is well within the reasonable range.

> ---
>  include/linux/udp.h                  | 2 +-
>  tools/testing/selftests/net/udpgso.c | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/include/linux/udp.h b/include/linux/udp.h
> index 3748e82b627b..7e75ccdf25fe 100644
> --- a/include/linux/udp.h
> +++ b/include/linux/udp.h
> @@ -108,7 +108,7 @@ struct udp_sock {
>  #define udp_assign_bit(nr, sk, val)		\
>  	assign_bit(UDP_FLAGS_##nr, &udp_sk(sk)->udp_flags, val)
>  
> -#define UDP_MAX_SEGMENTS	(1 << 6UL)
> +#define UDP_MAX_SEGMENTS	(1 << 7UL)
>  
>  #define udp_sk(ptr) container_of_const(ptr, struct udp_sock, inet.sk)
>  
> diff --git a/tools/testing/selftests/net/udpgso.c b/tools/testing/selftests/net/udpgso.c
> index 1d975bf52af3..85b3baa3f7f3 100644
> --- a/tools/testing/selftests/net/udpgso.c
> +++ b/tools/testing/selftests/net/udpgso.c
> @@ -34,7 +34,7 @@
>  #endif
>  
>  #ifndef UDP_MAX_SEGMENTS
> -#define UDP_MAX_SEGMENTS	(1 << 6UL)
> +#define UDP_MAX_SEGMENTS	(1 << 7UL)
>  #endif
>  
>  #define CONST_MTU_TEST	1500
> -- 
> 2.34.3
>