[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CANn89i+RbD3Dr+ca1+HErkW099DgZXQ7VF=vcY3zAd-xViFusg@mail.gmail.com>
Date: Mon, 11 Sep 2023 12:03:06 +0200
From: Eric Dumazet <edumazet@...gle.com>
To: Mika Westerberg <mika.westerberg@...ux.intel.com>
Cc: Michael Jamet <michael.jamet@...el.com>, Yehezkel Bernat <YehezkelShB@...il.com>,
"David S . Miller" <davem@...emloft.net>, Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
Alex Balcanquall <alex@...xbal.com>, netdev@...r.kernel.org
Subject: Re: [PATCH] net: thunderbolt: Fix TCP/UDPv6 GSO checksum calculation
On Mon, Sep 11, 2023 at 11:50 AM Mika Westerberg
<mika.westerberg@...ux.intel.com> wrote:
>
> Alex reported that running ssh over IPv6 does not work with
> Thunderbolt/USB4 networking driver. The reason for that is that driver
> should call skb_is_gso() before calling skb_is_gso_v6(), and it should
> not return false after calculates the checksum successfully. This probably
> was a copy paste error from the original driver where it was done
> properly.
>
> While there add checksum calculation for UDPv6 GSO packets as well.
This part does not belong to this patch, and should be submitted for net-next.
Note that this driver is not supposed to receive UDP GSO packets.
>
> Cc: stable@...r.kernel.org
What would be the Fixes: tag for this patch ?
> Reported-by: Alex Balcanquall <alex@...xbal.com>
> Signed-off-by: Mika Westerberg <mika.westerberg@...ux.intel.com>
> ---
> drivers/net/thunderbolt/main.c | 21 +++++++++++++++------
> 1 file changed, 15 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/net/thunderbolt/main.c b/drivers/net/thunderbolt/main.c
> index 0c1e8970ee58..ba50a554478f 100644
> --- a/drivers/net/thunderbolt/main.c
> +++ b/drivers/net/thunderbolt/main.c
> @@ -1049,12 +1049,21 @@ static bool tbnet_xmit_csum_and_map(struct tbnet *net, struct sk_buff *skb,
> *tucso = ~csum_tcpudp_magic(ip_hdr(skb)->saddr,
> ip_hdr(skb)->daddr, 0,
> ip_hdr(skb)->protocol, 0);
> - } else if (skb_is_gso_v6(skb)) {
> - tucso = dest + ((void *)&(tcp_hdr(skb)->check) - data);
> - *tucso = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
> - &ipv6_hdr(skb)->daddr, 0,
> - IPPROTO_TCP, 0);
> - return false;
> + } else if (skb_is_gso(skb)) {
> + if (skb_is_gso_v6(skb)) {
> + tucso = dest + ((void *)&(tcp_hdr(skb)->check) - data);
> + *tucso = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
> + &ipv6_hdr(skb)->daddr, 0,
> + IPPROTO_TCP, 0);
> + } else if (protocol == htons(ETH_P_IPV6) &&
> + (skb_shinfo(skb)->gso_type & SKB_GSO_UDP)) {
> + tucso = dest + ((void *)&(udp_hdr(skb)->check) - data);
> + *tucso = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
> + &ipv6_hdr(skb)->daddr, 0,
> + IPPROTO_UDP, 0);
This is dead code in the current state of this driver.
> + } else {
> + return false;
> + }
> } else if (protocol == htons(ETH_P_IPV6)) {
> tucso = dest + skb_checksum_start_offset(skb) + skb->csum_offset;
> *tucso = ~csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
> --
> 2.40.1
>
Powered by blists - more mailing lists