netdev - Re: [PATCH net] net: thunderx: Fix TCP/UDP checksum offload for IPv6 pkts

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: <1511366238.16595.1.camel@gmail.com>
Date:   Wed, 22 Nov 2017 07:57:18 -0800
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     Aleksey Makarov <aleksey.makarov@...iga.com>,
        netdev@...r.kernel.org
Cc:     linux-kernel@...r.kernel.org,
        "Goutham, Sunil" <Sunil.Goutham@...ium.com>,
        Robert Richter <rric@...nel.org>,
        Florian Westphal <fw@...len.de>,
        Aleksey Makarov <aleksey.makarov@...ium.com>,
        Sunil Goutham <sgoutham@...ium.com>,
        linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH net] net: thunderx: Fix TCP/UDP checksum offload for
 IPv6 pkts

On Wed, 2017-11-22 at 15:37 +0300, Aleksey Makarov wrote:
> From: Sunil Goutham <sgoutham@...ium.com>
> 
> This fixes a previous patch which missed some changes
> and due to which L3 checksum offload was getting enabled
> for IPv6 pkts. And HW is dropping these pkts as it assumes
> the pkt is IPv4 when IP csum offload is set in the SQ
> descriptor.
> 
> Fixes: bbbb494fd005 ("net: thunderx: Enable TSO and checksum offloads
> for ipv6")
> Signed-off-by: Sunil Goutham <sgoutham@...ium.com>
> Signed-off-by: Aleksey Makarov <aleksey.makarov@...iga.com>
> ---
>  drivers/net/ethernet/cavium/thunder/nicvf_queues.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
> b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
> index d4496e9afcdf..184d5bdbe7e0 100644
> --- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
> +++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
> @@ -1355,10 +1355,11 @@ nicvf_sq_add_hdr_subdesc(struct nicvf *nic,
> struct snd_queue *sq, int qentry,
>  
>  	/* Offload checksum calculation to HW */
>  	if (skb->ip_summed == CHECKSUM_PARTIAL) {
> -		hdr->csum_l3 = 1; /* Enable IP csum calculation */
>  		hdr->l3_offset = skb_network_offset(skb);
>  		hdr->l4_offset = skb_transport_offset(skb);
>  
> +		/* Enable IP HDR csum calculation for V4 pkts */
> +		hdr->csum_l3 = (ip.v4->version == 4) ? 1 : 0;

Have you tried to set hdr->csum_l3 to 0 regardless of version being 4
or 6 ?

This would remove the need for yet another conditional.

AFAIK, linux does not offload IPv4 header checksums to NIC, it is not
worth the trouble.