[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1418920017.9773.80.camel@edumazet-glaptop2.roam.corp.google.com>
Date: Thu, 18 Dec 2014 08:26:57 -0800
From: Eric Dumazet <eric.dumazet@...il.com>
To: Govindarajulu Varadarajan <_govind@....com>
Cc: davem@...emloft.net, netdev@...r.kernel.org, ssujith@...co.com,
benve@...co.com, Jiri Benc <jbenc@...hat.com>,
Stefan Assmann <sassmann@...hat.com>
Subject: Re: [PATCH net] enic: fix rx skb checksum
On Thu, 2014-12-18 at 15:58 +0530, Govindarajulu Varadarajan wrote:
> Hardware always provides compliment of IP pseudo checksum. Stack expects
> whole packet checksum without pseudo checksum if CHECKSUM_COMPLETE is set.
>
> This causes checksum error in nf & ovs.
>
> kernel: qg-19546f09-f2: hw csum failure
> kernel: CPU: 9 PID: 0 Comm: swapper/9 Tainted: GF O-------------- 3.10.0-123.8.1.el7.x86_64 #1
> kernel: Hardware name: Cisco Systems Inc UCSB-B200-M3/UCSB-B200-M3, BIOS B200M3.2.2.3.0.080820141339 08/08/2014
> kernel: ffff881218f40000 df68243feb35e3a8 ffff881237a43ab8 ffffffff815e237b
> kernel: ffff881237a43ad0 ffffffff814cd4ca ffff8829ec71eb00 ffff881237a43af0
> kernel: ffffffff814c6232 0000000000000286 ffff8829ec71eb00 ffff881237a43b00
> kernel: Call Trace:
> kernel: <IRQ> [<ffffffff815e237b>] dump_stack+0x19/0x1b
> kernel: [<ffffffff814cd4ca>] netdev_rx_csum_fault+0x3a/0x40
> kernel: [<ffffffff814c6232>] __skb_checksum_complete_head+0x62/0x70
> kernel: [<ffffffff814c6251>] __skb_checksum_complete+0x11/0x20
> kernel: [<ffffffff8155a20c>] nf_ip_checksum+0xcc/0x100
> kernel: [<ffffffffa049edc7>] icmp_error+0x1f7/0x35c [nf_conntrack_ipv4]
> kernel: [<ffffffff814cf419>] ? netif_rx+0xb9/0x1d0
> kernel: [<ffffffffa040eb7b>] ? internal_dev_recv+0xdb/0x130 [openvswitch]
> kernel: [<ffffffffa04c8330>] nf_conntrack_in+0xf0/0xa80 [nf_conntrack]
> kernel: [<ffffffff81509380>] ? inet_del_offload+0x40/0x40
> kernel: [<ffffffffa049e302>] ipv4_conntrack_in+0x22/0x30 [nf_conntrack_ipv4]
> kernel: [<ffffffff815005ca>] nf_iterate+0xaa/0xc0
> kernel: [<ffffffff81509380>] ? inet_del_offload+0x40/0x40
> kernel: [<ffffffff81500664>] nf_hook_slow+0x84/0x140
> kernel: [<ffffffff81509380>] ? inet_del_offload+0x40/0x40
> kernel: [<ffffffff81509dd4>] ip_rcv+0x344/0x380
>
> Hardware verifies IP & tcp/udp header checksum but does not provide payload
> checksum, use CHECKSUM_UNNECESSARY. Set it only if its valid IP tcp/udp packet.
>
> Cc: Jiri Benc <jbenc@...hat.com>
> Cc: Stefan Assmann <sassmann@...hat.com>
> Reported-by: Sunil Choudhary <schoudha@...hat.com>
> Signed-off-by: Govindarajulu Varadarajan <_govind@....com>
> ---
> drivers/net/ethernet/cisco/enic/enic_main.c | 12 ++++++++----
> 1 file changed, 8 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c
> index 868d0f6..705f334 100644
> --- a/drivers/net/ethernet/cisco/enic/enic_main.c
> +++ b/drivers/net/ethernet/cisco/enic/enic_main.c
> @@ -1060,10 +1060,14 @@ static void enic_rq_indicate_buf(struct vnic_rq *rq,
> PKT_HASH_TYPE_L4 : PKT_HASH_TYPE_L3);
> }
>
> - if ((netdev->features & NETIF_F_RXCSUM) && !csum_not_calc) {
> - skb->csum = htons(checksum);
> - skb->ip_summed = CHECKSUM_COMPLETE;
> - }
> + /* Hardware does not provide whole packet checksum. It only
> + * provides pseudo checksum. Since hw validates the packet
> + * checksum but not provide us the checksum value. use
> + * CHECSUM_UNNECESSARY.
> + */
> + if ((netdev->features & NETIF_F_RXCSUM) && tcp_udp_csum_ok &&
> + ipv4_csum_ok)
> + skb->ip_summed = CHECKSUM_UNNECESSARY;
>
> if (vlan_stripped)
> __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), vlan_tci);
Hmm.. this looks like CHECKSUM_COMPLETE could be kept for tunneling
cases.
Check commit f8c6455bb04b944e ("net/mlx4_en: Extend checksum offloading
by CHECKSUM COMPLETE") for a start.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists