Date:   Mon, 12 Mar 2018 23:04:48 -0700
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     Yonghong Song <yhs@...com>, Daniel Borkmann <daniel@...earbox.net>,
        Eric Dumazet <eric.dumazet@...il.com>,
        Alexei Starovoitov <ast@...com>,
        netdev <netdev@...r.kernel.org>, Martin Lau <kafai@...com>
Subject: Re: BUG_ON triggered in skb_segment



On 03/12/2018 10:45 PM, Yonghong Song wrote:
> Hi,
> 
> One of our in-house projects, bpf-based NAT, hits a kernel BUG_ON at
> net-next function skb_segment, line 3667.
> 
> 3472 struct sk_buff *skb_segment(struct sk_buff *head_skb,
> 3473                             netdev_features_t features)
> 3474 {
> 3475         struct sk_buff *segs = NULL;
> 3476         struct sk_buff *tail = NULL;
> ...
> 3665                 while (pos < offset + len) {
> 3666                         if (i >= nfrags) {
> 3667                                 BUG_ON(skb_headlen(list_skb));
> 3668
> 3669                                 i = 0;
> 3670                                 nfrags = 
> skb_shinfo(list_skb)->nr_frags;
> 3671                                 frag = skb_shinfo(list_skb)->frags;
> 3672                                 frag_skb = list_skb;
> ...
> 
> call stack:
> ...
> #0 [ffff883ffef034f8] machine_kexec at ffffffff81044c41
>   #1 [ffff883ffef03558] __crash_kexec at ffffffff8110c525
>   #2 [ffff883ffef03620] crash_kexec at ffffffff8110d5cc
>   #3 [ffff883ffef03640] oops_end at ffffffff8101d7e7
>   #4 [ffff883ffef03668] die at ffffffff8101deb2
>   #5 [ffff883ffef03698] do_trap at ffffffff8101a700
>   #6 [ffff883ffef036e8] do_error_trap at ffffffff8101abfe
>   #7 [ffff883ffef037a0] do_invalid_op at ffffffff8101acd0
>   #8 [ffff883ffef037b0] invalid_op at ffffffff81a00bab
>      [exception RIP: skb_segment+3044]
>      RIP: ffffffff817e4dd4  RSP: ffff883ffef03860  RFLAGS: 00010216
>      RAX: 0000000000002bf6  RBX: ffff883feb7aaa00  RCX: 0000000000000011
>      RDX: ffff883fb87910c0  RSI: 0000000000000011  RDI: ffff883feb7ab500
>      RBP: ffff883ffef03928   R8: 0000000000002ce2   R9: 00000000000027da
>      R10: 000001ea00000000  R11: 0000000000002d82  R12: ffff883f90a1ee80
>      R13: ffff883fb8791120  R14: ffff883feb7abc00  R15: 0000000000002ce2
>      ORIG_RAX: ffffffffffffffff  CS: 0010  SS: 0018
>   #9 [ffff883ffef03930] tcp_gso_segment at ffffffff818713e7
> #10 [ffff883ffef03990] tcp4_gso_segment at ffffffff818717d8
> #11 [ffff883ffef039b0] inet_gso_segment at ffffffff81882c9b
> #12 [ffff883ffef03a10] skb_mac_gso_segment at ffffffff817f39b8
> #13 [ffff883ffef03a38] __skb_gso_segment at ffffffff817f3ac9
> #14 [ffff883ffef03a68] validate_xmit_skb at ffffffff817f3eed
> #15 [ffff883ffef03aa8] validate_xmit_skb_list at ffffffff817f40a2
> #16 [ffff883ffef03ad8] sch_direct_xmit at ffffffff81824efb
> #17 [ffff883ffef03b20] __qdisc_run at ffffffff818251aa
> #18 [ffff883ffef03b90] __dev_queue_xmit at ffffffff817f45ed
> #19 [ffff883ffef03c08] dev_queue_xmit at ffffffff817f4b90
> #20 [ffff883ffef03c18] __bpf_redirect at ffffffff81812b66
> #21 [ffff883ffef03c40] skb_do_redirect at ffffffff81813209
> #22 [ffff883ffef03c60] __netif_receive_skb_core at ffffffff817f310d
> #23 [ffff883ffef03cc8] __netif_receive_skb at ffffffff817f32e8
> #24 [ffff883ffef03ce8] netif_receive_skb_internal at ffffffff817f5538
> #25 [ffff883ffef03d10] napi_gro_complete at ffffffff817f56c0
> #26 [ffff883ffef03d28] dev_gro_receive at ffffffff817f5ea6
> #27 [ffff883ffef03d78] napi_gro_receive at ffffffff817f6168
> #28 [ffff883ffef03da0] mlx5e_handle_rx_cqe_mpwrq at ffffffff817381c2
> #29 [ffff883ffef03e30] mlx5e_poll_rx_cq at ffffffff817386c2
> #30 [ffff883ffef03e80] mlx5e_napi_poll at ffffffff8173926e
> #31 [ffff883ffef03ed0] net_rx_action at ffffffff817f5a6e
> #32 [ffff883ffef03f48] __softirqentry_text_start at ffffffff81c000c3
> #33 [ffff883ffef03fa8] irq_exit at ffffffff8108f515
> #34 [ffff883ffef03fb8] do_IRQ at ffffffff81a01b11
> --- <IRQ stack> ---
> bt: cannot transition from IRQ stack to current process stack:
>          IRQ stack pointer: ffff883ffef034f8
>      process stack pointer: ffffffff81a01ae9
>         current stack base: ffffc9000c5c4000
> ...
> Setup:
> =====
> 
> The test will involve three machines:
>    M_ipv6 <-> M_nat <-> M_ipv4
> 
> The M_nat will do ipv4<->ipv6 address translation and then forward packet
> to proper destination. The control plane will configure M_nat properly
> will understand virtual ipv4 address for machine M_ipv6, and
> virtual ipv6 address for machine M_ipv4.
> 
> M_nat runs a bpf program, which is attached to clsact (ingress) qdisc.
> The program uses bpf_skb_change_proto to do protocol conversion.
> bpf_skb_change_proto will adjust skb header_len and len properly
> based on protocol change.
> After the conversion, the program will make proper change on
> ethhdr and ip4/6 header, recalculate checksum, and send the packet out
> through bpf_redirect.
> 
> Experiment:
> ===========
> 
> MTU: 1500B for all three machines.
> 
> The tso/lro/gro are enabled on the M_nat box.
> 
> ping works on both ways of M_ipv6 <-> M_ipv4.
> It works for transfering a small file (4KB) between M_ipv6 and M_ipv4 
> (both ways).
> Transfering a large file (e.g., 4MB) from M_ipv6 to M_ipv4, failed with 
> the above BUG_ON, really fast.
> Did not really test from M_ipv4 to M_ipv6 with large file.
> 
> The error path likely to be (also from the above call stack):
>    nic -> lro/gro -> bpf_program -> gso (BUG_ON)
> 
> In one of experiments, I explicitly printed the skb->len and 
> skb->data_len. The values are below:
>    skb_segment: len 2856, data_len 2686
> They should be equal to avoid BUG.
> 
> In another experiment, I got:
>    skb_segment: len 1428, data_len 1258
> 
> In both cases, the difference is 170 bytes. Not sure whether
> this is just a coincidence or not.
> 
> Workaround:
> ===========
> 
> A workaround to avoid BUG_ON is to disable lro/gro. This way,
> kernel will not receive big packets and hence gso is not really called.
> 
> I am not familiar with gso code. Does anybody hit this BUG_ON before?
> Any suggestion on how to debug this?
> 

skb_segment() works if incoming GRO packet is not modified in its geometry.

In your case it seems you had to adjust gso_size (calling 
skb_decrease_gso_size() or skb_increase_gso_size()), and this breaks 
skb_segment() badly, because geometry changes, unless you had specific 
MTU/MSS restrictions.

You will have to make skb_segment() more generic if you really want this.

Powered by blists - more mailing lists