Re: [PATCH net 2/7] net: ip6_gre: Fix headroom request in ip6erspan_tunnel_xmit()

2018-05-17 Thread Petr Machata
Fixes: e41c7c68ea77 ("ip6erspan: make sure enough headroom at xmit.")


Re: [PATCH net 2/7] net: ip6_gre: Fix headroom request in ip6erspan_tunnel_xmit()

2018-05-17 Thread William Tu
On Thu, May 17, 2018 at 7:36 AM, Petr Machata  wrote:
> dev->needed_headroom is not primed until ip6_tnl_xmit(), so it starts
> out zero. Thus the call to skb_cow_head() fails to actually make sure
> there's enough headroom to push the ERSPAN headers to. That can lead to
> the panic cited below. (Reproducer below that).
>
> Fix by requesting either needed_headroom if already primed, or just the
> bare minimum needed for the header otherwise.
>
> [  190.703567] kernel BUG at net/core/skbuff.c:104!
> [  190.708384] invalid opcode:  [#1] PREEMPT SMP KASAN PTI
> [  190.714007] Modules linked in: act_mirred cls_matchall ip6_gre ip6_tunnel 
> tunnel6 gre sch_ingress vrf veth x86_pkg_temp_thermal mlx_platform nfsd 
> e1000e leds_mlxcpld
> [  190.728975] CPU: 1 PID: 959 Comm: kworker/1:2 Not tainted 
> 4.17.0-rc4-net_master-custom-139 #10
> [  190.737647] Hardware name: Mellanox Technologies Ltd. 
> "MSN2410-CB2F"/"SA000874", BIOS 4.6.5 03/08/2016
> [  190.747006] Workqueue: ipv6_addrconf addrconf_dad_work
> [  190.75] RIP: 0010:skb_panic+0xc3/0x100
> [  190.756358] RSP: 0018:8801d54072f0 EFLAGS: 00010282
> [  190.761629] RAX: 0085 RBX: 8801c1a8ecc0 RCX: 
> 
> [  190.768830] RDX: 0085 RSI: dc00 RDI: 
> ed003aa80e54
> [  190.776025] RBP: 8801bd1ec5a0 R08: ed003aabce19 R09: 
> ed003aabce19
> [  190.783226] R10: 0001 R11: ed003aabce18 R12: 
> 8801bf695dbe
> [  190.790418] R13: 0084 R14: 06c0 R15: 
> 8801bf695dc8
> [  190.797621] FS:  () GS:8801d540() 
> knlGS:
> [  190.805786] CS:  0010 DS:  ES:  CR0: 80050033
> [  190.811582] CR2: 55fa929aced0 CR3: 03228004 CR4: 
> 001606e0
> [  190.818790] Call Trace:
> [  190.821264]  
> [  190.823314]  ? ip6erspan_tunnel_xmit+0x5e4/0x1982 [ip6_gre]
> [  190.828940]  ? ip6erspan_tunnel_xmit+0x5e4/0x1982 [ip6_gre]
> [  190.834562]  skb_push+0x78/0x90
> [  190.837749]  ip6erspan_tunnel_xmit+0x5e4/0x1982 [ip6_gre]
> [  190.843219]  ? ip6gre_tunnel_ioctl+0xd90/0xd90 [ip6_gre]
> [  190.848577]  ? debug_check_no_locks_freed+0x210/0x210
> [  190.853679]  ? debug_check_no_locks_freed+0x210/0x210
> [  190.858783]  ? print_irqtrace_events+0x120/0x120
> [  190.863451]  ? sched_clock_cpu+0x18/0x210
> [  190.867496]  ? cyc2ns_read_end+0x10/0x10
> [  190.871474]  ? skb_network_protocol+0x76/0x200
> [  190.875977]  dev_hard_start_xmit+0x137/0x770
> [  190.880317]  ? do_raw_spin_trylock+0x6d/0xa0
> [  190.884624]  sch_direct_xmit+0x2ef/0x5d0
> [  190.888589]  ? pfifo_fast_dequeue+0x3fa/0x670
> [  190.892994]  ? pfifo_fast_change_tx_queue_len+0x810/0x810
> [  190.898455]  ? __lock_is_held+0xa0/0x160
> [  190.902422]  __qdisc_run+0x39e/0xfc0
> [  190.906041]  ? _raw_spin_unlock+0x29/0x40
> [  190.910090]  ? pfifo_fast_enqueue+0x24b/0x3e0
> [  190.914501]  ? sch_direct_xmit+0x5d0/0x5d0
> [  190.918658]  ? pfifo_fast_dequeue+0x670/0x670
> [  190.923047]  ? __dev_queue_xmit+0x172/0x1770
> [  190.927365]  ? preempt_count_sub+0xf/0xd0
> [  190.931421]  __dev_queue_xmit+0x410/0x1770
> [  190.935553]  ? ___slab_alloc+0x605/0x930
> [  190.939524]  ? print_irqtrace_events+0x120/0x120
> [  190.944186]  ? memcpy+0x34/0x50
> [  190.947364]  ? netdev_pick_tx+0x1c0/0x1c0
> [  190.951428]  ? __skb_clone+0x2fd/0x3d0
> [  190.955218]  ? __copy_skb_header+0x270/0x270
> [  190.959537]  ? rcu_read_lock_sched_held+0x93/0xa0
> [  190.964282]  ? kmem_cache_alloc+0x344/0x4d0
> [  190.968520]  ? cyc2ns_read_end+0x10/0x10
> [  190.972495]  ? skb_clone+0x123/0x230
> [  190.976112]  ? skb_split+0x820/0x820
> [  190.979747]  ? tcf_mirred+0x554/0x930 [act_mirred]
> [  190.984582]  tcf_mirred+0x554/0x930 [act_mirred]
> [  190.989252]  ? tcf_mirred_act_wants_ingress.part.2+0x10/0x10 [act_mirred]
> [  190.996109]  ? __lock_acquire+0x706/0x26e0
> [  191.000239]  ? sched_clock_cpu+0x18/0x210
> [  191.004294]  tcf_action_exec+0xcf/0x2a0
> [  191.008179]  tcf_classify+0xfa/0x340
> [  191.011794]  __netif_receive_skb_core+0x8e1/0x1c60
> [  191.016630]  ? debug_check_no_locks_freed+0x210/0x210
> [  191.021732]  ? nf_ingress+0x500/0x500
> [  191.025458]  ? process_backlog+0x347/0x4b0
> [  191.029619]  ? print_irqtrace_events+0x120/0x120
> [  191.034302]  ? lock_acquire+0xd8/0x320
> [  191.038089]  ? process_backlog+0x1b6/0x4b0
> [  191.042246]  ? process_backlog+0xc2/0x4b0
> [  191.046303]  process_backlog+0xc2/0x4b0
> [  191.050189]  net_rx_action+0x5cc/0x980
> [  191.053991]  ? napi_complete_done+0x2c0/0x2c0
> [  191.058386]  ? mark_lock+0x13d/0xb40
> [  191.062001]  ? clockevents_program_event+0x6b/0x1d0
> [  191.066922]  ? print_irqtrace_events+0x120/0x120
> [  191.071593]  ? __lock_is_held+0xa0/0x160
> [  191.075566]  __do_softirq+0x1d4/0x9d2
> [  191.079282]  ? ip6_finish_output2+0x524/0x1460
> [  191.083771]  do_softirq_own_stack+0x2a/0x40
> [  191.087994]  
> [  191.090130]  do_softirq.part.13+0x38/0x40
> [  191.09417

[PATCH net 2/7] net: ip6_gre: Fix headroom request in ip6erspan_tunnel_xmit()

2018-05-17 Thread Petr Machata
dev->needed_headroom is not primed until ip6_tnl_xmit(), so it starts
out zero. Thus the call to skb_cow_head() fails to actually make sure
there's enough headroom to push the ERSPAN headers to. That can lead to
the panic cited below. (Reproducer below that).

Fix by requesting either needed_headroom if already primed, or just the
bare minimum needed for the header otherwise.

[  190.703567] kernel BUG at net/core/skbuff.c:104!
[  190.708384] invalid opcode:  [#1] PREEMPT SMP KASAN PTI
[  190.714007] Modules linked in: act_mirred cls_matchall ip6_gre ip6_tunnel 
tunnel6 gre sch_ingress vrf veth x86_pkg_temp_thermal mlx_platform nfsd e1000e 
leds_mlxcpld
[  190.728975] CPU: 1 PID: 959 Comm: kworker/1:2 Not tainted 
4.17.0-rc4-net_master-custom-139 #10
[  190.737647] Hardware name: Mellanox Technologies Ltd. 
"MSN2410-CB2F"/"SA000874", BIOS 4.6.5 03/08/2016
[  190.747006] Workqueue: ipv6_addrconf addrconf_dad_work
[  190.75] RIP: 0010:skb_panic+0xc3/0x100
[  190.756358] RSP: 0018:8801d54072f0 EFLAGS: 00010282
[  190.761629] RAX: 0085 RBX: 8801c1a8ecc0 RCX: 
[  190.768830] RDX: 0085 RSI: dc00 RDI: ed003aa80e54
[  190.776025] RBP: 8801bd1ec5a0 R08: ed003aabce19 R09: ed003aabce19
[  190.783226] R10: 0001 R11: ed003aabce18 R12: 8801bf695dbe
[  190.790418] R13: 0084 R14: 06c0 R15: 8801bf695dc8
[  190.797621] FS:  () GS:8801d540() 
knlGS:
[  190.805786] CS:  0010 DS:  ES:  CR0: 80050033
[  190.811582] CR2: 55fa929aced0 CR3: 03228004 CR4: 001606e0
[  190.818790] Call Trace:
[  190.821264]  
[  190.823314]  ? ip6erspan_tunnel_xmit+0x5e4/0x1982 [ip6_gre]
[  190.828940]  ? ip6erspan_tunnel_xmit+0x5e4/0x1982 [ip6_gre]
[  190.834562]  skb_push+0x78/0x90
[  190.837749]  ip6erspan_tunnel_xmit+0x5e4/0x1982 [ip6_gre]
[  190.843219]  ? ip6gre_tunnel_ioctl+0xd90/0xd90 [ip6_gre]
[  190.848577]  ? debug_check_no_locks_freed+0x210/0x210
[  190.853679]  ? debug_check_no_locks_freed+0x210/0x210
[  190.858783]  ? print_irqtrace_events+0x120/0x120
[  190.863451]  ? sched_clock_cpu+0x18/0x210
[  190.867496]  ? cyc2ns_read_end+0x10/0x10
[  190.871474]  ? skb_network_protocol+0x76/0x200
[  190.875977]  dev_hard_start_xmit+0x137/0x770
[  190.880317]  ? do_raw_spin_trylock+0x6d/0xa0
[  190.884624]  sch_direct_xmit+0x2ef/0x5d0
[  190.888589]  ? pfifo_fast_dequeue+0x3fa/0x670
[  190.892994]  ? pfifo_fast_change_tx_queue_len+0x810/0x810
[  190.898455]  ? __lock_is_held+0xa0/0x160
[  190.902422]  __qdisc_run+0x39e/0xfc0
[  190.906041]  ? _raw_spin_unlock+0x29/0x40
[  190.910090]  ? pfifo_fast_enqueue+0x24b/0x3e0
[  190.914501]  ? sch_direct_xmit+0x5d0/0x5d0
[  190.918658]  ? pfifo_fast_dequeue+0x670/0x670
[  190.923047]  ? __dev_queue_xmit+0x172/0x1770
[  190.927365]  ? preempt_count_sub+0xf/0xd0
[  190.931421]  __dev_queue_xmit+0x410/0x1770
[  190.935553]  ? ___slab_alloc+0x605/0x930
[  190.939524]  ? print_irqtrace_events+0x120/0x120
[  190.944186]  ? memcpy+0x34/0x50
[  190.947364]  ? netdev_pick_tx+0x1c0/0x1c0
[  190.951428]  ? __skb_clone+0x2fd/0x3d0
[  190.955218]  ? __copy_skb_header+0x270/0x270
[  190.959537]  ? rcu_read_lock_sched_held+0x93/0xa0
[  190.964282]  ? kmem_cache_alloc+0x344/0x4d0
[  190.968520]  ? cyc2ns_read_end+0x10/0x10
[  190.972495]  ? skb_clone+0x123/0x230
[  190.976112]  ? skb_split+0x820/0x820
[  190.979747]  ? tcf_mirred+0x554/0x930 [act_mirred]
[  190.984582]  tcf_mirred+0x554/0x930 [act_mirred]
[  190.989252]  ? tcf_mirred_act_wants_ingress.part.2+0x10/0x10 [act_mirred]
[  190.996109]  ? __lock_acquire+0x706/0x26e0
[  191.000239]  ? sched_clock_cpu+0x18/0x210
[  191.004294]  tcf_action_exec+0xcf/0x2a0
[  191.008179]  tcf_classify+0xfa/0x340
[  191.011794]  __netif_receive_skb_core+0x8e1/0x1c60
[  191.016630]  ? debug_check_no_locks_freed+0x210/0x210
[  191.021732]  ? nf_ingress+0x500/0x500
[  191.025458]  ? process_backlog+0x347/0x4b0
[  191.029619]  ? print_irqtrace_events+0x120/0x120
[  191.034302]  ? lock_acquire+0xd8/0x320
[  191.038089]  ? process_backlog+0x1b6/0x4b0
[  191.042246]  ? process_backlog+0xc2/0x4b0
[  191.046303]  process_backlog+0xc2/0x4b0
[  191.050189]  net_rx_action+0x5cc/0x980
[  191.053991]  ? napi_complete_done+0x2c0/0x2c0
[  191.058386]  ? mark_lock+0x13d/0xb40
[  191.062001]  ? clockevents_program_event+0x6b/0x1d0
[  191.066922]  ? print_irqtrace_events+0x120/0x120
[  191.071593]  ? __lock_is_held+0xa0/0x160
[  191.075566]  __do_softirq+0x1d4/0x9d2
[  191.079282]  ? ip6_finish_output2+0x524/0x1460
[  191.083771]  do_softirq_own_stack+0x2a/0x40
[  191.087994]  
[  191.090130]  do_softirq.part.13+0x38/0x40
[  191.094178]  __local_bh_enable_ip+0x135/0x190
[  191.098591]  ip6_finish_output2+0x54d/0x1460
[  191.102916]  ? ip6_forward_finish+0x2f0/0x2f0
[  191.107314]  ? ip6_mtu+0x3c/0x2c0
[  191.110674]  ? ip6_finish_output+0x2f8/0x650
[  191.114992]  ? ip6_output+0x12a/0x500