general protection fault in ip6_sublist_rcv

general protection fault in ip6_sublist_rcv

[syzbot]

Hello,

syzbot found the following crash on:

HEAD commit:    12d61c69 Add linux-next specific files for 20191024
git tree:       linux-next
console output: https://syzkaller.appspot.com/x/log.txt?x=10608674e00000
kernel config:  https://syzkaller.appspot.com/x/.config?x=afb75fd8c9fd5ed8
dashboard link: https://syzkaller.appspot.com/bug?extid=c54f457cad330e57e967
compiler:       gcc (GCC) 9.0.0 20181231 (experimental)
syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=134be44ce00000
C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=15938e18e00000

The bug was bisected to:

commit ca58fbe06c54795f00db79e447f94c2028d30124
Author: Florian Westphal <fw@strlen.de>
Date:   Thu Oct 10 22:30:37 2019 +0000

     netfilter: add and use nf_hook_slow_list()

bisection log:  https://syzkaller.appspot.com/x/bisect.txt?x=16366800e00000
final crash:    https://syzkaller.appspot.com/x/report.txt?x=15366800e00000
console output: https://syzkaller.appspot.com/x/log.txt?x=11366800e00000

IMPORTANT: if you fix the bug, please add the following tag to the commit:
Reported-by: syzbot+c54f457cad330e57e967@syzkaller.appspotmail.com
Fixes: ca58fbe06c54 ("netfilter: add and use nf_hook_slow_list()")

kasan: CONFIG_KASAN_INLINE enabled
kasan: GPF could be caused by NULL-ptr deref or user memory access
general protection fault: 0000 [#1] PREEMPT SMP KASAN
CPU: 1 PID: 9027 Comm: syz-executor906 Not tainted 5.4.0-rc4-next-20191024  
#0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS  
Google 01/01/2011
RIP: 0010:__read_once_size include/linux/compiler.h:199 [inline]
RIP: 0010:NF_HOOK_LIST include/linux/netfilter.h:331 [inline]
RIP: 0010:ip6_sublist_rcv+0x5c9/0x930 net/ipv6/ip6_input.c:292
Code: 0f 85 73 01 00 00 e8 06 63 24 fb 48 8b 85 00 ff ff ff 48 8d b8 10 0f  
00 00 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f  
85 44 03 00 00 48 8b 85 00 ff ff ff 4c 8b a0 10 0f
RSP: 0018:ffff8880956af3c0 EFLAGS: 00010206
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffffff864f0e0e
RDX: 00000000000001e2 RSI: ffffffff864f0c6a RDI: 0000000000000f10
RBP: ffff8880956af4f0 R08: ffff8880a1778040 R09: ffffed1015d26b7d
R10: ffffed1015d26b7c R11: ffff8880ae935be3 R12: 0000000000000001
R13: ffff8880956af4c8 R14: ffff8880956af558 R15: ffff8880956af680
FS:  0000000000819880(0000) GS:ffff8880ae900000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000020000780 CR3: 0000000089dec000 CR4: 00000000001406e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
  ipv6_list_rcv+0x373/0x4b0 net/ipv6/ip6_input.c:328
  __netif_receive_skb_list_ptype net/core/dev.c:5274 [inline]
  __netif_receive_skb_list_core+0x5fc/0x9d0 net/core/dev.c:5322
  __netif_receive_skb_list net/core/dev.c:5374 [inline]
  netif_receive_skb_list_internal+0x7eb/0xe50 net/core/dev.c:5469
  gro_normal_list.part.0+0x1e/0xb0 net/core/dev.c:5892
  gro_normal_list net/core/dev.c:5905 [inline]
  gro_normal_one+0x184/0x1d0 net/core/dev.c:5904
  napi_frags_finish net/core/dev.c:6008 [inline]
  napi_gro_frags+0x915/0xd00 net/core/dev.c:6081
  tun_get_user+0x2e8e/0x3f80 drivers/net/tun.c:1976
  tun_chr_write_iter+0xbd/0x156 drivers/net/tun.c:2022
  call_write_iter include/linux/fs.h:1902 [inline]
  do_iter_readv_writev+0x5f8/0x8f0 fs/read_write.c:693
  do_iter_write fs/read_write.c:970 [inline]
  do_iter_write+0x184/0x610 fs/read_write.c:951
  vfs_writev+0x1b3/0x2f0 fs/read_write.c:1015
  do_writev+0x15b/0x330 fs/read_write.c:1058
  __do_sys_writev fs/read_write.c:1131 [inline]
  __se_sys_writev fs/read_write.c:1128 [inline]
  __x64_sys_writev+0x75/0xb0 fs/read_write.c:1128
  do_syscall_64+0xfa/0x760 arch/x86/entry/common.c:290
  entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x441d80
Code: 05 48 3d 01 f0 ff ff 0f 83 5d 09 fc ff c3 66 2e 0f 1f 84 00 00 00 00  
00 66 90 83 3d 91 92 29 00 00 75 14 b8 14 00 00 00 0f 05 <48> 3d 01 f0 ff  
ff 0f 83 34 09 fc ff c3 48 83 ec 08 e8 ba 2b 00 00
RSP: 002b:00007ffc73f8e2e8 EFLAGS: 00000246 ORIG_RAX: 0000000000000014
RAX: ffffffffffffffda RBX: 00007ffc73f8e300 RCX: 0000000000441d80
RDX: 0000000000000001 RSI: 00007ffc73f8e330 RDI: 00000000000000f0
RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000004
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000012dae
R13: 0000000000402bd0 R14: 0000000000000000 R15: 0000000000000000
Modules linked in:
---[ end trace b92638c9e1a03392 ]---
RIP: 0010:__read_once_size include/linux/compiler.h:199 [inline]
RIP: 0010:NF_HOOK_LIST include/linux/netfilter.h:331 [inline]
RIP: 0010:ip6_sublist_rcv+0x5c9/0x930 net/ipv6/ip6_input.c:292
Code: 0f 85 73 01 00 00 e8 06 63 24 fb 48 8b 85 00 ff ff ff 48 8d b8 10 0f  
00 00 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f  
85 44 03 00 00 48 8b 85 00 ff ff ff 4c 8b a0 10 0f
RSP: 0018:ffff8880956af3c0 EFLAGS: 00010206
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffffff864f0e0e
RDX: 00000000000001e2 RSI: ffffffff864f0c6a RDI: 0000000000000f10
RBP: ffff8880956af4f0 R08: ffff8880a1778040 R09: ffffed1015d26b7d
R10: ffffed1015d26b7c R11: ffff8880ae935be3 R12: 0000000000000001
R13: ffff8880956af4c8 R14: ffff8880956af558 R15: ffff8880956af680
FS:  0000000000819880(0000) GS:ffff8880ae900000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000020000780 CR3: 0000000089dec000 CR4: 00000000001406e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400


---
This bug is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzkaller@googlegroups.com.

syzbot will keep track of this bug report. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
For information about bisection process see: https://goo.gl/tpsmEJ#bisection
syzbot can test patches for this bug, for details see:
https://goo.gl/tpsmEJ#testing-patches

[PATCH net-next] inet: do not call sublist_rcv on empty list

[Florian Westphal]

syzbot triggered struct net NULL deref in NF_HOOK_LIST:
RIP: 0010:NF_HOOK_LIST include/linux/netfilter.h:331 [inline]
RIP: 0010:ip6_sublist_rcv+0x5c9/0x930 net/ipv6/ip6_input.c:292
 ipv6_list_rcv+0x373/0x4b0 net/ipv6/ip6_input.c:328
 __netif_receive_skb_list_ptype net/core/dev.c:5274 [inline]

Reason:
void ipv6_list_rcv(struct list_head *head, struct packet_type *pt,
                   struct net_device *orig_dev)
[..]
        list_for_each_entry_safe(skb, next, head, list) {
		/* iterates list */
                skb = ip6_rcv_core(skb, dev, net);
		/* ip6_rcv_core drops skb -> NULL is returned */
                if (skb == NULL)
                        continue;
	[..]
	}
	/* sublist is empty -> curr_net is NULL */
        ip6_sublist_rcv(&sublist, curr_dev, curr_net);

Before the recent change NF_HOOK_LIST did a list iteration before
struct net deref, i.e. it was a no-op in the empty list case.

List iteration now happens after *net deref, causing crash.

Follow the same pattern as the ip(v6)_list_rcv loop and add a list_empty
test for the final sublist dispatch too.

Cc: Edward Cree <ecree@solarflare.com>
Reported-by: syzbot+c54f457cad330e57e967@syzkaller.appspotmail.com
Fixes: ca58fbe06c54 ("netfilter: add and use nf_hook_slow_list()")
Signed-off-by: Florian Westphal <fw@strlen.de>
---
 net/ipv4/ip_input.c  | 3 ++-
 net/ipv6/ip6_input.c | 3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/ip_input.c b/net/ipv4/ip_input.c
index c59a78a267c3..24a95126e698 100644
--- a/net/ipv4/ip_input.c
+++ b/net/ipv4/ip_input.c
@@ -611,5 +611,6 @@ void ip_list_rcv(struct list_head *head, struct packet_type *pt,
 		list_add_tail(&skb->list, &sublist);
 	}
 	/* dispatch final sublist */
-	ip_sublist_rcv(&sublist, curr_dev, curr_net);
+	if (!list_empty(&sublist))
+		ip_sublist_rcv(&sublist, curr_dev, curr_net);
 }
diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c
index 3d71c7d6102c..ef7f707d9ae3 100644
--- a/net/ipv6/ip6_input.c
+++ b/net/ipv6/ip6_input.c
@@ -325,7 +325,8 @@ void ipv6_list_rcv(struct list_head *head, struct packet_type *pt,
 		list_add_tail(&skb->list, &sublist);
 	}
 	/* dispatch final sublist */
-	ip6_sublist_rcv(&sublist, curr_dev, curr_net);
+	if (!list_empty(&sublist))
+		ip6_sublist_rcv(&sublist, curr_dev, curr_net);
 }
 
 INDIRECT_CALLABLE_DECLARE(int udpv6_rcv(struct sk_buff *));
-- 
2.23.0

Re: [PATCH net-next] inet: do not call sublist_rcv on empty list

[Leon Romanovsky]

On Tue, Oct 29, 2019 at 01:44:04AM +0100, Florian Westphal wrote:
> syzbot triggered struct net NULL deref in NF_HOOK_LIST:
> RIP: 0010:NF_HOOK_LIST include/linux/netfilter.h:331 [inline]
> RIP: 0010:ip6_sublist_rcv+0x5c9/0x930 net/ipv6/ip6_input.c:292
>  ipv6_list_rcv+0x373/0x4b0 net/ipv6/ip6_input.c:328
>  __netif_receive_skb_list_ptype net/core/dev.c:5274 [inline]
>
> Reason:
> void ipv6_list_rcv(struct list_head *head, struct packet_type *pt,
>                    struct net_device *orig_dev)
> [..]
>         list_for_each_entry_safe(skb, next, head, list) {
> 		/* iterates list */
>                 skb = ip6_rcv_core(skb, dev, net);
> 		/* ip6_rcv_core drops skb -> NULL is returned */
>                 if (skb == NULL)
>                         continue;
> 	[..]
> 	}
> 	/* sublist is empty -> curr_net is NULL */
>         ip6_sublist_rcv(&sublist, curr_dev, curr_net);
>
> Before the recent change NF_HOOK_LIST did a list iteration before
> struct net deref, i.e. it was a no-op in the empty list case.
>
> List iteration now happens after *net deref, causing crash.
>
> Follow the same pattern as the ip(v6)_list_rcv loop and add a list_empty
> test for the final sublist dispatch too.
>
> Cc: Edward Cree <ecree@solarflare.com>
> Reported-by: syzbot+c54f457cad330e57e967@syzkaller.appspotmail.com
> Fixes: ca58fbe06c54 ("netfilter: add and use nf_hook_slow_list()")
> Signed-off-by: Florian Westphal <fw@strlen.de>
> ---
>  net/ipv4/ip_input.c  | 3 ++-
>  net/ipv6/ip6_input.c | 3 ++-
>  2 files changed, 4 insertions(+), 2 deletions(-)
>

It fixed my crash on boot.

Thanks,
Tested-by: Leon Romanovsky <leonro@mellanox.com>

Re: [PATCH net-next] inet: do not call sublist_rcv on empty list

[Nikolay Aleksandrov]

On 29/10/2019 02:44, Florian Westphal wrote:
> syzbot triggered struct net NULL deref in NF_HOOK_LIST:
> RIP: 0010:NF_HOOK_LIST include/linux/netfilter.h:331 [inline]
> RIP: 0010:ip6_sublist_rcv+0x5c9/0x930 net/ipv6/ip6_input.c:292
>  ipv6_list_rcv+0x373/0x4b0 net/ipv6/ip6_input.c:328
>  __netif_receive_skb_list_ptype net/core/dev.c:5274 [inline]
> 
> Reason:
> void ipv6_list_rcv(struct list_head *head, struct packet_type *pt,
>                    struct net_device *orig_dev)
> [..]
>         list_for_each_entry_safe(skb, next, head, list) {
> 		/* iterates list */
>                 skb = ip6_rcv_core(skb, dev, net);
> 		/* ip6_rcv_core drops skb -> NULL is returned */
>                 if (skb == NULL)
>                         continue;
> 	[..]
> 	}
> 	/* sublist is empty -> curr_net is NULL */
>         ip6_sublist_rcv(&sublist, curr_dev, curr_net);
> 
> Before the recent change NF_HOOK_LIST did a list iteration before
> struct net deref, i.e. it was a no-op in the empty list case.
> 
> List iteration now happens after *net deref, causing crash.
> 
> Follow the same pattern as the ip(v6)_list_rcv loop and add a list_empty
> test for the final sublist dispatch too.
> 
> Cc: Edward Cree <ecree@solarflare.com>
> Reported-by: syzbot+c54f457cad330e57e967@syzkaller.appspotmail.com
> Fixes: ca58fbe06c54 ("netfilter: add and use nf_hook_slow_list()")
> Signed-off-by: Florian Westphal <fw@strlen.de>
> ---
>  net/ipv4/ip_input.c  | 3 ++-
>  net/ipv6/ip6_input.c | 3 ++-
>  2 files changed, 4 insertions(+), 2 deletions(-)
> 

Fixed my boot problem as well.

Tested-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>

Re: [PATCH net-next] inet: do not call sublist_rcv on empty list

[David Miller]

From: Florian Westphal <fw@strlen.de>
Date: Tue, 29 Oct 2019 01:44:04 +0100

> syzbot triggered struct net NULL deref in NF_HOOK_LIST:
> RIP: 0010:NF_HOOK_LIST include/linux/netfilter.h:331 [inline]
> RIP: 0010:ip6_sublist_rcv+0x5c9/0x930 net/ipv6/ip6_input.c:292
>  ipv6_list_rcv+0x373/0x4b0 net/ipv6/ip6_input.c:328
>  __netif_receive_skb_list_ptype net/core/dev.c:5274 [inline]
> 
> Reason:
> void ipv6_list_rcv(struct list_head *head, struct packet_type *pt,
>                    struct net_device *orig_dev)
> [..]
>         list_for_each_entry_safe(skb, next, head, list) {
> 		/* iterates list */
>                 skb = ip6_rcv_core(skb, dev, net);
> 		/* ip6_rcv_core drops skb -> NULL is returned */
>                 if (skb == NULL)
>                         continue;
> 	[..]
> 	}
> 	/* sublist is empty -> curr_net is NULL */
>         ip6_sublist_rcv(&sublist, curr_dev, curr_net);
> 
> Before the recent change NF_HOOK_LIST did a list iteration before
> struct net deref, i.e. it was a no-op in the empty list case.
> 
> List iteration now happens after *net deref, causing crash.
> 
> Follow the same pattern as the ip(v6)_list_rcv loop and add a list_empty
> test for the final sublist dispatch too.
> 
> Cc: Edward Cree <ecree@solarflare.com>
> Reported-by: syzbot+c54f457cad330e57e967@syzkaller.appspotmail.com
> Fixes: ca58fbe06c54 ("netfilter: add and use nf_hook_slow_list()")
> Signed-off-by: Florian Westphal <fw@strlen.de>

Applied.