Re: WARNING in ip_recv_error

[Eric Dumazet <eric.dumazet@gmail.com>]

On 05/18/2018 05:08 AM, DaeRyong Jeong wrote:
> We report the crash: WARNING in ip_recv_error
> (I resend the email since I mistakenly missed the subject in my previous
> email. I'm sorry.)
> 
> 
> This crash has been found in v4.17-rc1 using RaceFuzzer (a modified
> version of Syzkaller), which we describe more at the end of this
> report. Our analysis shows that the race occurs when invoking two
> syscalls concurrently, setsockopt$inet6_IPV6_ADDRFORM and recvmsg.
> 
> 
> Diagnosis:
> We think the concurrent execution of do_ipv6_setsockopt() with optname
> IPV6_ADDRFORM and inet_recvmsg() causes the crash. do_ipv6_setsockopt()
> can update sk->prot to &udp_prot and sk->sk_family to PF_INET. But
> inet_recvmsg() can execute sk->sk_prot->recvmsg() right after that
> sk->prot is updated and sk->sk_family is not updated by
> do_ipv6_setsockopt(). This will lead WARN_ON in ip_recv_error().
> 
> 
> Thread interleaving:
> CPU0 (do_ipv6_setsockopt)			CPU1 (inet_recvmsg)
> =====						=====
> struct proto *prot = &udp_prot;
> ...
> sk->sk_prot = prot;
> sk->sk_socket->ops = &inet_dgram_ops;
> 						err = sk->sk_prot->recvmsg(sk, msg, size, flags & MSG_DONTWAIT,
> 									flags & ~MSG_DONTWAIT, &addr_len);
> 
> 						(in udp_recvmsg)
> 						if (flags & MSG_ERRQUEUE)
> 							return ip_recv_error(sk, msg, len, addr_len);
> 	
> 						(in ip_recv_error)
> 						WARN_ON_ONCE(sk->sk_family == AF_INET6);
> sk->sk_family = PF_INET;
> 
> 
> Call Sequence:
> CPU0
> =====
> udpv6_setsockopt
> 	ipv6_setsockopt
> 		do_ipv6_setsockopt
> 
> CPU1
> =====
> sock_recvmsg
> 	sock_recvmsg_nosec
> 		inet_recvmsg
> 			udp_recvmsg
> 
> 
> ==================================================================
> WARNING: CPU: 1 PID: 32600 at /home/daeryong/workspace/new-race-fuzzer/kernels_repo/kernel_v4.17-rc1/net/ipv4/ip_sockglue.c:508 ip_recv_error+0x6f2/0x720 net/ipv4/ip_sockglue.c:508
> Kernel panic - not syncing: panic_on_warn set ...
> 
> CPU: 1 PID: 32600 Comm: syz-executor0 Not tainted 4.17.0-rc1 #1
> Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.8.2-0-g33fbe13 by qemu-project.org 04/01/2014
> Call Trace:
>  __dump_stack lib/dump_stack.c:77 [inline]
>  dump_stack+0x166/0x21c lib/dump_stack.c:113
>  panic+0x1a0/0x3a7 kernel/panic.c:184
>  __warn+0x191/0x1a0 kernel/panic.c:536
>  report_bug+0x132/0x1b0 lib/bug.c:186
>  fixup_bug.part.11+0x28/0x50 arch/x86/kernel/traps.c:178
>  fixup_bug arch/x86/kernel/traps.c:247 [inline]
>  do_error_trap+0x28b/0x2d0 arch/x86/kernel/traps.c:296
>  do_invalid_op+0x1b/0x20 arch/x86/kernel/traps.c:315
>  invalid_op+0x14/0x20 arch/x86/entry/entry_64.S:992
> RIP: 0010:ip_recv_error+0x6f2/0x720 net/ipv4/ip_sockglue.c:508
> RSP: 0018:ffff8801dadff630 EFLAGS: 00010212
> RAX: 0000000000040000 RBX: 0000000000002002 RCX: ffffffff8327de12
> RDX: 000000000000008a RSI: ffffc90001a0c000 RDI: ffff8801be615010
> RBP: ffff8801dadff720 R08: 0000000000002002 R09: ffff8801dadff918
> R10: ffff8801dadff738 R11: ffff8801dadffaff R12: ffff8801be615000
> R13: ffff8801dadffd50 R14: 1ffff1003b5bfece R15: ffff8801dadffb90
>  udp_recvmsg+0x834/0xa10 net/ipv4/udp.c:1571
>  inet_recvmsg+0x121/0x420 net/ipv4/af_inet.c:830
>  sock_recvmsg_nosec net/socket.c:802 [inline]
>  sock_recvmsg+0x7f/0xa0 net/socket.c:809
>  ___sys_recvmsg+0x1f0/0x430 net/socket.c:2279
>  __sys_recvmsg+0xfc/0x1c0 net/socket.c:2328
>  __do_sys_recvmsg net/socket.c:2338 [inline]
>  __se_sys_recvmsg net/socket.c:2335 [inline]
>  __x64_sys_recvmsg+0x48/0x50 net/socket.c:2335
>  do_syscall_64+0x15f/0x4a0 arch/x86/entry/common.c:287
>  entry_SYSCALL_64_after_hwframe+0x49/0xbe
> RIP: 0033:0x4563f9
> RSP: 002b:00007f24f6927b28 EFLAGS: 00000246 ORIG_RAX: 000000000000002f
> RAX: ffffffffffffffda RBX: 000000000072bfa0 RCX: 00000000004563f9
> RDX: 0000000000002002 RSI: 0000000020000240 RDI: 0000000000000016
> RBP: 00000000000004e4 R08: 0000000000000000 R09: 0000000000000000
> R10: 0000000000000000 R11: 0000000000000246 R12: 00007f24f69286d4
> R13: 00000000ffffffff R14: 00000000006fc600 R15: 0000000000000000
> Dumping ftrace buffer:
>    (ftrace buffer empty)
> Kernel Offset: disabled
> Rebooting in 86400 seconds..
> ==================================================================
> 
> 
> = About RaceFuzzer
> 
> RaceFuzzer is a customized version of Syzkaller, specifically tailored
> to find race condition bugs in the Linux kernel. While we leverage
> many different technique, the notable feature of RaceFuzzer is in
> leveraging a custom hypervisor (QEMU/KVM) to interleave the
> scheduling. In particular, we modified the hypervisor to intentionally
> stall a per-core execution, which is similar to supporting per-core
> breakpoint functionality. This allows RaceFuzzer to force the kernel
> to deterministically trigger racy condition (which may rarely happen
> in practice due to randomness in scheduling).
> 
> RaceFuzzer's C repro always pinpoints two racy syscalls. Since C
> repro's scheduling synchronization should be performed at the user
> space, its reproducibility is limited (reproduction may take from 1
> second to 10 minutes (or even more), depending on a bug). This is
> because, while RaceFuzzer precisely interleaves the scheduling at the
> kernel's instruction level when finding this bug, C repro cannot fully
> utilize such a feature. Please disregard all code related to
> "should_hypercall" in the C repro, as this is only for our debugging
> purposes using our own hypervisor.
> 


We probably need to revert Willem patch (7ce875e5ecb8562fd44040f69bda96c999e38bbc)