kernel bug found and suggestions for fixing it

Commit Message

ffhgfv March 4, 2025, 7:31 a.m. UTC

Hello, I found a bug titled "KASAN: null-ptr-deref Read in smc_tcp_syn_recv_sock" with modified syzkaller in the lasted upstream related to bcachefs file system.
If you fix this issue, please add the following tag to the commit:  Reported-by: Jianzhou Zhao <[e-mail]>,    xingwei lee <[e-mail]>, Zhizhuo Tang <[e-mail]>

------------[ cut here ]------------
TITLE: KASAN: null-ptr-deref Read in smc_tcp_syn_recv_sock
==================================================================
BUG: KASAN: null-ptr-deref in instrument_atomic_read include/linux/instrumented.h:68 [inline]
BUG: KASAN: null-ptr-deref in atomic_read include/linux/atomic/atomic-instrumented.h:32 [inline]
BUG: KASAN: null-ptr-deref in smc_tcp_syn_recv_sock+0xa7/0x4c0 net/smc/af_smc.c:131
Read of size 4 at addr 0000000000000a04 by task syz.7.21/12319

CPU: 1 UID: 0 PID: 12319 Comm: syz.7.21 Not tainted 6.14.0-rc5-dirty #2
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
Call Trace:
 <irq>
 __dump_stack lib/dump_stack.c:94 [inline]
 dump_stack_lvl+0x116/0x1b0 lib/dump_stack.c:120
 kasan_report+0xbd/0xf0 mm/kasan/report.c:634
 check_region_inline mm/kasan/generic.c:183 [inline]
 kasan_check_range+0xf4/0x1a0 mm/kasan/generic.c:189
 instrument_atomic_read include/linux/instrumented.h:68 [inline]
 atomic_read include/linux/atomic/atomic-instrumented.h:32 [inline]
 smc_tcp_syn_recv_sock+0xa7/0x4c0 net/smc/af_smc.c:131
 tcp_check_req+0x5e4/0x1a90 net/ipv4/tcp_minisocks.c:861
 tcp_v4_rcv+0x1753/0x44e0 net/ipv4/tcp_ipv4.c:2274
 ip_protocol_deliver_rcu+0xba/0x4c0 net/ipv4/ip_input.c:205
 ip_local_deliver_finish+0x320/0x570 net/ipv4/ip_input.c:233
 NF_HOOK include/linux/netfilter.h:314 [inline]
 NF_HOOK include/linux/netfilter.h:308 [inline]
 ip_local_deliver+0x19a/0x200 net/ipv4/ip_input.c:254
 dst_input include/net/dst.h:469 [inline]
 ip_rcv_finish net/ipv4/ip_input.c:447 [inline]
 NF_HOOK include/linux/netfilter.h:314 [inline]
 NF_HOOK include/linux/netfilter.h:308 [inline]
 ip_rcv+0x2be/0x5d0 net/ipv4/ip_input.c:567
 __netif_receive_skb_one_core+0x19b/0x1f0 net/core/dev.c:5893
 __netif_receive_skb+0x1d/0x170 net/core/dev.c:6006
 process_backlog+0x319/0x1460 net/core/dev.c:6354
 __napi_poll.constprop.0+0xb6/0x540 net/core/dev.c:7188
 napi_poll net/core/dev.c:7257 [inline]
 net_rx_action+0x9d2/0xe30 net/core/dev.c:7379
 handle_softirqs+0x1d1/0x870 kernel/softirq.c:561
 do_softirq kernel/softirq.c:462 [inline]
 do_softirq+0xac/0xe0 kernel/softirq.c:449
 </irq>
 <task>
 __local_bh_enable_ip+0x100/0x120 kernel/softirq.c:389
 local_bh_enable include/linux/bottom_half.h:33 [inline]
 rcu_read_unlock_bh include/linux/rcupdate.h:919 [inline]
 __dev_queue_xmit+0x1b7a/0x4120 net/core/dev.c:4676
 dev_queue_xmit include/linux/netdevice.h:3313 [inline]
 neigh_hh_output include/net/neighbour.h:523 [inline]
 neigh_output include/net/neighbour.h:537 [inline]
 ip_finish_output2+0xc1c/0x1f10 net/ipv4/ip_output.c:236
 __ip_finish_output net/ipv4/ip_output.c:314 [inline]
 __ip_finish_output+0x442/0x940 net/ipv4/ip_output.c:296
 ip_finish_output+0x35/0x380 net/ipv4/ip_output.c:324
 NF_HOOK_COND include/linux/netfilter.h:303 [inline]
 ip_output+0x146/0x2b0 net/ipv4/ip_output.c:434
 dst_output include/net/dst.h:459 [inline]
 ip_local_out net/ipv4/ip_output.c:130 [inline]
 __ip_queue_xmit+0x19ee/0x21f0 net/ipv4/ip_output.c:528
 __tcp_transmit_skb+0x2a55/0x3e70 net/ipv4/tcp_output.c:1471
 __tcp_send_ack.part.0+0x39c/0x720 net/ipv4/tcp_output.c:4275
 __tcp_send_ack net/ipv4/tcp_output.c:4281 [inline]
 tcp_send_ack+0x81/0xa0 net/ipv4/tcp_output.c:4281
 tcp_rcv_synsent_state_process net/ipv4/tcp_input.c:6600 [inline]
 tcp_rcv_state_process+0x40e2/0x4c80 net/ipv4/tcp_input.c:6794
 tcp_v4_do_rcv+0x1a8/0xa70 net/ipv4/tcp_ipv4.c:1941
 sk_backlog_rcv include/net/sock.h:1122 [inline]
 __release_sock+0x31d/0x400 net/core/sock.c:3123
 release_sock+0x5a/0x220 net/core/sock.c:3677
 tcp_sendmsg+0x3a/0x50 net/ipv4/tcp.c:1359
 inet_sendmsg+0xb9/0x150 net/ipv4/af_inet.c:851
 smc_sendmsg+0x22a/0x530 net/smc/af_smc.c:2796
 sock_sendmsg_nosec net/socket.c:718 [inline]
 __sock_sendmsg net/socket.c:733 [inline]
 ____sys_sendmsg+0xab8/0xc70 net/socket.c:2573
 ___sys_sendmsg+0x11d/0x1c0 net/socket.c:2627
 __sys_sendmsg+0x151/0x200 net/socket.c:2659
 do_syscall_x64 arch/x86/entry/common.c:52 [inline]
 do_syscall_64+0xcb/0x250 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f36ef9a962d
Code: 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 &lt;48&gt; 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f36f08c5f98 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
RAX: ffffffffffffffda RBX: 00007f36efbc5f80 RCX: 00007f36ef9a962d
RDX: 0000000024040049 RSI: 0000000020000200 RDI: 0000000000000004
RBP: 00007f36efa4e373 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 0000000000000000 R14: 00007f36efbc5f80 R15: 00007f36f08a6000
 </task>
==================================================================
I use the same kernel as syzbot instance upstream: 7eb172143d5508b4da468ed59ee857c6e5e01da6
kernel config: https://syzkaller.appspot.com/text?tag=KernelConfig&amp;x=da4b04ae798b7ef6
compiler: gcc version 11.4.0
===============================================================================
Unfortunately, the modified syzkaller does not generate an effective repeat program.
The following is my analysis of the bug and repair suggestions, hoping to help with the repair of the bug:
## Root cause analysis
1. ** null pointer access ** : 'smc_tcp_syn_recv_sock' function does not check whether the 'sk' pointer is valid, direct access to its member 'sk_wmem_alloc'.
2. ** Missing initialization ** : The 'sk' pointer may not be initialized correctly or returned in advance in the wrong path (such as memory allocation failure or SMC negotiation failure), resulting in a null pointer for subsequent operations.

### Repair suggestions
1. ** null pointer check ** : add 'if (!) before accessing' sk 'member sk) return NULL; '.
2. ** Error path handling ** : Ensure that resources are cleaned up in time when 'inet_csk_clone' fails to avoid passing invalid Pointers.
Patch example:

=========================================================================
I hope it helps.
Best regards
Jianzhou Zhao
xingwei lee
Zhizhuo Tang</[e-mail]></[e-mail]></[e-mail]>

Patch

diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -128,6 +128,8 @@ 
 	struct sock *child;
 
 	smc = smc_clcsock_user_data(sk);
+	if (!smc)
+		goto drop;  // Ensure that the smc pointer is valid before accessing its members
 
 	if (READ_ONCE(sk->sk_ack_backlog) + atomic_read(&smc->queued_smc_hs) >
 	    sk->sk_max_ack_backlog)