[syzbot ci] Re: mm: Unconditional per-VMA locks and cleanups

1 view
Skip to first unread message

syzbot ci

unread,
4:24 PM (5 hours ago) 4:24 PM
to ak...@linux-foundation.org, alic...@google.com, ar...@android.com, chri...@brauner.io, cmll...@google.com, dave....@linux.intel.com, da...@davemloft.net, dsa...@kernel.org, gre...@linuxfoundation.org, liam.h...@oracle.com, linux-...@vger.kernel.org, linu...@kvack.org, l...@kernel.org, net...@vger.kernel.org, shakee...@linux.dev, sur...@google.com, tk...@android.com, vba...@kernel.org, syz...@lists.linux.dev, syzkall...@googlegroups.com
syzbot ci has tested the following series

[v2] mm: Unconditional per-VMA locks and cleanups
https://lore.kernel.org/all/20260610230...@davehans-spike.ostc.intel.com
* [PATCH v2 1/5] mm: Make per-VMA locks available universally
* [PATCH v2 2/5] binder: Make shrinker rely solely on per-VMA lock
* [PATCH v2 3/5] mm: Add RCU-based VMA lookup helper that waits for writers
* [PATCH v2 4/5] binder: Remove mmap_lock fallback
* [PATCH v2 5/5] tcp: Remove mmap_lock fallback path

and found the following issue:
general protection fault in tcp_zerocopy_receive

Full report is available here:
https://ci.syzbot.org/series/3e6d125a-b2ae-49a4-b833-babfb8bc9150

***

general protection fault in tcp_zerocopy_receive

tree: net-next
URL: https://kernel.googlesource.com/pub/scm/linux/kernel/git/netdev/net-next.git
base: c8459ee2fef502d6ef6c063751c33d9ac7943eab
arch: amd64
compiler: Debian clang version 21.1.8 (++20251221033036+2078da43e25a-1~exp1~20251221153213.50), Debian LLD 21.1.8
config: https://ci.syzbot.org/builds/e7d66981-7900-4c3d-b992-664ccd13a57e/config
syz repro: https://ci.syzbot.org/findings/59d09544-f280-48fe-8ca9-a2fd8225e9df/syz_repro

Oops: general protection fault, probably for non-canonical address 0xdffffc0000000002: 0000 [#1] SMP KASAN PTI
KASAN: null-ptr-deref in range [0x0000000000000010-0x0000000000000017]
CPU: 0 UID: 0 PID: 5876 Comm: syz.1.18 Not tainted syzkaller #0 PREEMPT(full)
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.2-debian-1.16.2-1 04/01/2014
RIP: 0010:vma_start_read_locked_nested include/linux/mmap_lock.h:240 [inline]
RIP: 0010:vma_start_read_locked+0xa0/0x300 include/linux/mmap_lock.h:257
Code: 28 84 c0 0f 85 2b 02 00 00 44 8b 35 ba 0d 1b 0e 31 ff 44 89 f6 e8 c0 e2 af ff 45 85 f6 74 48 4c 8d 73 10 4c 89 f0 48 c1 e8 03 <42> 80 3c 28 00 74 08 4c 89 f7 e8 b1 37 1b 00 bf 38 03 00 00 49 03
RSP: 0018:ffffc9000399f4a0 EFLAGS: 00010202
RAX: 0000000000000002 RBX: 0000000000000000 RCX: 0000000000000000
RDX: ffff88816d558000 RSI: 0000000000000001 RDI: 0000000000000000
RBP: ffffc9000399f538 R08: ffff8881076c64df R09: 1ffff11020ed8c9b
R10: dffffc0000000000 R11: ffffed1020ed8c9c R12: 1ffff92000733e94
R13: dffffc0000000000 R14: 0000000000000010 R15: 0000000000011000
FS: 00007f4866b966c0(0000) GS:ffff88818dc86000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000200000011000 CR3: 0000000117c8a000 CR4: 00000000000006f0
Call Trace:
<TASK>
vma_start_read_unlocked+0x3f/0x70 mm/mmap_lock.c:362
find_tcp_vma net/ipv4/tcp.c:2173 [inline]
tcp_zerocopy_receive+0x762/0x2200 net/ipv4/tcp.c:2227
do_tcp_getsockopt+0x2079/0x2940 net/ipv4/tcp.c:4758
tcp_getsockopt+0x83/0x130 net/ipv4/tcp.c:4856
do_sock_getsockopt+0x51d/0x7e0 net/socket.c:2487
__sys_getsockopt net/socket.c:2518 [inline]
__do_sys_getsockopt net/socket.c:2525 [inline]
__se_sys_getsockopt net/socket.c:2522 [inline]
__x64_sys_getsockopt+0x1a4/0x240 net/socket.c:2522
do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
do_syscall_64+0x174/0x580 arch/x86/entry/syscall_64.c:94
entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f4865d9ce59
Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 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 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 e8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f4866b96028 EFLAGS: 00000246 ORIG_RAX: 0000000000000037
RAX: ffffffffffffffda RBX: 00007f4866015fa0 RCX: 00007f4865d9ce59
RDX: 0000000000000023 RSI: 0000000000000006 RDI: 0000000000000003
RBP: 00007f4865e32d6f R08: 0000200000000380 R09: 0000000000000000
R10: 0000200000000340 R11: 0000000000000246 R12: 0000000000000000
R13: 00007f4866016038 R14: 00007f4866015fa0 R15: 00007ffc748de268
</TASK>
Modules linked in:
---[ end trace 0000000000000000 ]---
RIP: 0010:vma_start_read_locked_nested include/linux/mmap_lock.h:240 [inline]
RIP: 0010:vma_start_read_locked+0xa0/0x300 include/linux/mmap_lock.h:257
Code: 28 84 c0 0f 85 2b 02 00 00 44 8b 35 ba 0d 1b 0e 31 ff 44 89 f6 e8 c0 e2 af ff 45 85 f6 74 48 4c 8d 73 10 4c 89 f0 48 c1 e8 03 <42> 80 3c 28 00 74 08 4c 89 f7 e8 b1 37 1b 00 bf 38 03 00 00 49 03
RSP: 0018:ffffc9000399f4a0 EFLAGS: 00010202
RAX: 0000000000000002 RBX: 0000000000000000 RCX: 0000000000000000
RDX: ffff88816d558000 RSI: 0000000000000001 RDI: 0000000000000000
RBP: ffffc9000399f538 R08: ffff8881076c64df R09: 1ffff11020ed8c9b
R10: dffffc0000000000 R11: ffffed1020ed8c9c R12: 1ffff92000733e94
R13: dffffc0000000000 R14: 0000000000000010 R15: 0000000000011000
FS: 00007f4866b966c0(0000) GS:ffff88818dc86000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f4865dea540 CR3: 0000000117c8a000 CR4: 00000000000006f0
----------------
Code disassembly (best guess):
0: 28 84 c0 0f 85 2b 02 sub %al,0x22b850f(%rax,%rax,8)
7: 00 00 add %al,(%rax)
9: 44 8b 35 ba 0d 1b 0e mov 0xe1b0dba(%rip),%r14d # 0xe1b0dca
10: 31 ff xor %edi,%edi
12: 44 89 f6 mov %r14d,%esi
15: e8 c0 e2 af ff call 0xffafe2da
1a: 45 85 f6 test %r14d,%r14d
1d: 74 48 je 0x67
1f: 4c 8d 73 10 lea 0x10(%rbx),%r14
23: 4c 89 f0 mov %r14,%rax
26: 48 c1 e8 03 shr $0x3,%rax
* 2a: 42 80 3c 28 00 cmpb $0x0,(%rax,%r13,1) <-- trapping instruction
2f: 74 08 je 0x39
31: 4c 89 f7 mov %r14,%rdi
34: e8 b1 37 1b 00 call 0x1b37ea
39: bf 38 03 00 00 mov $0x338,%edi
3e: 49 rex.WB
3f: 03 .byte 0x3


***

If these findings have caused you to resend the series or submit a
separate fix, please add the following tag to your commit message:
Tested-by: syz...@syzkaller.appspotmail.com

---
This report is generated by a bot. It may contain errors.
syzbot ci engineers can be reached at syzk...@googlegroups.com.

To test a patch for this bug, please reply with `#syz test`
(should be on a separate line).

The patch should be attached to the email.
Note: arguments like custom git repos and branches are not supported.
Reply all
Reply to author
Forward
0 new messages