Re: [syzbot] [mm?] WARNING in __folio_rmap_sanity_checks (2)

[David Hildenbrand <david@redhat.com>]

On 10.01.25 17:14, Matthew Wilcox wrote:
> On Fri, Jan 10, 2025 at 04:48:03PM +0100, David Hildenbrand wrote:
>> On 28.12.24 13:25, syzbot wrote:
>>> syzbot has found a reproducer for the following issue on:
>>>
>>> HEAD commit:    8155b4ef3466 Add linux-next specific files for 20241220
>>> git tree:       linux-next
>>> console output: https://syzkaller.appspot.com/x/log.txt?x=1661050f980000
>>> kernel config:  https://syzkaller.appspot.com/x/.config?x=9c90bb7161a56c88
>>> dashboard link: https://syzkaller.appspot.com/bug?extid=c0673e1f1f054fac28c2
>>> compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
>>> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=17438af8580000
>>> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=101006df980000
>>>
>>> Downloadable assets:
>>> disk image: https://storage.googleapis.com/syzbot-assets/98a974fc662d/disk-8155b4ef.raw.xz
>>> vmlinux: https://storage.googleapis.com/syzbot-assets/2dea9b72f624/vmlinux-8155b4ef.xz
>>> kernel image: https://storage.googleapis.com/syzbot-assets/593a42b9eb34/bzImage-8155b4ef.xz
>>> mounted in repro: https://storage.googleapis.com/syzbot-assets/5f780361c9ef/mount_0.gz
>>>
>>> IMPORTANT: if you fix the issue, please add the following tag to the commit:
>>> Reported-by: syzbot+c0673e1f1f054fac28c2@syzkaller.appspotmail.com
>>>
>>>    xfs_vn_setattr+0x25d/0x320 fs/xfs/xfs_iops.c:1065
>>>    notify_change+0xbca/0xe90 fs/attr.c:552
>>>    do_truncate+0x220/0x310 fs/open.c:65
>>>    do_ftruncate+0x4a1/0x540 fs/open.c:192
>>>    do_sys_ftruncate fs/open.c:207 [inline]
>>>    __do_sys_ftruncate fs/open.c:212 [inline]
>>>    __se_sys_ftruncate fs/open.c:210 [inline]
>>>    __x64_sys_ftruncate+0x94/0xf0 fs/open.c:210
>>>    do_syscall_x64 arch/x86/entry/common.c:52 [inline]
>>>    do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
>>>    entry_SYSCALL_64_after_hwframe+0x77/0x7f
>>> ------------[ cut here ]------------
>>> WARNING: CPU: 1 PID: 11276 at ./include/linux/rmap.h:217 __folio_rmap_sanity_checks+0x369/0x590 include/linux/rmap.h:217
>>> Modules linked in:
>>> CPU: 1 UID: 0 PID: 11276 Comm: syz-executor139 Not tainted 6.13.0-rc3-next-20241220-syzkaller #0
>>> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 09/13/2024
>>> RIP: 0010:__folio_rmap_sanity_checks+0x369/0x590 include/linux/rmap.h:217
>>> Code: 0f 0b 90 e9 e9 fd ff ff e8 64 cb ab ff 48 ff cb e9 34 fe ff ff e8 57 cb ab ff 4c 89 e7 48 c7 c6 e0 a7 15 8c e8 08 a4 f5 ff 90 <0f> 0b 90 e9 25 fe ff ff e8 3a cb ab ff 4c 89 e7 48 c7 c6 40 a9 15
>>> RSP: 0018:ffffc9000e67efd8 EFLAGS: 00010246
>>> RAX: 8577b516ce8a9400 RBX: ffffea0001a58080 RCX: ffffc9000e67eb03
>>> RDX: 0000000000000005 RSI: ffffffff8c0aaba0 RDI: ffffffff8c5fed00
>>> RBP: 00000000000024c0 R08: ffffffff901ab1f7 R09: 1ffffffff203563e
>>> R10: dffffc0000000000 R11: fffffbfff203563f R12: ffffea0001a50000
>>> R13: ffffea0001a55c00 R14: 0000000000000000 R15: 0000000000000093
>>> FS:  00007f885c85f6c0(0000) GS:ffff8880b8700000(0000) knlGS:0000000000000000
>>> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>> CR2: 00007f88545b7000 CR3: 000000007fea2000 CR4: 00000000003526f0
>>> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
>>> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>>> Call Trace:
>>>    <TASK>
>>>    __folio_add_rmap mm/rmap.c:1170 [inline]
>>>    __folio_add_file_rmap mm/rmap.c:1489 [inline]
>>>    folio_add_file_rmap_ptes+0x82/0x380 mm/rmap.c:1511
>>>    set_pte_range+0x30c/0x750 mm/memory.c:5136
>>
>> If I would have to guess, I would assume that we have a refcount issue such
>> that we succeed in splitting a folio while concurrently mapping it.
> 
> That would seem hard to accomplish, because both hold the folio lock,
> so it wouldn't be just a refcount bug but also a locking bug.  Not sure
> what this is though.

Yeah, but we also have

https://lkml.kernel.org/r/6774bf44.050a0220.25abdd.098a.GAE@google.com

-- 
Cheers,

David / dhildenb