| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <CAKFNMomJJm7cM5K5bjpPV_wcyUfi5Qvuasob=weB_7MMRtOQxw@mail.gmail.com>
Date: Sat, 10 May 2025 22:49:31 +0900
From: Ryusuke Konishi <konishi.ryusuke@...il.com>
To: syzbot <syzbot+81394db39b0e2ed2db06@...kaller.appspotmail.com>
Cc: eadavis@...com, syzkaller-bugs@...glegroups.com,
linux-kernel@...r.kernel.org
Subject: Re: [syzbot] [nilfs?] possible deadlock in nilfs_segctor_construct
On Sat, May 10, 2025 at 10:31 PM Ryusuke Konishi wrote:
>
> On Sat, May 10, 2025 at 9:46 PM syzbot
> <syzbot+81394db39b0e2ed2db06@...kaller.appspotmail.com> wrote:
> >
> > Hello,
> >
> > syzbot has tested the proposed patch but the reproducer is still triggering an issue:
> > possible deadlock in nilfs_mdt_get_block
> >
> > ======================================================
> > WARNING: possible circular locking dependency detected
> > 6.15.0-rc5-syzkaller-00136-g9c69f8884904-dirty #0 Not tainted
> > ------------------------------------------------------
> > segctord/5927 is trying to acquire lock:
> > ffff88803ffcc610 (sb_internal#2){.+.+}-{0:0}, at: nilfs_mdt_create_block fs/nilfs2/mdt.c:80 [inline]
> > ffff88803ffcc610 (sb_internal#2){.+.+}-{0:0}, at: nilfs_mdt_get_block+0x193/0xab0 fs/nilfs2/mdt.c:258
> >
> > but task is already holding lock:
> > ffff8880422660a0 (&nilfs_bmap_mdt_lock_key){++++}-{4:4}, at: nilfs_bmap_propagate+0x2f/0x130 fs/nilfs2/bmap.c:328
> >
> > which lock already depends on the new lock.
> >
> >
> > the existing dependency chain (in reverse order) is:
> >
> > -> #2 (&nilfs_bmap_mdt_lock_key){++++}-{4:4}:
> > lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5866
> > down_read+0x46/0x2e0 kernel/locking/rwsem.c:1524
> > nilfs_bmap_lookup_at_level+0x92/0x3d0 fs/nilfs2/bmap.c:65
> > nilfs_bmap_lookup fs/nilfs2/bmap.h:182 [inline]
> > nilfs_mdt_submit_block+0x3a3/0x830 fs/nilfs2/mdt.c:144
> > nilfs_mdt_read_block+0x46/0x320 fs/nilfs2/mdt.c:178
> > nilfs_mdt_get_block+0x11f/0xab0 fs/nilfs2/mdt.c:254
> > nilfs_palloc_get_block+0x18d/0x2b0 fs/nilfs2/alloc.c:239
> > nilfs_palloc_get_desc_block fs/nilfs2/alloc.c:296 [inline]
> > nilfs_palloc_prepare_alloc_entry+0x3a3/0xf40 fs/nilfs2/alloc.c:607
> > nilfs_ifile_create_inode+0x111/0x2e0 fs/nilfs2/ifile.c:61
> > nilfs_new_inode+0x1ed/0x750 fs/nilfs2/inode.c:321
> > nilfs_create+0x100/0x2f0 fs/nilfs2/namei.c:98
> > lookup_open fs/namei.c:3701 [inline]
> > open_last_lookups fs/namei.c:3800 [inline]
> > path_openat+0x14f1/0x3830 fs/namei.c:4036
> > do_filp_open+0x1fa/0x410 fs/namei.c:4066
> > do_sys_openat2+0x121/0x1c0 fs/open.c:1429
> > do_sys_open fs/open.c:1444 [inline]
> > __do_sys_openat fs/open.c:1460 [inline]
> > __se_sys_openat fs/open.c:1455 [inline]
> > __x64_sys_openat+0x138/0x170 fs/open.c:1455
> > do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
> > do_syscall_64+0xf6/0x210 arch/x86/entry/syscall_64.c:94
> > entry_SYSCALL_64_after_hwframe+0x77/0x7f
> >
> > -> #1 (&nilfs->ns_segctor_sem){++++}-{4:4}:
> > lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5866
> > down_read+0x46/0x2e0 kernel/locking/rwsem.c:1524
> > nilfs_transaction_begin+0x365/0x710 fs/nilfs2/segment.c:221
> > nilfs_create+0xc9/0x2f0 fs/nilfs2/namei.c:95
> > lookup_open fs/namei.c:3701 [inline]
> > open_last_lookups fs/namei.c:3800 [inline]
> > path_openat+0x14f1/0x3830 fs/namei.c:4036
> > do_filp_open+0x1fa/0x410 fs/namei.c:4066
> > do_sys_openat2+0x121/0x1c0 fs/open.c:1429
> > do_sys_open fs/open.c:1444 [inline]
> > __do_sys_openat fs/open.c:1460 [inline]
> > __se_sys_openat fs/open.c:1455 [inline]
> > __x64_sys_openat+0x138/0x170 fs/open.c:1455
> > do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
> > do_syscall_64+0xf6/0x210 arch/x86/entry/syscall_64.c:94
> > entry_SYSCALL_64_after_hwframe+0x77/0x7f
> >
> > -> #0 (sb_internal#2){.+.+}-{0:0}:
> > check_prev_add kernel/locking/lockdep.c:3166 [inline]
> > check_prevs_add kernel/locking/lockdep.c:3285 [inline]
> > validate_chain+0xb9b/0x2140 kernel/locking/lockdep.c:3909
> > __lock_acquire+0xaac/0xd20 kernel/locking/lockdep.c:5235
> > lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5866
> > percpu_down_read include/linux/percpu-rwsem.h:52 [inline]
> > __sb_start_write include/linux/fs.h:1783 [inline]
> > sb_start_intwrite include/linux/fs.h:1966 [inline]
> > nilfs_transaction_begin+0x268/0x710 fs/nilfs2/segment.c:218
> > nilfs_mdt_create_block fs/nilfs2/mdt.c:80 [inline]
> > nilfs_mdt_get_block+0x193/0xab0 fs/nilfs2/mdt.c:258
> > nilfs_palloc_get_block+0x18d/0x2b0 fs/nilfs2/alloc.c:239
> > nilfs_palloc_get_entry_block+0x8c/0xa0 fs/nilfs2/alloc.c:355
> > nilfs_dat_prepare_entry fs/nilfs2/dat.c:45 [inline]
> > nilfs_dat_prepare_alloc fs/nilfs2/dat.c:82 [inline]
> > nilfs_dat_prepare_update+0xb1/0x300 fs/nilfs2/dat.c:250
> > nilfs_direct_propagate+0x1ac/0x370 fs/nilfs2/direct.c:279
> > nilfs_bmap_propagate+0x77/0x130 fs/nilfs2/bmap.c:329
> > nilfs_collect_file_data+0x4f/0xd0 fs/nilfs2/segment.c:589
> > nilfs_segctor_apply_buffers+0x161/0x330 fs/nilfs2/segment.c:1010
> > nilfs_segctor_scan_file+0x68e/0x8e0 fs/nilfs2/segment.c:1059
> > nilfs_segctor_collect_blocks fs/nilfs2/segment.c:1267 [inline]
> > nilfs_segctor_collect fs/nilfs2/segment.c:1547 [inline]
> > nilfs_segctor_do_construct+0x1eac/0x6970 fs/nilfs2/segment.c:2122
> > nilfs_segctor_construct+0x17b/0x690 fs/nilfs2/segment.c:2478
> > nilfs_segctor_thread_construct fs/nilfs2/segment.c:2585 [inline]
> > nilfs_segctor_thread+0x6cd/0xe00 fs/nilfs2/segment.c:2700
> > kthread+0x70e/0x8a0 kernel/kthread.c:464
> > ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:153
> > ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
> >
> > other info that might help us debug this:
> >
> > Chain exists of:
> > sb_internal#2 --> &nilfs->ns_segctor_sem --> &nilfs_bmap_mdt_lock_key
> >
> > Possible unsafe locking scenario:
> >
> > CPU0 CPU1
> > ---- ----
> > lock(&nilfs_bmap_mdt_lock_key);
> > lock(&nilfs->ns_segctor_sem);
> > lock(&nilfs_bmap_mdt_lock_key);
> > rlock(sb_internal#2);
> >
> > *** DEADLOCK ***
> >
> > 1 lock held by segctord/5927:
> > #0: ffff8880422660a0 (&nilfs_bmap_mdt_lock_key){++++}-{4:4}, at: nilfs_bmap_propagate+0x2f/0x130 fs/nilfs2/bmap.c:328
> >
> > stack backtrace:
> > CPU: 0 UID: 0 PID: 5927 Comm: segctord Not tainted 6.15.0-rc5-syzkaller-00136-g9c69f8884904-dirty #0 PREEMPT(full)
> > Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014
> > Call Trace:
> > <TASK>
> > dump_stack_lvl+0x189/0x250 lib/dump_stack.c:120
> > print_circular_bug+0x2ee/0x310 kernel/locking/lockdep.c:2079
> > check_noncircular+0x134/0x160 kernel/locking/lockdep.c:2211
> > check_prev_add kernel/locking/lockdep.c:3166 [inline]
> > check_prevs_add kernel/locking/lockdep.c:3285 [inline]
> > validate_chain+0xb9b/0x2140 kernel/locking/lockdep.c:3909
> > __lock_acquire+0xaac/0xd20 kernel/locking/lockdep.c:5235
> > lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5866
> > percpu_down_read include/linux/percpu-rwsem.h:52 [inline]
> > __sb_start_write include/linux/fs.h:1783 [inline]
> > sb_start_intwrite include/linux/fs.h:1966 [inline]
> > nilfs_transaction_begin+0x268/0x710 fs/nilfs2/segment.c:218
> > nilfs_mdt_create_block fs/nilfs2/mdt.c:80 [inline]
> > nilfs_mdt_get_block+0x193/0xab0 fs/nilfs2/mdt.c:258
> > nilfs_palloc_get_block+0x18d/0x2b0 fs/nilfs2/alloc.c:239
> > nilfs_palloc_get_entry_block+0x8c/0xa0 fs/nilfs2/alloc.c:355
> > nilfs_dat_prepare_entry fs/nilfs2/dat.c:45 [inline]
> > nilfs_dat_prepare_alloc fs/nilfs2/dat.c:82 [inline]
> > nilfs_dat_prepare_update+0xb1/0x300 fs/nilfs2/dat.c:250
> > nilfs_direct_propagate+0x1ac/0x370 fs/nilfs2/direct.c:279
> > nilfs_bmap_propagate+0x77/0x130 fs/nilfs2/bmap.c:329
> > nilfs_collect_file_data+0x4f/0xd0 fs/nilfs2/segment.c:589
> > nilfs_segctor_apply_buffers+0x161/0x330 fs/nilfs2/segment.c:1010
> > nilfs_segctor_scan_file+0x68e/0x8e0 fs/nilfs2/segment.c:1059
> > nilfs_segctor_collect_blocks fs/nilfs2/segment.c:1267 [inline]
> > nilfs_segctor_collect fs/nilfs2/segment.c:1547 [inline]
> > nilfs_segctor_do_construct+0x1eac/0x6970 fs/nilfs2/segment.c:2122
> > nilfs_segctor_construct+0x17b/0x690 fs/nilfs2/segment.c:2478
> > nilfs_segctor_thread_construct fs/nilfs2/segment.c:2585 [inline]
> > nilfs_segctor_thread+0x6cd/0xe00 fs/nilfs2/segment.c:2700
> > kthread+0x70e/0x8a0 kernel/kthread.c:464
> > ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:153
> > ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245
> > </TASK>
> >
> >
> > Tested on:
> >
> > commit: 9c69f888 Merge tag 'bcachefs-2025-05-08' of git://evil..
> > git tree: upstream
> > console output: https://syzkaller.appspot.com/x/log.txt?x=11a694d4580000
> > kernel config: https://syzkaller.appspot.com/x/.config?x=91c351a0f6229e67
> > dashboard link: https://syzkaller.appspot.com/bug?extid=81394db39b0e2ed2db06
> > compiler: Debian clang version 20.1.2 (++20250402124445+58df0ef89dd6-1~exp1~20250402004600.97), Debian LLD 20.1.2
> > patch: https://syzkaller.appspot.com/x/patch.diff?x=10c74768580000
> >
> >
>
> This reports a different circular lock dependency than all other
> reports in this issue, so it appears to be a separate issue. I'd like
> to dig deeper.
>
> Ryusuke Konishi
Oops, I misunderstood.
This was the result of testing the following patch:
https://syzkaller.appspot.com/x/patch.diff?x=10c74768580000
Just sharing the info, the original issue (extid=81394db39b0e2ed2db06)
should be resolved with the following patch sent upstream:
https://lore.kernel.org/all/20250503053327.12294-1-konishi.ryusuke@gmail.com/
Ryusuke Konishi
Powered by blists - more mailing lists