* [syzbot] [nilfs?] possible deadlock in nilfs_segctor_construct
@ 2025-05-08 16:22 syzbot
2025-05-08 16:37 ` Ryusuke Konishi
2025-05-09 13:29 ` Edward Adam Davis
0 siblings, 2 replies; 6+ messages in thread
From: syzbot @ 2025-05-08 16:22 UTC (permalink / raw)
To: konishi.ryusuke, linux-kernel, linux-nilfs, syzkaller-bugs
Hello,
syzbot found the following issue on:
HEAD commit: e8ab83e34bdc Merge tag 'arm64-fixes' of git://git.kernel.o..
git tree: upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=146e70f4580000
kernel config: https://syzkaller.appspot.com/x/.config?x=a9a25b7a36123454
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
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=148c9a70580000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=1530b8d4580000
Downloadable assets:
disk image (non-bootable): https://storage.googleapis.com/syzbot-assets/7feb34a89c2a/non_bootable_disk-e8ab83e3.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/b26f15c51ac7/vmlinux-e8ab83e3.xz
kernel image: https://storage.googleapis.com/syzbot-assets/05e91bf788d8/bzImage-e8ab83e3.xz
mounted in repro: https://storage.googleapis.com/syzbot-assets/09efd3b532ea/mount_0.gz
IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+81394db39b0e2ed2db06@syzkaller.appspotmail.com
======================================================
WARNING: possible circular locking dependency detected
6.15.0-rc4-syzkaller-00296-ge8ab83e34bdc #0 Not tainted
------------------------------------------------------
segctord/5299 is trying to acquire lock:
ffff88801189e090 (&nilfs->ns_sem){++++}-{4:4}, at: nilfs_segctor_construct+0x2b1/0x690 fs/nilfs2/segment.c:2485
but task is already holding lock:
ffff88801189e2a0 (&nilfs->ns_segctor_sem){++++}-{4:4}, at: nilfs_transaction_lock+0x253/0x4c0 fs/nilfs2/segment.c:357
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #5 (&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
-> #4 (sb_internal#2){.+.+}-{0:0}:
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_page_mkwrite+0x8b0/0xc20 fs/nilfs2/file.c:95
do_page_mkwrite+0x14a/0x310 mm/memory.c:3287
do_shared_fault mm/memory.c:5594 [inline]
do_fault mm/memory.c:5656 [inline]
do_pte_missing mm/memory.c:4160 [inline]
handle_pte_fault mm/memory.c:5997 [inline]
__handle_mm_fault+0x18d2/0x5380 mm/memory.c:6140
handle_mm_fault+0x3f6/0x8c0 mm/memory.c:6309
do_user_addr_fault+0x764/0x1390 arch/x86/mm/fault.c:1388
handle_page_fault arch/x86/mm/fault.c:1480 [inline]
exc_page_fault+0x68/0x110 arch/x86/mm/fault.c:1538
asm_exc_page_fault+0x26/0x30 arch/x86/include/asm/idtentry.h:623
-> #3 (sb_pagefaults){.+.+}-{0:0}:
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_pagefault include/linux/fs.h:1948 [inline]
nilfs_page_mkwrite+0x21e/0xc20 fs/nilfs2/file.c:57
do_page_mkwrite+0x14a/0x310 mm/memory.c:3287
do_shared_fault mm/memory.c:5594 [inline]
do_fault mm/memory.c:5656 [inline]
do_pte_missing mm/memory.c:4160 [inline]
handle_pte_fault mm/memory.c:5997 [inline]
__handle_mm_fault+0x18d2/0x5380 mm/memory.c:6140
handle_mm_fault+0x3f6/0x8c0 mm/memory.c:6309
do_user_addr_fault+0x764/0x1390 arch/x86/mm/fault.c:1388
handle_page_fault arch/x86/mm/fault.c:1480 [inline]
exc_page_fault+0x68/0x110 arch/x86/mm/fault.c:1538
asm_exc_page_fault+0x26/0x30 arch/x86/include/asm/idtentry.h:623
-> #2 (&mm->mmap_lock){++++}-{4:4}:
lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5866
__might_fault+0xcc/0x130 mm/memory.c:7151
_copy_to_iter+0xf3/0x15a0 lib/iov_iter.c:184
copy_page_to_iter+0xa7/0x150 lib/iov_iter.c:362
copy_folio_to_iter include/linux/uio.h:198 [inline]
filemap_read+0x78d/0x11d0 mm/filemap.c:2753
blkdev_read_iter+0x30a/0x440 block/fops.c:809
new_sync_read fs/read_write.c:489 [inline]
vfs_read+0x4cd/0x980 fs/read_write.c:570
ksys_read+0x145/0x250 fs/read_write.c:713
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 (&sb->s_type->i_mutex_key#8){++++}-{4:4}:
lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5866
down_write+0x96/0x1f0 kernel/locking/rwsem.c:1577
inode_lock include/linux/fs.h:867 [inline]
set_blocksize+0x23b/0x500 block/bdev.c:203
sb_set_blocksize block/bdev.c:224 [inline]
sb_min_blocksize+0x119/0x210 block/bdev.c:239
init_nilfs+0x43/0x690 fs/nilfs2/the_nilfs.c:710
nilfs_fill_super+0x8f/0x650 fs/nilfs2/super.c:1060
nilfs_get_tree+0x4f4/0x870 fs/nilfs2/super.c:1228
vfs_get_tree+0x8f/0x2b0 fs/super.c:1759
do_new_mount+0x24a/0xa40 fs/namespace.c:3884
do_mount fs/namespace.c:4224 [inline]
__do_sys_mount fs/namespace.c:4435 [inline]
__se_sys_mount+0x317/0x410 fs/namespace.c:4412
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 (&nilfs->ns_sem){++++}-{4:4}:
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
down_write+0x96/0x1f0 kernel/locking/rwsem.c:1577
nilfs_segctor_construct+0x2b1/0x690 fs/nilfs2/segment.c:2485
nilfs_segctor_thread_construct fs/nilfs2/segment.c:2586 [inline]
nilfs_segctor_thread+0x6f7/0xe00 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:
&nilfs->ns_sem --> sb_internal#2 --> &nilfs->ns_segctor_sem
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(&nilfs->ns_segctor_sem);
lock(sb_internal#2);
lock(&nilfs->ns_segctor_sem);
lock(&nilfs->ns_sem);
*** DEADLOCK ***
1 lock held by segctord/5299:
#0: ffff88801189e2a0 (&nilfs->ns_segctor_sem){++++}-{4:4}, at: nilfs_transaction_lock+0x253/0x4c0 fs/nilfs2/segment.c:357
stack backtrace:
CPU: 0 UID: 0 PID: 5299 Comm: segctord Not tainted 6.15.0-rc4-syzkaller-00296-ge8ab83e34bdc #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
down_write+0x96/0x1f0 kernel/locking/rwsem.c:1577
nilfs_segctor_construct+0x2b1/0x690 fs/nilfs2/segment.c:2485
nilfs_segctor_thread_construct fs/nilfs2/segment.c:2586 [inline]
nilfs_segctor_thread+0x6f7/0xe00 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>
---
This report is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzkaller@googlegroups.com.
syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-title
If you want syzbot to run the reproducer, reply with:
#syz test: git://repo/address.git branch-or-commit-hash
If you attach or paste a git patch, syzbot will apply it before testing.
If you want to overwrite report's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)
If the report is a duplicate of another one, reply with:
#syz dup: exact-subject-of-another-report
If you want to undo deduplication, reply with:
#syz undup
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [syzbot] [nilfs?] possible deadlock in nilfs_segctor_construct
2025-05-08 16:22 [syzbot] [nilfs?] possible deadlock in nilfs_segctor_construct syzbot
@ 2025-05-08 16:37 ` Ryusuke Konishi
2025-05-09 13:29 ` Edward Adam Davis
1 sibling, 0 replies; 6+ messages in thread
From: Ryusuke Konishi @ 2025-05-08 16:37 UTC (permalink / raw)
To: syzbot; +Cc: linux-kernel, linux-nilfs, syzkaller-bugs
On Fri, May 9, 2025 at 1:22 AM syzbot
<syzbot+81394db39b0e2ed2db06@syzkaller.appspotmail.com> wrote:
>
> Hello,
>
> syzbot found the following issue on:
>
> HEAD commit: e8ab83e34bdc Merge tag 'arm64-fixes' of git://git.kernel.o..
> git tree: upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=146e70f4580000
> kernel config: https://syzkaller.appspot.com/x/.config?x=a9a25b7a36123454
> 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
> syz repro: https://syzkaller.appspot.com/x/repro.syz?x=148c9a70580000
> C reproducer: https://syzkaller.appspot.com/x/repro.c?x=1530b8d4580000
>
> Downloadable assets:
> disk image (non-bootable): https://storage.googleapis.com/syzbot-assets/7feb34a89c2a/non_bootable_disk-e8ab83e3.raw.xz
> vmlinux: https://storage.googleapis.com/syzbot-assets/b26f15c51ac7/vmlinux-e8ab83e3.xz
> kernel image: https://storage.googleapis.com/syzbot-assets/05e91bf788d8/bzImage-e8ab83e3.xz
> mounted in repro: https://storage.googleapis.com/syzbot-assets/09efd3b532ea/mount_0.gz
>
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+81394db39b0e2ed2db06@syzkaller.appspotmail.com
>
> ======================================================
> WARNING: possible circular locking dependency detected
> 6.15.0-rc4-syzkaller-00296-ge8ab83e34bdc #0 Not tainted
> ------------------------------------------------------
> segctord/5299 is trying to acquire lock:
> ffff88801189e090 (&nilfs->ns_sem){++++}-{4:4}, at: nilfs_segctor_construct+0x2b1/0x690 fs/nilfs2/segment.c:2485
>
> but task is already holding lock:
> ffff88801189e2a0 (&nilfs->ns_segctor_sem){++++}-{4:4}, at: nilfs_transaction_lock+0x253/0x4c0 fs/nilfs2/segment.c:357
>
> which lock already depends on the new lock.
>
>
> the existing dependency chain (in reverse order) is:
>
> -> #5 (&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
>
> -> #4 (sb_internal#2){.+.+}-{0:0}:
> 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_page_mkwrite+0x8b0/0xc20 fs/nilfs2/file.c:95
> do_page_mkwrite+0x14a/0x310 mm/memory.c:3287
> do_shared_fault mm/memory.c:5594 [inline]
> do_fault mm/memory.c:5656 [inline]
> do_pte_missing mm/memory.c:4160 [inline]
> handle_pte_fault mm/memory.c:5997 [inline]
> __handle_mm_fault+0x18d2/0x5380 mm/memory.c:6140
> handle_mm_fault+0x3f6/0x8c0 mm/memory.c:6309
> do_user_addr_fault+0x764/0x1390 arch/x86/mm/fault.c:1388
> handle_page_fault arch/x86/mm/fault.c:1480 [inline]
> exc_page_fault+0x68/0x110 arch/x86/mm/fault.c:1538
> asm_exc_page_fault+0x26/0x30 arch/x86/include/asm/idtentry.h:623
>
> -> #3 (sb_pagefaults){.+.+}-{0:0}:
> 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_pagefault include/linux/fs.h:1948 [inline]
> nilfs_page_mkwrite+0x21e/0xc20 fs/nilfs2/file.c:57
> do_page_mkwrite+0x14a/0x310 mm/memory.c:3287
> do_shared_fault mm/memory.c:5594 [inline]
> do_fault mm/memory.c:5656 [inline]
> do_pte_missing mm/memory.c:4160 [inline]
> handle_pte_fault mm/memory.c:5997 [inline]
> __handle_mm_fault+0x18d2/0x5380 mm/memory.c:6140
> handle_mm_fault+0x3f6/0x8c0 mm/memory.c:6309
> do_user_addr_fault+0x764/0x1390 arch/x86/mm/fault.c:1388
> handle_page_fault arch/x86/mm/fault.c:1480 [inline]
> exc_page_fault+0x68/0x110 arch/x86/mm/fault.c:1538
> asm_exc_page_fault+0x26/0x30 arch/x86/include/asm/idtentry.h:623
>
> -> #2 (&mm->mmap_lock){++++}-{4:4}:
> lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5866
> __might_fault+0xcc/0x130 mm/memory.c:7151
> _copy_to_iter+0xf3/0x15a0 lib/iov_iter.c:184
> copy_page_to_iter+0xa7/0x150 lib/iov_iter.c:362
> copy_folio_to_iter include/linux/uio.h:198 [inline]
> filemap_read+0x78d/0x11d0 mm/filemap.c:2753
> blkdev_read_iter+0x30a/0x440 block/fops.c:809
> new_sync_read fs/read_write.c:489 [inline]
> vfs_read+0x4cd/0x980 fs/read_write.c:570
> ksys_read+0x145/0x250 fs/read_write.c:713
> 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 (&sb->s_type->i_mutex_key#8){++++}-{4:4}:
> lock_acquire+0x120/0x360 kernel/locking/lockdep.c:5866
> down_write+0x96/0x1f0 kernel/locking/rwsem.c:1577
> inode_lock include/linux/fs.h:867 [inline]
> set_blocksize+0x23b/0x500 block/bdev.c:203
> sb_set_blocksize block/bdev.c:224 [inline]
> sb_min_blocksize+0x119/0x210 block/bdev.c:239
> init_nilfs+0x43/0x690 fs/nilfs2/the_nilfs.c:710
> nilfs_fill_super+0x8f/0x650 fs/nilfs2/super.c:1060
> nilfs_get_tree+0x4f4/0x870 fs/nilfs2/super.c:1228
> vfs_get_tree+0x8f/0x2b0 fs/super.c:1759
> do_new_mount+0x24a/0xa40 fs/namespace.c:3884
> do_mount fs/namespace.c:4224 [inline]
> __do_sys_mount fs/namespace.c:4435 [inline]
> __se_sys_mount+0x317/0x410 fs/namespace.c:4412
> 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 (&nilfs->ns_sem){++++}-{4:4}:
> 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
> down_write+0x96/0x1f0 kernel/locking/rwsem.c:1577
> nilfs_segctor_construct+0x2b1/0x690 fs/nilfs2/segment.c:2485
> nilfs_segctor_thread_construct fs/nilfs2/segment.c:2586 [inline]
> nilfs_segctor_thread+0x6f7/0xe00 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:
> &nilfs->ns_sem --> sb_internal#2 --> &nilfs->ns_segctor_sem
>
> Possible unsafe locking scenario:
>
> CPU0 CPU1
> ---- ----
> lock(&nilfs->ns_segctor_sem);
> lock(sb_internal#2);
> lock(&nilfs->ns_segctor_sem);
> lock(&nilfs->ns_sem);
>
> *** DEADLOCK ***
>
> 1 lock held by segctord/5299:
> #0: ffff88801189e2a0 (&nilfs->ns_segctor_sem){++++}-{4:4}, at: nilfs_transaction_lock+0x253/0x4c0 fs/nilfs2/segment.c:357
>
> stack backtrace:
> CPU: 0 UID: 0 PID: 5299 Comm: segctord Not tainted 6.15.0-rc4-syzkaller-00296-ge8ab83e34bdc #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
> down_write+0x96/0x1f0 kernel/locking/rwsem.c:1577
> nilfs_segctor_construct+0x2b1/0x690 fs/nilfs2/segment.c:2485
> nilfs_segctor_thread_construct fs/nilfs2/segment.c:2586 [inline]
> nilfs_segctor_thread+0x6f7/0xe00 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>
>
>
> ---
> This report is generated by a bot. It may contain errors.
> See https://goo.gl/tpsmEJ for more information about syzbot.
> syzbot engineers can be reached at syzkaller@googlegroups.com.
>
> syzbot will keep track of this issue. See:
> https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
>
> If the report is already addressed, let syzbot know by replying with:
> #syz fix: exact-commit-title
>
> If you want syzbot to run the reproducer, reply with:
> #syz test: git://repo/address.git branch-or-commit-hash
> If you attach or paste a git patch, syzbot will apply it before testing.
>
> If you want to overwrite report's subsystems, reply with:
> #syz set subsystems: new-subsystem
> (See the list of subsystem names on the web dashboard)
>
> If the report is a duplicate of another one, reply with:
> #syz dup: exact-subject-of-another-report
>
> If you want to undo deduplication, reply with:
> #syz undup
This looks a false positive deadlock warning similar to the one in the
syzbot report
below, caused by inode locking introduced in set_blocksize():
#syz dup: possible deadlock in __nilfs_error (3)
A patch to fix this warning is already on its way to the mainline.
Ryusuke Konishi
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [syzbot] [nilfs?] possible deadlock in nilfs_segctor_construct
2025-05-08 16:22 [syzbot] [nilfs?] possible deadlock in nilfs_segctor_construct syzbot
2025-05-08 16:37 ` Ryusuke Konishi
@ 2025-05-09 13:29 ` Edward Adam Davis
2025-05-09 13:46 ` syzbot
1 sibling, 1 reply; 6+ messages in thread
From: Edward Adam Davis @ 2025-05-09 13:29 UTC (permalink / raw)
To: syzbot+81394db39b0e2ed2db06; +Cc: linux-kernel, syzkaller-bugs
#syz test
diff --git a/fs/nilfs2/segment.c b/fs/nilfs2/segment.c
index 83970d97840b..d3076015771f 100644
--- a/fs/nilfs2/segment.c
+++ b/fs/nilfs2/segment.c
@@ -2582,8 +2582,8 @@ static void nilfs_segctor_thread_construct(struct nilfs_sc_info *sci, int mode)
{
struct nilfs_transaction_info ti;
- nilfs_transaction_lock(sci->sc_super, &ti, 0);
nilfs_segctor_construct(sci, mode);
+ nilfs_transaction_lock(sci->sc_super, &ti, 0);
/*
* Unclosed segment should be retried. We do this using sc_timer.
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [syzbot] [nilfs?] possible deadlock in nilfs_segctor_construct
2025-05-09 13:29 ` Edward Adam Davis
@ 2025-05-09 13:46 ` syzbot
2025-05-10 13:31 ` Ryusuke Konishi
0 siblings, 1 reply; 6+ messages in thread
From: syzbot @ 2025-05-09 13:46 UTC (permalink / raw)
To: eadavis, linux-kernel, syzkaller-bugs
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
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [syzbot] [nilfs?] possible deadlock in nilfs_segctor_construct
2025-05-09 13:46 ` syzbot
@ 2025-05-10 13:31 ` Ryusuke Konishi
2025-05-10 13:49 ` Ryusuke Konishi
0 siblings, 1 reply; 6+ messages in thread
From: Ryusuke Konishi @ 2025-05-10 13:31 UTC (permalink / raw)
To: syzbot; +Cc: syzkaller-bugs, linux-kernel, eadavis
On Sat, May 10, 2025 at 9:46 PM syzbot
<syzbot+81394db39b0e2ed2db06@syzkaller.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
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [syzbot] [nilfs?] possible deadlock in nilfs_segctor_construct
2025-05-10 13:31 ` Ryusuke Konishi
@ 2025-05-10 13:49 ` Ryusuke Konishi
0 siblings, 0 replies; 6+ messages in thread
From: Ryusuke Konishi @ 2025-05-10 13:49 UTC (permalink / raw)
To: syzbot; +Cc: eadavis, syzkaller-bugs, linux-kernel
On Sat, May 10, 2025 at 10:31 PM Ryusuke Konishi wrote:
>
> On Sat, May 10, 2025 at 9:46 PM syzbot
> <syzbot+81394db39b0e2ed2db06@syzkaller.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
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2025-05-10 13:49 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-05-08 16:22 [syzbot] [nilfs?] possible deadlock in nilfs_segctor_construct syzbot
2025-05-08 16:37 ` Ryusuke Konishi
2025-05-09 13:29 ` Edward Adam Davis
2025-05-09 13:46 ` syzbot
2025-05-10 13:31 ` Ryusuke Konishi
2025-05-10 13:49 ` Ryusuke Konishi
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.