Re: [syzbot] [hfs?] possible deadlock in hfsplus_block_free

From: syzbot
Date: Fri Jan 19 2024 - 23:24:35 EST

Next message: Lars-Peter Clausen: "Re: [PATCH v6 3/4] iio: humidity: Add driver for ti HDC302x humidity sensors"
Previous message: Chenyuan Yang: "[Linux Kernel Bug] kernel BUG in btrfs_get_root_ref (assertion failed: !anon_dev) and general protection fault in btrfs_update_reloc_root"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

syzbot has found a reproducer for the following issue on:

HEAD commit: 9d1694dc91ce Merge tag 'for-6.8/block-2024-01-18' of git:/..
git tree: upstream
console+strace: https://syzkaller.appspot.com/x/log.txt?x=13ac11abe80000
kernel config: https://syzkaller.appspot.com/x/.config?x=5142c864fad2684b
dashboard link: https://syzkaller.appspot.com/bug?extid=8fae81a1f77bf28ef3b5
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=1519994fe80000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=138ba46fe80000

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/4c0e4097a70a/disk-9d1694dc.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/9b551b9fa3e4/vmlinux-9d1694dc.xz
kernel image: https://storage.googleapis.com/syzbot-assets/6ecfef59fe31/bzImage-9d1694dc.xz
mounted in repro #1: https://storage.googleapis.com/syzbot-assets/2f9f30811e75/mount_0.gz
mounted in repro #2: https://storage.googleapis.com/syzbot-assets/e5b402bef2f2/mount_2.gz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+8fae81a1f77bf28ef3b5@xxxxxxxxxxxxxxxxxxxxxxxxx

======================================================
WARNING: possible circular locking dependency detected
6.7.0-syzkaller-12377-g9d1694dc91ce #0 Not tainted
------------------------------------------------------
syz-executor275/5161 is trying to acquire lock:
ffff888029d7a8f8 (&sbi->alloc_mutex){+.+.}-{3:3}, at: hfsplus_block_free+0xbb/0x4d0 fs/hfsplus/bitmap.c:182

but task is already holding lock:
ffff888079ada988 (&HFSPLUS_I(inode)->extents_lock){+.+.}-{3:3}, at: hfsplus_file_truncate+0x2da/0xb40 fs/hfsplus/extents.c:576

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #1 (&HFSPLUS_I(inode)->extents_lock){+.+.}-{3:3}:
lock_acquire+0x1e3/0x530 kernel/locking/lockdep.c:5754
__mutex_lock_common kernel/locking/mutex.c:608 [inline]
__mutex_lock+0x136/0xd60 kernel/locking/mutex.c:752
hfsplus_get_block+0x383/0x14e0 fs/hfsplus/extents.c:260
block_read_full_folio+0x422/0xe10 fs/buffer.c:2382
filemap_read_folio+0x19c/0x780 mm/filemap.c:2324
do_read_cache_folio+0x134/0x810 mm/filemap.c:3701
do_read_cache_page+0x30/0x200 mm/filemap.c:3767
read_mapping_page include/linux/pagemap.h:888 [inline]
hfsplus_block_allocate+0xee/0x8b0 fs/hfsplus/bitmap.c:37
hfsplus_file_extend+0xade/0x1b70 fs/hfsplus/extents.c:468
hfsplus_get_block+0x406/0x14e0 fs/hfsplus/extents.c:245
__block_write_begin_int+0x50b/0x1a70 fs/buffer.c:2103
__block_write_begin fs/buffer.c:2152 [inline]
block_write_begin+0x9b/0x1e0 fs/buffer.c:2211
cont_write_begin+0x643/0x880 fs/buffer.c:2565
hfsplus_write_begin+0x8a/0xd0 fs/hfsplus/inode.c:47
page_symlink+0x2c5/0x4e0 fs/namei.c:5228
hfsplus_symlink+0xca/0x260 fs/hfsplus/dir.c:449
vfs_symlink+0x12f/0x2a0 fs/namei.c:4480
do_symlinkat+0x222/0x3a0 fs/namei.c:4506
__do_sys_symlink fs/namei.c:4527 [inline]
__se_sys_symlink fs/namei.c:4525 [inline]
__x64_sys_symlink+0x7e/0x90 fs/namei.c:4525
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0xf5/0x230 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x63/0x6b

-> #0 (&sbi->alloc_mutex){+.+.}-{3:3}:
check_prev_add kernel/locking/lockdep.c:3134 [inline]
check_prevs_add kernel/locking/lockdep.c:3253 [inline]
validate_chain+0x1909/0x5ab0 kernel/locking/lockdep.c:3869
__lock_acquire+0x1345/0x1fd0 kernel/locking/lockdep.c:5137
lock_acquire+0x1e3/0x530 kernel/locking/lockdep.c:5754
__mutex_lock_common kernel/locking/mutex.c:608 [inline]
__mutex_lock+0x136/0xd60 kernel/locking/mutex.c:752
hfsplus_block_free+0xbb/0x4d0 fs/hfsplus/bitmap.c:182
hfsplus_free_extents+0x17a/0xae0 fs/hfsplus/extents.c:363
hfsplus_file_truncate+0x7d0/0xb40 fs/hfsplus/extents.c:591
hfsplus_delete_inode+0x174/0x220
hfsplus_unlink+0x512/0x790 fs/hfsplus/dir.c:405
hfsplus_rename+0xc8/0x1c0 fs/hfsplus/dir.c:547
vfs_rename+0xbd3/0xef0 fs/namei.c:4879
do_renameat2+0xd94/0x13f0 fs/namei.c:5036
__do_sys_renameat2 fs/namei.c:5070 [inline]
__se_sys_renameat2 fs/namei.c:5067 [inline]
__x64_sys_renameat2+0xd2/0xe0 fs/namei.c:5067
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0xf5/0x230 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x63/0x6b

other info that might help us debug this:

Possible unsafe locking scenario:

CPU0 CPU1
---- ----
lock(&HFSPLUS_I(inode)->extents_lock);
lock(&sbi->alloc_mutex);
lock(&HFSPLUS_I(inode)->extents_lock);
lock(&sbi->alloc_mutex);

*** DEADLOCK ***

6 locks held by syz-executor275/5161:
#0: ffff888029204420 (sb_writers#9){.+.+}-{0:0}, at: mnt_want_write+0x3f/0x90 fs/namespace.c:409
#1: ffff888079ad9740 (&type->i_mutex_dir_key#6/1){+.+.}-{3:3}, at: inode_lock_nested include/linux/fs.h:837 [inline]
#1: ffff888079ad9740 (&type->i_mutex_dir_key#6/1){+.+.}-{3:3}, at: lock_rename fs/namei.c:3065 [inline]
#1: ffff888079ad9740 (&type->i_mutex_dir_key#6/1){+.+.}-{3:3}, at: do_renameat2+0x62c/0x13f0 fs/namei.c:4971
#2: ffff888079ada4c0 (&sb->s_type->i_mutex_key#14){+.+.}-{3:3}, at: inode_lock include/linux/fs.h:802 [inline]
#2: ffff888079ada4c0 (&sb->s_type->i_mutex_key#14){+.+.}-{3:3}, at: lock_two_nondirectories+0xe1/0x170 fs/inode.c:1109
#3: ffff888079adab80 (&sb->s_type->i_mutex_key#14/4){+.+.}-{3:3}, at: vfs_rename+0x69e/0xef0 fs/namei.c:4850
#4: ffff888029d7a998 (&sbi->vh_mutex){+.+.}-{3:3}, at: hfsplus_unlink+0x161/0x790 fs/hfsplus/dir.c:370
#5: ffff888079ada988 (&HFSPLUS_I(inode)->extents_lock){+.+.}-{3:3}, at: hfsplus_file_truncate+0x2da/0xb40 fs/hfsplus/extents.c:576

stack backtrace:
CPU: 1 PID: 5161 Comm: syz-executor275 Not tainted 6.7.0-syzkaller-12377-g9d1694dc91ce #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 11/17/2023
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0x1e7/0x2d0 lib/dump_stack.c:106
check_noncircular+0x366/0x490 kernel/locking/lockdep.c:2187
check_prev_add kernel/locking/lockdep.c:3134 [inline]
check_prevs_add kernel/locking/lockdep.c:3253 [inline]
validate_chain+0x1909/0x5ab0 kernel/locking/lockdep.c:3869
__lock_acquire+0x1345/0x1fd0 kernel/locking/lockdep.c:5137
lock_acquire+0x1e3/0x530 kernel/locking/lockdep.c:5754
__mutex_lock_common kernel/locking/mutex.c:608 [inline]
__mutex_lock+0x136/0xd60 kernel/locking/mutex.c:752
hfsplus_block_free+0xbb/0x4d0 fs/hfsplus/bitmap.c:182
hfsplus_free_extents+0x17a/0xae0 fs/hfsplus/extents.c:363
hfsplus_file_truncate+0x7d0/0xb40 fs/hfsplus/extents.c:591
hfsplus_delete_inode+0x174/0x220
hfsplus_unlink+0x512/0x790 fs/hfsplus/dir.c:405
hfsplus_rename+0xc8/0x1c0 fs/hfsplus/dir.c:547
vfs_rename+0xbd3/0xef0 fs/namei.c:4879
do_renameat2+0xd94/0x13f0 fs/namei.c:5036
__do_sys_renameat2 fs/namei.c:5070 [inline]
__se_sys_renameat2 fs/namei.c:5067 [inline]
__x64_sys_renameat2+0xd2/0xe0 fs/namei.c:5067
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0xf5/0x230 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x63/0x6b
RIP: 0033:0x7f6db63c6e89
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 b1 18 00 00 90 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 b0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f6db6383218 EFLAGS: 00000246 ORIG_RAX: 000000000000013c
RAX: ffffffffffffffda RBX: 00007f6db644f6c8 RCX: 00007f6db63c6e89
RDX: 0000000000000004 RSI: 00000000200000c0 RDI: 0000000000000005
RBP: 00007f6db644f6c0 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000020000180 R11: 0000000000000246 R12: 00007f6db641c168
R13: 0032656c69662f2e R14: 0073756c70736668 R15: 0031656c69662f2e
</TASK>

---
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.

Next message: Lars-Peter Clausen: "Re: [PATCH v6 3/4] iio: humidity: Add driver for ti HDC302x humidity sensors"
Previous message: Chenyuan Yang: "[Linux Kernel Bug] kernel BUG in btrfs_get_root_ref (assertion failed: !anon_dev) and general protection fault in btrfs_update_reloc_root"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]