[syzbot] [hfs?] possible deadlock in hfsplus_file_extend (2)

[syzbot] [hfs?] possible deadlock in hfsplus_file_extend (2)

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit:    ff7afaeca1a1 Merge tag 'nfs-for-6.12-3' of git://git.linux..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=13d43f40580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=11254d3590b16717
dashboard link: https://syzkaller.appspot.com/bug?extid=4cba2fd444e9a16ae758
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=135dae30580000
C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=15142d87980000

Downloadable assets:
disk image (non-bootable): https://storage.googleapis.com/syzbot-assets/7feb34a89c2a/non_bootable_disk-ff7afaec.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/d8c6d968e5c0/vmlinux-ff7afaec.xz
kernel image: https://storage.googleapis.com/syzbot-assets/dcee7f1927a8/bzImage-ff7afaec.xz
mounted in repro: https://storage.googleapis.com/syzbot-assets/55a0799266ca/mount_0.gz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+4cba2fd444e9a16ae758@syzkaller.appspotmail.com

loop0: detected capacity change from 0 to 1024
hfsplus: inconsistency in B*Tree (128,1,255,1,0)
============================================
WARNING: possible recursive locking detected
6.12.0-rc6-syzkaller-00110-gff7afaeca1a1 #0 Not tainted
--------------------------------------------
syz-executor395/5309 is trying to acquire lock:
ffff8880422d0108 (&HFSPLUS_I(inode)->extents_lock){+.+.}-{3:3}, at: hfsplus_file_extend+0x21b/0x1b70 fs/hfsplus/extents.c:458

but task is already holding lock:
ffff8880422d0e88 (&HFSPLUS_I(inode)->extents_lock){+.+.}-{3:3}, at: hfsplus_file_extend+0x21b/0x1b70 fs/hfsplus/extents.c:458

other info that might help us debug this:
 Possible unsafe locking scenario:

       CPU0
       ----
  lock(&HFSPLUS_I(inode)->extents_lock);
  lock(&HFSPLUS_I(inode)->extents_lock);

 *** DEADLOCK ***

 May be due to missing lock nesting notation

6 locks held by syz-executor395/5309:
 #0: ffff88804026a420 (sb_writers#9){.+.+}-{0:0}, at: mnt_want_write+0x3f/0x90 fs/namespace.c:515
 #1: ffff8880422d24b8 (&sb->s_type->i_mutex_key#15){+.+.}-{3:3}, at: inode_lock include/linux/fs.h:815 [inline]
 #1: ffff8880422d24b8 (&sb->s_type->i_mutex_key#15){+.+.}-{3:3}, at: vfs_setxattr+0x1e1/0x430 fs/xattr.c:320
 #2: ffff8880118740b0 (&tree->tree_lock){+.+.}-{3:3}, at: hfsplus_find_init+0x14a/0x1c0 fs/hfsplus/bfind.c:28
 #3: ffff88801ebc80b0 (&tree->tree_lock/2){+.+.}-{3:3}, at: hfsplus_find_init+0x14a/0x1c0 fs/hfsplus/bfind.c:28
 #4: ffff8880422d0e88 (&HFSPLUS_I(inode)->extents_lock){+.+.}-{3:3}, at: hfsplus_file_extend+0x21b/0x1b70 fs/hfsplus/extents.c:458
 #5: ffff8880118720b0 (&tree->tree_lock/1){+.+.}-{3:3}, at: hfsplus_find_init+0x14a/0x1c0 fs/hfsplus/bfind.c:28

stack backtrace:
CPU: 0 UID: 0 PID: 5309 Comm: syz-executor395 Not tainted 6.12.0-rc6-syzkaller-00110-gff7afaeca1a1 #0
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 lib/dump_stack.c:94 [inline]
 dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120
 print_deadlock_bug+0x483/0x620 kernel/locking/lockdep.c:3037
 check_deadlock kernel/locking/lockdep.c:3089 [inline]
 validate_chain+0x15e2/0x5920 kernel/locking/lockdep.c:3891
 __lock_acquire+0x1384/0x2050 kernel/locking/lockdep.c:5202
 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5825
 __mutex_lock_common kernel/locking/mutex.c:608 [inline]
 __mutex_lock+0x136/0xd70 kernel/locking/mutex.c:752
 hfsplus_file_extend+0x21b/0x1b70 fs/hfsplus/extents.c:458
 hfsplus_bmap_reserve+0x105/0x4e0 fs/hfsplus/btree.c:358
 __hfsplus_ext_write_extent+0x2a4/0x5c0 fs/hfsplus/extents.c:104
 __hfsplus_ext_cache_extent+0x84/0xe10 fs/hfsplus/extents.c:186
 hfsplus_ext_read_extent fs/hfsplus/extents.c:218 [inline]
 hfsplus_file_extend+0x48c/0x1b70 fs/hfsplus/extents.c:462
 hfsplus_bmap_reserve+0x105/0x4e0 fs/hfsplus/btree.c:358
 hfsplus_create_attr+0x1c8/0x640 fs/hfsplus/attributes.c:221
 __hfsplus_setxattr+0x6fe/0x22d0 fs/hfsplus/xattr.c:354
 hfsplus_setxattr+0xb0/0xe0 fs/hfsplus/xattr.c:432
 hfsplus_trusted_setxattr+0x40/0x60 fs/hfsplus/xattr_trusted.c:30
 __vfs_setxattr+0x468/0x4a0 fs/xattr.c:200
 __vfs_setxattr_noperm+0x12e/0x660 fs/xattr.c:234
 vfs_setxattr+0x221/0x430 fs/xattr.c:321
 do_setxattr fs/xattr.c:629 [inline]
 path_setxattr+0x37e/0x4d0 fs/xattr.c:658
 __do_sys_lsetxattr fs/xattr.c:683 [inline]
 __se_sys_lsetxattr fs/xattr.c:679 [inline]
 __x64_sys_lsetxattr+0xb8/0xd0 fs/xattr.c:679
 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
RIP: 0033:0x7f704e2aca99
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 f1 17 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 b8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffd231fbc58 EFLAGS: 00000246 ORIG_RAX: 00000000000000bd
RAX: ffffffffffffffda RBX: 0031656c69662f2e RCX: 00007f704e2aca99
RDX: 0000000000000000 RSI: 0000000020000200 RDI: 0000000020000040
RBP: 00007f704e3205f0 R08: 0000000000000000 R09: 00005555908174c0
R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffd231fbc80
R13: 00007ffd231fbea8 R14: 431bde82d7b634db R15: 00007f704e2f503b
 </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

Re: [syzbot] [hfs?] possible deadlock in hfsplus_file_extend (2)

[Edward Adam Davis]

AA lock, move extents_lock out of the hfsplus_file_extend.

#syz test

diff --git a/fs/hfsplus/btree.c b/fs/hfsplus/btree.c
index 9e1732a2b92a..ec6e9bdfe7b4 100644
--- a/fs/hfsplus/btree.c
+++ b/fs/hfsplus/btree.c
@@ -355,7 +355,9 @@ int hfs_bmap_reserve(struct hfs_btree *tree, int rsvd_nodes)
 		return 0;
 
 	while (tree->free_nodes < rsvd_nodes) {
+		mutex_lock(&hip->extents_lock);
 		res = hfsplus_file_extend(inode, hfs_bnode_need_zeroout(tree));
+		mutex_unlock(&hip->extents_lock);
 		if (res)
 			return res;
 		hip->phys_size = inode->i_size =
diff --git a/fs/hfsplus/extents.c b/fs/hfsplus/extents.c
index a6d61685ae79..966252d129bc 100644
--- a/fs/hfsplus/extents.c
+++ b/fs/hfsplus/extents.c
@@ -242,7 +242,9 @@ int hfsplus_get_block(struct inode *inode, sector_t iblock,
 		if (iblock > hip->fs_blocks)
 			return -EIO;
 		if (ablock >= hip->alloc_blocks) {
+			mutex_lock(&hip->extents_lock);
 			res = hfsplus_file_extend(inode, false);
+			mutex_unlock(&hip->extents_lock);
 			if (res)
 				return res;
 		}
@@ -455,7 +457,6 @@ int hfsplus_file_extend(struct inode *inode, bool zeroout)
 		return -ENOSPC;
 	}
 
-	mutex_lock(&hip->extents_lock);
 	if (hip->alloc_blocks == hip->first_blocks)
 		goal = hfsplus_ext_lastblock(hip->first_extents);
 	else {
@@ -516,11 +517,9 @@ int hfsplus_file_extend(struct inode *inode, bool zeroout)
 out:
 	if (!res) {
 		hip->alloc_blocks += len;
-		mutex_unlock(&hip->extents_lock);
 		hfsplus_mark_inode_dirty(inode, HFSPLUS_I_ALLOC_DIRTY);
 		return 0;
 	}
-	mutex_unlock(&hip->extents_lock);
 	return res;
 
 insert_extent:
diff --git a/fs/hfsplus/xattr.c b/fs/hfsplus/xattr.c
index 9a1a93e3888b..a589130e42ea 100644
--- a/fs/hfsplus/xattr.c
+++ b/fs/hfsplus/xattr.c
@@ -191,7 +191,9 @@ static int hfsplus_create_attributes_file(struct super_block *sb)
 	}
 
 	while (hip->alloc_blocks < hip->clump_blocks) {
+		mutex_unlock(&hip->extents_lock);
 		err = hfsplus_file_extend(attr_file, false);
+		mutex_unlock(&hip->extents_lock);
 		if (unlikely(err)) {
 			pr_err("failed to extend attributes file\n");
 			goto end_attr_file_creation;

Re: [syzbot] [hfs?] possible deadlock in hfsplus_file_extend (2)

[syzbot]

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:
possible deadlock in hfsplus_bmap_reserve

loop0: detected capacity change from 0 to 1024
hfsplus: inconsistency in B*Tree (128,1,255,1,0)
============================================
WARNING: possible recursive locking detected
6.12.0-rc7-syzkaller-g2d5404caa8c7-dirty #0 Not tainted
--------------------------------------------
syz.0.15/5813 is trying to acquire lock:
ffff888036820108 (&HFSPLUS_I(inode)->extents_lock){+.+.}-{3:3}, at: hfsplus_bmap_reserve+0x119/0x500 fs/hfsplus/btree.c:358

but task is already holding lock:
ffff888036820e88 (&HFSPLUS_I(inode)->extents_lock){+.+.}-{3:3}, at: hfsplus_bmap_reserve+0x119/0x500 fs/hfsplus/btree.c:358

other info that might help us debug this:
 Possible unsafe locking scenario:

       CPU0
       ----
  lock(&HFSPLUS_I(inode)->extents_lock);
  lock(&HFSPLUS_I(inode)->extents_lock);

 *** DEADLOCK ***

 May be due to missing lock nesting notation

6 locks held by syz.0.15/5813:
 #0: ffff888056c0c420 (sb_writers#13){.+.+}-{0:0}, at: mnt_want_write+0x3f/0x90 fs/namespace.c:515
 #1: ffff8880368224b8 (&sb->s_type->i_mutex_key#21){+.+.}-{3:3}, at: inode_lock include/linux/fs.h:815 [inline]
 #1: ffff8880368224b8 (&sb->s_type->i_mutex_key#21){+.+.}-{3:3}, at: vfs_setxattr+0x1e1/0x430 fs/xattr.c:320
 #2: ffff888034b200b0 (&tree->tree_lock){+.+.}-{3:3}, at: hfsplus_find_init+0x14a/0x1c0 fs/hfsplus/bfind.c:28
 #3: ffff8880574a20b0 (&tree->tree_lock/2){+.+.}-{3:3}, at: hfsplus_find_init+0x14a/0x1c0 fs/hfsplus/bfind.c:28
 #4: ffff888036820e88 (&HFSPLUS_I(inode)->extents_lock){+.+.}-{3:3}, at: hfsplus_bmap_reserve+0x119/0x500 fs/hfsplus/btree.c:358
 #5: ffff888038c7e0b0 (&tree->tree_lock/1){+.+.}-{3:3}, at: hfsplus_find_init+0x14a/0x1c0 fs/hfsplus/bfind.c:28

stack backtrace:
CPU: 0 UID: 0 PID: 5813 Comm: syz.0.15 Not tainted 6.12.0-rc7-syzkaller-g2d5404caa8c7-dirty #0
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 lib/dump_stack.c:94 [inline]
 dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120
 print_deadlock_bug+0x483/0x620 kernel/locking/lockdep.c:3037
 check_deadlock kernel/locking/lockdep.c:3089 [inline]
 validate_chain+0x15e2/0x5920 kernel/locking/lockdep.c:3891
 __lock_acquire+0x1384/0x2050 kernel/locking/lockdep.c:5202
 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5825
 __mutex_lock_common kernel/locking/mutex.c:608 [inline]
 __mutex_lock+0x136/0xd70 kernel/locking/mutex.c:752
 hfsplus_bmap_reserve+0x119/0x500 fs/hfsplus/btree.c:358
 __hfsplus_ext_write_extent+0x2a4/0x5c0 fs/hfsplus/extents.c:104
 __hfsplus_ext_cache_extent+0x84/0xe10 fs/hfsplus/extents.c:186
 hfsplus_ext_read_extent fs/hfsplus/extents.c:218 [inline]
 hfsplus_file_extend+0x43f/0x1ac0 fs/hfsplus/extents.c:463
 hfsplus_bmap_reserve+0x12c/0x500 fs/hfsplus/btree.c:359
 hfsplus_create_attr+0x1c8/0x640 fs/hfsplus/attributes.c:221
 __hfsplus_setxattr+0x702/0x22d0 fs/hfsplus/xattr.c:356
 hfsplus_setxattr+0xb0/0xe0 fs/hfsplus/xattr.c:434
 hfsplus_trusted_setxattr+0x40/0x60 fs/hfsplus/xattr_trusted.c:30
 __vfs_setxattr+0x468/0x4a0 fs/xattr.c:200
 __vfs_setxattr_noperm+0x12e/0x660 fs/xattr.c:234
 vfs_setxattr+0x221/0x430 fs/xattr.c:321
 do_setxattr fs/xattr.c:629 [inline]
 path_setxattr+0x37e/0x4d0 fs/xattr.c:658
 __do_sys_lsetxattr fs/xattr.c:683 [inline]
 __se_sys_lsetxattr fs/xattr.c:679 [inline]
 __x64_sys_lsetxattr+0xb8/0xd0 fs/xattr.c:679
 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
RIP: 0033:0x7f2ade37e719
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 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 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f2adf11d038 EFLAGS: 00000246 ORIG_RAX: 00000000000000bd
RAX: ffffffffffffffda RBX: 00007f2ade535f80 RCX: 00007f2ade37e719
RDX: 0000000000000000 RSI: 0000000020000200 RDI: 0000000020000040
RBP: 00007f2ade3f139e R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 0000000000000000 R14: 00007f2ade535f80 R15: 00007ffd3661a948
 </TASK>


Tested on:

commit:         2d5404ca Linux 6.12-rc7
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=127734e8580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=d2aeec8c0b2e420c
dashboard link: https://syzkaller.appspot.com/bug?extid=4cba2fd444e9a16ae758
compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
patch:          https://syzkaller.appspot.com/x/patch.diff?x=10d9d8c0580000

Re: [syzbot] [hfs?] possible deadlock in hfsplus_file_extend (2)

[Edward Adam Davis]

AA lock, unlock for hfs_bmap_reserve.

#syz test

diff --git a/fs/hfsplus/extents.c b/fs/hfsplus/extents.c
index a6d61685ae79..1742e43ece58 100644
--- a/fs/hfsplus/extents.c
+++ b/fs/hfsplus/extents.c
@@ -101,7 +101,9 @@ static int __hfsplus_ext_write_extent(struct inode *inode,
 		if (res != -ENOENT)
 			return res;
 		/* Fail early and avoid ENOSPC during the btree operation */
+		mutex_unlock(&hip->extents_lock);
 		res = hfs_bmap_reserve(fd->tree, fd->tree->depth + 1);
+		mutex_lock(&hip->extents_lock);
 		if (res)
 			return res;
 		hfs_brec_insert(fd, hip->cached_extents,

Re: [syzbot] [hfs?] possible deadlock in hfsplus_file_extend (2)

[syzbot]

Hello,

syzbot has tested the proposed patch but the reproducer is still triggering an issue:
possible deadlock in hfsplus_file_extend

loop0: detected capacity change from 0 to 1024
hfsplus: inconsistency in B*Tree (128,1,255,1,0)
======================================================
WARNING: possible circular locking dependency detected
6.12.0-rc7-syzkaller-g2d5404caa8c7-dirty #0 Not tainted
------------------------------------------------------
syz.0.15/5813 is trying to acquire lock:
ffff8880569a0108 (&HFSPLUS_I(inode)->extents_lock){+.+.}-{3:3}, at: hfsplus_file_extend+0x21b/0x1b70 fs/hfsplus/extents.c:460

but task is already holding lock:
ffff88803625c0b0 (&tree->tree_lock/1){+.+.}-{3:3}, at: hfsplus_find_init+0x14a/0x1c0 fs/hfsplus/bfind.c:28

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #1 (&tree->tree_lock/1){+.+.}-{3:3}:
       lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5825
       __mutex_lock_common kernel/locking/mutex.c:608 [inline]
       __mutex_lock+0x136/0xd70 kernel/locking/mutex.c:752
       hfsplus_find_init+0x14a/0x1c0 fs/hfsplus/bfind.c:28
       hfsplus_ext_read_extent fs/hfsplus/extents.c:218 [inline]
       hfsplus_file_extend+0x454/0x1b70 fs/hfsplus/extents.c:464
       hfsplus_bmap_reserve+0x105/0x4e0 fs/hfsplus/btree.c:358
       hfsplus_create_attr+0x1c8/0x640 fs/hfsplus/attributes.c:221
       __hfsplus_setxattr+0x6fe/0x22d0 fs/hfsplus/xattr.c:354
       hfsplus_setxattr+0xb0/0xe0 fs/hfsplus/xattr.c:432
       hfsplus_trusted_setxattr+0x40/0x60 fs/hfsplus/xattr_trusted.c:30
       __vfs_setxattr+0x468/0x4a0 fs/xattr.c:200
       __vfs_setxattr_noperm+0x12e/0x660 fs/xattr.c:234
       vfs_setxattr+0x221/0x430 fs/xattr.c:321
       do_setxattr fs/xattr.c:629 [inline]
       path_setxattr+0x37e/0x4d0 fs/xattr.c:658
       __do_sys_lsetxattr fs/xattr.c:683 [inline]
       __se_sys_lsetxattr fs/xattr.c:679 [inline]
       __x64_sys_lsetxattr+0xb8/0xd0 fs/xattr.c:679
       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

-> #0 (&HFSPLUS_I(inode)->extents_lock){+.+.}-{3:3}:
       check_prev_add kernel/locking/lockdep.c:3161 [inline]
       check_prevs_add kernel/locking/lockdep.c:3280 [inline]
       validate_chain+0x18ef/0x5920 kernel/locking/lockdep.c:3904
       __lock_acquire+0x1384/0x2050 kernel/locking/lockdep.c:5202
       lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5825
       __mutex_lock_common kernel/locking/mutex.c:608 [inline]
       __mutex_lock+0x136/0xd70 kernel/locking/mutex.c:752
       hfsplus_file_extend+0x21b/0x1b70 fs/hfsplus/extents.c:460
       hfsplus_bmap_reserve+0x105/0x4e0 fs/hfsplus/btree.c:358
       __hfsplus_ext_write_extent+0x2af/0x5d0 fs/hfsplus/extents.c:105
       __hfsplus_ext_cache_extent+0x84/0xe10 fs/hfsplus/extents.c:188
       hfsplus_ext_read_extent fs/hfsplus/extents.c:220 [inline]
       hfsplus_file_extend+0x48c/0x1b70 fs/hfsplus/extents.c:464
       hfsplus_bmap_reserve+0x105/0x4e0 fs/hfsplus/btree.c:358
       hfsplus_create_attr+0x1c8/0x640 fs/hfsplus/attributes.c:221
       __hfsplus_setxattr+0x6fe/0x22d0 fs/hfsplus/xattr.c:354
       hfsplus_setxattr+0xb0/0xe0 fs/hfsplus/xattr.c:432
       hfsplus_trusted_setxattr+0x40/0x60 fs/hfsplus/xattr_trusted.c:30
       __vfs_setxattr+0x468/0x4a0 fs/xattr.c:200
       __vfs_setxattr_noperm+0x12e/0x660 fs/xattr.c:234
       vfs_setxattr+0x221/0x430 fs/xattr.c:321
       do_setxattr fs/xattr.c:629 [inline]
       path_setxattr+0x37e/0x4d0 fs/xattr.c:658
       __do_sys_lsetxattr fs/xattr.c:683 [inline]
       __se_sys_lsetxattr fs/xattr.c:679 [inline]
       __x64_sys_lsetxattr+0xb8/0xd0 fs/xattr.c:679
       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

other info that might help us debug this:

 Possible unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(&tree->tree_lock/1);
                               lock(&HFSPLUS_I(inode)->extents_lock);
                               lock(&tree->tree_lock/1);
  lock(&HFSPLUS_I(inode)->extents_lock);

 *** DEADLOCK ***

5 locks held by syz.0.15/5813:
 #0: ffff888011c4c420 (sb_writers#13){.+.+}-{0:0}, at: mnt_want_write+0x3f/0x90 fs/namespace.c:515
 #1: ffff8880569a24b8 (&sb->s_type->i_mutex_key#21){+.+.}-{3:3}, at: inode_lock include/linux/fs.h:815 [inline]
 #1: ffff8880569a24b8 (&sb->s_type->i_mutex_key#21){+.+.}-{3:3}, at: vfs_setxattr+0x1e1/0x430 fs/xattr.c:320
 #2: ffff888040a3a0b0 (&tree->tree_lock){+.+.}-{3:3}, at: hfsplus_find_init+0x14a/0x1c0 fs/hfsplus/bfind.c:28
 #3: ffff888040a3e0b0 (&tree->tree_lock/2){+.+.}-{3:3}, at: hfsplus_find_init+0x14a/0x1c0 fs/hfsplus/bfind.c:28
 #4: ffff88803625c0b0 (&tree->tree_lock/1){+.+.}-{3:3}, at: hfsplus_find_init+0x14a/0x1c0 fs/hfsplus/bfind.c:28

stack backtrace:
CPU: 0 UID: 0 PID: 5813 Comm: syz.0.15 Not tainted 6.12.0-rc7-syzkaller-g2d5404caa8c7-dirty #0
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 lib/dump_stack.c:94 [inline]
 dump_stack_lvl+0x241/0x360 lib/dump_stack.c:120
 print_circular_bug+0x13a/0x1b0 kernel/locking/lockdep.c:2074
 check_noncircular+0x36a/0x4a0 kernel/locking/lockdep.c:2206
 check_prev_add kernel/locking/lockdep.c:3161 [inline]
 check_prevs_add kernel/locking/lockdep.c:3280 [inline]
 validate_chain+0x18ef/0x5920 kernel/locking/lockdep.c:3904
 __lock_acquire+0x1384/0x2050 kernel/locking/lockdep.c:5202
 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5825
 __mutex_lock_common kernel/locking/mutex.c:608 [inline]
 __mutex_lock+0x136/0xd70 kernel/locking/mutex.c:752
 hfsplus_file_extend+0x21b/0x1b70 fs/hfsplus/extents.c:460
 hfsplus_bmap_reserve+0x105/0x4e0 fs/hfsplus/btree.c:358
 __hfsplus_ext_write_extent+0x2af/0x5d0 fs/hfsplus/extents.c:105
 __hfsplus_ext_cache_extent+0x84/0xe10 fs/hfsplus/extents.c:188
 hfsplus_ext_read_extent fs/hfsplus/extents.c:220 [inline]
 hfsplus_file_extend+0x48c/0x1b70 fs/hfsplus/extents.c:464
 hfsplus_bmap_reserve+0x105/0x4e0 fs/hfsplus/btree.c:358
 hfsplus_create_attr+0x1c8/0x640 fs/hfsplus/attributes.c:221
 __hfsplus_setxattr+0x6fe/0x22d0 fs/hfsplus/xattr.c:354
 hfsplus_setxattr+0xb0/0xe0 fs/hfsplus/xattr.c:432
 hfsplus_trusted_setxattr+0x40/0x60 fs/hfsplus/xattr_trusted.c:30
 __vfs_setxattr+0x468/0x4a0 fs/xattr.c:200
 __vfs_setxattr_noperm+0x12e/0x660 fs/xattr.c:234
 vfs_setxattr+0x221/0x430 fs/xattr.c:321
 do_setxattr fs/xattr.c:629 [inline]
 path_setxattr+0x37e/0x4d0 fs/xattr.c:658
 __do_sys_lsetxattr fs/xattr.c:683 [inline]
 __se_sys_lsetxattr fs/xattr.c:679 [inline]
 __x64_sys_lsetxattr+0xb8/0xd0 fs/xattr.c:679
 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
RIP: 0033:0x7fb2e7b7e719
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 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 a8 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fb2e88e4038 EFLAGS: 00000246 ORIG_RAX: 00000000000000bd
RAX: ffffffffffffffda RBX: 00007fb2e7d35f80 RCX: 00007fb2e7b7e719
RDX: 0000000000000000 RSI: 0000000020000200 RDI: 0000000020000040
RBP: 00007fb2e7bf139e R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
R13: 0000000000000000 R14: 00007fb2e7d35f80 R15: 00007ffc148ca008
 </TASK>


Tested on:

commit:         2d5404ca Linux 6.12-rc7
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=17a53ea7980000
kernel config:  https://syzkaller.appspot.com/x/.config?x=d2aeec8c0b2e420c
dashboard link: https://syzkaller.appspot.com/bug?extid=4cba2fd444e9a16ae758
compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
patch:          https://syzkaller.appspot.com/x/patch.diff?x=11f93ea7980000

Re: [syzbot] [hfs?] possible deadlock in hfsplus_file_extend (2)

[Edward Adam Davis]

AA lock, unlock for hfs_bmap_reserve.

#syz test

diff --git a/fs/hfsplus/extents.c b/fs/hfsplus/extents.c
index a6d61685ae79..90366e2da3c7 100644
--- a/fs/hfsplus/extents.c
+++ b/fs/hfsplus/extents.c
@@ -101,7 +101,12 @@ static int __hfsplus_ext_write_extent(struct inode *inode,
 		if (res != -ENOENT)
 			return res;
 		/* Fail early and avoid ENOSPC during the btree operation */
+		mutex_unlock(&hip->extents_lock);
+		mutex_unlock(&fd->tree->tree_lock);
 		res = hfs_bmap_reserve(fd->tree, fd->tree->depth + 1);
+		mutex_lock(&hip->extents_lock);
+		mutex_lock_nested(&fd->tree->tree_lock,
+			hfsplus_btree_lock_class(fd->tree));
 		if (res)
 			return res;
 		hfs_brec_insert(fd, hip->cached_extents,

Re: [syzbot] [hfs?] possible deadlock in hfsplus_file_extend (2)

[syzbot]

Hello,

syzbot has tested the proposed patch and the reproducer did not trigger any issue:

Reported-by: syzbot+4cba2fd444e9a16ae758@syzkaller.appspotmail.com
Tested-by: syzbot+4cba2fd444e9a16ae758@syzkaller.appspotmail.com

Tested on:

commit:         2d5404ca Linux 6.12-rc7
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=11d078c0580000
kernel config:  https://syzkaller.appspot.com/x/.config?x=d2aeec8c0b2e420c
dashboard link: https://syzkaller.appspot.com/bug?extid=4cba2fd444e9a16ae758
compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
patch:          https://syzkaller.appspot.com/x/patch.diff?x=135afe30580000

Note: testing is done by a robot and is best-effort only.