[syzbot] [cgroups?] possible deadlock in static_key_slow_inc (3)

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: 5f63595ebd82 Merge tag 'input-for-v6.4-rc5' of git://git.k..
git tree: upstream
console+strace: https://syzkaller.appspot.com/x/log.txt?x=153fcc63280000
kernel config: https://syzkaller.appspot.com/x/.config?x=7474de833c217bf4
dashboard link: https://syzkaller.appspot.com/bug?extid=2ab700fe1829880a2ec6
compiler: Debian clang version 15.0.7, GNU ld (GNU Binutils for Debian) 2.35.2
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=1321e2fd280000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=11946afd280000

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/d12b9e46ffe8/disk-5f63595e.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/c9044ded7edd/vmlinux-5f63595e.xz
kernel image: https://storage.googleapis.com/syzbot-assets/09f0fd3926e8/bzImage-5f63595e.xz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+2ab700...@syzkaller.appspotmail.com

======================================================
WARNING: possible circular locking dependency detected
6.4.0-rc5-syzkaller-00024-g5f63595ebd82 #0 Not tainted
------------------------------------------------------
syz-executor324/4995 is trying to acquire lock:
ffffffff8cdc3ff0 (cpu_hotplug_lock){++++}-{0:0}, at: static_key_slow_inc+0x12/0x30 kernel/jump_label.c:185

but task is already holding lock:
ffffffff8cf5a688 (freezer_mutex){+.+.}-{3:3}, at: freezer_css_online+0x4f/0x150 kernel/cgroup/legacy_freezer.c:111

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #1 (freezer_mutex){+.+.}-{3:3}:
lock_acquire+0x1e3/0x520 kernel/locking/lockdep.c:5705
__mutex_lock_common+0x1d8/0x2530 kernel/locking/mutex.c:603
__mutex_lock kernel/locking/mutex.c:747 [inline]
mutex_lock_nested+0x1b/0x20 kernel/locking/mutex.c:799
freezer_change_state kernel/cgroup/legacy_freezer.c:389 [inline]
freezer_write+0xa8/0x3f0 kernel/cgroup/legacy_freezer.c:429
cgroup_file_write+0x2ca/0x6a0 kernel/cgroup/cgroup.c:4071
kernfs_fop_write_iter+0x3a6/0x4f0 fs/kernfs/file.c:334
call_write_iter include/linux/fs.h:1868 [inline]
new_sync_write fs/read_write.c:491 [inline]
vfs_write+0x790/0xb20 fs/read_write.c:584
ksys_write+0x1a0/0x2c0 fs/read_write.c:637
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

-> #0 (cpu_hotplug_lock){++++}-{0:0}:
check_prev_add kernel/locking/lockdep.c:3113 [inline]
check_prevs_add kernel/locking/lockdep.c:3232 [inline]
validate_chain+0x166b/0x58f0 kernel/locking/lockdep.c:3847
__lock_acquire+0x1316/0x2070 kernel/locking/lockdep.c:5088
lock_acquire+0x1e3/0x520 kernel/locking/lockdep.c:5705
percpu_down_read include/linux/percpu-rwsem.h:51 [inline]
cpus_read_lock+0x42/0x150 kernel/cpu.c:310
static_key_slow_inc+0x12/0x30 kernel/jump_label.c:185
freezer_css_online+0xc6/0x150 kernel/cgroup/legacy_freezer.c:117
online_css+0xba/0x260 kernel/cgroup/cgroup.c:5491
css_create kernel/cgroup/cgroup.c:5562 [inline]
cgroup_apply_control_enable+0x7d1/0xae0 kernel/cgroup/cgroup.c:3249
cgroup_mkdir+0xd8f/0x1000 kernel/cgroup/cgroup.c:5758
kernfs_iop_mkdir+0x279/0x400 fs/kernfs/dir.c:1219
vfs_mkdir+0x29d/0x450 fs/namei.c:4115
do_mkdirat+0x264/0x520 fs/namei.c:4138
__do_sys_mkdirat fs/namei.c:4153 [inline]
__se_sys_mkdirat fs/namei.c:4151 [inline]
__x64_sys_mkdirat+0x89/0xa0 fs/namei.c:4151
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd

other info that might help us debug this:

Possible unsafe locking scenario:

CPU0 CPU1
---- ----
lock(freezer_mutex);
lock(cpu_hotplug_lock);
lock(freezer_mutex);
rlock(cpu_hotplug_lock);

*** DEADLOCK ***

4 locks held by syz-executor324/4995:
#0: ffff88807da6c460 (sb_writers#10){.+.+}-{0:0}, at: mnt_want_write+0x3f/0x90 fs/namespace.c:394
#1: ffff888075c6eee0 (&type->i_mutex_dir_key#7/1){+.+.}-{3:3}, at: inode_lock_nested include/linux/fs.h:810 [inline]
#1: ffff888075c6eee0 (&type->i_mutex_dir_key#7/1){+.+.}-{3:3}, at: filename_create+0x260/0x530 fs/namei.c:3884
#2: ffffffff8cf4f768 (cgroup_mutex){+.+.}-{3:3}, at: cgroup_lock include/linux/cgroup.h:370 [inline]
#2: ffffffff8cf4f768 (cgroup_mutex){+.+.}-{3:3}, at: cgroup_kn_lock_live+0xe9/0x290 kernel/cgroup/cgroup.c:1683
#3: ffffffff8cf5a688 (freezer_mutex){+.+.}-{3:3}, at: freezer_css_online+0x4f/0x150 kernel/cgroup/legacy_freezer.c:111

stack backtrace:
CPU: 0 PID: 4995 Comm: syz-executor324 Not tainted 6.4.0-rc5-syzkaller-00024-g5f63595ebd82 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/25/2023
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0x1e7/0x2d0 lib/dump_stack.c:106
check_noncircular+0x2fe/0x3b0 kernel/locking/lockdep.c:2188
check_prev_add kernel/locking/lockdep.c:3113 [inline]
check_prevs_add kernel/locking/lockdep.c:3232 [inline]
validate_chain+0x166b/0x58f0 kernel/locking/lockdep.c:3847
__lock_acquire+0x1316/0x2070 kernel/locking/lockdep.c:5088
lock_acquire+0x1e3/0x520 kernel/locking/lockdep.c:5705
percpu_down_read include/linux/percpu-rwsem.h:51 [inline]
cpus_read_lock+0x42/0x150 kernel/cpu.c:310
static_key_slow_inc+0x12/0x30 kernel/jump_label.c:185
freezer_css_online+0xc6/0x150 kernel/cgroup/legacy_freezer.c:117
online_css+0xba/0x260 kernel/cgroup/cgroup.c:5491
css_create kernel/cgroup/cgroup.c:5562 [inline]
cgroup_apply_control_enable+0x7d1/0xae0 kernel/cgroup/cgroup.c:3249
cgroup_mkdir+0xd8f/0x1000 kernel/cgroup/cgroup.c:5758
kernfs_iop_mkdir+0x279/0x400 fs/kernfs/dir.c:1219
vfs_mkdir+0x29d/0x450 fs/namei.c:4115
do_mkdirat+0x264/0x520 fs/namei.c:4138
__do_sys_mkdirat fs/namei.c:4153 [inline]
__se_sys_mkdirat fs/namei.c:4151 [inline]
__x64_sys_mkdirat+0x89/0xa0 fs/namei.c:4151
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x41/0xc0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x63/0xcd
RIP: 0033:0x7f7e55624e09
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 51 15 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 c0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007fff7236ac98 EFLAGS: 00000246 ORIG_RAX: 0000000000000102
RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007f7e55624e09
RDX: 00000000000001ff RSI: 0000000020000180 RDI: 0000000000000003
RBP: 0000000000000000 R08: 00007fff7236acc0 R09: 00007fff7236acc0
R10: 00007fff7236acc0 R11: 0000000000000246 R12: 00007fff7236acbc
R13: 00007fff7236acd0 R14: 00007fff7236ad10 R15: 0000000000000000
</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 syzk...@googlegroups.com.

syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.

If the bug is already fixed, 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 change bug's subsystems, reply with:
#syz set subsystems: new-subsystem
(See the list of subsystem names on the web dashboard)

If the bug is a duplicate of another bug, reply with:
#syz dup: exact-subject-of-another-report

If you want to undo deduplication, reply with:
#syz undup

[syzbot]

syzbot has bisected this issue to:

commit eb1d969203eb8212741751f88dcf5cb56bb11830
Author: Hans Verkuil <hver...@xs4all.nl>
Date: Fri Oct 21 12:21:25 2022 +0000

media: vivid: fix control handler mutex deadlock

bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=16af9d95280000
start commit: 5f63595ebd82 Merge tag 'input-for-v6.4-rc5' of git://git.k..
git tree: upstream
final oops: https://syzkaller.appspot.com/x/report.txt?x=15af9d95280000
console output: https://syzkaller.appspot.com/x/log.txt?x=11af9d95280000
kernel config: https://syzkaller.appspot.com/x/.config?x=3c980bfe8b399968
dashboard link: https://syzkaller.appspot.com/bug?extid=2ab700fe1829880a2ec6
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=142e0463280000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=10a32add280000

Reported-by: syzbot+2ab700...@syzkaller.appspotmail.com
Fixes: eb1d969203eb ("media: vivid: fix control handler mutex deadlock")

For information about bisection process see: https://goo.gl/tpsmEJ#bisection

[Tetsuo Handa]

#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master

Subject: [PATCH] cgroup,freezer: hold cpu_hotplug_lock before freezer_mutex in freezer_css_{online,offline}()

syzbot is again reporting circular locking dependency between
cpu_hotplug_lock and freezer_mutex. Do like what we did with
commit 57dcd64c7e036299 ("cgroup,freezer: hold cpu_hotplug_lock
before freezer_mutex").

Reported-by: syzbot <syzbot+2ab700...@syzkaller.appspotmail.com>
Closes: https://syzkaller.appspot.com/bug?extid=2ab700fe1829880a2ec6
Signed-off-by: Tetsuo Handa <penguin...@I-love.SAKURA.ne.jp>
---
kernel/cgroup/legacy_freezer.c | 8 ++++++--
1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/kernel/cgroup/legacy_freezer.c b/kernel/cgroup/legacy_freezer.c
index 936473203a6b..122dacb3a443 100644
--- a/kernel/cgroup/legacy_freezer.c
+++ b/kernel/cgroup/legacy_freezer.c
@@ -108,16 +108,18 @@ static int freezer_css_online(struct cgroup_subsys_state *css)
struct freezer *freezer = css_freezer(css);
struct freezer *parent = parent_freezer(freezer);

+ cpus_read_lock();
mutex_lock(&freezer_mutex);

freezer->state |= CGROUP_FREEZER_ONLINE;

if (parent && (parent->state & CGROUP_FREEZING)) {
freezer->state |= CGROUP_FREEZING_PARENT | CGROUP_FROZEN;
- static_branch_inc(&freezer_active);
+ static_branch_inc_cpuslocked(&freezer_active);
}

mutex_unlock(&freezer_mutex);
+ cpus_read_unlock();
return 0;
}

@@ -132,14 +134,16 @@ static void freezer_css_offline(struct cgroup_subsys_state *css)
{
struct freezer *freezer = css_freezer(css);

+ cpus_read_lock();
mutex_lock(&freezer_mutex);

if (freezer->state & CGROUP_FREEZING)
- static_branch_dec(&freezer_active);
+ static_branch_dec_cpuslocked(&freezer_active);

freezer->state = 0;

mutex_unlock(&freezer_mutex);
+ cpus_read_unlock();
}

static void freezer_css_free(struct cgroup_subsys_state *css)
--
2.18.4

[Tetsuo Handa]

[syzbot]

Hello,

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

Reported-and-tested-by: syzbot+2ab700...@syzkaller.appspotmail.com

Tested on:

commit: 022ce886 Merge tag 'i2c-for-6.4-rc6' of git://git.kern..
git tree: upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=158d352d280000

kernel config: https://syzkaller.appspot.com/x/.config?x=7474de833c217bf4
dashboard link: https://syzkaller.appspot.com/bug?extid=2ab700fe1829880a2ec6
compiler: Debian clang version 15.0.7, GNU ld (GNU Binutils for Debian) 2.35.2

patch: https://syzkaller.appspot.com/x/patch.diff?x=1567d1fd280000

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