Files
linux/include/linux
Roman Gushchin 956cf21cd1 mm: memcg: synchronize objcg lists with a dedicated spinlock
commit 0764db9b49 upstream.

Alexander reported a circular lock dependency revealed by the mmap1 ltp
test:

  LOCKDEP_CIRCULAR (suite: ltp, case: mtest06 (mmap1))
          WARNING: possible circular locking dependency detected
          5.17.0-20220113.rc0.git0.f2211f194038.300.fc35.s390x+debug #1 Not tainted
          ------------------------------------------------------
          mmap1/202299 is trying to acquire lock:
          00000001892c0188 (css_set_lock){..-.}-{2:2}, at: obj_cgroup_release+0x4a/0xe0
          but task is already holding lock:
          00000000ca3b3818 (&sighand->siglock){-.-.}-{2:2}, at: force_sig_info_to_task+0x38/0x180
          which lock already depends on the new lock.
          the existing dependency chain (in reverse order) is:
          -> #1 (&sighand->siglock){-.-.}-{2:2}:
                 __lock_acquire+0x604/0xbd8
                 lock_acquire.part.0+0xe2/0x238
                 lock_acquire+0xb0/0x200
                 _raw_spin_lock_irqsave+0x6a/0xd8
                 __lock_task_sighand+0x90/0x190
                 cgroup_freeze_task+0x2e/0x90
                 cgroup_migrate_execute+0x11c/0x608
                 cgroup_update_dfl_csses+0x246/0x270
                 cgroup_subtree_control_write+0x238/0x518
                 kernfs_fop_write_iter+0x13e/0x1e0
                 new_sync_write+0x100/0x190
                 vfs_write+0x22c/0x2d8
                 ksys_write+0x6c/0xf8
                 __do_syscall+0x1da/0x208
                 system_call+0x82/0xb0
          -> #0 (css_set_lock){..-.}-{2:2}:
                 check_prev_add+0xe0/0xed8
                 validate_chain+0x736/0xb20
                 __lock_acquire+0x604/0xbd8
                 lock_acquire.part.0+0xe2/0x238
                 lock_acquire+0xb0/0x200
                 _raw_spin_lock_irqsave+0x6a/0xd8
                 obj_cgroup_release+0x4a/0xe0
                 percpu_ref_put_many.constprop.0+0x150/0x168
                 drain_obj_stock+0x94/0xe8
                 refill_obj_stock+0x94/0x278
                 obj_cgroup_charge+0x164/0x1d8
                 kmem_cache_alloc+0xac/0x528
                 __sigqueue_alloc+0x150/0x308
                 __send_signal+0x260/0x550
                 send_signal+0x7e/0x348
                 force_sig_info_to_task+0x104/0x180
                 force_sig_fault+0x48/0x58
                 __do_pgm_check+0x120/0x1f0
                 pgm_check_handler+0x11e/0x180
          other info that might help us debug this:
           Possible unsafe locking scenario:
                 CPU0                    CPU1
                 ----                    ----
            lock(&sighand->siglock);
                                         lock(css_set_lock);
                                         lock(&sighand->siglock);
            lock(css_set_lock);
           *** DEADLOCK ***
          2 locks held by mmap1/202299:
           #0: 00000000ca3b3818 (&sighand->siglock){-.-.}-{2:2}, at: force_sig_info_to_task+0x38/0x180
           #1: 00000001892ad560 (rcu_read_lock){....}-{1:2}, at: percpu_ref_put_many.constprop.0+0x0/0x168
          stack backtrace:
          CPU: 15 PID: 202299 Comm: mmap1 Not tainted 5.17.0-20220113.rc0.git0.f2211f194038.300.fc35.s390x+debug #1
          Hardware name: IBM 3906 M04 704 (LPAR)
          Call Trace:
            dump_stack_lvl+0x76/0x98
            check_noncircular+0x136/0x158
            check_prev_add+0xe0/0xed8
            validate_chain+0x736/0xb20
            __lock_acquire+0x604/0xbd8
            lock_acquire.part.0+0xe2/0x238
            lock_acquire+0xb0/0x200
            _raw_spin_lock_irqsave+0x6a/0xd8
            obj_cgroup_release+0x4a/0xe0
            percpu_ref_put_many.constprop.0+0x150/0x168
            drain_obj_stock+0x94/0xe8
            refill_obj_stock+0x94/0x278
            obj_cgroup_charge+0x164/0x1d8
            kmem_cache_alloc+0xac/0x528
            __sigqueue_alloc+0x150/0x308
            __send_signal+0x260/0x550
            send_signal+0x7e/0x348
            force_sig_info_to_task+0x104/0x180
            force_sig_fault+0x48/0x58
            __do_pgm_check+0x120/0x1f0
            pgm_check_handler+0x11e/0x180
          INFO: lockdep is turned off.

In this example a slab allocation from __send_signal() caused a
refilling and draining of a percpu objcg stock, resulted in a releasing
of another non-related objcg.  Objcg release path requires taking the
css_set_lock, which is used to synchronize objcg lists.

This can create a circular dependency with the sighandler lock, which is
taken with the locked css_set_lock by the freezer code (to freeze a
task).

In general it seems that using css_set_lock to synchronize objcg lists
makes any slab allocations and deallocation with the locked css_set_lock
and any intervened locks risky.

To fix the problem and make the code more robust let's stop using
css_set_lock to synchronize objcg lists and use a new dedicated spinlock
instead.

Link: https://lkml.kernel.org/r/Yfm1IHmoGdyUR81T@carbon.dhcp.thefacebook.com
Fixes: bf4f059954 ("mm: memcg/slab: obj_cgroup API")
Signed-off-by: Roman Gushchin <guro@fb.com>
Reported-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Tested-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Reviewed-by: Waiman Long <longman@redhat.com>
Acked-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Reviewed-by: Jeremy Linton <jeremy.linton@arm.com>
Tested-by: Jeremy Linton <jeremy.linton@arm.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2022-02-16 12:56:38 +01:00
..
2021-09-02 22:49:16 +02:00
2021-07-21 19:54:21 -07:00
2021-02-02 00:16:57 +01:00
2021-09-20 12:43:34 +01:00
2021-06-03 14:04:18 -07:00
2020-06-16 14:19:57 +02:00
2021-01-23 14:57:21 +01:00
2021-08-29 14:47:42 +03:00
2021-05-06 19:24:11 -07:00
2021-10-07 16:51:57 +02:00
2021-04-14 16:30:30 +03:00
2021-05-24 21:13:05 -07:00
2021-08-16 10:50:32 -06:00
2021-01-24 14:27:17 +01:00
2021-02-26 09:41:03 -08:00
2021-04-08 16:04:20 -07:00
2021-09-08 15:32:35 -07:00
2021-01-21 14:06:00 -07:00
2021-08-26 16:52:03 -07:00
2021-12-14 10:57:11 +01:00
2021-11-25 09:48:41 +01:00
2021-07-20 09:20:49 -07:00
2021-09-07 21:17:28 +02:00
2021-03-22 03:57:39 +01:00
2021-08-26 15:32:28 -04:00
2020-12-10 12:42:59 -06:00
2021-07-27 11:00:36 +02:00
2021-09-06 07:20:56 -04:00
2021-04-12 15:04:23 +02:00
2020-11-16 16:31:41 +01:00
2021-11-25 09:49:08 +01:00
2021-06-25 19:57:01 -04:00
2021-07-26 15:09:44 +02:00
2021-07-27 20:11:45 +01:00
2021-07-27 20:11:44 +01:00
2021-09-03 09:58:13 -07:00
2020-07-24 17:12:41 -07:00
2021-02-11 13:24:44 -08:00
2021-08-05 11:46:42 +01:00
2020-08-04 21:02:38 -04:00
2020-12-02 22:10:35 -08:00
2021-04-30 11:20:40 -07:00
2020-11-23 14:16:55 +00:00
2021-08-16 21:26:33 -07:00
2020-09-16 08:54:53 -05:00
2020-11-24 16:47:48 +01:00
2021-07-15 10:13:49 -07:00
2021-07-27 17:05:06 +01:00
2021-08-19 09:02:55 +09:00
2021-07-27 09:29:15 +02:00
2021-01-21 16:16:10 +00:00
2021-06-01 10:29:21 +01:00
2021-02-26 09:41:03 -08:00
2020-09-04 09:25:20 -07:00
2021-11-18 19:15:51 +01:00
2021-05-12 20:11:17 +02:00
2021-06-17 13:09:27 -04:00
2021-06-15 17:46:57 +02:00
2021-07-01 11:06:02 -07:00
2021-06-29 10:53:53 -07:00
2021-07-01 11:06:05 -07:00
2021-08-23 13:19:12 +02:00
2020-07-01 10:49:02 +02:00
2021-09-17 13:52:17 +01:00
2021-02-26 09:40:59 -08:00
2021-02-03 19:05:50 +01:00
2021-06-16 17:20:40 -05:00
2021-07-06 10:37:46 -05:00
2020-08-31 12:52:33 -07:00
2021-07-20 06:55:20 -07:00
2021-06-07 14:11:47 -07:00
2021-01-24 14:27:17 +01:00
2021-08-18 22:08:24 +02:00
2021-02-26 09:41:03 -08:00
2021-09-02 21:38:56 +02:00
2021-10-07 16:51:57 +02:00
2020-12-03 11:20:52 +01:00
2021-01-16 23:19:26 +01:00
2021-08-17 17:50:51 +02:00
2021-03-18 12:58:27 -04:00
2020-12-10 16:17:15 +01:00
2020-12-10 10:45:36 +01:00
2021-05-10 16:03:35 -07:00
2021-08-06 13:41:48 -07:00
2021-08-19 09:02:55 +09:00
2021-06-24 15:49:32 +02:00
2021-11-21 13:44:12 +01:00
2021-07-01 11:06:03 -07:00
2020-10-30 21:57:04 +01:00
2021-07-27 12:17:21 +02:00
2021-07-27 12:12:08 +02:00
2021-02-13 17:17:53 +01:00
2021-09-08 15:32:35 -07:00
2021-03-30 17:06:49 -07:00
2020-11-09 18:39:36 +01:00
2021-09-15 09:22:35 -06:00
2020-07-04 09:35:36 -05:00
2020-09-10 14:03:31 -07:00
2020-08-19 14:13:20 +02:00
2021-01-21 16:16:10 +00:00
2021-08-10 11:50:55 -04:00
2021-02-09 12:15:07 +01:00
2021-08-11 06:44:24 -04:00
2021-01-21 14:06:00 -07:00
2021-12-14 10:57:15 +01:00
2021-03-30 13:42:33 -04:00
2021-01-24 14:27:17 +01:00
2021-02-08 22:58:55 +01:00