linux_dsm_epyc7002/kernel/rcu
Paul E. McKenney 35a35909ec rcu: Reject RCU_LOCKDEP_WARN() false positives
[ Upstream commit 3066820034b5dd4e89bd74a7739c51c2d6f5e554 ]

If another lockdep report runs concurrently with an RCU lockdep report
from RCU_LOCKDEP_WARN(), the following sequence of events can occur:

1.	debug_lockdep_rcu_enabled() sees that lockdep is enabled
	when called from (say) synchronize_rcu().

2.	Lockdep is disabled by a concurrent lockdep report.

3.	debug_lockdep_rcu_enabled() evaluates its lockdep-expression
	argument, for example, lock_is_held(&rcu_bh_lock_map).

4.	Because lockdep is now disabled, lock_is_held() plays it safe and
	returns the constant 1.

5.	But in this case, the constant 1 is not safe, because invoking
	synchronize_rcu() under rcu_read_lock_bh() is disallowed.

6.	debug_lockdep_rcu_enabled() wrongly invokes lockdep_rcu_suspicious(),
	resulting in a false-positive splat.

This commit therefore changes RCU_LOCKDEP_WARN() to check
debug_lockdep_rcu_enabled() after checking the lockdep expression,
so that any "safe" returns from lock_is_held() are rejected by
debug_lockdep_rcu_enabled().  This requires memory ordering, which is
supplied by READ_ONCE(debug_locks).  The resulting volatile accesses
prevent the compiler from reordering and the fact that only one variable
is being accessed prevents the underlying hardware from reordering.
The combination works for IA64, which can reorder reads to the same
location, but this is defeated by the volatile accesses, which compile
to load instructions that provide ordering.

Reported-by: syzbot+dde0cc33951735441301@syzkaller.appspotmail.com
Reported-by: Matthew Wilcox <willy@infradead.org>
Reported-by: syzbot+88e4f02896967fe1ab0d@syzkaller.appspotmail.com
Reported-by: Thomas Gleixner <tglx@linutronix.de>
Suggested-by: Boqun Feng <boqun.feng@gmail.com>
Reviewed-by: Boqun Feng <boqun.feng@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
2021-07-20 16:05:38 +02:00
..
Kconfig rcu: Reduce leaf fanout for strict RCU grace periods 2020-08-24 18:40:23 -07:00
Kconfig.debug Merge branch 'strictgp.2020.08.24a' into HEAD 2020-09-03 09:47:42 -07:00
Makefile rcuperf: Change rcuperf to rcuscale 2020-08-24 18:39:24 -07:00
rcu_segcblist.c rcu/segcblist: Prevent useless GP start if no CBs to accelerate 2020-09-03 09:39:59 -07:00
rcu_segcblist.h rcu: Remove kfree_rcu() special casing and lazy-callback handling 2020-01-24 10:24:31 -08:00
rcu.h srcu: Fix broken node geometry after early ssp init 2021-07-20 16:05:38 +02:00
rcuscale.c rcuperf: Change rcuperf to rcuscale 2020-08-24 18:39:24 -07:00
rcutorture.c rcutorture: Allow pointer leaks to test diagnostic code 2020-08-24 18:45:36 -07:00
refscale.c refperf: Avoid null pointer dereference when buf fails to allocate 2020-08-24 18:45:35 -07:00
srcutiny.c rcu: Use CONFIG_PREEMPTION where appropriate 2019-12-09 12:37:51 -08:00
srcutree.c srcu: Fix broken node geometry after early ssp init 2021-07-20 16:05:38 +02:00
sync.c rcu/sync: Simplify the state machine 2019-05-28 09:05:23 -07:00
tasks.h rcu-tasks: Move RCU-tasks initialization to before early_initcall() 2021-01-19 18:27:28 +01:00
tiny.c rcu: Rename *_kfree_callback/*_kfree_rcu_offset/kfree_call_* 2020-06-29 11:59:25 -07:00
tree_exp.h rcu: Initialize at declaration time in rcu_exp_handler() 2020-08-24 18:36:03 -07:00
tree_plugin.h rcu/nocb: Perform deferred wake up before last idle's need_resched() check 2021-03-04 11:38:35 +01:00
tree_stall.h rcu: Don't invoke try_invoke_on_locked_down_task() with irqs disabled 2020-11-10 17:10:38 -08:00
tree.c srcu: Fix broken node geometry after early ssp init 2021-07-20 16:05:38 +02:00
tree.h Merge branch 'strictgp.2020.08.24a' into HEAD 2020-09-03 09:47:42 -07:00
update.c rcu: Reject RCU_LOCKDEP_WARN() false positives 2021-07-20 16:05:38 +02:00