seqlock: Do the lockdep annotation before locking in do_write_seqcount_begin_nested()
PATCH 1/1 seqlock: Do the lockdep annotation before locking in do_write_seqcount_begin_nested()
It was brought up by Tetsuo that the following sequence
write_seqlock_irqsave() printk_deferred_enter()
could lead to a deadlock if the lockdep annotation within write_seqlock_irqsave() triggers. The problem is that the sequence counter is incremented before the lockdep annotation is performed. The lockdep splat would then attempt to invoke printk() but the reader side, of the same seqcount, could have a tty_port::lock acquired waiting for the sequence number to become even again.
The other lockdep annotations come before the actual locking because "we want to see the locking error before it happens". There is no reason why seqcount should be different here.
Do the lockdep annotation first then perform the locking operation (the sequence increment).
---
The previous submission was in
https://lore.kernel.org/r/[email protected]
The mm bits in the referenced thread have been merged as of -rc1.
include/linux/seqlock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/seqlock.h +1 -1
@@ -512,8 +512,8 @@ do { \static inline void do_write_seqcount_begin_nested (seqcount_t *s,int subclass) { -do_raw_write_seqcount_begin (s);seqcount_acquire (&s->dep_map, subclass,0 , _RET_IP_); +do_raw_write_seqcount_begin (s); }/**
[tip: locking/urgent] locking/seqlock: Do the lockdep annotation before locking in do_write_seqcount_begin_nested()
The following commit has been merged into the locking/urgent branch of tip:
The following commit has been merged into the locking/urgent branch of tip:
Commit-ID: 41b43b6c6e30a832c790b010a06772e793bca193 Gitweb: https://git.kernel.org/tip/41b43b6c6e30a832c790b010a06772e793bca193 Author: Sebastian Andrzej Siewior <[email protected]> AuthorDate: Wed, 20 Sep 2023 12:46:27 +02:00 Committer: Ingo Molnar <[email protected]> CommitterDate: Thu, 21 Sep 2023 08:37:44 +02:00
locking/seqlock: Do the lockdep annotation before locking in do_write_seqcount_begin_nested()
It was brought up by Tetsuo that the following sequence:
write_seqlock_irqsave()
printk_deferred_enter()
could lead to a deadlock if the lockdep annotation within write_seqlock_irqsave() triggers.
The problem is that the sequence counter is incremented before the lockdep annotation is performed. The lockdep splat would then attempt to invoke printk() but the reader side, of the same seqcount, could have a tty_port::lock acquired waiting for the sequence number to become even again.
The other lockdep annotations come before the actual locking because "we want to see the locking error before it happens". There is no reason why seqcount should be different here.
Do the lockdep annotation first then perform the locking operation (the sequence increment).
--- include/linux/seqlock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/seqlock.h +1 -1
@@ -512,8 +512,8 @@ do { \static inline void do_write_seqcount_begin_nested (seqcount_t *s,int subclass) { -do_raw_write_seqcount_begin (s);seqcount_acquire (&s->dep_map, subclass,0 , _RET_IP_); +do_raw_write_seqcount_begin (s); }/**