From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
David Miller <davem@davemloft.net>,
hugh@veritas.com, mingo@elte.hu, akpm@linux-foundation.org,
linux-kernel@vger.kernel.org, davej@redhat.com
Subject: Re: [RFC][PATCH 7/7] lockdep: spin_lock_nest_lock()
Date: Mon, 04 Aug 2008 11:06:10 -0700 [thread overview]
Message-ID: <48974512.7000206@goop.org> (raw)
In-Reply-To: <20080804131012.246115111@chello.nl>
Peter Zijlstra wrote:
> Expose the new lock protection lock.
>
> This can be used to annotate places where we take multiple locks of the
> same class and avoid deadlocks by always taking another (top-level) lock
> first.
>
OK, so the expected usage is:
spin_lock(&outer_lock);
/* take in any order */
spin_lock_nest_lock(&inner[0], &outer_lock);
spin_lock_nest_lock(&inner[2], &outer_lock);
spin_lock_nest_lock(&inner[1], &outer_lock);
...
?
And it's OK to
1. take inner locks one at a time without holding the outer lock
2. use plain spin_lock on inner locks when you're taking them one at
a time, and
3. release the outer lock before releasing the inner locks
but it's not OK to try to use different outer locks for a given inner lock.
Yes?
J
> NOTE: we're still bound to the MAX_LOCK_DEPTH (48) limit.
>
> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> ---
> include/linux/lockdep.h | 2 ++
> include/linux/spinlock.h | 6 ++++++
> kernel/spinlock.c | 11 +++++++++++
> 3 files changed, 19 insertions(+)
>
> Index: linux-2.6/include/linux/lockdep.h
> ===================================================================
> --- linux-2.6.orig/include/linux/lockdep.h
> +++ linux-2.6/include/linux/lockdep.h
> @@ -410,8 +410,10 @@ static inline void print_irqtrace_events
> #ifdef CONFIG_DEBUG_LOCK_ALLOC
> # ifdef CONFIG_PROVE_LOCKING
> # define spin_acquire(l, s, t, i) lock_acquire(l, s, t, 0, 2, NULL, i)
> +# define spin_acquire_nest(l, s, t, n, i) lock_acquire(l, s, t, 0, 2, n, i)
> # else
> # define spin_acquire(l, s, t, i) lock_acquire(l, s, t, 0, 1, NULL, i)
> +# define spin_acquire_nest(l, s, t, n, i) lock_acquire(l, s, t, 0, 1, NULL, i)
> # endif
> # define spin_release(l, n, i) lock_release(l, n, i)
> #else
> Index: linux-2.6/include/linux/spinlock.h
> ===================================================================
> --- linux-2.6.orig/include/linux/spinlock.h
> +++ linux-2.6/include/linux/spinlock.h
> @@ -183,8 +183,14 @@ do { \
>
> #ifdef CONFIG_DEBUG_LOCK_ALLOC
> # define spin_lock_nested(lock, subclass) _spin_lock_nested(lock, subclass)
> +# define spin_lock_nest_lock(lock, nest_lock) \
> + do { \
> + typecheck(struct lockdep_map *, &nest_lock->dep_map); \
> + _spin_lock_nest_lock(lock, &nest_lock->dep_map); \
> + } while (0)
> #else
> # define spin_lock_nested(lock, subclass) _spin_lock(lock)
> +# define spin_lock_nest_lock(lock, nest_lock) _spin_lock(lock)
> #endif
>
> #define write_lock(lock) _write_lock(lock)
> Index: linux-2.6/kernel/spinlock.c
> ===================================================================
> --- linux-2.6.orig/kernel/spinlock.c
> +++ linux-2.6/kernel/spinlock.c
> @@ -292,6 +292,7 @@ void __lockfunc _spin_lock_nested(spinlo
> }
>
> EXPORT_SYMBOL(_spin_lock_nested);
> +
> unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, int subclass)
> {
> unsigned long flags;
> @@ -314,6 +315,16 @@ unsigned long __lockfunc _spin_lock_irqs
>
> EXPORT_SYMBOL(_spin_lock_irqsave_nested);
>
> +void __lockfunc _spin_lock_nest_lock(spinlock_t *lock,
> + struct lockdep_map *nest_lock)
> +{
> + preempt_disable();
> + spin_acquire_nest(&lock->dep_map, 0, 0, nest_lock, _RET_IP_);
> + LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock);
> +}
> +
> +EXPORT_SYMBOL(_spin_lock_nest_lock)
> +
> #endif
>
> void __lockfunc _spin_unlock(spinlock_t *lock)
>
>
next prev parent reply other threads:[~2008-08-04 18:06 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-08-04 13:03 [RFC][PATCH 0/7] lockdep Peter Zijlstra
2008-08-04 13:03 ` [RFC][PATCH 1/7] lockdep: Fix combinatorial explosion in lock subgraph traversal Peter Zijlstra
2008-08-05 8:34 ` David Miller
2008-08-05 8:46 ` Peter Zijlstra
2008-08-13 3:48 ` Tim Pepper
2008-08-13 10:56 ` Ingo Molnar
2008-08-04 13:03 ` [RFC][PATCH 2/7] lockdep: lock_set_subclass - reset a held locks subclass Peter Zijlstra
2008-08-05 8:35 ` David Miller
2008-08-04 13:03 ` [RFC][PATCH 3/7] lockdep: re-annotate scheduler runqueues Peter Zijlstra
2008-08-05 8:35 ` David Miller
2008-08-04 13:03 ` [RFC][PATCH 4/7] lockdep: shrink held_lock structure Peter Zijlstra
2008-08-05 16:08 ` Peter Zijlstra
2008-08-06 7:17 ` Peter Zijlstra
2008-08-04 13:03 ` [RFC][PATCH 5/7] lockdep: map_acquire Peter Zijlstra
2008-08-04 13:03 ` [RFC][PATCH 6/7] lockdep: lock protection locks Peter Zijlstra
2008-08-04 13:03 ` [RFC][PATCH 7/7] lockdep: spin_lock_nest_lock() Peter Zijlstra
2008-08-04 14:07 ` Roland Dreier
2008-08-04 14:19 ` Peter Zijlstra
2008-08-04 14:26 ` Roland Dreier
2008-08-04 14:32 ` Peter Zijlstra
2008-08-04 14:53 ` Dave Jones
2008-08-04 14:56 ` Peter Zijlstra
2008-08-04 16:26 ` Andrea Arcangeli
2008-08-04 16:38 ` Peter Zijlstra
2008-08-04 17:27 ` Andrea Arcangeli
2008-08-04 17:46 ` Andrea Arcangeli
2008-08-04 17:57 ` [PATCH] workaround minor lockdep bug triggered by mm_take_all_locks Andrea Arcangeli
2008-08-04 18:48 ` Peter Zijlstra
2008-08-04 18:56 ` Roland Dreier
2008-08-04 19:05 ` Peter Zijlstra
2008-08-04 20:15 ` Andrea Arcangeli
2008-08-04 20:37 ` Peter Zijlstra
2008-08-04 21:09 ` Andrea Arcangeli
2008-08-04 21:14 ` Pekka Enberg
2008-08-04 21:30 ` Andrea Arcangeli
2008-08-04 21:41 ` Andrew Morton
2008-08-04 22:12 ` Andrea Arcangeli
2008-08-04 21:42 ` Arjan van de Ven
2008-08-04 22:30 ` Andrea Arcangeli
2008-08-04 23:38 ` Arjan van de Ven
2008-08-05 0:47 ` Andrea Arcangeli
2008-08-04 21:27 ` Arjan van de Ven
2008-08-04 21:54 ` Andrea Arcangeli
2008-08-04 21:57 ` David Miller
2008-08-05 2:00 ` Roland Dreier
2008-08-05 2:18 ` Andrea Arcangeli
2008-08-05 12:02 ` Roland Dreier
2008-08-05 12:20 ` Andrea Arcangeli
2008-08-04 18:48 ` [RFC][PATCH 7/7] lockdep: spin_lock_nest_lock() Peter Zijlstra
2008-08-04 21:32 ` David Miller
2008-08-04 18:06 ` Jeremy Fitzhardinge [this message]
2008-08-04 18:54 ` Peter Zijlstra
2008-08-04 19:26 ` Jeremy Fitzhardinge
2008-08-04 19:31 ` Linus Torvalds
2008-08-04 19:39 ` Peter Zijlstra
2008-08-04 20:16 ` Jeremy Fitzhardinge
2008-10-08 15:27 ` Steven Rostedt
2008-10-08 15:43 ` Linus Torvalds
2008-10-08 16:03 ` Steven Rostedt
2008-10-08 16:19 ` Linus Torvalds
2008-10-08 16:53 ` Steven Rostedt
2008-10-08 15:52 ` Nick Piggin
2008-10-08 17:18 ` Steven Rostedt
2008-08-07 11:25 ` Peter Zijlstra
2008-08-07 11:25 ` [RFC][PATCH 8/7] lockdep: annotate mm_take_all_locks() Peter Zijlstra
2008-08-07 11:25 ` [RFC][PATCH 9/7] mm: fix mm_take_all_locks() locking order Peter Zijlstra
2008-08-07 12:14 ` Hugh Dickins
2008-08-07 12:41 ` Peter Zijlstra
2008-08-07 13:27 ` Hugh Dickins
2008-08-07 21:46 ` Andrea Arcangeli
2008-08-08 1:34 ` Andrea Arcangeli
2008-08-08 7:16 ` Peter Zijlstra
2008-08-11 10:08 ` [RFC][PATCH 0/7] lockdep Ingo Molnar
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=48974512.7000206@goop.org \
--to=jeremy@goop.org \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=davej@redhat.com \
--cc=davem@davemloft.net \
--cc=hugh@veritas.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox