public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Yong Zhang <yong.zhang0@gmail.com>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Vegard Nossum <vegard.nossum@gmail.com>,
	linux-kernel@vger.kernel.org, sergey.senozhatsky@gmail.com,
	bp@alien8.de, Ingo Molnar <mingo@elte.hu>,
	Tejun Heo <tj@kernel.org>, David Rientjes <rientjes@google.com>,
	casteyde.christian@free.fr,
	Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Subject: [PATCH 2/2] lockdep: always try to set ->class_cache in register_lock_class() lockdep_init_map()
Date: Wed, 9 Nov 2011 16:07:14 +0800	[thread overview]
Message-ID: <20111109080714.GC8124@zhy> (raw)
In-Reply-To: <20111109080451.GB8124@zhy>

From: Yong Zhang <yong.zhang0@gmail.com>

commit [62016250 lockdep: Add improved subclass caching] try to
improve performance (expecially to reduce the cost of rq->lock)
when using lockdep, but it fails due to lockdep_init_map() in
which ->class_cache is cleared. The typical caller is
lock_set_subclass(), after that class will not be cached anymore.

This patch try to achive the goal of commit 62016250 by always
setting ->class_cache in register_lock_class().

=== Score comparison of benchmarks ===

for i in `seq 1 10`; do ./perf bench -f simple sched messaging; done

before:  min: 0.604, max: 0.660, avg: 0.622
after:   min: 0.414, max: 0.473, avg: 0.427

for i in `seq 1 10`; do ./perf bench -f simple sched messaging -g 40; done

before:  min: 2.347, max: 2.421, avg: 2.391
after:   min: 1.652, max: 1.699, avg: 1.671

Signed-off-by: Yong Zhang <yong.zhang0@gmail.com>
Cc: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Ingo Molnar <mingo@elte.hu>
---
 kernel/lockdep.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 21ea1dc..e9cbb99 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -722,7 +722,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force)
 
 	class = look_up_lock_class(lock, subclass);
 	if (likely(class))
-		return class;
+		goto out_set_class_cache;
 
 	/*
 	 * Debug-check: all keys must be persistent!
@@ -807,6 +807,7 @@ out_unlock_set:
 	graph_unlock();
 	raw_local_irq_restore(flags);
 
+out_set_class_cache:
 	if (!subclass || force)
 		lock->class_cache[0] = class;
 	else if (subclass < NR_LOCKDEP_CACHING_CLASSES)
-- 
1.7.5.4


  reply	other threads:[~2011-11-09  8:07 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-11-04  9:26 [PATCH 0/4] patches to cure race in lock_set_class() Yong Zhang
2011-11-04  9:26 ` [PATCH 1/4] lockdep: lock_set_subclass() fix Yong Zhang
2011-11-07 12:34   ` Peter Zijlstra
2011-11-07 13:31     ` Yong Zhang
2011-11-07 14:03       ` Tejun Heo
2011-11-07 13:54     ` Borislav Petkov
2011-11-07 15:28     ` Vegard Nossum
2011-11-07 16:10       ` Peter Zijlstra
2011-11-07 16:21         ` Tejun Heo
2011-11-07 16:26           ` Peter Zijlstra
2011-11-08  2:58         ` Yong Zhang
2011-11-08  3:02           ` Yong Zhang
2011-11-08  7:56           ` Peter Zijlstra
2011-11-08  8:14             ` Yong Zhang
2011-11-08  8:46               ` Peter Zijlstra
2011-11-08  9:07                 ` Yong Zhang
2011-11-08  9:37                   ` Yong Zhang
2011-11-08  9:40                   ` Peter Zijlstra
2011-11-09  8:04                     ` [PATCH 1/2] lockdep: kmemcheck: annotate ->lock in lockdep_init_map() Yong Zhang
2011-11-09  8:07                       ` Yong Zhang [this message]
2011-11-18 23:39                         ` [tip:core/locking] lockdep: Always try to set ->class_cache in register_lock_class() lockdep_init_map() tip-bot for Yong Zhang
2011-12-06  9:39                       ` [tip:core/locking] lockdep, kmemcheck: Annotate ->lock in lockdep_init_map() tip-bot for Yong Zhang
2011-12-06 19:56                         ` David Rientjes
2011-12-06 20:14                       ` [tip:perf/urgent] " tip-bot for Yong Zhang
2011-11-08  2:22       ` [PATCH 1/4] lockdep: lock_set_subclass() fix Yong Zhang
2011-11-04  9:26 ` [RFC PATCH 2/4] lockdep: Let register_lock_class() can be called with/without graph_lock Yong Zhang
2011-11-04  9:26 ` [RFC PATCH 3/4] lockdep: split lockdep_init_map() Yong Zhang
2011-11-04  9:26 ` [RFC PATCH 4/4] lockdep: fix race condition in __lock_set_class() Yong Zhang
2011-11-07 12:30   ` Peter Zijlstra
2011-11-07 13:26     ` Yong Zhang
2011-11-06 11:52 ` [PATCH 0/4] patches to cure race in lock_set_class() Borislav Petkov

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=20111109080714.GC8124@zhy \
    --to=yong.zhang0@gmail.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=bp@alien8.de \
    --cc=casteyde.christian@free.fr \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=mitake@dcl.info.waseda.ac.jp \
    --cc=rientjes@google.com \
    --cc=sergey.senozhatsky@gmail.com \
    --cc=tj@kernel.org \
    --cc=vegard.nossum@gmail.com \
    /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