linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Ingo Molnar <mingo@elte.hu>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
	Gregory Haskins <ghaskins@novell.com>,
	Matthew Wilcox <matthew@wil.cx>, Andi Kleen <andi@firstfloor.org>,
	Chris Mason <chris.mason@oracle.com>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>,
	linux-btrfs <linux-btrfs@vger.kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Nick Piggin <npiggin@suse.de>,
	Peter Morreale <pmorreale@novell.com>,
	Sven Dietrich <SDietrich@novell.com>,
	Dmitry Adamushko <dmitry.adamushko@gmail.com>,
	Johannes Weiner <hannes@cmpxchg.org>
Subject: Re: [GIT PULL] adaptive spinning mutexes
Date: Wed, 14 Jan 2009 20:23:06 +0100	[thread overview]
Message-ID: <1231960986.14825.45.camel@laptop> (raw)
In-Reply-To: <20090114105300.66bd014d.akpm@linux-foundation.org>

On Wed, 2009-01-14 at 10:53 -0800, Andrew Morton wrote:
> On Wed, 14 Jan 2009 19:33:19 +0100 Ingo Molnar <mingo@elte.hu> wrote:
> 
> > Please pull the adaptive-mutexes-for-linus git tree
> 
> <fear>
> 
> - It seems a major shortcoming that the feature is disabled if
>   CONFIG_DEBUG_MUTEXES=y.  It means that lots of people won't test it.

Yes, that's a bit unfortunate, a simple patch to enable that is:

I must admit I'm a bit stumped on why that debug check triggers, I
couldn't reproduce today, but Ingo ran into it quite quickly :/

Index: linux-2.6/kernel/mutex-debug.c
===================================================================
--- linux-2.6.orig/kernel/mutex-debug.c
+++ linux-2.6/kernel/mutex-debug.c
@@ -74,7 +74,7 @@ void debug_mutex_unlock(struct mutex *lo
 		return;
 
 	DEBUG_LOCKS_WARN_ON(lock->magic != lock);
-	DEBUG_LOCKS_WARN_ON(lock->owner != current_thread_info());
+	/* DEBUG_LOCKS_WARN_ON(lock->owner != current_thread_info()); */
 	DEBUG_LOCKS_WARN_ON(!lock->wait_list.prev && !lock->wait_list.next);
 	mutex_clear_owner(lock);
 }
Index: linux-2.6/kernel/mutex.c
===================================================================
--- linux-2.6.orig/kernel/mutex.c
+++ linux-2.6/kernel/mutex.c
@@ -148,7 +148,7 @@ __mutex_lock_common(struct mutex *lock, 
 
 	preempt_disable();
 	mutex_acquire(&lock->dep_map, subclass, 0, ip);
-#if defined(CONFIG_SMP) && !defined(CONFIG_DEBUG_MUTEXES)
+#ifdef CONFIG_SMP
 	/*
 	 * Optimistic spinning.
 	 *
@@ -162,9 +162,6 @@ __mutex_lock_common(struct mutex *lock, 
 	 * Since this needs the lock owner, and this mutex implementation
 	 * doesn't track the owner atomically in the lock field, we need to
 	 * track it non-atomically.
-	 *
-	 * We can't do this for DEBUG_MUTEXES because that relies on wait_lock
-	 * to serialize everything.
 	 */
 
 	for (;;) {


> - When people hit performance/latency oddities, it would be nice if
>   they had a /proc knob with which they could disable this feature at
>   runtime.
> 
>   This would also be useful for comparative performance testing.

CONFIG_SCHED_DEBUG=y

echo NO_MUTEX_SPIN > /debug/sched_features



  parent reply	other threads:[~2009-01-14 19:23 UTC|newest]

Thread overview: 92+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-01-12 15:37 [PATCH -v8][RFC] mutex: implement adaptive spinning Peter Zijlstra
2009-01-12 16:04 ` Linus Torvalds
2009-01-12 16:20   ` Linus Torvalds
2009-01-12 16:45     ` Chris Mason
2009-01-12 16:50       ` Peter Zijlstra
2009-01-12 17:14         ` Chris Mason
2009-01-12 17:24           ` Peter Zijlstra
2009-01-12 17:30             ` Chris Mason
2009-01-12 17:16     ` Peter Zijlstra
2009-01-12 17:33       ` Boaz Harrosh
2009-01-12 18:07         ` Peter Zijlstra
2009-01-12 16:13 ` Avi Kivity
2009-01-12 17:13   ` Peter Zijlstra
2009-01-12 17:23     ` Avi Kivity
2009-01-12 17:32     ` Avi Kivity
2009-01-14 16:46       ` Peter Zijlstra
2009-01-14 17:04         ` Nick Piggin
2009-01-14 17:23           ` Avi Kivity
2009-01-15  0:50             ` Nick Piggin
2009-01-13 15:15 ` [PATCH -v9][RFC] " Peter Zijlstra
2009-01-13 16:16   ` Linus Torvalds
2009-01-13 16:21     ` Peter Zijlstra
2009-01-13 16:39       ` Ingo Molnar
2009-01-13 16:40       ` Peter Zijlstra
2009-01-13 16:49         ` Linus Torvalds
2009-01-13 17:21           ` Peter Zijlstra
2009-01-13 18:33             ` Ingo Molnar
2009-01-13 18:40               ` Linus Torvalds
2009-01-13 19:01                 ` Ingo Molnar
2009-01-14  2:58             ` Chris Mason
2009-01-14 11:18               ` Dmitry Adamushko
2009-01-14 16:47                 ` Chris Mason
2009-01-14 17:32                   ` Dmitry Adamushko
2009-01-14 11:21               ` Ingo Molnar
2009-01-14 15:43                 ` Linus Torvalds
2009-01-14 16:23                   ` Chris Mason
2009-01-14 17:06                   ` [PATCH -v11 delta] " Ingo Molnar
2009-01-14 17:00             ` [PATCH -v11][RFC] " Peter Zijlstra
2009-01-14 17:18               ` Nick Piggin
2009-01-14 17:22                 ` Peter Zijlstra
2009-01-15  0:46                   ` Nick Piggin
2009-01-15  7:44                     ` Peter Zijlstra
2009-01-15  7:52                       ` Nick Piggin
2009-01-14 18:33               ` [GIT PULL] adaptive spinning mutexes Ingo Molnar
2009-01-14 18:40                 ` Chris Mason
2009-01-15  9:53                   ` Ingo Molnar
2009-01-14 18:47                 ` Ingo Molnar
2009-01-14 19:28                   ` Ingo Molnar
2009-01-15 17:44                     ` Matthew Wilcox
2009-01-15 18:05                       ` Linus Torvalds
2009-01-15 18:08                         ` Ingo Molnar
2009-01-15 18:16                           ` Linus Torvalds
2009-01-15 19:26                             ` Chris Mason
2009-01-15 20:13                               ` Linus Torvalds
2009-01-15 21:04                                 ` Chris Mason
2009-01-15 22:03                                   ` Ingo Molnar
2009-01-16 13:32                               ` Folkert van Heusden
2009-01-16 13:57                                 ` Folkert van Heusden
2009-01-16 18:37                               ` Bill Davidsen
2009-01-16  0:53                             ` Paul E. McKenney
2009-01-16  1:01                               ` Linus Torvalds
2009-01-16  1:34                                 ` Paul E. McKenney
2009-01-16 14:07                                 ` Folkert van Heusden
2009-01-16  3:03                             ` Nick Piggin
2009-01-15 18:06                       ` Ingo Molnar
2009-01-14 18:53                 ` Andrew Morton
2009-01-14 19:00                   ` Ingo Molnar
2009-01-14 19:36                     ` Andrew Morton
2009-01-14 19:50                       ` Peter Zijlstra
2009-01-14 20:21                         ` Andrew Morton
2009-01-14 20:27                         ` Ingo Molnar
2009-01-14 20:44                           ` Andrew Morton
2009-01-14 20:14                       ` Ingo Molnar
2009-01-14 20:30                         ` Andrew Morton
2009-01-14 20:51                           ` Ingo Molnar
2009-01-14 21:06                             ` Andrew Morton
2009-01-14 21:14                               ` Ingo Molnar
2009-01-14 21:35                                 ` Andrew Morton
2009-01-14 23:23                                   ` Ingo Molnar
2009-01-15  0:55                                   ` Nick Piggin
2009-01-14 21:41                                 ` Ingo Molnar
2009-01-14 21:50                                   ` Kay Sievers
2009-01-14 22:34                                     ` Ingo Molnar
2009-01-15 11:45                                       ` Folkert van Heusden
2009-01-15 12:53                                       ` Chris Samuel
2009-01-14 19:23                   ` Peter Zijlstra [this message]
2009-01-14 19:33                     ` Ingo Molnar
2009-01-15  8:41             ` [PATCH] mutex: set owner only once on acquisition Johannes Weiner
2009-01-15  8:56               ` Johannes Weiner
2009-01-13 18:12           ` [PATCH -v9][RFC] mutex: implement adaptive spinning Ingo Molnar
2009-01-13 18:21             ` Linus Torvalds
2009-01-13 18:24               ` 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=1231960986.14825.45.camel@laptop \
    --to=a.p.zijlstra@chello.nl \
    --cc=SDietrich@novell.com \
    --cc=akpm@linux-foundation.org \
    --cc=andi@firstfloor.org \
    --cc=chris.mason@oracle.com \
    --cc=dmitry.adamushko@gmail.com \
    --cc=ghaskins@novell.com \
    --cc=hannes@cmpxchg.org \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=matthew@wil.cx \
    --cc=mingo@elte.hu \
    --cc=npiggin@suse.de \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=pmorreale@novell.com \
    --cc=tglx@linutronix.de \
    --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;
as well as URLs for NNTP newsgroup(s).