public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Paul Jackson <pj@sgi.com>
To: Andrew Morton <akpm@osdl.org>
Cc: Linus Torvalds <torvalds@osdl.org>,
	Simon Derr <Simon.Derr@bull.net>, Paul Jackson <pj@sgi.com>,
	linux-kernel@vger.kernel.org,
	Nikita Danilov <nikita@clusterfs.com>
Subject: [PATCH] cpuset semaphore depth check optimize
Date: Mon, 12 Sep 2005 04:30:30 -0700 (PDT)	[thread overview]
Message-ID: <20050912113030.15934.9433.sendpatchset@jackhammer.engr.sgi.com> (raw)

Optimize the deadlock avoidance check on the global cpuset
semaphore cpuset_sem.  Instead of adding a depth counter to the
task struct of each task, rather just two words are enough, one
to store the depth and the other the current cpuset_sem holder.

Thanks to Nikita Danilov for the idea.

Signed-off-by: Paul Jackson <pj@sgi.com>

Index: linux-2.6.13-mem_exclusive_oom/include/linux/sched.h
===================================================================
--- linux-2.6.13-mem_exclusive_oom.orig/include/linux/sched.h
+++ linux-2.6.13-mem_exclusive_oom/include/linux/sched.h
@@ -765,7 +765,6 @@ struct task_struct {
 	short il_next;
 #endif
 #ifdef CONFIG_CPUSETS
-	short cpuset_sem_nest_depth;
 	struct cpuset *cpuset;
 	nodemask_t mems_allowed;
 	int cpuset_mems_generation;
Index: linux-2.6.13-mem_exclusive_oom/kernel/cpuset.c
===================================================================
--- linux-2.6.13-mem_exclusive_oom.orig/kernel/cpuset.c
+++ linux-2.6.13-mem_exclusive_oom/kernel/cpuset.c
@@ -180,6 +180,8 @@ static struct super_block *cpuset_sb = N
  */
 
 static DECLARE_MUTEX(cpuset_sem);
+static struct task_struct *cpuset_sem_owner;
+static int cpuset_sem_depth;
 
 /*
  * The global cpuset semaphore cpuset_sem can be needed by the
@@ -200,16 +202,19 @@ static DECLARE_MUTEX(cpuset_sem);
 
 static inline void cpuset_down(struct semaphore *psem)
 {
-	if (current->cpuset_sem_nest_depth == 0)
+	if (cpuset_sem_owner != current) {
 		down(psem);
-	current->cpuset_sem_nest_depth++;
+		cpuset_sem_owner = current;
+	}
+	cpuset_sem_depth++;
 }
 
 static inline void cpuset_up(struct semaphore *psem)
 {
-	current->cpuset_sem_nest_depth--;
-	if (current->cpuset_sem_nest_depth == 0)
+	if (--cpuset_sem_depth == 0) {
+		cpuset_sem_owner = NULL;
 		up(psem);
+	}
 }
 
 /*

-- 
                          I won't rest till it's the best ...
                          Programmer, Linux Scalability
                          Paul Jackson <pj@sgi.com> 1.650.933.1373

             reply	other threads:[~2005-09-12 11:30 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-09-12 11:30 Paul Jackson [this message]
2005-09-12 11:39 ` [PATCH] cpuset semaphore depth check optimize Andrew Morton
2005-09-12 14:38   ` Linus Torvalds
2005-09-12 15:32     ` Paul Jackson
     [not found]     ` <17189.39100.862561.865802@gargle.gargle.HOWL>
2005-09-12 15:38       ` Paul Jackson
2005-09-13  7:04     ` Ingo Molnar
2005-09-13 22:21       ` Paul Jackson
2005-09-12 14:51   ` Paul Jackson
2005-09-12 17:07     ` Roman Zippel
     [not found]       ` <20050912153135.3812d8e2.pj@sgi.com>
2005-09-13 11:42         ` Roman Zippel
2005-09-13 17:37           ` Paul Jackson
2005-09-14  6:01             ` Paul Jackson
2005-09-14 15:56             ` Roman Zippel
2005-09-14 19:46               ` Paul Jackson
2005-09-15 10:51                 ` Roman Zippel
2005-09-15 17:45                   ` Paul Jackson
2005-09-15 19:18                     ` Paul Jackson
2005-09-17  2:06                     ` Roman Zippel
2005-09-17  2:27                       ` Roman Zippel
2005-09-20  7:57                     ` Paul Jackson
2005-09-20 12:05                       ` Robin Holt
2005-09-20 12:21                         ` Robin Holt
2005-09-20 13:10                         ` Simon Derr
2005-09-20 14:22                         ` Paul Jackson
2005-09-20 14:54                           ` Robin Holt
2005-09-20 15:07                             ` Linus Torvalds
2005-09-20 15:14                             ` Paul Jackson
2005-09-20 15:17                             ` Simon Derr
2005-09-12 11:48 ` Nikita Danilov
2005-09-12 13:25   ` Paul Jackson

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=20050912113030.15934.9433.sendpatchset@jackhammer.engr.sgi.com \
    --to=pj@sgi.com \
    --cc=Simon.Derr@bull.net \
    --cc=akpm@osdl.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nikita@clusterfs.com \
    --cc=torvalds@osdl.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