All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: linux-mm@kvack.org, linux-kernel@vger.kernel.org
Cc: miklos@szeredi.hu, akpm@linux-foundation.org, neilb@suse.de,
	dgc@sgi.com, tomoki.sekiyama.qu@hitachi.com,
	a.p.zijlstra@chello.nl, nikita@clusterfs.com,
	trond.myklebust@fys.uio.no, yingchao.zhou@gmail.com
Subject: [PATCH 03/10] lib: dampen the percpu_counter FBC_BATCH
Date: Fri, 20 Apr 2007 17:51:57 +0200	[thread overview]
Message-ID: <20070420155502.679143273@chello.nl> (raw)
In-Reply-To: 20070420155154.898600123@chello.nl

[-- Attachment #1: percpu_counter_batch.patch --]
[-- Type: text/plain, Size: 949 bytes --]

With the current logic the percpu_counter's accuracy delta is quadric
wrt the number of cpus in the system, reduce this to O(n ln n).

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 include/linux/percpu_counter.h |    7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

Index: linux-2.6-mm/include/linux/percpu_counter.h
===================================================================
--- linux-2.6-mm.orig/include/linux/percpu_counter.h
+++ linux-2.6-mm/include/linux/percpu_counter.h
@@ -11,6 +11,7 @@
 #include <linux/threads.h>
 #include <linux/percpu.h>
 #include <linux/types.h>
+#include <linux/log2.h>
 
 #ifdef CONFIG_SMP
 
@@ -20,11 +21,7 @@ struct percpu_counter {
 	s32 *counters;
 };
 
-#if NR_CPUS >= 16
-#define FBC_BATCH	(NR_CPUS*2)
-#else
-#define FBC_BATCH	(NR_CPUS*4)
-#endif
+#define FBC_BATCH	(8*ilog2(NR_CPUS))
 
 static inline void percpu_counter_init(struct percpu_counter *fbc, s64 amount)
 {

-- 


WARNING: multiple messages have this Message-ID (diff)
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: linux-mm@kvack.org, linux-kernel@vger.kernel.org
Cc: miklos@szeredi.hu, akpm@linux-foundation.org, neilb@suse.de,
	dgc@sgi.com, tomoki.sekiyama.qu@hitachi.com,
	a.p.zijlstra@chello.nl, nikita@clusterfs.com,
	trond.myklebust@fys.uio.no, yingchao.zhou@gmail.com
Subject: [PATCH 03/10] lib: dampen the percpu_counter FBC_BATCH
Date: Fri, 20 Apr 2007 17:51:57 +0200	[thread overview]
Message-ID: <20070420155502.679143273@chello.nl> (raw)
In-Reply-To: 20070420155154.898600123@chello.nl

[-- Attachment #1: percpu_counter_batch.patch --]
[-- Type: text/plain, Size: 1174 bytes --]

With the current logic the percpu_counter's accuracy delta is quadric
wrt the number of cpus in the system, reduce this to O(n ln n).

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 include/linux/percpu_counter.h |    7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)

Index: linux-2.6-mm/include/linux/percpu_counter.h
===================================================================
--- linux-2.6-mm.orig/include/linux/percpu_counter.h
+++ linux-2.6-mm/include/linux/percpu_counter.h
@@ -11,6 +11,7 @@
 #include <linux/threads.h>
 #include <linux/percpu.h>
 #include <linux/types.h>
+#include <linux/log2.h>
 
 #ifdef CONFIG_SMP
 
@@ -20,11 +21,7 @@ struct percpu_counter {
 	s32 *counters;
 };
 
-#if NR_CPUS >= 16
-#define FBC_BATCH	(NR_CPUS*2)
-#else
-#define FBC_BATCH	(NR_CPUS*4)
-#endif
+#define FBC_BATCH	(8*ilog2(NR_CPUS))
 
 static inline void percpu_counter_init(struct percpu_counter *fbc, s64 amount)
 {

-- 

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2007-04-20 16:01 UTC|newest]

Thread overview: 110+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-04-20 15:51 [PATCH 00/10] per device dirty throttling -v5 Peter Zijlstra
2007-04-20 15:51 ` Peter Zijlstra
2007-04-20 15:51 ` [PATCH 01/10] revert per-backing_dev-dirty-and-writeback-page-accounting Peter Zijlstra
2007-04-20 15:51   ` Peter Zijlstra
2007-04-20 15:51 ` [PATCH 02/10] nfs: remove congestion_end() Peter Zijlstra
2007-04-20 15:51   ` Peter Zijlstra
2007-04-20 15:51 ` Peter Zijlstra [this message]
2007-04-20 15:51   ` [PATCH 03/10] lib: dampen the percpu_counter FBC_BATCH Peter Zijlstra
2007-04-21  9:55   ` Andrew Morton
2007-04-21  9:55     ` Andrew Morton
2007-04-21 10:58     ` Peter Zijlstra
2007-04-21 10:58       ` Peter Zijlstra
2007-04-20 15:51 ` [PATCH 04/10] lib: percpu_counter_mod64 Peter Zijlstra
2007-04-20 15:51   ` Peter Zijlstra
2007-04-21  9:55   ` Andrew Morton
2007-04-21  9:55     ` Andrew Morton
2007-04-21 11:02     ` Peter Zijlstra
2007-04-21 11:02       ` Peter Zijlstra
2007-04-21 19:21       ` Andrew Morton
2007-04-21 19:21         ` Andrew Morton
2007-04-21 19:30         ` Peter Zijlstra
2007-04-21 19:30           ` Peter Zijlstra
2007-04-20 15:51 ` [PATCH 05/10] mm: bdi init hooks Peter Zijlstra
2007-04-20 15:51   ` Peter Zijlstra
2007-04-20 15:52 ` [PATCH 06/10] mm: scalable bdi statistics counters Peter Zijlstra
2007-04-20 15:52   ` Peter Zijlstra
2007-04-20 15:52 ` [PATCH 07/10] mm: count reclaimable pages per BDI Peter Zijlstra
2007-04-20 15:52   ` Peter Zijlstra
2007-04-21  9:55   ` Andrew Morton
2007-04-21  9:55     ` Andrew Morton
2007-04-21 11:04     ` Peter Zijlstra
2007-04-21 11:04       ` Peter Zijlstra
2007-04-20 15:52 ` [PATCH 08/10] mm: count writeback " Peter Zijlstra
2007-04-20 15:52   ` Peter Zijlstra
2007-04-21  9:55   ` Andrew Morton
2007-04-21  9:55     ` Andrew Morton
2007-04-21 11:07     ` Peter Zijlstra
2007-04-21 11:07       ` Peter Zijlstra
2007-04-22  7:19       ` Andrew Morton
2007-04-22  7:19         ` Andrew Morton
2007-04-22  9:08         ` Peter Zijlstra
2007-04-22  9:08           ` Peter Zijlstra
2007-04-20 15:52 ` [PATCH 09/10] mm: expose BDI statistics in sysfs Peter Zijlstra
2007-04-20 15:52   ` Peter Zijlstra
2007-04-21  9:55   ` Andrew Morton
2007-04-21  9:55     ` Andrew Morton
2007-04-21 11:08     ` Peter Zijlstra
2007-04-21 11:08       ` Peter Zijlstra
2007-04-20 15:52 ` [PATCH 10/10] mm: per device dirty threshold Peter Zijlstra
2007-04-20 15:52   ` Peter Zijlstra
2007-04-21  9:55   ` Andrew Morton
2007-04-21  9:55     ` Andrew Morton
2007-04-21 10:38     ` Miklos Szeredi
2007-04-21 10:38       ` Miklos Szeredi
2007-04-21 10:54       ` Andrew Morton
2007-04-21 10:54         ` Andrew Morton
2007-04-21 20:25         ` Miklos Szeredi
2007-04-21 20:25           ` Miklos Szeredi
2007-04-23  6:14           ` Peter Zijlstra
2007-04-23  6:14             ` Peter Zijlstra
2007-04-23  6:29             ` Miklos Szeredi
2007-04-23  6:29               ` Miklos Szeredi
2007-04-23  6:39               ` Andrew Morton
2007-04-23  6:39                 ` Andrew Morton
2007-04-21 12:01     ` Peter Zijlstra
2007-04-21 12:01       ` Peter Zijlstra
2007-04-21 12:15       ` Peter Zijlstra
2007-04-21 12:15         ` Peter Zijlstra
2007-04-21 19:50         ` Peter Zijlstra
2007-04-21 19:50           ` Peter Zijlstra
2007-04-23 15:48         ` Christoph Lameter
2007-04-23 15:48           ` Christoph Lameter
2007-04-23 15:58           ` Peter Zijlstra
2007-04-23 15:58             ` Peter Zijlstra
2007-04-23 16:08             ` Christoph Lameter
2007-04-23 16:08               ` Christoph Lameter
2007-04-22  7:26       ` Andrew Morton
2007-04-22  7:26         ` Andrew Morton
2007-04-24  2:58   ` Neil Brown
2007-04-24  2:58     ` Neil Brown
2007-04-24  7:09     ` Peter Zijlstra
2007-04-24  7:09       ` Peter Zijlstra
2007-04-24  8:19       ` Miklos Szeredi
2007-04-24  8:19         ` Miklos Szeredi
2007-04-24  8:31         ` Peter Zijlstra
2007-04-24  8:31           ` Peter Zijlstra
2007-04-24  9:14           ` Miklos Szeredi
2007-04-24  9:14             ` Miklos Szeredi
2007-04-24  9:26             ` Peter Zijlstra
2007-04-24  9:26               ` Peter Zijlstra
2007-04-24  9:47               ` Miklos Szeredi
2007-04-24  9:47                 ` Miklos Szeredi
2007-04-24 10:00                 ` Andrew Morton
2007-04-24 10:00                   ` Andrew Morton
2007-04-24 10:12                   ` Peter Zijlstra
2007-04-24 10:12                     ` Peter Zijlstra
2007-04-24 10:19                     ` Miklos Szeredi
2007-04-24 10:19                       ` Miklos Szeredi
2007-04-24 10:24                       ` Peter Zijlstra
2007-04-24 10:24                         ` Peter Zijlstra
2007-04-24 10:40                     ` Andrew Morton
2007-04-24 10:40                       ` Andrew Morton
2007-04-24 11:22                       ` Miklos Szeredi
2007-04-24 11:22                         ` Miklos Szeredi
2007-04-24 11:50                         ` Andrew Morton
2007-04-24 11:50                           ` Andrew Morton
2007-04-24 12:07                           ` Miklos Szeredi
2007-04-24 12:07                             ` Miklos Szeredi
2007-04-22  9:57 ` [PATCH 00/10] per device dirty throttling -v5 Andrew Morton
2007-04-22  9:57   ` Andrew Morton

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=20070420155502.679143273@chello.nl \
    --to=a.p.zijlstra@chello.nl \
    --cc=akpm@linux-foundation.org \
    --cc=dgc@sgi.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=miklos@szeredi.hu \
    --cc=neilb@suse.de \
    --cc=nikita@clusterfs.com \
    --cc=tomoki.sekiyama.qu@hitachi.com \
    --cc=trond.myklebust@fys.uio.no \
    --cc=yingchao.zhou@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.