From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
To: Marcelo Tosatti <marcelo@kvack.org>
Cc: kosaki.motohiro@jp.fujitsu.com, linux-mm@kvack.org,
"Daniel Sp蚣g" <daniel.spang@gmail.com>,
"Rik van Riel" <riel@redhat.com>,
"Andrew Morton" <akpm@linux-foundation.org>
Subject: [RFC][patch] mem_notify more faster reduce load average
Date: Thu, 27 Dec 2007 13:49:09 +0900 [thread overview]
Message-ID: <20071227134854.7F0E.KOSAKI.MOTOHIRO@jp.fujitsu.com> (raw)
In-Reply-To: <20071225192625.D273.KOSAKI.MOTOHIRO@jp.fujitsu.com>
Hi, Marcelo-san
this patch is a bit improvement against my mem notifications large system patch.
original my patch is too slower reduce load average at after free memory increased.
this patch fixed it.
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Index: linux-2.6.23-mem_notify_v3/mm/mem_notify.c
===================================================================
--- linux-2.6.23-mem_notify_v3.orig/mm/mem_notify.c
+++ linux-2.6.23-mem_notify_v3/mm/mem_notify.c
@@ -20,7 +20,8 @@ struct mem_notify_file_info {
long last_event;
};
-atomic_t mem_notify_event = ATOMIC_INIT(0);
+static atomic_t mem_notify_event = ATOMIC_INIT(0);
+static atomic_t mem_notify_event_end = ATOMIC_INIT(0);
static DECLARE_WAIT_QUEUE_HEAD(mem_wait);
static atomic_long_t last_mem_notify = ATOMIC_LONG_INIT(INITIAL_JIFFIES);
@@ -76,13 +77,18 @@ static unsigned int mem_notify_poll(stru
unsigned long bonus;
unsigned long now;
unsigned long last;
+ unsigned long event;
poll_wait(file, &mem_wait, wait);
- if (file_info->last_event == atomic_read(&mem_notify_event))
+retry:
+ event = atomic_read(&mem_notify_event);
+ if (event == file_info->last_event)
goto out;
-retry:
+ if (event == atomic_read(&mem_notify_event_end))
+ goto out;
+
/* Ugly trick:
when too many task wakeup,
control function exit rate for prevent too much freed.
@@ -114,6 +120,8 @@ retry:
if (pages_free < (pages_high+pages_reserve)*2)
val = POLLIN;
+ else
+ atomic_set(&mem_notify_event_end, event);
out:
- kosaki
--
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>
next prev parent reply other threads:[~2007-12-27 4:49 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-12-24 20:32 [PATCH] mem notifications v3 Marcelo Tosatti
2007-12-25 3:47 ` KOSAKI Motohiro
2007-12-25 4:56 ` [RFC] add poll_wait_exclusive() API KOSAKI Motohiro
2007-12-27 21:05 ` Marcelo Tosatti
2007-12-25 8:31 ` [PATCH] mem notifications v3 KOSAKI Motohiro
2007-12-25 10:31 ` [RFC][patch 1/2] mem notifications v3 improvement for large system KOSAKI Motohiro
2007-12-27 21:04 ` Marcelo Tosatti
2007-12-28 0:38 ` KOSAKI Motohiro
2007-12-25 10:31 ` [RFC][patch 2/2] " KOSAKI Motohiro
2007-12-25 10:41 ` KOSAKI Motohiro
2007-12-27 4:49 ` KOSAKI Motohiro [this message]
2007-12-27 20:13 ` [PATCH] mem notifications v3 Marcelo Tosatti
2007-12-28 1:44 ` KOSAKI Motohiro
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=20071227134854.7F0E.KOSAKI.MOTOHIRO@jp.fujitsu.com \
--to=kosaki.motohiro@jp.fujitsu.com \
--cc=akpm@linux-foundation.org \
--cc=daniel.spang@gmail.com \
--cc=linux-mm@kvack.org \
--cc=marcelo@kvack.org \
--cc=riel@redhat.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