From: Konstantin Khlebnikov <khlebnikov@openvz.org>
To: Hugh Dickins <hughd@google.com>,
linux-kernel@vger.kernel.org, linux-mm@kvack.org,
Johannes Weiner <hannes@cmpxchg.org>,
Andrew Morton <akpm@linux-foundation.org>,
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Andi Kleen <andi@firstfloor.org>
Subject: [PATCH 1/2] mm: configure lruvec split by boot options
Date: Thu, 23 Feb 2012 22:48:24 +0400 [thread overview]
Message-ID: <20120223184824.7184.78353.stgit@zurg> (raw)
In-Reply-To: <20120223162111.GA4713@one.firstfloor.org>
This patch adds boot options:
lruvec_split=%u by default 1, limited by CONFIG_PAGE_LRU_SPLIT
lruvec_interleaving=%u by default CONFIG_PAGE_LRU_INTERLEAVING
Signed-off-by: Konstantin Khlebnikov <khlebnikov@openvz.org>
---
include/linux/mm.h | 5 ++++-
mm/internal.h | 2 +-
mm/page_alloc.c | 29 +++++++++++++++++++++++++++++
3 files changed, 34 insertions(+), 2 deletions(-)
diff --git a/include/linux/mm.h b/include/linux/mm.h
index d14db10..f042a34 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -737,12 +737,15 @@ static inline int page_lruvec_id(struct page *page)
#else /* CONFIG_PAGE_LRU_SPLIT */
+extern unsigned lruvec_split;
+extern unsigned lruvec_interleaving;
+
static inline int page_lruvec_id(struct page *page)
{
unsigned long pfn = page_to_pfn(page);
- return (pfn >> CONFIG_PAGE_LRU_INTERLEAVING) % CONFIG_PAGE_LRU_SPLIT;
+ return (pfn >> lruvec_interleaving) % lruvec_split;
}
#endif /* CONFIG_PAGE_LRU_SPLIT */
diff --git a/mm/internal.h b/mm/internal.h
index f429911..be7415b 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -17,7 +17,7 @@
for ( zone_id = 0 ; zone_id < MAX_NR_ZONES ; zone_id++ )
#define for_each_lruvec_id(lruvec_id) \
- for ( lruvec_id = 0 ; lruvec_id < CONFIG_PAGE_LRU_SPLIT ; lruvec_id++ )
+ for ( lruvec_id = 0 ; lruvec_id < lruvec_split ; lruvec_id++ )
#define for_each_zone_and_lruvec_id(zone_id, lruvec_id) \
for_each_zone_id(zone_id) for_each_lruvec_id(lruvec_id)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 9b0cc92..1a899fa 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -4303,6 +4303,35 @@ void init_zone_lruvec(struct zone *zone, struct lruvec *lruvec)
#endif
}
+#if CONFIG_PAGE_LRU_SPLIT != 1
+
+unsigned lruvec_split = 1;
+unsigned lruvec_interleaving = CONFIG_PAGE_LRU_INTERLEAVING;
+
+static int __init set_lruvec_split(char *arg)
+{
+ if (!kstrtouint(arg, 0, &lruvec_split) &&
+ lruvec_split >= 1 &&
+ lruvec_split <= CONFIG_PAGE_LRU_SPLIT)
+ return 0;
+ lruvec_split = 1;
+ return 1;
+}
+early_param("lruvec_split", set_lruvec_split);
+
+static int __init set_lruvec_interleaving(char *arg)
+{
+ if (!kstrtouint(arg, 0, &lruvec_interleaving) &&
+ lruvec_interleaving >= HPAGE_PMD_ORDER &&
+ lruvec_interleaving <= BITS_PER_LONG)
+ return 0;
+ lruvec_split = 1;
+ return 1;
+}
+early_param("lruvec_interleaving", set_lruvec_interleaving);
+
+#endif
+
/*
* Set up the zone data structures:
* - mark all pages reserved
--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2012-02-23 18:48 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-23 13:51 [PATCH v3 00/21] mm: lru_lock splitting Konstantin Khlebnikov
2012-02-23 13:51 ` [PATCH v3 01/21] memcg: unify inactive_ratio calculation Konstantin Khlebnikov
2012-02-28 0:05 ` KAMEZAWA Hiroyuki
2012-02-23 13:51 ` [PATCH v3 02/21] memcg: make mm_match_cgroup() hirarchical Konstantin Khlebnikov
2012-02-23 18:03 ` Johannes Weiner
2012-02-23 19:46 ` Konstantin Khlebnikov
2012-02-23 22:06 ` Johannes Weiner
2012-02-28 0:11 ` KAMEZAWA Hiroyuki
2012-02-28 6:31 ` Konstantin Khlebnikov
2012-02-23 13:51 ` [PATCH v3 03/21] memcg: fix page_referencies cgroup filter on global reclaim Konstantin Khlebnikov
2012-02-28 0:13 ` KAMEZAWA Hiroyuki
2012-02-23 13:51 ` [PATCH v3 04/21] memcg: use vm_swappiness from target memory cgroup Konstantin Khlebnikov
2012-02-28 0:15 ` KAMEZAWA Hiroyuki
2012-02-23 13:52 ` [PATCH v3 05/21] mm: rename lruvec->lists into lruvec->pages_lru Konstantin Khlebnikov
2012-02-28 0:20 ` KAMEZAWA Hiroyuki
2012-02-28 6:04 ` Konstantin Khlebnikov
2012-02-23 13:52 ` [PATCH v3 06/21] mm: lruvec linking functions Konstantin Khlebnikov
2012-02-28 0:27 ` KAMEZAWA Hiroyuki
2012-02-28 6:09 ` Konstantin Khlebnikov
2012-02-23 13:52 ` [PATCH v3 07/21] mm: add lruvec->pages_count Konstantin Khlebnikov
2012-02-28 0:35 ` KAMEZAWA Hiroyuki
2012-02-28 6:16 ` Konstantin Khlebnikov
2012-02-23 13:52 ` [PATCH v3 08/21] mm: unify inactive_list_is_low() Konstantin Khlebnikov
2012-02-28 0:36 ` KAMEZAWA Hiroyuki
2012-02-23 13:52 ` [PATCH v3 09/21] mm: add lruvec->reclaim_stat Konstantin Khlebnikov
2012-02-28 0:38 ` KAMEZAWA Hiroyuki
2012-02-23 13:52 ` [PATCH v3 10/21] mm: kill struct mem_cgroup_zone Konstantin Khlebnikov
2012-02-28 0:41 ` KAMEZAWA Hiroyuki
2012-02-23 13:52 ` [PATCH v3 11/21] mm: move page-to-lruvec translation upper Konstantin Khlebnikov
2012-02-28 0:42 ` KAMEZAWA Hiroyuki
2012-02-23 13:52 ` [PATCH v3 12/21] mm: push lruvec into update_page_reclaim_stat() Konstantin Khlebnikov
2012-02-28 0:44 ` KAMEZAWA Hiroyuki
2012-02-23 13:52 ` [PATCH v3 13/21] mm: push lruvecs from pagevec_lru_move_fn() to iterator Konstantin Khlebnikov
2012-02-28 0:45 ` KAMEZAWA Hiroyuki
2012-02-23 13:52 ` [PATCH v3 14/21] mm: introduce lruvec locking primitives Konstantin Khlebnikov
2012-02-28 0:56 ` KAMEZAWA Hiroyuki
2012-02-28 6:23 ` Konstantin Khlebnikov
2012-02-23 13:52 ` [PATCH v3 15/21] mm: handle lruvec relocks on lumpy reclaim Konstantin Khlebnikov
2012-02-28 1:01 ` KAMEZAWA Hiroyuki
2012-02-28 6:25 ` Konstantin Khlebnikov
2012-02-23 13:52 ` [PATCH v3 16/21] mm: handle lruvec relocks in compaction Konstantin Khlebnikov
2012-02-28 1:13 ` KAMEZAWA Hiroyuki
2012-02-28 6:31 ` Konstantin Khlebnikov
2012-02-23 13:53 ` [PATCH v3 17/21] mm: handle lruvec relock in memory controller Konstantin Khlebnikov
2012-02-28 1:22 ` KAMEZAWA Hiroyuki
2012-02-23 13:53 ` [PATCH v3 18/21] mm: add to lruvec isolated pages counters Konstantin Khlebnikov
2012-02-24 5:32 ` Konstantin Khlebnikov
2012-02-28 1:38 ` KAMEZAWA Hiroyuki
2012-02-23 13:53 ` [PATCH v3 19/21] memcg: check lru vectors emptiness in pre-destroy Konstantin Khlebnikov
2012-02-28 1:43 ` KAMEZAWA Hiroyuki
2012-02-23 13:53 ` [PATCH v3 20/21] mm: split zone->lru_lock Konstantin Khlebnikov
2012-02-28 1:49 ` KAMEZAWA Hiroyuki
2012-02-28 6:39 ` Konstantin Khlebnikov
2012-02-23 13:53 ` [PATCH v3 21/21] mm: zone lru vectors interleaving Konstantin Khlebnikov
2012-02-23 14:44 ` Hillf Danton
2012-02-23 16:21 ` Andi Kleen
2012-02-23 18:48 ` Konstantin Khlebnikov [this message]
2012-02-23 18:48 ` [PATCH 2/2] mm: show zone lruvec state in /proc/zoneinfo Konstantin Khlebnikov
2012-02-25 0:05 ` [PATCH v3 00/21] mm: lru_lock splitting Tim Chen
2012-02-25 5:34 ` Konstantin Khlebnikov
2012-02-25 2:15 ` KAMEZAWA Hiroyuki
2012-02-25 5:31 ` Konstantin Khlebnikov
2012-02-26 23:54 ` KAMEZAWA Hiroyuki
2012-02-28 1:52 ` KAMEZAWA Hiroyuki
2012-02-28 6:49 ` Konstantin Khlebnikov
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=20120223184824.7184.78353.stgit@zurg \
--to=khlebnikov@openvz.org \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=hannes@cmpxchg.org \
--cc=hughd@google.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.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).