From: Christoph Lameter <clameter@sgi.com>
From: Christoph Lameter <clameter@sgi.com>
To: linux-kernel@vger.kernel.org
Cc: akpm@osdl.org, Hugh Dickins <hugh@veritas.com>,
Con Kolivas <kernel@kolivas.org>,
Marcelo Tosatti <marcelo@kvack.org>,
Nick Piggin <nickpiggin@yahoo.com.au>,
linux-mm@kvack.org, Andi Kleen <ak@suse.de>,
Dave Chinner <dgc@sgi.com>, Christoph Lameter <clameter@sgi.com>
Subject: [PATCH 09/21] zone_reclaim: remove /proc/sys/vm/zone_reclaim_interval
Subject: zoned vm counters: use per zone counters to remove zone_reclaim_interval
Date: Mon, 12 Jun 2006 14:13:31 -0700 (PDT) [thread overview]
Message-ID: <20060612211331.20862.69229.sendpatchset@schroedinger.engr.sgi.com> (raw)
In-Reply-To: <20060612211244.20862.41106.sendpatchset@schroedinger.engr.sgi.com>
The zone_reclaim_interval was necessary because we were not able to determine
how many unmapped pages exist in a zone. Therefore we had to scan in
intervals to figure out if any pages were unmapped.
With the zoned counters and NR_ANON we now know the number of pagecache pages
and the number of mapped pages in a zone. So we can simply skip the reclaim
if there is an insufficient number of unmapped pages. We use SWAP_CLUSTER_MAX
as the boundary.
Drop all support for zone_reclaim_interval.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Index: linux-2.6.17-rc6-cl/include/linux/mmzone.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/mmzone.h 2006-06-12 13:33:11.045379511 -0700
+++ linux-2.6.17-rc6-cl/include/linux/mmzone.h 2006-06-12 13:37:33.840632482 -0700
@@ -183,12 +183,6 @@ struct zone {
/* Zone statistics */
atomic_long_t vm_stat[NR_STAT_ITEMS];
- /*
- * timestamp (in jiffies) of the last zone reclaim that did not
- * result in freeing of pages. This is used to avoid repeated scans
- * if all memory in the zone is in use.
- */
- unsigned long last_unsuccessful_zone_reclaim;
/*
* prev_priority holds the scanning priority for this zone. It is
Index: linux-2.6.17-rc6-cl/include/linux/swap.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/swap.h 2006-06-12 12:42:51.046775908 -0700
+++ linux-2.6.17-rc6-cl/include/linux/swap.h 2006-06-12 13:33:11.120570168 -0700
@@ -190,7 +190,6 @@ extern long vm_total_pages;
#ifdef CONFIG_NUMA
extern int zone_reclaim_mode;
-extern int zone_reclaim_interval;
extern int zone_reclaim(struct zone *, gfp_t, unsigned int);
#else
#define zone_reclaim_mode 0
Index: linux-2.6.17-rc6-cl/include/linux/sysctl.h
===================================================================
--- linux-2.6.17-rc6-cl.orig/include/linux/sysctl.h 2006-06-12 12:42:51.074117966 -0700
+++ linux-2.6.17-rc6-cl/include/linux/sysctl.h 2006-06-12 13:33:11.121546670 -0700
@@ -191,7 +191,6 @@ enum
VM_DROP_PAGECACHE=29, /* int: nuke lots of pagecache */
VM_PERCPU_PAGELIST_FRACTION=30,/* int: fraction of pages in each percpu_pagelist */
VM_ZONE_RECLAIM_MODE=31, /* reclaim local zone memory before going off node */
- VM_ZONE_RECLAIM_INTERVAL=32, /* time period to wait after reclaim failure */
VM_PANIC_ON_OOM=33, /* panic at out-of-memory */
VM_VDSO_ENABLED=34, /* map VDSO into new processes? */
VM_SWAP_PREFETCH=35, /* swap prefetch */
Index: linux-2.6.17-rc6-cl/kernel/sysctl.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/kernel/sysctl.c 2006-06-12 12:42:51.948087320 -0700
+++ linux-2.6.17-rc6-cl/kernel/sysctl.c 2006-06-12 13:33:11.123499674 -0700
@@ -1015,15 +1015,6 @@ static ctl_table vm_table[] = {
.strategy = &sysctl_intvec,
.extra1 = &zero,
},
- {
- .ctl_name = VM_ZONE_RECLAIM_INTERVAL,
- .procname = "zone_reclaim_interval",
- .data = &zone_reclaim_interval,
- .maxlen = sizeof(zone_reclaim_interval),
- .mode = 0644,
- .proc_handler = &proc_dointvec_jiffies,
- .strategy = &sysctl_jiffies,
- },
#endif
#ifdef CONFIG_X86_32
{
Index: linux-2.6.17-rc6-cl/mm/vmscan.c
===================================================================
--- linux-2.6.17-rc6-cl.orig/mm/vmscan.c 2006-06-12 13:33:11.047332515 -0700
+++ linux-2.6.17-rc6-cl/mm/vmscan.c 2006-06-12 13:33:11.124476176 -0700
@@ -1530,11 +1530,6 @@ int zone_reclaim_mode __read_mostly;
#define RECLAIM_SLAB (1<<3) /* Do a global slab shrink if the zone is out of memory */
/*
- * Mininum time between zone reclaim scans
- */
-int zone_reclaim_interval __read_mostly = 30*HZ;
-
-/*
* Priority for ZONE_RECLAIM. This determines the fraction of pages
* of a node considered for each zone_reclaim. 4 scans 1/16th of
* a zone.
@@ -1599,16 +1594,6 @@ static int __zone_reclaim(struct zone *z
p->reclaim_state = NULL;
current->flags &= ~(PF_MEMALLOC | PF_SWAPWRITE);
-
- if (nr_reclaimed == 0) {
- /*
- * We were unable to reclaim enough pages to stay on node. We
- * now allow off node accesses for a certain time period before
- * trying again to reclaim pages from the local zone.
- */
- zone->last_unsuccessful_zone_reclaim = jiffies;
- }
-
return nr_reclaimed >= nr_pages;
}
@@ -1618,13 +1603,17 @@ int zone_reclaim(struct zone *zone, gfp_
int node_id;
/*
- * Do not reclaim if there was a recent unsuccessful attempt at zone
- * reclaim. In that case we let allocations go off node for the
- * zone_reclaim_interval. Otherwise we would scan for each off-node
- * page allocation.
+ * Do not reclaim if there are not enough reclaimable pages in this
+ * zone that would satify this allocations.
+ *
+ * All unmapped pagecache pages are reclaimable.
+ *
+ * Both counters may be temporarily off a bit so we use
+ * SWAP_CLUSTER_MAX as the boundary. It may also be good to
+ * leave a few frequently used unmapped pagecache pages around.
*/
- if (time_before(jiffies,
- zone->last_unsuccessful_zone_reclaim + zone_reclaim_interval))
+ if (zone_page_state(zone, NR_PAGECACHE) -
+ zone_page_state(zone, NR_MAPPED) < SWAP_CLUSTER_MAX)
return 0;
/*
Index: linux-2.6.17-rc6-cl/Documentation/sysctl/vm.txt
===================================================================
--- linux-2.6.17-rc6-cl.orig/Documentation/sysctl/vm.txt 2006-06-12 12:42:42.519959822 -0700
+++ linux-2.6.17-rc6-cl/Documentation/sysctl/vm.txt 2006-06-12 13:36:53.817719845 -0700
@@ -28,7 +28,6 @@ Currently, these files are in /proc/sys/
- block_dump
- drop-caches
- zone_reclaim_mode
-- zone_reclaim_interval
- panic_on_oom
- swap_prefetch
- readahead_ratio
@@ -172,18 +171,6 @@ in all nodes of the system.
================================================================
-zone_reclaim_interval:
-
-The time allowed for off node allocations after zone reclaim
-has failed to reclaim enough pages to allow a local allocation.
-
-Time is set in seconds and set by default to 30 seconds.
-
-Reduce the interval if undesired off node allocations occur. However, too
-frequent scans will have a negative impact onoff node allocation performance.
-
-=============================================================
-
panic_on_oom
This enables or disables panic on out-of-memory feature. If this is set to 1,
--
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:[~2006-06-12 21:13 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-06-12 21:12 [PATCH 00/21] Zoned VM counters V3 Christoph Lameter
2006-06-12 21:12 ` [PATCH 01/21] Create vmstat.c/.h from page_alloc.c/.h Christoph Lameter
2006-06-12 21:12 ` [PATCH 02/21] Basic ZVC (zoned vm counter) implementation, zoned vm counters: per zone counter functionality Christoph Lameter, Christoph Lameter
2006-06-13 5:37 ` Nick Piggin
2006-06-13 15:55 ` Christoph Lameter
2006-06-14 1:21 ` Nick Piggin
2006-06-14 5:37 ` Hugh Dickins
2006-06-14 5:53 ` Andi Kleen
2006-06-14 16:14 ` Christoph Lameter
2006-06-12 21:13 ` [PATCH 03/21] Convert nr_mapped to per zone counter, zoned vm counters: conversion of nr_mapped to per zone counter Christoph Lameter, Christoph Lameter
2006-06-12 21:13 ` [PATCH 04/21] swap_prefetch: Convert nr_mapped to ZVC Christoph Lameter
2006-06-12 23:36 ` Con Kolivas
2006-06-12 21:13 ` [PATCH 05/21] Conversion of nr_pagecache to per zone counter, zoned vm counters: conversion of nr_pagecache to per zone counter Christoph Lameter, Christoph Lameter
2006-06-12 21:13 ` [PATCH 06/21] Remove nr_mapped from scan controls structure, zoned VM stats: Remove nr_mapped from scan control Christoph Lameter, Christoph Lameter
2006-06-12 21:13 ` [PATCH 07/21] Split NR_ANON off from NR_MAPPED, zoned VM stats: Add NR_ANON Christoph Lameter, Christoph Lameter
2006-06-12 21:13 ` [PATCH 08/21] swap_prefetch: Split NR_ANON off NR_MAPPED Christoph Lameter
2006-06-12 23:36 ` Con Kolivas
2006-06-12 21:13 ` Christoph Lameter, Christoph Lameter [this message]
2006-06-12 21:13 ` [PATCH 10/21] Conversion of nr_slab to per zone counter, zoned vm counters: conversion of nr_slab to per zone counter Christoph Lameter, Christoph Lameter
2006-06-12 21:13 ` [PATCH 11/21] swap_prefetch: Conversion of nr_slab to ZVC Christoph Lameter
2006-06-12 21:13 ` [PATCH 12/21] Conversion of nr_pagetables to per zone counter, zoned vm counters: conversion of nr_pagetable to per zone counter Christoph Lameter, Christoph Lameter
2006-06-12 21:13 ` [PATCH 13/21] Conversion of nr_dirty to per zone counter, zoned vm counters: conversion of nr_dirty " Christoph Lameter, Christoph Lameter
2006-06-12 21:13 ` [PATCH 14/21] swap_prefetch: Conversion of nr_dirty to ZVC Christoph Lameter
2006-06-12 23:38 ` Con Kolivas
2006-06-12 21:14 ` [PATCH 15/21] reiser4: Conversiion of nr_dirty to ZVC, reiser4: conversion of nr_dirty to per zone counter Christoph Lameter, Christoph Lameter
2006-06-12 21:14 ` [PATCH 16/21] Conversion of nr_writeback to per zone counter, zoned vm counters: conversion of nr_writeback " Christoph Lameter, Christoph Lameter
2006-06-12 21:14 ` [PATCH 17/21] swap_prefetch: Conversion of nr_writeback to ZVC, swap_prefetch: " Christoph Lameter, Christoph Lameter
2006-06-12 23:38 ` [PATCH 17/21] swap_prefetch: Conversion of nr_writeback to ZVC Con Kolivas
2006-06-12 21:14 ` [PATCH 18/21] Conversion of nr_unstable to per zone counter, zoned vm counters: conversion of nr_unstable to per zone counter Christoph Lameter, Christoph Lameter
2006-06-12 21:14 ` [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC, swap_prefetch: " Christoph Lameter, Christoph Lameter
2006-06-12 23:40 ` [PATCH 19/21] swap_prefetch: Conversion of nr_unstable to ZVC Con Kolivas
2006-06-12 23:48 ` Christoph Lameter
2006-06-12 23:57 ` Con Kolivas
2006-06-12 23:59 ` Con Kolivas
2006-06-13 0:08 ` Christoph Lameter
2006-06-13 0:19 ` Con Kolivas
2006-06-12 21:14 ` [PATCH 20/21] Conversion of nr_bounce to per zone counter, zoned vm counters: conversion of nr_bounce to per zone counter Christoph Lameter, Christoph Lameter
2006-06-12 21:14 ` [PATCH 21/21] Remove useless struct wbs, zoned vm counters: remove useless writeback structure Christoph Lameter, Christoph Lameter
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=20060612211331.20862.69229.sendpatchset@schroedinger.engr.sgi.com \
--to=clameter@sgi.com \
--cc=ak@suse.de \
--cc=akpm@osdl.org \
--cc=dgc@sgi.com \
--cc=hugh@veritas.com \
--cc=kernel@kolivas.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=marcelo@kvack.org \
--cc=nickpiggin@yahoo.com.au \
/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