From: "Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com>
To: akpm@linux-foundation.org, mgorman@suse.de, dave@sr71.net,
hannes@cmpxchg.org, tony.luck@intel.com,
matthew.garrett@nebula.com, riel@redhat.com,
arjan@linux.intel.com, srinivas.pandruvada@linux.intel.com,
willy@linux.intel.com, kamezawa.hiroyu@jp.fujitsu.com,
lenb@kernel.org, rjw@sisk.pl
Cc: gargankita@gmail.com, paulmck@linux.vnet.ibm.com,
svaidy@linux.vnet.ibm.com, andi@firstfloor.org,
isimatu.yasuaki@jp.fujitsu.com, santosh.shilimkar@ti.com,
kosaki.motohiro@gmail.com, srivatsa.bhat@linux.vnet.ibm.com,
linux-pm@vger.kernel.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org
Subject: [RFC PATCH v4 39/40] mm: Add intelligence in kmempowerd to ignore regions unsuitable for evacuation
Date: Thu, 26 Sep 2013 04:52:42 +0530 [thread overview]
Message-ID: <20130925232240.26184.54998.stgit@srivatsabhat.in.ibm.com> (raw)
In-Reply-To: <20130925231250.26184.31438.stgit@srivatsabhat.in.ibm.com>
Enhance kmempowerd to determine situations where evacuating a region would prove
to be too costly or counter-productive, and ignore those regions for region
evacuation.
For example, if the region has a significant number of used pages (say more than
32), then evacuation will involve more work and might not be justifiable. Also,
compacting region 0 would be pointless, since that is the target of all our
compaction runs. Add these checks in the region-evacuator.
Signed-off-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
---
include/linux/mmzone.h | 2 ++
mm/compaction.c | 25 +++++++++++++++++++++++--
mm/internal.h | 2 ++
3 files changed, 27 insertions(+), 2 deletions(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 257afdf..f383cc8d4 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -84,6 +84,8 @@ static inline int get_pageblock_migratetype(struct page *page)
return get_pageblock_flags_group(page, PB_migrate, PB_migrate_end);
}
+#define MAX_MEMPWR_MIGRATE_PAGES 32
+
struct mem_region_list {
struct list_head *page_block;
unsigned long nr_free;
diff --git a/mm/compaction.c b/mm/compaction.c
index b56be89..41585b0 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -1297,9 +1297,26 @@ void queue_mempower_work(struct pglist_data *pgdat, struct zone *zone,
queue_kthread_work(&pgdat->mempower_worker, &mpwork->work);
}
+int should_evacuate_region(struct zone *z, struct zone_mem_region *region)
+{
+ unsigned long pages_in_use;
+
+ /* Don't try to evacuate region 0, since its the target of migration */
+ if (region == z->zone_regions)
+ return 0;
+
+ pages_in_use = region->present_pages - region->nr_free;
+
+ if (pages_in_use > 0 && pages_in_use <= MAX_MEMPWR_MIGRATE_PAGES)
+ return 1;
+
+ return 0;
+}
+
static void kmempowerd(struct kthread_work *work)
{
struct mempower_work *mpwork;
+ struct zone_mem_region *zmr;
struct zone *zone;
unsigned long flags;
int region_id;
@@ -1315,8 +1332,12 @@ repeat:
if (bitmap_empty(mpwork_mask, nr_zone_region_bits))
return;
- for_each_set_bit(region_id, mpwork_mask, nr_zone_region_bits)
- evacuate_mem_region(zone, &zone->zone_regions[region_id]);
+ for_each_set_bit(region_id, mpwork_mask, nr_zone_region_bits) {
+ zmr = &zone->zone_regions[region_id];
+
+ if (should_evacuate_region(zone, zmr))
+ evacuate_mem_region(zone, zmr);
+ }
spin_lock_irqsave(&mpwork->lock, flags);
diff --git a/mm/internal.h b/mm/internal.h
index 3fbc9f6..5b4658c 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -184,6 +184,8 @@ int compact_range(struct compact_control *cc, struct aggression_control *ac,
void queue_mempower_work(struct pglist_data *pgdat, struct zone *zone,
int region_id);
+int should_evacuate_region(struct zone *z, struct zone_mem_region *region);
+
#endif
/*
--
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:[~2013-09-25 23:26 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-25 23:13 [RFC PATCH v4 00/40] mm: Memory Power Management Srivatsa S. Bhat
2013-09-25 23:13 ` [RFC PATCH v4 01/40] mm: Introduce memory regions data-structure to capture region boundaries within nodes Srivatsa S. Bhat
2013-10-23 9:54 ` Johannes Weiner
2013-10-23 14:38 ` Srivatsa S. Bhat
2013-09-25 23:14 ` [RFC PATCH v4 02/40] mm: Initialize node memory regions during boot Srivatsa S. Bhat
2013-09-25 23:14 ` [RFC PATCH v4 03/40] mm: Introduce and initialize zone memory regions Srivatsa S. Bhat
2013-09-25 23:14 ` [RFC PATCH v4 04/40] mm: Add helpers to retrieve node region and zone region for a given page Srivatsa S. Bhat
2013-09-25 23:14 ` [RFC PATCH v4 05/40] mm: Add data-structures to describe memory regions within the zones' freelists Srivatsa S. Bhat
2013-09-25 23:14 ` [RFC PATCH v4 06/40] mm: Demarcate and maintain pageblocks in region-order in " Srivatsa S. Bhat
2013-09-26 22:16 ` Dave Hansen
2013-09-27 6:34 ` Srivatsa S. Bhat
2013-10-23 10:17 ` Johannes Weiner
2013-10-23 16:09 ` Srivatsa S. Bhat
2013-09-25 23:15 ` [RFC PATCH v4 07/40] mm: Track the freepage migratetype of pages accurately Srivatsa S. Bhat
2013-09-25 23:15 ` [RFC PATCH v4 08/40] mm: Use the correct migratetype during buddy merging Srivatsa S. Bhat
2013-09-25 23:15 ` [RFC PATCH v4 09/40] mm: Add an optimized version of del_from_freelist to keep page allocation fast Srivatsa S. Bhat
2013-09-25 23:15 ` [RFC PATCH v4 10/40] bitops: Document the difference in indexing between fls() and __fls() Srivatsa S. Bhat
2013-09-25 23:16 ` [RFC PATCH v4 11/40] mm: A new optimized O(log n) sorting algo to speed up buddy-sorting Srivatsa S. Bhat
2013-09-25 23:16 ` [RFC PATCH v4 12/40] mm: Add support to accurately track per-memory-region allocation Srivatsa S. Bhat
2013-09-25 23:16 ` [RFC PATCH v4 13/40] mm: Print memory region statistics to understand the buddy allocator behavior Srivatsa S. Bhat
2013-09-25 23:17 ` [RFC PATCH v4 14/40] mm: Enable per-memory-region fragmentation stats in pagetypeinfo Srivatsa S. Bhat
2013-09-25 23:17 ` [RFC PATCH v4 15/40] mm: Add aggressive bias to prefer lower regions during page allocation Srivatsa S. Bhat
2013-09-25 23:17 ` [RFC PATCH v4 16/40] mm: Introduce a "Region Allocator" to manage entire memory regions Srivatsa S. Bhat
2013-10-23 10:10 ` Johannes Weiner
2013-10-23 16:22 ` Srivatsa S. Bhat
2013-09-25 23:17 ` [RFC PATCH v4 17/40] mm: Add a mechanism to add pages to buddy freelists in bulk Srivatsa S. Bhat
2013-09-25 23:18 ` [RFC PATCH v4 18/40] mm: Provide a mechanism to delete pages from " Srivatsa S. Bhat
2013-09-25 23:18 ` [RFC PATCH v4 19/40] mm: Provide a mechanism to release free memory to the region allocator Srivatsa S. Bhat
2013-09-25 23:18 ` [RFC PATCH v4 20/40] mm: Provide a mechanism to request free memory from " Srivatsa S. Bhat
2013-09-25 23:18 ` [RFC PATCH v4 21/40] mm: Maintain the counter for freepages in " Srivatsa S. Bhat
2013-09-25 23:18 ` [RFC PATCH v4 22/40] mm: Propagate the sorted-buddy bias for picking free regions, to " Srivatsa S. Bhat
2013-09-25 23:19 ` [RFC PATCH v4 23/40] mm: Fix vmstat to also account for freepages in the " Srivatsa S. Bhat
2013-09-25 23:19 ` [RFC PATCH v4 24/40] mm: Drop some very expensive sorted-buddy related checks under DEBUG_PAGEALLOC Srivatsa S. Bhat
2013-09-25 23:19 ` [RFC PATCH v4 25/40] mm: Connect Page Allocator(PA) to Region Allocator(RA); add PA => RA flow Srivatsa S. Bhat
2013-09-25 23:19 ` [RFC PATCH v4 26/40] mm: Connect Page Allocator(PA) to Region Allocator(RA); add PA <= " Srivatsa S. Bhat
2013-09-25 23:19 ` [RFC PATCH v4 27/40] mm: Update the freepage migratetype of pages during region allocation Srivatsa S. Bhat
2013-09-25 23:20 ` [RFC PATCH v4 28/40] mm: Provide a mechanism to check if a given page is in the region allocator Srivatsa S. Bhat
2013-09-25 23:20 ` [RFC PATCH v4 29/40] mm: Add a way to request pages of a particular region from " Srivatsa S. Bhat
2013-09-25 23:20 ` [RFC PATCH v4 30/40] mm: Modify move_freepages() to handle pages in the region allocator properly Srivatsa S. Bhat
2013-09-25 23:20 ` [RFC PATCH v4 31/40] mm: Never change migratetypes of pageblocks during freepage stealing Srivatsa S. Bhat
2013-09-25 23:20 ` [RFC PATCH v4 32/40] mm: Set pageblock migratetype when allocating regions from region allocator Srivatsa S. Bhat
2013-09-25 23:21 ` [RFC PATCH v4 33/40] mm: Use a cache between page-allocator and region-allocator Srivatsa S. Bhat
2013-09-25 23:21 ` [RFC PATCH v4 34/40] mm: Restructure the compaction part of CMA for wider use Srivatsa S. Bhat
2013-09-25 23:21 ` [RFC PATCH v4 35/40] mm: Add infrastructure to evacuate memory regions using compaction Srivatsa S. Bhat
2013-09-25 23:21 ` [RFC PATCH v4 36/40] kthread: Split out kthread-worker bits to avoid circular header-file dependency Srivatsa S. Bhat
2013-09-25 23:22 ` [RFC PATCH v4 37/40] mm: Add a kthread to perform targeted compaction for memory power management Srivatsa S. Bhat
2013-09-25 23:22 ` [RFC PATCH v4 38/40] mm: Add a mechanism to queue work to the kmempowerd kthread Srivatsa S. Bhat
2013-09-25 23:22 ` Srivatsa S. Bhat [this message]
2013-09-25 23:22 ` [RFC PATCH v4 40/40] mm: Add triggers in the page-allocator to kick off region evacuation Srivatsa S. Bhat
2013-09-25 23:26 ` [Results] [RFC PATCH v4 00/40] mm: Memory Power Management Srivatsa S. Bhat
2013-09-25 23:40 ` Andrew Morton
2013-09-25 23:47 ` Andi Kleen
2013-09-26 1:14 ` Arjan van de Ven
2013-09-26 13:09 ` Srivatsa S. Bhat
2013-09-26 1:15 ` Arjan van de Ven
2013-09-26 1:21 ` Andrew Morton
2013-09-26 1:50 ` Andi Kleen
2013-09-26 2:59 ` Andrew Morton
2013-09-26 13:42 ` Srivatsa S. Bhat
2013-09-26 15:58 ` Arjan van de Ven
2013-09-26 17:00 ` Srivatsa S. Bhat
2013-09-26 18:06 ` Arjan van de Ven
2013-09-26 18:33 ` Srivatsa S. Bhat
2013-09-26 18:50 ` Luck, Tony
2013-09-26 18:56 ` Srivatsa S. Bhat
2013-09-26 13:37 ` Srivatsa S. Bhat
2013-09-26 15:23 ` Arjan van de Ven
2013-09-26 13:16 ` Srivatsa S. Bhat
2013-09-26 12:58 ` Srivatsa S. Bhat
2013-09-26 15:29 ` Arjan van de Ven
2013-09-26 17:22 ` Luck, Tony
2013-09-26 17:54 ` Srivatsa S. Bhat
2013-09-26 19:38 ` Andi Kleen
2013-11-12 8:02 ` Srivatsa S. Bhat
2013-11-12 17:34 ` Dave Hansen
2013-11-12 18:44 ` Srivatsa S. Bhat
2013-11-12 18:49 ` Srivatsa S. Bhat
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=20130925232240.26184.54998.stgit@srivatsabhat.in.ibm.com \
--to=srivatsa.bhat@linux.vnet.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=andi@firstfloor.org \
--cc=arjan@linux.intel.com \
--cc=dave@sr71.net \
--cc=gargankita@gmail.com \
--cc=hannes@cmpxchg.org \
--cc=isimatu.yasuaki@jp.fujitsu.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=kosaki.motohiro@gmail.com \
--cc=lenb@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-pm@vger.kernel.org \
--cc=matthew.garrett@nebula.com \
--cc=mgorman@suse.de \
--cc=paulmck@linux.vnet.ibm.com \
--cc=riel@redhat.com \
--cc=rjw@sisk.pl \
--cc=santosh.shilimkar@ti.com \
--cc=srinivas.pandruvada@linux.intel.com \
--cc=svaidy@linux.vnet.ibm.com \
--cc=tony.luck@intel.com \
--cc=willy@linux.intel.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;
as well as URLs for NNTP newsgroup(s).