From: Mel Gorman <mgorman@suse.de>
To: Linux-MM <linux-mm@kvack.org>
Cc: Rik van Riel <riel@redhat.com>,
Johannes Weiner <hannes@cmpxchg.org>,
Michal Hocko <mhocko@suse.cz>,
LKML <linux-kernel@vger.kernel.org>, Mel Gorman <mgorman@suse.de>
Subject: [PATCH 11/25] mm, workingset: Make working set detection node-aware
Date: Mon, 8 Jun 2015 14:56:17 +0100 [thread overview]
Message-ID: <1433771791-30567-12-git-send-email-mgorman@suse.de> (raw)
In-Reply-To: <1433771791-30567-1-git-send-email-mgorman@suse.de>
Working set and refault detection is still zone-based, fix it.
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
include/linux/mmzone.h | 6 +++---
mm/vmstat.c | 6 +++---
mm/workingset.c | 47 +++++++++++++++++++++--------------------------
3 files changed, 27 insertions(+), 32 deletions(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 1830c2180555..4c761809d151 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -143,9 +143,6 @@ enum zone_stat_item {
NUMA_LOCAL, /* allocation from local node */
NUMA_OTHER, /* allocation from other node */
#endif
- WORKINGSET_REFAULT,
- WORKINGSET_ACTIVATE,
- WORKINGSET_NODERECLAIM,
NR_ANON_TRANSPARENT_HUGEPAGES,
NR_FREE_CMA_PAGES,
NR_VM_ZONE_STAT_ITEMS };
@@ -160,6 +157,9 @@ enum node_stat_item {
NR_ISOLATED_ANON, /* Temporary isolated pages from anon lru */
NR_ISOLATED_FILE, /* Temporary isolated pages from file lru */
NR_PAGES_SCANNED, /* pages scanned since last reclaim */
+ WORKINGSET_REFAULT,
+ WORKINGSET_ACTIVATE,
+ WORKINGSET_NODERECLAIM,
NR_VM_NODE_STAT_ITEMS
};
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 36897da22792..054ee50974c9 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -921,9 +921,6 @@ const char * const vmstat_text[] = {
"numa_local",
"numa_other",
#endif
- "workingset_refault",
- "workingset_activate",
- "workingset_nodereclaim",
"nr_anon_transparent_hugepages",
"nr_free_cma",
@@ -936,6 +933,9 @@ const char * const vmstat_text[] = {
"nr_isolated_anon",
"nr_isolated_file",
"nr_pages_scanned",
+ "workingset_refault",
+ "workingset_activate",
+ "workingset_nodereclaim",
/* enum writeback_stat_item counters */
"nr_dirty_threshold",
diff --git a/mm/workingset.c b/mm/workingset.c
index ca080cc11797..1cc71f1ca7fc 100644
--- a/mm/workingset.c
+++ b/mm/workingset.c
@@ -16,7 +16,7 @@
/*
* Double CLOCK lists
*
- * Per zone, two clock lists are maintained for file pages: the
+ * Per node, two clock lists are maintained for file pages: the
* inactive and the active list. Freshly faulted pages start out at
* the head of the inactive list and page reclaim scans pages from the
* tail. Pages that are accessed multiple times on the inactive list
@@ -141,48 +141,43 @@
*
* Implementation
*
- * For each zone's file LRU lists, a counter for inactive evictions
- * and activations is maintained (zone->inactive_age).
+ * For each node's file LRU lists, a counter for inactive evictions
+ * and activations is maintained (node->inactive_age).
*
* On eviction, a snapshot of this counter (along with some bits to
- * identify the zone) is stored in the now empty page cache radix tree
+ * identify the node) is stored in the now empty page cache radix tree
* slot of the evicted page. This is called a shadow entry.
*
* On cache misses for which there are shadow entries, an eligible
* refault distance will immediately activate the refaulting page.
*/
-static void *pack_shadow(unsigned long eviction, struct zone *zone)
+static void *pack_shadow(unsigned long eviction, struct pglist_data *pgdat)
{
- eviction = (eviction << NODES_SHIFT) | zone_to_nid(zone);
- eviction = (eviction << ZONES_SHIFT) | zone_idx(zone);
+ eviction = (eviction << NODES_SHIFT) | pgdat->node_id;
eviction = (eviction << RADIX_TREE_EXCEPTIONAL_SHIFT);
return (void *)(eviction | RADIX_TREE_EXCEPTIONAL_ENTRY);
}
static void unpack_shadow(void *shadow,
- struct zone **zone,
+ struct pglist_data **pgdat,
unsigned long *distance)
{
unsigned long entry = (unsigned long)shadow;
unsigned long eviction;
unsigned long refault;
unsigned long mask;
- int zid, nid;
+ int nid;
entry >>= RADIX_TREE_EXCEPTIONAL_SHIFT;
- zid = entry & ((1UL << ZONES_SHIFT) - 1);
- entry >>= ZONES_SHIFT;
nid = entry & ((1UL << NODES_SHIFT) - 1);
entry >>= NODES_SHIFT;
eviction = entry;
- *zone = NODE_DATA(nid)->node_zones + zid;
-
- refault = atomic_long_read(&(*zone)->zone_pgdat->inactive_age);
- mask = ~0UL >> (NODES_SHIFT + ZONES_SHIFT +
- RADIX_TREE_EXCEPTIONAL_SHIFT);
+ *pgdat = NODE_DATA(nid);
+ refault = atomic_long_read(&(*pgdat)->inactive_age);
+ mask = ~0UL >> (NODES_SHIFT + RADIX_TREE_EXCEPTIONAL_SHIFT);
/*
* The unsigned subtraction here gives an accurate distance
* across inactive_age overflows in most cases.
@@ -212,11 +207,11 @@ static void unpack_shadow(void *shadow,
*/
void *workingset_eviction(struct address_space *mapping, struct page *page)
{
- struct zone *zone = page_zone(page);
+ struct pglist_data *pgdat = page_zone(page)->zone_pgdat;
unsigned long eviction;
- eviction = atomic_long_inc_return(&zone->zone_pgdat->inactive_age);
- return pack_shadow(eviction, zone);
+ eviction = atomic_long_inc_return(&pgdat->inactive_age);
+ return pack_shadow(eviction, pgdat);
}
/**
@@ -224,20 +219,20 @@ void *workingset_eviction(struct address_space *mapping, struct page *page)
* @shadow: shadow entry of the evicted page
*
* Calculates and evaluates the refault distance of the previously
- * evicted page in the context of the zone it was allocated in.
+ * evicted page in the context of the node it was allocated in.
*
* Returns %true if the page should be activated, %false otherwise.
*/
bool workingset_refault(void *shadow)
{
unsigned long refault_distance;
- struct zone *zone;
+ struct pglist_data *pgdat;
- unpack_shadow(shadow, &zone, &refault_distance);
- inc_zone_state(zone, WORKINGSET_REFAULT);
+ unpack_shadow(shadow, &pgdat, &refault_distance);
+ inc_node_state(pgdat, WORKINGSET_REFAULT);
- if (refault_distance <= node_page_state(zone->zone_pgdat, NR_ACTIVE_FILE)) {
- inc_zone_state(zone, WORKINGSET_ACTIVATE);
+ if (refault_distance <= node_page_state(pgdat, NR_ACTIVE_FILE)) {
+ inc_node_state(pgdat, WORKINGSET_ACTIVATE);
return true;
}
return false;
@@ -356,7 +351,7 @@ static enum lru_status shadow_lru_isolate(struct list_head *item,
}
}
BUG_ON(node->count);
- inc_zone_state(page_zone(virt_to_page(node)), WORKINGSET_NODERECLAIM);
+ inc_node_state(page_zone(virt_to_page(node))->zone_pgdat, WORKINGSET_NODERECLAIM);
if (!__radix_tree_delete_node(&mapping->page_tree, node))
BUG();
--
2.3.5
next prev parent reply other threads:[~2015-06-08 14:00 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-08 13:56 [RFC PATCH 00/25] Move LRU page reclaim from zones to nodes Mel Gorman
2015-06-08 13:56 ` [PATCH 01/25] mm, vmstat: Add infrastructure for per-node vmstats Mel Gorman
2015-06-08 13:56 ` [PATCH 02/25] mm, vmscan: Move lru_lock to the node Mel Gorman
2015-06-08 13:56 ` [PATCH 03/25] mm, vmscan: Move LRU lists to node Mel Gorman
2015-06-08 13:56 ` [PATCH 04/25] mm, vmscan: Begin reclaiming pages on a per-node basis Mel Gorman
2015-06-08 13:56 ` [PATCH 05/25] mm, vmscan: Have kswapd only scan based on the highest requested zone Mel Gorman
2015-06-08 13:56 ` [PATCH 06/25] mm, vmscan: Avoid a second search through zones checking if compaction is required Mel Gorman
2015-06-08 13:56 ` [PATCH 07/25] mm, vmscan: Make kswapd think of reclaim in terms of nodes Mel Gorman
2015-06-08 13:56 ` [PATCH 08/25] mm, vmscan: By default have direct reclaim only shrink once per node Mel Gorman
2015-06-08 13:56 ` [PATCH 09/25] mm, vmscan: Clear congestion, dirty and need for compaction on a per-node basis Mel Gorman
2015-06-08 13:56 ` [PATCH 10/25] mm, vmscan: Make shrink_node decisions more node-centric Mel Gorman
2015-06-08 13:56 ` Mel Gorman [this message]
2015-06-08 13:56 ` [PATCH 12/25] mm, page_alloc: Consider dirtyable memory in terms of nodes Mel Gorman
2015-06-08 13:56 ` [PATCH 13/25] mm: Move NR_FILE_MAPPED accounting to the node Mel Gorman
2015-06-08 13:56 ` [PATCH 14/25] mm: Rename NR_ANON_PAGES to NR_ANON_MAPPED Mel Gorman
2015-06-08 13:56 ` [PATCH 15/25] mm: Move most file-based accounting to the node Mel Gorman
2015-06-08 13:56 ` [PATCH 16/25] mm, vmscan: Update classzone_idx if buffer_heads_over_limit Mel Gorman
2015-06-08 13:56 ` [PATCH 17/25] mm, vmscan: Check if cpusets are enabled during direct reclaim Mel Gorman
2015-06-08 13:56 ` [PATCH 18/25] mm, vmscan: Only wakeup kswapd once per node for the requested classzone Mel Gorman
2015-06-08 13:56 ` [PATCH 19/25] mm, vmscan: Account in vmstat for pages skipped during reclaim Mel Gorman
2015-06-08 13:56 ` [PATCH 20/25] mm, page_alloc: Remove fair zone allocation policy Mel Gorman
2015-06-08 13:56 ` [PATCH 21/25] mm, page_alloc: Defer zlc_setup until it is known it is required Mel Gorman
2015-06-08 13:56 ` [PATCH 22/25] mm: Convert zone_reclaim to node_reclaim Mel Gorman
2015-06-08 13:56 ` [PATCH 23/25] mm, page_alloc: Delete the zonelist_cache Mel Gorman
2015-06-08 13:56 ` [PATCH 24/25] mm, page_alloc: Use ac->classzone_idx instead of zone_idx(preferred_zone) Mel Gorman
2015-06-08 13:56 ` [PATCH 25/25] mm: page_alloc: Take fewer passes when allocating to the low watermark Mel Gorman
2015-06-19 17:01 ` [RFC PATCH 00/25] Move LRU page reclaim from zones to nodes Johannes Weiner
2015-06-21 14:04 ` Mel Gorman
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=1433771791-30567-12-git-send-email-mgorman@suse.de \
--to=mgorman@suse.de \
--cc=hannes@cmpxchg.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@suse.cz \
--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
all inboxes | Powered by JetHome®