[PATCH v1 00/10] Prevent LRU churning

[PATCH v1 00/10] Prevent LRU churning

[Minchan Kim]

There are some places to isolated and putback pages.
For example, compaction does it for getting contiguous page.
The problem is that if we isolate page in the middle of LRU and putback it, 
we lose LRU history as putback_lru_page inserts the page into head of LRU list. 
It means we can evict working set pages. This problem is discussed at LSF/MM.

This patch is for solving the problem as two methods.

 * Anti-churning
   when we isolate page on LRU list, let's not isolate page we can't handle
 * De-churning
   when we putback page on LRU list in migration, let's insert new page into old page's lru position.

[1,2,3/10] is just clean up.
[4,5,6/10] is related to Anti-churning. 
[7,8,9/10] is related to De-churning. 
[10/10] is adding to new tracepoints which is never for merge but just show the effect.

[7/10] is core of in-order putback support but it has a problem on hugepage like
physicall contiguos page stream in my previous RFC version. 
It was already pointed out by Rik. I have another idea to solve it.
Please see description of [7/10].

I test this series in my machine.(1G DRAM, Intel Core 2 Duo)
Test scenario is following as. 

1) Boot up
2) qemu ubuntu start up 
3) Run many applications and switch attached script(which is made by Wu)

I think this scenario is worst case since there are many contigous pages
when mahchine boots up by not aging(ie, not-fragment).

Test result is following as. 
For compaction, it isolated about 20000 pages. Only 10 pages are put backed with
out-of-order(ie, head of LRU)
Others, about 19990 pages are put backed with in-order(ie, position of old page while
migration happens)

Thanks. 

Minchan Kim (10):
  [1/10] Make clear description of isolate/putback functions
  [2/10] compaction: trivial clean up acct_isolated
  [3/10] Change int mode for isolate mode with enum ISOLATE_PAGE_MODE
  [4/10] Add additional isolation mode
  [5/10] compaction: make isolate_lru_page with filter aware
  [6/10] vmscan: make isolate_lru_page with filter aware
  [7/10] In order putback lru core
  [8/10] migration: make in-order-putback aware
  [9/10] compaction: make compaction use in-order putback
  [10/10] add tracepoints

 include/linux/memcontrol.h    |    5 +-
 include/linux/migrate.h       |   40 +++++
 include/linux/mm_types.h      |   16 ++-
 include/linux/swap.h          |   18 ++-
 include/trace/events/vmscan.h |    8 +-
 mm/compaction.c               |   47 +++---
 mm/internal.h                 |    2 +
 mm/memcontrol.c               |    3 +-
 mm/migrate.c                  |  389 ++++++++++++++++++++++++++++++++++++++++-
 mm/swap.c                     |    2 +-
 mm/vmscan.c                   |  115 ++++++++++--
 11 files changed, 590 insertions(+), 55 deletions(-)

--
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>

[PATCH v1 01/10] Make clear description of isolate/putback functions

[Minchan Kim]

Commonly, putback_lru_page is used with isolated_lru_page.
The isolated_lru_page picks the page in middle of LRU and
putback_lru_page insert the lru in head of LRU.
It means it could make LRU churning so we have to be very careful.
Let's clear description of isolate/putback functions.

Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 mm/migrate.c |    2 +-
 mm/vmscan.c  |    8 ++++++--
 2 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/mm/migrate.c b/mm/migrate.c
index e4a5c91..a04f68a 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -68,7 +68,7 @@ int migrate_prep_local(void)
 }
 
 /*
- * Add isolated pages on the list back to the LRU under page lock
+ * Add isolated pages on the list back to the LRU's head under page lock
  * to avoid leaking evictable pages back onto unevictable list.
  */
 void putback_lru_pages(struct list_head *l)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 292582c..a6a87c0 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -555,10 +555,10 @@ int remove_mapping(struct address_space *mapping, struct page *page)
 }
 
 /**
- * putback_lru_page - put previously isolated page onto appropriate LRU list
+ * putback_lru_page - put previously isolated page onto appropriate LRU list's head
  * @page: page to be put back to appropriate lru list
  *
- * Add previously isolated @page to appropriate LRU list.
+ * Add previously isolated @page to appropriate LRU list's head
  * Page may still be unevictable for other reasons.
  *
  * lru_lock must not be held, interrupts must be enabled.
@@ -1200,6 +1200,10 @@ static unsigned long clear_active_flags(struct list_head *page_list,
  *     without a stable reference).
  * (2) the lru_lock must not be held.
  * (3) interrupts must be enabled.
+ *
+ * NOTE : This function removes the page from LRU list and putback_lru_page
+ * insert the page to LRU list's head. It means it makes LRU churing so you
+ * have to use the function carefully.
  */
 int isolate_lru_page(struct page *page)
 {
-- 
1.7.1

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH v1 02/10] compaction: trivial clean up acct_isolated

[Minchan Kim]

acct_isolated of compaction uses page_lru_base_type which returns only
base type of LRU list so it never returns LRU_ACTIVE_ANON or LRU_ACTIVE_FILE.
In addtion, cc->nr_[anon|file] is used in only acct_isolated so it doesn't have
fields in conpact_control.
This patch removes fields from compact_control and makes clear function of
acct_issolated which counts the number of anon|file pages isolated.

Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 mm/compaction.c |   18 +++++-------------
 1 files changed, 5 insertions(+), 13 deletions(-)

diff --git a/mm/compaction.c b/mm/compaction.c
index 021a296..61eab88 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -35,10 +35,6 @@ struct compact_control {
 	unsigned long migrate_pfn;	/* isolate_migratepages search base */
 	bool sync;			/* Synchronous migration */
 
-	/* Account for isolated anon and file pages */
-	unsigned long nr_anon;
-	unsigned long nr_file;
-
 	unsigned int order;		/* order a direct compactor needs */
 	int migratetype;		/* MOVABLE, RECLAIMABLE etc */
 	struct zone *zone;
@@ -212,17 +208,13 @@ static void isolate_freepages(struct zone *zone,
 static void acct_isolated(struct zone *zone, struct compact_control *cc)
 {
 	struct page *page;
-	unsigned int count[NR_LRU_LISTS] = { 0, };
+	unsigned int count[2] = { 0, };
 
-	list_for_each_entry(page, &cc->migratepages, lru) {
-		int lru = page_lru_base_type(page);
-		count[lru]++;
-	}
+	list_for_each_entry(page, &cc->migratepages, lru)
+		count[!!page_is_file_cache(page)]++;
 
-	cc->nr_anon = count[LRU_ACTIVE_ANON] + count[LRU_INACTIVE_ANON];
-	cc->nr_file = count[LRU_ACTIVE_FILE] + count[LRU_INACTIVE_FILE];
-	__mod_zone_page_state(zone, NR_ISOLATED_ANON, cc->nr_anon);
-	__mod_zone_page_state(zone, NR_ISOLATED_FILE, cc->nr_file);
+	__mod_zone_page_state(zone, NR_ISOLATED_ANON, count[0]);
+	__mod_zone_page_state(zone, NR_ISOLATED_FILE, count[1]);
 }
 
 /* Similar to reclaim, but different enough that they don't share logic */
-- 
1.7.1

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH v1 03/10] Change int mode for isolate mode with enum ISOLATE_PAGE_MODE

[Minchan Kim]

This patch changes macro define with enum variable.
Normally, enum is preferred as it's type-safe and making debugging easier
as symbol can be passed throught to the debugger.

This patch doesn't change old behavior and it will be used by next patches.

Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 include/linux/memcontrol.h    |    5 ++++-
 include/linux/swap.h          |   12 ++++++++----
 include/trace/events/vmscan.h |    8 ++++----
 mm/memcontrol.c               |    3 ++-
 mm/vmscan.c                   |   19 +++++++++++++------
 5 files changed, 31 insertions(+), 16 deletions(-)

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 77e47f5..81ec507 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -32,10 +32,13 @@ enum mem_cgroup_page_stat_item {
 	MEMCG_NR_FILE_MAPPED, /* # of pages charged as file rss */
 };
 
+enum ISOLATE_PAGE_MODE;
+
 extern unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
 					struct list_head *dst,
 					unsigned long *scanned, int order,
-					int mode, struct zone *z,
+					enum ISOLATE_PAGE_MODE mode,
+					struct zone *z,
 					struct mem_cgroup *mem_cont,
 					int active, int file);
 
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 384eb5f..a7cc199 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -244,9 +244,12 @@ static inline void lru_cache_add_file(struct page *page)
 }
 
 /* LRU Isolation modes. */
-#define ISOLATE_INACTIVE 0	/* Isolate inactive pages. */
-#define ISOLATE_ACTIVE 1	/* Isolate active pages. */
-#define ISOLATE_BOTH 2		/* Isolate both active and inactive pages. */
+enum ISOLATE_PAGE_MODE {
+	ISOLATE_NONE,
+	ISOLATE_INACTIVE = 1,	/* Isolate inactive pages */
+	ISOLATE_ACTIVE = 2,	/* Isolate active pages */
+	ISOLATE_BOTH = 4,	/* Isolate both active and inactive pages */
+};
 
 /* linux/mm/vmscan.c */
 extern unsigned long try_to_free_pages(struct zonelist *zonelist, int order,
@@ -259,7 +262,8 @@ extern unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *mem,
 						unsigned int swappiness,
 						struct zone *zone,
 						unsigned long *nr_scanned);
-extern int __isolate_lru_page(struct page *page, int mode, int file);
+extern int __isolate_lru_page(struct page *page, enum ISOLATE_PAGE_MODE mode,
+						int file);
 extern unsigned long shrink_all_memory(unsigned long nr_pages);
 extern int vm_swappiness;
 extern int remove_mapping(struct address_space *mapping, struct page *page);
diff --git a/include/trace/events/vmscan.h b/include/trace/events/vmscan.h
index ea422aa..a20d766 100644
--- a/include/trace/events/vmscan.h
+++ b/include/trace/events/vmscan.h
@@ -187,7 +187,7 @@ DECLARE_EVENT_CLASS(mm_vmscan_lru_isolate_template,
 		unsigned long nr_lumpy_taken,
 		unsigned long nr_lumpy_dirty,
 		unsigned long nr_lumpy_failed,
-		int isolate_mode),
+		enum ISOLATE_PAGE_MODE isolate_mode),
 
 	TP_ARGS(order, nr_requested, nr_scanned, nr_taken, nr_lumpy_taken, nr_lumpy_dirty, nr_lumpy_failed, isolate_mode),
 
@@ -199,7 +199,7 @@ DECLARE_EVENT_CLASS(mm_vmscan_lru_isolate_template,
 		__field(unsigned long, nr_lumpy_taken)
 		__field(unsigned long, nr_lumpy_dirty)
 		__field(unsigned long, nr_lumpy_failed)
-		__field(int, isolate_mode)
+		__field(enum ISOLATE_PAGE_MODE, isolate_mode)
 	),
 
 	TP_fast_assign(
@@ -233,7 +233,7 @@ DEFINE_EVENT(mm_vmscan_lru_isolate_template, mm_vmscan_lru_isolate,
 		unsigned long nr_lumpy_taken,
 		unsigned long nr_lumpy_dirty,
 		unsigned long nr_lumpy_failed,
-		int isolate_mode),
+		enum ISOLATE_PAGE_MODE isolate_mode),
 
 	TP_ARGS(order, nr_requested, nr_scanned, nr_taken, nr_lumpy_taken, nr_lumpy_dirty, nr_lumpy_failed, isolate_mode)
 
@@ -248,7 +248,7 @@ DEFINE_EVENT(mm_vmscan_lru_isolate_template, mm_vmscan_memcg_isolate,
 		unsigned long nr_lumpy_taken,
 		unsigned long nr_lumpy_dirty,
 		unsigned long nr_lumpy_failed,
-		int isolate_mode),
+		enum ISOLATE_PAGE_MODE isolate_mode),
 
 	TP_ARGS(order, nr_requested, nr_scanned, nr_taken, nr_lumpy_taken, nr_lumpy_dirty, nr_lumpy_failed, isolate_mode)
 
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 95aecca..0400d32 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1193,7 +1193,8 @@ mem_cgroup_get_reclaim_stat_from_page(struct page *page)
 unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
 					struct list_head *dst,
 					unsigned long *scanned, int order,
-					int mode, struct zone *z,
+					enum ISOLATE_PAGE_MODE mode,
+					struct zone *z,
 					struct mem_cgroup *mem_cont,
 					int active, int file)
 {
diff --git a/mm/vmscan.c b/mm/vmscan.c
index a6a87c0..5d83e06 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -961,23 +961,29 @@ keep_lumpy:
  *
  * returns 0 on success, -ve errno on failure.
  */
-int __isolate_lru_page(struct page *page, int mode, int file)
+int __isolate_lru_page(struct page *page, enum ISOLATE_PAGE_MODE mode,
+							int file)
 {
+	int active;
 	int ret = -EINVAL;
+	BUG_ON(mode & ISOLATE_BOTH &&
+		(mode & ISOLATE_INACTIVE || mode & ISOLATE_ACTIVE));
 
 	/* Only take pages on the LRU. */
 	if (!PageLRU(page))
 		return ret;
 
+	active = PageActive(page);
+
 	/*
 	 * When checking the active state, we need to be sure we are
 	 * dealing with comparible boolean values.  Take the logical not
 	 * of each.
 	 */
-	if (mode != ISOLATE_BOTH && (!PageActive(page) != !mode))
+	if (mode & ISOLATE_ACTIVE && !active)
 		return ret;
 
-	if (mode != ISOLATE_BOTH && page_is_file_cache(page) != file)
+	if (mode & ISOLATE_INACTIVE && active)
 		return ret;
 
 	/*
@@ -1025,7 +1031,8 @@ int __isolate_lru_page(struct page *page, int mode, int file)
  */
 static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
 		struct list_head *src, struct list_head *dst,
-		unsigned long *scanned, int order, int mode, int file)
+		unsigned long *scanned, int order, enum ISOLATE_PAGE_MODE mode,
+		int file)
 {
 	unsigned long nr_taken = 0;
 	unsigned long nr_lumpy_taken = 0;
@@ -1138,8 +1145,8 @@ static unsigned long isolate_lru_pages(unsigned long nr_to_scan,
 static unsigned long isolate_pages_global(unsigned long nr,
 					struct list_head *dst,
 					unsigned long *scanned, int order,
-					int mode, struct zone *z,
-					int active, int file)
+					enum ISOLATE_PAGE_MODE mode,
+					struct zone *z,	int active, int file)
 {
 	int lru = LRU_BASE;
 	if (active)
-- 
1.7.1

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH v1 04/10] Add additional isolation mode

[Minchan Kim]

There are some places to isolate lru page and I believe
users of isolate_lru_page will be growing.
The purpose of them is each different so part of isolated pages
should put back to LRU, again.

The problem is when we put back the page into LRU,
we lose LRU ordering and the page is inserted at head of LRU list.
It makes unnecessary LRU churning so that vm can evict working set pages
rather than idle pages.

This patch adds new modes when we isolate page in LRU so we don't isolate pages
if we can't handle it. It could reduce LRU churning.

This patch doesn't change old behavior. It's just used by next patches.

Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 include/linux/swap.h |    2 ++
 mm/vmscan.c          |    6 ++++++
 2 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/include/linux/swap.h b/include/linux/swap.h
index a7cc199..0badb13 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -249,6 +249,8 @@ enum ISOLATE_PAGE_MODE {
 	ISOLATE_INACTIVE = 1,	/* Isolate inactive pages */
 	ISOLATE_ACTIVE = 2,	/* Isolate active pages */
 	ISOLATE_BOTH = 4,	/* Isolate both active and inactive pages */
+	ISOLATE_CLEAN = 8,	/* Isolate clean file */
+	ISOLATE_UNMAPPED = 16,	/* Isolate unmapped file */
 };
 
 /* linux/mm/vmscan.c */
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 5d83e06..4bd5513 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -996,6 +996,12 @@ int __isolate_lru_page(struct page *page, enum ISOLATE_PAGE_MODE mode,
 
 	ret = -EBUSY;
 
+	if (mode & ISOLATE_CLEAN && (PageDirty(page) || PageWriteback(page)))
+		return ret;
+
+	if (mode & ISOLATE_UNMAPPED && page_mapped(page))
+		return ret;
+
 	if (likely(get_page_unless_zero(page))) {
 		/*
 		 * Be careful not to clear PageLRU until after we're
-- 
1.7.1

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH v1 05/10] compaction: make isolate_lru_page with filter aware

[Minchan Kim]

In async mode, compaction doesn't migrate dirty or writeback pages.
So, it's meaningless to pick the page and re-add it to lru list.

Of course, when we isolate the page in compaction, the page might
be dirty or writeback but when we try to migrate the page, the page
would be not dirty, writeback. So it could be migrated. But it's
very unlikely as isolate and migration cycle is much faster than
writeout.

So, this patch helps cpu and prevent unnecessary LRU churning.

Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 mm/compaction.c |    5 ++++-
 1 files changed, 4 insertions(+), 1 deletions(-)

diff --git a/mm/compaction.c b/mm/compaction.c
index 61eab88..e218562 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -243,6 +243,7 @@ static unsigned long isolate_migratepages(struct zone *zone,
 	unsigned long last_pageblock_nr = 0, pageblock_nr;
 	unsigned long nr_scanned = 0, nr_isolated = 0;
 	struct list_head *migratelist = &cc->migratepages;
+	enum ISOLATE_PAGE_MODE mode = ISOLATE_BOTH;
 
 	/* Do not scan outside zone boundaries */
 	low_pfn = max(cc->migrate_pfn, zone->zone_start_pfn);
@@ -327,7 +328,9 @@ static unsigned long isolate_migratepages(struct zone *zone,
 		}
 
 		/* Try isolate the page */
-		if (__isolate_lru_page(page, ISOLATE_BOTH, 0) != 0)
+		if (!cc->sync)
+			mode |= ISOLATE_CLEAN;
+		if (__isolate_lru_page(page, mode, 0) != 0)
 			continue;
 
 		VM_BUG_ON(PageTransCompound(page));
-- 
1.7.1

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH v1 06/10] vmscan: make isolate_lru_page with filter aware

[Minchan Kim]

In some __zone_reclaim case, we don't want to shrink mapped page.
Nonetheless, we have isolated mapped page and re-add it into
LRU's head. It's unnecessary CPU overhead and makes LRU churning.

Of course, when we isolate the page, the page might be mapped but
when we try to migrate the page, the page would be not mapped.
So it could be migrated. But race is rare and although it happens,
it's no big deal.

Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 mm/vmscan.c |   29 +++++++++++++++++++++--------
 1 files changed, 21 insertions(+), 8 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 4bd5513..88c6baf 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1402,6 +1402,7 @@ shrink_inactive_list(unsigned long nr_to_scan, struct zone *zone,
 	unsigned long nr_taken;
 	unsigned long nr_anon;
 	unsigned long nr_file;
+	enum ISOLATE_PAGE_MODE mode = ISOLATE_NONE;
 
 	while (unlikely(too_many_isolated(zone, file, sc))) {
 		congestion_wait(BLK_RW_ASYNC, HZ/10);
@@ -1413,13 +1414,20 @@ shrink_inactive_list(unsigned long nr_to_scan, struct zone *zone,
 
 	set_reclaim_mode(priority, sc, false);
 	lru_add_drain();
+
+	if (!sc->may_unmap)
+		mode |= ISOLATE_UNMAPPED;
+	if (!sc->may_writepage)
+		mode |= ISOLATE_CLEAN;
+	mode |= sc->reclaim_mode & RECLAIM_MODE_LUMPYRECLAIM ?
+				ISOLATE_BOTH : ISOLATE_INACTIVE;
+
 	spin_lock_irq(&zone->lru_lock);
 
+
 	if (scanning_global_lru(sc)) {
 		nr_taken = isolate_pages_global(nr_to_scan,
-			&page_list, &nr_scanned, sc->order,
-			sc->reclaim_mode & RECLAIM_MODE_LUMPYRECLAIM ?
-					ISOLATE_BOTH : ISOLATE_INACTIVE,
+			&page_list, &nr_scanned, sc->order, mode,
 			zone, 0, file);
 		zone->pages_scanned += nr_scanned;
 		if (current_is_kswapd())
@@ -1430,9 +1438,7 @@ shrink_inactive_list(unsigned long nr_to_scan, struct zone *zone,
 					       nr_scanned);
 	} else {
 		nr_taken = mem_cgroup_isolate_pages(nr_to_scan,
-			&page_list, &nr_scanned, sc->order,
-			sc->reclaim_mode & RECLAIM_MODE_LUMPYRECLAIM ?
-					ISOLATE_BOTH : ISOLATE_INACTIVE,
+			&page_list, &nr_scanned, sc->order, mode,
 			zone, sc->mem_cgroup,
 			0, file);
 		/*
@@ -1536,19 +1542,26 @@ static void shrink_active_list(unsigned long nr_pages, struct zone *zone,
 	struct page *page;
 	struct zone_reclaim_stat *reclaim_stat = get_reclaim_stat(zone, sc);
 	unsigned long nr_rotated = 0;
+	enum ISOLATE_PAGE_MODE mode = ISOLATE_ACTIVE;
 
 	lru_add_drain();
+
+	if (!sc->may_unmap)
+		mode |= ISOLATE_UNMAPPED;
+	if (!sc->may_writepage)
+		mode |= ISOLATE_CLEAN;
+
 	spin_lock_irq(&zone->lru_lock);
 	if (scanning_global_lru(sc)) {
 		nr_taken = isolate_pages_global(nr_pages, &l_hold,
 						&pgscanned, sc->order,
-						ISOLATE_ACTIVE, zone,
+						mode, zone,
 						1, file);
 		zone->pages_scanned += pgscanned;
 	} else {
 		nr_taken = mem_cgroup_isolate_pages(nr_pages, &l_hold,
 						&pgscanned, sc->order,
-						ISOLATE_ACTIVE, zone,
+						mode, zone,
 						sc->mem_cgroup, 1, file);
 		/*
 		 * mem_cgroup_isolate_pages() keeps track of
-- 
1.7.1

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH v1 07/10] In order putback lru core

[Minchan Kim]

This patch defines new APIs to put back the page into previous position of LRU.
The idea I suggested in LSF/MM is simple.

When we try to put back the page into lru list and if friends(prev, next) of the page
still is nearest neighbor, we can insert isolated page into prev's next instead of
head of LRU list. So it keeps LRU history without losing the LRU information.

Before :
       LRU POV : H - P1 - P2 - P3 - P4 -T

Isolate P3 :
       LRU POV : H - P1 - P2 - P4 - T

Putback P3 :
       if (P2->next == P4)
               putback(P3, P2);
       So,
       LRU POV : H - P1 - P2 - P3 - P4 -T

I implemented this idea in RFC but it had two problems.

1)
For implement, I defined new structure _pages_lru_ which remembers
both lru friend pages of isolated one and handling functions.
For space of pages_lru, I allocated the space dynamically in kmalloc(GFP_AOTMIC)
but as we know, compaction is a reclaim path so it's not good idea to allocate memory
dynamically in the path. The space need to store pages_lru is enough to allocate just a page
as current compaction migrates unit of chunk of 32 pages.
In addition, compaction makes sure lots of order-0 free pages before starting
so it wouldn't a big problem, I think. But I admit it can pin some pages
so migration successful ratio might be down if concurrent compaction happens.

I decide changing my mind. I don't use dynamic memory space any more.
As I see migration, we don't need doubly linked list of page->lru.
Whole of operation is performed with enumeration so I think singly linked list is enough.
If we can use singly linked list, we can use a pointer as another buffer.
In here, we use it to store prev LRU page of page isolated.

2)
The page-relation approach had a problem on contiguous pages.
That's because the idea can not work since friend pages are isolated, too.
It means prev_page->next == next_page always is _false_ and both pages are not
LRU any more at that time. It's pointed out by Rik at LSF/MM summit.
So for solving the problem, I changed the idea.
We don't need both friend(prev, next) pages relation but just consider
either prev or next page that it is still same LRU

Worst case in this approach, prev or next page is free and allocate new
so it's in head of LRU and our isolated page is located on next of head.
But it's almost same situation with current problem. So it doesn't make worse
than now.
New idea works below.

===

assume : we isolate pages P3~P7 and we consider only prev LRU pointer.
notation : (P3,P2) = (isolated page,prev LRU page of isolated page)

H - P1 - P2 - P3 - P4 - P5 - P6 - P7 - P8 - P9 - P10 - T

If we isolate P3, following as

H - P1 - P2 - P4 - P5 - P6 - P7 - P8 - P9 - P10 - T
Isolated page list - (P3,P2)

If we isolate P4, following as

H - P1 - P2 - P5 - P6 - P7 - P8 - P9 - P10 - T
Isolated page list - (P4,P2) - (P3,P2)

If we isolate P5, following as

H - P1 - P2 - P6 - P7 - P8 - P9 - P10 - T
Isolated page list - (P5,P2) - (P4,P2) - (P3,P2)

..
..

If we isolate P7, following as

H - P1 - P2 - P8 - P9 - P10 - T

Isolated page list - (P7,P2) - (P6,P2) - (P5,P2) - (P4,P2) - (P3,P2)

Let's start putback from P7

P7)

H - P1 - P2 - P8 - P9 - P10 - T
prev P2 is valid, too. So,

H - P1 - P2 - P7 - P8 - P9 - P10 - T

P6)

H - P1 - P2 - P7 - P8 - P9 - P10 - T
Prev P2 is valid, too. So,

H - P1 - P2 - P6 - P7 - P8 - P9 - P10 - T

..
..

P3)
H - P1 - P2 - P4 - P5 - P6 - P7 - P8 - P9 - P10 - T
Prev P2 is valid, too. So,

H - P1 - P2 - P3 - P4 - P5 - P6 - P7 - P8 - P9 - P10 - T

===

Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 include/linux/migrate.h  |   35 ++++++++++++++++++
 include/linux/mm_types.h |   16 ++++++++-
 include/linux/swap.h     |    4 ++
 mm/internal.h            |    2 +
 mm/migrate.c             |   90 ++++++++++++++++++++++++++++++++++++++++++++++
 mm/swap.c                |    2 +-
 mm/vmscan.c              |   50 +++++++++++++++++++++++++
 7 files changed, 197 insertions(+), 2 deletions(-)

diff --git a/include/linux/migrate.h b/include/linux/migrate.h
index e39aeec..ca20500 100644
--- a/include/linux/migrate.h
+++ b/include/linux/migrate.h
@@ -9,7 +9,42 @@ typedef struct page *new_page_t(struct page *, unsigned long private, int **);
 #ifdef CONFIG_MIGRATION
 #define PAGE_MIGRATION 1
 
+/*
+ * Migratelist for compaction is singly linked list instead of double linked list.
+ * Current list utility is useful in some sense but we can't make sure compatibilty.
+ * Please use below functions instead of common list's ones.
+ */
+static inline void INIT_MIGRATE_LIST(struct inorder_lru *list)
+{
+	list->prev_page = NULL;
+	list->next = list;
+}
+
+static inline int migratelist_empty(const struct inorder_lru *head)
+{
+	return head->next == head;
+}
+
+static inline void migratelist_add(struct page *page,
+			struct page *prev_page, struct inorder_lru *head)
+{
+	VM_BUG_ON(PageLRU(page));
+
+	page->ilru.prev_page = prev_page;
+	page->ilru.next = head->next;
+	head->next = &page->ilru;
+}
+
+static inline void migratelist_del(struct page *page, struct inorder_lru *head)
+{
+	head->next = page->ilru.next;
+}
+
+#define list_for_each_migrate_entry		list_for_each_entry
+#define list_for_each_migrate_entry_safe	list_for_each_entry_safe
+
 extern void putback_lru_pages(struct list_head *l);
+extern void putback_inorder_lru_pages(struct inorder_lru *l);
 extern int migrate_page(struct address_space *,
 			struct page *, struct page *);
 extern int migrate_pages(struct list_head *l, new_page_t x,
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 27c498b..2b5fbe9 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -24,6 +24,17 @@ struct address_space;
 
 #define USE_SPLIT_PTLOCKS	(NR_CPUS >= CONFIG_SPLIT_PTLOCK_CPUS)
 
+struct page;
+
+/*
+ * The inorder_lru is used by compaction for keeping LRU order
+ * during migration.
+ */
+struct inorder_lru {
+	struct page *prev_page; 	/* prev LRU page of isolated page */
+	struct inorder_lru *next;	/* next pointer for singly linked list*/
+};
+
 /*
  * Each physical page in the system has a struct page associated with
  * it to keep track of whatever it is we are using the page for at the
@@ -72,9 +83,12 @@ struct page {
 		pgoff_t index;		/* Our offset within mapping. */
 		void *freelist;		/* SLUB: freelist req. slab lock */
 	};
-	struct list_head lru;		/* Pageout list, eg. active_list
+	union {
+		struct inorder_lru ilru;/* compaction: migrated page list */
+		struct list_head lru;	/* Pageout list, eg. active_list
 					 * protected by zone->lru_lock !
 					 */
+	};
 	/*
 	 * On machines where all RAM is mapped into kernel address space,
 	 * we can simply calculate the virtual address. On machines with
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 0badb13..5fe6919 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -226,6 +226,8 @@ extern int lru_add_drain_all(void);
 extern void rotate_reclaimable_page(struct page *page);
 extern void deactivate_page(struct page *page);
 extern void swap_setup(void);
+extern void update_page_reclaim_stat(struct zone *zone, struct page *page,
+		int file, int rotated);
 
 extern void add_page_to_unevictable_list(struct page *page);
 
@@ -264,6 +266,8 @@ extern unsigned long mem_cgroup_shrink_node_zone(struct mem_cgroup *mem,
 						unsigned int swappiness,
 						struct zone *zone,
 						unsigned long *nr_scanned);
+extern int __isolate_inorder_lru_page(struct page *page, enum ISOLATE_PAGE_MODE mode,
+				int file, struct page **lru_p_page);
 extern int __isolate_lru_page(struct page *page, enum ISOLATE_PAGE_MODE mode,
 						int file);
 extern unsigned long shrink_all_memory(unsigned long nr_pages);
diff --git a/mm/internal.h b/mm/internal.h
index d071d38..3aa15e0 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -42,6 +42,8 @@ extern unsigned long highest_memmap_pfn;
 /*
  * in mm/vmscan.c:
  */
+extern bool keep_lru_order(struct page *page, struct page *prev);
+extern void putback_page_to_lru(struct page *page, struct page *head_page);
 extern int isolate_lru_page(struct page *page);
 extern void putback_lru_page(struct page *page);
 
diff --git a/mm/migrate.c b/mm/migrate.c
index a04f68a..8986469 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -84,6 +84,30 @@ void putback_lru_pages(struct list_head *l)
 	}
 }
 
+void putback_inorder_lru_pages(struct inorder_lru *l)
+{
+	struct zone *zone;
+	struct page *page, *page2, *prev;
+
+	list_for_each_migrate_entry_safe(page, page2, l, ilru) {
+		dec_zone_page_state(page, NR_ISOLATED_ANON +
+				page_is_file_cache(page));
+		zone = page_zone(page);
+		spin_lock_irq(&zone->lru_lock);
+		prev = page->ilru.prev_page;
+		if (keep_lru_order(page, prev)) {
+			putback_page_to_lru(page, prev);
+			spin_unlock_irq(&zone->lru_lock);
+		}
+		else {
+			spin_unlock_irq(&zone->lru_lock);
+			putback_lru_page(page);
+		}
+
+		l->next = &page2->ilru;
+	}
+}
+
 /*
  * Restore a potential migration pte to a working pte entry
  */
@@ -995,6 +1019,72 @@ out:
 	return nr_failed + retry;
 }
 
+int __isolate_inorder_lru_page(struct page *page, enum ISOLATE_PAGE_MODE mode,
+		int file, struct page **lru_p_page)
+{
+	int active;
+	int ret = -EINVAL;
+	BUG_ON(mode & ISOLATE_BOTH &&
+			(mode & ISOLATE_INACTIVE || mode & ISOLATE_ACTIVE));
+
+	/* Only take pages on the LRU. */
+	if (!PageLRU(page))
+		return ret;
+
+	active = PageActive(page);
+
+	/*
+	 * When checking the active state, we need to be sure we are
+	 * dealing with comparible boolean values.  Take the logical not
+	 * of each.
+	 */
+	if (mode & ISOLATE_ACTIVE && !active)
+		return ret;
+
+	if (mode & ISOLATE_INACTIVE && active)
+		return ret;
+
+	/*
+	 * When this function is being called for lumpy reclaim, we
+	 * initially look into all LRU pages, active, inactive and
+	 * unevictable; only give shrink_page_list evictable pages.
+	 */
+	if (PageUnevictable(page))
+		return ret;
+
+	ret = -EBUSY;
+
+	if (mode & ISOLATE_CLEAN && (PageDirty(page) || PageWriteback(page)))
+		return ret;
+
+	if (mode & ISOLATE_UNMAPPED && page_mapped(page))
+		return ret;
+
+	if (likely(get_page_unless_zero(page))) {
+		struct zone *zone = page_zone(page);
+		struct page *prev_page;
+		enum lru_list l = page_lru(page);
+		/*
+		 * Be careful not to clear PageLRU until after we're
+		 * sure the page is not being freed elsewhere -- the
+		 * page release code relies on it.
+		 */
+		ClearPageLRU(page);
+
+		if (&zone->lru[l].list == page->lru.prev) {
+			*lru_p_page = NULL;
+			goto out;
+		}
+
+		prev_page = list_entry(page->lru.prev, struct page, lru);
+		*lru_p_page = prev_page;
+out:
+		ret = 0;
+	}
+
+	return ret;
+}
+
 #ifdef CONFIG_NUMA
 /*
  * Move a list of individual pages
diff --git a/mm/swap.c b/mm/swap.c
index 3a442f1..bdaf329 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -252,7 +252,7 @@ void rotate_reclaimable_page(struct page *page)
 	}
 }
 
-static void update_page_reclaim_stat(struct zone *zone, struct page *page,
+void update_page_reclaim_stat(struct zone *zone, struct page *page,
 				     int file, int rotated)
 {
 	struct zone_reclaim_stat *reclaim_stat = &zone->reclaim_stat;
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 88c6baf..62d5186 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -554,6 +554,56 @@ int remove_mapping(struct address_space *mapping, struct page *page)
 	return 0;
 }
 
+/*
+ * If prev_page is same LRU with page, we can keep LRU order of page.
+ * zone->lru_lock must be hold.
+ */
+bool keep_lru_order(struct page *page, struct page *prev)
+{
+	bool ret = false;
+	if (!prev || !PageLRU(prev))
+		goto out;
+
+	if (unlikely(PageUnevictable(prev)))
+		goto out;
+
+	if (page_lru_base_type(page) != page_lru_base_type(prev))
+		goto out;
+
+	ret = true;
+out:
+	return ret;
+}
+
+/**
+ * putback_page_to_lru - put isolated @page onto @head
+ * @page: page to be put back to appropriate lru list
+ * @head_page: lru position to be put back
+ *
+ * Insert previously isolated @page to appropriate position of lru list
+ * zone->lru_lock must be hold.
+ */
+void putback_page_to_lru(struct page *page, struct page *head_page)
+{
+	int lru, active, file;
+	struct zone *zone = page_zone(page);
+
+	VM_BUG_ON(PageLRU(page));
+
+	lru = page_lru(head_page);
+	active = is_active_lru(lru);
+	file = is_file_lru(lru);
+
+	if (active)
+		SetPageActive(page);
+	else
+		ClearPageActive(page);
+
+	update_page_reclaim_stat(zone, page, file, active);
+	SetPageLRU(page);
+	__add_page_to_lru_list(zone, page, lru, &head_page->lru);
+}
+
 /**
  * putback_lru_page - put previously isolated page onto appropriate LRU list's head
  * @page: page to be put back to appropriate lru list
-- 
1.7.1

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH v1 08/10] migration: make in-order-putback aware

[Minchan Kim]

This patch makes migrate_pages_inorder_lru which is aware of in-order putback.
So newpage is located at old page's LRU position.

The logic is following as.

This patch creates new API migrate_pages_inorder_lru for compaction.
We need it because

1. inorder_lru uses singly linked list but migrate_pages doesn't support it
2. I need defer old page's putback.(see below)
3. I don't want to bother generic migrate_pages. Maybe there are some points
   we can unify but I want to defer it after review/merge/stable this series.

For in-order-putback of migration, we need some tweak.
First of all, we need defer old page's putback.
At present, during migration, old page would be freed through unmap_and_move's putback_lru_page.
It has a problem in inorder-putback's keep_lru_order logic.
It does check PageLRU and so on. If old page would be freed, it doesn't have PageLRU any more
so keep_lru_order returns false so inorder putback would become nop.

Second, we need adjust prev_page of inorder_lru pages when we putback newpage and
free old page.
For example,

PHY : H - P1 - P2 - P3 - P4 - P5 - T
LRU : H - P5 - P4 - P3 - P2 - P1 - T
inorder_lru : 0

We isolate P2,P3,P4 so inorder_lru has following list

PHY : H - P1 - P2 - P3 - P4 - P5 - T
LRU : H - P5 - P1 - T
inorder_lru : (P4,P5) - (P3,P4) - (P2,P3)

After 1st putback,

PHY : H - P1 - P2 - P3 - P4 - P5 - T
LRU : H - P5 - P4' - P1 - T
inorder_lru : (P3,P4) - (P2,P3)
P4' is newpage and P4(ie, old page) would freed

In 2nd putback, P3 would find P4 in keep_order_lru but P4 is in buddy
so it returns false then inorder_lru doesn't work any more.
The bad effect continues until P2. That's too bad.
For fixing, this patch defines adjust_inorder_prev_page.
It works following as.

After 1st putback,

PHY : H - P1 - P2 - P3 - P4 - P5 - T
LRU : H - P5 - P4' - P1 - T
inorder_lru : (P3,P4') - (P2,P3)

It replaces old page's pointer with new one's so

In 2nd putback,

PHY : H - P1 - P2 - P3 - P4 - P5 - T
LRU : H - P5 - P4' - P3' - P1 -  T
inorder_lru : (P2,P3')

In 3rd putback,

PHY : H - P1 - P2 - P3 - P4 - P5 - T
LRU : H - P5 - P4' - P3' - P2' - P1 - T
inorder_lru : 0

Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 include/linux/migrate.h |    5 +
 mm/migrate.c            |  290 +++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 295 insertions(+), 0 deletions(-)

diff --git a/include/linux/migrate.h b/include/linux/migrate.h
index ca20500..8e96d92 100644
--- a/include/linux/migrate.h
+++ b/include/linux/migrate.h
@@ -50,6 +50,11 @@ extern int migrate_page(struct address_space *,
 extern int migrate_pages(struct list_head *l, new_page_t x,
 			unsigned long private, bool offlining,
 			bool sync);
+
+extern int migrate_inorder_lru_pages(struct inorder_lru *l, new_page_t x,
+			unsigned long private, bool offlining,
+			bool sync);
+
 extern int migrate_huge_pages(struct list_head *l, new_page_t x,
 			unsigned long private, bool offlining,
 			bool sync);
diff --git a/mm/migrate.c b/mm/migrate.c
index 8986469..f94fe65 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -839,6 +839,246 @@ move_newpage:
 	return rc;
 }
 
+static inline void adjust_inorder_prev_page(struct inorder_lru *head,
+			struct page *prev_page, struct page *new_page)
+{
+	struct page *page;
+	list_for_each_migrate_entry(page, head, ilru)
+		if (page->ilru.prev_page == prev_page)
+			page->ilru.prev_page = new_page;
+}
+
+/*
+ * Counterpart of unmap_and_move() for compaction.
+ * The logic is almost same with unmap_and_move. The difference is
+ * this function handles prev_lru. For inorder-lru compaction, we use
+ * singly linked list so we need prev pointer handling to delete entry.
+ */
+static int unmap_and_move_inorder_lru(new_page_t get_new_page, unsigned long private,
+		struct page *page, int force, bool offlining, bool sync,
+		struct inorder_lru **prev_lru, struct inorder_lru *head)
+{
+	int rc = 0;
+	int *result = NULL;
+	struct page *newpage = get_new_page(page, private, &result);
+	int remap_swapcache = 1;
+	int charge = 0;
+	struct mem_cgroup *mem;
+	struct anon_vma *anon_vma = NULL;
+	struct page *prev_page;
+	struct zone *zone;
+	bool del = false;
+
+	VM_BUG_ON(!prev_lru);
+
+	if (!newpage)
+		return -ENOMEM;
+
+	prev_page = page->ilru.prev_page;
+	if (page_count(page) == 1) {
+		/* page was freed from under us. So we are done. */
+		goto move_newpage;
+	}
+	if (unlikely(PageTransHuge(page)))
+		if (unlikely(split_huge_page(page)))
+			goto move_newpage;
+
+	/* prepare cgroup just returns 0 or -ENOMEM */
+	rc = -EAGAIN;
+
+	if (!trylock_page(page)) {
+		if (!force || !sync)
+			goto move_newpage;
+
+		/*
+		 * It's not safe for direct compaction to call lock_page.
+		 * For example, during page readahead pages are added locked
+		 * to the LRU. Later, when the IO completes the pages are
+		 * marked uptodate and unlocked. However, the queueing
+		 * could be merging multiple pages for one bio (e.g.
+		 * mpage_readpages). If an allocation happens for the
+		 * second or third page, the process can end up locking
+		 * the same page twice and deadlocking. Rather than
+		 * trying to be clever about what pages can be locked,
+		 * avoid the use of lock_page for direct compaction
+		 * altogether.
+		 */
+		if (current->flags & PF_MEMALLOC)
+			goto move_newpage;
+		lock_page(page);
+	}
+
+	/*
+	 * Only memory hotplug's offline_pages() caller has locked out KSM,
+	 * and can safely migrate a KSM page.  The other cases have skipped
+	 * PageKsm along with PageReserved - but it is only now when we have
+	 * the page lock that we can be certain it will not go KSM beneath us
+	 * (KSM will not upgrade a page from PageAnon to PageKsm when it sees
+	 * its pagecount raised, but only here do we take the page lock which
+	 * serializes that).
+	 */
+	if (PageKsm(page) && !offlining) {
+		rc = -EBUSY;
+		goto unlock;
+	}
+
+	/* charge against new page */
+	charge = mem_cgroup_prepare_migration(page, newpage, &mem, GFP_KERNEL);
+	if (charge == -ENOMEM) {
+		rc = -ENOMEM;
+		goto unlock;
+	}
+	BUG_ON(charge);
+
+	if (PageWriteback(page)) {
+		/*
+		 * For !sync, there is no point retrying as the retry loop
+		 * is expected to be too short for PageWriteback to be cleared
+		 */
+		if (!sync) {
+			rc = -EBUSY;
+			goto uncharge;
+		}
+		if (!force)
+			goto uncharge;
+		wait_on_page_writeback(page);
+	}
+	/*
+	 * By try_to_unmap(), page->mapcount goes down to 0 here. In this case,
+	 * we cannot notice that anon_vma is freed while we migrates a page.
+	 * This get_anon_vma() delays freeing anon_vma pointer until the end
+	 * of migration. File cache pages are no problem because of page_lock()
+	 * File Caches may use write_page() or lock_page() in migration, then,
+	 * just care Anon page here.
+	 */
+	if (PageAnon(page)) {
+		/*
+		 * Only page_lock_anon_vma() understands the subtleties of
+		 * getting a hold on an anon_vma from outside one of its mms.
+		 */
+		anon_vma = page_get_anon_vma(page);
+		if (anon_vma) {
+			/*
+			 * Anon page
+			 */
+		} else if (PageSwapCache(page)) {
+			/*
+			 * We cannot be sure that the anon_vma of an unmapped
+			 * swapcache page is safe to use because we don't
+			 * know in advance if the VMA that this page belonged
+			 * to still exists. If the VMA and others sharing the
+			 * data have been freed, then the anon_vma could
+			 * already be invalid.
+			 *
+			 * To avoid this possibility, swapcache pages get
+			 * migrated but are not remapped when migration
+			 * completes
+			 */
+			remap_swapcache = 0;
+		} else {
+			goto uncharge;
+		}
+	}
+
+	/*
+	 * Corner case handling:
+	 * 1. When a new swap-cache page is read into, it is added to the LRU
+	 * and treated as swapcache but it has no rmap yet.
+	 * Calling try_to_unmap() against a page->mapping==NULL page will
+	 * trigger a BUG.  So handle it here.
+	 * 2. An orphaned page (see truncate_complete_page) might have
+	 * fs-private metadata. The page can be picked up due to memory
+	 * offlining.  Everywhere else except page reclaim, the page is
+	 * invisible to the vm, so the page can not be migrated.  So try to
+	 * free the metadata, so the page can be freed.
+	 */
+	if (!page->mapping) {
+		VM_BUG_ON(PageAnon(page));
+		if (page_has_private(page)) {
+			try_to_free_buffers(page);
+			goto uncharge;
+		}
+		goto skip_unmap;
+	}
+
+	/* Establish migration ptes or remove ptes */
+	try_to_unmap(page, TTU_MIGRATION|TTU_IGNORE_MLOCK|TTU_IGNORE_ACCESS);
+
+skip_unmap:
+	if (!page_mapped(page))
+		rc = move_to_new_page(newpage, page, remap_swapcache, sync);
+
+	if (rc && remap_swapcache)
+		remove_migration_ptes(page, page);
+
+	/* Drop an anon_vma reference if we took one */
+	if (anon_vma)
+		put_anon_vma(anon_vma);
+
+uncharge:
+	if (!charge)
+		mem_cgroup_end_migration(mem, page, newpage, rc == 0);
+unlock:
+	unlock_page(page);
+
+move_newpage:
+	if (rc != -EAGAIN) {
+		/*
+		 * A page that has been migrated has all references
+		 * removed and will be freed. A page that has not been
+		 * migrated will have kepts its references and be
+		 * restored.
+		 */
+		migratelist_del(page, *prev_lru);
+		dec_zone_page_state(page, NR_ISOLATED_ANON +
+				page_is_file_cache(page));
+		/*
+		 * Unlike unmap_and_move, we defer putback page
+		 * after inorder handling. Because the page would
+		 * be freed so it doesn't have PG_lru. Then,
+		 * keep_lru_order doesn't work correctly.
+		 */
+		del = true;
+	}
+	else
+		*prev_lru = &page->ilru;
+
+	/*
+	 * Move the new page to the LRU. If migration was not successful
+	 * then this will free the page.
+	 */
+	zone = page_zone(page);
+	spin_lock_irq(&zone->lru_lock);
+	if (keep_lru_order(page, prev_page)) {
+		putback_page_to_lru(newpage, prev_page);
+		spin_unlock_irq(&zone->lru_lock);
+		/*
+		 * The newpage will replace LRU position of old page and
+		 * old one would be freed. So let's adjust prev_page of pages
+		 * remained in migratelist for keep_lru_order.
+		 */
+		adjust_inorder_prev_page(head, page, newpage);
+		put_page(newpage); /* drop ref from isolate */
+	}
+	else {
+
+		spin_unlock_irq(&zone->lru_lock);
+		putback_lru_page(newpage);
+	}
+
+	if (del)
+		putback_lru_page(page);
+
+	if (result) {
+		if (rc)
+			*result = rc;
+		else
+			*result = page_to_nid(newpage);
+	}
+	return rc;
+}
+
+
 /*
  * Counterpart of unmap_and_move_page() for hugepage migration.
  *
@@ -975,6 +1215,56 @@ out:
 	return nr_failed + retry;
 }
 
+int migrate_inorder_lru_pages(struct inorder_lru *head, new_page_t get_new_page,
+		unsigned long private, bool offlining, bool sync)
+{
+	int retry = 1;
+	int nr_failed = 0;
+	int pass = 0;
+	struct page *page, *page2;
+	struct inorder_lru *prev;
+	int swapwrite = current->flags & PF_SWAPWRITE;
+	int rc;
+
+	if (!swapwrite)
+		current->flags |= PF_SWAPWRITE;
+
+	for(pass = 0; pass < 10 && retry; pass++) {
+		retry = 0;
+		list_for_each_migrate_entry_safe(page, page2, head, ilru) {
+			cond_resched();
+
+			prev = head;
+			rc = unmap_and_move_inorder_lru(get_new_page, private,
+					page, pass > 2, offlining,
+					sync, &prev, head);
+
+			switch(rc) {
+				case -ENOMEM:
+					goto out;
+				case -EAGAIN:
+					retry++;
+					break;
+				case 0:
+					break;
+				default:
+					/* Permanent failure */
+					nr_failed++;
+					break;
+			}
+		}
+	}
+	rc = 0;
+out:
+	if (!swapwrite)
+		current->flags &= ~PF_SWAPWRITE;
+
+	if (rc)
+		return rc;
+
+	return nr_failed + retry;
+}
+
 int migrate_huge_pages(struct list_head *from,
 		new_page_t get_new_page, unsigned long private, bool offlining,
 		bool sync)
-- 
1.7.1

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH v1 09/10] compaction: make compaction use in-order putback

[Minchan Kim]

Compaction is good solution to get contiguous page but it makes
LRU churing which is not good.
This patch makes that compaction code use in-order putback so
after compaction completion, migrated pages are keeping LRU ordering.

Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 mm/compaction.c |   26 ++++++++++++++------------
 1 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/mm/compaction.c b/mm/compaction.c
index e218562..00e710a 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -28,7 +28,7 @@
  */
 struct compact_control {
 	struct list_head freepages;	/* List of free pages to migrate to */
-	struct list_head migratepages;	/* List of pages being migrated */
+	struct inorder_lru migratepages;/* List of pages being migrated */
 	unsigned long nr_freepages;	/* Number of isolated free pages */
 	unsigned long nr_migratepages;	/* Number of pages to migrate */
 	unsigned long free_pfn;		/* isolate_freepages search base */
@@ -210,7 +210,7 @@ static void acct_isolated(struct zone *zone, struct compact_control *cc)
 	struct page *page;
 	unsigned int count[2] = { 0, };
 
-	list_for_each_entry(page, &cc->migratepages, lru)
+	list_for_each_migrate_entry(page, &cc->migratepages, ilru)
 		count[!!page_is_file_cache(page)]++;
 
 	__mod_zone_page_state(zone, NR_ISOLATED_ANON, count[0]);
@@ -242,7 +242,7 @@ static unsigned long isolate_migratepages(struct zone *zone,
 	unsigned long low_pfn, end_pfn;
 	unsigned long last_pageblock_nr = 0, pageblock_nr;
 	unsigned long nr_scanned = 0, nr_isolated = 0;
-	struct list_head *migratelist = &cc->migratepages;
+	struct inorder_lru *migratelist = &cc->migratepages;
 	enum ISOLATE_PAGE_MODE mode = ISOLATE_BOTH;
 
 	/* Do not scan outside zone boundaries */
@@ -273,7 +273,7 @@ static unsigned long isolate_migratepages(struct zone *zone,
 	cond_resched();
 	spin_lock_irq(&zone->lru_lock);
 	for (; low_pfn < end_pfn; low_pfn++) {
-		struct page *page;
+		struct page *page, *prev_page;
 		bool locked = true;
 
 		/* give a chance to irqs before checking need_resched() */
@@ -330,14 +330,15 @@ static unsigned long isolate_migratepages(struct zone *zone,
 		/* Try isolate the page */
 		if (!cc->sync)
 			mode |= ISOLATE_CLEAN;
-		if (__isolate_lru_page(page, mode, 0) != 0)
+		if (__isolate_inorder_lru_page(page, mode, 0, &prev_page) != 0)
 			continue;
 
 		VM_BUG_ON(PageTransCompound(page));
 
 		/* Successfully isolated */
 		del_page_from_lru_list(zone, page, page_lru(page));
-		list_add(&page->lru, migratelist);
+		migratelist_add(page, prev_page, migratelist);
+
 		cc->nr_migratepages++;
 		nr_isolated++;
 
@@ -393,7 +394,7 @@ static void update_nr_listpages(struct compact_control *cc)
 	int nr_freepages = 0;
 	struct page *page;
 
-	list_for_each_entry(page, &cc->migratepages, lru)
+	list_for_each_migrate_entry(page, &cc->migratepages, ilru)
 		nr_migratepages++;
 	list_for_each_entry(page, &cc->freepages, lru)
 		nr_freepages++;
@@ -521,7 +522,8 @@ static int compact_zone(struct zone *zone, struct compact_control *cc)
 			continue;
 
 		nr_migrate = cc->nr_migratepages;
-		err = migrate_pages(&cc->migratepages, compaction_alloc,
+		err = migrate_inorder_lru_pages(&cc->migratepages,
+				compaction_alloc,
 				(unsigned long)cc, false,
 				cc->sync);
 		update_nr_listpages(cc);
@@ -536,7 +538,7 @@ static int compact_zone(struct zone *zone, struct compact_control *cc)
 
 		/* Release LRU pages not migrated */
 		if (err) {
-			putback_lru_pages(&cc->migratepages);
+			putback_inorder_lru_pages(&cc->migratepages);
 			cc->nr_migratepages = 0;
 		}
 
@@ -562,7 +564,7 @@ unsigned long compact_zone_order(struct zone *zone,
 		.sync = sync,
 	};
 	INIT_LIST_HEAD(&cc.freepages);
-	INIT_LIST_HEAD(&cc.migratepages);
+	INIT_MIGRATE_LIST(&cc.migratepages);
 
 	return compact_zone(zone, &cc);
 }
@@ -644,12 +646,12 @@ static int compact_node(int nid)
 
 		cc.zone = zone;
 		INIT_LIST_HEAD(&cc.freepages);
-		INIT_LIST_HEAD(&cc.migratepages);
+		INIT_MIGRATE_LIST(&cc.migratepages);
 
 		compact_zone(zone, &cc);
 
 		VM_BUG_ON(!list_empty(&cc.freepages));
-		VM_BUG_ON(!list_empty(&cc.migratepages));
+		VM_BUG_ON(!migratelist_empty(&cc.migratepages));
 	}
 
 	return 0;
-- 
1.7.1

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH v1 10/10] add tracepoints

[Minchan Kim]

This patch adds some tracepints for see the effect this patch
series.
This tracepoints isn't for merge but just see the effect.

Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 mm/compaction.c |    2 ++
 mm/migrate.c    |    7 +++++++
 mm/vmscan.c     |    3 +--
 3 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/mm/compaction.c b/mm/compaction.c
index 00e710a..92c180d 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -16,6 +16,7 @@
 #include <linux/sysfs.h>
 #include "internal.h"
 
+#include <trace/events/inorder_putback.h>
 #define CREATE_TRACE_POINTS
 #include <trace/events/compaction.h>
 
@@ -333,6 +334,7 @@ static unsigned long isolate_migratepages(struct zone *zone,
 		if (__isolate_inorder_lru_page(page, mode, 0, &prev_page) != 0)
 			continue;
 
+		trace_mm_compact_isolate(prev_page, page);
 		VM_BUG_ON(PageTransCompound(page));
 
 		/* Successfully isolated */
diff --git a/mm/migrate.c b/mm/migrate.c
index f94fe65..6a2eca9 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -39,6 +39,9 @@
 
 #include "internal.h"
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/inorder_putback.h>
+
 #define lru_to_page(_head) (list_entry((_head)->prev, struct page, lru))
 
 /*
@@ -96,10 +99,12 @@ void putback_inorder_lru_pages(struct inorder_lru *l)
 		spin_lock_irq(&zone->lru_lock);
 		prev = page->ilru.prev_page;
 		if (keep_lru_order(page, prev)) {
+			trace_mm_compaction_inorder(page, page);
 			putback_page_to_lru(page, prev);
 			spin_unlock_irq(&zone->lru_lock);
 		}
 		else {
+			trace_mm_compaction_outoforder(page, page);
 			spin_unlock_irq(&zone->lru_lock);
 			putback_lru_page(page);
 		}
@@ -1052,6 +1057,7 @@ move_newpage:
 	if (keep_lru_order(page, prev_page)) {
 		putback_page_to_lru(newpage, prev_page);
 		spin_unlock_irq(&zone->lru_lock);
+		trace_mm_compaction_inorder(page, newpage);
 		/*
 		 * The newpage will replace LRU position of old page and
 		 * old one would be freed. So let's adjust prev_page of pages
@@ -1063,6 +1069,7 @@ move_newpage:
 	else {
 
 		spin_unlock_irq(&zone->lru_lock);
+		trace_mm_compaction_inorder(page, newpage);
 		putback_lru_page(newpage);
 	}
 
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 62d5186..2e7cbad 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -49,10 +49,9 @@
 #include <linux/swapops.h>
 
 #include "internal.h"
-
+#include <trace/events/inorder_putback.h>
 #define CREATE_TRACE_POINTS
 #include <trace/events/vmscan.h>
-
 /*
  * reclaim_mode determines how the inactive list is shrunk
  * RECLAIM_MODE_SINGLE: Reclaim only order-0 pages
-- 
1.7.1

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH v1 00/10] Prevent LRU churning

[Minchan Kim]

[-- Attachment #1: Type: text/plain, Size: 77 bytes --]

I missed script used for testing.
Attached.



-- 
Kind regards,
Minchan Kim

[-- Attachment #2: run-many-x-apps.sh --]
[-- Type: application/x-sh, Size: 1345 bytes --]