mm: avoid false sharing of mm_counter

Considering the nature of per mm stats, it's the shared object among
threads and can be a cache-miss point in the page fault path.

This patch adds per-thread cache for mm_counter.  RSS value will be
counted into a struct in task_struct and synchronized with mm's one at
events.

Now, in this patch, the event is the number of calls to handle_mm_fault.
Per-thread value is added to mm at each 64 calls.

 rough estimation with small benchmark on parallel thread (2threads) shows
 [before]
     4.5 cache-miss/faults
 [after]
     4.0 cache-miss/faults
 Anyway, the most contended object is mmap_sem if the number of threads grows.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

3 files changed, 12 insertions, 6 deletions

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 2124cdb2d1d0..8e580c07d171 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -873,7 +873,7 @@ int __get_user_pages_fast(unsigned long start, int nr_pages, int write,
 /*
  * per-process(per-mm_struct) statistics.
  */
-#if USE_SPLIT_PTLOCKS
+#if defined(SPLIT_RSS_COUNTING)
 /*
  * The mm counters are not protected by its page_table_lock,
  * so must be incremented atomically.
@@ -883,10 +883,7 @@ static inline void set_mm_counter(struct mm_struct *mm, int member, long value)
         atomic_long_set(&mm->rss_stat.count[member], value);
 }
 
-static inline unsigned long get_mm_counter(struct mm_struct *mm, int member)
-{
-        return (unsigned long)atomic_long_read(&mm->rss_stat.count[member]);
-}
+unsigned long get_mm_counter(struct mm_struct *mm, int member);
 
 static inline void add_mm_counter(struct mm_struct *mm, int member, long value)
 {
@@ -974,6 +971,7 @@ static inline void setmax_mm_hiwater_rss(unsigned long *maxrss,
                 *maxrss = hiwater_rss;
 }
 
+void sync_mm_rss(struct task_struct *task, struct mm_struct *mm);
 
 /*
  * A callback you can register to apply pressure to ageable caches.
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index e1ca64be6678..21861239ab0c 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -202,9 +202,15 @@ enum {
 };
 
 #if USE_SPLIT_PTLOCKS
+#define SPLIT_RSS_COUNTING
 struct mm_rss_stat {
         atomic_long_t count[NR_MM_COUNTERS];
 };
+/* per-thread cached information, */
+struct task_rss_stat {
+        int events;     /* for synchronization threshold */
+        int count[NR_MM_COUNTERS];
+};
 #else  /* !USE_SPLIT_PTLOCKS */
 struct mm_rss_stat {
         unsigned long count[NR_MM_COUNTERS];
diff --git a/include/linux/sched.h b/include/linux/sched.h
index cbeafa49a53b..46c6f8d5dc06 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1220,7 +1220,9 @@ struct task_struct {
         struct plist_node pushable_tasks;
 
         struct mm_struct *mm, *active_mm;
-
+#if defined(SPLIT_RSS_COUNTING)
+        struct task_rss_stat    rss_stat;
+#endif
 /* task state */
         int exit_state;
         int exit_code, exit_signal;