oom: make oom_score to per-process value

oom-killer kills a process, not task.  Then oom_score should be calculated
as per-process too.  it makes consistency more and makes speed up
select_bad_process().
Signed-off-by: NKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Paul Menage <menage@google.com>
Cc: David Rientjes <rientjes@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Documentation/filesystems/proc.txt

@@ -1205,7 +1205,7 @@ The following heuristics are then applied:
  * if the task was reniced, its score doubles
  * superuser or direct hardware access tasks (CAP_SYS_ADMIN, CAP_SYS_RESOURCE
  	or CAP_SYS_RAWIO) have their score divided by 4
- * if oom condition happened in one cpuset and checked task does not belong
+ * if oom condition happened in one cpuset and checked process does not belong
  	to it, its score is divided by 8
  * the resulting score is multiplied by two to the power of oom_adj, i.e.
 	points <<= oom_adj when it is positive and

fs/proc/base.c

@@ -447,7 +447,7 @@ static int proc_oom_score(struct task_struct *task, char *buffer)
 
 	do_posix_clock_monotonic_gettime(&uptime);
 	read_lock(&tasklist_lock);
-	points = badness(task, uptime.tv_sec);
+	points = badness(task->group_leader, uptime.tv_sec);
 	read_unlock(&tasklist_lock);
 	return sprintf(buffer, "%lu\n", points);
 }

mm/oom_kill.c

@@ -34,6 +34,23 @@ int sysctl_oom_dump_tasks;
 static DEFINE_SPINLOCK(zone_scan_lock);
 /* #define DEBUG */
 
+/*
+ * Is all threads of the target process nodes overlap ours?
+ */
+static int has_intersects_mems_allowed(struct task_struct *tsk)
+{
+	struct task_struct *t;
+
+	t = tsk;
+	do {
+		if (cpuset_mems_allowed_intersects(current, t))
+			return 1;
+		t = next_thread(t);
+	} while (t != tsk);
+
+	return 0;
+}
+
 /**
  * badness - calculate a numeric value for how bad this task has been
  * @p: task struct of which task we should calculate
@@ -59,6 +76,9 @@ unsigned long badness(struct task_struct *p, unsigned long uptime)
 	struct mm_struct *mm;
 	struct task_struct *child;
 	int oom_adj = p->signal->oom_adj;
+	struct task_cputime task_time;
+	unsigned long utime;
+	unsigned long stime;
 
 	if (oom_adj == OOM_DISABLE)
 		return 0;
@@ -106,8 +126,11 @@ unsigned long badness(struct task_struct *p, unsigned long uptime)
          * of seconds. There is no particular reason for this other than
          * that it turned out to work very well in practice.
 	 */
-	cpu_time = (cputime_to_jiffies(p->utime) + cputime_to_jiffies(p->stime))
-		>> (SHIFT_HZ + 3);
+	thread_group_cputime(p, &task_time);
+	utime = cputime_to_jiffies(task_time.utime);
+	stime = cputime_to_jiffies(task_time.stime);
+	cpu_time = (utime + stime) >> (SHIFT_HZ + 3);
+
 
 	if (uptime >= p->start_time.tv_sec)
 		run_time = (uptime - p->start_time.tv_sec) >> 10;
@@ -148,7 +171,7 @@ unsigned long badness(struct task_struct *p, unsigned long uptime)
 	 * because p may have allocated or otherwise mapped memory on
 	 * this node before. However it will be less likely.
 	 */
-	if (!cpuset_mems_allowed_intersects(current, p))
+	if (!has_intersects_mems_allowed(p))
 		points /= 8;
 
 	/*
@@ -204,13 +227,13 @@ static inline enum oom_constraint constrained_alloc(struct zonelist *zonelist,
 static struct task_struct *select_bad_process(unsigned long *ppoints,
 						struct mem_cgroup *mem)
 {
-	struct task_struct *g, *p;
+	struct task_struct *p;
 	struct task_struct *chosen = NULL;
 	struct timespec uptime;
 	*ppoints = 0;
 
 	do_posix_clock_monotonic_gettime(&uptime);
-	do_each_thread(g, p) {
+	for_each_process(p) {
 		unsigned long points;
 
 		/*
@@ -263,7 +286,7 @@ static struct task_struct *select_bad_process(unsigned long *ppoints,
 			chosen = p;
 			*ppoints = points;
 		}
-	} while_each_thread(g, p);
+	}
 
 	return chosen;
 }