oom: /proc/<pid>/oom_score treat kernel thread honestly

If a kernel thread is using use_mm(), badness() returns a positive value.
This is not a big issue because caller take care of it correctly.  But
there is one exception, /proc/<pid>/oom_score calls badness() directly and
doesn't care that the task is a regular process.

Another example, /proc/1/oom_score return !0 value.  But it's unkillable.
This incorrectness makes administration a little confusing.

This patch fixes it.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: David Rientjes <rientjes@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This commit is contained in:
KOSAKI Motohiro 2010-08-09 17:19:37 -07:00 committed by Linus Torvalds
parent f88ccad588
commit 26ebc98491
2 changed files with 10 additions and 8 deletions

View File

@ -428,7 +428,8 @@ static const struct file_operations proc_lstats_operations = {
#endif #endif
/* The badness from the OOM killer */ /* The badness from the OOM killer */
unsigned long badness(struct task_struct *p, unsigned long uptime); unsigned long badness(struct task_struct *p, struct mem_cgroup *mem,
nodemask_t *nodemask, unsigned long uptime);
static int proc_oom_score(struct task_struct *task, char *buffer) static int proc_oom_score(struct task_struct *task, char *buffer)
{ {
unsigned long points = 0; unsigned long points = 0;
@ -437,7 +438,7 @@ static int proc_oom_score(struct task_struct *task, char *buffer)
do_posix_clock_monotonic_gettime(&uptime); do_posix_clock_monotonic_gettime(&uptime);
read_lock(&tasklist_lock); read_lock(&tasklist_lock);
if (pid_alive(task)) if (pid_alive(task))
points = badness(task, uptime.tv_sec); points = badness(task, NULL, NULL, uptime.tv_sec);
read_unlock(&tasklist_lock); read_unlock(&tasklist_lock);
return sprintf(buffer, "%lu\n", points); return sprintf(buffer, "%lu\n", points);
} }

View File

@ -139,8 +139,8 @@ static bool oom_unkillable_task(struct task_struct *p, struct mem_cgroup *mem,
* algorithm has been meticulously tuned to meet the principle * algorithm has been meticulously tuned to meet the principle
* of least surprise ... (be careful when you change it) * of least surprise ... (be careful when you change it)
*/ */
unsigned long badness(struct task_struct *p, struct mem_cgroup *mem,
unsigned long badness(struct task_struct *p, unsigned long uptime) const nodemask_t *nodemask, unsigned long uptime)
{ {
unsigned long points, cpu_time, run_time; unsigned long points, cpu_time, run_time;
struct task_struct *child; struct task_struct *child;
@ -150,6 +150,8 @@ unsigned long badness(struct task_struct *p, unsigned long uptime)
unsigned long utime; unsigned long utime;
unsigned long stime; unsigned long stime;
if (oom_unkillable_task(p, mem, nodemask))
return 0;
if (oom_adj == OOM_DISABLE) if (oom_adj == OOM_DISABLE)
return 0; return 0;
@ -351,7 +353,7 @@ static struct task_struct *select_bad_process(unsigned long *ppoints,
if (p->signal->oom_adj == OOM_DISABLE) if (p->signal->oom_adj == OOM_DISABLE)
continue; continue;
points = badness(p, uptime.tv_sec); points = badness(p, mem, nodemask, uptime.tv_sec);
if (points > *ppoints || !chosen) { if (points > *ppoints || !chosen) {
chosen = p; chosen = p;
*ppoints = points; *ppoints = points;
@ -482,11 +484,10 @@ static int oom_kill_process(struct task_struct *p, gfp_t gfp_mask, int order,
if (child->mm == p->mm) if (child->mm == p->mm)
continue; continue;
if (oom_unkillable_task(p, mem, nodemask))
continue;
/* badness() returns 0 if the thread is unkillable */ /* badness() returns 0 if the thread is unkillable */
child_points = badness(child, uptime.tv_sec); child_points = badness(child, mem, nodemask,
uptime.tv_sec);
if (child_points > victim_points) { if (child_points > victim_points) {
victim = child; victim = child;
victim_points = child_points; victim_points = child_points;