sched: Don't scan all-offline ->cpus_allowed twice if !CONFIG_CPUSETS
authorOleg Nesterov <[email protected]>
Sat, 10 Oct 2015 18:53:15 +0000 (20:53 +0200)
committerIngo Molnar <[email protected]>
Tue, 20 Oct 2015 08:25:57 +0000 (10:25 +0200)
If CONFIG_CPUSETS=n then "case cpuset" changes the state and runs
the already failed for_each_cpu() loop again for no reason.

Signed-off-by: Oleg Nesterov <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Cc: Andrew Morton <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Mike Galbraith <[email protected]>
Cc: Paul E. McKenney <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Tejun Heo <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Vitaly Kuznetsov <[email protected]>
Cc: [email protected]
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>
kernel/sched/core.c

index a7b368e51f69ab763e617fddc3f403a0a39ee487..b4d263db52a6f6d7b09dab05508c91fb526af929 100644 (file)
@@ -1580,13 +1580,15 @@ static int select_fallback_rq(int cpu, struct task_struct *p)
                        goto out;
                }
 
+               /* No more Mr. Nice Guy. */
                switch (state) {
                case cpuset:
-                       /* No more Mr. Nice Guy. */
-                       cpuset_cpus_allowed_fallback(p);
-                       state = possible;
-                       break;
-
+                       if (IS_ENABLED(CONFIG_CPUSETS)) {
+                               cpuset_cpus_allowed_fallback(p);
+                               state = possible;
+                               break;
+                       }
+                       /* fall-through */
                case possible:
                        do_set_cpus_allowed(p, cpu_possible_mask);
                        state = fail;