* Siddha, Suresh B <[EMAIL PROTECTED]> wrote:

> nr_moved is not the correct check for triggering all pinned logic. Fix 
> the all pinned logic in the case of load_balance_newidle().

indeeed.

i've applied your patch to my queue. (I've done a trivial style change 
to it, see below.)

        Ingo

--------------------->
Subject: sched: fix the all pinned logic in load_balance_newidle()
From: Suresh Siddha <[EMAIL PROTECTED]>

nr_moved is not the correct check for triggering all pinned logic. Fix
the all pinned logic in the case of load_balance_newidle().

Signed-off-by: Suresh Siddha <[EMAIL PROTECTED]>
Signed-off-by: Ingo Molnar <[EMAIL PROTECTED]>
---
 kernel/sched.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

Index: linux/kernel/sched.c
===================================================================
--- linux.orig/kernel/sched.c
+++ linux/kernel/sched.c
@@ -2679,6 +2679,7 @@ load_balance_newidle(int this_cpu, struc
        unsigned long imbalance;
        int nr_moved = 0;
        int sd_idle = 0;
+       int all_pinned = 0;
        cpumask_t cpus = CPU_MASK_ALL;
 
        /*
@@ -2717,10 +2718,11 @@ redo:
                double_lock_balance(this_rq, busiest);
                nr_moved = move_tasks(this_rq, this_cpu, busiest,
                                        minus_1_or_zero(busiest->nr_running),
-                                       imbalance, sd, CPU_NEWLY_IDLE, NULL);
+                                       imbalance, sd, CPU_NEWLY_IDLE,
+                                       &all_pinned);
                spin_unlock(&busiest->lock);
 
-               if (!nr_moved) {
+               if (unlikely(all_pinned)) {
                        cpu_clear(cpu_of(busiest), cpus);
                        if (!cpus_empty(cpus))
                                goto redo;
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Reply via email to