sched: Fix sign under-flows in wake_affine

While care is taken around the zero-point in effective_load to not exceed
the instantaneous rq->weight, it's still possible (e.g. using wake_idx != 0)
for (load + effective_load) to underflow.

In this case the comparing the unsigned values can result in incorrect balanced
decisions.

Signed-off-by: Paul Turner <pjt@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20110122044851.734245014@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 354769979c021cd6392714235043ee4dfcdb89ea..ccecfec02a7039bd1360f734508bbe6a807594b6 100644 (file)
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1432,7 +1432,7 @@ static inline unsigned long effective_load(struct task_group *tg, int cpu,
 
 static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
 {
-       unsigned long this_load, load;
+       s64 this_load, load;
        int idx, this_cpu, prev_cpu;
        unsigned long tl_per_task;
        struct task_group *tg;
@@ -1471,8 +1471,8 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
         * Otherwise check if either cpus are near enough in load to allow this
         * task to be woken on this_cpu.
         */
-       if (this_load) {
-               unsigned long this_eff_load, prev_eff_load;
+       if (this_load > 0) {
+               s64 this_eff_load, prev_eff_load;
 
                this_eff_load = 100;
                this_eff_load *= power_of(prev_cpu);