On 22 May 2013 17:46, Morten Rasmussen morten.rasmussen@arm.com wrote:
On Thu, Apr 25, 2013 at 06:23:29PM +0100, Vincent Guittot wrote:
@@ -4596,8 +4597,12 @@ static void update_cpu_power(struct sched_domain *sd, int cpu) cpu_rq(cpu)->cpu_available = power; sdg->sgp->power_available = power;
if (!is_my_buddy(cpu, cpu))power = 1;Using your fix for is_my_buddy() for patch 11 the above will always be true when sched_packing_mode is to anything else than SCHED_PACKING_FULL. So cpu_power for all cpus in SCHED_PACKING_{DEFAULT,NONE} is 1.
As far is I understand, this is not the intention?
yes, the too quick fix for 11/14 has not fixed anything... the final fix is available in git://git.linaro.org/people/vingu/kernel.git sched-pack-small-tasks-v4-fixed.
And it will be part of the next version that i prepare
Vincent
Cheers, Morten
cpu_rq(cpu)->cpu_power = power; sdg->sgp->power = power;
}
void update_group_power(struct sched_domain *sd, int cpu)
1.7.9.5