Thread (66 messages) 66 messages, 8 authors, 2014-07-18

[PATCH v3 04/12] sched: Allow all archs to set the power_orig

From: peterz@infradead.org (Peter Zijlstra)
Date: 2014-07-09 10:57:39
Also in: lkml

On Mon, Jun 30, 2014 at 06:05:35PM +0200, Vincent Guittot wrote:
quoted hunk ↗ jump to hunk
power_orig is only changed for system with a SMT sched_domain level in order to
reflect the lower capacity of CPUs. Heterogenous system also have to reflect an
original capacity that is different from the default value.

Create a more generic function arch_scale_cpu_power that can be also used by
non SMT platform to set power_orig.

The weak behavior of arch_scale_cpu_power is the previous SMT one in order to
keep backward compatibility in the use of power_orig.

Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
---
 kernel/sched/fair.c | 24 ++++++++++++++++--------
 1 file changed, 16 insertions(+), 8 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 148b277..f0bba5f 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5614,6 +5614,20 @@ unsigned long __weak arch_scale_smt_capacity(struct sched_domain *sd, int cpu)
 	return default_scale_smt_capacity(sd, cpu);
 }
 
+unsigned long __weak arch_scale_cpu_capacity(struct sched_domain *sd, int cpu)
+{
+	unsigned long weight = sd->span_weight;
+
+	if ((sd->flags & SD_SHARE_CPUCAPACITY) && weight > 1) {
+		if (sched_feat(ARCH_CAPACITY))
+			return arch_scale_smt_capacity(sd, cpu);
+		else
+			return default_scale_smt_capacity(sd, cpu);
+	}
+
+	return SCHED_CAPACITY_SCALE;
+}
The only caller of arch_scale_smt_capacity is now an arch_ function
itself; which makes it entirely redundant, no?

Also, sched_feat() and default_scale_smt_capability() aren't available
to arch implementations of this function and can thus not retain
semantics.

Seeing how we currently don't have any arch_scale_smt_capacity
implementations other than the default we can easily replace it entirely
with something like:

unsigned long __weak arch_scale_cpu_capacity()
{
	if ((sd->flags & SD_SHARE_CPUCAPACITY) && (sd->span_weight > 1))
		return sd->smt_gain / sd->span_weight;

	return SCHED_CAPACITY_SCALE;
}

Hmm?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20140709/56f24bf2/attachment.sig>
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help