[Patch v4 2/6] sched: Add infrastructure to store and update instantaneous thermal pressure

[Thara Gopinath]

Add thermal.c and thermal.h files that provides interface
APIs to initialize, update/average, track, accumulate and decay
thermal pressure per cpu basis. A per cpu variable delta_capacity is
introduced to keep track of instantaneous per cpu thermal pressure.
Thermal pressure is the delta between maximum capacity and capped
capacity due to a thermal event.
API trigger_thermal_pressure_average is called for periodic accumulate
and decay of the thermal pressure. It is to to be called from a
periodic tick function. This API passes on the instantaneous delta
capacity of a cpu to update_thermal_load_avg to do the necessary
accumulate, decay and average. 
API update_thermal_pressure is for the system to update the thermal
pressure by providing a capped frequency ratio.
Considering, trigger_thermal_pressure_average reads delta_capacity and
update_thermal_pressure writes into delta_capacity, one can argue for
some sort of locking mechanism to avoid a stale value.
But considering trigger_thermal_pressure_average can be called from a
system critical path like scheduler tick function, a locking mechanism
is not ideal. This means that it is possible the delta_capacity value
used to calculate average thermal pressure for a cpu can be
stale for upto 1 tick period.

Signed-off-by: Thara Gopinath <thara.gopinath@xxxxxxxxxx>
---
v3->v4:
	- Dropped per cpu max_capacity_info struct and instead added a per
	  delta_capacity variable to store the delta between maximum
	  capacity and capped capacity. The delta is now calculated when
	  thermal pressure is updated and not every tick.
	- Dropped populate_max_capacity_info api as only per cpu delta
	  capacity is stored.
	- Renamed update_periodic_maxcap to
	  trigger_thermal_pressure_average and update_maxcap_capacity to
	  update_thermal_pressure.

 include/linux/sched.h  |  8 ++++++++
 kernel/sched/Makefile  |  2 +-
 kernel/sched/thermal.c | 45 +++++++++++++++++++++++++++++++++++++++++++++
 kernel/sched/thermal.h | 13 +++++++++++++
 4 files changed, 67 insertions(+), 1 deletion(-)
 create mode 100644 kernel/sched/thermal.c
 create mode 100644 kernel/sched/thermal.h

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2c2e56b..d7ef543 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1983,6 +1983,14 @@ static inline void rseq_syscall(struct pt_regs *regs)
 
 #endif
 
+#ifdef CONFIG_SMP
+void update_thermal_pressure(int cpu, u64 capacity);
+#else
+static inline void update_thermal_pressure(int cpu, u64 capacity)
+{
+}
+#endif
+
 const struct sched_avg *sched_trace_cfs_rq_avg(struct cfs_rq *cfs_rq);
 char *sched_trace_cfs_rq_path(struct cfs_rq *cfs_rq, char *str, int len);
 int sched_trace_cfs_rq_cpu(struct cfs_rq *cfs_rq);
diff --git a/kernel/sched/Makefile b/kernel/sched/Makefile
index 21fb5a5..4d3b820 100644
--- a/kernel/sched/Makefile
+++ b/kernel/sched/Makefile
@@ -20,7 +20,7 @@ obj-y += core.o loadavg.o clock.o cputime.o
 obj-y += idle.o fair.o rt.o deadline.o
 obj-y += wait.o wait_bit.o swait.o completion.o
 
-obj-$(CONFIG_SMP) += cpupri.o cpudeadline.o topology.o stop_task.o pelt.o
+obj-$(CONFIG_SMP) += cpupri.o cpudeadline.o topology.o stop_task.o pelt.o thermal.o
 obj-$(CONFIG_SCHED_AUTOGROUP) += autogroup.o
 obj-$(CONFIG_SCHEDSTATS) += stats.o
 obj-$(CONFIG_SCHED_DEBUG) += debug.o
diff --git a/kernel/sched/thermal.c b/kernel/sched/thermal.c
new file mode 100644
index 0000000..0c84960
--- /dev/null
+++ b/kernel/sched/thermal.c
@@ -0,0 +1,45 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Scheduler Thermal Interactions
+ *
+ *  Copyright (C) 2018 Linaro, Inc., Thara Gopinath <thara.gopinath@xxxxxxxxxx>
+ */
+
+#include <linux/sched.h>
+#include "sched.h"
+#include "pelt.h"
+#include "thermal.h"
+
+static DEFINE_PER_CPU(unsigned long, delta_capacity);
+
+/**
+ * update_thermal_pressure: Update thermal pressure
+ * @cpu: the cpu for which thermal pressure is to be updated for
+ * @capped_freq_ratio: capped max frequency << SCHED_CAPACITY_SHIFT / max freq
+ *
+ * capped_freq_ratio is normalized into capped capacity and the delta between
+ * the arch_scale_cpu_capacity and capped capacity is stored in per cpu
+ * delta_capacity.
+ */
+void update_thermal_pressure(int cpu, u64 capped_freq_ratio)
+{
+	unsigned long __capacity, delta;
+
+	/* Normalize the capped freq ratio */
+	__capacity = (capped_freq_ratio * arch_scale_cpu_capacity(cpu)) >>
+							SCHED_CAPACITY_SHIFT;
+	delta = arch_scale_cpu_capacity(cpu) -  __capacity;
+	pr_debug("updating cpu%d thermal pressure to %lu\n", cpu, delta);
+
+	per_cpu(delta_capacity, cpu) = delta;
+}
+
+/**
+ * trigger_thermal_pressure_average: Trigger the thermal pressure accumulate
+ *				     and average algorithm
+ */
+void trigger_thermal_pressure_average(struct rq *rq)
+{
+	update_thermal_load_avg(rq_clock_task(rq), rq,
+				per_cpu(delta_capacity, cpu_of(rq)));
+}
diff --git a/kernel/sched/thermal.h b/kernel/sched/thermal.h
new file mode 100644
index 0000000..26e5b07
--- /dev/null
+++ b/kernel/sched/thermal.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Scheduler thermal interaction internal methods.
+ */
+
+#ifdef CONFIG_SMP
+void trigger_thermal_pressure_average(struct rq *rq);
+
+#else
+static inline void trigger_thermal_pressure_average(struct rq *rq)
+{
+}
+#endif
-- 
2.1.4