Skip to contentSkip to conversation

powernv: cpufreq support for IBM POWERNV platform

PATCH In review Vaidyanathan Srinivasan 2 participants 5 messages Feb 11, 2014

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

cover powernv: cpufreq support for IBM POWERNV platform

Hi,

The following patch series implements the platform driver to support dynamic CPU frequency scaling on IBM POWERNV platforms.

This patch series is based on Linux kernel 3.14-rc2 and tested on OPAL v3 based IBM POWERNV platform and IBM POWER8 processor.

--Vaidy

---

Srivatsa S. Bhat (1):
powernv, cpufreq: Add per-core locking to serialize frequency transitions

Vaidyanathan Srinivasan (1):
powernv: cpufreq driver for powernv platform

 arch/powerpc/include/asm/reg.h    |    4 +
 drivers/cpufreq/Kconfig.powerpc   |    9 +
 drivers/cpufreq/Makefile          |    1 
 drivers/cpufreq/powernv-cpufreq.c |  286 +++++++++++++++++++++++++++++++++++++
 4 files changed, 300 insertions(+)
 create mode 100644 drivers/cpufreq/powernv-cpufreq.c
#1 lore raw

PATCH 1/2 powernv: cpufreq driver for powernv platform

Backend driver to dynamically set voltage and frequency on IBM POWER non-virtualized platforms. Power management SPRs are used to set the required PState.

This driver works in conjunction with cpufreq governors like 'ondemand' to provide a demand based frequency and voltage setting on IBM POWER non-virtualized platforms.

PState table is obtained from OPAL v3 firmware through device tree.

powernv_cpufreq back-end driver would parse the relevant device-tree nodes and initialise the cpufreq subsystem on powernv platform.

---
 arch/powerpc/include/asm/reg.h    |    4 +
 drivers/cpufreq/Kconfig.powerpc   |    9 +
 drivers/cpufreq/Makefile          |    1 
 drivers/cpufreq/powernv-cpufreq.c |  275 +++++++++++++++++++++++++++++++++++++
 4 files changed, 289 insertions(+)
 create mode 100644 drivers/cpufreq/powernv-cpufreq.c
arch/powerpc/include/asm/reg.h +4 -0
index 90c06ec..84f92ca 100644
@@ -271,6 +271,10 @@
 #define SPRN_HSRR1	0x13B	/* Hypervisor Save/Restore 1 */
 #define SPRN_IC		0x350	/* Virtual Instruction Count */
 #define SPRN_VTB	0x351	/* Virtual Time Base */
+#define SPRN_PMICR	0x354   /* Power Management Idle Control Reg */
+#define SPRN_PMSR	0x355   /* Power Management Status Reg */
+#define SPRN_PMCR	0x374	/* Power Management Control Register */
+
 /* HFSCR and FSCR bit numbers are the same */
 #define FSCR_TAR_LG	8	/* Enable Target Address Register */
 #define FSCR_EBB_LG	7	/* Enable Event Based Branching */
drivers/cpufreq/Kconfig.powerpc +9 -0
index ca0021a..4a91ab1 100644
@@ -54,3 +54,12 @@ config PPC_PASEMI_CPUFREQ
 	help
 	  This adds the support for frequency switching on PA Semi
 	  PWRficient processors.
+
+config POWERNV_CPUFREQ
+       tristate "CPU frequency scaling for IBM POWERNV platform"
+       depends on PPC_POWERNV
+       select CPU_FREQ_TABLE
+       default y
+       help
+	 This adds support for CPU frequency switching on IBM POWERNV
+	 platform
drivers/cpufreq/Makefile +1 -0
index 7494565..0dbb963 100644
@@ -86,6 +86,7 @@ obj-$(CONFIG_PPC_CORENET_CPUFREQ)   += ppc-corenet-cpufreq.o
 obj-$(CONFIG_CPU_FREQ_PMAC)		+= pmac32-cpufreq.o
 obj-$(CONFIG_CPU_FREQ_PMAC64)		+= pmac64-cpufreq.o
 obj-$(CONFIG_PPC_PASEMI_CPUFREQ)	+= pasemi-cpufreq.o
+obj-$(CONFIG_POWERNV_CPUFREQ)		+= powernv-cpufreq.o
 
 ##################################################################################
 # Other platform drivers
drivers/cpufreq/powernv-cpufreq.c new file +275 -0

275 lines not on this page · raw message

#2 lore raw

Re: [PATCH v1 1/2] powernv: cpufreq driver for powernv platform

Hi Vaidy,

Quoted 18 lines · Vaidyanathan Srinivasan

Backend driver to dynamically set voltage and frequency on IBM POWER non-virtualized platforms. Power management SPRs are used to set the required PState.

This driver works in conjunction with cpufreq governors like 'ondemand' to provide a demand based frequency and voltage setting on IBM POWER non-virtualized platforms.

PState table is obtained from OPAL v3 firmware through device tree.

powernv_cpufreq back-end driver would parse the relevant device-tree nodes and initialise the cpufreq subsystem on powernv platform.

---

<snip>

drivers/cpufreq/powernv-cpufreq.c line 185

Quoted 11 lines
drivers/cpufreq/powernv-cpufreq.c new file +11 -0
+static int powernv_cpufreq_cpu_init(struct cpufreq_policy *policy)
+{
+	int base, i;
+
+#ifdef CONFIG_SMP
+	base = cpu_first_thread_sibling(policy->cpu);
+
+	for (i = 0; i < threads_per_core; i++)
+		cpumask_set_cpu(base + i, policy->cpus);
+#endif
+	policy->cpuinfo.transition_latency = 25000;

Is it ok to hard code this field? How about getting this also from the device tree?

drivers/cpufreq/powernv-cpufreq.c line 189

Quoted 4 lines
drivers/cpufreq/powernv-cpufreq.c new file +4 -0
+
+	/* Print frequency table */
+	for (i = 0; powernv_freqs[i].frequency != CPUFREQ_TABLE_END; i++)
+		pr_debug("%d: %d\n", i, powernv_freqs[i].frequency);

The frequency table as a result will be printed on every cpu when cpufreq gets initialized. Considering this information will not vary across CPUs, can we print this during powernv_cpufreq_init() after parsing the device tree for the pstates?

Thanks

Regards
Preeti U Murthy

#3 lore raw

PATCH 2/2 powernv, cpufreq: Add per-core locking to serialize frequency transitions

In reply to Vaidyanathan Srinivasan

From: Srivatsa S. Bhat <[email protected]>

On POWER systems, the CPU frequency is controlled at a core-level and hence we need to serialize so that only one of the threads in the core switches the core's frequency at a time.

Using a global mutex lock would needlessly serialize _all_ frequency transitions in the system (across all cores). So introduce per-core locking to enable finer-grained synchronization and thereby enhance the speed and responsiveness of the cpufreq driver to varying workload demands.

The design of per-core locking is very simple and straight-forward: we first define a Per-CPU lock and use the ones that belongs to the first thread sibling of the core.

cpu_first_thread_sibling() macro is used to find the *common* lock for all thread siblings belonging to a core.

---
 drivers/cpufreq/powernv-cpufreq.c |   21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)
drivers/cpufreq/powernv-cpufreq.c +16 -5
index ea3b630..8240e90 100644
@@ -24,8 +24,15 @@
 #include <linux/of.h>
 #include <asm/cputhreads.h>
 
-/* FIXME: Make this per-core */
-static DEFINE_MUTEX(freq_switch_mutex);
+/* Per-Core locking for frequency transitions */
+static DEFINE_PER_CPU(struct mutex, freq_switch_lock);
+
+#define lock_core_freq(cpu)				\
+			mutex_lock(&per_cpu(freq_switch_lock,\
+				cpu_first_thread_sibling(cpu)));
+#define unlock_core_freq(cpu)				\
+			mutex_unlock(&per_cpu(freq_switch_lock,\
+				cpu_first_thread_sibling(cpu)));
 
 #define POWERNV_MAX_PSTATES	256
 
@@ -219,7 +226,7 @@ static int powernv_cpufreq_target(struct cpufreq_policy *policy,
 	freqs.new = powernv_freqs[new_index].frequency;
 	freqs.cpu = policy->cpu;
 
-	mutex_lock(&freq_switch_mutex);
+	lock_core_freq(policy->cpu);
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 
 	pr_debug("setting frequency for cpu %d to %d kHz index %d pstate %d",
@@ -231,7 +238,7 @@ static int powernv_cpufreq_target(struct cpufreq_policy *policy,
 	rc = powernv_set_freq(policy->cpus, new_index);
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
-	mutex_unlock(&freq_switch_mutex);
+	unlock_core_freq(policy->cpu);
 
 	return rc;
 }
@@ -248,7 +255,7 @@ static struct cpufreq_driver powernv_cpufreq_driver = {
 
 static int __init powernv_cpufreq_init(void)
 {
-	int rc = 0;
+	int cpu, rc = 0;
 
 	/* Discover pstates from device tree and init */
 
@@ -258,6 +265,10 @@ static int __init powernv_cpufreq_init(void)
 		pr_info("powernv-cpufreq disabled\n");
 		return rc;
 	}
+	/* Init per-core mutex */
+	for_each_possible_cpu(cpu) {
+		mutex_init(&per_cpu(freq_switch_lock, cpu));
+	}
 
 	rc = cpufreq_register_driver(&powernv_cpufreq_driver);
 	return rc;
Reviewed-by on v1 2/2
#4 lore raw

Re: [PATCH v1 2/2] powernv, cpufreq: Add per-core locking to serialize frequency transitions

Hi Vaidy,

Quoted 75 lines · Vaidyanathan Srinivasan

From: Srivatsa S. Bhat <[email protected]>

On POWER systems, the CPU frequency is controlled at a core-level and hence we need to serialize so that only one of the threads in the core switches the core's frequency at a time.

Using a global mutex lock would needlessly serialize _all_ frequency transitions in the system (across all cores). So introduce per-core locking to enable finer-grained synchronization and thereby enhance the speed and responsiveness of the cpufreq driver to varying workload demands.

The design of per-core locking is very simple and straight-forward: we first define a Per-CPU lock and use the ones that belongs to the first thread sibling of the core.

cpu_first_thread_sibling() macro is used to find the *common* lock for all thread siblings belonging to a core.

---
 drivers/cpufreq/powernv-cpufreq.c |   21 ++++++++++++++++-----
 1 file changed, 16 insertions(+), 5 deletions(-)
drivers/cpufreq/powernv-cpufreq.c +12 -5
index ea3b630..8240e90 100644
@@ -24,8 +24,15 @@
 #include <linux/of.h>
 #include <asm/cputhreads.h>
 
-/* FIXME: Make this per-core */
-static DEFINE_MUTEX(freq_switch_mutex);
+/* Per-Core locking for frequency transitions */
+static DEFINE_PER_CPU(struct mutex, freq_switch_lock);
+
+#define lock_core_freq(cpu)				\
+			mutex_lock(&per_cpu(freq_switch_lock,\
+				cpu_first_thread_sibling(cpu)));
+#define unlock_core_freq(cpu)				\
+			mutex_unlock(&per_cpu(freq_switch_lock,\
+				cpu_first_thread_sibling(cpu)));
 
 #define POWERNV_MAX_PSTATES	256
 
@@ -219,7 +226,7 @@ static int powernv_cpufreq_target(struct cpufreq_policy *policy,
 	freqs.new = powernv_freqs[new_index].frequency;
 	freqs.cpu = policy->cpu;
 
-	mutex_lock(&freq_switch_mutex);
+	lock_core_freq(policy->cpu);
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE);
 
 	pr_debug("setting frequency for cpu %d to %d kHz index %d pstate %d",
@@ -231,7 +238,7 @@ static int powernv_cpufreq_target(struct cpufreq_policy *policy,
 	rc = powernv_set_freq(policy->cpus, new_index);
 
 	cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE);
-	mutex_unlock(&freq_switch_mutex);
+	unlock_core_freq(policy->cpu);
 
 	return rc;
 }
@@ -248,7 +255,7 @@ static struct cpufreq_driver powernv_cpufreq_driver = {
 
 static int __init powernv_cpufreq_init(void)
 {
-	int rc = 0;
+	int cpu, rc = 0;
 
 	/* Discover pstates from device tree and init */
 

drivers/cpufreq/powernv-cpufreq.c static int __init powernv_cpufreq_init(void) · line 274

Quoted 10 lines
drivers/cpufreq/powernv-cpufreq.c +4 -0
@@ -258,6 +265,10 @@ static int __init powernv_cpufreq_init(void)
 		pr_info("powernv-cpufreq disabled\n");
 		return rc;
 	}
+	/* Init per-core mutex */
+	for_each_possible_cpu(cpu) {
+		mutex_init(&per_cpu(freq_switch_lock, cpu));
+	}
 
 	rc = cpufreq_register_driver(&powernv_cpufreq_driver);
 	return rc;

This looks good to me.

Thanks

Regards
Preeti U Murthy

Quoted 6 lines

_______________________________________________
Linuxppc-dev mailing list
[email protected]
https://lists.ozlabs.org/listinfo/linuxppc-dev