[PATCH v4 0/6] hwmon: (fam15h_power) Introduce an accumulated power reporting algorithm

[PATCH v4 0/6] hwmon: (fam15h_power) Introduce an accumulated power reporting algorithm

[Huang Rui]

Hi Guenter,

This serial of patches introduces an accumulated power reporting
algorithm. It will calculate the average power consumption for the
processor. The cpu feature flag is CPUID.8000_0007H:EDX[12].

This algorithm is used to test the comparison of processor power
consumption with between MWAITX delay and TSC delay on AMD Carrizo
platforms.

Reference:
https://lkml.kernel.org/r/1438744732-1459-1-git-send-email-ray.huang@amd.com

Commit f96756 at tip ("x86/asm: Add MONITORX/MWAITX instruction support")
Commit b466bd at tip ("x86/asm/delay: Introduce an MWAITX-based delay with a configurable timer")

V1: https://lkml.kernel.org/r/1440662866-28716-1-git-send-email-ray.huang@amd.com
V2: https://lkml.kernel.org/r/1445308109-17970-1-git-send-email-ray.huang@amd.com
V3: https://lkml.kernel.org/r/1446199024-1896-1-git-send-email-ray.huang@amd.com

V3 is blocked by the patch "x86, amd: add accessor for number of cores
per compute unit", but now we can use smp_num_siblings instead of
cores_per_cu accessor. So resend the V4 now, for detailed info, please
check:

https://lkml.kernel.org/r/1449719771-12837-1-git-send-email-ray.huang@amd.com


Changes from v1 -> v2:
- Move fam15h_power_groups and fam15h_power_group into fam15h_power_data to
  avoid overwrite on multi-CPU system.
- Rename FAM15H_MIN_NUM_ATTRS macro and fix return error code.
- Remove unnecessary warning print.
- Adds do_read_registers_on_cu to do all the read to all MSRs and run it on one
  of the online cores on each compute unit with smp_call_function_many().
- Use power1_average and power1_average_interval standard entry
  instread of power1_acc
- Fix the CPU-hotplug case.

Changes from v2 -> v3:
- As Guenter's suggestion, remove typecast, use &data->groups[0].
- Remove all "fam15_power_*" prefix at data.
- Remove unnecessary ( ).
- Fix the issue that is reported by build test robot, and add
  CPU_SUP_AMD as the dependence of fam15h_power
- Remove the WARN_ON at do_read_registers_on_cu, because it must be
  behind CPUID check. The MSR must be available since
  CPUID.8000_0007H:EDX[12] is set 
- Add get_online_cpus()/put_online_cpus() functions.
- Refine commments and the method which generate cpumask for cu.
- Add the interval scope to make the value suitable for user
  experience
- Remove the useless mutex.

Changes from v3 -> v4:
- Rebase the patches to latest groeck/hwmon-next.
- Use smp_num_siblings instead of cores_per_cu accessor.
- Refine the cpumask method which is inspired by perf solution.
- Fix some typo and errors.


A simple example:

ray@hr-ub:~/tip$ sensors
fam15h_power-pci-00c4
Adapter: PCI adapter
power1:       19.58 mW (avg =   2.55 mW, interval =   0.01 s)
                       (crit =  15.00 W)

...

These patches are rebased on groeck/hwmon-next.

Thanks,
Rui


Huang Rui (6):
  hwmon: (fam15h_power) Add CPU_SUP_AMD as the dependence
  hwmon: (fam15h_power) Add compute unit accumulated power
  hwmon: (fam15h_power) Add ptsc counter value for accumulated power
  hwmon: (fam15h_power) Introduce a cpu accumulated power reporting
    algorithm
  hwmon: (fam15h_power) Add documentation for TDP and accumulated power
    algorithm
  hwmon: (fam15h_power) Add platform check function

 Documentation/hwmon/fam15h_power |  57 +++++++++++-
 drivers/hwmon/Kconfig            |   2 +-
 drivers/hwmon/fam15h_power.c     | 191 ++++++++++++++++++++++++++++++++++++++-
 3 files changed, 244 insertions(+), 6 deletions(-)

-- 
1.9.1

[lm-sensors] [PATCH v4 0/6] hwmon: (fam15h_power) Introduce an accumulated power reporting algorithm

[Huang Rui]

Hi Guenter,

This serial of patches introduces an accumulated power reporting
algorithm. It will calculate the average power consumption for the
processor. The cpu feature flag is CPUID.8000_0007H:EDX[12].

This algorithm is used to test the comparison of processor power
consumption with between MWAITX delay and TSC delay on AMD Carrizo
platforms.

Reference:
https://lkml.kernel.org/r/1438744732-1459-1-git-send-email-ray.huang@amd.com

Commit f96756 at tip ("x86/asm: Add MONITORX/MWAITX instruction support")
Commit b466bd at tip ("x86/asm/delay: Introduce an MWAITX-based delay with a configurable timer")

V1: https://lkml.kernel.org/r/1440662866-28716-1-git-send-email-ray.huang@amd.com
V2: https://lkml.kernel.org/r/1445308109-17970-1-git-send-email-ray.huang@amd.com
V3: https://lkml.kernel.org/r/1446199024-1896-1-git-send-email-ray.huang@amd.com

V3 is blocked by the patch "x86, amd: add accessor for number of cores
per compute unit", but now we can use smp_num_siblings instead of
cores_per_cu accessor. So resend the V4 now, for detailed info, please
check:

https://lkml.kernel.org/r/1449719771-12837-1-git-send-email-ray.huang@amd.com


Changes from v1 -> v2:
- Move fam15h_power_groups and fam15h_power_group into fam15h_power_data to
  avoid overwrite on multi-CPU system.
- Rename FAM15H_MIN_NUM_ATTRS macro and fix return error code.
- Remove unnecessary warning print.
- Adds do_read_registers_on_cu to do all the read to all MSRs and run it on one
  of the online cores on each compute unit with smp_call_function_many().
- Use power1_average and power1_average_interval standard entry
  instread of power1_acc
- Fix the CPU-hotplug case.

Changes from v2 -> v3:
- As Guenter's suggestion, remove typecast, use &data->groups[0].
- Remove all "fam15_power_*" prefix at data.
- Remove unnecessary ( ).
- Fix the issue that is reported by build test robot, and add
  CPU_SUP_AMD as the dependence of fam15h_power
- Remove the WARN_ON at do_read_registers_on_cu, because it must be
  behind CPUID check. The MSR must be available since
  CPUID.8000_0007H:EDX[12] is set 
- Add get_online_cpus()/put_online_cpus() functions.
- Refine commments and the method which generate cpumask for cu.
- Add the interval scope to make the value suitable for user
  experience
- Remove the useless mutex.

Changes from v3 -> v4:
- Rebase the patches to latest groeck/hwmon-next.
- Use smp_num_siblings instead of cores_per_cu accessor.
- Refine the cpumask method which is inspired by perf solution.
- Fix some typo and errors.


A simple example:

ray@hr-ub:~/tip$ sensors
fam15h_power-pci-00c4
Adapter: PCI adapter
power1:       19.58 mW (avg =   2.55 mW, interval =   0.01 s)
                       (crit =  15.00 W)

...

These patches are rebased on groeck/hwmon-next.

Thanks,
Rui


Huang Rui (6):
  hwmon: (fam15h_power) Add CPU_SUP_AMD as the dependence
  hwmon: (fam15h_power) Add compute unit accumulated power
  hwmon: (fam15h_power) Add ptsc counter value for accumulated power
  hwmon: (fam15h_power) Introduce a cpu accumulated power reporting
    algorithm
  hwmon: (fam15h_power) Add documentation for TDP and accumulated power
    algorithm
  hwmon: (fam15h_power) Add platform check function

 Documentation/hwmon/fam15h_power |  57 +++++++++++-
 drivers/hwmon/Kconfig            |   2 +-
 drivers/hwmon/fam15h_power.c     | 191 ++++++++++++++++++++++++++++++++++++++-
 3 files changed, 244 insertions(+), 6 deletions(-)

-- 
1.9.1


_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

[PATCH v4 1/6] hwmon: (fam15h_power) Add CPU_SUP_AMD as the dependence

[Huang Rui]

This patch adds CONFIG_CPU_SUP_AMD as the dependence of fam15h_power
driver. Because the following patch will use the interface from
x86/kernel/cpu/amd.c.

Otherwise, the below error might be encountered:

All errors (new ones prefixed by >>):

   drivers/built-in.o: In function `fam15h_power_probe':
>> fam15h_power.c:(.text+0x26e3a3): undefined reference to
>> `amd_get_cores_per_cu'
   fam15h_power.c:(.text+0x26e41e): undefined reference to
`amd_get_cores_per_cu'

Reported-by: build test robot <lkp@intel.com>
Signed-off-by: Huang Rui <ray.huang@amd.com>
---
 drivers/hwmon/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
index 5c2d13a..4be3792 100644
--- a/drivers/hwmon/Kconfig
+++ b/drivers/hwmon/Kconfig
@@ -288,7 +288,7 @@ config SENSORS_K10TEMP
 
 config SENSORS_FAM15H_POWER
 	tristate "AMD Family 15h processor power"
-	depends on X86 && PCI
+	depends on X86 && PCI && CPU_SUP_AMD
 	help
 	  If you say yes here you get support for processor power
 	  information of your AMD family 15h CPU.
-- 
1.9.1

[lm-sensors] [PATCH v4 1/6] hwmon: (fam15h_power) Add CPU_SUP_AMD as the dependence

[Huang Rui]

This patch adds CONFIG_CPU_SUP_AMD as the dependence of fam15h_power
driver. Because the following patch will use the interface from
x86/kernel/cpu/amd.c.

Otherwise, the below error might be encountered:

All errors (new ones prefixed by >>):

   drivers/built-in.o: In function `fam15h_power_probe':
>> fam15h_power.c:(.text+0x26e3a3): undefined reference to
>> `amd_get_cores_per_cu'
   fam15h_power.c:(.text+0x26e41e): undefined reference to
`amd_get_cores_per_cu'

Reported-by: build test robot <lkp@intel.com>
Signed-off-by: Huang Rui <ray.huang@amd.com>
---
 drivers/hwmon/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
index 5c2d13a..4be3792 100644
--- a/drivers/hwmon/Kconfig
+++ b/drivers/hwmon/Kconfig
@@ -288,7 +288,7 @@ config SENSORS_K10TEMP
 
 config SENSORS_FAM15H_POWER
 	tristate "AMD Family 15h processor power"
-	depends on X86 && PCI
+	depends on X86 && PCI && CPU_SUP_AMD
 	help
 	  If you say yes here you get support for processor power
 	  information of your AMD family 15h CPU.
-- 
1.9.1


_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

[PATCH v4 2/6] hwmon: (fam15h_power) Add compute unit accumulated power

[Huang Rui]

This patch adds a member in fam15h_power_data which specifies the
compute unit accumulated power. It adds do_read_registers_on_cu to do
all the read to all MSRs and run it on one of the online cores on each
compute unit with smp_call_function_many(). This behavior can decrease
IPI numbers.

Suggested-by: Borislav Petkov <bp@alien8.de>
Signed-off-by: Huang Rui <ray.huang@amd.com>
---
 drivers/hwmon/fam15h_power.c | 61 +++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 60 insertions(+), 1 deletion(-)

diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
index 4f695d8..c5e2297 100644
--- a/drivers/hwmon/fam15h_power.c
+++ b/drivers/hwmon/fam15h_power.c
@@ -25,6 +25,8 @@
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/bitops.h>
+#include <linux/cpu.h>
+#include <linux/cpumask.h>
 #include <asm/processor.h>
 #include <asm/msr.h>
 
@@ -44,7 +46,9 @@ MODULE_LICENSE("GPL");
 
 #define FAM15H_MIN_NUM_ATTRS		2
 #define FAM15H_NUM_GROUPS		2
+#define MAX_CUS				8
 
+#define MSR_F15H_CU_PWR_ACCUMULATOR	0xc001007a
 #define MSR_F15H_CU_MAX_PWR_ACCUMULATOR	0xc001007b
 
 #define PCI_DEVICE_ID_AMD_15H_M70H_NB_F4 0x15b4
@@ -59,6 +63,8 @@ struct fam15h_power_data {
 	struct attribute_group group;
 	/* maximum accumulated power of a compute unit */
 	u64 max_cu_acc_power;
+	/* accumulated power of the compute units */
+	u64 cu_acc_power[MAX_CUS];
 };
 
 static ssize_t show_power(struct device *dev,
@@ -125,6 +131,59 @@ static ssize_t show_power_crit(struct device *dev,
 }
 static DEVICE_ATTR(power1_crit, S_IRUGO, show_power_crit, NULL);
 
+static void do_read_registers_on_cu(void *_data)
+{
+	struct fam15h_power_data *data = _data;
+	int cpu, cu;
+
+	cpu = smp_processor_id();
+
+	cu = cpu / smp_num_siblings;
+
+	rdmsrl_safe(MSR_F15H_CU_PWR_ACCUMULATOR, &data->cu_acc_power[cu]);
+}
+
+/*
+ * This function is only able to be called when CPUID
+ * Fn8000_0007:EDX[12] is set.
+ */
+static int read_registers(struct fam15h_power_data *data)
+{
+	int this_cpu, ret, cpu;
+	int target;
+	cpumask_var_t mask;
+
+	ret = zalloc_cpumask_var(&mask, GFP_KERNEL);
+	if (!ret)
+		return -ENOMEM;
+
+	get_online_cpus();
+	this_cpu = get_cpu();
+
+	/*
+	 * Choose the first online core of each compute unit, and then
+	 * read their MSR value of power and ptsc in a single IPI,
+	 * because the MSR value of CPU core represent the compute
+	 * unit's.
+	 */
+	for_each_online_cpu(cpu) {
+		target = cpumask_first(topology_sibling_cpumask(cpu));
+		if (!cpumask_test_cpu(target, mask))
+			cpumask_set_cpu(target, mask);
+	}
+
+	if (cpumask_test_cpu(this_cpu, mask))
+		do_read_registers_on_cu(data);
+
+	smp_call_function_many(mask, do_read_registers_on_cu, data, true);
+	put_cpu();
+	put_online_cpus();
+
+	free_cpumask_var(mask);
+
+	return 0;
+}
+
 static int fam15h_power_init_attrs(struct pci_dev *pdev,
 				   struct fam15h_power_data *data)
 {
@@ -263,7 +322,7 @@ static int fam15h_power_init_data(struct pci_dev *f4,
 
 	data->max_cu_acc_power = tmp;
 
-	return 0;
+	return read_registers(data);
 }
 
 static int fam15h_power_probe(struct pci_dev *pdev,
-- 
1.9.1

[lm-sensors] [PATCH v4 2/6] hwmon: (fam15h_power) Add compute unit accumulated power

[Huang Rui]

This patch adds a member in fam15h_power_data which specifies the
compute unit accumulated power. It adds do_read_registers_on_cu to do
all the read to all MSRs and run it on one of the online cores on each
compute unit with smp_call_function_many(). This behavior can decrease
IPI numbers.

Suggested-by: Borislav Petkov <bp@alien8.de>
Signed-off-by: Huang Rui <ray.huang@amd.com>
---
 drivers/hwmon/fam15h_power.c | 61 +++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 60 insertions(+), 1 deletion(-)

diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
index 4f695d8..c5e2297 100644
--- a/drivers/hwmon/fam15h_power.c
+++ b/drivers/hwmon/fam15h_power.c
@@ -25,6 +25,8 @@
 #include <linux/module.h>
 #include <linux/pci.h>
 #include <linux/bitops.h>
+#include <linux/cpu.h>
+#include <linux/cpumask.h>
 #include <asm/processor.h>
 #include <asm/msr.h>
 
@@ -44,7 +46,9 @@ MODULE_LICENSE("GPL");
 
 #define FAM15H_MIN_NUM_ATTRS		2
 #define FAM15H_NUM_GROUPS		2
+#define MAX_CUS				8
 
+#define MSR_F15H_CU_PWR_ACCUMULATOR	0xc001007a
 #define MSR_F15H_CU_MAX_PWR_ACCUMULATOR	0xc001007b
 
 #define PCI_DEVICE_ID_AMD_15H_M70H_NB_F4 0x15b4
@@ -59,6 +63,8 @@ struct fam15h_power_data {
 	struct attribute_group group;
 	/* maximum accumulated power of a compute unit */
 	u64 max_cu_acc_power;
+	/* accumulated power of the compute units */
+	u64 cu_acc_power[MAX_CUS];
 };
 
 static ssize_t show_power(struct device *dev,
@@ -125,6 +131,59 @@ static ssize_t show_power_crit(struct device *dev,
 }
 static DEVICE_ATTR(power1_crit, S_IRUGO, show_power_crit, NULL);
 
+static void do_read_registers_on_cu(void *_data)
+{
+	struct fam15h_power_data *data = _data;
+	int cpu, cu;
+
+	cpu = smp_processor_id();
+
+	cu = cpu / smp_num_siblings;
+
+	rdmsrl_safe(MSR_F15H_CU_PWR_ACCUMULATOR, &data->cu_acc_power[cu]);
+}
+
+/*
+ * This function is only able to be called when CPUID
+ * Fn8000_0007:EDX[12] is set.
+ */
+static int read_registers(struct fam15h_power_data *data)
+{
+	int this_cpu, ret, cpu;
+	int target;
+	cpumask_var_t mask;
+
+	ret = zalloc_cpumask_var(&mask, GFP_KERNEL);
+	if (!ret)
+		return -ENOMEM;
+
+	get_online_cpus();
+	this_cpu = get_cpu();
+
+	/*
+	 * Choose the first online core of each compute unit, and then
+	 * read their MSR value of power and ptsc in a single IPI,
+	 * because the MSR value of CPU core represent the compute
+	 * unit's.
+	 */
+	for_each_online_cpu(cpu) {
+		target = cpumask_first(topology_sibling_cpumask(cpu));
+		if (!cpumask_test_cpu(target, mask))
+			cpumask_set_cpu(target, mask);
+	}
+
+	if (cpumask_test_cpu(this_cpu, mask))
+		do_read_registers_on_cu(data);
+
+	smp_call_function_many(mask, do_read_registers_on_cu, data, true);
+	put_cpu();
+	put_online_cpus();
+
+	free_cpumask_var(mask);
+
+	return 0;
+}
+
 static int fam15h_power_init_attrs(struct pci_dev *pdev,
 				   struct fam15h_power_data *data)
 {
@@ -263,7 +322,7 @@ static int fam15h_power_init_data(struct pci_dev *f4,
 
 	data->max_cu_acc_power = tmp;
 
-	return 0;
+	return read_registers(data);
 }
 
 static int fam15h_power_probe(struct pci_dev *pdev,
-- 
1.9.1


_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

[PATCH v4 3/6] hwmon: (fam15h_power) Add ptsc counter value for accumulated power

[Huang Rui]

PTSC is the performance timestamp counter value in a cpu core and the
cores in one compute unit have the fixed frequency. So it picks up the
performance timestamp counter value of the first core per compute unit
to measure the interval for average power per compute unit.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Cc: Borislav Petkov <bp@alien8.de>
---
 drivers/hwmon/fam15h_power.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
index c5e2297..72aef47 100644
--- a/drivers/hwmon/fam15h_power.c
+++ b/drivers/hwmon/fam15h_power.c
@@ -50,6 +50,7 @@ MODULE_LICENSE("GPL");
 
 #define MSR_F15H_CU_PWR_ACCUMULATOR	0xc001007a
 #define MSR_F15H_CU_MAX_PWR_ACCUMULATOR	0xc001007b
+#define MSR_F15H_PTSC			0xc0010280
 
 #define PCI_DEVICE_ID_AMD_15H_M70H_NB_F4 0x15b4
 
@@ -65,6 +66,8 @@ struct fam15h_power_data {
 	u64 max_cu_acc_power;
 	/* accumulated power of the compute units */
 	u64 cu_acc_power[MAX_CUS];
+	/* performance timestamp counter */
+	u64 cpu_sw_pwr_ptsc[MAX_CUS];
 };
 
 static ssize_t show_power(struct device *dev,
@@ -141,6 +144,7 @@ static void do_read_registers_on_cu(void *_data)
 	cu = cpu / smp_num_siblings;
 
 	rdmsrl_safe(MSR_F15H_CU_PWR_ACCUMULATOR, &data->cu_acc_power[cu]);
+	rdmsrl_safe(MSR_F15H_PTSC, &data->cpu_sw_pwr_ptsc[cu]);
 }
 
 /*
-- 
1.9.1

[lm-sensors] [PATCH v4 3/6] hwmon: (fam15h_power) Add ptsc counter value for accumulated power

[Huang Rui]

PTSC is the performance timestamp counter value in a cpu core and the
cores in one compute unit have the fixed frequency. So it picks up the
performance timestamp counter value of the first core per compute unit
to measure the interval for average power per compute unit.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Cc: Borislav Petkov <bp@alien8.de>
---
 drivers/hwmon/fam15h_power.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
index c5e2297..72aef47 100644
--- a/drivers/hwmon/fam15h_power.c
+++ b/drivers/hwmon/fam15h_power.c
@@ -50,6 +50,7 @@ MODULE_LICENSE("GPL");
 
 #define MSR_F15H_CU_PWR_ACCUMULATOR	0xc001007a
 #define MSR_F15H_CU_MAX_PWR_ACCUMULATOR	0xc001007b
+#define MSR_F15H_PTSC			0xc0010280
 
 #define PCI_DEVICE_ID_AMD_15H_M70H_NB_F4 0x15b4
 
@@ -65,6 +66,8 @@ struct fam15h_power_data {
 	u64 max_cu_acc_power;
 	/* accumulated power of the compute units */
 	u64 cu_acc_power[MAX_CUS];
+	/* performance timestamp counter */
+	u64 cpu_sw_pwr_ptsc[MAX_CUS];
 };
 
 static ssize_t show_power(struct device *dev,
@@ -141,6 +144,7 @@ static void do_read_registers_on_cu(void *_data)
 	cu = cpu / smp_num_siblings;
 
 	rdmsrl_safe(MSR_F15H_CU_PWR_ACCUMULATOR, &data->cu_acc_power[cu]);
+	rdmsrl_safe(MSR_F15H_PTSC, &data->cpu_sw_pwr_ptsc[cu]);
 }
 
 /*
-- 
1.9.1


_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

[PATCH v4 4/6] hwmon: (fam15h_power) Introduce a cpu accumulated power reporting algorithm

[Huang Rui]

This patch introduces an algorithm that computes the average power by
reading a delta value of “core power accumulator” register during
measurement interval, and then dividing delta value by the length of
the time interval.

User is able to use power1_average entry to measure the processor power
consumption and power1_average_interval entry to set the interval.

A simple example:

ray@hr-ub:~/tip$ sensors
fam15h_power-pci-00c4
Adapter: PCI adapter
power1:       19.58 mW (avg =   2.55 mW, interval =   0.01 s)
                       (crit =  15.00 W)

...

The result is current average processor power consumption in 10
millisecond. The unit of the result is uWatt.

Suggested-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Huang Rui <ray.huang@amd.com>
Cc: Borislav Petkov <bp@alien8.de>
---
 drivers/hwmon/fam15h_power.c | 115 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 115 insertions(+)

diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
index 72aef47..46cbca0 100644
--- a/drivers/hwmon/fam15h_power.c
+++ b/drivers/hwmon/fam15h_power.c
@@ -27,6 +27,8 @@
 #include <linux/bitops.h>
 #include <linux/cpu.h>
 #include <linux/cpumask.h>
+#include <linux/time.h>
+#include <linux/sched.h>
 #include <asm/processor.h>
 #include <asm/msr.h>
 
@@ -48,6 +50,9 @@ MODULE_LICENSE("GPL");
 #define FAM15H_NUM_GROUPS		2
 #define MAX_CUS				8
 
+/* set maximum interval as 1 second */
+#define MAX_INTERVAL			1000
+
 #define MSR_F15H_CU_PWR_ACCUMULATOR	0xc001007a
 #define MSR_F15H_CU_MAX_PWR_ACCUMULATOR	0xc001007b
 #define MSR_F15H_PTSC			0xc0010280
@@ -68,6 +73,9 @@ struct fam15h_power_data {
 	u64 cu_acc_power[MAX_CUS];
 	/* performance timestamp counter */
 	u64 cpu_sw_pwr_ptsc[MAX_CUS];
+	/* online/offline status of current compute unit */
+	int cu_on[MAX_CUS];
+	unsigned long power_period;
 };
 
 static ssize_t show_power(struct device *dev,
@@ -145,6 +153,8 @@ static void do_read_registers_on_cu(void *_data)
 
 	rdmsrl_safe(MSR_F15H_CU_PWR_ACCUMULATOR, &data->cu_acc_power[cu]);
 	rdmsrl_safe(MSR_F15H_PTSC, &data->cpu_sw_pwr_ptsc[cu]);
+
+	data->cu_on[cu] = 1;
 }
 
 /*
@@ -161,6 +171,8 @@ static int read_registers(struct fam15h_power_data *data)
 	if (!ret)
 		return -ENOMEM;
 
+	memset(data->cu_on, 0, sizeof(int) * MAX_CUS);
+
 	get_online_cpus();
 	this_cpu = get_cpu();
 
@@ -188,18 +200,113 @@ static int read_registers(struct fam15h_power_data *data)
 	return 0;
 }
 
+static ssize_t acc_show_power(struct device *dev,
+			      struct device_attribute *attr,
+			      char *buf)
+{
+	struct fam15h_power_data *data = dev_get_drvdata(dev);
+	u64 prev_cu_acc_power[MAX_CUS], prev_ptsc[MAX_CUS],
+	    jdelta[MAX_CUS];
+	u64 tdelta, avg_acc;
+	int cu, cu_num, ret;
+	signed long leftover;
+
+	cu_num = boot_cpu_data.x86_max_cores / smp_num_siblings;
+
+	ret = read_registers(data);
+	if (ret)
+		return 0;
+
+	for (cu = 0; cu < cu_num; cu++) {
+		prev_cu_acc_power[cu] = data->cu_acc_power[cu];
+		prev_ptsc[cu] = data->cpu_sw_pwr_ptsc[cu];
+	}
+
+	leftover = schedule_timeout_interruptible(msecs_to_jiffies(data->power_period));
+	if (leftover)
+		return 0;
+
+	ret = read_registers(data);
+	if (ret)
+		return 0;
+
+	for (cu = 0, avg_acc = 0; cu < cu_num; cu++) {
+		/* check if current compute unit is online */
+		if (data->cu_on[cu] == 0)
+			continue;
+
+		if (data->cu_acc_power[cu] < prev_cu_acc_power[cu]) {
+			jdelta[cu] = data->max_cu_acc_power + data->cu_acc_power[cu];
+			jdelta[cu] -= prev_cu_acc_power[cu];
+		} else {
+			jdelta[cu] = data->cu_acc_power[cu] - prev_cu_acc_power[cu];
+		}
+		tdelta = data->cpu_sw_pwr_ptsc[cu] - prev_ptsc[cu];
+		jdelta[cu] *= data->cpu_pwr_sample_ratio * 1000;
+		do_div(jdelta[cu], tdelta);
+
+		/* the unit is microWatt */
+		avg_acc += jdelta[cu];
+	}
+
+	return sprintf(buf, "%llu\n", (unsigned long long)avg_acc);
+}
+static DEVICE_ATTR(power1_average, S_IRUGO, acc_show_power, NULL);
+
+static ssize_t acc_show_power_period(struct device *dev,
+				     struct device_attribute *attr,
+				     char *buf)
+{
+	struct fam15h_power_data *data = dev_get_drvdata(dev);
+
+	return sprintf(buf, "%lu\n", data->power_period);
+}
+
+static ssize_t acc_set_power_period(struct device *dev,
+				    struct device_attribute *attr,
+				    const char *buf, size_t count)
+{
+	struct fam15h_power_data *data = dev_get_drvdata(dev);
+	unsigned long temp;
+	int ret;
+
+	ret = kstrtoul(buf, 10, &temp);
+	if (ret)
+		return ret;
+
+	if (temp > MAX_INTERVAL)
+		return -EINVAL;
+
+	/* the interval value should be greater than 0 */
+	if (temp <= 0)
+		return -EINVAL;
+
+	data->power_period = temp;
+
+	return count;
+}
+static DEVICE_ATTR(power1_average_interval, S_IRUGO | S_IWUSR,
+		   acc_show_power_period, acc_set_power_period);
+
 static int fam15h_power_init_attrs(struct pci_dev *pdev,
 				   struct fam15h_power_data *data)
 {
 	int n = FAM15H_MIN_NUM_ATTRS;
 	struct attribute **fam15h_power_attrs;
 	struct cpuinfo_x86 *c = &boot_cpu_data;
+	u32 cpuid;
 
 	if (c->x86 == 0x15 &&
 	    (c->x86_model <= 0xf ||
 	     (c->x86_model >= 0x60 && c->x86_model <= 0x7f)))
 		n += 1;
 
+	cpuid = cpuid_edx(0x80000007);
+
+	/* check if processor supports accumulated power */
+	if (cpuid & BIT(12))
+		n += 2;
+
 	fam15h_power_attrs = devm_kcalloc(&pdev->dev, n,
 					  sizeof(*fam15h_power_attrs),
 					  GFP_KERNEL);
@@ -214,6 +321,11 @@ static int fam15h_power_init_attrs(struct pci_dev *pdev,
 	     (c->x86_model >= 0x60 && c->x86_model <= 0x7f)))
 		fam15h_power_attrs[n++] = &dev_attr_power1_input.attr;
 
+	if (cpuid & BIT(12)) {
+		fam15h_power_attrs[n++] = &dev_attr_power1_average.attr;
+		fam15h_power_attrs[n++] = &dev_attr_power1_average_interval.attr;
+	}
+
 	data->group.attrs = fam15h_power_attrs;
 
 	return 0;
@@ -326,6 +438,9 @@ static int fam15h_power_init_data(struct pci_dev *f4,
 
 	data->max_cu_acc_power = tmp;
 
+	/* set default interval as 10 ms */
+	data->power_period = 10;
+
 	return read_registers(data);
 }
 
-- 
1.9.1

[lm-sensors] [PATCH v4 4/6] hwmon: (fam15h_power) Introduce a cpu accumulated power reporting algori

[Huang Rui]
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[PATCH v4 5/6] hwmon: (fam15h_power) Add documentation for TDP and accumulated power algorithm

[Huang Rui]

This patch adds the description to explain the TDP reporting mechanism
and accumulated power algorithm.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Cc: Borislav Petkov <bp@alien8.de>
---
 Documentation/hwmon/fam15h_power | 57 +++++++++++++++++++++++++++++++++++++++-
 drivers/hwmon/fam15h_power.c     |  2 +-
 2 files changed, 57 insertions(+), 2 deletions(-)

diff --git a/Documentation/hwmon/fam15h_power b/Documentation/hwmon/fam15h_power
index e2b1b69..2c4fbee 100644
--- a/Documentation/hwmon/fam15h_power
+++ b/Documentation/hwmon/fam15h_power
@@ -10,14 +10,22 @@ Supported chips:
   Datasheets:
   BIOS and Kernel Developer's Guide (BKDG) For AMD Family 15h Processors
   BIOS and Kernel Developer's Guide (BKDG) For AMD Family 16h Processors
+  AMD64 Architecture Programmer's Manual Volume 2: System Programming
 
 Author: Andreas Herrmann <herrmann.der.user@googlemail.com>
 
 Description
 -----------
 
+1) Processor TDP (Thermal design power)
+
+Given a fixed frequency and voltage, the power consumption of a
+processor varies based on the workload being executed. Derated power
+is the power consumed when running a specific application. Thermal
+design power (TDP) is an example of derated power.
+
 This driver permits reading of registers providing power information
-of AMD Family 15h and 16h processors.
+of AMD Family 15h and 16h processors via TDP algorithm.
 
 For AMD Family 15h and 16h processors the following power values can
 be calculated using different processor northbridge function
@@ -37,3 +45,50 @@ This driver provides ProcessorPwrWatts and CurrPwrWatts:
 On multi-node processors the calculated value is for the entire
 package and not for a single node. Thus the driver creates sysfs
 attributes only for internal node0 of a multi-node processor.
+
+2) Accumulated Power Mechanism
+
+This driver also introduces an algorithm that should be used to
+calculate the average power consumed by a processor during a
+measurement interval Tm. The feature of accumulated power mechanism is
+indicated by CPUID Fn8000_0007_EDX[12].
+
+* Tsample: compute unit power accumulator sample period
+* Tref: the PTSC counter period
+* PTSC: performance timestamp counter
+* N: the ratio of compute unit power accumulator sample period to the
+  PTSC period
+* Jmax: max compute unit accumulated power which is indicated by
+  MaxCpuSwPwrAcc MSR C001007b
+* Jx/Jy: compute unit accumulated power which is indicated by
+  CpuSwPwrAcc MSR C001007a
+* Tx/Ty: the value of performance timestamp counter which is indicated
+  by CU_PTSC MSR C0010280
+* PwrCPUave: CPU average power
+
+i. Determine the ratio of Tsample to Tref by executing CPUID Fn8000_0007.
+	N = value of CPUID Fn8000_0007_ECX[CpuPwrSampleTimeRatio[15:0]].
+
+ii. Read the full range of the cumulative energy value from the new
+MSR MaxCpuSwPwrAcc.
+	Jmax = value returned.
+iii. At time x, SW reads CpuSwPwrAcc MSR and samples the PTSC.
+	Jx = value read from CpuSwPwrAcc and Tx = value read from
+PTSC.
+
+iv. At time y, SW reads CpuSwPwrAcc MSR and samples the PTSC.
+	Jy = value read from CpuSwPwrAcc and Ty = value read from
+PTSC.
+
+v. Calculate the average power consumption for a compute unit over
+time period (y-x). Unit of result is uWatt.
+	if (Jy < Jx) // Rollover has occurred
+		Jdelta = (Jy + Jmax) - Jx
+	else
+		Jdelta = Jy - Jx
+	PwrCPUave = N * Jdelta * 1000 / (Ty - Tx)
+
+This driver provides PwrCPUave and interval(default is 10 millisecond
+and maximum is 1 second):
+* power1_average (PwrCPUave)
+* power1_average_interval (Interval)
diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
index 46cbca0..35800cd 100644
--- a/drivers/hwmon/fam15h_power.c
+++ b/drivers/hwmon/fam15h_power.c
@@ -1,7 +1,7 @@
 /*
  * fam15h_power.c - AMD Family 15h processor power monitoring
  *
- * Copyright (c) 2011 Advanced Micro Devices, Inc.
+ * Copyright (c) 2011-2016 Advanced Micro Devices, Inc.
  * Author: Andreas Herrmann <herrmann.der.user@googlemail.com>
  *
  *
-- 
1.9.1

[lm-sensors] [PATCH v4 5/6] hwmon: (fam15h_power) Add documentation for TDP and accumulated power al

[Huang Rui]

This patch adds the description to explain the TDP reporting mechanism
and accumulated power algorithm.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Cc: Borislav Petkov <bp@alien8.de>
---
 Documentation/hwmon/fam15h_power | 57 +++++++++++++++++++++++++++++++++++++++-
 drivers/hwmon/fam15h_power.c     |  2 +-
 2 files changed, 57 insertions(+), 2 deletions(-)

diff --git a/Documentation/hwmon/fam15h_power b/Documentation/hwmon/fam15h_power
index e2b1b69..2c4fbee 100644
--- a/Documentation/hwmon/fam15h_power
+++ b/Documentation/hwmon/fam15h_power
@@ -10,14 +10,22 @@ Supported chips:
   Datasheets:
   BIOS and Kernel Developer's Guide (BKDG) For AMD Family 15h Processors
   BIOS and Kernel Developer's Guide (BKDG) For AMD Family 16h Processors
+  AMD64 Architecture Programmer's Manual Volume 2: System Programming
 
 Author: Andreas Herrmann <herrmann.der.user@googlemail.com>
 
 Description
 -----------
 
+1) Processor TDP (Thermal design power)
+
+Given a fixed frequency and voltage, the power consumption of a
+processor varies based on the workload being executed. Derated power
+is the power consumed when running a specific application. Thermal
+design power (TDP) is an example of derated power.
+
 This driver permits reading of registers providing power information
-of AMD Family 15h and 16h processors.
+of AMD Family 15h and 16h processors via TDP algorithm.
 
 For AMD Family 15h and 16h processors the following power values can
 be calculated using different processor northbridge function
@@ -37,3 +45,50 @@ This driver provides ProcessorPwrWatts and CurrPwrWatts:
 On multi-node processors the calculated value is for the entire
 package and not for a single node. Thus the driver creates sysfs
 attributes only for internal node0 of a multi-node processor.
+
+2) Accumulated Power Mechanism
+
+This driver also introduces an algorithm that should be used to
+calculate the average power consumed by a processor during a
+measurement interval Tm. The feature of accumulated power mechanism is
+indicated by CPUID Fn8000_0007_EDX[12].
+
+* Tsample: compute unit power accumulator sample period
+* Tref: the PTSC counter period
+* PTSC: performance timestamp counter
+* N: the ratio of compute unit power accumulator sample period to the
+  PTSC period
+* Jmax: max compute unit accumulated power which is indicated by
+  MaxCpuSwPwrAcc MSR C001007b
+* Jx/Jy: compute unit accumulated power which is indicated by
+  CpuSwPwrAcc MSR C001007a
+* Tx/Ty: the value of performance timestamp counter which is indicated
+  by CU_PTSC MSR C0010280
+* PwrCPUave: CPU average power
+
+i. Determine the ratio of Tsample to Tref by executing CPUID Fn8000_0007.
+	N = value of CPUID Fn8000_0007_ECX[CpuPwrSampleTimeRatio[15:0]].
+
+ii. Read the full range of the cumulative energy value from the new
+MSR MaxCpuSwPwrAcc.
+	Jmax = value returned.
+iii. At time x, SW reads CpuSwPwrAcc MSR and samples the PTSC.
+	Jx = value read from CpuSwPwrAcc and Tx = value read from
+PTSC.
+
+iv. At time y, SW reads CpuSwPwrAcc MSR and samples the PTSC.
+	Jy = value read from CpuSwPwrAcc and Ty = value read from
+PTSC.
+
+v. Calculate the average power consumption for a compute unit over
+time period (y-x). Unit of result is uWatt.
+	if (Jy < Jx) // Rollover has occurred
+		Jdelta = (Jy + Jmax) - Jx
+	else
+		Jdelta = Jy - Jx
+	PwrCPUave = N * Jdelta * 1000 / (Ty - Tx)
+
+This driver provides PwrCPUave and interval(default is 10 millisecond
+and maximum is 1 second):
+* power1_average (PwrCPUave)
+* power1_average_interval (Interval)
diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
index 46cbca0..35800cd 100644
--- a/drivers/hwmon/fam15h_power.c
+++ b/drivers/hwmon/fam15h_power.c
@@ -1,7 +1,7 @@
 /*
  * fam15h_power.c - AMD Family 15h processor power monitoring
  *
- * Copyright (c) 2011 Advanced Micro Devices, Inc.
+ * Copyright (c) 2011-2016 Advanced Micro Devices, Inc.
  * Author: Andreas Herrmann <herrmann.der.user@googlemail.com>
  *
  *
-- 
1.9.1


_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

[PATCH v4 6/6] hwmon: (fam15h_power) Add platform check function

[Huang Rui]

This patch adds a platform check function to make code more readable.

Signed-off-by: Huang Rui <ray.huang@amd.com>
---
 drivers/hwmon/fam15h_power.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
index 35800cd..a0bea56 100644
--- a/drivers/hwmon/fam15h_power.c
+++ b/drivers/hwmon/fam15h_power.c
@@ -78,6 +78,11 @@ struct fam15h_power_data {
 	unsigned long power_period;
 };
 
+static bool is_carrizo_or_later(void)
+{
+	return boot_cpu_data.x86 == 0x15 && boot_cpu_data.x86_model >= 0x60;
+}
+
 static ssize_t show_power(struct device *dev,
 			  struct device_attribute *attr, char *buf)
 {
@@ -94,7 +99,7 @@ static ssize_t show_power(struct device *dev,
 	 * On Carrizo and later platforms, TdpRunAvgAccCap bit field
 	 * is extended to 4:31 from 4:25.
 	 */
-	if (boot_cpu_data.x86 == 0x15 && boot_cpu_data.x86_model >= 0x60) {
+	if (is_carrizo_or_later()) {
 		running_avg_capture = val >> 4;
 		running_avg_capture = sign_extend32(running_avg_capture, 27);
 	} else {
@@ -111,7 +116,7 @@ static ssize_t show_power(struct device *dev,
 	 * On Carrizo and later platforms, ApmTdpLimit bit field
 	 * is extended to 16:31 from 16:28.
 	 */
-	if (boot_cpu_data.x86 == 0x15 && boot_cpu_data.x86_model >= 0x60)
+	if (is_carrizo_or_later())
 		tdp_limit = val >> 16;
 	else
 		tdp_limit = (val >> 16) & 0x1fff;
-- 
1.9.1

[lm-sensors] [PATCH v4 6/6] hwmon: (fam15h_power) Add platform check function

[Huang Rui]

This patch adds a platform check function to make code more readable.

Signed-off-by: Huang Rui <ray.huang@amd.com>
---
 drivers/hwmon/fam15h_power.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
index 35800cd..a0bea56 100644
--- a/drivers/hwmon/fam15h_power.c
+++ b/drivers/hwmon/fam15h_power.c
@@ -78,6 +78,11 @@ struct fam15h_power_data {
 	unsigned long power_period;
 };
 
+static bool is_carrizo_or_later(void)
+{
+	return boot_cpu_data.x86 = 0x15 && boot_cpu_data.x86_model >= 0x60;
+}
+
 static ssize_t show_power(struct device *dev,
 			  struct device_attribute *attr, char *buf)
 {
@@ -94,7 +99,7 @@ static ssize_t show_power(struct device *dev,
 	 * On Carrizo and later platforms, TdpRunAvgAccCap bit field
 	 * is extended to 4:31 from 4:25.
 	 */
-	if (boot_cpu_data.x86 = 0x15 && boot_cpu_data.x86_model >= 0x60) {
+	if (is_carrizo_or_later()) {
 		running_avg_capture = val >> 4;
 		running_avg_capture = sign_extend32(running_avg_capture, 27);
 	} else {
@@ -111,7 +116,7 @@ static ssize_t show_power(struct device *dev,
 	 * On Carrizo and later platforms, ApmTdpLimit bit field
 	 * is extended to 16:31 from 16:28.
 	 */
-	if (boot_cpu_data.x86 = 0x15 && boot_cpu_data.x86_model >= 0x60)
+	if (is_carrizo_or_later())
 		tdp_limit = val >> 16;
 	else
 		tdp_limit = (val >> 16) & 0x1fff;
-- 
1.9.1


_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [PATCH v4 2/6] hwmon: (fam15h_power) Add compute unit accumulated power

[Guenter Roeck]

On 03/10/2016 06:17 PM, Huang Rui wrote:
> This patch adds a member in fam15h_power_data which specifies the
> compute unit accumulated power. It adds do_read_registers_on_cu to do
> all the read to all MSRs and run it on one of the online cores on each
> compute unit with smp_call_function_many(). This behavior can decrease
> IPI numbers.
>
> Suggested-by: Borislav Petkov <bp@alien8.de>
> Signed-off-by: Huang Rui <ray.huang@amd.com>
> ---
>   drivers/hwmon/fam15h_power.c | 61 +++++++++++++++++++++++++++++++++++++++++++-
>   1 file changed, 60 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
> index 4f695d8..c5e2297 100644
> --- a/drivers/hwmon/fam15h_power.c
> +++ b/drivers/hwmon/fam15h_power.c
> @@ -25,6 +25,8 @@
>   #include <linux/module.h>
>   #include <linux/pci.h>
>   #include <linux/bitops.h>
> +#include <linux/cpu.h>
> +#include <linux/cpumask.h>
>   #include <asm/processor.h>
>   #include <asm/msr.h>
>
> @@ -44,7 +46,9 @@ MODULE_LICENSE("GPL");
>
>   #define FAM15H_MIN_NUM_ATTRS		2
>   #define FAM15H_NUM_GROUPS		2
> +#define MAX_CUS				8
>
> +#define MSR_F15H_CU_PWR_ACCUMULATOR	0xc001007a
>   #define MSR_F15H_CU_MAX_PWR_ACCUMULATOR	0xc001007b
>
>   #define PCI_DEVICE_ID_AMD_15H_M70H_NB_F4 0x15b4
> @@ -59,6 +63,8 @@ struct fam15h_power_data {
>   	struct attribute_group group;
>   	/* maximum accumulated power of a compute unit */
>   	u64 max_cu_acc_power;
> +	/* accumulated power of the compute units */
> +	u64 cu_acc_power[MAX_CUS];
>   };
>
>   static ssize_t show_power(struct device *dev,
> @@ -125,6 +131,59 @@ static ssize_t show_power_crit(struct device *dev,
>   }
>   static DEVICE_ATTR(power1_crit, S_IRUGO, show_power_crit, NULL);
>
> +static void do_read_registers_on_cu(void *_data)
> +{
> +	struct fam15h_power_data *data = _data;
> +	int cpu, cu;
> +
> +	cpu = smp_processor_id();
> +
> +	cu = cpu / smp_num_siblings;
> +

If smp is not configured:

drivers/hwmon/fam15h_power.c: In function ‘do_read_registers_on_cu’:
drivers/hwmon/fam15h_power.c:144:13: error: ‘smp_num_siblings’ undeclared (first use in this function)

Guenter

> +	rdmsrl_safe(MSR_F15H_CU_PWR_ACCUMULATOR, &data->cu_acc_power[cu]);
> +}
> +
> +/*
> + * This function is only able to be called when CPUID
> + * Fn8000_0007:EDX[12] is set.
> + */
> +static int read_registers(struct fam15h_power_data *data)
> +{
> +	int this_cpu, ret, cpu;
> +	int target;
> +	cpumask_var_t mask;
> +
> +	ret = zalloc_cpumask_var(&mask, GFP_KERNEL);
> +	if (!ret)
> +		return -ENOMEM;
> +
> +	get_online_cpus();
> +	this_cpu = get_cpu();
> +
> +	/*
> +	 * Choose the first online core of each compute unit, and then
> +	 * read their MSR value of power and ptsc in a single IPI,
> +	 * because the MSR value of CPU core represent the compute
> +	 * unit's.
> +	 */
> +	for_each_online_cpu(cpu) {
> +		target = cpumask_first(topology_sibling_cpumask(cpu));
> +		if (!cpumask_test_cpu(target, mask))
> +			cpumask_set_cpu(target, mask);
> +	}
> +
> +	if (cpumask_test_cpu(this_cpu, mask))
> +		do_read_registers_on_cu(data);
> +
> +	smp_call_function_many(mask, do_read_registers_on_cu, data, true);
> +	put_cpu();
> +	put_online_cpus();
> +
> +	free_cpumask_var(mask);
> +
> +	return 0;
> +}
> +
>   static int fam15h_power_init_attrs(struct pci_dev *pdev,
>   				   struct fam15h_power_data *data)
>   {
> @@ -263,7 +322,7 @@ static int fam15h_power_init_data(struct pci_dev *f4,
>
>   	data->max_cu_acc_power = tmp;
>
> -	return 0;
> +	return read_registers(data);
>   }
>
>   static int fam15h_power_probe(struct pci_dev *pdev,
>

Re: [lm-sensors] [PATCH v4 2/6] hwmon: (fam15h_power) Add compute unit accumulated power

[Guenter Roeck]

On 03/10/2016 06:17 PM, Huang Rui wrote:
> This patch adds a member in fam15h_power_data which specifies the
> compute unit accumulated power. It adds do_read_registers_on_cu to do
> all the read to all MSRs and run it on one of the online cores on each
> compute unit with smp_call_function_many(). This behavior can decrease
> IPI numbers.
>
> Suggested-by: Borislav Petkov <bp@alien8.de>
> Signed-off-by: Huang Rui <ray.huang@amd.com>
> ---
>   drivers/hwmon/fam15h_power.c | 61 +++++++++++++++++++++++++++++++++++++++++++-
>   1 file changed, 60 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
> index 4f695d8..c5e2297 100644
> --- a/drivers/hwmon/fam15h_power.c
> +++ b/drivers/hwmon/fam15h_power.c
> @@ -25,6 +25,8 @@
>   #include <linux/module.h>
>   #include <linux/pci.h>
>   #include <linux/bitops.h>
> +#include <linux/cpu.h>
> +#include <linux/cpumask.h>
>   #include <asm/processor.h>
>   #include <asm/msr.h>
>
> @@ -44,7 +46,9 @@ MODULE_LICENSE("GPL");
>
>   #define FAM15H_MIN_NUM_ATTRS		2
>   #define FAM15H_NUM_GROUPS		2
> +#define MAX_CUS				8
>
> +#define MSR_F15H_CU_PWR_ACCUMULATOR	0xc001007a
>   #define MSR_F15H_CU_MAX_PWR_ACCUMULATOR	0xc001007b
>
>   #define PCI_DEVICE_ID_AMD_15H_M70H_NB_F4 0x15b4
> @@ -59,6 +63,8 @@ struct fam15h_power_data {
>   	struct attribute_group group;
>   	/* maximum accumulated power of a compute unit */
>   	u64 max_cu_acc_power;
> +	/* accumulated power of the compute units */
> +	u64 cu_acc_power[MAX_CUS];
>   };
>
>   static ssize_t show_power(struct device *dev,
> @@ -125,6 +131,59 @@ static ssize_t show_power_crit(struct device *dev,
>   }
>   static DEVICE_ATTR(power1_crit, S_IRUGO, show_power_crit, NULL);
>
> +static void do_read_registers_on_cu(void *_data)
> +{
> +	struct fam15h_power_data *data = _data;
> +	int cpu, cu;
> +
> +	cpu = smp_processor_id();
> +
> +	cu = cpu / smp_num_siblings;
> +

If smp is not configured:

drivers/hwmon/fam15h_power.c: In function ‘do_read_registers_on_cu’:
drivers/hwmon/fam15h_power.c:144:13: error: ‘smp_num_siblings’ undeclared (first use in this function)

Guenter

> +	rdmsrl_safe(MSR_F15H_CU_PWR_ACCUMULATOR, &data->cu_acc_power[cu]);
> +}
> +
> +/*
> + * This function is only able to be called when CPUID
> + * Fn8000_0007:EDX[12] is set.
> + */
> +static int read_registers(struct fam15h_power_data *data)
> +{
> +	int this_cpu, ret, cpu;
> +	int target;
> +	cpumask_var_t mask;
> +
> +	ret = zalloc_cpumask_var(&mask, GFP_KERNEL);
> +	if (!ret)
> +		return -ENOMEM;
> +
> +	get_online_cpus();
> +	this_cpu = get_cpu();
> +
> +	/*
> +	 * Choose the first online core of each compute unit, and then
> +	 * read their MSR value of power and ptsc in a single IPI,
> +	 * because the MSR value of CPU core represent the compute
> +	 * unit's.
> +	 */
> +	for_each_online_cpu(cpu) {
> +		target = cpumask_first(topology_sibling_cpumask(cpu));
> +		if (!cpumask_test_cpu(target, mask))
> +			cpumask_set_cpu(target, mask);
> +	}
> +
> +	if (cpumask_test_cpu(this_cpu, mask))
> +		do_read_registers_on_cu(data);
> +
> +	smp_call_function_many(mask, do_read_registers_on_cu, data, true);
> +	put_cpu();
> +	put_online_cpus();
> +
> +	free_cpumask_var(mask);
> +
> +	return 0;
> +}
> +
>   static int fam15h_power_init_attrs(struct pci_dev *pdev,
>   				   struct fam15h_power_data *data)
>   {
> @@ -263,7 +322,7 @@ static int fam15h_power_init_data(struct pci_dev *f4,
>
>   	data->max_cu_acc_power = tmp;
>
> -	return 0;
> +	return read_registers(data);
>   }
>
>   static int fam15h_power_probe(struct pci_dev *pdev,
>


_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [PATCH v4 2/6] hwmon: (fam15h_power) Add compute unit accumulated power

[Huang Rui]

On Fri, Mar 11, 2016 at 07:33:55PM -0800, Guenter Roeck wrote:
> On 03/10/2016 06:17 PM, Huang Rui wrote:
> >This patch adds a member in fam15h_power_data which specifies the
> >compute unit accumulated power. It adds do_read_registers_on_cu to do
> >all the read to all MSRs and run it on one of the online cores on each
> >compute unit with smp_call_function_many(). This behavior can decrease
> >IPI numbers.
> >
> >Suggested-by: Borislav Petkov <bp@alien8.de>
> >Signed-off-by: Huang Rui <ray.huang@amd.com>
> >---
> >  drivers/hwmon/fam15h_power.c | 61 +++++++++++++++++++++++++++++++++++++++++++-
> >  1 file changed, 60 insertions(+), 1 deletion(-)
> >
> >diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
> >index 4f695d8..c5e2297 100644
> >--- a/drivers/hwmon/fam15h_power.c
> >+++ b/drivers/hwmon/fam15h_power.c
> >@@ -25,6 +25,8 @@
> >  #include <linux/module.h>
> >  #include <linux/pci.h>
> >  #include <linux/bitops.h>
> >+#include <linux/cpu.h>
> >+#include <linux/cpumask.h>
> >  #include <asm/processor.h>
> >  #include <asm/msr.h>
> >
> >@@ -44,7 +46,9 @@ MODULE_LICENSE("GPL");
> >
> >  #define FAM15H_MIN_NUM_ATTRS		2
> >  #define FAM15H_NUM_GROUPS		2
> >+#define MAX_CUS				8
> >
> >+#define MSR_F15H_CU_PWR_ACCUMULATOR	0xc001007a
> >  #define MSR_F15H_CU_MAX_PWR_ACCUMULATOR	0xc001007b
> >
> >  #define PCI_DEVICE_ID_AMD_15H_M70H_NB_F4 0x15b4
> >@@ -59,6 +63,8 @@ struct fam15h_power_data {
> >  	struct attribute_group group;
> >  	/* maximum accumulated power of a compute unit */
> >  	u64 max_cu_acc_power;
> >+	/* accumulated power of the compute units */
> >+	u64 cu_acc_power[MAX_CUS];
> >  };
> >
> >  static ssize_t show_power(struct device *dev,
> >@@ -125,6 +131,59 @@ static ssize_t show_power_crit(struct device *dev,
> >  }
> >  static DEVICE_ATTR(power1_crit, S_IRUGO, show_power_crit, NULL);
> >
> >+static void do_read_registers_on_cu(void *_data)
> >+{
> >+	struct fam15h_power_data *data = _data;
> >+	int cpu, cu;
> >+
> >+	cpu = smp_processor_id();
> >+
> >+	cu = cpu / smp_num_siblings;
> >+
> 
> If smp is not configured:
> 
> drivers/hwmon/fam15h_power.c: In function ?do_read_registers_on_cu?:
> drivers/hwmon/fam15h_power.c:144:13: error: ?smp_num_siblings? undeclared (first use in this function)
> 

Nice catch, how about define a static variable like below:

#ifdef CONFIG_SMP
static int cores_per_cu = smp_num_siblings;
#else
static int cores_per_cu = 1;
#endif

Thanks,
Rui

Re: [lm-sensors] [PATCH v4 2/6] hwmon: (fam15h_power) Add compute unit accumulated power

[Huang Rui]

On Fri, Mar 11, 2016 at 07:33:55PM -0800, Guenter Roeck wrote:
> On 03/10/2016 06:17 PM, Huang Rui wrote:
> >This patch adds a member in fam15h_power_data which specifies the
> >compute unit accumulated power. It adds do_read_registers_on_cu to do
> >all the read to all MSRs and run it on one of the online cores on each
> >compute unit with smp_call_function_many(). This behavior can decrease
> >IPI numbers.
> >
> >Suggested-by: Borislav Petkov <bp@alien8.de>
> >Signed-off-by: Huang Rui <ray.huang@amd.com>
> >---
> >  drivers/hwmon/fam15h_power.c | 61 +++++++++++++++++++++++++++++++++++++++++++-
> >  1 file changed, 60 insertions(+), 1 deletion(-)
> >
> >diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
> >index 4f695d8..c5e2297 100644
> >--- a/drivers/hwmon/fam15h_power.c
> >+++ b/drivers/hwmon/fam15h_power.c
> >@@ -25,6 +25,8 @@
> >  #include <linux/module.h>
> >  #include <linux/pci.h>
> >  #include <linux/bitops.h>
> >+#include <linux/cpu.h>
> >+#include <linux/cpumask.h>
> >  #include <asm/processor.h>
> >  #include <asm/msr.h>
> >
> >@@ -44,7 +46,9 @@ MODULE_LICENSE("GPL");
> >
> >  #define FAM15H_MIN_NUM_ATTRS		2
> >  #define FAM15H_NUM_GROUPS		2
> >+#define MAX_CUS				8
> >
> >+#define MSR_F15H_CU_PWR_ACCUMULATOR	0xc001007a
> >  #define MSR_F15H_CU_MAX_PWR_ACCUMULATOR	0xc001007b
> >
> >  #define PCI_DEVICE_ID_AMD_15H_M70H_NB_F4 0x15b4
> >@@ -59,6 +63,8 @@ struct fam15h_power_data {
> >  	struct attribute_group group;
> >  	/* maximum accumulated power of a compute unit */
> >  	u64 max_cu_acc_power;
> >+	/* accumulated power of the compute units */
> >+	u64 cu_acc_power[MAX_CUS];
> >  };
> >
> >  static ssize_t show_power(struct device *dev,
> >@@ -125,6 +131,59 @@ static ssize_t show_power_crit(struct device *dev,
> >  }
> >  static DEVICE_ATTR(power1_crit, S_IRUGO, show_power_crit, NULL);
> >
> >+static void do_read_registers_on_cu(void *_data)
> >+{
> >+	struct fam15h_power_data *data = _data;
> >+	int cpu, cu;
> >+
> >+	cpu = smp_processor_id();
> >+
> >+	cu = cpu / smp_num_siblings;
> >+
> 
> If smp is not configured:
> 
> drivers/hwmon/fam15h_power.c: In function ?do_read_registers_on_cu?:
> drivers/hwmon/fam15h_power.c:144:13: error: ?smp_num_siblings? undeclared (first use in this function)
> 

Nice catch, how about define a static variable like below:

#ifdef CONFIG_SMP
static int cores_per_cu = smp_num_siblings;
#else
static int cores_per_cu = 1;
#endif

Thanks,
Rui

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors

Re: [PATCH v4 2/6] hwmon: (fam15h_power) Add compute unit accumulated power

[Borislav Petkov]

On Sat, Mar 12, 2016 at 11:35:24PM +0800, Huang Rui wrote:
> On Fri, Mar 11, 2016 at 07:33:55PM -0800, Guenter Roeck wrote:
> > On 03/10/2016 06:17 PM, Huang Rui wrote:
> > >This patch adds a member in fam15h_power_data which specifies the
> > >compute unit accumulated power. It adds do_read_registers_on_cu to do
> > >all the read to all MSRs and run it on one of the online cores on each
> > >compute unit with smp_call_function_many(). This behavior can decrease
> > >IPI numbers.
> > >
> > >Suggested-by: Borislav Petkov <bp@alien8.de>
> > >Signed-off-by: Huang Rui <ray.huang@amd.com>
> > >---
> > >  drivers/hwmon/fam15h_power.c | 61 +++++++++++++++++++++++++++++++++++++++++++-
> > >  1 file changed, 60 insertions(+), 1 deletion(-)
> > >
> > >diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
> > >index 4f695d8..c5e2297 100644
> > >--- a/drivers/hwmon/fam15h_power.c
> > >+++ b/drivers/hwmon/fam15h_power.c
> > >@@ -25,6 +25,8 @@
> > >  #include <linux/module.h>
> > >  #include <linux/pci.h>
> > >  #include <linux/bitops.h>
> > >+#include <linux/cpu.h>
> > >+#include <linux/cpumask.h>
> > >  #include <asm/processor.h>
> > >  #include <asm/msr.h>
> > >
> > >@@ -44,7 +46,9 @@ MODULE_LICENSE("GPL");
> > >
> > >  #define FAM15H_MIN_NUM_ATTRS		2
> > >  #define FAM15H_NUM_GROUPS		2
> > >+#define MAX_CUS				8
> > >
> > >+#define MSR_F15H_CU_PWR_ACCUMULATOR	0xc001007a
> > >  #define MSR_F15H_CU_MAX_PWR_ACCUMULATOR	0xc001007b
> > >
> > >  #define PCI_DEVICE_ID_AMD_15H_M70H_NB_F4 0x15b4
> > >@@ -59,6 +63,8 @@ struct fam15h_power_data {
> > >  	struct attribute_group group;
> > >  	/* maximum accumulated power of a compute unit */
> > >  	u64 max_cu_acc_power;
> > >+	/* accumulated power of the compute units */
> > >+	u64 cu_acc_power[MAX_CUS];
> > >  };
> > >
> > >  static ssize_t show_power(struct device *dev,
> > >@@ -125,6 +131,59 @@ static ssize_t show_power_crit(struct device *dev,
> > >  }
> > >  static DEVICE_ATTR(power1_crit, S_IRUGO, show_power_crit, NULL);
> > >
> > >+static void do_read_registers_on_cu(void *_data)
> > >+{
> > >+	struct fam15h_power_data *data = _data;
> > >+	int cpu, cu;
> > >+
> > >+	cpu = smp_processor_id();
> > >+
> > >+	cu = cpu / smp_num_siblings;
> > >+
> > 
> > If smp is not configured:
> > 
> > drivers/hwmon/fam15h_power.c: In function ?do_read_registers_on_cu?:
> > drivers/hwmon/fam15h_power.c:144:13: error: ?smp_num_siblings? undeclared (first use in this function)
> > 
> 
> Nice catch, how about define a static variable like below:
> 
> #ifdef CONFIG_SMP
> static int cores_per_cu = smp_num_siblings;
> #else
> static int cores_per_cu = 1;
> #endif

I think a cleaner solution would be to move the three lines:

/* Number of siblings per CPU package */
int smp_num_siblings = 1;
EXPORT_SYMBOL(smp_num_siblings);

to a compilation unit which is built unconditionally because even in
the !CONFIG_SMP case, we have exactly 1 smp siblings per package and
smp_num_siblings is valid then too.

Thomas?

-- 
Regards/Gruss,
    Boris.

ECO tip #101: Trim your mails when you reply.

Re: [lm-sensors] [PATCH v4 2/6] hwmon: (fam15h_power) Add compute unit accumulated power

[Borislav Petkov]

On Sat, Mar 12, 2016 at 11:35:24PM +0800, Huang Rui wrote:
> On Fri, Mar 11, 2016 at 07:33:55PM -0800, Guenter Roeck wrote:
> > On 03/10/2016 06:17 PM, Huang Rui wrote:
> > >This patch adds a member in fam15h_power_data which specifies the
> > >compute unit accumulated power. It adds do_read_registers_on_cu to do
> > >all the read to all MSRs and run it on one of the online cores on each
> > >compute unit with smp_call_function_many(). This behavior can decrease
> > >IPI numbers.
> > >
> > >Suggested-by: Borislav Petkov <bp@alien8.de>
> > >Signed-off-by: Huang Rui <ray.huang@amd.com>
> > >---
> > >  drivers/hwmon/fam15h_power.c | 61 +++++++++++++++++++++++++++++++++++++++++++-
> > >  1 file changed, 60 insertions(+), 1 deletion(-)
> > >
> > >diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
> > >index 4f695d8..c5e2297 100644
> > >--- a/drivers/hwmon/fam15h_power.c
> > >+++ b/drivers/hwmon/fam15h_power.c
> > >@@ -25,6 +25,8 @@
> > >  #include <linux/module.h>
> > >  #include <linux/pci.h>
> > >  #include <linux/bitops.h>
> > >+#include <linux/cpu.h>
> > >+#include <linux/cpumask.h>
> > >  #include <asm/processor.h>
> > >  #include <asm/msr.h>
> > >
> > >@@ -44,7 +46,9 @@ MODULE_LICENSE("GPL");
> > >
> > >  #define FAM15H_MIN_NUM_ATTRS		2
> > >  #define FAM15H_NUM_GROUPS		2
> > >+#define MAX_CUS				8
> > >
> > >+#define MSR_F15H_CU_PWR_ACCUMULATOR	0xc001007a
> > >  #define MSR_F15H_CU_MAX_PWR_ACCUMULATOR	0xc001007b
> > >
> > >  #define PCI_DEVICE_ID_AMD_15H_M70H_NB_F4 0x15b4
> > >@@ -59,6 +63,8 @@ struct fam15h_power_data {
> > >  	struct attribute_group group;
> > >  	/* maximum accumulated power of a compute unit */
> > >  	u64 max_cu_acc_power;
> > >+	/* accumulated power of the compute units */
> > >+	u64 cu_acc_power[MAX_CUS];
> > >  };
> > >
> > >  static ssize_t show_power(struct device *dev,
> > >@@ -125,6 +131,59 @@ static ssize_t show_power_crit(struct device *dev,
> > >  }
> > >  static DEVICE_ATTR(power1_crit, S_IRUGO, show_power_crit, NULL);
> > >
> > >+static void do_read_registers_on_cu(void *_data)
> > >+{
> > >+	struct fam15h_power_data *data = _data;
> > >+	int cpu, cu;
> > >+
> > >+	cpu = smp_processor_id();
> > >+
> > >+	cu = cpu / smp_num_siblings;
> > >+
> > 
> > If smp is not configured:
> > 
> > drivers/hwmon/fam15h_power.c: In function ?do_read_registers_on_cu?:
> > drivers/hwmon/fam15h_power.c:144:13: error: ?smp_num_siblings? undeclared (first use in this function)
> > 
> 
> Nice catch, how about define a static variable like below:
> 
> #ifdef CONFIG_SMP
> static int cores_per_cu = smp_num_siblings;
> #else
> static int cores_per_cu = 1;
> #endif

I think a cleaner solution would be to move the three lines:

/* Number of siblings per CPU package */
int smp_num_siblings = 1;
EXPORT_SYMBOL(smp_num_siblings);

to a compilation unit which is built unconditionally because even in
the !CONFIG_SMP case, we have exactly 1 smp siblings per package and
smp_num_siblings is valid then too.

Thomas?

-- 
Regards/Gruss,
    Boris.

ECO tip #101: Trim your mails when you reply.

_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors