path: root/meta-openbmc-mods/meta-common/recipes-kernel/linux/linux-aspeed/0090-peci-cpupower-driver-1.patch

From e40f7d10d5306c075e2fc7e538ce1efc70eeb448 Mon Sep 17 00:00:00 2001
From: ZhikuiRen <zhikui.ren@intel.com>
Date: Thu, 9 Jan 2020 10:48:00 -0800
Subject: [PATCH] Add peci-cpupower driver

peci-cpupower reads CPU energy counter through peci
and computes average power in mW since last read.

Signed-off-by: ZhikuiRen <zhikui.ren@intel.com>
---
 Documentation/hwmon/index.rst         |   1 +
 Documentation/hwmon/peci-cpupower.rst |  52 ++++++++
 arch/arm/configs/aspeed_g5_defconfig  |   1 +
 drivers/hwmon/Kconfig                 |  14 +++
 drivers/hwmon/Makefile                |   1 +
 drivers/hwmon/peci-cpupower.c         | 231 ++++++++++++++++++++++++++++++++++
 drivers/mfd/intel-peci-client.c       |   1 +
 include/uapi/linux/peci-ioctl.h       |   1 +
 8 files changed, 302 insertions(+)
 create mode 100644 Documentation/hwmon/peci-cpupower.rst
 create mode 100644 drivers/hwmon/peci-cpupower.c

diff --git a/Documentation/hwmon/index.rst b/Documentation/hwmon/index.rst
index 7d894c9..0bde0ef 100644
--- a/Documentation/hwmon/index.rst
+++ b/Documentation/hwmon/index.rst
@@ -130,6 +130,7 @@ Hardware Monitoring Kernel Drivers
    pcf8591
    peci-cputemp
    peci-dimmtemp
+   peci-cpupower
    pmbus
    powr1220
    pxe1610
diff --git a/Documentation/hwmon/peci-cpupower.rst b/Documentation/hwmon/peci-cpupower.rst
new file mode 100644
index 0000000..4d7bd61
--- /dev/null
+++ b/Documentation/hwmon/peci-cpupower.rst
@@ -0,0 +1,52 @@
+.. SPDX-License-Identifier: GPL-2.0
+
+Kernel driver peci-cpupower
+==========================
+
+:Copyright: |copy| 2018-2020 Intel Corporation
+
+Supported chips:
+	One of Intel server CPUs listed below which is connected to a PECI bus.
+		* Intel Xeon E5/E7 v3 server processors
+			Intel Xeon E5-14xx v3 family
+			Intel Xeon E5-24xx v3 family
+			Intel Xeon E5-16xx v3 family
+			Intel Xeon E5-26xx v3 family
+			Intel Xeon E5-46xx v3 family
+			Intel Xeon E7-48xx v3 family
+			Intel Xeon E7-88xx v3 family
+		* Intel Xeon E5/E7 v4 server processors
+			Intel Xeon E5-16xx v4 family
+			Intel Xeon E5-26xx v4 family
+			Intel Xeon E5-46xx v4 family
+			Intel Xeon E7-48xx v4 family
+			Intel Xeon E7-88xx v4 family
+		* Intel Xeon Scalable server processors
+			Intel Xeon D family
+			Intel Xeon Bronze family
+			Intel Xeon Silver family
+			Intel Xeon Gold family
+			Intel Xeon Platinum family
+
+	Addresses scanned: PECI client address 0x30 - 0x37
+	Datasheet: Available from http://www.intel.com/design/literature.htm
+
+Author:
+	Zhikui Ren <zhikui.ren@.intel.com>
+
+Description
+-----------
+
+This driver implements a generic PECI hwmon feature which provides
+average power consumption readings of the CPU package based on energy counter
+accessible using the PECI Client Command Suite via the processor PECI client.
+
+Power values are average power since last measure given in milli Watt and
+will be measurable only when the target CPU is powered on.
+
+``sysfs`` interface
+-------------------
+======================= =======================================================
+power1_average		Provides average power since last read in milli Watt.
+power1_label		Provides string "Average Power".
+======================= =======================================================
diff --git a/arch/arm/configs/aspeed_g5_defconfig b/arch/arm/configs/aspeed_g5_defconfig
index 1e589a0..0e50040 100644
--- a/arch/arm/configs/aspeed_g5_defconfig
+++ b/arch/arm/configs/aspeed_g5_defconfig
@@ -177,6 +177,7 @@ CONFIG_SENSORS_OCC_P8_I2C=y
 CONFIG_SENSORS_OCC_P9_SBE=y
 CONFIG_SENSORS_PECI_CPUTEMP=y
 CONFIG_SENSORS_PECI_DIMMTEMP=y
+CONFIG_SENSORS_PECI_CPUPOWER=y
 CONFIG_PMBUS=y
 CONFIG_SENSORS_ADM1275=y
 CONFIG_SENSORS_IBM_CFFPS=y
diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
index 8312b37..07d8826 100644
--- a/drivers/hwmon/Kconfig
+++ b/drivers/hwmon/Kconfig
@@ -1361,6 +1361,20 @@ config SENSORS_PECI_DIMMTEMP
 	  This driver can also be built as a module. If so, the module
 	  will be called peci-dimmtemp.
 
+config SENSORS_PECI_CPUPOWER
+	tristate "PECI CPU power monitoring support"
+	depends on PECI
+	select MFD_INTEL_PECI_CLIENT
+	help
+	  If you say yes here you get support for the generic Intel PECI
+	  cputemp driver which provides average engergy
+	  readings of the CPU package using
+	  the PECI Client Command Suite via the processor PECI client.
+	  Check Documentation/hwmon/peci-cpupower for details.
+
+	  This driver can also be built as a module. If so, the module
+	  will be called peci-cpupower.
+
 source "drivers/hwmon/pmbus/Kconfig"
 
 config SENSORS_PWM_FAN
diff --git a/drivers/hwmon/Makefile b/drivers/hwmon/Makefile
index e74ea92..fab43fd 100644
--- a/drivers/hwmon/Makefile
+++ b/drivers/hwmon/Makefile
@@ -144,6 +144,7 @@ obj-$(CONFIG_SENSORS_PC87427)	+= pc87427.o
 obj-$(CONFIG_SENSORS_PCF8591)	+= pcf8591.o
 obj-$(CONFIG_SENSORS_PECI_CPUTEMP)	+= peci-cputemp.o
 obj-$(CONFIG_SENSORS_PECI_DIMMTEMP)	+= peci-dimmtemp.o
+obj-$(CONFIG_SENSORS_PECI_CPUPOWER)	+= peci-cpupower.o
 obj-$(CONFIG_SENSORS_POWR1220)  += powr1220.o
 obj-$(CONFIG_SENSORS_PWM_FAN)	+= pwm-fan.o
 obj-$(CONFIG_SENSORS_RASPBERRYPI_HWMON)	+= raspberrypi-hwmon.o
diff --git a/drivers/hwmon/peci-cpupower.c b/drivers/hwmon/peci-cpupower.c
new file mode 100644
index 0000000..6907696
--- /dev/null
+++ b/drivers/hwmon/peci-cpupower.c
@@ -0,0 +1,231 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (c) 2018-2020 Intel Corporation
+
+#include <linux/hwmon.h>
+#include <linux/jiffies.h>
+#include <linux/mfd/intel-peci-client.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include "peci-hwmon.h"
+
+#define POWER_DEFAULT_CHANNEL_NUMS   1
+
+struct peci_cpupower {
+	struct peci_client_manager *mgr;
+	struct device *dev;
+	char name[PECI_NAME_SIZE];
+	const struct cpu_gen_info *gen_info;
+	struct peci_sensor_data energy;
+	long avg_power_val;
+	u64 core_mask;
+	u32 power_config[POWER_DEFAULT_CHANNEL_NUMS + 1];
+	uint config_idx;
+	struct hwmon_channel_info power_info;
+	const struct hwmon_channel_info *info[2];
+	struct hwmon_chip_info chip;
+};
+
+enum cpupower_channels {
+	average_power,
+};
+
+static const u32 config_table[POWER_DEFAULT_CHANNEL_NUMS] = {
+	/* average power */
+	HWMON_P_LABEL | HWMON_P_AVERAGE,
+};
+
+static const char *cpupower_label[POWER_DEFAULT_CHANNEL_NUMS] = {
+	"Average Power",
+};
+
+static int get_average_power(struct peci_cpupower *priv)
+{
+	u8  pkg_cfg[4];
+	int ret;
+
+	if (!peci_sensor_need_update(&priv->energy))
+		return 0;
+
+	ret = peci_client_read_package_config(priv->mgr,
+					      PECI_MBX_INDEX_TDP_UNITS,
+					      PECI_PKG_ID_PKG_ENERGY_STATUS,
+					      pkg_cfg);
+
+	u32 power_unit = ((le32_to_cpup((__le32 *)pkg_cfg)) & 0x1f00) >> 8;
+
+	dev_dbg(priv->dev, "cpupower units %d  (1J/pow(2, unit))\n",
+		power_unit);
+
+	ret = peci_client_read_package_config(priv->mgr,
+					      PECI_MBX_INDEX_ENERGY_COUNTER,
+					      PECI_PKG_ID_PKG_ENERGY_STATUS,
+					      pkg_cfg);
+	if (!ret) {
+		u32 energy_cnt = le32_to_cpup((__le32 *)pkg_cfg);
+		ulong jif = jiffies;
+		ulong elapsed = (jif - priv->energy.last_updated);
+		long power_val = 0;
+		/*
+		 * Don't calculate average power for first counter read or
+		 * counter wrapped around or last counter read was more than
+		 * 60 minutes ago (jiffies did not wrap and power calculation
+		 * does not overflow or underflow
+		 */
+		if (priv->energy.last_updated > 0 &&
+		    energy_cnt > priv->energy.value &&
+		    (elapsed < (HZ * 3600))) {
+			power_val = (long)(energy_cnt - priv->energy.value)
+				/ elapsed * HZ;
+			dev_dbg(priv->dev, "countDiff %d, jiffes elapsed %d, raw powerValue %d scale to %d mW\n",
+				(long)(energy_cnt - priv->energy.value),
+				elapsed, power_val,
+				power_val >> (power_unit - 10));
+		} else {
+			dev_dbg(priv->dev, "countDiff %d, jiffes elapsed %d, skipping calculate power, try agin\n",
+				(long)(energy_cnt - priv->energy.value),
+				elapsed);
+			ret = -EAGAIN;
+		}
+
+		priv->energy.value = energy_cnt;
+		priv->avg_power_val = power_val >> ((power_unit - 10));
+		peci_sensor_mark_updated(&priv->energy);
+
+		dev_dbg(priv->dev, "energy counter 0x%8x, average power %dmW, jif %u, HZ is %d jiffies\n",
+			priv->energy.value, priv->avg_power_val,
+			jif, HZ);
+	}
+	return ret;
+}
+
+static int cpupower_read_string(struct device *dev,
+				enum hwmon_sensor_types type,
+				u32 attr, int channel, const char **str)
+{
+	if (attr != hwmon_power_label)
+		return -EOPNOTSUPP;
+	if (channel >= POWER_DEFAULT_CHANNEL_NUMS)
+		return -EOPNOTSUPP;
+	*str = cpupower_label[channel];
+
+	return 0;
+}
+
+static int cpupower_read(struct device *dev,
+			 enum hwmon_sensor_types type,
+			 u32 attr, int channel, long *val)
+{
+	struct peci_cpupower *priv = dev_get_drvdata(dev);
+	int ret;
+
+	if (channel >= POWER_DEFAULT_CHANNEL_NUMS ||
+	    !(priv->power_config[channel] & BIT(attr)))
+		return -EOPNOTSUPP;
+
+	switch (attr) {
+	case hwmon_power_average:
+		switch (channel) {
+		case average_power:
+			ret = get_average_power(priv);
+			if (ret)
+				break;
+
+			*val = priv->avg_power_val;
+			break;
+		default:
+			break;
+		}
+		break;
+	default:
+		ret = -EOPNOTSUPP;
+		break;
+	}
+
+	return ret;
+}
+
+static umode_t cpupower_is_visible(const void *data,
+				   enum hwmon_sensor_types type,
+				   u32 attr, int channel)
+{
+	const struct peci_cpupower *priv = data;
+
+	if (channel < POWER_DEFAULT_CHANNEL_NUMS ||
+	    (priv->power_config[channel] & BIT(attr)))
+		return 0444;
+
+	return 0;
+}
+
+static const struct hwmon_ops cpupower_ops = {
+	.is_visible = cpupower_is_visible,
+	.read_string = cpupower_read_string,
+	.read = cpupower_read,
+};
+
+static int peci_cpupower_probe(struct platform_device *pdev)
+{
+	struct peci_client_manager *mgr = dev_get_drvdata(pdev->dev.parent);
+	struct device *dev = &pdev->dev;
+	struct peci_cpupower *priv;
+	struct device *hwmon_dev;
+
+	if ((mgr->client->adapter->cmd_mask &
+				(BIT(PECI_CMD_RD_PKG_CFG))) !=
+			(BIT(PECI_CMD_RD_PKG_CFG))) {
+		return -ENODEV;
+	}
+
+	priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
+	if (!priv)
+		return -ENOMEM;
+
+	dev_set_drvdata(dev, priv);
+	priv->mgr = mgr;
+	priv->dev = dev;
+	priv->gen_info = mgr->gen_info;
+
+	snprintf(priv->name, PECI_NAME_SIZE, "peci_cpupower.cpu%d",
+		 mgr->client->addr - PECI_BASE_ADDR);
+
+	priv->power_config[priv->config_idx++] = config_table[average_power];
+
+	priv->chip.ops = &cpupower_ops;
+	priv->chip.info = priv->info;
+
+	priv->info[0] = &priv->power_info;
+
+	priv->power_info.type = hwmon_power;
+	priv->power_info.config = priv->power_config;
+
+	hwmon_dev = devm_hwmon_device_register_with_info(priv->dev,
+							 priv->name,
+							 priv,
+							 &priv->chip,
+							 NULL);
+
+	if (IS_ERR(hwmon_dev))
+		return PTR_ERR(hwmon_dev);
+
+	dev_dbg(dev, "%s: sensor '%s'\n", dev_name(hwmon_dev), priv->name);
+
+	return 0;
+}
+
+static const struct platform_device_id peci_cpupower_ids[] = {
+	{ .name = "peci-cpupower", .driver_data = 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(platform, peci_cpupower_ids);
+
+static struct platform_driver peci_cpupower_driver = {
+	.probe    = peci_cpupower_probe,
+	.id_table = peci_cpupower_ids,
+	.driver   = { .name = KBUILD_MODNAME, },
+};
+module_platform_driver(peci_cpupower_driver);
+
+MODULE_AUTHOR("Zhikui Ren <zhikui.ren@intel.com>");
+MODULE_DESCRIPTION("PECI cpupower driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/mfd/intel-peci-client.c b/drivers/mfd/intel-peci-client.c
index 9751b04..0c62ad6 100644
--- a/drivers/mfd/intel-peci-client.c
+++ b/drivers/mfd/intel-peci-client.c
@@ -21,6 +21,7 @@
 static struct mfd_cell peci_functions[] = {
 	{ .name = "peci-cputemp", },
 	{ .name = "peci-dimmtemp", },
+	{ .name = "peci-cpupower", },
 };
 
 static const struct cpu_gen_info cpu_gen_info_table[] = {
diff --git a/include/uapi/linux/peci-ioctl.h b/include/uapi/linux/peci-ioctl.h
index 843930f..d16f7c9 100644
--- a/include/uapi/linux/peci-ioctl.h
+++ b/include/uapi/linux/peci-ioctl.h
@@ -231,6 +231,7 @@ struct peci_rd_pkg_cfg_msg {
 #define PECI_PKG_ID_MAX_THREAD_ID		0x0003  /* Max Thread ID */
 #define PECI_PKG_ID_MICROCODE_REV		0x0004  /* CPU Microcode Update Revision */
 #define PECI_PKG_ID_MACHINE_CHECK_STATUS	0x0005  /* Machine Check Status */
+#define PECI_PKG_ID_PKG_ENERGY_STATUS		0x00ff  /* Average Energy  */
 
 	__u8	rx_len;
 	__u8	cc;
-- 
2.7.4