From 40626a1bf657eef557fcee9e1b8ef5b4f5b56dcd Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Sun, 29 Apr 2018 08:08:24 -0700 Subject: [PATCH 1/3] hwmon: (k10temp) Fix reading critical temperature register The HTC (Hardware Temperature Control) register has moved for recent chips. Cc: stable@vger.kernel.org # v4.16+ Tested-by: Gabriel Craciunescu Signed-off-by: Guenter Roeck --- drivers/hwmon/k10temp.c | 40 ++++++++++++++++++++++++++++++---------- 1 file changed, 30 insertions(+), 10 deletions(-) diff --git a/drivers/hwmon/k10temp.c b/drivers/hwmon/k10temp.c index d2cc55e21374..34b5448b00be 100644 --- a/drivers/hwmon/k10temp.c +++ b/drivers/hwmon/k10temp.c @@ -63,10 +63,12 @@ static DEFINE_MUTEX(nb_smu_ind_mutex); #define NB_CAP_HTC 0x00000400 /* - * For F15h M60h, functionality of REG_REPORTED_TEMPERATURE - * has been moved to D0F0xBC_xD820_0CA4 [Reported Temperature - * Control] + * For F15h M60h and M70h, REG_HARDWARE_THERMAL_CONTROL + * and REG_REPORTED_TEMPERATURE have been moved to + * D0F0xBC_xD820_0C64 [Hardware Temperature Control] + * D0F0xBC_xD820_0CA4 [Reported Temperature Control] */ +#define F15H_M60H_HARDWARE_TEMP_CTRL_OFFSET 0xd8200c64 #define F15H_M60H_REPORTED_TEMP_CTRL_OFFSET 0xd8200ca4 /* F17h M01h Access througn SMN */ @@ -74,6 +76,7 @@ static DEFINE_MUTEX(nb_smu_ind_mutex); struct k10temp_data { struct pci_dev *pdev; + void (*read_htcreg)(struct pci_dev *pdev, u32 *regval); void (*read_tempreg)(struct pci_dev *pdev, u32 *regval); int temp_offset; u32 temp_adjust_mask; @@ -98,6 +101,11 @@ static const struct tctl_offset tctl_offset_table[] = { { 0x17, "AMD Ryzen Threadripper 1910", 10000 }, }; +static void read_htcreg_pci(struct pci_dev *pdev, u32 *regval) +{ + pci_read_config_dword(pdev, REG_HARDWARE_THERMAL_CONTROL, regval); +} + static void read_tempreg_pci(struct pci_dev *pdev, u32 *regval) { pci_read_config_dword(pdev, REG_REPORTED_TEMPERATURE, regval); @@ -114,6 +122,12 @@ static void amd_nb_index_read(struct pci_dev *pdev, unsigned int devfn, mutex_unlock(&nb_smu_ind_mutex); } +static void read_htcreg_nb_f15(struct pci_dev *pdev, u32 *regval) +{ + amd_nb_index_read(pdev, PCI_DEVFN(0, 0), 0xb8, + F15H_M60H_HARDWARE_TEMP_CTRL_OFFSET, regval); +} + static void read_tempreg_nb_f15(struct pci_dev *pdev, u32 *regval) { amd_nb_index_read(pdev, PCI_DEVFN(0, 0), 0xb8, @@ -160,8 +174,7 @@ static ssize_t show_temp_crit(struct device *dev, u32 regval; int value; - pci_read_config_dword(data->pdev, - REG_HARDWARE_THERMAL_CONTROL, ®val); + data->read_htcreg(data->pdev, ®val); value = ((regval >> 16) & 0x7f) * 500 + 52000; if (show_hyst) value -= ((regval >> 24) & 0xf) * 500; @@ -181,13 +194,18 @@ static umode_t k10temp_is_visible(struct kobject *kobj, struct pci_dev *pdev = data->pdev; if (index >= 2) { - u32 reg_caps, reg_htc; + u32 reg; + + if (!data->read_htcreg) + return 0; pci_read_config_dword(pdev, REG_NORTHBRIDGE_CAPABILITIES, - ®_caps); - pci_read_config_dword(pdev, REG_HARDWARE_THERMAL_CONTROL, - ®_htc); - if (!(reg_caps & NB_CAP_HTC) || !(reg_htc & HTC_ENABLE)) + ®); + if (!(reg & NB_CAP_HTC)) + return 0; + + data->read_htcreg(data->pdev, ®); + if (!(reg & HTC_ENABLE)) return 0; } return attr->mode; @@ -268,11 +286,13 @@ static int k10temp_probe(struct pci_dev *pdev, if (boot_cpu_data.x86 == 0x15 && (boot_cpu_data.x86_model == 0x60 || boot_cpu_data.x86_model == 0x70)) { + data->read_htcreg = read_htcreg_nb_f15; data->read_tempreg = read_tempreg_nb_f15; } else if (boot_cpu_data.x86 == 0x17) { data->temp_adjust_mask = 0x80000; data->read_tempreg = read_tempreg_nb_f17; } else { + data->read_htcreg = read_htcreg_pci; data->read_tempreg = read_tempreg_pci; } From f9bc6b2dd9cf025f827f471769e1d88b527bfb91 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Fri, 4 May 2018 13:01:32 -0700 Subject: [PATCH 2/3] x86/amd_nb: Add support for Raven Ridge CPUs Add Raven Ridge root bridge and data fabric PCI IDs. This is required for amd_pci_dev_to_node_id() and amd_smn_read(). Cc: stable@vger.kernel.org # v4.16+ Tested-by: Gabriel Craciunescu Acked-by: Thomas Gleixner Signed-off-by: Guenter Roeck --- arch/x86/kernel/amd_nb.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/x86/kernel/amd_nb.c b/arch/x86/kernel/amd_nb.c index c88e0b127810..b481b95bd8f6 100644 --- a/arch/x86/kernel/amd_nb.c +++ b/arch/x86/kernel/amd_nb.c @@ -14,8 +14,11 @@ #include #define PCI_DEVICE_ID_AMD_17H_ROOT 0x1450 +#define PCI_DEVICE_ID_AMD_17H_M10H_ROOT 0x15d0 #define PCI_DEVICE_ID_AMD_17H_DF_F3 0x1463 #define PCI_DEVICE_ID_AMD_17H_DF_F4 0x1464 +#define PCI_DEVICE_ID_AMD_17H_M10H_DF_F3 0x15eb +#define PCI_DEVICE_ID_AMD_17H_M10H_DF_F4 0x15ec /* Protect the PCI config register pairs used for SMN and DF indirect access. */ static DEFINE_MUTEX(smn_mutex); @@ -24,6 +27,7 @@ static u32 *flush_words; static const struct pci_device_id amd_root_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_17H_ROOT) }, + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_17H_M10H_ROOT) }, {} }; @@ -39,6 +43,7 @@ const struct pci_device_id amd_nb_misc_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_16H_NB_F3) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_16H_M30H_NB_F3) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_17H_DF_F3) }, + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_17H_M10H_DF_F3) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CNB17H_F3) }, {} }; @@ -51,6 +56,7 @@ static const struct pci_device_id amd_nb_link_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_16H_NB_F4) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_16H_M30H_NB_F4) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_17H_DF_F4) }, + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_17H_M10H_DF_F4) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CNB17H_F4) }, {} }; From 3b031622f598481970400519bd5abc2a16708282 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Fri, 4 May 2018 13:01:33 -0700 Subject: [PATCH 3/3] hwmon: (k10temp) Use API function to access System Management Network The SMN (System Management Network) on Family 17h AMD CPUs is also accessed from other drivers, specifically EDAC. Accessing it directly is racy. On top of that, accessing the SMN through root bridge 00:00 is wrong on multi-die CPUs and may result in reading the temperature from the wrong die. Use available API functions to fix the problem. For this to work, add dependency on AMD_NB. Also change the Raven Ridge PCI device ID to point to Data Fabric Function 3, since this ID is used by the API functions to find the CPU node. Cc: stable@vger.kernel.org # v4.16+ Tested-by: Gabriel Craciunescu Signed-off-by: Guenter Roeck --- drivers/hwmon/Kconfig | 2 +- drivers/hwmon/k10temp.c | 11 ++++++----- 2 files changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig index f249a4428458..6ec307c93ece 100644 --- a/drivers/hwmon/Kconfig +++ b/drivers/hwmon/Kconfig @@ -272,7 +272,7 @@ config SENSORS_K8TEMP config SENSORS_K10TEMP tristate "AMD Family 10h+ temperature sensor" - depends on X86 && PCI + depends on X86 && PCI && AMD_NB help If you say yes here you get support for the temperature sensor(s) inside your CPU. Supported are later revisions of diff --git a/drivers/hwmon/k10temp.c b/drivers/hwmon/k10temp.c index 34b5448b00be..3b73dee6fdc6 100644 --- a/drivers/hwmon/k10temp.c +++ b/drivers/hwmon/k10temp.c @@ -23,6 +23,7 @@ #include #include #include +#include #include MODULE_DESCRIPTION("AMD Family 10h+ CPU core temperature monitor"); @@ -40,8 +41,8 @@ static DEFINE_MUTEX(nb_smu_ind_mutex); #define PCI_DEVICE_ID_AMD_17H_DF_F3 0x1463 #endif -#ifndef PCI_DEVICE_ID_AMD_17H_RR_NB -#define PCI_DEVICE_ID_AMD_17H_RR_NB 0x15d0 +#ifndef PCI_DEVICE_ID_AMD_17H_M10H_DF_F3 +#define PCI_DEVICE_ID_AMD_17H_M10H_DF_F3 0x15eb #endif /* CPUID function 0x80000001, ebx */ @@ -136,8 +137,8 @@ static void read_tempreg_nb_f15(struct pci_dev *pdev, u32 *regval) static void read_tempreg_nb_f17(struct pci_dev *pdev, u32 *regval) { - amd_nb_index_read(pdev, PCI_DEVFN(0, 0), 0x60, - F17H_M01H_REPORTED_TEMP_CTRL_OFFSET, regval); + amd_smn_read(amd_pci_dev_to_node_id(pdev), + F17H_M01H_REPORTED_TEMP_CTRL_OFFSET, regval); } static ssize_t temp1_input_show(struct device *dev, @@ -322,7 +323,7 @@ static const struct pci_device_id k10temp_id_table[] = { { PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_16H_NB_F3) }, { PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_16H_M30H_NB_F3) }, { PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_17H_DF_F3) }, - { PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_17H_RR_NB) }, + { PCI_VDEVICE(AMD, PCI_DEVICE_ID_AMD_17H_M10H_DF_F3) }, {} }; MODULE_DEVICE_TABLE(pci, k10temp_id_table);