838 lines · 21.9 KB · C
RawDownload
// SPDX-License-Identifier: GPL-2.0-only/* * coretemp.c - Linux kernel module for hardware monitoring * * Copyright (C) 2007 Rudolf Marek <r.marek@assembler.cz> * * Inspired from many hwmon drivers */ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/module.h>#include <linux/init.h>#include <linux/slab.h>#include <linux/jiffies.h>#include <linux/hwmon.h>#include <linux/sysfs.h>#include <linux/hwmon-sysfs.h>#include <linux/err.h>#include <linux/mutex.h>#include <linux/list.h>#include <linux/platform_device.h>#include <linux/cpu.h>#include <linux/smp.h>#include <linux/moduleparam.h>#include <linux/pci.h>#include <asm/msr.h>#include <linux/processor.h>#include <asm/cpu_device_id.h>#include <linux/sched/isolation.h> #define DRVNAME	"coretemp" /* * force_tjmax only matters when TjMax can't be read from the CPU itself. * When set, it replaces the driver's suboptimal heuristic. */static int force_tjmax;module_param_named(tjmax, force_tjmax, int, 0444);MODULE_PARM_DESC(tjmax, "TjMax value in degrees Celsius"); #define CORETEMP_NAME_LENGTH	28	/* String Length of attrs */ enum coretemp_attr_index {	ATTR_LABEL,	ATTR_CRIT_ALARM,	ATTR_TEMP,	ATTR_TJMAX,	ATTR_TTARGET,	MAX_CORE_ATTRS = ATTR_TJMAX + 1,	/* Maximum no of basic attrs */	TOTAL_ATTRS = ATTR_TTARGET + 1		/* Maximum no of possible attrs */}; #ifdef CONFIG_SMP#define for_each_sibling(i, cpu) \	for_each_cpu(i, topology_sibling_cpumask(cpu))#else#define for_each_sibling(i, cpu)	for (i = 0; false; )#endif /* * Per-Core Temperature Data * @tjmax: The static tjmax value when tjmax cannot be retrieved from *		IA32_TEMPERATURE_TARGET MSR. * @last_updated: The time when the current temperature value was updated *		earlier (in jiffies). * @cpu_core_id: The CPU Core from which temperature values should be read *		This value is passed as "id" field to rdmsr/wrmsr functions. * @status_reg: One of IA32_THERM_STATUS or IA32_PACKAGE_THERM_STATUS, *		from where the temperature values should be read. * @attr_size:  Number of sysfs attributes in this temperature data group. */struct temp_data {	int temp;	int tjmax;	unsigned long last_updated;	unsigned int cpu;	int index;	u32 cpu_core_id;	u32 status_reg;	int attr_size;	struct device_attribute sd_attrs[TOTAL_ATTRS];	char attr_name[TOTAL_ATTRS][CORETEMP_NAME_LENGTH];	struct attribute *attrs[TOTAL_ATTRS + 1];	struct attribute_group attr_group;	struct mutex update_lock;}; /* Platform Data per Physical CPU */struct platform_data {	struct device		*hwmon_dev;	u16			pkg_id;	int			nr_cores;	struct ida		ida;	struct cpumask		cpumask;	struct temp_data	*pkg_data;	struct temp_data	**core_data;	struct device_attribute name_attr;}; struct tjmax_pci {	unsigned int device;	int tjmax;}; static const struct tjmax_pci tjmax_pci_table[] = {	{ 0x0708, 110000 },	/* CE41x0 (Sodaville ) */	{ 0x0c72, 102000 },	/* Atom S1240 (Centerton) */	{ 0x0c73, 95000 },	/* Atom S1220 (Centerton) */	{ 0x0c75, 95000 },	/* Atom S1260 (Centerton) */}; struct tjmax {	char const *id;	int tjmax;}; static const struct tjmax tjmax_table[] = {	{ "CPU  230", 100000 },		/* Model 0x1c, stepping 2	*/	{ "CPU  330", 125000 },		/* Model 0x1c, stepping 2	*/}; struct tjmax_model {	u32 vfm;	u8 stepping_mask;	int tjmax;}; #define ANY 0xff static const struct tjmax_model tjmax_model_table[] = {	{ INTEL_ATOM_BONNELL,	      10,  100000 },	/* D4xx, K4xx, N4xx, D5xx, K5xx, N5xx */	{ INTEL_ATOM_BONNELL,	      ANY, 90000 },	/* Z5xx, N2xx, possibly others							 * Note: Also matches 230 and 330,							 * which are covered by tjmax_table							 */	{ INTEL_ATOM_BONNELL_MID,     ANY, 90000 },	/* Atom Tunnel Creek (Exx), Lincroft (Z6xx)							 * Note: TjMax for E6xxT is 110C, but CPU							 * type is undetectable by software							 */	{ INTEL_ATOM_SALTWELL_MID,    ANY, 90000 },	/* Atom Medfield (Z2460) */	{ INTEL_ATOM_SALTWELL_TABLET, ANY, 90000 },	/* Atom Clover Trail/Cloverview (Z27x0) */	{ INTEL_ATOM_SALTWELL,	      ANY, 100000 },	/* Atom Cedar Trail/Cedarview (N2xxx, D2xxx)							 * Also matches S12x0 (stepping 9), covered							 * by PCI table							 */	{ INTEL_ATOM_SILVERMONT,      9, 110000 },	/* Atom Bay Trail E38xx (embedded) */	{ INTEL_ATOM_SILVERMONT,      ANY, 90000 },	/* Atom Bay Trail Z37xx (tablet) */	{ INTEL_ATOM_SILVERMONT_MID,  ANY, 90000 },	/* Atom Merrifield (Z34xx) */	{ INTEL_ATOM_SILVERMONT_MID2, ANY, 90000 },	/* Atom Moorefield (Z35xx) */	{ INTEL_ATOM_AIRMONT,	      ANY, 90000 },	/* Atom Cherry Trail (Z8xxx) */	{ INTEL_ATOM_GOLDMONT,	      ANY, 105000 },	/* Atom Apollo Lake (J3xxx, N3xxx, E39xx) */	{ INTEL_ATOM_GOLDMONT_PLUS,   ANY, 105000 },	/* Atom Gemini Lake (J4xxx, N4xxx, N5xxx) */	{ INTEL_ATOM_TREMONT,	      ANY, 105000 },	/* Atom Elkhart Lake */	{ INTEL_ATOM_TREMONT_L,	      ANY, 105000 },	/* Atom Jasper Lake */}; static bool is_pkg_temp_data(struct temp_data *tdata){	return tdata->index < 0;} static int adjust_tjmax(struct cpuinfo_x86 *c, u32 id, struct device *dev){	/* The 100C is default for both mobile and non mobile CPUs */ 	int tjmax = 100000;	int tjmax_ee = 85000;	int usemsr_ee = 1;	int err;	u64 val;	int i;	u16 devfn = PCI_DEVFN(0, 0);	struct pci_dev *host_bridge = pci_get_domain_bus_and_slot(0, 0, devfn); 	/*	 * Explicit tjmax table entries override heuristics.	 * First try PCI host bridge IDs, followed by model ID strings	 * and model/stepping information.	 */	if (host_bridge && host_bridge->vendor == PCI_VENDOR_ID_INTEL) {		for (i = 0; i < ARRAY_SIZE(tjmax_pci_table); i++) {			if (host_bridge->device == tjmax_pci_table[i].device) {				pci_dev_put(host_bridge);				return tjmax_pci_table[i].tjmax;			}		}	}	pci_dev_put(host_bridge); 	/*	 * This is literally looking for "CPU  XXX" in the model string.	 * Not checking it against the model as well. Just purely a	 * string search.	 */	for (i = 0; i < ARRAY_SIZE(tjmax_table); i++) {		if (strstr(c->x86_model_id, tjmax_table[i].id))			return tjmax_table[i].tjmax;	} 	for (i = 0; i < ARRAY_SIZE(tjmax_model_table); i++) {		const struct tjmax_model *tm = &tjmax_model_table[i]; 		if (c->x86_vfm == tm->vfm &&		    (tm->stepping_mask == ANY ||		     tm->stepping_mask == c->x86_stepping))			return tm->tjmax;	} 	/* Early chips have no MSR for TjMax */ 	if (c->x86_vfm == INTEL_CORE2_MEROM && c->x86_stepping < 4)		usemsr_ee = 0; 	if (c->x86_vfm > INTEL_CORE_YONAH && usemsr_ee) {		u8 platform_id; 		/*		 * Now we can detect the mobile CPU using Intel provided table		 * http://softwarecommunity.intel.com/Wiki/Mobility/720.htm		 * For Core2 cores, check MSR 0x17, bit 28 1 = Mobile CPU		 */		err = rdmsrq_safe_on_cpu(id, 0x17, &val);		if (err) {			dev_warn(dev,				 "Unable to access MSR 0x17, assuming desktop CPU\n");			usemsr_ee = 0;		} else if (c->x86_vfm < INTEL_CORE2_PENRYN &&			   !(val & 0x10000000)) {			/*			 * Trust bit 28 up to Penryn, I could not find any			 * documentation on that; if you happen to know			 * someone at Intel please ask			 */			usemsr_ee = 0;		} else {			/* Platform ID bits 52:50 */			platform_id = (val >> 50) & 0x7; 			/*			 * Mobile Penryn CPU seems to be platform ID 7 or 5			 * (guesswork)			 */			if (c->x86_vfm == INTEL_CORE2_PENRYN &&			    (platform_id == 5 || platform_id == 7)) {				/*				 * If MSR EE bit is set, set it to 90 degrees C,				 * otherwise 105 degrees C				 */				tjmax_ee = 90000;				tjmax = 105000;			}		}	} 	if (usemsr_ee) {		err = rdmsrq_safe_on_cpu(id, 0xee, &val);		if (err) {			dev_warn(dev,				 "Unable to access MSR 0xEE, for Tjmax, left at default\n");		} else if (val & 0x40000000) {			tjmax = tjmax_ee;		}	} else if (tjmax == 100000) {		/*		 * If we don't use msr EE it means we are desktop CPU		 * (with exeception of Atom)		 */		dev_warn(dev, "Using relative temperature scale!\n");	} 	return tjmax;} static int get_tjmax(struct temp_data *tdata, struct device *dev){	struct cpuinfo_x86 *c = &cpu_data(tdata->cpu);	int err;	u64 msrval;	u32 val; 	/* use static tjmax once it is set */	if (tdata->tjmax)		return tdata->tjmax; 	/*	 * A new feature of current Intel(R) processors, the	 * IA32_TEMPERATURE_TARGET contains the TjMax value	 */	err = rdmsrq_safe_on_cpu(tdata->cpu, MSR_IA32_TEMPERATURE_TARGET, &msrval);	if (err) {		dev_warn_once(dev, "Unable to read TjMax from CPU %u\n", tdata->cpu);	} else {		val = (msrval >> 16) & 0xff;		if (val)			return val * 1000;	} 	if (force_tjmax) {		dev_notice(dev, "TjMax forced to %d degrees C by user\n",			   force_tjmax);		tdata->tjmax = force_tjmax * 1000;	} else {		/*		 * An assumption is made for early CPUs and unreadable MSR.		 * NOTE: the calculated value may not be correct.		 */		tdata->tjmax = adjust_tjmax(c, tdata->cpu, dev);	}	return tdata->tjmax;} static int get_ttarget(struct temp_data *tdata, struct device *dev){	u64 val;	int tjmax, ttarget_offset, ret; 	/*	 * ttarget is valid only if tjmax can be retrieved from	 * MSR_IA32_TEMPERATURE_TARGET	 */	if (tdata->tjmax)		return -ENODEV; 	ret = rdmsrq_safe_on_cpu(tdata->cpu, MSR_IA32_TEMPERATURE_TARGET, &val);	if (ret)		return ret; 	tjmax = (val >> 16) & 0xff; 	/* Read the still undocumented bits 8:15 of IA32_TEMPERATURE_TARGET. */	ttarget_offset = (val >> 8) & 0xff; 	return (tjmax - ttarget_offset) * 1000;} /* Keep track of how many zone pointers we allocated in init() */static int max_zones __read_mostly;/* Array of zone pointers. Serialized by cpu hotplug lock */static struct platform_device **zone_devices; static ssize_t show_label(struct device *dev,			  struct device_attribute *devattr, char *buf){	struct platform_data *pdata = dev_get_drvdata(dev);	struct temp_data *tdata = container_of(devattr, struct temp_data, sd_attrs[ATTR_LABEL]); 	if (is_pkg_temp_data(tdata))		return sprintf(buf, "Package id %u\n", pdata->pkg_id); 	return sprintf(buf, "Core %u\n", tdata->cpu_core_id);} static ssize_t show_crit_alarm(struct device *dev,			       struct device_attribute *devattr, char *buf){	struct msr val;	struct temp_data *tdata = container_of(devattr, struct temp_data,						sd_attrs[ATTR_CRIT_ALARM]); 	mutex_lock(&tdata->update_lock);	rdmsrq_on_cpu(tdata->cpu, tdata->status_reg, &val.q);	mutex_unlock(&tdata->update_lock); 	return sprintf(buf, "%d\n", (val.l >> 5) & 1);} static ssize_t show_tjmax(struct device *dev,			  struct device_attribute *devattr, char *buf){	struct temp_data *tdata = container_of(devattr, struct temp_data, sd_attrs[ATTR_TJMAX]);	int tjmax; 	mutex_lock(&tdata->update_lock);	tjmax = get_tjmax(tdata, dev);	mutex_unlock(&tdata->update_lock); 	return sprintf(buf, "%d\n", tjmax);} static ssize_t show_ttarget(struct device *dev,			    struct device_attribute *devattr, char *buf){	struct temp_data *tdata = container_of(devattr, struct temp_data, sd_attrs[ATTR_TTARGET]);	int ttarget; 	mutex_lock(&tdata->update_lock);	ttarget = get_ttarget(tdata, dev);	mutex_unlock(&tdata->update_lock); 	if (ttarget < 0)		return ttarget;	return sprintf(buf, "%d\n", ttarget);} static ssize_t show_temp(struct device *dev,			 struct device_attribute *devattr, char *buf){	struct msr val;	struct temp_data *tdata = container_of(devattr, struct temp_data, sd_attrs[ATTR_TEMP]);	int tjmax; 	mutex_lock(&tdata->update_lock); 	tjmax = get_tjmax(tdata, dev);	/* Check whether the time interval has elapsed */	if (time_after(jiffies, tdata->last_updated + HZ)) {		rdmsrq_on_cpu(tdata->cpu, tdata->status_reg, &val.q);		/*		 * Ignore the valid bit. In all observed cases the register		 * value is either low or zero if the valid bit is 0.		 * Return it instead of reporting an error which doesn't		 * really help at all.		 */		tdata->temp = tjmax - ((val.l >> 16) & 0xff) * 1000;		tdata->last_updated = jiffies;	} 	mutex_unlock(&tdata->update_lock);	return sprintf(buf, "%d\n", tdata->temp);} static int create_core_attrs(struct temp_data *tdata, struct device *dev){	static ssize_t (*const rd_ptr[TOTAL_ATTRS]) (struct device *dev,			struct device_attribute *devattr, char *buf) = {			show_label, show_crit_alarm, show_temp, show_tjmax,			show_ttarget };	static const char *const suffixes[TOTAL_ATTRS] = {		"label", "crit_alarm", "input", "crit", "max"	};	int i; 	for (i = 0; i < tdata->attr_size; i++) {		/*		 * We map the attr number to core id of the CPU		 * The attr number is always core id + 2		 * The Pkgtemp will always show up as temp1_*, if available		 */		int attr_no = is_pkg_temp_data(tdata) ? 1 : tdata->cpu_core_id + 2; 		snprintf(tdata->attr_name[i], CORETEMP_NAME_LENGTH,			 "temp%d_%s", attr_no, suffixes[i]);		sysfs_attr_init(&tdata->sd_attrs[i].attr);		tdata->sd_attrs[i].attr.name = tdata->attr_name[i];		tdata->sd_attrs[i].attr.mode = 0444;		tdata->sd_attrs[i].show = rd_ptr[i];		tdata->attrs[i] = &tdata->sd_attrs[i].attr;	}	tdata->attr_group.attrs = tdata->attrs;	return sysfs_create_group(&dev->kobj, &tdata->attr_group);} static int chk_ucode_version(unsigned int cpu){	struct cpuinfo_x86 *c = &cpu_data(cpu); 	/*	 * Check if we have problem with errata AE18 of Core processors:	 * Readings might stop update when processor visited too deep sleep,	 * fixed for stepping D0 (6EC).	 */	if (c->x86_vfm == INTEL_CORE_YONAH && c->x86_stepping < 0xc && c->microcode < 0x39) {		pr_err("Errata AE18 not fixed, update BIOS or microcode of the CPU!\n");		return -ENODEV;	}	return 0;} static struct platform_device *coretemp_get_pdev(unsigned int cpu){	int id = topology_logical_die_id(cpu); 	if (id >= 0 && id < max_zones)		return zone_devices[id];	return NULL;} static struct temp_data *init_temp_data(struct platform_data *pdata, unsigned int cpu, int pkg_flag){	struct temp_data *tdata; 	if (!pdata->core_data) {		pdata->nr_cores = topology_num_cores_per_package();		pdata->core_data = kzalloc_objs(struct temp_data *,						pdata->nr_cores);		if (!pdata->core_data)			return NULL;	} 	tdata = kzalloc_obj(struct temp_data);	if (!tdata)		return NULL; 	if (pkg_flag) {		pdata->pkg_data = tdata;		/* Use tdata->index as indicator of package temp data */		tdata->index = -1;	} else {

Lines 1–500 of 838.