⭐ 欢迎来到虫虫下载站! | 📦 资源下载 📁 资源专辑 ℹ️ 关于我们
⭐ 虫虫下载站

📄 processor_thermal.c

📁 linux 内核源代码
💻 C
字号:
/* * processor_thermal.c - Passive cooling submodule of the ACPI processor driver * *  Copyright (C) 2001, 2002 Andy Grover <andrew.grover@intel.com> *  Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com> *  Copyright (C) 2004       Dominik Brodowski <linux@brodo.de> *  Copyright (C) 2004  Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com> *  			- Added processor hotplug support * * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ * *  This program is free software; you can redistribute it and/or modify *  it under the terms of the GNU General Public License as published by *  the Free Software Foundation; either version 2 of the License, or (at *  your option) any later version. * *  This program is distributed in the hope that it will be useful, but *  WITHOUT ANY WARRANTY; without even the implied warranty of *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU *  General Public License for more details. * *  You should have received a copy of the GNU General Public License along *  with this program; if not, write to the Free Software Foundation, Inc., *  59 Temple Place, Suite 330, Boston, MA 02111-1307 USA. * * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ */#include <linux/kernel.h>#include <linux/module.h>#include <linux/init.h>#include <linux/cpufreq.h>#include <linux/proc_fs.h>#include <linux/seq_file.h>#include <asm/uaccess.h>#include <acpi/acpi_bus.h>#include <acpi/processor.h>#include <acpi/acpi_drivers.h>#define ACPI_PROCESSOR_COMPONENT        0x01000000#define ACPI_PROCESSOR_CLASS            "processor"#define _COMPONENT              ACPI_PROCESSOR_COMPONENTACPI_MODULE_NAME("processor_thermal");/* --------------------------------------------------------------------------                                 Limit Interface   -------------------------------------------------------------------------- */static int acpi_processor_apply_limit(struct acpi_processor *pr){	int result = 0;	u16 px = 0;	u16 tx = 0;	if (!pr)		return -EINVAL;	if (!pr->flags.limit)		return -ENODEV;	if (pr->flags.throttling) {		if (pr->limit.user.tx > tx)			tx = pr->limit.user.tx;		if (pr->limit.thermal.tx > tx)			tx = pr->limit.thermal.tx;		result = acpi_processor_set_throttling(pr, tx);		if (result)			goto end;	}	pr->limit.state.px = px;	pr->limit.state.tx = tx;	ACPI_DEBUG_PRINT((ACPI_DB_INFO,			  "Processor [%d] limit set to (P%d:T%d)\n", pr->id,			  pr->limit.state.px, pr->limit.state.tx));      end:	if (result)		printk(KERN_ERR PREFIX "Unable to set limit\n");	return result;}#ifdef CONFIG_CPU_FREQ/* If a passive cooling situation is detected, primarily CPUfreq is used, as it * offers (in most cases) voltage scaling in addition to frequency scaling, and * thus a cubic (instead of linear) reduction of energy. Also, we allow for * _any_ cpufreq driver and not only the acpi-cpufreq driver. */static unsigned int cpufreq_thermal_reduction_pctg[NR_CPUS];static unsigned int acpi_thermal_cpufreq_is_init = 0;static int cpu_has_cpufreq(unsigned int cpu){	struct cpufreq_policy policy;	if (!acpi_thermal_cpufreq_is_init || cpufreq_get_policy(&policy, cpu))		return 0;	return 1;}static int acpi_thermal_cpufreq_increase(unsigned int cpu){	if (!cpu_has_cpufreq(cpu))		return -ENODEV;	if (cpufreq_thermal_reduction_pctg[cpu] < 60) {		cpufreq_thermal_reduction_pctg[cpu] += 20;		cpufreq_update_policy(cpu);		return 0;	}	return -ERANGE;}static int acpi_thermal_cpufreq_decrease(unsigned int cpu){	if (!cpu_has_cpufreq(cpu))		return -ENODEV;	if (cpufreq_thermal_reduction_pctg[cpu] > 20)		cpufreq_thermal_reduction_pctg[cpu] -= 20;	else		cpufreq_thermal_reduction_pctg[cpu] = 0;	cpufreq_update_policy(cpu);	/* We reached max freq again and can leave passive mode */	return !cpufreq_thermal_reduction_pctg[cpu];}static int acpi_thermal_cpufreq_notifier(struct notifier_block *nb,					 unsigned long event, void *data){	struct cpufreq_policy *policy = data;	unsigned long max_freq = 0;	if (event != CPUFREQ_ADJUST)		goto out;	max_freq =	    (policy->cpuinfo.max_freq *	     (100 - cpufreq_thermal_reduction_pctg[policy->cpu])) / 100;	cpufreq_verify_within_limits(policy, 0, max_freq);      out:	return 0;}static struct notifier_block acpi_thermal_cpufreq_notifier_block = {	.notifier_call = acpi_thermal_cpufreq_notifier,};void acpi_thermal_cpufreq_init(void){	int i;	for (i = 0; i < NR_CPUS; i++)		cpufreq_thermal_reduction_pctg[i] = 0;	i = cpufreq_register_notifier(&acpi_thermal_cpufreq_notifier_block,				      CPUFREQ_POLICY_NOTIFIER);	if (!i)		acpi_thermal_cpufreq_is_init = 1;}void acpi_thermal_cpufreq_exit(void){	if (acpi_thermal_cpufreq_is_init)		cpufreq_unregister_notifier		    (&acpi_thermal_cpufreq_notifier_block,		     CPUFREQ_POLICY_NOTIFIER);	acpi_thermal_cpufreq_is_init = 0;}#else				/* ! CONFIG_CPU_FREQ */static int acpi_thermal_cpufreq_increase(unsigned int cpu){	return -ENODEV;}static int acpi_thermal_cpufreq_decrease(unsigned int cpu){	return -ENODEV;}#endifint acpi_processor_set_thermal_limit(acpi_handle handle, int type){	int result = 0;	struct acpi_processor *pr = NULL;	struct acpi_device *device = NULL;	int tx = 0, max_tx_px = 0;	if ((type < ACPI_PROCESSOR_LIMIT_NONE)	    || (type > ACPI_PROCESSOR_LIMIT_DECREMENT))		return -EINVAL;	result = acpi_bus_get_device(handle, &device);	if (result)		return result;	pr = acpi_driver_data(device);	if (!pr)		return -ENODEV;	/* Thermal limits are always relative to the current Px/Tx state. */	if (pr->flags.throttling)		pr->limit.thermal.tx = pr->throttling.state;	/*	 * Our default policy is to only use throttling at the lowest	 * performance state.	 */	tx = pr->limit.thermal.tx;	switch (type) {	case ACPI_PROCESSOR_LIMIT_NONE:		do {			result = acpi_thermal_cpufreq_decrease(pr->id);		} while (!result);		tx = 0;		break;	case ACPI_PROCESSOR_LIMIT_INCREMENT:		/* if going up: P-states first, T-states later */		result = acpi_thermal_cpufreq_increase(pr->id);		if (!result)			goto end;		else if (result == -ERANGE)			ACPI_DEBUG_PRINT((ACPI_DB_INFO,					  "At maximum performance state\n"));		if (pr->flags.throttling) {			if (tx == (pr->throttling.state_count - 1))				ACPI_DEBUG_PRINT((ACPI_DB_INFO,						  "At maximum throttling state\n"));			else				tx++;		}		break;	case ACPI_PROCESSOR_LIMIT_DECREMENT:		/* if going down: T-states first, P-states later */		if (pr->flags.throttling) {			if (tx == 0) {				max_tx_px = 1;				ACPI_DEBUG_PRINT((ACPI_DB_INFO,						  "At minimum throttling state\n"));			} else {				tx--;				goto end;			}		}		result = acpi_thermal_cpufreq_decrease(pr->id);		if (result) {			/*			 * We only could get -ERANGE, 1 or 0.			 * In the first two cases we reached max freq again.			 */			ACPI_DEBUG_PRINT((ACPI_DB_INFO,					  "At minimum performance state\n"));			max_tx_px = 1;		} else			max_tx_px = 0;		break;	}      end:	if (pr->flags.throttling) {		pr->limit.thermal.px = 0;		pr->limit.thermal.tx = tx;		result = acpi_processor_apply_limit(pr);		if (result)			printk(KERN_ERR PREFIX "Unable to set thermal limit\n");		ACPI_DEBUG_PRINT((ACPI_DB_INFO, "Thermal limit now (P%d:T%d)\n",				  pr->limit.thermal.px, pr->limit.thermal.tx));	} else		result = 0;	if (max_tx_px)		return 1;	else		return result;}int acpi_processor_get_limit_info(struct acpi_processor *pr){	if (!pr)		return -EINVAL;	if (pr->flags.throttling)		pr->flags.limit = 1;	return 0;}/* /proc interface */static int acpi_processor_limit_seq_show(struct seq_file *seq, void *offset){	struct acpi_processor *pr = (struct acpi_processor *)seq->private;	if (!pr)		goto end;	if (!pr->flags.limit) {		seq_puts(seq, "<not supported>\n");		goto end;	}	seq_printf(seq, "active limit:            P%d:T%d\n"		   "user limit:              P%d:T%d\n"		   "thermal limit:           P%d:T%d\n",		   pr->limit.state.px, pr->limit.state.tx,		   pr->limit.user.px, pr->limit.user.tx,		   pr->limit.thermal.px, pr->limit.thermal.tx);      end:	return 0;}static int acpi_processor_limit_open_fs(struct inode *inode, struct file *file){	return single_open(file, acpi_processor_limit_seq_show,			   PDE(inode)->data);}static ssize_t acpi_processor_write_limit(struct file * file,					  const char __user * buffer,					  size_t count, loff_t * data){	int result = 0;	struct seq_file *m = file->private_data;	struct acpi_processor *pr = m->private;	char limit_string[25] = { '\0' };	int px = 0;	int tx = 0;	if (!pr || (count > sizeof(limit_string) - 1)) {		return -EINVAL;	}	if (copy_from_user(limit_string, buffer, count)) {		return -EFAULT;	}	limit_string[count] = '\0';	if (sscanf(limit_string, "%d:%d", &px, &tx) != 2) {		printk(KERN_ERR PREFIX "Invalid data format\n");		return -EINVAL;	}	if (pr->flags.throttling) {		if ((tx < 0) || (tx > (pr->throttling.state_count - 1))) {			printk(KERN_ERR PREFIX "Invalid tx\n");			return -EINVAL;		}		pr->limit.user.tx = tx;	}	result = acpi_processor_apply_limit(pr);	return count;}struct file_operations acpi_processor_limit_fops = {	.open = acpi_processor_limit_open_fs,	.read = seq_read,	.write = acpi_processor_write_limit,	.llseek = seq_lseek,	.release = single_release,};

⌨️ 快捷键说明

复制代码 Ctrl + C
搜索代码 Ctrl + F
全屏模式 F11
切换主题 Ctrl + Shift + D
显示快捷键 ?
增大字号 Ctrl + =
减小字号 Ctrl + -