linux电源管理架构

电源

linux电源管理架构

2019-07-13 22:07发布生成海报

站内文章 / 电源技术

11545 0

linux电源管理非常复杂，牵扯到系统机的待机、频率电压变换、系统空闲时间的处理以及每个设备驱动对系统待机的支持和每个设备的运行时runtime电源管理，可以说它和系统中的每个设备驱动都息息相关。 linux内核电源管理的整体架构，
这里写图片描述

1）CPU在运行时根据系统负载进行动态电压和频率变换的CPUFreq
2) CPU在系统空闲时根据空闲的情况进行低功耗模式的CPUIdle
3）多核系统内下CPU的热插拔支持
4）系统和设备针对延迟的特别需求而提出申请的PM QoS，它会作用于CPUIdle的具体策略
5）设备驱动针对系统挂起到RAM/硬盘的一系列入口函数
6）SoC进入挂起状态，SDRAM自刷新的入口。
7）设备运行时动态电源管理，根据使用情况动态开关设备。
8）底层的时钟，稳压器，频率/电压表（OPP模块完成）支撑，个驱动子系统都可能用到。

CPUFreq驱动

CPUFreq子系统位于drivers/cpufreq目录下，复杂进行运行过程中CPU频率和电压的动态调整，即DVFS（Dynamic Voltage Frequency Scaling，动态电压频率调整）。运行时进行CPU电压和频率调整的原因是：CMOS电路的功耗与电压的平方成正比、与频率成正比，因此降低电压和频率可降低功耗。
CPUFreq的核心层位于drivers/cpufreq/cpufreq.c下，它为各个SoC的CPUFreq驱动的实现提供了一套统一的接口，并实现了一套notifier机制，可以在CPUFrq的策略和频率改变的时候向其他模块发出通知。另外，在CPU运行频率发送变化的时候，内核的loops_per_jiffy常数也会发生相应的变化。 SoC的CPUFreq驱动实现
每个SoC的具体CPUFreq驱动实例只需要实现电压、频率表，以及从硬件层面完整这些变化。 CPUFreq核心层提供了如下API以供SoC注册自身的CPUFreq驱动：

int cpufreq_register_driver(struct cpufreq_driver *driver_data)

其参数为一个cpufreq_driver结构体指针，实际上，cpufreq_driver封装了一个具体的SoC的CPUFreq驱动的主体，该结构体代码如下：



struct cpufreq_driver {
    char        name[CPUFREQ_NAME_LEN];
    u8      flags;
    void        *driver_data;

    /* needed by all drivers */
    int     (*init)(struct cpufreq_policy *policy);
    int     (*verify)(struct cpufreq_policy *policy);

    /* define one out of two */
    int     (*setpolicy)(struct cpufreq_policy *policy);

    /*
     * On failure, should always restore frequency to policy->restore_freq
     * (i.e. old freq).
     */
    int     (*target)(struct cpufreq_policy *policy,
                  unsigned int target_freq,
                  unsigned int relation);   /* Deprecated */
    int     (*target_index)(struct cpufreq_policy *policy,
                    unsigned int index);
    unsigned int    (*fast_switch)(struct cpufreq_policy *policy,
                       unsigned int target_freq);
    /*
     * Only for drivers with target_index() and CPUFREQ_ASYNC_NOTIFICATION
     * unset.
     *
     * get_intermediate should return a stable intermediate frequency
     * platform wants to switch to and target_intermediate() should set CPU
     * to to that frequency, before jumping to the frequency corresponding
     * to 'index'. Core will take care of sending notifications and driver
     * doesn't have to handle them in target_intermediate() or
     * target_index().
     *
     * Drivers can return '0' from get_intermediate() in case they don't
     * wish to switch to intermediate frequency for some target frequency.
     * In that case core will directly call ->target_index().
     */
    unsigned int    (*get_intermediate)(struct cpufreq_policy *policy,
                        unsigned int index);
    int     (*target_intermediate)(struct cpufreq_policy *policy,
                           unsigned int index);

    /* should be defined, if possible */
    unsigned int    (*get)(unsigned int cpu);

    /* optional */
    int     (*bios_limit)(int cpu, unsigned int *limit);

    int     (*exit)(struct cpufreq_policy *policy);
    void        (*stop_cpu)(struct cpufreq_policy *policy);
    int     (*suspend)(struct cpufreq_policy *policy);
    int     (*resume)(struct cpufreq_policy *policy);

    /* Will be called after the driver is fully initialized */
    void        (*ready)(struct cpufreq_policy *policy);

    struct freq_attr **attr;

    /* platform specific boost support code */
    bool        boost_enabled;
    int     (*set_boost)(int state);
};

其中name成员是CPUFreq驱动的名字，如drivers/cpufreq/s5pv210-cpufreq.c设置那么为s5pv210，drivers/cpufreq/omap-cpufreq.c设置name为omap；
init（）成员是一个per-CPU初始化函数指针，每当一个新的CPU被注册进系统的时候，该函数就被调用，该函数接受一个cpufreq_policy的指针参数。
verify()成员函数用于对用户的CPUFreq策略设置进行有效性验证和数据修正，每当用户设定一个新策略时，该函数根据老的策略和新的策略，检验新策略设置的有效性并对无效设置进行表要的修正，该成员函数的具体实现，常用到如下辅助函数

static inline void cpufreq_verify_within_limits(struct cpufreq_policy *policy,
    unsigned int min, unsigned int max)

setpolicy()成员函数接受一个policy参数，实现了这个成员函数的CPU一遍具备在一个范围里的自耦东调整频率的能力。
target()成员函数用于将频率调整到一个指定的值。 cpufreq_register_driver 函数，具体操作内容如下：


/*********************************************************************
 *               REGISTER / UNREGISTER CPUFREQ DRIVER                *
 *********************************************************************/

/**
 * cpufreq_register_driver - register a CPU Frequency driver
 * @driver_data: A struct cpufreq_driver containing the values#
 * submitted by the CPU Frequency driver.
 *
 * Registers a CPU Frequency driver to this core code. This code
 * returns zero on success, -EEXIST when another driver got here first
 * (and isn't unregistered in the meantime).
 *
 */
int cpufreq_register_driver(struct cpufreq_driver *driver_data)
{
    unsigned long flags;
    int ret;

    if (cpufreq_disabled())
        return -ENODEV;

    if (!driver_data || !driver_data->verify || !driver_data->init ||
        !(driver_data->setpolicy || driver_data->target_index ||
            driver_data->target) ||
         (driver_data->setpolicy && (driver_data->target_index ||
            driver_data->target)) ||
         (!!driver_data->get_intermediate != !!driver_data->target_intermediate))
        return -EINVAL;

    pr_debug("trying to register driver %s
", driver_data->name);

    /* Protect against concurrent CPU online/offline. */
    get_online_cpus();

    write_lock_irqsave(&cpufreq_driver_lock, flags);
    if (cpufreq_driver) {
        write_unlock_irqrestore(&cpufreq_driver_lock, flags);
        ret = -EEXIST;
        goto out;
    }
    cpufreq_driver = driver_data;
    write_unlock_irqrestore(&cpufreq_driver_lock, flags);

    if (driver_data->setpolicy)
        driver_data->flags |= CPUFREQ_CONST_LOOPS;

    if (cpufreq_boost_supported()) {
        ret = create_boost_sysfs_file();
        if (ret)
            goto err_null_driver;
    }

    ret = subsys_interface_register(&cpufreq_interface);
    if (ret)
        goto err_boost_unreg;

    if (!(cpufreq_driver->flags & CPUFREQ_STICKY) &&
        list_empty(&cpufreq_policy_list)) {
        /* if all ->init() calls failed, unregister */
        pr_debug("%s: No CPU initialized for driver %s
", __func__,
             driver_data->name);
        goto err_if_unreg;
    }

    register_hotcpu_notifier(&cpufreq_cpu_notifier);
    pr_debug("driver %s up and running
", driver_data->name);
    goto out;

err_if_unreg:
    subsys_interface_unregister(&cpufreq_interface);
err_boost_unreg:
    remove_boost_sysfs_file();
err_null_driver:
    write_lock_irqsave(&cpufreq_driver_lock, flags);
    cpufreq_driver = NULL;
    write_unlock_irqrestore(&cpufreq_driver_lock, flags);
out:
    put_online_cpus();
    return ret;
}

一个SoC的CPUFreq驱动实例drivers/cpufreq/s3c64xx-cpufreq.c的核心结构代码如下：

/*
 * Copyright 2009 Wolfson Microelectronics plc
 *
 * S3C64xx CPUfreq Support
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 */

#define pr_fmt(fmt) "cpufreq: " fmt

#include 
#include 
#include 
#include 
#include 
#include 
#include 
#include 

static struct regulator *vddarm;
static unsigned long regulator_latency;

#ifdef CONFIG_CPU_S3C6410
struct s3c64xx_dvfs {
    unsigned int vddarm_min;
    unsigned int vddarm_max;
};

static struct s3c64xx_dvfs s3c64xx_dvfs_table[] = {
    [0] = { 1000000, 1150000 },
    [1] = { 1050000, 1150000 },
    [2] = { 1100000, 1150000 },
    [3] = { 1200000, 1350000 },
    [4] = { 1300000, 1350000 },
};

static struct cpufreq_frequency_table s3c64xx_freq_table[] = {
    { 0, 0,  66000 },
    { 0, 0, 100000 },
    { 0, 0, 133000 },
    { 0, 1, 200000 },
    { 0, 1, 222000 },
    { 0, 1, 266000 },
    { 0, 2, 333000 },
    { 0, 2, 400000 },
    { 0, 2, 532000 },
    { 0, 2, 533000 },
    { 0, 3, 667000 },
    { 0, 4, 800000 },
    { 0, 0, CPUFREQ_TABLE_END },
};
#endif

static int s3c64xx_cpufreq_set_target(struct cpufreq_policy *policy,
                      unsigned int index)
{
    struct s3c64xx_dvfs *dvfs;
    unsigned int old_freq, new_freq;
    int ret;

    old_freq = clk_get_rate(policy->clk) / 1000;
    new_freq = s3c64xx_freq_table[index].frequency;
    dvfs = &s3c64xx_dvfs_table[s3c64xx_freq_table[index].driver_data];

#ifdef CONFIG_REGULATOR
    if (vddarm && new_freq > old_freq) {
        ret = regulator_set_voltage(vddarm,
                        dvfs->vddarm_min,
                        dvfs->vddarm_max);
        if (ret != 0) {
            pr_err("Failed to set VDDARM for %dkHz: %d
",
                   new_freq, ret);
            return ret;
        }
    }
#endif

    ret = clk_set_rate(policy->clk, new_freq * 1000);
    if (ret < 0) {
        pr_err("Failed to set rate %dkHz: %d
",
               new_freq, ret);
        return ret;
    }

#ifdef CONFIG_REGULATOR
    if (vddarm && new_freq < old_freq) {
        ret = regulator_set_voltage(vddarm,
                        dvfs->vddarm_min,
                        dvfs->vddarm_max);
        if (ret != 0) {
            pr_err("Failed to set VDDARM for %dkHz: %d
",
                   new_freq, ret);
            if (clk_set_rate(policy->clk, old_freq * 1000) < 0)
                pr_err("Failed to restore original clock rate
");

            return ret;
        }
    }
#endif

    pr_debug("Set actual frequency %lukHz
",
         clk_get_rate(policy->clk) / 1000);

    return 0;
}

#ifdef CONFIG_REGULATOR
static void __init s3c64xx_cpufreq_config_regulator(void)
{
    int count, v, i, found;
    struct cpufreq_frequency_table *freq;
    struct s3c64xx_dvfs *dvfs;

    count = regulator_count_voltages(vddarm);
    if (count < 0) {
        pr_err("Unable to check supported voltages
");
    }

    if (!count)
        goto out;

    cpufreq_for_each_valid_entry(freq, s3c64xx_freq_table) {
        dvfs = &s3c64xx_dvfs_table[freq->driver_data];
        found = 0;

        for (i = 0; i < count; i++) {
            v = regulator_list_voltage(vddarm, i);
            if (v >= dvfs->vddarm_min && v <= dvfs->vddarm_max)
                found = 1;
        }

        if (!found) {
            pr_debug("%dkHz unsupported by regulator
",
                 freq->frequency);
            freq->frequency = CPUFREQ_ENTRY_INVALID;
        }
    }

out:
    /* Guess based on having to do an I2C/SPI write; in future we
     * will be able to query the regulator performance here. */
    regulator_latency = 1 * 1000 * 1000;
}
#endif

static int s3c64xx_cpufreq_driver_init(struct cpufreq_policy *policy)
{
    int ret;
    struct cpufreq_frequency_table *freq;

    if (policy->cpu != 0)
        return -EINVAL;

    if (s3c64xx_freq_table == NULL) {
        pr_err("No frequency information for this CPU
");
        return -ENODEV;
    }

    policy->clk = clk_get(NULL, "armclk");
    if (IS_ERR(policy->clk)) {
        pr_err("Unable to obtain ARMCLK: %ld
",
               PTR_ERR(policy->clk));
        return PTR_ERR(policy->clk);
    }

#ifdef CONFIG_REGULATOR
    vddarm = regulator_get(NULL, "vddarm");
    if (IS_ERR(vddarm)) {
        ret = PTR_ERR(vddarm);
        pr_err("Failed to obtain VDDARM: %d
", ret);
        pr_err("Only frequency scaling available
");
        vddarm = NULL;
    } else {
        s3c64xx_cpufreq_config_regulator();
    }
#endif

    cpufreq_for_each_entry(freq, s3c64xx_freq_table) {
        unsigned long r;

        /* Check for frequencies we can generate */
        r = clk_round_rate(policy->clk, freq->frequency * 1000);
        r /= 1000;
        if (r != freq->frequency) {
            pr_debug("%dkHz unsupported by clock
",
                 freq->frequency);
            freq->frequency = CPUFREQ_ENTRY_INVALID;
        }

        /* If we have no regulator then assume startup
         * frequency is the maximum we can support. */
        if (!vddarm && freq->frequency > clk_get_rate(policy->clk) / 1000)
            freq->frequency = CPUFREQ_ENTRY_INVALID;
    }

    /* Datasheet says PLL stabalisation time (if we were to use
     * the PLLs, which we don't currently) is ~300us worst case,
     * but add some fudge.
     */
    ret = cpufreq_generic_init(policy, s3c64xx_freq_table,
            (500 * 1000) + regulator_latency);
    if (ret != 0) {
        pr_err("Failed to configure frequency table: %d
",
               ret);
        regulator_put(vddarm);
        clk_put(policy->clk);
    }

    return ret;
}

static struct cpufreq_driver s3c64xx_cpufreq_driver = {
    .flags      = CPUFREQ_NEED_INITIAL_FREQ_CHECK,
    .verify     = cpufreq_generic_frequency_table_verify,
    .target_index   = s3c64xx_cpufreq_set_target,
    .get        = cpufreq_generic_get,
    .init       = s3c64xx_cpufreq_driver_init,
    .name       = "s3c",
};

static int __init s3c64xx_cpufreq_init(void)
{
    return cpufreq_register_driver(&s3c64xx_cpufreq_driver);
}
module_init(s3c64xx_cpufreq_init);

CPUFreq的策略
系统的状态以及CPUFreq的策略共同决定了CPU频率跳变的目标，CPUFreq核心层将目标频率传递给底层具体SoC的CPUFreq驱动，该驱动修改硬件，完成频率的变化。如下图所示：
这里写图片描述

用户空间一般可通过/sys/devices/system/cpu/cpux/cpufreq节点来设置CPUFreq，例如，设置CPUFreq到700MHz，采用userspace策略，则运行如下命令：

# echo userspace > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor 
# echo 700000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed

CPUIdele驱动

目前的ARM SoC大多支持几个不同的Idle级别，CPUIdle驱动子系统存在的目的就是对这些Idle状态进行管理，并根据系统运行的情况进入不同的Idle级别。
与CPUFreq类似，CPUIdle的核心层提供了用户注册cpuidle_driver的注册函数：

int cpuidle_register_driver(struct cpuidle_driver *drv)

cpuidle_device注册函数：

int cpuidle_register_device(struct cpuidle_device *dev)

cpuidle_driver结构体是CPUIdle驱动的主体



/****************************
 * CPUIDLE DRIVER INTERFACE *
 ****************************/

struct cpuidle_driver {
    const char      *name;
    struct module       *owner;
    int                     refcnt;

        /* used by the cpuidle framework to setup the broadcast timer */
    unsigned int            bctimer:1;
    /* states array must be ordered in decreasing power consumption */
    struct cpuidle_state    states[CPUIDLE_STATE_MAX];
    int         state_count;
    int         safe_state_index;

    /* the driver handles the cpus in cpumask */
    struct cpumask      *cpumask;
};

其中关键结构体是成员cpuidle_state，states[CPUIDLE_STATE_MAX] 该结构体数组就用于存储各种不同的Idle级别的信息，定义如下：


struct cpuidle_state {
    char        name[CPUIDLE_NAME_LEN];
    char        desc[CPUIDLE_DESC_LEN];

    unsigned int    flags;
    unsigned int    exit_latency; /* in US */
    int     power_usage; /* in mW */
    unsigned int    target_residency; /* in US */
    bool        disabled; /* disabled on all CPUs */

    int (*enter)    (struct cpuidle_device *dev,
            struct cpuidle_driver *drv,
            int index);

    int (*enter_dead) (struct cpuidle_device *dev, int index);

    /*
     * CPUs execute ->enter_freeze with the local tick or entire timekeeping
     * suspended, so it must not re-enable interrupts at any point (even
     * temporarily) or attempt to change states of clock event devices.
     */
    void (*enter_freeze) (struct cpuidle_device *dev,
                  struct cpuidle_driver *drv,
                  int index);
};

一个具体的SoC驱动实例，drivers/cpuidle/cpuidle-ux500.c，它有两个Idle级别，即WFI和ApIdle，具体代码如下：

/*
 * Copyright (c) 2012 Linaro : Daniel Lezcano  (IBM)
 *
 * Based on the work of Rickard Andersson 
 * and Jonas Aaberg .
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License version 2 as
 * published by the Free Software Foundation.
 */

#include 
#include 
#include 
#include 
#include 
#include 
#include 
#include 

#include 

static atomic_t master = ATOMIC_INIT(0);
static DEFINE_SPINLOCK(master_lock);

static inline int ux500_enter_idle(struct cpuidle_device *dev,
                   struct cpuidle_driver *drv, int index)
{
    int this_cpu = smp_processor_id();
    bool recouple = false;

    if (atomic_inc_return(&master) == num_online_cpus()) {

        /* With this lock, we prevent the other cpu to exit and enter
         * this function again and become the master */
        if (!spin_trylock(&master_lock))
            goto wfi;

        /* decouple the gic from the A9 cores */
        if (prcmu_gic_decouple()) {
            spin_unlock(&master_lock);
            goto out;
        }

        /* If an error occur, we will have to recouple the gic
         * manually */
        recouple = true;

        /* At this state, as the gic is decoupled, if the other
         * cpu is in WFI, we have the guarantee it won't be wake
         * up, so we can safely go to retention */
        if (!prcmu_is_cpu_in_wfi(this_cpu ? 0 : 1))
            goto out;

        /* The prcmu will be in charge of watching the interrupts
         * and wake up the cpus */
        if (prcmu_copy_gic_settings())
            goto out;

        /* Check in the meantime an interrupt did
         * not occur on the gic ... */
        if (prcmu_gic_pending_irq())
            goto out;

        /* ... and the prcmu */
        if (prcmu_pending_irq())
            goto out;

        /* Go to the retention state, the prcmu will wait for the
         * cpu to go WFI and this is what happens after exiting this
         * 'master' critical section */
        if (prcmu_set_power_state(PRCMU_AP_IDLE, true, true))
            goto out;

        /* When we switch to retention, the prcmu is in charge
         * of recoupling the gic automatically */
        recouple = false;

        spin_unlock(&master_lock);
    }
wfi:
    cpu_do_idle();
out:
    atomic_dec(&master);

    if (recouple) {
        prcmu_gic_recouple();
        spin_unlock(&master_lock);
    }

    return index;
}

static struct cpuidle_driver ux500_idle_driver = {
    .name = "ux500_idle",
    .owner = THIS_MODULE,
    .states = {
        ARM_CPUIDLE_WFI_STATE,
        {
            .enter        = ux500_enter_idle,
            .exit_latency     = 70,
            .target_residency = 260,
            .flags        = CPUIDLE_FLAG_TIMER_STOP,
            .name         = "ApIdle",
            .desc         = "ARM Retention",
        },
    },
    .safe_state_index = 0,
    .state_count = 2,
};

static int dbx500_cpuidle_probe(struct platform_device *pdev)
{
    /* Configure wake up reasons */
    prcmu_enable_wakeups(PRCMU_WAKEUP(ARM) | PRCMU_WAKEUP(RTC) |
                 PRCMU_WAKEUP(ABB));

    return cpuidle_register(&ux500_idle_driver, NULL);
}

static struct platform_driver dbx500_cpuidle_plat_driver = {
    .driver = {
        .name = "cpuidle-dbx500",
    },
    .probe = dbx500_cpuidle_probe,
};
builtin_platform_driver(dbx500_cpuidle_plat_driver);

综上分析，可以给出 Linux CPUIdle子系通总体架构，如下图：
这里写图片描述