Merge branch 'pm-cpufreq'

* pm-cpufreq:
  cpufreq: OMAP: remove loops_per_jiffy recalculate for smp
  sections: fix section conflicts in drivers/cpufreq
  cpufreq: conservative: update frequency when limits are relaxed
  cpufreq / ondemand: update frequency when limits are relaxed
  cpufreq: Add a generic cpufreq-cpu0 driver
  PM / OPP: Initialize OPP table from device tree
  ARM: add cpufreq transiton notifier to adjust loops_per_jiffy for smp
  cpufreq: Remove support for hardware P-state chips from powernow-k8
  acpi-cpufreq: Add compatibility for legacy AMD cpb sysfs knob
  acpi-cpufreq: Add support for disabling dynamic overclocking
  ACPI: Add fixups for AMD P-state figures
  powernow-k8: delay info messages until initialization has succeeded
  cpufreq: Add warning message to powernow-k8
  acpi-cpufreq: Add quirk to disable _PSD usage on all AMD CPUs
  acpi-cpufreq: Add support for modern AMD CPUs
  cpufreq / powernow-k8: Fixup missing _PSS objects message
  PM / cpufreq: Initialise the cpu field during conservative governor start
This commit is contained in:
Rafael J. Wysocki 2012-09-17 20:26:02 +02:00
commit fa373abbbd
20 changed files with 948 additions and 456 deletions

View File

@ -176,3 +176,14 @@ Description: Disable L3 cache indices
All AMD processors with L3 caches provide this functionality.
For details, see BKDGs at
http://developer.amd.com/documentation/guides/Pages/default.aspx
What: /sys/devices/system/cpu/cpufreq/boost
Date: August 2012
Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org>
Description: Processor frequency boosting control
This switch controls the boost setting for the whole system.
Boosting allows the CPU and the firmware to run at a frequency
beyound it's nominal limit.
More details can be found in Documentation/cpu-freq/boost.txt

View File

@ -0,0 +1,93 @@
Processor boosting control
- information for users -
Quick guide for the impatient:
--------------------
/sys/devices/system/cpu/cpufreq/boost
controls the boost setting for the whole system. You can read and write
that file with either "0" (boosting disabled) or "1" (boosting allowed).
Reading or writing 1 does not mean that the system is boosting at this
very moment, but only that the CPU _may_ raise the frequency at it's
discretion.
--------------------
Introduction
-------------
Some CPUs support a functionality to raise the operating frequency of
some cores in a multi-core package if certain conditions apply, mostly
if the whole chip is not fully utilized and below it's intended thermal
budget. This is done without operating system control by a combination
of hardware and firmware.
On Intel CPUs this is called "Turbo Boost", AMD calls it "Turbo-Core",
in technical documentation "Core performance boost". In Linux we use
the term "boost" for convenience.
Rationale for disable switch
----------------------------
Though the idea is to just give better performance without any user
intervention, sometimes the need arises to disable this functionality.
Most systems offer a switch in the (BIOS) firmware to disable the
functionality at all, but a more fine-grained and dynamic control would
be desirable:
1. While running benchmarks, reproducible results are important. Since
the boosting functionality depends on the load of the whole package,
single thread performance can vary. By explicitly disabling the boost
functionality at least for the benchmark's run-time the system will run
at a fixed frequency and results are reproducible again.
2. To examine the impact of the boosting functionality it is helpful
to do tests with and without boosting.
3. Boosting means overclocking the processor, though under controlled
conditions. By raising the frequency and the voltage the processor
will consume more power than without the boosting, which may be
undesirable for instance for mobile users. Disabling boosting may
save power here, though this depends on the workload.
User controlled switch
----------------------
To allow the user to toggle the boosting functionality, the acpi-cpufreq
driver exports a sysfs knob to disable it. There is a file:
/sys/devices/system/cpu/cpufreq/boost
which can either read "0" (boosting disabled) or "1" (boosting enabled).
Reading the file is always supported, even if the processor does not
support boosting. In this case the file will be read-only and always
reads as "0". Explicitly changing the permissions and writing to that
file anyway will return EINVAL.
On supported CPUs one can write either a "0" or a "1" into this file.
This will either disable the boost functionality on all cores in the
whole system (0) or will allow the hardware to boost at will (1).
Writing a "1" does not explicitly boost the system, but just allows the
CPU (and the firmware) to boost at their discretion. Some implementations
take external factors like the chip's temperature into account, so
boosting once does not necessarily mean that it will occur every time
even using the exact same software setup.
AMD legacy cpb switch
---------------------
The AMD powernow-k8 driver used to support a very similar switch to
disable or enable the "Core Performance Boost" feature of some AMD CPUs.
This switch was instantiated in each CPU's cpufreq directory
(/sys/devices/system/cpu[0-9]*/cpufreq) and was called "cpb".
Though the per CPU existence hints at a more fine grained control, the
actual implementation only supported a system-global switch semantics,
which was simply reflected into each CPU's file. Writing a 0 or 1 into it
would pull the other CPUs to the same state.
For compatibility reasons this file and its behavior is still supported
on AMD CPUs, though it is now protected by a config switch
(X86_ACPI_CPUFREQ_CPB). On Intel CPUs this file will never be created,
even with the config option set.
This functionality is considered legacy and will be removed in some future
kernel version.
More fine grained boosting control
----------------------------------
Technically it is possible to switch the boosting functionality at least
on a per package basis, for some CPUs even per core. Currently the driver
does not support it, but this may be implemented in the future.

View File

@ -0,0 +1,55 @@
Generic CPU0 cpufreq driver
It is a generic cpufreq driver for CPU0 frequency management. It
supports both uniprocessor (UP) and symmetric multiprocessor (SMP)
systems which share clock and voltage across all CPUs.
Both required and optional properties listed below must be defined
under node /cpus/cpu@0.
Required properties:
- operating-points: Refer to Documentation/devicetree/bindings/power/opp.txt
for details
Optional properties:
- clock-latency: Specify the possible maximum transition latency for clock,
in unit of nanoseconds.
- voltage-tolerance: Specify the CPU voltage tolerance in percentage.
Examples:
cpus {
#address-cells = <1>;
#size-cells = <0>;
cpu@0 {
compatible = "arm,cortex-a9";
reg = <0>;
next-level-cache = <&L2>;
operating-points = <
/* kHz uV */
792000 1100000
396000 950000
198000 850000
>;
transition-latency = <61036>; /* two CLK32 periods */
};
cpu@1 {
compatible = "arm,cortex-a9";
reg = <1>;
next-level-cache = <&L2>;
};
cpu@2 {
compatible = "arm,cortex-a9";
reg = <2>;
next-level-cache = <&L2>;
};
cpu@3 {
compatible = "arm,cortex-a9";
reg = <3>;
next-level-cache = <&L2>;
};
};

View File

@ -0,0 +1,25 @@
* Generic OPP Interface
SoCs have a standard set of tuples consisting of frequency and
voltage pairs that the device will support per voltage domain. These
are called Operating Performance Points or OPPs.
Properties:
- operating-points: An array of 2-tuples items, and each item consists
of frequency and voltage like <freq-kHz vol-uV>.
freq: clock frequency in kHz
vol: voltage in microvolt
Examples:
cpu@0 {
compatible = "arm,cortex-a9";
reg = <0>;
next-level-cache = <&L2>;
operating-points = <
/* kHz uV */
792000 1100000
396000 950000
198000 850000
>;
};

View File

@ -25,6 +25,7 @@
#include <linux/percpu.h>
#include <linux/clockchips.h>
#include <linux/completion.h>
#include <linux/cpufreq.h>
#include <linux/atomic.h>
#include <asm/cacheflush.h>
@ -584,3 +585,56 @@ int setup_profiling_timer(unsigned int multiplier)
{
return -EINVAL;
}
#ifdef CONFIG_CPU_FREQ
static DEFINE_PER_CPU(unsigned long, l_p_j_ref);
static DEFINE_PER_CPU(unsigned long, l_p_j_ref_freq);
static unsigned long global_l_p_j_ref;
static unsigned long global_l_p_j_ref_freq;
static int cpufreq_callback(struct notifier_block *nb,
unsigned long val, void *data)
{
struct cpufreq_freqs *freq = data;
int cpu = freq->cpu;
if (freq->flags & CPUFREQ_CONST_LOOPS)
return NOTIFY_OK;
if (!per_cpu(l_p_j_ref, cpu)) {
per_cpu(l_p_j_ref, cpu) =
per_cpu(cpu_data, cpu).loops_per_jiffy;
per_cpu(l_p_j_ref_freq, cpu) = freq->old;
if (!global_l_p_j_ref) {
global_l_p_j_ref = loops_per_jiffy;
global_l_p_j_ref_freq = freq->old;
}
}
if ((val == CPUFREQ_PRECHANGE && freq->old < freq->new) ||
(val == CPUFREQ_POSTCHANGE && freq->old > freq->new) ||
(val == CPUFREQ_RESUMECHANGE || val == CPUFREQ_SUSPENDCHANGE)) {
loops_per_jiffy = cpufreq_scale(global_l_p_j_ref,
global_l_p_j_ref_freq,
freq->new);
per_cpu(cpu_data, cpu).loops_per_jiffy =
cpufreq_scale(per_cpu(l_p_j_ref, cpu),
per_cpu(l_p_j_ref_freq, cpu),
freq->new);
}
return NOTIFY_OK;
}
static struct notifier_block cpufreq_notifier = {
.notifier_call = cpufreq_callback,
};
static int __init register_cpufreq_notifier(void)
{
return cpufreq_register_notifier(&cpufreq_notifier,
CPUFREQ_TRANSITION_NOTIFIER);
}
core_initcall(register_cpufreq_notifier);
#endif

View File

@ -248,6 +248,9 @@
#define MSR_IA32_PERF_STATUS 0x00000198
#define MSR_IA32_PERF_CTL 0x00000199
#define MSR_AMD_PSTATE_DEF_BASE 0xc0010064
#define MSR_AMD_PERF_STATUS 0xc0010063
#define MSR_AMD_PERF_CTL 0xc0010062
#define MSR_IA32_MPERF 0x000000e7
#define MSR_IA32_APERF 0x000000e8

View File

@ -324,6 +324,34 @@ static int acpi_processor_get_performance_control(struct acpi_processor *pr)
return result;
}
#ifdef CONFIG_X86
/*
* Some AMDs have 50MHz frequency multiples, but only provide 100MHz rounding
* in their ACPI data. Calculate the real values and fix up the _PSS data.
*/
static void amd_fixup_frequency(struct acpi_processor_px *px, int i)
{
u32 hi, lo, fid, did;
int index = px->control & 0x00000007;
if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD)
return;
if ((boot_cpu_data.x86 == 0x10 && boot_cpu_data.x86_model < 10)
|| boot_cpu_data.x86 == 0x11) {
rdmsr(MSR_AMD_PSTATE_DEF_BASE + index, lo, hi);
fid = lo & 0x3f;
did = (lo >> 6) & 7;
if (boot_cpu_data.x86 == 0x10)
px->core_frequency = (100 * (fid + 0x10)) >> did;
else
px->core_frequency = (100 * (fid + 8)) >> did;
}
}
#else
static void amd_fixup_frequency(struct acpi_processor_px *px, int i) {};
#endif
static int acpi_processor_get_performance_states(struct acpi_processor *pr)
{
int result = 0;
@ -379,6 +407,8 @@ static int acpi_processor_get_performance_states(struct acpi_processor *pr)
goto end;
}
amd_fixup_frequency(px, i);
ACPI_DEBUG_PRINT((ACPI_DB_INFO,
"State [%d]: core_frequency[%d] power[%d] transition_latency[%d] bus_master_latency[%d] control[0x%x] status[0x%x]\n",
i,

View File

@ -22,6 +22,7 @@
#include <linux/rculist.h>
#include <linux/rcupdate.h>
#include <linux/opp.h>
#include <linux/of.h>
/*
* Internal data structure organization with the OPP layer library is as
@ -674,3 +675,49 @@ struct srcu_notifier_head *opp_get_notifier(struct device *dev)
return &dev_opp->head;
}
#ifdef CONFIG_OF
/**
* of_init_opp_table() - Initialize opp table from device tree
* @dev: device pointer used to lookup device OPPs.
*
* Register the initial OPP table with the OPP library for given device.
*/
int of_init_opp_table(struct device *dev)
{
const struct property *prop;
const __be32 *val;
int nr;
prop = of_find_property(dev->of_node, "operating-points", NULL);
if (!prop)
return -ENODEV;
if (!prop->value)
return -ENODATA;
/*
* Each OPP is a set of tuples consisting of frequency and
* voltage like <freq-kHz vol-uV>.
*/
nr = prop->length / sizeof(u32);
if (nr % 2) {
dev_err(dev, "%s: Invalid OPP list\n", __func__);
return -EINVAL;
}
val = prop->value;
while (nr) {
unsigned long freq = be32_to_cpup(val++) * 1000;
unsigned long volt = be32_to_cpup(val++);
if (opp_add(dev, freq, volt)) {
dev_warn(dev, "%s: Failed to add OPP %ld\n",
__func__, freq);
continue;
}
nr -= 2;
}
return 0;
}
#endif

View File

@ -179,6 +179,17 @@ config CPU_FREQ_GOV_CONSERVATIVE
If in doubt, say N.
config GENERIC_CPUFREQ_CPU0
bool "Generic CPU0 cpufreq driver"
depends on HAVE_CLK && REGULATOR && PM_OPP && OF
select CPU_FREQ_TABLE
help
This adds a generic cpufreq driver for CPU0 frequency management.
It supports both uniprocessor (UP) and symmetric multiprocessor (SMP)
systems which share clock and voltage across all CPUs.
If in doubt, say N.
menu "x86 CPU frequency scaling drivers"
depends on X86
source "drivers/cpufreq/Kconfig.x86"

View File

@ -23,7 +23,8 @@ config X86_ACPI_CPUFREQ
help
This driver adds a CPUFreq driver which utilizes the ACPI
Processor Performance States.
This driver also supports Intel Enhanced Speedstep.
This driver also supports Intel Enhanced Speedstep and newer
AMD CPUs.
To compile this driver as a module, choose M here: the
module will be called acpi-cpufreq.
@ -32,6 +33,18 @@ config X86_ACPI_CPUFREQ
If in doubt, say N.
config X86_ACPI_CPUFREQ_CPB
default y
bool "Legacy cpb sysfs knob support for AMD CPUs"
depends on X86_ACPI_CPUFREQ && CPU_SUP_AMD
help
The powernow-k8 driver used to provide a sysfs knob called "cpb"
to disable the Core Performance Boosting feature of AMD CPUs. This
file has now been superseeded by the more generic "boost" entry.
By enabling this option the acpi_cpufreq driver provides the old
entry in addition to the new boost ones, for compatibility reasons.
config ELAN_CPUFREQ
tristate "AMD Elan SC400 and SC410"
select CPU_FREQ_TABLE
@ -95,7 +108,8 @@ config X86_POWERNOW_K8
select CPU_FREQ_TABLE
depends on ACPI && ACPI_PROCESSOR
help
This adds the CPUFreq driver for K8/K10 Opteron/Athlon64 processors.
This adds the CPUFreq driver for K8/early Opteron/Athlon64 processors.
Support for K10 and newer processors is now in acpi-cpufreq.
To compile this driver as a module, choose M here: the
module will be called powernow-k8.

View File

@ -13,13 +13,15 @@ obj-$(CONFIG_CPU_FREQ_GOV_CONSERVATIVE) += cpufreq_conservative.o
# CPUfreq cross-arch helpers
obj-$(CONFIG_CPU_FREQ_TABLE) += freq_table.o
obj-$(CONFIG_GENERIC_CPUFREQ_CPU0) += cpufreq-cpu0.o
##################################################################################
# x86 drivers.
# Link order matters. K8 is preferred to ACPI because of firmware bugs in early
# K8 systems. ACPI is preferred to all other hardware-specific drivers.
# speedstep-* is preferred over p4-clockmod.
obj-$(CONFIG_X86_POWERNOW_K8) += powernow-k8.o mperf.o
obj-$(CONFIG_X86_POWERNOW_K8) += powernow-k8.o
obj-$(CONFIG_X86_ACPI_CPUFREQ) += acpi-cpufreq.o mperf.o
obj-$(CONFIG_X86_PCC_CPUFREQ) += pcc-cpufreq.o
obj-$(CONFIG_X86_POWERNOW_K6) += powernow-k6.o

View File

@ -51,13 +51,19 @@ MODULE_AUTHOR("Paul Diefenbaugh, Dominik Brodowski");
MODULE_DESCRIPTION("ACPI Processor P-States Driver");
MODULE_LICENSE("GPL");
#define PFX "acpi-cpufreq: "
enum {
UNDEFINED_CAPABLE = 0,
SYSTEM_INTEL_MSR_CAPABLE,
SYSTEM_AMD_MSR_CAPABLE,
SYSTEM_IO_CAPABLE,
};
#define INTEL_MSR_RANGE (0xffff)
#define AMD_MSR_RANGE (0x7)
#define MSR_K7_HWCR_CPB_DIS (1ULL << 25)
struct acpi_cpufreq_data {
struct acpi_processor_performance *acpi_data;
@ -74,6 +80,116 @@ static struct acpi_processor_performance __percpu *acpi_perf_data;
static struct cpufreq_driver acpi_cpufreq_driver;
static unsigned int acpi_pstate_strict;
static bool boost_enabled, boost_supported;
static struct msr __percpu *msrs;
static bool boost_state(unsigned int cpu)
{
u32 lo, hi;
u64 msr;
switch (boot_cpu_data.x86_vendor) {
case X86_VENDOR_INTEL:
rdmsr_on_cpu(cpu, MSR_IA32_MISC_ENABLE, &lo, &hi);
msr = lo | ((u64)hi << 32);
return !(msr & MSR_IA32_MISC_ENABLE_TURBO_DISABLE);
case X86_VENDOR_AMD:
rdmsr_on_cpu(cpu, MSR_K7_HWCR, &lo, &hi);
msr = lo | ((u64)hi << 32);
return !(msr & MSR_K7_HWCR_CPB_DIS);
}
return false;
}
static void boost_set_msrs(bool enable, const struct cpumask *cpumask)
{
u32 cpu;
u32 msr_addr;
u64 msr_mask;
switch (boot_cpu_data.x86_vendor) {
case X86_VENDOR_INTEL:
msr_addr = MSR_IA32_MISC_ENABLE;
msr_mask = MSR_IA32_MISC_ENABLE_TURBO_DISABLE;
break;
case X86_VENDOR_AMD:
msr_addr = MSR_K7_HWCR;
msr_mask = MSR_K7_HWCR_CPB_DIS;
break;
default:
return;
}
rdmsr_on_cpus(cpumask, msr_addr, msrs);
for_each_cpu(cpu, cpumask) {
struct msr *reg = per_cpu_ptr(msrs, cpu);
if (enable)
reg->q &= ~msr_mask;
else
reg->q |= msr_mask;
}
wrmsr_on_cpus(cpumask, msr_addr, msrs);
}
static ssize_t _store_boost(const char *buf, size_t count)
{
int ret;
unsigned long val = 0;
if (!boost_supported)
return -EINVAL;
ret = kstrtoul(buf, 10, &val);
if (ret || (val > 1))
return -EINVAL;
if ((val && boost_enabled) || (!val && !boost_enabled))
return count;
get_online_cpus();
boost_set_msrs(val, cpu_online_mask);
put_online_cpus();
boost_enabled = val;
pr_debug("Core Boosting %sabled.\n", val ? "en" : "dis");
return count;
}
static ssize_t store_global_boost(struct kobject *kobj, struct attribute *attr,
const char *buf, size_t count)
{
return _store_boost(buf, count);
}
static ssize_t show_global_boost(struct kobject *kobj,
struct attribute *attr, char *buf)
{
return sprintf(buf, "%u\n", boost_enabled);
}
static struct global_attr global_boost = __ATTR(boost, 0644,
show_global_boost,
store_global_boost);
#ifdef CONFIG_X86_ACPI_CPUFREQ_CPB
static ssize_t store_cpb(struct cpufreq_policy *policy, const char *buf,
size_t count)
{
return _store_boost(buf, count);
}
static ssize_t show_cpb(struct cpufreq_policy *policy, char *buf)
{
return sprintf(buf, "%u\n", boost_enabled);
}
static struct freq_attr cpb = __ATTR(cpb, 0644, show_cpb, store_cpb);
#endif
static int check_est_cpu(unsigned int cpuid)
{
@ -82,6 +198,13 @@ static int check_est_cpu(unsigned int cpuid)
return cpu_has(cpu, X86_FEATURE_EST);
}
static int check_amd_hwpstate_cpu(unsigned int cpuid)
{
struct cpuinfo_x86 *cpu = &cpu_data(cpuid);
return cpu_has(cpu, X86_FEATURE_HW_PSTATE);
}
static unsigned extract_io(u32 value, struct acpi_cpufreq_data *data)
{
struct acpi_processor_performance *perf;
@ -101,7 +224,11 @@ static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data)
int i;
struct acpi_processor_performance *perf;
if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD)
msr &= AMD_MSR_RANGE;
else
msr &= INTEL_MSR_RANGE;
perf = data->acpi_data;
for (i = 0; data->freq_table[i].frequency != CPUFREQ_TABLE_END; i++) {
@ -115,6 +242,7 @@ static unsigned extract_freq(u32 val, struct acpi_cpufreq_data *data)
{
switch (data->cpu_feature) {
case SYSTEM_INTEL_MSR_CAPABLE:
case SYSTEM_AMD_MSR_CAPABLE:
return extract_msr(val, data);
case SYSTEM_IO_CAPABLE:
return extract_io(val, data);
@ -150,6 +278,7 @@ static void do_drv_read(void *_cmd)
switch (cmd->type) {
case SYSTEM_INTEL_MSR_CAPABLE:
case SYSTEM_AMD_MSR_CAPABLE:
rdmsr(cmd->addr.msr.reg, cmd->val, h);
break;
case SYSTEM_IO_CAPABLE:
@ -174,6 +303,9 @@ static void do_drv_write(void *_cmd)
lo = (lo & ~INTEL_MSR_RANGE) | (cmd->val & INTEL_MSR_RANGE);
wrmsr(cmd->addr.msr.reg, lo, hi);
break;
case SYSTEM_AMD_MSR_CAPABLE:
wrmsr(cmd->addr.msr.reg, cmd->val, 0);
break;
case SYSTEM_IO_CAPABLE:
acpi_os_write_port((acpi_io_address)cmd->addr.io.port,
cmd->val,
@ -217,6 +349,10 @@ static u32 get_cur_val(const struct cpumask *mask)
cmd.type = SYSTEM_INTEL_MSR_CAPABLE;
cmd.addr.msr.reg = MSR_IA32_PERF_STATUS;
break;
case SYSTEM_AMD_MSR_CAPABLE:
cmd.type = SYSTEM_AMD_MSR_CAPABLE;
cmd.addr.msr.reg = MSR_AMD_PERF_STATUS;
break;
case SYSTEM_IO_CAPABLE:
cmd.type = SYSTEM_IO_CAPABLE;
perf = per_cpu(acfreq_data, cpumask_first(mask))->acpi_data;
@ -326,6 +462,11 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy,
cmd.addr.msr.reg = MSR_IA32_PERF_CTL;
cmd.val = (u32) perf->states[next_perf_state].control;
break;
case SYSTEM_AMD_MSR_CAPABLE:
cmd.type = SYSTEM_AMD_MSR_CAPABLE;
cmd.addr.msr.reg = MSR_AMD_PERF_CTL;
cmd.val = (u32) perf->states[next_perf_state].control;
break;
case SYSTEM_IO_CAPABLE:
cmd.type = SYSTEM_IO_CAPABLE;
cmd.addr.io.port = perf->control_register.address;
@ -419,6 +560,44 @@ static void free_acpi_perf_data(void)
free_percpu(acpi_perf_data);
}
static int boost_notify(struct notifier_block *nb, unsigned long action,
void *hcpu)
{
unsigned cpu = (long)hcpu;
const struct cpumask *cpumask;
cpumask = get_cpu_mask(cpu);
/*
* Clear the boost-disable bit on the CPU_DOWN path so that
* this cpu cannot block the remaining ones from boosting. On
* the CPU_UP path we simply keep the boost-disable flag in
* sync with the current global state.
*/
switch (action) {
case CPU_UP_PREPARE:
case CPU_UP_PREPARE_FROZEN:
boost_set_msrs(boost_enabled, cpumask);
break;
case CPU_DOWN_PREPARE:
case CPU_DOWN_PREPARE_FROZEN:
boost_set_msrs(1, cpumask);
break;
default:
break;
}
return NOTIFY_OK;
}
static struct notifier_block boost_nb = {
.notifier_call = boost_notify,
};
/*
* acpi_cpufreq_early_init - initialize ACPI P-States library
*
@ -559,6 +738,14 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
policy->shared_type = CPUFREQ_SHARED_TYPE_ALL;
cpumask_copy(policy->cpus, cpu_core_mask(cpu));
}
if (check_amd_hwpstate_cpu(cpu) && !acpi_pstate_strict) {
cpumask_clear(policy->cpus);
cpumask_set_cpu(cpu, policy->cpus);
cpumask_copy(policy->related_cpus, cpu_sibling_mask(cpu));
policy->shared_type = CPUFREQ_SHARED_TYPE_HW;
pr_info_once(PFX "overriding BIOS provided _PSD data\n");
}
#endif
/* capability check */
@ -580,12 +767,16 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
break;
case ACPI_ADR_SPACE_FIXED_HARDWARE:
pr_debug("HARDWARE addr space\n");
if (!check_est_cpu(cpu)) {
result = -ENODEV;
goto err_unreg;
}
if (check_est_cpu(cpu)) {
data->cpu_feature = SYSTEM_INTEL_MSR_CAPABLE;
break;
}
if (check_amd_hwpstate_cpu(cpu)) {
data->cpu_feature = SYSTEM_AMD_MSR_CAPABLE;
break;
}
result = -ENODEV;
goto err_unreg;
default:
pr_debug("Unknown addr space %d\n",
(u32) (perf->control_register.space_id));
@ -718,6 +909,7 @@ static int acpi_cpufreq_resume(struct cpufreq_policy *policy)
static struct freq_attr *acpi_cpufreq_attr[] = {
&cpufreq_freq_attr_scaling_available_freqs,
NULL, /* this is a placeholder for cpb, do not remove */
NULL,
};
@ -733,6 +925,49 @@ static struct cpufreq_driver acpi_cpufreq_driver = {
.attr = acpi_cpufreq_attr,
};
static void __init acpi_cpufreq_boost_init(void)
{
if (boot_cpu_has(X86_FEATURE_CPB) || boot_cpu_has(X86_FEATURE_IDA)) {
msrs = msrs_alloc();
if (!msrs)
return;
boost_supported = true;
boost_enabled = boost_state(0);
get_online_cpus();
/* Force all MSRs to the same value */
boost_set_msrs(boost_enabled, cpu_online_mask);
register_cpu_notifier(&boost_nb);
put_online_cpus();
} else
global_boost.attr.mode = 0444;
/* We create the boost file in any case, though for systems without
* hardware support it will be read-only and hardwired to return 0.
*/
if (sysfs_create_file(cpufreq_global_kobject, &(global_boost.attr)))
pr_warn(PFX "could not register global boost sysfs file\n");
else
pr_debug("registered global boost sysfs file\n");
}
static void __exit acpi_cpufreq_boost_exit(void)
{
sysfs_remove_file(cpufreq_global_kobject, &(global_boost.attr));
if (msrs) {
unregister_cpu_notifier(&boost_nb);
msrs_free(msrs);
msrs = NULL;
}
}
static int __init acpi_cpufreq_init(void)
{
int ret;
@ -746,9 +981,32 @@ static int __init acpi_cpufreq_init(void)
if (ret)
return ret;
#ifdef CONFIG_X86_ACPI_CPUFREQ_CPB
/* this is a sysfs file with a strange name and an even stranger
* semantic - per CPU instantiation, but system global effect.
* Lets enable it only on AMD CPUs for compatibility reasons and
* only if configured. This is considered legacy code, which
* will probably be removed at some point in the future.
*/
if (check_amd_hwpstate_cpu(0)) {
struct freq_attr **iter;
pr_debug("adding sysfs entry for cpb\n");
for (iter = acpi_cpufreq_attr; *iter != NULL; iter++)
;
/* make sure there is a terminator behind it */
if (iter[1] == NULL)
*iter = &cpb;
}
#endif
ret = cpufreq_register_driver(&acpi_cpufreq_driver);
if (ret)
free_acpi_perf_data();
else
acpi_cpufreq_boost_init();
return ret;
}
@ -757,6 +1015,8 @@ static void __exit acpi_cpufreq_exit(void)
{
pr_debug("acpi_cpufreq_exit\n");
acpi_cpufreq_boost_exit();
cpufreq_unregister_driver(&acpi_cpufreq_driver);
free_acpi_perf_data();

View File

@ -0,0 +1,269 @@
/*
* Copyright (C) 2012 Freescale Semiconductor, Inc.
*
* The OPP code in function cpu0_set_target() is reused from
* drivers/cpufreq/omap-cpufreq.c
*
* 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) KBUILD_MODNAME ": " fmt
#include <linux/clk.h>
#include <linux/cpu.h>
#include <linux/cpufreq.h>
#include <linux/err.h>
#include <linux/module.h>
#include <linux/of.h>
#include <linux/opp.h>
#include <linux/regulator/consumer.h>
#include <linux/slab.h>
static unsigned int transition_latency;
static unsigned int voltage_tolerance; /* in percentage */
static struct device *cpu_dev;
static struct clk *cpu_clk;
static struct regulator *cpu_reg;
static struct cpufreq_frequency_table *freq_table;
static int cpu0_verify_speed(struct cpufreq_policy *policy)
{
return cpufreq_frequency_table_verify(policy, freq_table);
}
static unsigned int cpu0_get_speed(unsigned int cpu)
{
return clk_get_rate(cpu_clk) / 1000;
}
static int cpu0_set_target(struct cpufreq_policy *policy,
unsigned int target_freq, unsigned int relation)
{
struct cpufreq_freqs freqs;
struct opp *opp;
unsigned long freq_Hz, volt = 0, volt_old = 0, tol = 0;
unsigned int index, cpu;
int ret;
ret = cpufreq_frequency_table_target(policy, freq_table, target_freq,
relation, &index);
if (ret) {
pr_err("failed to match target freqency %d: %d\n",
target_freq, ret);
return ret;
}
freq_Hz = clk_round_rate(cpu_clk, freq_table[index].frequency * 1000);
if (freq_Hz < 0)
freq_Hz = freq_table[index].frequency * 1000;
freqs.new = freq_Hz / 1000;
freqs.old = clk_get_rate(cpu_clk) / 1000;
if (freqs.old == freqs.new)
return 0;
for_each_online_cpu(cpu) {
freqs.cpu = cpu;
cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
}
if (cpu_reg) {
opp = opp_find_freq_ceil(cpu_dev, &freq_Hz);
if (IS_ERR(opp)) {
pr_err("failed to find OPP for %ld\n", freq_Hz);
return PTR_ERR(opp);
}
volt = opp_get_voltage(opp);
tol = volt * voltage_tolerance / 100;
volt_old = regulator_get_voltage(cpu_reg);
}
pr_debug("%u MHz, %ld mV --> %u MHz, %ld mV\n",
freqs.old / 1000, volt_old ? volt_old / 1000 : -1,
freqs.new / 1000, volt ? volt / 1000 : -1);
/* scaling up? scale voltage before frequency */
if (cpu_reg && freqs.new > freqs.old) {
ret = regulator_set_voltage_tol(cpu_reg, volt, tol);
if (ret) {
pr_err("failed to scale voltage up: %d\n", ret);
freqs.new = freqs.old;
return ret;
}
}
ret = clk_set_rate(cpu_clk, freqs.new * 1000);
if (ret) {
pr_err("failed to set clock rate: %d\n", ret);
if (cpu_reg)
regulator_set_voltage_tol(cpu_reg, volt_old, tol);
return ret;
}
/* scaling down? scale voltage after frequency */
if (cpu_reg && freqs.new < freqs.old) {
ret = regulator_set_voltage_tol(cpu_reg, volt, tol);
if (ret) {
pr_err("failed to scale voltage down: %d\n", ret);
clk_set_rate(cpu_clk, freqs.old * 1000);
freqs.new = freqs.old;
return ret;
}
}
for_each_online_cpu(cpu) {
freqs.cpu = cpu;
cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
}
return 0;
}
static int cpu0_cpufreq_init(struct cpufreq_policy *policy)
{
int ret;
if (policy->cpu != 0)
return -EINVAL;
ret = cpufreq_frequency_table_cpuinfo(policy, freq_table);
if (ret) {
pr_err("invalid frequency table: %d\n", ret);
return ret;
}
policy->cpuinfo.transition_latency = transition_latency;
policy->cur = clk_get_rate(cpu_clk) / 1000;
/*
* The driver only supports the SMP configuartion where all processors
* share the clock and voltage and clock. Use cpufreq affected_cpus
* interface to have all CPUs scaled together.
*/
policy->shared_type = CPUFREQ_SHARED_TYPE_ANY;
cpumask_setall(policy->cpus);
cpufreq_frequency_table_get_attr(freq_table, policy->cpu);
return 0;
}
static int cpu0_cpufreq_exit(struct cpufreq_policy *policy)
{
cpufreq_frequency_table_put_attr(policy->cpu);
return 0;
}
static struct freq_attr *cpu0_cpufreq_attr[] = {
&cpufreq_freq_attr_scaling_available_freqs,
NULL,
};
static struct cpufreq_driver cpu0_cpufreq_driver = {
.flags = CPUFREQ_STICKY,
.verify = cpu0_verify_speed,
.target = cpu0_set_target,
.get = cpu0_get_speed,
.init = cpu0_cpufreq_init,
.exit = cpu0_cpufreq_exit,
.name = "generic_cpu0",
.attr = cpu0_cpufreq_attr,
};
static int __devinit cpu0_cpufreq_driver_init(void)
{
struct device_node *np;
int ret;
np = of_find_node_by_path("/cpus/cpu@0");
if (!np) {
pr_err("failed to find cpu0 node\n");
return -ENOENT;
}
cpu_dev = get_cpu_device(0);
if (!cpu_dev) {
pr_err("failed to get cpu0 device\n");
ret = -ENODEV;
goto out_put_node;
}
cpu_dev->of_node = np;
cpu_clk = clk_get(cpu_dev, NULL);
if (IS_ERR(cpu_clk)) {
ret = PTR_ERR(cpu_clk);
pr_err("failed to get cpu0 clock: %d\n", ret);
goto out_put_node;
}
cpu_reg = regulator_get(cpu_dev, "cpu0");
if (IS_ERR(cpu_reg)) {
pr_warn("failed to get cpu0 regulator\n");
cpu_reg = NULL;
}
ret = of_init_opp_table(cpu_dev);
if (ret) {
pr_err("failed to init OPP table: %d\n", ret);
goto out_put_node;
}
ret = opp_init_cpufreq_table(cpu_dev, &freq_table);
if (ret) {
pr_err("failed to init cpufreq table: %d\n", ret);
goto out_put_node;
}
of_property_read_u32(np, "voltage-tolerance", &voltage_tolerance);
if (of_property_read_u32(np, "clock-latency", &transition_latency))
transition_latency = CPUFREQ_ETERNAL;
if (cpu_reg) {
struct opp *opp;
unsigned long min_uV, max_uV;
int i;
/*
* OPP is maintained in order of increasing frequency, and
* freq_table initialised from OPP is therefore sorted in the
* same order.
*/
for (i = 0; freq_table[i].frequency != CPUFREQ_TABLE_END; i++)
;
opp = opp_find_freq_exact(cpu_dev,
freq_table[0].frequency * 1000, true);
min_uV = opp_get_voltage(opp);
opp = opp_find_freq_exact(cpu_dev,
freq_table[i-1].frequency * 1000, true);
max_uV = opp_get_voltage(opp);
ret = regulator_set_voltage_time(cpu_reg, min_uV, max_uV);
if (ret > 0)
transition_latency += ret * 1000;
}
ret = cpufreq_register_driver(&cpu0_cpufreq_driver);
if (ret) {
pr_err("failed register driver: %d\n", ret);
goto out_free_table;
}
of_node_put(np);
return 0;
out_free_table:
opp_free_cpufreq_table(cpu_dev, &freq_table);
out_put_node:
of_node_put(np);
return ret;
}
late_initcall(cpu0_cpufreq_driver_init);
MODULE_AUTHOR("Shawn Guo <shawn.guo@linaro.org>");
MODULE_DESCRIPTION("Generic CPU0 cpufreq driver");
MODULE_LICENSE("GPL");

View File

@ -504,6 +504,7 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy,
j_dbs_info->prev_cpu_nice =
kcpustat_cpu(j).cpustat[CPUTIME_NICE];
}
this_dbs_info->cpu = cpu;
this_dbs_info->down_skip = 0;
this_dbs_info->requested_freq = policy->cur;
@ -583,6 +584,7 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy,
__cpufreq_driver_target(
this_dbs_info->cur_policy,
policy->min, CPUFREQ_RELATION_L);
dbs_check_cpu(this_dbs_info);
mutex_unlock(&this_dbs_info->timer_mutex);
break;

View File

@ -761,6 +761,7 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy,
else if (policy->min > this_dbs_info->cur_policy->cur)
__cpufreq_driver_target(this_dbs_info->cur_policy,
policy->min, CPUFREQ_RELATION_L);
dbs_check_cpu(this_dbs_info);
mutex_unlock(&this_dbs_info->timer_mutex);
break;
}

View File

@ -56,7 +56,7 @@ union msr_longhaul {
/*
* VIA C3 Samuel 1 & Samuel 2 (stepping 0)
*/
static const int __cpuinitdata samuel1_mults[16] = {
static const int __cpuinitconst samuel1_mults[16] = {
-1, /* 0000 -> RESERVED */
30, /* 0001 -> 3.0x */
40, /* 0010 -> 4.0x */
@ -75,7 +75,7 @@ static const int __cpuinitdata samuel1_mults[16] = {
-1, /* 1111 -> RESERVED */
};
static const int __cpuinitdata samuel1_eblcr[16] = {
static const int __cpuinitconst samuel1_eblcr[16] = {
50, /* 0000 -> RESERVED */
30, /* 0001 -> 3.0x */
40, /* 0010 -> 4.0x */
@ -97,7 +97,7 @@ static const int __cpuinitdata samuel1_eblcr[16] = {
/*
* VIA C3 Samuel2 Stepping 1->15
*/
static const int __cpuinitdata samuel2_eblcr[16] = {
static const int __cpuinitconst samuel2_eblcr[16] = {
50, /* 0000 -> 5.0x */
30, /* 0001 -> 3.0x */
40, /* 0010 -> 4.0x */
@ -119,7 +119,7 @@ static const int __cpuinitdata samuel2_eblcr[16] = {
/*
* VIA C3 Ezra
*/
static const int __cpuinitdata ezra_mults[16] = {
static const int __cpuinitconst ezra_mults[16] = {
100, /* 0000 -> 10.0x */
30, /* 0001 -> 3.0x */
40, /* 0010 -> 4.0x */
@ -138,7 +138,7 @@ static const int __cpuinitdata ezra_mults[16] = {
120, /* 1111 -> 12.0x */
};
static const int __cpuinitdata ezra_eblcr[16] = {
static const int __cpuinitconst ezra_eblcr[16] = {
50, /* 0000 -> 5.0x */
30, /* 0001 -> 3.0x */
40, /* 0010 -> 4.0x */
@ -160,7 +160,7 @@ static const int __cpuinitdata ezra_eblcr[16] = {
/*
* VIA C3 (Ezra-T) [C5M].
*/
static const int __cpuinitdata ezrat_mults[32] = {
static const int __cpuinitconst ezrat_mults[32] = {
100, /* 0000 -> 10.0x */
30, /* 0001 -> 3.0x */
40, /* 0010 -> 4.0x */
@ -196,7 +196,7 @@ static const int __cpuinitdata ezrat_mults[32] = {
-1, /* 1111 -> RESERVED (12.0x) */
};
static const int __cpuinitdata ezrat_eblcr[32] = {
static const int __cpuinitconst ezrat_eblcr[32] = {
50, /* 0000 -> 5.0x */
30, /* 0001 -> 3.0x */
40, /* 0010 -> 4.0x */
@ -235,7 +235,7 @@ static const int __cpuinitdata ezrat_eblcr[32] = {
/*
* VIA C3 Nehemiah */
static const int __cpuinitdata nehemiah_mults[32] = {
static const int __cpuinitconst nehemiah_mults[32] = {
100, /* 0000 -> 10.0x */
-1, /* 0001 -> 16.0x */
40, /* 0010 -> 4.0x */
@ -270,7 +270,7 @@ static const int __cpuinitdata nehemiah_mults[32] = {
-1, /* 1111 -> 12.0x */
};
static const int __cpuinitdata nehemiah_eblcr[32] = {
static const int __cpuinitconst nehemiah_eblcr[32] = {
50, /* 0000 -> 5.0x */
160, /* 0001 -> 16.0x */
40, /* 0010 -> 4.0x */
@ -315,7 +315,7 @@ struct mV_pos {
unsigned short pos;
};
static const struct mV_pos __cpuinitdata vrm85_mV[32] = {
static const struct mV_pos __cpuinitconst vrm85_mV[32] = {
{1250, 8}, {1200, 6}, {1150, 4}, {1100, 2},
{1050, 0}, {1800, 30}, {1750, 28}, {1700, 26},
{1650, 24}, {1600, 22}, {1550, 20}, {1500, 18},
@ -326,14 +326,14 @@ static const struct mV_pos __cpuinitdata vrm85_mV[32] = {
{1475, 17}, {1425, 15}, {1375, 13}, {1325, 11}
};
static const unsigned char __cpuinitdata mV_vrm85[32] = {
static const unsigned char __cpuinitconst mV_vrm85[32] = {
0x04, 0x14, 0x03, 0x13, 0x02, 0x12, 0x01, 0x11,
0x00, 0x10, 0x0f, 0x1f, 0x0e, 0x1e, 0x0d, 0x1d,
0x0c, 0x1c, 0x0b, 0x1b, 0x0a, 0x1a, 0x09, 0x19,
0x08, 0x18, 0x07, 0x17, 0x06, 0x16, 0x05, 0x15
};
static const struct mV_pos __cpuinitdata mobilevrm_mV[32] = {
static const struct mV_pos __cpuinitconst mobilevrm_mV[32] = {
{1750, 31}, {1700, 30}, {1650, 29}, {1600, 28},
{1550, 27}, {1500, 26}, {1450, 25}, {1400, 24},
{1350, 23}, {1300, 22}, {1250, 21}, {1200, 20},
@ -344,7 +344,7 @@ static const struct mV_pos __cpuinitdata mobilevrm_mV[32] = {
{675, 3}, {650, 2}, {625, 1}, {600, 0}
};
static const unsigned char __cpuinitdata mV_mobilevrm[32] = {
static const unsigned char __cpuinitconst mV_mobilevrm[32] = {
0x1f, 0x1e, 0x1d, 0x1c, 0x1b, 0x1a, 0x19, 0x18,
0x17, 0x16, 0x15, 0x14, 0x13, 0x12, 0x11, 0x10,
0x0f, 0x0e, 0x0d, 0x0c, 0x0b, 0x0a, 0x09, 0x08,

View File

@ -40,16 +40,6 @@
/* OPP tolerance in percentage */
#define OPP_TOLERANCE 4
#ifdef CONFIG_SMP
struct lpj_info {
unsigned long ref;
unsigned int freq;
};
static DEFINE_PER_CPU(struct lpj_info, lpj_ref);
static struct lpj_info global_lpj_ref;
#endif
static struct cpufreq_frequency_table *freq_table;
static atomic_t freq_table_users = ATOMIC_INIT(0);
static struct clk *mpu_clk;
@ -161,31 +151,6 @@ static int omap_target(struct cpufreq_policy *policy,
}
freqs.new = omap_getspeed(policy->cpu);
#ifdef CONFIG_SMP
/*
* Note that loops_per_jiffy is not updated on SMP systems in
* cpufreq driver. So, update the per-CPU loops_per_jiffy value
* on frequency transition. We need to update all dependent CPUs.
*/
for_each_cpu(i, policy->cpus) {
struct lpj_info *lpj = &per_cpu(lpj_ref, i);
if (!lpj->freq) {
lpj->ref = per_cpu(cpu_data, i).loops_per_jiffy;
lpj->freq = freqs.old;
}
per_cpu(cpu_data, i).loops_per_jiffy =
cpufreq_scale(lpj->ref, lpj->freq, freqs.new);
}
/* And don't forget to adjust the global one */
if (!global_lpj_ref.freq) {
global_lpj_ref.ref = loops_per_jiffy;
global_lpj_ref.freq = freqs.old;
}
loops_per_jiffy = cpufreq_scale(global_lpj_ref.ref, global_lpj_ref.freq,
freqs.new);
#endif
done:
/* notifiers */

View File

@ -49,22 +49,12 @@
#define PFX "powernow-k8: "
#define VERSION "version 2.20.00"
#include "powernow-k8.h"
#include "mperf.h"
/* serialize freq changes */
static DEFINE_MUTEX(fidvid_mutex);
static DEFINE_PER_CPU(struct powernow_k8_data *, powernow_data);
static int cpu_family = CPU_OPTERON;
/* array to map SW pstate number to acpi state */
static u32 ps_to_as[8];
/* core performance boost */
static bool cpb_capable, cpb_enabled;
static struct msr __percpu *msrs;
static struct cpufreq_driver cpufreq_amd64_driver;
#ifndef CONFIG_SMP
@ -86,12 +76,6 @@ static u32 find_khz_freq_from_fid(u32 fid)
return 1000 * find_freq_from_fid(fid);
}
static u32 find_khz_freq_from_pstate(struct cpufreq_frequency_table *data,
u32 pstate)
{
return data[ps_to_as[pstate]].frequency;
}
/* Return the vco fid for an input fid
*
* Each "low" fid has corresponding "high" fid, and you can get to "low" fids
@ -114,9 +98,6 @@ static int pending_bit_stuck(void)
{
u32 lo, hi;
if (cpu_family == CPU_HW_PSTATE)
return 0;
rdmsr(MSR_FIDVID_STATUS, lo, hi);
return lo & MSR_S_LO_CHANGE_PENDING ? 1 : 0;
}
@ -130,20 +111,6 @@ static int query_current_values_with_pending_wait(struct powernow_k8_data *data)
u32 lo, hi;
u32 i = 0;
if (cpu_family == CPU_HW_PSTATE) {
rdmsr(MSR_PSTATE_STATUS, lo, hi);
i = lo & HW_PSTATE_MASK;
data->currpstate = i;
/*
* a workaround for family 11h erratum 311 might cause
* an "out-of-range Pstate if the core is in Pstate-0
*/
if ((boot_cpu_data.x86 == 0x11) && (i >= data->numps))
data->currpstate = HW_PSTATE_0;
return 0;
}
do {
if (i++ > 10000) {
pr_debug("detected change pending stuck\n");
@ -300,14 +267,6 @@ static int decrease_vid_code_by_step(struct powernow_k8_data *data,
return 0;
}
/* Change hardware pstate by single MSR write */
static int transition_pstate(struct powernow_k8_data *data, u32 pstate)
{
wrmsr(MSR_PSTATE_CTRL, pstate, 0);
data->currpstate = pstate;
return 0;
}
/* Change Opteron/Athlon64 fid and vid, by the 3 phases. */
static int transition_fid_vid(struct powernow_k8_data *data,
u32 reqfid, u32 reqvid)
@ -524,8 +483,6 @@ static int core_voltage_post_transition(struct powernow_k8_data *data,
static const struct x86_cpu_id powernow_k8_ids[] = {
/* IO based frequency switching */
{ X86_VENDOR_AMD, 0xf },
/* MSR based frequency switching supported */
X86_FEATURE_MATCH(X86_FEATURE_HW_PSTATE),
{}
};
MODULE_DEVICE_TABLE(x86cpu, powernow_k8_ids);
@ -561,15 +518,8 @@ static void check_supported_cpu(void *_rc)
"Power state transitions not supported\n");
return;
}
} else { /* must be a HW Pstate capable processor */
cpuid(CPUID_FREQ_VOLT_CAPABILITIES, &eax, &ebx, &ecx, &edx);
if ((edx & USE_HW_PSTATE) == USE_HW_PSTATE)
cpu_family = CPU_HW_PSTATE;
else
return;
}
*rc = 0;
}
}
static int check_pst_table(struct powernow_k8_data *data, struct pst_s *pst,
@ -633,12 +583,6 @@ static void print_basics(struct powernow_k8_data *data)
for (j = 0; j < data->numps; j++) {
if (data->powernow_table[j].frequency !=
CPUFREQ_ENTRY_INVALID) {
if (cpu_family == CPU_HW_PSTATE) {
printk(KERN_INFO PFX
" %d : pstate %d (%d MHz)\n", j,
data->powernow_table[j].index,
data->powernow_table[j].frequency/1000);
} else {
printk(KERN_INFO PFX
"fid 0x%x (%d MHz), vid 0x%x\n",
data->powernow_table[j].index & 0xff,
@ -646,26 +590,11 @@ static void print_basics(struct powernow_k8_data *data)
data->powernow_table[j].index >> 8);
}
}
}
if (data->batps)
printk(KERN_INFO PFX "Only %d pstates on battery\n",
data->batps);
}
static u32 freq_from_fid_did(u32 fid, u32 did)
{
u32 mhz = 0;
if (boot_cpu_data.x86 == 0x10)
mhz = (100 * (fid + 0x10)) >> did;
else if (boot_cpu_data.x86 == 0x11)
mhz = (100 * (fid + 8)) >> did;
else
BUG();
return mhz * 1000;
}
static int fill_powernow_table(struct powernow_k8_data *data,
struct pst_s *pst, u8 maxvid)
{
@ -825,7 +754,7 @@ static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data,
{
u64 control;
if (!data->acpi_data.state_count || (cpu_family == CPU_HW_PSTATE))
if (!data->acpi_data.state_count)
return;
control = data->acpi_data.states[index].control;
@ -876,9 +805,6 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data)
data->numps = data->acpi_data.state_count;
powernow_k8_acpi_pst_values(data, 0);
if (cpu_family == CPU_HW_PSTATE)
ret_val = fill_powernow_table_pstate(data, powernow_table);
else
ret_val = fill_powernow_table_fidvid(data, powernow_table);
if (ret_val)
goto err_out_mem;
@ -916,51 +842,6 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data)
return ret_val;
}
static int fill_powernow_table_pstate(struct powernow_k8_data *data,
struct cpufreq_frequency_table *powernow_table)
{
int i;
u32 hi = 0, lo = 0;
rdmsr(MSR_PSTATE_CUR_LIMIT, lo, hi);
data->max_hw_pstate = (lo & HW_PSTATE_MAX_MASK) >> HW_PSTATE_MAX_SHIFT;
for (i = 0; i < data->acpi_data.state_count; i++) {
u32 index;
index = data->acpi_data.states[i].control & HW_PSTATE_MASK;
if (index > data->max_hw_pstate) {
printk(KERN_ERR PFX "invalid pstate %d - "
"bad value %d.\n", i, index);
printk(KERN_ERR PFX "Please report to BIOS "
"manufacturer\n");
invalidate_entry(powernow_table, i);
continue;
}
ps_to_as[index] = i;
/* Frequency may be rounded for these */
if ((boot_cpu_data.x86 == 0x10 && boot_cpu_data.x86_model < 10)
|| boot_cpu_data.x86 == 0x11) {
rdmsr(MSR_PSTATE_DEF_BASE + index, lo, hi);
if (!(hi & HW_PSTATE_VALID_MASK)) {
pr_debug("invalid pstate %d, ignoring\n", index);
invalidate_entry(powernow_table, i);
continue;
}
powernow_table[i].frequency =
freq_from_fid_did(lo & 0x3f, (lo >> 6) & 7);
} else
powernow_table[i].frequency =
data->acpi_data.states[i].core_frequency * 1000;
powernow_table[i].index = index;
}
return 0;
}
static int fill_powernow_table_fidvid(struct powernow_k8_data *data,
struct cpufreq_frequency_table *powernow_table)
{
@ -1037,15 +918,7 @@ static int get_transition_latency(struct powernow_k8_data *data)
max_latency = cur_latency;
}
if (max_latency == 0) {
/*
* Fam 11h and later may return 0 as transition latency. This
* is intended and means "very fast". While cpufreq core and
* governors currently can handle that gracefully, better set it
* to 1 to avoid problems in the future.
*/
if (boot_cpu_data.x86 < 0x11)
printk(KERN_ERR FW_WARN PFX "Invalid zero transition "
"latency\n");
pr_err(FW_WARN PFX "Invalid zero transition latency\n");
max_latency = 1;
}
/* value in usecs, needs to be in nanoseconds */
@ -1105,40 +978,6 @@ static int transition_frequency_fidvid(struct powernow_k8_data *data,
return res;
}
/* Take a frequency, and issue the hardware pstate transition command */
static int transition_frequency_pstate(struct powernow_k8_data *data,
unsigned int index)
{
u32 pstate = 0;
int res, i;
struct cpufreq_freqs freqs;
pr_debug("cpu %d transition to index %u\n", smp_processor_id(), index);
/* get MSR index for hardware pstate transition */
pstate = index & HW_PSTATE_MASK;
if (pstate > data->max_hw_pstate)
return -EINVAL;
freqs.old = find_khz_freq_from_pstate(data->powernow_table,
data->currpstate);
freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate);
for_each_cpu(i, data->available_cores) {
freqs.cpu = i;
cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE);
}
res = transition_pstate(data, pstate);
freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate);
for_each_cpu(i, data->available_cores) {
freqs.cpu = i;
cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE);
}
return res;
}
/* Driver entry point to switch to the target frequency */
static int powernowk8_target(struct cpufreq_policy *pol,
unsigned targfreq, unsigned relation)
@ -1180,19 +1019,16 @@ static int powernowk8_target(struct cpufreq_policy *pol,
if (query_current_values_with_pending_wait(data))
goto err_out;
if (cpu_family != CPU_HW_PSTATE) {
pr_debug("targ: curr fid 0x%x, vid 0x%x\n",
data->currfid, data->currvid);
if ((checkvid != data->currvid) ||
(checkfid != data->currfid)) {
printk(KERN_INFO PFX
"error - out of sync, fix 0x%x 0x%x, "
"vid 0x%x 0x%x\n",
pr_info(PFX
"error - out of sync, fix 0x%x 0x%x, vid 0x%x 0x%x\n",
checkfid, data->currfid,
checkvid, data->currvid);
}
}
if (cpufreq_frequency_table_target(pol, data->powernow_table,
targfreq, relation, &newstate))
@ -1202,11 +1038,8 @@ static int powernowk8_target(struct cpufreq_policy *pol,
powernow_k8_acpi_pst_values(data, newstate);
if (cpu_family == CPU_HW_PSTATE)
ret = transition_frequency_pstate(data,
data->powernow_table[newstate].index);
else
ret = transition_frequency_fidvid(data, newstate);
if (ret) {
printk(KERN_ERR PFX "transition frequency failed\n");
ret = 1;
@ -1215,10 +1048,6 @@ static int powernowk8_target(struct cpufreq_policy *pol,
}
mutex_unlock(&fidvid_mutex);
if (cpu_family == CPU_HW_PSTATE)
pol->cur = find_khz_freq_from_pstate(data->powernow_table,
data->powernow_table[newstate].index);
else
pol->cur = find_khz_freq_from_fid(data->currfid);
ret = 0;
@ -1259,22 +1088,23 @@ static void __cpuinit powernowk8_cpu_init_on_cpu(void *_init_on_cpu)
return;
}
if (cpu_family == CPU_OPTERON)
fidvid_msr_init();
init_on_cpu->rc = 0;
}
static const char missing_pss_msg[] =
KERN_ERR
FW_BUG PFX "No compatible ACPI _PSS objects found.\n"
FW_BUG PFX "First, make sure Cool'N'Quiet is enabled in the BIOS.\n"
FW_BUG PFX "If that doesn't help, try upgrading your BIOS.\n";
/* per CPU init entry point to the driver */
static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
{
static const char ACPI_PSS_BIOS_BUG_MSG[] =
KERN_ERR FW_BUG PFX "No compatible ACPI _PSS objects found.\n"
FW_BUG PFX "Try again with latest BIOS.\n";
struct powernow_k8_data *data;
struct init_on_cpu init_on_cpu;
int rc;
struct cpuinfo_x86 *c = &cpu_data(pol->cpu);
if (!cpu_online(pol->cpu))
return -ENODEV;
@ -1290,7 +1120,6 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
}
data->cpu = pol->cpu;
data->currpstate = HW_PSTATE_INVALID;
if (powernow_k8_cpu_init_acpi(data)) {
/*
@ -1298,7 +1127,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
* an UP version, and is deprecated by AMD.
*/
if (num_online_cpus() != 1) {
printk_once(ACPI_PSS_BIOS_BUG_MSG);
printk_once(missing_pss_msg);
goto err_out;
}
if (pol->cpu != 0) {
@ -1327,16 +1156,9 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
if (rc != 0)
goto err_out_exit_acpi;
if (cpu_family == CPU_HW_PSTATE)
cpumask_copy(pol->cpus, cpumask_of(pol->cpu));
else
cpumask_copy(pol->cpus, cpu_core_mask(pol->cpu));
data->available_cores = pol->cpus;
if (cpu_family == CPU_HW_PSTATE)
pol->cur = find_khz_freq_from_pstate(data->powernow_table,
data->currpstate);
else
pol->cur = find_khz_freq_from_fid(data->currfid);
pr_debug("policy current frequency %d kHz\n", pol->cur);
@ -1349,16 +1171,8 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
return -EINVAL;
}
/* Check for APERF/MPERF support in hardware */
if (cpu_has(c, X86_FEATURE_APERFMPERF))
cpufreq_amd64_driver.getavg = cpufreq_get_measured_perf;
cpufreq_frequency_table_get_attr(data->powernow_table, pol->cpu);
if (cpu_family == CPU_HW_PSTATE)
pr_debug("cpu_init done, current pstate 0x%x\n",
data->currpstate);
else
pr_debug("cpu_init done, current fid 0x%x, vid 0x%x\n",
data->currfid, data->currvid);
@ -1413,10 +1227,6 @@ static unsigned int powernowk8_get(unsigned int cpu)
if (err)
goto out;
if (cpu_family == CPU_HW_PSTATE)
khz = find_khz_freq_from_pstate(data->powernow_table,
data->currpstate);
else
khz = find_khz_freq_from_fid(data->currfid);
@ -1424,77 +1234,8 @@ static unsigned int powernowk8_get(unsigned int cpu)
return khz;
}
static void _cpb_toggle_msrs(bool t)
{
int cpu;
get_online_cpus();
rdmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs);
for_each_cpu(cpu, cpu_online_mask) {
struct msr *reg = per_cpu_ptr(msrs, cpu);
if (t)
reg->l &= ~BIT(25);
else
reg->l |= BIT(25);
}
wrmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs);
put_online_cpus();
}
/*
* Switch on/off core performance boosting.
*
* 0=disable
* 1=enable.
*/
static void cpb_toggle(bool t)
{
if (!cpb_capable)
return;
if (t && !cpb_enabled) {
cpb_enabled = true;
_cpb_toggle_msrs(t);
printk(KERN_INFO PFX "Core Boosting enabled.\n");
} else if (!t && cpb_enabled) {
cpb_enabled = false;
_cpb_toggle_msrs(t);
printk(KERN_INFO PFX "Core Boosting disabled.\n");
}
}
static ssize_t store_cpb(struct cpufreq_policy *policy, const char *buf,
size_t count)
{
int ret = -EINVAL;
unsigned long val = 0;
ret = strict_strtoul(buf, 10, &val);
if (!ret && (val == 0 || val == 1) && cpb_capable)
cpb_toggle(val);
else
return -EINVAL;
return count;
}
static ssize_t show_cpb(struct cpufreq_policy *policy, char *buf)
{
return sprintf(buf, "%u\n", cpb_enabled);
}
#define define_one_rw(_name) \
static struct freq_attr _name = \
__ATTR(_name, 0644, show_##_name, store_##_name)
define_one_rw(cpb);
static struct freq_attr *powernow_k8_attr[] = {
&cpufreq_freq_attr_scaling_available_freqs,
&cpb,
NULL,
};
@ -1510,53 +1251,18 @@ static struct cpufreq_driver cpufreq_amd64_driver = {
.attr = powernow_k8_attr,
};
/*
* Clear the boost-disable flag on the CPU_DOWN path so that this cpu
* cannot block the remaining ones from boosting. On the CPU_UP path we
* simply keep the boost-disable flag in sync with the current global
* state.
*/
static int cpb_notify(struct notifier_block *nb, unsigned long action,
void *hcpu)
{
unsigned cpu = (long)hcpu;
u32 lo, hi;
switch (action) {
case CPU_UP_PREPARE:
case CPU_UP_PREPARE_FROZEN:
if (!cpb_enabled) {
rdmsr_on_cpu(cpu, MSR_K7_HWCR, &lo, &hi);
lo |= BIT(25);
wrmsr_on_cpu(cpu, MSR_K7_HWCR, lo, hi);
}
break;
case CPU_DOWN_PREPARE:
case CPU_DOWN_PREPARE_FROZEN:
rdmsr_on_cpu(cpu, MSR_K7_HWCR, &lo, &hi);
lo &= ~BIT(25);
wrmsr_on_cpu(cpu, MSR_K7_HWCR, lo, hi);
break;
default:
break;
}
return NOTIFY_OK;
}
static struct notifier_block cpb_nb = {
.notifier_call = cpb_notify,
};
/* driver entry point for init */
static int __cpuinit powernowk8_init(void)
{
unsigned int i, supported_cpus = 0, cpu;
unsigned int i, supported_cpus = 0;
int rv;
if (static_cpu_has(X86_FEATURE_HW_PSTATE)) {
pr_warn(PFX "this CPU is not supported anymore, using acpi-cpufreq instead.\n");
request_module("acpi-cpufreq");
return -ENODEV;
}
if (!x86_match_cpu(powernow_k8_ids))
return -ENODEV;
@ -1570,38 +1276,13 @@ static int __cpuinit powernowk8_init(void)
if (supported_cpus != num_online_cpus())
return -ENODEV;
printk(KERN_INFO PFX "Found %d %s (%d cpu cores) (" VERSION ")\n",
num_online_nodes(), boot_cpu_data.x86_model_id, supported_cpus);
if (boot_cpu_has(X86_FEATURE_CPB)) {
cpb_capable = true;
msrs = msrs_alloc();
if (!msrs) {
printk(KERN_ERR "%s: Error allocating msrs!\n", __func__);
return -ENOMEM;
}
register_cpu_notifier(&cpb_nb);
rdmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs);
for_each_cpu(cpu, cpu_online_mask) {
struct msr *reg = per_cpu_ptr(msrs, cpu);
cpb_enabled |= !(!!(reg->l & BIT(25)));
}
printk(KERN_INFO PFX "Core Performance Boosting: %s.\n",
(cpb_enabled ? "on" : "off"));
}
rv = cpufreq_register_driver(&cpufreq_amd64_driver);
if (rv < 0 && boot_cpu_has(X86_FEATURE_CPB)) {
unregister_cpu_notifier(&cpb_nb);
msrs_free(msrs);
msrs = NULL;
}
if (!rv)
pr_info(PFX "Found %d %s (%d cpu cores) (" VERSION ")\n",
num_online_nodes(), boot_cpu_data.x86_model_id,
supported_cpus);
return rv;
}
@ -1610,13 +1291,6 @@ static void __exit powernowk8_exit(void)
{
pr_debug("exit\n");
if (boot_cpu_has(X86_FEATURE_CPB)) {
msrs_free(msrs);
msrs = NULL;
unregister_cpu_notifier(&cpb_nb);
}
cpufreq_unregister_driver(&cpufreq_amd64_driver);
}

View File

@ -5,24 +5,11 @@
* http://www.gnu.org/licenses/gpl.html
*/
enum pstate {
HW_PSTATE_INVALID = 0xff,
HW_PSTATE_0 = 0,
HW_PSTATE_1 = 1,
HW_PSTATE_2 = 2,
HW_PSTATE_3 = 3,
HW_PSTATE_4 = 4,
HW_PSTATE_5 = 5,
HW_PSTATE_6 = 6,
HW_PSTATE_7 = 7,
};
struct powernow_k8_data {
unsigned int cpu;
u32 numps; /* number of p-states */
u32 batps; /* number of p-states supported on battery */
u32 max_hw_pstate; /* maximum legal hardware pstate */
/* these values are constant when the PSB is used to determine
* vid/fid pairings, but are modified during the ->target() call
@ -37,7 +24,6 @@ struct powernow_k8_data {
/* keep track of the current fid / vid or pstate */
u32 currvid;
u32 currfid;
enum pstate currpstate;
/* the powernow_table includes all frequency and vid/fid pairings:
* fid are the lower 8 bits of the index, vid are the upper 8 bits.
@ -97,23 +83,6 @@ struct powernow_k8_data {
#define MSR_S_HI_CURRENT_VID 0x0000003f
#define MSR_C_HI_STP_GNT_BENIGN 0x00000001
/* Hardware Pstate _PSS and MSR definitions */
#define USE_HW_PSTATE 0x00000080
#define HW_PSTATE_MASK 0x00000007
#define HW_PSTATE_VALID_MASK 0x80000000
#define HW_PSTATE_MAX_MASK 0x000000f0
#define HW_PSTATE_MAX_SHIFT 4
#define MSR_PSTATE_DEF_BASE 0xc0010064 /* base of Pstate MSRs */
#define MSR_PSTATE_STATUS 0xc0010063 /* Pstate Status MSR */
#define MSR_PSTATE_CTRL 0xc0010062 /* Pstate control MSR */
#define MSR_PSTATE_CUR_LIMIT 0xc0010061 /* pstate current limit MSR */
/* define the two driver architectures */
#define CPU_OPTERON 0
#define CPU_HW_PSTATE 1
/*
* There are restrictions frequencies have to follow:
* - only 1 entry in the low fid table ( <=1.4GHz )
@ -218,5 +187,4 @@ static int core_frequency_transition(struct powernow_k8_data *data, u32 reqfid);
static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned int index);
static int fill_powernow_table_pstate(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table);
static int fill_powernow_table_fidvid(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table);

View File

@ -48,6 +48,14 @@ int opp_disable(struct device *dev, unsigned long freq);
struct srcu_notifier_head *opp_get_notifier(struct device *dev);
#ifdef CONFIG_OF
int of_init_opp_table(struct device *dev);
#else
static inline int of_init_opp_table(struct device *dev)
{
return -EINVAL;
}
#endif /* CONFIG_OF */
#else
static inline unsigned long opp_get_voltage(struct opp *opp)
{