#include <linux/cpufreq.h>
#include <linux/compiler.h>
#include <linux/dmi.h>
-#include <trace/power.h>
+#include <linux/slab.h>
+#include <trace/events/power.h>
#include <linux/acpi.h>
#include <linux/io.h>
};
#define INTEL_MSR_RANGE (0xffff)
-#define CPUID_6_ECX_APERFMPERF_CAPABILITY (0x1)
struct acpi_cpufreq_data {
struct acpi_processor_performance *acpi_data;
struct cpufreq_frequency_table *freq_table;
- unsigned int max_freq;
unsigned int resume;
unsigned int cpu_feature;
};
-static DEFINE_PER_CPU(struct acpi_cpufreq_data *, drv_data);
+static DEFINE_PER_CPU(struct acpi_cpufreq_data *, acfreq_data);
-DEFINE_TRACE(power_mark);
+static DEFINE_PER_CPU(struct aperfmperf, acfreq_old_perf);
/* acpi_perf_data is a pointer to percpu data. */
static struct acpi_processor_performance *acpi_perf_data;
{
struct cpuinfo_x86 *cpu = &cpu_data(cpuid);
- if (cpu->x86_vendor != X86_VENDOR_INTEL ||
- !cpu_has(cpu, X86_FEATURE_EST))
- return 0;
-
- return 1;
+ return cpu_has(cpu, X86_FEATURE_EST);
}
static unsigned extract_io(u32 value, struct acpi_cpufreq_data *data)
u32 val;
};
-static long do_drv_read(void *_cmd)
+/* Called via smp_call_function_single(), on the target CPU */
+static void do_drv_read(void *_cmd)
{
struct drv_cmd *cmd = _cmd;
u32 h;
default:
break;
}
- return 0;
}
-static long do_drv_write(void *_cmd)
+/* Called via smp_call_function_many(), on the target CPUs */
+static void do_drv_write(void *_cmd)
{
struct drv_cmd *cmd = _cmd;
u32 lo, hi;
default:
break;
}
- return 0;
}
static void drv_read(struct drv_cmd *cmd)
{
+ int err;
cmd->val = 0;
- work_on_cpu(cpumask_any(cmd->mask), do_drv_read, cmd);
+ err = smp_call_function_any(cmd->mask, do_drv_read, cmd, 1);
+ WARN_ON_ONCE(err); /* smp_call_function_any() was buggy? */
}
static void drv_write(struct drv_cmd *cmd)
{
- unsigned int i;
+ int this_cpu;
- for_each_cpu(i, cmd->mask) {
- work_on_cpu(i, do_drv_write, cmd);
- }
+ this_cpu = get_cpu();
+ if (cpumask_test_cpu(this_cpu, cmd->mask))
+ do_drv_write(cmd);
+ smp_call_function_many(cmd->mask, do_drv_write, cmd, 1);
+ put_cpu();
}
static u32 get_cur_val(const struct cpumask *mask)
if (unlikely(cpumask_empty(mask)))
return 0;
- switch (per_cpu(drv_data, cpumask_first(mask))->cpu_feature) {
+ switch (per_cpu(acfreq_data, cpumask_first(mask))->cpu_feature) {
case SYSTEM_INTEL_MSR_CAPABLE:
cmd.type = SYSTEM_INTEL_MSR_CAPABLE;
cmd.addr.msr.reg = MSR_IA32_PERF_STATUS;
break;
case SYSTEM_IO_CAPABLE:
cmd.type = SYSTEM_IO_CAPABLE;
- perf = per_cpu(drv_data, cpumask_first(mask))->acpi_data;
+ perf = per_cpu(acfreq_data, cpumask_first(mask))->acpi_data;
cmd.addr.io.port = perf->control_register.address;
cmd.addr.io.bit_width = perf->control_register.bit_width;
break;
return cmd.val;
}
-struct perf_cur {
- union {
- struct {
- u32 lo;
- u32 hi;
- } split;
- u64 whole;
- } aperf_cur, mperf_cur;
-};
-
-
-static long read_measured_perf_ctrs(void *_cur)
+/* Called via smp_call_function_single(), on the target CPU */
+static void read_measured_perf_ctrs(void *_cur)
{
- struct perf_cur *cur = _cur;
-
- rdmsr(MSR_IA32_APERF, cur->aperf_cur.split.lo, cur->aperf_cur.split.hi);
- rdmsr(MSR_IA32_MPERF, cur->mperf_cur.split.lo, cur->mperf_cur.split.hi);
+ struct aperfmperf *am = _cur;
- wrmsr(MSR_IA32_APERF, 0, 0);
- wrmsr(MSR_IA32_MPERF, 0, 0);
-
- return 0;
+ get_aperfmperf(am);
}
/*
static unsigned int get_measured_perf(struct cpufreq_policy *policy,
unsigned int cpu)
{
- struct perf_cur cur;
- unsigned int perf_percent;
+ struct aperfmperf perf;
+ unsigned long ratio;
unsigned int retval;
- if (!work_on_cpu(cpu, read_measured_perf_ctrs, &cur))
+ if (smp_call_function_single(cpu, read_measured_perf_ctrs, &perf, 1))
return 0;
-#ifdef __i386__
- /*
- * We dont want to do 64 bit divide with 32 bit kernel
- * Get an approximate value. Return failure in case we cannot get
- * an approximate value.
- */
- if (unlikely(cur.aperf_cur.split.hi || cur.mperf_cur.split.hi)) {
- int shift_count;
- u32 h;
-
- h = max_t(u32, cur.aperf_cur.split.hi, cur.mperf_cur.split.hi);
- shift_count = fls(h);
-
- cur.aperf_cur.whole >>= shift_count;
- cur.mperf_cur.whole >>= shift_count;
- }
-
- if (((unsigned long)(-1) / 100) < cur.aperf_cur.split.lo) {
- int shift_count = 7;
- cur.aperf_cur.split.lo >>= shift_count;
- cur.mperf_cur.split.lo >>= shift_count;
- }
-
- if (cur.aperf_cur.split.lo && cur.mperf_cur.split.lo)
- perf_percent = (cur.aperf_cur.split.lo * 100) /
- cur.mperf_cur.split.lo;
- else
- perf_percent = 0;
-
-#else
- if (unlikely(((unsigned long)(-1) / 100) < cur.aperf_cur.whole)) {
- int shift_count = 7;
- cur.aperf_cur.whole >>= shift_count;
- cur.mperf_cur.whole >>= shift_count;
- }
+ ratio = calc_aperfmperf_ratio(&per_cpu(acfreq_old_perf, cpu), &perf);
+ per_cpu(acfreq_old_perf, cpu) = perf;
- if (cur.aperf_cur.whole && cur.mperf_cur.whole)
- perf_percent = (cur.aperf_cur.whole * 100) /
- cur.mperf_cur.whole;
- else
- perf_percent = 0;
-
-#endif
-
- retval = per_cpu(drv_data, policy->cpu)->max_freq * perf_percent / 100;
+ retval = (policy->cpuinfo.max_freq * ratio) >> APERFMPERF_SHIFT;
return retval;
}
static unsigned int get_cur_freq_on_cpu(unsigned int cpu)
{
- struct acpi_cpufreq_data *data = per_cpu(drv_data, cpu);
+ struct acpi_cpufreq_data *data = per_cpu(acfreq_data, cpu);
unsigned int freq;
unsigned int cached_freq;
static int acpi_cpufreq_target(struct cpufreq_policy *policy,
unsigned int target_freq, unsigned int relation)
{
- struct acpi_cpufreq_data *data = per_cpu(drv_data, policy->cpu);
+ struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu);
struct acpi_processor_performance *perf;
struct cpufreq_freqs freqs;
struct drv_cmd cmd;
unsigned int next_perf_state = 0; /* Index into perf table */
unsigned int i;
int result = 0;
- struct power_trace it;
dprintk("acpi_cpufreq_target %d (%d)\n", target_freq, policy->cpu);
}
}
- trace_power_mark(&it, POWER_PSTATE, next_perf_state);
+ trace_power_frequency(POWER_PSTATE, data->freq_table[next_state].frequency);
switch (data->cpu_feature) {
case SYSTEM_INTEL_MSR_CAPABLE:
static int acpi_cpufreq_verify(struct cpufreq_policy *policy)
{
- struct acpi_cpufreq_data *data = per_cpu(drv_data, policy->cpu);
+ struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu);
dprintk("acpi_cpufreq_verify\n");
return -ENOMEM;
}
for_each_possible_cpu(i) {
- if (!alloc_cpumask_var_node(
+ if (!zalloc_cpumask_var_node(
&per_cpu_ptr(acpi_perf_data, i)->shared_cpu_map,
GFP_KERNEL, cpu_to_node(i))) {
},
{ }
};
+
+static int acpi_cpufreq_blacklist(struct cpuinfo_x86 *c)
+{
+ /* Intel Xeon Processor 7100 Series Specification Update
+ * http://www.intel.com/Assets/PDF/specupdate/314554.pdf
+ * AL30: A Machine Check Exception (MCE) Occurring during an
+ * Enhanced Intel SpeedStep Technology Ratio Change May Cause
+ * Both Processor Cores to Lock Up. */
+ if (c->x86_vendor == X86_VENDOR_INTEL) {
+ if ((c->x86 == 15) &&
+ (c->x86_model == 6) &&
+ (c->x86_mask == 8)) {
+ printk(KERN_INFO "acpi-cpufreq: Intel(R) "
+ "Xeon(R) 7100 Errata AL30, processors may "
+ "lock up on frequency changes: disabling "
+ "acpi-cpufreq.\n");
+ return -ENODEV;
+ }
+ }
+ return 0;
+}
#endif
static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy)
unsigned int result = 0;
struct cpuinfo_x86 *c = &cpu_data(policy->cpu);
struct acpi_processor_performance *perf;
+#ifdef CONFIG_SMP
+ static int blacklisted;
+#endif
dprintk("acpi_cpufreq_cpu_init\n");
+#ifdef CONFIG_SMP
+ if (blacklisted)
+ return blacklisted;
+ blacklisted = acpi_cpufreq_blacklist(c);
+ if (blacklisted)
+ return blacklisted;
+#endif
+
data = kzalloc(sizeof(struct acpi_cpufreq_data), GFP_KERNEL);
if (!data)
return -ENOMEM;
data->acpi_data = per_cpu_ptr(acpi_perf_data, cpu);
- per_cpu(drv_data, cpu) = data;
+ per_cpu(acfreq_data, cpu) = data;
if (cpu_has(c, X86_FEATURE_CONSTANT_TSC))
acpi_cpufreq_driver.flags |= CPUFREQ_CONST_LOOPS;
perf->states[i].transition_latency * 1000;
}
- data->max_freq = perf->states[0].core_frequency * 1000;
+ /* Check for high latency (>20uS) from buggy BIOSes, like on T42 */
+ if (perf->control_register.space_id == ACPI_ADR_SPACE_FIXED_HARDWARE &&
+ policy->cpuinfo.transition_latency > 20 * 1000) {
+ policy->cpuinfo.transition_latency = 20 * 1000;
+ printk_once(KERN_INFO
+ "P-state transition latency capped at 20 uS\n");
+ }
+
/* table init */
for (i = 0; i < perf->state_count; i++) {
if (i > 0 && perf->states[i].core_frequency >=
if (result)
goto err_freqfree;
+ if (perf->states[0].core_frequency * 1000 != policy->cpuinfo.max_freq)
+ printk(KERN_WARNING FW_WARN "P-state 0 is not max freq\n");
+
switch (perf->control_register.space_id) {
case ACPI_ADR_SPACE_SYSTEM_IO:
/* Current speed is unknown and not detectable by IO port */
acpi_processor_notify_smm(THIS_MODULE);
/* Check for APERF/MPERF support in hardware */
- if (c->x86_vendor == X86_VENDOR_INTEL && c->cpuid_level >= 6) {
- unsigned int ecx;
- ecx = cpuid_ecx(6);
- if (ecx & CPUID_6_ECX_APERFMPERF_CAPABILITY)
- acpi_cpufreq_driver.getavg = get_measured_perf;
- }
+ if (cpu_has(c, X86_FEATURE_APERFMPERF))
+ acpi_cpufreq_driver.getavg = get_measured_perf;
dprintk("CPU%u - ACPI performance management activated.\n", cpu);
for (i = 0; i < perf->state_count; i++)
acpi_processor_unregister_performance(perf, cpu);
err_free:
kfree(data);
- per_cpu(drv_data, cpu) = NULL;
+ per_cpu(acfreq_data, cpu) = NULL;
return result;
}
static int acpi_cpufreq_cpu_exit(struct cpufreq_policy *policy)
{
- struct acpi_cpufreq_data *data = per_cpu(drv_data, policy->cpu);
+ struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu);
dprintk("acpi_cpufreq_cpu_exit\n");
if (data) {
cpufreq_frequency_table_put_attr(policy->cpu);
- per_cpu(drv_data, policy->cpu) = NULL;
+ per_cpu(acfreq_data, policy->cpu) = NULL;
acpi_processor_unregister_performance(data->acpi_data,
policy->cpu);
kfree(data);
static int acpi_cpufreq_resume(struct cpufreq_policy *policy)
{
- struct acpi_cpufreq_data *data = per_cpu(drv_data, policy->cpu);
+ struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu);
dprintk("acpi_cpufreq_resume\n");
};
static struct cpufreq_driver acpi_cpufreq_driver = {
- .verify = acpi_cpufreq_verify,
- .target = acpi_cpufreq_target,
- .init = acpi_cpufreq_cpu_init,
- .exit = acpi_cpufreq_cpu_exit,
- .resume = acpi_cpufreq_resume,
- .name = "acpi-cpufreq",
- .owner = THIS_MODULE,
- .attr = acpi_cpufreq_attr,
+ .verify = acpi_cpufreq_verify,
+ .target = acpi_cpufreq_target,
+ .bios_limit = acpi_processor_get_bios_limit,
+ .init = acpi_cpufreq_cpu_init,
+ .exit = acpi_cpufreq_cpu_exit,
+ .resume = acpi_cpufreq_resume,
+ .name = "acpi-cpufreq",
+ .owner = THIS_MODULE,
+ .attr = acpi_cpufreq_attr,
};
static int __init acpi_cpufreq_init(void)