#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
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
{
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;
}
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");
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)
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);
"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;
}
-
- *rc = 0;
}
static int check_pst_table(struct powernow_k8_data *data, struct pst_s *pst,
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,
data->powernow_table[j].frequency/1000,
data->powernow_table[j].index >> 8);
- }
}
}
if (data->batps)
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)
{
{
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;
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);
+ ret_val = fill_powernow_table_fidvid(data, powernow_table);
if (ret_val)
goto err_out_mem;
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)
{
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 */
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)
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);
+ 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",
- checkfid, data->currfid,
- checkvid, data->currvid);
- }
+ if ((checkvid != data->currvid) ||
+ (checkfid != data->currfid)) {
+ 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,
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);
+ ret = transition_frequency_fidvid(data, newstate);
+
if (ret) {
printk(KERN_ERR PFX "transition frequency failed\n");
ret = 1;
}
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);
+ pol->cur = find_khz_freq_from_fid(data->currfid);
ret = 0;
err_out:
return;
}
- if (cpu_family == CPU_OPTERON)
- fidvid_msr_init();
+ fidvid_msr_init();
init_on_cpu->rc = 0;
}
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;
}
data->cpu = pol->cpu;
- data->currpstate = HW_PSTATE_INVALID;
if (powernow_k8_cpu_init_acpi(data)) {
/*
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));
+ 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);
+ pol->cur = find_khz_freq_from_fid(data->currfid);
pr_debug("policy current frequency %d kHz\n", pol->cur);
/* min/max the cpu is capable of */
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);
+ pr_debug("cpu_init done, current fid 0x%x, vid 0x%x\n",
+ data->currfid, data->currvid);
per_cpu(powernow_data, pol->cpu) = data;
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);
+ khz = find_khz_freq_from_fid(data->currfid);
out:
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,
};
.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 (!x86_match_cpu(powernow_k8_ids))
+ 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 (static_cpu_has(X86_FEATURE_HW_PSTATE))
- pr_warn(PFX "support for this CPU is deprecated, use acpi-cpufreq instead.\n");
+ if (!x86_match_cpu(powernow_k8_ids))
+ return -ENODEV;
for_each_online_cpu(i) {
int rc;
if (supported_cpus != num_online_cpus())
return -ENODEV;
- 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)));
- }
- }
-
rv = cpufreq_register_driver(&cpufreq_amd64_driver);
if (!rv)
num_online_nodes(), boot_cpu_data.x86_model_id,
supported_cpus);
- if (boot_cpu_has(X86_FEATURE_CPB)) {
- if (rv < 0) {
- unregister_cpu_notifier(&cpb_nb);
- msrs_free(msrs);
- msrs = NULL;
- } else
- pr_info(PFX "Core Performance Boosting: %s.\n",
- (cpb_enabled ? "on" : "off"));
- }
return rv;
}
{
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);
}
git.openpandora.org / pandora-kernel.git / commitdiff