diff options
Diffstat (limited to 'drivers/cpufreq/intel_pstate.c')
-rw-r--r-- | drivers/cpufreq/intel_pstate.c | 148 |
1 files changed, 77 insertions, 71 deletions
diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index 0d159b513469..9ec033b4f2d9 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -35,6 +35,7 @@ #include <asm/msr.h> #include <asm/cpu_device_id.h> #include <asm/cpufeature.h> +#include <asm/intel-family.h> #define ATOM_RATIOS 0x66a #define ATOM_VIDS 0x66b @@ -96,7 +97,6 @@ static inline u64 div_ext_fp(u64 x, u64 y) * read from MPERF MSR between last and current sample * @tsc: Difference of time stamp counter between last and * current sample - * @freq: Effective frequency calculated from APERF/MPERF * @time: Current time from scheduler * * This structure is used in the cpudata structure to store performance sample @@ -108,7 +108,6 @@ struct sample { u64 aperf; u64 mperf; u64 tsc; - int freq; u64 time; }; @@ -281,9 +280,9 @@ struct cpu_defaults { static inline int32_t get_target_pstate_use_performance(struct cpudata *cpu); static inline int32_t get_target_pstate_use_cpu_load(struct cpudata *cpu); -static struct pstate_adjust_policy pid_params; -static struct pstate_funcs pstate_funcs; -static int hwp_active; +static struct pstate_adjust_policy pid_params __read_mostly; +static struct pstate_funcs pstate_funcs __read_mostly; +static int hwp_active __read_mostly; #ifdef CONFIG_ACPI static bool acpi_ppc; @@ -372,26 +371,9 @@ static bool intel_pstate_get_ppc_enable_status(void) return acpi_ppc; } -/* - * The max target pstate ratio is a 8 bit value in both PLATFORM_INFO MSR and - * in TURBO_RATIO_LIMIT MSR, which pstate driver stores in max_pstate and - * max_turbo_pstate fields. The PERF_CTL MSR contains 16 bit value for P state - * ratio, out of it only high 8 bits are used. For example 0x1700 is setting - * target ratio 0x17. The _PSS control value stores in a format which can be - * directly written to PERF_CTL MSR. But in intel_pstate driver this shift - * occurs during write to PERF_CTL (E.g. for cores core_set_pstate()). - * This function converts the _PSS control value to intel pstate driver format - * for comparison and assignment. - */ -static int convert_to_native_pstate_format(struct cpudata *cpu, int index) -{ - return cpu->acpi_perf_data.states[index].control >> 8; -} - static void intel_pstate_init_acpi_perf_limits(struct cpufreq_policy *policy) { struct cpudata *cpu; - int turbo_pss_ctl; int ret; int i; @@ -441,11 +423,10 @@ static void intel_pstate_init_acpi_perf_limits(struct cpufreq_policy *policy) * max frequency, which will cause a reduced performance as * this driver uses real max turbo frequency as the max * frequency. So correct this frequency in _PSS table to - * correct max turbo frequency based on the turbo ratio. + * correct max turbo frequency based on the turbo state. * Also need to convert to MHz as _PSS freq is in MHz. */ - turbo_pss_ctl = convert_to_native_pstate_format(cpu, 0); - if (turbo_pss_ctl > cpu->pstate.max_pstate) + if (!limits->turbo_disabled) cpu->acpi_perf_data.states[0].core_frequency = policy->cpuinfo.max_freq / 1000; cpu->valid_pss_table = true; @@ -825,7 +806,8 @@ static void __init intel_pstate_sysfs_expose_params(void) static void intel_pstate_hwp_enable(struct cpudata *cpudata) { /* First disable HWP notification interrupt as we don't process them */ - wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_INTERRUPT, 0x00); + if (static_cpu_has(X86_FEATURE_HWP_NOTIFY)) + wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_INTERRUPT, 0x00); wrmsrl_on_cpu(cpudata->cpu, MSR_PM_ENABLE, 0x1); } @@ -962,7 +944,7 @@ static int core_get_max_pstate(void) if (err) goto skip_tar; - tdp_msr = MSR_CONFIG_TDP_NOMINAL + tdp_ctrl; + tdp_msr = MSR_CONFIG_TDP_NOMINAL + (tdp_ctrl & 0x3); err = rdmsrl_safe(tdp_msr, &tdp_ratio); if (err) goto skip_tar; @@ -990,7 +972,7 @@ static int core_get_turbo_pstate(void) u64 value; int nont, ret; - rdmsrl(MSR_NHM_TURBO_RATIO_LIMIT, value); + rdmsrl(MSR_TURBO_RATIO_LIMIT, value); nont = core_get_max_pstate(); ret = (value) & 255; if (ret <= nont) @@ -1019,7 +1001,7 @@ static int knl_get_turbo_pstate(void) u64 value; int nont, ret; - rdmsrl(MSR_NHM_TURBO_RATIO_LIMIT, value); + rdmsrl(MSR_TURBO_RATIO_LIMIT, value); nont = core_get_max_pstate(); ret = (((value) >> 8) & 0xFF); if (ret <= nont) @@ -1109,6 +1091,26 @@ static struct cpu_defaults knl_params = { }, }; +static struct cpu_defaults bxt_params = { + .pid_policy = { + .sample_rate_ms = 10, + .deadband = 0, + .setpoint = 60, + .p_gain_pct = 14, + .d_gain_pct = 0, + .i_gain_pct = 4, + }, + .funcs = { + .get_max = core_get_max_pstate, + .get_max_physical = core_get_max_pstate_physical, + .get_min = core_get_min_pstate, + .get_turbo = core_get_turbo_pstate, + .get_scaling = core_get_scaling, + .get_val = core_get_val, + .get_target_pstate = get_target_pstate_use_cpu_load, + }, +}; + static void intel_pstate_get_min_max(struct cpudata *cpu, int *min, int *max) { int max_perf = cpu->pstate.turbo_pstate; @@ -1131,17 +1133,12 @@ static void intel_pstate_get_min_max(struct cpudata *cpu, int *min, int *max) *min = clamp_t(int, min_perf, cpu->pstate.min_pstate, max_perf); } -static inline void intel_pstate_record_pstate(struct cpudata *cpu, int pstate) -{ - trace_cpu_frequency(pstate * cpu->pstate.scaling, cpu->cpu); - cpu->pstate.current_pstate = pstate; -} - static void intel_pstate_set_min_pstate(struct cpudata *cpu) { int pstate = cpu->pstate.min_pstate; - intel_pstate_record_pstate(cpu, pstate); + trace_cpu_frequency(pstate * cpu->pstate.scaling, cpu->cpu); + cpu->pstate.current_pstate = pstate; /* * Generally, there is no guarantee that this code will always run on * the CPU being updated, so force the register update to run on the @@ -1301,10 +1298,11 @@ static inline void intel_pstate_update_pstate(struct cpudata *cpu, int pstate) intel_pstate_get_min_max(cpu, &min_perf, &max_perf); pstate = clamp_t(int, pstate, min_perf, max_perf); + trace_cpu_frequency(pstate * cpu->pstate.scaling, cpu->cpu); if (pstate == cpu->pstate.current_pstate) return; - intel_pstate_record_pstate(cpu, pstate); + cpu->pstate.current_pstate = pstate; wrmsrl(MSR_IA32_PERF_CTL, pstate_funcs.get_val(cpu, pstate)); } @@ -1352,29 +1350,30 @@ static void intel_pstate_update_util(struct update_util_data *data, u64 time, (unsigned long)&policy } static const struct x86_cpu_id intel_pstate_cpu_ids[] = { - ICPU(0x2a, core_params), - ICPU(0x2d, core_params), - ICPU(0x37, silvermont_params), - ICPU(0x3a, core_params), - ICPU(0x3c, core_params), - ICPU(0x3d, core_params), - ICPU(0x3e, core_params), - ICPU(0x3f, core_params), - ICPU(0x45, core_params), - ICPU(0x46, core_params), - ICPU(0x47, core_params), - ICPU(0x4c, airmont_params), - ICPU(0x4e, core_params), - ICPU(0x4f, core_params), - ICPU(0x5e, core_params), - ICPU(0x56, core_params), - ICPU(0x57, knl_params), + ICPU(INTEL_FAM6_SANDYBRIDGE, core_params), + ICPU(INTEL_FAM6_SANDYBRIDGE_X, core_params), + ICPU(INTEL_FAM6_ATOM_SILVERMONT1, silvermont_params), + ICPU(INTEL_FAM6_IVYBRIDGE, core_params), + ICPU(INTEL_FAM6_HASWELL_CORE, core_params), + ICPU(INTEL_FAM6_BROADWELL_CORE, core_params), + ICPU(INTEL_FAM6_IVYBRIDGE_X, core_params), + ICPU(INTEL_FAM6_HASWELL_X, core_params), + ICPU(INTEL_FAM6_HASWELL_ULT, core_params), + ICPU(INTEL_FAM6_HASWELL_GT3E, core_params), + ICPU(INTEL_FAM6_BROADWELL_GT3E, core_params), + ICPU(INTEL_FAM6_ATOM_AIRMONT, airmont_params), + ICPU(INTEL_FAM6_SKYLAKE_MOBILE, core_params), + ICPU(INTEL_FAM6_BROADWELL_X, core_params), + ICPU(INTEL_FAM6_SKYLAKE_DESKTOP, core_params), + ICPU(INTEL_FAM6_BROADWELL_XEON_D, core_params), + ICPU(INTEL_FAM6_XEON_PHI_KNL, knl_params), + ICPU(INTEL_FAM6_ATOM_GOLDMONT, bxt_params), {} }; MODULE_DEVICE_TABLE(x86cpu, intel_pstate_cpu_ids); -static const struct x86_cpu_id intel_pstate_cpu_oob_ids[] = { - ICPU(0x56, core_params), +static const struct x86_cpu_id intel_pstate_cpu_oob_ids[] __initconst = { + ICPU(INTEL_FAM6_BROADWELL_XEON_D, core_params), {} }; @@ -1418,6 +1417,9 @@ static void intel_pstate_set_update_util_hook(unsigned int cpu_num) { struct cpudata *cpu = all_cpu_data[cpu_num]; + if (cpu->update_util_set) + return; + /* Prevent intel_pstate_update_util() from using stale data. */ cpu->sample.time = 0; cpufreq_add_update_util_hook(cpu_num, &cpu->update_util, @@ -1458,7 +1460,8 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy) if (!policy->cpuinfo.max_freq) return -ENODEV; - intel_pstate_clear_update_util_hook(policy->cpu); + pr_debug("set_policy cpuinfo.max %u policy->max %u\n", + policy->cpuinfo.max_freq, policy->max); cpu = all_cpu_data[0]; if (cpu->pstate.max_pstate_physical > cpu->pstate.max_pstate && @@ -1495,13 +1498,13 @@ static int intel_pstate_set_policy(struct cpufreq_policy *policy) limits->max_sysfs_pct); limits->max_perf_pct = max(limits->min_policy_pct, limits->max_perf_pct); - limits->max_perf = round_up(limits->max_perf, FRAC_BITS); /* Make sure min_perf_pct <= max_perf_pct */ limits->min_perf_pct = min(limits->max_perf_pct, limits->min_perf_pct); limits->min_perf = div_fp(limits->min_perf_pct, 100); limits->max_perf = div_fp(limits->max_perf_pct, 100); + limits->max_perf = round_up(limits->max_perf, FRAC_BITS); out: intel_pstate_set_update_util_hook(policy->cpu); @@ -1558,8 +1561,11 @@ static int intel_pstate_cpu_init(struct cpufreq_policy *policy) /* cpuinfo and default policy values */ policy->cpuinfo.min_freq = cpu->pstate.min_pstate * cpu->pstate.scaling; - policy->cpuinfo.max_freq = - cpu->pstate.turbo_pstate * cpu->pstate.scaling; + update_turbo_state(); + policy->cpuinfo.max_freq = limits->turbo_disabled ? + cpu->pstate.max_pstate : cpu->pstate.turbo_pstate; + policy->cpuinfo.max_freq *= cpu->pstate.scaling; + intel_pstate_init_acpi_perf_limits(policy); policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL; cpumask_set_cpu(policy->cpu, policy->cpus); @@ -1586,12 +1592,12 @@ static struct cpufreq_driver intel_pstate_driver = { .name = "intel_pstate", }; -static int __initdata no_load; -static int __initdata no_hwp; -static int __initdata hwp_only; -static unsigned int force_load; +static int no_load __initdata; +static int no_hwp __initdata; +static int hwp_only __initdata; +static unsigned int force_load __initdata; -static int intel_pstate_msrs_not_valid(void) +static int __init intel_pstate_msrs_not_valid(void) { if (!pstate_funcs.get_max() || !pstate_funcs.get_min() || @@ -1601,7 +1607,7 @@ static int intel_pstate_msrs_not_valid(void) return 0; } -static void copy_pid_params(struct pstate_adjust_policy *policy) +static void __init copy_pid_params(struct pstate_adjust_policy *policy) { pid_params.sample_rate_ms = policy->sample_rate_ms; pid_params.sample_rate_ns = pid_params.sample_rate_ms * NSEC_PER_MSEC; @@ -1612,7 +1618,7 @@ static void copy_pid_params(struct pstate_adjust_policy *policy) pid_params.setpoint = policy->setpoint; } -static void copy_cpu_funcs(struct pstate_funcs *funcs) +static void __init copy_cpu_funcs(struct pstate_funcs *funcs) { pstate_funcs.get_max = funcs->get_max; pstate_funcs.get_max_physical = funcs->get_max_physical; @@ -1627,7 +1633,7 @@ static void copy_cpu_funcs(struct pstate_funcs *funcs) #ifdef CONFIG_ACPI -static bool intel_pstate_no_acpi_pss(void) +static bool __init intel_pstate_no_acpi_pss(void) { int i; @@ -1656,7 +1662,7 @@ static bool intel_pstate_no_acpi_pss(void) return true; } -static bool intel_pstate_has_acpi_ppc(void) +static bool __init intel_pstate_has_acpi_ppc(void) { int i; @@ -1684,7 +1690,7 @@ struct hw_vendor_info { }; /* Hardware vendor-specific info that has its own power management modes */ -static struct hw_vendor_info vendor_info[] = { +static struct hw_vendor_info vendor_info[] __initdata = { {1, "HP ", "ProLiant", PSS}, {1, "ORACLE", "X4-2 ", PPC}, {1, "ORACLE", "X4-2L ", PPC}, @@ -1703,7 +1709,7 @@ static struct hw_vendor_info vendor_info[] = { {0, "", ""}, }; -static bool intel_pstate_platform_pwr_mgmt_exists(void) +static bool __init intel_pstate_platform_pwr_mgmt_exists(void) { struct acpi_table_header hdr; struct hw_vendor_info *v_info; |