diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2024-09-16 07:47:50 +0200 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2024-09-16 07:47:50 +0200 |
| commit | 02824a5fd11f99b4637668926a59aab3698b46a9 (patch) | |
| tree | 66fa92cf244cc9a6be16486b0820802f656dd65f | |
| parent | 11b3125073d16929403d3aa7b2ae6a482060a937 (diff) | |
| parent | 0a06811d664b8695a7612d3e59c1defb4382f4e0 (diff) | |
Merge tag 'pm-6.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull power management updates from Rafael Wysocki:
"By the number of new lines of code, the most visible change here is
the addition of hybrid CPU capacity scaling support to the
intel_pstate driver. Next are the amd-pstate driver changes related to
the calculation of the AMD boost numerator and preferred core
detection.
As far as new hardware support is concerned, the intel_idle driver
will now handle Granite Rapids Xeon processors natively, the
intel_rapl power capping driver will recognize family 1Ah of AMD
processors and Intel ArrowLake-U chipos, and intel_pstate will handle
Granite Rapids and Sierra Forest chips in the out-of-band (OOB) mode.
Apart from the above, there is a usual collection of assorted fixes
and code cleanups in many places and there are tooling updates.
Specifics:
- Remove LATENCY_MULTIPLIER from cpufreq (Qais Yousef)
- Add support for Granite Rapids and Sierra Forest in OOB mode to the
intel_pstate cpufreq driver (Srinivas Pandruvada)
- Add basic support for CPU capacity scaling on x86 and make the
intel_pstate driver set asymmetric CPU capacity on hybrid systems
without SMT (Rafael Wysocki)
- Add missing MODULE_DESCRIPTION() macros to the powerpc cpufreq
driver (Jeff Johnson)
- Several OF related cleanups in cpufreq drivers (Rob Herring)
- Enable COMPILE_TEST for ARM drivers (Rob Herrring)
- Introduce quirks for syscon failures and use socinfo to get
revision for TI cpufreq driver (Dhruva Gole, Nishanth Menon)
- Minor cleanups in amd-pstate driver (Anastasia Belova, Dhananjay
Ugwekar)
- Minor cleanups for loongson, cpufreq-dt and powernv cpufreq drivers
(Danila Tikhonov, Huacai Chen, and Liu Jing)
- Make amd-pstate validate return of any attempt to update EPP
limits, which fixes the masking hardware problems (Mario
Limonciello)
- Move the calculation of the AMD boost numerator outside of
amd-pstate, correcting acpi-cpufreq on systems with preferred cores
(Mario Limonciello)
- Harden preferred core detection in amd-pstate to avoid potential
false positives (Mario Limonciello)
- Add extra unit test coverage for mode state machine (Mario
Limonciello)
- Fix an "Uninitialized variables" issue in amd-pstste (Qianqiang
Liu)
- Add Granite Rapids Xeon support to intel_idle (Artem Bityutskiy)
- Disable promotion to C1E on Jasper Lake and Elkhart Lake in
intel_idle (Kai-Heng Feng)
- Use scoped device node handling to fix missing of_node_put() and
simplify walking OF children in the riscv-sbi cpuidle driver
(Krzysztof Kozlowski)
- Remove dead code from cpuidle_enter_state() (Dhruva Gole)
- Change an error pointer to NULL to fix error handling in the
intel_rapl power capping driver (Dan Carpenter)
- Fix off by one in get_rpi() in the intel_rapl power capping driver
(Dan Carpenter)
- Add support for ArrowLake-U to the intel_rapl power capping driver
(Sumeet Pawnikar)
- Fix the energy-pkg event for AMD CPUs in the intel_rapl power
capping driver (Dhananjay Ugwekar)
- Add support for AMD family 1Ah processors to the intel_rapl power
capping driver (Dhananjay Ugwekar)
- Remove unused stub for saveable_highmem_page() and remove
deprecated macros from power management documentation (Andy
Shevchenko)
- Use ysfs_emit() and sysfs_emit_at() in "show" functions in the PM
sysfs interface (Xueqin Luo)
- Update the maintainers information for the
operating-points-v2-ti-cpu DT binding (Dhruva Gole)
- Drop unnecessary of_match_ptr() from ti-opp-supply (Rob Herring)
- Add missing MODULE_DESCRIPTION() macros to devfreq governors (Jeff
Johnson)
- Use devm_clk_get_enabled() in the exynos-bus devfreq driver (Anand
Moon)
- Use of_property_present() instead of of_get_property() in the
imx-bus devfreq driver (Rob Herring)
- Update directory handling and installation process in the pm-graph
Makefile and add .gitignore to ignore sleepgraph.py artifacts to
pm-graph (Amit Vadhavana, Yo-Jung Lin)
- Make cpupower display residency value in idle-info (Aboorva
Devarajan)
- Add missing powercap_set_enabled() stub function to cpupower (John
B. Wyatt IV)
- Add SWIG support to cpupower (John B. Wyatt IV)"
* tag 'pm-6.12-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (62 commits)
cpufreq/amd-pstate-ut: Fix an "Uninitialized variables" issue
cpufreq/amd-pstate-ut: Add test case for mode switches
cpufreq/amd-pstate: Export symbols for changing modes
amd-pstate: Add missing documentation for `amd_pstate_prefcore_ranking`
cpufreq: amd-pstate: Add documentation for `amd_pstate_hw_prefcore`
cpufreq: amd-pstate: Optimize amd_pstate_update_limits()
cpufreq: amd-pstate: Merge amd_pstate_highest_perf_set() into amd_get_boost_ratio_numerator()
x86/amd: Detect preferred cores in amd_get_boost_ratio_numerator()
x86/amd: Move amd_get_highest_perf() out of amd-pstate
ACPI: CPPC: Adjust debug messages in amd_set_max_freq_ratio() to warn
ACPI: CPPC: Drop check for non zero perf ratio
x86/amd: Rename amd_get_highest_perf() to amd_get_boost_ratio_numerator()
ACPI: CPPC: Adjust return code for inline functions in !CONFIG_ACPI_CPPC_LIB
x86/amd: Move amd_get_highest_perf() from amd.c to cppc.c
PM: hibernate: Remove unused stub for saveable_highmem_page()
pm:cpupower: Add error warning when SWIG is not installed
MAINTAINERS: Add Maintainers for SWIG Python bindings
pm:cpupower: Include test_raw_pylibcpupower.py
pm:cpupower: Add SWIG bindings files for libcpupower
pm:cpupower: Add missing powercap_set_enabled() stub function
...
63 files changed, 1416 insertions, 407 deletions
diff --git a/Documentation/admin-guide/pm/amd-pstate.rst b/Documentation/admin-guide/pm/amd-pstate.rst index d0324d44f548..210a808b74ec 100644 --- a/Documentation/admin-guide/pm/amd-pstate.rst +++ b/Documentation/admin-guide/pm/amd-pstate.rst @@ -251,7 +251,9 @@ performance supported in `AMD CPPC Performance Capability <perf_cap_>`_). In some ASICs, the highest CPPC performance is not the one in the ``_CPC`` table, so we need to expose it to sysfs. If boost is not active, but still supported, this maximum frequency will be larger than the one in -``cpuinfo``. +``cpuinfo``. On systems that support preferred core, the driver will have +different values for some cores than others and this will reflect the values +advertised by the platform at bootup. This attribute is read-only. ``amd_pstate_lowest_nonlinear_freq`` @@ -262,6 +264,17 @@ lowest non-linear performance in `AMD CPPC Performance Capability <perf_cap_>`_.) This attribute is read-only. +``amd_pstate_hw_prefcore`` + +Whether the platform supports the preferred core feature and it has been +enabled. This attribute is read-only. + +``amd_pstate_prefcore_ranking`` + +The performance ranking of the core. This number doesn't have any unit, but +larger numbers are preferred at the time of reading. This can change at +runtime based on platform conditions. This attribute is read-only. + ``energy_performance_available_preferences`` A list of all the supported EPP preferences that could be used for diff --git a/Documentation/devicetree/bindings/opp/operating-points-v2-ti-cpu.yaml b/Documentation/devicetree/bindings/opp/operating-points-v2-ti-cpu.yaml index 02d1d2c17129..fd0c8d5c5f3e 100644 --- a/Documentation/devicetree/bindings/opp/operating-points-v2-ti-cpu.yaml +++ b/Documentation/devicetree/bindings/opp/operating-points-v2-ti-cpu.yaml @@ -19,7 +19,7 @@ description: the hardware description for the scheme mentioned above. maintainers: - - Nishanth Menon <nm@ti.com> + - Dhruva Gole <d-gole@ti.com> allOf: - $ref: opp-v2-base.yaml# diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index e2c1fb8a569a..9ebecb7b00b2 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -979,18 +979,17 @@ subsections can be defined as a separate function, it often is convenient to point two or more members of struct dev_pm_ops to the same routine. There are a few convenience macros that can be used for this purpose. -The SIMPLE_DEV_PM_OPS macro declares a struct dev_pm_ops object with one +The DEFINE_SIMPLE_DEV_PM_OPS() declares a struct dev_pm_ops object with one suspend routine pointed to by the .suspend(), .freeze(), and .poweroff() members and one resume routine pointed to by the .resume(), .thaw(), and .restore() members. The other function pointers in this struct dev_pm_ops are unset. -The UNIVERSAL_DEV_PM_OPS macro is similar to SIMPLE_DEV_PM_OPS, but it -additionally sets the .runtime_resume() pointer to the same value as -.resume() (and .thaw(), and .restore()) and the .runtime_suspend() pointer to -the same value as .suspend() (and .freeze() and .poweroff()). +The DEFINE_RUNTIME_DEV_PM_OPS() is similar to DEFINE_SIMPLE_DEV_PM_OPS(), but it +additionally sets the .runtime_resume() pointer to pm_runtime_force_resume() +and the .runtime_suspend() pointer to pm_runtime_force_suspend(). -The SET_SYSTEM_SLEEP_PM_OPS can be used inside of a declaration of struct +The SYSTEM_SLEEP_PM_OPS() can be used inside of a declaration of struct dev_pm_ops to indicate that one suspend routine is to be pointed to by the .suspend(), .freeze(), and .poweroff() members and one resume routine is to be pointed to by the .resume(), .thaw(), and .restore() members. diff --git a/Documentation/power/runtime_pm.rst b/Documentation/power/runtime_pm.rst index 5c4e730f38d0..53d1996460ab 100644 --- a/Documentation/power/runtime_pm.rst +++ b/Documentation/power/runtime_pm.rst @@ -811,8 +811,8 @@ subsystem-level dev_pm_ops structure. Device drivers that wish to use the same function as a system suspend, freeze, poweroff and runtime suspend callback, and similarly for system resume, thaw, -restore, and runtime resume, can achieve this with the help of the -UNIVERSAL_DEV_PM_OPS macro defined in include/linux/pm.h (possibly setting its +restore, and runtime resume, can achieve similar behaviour with the help of the +DEFINE_RUNTIME_DEV_PM_OPS() defined in include/linux/pm_runtime.h (possibly setting its last argument to NULL). 8. "No-Callback" Devices diff --git a/MAINTAINERS b/MAINTAINERS index 6c5dc4b24d23..988bc1e94c3c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5851,6 +5851,9 @@ CPU POWER MONITORING SUBSYSTEM M: Thomas Renninger <trenn@suse.com> M: Shuah Khan <shuah@kernel.org> M: Shuah Khan <skhan@linuxfoundation.org> +M: John B. Wyatt IV <jwyatt@redhat.com> +M: John B. Wyatt IV <sageofredondo@gmail.com> +M: John Kacur <jkacur@redhat.com> L: linux-pm@vger.kernel.org S: Maintained F: tools/power/cpupower/ diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index a75a07f4931f..775acbdea1a9 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -691,8 +691,6 @@ static inline u32 per_cpu_l2c_id(unsigned int cpu) } #ifdef CONFIG_CPU_SUP_AMD -extern u32 amd_get_highest_perf(void); - /* * Issue a DIV 0/1 insn to clear any division data from previous DIV * operations. @@ -705,7 +703,6 @@ static __always_inline void amd_clear_divider(void) extern void amd_check_microcode(void); #else -static inline u32 amd_get_highest_perf(void) { return 0; } static inline void amd_clear_divider(void) { } static inline void amd_check_microcode(void) { } #endif diff --git a/arch/x86/include/asm/topology.h b/arch/x86/include/asm/topology.h index abe3a8f22cbd..aef70336d624 100644 --- a/arch/x86/include/asm/topology.h +++ b/arch/x86/include/asm/topology.h @@ -282,9 +282,22 @@ static inline long arch_scale_freq_capacity(int cpu) } #define arch_scale_freq_capacity arch_scale_freq_capacity +bool arch_enable_hybrid_capacity_scale(void); +void arch_set_cpu_capacity(int cpu, unsigned long cap, unsigned long max_cap, + unsigned long cap_freq, unsigned long base_freq); + +unsigned long arch_scale_cpu_capacity(int cpu); +#define arch_scale_cpu_capacity arch_scale_cpu_capacity + extern void arch_set_max_freq_ratio(bool turbo_disabled); extern void freq_invariance_set_perf_ratio(u64 ratio, bool turbo_disabled); #else +static inline bool arch_enable_hybrid_capacity_scale(void) { return false; } +static inline void arch_set_cpu_capacity(int cpu, unsigned long cap, + unsigned long max_cap, + unsigned long cap_freq, + unsigned long base_freq) { } + static inline void arch_set_max_freq_ratio(bool turbo_disabled) { } static inline void freq_invariance_set_perf_ratio(u64 ratio, bool turbo_disabled) { } #endif diff --git a/arch/x86/kernel/acpi/cppc.c b/arch/x86/kernel/acpi/cppc.c index ff8f25faca3d..956984054bf3 100644 --- a/arch/x86/kernel/acpi/cppc.c +++ b/arch/x86/kernel/acpi/cppc.c @@ -9,6 +9,17 @@ #include <asm/processor.h> #include <asm/topology.h> +#define CPPC_HIGHEST_PERF_PERFORMANCE 196 +#define CPPC_HIGHEST_PERF_PREFCORE 166 + +enum amd_pref_core { + AMD_PREF_CORE_UNKNOWN = 0, + AMD_PREF_CORE_SUPPORTED, + AMD_PREF_CORE_UNSUPPORTED, +}; +static enum amd_pref_core amd_pref_core_detected; +static u64 boost_numerator; + /* Refer to drivers/acpi/cppc_acpi.c for the description of functions */ bool cpc_supported_by_cpu(void) @@ -69,31 +80,30 @@ int cpc_write_ffh(int cpunum, struct cpc_reg *reg, u64 val) static void amd_set_max_freq_ratio(void) { struct cppc_perf_caps perf_caps; - u64 highest_perf, nominal_perf; + u64 numerator, nominal_perf; u64 perf_ratio; int rc; rc = cppc_get_perf_caps(0, &perf_caps); if (rc) { - pr_debug("Could not retrieve perf counters (%d)\n", rc); + pr_warn("Could not retrieve perf counters (%d)\n", rc); return; } - highest_perf = amd_get_highest_perf(); + rc = amd_get_boost_ratio_numerator(0, &numerator); + if (rc) { + pr_warn("Could not retrieve highest performance (%d)\n", rc); + return; + } nominal_perf = perf_caps.nominal_perf; - if (!highest_perf || !nominal_perf) { - pr_debug("Could not retrieve highest or nominal performance\n"); + if (!nominal_perf) { + pr_warn("Could not retrieve nominal performance\n"); return; } - perf_ratio = div_u64(highest_perf * SCHED_CAPACITY_SCALE, nominal_perf); /* midpoint between max_boost and max_P */ - perf_ratio = (perf_ratio + SCHED_CAPACITY_SCALE) >> 1; - if (!perf_ratio) { - pr_debug("Non-zero highest/nominal perf values led to a 0 ratio\n"); - return; - } + perf_ratio = (div_u64(numerator * SCHED_CAPACITY_SCALE, nominal_perf) + SCHED_CAPACITY_SCALE) >> 1; freq_invariance_set_perf_ratio(perf_ratio, false); } @@ -116,3 +126,143 @@ void init_freq_invariance_cppc(void) init_done = true; mutex_unlock(&freq_invariance_lock); } + +/* + * Get the highest performance register value. + * @cpu: CPU from which to get highest performance. + * @highest_perf: Return address for highest performance value. + * + * Return: 0 for success, negative error code otherwise. + */ +int amd_get_highest_perf(unsigned int cpu, u32 *highest_perf) +{ + u64 val; + int ret; + + if (cpu_feature_enabled(X86_FEATURE_CPPC)) { + ret = rdmsrl_safe_on_cpu(cpu, MSR_AMD_CPPC_CAP1, &val); + if (ret) + goto out; + + val = AMD_CPPC_HIGHEST_PERF(val); + } else { + ret = cppc_get_highest_perf(cpu, &val); + if (ret) + goto out; + } + + WRITE_ONCE(*highest_perf, (u32)val); +out: + return ret; +} +EXPORT_SYMBOL_GPL(amd_get_highest_perf); + +/** + * amd_detect_prefcore: Detect if CPUs in the system support preferred cores + * @detected: Output variable for the result of the detection. + * + * Determine whether CPUs in the system support preferred cores. On systems + * that support preferred cores, different highest perf values will be found + * on different cores. On other systems, the highest perf value will be the + * same on all cores. + * + * The result of the detection will be stored in the 'detected' parameter. + * + * Return: 0 for success, negative error code otherwise + */ +int amd_detect_prefcore(bool *detected) +{ + int cpu, count = 0; + u64 highest_perf[2] = {0}; + + if (WARN_ON(!detected)) + return -EINVAL; + + switch (amd_pref_core_detected) { + case AMD_PREF_CORE_SUPPORTED: + *detected = true; + return 0; + case AMD_PREF_CORE_UNSUPPORTED: + *detected = false; + return 0; + default: + break; + } + + for_each_present_cpu(cpu) { + u32 tmp; + int ret; + + ret = amd_get_highest_perf(cpu, &tmp); + if (ret) + return ret; + + if (!count || (count == 1 && tmp != highest_perf[0])) + highest_perf[count++] = tmp; + + if (count == 2) + break; + } + + *detected = (count == 2); + boost_numerator = highest_perf[0]; + + amd_pref_core_detected = *detected ? AMD_PREF_CORE_SUPPORTED : + AMD_PREF_CORE_UNSUPPORTED; + + pr_debug("AMD CPPC preferred core is %ssupported (highest perf: 0x%llx)\n", + *detected ? "" : "un", highest_perf[0]); + + return 0; +} +EXPORT_SYMBOL_GPL(amd_detect_prefcore); + +/** + * amd_get_boost_ratio_numerator: Get the numerator to use for boost ratio calculation + * @cpu: CPU to get numerator for. + * @numerator: Output variable for numerator. + * + * Determine the numerator to use for calculating the boost ratio on + * a CPU. On systems that support preferred cores, this will be a hardcoded + * value. On other systems this will the highest performance register value. + * + * If booting the system with amd-pstate enabled but preferred cores disabled then + * the correct boost numerator will be returned to match hardware capabilities + * even if the preferred cores scheduling hints are not enabled. + * + * Return: 0 for success, negative error code otherwise. + */ +int amd_get_boost_ratio_numerator(unsigned int cpu, u64 *numerator) +{ + bool prefcore; + int ret; + + ret = amd_detect_prefcore(&prefcore); + if (ret) + return ret; + + /* without preferred cores, return the highest perf register value */ + if (!prefcore) { + *numerator = boost_numerator; + return 0; + } + + /* + * For AMD CPUs with Family ID 19H and Model ID range 0x70 to 0x7f, + * the highest performance level is set to 196. + * https://bugzilla.kernel.org/show_bug.cgi?id=218759 + */ + if (cpu_feature_enabled(X86_FEATURE_ZEN4)) { + switch (boot_cpu_data.x86_model) { + case 0x70 ... 0x7f: + *numerator = CPPC_HIGHEST_PERF_PERFORMANCE; + return 0; + default: + break; + } + } + *numerator = CPPC_HIGHEST_PERF_PREFCORE; + + return 0; +} +EXPORT_SYMBOL_GPL(amd_get_boost_ratio_numerator); diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 1e0fe5f8ab84..015971adadfc 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -1190,22 +1190,6 @@ unsigned long amd_get_dr_addr_mask(unsigned int dr) } EXPORT_SYMBOL_GPL(amd_get_dr_addr_mask); -u32 amd_get_highest_perf(void) -{ - struct cpuinfo_x86 *c = &boot_cpu_data; - - if (c->x86 == 0x17 && ((c->x86_model >= 0x30 && c->x86_model < 0x40) || - (c->x86_model >= 0x70 && c->x86_model < 0x80))) - return 166; - - if (c->x86 == 0x19 && ((c->x86_model >= 0x20 && c->x86_model < 0x30) || - (c->x86_model >= 0x40 && c->x86_model < 0x70))) - return 166; - - return 255; -} -EXPORT_SYMBOL_GPL(amd_get_highest_perf); - static void zenbleed_check_cpu(void *unused) { struct cpuinfo_x86 *c = &cpu_data(smp_processor_id()); diff --git a/arch/x86/kernel/cpu/aperfmperf.c b/arch/x86/kernel/cpu/aperfmperf.c index 0b69bfbf345d..f642de2ebdac 100644 --- a/arch/x86/kernel/cpu/aperfmperf.c +++ b/arch/x86/kernel/cpu/aperfmperf.c @@ -349,9 +349,89 @@ static DECLARE_WORK(disable_freq_invariance_work, DEFINE_PER_CPU(unsigned long, arch_freq_scale) = SCHED_CAPACITY_SCALE; EXPORT_PER_CPU_SYMBOL_GPL(arch_freq_scale); +static DEFINE_STATIC_KEY_FALSE(arch_hybrid_cap_scale_key); + +struct arch_hybrid_cpu_scale { + unsigned long capacity; + unsigned long freq_ratio; +}; + +static struct arch_hybrid_cpu_scale __percpu *arch_cpu_scale; + +/** + * arch_enable_hybrid_capacity_scale() - Enable hybrid CPU capacity scaling + * + * Allocate memory for per-CPU data used by hybrid CPU capacity scaling, + * initialize it and set the static key controlling its code paths. + * + * Must be called before arch_set_cpu_capacity(). + */ +bool arch_enable_hybrid_capacity_scale(void) +{ + int cpu; + + if (static_branch_unlikely(&arch_hybrid_cap_scale_key)) { + WARN_ONCE(1, "Hybrid CPU capacity scaling already enabled"); + return true; + } + + arch_cpu_scale = alloc_percpu(struct arch_hybrid_cpu_scale); + if (!arch_cpu_scale) + return false; + + for_each_possible_cpu(cpu) { + per_cpu_ptr(arch_cpu_scale, cpu)->capacity = SCHED_CAPACITY_SCALE; + per_cpu_ptr(arch_cpu_scale, cpu)->freq_ratio = arch_max_freq_ratio; + } + + static_branch_enable(&arch_hybrid_cap_scale_key); + + pr_info("Hybrid CPU capacity scaling enabled\n"); + + return true; +} + +/** + * arch_set_cpu_capacity() - Set scale-invariance parameters for a CPU + * @cpu: Target CPU. + * @cap: Capacity of @cpu at its maximum frequency, relative to @max_cap. + * @max_cap: System-wide maximum CPU capacity. + * @cap_freq: Frequency of @cpu corresponding to @cap. + * @base_freq: Frequency of @cpu at which MPERF counts. + * + * The units in which @cap and @max_cap are expressed do not matter, so long + * as they are consistent, because the former is effectively divided by the + * latter. Analogously for @cap_freq and @base_freq. + * + * After calling this function for all CPUs, call arch_rebuild_sched_domains() + * to let the scheduler know that capacity-aware scheduling can be used going + * forward. + */ +void arch_set_cpu_capacity(int cpu, unsigned long cap, unsigned long max_cap, + unsigned long cap_freq, unsigned long base_freq) +{ + if (static_branch_likely(&arch_hybrid_cap_scale_key)) { + WRITE_ONCE(per_cpu_ptr(arch_cpu_scale, cpu)->capacity, + div_u64(cap << SCHED_CAPACITY_SHIFT, max_cap)); + WRITE_ONCE(per_cpu_ptr(arch_cpu_scale, cpu)->freq_ratio, + div_u64(cap_freq << SCHED_CAPACITY_SHIFT, base_freq)); + } else { + WARN_ONCE(1, "Hybrid CPU capacity scaling not enabled"); + } +} + +unsigned long arch_scale_cpu_capacity(int cpu) +{ + if (static_branch_unlikely(&arch_hybrid_cap_scale_key)) + return READ_ONCE(per_cpu_ptr(arch_cpu_scale, cpu)->capacity); + + return SCHED_CAPACITY_SCALE; +} +EXPORT_SYMBOL_GPL(arch_scale_cpu_capacity); + static void scale_freq_tick(u64 acnt, u64 mcnt) { - u64 freq_scale; + u64 freq_scale, freq_ratio; if (!arch_scale_freq_invariant()) return; @@ -359,7 +439,12 @@ static void scale_freq_tick(u64 acnt, u64 mcnt) if (check_shl_overflow(acnt, 2*SCHED_CAPACITY_SHIFT, &acnt)) goto error; - if (check_mul_overflow(mcnt, arch_max_freq_ratio, &mcnt) || !mcnt) + if (static_branch_unlikely(&arch_hybrid_cap_scale_key)) + freq_ratio = READ_ONCE(this_cpu_ptr(arch_cpu_scale)->freq_ratio); + else + freq_ratio = arch_max_freq_ratio; + + if (check_mul_overflow(mcnt, freq_ratio, &mcnt) || !mcnt) goto error; freq_scale = div64_u64(acnt, mcnt); diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig index 10cda6f2fe1d..2561b215432a 100644 --- a/drivers/cpufreq/Kconfig +++ b/drivers/cpufreq/Kconfig @@ -231,9 +231,7 @@ if X86 source "drivers/cpufreq/Kconfig.x86" endif -if ARM || ARM64 source "drivers/cpufreq/Kconfig.arm" -endif if PPC32 || PPC64 source "drivers/cpufreq/Kconfig.powerpc" diff --git a/drivers/cpufreq/Kconfig.arm b/drivers/cpufreq/Kconfig.arm index 96b404ce829f..5f7e13e60c80 100644 --- a/drivers/cpufreq/Kconfig.arm +++ b/drivers/cpufreq/Kconfig.arm @@ -5,7 +5,7 @@ config ARM_ALLWINNER_SUN50I_CPUFREQ_NVMEM tristate "Allwinner nvmem based SUN50I CPUFreq driver" - depends on ARCH_SUNXI + depends on ARCH_SUNXI || COMPILE_TEST depends on NVMEM_SUNXI_SID select PM_OPP help @@ -26,15 +26,17 @@ config ARM_APPLE_SOC_CPUFREQ config ARM_ARMADA_37XX_CPUFREQ tristate "Armada 37xx CPUFreq support" - depends on ARCH_MVEBU && CPUFREQ_DT + depends on ARCH_MVEBU || COMPILE_TEST + depends on CPUFREQ_DT help This adds the CPUFreq driver support for Marvell Armada 37xx SoCs. The Armada 37xx PMU supports 4 frequency and VDD levels. config ARM_ARMADA_8K_CPUFREQ tristate "Armada 8K CPUFreq driver" - depends on ARCH_MVEBU && CPUFREQ_DT - select ARMADA_AP_CPU_CLK + depends on ARCH_MVEBU || COMPILE_TEST + depends on CPUFREQ_DT + select ARMADA_AP_CPU_CLK if COMMON_CLK help This enables the CPUFreq driver support for Marvell Armada8k SOCs. @@ -56,7 +58,7 @@ config ARM_SCPI_CPUFREQ config ARM_VEXPRESS_SPC_CPUFREQ tristate "Versatile Express SPC based CPUfreq driver" depends on ARM_CPU_TOPOLOGY && HAVE_CLK - depends on ARCH_VEXPRESS_SPC + depends on ARCH_VEXPRESS_SPC || COMPILE_TEST select PM_OPP help This add the CPUfreq driver support for Versatile Express @@ -75,7 +77,8 @@ config ARM_BRCMSTB_AVS_CPUFREQ config ARM_HIGHBANK_CPUFREQ tristate "Calxeda Highbank-based" - depends on ARCH_HIGHBANK && CPUFREQ_DT && REGULATOR + depends on ARCH_HIGHBANK || COMPILE_TEST + depends on CPUFREQ_DT && REGULATOR && PL320_MBOX default m help This adds the CPUFreq driver for Calxeda Highbank SoC @@ -96,7 +99,8 @@ config ARM_IMX6Q_CPUFREQ config ARM_IMX_CPUFREQ_DT tristate "Freescale i.MX8M cpufreq support" - depends on ARCH_MXC && CPUFREQ_DT + depends on CPUFREQ_DT + depends on ARCH_MXC || COMPILE_TEST help This adds cpufreq driver support for Freescale i.MX7/i.MX8M series SoCs, based on cpufreq-dt. @@ -111,7 +115,8 @@ config ARM_KIRKWOOD_CPUFREQ config ARM_MEDIATEK_CPUFREQ tristate "CPU Frequency scaling support for MediaTek SoCs" - depends on ARCH_MEDIATEK && REGULATOR + depends on ARCH_MEDIATEK || COMPILE_TEST + depends on REGULATOR select PM_OPP help This adds the CPUFreq driver support for MediaTek SoCs. @@ -130,12 +135,12 @@ config ARM_MEDIATEK_CPUFREQ_HW config ARM_OMAP2PLUS_CPUFREQ bool "TI OMAP2+" - depends on ARCH_OMAP2PLUS + depends on ARCH_OMAP2PLUS || COMPILE_TEST default ARCH_OMAP2PLUS config ARM_QCOM_CPUFREQ_NVMEM tristate "Qualcomm nvmem based CPUFreq" - depends on ARCH_QCOM + depends on ARCH_QCOM || COMPILE_TEST depends on NVMEM_QCOM_QFPROM depends on QCOM_SMEM select PM_OPP @@ -166,7 +171,7 @@ config ARM_RASPBERRYPI_CPUFREQ config ARM_S3C64XX_CPUFREQ bool "Samsung S3C64XX" - depends on CPU_S3C6410 + depends on CPU_S3C6410 || COMPILE_TEST default y help This adds the CPUFreq driver for Samsung S3C6410 SoC. @@ -175,7 +180,7 @@ config ARM_S3C64XX_CPUFREQ config ARM_S5PV210_CPUFREQ bool "Samsung S5PV210 and S5PC110" - depends on CPU_S5PV210 + depends on CPU_S5PV210 || COMPILE_TEST default y help This adds the CPUFreq driver for Samsung S5PV210 and @@ -199,14 +204,15 @@ config ARM_SCMI_CPUFREQ config ARM_SPEAR_CPUFREQ bool "SPEAr CPUFreq support" - depends on PLAT_SPEAR + depends on PLAT_SPEAR || COMPILE_TEST default y help This adds the CPUFreq driver support for SPEAr SOCs. config ARM_STI_CPUFREQ tristate "STi CPUFreq support" - depends on CPUFREQ_DT && SOC_STIH407 + depends on CPUFREQ_DT + depends on SOC_STIH407 || COMPILE_TEST help This driver uses the generic OPP framework to match the running platform with a predefined set of suitable values. If not provided @@ -216,34 +222,38 @@ config ARM_STI_CPUFREQ config ARM_TEGRA20_CPUFREQ tristate "Tegra20/30 CPUFreq support" - depends on ARCH_TEGRA && CPUFREQ_DT + depends on ARCH_TEGRA || COMPILE_TEST + depends on CPUFREQ_DT default y help This adds the CPUFreq driver support for Tegra20/30 SOCs. config ARM_TEGRA124_CPUFREQ bool "Tegra124 CPUFreq support" - depends on ARCH_TEGRA && CPUFREQ_DT + depends on ARCH_TEGRA || COMPILE_TEST + depends on CPUFREQ_DT default y help This adds the CPUFreq driver support for Tegra124 SOCs. config ARM_TEGRA186_CPUFREQ tristate "Tegra186 CPUFreq support" - depends on ARCH_TEGRA && TEGRA_BPMP + depends on ARCH_TEGRA || COMPILE_TEST + depends on TEGRA_BPMP help This adds the CPUFreq driver support for Tegra186 SOCs. config ARM_TEGRA194_CPUFREQ tristate "Tegra194 CPUFreq support" - depends on ARCH_TEGRA_194_SOC && TEGRA_BPMP + depends on ARCH_TEGRA_194_SOC || (64BIT && COMPILE_TEST) + depends on TEGRA_BPMP default y help This adds CPU frequency driver support for Tegra194 SOCs. config ARM_TI_CPUFREQ bool "Texas Instruments CPUFreq support" - depends on ARCH_OMAP2PLUS || ARCH_K3 + depends on ARCH_OMAP2PLUS || ARCH_K3 || COMPILE_TEST default y help This driver enables valid OPPs on the running platform based on @@ -255,7 +265,7 @@ config ARM_TI_CPUFREQ config ARM_PXA2xx_CPUFREQ tristate "Intel PXA2xx CPUfreq driver" - depends on PXA27x || PXA25x + depends on PXA27x || PXA25x || COMPILE_TEST help This add the CPUFreq driver support for Intel PXA2xx SOCs. diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c index a8ca625a98b8..0f04feb6cafa 100644 --- a/drivers/cpufreq/acpi-cpufreq.c +++ b/drivers/cpufreq/acpi-cpufreq.c @@ -642,10 +642,16 @@ static u64 get_max_boost_ratio(unsigned int cpu) return 0; } - if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) - highest_perf = amd_get_highest_perf(); - else + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) { + ret = amd_get_boost_ratio_numerator(cpu, &highest_perf); + if (ret) { + pr_debug("CPU%d: Unable to get boost ratio numerator (%d)\n", + cpu, ret); + return 0; + } + } else { highest_perf = perf_caps.highest_perf; + } nominal_perf = perf_caps.nominal_perf; diff --git a/drivers/cpufreq/amd-pstate-ut.c b/drivers/cpufreq/amd-pstate-ut.c index b7318669485e..f66701514d90 100644 --- a/drivers/cpufreq/amd-pstate-ut.c +++ b/drivers/cpufreq/amd-pstate-ut.c @@ -54,12 +54,14 @@ static void amd_pstate_ut_acpi_cpc_valid(u32 index); static void amd_pstate_ut_check_enabled(u32 index); static void amd_pstate_ut_check_perf(u32 index); static void amd_pstate_ut_check_freq(u32 index); +static void amd_pstate_ut_check_driver(u32 index); static struct amd_pstate_ut_struct amd_pstate_ut_cases[] = { {"amd_pstate_ut_acpi_cpc_valid", amd_pstate_ut_acpi_cpc_valid }, {"amd_pstate_ut_check_enabled", amd_pstate_ut_check_enabled }, {"amd_pstate_ut_check_perf", amd_pstate_ut_check_perf }, - {"amd_pstate_ut_check_freq", amd_pstate_ut_check_freq } + {"amd_pstate_ut_check_freq", amd_pstate_ut_check_freq }, + {"amd_pstate_ut_check_driver", amd_pstate_ut_check_driver } }; static bool get_shared_mem(void) @@ -257,6 +259,43 @@ skip_test: cpufreq_cpu_put(policy); } +static int amd_pstate_set_mode(enum amd_pstate_mode mode) +{ + const char *mode_str = amd_pstate_get_mode_string(mode); + + pr_debug("->setting mode to %s\n", mode_str); + + return amd_pstate_update_status(mode_str, strlen(mode_str)); +} + +static void amd_pstate_ut_check_driver(u32 index) +{ + enum amd_pstate_mode mode1, mode2 = AMD_PSTATE_DISABLE; + int ret; + + for (mode1 = AMD_PSTATE_DISABLE; mode1 < AMD_PSTATE_MAX; mode1++) { + ret = amd_pstate_set_mode(mode1); + if (ret) + goto out; + for (mode2 = AMD_PSTATE_DISABLE; mode2 < AMD_PSTATE_MAX; mode2++) { + if (mode1 == mode2) + continue; + ret = amd_pstate_set_mode(mode2); + if (ret) + goto out; + } + } +out: + if (ret) + pr_warn("%s: failed to update status for %s->%s: %d\n", __func__, + amd_pstate_get_mode_string(mode1), + amd_pstate_get_mode_string(mode2), ret); + + amd_pstate_ut_cases[index].result = ret ? + AMD_PSTATE_UT_RESULT_FAIL : + AMD_PSTATE_UT_RESULT_PASS; +} + static int __init amd_pstate_ut_init(void) { u32 i = 0, arr_size = ARRAY_SIZE(amd_pstate_ut_cases); diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index 259a917da75f..15e201d5e911 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -52,26 +52,12 @@ #define AMD_PSTATE_TRANSITION_LATENCY 20000 #define AMD_PSTATE_TRANSITION_DELAY 1000 #define AMD_PSTATE_FAST_CPPC_TRANSITION_DELAY 600 -#define CPPC_HIGHEST_PERF_PERFORMANCE 196 -#define CPPC_HIGHEST_PERF_DEFAULT 166 #define AMD_CPPC_EPP_PERFORMANCE 0x00 #define AMD_CPPC_EPP_BALANCE_PERFORMANCE 0x80 #define AMD_CPPC_EPP_BALANCE_POWERSAVE 0xBF #define AMD_CPPC_EPP_POWERSAVE 0xFF -/* - * enum amd_pstate_mode - driver working mode of amd pstate - */ -enum amd_pstate_mode { - AMD_PSTATE_UNDEFINED = 0, - AMD_PSTATE_DISABLE, - AMD_PSTATE_PASSIVE, - AMD_PSTATE_ACTIVE, - AMD_PSTATE_GUIDED, - AMD_PSTATE_MAX, -}; - static const char * const amd_pstate_mode_string[] = { [AMD_PSTATE_UNDEFINED] = "undefined", [AMD_PSTATE_DISABLE] = "disable", @@ -81,6 +67,14 @@ static const char * const amd_pstate_mode_string[] = { NULL, }; +const char *amd_pstate_get_mode_string(enum amd_pstate_mode mode) +{ + if (mode < 0 || mode >= AMD_PSTATE_MAX) + return NULL; + return amd_pstate_mode_string[mode]; +} +EXPORT_SYMBOL_GPL(amd_pstate_get_mode_string); + struct quirk_entry { u32 nominal_freq; u32 lowest_freq; @@ -372,43 +366,17 @@ static inline int amd_pstate_enable(bool enable) return static_call(amd_pstate_enable)(enable); } -static u32 amd_pstate_highest_perf_set(struct amd_cpudata *cpudata) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - - /* - * For AMD CPUs with Family ID 19H and Model ID range 0x70 to 0x7f, - * the highest performance level is set to 196. - * https://bugzilla.kernel.org/show_bug.cgi?id=218759 - */ - if (c->x86 == 0x19 && (c->x86_model >= 0x70 && c->x86_model <= 0x7f)) - return CPPC_HIGHEST_PERF_PERFORMANCE; - - return CPPC_HIGHEST_PERF_DEFAULT; -} - static int pstate_init_perf(struct amd_cpudata *cpudata) { u64 cap1; - u32 highest_perf; int ret = rdmsrl_safe_on_cpu(cpudata->cpu, MSR_AMD_CPPC_CAP1, &cap1); if (ret) return ret; - /* For platforms that do not support the preferred core feature, the - * highest_pef may be configured with 166 or 255, to avoid max frequency - * calculated wrongly. we take the AMD_CPPC_HIGHEST_PERF(cap1) value as - * the default max perf. - */ - if (cpudata->hw_prefcore) - highest_perf = amd_pstate_highest_perf_set(cpudata); - else - highest_perf = AMD_CPPC_HIGHEST_PERF(cap1); - - WRITE_ONCE(cpudata->highest_perf, highest_perf); - WRITE_ONCE(cpudata->max_limit_perf, highest_perf); + WRITE_ONCE(cpudata->highest_perf, AMD_CPPC_HIGHEST_PERF(cap1)); + WRITE_ONCE(cpudata->max_limit_perf, AMD_CPPC_HIGHEST_PERF(cap1)); WRITE_ONCE(cpudata->nominal_perf, AMD_CPPC_NOMINAL_PERF(cap1)); WRITE_ONCE(cpudata->lowest_nonlinear_perf, AMD_CPPC_LOWNONLIN_PERF(cap1)); WRITE_ONCE(cpudata->lowest_perf, AMD_CPPC_LOWEST_PERF(cap1)); @@ -420,19 +388,13 @@ static int pstate_init_perf(struct amd_cpudata *cpudata) static int cppc_init_perf(struct amd_cpudata *cpudata) { struct cppc_perf_caps cppc_perf; - u32 highest_perf; int ret = cppc_get_perf_caps(cpudata->cpu, &cppc_perf); if (ret) return ret; - if (cpudata->hw_prefcore) - highest_perf = amd_pstate_highest_perf_set(cpudata); - else - highest_perf = cppc_perf.highest_perf; - - WRITE_ONCE(cpudata->highest_perf, highest_perf); - WRITE_ONCE(cpudata->max_limit_perf, highest_perf); + WRITE_ONCE(cpudata->highest_perf, cppc_perf.highest_perf); + WRITE_ONCE(cpudata->max_limit_perf, cppc_perf.highest_perf); WRITE_ONCE(cpudata->nominal_perf, cppc_perf.nominal_perf); WRITE_ONCE(cpudata->lowest_nonlinear_perf, cppc_perf.lowest_nonlinear_perf); @@ -554,12 +516,15 @@ static void amd_pstate_update(struct amd_cpudata *cpudata, u32 min_perf, } if (value == prev) - return; + goto cpufreq_policy_put; WRITE_ONCE(cpudata->cppc_req_cached, value); amd_pstate_update_perf(cpudata, min_perf, des_perf, max_perf, fast_switch); + +cpufreq_policy_put: + cpufreq_cpu_put(policy); } static int amd_pstate_verify(struct cpufreq_policy_data *policy) @@ -656,7 +621,12 @@ static void amd_pstate_adjust_perf(unsigned int cpu, unsigned long max_perf, min_perf, des_perf, cap_perf, lowest_nonlinear_perf; struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); - struct amd_cpudata *cpudata = policy->driver_data; + struct amd_cpudata *cpudata; + + if (!policy) + return; + + cpudata = policy->driver_data; if (policy->min != cpudata->min_limit_freq || policy->max != cpudata->max_limit_freq) amd_pstate_update_min_max_limit(policy); @@ -803,66 +773,22 @@ static void amd_pstste_sched_prefcore_workfn(struct work_struct *work) } static DECLARE_WORK(sched_prefcore_work, amd_pstste_sched_prefcore_workfn); -/* - * Get the highest performance register value. - * @cpu: CPU from which to get highest performance. - * @highest_perf: Return address. - * - * Return: 0 for success, -EIO otherwise. - */ -static int amd_pstate_get_highest_perf(int cpu, u32 *highest_perf) -{ - int ret; - - if (cpu_feature_enabled(X86_FEATURE_CPPC)) { - u64 cap1; - - ret = rdmsrl_safe_on_cpu(cpu, MSR_AMD_CPPC_CAP1, &cap1); - if (ret) - return ret; - WRITE_ONCE(*highest_perf, AMD_CPPC_HIGHEST_PERF(cap1)); - } else { - u64 cppc_highest_perf; - - ret = cppc_get_highest_perf(cpu, &cppc_highest_perf); - if (ret) - return ret; - WRITE_ONCE(*highest_perf, cppc_highest_perf); - } - - return (ret); -} - #define CPPC_MAX_PERF U8_MAX static void amd_pstate_init_prefcore(struct amd_cpudata *cpudata) { - int ret, prio; - u32 highest_perf; - - ret = amd_pstate_get_highest_perf(cpudata->cpu, &highest_perf); - if (ret) + /* user disabled or not detected */ + if (!amd_pstate_prefcore) return; cpudata->hw_prefcore = true; - /* check if CPPC preferred core feature is enabled*/ - if (highest_perf < CPPC_MAX_PERF) - prio = (int)highest_perf; - else { - pr_debug("AMD CPPC preferred core is unsupported!\n"); - cpudata->hw_prefcore = false; - return; - } - - if (!amd_pstate_prefcore) - return; /* * The priorities can be set regardless of whether or not * sched_set_itmt_support(true) has been called and it is valid to * update them at any time after it has been called. */ - sched_set_itmt_core_prio(prio, cpudata->cpu); + sched_set_itmt_core_prio((int)READ_ONCE(cpudata->highest_perf), cpudata->cpu); schedule_work(&sched_prefcore_work); } @@ -870,22 +796,27 @@ static void amd_pstate_init_prefcore(struct amd_cpudata *cpudata) static void amd_pstate_update_limits(unsigned int cpu) { struct cpufreq_policy *policy = cpufreq_cpu_get(cpu); - struct amd_cpudata *cpudata = policy->driver_data; + struct amd_cpudata *cpudata; u32 prev_high = 0, cur_high = 0; int ret; bool highest_perf_changed = false; - mutex_lock(&amd_pstate_driver_lock); - if ((!amd_pstate_prefcore) || (!cpudata->hw_prefcore)) - goto free_cpufreq_put; + if (!policy) + return; + + cpudata = policy->driver_data; + + if (!amd_pstate_prefcore) + return; - ret = amd_pstate_get_highest_perf(cpu, &cur_high); + mutex_lock(&amd_pstate_driver_lock); + ret = amd_get_highest_perf(cpu, &cur_high); if (ret) goto free_cpufreq_put; prev_high = READ_ONCE(cpudata->prefcore_ranking); - if (prev_high != cur_high) { - highest_perf_changed = true; + highest_perf_changed = (prev_high != cur_high); + if (highest_perf_changed) { WRITE_ONCE(cpudata->prefcore_ranking, cur_high); if (cur_high < CPPC_MAX_PERF) @@ -949,8 +880,8 @@ static u32 amd_pstate_get_transition_latency(unsigned int cpu) static int amd_pstate_init_freq(struct amd_cpudata *cpudata) { int ret; - u32 min_freq; - u32 highest_perf, max_freq; + u32 min_freq, max_freq; + u64 numerator; u32 nominal_perf, nominal_freq; u32 lowest_nonlinear_perf, lowest_nonlinear_freq; u32 boost_ratio, lowest_nonlinear_ratio; @@ -972,8 +903,10 @@ static int amd_pstate_init_freq(struct amd_cpudata *cpudata) nominal_perf = READ_ONCE(cpudata->nominal_perf); - highest_perf = READ_ONCE(cpudata->highest_perf); - boost_ratio = div_u64(highest_perf << SCHED_CAPACITY_SHIFT, nominal_perf); + ret = amd_get_boost_ratio_numerator(cpudata->cpu, &numerator); + if (ret) + return ret; + boost_ratio = div_u64(numerator << SCHED_CAPACITY_SHIFT, nominal_perf); max_freq = (nominal_freq * boost_ratio >> SCHED_CAPACITY_SHIFT) * 1000; lowest_nonlinear_perf = READ_ONCE(cpudata->lowest_nonlinear_perf); @@ -1028,12 +961,12 @@ static int amd_pstate_cpu_init(struct cpufreq_policy *policy) cpudata->cpu = policy->cpu; - amd_pstate_init_prefcore(cpudata); - ret = amd_pstate_init_perf(cpudata); if (ret) goto free_cpudata1; + amd_pstate_init_prefcore(cpudata); + ret = amd_pstate_init_freq(cpudata); if (ret) goto free_cpudata1; @@ -1349,7 +1282,7 @@ static ssize_t amd_pstate_show_status(char *buf) return sysfs_emit(buf, "%s\n", amd_pstate_mode_string[cppc_state]); } -static int amd_pstate_update_status(const char *buf, size_t size) +int amd_pstate_update_status(const char *buf, size_t size) { int mode_idx; @@ -1366,6 +1299,7 @@ static int amd_pstate_update_status(const char *buf, size_t size) return 0; } +EXPORT_SYMBOL_GPL(amd_pstate_update_status); static ssize_t status_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -1483,12 +1417,12 @@ static int amd_pstate_epp_cpu_init(struct cpufreq_policy *policy) cpudata->cpu = policy->cpu; cpudata->epp_policy = 0; - amd_pstate_init_prefcore(cpudata); - ret = amd_pstate_init_perf(cpudata); if (ret) goto free_cpudata1; + amd_pstate_init_prefcore(cpudata); + ret = amd_pstate_init_freq(cpudata); if (ret) goto free_cpudata1; @@ -1555,7 +1489,7 @@ static void amd_pstate_epp_cpu_exit(struct cpufreq_policy *policy) pr_debug("CPU %d exiting\n", policy->cpu); } -static void amd_pstate_epp_update_limit(struct cpufreq_policy *policy) +static int amd_pstate_epp_update_limit(struct cpufreq_policy *policy) { struct amd_cpudata *cpudata = policy->driver_data; u32 max_perf, min_perf, min_limit_perf, max_limit_perf; @@ -1605,7 +1539,7 @@ static void amd_pstate_epp_update_limit(struct cpufreq_policy *policy) * This return value can only be negative for shared_memory * systems where EPP register read/write not supported. */ - return; + return epp; } if (cpudata->policy == CPUFREQ_POLICY_PERFORMANCE) @@ -1618,12 +1552,13 @@ static void amd_pstate_epp_update_limit(struct cpufreq_policy *policy) } WRITE_ONCE(cpudata->cppc_req_cached, value); - amd_pstate_set_epp(cpudata, epp); + return amd_pstate_set_epp(cpudata, epp); } static int amd_pstate_epp_set_policy(struct cpufreq_policy *policy) { struct amd_cpudata *cpudata = policy->driver_data; + int ret; if (!policy->cpuinfo.max_freq) return -ENODEV; @@ -1633,7 +1568,9 @@ static int amd_pstate_epp_set_policy(struct cpufreq_policy *policy) cpudata->policy = policy->policy; - amd_pstate_epp_update_limit(policy); + ret = amd_pstate_epp_update_limit(policy); + if (ret) + return ret; /* * policy->cur is never updated with the amd_pstate_epp driver, but it @@ -1947,6 +1884,12 @@ static int __init amd_pstate_init(void) static_call_update(amd_pstate_update_perf, cppc_update_perf); } + if (amd_pstate_prefcore) { + ret = amd_detect_prefcore(&amd_pstate_prefcore); + if (ret) + return ret; + } + /* enable amd pstate feature */ ret = amd_pstate_enable(true); if (ret) { diff --git a/drivers/cpufreq/amd-pstate.h b/drivers/cpufreq/amd-pstate.h index cc8bb2bc325a..cd573bc6b6db 100644 --- a/drivers/cpufreq/amd-pstate.h +++ b/drivers/cpufreq/amd-pstate.h @@ -103,4 +103,18 @@ struct amd_cpudata { bool boost_state; }; +/* + * enum amd_pstate_mode - driver working mode of amd pstate + */ +enum amd_pstate_mode { + AMD_PSTATE_UNDEFINED = 0, + AMD_PSTATE_DISABLE, + AMD_PSTATE_PASSIVE, + AMD_PSTATE_ACTIVE, + AMD_PSTATE_GUIDED, + AMD_PSTATE_MAX, +}; +const char *amd_pstate_get_mode_string(enum amd_pstate_mode mode); +int amd_pstate_update_status(const char *buf, size_t size); + #endif /* _LINUX_AMD_PSTATE_H */ diff --git a/drivers/cpufreq/apple-soc-cpufreq.c b/drivers/cpufreq/apple-soc-cpufreq.c index af34c22fa273..4dcacab9b4bf 100644 --- a/drivers/cpufreq/apple-soc-cpufreq.c +++ b/drivers/cpufreq/apple-soc-cpufreq.c @@ -85,7 +85,7 @@ static const struct apple_soc_cpufreq_info soc_default_info = { .cur_pstate_mask = 0, /* fallback */ }; -static const struct of_device_id apple_soc_cpufreq_of_match[] = { +static const struct of_device_id apple_soc_cpufreq_of_match[] __maybe_unused = { { .compatible = "apple,t8103-cluster-cpufreq", .data = &soc_t8103_info, diff --git a/drivers/cpufreq/armada-8k-cpufreq.c b/drivers/cpufreq/armada-8k-cpufreq.c index ce5a5641b6dd..7a979db81f09 100644 --- a/drivers/cpufreq/armada-8k-cpufreq.c +++ b/drivers/cpufreq/armada-8k-cpufreq.c @@ -132,7 +132,7 @@ static int __init armada_8k_cpufreq_init(void) int ret = 0, opps_index = 0, cpu, nb_cpus; struct freq_table *freq_tables; struct device_node *node; - struct cpumask cpus; + static struct cpumask cpus; node = of_find_matching_node_and_match(NULL, armada_8k_cpufreq_of_match, NULL); diff --git a/drivers/cpufreq/cpufreq-dt-platdev.c b/drivers/cpufreq/cpufreq-dt-platdev.c index cac379ba006d..18942bfe9c95 100644 --- a/drivers/cpufreq/cpufreq-dt-platdev.c +++ b/drivers/cpufreq/cpufreq-dt-platdev.c @@ -166,6 +166,7 @@ static const struct of_device_id blocklist[] __initconst = { { .compatible = "qcom,sm6350", }, { .compatible = "qcom,sm6375", }, { .compatible = "qcom,sm7225", }, + { .compatible = "qcom,sm7325", }, { .compatible = "qcom,sm8150", }, { .compatible = "qcom,sm8250", }, { .compatible = "qcom,sm8350", }, diff --git a/drivers/cpufreq/cpufreq-dt.c b/drivers/cpufreq/cpufreq-dt.c index 6532c4d71338..983443396f8f 100644 --- a/drivers/cpufreq/cpufreq-dt.c +++ b/drivers/cpufreq/cpufreq-dt.c @@ -69,7 +69,6 @@ static int set_target(struct cpufreq_policy *policy, unsigned int index) static const char *find_supply_name(struct device *dev) { struct device_node *np __free(device_node) = of_node_get(dev->of_node); - struct property *pp; int cpu = dev->id; /* This must be valid for sure */ @@ -77,14 +76,10 @@ static const char *find_supply_name(struct device *dev) return NULL; /* Try "cpu0" for older DTs */ - if (!cpu) { - pp = of_find_property(np, "cpu0-supply", NULL); - if (pp) - return "cpu0"; - } + if (!cpu && of_property_present(np, "cpu0-supply")) + return "cpu0"; - pp = of_find_property(np, "cpu-supply", NULL); - if (pp) + if (of_property_present(np, "cpu-supply")) return "cpu"; dev_dbg(dev, "no regulator for cpu%d\n", cpu); diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 04fc786dd2c0..f98c9438760c 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -575,30 +575,11 @@ unsigned int cpufreq_policy_transition_delay_us(struct cpufreq_policy *policy) return policy->transition_delay_us; latency = policy->cpuinfo.transition_latency / NSEC_PER_USEC; - if (latency) { - unsigned int max_delay_us = 2 * MSEC_PER_SEC; + if (latency) + /* Give a 50% breathing room between updates */ + return latency + (latency >> 1); - /* - * If the platform already has high transition_latency, use it - * as-is. - */ - if (latency > max_delay_us) - return latency; - - /* - * For platforms that can change the frequency very fast (< 2 - * us), the above formula gives a decent transition delay. But - * for platforms where transition_latency is in milliseconds, it - * ends up giving unrealistic values. - * - * Cap the default transition delay to 2 ms, which seems to be - * a reasonable amount of time after which we should reevaluate - * the frequency. - */ - return min(latency * LATENCY_MULTIPLIER, max_delay_us); - } - - return LATENCY_MULTIPLIER; + return USEC_PER_MSEC; } EXPORT_SYMBOL_GPL(cpufreq_policy_transition_delay_us); diff --git a/drivers/cpufreq/intel_pstate.c b/drivers/cpufreq/intel_pstate.c index c0278d023cfc..aaea9a39eced 100644 --- a/drivers/cpufreq/intel_pstate.c +++ b/drivers/cpufreq/intel_pstate.c @@ -16,6 +16,7 @@ #include <linux/tick.h> #include <linux/slab.h> #include <linux/sched/cpufreq.h> +#include <linux/sched/smt.h> #include <linux/list.h> #include <linux/cpu.h> #include <linux/cpufreq.h> @@ -215,6 +216,7 @@ struct global_params { * @hwp_req_cached: Cached value of the last HWP Request MSR * @hwp_cap_cached: Cached value of the last HWP Capabilities MSR * @last_io_update: Last time when IO wake flag was set + * @capacity_perf: Highest perf used for scale invariance * @sched_flags: Store scheduler flags for possible cross CPU update * @hwp_boost_min: Last HWP boosted min performance * @suspended: Whether or not the driver has been suspended. @@ -253,6 +255,7 @@ struct cpudata { u64 hwp_req_cached; u64 hwp_cap_cached; u64 last_io_update; + unsigned int capacity_perf; unsigned int sched_flags; u32 hwp_boost_min; bool suspended; @@ -295,6 +298,7 @@ static int hwp_mode_bdw __ro_after_init; static bool per_cpu_limits __ro_after_init; static bool hwp_forced __ro_after_init; static bool hwp_boost __read_mostly; +static bool hwp_is_hybrid; static struct cpufreq_driver *intel_pstate_driver __read_mostly; @@ -934,6 +938,139 @@ static struct freq_attr *hwp_cpufreq_attrs[] = { NULL, }; +static struct cpudata *hybrid_max_perf_cpu __read_mostly; +/* + * Protects hybrid_max_perf_cpu, the capacity_perf fields in struct cpudata, + * and the x86 arch scale-invariance information from concurrent updates. + */ +static DEFINE_MUTEX(hybrid_capacity_lock); + +static void hybrid_set_cpu_capacity(struct cpudata *cpu) +{ + arch_set_cpu_capacity(cpu->cpu, cpu->capacity_perf, + hybrid_max_perf_cpu->capacity_perf, + cpu->capacity_perf, + cpu->pstate.max_pstate_physical); + + pr_debug("CPU%d: perf = %u, max. perf = %u, base perf = %d\n", cpu->cpu, + cpu->capacity_perf, hybrid_max_perf_cpu->capacity_perf, + cpu->pstate.max_pstate_physical); +} + +static void hybrid_clear_cpu_capacity(unsigned int cpunum) +{ + arch_set_cpu_capacity(cpunum, 1, 1, 1, 1); +} + +static void hybrid_get_capacity_perf(struct cpudata *cpu) +{ + if (READ_ONCE(global.no_turbo)) { + cpu->capacity_perf = cpu->pstate.max_pstate_physical; + return; + } + + cpu->capacity_perf = HWP_HIGHEST_PERF(READ_ONCE(cpu->hwp_cap_cached)); +} + +static void hybrid_set_capacity_of_cpus(void) +{ + int cpunum; + + for_each_online_cpu(cpunum) { + struct cpudata *cpu = all_cpu_data[cpunum]; + + if (cpu) + hybrid_set_cpu_capacity(cpu); + } +} + +static void hybrid_update_cpu_capacity_scaling(void) +{ + struct cpudata *max_perf_cpu = NULL; + unsigned int max_cap_perf = 0; + int cpunum; + + for_each_online_cpu(cpunum) { + struct cpudata *cpu = all_cpu_data[cpunum]; + + if (!cpu) + continue; + + /* + * During initialization, CPU performance at full capacity needs + * to be determined. + */ + if (!hybrid_max_perf_cpu) + hybrid_get_capacity_perf(cpu); + + /* + * If hybrid_max_perf_cpu is not NULL at this point, it is + * being replaced, so don't take it into account when looking + * for the new one. + */ + if (cpu == hybrid_max_perf_cpu) + continue; + + if (cpu->capacity_perf > max_cap_perf) { + max_cap_perf = cpu->capacity_perf; + max_perf_cpu = cpu; + } + } + + if (max_perf_cpu) { + hybrid_max_perf_cpu = max_perf_cpu; + hybrid_set_capacity_of_cpus(); + } else { + pr_info("Found no CPUs with nonzero maximum performance\n"); + /* Revert to the flat CPU capacity structure. */ + for_each_online_cpu(cpunum) + hybrid_clear_cpu_capacity(cpunum); + } +} + +static void __hybrid_init_cpu_capacity_scaling(void) +{ + hybrid_max_perf_cpu = NULL; + hybrid_update_cpu_capacity_scaling(); +} + +static void hybrid_init_cpu_capacity_scaling(void) +{ + bool disable_itmt = false; + + mutex_lock(&hybrid_capacity_lock); + + /* + * If hybrid_max_perf_cpu is set at this point, the hybrid CPU capacity + * scaling has been enabled already and the driver is just changing the + * operation mode. + */ + if (hybrid_max_perf_cpu) { + __hybrid_init_cpu_capacity_scaling(); + goto unlock; + } + + /* + * On hybrid systems, use asym capacity instead of ITMT, but because + * the capacity of SMT threads is not deterministic even approximately, + * do not do that when SMT is in use. + */ + if (hwp_is_hybrid && !sched_smt_active() && arch_enable_hybrid_capacity_scale()) { + __hybrid_init_cpu_capacity_scaling(); + disable_itmt = true; + } + +unlock: + mutex_unlock(&hybrid_capacity_lock); + + /* + * Disabling ITMT causes sched domains to be rebuilt to disable asym + * packing and enable asym capacity. + */ + if (disable_itmt) + sched_clear_itmt_support(); +} + static void __intel_pstate_get_hwp_cap(struct cpudata *cpu) { u64 cap; @@ -962,6 +1099,43 @@ static void intel_pstate_get_hwp_cap(struct cpudata *cpu) } } +static void hybrid_update_capacity(struct cpudata *cpu) +{ + unsigned int max_cap_perf; + + mutex_lock(&hybrid_capacity_lock); + + if (!hybrid_max_perf_cpu) + goto unlock; + + /* + * The maximum performance of the CPU may have changed, but assume + * that the performance of the other CPUs has not changed. + */ + max_cap_perf = hybrid_max_perf_cpu->capacity_perf; + + intel_pstate_get_hwp_cap(cpu); + + hybrid_get_capacity_perf(cpu); + /* Should hybrid_max_perf_cpu be replaced by this CPU? */ + if (cpu->capacity_perf > max_cap_perf) { + hybrid_max_perf_cpu = cpu; + hybrid_set_capacity_of_cpus(); + goto unlock; + } + + /* If this CPU is hybrid_max_perf_cpu, should it be replaced? */ + if (cpu == hybrid_max_perf_cpu && cpu->capacity_perf < max_cap_perf) { + hybrid_update_cpu_capacity_scaling(); + goto unlock; + } + + hybrid_set_cpu_capacity(cpu); + +unlock: + mutex_unlock(&hybrid_capacity_lock); +} + static void intel_pstate_hwp_set(unsigned int cpu) { struct cpudata *cpu_data = all_cpu_data[cpu]; @@ -1070,6 +1244,22 @@ static void intel_pstate_hwp_offline(struct cpudata *cpu) value |= HWP_ENERGY_PERF_PREFERENCE(HWP_EPP_POWERSAVE); wrmsrl_on_cpu(cpu->cpu, MSR_HWP_REQUEST, value); + + mutex_lock(&hybrid_capacity_lock); + + if (!hybrid_max_perf_cpu) { + mutex_unlock(&hybrid_capacity_lock); + + return; + } + + if (hybrid_max_perf_cpu == cpu) + hybrid_update_cpu_capacity_scaling(); + + mutex_unlock(&hybrid_capacity_lock); + + /* Reset the capacity of the CPU going offline to the initial value. */ + hybrid_clear_cpu_capacity(cpu->cpu); } #define POWER_CTL_EE_ENABLE 1 @@ -1165,21 +1355,46 @@ static void __intel_pstate_update_max_freq(struct cpudata *cpudata, static void intel_pstate_update_limits(unsigned int cpu) { struct cpufreq_policy *policy = cpufreq_cpu_acquire(cpu); + struct cpudata *cpudata; if (!policy) return; - __intel_pstate_update_max_freq(all_cpu_data[cpu], policy); + cpudata = all_cpu_data[cpu]; + + __intel_pstate_update_max_freq(cpudata, policy); + + /* Prevent the driver from being unregistered now. */ + mutex_lock(&intel_pstate_driver_lock); cpufreq_cpu_release(policy); + + hybrid_update_capacity(cpudata); + + mutex_unlock(&intel_pstate_driver_lock); } static void intel_pstate_update_limits_for_all(void) { int cpu; - for_each_possible_cpu(cpu) - intel_pstate_update_limits(cpu); + for_each_possible_cpu(cpu) { + struct cpufreq_policy *policy = cpufreq_cpu_acquire(cpu); + + if (!policy) + continue; + + __intel_pstate_update_max_freq(all_cpu_data[cpu], policy); + + cpufreq_cpu_release(policy); + } + + mutex_lock(&hybrid_capacity_lock); + + if (hybrid_max_perf_cpu) + __hybrid_init_cpu_capacity_scaling(); + + mutex_unlock(&hybrid_capacity_lock); } /************************** sysfs begin ************************/ @@ -1618,6 +1833,13 @@ static void intel_pstate_notify_work(struct work_struct *work) __intel_pstate_update_max_freq(cpudata, policy); cpufreq_cpu_release(policy); + + /* + * The driver will not be unregistered while this function is + * running, so update the capacity without acquiring the driver + * lock. + */ + hybrid_update_capacity(cpudata); } wrmsrl_on_cpu(cpudata->cpu, MSR_HWP_STATUS, 0); @@ -2034,8 +2256,10 @@ static void intel_pstate_get_cpu_pstates(struct cpudata *cpu) if (pstate_funcs.get_cpu_scaling) { cpu->pstate.scaling = pstate_funcs.get_cpu_scaling(cpu->cpu); - if (cpu->pstate.scaling != perf_ctl_scaling) + if (cpu->pstate.scaling != perf_ctl_scaling) { intel_pstate_hybrid_hwp_adjust(cpu); + hwp_is_hybrid = true; + } } else { cpu->pstate.scaling = perf_ctl_scaling; } @@ -2425,6 +2649,10 @@ static const struct x86_cpu_id intel_pstate_cpu_oob_ids[] __initconst = { X86_MATCH(INTEL_ICELAKE_X, core_funcs), X86_MATCH(INTEL_SAPPHIRERAPIDS_X, core_funcs), X86_MATCH(INTEL_EMERALDRAPIDS_X, core_funcs), + X86_MATCH(INTEL_GRANITERAPIDS_D, core_funcs), + X86_MATCH(INTEL_GRANITERAPIDS_X, core_funcs), + X86_MATCH(INTEL_ATOM_CRESTMONT, core_funcs), + X86_MATCH(INTEL_ATOM_CRESTMONT_X, core_funcs), {} }; #endif @@ -2703,6 +2931,8 @@ static int intel_pstate_cpu_online(struct cpufreq_policy *policy) */ intel_pstate_hwp_reenable(cpu); cpu->suspended = false; + + hybrid_update_capacity(cpu); } return 0; @@ -3143,6 +3373,8 @@ static int intel_pstate_register_driver(struct cpufreq_driver *driver) global.min_perf_pct = min_perf_pct_min(); + hybrid_init_cpu_capacity_scaling(); + return 0; } diff --git a/drivers/cpufreq/loongson3_cpufreq.c b/drivers/cpufreq/loongson3_cpufreq.c index 5f79b6de127c..6b5e6798d9a2 100644 --- a/drivers/cpufreq/loongson3_cpufreq.c +++ b/drivers/cpufreq/loongson3_cpufreq.c @@ -176,7 +176,7 @@ static DEFINE_PER_CPU(struct loongson3_freq_data *, freq_data); static inline int do_service_request(u32 id, u32 info, u32 cmd, u32 val, u32 extra) { int retries; - unsigned int cpu = smp_processor_id(); + unsigned int cpu = raw_smp_processor_id(); unsigned int package = cpu_data[cpu].package; union smc_message msg, last; diff --git a/drivers/cpufreq/maple-cpufreq.c b/drivers/cpufreq/maple-cpufreq.c index f9306410a07f..690da85c4865 100644 --- a/drivers/cpufreq/maple-cpufreq.c +++ b/drivers/cpufreq/maple-cpufreq.c @@ -238,4 +238,5 @@ bail_noprops: module_init(maple_cpufreq_init); +MODULE_DESCRIPTION("cpufreq driver for Maple 970FX/970MP boards"); MODULE_LICENSE("GPL"); diff --git a/drivers/cpufreq/mediatek-cpufreq.c b/drivers/cpufreq/mediatek-cpufreq.c index 3a1aadaa723c..663f61565cf7 100644 --- a/drivers/cpufreq/mediatek-cpufreq.c +++ b/drivers/cpufreq/mediatek-cpufreq.c @@ -738,7 +738,7 @@ static const struct mtk_cpufreq_platform_data mt8516_platform_data = { }; /* List of machines supported by this driver */ -static const struct of_device_id mtk_cpufreq_machines[] __initconst = { +static const struct of_device_id mtk_cpufreq_machines[] __initconst __maybe_unused = { { .compatible = "mediatek,mt2701", .data = &mt2701_platform_data }, { .compatible = "mediatek,mt2712", .data = &mt2701_platform_data }, { .compatible = "mediatek,mt7622", .data = &mt7622_platform_data }, diff --git a/drivers/cpufreq/omap-cpufreq.c b/drivers/cpufreq/omap-cpufreq.c index 3458d5cc9b7f..de8be0a8932d 100644 --- a/drivers/cpufreq/omap-cpufreq.c +++ b/drivers/cpufreq/omap-cpufreq.c @@ -28,9 +28,6 @@ #include <linux/platform_device.h> #include <linux/regulator/consumer.h> -#include <asm/smp_plat.h> -#include <asm/cpu.h> - /* OPP tolerance in percentage */ #define OPP_TOLERANCE 4 diff --git a/drivers/cpufreq/pasemi-cpufreq.c b/drivers/cpufreq/pasemi-cpufreq.c index ee925b53b6b9..5fc9cb480516 100644 --- a/drivers/cpufreq/pasemi-cpufreq.c +++ b/drivers/cpufreq/pasemi-cpufreq.c @@ -269,5 +269,6 @@ static void __exit pas_cpufreq_exit(void) module_init(pas_cpufreq_init); module_exit(pas_cpufreq_exit); +MODULE_DESCRIPTION("cpufreq driver for PA Semi PWRficient"); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Egor Martovetsky <egor@pasemi.com>, Olof Johansson <olof@lixom.net>"); diff --git a/drivers/cpufreq/pmac64-cpufreq.c b/drivers/cpufreq/pmac64-cpufreq.c index 2cd2b06849a2..74ff6c47df29 100644 --- a/drivers/cpufreq/pmac64-cpufreq.c +++ b/drivers/cpufreq/pmac64-cpufreq.c @@ -505,7 +505,7 @@ static int __init g5_pm72_cpufreq_init(struct device_node *cpunode) continue; if (strcmp(loc, "CPU CLOCK")) continue; - if (!of_get_property(hwclock, "platform-get-frequency", NULL)) + if (!of_property_present(hwclock, "platform-get-frequency")) continue; break; } @@ -671,4 +671,5 @@ static int __init g5_cpufreq_init(void) module_init(g5_cpufreq_init); +MODULE_DESCRIPTION("cpufreq driver for SMU & 970FX based G5 Macs"); MODULE_LICENSE("GPL"); diff --git a/drivers/cpufreq/powernv-cpufreq.c b/drivers/cpufreq/powernv-cpufreq.c index 50c62929f7ca..8de759247771 100644 --- a/drivers/cpufreq/powernv-cpufreq.c +++ b/drivers/cpufreq/powernv-cpufreq.c @@ -692,7 +692,7 @@ static void gpstate_timer_handler(struct timer_list *t) } /* - * If PMCR was last updated was using fast_swtich then + * If PMCR was last updated was using fast_switch then * We may have wrong in gpstate->last_lpstate_idx * value. Hence, read from PMCR to get correct data. */ @@ -1160,5 +1160,6 @@ static void __exit powernv_cpufreq_exit(void) } module_exit(powernv_cpufreq_exit); +MODULE_DESCRIPTION("cpufreq driver for IBM/OpenPOWER powernv systems"); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Vaidyanathan Srinivasan <svaidy at linux.vnet.ibm.com>"); diff --git a/drivers/cpufreq/ppc_cbe_cpufreq.c b/drivers/cpufreq/ppc_cbe_cpufreq.c index 5ee4c7bfdcc5..98595b3ea13f 100644 --- a/drivers/cpufreq/ppc_cbe_cpufreq.c +++ b/drivers/cpufreq/ppc_cbe_cpufreq.c @@ -168,5 +168,6 @@ static void __exit cbe_cpufreq_exit(void) module_init(cbe_cpufreq_init); module_exit(cbe_cpufreq_exit); +MODULE_DESCRIPTION("cpufreq driver for Cell BE processors"); MODULE_LICENSE("GPL"); MODULE_AUTHOR("Christian Krafft <krafft@de.ibm.com>"); diff --git a/drivers/cpufreq/qcom-cpufreq-hw.c b/drivers/cpufreq/qcom-cpufreq-hw.c index 370fe6a0104b..900d6844c43d 100644 --- a/drivers/cpufreq/qcom-cpufreq-hw.c +++ b/drivers/cpufreq/qcom-cpufreq-hw.c @@ -9,6 +9,7 @@ #include <linux/init.h> #include <linux/interconnect.h> #include <linux/interrupt.h> +#include <linux/io.h> #include <linux/kernel.h> #include <linux/module.h> #include <linux/of.h> diff --git a/drivers/cpufreq/qcom-cpufreq-nvmem.c b/drivers/cpufreq/qcom-cpufreq-nvmem.c index 939702dfa73f..703308fb891a 100644 --- a/drivers/cpufreq/qcom-cpufreq-nvmem.c +++ b/drivers/cpufreq/qcom-cpufreq-nvmem.c @@ -611,7 +611,7 @@ static struct platform_driver qcom_cpufreq_driver = { }, }; -static const struct of_device_id qcom_cpufreq_match_list[] __initconst = { +static const struct of_device_id qcom_cpufreq_match_list[] __initconst __maybe_unused = { { .compatible = "qcom,apq8096", .data = &match_data_kryo }, { .compatible = "qcom,msm8909", .data = &match_data_msm8909 }, { .compatible = "qcom,msm8996", .data = &match_data_kryo }, diff --git a/drivers/cpufreq/spear-cpufreq.c b/drivers/cpufreq/spear-cpufreq.c index 78b875db6b66..d8ab5b01d46d 100644 --- a/drivers/cpufreq/spear-cpufreq.c +++ b/drivers/cpufreq/spear-cpufreq.c @@ -171,10 +171,9 @@ static struct cpufreq_driver spear_cpufreq_driver = { static int spear_cpufreq_probe(struct platform_device *pdev) { struct device_node *np; - const struct property *prop; struct cpufreq_frequency_table *freq_tbl; - const __be32 *val; - int cnt, i, ret; + u32 val; + int cnt, ret, i = 0; np = of_cpu_device_node_get(0); if (!np) { @@ -186,26 +185,23 @@ static int spear_cpufreq_probe(struct platform_device *pdev) &spear_cpufreq.transition_latency)) spear_cpufreq.transition_latency = CPUFREQ_ETERNAL; - prop = of_find_property(np, "cpufreq_tbl", NULL); - if (!prop || !prop->value) { + cnt = of_property_count_u32_elems(np, "cpufreq_tbl"); + if (cnt <= 0) { pr_err("Invalid cpufreq_tbl\n"); ret = -ENODEV; goto out_put_node; } - cnt = prop->length / sizeof(u32); - val = prop->value; - freq_tbl = kcalloc(cnt + 1, sizeof(*freq_tbl), GFP_KERNEL); if (!freq_tbl) { ret = -ENOMEM; goto out_put_node; } - for (i = 0; i < cnt; i++) - freq_tbl[i].frequency = be32_to_cpup(val++); + of_property_for_each_u32(np, "cpufreq_tbl", val) + freq_tbl[i++].frequency = val; - freq_tbl[i].frequency = CPUFREQ_TABLE_END; + freq_tbl[cnt].frequency = CPUFREQ_TABLE_END; spear_cpufreq.freq_tbl = freq_tbl; diff --git a/drivers/cpufreq/sti-cpufreq.c b/drivers/cpufreq/sti-cpufreq.c index 8e2e703c3865..b15b3142b5fe 100644 --- a/drivers/cpufreq/sti-cpufreq.c +++ b/drivers/cpufreq/sti-cpufreq.c @@ -267,7 +267,7 @@ static int __init sti_cpufreq_init(void) goto skip_voltage_scaling; } - if (!of_get_property(ddata.cpu->of_node, "operating-points-v2", NULL)) { + if (!of_property_present(ddata.cpu->of_node, "operating-points-v2")) { dev_err(ddata.cpu, "OPP-v2 not supported\n"); goto skip_voltage_scaling; } diff --git a/drivers/cpufreq/sun50i-cpufreq-nvmem.c b/drivers/cpufreq/sun50i-cpufreq-nvmem.c index 95ac8d46c156..293921acec93 100644 --- a/drivers/cpufreq/sun50i-cpufreq-nvmem.c +++ b/drivers/cpufreq/sun50i-cpufreq-nvmem.c @@ -146,7 +146,7 @@ static bool dt_has_supported_hw(void) return false; for_each_child_of_node_scoped(np, opp) { - if (of_find_property(opp, "opp-supported-hw", NULL)) { + if (of_property_present(opp, "opp-supported-hw")) { has_opp_supported_hw = true; break; } diff --git a/drivers/cpufreq/ti-cpufreq.c b/drivers/cpufreq/ti-cpufreq.c index 4d3f27958fbd..ba621ce1cdda 100644 --- a/drivers/cpufreq/ti-cpufreq.c +++ b/drivers/cpufreq/ti-cpufreq.c @@ -16,6 +16,7 @@ #include <linux/pm_opp.h> #include <linux/regmap.h> #include <linux/slab.h> +#include <linux/sys_soc.h> #define REVISION_MASK 0xF #define REVISION_SHIFT 28 @@ -90,6 +91,9 @@ struct ti_cpufreq_soc_data { unsigned long efuse_shift; unsigned long rev_offset; bool multi_regulator; +/* Backward compatibility hack: Might have missing syscon */ +#define TI_QUIRK_SYSCON_MAY_BE_MISSING 0x1 + u8 quirks; }; struct ti_cpufreq_data { @@ -254,6 +258,7 @@ static struct ti_cpufreq_soc_data omap34xx_soc_data = { .efuse_mask = BIT(3), .rev_offset = OMAP3_CONTROL_IDCODE - OMAP3_SYSCON_BASE, .multi_regulator = false, + .quirks = TI_QUIRK_SYSCON_MAY_BE_MISSING, }; /* @@ -281,6 +286,7 @@ static struct ti_cpufreq_soc_data omap36xx_soc_data = { .efuse_mask = BIT(9), .rev_offset = OMAP3_CONTROL_IDCODE - OMAP3_SYSCON_BASE, .multi_regulator = true, + .quirks = TI_QUIRK_SYSCON_MAY_BE_MISSING, }; /* @@ -295,6 +301,14 @@ static struct ti_cpufreq_soc_data am3517_soc_data = { .efuse_mask = 0, .rev_offset = OMAP3_CONTROL_IDCODE - OMAP3_SYSCON_BASE, .multi_regulator = false, + .quirks = TI_QUIRK_SYSCON_MAY_BE_MISSING, +}; + +static const struct soc_device_attribute k3_cpufreq_soc[] = { + { .family = "AM62X", .revision = "SR1.0" }, + { .family = "AM62AX", .revision = "SR1.0" }, + { .family = "AM62PX", .revision = "SR1.0" }, + { /* sentinel */ } }; static struct ti_cpufreq_soc_data am625_soc_data = { @@ -340,7 +354,7 @@ static int ti_cpufreq_get_efuse(struct ti_cpufreq_data *opp_data, ret = regmap_read(opp_data->syscon, opp_data->soc_data->efuse_offset, &efuse); - if (ret == -EIO) { + if (opp_data->soc_data->quirks & TI_QUIRK_SYSCON_MAY_BE_MISSING && ret == -EIO) { /* not a syscon register! */ void __iomem *regs = ioremap(OMAP3_SYSCON_BASE + opp_data->soc_data->efuse_offset, 4); @@ -378,10 +392,20 @@ static int ti_cpufreq_get_rev(struct ti_cpufreq_data *opp_data, struct device *dev = opp_data->cpu_dev; u32 revision; int ret; + if (soc_device_match(k3_cpufreq_soc)) { + /* + * Since the SR is 1.0, hard code the revision_value as + * 0x1 here. This way we avoid re using the same register + * that is giving us required information inside socinfo + * anyway. + */ + *revision_value = 0x1; + goto done; + } ret = regmap_read(opp_data->syscon, opp_data->soc_data->rev_offset, &revision); - if (ret == -EIO) { + if (opp_data->soc_data->quirks & TI_QUIRK_SYSCON_MAY_BE_MISSING && ret == -EIO) { /* not a syscon register! */ void __iomem *regs = ioremap(OMAP3_SYSCON_BASE + opp_data->soc_data->rev_offset, 4); @@ -400,6 +424,7 @@ static int ti_cpufreq_get_rev(struct ti_cpufreq_data *opp_data, *revision_value = BIT((revision >> REVISION_SHIFT) & REVISION_MASK); +done: return 0; } @@ -419,7 +444,7 @@ static int ti_cpufreq_setup_syscon_register(struct ti_cpufreq_data *opp_data) return 0; } -static const struct of_device_id ti_cpufreq_of_match[] = { +static const struct of_device_id ti_cpufreq_of_match[] __maybe_unused = { { .compatible = "ti,am33xx", .data = &am3x_soc_data, }, { .compatible = "ti,am3517", .data = &am3517_soc_data, }, { .compatible = "ti,am43", .data = &am4x_soc_data, }, diff --git a/drivers/cpuidle/cpuidle-riscv-sbi.c b/drivers/cpuidle/cpuidle-riscv-sbi.c index a6e123dfe394..d228b4d18d56 100644 --- a/drivers/cpuidle/cpuidle-riscv-sbi.c +++ b/drivers/cpuidle/cpuidle-riscv-sbi.c @@ -8,6 +8,7 @@ #define pr_fmt(fmt) "cpuidle-riscv-sbi: " fmt +#include <linux/cleanup.h> #include <linux/cpuhotplug.h> #include <linux/cpuidle.h> #include <linux/cpumask.h> @@ -236,19 +237,16 @@ static int sbi_cpuidle_dt_init_states(struct device *dev, { struct sbi_cpuidle_data *data = per_cpu_ptr(&sbi_cpuidle_data, cpu); struct device_node *state_node; - struct device_node *cpu_node; u32 *states; int i, ret; - cpu_node = of_cpu_device_node_get(cpu); + struct device_node *cpu_node __free(device_node) = of_cpu_device_node_get(cpu); if (!cpu_node) return -ENODEV; states = devm_kcalloc(dev, state_count, sizeof(*states), GFP_KERNEL); - if (!states) { - ret = -ENOMEM; - goto fail; - } + if (!states) + return -ENOMEM; /* Parse SBI specific details from state DT nodes */ for (i = 1; i < state_count; i++) { @@ -264,10 +262,8 @@ static int sbi_cpuidle_dt_init_states(struct device *dev, pr_debug("sbi-state %#x index %d\n", states[i], i); } - if (i != state_count) { - ret = -ENODEV; - goto fail; - } + if (i != state_count) + return -ENODEV; /* Initialize optional data, used for the hierarchical topology. */ ret = sbi_dt_cpu_init_topology(drv, data, state_count, cpu); @@ -277,10 +273,7 @@ static int sbi_cpuidle_dt_init_states(struct device *dev, /* Store states in the per-cpu struct. */ data->states = states; -fail: - of_node_put(cpu_node); - - return ret; + return 0; } static void sbi_cpuidle_deinit_cpu(int cpu) @@ -455,7 +448,6 @@ static void sbi_pd_remove(void) static int sbi_genpd_probe(struct device_node *np) { - struct device_node *node; int ret = 0, pd_count = 0; if (!np) @@ -465,13 +457,13 @@ static int sbi_genpd_probe(struct device_node *np) * Parse child nodes for the "#power-domain-cells" property and * initialize a genpd/genpd-of-provider pair when it's found. */ - for_each_child_of_node(np, node) { + for_each_child_of_node_scoped(np, node) { if (!of_property_present(node, "#power-domain-cells")) continue; ret = sbi_pd_init(node); if (ret) - goto put_node; + goto remove_pd; pd_count++; } @@ -487,8 +479,6 @@ static int sbi_genpd_probe(struct device_node *np) return 0; -put_node: - of_node_put(node); remove_pd: sbi_pd_remove(); pr_err("failed to create CPU PM domains ret=%d\n", ret); diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c index 02e40fd7d948..9e418aec1755 100644 --- a/drivers/cpuidle/cpuidle.c +++ b/drivers/cpuidle/cpuidle.c @@ -228,10 +228,7 @@ noinstr int cpuidle_enter_state(struct cpuidle_device *dev, if (broadcast && tick_broadcast_enter()) { index = find_deepest_state(drv, dev, target_state->exit_latency_ns, CPUIDLE_FLAG_TIMER_STOP, false); - if (index < 0) { - default_idle_call(); - return -EBUSY; - } + target_state = &drv->states[index]; broadcast = false; } diff --git a/drivers/devfreq/exynos-bus.c b/drivers/devfreq/exynos-bus.c index 00118580905a..7d06c476d8e9 100644 --- a/drivers/devfreq/exynos-bus.c +++ b/drivers/devfreq/exynos-bus.c @@ -160,7 +160,6 @@ static void exynos_bus_exit(struct device *dev) platform_device_unregister(bus->icc_pdev); dev_pm_opp_of_remove_table(dev); - clk_disable_unprepare(bus->clk); dev_pm_opp_put_regulators(bus->opp_token); } @@ -171,7 +170,6 @@ static void exynos_bus_passive_exit(struct device *dev) platform_device_unregister(bus->icc_pdev); dev_pm_opp_of_remove_table(dev); - clk_disable_unprepare(bus->clk); } static int exynos_bus_parent_parse_of(struct device_node *np, @@ -247,23 +245,16 @@ static int exynos_bus_parse_of(struct device_node *np, int ret; /* Get the clock to provide each bus with source clock */ - bus->clk = devm_clk_get(dev, "bus"); - if (IS_ERR(bus->clk)) { - dev_err(dev, "failed to get bus clock\n"); - return PTR_ERR(bus->clk); - } - - ret = clk_prepare_enable(bus->clk); - if (ret < 0) { - dev_err(dev, "failed to get enable clock\n"); - return ret; - } + bus->clk = devm_clk_get_enabled(dev, "bus"); + if (IS_ERR(bus->clk)) + return dev_err_probe(dev, PTR_ERR(bus->clk), + "failed to get bus clock\n"); /* Get the freq and voltage from OPP table to scale the bus freq */ ret = dev_pm_opp_of_add_table(dev); if (ret < 0) { dev_err(dev, "failed to get OPP table\n"); - goto err_clk; + return ret; } rate = clk_get_rate(bus->clk); @@ -281,8 +272,6 @@ static int exynos_bus_parse_of(struct device_node *np, err_opp: dev_pm_opp_of_remove_table(dev); -err_clk: - clk_disable_unprepare(bus->clk); return ret; } @@ -453,7 +442,6 @@ static int exynos_bus_probe(struct platform_device *pdev) err: dev_pm_opp_of_remove_table(dev); - clk_disable_unprepare(bus->clk); err_reg: dev_pm_opp_put_regulators(bus->opp_token); diff --git a/drivers/devfreq/governor_performance.c b/drivers/devfreq/governor_performance.c index 5dbc1e56ec08..2e4e981446fa 100644 --- a/drivers/devfreq/governor_performance.c +++ b/drivers/devfreq/governor_performance.c @@ -58,4 +58,5 @@ static void __exit devfreq_performance_exit(void) return; } module_exit(devfreq_performance_exit); +MODULE_DESCRIPTION("DEVFREQ Performance governor"); MODULE_LICENSE("GPL"); diff --git a/drivers/devfreq/governor_powersave.c b/drivers/devfreq/governor_powersave.c index 4746af2435b0..f059e8814804 100644 --- a/drivers/devfreq/governor_powersave.c +++ b/drivers/devfreq/governor_powersave.c @@ -58,4 +58,5 @@ static void __exit devfreq_powersave_exit(void) return; } module_exit(devfreq_powersave_exit); +MODULE_DESCRIPTION("DEVFREQ Powersave governor"); MODULE_LICENSE("GPL"); diff --git a/drivers/devfreq/governor_simpleondemand.c b/drivers/devfreq/governor_simpleondemand.c index d57b82a2b570..c23435736367 100644 --- a/drivers/devfreq/governor_simpleondemand.c +++ b/drivers/devfreq/governor_simpleondemand.c @@ -140,4 +140,5 @@ static void __exit devfreq_simple_ondemand_exit(void) return; } module_exit(devfreq_simple_ondemand_exit); +MODULE_DESCRIPTION("DEVFREQ Simple On-demand governor"); MODULE_LICENSE("GPL"); diff --git a/drivers/devfreq/governor_userspace.c b/drivers/devfreq/governor_userspace.c index d69672ccacc4..d1aa6806b683 100644 --- a/drivers/devfreq/governor_userspace.c +++ b/drivers/devfreq/governor_userspace.c @@ -153,4 +153,5 @@ static void __exit devfreq_userspace_exit(void) return; } module_exit(devfreq_userspace_exit); +MODULE_DESCRIPTION("DEVFREQ Userspace governor"); MODULE_LICENSE("GPL"); diff --git a/drivers/devfreq/imx-bus.c b/drivers/devfreq/imx-bus.c index 86850b7dea09..49798f542d68 100644 --- a/drivers/devfreq/imx-bus.c +++ b/drivers/devfreq/imx-bus.c @@ -59,7 +59,7 @@ static int imx_bus_init_icc(struct device *dev) struct imx_bus *priv = dev_get_drvdata(dev); const char *icc_driver_name; - if (!of_get_property(dev->of_node, "#interconnect-cells", NULL)) + if (!of_property_present(dev->of_node, "#interconnect-cells")) return 0; if (!IS_ENABLED(CONFIG_INTERCONNECT_IMX)) { dev_warn(dev, "imx interconnect drivers disabled\n"); diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index 9aab7abc2ae9..9457e34b9e32 100644 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -1022,6 +1022,45 @@ static struct cpuidle_state spr_cstates[] __initdata = { .enter = NULL } }; +static struct cpuidle_state gnr_cstates[] __initdata = { + { + .name = "C1", + .desc = "MWAIT 0x00", + .flags = MWAIT2flg(0x00), + .exit_latency = 1, + .target_residency = 1, + .enter = &intel_idle, + .enter_s2idle = intel_idle_s2idle, }, + { + .name = "C1E", + .desc = "MWAIT 0x01", + .flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE, + .exit_latency = 4, + .target_residency = 4, + .enter = &intel_idle, + .enter_s2idle = intel_idle_s2idle, }, + { + .name = "C6", + .desc = "MWAIT 0x20", + .flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED | + CPUIDLE_FLAG_INIT_XSTATE, + .exit_latency = 170, + .target_residency = 650, + .enter = &intel_idle, + .enter_s2idle = intel_idle_s2idle, }, + { + .name = "C6P", + .desc = "MWAIT 0x21", + .flags = MWAIT2flg(0x21) | CPUIDLE_FLAG_TLB_FLUSHED | + CPUIDLE_FLAG_INIT_XSTATE, + .exit_latency = 210, + .target_residency = 1000, + .enter = &intel_idle, + .enter_s2idle = intel_idle_s2idle, }, + { + .enter = NULL } +}; + static struct cpuidle_state atom_cstates[] __initdata = { { .name = "C1E", @@ -1453,6 +1492,12 @@ static const struct idle_cpu idle_cpu_spr __initconst = { .use_acpi = true, }; +static const struct idle_cpu idle_cpu_gnr __initconst = { + .state_table = gnr_cstates, + .disable_promotion_to_c1e = true, + .use_acpi = true, +}; + static const struct idle_cpu idle_cpu_avn __initconst = { .state_table = avn_cstates, .disable_promotion_to_c1e = true, @@ -1475,6 +1520,10 @@ static const struct idle_cpu idle_cpu_dnv __initconst = { .use_acpi = true, }; +static const struct idle_cpu idle_cpu_tmt __initconst = { + .disable_promotion_to_c1e = true, +}; + static const struct idle_cpu idle_cpu_snr __initconst = { .state_table = snr_cstates, .disable_promotion_to_c1e = true, @@ -1533,11 +1582,14 @@ static const struct x86_cpu_id intel_idle_ids[] __initconst = { X86_MATCH_VFM(INTEL_ATOM_GRACEMONT, &idle_cpu_gmt), X86_MATCH_VFM(INTEL_SAPPHIRERAPIDS_X, &idle_cpu_spr), X86_MATCH_VFM(INTEL_EMERALDRAPIDS_X, &idle_cpu_spr), + X86_MATCH_VFM(INTEL_GRANITERAPIDS_X, &idle_cpu_gnr), X86_MATCH_VFM(INTEL_XEON_PHI_KNL, &idle_cpu_knl), X86_MATCH_VFM(INTEL_XEON_PHI_KNM, &idle_cpu_knl), X86_MATCH_VFM(INTEL_ATOM_GOLDMONT, &idle_cpu_bxt), X86_MATCH_VFM(INTEL_ATOM_GOLDMONT_PLUS, &idle_cpu_bxt), X86_MATCH_VFM(INTEL_ATOM_GOLDMONT_D, &idle_cpu_dnv), + X86_MATCH_VFM(INTEL_ATOM_TREMONT, &idle_cpu_tmt), + X86_MATCH_VFM(INTEL_ATOM_TREMONT_L, &idle_cpu_tmt), X86_MATCH_VFM(INTEL_ATOM_TREMONT_D, &idle_cpu_snr), X86_MATCH_VFM(INTEL_ATOM_CRESTMONT, &idle_cpu_grr), X86_MATCH_VFM(INTEL_ATOM_CRESTMONT_X, &idle_cpu_srf), @@ -2075,7 +2127,7 @@ static void __init intel_idle_cpuidle_driver_init(struct cpuidle_driver *drv) drv->state_count = 1; - if (icpu) + if (icpu && icpu->state_table) intel_idle_init_cstates_icpu(drv); else intel_idle_init_cstates_acpi(drv); @@ -2209,7 +2261,11 @@ static int __init intel_idle_init(void) icpu = (const struct idle_cpu *)id->driver_data; if (icpu) { - cpuidle_state_table = icpu->state_table; + if (icpu->state_table) + cpuidle_state_table = icpu->state_table; + else if (!intel_idle_acpi_cst_extract()) + return -ENODEV; + auto_demotion_disable_flags = icpu->auto_demotion_disable_flags; if (icpu->disable_promotion_to_c1e) c1e_promotion = C1E_PROMOTION_DISABLE; diff --git a/drivers/opp/ti-opp-supply.c b/drivers/opp/ti-opp-supply.c index ec0056a4bb13..5f0fb3ea385b 100644 --- a/drivers/opp/ti-opp-supply.c +++ b/drivers/opp/ti-opp-supply.c @@ -405,7 +405,7 @@ static struct platform_driver ti_opp_supply_driver = { .probe = ti_opp_supply_probe, .driver = { .name = "ti_opp_supply", - .of_match_table = of_match_ptr(ti_opp_supply_of_match), + .of_match_table = ti_opp_supply_of_match, }, }; module_platform_driver(ti_opp_supply_driver); diff --git a/drivers/powercap/intel_rapl_common.c b/drivers/powercap/intel_rapl_common.c index 3cffa6c79538..5e793b80fd6b 100644 --- a/drivers/powercap/intel_rapl_common.c +++ b/drivers/powercap/intel_rapl_common.c @@ -740,7 +740,7 @@ static struct rapl_primitive_info *get_rpi(struct rapl_package *rp, int prim) { struct rapl_primitive_info *rpi = rp->priv->rpi; - if (prim < 0 || prim > NR_RAPL_PRIMITIVES || !rpi) + if (prim < 0 || prim >= NR_RAPL_PRIMITIVES || !rpi) return NULL; return &rpi[prim]; @@ -1267,6 +1267,7 @@ static const struct x86_cpu_id rapl_ids[] __initconst = { X86_MATCH_VFM(INTEL_LUNARLAKE_M, &rapl_defaults_core), X86_MATCH_VFM(INTEL_ARROWLAKE_H, &rapl_defaults_core), X86_MATCH_VFM(INTEL_ARROWLAKE, &rapl_defaults_core), + X86_MATCH_VFM(INTEL_ARROWLAKE_U, &rapl_defaults_core), X86_MATCH_VFM(INTEL_LAKEFIELD, &rapl_defaults_core), X86_MATCH_VFM(INTEL_ATOM_SILVERMONT, &rapl_defaults_byt), @@ -1285,6 +1286,7 @@ static const struct x86_cpu_id rapl_ids[] __initconst = { X86_MATCH_VENDOR_FAM(AMD, 0x17, &rapl_defaults_amd), X86_MATCH_VENDOR_FAM(AMD, 0x19, &rapl_defaults_amd), + X86_MATCH_VENDOR_FAM(AMD, 0x1A, &rapl_defaults_amd), X86_MATCH_VENDOR_FAM(HYGON, 0x18, &rapl_defaults_amd), {} }; @@ -2128,6 +2130,21 @@ void rapl_remove_package(struct rapl_package *rp) } EXPORT_SYMBOL_GPL(rapl_remove_package); +/* + * RAPL Package energy counter scope: + * 1. AMD/HYGON platforms use per-PKG package energy counter + * 2. For Intel platforms + * 2.1 CLX-AP platform has per-DIE package energy counter + * 2.2 Other platforms that uses MSR RAPL are single die systems so the + * package energy counter can be considered as per-PKG/per-DIE, + * here it is considered as per-DIE. + * 2.3 New platforms that use TPMI RAPL doesn't care about the + * scope because they are not MSR/CPU based. + */ +#define rapl_msrs_are_pkg_scope() \ + (boot_cpu_data.x86_vendor == X86_VENDOR_AMD || \ + boot_cpu_data.x86_vendor == X86_VENDOR_HYGON) + /* caller to ensure CPU hotplug lock is held */ struct rapl_package *rapl_find_package_domain_cpuslocked(int id, struct rapl_if_priv *priv, bool id_is_cpu) @@ -2135,8 +2152,14 @@ struct rapl_package *rapl_find_package_domain_cpuslocked(int id, struct rapl_if_ struct rapl_package *rp; int uid; - if (id_is_cpu) - uid = topology_logical_die_id(id); + if (id_is_cpu) { + uid = rapl_msrs_are_pkg_scope() ? + topology_physical_package_id(id) : topology_logical_die_id(id); + if (uid < 0) { + pr_err("topology_logical_(package/die)_id() returned a negative value"); + return NULL; + } + } else uid = id; @@ -2168,9 +2191,14 @@ struct rapl_package *rapl_add_package_cpuslocked(int id, struct rapl_if_priv *pr return ERR_PTR(-ENOMEM); if (id_is_cpu) { - rp->id = topology_logical_die_id(id); + rp->id = rapl_msrs_are_pkg_scope() ? + topology_physical_package_id(id) : topology_logical_die_id(id); + if ((int)(rp->id) < 0) { + pr_err("topology_logical_(package/die)_id() returned a negative value"); + return ERR_PTR(-EINVAL); + } rp->lead_cpu = id; - if (topology_max_dies_per_package() > 1) + if (!rapl_msrs_are_pkg_scope() && topology_max_dies_per_package() > 1) snprintf(rp->name, PACKAGE_DOMAIN_NAME_LENGTH, "package-%d-die-%d", topology_physical_package_id(id), topology_die_id(id)); else diff --git a/include/acpi/cppc_acpi.h b/include/acpi/cppc_acpi.h index e1720d930666..76e44e102780 100644 --- a/include/acpi/cppc_acpi.h +++ b/include/acpi/cppc_acpi.h @@ -161,34 +161,37 @@ extern int cppc_get_epp_perf(int cpunum, u64 *epp_perf); extern int cppc_set_epp_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls, bool enable); extern int cppc_get_auto_sel_caps(int cpunum, struct cppc_perf_caps *perf_caps); extern int cppc_set_auto_sel(int cpu, bool enable); +extern int amd_get_highest_perf(unsigned int cpu, u32 *highest_perf); +extern int amd_get_boost_ratio_numerator(unsigned int cpu, u64 *numerator); +extern int amd_detect_prefcore(bool *detected); #else /* !CONFIG_ACPI_CPPC_LIB */ static inline int cppc_get_desired_perf(int cpunum, u64 *desired_perf) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int cppc_get_nominal_perf(int cpunum, u64 *nominal_perf) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int cppc_get_highest_perf(int cpunum, u64 *highest_perf) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int cppc_get_perf_ctrs(int cpu, struct cppc_perf_fb_ctrs *perf_fb_ctrs) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int cppc_set_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int cppc_set_enable(int cpu, bool enable) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int cppc_get_perf_caps(int cpu, struct cppc_perf_caps *caps) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline bool cppc_perf_ctrs_in_pcc(void) { @@ -212,27 +215,39 @@ static inline bool cpc_ffh_supported(void) } static inline int cpc_read_ffh(int cpunum, struct cpc_reg *reg, u64 *val) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int cpc_write_ffh(int cpunum, struct cpc_reg *reg, u64 val) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int cppc_set_epp_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls, bool enable) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int cppc_get_epp_perf(int cpunum, u64 *epp_perf) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int cppc_set_auto_sel(int cpu, bool enable) { - return -ENOTSUPP; + return -EOPNOTSUPP; } static inline int cppc_get_auto_sel_caps(int cpunum, struct cppc_perf_caps *perf_caps) { - return -ENOTSUPP; + return -EOPNOTSUPP; +} +static inline int amd_get_highest_perf(unsigned int cpu, u32 *highest_perf) +{ + return -ENODEV; +} +static inline int amd_get_boost_ratio_numerator(unsigned int cpu, u64 *numerator) +{ + return -EOPNOTSUPP; +} +static inline int amd_detect_prefcore(bool *detected) +{ + return -ENODEV; } #endif /* !CONFIG_ACPI_CPPC_LIB */ diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h index d4d2f4d1d7cb..e0e19d9c1323 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -577,12 +577,6 @@ static inline unsigned long cpufreq_scale(unsigned long old, u_int div, #define CPUFREQ_POLICY_POWERSAVE (1) #define CPUFREQ_POLICY_PERFORMANCE (2) -/* - * The polling frequency depends on the capability of the processor. Default - * polling frequency is 1000 times the transition latency of the processor. - */ -#define LATENCY_MULTIPLIER (1000) - struct cpufreq_governor { char name[CPUFREQ_NAME_LEN]; int (*init)(struct cpufreq_policy *policy); diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 0a213f69a9e4..e35829d36039 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -1123,11 +1123,11 @@ static const char * const hibernation_modes[] = { static ssize_t disk_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { + ssize_t count = 0; int i; - char *start = buf; if (!hibernation_available()) - return sprintf(buf, "[disabled]\n"); + return sysfs_emit(buf, "[disabled]\n"); for (i = HIBERNATION_FIRST; i <= HIBERNATION_MAX; i++) { if (!hibernation_modes[i]) @@ -1147,12 +1147,16 @@ static ssize_t disk_show(struct kobject *kobj, struct kobj_attribute *attr, continue; } if (i == hibernation_mode) - buf += sprintf(buf, "[%s] ", hibernation_modes[i]); + count += sysfs_emit_at(buf, count, "[%s] ", hibernation_modes[i]); else - buf += sprintf(buf, "%s ", hibernation_modes[i]); + count += sysfs_emit_at(buf, count, "%s ", hibernation_modes[i]); } - buf += sprintf(buf, "\n"); - return buf-start; + + /* Convert the last space to a newline if needed. */ + if (count > 0) + buf[count - 1] = '\n'; + + return count; } static ssize_t disk_store(struct kobject *kobj, struct kobj_attribute *attr, @@ -1210,8 +1214,8 @@ power_attr(disk); static ssize_t resume_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - return sprintf(buf, "%d:%d\n", MAJOR(swsusp_resume_device), - MINOR(swsusp_resume_device)); + return sysfs_emit(buf, "%d:%d\n", MAJOR(swsusp_resume_device), + MINOR(swsusp_resume_device)); } static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr, @@ -1270,7 +1274,7 @@ power_attr(resume); static ssize_t resume_offset_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - return sprintf(buf, "%llu\n", (unsigned long long)swsusp_resume_block); + return sysfs_emit(buf, "%llu\n", (unsigned long long)swsusp_resume_block); } static ssize_t resume_offset_store(struct kobject *kobj, @@ -1293,7 +1297,7 @@ power_attr(resume_offset); static ssize_t image_size_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - return sprintf(buf, "%lu\n", image_size); + return sysfs_emit(buf, "%lu\n", image_size); } static ssize_t image_size_store(struct kobject *kobj, struct kobj_attribute *attr, @@ -1314,7 +1318,7 @@ power_attr(image_size); static ssize_t reserved_size_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - return sprintf(buf, "%lu\n", reserved_size); + return sysfs_emit(buf, "%lu\n", reserved_size); } static ssize_t reserved_size_store(struct kobject *kobj, diff --git a/kernel/power/main.c b/kernel/power/main.c index a9e0693aaf69..6254814d4817 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -115,7 +115,7 @@ int pm_async_enabled = 1; static ssize_t pm_async_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - return sprintf(buf, "%d\n", pm_async_enabled); + return sysfs_emit(buf, "%d\n", pm_async_enabled); } static ssize_t pm_async_store(struct kobject *kobj, struct kobj_attribute *attr, @@ -139,7 +139,7 @@ power_attr(pm_async); static ssize_t mem_sleep_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - char *s = buf; + ssize_t count = 0; suspend_state_t i; for (i = PM_SUSPEND_MIN; i < PM_SUSPEND_MAX; i++) { @@ -149,17 +149,17 @@ static ssize_t mem_sleep_show(struct kobject *kobj, struct kobj_attribute *attr, const char *label = mem_sleep_states[i]; if (mem_sleep_current == i) - s += sprintf(s, "[%s] ", label); + count += sysfs_emit_at(buf, count, "[%s] ", label); else - s += sprintf(s, "%s ", label); + count += sysfs_emit_at(buf, count, "%s ", label); } } /* Convert the last space to a newline if needed. */ - if (s != buf) - *(s-1) = '\n'; + if (count > 0) + buf[count - 1] = '\n'; - return (s - buf); + return count; } static suspend_state_t decode_suspend_state(const char *buf, size_t n) @@ -220,7 +220,7 @@ bool sync_on_suspend_enabled = !IS_ENABLED(CONFIG_SUSPEND_SKIP_SYNC); static ssize_t sync_on_suspend_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - return sprintf(buf, "%d\n", sync_on_suspend_enabled); + return sysfs_emit(buf, "%d\n", sync_on_suspend_enabled); } static ssize_t sync_on_suspend_store(struct kobject *kobj, @@ -257,22 +257,22 @@ static const char * const pm_tests[__TEST_AFTER_LAST] = { static ssize_t pm_test_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - char *s = buf; + ssize_t count = 0; int level; for (level = TEST_FIRST; level <= TEST_MAX; level++) if (pm_tests[level]) { if (level == pm_test_level) - s += sprintf(s, "[%s] ", pm_tests[level]); + count += sysfs_emit_at(buf, count, "[%s] ", pm_tests[level]); else - s += sprintf(s, "%s ", pm_tests[level]); + count += sysfs_emit_at(buf, count, "%s ", pm_tests[level]); } - if (s != buf) - /* convert the last space to a newline */ - *(s-1) = '\n'; + /* Convert the last space to a newline if needed. */ + if (count > 0) + buf[count - 1] = '\n'; - return (s - buf); + return count; } static ssize_t pm_test_store(struct kobject *kobj, struct kobj_attribute *attr, @@ -390,7 +390,7 @@ static const char * const suspend_step_names[] = { static ssize_t _name##_show(struct kobject *kobj, \ struct kobj_attribute *attr, char *buf) \ { \ - return sprintf(buf, format_str, suspend_stats._name); \ + return sysfs_emit(buf, format_str, suspend_stats._name);\ } \ static struct kobj_attribute _name = __ATTR_RO(_name) @@ -404,7 +404,7 @@ suspend_attr(max_hw_sleep, "%llu\n"); static ssize_t _name##_show(struct kobject *kobj, \ struct kobj_attribute *attr, char *buf) \ { \ - return sprintf(buf, "%u\n", \ + return sysfs_emit(buf, "%u\n", \ suspend_stats.step_failures[step-1]); \ } \ static struct kobj_attribute _name = __ATTR_RO(_name) @@ -428,7 +428,7 @@ static ssize_t last_failed_dev_show(struct kobject *kobj, index %= REC_FAILED_NUM; last_failed_dev = suspend_stats.failed_devs[index]; - return sprintf(buf, "%s\n", last_failed_dev); + return sysfs_emit(buf, "%s\n", last_failed_dev); } static struct kobj_attribute last_failed_dev = __ATTR_RO(last_failed_dev); @@ -442,7 +442,7 @@ static ssize_t last_failed_errno_show(struct kobject *kobj, index %= REC_FAILED_NUM; last_failed_errno = suspend_stats.errno[index]; - return sprintf(buf, "%d\n", last_failed_errno); + return sysfs_emit(buf, "%d\n", last_failed_errno); } static struct kobj_attribute last_failed_errno = __ATTR_RO(last_failed_errno); @@ -456,7 +456,7 @@ static ssize_t last_failed_step_show(struct kobject *kobj, index %= REC_FAILED_NUM; step = suspend_stats.failed_steps[index]; - return sprintf(buf, "%s\n", suspend_step_names[step]); + return sysfs_emit(buf, "%s\n", suspend_step_names[step]); } static struct kobj_attribute last_failed_step = __ATTR_RO(last_failed_step); @@ -571,7 +571,7 @@ bool pm_print_times_enabled; static ssize_t pm_print_times_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - return sprintf(buf, "%d\n", pm_print_times_enabled); + return sysfs_emit(buf, "%d\n", pm_print_times_enabled); } static ssize_t pm_print_times_store(struct kobject *kobj, @@ -604,7 +604,7 @@ static ssize_t pm_wakeup_irq_show(struct kobject *kobj, if (!pm_wakeup_irq()) return -ENODATA; - return sprintf(buf, "%u\n", pm_wakeup_irq()); + return sysfs_emit(buf, "%u\n", pm_wakeup_irq()); } power_attr_ro(pm_wakeup_irq); @@ -620,7 +620,7 @@ EXPORT_SYMBOL_GPL(pm_debug_messages_should_print); static ssize_t pm_debug_messages_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - return sprintf(buf, "%d\n", pm_debug_messages_on); + return sysfs_emit(buf, "%d\n", pm_debug_messages_on); } static ssize_t pm_debug_messages_store(struct kobject *kobj, @@ -668,21 +668,23 @@ struct kobject *power_kobj; static ssize_t state_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - char *s = buf; + ssize_t count = 0; #ifdef CONFIG_SUSPEND suspend_state_t i; for (i = PM_SUSPEND_MIN; i < PM_SUSPEND_MAX; i++) if (pm_states[i]) - s += sprintf(s,"%s ", pm_states[i]); + count += sysfs_emit_at(buf, count, "%s ", pm_states[i]); #endif if (hibernation_available()) - s += sprintf(s, "disk "); - if (s != buf) - /* convert the last space to a newline */ - *(s-1) = '\n'; - return (s - buf); + count += sysfs_emit_at(buf, count, "disk "); + + /* Convert the last space to a newline if needed. */ + if (count > 0) + buf[count - 1] = '\n'; + + return count; } static suspend_state_t decode_state(const char *buf, size_t n) @@ -782,7 +784,7 @@ static ssize_t wakeup_count_show(struct kobject *kobj, unsigned int val; return pm_get_wakeup_count(&val, true) ? - sprintf(buf, "%u\n", val) : -EINTR; + sysfs_emit(buf, "%u\n", val) : -EINTR; } static ssize_t wakeup_count_store(struct kobject *kobj, @@ -824,17 +826,17 @@ static ssize_t autosleep_show(struct kobject *kobj, suspend_state_t state = pm_autosleep_state(); if (state == PM_SUSPEND_ON) - return sprintf(buf, "off\n"); + return sysfs_emit(buf, "off\n"); #ifdef CONFIG_SUSPEND if (state < PM_SUSPEND_MAX) - return sprintf(buf, "%s\n", pm_states[state] ? + return sysfs_emit(buf, "%s\n", pm_states[state] ? pm_states[state] : "error"); #endif #ifdef CONFIG_HIBERNATION - return sprintf(buf, "disk\n"); + return sysfs_emit(buf, "disk\n"); #else - return sprintf(buf, "error"); + return sysfs_emit(buf, "error\n"); #endif } @@ -903,7 +905,7 @@ int pm_trace_enabled; static ssize_t pm_trace_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - return sprintf(buf, "%d\n", pm_trace_enabled); + return sysfs_emit(buf, "%d\n", pm_trace_enabled); } static ssize_t @@ -940,7 +942,7 @@ power_attr_ro(pm_trace_dev_match); static ssize_t pm_freeze_timeout_show(struct kobject *kobj, struct kobj_attribute *attr, char *buf) { - return sprintf(buf, "%u\n", freeze_timeout_msecs); + return sysfs_emit(buf, "%u\n", freeze_timeout_msecs); } static ssize_t pm_freeze_timeout_store(struct kobject *kobj, diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 405eddbda4fc..30894d8f0a78 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -1365,11 +1365,6 @@ static unsigned int count_highmem_pages(void) } return n; } -#else -static inline void *saveable_highmem_page(struct zone *z, unsigned long p) -{ - return NULL; -} #endif /* CONFIG_HIGHMEM */ /** diff --git a/tools/power/cpupower/bindings/python/.gitignore b/tools/power/cpupower/bindings/python/.gitignore new file mode 100644 index 000000000000..5c9a1f0212dd --- /dev/null +++ b/tools/power/cpupower/bindings/python/.gitignore @@ -0,0 +1,8 @@ +__pycache__/ +raw_pylibcpupower_wrap.c +*.o +*.so +*.py +!test_raw_pylibcpupower.py +# git keeps ignoring this file, use git add -f raw_libcpupower.i +!raw_pylibcpupower.i diff --git a/tools/power/cpupower/bindings/python/Makefile b/tools/power/cpupower/bindings/python/Makefile new file mode 100644 index 000000000000..dc09c5b66ead --- /dev/null +++ b/tools/power/cpupower/bindings/python/Makefile @@ -0,0 +1,33 @@ +# SPDX-License-Identifier: GPL-2.0-only +# Makefile for libcpupower's Python bindings +# +# This Makefile expects you have already run the makefile for cpupower to build +# the .o files in the lib directory for the bindings to be created. + +CC := gcc +HAVE_SWIG := $(shell if which swig >/dev/null 2>&1; then echo 1; else echo 0; fi) +HAVE_PYCONFIG := $(shell if which python-config >/dev/null 2>&1; then echo 1; else echo 0; fi) + +LIB_DIR := ../../lib +PY_INCLUDE = $(firstword $(shell python-config --includes)) +OBJECTS_LIB = $(wildcard $(LIB_DIR)/*.o) + +all: _raw_pylibcpupower.so + +_raw_pylibcpupower.so: raw_pylibcpupower_wrap.o + $(CC) -shared $(OBJECTS_LIB) raw_pylibcpupower_wrap.o -o _raw_pylibcpupower.so + +raw_pylibcpupower_wrap.o: raw_pylibcpupower_wrap.c + $(CC) -fPIC -c raw_pylibcpupower_wrap.c $(PY_INCLUDE) + +raw_pylibcpupower_wrap.c: raw_pylibcpupower.i +ifeq ($(HAVE_SWIG),0) + $(error "swig was not found. Make sure you have it installed and in the PATH to generate the bindings.") +else ifeq ($(HAVE_PYCONFIG),0) + $(error "python-config was not found. Make sure you have it installed and in the PATH to generate the bindings.") +endif + swig -python raw_pylibcpupower.i + +# Will only clean the bindings folder; will not clean the actual cpupower folder +clean: + rm -f raw_pylibcpupower.py raw_pylibcpupower_wrap.c raw_pylibcpupower_wrap.o _raw_pylibcpupower.so diff --git a/tools/power/cpupower/bindings/python/README b/tools/power/cpupower/bindings/python/README new file mode 100644 index 000000000000..0a4bb2581e8a --- /dev/null +++ b/tools/power/cpupower/bindings/python/README @@ -0,0 +1,59 @@ +This folder contains the necessary files to build the Python bindings for +libcpupower (aside from the libcpupower object files). + + +requirements +------------ + +* You need the object files in the libcpupower directory compiled by +cpupower's makefile. +* The SWIG program must be installed. +* The Python's development libraries installed. + +Please check that your version of SWIG is compatible with the version of Python +installed on your machine by checking the SWIG changelog on their website. +https://swig.org/ + +Note that while SWIG itself is GPL v3+ licensed; the resulting output, +the bindings code: is permissively licensed + the license of libcpupower's .o +files. For these bindings that means GPL v2. + +Please see https://swig.org/legal.html and the discussion [1] for more details. + +[1] +https://lore.kernel.org/linux-pm/Zqv9BOjxLAgyNP5B@hatbackup/ + + +build +----- + +Install SWIG and the Python development files provided by your distribution. + +Build the object files for libcpupower by running make in the cpupower +directory. + +Return to the directory this README is in to run: + +$ make + + +testing +------- + +Please verify the _raw_pylibcpupower.so and raw_pylibcpupower.py files have +been created. + +To run the test script: + +$ python test_raw_pylibcpupower.py + + +credits +------- + +Original Bindings Author: +John B. Wyatt IV +jwyatt@redhat.com +sageofredondo@gmail.com + +Copyright (C) 2024 Red Hat diff --git a/tools/power/cpupower/bindings/python/raw_pylibcpupower.i b/tools/power/cpupower/bindings/python/raw_pylibcpupower.i new file mode 100644 index 000000000000..96556d87a745 --- /dev/null +++ b/tools/power/cpupower/bindings/python/raw_pylibcpupower.i @@ -0,0 +1,247 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ + +%module raw_pylibcpupower +%{ +#include "../../lib/cpupower_intern.h" +#include "../../lib/acpi_cppc.h" +#include "../../lib/cpufreq.h" +#include "../../lib/cpuidle.h" +#include "../../lib/cpupower.h" +#include "../../lib/powercap.h" +%} + +/* + * cpupower_intern.h + */ + +#define PATH_TO_CPU "/sys/devices/system/cpu/" +#define MAX_LINE_LEN 4096 +#define SYSFS_PATH_MAX 255 + +int is_valid_path(const char *path); + +unsigned int cpupower_read_sysfs(const char *path, char *buf, size_t buflen); + +unsigned int cpupower_write_sysfs(const char *path, char *buf, size_t buflen); + +/* + * acpi_cppc.h + */ + +enum acpi_cppc_value { + HIGHEST_PERF, + LOWEST_PERF, + NOMINAL_PERF, + LOWEST_NONLINEAR_PERF, + LOWEST_FREQ, + NOMINAL_FREQ, + REFERENCE_PERF, + WRAPAROUND_TIME, + MAX_CPPC_VALUE_FILES +}; + +unsigned long acpi_cppc_get_data(unsigned int cpu, + enum acpi_cppc_value which); + +/* + * cpufreq.h + */ + +struct cpufreq_policy { + unsigned long min; + unsigned long max; + char *governor; +}; + +struct cpufreq_available_governors { + char *governor; + struct cpufreq_available_governors *next; + struct cpufreq_available_governors *first; +}; + +struct cpufreq_available_frequencies { + unsigned long frequency; + struct cpufreq_available_frequencies *next; + struct cpufreq_available_frequencies *first; +}; + + +struct cpufreq_affected_cpus { + unsigned int cpu; + struct cpufreq_affected_cpus *next; + struct cpufreq_affected_cpus *first; +}; + +struct cpufreq_stats { + unsigned long frequency; + unsigned long long time_in_state; + struct cpufreq_stats *next; + struct cpufreq_stats *first; +}; + +unsigned long cpufreq_get_freq_kernel(unsigned int cpu); + +unsigned long cpufreq_get_freq_hardware(unsigned int cpu); + +#define cpufreq_get(cpu) cpufreq_get_freq_kernel(cpu); + +unsigned long cpufreq_get_transition_latency(unsigned int cpu); + +int cpufreq_get_hardware_limits(unsigned int cpu, + unsigned long *min, + unsigned long *max); + +char *cpufreq_get_driver(unsigned int cpu); + +void cpufreq_put_driver(char *ptr); + +struct cpufreq_policy *cpufreq_get_policy(unsigned int cpu); + +void cpufreq_put_policy(struct cpufreq_policy *policy); + +struct cpufreq_available_governors +*cpufreq_get_available_governors(unsigned int cpu); + +void cpufreq_put_available_governors( + struct cpufreq_available_governors *first); + +struct cpufreq_available_frequencies +*cpufreq_get_available_frequencies(unsigned int cpu); + +void cpufreq_put_available_frequencies( + struct cpufreq_available_frequencies *first); + +struct cpufreq_available_frequencies +*cpufreq_get_boost_frequencies(unsigned int cpu); + +void cpufreq_put_boost_frequencies( + struct cpufreq_available_frequencies *first); + +struct cpufreq_affected_cpus *cpufreq_get_affected_cpus(unsigned + int cpu); + +void cpufreq_put_affected_cpus(struct cpufreq_affected_cpus *first); + +struct cpufreq_affected_cpus *cpufreq_get_related_cpus(unsigned + int cpu); + +void cpufreq_put_related_cpus(struct cpufreq_affected_cpus *first); + +struct cpufreq_stats *cpufreq_get_stats(unsigned int cpu, + unsigned long long *total_time); + +void cpufreq_put_stats(struct cpufreq_stats *stats); + +unsigned long cpufreq_get_transitions(unsigned int cpu); + +int cpufreq_set_policy(unsigned int cpu, struct cpufreq_policy *policy); + +int cpufreq_modify_policy_min(unsigned int cpu, unsigned long min_freq); + +int cpufreq_modify_policy_max(unsigned int cpu, unsigned long max_freq); + +int cpufreq_modify_policy_governor(unsigned int cpu, char *governor); + +int cpufreq_set_frequency(unsigned int cpu, + unsigned long target_frequency); + +unsigned long cpufreq_get_sysfs_value_from_table(unsigned int cpu, + const char **table, + unsigned int index, + unsigned int size); + +/* + * cpuidle.h + */ + +int cpuidle_is_state_disabled(unsigned int cpu, + unsigned int idlestate); +int cpuidle_state_disable(unsigned int cpu, unsigned int idlestate, + unsigned int disable); +unsigned long cpuidle_state_latency(unsigned int cpu, + unsigned int idlestate); +unsigned long cpuidle_state_usage(unsigned int cpu, + unsigned int idlestate); +unsigned long long cpuidle_state_time(unsigned int cpu, + unsigned int idlestate); +char *cpuidle_state_name(unsigned int cpu, + unsigned int idlestate); +char *cpuidle_state_desc(unsigned int cpu, + unsigned int idlestate); +unsigned int cpuidle_state_count(unsigned int cpu); + +char *cpuidle_get_governor(void); + +char *cpuidle_get_driver(void); + +/* + * cpupower.h + */ + +struct cpupower_topology { + /* Amount of CPU cores, packages and threads per core in the system */ + unsigned int cores; + unsigned int pkgs; + unsigned int threads; /* per core */ + + /* Array gets mallocated with cores entries, holding per core info */ + struct cpuid_core_info *core_info; +}; + +struct cpuid_core_info { + int pkg; + int core; + int cpu; + + /* flags */ + unsigned int is_online:1; +}; + +int get_cpu_topology(struct cpupower_topology *cpu_top); + +void cpu_topology_release(struct cpupower_topology cpu_top); + +int cpupower_is_cpu_online(unsigned int cpu); + +/* + * powercap.h + */ + +struct powercap_zone { + char name[MAX_LINE_LEN]; + /* + * sys_name relative to PATH_TO_POWERCAP, + * do not forget the / in between + */ + char sys_name[SYSFS_PATH_MAX]; + int tree_depth; + struct powercap_zone *parent; + struct powercap_zone *children[POWERCAP_MAX_CHILD_ZONES]; + /* More possible caps or attributes to be added? */ + uint32_t has_power_uw:1, + has_energy_uj:1; + +}; + +int powercap_walk_zones(struct powercap_zone *zone, + int (*f)(struct powercap_zone *zone)); + +struct powercap_zone *powercap_init_zones(void); + +int powercap_get_enabled(int *mode); + +int powercap_set_enabled(int mode); + +int powercap_get_driver(char *driver, int buflen); + +int powercap_get_max_energy_range_uj(struct powercap_zone *zone, uint64_t *val); + +int powercap_get_energy_uj(struct powercap_zone *zone, uint64_t *val); + +int powercap_get_max_power_range_uw(struct powercap_zone *zone, uint64_t *val); + +int powercap_get_power_uw(struct powercap_zone *zone, uint64_t *val); + +int powercap_zone_get_enabled(struct powercap_zone *zone, int *mode); + +int powercap_zone_set_enabled(struct powercap_zone *zone, int mode); diff --git a/tools/power/cpupower/bindings/python/test_raw_pylibcpupower.py b/tools/power/cpupower/bindings/python/test_raw_pylibcpupower.py new file mode 100755 index 000000000000..3d6f62b9556a --- /dev/null +++ b/tools/power/cpupower/bindings/python/test_raw_pylibcpupower.py @@ -0,0 +1,42 @@ +#!/usr/bin/env python3 +# SPDX-License-Identifier: GPL-2.0-only + +import raw_pylibcpupower as p + +# Simple function call + +""" +Get cstate count +""" +cpu_cstates_count = p.cpuidle_state_count(0) +if cpu_cstates_count > -1: + print(f"CPU 0 has {cpu_cstates_count} c-states") +else: + print(f"cstate count error: return code: {cpu_cstates_count}") + +""" +Disable cstate (will fail if the above is 0, ex: a virtual machine) +""" +cstate_disabled = p.cpuidle_state_disable(0, 0, 1) +if cpu_cstates_count == 0: + print(f"CPU 0 has {cpu_cstates_count} c-states") +else: + print(f"cstate count error: return code: {cpu_cstates_count}") + +match cstate_disabled: + case 0: + print(f"CPU state disabled") + case -1: + print(f"Idlestate not available") + case _: + print(f"Not documented") + + +# Pointer example + +topo = p.cpupower_topology() +total_cpus = p.get_cpu_topology(topo) +if total_cpus > 0: + print(f"Number of total cpus: {total_cpus} and number of cores: {topo.cores}") +else: + print(f"Error: could not get cpu topology") diff --git a/tools/power/cpupower/lib/cpuidle.c b/tools/power/cpupower/lib/cpuidle.c index 479c5971aa6d..0ecac009273c 100644 --- a/tools/power/cpupower/lib/cpuidle.c +++ b/tools/power/cpupower/lib/cpuidle.c @@ -116,6 +116,7 @@ enum idlestate_value { IDLESTATE_USAGE, IDLESTATE_POWER, IDLESTATE_LATENCY, + IDLESTATE_RESIDENCY, IDLESTATE_TIME, IDLESTATE_DISABLE, MAX_IDLESTATE_VALUE_FILES @@ -125,6 +126,7 @@ static const char *idlestate_value_files[MAX_IDLESTATE_VALUE_FILES] = { [IDLESTATE_USAGE] = "usage", [IDLESTATE_POWER] = "power", [IDLESTATE_LATENCY] = "latency", + [IDLESTATE_RESIDENCY] = "residency", [IDLESTATE_TIME] = "time", [IDLESTATE_DISABLE] = "disable", }; @@ -254,6 +256,12 @@ unsigned long cpuidle_state_latency(unsigned int cpu, return cpuidle_state_get_one_value(cpu, idlestate, IDLESTATE_LATENCY); } +unsigned long cpuidle_state_residency(unsigned int cpu, + unsigned int idlestate) +{ + return cpuidle_state_get_one_value(cpu, idlestate, IDLESTATE_RESIDENCY); +} + unsigned long cpuidle_state_usage(unsigned int cpu, unsigned int idlestate) { diff --git a/tools/power/cpupower/lib/cpuidle.h b/tools/power/cpupower/lib/cpuidle.h index 2e10fead2e1e..2ab404d40259 100644 --- a/tools/power/cpupower/lib/cpuidle.h +++ b/tools/power/cpupower/lib/cpuidle.h @@ -8,6 +8,8 @@ int cpuidle_state_disable(unsigned int cpu, unsigned int idlestate, unsigned int disable); unsigned long cpuidle_state_latency(unsigned int cpu, unsigned int idlestate); +unsigned long cpuidle_state_residency(unsigned int cpu, + unsigned int idlestate); unsigned long cpuidle_state_usage(unsigned int cpu, unsigned int idlestate); unsigned long long cpuidle_state_time(unsigned int cpu, diff --git a/tools/power/cpupower/lib/powercap.c b/tools/power/cpupower/lib/powercap.c index a7a59c6bacda..94a0c69e55ef 100644 --- a/tools/power/cpupower/lib/powercap.c +++ b/tools/power/cpupower/lib/powercap.c @@ -78,6 +78,14 @@ int powercap_get_enabled(int *mode) } /* + * TODO: implement function. Returns dummy 0 for now. + */ +int powercap_set_enabled(int mode) +{ + return 0; +} + +/* * Hardcoded, because rapl is the only powercap implementation - * this needs to get more generic if more powercap implementations * should show up diff --git a/tools/power/cpupower/utils/cpuidle-info.c b/tools/power/cpupower/utils/cpuidle-info.c index 44126a87fa7a..e0d17f0de3fe 100644 --- a/tools/power/cpupower/utils/cpuidle-info.c +++ b/tools/power/cpupower/utils/cpuidle-info.c @@ -64,6 +64,8 @@ static void cpuidle_cpu_output(unsigned int cpu, int verbose) printf(_("Latency: %lu\n"), cpuidle_state_latency(cpu, idlestate)); + printf(_("Residency: %lu\n"), + cpuidle_state_residency(cpu, idlestate)); printf(_("Usage: %lu\n"), cpuidle_state_usage(cpu, idlestate)); printf(_("Duration: %llu\n"), @@ -115,6 +117,8 @@ static void proc_cpuidle_cpu_output(unsigned int cpu) printf(_("promotion[--] demotion[--] ")); printf(_("latency[%03lu] "), cpuidle_state_latency(cpu, cstate)); + printf(_("residency[%05lu] "), + cpuidle_state_residency(cpu, cstate)); printf(_("usage[%08lu] "), cpuidle_state_usage(cpu, cstate)); printf(_("duration[%020Lu] \n"), diff --git a/tools/power/pm-graph/.gitignore b/tools/power/pm-graph/.gitignore new file mode 100644 index 000000000000..37762a8a06d6 --- /dev/null +++ b/tools/power/pm-graph/.gitignore @@ -0,0 +1,3 @@ +# sleepgraph.py artifacts +suspend-[0-9]*-[0-9]* +suspend-[0-9]*-[0-9]*-x[0-9]* diff --git a/tools/power/pm-graph/Makefile b/tools/power/pm-graph/Makefile index b5310832c19c..aeddbaf2d4c4 100644 --- a/tools/power/pm-graph/Makefile +++ b/tools/power/pm-graph/Makefile @@ -1,51 +1,86 @@ # SPDX-License-Identifier: GPL-2.0 -PREFIX ?= /usr -DESTDIR ?= +# +# Copyright (c) 2013, Intel Corporation. +# +# This program is free software; you can redistribute it and/or modify it +# under the terms and conditions of the GNU General Public License, +# version 2, as published by the Free Software Foundation. +# +# This program is distributed in the hope it will be useful, but WITHOUT +# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or +# FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for +# more details. +# +# Authors: +# Todd Brandt <todd.e.brandt@linux.intel.com> + +# Prefix to the directories we're installing to +DESTDIR ?= + +# Directory definitions. These are default and most probably +# do not need to be changed. Please note that DESTDIR is +# added in front of any of them + +BINDIR ?= /usr/bin +MANDIR ?= /usr/share/man +LIBDIR ?= /usr/lib + +# Toolchain: what tools do we use, and what options do they need: +INSTALL = /usr/bin/install +INSTALL_DATA = ${INSTALL} -m 644 all: @echo "Nothing to build" install : uninstall - install -d $(DESTDIR)$(PREFIX)/lib/pm-graph - install sleepgraph.py $(DESTDIR)$(PREFIX)/lib/pm-graph - install bootgraph.py $(DESTDIR)$(PREFIX)/lib/pm-graph - install -d $(DESTDIR)$(PREFIX)/lib/pm-graph/config - install -m 644 config/cgskip.txt $(DESTDIR)$(PREFIX)/lib/pm-graph/config - install -m 644 config/freeze-callgraph.cfg $(DESTDIR)$(PREFIX)/lib/pm-graph/config - install -m 644 config/freeze.cfg $(DESTDIR)$(PREFIX)/lib/pm-graph/config - install -m 644 config/freeze-dev.cfg $(DESTDIR)$(PREFIX)/lib/pm-graph/config - install -m 644 config/standby-callgraph.cfg $(DESTDIR)$(PREFIX)/lib/pm-graph/config - install -m 644 config/standby.cfg $(DESTDIR)$(PREFIX)/lib/pm-graph/config - install -m 644 config/standby-dev.cfg $(DESTDIR)$(PREFIX)/lib/pm-graph/config - install -m 644 config/suspend-callgraph.cfg $(DESTDIR)$(PREFIX)/lib/pm-graph/config - install -m 644 config/suspend.cfg $(DESTDIR)$(PREFIX)/lib/pm-graph/config - install -m 644 config/suspend-dev.cfg $(DESTDIR)$(PREFIX)/lib/pm-graph/config - install -m 644 config/suspend-x2-proc.cfg $(DESTDIR)$(PREFIX)/lib/pm-graph/config - - install -d $(DESTDIR)$(PREFIX)/bin - ln -s ../lib/pm-graph/bootgraph.py $(DESTDIR)$(PREFIX)/bin/bootgraph - ln -s ../lib/pm-graph/sleepgraph.py $(DESTDIR)$(PREFIX)/bin/sleepgraph - - install -d $(DESTDIR)$(PREFIX)/share/man/man8 - install bootgraph.8 $(DESTDIR)$(PREFIX)/share/man/man8 - install sleepgraph.8 $(DESTDIR)$(PREFIX)/share/man/man8 + $(INSTALL) -d $(DESTDIR)$(LIBDIR)/pm-graph + $(INSTALL) sleepgraph.py $(DESTDIR)$(LIBDIR)/pm-graph + $(INSTALL) bootgraph.py $(DESTDIR)$(LIBDIR)/pm-graph + $(INSTALL) -d $(DESTDIR)$(LIBDIR)/pm-graph/config + $(INSTALL_DATA) config/cgskip.txt $(DESTDIR)$(LIBDIR)/pm-graph/config + $(INSTALL_DATA) config/freeze-callgraph.cfg $(DESTDIR)$(LIBDIR)/pm-graph/config + $(INSTALL_DATA) config/freeze.cfg $(DESTDIR)$(LIBDIR)/pm-graph/config + $(INSTALL_DATA) config/freeze-dev.cfg $(DESTDIR)$(LIBDIR)/pm-graph/config + $(INSTALL_DATA) config/standby-callgraph.cfg $(DESTDIR)$(LIBDIR)/pm-graph/config + $(INSTALL_DATA) config/standby.cfg $(DESTDIR)$(LIBDIR)/pm-graph/config + $(INSTALL_DATA) config/standby-dev.cfg $(DESTDIR)$(LIBDIR)/pm-graph/config + $(INSTALL_DATA) config/suspend-callgraph.cfg $(DESTDIR)$(LIBDIR)/pm-graph/config + $(INSTALL_DATA) config/suspend.cfg $(DESTDIR)$(LIBDIR)/pm-graph/config + $(INSTALL_DATA) config/suspend-dev.cfg $(DESTDIR)$(LIBDIR)/pm-graph/config + $(INSTALL_DATA) config/suspend-x2-proc.cfg $(DESTDIR)$(LIBDIR)/pm-graph/config + + $(INSTALL) -d $(DESTDIR)$(BINDIR) + ln -s ../lib/pm-graph/bootgraph.py $(DESTDIR)$(BINDIR)/bootgraph + ln -s ../lib/pm-graph/sleepgraph.py $(DESTDIR)$(BINDIR)/sleepgraph + + $(INSTALL) -d $(DESTDIR)$(MANDIR)/man8 + $(INSTALL) bootgraph.8 $(DESTDIR)$(MANDIR)/man8 + $(INSTALL) sleepgraph.8 $(DESTDIR)$(MANDIR)/man8 uninstall : - rm -f $(DESTDIR)$(PREFIX)/share/man/man8/bootgraph.8 - rm -f $(DESTDIR)$(PREFIX)/share/man/man8/sleepgraph.8 + rm -f $(DESTDIR)$(MANDIR)/man8/bootgraph.8 + rm -f $(DESTDIR)$(MANDIR)/man8/sleepgraph.8 - rm -f $(DESTDIR)$(PREFIX)/bin/bootgraph - rm -f $(DESTDIR)$(PREFIX)/bin/sleepgraph + rm -f $(DESTDIR)$(BINDIR)/bootgraph + rm -f $(DESTDIR)$(BINDIR)/sleepgraph - rm -f $(DESTDIR)$(PREFIX)/lib/pm-graph/config/* - if [ -d $(DESTDIR)$(PREFIX)/lib/pm-graph/config ] ; then \ - rmdir $(DESTDIR)$(PREFIX)/lib/pm-graph/config; \ + rm -f $(DESTDIR)$(LIBDIR)/pm-graph/config/* + if [ -d $(DESTDIR)$(LIBDIR)/pm-graph/config ] ; then \ + rmdir $(DESTDIR)$(LIBDIR)/pm-graph/config; \ fi; - rm -f $(DESTDIR)$(PREFIX)/lib/pm-graph/__pycache__/* - if [ -d $(DESTDIR)$(PREFIX)/lib/pm-graph/__pycache__ ] ; then \ - rmdir $(DESTDIR)$(PREFIX)/lib/pm-graph/__pycache__; \ + rm -f $(DESTDIR)$(LIBDIR)/pm-graph/__pycache__/* + if [ -d $(DESTDIR)$(LIBDIR)/pm-graph/__pycache__ ] ; then \ + rmdir $(DESTDIR)$(LIBDIR)/pm-graph/__pycache__; \ fi; - rm -f $(DESTDIR)$(PREFIX)/lib/pm-graph/* - if [ -d $(DESTDIR)$(PREFIX)/lib/pm-graph ] ; then \ - rmdir $(DESTDIR)$(PREFIX)/lib/pm-graph; \ + rm -f $(DESTDIR)$(LIBDIR)/pm-graph/* + if [ -d $(DESTDIR)$(LIBDIR)/pm-graph ] ; then \ + rmdir $(DESTDIR)$(LIBDIR)/pm-graph; \ fi; + +help: + @echo 'Building targets:' + @echo ' all - Nothing to build' + @echo ' install - Install the program and create necessary directories' + @echo ' uninstall - Remove installed files and directories' + +.PHONY: all install uninstall help |