Commit 312cd1f5 authored by Kan Liang's avatar Kan Liang Committed by Yunying Sun
Browse files

perf/x86/intel: Add common intel_pmu_init_hybrid() 

mainline inclusion
from mainline-v6.10-rc1
commit 97588df87b56e27fd2b5d928d61c7a53e38afbb0
category: bugfix
bugzilla: https://gitee.com/openeuler/intel-kernel/issues/IAGLFT
CVE: NA

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=97588df87b56e27fd2b5d928d61c7a53e38afbb0



-------------------------------------

The current hybrid initialization codes aren't well organized and are
hard to read.

Factor out intel_pmu_init_hybrid() to do a common setup for each
hybrid PMU. The PMU-specific capability will be updated later via either
hard code (ADL) or CPUID hybrid enumeration (MTL).

Splitting the ADL and MTL initialization codes, since they have
different uarches. The hard code PMU capabilities are not required for
MTL either. They can be enumerated by the new leaf 0x23 and
IA32_PERF_CAPABILITIES MSR.

The hybrid enumeration of the IA32_PERF_CAPABILITIES MSR is broken on
MTL. Using the default value.

Intel-SIG: commit 97588df87b56 perf/x86/intel: Add common intel_pmu_init_hybrid()
Backport as a dependency needed by the GNR distinct pmu name fix

Signed-off-by: default avatarKan Liang <kan.liang@linux.intel.com>
Signed-off-by: default avatarIngo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230829125806.3016082-7-kan.liang@linux.intel.com


Signed-off-by: default avatarYunying Sun <yunying.sun@intel.com>
parent d6caf70e

arch/x86/events/intel/core.c

+111 −51
Original line number Diff line number Diff line
@@ -4678,6 +4678,16 @@ static void intel_pmu_check_num_counters(int *num_counters, 
					 int *num_counters_fixed,

					 u64 *intel_ctrl, u64 fixed_mask);



static inline bool intel_pmu_broken_perf_cap(void)

{

	/* The Perf Metric (Bit 15) is always cleared */

	if ((boot_cpu_data.x86_model == INTEL_FAM6_METEORLAKE) ||

	    (boot_cpu_data.x86_model == INTEL_FAM6_METEORLAKE_L))

		return true;



	return false;

}



static void update_pmu_cap(struct x86_hybrid_pmu *pmu)

{

	unsigned int sub_bitmaps = cpuid_eax(ARCH_PERFMON_EXT_LEAF);
@@ -4690,7 +4700,27 @@ static void update_pmu_cap(struct x86_hybrid_pmu *pmu) 
		pmu->num_counters_fixed = fls(ebx);

		intel_pmu_check_num_counters(&pmu->num_counters, &pmu->num_counters_fixed,

					     &pmu->intel_ctrl, ebx);

		pmu->max_pebs_events = min_t(unsigned, MAX_PEBS_EVENTS, pmu->num_counters);

		pmu->unconstrained = (struct event_constraint)

				     __EVENT_CONSTRAINT(0, (1ULL << pmu->num_counters) - 1,

							0, pmu->num_counters, 0, 0);

	}





	if (!intel_pmu_broken_perf_cap()) {

		/* Perf Metric (Bit 15) and PEBS via PT (Bit 16) are hybrid enumeration */

		rdmsrl(MSR_IA32_PERF_CAPABILITIES, pmu->intel_cap.capabilities);

	}



	if (pmu->intel_cap.perf_metrics)

		pmu->intel_ctrl |= 1ULL << GLOBAL_CTRL_EN_PERF_METRICS;

	else

		pmu->intel_ctrl &= ~(1ULL << GLOBAL_CTRL_EN_PERF_METRICS);



	if (pmu->intel_cap.pebs_output_pt_available)

		pmu->pmu.capabilities |= PERF_PMU_CAP_AUX_OUTPUT;

	else

		pmu->pmu.capabilities |= ~PERF_PMU_CAP_AUX_OUTPUT;

}



static struct x86_hybrid_pmu *find_hybrid_pmu_for_cpu(void)
@@ -6019,10 +6049,52 @@ static void intel_pmu_check_hybrid_pmus(u64 fixed_mask) 
	}

}



static __always_inline bool is_mtl(u8 x86_model)

static const struct { enum hybrid_pmu_type id; char *name; } intel_hybrid_pmu_type_map[] __initconst = {

	{ hybrid_small, "cpu_atom" },

	{ hybrid_big, "cpu_core" },

};



static __always_inline int intel_pmu_init_hybrid(enum hybrid_pmu_type pmus)

{

	return (x86_model == INTEL_FAM6_METEORLAKE) ||

	       (x86_model == INTEL_FAM6_METEORLAKE_L);

	unsigned long pmus_mask = pmus;

	struct x86_hybrid_pmu *pmu;

	int idx = 0, bit;



	x86_pmu.num_hybrid_pmus = hweight_long(pmus_mask);

	x86_pmu.hybrid_pmu = kcalloc(x86_pmu.num_hybrid_pmus,

				     sizeof(struct x86_hybrid_pmu),

				     GFP_KERNEL);

	if (!x86_pmu.hybrid_pmu)

		return -ENOMEM;



	static_branch_enable(&perf_is_hybrid);

	x86_pmu.filter = intel_pmu_filter;



	for_each_set_bit(bit, &pmus_mask, ARRAY_SIZE(intel_hybrid_pmu_type_map)) {

		pmu = &x86_pmu.hybrid_pmu[idx++];

		pmu->pmu_type = intel_hybrid_pmu_type_map[bit].id;

		pmu->name = intel_hybrid_pmu_type_map[bit].name;



		pmu->num_counters = x86_pmu.num_counters;

		pmu->num_counters_fixed = x86_pmu.num_counters_fixed;

		pmu->max_pebs_events = min_t(unsigned, MAX_PEBS_EVENTS, pmu->num_counters);

		pmu->unconstrained = (struct event_constraint)

				     __EVENT_CONSTRAINT(0, (1ULL << pmu->num_counters) - 1,

							0, pmu->num_counters, 0, 0);



		pmu->intel_cap.capabilities = x86_pmu.intel_cap.capabilities;

		if (pmu->pmu_type & hybrid_small) {

			pmu->intel_cap.perf_metrics = 0;

			pmu->intel_cap.pebs_output_pt_available = 1;

			pmu->mid_ack = true;

		} else if (pmu->pmu_type & hybrid_big) {

			pmu->intel_cap.perf_metrics = 1;

			pmu->intel_cap.pebs_output_pt_available = 0;

			pmu->late_ack = true;

		}

	}



	return 0;

}



static __always_inline void intel_pmu_init_glc(struct pmu *pmu)
@@ -6706,23 +6778,14 @@ __init int intel_pmu_init(void) 
	case INTEL_FAM6_RAPTORLAKE:

	case INTEL_FAM6_RAPTORLAKE_P:

	case INTEL_FAM6_RAPTORLAKE_S:

	case INTEL_FAM6_METEORLAKE:

	case INTEL_FAM6_METEORLAKE_L:

		/*

		 * Alder Lake has 2 types of CPU, core and atom.

		 *

		 * Initialize the common PerfMon capabilities here.

		 */

		x86_pmu.hybrid_pmu = kcalloc(X86_HYBRID_NUM_PMUS,

					     sizeof(struct x86_hybrid_pmu),

					     GFP_KERNEL);

		if (!x86_pmu.hybrid_pmu)

			return -ENOMEM;

		static_branch_enable(&perf_is_hybrid);

		x86_pmu.num_hybrid_pmus = X86_HYBRID_NUM_PMUS;

		intel_pmu_init_hybrid(hybrid_big_small);



		x86_pmu.pebs_latency_data = adl_latency_data_small;

		x86_pmu.filter = intel_pmu_filter;

		x86_pmu.get_event_constraints = adl_get_event_constraints;

		x86_pmu.hw_config = adl_hw_config;

		x86_pmu.get_hybrid_cpu_type = adl_get_hybrid_cpu_type;
@@ -6735,10 +6798,7 @@ __init int intel_pmu_init(void) 


		/* Initialize big core specific PerfMon capabilities.*/

		pmu = &x86_pmu.hybrid_pmu[X86_HYBRID_PMU_CORE_IDX];

		pmu->name = "cpu_core";

		pmu->pmu_type = hybrid_big;

		intel_pmu_init_glc(&pmu->pmu);

		pmu->late_ack = true;

		if (cpu_feature_enabled(X86_FEATURE_HYBRID_CPU)) {

			pmu->num_counters = x86_pmu.num_counters + 2;

			pmu->num_counters_fixed = x86_pmu.num_counters_fixed + 1;
@@ -6763,45 +6823,45 @@ __init int intel_pmu_init(void) 
		pmu->unconstrained = (struct event_constraint)

					__EVENT_CONSTRAINT(0, (1ULL << pmu->num_counters) - 1,

							   0, pmu->num_counters, 0, 0);

		pmu->intel_cap.capabilities = x86_pmu.intel_cap.capabilities;

		pmu->intel_cap.perf_metrics = 1;

		pmu->intel_cap.pebs_output_pt_available = 0;



		pmu->extra_regs = intel_glc_extra_regs;



		/* Initialize Atom core specific PerfMon capabilities.*/

		pmu = &x86_pmu.hybrid_pmu[X86_HYBRID_PMU_ATOM_IDX];

		pmu->name = "cpu_atom";

		pmu->pmu_type = hybrid_small;

		intel_pmu_init_grt(&pmu->pmu);

		pmu->mid_ack = true;

		pmu->num_counters = x86_pmu.num_counters;

		pmu->num_counters_fixed = x86_pmu.num_counters_fixed;

		pmu->max_pebs_events = x86_pmu.max_pebs_events;

		pmu->unconstrained = (struct event_constraint)

					__EVENT_CONSTRAINT(0, (1ULL << pmu->num_counters) - 1,

							   0, pmu->num_counters, 0, 0);

		pmu->intel_cap.capabilities = x86_pmu.intel_cap.capabilities;

		pmu->intel_cap.perf_metrics = 0;

		pmu->intel_cap.pebs_output_pt_available = 1;



		if (is_mtl(boot_cpu_data.x86_model)) {

			x86_pmu.hybrid_pmu[X86_HYBRID_PMU_CORE_IDX].extra_regs = intel_rwc_extra_regs;

		x86_pmu.flags |= PMU_FL_MEM_LOADS_AUX;

		intel_pmu_pebs_data_source_adl();

		pr_cont("Alderlake Hybrid events, ");

		name = "alderlake_hybrid";

		break;



	case INTEL_FAM6_METEORLAKE:

	case INTEL_FAM6_METEORLAKE_L:

		intel_pmu_init_hybrid(hybrid_big_small);



		x86_pmu.pebs_latency_data = mtl_latency_data_small;

		x86_pmu.get_event_constraints = mtl_get_event_constraints;

		x86_pmu.hw_config = adl_hw_config;



		td_attr = adl_hybrid_events_attrs;

		mem_attr = mtl_hybrid_mem_attrs;

		tsx_attr = adl_hybrid_tsx_attrs;

		extra_attr = boot_cpu_has(X86_FEATURE_RTM) ?

			mtl_hybrid_extra_attr_rtm : mtl_hybrid_extra_attr;

			mem_attr = mtl_hybrid_mem_attrs;

			intel_pmu_pebs_data_source_mtl();

			x86_pmu.get_event_constraints = mtl_get_event_constraints;



		/* Initialize big core specific PerfMon capabilities.*/

		pmu = &x86_pmu.hybrid_pmu[X86_HYBRID_PMU_CORE_IDX];

		intel_pmu_init_glc(&pmu->pmu);

		pmu->extra_regs = intel_rwc_extra_regs;



		/* Initialize Atom core specific PerfMon capabilities.*/

		pmu = &x86_pmu.hybrid_pmu[X86_HYBRID_PMU_ATOM_IDX];

		intel_pmu_init_grt(&pmu->pmu);

		pmu->extra_regs = intel_cmt_extra_regs;



		intel_pmu_pebs_data_source_mtl();

		pr_cont("Meteorlake Hybrid events, ");

		name = "meteorlake_hybrid";

		} else {

			x86_pmu.flags |= PMU_FL_MEM_LOADS_AUX;

			intel_pmu_pebs_data_source_adl();

			pr_cont("Alderlake Hybrid events, ");

			name = "alderlake_hybrid";

		}

		break;



	default:
@@ -6913,7 +6973,7 @@ __init int intel_pmu_init(void) 
	if (!is_hybrid() && x86_pmu.intel_cap.perf_metrics)

		x86_pmu.intel_ctrl |= 1ULL << GLOBAL_CTRL_EN_PERF_METRICS;



	if (is_hybrid())

	if (is_hybrid() && !boot_cpu_has(X86_FEATURE_ARCH_PERFMON_EXT))

		intel_pmu_check_hybrid_pmus((u64)fixed_mask);



	if (x86_pmu.intel_cap.pebs_timing_info)