diff options
Diffstat (limited to 'arch/x86/events')
-rw-r--r-- | arch/x86/events/core.c | 7 | ||||
-rw-r--r-- | arch/x86/events/intel/bts.c | 6 | ||||
-rw-r--r-- | arch/x86/events/intel/core.c | 89 | ||||
-rw-r--r-- | arch/x86/events/intel/ds.c | 7 | ||||
-rw-r--r-- | arch/x86/events/intel/lbr.c | 20 | ||||
-rw-r--r-- | arch/x86/events/intel/uncore_discovery.h | 2 | ||||
-rw-r--r-- | arch/x86/events/intel/uncore_snbep.c | 16 | ||||
-rw-r--r-- | arch/x86/events/msr.c | 1 | ||||
-rw-r--r-- | arch/x86/events/perf_event.h | 21 |
9 files changed, 140 insertions, 29 deletions
diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c index 2a57dbed4894..38b2c779146f 100644 --- a/arch/x86/events/core.c +++ b/arch/x86/events/core.c @@ -66,6 +66,8 @@ DEFINE_STATIC_CALL_NULL(x86_pmu_enable_all, *x86_pmu.enable_all); DEFINE_STATIC_CALL_NULL(x86_pmu_enable, *x86_pmu.enable); DEFINE_STATIC_CALL_NULL(x86_pmu_disable, *x86_pmu.disable); +DEFINE_STATIC_CALL_NULL(x86_pmu_assign, *x86_pmu.assign); + DEFINE_STATIC_CALL_NULL(x86_pmu_add, *x86_pmu.add); DEFINE_STATIC_CALL_NULL(x86_pmu_del, *x86_pmu.del); DEFINE_STATIC_CALL_NULL(x86_pmu_read, *x86_pmu.read); @@ -1215,6 +1217,8 @@ static inline void x86_assign_hw_event(struct perf_event *event, hwc->last_cpu = smp_processor_id(); hwc->last_tag = ++cpuc->tags[i]; + static_call_cond(x86_pmu_assign)(event, idx); + switch (hwc->idx) { case INTEL_PMC_IDX_FIXED_BTS: case INTEL_PMC_IDX_FIXED_VLBR: @@ -2005,6 +2009,8 @@ static void x86_pmu_static_call_update(void) static_call_update(x86_pmu_enable, x86_pmu.enable); static_call_update(x86_pmu_disable, x86_pmu.disable); + static_call_update(x86_pmu_assign, x86_pmu.assign); + static_call_update(x86_pmu_add, x86_pmu.add); static_call_update(x86_pmu_del, x86_pmu.del); static_call_update(x86_pmu_read, x86_pmu.read); @@ -2465,6 +2471,7 @@ static int x86_pmu_event_init(struct perf_event *event) if (err) { if (event->destroy) event->destroy(event); + event->destroy = NULL; } if (READ_ONCE(x86_pmu.attr_rdpmc) && diff --git a/arch/x86/events/intel/bts.c b/arch/x86/events/intel/bts.c index 6320d2cfd9d3..974e917e65b2 100644 --- a/arch/x86/events/intel/bts.c +++ b/arch/x86/events/intel/bts.c @@ -209,6 +209,12 @@ static void bts_update(struct bts_ctx *bts) } else { local_set(&buf->data_size, head); } + + /* + * Since BTS is coherent, just add compiler barrier to ensure + * BTS updating is ordered against bts::handle::event. + */ + barrier(); } static int diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index 7011e87be6d0..603964408d2d 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -243,7 +243,8 @@ static struct extra_reg intel_skl_extra_regs[] __read_mostly = { static struct event_constraint intel_icl_event_constraints[] = { FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ - FIXED_EVENT_CONSTRAINT(0x01c0, 0), /* INST_RETIRED.PREC_DIST */ + FIXED_EVENT_CONSTRAINT(0x01c0, 0), /* old INST_RETIRED.PREC_DIST */ + FIXED_EVENT_CONSTRAINT(0x0100, 0), /* INST_RETIRED.PREC_DIST */ FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */ FIXED_EVENT_CONSTRAINT(0x0400, 3), /* SLOTS */ @@ -263,6 +264,7 @@ static struct event_constraint intel_icl_event_constraints[] = { INTEL_EVENT_CONSTRAINT_RANGE(0xa8, 0xb0, 0xf), INTEL_EVENT_CONSTRAINT_RANGE(0xb7, 0xbd, 0xf), INTEL_EVENT_CONSTRAINT_RANGE(0xd0, 0xe6, 0xf), + INTEL_EVENT_CONSTRAINT(0xef, 0xf), INTEL_EVENT_CONSTRAINT_RANGE(0xf0, 0xf4, 0xf), EVENT_CONSTRAINT_END }; @@ -287,7 +289,7 @@ static struct extra_reg intel_spr_extra_regs[] __read_mostly = { static struct event_constraint intel_spr_event_constraints[] = { FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ - FIXED_EVENT_CONSTRAINT(0x01c0, 0), /* INST_RETIRED.PREC_DIST */ + FIXED_EVENT_CONSTRAINT(0x0100, 0), /* INST_RETIRED.PREC_DIST */ FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */ FIXED_EVENT_CONSTRAINT(0x0400, 3), /* SLOTS */ @@ -2143,19 +2145,19 @@ static __initconst const u64 knl_hw_cache_extra_regs * However, there are some cases which may change PEBS status, e.g. PMI * throttle. The PEBS_ENABLE should be updated where the status changes. */ -static void __intel_pmu_disable_all(void) +static __always_inline void __intel_pmu_disable_all(bool bts) { struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0); - if (test_bit(INTEL_PMC_IDX_FIXED_BTS, cpuc->active_mask)) + if (bts && test_bit(INTEL_PMC_IDX_FIXED_BTS, cpuc->active_mask)) intel_pmu_disable_bts(); } -static void intel_pmu_disable_all(void) +static __always_inline void intel_pmu_disable_all(void) { - __intel_pmu_disable_all(); + __intel_pmu_disable_all(true); intel_pmu_pebs_disable_all(); intel_pmu_lbr_disable_all(); } @@ -2186,6 +2188,49 @@ static void intel_pmu_enable_all(int added) __intel_pmu_enable_all(added, false); } +static noinline int +__intel_pmu_snapshot_branch_stack(struct perf_branch_entry *entries, + unsigned int cnt, unsigned long flags) +{ + struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); + + intel_pmu_lbr_read(); + cnt = min_t(unsigned int, cnt, x86_pmu.lbr_nr); + + memcpy(entries, cpuc->lbr_entries, sizeof(struct perf_branch_entry) * cnt); + intel_pmu_enable_all(0); + local_irq_restore(flags); + return cnt; +} + +static int +intel_pmu_snapshot_branch_stack(struct perf_branch_entry *entries, unsigned int cnt) +{ + unsigned long flags; + + /* must not have branches... */ + local_irq_save(flags); + __intel_pmu_disable_all(false); /* we don't care about BTS */ + __intel_pmu_pebs_disable_all(); + __intel_pmu_lbr_disable(); + /* ... until here */ + return __intel_pmu_snapshot_branch_stack(entries, cnt, flags); +} + +static int +intel_pmu_snapshot_arch_branch_stack(struct perf_branch_entry *entries, unsigned int cnt) +{ + unsigned long flags; + + /* must not have branches... */ + local_irq_save(flags); + __intel_pmu_disable_all(false); /* we don't care about BTS */ + __intel_pmu_pebs_disable_all(); + __intel_pmu_arch_lbr_disable(); + /* ... until here */ + return __intel_pmu_snapshot_branch_stack(entries, cnt, flags); +} + /* * Workaround for: * Intel Errata AAK100 (model 26) @@ -2402,6 +2447,12 @@ static void intel_pmu_disable_event(struct perf_event *event) intel_pmu_pebs_disable(event); } +static void intel_pmu_assign_event(struct perf_event *event, int idx) +{ + if (is_pebs_pt(event)) + perf_report_aux_output_id(event, idx); +} + static void intel_pmu_del_event(struct perf_event *event) { if (needs_branch_stack(event)) @@ -2929,7 +2980,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) apic_write(APIC_LVTPC, APIC_DM_NMI); intel_bts_disable_local(); cpuc->enabled = 0; - __intel_pmu_disable_all(); + __intel_pmu_disable_all(true); handled = intel_pmu_drain_bts_buffer(); handled += intel_bts_interrupt(); status = intel_pmu_get_status(); @@ -4494,8 +4545,16 @@ static int intel_pmu_check_period(struct perf_event *event, u64 value) return intel_pmu_has_bts_period(event, value) ? -EINVAL : 0; } +static void intel_aux_output_init(void) +{ + /* Refer also intel_pmu_aux_output_match() */ + if (x86_pmu.intel_cap.pebs_output_pt_available) + x86_pmu.assign = intel_pmu_assign_event; +} + static int intel_pmu_aux_output_match(struct perf_event *event) { + /* intel_pmu_assign_event() is needed, refer intel_aux_output_init() */ if (!x86_pmu.intel_cap.pebs_output_pt_available) return 0; @@ -6283,9 +6342,21 @@ __init int intel_pmu_init(void) x86_pmu.lbr_nr = 0; } - if (x86_pmu.lbr_nr) + if (x86_pmu.lbr_nr) { pr_cont("%d-deep LBR, ", x86_pmu.lbr_nr); + /* only support branch_stack snapshot for perfmon >= v2 */ + if (x86_pmu.disable_all == intel_pmu_disable_all) { + if (boot_cpu_has(X86_FEATURE_ARCH_LBR)) { + static_call_update(perf_snapshot_branch_stack, + intel_pmu_snapshot_arch_branch_stack); + } else { + static_call_update(perf_snapshot_branch_stack, + intel_pmu_snapshot_branch_stack); + } + } + } + intel_pmu_check_extra_regs(x86_pmu.extra_regs); /* Support full width counters using alternative MSR range */ @@ -6301,6 +6372,8 @@ __init int intel_pmu_init(void) if (is_hybrid()) intel_pmu_check_hybrid_pmus((u64)fixed_mask); + intel_aux_output_init(); + return 0; } diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index 8647713276a7..2e215369df4a 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -923,7 +923,8 @@ struct event_constraint intel_skl_pebs_event_constraints[] = { }; struct event_constraint intel_icl_pebs_event_constraints[] = { - INTEL_FLAGS_UEVENT_CONSTRAINT(0x1c0, 0x100000000ULL), /* INST_RETIRED.PREC_DIST */ + INTEL_FLAGS_UEVENT_CONSTRAINT(0x01c0, 0x100000000ULL), /* old INST_RETIRED.PREC_DIST */ + INTEL_FLAGS_UEVENT_CONSTRAINT(0x0100, 0x100000000ULL), /* INST_RETIRED.PREC_DIST */ INTEL_FLAGS_UEVENT_CONSTRAINT(0x0400, 0x800000000ULL), /* SLOTS */ INTEL_PLD_CONSTRAINT(0x1cd, 0xff), /* MEM_TRANS_RETIRED.LOAD_LATENCY */ @@ -943,7 +944,7 @@ struct event_constraint intel_icl_pebs_event_constraints[] = { }; struct event_constraint intel_spr_pebs_event_constraints[] = { - INTEL_FLAGS_UEVENT_CONSTRAINT(0x1c0, 0x100000000ULL), + INTEL_FLAGS_UEVENT_CONSTRAINT(0x100, 0x100000000ULL), /* INST_RETIRED.PREC_DIST */ INTEL_FLAGS_UEVENT_CONSTRAINT(0x0400, 0x800000000ULL), INTEL_FLAGS_EVENT_CONSTRAINT(0xc0, 0xfe), @@ -1301,7 +1302,7 @@ void intel_pmu_pebs_disable_all(void) struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); if (cpuc->pebs_enabled) - wrmsrl(MSR_IA32_PEBS_ENABLE, 0); + __intel_pmu_pebs_disable_all(); } static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) diff --git a/arch/x86/events/intel/lbr.c b/arch/x86/events/intel/lbr.c index 9e6d6eaeb4cb..6b72e9b55c69 100644 --- a/arch/x86/events/intel/lbr.c +++ b/arch/x86/events/intel/lbr.c @@ -228,20 +228,6 @@ static void __intel_pmu_lbr_enable(bool pmi) wrmsrl(MSR_ARCH_LBR_CTL, lbr_select | ARCH_LBR_CTL_LBREN); } -static void __intel_pmu_lbr_disable(void) -{ - u64 debugctl; - - if (static_cpu_has(X86_FEATURE_ARCH_LBR)) { - wrmsrl(MSR_ARCH_LBR_CTL, 0); - return; - } - - rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); - debugctl &= ~(DEBUGCTLMSR_LBR | DEBUGCTLMSR_FREEZE_LBRS_ON_PMI); - wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); -} - void intel_pmu_lbr_reset_32(void) { int i; @@ -779,8 +765,12 @@ void intel_pmu_lbr_disable_all(void) { struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events); - if (cpuc->lbr_users && !vlbr_exclude_host()) + if (cpuc->lbr_users && !vlbr_exclude_host()) { + if (static_cpu_has(X86_FEATURE_ARCH_LBR)) + return __intel_pmu_arch_lbr_disable(); + __intel_pmu_lbr_disable(); + } } void intel_pmu_lbr_read_32(struct cpu_hw_events *cpuc) diff --git a/arch/x86/events/intel/uncore_discovery.h b/arch/x86/events/intel/uncore_discovery.h index 7280c8a3c831..6d735611c281 100644 --- a/arch/x86/events/intel/uncore_discovery.h +++ b/arch/x86/events/intel/uncore_discovery.h @@ -30,7 +30,7 @@ #define uncore_discovery_invalid_unit(unit) \ - (!unit.table1 || !unit.ctl || !unit.table3 || \ + (!unit.table1 || !unit.ctl || \ unit.table1 == -1ULL || unit.ctl == -1ULL || \ unit.table3 == -1ULL) diff --git a/arch/x86/events/intel/uncore_snbep.c b/arch/x86/events/intel/uncore_snbep.c index 5ddc0f30db6f..eb2c6cea9d0d 100644 --- a/arch/x86/events/intel/uncore_snbep.c +++ b/arch/x86/events/intel/uncore_snbep.c @@ -452,7 +452,7 @@ #define ICX_M3UPI_PCI_PMON_BOX_CTL 0xa0 /* ICX IMC */ -#define ICX_NUMBER_IMC_CHN 2 +#define ICX_NUMBER_IMC_CHN 3 #define ICX_IMC_MEM_STRIDE 0x4 /* SPR */ @@ -5076,8 +5076,10 @@ static struct event_constraint icx_uncore_iio_constraints[] = { UNCORE_EVENT_CONSTRAINT(0x02, 0x3), UNCORE_EVENT_CONSTRAINT(0x03, 0x3), UNCORE_EVENT_CONSTRAINT(0x83, 0x3), + UNCORE_EVENT_CONSTRAINT(0x88, 0xc), UNCORE_EVENT_CONSTRAINT(0xc0, 0xc), UNCORE_EVENT_CONSTRAINT(0xc5, 0xc), + UNCORE_EVENT_CONSTRAINT(0xd5, 0xc), EVENT_CONSTRAINT_END }; @@ -5463,7 +5465,7 @@ static struct intel_uncore_ops icx_uncore_mmio_ops = { static struct intel_uncore_type icx_uncore_imc = { .name = "imc", .num_counters = 4, - .num_boxes = 8, + .num_boxes = 12, .perf_ctr_bits = 48, .fixed_ctr_bits = 48, .fixed_ctr = SNR_IMC_MMIO_PMON_FIXED_CTR, @@ -5647,6 +5649,7 @@ static struct intel_uncore_type spr_uncore_chabox = { .event_mask = SPR_CHA_PMON_EVENT_MASK, .event_mask_ext = SPR_RAW_EVENT_MASK_EXT, .num_shared_regs = 1, + .constraints = skx_uncore_chabox_constraints, .ops = &spr_uncore_chabox_ops, .format_group = &spr_uncore_chabox_format_group, .attr_update = uncore_alias_groups, @@ -5658,6 +5661,7 @@ static struct intel_uncore_type spr_uncore_iio = { .event_mask_ext = SNR_IIO_PMON_RAW_EVENT_MASK_EXT, .format_group = &snr_uncore_iio_format_group, .attr_update = uncore_alias_groups, + .constraints = icx_uncore_iio_constraints, }; static struct attribute *spr_uncore_raw_formats_attr[] = { @@ -5686,9 +5690,16 @@ static struct intel_uncore_type spr_uncore_irp = { }; +static struct event_constraint spr_uncore_m2pcie_constraints[] = { + UNCORE_EVENT_CONSTRAINT(0x14, 0x3), + UNCORE_EVENT_CONSTRAINT(0x2d, 0x3), + EVENT_CONSTRAINT_END +}; + static struct intel_uncore_type spr_uncore_m2pcie = { SPR_UNCORE_COMMON_FORMAT(), .name = "m2pcie", + .constraints = spr_uncore_m2pcie_constraints, }; static struct intel_uncore_type spr_uncore_pcu = { @@ -5765,6 +5776,7 @@ static struct intel_uncore_type spr_uncore_upi = { static struct intel_uncore_type spr_uncore_m3upi = { SPR_UNCORE_PCI_COMMON_FORMAT(), .name = "m3upi", + .constraints = icx_uncore_m3upi_constraints, }; static struct intel_uncore_type spr_uncore_mdf = { diff --git a/arch/x86/events/msr.c b/arch/x86/events/msr.c index c853b28efa33..96c775abe31f 100644 --- a/arch/x86/events/msr.c +++ b/arch/x86/events/msr.c @@ -68,6 +68,7 @@ static bool test_intel(int idx, void *data) case INTEL_FAM6_BROADWELL_D: case INTEL_FAM6_BROADWELL_G: case INTEL_FAM6_BROADWELL_X: + case INTEL_FAM6_SAPPHIRERAPIDS_X: case INTEL_FAM6_ATOM_SILVERMONT: case INTEL_FAM6_ATOM_SILVERMONT_D: diff --git a/arch/x86/events/perf_event.h b/arch/x86/events/perf_event.h index e3ac05c97b5e..5480db242083 100644 --- a/arch/x86/events/perf_event.h +++ b/arch/x86/events/perf_event.h @@ -14,6 +14,7 @@ #include <linux/perf_event.h> +#include <asm/fpu/xstate.h> #include <asm/intel_ds.h> #include <asm/cpu.h> @@ -726,6 +727,7 @@ struct x86_pmu { void (*enable_all)(int added); void (*enable)(struct perf_event *); void (*disable)(struct perf_event *); + void (*assign)(struct perf_event *event, int idx); void (*add)(struct perf_event *); void (*del)(struct perf_event *); void (*read)(struct perf_event *event); @@ -1240,6 +1242,25 @@ static inline bool intel_pmu_has_bts(struct perf_event *event) return intel_pmu_has_bts_period(event, hwc->sample_period); } +static __always_inline void __intel_pmu_pebs_disable_all(void) +{ + wrmsrl(MSR_IA32_PEBS_ENABLE, 0); +} + +static __always_inline void __intel_pmu_arch_lbr_disable(void) +{ + wrmsrl(MSR_ARCH_LBR_CTL, 0); +} + +static __always_inline void __intel_pmu_lbr_disable(void) +{ + u64 debugctl; + + rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); + debugctl &= ~(DEBUGCTLMSR_LBR | DEBUGCTLMSR_FREEZE_LBRS_ON_PMI); + wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl); +} + int intel_pmu_save_and_restart(struct perf_event *event); struct event_constraint * |