diff options
Diffstat (limited to 'arch/x86/events/intel/pt.c')
| -rw-r--r-- | arch/x86/events/intel/pt.c | 775 |
1 files changed, 560 insertions, 215 deletions
diff --git a/arch/x86/events/intel/pt.c b/arch/x86/events/intel/pt.c index 9494ca68fd9d..44524a387c58 100644 --- a/arch/x86/events/intel/pt.c +++ b/arch/x86/events/intel/pt.c @@ -1,16 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Intel(R) Processor Trace PMU driver for perf * Copyright (c) 2013-2014, 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. - * * Intel PT is specified in the Intel Architecture Instruction Set Extensions * Programming Reference: * http://software.intel.com/en-us/intel-isa-extensions @@ -21,14 +13,19 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt #include <linux/types.h> +#include <linux/bits.h> +#include <linux/limits.h> #include <linux/slab.h> #include <linux/device.h> +#include <linux/kvm_types.h> +#include <asm/cpuid/api.h> #include <asm/perf_event.h> #include <asm/insn.h> #include <asm/io.h> #include <asm/intel_pt.h> -#include <asm/intel-family.h> +#include <asm/cpu_device_id.h> +#include <asm/msr.h> #include "../perf_event.h" #include "pt.h" @@ -65,12 +62,14 @@ static struct pt_cap_desc { PT_CAP(mtc, 0, CPUID_EBX, BIT(3)), PT_CAP(ptwrite, 0, CPUID_EBX, BIT(4)), PT_CAP(power_event_trace, 0, CPUID_EBX, BIT(5)), + PT_CAP(event_trace, 0, CPUID_EBX, BIT(7)), + PT_CAP(tnt_disable, 0, CPUID_EBX, BIT(8)), PT_CAP(topa_output, 0, CPUID_ECX, BIT(0)), PT_CAP(topa_multiple_entries, 0, CPUID_ECX, BIT(1)), PT_CAP(single_range_output, 0, CPUID_ECX, BIT(2)), PT_CAP(output_subsys, 0, CPUID_ECX, BIT(3)), PT_CAP(payloads_lip, 0, CPUID_ECX, BIT(31)), - PT_CAP(num_address_ranges, 1, CPUID_EAX, 0x3), + PT_CAP(num_address_ranges, 1, CPUID_EAX, 0x7), PT_CAP(mtc_periods, 1, CPUID_EAX, 0xffff0000), PT_CAP(cycle_thresholds, 1, CPUID_EBX, 0xffff), PT_CAP(psb_periods, 1, CPUID_EBX, 0xffff0000), @@ -84,13 +83,13 @@ u32 intel_pt_validate_cap(u32 *caps, enum pt_capabilities capability) return (c & cd->mask) >> shift; } -EXPORT_SYMBOL_GPL(intel_pt_validate_cap); +EXPORT_SYMBOL_FOR_KVM(intel_pt_validate_cap); u32 intel_pt_validate_hw_cap(enum pt_capabilities cap) { return intel_pt_validate_cap(pt_pmu.caps, cap); } -EXPORT_SYMBOL_GPL(intel_pt_validate_hw_cap); +EXPORT_SYMBOL_FOR_KVM(intel_pt_validate_hw_cap); static ssize_t pt_cap_show(struct device *cdev, struct device_attribute *attr, @@ -116,6 +115,8 @@ PMU_FORMAT_ATTR(tsc, "config:10" ); PMU_FORMAT_ATTR(noretcomp, "config:11" ); PMU_FORMAT_ATTR(ptw, "config:12" ); PMU_FORMAT_ATTR(branch, "config:13" ); +PMU_FORMAT_ATTR(event, "config:31" ); +PMU_FORMAT_ATTR(notnt, "config:55" ); PMU_FORMAT_ATTR(mtc_period, "config:14-17" ); PMU_FORMAT_ATTR(cyc_thresh, "config:19-22" ); PMU_FORMAT_ATTR(psb_period, "config:24-27" ); @@ -124,6 +125,8 @@ static struct attribute *pt_formats_attr[] = { &format_attr_pt.attr, &format_attr_cyc.attr, &format_attr_pwr_evt.attr, + &format_attr_event.attr, + &format_attr_notnt.attr, &format_attr_fup_on_ptw.attr, &format_attr_mtc.attr, &format_attr_tsc.attr, @@ -193,7 +196,7 @@ static int __init pt_pmu_hw_init(void) int ret; long i; - rdmsrl(MSR_PLATFORM_INFO, reg); + rdmsrq(MSR_PLATFORM_INFO, reg); pt_pmu.max_nonturbo_ratio = (reg & 0xff00) >> 8; /* @@ -201,21 +204,21 @@ static int __init pt_pmu_hw_init(void) * otherwise, zero for numerator stands for "not enumerated" * as per SDM */ - if (boot_cpu_data.cpuid_level >= CPUID_TSC_LEAF) { + if (boot_cpu_data.cpuid_level >= CPUID_LEAF_TSC) { u32 eax, ebx, ecx, edx; - cpuid(CPUID_TSC_LEAF, &eax, &ebx, &ecx, &edx); + cpuid(CPUID_LEAF_TSC, &eax, &ebx, &ecx, &edx); pt_pmu.tsc_art_num = ebx; pt_pmu.tsc_art_den = eax; } /* model-specific quirks */ - switch (boot_cpu_data.x86_model) { - case INTEL_FAM6_BROADWELL_CORE: - case INTEL_FAM6_BROADWELL_XEON_D: - case INTEL_FAM6_BROADWELL_GT3E: - case INTEL_FAM6_BROADWELL_X: + switch (boot_cpu_data.x86_vfm) { + case INTEL_BROADWELL: + case INTEL_BROADWELL_D: + case INTEL_BROADWELL_G: + case INTEL_BROADWELL_X: /* not setting BRANCH_EN will #GP, erratum BDM106 */ pt_pmu.branch_en_always_on = true; break; @@ -229,13 +232,11 @@ static int __init pt_pmu_hw_init(void) * "IA32_VMX_MISC[bit 14]" being 1 means PT can trace * post-VMXON. */ - rdmsrl(MSR_IA32_VMX_MISC, reg); + rdmsrq(MSR_IA32_VMX_MISC, reg); if (reg & BIT(14)) pt_pmu.vmx = true; } - attrs = NULL; - for (i = 0; i < PT_CPUID_LEAVES; i++) { cpuid_count(20, i, &pt_pmu.caps[CPUID_EAX + i*PT_CPUID_REGS_NUM], @@ -306,6 +307,8 @@ fail: RTIT_CTL_CYC_PSB | \ RTIT_CTL_MTC | \ RTIT_CTL_PWR_EVT_EN | \ + RTIT_CTL_EVENT_EN | \ + RTIT_CTL_NOTNT | \ RTIT_CTL_FUP_ON_PTW | \ RTIT_CTL_PTW_EN) @@ -360,6 +363,14 @@ static bool pt_event_valid(struct perf_event *event) !intel_pt_validate_hw_cap(PT_CAP_power_event_trace)) return false; + if (config & RTIT_CTL_EVENT_EN && + !intel_pt_validate_hw_cap(PT_CAP_event_trace)) + return false; + + if (config & RTIT_CTL_NOTNT && + !intel_pt_validate_hw_cap(PT_CAP_tnt_disable)) + return false; + if (config & RTIT_CTL_PTW) { if (!intel_pt_validate_hw_cap(PT_CAP_ptwrite)) return false; @@ -372,7 +383,7 @@ static bool pt_event_valid(struct perf_event *event) /* * Setting bit 0 (TraceEn in RTIT_CTL MSR) in the attr.config - * clears the assomption that BranchEn must always be enabled, + * clears the assumption that BranchEn must always be enabled, * as was the case with the first implementation of PT. * If this bit is not set, the legacy behavior is preserved * for compatibility with the older userspace. @@ -405,6 +416,23 @@ static bool pt_event_valid(struct perf_event *event) * These all are cpu affine and operate on a local PT */ +static void pt_config_start(struct perf_event *event) +{ + struct pt *pt = this_cpu_ptr(&pt_ctx); + u64 ctl = event->hw.aux_config; + + if (READ_ONCE(event->hw.aux_paused)) + return; + + ctl |= RTIT_CTL_TRACEEN; + if (READ_ONCE(pt->vmx_on)) + perf_aux_output_flag(&pt->handle, PERF_AUX_FLAG_PARTIAL); + else + wrmsrq(MSR_IA32_RTIT_CTL, ctl); + + WRITE_ONCE(event->hw.aux_config, ctl); +} + /* Address ranges and their corresponding msr configuration registers */ static const struct pt_address_range { unsigned long msr_a; @@ -459,16 +487,16 @@ static u64 pt_config_filters(struct perf_event *event) /* avoid redundant msr writes */ if (pt->filters.filter[range].msr_a != filter->msr_a) { - wrmsrl(pt_address_ranges[range].msr_a, filter->msr_a); + wrmsrq(pt_address_ranges[range].msr_a, filter->msr_a); pt->filters.filter[range].msr_a = filter->msr_a; } if (pt->filters.filter[range].msr_b != filter->msr_b) { - wrmsrl(pt_address_ranges[range].msr_b, filter->msr_b); + wrmsrq(pt_address_ranges[range].msr_b, filter->msr_b); pt->filters.filter[range].msr_b = filter->msr_b; } - rtit_ctl |= filter->config << pt_address_ranges[range].reg_off; + rtit_ctl |= (u64)filter->config << pt_address_ranges[range].reg_off; } return rtit_ctl; @@ -477,16 +505,19 @@ static u64 pt_config_filters(struct perf_event *event) static void pt_config(struct perf_event *event) { struct pt *pt = this_cpu_ptr(&pt_ctx); + struct pt_buffer *buf = perf_get_aux(&pt->handle); u64 reg; /* First round: clear STATUS, in particular the PSB byte counter. */ - if (!event->hw.config) { + if (!event->hw.aux_config) { perf_event_itrace_started(event); - wrmsrl(MSR_IA32_RTIT_STATUS, 0); + wrmsrq(MSR_IA32_RTIT_STATUS, 0); } reg = pt_config_filters(event); - reg |= RTIT_CTL_TOPA | RTIT_CTL_TRACEEN; + reg |= RTIT_CTL_TRACEEN; + if (!buf->single) + reg |= RTIT_CTL_TOPA; /* * Previously, we had BRANCH_EN on by default, but now that PT has @@ -508,17 +539,31 @@ static void pt_config(struct perf_event *event) reg |= (event->attr.config & PT_CONFIG_MASK); - event->hw.config = reg; - if (READ_ONCE(pt->vmx_on)) - perf_aux_output_flag(&pt->handle, PERF_AUX_FLAG_PARTIAL); - else - wrmsrl(MSR_IA32_RTIT_CTL, reg); + event->hw.aux_config = reg; + + /* + * Allow resume before starting so as not to overwrite a value set by a + * PMI. + */ + barrier(); + WRITE_ONCE(pt->resume_allowed, 1); + /* Configuration is complete, it is now OK to handle an NMI */ + barrier(); + WRITE_ONCE(pt->handle_nmi, 1); + barrier(); + pt_config_start(event); + barrier(); + /* + * Allow pause after starting so its pt_config_stop() doesn't race with + * pt_config_start(). + */ + WRITE_ONCE(pt->pause_allowed, 1); } static void pt_config_stop(struct perf_event *event) { struct pt *pt = this_cpu_ptr(&pt_ctx); - u64 ctl = READ_ONCE(event->hw.config); + u64 ctl = READ_ONCE(event->hw.aux_config); /* may be already stopped by a PMI */ if (!(ctl & RTIT_CTL_TRACEEN)) @@ -526,9 +571,9 @@ static void pt_config_stop(struct perf_event *event) ctl &= ~RTIT_CTL_TRACEEN; if (!READ_ONCE(pt->vmx_on)) - wrmsrl(MSR_IA32_RTIT_CTL, ctl); + wrmsrq(MSR_IA32_RTIT_CTL, ctl); - WRITE_ONCE(event->hw.config, ctl); + WRITE_ONCE(event->hw.aux_config, ctl); /* * A wrmsr that disables trace generation serializes other PT @@ -541,45 +586,89 @@ static void pt_config_stop(struct perf_event *event) wmb(); } -static void pt_config_buffer(void *buf, unsigned int topa_idx, - unsigned int output_off) -{ - u64 reg; - - wrmsrl(MSR_IA32_RTIT_OUTPUT_BASE, virt_to_phys(buf)); - - reg = 0x7f | ((u64)topa_idx << 7) | ((u64)output_off << 32); - - wrmsrl(MSR_IA32_RTIT_OUTPUT_MASK, reg); -} - -/* - * Keep ToPA table-related metadata on the same page as the actual table, - * taking up a few words from the top - */ - -#define TENTS_PER_PAGE (((PAGE_SIZE - 40) / sizeof(struct topa_entry)) - 1) - /** - * struct topa - page-sized ToPA table with metadata at the top - * @table: actual ToPA table entries, as understood by PT hardware + * struct topa - ToPA metadata * @list: linkage to struct pt_buffer's list of tables - * @phys: physical address of this page * @offset: offset of the first entry in this table in the buffer * @size: total size of all entries in this table * @last: index of the last initialized entry in this table + * @z_count: how many times the first entry repeats */ struct topa { - struct topa_entry table[TENTS_PER_PAGE]; struct list_head list; - u64 phys; u64 offset; size_t size; int last; + unsigned int z_count; }; +/* + * Keep ToPA table-related metadata on the same page as the actual table, + * taking up a few words from the top + */ + +#define TENTS_PER_PAGE \ + ((PAGE_SIZE - sizeof(struct topa)) / sizeof(struct topa_entry)) + +/** + * struct topa_page - page-sized ToPA table with metadata at the top + * @table: actual ToPA table entries, as understood by PT hardware + * @topa: metadata + */ +struct topa_page { + struct topa_entry table[TENTS_PER_PAGE]; + struct topa topa; +}; + +static inline struct topa_page *topa_to_page(struct topa *topa) +{ + return container_of(topa, struct topa_page, topa); +} + +static inline struct topa_page *topa_entry_to_page(struct topa_entry *te) +{ + return (struct topa_page *)((unsigned long)te & PAGE_MASK); +} + +static inline phys_addr_t topa_pfn(struct topa *topa) +{ + return PFN_DOWN(virt_to_phys(topa_to_page(topa))); +} + /* make -1 stand for the last table entry */ -#define TOPA_ENTRY(t, i) ((i) == -1 ? &(t)->table[(t)->last] : &(t)->table[(i)]) +#define TOPA_ENTRY(t, i) \ + ((i) == -1 \ + ? &topa_to_page(t)->table[(t)->last] \ + : &topa_to_page(t)->table[(i)]) +#define TOPA_ENTRY_SIZE(t, i) (sizes(TOPA_ENTRY((t), (i))->size)) +#define TOPA_ENTRY_PAGES(t, i) (1 << TOPA_ENTRY((t), (i))->size) + +static void pt_config_buffer(struct pt_buffer *buf) +{ + struct pt *pt = this_cpu_ptr(&pt_ctx); + u64 reg, mask; + void *base; + + if (buf->single) { + base = buf->data_pages[0]; + mask = (buf->nr_pages * PAGE_SIZE - 1) >> 7; + } else { + base = topa_to_page(buf->cur)->table; + mask = (u64)buf->cur_idx; + } + + reg = virt_to_phys(base); + if (pt->output_base != reg) { + pt->output_base = reg; + wrmsrq(MSR_IA32_RTIT_OUTPUT_BASE, reg); + } + + reg = 0x7f | (mask << 7) | ((u64)buf->output_off << 32); + if (pt->output_mask != reg) { + pt->output_mask = reg; + wrmsrq(MSR_IA32_RTIT_OUTPUT_MASK, reg); + } +} /** * topa_alloc() - allocate page-sized ToPA table @@ -591,27 +680,26 @@ struct topa { static struct topa *topa_alloc(int cpu, gfp_t gfp) { int node = cpu_to_node(cpu); - struct topa *topa; + struct topa_page *tp; struct page *p; p = alloc_pages_node(node, gfp | __GFP_ZERO, 0); if (!p) return NULL; - topa = page_address(p); - topa->last = 0; - topa->phys = page_to_phys(p); + tp = page_address(p); + tp->topa.last = 0; /* * In case of singe-entry ToPA, always put the self-referencing END * link as the 2nd entry in the table */ if (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries)) { - TOPA_ENTRY(topa, 1)->base = topa->phys >> TOPA_SHIFT; - TOPA_ENTRY(topa, 1)->end = 1; + TOPA_ENTRY(&tp->topa, 1)->base = page_to_phys(p) >> TOPA_SHIFT; + TOPA_ENTRY(&tp->topa, 1)->end = 1; } - return topa; + return &tp->topa; } /** @@ -651,7 +739,7 @@ static void topa_insert_table(struct pt_buffer *buf, struct topa *topa) BUG_ON(last->last != TENTS_PER_PAGE - 1); - TOPA_ENTRY(last, -1)->base = topa->phys >> TOPA_SHIFT; + TOPA_ENTRY(last, -1)->base = topa_pfn(topa); TOPA_ENTRY(last, -1)->end = 1; } @@ -671,6 +759,7 @@ static bool topa_table_full(struct topa *topa) /** * topa_insert_pages() - create a list of ToPA tables * @buf: PT buffer being initialized. + * @cpu: CPU on which to allocate. * @gfp: Allocation flags. * * This initializes a list of ToPA tables with entries from @@ -678,7 +767,7 @@ static bool topa_table_full(struct topa *topa) * * Return: 0 on success or error code. */ -static int topa_insert_pages(struct pt_buffer *buf, gfp_t gfp) +static int topa_insert_pages(struct pt_buffer *buf, int cpu, gfp_t gfp) { struct topa *topa = buf->last; int order = 0; @@ -689,13 +778,18 @@ static int topa_insert_pages(struct pt_buffer *buf, gfp_t gfp) order = page_private(p); if (topa_table_full(topa)) { - topa = topa_alloc(buf->cpu, gfp); + topa = topa_alloc(cpu, gfp); if (!topa) return -ENOMEM; topa_insert_table(buf, topa); } + if (topa->z_count == topa->last - 1) { + if (order == TOPA_ENTRY(topa, topa->last - 1)->size) + topa->z_count++; + } + TOPA_ENTRY(topa, -1)->base = page_to_phys(p) >> TOPA_SHIFT; TOPA_ENTRY(topa, -1)->size = order; if (!buf->snapshot && @@ -721,23 +815,26 @@ static void pt_topa_dump(struct pt_buffer *buf) struct topa *topa; list_for_each_entry(topa, &buf->tables, list) { + struct topa_page *tp = topa_to_page(topa); int i; - pr_debug("# table @%p (%016Lx), off %llx size %zx\n", topa->table, - topa->phys, topa->offset, topa->size); + pr_debug("# table @%p, off %llx size %zx\n", tp->table, + topa->offset, topa->size); for (i = 0; i < TENTS_PER_PAGE; i++) { pr_debug("# entry @%p (%lx sz %u %c%c%c) raw=%16llx\n", - &topa->table[i], - (unsigned long)topa->table[i].base << TOPA_SHIFT, - sizes(topa->table[i].size), - topa->table[i].end ? 'E' : ' ', - topa->table[i].intr ? 'I' : ' ', - topa->table[i].stop ? 'S' : ' ', - *(u64 *)&topa->table[i]); + &tp->table[i], + (unsigned long)tp->table[i].base << TOPA_SHIFT, + sizes(tp->table[i].size), + tp->table[i].end ? 'E' : ' ', + tp->table[i].intr ? 'I' : ' ', + tp->table[i].stop ? 'S' : ' ', + *(u64 *)&tp->table[i]); if ((intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries) && - topa->table[i].stop) || - topa->table[i].end) + tp->table[i].stop) || + tp->table[i].end) break; + if (!i && topa->z_count) + i += topa->z_count; } } } @@ -754,11 +851,13 @@ static void pt_buffer_advance(struct pt_buffer *buf) buf->cur_idx++; if (buf->cur_idx == buf->cur->last) { - if (buf->cur == buf->last) + if (buf->cur == buf->last) { buf->cur = buf->first; - else + buf->wrapped = true; + } else { buf->cur = list_entry(buf->cur->list.next, struct topa, list); + } buf->cur_idx = 0; } } @@ -772,21 +871,29 @@ static void pt_buffer_advance(struct pt_buffer *buf) static void pt_update_head(struct pt *pt) { struct pt_buffer *buf = perf_get_aux(&pt->handle); + bool wrapped = buf->wrapped; u64 topa_idx, base, old; + buf->wrapped = false; + + if (buf->single) { + local_set(&buf->data_size, buf->output_off); + return; + } + /* offset of the first region in this table from the beginning of buf */ base = buf->cur->offset + buf->output_off; /* offset of the current output region within this table */ for (topa_idx = 0; topa_idx < buf->cur_idx; topa_idx++) - base += sizes(buf->cur->table[topa_idx].size); + base += TOPA_ENTRY_SIZE(buf->cur, topa_idx); if (buf->snapshot) { local_set(&buf->data_size, base); } else { old = (local64_xchg(&buf->head, base) & ((buf->nr_pages << PAGE_SHIFT) - 1)); - if (base < old) + if (base < old || (base == old && wrapped)) base += buf->nr_pages << PAGE_SHIFT; local_add(base - old, &buf->data_size); @@ -799,7 +906,7 @@ static void pt_update_head(struct pt *pt) */ static void *pt_buffer_region(struct pt_buffer *buf) { - return phys_to_virt(buf->cur->table[buf->cur_idx].base << TOPA_SHIFT); + return phys_to_virt((phys_addr_t)TOPA_ENTRY(buf->cur, buf->cur_idx)->base << TOPA_SHIFT); } /** @@ -808,7 +915,7 @@ static void *pt_buffer_region(struct pt_buffer *buf) */ static size_t pt_buffer_region_size(struct pt_buffer *buf) { - return sizes(buf->cur->table[buf->cur_idx].size); + return TOPA_ENTRY_SIZE(buf->cur, buf->cur_idx); } /** @@ -821,7 +928,7 @@ static void pt_handle_status(struct pt *pt) int advance = 0; u64 status; - rdmsrl(MSR_IA32_RTIT_STATUS, status); + rdmsrq(MSR_IA32_RTIT_STATUS, status); if (status & RTIT_STATUS_ERROR) { pr_err_ratelimited("ToPA ERROR encountered, trying to recover\n"); @@ -837,8 +944,9 @@ static void pt_handle_status(struct pt *pt) * means we are already losing data; need to let the decoder * know. */ - if (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries) || - buf->output_off == sizes(TOPA_ENTRY(buf->cur, buf->cur_idx)->size)) { + if (!buf->single && + (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries) || + buf->output_off == pt_buffer_region_size(buf))) { perf_aux_output_flag(&pt->handle, PERF_AUX_FLAG_TRUNCATED); advance++; @@ -864,7 +972,7 @@ static void pt_handle_status(struct pt *pt) if (advance) pt_buffer_advance(buf); - wrmsrl(MSR_IA32_RTIT_STATUS, status); + wrmsrq(MSR_IA32_RTIT_STATUS, status); } /** @@ -875,41 +983,114 @@ static void pt_handle_status(struct pt *pt) */ static void pt_read_offset(struct pt_buffer *buf) { - u64 offset, base_topa; + struct pt *pt = this_cpu_ptr(&pt_ctx); + struct topa_page *tp; - rdmsrl(MSR_IA32_RTIT_OUTPUT_BASE, base_topa); - buf->cur = phys_to_virt(base_topa); + if (!buf->single) { + rdmsrq(MSR_IA32_RTIT_OUTPUT_BASE, pt->output_base); + tp = phys_to_virt(pt->output_base); + buf->cur = &tp->topa; + } - rdmsrl(MSR_IA32_RTIT_OUTPUT_MASK, offset); + rdmsrq(MSR_IA32_RTIT_OUTPUT_MASK, pt->output_mask); /* offset within current output region */ - buf->output_off = offset >> 32; + buf->output_off = pt->output_mask >> 32; /* index of current output region within this table */ - buf->cur_idx = (offset & 0xffffff80) >> 7; + if (!buf->single) + buf->cur_idx = (pt->output_mask & 0xffffff80) >> 7; } -/** - * pt_topa_next_entry() - obtain index of the first page in the next ToPA entry - * @buf: PT buffer. - * @pg: Page offset in the buffer. - * - * When advancing to the next output region (ToPA entry), given a page offset - * into the buffer, we need to find the offset of the first page in the next - * region. - */ -static unsigned int pt_topa_next_entry(struct pt_buffer *buf, unsigned int pg) +static struct topa_entry * +pt_topa_entry_for_page(struct pt_buffer *buf, unsigned int pg) +{ + struct topa_page *tp; + struct topa *topa; + unsigned int idx, cur_pg = 0, z_pg = 0, start_idx = 0; + + /* + * Indicates a bug in the caller. + */ + if (WARN_ON_ONCE(pg >= buf->nr_pages)) + return NULL; + + /* + * First, find the ToPA table where @pg fits. With high + * order allocations, there shouldn't be many of these. + */ + list_for_each_entry(topa, &buf->tables, list) { + if (topa->offset + topa->size > (unsigned long)pg << PAGE_SHIFT) + goto found; + } + + /* + * Hitting this means we have a problem in the ToPA + * allocation code. + */ + WARN_ON_ONCE(1); + + return NULL; + +found: + /* + * Indicates a problem in the ToPA allocation code. + */ + if (WARN_ON_ONCE(topa->last == -1)) + return NULL; + + tp = topa_to_page(topa); + cur_pg = PFN_DOWN(topa->offset); + if (topa->z_count) { + z_pg = TOPA_ENTRY_PAGES(topa, 0) * (topa->z_count + 1); + start_idx = topa->z_count + 1; + } + + /* + * Multiple entries at the beginning of the table have the same size, + * ideally all of them; if @pg falls there, the search is done. + */ + if (pg >= cur_pg && pg < cur_pg + z_pg) { + idx = (pg - cur_pg) / TOPA_ENTRY_PAGES(topa, 0); + return &tp->table[idx]; + } + + /* + * Otherwise, slow path: iterate through the remaining entries. + */ + for (idx = start_idx, cur_pg += z_pg; idx < topa->last; idx++) { + if (cur_pg + TOPA_ENTRY_PAGES(topa, idx) > pg) + return &tp->table[idx]; + + cur_pg += TOPA_ENTRY_PAGES(topa, idx); + } + + /* + * Means we couldn't find a ToPA entry in the table that does match. + */ + WARN_ON_ONCE(1); + + return NULL; +} + +static struct topa_entry * +pt_topa_prev_entry(struct pt_buffer *buf, struct topa_entry *te) { - struct topa_entry *te = buf->topa_index[pg]; + unsigned long table = (unsigned long)te & ~(PAGE_SIZE - 1); + struct topa_page *tp; + struct topa *topa; - /* one region */ - if (buf->first == buf->last && buf->first->last == 1) - return pg; + tp = (struct topa_page *)table; + if (tp->table != te) + return --te; - do { - pg++; - pg &= buf->nr_pages - 1; - } while (buf->topa_index[pg] == te); + topa = &tp->topa; + if (topa == buf->first) + topa = buf->last; + else + topa = list_prev_entry(topa, list); - return pg; + tp = topa_to_page(topa); + + return &tp->table[topa->last - 1]; } /** @@ -932,9 +1113,11 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf, unsigned long head = local64_read(&buf->head); unsigned long idx, npages, wakeup; + if (buf->single) + return 0; + /* can't stop in the middle of an output region */ - if (buf->output_off + handle->size + 1 < - sizes(TOPA_ENTRY(buf->cur, buf->cur_idx)->size)) { + if (buf->output_off + handle->size + 1 < pt_buffer_region_size(buf)) { perf_aux_output_flag(handle, PERF_AUX_FLAG_TRUNCATED); return -EINVAL; } @@ -945,9 +1128,13 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf, return 0; /* clear STOP and INT from current entry */ - buf->topa_index[buf->stop_pos]->stop = 0; - buf->topa_index[buf->stop_pos]->intr = 0; - buf->topa_index[buf->intr_pos]->intr = 0; + if (buf->stop_te) { + buf->stop_te->stop = 0; + buf->stop_te->intr = 0; + } + + if (buf->intr_te) + buf->intr_te->intr = 0; /* how many pages till the STOP marker */ npages = handle->size >> PAGE_SHIFT; @@ -958,7 +1145,12 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf, idx = (head >> PAGE_SHIFT) + npages; idx &= buf->nr_pages - 1; - buf->stop_pos = idx; + + if (idx != buf->stop_pos) { + buf->stop_pos = idx; + buf->stop_te = pt_topa_entry_for_page(buf, idx); + buf->stop_te = pt_topa_prev_entry(buf, buf->stop_te); + } wakeup = handle->wakeup >> PAGE_SHIFT; @@ -968,51 +1160,20 @@ static int pt_buffer_reset_markers(struct pt_buffer *buf, idx = wakeup; idx &= buf->nr_pages - 1; - buf->intr_pos = idx; + if (idx != buf->intr_pos) { + buf->intr_pos = idx; + buf->intr_te = pt_topa_entry_for_page(buf, idx); + buf->intr_te = pt_topa_prev_entry(buf, buf->intr_te); + } - buf->topa_index[buf->stop_pos]->stop = 1; - buf->topa_index[buf->stop_pos]->intr = 1; - buf->topa_index[buf->intr_pos]->intr = 1; + buf->stop_te->stop = 1; + buf->stop_te->intr = 1; + buf->intr_te->intr = 1; return 0; } /** - * pt_buffer_setup_topa_index() - build topa_index[] table of regions - * @buf: PT buffer. - * - * topa_index[] references output regions indexed by offset into the - * buffer for purposes of quick reverse lookup. - */ -static void pt_buffer_setup_topa_index(struct pt_buffer *buf) -{ - struct topa *cur = buf->first, *prev = buf->last; - struct topa_entry *te_cur = TOPA_ENTRY(cur, 0), - *te_prev = TOPA_ENTRY(prev, prev->last - 1); - int pg = 0, idx = 0; - - while (pg < buf->nr_pages) { - int tidx; - - /* pages within one topa entry */ - for (tidx = 0; tidx < 1 << te_cur->size; tidx++, pg++) - buf->topa_index[pg] = te_prev; - - te_prev = te_cur; - - if (idx == cur->last - 1) { - /* advance to next topa table */ - idx = 0; - cur = list_entry(cur->list.next, struct topa, list); - } else { - idx++; - } - te_cur = TOPA_ENTRY(cur, idx); - } - -} - -/** * pt_buffer_reset_offsets() - adjust buffer's write pointers from aux_head * @buf: PT buffer. * @head: Write pointer (aux_head) from AUX buffer. @@ -1029,18 +1190,24 @@ static void pt_buffer_setup_topa_index(struct pt_buffer *buf) */ static void pt_buffer_reset_offsets(struct pt_buffer *buf, unsigned long head) { + struct topa_page *cur_tp; + struct topa_entry *te; int pg; if (buf->snapshot) head &= (buf->nr_pages << PAGE_SHIFT) - 1; - pg = (head >> PAGE_SHIFT) & (buf->nr_pages - 1); - pg = pt_topa_next_entry(buf, pg); + if (!buf->single) { + pg = (head >> PAGE_SHIFT) & (buf->nr_pages - 1); + te = pt_topa_entry_for_page(buf, pg); - buf->cur = (struct topa *)((unsigned long)buf->topa_index[pg] & PAGE_MASK); - buf->cur_idx = ((unsigned long)buf->topa_index[pg] - - (unsigned long)buf->cur) / sizeof(struct topa_entry); - buf->output_off = head & (sizes(buf->cur->table[buf->cur_idx].size) - 1); + cur_tp = topa_entry_to_page(te); + buf->cur = &cur_tp->topa; + buf->cur_idx = te - TOPA_ENTRY(buf->cur, 0); + buf->output_off = head & (pt_buffer_region_size(buf) - 1); + } else { + buf->output_off = head; + } local64_set(&buf->head, head); local_set(&buf->data_size, 0); @@ -1054,6 +1221,9 @@ static void pt_buffer_fini_topa(struct pt_buffer *buf) { struct topa *topa, *iter; + if (buf->single) + return; + list_for_each_entry_safe(topa, iter, &buf->tables, list) { /* * right now, this is in free_aux() path only, so @@ -1066,34 +1236,35 @@ static void pt_buffer_fini_topa(struct pt_buffer *buf) /** * pt_buffer_init_topa() - initialize ToPA table for pt buffer * @buf: PT buffer. - * @size: Total size of all regions within this ToPA. + * @cpu: CPU on which to allocate. + * @nr_pages: No. of pages to allocate. * @gfp: Allocation flags. + * + * Return: 0 on success or error code. */ -static int pt_buffer_init_topa(struct pt_buffer *buf, unsigned long nr_pages, - gfp_t gfp) +static int pt_buffer_init_topa(struct pt_buffer *buf, int cpu, + unsigned long nr_pages, gfp_t gfp) { struct topa *topa; int err; - topa = topa_alloc(buf->cpu, gfp); + topa = topa_alloc(cpu, gfp); if (!topa) return -ENOMEM; topa_insert_table(buf, topa); while (buf->nr_pages < nr_pages) { - err = topa_insert_pages(buf, gfp); + err = topa_insert_pages(buf, cpu, gfp); if (err) { pt_buffer_fini_topa(buf); return -ENOMEM; } } - pt_buffer_setup_topa_index(buf); - /* link last table to the first one, unless we're double buffering */ if (intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries)) { - TOPA_ENTRY(buf->last, -1)->base = buf->first->phys >> TOPA_SHIFT; + TOPA_ENTRY(buf->last, -1)->base = topa_pfn(buf->first); TOPA_ENTRY(buf->last, -1)->end = 1; } @@ -1101,9 +1272,48 @@ static int pt_buffer_init_topa(struct pt_buffer *buf, unsigned long nr_pages, return 0; } +static int pt_buffer_try_single(struct pt_buffer *buf, int nr_pages) +{ + struct page *p = virt_to_page(buf->data_pages[0]); + int ret = -ENOTSUPP, order = 0; + + /* + * We can use single range output mode + * + in snapshot mode, where we don't need interrupts; + * + if the hardware supports it; + * + if the entire buffer is one contiguous allocation. + */ + if (!buf->snapshot) + goto out; + + if (!intel_pt_validate_hw_cap(PT_CAP_single_range_output)) + goto out; + + if (PagePrivate(p)) + order = page_private(p); + + if (1 << order != nr_pages) + goto out; + + /* + * Some processors cannot always support single range for more than + * 4KB - refer errata TGL052, ADL037 and RPL017. Future processors might + * also be affected, so for now rather than trying to keep track of + * which ones, just disable it for all. + */ + if (nr_pages > 1) + goto out; + + buf->single = true; + buf->nr_pages = nr_pages; + ret = 0; +out: + return ret; +} + /** * pt_buffer_setup_aux() - set up topa tables for a PT buffer - * @cpu: Cpu on which to allocate, -1 means current. + * @event: Performance event * @pages: Array of pointers to buffer pages passed from perf core. * @nr_pages: Number of pages in the buffer. * @snapshot: If this is a snapshot/overwrite counter. @@ -1114,30 +1324,42 @@ static int pt_buffer_init_topa(struct pt_buffer *buf, unsigned long nr_pages, * Return: Our private PT buffer structure. */ static void * -pt_buffer_setup_aux(int cpu, void **pages, int nr_pages, bool snapshot) +pt_buffer_setup_aux(struct perf_event *event, void **pages, + int nr_pages, bool snapshot) { struct pt_buffer *buf; - int node, ret; + int node, ret, cpu = event->cpu; if (!nr_pages) return NULL; + /* + * Only support AUX sampling in snapshot mode, where we don't + * generate NMIs. + */ + if (event->attr.aux_sample_size && !snapshot) + return NULL; + if (cpu == -1) cpu = raw_smp_processor_id(); node = cpu_to_node(cpu); - buf = kzalloc_node(offsetof(struct pt_buffer, topa_index[nr_pages]), - GFP_KERNEL, node); + buf = kzalloc_node(sizeof(struct pt_buffer), GFP_KERNEL, node); if (!buf) return NULL; - buf->cpu = cpu; buf->snapshot = snapshot; buf->data_pages = pages; + buf->stop_pos = -1; + buf->intr_pos = -1; INIT_LIST_HEAD(&buf->tables); - ret = pt_buffer_init_topa(buf, nr_pages, GFP_KERNEL); + ret = pt_buffer_try_single(buf, nr_pages); + if (!ret) + return buf; + + ret = pt_buffer_init_topa(buf, cpu, nr_pages, GFP_KERNEL); if (ret) { kfree(buf); return NULL; @@ -1185,11 +1407,27 @@ static void pt_addr_filters_fini(struct perf_event *event) event->hw.addr_filters = NULL; } -static inline bool valid_kernel_ip(unsigned long ip) +#ifdef CONFIG_X86_64 +/* Clamp to a canonical address greater-than-or-equal-to the address given */ +static u64 clamp_to_ge_canonical_addr(u64 vaddr, u8 vaddr_bits) { - return virt_addr_valid(ip) && kernel_ip(ip); + return __is_canonical_address(vaddr, vaddr_bits) ? + vaddr : + -BIT_ULL(vaddr_bits - 1); } +/* Clamp to a canonical address less-than-or-equal-to the address given */ +static u64 clamp_to_le_canonical_addr(u64 vaddr, u8 vaddr_bits) +{ + return __is_canonical_address(vaddr, vaddr_bits) ? + vaddr : + BIT_ULL(vaddr_bits - 1) - 1; +} +#else +#define clamp_to_ge_canonical_addr(x, y) (x) +#define clamp_to_le_canonical_addr(x, y) (x) +#endif + static int pt_event_addr_filters_validate(struct list_head *filters) { struct perf_addr_filter *filter; @@ -1204,14 +1442,6 @@ static int pt_event_addr_filters_validate(struct list_head *filters) filter->action == PERF_ADDR_FILTER_ACTION_START) return -EOPNOTSUPP; - if (!filter->path.dentry) { - if (!valid_kernel_ip(filter->offset)) - return -EINVAL; - - if (!valid_kernel_ip(filter->offset + filter->size)) - return -EINVAL; - } - if (++range > intel_pt_validate_hw_cap(PT_CAP_num_address_ranges)) return -EOPNOTSUPP; } @@ -1222,7 +1452,8 @@ static int pt_event_addr_filters_validate(struct list_head *filters) static void pt_event_addr_filters_sync(struct perf_event *event) { struct perf_addr_filters_head *head = perf_event_addr_filters(event); - unsigned long msr_a, msr_b, *offs = event->addr_filters_offs; + unsigned long msr_a, msr_b; + struct perf_addr_filter_range *fr = event->addr_filter_ranges; struct pt_filters *filters = event->hw.addr_filters; struct perf_addr_filter *filter; int range = 0; @@ -1231,12 +1462,29 @@ static void pt_event_addr_filters_sync(struct perf_event *event) return; list_for_each_entry(filter, &head->list, entry) { - if (filter->path.dentry && !offs[range]) { + if (filter->path.dentry && !fr[range].start) { msr_a = msr_b = 0; } else { - /* apply the offset */ - msr_a = filter->offset + offs[range]; - msr_b = filter->size + msr_a - 1; + unsigned long n = fr[range].size - 1; + unsigned long a = fr[range].start; + unsigned long b; + + if (a > ULONG_MAX - n) + b = ULONG_MAX; + else + b = a + n; + /* + * Apply the offset. 64-bit addresses written to the + * MSRs must be canonical, but the range can encompass + * non-canonical addresses. Since software cannot + * execute at non-canonical addresses, adjusting to + * canonical addresses does not affect the result of the + * address filter. + */ + msr_a = clamp_to_ge_canonical_addr(a, boot_cpu_data.x86_virt_bits); + msr_b = clamp_to_le_canonical_addr(b, boot_cpu_data.x86_virt_bits); + if (msr_b < msr_a) + msr_a = msr_b = 0; } filters->filter[range].msr_a = msr_a; @@ -1291,6 +1539,7 @@ void intel_pt_interrupt(void) buf = perf_aux_output_begin(&pt->handle, event); if (!buf) { event->hw.state = PERF_HES_STOPPED; + WRITE_ONCE(pt->resume_allowed, 0); return; } @@ -1299,12 +1548,12 @@ void intel_pt_interrupt(void) ret = pt_buffer_reset_markers(buf, &pt->handle); if (ret) { perf_aux_output_end(&pt->handle, 0); + WRITE_ONCE(pt->resume_allowed, 0); return; } - pt_config_buffer(buf->cur->table, buf->cur_idx, - buf->output_off); - pt_config(event); + pt_config_buffer(buf); + pt_config_start(event); } } @@ -1338,11 +1587,11 @@ void intel_pt_handle_vmx(int on) /* Turn PTs back on */ if (!on && event) - wrmsrl(MSR_IA32_RTIT_CTL, event->hw.config); + wrmsrq(MSR_IA32_RTIT_CTL, event->hw.aux_config); local_irq_restore(flags); } -EXPORT_SYMBOL_GPL(intel_pt_handle_vmx); +EXPORT_SYMBOL_FOR_KVM(intel_pt_handle_vmx); /* * PMU callbacks @@ -1354,6 +1603,26 @@ static void pt_event_start(struct perf_event *event, int mode) struct pt *pt = this_cpu_ptr(&pt_ctx); struct pt_buffer *buf; + if (mode & PERF_EF_RESUME) { + if (READ_ONCE(pt->resume_allowed)) { + u64 status; + + /* + * Only if the trace is not active and the error and + * stopped bits are clear, is it safe to start, but a + * PMI might have just cleared these, so resume_allowed + * must be checked again also. + */ + rdmsrq(MSR_IA32_RTIT_STATUS, status); + if (!(status & (RTIT_STATUS_TRIGGEREN | + RTIT_STATUS_ERROR | + RTIT_STATUS_STOPPED)) && + READ_ONCE(pt->resume_allowed)) + pt_config_start(event); + } + return; + } + buf = perf_aux_output_begin(&pt->handle, event); if (!buf) goto fail_stop; @@ -1364,11 +1633,9 @@ static void pt_event_start(struct perf_event *event, int mode) goto fail_end_stop; } - WRITE_ONCE(pt->handle_nmi, 1); hwc->state = 0; - pt_config_buffer(buf->cur->table, buf->cur_idx, - buf->output_off); + pt_config_buffer(buf); pt_config(event); return; @@ -1383,11 +1650,27 @@ static void pt_event_stop(struct perf_event *event, int mode) { struct pt *pt = this_cpu_ptr(&pt_ctx); + if (mode & PERF_EF_PAUSE) { + if (READ_ONCE(pt->pause_allowed)) + pt_config_stop(event); + return; + } + /* * Protect against the PMI racing with disabling wrmsr, * see comment in intel_pt_interrupt(). */ WRITE_ONCE(pt->handle_nmi, 0); + barrier(); + + /* + * Prevent a resume from attempting to restart tracing, or a pause + * during a subsequent start. Do this after clearing handle_nmi so that + * pt_event_snapshot_aux() will not re-allow them. + */ + WRITE_ONCE(pt->pause_allowed, 0); + WRITE_ONCE(pt->resume_allowed, 0); + barrier(); pt_config_stop(event); @@ -1419,6 +1702,59 @@ static void pt_event_stop(struct perf_event *event, int mode) } } +static long pt_event_snapshot_aux(struct perf_event *event, + struct perf_output_handle *handle, + unsigned long size) +{ + struct pt *pt = this_cpu_ptr(&pt_ctx); + struct pt_buffer *buf = perf_get_aux(&pt->handle); + unsigned long from = 0, to; + long ret; + + if (WARN_ON_ONCE(!buf)) + return 0; + + /* + * Sampling is only allowed on snapshot events; + * see pt_buffer_setup_aux(). + */ + if (WARN_ON_ONCE(!buf->snapshot)) + return 0; + + /* Prevent pause/resume from attempting to start/stop tracing */ + WRITE_ONCE(pt->pause_allowed, 0); + WRITE_ONCE(pt->resume_allowed, 0); + barrier(); + /* + * There is no PT interrupt in this mode, so stop the trace and it will + * remain stopped while the buffer is copied. + */ + pt_config_stop(event); + pt_read_offset(buf); + pt_update_head(pt); + + to = local_read(&buf->data_size); + if (to < size) + from = buf->nr_pages << PAGE_SHIFT; + from += to - size; + + ret = perf_output_copy_aux(&pt->handle, handle, from, to); + + /* + * Here, handle_nmi tells us if the tracing was on. + * If the tracing was on, restart it. + */ + if (READ_ONCE(pt->handle_nmi)) { + WRITE_ONCE(pt->resume_allowed, 1); + barrier(); + pt_config_start(event); + barrier(); + WRITE_ONCE(pt->pause_allowed, 1); + } + + return ret; +} + static void pt_event_del(struct perf_event *event, int mode) { pt_event_stop(event, PERF_EF_UPDATE); @@ -1487,6 +1823,11 @@ void cpu_emergency_stop_pt(void) pt_event_stop(pt->handle.event, PERF_EF_UPDATE); } +int is_intel_pt_event(struct perf_event *event) +{ + return event->pmu == &pt_pmu.pmu; +} + static __init int pt_init(void) { int ret, cpu, prior_warn = 0; @@ -1496,15 +1837,15 @@ static __init int pt_init(void) if (!boot_cpu_has(X86_FEATURE_INTEL_PT)) return -ENODEV; - get_online_cpus(); + cpus_read_lock(); for_each_online_cpu(cpu) { u64 ctl; - ret = rdmsrl_safe_on_cpu(cpu, MSR_IA32_RTIT_CTL, &ctl); + ret = rdmsrq_safe_on_cpu(cpu, MSR_IA32_RTIT_CTL, &ctl); if (!ret && (ctl & RTIT_CTL_TRACEEN)) prior_warn++; } - put_online_cpus(); + cpus_read_unlock(); if (prior_warn) { x86_add_exclusive(x86_lbr_exclusive_pt); @@ -1523,10 +1864,13 @@ static __init int pt_init(void) } if (!intel_pt_validate_hw_cap(PT_CAP_topa_multiple_entries)) - pt_pmu.pmu.capabilities = - PERF_PMU_CAP_AUX_NO_SG | PERF_PMU_CAP_AUX_SW_DOUBLEBUF; + pt_pmu.pmu.capabilities = PERF_PMU_CAP_AUX_NO_SG; + else + pt_pmu.pmu.capabilities = PERF_PMU_CAP_AUX_PREFER_LARGE; - pt_pmu.pmu.capabilities |= PERF_PMU_CAP_EXCLUSIVE | PERF_PMU_CAP_ITRACE; + pt_pmu.pmu.capabilities |= PERF_PMU_CAP_EXCLUSIVE | + PERF_PMU_CAP_ITRACE | + PERF_PMU_CAP_AUX_PAUSE; pt_pmu.pmu.attr_groups = pt_attr_groups; pt_pmu.pmu.task_ctx_nr = perf_sw_context; pt_pmu.pmu.event_init = pt_event_init; @@ -1534,6 +1878,7 @@ static __init int pt_init(void) pt_pmu.pmu.del = pt_event_del; pt_pmu.pmu.start = pt_event_start; pt_pmu.pmu.stop = pt_event_stop; + pt_pmu.pmu.snapshot_aux = pt_event_snapshot_aux; pt_pmu.pmu.read = pt_event_read; pt_pmu.pmu.setup_aux = pt_buffer_setup_aux; pt_pmu.pmu.free_aux = pt_buffer_free_aux; |