diff options
Diffstat (limited to 'tools/perf/util/evsel.c')
| -rw-r--r-- | tools/perf/util/evsel.c | 1922 |
1 files changed, 1575 insertions, 347 deletions
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c index 22d3267ce294..9cd706f62793 100644 --- a/tools/perf/util/evsel.c +++ b/tools/perf/util/evsel.c @@ -5,6 +5,11 @@ * Parts came from builtin-{top,stat,record}.c, see those files for further * copyright notes. */ +/* + * Powerpc needs __SANE_USERSPACE_TYPES__ before <linux/types.h> to select + * 'int-ll64.h' and avoid compile warnings when printing __u64 with %llu. + */ +#define __SANE_USERSPACE_TYPES__ #include <byteswap.h> #include <errno.h> @@ -12,7 +17,6 @@ #include <linux/bitops.h> #include <api/fs/fs.h> #include <api/fs/tracing_path.h> -#include <traceevent/event-parse.h> #include <linux/hw_breakpoint.h> #include <linux/perf_event.h> #include <linux/compiler.h> @@ -20,6 +24,7 @@ #include <linux/zalloc.h> #include <sys/ioctl.h> #include <sys/resource.h> +#include <sys/syscall.h> #include <sys/types.h> #include <dirent.h> #include <stdlib.h> @@ -31,6 +36,7 @@ #include "counts.h" #include "event.h" #include "evsel.h" +#include "time-utils.h" #include "util/env.h" #include "util/evsel_config.h" #include "util/evsel_fprintf.h" @@ -42,19 +48,35 @@ #include "record.h" #include "debug.h" #include "trace-event.h" +#include "session.h" #include "stat.h" #include "string2.h" #include "memswap.h" #include "util.h" -#include "hashmap.h" -#include "pmu-hybrid.h" +#include "util/hashmap.h" +#include "off_cpu.h" +#include "pmu.h" +#include "pmus.h" +#include "drm_pmu.h" +#include "hwmon_pmu.h" +#include "tool_pmu.h" +#include "tp_pmu.h" +#include "rlimit.h" #include "../perf-sys.h" #include "util/parse-branch-options.h" +#include "util/bpf-filter.h" +#include "util/hist.h" #include <internal/xyarray.h> #include <internal/lib.h> +#include <internal/threadmap.h> +#include "util/intel-tpebs.h" #include <linux/ctype.h> +#ifdef HAVE_LIBTRACEEVENT +#include <event-parse.h> +#endif + struct perf_missing_features perf_missing_features; static clockid_t clockid; @@ -64,7 +86,123 @@ static int evsel__no_extra_init(struct evsel *evsel __maybe_unused) return 0; } -void __weak test_attr__ready(void) { } +static bool test_attr__enabled(void) +{ + static bool test_attr__enabled; + static bool test_attr__enabled_tested; + + if (!test_attr__enabled_tested) { + char *dir = getenv("PERF_TEST_ATTR"); + + test_attr__enabled = (dir != NULL); + test_attr__enabled_tested = true; + } + return test_attr__enabled; +} + +#define __WRITE_ASS(str, fmt, data) \ +do { \ + if (fprintf(file, #str "=%"fmt "\n", data) < 0) { \ + perror("test attr - failed to write event file"); \ + fclose(file); \ + return -1; \ + } \ +} while (0) + +#define WRITE_ASS(field, fmt) __WRITE_ASS(field, fmt, attr->field) + +static int store_event(struct perf_event_attr *attr, pid_t pid, struct perf_cpu cpu, + int fd, int group_fd, unsigned long flags) +{ + FILE *file; + char path[PATH_MAX]; + char *dir = getenv("PERF_TEST_ATTR"); + + snprintf(path, PATH_MAX, "%s/event-%d-%llu-%d", dir, + attr->type, attr->config, fd); + + file = fopen(path, "w+"); + if (!file) { + perror("test attr - failed to open event file"); + return -1; + } + + if (fprintf(file, "[event-%d-%llu-%d]\n", + attr->type, attr->config, fd) < 0) { + perror("test attr - failed to write event file"); + fclose(file); + return -1; + } + + /* syscall arguments */ + __WRITE_ASS(fd, "d", fd); + __WRITE_ASS(group_fd, "d", group_fd); + __WRITE_ASS(cpu, "d", cpu.cpu); + __WRITE_ASS(pid, "d", pid); + __WRITE_ASS(flags, "lu", flags); + + /* struct perf_event_attr */ + WRITE_ASS(type, PRIu32); + WRITE_ASS(size, PRIu32); + WRITE_ASS(config, "llu"); + WRITE_ASS(sample_period, "llu"); + WRITE_ASS(sample_type, "llu"); + WRITE_ASS(read_format, "llu"); + WRITE_ASS(disabled, "d"); + WRITE_ASS(inherit, "d"); + WRITE_ASS(pinned, "d"); + WRITE_ASS(exclusive, "d"); + WRITE_ASS(exclude_user, "d"); + WRITE_ASS(exclude_kernel, "d"); + WRITE_ASS(exclude_hv, "d"); + WRITE_ASS(exclude_idle, "d"); + WRITE_ASS(mmap, "d"); + WRITE_ASS(comm, "d"); + WRITE_ASS(freq, "d"); + WRITE_ASS(inherit_stat, "d"); + WRITE_ASS(enable_on_exec, "d"); + WRITE_ASS(task, "d"); + WRITE_ASS(watermark, "d"); + WRITE_ASS(precise_ip, "d"); + WRITE_ASS(mmap_data, "d"); + WRITE_ASS(sample_id_all, "d"); + WRITE_ASS(exclude_host, "d"); + WRITE_ASS(exclude_guest, "d"); + WRITE_ASS(exclude_callchain_kernel, "d"); + WRITE_ASS(exclude_callchain_user, "d"); + WRITE_ASS(mmap2, "d"); + WRITE_ASS(comm_exec, "d"); + WRITE_ASS(context_switch, "d"); + WRITE_ASS(write_backward, "d"); + WRITE_ASS(namespaces, "d"); + WRITE_ASS(use_clockid, "d"); + WRITE_ASS(wakeup_events, PRIu32); + WRITE_ASS(bp_type, PRIu32); + WRITE_ASS(config1, "llu"); + WRITE_ASS(config2, "llu"); + WRITE_ASS(branch_sample_type, "llu"); + WRITE_ASS(sample_regs_user, "llu"); + WRITE_ASS(sample_stack_user, PRIu32); + + fclose(file); + return 0; +} + +#undef __WRITE_ASS +#undef WRITE_ASS + +static void test_attr__open(struct perf_event_attr *attr, pid_t pid, struct perf_cpu cpu, + int fd, int group_fd, unsigned long flags) +{ + int errno_saved = errno; + + if ((fd != -1) && store_event(attr, pid, cpu, fd, group_fd, flags)) { + pr_err("test attr FAILED"); + exit(128); + } + + errno = errno_saved; +} static void evsel__no_extra_fini(struct evsel *evsel __maybe_unused) { @@ -102,6 +240,16 @@ set_methods: return 0; } +const char *evsel__pmu_name(const struct evsel *evsel) +{ + struct perf_pmu *pmu = evsel__find_pmu(evsel); + + if (pmu) + return pmu->name; + + return event_type(evsel->core.attr.type); +} + #define FD(e, x, y) (*(int *)xyarray__entry(e->core.fd, x, y)) int __evsel__sample_size(u64 sample_type) @@ -249,16 +397,18 @@ void evsel__init(struct evsel *evsel, evsel->bpf_fd = -1; INIT_LIST_HEAD(&evsel->config_terms); INIT_LIST_HEAD(&evsel->bpf_counter_list); + INIT_LIST_HEAD(&evsel->bpf_filters); perf_evsel__object.init(evsel); evsel->sample_size = __evsel__sample_size(attr->sample_type); evsel__calc_id_pos(evsel); evsel->cmdline_group_boundary = false; - evsel->metric_expr = NULL; - evsel->metric_name = NULL; - evsel->metric_events = NULL; evsel->per_pkg_mask = NULL; evsel->collect_stat = false; - evsel->pmu_name = NULL; + evsel->group_pmu_name = NULL; + evsel->skippable = false; + evsel->supported = true; + evsel->alternate_hw_config = PERF_COUNT_HW_MAX; + evsel->script_output_type = -1; // FIXME: OUTPUT_TYPE_UNSET, see builtin-script.c } struct evsel *evsel__new_idx(struct perf_event_attr *attr, int idx) @@ -269,8 +419,8 @@ struct evsel *evsel__new_idx(struct perf_event_attr *attr, int idx) return NULL; evsel__init(evsel, attr, idx); - if (evsel__is_bpf_output(evsel)) { - evsel->core.attr.sample_type |= (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME | + if (evsel__is_bpf_output(evsel) && !attr->sample_type) { + evsel->core.attr.sample_type = (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME | PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD), evsel->core.attr.sample_period = 1; } @@ -284,48 +434,6 @@ struct evsel *evsel__new_idx(struct perf_event_attr *attr, int idx) return evsel; } -static bool perf_event_can_profile_kernel(void) -{ - return perf_event_paranoid_check(1); -} - -struct evsel *evsel__new_cycles(bool precise __maybe_unused, __u32 type, __u64 config) -{ - struct perf_event_attr attr = { - .type = type, - .config = config, - .exclude_kernel = !perf_event_can_profile_kernel(), - }; - struct evsel *evsel; - - event_attr_init(&attr); - - /* - * Now let the usual logic to set up the perf_event_attr defaults - * to kick in when we return and before perf_evsel__open() is called. - */ - evsel = evsel__new(&attr); - if (evsel == NULL) - goto out; - - arch_evsel__fixup_new_cycles(&evsel->core.attr); - - evsel->precise_max = true; - - /* use asprintf() because free(evsel) assumes name is allocated */ - if (asprintf(&evsel->name, "cycles%s%s%.*s", - (attr.precise_ip || attr.exclude_kernel) ? ":" : "", - attr.exclude_kernel ? "u" : "", - attr.precise_ip ? attr.precise_ip + 1 : 0, "ppp") < 0) - goto error_free; -out: - return evsel; -error_free: - evsel__delete(evsel); - evsel = NULL; - goto out; -} - int copy_config_terms(struct list_head *dst, struct list_head *src) { struct evsel_config_term *pos, *tmp; @@ -360,7 +468,7 @@ static int evsel__copy_config_terms(struct evsel *dst, struct evsel *src) * The assumption is that @orig is not configured nor opened yet. * So we only care about the attributes that can be set while it's parsed. */ -struct evsel *evsel__clone(struct evsel *orig) +struct evsel *evsel__clone(struct evsel *dest, struct evsel *orig) { struct evsel *evsel; @@ -373,15 +481,21 @@ struct evsel *evsel__clone(struct evsel *orig) if (orig->bpf_obj) return NULL; - evsel = evsel__new(&orig->core.attr); + if (dest) + evsel = dest; + else + evsel = evsel__new(&orig->core.attr); + if (evsel == NULL) return NULL; evsel->core.cpus = perf_cpu_map__get(orig->core.cpus); - evsel->core.own_cpus = perf_cpu_map__get(orig->core.own_cpus); + evsel->core.pmu_cpus = perf_cpu_map__get(orig->core.pmu_cpus); evsel->core.threads = perf_thread_map__get(orig->core.threads); evsel->core.nr_members = orig->core.nr_members; evsel->core.system_wide = orig->core.system_wide; + evsel->core.requires_cpu = orig->core.requires_cpu; + evsel->core.is_pmu_core = orig->core.is_pmu_core; if (orig->name) { evsel->name = strdup(orig->name); @@ -393,9 +507,9 @@ struct evsel *evsel__clone(struct evsel *orig) if (evsel->group_name == NULL) goto out_err; } - if (orig->pmu_name) { - evsel->pmu_name = strdup(orig->pmu_name); - if (evsel->pmu_name == NULL) + if (orig->group_pmu_name) { + evsel->group_pmu_name = strdup(orig->group_pmu_name); + if (evsel->group_pmu_name == NULL) goto out_err; } if (orig->filter) { @@ -409,35 +523,50 @@ struct evsel *evsel__clone(struct evsel *orig) goto out_err; } evsel->cgrp = cgroup__get(orig->cgrp); +#ifdef HAVE_LIBTRACEEVENT + if (orig->tp_sys) { + evsel->tp_sys = strdup(orig->tp_sys); + if (evsel->tp_sys == NULL) + goto out_err; + } + if (orig->tp_name) { + evsel->tp_name = strdup(orig->tp_name); + if (evsel->tp_name == NULL) + goto out_err; + } evsel->tp_format = orig->tp_format; +#endif evsel->handler = orig->handler; evsel->core.leader = orig->core.leader; + evsel->metric_leader = orig->metric_leader; evsel->max_events = orig->max_events; - evsel->tool_event = orig->tool_event; - free((char *)evsel->unit); - evsel->unit = strdup(orig->unit); - if (evsel->unit == NULL) - goto out_err; - + zfree(&evsel->unit); + if (orig->unit) { + evsel->unit = strdup(orig->unit); + if (evsel->unit == NULL) + goto out_err; + } evsel->scale = orig->scale; evsel->snapshot = orig->snapshot; evsel->per_pkg = orig->per_pkg; evsel->percore = orig->percore; evsel->precise_max = orig->precise_max; - evsel->use_uncore_alias = orig->use_uncore_alias; evsel->is_libpfm_event = orig->is_libpfm_event; evsel->exclude_GH = orig->exclude_GH; evsel->sample_read = orig->sample_read; - evsel->auto_merge_stats = orig->auto_merge_stats; evsel->collect_stat = orig->collect_stat; evsel->weak_group = orig->weak_group; evsel->use_config_name = orig->use_config_name; + evsel->pmu = orig->pmu; + evsel->first_wildcard_match = orig->first_wildcard_match; if (evsel__copy_config_terms(evsel, orig) < 0) goto out_err; + evsel->alternate_hw_config = orig->alternate_hw_config; + return evsel; out_err: @@ -448,45 +577,87 @@ out_err: /* * Returns pointer with encoded error via <linux/err.h> interface. */ -struct evsel *evsel__newtp_idx(const char *sys, const char *name, int idx) +struct evsel *evsel__newtp_idx(const char *sys, const char *name, int idx, bool format) { + struct perf_event_attr attr = { + .type = PERF_TYPE_TRACEPOINT, + .sample_type = (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME | + PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD), + }; struct evsel *evsel = zalloc(perf_evsel__object.size); - int err = -ENOMEM; + int err = -ENOMEM, id = -1; - if (evsel == NULL) { + if (evsel == NULL) goto out_err; - } else { - struct perf_event_attr attr = { - .type = PERF_TYPE_TRACEPOINT, - .sample_type = (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME | - PERF_SAMPLE_CPU | PERF_SAMPLE_PERIOD), - }; - if (asprintf(&evsel->name, "%s:%s", sys, name) < 0) - goto out_free; - evsel->tp_format = trace_event__tp_format(sys, name); - if (IS_ERR(evsel->tp_format)) { - err = PTR_ERR(evsel->tp_format); + if (asprintf(&evsel->name, "%s:%s", sys, name) < 0) + goto out_free; + +#ifdef HAVE_LIBTRACEEVENT + evsel->tp_sys = strdup(sys); + if (!evsel->tp_sys) + goto out_free; + + evsel->tp_name = strdup(name); + if (!evsel->tp_name) + goto out_free; +#endif + + event_attr_init(&attr); + + if (format) { + id = tp_pmu__id(sys, name); + if (id < 0) { + err = id; goto out_free; } - - event_attr_init(&attr); - attr.config = evsel->tp_format->id; - attr.sample_period = 1; - evsel__init(evsel, &attr, idx); } - + attr.config = (__u64)id; + attr.sample_period = 1; + evsel__init(evsel, &attr, idx); return evsel; out_free: zfree(&evsel->name); +#ifdef HAVE_LIBTRACEEVENT + zfree(&evsel->tp_sys); + zfree(&evsel->tp_name); +#endif free(evsel); out_err: return ERR_PTR(err); } -const char *evsel__hw_names[PERF_COUNT_HW_MAX] = { +#ifdef HAVE_LIBTRACEEVENT +struct tep_event *evsel__tp_format(struct evsel *evsel) +{ + struct tep_event *tp_format = evsel->tp_format; + + if (tp_format) + return tp_format; + + if (evsel->core.attr.type != PERF_TYPE_TRACEPOINT) + return NULL; + + if (!evsel->tp_sys) + tp_format = trace_event__tp_format_id(evsel->core.attr.config); + else + tp_format = trace_event__tp_format(evsel->tp_sys, evsel->tp_name); + + if (IS_ERR(tp_format)) { + int err = -PTR_ERR(evsel->tp_format); + + pr_err("Error getting tracepoint format '%s' '%s'(%d)\n", + evsel__name(evsel), strerror(err), err); + return NULL; + } + evsel->tp_format = tp_format; + return evsel->tp_format; +} +#endif + +const char *const evsel__hw_names[PERF_COUNT_HW_MAX] = { "cycles", "instructions", "cache-references", @@ -533,7 +704,6 @@ static int evsel__add_modifiers(struct evsel *evsel, char *bf, size_t size) { int colon = 0, r = 0; struct perf_event_attr *attr = &evsel->core.attr; - bool exclude_guest_default = false; #define MOD_PRINT(context, mod) do { \ if (!attr->exclude_##context) { \ @@ -545,17 +715,15 @@ static int evsel__add_modifiers(struct evsel *evsel, char *bf, size_t size) MOD_PRINT(kernel, 'k'); MOD_PRINT(user, 'u'); MOD_PRINT(hv, 'h'); - exclude_guest_default = true; } if (attr->precise_ip) { if (!colon) colon = ++r; r += scnprintf(bf + r, size - r, "%.*s", attr->precise_ip, "ppp"); - exclude_guest_default = true; } - if (attr->exclude_host || attr->exclude_guest == exclude_guest_default) { + if (attr->exclude_host || attr->exclude_guest) { MOD_PRINT(host, 'H'); MOD_PRINT(guest, 'G'); } @@ -565,13 +733,18 @@ static int evsel__add_modifiers(struct evsel *evsel, char *bf, size_t size) return r; } +int __weak arch_evsel__hw_name(struct evsel *evsel, char *bf, size_t size) +{ + return scnprintf(bf, size, "%s", __evsel__hw_name(evsel->core.attr.config)); +} + static int evsel__hw_name(struct evsel *evsel, char *bf, size_t size) { - int r = scnprintf(bf, size, "%s", __evsel__hw_name(evsel->core.attr.config)); + int r = arch_evsel__hw_name(evsel, bf, size); return r + evsel__add_modifiers(evsel, bf + r, size - r); } -const char *evsel__sw_names[PERF_COUNT_SW_MAX] = { +const char *const evsel__sw_names[PERF_COUNT_SW_MAX] = { "cpu-clock", "task-clock", "page-faults", @@ -622,7 +795,7 @@ static int evsel__bp_name(struct evsel *evsel, char *bf, size_t size) return r + evsel__add_modifiers(evsel, bf + r, size - r); } -const char *evsel__hw_cache[PERF_COUNT_HW_CACHE_MAX][EVSEL__MAX_ALIASES] = { +const char *const evsel__hw_cache[PERF_COUNT_HW_CACHE_MAX][EVSEL__MAX_ALIASES] = { { "L1-dcache", "l1-d", "l1d", "L1-data", }, { "L1-icache", "l1-i", "l1i", "L1-instruction", }, { "LLC", "L2", }, @@ -632,13 +805,13 @@ const char *evsel__hw_cache[PERF_COUNT_HW_CACHE_MAX][EVSEL__MAX_ALIASES] = { { "node", }, }; -const char *evsel__hw_cache_op[PERF_COUNT_HW_CACHE_OP_MAX][EVSEL__MAX_ALIASES] = { +const char *const evsel__hw_cache_op[PERF_COUNT_HW_CACHE_OP_MAX][EVSEL__MAX_ALIASES] = { { "load", "loads", "read", }, { "store", "stores", "write", }, { "prefetch", "prefetches", "speculative-read", "speculative-load", }, }; -const char *evsel__hw_cache_result[PERF_COUNT_HW_CACHE_RESULT_MAX][EVSEL__MAX_ALIASES] = { +const char *const evsel__hw_cache_result[PERF_COUNT_HW_CACHE_RESULT_MAX][EVSEL__MAX_ALIASES] = { { "refs", "Reference", "ops", "access", }, { "misses", "miss", }, }; @@ -654,7 +827,7 @@ const char *evsel__hw_cache_result[PERF_COUNT_HW_CACHE_RESULT_MAX][EVSEL__MAX_AL * L1I : Read and prefetch only * ITLB and BPU : Read-only */ -static unsigned long evsel__hw_cache_stat[C(MAX)] = { +static const unsigned long evsel__hw_cache_stat[C(MAX)] = { [C(L1D)] = (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH), [C(L1I)] = (CACHE_READ | CACHE_PREFETCH), [C(LL)] = (CACHE_READ | CACHE_WRITE | CACHE_PREFETCH), @@ -723,12 +896,6 @@ static int evsel__raw_name(struct evsel *evsel, char *bf, size_t size) return ret + evsel__add_modifiers(evsel, bf + ret, size - ret); } -static int evsel__tool_name(char *bf, size_t size) -{ - int ret = scnprintf(bf, size, "duration_time"); - return ret; -} - const char *evsel__name(struct evsel *evsel) { char bf[128]; @@ -753,10 +920,7 @@ const char *evsel__name(struct evsel *evsel) break; case PERF_TYPE_SOFTWARE: - if (evsel->tool_event) - evsel__tool_name(bf, sizeof(bf)); - else - evsel__sw_name(evsel, bf, sizeof(bf)); + evsel__sw_name(evsel, bf, sizeof(bf)); break; case PERF_TYPE_TRACEPOINT: @@ -767,6 +931,10 @@ const char *evsel__name(struct evsel *evsel) evsel__bp_name(evsel, bf, sizeof(bf)); break; + case PERF_PMU_TYPE_TOOL: + scnprintf(bf, sizeof(bf), "%s", evsel__tool_pmu_event_name(evsel)); + break; + default: scnprintf(bf, sizeof(bf), "unknown attr type: %d", evsel->core.attr.type); @@ -781,13 +949,18 @@ out_unknown: return "unknown"; } +bool evsel__name_is(struct evsel *evsel, const char *name) +{ + return !strcmp(evsel__name(evsel), name); +} + const char *evsel__metric_id(const struct evsel *evsel) { if (evsel->metric_id) return evsel->metric_id; - if (evsel->core.attr.type == PERF_TYPE_SOFTWARE && evsel->tool_event) - return "duration_time"; + if (evsel__is_tool(evsel)) + return evsel__tool_pmu_event_name(evsel); return "unknown"; } @@ -810,16 +983,22 @@ const char *evsel__group_name(struct evsel *evsel) int evsel__group_desc(struct evsel *evsel, char *buf, size_t size) { int ret = 0; + bool first = true; struct evsel *pos; const char *group_name = evsel__group_name(evsel); if (!evsel->forced_leader) ret = scnprintf(buf, size, "%s { ", group_name); - ret += scnprintf(buf + ret, size - ret, "%s", evsel__name(evsel)); + for_each_group_evsel(pos, evsel) { + if (symbol_conf.skip_empty && + evsel__hists(pos)->stats.nr_samples == 0) + continue; - for_each_group_member(pos, evsel) - ret += scnprintf(buf + ret, size - ret, ", %s", evsel__name(pos)); + ret += scnprintf(buf + ret, size - ret, "%s%s", + first ? "" : ", ", evsel__name(pos)); + first = false; + } if (!evsel->forced_leader) ret += scnprintf(buf + ret, size - ret, " }"); @@ -862,15 +1041,18 @@ static void __evsel__config_callchain(struct evsel *evsel, struct record_opts *o if (param->record_mode == CALLCHAIN_DWARF) { if (!function) { + const char *arch = perf_env__arch(evsel__env(evsel)); + evsel__set_sample_bit(evsel, REGS_USER); evsel__set_sample_bit(evsel, STACK_USER); - if (opts->sample_user_regs && DWARF_MINIMAL_REGS != PERF_REGS_MASK) { - attr->sample_regs_user |= DWARF_MINIMAL_REGS; + if (opts->sample_user_regs && + DWARF_MINIMAL_REGS(arch) != arch__user_reg_mask()) { + attr->sample_regs_user |= DWARF_MINIMAL_REGS(arch); pr_warning("WARNING: The use of --call-graph=dwarf may require all the user registers, " "specifying a subset with --user-regs may render DWARF unwinding unreliable, " "so the minimal registers set (IP, SP) is explicitly forced.\n"); } else { - attr->sample_regs_user |= PERF_REGS_MASK; + attr->sample_regs_user |= arch__user_reg_mask(); } attr->sample_stack_user = param->dump_size; attr->exclude_callchain_user = 1; @@ -884,6 +1066,9 @@ static void __evsel__config_callchain(struct evsel *evsel, struct record_opts *o pr_info("Disabling user space callchains for function trace event.\n"); attr->exclude_callchain_user = 1; } + + if (param->defer && !attr->exclude_callchain_user) + attr->defer_callchain = 1; } void evsel__config_callchain(struct evsel *evsel, struct record_opts *opts, @@ -910,6 +1095,71 @@ static void evsel__reset_callgraph(struct evsel *evsel, struct callchain_param * } } +static void evsel__apply_ratio_to_prev(struct evsel *evsel, + struct perf_event_attr *attr, + struct record_opts *opts, + const char *buf) +{ + struct perf_event_attr *prev_attr = NULL; + struct evsel *evsel_prev = NULL; + u64 type = evsel->core.attr.sample_type; + u64 prev_type = 0; + double rtp; + + rtp = strtod(buf, NULL); + if (rtp <= 0) { + pr_err("Invalid ratio-to-prev value %lf\n", rtp); + return; + } + if (evsel == evsel__leader(evsel)) { + pr_err("Invalid use of ratio-to-prev term without preceding element in group\n"); + return; + } + if (!evsel->pmu->is_core) { + pr_err("Event using ratio-to-prev term must have a core PMU\n"); + return; + } + + evsel_prev = evsel__prev(evsel); + if (!evsel_prev) { + pr_err("Previous event does not exist.\n"); + return; + } + + if (evsel_prev->pmu->type != evsel->pmu->type) { + pr_err("Compared events (\"%s\", \"%s\") must have same PMU\n", + evsel->name, evsel_prev->name); + return; + } + + prev_attr = &evsel_prev->core.attr; + prev_type = evsel_prev->core.attr.sample_type; + + if (!(prev_type & PERF_SAMPLE_PERIOD)) { + attr->sample_period = prev_attr->sample_period * rtp; + attr->freq = 0; + evsel__reset_sample_bit(evsel, PERIOD); + } else if (!(type & PERF_SAMPLE_PERIOD)) { + prev_attr->sample_period = attr->sample_period / rtp; + prev_attr->freq = 0; + evsel__reset_sample_bit(evsel_prev, PERIOD); + } else { + if (opts->user_interval != ULLONG_MAX) { + prev_attr->sample_period = opts->user_interval; + attr->sample_period = prev_attr->sample_period * rtp; + prev_attr->freq = 0; + attr->freq = 0; + evsel__reset_sample_bit(evsel_prev, PERIOD); + evsel__reset_sample_bit(evsel, PERIOD); + } else { + pr_err("Event period term or count (-c) must be set when using ratio-to-prev term.\n"); + return; + } + } + + arch_evsel__apply_ratio_to_prev(evsel, attr); +} + static void evsel__apply_config_terms(struct evsel *evsel, struct record_opts *opts, bool track) { @@ -923,6 +1173,7 @@ static void evsel__apply_config_terms(struct evsel *evsel, u32 dump_size = 0; int max_stack = 0; const char *callgraph_buf = NULL; + const char *rtp_buf = NULL; list_for_each_entry(term, config_terms, list) { switch (term->type) { @@ -985,11 +1236,17 @@ static void evsel__apply_config_terms(struct evsel *evsel, case EVSEL__CONFIG_TERM_AUX_OUTPUT: attr->aux_output = term->val.aux_output ? 1 : 0; break; + case EVSEL__CONFIG_TERM_AUX_ACTION: + /* Already applied by auxtrace */ + break; case EVSEL__CONFIG_TERM_AUX_SAMPLE_SIZE: /* Already applied by auxtrace */ break; case EVSEL__CONFIG_TERM_CFG_CHG: break; + case EVSEL__CONFIG_TERM_RATIO_TO_PREV: + rtp_buf = term->val.str; + break; default: break; } @@ -1041,6 +1298,8 @@ static void evsel__apply_config_terms(struct evsel *evsel, evsel__config_callchain(evsel, opts, ¶m); } } + if (rtp_buf) + evsel__apply_ratio_to_prev(evsel, attr, opts, rtp_buf); } struct evsel_config_term *__evsel__get_config_term(struct evsel *evsel, enum evsel_term_type type) @@ -1060,7 +1319,13 @@ void __weak arch_evsel__set_sample_weight(struct evsel *evsel) evsel__set_sample_bit(evsel, WEIGHT); } -void __weak arch_evsel__fixup_new_cycles(struct perf_event_attr *attr __maybe_unused) +void __weak arch__post_evsel_config(struct evsel *evsel __maybe_unused, + struct perf_event_attr *attr __maybe_unused) +{ +} + +void __weak arch_evsel__apply_ratio_to_prev(struct evsel *evsel __maybe_unused, + struct perf_event_attr *attr __maybe_unused) { } @@ -1075,6 +1340,12 @@ static void evsel__set_default_freq_period(struct record_opts *opts, } } +bool evsel__is_offcpu_event(struct evsel *evsel) +{ + return evsel__is_bpf_output(evsel) && evsel__name_is(evsel, OFFCPU_EVENT) && + evsel->core.attr.sample_type & PERF_SAMPLE_RAW; +} + /* * The enable_on_exec/disabled value strategy: * @@ -1112,8 +1383,9 @@ void evsel__config(struct evsel *evsel, struct record_opts *opts, bool per_cpu = opts->target.default_per_cpu && !opts->target.per_thread; attr->sample_id_all = perf_missing_features.sample_id_all ? 0 : 1; - attr->inherit = !opts->no_inherit; + attr->inherit = target__has_cpu(&opts->target) ? 0 : !opts->no_inherit; attr->write_backward = opts->overwrite ? 1 : 0; + attr->read_format = PERF_FORMAT_LOST; evsel__set_sample_bit(evsel, IP); evsel__set_sample_bit(evsel, TID); @@ -1133,7 +1405,15 @@ void evsel__config(struct evsel *evsel, struct record_opts *opts, */ if (leader->core.nr_members > 1) { attr->read_format |= PERF_FORMAT_GROUP; - attr->inherit = 0; + } + + /* + * Inherit + SAMPLE_READ requires SAMPLE_TID in the read_format + */ + if (attr->inherit) { + evsel__set_sample_bit(evsel, TID); + evsel->core.attr.read_format |= + PERF_FORMAT_ID; } } @@ -1211,7 +1491,7 @@ void evsel__config(struct evsel *evsel, struct record_opts *opts, evsel__set_sample_bit(evsel, CPU); } - if (opts->sample_address) + if (opts->sample_data_src) evsel__set_sample_bit(evsel, DATA_SRC); if (opts->sample_phys_addr) @@ -1226,14 +1506,16 @@ void evsel__config(struct evsel *evsel, struct record_opts *opts, attr->branch_sample_type = opts->branch_stack; } - if (opts->sample_weight) + if (opts->sample_weight || evsel->retire_lat) { arch_evsel__set_sample_weight(evsel); - + evsel->retire_lat = false; + } attr->task = track; attr->mmap = track; attr->mmap2 = track && !perf_missing_features.mmap2; attr->comm = track; attr->build_id = track && opts->build_id; + attr->defer_output = track && callchain && callchain->defer; /* * ksymbol is tracked separately with text poke because it needs to be @@ -1283,7 +1565,7 @@ void evsel__config(struct evsel *evsel, struct record_opts *opts, * group leaders for traced executed by perf. */ if (target__none(&opts->target) && evsel__is_group_leader(evsel) && - !opts->initial_delay) + !opts->target.initial_delay) attr->enable_on_exec = 1; if (evsel->immediate) { @@ -1310,7 +1592,7 @@ void evsel__config(struct evsel *evsel, struct record_opts *opts, attr->exclude_user = 1; } - if (evsel->core.own_cpus || evsel->unit) + if (evsel->core.pmu_cpus || evsel->unit) evsel->core.attr.read_format |= PERF_FORMAT_ID; /* @@ -1339,6 +1621,13 @@ void evsel__config(struct evsel *evsel, struct record_opts *opts, */ if (evsel__is_dummy_event(evsel)) evsel__reset_sample_bit(evsel, BRANCH_STACK); + + if (evsel__is_offcpu_event(evsel)) { + evsel->core.attr.sample_type &= OFFCPU_SAMPLE_TYPES; + attr->inherit = 0; + } + + arch__post_evsel_config(evsel, attr); } int evsel__set_filter(struct evsel *evsel, const char *filter) @@ -1433,33 +1722,55 @@ static void evsel__free_config_terms(struct evsel *evsel) free_config_terms(&evsel->config_terms); } +static void (*evsel__priv_destructor)(void *priv); + +void evsel__set_priv_destructor(void (*destructor)(void *priv)) +{ + assert(evsel__priv_destructor == NULL); + + evsel__priv_destructor = destructor; +} + void evsel__exit(struct evsel *evsel) { assert(list_empty(&evsel->core.node)); assert(evsel->evlist == NULL); + if (evsel__is_retire_lat(evsel)) + evsel__tpebs_close(evsel); bpf_counter__destroy(evsel); + perf_bpf_filter__destroy(evsel); evsel__free_counts(evsel); perf_evsel__free_fd(&evsel->core); perf_evsel__free_id(&evsel->core); evsel__free_config_terms(evsel); cgroup__put(evsel->cgrp); - perf_cpu_map__put(evsel->core.cpus); - perf_cpu_map__put(evsel->core.own_cpus); - perf_thread_map__put(evsel->core.threads); + perf_evsel__exit(&evsel->core); zfree(&evsel->group_name); zfree(&evsel->name); - zfree(&evsel->pmu_name); +#ifdef HAVE_LIBTRACEEVENT + zfree(&evsel->tp_sys); + zfree(&evsel->tp_name); +#endif + zfree(&evsel->filter); + zfree(&evsel->group_pmu_name); zfree(&evsel->unit); zfree(&evsel->metric_id); evsel__zero_per_pkg(evsel); hashmap__free(evsel->per_pkg_mask); evsel->per_pkg_mask = NULL; - zfree(&evsel->metric_events); + if (evsel__priv_destructor) + evsel__priv_destructor(evsel->priv); perf_evsel__object.fini(evsel); + if (evsel__tool_event(evsel) == TOOL_PMU__EVENT_SYSTEM_TIME || + evsel__tool_event(evsel) == TOOL_PMU__EVENT_USER_TIME) + xyarray__delete(evsel->start_times); } void evsel__delete(struct evsel *evsel) { + if (!evsel) + return; + evsel__exit(evsel); free(evsel); } @@ -1472,13 +1783,8 @@ void evsel__compute_deltas(struct evsel *evsel, int cpu_map_idx, int thread, if (!evsel->prev_raw_counts) return; - if (cpu_map_idx == -1) { - tmp = evsel->prev_raw_counts->aggr; - evsel->prev_raw_counts->aggr = *count; - } else { - tmp = *perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread); - *perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread) = *count; - } + tmp = *perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread); + *perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread) = *count; count->val = count->val - tmp.val; count->ena = count->ena - tmp.ena; @@ -1493,28 +1799,89 @@ static int evsel__read_one(struct evsel *evsel, int cpu_map_idx, int thread) } static void evsel__set_count(struct evsel *counter, int cpu_map_idx, int thread, - u64 val, u64 ena, u64 run) + u64 val, u64 ena, u64 run, u64 lost) { struct perf_counts_values *count; count = perf_counts(counter->counts, cpu_map_idx, thread); + if (evsel__is_retire_lat(counter)) { + evsel__tpebs_read(counter, cpu_map_idx, thread); + perf_counts__set_loaded(counter->counts, cpu_map_idx, thread, true); + return; + } + count->val = val; count->ena = ena; count->run = run; + count->lost = lost; perf_counts__set_loaded(counter->counts, cpu_map_idx, thread, true); } +static bool evsel__group_has_tpebs(struct evsel *leader) +{ + struct evsel *evsel; + + for_each_group_evsel(evsel, leader) { + if (evsel__is_retire_lat(evsel)) + return true; + } + return false; +} + +static u64 evsel__group_read_nr_members(struct evsel *leader) +{ + u64 nr = leader->core.nr_members; + struct evsel *evsel; + + for_each_group_evsel(evsel, leader) { + if (evsel__is_retire_lat(evsel)) + nr--; + } + return nr; +} + +static u64 evsel__group_read_size(struct evsel *leader) +{ + u64 read_format = leader->core.attr.read_format; + int entry = sizeof(u64); /* value */ + int size = 0; + int nr = 1; + + if (!evsel__group_has_tpebs(leader)) + return perf_evsel__read_size(&leader->core); + + if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) + size += sizeof(u64); + + if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) + size += sizeof(u64); + + if (read_format & PERF_FORMAT_ID) + entry += sizeof(u64); + + if (read_format & PERF_FORMAT_LOST) + entry += sizeof(u64); + + if (read_format & PERF_FORMAT_GROUP) { + nr = evsel__group_read_nr_members(leader); + size += sizeof(u64); + } + + size += entry * nr; + return size; +} + static int evsel__process_group_data(struct evsel *leader, int cpu_map_idx, int thread, u64 *data) { u64 read_format = leader->core.attr.read_format; struct sample_read_value *v; - u64 nr, ena = 0, run = 0, i; + u64 nr, ena = 0, run = 0, lost = 0; nr = *data++; - if (nr != (u64) leader->core.nr_members) + if (nr != evsel__group_read_nr_members(leader)) return -EINVAL; if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) @@ -1523,18 +1890,18 @@ static int evsel__process_group_data(struct evsel *leader, int cpu_map_idx, int if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) run = *data++; - v = (struct sample_read_value *) data; - - evsel__set_count(leader, cpu_map_idx, thread, v[0].value, ena, run); - - for (i = 1; i < nr; i++) { + v = (void *)data; + sample_read_group__for_each(v, nr, read_format) { struct evsel *counter; - counter = evlist__id2evsel(leader->evlist, v[i].id); + counter = evlist__id2evsel(leader->evlist, v->id); if (!counter) return -EINVAL; - evsel__set_count(counter, cpu_map_idx, thread, v[i].value, ena, run); + if (read_format & PERF_FORMAT_LOST) + lost = v->lost; + + evsel__set_count(counter, cpu_map_idx, thread, v->value, ena, run, lost); } return 0; @@ -1544,7 +1911,7 @@ static int evsel__read_group(struct evsel *leader, int cpu_map_idx, int thread) { struct perf_stat_evsel *ps = leader->stats; u64 read_format = leader->core.attr.read_format; - int size = perf_evsel__read_size(&leader->core); + int size = evsel__group_read_size(leader); u64 *data = ps->group_data; if (!(read_format & PERF_FORMAT_ID)) @@ -1570,11 +1937,42 @@ static int evsel__read_group(struct evsel *leader, int cpu_map_idx, int thread) return evsel__process_group_data(leader, cpu_map_idx, thread, data); } +bool __evsel__match(const struct evsel *evsel, u32 type, u64 config) +{ + + u32 e_type = evsel->core.attr.type; + u64 e_config = evsel->core.attr.config; + + if (e_type == type && e_config == config) + return true; + if (type != PERF_TYPE_HARDWARE && type != PERF_TYPE_HW_CACHE) + return false; + if ((e_type == PERF_TYPE_HARDWARE || e_type == PERF_TYPE_HW_CACHE) && + perf_pmus__supports_extended_type()) + e_config &= PERF_HW_EVENT_MASK; + if (e_type == type && e_config == config) + return true; + if (type == PERF_TYPE_HARDWARE && evsel->pmu && evsel->pmu->is_core && + evsel->alternate_hw_config == config) + return true; + return false; +} + int evsel__read_counter(struct evsel *evsel, int cpu_map_idx, int thread) { - u64 read_format = evsel->core.attr.read_format; + if (evsel__is_tool(evsel)) + return evsel__tool_pmu_read(evsel, cpu_map_idx, thread); - if (read_format & PERF_FORMAT_GROUP) + if (evsel__is_hwmon(evsel)) + return evsel__hwmon_pmu_read(evsel, cpu_map_idx, thread); + + if (evsel__is_drm(evsel)) + return evsel__drm_pmu_read(evsel, cpu_map_idx, thread); + + if (evsel__is_retire_lat(evsel)) + return evsel__tpebs_read(evsel, cpu_map_idx, thread); + + if (evsel->core.attr.read_format & PERF_FORMAT_GROUP) return evsel__read_group(evsel, cpu_map_idx, thread); return evsel__read_one(evsel, cpu_map_idx, thread); @@ -1626,7 +2024,7 @@ static int get_group_fd(struct evsel *evsel, int cpu_map_idx, int thread) struct evsel *leader = evsel__leader(evsel); int fd; - if (evsel__is_group_leader(evsel)) + if (!evsel->supported || evsel__is_group_leader(evsel)) return -1; /* @@ -1640,9 +2038,13 @@ static int get_group_fd(struct evsel *evsel, int cpu_map_idx, int thread) return -1; fd = FD(leader, cpu_map_idx, thread); - BUG_ON(fd == -1); + BUG_ON(fd == -1 && leader->supported); - return fd; + /* + * When the leader has been skipped, return -2 to distinguish from no + * group leader case. + */ + return fd == -1 ? -2 : fd; } static void evsel__remove_fd(struct evsel *pos, int nr_cpus, int nr_threads, int thread_idx) @@ -1759,7 +2161,8 @@ static struct perf_thread_map *empty_thread_map; static int __evsel__prepare_open(struct evsel *evsel, struct perf_cpu_map *cpus, struct perf_thread_map *threads) { - int nthreads; + int ret = 0; + int nthreads = perf_thread_map__nr(threads); if ((perf_missing_features.write_backward && evsel->core.attr.write_backward) || (perf_missing_features.aux_output && evsel->core.attr.aux_output)) @@ -1767,7 +2170,7 @@ static int __evsel__prepare_open(struct evsel *evsel, struct perf_cpu_map *cpus, if (cpus == NULL) { if (empty_cpu_map == NULL) { - empty_cpu_map = perf_cpu_map__dummy_new(); + empty_cpu_map = perf_cpu_map__new_any_cpu(); if (empty_cpu_map == NULL) return -ENOMEM; } @@ -1785,24 +2188,33 @@ static int __evsel__prepare_open(struct evsel *evsel, struct perf_cpu_map *cpus, threads = empty_thread_map; } - if (evsel->core.system_wide) - nthreads = 1; - else - nthreads = threads->nr; - if (evsel->core.fd == NULL && perf_evsel__alloc_fd(&evsel->core, perf_cpu_map__nr(cpus), nthreads) < 0) return -ENOMEM; + if (evsel__is_tool(evsel)) + ret = evsel__tool_pmu_prepare_open(evsel, cpus, nthreads); + evsel->open_flags = PERF_FLAG_FD_CLOEXEC; if (evsel->cgrp) evsel->open_flags |= PERF_FLAG_PID_CGROUP; - return 0; + return ret; } static void evsel__disable_missing_features(struct evsel *evsel) { + if (perf_missing_features.defer_callchain && evsel->core.attr.defer_callchain) + evsel->core.attr.defer_callchain = 0; + if (perf_missing_features.defer_callchain && evsel->core.attr.defer_output) + evsel->core.attr.defer_output = 0; + if (perf_missing_features.inherit_sample_read && evsel->core.attr.inherit && + (evsel->core.attr.sample_type & PERF_SAMPLE_READ)) + evsel->core.attr.inherit = 0; + if (perf_missing_features.branch_counters) + evsel->core.attr.branch_sample_type &= ~PERF_SAMPLE_BRANCH_COUNTERS; + if (perf_missing_features.read_lost) + evsel->core.attr.read_format &= ~PERF_FORMAT_LOST; if (perf_missing_features.weight_struct) { evsel__set_sample_bit(evsel, WEIGHT); evsel__reset_sample_bit(evsel, WEIGHT_STRUCT); @@ -1848,136 +2260,404 @@ int evsel__prepare_open(struct evsel *evsel, struct perf_cpu_map *cpus, return err; } -bool evsel__detect_missing_features(struct evsel *evsel) +static bool __has_attr_feature(struct perf_event_attr *attr, + struct perf_cpu cpu, unsigned long flags) +{ + int fd = syscall(SYS_perf_event_open, attr, /*pid=*/0, cpu.cpu, + /*group_fd=*/-1, flags); + close(fd); + + if (fd < 0) { + attr->exclude_kernel = 1; + + fd = syscall(SYS_perf_event_open, attr, /*pid=*/0, cpu.cpu, + /*group_fd=*/-1, flags); + close(fd); + } + + if (fd < 0) { + attr->exclude_hv = 1; + + fd = syscall(SYS_perf_event_open, attr, /*pid=*/0, cpu.cpu, + /*group_fd=*/-1, flags); + close(fd); + } + + if (fd < 0) { + attr->exclude_guest = 1; + + fd = syscall(SYS_perf_event_open, attr, /*pid=*/0, cpu.cpu, + /*group_fd=*/-1, flags); + close(fd); + } + + attr->exclude_kernel = 0; + attr->exclude_guest = 0; + attr->exclude_hv = 0; + + return fd >= 0; +} + +static bool has_attr_feature(struct perf_event_attr *attr, unsigned long flags) +{ + struct perf_cpu cpu = {.cpu = -1}; + + return __has_attr_feature(attr, cpu, flags); +} + +static void evsel__detect_missing_pmu_features(struct evsel *evsel) +{ + struct perf_event_attr attr = { + .type = evsel->core.attr.type, + .config = evsel->core.attr.config, + .disabled = 1, + }; + struct perf_pmu *pmu = evsel->pmu; + int old_errno; + + old_errno = errno; + + if (pmu == NULL) + pmu = evsel->pmu = evsel__find_pmu(evsel); + + if (pmu == NULL || pmu->missing_features.checked) + goto out; + + /* + * Must probe features in the order they were added to the + * perf_event_attr interface. These are kernel core limitation but + * specific to PMUs with branch stack. So we can detect with the given + * hardware event and stop on the first one succeeded. + */ + + /* Please add new feature detection here. */ + + attr.exclude_guest = 1; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + pmu->missing_features.exclude_guest = true; + pr_debug2("switching off exclude_guest for PMU %s\n", pmu->name); + +found: + pmu->missing_features.checked = true; +out: + errno = old_errno; +} + +static void evsel__detect_missing_brstack_features(struct evsel *evsel) { + static bool detection_done = false; + struct perf_event_attr attr = { + .type = evsel->core.attr.type, + .config = evsel->core.attr.config, + .disabled = 1, + .sample_type = PERF_SAMPLE_BRANCH_STACK, + .sample_period = 1000, + }; + int old_errno; + + if (detection_done) + return; + + old_errno = errno; + /* * Must probe features in the order they were added to the - * perf_event_attr interface. + * perf_event_attr interface. These are PMU specific limitation + * so we can detect with the given hardware event and stop on the + * first one succeeded. */ - if (!perf_missing_features.weight_struct && - (evsel->core.attr.sample_type & PERF_SAMPLE_WEIGHT_STRUCT)) { - perf_missing_features.weight_struct = true; - pr_debug2("switching off weight struct support\n"); + + /* Please add new feature detection here. */ + + attr.branch_sample_type = PERF_SAMPLE_BRANCH_COUNTERS; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.branch_counters = true; + pr_debug2("switching off branch counters support\n"); + + attr.branch_sample_type = PERF_SAMPLE_BRANCH_HW_INDEX; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.branch_hw_idx = true; + pr_debug2("switching off branch HW index support\n"); + + attr.branch_sample_type = PERF_SAMPLE_BRANCH_NO_CYCLES | PERF_SAMPLE_BRANCH_NO_FLAGS; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.lbr_flags = true; + pr_debug2_peo("switching off branch sample type no (cycles/flags)\n"); + +found: + detection_done = true; + errno = old_errno; +} + +static bool evsel__probe_aux_action(struct evsel *evsel, struct perf_cpu cpu) +{ + struct perf_event_attr attr = evsel->core.attr; + int old_errno = errno; + + attr.disabled = 1; + attr.aux_start_paused = 1; + + if (__has_attr_feature(&attr, cpu, /*flags=*/0)) { + errno = old_errno; return true; - } else if (!perf_missing_features.code_page_size && - (evsel->core.attr.sample_type & PERF_SAMPLE_CODE_PAGE_SIZE)) { - perf_missing_features.code_page_size = true; - pr_debug2_peo("Kernel has no PERF_SAMPLE_CODE_PAGE_SIZE support, bailing out\n"); - return false; - } else if (!perf_missing_features.data_page_size && - (evsel->core.attr.sample_type & PERF_SAMPLE_DATA_PAGE_SIZE)) { - perf_missing_features.data_page_size = true; - pr_debug2_peo("Kernel has no PERF_SAMPLE_DATA_PAGE_SIZE support, bailing out\n"); - return false; - } else if (!perf_missing_features.cgroup && evsel->core.attr.cgroup) { - perf_missing_features.cgroup = true; - pr_debug2_peo("Kernel has no cgroup sampling support, bailing out\n"); - return false; - } else if (!perf_missing_features.branch_hw_idx && - (evsel->core.attr.branch_sample_type & PERF_SAMPLE_BRANCH_HW_INDEX)) { - perf_missing_features.branch_hw_idx = true; - pr_debug2("switching off branch HW index support\n"); + } + + /* + * EOPNOTSUPP means the kernel supports the feature but the PMU does + * not, so keep that distinction if possible. + */ + if (errno != EOPNOTSUPP) + errno = old_errno; + + return false; +} + +static void evsel__detect_missing_aux_action_feature(struct evsel *evsel, struct perf_cpu cpu) +{ + static bool detection_done; + struct evsel *leader; + + /* + * Don't bother probing aux_action if it is not being used or has been + * probed before. + */ + if (!evsel->core.attr.aux_action || detection_done) + return; + + detection_done = true; + + /* + * The leader is an AUX area event. If it has failed, assume the feature + * is not supported. + */ + leader = evsel__leader(evsel); + if (evsel == leader) { + perf_missing_features.aux_action = true; + return; + } + + /* + * AUX area event with aux_action must have been opened successfully + * already, so feature is supported. + */ + if (leader->core.attr.aux_action) + return; + + if (!evsel__probe_aux_action(leader, cpu)) + perf_missing_features.aux_action = true; +} + +static bool evsel__detect_missing_features(struct evsel *evsel, struct perf_cpu cpu) +{ + static bool detection_done = false; + struct perf_event_attr attr = { + .type = PERF_TYPE_SOFTWARE, + .config = PERF_COUNT_SW_TASK_CLOCK, + .disabled = 1, + }; + int old_errno; + + evsel__detect_missing_aux_action_feature(evsel, cpu); + + evsel__detect_missing_pmu_features(evsel); + + if (evsel__has_br_stack(evsel)) + evsel__detect_missing_brstack_features(evsel); + + if (detection_done) + goto check; + + old_errno = errno; + + /* + * Must probe features in the order they were added to the + * perf_event_attr interface. These are kernel core limitation + * not PMU-specific so we can detect with a software event and + * stop on the first one succeeded. + */ + + /* Please add new feature detection here. */ + + attr.defer_callchain = true; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.defer_callchain = true; + pr_debug2("switching off deferred callchain support\n"); + attr.defer_callchain = false; + + attr.inherit = true; + attr.sample_type = PERF_SAMPLE_READ | PERF_SAMPLE_TID; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.inherit_sample_read = true; + pr_debug2("Using PERF_SAMPLE_READ / :S modifier is not compatible with inherit, falling back to no-inherit.\n"); + attr.inherit = false; + attr.sample_type = 0; + + attr.read_format = PERF_FORMAT_LOST; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.read_lost = true; + pr_debug2("switching off PERF_FORMAT_LOST support\n"); + attr.read_format = 0; + + attr.sample_type = PERF_SAMPLE_WEIGHT_STRUCT; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.weight_struct = true; + pr_debug2("switching off weight struct support\n"); + attr.sample_type = 0; + + attr.sample_type = PERF_SAMPLE_CODE_PAGE_SIZE; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.code_page_size = true; + pr_debug2_peo("Kernel has no PERF_SAMPLE_CODE_PAGE_SIZE support\n"); + attr.sample_type = 0; + + attr.sample_type = PERF_SAMPLE_DATA_PAGE_SIZE; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.data_page_size = true; + pr_debug2_peo("Kernel has no PERF_SAMPLE_DATA_PAGE_SIZE support\n"); + attr.sample_type = 0; + + attr.cgroup = 1; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.cgroup = true; + pr_debug2_peo("Kernel has no cgroup sampling support\n"); + attr.cgroup = 0; + + attr.aux_output = 1; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.aux_output = true; + pr_debug2_peo("Kernel has no attr.aux_output support\n"); + attr.aux_output = 0; + + attr.bpf_event = 1; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.bpf = true; + pr_debug2_peo("switching off bpf_event\n"); + attr.bpf_event = 0; + + attr.ksymbol = 1; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.ksymbol = true; + pr_debug2_peo("switching off ksymbol\n"); + attr.ksymbol = 0; + + attr.write_backward = 1; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.write_backward = true; + pr_debug2_peo("switching off write_backward\n"); + attr.write_backward = 0; + + attr.use_clockid = 1; + attr.clockid = CLOCK_MONOTONIC; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.clockid = true; + pr_debug2_peo("switching off clockid\n"); + attr.use_clockid = 0; + attr.clockid = 0; + + if (has_attr_feature(&attr, /*flags=*/PERF_FLAG_FD_CLOEXEC)) + goto found; + perf_missing_features.cloexec = true; + pr_debug2_peo("switching off cloexec flag\n"); + + attr.mmap2 = 1; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.mmap2 = true; + pr_debug2_peo("switching off mmap2\n"); + attr.mmap2 = 0; + + /* set this unconditionally? */ + perf_missing_features.sample_id_all = true; + pr_debug2_peo("switching off sample_id_all\n"); + + attr.inherit = 1; + attr.read_format = PERF_FORMAT_GROUP; + if (has_attr_feature(&attr, /*flags=*/0)) + goto found; + perf_missing_features.group_read = true; + pr_debug2_peo("switching off group read\n"); + attr.inherit = 0; + attr.read_format = 0; + +found: + detection_done = true; + errno = old_errno; + +check: + if ((evsel->core.attr.defer_callchain || evsel->core.attr.defer_output) && + perf_missing_features.defer_callchain) return true; - } else if (!perf_missing_features.aux_output && evsel->core.attr.aux_output) { - perf_missing_features.aux_output = true; - pr_debug2_peo("Kernel has no attr.aux_output support, bailing out\n"); - return false; - } else if (!perf_missing_features.bpf && evsel->core.attr.bpf_event) { - perf_missing_features.bpf = true; - pr_debug2_peo("switching off bpf_event\n"); + + if (evsel->core.attr.inherit && + (evsel->core.attr.sample_type & PERF_SAMPLE_READ) && + perf_missing_features.inherit_sample_read) return true; - } else if (!perf_missing_features.ksymbol && evsel->core.attr.ksymbol) { - perf_missing_features.ksymbol = true; - pr_debug2_peo("switching off ksymbol\n"); + + if ((evsel->core.attr.branch_sample_type & PERF_SAMPLE_BRANCH_COUNTERS) && + perf_missing_features.branch_counters) return true; - } else if (!perf_missing_features.write_backward && evsel->core.attr.write_backward) { - perf_missing_features.write_backward = true; - pr_debug2_peo("switching off write_backward\n"); - return false; - } else if (!perf_missing_features.clockid_wrong && evsel->core.attr.use_clockid) { - perf_missing_features.clockid_wrong = true; - pr_debug2_peo("switching off clockid\n"); + + if ((evsel->core.attr.read_format & PERF_FORMAT_LOST) && + perf_missing_features.read_lost) return true; - } else if (!perf_missing_features.clockid && evsel->core.attr.use_clockid) { - perf_missing_features.clockid = true; - pr_debug2_peo("switching off use_clockid\n"); + + if ((evsel->core.attr.sample_type & PERF_SAMPLE_WEIGHT_STRUCT) && + perf_missing_features.weight_struct) return true; - } else if (!perf_missing_features.cloexec && (evsel->open_flags & PERF_FLAG_FD_CLOEXEC)) { - perf_missing_features.cloexec = true; - pr_debug2_peo("switching off cloexec flag\n"); + + if (evsel->core.attr.use_clockid && evsel->core.attr.clockid != CLOCK_MONOTONIC && + !perf_missing_features.clockid) { + perf_missing_features.clockid_wrong = true; return true; - } else if (!perf_missing_features.mmap2 && evsel->core.attr.mmap2) { - perf_missing_features.mmap2 = true; - pr_debug2_peo("switching off mmap2\n"); + } + + if (evsel->core.attr.use_clockid && perf_missing_features.clockid) return true; - } else if ((evsel->core.attr.exclude_guest || evsel->core.attr.exclude_host) && - (evsel->pmu == NULL || evsel->pmu->missing_features.exclude_guest)) { - if (evsel->pmu == NULL) { - evsel->pmu = evsel__find_pmu(evsel); - if (evsel->pmu) - evsel->pmu->missing_features.exclude_guest = true; - else { - /* we cannot find PMU, disable attrs now */ - evsel->core.attr.exclude_host = false; - evsel->core.attr.exclude_guest = false; - } - } - if (evsel->exclude_GH) { - pr_debug2_peo("PMU has no exclude_host/guest support, bailing out\n"); - return false; - } - if (!perf_missing_features.exclude_guest) { - perf_missing_features.exclude_guest = true; - pr_debug2_peo("switching off exclude_guest, exclude_host\n"); - } + if ((evsel->open_flags & PERF_FLAG_FD_CLOEXEC) && + perf_missing_features.cloexec) return true; - } else if (!perf_missing_features.sample_id_all) { - perf_missing_features.sample_id_all = true; - pr_debug2_peo("switching off sample_id_all\n"); + + if (evsel->core.attr.mmap2 && perf_missing_features.mmap2) return true; - } else if (!perf_missing_features.lbr_flags && - (evsel->core.attr.branch_sample_type & - (PERF_SAMPLE_BRANCH_NO_CYCLES | - PERF_SAMPLE_BRANCH_NO_FLAGS))) { - perf_missing_features.lbr_flags = true; - pr_debug2_peo("switching off branch sample type no (cycles/flags)\n"); + + if ((evsel->core.attr.branch_sample_type & (PERF_SAMPLE_BRANCH_NO_FLAGS | + PERF_SAMPLE_BRANCH_NO_CYCLES)) && + perf_missing_features.lbr_flags) return true; - } else if (!perf_missing_features.group_read && - evsel->core.attr.inherit && - (evsel->core.attr.read_format & PERF_FORMAT_GROUP) && - evsel__is_group_leader(evsel)) { - perf_missing_features.group_read = true; - pr_debug2_peo("switching off group read\n"); + + if (evsel->core.attr.inherit && (evsel->core.attr.read_format & PERF_FORMAT_GROUP) && + perf_missing_features.group_read) return true; - } else { - return false; - } -} -bool evsel__increase_rlimit(enum rlimit_action *set_rlimit) -{ - int old_errno; - struct rlimit l; + if (evsel->core.attr.ksymbol && perf_missing_features.ksymbol) + return true; - if (*set_rlimit < INCREASED_MAX) { - old_errno = errno; + if (evsel->core.attr.bpf_event && perf_missing_features.bpf) + return true; - if (getrlimit(RLIMIT_NOFILE, &l) == 0) { - if (*set_rlimit == NO_CHANGE) { - l.rlim_cur = l.rlim_max; - } else { - l.rlim_cur = l.rlim_max + 1000; - l.rlim_max = l.rlim_cur; - } - if (setrlimit(RLIMIT_NOFILE, &l) == 0) { - (*set_rlimit) += 1; - errno = old_errno; - return true; - } - } - errno = old_errno; - } + if ((evsel->core.attr.branch_sample_type & PERF_SAMPLE_BRANCH_HW_INDEX) && + perf_missing_features.branch_hw_idx) + return true; + + if (evsel->core.attr.sample_id_all && perf_missing_features.sample_id_all) + return true; return false; } @@ -1989,10 +2669,16 @@ static int evsel__open_cpu(struct evsel *evsel, struct perf_cpu_map *cpus, int idx, thread, nthreads; int pid = -1, err, old_errno; enum rlimit_action set_rlimit = NO_CHANGE; + struct perf_cpu cpu; + + if (evsel__is_retire_lat(evsel)) { + err = evsel__tpebs_open(evsel); + goto out; + } err = __evsel__prepare_open(evsel, cpus, threads); if (err) - return err; + goto out; if (cpus == NULL) cpus = empty_cpu_map; @@ -2000,10 +2686,7 @@ static int evsel__open_cpu(struct evsel *evsel, struct perf_cpu_map *cpus, if (threads == NULL) threads = empty_thread_map; - if (evsel->core.system_wide) - nthreads = 1; - else - nthreads = threads->nr; + nthreads = perf_thread_map__nr(threads); if (evsel->cgrp) pid = evsel->cgrp->fd; @@ -2011,9 +2694,30 @@ static int evsel__open_cpu(struct evsel *evsel, struct perf_cpu_map *cpus, fallback_missing_features: evsel__disable_missing_features(evsel); + pr_debug3("Opening: %s\n", evsel__name(evsel)); display_attr(&evsel->core.attr); + if (evsel__is_tool(evsel)) { + err = evsel__tool_pmu_open(evsel, threads, + start_cpu_map_idx, + end_cpu_map_idx); + goto out; + } + if (evsel__is_hwmon(evsel)) { + err = evsel__hwmon_pmu_open(evsel, threads, + start_cpu_map_idx, + end_cpu_map_idx); + goto out; + } + if (evsel__is_drm(evsel)) { + err = evsel__drm_pmu_open(evsel, threads, + start_cpu_map_idx, + end_cpu_map_idx); + goto out; + } + for (idx = start_cpu_map_idx; idx < end_cpu_map_idx; idx++) { + cpu = perf_cpu_map__cpu(cpus, idx); for (thread = 0; thread < nthreads; thread++) { int fd, group_fd; @@ -2026,13 +2730,17 @@ retry_open: group_fd = get_group_fd(evsel, idx, thread); - test_attr__ready(); + if (group_fd == -2) { + pr_debug("broken group leader for %s\n", evsel->name); + err = -EINVAL; + goto out_close; + } + /* Debug message used by test scripts */ pr_debug2_peo("sys_perf_event_open: pid %d cpu %d group_fd %d flags %#lx", - pid, perf_cpu_map__cpu(cpus, idx).cpu, group_fd, evsel->open_flags); + pid, cpu.cpu, group_fd, evsel->open_flags); - fd = sys_perf_event_open(&evsel->core.attr, pid, - perf_cpu_map__cpu(cpus, idx).cpu, + fd = sys_perf_event_open(&evsel->core.attr, pid, cpu.cpu, group_fd, evsel->open_flags); FD(evsel, idx, thread) = fd; @@ -2047,12 +2755,12 @@ retry_open: bpf_counter__install_pe(evsel, idx, fd); - if (unlikely(test_attr__enabled)) { - test_attr__open(&evsel->core.attr, pid, - perf_cpu_map__cpu(cpus, idx), + if (unlikely(test_attr__enabled())) { + test_attr__open(&evsel->core.attr, pid, cpu, fd, group_fd, evsel->open_flags); } + /* Debug message used by test scripts */ pr_debug2_peo(" = %d\n", fd); if (evsel->bpf_fd >= 0) { @@ -2084,12 +2792,10 @@ retry_open: } } - return 0; + err = 0; + goto out; try_fallback: - if (evsel__precise_ip_fallback(evsel)) - goto retry_open; - if (evsel__ignore_missing_thread(evsel, perf_cpu_map__nr(cpus), idx, threads, thread, err)) { /* We just removed 1 thread, so lower the upper nthreads limit. */ @@ -2103,14 +2809,15 @@ try_fallback: * perf stat needs between 5 and 22 fds per CPU. When we run out * of them try to increase the limits. */ - if (err == -EMFILE && evsel__increase_rlimit(&set_rlimit)) + if (err == -EMFILE && rlimit__increase_nofile(&set_rlimit)) goto retry_open; - if (err != -EINVAL || idx > 0 || thread > 0) - goto out_close; - - if (evsel__detect_missing_features(evsel)) + if (err == -EINVAL && evsel__detect_missing_features(evsel, cpu)) goto fallback_missing_features; + + if (evsel__precise_ip_fallback(evsel)) + goto retry_open; + out_close: if (err) threads->err_thread = thread; @@ -2125,6 +2832,9 @@ out_close: thread = nthreads; } while (--idx >= 0); errno = old_errno; +out: + if (err) + evsel->supported = false; return err; } @@ -2136,21 +2846,38 @@ int evsel__open(struct evsel *evsel, struct perf_cpu_map *cpus, void evsel__close(struct evsel *evsel) { + if (evsel__is_retire_lat(evsel)) + evsel__tpebs_close(evsel); perf_evsel__close(&evsel->core); perf_evsel__free_id(&evsel->core); } -int evsel__open_per_cpu(struct evsel *evsel, struct perf_cpu_map *cpus, int cpu_map_idx) +int evsel__open_per_cpu_and_thread(struct evsel *evsel, + struct perf_cpu_map *cpus, int cpu_map_idx, + struct perf_thread_map *threads) { if (cpu_map_idx == -1) - return evsel__open_cpu(evsel, cpus, NULL, 0, perf_cpu_map__nr(cpus)); + return evsel__open_cpu(evsel, cpus, threads, 0, perf_cpu_map__nr(cpus)); + + return evsel__open_cpu(evsel, cpus, threads, cpu_map_idx, cpu_map_idx + 1); +} - return evsel__open_cpu(evsel, cpus, NULL, cpu_map_idx, cpu_map_idx + 1); +int evsel__open_per_cpu(struct evsel *evsel, struct perf_cpu_map *cpus, int cpu_map_idx) +{ + struct perf_thread_map *threads = thread_map__new_by_tid(-1); + int ret = evsel__open_per_cpu_and_thread(evsel, cpus, cpu_map_idx, threads); + + perf_thread_map__put(threads); + return ret; } int evsel__open_per_thread(struct evsel *evsel, struct perf_thread_map *threads) { - return evsel__open(evsel, NULL, threads); + struct perf_cpu_map *cpus = perf_cpu_map__new_any_cpu(); + int ret = evsel__open_per_cpu_and_thread(evsel, cpus, -1, threads); + + perf_cpu_map__put(cpus); + return ret; } static int perf_evsel__parse_id_sample(const struct evsel *evsel, @@ -2243,11 +2970,18 @@ perf_event__check_size(union perf_event *event, unsigned int sample_size) return 0; } -void __weak arch_perf_parse_sample_weight(struct perf_sample *data, - const __u64 *array, - u64 type __maybe_unused) +static void perf_parse_sample_weight(struct perf_sample *data, const __u64 *array, u64 type) { - data->weight = *array; + union perf_sample_weight weight; + + weight.full = *array; + if (type & PERF_SAMPLE_WEIGHT_STRUCT) { + data->weight = weight.var1_dw; + data->ins_lat = weight.var2_w; + data->weight3 = weight.var3_w; + } else { + data->weight = weight.full; + } } u64 evsel__bitfield_swap_branch_flags(u64 value) @@ -2265,7 +2999,10 @@ u64 evsel__bitfield_swap_branch_flags(u64 value) * abort:1 //transaction abort * cycles:16 //cycle count to last branch * type:4 //branch type - * reserved:40 + * spec:2 //branch speculation info + * new_type:4 //additional branch type + * priv:3 //privilege level + * reserved:31 * } * } * @@ -2273,18 +3010,18 @@ u64 evsel__bitfield_swap_branch_flags(u64 value) * as it has variable bit-field sizes. Instead the * macro takes the bit-field position/size, * swaps it based on the host endianness. - * - * tep_is_bigendian() is used here instead of - * bigendian() to avoid python test fails. */ - if (tep_is_bigendian()) { + if (host_is_bigendian()) { new_val = bitfield_swap(value, 0, 1); new_val |= bitfield_swap(value, 1, 1); new_val |= bitfield_swap(value, 2, 1); new_val |= bitfield_swap(value, 3, 1); new_val |= bitfield_swap(value, 4, 16); new_val |= bitfield_swap(value, 20, 4); - new_val |= bitfield_swap(value, 24, 40); + new_val |= bitfield_swap(value, 24, 2); + new_val |= bitfield_swap(value, 26, 4); + new_val |= bitfield_swap(value, 30, 3); + new_val |= bitfield_swap(value, 33, 31); } else { new_val = bitfield_swap(value, 63, 1); new_val |= bitfield_swap(value, 62, 1); @@ -2292,12 +3029,61 @@ u64 evsel__bitfield_swap_branch_flags(u64 value) new_val |= bitfield_swap(value, 60, 1); new_val |= bitfield_swap(value, 44, 16); new_val |= bitfield_swap(value, 40, 4); - new_val |= bitfield_swap(value, 0, 40); + new_val |= bitfield_swap(value, 38, 2); + new_val |= bitfield_swap(value, 34, 4); + new_val |= bitfield_swap(value, 31, 3); + new_val |= bitfield_swap(value, 0, 31); } return new_val; } +static inline bool evsel__has_branch_counters(const struct evsel *evsel) +{ + struct evsel *leader = evsel__leader(evsel); + + /* The branch counters feature only supports group */ + if (!leader || !evsel->evlist) + return false; + + if (evsel->evlist->nr_br_cntr < 0) + evlist__update_br_cntr(evsel->evlist); + + if (leader->br_cntr_nr > 0) + return true; + + return false; +} + +static int __set_offcpu_sample(struct perf_sample *data) +{ + u64 *array = data->raw_data; + u32 max_size = data->raw_size, *p32; + const void *endp = (void *)array + max_size; + + if (array == NULL) + return -EFAULT; + + OVERFLOW_CHECK_u64(array); + p32 = (void *)array++; + data->pid = p32[0]; + data->tid = p32[1]; + + OVERFLOW_CHECK_u64(array); + data->period = *array++; + + OVERFLOW_CHECK_u64(array); + data->callchain = (struct ip_callchain *)array++; + OVERFLOW_CHECK(array, data->callchain->nr * sizeof(u64), max_size); + data->ip = data->callchain->ips[1]; + array += data->callchain->nr; + + OVERFLOW_CHECK_u64(array); + data->cgroup = *array; + + return 0; +} + int evsel__parse_sample(struct evsel *evsel, union perf_event *event, struct perf_sample *data) { @@ -2320,8 +3106,22 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event, data->period = evsel->core.attr.sample_period; data->cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; data->misc = event->header.misc; - data->id = -1ULL; data->data_src = PERF_MEM_DATA_SRC_NONE; + data->vcpu = -1; + + if (event->header.type == PERF_RECORD_CALLCHAIN_DEFERRED) { + const u64 max_callchain_nr = UINT64_MAX / sizeof(u64); + + data->callchain = (struct ip_callchain *)&event->callchain_deferred.nr; + if (data->callchain->nr > max_callchain_nr) + return -EFAULT; + + data->deferred_cookie = event->callchain_deferred.cookie; + + if (evsel->core.attr.sample_id_all) + perf_evsel__parse_id_sample(evsel, event, data); + return 0; + } if (event->header.type != PERF_RECORD_SAMPLE) { if (!evsel->core.attr.sample_id_all) @@ -2426,8 +3226,8 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event, if (data->read.group.nr > max_group_nr) return -EFAULT; - sz = data->read.group.nr * - sizeof(struct sample_read_value); + + sz = data->read.group.nr * sample_read_value_size(read_format); OVERFLOW_CHECK(array, sz, max_size); data->read.group.values = (struct sample_read_value *)array; @@ -2436,17 +3236,36 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event, OVERFLOW_CHECK_u64(array); data->read.one.id = *array; array++; + + if (read_format & PERF_FORMAT_LOST) { + OVERFLOW_CHECK_u64(array); + data->read.one.lost = *array; + array++; + } } } if (type & PERF_SAMPLE_CALLCHAIN) { const u64 max_callchain_nr = UINT64_MAX / sizeof(u64); + u64 callchain_nr; OVERFLOW_CHECK_u64(array); data->callchain = (struct ip_callchain *)array++; - if (data->callchain->nr > max_callchain_nr) + callchain_nr = data->callchain->nr; + if (callchain_nr > max_callchain_nr) return -EFAULT; - sz = data->callchain->nr * sizeof(u64); + sz = callchain_nr * sizeof(u64); + /* + * Save the cookie for the deferred user callchain. The last 2 + * entries in the callchain should be the context marker and the + * cookie. The cookie will be used to match PERF_RECORD_ + * CALLCHAIN_DEFERRED later. + */ + if (evsel->core.attr.defer_callchain && callchain_nr >= 2 && + data->callchain->ips[callchain_nr - 2] == PERF_CONTEXT_USER_DEFERRED) { + data->deferred_cookie = data->callchain->ips[callchain_nr - 1]; + data->deferred_callchain = true; + } OVERFLOW_CHECK(array, sz, max_size); array = (void *)array + sz; } @@ -2524,20 +3343,30 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event, OVERFLOW_CHECK(array, sz, max_size); array = (void *)array + sz; + + if (evsel__has_branch_counters(evsel)) { + data->branch_stack_cntr = (u64 *)array; + sz = data->branch_stack->nr * sizeof(u64); + + OVERFLOW_CHECK(array, sz, max_size); + array = (void *)array + sz; + } } if (type & PERF_SAMPLE_REGS_USER) { + struct regs_dump *regs = perf_sample__user_regs(data); + OVERFLOW_CHECK_u64(array); - data->user_regs.abi = *array; + regs->abi = *array; array++; - if (data->user_regs.abi) { + if (regs->abi) { u64 mask = evsel->core.attr.sample_regs_user; sz = hweight64(mask) * sizeof(u64); OVERFLOW_CHECK(array, sz, max_size); - data->user_regs.mask = mask; - data->user_regs.regs = (u64 *)array; + regs->mask = mask; + regs->regs = (u64 *)array; array = (void *)array + sz; } } @@ -2565,7 +3394,7 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event, if (type & PERF_SAMPLE_WEIGHT_TYPE) { OVERFLOW_CHECK_u64(array); - arch_perf_parse_sample_weight(data, array, type); + perf_parse_sample_weight(data, array, type); array++; } @@ -2581,19 +3410,20 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event, array++; } - data->intr_regs.abi = PERF_SAMPLE_REGS_ABI_NONE; if (type & PERF_SAMPLE_REGS_INTR) { + struct regs_dump *regs = perf_sample__intr_regs(data); + OVERFLOW_CHECK_u64(array); - data->intr_regs.abi = *array; + regs->abi = *array; array++; - if (data->intr_regs.abi != PERF_SAMPLE_REGS_ABI_NONE) { + if (regs->abi != PERF_SAMPLE_REGS_ABI_NONE) { u64 mask = evsel->core.attr.sample_regs_intr; sz = hweight64(mask) * sizeof(u64); OVERFLOW_CHECK(array, sz, max_size); - data->intr_regs.mask = mask; - data->intr_regs.regs = (u64 *)array; + regs->mask = mask; + regs->regs = (u64 *)array; array = (void *)array + sz; } } @@ -2635,6 +3465,9 @@ int evsel__parse_sample(struct evsel *evsel, union perf_event *event, array = (void *)array + sz; } + if (evsel__is_offcpu_event(evsel)) + return __set_offcpu_sample(data); + return 0; } @@ -2681,9 +3514,45 @@ int evsel__parse_sample_timestamp(struct evsel *evsel, union perf_event *event, return 0; } +u16 evsel__id_hdr_size(const struct evsel *evsel) +{ + u64 sample_type = evsel->core.attr.sample_type; + u16 size = 0; + + if (sample_type & PERF_SAMPLE_TID) + size += sizeof(u64); + + if (sample_type & PERF_SAMPLE_TIME) + size += sizeof(u64); + + if (sample_type & PERF_SAMPLE_ID) + size += sizeof(u64); + + if (sample_type & PERF_SAMPLE_STREAM_ID) + size += sizeof(u64); + + if (sample_type & PERF_SAMPLE_CPU) + size += sizeof(u64); + + if (sample_type & PERF_SAMPLE_IDENTIFIER) + size += sizeof(u64); + + return size; +} + +#ifdef HAVE_LIBTRACEEVENT struct tep_format_field *evsel__field(struct evsel *evsel, const char *name) { - return tep_find_field(evsel->tp_format, name); + struct tep_event *tp_format = evsel__tp_format(evsel); + + return tp_format ? tep_find_field(tp_format, name) : NULL; +} + +struct tep_format_field *evsel__common_field(struct evsel *evsel, const char *name) +{ + struct tep_event *tp_format = evsel__tp_format(evsel); + + return tp_format ? tep_find_common_field(tp_format, name) : NULL; } void *evsel__rawptr(struct evsel *evsel, struct perf_sample *sample, const char *name) @@ -2699,7 +3568,7 @@ void *evsel__rawptr(struct evsel *evsel, struct perf_sample *sample, const char if (field->flags & TEP_FIELD_IS_DYNAMIC) { offset = *(int *)(sample->raw_data + field->offset); offset &= 0xffff; - if (field->flags & TEP_FIELD_IS_RELATIVE) + if (tep_field_is_relative(field->flags)) offset += field->offset + field->size; } @@ -2749,13 +3618,53 @@ u64 evsel__intval(struct evsel *evsel, struct perf_sample *sample, const char *n { struct tep_format_field *field = evsel__field(evsel, name); - if (!field) - return 0; + return field ? format_field__intval(field, sample, evsel->needs_swap) : 0; +} + +u64 evsel__intval_common(struct evsel *evsel, struct perf_sample *sample, const char *name) +{ + struct tep_format_field *field = evsel__common_field(evsel, name); return field ? format_field__intval(field, sample, evsel->needs_swap) : 0; } -bool evsel__fallback(struct evsel *evsel, int err, char *msg, size_t msgsize) +char evsel__taskstate(struct evsel *evsel, struct perf_sample *sample, const char *name) +{ + static struct tep_format_field *prev_state_field; + static const char *states; + struct tep_format_field *field; + unsigned long long val; + unsigned int bit; + char state = '?'; /* '?' denotes unknown task state */ + + field = evsel__field(evsel, name); + + if (!field) + return state; + + if (!states || field != prev_state_field) { + states = parse_task_states(field); + if (!states) + return state; + prev_state_field = field; + } + + /* + * Note since the kernel exposes TASK_REPORT_MAX to userspace + * to denote the 'preempted' state, we might as welll report + * 'R' for this case, which make senses to users as well. + * + * We can change this if we have a good reason in the future. + */ + val = evsel__intval(evsel, sample, name); + bit = val ? ffs(val) : 0; + state = (!bit || bit > strlen(states)) ? 'R' : states[bit-1]; + return state; +} +#endif + +bool evsel__fallback(struct evsel *evsel, struct target *target, int err, + char *msg, size_t msgsize) { int paranoid; @@ -2763,18 +3672,19 @@ bool evsel__fallback(struct evsel *evsel, int err, char *msg, size_t msgsize) evsel->core.attr.type == PERF_TYPE_HARDWARE && evsel->core.attr.config == PERF_COUNT_HW_CPU_CYCLES) { /* - * If it's cycles then fall back to hrtimer based - * cpu-clock-tick sw counter, which is always available even if - * no PMU support. + * If it's cycles then fall back to hrtimer based cpu-clock sw + * counter, which is always available even if no PMU support. * * PPC returns ENXIO until 2.6.37 (behavior changed with commit * b0a873e). */ - scnprintf(msg, msgsize, "%s", -"The cycles event is not supported, trying to fall back to cpu-clock-ticks"); - evsel->core.attr.type = PERF_TYPE_SOFTWARE; - evsel->core.attr.config = PERF_COUNT_SW_CPU_CLOCK; + evsel->core.attr.config = target__has_cpu(target) + ? PERF_COUNT_SW_CPU_CLOCK + : PERF_COUNT_SW_TASK_CLOCK; + scnprintf(msg, msgsize, + "The cycles event is not supported, trying to fall back to %s", + target__has_cpu(target) ? "cpu-clock" : "task-clock"); zfree(&evsel->name); return true; @@ -2786,7 +3696,7 @@ bool evsel__fallback(struct evsel *evsel, int err, char *msg, size_t msgsize) /* If event has exclude user then don't exclude kernel. */ if (evsel->core.attr.exclude_user) - return false; + goto no_fallback; /* Is there already the separator in the name. */ if (strchr(name, '/') || @@ -2794,10 +3704,9 @@ bool evsel__fallback(struct evsel *evsel, int err, char *msg, size_t msgsize) sep = ""; if (asprintf(&new_name, "%s%su", name, sep) < 0) - return false; + goto no_fallback; - if (evsel->name) - free(evsel->name); + free(evsel->name); evsel->name = new_name; scnprintf(msg, msgsize, "kernel.perf_event_paranoid=%d, trying " "to fall back to excluding kernel and hypervisor " @@ -2806,8 +3715,31 @@ bool evsel__fallback(struct evsel *evsel, int err, char *msg, size_t msgsize) evsel->core.attr.exclude_hv = 1; return true; - } + } else if (err == EOPNOTSUPP && !evsel->core.attr.exclude_guest && + !evsel->exclude_GH) { + const char *name = evsel__name(evsel); + char *new_name; + const char *sep = ":"; + + /* Is there already the separator in the name. */ + if (strchr(name, '/') || + (strchr(name, ':') && !evsel->is_libpfm_event)) + sep = ""; + + if (asprintf(&new_name, "%s%sH", name, sep) < 0) + goto no_fallback; + + free(evsel->name); + evsel->name = new_name; + /* Apple M1 requires exclude_guest */ + scnprintf(msg, msgsize, "Trying to fall back to excluding guest samples"); + evsel->core.attr.exclude_guest = 1; + return true; + } +no_fallback: + scnprintf(msg, msgsize, "No fallback found for '%s' for error %d", + evsel__name(evsel), err); return false; } @@ -2847,11 +3779,93 @@ static bool find_process(const char *name) return ret ? false : true; } +static int dump_perf_event_processes(char *msg, size_t size) +{ + DIR *proc_dir; + struct dirent *proc_entry; + int printed = 0; + + proc_dir = opendir(procfs__mountpoint()); + if (!proc_dir) + return 0; + + /* Walk through the /proc directory. */ + while ((proc_entry = readdir(proc_dir)) != NULL) { + char buf[256]; + DIR *fd_dir; + struct dirent *fd_entry; + int fd_dir_fd; + + if (proc_entry->d_type != DT_DIR || + !isdigit(proc_entry->d_name[0]) || + strlen(proc_entry->d_name) > sizeof(buf) - 4) + continue; + + scnprintf(buf, sizeof(buf), "%s/fd", proc_entry->d_name); + fd_dir_fd = openat(dirfd(proc_dir), buf, O_DIRECTORY); + if (fd_dir_fd == -1) + continue; + fd_dir = fdopendir(fd_dir_fd); + if (!fd_dir) { + close(fd_dir_fd); + continue; + } + while ((fd_entry = readdir(fd_dir)) != NULL) { + ssize_t link_size; + + if (fd_entry->d_type != DT_LNK) + continue; + link_size = readlinkat(fd_dir_fd, fd_entry->d_name, buf, sizeof(buf)); + if (link_size < 0) + continue; + /* Take care as readlink doesn't null terminate the string. */ + if (!strncmp(buf, "anon_inode:[perf_event]", link_size)) { + int cmdline_fd; + ssize_t cmdline_size; + + scnprintf(buf, sizeof(buf), "%s/cmdline", proc_entry->d_name); + cmdline_fd = openat(dirfd(proc_dir), buf, O_RDONLY); + if (cmdline_fd == -1) + continue; + cmdline_size = read(cmdline_fd, buf, sizeof(buf) - 1); + close(cmdline_fd); + if (cmdline_size < 0) + continue; + buf[cmdline_size] = '\0'; + for (ssize_t i = 0; i < cmdline_size; i++) { + if (buf[i] == '\0') + buf[i] = ' '; + } + + if (printed == 0) + printed += scnprintf(msg, size, "Possible processes:\n"); + + printed += scnprintf(msg + printed, size - printed, + "%s %s\n", proc_entry->d_name, buf); + break; + } + } + closedir(fd_dir); + } + closedir(proc_dir); + return printed; +} + +int __weak arch_evsel__open_strerror(struct evsel *evsel __maybe_unused, + int err __maybe_unused, + char *msg __maybe_unused, + size_t size __maybe_unused) +{ + return 0; +} + int evsel__open_strerror(struct evsel *evsel, struct target *target, int err, char *msg, size_t size) { + struct perf_pmu *pmu; char sbuf[STRERR_BUFSIZE]; int printed = 0, enforced = 0; + int ret; switch (err) { case EPERM: @@ -2872,7 +3886,7 @@ int evsel__open_strerror(struct evsel *evsel, struct target *target, printed += scnprintf(msg, size, "No permission to enable %s event.\n\n", evsel__name(evsel)); - return scnprintf(msg + printed, size - printed, + return printed + scnprintf(msg + printed, size - printed, "Consider adjusting /proc/sys/kernel/perf_event_paranoid setting to open\n" "access to performance monitoring and observability operations for processes\n" "without CAP_PERFMON, CAP_SYS_PTRACE or CAP_SYS_ADMIN Linux capability.\n" @@ -2909,10 +3923,18 @@ int evsel__open_strerror(struct evsel *evsel, struct target *target, "No such device - did you specify an out-of-range profile CPU?"); break; case EOPNOTSUPP: + if (evsel->core.attr.sample_type & PERF_SAMPLE_BRANCH_STACK) + return scnprintf(msg, size, + "%s: PMU Hardware or event type doesn't support branch stack sampling.", + evsel__name(evsel)); if (evsel->core.attr.aux_output) return scnprintf(msg, size, "%s: PMU Hardware doesn't support 'aux_output' feature", evsel__name(evsel)); + if (evsel->core.attr.aux_action) + return scnprintf(msg, size, + "%s: PMU Hardware doesn't support 'aux_action' feature", + evsel__name(evsel)); if (evsel->core.attr.sample_period != 0) return scnprintf(msg, size, "%s: PMU Hardware doesn't support sampling/overflow-interrupts. Try 'perf stat'", @@ -2925,12 +3947,21 @@ int evsel__open_strerror(struct evsel *evsel, struct target *target, return scnprintf(msg, size, "%s", "No hardware sampling interrupt available.\n"); #endif + if (!target__has_cpu(target)) + return scnprintf(msg, size, + "Unsupported event (%s) in per-thread mode, enable system wide with '-a'.", + evsel__name(evsel)); break; case EBUSY: if (find_process("oprofiled")) return scnprintf(msg, size, "The PMU counters are busy/taken by another profiler.\n" "We found oprofile daemon running, please stop it and try again."); + printed += scnprintf( + msg, size, + "The PMU %s counters are busy and in use by another process.\n", + evsel->pmu ? evsel->pmu->name : ""); + return printed + dump_perf_event_processes(msg + printed, size - printed); break; case EINVAL: if (evsel->core.attr.sample_type & PERF_SAMPLE_CODE_PAGE_SIZE && perf_missing_features.code_page_size) @@ -2943,12 +3974,16 @@ int evsel__open_strerror(struct evsel *evsel, struct target *target, return scnprintf(msg, size, "clockid feature not supported."); if (perf_missing_features.clockid_wrong) return scnprintf(msg, size, "wrong clockid (%d).", clockid); + if (perf_missing_features.aux_action) + return scnprintf(msg, size, "The 'aux_action' feature is not supported, update the kernel."); if (perf_missing_features.aux_output) return scnprintf(msg, size, "The 'aux_output' feature is not supported, update the kernel."); - if (!target__has_cpu(target)) + pmu = evsel__find_pmu(evsel); + if (!pmu->is_core && !target__has_cpu(target)) return scnprintf(msg, size, "Invalid event (%s) in per-thread mode, enable system wide with '-a'.", evsel__name(evsel)); + break; case ENODATA: return scnprintf(msg, size, "Cannot collect data source with the load latency event alone. " @@ -2957,23 +3992,38 @@ int evsel__open_strerror(struct evsel *evsel, struct target *target, break; } + ret = arch_evsel__open_strerror(evsel, err, msg, size); + if (ret) + return ret; + return scnprintf(msg, size, "The sys_perf_event_open() syscall returned with %d (%s) for event (%s).\n" - "/bin/dmesg | grep -i perf may provide additional information.\n", + "\"dmesg | grep -i perf\" may provide additional information.\n", err, str_error_r(err, sbuf, sizeof(sbuf)), evsel__name(evsel)); } +struct perf_session *evsel__session(struct evsel *evsel) +{ + return evsel && evsel->evlist ? evsel->evlist->session : NULL; +} + struct perf_env *evsel__env(struct evsel *evsel) { - if (evsel && evsel->evlist) - return evsel->evlist->env; - return &perf_env; + struct perf_session *session = evsel__session(evsel); + + return session ? perf_session__env(session) : NULL; } static int store_evsel_ids(struct evsel *evsel, struct evlist *evlist) { int cpu_map_idx, thread; + if (evsel__is_retire_lat(evsel)) + return 0; + + if (perf_pmu__kind(evsel->pmu) != PERF_PMU_KIND_PE) + return 0; + for (cpu_map_idx = 0; cpu_map_idx < xyarray__max_x(evsel->core.fd); cpu_map_idx++) { for (thread = 0; thread < xyarray__max_y(evsel->core.fd); thread++) { @@ -3006,19 +4056,29 @@ void evsel__zero_per_pkg(struct evsel *evsel) if (evsel->per_pkg_mask) { hashmap__for_each_entry(evsel->per_pkg_mask, cur, bkt) - free((char *)cur->key); + zfree(&cur->pkey); hashmap__clear(evsel->per_pkg_mask); } } -bool evsel__is_hybrid(struct evsel *evsel) +/** + * evsel__is_hybrid - does the evsel have a known PMU that is hybrid. Note, this + * will be false on hybrid systems for hardware and legacy + * cache events. + */ +bool evsel__is_hybrid(const struct evsel *evsel) { - return evsel->pmu_name && perf_pmu__is_hybrid(evsel->pmu_name); + if (!evsel->core.is_pmu_core) + return false; + + return perf_pmus__num_core_pmus() > 1; } -struct evsel *evsel__leader(struct evsel *evsel) +struct evsel *evsel__leader(const struct evsel *evsel) { + if (evsel->core.leader == NULL) + return NULL; return container_of(evsel->core.leader, struct evsel, core); } @@ -3048,3 +4108,171 @@ int evsel__source_count(const struct evsel *evsel) } return count; } + +bool __weak arch_evsel__must_be_in_group(const struct evsel *evsel __maybe_unused) +{ + return false; +} + +/* + * Remove an event from a given group (leader). + * Some events, e.g., perf metrics Topdown events, + * must always be grouped. Ignore the events. + */ +void evsel__remove_from_group(struct evsel *evsel, struct evsel *leader) +{ + if (!arch_evsel__must_be_in_group(evsel) && evsel != leader) { + evsel__set_leader(evsel, evsel); + evsel->core.nr_members = 0; + leader->core.nr_members--; + } +} + +bool evsel__set_needs_uniquify(struct evsel *counter, const struct perf_stat_config *config) +{ + struct evsel *evsel; + + if (counter->needs_uniquify) { + /* Already set. */ + return true; + } + + if (counter->use_config_name || counter->is_libpfm_event) { + /* Original name will be used. */ + return false; + } + + if (!config->hybrid_merge && evsel__is_hybrid(counter)) { + /* Unique hybrid counters necessary. */ + counter->needs_uniquify = true; + return true; + } + + if (counter->core.attr.type < PERF_TYPE_MAX && counter->core.attr.type != PERF_TYPE_RAW) { + /* Legacy event, don't uniquify. */ + return false; + } + + if (counter->pmu && counter->pmu->is_core && + counter->alternate_hw_config != PERF_COUNT_HW_MAX) { + /* A sysfs or json event replacing a legacy event, don't uniquify. */ + return false; + } + + if (config->aggr_mode == AGGR_NONE) { + /* Always unique with no aggregation. */ + counter->needs_uniquify = true; + return true; + } + + if (counter->first_wildcard_match != NULL) { + /* + * If stats are merged then only the first_wildcard_match is + * displayed, there is no need to uniquify this evsel as the + * name won't be shown. + */ + return false; + } + + /* + * Do other non-merged events in the evlist have the same name? If so + * uniquify is necessary. + */ + evlist__for_each_entry(counter->evlist, evsel) { + if (evsel == counter || evsel->first_wildcard_match || evsel->pmu == counter->pmu) + continue; + + if (evsel__name_is(counter, evsel__name(evsel))) { + counter->needs_uniquify = true; + return true; + } + } + return false; +} + +void evsel__uniquify_counter(struct evsel *counter) +{ + const char *name, *pmu_name, *config; + char *new_name; + int len, ret; + + /* No uniquification necessary. */ + if (!counter->needs_uniquify) + return; + + /* The evsel was already uniquified. */ + if (counter->uniquified_name) + return; + + /* Avoid checking to uniquify twice. */ + counter->uniquified_name = true; + + name = evsel__name(counter); + config = strchr(name, '/'); + pmu_name = counter->pmu->name; + + /* Already prefixed by the PMU name? */ + len = pmu_name_len_no_suffix(pmu_name); + + if (!strncmp(name, pmu_name, len)) { + /* + * If the PMU name is there, then there is no sense in not + * having a slash. Do this for robustness. + */ + if (config == NULL) + config = name - 1; + + ret = asprintf(&new_name, "%s/%s", pmu_name, config + 1); + } else if (config) { + len = config - name; + if (config[1] == '/') { + /* case: event// */ + ret = asprintf(&new_name, "%s/%.*s/%s", pmu_name, len, name, config + 2); + } else { + /* case: event/.../ */ + ret = asprintf(&new_name, "%s/%.*s,%s", pmu_name, len, name, config + 1); + } + } else { + config = strchr(name, ':'); + if (config) { + /* case: event:.. */ + len = config - name; + + ret = asprintf(&new_name, "%s/%.*s/%s", pmu_name, len, name, config + 1); + } else { + /* case: event */ + ret = asprintf(&new_name, "%s/%s/", pmu_name, name); + } + } + if (ret > 0) { + free(counter->name); + counter->name = new_name; + } else { + /* ENOMEM from asprintf. */ + counter->uniquified_name = false; + } +} + +void evsel__warn_user_requested_cpus(struct evsel *evsel, struct perf_cpu_map *user_requested_cpus) +{ + struct perf_cpu_map *intersect, *online = NULL; + const struct perf_pmu *pmu = evsel__find_pmu(evsel); + + if (pmu && pmu->is_core) { + intersect = perf_cpu_map__intersect(pmu->cpus, user_requested_cpus); + } else { + online = cpu_map__online(); + intersect = perf_cpu_map__intersect(online, user_requested_cpus); + } + if (!perf_cpu_map__equal(intersect, user_requested_cpus)) { + char buf1[128]; + char buf2[128]; + + cpu_map__snprint(user_requested_cpus, buf1, sizeof(buf1)); + cpu_map__snprint(online ?: pmu->cpus, buf2, sizeof(buf2)); + pr_warning("WARNING: A requested CPU in '%s' is not supported by PMU '%s' (CPUs %s) for event '%s'\n", + buf1, pmu ? pmu->name : "cpu", buf2, evsel__name(evsel)); + } + perf_cpu_map__put(intersect); + perf_cpu_map__put(online); +} |