diff options
Diffstat (limited to 'tools/perf/util')
119 files changed, 3787 insertions, 1485 deletions
diff --git a/tools/perf/util/Build b/tools/perf/util/Build index 7910d908c814..4959e7a990e4 100644 --- a/tools/perf/util/Build +++ b/tools/perf/util/Build @@ -41,6 +41,7 @@ perf-util-y += rbtree.o perf-util-y += libstring.o perf-util-y += bitmap.o perf-util-y += hweight.o +perf-util-y += sha1.o perf-util-y += smt.o perf-util-y += strbuf.o perf-util-y += string.o @@ -84,8 +85,10 @@ perf-util-y += pmu.o perf-util-y += pmus.o perf-util-y += pmu-flex.o perf-util-y += pmu-bison.o +perf-util-y += drm_pmu.o perf-util-y += hwmon_pmu.o perf-util-y += tool_pmu.o +perf-util-y += tp_pmu.o perf-util-y += svghelper.o perf-util-y += trace-event-info.o perf-util-y += trace-event-scripting.o @@ -175,6 +178,7 @@ perf-util-$(CONFIG_PERF_BPF_SKEL) += btf.o ifeq ($(CONFIG_TRACE),y) perf-util-$(CONFIG_PERF_BPF_SKEL) += bpf-trace-summary.o + perf-util-$(CONFIG_PERF_BPF_SKEL) += bpf_trace_augment.o endif ifeq ($(CONFIG_LIBTRACEEVENT),y) @@ -421,7 +425,7 @@ endif $(OUTPUT)%.shellcheck_log: % $(call rule_mkdir) - $(Q)$(call echo-cmd,test)shellcheck -a -S warning "$<" > $@ || (cat $@ && rm $@ && false) + $(Q)$(call echo-cmd,test)$(SHELLCHECK) "$<" > $@ || (cat $@ && rm $@ && false) perf-util-y += $(SHELL_TEST_LOGS) diff --git a/tools/perf/util/affinity.c b/tools/perf/util/affinity.c index 38dc4524b7e8..4fe851334296 100644 --- a/tools/perf/util/affinity.c +++ b/tools/perf/util/affinity.c @@ -5,6 +5,7 @@ #include <stdlib.h> #include <linux/bitmap.h> #include <linux/zalloc.h> +#include <perf/cpumap.h> #include "perf.h" #include "cpumap.h" #include "affinity.h" @@ -83,3 +84,20 @@ void affinity__cleanup(struct affinity *a) if (a != NULL) __affinity__cleanup(a); } + +void cpu_map__set_affinity(const struct perf_cpu_map *cpumap) +{ + int cpu_set_size = get_cpu_set_size(); + unsigned long *cpuset = bitmap_zalloc(cpu_set_size * 8); + struct perf_cpu cpu; + int idx; + + if (!cpuset) + return; + + perf_cpu_map__for_each_cpu_skip_any(cpu, idx, cpumap) + __set_bit(cpu.cpu, cpuset); + + sched_setaffinity(0, cpu_set_size, (cpu_set_t *)cpuset); + zfree(&cpuset); +} diff --git a/tools/perf/util/affinity.h b/tools/perf/util/affinity.h index 0ad6a18ef20c..7341194b2298 100644 --- a/tools/perf/util/affinity.h +++ b/tools/perf/util/affinity.h @@ -4,6 +4,7 @@ #include <stdbool.h> +struct perf_cpu_map; struct affinity { unsigned long *orig_cpus; unsigned long *sched_cpus; @@ -13,5 +14,6 @@ struct affinity { void affinity__cleanup(struct affinity *a); void affinity__set(struct affinity *a, int cpu); int affinity__setup(struct affinity *a); +void cpu_map__set_affinity(const struct perf_cpu_map *cpumap); #endif // PERF_AFFINITY_H diff --git a/tools/perf/util/amd-sample-raw.c b/tools/perf/util/amd-sample-raw.c index 4b540e6fb42d..b084dee76b1a 100644 --- a/tools/perf/util/amd-sample-raw.c +++ b/tools/perf/util/amd-sample-raw.c @@ -354,7 +354,7 @@ static void parse_cpuid(struct perf_env *env) */ bool evlist__has_amd_ibs(struct evlist *evlist) { - struct perf_env *env = evlist->env; + struct perf_env *env = perf_session__env(evlist->session); int ret, nr_pmu_mappings = perf_env__nr_pmu_mappings(env); const char *pmu_mapping = perf_env__pmu_mappings(env); char name[sizeof("ibs_fetch")]; diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c index 264a212b47df..0dd475a744b6 100644 --- a/tools/perf/util/annotate.c +++ b/tools/perf/util/annotate.c @@ -1451,6 +1451,7 @@ void annotated_source__purge(struct annotated_source *as) list_del_init(&al->node); disasm_line__free(disasm_line(al)); } + as->tried_source = false; } static size_t disasm_line__fprintf(struct disasm_line *dl, FILE *fp) @@ -2280,6 +2281,7 @@ void annotation_options__init(void) opt->annotate_src = true; opt->offset_level = ANNOTATION__OFFSET_JUMP_TARGETS; opt->percent_type = PERCENT_PERIOD_LOCAL; + opt->hide_src_code = true; opt->hide_src_code_on_title = true; } diff --git a/tools/perf/util/annotate.h b/tools/perf/util/annotate.h index bbb89b32f398..8b5131d257b0 100644 --- a/tools/perf/util/annotate.h +++ b/tools/perf/util/annotate.h @@ -294,6 +294,7 @@ struct annotated_source { int nr_entries; int nr_asm_entries; int max_jump_sources; + bool tried_source; u64 start; struct { u8 addr; diff --git a/tools/perf/util/arm-spe.c b/tools/perf/util/arm-spe.c index d46e0cccac99..8942fa598a84 100644 --- a/tools/perf/util/arm-spe.c +++ b/tools/perf/util/arm-spe.c @@ -856,7 +856,7 @@ static bool arm_spe__synth_ds(struct arm_spe_queue *speq, const char *cpuid; pr_warning_once("Old SPE metadata, re-record to improve decode accuracy\n"); - cpuid = perf_env__cpuid(spe->session->evlist->env); + cpuid = perf_env__cpuid(perf_session__env(spe->session)); midr = strtol(cpuid, NULL, 16); } else { /* CPU ID is -1 for per-thread mode */ diff --git a/tools/perf/util/auxtrace.c b/tools/perf/util/auxtrace.c index 03211c2623de..ebd32f1b8f12 100644 --- a/tools/perf/util/auxtrace.c +++ b/tools/perf/util/auxtrace.c @@ -1890,7 +1890,7 @@ int __weak compat_auxtrace_mmap__write_tail(struct auxtrace_mmap *mm, u64 tail) } static int __auxtrace_mmap__read(struct mmap *map, - struct auxtrace_record *itr, + struct auxtrace_record *itr, struct perf_env *env, const struct perf_tool *tool, process_auxtrace_t fn, bool snapshot, size_t snapshot_size) { @@ -1900,7 +1900,7 @@ static int __auxtrace_mmap__read(struct mmap *map, size_t size, head_off, old_off, len1, len2, padding; union perf_event ev; void *data1, *data2; - int kernel_is_64_bit = perf_env__kernel_is_64_bit(evsel__env(NULL)); + int kernel_is_64_bit = perf_env__kernel_is_64_bit(env); head = auxtrace_mmap__read_head(mm, kernel_is_64_bit); @@ -2002,17 +2002,18 @@ static int __auxtrace_mmap__read(struct mmap *map, } int auxtrace_mmap__read(struct mmap *map, struct auxtrace_record *itr, - const struct perf_tool *tool, process_auxtrace_t fn) + struct perf_env *env, const struct perf_tool *tool, + process_auxtrace_t fn) { - return __auxtrace_mmap__read(map, itr, tool, fn, false, 0); + return __auxtrace_mmap__read(map, itr, env, tool, fn, false, 0); } int auxtrace_mmap__read_snapshot(struct mmap *map, - struct auxtrace_record *itr, + struct auxtrace_record *itr, struct perf_env *env, const struct perf_tool *tool, process_auxtrace_t fn, size_t snapshot_size) { - return __auxtrace_mmap__read(map, itr, tool, fn, true, snapshot_size); + return __auxtrace_mmap__read(map, itr, env, tool, fn, true, snapshot_size); } /** diff --git a/tools/perf/util/auxtrace.h b/tools/perf/util/auxtrace.h index b0db84d27b25..f001cbb68f8e 100644 --- a/tools/perf/util/auxtrace.h +++ b/tools/perf/util/auxtrace.h @@ -23,6 +23,7 @@ union perf_event; struct perf_session; struct evlist; struct evsel; +struct perf_env; struct perf_tool; struct mmap; struct perf_sample; @@ -512,10 +513,11 @@ typedef int (*process_auxtrace_t)(const struct perf_tool *tool, size_t len1, void *data2, size_t len2); int auxtrace_mmap__read(struct mmap *map, struct auxtrace_record *itr, - const struct perf_tool *tool, process_auxtrace_t fn); + struct perf_env *env, const struct perf_tool *tool, + process_auxtrace_t fn); int auxtrace_mmap__read_snapshot(struct mmap *map, - struct auxtrace_record *itr, + struct auxtrace_record *itr, struct perf_env *env, const struct perf_tool *tool, process_auxtrace_t fn, size_t snapshot_size); diff --git a/tools/perf/util/bpf-event.c b/tools/perf/util/bpf-event.c index c81444059ad0..5b6d3e899e11 100644 --- a/tools/perf/util/bpf-event.c +++ b/tools/perf/util/bpf-event.c @@ -1,13 +1,21 @@ // SPDX-License-Identifier: GPL-2.0 #include <errno.h> +#include <stddef.h> +#include <stdint.h> +#include <stdio.h> #include <stdlib.h> +#include <string.h> #include <bpf/bpf.h> #include <bpf/btf.h> #include <bpf/libbpf.h> +#include <linux/bpf.h> #include <linux/btf.h> #include <linux/err.h> +#include <linux/perf_event.h> #include <linux/string.h> +#include <linux/zalloc.h> #include <internal/lib.h> +#include <perf/event.h> #include <symbol/kallsyms.h> #include "bpf-event.h" #include "bpf-utils.h" @@ -151,6 +159,362 @@ static int synthesize_bpf_prog_name(char *buf, int size, return name_len; } +#ifdef HAVE_LIBBPF_STRINGS_SUPPORT + +#define BPF_METADATA_PREFIX "bpf_metadata_" +#define BPF_METADATA_PREFIX_LEN (sizeof(BPF_METADATA_PREFIX) - 1) + +static bool name_has_bpf_metadata_prefix(const char **s) +{ + if (strncmp(*s, BPF_METADATA_PREFIX, BPF_METADATA_PREFIX_LEN) != 0) + return false; + *s += BPF_METADATA_PREFIX_LEN; + return true; +} + +struct bpf_metadata_map { + struct btf *btf; + const struct btf_type *datasec; + void *rodata; + size_t rodata_size; + unsigned int num_vars; +}; + +static int bpf_metadata_read_map_data(__u32 map_id, struct bpf_metadata_map *map) +{ + int map_fd; + struct bpf_map_info map_info; + __u32 map_info_len; + int key; + struct btf *btf; + const struct btf_type *datasec; + struct btf_var_secinfo *vsi; + unsigned int vlen, vars; + void *rodata; + + map_fd = bpf_map_get_fd_by_id(map_id); + if (map_fd < 0) + return -1; + + memset(&map_info, 0, sizeof(map_info)); + map_info_len = sizeof(map_info); + if (bpf_obj_get_info_by_fd(map_fd, &map_info, &map_info_len) < 0) + goto out_close; + + /* If it's not an .rodata map, don't bother. */ + if (map_info.type != BPF_MAP_TYPE_ARRAY || + map_info.key_size != sizeof(int) || + map_info.max_entries != 1 || + !map_info.btf_value_type_id || + !strstr(map_info.name, ".rodata")) { + goto out_close; + } + + btf = btf__load_from_kernel_by_id(map_info.btf_id); + if (!btf) + goto out_close; + datasec = btf__type_by_id(btf, map_info.btf_value_type_id); + if (!btf_is_datasec(datasec)) + goto out_free_btf; + + /* + * If there aren't any variables with the "bpf_metadata_" prefix, + * don't bother. + */ + vlen = btf_vlen(datasec); + vsi = btf_var_secinfos(datasec); + vars = 0; + for (unsigned int i = 0; i < vlen; i++, vsi++) { + const struct btf_type *t_var = btf__type_by_id(btf, vsi->type); + const char *name = btf__name_by_offset(btf, t_var->name_off); + + if (name_has_bpf_metadata_prefix(&name)) + vars++; + } + if (vars == 0) + goto out_free_btf; + + rodata = zalloc(map_info.value_size); + if (!rodata) + goto out_free_btf; + key = 0; + if (bpf_map_lookup_elem(map_fd, &key, rodata)) { + free(rodata); + goto out_free_btf; + } + close(map_fd); + + map->btf = btf; + map->datasec = datasec; + map->rodata = rodata; + map->rodata_size = map_info.value_size; + map->num_vars = vars; + return 0; + +out_free_btf: + btf__free(btf); +out_close: + close(map_fd); + return -1; +} + +struct format_btf_ctx { + char *buf; + size_t buf_size; + size_t buf_idx; +}; + +static void format_btf_cb(void *arg, const char *fmt, va_list ap) +{ + int n; + struct format_btf_ctx *ctx = (struct format_btf_ctx *)arg; + + n = vsnprintf(ctx->buf + ctx->buf_idx, ctx->buf_size - ctx->buf_idx, + fmt, ap); + ctx->buf_idx += n; + if (ctx->buf_idx >= ctx->buf_size) + ctx->buf_idx = ctx->buf_size; +} + +static void format_btf_variable(struct btf *btf, char *buf, size_t buf_size, + const struct btf_type *t, const void *btf_data) +{ + struct format_btf_ctx ctx = { + .buf = buf, + .buf_idx = 0, + .buf_size = buf_size, + }; + const struct btf_dump_type_data_opts opts = { + .sz = sizeof(struct btf_dump_type_data_opts), + .skip_names = 1, + .compact = 1, + .emit_strings = 1, + }; + struct btf_dump *d; + size_t btf_size; + + d = btf_dump__new(btf, format_btf_cb, &ctx, NULL); + btf_size = btf__resolve_size(btf, t->type); + btf_dump__dump_type_data(d, t->type, btf_data, btf_size, &opts); + btf_dump__free(d); +} + +static void bpf_metadata_fill_event(struct bpf_metadata_map *map, + struct perf_record_bpf_metadata *bpf_metadata_event) +{ + struct btf_var_secinfo *vsi; + unsigned int i, vlen; + + memset(bpf_metadata_event->prog_name, 0, BPF_PROG_NAME_LEN); + vlen = btf_vlen(map->datasec); + vsi = btf_var_secinfos(map->datasec); + + for (i = 0; i < vlen; i++, vsi++) { + const struct btf_type *t_var = btf__type_by_id(map->btf, + vsi->type); + const char *name = btf__name_by_offset(map->btf, + t_var->name_off); + const __u64 nr_entries = bpf_metadata_event->nr_entries; + struct perf_record_bpf_metadata_entry *entry; + + if (!name_has_bpf_metadata_prefix(&name)) + continue; + + if (nr_entries >= (__u64)map->num_vars) + break; + + entry = &bpf_metadata_event->entries[nr_entries]; + memset(entry, 0, sizeof(*entry)); + snprintf(entry->key, BPF_METADATA_KEY_LEN, "%s", name); + format_btf_variable(map->btf, entry->value, + BPF_METADATA_VALUE_LEN, t_var, + map->rodata + vsi->offset); + bpf_metadata_event->nr_entries++; + } +} + +static void bpf_metadata_free_map_data(struct bpf_metadata_map *map) +{ + btf__free(map->btf); + free(map->rodata); +} + +static struct bpf_metadata *bpf_metadata_alloc(__u32 nr_prog_tags, + __u32 nr_variables) +{ + struct bpf_metadata *metadata; + size_t event_size; + + metadata = zalloc(sizeof(struct bpf_metadata)); + if (!metadata) + return NULL; + + metadata->prog_names = zalloc(nr_prog_tags * sizeof(char *)); + if (!metadata->prog_names) { + bpf_metadata_free(metadata); + return NULL; + } + for (__u32 prog_index = 0; prog_index < nr_prog_tags; prog_index++) { + metadata->prog_names[prog_index] = zalloc(BPF_PROG_NAME_LEN); + if (!metadata->prog_names[prog_index]) { + bpf_metadata_free(metadata); + return NULL; + } + metadata->nr_prog_names++; + } + + event_size = sizeof(metadata->event->bpf_metadata) + + nr_variables * sizeof(metadata->event->bpf_metadata.entries[0]); + metadata->event = zalloc(event_size); + if (!metadata->event) { + bpf_metadata_free(metadata); + return NULL; + } + metadata->event->bpf_metadata = (struct perf_record_bpf_metadata) { + .header = { + .type = PERF_RECORD_BPF_METADATA, + .size = event_size, + }, + .nr_entries = 0, + }; + + return metadata; +} + +static struct bpf_metadata *bpf_metadata_create(struct bpf_prog_info *info) +{ + struct bpf_metadata *metadata; + const __u32 *map_ids = (__u32 *)(uintptr_t)info->map_ids; + + for (__u32 map_index = 0; map_index < info->nr_map_ids; map_index++) { + struct bpf_metadata_map map; + + if (bpf_metadata_read_map_data(map_ids[map_index], &map) != 0) + continue; + + metadata = bpf_metadata_alloc(info->nr_prog_tags, map.num_vars); + if (!metadata) + continue; + + bpf_metadata_fill_event(&map, &metadata->event->bpf_metadata); + + for (__u32 index = 0; index < info->nr_prog_tags; index++) { + synthesize_bpf_prog_name(metadata->prog_names[index], + BPF_PROG_NAME_LEN, info, + map.btf, index); + } + + bpf_metadata_free_map_data(&map); + + return metadata; + } + + return NULL; +} + +static int synthesize_perf_record_bpf_metadata(const struct bpf_metadata *metadata, + const struct perf_tool *tool, + perf_event__handler_t process, + struct machine *machine) +{ + const size_t event_size = metadata->event->header.size; + union perf_event *event; + int err = 0; + + event = zalloc(event_size + machine->id_hdr_size); + if (!event) + return -1; + memcpy(event, metadata->event, event_size); + memset((void *)event + event->header.size, 0, machine->id_hdr_size); + event->header.size += machine->id_hdr_size; + for (__u32 index = 0; index < metadata->nr_prog_names; index++) { + memcpy(event->bpf_metadata.prog_name, + metadata->prog_names[index], BPF_PROG_NAME_LEN); + err = perf_tool__process_synth_event(tool, event, machine, + process); + if (err != 0) + break; + } + + free(event); + return err; +} + +void bpf_metadata_free(struct bpf_metadata *metadata) +{ + if (metadata == NULL) + return; + for (__u32 index = 0; index < metadata->nr_prog_names; index++) + free(metadata->prog_names[index]); + free(metadata->prog_names); + free(metadata->event); + free(metadata); +} + +#else /* HAVE_LIBBPF_STRINGS_SUPPORT */ + +static struct bpf_metadata *bpf_metadata_create(struct bpf_prog_info *info __maybe_unused) +{ + return NULL; +} + +static int synthesize_perf_record_bpf_metadata(const struct bpf_metadata *metadata __maybe_unused, + const struct perf_tool *tool __maybe_unused, + perf_event__handler_t process __maybe_unused, + struct machine *machine __maybe_unused) +{ + return 0; +} + +void bpf_metadata_free(struct bpf_metadata *metadata __maybe_unused) +{ +} + +#endif /* HAVE_LIBBPF_STRINGS_SUPPORT */ + +struct bpf_metadata_final_ctx { + const struct perf_tool *tool; + perf_event__handler_t process; + struct machine *machine; +}; + +static void synthesize_final_bpf_metadata_cb(struct bpf_prog_info_node *node, + void *data) +{ + struct bpf_metadata_final_ctx *ctx = (struct bpf_metadata_final_ctx *)data; + struct bpf_metadata *metadata = node->metadata; + int err; + + if (metadata == NULL) + return; + err = synthesize_perf_record_bpf_metadata(metadata, ctx->tool, + ctx->process, ctx->machine); + if (err != 0) { + const char *prog_name = metadata->prog_names[0]; + + if (prog_name != NULL) + pr_warning("Couldn't synthesize final BPF metadata for %s.\n", prog_name); + else + pr_warning("Couldn't synthesize final BPF metadata.\n"); + } + bpf_metadata_free(metadata); + node->metadata = NULL; +} + +void perf_event__synthesize_final_bpf_metadata(struct perf_session *session, + perf_event__handler_t process) +{ + struct perf_env *env = &session->header.env; + struct bpf_metadata_final_ctx ctx = { + .tool = session->tool, + .process = process, + .machine = &session->machines.host, + }; + + perf_env__iterate_bpf_prog_info(env, synthesize_final_bpf_metadata_cb, + &ctx); +} + /* * Synthesize PERF_RECORD_KSYMBOL and PERF_RECORD_BPF_EVENT for one bpf * program. One PERF_RECORD_BPF_EVENT is generated for the program. And @@ -173,6 +537,7 @@ static int perf_event__synthesize_one_bpf_prog(struct perf_session *session, const struct perf_tool *tool = session->tool; struct bpf_prog_info_node *info_node; struct perf_bpil *info_linear; + struct bpf_metadata *metadata; struct bpf_prog_info *info; struct btf *btf = NULL; struct perf_env *env; @@ -184,7 +549,7 @@ static int perf_event__synthesize_one_bpf_prog(struct perf_session *session, * for perf-record and perf-report use header.env; * otherwise, use global perf_env. */ - env = session->data ? &session->header.env : &perf_env; + env = perf_session__env(session); arrays = 1UL << PERF_BPIL_JITED_KSYMS; arrays |= 1UL << PERF_BPIL_JITED_FUNC_LENS; @@ -193,6 +558,7 @@ static int perf_event__synthesize_one_bpf_prog(struct perf_session *session, arrays |= 1UL << PERF_BPIL_JITED_INSNS; arrays |= 1UL << PERF_BPIL_LINE_INFO; arrays |= 1UL << PERF_BPIL_JITED_LINE_INFO; + arrays |= 1UL << PERF_BPIL_MAP_IDS; info_linear = get_bpf_prog_info_linear(fd, arrays); if (IS_ERR_OR_NULL(info_linear)) { @@ -289,6 +655,7 @@ static int perf_event__synthesize_one_bpf_prog(struct perf_session *session, } info_node->info_linear = info_linear; + info_node->metadata = NULL; if (!perf_env__insert_bpf_prog_info(env, info_node)) { free(info_linear); free(info_node); @@ -301,6 +668,15 @@ static int perf_event__synthesize_one_bpf_prog(struct perf_session *session, */ err = perf_tool__process_synth_event(tool, event, machine, process); + + /* Synthesize PERF_RECORD_BPF_METADATA */ + metadata = bpf_metadata_create(info); + if (metadata != NULL) { + err = synthesize_perf_record_bpf_metadata(metadata, + tool, process, + machine); + bpf_metadata_free(metadata); + } } out: @@ -471,6 +847,7 @@ static void perf_env__add_bpf_info(struct perf_env *env, u32 id) arrays |= 1UL << PERF_BPIL_JITED_INSNS; arrays |= 1UL << PERF_BPIL_LINE_INFO; arrays |= 1UL << PERF_BPIL_JITED_LINE_INFO; + arrays |= 1UL << PERF_BPIL_MAP_IDS; info_linear = get_bpf_prog_info_linear(fd, arrays); if (IS_ERR_OR_NULL(info_linear)) { @@ -483,6 +860,7 @@ static void perf_env__add_bpf_info(struct perf_env *env, u32 id) info_node = malloc(sizeof(struct bpf_prog_info_node)); if (info_node) { info_node->info_linear = info_linear; + info_node->metadata = bpf_metadata_create(&info_linear->info); if (!perf_env__insert_bpf_prog_info(env, info_node)) { free(info_linear); free(info_node); diff --git a/tools/perf/util/bpf-event.h b/tools/perf/util/bpf-event.h index e2f0420905f5..60d2c6637af5 100644 --- a/tools/perf/util/bpf-event.h +++ b/tools/perf/util/bpf-event.h @@ -17,8 +17,15 @@ struct record_opts; struct evlist; struct target; +struct bpf_metadata { + union perf_event *event; + char **prog_names; + __u64 nr_prog_names; +}; + struct bpf_prog_info_node { struct perf_bpil *info_linear; + struct bpf_metadata *metadata; struct rb_node rb_node; }; @@ -36,6 +43,7 @@ int evlist__add_bpf_sb_event(struct evlist *evlist, struct perf_env *env); void __bpf_event__print_bpf_prog_info(struct bpf_prog_info *info, struct perf_env *env, FILE *fp); +void bpf_metadata_free(struct bpf_metadata *metadata); #else static inline int machine__process_bpf(struct machine *machine __maybe_unused, union perf_event *event __maybe_unused, @@ -56,5 +64,10 @@ static inline void __bpf_event__print_bpf_prog_info(struct bpf_prog_info *info _ { } + +static inline void bpf_metadata_free(struct bpf_metadata *metadata __maybe_unused) +{ + +} #endif // HAVE_LIBBPF_SUPPORT #endif diff --git a/tools/perf/util/bpf-filter.c b/tools/perf/util/bpf-filter.c index a4fdf6911ec1..d0e013eeb0f7 100644 --- a/tools/perf/util/bpf-filter.c +++ b/tools/perf/util/bpf-filter.c @@ -52,6 +52,7 @@ #include <internal/xyarray.h> #include <perf/threadmap.h> +#include "util/cap.h" #include "util/debug.h" #include "util/evsel.h" #include "util/target.h" @@ -449,7 +450,7 @@ int perf_bpf_filter__prepare(struct evsel *evsel, struct target *target) struct bpf_program *prog; struct bpf_link *link; struct perf_bpf_filter_entry *entry; - bool needs_idx_hash = !target__has_cpu(target) && !target->uid_str; + bool needs_idx_hash = !target__has_cpu(target); entry = calloc(MAX_FILTERS, sizeof(*entry)); if (entry == NULL) @@ -618,11 +619,38 @@ struct perf_bpf_filter_expr *perf_bpf_filter_expr__new(enum perf_bpf_filter_term return expr; } +static bool check_bpf_filter_capable(void) +{ + bool used_root; + + if (perf_cap__capable(CAP_BPF, &used_root)) + return true; + + if (!used_root) { + /* Check if root already pinned the filter programs and maps */ + int fd = get_pinned_fd("filters"); + + if (fd >= 0) { + close(fd); + return true; + } + } + + pr_err("Error: BPF filter only works for %s!\n" + "\tPlease run 'perf record --setup-filter pin' as root first.\n", + used_root ? "root" : "users with the CAP_BPF capability"); + + return false; +} + int perf_bpf_filter__parse(struct list_head *expr_head, const char *str) { YY_BUFFER_STATE buffer; int ret; + if (!check_bpf_filter_capable()) + return -EPERM; + buffer = perf_bpf_filter__scan_string(str); ret = perf_bpf_filter_parse(expr_head); diff --git a/tools/perf/util/bpf-filter.h b/tools/perf/util/bpf-filter.h index 916ed7770b73..122477f2de44 100644 --- a/tools/perf/util/bpf-filter.h +++ b/tools/perf/util/bpf-filter.h @@ -5,6 +5,7 @@ #include <linux/list.h> #include "bpf_skel/sample-filter.h" +#include "util/debug.h" struct perf_bpf_filter_expr { struct list_head list; @@ -38,6 +39,8 @@ int perf_bpf_filter__unpin(void); static inline int perf_bpf_filter__parse(struct list_head *expr_head __maybe_unused, const char *str __maybe_unused) { + pr_err("Error: BPF filter is requested but perf is not built with BPF.\n" + "\tPlease make sure to build with libbpf and BPF skeleton.\n"); return -EOPNOTSUPP; } static inline int perf_bpf_filter__prepare(struct evsel *evsel __maybe_unused, diff --git a/tools/perf/util/bpf_ftrace.c b/tools/perf/util/bpf_ftrace.c index 7324668cc83e..0cb02412043c 100644 --- a/tools/perf/util/bpf_ftrace.c +++ b/tools/perf/util/bpf_ftrace.c @@ -21,15 +21,26 @@ int perf_ftrace__latency_prepare_bpf(struct perf_ftrace *ftrace) { int fd, err; int i, ncpus = 1, ntasks = 1; - struct filter_entry *func; + struct filter_entry *func = NULL; - if (!list_is_singular(&ftrace->filters)) { - pr_err("ERROR: %s target function(s).\n", - list_empty(&ftrace->filters) ? "No" : "Too many"); - return -1; - } + if (!list_empty(&ftrace->filters)) { + if (!list_is_singular(&ftrace->filters)) { + pr_err("ERROR: Too many target functions.\n"); + return -1; + } + func = list_first_entry(&ftrace->filters, struct filter_entry, list); + } else { + int count = 0; + struct list_head *pos; - func = list_first_entry(&ftrace->filters, struct filter_entry, list); + list_for_each(pos, &ftrace->event_pair) + count++; + + if (count != 2) { + pr_err("ERROR: Needs two target events.\n"); + return -1; + } + } skel = func_latency_bpf__open(); if (!skel) { @@ -93,20 +104,44 @@ int perf_ftrace__latency_prepare_bpf(struct perf_ftrace *ftrace) skel->bss->min = INT64_MAX; - skel->links.func_begin = bpf_program__attach_kprobe(skel->progs.func_begin, - false, func->name); - if (IS_ERR(skel->links.func_begin)) { - pr_err("Failed to attach fentry program\n"); - err = PTR_ERR(skel->links.func_begin); - goto out; - } + if (func) { + skel->links.func_begin = bpf_program__attach_kprobe(skel->progs.func_begin, + false, func->name); + if (IS_ERR(skel->links.func_begin)) { + pr_err("Failed to attach fentry program\n"); + err = PTR_ERR(skel->links.func_begin); + goto out; + } - skel->links.func_end = bpf_program__attach_kprobe(skel->progs.func_end, - true, func->name); - if (IS_ERR(skel->links.func_end)) { - pr_err("Failed to attach fexit program\n"); - err = PTR_ERR(skel->links.func_end); - goto out; + skel->links.func_end = bpf_program__attach_kprobe(skel->progs.func_end, + true, func->name); + if (IS_ERR(skel->links.func_end)) { + pr_err("Failed to attach fexit program\n"); + err = PTR_ERR(skel->links.func_end); + goto out; + } + } else { + struct filter_entry *event; + + event = list_first_entry(&ftrace->event_pair, struct filter_entry, list); + + skel->links.event_begin = bpf_program__attach_raw_tracepoint(skel->progs.event_begin, + event->name); + if (IS_ERR(skel->links.event_begin)) { + pr_err("Failed to attach first tracepoint program\n"); + err = PTR_ERR(skel->links.event_begin); + goto out; + } + + event = list_next_entry(event, list); + + skel->links.event_end = bpf_program__attach_raw_tracepoint(skel->progs.event_end, + event->name); + if (IS_ERR(skel->links.event_end)) { + pr_err("Failed to attach second tracepoint program\n"); + err = PTR_ERR(skel->links.event_end); + goto out; + } } /* XXX: we don't actually use this fd - just for poll() */ diff --git a/tools/perf/util/bpf_skel/augmented_raw_syscalls.bpf.c b/tools/perf/util/bpf_skel/augmented_raw_syscalls.bpf.c index e4352881e3fa..cb86e261b4de 100644 --- a/tools/perf/util/bpf_skel/augmented_raw_syscalls.bpf.c +++ b/tools/perf/util/bpf_skel/augmented_raw_syscalls.bpf.c @@ -7,7 +7,6 @@ */ #include "vmlinux.h" -#include "../trace_augment.h" #include <bpf/bpf_helpers.h> #include <linux/limits.h> @@ -27,6 +26,8 @@ #define MAX_CPUS 4096 +#define TRACE_AUG_MAX_BUF 32 /* for buffer augmentation in perf trace */ + /* bpf-output associated map */ struct __augmented_syscalls__ { __uint(type, BPF_MAP_TYPE_PERF_EVENT_ARRAY); diff --git a/tools/perf/util/bpf_skel/func_latency.bpf.c b/tools/perf/util/bpf_skel/func_latency.bpf.c index e731a79a753a..621e2022c8bc 100644 --- a/tools/perf/util/bpf_skel/func_latency.bpf.c +++ b/tools/perf/util/bpf_skel/func_latency.bpf.c @@ -52,34 +52,89 @@ const volatile unsigned int min_latency; const volatile unsigned int max_latency; const volatile unsigned int bucket_num = NUM_BUCKET; -SEC("kprobe/func") -int BPF_PROG(func_begin) +static bool can_record(void) { - __u64 key, now; - - if (!enabled) - return 0; - - key = bpf_get_current_pid_tgid(); - if (has_cpu) { __u32 cpu = bpf_get_smp_processor_id(); __u8 *ok; ok = bpf_map_lookup_elem(&cpu_filter, &cpu); if (!ok) - return 0; + return false; } if (has_task) { - __u32 pid = key & 0xffffffff; + __u32 pid = bpf_get_current_pid_tgid(); __u8 *ok; ok = bpf_map_lookup_elem(&task_filter, &pid); if (!ok) - return 0; + return false; } + return true; +} + +static void update_latency(__s64 delta) +{ + __u64 val = delta; + __u32 key = 0; + __u64 *hist; + __u64 cmp_base = use_nsec ? 1 : 1000; + + if (delta < 0) + return; + if (bucket_range != 0) { + val = delta / cmp_base; + + if (min_latency > 0) { + if (val > min_latency) + val -= min_latency; + else + goto do_lookup; + } + + // Less than 1 unit (ms or ns), or, in the future, + // than the min latency desired. + if (val > 0) { // 1st entry: [ 1 unit .. bucket_range units ) + key = val / bucket_range + 1; + if (key >= bucket_num) + key = bucket_num - 1; + } + + goto do_lookup; + } + // calculate index using delta + for (key = 0; key < (bucket_num - 1); key++) { + if (delta < (cmp_base << key)) + break; + } + +do_lookup: + hist = bpf_map_lookup_elem(&latency, &key); + if (!hist) + return; + + __sync_fetch_and_add(hist, 1); + + __sync_fetch_and_add(&total, delta); // always in nsec + __sync_fetch_and_add(&count, 1); + + if (delta > max) + max = delta; + if (delta < min) + min = delta; +} + +SEC("kprobe/func") +int BPF_PROG(func_begin) +{ + __u64 key, now; + + if (!enabled || !can_record()) + return 0; + + key = bpf_get_current_pid_tgid(); now = bpf_ktime_get_ns(); // overwrite timestamp for nested functions @@ -92,7 +147,6 @@ int BPF_PROG(func_end) { __u64 tid; __u64 *start; - __u64 cmp_base = use_nsec ? 1 : 1000; if (!enabled) return 0; @@ -101,56 +155,44 @@ int BPF_PROG(func_end) start = bpf_map_lookup_elem(&functime, &tid); if (start) { - __s64 delta = bpf_ktime_get_ns() - *start; - __u64 val = delta; - __u32 key = 0; - __u64 *hist; - + update_latency(bpf_ktime_get_ns() - *start); bpf_map_delete_elem(&functime, &tid); + } - if (delta < 0) - return 0; + return 0; +} - if (bucket_range != 0) { - val = delta / cmp_base; +SEC("raw_tp") +int BPF_PROG(event_begin) +{ + __u64 key, now; - if (min_latency > 0) { - if (val > min_latency) - val -= min_latency; - else - goto do_lookup; - } + if (!enabled || !can_record()) + return 0; - // Less than 1 unit (ms or ns), or, in the future, - // than the min latency desired. - if (val > 0) { // 1st entry: [ 1 unit .. bucket_range units ) - key = val / bucket_range + 1; - if (key >= bucket_num) - key = bucket_num - 1; - } + key = bpf_get_current_pid_tgid(); + now = bpf_ktime_get_ns(); - goto do_lookup; - } - // calculate index using delta - for (key = 0; key < (bucket_num - 1); key++) { - if (delta < (cmp_base << key)) - break; - } + // overwrite timestamp for nested events + bpf_map_update_elem(&functime, &key, &now, BPF_ANY); + return 0; +} -do_lookup: - hist = bpf_map_lookup_elem(&latency, &key); - if (!hist) - return 0; +SEC("raw_tp") +int BPF_PROG(event_end) +{ + __u64 tid; + __u64 *start; - __sync_fetch_and_add(hist, 1); + if (!enabled) + return 0; - __sync_fetch_and_add(&total, delta); // always in nsec - __sync_fetch_and_add(&count, 1); + tid = bpf_get_current_pid_tgid(); - if (delta > max) - max = delta; - if (delta < min) - min = delta; + start = bpf_map_lookup_elem(&functime, &tid); + if (start) { + update_latency(bpf_ktime_get_ns() - *start); + bpf_map_delete_elem(&functime, &tid); } return 0; diff --git a/tools/perf/util/bpf_skel/perf_version.h b/tools/perf/util/bpf_skel/perf_version.h new file mode 100644 index 000000000000..1ed5b2e59bf5 --- /dev/null +++ b/tools/perf/util/bpf_skel/perf_version.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) */ + +#ifndef __PERF_VERSION_H__ +#define __PERF_VERSION_H__ + +#include "vmlinux.h" +#include <bpf/bpf_helpers.h> + +/* + * This is used by tests/shell/record_bpf_metadata.sh + * to verify that BPF metadata generation works. + * + * PERF_VERSION is defined by a build rule at compile time. + */ +const char bpf_metadata_perf_version[] SEC(".rodata") = PERF_VERSION; + +#endif /* __PERF_VERSION_H__ */ diff --git a/tools/perf/util/bpf_trace_augment.c b/tools/perf/util/bpf_trace_augment.c new file mode 100644 index 000000000000..56ed17534caa --- /dev/null +++ b/tools/perf/util/bpf_trace_augment.c @@ -0,0 +1,143 @@ +#include <bpf/libbpf.h> +#include <internal/xyarray.h> + +#include "util/debug.h" +#include "util/evlist.h" +#include "util/trace_augment.h" + +#include "bpf_skel/augmented_raw_syscalls.skel.h" + +static struct augmented_raw_syscalls_bpf *skel; +static struct evsel *bpf_output; + +int augmented_syscalls__prepare(void) +{ + struct bpf_program *prog; + char buf[128]; + int err; + + skel = augmented_raw_syscalls_bpf__open(); + if (!skel) { + pr_debug("Failed to open augmented syscalls BPF skeleton\n"); + return -errno; + } + + /* + * Disable attaching the BPF programs except for sys_enter and + * sys_exit that tail call into this as necessary. + */ + bpf_object__for_each_program(prog, skel->obj) { + if (prog != skel->progs.sys_enter && prog != skel->progs.sys_exit) + bpf_program__set_autoattach(prog, /*autoattach=*/false); + } + + err = augmented_raw_syscalls_bpf__load(skel); + if (err < 0) { + libbpf_strerror(err, buf, sizeof(buf)); + pr_debug("Failed to load augmented syscalls BPF skeleton: %s\n", buf); + return err; + } + + augmented_raw_syscalls_bpf__attach(skel); + return 0; +} + +int augmented_syscalls__create_bpf_output(struct evlist *evlist) +{ + int err = parse_event(evlist, "bpf-output/no-inherit=1,name=__augmented_syscalls__/"); + + if (err) { + pr_err("ERROR: Setup BPF output event failed: %d\n", err); + return err; + } + + bpf_output = evlist__last(evlist); + assert(evsel__name_is(bpf_output, "__augmented_syscalls__")); + + return 0; +} + +void augmented_syscalls__setup_bpf_output(void) +{ + struct perf_cpu cpu; + int i; + + if (bpf_output == NULL) + return; + + /* + * Set up the __augmented_syscalls__ BPF map to hold for each + * CPU the bpf-output event's file descriptor. + */ + perf_cpu_map__for_each_cpu(cpu, i, bpf_output->core.cpus) { + int mycpu = cpu.cpu; + + bpf_map__update_elem(skel->maps.__augmented_syscalls__, + &mycpu, sizeof(mycpu), + xyarray__entry(bpf_output->core.fd, + mycpu, 0), + sizeof(__u32), BPF_ANY); + } +} + +int augmented_syscalls__set_filter_pids(unsigned int nr, pid_t *pids) +{ + bool value = true; + int err = 0; + + if (skel == NULL) + return 0; + + for (size_t i = 0; i < nr; ++i) { + err = bpf_map__update_elem(skel->maps.pids_filtered, &pids[i], + sizeof(*pids), &value, sizeof(value), + BPF_ANY); + if (err) + break; + } + return err; +} + +int augmented_syscalls__get_map_fds(int *enter_fd, int *exit_fd, int *beauty_fd) +{ + if (skel == NULL) + return -1; + + *enter_fd = bpf_map__fd(skel->maps.syscalls_sys_enter); + *exit_fd = bpf_map__fd(skel->maps.syscalls_sys_exit); + *beauty_fd = bpf_map__fd(skel->maps.beauty_map_enter); + + if (*enter_fd < 0 || *exit_fd < 0 || *beauty_fd < 0) { + pr_err("Error: failed to get syscall or beauty map fd\n"); + return -1; + } + + return 0; +} + +struct bpf_program *augmented_syscalls__unaugmented(void) +{ + return skel->progs.syscall_unaugmented; +} + +struct bpf_program *augmented_syscalls__find_by_title(const char *name) +{ + struct bpf_program *pos; + const char *sec_name; + + if (skel->obj == NULL) + return NULL; + + bpf_object__for_each_program(pos, skel->obj) { + sec_name = bpf_program__section_name(pos); + if (sec_name && !strcmp(sec_name, name)) + return pos; + } + + return NULL; +} + +void augmented_syscalls__cleanup(void) +{ + augmented_raw_syscalls_bpf__destroy(skel); +} diff --git a/tools/perf/util/branch.c b/tools/perf/util/branch.c index ab760e267d41..3712be067464 100644 --- a/tools/perf/util/branch.c +++ b/tools/perf/util/branch.c @@ -46,7 +46,7 @@ const char *branch_new_type_name(int new_type) "FAULT_DATA", "FAULT_INST", /* - * TODO: This switch should happen on 'session->header.env.arch' + * TODO: This switch should happen on 'perf_session__env(session)->arch' * instead, because an arm64 platform perf recording could be * opened for analysis on other platforms as well. */ diff --git a/tools/perf/util/build-id.c b/tools/perf/util/build-id.c index e763e8d99a43..a7018a3b0437 100644 --- a/tools/perf/util/build-id.c +++ b/tools/perf/util/build-id.c @@ -42,10 +42,20 @@ static bool no_buildid_cache; +static int mark_dso_hit_callback(struct callchain_cursor_node *node, void *data __maybe_unused) +{ + struct map *map = node->ms.map; + + if (map) + dso__set_hit(map__dso(map)); + + return 0; +} + int build_id__mark_dso_hit(const struct perf_tool *tool __maybe_unused, union perf_event *event, struct perf_sample *sample, - struct evsel *evsel __maybe_unused, + struct evsel *evsel, struct machine *machine) { struct addr_location al; @@ -63,31 +73,29 @@ int build_id__mark_dso_hit(const struct perf_tool *tool __maybe_unused, dso__set_hit(map__dso(al.map)); addr_location__exit(&al); + + sample__for_each_callchain_node(thread, evsel, sample, PERF_MAX_STACK_DEPTH, + /*symbols=*/false, mark_dso_hit_callback, /*data=*/NULL); + + thread__put(thread); return 0; } -int build_id__sprintf(const struct build_id *build_id, char *bf) +int build_id__snprintf(const struct build_id *build_id, char *bf, size_t bf_size) { - char *bid = bf; - const u8 *raw = build_id->data; - size_t i; + size_t offs = 0; - bf[0] = 0x0; + for (size_t i = 0; i < build_id->size && offs < bf_size; ++i) + offs += snprintf(bf + offs, bf_size - offs, "%02x", build_id->data[i]); - for (i = 0; i < build_id->size; ++i) { - sprintf(bid, "%02x", *raw); - ++raw; - bid += 2; - } - - return (bid - bf) + 1; + return offs; } -int sysfs__sprintf_build_id(const char *root_dir, char *sbuild_id) +int sysfs__snprintf_build_id(const char *root_dir, char *sbuild_id, size_t sbuild_id_size) { char notes[PATH_MAX]; - struct build_id bid; + struct build_id bid = { .size = 0, }; int ret; if (!root_dir) @@ -99,19 +107,19 @@ int sysfs__sprintf_build_id(const char *root_dir, char *sbuild_id) if (ret < 0) return ret; - return build_id__sprintf(&bid, sbuild_id); + return build_id__snprintf(&bid, sbuild_id, sbuild_id_size); } -int filename__sprintf_build_id(const char *pathname, char *sbuild_id) +int filename__snprintf_build_id(const char *pathname, char *sbuild_id, size_t sbuild_id_size) { - struct build_id bid; + struct build_id bid = { .size = 0, }; int ret; ret = filename__read_build_id(pathname, &bid); if (ret < 0) return ret; - return build_id__sprintf(&bid, sbuild_id); + return build_id__snprintf(&bid, sbuild_id, sbuild_id_size); } /* asnprintf consolidates asprintf and snprintf */ @@ -212,9 +220,9 @@ static bool build_id_cache__valid_id(char *sbuild_id) return false; if (!strcmp(pathname, DSO__NAME_KALLSYMS)) - ret = sysfs__sprintf_build_id("/", real_sbuild_id); + ret = sysfs__snprintf_build_id("/", real_sbuild_id, sizeof(real_sbuild_id)); else if (pathname[0] == '/') - ret = filename__sprintf_build_id(pathname, real_sbuild_id); + ret = filename__snprintf_build_id(pathname, real_sbuild_id, sizeof(real_sbuild_id)); else ret = -EINVAL; /* Should we support other special DSO cache? */ if (ret >= 0) @@ -243,7 +251,7 @@ char *__dso__build_id_filename(const struct dso *dso, char *bf, size_t size, if (!dso__has_build_id(dso)) return NULL; - build_id__sprintf(dso__bid_const(dso), sbuild_id); + build_id__snprintf(dso__bid(dso), sbuild_id, sizeof(sbuild_id)); linkname = build_id_cache__linkname(sbuild_id, NULL, 0); if (!linkname) return NULL; @@ -326,7 +334,7 @@ static int machine__write_buildid_table_cb(struct dso *dso, void *data) } in_kernel = dso__kernel(dso) || is_kernel_module(name, PERF_RECORD_MISC_CPUMODE_UNKNOWN); - return write_buildid(name, name_len, dso__bid(dso), args->machine->pid, + return write_buildid(name, name_len, &dso__id(dso)->build_id, args->machine->pid, in_kernel ? args->kmisc : args->umisc, args->fd); } @@ -769,7 +777,7 @@ static int build_id_cache__add_b(const struct build_id *bid, { char sbuild_id[SBUILD_ID_SIZE]; - build_id__sprintf(bid, sbuild_id); + build_id__snprintf(bid, sbuild_id, sizeof(sbuild_id)); return __build_id_cache__add_s(sbuild_id, name, nsi, is_kallsyms, is_vdso, proper_name, root_dir); @@ -841,7 +849,7 @@ static int filename__read_build_id_ns(const char *filename, static bool dso__build_id_mismatch(struct dso *dso, const char *name) { - struct build_id bid; + struct build_id bid = { .size = 0, }; bool ret = false; mutex_lock(dso__lock(dso)); @@ -864,7 +872,7 @@ static int dso__cache_build_id(struct dso *dso, struct machine *machine, char *allocated_name = NULL; int ret = 0; - if (!dso__has_build_id(dso)) + if (!dso__has_build_id(dso) || !dso__hit(dso)) return 0; if (dso__is_kcore(dso)) { @@ -951,7 +959,10 @@ bool perf_session__read_build_ids(struct perf_session *session, bool with_hits) void build_id__init(struct build_id *bid, const u8 *data, size_t size) { - WARN_ON(size > BUILD_ID_SIZE); + if (size > BUILD_ID_SIZE) { + pr_debug("Truncating build_id size from %zd\n", size); + size = BUILD_ID_SIZE; + } memcpy(bid->data, data, size); bid->size = size; } diff --git a/tools/perf/util/build-id.h b/tools/perf/util/build-id.h index a212497bfdb0..47e621cebe1b 100644 --- a/tools/perf/util/build-id.h +++ b/tools/perf/util/build-id.h @@ -13,7 +13,7 @@ struct build_id { u8 data[BUILD_ID_SIZE]; - size_t size; + u8 size; }; struct dso; @@ -21,10 +21,10 @@ struct feat_fd; struct nsinfo; void build_id__init(struct build_id *bid, const u8 *data, size_t size); -int build_id__sprintf(const struct build_id *build_id, char *bf); +int build_id__snprintf(const struct build_id *build_id, char *bf, size_t bf_size); bool build_id__is_defined(const struct build_id *bid); -int sysfs__sprintf_build_id(const char *root_dir, char *sbuild_id); -int filename__sprintf_build_id(const char *pathname, char *sbuild_id); +int sysfs__snprintf_build_id(const char *root_dir, char *sbuild_id, size_t sbuild_id_size); +int filename__snprintf_build_id(const char *pathname, char *sbuild_id, size_t sbuild_id_size); char *build_id_cache__kallsyms_path(const char *sbuild_id, char *bf, size_t size); diff --git a/tools/perf/util/cap.c b/tools/perf/util/cap.c index 69d9a2bcd40b..24a0ea7e6d97 100644 --- a/tools/perf/util/cap.c +++ b/tools/perf/util/cap.c @@ -7,7 +7,6 @@ #include "debug.h" #include <errno.h> #include <string.h> -#include <linux/capability.h> #include <sys/syscall.h> #include <unistd.h> diff --git a/tools/perf/util/cap.h b/tools/perf/util/cap.h index 0c6a1ff55f07..c1b8ac033ccc 100644 --- a/tools/perf/util/cap.h +++ b/tools/perf/util/cap.h @@ -3,6 +3,7 @@ #define __PERF_CAP_H #include <stdbool.h> +#include <linux/capability.h> /* For older systems */ #ifndef CAP_SYSLOG @@ -13,6 +14,10 @@ #define CAP_PERFMON 38 #endif +#ifndef CAP_BPF +#define CAP_BPF 39 +#endif + /* Query if a capability is supported, used_root is set if the fallback root check was used. */ bool perf_cap__capable(int cap, bool *used_root); diff --git a/tools/perf/util/cgroup.c b/tools/perf/util/cgroup.c index fbcc0626f9ce..25e2769b5e74 100644 --- a/tools/perf/util/cgroup.c +++ b/tools/perf/util/cgroup.c @@ -413,8 +413,7 @@ static bool has_pattern_string(const char *str) return !!strpbrk(str, "{}[]()|*+?^$"); } -int evlist__expand_cgroup(struct evlist *evlist, const char *str, - struct rblist *metric_events, bool open_cgroup) +int evlist__expand_cgroup(struct evlist *evlist, const char *str, bool open_cgroup) { struct evlist *orig_list, *tmp_list; struct evsel *pos, *evsel, *leader; @@ -440,12 +439,8 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, evlist__splice_list_tail(orig_list, &evlist->core.entries); evlist->core.nr_entries = 0; - if (metric_events) { - orig_metric_events = *metric_events; - rblist__init(metric_events); - } else { - rblist__init(&orig_metric_events); - } + orig_metric_events = evlist->metric_events; + metricgroup__rblist_init(&evlist->metric_events); if (has_pattern_string(str)) prefix_len = match_cgroups(str); @@ -490,12 +485,10 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, cgroup__put(cgrp); nr_cgroups++; - if (metric_events) { - if (metricgroup__copy_metric_events(tmp_list, cgrp, - metric_events, - &orig_metric_events) < 0) - goto out_err; - } + if (metricgroup__copy_metric_events(tmp_list, cgrp, + &evlist->metric_events, + &orig_metric_events) < 0) + goto out_err; evlist__splice_list_tail(evlist, &tmp_list->core.entries); tmp_list->core.nr_entries = 0; @@ -512,7 +505,7 @@ int evlist__expand_cgroup(struct evlist *evlist, const char *str, out_err: evlist__delete(orig_list); evlist__delete(tmp_list); - rblist__exit(&orig_metric_events); + metricgroup__rblist_exit(&orig_metric_events); release_cgroup_list(); return ret; diff --git a/tools/perf/util/cgroup.h b/tools/perf/util/cgroup.h index de8882d6e8d3..7b1bda22878c 100644 --- a/tools/perf/util/cgroup.h +++ b/tools/perf/util/cgroup.h @@ -28,8 +28,7 @@ struct rblist; struct cgroup *cgroup__new(const char *name, bool do_open); struct cgroup *evlist__findnew_cgroup(struct evlist *evlist, const char *name); -int evlist__expand_cgroup(struct evlist *evlist, const char *cgroups, - struct rblist *metric_events, bool open_cgroup); +int evlist__expand_cgroup(struct evlist *evlist, const char *cgroups, bool open_cgroup); void evlist__set_default_cgroup(struct evlist *evlist, struct cgroup *cgroup); diff --git a/tools/perf/util/comm.c b/tools/perf/util/comm.c index 8aa456d7c2cd..9880247a2c33 100644 --- a/tools/perf/util/comm.c +++ b/tools/perf/util/comm.c @@ -24,6 +24,7 @@ static struct comm_strs { static void comm_strs__remove_if_last(struct comm_str *cs); static void comm_strs__init(void) + NO_THREAD_SAFETY_ANALYSIS /* Inherently single threaded due to pthread_once. */ { init_rwsem(&_comm_strs.lock); _comm_strs.capacity = 16; @@ -119,6 +120,7 @@ static void comm_strs__remove_if_last(struct comm_str *cs) } static struct comm_str *__comm_strs__find(struct comm_strs *comm_strs, const char *str) + SHARED_LOCKS_REQUIRED(comm_strs->lock) { struct comm_str **result; diff --git a/tools/perf/util/data-convert-bt.c b/tools/perf/util/data-convert-bt.c index 5e7ff09fbc95..3d2e437e1354 100644 --- a/tools/perf/util/data-convert-bt.c +++ b/tools/perf/util/data-convert-bt.c @@ -1338,14 +1338,14 @@ static void cleanup_events(struct perf_session *session) static int setup_streams(struct ctf_writer *cw, struct perf_session *session) { struct ctf_stream **stream; - struct perf_header *ph = &session->header; + struct perf_env *env = perf_session__env(session); int ncpus; /* * Try to get the number of cpus used in the data file, * if not present fallback to the MAX_CPUS. */ - ncpus = ph->env.nr_cpus_avail ?: MAX_CPUS; + ncpus = env->nr_cpus_avail ?: MAX_CPUS; stream = zalloc(sizeof(*stream) * ncpus); if (!stream) { @@ -1371,7 +1371,7 @@ static void free_streams(struct ctf_writer *cw) static int ctf_writer__setup_env(struct ctf_writer *cw, struct perf_session *session) { - struct perf_header *header = &session->header; + struct perf_env *env = perf_session__env(session); struct bt_ctf_writer *writer = cw->writer; #define ADD(__n, __v) \ @@ -1380,11 +1380,11 @@ do { \ return -1; \ } while (0) - ADD("host", header->env.hostname); + ADD("host", env->hostname); ADD("sysname", "Linux"); - ADD("release", header->env.os_release); - ADD("version", header->env.version); - ADD("machine", header->env.arch); + ADD("release", env->os_release); + ADD("version", env->version); + ADD("machine", env->arch); ADD("domain", "kernel"); ADD("tracer_name", "perf"); @@ -1401,7 +1401,7 @@ static int ctf_writer__setup_clock(struct ctf_writer *cw, int64_t offset = 0; if (tod) { - struct perf_env *env = &session->header.env; + struct perf_env *env = perf_session__env(session); if (!env->clock.enabled) { pr_err("Can't provide --tod time, missing clock data. " diff --git a/tools/perf/util/data-convert-json.c b/tools/perf/util/data-convert-json.c index d9f805bf6fb0..9dc1e184cf3c 100644 --- a/tools/perf/util/data-convert-json.c +++ b/tools/perf/util/data-convert-json.c @@ -257,7 +257,8 @@ static int process_sample_event(const struct perf_tool *tool, static void output_headers(struct perf_session *session, struct convert_json *c) { struct stat st; - struct perf_header *header = &session->header; + const struct perf_header *header = &session->header; + const struct perf_env *env = perf_session__env(session); int ret; int fd = perf_data__fd(session->data); int i; @@ -280,32 +281,32 @@ static void output_headers(struct perf_session *session, struct convert_json *c) output_json_key_format(out, true, 2, "data-size", "%" PRIu64, header->data_size); output_json_key_format(out, true, 2, "feat-offset", "%" PRIu64, header->feat_offset); - output_json_key_string(out, true, 2, "hostname", header->env.hostname); - output_json_key_string(out, true, 2, "os-release", header->env.os_release); - output_json_key_string(out, true, 2, "arch", header->env.arch); + output_json_key_string(out, true, 2, "hostname", env->hostname); + output_json_key_string(out, true, 2, "os-release", env->os_release); + output_json_key_string(out, true, 2, "arch", env->arch); - if (header->env.cpu_desc) - output_json_key_string(out, true, 2, "cpu-desc", header->env.cpu_desc); + if (env->cpu_desc) + output_json_key_string(out, true, 2, "cpu-desc", env->cpu_desc); - output_json_key_string(out, true, 2, "cpuid", header->env.cpuid); - output_json_key_format(out, true, 2, "nrcpus-online", "%u", header->env.nr_cpus_online); - output_json_key_format(out, true, 2, "nrcpus-avail", "%u", header->env.nr_cpus_avail); + output_json_key_string(out, true, 2, "cpuid", env->cpuid); + output_json_key_format(out, true, 2, "nrcpus-online", "%u", env->nr_cpus_online); + output_json_key_format(out, true, 2, "nrcpus-avail", "%u", env->nr_cpus_avail); - if (header->env.clock.enabled) { + if (env->clock.enabled) { output_json_key_format(out, true, 2, "clockid", - "%u", header->env.clock.clockid); + "%u", env->clock.clockid); output_json_key_format(out, true, 2, "clock-time", - "%" PRIu64, header->env.clock.clockid_ns); + "%" PRIu64, env->clock.clockid_ns); output_json_key_format(out, true, 2, "real-time", - "%" PRIu64, header->env.clock.tod_ns); + "%" PRIu64, env->clock.tod_ns); } - output_json_key_string(out, true, 2, "perf-version", header->env.version); + output_json_key_string(out, true, 2, "perf-version", env->version); output_json_key_format(out, true, 2, "cmdline", "["); - for (i = 0; i < header->env.nr_cmdline; i++) { + for (i = 0; i < env->nr_cmdline; i++) { output_json_delimiters(out, i != 0, 3); - output_json_string(c->out, header->env.cmdline_argv[i]); + output_json_string(c->out, env->cmdline_argv[i]); } output_json_format(out, false, 2, "]"); } @@ -376,8 +377,7 @@ int bt_convert__perf2json(const char *input_name, const char *output_name, fprintf(stderr, "Error creating perf session!\n"); goto err_fclose; } - - if (symbol__init(&session->header.env) < 0) { + if (symbol__init(perf_session__env(session)) < 0) { fprintf(stderr, "Symbol init error!\n"); goto err_session_delete; } diff --git a/tools/perf/util/db-export.c b/tools/perf/util/db-export.c index 50f916374d87..8f52e8cefcf3 100644 --- a/tools/perf/util/db-export.c +++ b/tools/perf/util/db-export.c @@ -181,7 +181,7 @@ static int db_ids_from_al(struct db_export *dbe, struct addr_location *al, if (al->map) { struct dso *dso = map__dso(al->map); - err = db_export__dso(dbe, dso, maps__machine(al->maps)); + err = db_export__dso(dbe, dso, maps__machine(thread__maps(al->thread))); if (err) return err; *dso_db_id = dso__db_id(dso); @@ -256,6 +256,7 @@ static struct call_path *call_path_from_sample(struct db_export *dbe, al.map = map__get(node->ms.map); al.maps = maps__get(thread__maps(thread)); al.addr = node->ip; + al.thread = thread__get(thread); if (al.map && !al.sym) al.sym = dso__find_symbol(map__dso(al.map), al.addr); @@ -358,14 +359,18 @@ int db_export__sample(struct db_export *dbe, union perf_event *event, }; struct thread *main_thread; struct comm *comm = NULL; - struct machine *machine; + struct machine *machine = NULL; int err; + if (thread__maps(thread)) + machine = maps__machine(thread__maps(thread)); + if (!machine) + return -1; + err = db_export__evsel(dbe, evsel); if (err) return err; - machine = maps__machine(al->maps); err = db_export__machine(dbe, machine); if (err) return err; diff --git a/tools/perf/util/debug.c b/tools/perf/util/debug.c index f9ef7d045c92..1dfa4d0eec4d 100644 --- a/tools/perf/util/debug.c +++ b/tools/perf/util/debug.c @@ -14,11 +14,19 @@ #ifdef HAVE_BACKTRACE_SUPPORT #include <execinfo.h> #endif +#include "addr_location.h" #include "color.h" -#include "event.h" #include "debug.h" +#include "env.h" +#include "event.h" +#include "machine.h" +#include "map.h" #include "print_binary.h" +#include "srcline.h" +#include "symbol.h" +#include "synthetic-events.h" #include "target.h" +#include "thread.h" #include "trace-event.h" #include "ui/helpline.h" #include "ui/ui.h" @@ -298,21 +306,66 @@ void perf_debug_setup(void) libapi_set_print(pr_warning_wrapper, pr_warning_wrapper, pr_debug_wrapper); } +void __dump_stack(FILE *file, void **stackdump, size_t stackdump_size) +{ + /* TODO: async safety. printf, malloc, etc. aren't safe inside a signal handler. */ + pid_t pid = getpid(); + struct machine *machine; + struct thread *thread = NULL; + struct perf_env host_env; + + perf_env__init(&host_env); + machine = machine__new_live(&host_env, /*kernel_maps=*/false, pid); + + if (machine) + thread = machine__find_thread(machine, pid, pid); + +#ifdef HAVE_BACKTRACE_SUPPORT + if (!machine || !thread) { + /* + * Backtrace functions are async signal safe. Fall back on them + * if machine/thread creation fails. + */ + backtrace_symbols_fd(stackdump, stackdump_size, fileno(file)); + machine__delete(machine); + perf_env__exit(&host_env); + return; + } +#endif + + for (size_t i = 0; i < stackdump_size; i++) { + struct addr_location al; + u64 addr = (u64)(uintptr_t)stackdump[i]; + bool printed = false; + + addr_location__init(&al); + if (thread && thread__find_map(thread, PERF_RECORD_MISC_USER, addr, &al)) { + al.sym = map__find_symbol(al.map, al.addr); + if (al.sym) { + fprintf(file, " #%zd %p in %s ", i, stackdump[i], al.sym->name); + printed = true; + } + } + if (!printed) + fprintf(file, " #%zd %p ", i, stackdump[i]); + + map__fprintf_srcline(al.map, al.addr, "", file); + fprintf(file, "\n"); + addr_location__exit(&al); + } + thread__put(thread); + machine__delete(machine); + perf_env__exit(&host_env); +} + /* Obtain a backtrace and print it to stdout. */ #ifdef HAVE_BACKTRACE_SUPPORT void dump_stack(void) { - void *array[16]; - size_t size = backtrace(array, ARRAY_SIZE(array)); - char **strings = backtrace_symbols(array, size); - size_t i; - - printf("Obtained %zd stack frames.\n", size); - - for (i = 0; i < size; i++) - printf("%s\n", strings[i]); + void *stackdump[32]; + size_t size = backtrace(stackdump, ARRAY_SIZE(stackdump)); - free(strings); + __dump_stack(stdout, stackdump, size); } #else void dump_stack(void) {} diff --git a/tools/perf/util/debug.h b/tools/perf/util/debug.h index a4026d1fd6a3..6b737e195ce1 100644 --- a/tools/perf/util/debug.h +++ b/tools/perf/util/debug.h @@ -85,6 +85,7 @@ void debug_set_display_time(bool set); void perf_debug_setup(void); int perf_quiet_option(void); +void __dump_stack(FILE *file, void **stackdump, size_t stackdump_size); void dump_stack(void); void sighandler_dump_stack(int sig); diff --git a/tools/perf/util/debuginfo.c b/tools/perf/util/debuginfo.c index b5deea7cbdf2..a44c70f93156 100644 --- a/tools/perf/util/debuginfo.c +++ b/tools/perf/util/debuginfo.c @@ -103,7 +103,7 @@ struct debuginfo *debuginfo__new(const char *path) char buf[PATH_MAX], nil = '\0'; struct dso *dso; struct debuginfo *dinfo = NULL; - struct build_id bid; + struct build_id bid = { .size = 0}; /* Try to open distro debuginfo files */ dso = dso__new(path); diff --git a/tools/perf/util/disasm.c b/tools/perf/util/disasm.c index 8f0eb56c6fc6..b1e4919d016f 100644 --- a/tools/perf/util/disasm.c +++ b/tools/perf/util/disasm.c @@ -1218,7 +1218,7 @@ int symbol__strerror_disassemble(struct map_symbol *ms, int errnum, char *buf, s char *build_id_msg = NULL; if (dso__has_build_id(dso)) { - build_id__sprintf(dso__bid(dso), bf + 15); + build_id__snprintf(dso__bid(dso), bf + 15, sizeof(bf) - 15); build_id_msg = bf; } scnprintf(buf, buflen, @@ -2284,6 +2284,13 @@ int symbol__disassemble(struct symbol *sym, struct annotate_args *args) } } + /* FIXME: LLVM and CAPSTONE should support source code */ + if (options->annotate_src && !options->hide_src_code) { + err = symbol__disassemble_objdump(symfs_filename, sym, args); + if (err == 0) + goto out_remove_tmp; + } + err = -1; for (u8 i = 0; i < ARRAY_SIZE(options->disassemblers) && err != 0; i++) { enum perf_disassembler dis = options->disassemblers[i]; diff --git a/tools/perf/util/dlfilter.c b/tools/perf/util/dlfilter.c index ddacef881af2..c0afcbd954f8 100644 --- a/tools/perf/util/dlfilter.c +++ b/tools/perf/util/dlfilter.c @@ -513,6 +513,7 @@ int dlfilter__do_filter_event(struct dlfilter *d, d->d_addr_al = &d_addr_al; d_sample.size = sizeof(d_sample); + d_sample.p_stage_cyc = sample->weight3; d_ip_al.size = 0; /* To indicate d_ip_al is not initialized */ d_addr_al.size = 0; /* To indicate d_addr_al is not initialized */ @@ -526,7 +527,6 @@ int dlfilter__do_filter_event(struct dlfilter *d, ASSIGN(period); ASSIGN(weight); ASSIGN(ins_lat); - ASSIGN(p_stage_cyc); ASSIGN(transaction); ASSIGN(insn_cnt); ASSIGN(cyc_cnt); diff --git a/tools/perf/util/drm_pmu.c b/tools/perf/util/drm_pmu.c new file mode 100644 index 000000000000..988890f37ba7 --- /dev/null +++ b/tools/perf/util/drm_pmu.c @@ -0,0 +1,686 @@ +// SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) +#include "drm_pmu.h" +#include "counts.h" +#include "cpumap.h" +#include "debug.h" +#include "evsel.h" +#include "pmu.h" +#include <perf/threadmap.h> +#include <api/fs/fs.h> +#include <api/io.h> +#include <ctype.h> +#include <dirent.h> +#include <fcntl.h> +#include <unistd.h> +#include <linux/unistd.h> +#include <linux/kcmp.h> +#include <linux/zalloc.h> +#include <sys/stat.h> +#include <sys/syscall.h> +#include <sys/sysmacros.h> +#include <sys/types.h> + +enum drm_pmu_unit { + DRM_PMU_UNIT_BYTES, + DRM_PMU_UNIT_CAPACITY, + DRM_PMU_UNIT_CYCLES, + DRM_PMU_UNIT_HZ, + DRM_PMU_UNIT_NS, + + DRM_PMU_UNIT_MAX, +}; + +struct drm_pmu_event { + const char *name; + const char *desc; + enum drm_pmu_unit unit; +}; + +struct drm_pmu { + struct perf_pmu pmu; + struct drm_pmu_event *events; + int num_events; +}; + +static const char * const drm_pmu_unit_strs[DRM_PMU_UNIT_MAX] = { + "bytes", + "capacity", + "cycles", + "hz", + "ns", +}; + +static const char * const drm_pmu_scale_unit_strs[DRM_PMU_UNIT_MAX] = { + "1bytes", + "1capacity", + "1cycles", + "1hz", + "1ns", +}; + +bool perf_pmu__is_drm(const struct perf_pmu *pmu) +{ + return pmu && pmu->type >= PERF_PMU_TYPE_DRM_START && + pmu->type <= PERF_PMU_TYPE_DRM_END; +} + +bool evsel__is_drm(const struct evsel *evsel) +{ + return perf_pmu__is_drm(evsel->pmu); +} + +static struct drm_pmu *add_drm_pmu(struct list_head *pmus, char *line, size_t line_len) +{ + struct drm_pmu *drm; + struct perf_pmu *pmu; + const char *name; + __u32 max_drm_pmu_type = 0, type; + int i = 12; + + if (line[line_len - 1] == '\n') + line[line_len - 1] = '\0'; + while (isspace(line[i])) + i++; + + line[--i] = '_'; + line[--i] = 'm'; + line[--i] = 'r'; + line[--i] = 'd'; + name = &line[i]; + + list_for_each_entry(pmu, pmus, list) { + if (!perf_pmu__is_drm(pmu)) + continue; + if (pmu->type > max_drm_pmu_type) + max_drm_pmu_type = pmu->type; + if (!strcmp(pmu->name, name)) { + /* PMU already exists. */ + return NULL; + } + } + + if (max_drm_pmu_type != 0) + type = max_drm_pmu_type + 1; + else + type = PERF_PMU_TYPE_DRM_START; + + if (type > PERF_PMU_TYPE_DRM_END) { + zfree(&drm); + pr_err("Unable to encode DRM PMU type for %s\n", name); + return NULL; + } + + drm = zalloc(sizeof(*drm)); + if (!drm) + return NULL; + + if (perf_pmu__init(&drm->pmu, type, name) != 0) { + perf_pmu__delete(&drm->pmu); + return NULL; + } + + drm->pmu.cpus = perf_cpu_map__new("0"); + if (!drm->pmu.cpus) { + perf_pmu__delete(&drm->pmu); + return NULL; + } + return drm; +} + + +static bool starts_with(const char *str, const char *prefix) +{ + return !strncmp(prefix, str, strlen(prefix)); +} + +static int add_event(struct drm_pmu_event **events, int *num_events, + const char *line, enum drm_pmu_unit unit, const char *desc) +{ + const char *colon = strchr(line, ':'); + struct drm_pmu_event *tmp; + + if (!colon) + return -EINVAL; + + tmp = reallocarray(*events, *num_events + 1, sizeof(struct drm_pmu_event)); + if (!tmp) + return -ENOMEM; + tmp[*num_events].unit = unit; + tmp[*num_events].desc = desc; + tmp[*num_events].name = strndup(line, colon - line); + if (!tmp[*num_events].name) + return -ENOMEM; + (*num_events)++; + *events = tmp; + return 0; +} + +static int read_drm_pmus_cb(void *args, int fdinfo_dir_fd, const char *fd_name) +{ + struct list_head *pmus = args; + char buf[640]; + struct io io; + char *line = NULL; + size_t line_len; + struct drm_pmu *drm = NULL; + struct drm_pmu_event *events = NULL; + int num_events = 0; + + io__init(&io, openat(fdinfo_dir_fd, fd_name, O_RDONLY), buf, sizeof(buf)); + if (io.fd == -1) { + /* Failed to open file, ignore. */ + return 0; + } + + while (io__getline(&io, &line, &line_len) > 0) { + if (starts_with(line, "drm-driver:")) { + drm = add_drm_pmu(pmus, line, line_len); + if (!drm) + break; + continue; + } + /* + * Note the string matching below is alphabetical, with more + * specific matches appearing before less specific. + */ + if (starts_with(line, "drm-active-")) { + add_event(&events, &num_events, line, DRM_PMU_UNIT_BYTES, + "Total memory active in one or more engines"); + continue; + } + if (starts_with(line, "drm-cycles-")) { + add_event(&events, &num_events, line, DRM_PMU_UNIT_CYCLES, + "Busy cycles"); + continue; + } + if (starts_with(line, "drm-engine-capacity-")) { + add_event(&events, &num_events, line, DRM_PMU_UNIT_CAPACITY, + "Engine capacity"); + continue; + } + if (starts_with(line, "drm-engine-")) { + add_event(&events, &num_events, line, DRM_PMU_UNIT_NS, + "Utilization in ns"); + continue; + } + if (starts_with(line, "drm-maxfreq-")) { + add_event(&events, &num_events, line, DRM_PMU_UNIT_HZ, + "Maximum frequency"); + continue; + } + if (starts_with(line, "drm-purgeable-")) { + add_event(&events, &num_events, line, DRM_PMU_UNIT_BYTES, + "Size of resident and purgeable memory buffers"); + continue; + } + if (starts_with(line, "drm-resident-")) { + add_event(&events, &num_events, line, DRM_PMU_UNIT_BYTES, + "Size of resident memory buffers"); + continue; + } + if (starts_with(line, "drm-shared-")) { + add_event(&events, &num_events, line, DRM_PMU_UNIT_BYTES, + "Size of shared memory buffers"); + continue; + } + if (starts_with(line, "drm-total-cycles-")) { + add_event(&events, &num_events, line, DRM_PMU_UNIT_BYTES, + "Total busy cycles"); + continue; + } + if (starts_with(line, "drm-total-")) { + add_event(&events, &num_events, line, DRM_PMU_UNIT_BYTES, + "Size of shared and private memory"); + continue; + } + if (verbose > 1 && starts_with(line, "drm-") && + !starts_with(line, "drm-client-id:") && + !starts_with(line, "drm-pdev:")) + pr_debug("Unhandled DRM PMU fdinfo line match '%s'\n", line); + } + if (drm) { + drm->events = events; + drm->num_events = num_events; + list_add_tail(&drm->pmu.list, pmus); + } + free(line); + if (io.fd != -1) + close(io.fd); + return 0; +} + +void drm_pmu__exit(struct perf_pmu *pmu) +{ + struct drm_pmu *drm = container_of(pmu, struct drm_pmu, pmu); + + free(drm->events); +} + +bool drm_pmu__have_event(const struct perf_pmu *pmu, const char *name) +{ + struct drm_pmu *drm = container_of(pmu, struct drm_pmu, pmu); + + if (!starts_with(name, "drm-")) + return false; + + for (int i = 0; i < drm->num_events; i++) { + if (!strcasecmp(drm->events[i].name, name)) + return true; + } + return false; +} + +int drm_pmu__for_each_event(const struct perf_pmu *pmu, void *state, pmu_event_callback cb) +{ + struct drm_pmu *drm = container_of(pmu, struct drm_pmu, pmu); + + for (int i = 0; i < drm->num_events; i++) { + char encoding_buf[128]; + struct pmu_event_info info = { + .pmu = pmu, + .name = drm->events[i].name, + .alias = NULL, + .scale_unit = drm_pmu_scale_unit_strs[drm->events[i].unit], + .desc = drm->events[i].desc, + .long_desc = NULL, + .encoding_desc = encoding_buf, + .topic = "drm", + .pmu_name = pmu->name, + .event_type_desc = "DRM event", + }; + int ret; + + snprintf(encoding_buf, sizeof(encoding_buf), "%s/config=0x%x/", pmu->name, i); + + ret = cb(state, &info); + if (ret) + return ret; + } + return 0; +} + +size_t drm_pmu__num_events(const struct perf_pmu *pmu) +{ + const struct drm_pmu *drm = container_of(pmu, struct drm_pmu, pmu); + + return drm->num_events; +} + +static int drm_pmu__index_for_event(const struct drm_pmu *drm, const char *name) +{ + for (int i = 0; i < drm->num_events; i++) { + if (!strcmp(drm->events[i].name, name)) + return i; + } + return -1; +} + +static int drm_pmu__config_term(const struct drm_pmu *drm, + struct perf_event_attr *attr, + struct parse_events_term *term, + struct parse_events_error *err) +{ + if (term->type_term == PARSE_EVENTS__TERM_TYPE_USER) { + int i = drm_pmu__index_for_event(drm, term->config); + + if (i >= 0) { + attr->config = i; + return 0; + } + } + if (err) { + char *err_str; + + parse_events_error__handle(err, term->err_val, + asprintf(&err_str, + "unexpected drm event term (%s) %s", + parse_events__term_type_str(term->type_term), + term->config) < 0 + ? strdup("unexpected drm event term") + : err_str, + NULL); + } + return -EINVAL; +} + +int drm_pmu__config_terms(const struct perf_pmu *pmu, + struct perf_event_attr *attr, + struct parse_events_terms *terms, + struct parse_events_error *err) +{ + struct drm_pmu *drm = container_of(pmu, struct drm_pmu, pmu); + struct parse_events_term *term; + + list_for_each_entry(term, &terms->terms, list) { + if (drm_pmu__config_term(drm, attr, term, err)) + return -EINVAL; + } + + return 0; +} + +int drm_pmu__check_alias(const struct perf_pmu *pmu, struct parse_events_terms *terms, + struct perf_pmu_info *info, struct parse_events_error *err) +{ + struct drm_pmu *drm = container_of(pmu, struct drm_pmu, pmu); + struct parse_events_term *term = + list_first_entry(&terms->terms, struct parse_events_term, list); + + if (term->type_term == PARSE_EVENTS__TERM_TYPE_USER) { + int i = drm_pmu__index_for_event(drm, term->config); + + if (i >= 0) { + info->unit = drm_pmu_unit_strs[drm->events[i].unit]; + info->scale = 1; + return 0; + } + } + if (err) { + char *err_str; + + parse_events_error__handle(err, term->err_val, + asprintf(&err_str, + "unexpected drm event term (%s) %s", + parse_events__term_type_str(term->type_term), + term->config) < 0 + ? strdup("unexpected drm event term") + : err_str, + NULL); + } + return -EINVAL; +} + +struct minor_info { + unsigned int *minors; + int minors_num, minors_len; +}; + +static int for_each_drm_fdinfo_in_dir(int (*cb)(void *args, int fdinfo_dir_fd, const char *fd_name), + void *args, int proc_dir, const char *pid_name, + struct minor_info *minors) +{ + char buf[256]; + DIR *fd_dir; + struct dirent *fd_entry; + int fd_dir_fd, fdinfo_dir_fd = -1; + + + scnprintf(buf, sizeof(buf), "%s/fd", pid_name); + fd_dir_fd = openat(proc_dir, buf, O_DIRECTORY); + if (fd_dir_fd == -1) + return 0; /* Presumably lost race to open. */ + fd_dir = fdopendir(fd_dir_fd); + if (!fd_dir) { + close(fd_dir_fd); + return -ENOMEM; + } + while ((fd_entry = readdir(fd_dir)) != NULL) { + struct stat stat; + unsigned int minor; + bool is_dup = false; + int ret; + + if (fd_entry->d_type != DT_LNK) + continue; + + if (fstatat(fd_dir_fd, fd_entry->d_name, &stat, 0) != 0) + continue; + + if ((stat.st_mode & S_IFMT) != S_IFCHR || major(stat.st_rdev) != 226) + continue; + + minor = minor(stat.st_rdev); + for (int i = 0; i < minors->minors_num; i++) { + if (minor(stat.st_rdev) == minors->minors[i]) { + is_dup = true; + break; + } + } + if (is_dup) + continue; + + if (minors->minors_num == minors->minors_len) { + unsigned int *tmp = reallocarray(minors->minors, minors->minors_len + 4, + sizeof(unsigned int)); + + if (tmp) { + minors->minors = tmp; + minors->minors_len += 4; + } + } + minors->minors[minors->minors_num++] = minor; + if (fdinfo_dir_fd == -1) { + /* Open fdinfo dir if we have a DRM fd. */ + scnprintf(buf, sizeof(buf), "%s/fdinfo", pid_name); + fdinfo_dir_fd = openat(proc_dir, buf, O_DIRECTORY); + if (fdinfo_dir_fd == -1) + continue; + } + ret = cb(args, fdinfo_dir_fd, fd_entry->d_name); + if (ret) + return ret; + } + if (fdinfo_dir_fd != -1) + close(fdinfo_dir_fd); + closedir(fd_dir); + return 0; +} + +static int for_each_drm_fdinfo(bool skip_all_duplicates, + int (*cb)(void *args, int fdinfo_dir_fd, const char *fd_name), + void *args) +{ + DIR *proc_dir; + struct dirent *proc_entry; + int ret; + /* + * minors maintains an array of DRM minor device numbers seen for a pid, + * or for all pids if skip_all_duplicates is true, so that duplicates + * are ignored. + */ + struct minor_info minors = { + .minors = NULL, + .minors_num = 0, + .minors_len = 0, + }; + + proc_dir = opendir(procfs__mountpoint()); + if (!proc_dir) + return 0; + + /* Walk through the /proc directory. */ + while ((proc_entry = readdir(proc_dir)) != NULL) { + if (proc_entry->d_type != DT_DIR || + !isdigit(proc_entry->d_name[0])) + continue; + if (!skip_all_duplicates) { + /* Reset the seen minor numbers for each pid. */ + minors.minors_num = 0; + } + ret = for_each_drm_fdinfo_in_dir(cb, args, + dirfd(proc_dir), proc_entry->d_name, + &minors); + if (ret) + break; + } + free(minors.minors); + closedir(proc_dir); + return ret; +} + +int perf_pmus__read_drm_pmus(struct list_head *pmus) +{ + return for_each_drm_fdinfo(/*skip_all_duplicates=*/true, read_drm_pmus_cb, pmus); +} + +int evsel__drm_pmu_open(struct evsel *evsel, + struct perf_thread_map *threads, + int start_cpu_map_idx, int end_cpu_map_idx) +{ + (void)evsel; + (void)threads; + (void)start_cpu_map_idx; + (void)end_cpu_map_idx; + return 0; +} + +static uint64_t read_count_and_apply_unit(const char *count_and_unit, enum drm_pmu_unit unit) +{ + char *unit_ptr = NULL; + uint64_t count = strtoul(count_and_unit, &unit_ptr, 10); + + if (!unit_ptr) + return 0; + + while (isblank(*unit_ptr)) + unit_ptr++; + + switch (unit) { + case DRM_PMU_UNIT_BYTES: + if (*unit_ptr == '\0') + assert(count == 0); /* Generally undocumented, happens for 0. */ + else if (!strcmp(unit_ptr, "KiB")) + count *= 1024; + else if (!strcmp(unit_ptr, "MiB")) + count *= 1024 * 1024; + else + pr_err("Unexpected bytes unit '%s'\n", unit_ptr); + break; + case DRM_PMU_UNIT_CAPACITY: + /* No units expected. */ + break; + case DRM_PMU_UNIT_CYCLES: + /* No units expected. */ + break; + case DRM_PMU_UNIT_HZ: + if (!strcmp(unit_ptr, "Hz")) + count *= 1; + else if (!strcmp(unit_ptr, "KHz")) + count *= 1000; + else if (!strcmp(unit_ptr, "MHz")) + count *= 1000000; + else + pr_err("Unexpected hz unit '%s'\n", unit_ptr); + break; + case DRM_PMU_UNIT_NS: + /* Only unit ns expected. */ + break; + case DRM_PMU_UNIT_MAX: + default: + break; + } + return count; +} + +static uint64_t read_drm_event(int fdinfo_dir_fd, const char *fd_name, + const char *match, enum drm_pmu_unit unit) +{ + char buf[640]; + struct io io; + char *line = NULL; + size_t line_len; + uint64_t count = 0; + + io__init(&io, openat(fdinfo_dir_fd, fd_name, O_RDONLY), buf, sizeof(buf)); + if (io.fd == -1) { + /* Failed to open file, ignore. */ + return 0; + } + while (io__getline(&io, &line, &line_len) > 0) { + size_t i = strlen(match); + + if (strncmp(line, match, i)) + continue; + if (line[i] != ':') + continue; + while (isblank(line[++i])) + ; + if (line[line_len - 1] == '\n') + line[line_len - 1] = '\0'; + count = read_count_and_apply_unit(&line[i], unit); + break; + } + free(line); + close(io.fd); + return count; +} + +struct read_drm_event_cb_args { + const char *match; + uint64_t count; + enum drm_pmu_unit unit; +}; + +static int read_drm_event_cb(void *vargs, int fdinfo_dir_fd, const char *fd_name) +{ + struct read_drm_event_cb_args *args = vargs; + + args->count += read_drm_event(fdinfo_dir_fd, fd_name, args->match, args->unit); + return 0; +} + +static uint64_t drm_pmu__read_system_wide(struct drm_pmu *drm, struct evsel *evsel) +{ + struct read_drm_event_cb_args args = { + .count = 0, + .match = drm->events[evsel->core.attr.config].name, + .unit = drm->events[evsel->core.attr.config].unit, + }; + + for_each_drm_fdinfo(/*skip_all_duplicates=*/false, read_drm_event_cb, &args); + return args.count; +} + +static uint64_t drm_pmu__read_for_pid(struct drm_pmu *drm, struct evsel *evsel, int pid) +{ + struct read_drm_event_cb_args args = { + .count = 0, + .match = drm->events[evsel->core.attr.config].name, + .unit = drm->events[evsel->core.attr.config].unit, + }; + struct minor_info minors = { + .minors = NULL, + .minors_num = 0, + .minors_len = 0, + }; + int proc_dir = open(procfs__mountpoint(), O_DIRECTORY); + char pid_name[12]; + int ret; + + if (proc_dir < 0) + return 0; + + snprintf(pid_name, sizeof(pid_name), "%d", pid); + ret = for_each_drm_fdinfo_in_dir(read_drm_event_cb, &args, proc_dir, pid_name, &minors); + free(minors.minors); + close(proc_dir); + return ret == 0 ? args.count : 0; +} + +int evsel__drm_pmu_read(struct evsel *evsel, int cpu_map_idx, int thread) +{ + struct drm_pmu *drm = container_of(evsel->pmu, struct drm_pmu, pmu); + struct perf_counts_values *count, *old_count = NULL; + int pid = perf_thread_map__pid(evsel->core.threads, thread); + uint64_t counter; + + if (pid != -1) + counter = drm_pmu__read_for_pid(drm, evsel, pid); + else + counter = drm_pmu__read_system_wide(drm, evsel); + + if (evsel->prev_raw_counts) + old_count = perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread); + + count = perf_counts(evsel->counts, cpu_map_idx, thread); + if (old_count) { + count->val = old_count->val + counter; + count->run = old_count->run + 1; + count->ena = old_count->ena + 1; + } else { + count->val = counter; + count->run++; + count->ena++; + } + return 0; +} diff --git a/tools/perf/util/drm_pmu.h b/tools/perf/util/drm_pmu.h new file mode 100644 index 000000000000..e7f366fca8a4 --- /dev/null +++ b/tools/perf/util/drm_pmu.h @@ -0,0 +1,39 @@ +/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */ +#ifndef __DRM_PMU_H +#define __DRM_PMU_H +/* + * Linux DRM clients expose information through usage stats as documented in + * Documentation/gpu/drm-usage-stats.rst (available online at + * https://docs.kernel.org/gpu/drm-usage-stats.html). This is a tool like PMU + * that exposes DRM information. + */ + +#include "pmu.h" +#include <stdbool.h> + +struct list_head; +struct perf_thread_map; + +void drm_pmu__exit(struct perf_pmu *pmu); +bool drm_pmu__have_event(const struct perf_pmu *pmu, const char *name); +int drm_pmu__for_each_event(const struct perf_pmu *pmu, void *state, pmu_event_callback cb); +size_t drm_pmu__num_events(const struct perf_pmu *pmu); +int drm_pmu__config_terms(const struct perf_pmu *pmu, + struct perf_event_attr *attr, + struct parse_events_terms *terms, + struct parse_events_error *err); +int drm_pmu__check_alias(const struct perf_pmu *pmu, struct parse_events_terms *terms, + struct perf_pmu_info *info, struct parse_events_error *err); + + +bool perf_pmu__is_drm(const struct perf_pmu *pmu); +bool evsel__is_drm(const struct evsel *evsel); + +int perf_pmus__read_drm_pmus(struct list_head *pmus); + +int evsel__drm_pmu_open(struct evsel *evsel, + struct perf_thread_map *threads, + int start_cpu_map_idx, int end_cpu_map_idx); +int evsel__drm_pmu_read(struct evsel *evsel, int cpu_map_idx, int thread); + +#endif /* __DRM_PMU_H */ diff --git a/tools/perf/util/dso.c b/tools/perf/util/dso.c index 057fcf4225ac..282e3af85d5a 100644 --- a/tools/perf/util/dso.c +++ b/tools/perf/util/dso.c @@ -217,7 +217,7 @@ int dso__read_binary_type_filename(const struct dso *dso, break; } - build_id__sprintf(dso__bid_const(dso), build_id_hex); + build_id__snprintf(dso__bid(dso), build_id_hex, sizeof(build_id_hex)); len = __symbol__join_symfs(filename, size, "/usr/lib/debug/.build-id/"); snprintf(filename + len, size - len, "%.2s/%s.debug", build_id_hex, build_id_hex + 2); @@ -1382,64 +1382,76 @@ static void dso__set_long_name_id(struct dso *dso, const char *name, bool name_a static int __dso_id__cmp(const struct dso_id *a, const struct dso_id *b) { - if (a->maj > b->maj) return -1; - if (a->maj < b->maj) return 1; + if (a->mmap2_valid && b->mmap2_valid) { + if (a->maj > b->maj) return -1; + if (a->maj < b->maj) return 1; - if (a->min > b->min) return -1; - if (a->min < b->min) return 1; + if (a->min > b->min) return -1; + if (a->min < b->min) return 1; - if (a->ino > b->ino) return -1; - if (a->ino < b->ino) return 1; - - /* - * Synthesized MMAP events have zero ino_generation, avoid comparing - * them with MMAP events with actual ino_generation. - * - * I found it harmful because the mismatch resulted in a new - * dso that did not have a build ID whereas the original dso did have a - * build ID. The build ID was essential because the object was not found - * otherwise. - Adrian - */ - if (a->ino_generation && b->ino_generation) { + if (a->ino > b->ino) return -1; + if (a->ino < b->ino) return 1; + } + if (a->mmap2_ino_generation_valid && b->mmap2_ino_generation_valid) { if (a->ino_generation > b->ino_generation) return -1; if (a->ino_generation < b->ino_generation) return 1; } - + if (build_id__is_defined(&a->build_id) && build_id__is_defined(&b->build_id)) { + if (a->build_id.size != b->build_id.size) + return a->build_id.size < b->build_id.size ? -1 : 1; + return memcmp(a->build_id.data, b->build_id.data, a->build_id.size); + } return 0; } -bool dso_id__empty(const struct dso_id *id) -{ - if (!id) - return true; - - return !id->maj && !id->min && !id->ino && !id->ino_generation; -} +const struct dso_id dso_id_empty = { + { + .maj = 0, + .min = 0, + .ino = 0, + .ino_generation = 0, + }, + .mmap2_valid = false, + .mmap2_ino_generation_valid = false, + { + .size = 0, + } +}; -void __dso__inject_id(struct dso *dso, const struct dso_id *id) +void __dso__improve_id(struct dso *dso, const struct dso_id *id) { struct dsos *dsos = dso__dsos(dso); struct dso_id *dso_id = dso__id(dso); + bool changed = false; /* dsos write lock held by caller. */ - dso_id->maj = id->maj; - dso_id->min = id->min; - dso_id->ino = id->ino; - dso_id->ino_generation = id->ino_generation; - - if (dsos) + if (id->mmap2_valid && !dso_id->mmap2_valid) { + dso_id->maj = id->maj; + dso_id->min = id->min; + dso_id->ino = id->ino; + dso_id->mmap2_valid = true; + changed = true; + } + if (id->mmap2_ino_generation_valid && !dso_id->mmap2_ino_generation_valid) { + dso_id->ino_generation = id->ino_generation; + dso_id->mmap2_ino_generation_valid = true; + changed = true; + } + if (build_id__is_defined(&id->build_id) && !build_id__is_defined(&dso_id->build_id)) { + dso_id->build_id = id->build_id; + changed = true; + } + if (changed && dsos) dsos->sorted = false; } int dso_id__cmp(const struct dso_id *a, const struct dso_id *b) { - /* - * The second is always dso->id, so zeroes if not set, assume passing - * NULL for a means a zeroed id - */ - if (dso_id__empty(a) || dso_id__empty(b)) + if (a == &dso_id_empty || b == &dso_id_empty) { + /* There is no valid data to compare so the comparison always returns identical. */ return 0; + } return __dso_id__cmp(a, b); } @@ -1540,7 +1552,6 @@ struct dso *dso__new_id(const char *name, const struct dso_id *id) dso->loaded = 0; dso->rel = 0; dso->sorted_by_name = 0; - dso->has_build_id = 0; dso->has_srcline = 1; dso->a2l_fails = 1; dso->kernel = DSO_SPACE__USER; @@ -1612,6 +1623,10 @@ struct dso *dso__get(struct dso *dso) void dso__put(struct dso *dso) { +#ifdef REFCNT_CHECKING + if (dso && dso__data(dso) && refcount_read(&RC_CHK_ACCESS(dso)->refcnt) == 2) + dso__data_close(dso); +#endif if (dso && refcount_dec_and_test(&RC_CHK_ACCESS(dso)->refcnt)) dso__delete(dso); else @@ -1645,15 +1660,14 @@ int dso__swap_init(struct dso *dso, unsigned char eidata) return 0; } -void dso__set_build_id(struct dso *dso, struct build_id *bid) +void dso__set_build_id(struct dso *dso, const struct build_id *bid) { - RC_CHK_ACCESS(dso)->bid = *bid; - RC_CHK_ACCESS(dso)->has_build_id = 1; + dso__id(dso)->build_id = *bid; } -bool dso__build_id_equal(const struct dso *dso, struct build_id *bid) +bool dso__build_id_equal(const struct dso *dso, const struct build_id *bid) { - const struct build_id *dso_bid = dso__bid_const(dso); + const struct build_id *dso_bid = dso__bid(dso); if (dso_bid->size > bid->size && dso_bid->size == BUILD_ID_SIZE) { /* @@ -1672,18 +1686,20 @@ bool dso__build_id_equal(const struct dso *dso, struct build_id *bid) void dso__read_running_kernel_build_id(struct dso *dso, struct machine *machine) { char path[PATH_MAX]; + struct build_id bid = { .size = 0, }; if (machine__is_default_guest(machine)) return; sprintf(path, "%s/sys/kernel/notes", machine->root_dir); - if (sysfs__read_build_id(path, dso__bid(dso)) == 0) - dso__set_has_build_id(dso); + sysfs__read_build_id(path, &bid); + dso__set_build_id(dso, &bid); } int dso__kernel_module_get_build_id(struct dso *dso, const char *root_dir) { char filename[PATH_MAX]; + struct build_id bid = { .size = 0, }; /* * kernel module short names are of the form "[module]" and * we need just "module" here. @@ -1694,9 +1710,8 @@ int dso__kernel_module_get_build_id(struct dso *dso, "%s/sys/module/%.*s/notes/.note.gnu.build-id", root_dir, (int)strlen(name) - 1, name); - if (sysfs__read_build_id(filename, dso__bid(dso)) == 0) - dso__set_has_build_id(dso); - + sysfs__read_build_id(filename, &bid); + dso__set_build_id(dso, &bid); return 0; } @@ -1704,7 +1719,7 @@ static size_t dso__fprintf_buildid(struct dso *dso, FILE *fp) { char sbuild_id[SBUILD_ID_SIZE]; - build_id__sprintf(dso__bid(dso), sbuild_id); + build_id__snprintf(dso__bid(dso), sbuild_id, sizeof(sbuild_id)); return fprintf(fp, "%s", sbuild_id); } diff --git a/tools/perf/util/dso.h b/tools/perf/util/dso.h index c87564471f9b..3457d713d3c5 100644 --- a/tools/perf/util/dso.h +++ b/tools/perf/util/dso.h @@ -185,14 +185,33 @@ enum dso_load_errno { #define DSO__DATA_CACHE_SIZE 4096 #define DSO__DATA_CACHE_MASK ~(DSO__DATA_CACHE_SIZE - 1) -/* - * Data about backing storage DSO, comes from PERF_RECORD_MMAP2 meta events +/** + * struct dso_id + * + * Data about backing storage DSO, comes from PERF_RECORD_MMAP2 meta events, + * reading from /proc/pid/maps or synthesis of build_ids from DSOs. Possibly + * incomplete at any particular use. */ struct dso_id { - u32 maj; - u32 min; - u64 ino; - u64 ino_generation; + /* Data related to the mmap2 event or read from /proc/pid/maps. */ + struct { + u32 maj; + u32 min; + u64 ino; + u64 ino_generation; + }; + /** @mmap2_valid: Are the maj, min and ino fields valid? */ + bool mmap2_valid; + /** + * @mmap2_ino_generation_valid: Is the ino_generation valid? Generally + * false for /proc/pid/maps mmap event. + */ + bool mmap2_ino_generation_valid; + /** + * @build_id: A possibly populated build_id. build_id__is_defined checks + * whether it is populated. + */ + struct build_id build_id; }; struct dso_cache { @@ -243,7 +262,6 @@ DECLARE_RC_STRUCT(dso) { u64 addr; struct symbol *symbol; } last_find_result; - struct build_id bid; u64 text_offset; u64 text_end; const char *short_name; @@ -276,7 +294,6 @@ DECLARE_RC_STRUCT(dso) { enum dso_swap_type needs_swap:2; bool is_kmod:1; u8 adjust_symbols:1; - u8 has_build_id:1; u8 header_build_id:1; u8 has_srcline:1; u8 hit:1; @@ -292,6 +309,9 @@ DECLARE_RC_STRUCT(dso) { }; extern struct mutex _dso__data_open_lock; +extern const struct dso_id dso_id_empty; + +int dso_id__cmp(const struct dso_id *a, const struct dso_id *b); /* dso__for_each_symbol - iterate over the symbols of given type * @@ -362,31 +382,11 @@ static inline void dso__set_auxtrace_cache(struct dso *dso, struct auxtrace_cach RC_CHK_ACCESS(dso)->auxtrace_cache = cache; } -static inline struct build_id *dso__bid(struct dso *dso) -{ - return &RC_CHK_ACCESS(dso)->bid; -} - -static inline const struct build_id *dso__bid_const(const struct dso *dso) -{ - return &RC_CHK_ACCESS(dso)->bid; -} - static inline struct dso_bpf_prog *dso__bpf_prog(struct dso *dso) { return &RC_CHK_ACCESS(dso)->bpf_prog; } -static inline bool dso__has_build_id(const struct dso *dso) -{ - return RC_CHK_ACCESS(dso)->has_build_id; -} - -static inline void dso__set_has_build_id(struct dso *dso) -{ - RC_CHK_ACCESS(dso)->has_build_id = true; -} - static inline bool dso__has_srcline(const struct dso *dso) { return RC_CHK_ACCESS(dso)->has_srcline; @@ -462,6 +462,16 @@ static inline const struct dso_id *dso__id_const(const struct dso *dso) return &RC_CHK_ACCESS(dso)->id; } +static inline const struct build_id *dso__bid(const struct dso *dso) +{ + return &dso__id_const(dso)->build_id; +} + +static inline bool dso__has_build_id(const struct dso *dso) +{ + return build_id__is_defined(dso__bid(dso)); +} + static inline struct rb_root_cached *dso__inlined_nodes(struct dso *dso) { return &RC_CHK_ACCESS(dso)->inlined_nodes; @@ -699,9 +709,6 @@ static inline void dso__set_text_offset(struct dso *dso, u64 val) RC_CHK_ACCESS(dso)->text_offset = val; } -int dso_id__cmp(const struct dso_id *a, const struct dso_id *b); -bool dso_id__empty(const struct dso_id *id); - struct dso *dso__new_id(const char *name, const struct dso_id *id); struct dso *dso__new(const char *name); void dso__delete(struct dso *dso); @@ -709,7 +716,7 @@ void dso__delete(struct dso *dso); int dso__cmp_id(struct dso *a, struct dso *b); void dso__set_short_name(struct dso *dso, const char *name, bool name_allocated); void dso__set_long_name(struct dso *dso, const char *name, bool name_allocated); -void __dso__inject_id(struct dso *dso, const struct dso_id *id); +void __dso__improve_id(struct dso *dso, const struct dso_id *id); int dso__name_len(const struct dso *dso); @@ -739,8 +746,8 @@ void dso__sort_by_name(struct dso *dso); int dso__swap_init(struct dso *dso, unsigned char eidata); -void dso__set_build_id(struct dso *dso, struct build_id *bid); -bool dso__build_id_equal(const struct dso *dso, struct build_id *bid); +void dso__set_build_id(struct dso *dso, const struct build_id *bid); +bool dso__build_id_equal(const struct dso *dso, const struct build_id *bid); void dso__read_running_kernel_build_id(struct dso *dso, struct machine *machine); int dso__kernel_module_get_build_id(struct dso *dso, const char *root_dir); diff --git a/tools/perf/util/dsos.c b/tools/perf/util/dsos.c index 4d213017d202..0a7645c7fae7 100644 --- a/tools/perf/util/dsos.c +++ b/tools/perf/util/dsos.c @@ -72,6 +72,7 @@ static int dsos__read_build_ids_cb(struct dso *dso, void *data) { struct dsos__read_build_ids_cb_args *args = data; struct nscookie nsc; + struct build_id bid = { .size = 0, }; if (args->with_hits && !dso__hit(dso) && !dso__is_vdso(dso)) return 0; @@ -80,15 +81,15 @@ static int dsos__read_build_ids_cb(struct dso *dso, void *data) return 0; } nsinfo__mountns_enter(dso__nsinfo(dso), &nsc); - if (filename__read_build_id(dso__long_name(dso), dso__bid(dso)) > 0) { + if (filename__read_build_id(dso__long_name(dso), &bid) > 0) { + dso__set_build_id(dso, &bid); args->have_build_id = true; - dso__set_has_build_id(dso); } else if (errno == ENOENT && dso__nsinfo(dso)) { char *new_name = dso__filename_with_chroot(dso, dso__long_name(dso)); - if (new_name && filename__read_build_id(new_name, dso__bid(dso)) > 0) { + if (new_name && filename__read_build_id(new_name, &bid) > 0) { + dso__set_build_id(dso, &bid); args->have_build_id = true; - dso__set_has_build_id(dso); } free(new_name); } @@ -286,7 +287,7 @@ struct dso *dsos__find(struct dsos *dsos, const char *name, bool cmp_short) struct dso *res; down_read(&dsos->lock); - res = __dsos__find_id(dsos, name, NULL, cmp_short, /*write_locked=*/false); + res = __dsos__find_id(dsos, name, &dso_id_empty, cmp_short, /*write_locked=*/false); up_read(&dsos->lock); return res; } @@ -344,8 +345,8 @@ static struct dso *__dsos__findnew_id(struct dsos *dsos, const char *name, const { struct dso *dso = __dsos__find_id(dsos, name, id, false, /*write_locked=*/true); - if (dso && dso_id__empty(dso__id(dso)) && !dso_id__empty(id)) - __dso__inject_id(dso, id); + if (dso) + __dso__improve_id(dso, id); return dso ? dso : __dsos__addnew_id(dsos, name, id); } @@ -373,7 +374,7 @@ static int dsos__fprintf_buildid_cb(struct dso *dso, void *data) if (args->skip && args->skip(dso, args->parm)) return 0; - build_id__sprintf(dso__bid(dso), sbuild_id); + build_id__snprintf(dso__bid(dso), sbuild_id, sizeof(sbuild_id)); args->ret += fprintf(args->fp, "%-40s %s\n", sbuild_id, dso__long_name(dso)); return 0; } @@ -436,7 +437,8 @@ struct dso *dsos__findnew_module_dso(struct dsos *dsos, down_write(&dsos->lock); - dso = __dsos__find_id(dsos, m->name, NULL, /*cmp_short=*/true, /*write_locked=*/true); + dso = __dsos__find_id(dsos, m->name, &dso_id_empty, /*cmp_short=*/true, + /*write_locked=*/true); if (dso) { up_write(&dsos->lock); return dso; diff --git a/tools/perf/util/env.c b/tools/perf/util/env.c index 36411749e007..c8c248754621 100644 --- a/tools/perf/util/env.c +++ b/tools/perf/util/env.c @@ -3,8 +3,10 @@ #include "debug.h" #include "env.h" #include "util/header.h" -#include "linux/compiler.h" +#include "util/rwsem.h" +#include <linux/compiler.h> #include <linux/ctype.h> +#include <linux/rbtree.h> #include <linux/string.h> #include <linux/zalloc.h> #include "cgroup.h" @@ -17,8 +19,6 @@ #include "strbuf.h" #include "trace/beauty/beauty.h" -struct perf_env perf_env; - #ifdef HAVE_LIBBPF_SUPPORT #include "bpf-event.h" #include "bpf-utils.h" @@ -89,6 +89,20 @@ out: return node; } +void perf_env__iterate_bpf_prog_info(struct perf_env *env, + void (*cb)(struct bpf_prog_info_node *node, + void *data), + void *data) +{ + struct rb_node *first; + + down_read(&env->bpf_progs.lock); + first = rb_first(&env->bpf_progs.infos); + for (struct rb_node *node = first; node != NULL; node = rb_next(node)) + (*cb)(rb_entry(node, struct bpf_prog_info_node, rb_node), data); + up_read(&env->bpf_progs.lock); +} + bool perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node) { bool ret; @@ -174,6 +188,7 @@ static void perf_env__purge_bpf(struct perf_env *env) next = rb_next(&node->rb_node); rb_erase(&node->rb_node, root); zfree(&node->info_linear); + bpf_metadata_free(node->metadata); free(node); } @@ -254,6 +269,7 @@ void perf_env__exit(struct perf_env *env) void perf_env__init(struct perf_env *env) { + memset(env, 0, sizeof(*env)); #ifdef HAVE_LIBBPF_SUPPORT env->bpf_progs.infos = RB_ROOT; env->bpf_progs.btfs = RB_ROOT; @@ -416,6 +432,116 @@ static int perf_env__read_nr_cpus_avail(struct perf_env *env) return env->nr_cpus_avail ? 0 : -ENOENT; } +static int __perf_env__read_core_pmu_caps(const struct perf_pmu *pmu, + int *nr_caps, char ***caps, + unsigned int *max_branches, + unsigned int *br_cntr_nr, + unsigned int *br_cntr_width) +{ + struct perf_pmu_caps *pcaps = NULL; + char *ptr, **tmp; + int ret = 0; + + *nr_caps = 0; + *caps = NULL; + + if (!pmu->nr_caps) + return 0; + + *caps = calloc(pmu->nr_caps, sizeof(char *)); + if (!*caps) + return -ENOMEM; + + tmp = *caps; + list_for_each_entry(pcaps, &pmu->caps, list) { + if (asprintf(&ptr, "%s=%s", pcaps->name, pcaps->value) < 0) { + ret = -ENOMEM; + goto error; + } + + *tmp++ = ptr; + + if (!strcmp(pcaps->name, "branches")) + *max_branches = atoi(pcaps->value); + else if (!strcmp(pcaps->name, "branch_counter_nr")) + *br_cntr_nr = atoi(pcaps->value); + else if (!strcmp(pcaps->name, "branch_counter_width")) + *br_cntr_width = atoi(pcaps->value); + } + *nr_caps = pmu->nr_caps; + return 0; +error: + while (tmp-- != *caps) + zfree(tmp); + zfree(caps); + *nr_caps = 0; + return ret; +} + +int perf_env__read_core_pmu_caps(struct perf_env *env) +{ + struct pmu_caps *pmu_caps; + struct perf_pmu *pmu = NULL; + int nr_pmu, i = 0, j; + int ret; + + nr_pmu = perf_pmus__num_core_pmus(); + + if (!nr_pmu) + return -ENODEV; + + if (nr_pmu == 1) { + pmu = perf_pmus__find_core_pmu(); + if (!pmu) + return -ENODEV; + ret = perf_pmu__caps_parse(pmu); + if (ret < 0) + return ret; + return __perf_env__read_core_pmu_caps(pmu, &env->nr_cpu_pmu_caps, + &env->cpu_pmu_caps, + &env->max_branches, + &env->br_cntr_nr, + &env->br_cntr_width); + } + + pmu_caps = calloc(nr_pmu, sizeof(*pmu_caps)); + if (!pmu_caps) + return -ENOMEM; + + while ((pmu = perf_pmus__scan_core(pmu)) != NULL) { + if (perf_pmu__caps_parse(pmu) <= 0) + continue; + ret = __perf_env__read_core_pmu_caps(pmu, &pmu_caps[i].nr_caps, + &pmu_caps[i].caps, + &pmu_caps[i].max_branches, + &pmu_caps[i].br_cntr_nr, + &pmu_caps[i].br_cntr_width); + if (ret) + goto error; + + pmu_caps[i].pmu_name = strdup(pmu->name); + if (!pmu_caps[i].pmu_name) { + ret = -ENOMEM; + goto error; + } + i++; + } + + env->nr_pmus_with_caps = nr_pmu; + env->pmu_caps = pmu_caps; + + return 0; +error: + for (i = 0; i < nr_pmu; i++) { + for (j = 0; j < pmu_caps[i].nr_caps; j++) + zfree(&pmu_caps[i].caps[j]); + zfree(&pmu_caps[i].caps); + zfree(&pmu_caps[i].pmu_name); + } + zfree(&pmu_caps); + return ret; +} + const char *perf_env__raw_arch(struct perf_env *env) { return env && !perf_env__read_arch(env) ? env->arch : "unknown"; diff --git a/tools/perf/util/env.h b/tools/perf/util/env.h index d90e343cf1fa..e00179787a34 100644 --- a/tools/perf/util/env.h +++ b/tools/perf/util/env.h @@ -150,8 +150,7 @@ enum perf_compress_type { struct bpf_prog_info_node; struct btf_node; -extern struct perf_env perf_env; - +int perf_env__read_core_pmu_caps(struct perf_env *env); void perf_env__exit(struct perf_env *env); int perf_env__kernel_is_64_bit(struct perf_env *env); @@ -174,16 +173,22 @@ const char *perf_env__raw_arch(struct perf_env *env); int perf_env__nr_cpus_avail(struct perf_env *env); void perf_env__init(struct perf_env *env); +#ifdef HAVE_LIBBPF_SUPPORT bool __perf_env__insert_bpf_prog_info(struct perf_env *env, struct bpf_prog_info_node *info_node); bool perf_env__insert_bpf_prog_info(struct perf_env *env, struct bpf_prog_info_node *info_node); struct bpf_prog_info_node *perf_env__find_bpf_prog_info(struct perf_env *env, __u32 prog_id); +void perf_env__iterate_bpf_prog_info(struct perf_env *env, + void (*cb)(struct bpf_prog_info_node *node, + void *data), + void *data); bool perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node); bool __perf_env__insert_btf(struct perf_env *env, struct btf_node *btf_node); struct btf_node *perf_env__find_btf(struct perf_env *env, __u32 btf_id); struct btf_node *__perf_env__find_btf(struct perf_env *env, __u32 btf_id); +#endif // HAVE_LIBBPF_SUPPORT int perf_env__numa_node(struct perf_env *env, struct perf_cpu cpu); char *perf_env__find_pmu_cap(struct perf_env *env, const char *pmu_name, diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index 7544a3104e21..fcf44149feb2 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -1,9 +1,12 @@ #include <errno.h> #include <fcntl.h> #include <inttypes.h> +#include <linux/compiler.h> #include <linux/kernel.h> #include <linux/types.h> #include <perf/cpumap.h> +#include <perf/event.h> +#include <stdio.h> #include <sys/types.h> #include <sys/stat.h> #include <unistd.h> @@ -78,6 +81,7 @@ static const char *perf_event__names[] = { [PERF_RECORD_COMPRESSED] = "COMPRESSED", [PERF_RECORD_FINISHED_INIT] = "FINISHED_INIT", [PERF_RECORD_COMPRESSED2] = "COMPRESSED2", + [PERF_RECORD_BPF_METADATA] = "BPF_METADATA", }; const char *perf_event__name(unsigned int id) @@ -330,7 +334,7 @@ size_t perf_event__fprintf_mmap2(union perf_event *event, FILE *fp) build_id__init(&bid, event->mmap2.build_id, event->mmap2.build_id_size); - build_id__sprintf(&bid, sbuild_id); + build_id__snprintf(&bid, sbuild_id, sizeof(sbuild_id)); return fprintf(fp, " %d/%d: [%#" PRI_lx64 "(%#" PRI_lx64 ") @ %#" PRI_lx64 " <%s>]: %c%c%c%c %s\n", @@ -505,6 +509,20 @@ size_t perf_event__fprintf_bpf(union perf_event *event, FILE *fp) event->bpf.type, event->bpf.flags, event->bpf.id); } +size_t perf_event__fprintf_bpf_metadata(union perf_event *event, FILE *fp) +{ + struct perf_record_bpf_metadata *metadata = &event->bpf_metadata; + size_t ret; + + ret = fprintf(fp, " prog %s\n", metadata->prog_name); + for (__u32 i = 0; i < metadata->nr_entries; i++) { + ret += fprintf(fp, " entry %d: %20s = %s\n", i, + metadata->entries[i].key, + metadata->entries[i].value); + } + return ret; +} + static int text_poke_printer(enum binary_printer_ops op, unsigned int val, void *extra, FILE *fp) { @@ -602,6 +620,9 @@ size_t perf_event__fprintf(union perf_event *event, struct machine *machine, FIL case PERF_RECORD_AUX_OUTPUT_HW_ID: ret += perf_event__fprintf_aux_output_hw_id(event, fp); break; + case PERF_RECORD_BPF_METADATA: + ret += perf_event__fprintf_bpf_metadata(event, fp); + break; default: ret += fprintf(fp, "\n"); } diff --git a/tools/perf/util/event.h b/tools/perf/util/event.h index 664bf39567ce..e40d16d3246c 100644 --- a/tools/perf/util/event.h +++ b/tools/perf/util/event.h @@ -370,6 +370,7 @@ size_t perf_event__fprintf_namespaces(union perf_event *event, FILE *fp); size_t perf_event__fprintf_cgroup(union perf_event *event, FILE *fp); size_t perf_event__fprintf_ksymbol(union perf_event *event, FILE *fp); size_t perf_event__fprintf_bpf(union perf_event *event, FILE *fp); +size_t perf_event__fprintf_bpf_metadata(union perf_event *event, FILE *fp); size_t perf_event__fprintf_text_poke(union perf_event *event, struct machine *machine,FILE *fp); size_t perf_event__fprintf(union perf_event *event, struct machine *machine, FILE *fp); @@ -390,11 +391,6 @@ extern unsigned int proc_map_timeout; #define PAGE_SIZE_NAME_LEN 32 char *get_page_size_name(u64 size, char *str); -void arch_perf_parse_sample_weight(struct perf_sample *data, const __u64 *array, u64 type); -void arch_perf_synthesize_sample_weight(const struct perf_sample *data, __u64 *array, u64 type); -const char *arch_perf_header_entry(const char *se_header); -int arch_support_sort_key(const char *sort_key); - static inline bool perf_event_header__cpumode_is_guest(u8 cpumode) { return cpumode == PERF_RECORD_MISC_GUEST_KERNEL || diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index dcd1130502df..80d8387e6b97 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -35,6 +35,7 @@ #include "util/util.h" #include "util/env.h" #include "util/intel-tpebs.h" +#include "util/metricgroup.h" #include "util/strbuf.h" #include <signal.h> #include <unistd.h> @@ -83,6 +84,7 @@ void evlist__init(struct evlist *evlist, struct perf_cpu_map *cpus, evlist->ctl_fd.ack = -1; evlist->ctl_fd.pos = -1; evlist->nr_br_cntr = -1; + metricgroup__rblist_init(&evlist->metric_events); } struct evlist *evlist__new(void) @@ -173,6 +175,7 @@ static void evlist__purge(struct evlist *evlist) void evlist__exit(struct evlist *evlist) { + metricgroup__rblist_exit(&evlist->metric_events); event_enable_timer__exit(&evlist->eet); zfree(&evlist->mmap); zfree(&evlist->overwrite_mmap); @@ -1006,8 +1009,7 @@ int evlist__create_maps(struct evlist *evlist, struct target *target) * per-thread data. thread_map__new_str will call * thread_map__new_all_cpus to enumerate all threads. */ - threads = thread_map__new_str(target->pid, target->tid, target->uid, - all_threads); + threads = thread_map__new_str(target->pid, target->tid, all_threads); if (!threads) return -1; @@ -2547,20 +2549,7 @@ void evlist__warn_user_requested_cpus(struct evlist *evlist, const char *cpu_lis return; evlist__for_each_entry(evlist, pos) { - struct perf_cpu_map *intersect, *to_test, *online = cpu_map__online(); - const struct perf_pmu *pmu = evsel__find_pmu(pos); - - to_test = pmu && pmu->is_core ? pmu->cpus : online; - intersect = perf_cpu_map__intersect(to_test, user_requested_cpus); - if (!perf_cpu_map__equal(intersect, user_requested_cpus)) { - char buf[128]; - - cpu_map__snprint(to_test, buf, sizeof(buf)); - pr_warning("WARNING: A requested CPU in '%s' is not supported by PMU '%s' (CPUs %s) for event '%s'\n", - cpu_list, pmu ? pmu->name : "cpu", buf, evsel__name(pos)); - } - perf_cpu_map__put(intersect); - perf_cpu_map__put(online); + evsel__warn_user_requested_cpus(pos, user_requested_cpus); } perf_cpu_map__put(user_requested_cpus); } diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h index 85859708393e..5e71e3dc6042 100644 --- a/tools/perf/util/evlist.h +++ b/tools/perf/util/evlist.h @@ -12,6 +12,7 @@ #include <perf/evlist.h> #include "events_stats.h" #include "evsel.h" +#include "rblist.h" #include <pthread.h> #include <signal.h> #include <unistd.h> @@ -70,7 +71,7 @@ struct evlist { struct mmap *overwrite_mmap; struct evsel *selected; struct events_stats stats; - struct perf_env *env; + struct perf_session *session; void (*trace_event_sample_raw)(struct evlist *evlist, union perf_event *event, struct perf_sample *sample); @@ -86,6 +87,11 @@ struct evlist { int pos; /* index at evlist core object to check signals */ } ctl_fd; struct event_enable_timer *eet; + /** + * @metric_events: A list of struct metric_event which each have a list + * of struct metric_expr. + */ + struct rblist metric_events; }; struct evsel_str_handler { @@ -105,6 +111,7 @@ void evlist__add(struct evlist *evlist, struct evsel *entry); void evlist__remove(struct evlist *evlist, struct evsel *evsel); int arch_evlist__cmp(const struct evsel *lhs, const struct evsel *rhs); +int arch_evlist__add_required_events(struct list_head *list); int evlist__add_dummy(struct evlist *evlist); struct evsel *evlist__add_aux_dummy(struct evlist *evlist, bool system_wide); diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c index d55482f094bf..d264c143b592 100644 --- a/tools/perf/util/evsel.c +++ b/tools/perf/util/evsel.c @@ -48,6 +48,7 @@ #include "record.h" #include "debug.h" #include "trace-event.h" +#include "session.h" #include "stat.h" #include "string2.h" #include "memswap.h" @@ -56,8 +57,10 @@ #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" @@ -487,7 +490,7 @@ struct evsel *evsel__clone(struct evsel *dest, struct evsel *orig) 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; @@ -570,24 +573,6 @@ out_err: return NULL; } -static int trace_event__id(const char *sys, const char *name) -{ - char *tp_dir = get_events_file(sys); - char path[PATH_MAX]; - int id, err; - - if (!tp_dir) - return -1; - - scnprintf(path, PATH_MAX, "%s/%s/id", tp_dir, name); - put_events_file(tp_dir); - err = filename__read_int(path, &id); - if (err) - return err; - - return id; -} - /* * Returns pointer with encoded error via <linux/err.h> interface. */ @@ -621,7 +606,7 @@ struct evsel *evsel__newtp_idx(const char *sys, const char *name, int idx, bool event_attr_init(&attr); if (format) { - id = trace_event__id(sys, name); + id = tp_pmu__id(sys, name); if (id < 0) { err = id; goto out_free; @@ -1526,7 +1511,7 @@ 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; /* @@ -1656,6 +1641,15 @@ 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)); @@ -1669,9 +1663,7 @@ void evsel__exit(struct evsel *evsel) 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); #ifdef HAVE_LIBTRACEEVENT @@ -1686,6 +1678,8 @@ void evsel__exit(struct evsel *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) @@ -1889,6 +1883,9 @@ int evsel__read_counter(struct evsel *evsel, int cpu_map_idx, int thread) 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); @@ -2610,6 +2607,11 @@ fallback_missing_features: start_cpu_map_idx, end_cpu_map_idx); } + if (evsel__is_drm(evsel)) { + return evsel__drm_pmu_open(evsel, threads, + start_cpu_map_idx, + end_cpu_map_idx); + } for (idx = start_cpu_map_idx; idx < end_cpu_map_idx; idx++) { cpu = perf_cpu_map__cpu(cpus, idx); @@ -2743,17 +2745,32 @@ void evsel__close(struct evsel *evsel) 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, NULL, cpu_map_idx, cpu_map_idx + 1); + return evsel__open_cpu(evsel, cpus, threads, 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, @@ -2846,11 +2863,18 @@ 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) @@ -3236,7 +3260,7 @@ 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++; } @@ -3839,11 +3863,16 @@ int evsel__open_strerror(struct evsel *evsel, struct target *target, 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 && evsel->evlist->env) - 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) @@ -4071,3 +4100,27 @@ void evsel__uniquify_counter(struct evsel *counter) 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); +} diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h index 6dbc9690e0c9..5797a02e5d6a 100644 --- a/tools/perf/util/evsel.h +++ b/tools/perf/util/evsel.h @@ -280,6 +280,8 @@ void evsel__init(struct evsel *evsel, struct perf_event_attr *attr, int idx); void evsel__exit(struct evsel *evsel); void evsel__delete(struct evsel *evsel); +void evsel__set_priv_destructor(void (*destructor)(void *priv)); + struct callchain_param; void evsel__config(struct evsel *evsel, struct record_opts *opts, @@ -349,6 +351,9 @@ int evsel__enable(struct evsel *evsel); int evsel__disable(struct evsel *evsel); int evsel__disable_cpu(struct evsel *evsel, 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); int evsel__open_per_cpu(struct evsel *evsel, struct perf_cpu_map *cpus, int cpu_map_idx); int evsel__open_per_thread(struct evsel *evsel, struct perf_thread_map *threads); int evsel__open(struct evsel *evsel, struct perf_cpu_map *cpus, @@ -537,6 +542,7 @@ static inline bool evsel__is_dummy_event(struct evsel *evsel) (evsel->core.attr.config == PERF_COUNT_SW_DUMMY); } +struct perf_session *evsel__session(struct evsel *evsel); struct perf_env *evsel__env(struct evsel *evsel); int evsel__store_ids(struct evsel *evsel, struct evlist *evlist); @@ -572,4 +578,6 @@ void evsel__set_config_if_unset(struct perf_pmu *pmu, struct evsel *evsel, bool evsel__is_offcpu_event(struct evsel *evsel); +void evsel__warn_user_requested_cpus(struct evsel *evsel, struct perf_cpu_map *user_requested_cpus); + #endif /* __PERF_EVSEL_H */ diff --git a/tools/perf/util/expr.c b/tools/perf/util/expr.c index 6413537442aa..7fda0ff89c16 100644 --- a/tools/perf/util/expr.c +++ b/tools/perf/util/expr.c @@ -166,8 +166,12 @@ int expr__add_id_val_source_count(struct expr_parse_ctx *ctx, const char *id, data_ptr->kind = EXPR_ID_DATA__VALUE; ret = hashmap__set(ctx->ids, id, data_ptr, &old_key, &old_data); - if (ret) + if (ret) { free(data_ptr); + } else if (old_data) { + data_ptr->val.val += old_data->val.val; + data_ptr->val.source_count += old_data->val.source_count; + } free(old_key); free(old_data); return ret; @@ -397,7 +401,7 @@ double expr__get_literal(const char *literal, const struct expr_scanner_ctx *ctx if (ev != TOOL_PMU__EVENT_NONE) { u64 count; - if (tool_pmu__read_event(ev, &count)) + if (tool_pmu__read_event(ev, /*evsel=*/NULL, &count)) result = count; else pr_err("Failure to read '%s'", literal); diff --git a/tools/perf/util/ftrace.h b/tools/perf/util/ftrace.h index a9bc47da83a5..950f2efafad2 100644 --- a/tools/perf/util/ftrace.h +++ b/tools/perf/util/ftrace.h @@ -17,6 +17,7 @@ struct perf_ftrace { struct list_head notrace; struct list_head graph_funcs; struct list_head nograph_funcs; + struct list_head event_pair; struct hashmap *profile_hash; unsigned long percpu_buffer_size; bool inherit; @@ -29,6 +30,10 @@ struct perf_ftrace { int graph_depth; int func_stack_trace; int func_irq_info; + int graph_args; + int graph_retval; + int graph_retval_hex; + int graph_retaddr; int graph_nosleep_time; int graph_noirqs; int graph_verbose; diff --git a/tools/perf/util/genelf.c b/tools/perf/util/genelf.c index cdce7f173d00..591548b10e34 100644 --- a/tools/perf/util/genelf.c +++ b/tools/perf/util/genelf.c @@ -12,15 +12,14 @@ #include <libelf.h> #include <string.h> #include <stdlib.h> -#include <unistd.h> #include <inttypes.h> -#include <fcntl.h> #include <err.h> #ifdef HAVE_LIBDW_SUPPORT #include <dwarf.h> #endif #include "genelf.h" +#include "sha1.h" #include "../util/jitdump.h" #include <linux/compiler.h> @@ -28,25 +27,6 @@ #define NT_GNU_BUILD_ID 3 #endif -#define BUILD_ID_URANDOM /* different uuid for each run */ - -#ifdef HAVE_LIBCRYPTO_SUPPORT - -#define BUILD_ID_MD5 -#undef BUILD_ID_SHA /* does not seem to work well when linked with Java */ -#undef BUILD_ID_URANDOM /* different uuid for each run */ - -#ifdef BUILD_ID_SHA -#include <openssl/sha.h> -#endif - -#ifdef BUILD_ID_MD5 -#include <openssl/evp.h> -#include <openssl/md5.h> -#endif -#endif - - typedef struct { unsigned int namesz; /* Size of entry's owner string */ unsigned int descsz; /* Size of the note descriptor */ @@ -71,7 +51,7 @@ static char shd_string_table[] = { static struct buildid_note { Elf_Note desc; /* descsz: size of build-id, must be multiple of 4 */ char name[4]; /* GNU\0 */ - char build_id[20]; + u8 build_id[SHA1_DIGEST_SIZE]; } bnote; static Elf_Sym symtab[]={ @@ -92,65 +72,6 @@ static Elf_Sym symtab[]={ } }; -#ifdef BUILD_ID_URANDOM -static void -gen_build_id(struct buildid_note *note, - unsigned long load_addr __maybe_unused, - const void *code __maybe_unused, - size_t csize __maybe_unused) -{ - int fd; - size_t sz = sizeof(note->build_id); - ssize_t sret; - - fd = open("/dev/urandom", O_RDONLY); - if (fd == -1) - err(1, "cannot access /dev/urandom for buildid"); - - sret = read(fd, note->build_id, sz); - - close(fd); - - if (sret != (ssize_t)sz) - memset(note->build_id, 0, sz); -} -#endif - -#ifdef BUILD_ID_SHA -static void -gen_build_id(struct buildid_note *note, - unsigned long load_addr __maybe_unused, - const void *code, - size_t csize) -{ - if (sizeof(note->build_id) < SHA_DIGEST_LENGTH) - errx(1, "build_id too small for SHA1"); - - SHA1(code, csize, (unsigned char *)note->build_id); -} -#endif - -#ifdef BUILD_ID_MD5 -static void -gen_build_id(struct buildid_note *note, unsigned long load_addr, const void *code, size_t csize) -{ - EVP_MD_CTX *mdctx; - - if (sizeof(note->build_id) < 16) - errx(1, "build_id too small for MD5"); - - mdctx = EVP_MD_CTX_new(); - if (!mdctx) - errx(2, "failed to create EVP_MD_CTX"); - - EVP_DigestInit_ex(mdctx, EVP_md5(), NULL); - EVP_DigestUpdate(mdctx, &load_addr, sizeof(load_addr)); - EVP_DigestUpdate(mdctx, code, csize); - EVP_DigestFinal_ex(mdctx, (unsigned char *)note->build_id, NULL); - EVP_MD_CTX_free(mdctx); -} -#endif - static int jit_add_eh_frame_info(Elf *e, void* unwinding, uint64_t unwinding_header_size, uint64_t unwinding_size, uint64_t base_offset) @@ -239,7 +160,7 @@ jit_add_eh_frame_info(Elf *e, void* unwinding, uint64_t unwinding_header_size, * csize: the code size in bytes */ int -jit_write_elf(int fd, uint64_t load_addr, const char *sym, +jit_write_elf(int fd, uint64_t load_addr __maybe_unused, const char *sym, const void *code, int csize, void *debug __maybe_unused, int nr_debug_entries __maybe_unused, void *unwinding, uint64_t unwinding_header_size, uint64_t unwinding_size) @@ -473,7 +394,7 @@ jit_write_elf(int fd, uint64_t load_addr, const char *sym, /* * build-id generation */ - gen_build_id(&bnote, load_addr, code, csize); + sha1(code, csize, bnote.build_id); bnote.desc.namesz = sizeof(bnote.name); /* must include 0 termination */ bnote.desc.descsz = sizeof(bnote.build_id); bnote.desc.type = NT_GNU_BUILD_ID; diff --git a/tools/perf/util/header.c b/tools/perf/util/header.c index e3cdc3b7b4ab..4f2a6e10ed5c 100644 --- a/tools/perf/util/header.c +++ b/tools/perf/util/header.c @@ -557,6 +557,7 @@ static int write_event_desc(struct feat_fd *ff, static int write_cmdline(struct feat_fd *ff, struct evlist *evlist __maybe_unused) { + struct perf_env *env = &ff->ph->env; char pbuf[MAXPATHLEN], *buf; int i, ret, n; @@ -564,7 +565,7 @@ static int write_cmdline(struct feat_fd *ff, buf = perf_exe(pbuf, MAXPATHLEN); /* account for binary path */ - n = perf_env.nr_cmdline + 1; + n = env->nr_cmdline + 1; ret = do_write(ff, &n, sizeof(n)); if (ret < 0) @@ -574,8 +575,8 @@ static int write_cmdline(struct feat_fd *ff, if (ret < 0) return ret; - for (i = 0 ; i < perf_env.nr_cmdline; i++) { - ret = do_write_string(ff, perf_env.cmdline_argv[i]); + for (i = 0 ; i < env->nr_cmdline; i++) { + ret = do_write_string(ff, env->cmdline_argv[i]); if (ret < 0) return ret; } @@ -586,6 +587,7 @@ static int write_cmdline(struct feat_fd *ff, static int write_cpu_topology(struct feat_fd *ff, struct evlist *evlist __maybe_unused) { + struct perf_env *env = &ff->ph->env; struct cpu_topology *tp; u32 i; int ret, j; @@ -613,17 +615,17 @@ static int write_cpu_topology(struct feat_fd *ff, break; } - ret = perf_env__read_cpu_topology_map(&perf_env); + ret = perf_env__read_cpu_topology_map(env); if (ret < 0) goto done; - for (j = 0; j < perf_env.nr_cpus_avail; j++) { - ret = do_write(ff, &perf_env.cpu[j].core_id, - sizeof(perf_env.cpu[j].core_id)); + for (j = 0; j < env->nr_cpus_avail; j++) { + ret = do_write(ff, &env->cpu[j].core_id, + sizeof(env->cpu[j].core_id)); if (ret < 0) return ret; - ret = do_write(ff, &perf_env.cpu[j].socket_id, - sizeof(perf_env.cpu[j].socket_id)); + ret = do_write(ff, &env->cpu[j].socket_id, + sizeof(env->cpu[j].socket_id)); if (ret < 0) return ret; } @@ -641,9 +643,9 @@ static int write_cpu_topology(struct feat_fd *ff, goto done; } - for (j = 0; j < perf_env.nr_cpus_avail; j++) { - ret = do_write(ff, &perf_env.cpu[j].die_id, - sizeof(perf_env.cpu[j].die_id)); + for (j = 0; j < env->nr_cpus_avail; j++) { + ret = do_write(ff, &env->cpu[j].die_id, + sizeof(env->cpu[j].die_id)); if (ret < 0) return ret; } @@ -1016,10 +1018,13 @@ static int write_bpf_prog_info(struct feat_fd *ff, struct perf_env *env = &ff->ph->env; struct rb_root *root; struct rb_node *next; - int ret; + int ret = 0; down_read(&env->bpf_progs.lock); + if (env->bpf_progs.infos_cnt == 0) + goto out; + ret = do_write(ff, &env->bpf_progs.infos_cnt, sizeof(env->bpf_progs.infos_cnt)); if (ret < 0) @@ -1058,10 +1063,13 @@ static int write_bpf_btf(struct feat_fd *ff, struct perf_env *env = &ff->ph->env; struct rb_root *root; struct rb_node *next; - int ret; + int ret = 0; down_read(&env->bpf_progs.lock); + if (env->bpf_progs.btfs_cnt == 0) + goto out; + ret = do_write(ff, &env->bpf_progs.btfs_cnt, sizeof(env->bpf_progs.btfs_cnt)); @@ -1814,6 +1822,9 @@ static void print_bpf_prog_info(struct feat_fd *ff, FILE *fp) root = &env->bpf_progs.infos; next = rb_first(root); + if (!next) + printf("# bpf_prog_info empty\n"); + while (next) { struct bpf_prog_info_node *node; @@ -1838,6 +1849,9 @@ static void print_bpf_btf(struct feat_fd *ff, FILE *fp) root = &env->bpf_progs.btfs; next = rb_first(root); + if (!next) + printf("# btf info empty\n"); + while (next) { struct btf_node *node; @@ -2111,17 +2125,18 @@ static void print_cpu_pmu_caps(struct feat_fd *ff, FILE *fp) static void print_pmu_caps(struct feat_fd *ff, FILE *fp) { + struct perf_env *env = &ff->ph->env; struct pmu_caps *pmu_caps; - for (int i = 0; i < ff->ph->env.nr_pmus_with_caps; i++) { - pmu_caps = &ff->ph->env.pmu_caps[i]; + for (int i = 0; i < env->nr_pmus_with_caps; i++) { + pmu_caps = &env->pmu_caps[i]; __print_pmu_caps(fp, pmu_caps->nr_caps, pmu_caps->caps, pmu_caps->pmu_name); } - if (strcmp(perf_env__arch(&ff->ph->env), "x86") == 0 && - perf_env__has_pmu_mapping(&ff->ph->env, "ibs_op")) { - char *max_precise = perf_env__find_pmu_cap(&ff->ph->env, "cpu", "max_precise"); + if (strcmp(perf_env__arch(env), "x86") == 0 && + perf_env__has_pmu_mapping(env, "ibs_op")) { + char *max_precise = perf_env__find_pmu_cap(env, "cpu", "max_precise"); if (max_precise != NULL && atoi(max_precise) == 0) fprintf(fp, "# AMD systems uses ibs_op// PMU for some precise events, e.g.: cycles:p, see the 'perf list' man page for further details.\n"); @@ -2130,18 +2145,19 @@ static void print_pmu_caps(struct feat_fd *ff, FILE *fp) static void print_pmu_mappings(struct feat_fd *ff, FILE *fp) { + struct perf_env *env = &ff->ph->env; const char *delimiter = "# pmu mappings: "; char *str, *tmp; u32 pmu_num; u32 type; - pmu_num = ff->ph->env.nr_pmu_mappings; + pmu_num = env->nr_pmu_mappings; if (!pmu_num) { fprintf(fp, "# pmu mappings: not available\n"); return; } - str = ff->ph->env.pmu_mappings; + str = env->pmu_mappings; while (pmu_num) { type = strtoul(str, &tmp, 0); @@ -2223,17 +2239,18 @@ static void memory_node__fprintf(struct memory_node *n, static void print_mem_topology(struct feat_fd *ff, FILE *fp) { + struct perf_env *env = &ff->ph->env; struct memory_node *nodes; int i, nr; - nodes = ff->ph->env.memory_nodes; - nr = ff->ph->env.nr_memory_nodes; + nodes = env->memory_nodes; + nr = env->nr_memory_nodes; fprintf(fp, "# memory nodes (nr %d, block size 0x%llx):\n", - nr, ff->ph->env.memory_bsize); + nr, env->memory_bsize); for (i = 0; i < nr; i++) { - memory_node__fprintf(&nodes[i], ff->ph->env.memory_bsize, fp); + memory_node__fprintf(&nodes[i], env->memory_bsize, fp); } } @@ -2291,7 +2308,7 @@ static int __event_process_build_id(struct perf_record_header_build_id *bev, free(m.name); } - build_id__sprintf(dso__bid(dso), sbuild_id); + build_id__snprintf(dso__bid(dso), sbuild_id, sizeof(sbuild_id)); pr_debug("build id event received for %s: %s [%zu]\n", dso__long_name(dso), sbuild_id, size); dso__put(dso); @@ -2431,6 +2448,7 @@ static int process_build_id(struct feat_fd *ff, void *data __maybe_unused) static int process_nrcpus(struct feat_fd *ff, void *data __maybe_unused) { + struct perf_env *env = &ff->ph->env; int ret; u32 nr_cpus_avail, nr_cpus_online; @@ -2441,20 +2459,21 @@ static int process_nrcpus(struct feat_fd *ff, void *data __maybe_unused) ret = do_read_u32(ff, &nr_cpus_online); if (ret) return ret; - ff->ph->env.nr_cpus_avail = (int)nr_cpus_avail; - ff->ph->env.nr_cpus_online = (int)nr_cpus_online; + env->nr_cpus_avail = (int)nr_cpus_avail; + env->nr_cpus_online = (int)nr_cpus_online; return 0; } static int process_total_mem(struct feat_fd *ff, void *data __maybe_unused) { + struct perf_env *env = &ff->ph->env; u64 total_mem; int ret; ret = do_read_u64(ff, &total_mem); if (ret) return -1; - ff->ph->env.total_mem = (unsigned long long)total_mem; + env->total_mem = (unsigned long long)total_mem; return 0; } @@ -2515,13 +2534,14 @@ process_event_desc(struct feat_fd *ff, void *data __maybe_unused) static int process_cmdline(struct feat_fd *ff, void *data __maybe_unused) { + struct perf_env *env = &ff->ph->env; char *str, *cmdline = NULL, **argv = NULL; u32 nr, i, len = 0; if (do_read_u32(ff, &nr)) return -1; - ff->ph->env.nr_cmdline = nr; + env->nr_cmdline = nr; cmdline = zalloc(ff->size + nr + 1); if (!cmdline) @@ -2541,8 +2561,8 @@ static int process_cmdline(struct feat_fd *ff, void *data __maybe_unused) len += strlen(str) + 1; free(str); } - ff->ph->env.cmdline = cmdline; - ff->ph->env.cmdline_argv = (const char **) argv; + env->cmdline = cmdline; + env->cmdline_argv = (const char **) argv; return 0; error: @@ -2556,19 +2576,18 @@ static int process_cpu_topology(struct feat_fd *ff, void *data __maybe_unused) u32 nr, i; char *str = NULL; struct strbuf sb; - int cpu_nr = ff->ph->env.nr_cpus_avail; + struct perf_env *env = &ff->ph->env; + int cpu_nr = env->nr_cpus_avail; u64 size = 0; - struct perf_header *ph = ff->ph; - bool do_core_id_test = true; - ph->env.cpu = calloc(cpu_nr, sizeof(*ph->env.cpu)); - if (!ph->env.cpu) + env->cpu = calloc(cpu_nr, sizeof(*env->cpu)); + if (!env->cpu) return -1; if (do_read_u32(ff, &nr)) goto free_cpu; - ph->env.nr_sibling_cores = nr; + env->nr_sibling_cores = nr; size += sizeof(u32); if (strbuf_init(&sb, 128) < 0) goto free_cpu; @@ -2584,12 +2603,12 @@ static int process_cpu_topology(struct feat_fd *ff, void *data __maybe_unused) size += string_size(str); zfree(&str); } - ph->env.sibling_cores = strbuf_detach(&sb, NULL); + env->sibling_cores = strbuf_detach(&sb, NULL); if (do_read_u32(ff, &nr)) return -1; - ph->env.nr_sibling_threads = nr; + env->nr_sibling_threads = nr; size += sizeof(u32); for (i = 0; i < nr; i++) { @@ -2603,43 +2622,28 @@ static int process_cpu_topology(struct feat_fd *ff, void *data __maybe_unused) size += string_size(str); zfree(&str); } - ph->env.sibling_threads = strbuf_detach(&sb, NULL); + env->sibling_threads = strbuf_detach(&sb, NULL); /* * The header may be from old perf, * which doesn't include core id and socket id information. */ if (ff->size <= size) { - zfree(&ph->env.cpu); + zfree(&env->cpu); return 0; } - /* On s390 the socket_id number is not related to the numbers of cpus. - * The socket_id number might be higher than the numbers of cpus. - * This depends on the configuration. - * AArch64 is the same. - */ - if (ph->env.arch && (!strncmp(ph->env.arch, "s390", 4) - || !strncmp(ph->env.arch, "aarch64", 7))) - do_core_id_test = false; - for (i = 0; i < (u32)cpu_nr; i++) { if (do_read_u32(ff, &nr)) goto free_cpu; - ph->env.cpu[i].core_id = nr; + env->cpu[i].core_id = nr; size += sizeof(u32); if (do_read_u32(ff, &nr)) goto free_cpu; - if (do_core_id_test && nr != (u32)-1 && nr > (u32)cpu_nr) { - pr_debug("socket_id number is too big." - "You may need to upgrade the perf tool.\n"); - goto free_cpu; - } - - ph->env.cpu[i].socket_id = nr; + env->cpu[i].socket_id = nr; size += sizeof(u32); } @@ -2653,7 +2657,7 @@ static int process_cpu_topology(struct feat_fd *ff, void *data __maybe_unused) if (do_read_u32(ff, &nr)) return -1; - ph->env.nr_sibling_dies = nr; + env->nr_sibling_dies = nr; size += sizeof(u32); for (i = 0; i < nr; i++) { @@ -2667,13 +2671,13 @@ static int process_cpu_topology(struct feat_fd *ff, void *data __maybe_unused) size += string_size(str); zfree(&str); } - ph->env.sibling_dies = strbuf_detach(&sb, NULL); + env->sibling_dies = strbuf_detach(&sb, NULL); for (i = 0; i < (u32)cpu_nr; i++) { if (do_read_u32(ff, &nr)) goto free_cpu; - ph->env.cpu[i].die_id = nr; + env->cpu[i].die_id = nr; } return 0; @@ -2682,12 +2686,13 @@ error: strbuf_release(&sb); zfree(&str); free_cpu: - zfree(&ph->env.cpu); + zfree(&env->cpu); return -1; } static int process_numa_topology(struct feat_fd *ff, void *data __maybe_unused) { + struct perf_env *env = &ff->ph->env; struct numa_node *nodes, *n; u32 nr, i; char *str; @@ -2722,8 +2727,8 @@ static int process_numa_topology(struct feat_fd *ff, void *data __maybe_unused) if (!n->map) goto error; } - ff->ph->env.nr_numa_nodes = nr; - ff->ph->env.numa_nodes = nodes; + env->nr_numa_nodes = nr; + env->numa_nodes = nodes; return 0; error: @@ -2733,6 +2738,7 @@ error: static int process_pmu_mappings(struct feat_fd *ff, void *data __maybe_unused) { + struct perf_env *env = &ff->ph->env; char *name; u32 pmu_num; u32 type; @@ -2746,7 +2752,7 @@ static int process_pmu_mappings(struct feat_fd *ff, void *data __maybe_unused) return 0; } - ff->ph->env.nr_pmu_mappings = pmu_num; + env->nr_pmu_mappings = pmu_num; if (strbuf_init(&sb, 128) < 0) return -1; @@ -2765,14 +2771,14 @@ static int process_pmu_mappings(struct feat_fd *ff, void *data __maybe_unused) goto error; if (!strcmp(name, "msr")) - ff->ph->env.msr_pmu_type = type; + env->msr_pmu_type = type; free(name); pmu_num--; } /* AMD may set it by evlist__has_amd_ibs() from perf_session__new() */ - free(ff->ph->env.pmu_mappings); - ff->ph->env.pmu_mappings = strbuf_detach(&sb, NULL); + free(env->pmu_mappings); + env->pmu_mappings = strbuf_detach(&sb, NULL); return 0; error: @@ -2782,6 +2788,7 @@ error: static int process_group_desc(struct feat_fd *ff, void *data __maybe_unused) { + struct perf_env *env = &ff->ph->env; size_t ret = -1; u32 i, nr, nr_groups; struct perf_session *session; @@ -2795,7 +2802,7 @@ static int process_group_desc(struct feat_fd *ff, void *data __maybe_unused) if (do_read_u32(ff, &nr_groups)) return -1; - ff->ph->env.nr_groups = nr_groups; + env->nr_groups = nr_groups; if (!nr_groups) { pr_debug("group desc not available\n"); return 0; @@ -2879,6 +2886,7 @@ static int process_auxtrace(struct feat_fd *ff, void *data __maybe_unused) static int process_cache(struct feat_fd *ff, void *data __maybe_unused) { + struct perf_env *env = &ff->ph->env; struct cpu_cache_level *caches; u32 cnt, i, version; @@ -2919,8 +2927,8 @@ static int process_cache(struct feat_fd *ff, void *data __maybe_unused) #undef _R } - ff->ph->env.caches = caches; - ff->ph->env.caches_cnt = cnt; + env->caches = caches; + env->caches_cnt = cnt; return 0; out_free_caches: for (i = 0; i < cnt; i++) { @@ -2956,6 +2964,7 @@ static int process_sample_time(struct feat_fd *ff, void *data __maybe_unused) static int process_mem_topology(struct feat_fd *ff, void *data __maybe_unused) { + struct perf_env *env = &ff->ph->env; struct memory_node *nodes; u64 version, i, nr, bsize; int ret = -1; @@ -2994,9 +3003,9 @@ static int process_mem_topology(struct feat_fd *ff, nodes[i] = n; } - ff->ph->env.memory_bsize = bsize; - ff->ph->env.memory_nodes = nodes; - ff->ph->env.nr_memory_nodes = nr; + env->memory_bsize = bsize; + env->memory_nodes = nodes; + env->nr_memory_nodes = nr; ret = 0; out: @@ -3008,7 +3017,9 @@ out: static int process_clockid(struct feat_fd *ff, void *data __maybe_unused) { - if (do_read_u64(ff, &ff->ph->env.clock.clockid_res_ns)) + struct perf_env *env = &ff->ph->env; + + if (do_read_u64(ff, &env->clock.clockid_res_ns)) return -1; return 0; @@ -3017,6 +3028,7 @@ static int process_clockid(struct feat_fd *ff, static int process_clock_data(struct feat_fd *ff, void *_data __maybe_unused) { + struct perf_env *env = &ff->ph->env; u32 data32; u64 data64; @@ -3031,26 +3043,27 @@ static int process_clock_data(struct feat_fd *ff, if (do_read_u32(ff, &data32)) return -1; - ff->ph->env.clock.clockid = data32; + env->clock.clockid = data32; /* TOD ref time */ if (do_read_u64(ff, &data64)) return -1; - ff->ph->env.clock.tod_ns = data64; + env->clock.tod_ns = data64; /* clockid ref time */ if (do_read_u64(ff, &data64)) return -1; - ff->ph->env.clock.clockid_ns = data64; - ff->ph->env.clock.enabled = true; + env->clock.clockid_ns = data64; + env->clock.enabled = true; return 0; } static int process_hybrid_topology(struct feat_fd *ff, void *data __maybe_unused) { + struct perf_env *env = &ff->ph->env; struct hybrid_node *nodes, *n; u32 nr, i; @@ -3074,8 +3087,8 @@ static int process_hybrid_topology(struct feat_fd *ff, goto error; } - ff->ph->env.nr_hybrid_nodes = nr; - ff->ph->env.hybrid_nodes = nodes; + env->nr_hybrid_nodes = nr; + env->hybrid_nodes = nodes; return 0; error: @@ -3161,6 +3174,7 @@ static int process_bpf_prog_info(struct feat_fd *ff, void *data __maybe_unused) /* after reading from file, translate offset to address */ bpil_offs_to_addr(info_linear); info_node->info_linear = info_linear; + info_node->metadata = NULL; if (!__perf_env__insert_bpf_prog_info(env, info_node)) { free(info_linear); free(info_node); @@ -3227,19 +3241,21 @@ out: static int process_compressed(struct feat_fd *ff, void *data __maybe_unused) { - if (do_read_u32(ff, &(ff->ph->env.comp_ver))) + struct perf_env *env = &ff->ph->env; + + if (do_read_u32(ff, &(env->comp_ver))) return -1; - if (do_read_u32(ff, &(ff->ph->env.comp_type))) + if (do_read_u32(ff, &(env->comp_type))) return -1; - if (do_read_u32(ff, &(ff->ph->env.comp_level))) + if (do_read_u32(ff, &(env->comp_level))) return -1; - if (do_read_u32(ff, &(ff->ph->env.comp_ratio))) + if (do_read_u32(ff, &(env->comp_ratio))) return -1; - if (do_read_u32(ff, &(ff->ph->env.comp_mmap_len))) + if (do_read_u32(ff, &(env->comp_mmap_len))) return -1; return 0; @@ -3311,19 +3327,21 @@ error: static int process_cpu_pmu_caps(struct feat_fd *ff, void *data __maybe_unused) { - int ret = __process_pmu_caps(ff, &ff->ph->env.nr_cpu_pmu_caps, - &ff->ph->env.cpu_pmu_caps, - &ff->ph->env.max_branches, - &ff->ph->env.br_cntr_nr, - &ff->ph->env.br_cntr_width); + struct perf_env *env = &ff->ph->env; + int ret = __process_pmu_caps(ff, &env->nr_cpu_pmu_caps, + &env->cpu_pmu_caps, + &env->max_branches, + &env->br_cntr_nr, + &env->br_cntr_width); - if (!ret && !ff->ph->env.cpu_pmu_caps) + if (!ret && !env->cpu_pmu_caps) pr_debug("cpu pmu capabilities not available\n"); return ret; } static int process_pmu_caps(struct feat_fd *ff, void *data __maybe_unused) { + struct perf_env *env = &ff->ph->env; struct pmu_caps *pmu_caps; u32 nr_pmu, i; int ret; @@ -3361,8 +3379,8 @@ static int process_pmu_caps(struct feat_fd *ff, void *data __maybe_unused) } } - ff->ph->env.nr_pmus_with_caps = nr_pmu; - ff->ph->env.pmu_caps = pmu_caps; + env->nr_pmus_with_caps = nr_pmu; + env->pmu_caps = pmu_caps; return 0; err: @@ -3660,6 +3678,7 @@ static int perf_session__do_write_header(struct perf_session *session, struct perf_header *header = &session->header; struct evsel *evsel; struct feat_fd ff = { + .ph = header, .fd = fd, }; u64 attr_offset = sizeof(f_header), attr_size = 0; @@ -4228,7 +4247,7 @@ int perf_session__read_header(struct perf_session *session) if (session->evlist == NULL) return -ENOMEM; - session->evlist->env = &header->env; + session->evlist->session = session; session->machines.host.env = &header->env; /* @@ -4341,12 +4360,12 @@ out_delete_evlist: int perf_event__process_feature(struct perf_session *session, union perf_event *event) { - const struct perf_tool *tool = session->tool; struct feat_fd ff = { .fd = 0 }; struct perf_record_header_feature *fe = (struct perf_record_header_feature *)event; int type = fe->header.type; u64 feat = fe->feat_id; int ret = 0; + bool print = dump_trace; if (type < 0 || type >= PERF_RECORD_HEADER_MAX) { pr_warning("invalid record type %d in pipe-mode\n", type); @@ -4357,28 +4376,35 @@ int perf_event__process_feature(struct perf_session *session, return -1; } - if (!feat_ops[feat].process) - return 0; - ff.buf = (void *)fe->data; ff.size = event->header.size - sizeof(*fe); ff.ph = &session->header; - if (feat_ops[feat].process(&ff, NULL)) { + if (feat_ops[feat].process && feat_ops[feat].process(&ff, NULL)) { ret = -1; goto out; } - if (!feat_ops[feat].print || !tool->show_feat_hdr) - goto out; + if (session->tool->show_feat_hdr) { + if (!feat_ops[feat].full_only || + session->tool->show_feat_hdr >= SHOW_FEAT_HEADER_FULL_INFO) { + print = true; + } else { + fprintf(stdout, "# %s info available, use -I to display\n", + feat_ops[feat].name); + } + } - if (!feat_ops[feat].full_only || - tool->show_feat_hdr >= SHOW_FEAT_HEADER_FULL_INFO) { - feat_ops[feat].print(&ff, stdout); - } else { - fprintf(stdout, "# %s info available, use -I to display\n", - feat_ops[feat].name); + if (dump_trace) + printf(", "); + + if (print) { + if (feat_ops[feat].print) + feat_ops[feat].print(&ff, stdout); + else + printf("# %s", feat_ops[feat].name); } + out: free_event_desc(ff.events); return ret; @@ -4420,6 +4446,11 @@ size_t perf_event__fprintf_event_update(union perf_event *event, FILE *fp) return ret; } +size_t perf_event__fprintf_attr(union perf_event *event, FILE *fp) +{ + return perf_event_attr__fprintf(fp, &event->attr.attr, __desc_attr__fprintf, NULL); +} + int perf_event__process_attr(const struct perf_tool *tool __maybe_unused, union perf_event *event, struct evlist **pevlist) @@ -4429,6 +4460,9 @@ int perf_event__process_attr(const struct perf_tool *tool __maybe_unused, struct evsel *evsel; struct evlist *evlist = *pevlist; + if (dump_trace) + perf_event__fprintf_attr(event, stdout); + if (evlist == NULL) { *pevlist = evlist = evlist__new(); if (evlist == NULL) @@ -4495,8 +4529,8 @@ int perf_event__process_event_update(const struct perf_tool *tool __maybe_unused case PERF_EVENT_UPDATE__CPUS: map = cpu_map__new_data(&ev->cpus.cpus); if (map) { - perf_cpu_map__put(evsel->core.own_cpus); - evsel->core.own_cpus = map; + perf_cpu_map__put(evsel->core.pmu_cpus); + evsel->core.pmu_cpus = map; } else pr_err("failed to get event_update cpus\n"); default: diff --git a/tools/perf/util/header.h b/tools/perf/util/header.h index 5201af6305f4..d16dfceccd74 100644 --- a/tools/perf/util/header.h +++ b/tools/perf/util/header.h @@ -175,6 +175,7 @@ int perf_event__process_attr(const struct perf_tool *tool, union perf_event *eve int perf_event__process_event_update(const struct perf_tool *tool, union perf_event *event, struct evlist **pevlist); +size_t perf_event__fprintf_attr(union perf_event *event, FILE *fp); size_t perf_event__fprintf_event_update(union perf_event *event, FILE *fp); #ifdef HAVE_LIBTRACEEVENT int perf_event__process_tracing_data(struct perf_session *session, diff --git a/tools/perf/util/hist.c b/tools/perf/util/hist.c index afc6855327ab..64ff427040c3 100644 --- a/tools/perf/util/hist.c +++ b/tools/perf/util/hist.c @@ -829,7 +829,7 @@ __hists__add_entry(struct hists *hists, .period = sample->period, .weight1 = sample->weight, .weight2 = sample->ins_lat, - .weight3 = sample->p_stage_cyc, + .weight3 = sample->weight3, .latency = al->latency, }, .parent = sym_parent, @@ -846,7 +846,7 @@ __hists__add_entry(struct hists *hists, .time = hist_time(sample->time), .weight = sample->weight, .ins_lat = sample->ins_lat, - .p_stage_cyc = sample->p_stage_cyc, + .weight3 = sample->weight3, .simd_flags = sample->simd_flags, }, *he = hists__findnew_entry(hists, &entry, al, sample_self); diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h index c64254088fc7..70438d03ca9c 100644 --- a/tools/perf/util/hist.h +++ b/tools/perf/util/hist.h @@ -255,7 +255,8 @@ struct hist_entry { u64 code_page_size; u64 weight; u64 ins_lat; - u64 p_stage_cyc; + /** @weight3: On x86 holds retire_lat, on powerpc holds p_stage_cyc. */ + u64 weight3; s32 socket; s32 cpu; int parallelism; diff --git a/tools/perf/util/hwmon_pmu.c b/tools/perf/util/hwmon_pmu.c index c25e7296f1c1..416dfea9ffff 100644 --- a/tools/perf/util/hwmon_pmu.c +++ b/tools/perf/util/hwmon_pmu.c @@ -104,7 +104,7 @@ static const char *const hwmon_units[HWMON_TYPE_MAX] = { struct hwmon_pmu { struct perf_pmu pmu; struct hashmap events; - int hwmon_dir_fd; + char *hwmon_dir; }; /** @@ -245,7 +245,7 @@ static int hwmon_pmu__read_events(struct hwmon_pmu *pmu) return 0; /* Use openat so that the directory contents are refreshed. */ - io_dir__init(&dir, openat(pmu->hwmon_dir_fd, ".", O_CLOEXEC | O_DIRECTORY | O_RDONLY)); + io_dir__init(&dir, open(pmu->hwmon_dir, O_CLOEXEC | O_DIRECTORY | O_RDONLY)); if (dir.dirfd < 0) return -ENOENT; @@ -283,7 +283,7 @@ static int hwmon_pmu__read_events(struct hwmon_pmu *pmu) __set_bit(item, alarm ? value->alarm_items : value->items); if (item == HWMON_ITEM_LABEL) { char buf[128]; - int fd = openat(pmu->hwmon_dir_fd, ent->d_name, O_RDONLY); + int fd = openat(dir.dirfd, ent->d_name, O_RDONLY); ssize_t read_len; if (fd < 0) @@ -342,9 +342,10 @@ err_out: return err; } -struct perf_pmu *hwmon_pmu__new(struct list_head *pmus, int hwmon_dir, const char *sysfs_name, const char *name) +struct perf_pmu *hwmon_pmu__new(struct list_head *pmus, const char *hwmon_dir, + const char *sysfs_name, const char *name) { - char buf[32]; + char buf[64]; struct hwmon_pmu *hwm; __u32 type = PERF_PMU_TYPE_HWMON_START + strtoul(sysfs_name + 5, NULL, 10); @@ -365,7 +366,11 @@ struct perf_pmu *hwmon_pmu__new(struct list_head *pmus, int hwmon_dir, const cha return NULL; } - hwm->hwmon_dir_fd = hwmon_dir; + hwm->hwmon_dir = strdup(hwmon_dir); + if (!hwm->hwmon_dir) { + perf_pmu__delete(&hwm->pmu); + return NULL; + } hwm->pmu.alias_name = strdup(sysfs_name); if (!hwm->pmu.alias_name) { perf_pmu__delete(&hwm->pmu); @@ -399,7 +404,7 @@ void hwmon_pmu__exit(struct perf_pmu *pmu) free(value); } hashmap__clear(&hwm->events); - close(hwm->hwmon_dir_fd); + zfree(&hwm->hwmon_dir); } static size_t hwmon_pmu__describe_items(struct hwmon_pmu *hwm, char *out_buf, size_t out_buf_len, @@ -409,6 +414,10 @@ static size_t hwmon_pmu__describe_items(struct hwmon_pmu *hwm, char *out_buf, si size_t bit; char buf[64]; size_t len = 0; + int dir = open(hwm->hwmon_dir, O_CLOEXEC | O_DIRECTORY | O_RDONLY); + + if (dir < 0) + return 0; for_each_set_bit(bit, items, HWMON_ITEM__MAX) { int fd; @@ -421,7 +430,7 @@ static size_t hwmon_pmu__describe_items(struct hwmon_pmu *hwm, char *out_buf, si key.num, hwmon_item_strs[bit], is_alarm ? "_alarm" : ""); - fd = openat(hwm->hwmon_dir_fd, buf, O_RDONLY); + fd = openat(dir, buf, O_RDONLY); if (fd > 0) { ssize_t read_len = read(fd, buf, sizeof(buf)); @@ -443,6 +452,7 @@ static size_t hwmon_pmu__describe_items(struct hwmon_pmu *hwm, char *out_buf, si close(fd); } } + close(dir); return len; } @@ -712,6 +722,7 @@ int perf_pmus__read_hwmon_pmus(struct list_head *pmus) size_t line_len; int hwmon_dir, name_fd; struct io io; + char buf2[128]; if (class_hwmon_ent->d_type != DT_LNK) continue; @@ -730,12 +741,13 @@ int perf_pmus__read_hwmon_pmus(struct list_head *pmus) close(hwmon_dir); continue; } - io__init(&io, name_fd, buf, sizeof(buf)); + io__init(&io, name_fd, buf2, sizeof(buf2)); io__getline(&io, &line, &line_len); if (line_len > 0 && line[line_len - 1] == '\n') line[line_len - 1] = '\0'; - hwmon_pmu__new(pmus, hwmon_dir, class_hwmon_ent->d_name, line); + hwmon_pmu__new(pmus, buf, class_hwmon_ent->d_name, line); close(name_fd); + close(hwmon_dir); } free(line); close(class_hwmon_dir.dirfd); @@ -753,6 +765,10 @@ int evsel__hwmon_pmu_open(struct evsel *evsel, .type_and_num = evsel->core.attr.config, }; int idx = 0, thread = 0, nthreads, err = 0; + int dir = open(hwm->hwmon_dir, O_CLOEXEC | O_DIRECTORY | O_RDONLY); + + if (dir < 0) + return -errno; nthreads = perf_thread_map__nr(threads); for (idx = start_cpu_map_idx; idx < end_cpu_map_idx; idx++) { @@ -763,7 +779,7 @@ int evsel__hwmon_pmu_open(struct evsel *evsel, snprintf(buf, sizeof(buf), "%s%d_input", hwmon_type_strs[key.type], key.num); - fd = openat(hwm->hwmon_dir_fd, buf, O_RDONLY); + fd = openat(dir, buf, O_RDONLY); FD(evsel, idx, thread) = fd; if (fd < 0) { err = -errno; @@ -771,6 +787,7 @@ int evsel__hwmon_pmu_open(struct evsel *evsel, } } } + close(dir); return 0; out_close: if (err) @@ -784,6 +801,7 @@ out_close: } thread = nthreads; } while (--idx >= 0); + close(dir); return err; } diff --git a/tools/perf/util/hwmon_pmu.h b/tools/perf/util/hwmon_pmu.h index b3329774d2b2..dc711b289ff5 100644 --- a/tools/perf/util/hwmon_pmu.h +++ b/tools/perf/util/hwmon_pmu.h @@ -135,14 +135,14 @@ bool parse_hwmon_filename(const char *filename, * hwmon_pmu__new() - Allocate and construct a hwmon PMU. * * @pmus: The list of PMUs to be added to. - * @hwmon_dir: An O_DIRECTORY file descriptor for a hwmon directory. + * @hwmon_dir: The path to a hwmon directory. * @sysfs_name: Name of the hwmon sysfs directory like hwmon0. * @name: The contents of the "name" file in the hwmon directory. * * Exposed for testing. Regular construction should happen via * perf_pmus__read_hwmon_pmus. */ -struct perf_pmu *hwmon_pmu__new(struct list_head *pmus, int hwmon_dir, +struct perf_pmu *hwmon_pmu__new(struct list_head *pmus, const char *hwmon_dir, const char *sysfs_name, const char *name); void hwmon_pmu__exit(struct perf_pmu *pmu); diff --git a/tools/perf/util/include/linux/linkage.h b/tools/perf/util/include/linux/linkage.h index 178b00205fe6..89979ca23c3f 100644 --- a/tools/perf/util/include/linux/linkage.h +++ b/tools/perf/util/include/linux/linkage.h @@ -132,4 +132,8 @@ SYM_TYPED_START(name, SYM_L_GLOBAL, SYM_A_ALIGN) #endif +#ifndef SYM_PIC_ALIAS +#define SYM_PIC_ALIAS(sym) SYM_ALIAS(__pi_ ## sym, sym, SYM_T_FUNC, SYM_L_GLOBAL) +#endif + #endif /* PERF_LINUX_LINKAGE_H_ */ diff --git a/tools/perf/util/intel-tpebs.c b/tools/perf/util/intel-tpebs.c index 3b92ebf5c112..8c9aee157ec4 100644 --- a/tools/perf/util/intel-tpebs.c +++ b/tools/perf/util/intel-tpebs.c @@ -210,8 +210,8 @@ static int process_sample_event(const struct perf_tool *tool __maybe_unused, * latency value will be used. Save the number of samples and the sum of * retire latency value for each event. */ - t->last = sample->retire_lat; - update_stats(&t->stats, sample->retire_lat); + t->last = sample->weight3; + update_stats(&t->stats, sample->weight3); mutex_unlock(tpebs_mtx_get()); return 0; } diff --git a/tools/perf/util/jitdump.c b/tools/perf/util/jitdump.c index 624964f01b5f..b062b1f234b6 100644 --- a/tools/perf/util/jitdump.c +++ b/tools/perf/util/jitdump.c @@ -14,9 +14,9 @@ #include <sys/mman.h> #include <linux/stringify.h> -#include "build-id.h" #include "event.h" #include "debug.h" +#include "dso.h" #include "evlist.h" #include "namespaces.h" #include "symbol.h" @@ -531,9 +531,22 @@ static int jit_repipe_code_load(struct jit_buf_desc *jd, union jr_entry *jr) /* * mark dso as use to generate buildid in the header */ - if (!ret) - build_id__mark_dso_hit(tool, event, &sample, NULL, jd->machine); - + if (!ret) { + struct dso_id dso_id = { + { + .maj = event->mmap2.maj, + .min = event->mmap2.min, + .ino = event->mmap2.ino, + .ino_generation = event->mmap2.ino_generation, + }, + .mmap2_valid = true, + .mmap2_ino_generation_valid = true, + }; + struct dso *dso = machine__findnew_dso_id(jd->machine, filename, &dso_id); + + if (dso) + dso__set_hit(dso); + } out: perf_sample__exit(&sample); free(event); diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 7ec12c207970..b5dd42588c91 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -129,7 +129,7 @@ out: return 0; } -static struct machine *__machine__new_host(bool kernel_maps) +static struct machine *__machine__new_host(struct perf_env *host_env, bool kernel_maps) { struct machine *machine = malloc(sizeof(*machine)); @@ -142,13 +142,13 @@ static struct machine *__machine__new_host(bool kernel_maps) free(machine); return NULL; } - machine->env = &perf_env; + machine->env = host_env; return machine; } -struct machine *machine__new_host(void) +struct machine *machine__new_host(struct perf_env *host_env) { - return __machine__new_host(/*kernel_maps=*/true); + return __machine__new_host(host_env, /*kernel_maps=*/true); } static int mmap_handler(const struct perf_tool *tool __maybe_unused, @@ -168,9 +168,9 @@ static int machine__init_live(struct machine *machine, pid_t pid) mmap_handler, machine, true); } -struct machine *machine__new_live(bool kernel_maps, pid_t pid) +struct machine *machine__new_live(struct perf_env *host_env, bool kernel_maps, pid_t pid) { - struct machine *machine = __machine__new_host(kernel_maps); + struct machine *machine = __machine__new_host(host_env, kernel_maps); if (!machine) return NULL; @@ -182,9 +182,9 @@ struct machine *machine__new_live(bool kernel_maps, pid_t pid) return machine; } -struct machine *machine__new_kallsyms(void) +struct machine *machine__new_kallsyms(struct perf_env *host_env) { - struct machine *machine = machine__new_host(); + struct machine *machine = machine__new_host(host_env); /* * FIXME: * 1) We should switch to machine__load_kallsyms(), i.e. not explicitly @@ -1731,21 +1731,21 @@ int machine__process_mmap2_event(struct machine *machine, { struct thread *thread; struct map *map; - struct dso_id dso_id = { - .maj = event->mmap2.maj, - .min = event->mmap2.min, - .ino = event->mmap2.ino, - .ino_generation = event->mmap2.ino_generation, - }; - struct build_id __bid, *bid = NULL; + struct dso_id dso_id = dso_id_empty; int ret = 0; if (dump_trace) perf_event__fprintf_mmap2(event, stdout); if (event->header.misc & PERF_RECORD_MISC_MMAP_BUILD_ID) { - bid = &__bid; - build_id__init(bid, event->mmap2.build_id, event->mmap2.build_id_size); + build_id__init(&dso_id.build_id, event->mmap2.build_id, event->mmap2.build_id_size); + } else { + dso_id.maj = event->mmap2.maj; + dso_id.min = event->mmap2.min; + dso_id.ino = event->mmap2.ino; + dso_id.ino_generation = event->mmap2.ino_generation; + dso_id.mmap2_valid = true; + dso_id.mmap2_ino_generation_valid = true; } if (sample->cpumode == PERF_RECORD_MISC_GUEST_KERNEL || @@ -1757,7 +1757,7 @@ int machine__process_mmap2_event(struct machine *machine, }; strlcpy(xm.name, event->mmap2.filename, KMAP_NAME_LEN); - ret = machine__process_kernel_mmap_event(machine, &xm, bid); + ret = machine__process_kernel_mmap_event(machine, &xm, &dso_id.build_id); if (ret < 0) goto out_problem; return 0; @@ -1771,7 +1771,7 @@ int machine__process_mmap2_event(struct machine *machine, map = map__new(machine, event->mmap2.start, event->mmap2.len, event->mmap2.pgoff, &dso_id, event->mmap2.prot, - event->mmap2.flags, bid, + event->mmap2.flags, event->mmap2.filename, thread); if (map == NULL) @@ -1829,8 +1829,8 @@ int machine__process_mmap_event(struct machine *machine, union perf_event *event prot = PROT_EXEC; map = map__new(machine, event->mmap.start, - event->mmap.len, event->mmap.pgoff, - NULL, prot, 0, NULL, event->mmap.filename, thread); + event->mmap.len, event->mmap.pgoff, + &dso_id_empty, prot, /*flags=*/0, event->mmap.filename, thread); if (map == NULL) goto out_problem_map; @@ -3192,7 +3192,7 @@ struct dso *machine__findnew_dso_id(struct machine *machine, const char *filenam struct dso *machine__findnew_dso(struct machine *machine, const char *filename) { - return machine__findnew_dso_id(machine, filename, NULL); + return machine__findnew_dso_id(machine, filename, &dso_id_empty); } char *machine__resolve_kernel_addr(void *vmachine, unsigned long long *addrp, char **modp) diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h index 180b369c366c..22a42c5825fa 100644 --- a/tools/perf/util/machine.h +++ b/tools/perf/util/machine.h @@ -169,9 +169,9 @@ struct thread *machine__findnew_guest_code(struct machine *machine, pid_t pid); void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size); void machines__set_comm_exec(struct machines *machines, bool comm_exec); -struct machine *machine__new_host(void); -struct machine *machine__new_kallsyms(void); -struct machine *machine__new_live(bool kernel_maps, pid_t pid); +struct machine *machine__new_host(struct perf_env *host_env); +struct machine *machine__new_kallsyms(struct perf_env *host_env); +struct machine *machine__new_live(struct perf_env *host_env, bool kernel_maps, pid_t pid); int machine__init(struct machine *machine, const char *root_dir, pid_t pid); void machine__exit(struct machine *machine); void machine__delete_threads(struct machine *machine); diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c index d729438b7d65..b46c68c24d1c 100644 --- a/tools/perf/util/map.c +++ b/tools/perf/util/map.c @@ -120,8 +120,8 @@ static void map__init(struct map *map, u64 start, u64 end, u64 pgoff, } struct map *map__new(struct machine *machine, u64 start, u64 len, - u64 pgoff, struct dso_id *id, - u32 prot, u32 flags, struct build_id *bid, + u64 pgoff, const struct dso_id *id, + u32 prot, u32 flags, char *filename, struct thread *thread) { struct map *result; @@ -132,7 +132,7 @@ struct map *map__new(struct machine *machine, u64 start, u64 len, map = zalloc(sizeof(*map)); if (ADD_RC_CHK(result, map)) { char newfilename[PATH_MAX]; - struct dso *dso, *header_bid_dso; + struct dso *dso; int anon, no_dso, vdso, android; android = is_android_lib(filename); @@ -189,16 +189,15 @@ struct map *map__new(struct machine *machine, u64 start, u64 len, dso__set_nsinfo(dso, nsi); mutex_unlock(dso__lock(dso)); - if (build_id__is_defined(bid)) { - dso__set_build_id(dso, bid); - } else { + if (!build_id__is_defined(&id->build_id)) { /* * If the mmap event had no build ID, search for an existing dso from the * build ID header by name. Otherwise only the dso loaded at the time of * reading the header will have the build ID set and all future mmaps will * have it missing. */ - header_bid_dso = dsos__find(&machine->dsos, filename, false); + struct dso *header_bid_dso = dsos__find(&machine->dsos, filename, false); + if (header_bid_dso && dso__header_build_id(header_bid_dso)) { dso__set_build_id(dso, dso__bid(header_bid_dso)); dso__set_header_build_id(dso, 1); @@ -354,7 +353,7 @@ int map__load(struct map *map) if (dso__has_build_id(dso)) { char sbuild_id[SBUILD_ID_SIZE]; - build_id__sprintf(dso__bid(dso), sbuild_id); + build_id__snprintf(dso__bid(dso), sbuild_id, sizeof(sbuild_id)); pr_debug("%s with build id %s not found", name, sbuild_id); } else pr_debug("Failed to open %s", name); diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h index 4262f5a143be..9cadf533a561 100644 --- a/tools/perf/util/map.h +++ b/tools/perf/util/map.h @@ -173,11 +173,10 @@ struct thread; __map__for_each_symbol_by_name(map, sym_name, (pos), idx) struct dso_id; -struct build_id; struct map *map__new(struct machine *machine, u64 start, u64 len, - u64 pgoff, struct dso_id *id, u32 prot, u32 flags, - struct build_id *bid, char *filename, struct thread *thread); + u64 pgoff, const struct dso_id *id, u32 prot, u32 flags, + char *filename, struct thread *thread); struct map *map__new2(u64 start, struct dso *dso); void map__delete(struct map *map); struct map *map__clone(struct map *map); diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util/metricgroup.c index 43d35f956a33..595b83142d2c 100644 --- a/tools/perf/util/metricgroup.c +++ b/tools/perf/util/metricgroup.c @@ -103,7 +103,7 @@ static void metric_event_delete(struct rblist *rblist __maybe_unused, free(me); } -static void metricgroup__rblist_init(struct rblist *metric_events) +void metricgroup__rblist_init(struct rblist *metric_events) { rblist__init(metric_events); metric_events->node_cmp = metric_event_cmp; @@ -179,7 +179,7 @@ static void metric__watchdog_constraint_hint(const char *name, bool foot) " echo 1 > /proc/sys/kernel/nmi_watchdog\n"); } -static bool metric__group_events(const struct pmu_metric *pm) +static bool metric__group_events(const struct pmu_metric *pm, bool metric_no_threshold) { switch (pm->event_grouping) { case MetricNoGroupEvents: @@ -191,6 +191,13 @@ static bool metric__group_events(const struct pmu_metric *pm) return false; case MetricNoGroupEventsSmt: return !smt_on(); + case MetricNoGroupEventsThresholdAndNmi: + if (metric_no_threshold) + return true; + if (!sysctl__nmi_watchdog_enabled()) + return true; + metric__watchdog_constraint_hint(pm->metric_name, /*foot=*/false); + return false; case MetricGroupEvents: default: return true; @@ -212,6 +219,7 @@ static void metric__free(struct metric *m) static struct metric *metric__new(const struct pmu_metric *pm, const char *modifier, bool metric_no_group, + bool metric_no_threshold, int runtime, const char *user_requested_cpu_list, bool system_wide) @@ -246,7 +254,7 @@ static struct metric *metric__new(const struct pmu_metric *pm, } m->pctx->sctx.runtime = runtime; m->pctx->sctx.system_wide = system_wide; - m->group_events = !metric_no_group && metric__group_events(pm); + m->group_events = !metric_no_group && metric__group_events(pm, metric_no_threshold); m->metric_refs = NULL; m->evlist = NULL; @@ -384,107 +392,6 @@ static bool match_pm_metric_or_groups(const struct pmu_metric *pm, const char *p match_metric_or_groups(pm->metric_name, metric_or_groups); } -/** struct mep - RB-tree node for building printing information. */ -struct mep { - /** nd - RB-tree element. */ - struct rb_node nd; - /** @metric_group: Owned metric group name, separated others with ';'. */ - char *metric_group; - const char *metric_name; - const char *metric_desc; - const char *metric_long_desc; - const char *metric_expr; - const char *metric_threshold; - const char *metric_unit; - const char *pmu_name; -}; - -static int mep_cmp(struct rb_node *rb_node, const void *entry) -{ - struct mep *a = container_of(rb_node, struct mep, nd); - struct mep *b = (struct mep *)entry; - int ret; - - ret = strcmp(a->metric_group, b->metric_group); - if (ret) - return ret; - - return strcmp(a->metric_name, b->metric_name); -} - -static struct rb_node *mep_new(struct rblist *rl __maybe_unused, const void *entry) -{ - struct mep *me = malloc(sizeof(struct mep)); - - if (!me) - return NULL; - - memcpy(me, entry, sizeof(struct mep)); - return &me->nd; -} - -static void mep_delete(struct rblist *rl __maybe_unused, - struct rb_node *nd) -{ - struct mep *me = container_of(nd, struct mep, nd); - - zfree(&me->metric_group); - free(me); -} - -static struct mep *mep_lookup(struct rblist *groups, const char *metric_group, - const char *metric_name) -{ - struct rb_node *nd; - struct mep me = { - .metric_group = strdup(metric_group), - .metric_name = metric_name, - }; - nd = rblist__find(groups, &me); - if (nd) { - free(me.metric_group); - return container_of(nd, struct mep, nd); - } - rblist__add_node(groups, &me); - nd = rblist__find(groups, &me); - if (nd) - return container_of(nd, struct mep, nd); - return NULL; -} - -static int metricgroup__add_to_mep_groups(const struct pmu_metric *pm, - struct rblist *groups) -{ - const char *g; - char *omg, *mg; - - mg = strdup(pm->metric_group ?: pm->metric_name); - if (!mg) - return -ENOMEM; - omg = mg; - while ((g = strsep(&mg, ";")) != NULL) { - struct mep *me; - - g = skip_spaces(g); - if (strlen(g)) - me = mep_lookup(groups, g, pm->metric_name); - else - me = mep_lookup(groups, pm->metric_name, pm->metric_name); - - if (me) { - me->metric_desc = pm->desc; - me->metric_long_desc = pm->long_desc; - me->metric_expr = pm->metric_expr; - me->metric_threshold = pm->metric_threshold; - me->metric_unit = pm->unit; - me->pmu_name = pm->pmu; - } - } - free(omg); - - return 0; -} - struct metricgroup_iter_data { pmu_metric_iter_fn fn; void *data; @@ -510,54 +417,22 @@ static int metricgroup__sys_event_iter(const struct pmu_metric *pm, return 0; } -static int metricgroup__add_to_mep_groups_callback(const struct pmu_metric *pm, - const struct pmu_metrics_table *table __maybe_unused, - void *vdata) -{ - struct rblist *groups = vdata; - - return metricgroup__add_to_mep_groups(pm, groups); -} - -void metricgroup__print(const struct print_callbacks *print_cb, void *print_state) +int metricgroup__for_each_metric(const struct pmu_metrics_table *table, pmu_metric_iter_fn fn, + void *data) { - struct rblist groups; - const struct pmu_metrics_table *table; - struct rb_node *node, *next; + struct metricgroup_iter_data sys_data = { + .fn = fn, + .data = data, + }; - rblist__init(&groups); - groups.node_new = mep_new; - groups.node_cmp = mep_cmp; - groups.node_delete = mep_delete; - table = pmu_metrics_table__find(); if (table) { - pmu_metrics_table__for_each_metric(table, - metricgroup__add_to_mep_groups_callback, - &groups); - } - { - struct metricgroup_iter_data data = { - .fn = metricgroup__add_to_mep_groups_callback, - .data = &groups, - }; - pmu_for_each_sys_metric(metricgroup__sys_event_iter, &data); - } + int ret = pmu_metrics_table__for_each_metric(table, fn, data); - for (node = rb_first_cached(&groups.entries); node; node = next) { - struct mep *me = container_of(node, struct mep, nd); - - print_cb->print_metric(print_state, - me->metric_group, - me->metric_name, - me->metric_desc, - me->metric_long_desc, - me->metric_expr, - me->metric_threshold, - me->metric_unit, - me->pmu_name); - next = rb_next(node); - rblist__remove_node(&groups, node); + if (ret) + return ret; } + + return pmu_for_each_sys_metric(metricgroup__sys_event_iter, &sys_data); } static const char *code_characters = ",-=@"; @@ -964,8 +839,8 @@ static int __add_metric(struct list_head *metric_list, * This metric is the root of a tree and may reference other * metrics that are added recursively. */ - root_metric = metric__new(pm, modifier, metric_no_group, runtime, - user_requested_cpu_list, system_wide); + root_metric = metric__new(pm, modifier, metric_no_group, metric_no_threshold, + runtime, user_requested_cpu_list, system_wide); if (!root_metric) return -ENOMEM; @@ -1090,29 +965,6 @@ static int add_metric(struct list_head *metric_list, return ret; } -static int metricgroup__add_metric_sys_event_iter(const struct pmu_metric *pm, - const struct pmu_metrics_table *table __maybe_unused, - void *data) -{ - struct metricgroup_add_iter_data *d = data; - int ret; - - if (!match_pm_metric_or_groups(pm, d->pmu, d->metric_name)) - return 0; - - ret = add_metric(d->metric_list, pm, d->modifier, d->metric_no_group, - d->metric_no_threshold, d->user_requested_cpu_list, - d->system_wide, d->root_metric, d->visited, d->table); - if (ret) - goto out; - - *(d->has_match) = true; - -out: - *(d->ret) = ret; - return ret; -} - /** * metric_list_cmp - list_sort comparator that sorts metrics with more events to * the front. tool events are excluded from the count. @@ -1216,55 +1068,26 @@ static int metricgroup__add_metric(const char *pmu, const char *metric_name, con { LIST_HEAD(list); int ret; - bool has_match = false; - - { - struct metricgroup__add_metric_data data = { - .list = &list, - .pmu = pmu, - .metric_name = metric_name, - .modifier = modifier, - .metric_no_group = metric_no_group, - .metric_no_threshold = metric_no_threshold, - .user_requested_cpu_list = user_requested_cpu_list, - .system_wide = system_wide, - .has_match = false, - }; - /* - * Iterate over all metrics seeing if metric matches either the - * name or group. When it does add the metric to the list. - */ - ret = pmu_metrics_table__for_each_metric(table, metricgroup__add_metric_callback, - &data); - if (ret) - goto out; + struct metricgroup__add_metric_data data = { + .list = &list, + .pmu = pmu, + .metric_name = metric_name, + .modifier = modifier, + .metric_no_group = metric_no_group, + .metric_no_threshold = metric_no_threshold, + .user_requested_cpu_list = user_requested_cpu_list, + .system_wide = system_wide, + .has_match = false, + }; - has_match = data.has_match; - } - { - struct metricgroup_iter_data data = { - .fn = metricgroup__add_metric_sys_event_iter, - .data = (void *) &(struct metricgroup_add_iter_data) { - .metric_list = &list, - .pmu = pmu, - .metric_name = metric_name, - .modifier = modifier, - .metric_no_group = metric_no_group, - .user_requested_cpu_list = user_requested_cpu_list, - .system_wide = system_wide, - .has_match = &has_match, - .ret = &ret, - .table = table, - }, - }; - - pmu_for_each_sys_metric(metricgroup__sys_event_iter, &data); - } - /* End of pmu events. */ - if (!has_match) + /* + * Iterate over all metrics seeing if metric matches either the + * name or group. When it does add the metric to the list. + */ + ret = metricgroup__for_each_metric(table, metricgroup__add_metric_callback, &data); + if (!ret && !data.has_match) ret = -EINVAL; -out: /* * add to metric_list so that they can be released * even if it's failed @@ -1508,7 +1331,6 @@ static int parse_groups(struct evlist *perf_evlist, const char *user_requested_cpu_list, bool system_wide, bool fake_pmu, - struct rblist *metric_events_list, const struct pmu_metrics_table *table) { struct evlist *combined_evlist = NULL; @@ -1518,8 +1340,6 @@ static int parse_groups(struct evlist *perf_evlist, bool is_default = !strcmp(str, "Default"); int ret; - if (metric_events_list->nr_entries == 0) - metricgroup__rblist_init(metric_events_list); ret = metricgroup__add_metric_list(pmu, str, metric_no_group, metric_no_threshold, user_requested_cpu_list, system_wide, &metric_list, table); @@ -1610,7 +1430,8 @@ static int parse_groups(struct evlist *perf_evlist, goto out; } - me = metricgroup__lookup(metric_events_list, metric_events[0], true); + me = metricgroup__lookup(&perf_evlist->metric_events, metric_events[0], + /*create=*/true); expr = malloc(sizeof(struct metric_expr)); if (!expr) { @@ -1670,8 +1491,7 @@ int metricgroup__parse_groups(struct evlist *perf_evlist, bool metric_no_threshold, const char *user_requested_cpu_list, bool system_wide, - bool hardware_aware_grouping, - struct rblist *metric_events) + bool hardware_aware_grouping) { const struct pmu_metrics_table *table = pmu_metrics_table__find(); @@ -1682,13 +1502,12 @@ int metricgroup__parse_groups(struct evlist *perf_evlist, return parse_groups(perf_evlist, pmu, str, metric_no_group, metric_no_merge, metric_no_threshold, user_requested_cpu_list, system_wide, - /*fake_pmu=*/false, metric_events, table); + /*fake_pmu=*/false, table); } int metricgroup__parse_groups_test(struct evlist *evlist, const struct pmu_metrics_table *table, - const char *str, - struct rblist *metric_events) + const char *str) { return parse_groups(evlist, "all", str, /*metric_no_group=*/false, @@ -1696,7 +1515,7 @@ int metricgroup__parse_groups_test(struct evlist *evlist, /*metric_no_threshold=*/false, /*user_requested_cpu_list=*/NULL, /*system_wide=*/false, - /*fake_pmu=*/true, metric_events, table); + /*fake_pmu=*/true, table); } struct metricgroup__has_metric_data { @@ -1781,7 +1600,7 @@ int metricgroup__copy_metric_events(struct evlist *evlist, struct cgroup *cgrp, evsel = evlist__find_evsel(evlist, old_me->evsel->core.idx); if (!evsel) return -EINVAL; - new_me = metricgroup__lookup(new_metric_events, evsel, true); + new_me = metricgroup__lookup(new_metric_events, evsel, /*create=*/true); if (!new_me) return -ENOMEM; diff --git a/tools/perf/util/metricgroup.h b/tools/perf/util/metricgroup.h index a04ac1afa6cc..324880b2ed8f 100644 --- a/tools/perf/util/metricgroup.h +++ b/tools/perf/util/metricgroup.h @@ -77,17 +77,17 @@ int metricgroup__parse_groups(struct evlist *perf_evlist, bool metric_no_threshold, const char *user_requested_cpu_list, bool system_wide, - bool hardware_aware_grouping, - struct rblist *metric_events); + bool hardware_aware_grouping); int metricgroup__parse_groups_test(struct evlist *evlist, const struct pmu_metrics_table *table, - const char *str, - struct rblist *metric_events); + const char *str); -void metricgroup__print(const struct print_callbacks *print_cb, void *print_state); +int metricgroup__for_each_metric(const struct pmu_metrics_table *table, pmu_metric_iter_fn fn, + void *data); bool metricgroup__has_metric_or_groups(const char *pmu, const char *metric_or_groups); unsigned int metricgroups__topdown_max_level(void); int arch_get_runtimeparam(const struct pmu_metric *pm); +void metricgroup__rblist_init(struct rblist *metric_events); void metricgroup__rblist_exit(struct rblist *metric_events); int metricgroup__copy_metric_events(struct evlist *evlist, struct cgroup *cgrp, diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index 2380de56a207..8282ddf68b98 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -17,14 +17,14 @@ #include "string2.h" #include "strbuf.h" #include "debug.h" -#include <api/fs/tracing_path.h> -#include <api/io_dir.h> #include <perf/cpumap.h> #include <util/parse-events-bison.h> #include <util/parse-events-flex.h> #include "pmu.h" #include "pmus.h" +#include "tp_pmu.h" #include "asm/bug.h" +#include "ui/ui.h" #include "util/parse-branch-options.h" #include "util/evsel_config.h" #include "util/event.h" @@ -32,6 +32,7 @@ #include "util/stat.h" #include "util/util.h" #include "tracepoint.h" +#include <api/fs/tracing_path.h> #define MAX_NAME_LEN 100 @@ -83,77 +84,21 @@ const struct event_symbol event_symbols_hw[PERF_COUNT_HW_MAX] = { }, }; -const struct event_symbol event_symbols_sw[PERF_COUNT_SW_MAX] = { - [PERF_COUNT_SW_CPU_CLOCK] = { - .symbol = "cpu-clock", - .alias = "", - }, - [PERF_COUNT_SW_TASK_CLOCK] = { - .symbol = "task-clock", - .alias = "", - }, - [PERF_COUNT_SW_PAGE_FAULTS] = { - .symbol = "page-faults", - .alias = "faults", - }, - [PERF_COUNT_SW_CONTEXT_SWITCHES] = { - .symbol = "context-switches", - .alias = "cs", - }, - [PERF_COUNT_SW_CPU_MIGRATIONS] = { - .symbol = "cpu-migrations", - .alias = "migrations", - }, - [PERF_COUNT_SW_PAGE_FAULTS_MIN] = { - .symbol = "minor-faults", - .alias = "", - }, - [PERF_COUNT_SW_PAGE_FAULTS_MAJ] = { - .symbol = "major-faults", - .alias = "", - }, - [PERF_COUNT_SW_ALIGNMENT_FAULTS] = { - .symbol = "alignment-faults", - .alias = "", - }, - [PERF_COUNT_SW_EMULATION_FAULTS] = { - .symbol = "emulation-faults", - .alias = "", - }, - [PERF_COUNT_SW_DUMMY] = { - .symbol = "dummy", - .alias = "", - }, - [PERF_COUNT_SW_BPF_OUTPUT] = { - .symbol = "bpf-output", - .alias = "", - }, - [PERF_COUNT_SW_CGROUP_SWITCHES] = { - .symbol = "cgroup-switches", - .alias = "", - }, +static const char *const event_types[] = { + [PERF_TYPE_HARDWARE] = "hardware", + [PERF_TYPE_SOFTWARE] = "software", + [PERF_TYPE_TRACEPOINT] = "tracepoint", + [PERF_TYPE_HW_CACHE] = "hardware-cache", + [PERF_TYPE_RAW] = "raw", + [PERF_TYPE_BREAKPOINT] = "breakpoint", }; -const char *event_type(int type) +const char *event_type(size_t type) { - switch (type) { - case PERF_TYPE_HARDWARE: - return "hardware"; - - case PERF_TYPE_SOFTWARE: - return "software"; - - case PERF_TYPE_TRACEPOINT: - return "tracepoint"; - - case PERF_TYPE_HW_CACHE: - return "hardware-cache"; - - default: - break; - } + if (type >= PERF_TYPE_MAX) + return "unknown"; - return "unknown"; + return event_types[type]; } static char *get_config_str(const struct parse_events_terms *head_terms, @@ -191,10 +136,22 @@ static struct perf_cpu_map *get_config_cpu(const struct parse_events_terms *head list_for_each_entry(term, &head_terms->terms, list) { if (term->type_term == PARSE_EVENTS__TERM_TYPE_CPU) { - struct perf_cpu_map *cpu = perf_cpu_map__new_int(term->val.num); - - perf_cpu_map__merge(&cpus, cpu); - perf_cpu_map__put(cpu); + struct perf_cpu_map *term_cpus; + + if (term->type_val == PARSE_EVENTS__TERM_TYPE_NUM) { + term_cpus = perf_cpu_map__new_int(term->val.num); + } else { + struct perf_pmu *pmu = perf_pmus__find(term->val.str); + + if (pmu && perf_cpu_map__is_empty(pmu->cpus)) + term_cpus = pmu->is_core ? cpu_map__online() : NULL; + else if (pmu) + term_cpus = perf_cpu_map__get(pmu->cpus); + else + term_cpus = perf_cpu_map__new(term->val.str); + } + perf_cpu_map__merge(&cpus, term_cpus); + perf_cpu_map__put(term_cpus); } } @@ -251,11 +208,12 @@ __add_event(struct list_head *list, int *idx, bool init_attr, const char *name, const char *metric_id, struct perf_pmu *pmu, struct list_head *config_terms, struct evsel *first_wildcard_match, - struct perf_cpu_map *cpu_list, u64 alternate_hw_config) + struct perf_cpu_map *user_cpus, u64 alternate_hw_config) { struct evsel *evsel; bool is_pmu_core; - struct perf_cpu_map *cpus; + struct perf_cpu_map *cpus, *pmu_cpus; + bool has_user_cpus = !perf_cpu_map__is_empty(user_cpus); /* * Ensure the first_wildcard_match's PMU matches that of the new event @@ -279,8 +237,6 @@ __add_event(struct list_head *list, int *idx, } if (pmu) { - is_pmu_core = pmu->is_core; - cpus = perf_cpu_map__get(perf_cpu_map__is_empty(cpu_list) ? pmu->cpus : cpu_list); perf_pmu__warn_invalid_formats(pmu); if (attr->type == PERF_TYPE_RAW || attr->type >= PERF_TYPE_MAX) { perf_pmu__warn_invalid_config(pmu, attr->config, name, @@ -292,45 +248,75 @@ __add_event(struct list_head *list, int *idx, perf_pmu__warn_invalid_config(pmu, attr->config3, name, PERF_PMU_FORMAT_VALUE_CONFIG3, "config3"); } + } + /* + * If a PMU wasn't given, such as for legacy events, find now that + * warnings won't be generated. + */ + if (!pmu) + pmu = perf_pmus__find_by_attr(attr); + + if (pmu) { + is_pmu_core = pmu->is_core; + pmu_cpus = perf_cpu_map__get(pmu->cpus); + if (perf_cpu_map__is_empty(pmu_cpus)) + pmu_cpus = cpu_map__online(); } else { is_pmu_core = (attr->type == PERF_TYPE_HARDWARE || attr->type == PERF_TYPE_HW_CACHE); - if (perf_cpu_map__is_empty(cpu_list)) - cpus = is_pmu_core ? perf_cpu_map__new_online_cpus() : NULL; - else - cpus = perf_cpu_map__get(cpu_list); + pmu_cpus = is_pmu_core ? cpu_map__online() : NULL; } + + if (has_user_cpus) + cpus = perf_cpu_map__get(user_cpus); + else + cpus = perf_cpu_map__get(pmu_cpus); + if (init_attr) event_attr_init(attr); evsel = evsel__new_idx(attr, *idx); - if (!evsel) { - perf_cpu_map__put(cpus); - return NULL; + if (!evsel) + goto out_err; + + if (name) { + evsel->name = strdup(name); + if (!evsel->name) + goto out_err; + } + + if (metric_id) { + evsel->metric_id = strdup(metric_id); + if (!evsel->metric_id) + goto out_err; } (*idx)++; evsel->core.cpus = cpus; - evsel->core.own_cpus = perf_cpu_map__get(cpus); + evsel->core.pmu_cpus = pmu_cpus; evsel->core.requires_cpu = pmu ? pmu->is_uncore : false; evsel->core.is_pmu_core = is_pmu_core; evsel->pmu = pmu; evsel->alternate_hw_config = alternate_hw_config; evsel->first_wildcard_match = first_wildcard_match; - if (name) - evsel->name = strdup(name); - - if (metric_id) - evsel->metric_id = strdup(metric_id); - if (config_terms) list_splice_init(config_terms, &evsel->config_terms); if (list) list_add_tail(&evsel->core.node, list); + if (has_user_cpus) + evsel__warn_user_requested_cpus(evsel, user_cpus); + return evsel; +out_err: + perf_cpu_map__put(cpus); + perf_cpu_map__put(pmu_cpus); + zfree(&evsel->name); + zfree(&evsel->metric_id); + free(evsel); + return NULL; } struct evsel *parse_events__add_event(int idx, struct perf_event_attr *attr, @@ -489,7 +475,7 @@ int parse_events_add_cache(struct list_head *list, int *idx, const char *name, int ret = 0; struct evsel *first_wildcard_match = NULL; - while ((pmu = perf_pmus__scan(pmu)) != NULL) { + while ((pmu = perf_pmus__scan_for_event(pmu, name)) != NULL) { LIST_HEAD(config_terms); struct perf_event_attr attr; @@ -613,105 +599,82 @@ static int add_tracepoint(struct parse_events_state *parse_state, return 0; } -static int add_tracepoint_multi_event(struct parse_events_state *parse_state, - struct list_head *list, - const char *sys_name, const char *evt_name, - struct parse_events_error *err, - struct parse_events_terms *head_config, YYLTYPE *loc) -{ - char *evt_path; - struct io_dirent64 *evt_ent; - struct io_dir evt_dir; - int ret = 0, found = 0; - - evt_path = get_events_file(sys_name); - if (!evt_path) { - tracepoint_error(err, errno, sys_name, evt_name, loc->first_column); - return -1; - } - io_dir__init(&evt_dir, open(evt_path, O_CLOEXEC | O_DIRECTORY | O_RDONLY)); - if (evt_dir.dirfd < 0) { - put_events_file(evt_path); - tracepoint_error(err, errno, sys_name, evt_name, loc->first_column); - return -1; - } +struct add_tracepoint_multi_args { + struct parse_events_state *parse_state; + struct list_head *list; + const char *sys_glob; + const char *evt_glob; + struct parse_events_error *err; + struct parse_events_terms *head_config; + YYLTYPE *loc; + int found; +}; - while (!ret && (evt_ent = io_dir__readdir(&evt_dir))) { - if (!strcmp(evt_ent->d_name, ".") - || !strcmp(evt_ent->d_name, "..") - || !strcmp(evt_ent->d_name, "enable") - || !strcmp(evt_ent->d_name, "filter")) - continue; +static int add_tracepoint_multi_event_cb(void *state, const char *sys_name, const char *evt_name) +{ + struct add_tracepoint_multi_args *args = state; + int ret; - if (!strglobmatch(evt_ent->d_name, evt_name)) - continue; + if (!strglobmatch(evt_name, args->evt_glob)) + return 0; - found++; + args->found++; + ret = add_tracepoint(args->parse_state, args->list, sys_name, evt_name, + args->err, args->head_config, args->loc); - ret = add_tracepoint(parse_state, list, sys_name, evt_ent->d_name, - err, head_config, loc); - } + return ret; +} - if (!found) { - tracepoint_error(err, ENOENT, sys_name, evt_name, loc->first_column); - ret = -1; +static int add_tracepoint_multi_event(struct add_tracepoint_multi_args *args, const char *sys_name) +{ + if (strpbrk(args->evt_glob, "*?") == NULL) { + /* Not a glob. */ + args->found++; + return add_tracepoint(args->parse_state, args->list, sys_name, args->evt_glob, + args->err, args->head_config, args->loc); } - put_events_file(evt_path); - close(evt_dir.dirfd); - return ret; + return tp_pmu__for_each_tp_event(sys_name, args, add_tracepoint_multi_event_cb); } -static int add_tracepoint_event(struct parse_events_state *parse_state, - struct list_head *list, - const char *sys_name, const char *evt_name, - struct parse_events_error *err, - struct parse_events_terms *head_config, YYLTYPE *loc) +static int add_tracepoint_multi_sys_cb(void *state, const char *sys_name) { - return strpbrk(evt_name, "*?") ? - add_tracepoint_multi_event(parse_state, list, sys_name, evt_name, - err, head_config, loc) : - add_tracepoint(parse_state, list, sys_name, evt_name, - err, head_config, loc); + struct add_tracepoint_multi_args *args = state; + + if (!strglobmatch(sys_name, args->sys_glob)) + return 0; + + return add_tracepoint_multi_event(args, sys_name); } static int add_tracepoint_multi_sys(struct parse_events_state *parse_state, struct list_head *list, - const char *sys_name, const char *evt_name, + const char *sys_glob, const char *evt_glob, struct parse_events_error *err, struct parse_events_terms *head_config, YYLTYPE *loc) { - struct io_dirent64 *events_ent; - struct io_dir events_dir; - int ret = 0; - char *events_dir_path = get_tracing_file("events"); + struct add_tracepoint_multi_args args = { + .parse_state = parse_state, + .list = list, + .sys_glob = sys_glob, + .evt_glob = evt_glob, + .err = err, + .head_config = head_config, + .loc = loc, + .found = 0, + }; + int ret; - if (!events_dir_path) { - tracepoint_error(err, errno, sys_name, evt_name, loc->first_column); - return -1; - } - io_dir__init(&events_dir, open(events_dir_path, O_CLOEXEC | O_DIRECTORY | O_RDONLY)); - put_events_file(events_dir_path); - if (events_dir.dirfd < 0) { - tracepoint_error(err, errno, sys_name, evt_name, loc->first_column); - return -1; + if (strpbrk(sys_glob, "*?") == NULL) { + /* Not a glob. */ + ret = add_tracepoint_multi_event(&args, sys_glob); + } else { + ret = tp_pmu__for_each_tp_sys(&args, add_tracepoint_multi_sys_cb); } - - while (!ret && (events_ent = io_dir__readdir(&events_dir))) { - if (!strcmp(events_ent->d_name, ".") - || !strcmp(events_ent->d_name, "..") - || !strcmp(events_ent->d_name, "enable") - || !strcmp(events_ent->d_name, "header_event") - || !strcmp(events_ent->d_name, "header_page")) - continue; - - if (!strglobmatch(events_ent->d_name, sys_name)) - continue; - - ret = add_tracepoint_event(parse_state, list, events_ent->d_name, - evt_name, err, head_config, loc); + if (args.found == 0) { + tracepoint_error(err, ENOENT, sys_glob, evt_glob, loc->first_column); + return -ENOENT; } - close(events_dir.dirfd); return ret; } @@ -1048,15 +1011,32 @@ do { \ return -EINVAL; } break; - case PARSE_EVENTS__TERM_TYPE_CPU: - CHECK_TYPE_VAL(NUM); - if (term->val.num >= (u64)cpu__max_present_cpu().cpu) { + case PARSE_EVENTS__TERM_TYPE_CPU: { + struct perf_cpu_map *map; + + if (term->type_val == PARSE_EVENTS__TERM_TYPE_NUM) { + if (term->val.num >= (u64)cpu__max_present_cpu().cpu) { + parse_events_error__handle(err, term->err_val, + strdup("too big"), + /*help=*/NULL); + return -EINVAL; + } + break; + } + assert(term->type_val == PARSE_EVENTS__TERM_TYPE_STR); + if (perf_pmus__find(term->val.str) != NULL) + break; + + map = perf_cpu_map__new(term->val.str); + if (!map) { parse_events_error__handle(err, term->err_val, - strdup("too big"), - NULL); + strdup("not a valid PMU or CPU number"), + /*help=*/NULL); return -EINVAL; } + perf_cpu_map__put(map); break; + } case PARSE_EVENTS__TERM_TYPE_DRV_CFG: case PARSE_EVENTS__TERM_TYPE_USER: case PARSE_EVENTS__TERM_TYPE_LEGACY_CACHE: @@ -1403,12 +1383,8 @@ int parse_events_add_tracepoint(struct parse_events_state *parse_state, return -EINVAL; } - if (strpbrk(sys, "*?")) - return add_tracepoint_multi_sys(parse_state, list, sys, event, - err, head_config, loc); - else - return add_tracepoint_event(parse_state, list, sys, event, - err, head_config, loc); + return add_tracepoint_multi_sys(parse_state, list, sys, event, + err, head_config, loc); } static int __parse_events_add_numeric(struct parse_events_state *parse_state, @@ -1680,7 +1656,8 @@ int parse_events_multi_pmu_add(struct parse_events_state *parse_state, INIT_LIST_HEAD(list); - while ((pmu = perf_pmus__scan(pmu)) != NULL) { + while ((pmu = perf_pmus__scan_for_event(pmu, event_name)) != NULL) { + if (parse_events__filter_pmu(parse_state, pmu)) continue; @@ -1759,19 +1736,21 @@ int parse_events_multi_pmu_add_or_add_pmu(struct parse_events_state *parse_state pmu = NULL; /* Failed to add, try wildcard expansion of event_or_pmu as a PMU name. */ - while ((pmu = perf_pmus__scan(pmu)) != NULL) { - if (!parse_events__filter_pmu(parse_state, pmu) && - perf_pmu__wildcard_match(pmu, event_or_pmu)) { - if (!parse_events_add_pmu(parse_state, *listp, pmu, - const_parsed_terms, - first_wildcard_match, - /*alternate_hw_config=*/PERF_COUNT_HW_MAX)) { - ok++; - parse_state->wild_card_pmus = true; - } - if (first_wildcard_match == NULL) - first_wildcard_match = - container_of((*listp)->prev, struct evsel, core.node); + while ((pmu = perf_pmus__scan_matching_wildcard(pmu, event_or_pmu)) != NULL) { + + if (parse_events__filter_pmu(parse_state, pmu)) + continue; + + if (!parse_events_add_pmu(parse_state, *listp, pmu, + const_parsed_terms, + first_wildcard_match, + /*alternate_hw_config=*/PERF_COUNT_HW_MAX)) { + ok++; + parse_state->wild_card_pmus = true; + } + if (first_wildcard_match == NULL) { + first_wildcard_match = + container_of((*listp)->prev, struct evsel, core.node); } } if (ok) @@ -1829,13 +1808,11 @@ static int parse_events__modifier_list(struct parse_events_state *parse_state, int eH = group ? evsel->core.attr.exclude_host : 0; int eG = group ? evsel->core.attr.exclude_guest : 0; int exclude = eu | ek | eh; - int exclude_GH = group ? evsel->exclude_GH : 0; + int exclude_GH = eG | eH; if (mod.user) { if (!exclude) exclude = eu = ek = eh = 1; - if (!exclude_GH && !perf_guest && exclude_GH_default) - eG = 1; eu = 0; } if (mod.kernel) { @@ -1858,6 +1835,13 @@ static int parse_events__modifier_list(struct parse_events_state *parse_state, exclude_GH = eG = eH = 1; eH = 0; } + if (!exclude_GH && exclude_GH_default) { + if (perf_host) + eG = 1; + else if (perf_guest) + eH = 1; + } + evsel->core.attr.exclude_user = eu; evsel->core.attr.exclude_kernel = ek; evsel->core.attr.exclude_hv = eh; @@ -2128,6 +2112,11 @@ static int evlist__cmp(void *_fg_idx, const struct list_head *l, const struct li return arch_evlist__cmp(lhs, rhs); } +int __weak arch_evlist__add_required_events(struct list_head *list __always_unused) +{ + return 0; +} + static int parse_events__sort_events_and_fix_groups(struct list_head *list) { int idx = 0, force_grouped_idx = -1; @@ -2139,6 +2128,11 @@ static int parse_events__sort_events_and_fix_groups(struct list_head *list) struct evsel *force_grouped_leader = NULL; bool last_event_was_forced_leader = false; + /* On x86 topdown metrics events require a slots event. */ + ret = arch_evlist__add_required_events(list); + if (ret) + return ret; + /* * Compute index to insert ungrouped events at. Place them where the * first ungrouped event appears. @@ -2561,12 +2555,17 @@ foreach_evsel_in_last_glob(struct evlist *evlist, return 0; } +/* Will a tracepoint filter work for str or should a BPF filter be used? */ +static bool is_possible_tp_filter(const char *str) +{ + return strstr(str, "uid") == NULL; +} + static int set_filter(struct evsel *evsel, const void *arg) { const char *str = arg; - bool found = false; int nr_addr_filters = 0; - struct perf_pmu *pmu = NULL; + struct perf_pmu *pmu; if (evsel == NULL) { fprintf(stderr, @@ -2574,7 +2573,7 @@ static int set_filter(struct evsel *evsel, const void *arg) return -1; } - if (evsel->core.attr.type == PERF_TYPE_TRACEPOINT) { + if (evsel->core.attr.type == PERF_TYPE_TRACEPOINT && is_possible_tp_filter(str)) { if (evsel__append_tp_filter(evsel, str) < 0) { fprintf(stderr, "not enough memory to hold filter string\n"); @@ -2584,16 +2583,11 @@ static int set_filter(struct evsel *evsel, const void *arg) return 0; } - while ((pmu = perf_pmus__scan(pmu)) != NULL) - if (pmu->type == evsel->core.attr.type) { - found = true; - break; - } - - if (found) + pmu = evsel__find_pmu(evsel); + if (pmu) { perf_pmu__scan_file(pmu, "nr_addr_filters", "%d", &nr_addr_filters); - + } if (!nr_addr_filters) return perf_bpf_filter__parse(&evsel->bpf_filters, str); @@ -2615,6 +2609,30 @@ int parse_filter(const struct option *opt, const char *str, (const void *)str); } +int parse_uid_filter(struct evlist *evlist, uid_t uid) +{ + struct option opt = { + .value = &evlist, + }; + char buf[128]; + int ret; + + snprintf(buf, sizeof(buf), "uid == %d", uid); + ret = parse_filter(&opt, buf, /*unset=*/0); + if (ret) { + if (use_browser >= 1) { + /* + * Use ui__warning so a pop up appears above the + * underlying BPF error message. + */ + ui__warning("Failed to add UID filtering that uses BPF filtering.\n"); + } else { + fprintf(stderr, "Failed to add UID filtering that uses BPF filtering.\n"); + } + } + return ret; +} + static int add_exclude_perf_filter(struct evsel *evsel, const void *arg __maybe_unused) { diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h index ab242f671031..62dc7202e3ba 100644 --- a/tools/perf/util/parse-events.h +++ b/tools/perf/util/parse-events.h @@ -11,6 +11,7 @@ #include <linux/perf_event.h> #include <stdio.h> #include <string.h> +#include <sys/types.h> struct evsel; struct evlist; @@ -20,7 +21,7 @@ struct option; struct perf_pmu; struct strbuf; -const char *event_type(int type); +const char *event_type(size_t type); /* Arguments encoded in opt->value. */ struct parse_events_option_args { @@ -45,6 +46,7 @@ static inline int parse_events(struct evlist *evlist, const char *str, int parse_event(struct evlist *evlist, const char *str); int parse_filter(const struct option *opt, const char *str, int unset); +int parse_uid_filter(struct evlist *evlist, uid_t uid); int exclude_perf(const struct option *opt, const char *arg, int unset); enum parse_events__term_val_type { @@ -262,7 +264,6 @@ struct event_symbol { const char *alias; }; extern const struct event_symbol event_symbols_hw[]; -extern const struct event_symbol event_symbols_sw[]; char *parse_events_formats_error_string(char *additional_terms); diff --git a/tools/perf/util/parse-events.l b/tools/perf/util/parse-events.l index 4af7b9c1f44d..2034590eb789 100644 --- a/tools/perf/util/parse-events.l +++ b/tools/perf/util/parse-events.l @@ -117,12 +117,12 @@ do { \ yyless(0); \ } while (0) -static int sym(yyscan_t scanner, int type, int config) +static int sym(yyscan_t scanner, int config) { YYSTYPE *yylval = parse_events_get_lval(scanner); - yylval->num = (type << 16) + config; - return type == PERF_TYPE_HARDWARE ? PE_VALUE_SYM_HW : PE_VALUE_SYM_SW; + yylval->num = config; + return PE_VALUE_SYM_HW; } static int term(yyscan_t scanner, enum parse_events__term_type type) @@ -391,28 +391,16 @@ r0x{num_raw_hex} { return str(yyscanner, PE_RAW); } <<EOF>> { BEGIN(INITIAL); } } -cpu-cycles|cycles { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_CPU_CYCLES); } -stalled-cycles-frontend|idle-cycles-frontend { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_STALLED_CYCLES_FRONTEND); } -stalled-cycles-backend|idle-cycles-backend { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_STALLED_CYCLES_BACKEND); } -instructions { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_INSTRUCTIONS); } -cache-references { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_CACHE_REFERENCES); } -cache-misses { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_CACHE_MISSES); } -branch-instructions|branches { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_BRANCH_INSTRUCTIONS); } -branch-misses { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_BRANCH_MISSES); } -bus-cycles { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_BUS_CYCLES); } -ref-cycles { return sym(yyscanner, PERF_TYPE_HARDWARE, PERF_COUNT_HW_REF_CPU_CYCLES); } -cpu-clock { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_CPU_CLOCK); } -task-clock { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_TASK_CLOCK); } -page-faults|faults { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_PAGE_FAULTS); } -minor-faults { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_PAGE_FAULTS_MIN); } -major-faults { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_PAGE_FAULTS_MAJ); } -context-switches|cs { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_CONTEXT_SWITCHES); } -cpu-migrations|migrations { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_CPU_MIGRATIONS); } -alignment-faults { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_ALIGNMENT_FAULTS); } -emulation-faults { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_EMULATION_FAULTS); } -dummy { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_DUMMY); } -bpf-output { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_BPF_OUTPUT); } -cgroup-switches { return sym(yyscanner, PERF_TYPE_SOFTWARE, PERF_COUNT_SW_CGROUP_SWITCHES); } +cpu-cycles|cycles { return sym(yyscanner, PERF_COUNT_HW_CPU_CYCLES); } +stalled-cycles-frontend|idle-cycles-frontend { return sym(yyscanner, PERF_COUNT_HW_STALLED_CYCLES_FRONTEND); } +stalled-cycles-backend|idle-cycles-backend { return sym(yyscanner, PERF_COUNT_HW_STALLED_CYCLES_BACKEND); } +instructions { return sym(yyscanner, PERF_COUNT_HW_INSTRUCTIONS); } +cache-references { return sym(yyscanner, PERF_COUNT_HW_CACHE_REFERENCES); } +cache-misses { return sym(yyscanner, PERF_COUNT_HW_CACHE_MISSES); } +branch-instructions|branches { return sym(yyscanner, PERF_COUNT_HW_BRANCH_INSTRUCTIONS); } +branch-misses { return sym(yyscanner, PERF_COUNT_HW_BRANCH_MISSES); } +bus-cycles { return sym(yyscanner, PERF_COUNT_HW_BUS_CYCLES); } +ref-cycles { return sym(yyscanner, PERF_COUNT_HW_REF_CPU_CYCLES); } {lc_type} { return str(yyscanner, PE_LEGACY_CACHE); } {lc_type}-{lc_op_result} { return str(yyscanner, PE_LEGACY_CACHE); } diff --git a/tools/perf/util/parse-events.y b/tools/perf/util/parse-events.y index f888cbb076d6..a2361c0040d7 100644 --- a/tools/perf/util/parse-events.y +++ b/tools/perf/util/parse-events.y @@ -55,7 +55,7 @@ static void free_list_evsel(struct list_head* list_evsel) %} %token PE_START_EVENTS PE_START_TERMS -%token PE_VALUE PE_VALUE_SYM_HW PE_VALUE_SYM_SW PE_TERM +%token PE_VALUE PE_VALUE_SYM_HW PE_TERM %token PE_EVENT_NAME %token PE_RAW PE_NAME %token PE_MODIFIER_EVENT PE_MODIFIER_BP PE_BP_COLON PE_BP_SLASH @@ -66,10 +66,8 @@ static void free_list_evsel(struct list_head* list_evsel) %token PE_TERM_HW %type <num> PE_VALUE %type <num> PE_VALUE_SYM_HW -%type <num> PE_VALUE_SYM_SW %type <mod> PE_MODIFIER_EVENT %type <term_type> PE_TERM -%type <num> value_sym %type <str> PE_RAW %type <str> PE_NAME %type <str> PE_LEGACY_CACHE @@ -306,24 +304,19 @@ PE_NAME sep_dc $$ = list; } -value_sym: -PE_VALUE_SYM_HW -| -PE_VALUE_SYM_SW - event_legacy_symbol: -value_sym '/' event_config '/' +PE_VALUE_SYM_HW '/' event_config '/' { struct list_head *list; - int type = $1 >> 16; - int config = $1 & 255; int err; - bool wildcard = (type == PERF_TYPE_HARDWARE || type == PERF_TYPE_HW_CACHE); list = alloc_list(); if (!list) YYNOMEM; - err = parse_events_add_numeric(_parse_state, list, type, config, $3, wildcard); + err = parse_events_add_numeric(_parse_state, list, + PERF_TYPE_HARDWARE, $1, + $3, + /*wildcard=*/true); parse_events_terms__delete($3); if (err) { free_list_evsel(list); @@ -332,18 +325,18 @@ value_sym '/' event_config '/' $$ = list; } | -value_sym sep_slash_slash_dc +PE_VALUE_SYM_HW sep_slash_slash_dc { struct list_head *list; - int type = $1 >> 16; - int config = $1 & 255; - bool wildcard = (type == PERF_TYPE_HARDWARE || type == PERF_TYPE_HW_CACHE); int err; list = alloc_list(); if (!list) YYNOMEM; - err = parse_events_add_numeric(_parse_state, list, type, config, /*head_config=*/NULL, wildcard); + err = parse_events_add_numeric(_parse_state, list, + PERF_TYPE_HARDWARE, $1, + /*head_config=*/NULL, + /*wildcard=*/true); if (err) PE_ABORT(err); $$ = list; diff --git a/tools/perf/util/pfm.c b/tools/perf/util/pfm.c index 0dacc133ed39..e5b3a2a5ddef 100644 --- a/tools/perf/util/pfm.c +++ b/tools/perf/util/pfm.c @@ -47,10 +47,6 @@ int parse_libpfm_events_option(const struct option *opt, const char *str, p_orig = p = strdup(str); if (!p) return -1; - /* - * force loading of the PMU list - */ - perf_pmus__scan(NULL); for (q = p; strsep(&p, ",{}"); q = p) { sep = p ? str + (p - p_orig - 1) : ""; @@ -234,6 +230,7 @@ print_libpfm_event(const struct print_callbacks *print_cb, void *print_state, if (is_libpfm_event_supported(name, cpus, threads)) { print_cb->print_event(print_state, topic, pinfo->name, + /*pmu_type=*/PERF_TYPE_RAW, name, info->equiv, /*scale_unit=*/NULL, /*deprecated=*/NULL, "PFM event", @@ -269,6 +266,7 @@ print_libpfm_event(const struct print_callbacks *print_cb, void *print_state, print_cb->print_event(print_state, topic, pinfo->name, + /*pmu_type=*/PERF_TYPE_RAW, name, /*alias=*/NULL, /*scale_unit=*/NULL, /*deprecated=*/NULL, "PFM event", diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c index 609828513f6c..5a291f1380ed 100644 --- a/tools/perf/util/pmu.c +++ b/tools/perf/util/pmu.c @@ -20,9 +20,11 @@ #include "debug.h" #include "evsel.h" #include "pmu.h" +#include "drm_pmu.h" #include "hwmon_pmu.h" #include "pmus.h" #include "tool_pmu.h" +#include "tp_pmu.h" #include <util/pmu-bison.h> #include <util/pmu-flex.h> #include "parse-events.h" @@ -452,7 +454,7 @@ static struct perf_pmu_alias *perf_pmu__find_alias(struct perf_pmu *pmu, { struct perf_pmu_alias *alias; bool has_sysfs_event; - char event_file_name[FILENAME_MAX + 8]; + char event_file_name[NAME_MAX + 8]; if (hashmap__find(pmu->aliases, name, &alias)) return alias; @@ -622,8 +624,7 @@ static int perf_pmu__new_alias(struct perf_pmu *pmu, const char *name, alias->name = strdup(name); alias->desc = desc ? strdup(desc) : NULL; - alias->long_desc = long_desc ? strdup(long_desc) : - desc ? strdup(desc) : NULL; + alias->long_desc = long_desc ? strdup(long_desc) : NULL; alias->topic = topic ? strdup(topic) : NULL; alias->pmu_name = pmu_name ? strdup(pmu_name) : NULL; if (unit) { @@ -752,7 +753,7 @@ static int pmu_aliases_parse(struct perf_pmu *pmu) static int pmu_aliases_parse_eager(struct perf_pmu *pmu, int sysfs_fd) { - char path[FILENAME_MAX + 7]; + char path[NAME_MAX + 8]; int ret, events_dir_fd; scnprintf(path, sizeof(path), "%s/events", pmu->name); @@ -1181,6 +1182,32 @@ int perf_pmu__init(struct perf_pmu *pmu, __u32 type, const char *name) return 0; } +static __u32 wellknown_pmu_type(const char *pmu_name) +{ + struct { + const char *pmu_name; + __u32 type; + } wellknown_pmus[] = { + { + "software", + PERF_TYPE_SOFTWARE + }, + { + "tracepoint", + PERF_TYPE_TRACEPOINT + }, + { + "breakpoint", + PERF_TYPE_BREAKPOINT + }, + }; + for (size_t i = 0; i < ARRAY_SIZE(wellknown_pmus); i++) { + if (!strcmp(wellknown_pmus[i].pmu_name, pmu_name)) + return wellknown_pmus[i].type; + } + return PERF_TYPE_MAX; +} + struct perf_pmu *perf_pmu__lookup(struct list_head *pmus, int dirfd, const char *name, bool eager_load) { @@ -1200,8 +1227,12 @@ struct perf_pmu *perf_pmu__lookup(struct list_head *pmus, int dirfd, const char * that type value is successfully assigned (return 1). */ if (perf_pmu__scan_file_at(pmu, dirfd, "type", "%u", &pmu->type) != 1) { - perf_pmu__delete(pmu); - return NULL; + /* Double check the PMU's name isn't wellknown. */ + pmu->type = wellknown_pmu_type(name); + if (pmu->type == PERF_TYPE_MAX) { + perf_pmu__delete(pmu); + return NULL; + } } /* @@ -1627,6 +1658,8 @@ int perf_pmu__config_terms(const struct perf_pmu *pmu, if (perf_pmu__is_hwmon(pmu)) return hwmon_pmu__config_terms(pmu, attr, terms, err); + if (perf_pmu__is_drm(pmu)) + return drm_pmu__config_terms(pmu, attr, terms, err); list_for_each_entry(term, &terms->terms, list) { if (pmu_config_term(pmu, attr, term, terms, zero, apply_hardcoded, err)) @@ -1767,6 +1800,10 @@ int perf_pmu__check_alias(struct perf_pmu *pmu, struct parse_events_terms *head_ ret = hwmon_pmu__check_alias(head_terms, info, err); goto out; } + if (perf_pmu__is_drm(pmu)) { + ret = drm_pmu__check_alias(pmu, head_terms, info, err); + goto out; + } /* Fake PMU doesn't rewrite terms. */ if (perf_pmu__is_fake(pmu)) @@ -1947,8 +1984,12 @@ bool perf_pmu__have_event(struct perf_pmu *pmu, const char *name) return false; if (perf_pmu__is_tool(pmu) && tool_pmu__skip_event(name)) return false; + if (perf_pmu__is_tracepoint(pmu)) + return tp_pmu__have_event(pmu, name); if (perf_pmu__is_hwmon(pmu)) return hwmon_pmu__have_event(pmu, name); + if (perf_pmu__is_drm(pmu)) + return drm_pmu__have_event(pmu, name); if (perf_pmu__find_alias(pmu, name, /*load=*/ true) != NULL) return true; if (pmu->cpu_aliases_added || !pmu->events_table) @@ -1960,8 +2001,12 @@ size_t perf_pmu__num_events(struct perf_pmu *pmu) { size_t nr; + if (perf_pmu__is_tracepoint(pmu)) + return tp_pmu__num_events(pmu); if (perf_pmu__is_hwmon(pmu)) return hwmon_pmu__num_events(pmu); + if (perf_pmu__is_drm(pmu)) + return drm_pmu__num_events(pmu); pmu_aliases_parse(pmu); nr = pmu->sysfs_aliases + pmu->sys_json_aliases; @@ -2028,8 +2073,12 @@ int perf_pmu__for_each_event(struct perf_pmu *pmu, bool skip_duplicate_pmus, struct hashmap_entry *entry; size_t bkt; + if (perf_pmu__is_tracepoint(pmu)) + return tp_pmu__for_each_event(pmu, state, cb); if (perf_pmu__is_hwmon(pmu)) return hwmon_pmu__for_each_event(pmu, state, cb); + if (perf_pmu__is_drm(pmu)) + return drm_pmu__for_each_event(pmu, state, cb); strbuf_init(&sb, /*hint=*/ 0); pmu_aliases_parse(pmu); @@ -2511,6 +2560,8 @@ void perf_pmu__delete(struct perf_pmu *pmu) if (perf_pmu__is_hwmon(pmu)) hwmon_pmu__exit(pmu); + else if (perf_pmu__is_drm(pmu)) + drm_pmu__exit(pmu); perf_pmu__del_formats(&pmu->format); perf_pmu__del_aliases(pmu); diff --git a/tools/perf/util/pmu.h b/tools/perf/util/pmu.h index 71b8636fd07d..1ebcf0242af8 100644 --- a/tools/perf/util/pmu.h +++ b/tools/perf/util/pmu.h @@ -39,7 +39,9 @@ struct perf_pmu_caps { enum { PERF_PMU_TYPE_PE_START = 0, - PERF_PMU_TYPE_PE_END = 0xFFFEFFFF, + PERF_PMU_TYPE_PE_END = 0xFFFDFFFF, + PERF_PMU_TYPE_DRM_START = 0xFFFE0000, + PERF_PMU_TYPE_DRM_END = 0xFFFEFFFF, PERF_PMU_TYPE_HWMON_START = 0xFFFF0000, PERF_PMU_TYPE_HWMON_END = 0xFFFFFFFD, PERF_PMU_TYPE_TOOL = 0xFFFFFFFE, @@ -300,7 +302,6 @@ struct perf_pmu *perf_pmu__lookup(struct list_head *pmus, int dirfd, const char bool eager_load); struct perf_pmu *perf_pmu__create_placeholder_core_pmu(struct list_head *core_pmus); void perf_pmu__delete(struct perf_pmu *pmu); -struct perf_pmu *perf_pmus__find_core_pmu(void); const char *perf_pmu__name_from_config(struct perf_pmu *pmu, u64 config); bool perf_pmu__is_fake(const struct perf_pmu *pmu); diff --git a/tools/perf/util/pmus.c b/tools/perf/util/pmus.c index 3bbd26fec78a..98be2eb8f1f0 100644 --- a/tools/perf/util/pmus.c +++ b/tools/perf/util/pmus.c @@ -12,6 +12,7 @@ #include <unistd.h> #include "cpumap.h" #include "debug.h" +#include "drm_pmu.h" #include "evsel.h" #include "pmus.h" #include "pmu.h" @@ -19,6 +20,7 @@ #include "tool_pmu.h" #include "print-events.h" #include "strbuf.h" +#include "string2.h" /* * core_pmus: A PMU belongs to core_pmus if it's name is "cpu" or it's sysfs @@ -42,16 +44,19 @@ enum perf_tool_pmu_type { PERF_TOOL_PMU_TYPE_PE_OTHER, PERF_TOOL_PMU_TYPE_TOOL, PERF_TOOL_PMU_TYPE_HWMON, + PERF_TOOL_PMU_TYPE_DRM, #define PERF_TOOL_PMU_TYPE_PE_CORE_MASK (1 << PERF_TOOL_PMU_TYPE_PE_CORE) #define PERF_TOOL_PMU_TYPE_PE_OTHER_MASK (1 << PERF_TOOL_PMU_TYPE_PE_OTHER) #define PERF_TOOL_PMU_TYPE_TOOL_MASK (1 << PERF_TOOL_PMU_TYPE_TOOL) #define PERF_TOOL_PMU_TYPE_HWMON_MASK (1 << PERF_TOOL_PMU_TYPE_HWMON) +#define PERF_TOOL_PMU_TYPE_DRM_MASK (1 << PERF_TOOL_PMU_TYPE_DRM) #define PERF_TOOL_PMU_TYPE_ALL_MASK (PERF_TOOL_PMU_TYPE_PE_CORE_MASK | \ PERF_TOOL_PMU_TYPE_PE_OTHER_MASK | \ PERF_TOOL_PMU_TYPE_TOOL_MASK | \ - PERF_TOOL_PMU_TYPE_HWMON_MASK) + PERF_TOOL_PMU_TYPE_HWMON_MASK | \ + PERF_TOOL_PMU_TYPE_DRM_MASK) }; static unsigned int read_pmu_types; @@ -172,6 +177,8 @@ struct perf_pmu *perf_pmus__find(const char *name) /* Looking up an individual perf event PMU failed, check if a tool PMU should be read. */ if (!strncmp(name, "hwmon_", 6)) to_read_pmus |= PERF_TOOL_PMU_TYPE_HWMON_MASK; + else if (!strncmp(name, "drm_", 4)) + to_read_pmus |= PERF_TOOL_PMU_TYPE_DRM_MASK; else if (!strcmp(name, "tool")) to_read_pmus |= PERF_TOOL_PMU_TYPE_TOOL_MASK; @@ -272,6 +279,10 @@ skip_pe_pmus: (read_pmu_types & PERF_TOOL_PMU_TYPE_HWMON_MASK) == 0) perf_pmus__read_hwmon_pmus(&other_pmus); + if ((to_read_types & PERF_TOOL_PMU_TYPE_DRM_MASK) != 0 && + (read_pmu_types & PERF_TOOL_PMU_TYPE_DRM_MASK) == 0) + perf_pmus__read_drm_pmus(&other_pmus); + list_sort(NULL, &other_pmus, pmus_cmp); read_pmu_types |= to_read_types; @@ -304,6 +315,8 @@ struct perf_pmu *perf_pmus__find_by_type(unsigned int type) if (type >= PERF_PMU_TYPE_PE_START && type <= PERF_PMU_TYPE_PE_END) { to_read_pmus = PERF_TOOL_PMU_TYPE_PE_CORE_MASK | PERF_TOOL_PMU_TYPE_PE_OTHER_MASK; + } else if (type >= PERF_PMU_TYPE_DRM_START && type <= PERF_PMU_TYPE_DRM_END) { + to_read_pmus = PERF_TOOL_PMU_TYPE_DRM_MASK; } else if (type >= PERF_PMU_TYPE_HWMON_START && type <= PERF_PMU_TYPE_HWMON_END) { to_read_pmus = PERF_TOOL_PMU_TYPE_HWMON_MASK; } else { @@ -350,6 +363,92 @@ struct perf_pmu *perf_pmus__scan_core(struct perf_pmu *pmu) return NULL; } +struct perf_pmu *perf_pmus__scan_for_event(struct perf_pmu *pmu, const char *event) +{ + bool use_core_pmus = !pmu || pmu->is_core; + + if (!pmu) { + /* Hwmon filename values that aren't used. */ + enum hwmon_type type; + int number; + /* + * Core PMUs, other sysfs PMUs and tool PMU can take all event + * types or aren't wother optimizing for. + */ + unsigned int to_read_pmus = PERF_TOOL_PMU_TYPE_PE_CORE_MASK | + PERF_TOOL_PMU_TYPE_PE_OTHER_MASK | + PERF_TOOL_PMU_TYPE_TOOL_MASK; + + /* Could the event be a hwmon event? */ + if (parse_hwmon_filename(event, &type, &number, /*item=*/NULL, /*alarm=*/NULL)) + to_read_pmus |= PERF_TOOL_PMU_TYPE_HWMON_MASK; + + /* Could the event be a DRM event? */ + if (strlen(event) > 4 && strncmp("drm-", event, 4) == 0) + to_read_pmus |= PERF_TOOL_PMU_TYPE_DRM_MASK; + + pmu_read_sysfs(to_read_pmus); + pmu = list_prepare_entry(pmu, &core_pmus, list); + } + if (use_core_pmus) { + list_for_each_entry_continue(pmu, &core_pmus, list) + return pmu; + + pmu = NULL; + pmu = list_prepare_entry(pmu, &other_pmus, list); + } + list_for_each_entry_continue(pmu, &other_pmus, list) + return pmu; + return NULL; +} + +struct perf_pmu *perf_pmus__scan_matching_wildcard(struct perf_pmu *pmu, const char *wildcard) +{ + bool use_core_pmus = !pmu || pmu->is_core; + + if (!pmu) { + /* + * Core PMUs, other sysfs PMUs and tool PMU can have any name or + * aren't wother optimizing for. + */ + unsigned int to_read_pmus = PERF_TOOL_PMU_TYPE_PE_CORE_MASK | + PERF_TOOL_PMU_TYPE_PE_OTHER_MASK | + PERF_TOOL_PMU_TYPE_TOOL_MASK; + + /* + * Hwmon PMUs have an alias from a sysfs name like hwmon0, + * hwmon1, etc. or have a name of hwmon_<name>. They therefore + * can only have a wildcard match if the wildcard begins with + * "hwmon". Similarly drm PMUs must start "drm_", avoid reading + * such events unless the PMU could match. + */ + if (strisglob(wildcard)) { + to_read_pmus |= PERF_TOOL_PMU_TYPE_HWMON_MASK | + PERF_TOOL_PMU_TYPE_DRM_MASK; + } else if (strlen(wildcard) >= 4 && strncmp("drm_", wildcard, 4) == 0) { + to_read_pmus |= PERF_TOOL_PMU_TYPE_DRM_MASK; + } else if (strlen(wildcard) >= 5 && strncmp("hwmon", wildcard, 5) == 0) { + to_read_pmus |= PERF_TOOL_PMU_TYPE_HWMON_MASK; + } + + pmu_read_sysfs(to_read_pmus); + pmu = list_prepare_entry(pmu, &core_pmus, list); + } + if (use_core_pmus) { + list_for_each_entry_continue(pmu, &core_pmus, list) { + if (perf_pmu__wildcard_match(pmu, wildcard)) + return pmu; + } + pmu = NULL; + pmu = list_prepare_entry(pmu, &other_pmus, list); + } + list_for_each_entry_continue(pmu, &other_pmus, list) { + if (perf_pmu__wildcard_match(pmu, wildcard)) + return pmu; + } + return NULL; +} + static struct perf_pmu *perf_pmus__scan_skip_duplicates(struct perf_pmu *pmu) { bool use_core_pmus = !pmu || pmu->is_core; @@ -546,6 +645,7 @@ void perf_pmus__print_pmu_events(const struct print_callbacks *print_cb, void *p print_cb->print_event(print_state, aliases[j].topic, aliases[j].pmu_name, + aliases[j].pmu->type, aliases[j].name, aliases[j].alias, aliases[j].scale_unit, @@ -650,6 +750,7 @@ void perf_pmus__print_raw_pmu_events(const struct print_callbacks *print_cb, voi print_cb->print_event(print_state, /*topic=*/NULL, /*pmu_name=*/NULL, + pmu->type, format_args.short_string.buf, /*event_alias=*/NULL, /*scale_unit=*/NULL, @@ -715,24 +816,18 @@ bool perf_pmus__supports_extended_type(void) return perf_pmus__do_support_extended_type; } -struct perf_pmu *evsel__find_pmu(const struct evsel *evsel) +struct perf_pmu *perf_pmus__find_by_attr(const struct perf_event_attr *attr) { - struct perf_pmu *pmu = evsel->pmu; - bool legacy_core_type; - - if (pmu) - return pmu; + struct perf_pmu *pmu = perf_pmus__find_by_type(attr->type); + u32 type = attr->type; + bool legacy_core_type = type == PERF_TYPE_HARDWARE || type == PERF_TYPE_HW_CACHE; - pmu = perf_pmus__find_by_type(evsel->core.attr.type); - legacy_core_type = - evsel->core.attr.type == PERF_TYPE_HARDWARE || - evsel->core.attr.type == PERF_TYPE_HW_CACHE; if (!pmu && legacy_core_type && perf_pmus__supports_extended_type()) { - u32 type = evsel->core.attr.config >> PERF_PMU_TYPE_SHIFT; + type = attr->config >> PERF_PMU_TYPE_SHIFT; pmu = perf_pmus__find_by_type(type); } - if (!pmu && (legacy_core_type || evsel->core.attr.type == PERF_TYPE_RAW)) { + if (!pmu && (legacy_core_type || type == PERF_TYPE_RAW)) { /* * For legacy events, if there was no extended type info then * assume the PMU is the first core PMU. @@ -743,6 +838,17 @@ struct perf_pmu *evsel__find_pmu(const struct evsel *evsel) */ pmu = perf_pmus__find_core_pmu(); } + return pmu; +} + +struct perf_pmu *evsel__find_pmu(const struct evsel *evsel) +{ + struct perf_pmu *pmu = evsel->pmu; + + if (pmu) + return pmu; + + pmu = perf_pmus__find_by_attr(&evsel->core.attr); ((struct evsel *)evsel)->pmu = pmu; return pmu; } @@ -762,7 +868,7 @@ struct perf_pmu *perf_pmus__add_test_pmu(int test_sysfs_dirfd, const char *name) return perf_pmu__lookup(&other_pmus, test_sysfs_dirfd, name, /*eager_load=*/true); } -struct perf_pmu *perf_pmus__add_test_hwmon_pmu(int hwmon_dir, +struct perf_pmu *perf_pmus__add_test_hwmon_pmu(const char *hwmon_dir, const char *sysfs_name, const char *name) { diff --git a/tools/perf/util/pmus.h b/tools/perf/util/pmus.h index 8def20e615ad..7cb36863711a 100644 --- a/tools/perf/util/pmus.h +++ b/tools/perf/util/pmus.h @@ -5,6 +5,7 @@ #include <stdbool.h> #include <stddef.h> +struct perf_event_attr; struct perf_pmu; struct print_callbacks; @@ -16,9 +17,12 @@ void perf_pmus__destroy(void); struct perf_pmu *perf_pmus__find(const char *name); struct perf_pmu *perf_pmus__find_by_type(unsigned int type); +struct perf_pmu *perf_pmus__find_by_attr(const struct perf_event_attr *attr); struct perf_pmu *perf_pmus__scan(struct perf_pmu *pmu); struct perf_pmu *perf_pmus__scan_core(struct perf_pmu *pmu); +struct perf_pmu *perf_pmus__scan_for_event(struct perf_pmu *pmu, const char *event); +struct perf_pmu *perf_pmus__scan_matching_wildcard(struct perf_pmu *pmu, const char *wildcard); const struct perf_pmu *perf_pmus__pmu_for_pmu_filter(const char *str); @@ -29,9 +33,10 @@ int perf_pmus__num_core_pmus(void); bool perf_pmus__supports_extended_type(void); struct perf_pmu *perf_pmus__add_test_pmu(int test_sysfs_dirfd, const char *name); -struct perf_pmu *perf_pmus__add_test_hwmon_pmu(int hwmon_dir, +struct perf_pmu *perf_pmus__add_test_hwmon_pmu(const char *hwmon_dir, const char *sysfs_name, const char *name); struct perf_pmu *perf_pmus__fake_pmu(void); +struct perf_pmu *perf_pmus__find_core_pmu(void); #endif /* __PMUS_H */ diff --git a/tools/perf/util/print-events.c b/tools/perf/util/print-events.c index a786cbfb0ff5..4153124a9948 100644 --- a/tools/perf/util/print-events.c +++ b/tools/perf/util/print-events.c @@ -44,97 +44,6 @@ static const char * const event_type_descriptors[] = { "Hardware breakpoint", }; -/* - * Print the events from <debugfs_mount_point>/tracing/events - */ -void print_tracepoint_events(const struct print_callbacks *print_cb __maybe_unused, void *print_state __maybe_unused) -{ - char *events_path = get_tracing_file("events"); - int events_fd = open(events_path, O_PATH); - struct dirent **sys_namelist = NULL; - int sys_items; - - if (events_fd < 0) { - pr_err("Error: failed to open tracing events directory\n"); - pr_err("%s: %s\n", events_path, strerror(errno)); - return; - } - put_tracing_file(events_path); - - sys_items = tracing_events__scandir_alphasort(&sys_namelist); - - for (int i = 0; i < sys_items; i++) { - struct dirent *sys_dirent = sys_namelist[i]; - struct dirent **evt_namelist = NULL; - int dir_fd; - int evt_items; - - if (sys_dirent->d_type != DT_DIR || - !strcmp(sys_dirent->d_name, ".") || - !strcmp(sys_dirent->d_name, "..")) - goto next_sys; - - dir_fd = openat(events_fd, sys_dirent->d_name, O_PATH); - if (dir_fd < 0) - goto next_sys; - - evt_items = scandirat(events_fd, sys_dirent->d_name, &evt_namelist, NULL, alphasort); - for (int j = 0; j < evt_items; j++) { - /* - * Buffer sized at twice the max filename length + 1 - * separator + 1 \0 terminator. - */ - char buf[NAME_MAX * 2 + 2]; - /* 16 possible hex digits and 22 other characters and \0. */ - char encoding[16 + 22]; - struct dirent *evt_dirent = evt_namelist[j]; - struct io id; - __u64 config; - - if (evt_dirent->d_type != DT_DIR || - !strcmp(evt_dirent->d_name, ".") || - !strcmp(evt_dirent->d_name, "..")) - goto next_evt; - - snprintf(buf, sizeof(buf), "%s/id", evt_dirent->d_name); - io__init(&id, openat(dir_fd, buf, O_RDONLY), buf, sizeof(buf)); - - if (id.fd < 0) - goto next_evt; - - if (io__get_dec(&id, &config) < 0) { - close(id.fd); - goto next_evt; - } - close(id.fd); - - snprintf(buf, sizeof(buf), "%s:%s", - sys_dirent->d_name, evt_dirent->d_name); - snprintf(encoding, sizeof(encoding), "tracepoint/config=0x%llx/", config); - print_cb->print_event(print_state, - /*topic=*/NULL, - /*pmu_name=*/NULL, /* really "tracepoint" */ - /*event_name=*/buf, - /*event_alias=*/NULL, - /*scale_unit=*/NULL, - /*deprecated=*/false, - "Tracepoint event", - /*desc=*/NULL, - /*long_desc=*/NULL, - encoding); -next_evt: - free(evt_namelist[j]); - } - close(dir_fd); - free(evt_namelist); -next_sys: - free(sys_namelist[i]); - } - - free(sys_namelist); - close(events_fd); -} - void print_sdt_events(const struct print_callbacks *print_cb, void *print_state) { struct strlist *bidlist, *sdtlist; @@ -212,6 +121,7 @@ void print_sdt_events(const struct print_callbacks *print_cb, void *print_state) print_cb->print_event(print_state, /*topic=*/NULL, /*pmu_name=*/NULL, + PERF_TYPE_TRACEPOINT, evt_name ?: sdt_name->s, /*event_alias=*/NULL, /*deprecated=*/false, @@ -268,6 +178,7 @@ bool is_event_supported(u8 type, u64 config) ret = evsel__open(evsel, NULL, tmap) >= 0; } + evsel__close(evsel); evsel__delete(evsel); } @@ -312,6 +223,7 @@ int print_hwcache_events(const struct print_callbacks *print_cb, void *print_sta print_cb->print_event(print_state, "cache", pmu->name, + pmu->type, name, alias_name, /*scale_unit=*/NULL, @@ -368,6 +280,7 @@ void print_symbol_events(const struct print_callbacks *print_cb, void *print_sta print_cb->print_event(print_state, /*topic=*/NULL, /*pmu_name=*/NULL, + type, nd->s, alias, /*scale_unit=*/NULL, @@ -380,6 +293,139 @@ void print_symbol_events(const struct print_callbacks *print_cb, void *print_sta strlist__delete(evt_name_list); } +/** struct mep - RB-tree node for building printing information. */ +struct mep { + /** nd - RB-tree element. */ + struct rb_node nd; + /** @metric_group: Owned metric group name, separated others with ';'. */ + char *metric_group; + const char *metric_name; + const char *metric_desc; + const char *metric_long_desc; + const char *metric_expr; + const char *metric_threshold; + const char *metric_unit; + const char *pmu_name; +}; + +static int mep_cmp(struct rb_node *rb_node, const void *entry) +{ + struct mep *a = container_of(rb_node, struct mep, nd); + struct mep *b = (struct mep *)entry; + int ret; + + ret = strcmp(a->metric_group, b->metric_group); + if (ret) + return ret; + + return strcmp(a->metric_name, b->metric_name); +} + +static struct rb_node *mep_new(struct rblist *rl __maybe_unused, const void *entry) +{ + struct mep *me = malloc(sizeof(struct mep)); + + if (!me) + return NULL; + + memcpy(me, entry, sizeof(struct mep)); + return &me->nd; +} + +static void mep_delete(struct rblist *rl __maybe_unused, + struct rb_node *nd) +{ + struct mep *me = container_of(nd, struct mep, nd); + + zfree(&me->metric_group); + free(me); +} + +static struct mep *mep_lookup(struct rblist *groups, const char *metric_group, + const char *metric_name) +{ + struct rb_node *nd; + struct mep me = { + .metric_group = strdup(metric_group), + .metric_name = metric_name, + }; + nd = rblist__find(groups, &me); + if (nd) { + free(me.metric_group); + return container_of(nd, struct mep, nd); + } + rblist__add_node(groups, &me); + nd = rblist__find(groups, &me); + if (nd) + return container_of(nd, struct mep, nd); + return NULL; +} + +static int metricgroup__add_to_mep_groups_callback(const struct pmu_metric *pm, + const struct pmu_metrics_table *table __maybe_unused, + void *vdata) +{ + struct rblist *groups = vdata; + const char *g; + char *omg, *mg; + + mg = strdup(pm->metric_group ?: pm->metric_name); + if (!mg) + return -ENOMEM; + omg = mg; + while ((g = strsep(&mg, ";")) != NULL) { + struct mep *me; + + g = skip_spaces(g); + if (strlen(g)) + me = mep_lookup(groups, g, pm->metric_name); + else + me = mep_lookup(groups, pm->metric_name, pm->metric_name); + + if (me) { + me->metric_desc = pm->desc; + me->metric_long_desc = pm->long_desc; + me->metric_expr = pm->metric_expr; + me->metric_threshold = pm->metric_threshold; + me->metric_unit = pm->unit; + me->pmu_name = pm->pmu; + } + } + free(omg); + + return 0; +} + +void metricgroup__print(const struct print_callbacks *print_cb, void *print_state) +{ + struct rblist groups; + struct rb_node *node, *next; + const struct pmu_metrics_table *table = pmu_metrics_table__find(); + + rblist__init(&groups); + groups.node_new = mep_new; + groups.node_cmp = mep_cmp; + groups.node_delete = mep_delete; + + metricgroup__for_each_metric(table, metricgroup__add_to_mep_groups_callback, &groups); + + for (node = rb_first_cached(&groups.entries); node; node = next) { + struct mep *me = container_of(node, struct mep, nd); + + print_cb->print_metric(print_state, + me->metric_group, + me->metric_name, + me->metric_desc, + me->metric_long_desc, + me->metric_expr, + me->metric_threshold, + me->metric_unit, + me->pmu_name); + next = rb_next(node); + rblist__remove_node(&groups, node); + } +} + /* * Print the help text for the event symbols: */ @@ -387,8 +433,6 @@ void print_events(const struct print_callbacks *print_cb, void *print_state) { print_symbol_events(print_cb, print_state, PERF_TYPE_HARDWARE, event_symbols_hw, PERF_COUNT_HW_MAX); - print_symbol_events(print_cb, print_state, PERF_TYPE_SOFTWARE, - event_symbols_sw, PERF_COUNT_SW_MAX); print_hwcache_events(print_cb, print_state); @@ -397,6 +441,7 @@ void print_events(const struct print_callbacks *print_cb, void *print_state) print_cb->print_event(print_state, /*topic=*/NULL, /*pmu_name=*/NULL, + PERF_TYPE_RAW, "rNNN", /*event_alias=*/NULL, /*scale_unit=*/NULL, @@ -411,6 +456,7 @@ void print_events(const struct print_callbacks *print_cb, void *print_state) print_cb->print_event(print_state, /*topic=*/NULL, /*pmu_name=*/NULL, + PERF_TYPE_BREAKPOINT, "mem:<addr>[/len][:access]", /*scale_unit=*/NULL, /*event_alias=*/NULL, @@ -420,8 +466,6 @@ void print_events(const struct print_callbacks *print_cb, void *print_state) /*long_desc=*/NULL, /*encoding_desc=*/NULL); - print_tracepoint_events(print_cb, print_state); - print_sdt_events(print_cb, print_state); metricgroup__print(print_cb, print_state); diff --git a/tools/perf/util/print-events.h b/tools/perf/util/print-events.h index 8f19c2bea64a..d6ba384f0c66 100644 --- a/tools/perf/util/print-events.h +++ b/tools/perf/util/print-events.h @@ -12,7 +12,7 @@ struct print_callbacks { void (*print_start)(void *print_state); void (*print_end)(void *print_state); void (*print_event)(void *print_state, const char *topic, - const char *pmu_name, + const char *pmu_name, u32 pmu_type, const char *event_name, const char *event_alias, const char *scale_unit, bool deprecated, const char *event_type_desc, @@ -37,7 +37,7 @@ void print_sdt_events(const struct print_callbacks *print_cb, void *print_state) void print_symbol_events(const struct print_callbacks *print_cb, void *print_state, unsigned int type, const struct event_symbol *syms, unsigned int max); -void print_tracepoint_events(const struct print_callbacks *print_cb, void *print_state); +void metricgroup__print(const struct print_callbacks *print_cb, void *print_state); bool is_event_supported(u8 type, u64 config); #endif /* __PERF_PRINT_EVENTS_H */ diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index 307ad6242a4e..6ab2eb551b6c 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -75,12 +75,14 @@ int e_snprintf(char *str, size_t size, const char *format, ...) } static struct machine *host_machine; +static struct perf_env host_env; /* Initialize symbol maps and path of vmlinux/modules */ int init_probe_symbol_maps(bool user_only) { int ret; + perf_env__init(&host_env); symbol_conf.allow_aliases = true; ret = symbol__init(NULL); if (ret < 0) { @@ -94,7 +96,7 @@ int init_probe_symbol_maps(bool user_only) if (symbol_conf.vmlinux_name) pr_debug("Use vmlinux: %s\n", symbol_conf.vmlinux_name); - host_machine = machine__new_host(); + host_machine = machine__new_host(&host_env); if (!host_machine) { pr_debug("machine__new_host() failed.\n"); symbol__exit(); @@ -111,6 +113,7 @@ void exit_probe_symbol_maps(void) machine__delete(host_machine); host_machine = NULL; symbol__exit(); + perf_env__exit(&host_env); } static struct ref_reloc_sym *kernel_get_ref_reloc_sym(struct map **pmap) @@ -502,7 +505,7 @@ static struct debuginfo *open_from_debuginfod(struct dso *dso, struct nsinfo *ns if (!c) return NULL; - build_id__sprintf(dso__bid(dso), sbuild_id); + build_id__snprintf(dso__bid(dso), sbuild_id, sizeof(sbuild_id)); fd = debuginfod_find_debuginfo(c, (const unsigned char *)sbuild_id, 0, &path); if (fd >= 0) @@ -1063,7 +1066,6 @@ static int sprint_line_description(char *sbuf, size_t size, struct line_range *l static int __show_line_range(struct line_range *lr, const char *module, bool user) { - struct build_id bid; int l = 1; struct int_node *ln; struct debuginfo *dinfo; @@ -1088,8 +1090,10 @@ static int __show_line_range(struct line_range *lr, const char *module, ret = -ENOENT; } if (dinfo->build_id) { + struct build_id bid; + build_id__init(&bid, dinfo->build_id, BUILD_ID_SIZE); - build_id__sprintf(&bid, sbuild_id); + build_id__snprintf(&bid, sbuild_id, sizeof(sbuild_id)); } debuginfo__delete(dinfo); if (ret == 0 || ret == -ENOENT) { diff --git a/tools/perf/util/probe-file.c b/tools/perf/util/probe-file.c index ec8ac242fedb..5069fb61f48c 100644 --- a/tools/perf/util/probe-file.c +++ b/tools/perf/util/probe-file.c @@ -448,10 +448,10 @@ static int probe_cache__open(struct probe_cache *pcache, const char *target, if (!target || !strcmp(target, DSO__NAME_KALLSYMS)) { target = DSO__NAME_KALLSYMS; is_kallsyms = true; - ret = sysfs__sprintf_build_id("/", sbuildid); + ret = sysfs__snprintf_build_id("/", sbuildid, sizeof(sbuildid)); } else { nsinfo__mountns_enter(nsi, &nsc); - ret = filename__sprintf_build_id(target, sbuildid); + ret = filename__snprintf_build_id(target, sbuildid, sizeof(sbuildid)); nsinfo__mountns_exit(&nsc); } diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index 3cc7c40f5097..5ffd97ee4898 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -848,7 +848,6 @@ static int probe_point_lazy_walker(const char *fname, int lineno, /* Find probe points from lazy pattern */ static int find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf) { - struct build_id bid; char sbuild_id[SBUILD_ID_SIZE] = ""; int ret = 0; char *fpath; @@ -858,8 +857,10 @@ static int find_probe_point_lazy(Dwarf_Die *sp_die, struct probe_finder *pf) comp_dir = cu_get_comp_dir(&pf->cu_die); if (pf->dbg->build_id) { + struct build_id bid; + build_id__init(&bid, pf->dbg->build_id, BUILD_ID_SIZE); - build_id__sprintf(&bid, sbuild_id); + build_id__snprintf(&bid, sbuild_id, sizeof(sbuild_id)); } ret = find_source_path(pf->fname, sbuild_id, comp_dir, &fpath); if (ret < 0) { diff --git a/tools/perf/util/python.c b/tools/perf/util/python.c index 321c333877fa..ea77bea0306f 100644 --- a/tools/perf/util/python.c +++ b/tools/perf/util/python.c @@ -10,6 +10,7 @@ #endif #include <perf/mmap.h> #include "callchain.h" +#include "counts.h" #include "evlist.h" #include "evsel.h" #include "event.h" @@ -18,6 +19,7 @@ #include "strbuf.h" #include "thread_map.h" #include "trace-event.h" +#include "metricgroup.h" #include "mmap.h" #include "util/sample.h" #include <internal/lib.h> @@ -335,7 +337,6 @@ tracepoint_field(const struct pyrf_event *pe, struct tep_format_field *field) static PyObject* get_tracepoint_field(struct pyrf_event *pevent, PyObject *attr_name) { - const char *str = _PyUnicode_AsString(PyObject_Str(attr_name)); struct evsel *evsel = pevent->evsel; struct tep_event *tp_format = evsel__tp_format(evsel); struct tep_format_field *field; @@ -343,7 +344,18 @@ get_tracepoint_field(struct pyrf_event *pevent, PyObject *attr_name) if (IS_ERR_OR_NULL(tp_format)) return NULL; + PyObject *obj = PyObject_Str(attr_name); + if (obj == NULL) + return NULL; + + const char *str = PyUnicode_AsUTF8(obj); + if (str == NULL) { + Py_DECREF(obj); + return NULL; + } + field = tep_find_any_field(tp_format, str); + Py_DECREF(obj); return field ? tracepoint_field(pevent, field) : NULL; } #endif /* HAVE_LIBTRACEEVENT */ @@ -527,8 +539,10 @@ static PyObject *pyrf_cpu_map__item(PyObject *obj, Py_ssize_t i) { struct pyrf_cpu_map *pcpus = (void *)obj; - if (i >= perf_cpu_map__nr(pcpus->cpus)) + if (i >= perf_cpu_map__nr(pcpus->cpus)) { + PyErr_SetString(PyExc_IndexError, "Index out of range"); return NULL; + } return Py_BuildValue("i", perf_cpu_map__cpu(pcpus->cpus, i).cpu); } @@ -566,14 +580,14 @@ struct pyrf_thread_map { static int pyrf_thread_map__init(struct pyrf_thread_map *pthreads, PyObject *args, PyObject *kwargs) { - static char *kwlist[] = { "pid", "tid", "uid", NULL }; - int pid = -1, tid = -1, uid = UINT_MAX; + static char *kwlist[] = { "pid", "tid", NULL }; + int pid = -1, tid = -1; - if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|iii", - kwlist, &pid, &tid, &uid)) + if (!PyArg_ParseTupleAndKeywords(args, kwargs, "|ii", + kwlist, &pid, &tid)) return -1; - pthreads->threads = thread_map__new(pid, tid, uid); + pthreads->threads = thread_map__new(pid, tid); if (pthreads->threads == NULL) return -1; return 0; @@ -596,8 +610,10 @@ static PyObject *pyrf_thread_map__item(PyObject *obj, Py_ssize_t i) { struct pyrf_thread_map *pthreads = (void *)obj; - if (i >= perf_thread_map__nr(pthreads->threads)) + if (i >= perf_thread_map__nr(pthreads->threads)) { + PyErr_SetString(PyExc_IndexError, "Index out of range"); return NULL; + } return Py_BuildValue("i", perf_thread_map__pid(pthreads->threads, i)); } @@ -888,12 +904,38 @@ static PyObject *pyrf_evsel__threads(struct pyrf_evsel *pevsel) return (PyObject *)pthread_map; } +/* + * Ensure evsel's counts and prev_raw_counts are allocated, the latter + * used by tool PMUs to compute the cumulative count as expected by + * stat's process_counter_values. + */ +static int evsel__ensure_counts(struct evsel *evsel) +{ + int nthreads, ncpus; + + if (evsel->counts != NULL) + return 0; + + nthreads = perf_thread_map__nr(evsel->core.threads); + ncpus = perf_cpu_map__nr(evsel->core.cpus); + + evsel->counts = perf_counts__new(ncpus, nthreads); + if (evsel->counts == NULL) + return -ENOMEM; + + evsel->prev_raw_counts = perf_counts__new(ncpus, nthreads); + if (evsel->prev_raw_counts == NULL) + return -ENOMEM; + + return 0; +} + static PyObject *pyrf_evsel__read(struct pyrf_evsel *pevsel, PyObject *args, PyObject *kwargs) { struct evsel *evsel = &pevsel->evsel; int cpu = 0, cpu_idx, thread = 0, thread_idx; - struct perf_counts_values counts; + struct perf_counts_values *old_count, *new_count; struct pyrf_counts_values *count_values = PyObject_New(struct pyrf_counts_values, &pyrf_counts_values__type); @@ -909,13 +951,27 @@ static PyObject *pyrf_evsel__read(struct pyrf_evsel *pevsel, return NULL; } thread_idx = perf_thread_map__idx(evsel->core.threads, thread); - if (cpu_idx < 0) { + if (thread_idx < 0) { PyErr_Format(PyExc_TypeError, "Thread %d is not part of evsel's threads", thread); return NULL; } - perf_evsel__read(&(evsel->core), cpu_idx, thread_idx, &counts); - count_values->values = counts; + + if (evsel__ensure_counts(evsel)) + return PyErr_NoMemory(); + + /* Set up pointers to the old and newly read counter values. */ + old_count = perf_counts(evsel->prev_raw_counts, cpu_idx, thread_idx); + new_count = perf_counts(evsel->counts, cpu_idx, thread_idx); + /* Update the value in evsel->counts. */ + evsel__read_counter(evsel, cpu_idx, thread_idx); + /* Copy the value and turn it into the delta from old_count. */ + count_values->values = *new_count; + count_values->values.val -= old_count->val; + count_values->values.ena -= old_count->ena; + count_values->values.run -= old_count->run; + /* Save the new count over the old_count for the next read. */ + *old_count = *new_count; return (PyObject *)count_values; } @@ -924,10 +980,7 @@ static PyObject *pyrf_evsel__str(PyObject *self) struct pyrf_evsel *pevsel = (void *)self; struct evsel *evsel = &pevsel->evsel; - if (!evsel->pmu) - return PyUnicode_FromFormat("evsel(%s)", evsel__name(evsel)); - - return PyUnicode_FromFormat("evsel(%s/%s/)", evsel->pmu->name, evsel__name(evsel)); + return PyUnicode_FromFormat("evsel(%s/%s/)", evsel__pmu_name(evsel), evsel__name(evsel)); } static PyMethodDef pyrf_evsel__methods[] = { @@ -1529,10 +1582,37 @@ static PyObject *pyrf_evsel__from_evsel(struct evsel *evsel) return (PyObject *)pevsel; } +static int evlist__pos(struct evlist *evlist, struct evsel *evsel) +{ + struct evsel *pos; + int idx = 0; + + evlist__for_each_entry(evlist, pos) { + if (evsel == pos) + return idx; + idx++; + } + return -1; +} + +static struct evsel *evlist__at(struct evlist *evlist, int idx) +{ + struct evsel *pos; + int idx2 = 0; + + evlist__for_each_entry(evlist, pos) { + if (idx == idx2) + return pos; + idx2++; + } + return NULL; +} + static PyObject *pyrf_evlist__from_evlist(struct evlist *evlist) { struct pyrf_evlist *pevlist = PyObject_New(struct pyrf_evlist, &pyrf_evlist__type); struct evsel *pos; + struct rb_node *node; if (!pevlist) return NULL; @@ -1544,6 +1624,39 @@ static PyObject *pyrf_evlist__from_evlist(struct evlist *evlist) evlist__add(&pevlist->evlist, &pevsel->evsel); } + evlist__for_each_entry(&pevlist->evlist, pos) { + struct evsel *leader = evsel__leader(pos); + + if (pos != leader) { + int idx = evlist__pos(evlist, leader); + + if (idx >= 0) + evsel__set_leader(pos, evlist__at(&pevlist->evlist, idx)); + else if (leader == NULL) + evsel__set_leader(pos, pos); + } + } + metricgroup__copy_metric_events(&pevlist->evlist, /*cgrp=*/NULL, + &pevlist->evlist.metric_events, + &evlist->metric_events); + for (node = rb_first_cached(&pevlist->evlist.metric_events.entries); node; + node = rb_next(node)) { + struct metric_event *me = container_of(node, struct metric_event, nd); + struct list_head *mpos; + int idx = evlist__pos(evlist, me->evsel); + + if (idx >= 0) + me->evsel = evlist__at(&pevlist->evlist, idx); + list_for_each(mpos, &me->head) { + struct metric_expr *e = container_of(mpos, struct metric_expr, nd); + + for (int j = 0; e->metric_events[j]; j++) { + idx = evlist__pos(evlist, e->metric_events[j]); + if (idx >= 0) + e->metric_events[j] = evlist__at(&pevlist->evlist, idx); + } + } + } return (PyObject *)pevlist; } diff --git a/tools/perf/util/s390-cpumsf.c b/tools/perf/util/s390-cpumsf.c index 0ce52f0280b8..c17dbe232c54 100644 --- a/tools/perf/util/s390-cpumsf.c +++ b/tools/perf/util/s390-cpumsf.c @@ -1142,7 +1142,7 @@ int s390_cpumsf_process_auxtrace_info(union perf_event *event, sf->machine = &session->machines.host; /* No kvm support */ sf->auxtrace_type = auxtrace_info->type; sf->pmu_type = PERF_TYPE_RAW; - sf->machine_type = s390_cpumsf_get_type(session->evlist->env->cpuid); + sf->machine_type = s390_cpumsf_get_type(perf_session__env(session)->cpuid); sf->auxtrace.process_event = s390_cpumsf_process_event; sf->auxtrace.process_auxtrace_event = s390_cpumsf_process_auxtrace_event; diff --git a/tools/perf/util/sample-raw.c b/tools/perf/util/sample-raw.c index f3f6bd9d290e..bcf442574d6e 100644 --- a/tools/perf/util/sample-raw.c +++ b/tools/perf/util/sample-raw.c @@ -6,15 +6,16 @@ #include "env.h" #include "header.h" #include "sample-raw.h" +#include "session.h" /* * Check platform the perf data file was created on and perform platform * specific interpretation. */ -void evlist__init_trace_event_sample_raw(struct evlist *evlist) +void evlist__init_trace_event_sample_raw(struct evlist *evlist, struct perf_env *env) { - const char *arch_pf = perf_env__arch(evlist->env); - const char *cpuid = perf_env__cpuid(evlist->env); + const char *arch_pf = perf_env__arch(env); + const char *cpuid = perf_env__cpuid(env); if (arch_pf && !strcmp("s390", arch_pf)) evlist->trace_event_sample_raw = evlist__s390_sample_raw; diff --git a/tools/perf/util/sample-raw.h b/tools/perf/util/sample-raw.h index ea01c5811503..896e9a87e373 100644 --- a/tools/perf/util/sample-raw.h +++ b/tools/perf/util/sample-raw.h @@ -11,5 +11,5 @@ void evlist__s390_sample_raw(struct evlist *evlist, union perf_event *event, bool evlist__has_amd_ibs(struct evlist *evlist); void evlist__amd_sample_raw(struct evlist *evlist, union perf_event *event, struct perf_sample *sample); -void evlist__init_trace_event_sample_raw(struct evlist *evlist); +void evlist__init_trace_event_sample_raw(struct evlist *evlist, struct perf_env *env); #endif /* __PERF_EVLIST_H */ diff --git a/tools/perf/util/sample.h b/tools/perf/util/sample.h index 0e96240052e9..fae834144ef4 100644 --- a/tools/perf/util/sample.h +++ b/tools/perf/util/sample.h @@ -104,10 +104,8 @@ struct perf_sample { u8 cpumode; u16 misc; u16 ins_lat; - union { - u16 p_stage_cyc; - u16 retire_lat; - }; + /** @weight3: On x86 holds retire_lat, on powerpc holds p_stage_cyc. */ + u16 weight3; bool no_hw_idx; /* No hw_idx collected in branch_stack */ char insn[MAX_INSN]; void *raw_data; diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c index 520729e78965..6655c0bbe0d8 100644 --- a/tools/perf/util/scripting-engines/trace-event-python.c +++ b/tools/perf/util/scripting-engines/trace-event-python.c @@ -780,14 +780,13 @@ static void set_sym_in_dict(PyObject *dict, struct addr_location *al, const char *sym_field, const char *symoff_field, const char *map_pgoff) { - char sbuild_id[SBUILD_ID_SIZE]; - if (al->map) { + char sbuild_id[SBUILD_ID_SIZE]; struct dso *dso = map__dso(al->map); pydict_set_item_string_decref(dict, dso_field, _PyUnicode_FromString(dso__name(dso))); - build_id__sprintf(dso__bid(dso), sbuild_id); + build_id__snprintf(dso__bid(dso), sbuild_id, sizeof(sbuild_id)); pydict_set_item_string_decref(dict, dso_bid_field, _PyUnicode_FromString(sbuild_id)); pydict_set_item_string_decref(dict, dso_map_start, @@ -1238,7 +1237,7 @@ static int python_export_dso(struct db_export *dbe, struct dso *dso, char sbuild_id[SBUILD_ID_SIZE]; PyObject *t; - build_id__sprintf(dso__bid(dso), sbuild_id); + build_id__snprintf(dso__bid(dso), sbuild_id, sizeof(sbuild_id)); t = tuple_new(5); @@ -1306,7 +1305,7 @@ static void python_export_sample_table(struct db_export *dbe, tuple_set_d64(t, 0, es->db_id); tuple_set_d64(t, 1, es->evsel->db_id); - tuple_set_d64(t, 2, maps__machine(es->al->maps)->db_id); + tuple_set_d64(t, 2, maps__machine(thread__maps(es->al->thread))->db_id); tuple_set_d64(t, 3, thread__db_id(es->al->thread)); tuple_set_d64(t, 4, es->comm_db_id); tuple_set_d64(t, 5, es->dso_db_id); diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index a320672c264e..26ae078278cd 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -12,6 +12,7 @@ #include <sys/types.h> #include <sys/mman.h> #include <perf/cpumap.h> +#include <perf/event.h> #include "map_symbol.h" #include "branch.h" @@ -137,7 +138,8 @@ static int ordered_events__deliver_event(struct ordered_events *oe, struct perf_session *__perf_session__new(struct perf_data *data, struct perf_tool *tool, - bool trace_event_repipe) + bool trace_event_repipe, + struct perf_env *host_env) { int ret = -ENOMEM; struct perf_session *session = zalloc(sizeof(*session)); @@ -176,7 +178,7 @@ struct perf_session *__perf_session__new(struct perf_data *data, perf_session__set_comm_exec(session); } - evlist__init_trace_event_sample_raw(session->evlist); + evlist__init_trace_event_sample_raw(session->evlist, &session->header.env); /* Open the directory data. */ if (data->is_dir) { @@ -190,8 +192,11 @@ struct perf_session *__perf_session__new(struct perf_data *data, symbol_conf.kallsyms_name = perf_data__kallsyms_name(data); } } else { - session->machines.host.env = &perf_env; + assert(host_env != NULL); + session->machines.host.env = host_env; } + if (session->evlist) + session->evlist->session = session; session->machines.host.single_address_space = perf_env__single_address_space(session->machines.host.env); @@ -1094,7 +1099,7 @@ static void dump_sample(struct evsel *evsel, union perf_event *event, printf("... weight: %" PRIu64 "", sample->weight); if (sample_type & PERF_SAMPLE_WEIGHT_STRUCT) { printf(",0x%"PRIx16"", sample->ins_lat); - printf(",0x%"PRIx16"", sample->p_stage_cyc); + printf(",0x%"PRIx16"", sample->weight3); } printf("\n"); } @@ -1491,6 +1496,9 @@ static s64 perf_session__process_user_event(struct perf_session *session, case PERF_RECORD_FINISHED_INIT: err = tool->finished_init(session, event); break; + case PERF_RECORD_BPF_METADATA: + err = tool->bpf_metadata(session, event); + break; default: err = -EINVAL; break; @@ -2554,7 +2562,7 @@ int perf_session__cpu_bitmap(struct perf_session *session, { int i, err = -1; struct perf_cpu_map *map; - int nr_cpus = min(session->header.env.nr_cpus_avail, MAX_NR_CPUS); + int nr_cpus = min(perf_session__env(session)->nr_cpus_avail, MAX_NR_CPUS); struct perf_cpu cpu; for (i = 0; i < PERF_TYPE_MAX; ++i) { @@ -2743,3 +2751,8 @@ int perf_session__dsos_hit_all(struct perf_session *session) return 0; } + +struct perf_env *perf_session__env(struct perf_session *session) +{ + return &session->header.env; +} diff --git a/tools/perf/util/session.h b/tools/perf/util/session.h index db1c120a9e67..cf88d65a25cb 100644 --- a/tools/perf/util/session.h +++ b/tools/perf/util/session.h @@ -107,12 +107,13 @@ struct perf_tool; struct perf_session *__perf_session__new(struct perf_data *data, struct perf_tool *tool, - bool trace_event_repipe); + bool trace_event_repipe, + struct perf_env *host_env); static inline struct perf_session *perf_session__new(struct perf_data *data, struct perf_tool *tool) { - return __perf_session__new(data, tool, /*trace_event_repipe=*/false); + return __perf_session__new(data, tool, /*trace_event_repipe=*/false, /*host_env=*/NULL); } void perf_session__delete(struct perf_session *session); @@ -208,4 +209,6 @@ int perf_event__process_finished_round(const struct perf_tool *tool, union perf_event *event, struct ordered_events *oe); +struct perf_env *perf_session__env(struct perf_session *session); + #endif /* __PERF_SESSION_H */ diff --git a/tools/perf/util/sha1.c b/tools/perf/util/sha1.c new file mode 100644 index 000000000000..7032fa4ff3fd --- /dev/null +++ b/tools/perf/util/sha1.c @@ -0,0 +1,97 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * SHA-1 message digest algorithm + * + * Copyright 2025 Google LLC + */ +#include <linux/bitops.h> +#include <linux/kernel.h> +#include <linux/unaligned.h> +#include <string.h> + +#include "sha1.h" + +#define SHA1_BLOCK_SIZE 64 + +static const u32 sha1_K[4] = { 0x5A827999, 0x6ED9EBA1, 0x8F1BBCDC, 0xCA62C1D6 }; + +#define SHA1_ROUND(i, a, b, c, d, e) \ + do { \ + if ((i) >= 16) \ + w[i] = rol32(w[(i) - 16] ^ w[(i) - 14] ^ w[(i) - 8] ^ \ + w[(i) - 3], \ + 1); \ + e += w[i] + rol32(a, 5) + sha1_K[(i) / 20]; \ + if ((i) < 20) \ + e += (b & (c ^ d)) ^ d; \ + else if ((i) < 40 || (i) >= 60) \ + e += b ^ c ^ d; \ + else \ + e += (c & d) ^ (b & (c ^ d)); \ + b = rol32(b, 30); \ + /* The new (a, b, c, d, e) is the old (e, a, b, c, d). */ \ + } while (0) + +#define SHA1_5ROUNDS(i) \ + do { \ + SHA1_ROUND((i) + 0, a, b, c, d, e); \ + SHA1_ROUND((i) + 1, e, a, b, c, d); \ + SHA1_ROUND((i) + 2, d, e, a, b, c); \ + SHA1_ROUND((i) + 3, c, d, e, a, b); \ + SHA1_ROUND((i) + 4, b, c, d, e, a); \ + } while (0) + +#define SHA1_20ROUNDS(i) \ + do { \ + SHA1_5ROUNDS((i) + 0); \ + SHA1_5ROUNDS((i) + 5); \ + SHA1_5ROUNDS((i) + 10); \ + SHA1_5ROUNDS((i) + 15); \ + } while (0) + +static void sha1_blocks(u32 h[5], const u8 *data, size_t nblocks) +{ + while (nblocks--) { + u32 a = h[0]; + u32 b = h[1]; + u32 c = h[2]; + u32 d = h[3]; + u32 e = h[4]; + u32 w[80]; + + for (int i = 0; i < 16; i++) + w[i] = get_unaligned_be32(&data[i * 4]); + SHA1_20ROUNDS(0); + SHA1_20ROUNDS(20); + SHA1_20ROUNDS(40); + SHA1_20ROUNDS(60); + + h[0] += a; + h[1] += b; + h[2] += c; + h[3] += d; + h[4] += e; + data += SHA1_BLOCK_SIZE; + } +} + +/* Calculate the SHA-1 message digest of the given data. */ +void sha1(const void *data, size_t len, u8 out[SHA1_DIGEST_SIZE]) +{ + u32 h[5] = { 0x67452301, 0xEFCDAB89, 0x98BADCFE, 0x10325476, + 0xC3D2E1F0 }; + u8 final_data[2 * SHA1_BLOCK_SIZE] = { 0 }; + size_t final_len = len % SHA1_BLOCK_SIZE; + + sha1_blocks(h, data, len / SHA1_BLOCK_SIZE); + + memcpy(final_data, data + len - final_len, final_len); + final_data[final_len] = 0x80; + final_len = round_up(final_len + 9, SHA1_BLOCK_SIZE); + put_unaligned_be64((u64)len * 8, &final_data[final_len - 8]); + + sha1_blocks(h, final_data, final_len / SHA1_BLOCK_SIZE); + + for (int i = 0; i < 5; i++) + put_unaligned_be32(h[i], &out[i * 4]); +} diff --git a/tools/perf/util/sha1.h b/tools/perf/util/sha1.h new file mode 100644 index 000000000000..e92c9966e1d5 --- /dev/null +++ b/tools/perf/util/sha1.h @@ -0,0 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +#include <linux/types.h> + +#define SHA1_DIGEST_SIZE 20 + +void sha1(const void *data, size_t len, u8 out[SHA1_DIGEST_SIZE]); diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c index 45e654653960..f3a565b0e230 100644 --- a/tools/perf/util/sort.c +++ b/tools/perf/util/sort.c @@ -1746,22 +1746,27 @@ sort__dcacheline_cmp(struct hist_entry *left, struct hist_entry *right) if (rc) return rc; /* - * Addresses with no major/minor numbers are assumed to be + * Addresses with no major/minor numbers or build ID are assumed to be * anonymous in userspace. Sort those on pid then address. * * The kernel and non-zero major/minor mapped areas are * assumed to be unity mapped. Sort those on address. */ + if (left->cpumode != PERF_RECORD_MISC_KERNEL && (map__flags(l_map) & MAP_SHARED) == 0) { + const struct dso_id *dso_id = dso__id_const(l_dso); - if ((left->cpumode != PERF_RECORD_MISC_KERNEL) && - (!(map__flags(l_map) & MAP_SHARED)) && !dso__id(l_dso)->maj && !dso__id(l_dso)->min && - !dso__id(l_dso)->ino && !dso__id(l_dso)->ino_generation) { - /* userspace anonymous */ + if (!dso_id->mmap2_valid) + dso_id = dso__id_const(r_dso); - if (thread__pid(left->thread) > thread__pid(right->thread)) - return -1; - if (thread__pid(left->thread) < thread__pid(right->thread)) - return 1; + if (!build_id__is_defined(&dso_id->build_id) && + (!dso_id->mmap2_valid || (dso_id->maj == 0 && dso_id->min == 0))) { + /* userspace anonymous */ + + if (thread__pid(left->thread) > thread__pid(right->thread)) + return -1; + if (thread__pid(left->thread) < thread__pid(right->thread)) + return 1; + } } addr: @@ -1786,6 +1791,7 @@ static int hist_entry__dcacheline_snprintf(struct hist_entry *he, char *bf, if (he->mem_info) { struct map *map = mem_info__daddr(he->mem_info)->ms.map; struct dso *dso = map ? map__dso(map) : NULL; + const struct dso_id *dso_id = dso ? dso__id_const(dso) : &dso_id_empty; addr = cl_address(mem_info__daddr(he->mem_info)->al_addr, chk_double_cl); ms = &mem_info__daddr(he->mem_info)->ms; @@ -1794,8 +1800,7 @@ static int hist_entry__dcacheline_snprintf(struct hist_entry *he, char *bf, if ((he->cpumode != PERF_RECORD_MISC_KERNEL) && map && !(map__prot(map) & PROT_EXEC) && (map__flags(map) & MAP_SHARED) && - (dso__id(dso)->maj || dso__id(dso)->min || dso__id(dso)->ino || - dso__id(dso)->ino_generation)) + (!dso_id->mmap2_valid || (dso_id->maj == 0 && dso_id->min == 0))) level = 's'; else if (!map) level = 'X'; @@ -1879,21 +1884,20 @@ struct sort_entry sort_global_ins_lat = { static int64_t sort__p_stage_cyc_cmp(struct hist_entry *left, struct hist_entry *right) { - return left->p_stage_cyc - right->p_stage_cyc; + return left->weight3 - right->weight3; } static int hist_entry__global_p_stage_cyc_snprintf(struct hist_entry *he, char *bf, size_t size, unsigned int width) { - return repsep_snprintf(bf, size, "%-*u", width, - he->p_stage_cyc * he->stat.nr_events); + return repsep_snprintf(bf, size, "%-*u", width, he->weight3 * he->stat.nr_events); } static int hist_entry__p_stage_cyc_snprintf(struct hist_entry *he, char *bf, size_t size, unsigned int width) { - return repsep_snprintf(bf, size, "%-*u", width, he->p_stage_cyc); + return repsep_snprintf(bf, size, "%-*u", width, he->weight3); } struct sort_entry sort_local_p_stage_cyc = { @@ -2526,19 +2530,44 @@ struct sort_dimension { int taken; }; -int __weak arch_support_sort_key(const char *sort_key __maybe_unused) +static int arch_support_sort_key(const char *sort_key, struct perf_env *env) { + const char *arch = perf_env__arch(env); + + if (!strcmp("x86", arch) || !strcmp("powerpc", arch)) { + if (!strcmp(sort_key, "p_stage_cyc")) + return 1; + if (!strcmp(sort_key, "local_p_stage_cyc")) + return 1; + } return 0; } -const char * __weak arch_perf_header_entry(const char *se_header) -{ +static const char *arch_perf_header_entry(const char *se_header, struct perf_env *env) +{ + const char *arch = perf_env__arch(env); + + if (!strcmp("x86", arch)) { + if (!strcmp(se_header, "Local Pipeline Stage Cycle")) + return "Local Retire Latency"; + else if (!strcmp(se_header, "Pipeline Stage Cycle")) + return "Retire Latency"; + } else if (!strcmp("powerpc", arch)) { + if (!strcmp(se_header, "Local INSTR Latency")) + return "Finish Cyc"; + else if (!strcmp(se_header, "INSTR Latency")) + return "Global Finish_cyc"; + else if (!strcmp(se_header, "Local Pipeline Stage Cycle")) + return "Dispatch Cyc"; + else if (!strcmp(se_header, "Pipeline Stage Cycle")) + return "Global Dispatch_cyc"; + } return se_header; } -static void sort_dimension_add_dynamic_header(struct sort_dimension *sd) +static void sort_dimension_add_dynamic_header(struct sort_dimension *sd, struct perf_env *env) { - sd->entry->se_header = arch_perf_header_entry(sd->entry->se_header); + sd->entry->se_header = arch_perf_header_entry(sd->entry->se_header, env); } #define DIM(d, n, func) [d] = { .name = n, .entry = &(func) } @@ -3590,7 +3619,7 @@ int hpp_dimension__add_output(unsigned col, bool implicit) } int sort_dimension__add(struct perf_hpp_list *list, const char *tok, - struct evlist *evlist, + struct evlist *evlist, struct perf_env *env, int level) { unsigned int i, j; @@ -3603,7 +3632,7 @@ int sort_dimension__add(struct perf_hpp_list *list, const char *tok, */ for (j = 0; j < ARRAY_SIZE(arch_specific_sort_keys); j++) { if (!strcmp(arch_specific_sort_keys[j], tok) && - !arch_support_sort_key(tok)) { + !arch_support_sort_key(tok, env)) { return 0; } } @@ -3616,7 +3645,7 @@ int sort_dimension__add(struct perf_hpp_list *list, const char *tok, for (j = 0; j < ARRAY_SIZE(dynamic_headers); j++) { if (sd->name && !strcmp(dynamic_headers[j], sd->name)) - sort_dimension_add_dynamic_header(sd); + sort_dimension_add_dynamic_header(sd, env); } if (sd->entry == &sort_parent && parent_pattern) { @@ -3712,13 +3741,13 @@ int sort_dimension__add(struct perf_hpp_list *list, const char *tok, } /* This should match with sort_dimension__add() above */ -static bool is_hpp_sort_key(const char *key) +static bool is_hpp_sort_key(const char *key, struct perf_env *env) { unsigned i; for (i = 0; i < ARRAY_SIZE(arch_specific_sort_keys); i++) { if (!strcmp(arch_specific_sort_keys[i], key) && - !arch_support_sort_key(key)) { + !arch_support_sort_key(key, env)) { return false; } } @@ -3740,7 +3769,7 @@ static bool is_hpp_sort_key(const char *key) } static int setup_sort_list(struct perf_hpp_list *list, char *str, - struct evlist *evlist) + struct evlist *evlist, struct perf_env *env) { char *tmp, *tok; int ret = 0; @@ -3769,7 +3798,7 @@ static int setup_sort_list(struct perf_hpp_list *list, char *str, } if (*tok) { - if (is_hpp_sort_key(tok)) { + if (is_hpp_sort_key(tok, env)) { /* keep output (hpp) sort keys in the same level */ if (prev_was_hpp) { bool next_same = (level == next_level); @@ -3782,7 +3811,7 @@ static int setup_sort_list(struct perf_hpp_list *list, char *str, prev_was_hpp = false; } - ret = sort_dimension__add(list, tok, evlist, level); + ret = sort_dimension__add(list, tok, evlist, env, level); if (ret == -EINVAL) { if (!cacheline_size() && !strncasecmp(tok, "dcacheline", strlen(tok))) ui__error("The \"dcacheline\" --sort key needs to know the cacheline size and it couldn't be determined on this system"); @@ -3911,7 +3940,7 @@ static char *setup_overhead(char *keys) return keys; } -static int __setup_sorting(struct evlist *evlist) +static int __setup_sorting(struct evlist *evlist, struct perf_env *env) { char *str; const char *sort_keys; @@ -3951,7 +3980,7 @@ static int __setup_sorting(struct evlist *evlist) } } - ret = setup_sort_list(&perf_hpp_list, str, evlist); + ret = setup_sort_list(&perf_hpp_list, str, evlist, env); free(str); return ret; @@ -4187,16 +4216,16 @@ out: return ret; } -int setup_sorting(struct evlist *evlist) +int setup_sorting(struct evlist *evlist, struct perf_env *env) { int err; - err = __setup_sorting(evlist); + err = __setup_sorting(evlist, env); if (err < 0) return err; if (parent_pattern != default_parent_pattern) { - err = sort_dimension__add(&perf_hpp_list, "parent", evlist, -1); + err = sort_dimension__add(&perf_hpp_list, "parent", evlist, env, -1); if (err < 0) return err; } diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h index a742ab7f3c67..d7787958e06b 100644 --- a/tools/perf/util/sort.h +++ b/tools/perf/util/sort.h @@ -6,6 +6,7 @@ #include "hist.h" struct option; +struct perf_env; extern regex_t parent_regex; extern const char *sort_order; @@ -130,7 +131,7 @@ extern struct sort_entry sort_thread; struct evlist; struct tep_handle; -int setup_sorting(struct evlist *evlist); +int setup_sorting(struct evlist *evlist, struct perf_env *env); int setup_output_field(void); void reset_output_field(void); void sort__setup_elide(FILE *fp); @@ -145,7 +146,7 @@ bool is_strict_order(const char *order); int hpp_dimension__add_output(unsigned col, bool implicit); void reset_dimensions(void); int sort_dimension__add(struct perf_hpp_list *list, const char *tok, - struct evlist *evlist, + struct evlist *evlist, struct perf_env *env, int level); int output_field_add(struct perf_hpp_list *list, const char *tok, int *level); int64_t diff --git a/tools/perf/util/spark.c b/tools/perf/util/spark.c index 70272a8b81a6..65ca253cc22e 100644 --- a/tools/perf/util/spark.c +++ b/tools/perf/util/spark.c @@ -1,9 +1,7 @@ -#include <stdio.h> -#include <limits.h> -#include <string.h> -#include <stdlib.h> +// SPDX-License-Identifier: GPL-2.0 #include "spark.h" -#include "stat.h" +#include <limits.h> +#include <linux/kernel.h> #define SPARK_SHIFT 8 diff --git a/tools/perf/util/spark.h b/tools/perf/util/spark.h index 25402d7d7a64..78597c38ef35 100644 --- a/tools/perf/util/spark.h +++ b/tools/perf/util/spark.h @@ -1,3 +1,4 @@ +/* SPDX-License-Identifier: GPL-2.0 */ #ifndef SPARK_H #define SPARK_H 1 diff --git a/tools/perf/util/srcline.c b/tools/perf/util/srcline.c index f32d0d4f4bc9..3e3449e35dd4 100644 --- a/tools/perf/util/srcline.c +++ b/tools/perf/util/srcline.c @@ -524,12 +524,12 @@ static enum a2l_style addr2line_configure(struct child_process *a2l, const char style = LLVM; cached = true; lines = 1; - pr_debug("Detected LLVM addr2line style\n"); + pr_debug3("Detected LLVM addr2line style\n"); } else if (ch == '0') { style = GNU_BINUTILS; cached = true; lines = 3; - pr_debug("Detected binutils addr2line style\n"); + pr_debug3("Detected binutils addr2line style\n"); } else { if (!symbol_conf.disable_add2line_warn) { char *output = NULL; @@ -595,7 +595,7 @@ static int read_addr2line_record(struct io *io, if (io__getline(io, &line, &line_len) < 0 || !line_len) goto error; - pr_debug("%s %s: addr2line read address for sentinel: %s", __func__, dso_name, line); + pr_debug3("%s %s: addr2line read address for sentinel: %s", __func__, dso_name, line); if (style == LLVM && line_len == 2 && line[0] == ',') { /* Found the llvm-addr2line sentinel character. */ zfree(&line); @@ -641,7 +641,7 @@ static int read_addr2line_record(struct io *io, if (first && (io__getline(io, &line, &line_len) < 0 || !line_len)) goto error; - pr_debug("%s %s: addr2line read line: %s", __func__, dso_name, line); + pr_debug3("%s %s: addr2line read line: %s", __func__, dso_name, line); if (function != NULL) *function = strdup(strim(line)); @@ -652,7 +652,7 @@ static int read_addr2line_record(struct io *io, if (io__getline(io, &line, &line_len) < 0 || !line_len) goto error; - pr_debug("%s %s: addr2line filename:number : %s", __func__, dso_name, line); + pr_debug3("%s %s: addr2line filename:number : %s", __func__, dso_name, line); if (filename_split(line, line_nr == NULL ? &dummy_line_nr : line_nr) == 0 && style == GNU_BINUTILS) { ret = 0; diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c index 729ad5cd52cb..a67b991f4e81 100644 --- a/tools/perf/util/stat-display.c +++ b/tools/perf/util/stat-display.c @@ -50,15 +50,15 @@ static int aggr_header_lens[] = { }; static const char *aggr_header_csv[] = { - [AGGR_CORE] = "core,cpus,", - [AGGR_CACHE] = "cache,cpus,", - [AGGR_CLUSTER] = "cluster,cpus,", - [AGGR_DIE] = "die,cpus,", - [AGGR_SOCKET] = "socket,cpus,", - [AGGR_NONE] = "cpu,", - [AGGR_THREAD] = "comm-pid,", - [AGGR_NODE] = "node,", - [AGGR_GLOBAL] = "" + [AGGR_CORE] = "core,ctrs,", + [AGGR_CACHE] = "cache,ctrs,", + [AGGR_CLUSTER] = "cluster,ctrs,", + [AGGR_DIE] = "die,ctrs,", + [AGGR_SOCKET] = "socket,ctrs,", + [AGGR_NONE] = "cpu,", + [AGGR_THREAD] = "comm-pid,", + [AGGR_NODE] = "node,", + [AGGR_GLOBAL] = "" }; static const char *aggr_header_std[] = { @@ -304,7 +304,7 @@ static void print_aggr_id_std(struct perf_stat_config *config, return; } - fprintf(output, "%-*s %*d ", aggr_header_lens[idx], buf, 4, aggr_nr); + fprintf(output, "%-*s %*d ", aggr_header_lens[idx], buf, /*strlen("ctrs")*/ 4, aggr_nr); } static void print_aggr_id_csv(struct perf_stat_config *config, @@ -366,27 +366,27 @@ static void print_aggr_id_json(struct perf_stat_config *config, struct outstate { switch (config->aggr_mode) { case AGGR_CORE: - json_out(os, "\"core\" : \"S%d-D%d-C%d\", \"aggregate-number\" : %d", + json_out(os, "\"core\" : \"S%d-D%d-C%d\", \"counters\" : %d", id.socket, id.die, id.core, aggr_nr); break; case AGGR_CACHE: - json_out(os, "\"cache\" : \"S%d-D%d-L%d-ID%d\", \"aggregate-number\" : %d", + json_out(os, "\"cache\" : \"S%d-D%d-L%d-ID%d\", \"counters\" : %d", id.socket, id.die, id.cache_lvl, id.cache, aggr_nr); break; case AGGR_CLUSTER: - json_out(os, "\"cluster\" : \"S%d-D%d-CLS%d\", \"aggregate-number\" : %d", + json_out(os, "\"cluster\" : \"S%d-D%d-CLS%d\", \"counters\" : %d", id.socket, id.die, id.cluster, aggr_nr); break; case AGGR_DIE: - json_out(os, "\"die\" : \"S%d-D%d\", \"aggregate-number\" : %d", + json_out(os, "\"die\" : \"S%d-D%d\", \"counters\" : %d", id.socket, id.die, aggr_nr); break; case AGGR_SOCKET: - json_out(os, "\"socket\" : \"S%d\", \"aggregate-number\" : %d", + json_out(os, "\"socket\" : \"S%d\", \"counters\" : %d", id.socket, aggr_nr); break; case AGGR_NODE: - json_out(os, "\"node\" : \"N%d\", \"aggregate-number\" : %d", + json_out(os, "\"node\" : \"N%d\", \"counters\" : %d", id.node, aggr_nr); break; case AGGR_NONE: @@ -899,12 +899,11 @@ static void printout(struct perf_stat_config *config, struct outstate *os, print_noise(config, os, counter, noise, /*before_metric=*/true); print_running(config, os, run, ena, /*before_metric=*/true); from = perf_stat__print_shadow_stats_metricgroup(config, counter, aggr_idx, - &num, from, &out, - &config->metric_events); + &num, from, &out); } while (from != NULL); - } else - perf_stat__print_shadow_stats(config, counter, uval, aggr_idx, - &out, &config->metric_events); + } else { + perf_stat__print_shadow_stats(config, counter, uval, aggr_idx, &out); + } } else { pm(config, os, METRIC_THRESHOLD_UNKNOWN, /*format=*/NULL, /*unit=*/NULL, /*val=*/0); } @@ -1016,7 +1015,7 @@ static void print_counter_aggrdata(struct perf_stat_config *config, ena = aggr->counts.ena; run = aggr->counts.run; - if (perf_stat__skip_metric_event(counter, &config->metric_events, ena, run)) + if (perf_stat__skip_metric_event(counter, ena, run)) return; if (val == 0 && should_skip_zero_counter(config, counter, &id)) @@ -1275,10 +1274,7 @@ static void print_metric_headers(struct perf_stat_config *config, os.evsel = counter; - perf_stat__print_shadow_stats(config, counter, 0, - 0, - &out, - &config->metric_events); + perf_stat__print_shadow_stats(config, counter, 0, 0, &out); } if (!config->json_output) @@ -1317,7 +1313,7 @@ static void print_header_interval_std(struct perf_stat_config *config, case AGGR_CLUSTER: case AGGR_CACHE: case AGGR_CORE: - fprintf(output, "#%*s %-*s cpus", + fprintf(output, "#%*s %-*s ctrs", INTERVAL_LEN - 1, "time", aggr_header_lens[config->aggr_mode], aggr_header_std[config->aggr_mode]); diff --git a/tools/perf/util/stat-shadow.c b/tools/perf/util/stat-shadow.c index d83bda5824d2..abaf6b579bfc 100644 --- a/tools/perf/util/stat-shadow.c +++ b/tools/perf/util/stat-shadow.c @@ -635,14 +635,14 @@ void *perf_stat__print_shadow_stats_metricgroup(struct perf_stat_config *config, int aggr_idx, int *num, void *from, - struct perf_stat_output_ctx *out, - struct rblist *metric_events) + struct perf_stat_output_ctx *out) { struct metric_event *me; struct metric_expr *mexp = from; void *ctxp = out->ctx; bool header_printed = false; const char *name = NULL; + struct rblist *metric_events = &evsel->evlist->metric_events; me = metricgroup__lookup(metric_events, evsel, false); if (me == NULL) @@ -683,8 +683,7 @@ void *perf_stat__print_shadow_stats_metricgroup(struct perf_stat_config *config, void perf_stat__print_shadow_stats(struct perf_stat_config *config, struct evsel *evsel, double avg, int aggr_idx, - struct perf_stat_output_ctx *out, - struct rblist *metric_events) + struct perf_stat_output_ctx *out) { typedef void (*stat_print_function_t)(struct perf_stat_config *config, const struct evsel *evsel, @@ -735,7 +734,7 @@ void perf_stat__print_shadow_stats(struct perf_stat_config *config, } perf_stat__print_shadow_stats_metricgroup(config, evsel, aggr_idx, - &num, NULL, out, metric_events); + &num, NULL, out); if (num == 0) { print_metric(config, ctxp, METRIC_THRESHOLD_UNKNOWN, @@ -748,7 +747,6 @@ void perf_stat__print_shadow_stats(struct perf_stat_config *config, * if it's not running or not the metric event. */ bool perf_stat__skip_metric_event(struct evsel *evsel, - struct rblist *metric_events, u64 ena, u64 run) { if (!evsel->default_metricgroup) @@ -757,5 +755,5 @@ bool perf_stat__skip_metric_event(struct evsel *evsel, if (!ena || !run) return true; - return !metricgroup__lookup(metric_events, evsel, false); + return !metricgroup__lookup(&evsel->evlist->metric_events, evsel, false); } diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index 355a7d5c8ab8..50b1a92d16df 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -526,7 +526,7 @@ static int evsel__merge_aggr_counters(struct evsel *evsel, struct evsel *alias) struct perf_counts_values *aggr_counts_a = &ps_a->aggr[i].counts; struct perf_counts_values *aggr_counts_b = &ps_b->aggr[i].counts; - /* NB: don't increase aggr.nr for aliases */ + ps_a->aggr[i].nr += ps_b->aggr[i].nr; aggr_counts_a->val += aggr_counts_b->val; aggr_counts_a->ena += aggr_counts_b->ena; @@ -769,8 +769,6 @@ int create_perf_stat_counter(struct evsel *evsel, attr->enable_on_exec = 1; } - if (target__has_cpu(target) && !target__has_per_thread(target)) - return evsel__open_per_cpu(evsel, evsel__cpus(evsel), cpu_map_idx); - - return evsel__open_per_thread(evsel, evsel->core.threads); + return evsel__open_per_cpu_and_thread(evsel, evsel__cpus(evsel), cpu_map_idx, + evsel->core.threads); } diff --git a/tools/perf/util/stat.h b/tools/perf/util/stat.h index 1bcd7634bf47..4b0f14ae4e5f 100644 --- a/tools/perf/util/stat.h +++ b/tools/perf/util/stat.h @@ -7,7 +7,6 @@ #include <sys/types.h> #include <sys/resource.h> #include "cpumap.h" -#include "rblist.h" #include "counts.h" struct perf_cpu_map; @@ -108,7 +107,6 @@ struct perf_stat_config { aggr_get_id_t aggr_get_id; struct cpu_aggr_map *cpus_aggr_map; u64 *walltime_run; - struct rblist metric_events; int ctl_fd; int ctl_fd_ack; bool ctl_fd_close; @@ -187,18 +185,14 @@ struct perf_stat_output_ctx { void perf_stat__print_shadow_stats(struct perf_stat_config *config, struct evsel *evsel, double avg, int aggr_idx, - struct perf_stat_output_ctx *out, - struct rblist *metric_events); -bool perf_stat__skip_metric_event(struct evsel *evsel, - struct rblist *metric_events, - u64 ena, u64 run); + struct perf_stat_output_ctx *out); +bool perf_stat__skip_metric_event(struct evsel *evsel, u64 ena, u64 run); void *perf_stat__print_shadow_stats_metricgroup(struct perf_stat_config *config, struct evsel *evsel, int aggr_idx, int *num, void *from, - struct perf_stat_output_ctx *out, - struct rblist *metric_events); + struct perf_stat_output_ctx *out); int evlist__alloc_stats(struct perf_stat_config *config, struct evlist *evlist, bool alloc_raw); diff --git a/tools/perf/util/symbol-minimal.c b/tools/perf/util/symbol-minimal.c index c73fe2e09fe9..7201494c5c20 100644 --- a/tools/perf/util/symbol-minimal.c +++ b/tools/perf/util/symbol-minimal.c @@ -317,7 +317,7 @@ int dso__load_sym(struct dso *dso, struct map *map __maybe_unused, struct symsrc *runtime_ss __maybe_unused, int kmodule __maybe_unused) { - struct build_id bid; + struct build_id bid = { .size = 0, }; int ret; ret = fd__is_64_bit(ss->fd); diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index 8b30c6f16a9e..e816e4220d33 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -631,7 +631,7 @@ void dso__sort_by_name(struct dso *dso) { mutex_lock(dso__lock(dso)); if (!dso__sorted_by_name(dso)) { - size_t len; + size_t len = 0; dso__set_symbol_names(dso, symbols__sort_by_name(dso__symbols(dso), &len)); if (dso__symbol_names(dso)) { @@ -1422,6 +1422,7 @@ static int dso__load_kcore(struct dso *dso, struct map *map, goto out_err; } } + map__zput(new_node->map); free(new_node); } @@ -1812,7 +1813,6 @@ int dso__load(struct dso *dso, struct map *map) struct symsrc *syms_ss = NULL, *runtime_ss = NULL; bool kmod; bool perfmap; - struct build_id bid; struct nscookie nsc; char newmapname[PATH_MAX]; const char *map_path = dso__long_name(dso); @@ -1873,6 +1873,8 @@ int dso__load(struct dso *dso, struct map *map) */ if (!dso__has_build_id(dso) && is_regular_file(dso__long_name(dso))) { + struct build_id bid = { .size = 0, }; + __symbol__join_symfs(name, PATH_MAX, dso__long_name(dso)); if (filename__read_build_id(name, &bid) > 0) dso__set_build_id(dso, &bid); @@ -2121,7 +2123,7 @@ static bool filename__readable(const char *file) static char *dso__find_kallsyms(struct dso *dso, struct map *map) { - struct build_id bid; + struct build_id bid = { .size = 0, }; char sbuild_id[SBUILD_ID_SIZE]; bool is_host = false; char path[PATH_MAX]; @@ -2151,7 +2153,7 @@ static char *dso__find_kallsyms(struct dso *dso, struct map *map) goto proc_kallsyms; } - build_id__sprintf(dso__bid(dso), sbuild_id); + build_id__snprintf(dso__bid(dso), sbuild_id, sizeof(sbuild_id)); /* Find kallsyms in build-id cache with kcore */ scnprintf(path, sizeof(path), "%s/%s/%s", diff --git a/tools/perf/util/symbol_conf.h b/tools/perf/util/symbol_conf.h index cd9aa82c7d5a..7a80d2c14d9b 100644 --- a/tools/perf/util/symbol_conf.h +++ b/tools/perf/util/symbol_conf.h @@ -43,7 +43,7 @@ struct symbol_conf { report_individual_block, inline_name, disable_add2line_warn, - buildid_mmap2, + no_buildid_mmap2, guest_code, lazy_load_kernel_maps, keep_exited_threads, diff --git a/tools/perf/util/synthetic-events.c b/tools/perf/util/synthetic-events.c index 2fc4d0537840..cb2c1ace304a 100644 --- a/tools/perf/util/synthetic-events.c +++ b/tools/perf/util/synthetic-events.c @@ -368,11 +368,11 @@ static void perf_record_mmap2__read_build_id(struct perf_record_mmap2 *event, struct machine *machine, bool is_kernel) { - struct build_id bid; + struct build_id bid = { .size = 0, }; struct nsinfo *nsi; struct nscookie nc; struct dso *dso = NULL; - struct dso_id id; + struct dso_id dso_id = dso_id_empty; int rc; if (is_kernel) { @@ -380,12 +380,18 @@ static void perf_record_mmap2__read_build_id(struct perf_record_mmap2 *event, goto out; } - id.maj = event->maj; - id.min = event->min; - id.ino = event->ino; - id.ino_generation = event->ino_generation; + if (event->header.misc & PERF_RECORD_MISC_MMAP_BUILD_ID) { + build_id__init(&dso_id.build_id, event->build_id, event->build_id_size); + } else { + dso_id.maj = event->maj; + dso_id.min = event->min; + dso_id.ino = event->ino; + dso_id.ino_generation = event->ino_generation; + dso_id.mmap2_valid = true; + dso_id.mmap2_ino_generation_valid = true; + }; - dso = dsos__findnew_id(&machine->dsos, event->filename, &id); + dso = dsos__findnew_id(&machine->dsos, event->filename, &dso_id); if (dso && dso__has_build_id(dso)) { bid = *dso__bid(dso); rc = 0; @@ -526,7 +532,7 @@ out: event->mmap2.pid = tgid; event->mmap2.tid = pid; - if (symbol_conf.buildid_mmap2) + if (!symbol_conf.no_buildid_mmap2) perf_record_mmap2__read_build_id(&event->mmap2, machine, false); if (perf_tool__process_synth_event(tool, event, machine, process) != 0) { @@ -684,7 +690,7 @@ static int perf_event__synthesize_modules_maps_cb(struct map *map, void *data) return 0; dso = map__dso(map); - if (symbol_conf.buildid_mmap2) { + if (!symbol_conf.no_buildid_mmap2) { size = PERF_ALIGN(dso__long_name_len(dso) + 1, sizeof(u64)); event->mmap2.header.type = PERF_RECORD_MMAP2; event->mmap2.header.size = (sizeof(event->mmap2) - @@ -728,9 +734,9 @@ int perf_event__synthesize_modules(const struct perf_tool *tool, perf_event__han .process = process, .machine = machine, }; - size_t size = symbol_conf.buildid_mmap2 - ? sizeof(args.event->mmap2) - : sizeof(args.event->mmap); + size_t size = symbol_conf.no_buildid_mmap2 + ? sizeof(args.event->mmap) + : sizeof(args.event->mmap2); args.event = zalloc(size + machine->id_hdr_size); if (args.event == NULL) { @@ -1118,8 +1124,8 @@ static int __perf_event__synthesize_kernel_mmap(const struct perf_tool *tool, struct machine *machine) { union perf_event *event; - size_t size = symbol_conf.buildid_mmap2 ? - sizeof(event->mmap2) : sizeof(event->mmap); + size_t size = symbol_conf.no_buildid_mmap2 ? + sizeof(event->mmap) : sizeof(event->mmap2); struct map *map = machine__kernel_map(machine); struct kmap *kmap; int err; @@ -1153,7 +1159,7 @@ static int __perf_event__synthesize_kernel_mmap(const struct perf_tool *tool, event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL; } - if (symbol_conf.buildid_mmap2) { + if (!symbol_conf.no_buildid_mmap2) { size = snprintf(event->mmap2.filename, sizeof(event->mmap2.filename), "%s%s", machine->mmap_name, kmap->ref_reloc_sym->name) + 1; size = PERF_ALIGN(size, sizeof(u64)); @@ -1567,10 +1573,16 @@ size_t perf_event__sample_event_size(const struct perf_sample *sample, u64 type, return result; } -void __weak arch_perf_synthesize_sample_weight(const struct perf_sample *data, +static void perf_synthesize_sample_weight(const struct perf_sample *data, __u64 *array, u64 type __maybe_unused) { *array = data->weight; + + if (type & PERF_SAMPLE_WEIGHT_STRUCT) { + *array &= 0xffffffff; + *array |= ((u64)data->ins_lat << 32); + *array |= ((u64)data->weight3 << 48); + } } static __u64 *copy_read_group_values(__u64 *array, __u64 read_format, @@ -1730,7 +1742,7 @@ int perf_event__synthesize_sample(union perf_event *event, u64 type, u64 read_fo } if (type & PERF_SAMPLE_WEIGHT_TYPE) { - arch_perf_synthesize_sample_weight(sample, array, type); + perf_synthesize_sample_weight(sample, array, type); array++; } @@ -2045,7 +2057,7 @@ int perf_event__synthesize_event_update_name(const struct perf_tool *tool, struc int perf_event__synthesize_event_update_cpus(const struct perf_tool *tool, struct evsel *evsel, perf_event__handler_t process) { - struct synthesize_cpu_map_data syn_data = { .map = evsel->core.own_cpus }; + struct synthesize_cpu_map_data syn_data = { .map = evsel->core.pmu_cpus }; struct perf_record_event_update *ev; int err; @@ -2126,7 +2138,7 @@ int perf_event__synthesize_extra_attr(const struct perf_tool *tool, struct evlis } } - if (evsel->core.own_cpus) { + if (evsel->core.pmu_cpus) { err = perf_event__synthesize_event_update_cpus(tool, evsel, process); if (err < 0) { pr_err("Couldn't synthesize evsel cpus.\n"); @@ -2248,7 +2260,9 @@ int perf_event__synthesize_build_id(const struct perf_tool *tool, memset(&ev, 0, len); - ev.build_id.size = min(bid->size, sizeof(ev.build_id.build_id)); + ev.build_id.size = bid->size; + if (ev.build_id.size > sizeof(ev.build_id.build_id)) + ev.build_id.size = sizeof(ev.build_id.build_id); memcpy(ev.build_id.build_id, bid->data, ev.build_id.size); ev.build_id.header.type = PERF_RECORD_HEADER_BUILD_ID; ev.build_id.header.misc = misc | PERF_RECORD_MISC_BUILD_ID_SIZE; @@ -2308,7 +2322,9 @@ int perf_event__synthesize_mmap2_build_id(const struct perf_tool *tool, ev.mmap2.len = len; ev.mmap2.pgoff = pgoff; - ev.mmap2.build_id_size = min(bid->size, sizeof(ev.mmap2.build_id)); + ev.mmap2.build_id_size = bid->size; + if (ev.mmap2.build_id_size > sizeof(ev.mmap2.build_id)) + ev.build_id.size = sizeof(ev.mmap2.build_id); memcpy(ev.mmap2.build_id, bid->data, ev.mmap2.build_id_size); ev.mmap2.prot = prot; diff --git a/tools/perf/util/synthetic-events.h b/tools/perf/util/synthetic-events.h index b9c936b5cfeb..ee29615d68e5 100644 --- a/tools/perf/util/synthetic-events.h +++ b/tools/perf/util/synthetic-events.h @@ -92,6 +92,8 @@ int perf_event__synthesize_threads(const struct perf_tool *tool, perf_event__han int perf_event__synthesize_tracing_data(const struct perf_tool *tool, int fd, struct evlist *evlist, perf_event__handler_t process); int perf_event__synth_time_conv(const struct perf_event_mmap_page *pc, const struct perf_tool *tool, perf_event__handler_t process, struct machine *machine); pid_t perf_event__synthesize_comm(const struct perf_tool *tool, union perf_event *event, pid_t pid, perf_event__handler_t process, struct machine *machine); +void perf_event__synthesize_final_bpf_metadata(struct perf_session *session, + perf_event__handler_t process); int perf_tool__process_synth_event(const struct perf_tool *tool, union perf_event *event, struct machine *machine, perf_event__handler_t process); diff --git a/tools/perf/util/target.c b/tools/perf/util/target.c index 0f383418e3df..8cf71bea295a 100644 --- a/tools/perf/util/target.c +++ b/tools/perf/util/target.c @@ -28,20 +28,6 @@ enum target_errno target__validate(struct target *target) ret = TARGET_ERRNO__PID_OVERRIDE_CPU; } - /* UID and PID are mutually exclusive */ - if (target->tid && target->uid_str) { - target->uid_str = NULL; - if (ret == TARGET_ERRNO__SUCCESS) - ret = TARGET_ERRNO__PID_OVERRIDE_UID; - } - - /* UID and CPU are mutually exclusive */ - if (target->uid_str && target->cpu_list) { - target->cpu_list = NULL; - if (ret == TARGET_ERRNO__SUCCESS) - ret = TARGET_ERRNO__UID_OVERRIDE_CPU; - } - /* PID and SYSTEM are mutually exclusive */ if (target->tid && target->system_wide) { target->system_wide = false; @@ -49,13 +35,6 @@ enum target_errno target__validate(struct target *target) ret = TARGET_ERRNO__PID_OVERRIDE_SYSTEM; } - /* UID and SYSTEM are mutually exclusive */ - if (target->uid_str && target->system_wide) { - target->system_wide = false; - if (ret == TARGET_ERRNO__SUCCESS) - ret = TARGET_ERRNO__UID_OVERRIDE_SYSTEM; - } - /* BPF and CPU are mutually exclusive */ if (target->bpf_str && target->cpu_list) { target->cpu_list = NULL; @@ -70,13 +49,6 @@ enum target_errno target__validate(struct target *target) ret = TARGET_ERRNO__BPF_OVERRIDE_PID; } - /* BPF and UID are mutually exclusive */ - if (target->bpf_str && target->uid_str) { - target->uid_str = NULL; - if (ret == TARGET_ERRNO__SUCCESS) - ret = TARGET_ERRNO__BPF_OVERRIDE_UID; - } - /* BPF and THREADS are mutually exclusive */ if (target->bpf_str && target->per_thread) { target->per_thread = false; @@ -94,15 +66,13 @@ enum target_errno target__validate(struct target *target) return ret; } -enum target_errno target__parse_uid(struct target *target) +uid_t parse_uid(const char *str) { struct passwd pwd, *result; char buf[1024]; - const char *str = target->uid_str; - target->uid = UINT_MAX; if (str == NULL) - return TARGET_ERRNO__SUCCESS; + return UINT_MAX; /* Try user name first */ getpwnam_r(str, &pwd, buf, sizeof(buf), &result); @@ -115,16 +85,15 @@ enum target_errno target__parse_uid(struct target *target) int uid = strtol(str, &endptr, 10); if (*endptr != '\0') - return TARGET_ERRNO__INVALID_UID; + return UINT_MAX; getpwuid_r(uid, &pwd, buf, sizeof(buf), &result); if (result == NULL) - return TARGET_ERRNO__USER_NOT_FOUND; + return UINT_MAX; } - target->uid = result->pw_uid; - return TARGET_ERRNO__SUCCESS; + return result->pw_uid; } /* @@ -132,20 +101,14 @@ enum target_errno target__parse_uid(struct target *target) */ static const char *target__error_str[] = { "PID/TID switch overriding CPU", - "PID/TID switch overriding UID", - "UID switch overriding CPU", "PID/TID switch overriding SYSTEM", - "UID switch overriding SYSTEM", "SYSTEM/CPU switch overriding PER-THREAD", "BPF switch overriding CPU", "BPF switch overriding PID/TID", - "BPF switch overriding UID", "BPF switch overriding THREAD", - "Invalid User: %s", - "Problems obtaining information for user %s", }; -int target__strerror(struct target *target, int errnum, +int target__strerror(struct target *target __maybe_unused, int errnum, char *buf, size_t buflen) { int idx; @@ -170,11 +133,6 @@ int target__strerror(struct target *target, int errnum, snprintf(buf, buflen, "%s", msg); break; - case TARGET_ERRNO__INVALID_UID: - case TARGET_ERRNO__USER_NOT_FOUND: - snprintf(buf, buflen, msg, target->uid_str); - break; - default: /* cannot reach here */ break; diff --git a/tools/perf/util/target.h b/tools/perf/util/target.h index 2ee2cc30340f..84ebb9c940c6 100644 --- a/tools/perf/util/target.h +++ b/tools/perf/util/target.h @@ -9,9 +9,7 @@ struct target { const char *pid; const char *tid; const char *cpu_list; - const char *uid_str; const char *bpf_str; - uid_t uid; bool system_wide; bool uses_mmap; bool default_per_cpu; @@ -36,31 +34,24 @@ enum target_errno { /* for target__validate() */ TARGET_ERRNO__PID_OVERRIDE_CPU = __TARGET_ERRNO__START, - TARGET_ERRNO__PID_OVERRIDE_UID, - TARGET_ERRNO__UID_OVERRIDE_CPU, TARGET_ERRNO__PID_OVERRIDE_SYSTEM, - TARGET_ERRNO__UID_OVERRIDE_SYSTEM, TARGET_ERRNO__SYSTEM_OVERRIDE_THREAD, TARGET_ERRNO__BPF_OVERRIDE_CPU, TARGET_ERRNO__BPF_OVERRIDE_PID, - TARGET_ERRNO__BPF_OVERRIDE_UID, TARGET_ERRNO__BPF_OVERRIDE_THREAD, - /* for target__parse_uid() */ - TARGET_ERRNO__INVALID_UID, - TARGET_ERRNO__USER_NOT_FOUND, - __TARGET_ERRNO__END, }; enum target_errno target__validate(struct target *target); -enum target_errno target__parse_uid(struct target *target); + +uid_t parse_uid(const char *str); int target__strerror(struct target *target, int errnum, char *buf, size_t buflen); static inline bool target__has_task(struct target *target) { - return target->tid || target->pid || target->uid_str; + return target->tid || target->pid; } static inline bool target__has_cpu(struct target *target) diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c index ffb48cc2103f..aa9c58bbf9d3 100644 --- a/tools/perf/util/thread.c +++ b/tools/perf/util/thread.c @@ -41,6 +41,7 @@ int thread__init_maps(struct thread *thread, struct machine *machine) } struct thread *thread__new(pid_t pid, pid_t tid) + NO_THREAD_SAFETY_ANALYSIS /* Allocation/creation is inherently single threaded. */ { RC_STRUCT(thread) *_thread = zalloc(sizeof(*_thread)); struct thread *thread; @@ -200,7 +201,8 @@ int thread__set_namespaces(struct thread *thread, u64 timestamp, return ret; } -struct comm *thread__comm(struct thread *thread) +static struct comm *__thread__comm(struct thread *thread) + SHARED_LOCKS_REQUIRED(thread__comm_lock(thread)) { if (list_empty(thread__comm_list(thread))) return NULL; @@ -208,16 +210,30 @@ struct comm *thread__comm(struct thread *thread) return list_first_entry(thread__comm_list(thread), struct comm, list); } +struct comm *thread__comm(struct thread *thread) +{ + struct comm *res = NULL; + + down_read(thread__comm_lock(thread)); + res = __thread__comm(thread); + up_read(thread__comm_lock(thread)); + return res; +} + struct comm *thread__exec_comm(struct thread *thread) { struct comm *comm, *last = NULL, *second_last = NULL; + down_read(thread__comm_lock(thread)); list_for_each_entry(comm, thread__comm_list(thread), list) { - if (comm->exec) + if (comm->exec) { + up_read(thread__comm_lock(thread)); return comm; + } second_last = last; last = comm; } + up_read(thread__comm_lock(thread)); /* * 'last' with no start time might be the parent's comm of a synthesized @@ -233,8 +249,9 @@ struct comm *thread__exec_comm(struct thread *thread) static int ____thread__set_comm(struct thread *thread, const char *str, u64 timestamp, bool exec) + EXCLUSIVE_LOCKS_REQUIRED(thread__comm_lock(thread)) { - struct comm *new, *curr = thread__comm(thread); + struct comm *new, *curr = __thread__comm(thread); /* Override the default :tid entry */ if (!thread__comm_set(thread)) { @@ -285,8 +302,9 @@ int thread__set_comm_from_proc(struct thread *thread) } static const char *__thread__comm_str(struct thread *thread) + SHARED_LOCKS_REQUIRED(thread__comm_lock(thread)) { - const struct comm *comm = thread__comm(thread); + const struct comm *comm = __thread__comm(thread); if (!comm) return NULL; diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h index 2b90bbed7a61..310eaea344bb 100644 --- a/tools/perf/util/thread.h +++ b/tools/perf/util/thread.h @@ -236,14 +236,15 @@ static inline struct rw_semaphore *thread__namespaces_lock(struct thread *thread return &RC_CHK_ACCESS(thread)->namespaces_lock; } -static inline struct list_head *thread__comm_list(struct thread *thread) +static inline struct rw_semaphore *thread__comm_lock(struct thread *thread) { - return &RC_CHK_ACCESS(thread)->comm_list; + return &RC_CHK_ACCESS(thread)->comm_lock; } -static inline struct rw_semaphore *thread__comm_lock(struct thread *thread) +static inline struct list_head *thread__comm_list(struct thread *thread) + SHARED_LOCKS_REQUIRED(thread__comm_lock(thread)) { - return &RC_CHK_ACCESS(thread)->comm_lock; + return &RC_CHK_ACCESS(thread)->comm_list; } static inline u64 thread__db_id(const struct thread *thread) diff --git a/tools/perf/util/thread_map.c b/tools/perf/util/thread_map.c index b5f12390c355..ca193c1374ed 100644 --- a/tools/perf/util/thread_map.c +++ b/tools/perf/util/thread_map.c @@ -72,7 +72,7 @@ struct perf_thread_map *thread_map__new_by_tid(pid_t tid) return threads; } -static struct perf_thread_map *__thread_map__new_all_cpus(uid_t uid) +static struct perf_thread_map *thread_map__new_all_cpus(void) { DIR *proc; int max_threads = 32, items, i; @@ -98,15 +98,6 @@ static struct perf_thread_map *__thread_map__new_all_cpus(uid_t uid) if (*end) /* only interested in proper numerical dirents */ continue; - snprintf(path, sizeof(path), "/proc/%s", dirent->d_name); - - if (uid != UINT_MAX) { - struct stat st; - - if (stat(path, &st) != 0 || st.st_uid != uid) - continue; - } - snprintf(path, sizeof(path), "/proc/%d/task", pid); items = scandir(path, &namelist, filter, NULL); if (items <= 0) { @@ -157,24 +148,11 @@ out_free_namelist: goto out_closedir; } -struct perf_thread_map *thread_map__new_all_cpus(void) -{ - return __thread_map__new_all_cpus(UINT_MAX); -} - -struct perf_thread_map *thread_map__new_by_uid(uid_t uid) -{ - return __thread_map__new_all_cpus(uid); -} - -struct perf_thread_map *thread_map__new(pid_t pid, pid_t tid, uid_t uid) +struct perf_thread_map *thread_map__new(pid_t pid, pid_t tid) { if (pid != -1) return thread_map__new_by_pid(pid); - if (tid == -1 && uid != UINT_MAX) - return thread_map__new_by_uid(uid); - return thread_map__new_by_tid(tid); } @@ -289,15 +267,11 @@ out_free_threads: goto out; } -struct perf_thread_map *thread_map__new_str(const char *pid, const char *tid, - uid_t uid, bool all_threads) +struct perf_thread_map *thread_map__new_str(const char *pid, const char *tid, bool all_threads) { if (pid) return thread_map__new_by_pid_str(pid); - if (!tid && uid != UINT_MAX) - return thread_map__new_by_uid(uid); - if (all_threads) return thread_map__new_all_cpus(); diff --git a/tools/perf/util/thread_map.h b/tools/perf/util/thread_map.h index 00ec05fc1656..fc16d87f32fb 100644 --- a/tools/perf/util/thread_map.h +++ b/tools/perf/util/thread_map.h @@ -11,13 +11,11 @@ struct perf_record_thread_map; struct perf_thread_map *thread_map__new_dummy(void); struct perf_thread_map *thread_map__new_by_pid(pid_t pid); struct perf_thread_map *thread_map__new_by_tid(pid_t tid); -struct perf_thread_map *thread_map__new_by_uid(uid_t uid); -struct perf_thread_map *thread_map__new_all_cpus(void); -struct perf_thread_map *thread_map__new(pid_t pid, pid_t tid, uid_t uid); +struct perf_thread_map *thread_map__new(pid_t pid, pid_t tid); struct perf_thread_map *thread_map__new_event(struct perf_record_thread_map *event); struct perf_thread_map *thread_map__new_str(const char *pid, - const char *tid, uid_t uid, bool all_threads); + const char *tid, bool all_threads); struct perf_thread_map *thread_map__new_by_tid_str(const char *tid_str); diff --git a/tools/perf/util/tool.c b/tools/perf/util/tool.c index 37bd8ac63b01..e83c7ababc2a 100644 --- a/tools/perf/util/tool.c +++ b/tools/perf/util/tool.c @@ -1,12 +1,15 @@ // SPDX-License-Identifier: GPL-2.0 #include "data.h" #include "debug.h" +#include "event.h" #include "header.h" #include "session.h" #include "stat.h" #include "tool.h" #include "tsc.h" +#include <linux/compiler.h> #include <sys/mman.h> +#include <stddef.h> #include <unistd.h> #ifdef HAVE_ZSTD_SUPPORT @@ -17,7 +20,7 @@ static int perf_session__process_compressed_event(struct perf_session *session, void *src; size_t decomp_size, src_size; u64 decomp_last_rem = 0; - size_t mmap_len, decomp_len = session->header.env.comp_mmap_len; + size_t mmap_len, decomp_len = perf_session__env(session)->comp_mmap_len; struct decomp *decomp, *decomp_last = session->active_decomp->decomp_last; if (decomp_last) { @@ -237,6 +240,16 @@ static int perf_session__process_compressed_event_stub(struct perf_session *sess return 0; } +static int perf_event__process_bpf_metadata_stub(struct perf_session *perf_session __maybe_unused, + union perf_event *event) +{ + if (dump_trace) + perf_event__fprintf_bpf_metadata(event, stdout); + + dump_printf(": unhandled!\n"); + return 0; +} + void perf_tool__init(struct perf_tool *tool, bool ordered_events) { tool->ordered_events = ordered_events; @@ -293,6 +306,7 @@ void perf_tool__init(struct perf_tool *tool, bool ordered_events) tool->compressed = perf_session__process_compressed_event_stub; #endif tool->finished_init = process_event_op2_stub; + tool->bpf_metadata = perf_event__process_bpf_metadata_stub; } bool perf_tool__compressed_is_stub(const struct perf_tool *tool) diff --git a/tools/perf/util/tool.h b/tools/perf/util/tool.h index db1c7642b0d1..18b76ff0f26a 100644 --- a/tools/perf/util/tool.h +++ b/tools/perf/util/tool.h @@ -77,7 +77,8 @@ struct perf_tool { stat, stat_round, feature, - finished_init; + finished_init, + bpf_metadata; event_op4 compressed; event_op3 auxtrace; bool ordered_events; diff --git a/tools/perf/util/tool_pmu.c b/tools/perf/util/tool_pmu.c index 4630b8cc8e52..d99e699e646d 100644 --- a/tools/perf/util/tool_pmu.c +++ b/tools/perf/util/tool_pmu.c @@ -332,7 +332,7 @@ static bool has_pmem(void) return has_pmem; } -bool tool_pmu__read_event(enum tool_pmu_event ev, u64 *result) +bool tool_pmu__read_event(enum tool_pmu_event ev, struct evsel *evsel, u64 *result) { const struct cpu_topology *topology; @@ -347,18 +347,60 @@ bool tool_pmu__read_event(enum tool_pmu_event ev, u64 *result) return true; case TOOL_PMU__EVENT_NUM_CPUS: - *result = cpu__max_present_cpu().cpu; + if (!evsel || perf_cpu_map__is_empty(evsel->core.cpus)) { + /* No evsel to be specific to. */ + *result = cpu__max_present_cpu().cpu; + } else if (!perf_cpu_map__has_any_cpu(evsel->core.cpus)) { + /* Evsel just has specific CPUs. */ + *result = perf_cpu_map__nr(evsel->core.cpus); + } else { + /* + * "Any CPU" event that can be scheduled on any CPU in + * the PMU's cpumask. The PMU cpumask should be saved in + * pmu_cpus. If not present fall back to max. + */ + if (!perf_cpu_map__is_empty(evsel->core.pmu_cpus)) + *result = perf_cpu_map__nr(evsel->core.pmu_cpus); + else + *result = cpu__max_present_cpu().cpu; + } return true; case TOOL_PMU__EVENT_NUM_CPUS_ONLINE: { struct perf_cpu_map *online = cpu_map__online(); - if (online) { + if (!online) + return false; + + if (!evsel || perf_cpu_map__is_empty(evsel->core.cpus)) { + /* No evsel to be specific to. */ *result = perf_cpu_map__nr(online); - perf_cpu_map__put(online); - return true; + } else if (!perf_cpu_map__has_any_cpu(evsel->core.cpus)) { + /* Evsel just has specific CPUs. */ + struct perf_cpu_map *tmp = + perf_cpu_map__intersect(online, evsel->core.cpus); + + *result = perf_cpu_map__nr(tmp); + perf_cpu_map__put(tmp); + } else { + /* + * "Any CPU" event that can be scheduled on any CPU in + * the PMU's cpumask. The PMU cpumask should be saved in + * pmu_cpus, if not present then just the online cpu + * mask. + */ + if (!perf_cpu_map__is_empty(evsel->core.pmu_cpus)) { + struct perf_cpu_map *tmp = + perf_cpu_map__intersect(online, evsel->core.pmu_cpus); + + *result = perf_cpu_map__nr(tmp); + perf_cpu_map__put(tmp); + } else { + *result = perf_cpu_map__nr(online); + } } - return false; + perf_cpu_map__put(online); + return true; } case TOOL_PMU__EVENT_NUM_DIES: topology = online_topology(); @@ -417,7 +459,7 @@ int evsel__tool_pmu_read(struct evsel *evsel, int cpu_map_idx, int thread) old_count = perf_counts(evsel->prev_raw_counts, cpu_map_idx, thread); val = 0; if (cpu_map_idx == 0 && thread == 0) { - if (!tool_pmu__read_event(ev, &val)) { + if (!tool_pmu__read_event(ev, evsel, &val)) { count->lost++; val = 0; } diff --git a/tools/perf/util/tool_pmu.h b/tools/perf/util/tool_pmu.h index c6ad1dd90a56..d642e7d73910 100644 --- a/tools/perf/util/tool_pmu.h +++ b/tools/perf/util/tool_pmu.h @@ -34,7 +34,7 @@ enum tool_pmu_event tool_pmu__str_to_event(const char *str); bool tool_pmu__skip_event(const char *name); int tool_pmu__num_skip_events(void); -bool tool_pmu__read_event(enum tool_pmu_event ev, u64 *result); +bool tool_pmu__read_event(enum tool_pmu_event ev, struct evsel *evsel, u64 *result); u64 tool_pmu__cpu_slots_per_cycle(void); diff --git a/tools/perf/util/top.c b/tools/perf/util/top.c index 4db3d1bd686c..b06e10a116bb 100644 --- a/tools/perf/util/top.c +++ b/tools/perf/util/top.c @@ -88,9 +88,9 @@ size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size) else if (target->tid) ret += SNPRINTF(bf + ret, size - ret, " (target_tid: %s", target->tid); - else if (target->uid_str != NULL) + else if (top->uid_str != NULL) ret += SNPRINTF(bf + ret, size - ret, " (uid: %s", - target->uid_str); + top->uid_str); else ret += SNPRINTF(bf + ret, size - ret, " (all"); diff --git a/tools/perf/util/top.h b/tools/perf/util/top.h index 4c5588dbb131..04ff926846be 100644 --- a/tools/perf/util/top.h +++ b/tools/perf/util/top.h @@ -48,6 +48,7 @@ struct perf_top { const char *sym_filter; float min_percent; unsigned int nr_threads_synthesize; + const char *uid_str; struct { struct ordered_events *in; diff --git a/tools/perf/util/tp_pmu.c b/tools/perf/util/tp_pmu.c new file mode 100644 index 000000000000..e7534a973247 --- /dev/null +++ b/tools/perf/util/tp_pmu.c @@ -0,0 +1,210 @@ +// SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) +#include "tp_pmu.h" +#include "pmus.h" +#include <api/fs/fs.h> +#include <api/fs/tracing_path.h> +#include <api/io_dir.h> +#include <linux/kernel.h> +#include <errno.h> +#include <string.h> + +int tp_pmu__id(const char *sys, const char *name) +{ + char *tp_dir = get_events_file(sys); + char path[PATH_MAX]; + int id, err; + + if (!tp_dir) + return -1; + + scnprintf(path, PATH_MAX, "%s/%s/id", tp_dir, name); + put_events_file(tp_dir); + err = filename__read_int(path, &id); + if (err) + return err; + + return id; +} + + +int tp_pmu__for_each_tp_event(const char *sys, void *state, tp_event_callback cb) +{ + char *evt_path; + struct io_dirent64 *evt_ent; + struct io_dir evt_dir; + int ret = 0; + + evt_path = get_events_file(sys); + if (!evt_path) + return -errno; + + io_dir__init(&evt_dir, open(evt_path, O_CLOEXEC | O_DIRECTORY | O_RDONLY)); + if (evt_dir.dirfd < 0) { + ret = -errno; + put_events_file(evt_path); + return ret; + } + put_events_file(evt_path); + + while (!ret && (evt_ent = io_dir__readdir(&evt_dir))) { + if (!strcmp(evt_ent->d_name, ".") + || !strcmp(evt_ent->d_name, "..") + || !strcmp(evt_ent->d_name, "enable") + || !strcmp(evt_ent->d_name, "filter")) + continue; + + ret = cb(state, sys, evt_ent->d_name); + if (ret) + break; + } + close(evt_dir.dirfd); + return ret; +} + +int tp_pmu__for_each_tp_sys(void *state, tp_sys_callback cb) +{ + struct io_dirent64 *events_ent; + struct io_dir events_dir; + int ret = 0; + char *events_dir_path = get_tracing_file("events"); + + if (!events_dir_path) + return -errno; + + io_dir__init(&events_dir, open(events_dir_path, O_CLOEXEC | O_DIRECTORY | O_RDONLY)); + if (events_dir.dirfd < 0) { + ret = -errno; + put_events_file(events_dir_path); + return ret; + } + put_events_file(events_dir_path); + + while (!ret && (events_ent = io_dir__readdir(&events_dir))) { + if (!strcmp(events_ent->d_name, ".") || + !strcmp(events_ent->d_name, "..") || + !strcmp(events_ent->d_name, "enable") || + !strcmp(events_ent->d_name, "header_event") || + !strcmp(events_ent->d_name, "header_page")) + continue; + + ret = cb(state, events_ent->d_name); + if (ret) + break; + } + close(events_dir.dirfd); + return ret; +} + +bool perf_pmu__is_tracepoint(const struct perf_pmu *pmu) +{ + return pmu->type == PERF_TYPE_TRACEPOINT; +} + +struct for_each_event_args { + void *state; + pmu_event_callback cb; + const struct perf_pmu *pmu; +}; + +static int for_each_event_cb(void *state, const char *sys_name, const char *evt_name) +{ + struct for_each_event_args *args = state; + char name[2 * FILENAME_MAX + 2]; + /* 16 possible hex digits and 22 other characters and \0. */ + char encoding[16 + 22]; + char *format = NULL; + size_t format_size; + struct pmu_event_info info = { + .pmu = args->pmu, + .pmu_name = args->pmu->name, + .event_type_desc = "Tracepoint event", + }; + char *tp_dir = get_events_file(sys_name); + char path[PATH_MAX]; + int id, err; + + if (!tp_dir) + return -1; + + scnprintf(path, sizeof(path), "%s/%s/id", tp_dir, evt_name); + err = filename__read_int(path, &id); + if (err == 0) { + snprintf(encoding, sizeof(encoding), "tracepoint/config=0x%x/", id); + info.encoding_desc = encoding; + } + + scnprintf(path, sizeof(path), "%s/%s/format", tp_dir, evt_name); + put_events_file(tp_dir); + err = filename__read_str(path, &format, &format_size); + if (err == 0) { + info.long_desc = format; + for (size_t i = 0 ; i < format_size; i++) { + /* Swap tabs to spaces due to some rendering issues. */ + if (format[i] == '\t') + format[i] = ' '; + } + } + snprintf(name, sizeof(name), "%s:%s", sys_name, evt_name); + info.name = name; + err = args->cb(args->state, &info); + free(format); + return err; +} + +static int for_each_event_sys_cb(void *state, const char *sys_name) +{ + return tp_pmu__for_each_tp_event(sys_name, state, for_each_event_cb); +} + +int tp_pmu__for_each_event(struct perf_pmu *pmu, void *state, pmu_event_callback cb) +{ + struct for_each_event_args args = { + .state = state, + .cb = cb, + .pmu = pmu, + }; + + return tp_pmu__for_each_tp_sys(&args, for_each_event_sys_cb); +} + +static int num_events_cb(void *state, const char *sys_name __maybe_unused, + const char *evt_name __maybe_unused) +{ + size_t *count = state; + + (*count)++; + return 0; +} + +static int num_events_sys_cb(void *state, const char *sys_name) +{ + return tp_pmu__for_each_tp_event(sys_name, state, num_events_cb); +} + +size_t tp_pmu__num_events(struct perf_pmu *pmu __maybe_unused) +{ + size_t count = 0; + + tp_pmu__for_each_tp_sys(&count, num_events_sys_cb); + return count; +} + +bool tp_pmu__have_event(struct perf_pmu *pmu __maybe_unused, const char *name) +{ + char *dup_name, *colon; + int id; + + colon = strchr(name, ':'); + if (colon == NULL) + return false; + + dup_name = strdup(name); + if (!dup_name) + return false; + + colon = dup_name + (colon - name); + *colon = '\0'; + id = tp_pmu__id(dup_name, colon + 1); + free(dup_name); + return id >= 0; +} diff --git a/tools/perf/util/tp_pmu.h b/tools/perf/util/tp_pmu.h new file mode 100644 index 000000000000..30456bd6943d --- /dev/null +++ b/tools/perf/util/tp_pmu.h @@ -0,0 +1,19 @@ +/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */ +#ifndef __TP_PMU_H +#define __TP_PMU_H + +#include "pmu.h" + +typedef int (*tp_sys_callback)(void *state, const char *sys_name); +typedef int (*tp_event_callback)(void *state, const char *sys_name, const char *evt_name); + +int tp_pmu__id(const char *sys, const char *name); +int tp_pmu__for_each_tp_event(const char *sys, void *state, tp_event_callback cb); +int tp_pmu__for_each_tp_sys(void *state, tp_sys_callback cb); + +bool perf_pmu__is_tracepoint(const struct perf_pmu *pmu); +int tp_pmu__for_each_event(struct perf_pmu *pmu, void *state, pmu_event_callback cb); +size_t tp_pmu__num_events(struct perf_pmu *pmu); +bool tp_pmu__have_event(struct perf_pmu *pmu, const char *name); + +#endif /* __TP_PMU_H */ diff --git a/tools/perf/util/trace_augment.h b/tools/perf/util/trace_augment.h index 57a3e5045937..4f729bc67753 100644 --- a/tools/perf/util/trace_augment.h +++ b/tools/perf/util/trace_augment.h @@ -1,6 +1,66 @@ #ifndef TRACE_AUGMENT_H #define TRACE_AUGMENT_H -#define TRACE_AUG_MAX_BUF 32 /* for buffer augmentation in perf trace */ +#include <linux/compiler.h> + +struct bpf_program; +struct evlist; + +#ifdef HAVE_BPF_SKEL + +int augmented_syscalls__prepare(void); +int augmented_syscalls__create_bpf_output(struct evlist *evlist); +void augmented_syscalls__setup_bpf_output(void); +int augmented_syscalls__set_filter_pids(unsigned int nr, pid_t *pids); +int augmented_syscalls__get_map_fds(int *enter_fd, int *exit_fd, int *beauty_fd); +struct bpf_program *augmented_syscalls__find_by_title(const char *name); +struct bpf_program *augmented_syscalls__unaugmented(void); +void augmented_syscalls__cleanup(void); + +#else /* !HAVE_BPF_SKEL */ + +static inline int augmented_syscalls__prepare(void) +{ + return -1; +} + +static inline int augmented_syscalls__create_bpf_output(struct evlist *evlist __maybe_unused) +{ + return -1; +} + +static inline void augmented_syscalls__setup_bpf_output(void) +{ +} + +static inline int augmented_syscalls__set_filter_pids(unsigned int nr __maybe_unused, + pid_t *pids __maybe_unused) +{ + return 0; +} + +static inline int augmented_syscalls__get_map_fds(int *enter_fd __maybe_unused, + int *exit_fd __maybe_unused, + int *beauty_fd __maybe_unused) +{ + return -1; +} + +static inline struct bpf_program * +augmented_syscalls__find_by_title(const char *name __maybe_unused) +{ + return NULL; +} + +static inline struct bpf_program *augmented_syscalls__unaugmented(void) +{ + return NULL; +} + +static inline void augmented_syscalls__cleanup(void) +{ +} + +#endif /* HAVE_BPF_SKEL */ #endif diff --git a/tools/perf/util/unwind-libdw.c b/tools/perf/util/unwind-libdw.c index 793d11832694..ae70fb56a057 100644 --- a/tools/perf/util/unwind-libdw.c +++ b/tools/perf/util/unwind-libdw.c @@ -84,8 +84,11 @@ static int __report_module(struct addr_location *al, u64 ip, char filename[PATH_MAX]; __symbol__join_symfs(filename, sizeof(filename), dso__long_name(dso)); - mod = dwfl_report_elf(ui->dwfl, dso__short_name(dso), filename, -1, - base, false); + /* Don't hang up on device files like /dev/dri/renderD128. */ + if (is_regular_file(filename)) { + mod = dwfl_report_elf(ui->dwfl, dso__short_name(dso), filename, -1, + base, false); + } } if (!mod) { char filename[PATH_MAX]; |