diff options
Diffstat (limited to '')
-rw-r--r-- | tools/perf/util/synthetic-events.c | 822 |
1 files changed, 591 insertions, 231 deletions
diff --git a/tools/perf/util/synthetic-events.c b/tools/perf/util/synthetic-events.c index 8a23391558cf..2a0289c14959 100644 --- a/tools/perf/util/synthetic-events.c +++ b/tools/perf/util/synthetic-events.c @@ -1,5 +1,7 @@ // SPDX-License-Identifier: GPL-2.0-only +#include "util/cgroup.h" +#include "util/data.h" #include "util/debug.h" #include "util/dso.h" #include "util/event.h" @@ -16,7 +18,6 @@ #include "util/synthetic-events.h" #include "util/target.h" #include "util/time-utils.h" -#include "util/cgroup.h" #include <linux/bitops.h> #include <linux/kernel.h> #include <linux/string.h> @@ -24,7 +25,6 @@ #include <linux/perf_event.h> #include <asm/bug.h> #include <perf/evsel.h> -#include <internal/cpumap.h> #include <perf/cpumap.h> #include <internal/lib.h> // page_size #include <internal/threadmap.h> @@ -69,19 +69,22 @@ int perf_tool__process_synth_event(struct perf_tool *tool, * Assumes that the first 4095 bytes of /proc/pid/stat contains * the comm, tgid and ppid. */ -static int perf_event__get_comm_ids(pid_t pid, char *comm, size_t len, - pid_t *tgid, pid_t *ppid) +static int perf_event__get_comm_ids(pid_t pid, pid_t tid, char *comm, size_t len, + pid_t *tgid, pid_t *ppid, bool *kernel) { char bf[4096]; int fd; size_t size = 0; ssize_t n; - char *name, *tgids, *ppids; + char *name, *tgids, *ppids, *vmpeak, *threads; *tgid = -1; *ppid = -1; - snprintf(bf, sizeof(bf), "/proc/%d/status", pid); + if (pid) + snprintf(bf, sizeof(bf), "/proc/%d/task/%d/status", pid, tid); + else + snprintf(bf, sizeof(bf), "/proc/%d/status", tid); fd = open(bf, O_RDONLY); if (fd < 0) { @@ -93,14 +96,20 @@ static int perf_event__get_comm_ids(pid_t pid, char *comm, size_t len, close(fd); if (n <= 0) { pr_warning("Couldn't get COMM, tigd and ppid for pid %d\n", - pid); + tid); return -1; } bf[n] = '\0'; name = strstr(bf, "Name:"); - tgids = strstr(bf, "Tgid:"); - ppids = strstr(bf, "PPid:"); + tgids = strstr(name ?: bf, "Tgid:"); + ppids = strstr(tgids ?: bf, "PPid:"); + vmpeak = strstr(ppids ?: bf, "VmPeak:"); + + if (vmpeak) + threads = NULL; + else + threads = strstr(ppids ?: bf, "Threads:"); if (name) { char *nl; @@ -116,29 +125,34 @@ static int perf_event__get_comm_ids(pid_t pid, char *comm, size_t len, memcpy(comm, name, size); comm[size] = '\0'; } else { - pr_debug("Name: string not found for pid %d\n", pid); + pr_debug("Name: string not found for pid %d\n", tid); } if (tgids) { tgids += 5; /* strlen("Tgid:") */ *tgid = atoi(tgids); } else { - pr_debug("Tgid: string not found for pid %d\n", pid); + pr_debug("Tgid: string not found for pid %d\n", tid); } if (ppids) { ppids += 5; /* strlen("PPid:") */ *ppid = atoi(ppids); } else { - pr_debug("PPid: string not found for pid %d\n", pid); + pr_debug("PPid: string not found for pid %d\n", tid); } + if (!vmpeak && threads) + *kernel = true; + else + *kernel = false; + return 0; } -static int perf_event__prepare_comm(union perf_event *event, pid_t pid, +static int perf_event__prepare_comm(union perf_event *event, pid_t pid, pid_t tid, struct machine *machine, - pid_t *tgid, pid_t *ppid) + pid_t *tgid, pid_t *ppid, bool *kernel) { size_t size; @@ -147,9 +161,9 @@ static int perf_event__prepare_comm(union perf_event *event, pid_t pid, memset(&event->comm, 0, sizeof(event->comm)); if (machine__is_host(machine)) { - if (perf_event__get_comm_ids(pid, event->comm.comm, + if (perf_event__get_comm_ids(pid, tid, event->comm.comm, sizeof(event->comm.comm), - tgid, ppid) != 0) { + tgid, ppid, kernel) != 0) { return -1; } } else { @@ -168,7 +182,7 @@ static int perf_event__prepare_comm(union perf_event *event, pid_t pid, event->comm.header.size = (sizeof(event->comm) - (sizeof(event->comm.comm) - size) + machine->id_hdr_size); - event->comm.tid = pid; + event->comm.tid = tid; return 0; } @@ -179,8 +193,10 @@ pid_t perf_event__synthesize_comm(struct perf_tool *tool, struct machine *machine) { pid_t tgid, ppid; + bool kernel_thread; - if (perf_event__prepare_comm(event, pid, machine, &tgid, &ppid) != 0) + if (perf_event__prepare_comm(event, 0, pid, machine, &tgid, &ppid, + &kernel_thread) != 0) return -1; if (perf_tool__process_synth_event(tool, event, machine, process) != 0) @@ -347,6 +363,61 @@ static bool read_proc_maps_line(struct io *io, __u64 *start, __u64 *end, } } +static void perf_record_mmap2__read_build_id(struct perf_record_mmap2 *event, + struct machine *machine, + bool is_kernel) +{ + struct build_id bid; + struct nsinfo *nsi; + struct nscookie nc; + struct dso *dso = NULL; + struct dso_id id; + int rc; + + if (is_kernel) { + rc = sysfs__read_build_id("/sys/kernel/notes", &bid); + goto out; + } + + id.maj = event->maj; + id.min = event->min; + id.ino = event->ino; + id.ino_generation = event->ino_generation; + + dso = dsos__findnew_id(&machine->dsos, event->filename, &id); + if (dso && dso->has_build_id) { + bid = dso->bid; + rc = 0; + goto out; + } + + nsi = nsinfo__new(event->pid); + nsinfo__mountns_enter(nsi, &nc); + + rc = filename__read_build_id(event->filename, &bid) > 0 ? 0 : -1; + + nsinfo__mountns_exit(&nc); + nsinfo__put(nsi); + +out: + if (rc == 0) { + memcpy(event->build_id, bid.data, sizeof(bid.data)); + event->build_id_size = (u8) bid.size; + event->header.misc |= PERF_RECORD_MISC_MMAP_BUILD_ID; + event->__reserved_1 = 0; + event->__reserved_2 = 0; + + if (dso && !dso->has_build_id) + dso__set_build_id(dso, &bid); + } else { + if (event->filename[0] == '/') { + pr_debug2("Failed to read build ID for %s\n", + event->filename); + } + } + dso__put(dso); +} + int perf_event__synthesize_mmap_events(struct perf_tool *tool, union perf_event *event, pid_t pid, pid_t tgid, @@ -384,7 +455,7 @@ int perf_event__synthesize_mmap_events(struct perf_tool *tool, while (!io.eof) { static const char anonstr[] = "//anon"; - size_t size; + size_t size, aligned_size; /* ensure null termination since stack will be reused. */ event->mmap2.filename[0] = '\0'; @@ -444,15 +515,19 @@ out: } size = strlen(event->mmap2.filename) + 1; - size = PERF_ALIGN(size, sizeof(u64)); + aligned_size = PERF_ALIGN(size, sizeof(u64)); event->mmap2.len -= event->mmap.start; event->mmap2.header.size = (sizeof(event->mmap2) - - (sizeof(event->mmap2.filename) - size)); - memset(event->mmap2.filename + size, 0, machine->id_hdr_size); + (sizeof(event->mmap2.filename) - aligned_size)); + memset(event->mmap2.filename + size, 0, machine->id_hdr_size + + (aligned_size - size)); event->mmap2.header.size += machine->id_hdr_size; event->mmap2.pid = tgid; event->mmap2.tid = pid; + if (symbol_conf.buildid_mmap2) + perf_record_mmap2__read_build_id(&event->mmap2, machine, false); + if (perf_tool__process_synth_event(tool, event, machine, process) != 0) { rc = -1; break; @@ -563,6 +638,9 @@ int perf_event__synthesize_cgroups(struct perf_tool *tool, char cgrp_root[PATH_MAX]; size_t mount_len; /* length of mount point in the path */ + if (!tool || !tool->cgroup_events) + return 0; + if (cgroupfs_find_mountpoint(cgrp_root, PATH_MAX, "perf_event") < 0) { pr_debug("cannot find cgroup mount point\n"); return -1; @@ -587,71 +665,112 @@ int perf_event__synthesize_cgroups(struct perf_tool *tool __maybe_unused, } #endif +struct perf_event__synthesize_modules_maps_cb_args { + struct perf_tool *tool; + perf_event__handler_t process; + struct machine *machine; + union perf_event *event; +}; + +static int perf_event__synthesize_modules_maps_cb(struct map *map, void *data) +{ + struct perf_event__synthesize_modules_maps_cb_args *args = data; + union perf_event *event = args->event; + struct dso *dso; + size_t size; + + if (!__map__is_kmodule(map)) + return 0; + + dso = map__dso(map); + if (symbol_conf.buildid_mmap2) { + size = PERF_ALIGN(dso->long_name_len + 1, sizeof(u64)); + event->mmap2.header.type = PERF_RECORD_MMAP2; + event->mmap2.header.size = (sizeof(event->mmap2) - + (sizeof(event->mmap2.filename) - size)); + memset(event->mmap2.filename + size, 0, args->machine->id_hdr_size); + event->mmap2.header.size += args->machine->id_hdr_size; + event->mmap2.start = map__start(map); + event->mmap2.len = map__size(map); + event->mmap2.pid = args->machine->pid; + + memcpy(event->mmap2.filename, dso->long_name, dso->long_name_len + 1); + + perf_record_mmap2__read_build_id(&event->mmap2, args->machine, false); + } else { + size = PERF_ALIGN(dso->long_name_len + 1, sizeof(u64)); + event->mmap.header.type = PERF_RECORD_MMAP; + event->mmap.header.size = (sizeof(event->mmap) - + (sizeof(event->mmap.filename) - size)); + memset(event->mmap.filename + size, 0, args->machine->id_hdr_size); + event->mmap.header.size += args->machine->id_hdr_size; + event->mmap.start = map__start(map); + event->mmap.len = map__size(map); + event->mmap.pid = args->machine->pid; + + memcpy(event->mmap.filename, dso->long_name, dso->long_name_len + 1); + } + + if (perf_tool__process_synth_event(args->tool, event, args->machine, args->process) != 0) + return -1; + + return 0; +} + int perf_event__synthesize_modules(struct perf_tool *tool, perf_event__handler_t process, struct machine *machine) { - int rc = 0; - struct map *pos; + int rc; struct maps *maps = machine__kernel_maps(machine); - union perf_event *event = zalloc((sizeof(event->mmap) + - machine->id_hdr_size)); - if (event == NULL) { + struct perf_event__synthesize_modules_maps_cb_args args = { + .tool = tool, + .process = process, + .machine = machine, + }; + size_t size = symbol_conf.buildid_mmap2 + ? sizeof(args.event->mmap2) + : sizeof(args.event->mmap); + + args.event = zalloc(size + machine->id_hdr_size); + if (args.event == NULL) { pr_debug("Not enough memory synthesizing mmap event " "for kernel modules\n"); return -1; } - event->header.type = PERF_RECORD_MMAP; - /* * kernel uses 0 for user space maps, see kernel/perf_event.c * __perf_event_mmap */ if (machine__is_host(machine)) - event->header.misc = PERF_RECORD_MISC_KERNEL; + args.event->header.misc = PERF_RECORD_MISC_KERNEL; else - event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL; - - maps__for_each_entry(maps, pos) { - size_t size; - - if (!__map__is_kmodule(pos)) - continue; + args.event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL; - size = PERF_ALIGN(pos->dso->long_name_len + 1, sizeof(u64)); - event->mmap.header.type = PERF_RECORD_MMAP; - event->mmap.header.size = (sizeof(event->mmap) - - (sizeof(event->mmap.filename) - size)); - memset(event->mmap.filename + size, 0, machine->id_hdr_size); - event->mmap.header.size += machine->id_hdr_size; - event->mmap.start = pos->start; - event->mmap.len = pos->end - pos->start; - event->mmap.pid = machine->pid; + rc = maps__for_each_map(maps, perf_event__synthesize_modules_maps_cb, &args); - memcpy(event->mmap.filename, pos->dso->long_name, - pos->dso->long_name_len + 1); - if (perf_tool__process_synth_event(tool, event, machine, process) != 0) { - rc = -1; - break; - } - } - - free(event); + free(args.event); return rc; } +static int filter_task(const struct dirent *dirent) +{ + return isdigit(dirent->d_name[0]); +} + static int __event__synthesize_thread(union perf_event *comm_event, union perf_event *mmap_event, union perf_event *fork_event, union perf_event *namespaces_event, pid_t pid, int full, perf_event__handler_t process, - struct perf_tool *tool, struct machine *machine, bool mmap_data) + struct perf_tool *tool, struct machine *machine, + bool needs_mmap, bool mmap_data) { char filename[PATH_MAX]; - DIR *tasks; - struct dirent *dirent; + struct dirent **dirent; pid_t tgid, ppid; int rc = 0; + int i, n; /* special case: only send one comm event using passed in pid */ if (!full) { @@ -669,7 +788,7 @@ static int __event__synthesize_thread(union perf_event *comm_event, * send mmap only for thread group leader * see thread__init_maps() */ - if (pid == tgid && + if (pid == tgid && needs_mmap && perf_event__synthesize_mmap_events(tool, mmap_event, pid, tgid, process, machine, mmap_data)) return -1; @@ -683,25 +802,25 @@ static int __event__synthesize_thread(union perf_event *comm_event, snprintf(filename, sizeof(filename), "%s/proc/%d/task", machine->root_dir, pid); - tasks = opendir(filename); - if (tasks == NULL) { - pr_debug("couldn't open %s\n", filename); - return 0; - } + n = scandir(filename, &dirent, filter_task, NULL); + if (n < 0) + return n; - while ((dirent = readdir(tasks)) != NULL) { + for (i = 0; i < n; i++) { char *end; pid_t _pid; + bool kernel_thread = false; - _pid = strtol(dirent->d_name, &end, 10); + _pid = strtol(dirent[i]->d_name, &end, 10); if (*end) continue; - rc = -1; - if (perf_event__prepare_comm(comm_event, _pid, machine, - &tgid, &ppid) != 0) - break; + /* some threads may exit just after scan, ignore it */ + if (perf_event__prepare_comm(comm_event, pid, _pid, machine, + &tgid, &ppid, &kernel_thread) != 0) + continue; + rc = -1; if (perf_event__synthesize_fork(tool, fork_event, _pid, tgid, ppid, process, machine) < 0) break; @@ -717,7 +836,7 @@ static int __event__synthesize_thread(union perf_event *comm_event, break; rc = 0; - if (_pid == pid) { + if (_pid == pid && !kernel_thread && needs_mmap) { /* process the parent's maps too */ rc = perf_event__synthesize_mmap_events(tool, mmap_event, pid, tgid, process, machine, mmap_data); @@ -726,7 +845,10 @@ static int __event__synthesize_thread(union perf_event *comm_event, } } - closedir(tasks); + for (i = 0; i < n; i++) + zfree(&dirent[i]); + free(dirent); + return rc; } @@ -734,7 +856,7 @@ int perf_event__synthesize_thread_map(struct perf_tool *tool, struct perf_thread_map *threads, perf_event__handler_t process, struct machine *machine, - bool mmap_data) + bool needs_mmap, bool mmap_data) { union perf_event *comm_event, *mmap_event, *fork_event; union perf_event *namespaces_event; @@ -764,7 +886,7 @@ int perf_event__synthesize_thread_map(struct perf_tool *tool, fork_event, namespaces_event, perf_thread_map__pid(threads, thread), 0, process, tool, machine, - mmap_data)) { + needs_mmap, mmap_data)) { err = -1; break; } @@ -790,7 +912,7 @@ int perf_event__synthesize_thread_map(struct perf_tool *tool, fork_event, namespaces_event, comm_event->comm.pid, 0, process, tool, machine, - mmap_data)) { + needs_mmap, mmap_data)) { err = -1; break; } @@ -810,6 +932,7 @@ out: static int __perf_event__synthesize_threads(struct perf_tool *tool, perf_event__handler_t process, struct machine *machine, + bool needs_mmap, bool mmap_data, struct dirent **dirent, int start, @@ -854,7 +977,7 @@ static int __perf_event__synthesize_threads(struct perf_tool *tool, */ __event__synthesize_thread(comm_event, mmap_event, fork_event, namespaces_event, pid, 1, process, - tool, machine, mmap_data); + tool, machine, needs_mmap, mmap_data); } err = 0; @@ -873,6 +996,7 @@ struct synthesize_threads_arg { struct perf_tool *tool; perf_event__handler_t process; struct machine *machine; + bool needs_mmap; bool mmap_data; struct dirent **dirent; int num; @@ -884,7 +1008,8 @@ static void *synthesize_threads_worker(void *arg) struct synthesize_threads_arg *args = arg; __perf_event__synthesize_threads(args->tool, args->process, - args->machine, args->mmap_data, + args->machine, + args->needs_mmap, args->mmap_data, args->dirent, args->start, args->num); return NULL; @@ -893,7 +1018,7 @@ static void *synthesize_threads_worker(void *arg) int perf_event__synthesize_threads(struct perf_tool *tool, perf_event__handler_t process, struct machine *machine, - bool mmap_data, + bool needs_mmap, bool mmap_data, unsigned int nr_threads_synthesize) { struct synthesize_threads_arg *args = NULL; @@ -911,7 +1036,7 @@ int perf_event__synthesize_threads(struct perf_tool *tool, return 0; snprintf(proc_path, sizeof(proc_path), "%s/proc", machine->root_dir); - n = scandir(proc_path, &dirent, 0, alphasort); + n = scandir(proc_path, &dirent, filter_task, NULL); if (n < 0) return err; @@ -922,18 +1047,19 @@ int perf_event__synthesize_threads(struct perf_tool *tool, if (thread_nr <= 1) { err = __perf_event__synthesize_threads(tool, process, - machine, mmap_data, + machine, + needs_mmap, mmap_data, dirent, base, n); goto free_dirent; } if (thread_nr > n) thread_nr = n; - synthesize_threads = calloc(sizeof(pthread_t), thread_nr); + synthesize_threads = calloc(thread_nr, sizeof(pthread_t)); if (synthesize_threads == NULL) goto free_dirent; - args = calloc(sizeof(*args), thread_nr); + args = calloc(thread_nr, sizeof(*args)); if (args == NULL) goto free_threads; @@ -943,6 +1069,7 @@ int perf_event__synthesize_threads(struct perf_tool *tool, args[i].tool = tool; args[i].process = process; args[i].machine = machine; + args[i].needs_mmap = needs_mmap; args[i].mmap_data = mmap_data; args[i].dirent = dirent; } @@ -988,11 +1115,12 @@ static int __perf_event__synthesize_kernel_mmap(struct perf_tool *tool, perf_event__handler_t process, struct machine *machine) { - size_t size; + union perf_event *event; + size_t size = symbol_conf.buildid_mmap2 ? + sizeof(event->mmap2) : sizeof(event->mmap); struct map *map = machine__kernel_map(machine); struct kmap *kmap; int err; - union perf_event *event; if (map == NULL) return -1; @@ -1006,7 +1134,7 @@ static int __perf_event__synthesize_kernel_mmap(struct perf_tool *tool, * available use this, and after it is use this as a fallback for older * kernels. */ - event = zalloc((sizeof(event->mmap) + machine->id_hdr_size)); + event = zalloc(size + machine->id_hdr_size); if (event == NULL) { pr_debug("Not enough memory synthesizing mmap event " "for kernel modules\n"); @@ -1023,16 +1151,31 @@ static int __perf_event__synthesize_kernel_mmap(struct perf_tool *tool, event->header.misc = PERF_RECORD_MISC_GUEST_KERNEL; } - size = snprintf(event->mmap.filename, sizeof(event->mmap.filename), - "%s%s", machine->mmap_name, kmap->ref_reloc_sym->name) + 1; - size = PERF_ALIGN(size, sizeof(u64)); - event->mmap.header.type = PERF_RECORD_MMAP; - event->mmap.header.size = (sizeof(event->mmap) - - (sizeof(event->mmap.filename) - size) + machine->id_hdr_size); - event->mmap.pgoff = kmap->ref_reloc_sym->addr; - event->mmap.start = map->start; - event->mmap.len = map->end - event->mmap.start; - event->mmap.pid = machine->pid; + if (symbol_conf.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)); + event->mmap2.header.type = PERF_RECORD_MMAP2; + event->mmap2.header.size = (sizeof(event->mmap2) - + (sizeof(event->mmap2.filename) - size) + machine->id_hdr_size); + event->mmap2.pgoff = kmap->ref_reloc_sym->addr; + event->mmap2.start = map__start(map); + event->mmap2.len = map__end(map) - event->mmap.start; + event->mmap2.pid = machine->pid; + + perf_record_mmap2__read_build_id(&event->mmap2, machine, true); + } else { + size = snprintf(event->mmap.filename, sizeof(event->mmap.filename), + "%s%s", machine->mmap_name, kmap->ref_reloc_sym->name) + 1; + size = PERF_ALIGN(size, sizeof(u64)); + event->mmap.header.type = PERF_RECORD_MMAP; + event->mmap.header.size = (sizeof(event->mmap) - + (sizeof(event->mmap.filename) - size) + machine->id_hdr_size); + event->mmap.pgoff = kmap->ref_reloc_sym->addr; + event->mmap.start = map__start(map); + event->mmap.len = map__end(map) - event->mmap.start; + event->mmap.pid = machine->pid; + } err = perf_tool__process_synth_event(tool, event, machine, process); free(event); @@ -1089,122 +1232,122 @@ int perf_event__synthesize_thread_map2(struct perf_tool *tool, return err; } -static void synthesize_cpus(struct cpu_map_entries *cpus, - struct perf_cpu_map *map) -{ - int i; - - cpus->nr = map->nr; - - for (i = 0; i < map->nr; i++) - cpus->cpu[i] = map->map[i]; -} - -static void synthesize_mask(struct perf_record_record_cpu_map *mask, - struct perf_cpu_map *map, int max) -{ - int i; - - mask->nr = BITS_TO_LONGS(max); - mask->long_size = sizeof(long); - - for (i = 0; i < map->nr; i++) - set_bit(map->map[i], mask->mask); -} +struct synthesize_cpu_map_data { + const struct perf_cpu_map *map; + int nr; + int min_cpu; + int max_cpu; + int has_any_cpu; + int type; + size_t size; + struct perf_record_cpu_map_data *data; +}; -static size_t cpus_size(struct perf_cpu_map *map) +static void synthesize_cpus(struct synthesize_cpu_map_data *data) { - return sizeof(struct cpu_map_entries) + map->nr * sizeof(u16); + data->data->type = PERF_CPU_MAP__CPUS; + data->data->cpus_data.nr = data->nr; + for (int i = 0; i < data->nr; i++) + data->data->cpus_data.cpu[i] = perf_cpu_map__cpu(data->map, i).cpu; } -static size_t mask_size(struct perf_cpu_map *map, int *max) +static void synthesize_mask(struct synthesize_cpu_map_data *data) { - int i; + int idx; + struct perf_cpu cpu; - *max = 0; + /* Due to padding, the 4bytes per entry mask variant is always smaller. */ + data->data->type = PERF_CPU_MAP__MASK; + data->data->mask32_data.nr = BITS_TO_U32(data->max_cpu); + data->data->mask32_data.long_size = 4; - for (i = 0; i < map->nr; i++) { - /* bit possition of the cpu is + 1 */ - int bit = map->map[i] + 1; + perf_cpu_map__for_each_cpu(cpu, idx, data->map) { + int bit_word = cpu.cpu / 32; + u32 bit_mask = 1U << (cpu.cpu & 31); - if (bit > *max) - *max = bit; + data->data->mask32_data.mask[bit_word] |= bit_mask; } +} - return sizeof(struct perf_record_record_cpu_map) + BITS_TO_LONGS(*max) * sizeof(long); +static void synthesize_range_cpus(struct synthesize_cpu_map_data *data) +{ + data->data->type = PERF_CPU_MAP__RANGE_CPUS; + data->data->range_cpu_data.any_cpu = data->has_any_cpu; + data->data->range_cpu_data.start_cpu = data->min_cpu; + data->data->range_cpu_data.end_cpu = data->max_cpu; } -void *cpu_map_data__alloc(struct perf_cpu_map *map, size_t *size, u16 *type, int *max) +static void *cpu_map_data__alloc(struct synthesize_cpu_map_data *syn_data, + size_t header_size) { size_t size_cpus, size_mask; - bool is_dummy = perf_cpu_map__empty(map); - /* - * Both array and mask data have variable size based - * on the number of cpus and their actual values. - * The size of the 'struct perf_record_cpu_map_data' is: - * - * array = size of 'struct cpu_map_entries' + - * number of cpus * sizeof(u64) - * - * mask = size of 'struct perf_record_record_cpu_map' + - * maximum cpu bit converted to size of longs - * - * and finaly + the size of 'struct perf_record_cpu_map_data'. - */ - size_cpus = cpus_size(map); - size_mask = mask_size(map, max); - - if (is_dummy || (size_cpus < size_mask)) { - *size += size_cpus; - *type = PERF_CPU_MAP__CPUS; - } else { - *size += size_mask; - *type = PERF_CPU_MAP__MASK; - } - - *size += sizeof(struct perf_record_cpu_map_data); - *size = PERF_ALIGN(*size, sizeof(u64)); - return zalloc(*size); + syn_data->nr = perf_cpu_map__nr(syn_data->map); + syn_data->has_any_cpu = (perf_cpu_map__cpu(syn_data->map, 0).cpu == -1) ? 1 : 0; + + syn_data->min_cpu = perf_cpu_map__cpu(syn_data->map, syn_data->has_any_cpu).cpu; + syn_data->max_cpu = perf_cpu_map__max(syn_data->map).cpu; + if (syn_data->max_cpu - syn_data->min_cpu + 1 == syn_data->nr - syn_data->has_any_cpu) { + /* A consecutive range of CPUs can be encoded using a range. */ + assert(sizeof(u16) + sizeof(struct perf_record_range_cpu_map) == sizeof(u64)); + syn_data->type = PERF_CPU_MAP__RANGE_CPUS; + syn_data->size = header_size + sizeof(u64); + return zalloc(syn_data->size); + } + + size_cpus = sizeof(u16) + sizeof(struct cpu_map_entries) + syn_data->nr * sizeof(u16); + /* Due to padding, the 4bytes per entry mask variant is always smaller. */ + size_mask = sizeof(u16) + sizeof(struct perf_record_mask_cpu_map32) + + BITS_TO_U32(syn_data->max_cpu) * sizeof(__u32); + if (syn_data->has_any_cpu || size_cpus < size_mask) { + /* Follow the CPU map encoding. */ + syn_data->type = PERF_CPU_MAP__CPUS; + syn_data->size = header_size + PERF_ALIGN(size_cpus, sizeof(u64)); + return zalloc(syn_data->size); + } + /* Encode using a bitmask. */ + syn_data->type = PERF_CPU_MAP__MASK; + syn_data->size = header_size + PERF_ALIGN(size_mask, sizeof(u64)); + return zalloc(syn_data->size); } -void cpu_map_data__synthesize(struct perf_record_cpu_map_data *data, struct perf_cpu_map *map, - u16 type, int max) +static void cpu_map_data__synthesize(struct synthesize_cpu_map_data *data) { - data->type = type; - - switch (type) { + switch (data->type) { case PERF_CPU_MAP__CPUS: - synthesize_cpus((struct cpu_map_entries *) data->data, map); + synthesize_cpus(data); break; case PERF_CPU_MAP__MASK: - synthesize_mask((struct perf_record_record_cpu_map *)data->data, map, max); + synthesize_mask(data); + break; + case PERF_CPU_MAP__RANGE_CPUS: + synthesize_range_cpus(data); + break; default: break; } } -static struct perf_record_cpu_map *cpu_map_event__new(struct perf_cpu_map *map) +static struct perf_record_cpu_map *cpu_map_event__new(const struct perf_cpu_map *map) { - size_t size = sizeof(struct perf_record_cpu_map); + struct synthesize_cpu_map_data syn_data = { .map = map }; struct perf_record_cpu_map *event; - int max; - u16 type; - event = cpu_map_data__alloc(map, &size, &type, &max); + + event = cpu_map_data__alloc(&syn_data, sizeof(struct perf_event_header)); if (!event) return NULL; + syn_data.data = &event->data; event->header.type = PERF_RECORD_CPU_MAP; - event->header.size = size; - event->data.type = type; - - cpu_map_data__synthesize(&event->data, map, type, max); + event->header.size = syn_data.size; + cpu_map_data__synthesize(&syn_data); return event; } + int perf_event__synthesize_cpu_map(struct perf_tool *tool, - struct perf_cpu_map *map, + const struct perf_cpu_map *map, perf_event__handler_t process, struct machine *machine) { @@ -1248,6 +1391,7 @@ int perf_event__synthesize_stat_config(struct perf_tool *tool, ADD(AGGR_MODE, config->aggr_mode) ADD(INTERVAL, config->interval) ADD(SCALE, config->scale) + ADD(AGGR_LEVEL, config->aggr_level) WARN_ONCE(i != PERF_STAT_CONFIG_TERM__MAX, "stat config terms unbalanced\n"); @@ -1260,7 +1404,7 @@ int perf_event__synthesize_stat_config(struct perf_tool *tool, } int perf_event__synthesize_stat(struct perf_tool *tool, - u32 cpu, u32 thread, u64 id, + struct perf_cpu cpu, u32 thread, u64 id, struct perf_counts_values *count, perf_event__handler_t process, struct machine *machine) @@ -1272,7 +1416,7 @@ int perf_event__synthesize_stat(struct perf_tool *tool, event.header.misc = 0; event.id = id; - event.cpu = cpu; + event.cpu = cpu.cpu; event.thread = thread; event.val = count->val; event.ena = count->ena; @@ -1337,11 +1481,12 @@ size_t perf_event__sample_event_size(const struct perf_sample *sample, u64 type, result += sizeof(u64); /* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */ if (read_format & PERF_FORMAT_GROUP) { - sz = sample->read.group.nr * - sizeof(struct sample_read_value); - result += sz; + sz = sample_read_value_size(read_format); + result += sz * sample->read.group.nr; } else { result += sizeof(u64); + if (read_format & PERF_FORMAT_LOST) + result += sizeof(u64); } } @@ -1381,7 +1526,7 @@ size_t perf_event__sample_event_size(const struct perf_sample *sample, u64 type, } } - if (type & PERF_SAMPLE_WEIGHT) + if (type & PERF_SAMPLE_WEIGHT_TYPE) result += sizeof(u64); if (type & PERF_SAMPLE_DATA_SRC) @@ -1406,6 +1551,12 @@ size_t perf_event__sample_event_size(const struct perf_sample *sample, u64 type, if (type & PERF_SAMPLE_CGROUP) result += sizeof(u64); + if (type & PERF_SAMPLE_DATA_PAGE_SIZE) + result += sizeof(u64); + + if (type & PERF_SAMPLE_CODE_PAGE_SIZE) + result += sizeof(u64); + if (type & PERF_SAMPLE_AUX) { result += sizeof(u64); result += sample->aux_sample.size; @@ -1414,6 +1565,26 @@ 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, + __u64 *array, u64 type __maybe_unused) +{ + *array = data->weight; +} + +static __u64 *copy_read_group_values(__u64 *array, __u64 read_format, + const struct perf_sample *sample) +{ + size_t sz = sample_read_value_size(read_format); + struct sample_read_value *v = sample->read.group.values; + + sample_read_group__for_each(v, sample->read.group.nr, read_format) { + /* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */ + memcpy(array, v, sz); + array = (void *)array + sz; + } + return array; +} + int perf_event__synthesize_sample(union perf_event *event, u64 type, u64 read_format, const struct perf_sample *sample) { @@ -1495,13 +1666,16 @@ int perf_event__synthesize_sample(union perf_event *event, u64 type, u64 read_fo /* PERF_FORMAT_ID is forced for PERF_SAMPLE_READ */ if (read_format & PERF_FORMAT_GROUP) { - sz = sample->read.group.nr * - sizeof(struct sample_read_value); - memcpy(array, sample->read.group.values, sz); - array = (void *)array + sz; + array = copy_read_group_values(array, read_format, + sample); } else { *array = sample->read.one.id; array++; + + if (read_format & PERF_FORMAT_LOST) { + *array = sample->read.one.lost; + array++; + } } } @@ -1549,8 +1723,8 @@ int perf_event__synthesize_sample(union perf_event *event, u64 type, u64 read_fo } } - if (type & PERF_SAMPLE_WEIGHT) { - *array = sample->weight; + if (type & PERF_SAMPLE_WEIGHT_TYPE) { + arch_perf_synthesize_sample_weight(sample, array, type); array++; } @@ -1585,6 +1759,16 @@ int perf_event__synthesize_sample(union perf_event *event, u64 type, u64 read_fo array++; } + if (type & PERF_SAMPLE_DATA_PAGE_SIZE) { + *array = sample->data_page_size; + array++; + } + + if (type & PERF_SAMPLE_CODE_PAGE_SIZE) { + *array = sample->code_page_size; + array++; + } + if (type & PERF_SAMPLE_AUX) { sz = sample->aux_sample.size; *array++ = sz; @@ -1595,65 +1779,136 @@ int perf_event__synthesize_sample(union perf_event *event, u64 type, u64 read_fo return 0; } -int perf_event__synthesize_id_index(struct perf_tool *tool, perf_event__handler_t process, - struct evlist *evlist, struct machine *machine) +int perf_event__synthesize_id_sample(__u64 *array, u64 type, const struct perf_sample *sample) +{ + __u64 *start = array; + + /* + * used for cross-endian analysis. See git commit 65014ab3 + * for why this goofiness is needed. + */ + union u64_swap u; + + if (type & PERF_SAMPLE_TID) { + u.val32[0] = sample->pid; + u.val32[1] = sample->tid; + *array = u.val64; + array++; + } + + if (type & PERF_SAMPLE_TIME) { + *array = sample->time; + array++; + } + + if (type & PERF_SAMPLE_ID) { + *array = sample->id; + array++; + } + + if (type & PERF_SAMPLE_STREAM_ID) { + *array = sample->stream_id; + array++; + } + + if (type & PERF_SAMPLE_CPU) { + u.val32[0] = sample->cpu; + u.val32[1] = 0; + *array = u.val64; + array++; + } + + if (type & PERF_SAMPLE_IDENTIFIER) { + *array = sample->id; + array++; + } + + return (void *)array - (void *)start; +} + +int __perf_event__synthesize_id_index(struct perf_tool *tool, perf_event__handler_t process, + struct evlist *evlist, struct machine *machine, size_t from) { union perf_event *ev; struct evsel *evsel; - size_t nr = 0, i = 0, sz, max_nr, n; + size_t nr = 0, i = 0, sz, max_nr, n, pos; + size_t e1_sz = sizeof(struct id_index_entry); + size_t e2_sz = sizeof(struct id_index_entry_2); + size_t etot_sz = e1_sz + e2_sz; + bool e2_needed = false; int err; - pr_debug2("Synthesizing id index\n"); - - max_nr = (UINT16_MAX - sizeof(struct perf_record_id_index)) / - sizeof(struct id_index_entry); + max_nr = (UINT16_MAX - sizeof(struct perf_record_id_index)) / etot_sz; - evlist__for_each_entry(evlist, evsel) + pos = 0; + evlist__for_each_entry(evlist, evsel) { + if (pos++ < from) + continue; nr += evsel->core.ids; + } + + if (!nr) + return 0; + + pr_debug2("Synthesizing id index\n"); n = nr > max_nr ? max_nr : nr; - sz = sizeof(struct perf_record_id_index) + n * sizeof(struct id_index_entry); + sz = sizeof(struct perf_record_id_index) + n * etot_sz; ev = zalloc(sz); if (!ev) return -ENOMEM; + sz = sizeof(struct perf_record_id_index) + n * e1_sz; + ev->id_index.header.type = PERF_RECORD_ID_INDEX; - ev->id_index.header.size = sz; ev->id_index.nr = n; + pos = 0; evlist__for_each_entry(evlist, evsel) { u32 j; - for (j = 0; j < evsel->core.ids; j++) { + if (pos++ < from) + continue; + for (j = 0; j < evsel->core.ids; j++, i++) { struct id_index_entry *e; + struct id_index_entry_2 *e2; struct perf_sample_id *sid; if (i >= n) { + ev->id_index.header.size = sz + (e2_needed ? n * e2_sz : 0); err = process(tool, ev, NULL, machine); if (err) goto out_err; nr -= n; i = 0; + e2_needed = false; } - e = &ev->id_index.entries[i++]; + e = &ev->id_index.entries[i]; e->id = evsel->core.id[j]; - sid = perf_evlist__id2sid(evlist, e->id); + sid = evlist__id2sid(evlist, e->id); if (!sid) { free(ev); return -ENOENT; } e->idx = sid->idx; - e->cpu = sid->cpu; + e->cpu = sid->cpu.cpu; e->tid = sid->tid; + + if (sid->machine_pid) + e2_needed = true; + + e2 = (void *)ev + sz; + e2[i].machine_pid = sid->machine_pid; + e2[i].vcpu = sid->vcpu.cpu; } } - sz = sizeof(struct perf_record_id_index) + nr * sizeof(struct id_index_entry); - ev->id_index.header.size = sz; + sz = sizeof(struct perf_record_id_index) + nr * e1_sz; + ev->id_index.header.size = sz + (e2_needed ? nr * e2_sz : 0); ev->id_index.nr = nr; err = process(tool, ev, NULL, machine); @@ -1663,28 +1918,54 @@ out_err: return err; } +int perf_event__synthesize_id_index(struct perf_tool *tool, perf_event__handler_t process, + struct evlist *evlist, struct machine *machine) +{ + return __perf_event__synthesize_id_index(tool, process, evlist, machine, 0); +} + int __machine__synthesize_threads(struct machine *machine, struct perf_tool *tool, struct target *target, struct perf_thread_map *threads, - perf_event__handler_t process, bool data_mmap, - unsigned int nr_threads_synthesize) + perf_event__handler_t process, bool needs_mmap, + bool data_mmap, unsigned int nr_threads_synthesize) { + /* + * When perf runs in non-root PID namespace, and the namespace's proc FS + * is not mounted, nsinfo__is_in_root_namespace() returns false. + * In this case, the proc FS is coming for the parent namespace, thus + * perf tool will wrongly gather process info from its parent PID + * namespace. + * + * To avoid the confusion that the perf tool runs in a child PID + * namespace but it synthesizes thread info from its parent PID + * namespace, returns failure with warning. + */ + if (!nsinfo__is_in_root_namespace()) { + pr_err("Perf runs in non-root PID namespace but it tries to "); + pr_err("gather process info from its parent PID namespace.\n"); + pr_err("Please mount the proc file system properly, e.g. "); + pr_err("add the option '--mount-proc' for unshare command.\n"); + return -EPERM; + } + if (target__has_task(target)) - return perf_event__synthesize_thread_map(tool, threads, process, machine, data_mmap); + return perf_event__synthesize_thread_map(tool, threads, process, machine, + needs_mmap, data_mmap); else if (target__has_cpu(target)) - return perf_event__synthesize_threads(tool, process, - machine, data_mmap, + return perf_event__synthesize_threads(tool, process, machine, + needs_mmap, data_mmap, nr_threads_synthesize); /* command specified */ return 0; } int machine__synthesize_threads(struct machine *machine, struct target *target, - struct perf_thread_map *threads, bool data_mmap, - unsigned int nr_threads_synthesize) + struct perf_thread_map *threads, bool needs_mmap, + bool data_mmap, unsigned int nr_threads_synthesize) { return __machine__synthesize_threads(machine, NULL, target, threads, - perf_event__process, data_mmap, - nr_threads_synthesize); + perf_event__process, needs_mmap, + data_mmap, nr_threads_synthesize); } static struct perf_record_event_update *event_update_event__new(size_t size, u64 type, u64 id) @@ -1715,7 +1996,7 @@ int perf_event__synthesize_event_update_unit(struct perf_tool *tool, struct evse if (ev == NULL) return -ENOMEM; - strlcpy(ev->data, evsel->unit, size + 1); + strlcpy(ev->unit, evsel->unit, size + 1); err = process(tool, (union perf_event *)ev, NULL, NULL); free(ev); return err; @@ -1732,8 +2013,7 @@ int perf_event__synthesize_event_update_scale(struct perf_tool *tool, struct evs if (ev == NULL) return -ENOMEM; - ev_data = (struct perf_record_event_update_scale *)ev->data; - ev_data->scale = evsel->scale; + ev->scale.scale = evsel->scale; err = process(tool, (union perf_event *)ev, NULL, NULL); free(ev); return err; @@ -1743,14 +2023,14 @@ int perf_event__synthesize_event_update_name(struct perf_tool *tool, struct evse perf_event__handler_t process) { struct perf_record_event_update *ev; - size_t len = strlen(evsel->name); + size_t len = strlen(evsel__name(evsel)); int err; ev = event_update_event__new(len + 1, PERF_EVENT_UPDATE__NAME, evsel->core.id[0]); if (ev == NULL) return -ENOMEM; - strlcpy(ev->data, evsel->name, len + 1); + strlcpy(ev->name, evsel->name, len + 1); err = process(tool, (union perf_event *)ev, NULL, NULL); free(ev); return err; @@ -1759,25 +2039,20 @@ int perf_event__synthesize_event_update_name(struct perf_tool *tool, struct evse int perf_event__synthesize_event_update_cpus(struct perf_tool *tool, struct evsel *evsel, perf_event__handler_t process) { - size_t size = sizeof(struct perf_record_event_update); + struct synthesize_cpu_map_data syn_data = { .map = evsel->core.own_cpus }; struct perf_record_event_update *ev; - int max, err; - u16 type; - - if (!evsel->core.own_cpus) - return 0; + int err; - ev = cpu_map_data__alloc(evsel->core.own_cpus, &size, &type, &max); + ev = cpu_map_data__alloc(&syn_data, sizeof(struct perf_event_header) + 2 * sizeof(u64)); if (!ev) return -ENOMEM; + syn_data.data = &ev->cpus.cpus; ev->header.type = PERF_RECORD_EVENT_UPDATE; - ev->header.size = (u16)size; + ev->header.size = (u16)syn_data.size; ev->type = PERF_EVENT_UPDATE__CPUS; ev->id = evsel->core.id[0]; - - cpu_map_data__synthesize((struct perf_record_cpu_map_data *)ev->data, - evsel->core.own_cpus, type, max); + cpu_map_data__synthesize(&syn_data); err = process(tool, (union perf_event *)ev, NULL, NULL); free(ev); @@ -1886,7 +2161,7 @@ int perf_event__synthesize_attr(struct perf_tool *tool, struct perf_event_attr * return -ENOMEM; ev->attr.attr = *attr; - memcpy(ev->attr.id, id, ids * sizeof(u64)); + memcpy(perf_record_header_attr_id(ev), id, ids * sizeof(u64)); ev->attr.header.type = PERF_RECORD_HEADER_ATTR; ev->attr.header.size = (u16)size; @@ -1901,6 +2176,7 @@ int perf_event__synthesize_attr(struct perf_tool *tool, struct perf_event_attr * return err; } +#ifdef HAVE_LIBTRACEEVENT int perf_event__synthesize_tracing_data(struct perf_tool *tool, int fd, struct evlist *evlist, perf_event__handler_t process) { @@ -1947,6 +2223,7 @@ int perf_event__synthesize_tracing_data(struct perf_tool *tool, int fd, struct e return aligned_size; } +#endif int perf_event__synthesize_build_id(struct perf_tool *tool, struct dso *pos, u16 misc, perf_event__handler_t process, struct machine *machine) @@ -1961,9 +2238,10 @@ int perf_event__synthesize_build_id(struct perf_tool *tool, struct dso *pos, u16 len = pos->long_name_len + 1; len = PERF_ALIGN(len, NAME_ALIGN); - memcpy(&ev.build_id.build_id, pos->bid.data, sizeof(pos->bid.data)); + ev.build_id.size = min(pos->bid.size, sizeof(pos->bid.data)); + memcpy(&ev.build_id.build_id, pos->bid.data, ev.build_id.size); ev.build_id.header.type = PERF_RECORD_HEADER_BUILD_ID; - ev.build_id.header.misc = misc; + ev.build_id.header.misc = misc | PERF_RECORD_MISC_BUILD_ID_SIZE; ev.build_id.pid = machine->pid; ev.build_id.header.size = sizeof(ev.build_id) + len; memcpy(&ev.build_id.filename, pos->long_name, pos->long_name_len); @@ -1991,7 +2269,7 @@ int perf_event__synthesize_stat_events(struct perf_stat_config *config, struct p return err; } - err = perf_event__synthesize_cpu_map(tool, evlist->core.cpus, process, NULL); + err = perf_event__synthesize_cpu_map(tool, evlist->core.user_requested_cpus, process, NULL); if (err < 0) { pr_err("Couldn't synthesize thread map.\n"); return err; @@ -2070,3 +2348,85 @@ int perf_event__synthesize_features(struct perf_tool *tool, struct perf_session free(ff.buf); return ret; } + +int perf_event__synthesize_for_pipe(struct perf_tool *tool, + struct perf_session *session, + struct perf_data *data, + perf_event__handler_t process) +{ + int err; + int ret = 0; + struct evlist *evlist = session->evlist; + + /* + * We need to synthesize events first, because some + * features works on top of them (on report side). + */ + err = perf_event__synthesize_attrs(tool, evlist, process); + if (err < 0) { + pr_err("Couldn't synthesize attrs.\n"); + return err; + } + ret += err; + + err = perf_event__synthesize_features(tool, session, evlist, process); + if (err < 0) { + pr_err("Couldn't synthesize features.\n"); + return err; + } + ret += err; + +#ifdef HAVE_LIBTRACEEVENT + if (have_tracepoints(&evlist->core.entries)) { + int fd = perf_data__fd(data); + + /* + * FIXME err <= 0 here actually means that + * there were no tracepoints so its not really + * an error, just that we don't need to + * synthesize anything. We really have to + * return this more properly and also + * propagate errors that now are calling die() + */ + err = perf_event__synthesize_tracing_data(tool, fd, evlist, + process); + if (err <= 0) { + pr_err("Couldn't record tracing data.\n"); + return err; + } + ret += err; + } +#else + (void)data; +#endif + + return ret; +} + +int parse_synth_opt(char *synth) +{ + char *p, *q; + int ret = 0; + + if (synth == NULL) + return -1; + + for (q = synth; (p = strsep(&q, ",")); p = q) { + if (!strcasecmp(p, "no") || !strcasecmp(p, "none")) + return 0; + + if (!strcasecmp(p, "all")) + return PERF_SYNTH_ALL; + + if (!strcasecmp(p, "task")) + ret |= PERF_SYNTH_TASK; + else if (!strcasecmp(p, "mmap")) + ret |= PERF_SYNTH_TASK | PERF_SYNTH_MMAP; + else if (!strcasecmp(p, "cgroup")) + ret |= PERF_SYNTH_CGROUP; + else + return -1; + } + + return ret; +} |