diff options
Diffstat (limited to 'tools/perf/util/machine.c')
| -rw-r--r-- | tools/perf/util/machine.c | 2588 |
1 files changed, 1832 insertions, 756 deletions
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 2e9eb6aa3ce2..841b711d970e 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -1,17 +1,31 @@ +// SPDX-License-Identifier: GPL-2.0 #include <dirent.h> #include <errno.h> #include <inttypes.h> #include <regex.h> +#include <stdlib.h> #include "callchain.h" #include "debug.h" +#include "dso.h" +#include "env.h" #include "event.h" #include "evsel.h" #include "hist.h" #include "machine.h" #include "map.h" +#include "map_symbol.h" +#include "branch.h" +#include "mem-events.h" +#include "mem-info.h" +#include "path.h" +#include "srcline.h" +#include "symbol.h" +#include "synthetic-events.h" #include "sort.h" #include "strlist.h" +#include "target.h" #include "thread.h" +#include "util.h" #include "vdso.h" #include <stdbool.h> #include <sys/types.h> @@ -20,31 +34,57 @@ #include "unwind.h" #include "linux/hash.h" #include "asm/bug.h" +#include "bpf-event.h" +#include <internal/lib.h> // page_size +#include "cgroup.h" +#include "arm64-frame-pointer-unwind-support.h" +#include <api/io_dir.h> -#include "sane_ctype.h" +#include <linux/ctype.h> #include <symbol/kallsyms.h> +#include <linux/mman.h> +#include <linux/string.h> +#include <linux/zalloc.h> -static void __machine__remove_thread(struct machine *machine, struct thread *th, bool lock); +static struct dso *machine__kernel_dso(struct machine *machine) +{ + return map__dso(machine->vmlinux_map); +} -static void dsos__init(struct dsos *dsos) +static int machine__set_mmap_name(struct machine *machine) { - INIT_LIST_HEAD(&dsos->head); - dsos->root = RB_ROOT; - pthread_rwlock_init(&dsos->lock, NULL); + if (machine__is_host(machine)) + machine->mmap_name = strdup("[kernel.kallsyms]"); + else if (machine__is_default_guest(machine)) + machine->mmap_name = strdup("[guest.kernel.kallsyms]"); + else if (asprintf(&machine->mmap_name, "[guest.kernel.kallsyms.%d]", + machine->pid) < 0) + machine->mmap_name = NULL; + + return machine->mmap_name ? 0 : -ENOMEM; +} + +static void thread__set_guest_comm(struct thread *thread, pid_t pid) +{ + char comm[64]; + + snprintf(comm, sizeof(comm), "[guest/%d]", pid); + thread__set_comm(thread, comm, 0); } int machine__init(struct machine *machine, const char *root_dir, pid_t pid) { + int err = -ENOMEM; + memset(machine, 0, sizeof(*machine)); - map_groups__init(&machine->kmaps, machine); + machine->kmaps = maps__new(machine); + if (machine->kmaps == NULL) + return -ENOMEM; + RB_CLEAR_NODE(&machine->rb_node); dsos__init(&machine->dsos); - machine->threads = RB_ROOT; - pthread_rwlock_init(&machine->threads_lock, NULL); - machine->nr_threads = 0; - INIT_LIST_HEAD(&machine->dead_threads); - machine->last_match = NULL; + threads__init(&machine->threads); machine->vdso_info = NULL; machine->env = NULL; @@ -55,60 +95,103 @@ int machine__init(struct machine *machine, const char *root_dir, pid_t pid) machine->kptr_restrict_warned = false; machine->comm_exec = false; machine->kernel_start = 0; - - memset(machine->vmlinux_maps, 0, sizeof(machine->vmlinux_maps)); + machine->vmlinux_map = NULL; + /* There is no initial context switch in, so we start at 1. */ + machine->parallelism = 1; machine->root_dir = strdup(root_dir); if (machine->root_dir == NULL) - return -ENOMEM; + goto out; + + if (machine__set_mmap_name(machine)) + goto out; if (pid != HOST_KERNEL_ID) { struct thread *thread = machine__findnew_thread(machine, -1, pid); - char comm[64]; if (thread == NULL) - return -ENOMEM; + goto out; - snprintf(comm, sizeof(comm), "[guest/%d]", pid); - thread__set_comm(thread, comm, 0); + thread__set_guest_comm(thread, pid); thread__put(thread); } machine->current_tid = NULL; + err = 0; +out: + if (err) { + zfree(&machine->kmaps); + zfree(&machine->root_dir); + zfree(&machine->mmap_name); + } return 0; } -struct machine *machine__new_host(void) +static struct machine *__machine__new_host(struct perf_env *host_env, bool kernel_maps) { struct machine *machine = malloc(sizeof(*machine)); - if (machine != NULL) { - machine__init(machine, "", HOST_KERNEL_ID); + if (!machine) + return NULL; + + machine__init(machine, "", HOST_KERNEL_ID); - if (machine__create_kernel_maps(machine) < 0) - goto out_delete; + if (kernel_maps && machine__create_kernel_maps(machine) < 0) { + free(machine); + return NULL; } + machine->env = host_env; + return machine; +} +struct machine *machine__new_host(struct perf_env *host_env) +{ + return __machine__new_host(host_env, /*kernel_maps=*/true); +} + +static int mmap_handler(const struct perf_tool *tool __maybe_unused, + union perf_event *event, + struct perf_sample *sample, + struct machine *machine) +{ + return machine__process_mmap2_event(machine, event, sample); +} + +static int machine__init_live(struct machine *machine, pid_t pid) +{ + union perf_event event; + + memset(&event, 0, sizeof(event)); + return perf_event__synthesize_mmap_events(NULL, &event, pid, pid, + mmap_handler, machine, true); +} + +struct machine *machine__new_live(struct perf_env *host_env, bool kernel_maps, pid_t pid) +{ + struct machine *machine = __machine__new_host(host_env, kernel_maps); + + if (!machine) + return NULL; + + if (machine__init_live(machine, pid)) { + machine__delete(machine); + return NULL; + } return machine; -out_delete: - free(machine); - return NULL; } -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) MAP__FUNCTION will go away when we stop loading separate maps for - * functions and data objects. - * 2) We should switch to machine__load_kallsyms(), i.e. not explicitely + * 1) We should switch to machine__load_kallsyms(), i.e. not explicitly * ask for not using the kcore parsing code, once this one is fixed * to create a map per module. */ - if (machine && __machine__load_kallsyms(machine, "/proc/kallsyms", MAP__FUNCTION, true) <= 0) { + if (machine && machine__load_kallsyms(machine, "/proc/kallsyms") <= 0) { machine__delete(machine); machine = NULL; } @@ -116,52 +199,26 @@ struct machine *machine__new_kallsyms(void) return machine; } -static void dsos__purge(struct dsos *dsos) -{ - struct dso *pos, *n; - - pthread_rwlock_wrlock(&dsos->lock); - - list_for_each_entry_safe(pos, n, &dsos->head, node) { - RB_CLEAR_NODE(&pos->rb_node); - pos->root = NULL; - list_del_init(&pos->node); - dso__put(pos); - } - - pthread_rwlock_unlock(&dsos->lock); -} - -static void dsos__exit(struct dsos *dsos) -{ - dsos__purge(dsos); - pthread_rwlock_destroy(&dsos->lock); -} - void machine__delete_threads(struct machine *machine) { - struct rb_node *nd; - - pthread_rwlock_wrlock(&machine->threads_lock); - nd = rb_first(&machine->threads); - while (nd) { - struct thread *t = rb_entry(nd, struct thread, rb_node); - - nd = rb_next(nd); - __machine__remove_thread(machine, t, false); - } - pthread_rwlock_unlock(&machine->threads_lock); + threads__remove_all_threads(&machine->threads); } void machine__exit(struct machine *machine) { + if (machine == NULL) + return; + machine__destroy_kernel_maps(machine); - map_groups__exit(&machine->kmaps); + maps__zput(machine->kmaps); dsos__exit(&machine->dsos); machine__exit_vdso(machine); zfree(&machine->root_dir); + zfree(&machine->mmap_name); zfree(&machine->current_tid); - pthread_rwlock_destroy(&machine->threads_lock); + zfree(&machine->kallsyms_filename); + + threads__exit(&machine->threads); } void machine__delete(struct machine *machine) @@ -175,7 +232,7 @@ void machine__delete(struct machine *machine) void machines__init(struct machines *machines) { machine__init(&machines->host, "", HOST_KERNEL_ID); - machines->guests = RB_ROOT; + machines->guests = RB_ROOT_CACHED; } void machines__exit(struct machines *machines) @@ -187,9 +244,10 @@ void machines__exit(struct machines *machines) struct machine *machines__add(struct machines *machines, pid_t pid, const char *root_dir) { - struct rb_node **p = &machines->guests.rb_node; + struct rb_node **p = &machines->guests.rb_root.rb_node; struct rb_node *parent = NULL; struct machine *pos, *machine = malloc(sizeof(*machine)); + bool leftmost = true; if (machine == NULL) return NULL; @@ -204,12 +262,16 @@ struct machine *machines__add(struct machines *machines, pid_t pid, pos = rb_entry(parent, struct machine, rb_node); if (pid < pos->pid) p = &(*p)->rb_left; - else + else { p = &(*p)->rb_right; + leftmost = false; + } } rb_link_node(&machine->rb_node, parent, p); - rb_insert_color(&machine->rb_node, &machines->guests); + rb_insert_color_cached(&machine->rb_node, &machines->guests, leftmost); + + machine->machines = machines; return machine; } @@ -220,7 +282,7 @@ void machines__set_comm_exec(struct machines *machines, bool comm_exec) machines->host.comm_exec = comm_exec; - for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) { + for (nd = rb_first_cached(&machines->guests); nd; nd = rb_next(nd)) { struct machine *machine = rb_entry(nd, struct machine, rb_node); machine->comm_exec = comm_exec; @@ -229,7 +291,7 @@ void machines__set_comm_exec(struct machines *machines, bool comm_exec) struct machine *machines__find(struct machines *machines, pid_t pid) { - struct rb_node **p = &machines->guests.rb_node; + struct rb_node **p = &machines->guests.rb_root.rb_node; struct rb_node *parent = NULL; struct machine *machine; struct machine *default_machine = NULL; @@ -287,31 +349,113 @@ out: return machine; } +struct machine *machines__find_guest(struct machines *machines, pid_t pid) +{ + struct machine *machine = machines__find(machines, pid); + + if (!machine) + machine = machines__findnew(machines, DEFAULT_GUEST_KERNEL_ID); + return machine; +} + +/* + * A common case for KVM test programs is that the test program acts as the + * hypervisor, creating, running and destroying the virtual machine, and + * providing the guest object code from its own object code. In this case, + * the VM is not running an OS, but only the functions loaded into it by the + * hypervisor test program, and conveniently, loaded at the same virtual + * addresses. + * + * Normally to resolve addresses, MMAP events are needed to map addresses + * back to the object code and debug symbols for that object code. + * + * Currently, there is no way to get such mapping information from guests + * but, in the scenario described above, the guest has the same mappings + * as the hypervisor, so support for that scenario can be achieved. + * + * To support that, copy the host thread's maps to the guest thread's maps. + * Note, we do not discover the guest until we encounter a guest event, + * which works well because it is not until then that we know that the host + * thread's maps have been set up. + * + * This function returns the guest thread. Apart from keeping the data + * structures sane, using a thread belonging to the guest machine, instead + * of the host thread, allows it to have its own comm (refer + * thread__set_guest_comm()). + */ +static struct thread *findnew_guest_code(struct machine *machine, + struct machine *host_machine, + pid_t pid) +{ + struct thread *host_thread; + struct thread *thread; + int err; + + if (!machine) + return NULL; + + thread = machine__findnew_thread(machine, -1, pid); + if (!thread) + return NULL; + + /* Assume maps are set up if there are any */ + if (!maps__empty(thread__maps(thread))) + return thread; + + host_thread = machine__find_thread(host_machine, -1, pid); + if (!host_thread) + goto out_err; + + thread__set_guest_comm(thread, pid); + + /* + * Guest code can be found in hypervisor process at the same address + * so copy host maps. + */ + err = maps__copy_from(thread__maps(thread), thread__maps(host_thread)); + thread__put(host_thread); + if (err) + goto out_err; + + return thread; + +out_err: + thread__zput(thread); + return NULL; +} + +struct thread *machines__findnew_guest_code(struct machines *machines, pid_t pid) +{ + struct machine *host_machine = machines__find(machines, HOST_KERNEL_ID); + struct machine *machine = machines__findnew(machines, pid); + + return findnew_guest_code(machine, host_machine, pid); +} + +struct thread *machine__findnew_guest_code(struct machine *machine, pid_t pid) +{ + struct machines *machines = machine->machines; + struct machine *host_machine; + + if (!machines) + return NULL; + + host_machine = machines__find(machines, HOST_KERNEL_ID); + + return findnew_guest_code(machine, host_machine, pid); +} + void machines__process_guests(struct machines *machines, machine__process_t process, void *data) { struct rb_node *nd; - for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) { + for (nd = rb_first_cached(&machines->guests); nd; nd = rb_next(nd)) { struct machine *pos = rb_entry(nd, struct machine, rb_node); process(pos, data); } } -char *machine__mmap_name(struct machine *machine, char *bf, size_t size) -{ - if (machine__is_host(machine)) - snprintf(bf, size, "[%s]", "kernel.kallsyms"); - else if (machine__is_default_guest(machine)) - snprintf(bf, size, "[%s]", "guest.kernel.kallsyms"); - else { - snprintf(bf, size, "[%s.%d]", "guest.kernel.kallsyms", - machine->pid); - } - - return bf; -} - void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size) { struct rb_node *node; @@ -319,7 +463,8 @@ void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size) machines->host.id_hdr_size = id_hdr_size; - for (node = rb_first(&machines->guests); node; node = rb_next(node)) { + for (node = rb_first_cached(&machines->guests); node; + node = rb_next(node)) { machine = rb_entry(node, struct machine, rb_node); machine->id_hdr_size = id_hdr_size; } @@ -332,45 +477,45 @@ static void machine__update_thread_pid(struct machine *machine, { struct thread *leader; - if (pid == th->pid_ || pid == -1 || th->pid_ != -1) + if (pid == thread__pid(th) || pid == -1 || thread__pid(th) != -1) return; - th->pid_ = pid; + thread__set_pid(th, pid); - if (th->pid_ == th->tid) + if (thread__pid(th) == thread__tid(th)) return; - leader = __machine__findnew_thread(machine, th->pid_, th->pid_); + leader = machine__findnew_thread(machine, thread__pid(th), thread__pid(th)); if (!leader) goto out_err; - if (!leader->mg) - leader->mg = map_groups__new(machine); + if (!thread__maps(leader)) + thread__set_maps(leader, maps__new(machine)); - if (!leader->mg) + if (!thread__maps(leader)) goto out_err; - if (th->mg == leader->mg) - return; + if (thread__maps(th) == thread__maps(leader)) + goto out_put; - if (th->mg) { + if (thread__maps(th)) { /* * Maps are created from MMAP events which provide the pid and * tid. Consequently there never should be any maps on a thread * with an unknown pid. Just print an error if there are. */ - if (!map_groups__empty(th->mg)) + if (!maps__empty(thread__maps(th))) pr_err("Discarding thread maps for %d:%d\n", - th->pid_, th->tid); - map_groups__put(th->mg); + thread__pid(th), thread__tid(th)); + maps__put(thread__maps(th)); } - th->mg = map_groups__get(leader->mg); + thread__set_maps(th, maps__get(thread__maps(leader))); out_put: thread__put(leader); return; out_err: - pr_err("Failed to join map groups for %d:%d\n", th->pid_, th->tid); + pr_err("Failed to join map groups for %d:%d\n", thread__pid(th), thread__tid(th)); goto out_put; } @@ -378,102 +523,70 @@ out_err: * Caller must eventually drop thread->refcnt returned with a successful * lookup/new thread inserted. */ -static struct thread *____machine__findnew_thread(struct machine *machine, - pid_t pid, pid_t tid, - bool create) +static struct thread *__machine__findnew_thread(struct machine *machine, + pid_t pid, + pid_t tid, + bool create) { - struct rb_node **p = &machine->threads.rb_node; - struct rb_node *parent = NULL; - struct thread *th; + struct thread *th = threads__find(&machine->threads, tid); + bool created; - /* - * Front-end cache - TID lookups come in blocks, - * so most of the time we dont have to look up - * the full rbtree: - */ - th = machine->last_match; - if (th != NULL) { - if (th->tid == tid) { - machine__update_thread_pid(machine, th, pid); - return thread__get(th); - } - - machine->last_match = NULL; - } - - while (*p != NULL) { - parent = *p; - th = rb_entry(parent, struct thread, rb_node); - - if (th->tid == tid) { - machine->last_match = th; - machine__update_thread_pid(machine, th, pid); - return thread__get(th); - } - - if (tid < th->tid) - p = &(*p)->rb_left; - else - p = &(*p)->rb_right; + if (th) { + machine__update_thread_pid(machine, th, pid); + return th; } - if (!create) return NULL; - th = thread__new(pid, tid); - if (th != NULL) { - rb_link_node(&th->rb_node, parent, p); - rb_insert_color(&th->rb_node, &machine->threads); - + th = threads__findnew(&machine->threads, pid, tid, &created); + if (created) { /* - * We have to initialize map_groups separately - * after rb tree is updated. + * We have to initialize maps separately after rb tree is + * updated. * - * The reason is that we call machine__findnew_thread - * within thread__init_map_groups to find the thread - * leader and that would screwed the rb tree. + * The reason is that we call machine__findnew_thread within + * thread__init_maps to find the thread leader and that would + * screwed the rb tree. */ - if (thread__init_map_groups(th, machine)) { - rb_erase_init(&th->rb_node, &machine->threads); - RB_CLEAR_NODE(&th->rb_node); + if (thread__init_maps(th, machine)) { + pr_err("Thread init failed thread %d\n", pid); + threads__remove(&machine->threads, th); thread__put(th); return NULL; } - /* - * It is now in the rbtree, get a ref - */ - thread__get(th); - machine->last_match = th; - ++machine->nr_threads; - } + } else + machine__update_thread_pid(machine, th, pid); return th; } -struct thread *__machine__findnew_thread(struct machine *machine, pid_t pid, pid_t tid) +struct thread *machine__findnew_thread(struct machine *machine, pid_t pid, pid_t tid) { - return ____machine__findnew_thread(machine, pid, tid, true); + return __machine__findnew_thread(machine, pid, tid, /*create=*/true); } -struct thread *machine__findnew_thread(struct machine *machine, pid_t pid, - pid_t tid) +struct thread *machine__find_thread(struct machine *machine, pid_t pid, + pid_t tid) { - struct thread *th; - - pthread_rwlock_wrlock(&machine->threads_lock); - th = __machine__findnew_thread(machine, pid, tid); - pthread_rwlock_unlock(&machine->threads_lock); - return th; + return __machine__findnew_thread(machine, pid, tid, /*create=*/false); } -struct thread *machine__find_thread(struct machine *machine, pid_t pid, - pid_t tid) +/* + * Threads are identified by pid and tid, and the idle task has pid == tid == 0. + * So here a single thread is created for that, but actually there is a separate + * idle task per cpu, so there should be one 'struct thread' per cpu, but there + * is only 1. That causes problems for some tools, requiring workarounds. For + * example get_idle_thread() in builtin-sched.c, or thread_stack__per_cpu(). + */ +struct thread *machine__idle_thread(struct machine *machine) { - struct thread *th; - pthread_rwlock_rdlock(&machine->threads_lock); - th = ____machine__findnew_thread(machine, pid, tid, false); - pthread_rwlock_unlock(&machine->threads_lock); - return th; + struct thread *thread = machine__findnew_thread(machine, 0, 0); + + if (!thread || thread__set_comm(thread, "swapper", 0) || + thread__set_namespaces(thread, 0, NULL)) + pr_err("problem inserting idle task for machine pid %d\n", machine->pid); + + return thread; } struct comm *machine__thread_exec_comm(struct machine *machine, @@ -542,10 +655,26 @@ int machine__process_namespaces_event(struct machine *machine __maybe_unused, return err; } +int machine__process_cgroup_event(struct machine *machine, + union perf_event *event, + struct perf_sample *sample __maybe_unused) +{ + struct cgroup *cgrp; + + if (dump_trace) + perf_event__fprintf_cgroup(event, stdout); + + cgrp = cgroup__findnew(machine->env, event->cgroup.id, event->cgroup.path); + if (cgrp == NULL) + return -ENOMEM; + + return 0; +} + int machine__process_lost_event(struct machine *machine __maybe_unused, union perf_event *event, struct perf_sample *sample __maybe_unused) { - dump_printf(": id:%" PRIu64 ": lost:%" PRIu64 "\n", + dump_printf(": id:%" PRI_lu64 ": lost:%" PRI_lu64 "\n", event->lost.id, event->lost.lost); return 0; } @@ -553,35 +682,12 @@ int machine__process_lost_event(struct machine *machine __maybe_unused, int machine__process_lost_samples_event(struct machine *machine __maybe_unused, union perf_event *event, struct perf_sample *sample) { - dump_printf(": id:%" PRIu64 ": lost samples :%" PRIu64 "\n", - sample->id, event->lost_samples.lost); + dump_printf(": id:%" PRIu64 ": lost samples :%" PRI_lu64 "%s\n", + sample->id, event->lost_samples.lost, + event->header.misc & PERF_RECORD_MISC_LOST_SAMPLES_BPF ? " (BPF)" : ""); return 0; } -static struct dso *machine__findnew_module_dso(struct machine *machine, - struct kmod_path *m, - const char *filename) -{ - struct dso *dso; - - pthread_rwlock_wrlock(&machine->dsos.lock); - - dso = __dsos__find(&machine->dsos, m->name, true); - if (!dso) { - dso = __dsos__addnew(&machine->dsos, m->name); - if (dso == NULL) - goto out_unlock; - - dso__set_module_info(dso, m, machine); - dso__set_long_name(dso, strdup(filename), true); - } - - dso__get(dso); -out_unlock: - pthread_rwlock_unlock(&machine->dsos.lock); - return dso; -} - int machine__process_aux_event(struct machine *machine __maybe_unused, union perf_event *event) { @@ -598,81 +704,209 @@ int machine__process_itrace_start_event(struct machine *machine __maybe_unused, return 0; } +int machine__process_aux_output_hw_id_event(struct machine *machine __maybe_unused, + union perf_event *event) +{ + if (dump_trace) + perf_event__fprintf_aux_output_hw_id(event, stdout); + return 0; +} + int machine__process_switch_event(struct machine *machine __maybe_unused, union perf_event *event) { + bool out = event->header.misc & PERF_RECORD_MISC_SWITCH_OUT; + if (dump_trace) perf_event__fprintf_switch(event, stdout); + machine->parallelism += out ? -1 : 1; return 0; } -static void dso__adjust_kmod_long_name(struct dso *dso, const char *filename) +static int machine__process_ksymbol_register(struct machine *machine, + union perf_event *event, + struct perf_sample *sample __maybe_unused) { - const char *dup_filename; + struct symbol *sym; + struct dso *dso = NULL; + struct map *map = maps__find(machine__kernel_maps(machine), event->ksymbol.addr); + int err = 0; - if (!filename || !dso || !dso->long_name) - return; - if (dso->long_name[0] != '[') - return; - if (!strchr(filename, '/')) - return; + if (!map) { + dso = dso__new(event->ksymbol.name); - dup_filename = strdup(filename); - if (!dup_filename) - return; + if (!dso) { + err = -ENOMEM; + goto out; + } + dso__set_kernel(dso, DSO_SPACE__KERNEL); + map = map__new2(0, dso); + if (!map) { + err = -ENOMEM; + goto out; + } + if (event->ksymbol.ksym_type == PERF_RECORD_KSYMBOL_TYPE_OOL) { + dso__set_binary_type(dso, DSO_BINARY_TYPE__OOL); + dso__data(dso)->file_size = event->ksymbol.len; + dso__set_loaded(dso); + } - dso__set_long_name(dso, dup_filename, true); + map__set_start(map, event->ksymbol.addr); + map__set_end(map, map__start(map) + event->ksymbol.len); + err = maps__fixup_overlap_and_insert(machine__kernel_maps(machine), map); + if (err) { + err = -ENOMEM; + goto out; + } + + dso__set_loaded(dso); + + if (is_bpf_image(event->ksymbol.name)) { + dso__set_binary_type(dso, DSO_BINARY_TYPE__BPF_IMAGE); + dso__set_long_name(dso, "", false); + } + } else { + dso = dso__get(map__dso(map)); + } + + sym = symbol__new(map__map_ip(map, map__start(map)), + event->ksymbol.len, + 0, 0, event->ksymbol.name); + if (!sym) { + err = -ENOMEM; + goto out; + } + dso__insert_symbol(dso, sym); +out: + map__put(map); + dso__put(dso); + return err; } -struct map *machine__findnew_module_map(struct machine *machine, u64 start, - const char *filename) +static int machine__process_ksymbol_unregister(struct machine *machine, + union perf_event *event, + struct perf_sample *sample __maybe_unused) { - struct map *map = NULL; - struct dso *dso = NULL; - struct kmod_path m; + struct symbol *sym; + struct map *map; - if (kmod_path__parse_name(&m, filename)) - return NULL; + map = maps__find(machine__kernel_maps(machine), event->ksymbol.addr); + if (!map) + return 0; + + if (!RC_CHK_EQUAL(map, machine->vmlinux_map)) + maps__remove(machine__kernel_maps(machine), map); + else { + struct dso *dso = map__dso(map); + + sym = dso__find_symbol(dso, map__map_ip(map, map__start(map))); + if (sym) + dso__delete_symbol(dso, sym); + } + map__put(map); + return 0; +} + +int machine__process_ksymbol(struct machine *machine __maybe_unused, + union perf_event *event, + struct perf_sample *sample) +{ + if (dump_trace) + perf_event__fprintf_ksymbol(event, stdout); + + /* no need to process non-JIT BPF as it cannot get samples */ + if (event->ksymbol.len == 0) + return 0; + + if (event->ksymbol.flags & PERF_RECORD_KSYMBOL_FLAGS_UNREGISTER) + return machine__process_ksymbol_unregister(machine, event, + sample); + return machine__process_ksymbol_register(machine, event, sample); +} + +int machine__process_text_poke(struct machine *machine, union perf_event *event, + struct perf_sample *sample __maybe_unused) +{ + struct map *map = maps__find(machine__kernel_maps(machine), event->text_poke.addr); + u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; + struct dso *dso = map ? map__dso(map) : NULL; + + if (dump_trace) + perf_event__fprintf_text_poke(event, machine, stdout); + + if (!event->text_poke.new_len) + goto out; + + if (cpumode != PERF_RECORD_MISC_KERNEL) { + pr_debug("%s: unsupported cpumode - ignoring\n", __func__); + goto out; + } + + if (dso) { + u8 *new_bytes = event->text_poke.bytes + event->text_poke.old_len; + int ret; - map = map_groups__find_by_name(&machine->kmaps, MAP__FUNCTION, - m.name); - if (map) { /* - * If the map's dso is an offline module, give dso__load() - * a chance to find the file path of that module by fixing - * long_name. + * Kernel maps might be changed when loading symbols so loading + * must be done prior to using kernel maps. */ - dso__adjust_kmod_long_name(map->dso, filename); - goto out; + map__load(map); + ret = dso__data_write_cache_addr(dso, map, machine, + event->text_poke.addr, + new_bytes, + event->text_poke.new_len); + if (ret != event->text_poke.new_len) + pr_debug("Failed to write kernel text poke at %#" PRI_lx64 "\n", + event->text_poke.addr); + } else { + pr_debug("Failed to find kernel text poke address map for %#" PRI_lx64 "\n", + event->text_poke.addr); } +out: + map__put(map); + return 0; +} - dso = machine__findnew_module_dso(machine, &m, filename); +static struct map *machine__addnew_module_map(struct machine *machine, u64 start, + const char *filename) +{ + struct map *map = NULL; + struct kmod_path m; + struct dso *dso; + int err; + + if (kmod_path__parse_name(&m, filename)) + return NULL; + + dso = dsos__findnew_module_dso(&machine->dsos, machine, &m, filename); if (dso == NULL) goto out; - map = map__new2(start, dso, MAP__FUNCTION); + map = map__new2(start, dso); if (map == NULL) goto out; - map_groups__insert(&machine->kmaps, map); - - /* Put the map here because map_groups__insert alread got it */ - map__put(map); + err = maps__insert(machine__kernel_maps(machine), map); + /* If maps__insert failed, return NULL. */ + if (err) { + map__put(map); + map = NULL; + } out: /* put the dso here, corresponding to machine__findnew_module_dso */ dso__put(dso); - free(m.name); + zfree(&m.name); return map; } size_t machines__fprintf_dsos(struct machines *machines, FILE *fp) { struct rb_node *nd; - size_t ret = __dsos__fprintf(&machines->host.dsos.head, fp); + size_t ret = dsos__fprintf(&machines->host.dsos, fp); - for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) { + for (nd = rb_first_cached(&machines->guests); nd; nd = rb_next(nd)) { struct machine *pos = rb_entry(nd, struct machine, rb_node); - ret += __dsos__fprintf(&pos->dsos.head, fp); + ret += dsos__fprintf(&pos->dsos, fp); } return ret; @@ -681,7 +915,7 @@ size_t machines__fprintf_dsos(struct machines *machines, FILE *fp) size_t machine__fprintf_dsos_buildid(struct machine *m, FILE *fp, bool (skip)(struct dso *dso, int parm), int parm) { - return __dsos__fprintf_buildid(&m->dsos.head, fp, skip, parm); + return dsos__fprintf_buildid(&m->dsos, fp, skip, parm); } size_t machines__fprintf_dsos_buildid(struct machines *machines, FILE *fp, @@ -690,90 +924,67 @@ size_t machines__fprintf_dsos_buildid(struct machines *machines, FILE *fp, struct rb_node *nd; size_t ret = machine__fprintf_dsos_buildid(&machines->host, fp, skip, parm); - for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) { + for (nd = rb_first_cached(&machines->guests); nd; nd = rb_next(nd)) { struct machine *pos = rb_entry(nd, struct machine, rb_node); ret += machine__fprintf_dsos_buildid(pos, fp, skip, parm); } return ret; } -size_t machine__fprintf_vmlinux_path(struct machine *machine, FILE *fp) -{ - int i; - size_t printed = 0; - struct dso *kdso = machine__kernel_map(machine)->dso; - - if (kdso->has_build_id) { - char filename[PATH_MAX]; - if (dso__build_id_filename(kdso, filename, sizeof(filename))) - printed += fprintf(fp, "[0] %s\n", filename); - } +struct machine_fprintf_cb_args { + FILE *fp; + size_t printed; +}; - for (i = 0; i < vmlinux_path__nr_entries; ++i) - printed += fprintf(fp, "[%d] %s\n", - i + kdso->has_build_id, vmlinux_path[i]); +static int machine_fprintf_cb(struct thread *thread, void *data) +{ + struct machine_fprintf_cb_args *args = data; - return printed; + /* TODO: handle fprintf errors. */ + args->printed += thread__fprintf(thread, args->fp); + return 0; } size_t machine__fprintf(struct machine *machine, FILE *fp) { - size_t ret; - struct rb_node *nd; - - pthread_rwlock_rdlock(&machine->threads_lock); + struct machine_fprintf_cb_args args = { + .fp = fp, + .printed = 0, + }; + size_t ret = fprintf(fp, "Threads: %zu\n", threads__nr(&machine->threads)); - ret = fprintf(fp, "Threads: %u\n", machine->nr_threads); - - for (nd = rb_first(&machine->threads); nd; nd = rb_next(nd)) { - struct thread *pos = rb_entry(nd, struct thread, rb_node); - - ret += thread__fprintf(pos, fp); - } - - pthread_rwlock_unlock(&machine->threads_lock); - - return ret; + machine__for_each_thread(machine, machine_fprintf_cb, &args); + return ret + args.printed; } static struct dso *machine__get_kernel(struct machine *machine) { - const char *vmlinux_name = NULL; + const char *vmlinux_name = machine->mmap_name; struct dso *kernel; if (machine__is_host(machine)) { - vmlinux_name = symbol_conf.vmlinux_name; - if (!vmlinux_name) - vmlinux_name = DSO__NAME_KALLSYMS; + if (symbol_conf.vmlinux_name) + vmlinux_name = symbol_conf.vmlinux_name; kernel = machine__findnew_kernel(machine, vmlinux_name, - "[kernel]", DSO_TYPE_KERNEL); + "[kernel]", DSO_SPACE__KERNEL); } else { - char bf[PATH_MAX]; - - if (machine__is_default_guest(machine)) + if (symbol_conf.default_guest_vmlinux_name) vmlinux_name = symbol_conf.default_guest_vmlinux_name; - if (!vmlinux_name) - vmlinux_name = machine__mmap_name(machine, bf, - sizeof(bf)); kernel = machine__findnew_kernel(machine, vmlinux_name, "[guest.kernel]", - DSO_TYPE_GUEST_KERNEL); + DSO_SPACE__KERNEL_GUEST); } - if (kernel != NULL && (!kernel->has_build_id)) + if (kernel != NULL && (!dso__has_build_id(kernel))) dso__read_running_kernel_build_id(kernel, machine); return kernel; } -struct process_args { - u64 start; -}; - -static void machine__get_kallsyms_filename(struct machine *machine, char *buf, - size_t bufsz) +void machine__get_kallsyms_filename(struct machine *machine, char *buf, + size_t bufsz) { if (machine__is_default_guest(machine)) scnprintf(buf, bufsz, "%s", symbol_conf.default_guest_kallsyms); @@ -788,7 +999,8 @@ const char *ref_reloc_sym_names[] = {"_text", "_stext", NULL}; * symbol_name if it's not that important. */ static int machine__get_running_kernel_start(struct machine *machine, - const char **symbol_name, u64 *start) + const char **symbol_name, + u64 *start, u64 *end) { char filename[PATH_MAX]; int i, err = -1; @@ -813,71 +1025,186 @@ static int machine__get_running_kernel_start(struct machine *machine, *symbol_name = name; *start = addr; + + err = kallsyms__get_symbol_start(filename, "_edata", &addr); + if (err) + err = kallsyms__get_symbol_start(filename, "_etext", &addr); + if (!err) + *end = addr; + return 0; } -int __machine__create_kernel_maps(struct machine *machine, struct dso *kernel) +int machine__create_extra_kernel_map(struct machine *machine, + struct dso *kernel, + struct extra_kernel_map *xm) { - int type; - u64 start = 0; + struct kmap *kmap; + struct map *map; + int err; - if (machine__get_running_kernel_start(machine, NULL, &start)) - return -1; + map = map__new2(xm->start, kernel); + if (!map) + return -ENOMEM; - /* In case of renewal the kernel map, destroy previous one */ - machine__destroy_kernel_maps(machine); + map__set_end(map, xm->end); + map__set_pgoff(map, xm->pgoff); - for (type = 0; type < MAP__NR_TYPES; ++type) { - struct kmap *kmap; - struct map *map; + kmap = map__kmap(map); - machine->vmlinux_maps[type] = map__new2(start, kernel, type); - if (machine->vmlinux_maps[type] == NULL) - return -1; + strlcpy(kmap->name, xm->name, KMAP_NAME_LEN); - machine->vmlinux_maps[type]->map_ip = - machine->vmlinux_maps[type]->unmap_ip = - identity__map_ip; - map = __machine__kernel_map(machine, type); - kmap = map__kmap(map); - if (!kmap) - return -1; + err = maps__insert(machine__kernel_maps(machine), map); - kmap->kmaps = &machine->kmaps; - map_groups__insert(&machine->kmaps, map); + if (!err) { + pr_debug2("Added extra kernel map %s %" PRIx64 "-%" PRIx64 "\n", + kmap->name, map__start(map), map__end(map)); + } + + map__put(map); + + return err; +} + +static u64 find_entry_trampoline(struct dso *dso) +{ + /* Duplicates are removed so lookup all aliases */ + const char *syms[] = { + "_entry_trampoline", + "__entry_trampoline_start", + "entry_SYSCALL_64_trampoline", + }; + struct symbol *sym = dso__first_symbol(dso); + unsigned int i; + + for (; sym; sym = dso__next_symbol(sym)) { + if (sym->binding != STB_GLOBAL) + continue; + for (i = 0; i < ARRAY_SIZE(syms); i++) { + if (!strcmp(sym->name, syms[i])) + return sym->start; + } } return 0; } -void machine__destroy_kernel_maps(struct machine *machine) +/* + * These values can be used for kernels that do not have symbols for the entry + * trampolines in kallsyms. + */ +#define X86_64_CPU_ENTRY_AREA_PER_CPU 0xfffffe0000000000ULL +#define X86_64_CPU_ENTRY_AREA_SIZE 0x2c000 +#define X86_64_ENTRY_TRAMPOLINE 0x6000 + +struct machine__map_x86_64_entry_trampolines_args { + struct maps *kmaps; + bool found; +}; + +static int machine__map_x86_64_entry_trampolines_cb(struct map *map, void *data) { - int type; + struct machine__map_x86_64_entry_trampolines_args *args = data; + struct map *dest_map; + struct kmap *kmap = __map__kmap(map); - for (type = 0; type < MAP__NR_TYPES; ++type) { - struct kmap *kmap; - struct map *map = __machine__kernel_map(machine, type); + if (!kmap || !is_entry_trampoline(kmap->name)) + return 0; - if (map == NULL) - continue; + dest_map = maps__find(args->kmaps, map__pgoff(map)); + if (RC_CHK_ACCESS(dest_map) != RC_CHK_ACCESS(map)) + map__set_pgoff(map, map__map_ip(dest_map, map__pgoff(map))); - kmap = map__kmap(map); - map_groups__remove(&machine->kmaps, map); - if (kmap && kmap->ref_reloc_sym) { - /* - * ref_reloc_sym is shared among all maps, so free just - * on one of them. - */ - if (type == MAP__FUNCTION) { - zfree((char **)&kmap->ref_reloc_sym->name); - zfree(&kmap->ref_reloc_sym); - } else - kmap->ref_reloc_sym = NULL; - } + map__put(dest_map); + args->found = true; + return 0; +} + +/* Map x86_64 PTI entry trampolines */ +int machine__map_x86_64_entry_trampolines(struct machine *machine, + struct dso *kernel) +{ + struct machine__map_x86_64_entry_trampolines_args args = { + .kmaps = machine__kernel_maps(machine), + .found = false, + }; + int nr_cpus_avail, cpu; + u64 pgoff; + + /* + * In the vmlinux case, pgoff is a virtual address which must now be + * mapped to a vmlinux offset. + */ + maps__for_each_map(args.kmaps, machine__map_x86_64_entry_trampolines_cb, &args); + + if (args.found || machine->trampolines_mapped) + return 0; + + pgoff = find_entry_trampoline(kernel); + if (!pgoff) + return 0; + + nr_cpus_avail = machine__nr_cpus_avail(machine); + + /* Add a 1 page map for each CPU's entry trampoline */ + for (cpu = 0; cpu < nr_cpus_avail; cpu++) { + u64 va = X86_64_CPU_ENTRY_AREA_PER_CPU + + cpu * X86_64_CPU_ENTRY_AREA_SIZE + + X86_64_ENTRY_TRAMPOLINE; + struct extra_kernel_map xm = { + .start = va, + .end = va + page_size, + .pgoff = pgoff, + }; + + strlcpy(xm.name, ENTRY_TRAMPOLINE_NAME, KMAP_NAME_LEN); - map__put(machine->vmlinux_maps[type]); - machine->vmlinux_maps[type] = NULL; + if (machine__create_extra_kernel_map(machine, kernel, &xm) < 0) + return -1; + } + + machine->trampolines_mapped = nr_cpus_avail; + + return 0; +} + +int __weak machine__create_extra_kernel_maps(struct machine *machine __maybe_unused, + struct dso *kernel __maybe_unused) +{ + return 0; +} + +static int +__machine__create_kernel_maps(struct machine *machine, struct dso *kernel) +{ + /* In case of renewal the kernel map, destroy previous one */ + machine__destroy_kernel_maps(machine); + + map__put(machine->vmlinux_map); + machine->vmlinux_map = map__new2(0, kernel); + if (machine->vmlinux_map == NULL) + return -ENOMEM; + + map__set_mapping_type(machine->vmlinux_map, MAPPING_TYPE__IDENTITY); + return maps__insert(machine__kernel_maps(machine), machine->vmlinux_map); +} + +void machine__destroy_kernel_maps(struct machine *machine) +{ + struct kmap *kmap; + struct map *map = machine__kernel_map(machine); + + if (map == NULL) + return; + + kmap = map__kmap(map); + maps__remove(machine__kernel_maps(machine), map); + if (kmap && kmap->ref_reloc_sym) { + zfree((char **)&kmap->ref_reloc_sym->name); + zfree(&kmap->ref_reloc_sym); } + + map__zput(machine->vmlinux_map); } int machines__create_guest_kernel_maps(struct machines *machines) @@ -931,7 +1258,7 @@ failure: void machines__destroy_kernel_maps(struct machines *machines) { - struct rb_node *next = rb_first(&machines->guests); + struct rb_node *next = rb_first_cached(&machines->guests); machine__destroy_kernel_maps(&machines->host); @@ -939,7 +1266,7 @@ void machines__destroy_kernel_maps(struct machines *machines) struct machine *pos = rb_entry(next, struct machine, rb_node); next = rb_next(&pos->rb_node); - rb_erase(&pos->rb_node, &machines->guests); + rb_erase_cached(&pos->rb_node, &machines->guests); machine__delete(pos); } } @@ -954,49 +1281,37 @@ int machines__create_kernel_maps(struct machines *machines, pid_t pid) return machine__create_kernel_maps(machine); } -int __machine__load_kallsyms(struct machine *machine, const char *filename, - enum map_type type, bool no_kcore) +int machine__load_kallsyms(struct machine *machine, const char *filename) { struct map *map = machine__kernel_map(machine); - int ret = __dso__load_kallsyms(map->dso, filename, map, no_kcore); + struct dso *dso = map__dso(map); + int ret = __dso__load_kallsyms(dso, filename, map, true); if (ret > 0) { - dso__set_loaded(map->dso, type); + dso__set_loaded(dso); /* * Since /proc/kallsyms will have multiple sessions for the * kernel, with modules between them, fixup the end of all * sections. */ - __map_groups__fixup_end(&machine->kmaps, type); + maps__fixup_end(machine__kernel_maps(machine)); } return ret; } -int machine__load_kallsyms(struct machine *machine, const char *filename, - enum map_type type) -{ - return __machine__load_kallsyms(machine, filename, type, false); -} - -int machine__load_vmlinux_path(struct machine *machine, enum map_type type) +int machine__load_vmlinux_path(struct machine *machine) { struct map *map = machine__kernel_map(machine); - int ret = dso__load_vmlinux_path(map->dso, map); + struct dso *dso = map__dso(map); + int ret = dso__load_vmlinux_path(dso, map); if (ret > 0) - dso__set_loaded(map->dso, type); + dso__set_loaded(dso); return ret; } -static void map_groups__fixup_end(struct map_groups *mg) -{ - int i; - for (i = 0; i < MAP__NR_TYPES; ++i) - __map_groups__fixup_end(mg, i); -} - static char *get_kernel_version(const char *root_dir) { char version[PATH_MAX]; @@ -1009,9 +1324,10 @@ static char *get_kernel_version(const char *root_dir) if (!file) return NULL; - version[0] = '\0'; tmp = fgets(version, sizeof(version), file); fclose(file); + if (!tmp) + return NULL; name = strstr(version, prefix); if (!name) @@ -1026,59 +1342,59 @@ static char *get_kernel_version(const char *root_dir) static bool is_kmod_dso(struct dso *dso) { - return dso->symtab_type == DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE || - dso->symtab_type == DSO_BINARY_TYPE__GUEST_KMODULE; + return dso__symtab_type(dso) == DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE || + dso__symtab_type(dso) == DSO_BINARY_TYPE__GUEST_KMODULE; } -static int map_groups__set_module_path(struct map_groups *mg, const char *path, - struct kmod_path *m) +static int maps__set_module_path(struct maps *maps, const char *path, struct kmod_path *m) { - struct map *map; char *long_name; + struct dso *dso; + struct map *map = maps__find_by_name(maps, m->name); - map = map_groups__find_by_name(mg, MAP__FUNCTION, m->name); if (map == NULL) return 0; long_name = strdup(path); - if (long_name == NULL) + if (long_name == NULL) { + map__put(map); return -ENOMEM; + } - dso__set_long_name(map->dso, long_name, true); - dso__kernel_module_get_build_id(map->dso, ""); + dso = map__dso(map); + dso__set_long_name(dso, long_name, true); + dso__kernel_module_get_build_id(dso, ""); /* * Full name could reveal us kmod compression, so * we need to update the symtab_type if needed. */ - if (m->comp && is_kmod_dso(map->dso)) - map->dso->symtab_type++; - + if (m->comp && is_kmod_dso(dso)) { + dso__set_symtab_type(dso, dso__symtab_type(dso)+1); + dso__set_comp(dso, m->comp); + } + map__put(map); return 0; } -static int map_groups__set_modules_path_dir(struct map_groups *mg, - const char *dir_name, int depth) +static int maps__set_modules_path_dir(struct maps *maps, char *path, size_t path_size, int depth) { - struct dirent *dent; - DIR *dir = opendir(dir_name); + struct io_dirent64 *dent; + struct io_dir iod; + size_t root_len = strlen(path); int ret = 0; - if (!dir) { - pr_debug("%s: cannot open %s dir\n", __func__, dir_name); + io_dir__init(&iod, open(path, O_CLOEXEC | O_DIRECTORY | O_RDONLY)); + if (iod.dirfd < 0) { + pr_debug("%s: cannot open %s dir\n", __func__, path); return -1; } - - while ((dent = readdir(dir)) != NULL) { - char path[PATH_MAX]; - struct stat st; - - /*sshfs might return bad dent->d_type, so we have to stat*/ - snprintf(path, sizeof(path), "%s/%s", dir_name, dent->d_name); - if (stat(path, &st)) - continue; - - if (S_ISDIR(st.st_mode)) { + /* Bounds check, should never happen. */ + if (root_len >= path_size) + return -1; + path[root_len++] = '/'; + while ((dent = io_dir__readdir(&iod)) != NULL) { + if (io_dir__is_dir(&iod, dent)) { if (!strcmp(dent->d_name, ".") || !strcmp(dent->d_name, "..")) continue; @@ -1090,8 +1406,12 @@ static int map_groups__set_modules_path_dir(struct map_groups *mg, continue; } - ret = map_groups__set_modules_path_dir(mg, path, - depth + 1); + /* Bounds check, should never happen. */ + if (root_len + strlen(dent->d_name) >= path_size) + continue; + + strcpy(path + root_len, dent->d_name); + ret = maps__set_modules_path_dir(maps, path, path_size, depth + 1); if (ret < 0) goto out; } else { @@ -1101,10 +1421,15 @@ static int map_groups__set_modules_path_dir(struct map_groups *mg, if (ret) goto out; - if (m.kmod) - ret = map_groups__set_module_path(mg, path, &m); + if (m.kmod) { + /* Bounds check, should never happen. */ + if (root_len + strlen(dent->d_name) < path_size) { + strcpy(path + root_len, dent->d_name); + ret = maps__set_module_path(maps, path, &m); - free(m.name); + } + } + zfree(&m.name); if (ret) goto out; @@ -1112,7 +1437,7 @@ static int map_groups__set_modules_path_dir(struct map_groups *mg, } out: - closedir(dir); + close(iod.dirfd); return ret; } @@ -1129,28 +1454,32 @@ static int machine__set_modules_path(struct machine *machine) machine->root_dir, version); free(version); - return map_groups__set_modules_path_dir(&machine->kmaps, modules_path, 0); + return maps__set_modules_path_dir(machine__kernel_maps(machine), + modules_path, sizeof(modules_path), 0); } int __weak arch__fix_module_text_start(u64 *start __maybe_unused, + u64 *size __maybe_unused, const char *name __maybe_unused) { return 0; } -static int machine__create_module(void *arg, const char *name, u64 start) +static int machine__create_module(void *arg, const char *name, u64 start, + u64 size) { struct machine *machine = arg; struct map *map; - if (arch__fix_module_text_start(&start, name) < 0) + if (arch__fix_module_text_start(&start, &size, name) < 0) return -1; - map = machine__findnew_module_map(machine, start, name); + map = machine__addnew_module_map(machine, start, name); if (map == NULL) return -1; + map__set_end(map, start + size); - dso__kernel_module_get_build_id(map->dso, machine->root_dir); - + dso__kernel_module_get_build_id(map__dso(map), machine->root_dir); + map__put(map); return 0; } @@ -1180,20 +1509,50 @@ static int machine__create_modules(struct machine *machine) return 0; } +static void machine__set_kernel_mmap(struct machine *machine, + u64 start, u64 end) +{ + map__set_start(machine->vmlinux_map, start); + map__set_end(machine->vmlinux_map, end); + /* + * Be a bit paranoid here, some perf.data file came with + * a zero sized synthesized MMAP event for the kernel. + */ + if (start == 0 && end == 0) + map__set_end(machine->vmlinux_map, ~0ULL); +} + +static int machine__update_kernel_mmap(struct machine *machine, + u64 start, u64 end) +{ + struct map *orig, *updated; + int err; + + orig = machine->vmlinux_map; + updated = map__get(orig); + + machine->vmlinux_map = updated; + maps__remove(machine__kernel_maps(machine), orig); + machine__set_kernel_mmap(machine, start, end); + err = maps__insert(machine__kernel_maps(machine), updated); + map__put(orig); + + return err; +} + int machine__create_kernel_maps(struct machine *machine) { struct dso *kernel = machine__get_kernel(machine); const char *name = NULL; - u64 addr = 0; + u64 start = 0, end = ~0ULL; int ret; if (kernel == NULL) return -1; ret = __machine__create_kernel_maps(machine, kernel); - dso__put(kernel); if (ret < 0) - return -1; + goto out_put; if (symbol_conf.use_modules && machine__create_modules(machine) < 0) { if (machine__is_host(machine)) @@ -1204,148 +1563,151 @@ int machine__create_kernel_maps(struct machine *machine) "continuing anyway...\n", machine->pid); } - /* - * Now that we have all the maps created, just set the ->end of them: - */ - map_groups__fixup_end(&machine->kmaps); - - if (!machine__get_running_kernel_start(machine, &name, &addr)) { + if (!machine__get_running_kernel_start(machine, &name, &start, &end)) { if (name && - maps__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps, name, addr)) { + map__set_kallsyms_ref_reloc_sym(machine->vmlinux_map, name, start)) { machine__destroy_kernel_maps(machine); - return -1; + ret = -1; + goto out_put; } + + /* + * we have a real start address now, so re-order the kmaps + * assume it's the last in the kmaps + */ + ret = machine__update_kernel_mmap(machine, start, end); + if (ret < 0) + goto out_put; } - return 0; -} + if (machine__create_extra_kernel_maps(machine, kernel)) + pr_debug("Problems creating extra kernel maps, continuing anyway...\n"); -static void machine__set_kernel_mmap_len(struct machine *machine, - union perf_event *event) -{ - int i; + if (end == ~0ULL) { + /* update end address of the kernel map using adjacent module address */ + struct map *next = maps__find_next_entry(machine__kernel_maps(machine), + machine__kernel_map(machine)); - for (i = 0; i < MAP__NR_TYPES; i++) { - machine->vmlinux_maps[i]->start = event->mmap.start; - machine->vmlinux_maps[i]->end = (event->mmap.start + - event->mmap.len); - /* - * Be a bit paranoid here, some perf.data file came with - * a zero sized synthesized MMAP event for the kernel. - */ - if (machine->vmlinux_maps[i]->end == 0) - machine->vmlinux_maps[i]->end = ~0ULL; + if (next) { + machine__set_kernel_mmap(machine, start, map__start(next)); + map__put(next); + } } + + maps__fixup_end(machine__kernel_maps(machine)); + +out_put: + dso__put(kernel); + return ret; +} + +static int machine__uses_kcore_cb(struct dso *dso, void *data __maybe_unused) +{ + return dso__is_kcore(dso) ? 1 : 0; } static bool machine__uses_kcore(struct machine *machine) { - struct dso *dso; + return dsos__for_each_dso(&machine->dsos, machine__uses_kcore_cb, NULL) != 0 ? true : false; +} - list_for_each_entry(dso, &machine->dsos.head, node) { - if (dso__is_kcore(dso)) - return true; - } +static bool perf_event__is_extra_kernel_mmap(struct machine *machine, + struct extra_kernel_map *xm) +{ + return machine__is(machine, "x86_64") && + is_entry_trampoline(xm->name); +} - return false; +static int machine__process_extra_kernel_map(struct machine *machine, + struct extra_kernel_map *xm) +{ + struct dso *kernel = machine__kernel_dso(machine); + + if (kernel == NULL) + return -1; + + return machine__create_extra_kernel_map(machine, kernel, xm); } static int machine__process_kernel_mmap_event(struct machine *machine, - union perf_event *event) + struct extra_kernel_map *xm, + struct build_id *bid) { - struct map *map; - char kmmap_prefix[PATH_MAX]; - enum dso_kernel_type kernel_type; + enum dso_space_type dso_space; bool is_kernel_mmap; + const char *mmap_name = machine->mmap_name; /* If we have maps from kcore then we do not need or want any others */ if (machine__uses_kcore(machine)) return 0; - machine__mmap_name(machine, kmmap_prefix, sizeof(kmmap_prefix)); if (machine__is_host(machine)) - kernel_type = DSO_TYPE_KERNEL; + dso_space = DSO_SPACE__KERNEL; else - kernel_type = DSO_TYPE_GUEST_KERNEL; - - is_kernel_mmap = memcmp(event->mmap.filename, - kmmap_prefix, - strlen(kmmap_prefix) - 1) == 0; - if (event->mmap.filename[0] == '/' || - (!is_kernel_mmap && event->mmap.filename[0] == '[')) { - map = machine__findnew_module_map(machine, event->mmap.start, - event->mmap.filename); + dso_space = DSO_SPACE__KERNEL_GUEST; + + is_kernel_mmap = memcmp(xm->name, mmap_name, strlen(mmap_name) - 1) == 0; + if (!is_kernel_mmap && !machine__is_host(machine)) { + /* + * If the event was recorded inside the guest and injected into + * the host perf.data file, then it will match a host mmap_name, + * so try that - see machine__set_mmap_name(). + */ + mmap_name = "[kernel.kallsyms]"; + is_kernel_mmap = memcmp(xm->name, mmap_name, strlen(mmap_name) - 1) == 0; + } + if (xm->name[0] == '/' || + (!is_kernel_mmap && xm->name[0] == '[')) { + struct map *map = machine__addnew_module_map(machine, xm->start, xm->name); + if (map == NULL) goto out_problem; - map->end = map->start + event->mmap.len; + map__set_end(map, map__start(map) + xm->end - xm->start); + + if (build_id__is_defined(bid)) + dso__set_build_id(map__dso(map), bid); + + map__put(map); } else if (is_kernel_mmap) { - const char *symbol_name = (event->mmap.filename + - strlen(kmmap_prefix)); + const char *symbol_name = xm->name + strlen(mmap_name); /* * Should be there already, from the build-id table in * the header. */ - struct dso *kernel = NULL; - struct dso *dso; - - pthread_rwlock_rdlock(&machine->dsos.lock); - - list_for_each_entry(dso, &machine->dsos.head, node) { - - /* - * The cpumode passed to is_kernel_module is not the - * cpumode of *this* event. If we insist on passing - * correct cpumode to is_kernel_module, we should - * record the cpumode when we adding this dso to the - * linked list. - * - * However we don't really need passing correct - * cpumode. We know the correct cpumode must be kernel - * mode (if not, we should not link it onto kernel_dsos - * list). - * - * Therefore, we pass PERF_RECORD_MISC_CPUMODE_UNKNOWN. - * is_kernel_module() treats it as a kernel cpumode. - */ - - if (!dso->kernel || - is_kernel_module(dso->long_name, - PERF_RECORD_MISC_CPUMODE_UNKNOWN)) - continue; - - - kernel = dso; - break; - } - - pthread_rwlock_unlock(&machine->dsos.lock); + struct dso *kernel = dsos__find_kernel_dso(&machine->dsos); if (kernel == NULL) - kernel = machine__findnew_dso(machine, kmmap_prefix); + kernel = machine__findnew_dso(machine, machine->mmap_name); if (kernel == NULL) goto out_problem; - kernel->kernel = kernel_type; + dso__set_kernel(kernel, dso_space); if (__machine__create_kernel_maps(machine, kernel) < 0) { dso__put(kernel); goto out_problem; } - if (strstr(kernel->long_name, "vmlinux")) + if (strstr(dso__long_name(kernel), "vmlinux")) dso__set_short_name(kernel, "[kernel.vmlinux]", false); - machine__set_kernel_mmap_len(machine, event); + if (machine__update_kernel_mmap(machine, xm->start, xm->end) < 0) { + dso__put(kernel); + goto out_problem; + } + + if (build_id__is_defined(bid)) + dso__set_build_id(kernel, bid); /* * Avoid using a zero address (kptr_restrict) for the ref reloc * symbol. Effectively having zero here means that at record * time /proc/sys/kernel/kptr_restrict was non zero. */ - if (event->mmap.pgoff != 0) { - maps__set_kallsyms_ref_reloc_sym(machine->vmlinux_maps, - symbol_name, - event->mmap.pgoff); + if (xm->pgoff != 0) { + map__set_kallsyms_ref_reloc_sym(machine->vmlinux_map, + symbol_name, + xm->pgoff); } if (machine__is_default_guest(machine)) { @@ -1354,6 +1716,9 @@ static int machine__process_kernel_mmap_event(struct machine *machine, */ dso__load(kernel, machine__kernel_map(machine)); } + dso__put(kernel); + } else if (perf_event__is_extra_kernel_mmap(machine, xm)) { + return machine__process_extra_kernel_map(machine, xm); } return 0; out_problem: @@ -1366,15 +1731,33 @@ int machine__process_mmap2_event(struct machine *machine, { struct thread *thread; struct map *map; - enum map_type type; + 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) { + 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 || sample->cpumode == PERF_RECORD_MISC_KERNEL) { - ret = machine__process_kernel_mmap_event(machine, event); + struct extra_kernel_map xm = { + .start = event->mmap2.start, + .end = event->mmap2.start + event->mmap2.len, + .pgoff = event->mmap2.pgoff, + }; + + strlcpy(xm.name, event->mmap2.filename, KMAP_NAME_LEN); + ret = machine__process_kernel_mmap_event(machine, &xm, &dso_id.build_id); if (ret < 0) goto out_problem; return 0; @@ -1385,19 +1768,11 @@ int machine__process_mmap2_event(struct machine *machine, if (thread == NULL) goto out_problem; - if (event->header.misc & PERF_RECORD_MISC_MMAP_DATA) - type = MAP__VARIABLE; - else - type = MAP__FUNCTION; - map = map__new(machine, event->mmap2.start, event->mmap2.len, event->mmap2.pgoff, - event->mmap2.pid, event->mmap2.maj, - event->mmap2.min, event->mmap2.ino, - event->mmap2.ino_generation, - event->mmap2.prot, + &dso_id, event->mmap2.prot, event->mmap2.flags, - event->mmap2.filename, type, thread); + event->mmap2.filename, thread); if (map == NULL) goto out_problem_map; @@ -1424,7 +1799,7 @@ int machine__process_mmap_event(struct machine *machine, union perf_event *event { struct thread *thread; struct map *map; - enum map_type type; + u32 prot = 0; int ret = 0; if (dump_trace) @@ -1432,7 +1807,14 @@ int machine__process_mmap_event(struct machine *machine, union perf_event *event if (sample->cpumode == PERF_RECORD_MISC_GUEST_KERNEL || sample->cpumode == PERF_RECORD_MISC_KERNEL) { - ret = machine__process_kernel_mmap_event(machine, event); + struct extra_kernel_map xm = { + .start = event->mmap.start, + .end = event->mmap.start + event->mmap.len, + .pgoff = event->mmap.pgoff, + }; + + strlcpy(xm.name, event->mmap.filename, KMAP_NAME_LEN); + ret = machine__process_kernel_mmap_event(machine, &xm, NULL); if (ret < 0) goto out_problem; return 0; @@ -1443,16 +1825,12 @@ int machine__process_mmap_event(struct machine *machine, union perf_event *event if (thread == NULL) goto out_problem; - if (event->header.misc & PERF_RECORD_MISC_MMAP_DATA) - type = MAP__VARIABLE; - else - type = MAP__FUNCTION; + if (!(event->header.misc & PERF_RECORD_MISC_MMAP_DATA)) + prot = PROT_EXEC; map = map__new(machine, event->mmap.start, - event->mmap.len, event->mmap.pgoff, - event->mmap.pid, 0, 0, 0, 0, 0, 0, - event->mmap.filename, - type, thread); + event->mmap.len, event->mmap.pgoff, + &dso_id_empty, prot, /*flags=*/0, event->mmap.filename, thread); if (map == NULL) goto out_problem_map; @@ -1474,31 +1852,9 @@ out_problem: return 0; } -static void __machine__remove_thread(struct machine *machine, struct thread *th, bool lock) -{ - if (machine->last_match == th) - machine->last_match = NULL; - - BUG_ON(refcount_read(&th->refcnt) == 0); - if (lock) - pthread_rwlock_wrlock(&machine->threads_lock); - rb_erase_init(&th->rb_node, &machine->threads); - RB_CLEAR_NODE(&th->rb_node); - --machine->nr_threads; - /* - * Move it first to the dead_threads list, then drop the reference, - * if this is the last reference, then the thread__delete destructor - * will be called and we will remove it from the dead_threads list. - */ - list_add_tail(&th->node, &machine->dead_threads); - if (lock) - pthread_rwlock_unlock(&machine->threads_lock); - thread__put(th); -} - void machine__remove_thread(struct machine *machine, struct thread *th) { - return __machine__remove_thread(machine, th, true); + return threads__remove(&machine->threads, th); } int machine__process_fork_event(struct machine *machine, union perf_event *event, @@ -1510,6 +1866,7 @@ int machine__process_fork_event(struct machine *machine, union perf_event *event struct thread *parent = machine__findnew_thread(machine, event->fork.ppid, event->fork.ptid); + bool do_maps_clone = true; int err = 0; if (dump_trace) @@ -1521,9 +1878,9 @@ int machine__process_fork_event(struct machine *machine, union perf_event *event * (fork) event that would have removed the thread was lost. Assume the * latter case and continue on as best we can. */ - if (parent->pid_ != (pid_t)event->fork.ppid) { + if (thread__pid(parent) != (pid_t)event->fork.ppid) { dump_printf("removing erroneous parent thread %d/%d\n", - parent->pid_, parent->tid); + thread__pid(parent), thread__tid(parent)); machine__remove_thread(machine, parent); thread__put(parent); parent = machine__findnew_thread(machine, event->fork.ppid, @@ -1538,9 +1895,25 @@ int machine__process_fork_event(struct machine *machine, union perf_event *event thread = machine__findnew_thread(machine, event->fork.pid, event->fork.tid); + /* + * When synthesizing FORK events, we are trying to create thread + * objects for the already running tasks on the machine. + * + * Normally, for a kernel FORK event, we want to clone the parent's + * maps because that is what the kernel just did. + * + * But when synthesizing, this should not be done. If we do, we end up + * with overlapping maps as we process the synthesized MMAP2 events that + * get delivered shortly thereafter. + * + * Use the FORK event misc flags in an internal way to signal this + * situation, so we can elide the map clone when appropriate. + */ + if (event->fork.header.misc & PERF_RECORD_MISC_FORK_EXEC) + do_maps_clone = false; if (thread == NULL || parent == NULL || - thread__fork(thread, parent, sample->time) < 0) { + thread__fork(thread, parent, sample->time, do_maps_clone) < 0) { dump_printf("problem processing PERF_RECORD_FORK, skipping event.\n"); err = -1; } @@ -1560,11 +1933,15 @@ int machine__process_exit_event(struct machine *machine, union perf_event *event if (dump_trace) perf_event__fprintf_task(event, stdout); + /* There is no context switch out before exit, so we decrement here. */ + machine->parallelism--; if (thread != NULL) { - thread__exited(thread); - thread__put(thread); + if (symbol_conf.keep_exited_threads) + thread__set_exited(thread, /*exited=*/true); + else + machine__remove_thread(machine, thread); } - + thread__put(thread); return 0; } @@ -1580,6 +1957,8 @@ int machine__process_event(struct machine *machine, union perf_event *event, ret = machine__process_mmap_event(machine, event, sample); break; case PERF_RECORD_NAMESPACES: ret = machine__process_namespaces_event(machine, event, sample); break; + case PERF_RECORD_CGROUP: + ret = machine__process_cgroup_event(machine, event, sample); break; case PERF_RECORD_MMAP2: ret = machine__process_mmap2_event(machine, event, sample); break; case PERF_RECORD_FORK: @@ -1597,6 +1976,14 @@ int machine__process_event(struct machine *machine, union perf_event *event, case PERF_RECORD_SWITCH: case PERF_RECORD_SWITCH_CPU_WIDE: ret = machine__process_switch_event(machine, event); break; + case PERF_RECORD_KSYMBOL: + ret = machine__process_ksymbol(machine, event, sample); break; + case PERF_RECORD_BPF_EVENT: + ret = machine__process_bpf(machine, event, sample); break; + case PERF_RECORD_TEXT_POKE: + ret = machine__process_text_poke(machine, event, sample); break; + case PERF_RECORD_AUX_OUTPUT_HW_ID: + ret = machine__process_aux_output_hw_id_event(machine, event); break; default: ret = -1; break; @@ -1607,9 +1994,7 @@ int machine__process_event(struct machine *machine, union perf_event *event, static bool symbol__match_regex(struct symbol *sym, regex_t *regex) { - if (!regexec(regex, sym->name, 0, NULL, 0)) - return 1; - return 0; + return regexec(regex, sym->name, 0, NULL, 0) == 0; } static void ip__resolve_ams(struct thread *thread, @@ -1618,7 +2003,7 @@ static void ip__resolve_ams(struct thread *thread, { struct addr_location al; - memset(&al, 0, sizeof(al)); + addr_location__init(&al); /* * We cannot use the header.misc hint to determine whether a * branch stack address is user, kernel, guest, hypervisor. @@ -1626,52 +2011,85 @@ static void ip__resolve_ams(struct thread *thread, * Thus, we have to try consecutively until we find a match * or else, the symbol is unknown */ - thread__find_cpumode_addr_location(thread, MAP__FUNCTION, ip, &al); + thread__find_cpumode_addr_location(thread, ip, /*symbols=*/true, &al); ams->addr = ip; ams->al_addr = al.addr; - ams->sym = al.sym; - ams->map = al.map; + ams->al_level = al.level; + ams->ms.maps = maps__get(al.maps); + ams->ms.sym = al.sym; + ams->ms.map = map__get(al.map); + ams->phys_addr = 0; + ams->data_page_size = 0; + addr_location__exit(&al); } static void ip__resolve_data(struct thread *thread, - u8 m, struct addr_map_symbol *ams, u64 addr) + u8 m, struct addr_map_symbol *ams, + u64 addr, u64 phys_addr, u64 daddr_page_size) { struct addr_location al; - memset(&al, 0, sizeof(al)); + addr_location__init(&al); - thread__find_addr_location(thread, m, MAP__VARIABLE, addr, &al); - if (al.map == NULL) { - /* - * some shared data regions have execute bit set which puts - * their mapping in the MAP__FUNCTION type array. - * Check there as a fallback option before dropping the sample. - */ - thread__find_addr_location(thread, m, MAP__FUNCTION, addr, &al); - } + thread__find_symbol(thread, m, addr, &al); ams->addr = addr; ams->al_addr = al.addr; - ams->sym = al.sym; - ams->map = al.map; + ams->al_level = al.level; + ams->ms.maps = maps__get(al.maps); + ams->ms.sym = al.sym; + ams->ms.map = map__get(al.map); + ams->phys_addr = phys_addr; + ams->data_page_size = daddr_page_size; + addr_location__exit(&al); } struct mem_info *sample__resolve_mem(struct perf_sample *sample, struct addr_location *al) { - struct mem_info *mi = zalloc(sizeof(*mi)); + struct mem_info *mi = mem_info__new(); if (!mi) return NULL; - ip__resolve_ams(al->thread, &mi->iaddr, sample->ip); - ip__resolve_data(al->thread, al->cpumode, &mi->daddr, sample->addr); - mi->data_src.val = sample->data_src; + ip__resolve_ams(al->thread, mem_info__iaddr(mi), sample->ip); + ip__resolve_data(al->thread, al->cpumode, mem_info__daddr(mi), + sample->addr, sample->phys_addr, + sample->data_page_size); + mem_info__data_src(mi)->val = sample->data_src; return mi; } +static char *callchain_srcline(struct map_symbol *ms, u64 ip) +{ + struct map *map = ms->map; + char *srcline = NULL; + struct dso *dso; + + if (!map || callchain_param.key == CCKEY_FUNCTION) + return srcline; + + dso = map__dso(map); + srcline = srcline__tree_find(dso__srclines(dso), ip); + if (!srcline) { + bool show_sym = false; + bool show_addr = callchain_param.key == CCKEY_ADDRESS; + + srcline = get_srcline(dso, map__rip_2objdump(map, ip), + ms->sym, show_sym, show_addr, ip); + srcline__tree_insert(dso__srclines(dso), ip, srcline); + } + + return srcline; +} + +struct iterations { + int nr_loop_iter; + u64 cycles; +}; + static int add_callchain_ip(struct thread *thread, struct callchain_cursor *cursor, struct symbol **parent, @@ -1680,16 +2098,22 @@ static int add_callchain_ip(struct thread *thread, u64 ip, bool branch, struct branch_flags *flags, - int nr_loop_iter, - int samples) + struct iterations *iter, + u64 branch_from, + bool symbols) { + struct map_symbol ms = {}; struct addr_location al; + int nr_loop_iter = 0, err = 0; + u64 iter_cycles = 0; + const char *srcline = NULL; + addr_location__init(&al); al.filtered = 0; al.sym = NULL; + al.srcline = NULL; if (!cpumode) { - thread__find_cpumode_addr_location(thread, MAP__FUNCTION, - ip, &al); + thread__find_cpumode_addr_location(thread, ip, symbols, &al); } else { if (ip >= PERF_CONTEXT_MAX) { switch (ip) { @@ -1700,6 +2124,7 @@ static int add_callchain_ip(struct thread *thread, *cpumode = PERF_RECORD_MISC_KERNEL; break; case PERF_CONTEXT_USER: + case PERF_CONTEXT_USER_DEFERRED: *cpumode = PERF_RECORD_MISC_USER; break; default: @@ -1710,12 +2135,15 @@ static int add_callchain_ip(struct thread *thread, * Discard all. */ callchain_cursor_reset(cursor); - return 1; + err = 1; + goto out; } - return 0; + goto out; } - thread__find_addr_location(thread, *cpumode, MAP__FUNCTION, - ip, &al); + if (symbols) + thread__find_symbol(thread, *cpumode, ip, &al); + else + thread__find_map(thread, *cpumode, ip, &al); } if (al.sym != NULL) { @@ -1726,15 +2154,30 @@ static int add_callchain_ip(struct thread *thread, symbol__match_regex(al.sym, &ignore_callees_regex)) { /* Treat this symbol as the root, forgetting its callees. */ - *root_al = al; + addr_location__copy(root_al, &al); callchain_cursor_reset(cursor); } } if (symbol_conf.hide_unresolved && al.sym == NULL) - return 0; - return callchain_cursor_append(cursor, al.addr, al.map, al.sym, - branch, flags, nr_loop_iter, samples); + goto out; + + if (iter) { + nr_loop_iter = iter->nr_loop_iter; + iter_cycles = iter->cycles; + } + + ms.maps = maps__get(al.maps); + ms.map = map__get(al.map); + ms.sym = al.sym; + srcline = callchain_srcline(&ms, al.addr); + err = callchain_cursor_append(cursor, ip, &ms, + branch, flags, nr_loop_iter, + iter_cycles, branch_from, srcline); +out: + addr_location__exit(&al); + map_symbol__exit(&ms); + return err; } struct branch_info *sample__resolve_bstack(struct perf_sample *sample, @@ -1742,19 +2185,35 @@ struct branch_info *sample__resolve_bstack(struct perf_sample *sample, { unsigned int i; const struct branch_stack *bs = sample->branch_stack; + struct branch_entry *entries = perf_sample__branch_entries(sample); + u64 *branch_stack_cntr = sample->branch_stack_cntr; struct branch_info *bi = calloc(bs->nr, sizeof(struct branch_info)); if (!bi) return NULL; for (i = 0; i < bs->nr; i++) { - ip__resolve_ams(al->thread, &bi[i].to, bs->entries[i].to); - ip__resolve_ams(al->thread, &bi[i].from, bs->entries[i].from); - bi[i].flags = bs->entries[i].flags; + ip__resolve_ams(al->thread, &bi[i].to, entries[i].to); + ip__resolve_ams(al->thread, &bi[i].from, entries[i].from); + bi[i].flags = entries[i].flags; + if (branch_stack_cntr) + bi[i].branch_stack_cntr = branch_stack_cntr[i]; } return bi; } +static void save_iterations(struct iterations *iter, + struct branch_entry *be, int nr) +{ + int i; + + iter->nr_loop_iter++; + iter->cycles = 0; + + for (i = 0; i < nr; i++) + iter->cycles += be[i].flags.cycles; +} + #define CHASHSZ 127 #define CHASHBITS 7 #define NO_ENTRY 0xff @@ -1762,7 +2221,8 @@ struct branch_info *sample__resolve_bstack(struct perf_sample *sample, #define PERF_MAX_BRANCH_DEPTH 127 /* Remove loops. */ -static int remove_loops(struct branch_entry *l, int nr) +static int remove_loops(struct branch_entry *l, int nr, + struct iterations *iter) { int i, j, off; unsigned char chash[CHASHSZ]; @@ -1787,8 +2247,18 @@ static int remove_loops(struct branch_entry *l, int nr) break; } if (is_loop) { - memmove(l + i, l + i + off, - (nr - (i + off)) * sizeof(*l)); + j = nr - (i + off); + if (j > 0) { + save_iterations(iter + i + off, + l + i, off); + + memmove(iter + i, iter + i + off, + j * sizeof(*iter)); + + memmove(l + i, l + i + off, + j * sizeof(*l)); + } + nr -= off; } } @@ -1796,8 +2266,322 @@ static int remove_loops(struct branch_entry *l, int nr) return nr; } +static int lbr_callchain_add_kernel_ip(struct thread *thread, + struct callchain_cursor *cursor, + struct perf_sample *sample, + struct symbol **parent, + struct addr_location *root_al, + u64 branch_from, + bool callee, int end, + bool symbols) +{ + struct ip_callchain *chain = sample->callchain; + u8 cpumode = PERF_RECORD_MISC_USER; + int err, i; + + if (callee) { + for (i = 0; i < end + 1; i++) { + err = add_callchain_ip(thread, cursor, parent, + root_al, &cpumode, chain->ips[i], + false, NULL, NULL, branch_from, + symbols); + if (err) + return err; + } + return 0; + } + + for (i = end; i >= 0; i--) { + err = add_callchain_ip(thread, cursor, parent, + root_al, &cpumode, chain->ips[i], + false, NULL, NULL, branch_from, + symbols); + if (err) + return err; + } + + return 0; +} + +static void save_lbr_cursor_node(struct thread *thread, + struct callchain_cursor *cursor, + int idx) +{ + struct lbr_stitch *lbr_stitch = thread__lbr_stitch(thread); + + if (!lbr_stitch) + return; + + if (cursor->pos == cursor->nr) { + lbr_stitch->prev_lbr_cursor[idx].valid = false; + return; + } + + if (!cursor->curr) + cursor->curr = cursor->first; + else + cursor->curr = cursor->curr->next; + + map_symbol__exit(&lbr_stitch->prev_lbr_cursor[idx].ms); + memcpy(&lbr_stitch->prev_lbr_cursor[idx], cursor->curr, + sizeof(struct callchain_cursor_node)); + lbr_stitch->prev_lbr_cursor[idx].ms.maps = maps__get(cursor->curr->ms.maps); + lbr_stitch->prev_lbr_cursor[idx].ms.map = map__get(cursor->curr->ms.map); + + lbr_stitch->prev_lbr_cursor[idx].valid = true; + cursor->pos++; +} + +static int lbr_callchain_add_lbr_ip(struct thread *thread, + struct callchain_cursor *cursor, + struct perf_sample *sample, + struct symbol **parent, + struct addr_location *root_al, + u64 *branch_from, + bool callee, + bool symbols) +{ + struct branch_stack *lbr_stack = sample->branch_stack; + struct branch_entry *entries = perf_sample__branch_entries(sample); + u8 cpumode = PERF_RECORD_MISC_USER; + int lbr_nr = lbr_stack->nr; + struct branch_flags *flags; + int err, i; + u64 ip; + + /* + * The curr and pos are not used in writing session. They are cleared + * in callchain_cursor_commit() when the writing session is closed. + * Using curr and pos to track the current cursor node. + */ + if (thread__lbr_stitch(thread)) { + cursor->curr = NULL; + cursor->pos = cursor->nr; + if (cursor->nr) { + cursor->curr = cursor->first; + for (i = 0; i < (int)(cursor->nr - 1); i++) + cursor->curr = cursor->curr->next; + } + } + + if (callee) { + /* Add LBR ip from first entries.to */ + ip = entries[0].to; + flags = &entries[0].flags; + *branch_from = entries[0].from; + err = add_callchain_ip(thread, cursor, parent, + root_al, &cpumode, ip, + true, flags, NULL, + *branch_from, symbols); + if (err) + return err; + + /* + * The number of cursor node increases. + * Move the current cursor node. + * But does not need to save current cursor node for entry 0. + * It's impossible to stitch the whole LBRs of previous sample. + */ + if (thread__lbr_stitch(thread) && (cursor->pos != cursor->nr)) { + if (!cursor->curr) + cursor->curr = cursor->first; + else + cursor->curr = cursor->curr->next; + cursor->pos++; + } + + /* Add LBR ip from entries.from one by one. */ + for (i = 0; i < lbr_nr; i++) { + ip = entries[i].from; + flags = &entries[i].flags; + err = add_callchain_ip(thread, cursor, parent, + root_al, &cpumode, ip, + true, flags, NULL, + *branch_from, symbols); + if (err) + return err; + save_lbr_cursor_node(thread, cursor, i); + } + return 0; + } + + /* Add LBR ip from entries.from one by one. */ + for (i = lbr_nr - 1; i >= 0; i--) { + ip = entries[i].from; + flags = &entries[i].flags; + err = add_callchain_ip(thread, cursor, parent, + root_al, &cpumode, ip, + true, flags, NULL, + *branch_from, symbols); + if (err) + return err; + save_lbr_cursor_node(thread, cursor, i); + } + + if (lbr_nr > 0) { + /* Add LBR ip from first entries.to */ + ip = entries[0].to; + flags = &entries[0].flags; + *branch_from = entries[0].from; + err = add_callchain_ip(thread, cursor, parent, + root_al, &cpumode, ip, + true, flags, NULL, + *branch_from, symbols); + if (err) + return err; + } + + return 0; +} + +static int lbr_callchain_add_stitched_lbr_ip(struct thread *thread, + struct callchain_cursor *cursor) +{ + struct lbr_stitch *lbr_stitch = thread__lbr_stitch(thread); + struct callchain_cursor_node *cnode; + struct stitch_list *stitch_node; + int err; + + list_for_each_entry(stitch_node, &lbr_stitch->lists, node) { + cnode = &stitch_node->cursor; + + err = callchain_cursor_append(cursor, cnode->ip, + &cnode->ms, + cnode->branch, + &cnode->branch_flags, + cnode->nr_loop_iter, + cnode->iter_cycles, + cnode->branch_from, + cnode->srcline); + if (err) + return err; + } + return 0; +} + +static struct stitch_list *get_stitch_node(struct thread *thread) +{ + struct lbr_stitch *lbr_stitch = thread__lbr_stitch(thread); + struct stitch_list *stitch_node; + + if (!list_empty(&lbr_stitch->free_lists)) { + stitch_node = list_first_entry(&lbr_stitch->free_lists, + struct stitch_list, node); + list_del(&stitch_node->node); + + return stitch_node; + } + + return malloc(sizeof(struct stitch_list)); +} + +static bool has_stitched_lbr(struct thread *thread, + struct perf_sample *cur, + struct perf_sample *prev, + unsigned int max_lbr, + bool callee) +{ + struct branch_stack *cur_stack = cur->branch_stack; + struct branch_entry *cur_entries = perf_sample__branch_entries(cur); + struct branch_stack *prev_stack = prev->branch_stack; + struct branch_entry *prev_entries = perf_sample__branch_entries(prev); + struct lbr_stitch *lbr_stitch = thread__lbr_stitch(thread); + int i, j, nr_identical_branches = 0; + struct stitch_list *stitch_node; + u64 cur_base, distance; + + if (!cur_stack || !prev_stack) + return false; + + /* Find the physical index of the base-of-stack for current sample. */ + cur_base = max_lbr - cur_stack->nr + cur_stack->hw_idx + 1; + + distance = (prev_stack->hw_idx > cur_base) ? (prev_stack->hw_idx - cur_base) : + (max_lbr + prev_stack->hw_idx - cur_base); + /* Previous sample has shorter stack. Nothing can be stitched. */ + if (distance + 1 > prev_stack->nr) + return false; + + /* + * Check if there are identical LBRs between two samples. + * Identical LBRs must have same from, to and flags values. Also, + * they have to be saved in the same LBR registers (same physical + * index). + * + * Starts from the base-of-stack of current sample. + */ + for (i = distance, j = cur_stack->nr - 1; (i >= 0) && (j >= 0); i--, j--) { + if ((prev_entries[i].from != cur_entries[j].from) || + (prev_entries[i].to != cur_entries[j].to) || + (prev_entries[i].flags.value != cur_entries[j].flags.value)) + break; + nr_identical_branches++; + } + + if (!nr_identical_branches) + return false; + + /* + * Save the LBRs between the base-of-stack of previous sample + * and the base-of-stack of current sample into lbr_stitch->lists. + * These LBRs will be stitched later. + */ + for (i = prev_stack->nr - 1; i > (int)distance; i--) { + + if (!lbr_stitch->prev_lbr_cursor[i].valid) + continue; + + stitch_node = get_stitch_node(thread); + if (!stitch_node) + return false; + + memcpy(&stitch_node->cursor, &lbr_stitch->prev_lbr_cursor[i], + sizeof(struct callchain_cursor_node)); + + stitch_node->cursor.ms.maps = maps__get(lbr_stitch->prev_lbr_cursor[i].ms.maps); + stitch_node->cursor.ms.map = map__get(lbr_stitch->prev_lbr_cursor[i].ms.map); + + if (callee) + list_add(&stitch_node->node, &lbr_stitch->lists); + else + list_add_tail(&stitch_node->node, &lbr_stitch->lists); + } + + return true; +} + +static bool alloc_lbr_stitch(struct thread *thread, unsigned int max_lbr) +{ + if (thread__lbr_stitch(thread)) + return true; + + thread__set_lbr_stitch(thread, zalloc(sizeof(struct lbr_stitch))); + if (!thread__lbr_stitch(thread)) + goto err; + + thread__lbr_stitch(thread)->prev_lbr_cursor = + calloc(max_lbr + 1, sizeof(struct callchain_cursor_node)); + if (!thread__lbr_stitch(thread)->prev_lbr_cursor) + goto free_lbr_stitch; + + thread__lbr_stitch(thread)->prev_lbr_cursor_size = max_lbr + 1; + + INIT_LIST_HEAD(&thread__lbr_stitch(thread)->lists); + INIT_LIST_HEAD(&thread__lbr_stitch(thread)->free_lists); + + return true; + +free_lbr_stitch: + free(thread__lbr_stitch(thread)); + thread__set_lbr_stitch(thread, NULL); +err: + pr_warning("Failed to allocate space for stitched LBRs. Disable LBR stitch\n"); + thread__set_lbr_stitch_enable(thread, false); + return false; +} + /* - * Recolve LBR callstack chain sample + * Resolve LBR callstack chain sample * Return: * 1 on success get LBR callchain information * 0 no available LBR callchain information, should try fp @@ -1808,12 +2592,17 @@ static int resolve_lbr_callchain_sample(struct thread *thread, struct perf_sample *sample, struct symbol **parent, struct addr_location *root_al, - int max_stack) + int max_stack, + unsigned int max_lbr, + bool symbols) { + bool callee = (callchain_param.order == ORDER_CALLEE); struct ip_callchain *chain = sample->callchain; int chain_nr = min(max_stack, (int)chain->nr), i; - u8 cpumode = PERF_RECORD_MISC_USER; - u64 ip; + struct lbr_stitch *lbr_stitch; + bool stitched_lbr = false; + u64 branch_from = 0; + int err; for (i = 0; i < chain_nr; i++) { if (chain->ips[i] == PERF_CONTEXT_USER) @@ -1821,89 +2610,130 @@ static int resolve_lbr_callchain_sample(struct thread *thread, } /* LBR only affects the user callchain */ - if (i != chain_nr) { - struct branch_stack *lbr_stack = sample->branch_stack; - int lbr_nr = lbr_stack->nr, j, k; - bool branch; - struct branch_flags *flags; - /* - * LBR callstack can only get user call chain. - * The mix_chain_nr is kernel call chain - * number plus LBR user call chain number. - * i is kernel call chain number, - * 1 is PERF_CONTEXT_USER, - * lbr_nr + 1 is the user call chain number. - * For details, please refer to the comments - * in callchain__printf - */ - int mix_chain_nr = i + 1 + lbr_nr + 1; + if (i == chain_nr) + return 0; - for (j = 0; j < mix_chain_nr; j++) { - int err; - branch = false; - flags = NULL; + if (thread__lbr_stitch_enable(thread) && !sample->no_hw_idx && + (max_lbr > 0) && alloc_lbr_stitch(thread, max_lbr)) { + lbr_stitch = thread__lbr_stitch(thread); - if (callchain_param.order == ORDER_CALLEE) { - if (j < i + 1) - ip = chain->ips[j]; - else if (j > i + 1) { - k = j - i - 2; - ip = lbr_stack->entries[k].from; - branch = true; - flags = &lbr_stack->entries[k].flags; - } else { - ip = lbr_stack->entries[0].to; - branch = true; - flags = &lbr_stack->entries[0].flags; - } - } else { - if (j < lbr_nr) { - k = lbr_nr - j - 1; - ip = lbr_stack->entries[k].from; - branch = true; - flags = &lbr_stack->entries[k].flags; - } - else if (j > lbr_nr) - ip = chain->ips[i + 1 - (j - lbr_nr)]; - else { - ip = lbr_stack->entries[0].to; - branch = true; - flags = &lbr_stack->entries[0].flags; - } - } + stitched_lbr = has_stitched_lbr(thread, sample, + &lbr_stitch->prev_sample, + max_lbr, callee); - err = add_callchain_ip(thread, cursor, parent, - root_al, &cpumode, ip, - branch, flags, 0, 0); + if (!stitched_lbr && !list_empty(&lbr_stitch->lists)) { + struct stitch_list *stitch_node; + + list_for_each_entry(stitch_node, &lbr_stitch->lists, node) + map_symbol__exit(&stitch_node->cursor.ms); + + list_splice_init(&lbr_stitch->lists, &lbr_stitch->free_lists); + } + memcpy(&lbr_stitch->prev_sample, sample, sizeof(*sample)); + } + + if (callee) { + /* Add kernel ip */ + err = lbr_callchain_add_kernel_ip(thread, cursor, sample, + parent, root_al, branch_from, + true, i, symbols); + if (err) + goto error; + + err = lbr_callchain_add_lbr_ip(thread, cursor, sample, parent, + root_al, &branch_from, true, symbols); + if (err) + goto error; + + if (stitched_lbr) { + err = lbr_callchain_add_stitched_lbr_ip(thread, cursor); if (err) - return (err < 0) ? err : 0; + goto error; } - return 1; + + } else { + if (stitched_lbr) { + err = lbr_callchain_add_stitched_lbr_ip(thread, cursor); + if (err) + goto error; + } + err = lbr_callchain_add_lbr_ip(thread, cursor, sample, parent, + root_al, &branch_from, false, symbols); + if (err) + goto error; + + /* Add kernel ip */ + err = lbr_callchain_add_kernel_ip(thread, cursor, sample, + parent, root_al, branch_from, + false, i, symbols); + if (err) + goto error; } + return 1; - return 0; +error: + return (err < 0) ? err : 0; +} + +static int find_prev_cpumode(struct ip_callchain *chain, struct thread *thread, + struct callchain_cursor *cursor, + struct symbol **parent, + struct addr_location *root_al, + u8 *cpumode, int ent, bool symbols) +{ + int err = 0; + + while (--ent >= 0) { + u64 ip = chain->ips[ent]; + + if (ip >= PERF_CONTEXT_MAX) { + err = add_callchain_ip(thread, cursor, parent, + root_al, cpumode, ip, + false, NULL, NULL, 0, symbols); + break; + } + } + return err; +} + +static u64 get_leaf_frame_caller(struct perf_sample *sample, + struct thread *thread, int usr_idx) +{ + if (machine__normalized_is(maps__machine(thread__maps(thread)), "arm64")) + return get_leaf_frame_caller_aarch64(sample, thread, usr_idx); + else + return 0; } static int thread__resolve_callchain_sample(struct thread *thread, struct callchain_cursor *cursor, - struct perf_evsel *evsel, + struct evsel *evsel, struct perf_sample *sample, struct symbol **parent, struct addr_location *root_al, - int max_stack) + int max_stack, + bool symbols) { struct branch_stack *branch = sample->branch_stack; + struct branch_entry *entries = perf_sample__branch_entries(sample); struct ip_callchain *chain = sample->callchain; - int chain_nr = chain->nr; + int chain_nr = 0; u8 cpumode = PERF_RECORD_MISC_USER; - int i, j, err, nr_entries; + int i, j, err, nr_entries, usr_idx; int skip_idx = -1; int first_call = 0; - int nr_loop_iter; + u64 leaf_frame_caller; + + if (chain) + chain_nr = chain->nr; + + if (evsel__has_branch_callstack(evsel)) { + struct perf_env *env = evsel__env(evsel); - if (perf_evsel__has_branch_callstack(evsel)) { err = resolve_lbr_callchain_sample(thread, cursor, sample, parent, - root_al, max_stack); + root_al, max_stack, + !env ? 0 : env->max_branches, + symbols); if (err) return (err < 0) ? err : 0; } @@ -1929,6 +2759,7 @@ static int thread__resolve_callchain_sample(struct thread *thread, if (branch && callchain_param.branch_callstack) { int nr = min(max_stack, (int)branch->nr); struct branch_entry be[nr]; + struct iterations iter[nr]; if (branch->nr > PERF_MAX_BRANCH_DEPTH) { pr_warning("corrupted branch chain. skipping...\n"); @@ -1937,7 +2768,11 @@ static int thread__resolve_callchain_sample(struct thread *thread, for (i = 0; i < nr; i++) { if (callchain_param.order == ORDER_CALLEE) { - be[i] = branch->entries[i]; + be[i] = entries[i]; + + if (chain == NULL) + continue; + /* * Check for overlap into the callchain. * The return address is one off compared to @@ -1952,49 +2787,44 @@ static int thread__resolve_callchain_sample(struct thread *thread, be[i].from >= chain->ips[first_call] - 8) first_call++; } else - be[i] = branch->entries[branch->nr - i - 1]; + be[i] = entries[branch->nr - i - 1]; } - nr_loop_iter = nr; - nr = remove_loops(be, nr); - - /* - * Get the number of iterations. - * It's only approximation, but good enough in practice. - */ - if (nr_loop_iter > nr) - nr_loop_iter = nr_loop_iter - nr + 1; - else - nr_loop_iter = 0; + memset(iter, 0, sizeof(struct iterations) * nr); + nr = remove_loops(be, nr, iter); for (i = 0; i < nr; i++) { - if (i == nr - 1) - err = add_callchain_ip(thread, cursor, parent, - root_al, - NULL, be[i].to, - true, &be[i].flags, - nr_loop_iter, 1); - else - err = add_callchain_ip(thread, cursor, parent, - root_al, - NULL, be[i].to, - true, &be[i].flags, - 0, 0); + err = add_callchain_ip(thread, cursor, parent, + root_al, + NULL, be[i].to, + true, &be[i].flags, + NULL, be[i].from, symbols); - if (!err) + if (!err) { err = add_callchain_ip(thread, cursor, parent, root_al, NULL, be[i].from, true, &be[i].flags, - 0, 0); + &iter[i], 0, symbols); + } if (err == -EINVAL) break; if (err) return err; } + + if (chain_nr == 0) + return 0; + chain_nr -= nr; } check_calls: + if (chain && callchain_param.order != ORDER_CALLEE) { + err = find_prev_cpumode(chain, thread, cursor, parent, root_al, + &cpumode, chain->nr - first_call, symbols); + if (err) + return (err < 0) ? err : 0; + } for (i = first_call, nr_entries = 0; i < chain_nr && nr_entries < max_stack; i++) { u64 ip; @@ -2009,13 +2839,47 @@ check_calls: continue; #endif ip = chain->ips[j]; - if (ip < PERF_CONTEXT_MAX) ++nr_entries; + else if (callchain_param.order != ORDER_CALLEE) { + err = find_prev_cpumode(chain, thread, cursor, parent, + root_al, &cpumode, j, symbols); + if (err) + return (err < 0) ? err : 0; + continue; + } + + /* + * PERF_CONTEXT_USER allows us to locate where the user stack ends. + * Depending on callchain_param.order and the position of PERF_CONTEXT_USER, + * the index will be different in order to add the missing frame + * at the right place. + */ + + usr_idx = callchain_param.order == ORDER_CALLEE ? j-2 : j-1; + + if (usr_idx >= 0 && chain->ips[usr_idx] == PERF_CONTEXT_USER) { + + leaf_frame_caller = get_leaf_frame_caller(sample, thread, usr_idx); + + /* + * check if leaf_frame_Caller != ip to not add the same + * value twice. + */ + + if (leaf_frame_caller && leaf_frame_caller != ip) { + + err = add_callchain_ip(thread, cursor, parent, + root_al, &cpumode, leaf_frame_caller, + false, NULL, NULL, 0, symbols); + if (err) + return (err < 0) ? err : 0; + } + } err = add_callchain_ip(thread, cursor, parent, root_al, &cpumode, ip, - false, NULL, 0, 0); + false, NULL, NULL, 0, symbols); if (err) return (err < 0) ? err : 0; @@ -2024,69 +2888,132 @@ check_calls: return 0; } +static int append_inlines(struct callchain_cursor *cursor, struct map_symbol *ms, u64 ip) +{ + struct symbol *sym = ms->sym; + struct map *map = ms->map; + struct inline_node *inline_node; + struct inline_list *ilist; + struct dso *dso; + u64 addr; + int ret = 1; + struct map_symbol ilist_ms; + + if (!symbol_conf.inline_name || !map || !sym) + return ret; + + addr = map__dso_map_ip(map, ip); + addr = map__rip_2objdump(map, addr); + dso = map__dso(map); + + inline_node = inlines__tree_find(dso__inlined_nodes(dso), addr); + if (!inline_node) { + inline_node = dso__parse_addr_inlines(dso, addr, sym); + if (!inline_node) + return ret; + inlines__tree_insert(dso__inlined_nodes(dso), inline_node); + } + + ilist_ms = (struct map_symbol) { + .maps = maps__get(ms->maps), + .map = map__get(map), + }; + list_for_each_entry(ilist, &inline_node->val, list) { + ilist_ms.sym = ilist->symbol; + ret = callchain_cursor_append(cursor, ip, &ilist_ms, false, + NULL, 0, 0, 0, ilist->srcline); + + if (ret != 0) + return ret; + } + map_symbol__exit(&ilist_ms); + + return ret; +} + static int unwind_entry(struct unwind_entry *entry, void *arg) { struct callchain_cursor *cursor = arg; + const char *srcline = NULL; + u64 addr = entry->ip; + + if (symbol_conf.hide_unresolved && entry->ms.sym == NULL) + return 0; - if (symbol_conf.hide_unresolved && entry->sym == NULL) + if (append_inlines(cursor, &entry->ms, entry->ip) == 0) return 0; - return callchain_cursor_append(cursor, entry->ip, - entry->map, entry->sym, - false, NULL, 0, 0); + + /* + * Convert entry->ip from a virtual address to an offset in + * its corresponding binary. + */ + if (entry->ms.map) + addr = map__dso_map_ip(entry->ms.map, entry->ip); + + srcline = callchain_srcline(&entry->ms, addr); + return callchain_cursor_append(cursor, entry->ip, &entry->ms, + false, NULL, 0, 0, 0, srcline); } static int thread__resolve_callchain_unwind(struct thread *thread, struct callchain_cursor *cursor, - struct perf_evsel *evsel, + struct evsel *evsel, struct perf_sample *sample, - int max_stack) + int max_stack, bool symbols) { /* Can we do dwarf post unwind? */ - if (!((evsel->attr.sample_type & PERF_SAMPLE_REGS_USER) && - (evsel->attr.sample_type & PERF_SAMPLE_STACK_USER))) + if (!((evsel->core.attr.sample_type & PERF_SAMPLE_REGS_USER) && + (evsel->core.attr.sample_type & PERF_SAMPLE_STACK_USER))) return 0; /* Bail out if nothing was captured. */ - if ((!sample->user_regs.regs) || - (!sample->user_stack.size)) + if (!sample->user_regs || !sample->user_regs->regs || + !sample->user_stack.size) return 0; + if (!symbols) + pr_debug("Not resolving symbols with an unwinder isn't currently supported\n"); + return unwind__get_entries(unwind_entry, cursor, - thread, sample, max_stack); + thread, sample, max_stack, false); } -int thread__resolve_callchain(struct thread *thread, - struct callchain_cursor *cursor, - struct perf_evsel *evsel, - struct perf_sample *sample, - struct symbol **parent, - struct addr_location *root_al, - int max_stack) +int __thread__resolve_callchain(struct thread *thread, + struct callchain_cursor *cursor, + struct evsel *evsel, + struct perf_sample *sample, + struct symbol **parent, + struct addr_location *root_al, + int max_stack, + bool symbols) { int ret = 0; - callchain_cursor_reset(&callchain_cursor); + if (cursor == NULL) + return -ENOMEM; + + callchain_cursor_reset(cursor); if (callchain_param.order == ORDER_CALLEE) { ret = thread__resolve_callchain_sample(thread, cursor, evsel, sample, parent, root_al, - max_stack); + max_stack, symbols); if (ret) return ret; ret = thread__resolve_callchain_unwind(thread, cursor, evsel, sample, - max_stack); + max_stack, symbols); } else { ret = thread__resolve_callchain_unwind(thread, cursor, evsel, sample, - max_stack); + max_stack, symbols); if (ret) return ret; ret = thread__resolve_callchain_sample(thread, cursor, evsel, sample, parent, root_al, - max_stack); + max_stack, symbols); } return ret; @@ -2096,23 +3023,7 @@ int machine__for_each_thread(struct machine *machine, int (*fn)(struct thread *thread, void *p), void *priv) { - struct rb_node *nd; - struct thread *thread; - int rc = 0; - - for (nd = rb_first(&machine->threads); nd; nd = rb_next(nd)) { - thread = rb_entry(nd, struct thread, rb_node); - rc = fn(thread, priv); - if (rc != 0) - return rc; - } - - list_for_each_entry(thread, &machine->dead_threads, node) { - rc = fn(thread, priv); - if (rc != 0) - return rc; - } - return rc; + return threads__for_each_thread(&machine->threads, fn, priv); } int machines__for_each_thread(struct machines *machines, @@ -2126,7 +3037,7 @@ int machines__for_each_thread(struct machines *machines, if (rc != 0) return rc; - for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) { + for (nd = rb_first_cached(&machines->guests); nd; nd = rb_next(nd)) { struct machine *machine = rb_entry(nd, struct machine, rb_node); rc = machine__for_each_thread(machine, fn, priv); @@ -2136,22 +3047,39 @@ int machines__for_each_thread(struct machines *machines, return rc; } -int __machine__synthesize_threads(struct machine *machine, struct perf_tool *tool, - struct target *target, struct thread_map *threads, - perf_event__handler_t process, bool data_mmap, - unsigned int proc_map_timeout) + +static int thread_list_cb(struct thread *thread, void *data) { - if (target__has_task(target)) - return perf_event__synthesize_thread_map(tool, threads, process, machine, data_mmap, proc_map_timeout); - else if (target__has_cpu(target)) - return perf_event__synthesize_threads(tool, process, machine, data_mmap, proc_map_timeout); - /* command specified */ + struct list_head *list = data; + struct thread_list *entry = malloc(sizeof(*entry)); + + if (!entry) + return -ENOMEM; + + entry->thread = thread__get(thread); + list_add_tail(&entry->list, list); return 0; } +int machine__thread_list(struct machine *machine, struct list_head *list) +{ + return machine__for_each_thread(machine, thread_list_cb, list); +} + +void thread_list__delete(struct list_head *list) +{ + struct thread_list *pos, *next; + + list_for_each_entry_safe(pos, next, list, list) { + thread__zput(pos->thread); + list_del(&pos->list); + free(pos); + } +} + pid_t machine__get_current_tid(struct machine *machine, int cpu) { - if (cpu < 0 || cpu >= MAX_NR_CPUS || !machine->current_tid) + if (cpu < 0 || (size_t)cpu >= machine->current_tid_sz) return -1; return machine->current_tid[cpu]; @@ -2161,25 +3089,16 @@ int machine__set_current_tid(struct machine *machine, int cpu, pid_t pid, pid_t tid) { struct thread *thread; + const pid_t init_val = -1; if (cpu < 0) return -EINVAL; - if (!machine->current_tid) { - int i; - - machine->current_tid = calloc(MAX_NR_CPUS, sizeof(pid_t)); - if (!machine->current_tid) - return -ENOMEM; - for (i = 0; i < MAX_NR_CPUS; i++) - machine->current_tid[i] = -1; - } - - if (cpu >= MAX_NR_CPUS) { - pr_err("Requested CPU %d too large. ", cpu); - pr_err("Consider raising MAX_NR_CPUS\n"); - return -EINVAL; - } + if (realloc_array_as_needed(machine->current_tid, + machine->current_tid_sz, + (unsigned int)cpu, + &init_val)) + return -ENOMEM; machine->current_tid[cpu] = tid; @@ -2187,12 +3106,31 @@ int machine__set_current_tid(struct machine *machine, int cpu, pid_t pid, if (!thread) return -ENOMEM; - thread->cpu = cpu; + thread__set_cpu(thread, cpu); thread__put(thread); return 0; } +/* + * Compares the raw arch string. N.B. see instead perf_env__arch() or + * machine__normalized_is() if a normalized arch is needed. + */ +bool machine__is(struct machine *machine, const char *arch) +{ + return machine && !strcmp(perf_env__raw_arch(machine->env), arch); +} + +bool machine__normalized_is(struct machine *machine, const char *arch) +{ + return machine && !strcmp(perf_env__arch(machine->env), arch); +} + +int machine__nr_cpus_avail(struct machine *machine) +{ + return machine ? perf_env__nr_cpus_avail(machine->env) : 0; +} + int machine__get_kernel_start(struct machine *machine) { struct map *map = machine__kernel_map(machine); @@ -2209,27 +3147,165 @@ int machine__get_kernel_start(struct machine *machine) machine->kernel_start = 1ULL << 63; if (map) { err = map__load(map); - if (!err) - machine->kernel_start = map->start; + /* + * On x86_64, PTI entry trampolines are less than the + * start of kernel text, but still above 2^63. So leave + * kernel_start = 1ULL << 63 for x86_64. + */ + if (!err && !machine__is(machine, "x86_64")) + machine->kernel_start = map__start(map); } return err; } +u8 machine__addr_cpumode(struct machine *machine, u8 cpumode, u64 addr) +{ + u8 addr_cpumode = cpumode; + bool kernel_ip; + + if (!machine->single_address_space) + goto out; + + kernel_ip = machine__kernel_ip(machine, addr); + switch (cpumode) { + case PERF_RECORD_MISC_KERNEL: + case PERF_RECORD_MISC_USER: + addr_cpumode = kernel_ip ? PERF_RECORD_MISC_KERNEL : + PERF_RECORD_MISC_USER; + break; + case PERF_RECORD_MISC_GUEST_KERNEL: + case PERF_RECORD_MISC_GUEST_USER: + addr_cpumode = kernel_ip ? PERF_RECORD_MISC_GUEST_KERNEL : + PERF_RECORD_MISC_GUEST_USER; + break; + default: + break; + } +out: + return addr_cpumode; +} + +struct dso *machine__findnew_dso_id(struct machine *machine, const char *filename, + const struct dso_id *id) +{ + return dsos__findnew_id(&machine->dsos, filename, id); +} + struct dso *machine__findnew_dso(struct machine *machine, const char *filename) { - return dsos__findnew(&machine->dsos, filename); + return machine__findnew_dso_id(machine, filename, &dso_id_empty); } char *machine__resolve_kernel_addr(void *vmachine, unsigned long long *addrp, char **modp) { struct machine *machine = vmachine; struct map *map; - struct symbol *sym = map_groups__find_symbol(&machine->kmaps, MAP__FUNCTION, *addrp, &map); + struct symbol *sym = machine__find_kernel_symbol(machine, *addrp, &map); if (sym == NULL) return NULL; - *modp = __map__is_kmodule(map) ? (char *)map->dso->short_name : NULL; - *addrp = map->unmap_ip(map, sym->start); + *modp = __map__is_kmodule(map) ? (char *)dso__short_name(map__dso(map)) : NULL; + *addrp = map__unmap_ip(map, sym->start); return sym->name; } + +struct machine__for_each_dso_cb_args { + struct machine *machine; + machine__dso_t fn; + void *priv; +}; + +static int machine__for_each_dso_cb(struct dso *dso, void *data) +{ + struct machine__for_each_dso_cb_args *args = data; + + return args->fn(dso, args->machine, args->priv); +} + +int machine__for_each_dso(struct machine *machine, machine__dso_t fn, void *priv) +{ + struct machine__for_each_dso_cb_args args = { + .machine = machine, + .fn = fn, + .priv = priv, + }; + + return dsos__for_each_dso(&machine->dsos, machine__for_each_dso_cb, &args); +} + +int machine__for_each_kernel_map(struct machine *machine, machine__map_t fn, void *priv) +{ + struct maps *maps = machine__kernel_maps(machine); + + return maps__for_each_map(maps, fn, priv); +} + +bool machine__is_lock_function(struct machine *machine, u64 addr) +{ + if (!machine->sched.text_start) { + struct map *kmap; + struct symbol *sym = machine__find_kernel_symbol_by_name(machine, "__sched_text_start", &kmap); + + if (!sym) { + /* to avoid retry */ + machine->sched.text_start = 1; + return false; + } + + machine->sched.text_start = map__unmap_ip(kmap, sym->start); + + /* should not fail from here */ + sym = machine__find_kernel_symbol_by_name(machine, "__sched_text_end", &kmap); + machine->sched.text_end = map__unmap_ip(kmap, sym->start); + + sym = machine__find_kernel_symbol_by_name(machine, "__lock_text_start", &kmap); + machine->lock.text_start = map__unmap_ip(kmap, sym->start); + + sym = machine__find_kernel_symbol_by_name(machine, "__lock_text_end", &kmap); + machine->lock.text_end = map__unmap_ip(kmap, sym->start); + + sym = machine__find_kernel_symbol_by_name(machine, "__traceiter_contention_begin", &kmap); + if (sym) { + machine->traceiter.text_start = map__unmap_ip(kmap, sym->start); + machine->traceiter.text_end = map__unmap_ip(kmap, sym->end); + } + sym = machine__find_kernel_symbol_by_name(machine, "trace_contention_begin", &kmap); + if (sym) { + machine->trace.text_start = map__unmap_ip(kmap, sym->start); + machine->trace.text_end = map__unmap_ip(kmap, sym->end); + } + } + + /* failed to get kernel symbols */ + if (machine->sched.text_start == 1) + return false; + + /* mutex and rwsem functions are in sched text section */ + if (machine->sched.text_start <= addr && addr < machine->sched.text_end) + return true; + + /* spinlock functions are in lock text section */ + if (machine->lock.text_start <= addr && addr < machine->lock.text_end) + return true; + + /* traceiter functions currently don't have their own section + * but we consider them lock functions + */ + if (machine->traceiter.text_start != 0) { + if (machine->traceiter.text_start <= addr && addr < machine->traceiter.text_end) + return true; + } + + if (machine->trace.text_start != 0) { + if (machine->trace.text_start <= addr && addr < machine->trace.text_end) + return true; + } + + return false; +} + +int machine__hit_all_dsos(struct machine *machine) +{ + return dsos__hit_all(&machine->dsos); +} |