From fa1531fdd7b6332aa61bcc9fda495583acba460d Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Mon, 10 Jun 2013 16:31:28 -0400 Subject: perf tools: Remove callchain_cursor_reset call Removing callchain_cursor_reset call as it is called in subsequent machine__resolve_callchain_sample function. Signed-off-by: Jiri Olsa Cc: Andi Kleen Cc: Corey Ashford Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-ic53wabwmmgvvwve2ymv3yf7@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index b2ecad6ec46..93527afb09d 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -1242,8 +1242,6 @@ int machine__resolve_callchain(struct machine *machine, { int ret; - callchain_cursor_reset(&callchain_cursor); - ret = machine__resolve_callchain_sample(machine, thread, sample->callchain, parent); if (ret) -- cgit v1.2.3-70-g09d2 From 380512345e13c3af64e59627f1b993c4faa94a84 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 4 Jul 2013 16:20:31 +0300 Subject: perf tools: struct thread has a tid not a pid As evident from 'machine__process_fork_event()' and 'machine__process_exit_event()' the 'pid' member of struct thread is actually the tid. Rename 'pid' to 'tid' in struct thread accordingly. Signed-off-by: Adrian Hunter Acked-by: David Ahern Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1372944040-32690-13-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/builtin-kmem.c | 2 +- tools/perf/builtin-sched.c | 12 ++++++------ tools/perf/builtin-trace.c | 4 ++-- tools/perf/ui/browsers/hists.c | 6 +++--- tools/perf/util/event.c | 2 +- tools/perf/util/machine.c | 20 ++++++++++---------- tools/perf/util/machine.h | 4 ++-- tools/perf/util/sort.c | 6 +++--- tools/perf/util/thread.c | 10 +++++----- tools/perf/util/thread.h | 4 ++-- 10 files changed, 35 insertions(+), 35 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c index 0259502638b..b49f5c58e15 100644 --- a/tools/perf/builtin-kmem.c +++ b/tools/perf/builtin-kmem.c @@ -313,7 +313,7 @@ static int process_sample_event(struct perf_tool *tool __maybe_unused, return -1; } - dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid); + dump_printf(" ... thread: %s:%d\n", thread->comm, thread->tid); if (evsel->handler.func != NULL) { tracepoint_handler f = evsel->handler.func; diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c index fba4a940ba3..948183adb6e 100644 --- a/tools/perf/builtin-sched.c +++ b/tools/perf/builtin-sched.c @@ -1075,7 +1075,7 @@ static int latency_migrate_task_event(struct perf_sched *sched, if (!atoms) { if (thread_atoms_insert(sched, migrant)) return -1; - register_pid(sched, migrant->pid, migrant->comm); + register_pid(sched, migrant->tid, migrant->comm); atoms = thread_atoms_search(&sched->atom_root, migrant, &sched->cmp_pid); if (!atoms) { pr_err("migration-event: Internal tree error"); @@ -1115,7 +1115,7 @@ static void output_lat_thread(struct perf_sched *sched, struct work_atoms *work_ sched->all_runtime += work_list->total_runtime; sched->all_count += work_list->nb_atoms; - ret = printf(" %s:%d ", work_list->thread->comm, work_list->thread->pid); + ret = printf(" %s:%d ", work_list->thread->comm, work_list->thread->tid); for (i = 0; i < 24 - ret; i++) printf(" "); @@ -1131,9 +1131,9 @@ static void output_lat_thread(struct perf_sched *sched, struct work_atoms *work_ static int pid_cmp(struct work_atoms *l, struct work_atoms *r) { - if (l->thread->pid < r->thread->pid) + if (l->thread->tid < r->thread->tid) return -1; - if (l->thread->pid > r->thread->pid) + if (l->thread->tid > r->thread->tid) return 1; return 0; @@ -1321,7 +1321,7 @@ static int map_switch_event(struct perf_sched *sched, struct perf_evsel *evsel, printf("*"); if (sched->curr_thread[cpu]) { - if (sched->curr_thread[cpu]->pid) + if (sched->curr_thread[cpu]->tid) printf("%2s ", sched->curr_thread[cpu]->shortname); else printf(". "); @@ -1332,7 +1332,7 @@ static int map_switch_event(struct perf_sched *sched, struct perf_evsel *evsel, printf(" %12.6f secs ", (double)timestamp/1e9); if (new_shortname) { printf("%s => %s:%d\n", - sched_in->shortname, sched_in->comm, sched_in->pid); + sched_in->shortname, sched_in->comm, sched_in->tid); } else { printf("\n"); } diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index 87fc7d08ca0..0e4b67f6bbd 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -142,7 +142,7 @@ static size_t trace__fprintf_entry_head(struct trace *trace, struct thread *thre printed += fprintf_duration(duration, fp); if (trace->multiple_threads) - printed += fprintf(fp, "%d ", thread->pid); + printed += fprintf(fp, "%d ", thread->tid); return printed; } @@ -593,7 +593,7 @@ static size_t trace__fprintf_thread_summary(struct trace *trace, FILE *fp) color = PERF_COLOR_YELLOW; printed += color_fprintf(fp, color, "%20s", thread->comm); - printed += fprintf(fp, " - %-5d :%11lu [", thread->pid, ttrace->nr_events); + printed += fprintf(fp, " - %-5d :%11lu [", thread->tid, ttrace->nr_events); printed += color_fprintf(fp, color, "%5.1f%%", ratio); printed += fprintf(fp, " ] %10.3f ms\n", ttrace->runtime_ms); } diff --git a/tools/perf/ui/browsers/hists.c b/tools/perf/ui/browsers/hists.c index fc0bd3843d3..06e892f1f8c 100644 --- a/tools/perf/ui/browsers/hists.c +++ b/tools/perf/ui/browsers/hists.c @@ -1256,7 +1256,7 @@ static int hists__browser_title(struct hists *hists, char *bf, size_t size, printed += scnprintf(bf + printed, size - printed, ", Thread: %s(%d)", (thread->comm_set ? thread->comm : ""), - thread->pid); + thread->tid); if (dso) printed += scnprintf(bf + printed, size - printed, ", DSO: %s", dso->short_name); @@ -1579,7 +1579,7 @@ static int perf_evsel__hists_browse(struct perf_evsel *evsel, int nr_events, asprintf(&options[nr_options], "Zoom %s %s(%d) thread", (browser->hists->thread_filter ? "out of" : "into"), (thread->comm_set ? thread->comm : ""), - thread->pid) > 0) + thread->tid) > 0) zoom_thread = nr_options++; if (dso != NULL && @@ -1702,7 +1702,7 @@ zoom_out_thread: } else { ui_helpline__fpush("To zoom out press <- or -> + \"Zoom out of %s(%d) thread\"", thread->comm_set ? thread->comm : "", - thread->pid); + thread->tid); browser->hists->thread_filter = thread; sort_thread.elide = true; pstack__push(fstack, &browser->hists->thread_filter); diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index 5cd13d768ce..95412705d0d 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -686,7 +686,7 @@ int perf_event__preprocess_sample(const union perf_event *event, !strlist__has_entry(symbol_conf.comm_list, thread->comm)) goto out_filtered; - dump_printf(" ... thread: %s:%d\n", thread->comm, thread->pid); + dump_printf(" ... thread: %s:%d\n", thread->comm, thread->tid); /* * Have we already created the kernel maps for this machine? * diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 93527afb09d..5dd5026a82e 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -233,7 +233,7 @@ void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size) return; } -static struct thread *__machine__findnew_thread(struct machine *machine, pid_t pid, +static struct thread *__machine__findnew_thread(struct machine *machine, pid_t tid, bool create) { struct rb_node **p = &machine->threads.rb_node; @@ -241,23 +241,23 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t p struct thread *th; /* - * Font-end cache - PID lookups come in blocks, + * Front-end cache - TID lookups come in blocks, * so most of the time we dont have to look up * the full rbtree: */ - if (machine->last_match && machine->last_match->pid == pid) + if (machine->last_match && machine->last_match->tid == tid) return machine->last_match; while (*p != NULL) { parent = *p; th = rb_entry(parent, struct thread, rb_node); - if (th->pid == pid) { + if (th->tid == tid) { machine->last_match = th; return th; } - if (pid < th->pid) + if (tid < th->tid) p = &(*p)->rb_left; else p = &(*p)->rb_right; @@ -266,7 +266,7 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t p if (!create) return NULL; - th = thread__new(pid); + th = thread__new(tid); if (th != NULL) { rb_link_node(&th->rb_node, parent, p); rb_insert_color(&th->rb_node, &machine->threads); @@ -276,14 +276,14 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t p return th; } -struct thread *machine__findnew_thread(struct machine *machine, pid_t pid) +struct thread *machine__findnew_thread(struct machine *machine, pid_t tid) { - return __machine__findnew_thread(machine, pid, true); + return __machine__findnew_thread(machine, tid, true); } -struct thread *machine__find_thread(struct machine *machine, pid_t pid) +struct thread *machine__find_thread(struct machine *machine, pid_t tid) { - return __machine__findnew_thread(machine, pid, false); + return __machine__findnew_thread(machine, tid, false); } int machine__process_comm_event(struct machine *machine, union perf_event *event) diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h index 77940680f1f..e49ba01b793 100644 --- a/tools/perf/util/machine.h +++ b/tools/perf/util/machine.h @@ -36,7 +36,7 @@ struct map *machine__kernel_map(struct machine *machine, enum map_type type) return machine->vmlinux_maps[type]; } -struct thread *machine__find_thread(struct machine *machine, pid_t pid); +struct thread *machine__find_thread(struct machine *machine, pid_t tid); int machine__process_comm_event(struct machine *machine, union perf_event *event); int machine__process_exit_event(struct machine *machine, union perf_event *event); @@ -99,7 +99,7 @@ static inline bool machine__is_host(struct machine *machine) return machine ? machine->pid == HOST_KERNEL_ID : false; } -struct thread *machine__findnew_thread(struct machine *machine, pid_t pid); +struct thread *machine__findnew_thread(struct machine *machine, pid_t tid); size_t machine__fprintf(struct machine *machine, FILE *fp); diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c index 313a5a73011..8deee19d2e7 100644 --- a/tools/perf/util/sort.c +++ b/tools/perf/util/sort.c @@ -55,14 +55,14 @@ static int64_t cmp_null(void *l, void *r) static int64_t sort__thread_cmp(struct hist_entry *left, struct hist_entry *right) { - return right->thread->pid - left->thread->pid; + return right->thread->tid - left->thread->tid; } static int hist_entry__thread_snprintf(struct hist_entry *self, char *bf, size_t size, unsigned int width) { return repsep_snprintf(bf, size, "%*s:%5d", width - 6, - self->thread->comm ?: "", self->thread->pid); + self->thread->comm ?: "", self->thread->tid); } struct sort_entry sort_thread = { @@ -77,7 +77,7 @@ struct sort_entry sort_thread = { static int64_t sort__comm_cmp(struct hist_entry *left, struct hist_entry *right) { - return right->thread->pid - left->thread->pid; + return right->thread->tid - left->thread->tid; } static int64_t diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c index 40399cbcca7..6feeb88eb5b 100644 --- a/tools/perf/util/thread.c +++ b/tools/perf/util/thread.c @@ -7,17 +7,17 @@ #include "util.h" #include "debug.h" -struct thread *thread__new(pid_t pid) +struct thread *thread__new(pid_t tid) { struct thread *self = zalloc(sizeof(*self)); if (self != NULL) { map_groups__init(&self->mg); - self->pid = pid; + self->tid = tid; self->ppid = -1; self->comm = malloc(32); if (self->comm) - snprintf(self->comm, 32, ":%d", self->pid); + snprintf(self->comm, 32, ":%d", self->tid); } return self; @@ -57,7 +57,7 @@ int thread__comm_len(struct thread *self) size_t thread__fprintf(struct thread *thread, FILE *fp) { - return fprintf(fp, "Thread %d %s\n", thread->pid, thread->comm) + + return fprintf(fp, "Thread %d %s\n", thread->tid, thread->comm) + map_groups__fprintf(&thread->mg, verbose, fp); } @@ -84,7 +84,7 @@ int thread__fork(struct thread *self, struct thread *parent) if (map_groups__clone(&self->mg, &parent->mg, i) < 0) return -ENOMEM; - self->ppid = parent->pid; + self->ppid = parent->tid; return 0; } diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h index 5e7ba35a551..0fe1f9c0586 100644 --- a/tools/perf/util/thread.h +++ b/tools/perf/util/thread.h @@ -12,7 +12,7 @@ struct thread { struct list_head node; }; struct map_groups mg; - pid_t pid; + pid_t tid; pid_t ppid; char shortname[3]; bool comm_set; @@ -24,7 +24,7 @@ struct thread { struct machine; -struct thread *thread__new(pid_t pid); +struct thread *thread__new(pid_t tid); void thread__delete(struct thread *self); int thread__set_comm(struct thread *self, const char *comm); -- cgit v1.2.3-70-g09d2 From b21484f1a1f300d422cfe5d4f8f50015e22cea24 Mon Sep 17 00:00:00 2001 From: Greg Price Date: Thu, 6 Dec 2012 21:48:05 -0800 Subject: perf report/top: Add option to collapse undesired parts of call graph For example, in an application with an expensive function implemented with deeply nested recursive calls, the default call-graph presentation is dominated by the different callchains within that function. By ignoring these callees, we can collect the callchains leading into the function and compactly identify what to blame for expensive calls. For example, in this report the callers of garbage_collect() are scattered across the tree: $ perf report -d ruby 2>- | grep -m10 ^[^#]*[a-z] 22.03% ruby [.] gc_mark --- gc_mark |--59.40%-- mark_keyvalue | st_foreach | gc_mark_children | |--99.75%-- rb_gc_mark | | rb_vm_mark | | gc_mark_children | | gc_marks | | |--99.00%-- garbage_collect If we ignore the callees of garbage_collect(), its callers are coalesced: $ perf report --ignore-callees garbage_collect -d ruby 2>- | grep -m10 ^[^#]*[a-z] 72.92% ruby [.] garbage_collect --- garbage_collect vm_xmalloc |--47.08%-- ruby_xmalloc | st_insert2 | rb_hash_aset | |--98.45%-- features_index_add | | rb_provide_feature | | rb_require_safe | | vm_call_method Signed-off-by: Greg Price Tested-by: Jiri Olsa Cc: David Ahern Cc: Ingo Molnar Cc: Jiri Olsa Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20130623031720.GW22203@biohazard-cafe.mit.edu Link: http://lkml.kernel.org/r/20130708115746.GO22203@biohazard-cafe.mit.edu Cc: Fengguang Wu [ remove spaces at beginning of line, reported by Fengguang Wu ] Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/Documentation/perf-report.txt | 5 +++++ tools/perf/Documentation/perf-top.txt | 5 +++++ tools/perf/builtin-report.c | 27 ++++++++++++++++++++++++--- tools/perf/builtin-top.c | 6 ++++-- tools/perf/util/machine.c | 24 +++++++++++++++--------- tools/perf/util/machine.h | 4 +++- tools/perf/util/session.c | 3 +-- tools/perf/util/sort.c | 2 ++ tools/perf/util/sort.h | 4 ++++ 9 files changed, 63 insertions(+), 17 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt index 66dab7410c1..747ff50284b 100644 --- a/tools/perf/Documentation/perf-report.txt +++ b/tools/perf/Documentation/perf-report.txt @@ -135,6 +135,11 @@ OPTIONS --inverted:: alias for inverted caller based call graph. +--ignore-callees=:: + Ignore callees of the function(s) matching the given regex. + This has the effect of collecting the callers of each such + function into one place in the call-graph tree. + --pretty=:: Pretty printing style. key: normal, raw diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt index 7fdd1909e37..58d6598a968 100644 --- a/tools/perf/Documentation/perf-top.txt +++ b/tools/perf/Documentation/perf-top.txt @@ -155,6 +155,11 @@ Default is to monitor all CPUS. Default: fractal,0.5,callee. +--ignore-callees=:: + Ignore callees of the function(s) matching the given regex. + This has the effect of collecting the callers of each such + function into one place in the call-graph tree. + --percent-limit:: Do not show entries which have an overhead under that percent. (Default: 0). diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index ee2ca3eb22d..9a7e54d701c 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -89,7 +89,7 @@ static int perf_report__add_mem_hist_entry(struct perf_tool *tool, if ((sort__has_parent || symbol_conf.use_callchain) && sample->callchain) { err = machine__resolve_callchain(machine, evsel, al->thread, - sample, &parent); + sample, &parent, al); if (err) return err; } @@ -180,7 +180,7 @@ static int perf_report__add_branch_hist_entry(struct perf_tool *tool, if ((sort__has_parent || symbol_conf.use_callchain) && sample->callchain) { err = machine__resolve_callchain(machine, evsel, al->thread, - sample, &parent); + sample, &parent, al); if (err) return err; } @@ -254,7 +254,7 @@ static int perf_evsel__add_hist_entry(struct perf_evsel *evsel, if ((sort__has_parent || symbol_conf.use_callchain) && sample->callchain) { err = machine__resolve_callchain(machine, evsel, al->thread, - sample, &parent); + sample, &parent, al); if (err) return err; } @@ -681,6 +681,24 @@ setup: return 0; } +int +report_parse_ignore_callees_opt(const struct option *opt __maybe_unused, + const char *arg, int unset __maybe_unused) +{ + if (arg) { + int err = regcomp(&ignore_callees_regex, arg, REG_EXTENDED); + if (err) { + char buf[BUFSIZ]; + regerror(err, &ignore_callees_regex, buf, sizeof(buf)); + pr_err("Invalid --ignore-callees regex: %s\n%s", arg, buf); + return -1; + } + have_ignore_callees = 1; + } + + return 0; +} + static int parse_branch_mode(const struct option *opt __maybe_unused, const char *str __maybe_unused, int unset) @@ -771,6 +789,9 @@ int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused) "Default: fractal,0.5,callee", &parse_callchain_opt, callchain_default_opt), OPT_BOOLEAN('G', "inverted", &report.inverted_callchain, "alias for inverted call graph"), + OPT_CALLBACK(0, "ignore-callees", NULL, "regex", + "ignore callees of these functions in call graphs", + report_parse_ignore_callees_opt), OPT_STRING('d', "dsos", &symbol_conf.dso_list_str, "dso[,dso...]", "only consider symbols in these dsos"), OPT_STRING('c', "comms", &symbol_conf.comm_list_str, "comm[,comm...]", diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c index a237059f51c..bbf46357277 100644 --- a/tools/perf/builtin-top.c +++ b/tools/perf/builtin-top.c @@ -773,8 +773,7 @@ static void perf_event__process_sample(struct perf_tool *tool, sample->callchain) { err = machine__resolve_callchain(machine, evsel, al.thread, sample, - &parent); - + &parent, &al); if (err) return; } @@ -1109,6 +1108,9 @@ int cmd_top(int argc, const char **argv, const char *prefix __maybe_unused) OPT_CALLBACK_DEFAULT('G', "call-graph", &top.record_opts, "mode[,dump_size]", record_callchain_help, &parse_callchain_opt, "fp"), + OPT_CALLBACK(0, "ignore-callees", NULL, "regex", + "ignore callees of these functions in call graphs", + report_parse_ignore_callees_opt), OPT_BOOLEAN(0, "show-total-period", &symbol_conf.show_total_period, "Show a column with the sum of periods"), OPT_STRING(0, "dsos", &symbol_conf.dso_list_str, "dso[,dso...]", diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 5dd5026a82e..f9f9d6381b9 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -1058,11 +1058,10 @@ int machine__process_event(struct machine *machine, union perf_event *event) return ret; } -static bool symbol__match_parent_regex(struct symbol *sym) +static bool symbol__match_regex(struct symbol *sym, regex_t *regex) { - if (sym->name && !regexec(&parent_regex, sym->name, 0, NULL, 0)) + if (sym->name && !regexec(regex, sym->name, 0, NULL, 0)) return 1; - return 0; } @@ -1159,8 +1158,8 @@ struct branch_info *machine__resolve_bstack(struct machine *machine, static int machine__resolve_callchain_sample(struct machine *machine, struct thread *thread, struct ip_callchain *chain, - struct symbol **parent) - + struct symbol **parent, + struct addr_location *root_al) { u8 cpumode = PERF_RECORD_MISC_USER; unsigned int i; @@ -1211,8 +1210,15 @@ static int machine__resolve_callchain_sample(struct machine *machine, MAP__FUNCTION, ip, &al, NULL); if (al.sym != NULL) { if (sort__has_parent && !*parent && - symbol__match_parent_regex(al.sym)) + symbol__match_regex(al.sym, &parent_regex)) *parent = al.sym; + else if (have_ignore_callees && root_al && + symbol__match_regex(al.sym, &ignore_callees_regex)) { + /* Treat this symbol as the root, + forgetting its callees. */ + *root_al = al; + callchain_cursor_reset(&callchain_cursor); + } if (!symbol_conf.use_callchain) break; } @@ -1237,13 +1243,13 @@ int machine__resolve_callchain(struct machine *machine, struct perf_evsel *evsel, struct thread *thread, struct perf_sample *sample, - struct symbol **parent) - + struct symbol **parent, + struct addr_location *root_al) { int ret; ret = machine__resolve_callchain_sample(machine, thread, - sample->callchain, parent); + sample->callchain, parent, root_al); if (ret) return ret; diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h index e49ba01b793..5bb6244194d 100644 --- a/tools/perf/util/machine.h +++ b/tools/perf/util/machine.h @@ -5,6 +5,7 @@ #include #include "map.h" +struct addr_location; struct branch_stack; struct perf_evsel; struct perf_sample; @@ -83,7 +84,8 @@ int machine__resolve_callchain(struct machine *machine, struct perf_evsel *evsel, struct thread *thread, struct perf_sample *sample, - struct symbol **parent); + struct symbol **parent, + struct addr_location *root_al); /* * Default guest kernel is defined by parameter --guestkallsyms diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index 951a1cfb317..1eb58eedcac 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -1406,9 +1406,8 @@ void perf_evsel__print_ip(struct perf_evsel *evsel, union perf_event *event, if (symbol_conf.use_callchain && sample->callchain) { - if (machine__resolve_callchain(machine, evsel, al.thread, - sample, NULL) != 0) { + sample, NULL, NULL) != 0) { if (verbose) error("Failed to resolve callchain. Skipping\n"); return; diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c index 8deee19d2e7..cb2b108635e 100644 --- a/tools/perf/util/sort.c +++ b/tools/perf/util/sort.c @@ -7,6 +7,8 @@ const char default_parent_pattern[] = "^sys_|^do_page_fault"; const char *parent_pattern = default_parent_pattern; const char default_sort_order[] = "comm,dso,symbol"; const char *sort_order = default_sort_order; +regex_t ignore_callees_regex; +int have_ignore_callees = 0; int sort__need_collapse = 0; int sort__has_parent = 0; int sort__has_sym = 0; diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h index 45ac84c1e03..a4a6d0b1ea0 100644 --- a/tools/perf/util/sort.h +++ b/tools/perf/util/sort.h @@ -29,6 +29,8 @@ extern const char *sort_order; extern const char default_parent_pattern[]; extern const char *parent_pattern; extern const char default_sort_order[]; +extern regex_t ignore_callees_regex; +extern int have_ignore_callees; extern int sort__need_collapse; extern int sort__has_parent; extern int sort__has_sym; @@ -183,4 +185,6 @@ int setup_sorting(void); extern int sort_dimension__add(const char *); void sort__setup_elide(FILE *fp); +int report_parse_ignore_callees_opt(const struct option *opt, const char *arg, int unset); + #endif /* __PERF_SORT_H */ -- cgit v1.2.3-70-g09d2 From 39b12f7812710e9a5896805d96812b3ede7491e8 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 7 Aug 2013 14:38:47 +0300 Subject: perf tools: Make it possible to read object code from vmlinux The new "object code reading" test shows that it is not possible to read object code from vmlinux. That is because the mappings do not map to the dso. This patch fixes that. A side-effect of changing the kernel map is that the "reloc" offset must be taken into account. As a result of that separate map functions for relocation are no longer needed. Also fixing up the maps to match the symbols no longer makes sense and so is not done. The vmlinux dso data_type is now set to either DSO_BINARY_TYPE__VMLINUX or DSO_BINARY_TYPE__GUEST_VMLINUX as approprite, which enables the correct file name to be determined by dso__binary_type_file(). This patch breaks the "vmlinux symtab matches kallsyms" test. That is fixed in a following patch. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1375875537-4509-4-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/dso.c | 4 +- tools/perf/util/dso.h | 8 ++++ tools/perf/util/machine.c | 4 +- tools/perf/util/map.c | 35 --------------- tools/perf/util/symbol-elf.c | 100 +++++++++++++++++++++++++++++++++++++++---- tools/perf/util/symbol.c | 22 +++++----- 6 files changed, 112 insertions(+), 61 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/dso.c b/tools/perf/util/dso.c index c4374f07603..121583dbf34 100644 --- a/tools/perf/util/dso.c +++ b/tools/perf/util/dso.c @@ -78,6 +78,8 @@ int dso__binary_type_file(struct dso *dso, enum dso_binary_type type, symbol_conf.symfs, build_id_hex, build_id_hex + 2); break; + case DSO_BINARY_TYPE__VMLINUX: + case DSO_BINARY_TYPE__GUEST_VMLINUX: case DSO_BINARY_TYPE__SYSTEM_PATH_DSO: snprintf(file, size, "%s%s", symbol_conf.symfs, dso->long_name); @@ -95,9 +97,7 @@ int dso__binary_type_file(struct dso *dso, enum dso_binary_type type, default: case DSO_BINARY_TYPE__KALLSYMS: - case DSO_BINARY_TYPE__VMLINUX: case DSO_BINARY_TYPE__GUEST_KALLSYMS: - case DSO_BINARY_TYPE__GUEST_VMLINUX: case DSO_BINARY_TYPE__JAVA_JIT: case DSO_BINARY_TYPE__NOT_FOUND: ret = -1; diff --git a/tools/perf/util/dso.h b/tools/perf/util/dso.h index d51aaf272c6..02aadaf45c6 100644 --- a/tools/perf/util/dso.h +++ b/tools/perf/util/dso.h @@ -3,6 +3,7 @@ #include #include +#include #include "types.h" #include "map.h" @@ -146,4 +147,11 @@ size_t dso__fprintf_buildid(struct dso *dso, FILE *fp); size_t dso__fprintf_symbols_by_name(struct dso *dso, enum map_type type, FILE *fp); size_t dso__fprintf(struct dso *dso, enum map_type type, FILE *fp); + +static inline bool dso__is_vmlinux(struct dso *dso) +{ + return dso->data_type == DSO_BINARY_TYPE__VMLINUX || + dso->data_type == DSO_BINARY_TYPE__GUEST_VMLINUX; +} + #endif /* __PERF_DSO */ diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index f9f9d6381b9..dc35dcffbd0 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -628,10 +628,8 @@ int machine__load_vmlinux_path(struct machine *machine, enum map_type type, struct map *map = machine->vmlinux_maps[type]; int ret = dso__load_vmlinux_path(map->dso, map, filter); - if (ret > 0) { + if (ret > 0) dso__set_loaded(map->dso, type); - map__reloc_vmlinux(map); - } return ret; } diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c index 8bcdf9e5408..5f662a3a016 100644 --- a/tools/perf/util/map.c +++ b/tools/perf/util/map.c @@ -182,12 +182,6 @@ int map__load(struct map *map, symbol_filter_t filter) #endif return -1; } - /* - * Only applies to the kernel, as its symtabs aren't relative like the - * module ones. - */ - if (map->dso->kernel) - map__reloc_vmlinux(map); return 0; } @@ -513,35 +507,6 @@ int map_groups__clone(struct map_groups *mg, return 0; } -static u64 map__reloc_map_ip(struct map *map, u64 ip) -{ - return ip + (s64)map->pgoff; -} - -static u64 map__reloc_unmap_ip(struct map *map, u64 ip) -{ - return ip - (s64)map->pgoff; -} - -void map__reloc_vmlinux(struct map *map) -{ - struct kmap *kmap = map__kmap(map); - s64 reloc; - - if (!kmap->ref_reloc_sym || !kmap->ref_reloc_sym->unrelocated_addr) - return; - - reloc = (kmap->ref_reloc_sym->unrelocated_addr - - kmap->ref_reloc_sym->addr); - - if (!reloc) - return; - - map->map_ip = map__reloc_map_ip; - map->unmap_ip = map__reloc_unmap_ip; - map->pgoff = reloc; -} - void maps__insert(struct rb_root *maps, struct map *map) { struct rb_node **p = &maps->rb_node; diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c index 4b12bf85032..ed6f443a3e1 100644 --- a/tools/perf/util/symbol-elf.c +++ b/tools/perf/util/symbol-elf.c @@ -603,7 +603,7 @@ int symsrc__init(struct symsrc *ss, struct dso *dso, const char *name, ".gnu.prelink_undo", NULL) != NULL); } else { - ss->adjust_symbols = 0; + ss->adjust_symbols = ehdr.e_type == ET_EXEC; } ss->name = strdup(name); @@ -624,6 +624,37 @@ out_close: return err; } +/** + * ref_reloc_sym_not_found - has kernel relocation symbol been found. + * @kmap: kernel maps and relocation reference symbol + * + * This function returns %true if we are dealing with the kernel maps and the + * relocation reference symbol has not yet been found. Otherwise %false is + * returned. + */ +static bool ref_reloc_sym_not_found(struct kmap *kmap) +{ + return kmap && kmap->ref_reloc_sym && kmap->ref_reloc_sym->name && + !kmap->ref_reloc_sym->unrelocated_addr; +} + +/** + * ref_reloc - kernel relocation offset. + * @kmap: kernel maps and relocation reference symbol + * + * This function returns the offset of kernel addresses as determined by using + * the relocation reference symbol i.e. if the kernel has not been relocated + * then the return value is zero. + */ +static u64 ref_reloc(struct kmap *kmap) +{ + if (kmap && kmap->ref_reloc_sym && + kmap->ref_reloc_sym->unrelocated_addr) + return kmap->ref_reloc_sym->addr - + kmap->ref_reloc_sym->unrelocated_addr; + return 0; +} + int dso__load_sym(struct dso *dso, struct map *map, struct symsrc *syms_ss, struct symsrc *runtime_ss, symbol_filter_t filter, int kmodule) @@ -642,6 +673,7 @@ int dso__load_sym(struct dso *dso, struct map *map, Elf_Scn *sec, *sec_strndx; Elf *elf; int nr = 0; + bool remap_kernel = false, adjust_kernel_syms = false; dso->symtab_type = syms_ss->type; @@ -681,7 +713,31 @@ int dso__load_sym(struct dso *dso, struct map *map, nr_syms = shdr.sh_size / shdr.sh_entsize; memset(&sym, 0, sizeof(sym)); - dso->adjust_symbols = runtime_ss->adjust_symbols; + + /* + * The kernel relocation symbol is needed in advance in order to adjust + * kernel maps correctly. + */ + if (ref_reloc_sym_not_found(kmap)) { + elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) { + const char *elf_name = elf_sym__name(&sym, symstrs); + + if (strcmp(elf_name, kmap->ref_reloc_sym->name)) + continue; + kmap->ref_reloc_sym->unrelocated_addr = sym.st_value; + break; + } + } + + dso->adjust_symbols = runtime_ss->adjust_symbols || ref_reloc(kmap); + /* + * Initial kernel and module mappings do not map to the dso. For + * function mappings, flag the fixups. + */ + if (map->type == MAP__FUNCTION && (dso->kernel || kmodule)) { + remap_kernel = true; + adjust_kernel_syms = dso->adjust_symbols; + } elf_symtab__for_each_symbol(syms, nr_syms, idx, sym) { struct symbol *f; const char *elf_name = elf_sym__name(&sym, symstrs); @@ -690,10 +746,6 @@ int dso__load_sym(struct dso *dso, struct map *map, const char *section_name; bool used_opd = false; - if (kmap && kmap->ref_reloc_sym && kmap->ref_reloc_sym->name && - strcmp(elf_name, kmap->ref_reloc_sym->name) == 0) - kmap->ref_reloc_sym->unrelocated_addr = sym.st_value; - if (!is_label && !elf_sym__is_a(&sym, map->type)) continue; @@ -745,15 +797,37 @@ int dso__load_sym(struct dso *dso, struct map *map, (sym.st_value & 1)) --sym.st_value; - if (dso->kernel != DSO_TYPE_USER || kmodule) { + if (dso->kernel || kmodule) { char dso_name[PATH_MAX]; + /* Adjust symbol to map to file offset */ + if (adjust_kernel_syms) + sym.st_value -= shdr.sh_addr - shdr.sh_offset; + if (strcmp(section_name, (curr_dso->short_name + dso->short_name_len)) == 0) goto new_symbol; if (strcmp(section_name, ".text") == 0) { + /* + * The initial kernel mapping is based on + * kallsyms and identity maps. Overwrite it to + * map to the kernel dso. + */ + if (remap_kernel && dso->kernel) { + remap_kernel = false; + map->start = shdr.sh_addr + + ref_reloc(kmap); + map->end = map->start + shdr.sh_size; + map->pgoff = shdr.sh_offset; + map->map_ip = map__map_ip; + map->unmap_ip = map__unmap_ip; + /* Ensure maps are correctly ordered */ + map_groups__remove(kmap->kmaps, map); + map_groups__insert(kmap->kmaps, map); + } + curr_map = map; curr_dso = dso; goto new_symbol; @@ -781,8 +855,16 @@ int dso__load_sym(struct dso *dso, struct map *map, dso__delete(curr_dso); goto out_elf_end; } - curr_map->map_ip = identity__map_ip; - curr_map->unmap_ip = identity__map_ip; + if (adjust_kernel_syms) { + curr_map->start = shdr.sh_addr + + ref_reloc(kmap); + curr_map->end = curr_map->start + + shdr.sh_size; + curr_map->pgoff = shdr.sh_offset; + } else { + curr_map->map_ip = identity__map_ip; + curr_map->unmap_ip = identity__map_ip; + } curr_dso->symtab_type = dso->symtab_type; map_groups__insert(kmap->kmaps, curr_map); dsos__add(&dso->node, curr_dso); diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index ea62ecd191f..04300dd5221 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -917,6 +917,10 @@ int dso__load_vmlinux(struct dso *dso, struct map *map, symsrc__destroy(&ss); if (err > 0) { + if (dso->kernel == DSO_TYPE_GUEST_KERNEL) + dso->data_type = DSO_BINARY_TYPE__GUEST_VMLINUX; + else + dso->data_type = DSO_BINARY_TYPE__VMLINUX; dso__set_long_name(dso, (char *)vmlinux); dso__set_loaded(dso, map->type); pr_debug("Using %s for symbols\n", symfs_vmlinux); @@ -989,7 +993,7 @@ static int dso__load_kernel_sym(struct dso *dso, struct map *map, dso__set_long_name(dso, strdup(symbol_conf.vmlinux_name)); dso->lname_alloc = 1; - goto out_fixup; + return err; } return err; } @@ -997,7 +1001,7 @@ static int dso__load_kernel_sym(struct dso *dso, struct map *map, if (vmlinux_path != NULL) { err = dso__load_vmlinux_path(dso, map, filter); if (err > 0) - goto out_fixup; + return err; } /* do not try local files if a symfs was given */ @@ -1058,7 +1062,6 @@ do_kallsyms: if (err > 0) { dso__set_long_name(dso, strdup("[kernel.kallsyms]")); -out_fixup: map__fixup_start(map); map__fixup_end(map); } @@ -1089,7 +1092,7 @@ static int dso__load_guest_kernel_sym(struct dso *dso, struct map *map, if (symbol_conf.default_guest_vmlinux_name != NULL) { err = dso__load_vmlinux(dso, map, symbol_conf.default_guest_vmlinux_name, filter); - goto out_try_fixup; + return err; } kallsyms_filename = symbol_conf.default_guest_kallsyms; @@ -1101,15 +1104,10 @@ static int dso__load_guest_kernel_sym(struct dso *dso, struct map *map, } err = dso__load_kallsyms(dso, kallsyms_filename, map, filter); - if (err > 0) - pr_debug("Using %s for symbols\n", kallsyms_filename); - -out_try_fixup: if (err > 0) { - if (kallsyms_filename != NULL) { - machine__mmap_name(machine, path, sizeof(path)); - dso__set_long_name(dso, strdup(path)); - } + pr_debug("Using %s for symbols\n", kallsyms_filename); + machine__mmap_name(machine, path, sizeof(path)); + dso__set_long_name(dso, strdup(path)); map__fixup_start(map); map__fixup_end(map); } -- cgit v1.2.3-70-g09d2 From 8e0cf965f95edd41df11cca50b92b4cb6ea8d80a Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Wed, 7 Aug 2013 14:38:51 +0300 Subject: perf symbols: Add support for reading from /proc/kcore In the absence of vmlinux, perf tools uses kallsyms for symbols. If the user has access, now also map to /proc/kcore. The dso data_type is now set to either DSO_BINARY_TYPE__KCORE or DSO_BINARY_TYPE__GUEST_KCORE as approprite. This patch breaks the "vmlinux symtab matches kallsyms" test. That is fixed in a following patch. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1375875537-4509-8-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/dso.c | 5 + tools/perf/util/dso.h | 8 ++ tools/perf/util/machine.c | 16 +++ tools/perf/util/map.c | 18 ++++ tools/perf/util/map.h | 13 +++ tools/perf/util/symbol-elf.c | 51 +++++++++ tools/perf/util/symbol-minimal.c | 7 ++ tools/perf/util/symbol.c | 227 ++++++++++++++++++++++++++++++++++++++- tools/perf/util/symbol.h | 5 + 9 files changed, 347 insertions(+), 3 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/dso.c b/tools/perf/util/dso.c index 1955804e82f..e3c1ff8512c 100644 --- a/tools/perf/util/dso.c +++ b/tools/perf/util/dso.c @@ -95,6 +95,11 @@ int dso__binary_type_file(struct dso *dso, enum dso_binary_type type, dso->long_name); break; + case DSO_BINARY_TYPE__KCORE: + case DSO_BINARY_TYPE__GUEST_KCORE: + snprintf(file, size, "%s", dso->long_name); + break; + default: case DSO_BINARY_TYPE__KALLSYMS: case DSO_BINARY_TYPE__GUEST_KALLSYMS: diff --git a/tools/perf/util/dso.h b/tools/perf/util/dso.h index 735a83751b1..b793053335d 100644 --- a/tools/perf/util/dso.h +++ b/tools/perf/util/dso.h @@ -21,6 +21,8 @@ enum dso_binary_type { DSO_BINARY_TYPE__SYSTEM_PATH_DSO, DSO_BINARY_TYPE__GUEST_KMODULE, DSO_BINARY_TYPE__SYSTEM_PATH_KMODULE, + DSO_BINARY_TYPE__KCORE, + DSO_BINARY_TYPE__GUEST_KCORE, DSO_BINARY_TYPE__NOT_FOUND, }; @@ -155,4 +157,10 @@ static inline bool dso__is_vmlinux(struct dso *dso) dso->data_type == DSO_BINARY_TYPE__GUEST_VMLINUX; } +static inline bool dso__is_kcore(struct dso *dso) +{ + return dso->data_type == DSO_BINARY_TYPE__KCORE || + dso->data_type == DSO_BINARY_TYPE__GUEST_KCORE; +} + #endif /* __PERF_DSO */ diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index dc35dcffbd0..ef3b49cde04 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -856,6 +856,18 @@ static void machine__set_kernel_mmap_len(struct machine *machine, } } +static bool machine__uses_kcore(struct machine *machine) +{ + struct dso *dso; + + list_for_each_entry(dso, &machine->kernel_dsos, node) { + if (dso__is_kcore(dso)) + return true; + } + + return false; +} + static int machine__process_kernel_mmap_event(struct machine *machine, union perf_event *event) { @@ -864,6 +876,10 @@ static int machine__process_kernel_mmap_event(struct machine *machine, enum dso_kernel_type kernel_type; bool is_kernel_mmap; + /* 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; diff --git a/tools/perf/util/map.c b/tools/perf/util/map.c index 4d599febfb0..9e8304ca343 100644 --- a/tools/perf/util/map.c +++ b/tools/perf/util/map.c @@ -555,3 +555,21 @@ struct map *maps__find(struct rb_root *maps, u64 ip) return NULL; } + +struct map *maps__first(struct rb_root *maps) +{ + struct rb_node *first = rb_first(maps); + + if (first) + return rb_entry(first, struct map, rb_node); + return NULL; +} + +struct map *maps__next(struct map *map) +{ + struct rb_node *next = rb_next(&map->rb_node); + + if (next) + return rb_entry(next, struct map, rb_node); + return NULL; +} diff --git a/tools/perf/util/map.h b/tools/perf/util/map.h index a887f2c9dfb..2cc93cbf0e1 100644 --- a/tools/perf/util/map.h +++ b/tools/perf/util/map.h @@ -112,6 +112,8 @@ size_t __map_groups__fprintf_maps(struct map_groups *mg, void maps__insert(struct rb_root *maps, struct map *map); void maps__remove(struct rb_root *maps, struct map *map); struct map *maps__find(struct rb_root *maps, u64 addr); +struct map *maps__first(struct rb_root *maps); +struct map *maps__next(struct map *map); void map_groups__init(struct map_groups *mg); void map_groups__exit(struct map_groups *mg); int map_groups__clone(struct map_groups *mg, @@ -139,6 +141,17 @@ static inline struct map *map_groups__find(struct map_groups *mg, return maps__find(&mg->maps[type], addr); } +static inline struct map *map_groups__first(struct map_groups *mg, + enum map_type type) +{ + return maps__first(&mg->maps[type]); +} + +static inline struct map *map_groups__next(struct map *map) +{ + return maps__next(map); +} + struct symbol *map_groups__find_symbol(struct map_groups *mg, enum map_type type, u64 addr, struct map **mapp, diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c index 3eaa7b486e2..a7b9ab55738 100644 --- a/tools/perf/util/symbol-elf.c +++ b/tools/perf/util/symbol-elf.c @@ -951,6 +951,57 @@ out_elf_end: return err; } +static int elf_read_maps(Elf *elf, bool exe, mapfn_t mapfn, void *data) +{ + GElf_Phdr phdr; + size_t i, phdrnum; + int err; + u64 sz; + + if (elf_getphdrnum(elf, &phdrnum)) + return -1; + + for (i = 0; i < phdrnum; i++) { + if (gelf_getphdr(elf, i, &phdr) == NULL) + return -1; + if (phdr.p_type != PT_LOAD) + continue; + if (exe) { + if (!(phdr.p_flags & PF_X)) + continue; + } else { + if (!(phdr.p_flags & PF_R)) + continue; + } + sz = min(phdr.p_memsz, phdr.p_filesz); + if (!sz) + continue; + err = mapfn(phdr.p_vaddr, sz, phdr.p_offset, data); + if (err) + return err; + } + return 0; +} + +int file__read_maps(int fd, bool exe, mapfn_t mapfn, void *data, + bool *is_64_bit) +{ + int err; + Elf *elf; + + elf = elf_begin(fd, PERF_ELF_C_READ_MMAP, NULL); + if (elf == NULL) + return -1; + + if (is_64_bit) + *is_64_bit = (gelf_getclass(elf) == ELFCLASS64); + + err = elf_read_maps(elf, exe, mapfn, data); + + elf_end(elf); + return err; +} + void symbol__elf_init(void) { elf_version(EV_CURRENT); diff --git a/tools/perf/util/symbol-minimal.c b/tools/perf/util/symbol-minimal.c index a7390cde63b..3a802c300fc 100644 --- a/tools/perf/util/symbol-minimal.c +++ b/tools/perf/util/symbol-minimal.c @@ -301,6 +301,13 @@ int dso__load_sym(struct dso *dso, struct map *map __maybe_unused, return 0; } +int file__read_maps(int fd __maybe_unused, bool exe __maybe_unused, + mapfn_t mapfn __maybe_unused, void *data __maybe_unused, + bool *is_64_bit __maybe_unused) +{ + return -1; +} + void symbol__elf_init(void) { } diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index b9056a8c162..77f3b95bb46 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -327,6 +327,16 @@ static struct symbol *symbols__find(struct rb_root *symbols, u64 ip) return NULL; } +static struct symbol *symbols__first(struct rb_root *symbols) +{ + struct rb_node *n = rb_first(symbols); + + if (n) + return rb_entry(n, struct symbol, rb_node); + + return NULL; +} + struct symbol_name_rb_node { struct rb_node rb_node; struct symbol sym; @@ -397,6 +407,11 @@ struct symbol *dso__find_symbol(struct dso *dso, return symbols__find(&dso->symbols[type], addr); } +struct symbol *dso__first_symbol(struct dso *dso, enum map_type type) +{ + return symbols__first(&dso->symbols[type]); +} + struct symbol *dso__find_symbol_by_name(struct dso *dso, enum map_type type, const char *name) { @@ -533,6 +548,53 @@ static int dso__load_all_kallsyms(struct dso *dso, const char *filename, return kallsyms__parse(filename, &args, map__process_kallsym_symbol); } +static int dso__split_kallsyms_for_kcore(struct dso *dso, struct map *map, + symbol_filter_t filter) +{ + struct map_groups *kmaps = map__kmap(map)->kmaps; + struct map *curr_map; + struct symbol *pos; + int count = 0, moved = 0; + struct rb_root *root = &dso->symbols[map->type]; + struct rb_node *next = rb_first(root); + + while (next) { + char *module; + + pos = rb_entry(next, struct symbol, rb_node); + next = rb_next(&pos->rb_node); + + module = strchr(pos->name, '\t'); + if (module) + *module = '\0'; + + curr_map = map_groups__find(kmaps, map->type, pos->start); + + if (!curr_map || (filter && filter(curr_map, pos))) { + rb_erase(&pos->rb_node, root); + symbol__delete(pos); + } else { + pos->start -= curr_map->start - curr_map->pgoff; + if (pos->end) + pos->end -= curr_map->start - curr_map->pgoff; + if (curr_map != map) { + rb_erase(&pos->rb_node, root); + symbols__insert( + &curr_map->dso->symbols[curr_map->type], + pos); + ++moved; + } else { + ++count; + } + } + } + + /* Symbols have been adjusted */ + dso->adjust_symbols = 1; + + return count + moved; +} + /* * Split the symbols into maps, making sure there are no overlaps, i.e. the * kernel range is broken in several maps, named [kernel].N, as we don't have @@ -674,6 +736,161 @@ bool symbol__restricted_filename(const char *filename, return restricted; } +struct kcore_mapfn_data { + struct dso *dso; + enum map_type type; + struct list_head maps; +}; + +static int kcore_mapfn(u64 start, u64 len, u64 pgoff, void *data) +{ + struct kcore_mapfn_data *md = data; + struct map *map; + + map = map__new2(start, md->dso, md->type); + if (map == NULL) + return -ENOMEM; + + map->end = map->start + len; + map->pgoff = pgoff; + + list_add(&map->node, &md->maps); + + return 0; +} + +/* + * If kallsyms is referenced by name then we look for kcore in the same + * directory. + */ +static bool kcore_filename_from_kallsyms_filename(char *kcore_filename, + const char *kallsyms_filename) +{ + char *name; + + strcpy(kcore_filename, kallsyms_filename); + name = strrchr(kcore_filename, '/'); + if (!name) + return false; + + if (!strcmp(name, "/kallsyms")) { + strcpy(name, "/kcore"); + return true; + } + + return false; +} + +static int dso__load_kcore(struct dso *dso, struct map *map, + const char *kallsyms_filename) +{ + struct map_groups *kmaps = map__kmap(map)->kmaps; + struct machine *machine = kmaps->machine; + struct kcore_mapfn_data md; + struct map *old_map, *new_map, *replacement_map = NULL; + bool is_64_bit; + int err, fd; + char kcore_filename[PATH_MAX]; + struct symbol *sym; + + /* This function requires that the map is the kernel map */ + if (map != machine->vmlinux_maps[map->type]) + return -EINVAL; + + if (!kcore_filename_from_kallsyms_filename(kcore_filename, + kallsyms_filename)) + return -EINVAL; + + md.dso = dso; + md.type = map->type; + INIT_LIST_HEAD(&md.maps); + + fd = open(kcore_filename, O_RDONLY); + if (fd < 0) + return -EINVAL; + + /* Read new maps into temporary lists */ + err = file__read_maps(fd, md.type == MAP__FUNCTION, kcore_mapfn, &md, + &is_64_bit); + if (err) + goto out_err; + + if (list_empty(&md.maps)) { + err = -EINVAL; + goto out_err; + } + + /* Remove old maps */ + old_map = map_groups__first(kmaps, map->type); + while (old_map) { + struct map *next = map_groups__next(old_map); + + if (old_map != map) + map_groups__remove(kmaps, old_map); + old_map = next; + } + + /* Find the kernel map using the first symbol */ + sym = dso__first_symbol(dso, map->type); + list_for_each_entry(new_map, &md.maps, node) { + if (sym && sym->start >= new_map->start && + sym->start < new_map->end) { + replacement_map = new_map; + break; + } + } + + if (!replacement_map) + replacement_map = list_entry(md.maps.next, struct map, node); + + /* Add new maps */ + while (!list_empty(&md.maps)) { + new_map = list_entry(md.maps.next, struct map, node); + list_del(&new_map->node); + if (new_map == replacement_map) { + map->start = new_map->start; + map->end = new_map->end; + map->pgoff = new_map->pgoff; + map->map_ip = new_map->map_ip; + map->unmap_ip = new_map->unmap_ip; + map__delete(new_map); + /* Ensure maps are correctly ordered */ + map_groups__remove(kmaps, map); + map_groups__insert(kmaps, map); + } else { + map_groups__insert(kmaps, new_map); + } + } + + /* + * Set the data type and long name so that kcore can be read via + * dso__data_read_addr(). + */ + if (dso->kernel == DSO_TYPE_GUEST_KERNEL) + dso->data_type = DSO_BINARY_TYPE__GUEST_KCORE; + else + dso->data_type = DSO_BINARY_TYPE__KCORE; + dso__set_long_name(dso, strdup(kcore_filename)); + + close(fd); + + if (map->type == MAP__FUNCTION) + pr_debug("Using %s for kernel object code\n", kcore_filename); + else + pr_debug("Using %s for kernel data\n", kcore_filename); + + return 0; + +out_err: + while (!list_empty(&md.maps)) { + map = list_entry(md.maps.next, struct map, node); + list_del(&map->node); + map__delete(map); + } + close(fd); + return -EINVAL; +} + int dso__load_kallsyms(struct dso *dso, const char *filename, struct map *map, symbol_filter_t filter) { @@ -691,7 +908,10 @@ int dso__load_kallsyms(struct dso *dso, const char *filename, else dso->symtab_type = DSO_BINARY_TYPE__KALLSYMS; - return dso__split_kallsyms(dso, map, filter); + if (!dso__load_kcore(dso, map, filename)) + return dso__split_kallsyms_for_kcore(dso, map, filter); + else + return dso__split_kallsyms(dso, map, filter); } static int dso__load_perf_map(struct dso *dso, struct map *map, @@ -1065,7 +1285,7 @@ do_kallsyms: pr_debug("Using %s for symbols\n", kallsyms_filename); free(kallsyms_allocated_filename); - if (err > 0) { + if (err > 0 && !dso__is_kcore(dso)) { dso__set_long_name(dso, strdup("[kernel.kallsyms]")); map__fixup_start(map); map__fixup_end(map); @@ -1109,8 +1329,9 @@ static int dso__load_guest_kernel_sym(struct dso *dso, struct map *map, } err = dso__load_kallsyms(dso, kallsyms_filename, map, filter); - if (err > 0) { + if (err > 0) pr_debug("Using %s for symbols\n", kallsyms_filename); + if (err > 0 && !dso__is_kcore(dso)) { machine__mmap_name(machine, path, sizeof(path)); dso__set_long_name(dso, strdup(path)); map__fixup_start(map); diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h index 5f720dc076d..fd5b70ea298 100644 --- a/tools/perf/util/symbol.h +++ b/tools/perf/util/symbol.h @@ -215,6 +215,7 @@ struct symbol *dso__find_symbol(struct dso *dso, enum map_type type, u64 addr); struct symbol *dso__find_symbol_by_name(struct dso *dso, enum map_type type, const char *name); +struct symbol *dso__first_symbol(struct dso *dso, enum map_type type); int filename__read_build_id(const char *filename, void *bf, size_t size); int sysfs__read_build_id(const char *filename, void *bf, size_t size); @@ -247,4 +248,8 @@ void symbols__fixup_duplicate(struct rb_root *symbols); void symbols__fixup_end(struct rb_root *symbols); void __map_groups__fixup_end(struct map_groups *mg, enum map_type type); +typedef int (*mapfn_t)(u64 start, u64 len, u64 pgoff, void *data); +int file__read_maps(int fd, bool exe, mapfn_t mapfn, void *data, + bool *is_64_bit); + #endif /* __PERF_SYMBOL */ -- cgit v1.2.3-70-g09d2 From 8f76fcd902e3b3a7d6f6c695cc8bc053579eb179 Mon Sep 17 00:00:00 2001 From: Jason Wessel Date: Mon, 15 Jul 2013 15:27:53 -0500 Subject: perf machine: Do not require /lib/modules/* on a guest For some types of work loads and special guest environments, you might have a kernel that has no kernel modules. The perf kvm record tool fails instantiate vmlinux maps when the kernel modules directory cannot be opened, even though the kallsyms has been properly processed. This leads to a perf kvm report that has no guest symbols resolved. This patch changes the failure to locate kernel modules to be non-fatal. Signed-off-by: Jason Wessel Acked-by: David Ahern Cc: Paul Mackerras Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1373920073-4874-1-git-send-email-jason.wessel@windriver.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index ef3b49cde04..6fcc358138a 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -806,7 +806,10 @@ static int machine__create_modules(struct machine *machine) free(line); fclose(file); - return machine__set_modules_path(machine); + if (machine__set_modules_path(machine) < 0) { + pr_debug("Problems setting modules path maps, continuing anyway...\n"); + } + return 0; out_delete_line: free(line); -- cgit v1.2.3-70-g09d2 From 611a5ce8aa4cdb2daefbd9bed77ec3b3e9bd00ea Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 8 Aug 2013 14:32:20 +0300 Subject: perf machine: Add symbol filter to struct machine The symbol filter needs to be applied machine-wide, so add it to struct machine. Currently tools pass the symbol filter as a parameter to various map-related functions. However a need to load a map can occur anywhere in the code, at which point the filter is needed. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1375961547-30267-2-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 20 ++++++++++++++++++++ tools/perf/util/machine.h | 5 +++++ 2 files changed, 25 insertions(+) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 6fcc358138a..4c7e0a28bf3 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -25,6 +25,8 @@ int machine__init(struct machine *machine, const char *root_dir, pid_t pid) machine->kmaps.machine = machine; machine->pid = pid; + machine->symbol_filter = NULL; + machine->root_dir = strdup(root_dir); if (machine->root_dir == NULL) return -ENOMEM; @@ -95,6 +97,7 @@ void machines__init(struct machines *machines) { machine__init(&machines->host, "", HOST_KERNEL_ID); machines->guests = RB_ROOT; + machines->symbol_filter = NULL; } void machines__exit(struct machines *machines) @@ -118,6 +121,8 @@ struct machine *machines__add(struct machines *machines, pid_t pid, return NULL; } + machine->symbol_filter = machines->symbol_filter; + while (*p != NULL) { parent = *p; pos = rb_entry(parent, struct machine, rb_node); @@ -133,6 +138,21 @@ struct machine *machines__add(struct machines *machines, pid_t pid, return machine; } +void machines__set_symbol_filter(struct machines *machines, + symbol_filter_t symbol_filter) +{ + struct rb_node *nd; + + machines->symbol_filter = symbol_filter; + machines->host.symbol_filter = symbol_filter; + + for (nd = rb_first(&machines->guests); nd; nd = rb_next(nd)) { + struct machine *machine = rb_entry(nd, struct machine, rb_node); + + machine->symbol_filter = symbol_filter; + } +} + struct machine *machines__find(struct machines *machines, pid_t pid) { struct rb_node **p = &machines->guests.rb_node; diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h index 5bb6244194d..603ffba999d 100644 --- a/tools/perf/util/machine.h +++ b/tools/perf/util/machine.h @@ -29,6 +29,7 @@ struct machine { struct list_head kernel_dsos; struct map_groups kmaps; struct map *vmlinux_maps[MAP__NR_TYPES]; + symbol_filter_t symbol_filter; }; static inline @@ -51,6 +52,7 @@ typedef void (*machine__process_t)(struct machine *machine, void *data); struct machines { struct machine host; struct rb_root guests; + symbol_filter_t symbol_filter; }; void machines__init(struct machines *machines); @@ -68,6 +70,9 @@ struct machine *machines__findnew(struct machines *machines, pid_t pid); void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size); char *machine__mmap_name(struct machine *machine, char *bf, size_t size); +void machines__set_symbol_filter(struct machines *machines, + symbol_filter_t symbol_filter); + int machine__init(struct machine *machine, const char *root_dir, pid_t pid); void machine__exit(struct machine *machine); void machine__delete_dead_threads(struct machine *machine); -- cgit v1.2.3-70-g09d2 From 61710bdee324aab1c148c8573ee49cea59d05874 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Thu, 8 Aug 2013 14:32:26 +0300 Subject: perf tools: Remove filter parameter of thread__find_addr_location() Now that the symbol filter is recorded on the machine there is no need to pass it to thread__find_addr_location(). So remove it. Signed-off-by: Adrian Hunter Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1375961547-30267-8-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/event.c | 9 +++++---- tools/perf/util/machine.c | 8 ++++---- tools/perf/util/thread.h | 3 +-- tools/perf/util/unwind.c | 2 +- 4 files changed, 11 insertions(+), 11 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index f3cf771d362..9d301c92310 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -670,12 +670,13 @@ try_again: void thread__find_addr_location(struct thread *thread, struct machine *machine, u8 cpumode, enum map_type type, u64 addr, - struct addr_location *al, - symbol_filter_t filter) + struct addr_location *al) { - thread__find_addr_map(thread, machine, cpumode, type, addr, al, filter); + thread__find_addr_map(thread, machine, cpumode, type, addr, al, + machine->symbol_filter); if (al->map != NULL) - al->sym = map__find_symbol(al->map, al->addr, filter); + al->sym = map__find_symbol(al->map, al->addr, + machine->symbol_filter); else al->sym = NULL; } diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 4c7e0a28bf3..4514e7e9b65 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -1130,7 +1130,7 @@ static void ip__resolve_ams(struct machine *machine, struct thread *thread, * or else, the symbol is unknown */ thread__find_addr_location(thread, machine, m, MAP__FUNCTION, - ip, &al, NULL); + ip, &al); if (al.sym) goto found; } @@ -1148,8 +1148,8 @@ static void ip__resolve_data(struct machine *machine, struct thread *thread, memset(&al, 0, sizeof(al)); - thread__find_addr_location(thread, machine, m, MAP__VARIABLE, addr, &al, - NULL); + thread__find_addr_location(thread, machine, m, MAP__VARIABLE, addr, + &al); ams->addr = addr; ams->al_addr = al.addr; ams->sym = al.sym; @@ -1244,7 +1244,7 @@ static int machine__resolve_callchain_sample(struct machine *machine, al.filtered = false; thread__find_addr_location(thread, machine, cpumode, - MAP__FUNCTION, ip, &al, NULL); + MAP__FUNCTION, ip, &al); if (al.sym != NULL) { if (sort__has_parent && !*parent && symbol__match_regex(al.sym, &parent_regex)) diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h index f98d1d98354..0ab47d860d4 100644 --- a/tools/perf/util/thread.h +++ b/tools/perf/util/thread.h @@ -45,8 +45,7 @@ void thread__find_addr_map(struct thread *thread, struct machine *machine, void thread__find_addr_location(struct thread *thread, struct machine *machine, u8 cpumode, enum map_type type, u64 addr, - struct addr_location *al, - symbol_filter_t filter); + struct addr_location *al); static inline void *thread__priv(struct thread *thread) { diff --git a/tools/perf/util/unwind.c b/tools/perf/util/unwind.c index 5bbd4947c27..abac3f97d95 100644 --- a/tools/perf/util/unwind.c +++ b/tools/perf/util/unwind.c @@ -473,7 +473,7 @@ static int entry(u64 ip, struct thread *thread, struct machine *machine, thread__find_addr_location(thread, machine, PERF_RECORD_MISC_USER, - MAP__FUNCTION, ip, &al, NULL); + MAP__FUNCTION, ip, &al); e.ip = ip; e.map = al.map; -- cgit v1.2.3-70-g09d2 From 236a3bbd5cb51edbf9550f5a7df885665d18a271 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 14 Aug 2013 08:49:27 -0600 Subject: perf tools: Sample after exit loses thread correlation Occassionally events (e.g., context-switch, sched tracepoints) are losing the conversion of sample data associated with a thread. For example: $ perf record -e sched:sched_switch -c 1 -a -- sleep 5 $ perf script ls 30482 [000] 1379727.583037: sched:sched_switch: prev_comm=ls prev_pid=30482 ... ls 30482 [000] 1379727.586339: sched:sched_switch: prev_comm=ls prev_pid=30482 ... :30482 30482 [000] 1379727.589462: sched:sched_switch: prev_comm=ls prev_pid=30482 ... The last line lost the conversion from tid to comm. If you look at the events (perf script -D) you see why - a SAMPLE event is generated after the EXIT: 0 1379727589449774 0x1540b0 [0x38]: PERF_RECORD_EXIT(30482:30482):(30482:30482) 0 1379727589462497 0x1540e8 [0x80]: PERF_RECORD_SAMPLE(IP, 1): 30482/30482: 0xffffffff816416f1 period: 1 addr: 0 ... thread: :30482:30482 When perf processes the EXIT event the thread is moved to the dead_threads list. When the SAMPLE event is processed no thread exists for the pid so a new one is created by machine__findnew_thread. This patch address the problem by delaying the move to the dead_threads list until the tid is re-used (per Adrian's suggestion). With this patch we get the previous example shows: ls 30482 [000] 1379727.583037: sched:sched_switch: prev_comm=ls prev_pid=30482 ... ls 30482 [000] 1379727.586339: sched:sched_switch: prev_comm=ls prev_pid=30482 ... ls 30482 [000] 1379727.589462: sched:sched_switch: prev_comm=ls prev_pid=30482 ... and 0 1379727589449774 0x1540b0 [0x38]: PERF_RECORD_EXIT(30482:30482):(30482:30482) 0 1379727589462497 0x1540e8 [0x80]: PERF_RECORD_SAMPLE(IP, 1): 30482/30482: 0xffffffff816416f1 period: 1 addr: 0 ... thread: ls:30482 v4: per Arnaldo's request add dead flag to thread struct and set when task exits v3: re-do from a time based check to a delayed move to dead_threads list v2: Rebased to latest perf/core branch. Changed time comparison to use a macro which explicitly shows the time basis Signed-off-by: David Ahern Acked-by: Adrian Hunter Cc: Adrian Hunter Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1376491767-84171-1-git-send-email-dsahern@gmail.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 34 ++++++++++++++++++++-------------- tools/perf/util/thread.h | 5 +++++ 2 files changed, 25 insertions(+), 14 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 4514e7e9b65..574feb7003a 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -1031,11 +1031,27 @@ out_problem: return 0; } +static void machine__remove_thread(struct machine *machine, struct thread *th) +{ + machine->last_match = NULL; + rb_erase(&th->rb_node, &machine->threads); + /* + * We may have references to this thread, for instance in some hist_entry + * instances, so just move them to a separate list. + */ + list_add_tail(&th->node, &machine->dead_threads); +} + int machine__process_fork_event(struct machine *machine, union perf_event *event) { - struct thread *thread = machine__findnew_thread(machine, event->fork.tid); + struct thread *thread = machine__find_thread(machine, event->fork.tid); struct thread *parent = machine__findnew_thread(machine, event->fork.ptid); + /* if a thread currently exists for the thread id remove it */ + if (thread != NULL) + machine__remove_thread(machine, thread); + + thread = machine__findnew_thread(machine, event->fork.tid); if (dump_trace) perf_event__fprintf_task(event, stdout); @@ -1048,18 +1064,8 @@ int machine__process_fork_event(struct machine *machine, union perf_event *event return 0; } -static void machine__remove_thread(struct machine *machine, struct thread *th) -{ - machine->last_match = NULL; - rb_erase(&th->rb_node, &machine->threads); - /* - * We may have references to this thread, for instance in some hist_entry - * instances, so just move them to a separate list. - */ - list_add_tail(&th->node, &machine->dead_threads); -} - -int machine__process_exit_event(struct machine *machine, union perf_event *event) +int machine__process_exit_event(struct machine *machine __maybe_unused, + union perf_event *event) { struct thread *thread = machine__find_thread(machine, event->fork.tid); @@ -1067,7 +1073,7 @@ int machine__process_exit_event(struct machine *machine, union perf_event *event perf_event__fprintf_task(event, stdout); if (thread != NULL) - machine__remove_thread(machine, thread); + thread__exited(thread); return 0; } diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h index 13c62c90939..32d060164b5 100644 --- a/tools/perf/util/thread.h +++ b/tools/perf/util/thread.h @@ -16,6 +16,7 @@ struct thread { pid_t ppid; char shortname[3]; bool comm_set; + bool dead; /* if set thread has exited */ char *comm; int comm_len; @@ -26,6 +27,10 @@ struct machine; struct thread *thread__new(pid_t tid); void thread__delete(struct thread *self); +static inline void thread__exited(struct thread *thread) +{ + thread->dead = true; +} int thread__set_comm(struct thread *self, const char *comm); int thread__comm_len(struct thread *self); -- cgit v1.2.3-70-g09d2 From 99d725fc65563a85d4290342c81b00a673c6be66 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Mon, 26 Aug 2013 16:00:19 +0300 Subject: perf tools: Add pid to struct thread Record pid on struct thread. The member is named 'pid_' to avoid confusion with the 'tid' member which was previously named 'pid'. Signed-off-by: Adrian Hunter Acked-by: David Ahern Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1377522030-27870-3-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/machine.c | 16 +++++++++++----- tools/perf/util/thread.c | 3 ++- tools/perf/util/thread.h | 3 ++- 3 files changed, 15 insertions(+), 7 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 574feb7003a..59486c18062 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -253,7 +253,8 @@ void machines__set_id_hdr_size(struct machines *machines, u16 id_hdr_size) return; } -static struct thread *__machine__findnew_thread(struct machine *machine, pid_t tid, +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; @@ -265,8 +266,11 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t * so most of the time we dont have to look up * the full rbtree: */ - if (machine->last_match && machine->last_match->tid == tid) + if (machine->last_match && machine->last_match->tid == tid) { + if (pid && pid != machine->last_match->pid_) + machine->last_match->pid_ = pid; return machine->last_match; + } while (*p != NULL) { parent = *p; @@ -274,6 +278,8 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t if (th->tid == tid) { machine->last_match = th; + if (pid && pid != th->pid_) + th->pid_ = pid; return th; } @@ -286,7 +292,7 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t if (!create) return NULL; - th = thread__new(tid); + th = thread__new(pid, tid); if (th != NULL) { rb_link_node(&th->rb_node, parent, p); rb_insert_color(&th->rb_node, &machine->threads); @@ -298,12 +304,12 @@ static struct thread *__machine__findnew_thread(struct machine *machine, pid_t t struct thread *machine__findnew_thread(struct machine *machine, pid_t tid) { - return __machine__findnew_thread(machine, tid, true); + return __machine__findnew_thread(machine, 0, tid, true); } struct thread *machine__find_thread(struct machine *machine, pid_t tid) { - return __machine__findnew_thread(machine, tid, false); + return __machine__findnew_thread(machine, 0, tid, false); } int machine__process_comm_event(struct machine *machine, union perf_event *event) diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c index 6feeb88eb5b..e3d4a550a70 100644 --- a/tools/perf/util/thread.c +++ b/tools/perf/util/thread.c @@ -7,12 +7,13 @@ #include "util.h" #include "debug.h" -struct thread *thread__new(pid_t tid) +struct thread *thread__new(pid_t pid, pid_t tid) { struct thread *self = zalloc(sizeof(*self)); if (self != NULL) { map_groups__init(&self->mg); + self->pid_ = pid; self->tid = tid; self->ppid = -1; self->comm = malloc(32); diff --git a/tools/perf/util/thread.h b/tools/perf/util/thread.h index 32d060164b5..4ebbb40d46d 100644 --- a/tools/perf/util/thread.h +++ b/tools/perf/util/thread.h @@ -12,6 +12,7 @@ struct thread { struct list_head node; }; struct map_groups mg; + pid_t pid_; /* Not all tools update this */ pid_t tid; pid_t ppid; char shortname[3]; @@ -25,7 +26,7 @@ struct thread { struct machine; -struct thread *thread__new(pid_t tid); +struct thread *thread__new(pid_t pid, pid_t tid); void thread__delete(struct thread *self); static inline void thread__exited(struct thread *thread) { -- cgit v1.2.3-70-g09d2 From 314add6b1f045b59ca39683bd0cbc5310cd203f2 Mon Sep 17 00:00:00 2001 From: Adrian Hunter Date: Tue, 27 Aug 2013 11:23:03 +0300 Subject: perf tools: change machine__findnew_thread() to set thread pid Add a new parameter for 'pid' to machine__findnew_thread(). Change callers to pass 'pid' when it is known. Note that callers sometimes want to find the main thread which has the memory maps. The main thread has tid == pid so the usage in that case is: machine__findnew_thread(machine, pid, pid) whereas the usage to find the specific thread is: machine__findnew_thread(machine, pid, tid) Signed-off-by: Adrian Hunter Acked-by: David Ahern Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1377591794-30553-2-git-send-email-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/builtin-inject.c | 2 +- tools/perf/builtin-kmem.c | 3 ++- tools/perf/builtin-kvm.c | 2 +- tools/perf/builtin-lock.c | 3 ++- tools/perf/builtin-sched.c | 20 +++++++++++--------- tools/perf/builtin-script.c | 3 ++- tools/perf/builtin-trace.c | 10 +++++++--- tools/perf/tests/code-reading.c | 4 ++-- tools/perf/tests/hists_link.c | 3 ++- tools/perf/util/build-id.c | 7 +++++-- tools/perf/util/event.c | 3 ++- tools/perf/util/machine.c | 22 +++++++++++++++------- tools/perf/util/machine.h | 3 ++- tools/perf/util/session.c | 2 +- 14 files changed, 55 insertions(+), 32 deletions(-) (limited to 'tools/perf/util/machine.c') diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c index 1d8de2e4a40..0d4ae1dd7b6 100644 --- a/tools/perf/builtin-inject.c +++ b/tools/perf/builtin-inject.c @@ -198,7 +198,7 @@ static int perf_event__inject_buildid(struct perf_tool *tool, cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; - thread = machine__findnew_thread(machine, event->ip.pid); + thread = machine__findnew_thread(machine, event->ip.pid, event->ip.pid); if (thread == NULL) { pr_err("problem processing %d event, skipping it.\n", event->header.type); diff --git a/tools/perf/builtin-kmem.c b/tools/perf/builtin-kmem.c index b49f5c58e15..c32477837cb 100644 --- a/tools/perf/builtin-kmem.c +++ b/tools/perf/builtin-kmem.c @@ -305,7 +305,8 @@ static int process_sample_event(struct perf_tool *tool __maybe_unused, struct perf_evsel *evsel, struct machine *machine) { - struct thread *thread = machine__findnew_thread(machine, event->ip.pid); + struct thread *thread = machine__findnew_thread(machine, event->ip.pid, + event->ip.pid); if (thread == NULL) { pr_debug("problem processing %d event, skipping it.\n", diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c index 6cd4de59be2..47b35407c2f 100644 --- a/tools/perf/builtin-kvm.c +++ b/tools/perf/builtin-kvm.c @@ -815,7 +815,7 @@ static int process_sample_event(struct perf_tool *tool, if (skip_sample(kvm, sample)) return 0; - thread = machine__findnew_thread(machine, sample->tid); + thread = machine__findnew_thread(machine, sample->pid, sample->tid); if (thread == NULL) { pr_debug("problem processing %d event, skipping it.\n", event->header.type); diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c index 76543a4a7a3..ee33ba2f05d 100644 --- a/tools/perf/builtin-lock.c +++ b/tools/perf/builtin-lock.c @@ -805,7 +805,8 @@ static int process_sample_event(struct perf_tool *tool __maybe_unused, struct perf_evsel *evsel, struct machine *machine) { - struct thread *thread = machine__findnew_thread(machine, sample->tid); + struct thread *thread = machine__findnew_thread(machine, sample->pid, + sample->tid); if (thread == NULL) { pr_debug("problem processing %d event, skipping it.\n", diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c index f809cc7fb7d..d8c51b2f263 100644 --- a/tools/perf/builtin-sched.c +++ b/tools/perf/builtin-sched.c @@ -724,8 +724,10 @@ static int replay_fork_event(struct perf_sched *sched, { struct thread *child, *parent; - child = machine__findnew_thread(machine, event->fork.tid); - parent = machine__findnew_thread(machine, event->fork.ptid); + child = machine__findnew_thread(machine, event->fork.pid, + event->fork.tid); + parent = machine__findnew_thread(machine, event->fork.ppid, + event->fork.ptid); if (child == NULL || parent == NULL) { pr_debug("thread does not exist on fork event: child %p, parent %p\n", @@ -934,8 +936,8 @@ static int latency_switch_event(struct perf_sched *sched, return -1; } - sched_out = machine__findnew_thread(machine, prev_pid); - sched_in = machine__findnew_thread(machine, next_pid); + sched_out = machine__findnew_thread(machine, 0, prev_pid); + sched_in = machine__findnew_thread(machine, 0, next_pid); out_events = thread_atoms_search(&sched->atom_root, sched_out, &sched->cmp_pid); if (!out_events) { @@ -978,7 +980,7 @@ static int latency_runtime_event(struct perf_sched *sched, { const u32 pid = perf_evsel__intval(evsel, sample, "pid"); const u64 runtime = perf_evsel__intval(evsel, sample, "runtime"); - struct thread *thread = machine__findnew_thread(machine, pid); + struct thread *thread = machine__findnew_thread(machine, 0, pid); struct work_atoms *atoms = thread_atoms_search(&sched->atom_root, thread, &sched->cmp_pid); u64 timestamp = sample->time; int cpu = sample->cpu; @@ -1016,7 +1018,7 @@ static int latency_wakeup_event(struct perf_sched *sched, if (!success) return 0; - wakee = machine__findnew_thread(machine, pid); + wakee = machine__findnew_thread(machine, 0, pid); atoms = thread_atoms_search(&sched->atom_root, wakee, &sched->cmp_pid); if (!atoms) { if (thread_atoms_insert(sched, wakee)) @@ -1070,7 +1072,7 @@ static int latency_migrate_task_event(struct perf_sched *sched, if (sched->profile_cpu == -1) return 0; - migrant = machine__findnew_thread(machine, pid); + migrant = machine__findnew_thread(machine, 0, pid); atoms = thread_atoms_search(&sched->atom_root, migrant, &sched->cmp_pid); if (!atoms) { if (thread_atoms_insert(sched, migrant)) @@ -1289,8 +1291,8 @@ static int map_switch_event(struct perf_sched *sched, struct perf_evsel *evsel, return -1; } - sched_out = machine__findnew_thread(machine, prev_pid); - sched_in = machine__findnew_thread(machine, next_pid); + sched_out = machine__findnew_thread(machine, 0, prev_pid); + sched_in = machine__findnew_thread(machine, 0, next_pid); sched->curr_thread[this_cpu] = sched_in; diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c index 2ad9d5b6fb3..d82712f169b 100644 --- a/tools/perf/builtin-script.c +++ b/tools/perf/builtin-script.c @@ -501,7 +501,8 @@ static int process_sample_event(struct perf_tool *tool __maybe_unused, struct machine *machine) { struct addr_location al; - struct thread *thread = machine__findnew_thread(machine, event->ip.tid); + struct thread *thread = machine__findnew_thread(machine, event->ip.pid, + event->ip.tid); if (thread == NULL) { pr_debug("problem processing %d event, skipping it.\n", diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index b72afc73f9a..88387c56568 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -521,7 +521,8 @@ static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel, if (sc->filtered) return 0; - thread = machine__findnew_thread(&trace->host, sample->tid); + thread = machine__findnew_thread(&trace->host, sample->pid, + sample->tid); ttrace = thread__trace(thread, trace->output); if (ttrace == NULL) return -1; @@ -572,7 +573,8 @@ static int trace__sys_exit(struct trace *trace, struct perf_evsel *evsel, if (sc->filtered) return 0; - thread = machine__findnew_thread(&trace->host, sample->tid); + thread = machine__findnew_thread(&trace->host, sample->pid, + sample->tid); ttrace = thread__trace(thread, trace->output); if (ttrace == NULL) return -1; @@ -628,7 +630,9 @@ static int trace__sched_stat_runtime(struct trace *trace, struct perf_evsel *evs { u64 runtime = perf_evsel__intval(evsel, sample, "runtime"); double runtime_ms = (double)runtime / NSEC_PER_MSEC; - struct thread *thread = machine__findnew_thread(&trace->host, sample->tid); + struct thread *thread = machine__findnew_thread(&trace->host, + sample->pid, + sample->tid); struct thread_trace *ttrace = thread__trace(thread, trace->output); if (ttrace == NULL) diff --git a/tools/perf/tests/code-reading.c b/tools/perf/tests/code-reading.c index df9afd9cab4..6fb781d5586 100644 --- a/tools/perf/tests/code-reading.c +++ b/tools/perf/tests/code-reading.c @@ -257,7 +257,7 @@ static int process_sample_event(struct machine *machine, return -1; } - thread = machine__findnew_thread(machine, sample.pid); + thread = machine__findnew_thread(machine, sample.pid, sample.pid); if (!thread) { pr_debug("machine__findnew_thread failed\n"); return -1; @@ -446,7 +446,7 @@ static int do_test_code_reading(bool try_kcore) goto out_err; } - thread = machine__findnew_thread(machine, pid); + thread = machine__findnew_thread(machine, pid, pid); if (!thread) { pr_debug("machine__findnew_thread failed\n"); goto out_err; diff --git a/tools/perf/tests/hists_link.c b/tools/perf/tests/hists_link.c index 50bfb01183e..87f9f7280c4 100644 --- a/tools/perf/tests/hists_link.c +++ b/tools/perf/tests/hists_link.c @@ -88,7 +88,8 @@ static struct machine *setup_fake_machine(struct machines *machines) for (i = 0; i < ARRAY_SIZE(fake_threads); i++) { struct thread *thread; - thread = machine__findnew_thread(machine, fake_threads[i].pid); + thread = machine__findnew_thread(machine, fake_threads[i].pid, + fake_threads[i].pid); if (thread == NULL) goto out; diff --git a/tools/perf/util/build-id.c b/tools/perf/util/build-id.c index 5295625c0c0..0f9d27a6bc8 100644 --- a/tools/perf/util/build-id.c +++ b/tools/perf/util/build-id.c @@ -24,7 +24,8 @@ int build_id__mark_dso_hit(struct perf_tool *tool __maybe_unused, { struct addr_location al; u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; - struct thread *thread = machine__findnew_thread(machine, event->ip.pid); + struct thread *thread = machine__findnew_thread(machine, event->ip.pid, + event->ip.pid); if (thread == NULL) { pr_err("problem processing %d event, skipping it.\n", @@ -47,7 +48,9 @@ static int perf_event__exit_del_thread(struct perf_tool *tool __maybe_unused, __maybe_unused, struct machine *machine) { - struct thread *thread = machine__findnew_thread(machine, event->fork.tid); + struct thread *thread = machine__findnew_thread(machine, + event->fork.pid, + event->fork.tid); dump_printf("(%d:%d):(%d:%d)\n", event->fork.pid, event->fork.tid, event->fork.ppid, event->fork.ptid); diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index 49713ae4655..61cecf9caff 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -686,7 +686,8 @@ int perf_event__preprocess_sample(const union perf_event *event, struct perf_sample *sample) { u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK; - struct thread *thread = machine__findnew_thread(machine, event->ip.pid); + struct thread *thread = machine__findnew_thread(machine, event->ip.pid, + event->ip.pid); if (thread == NULL) return -1; diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c index 59486c18062..1dca61f0512 100644 --- a/tools/perf/util/machine.c +++ b/tools/perf/util/machine.c @@ -32,7 +32,8 @@ int machine__init(struct machine *machine, const char *root_dir, pid_t pid) return -ENOMEM; if (pid != HOST_KERNEL_ID) { - struct thread *thread = machine__findnew_thread(machine, pid); + struct thread *thread = machine__findnew_thread(machine, 0, + pid); char comm[64]; if (thread == NULL) @@ -302,9 +303,10 @@ static struct thread *__machine__findnew_thread(struct machine *machine, return th; } -struct thread *machine__findnew_thread(struct machine *machine, pid_t tid) +struct thread *machine__findnew_thread(struct machine *machine, pid_t pid, + pid_t tid) { - return __machine__findnew_thread(machine, 0, tid, true); + return __machine__findnew_thread(machine, pid, tid, true); } struct thread *machine__find_thread(struct machine *machine, pid_t tid) @@ -314,7 +316,9 @@ struct thread *machine__find_thread(struct machine *machine, pid_t tid) int machine__process_comm_event(struct machine *machine, union perf_event *event) { - struct thread *thread = machine__findnew_thread(machine, event->comm.tid); + struct thread *thread = machine__findnew_thread(machine, + event->comm.pid, + event->comm.tid); if (dump_trace) perf_event__fprintf_comm(event, stdout); @@ -1012,7 +1016,8 @@ int machine__process_mmap_event(struct machine *machine, union perf_event *event return 0; } - thread = machine__findnew_thread(machine, event->mmap.pid); + thread = machine__findnew_thread(machine, event->mmap.pid, + event->mmap.pid); if (thread == NULL) goto out_problem; @@ -1051,13 +1056,16 @@ static void machine__remove_thread(struct machine *machine, struct thread *th) int machine__process_fork_event(struct machine *machine, union perf_event *event) { struct thread *thread = machine__find_thread(machine, event->fork.tid); - struct thread *parent = machine__findnew_thread(machine, event->fork.ptid); + struct thread *parent = machine__findnew_thread(machine, + event->fork.ppid, + event->fork.ptid); /* if a thread currently exists for the thread id remove it */ if (thread != NULL) machine__remove_thread(machine, thread); - thread = machine__findnew_thread(machine, event->fork.tid); + thread = machine__findnew_thread(machine, event->fork.pid, + event->fork.tid); if (dump_trace) perf_event__fprintf_task(event, stdout); diff --git a/tools/perf/util/machine.h b/tools/perf/util/machine.h index 603ffba999d..0df925ba6a4 100644 --- a/tools/perf/util/machine.h +++ b/tools/perf/util/machine.h @@ -106,7 +106,8 @@ static inline bool machine__is_host(struct machine *machine) return machine ? machine->pid == HOST_KERNEL_ID : false; } -struct thread *machine__findnew_thread(struct machine *machine, pid_t tid); +struct thread *machine__findnew_thread(struct machine *machine, pid_t pid, + pid_t tid); size_t machine__fprintf(struct machine *machine, FILE *fp); diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index de16a773685..57b6f38f246 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -1099,7 +1099,7 @@ void perf_event_header__bswap(struct perf_event_header *self) struct thread *perf_session__findnew(struct perf_session *session, pid_t pid) { - return machine__findnew_thread(&session->machines.host, pid); + return machine__findnew_thread(&session->machines.host, 0, pid); } static struct thread *perf_session__register_idle_thread(struct perf_session *self) -- cgit v1.2.3-70-g09d2