From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934289Ab3GRT4P (ORCPT ); Thu, 18 Jul 2013 15:56:15 -0400 Received: from merlin.infradead.org ([205.233.59.134]:41672 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759472Ab3GRTu3 (ORCPT ); Thu, 18 Jul 2013 15:50:29 -0400 From: Arnaldo Carvalho de Melo To: Ingo Molnar Cc: linux-kernel@vger.kernel.org, Greg Price , Greg Price , David Ahern , Ingo Molnar , Jiri Olsa , Namhyung Kim , Paul Mackerras , Peter Zijlstra , Fengguang Wu , Arnaldo Carvalho de Melo Subject: [PATCH 50/76] perf report/top: Add option to collapse undesired parts of call graph Date: Thu, 18 Jul 2013 16:49:37 -0300 Message-Id: <1374177003-3706-51-git-send-email-acme@infradead.org> X-Mailer: git-send-email 1.8.1.4 In-Reply-To: <1374177003-3706-1-git-send-email-acme@infradead.org> References: <1374177003-3706-1-git-send-email-acme@infradead.org> X-SRS-Rewrite: SMTP reverse-path rewritten from by merlin.infradead.org See http://www.infradead.org/rpr.html Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Greg Price 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(-) diff --git a/tools/perf/Documentation/perf-report.txt b/tools/perf/Documentation/perf-report.txt index 66dab74..747ff50 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 7fdd190..58d6598 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 ee2ca3e..9a7e54d 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 a237059..bbf4635 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 5dd5026..f9f9d63 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 e49ba01..5bb6244 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 951a1cf..1eb58ee 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 8deee19..cb2b108 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 45ac84c..a4a6d0b 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 */ -- 1.8.1.4