From 2a8083f063472f27c253545dd64e1a7bbbb1ab61 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 11 Aug 2009 16:22:00 -0300 Subject: perf record: Fix .tid and .pid fill-in when synthesizing events MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Noticed when trying to record events for a firefox thread. We were synthesizing both .tid and .pid with the pid passed via --pid. Fix it by reading /proc/PID/status and getting the tgid to use in .pid, .tid gets the specified "pid". Signed-off-by: Arnaldo Carvalho de Melo Cc: "H. Peter Anvin" Cc: Frédéric Weisbecker Cc: Peter Zijlstra Cc: Mike Galbraith LKML-Reference: <20090811192200.GF18061@ghostprotocols.net> Signed-off-by: Ingo Molnar --- tools/perf/builtin-record.c | 71 ++++++++++++++++++++++++--------------------- 1 file changed, 38 insertions(+), 33 deletions(-) (limited to 'tools') diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 0345aad8eba5..30b83def03d4 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -203,46 +203,48 @@ static void sig_atexit(void) kill(getpid(), signr); } -static void pid_synthesize_comm_event(pid_t pid, int full) +static pid_t pid_synthesize_comm_event(pid_t pid, int full) { struct comm_event comm_ev; char filename[PATH_MAX]; char bf[BUFSIZ]; - int fd; - size_t size; - char *field, *sep; + FILE *fp; + size_t size = 0; DIR *tasks; struct dirent dirent, *next; + pid_t tgid = 0; - snprintf(filename, sizeof(filename), "/proc/%d/stat", pid); + snprintf(filename, sizeof(filename), "/proc/%d/status", pid); - fd = open(filename, O_RDONLY); - if (fd < 0) { + fp = fopen(filename, "r"); + if (fd == NULL) { /* * We raced with a task exiting - just return: */ if (verbose) fprintf(stderr, "couldn't open %s\n", filename); - return; + return 0; } - if (read(fd, bf, sizeof(bf)) < 0) { - fprintf(stderr, "couldn't read %s\n", filename); - exit(EXIT_FAILURE); - } - close(fd); - /* 9027 (cat) R 6747 9027 6747 34816 9027 ... */ memset(&comm_ev, 0, sizeof(comm_ev)); - field = strchr(bf, '('); - if (field == NULL) - goto out_failure; - sep = strchr(++field, ')'); - if (sep == NULL) - goto out_failure; - size = sep - field; - memcpy(comm_ev.comm, field, size++); - - comm_ev.pid = pid; + while (!comm_ev.comm[0] || !comm_ev.pid) { + if (fgets(bf, sizeof(bf), fp) == NULL) + goto out_failure; + + if (memcmp(bf, "Name:", 5) == 0) { + char *name = bf + 5; + while (*name && isspace(*name)) + ++name; + size = strlen(name) - 1; + memcpy(comm_ev.comm, name, size++); + } else if (memcmp(bf, "Tgid:", 5) == 0) { + char *tgids = bf + 5; + while (*tgids && isspace(*tgids)) + ++tgids; + tgid = comm_ev.pid = atoi(tgids); + } + } + comm_ev.header.type = PERF_EVENT_COMM; size = ALIGN(size, sizeof(u64)); comm_ev.header.size = sizeof(comm_ev) - (sizeof(comm_ev.comm) - size); @@ -251,7 +253,7 @@ static void pid_synthesize_comm_event(pid_t pid, int full) comm_ev.tid = pid; write_output(&comm_ev, comm_ev.header.size); - return; + goto out_fclose; } snprintf(filename, sizeof(filename), "/proc/%d/task", pid); @@ -268,7 +270,10 @@ static void pid_synthesize_comm_event(pid_t pid, int full) write_output(&comm_ev, comm_ev.header.size); } closedir(tasks); - return; + +out_fclose: + fclose(fp); + return tgid; out_failure: fprintf(stderr, "couldn't get COMM and pgid, malformed %s\n", @@ -276,7 +281,7 @@ out_failure: exit(EXIT_FAILURE); } -static void pid_synthesize_mmap_samples(pid_t pid) +static void pid_synthesize_mmap_samples(pid_t pid, pid_t tgid) { char filename[PATH_MAX]; FILE *fp; @@ -328,7 +333,7 @@ static void pid_synthesize_mmap_samples(pid_t pid) mmap_ev.len -= mmap_ev.start; mmap_ev.header.size = (sizeof(mmap_ev) - (sizeof(mmap_ev.filename) - size)); - mmap_ev.pid = pid; + mmap_ev.pid = tgid; mmap_ev.tid = pid; write_output(&mmap_ev, mmap_ev.header.size); @@ -347,14 +352,14 @@ static void synthesize_all(void) while (!readdir_r(proc, &dirent, &next) && next) { char *end; - pid_t pid; + pid_t pid, tgid; pid = strtol(dirent.d_name, &end, 10); if (*end) /* only interested in proper numerical dirents */ continue; - pid_synthesize_comm_event(pid, 1); - pid_synthesize_mmap_samples(pid); + tgid = pid_synthesize_comm_event(pid, 1); + pid_synthesize_mmap_samples(pid, tgid); } closedir(proc); @@ -567,8 +572,8 @@ static int __cmd_record(int argc, const char **argv) perf_header__write(header, output); if (!system_wide) { - pid_synthesize_comm_event(pid, 0); - pid_synthesize_mmap_samples(pid); + pid_t tgid = pid_synthesize_comm_event(pid, 0); + pid_synthesize_mmap_samples(pid, tgid); } else synthesize_all(); -- cgit v1.2.3 From 94a24752fe95ca1e7f98b197052d44e6a207740d Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 11 Aug 2009 16:21:38 -0300 Subject: perf report: Show the tid too in -D MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This made it easier to find the firefox threading related bug. Signed-off-by: Arnaldo Carvalho de Melo Cc: "H. Peter Anvin" Cc: Peter Zijlstra Cc: Mike Galbraith Cc: Frédéric Weisbecker LKML-Reference: <20090811192138.GE18061@ghostprotocols.net> Signed-off-by: Ingo Molnar --- tools/perf/builtin-report.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'tools') diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index 99274cec0adb..23e1457f2409 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -1526,11 +1526,11 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head) more_data += sizeof(u64); } - dprintf("%p [%p]: PERF_EVENT_SAMPLE (IP, %d): %d: %p period: %Ld\n", + dprintf("%p [%p]: PERF_EVENT_SAMPLE (IP, %d): %d/%d: %p period: %Ld\n", (void *)(offset + head), (void *)(long)(event->header.size), event->header.misc, - event->ip.pid, + event->ip.pid, event->ip.tid, (void *)(long)ip, (long long)period); @@ -1612,10 +1612,11 @@ process_mmap_event(event_t *event, unsigned long offset, unsigned long head) struct thread *thread = threads__findnew(event->mmap.pid); struct map *map = map__new(&event->mmap); - dprintf("%p [%p]: PERF_EVENT_MMAP %d: [%p(%p) @ %p]: %s\n", + dprintf("%p [%p]: PERF_EVENT_MMAP %d/%d: [%p(%p) @ %p]: %s\n", (void *)(offset + head), (void *)(long)(event->header.size), event->mmap.pid, + event->mmap.tid, (void *)(long)event->mmap.start, (void *)(long)event->mmap.len, (void *)(long)event->mmap.pgoff, -- cgit v1.2.3 From 247648e3742ded01e42a4b14c2da330b13cbb47f Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 11 Aug 2009 16:22:11 -0300 Subject: perf tools: Fix fallback to cplus_demangle() when bfd_demangle() is not available MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In old binutils we can't access bfd_demangle(), use cplus_demangle() just like oprofile. Signed-off-by: Arnaldo Carvalho de Melo Cc: Luis Claudio R. Gonçalves Cc: H. Peter Anvin Cc: Peter Zijlstra Cc: Mike Galbraith Cc: Frédéric Weisbecker LKML-Reference: <20090811192211.GG18061@ghostprotocols.net> Signed-off-by: Ingo Molnar --- tools/perf/Makefile | 29 ++++++++++++++++++----------- tools/perf/util/symbol.c | 15 --------------- tools/perf/util/symbol.h | 24 ++++++++++++++++++++++++ 3 files changed, 42 insertions(+), 26 deletions(-) (limited to 'tools') diff --git a/tools/perf/Makefile b/tools/perf/Makefile index 60411e94113b..c045b4271e57 100644 --- a/tools/perf/Makefile +++ b/tools/perf/Makefile @@ -382,22 +382,29 @@ endif ifdef NO_DEMANGLE BASIC_CFLAGS += -DNO_DEMANGLE else - has_bfd := $(shell sh -c "(echo '\#include '; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd > /dev/null 2>&1 && echo y") - has_bfd_iberty := $(shell sh -c "(echo '\#include '; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty > /dev/null 2>&1 && echo y") - - has_bfd_iberty_z := $(shell sh -c "(echo '\#include '; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty -lz > /dev/null 2>&1 && echo y") - ifeq ($(has_bfd),y) EXTLIBS += -lbfd - else ifeq ($(has_bfd_iberty),y) - EXTLIBS += -lbfd -liberty - else ifeq ($(has_bfd_iberty_z),y) - EXTLIBS += -lbfd -liberty -lz else - msg := $(warning No bfd.h/libbfd found, install binutils-dev[el] to gain symbol demangling) - BASIC_CFLAGS += -DNO_DEMANGLE + has_bfd_iberty := $(shell sh -c "(echo '\#include '; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty > /dev/null 2>&1 && echo y") + ifeq ($(has_bfd_iberty),y) + EXTLIBS += -lbfd -liberty + else + has_bfd_iberty_z := $(shell sh -c "(echo '\#include '; echo 'int main(void) { bfd_demangle(0, 0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -lbfd -liberty -lz > /dev/null 2>&1 && echo y") + ifeq ($(has_bfd_iberty_z),y) + EXTLIBS += -lbfd -liberty -lz + else + has_cplus_demangle := $(shell sh -c "(echo 'extern char *cplus_demangle(const char *, int);'; echo 'int main(void) { cplus_demangle(0, 0); return 0; }') | $(CC) -x c - $(ALL_CFLAGS) -o /dev/null $(ALL_LDFLAGS) -liberty > /dev/null 2>&1 && echo y") + ifeq ($(has_cplus_demangle),y) + EXTLIBS += -liberty + BASIC_CFLAGS += -DHAVE_CPLUS_DEMANGLE + else + msg := $(warning No bfd.h/libbfd found, install binutils-dev[el] to gain symbol demangling) + BASIC_CFLAGS += -DNO_DEMANGLE + endif + endif + endif endif endif diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index f1dcede14307..2473fd427beb 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -7,23 +7,8 @@ #include #include -#ifndef NO_DEMANGLE -#include -#else -static inline -char *bfd_demangle(void __used *v, const char __used *c, int __used i) -{ - return NULL; -} -#endif - const char *sym_hist_filter; -#ifndef DMGL_PARAMS -#define DMGL_PARAMS (1 << 0) /* Include function args */ -#define DMGL_ANSI (1 << 1) /* Include const, volatile, etc */ -#endif - enum dso_origin { DSO__ORIG_KERNEL = 0, DSO__ORIG_JAVA_JIT, diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h index 1e003ec2f4b1..b53bf0125c1b 100644 --- a/tools/perf/util/symbol.h +++ b/tools/perf/util/symbol.h @@ -7,6 +7,30 @@ #include #include "module.h" +#ifdef HAVE_CPLUS_DEMANGLE +extern char *cplus_demangle(const char *, int); + +static inline char *bfd_demangle(void __used *v, const char *c, int i) +{ + return cplus_demangle(c, i); +} +#else +#ifdef NO_DEMANGLE +static inline char *bfd_demangle(void __used *v, const char __used *c, + int __used i) +{ + return NULL; +} +#else +#include +#endif +#endif + +#ifndef DMGL_PARAMS +#define DMGL_PARAMS (1 << 0) /* Include function args */ +#define DMGL_ANSI (1 << 1) /* Include const, volatile, etc */ +#endif + struct symbol { struct rb_node rb_node; u64 start; -- cgit v1.2.3 From 1340e6bbaff7ff7f6f75eb4a5c34933efce84a84 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 11 Aug 2009 17:04:36 -0300 Subject: perf tools: Fix dso__new handle() to handle deleted DSOs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It is better than showing the map addr, this way at least we know that we can't get the symtabs because the DSO was deleted (system update) while an app still used such DSO. Yeah, don't do that, but if you do, you'll figure it out quicker this way. [acme@doppio linux-2.6-tip]$ perf report | head -15 # Samples: 3796 # # Overhead Command Shared Object Symbol # ........ ....... ................................................................... ...... # 23.55% pidgin /lib64/libglib-2.0.so.0.2000.4.#prelink#.Pd98lu (deleted) [.] 0x00000000038844 21.55% pidgin /lib64/libpthread-2.10.1.so.#prelink#.AFwK8Q (deleted) [.] 0x0000000000a42d 10.85% pidgin [kernel] [.] vread_hpet 7.85% pidgin /lib64/libgobject-2.0.so.0.2000.4.#prelink#.o1vpU7 (deleted) [.] 0x00000000014de8 3.35% pidgin /lib64/libc-2.10.1.so (deleted) [.] 0x0000000007a875 3.19% pidgin /lib64/libdbus-1.so.3.4.0.#prelink#.6mwgZP (deleted) [.] 0x0000000001d254 3.06% pidgin /usr/lib64/libgtk-x11-2.0.so.0.1600.5.#prelink#.511hAl (deleted) [.] 0x000000002334e7 2.90% pidgin /usr/lib64/libgdk-x11-2.0.so.0.1600.5.#prelink#.5qlMo1 (deleted) [.] 0x00000000037b2d 1.84% pidgin [kernel] [k] do_sys_poll 1.45% pidgin /usr/lib64/libX11.so.6.2.0.#prelink#.iR59Rx (deleted) [.] 0x0000000004c751 [acme@doppio linux-2.6-tip]$ Signed-off-by: Arnaldo Carvalho de Melo Cc: Luis Claudio R. Gonçalves Cc: Clark Williams Cc: H. Peter Anvin Cc: Peter Zijlstra Cc: Mike Galbraith Cc: Frédéric Weisbecker LKML-Reference: <20090811200436.GA3478@ghostprotocols.net> Signed-off-by: Ingo Molnar --- tools/perf/util/symbol.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'tools') diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index 2473fd427beb..5c0f42e6b33b 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -801,6 +801,8 @@ more: } out: free(name); + if (ret < 0 && strstr(self->name, " (deleted)") != NULL) + return 0; return ret; } -- cgit v1.2.3 From 0a5ac84650fb7a7f226814103d95724e34b012ae Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Wed, 12 Aug 2009 11:18:01 +0200 Subject: perf record: Add missing -C option support for specifying profile cpu perf top supports a -C for setting the profile CPU, but perf record does not. This adds the same option for perf record, allowing the user to specify a specific target profile CPU. Signed-off-by: Jens Axboe Cc: Peter Zijlstra Cc: Mike Galbraith Cc: Paul Mackerras Cc: Arnaldo Carvalho de Melo Cc: Frederic Weisbecker LKML-Reference: <20090812091801.GC12579@kernel.dk> Signed-off-by: Ingo Molnar --- tools/perf/builtin-record.c | 17 ++++++++++++++--- 1 file changed, 14 insertions(+), 3 deletions(-) (limited to 'tools') diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 30b83def03d4..de76008fbaa2 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -35,6 +35,7 @@ static const char *output_name = "perf.data"; static int group = 0; static unsigned int realtime_prio = 0; static int system_wide = 0; +static int profile_cpu = -1; static pid_t target_pid = -1; static int inherit = 1; static int force = 0; @@ -431,6 +432,8 @@ try_again: if (err == EPERM) die("Permission error - are you root?\n"); + else if (err == ENODEV && profile_cpu != -1) + die("No such device - did you specify an out-of-range profile CPU?\n"); /* * If it's cycles then fall back to hrtimer @@ -564,9 +567,15 @@ static int __cmd_record(int argc, const char **argv) if (pid == -1) pid = getpid(); - open_counters(-1, pid); - } else for (i = 0; i < nr_cpus; i++) - open_counters(i, target_pid); + open_counters(profile_cpu, pid); + } else { + if (profile_cpu != -1) { + open_counters(profile_cpu, target_pid); + } else { + for (i = 0; i < nr_cpus; i++) + open_counters(i, target_pid); + } + } if (file_new) perf_header__write(header, output); @@ -645,6 +654,8 @@ static const struct option options[] = { "system-wide collection from all CPUs"), OPT_BOOLEAN('A', "append", &append_file, "append to the output file to do incremental profiling"), + OPT_INTEGER('C', "profile_cpu", &profile_cpu, + "CPU to profile on"), OPT_BOOLEAN('f', "force", &force, "overwrite existing data file"), OPT_LONG('c', "count", &default_interval, -- cgit v1.2.3 From 8f7a0dc51674ad0dd06155291b0aed60d655943c Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 12 Aug 2009 14:44:59 -0300 Subject: perf list: Fix large list output by using the pager When /sys/kernel/debug is mounted the list can be imense, so use the pager like the other tools. Signed-off-by: Arnaldo Carvalho de Melo Acked-by: Frederic Weisbecker Cc: Peter Zijlstra LKML-Reference: <20090812174459.GB3495@ghostprotocols.net> Signed-off-by: Ingo Molnar --- tools/perf/builtin-list.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/builtin-list.c b/tools/perf/builtin-list.c index f990fa8a35c9..d88c6961274c 100644 --- a/tools/perf/builtin-list.c +++ b/tools/perf/builtin-list.c @@ -10,11 +10,12 @@ #include "perf.h" -#include "util/parse-options.h" #include "util/parse-events.h" +#include "util/cache.h" int cmd_list(int argc __used, const char **argv __used, const char *prefix __used) { + setup_pager(); print_events(); return 0; } -- cgit v1.2.3 From 3a9f131fb00b8ac5950a11ad1599e45edfb5ae44 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 13 Aug 2009 10:27:18 +0200 Subject: perf tools: Add a per tracepoint counter attribute to get raw sample Add a new flag field while opening a tracepoint perf counter: -e tracepoint_subsystem:tracepoint_name:flags This is intended to be generic although for now it only supports the r[e[c[o[r[d]]]]] flag: ./perf record -e workqueue:workqueue_insertion:record ./perf record -e workqueue:workqueue_insertion:r will have the same effect: enabling the raw samples record for the given tracepoint counter. In the future, we may want to support further flags, separated by commas. Signed-off-by: Frederic Weisbecker Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Mike Galbraith LKML-Reference: <1250152039-7284-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar --- tools/perf/builtin-record.c | 2 +- tools/perf/util/parse-events.c | 10 ++++++++++ 2 files changed, 11 insertions(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index de76008fbaa2..78adc47da869 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -398,7 +398,7 @@ static void create_counter(int counter, int cpu, pid_t pid) PERF_FORMAT_TOTAL_TIME_RUNNING | PERF_FORMAT_ID; - attr->sample_type = PERF_SAMPLE_IP | PERF_SAMPLE_TID; + attr->sample_type |= PERF_SAMPLE_IP | PERF_SAMPLE_TID; if (freq) { attr->sample_type |= PERF_SAMPLE_PERIOD; diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index 4858d83b3b67..044178408783 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -379,6 +379,7 @@ static int parse_tracepoint_event(const char **strp, struct perf_counter_attr *attr) { const char *evt_name; + char *flags; char sys_name[MAX_EVENT_LENGTH]; char id_buf[4]; int fd; @@ -400,6 +401,15 @@ static int parse_tracepoint_event(const char **strp, strncpy(sys_name, *strp, sys_length); sys_name[sys_length] = '\0'; evt_name = evt_name + 1; + + flags = strchr(evt_name, ':'); + if (flags) { + *flags = '\0'; + flags++; + if (!strncmp(flags, "record", strlen(flags))) + attr->sample_type |= PERF_SAMPLE_RAW; + } + evt_length = strlen(evt_name); if (evt_length >= MAX_EVENT_LENGTH) return 0; -- cgit v1.2.3 From daac07b2e6b77f1bd44104aa2f0593e5505f27d4 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Thu, 13 Aug 2009 10:27:19 +0200 Subject: perf tools: Add a general option to enable raw sample records While we can enable the perf sample records per tracepoint counter, we may also want to enable this option for every tracepoint counters to open, so that we don't need to add a :record flag for all of them. Add the -R, --raw-samples options for this purpose. Signed-off-by: Frederic Weisbecker Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Mike Galbraith LKML-Reference: <1250152039-7284-2-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar --- tools/perf/builtin-record.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'tools') diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 78adc47da869..3d051b9cf25f 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -34,6 +34,7 @@ static int output; static const char *output_name = "perf.data"; static int group = 0; static unsigned int realtime_prio = 0; +static int raw_samples = 0; static int system_wide = 0; static int profile_cpu = -1; static pid_t target_pid = -1; @@ -418,6 +419,8 @@ static void create_counter(int counter, int cpu, pid_t pid) if (call_graph) attr->sample_type |= PERF_SAMPLE_CALLCHAIN; + if (raw_samples) + attr->sample_type |= PERF_SAMPLE_RAW; attr->mmap = track; attr->comm = track; @@ -650,6 +653,8 @@ static const struct option options[] = { "record events on existing pid"), OPT_INTEGER('r', "realtime", &realtime_prio, "collect data with this RT SCHED_FIFO priority"), + OPT_BOOLEAN('R', "raw-samples", &raw_samples, + "collect raw sample records from all opened counters"), OPT_BOOLEAN('a', "all-cpus", &system_wide, "system-wide collection from all CPUs"), OPT_BOOLEAN('A', "append", &append_file, -- cgit v1.2.3 From 8fd101f20bdf771949a8f3a5a779877d09b2fb56 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 12 Aug 2009 18:19:57 -0300 Subject: perf report: Don't show unresolved DSOs and symbols when -S/-d is used We're interested in just those symbols/DSOs, so filter out the unresolved ones. Signed-off-by: Arnaldo Carvalho de Melo Cc: Peter Zijlstra LKML-Reference: <20090812211957.GE3495@ghostprotocols.net> Signed-off-by: Ingo Molnar --- tools/perf/builtin-report.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'tools') diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index 23e1457f2409..b53a60fc12de 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -1590,10 +1590,11 @@ process_sample_event(event_t *event, unsigned long offset, unsigned long head) if (show & show_mask) { struct symbol *sym = resolve_symbol(thread, &map, &dso, &ip); - if (dso_list && dso && dso->name && !strlist__has_entry(dso_list, dso->name)) + if (dso_list && (!dso || !dso->name || + !strlist__has_entry(dso_list, dso->name))) return 0; - if (sym_list && sym && !strlist__has_entry(sym_list, sym->name)) + if (sym_list && (!sym || !strlist__has_entry(sym_list, sym->name))) return 0; if (hist_entry__add(thread, map, dso, sym, ip, chain, level, period)) { -- cgit v1.2.3 From 39e6dd73502f64e2ae3236b304e160ae30de9384 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Fri, 14 Aug 2009 15:26:32 -0300 Subject: perf record: Fix typo in pid_synthesize_comm_event We were using 'fd' locally, but there was a global 'fd' too, so when converting from open to fopen the test made against fd should be made against 'fp', but since we have that global it didnt get discovered ... Reported-by: Ulrich Drepper Signed-off-by: Arnaldo Carvalho de Melo Cc: Peter Zijlstra LKML-Reference: <20090814182632.GF3490@ghostprotocols.net> Signed-off-by: Ingo Molnar --- tools/perf/builtin-record.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c index 3d051b9cf25f..89a5ddcd1ded 100644 --- a/tools/perf/builtin-record.c +++ b/tools/perf/builtin-record.c @@ -219,7 +219,7 @@ static pid_t pid_synthesize_comm_event(pid_t pid, int full) snprintf(filename, sizeof(filename), "/proc/%d/status", pid); fp = fopen(filename, "r"); - if (fd == NULL) { + if (fp == NULL) { /* * We raced with a task exiting - just return: */ -- cgit v1.2.3 From 2932cffc89e9a1476b28a59896fa4f81e0d4f131 Mon Sep 17 00:00:00 2001 From: "Carlos R. Mafra" Date: Mon, 17 Aug 2009 00:36:21 +0200 Subject: perf: Rename perf-examples.txt to examples.txt Rename it to examples.txt to avoid the perf-*.txt pattern in the Makefile, otherwise 'make doc' fails because perf-examples.txt is not formatted to be a man page: ERROR: perf-examples.txt: line 1: manpage document title is mandatory Signed-off-by: Carlos R. Mafra Cc: Peter Zijlstra Cc: Paul Mackerras LKML-Reference: Signed-off-by: Ingo Molnar --- tools/perf/Documentation/examples.txt | 225 +++++++++++++++++++++++++++++ tools/perf/Documentation/perf-examples.txt | 225 ----------------------------- 2 files changed, 225 insertions(+), 225 deletions(-) create mode 100644 tools/perf/Documentation/examples.txt delete mode 100644 tools/perf/Documentation/perf-examples.txt (limited to 'tools') diff --git a/tools/perf/Documentation/examples.txt b/tools/perf/Documentation/examples.txt new file mode 100644 index 000000000000..8eb6c489fb15 --- /dev/null +++ b/tools/perf/Documentation/examples.txt @@ -0,0 +1,225 @@ + + ------------------------------ + ****** perf by examples ****** + ------------------------------ + +[ From an e-mail by Ingo Molnar, http://lkml.org/lkml/2009/8/4/346 ] + + +First, discovery/enumeration of available counters can be done via +'perf list': + +titan:~> perf list + [...] + kmem:kmalloc [Tracepoint event] + kmem:kmem_cache_alloc [Tracepoint event] + kmem:kmalloc_node [Tracepoint event] + kmem:kmem_cache_alloc_node [Tracepoint event] + kmem:kfree [Tracepoint event] + kmem:kmem_cache_free [Tracepoint event] + kmem:mm_page_free_direct [Tracepoint event] + kmem:mm_pagevec_free [Tracepoint event] + kmem:mm_page_alloc [Tracepoint event] + kmem:mm_page_alloc_zone_locked [Tracepoint event] + kmem:mm_page_pcpu_drain [Tracepoint event] + kmem:mm_page_alloc_extfrag [Tracepoint event] + +Then any (or all) of the above event sources can be activated and +measured. For example the page alloc/free properties of a 'hackbench +run' are: + + titan:~> perf stat -e kmem:mm_page_pcpu_drain -e kmem:mm_page_alloc + -e kmem:mm_pagevec_free -e kmem:mm_page_free_direct ./hackbench 10 + Time: 0.575 + + Performance counter stats for './hackbench 10': + + 13857 kmem:mm_page_pcpu_drain + 27576 kmem:mm_page_alloc + 6025 kmem:mm_pagevec_free + 20934 kmem:mm_page_free_direct + + 0.613972165 seconds time elapsed + +You can observe the statistical properties as well, by using the +'repeat the workload N times' feature of perf stat: + + titan:~> perf stat --repeat 5 -e kmem:mm_page_pcpu_drain -e + kmem:mm_page_alloc -e kmem:mm_pagevec_free -e + kmem:mm_page_free_direct ./hackbench 10 + Time: 0.627 + Time: 0.644 + Time: 0.564 + Time: 0.559 + Time: 0.626 + + Performance counter stats for './hackbench 10' (5 runs): + + 12920 kmem:mm_page_pcpu_drain ( +- 3.359% ) + 25035 kmem:mm_page_alloc ( +- 3.783% ) + 6104 kmem:mm_pagevec_free ( +- 0.934% ) + 18376 kmem:mm_page_free_direct ( +- 4.941% ) + + 0.643954516 seconds time elapsed ( +- 2.363% ) + +Furthermore, these tracepoints can be used to sample the workload as +well. For example the page allocations done by a 'git gc' can be +captured the following way: + + titan:~/git> perf record -f -e kmem:mm_page_alloc -c 1 ./git gc + Counting objects: 1148, done. + Delta compression using up to 2 threads. + Compressing objects: 100% (450/450), done. + Writing objects: 100% (1148/1148), done. + Total 1148 (delta 690), reused 1148 (delta 690) + [ perf record: Captured and wrote 0.267 MB perf.data (~11679 samples) ] + +To check which functions generated page allocations: + + titan:~/git> perf report + # Samples: 10646 + # + # Overhead Command Shared Object + # ........ ............... .......................... + # + 23.57% git-repack /lib64/libc-2.5.so + 21.81% git /lib64/libc-2.5.so + 14.59% git ./git + 11.79% git-repack ./git + 7.12% git /lib64/ld-2.5.so + 3.16% git-repack /lib64/libpthread-2.5.so + 2.09% git-repack /bin/bash + 1.97% rm /lib64/libc-2.5.so + 1.39% mv /lib64/ld-2.5.so + 1.37% mv /lib64/libc-2.5.so + 1.12% git-repack /lib64/ld-2.5.so + 0.95% rm /lib64/ld-2.5.so + 0.90% git-update-serv /lib64/libc-2.5.so + 0.73% git-update-serv /lib64/ld-2.5.so + 0.68% perf /lib64/libpthread-2.5.so + 0.64% git-repack /usr/lib64/libz.so.1.2.3 + +Or to see it on a more finegrained level: + +titan:~/git> perf report --sort comm,dso,symbol +# Samples: 10646 +# +# Overhead Command Shared Object Symbol +# ........ ............... .......................... ...... +# + 9.35% git-repack ./git [.] insert_obj_hash + 9.12% git ./git [.] insert_obj_hash + 7.31% git /lib64/libc-2.5.so [.] memcpy + 6.34% git-repack /lib64/libc-2.5.so [.] _int_malloc + 6.24% git-repack /lib64/libc-2.5.so [.] memcpy + 5.82% git-repack /lib64/libc-2.5.so [.] __GI___fork + 5.47% git /lib64/libc-2.5.so [.] _int_malloc + 2.99% git /lib64/libc-2.5.so [.] memset + +Furthermore, call-graph sampling can be done too, of page +allocations - to see precisely what kind of page allocations there +are: + + titan:~/git> perf record -f -g -e kmem:mm_page_alloc -c 1 ./git gc + Counting objects: 1148, done. + Delta compression using up to 2 threads. + Compressing objects: 100% (450/450), done. + Writing objects: 100% (1148/1148), done. + Total 1148 (delta 690), reused 1148 (delta 690) + [ perf record: Captured and wrote 0.963 MB perf.data (~42069 samples) ] + + titan:~/git> perf report -g + # Samples: 10686 + # + # Overhead Command Shared Object + # ........ ............... .......................... + # + 23.25% git-repack /lib64/libc-2.5.so + | + |--50.00%-- _int_free + | + |--37.50%-- __GI___fork + | make_child + | + |--12.50%-- ptmalloc_unlock_all2 + | make_child + | + --6.25%-- __GI_strcpy + 21.61% git /lib64/libc-2.5.so + | + |--30.00%-- __GI_read + | | + | --83.33%-- git_config_from_file + | git_config + | | + [...] + +Or you can observe the whole system's page allocations for 10 +seconds: + +titan:~/git> perf stat -a -e kmem:mm_page_pcpu_drain -e +kmem:mm_page_alloc -e kmem:mm_pagevec_free -e +kmem:mm_page_free_direct sleep 10 + + Performance counter stats for 'sleep 10': + + 171585 kmem:mm_page_pcpu_drain + 322114 kmem:mm_page_alloc + 73623 kmem:mm_pagevec_free + 254115 kmem:mm_page_free_direct + + 10.000591410 seconds time elapsed + +Or observe how fluctuating the page allocations are, via statistical +analysis done over ten 1-second intervals: + + titan:~/git> perf stat --repeat 10 -a -e kmem:mm_page_pcpu_drain -e + kmem:mm_page_alloc -e kmem:mm_pagevec_free -e + kmem:mm_page_free_direct sleep 1 + + Performance counter stats for 'sleep 1' (10 runs): + + 17254 kmem:mm_page_pcpu_drain ( +- 3.709% ) + 34394 kmem:mm_page_alloc ( +- 4.617% ) + 7509 kmem:mm_pagevec_free ( +- 4.820% ) + 25653 kmem:mm_page_free_direct ( +- 3.672% ) + + 1.058135029 seconds time elapsed ( +- 3.089% ) + +Or you can annotate the recorded 'git gc' run on a per symbol basis +and check which instructions/source-code generated page allocations: + + titan:~/git> perf annotate __GI___fork + ------------------------------------------------ + Percent | Source code & Disassembly of libc-2.5.so + ------------------------------------------------ + : + : + : Disassembly of section .plt: + : Disassembly of section .text: + : + : 00000031a2e95560 <__fork>: + [...] + 0.00 : 31a2e95602: b8 38 00 00 00 mov $0x38,%eax + 0.00 : 31a2e95607: 0f 05 syscall + 83.42 : 31a2e95609: 48 3d 00 f0 ff ff cmp $0xfffffffffffff000,%rax + 0.00 : 31a2e9560f: 0f 87 4d 01 00 00 ja 31a2e95762 <__fork+0x202> + 0.00 : 31a2e95615: 85 c0 test %eax,%eax + +( this shows that 83.42% of __GI___fork's page allocations come from + the 0x38 system call it performs. ) + +etc. etc. - a lot more is possible. I could list a dozen of +other different usecases straight away - neither of which is +possible via /proc/vmstat. + +/proc/vmstat is not in the same league really, in terms of +expressive power of system analysis and performance +analysis. + +All that the above results needed were those new tracepoints +in include/tracing/events/kmem.h. + + Ingo + + diff --git a/tools/perf/Documentation/perf-examples.txt b/tools/perf/Documentation/perf-examples.txt deleted file mode 100644 index 8eb6c489fb15..000000000000 --- a/tools/perf/Documentation/perf-examples.txt +++ /dev/null @@ -1,225 +0,0 @@ - - ------------------------------ - ****** perf by examples ****** - ------------------------------ - -[ From an e-mail by Ingo Molnar, http://lkml.org/lkml/2009/8/4/346 ] - - -First, discovery/enumeration of available counters can be done via -'perf list': - -titan:~> perf list - [...] - kmem:kmalloc [Tracepoint event] - kmem:kmem_cache_alloc [Tracepoint event] - kmem:kmalloc_node [Tracepoint event] - kmem:kmem_cache_alloc_node [Tracepoint event] - kmem:kfree [Tracepoint event] - kmem:kmem_cache_free [Tracepoint event] - kmem:mm_page_free_direct [Tracepoint event] - kmem:mm_pagevec_free [Tracepoint event] - kmem:mm_page_alloc [Tracepoint event] - kmem:mm_page_alloc_zone_locked [Tracepoint event] - kmem:mm_page_pcpu_drain [Tracepoint event] - kmem:mm_page_alloc_extfrag [Tracepoint event] - -Then any (or all) of the above event sources can be activated and -measured. For example the page alloc/free properties of a 'hackbench -run' are: - - titan:~> perf stat -e kmem:mm_page_pcpu_drain -e kmem:mm_page_alloc - -e kmem:mm_pagevec_free -e kmem:mm_page_free_direct ./hackbench 10 - Time: 0.575 - - Performance counter stats for './hackbench 10': - - 13857 kmem:mm_page_pcpu_drain - 27576 kmem:mm_page_alloc - 6025 kmem:mm_pagevec_free - 20934 kmem:mm_page_free_direct - - 0.613972165 seconds time elapsed - -You can observe the statistical properties as well, by using the -'repeat the workload N times' feature of perf stat: - - titan:~> perf stat --repeat 5 -e kmem:mm_page_pcpu_drain -e - kmem:mm_page_alloc -e kmem:mm_pagevec_free -e - kmem:mm_page_free_direct ./hackbench 10 - Time: 0.627 - Time: 0.644 - Time: 0.564 - Time: 0.559 - Time: 0.626 - - Performance counter stats for './hackbench 10' (5 runs): - - 12920 kmem:mm_page_pcpu_drain ( +- 3.359% ) - 25035 kmem:mm_page_alloc ( +- 3.783% ) - 6104 kmem:mm_pagevec_free ( +- 0.934% ) - 18376 kmem:mm_page_free_direct ( +- 4.941% ) - - 0.643954516 seconds time elapsed ( +- 2.363% ) - -Furthermore, these tracepoints can be used to sample the workload as -well. For example the page allocations done by a 'git gc' can be -captured the following way: - - titan:~/git> perf record -f -e kmem:mm_page_alloc -c 1 ./git gc - Counting objects: 1148, done. - Delta compression using up to 2 threads. - Compressing objects: 100% (450/450), done. - Writing objects: 100% (1148/1148), done. - Total 1148 (delta 690), reused 1148 (delta 690) - [ perf record: Captured and wrote 0.267 MB perf.data (~11679 samples) ] - -To check which functions generated page allocations: - - titan:~/git> perf report - # Samples: 10646 - # - # Overhead Command Shared Object - # ........ ............... .......................... - # - 23.57% git-repack /lib64/libc-2.5.so - 21.81% git /lib64/libc-2.5.so - 14.59% git ./git - 11.79% git-repack ./git - 7.12% git /lib64/ld-2.5.so - 3.16% git-repack /lib64/libpthread-2.5.so - 2.09% git-repack /bin/bash - 1.97% rm /lib64/libc-2.5.so - 1.39% mv /lib64/ld-2.5.so - 1.37% mv /lib64/libc-2.5.so - 1.12% git-repack /lib64/ld-2.5.so - 0.95% rm /lib64/ld-2.5.so - 0.90% git-update-serv /lib64/libc-2.5.so - 0.73% git-update-serv /lib64/ld-2.5.so - 0.68% perf /lib64/libpthread-2.5.so - 0.64% git-repack /usr/lib64/libz.so.1.2.3 - -Or to see it on a more finegrained level: - -titan:~/git> perf report --sort comm,dso,symbol -# Samples: 10646 -# -# Overhead Command Shared Object Symbol -# ........ ............... .......................... ...... -# - 9.35% git-repack ./git [.] insert_obj_hash - 9.12% git ./git [.] insert_obj_hash - 7.31% git /lib64/libc-2.5.so [.] memcpy - 6.34% git-repack /lib64/libc-2.5.so [.] _int_malloc - 6.24% git-repack /lib64/libc-2.5.so [.] memcpy - 5.82% git-repack /lib64/libc-2.5.so [.] __GI___fork - 5.47% git /lib64/libc-2.5.so [.] _int_malloc - 2.99% git /lib64/libc-2.5.so [.] memset - -Furthermore, call-graph sampling can be done too, of page -allocations - to see precisely what kind of page allocations there -are: - - titan:~/git> perf record -f -g -e kmem:mm_page_alloc -c 1 ./git gc - Counting objects: 1148, done. - Delta compression using up to 2 threads. - Compressing objects: 100% (450/450), done. - Writing objects: 100% (1148/1148), done. - Total 1148 (delta 690), reused 1148 (delta 690) - [ perf record: Captured and wrote 0.963 MB perf.data (~42069 samples) ] - - titan:~/git> perf report -g - # Samples: 10686 - # - # Overhead Command Shared Object - # ........ ............... .......................... - # - 23.25% git-repack /lib64/libc-2.5.so - | - |--50.00%-- _int_free - | - |--37.50%-- __GI___fork - | make_child - | - |--12.50%-- ptmalloc_unlock_all2 - | make_child - | - --6.25%-- __GI_strcpy - 21.61% git /lib64/libc-2.5.so - | - |--30.00%-- __GI_read - | | - | --83.33%-- git_config_from_file - | git_config - | | - [...] - -Or you can observe the whole system's page allocations for 10 -seconds: - -titan:~/git> perf stat -a -e kmem:mm_page_pcpu_drain -e -kmem:mm_page_alloc -e kmem:mm_pagevec_free -e -kmem:mm_page_free_direct sleep 10 - - Performance counter stats for 'sleep 10': - - 171585 kmem:mm_page_pcpu_drain - 322114 kmem:mm_page_alloc - 73623 kmem:mm_pagevec_free - 254115 kmem:mm_page_free_direct - - 10.000591410 seconds time elapsed - -Or observe how fluctuating the page allocations are, via statistical -analysis done over ten 1-second intervals: - - titan:~/git> perf stat --repeat 10 -a -e kmem:mm_page_pcpu_drain -e - kmem:mm_page_alloc -e kmem:mm_pagevec_free -e - kmem:mm_page_free_direct sleep 1 - - Performance counter stats for 'sleep 1' (10 runs): - - 17254 kmem:mm_page_pcpu_drain ( +- 3.709% ) - 34394 kmem:mm_page_alloc ( +- 4.617% ) - 7509 kmem:mm_pagevec_free ( +- 4.820% ) - 25653 kmem:mm_page_free_direct ( +- 3.672% ) - - 1.058135029 seconds time elapsed ( +- 3.089% ) - -Or you can annotate the recorded 'git gc' run on a per symbol basis -and check which instructions/source-code generated page allocations: - - titan:~/git> perf annotate __GI___fork - ------------------------------------------------ - Percent | Source code & Disassembly of libc-2.5.so - ------------------------------------------------ - : - : - : Disassembly of section .plt: - : Disassembly of section .text: - : - : 00000031a2e95560 <__fork>: - [...] - 0.00 : 31a2e95602: b8 38 00 00 00 mov $0x38,%eax - 0.00 : 31a2e95607: 0f 05 syscall - 83.42 : 31a2e95609: 48 3d 00 f0 ff ff cmp $0xfffffffffffff000,%rax - 0.00 : 31a2e9560f: 0f 87 4d 01 00 00 ja 31a2e95762 <__fork+0x202> - 0.00 : 31a2e95615: 85 c0 test %eax,%eax - -( this shows that 83.42% of __GI___fork's page allocations come from - the 0x38 system call it performs. ) - -etc. etc. - a lot more is possible. I could list a dozen of -other different usecases straight away - neither of which is -possible via /proc/vmstat. - -/proc/vmstat is not in the same league really, in terms of -expressive power of system analysis and performance -analysis. - -All that the above results needed were those new tracepoints -in include/tracing/events/kmem.h. - - Ingo - - -- cgit v1.2.3 From 15f3fa4e7f608c5ce19187b3b4a953222fdfa751 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 18 Aug 2009 13:52:28 +0200 Subject: perf annotate: Fix segmentation fault Linus reported this perf annotate segfault: [torvalds@nehalem git]$ perf annotate unmap_vmas Segmentation fault #0 map__clone (self=) at builtin-annotate.c:236 #1 thread__fork (self=) at builtin-annotate.c:372 The bug here was that builtin-annotate.c was a copy of builtin-report.c and a threading related fix to builtin-report.c didnt get propagated to builtin-annotate.c ... Reported-by: Linus Torvalds Cc: Peter Zijlstra Cc: Mike Galbraith Cc: Paul Mackerras Cc: Arnaldo Carvalho de Melo Cc: Frederic Weisbecker LKML-Reference: Signed-off-by: Ingo Molnar --- tools/perf/builtin-annotate.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'tools') diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c index 1dba568e1941..343e7b14bf01 100644 --- a/tools/perf/builtin-annotate.c +++ b/tools/perf/builtin-annotate.c @@ -980,6 +980,13 @@ process_fork_event(event_t *event, unsigned long offset, unsigned long head) (void *)(long)(event->header.size), event->fork.pid, event->fork.ppid); + /* + * A thread clone will have the same PID for both + * parent and child. + */ + if (thread == parent) + return 0; + if (!thread || !parent || thread__fork(thread, parent)) { dprintf("problem processing PERF_EVENT_FORK, skipping event.\n"); return -1; -- cgit v1.2.3 From b395cd8a74b4a8d943dd4b5585e676f62f7350b3 Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Tue, 18 Aug 2009 12:41:25 -0400 Subject: perf tools: Make 'make html' work pushd tools/perf/Documentation make html popd is failing for me... ASCIIDOC perf-annotate.html ERROR: unsafe: include file: /etc/asciidoc/./stylesheets/xhtml11.css ERROR: unsafe: include file: /etc/asciidoc/./stylesheets/xhtml11-manpage.css ERROR: unsafe: include file: /etc/asciidoc/./stylesheets/xhtml11-quirks.css make: *** [perf-annotate.html] Error 1 Apparently asciidoc "unsafe" is the default mode of operation in practice. https://bugzilla.redhat.com/show_bug.cgi?id=506953 Works tidily now. Signed-off-by: Kyle McMartin Cc: Peter Zijlstra Cc: Mike Galbraith Cc: Paul Mackerras Cc: Arnaldo Carvalho de Melo Cc: Frederic Weisbecker LKML-Reference: <20090818164125.GM25206@bombadil.infradead.org> Signed-off-by: Ingo Molnar --- tools/perf/Documentation/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'tools') diff --git a/tools/perf/Documentation/Makefile b/tools/perf/Documentation/Makefile index 5457192e1b41..bdd3b7ecad0a 100644 --- a/tools/perf/Documentation/Makefile +++ b/tools/perf/Documentation/Makefile @@ -35,7 +35,7 @@ man7dir=$(mandir)/man7 # DESTDIR= ASCIIDOC=asciidoc -ASCIIDOC_EXTRA = +ASCIIDOC_EXTRA = --unsafe MANPAGE_XSL = manpage-normal.xsl XMLTO_EXTRA = INSTALL?=install -- cgit v1.2.3 From fa6963b2481beff8b11f76006fbb63fdbbf2d2d7 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 19 Aug 2009 11:18:26 +0200 Subject: perf tools: Check perf.data owner Add an owner check to opening perf.data files and a switch to silence it. Because perf-report/perf-annotate are binary parsers reading another users' perf.data file could be a security risk if the file were explicitly engineered to trigger bugs in the parser (we hope of course there are non such bugs, but you never know). Signed-off-by: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Frederic Weisbecker Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Paul Mackerras LKML-Reference: <20090819092023.896648538@chello.nl> Signed-off-by: Ingo Molnar --- tools/perf/builtin-annotate.c | 7 +++++++ tools/perf/builtin-report.c | 7 +++++++ 2 files changed, 14 insertions(+) (limited to 'tools') diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c index 343e7b14bf01..5e17de984dc8 100644 --- a/tools/perf/builtin-annotate.c +++ b/tools/perf/builtin-annotate.c @@ -31,6 +31,7 @@ static char *vmlinux = "vmlinux"; static char default_sort_order[] = "comm,symbol"; static char *sort_order = default_sort_order; +static int force; static int input; static int show_mask = SHOW_KERNEL | SHOW_USER | SHOW_HV; @@ -1334,6 +1335,11 @@ static int __cmd_annotate(void) exit(-1); } + if (!force && (stat.st_uid != geteuid())) { + fprintf(stderr, "file: %s not owned by current user\n", input_name); + exit(-1); + } + if (!stat.st_size) { fprintf(stderr, "zero-sized file, nothing to do!\n"); exit(0); @@ -1439,6 +1445,7 @@ static const struct option options[] = { "input file name"), OPT_STRING('s', "symbol", &sym_hist_filter, "symbol", "symbol to annotate"), + OPT_BOOLEAN('f', "force", &force, "don't complain, do it"), OPT_BOOLEAN('v', "verbose", &verbose, "be more verbose (show symbol address, etc)"), OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c index b53a60fc12de..8b2ec882e6e0 100644 --- a/tools/perf/builtin-report.c +++ b/tools/perf/builtin-report.c @@ -38,6 +38,7 @@ static char *dso_list_str, *comm_list_str, *sym_list_str, static struct strlist *dso_list, *comm_list, *sym_list; static char *field_sep; +static int force; static int input; static int show_mask = SHOW_KERNEL | SHOW_USER | SHOW_HV; @@ -1856,6 +1857,11 @@ static int __cmd_report(void) exit(-1); } + if (!force && (stat.st_uid != geteuid())) { + fprintf(stderr, "file: %s not owned by current user\n", input_name); + exit(-1); + } + if (!stat.st_size) { fprintf(stderr, "zero-sized file, nothing to do!\n"); exit(0); @@ -2064,6 +2070,7 @@ static const struct option options[] = { OPT_BOOLEAN('D', "dump-raw-trace", &dump_trace, "dump raw trace in ASCII"), OPT_STRING('k', "vmlinux", &vmlinux, "file", "vmlinux pathname"), + OPT_BOOLEAN('f', "force", &force, "don't complain, do it"), OPT_BOOLEAN('m', "modules", &modules, "load module symbols - WARNING: use only with -k and LIVE kernel"), OPT_BOOLEAN('n', "show-nr-samples", &show_nr_samples, -- cgit v1.2.3