perf/record improvements and fixes
Arnaldo Carvalho de Melo
maintainerHi Ingo,
Please consider pulling,
- Arnaldo
The following changes since commit 8a26ce4e544659256349551283414df504889a59:
Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core (2015-02-18 19:18:18 +0100)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo
for you to fetch changes up to 54cf776a9c5c2e6a91de31954bba4d3bad6c657c:
perf data: Add a 'perf' prefix to the generic fields (2015-02-25 16:14:33 -0300)
----------------------------------------------------------------
perf/core improvements and fixes:
New user selectable features:
- Support recording running/enabled time in 'perf record' (Andi Kleen)
- New tool: 'perf data' for converting perf.data to other formats, initially for the CTF (Common Trace Format) from LTTng (Jiri Olsa, Sebastian Siewior)
User visible:
- Only insert blank duration bracket when tracing syscalls in 'perf trace' (Arnaldo Carvalho de Melo)
- Filter out the trace pid when no threads are specified in 'perf trace' (Arnaldo Carvalho de Melo)
- Add 'perf trace' man page entry for --event (Arnaldo Carvalho de Melo)
- Dump stack on segfaults in 'perf trace' (Arnaldo Carvalho de Melo)
Infrastructure:
- Introduce set_filter_pid and set_filter_pids methods in the evlist class (Arnaldo Carvalho de Melo)
- Some perf_session untanglement patches, removing the need to pass a perf_session instance for things that are related to evlists, so that tools that don't deal with perf.data files like trace in live mode can make use of the ordered_events class (Arnaldo Carvalho de Melo)
----------------------------------------------------------------
Andi Kleen (1):
perf record: Support recording running/enabled time
Arnaldo Carvalho de Melo (16):
perf trace: Only insert blank duration bracket when tracing syscalls perf evlist: Introduce set_filter_pid method
perf trace: Filter out the trace pid when no threads are specified perf evlist: Introduce set_filter_pids method
perf trace: Introduce --filter-pids
perf trace: Add man page entry for --event
perf trace: Separate routine that handles an event from the one that reads it perf session: Remove wrappers to machines__find
perf evlist: Adopt events_stats from perf_session
perf session: Remove perf_session from warn_errors signature perf session: Remove perf_session from some deliver event routines perf session: Remove perf_session from dump_event
perf ordered_events: Stop using tool->ordered_events
perf tools: Introduce dump_stack signal helper
perf trace: Dump stack on segfaults
perf tools: Print the thread's tid on PERF_RECORD_COMM events when -D is asked
Jiri Olsa (3):
perf tools: Add feature check for libbabeltrace perf tools: Add new 'perf data' command perf data: Add perf data to CTF conversion support
Sebastian Andrzej Siewior (1):
perf data: Add a 'perf' prefix to the generic fields
tools/perf/Build | 1 + tools/perf/Documentation/perf-data.txt | 40 ++ tools/perf/Documentation/perf-record.txt | 3 + tools/perf/Documentation/perf-trace.txt | 6 + tools/perf/Documentation/perf.txt | 7 +- tools/perf/Makefile.perf | 4 +- tools/perf/builtin-data.c | 119 ++++ tools/perf/builtin-record.c | 2 + tools/perf/builtin-sched.c | 6 +- tools/perf/builtin-top.c | 4 +- tools/perf/builtin-trace.c | 121 +++- tools/perf/builtin.h | 1 + tools/perf/command-list.txt | 1 + tools/perf/config/Makefile | 25 + tools/perf/config/feature-checks/Makefile | 8 +- tools/perf/config/feature-checks/test-all.c | 5 + .../config/feature-checks/test-libbabeltrace.c | 8 + tools/perf/perf.c | 1 + tools/perf/perf.h | 1 + tools/perf/util/Build | 2 + tools/perf/util/data-convert-bt.c | 614 +++++++++++++++++++++ tools/perf/util/data-convert-bt.h | 8 + tools/perf/util/debug.c | 2 + tools/perf/util/debug.h | 1 + tools/perf/util/event.c | 2 +- tools/perf/util/evlist.c | 32 ++ tools/perf/util/evlist.h | 3 + tools/perf/util/evsel.c | 6 + tools/perf/util/ordered-events.c | 5 +- tools/perf/util/session.c | 115 ++-- tools/perf/util/session.h | 1 - tools/perf/util/util.c | 7 + tools/perf/util/util.h | 1 + 33 files changed, 1063 insertions(+), 99 deletions(-) create mode 100644 tools/perf/Documentation/perf-data.txt create mode 100644 tools/perf/builtin-data.c create mode 100644 tools/perf/config/feature-checks/test-libbabeltrace.c create mode 100644 tools/perf/util/data-convert-bt.c create mode 100644 tools/perf/util/data-convert-bt.h
Arnaldo Carvalho de Melo
maintainerPATCH 1/21 perf trace: Only insert blank duration bracket when tracing syscalls
From: Arnaldo Carvalho de Melo <[email protected]>
When printing just events, i.e. '--no-sys --ev some:events' it makes no sense to waste screen space.
Before:
# trace --no-sys --ev probe:* 84481.704 ( ): probe:vfs_getname:(ffffffff811ed023) pathname="/etc/services") 84481.892 ( ): probe:vfs_getname:(ffffffff811ed023) pathname="/etc/services") 84482.230 ( ): probe:vfs_getname:(ffffffff811ed023) pathname="/etc/resolv.conf") 84482.481 ( ): probe:vfs_getname:(ffffffff811ed023) pathname="/etc/hosts") 85097.725 ( ): probe:vfs_getname:(ffffffff811ed023) pathname="/root" #
After:
# trace --no-sys --ev probe:* 0.000 probe:vfs_getname:(ffffffff811ed023) pathname="/root") 1.711 probe:vfs_getname:(ffffffff811ed023) pathname="/etc/localtime") 2.103 probe:vfs_getname:(ffffffff811ed023) pathname="/etc/localtime") ^C#
--- tools/perf/builtin-trace.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
tools/perf/builtin-trace.c +5 -1
@@ -1840,7 +1840,11 @@ static int trace__event_handler(struct trace *trace, struct perf_evsel *evsel, {trace__printf_interrupted_entry (trace, sample);trace__fprintf_tstamp (trace, sample->time, trace->output); -fprintf (trace->output,"(%9.9s): %s:" ," " , evsel->name); + +if (trace->trace_syscalls) +fprintf (trace->output,"( ): " ); + +fprintf (trace->output,"%s:" , evsel->name);if (evsel->tp_format) { event_format__fprintf(evsel->tp_format, sample->cpu,
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 2/21 perf evlist: Introduce set_filter_pid method
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
To filter out events for a certain pid, for instance, when tracing system wide, so that the tracer itself doesn't creates an event loop.
--- tools/perf/util/evlist.c | 13 +++++++++++++ tools/perf/util/evlist.h | 1 + 2 files changed, 14 insertions(+)
tools/perf/util/evlist.c +13 -0
@@ -1085,6 +1085,19 @@ int perf_evlist__set_filter(struct perf_evlist *evlist, const char *filter)return err; } +int perf_evlist__set_filter_pid (struct perf_evlist *evlist,pid_t pid) +{ +char *filter; +int ret; + +if (asprintf (&filter,"common_pid != %d" , pid) <0 ) +return -1 ; + + ret =perf_evlist__set_filter (evlist, filter); +free (filter); +return ret; +} +bool perf_evlist__valid_sample_type (struct perf_evlist *evlist) {struct perf_evsel *pos;
tools/perf/util/evlist.h +1 -0
@@ -77,6 +77,7 @@ int perf_evlist__add_newtp(struct perf_evlist *evlist, const char *sys, const char *name, void *handler);int perf_evlist__set_filter (struct perf_evlist *evlist,const char *filter); +int perf_evlist__set_filter_pid (struct perf_evlist *evlist,pid_t pid);struct perf_evsel *perf_evlist__find_tracepoint_by_id (struct perf_evlist *evlist,int id);
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 3/21 perf trace: Filter out the trace pid when no threads are specified
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
To avoid tracing the tracer.
--- tools/perf/builtin-trace.c | 9 +++++++++ 1 file changed, 9 insertions(+)
tools/perf/builtin-trace.c +9 -0
@@ -2151,6 +2151,15 @@ static int trace__run(struct trace *trace, int argc, const char **argv)if (err <0 ) goto out_error_open; +/* +* Better not use !target__has_task() here because we need to cover the +* case where no threads were specified in the command line, but a +* workload was, and in that case we will fill in the thread_map when +* we fork the workload in perf_evlist__prepare_workload. +*/ +if (evlist->threads->map[0 ] ==-1 ) +perf_evlist__set_filter_pid (evlist,getpid ()); + err =perf_evlist__mmap (evlist, trace->opts.mmap_pages, false);if (err <0 ) goto out_error_mmap;
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 4/21 perf evlist: Introduce set_filter_pids method
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
We need to filter multiple pids in trace, i.e. trace itself, gnome-terminal, X.org, etc.
--- tools/perf/util/evlist.c | 27 +++++++++++++++++++++++---- tools/perf/util/evlist.h | 1 + 2 files changed, 24 insertions(+), 4 deletions(-)
tools/perf/util/evlist.c +23 -4
@@ -1085,19 +1085,38 @@ int perf_evlist__set_filter(struct perf_evlist *evlist, const char *filter)return err; } -int perf_evlist__set_filter_pid (struct perf_evlist *evlist,pid_t pid) +int perf_evlist__set_filter_pids (struct perf_evlist *evlist,size_t npids,pid_t *pids) {char *filter; -int ret; +int ret =-1 ; +size_t i; -if (asprintf (&filter,"common_pid != %d" , pid) <0 ) -return -1 ; +for (i =0 ; i < npids; ++i) { +if (i ==0 ) { +if (asprintf (&filter,"common_pid != %d" , pids[i]) <0 ) +return -1 ; + }else { +char *tmp; + +if (asprintf (&tmp,"%s && common_pid != %d" , filter, pids[i]) <0 ) + goto out_free; + +free (filter); + filter = tmp; + } + } ret =perf_evlist__set_filter (evlist, filter); +out_free:free (filter);return ret; } +int perf_evlist__set_filter_pid (struct perf_evlist *evlist,pid_t pid) +{ +return perf_evlist__set_filter_pids (evlist,1 , &pid); +} +bool perf_evlist__valid_sample_type (struct perf_evlist *evlist) {struct perf_evsel *pos;
tools/perf/util/evlist.h +1 -0
@@ -78,6 +78,7 @@ int perf_evlist__add_newtp(struct perf_evlist *evlist,int perf_evlist__set_filter (struct perf_evlist *evlist,const char *filter);int perf_evlist__set_filter_pid (struct perf_evlist *evlist,pid_t pid); +int perf_evlist__set_filter_pids (struct perf_evlist *evlist,size_t npids,pid_t *pids);struct perf_evsel *perf_evlist__find_tracepoint_by_id (struct perf_evlist *evlist,int id);
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 6/21 perf trace: Add man page entry for --event
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
Forgot to do it when adding the feature.
--- tools/perf/Documentation/perf-trace.txt | 3 +++ 1 file changed, 3 insertions(+)
tools/perf/Documentation/perf-trace.txt +3 -0
@@ -118,6 +118,9 @@ the thread executes on the designated CPUs. Default is to monitor all CPUs. --syscalls:: Trace system calls. This options is enabled by default. +--event:: + Trace other events, see 'perf list' for a complete list. + PAGEFAULTS ----------
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 7/21 perf trace: Separate routine that handles an event from the one that reads it
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
Because we need to use ordered_events in some cases, so we will need to first have them in a queue, order that queue, and then process the event.
--- tools/perf/builtin-trace.c | 58 +++++++++++++++++++++++++--------------------- 1 file changed, 31 insertions(+), 27 deletions(-)
tools/perf/builtin-trace.c +31 -27
@@ -2092,10 +2092,39 @@ static int perf_evlist__add_pgfault(struct perf_evlist *evlist,return 0 ; } +static void trace__handle_event (struct trace *trace,union perf_event *event,struct perf_sample *sample) +{ +const u32 type = event->header.type; +struct perf_evsel *evsel; + +if (!trace->full_time && trace->base_time ==0 ) + trace->base_time = sample->time; + +if (type !=PERF_RECORD_SAMPLE ) { +trace__process_event (trace, trace->host, event, sample); +return ; + } + + evsel =perf_evlist__id2evsel (trace->evlist, sample->id); +if (evsel ==NULL ) { +fprintf (trace->output,"Unknown tp ID %" PRIu64", skipping...\n" , sample->id); +return ; + } + +if (evsel->attr.type ==PERF_TYPE_TRACEPOINT && + sample->raw_data ==NULL ) { +fprintf (trace->output,"%s sample with no payload for tid: %d, cpu %d, raw_size=%d, skipping...\n" , +perf_evsel__name (evsel), sample->tid, + sample->cpu, sample->raw_size); + }else { +tracepoint_handler handler = evsel->handler; +handler (trace, evsel, event, sample); + } +} +static int trace__run (struct trace *trace,int argc,const char **argv) {struct perf_evlist *evlist = trace->evlist; -struct perf_evsel *evsel;int err =-1 , i;unsigned long before;const bool forks = argc >0 ;
@@ -2190,8 +2219,6 @@ again:union perf_event *event;while ((event =perf_evlist__mmap_read (evlist, i)) !=NULL ) { -const u32 type = event->header.type; -tracepoint_handler handler;struct perf_sample sample; ++trace->nr_events;
@@ -2202,30 +2229,7 @@ again: goto next_event; } -if (!trace->full_time && trace->base_time ==0 ) - trace->base_time = sample.time; - -if (type !=PERF_RECORD_SAMPLE ) { -trace__process_event (trace, trace->host, event, &sample); -continue ; - } - - evsel =perf_evlist__id2evsel (evlist, sample.id); -if (evsel ==NULL ) { -fprintf (trace->output,"Unknown tp ID %" PRIu64", skipping...\n" , sample.id); - goto next_event; - } - -if (evsel->attr.type ==PERF_TYPE_TRACEPOINT && - sample.raw_data ==NULL ) { -fprintf (trace->output,"%s sample with no payload for tid: %d, cpu %d, raw_size=%d, skipping...\n" , -perf_evsel__name (evsel), sample.tid, - sample.cpu, sample.raw_size); - goto next_event; - } - - handler = evsel->handler; -handler (trace, evsel, event, &sample); +trace__handle_event (trace, event, &sample); next_event:perf_evlist__mmap_consume (evlist, i);
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 8/21 perf session: Remove wrappers to machines__find
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
Start to untangle session from delivering samples, as there are tools that want to use ordered_events and don't use perf_session at all.
--- tools/perf/util/session.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-)
tools/perf/util/session.c +5 -8
@@ -797,8 +797,7 @@ static void dump_sample(struct perf_evsel *evsel, union perf_event *event,sample_read__printf (sample, evsel->attr.read_format); } -static struct machine * -perf_session__find_machine_for_cpumode (struct perf_session *session, +static struct machine *machines__find_for_cpumode (struct machines *machines,union perf_event *event,struct perf_sample *sample) {
@@ -816,14 +815,13 @@ static struct machine *else pid = sample->pid; - machine =perf_session__find_machine (session, pid); + machine =machines__find (machines, pid);if (!machine) - machine =perf_session__findnew_machine (session, -DEFAULT_GUEST_KERNEL_ID ); + machine =machines__find (machines,DEFAULT_GUEST_KERNEL_ID );return machine; } -return &session->machines.host; +return &machines->host; } static int deliver_sample_value(struct perf_session *session,
@@ -907,8 +905,7 @@ int perf_session__deliver_event(struct perf_session *session, evsel =perf_evlist__id2evsel (session->evlist, sample->id); - machine =perf_session__find_machine_for_cpumode (session, event, - sample); + machine =machines__find_for_cpumode (&session->machines, event, sample);switch (event->header.type) {case PERF_RECORD_SAMPLE :
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 9/21 perf evlist: Adopt events_stats from perf_session
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
For tools that don't deal with perf.data files, thus do not need to use perf_session.
--- tools/perf/builtin-sched.c | 6 +++--- tools/perf/builtin-top.c | 4 ++-- tools/perf/util/evlist.h | 1 + tools/perf/util/session.c | 54 +++++++++++++++++++++++----------------------- tools/perf/util/session.h | 1 - 5 files changed, 33 insertions(+), 33 deletions(-)
tools/perf/builtin-sched.c +3 -3
@@ -1473,9 +1473,9 @@ static int perf_sched__read_events(struct perf_sched *sched, goto out_delete; } - sched->nr_events = session->stats.nr_events[0 ]; - sched->nr_lost_events = session->stats.total_lost; - sched->nr_lost_chunks = session->stats.nr_events[PERF_RECORD_LOST ]; + sched->nr_events = session->evlist->stats.nr_events[0 ]; + sched->nr_lost_events = session->evlist->stats.total_lost; + sched->nr_lost_chunks = session->evlist->stats.nr_events[PERF_RECORD_LOST ]; }if (psession)
tools/perf/builtin-top.c +2 -2
@@ -716,7 +716,7 @@ static void perf_event__process_sample(struct perf_tool *tool,if (!machine) {pr_err ("%u unprocessable samples recorded.\r" , - top->session->stats.nr_unprocessable_samples++); + top->session->evlist->stats.nr_unprocessable_samples++);return ; }
@@ -856,7 +856,7 @@ static void perf_top__mmap_read_idx(struct perf_top *top, int idx)hists__inc_nr_events (evsel__hists (evsel), event->header.type);machine__process_event (machine, event, &sample); }else - ++session->stats.nr_unknown_events; + ++session->evlist->stats.nr_unknown_events; next_event:perf_evlist__mmap_consume (top->evlist, idx); }
tools/perf/util/evlist.h +1 -0
@@ -51,6 +51,7 @@ struct perf_evlist {struct thread_map *threads;struct cpu_map *cpus;struct perf_evsel *selected; +struct events_stats stats; };struct perf_evsel_str_handler {
tools/perf/util/session.c +27 -27
130 lines not on this page · raw message
tools/perf/util/session.h +0 -1
@@ -20,7 +20,6 @@ struct perf_session {struct machines machines;struct perf_evlist *evlist;struct trace_event tevent; -struct events_stats stats;bool repipe;bool one_mmap;void *one_mmap_addr;
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 10/21 perf session: Remove perf_session from warn_errors signature
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
--- tools/perf/util/session.c | 36 ++++++++++++++++++------------------ 1 file changed, 18 insertions(+), 18 deletions(-)
tools/perf/util/session.c +18 -18
@@ -1125,47 +1125,47 @@ static struct thread *perf_session__register_idle_thread(struct perf_session *sereturn thread; } -static void perf_session__warn_about_errors (const struct perf_session *session, -const struct perf_tool *tool) +static void perf_tool__warn_about_errors (const struct perf_tool *tool, +const struct events_stats *stats) {if (tool->lost == perf_event__process_lost && - session->evlist->stats.nr_events[PERF_RECORD_LOST ] !=0 ) { + stats->nr_events[PERF_RECORD_LOST ] !=0 ) {ui__warning ("Processed %d events and lost %d chunks!\n\n" "Check IO/CPU overload!\n\n" , - session->evlist->stats.nr_events[0 ], - session->evlist->stats.nr_events[PERF_RECORD_LOST ]); + stats->nr_events[0 ], + stats->nr_events[PERF_RECORD_LOST ]); } -if (session->evlist->stats.nr_unknown_events !=0 ) { +if (stats->nr_unknown_events !=0 ) {ui__warning ("Found %u unknown events!\n\n" "Is this an older tool processing a perf.data " "file generated by a more recent tool?\n\n" "If that is not the case, consider " "reporting to [email protected].\n\n" , - session->evlist->stats.nr_unknown_events); + stats->nr_unknown_events); } -if (session->evlist->stats.nr_unknown_id !=0 ) { +if (stats->nr_unknown_id !=0 ) {ui__warning ("%u samples with id not present in the header\n" , - session->evlist->stats.nr_unknown_id); + stats->nr_unknown_id); } -if (session->evlist->stats.nr_invalid_chains !=0 ) { +if (stats->nr_invalid_chains !=0 ) {ui__warning ("Found invalid callchains!\n\n" "%u out of %u events were discarded for this reason.\n\n" "Consider reporting to [email protected].\n\n" , - session->evlist->stats.nr_invalid_chains, - session->evlist->stats.nr_events[PERF_RECORD_SAMPLE ]); + stats->nr_invalid_chains, + stats->nr_events[PERF_RECORD_SAMPLE ]); } -if (session->evlist->stats.nr_unprocessable_samples !=0 ) { +if (stats->nr_unprocessable_samples !=0 ) {ui__warning ("%u unprocessable samples recorded.\n" "Do you have a KVM guest running and not using 'perf kvm'?\n" , - session->evlist->stats.nr_unprocessable_samples); + stats->nr_unprocessable_samples); } -if (session->evlist->stats.nr_unordered_events !=0 ) -ui__warning ("%u out of order events recorded.\n" , session->evlist->stats.nr_unordered_events); +if (stats->nr_unordered_events !=0 ) +ui__warning ("%u out of order events recorded.\n" , stats->nr_unordered_events); }volatile int session_done;
@@ -1255,7 +1255,7 @@ done: err =ordered_events__flush (session, tool,OE_FLUSH__FINAL ); out_err:free (buf); -perf_session__warn_about_errors (session, tool); +perf_tool__warn_about_errors (tool, &session->evlist->stats);ordered_events__free (&session->ordered_events);return err; }
@@ -1400,7 +1400,7 @@ out: err =ordered_events__flush (session, tool,OE_FLUSH__FINAL ); out_err:ui_progress__finish (); -perf_session__warn_about_errors (session, tool); +perf_tool__warn_about_errors (tool, &session->evlist->stats);ordered_events__free (&session->ordered_events); session->one_mmap = false;return err;
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 11/21 perf session: Remove perf_session from some deliver event routines
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
Further untangling perf_session from plain event delivery routines.
--- tools/perf/util/session.c | 38 +++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-)
tools/perf/util/session.c +19 -19
@@ -824,16 +824,15 @@ static struct machine *machines__find_for_cpumode(struct machines *machines,return &machines->host; } -static int deliver_sample_value (struct perf_session *session, +static int deliver_sample_value (struct perf_evlist *evlist,struct perf_tool *tool,union perf_event *event,struct perf_sample *sample,struct sample_read_value *v,struct machine *machine) { -struct perf_sample_id *sid; +struct perf_sample_id *sid =perf_evlist__id2sid (evlist, v->id); - sid =perf_evlist__id2sid (session->evlist, v->id);if (sid) { sample->id = v->id; sample->period = v->value - sid->period;
@@ -841,14 +840,14 @@ static int deliver_sample_value(struct perf_session *session, }if (!sid || sid->evsel ==NULL ) { - ++session->evlist->stats.nr_unknown_id; + ++evlist->stats.nr_unknown_id;return 0 ; }return tool->sample (tool, event, sample, sid->evsel, machine); } -static int deliver_sample_group(struct perf_session *session, +static int deliver_sample_group(struct perf_evlist *evlist, struct perf_tool *tool, union perf_event *event, struct perf_sample *sample,
@@ -858,7 +857,7 @@ static int deliver_sample_group(struct perf_session *session,u64 i;for (i =0 ; i < sample->read.group.nr; i++) { - ret =deliver_sample_value (session, tool, event, sample, + ret =deliver_sample_value (evlist, tool, event, sample, &sample->read.group.values[i], machine);if (ret)
@@ -869,7 +868,7 @@ static int deliver_sample_group(struct perf_session *session, } static int -perf_session__deliver_sample(struct perf_session *session, + perf_evlist__deliver_sample(struct perf_evlist *evlist, struct perf_tool *tool, union perf_event *event, struct perf_sample *sample,
@@ -886,10 +885,10 @@ perf_session__deliver_sample(struct perf_session *session,/* For PERF_SAMPLE_READ we have either single or group mode. */ if (read_format &PERF_FORMAT_GROUP ) -return deliver_sample_group (session, tool, event, sample, +return deliver_sample_group (evlist, tool, event, sample, machine);else -return deliver_sample_value (session, tool, event, sample, +return deliver_sample_value (evlist, tool, event, sample, &sample->read.one, machine); }
@@ -898,12 +897,13 @@ int perf_session__deliver_event(struct perf_session *session,struct perf_sample *sample, struct perf_tool *tool, u64 file_offset) { + struct perf_evlist *evlist = session->evlist;struct perf_evsel *evsel;struct machine *machine;dump_event (session, event, file_offset, sample); - evsel =perf_evlist__id2evsel (session->evlist, sample->id); + evsel =perf_evlist__id2evsel (evlist, sample->id); machine =machines__find_for_cpumode (&session->machines, event, sample);
@@ -911,15 +911,14 @@ int perf_session__deliver_event(struct perf_session *session,case PERF_RECORD_SAMPLE :dump_sample (evsel, event, sample);if (evsel ==NULL ) { - ++session->evlist->stats.nr_unknown_id; + ++evlist->stats.nr_unknown_id;return 0 ; }if (machine ==NULL ) { - ++session->evlist->stats.nr_unprocessable_samples; + ++evlist->stats.nr_unprocessable_samples;return 0 ; } -return perf_session__deliver_sample (session, tool, event, - sample, evsel, machine); +return perf_evlist__deliver_sample (evlist, tool, event, sample, evsel, machine);case PERF_RECORD_MMAP :return tool->mmap (tool, event, sample, machine);case PERF_RECORD_MMAP2 :
@@ -932,7 +931,7 @@ int perf_session__deliver_event(struct perf_session *session,return tool->exit (tool, event, sample, machine);case PERF_RECORD_LOST :if (tool->lost == perf_event__process_lost) - session->evlist->stats.total_lost += event->lost.lost; + evlist->stats.total_lost += event->lost.lost;return tool->lost (tool, event, sample, machine);case PERF_RECORD_READ :return tool->read (tool, event, sample, evsel, machine);
@@ -941,7 +940,7 @@ int perf_session__deliver_event(struct perf_session *session,case PERF_RECORD_UNTHROTTLE :return tool->unthrottle (tool, event, sample, machine);default : - ++session->evlist->stats.nr_unknown_events; + ++evlist->stats.nr_unknown_events;return -1 ; } }
@@ -1068,16 +1067,17 @@ static s64 perf_session__process_event(struct perf_session *session, struct perf_tool *tool, u64 file_offset) { +struct perf_evlist *evlist = session->evlist;struct perf_sample sample;int ret;if (session->header.needs_swap) -event_swap (event,perf_evlist__sample_id_all (session->evlist)); +event_swap (event,perf_evlist__sample_id_all (evlist));if (event->header.type >=PERF_RECORD_HEADER_MAX )return -EINVAL ; -events_stats__inc (&session->evlist->stats, event->header.type); +events_stats__inc (&evlist->stats, event->header.type);if (event->header.type >=PERF_RECORD_USER_TYPE_START )return perf_session__process_user_event (session, event, tool, file_offset);
@@ -1085,7 +1085,7 @@ static s64 perf_session__process_event(struct perf_session *session,/* * For all kernel events we get the sample data */ - ret =perf_evlist__parse_sample (session->evlist, event, &sample); + ret =perf_evlist__parse_sample (evlist, event, &sample);if (ret)return ret;
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 13/21 perf ordered_events: Stop using tool->ordered_events
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
To figure out if ordered_events are being used when doing a flush operation, it is enough to check if there were in fact some events queued, i.e. look at oe->nr_events.
--- tools/perf/util/ordered-events.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
tools/perf/util/ordered-events.c +4 -1
@@ -166,7 +166,7 @@ static int __ordered_events__flush(struct perf_session *s,struct ui_progress prog;int ret; -if (!tool->ordered_events || !limit) +if (!limit)return 0 ;if (show_progress)
@@ -216,6 +216,9 @@ int ordered_events__flush(struct perf_session *s, struct perf_tool *tool, };int err; +if (oe->nr_events ==0 ) +return 0 ; +switch (how) {case OE_FLUSH__FINAL : oe->next_flush =ULLONG_MAX ;
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 15/21 perf trace: Dump stack on segfaults
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
[root@ssdandy ~]# perf trace --filter-pids 16348
0.000 ( 0.000 ms): tuned/1027 ... [continued]: select()) = 0 Timeout
793.770 ( 0.000 ms): lsmd/895 ... [continued]: select()) = 0 Timeout 793.775 (793.724 ms): tuned/1027 select(tvp: 0x7f7655556e50) ...
perf: Segmentation fault
Obtained 15 stack frames.
perf(dump_stack+0x2e) [0x4ed330] perf(sighandler_dump_stack+0x2e) [0x4ed40f] /lib64/libc.so.6(+0x35640) [0x7fa2d5b69640]
perf() [0x4c2d35]
perf(machine__findnew_thread+0x39) [0x4c2ed6]
perf() [0x454a4d]
perf() [0x455f87]
perf() [0x456556]
perf(cmd_trace+0xa7e) [0x4580af]
perf() [0x4867bd]
perf() [0x486a1c]
perf() [0x486b68]
perf(main+0x23b) [0x486ec9] /lib64/libc.so.6(__libc_start_main+0xf5) [0x7fa2d5b55af5] perf() [0x41bd91]
[ root@ssdandy ~]#
--- tools/perf/builtin-trace.c | 3 +++ 1 file changed, 3 insertions(+)
tools/perf/builtin-trace.c +3 -0
@@ -2660,6 +2660,9 @@ int cmd_trace(int argc, const char **argv, const char *prefix __maybe_unused)int err;char bf[BUFSIZ ]; +signal (SIGSEGV , sighandler_dump_stack); +signal (SIGFPE , sighandler_dump_stack); + trace.evlist =perf_evlist__new ();if (trace.evlist ==NULL )return -ENOMEM ;
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 16/21 perf tools: Print the thread's tid on PERF_RECORD_COMM events when -D is asked
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
--- tools/perf/util/event.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/util/event.c +1 -1
@@ -615,7 +615,7 @@ size_t perf_event__fprintf_comm(union perf_event *event, FILE *fp)else s ="" ; -return fprintf (fp,"%s: %s:%d\n" , s, event->comm.comm, event->comm.tid); +return fprintf (fp,"%s: %s:%d/%d\n" , s, event->comm.comm, event->comm.pid, event->comm.tid); } int perf_event__process_comm(struct perf_tool *tool __maybe_unused,
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 17/21 perf record: Support recording running/enabled time
In reply to Arnaldo Carvalho de MeloFrom: Andi Kleen <[email protected]>
Add an option to perf record to record running/enabled time for read events, similar to what stat does.
This is useful to understand multiplexing problems.
Right now the report support is not great, but at least report -D already supports it.
--- tools/perf/Documentation/perf-record.txt | 3 +++ tools/perf/builtin-record.c | 2 ++ tools/perf/perf.h | 1 + tools/perf/util/evsel.c | 6 ++++++ 4 files changed, 12 insertions(+)
tools/perf/Documentation/perf-record.txt +3 -0
@@ -241,6 +241,9 @@ Capture machine state (registers) at interrupt, i.e., on counter overflows for each sample. List of captured registers depends on the architecture. This option is off by default. +--running-time:: +Record running and enabled time for read events (:S) + SEE ALSO -------- linkperf:perf-stat[1], linkperf:perf-list[1]
tools/perf/builtin-record.c +2 -0
@@ -839,6 +839,8 @@ struct option __record_options[] = { "use per-thread mmaps"),OPT_BOOLEAN ('I' ,"intr-regs" , &record.opts.sample_intr_regs,"Sample machine registers on interrupt" ), +OPT_BOOLEAN (0 ,"running-time" , &record.opts.running_time, +"Record running/enabled time of read (:S) events" ), OPT_END() };
tools/perf/perf.h +1 -0
@@ -53,6 +53,7 @@ struct record_opts {bool sample_time;bool period;bool sample_intr_regs; +bool running_time;unsigned int freq;unsigned int mmap_pages;unsigned int user_freq;
tools/perf/util/evsel.c +6 -0
@@ -734,6 +734,12 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts)if (opts->sample_transaction)perf_evsel__set_sample_bit (evsel,TRANSACTION ); +if (opts->running_time) { + evsel->attr.read_format |= +PERF_FORMAT_TOTAL_TIME_ENABLED | +PERF_FORMAT_TOTAL_TIME_RUNNING ; + } +/* * XXX see the function comment above *
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 18/21 perf tools: Add feature check for libbabeltrace
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Adding feature check for babeltrace library [1], which will be used for perf data file CTF [2] conversion in following patches.
The babeltrace library is now automatically detected as standard feature. It's possible to specify LIBBABELTRACE_DIR make variable to specify location of installed libbabeltrace, like:
$ make LIBBABELTRACE_DIR=/opt/libbabeltrace/
BUILD: Doing 'make -j4' parallel buildAuto-detecting system features:
... dwarf: [ on ] ... glibc: [ on ] ... gtk2: [ on ] ... libaudit: [ on ] ... libbfd: [ on ] ... libelf: [ on ] ... libnuma: [ on ] ... libperl: [ on ] ... libpython: [ on ] ... libslang: [ on ] ... libunwind: [ on ] ... libbabeltrace: [ on ] ... libdw-dwarf-unwind: [ on ] ... zlib: [ on ]
... DWARF post unwind library: libunwind
NOTE The installation of the [1] to to used by above make:
$ git clone git://git.efficios.com/babeltrace.git
$ cd babeltrace
$ vim README
$ ./bootstrap
$ ./configure --prefix=/opt/libbabeltrace
$ make prefix=/opt/libbabeltrace
$ sudo make install prefix=/opt/libbabeltracePlease make sure that the /opt/libbabeltrace/lib directory is in your LD_LIBRARY_PATH:
$ export LD_LIBRARY_PATH=/opt/libbabeltrace/lib
[1] babeltrace - http://www.efficios.com/babeltrace
[2] Common Trace Format - http://www.efficios.com/ctf
--- tools/perf/Makefile.perf | 4 +++- tools/perf/config/Makefile | 24 ++++++++++++++++++++++ tools/perf/config/feature-checks/Makefile | 8 ++++++-- tools/perf/config/feature-checks/test-all.c | 5 +++++ .../config/feature-checks/test-libbabeltrace.c | 8 ++++++++ 5 files changed, 46 insertions(+), 3 deletions(-) create mode 100644 tools/perf/config/feature-checks/test-libbabeltrace.c
tools/perf/Makefile.perf +3 -1
@@ -68,7 +68,9 @@ include config/utilities.mak # for reading the x32 mode 32-bit compatibility VDSO in 64-bit mode # # Define NO_ZLIB if you do not want to support compressed kernel modules - +# +# Define NO_LIBBABELTRACE if you do not want libbabeltrace support +# for CTF data format. ifeq ($(srctree),) srctree := $(patsubst %/,%,$(dir $(shell pwd)))
tools/perf/config/Makefile +24 -0
@@ -96,6 +96,17 @@ ifndef NO_LIBELF FEATURE_CHECK_LDFLAGS-libdw-dwarf-unwind := $(LIBDW_LDFLAGS) -ldw endif +ifndef NO_LIBBABELTRACE + # for linking with debug library, run like: + # make DEBUG=1 LIBBABELTRACE_DIR=/opt/libbabeltrace/ + ifdef LIBBABELTRACE_DIR + LIBBABELTRACE_CFLAGS := -I$(LIBBABELTRACE_DIR)/include + LIBBABELTRACE_LDFLAGS := -L$(LIBBABELTRACE_DIR)/lib + endif + FEATURE_CHECK_CFLAGS-libbabeltrace := $(LIBBABELTRACE_CFLAGS) + FEATURE_CHECK_LDFLAGS-libbabeltrace := $(LIBBABELTRACE_LDFLAGS) -lbabeltrace-ctf +endif + # include ARCH specific config -include $(src-perf)/arch/$(ARCH)/Makefile
@@ -216,6 +227,7 @@ CORE_FEATURE_TESTS = \ stackprotector-all \ timerfd \ libdw-dwarf-unwind \ + libbabeltrace \ zlib LIB_FEATURE_TESTS = \
@@ -231,6 +243,7 @@ LIB_FEATURE_TESTS = \ libslang \ libunwind \ libdw-dwarf-unwind \ + libbabeltrace \ zlib VF_FEATURE_TESTS = \
@@ -692,6 +705,17 @@ else NO_PERF_READ_VDSOX32 := 1 endif +ifndef NO_LIBBABELTRACE + ifeq ($(feature-libbabeltrace), 0) + msg := $(warning No libbabeltrace found, disables 'perf data' CTF format support, please install libbabeltrace-devel/libbabeltrace-ctf-dev); + NO_LIBBABELTRACE := 1 + else + CFLAGS += -DHAVE_LIBBABELTRACE_SUPPORT $(LIBBABELTRACE_CFLAGS) + LDFLAGS += $(LIBBABELTRACE_LDFLAGS) + EXTLIBS += -lbabeltrace-ctf + endif +endif + # Among the variables below, these: # perfexecdir # template_dir
tools/perf/config/feature-checks/Makefile +6 -2
@@ -29,6 +29,7 @@ FILES= \ test-stackprotector-all.bin \ test-timerfd.bin \ test-libdw-dwarf-unwind.bin \ + test-libbabeltrace.bin \ test-compile-32.bin \ test-compile-x32.bin \ test-zlib.bin
@@ -43,7 +44,7 @@ BUILD = $(CC) $(CFLAGS) -o $(OUTPUT)$@ $(patsubst %.bin,%.c,$@) $(LDFLAGS) ############################### test-all.bin: - $(BUILD) -Werror -fstack-protector-all -O2 -Werror -D_FORTIFY_SOURCE=2 -ldw -lelf -lnuma -lelf -laudit -I/usr/include/slang -lslang $(shell $(PKG_CONFIG) --libs --cflags gtk+-2.0 2>/dev/null) $(FLAGS_PERL_EMBED) $(FLAGS_PYTHON_EMBED) -DPACKAGE='"perf"' -lbfd -ldl -lz + $(BUILD) -Werror -fstack-protector-all -O2 -Werror -D_FORTIFY_SOURCE=2 -ldw -lelf -lnuma -lelf -laudit -I/usr/include/slang -lslang $(shell $(PKG_CONFIG) --libs --cflags gtk+-2.0 2>/dev/null) $(FLAGS_PERL_EMBED) $(FLAGS_PYTHON_EMBED) -DPACKAGE='"perf"' -lbfd -ldl -lz -lbabeltrace test-hello.bin: $(BUILD)
@@ -133,7 +134,10 @@ test-timerfd.bin: $(BUILD) test-libdw-dwarf-unwind.bin: - $(BUILD) + $(BUILD) # -ldw provided by $(FEATURE_CHECK_LDFLAGS-libdw-dwarf-unwind) + +test-libbabeltrace.bin: + $(BUILD) # -lbabeltrace provided by $(FEATURE_CHECK_LDFLAGS-libbabeltrace) test-sync-compare-and-swap.bin: $(BUILD) -Werror
tools/perf/config/feature-checks/test-all.c +5 -0
@@ -101,6 +101,10 @@# include "test-pthread_attr_setaffinity_np.c" #undef main +#define main main_test_libbabeltrace +# include "test-libbabeltrace.c" +#undef main +int main (int argc,char *argv[]) {main_test_libpython ();
@@ -126,6 +130,7 @@ int main(int argc, char *argv[])main_test_sync_compare_and_swap (argc, argv);main_test_zlib ();main_test_pthread_attr_setaffinity_np (); +main_test_libbabeltrace ();return 0 ; }
tools/perf/config/feature-checks/test-libbabeltrace.c new file +8 -0
@@ -0,0 +1,8 @@ + +#include <babeltrace/ctf-writer/writer.h> + +int main (void ) +{ +bt_ctf_stream_class_get_packet_context_type ((void *)0 ); +return 0 ; +}
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 20/21 perf data: Add perf data to CTF conversion support
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Adding 'perf data convert' to convert perf data file into different format. This patch adds support for CTF format conversion.
To convert perf.data into CTF run:
$ perf data convert --to-ctf=./ctf-data/ [ perf data convert: Converted 'perf.data' into CTF data './ctf-data/' ] [ perf data convert: Converted and wrote 11.268 MB (100230 samples) ]
The command will create CTF metadata out of perf.data file (or one specified via -i option) and then convert all sample events into single CTF stream.
Each sample_type bit is translated into separated CTF event field apart from following exceptions:
PERF_SAMPLE_RAW - added in next patch PERF_SAMPLE_READ - TODO PERF_SAMPLE_CALLCHAIN - TODO
PERF_SAMPLE_BRANCH_STACK - TODO
PERF_SAMPLE_REGS_USER - TODO
PERF_SAMPLE_STACK_USER - TODO
$ perf --debug=data-convert=2 data convert ...
The converted CTF data could be analyzed by CTF tools, like babletrace or tracecompass [1].
$ babeltrace ./ctf-data/
[03:19:13.962125533] (+?.?????????) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 1 }
[03:19:13.962130001] (+0.000004468) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 1 }
[03:19:13.962131936] (+0.000001935) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 8 }
[03:19:13.962133732] (+0.000001796) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 114 }
[03:19:13.962135557] (+0.000001825) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 2087 }
[03:19:13.962137627] (+0.000002070) cycles: { }, { ip = 0xFFFFFFFF81361938, tid = 20714, pid = 20714, period = 37582 }
[03:19:13.962161091] (+0.000023464) cycles: { }, { ip = 0xFFFFFFFF8124218F, tid = 20714, pid = 20714, period = 600246 }
[03:19:13.962517569] (+0.000356478) cycles: { }, { ip = 0xFFFFFFFF811A75DB, tid = 20714, pid = 20714, period = 1325731 }
[03:19:13.969518008] (+0.007000439) cycles: { }, { ip = 0x34080917B2, tid = 20714, pid = 20714, period = 1144298 }The following members to the ctf-environment were decided to be added to distinguish and specify perf CTF data:
- domain
It says "kernel" because it contains a kernel trace (not to be confused with a user space like lttng-ust does)
- tracer_name
It says perf. This can be used to distinguish between lttng and perf CTF based trace.
- version
The kernel version from stream. In addition to release, this is what it looks like on a Debian kernel:
release = "3.14-1-amd64";
version = "3.14.0";
[1] http://projects.eclipse.org/projects/tools.tracecompass
--- tools/perf/Documentation/perf-data.txt | 25 ++ tools/perf/Documentation/perf.txt | 7 +- tools/perf/builtin-data.c | 44 +++ tools/perf/config/Makefile | 1 + tools/perf/util/Build | 2 + tools/perf/util/data-convert-bt.c | 612 +++++++++++++++++++++++++++++++++ tools/perf/util/data-convert-bt.h | 8 + tools/perf/util/debug.c | 2 + tools/perf/util/debug.h | 1 + 9 files changed, 701 insertions(+), 1 deletion(-) create mode 100644 tools/perf/util/data-convert-bt.c create mode 100644 tools/perf/util/data-convert-bt.h
tools/perf/Documentation/perf-data.txt +25 -0
@@ -13,3 +13,28 @@ SYNOPSIS DESCRIPTION ----------- Data file related processing. + +COMMANDS +-------- +convert:: + Converts perf data file into another format (only CTF [1] format is support by now). + It's possible to set data-convert debug variable to get debug messages from conversion, + like: + perf --debug data-convert data convert ... + +OPTIONS for 'convert' +--------------------- +--to-ctf:: + Triggers the CTF conversion, specify the path of CTF data directory. + +-i:: + Specify input perf data file path. + +-v:: +--verbose:: + Be more verbose (show counter open errors, etc). + +SEE ALSO +-------- +linkperf:perf[1] +[1] Common Trace Format - http://www.efficios.com/ctf
tools/perf/Documentation/perf.txt +6 -1
@@ -13,11 +13,16 @@ SYNOPSIS OPTIONS ------- --debug:: - Setup debug variable (just verbose for now) in value + Setup debug variable (see list below) in value range (0, 10). Use like: --debug verbose # sets verbose = 1 --debug verbose=2 # sets verbose = 2 + List of debug variables allowed to set: + verbose - general debug messages + ordered-events - ordered events object debug messages + data-convert - data convert command debug messages + --buildid-dir:: Setup buildid cache directory. It has higher priority than buildid.dir config file option.
tools/perf/builtin-data.c +44 -0
57 lines not on this page · raw message
tools/perf/config/Makefile +1 -0
@@ -713,6 +713,7 @@ ifndef NO_LIBBABELTRACE CFLAGS += -DHAVE_LIBBABELTRACE_SUPPORT $(LIBBABELTRACE_CFLAGS) LDFLAGS += $(LIBBABELTRACE_LDFLAGS) EXTLIBS += -lbabeltrace-ctf + $(call detected,CONFIG_LIBBABELTRACE) endif endif
tools/perf/util/Build +2 -0
@@ -88,6 +88,8 @@ libperf-$(CONFIG_DWARF) += dwarf-aux.o libperf-$(CONFIG_LIBDW_DWARF_UNWIND) += unwind-libdw.o libperf-$(CONFIG_LIBUNWIND) += unwind-libunwind.o +libperf-$(CONFIG_LIBBABELTRACE) += data-convert-bt.o + libperf-y += scripting-engines/ libperf-$(CONFIG_PERF_REGS) += perf_regs.o
tools/perf/util/data-convert-bt.c new file +612 -0
612 lines not on this page · raw message
tools/perf/util/data-convert-bt.h new file +8 -0
@@ -0,0 +1,8 @@ +#ifndef __DATA_CONVERT_BT_H +#define __DATA_CONVERT_BT_H +#ifdef HAVE_LIBBABELTRACE_SUPPORT + +int bt_convert__perf2ctf (const char *input_name,const char *to_ctf); + +#endif /* HAVE_LIBBABELTRACE_SUPPORT */ +#endif /* __DATA_CONVERT_BT_H */
tools/perf/util/debug.c +2 -0
@@ -20,6 +20,7 @@ int verbose;bool dump_trace = false, quiet = false;int debug_ordered_events;static int redirect_to_stderr; +int debug_data_convert;static int _eprintf (int level,int var,const char *fmt,va_list args) {
@@ -147,6 +148,7 @@ static struct debug_variable { { .name = "verbose", .ptr = &verbose }, { .name = "ordered-events", .ptr = &debug_ordered_events}, { .name = "stderr", .ptr = &redirect_to_stderr}, + { .name = "data-convert", .ptr = &debug_data_convert }, { .name = NULL, } };
tools/perf/util/debug.h +1 -0
@@ -12,6 +12,7 @@extern int verbose;extern bool quiet, dump_trace;extern int debug_ordered_events; +extern int debug_data_convert;#ifndef pr_fmt#define pr_fmt (fmt) fmt
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 21/21 perf data: Add a 'perf' prefix to the generic fields
In reply to Arnaldo Carvalho de MeloFrom: Sebastian Andrzej Siewior <[email protected]>
Some of the tracers bring their own id or pid fields and we can end up having two of them. This patch adds a "perf_" prefix to the 'generic' fields so we avoid a clash of the member names.
The change is visible in the babeltrace output:
Before:
$ babeltrace ./ctf-data/
[03:19:13.962131936] (+0.000001935) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 8 }
[03:19:13.962133732] (+0.000001796) cycles: { }, { ip = 0xFFFFFFFF8105443A, tid = 20714, pid = 20714, period = 114 }
...Now:
$ babeltrace ./ctf-data/
[03:19:13.962131936] (+0.000001935) cycles: { }, { perf_ip = 0xFFFFFFFF8105443A, perf_tid = 20714, perf_pid = 20714, perf_period = 8 }
[03:19:13.962133732] (+0.000001796) cycles: { }, { perf_ip = 0xFFFFFFFF8105443A, perf_tid = 20714, perf_pid = 20714, perf_period = 114 }
...--- tools/perf/util/data-convert-bt.c | 42 ++++++++++++++++++++------------------- 1 file changed, 22 insertions(+), 20 deletions(-)
tools/perf/util/data-convert-bt.c +22 -20
@@ -147,60 +147,62 @@ static int add_generic_values(struct ctf_writer *cw,*/ if (type &PERF_SAMPLE_IP ) { - ret =value_set_u64_hex (cw, event,"ip" , sample->ip); + ret =value_set_u64_hex (cw, event,"perf_ip" , sample->ip);if (ret)return -1 ; }if (type &PERF_SAMPLE_TID ) { - ret =value_set_s32 (cw, event,"tid" , sample->tid); + ret =value_set_s32 (cw, event,"perf_tid" , sample->tid);if (ret)return -1 ; - ret =value_set_s32 (cw, event,"pid" , sample->pid); + ret =value_set_s32 (cw, event,"perf_pid" , sample->pid);if (ret)return -1 ; }if ((type &PERF_SAMPLE_ID ) || (type &PERF_SAMPLE_IDENTIFIER )) { - ret =value_set_u64 (cw, event,"id" , sample->id); + ret =value_set_u64 (cw, event,"perf_id" , sample->id);if (ret)return -1 ; }if (type &PERF_SAMPLE_STREAM_ID ) { - ret =value_set_u64 (cw, event,"stream_id" , sample->stream_id); + ret =value_set_u64 (cw, event,"perf_stream_id" , sample->stream_id);if (ret)return -1 ; }if (type &PERF_SAMPLE_CPU ) { - ret =value_set_u32 (cw, event,"cpu" , sample->cpu); + ret =value_set_u32 (cw, event,"perf_cpu" , sample->cpu);if (ret)return -1 ; }if (type &PERF_SAMPLE_PERIOD ) { - ret =value_set_u64 (cw, event,"period" , sample->period); + ret =value_set_u64 (cw, event,"perf_period" , sample->period);if (ret)return -1 ; }if (type &PERF_SAMPLE_WEIGHT ) { - ret =value_set_u64 (cw, event,"weight" , sample->weight); + ret =value_set_u64 (cw, event,"perf_weight" , sample->weight);if (ret)return -1 ; }if (type &PERF_SAMPLE_DATA_SRC ) { - ret =value_set_u64 (cw, event,"data_src" , sample->data_src); + ret =value_set_u64 (cw, event,"perf_data_src" , + sample->data_src);if (ret)return -1 ; }if (type &PERF_SAMPLE_TRANSACTION ) { - ret =value_set_u64 (cw, event,"transaction" , sample->transaction); + ret =value_set_u64 (cw, event,"perf_transaction" , + sample->transaction);if (ret)return -1 ; }
@@ -276,34 +278,34 @@ static int add_generic_types(struct ctf_writer *cw, struct perf_evsel *evsel, }while (0 )if (type &PERF_SAMPLE_IP ) -ADD_FIELD (event_class, cw->data.u64_hex,"ip" ); +ADD_FIELD (event_class, cw->data.u64_hex,"perf_ip" );if (type &PERF_SAMPLE_TID ) { -ADD_FIELD (event_class, cw->data.s32,"tid" ); -ADD_FIELD (event_class, cw->data.s32,"pid" ); +ADD_FIELD (event_class, cw->data.s32,"perf_tid" ); +ADD_FIELD (event_class, cw->data.s32,"perf_pid" ); }if ((type &PERF_SAMPLE_ID ) || (type &PERF_SAMPLE_IDENTIFIER )) -ADD_FIELD (event_class, cw->data.u64,"id" ); +ADD_FIELD (event_class, cw->data.u64,"perf_id" );if (type &PERF_SAMPLE_STREAM_ID ) -ADD_FIELD (event_class, cw->data.u64,"stream_id" ); +ADD_FIELD (event_class, cw->data.u64,"perf_stream_id" );if (type &PERF_SAMPLE_CPU ) -ADD_FIELD (event_class, cw->data.u32,"cpu" ); +ADD_FIELD (event_class, cw->data.u32,"perf_cpu" );if (type &PERF_SAMPLE_PERIOD ) -ADD_FIELD (event_class, cw->data.u64,"period" ); +ADD_FIELD (event_class, cw->data.u64,"perf_period" );if (type &PERF_SAMPLE_WEIGHT ) -ADD_FIELD (event_class, cw->data.u64,"weight" ); +ADD_FIELD (event_class, cw->data.u64,"perf_weight" );if (type &PERF_SAMPLE_DATA_SRC ) -ADD_FIELD (event_class, cw->data.u64,"data_src" ); +ADD_FIELD (event_class, cw->data.u64,"perf_data_src" );if (type &PERF_SAMPLE_TRANSACTION ) -ADD_FIELD (event_class, cw->data.u64,"transaction" ); +ADD_FIELD (event_class, cw->data.u64,"perf_transaction" );#undef ADD_FIELDreturn 0 ;
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 19/21 perf tools: Add new 'perf data' command
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Adding new 'perf data' command to provide operations over data files.
The 'perf data convert' sub command is coming in following patch, but there's possibility for other useful commands like 'perf data ls' (to display perf data file in directory in ls style).
--- tools/perf/Build | 1 + tools/perf/Documentation/perf-data.txt | 15 +++++++ tools/perf/builtin-data.c | 75 ++++++++++++++++++++++++++++++++++ tools/perf/builtin.h | 1 + tools/perf/command-list.txt | 1 + tools/perf/perf.c | 1 + 6 files changed, 94 insertions(+) create mode 100644 tools/perf/Documentation/perf-data.txt create mode 100644 tools/perf/builtin-data.c
tools/perf/Build +1 -0
@@ -18,6 +18,7 @@ perf-y += builtin-lock.o perf-y += builtin-kvm.o perf-y += builtin-inject.o perf-y += builtin-mem.o +perf-y += builtin-data.o perf-$(CONFIG_AUDIT) += builtin-trace.o perf-$(CONFIG_LIBELF) += builtin-probe.o
tools/perf/Documentation/perf-data.txt new file +15 -0
@@ -0,0 +1,15 @@ +perf-data(1) +============== + +NAME +---- +perf-data - Data file related processing + +SYNOPSIS +-------- +[verse] +'perf data' [<common options>] <command> [<options>]", + +DESCRIPTION +----------- +Data file related processing.
tools/perf/builtin-data.c new file +75 -0
75 lines not on this page · raw message
tools/perf/builtin.h +1 -0
@@ -37,6 +37,7 @@ extern int cmd_test(int argc, const char **argv, const char *prefix);extern int cmd_trace (int argc,const char **argv,const char *prefix);extern int cmd_inject (int argc,const char **argv,const char *prefix);extern int cmd_mem (int argc,const char **argv,const char *prefix); +extern int cmd_data (int argc,const char **argv,const char *prefix);extern int find_scripts (char **scripts_array,char **scripts_path_array);#endif
tools/perf/command-list.txt +1 -0
@@ -7,6 +7,7 @@ perf-archive mainporcelain common perf-bench mainporcelain common perf-buildid-cache mainporcelain common perf-buildid-list mainporcelain common +perf-data mainporcelain common perf-diff mainporcelain common perf-evlist mainporcelain common perf-inject mainporcelain common
tools/perf/perf.c +1 -0
@@ -62,6 +62,7 @@ static struct cmd_struct commands[] = {#endif { "inject", cmd_inject, 0 }, { "mem", cmd_mem, 0 }, + { "data", cmd_data, 0 }, }; struct pager_config {
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 14/21 perf tools: Introduce dump_stack signal helper
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
To use in stdio based tools, like 'trace'.
--- tools/perf/util/util.c | 7 +++++++ tools/perf/util/util.h | 1 + 2 files changed, 8 insertions(+)
tools/perf/util/util.c +7 -0
@@ -269,6 +269,13 @@ void dump_stack(void)void dump_stack (void ) {}#endif +void sighandler_dump_stack (int sig) +{ +psignal (sig,"perf" ); +dump_stack (); +exit (sig); +} +void get_term_dimensions (struct winsize *ws) {char *s =getenv ("LINES" );
tools/perf/util/util.h +1 -0
@@ -277,6 +277,7 @@ char *ltrim(char *s);char *rtrim (char *s);void dump_stack (void ); +void sighandler_dump_stack (int sig);extern unsigned int page_size;extern int cacheline_size;
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerBummer, the subject should've been:
[GIT PULL 00/21] perf/core improvements and fixes
;-\
- Arnaldo
Arnaldo Carvalho de Melo
maintainerPATCH 12/21 perf session: Remove perf_session from dump_event
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
All it wants is session->evlist.
--- tools/perf/util/session.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-)
tools/perf/util/session.c +7 -7
@@ -688,14 +688,14 @@ static void stack_user__printf(struct stack_dump *dump) dump->size, dump->offset); } -static void perf_session__print_tstamp (struct perf_session *session, +static void perf_evlist__print_tstamp (struct perf_evlist *evlist,union perf_event *event,struct perf_sample *sample) { -u64 sample_type =__perf_evlist__combined_sample_type (session->evlist); +u64 sample_type =__perf_evlist__combined_sample_type (evlist);if (event->header.type !=PERF_RECORD_SAMPLE && - !perf_evlist__sample_id_all (session->evlist)) { + !perf_evlist__sample_id_all (evlist)) {fputs ("-1 -1 " , stdout);return ; }
@@ -737,7 +737,7 @@ static void sample_read__printf(struct perf_sample *sample, u64 read_format) sample->read.one.id, sample->read.one.value); } -static void dump_event (struct perf_session *session,union perf_event *event, +static void dump_event (struct perf_evlist *evlist,union perf_event *event,u64 file_offset,struct perf_sample *sample) {if (!dump_trace)
@@ -749,7 +749,7 @@ static void dump_event(struct perf_session *session, union perf_event *event,trace_event (event);if (sample) -perf_session__print_tstamp (session, event, sample); +perf_evlist__print_tstamp (evlist, event, sample);printf ("%#" PRIx64" [%#x]: PERF_RECORD_%s" , file_offset, event->header.size,perf_event__name (event->header.type));
@@ -901,7 +901,7 @@ int perf_session__deliver_event(struct perf_session *session,struct perf_evsel *evsel;struct machine *machine; -dump_event (session, event, file_offset, sample); +dump_event (evlist, event, file_offset, sample); evsel =perf_evlist__id2evsel (evlist, sample->id);
@@ -953,7 +953,7 @@ static s64 perf_session__process_user_event(struct perf_session *session,int fd =perf_data_file__fd (session->file);int err; -dump_event (session, event, file_offset,NULL ); +dump_event (session->evlist, event, file_offset,NULL );/* These events are processed right away */ switch (event->header.type) {
Signature · 1 line
1.9.3
Arnaldo Carvalho de Melo
maintainerPATCH 5/21 perf trace: Introduce --filter-pids
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
When tracing in X we get event loops due to the tracing activity, i.e. updates to a gnome-terminal that generate syscalls for X.org, etc.
To get a more useful view of what is happening, syscall wise, system wide, we need to filter those, like in:
# ps ax|egrep '981|2296|1519' | grep -v egrep 981 tty1 Ss+ 5:40 /usr/bin/Xorg :0 -background none ... 1519 ? Sl 2:22 /usr/bin/gnome-shell 2296 ? Sl 4:16 /usr/libexec/gnome-terminal-server #
# trace -e write --filter-pids 981,2296,1519
0.385 ( 0.021 ms): goa-daemon/2061 write(fd: 1</dev/null>, buf: 0x7fbeb017b000, count: 136) = 136
0.922 ( 0.014 ms): goa-daemon/2061 write(fd: 1</dev/null>, buf: 0x7fbeb017b000, count: 140) = 140
5006.525 ( 0.029 ms): goa-daemon/2061 write(fd: 1</dev/null>, buf: 0x7fbeb017b000, count: 136) = 136
5007.235 ( 0.023 ms): goa-daemon/2061 write(fd: 1</dev/null>, buf: 0x7fbeb017b000, count: 140) = 140
5177.646 ( 0.018 ms): rtkit-daemon/782 write(fd: 5<anon_inode:[eventfd]>, buf: 0x7f7eea70be88, count: 8) = 8
8314.497 ( 0.004 ms): gsd-locate-poi/2084 write(fd: 5<anon_inode:[eventfd]>, buf: 0x7fffe96af7b0, count: 8) = 8
8314.518 ( 0.002 ms): gsd-locate-poi/2084 write(fd: 5<anon_inode:[eventfd]>, buf: 0x7fffe96af0e0, count: 8) = 8
^C#When this option is used the tracer pid is also filtered.
--- tools/perf/Documentation/perf-trace.txt | 3 ++ tools/perf/builtin-trace.c | 49 +++++++++++++++++++++++++++++++-- 2 files changed, 50 insertions(+), 2 deletions(-)
tools/perf/Documentation/perf-trace.txt +3 -0
@@ -55,6 +55,9 @@ OPTIONS --uid=:: Record events in threads owned by uid. Name or number. +--filter-pids=:: + Filter out events for these pids and for 'trace' itself (comma separated list). + -v:: --verbose=:: Verbosity level.
tools/perf/builtin-trace.c +47 -2
73 lines not on this page · raw message
Signature · 1 line
1.9.3
Quoted 111 lines · Arnaldo Carvalho de Melo
Hi Ingo,
Please consider pulling,
- Arnaldo
The following changes since commit
8a26ce4e544659256349551283414df504889a59:Merge tag 'perf-core-for-mingo' of
git://git.kernel.org/pub/scm/linux/kernel/git/acme/linuxinto perf/core (2015-02-18 19:18:18 +0100)are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.gittags/perf-core-for-mingofor you to fetch changes up to
54cf776a9c5c2e6a91de31954bba4d3bad6c657c:perf data: Add a 'perf' prefix to the generic fields (2015-02-25 16:14:33 -0300)
----------------------------------------------------------------
perf/core improvements and fixes:New user selectable features:
- Support recording running/enabled time in 'perf record' (Andi Kleen)
- New tool: 'perf data' for converting perf.data to other formats, initially for the CTF (Common Trace Format) from LTTng (Jiri Olsa, Sebastian Siewior)
User visible:
- Only insert blank duration bracket when tracing syscalls in 'perf trace' (Arnaldo Carvalho de Melo)
- Filter out the trace pid when no threads are specified in 'perf trace' (Arnaldo Carvalho de Melo)
- Add 'perf trace' man page entry for --event (Arnaldo Carvalho de Melo)
- Dump stack on segfaults in 'perf trace' (Arnaldo Carvalho de Melo)
Infrastructure:
- Introduce
set_filter_pidandset_filter_pidsmethods in the evlist class (Arnaldo Carvalho de Melo)- Some
perf_sessionuntanglement patches, removing the need to pass aperf_sessioninstance for things that are related to evlists, so that tools that don't deal with perf.data files like trace in live mode can make use of theordered_eventsclass (Arnaldo Carvalho de Melo)Signed-off-by Arnaldo Carvalho de Melo <[email protected]>----------------------------------------------------------------
Andi Kleen (1):
perf record: Support recording running/enabled timeArnaldo Carvalho de Melo (16):
perf trace: Only insert blank duration bracket when tracing syscalls perf evlist: Introduceset_filter_pidmethod
perf trace: Filter out the trace pid when no threads are specified perf evlist: Introduceset_filter_pidsmethod
perf trace: Introduce --filter-pids
perf trace: Add man page entry for --event
perf trace: Separate routine that handles an event from the one that reads it perf session: Remove wrappers tomachines__find
perf evlist: Adoptevents_statsfromperf_session
perf session: Removeperf_sessionfromwarn_errorssignature perf session: Removeperf_sessionfrom some deliver event routines perf session: Removeperf_sessionfromdump_event
perfordered_events: Stop usingtool->ordered_events
perf tools: Introducedump_stacksignal helper
perf trace: Dump stack on segfaults
perf tools: Print the thread's tid onPERF_RECORD_COMMevents when -D is askedJiri Olsa (3):
perf tools: Add feature check for libbabeltrace perf tools: Add new 'perf data' command perf data: Add perf data to CTF conversion supportSebastian Andrzej Siewior (1):
perf data: Add a 'perf' prefix to the generic fieldstools/perf/Build | 1 + tools/perf/Documentation/perf-data.txt | 40 ++ tools/perf/Documentation/perf-record.txt | 3 + tools/perf/Documentation/perf-trace.txt | 6 + tools/perf/Documentation/perf.txt | 7 +- tools/perf/Makefile.perf | 4 +- tools/perf/builtin-data.c | 119 ++++ tools/perf/builtin-record.c | 2 + tools/perf/builtin-sched.c | 6 +- tools/perf/builtin-top.c | 4 +- tools/perf/builtin-trace.c | 121 +++- tools/perf/builtin.h | 1 + tools/perf/command-list.txt | 1 + tools/perf/config/Makefile | 25 + tools/perf/config/feature-checks/Makefile | 8 +- tools/perf/config/feature-checks/test-all.c | 5 + .../config/feature-checks/test-libbabeltrace.c | 8 + tools/perf/perf.c | 1 + tools/perf/perf.h | 1 + tools/perf/util/Build | 2 + tools/perf/util/data-convert-bt.c | 614 +++++++++++++++++++++ tools/perf/util/data-convert-bt.h | 8 + tools/perf/util/debug.c | 2 + tools/perf/util/debug.h | 1 + tools/perf/util/event.c | 2 +- tools/perf/util/evlist.c | 32 ++ tools/perf/util/evlist.h | 3 + tools/perf/util/evsel.c | 6 + tools/perf/util/ordered-events.c | 5 +- tools/perf/util/session.c | 115 ++-- tools/perf/util/session.h | 1 -
Quoted 8 lines
tools/perf/util/util.c | 7 + tools/perf/util/util.h | 1 + 33 files changed, 1063 insertions(+), 99 deletions(-) create mode 100644 tools/perf/Documentation/perf-data.txt create mode 100644 tools/perf/builtin-data.c create mode 100644 tools/perf/config/feature-checks/test-libbabeltrace.c create mode 100644 tools/perf/util/data-convert-bt.c create mode 100644 tools/perf/util/data-convert-bt.h
Pulled, thanks a lot Arnaldo!
Ingo