* [PATCH 0/5] tools,libperf: Assorted fixes
@ 2019-08-22 11:11 Jiri Olsa
2019-08-22 11:11 ` [PATCH 1/5] tools: Add missing perf_event.h include Jiri Olsa
` (4 more replies)
0 siblings, 5 replies; 11+ messages in thread
From: Jiri Olsa @ 2019-08-22 11:11 UTC (permalink / raw)
To: Arnaldo Carvalho de Melo
Cc: lkml, Ingo Molnar, Namhyung Kim, Alexander Shishkin,
Peter Zijlstra, Michael Petlan
hi,
sending fixes for tools and libperf.
thanks,
jirka
---
Jiri Olsa (5):
tools: Add missing perf_event.h include
libperf: Use perf_cpu_map__nr instead of cpu_map__nr
libperf: Move cpu_map__empty to perf_cpu_map__empty
libperf: Move cpu_map__idx to perf_cpu_map__idx
libperf: Add perf_thread_map__nr/perf_thread_map__pid functions
tools/include/linux/ring_buffer.h | 1 +
tools/perf/arch/arm/util/cs-etm.c | 12 ++++++------
tools/perf/arch/x86/util/intel-bts.c | 4 ++--
tools/perf/arch/x86/util/intel-pt.c | 10 +++++-----
tools/perf/builtin-c2c.c | 2 +-
tools/perf/builtin-ftrace.c | 2 +-
tools/perf/builtin-script.c | 4 ++--
tools/perf/builtin-stat.c | 8 ++++----
tools/perf/builtin-trace.c | 4 ++--
tools/perf/lib/cpumap.c | 17 +++++++++++++++++
tools/perf/lib/include/internal/cpumap.h | 2 ++
tools/perf/lib/include/perf/cpumap.h | 2 ++
tools/perf/lib/include/perf/threadmap.h | 2 ++
tools/perf/lib/libperf.map | 3 +++
tools/perf/lib/threadmap.c | 10 ++++++++++
tools/perf/tests/thread-map.c | 6 +++---
tools/perf/util/auxtrace.c | 4 ++--
tools/perf/util/cpumap.c | 20 ++++----------------
tools/perf/util/cpumap.h | 13 +------------
tools/perf/util/event.c | 10 +++++-----
tools/perf/util/evlist.c | 30 +++++++++++++++---------------
tools/perf/util/evsel.c | 4 ++--
tools/perf/util/mmap.c | 2 +-
tools/perf/util/record.c | 2 +-
tools/perf/util/scripting-engines/trace-event-python.c | 2 +-
tools/perf/util/stat-display.c | 6 +++---
tools/perf/util/stat.c | 6 +++---
tools/perf/util/thread_map.c | 4 ++--
tools/perf/util/thread_map.h | 10 ----------
29 files changed, 103 insertions(+), 99 deletions(-)
^ permalink raw reply [flat|nested] 11+ messages in thread* [PATCH 1/5] tools: Add missing perf_event.h include 2019-08-22 11:11 [PATCH 0/5] tools,libperf: Assorted fixes Jiri Olsa @ 2019-08-22 11:11 ` Jiri Olsa 2019-08-22 14:13 ` Arnaldo Carvalho de Melo 2019-08-22 11:11 ` [PATCH 2/5] libperf: Use perf_cpu_map__nr instead of cpu_map__nr Jiri Olsa ` (3 subsequent siblings) 4 siblings, 1 reply; 11+ messages in thread From: Jiri Olsa @ 2019-08-22 11:11 UTC (permalink / raw) To: Arnaldo Carvalho de Melo Cc: lkml, Ingo Molnar, Namhyung Kim, Alexander Shishkin, Peter Zijlstra, Michael Petlan We need perf_event.h include for 'struct perf_event_mmap_page'. Link: http://lkml.kernel.org/n/tip-bolqkmqajexhccjb0ib0an8w@git.kernel.org Signed-off-by: Jiri Olsa <jolsa@kernel.org> --- tools/include/linux/ring_buffer.h | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/include/linux/ring_buffer.h b/tools/include/linux/ring_buffer.h index 9a083ae60473..6c02617377c2 100644 --- a/tools/include/linux/ring_buffer.h +++ b/tools/include/linux/ring_buffer.h @@ -2,6 +2,7 @@ #define _TOOLS_LINUX_RING_BUFFER_H_ #include <asm/barrier.h> +#include <linux/perf_event.h> /* * Contract with kernel for walking the perf ring buffer from -- 2.21.0 ^ permalink raw reply related [flat|nested] 11+ messages in thread
* Re: [PATCH 1/5] tools: Add missing perf_event.h include 2019-08-22 11:11 ` [PATCH 1/5] tools: Add missing perf_event.h include Jiri Olsa @ 2019-08-22 14:13 ` Arnaldo Carvalho de Melo 0 siblings, 0 replies; 11+ messages in thread From: Arnaldo Carvalho de Melo @ 2019-08-22 14:13 UTC (permalink / raw) To: Jiri Olsa Cc: lkml, Ingo Molnar, Namhyung Kim, Alexander Shishkin, Peter Zijlstra, Michael Petlan Em Thu, Aug 22, 2019 at 01:11:37PM +0200, Jiri Olsa escreveu: > We need perf_event.h include for 'struct perf_event_mmap_page'. > > Link: http://lkml.kernel.org/n/tip-bolqkmqajexhccjb0ib0an8w@git.kernel.org Thanks, applied. - Arnaldo > Signed-off-by: Jiri Olsa <jolsa@kernel.org> > --- > tools/include/linux/ring_buffer.h | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/tools/include/linux/ring_buffer.h b/tools/include/linux/ring_buffer.h > index 9a083ae60473..6c02617377c2 100644 > --- a/tools/include/linux/ring_buffer.h > +++ b/tools/include/linux/ring_buffer.h > @@ -2,6 +2,7 @@ > #define _TOOLS_LINUX_RING_BUFFER_H_ > > #include <asm/barrier.h> > +#include <linux/perf_event.h> > > /* > * Contract with kernel for walking the perf ring buffer from > -- > 2.21.0 -- - Arnaldo ^ permalink raw reply [flat|nested] 11+ messages in thread
* [PATCH 2/5] libperf: Use perf_cpu_map__nr instead of cpu_map__nr 2019-08-22 11:11 [PATCH 0/5] tools,libperf: Assorted fixes Jiri Olsa 2019-08-22 11:11 ` [PATCH 1/5] tools: Add missing perf_event.h include Jiri Olsa @ 2019-08-22 11:11 ` Jiri Olsa 2019-08-22 11:11 ` [PATCH 3/5] libperf: Move cpu_map__empty to perf_cpu_map__empty Jiri Olsa ` (2 subsequent siblings) 4 siblings, 0 replies; 11+ messages in thread From: Jiri Olsa @ 2019-08-22 11:11 UTC (permalink / raw) To: Arnaldo Carvalho de Melo Cc: lkml, Ingo Molnar, Namhyung Kim, Alexander Shishkin, Peter Zijlstra, Michael Petlan Switch the rest of the perf code to use perf_cpu_map__nr, which is the same as current cpu_map__nr and remove cpu_map__nr function. Link: http://lkml.kernel.org/n/tip-6e0guy75clis7nm0xpuz9fga@git.kernel.org Signed-off-by: Jiri Olsa <jolsa@kernel.org> --- tools/perf/arch/arm/util/cs-etm.c | 4 ++-- tools/perf/util/cpumap.h | 5 ----- tools/perf/util/evlist.c | 10 +++++----- tools/perf/util/mmap.c | 2 +- tools/perf/util/stat-display.c | 2 +- 5 files changed, 9 insertions(+), 14 deletions(-) diff --git a/tools/perf/arch/arm/util/cs-etm.c b/tools/perf/arch/arm/util/cs-etm.c index 5cb07e8cb296..c786ab095d15 100644 --- a/tools/perf/arch/arm/util/cs-etm.c +++ b/tools/perf/arch/arm/util/cs-etm.c @@ -653,7 +653,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr, cpu_map = online_cpus; } else { /* Make sure all specified CPUs are online */ - for (i = 0; i < cpu_map__nr(event_cpus); i++) { + for (i = 0; i < perf_cpu_map__nr(event_cpus); i++) { if (cpu_map__has(event_cpus, i) && !cpu_map__has(online_cpus, i)) return -EINVAL; @@ -662,7 +662,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr, cpu_map = event_cpus; } - nr_cpu = cpu_map__nr(cpu_map); + nr_cpu = perf_cpu_map__nr(cpu_map); /* Get PMU type as dynamically assigned by the core */ type = cs_etm_pmu->type; diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h index a3d27f4131be..77f85e9c88d4 100644 --- a/tools/perf/util/cpumap.h +++ b/tools/perf/util/cpumap.h @@ -49,11 +49,6 @@ static inline int cpu_map__id_to_cpu(int id) return id & 0xffff; } -static inline int cpu_map__nr(const struct perf_cpu_map *map) -{ - return map ? map->nr : 1; -} - static inline bool cpu_map__empty(const struct perf_cpu_map *map) { return map ? map->map[0] == -1 : true; diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index c4489a1ad6bc..15d1046014d7 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -370,7 +370,7 @@ static int perf_evlist__enable_event_thread(struct evlist *evlist, int thread) { int cpu; - int nr_cpus = cpu_map__nr(evlist->core.cpus); + int nr_cpus = perf_cpu_map__nr(evlist->core.cpus); if (!evsel->core.fd) return -EINVAL; @@ -396,7 +396,7 @@ int perf_evlist__enable_event_idx(struct evlist *evlist, int perf_evlist__alloc_pollfd(struct evlist *evlist) { - int nr_cpus = cpu_map__nr(evlist->core.cpus); + int nr_cpus = perf_cpu_map__nr(evlist->core.cpus); int nr_threads = thread_map__nr(evlist->core.threads); int nfds = 0; struct evsel *evsel; @@ -692,7 +692,7 @@ static struct perf_mmap *perf_evlist__alloc_mmap(struct evlist *evlist, int i; struct perf_mmap *map; - evlist->nr_mmaps = cpu_map__nr(evlist->core.cpus); + evlist->nr_mmaps = perf_cpu_map__nr(evlist->core.cpus); if (cpu_map__empty(evlist->core.cpus)) evlist->nr_mmaps = thread_map__nr(evlist->core.threads); map = zalloc(evlist->nr_mmaps * sizeof(struct perf_mmap)); @@ -807,7 +807,7 @@ static int perf_evlist__mmap_per_cpu(struct evlist *evlist, struct mmap_params *mp) { int cpu, thread; - int nr_cpus = cpu_map__nr(evlist->core.cpus); + int nr_cpus = perf_cpu_map__nr(evlist->core.cpus); int nr_threads = thread_map__nr(evlist->core.threads); pr_debug2("perf event ring buffer mmapped per cpu\n"); @@ -1014,7 +1014,7 @@ int perf_evlist__mmap_ex(struct evlist *evlist, unsigned int pages, evlist__for_each_entry(evlist, evsel) { if ((evsel->core.attr.read_format & PERF_FORMAT_ID) && evsel->sample_id == NULL && - perf_evsel__alloc_id(evsel, cpu_map__nr(cpus), threads->nr) < 0) + perf_evsel__alloc_id(evsel, perf_cpu_map__nr(cpus), threads->nr) < 0) return -ENOMEM; } diff --git a/tools/perf/util/mmap.c b/tools/perf/util/mmap.c index 42a5971146ae..5f3532e51ec9 100644 --- a/tools/perf/util/mmap.c +++ b/tools/perf/util/mmap.c @@ -331,7 +331,7 @@ static void build_node_mask(int node, cpu_set_t *mask) if (!cpu_map) return; - nr_cpus = cpu_map__nr(cpu_map); + nr_cpus = perf_cpu_map__nr(cpu_map); for (c = 0; c < nr_cpus; c++) { cpu = cpu_map->map[c]; /* map c index to online cpu index */ if (cpu__get_node(cpu) == node) diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c index f7b39f4bc51e..3df0e39ccd52 100644 --- a/tools/perf/util/stat-display.c +++ b/tools/perf/util/stat-display.c @@ -745,7 +745,7 @@ static void print_aggr_thread(struct perf_stat_config *config, { FILE *output = config->output; int nthreads = thread_map__nr(counter->core.threads); - int ncpus = cpu_map__nr(counter->core.cpus); + int ncpus = perf_cpu_map__nr(counter->core.cpus); int thread, sorted_threads, id; struct perf_aggr_thread_value *buf; -- 2.21.0 ^ permalink raw reply related [flat|nested] 11+ messages in thread
* [PATCH 3/5] libperf: Move cpu_map__empty to perf_cpu_map__empty 2019-08-22 11:11 [PATCH 0/5] tools,libperf: Assorted fixes Jiri Olsa 2019-08-22 11:11 ` [PATCH 1/5] tools: Add missing perf_event.h include Jiri Olsa 2019-08-22 11:11 ` [PATCH 2/5] libperf: Use perf_cpu_map__nr instead of cpu_map__nr Jiri Olsa @ 2019-08-22 11:11 ` Jiri Olsa 2019-08-23 12:21 ` [tip: perf/core] libperf: Move perf's cpu_map__empty() to perf_cpu_map__empty() tip-bot2 for Jiri Olsa 2019-08-22 11:11 ` [PATCH 4/5] libperf: Move cpu_map__idx to perf_cpu_map__idx Jiri Olsa 2019-08-22 11:11 ` [PATCH 5/5] libperf: Add perf_thread_map__nr/perf_thread_map__pid functions Jiri Olsa 4 siblings, 1 reply; 11+ messages in thread From: Jiri Olsa @ 2019-08-22 11:11 UTC (permalink / raw) To: Arnaldo Carvalho de Melo Cc: lkml, Ingo Molnar, Namhyung Kim, Alexander Shishkin, Peter Zijlstra, Michael Petlan So it's part of libperf library as one of basic functions over struct perf_cpu_map object. Link: http://lkml.kernel.org/n/tip-ircwjcd1xqvbf14lhtvwkw51@git.kernel.org Signed-off-by: Jiri Olsa <jolsa@kernel.org> --- tools/perf/arch/arm/util/cs-etm.c | 8 ++++---- tools/perf/arch/x86/util/intel-bts.c | 4 ++-- tools/perf/arch/x86/util/intel-pt.c | 10 +++++----- tools/perf/builtin-c2c.c | 2 +- tools/perf/builtin-stat.c | 4 ++-- tools/perf/lib/cpumap.c | 5 +++++ tools/perf/lib/include/perf/cpumap.h | 2 ++ tools/perf/lib/libperf.map | 1 + tools/perf/util/cpumap.c | 6 +++--- tools/perf/util/cpumap.h | 7 +------ tools/perf/util/event.c | 2 +- tools/perf/util/evlist.c | 6 +++--- tools/perf/util/record.c | 2 +- tools/perf/util/stat.c | 2 +- 14 files changed, 32 insertions(+), 29 deletions(-) diff --git a/tools/perf/arch/arm/util/cs-etm.c b/tools/perf/arch/arm/util/cs-etm.c index c786ab095d15..c73da3245b67 100644 --- a/tools/perf/arch/arm/util/cs-etm.c +++ b/tools/perf/arch/arm/util/cs-etm.c @@ -396,7 +396,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, * AUX event. We also need the contextID in order to be notified * when a context switch happened. */ - if (!cpu_map__empty(cpus)) { + if (!perf_cpu_map__empty(cpus)) { perf_evsel__set_sample_bit(cs_etm_evsel, CPU); err = cs_etm_set_option(itr, cs_etm_evsel, @@ -420,7 +420,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, tracking_evsel->core.attr.sample_period = 1; /* In per-cpu case, always need the time of mmap events etc */ - if (!cpu_map__empty(cpus)) + if (!perf_cpu_map__empty(cpus)) perf_evsel__set_sample_bit(tracking_evsel, TIME); } @@ -493,7 +493,7 @@ cs_etm_info_priv_size(struct auxtrace_record *itr __maybe_unused, struct perf_cpu_map *online_cpus = perf_cpu_map__new(NULL); /* cpu map is not empty, we have specific CPUs to work with */ - if (!cpu_map__empty(event_cpus)) { + if (!perf_cpu_map__empty(event_cpus)) { for (i = 0; i < cpu__max_cpu(); i++) { if (!cpu_map__has(event_cpus, i) || !cpu_map__has(online_cpus, i)) @@ -649,7 +649,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr, return -EINVAL; /* If the cpu_map is empty all online CPUs are involved */ - if (cpu_map__empty(event_cpus)) { + if (perf_cpu_map__empty(event_cpus)) { cpu_map = online_cpus; } else { /* Make sure all specified CPUs are online */ diff --git a/tools/perf/arch/x86/util/intel-bts.c b/tools/perf/arch/x86/util/intel-bts.c index 7b23318ebd7b..2d5d8a12dd1f 100644 --- a/tools/perf/arch/x86/util/intel-bts.c +++ b/tools/perf/arch/x86/util/intel-bts.c @@ -133,7 +133,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr, if (!opts->full_auxtrace) return 0; - if (opts->full_auxtrace && !cpu_map__empty(cpus)) { + if (opts->full_auxtrace && !perf_cpu_map__empty(cpus)) { pr_err(INTEL_BTS_PMU_NAME " does not support per-cpu recording\n"); return -EINVAL; } @@ -214,7 +214,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr, * In the case of per-cpu mmaps, we need the CPU on the * AUX event. */ - if (!cpu_map__empty(cpus)) + if (!perf_cpu_map__empty(cpus)) perf_evsel__set_sample_bit(intel_bts_evsel, CPU); } diff --git a/tools/perf/arch/x86/util/intel-pt.c b/tools/perf/arch/x86/util/intel-pt.c index a8e633aa278a..c72a77a82b39 100644 --- a/tools/perf/arch/x86/util/intel-pt.c +++ b/tools/perf/arch/x86/util/intel-pt.c @@ -365,7 +365,7 @@ static int intel_pt_info_fill(struct auxtrace_record *itr, ui__warning("Intel Processor Trace: TSC not available\n"); } - per_cpu_mmaps = !cpu_map__empty(session->evlist->core.cpus); + per_cpu_mmaps = !perf_cpu_map__empty(session->evlist->core.cpus); auxtrace_info->type = PERF_AUXTRACE_INTEL_PT; auxtrace_info->priv[INTEL_PT_PMU_TYPE] = intel_pt_pmu->type; @@ -702,7 +702,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * Per-cpu recording needs sched_switch events to distinguish different * threads. */ - if (have_timing_info && !cpu_map__empty(cpus)) { + if (have_timing_info && !perf_cpu_map__empty(cpus)) { if (perf_can_record_switch_events()) { bool cpu_wide = !target__none(&opts->target) && !target__has_task(&opts->target); @@ -760,7 +760,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * In the case of per-cpu mmaps, we need the CPU on the * AUX event. */ - if (!cpu_map__empty(cpus)) + if (!perf_cpu_map__empty(cpus)) perf_evsel__set_sample_bit(intel_pt_evsel, CPU); } @@ -784,7 +784,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, tracking_evsel->immediate = true; /* In per-cpu case, always need the time of mmap events etc */ - if (!cpu_map__empty(cpus)) { + if (!perf_cpu_map__empty(cpus)) { perf_evsel__set_sample_bit(tracking_evsel, TIME); /* And the CPU for switch events */ perf_evsel__set_sample_bit(tracking_evsel, CPU); @@ -796,7 +796,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * Warn the user when we do not have enough information to decode i.e. * per-cpu with no sched_switch (except workload-only). */ - if (!ptr->have_sched_switch && !cpu_map__empty(cpus) && + if (!ptr->have_sched_switch && !perf_cpu_map__empty(cpus) && !target__none(&opts->target)) ui__warning("Intel Processor Trace decoding will not be possible except for kernel tracing!\n"); diff --git a/tools/perf/builtin-c2c.c b/tools/perf/builtin-c2c.c index f0aae6e13a33..01629f5b6d1f 100644 --- a/tools/perf/builtin-c2c.c +++ b/tools/perf/builtin-c2c.c @@ -2059,7 +2059,7 @@ static int setup_nodes(struct perf_session *session) nodes[node] = set; /* empty node, skip */ - if (cpu_map__empty(map)) + if (perf_cpu_map__empty(map)) continue; for (cpu = 0; cpu < map->nr; cpu++) { diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index b19df671111e..90636a811b36 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -928,7 +928,7 @@ static int perf_stat_init_aggr_mode(void) * the aggregation translate cpumap. */ nr = cpu_map__get_max(evsel_list->core.cpus); - stat_config.cpus_aggr_map = cpu_map__empty_new(nr + 1); + stat_config.cpus_aggr_map = perf_cpu_map__empty_new(nr + 1); return stat_config.cpus_aggr_map ? 0 : -ENOMEM; } @@ -1493,7 +1493,7 @@ int process_stat_config_event(struct perf_session *session, perf_event__read_stat_config(&stat_config, &event->stat_config); - if (cpu_map__empty(st->cpus)) { + if (perf_cpu_map__empty(st->cpus)) { if (st->aggr_mode != AGGR_UNSET) pr_warning("warning: processing task data, aggregation mode not set\n"); return 0; diff --git a/tools/perf/lib/cpumap.c b/tools/perf/lib/cpumap.c index 1ddb69e796e5..63f7df7e47ff 100644 --- a/tools/perf/lib/cpumap.c +++ b/tools/perf/lib/cpumap.c @@ -237,3 +237,8 @@ int perf_cpu_map__nr(const struct perf_cpu_map *cpus) { return cpus ? cpus->nr : 1; } + +bool perf_cpu_map__empty(const struct perf_cpu_map *map) +{ + return map ? map->map[0] == -1 : true; +} diff --git a/tools/perf/lib/include/perf/cpumap.h b/tools/perf/lib/include/perf/cpumap.h index 1b6e7db3fa2b..8aa995c59498 100644 --- a/tools/perf/lib/include/perf/cpumap.h +++ b/tools/perf/lib/include/perf/cpumap.h @@ -4,6 +4,7 @@ #include <perf/core.h> #include <stdio.h> +#include <stdbool.h> struct perf_cpu_map; @@ -14,6 +15,7 @@ LIBPERF_API struct perf_cpu_map *perf_cpu_map__get(struct perf_cpu_map *map); LIBPERF_API void perf_cpu_map__put(struct perf_cpu_map *map); LIBPERF_API int perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx); LIBPERF_API int perf_cpu_map__nr(const struct perf_cpu_map *cpus); +LIBPERF_API bool perf_cpu_map__empty(const struct perf_cpu_map *map); #define perf_cpu_map__for_each_cpu(cpu, idx, cpus) \ for ((idx) = 0, (cpu) = perf_cpu_map__cpu(cpus, idx); \ diff --git a/tools/perf/lib/libperf.map b/tools/perf/lib/libperf.map index e24d3cec01c1..3373dd51fcda 100644 --- a/tools/perf/lib/libperf.map +++ b/tools/perf/lib/libperf.map @@ -8,6 +8,7 @@ LIBPERF_0.0.1 { perf_cpu_map__read; perf_cpu_map__nr; perf_cpu_map__cpu; + perf_cpu_map__empty; perf_thread_map__new_dummy; perf_thread_map__set_pid; perf_thread_map__comm; diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c index beb3525e9e45..4402e67445a4 100644 --- a/tools/perf/util/cpumap.c +++ b/tools/perf/util/cpumap.c @@ -21,7 +21,7 @@ static struct perf_cpu_map *cpu_map__from_entries(struct cpu_map_entries *cpus) { struct perf_cpu_map *map; - map = cpu_map__empty_new(cpus->nr); + map = perf_cpu_map__empty_new(cpus->nr); if (map) { unsigned i; @@ -48,7 +48,7 @@ static struct perf_cpu_map *cpu_map__from_mask(struct cpu_map_mask *mask) nr = bitmap_weight(mask->mask, nbits); - map = cpu_map__empty_new(nr); + map = perf_cpu_map__empty_new(nr); if (map) { int cpu, i = 0; @@ -77,7 +77,7 @@ size_t cpu_map__fprintf(struct perf_cpu_map *map, FILE *fp) #undef BUFSIZE } -struct perf_cpu_map *cpu_map__empty_new(int nr) +struct perf_cpu_map *perf_cpu_map__empty_new(int nr) { struct perf_cpu_map *cpus = malloc(sizeof(*cpus) + sizeof(int) * nr); diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h index 77f85e9c88d4..3e068090612f 100644 --- a/tools/perf/util/cpumap.h +++ b/tools/perf/util/cpumap.h @@ -11,7 +11,7 @@ #include "perf.h" #include "util/debug.h" -struct perf_cpu_map *cpu_map__empty_new(int nr); +struct perf_cpu_map *perf_cpu_map__empty_new(int nr); struct perf_cpu_map *cpu_map__new_data(struct cpu_map_data *data); size_t cpu_map__snprint(struct perf_cpu_map *map, char *buf, size_t size); size_t cpu_map__snprint_mask(struct perf_cpu_map *map, char *buf, size_t size); @@ -49,11 +49,6 @@ static inline int cpu_map__id_to_cpu(int id) return id & 0xffff; } -static inline bool cpu_map__empty(const struct perf_cpu_map *map) -{ - return map ? map->map[0] == -1 : true; -} - int cpu__setup_cpunode_map(void); int cpu__max_node(void); diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index f440fdc3e953..f433da85c45e 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -1055,7 +1055,7 @@ static size_t mask_size(struct perf_cpu_map *map, int *max) void *cpu_map_data__alloc(struct perf_cpu_map *map, size_t *size, u16 *type, int *max) { size_t size_cpus, size_mask; - bool is_dummy = cpu_map__empty(map); + bool is_dummy = perf_cpu_map__empty(map); /* * Both array and mask data have variable size based diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index 15d1046014d7..ba49b5ecffd0 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -386,7 +386,7 @@ static int perf_evlist__enable_event_thread(struct evlist *evlist, int perf_evlist__enable_event_idx(struct evlist *evlist, struct evsel *evsel, int idx) { - bool per_cpu_mmaps = !cpu_map__empty(evlist->core.cpus); + bool per_cpu_mmaps = !perf_cpu_map__empty(evlist->core.cpus); if (per_cpu_mmaps) return perf_evlist__enable_event_cpu(evlist, evsel, idx); @@ -693,7 +693,7 @@ static struct perf_mmap *perf_evlist__alloc_mmap(struct evlist *evlist, struct perf_mmap *map; evlist->nr_mmaps = perf_cpu_map__nr(evlist->core.cpus); - if (cpu_map__empty(evlist->core.cpus)) + if (perf_cpu_map__empty(evlist->core.cpus)) evlist->nr_mmaps = thread_map__nr(evlist->core.threads); map = zalloc(evlist->nr_mmaps * sizeof(struct perf_mmap)); if (!map) @@ -1018,7 +1018,7 @@ int perf_evlist__mmap_ex(struct evlist *evlist, unsigned int pages, return -ENOMEM; } - if (cpu_map__empty(cpus)) + if (perf_cpu_map__empty(cpus)) return perf_evlist__mmap_per_thread(evlist, &mp); return perf_evlist__mmap_per_cpu(evlist, &mp); diff --git a/tools/perf/util/record.c b/tools/perf/util/record.c index e59382d99196..51bbd0714e6d 100644 --- a/tools/perf/util/record.c +++ b/tools/perf/util/record.c @@ -275,7 +275,7 @@ bool perf_evlist__can_select_event(struct evlist *evlist, const char *str) evsel = perf_evlist__last(temp_evlist); - if (!evlist || cpu_map__empty(evlist->core.cpus)) { + if (!evlist || perf_cpu_map__empty(evlist->core.cpus)) { struct perf_cpu_map *cpus = perf_cpu_map__new(NULL); cpu = cpus ? cpus->map[0] : 0; diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index e4e4e3bf8b2b..2715112290cf 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -223,7 +223,7 @@ static int check_per_pkg(struct evsel *counter, if (!counter->per_pkg) return 0; - if (cpu_map__empty(cpus)) + if (perf_cpu_map__empty(cpus)) return 0; if (!mask) { -- 2.21.0 ^ permalink raw reply related [flat|nested] 11+ messages in thread
* [tip: perf/core] libperf: Move perf's cpu_map__empty() to perf_cpu_map__empty() 2019-08-22 11:11 ` [PATCH 3/5] libperf: Move cpu_map__empty to perf_cpu_map__empty Jiri Olsa @ 2019-08-23 12:21 ` tip-bot2 for Jiri Olsa 0 siblings, 0 replies; 11+ messages in thread From: tip-bot2 for Jiri Olsa @ 2019-08-23 12:21 UTC (permalink / raw) To: linux-tip-commits Cc: Jiri Olsa, Alexander Shishkin, Michael Petlan, Namhyung Kim, Peter Zijlstra, Arnaldo Carvalho de Melo, Ingo Molnar, linux-kernel The following commit has been merged into the perf/core branch of tip: Commit-ID: 315c0a1f0ccdd44c65f80ccbc62202fed8a23050 Gitweb: https://git.kernel.org/tip/315c0a1f0ccdd44c65f80ccbc62202fed8a23050 Author: Jiri Olsa <jolsa@kernel.org> AuthorDate: Thu, 22 Aug 2019 13:11:39 +02:00 Committer: Arnaldo Carvalho de Melo <acme@redhat.com> CommitterDate: Thu, 22 Aug 2019 11:17:03 -03:00 libperf: Move perf's cpu_map__empty() to perf_cpu_map__empty() So it's part of the libperf library as one of basic functions operating on the perf_cpu_map class. Signed-off-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Michael Petlan <mpetlan@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20190822111141.25823-4-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> --- tools/perf/arch/arm/util/cs-etm.c | 8 ++++---- tools/perf/arch/x86/util/intel-bts.c | 4 ++-- tools/perf/arch/x86/util/intel-pt.c | 10 +++++----- tools/perf/builtin-c2c.c | 2 +- tools/perf/builtin-stat.c | 4 ++-- tools/perf/lib/cpumap.c | 5 +++++ tools/perf/lib/include/perf/cpumap.h | 2 ++ tools/perf/lib/libperf.map | 1 + tools/perf/util/cpumap.c | 6 +++--- tools/perf/util/cpumap.h | 7 +------ tools/perf/util/event.c | 2 +- tools/perf/util/evlist.c | 6 +++--- tools/perf/util/record.c | 2 +- tools/perf/util/stat.c | 2 +- 14 files changed, 32 insertions(+), 29 deletions(-) diff --git a/tools/perf/arch/arm/util/cs-etm.c b/tools/perf/arch/arm/util/cs-etm.c index c786ab0..c73da32 100644 --- a/tools/perf/arch/arm/util/cs-etm.c +++ b/tools/perf/arch/arm/util/cs-etm.c @@ -396,7 +396,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, * AUX event. We also need the contextID in order to be notified * when a context switch happened. */ - if (!cpu_map__empty(cpus)) { + if (!perf_cpu_map__empty(cpus)) { perf_evsel__set_sample_bit(cs_etm_evsel, CPU); err = cs_etm_set_option(itr, cs_etm_evsel, @@ -420,7 +420,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr, tracking_evsel->core.attr.sample_period = 1; /* In per-cpu case, always need the time of mmap events etc */ - if (!cpu_map__empty(cpus)) + if (!perf_cpu_map__empty(cpus)) perf_evsel__set_sample_bit(tracking_evsel, TIME); } @@ -493,7 +493,7 @@ cs_etm_info_priv_size(struct auxtrace_record *itr __maybe_unused, struct perf_cpu_map *online_cpus = perf_cpu_map__new(NULL); /* cpu map is not empty, we have specific CPUs to work with */ - if (!cpu_map__empty(event_cpus)) { + if (!perf_cpu_map__empty(event_cpus)) { for (i = 0; i < cpu__max_cpu(); i++) { if (!cpu_map__has(event_cpus, i) || !cpu_map__has(online_cpus, i)) @@ -649,7 +649,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr, return -EINVAL; /* If the cpu_map is empty all online CPUs are involved */ - if (cpu_map__empty(event_cpus)) { + if (perf_cpu_map__empty(event_cpus)) { cpu_map = online_cpus; } else { /* Make sure all specified CPUs are online */ diff --git a/tools/perf/arch/x86/util/intel-bts.c b/tools/perf/arch/x86/util/intel-bts.c index 7b23318..2d5d8a1 100644 --- a/tools/perf/arch/x86/util/intel-bts.c +++ b/tools/perf/arch/x86/util/intel-bts.c @@ -133,7 +133,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr, if (!opts->full_auxtrace) return 0; - if (opts->full_auxtrace && !cpu_map__empty(cpus)) { + if (opts->full_auxtrace && !perf_cpu_map__empty(cpus)) { pr_err(INTEL_BTS_PMU_NAME " does not support per-cpu recording\n"); return -EINVAL; } @@ -214,7 +214,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr, * In the case of per-cpu mmaps, we need the CPU on the * AUX event. */ - if (!cpu_map__empty(cpus)) + if (!perf_cpu_map__empty(cpus)) perf_evsel__set_sample_bit(intel_bts_evsel, CPU); } diff --git a/tools/perf/arch/x86/util/intel-pt.c b/tools/perf/arch/x86/util/intel-pt.c index a8e633a..c72a77a 100644 --- a/tools/perf/arch/x86/util/intel-pt.c +++ b/tools/perf/arch/x86/util/intel-pt.c @@ -365,7 +365,7 @@ static int intel_pt_info_fill(struct auxtrace_record *itr, ui__warning("Intel Processor Trace: TSC not available\n"); } - per_cpu_mmaps = !cpu_map__empty(session->evlist->core.cpus); + per_cpu_mmaps = !perf_cpu_map__empty(session->evlist->core.cpus); auxtrace_info->type = PERF_AUXTRACE_INTEL_PT; auxtrace_info->priv[INTEL_PT_PMU_TYPE] = intel_pt_pmu->type; @@ -702,7 +702,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * Per-cpu recording needs sched_switch events to distinguish different * threads. */ - if (have_timing_info && !cpu_map__empty(cpus)) { + if (have_timing_info && !perf_cpu_map__empty(cpus)) { if (perf_can_record_switch_events()) { bool cpu_wide = !target__none(&opts->target) && !target__has_task(&opts->target); @@ -760,7 +760,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * In the case of per-cpu mmaps, we need the CPU on the * AUX event. */ - if (!cpu_map__empty(cpus)) + if (!perf_cpu_map__empty(cpus)) perf_evsel__set_sample_bit(intel_pt_evsel, CPU); } @@ -784,7 +784,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, tracking_evsel->immediate = true; /* In per-cpu case, always need the time of mmap events etc */ - if (!cpu_map__empty(cpus)) { + if (!perf_cpu_map__empty(cpus)) { perf_evsel__set_sample_bit(tracking_evsel, TIME); /* And the CPU for switch events */ perf_evsel__set_sample_bit(tracking_evsel, CPU); @@ -796,7 +796,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, * Warn the user when we do not have enough information to decode i.e. * per-cpu with no sched_switch (except workload-only). */ - if (!ptr->have_sched_switch && !cpu_map__empty(cpus) && + if (!ptr->have_sched_switch && !perf_cpu_map__empty(cpus) && !target__none(&opts->target)) ui__warning("Intel Processor Trace decoding will not be possible except for kernel tracing!\n"); diff --git a/tools/perf/builtin-c2c.c b/tools/perf/builtin-c2c.c index f0aae6e..01629f5 100644 --- a/tools/perf/builtin-c2c.c +++ b/tools/perf/builtin-c2c.c @@ -2059,7 +2059,7 @@ static int setup_nodes(struct perf_session *session) nodes[node] = set; /* empty node, skip */ - if (cpu_map__empty(map)) + if (perf_cpu_map__empty(map)) continue; for (cpu = 0; cpu < map->nr; cpu++) { diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index b19df67..90636a8 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -928,7 +928,7 @@ static int perf_stat_init_aggr_mode(void) * the aggregation translate cpumap. */ nr = cpu_map__get_max(evsel_list->core.cpus); - stat_config.cpus_aggr_map = cpu_map__empty_new(nr + 1); + stat_config.cpus_aggr_map = perf_cpu_map__empty_new(nr + 1); return stat_config.cpus_aggr_map ? 0 : -ENOMEM; } @@ -1493,7 +1493,7 @@ int process_stat_config_event(struct perf_session *session, perf_event__read_stat_config(&stat_config, &event->stat_config); - if (cpu_map__empty(st->cpus)) { + if (perf_cpu_map__empty(st->cpus)) { if (st->aggr_mode != AGGR_UNSET) pr_warning("warning: processing task data, aggregation mode not set\n"); return 0; diff --git a/tools/perf/lib/cpumap.c b/tools/perf/lib/cpumap.c index 1ddb69e..63f7df7 100644 --- a/tools/perf/lib/cpumap.c +++ b/tools/perf/lib/cpumap.c @@ -237,3 +237,8 @@ int perf_cpu_map__nr(const struct perf_cpu_map *cpus) { return cpus ? cpus->nr : 1; } + +bool perf_cpu_map__empty(const struct perf_cpu_map *map) +{ + return map ? map->map[0] == -1 : true; +} diff --git a/tools/perf/lib/include/perf/cpumap.h b/tools/perf/lib/include/perf/cpumap.h index 1b6e7db..8aa995c 100644 --- a/tools/perf/lib/include/perf/cpumap.h +++ b/tools/perf/lib/include/perf/cpumap.h @@ -4,6 +4,7 @@ #include <perf/core.h> #include <stdio.h> +#include <stdbool.h> struct perf_cpu_map; @@ -14,6 +15,7 @@ LIBPERF_API struct perf_cpu_map *perf_cpu_map__get(struct perf_cpu_map *map); LIBPERF_API void perf_cpu_map__put(struct perf_cpu_map *map); LIBPERF_API int perf_cpu_map__cpu(const struct perf_cpu_map *cpus, int idx); LIBPERF_API int perf_cpu_map__nr(const struct perf_cpu_map *cpus); +LIBPERF_API bool perf_cpu_map__empty(const struct perf_cpu_map *map); #define perf_cpu_map__for_each_cpu(cpu, idx, cpus) \ for ((idx) = 0, (cpu) = perf_cpu_map__cpu(cpus, idx); \ diff --git a/tools/perf/lib/libperf.map b/tools/perf/lib/libperf.map index e24d3ce..3373dd5 100644 --- a/tools/perf/lib/libperf.map +++ b/tools/perf/lib/libperf.map @@ -8,6 +8,7 @@ LIBPERF_0.0.1 { perf_cpu_map__read; perf_cpu_map__nr; perf_cpu_map__cpu; + perf_cpu_map__empty; perf_thread_map__new_dummy; perf_thread_map__set_pid; perf_thread_map__comm; diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c index beb3525..4402e67 100644 --- a/tools/perf/util/cpumap.c +++ b/tools/perf/util/cpumap.c @@ -21,7 +21,7 @@ static struct perf_cpu_map *cpu_map__from_entries(struct cpu_map_entries *cpus) { struct perf_cpu_map *map; - map = cpu_map__empty_new(cpus->nr); + map = perf_cpu_map__empty_new(cpus->nr); if (map) { unsigned i; @@ -48,7 +48,7 @@ static struct perf_cpu_map *cpu_map__from_mask(struct cpu_map_mask *mask) nr = bitmap_weight(mask->mask, nbits); - map = cpu_map__empty_new(nr); + map = perf_cpu_map__empty_new(nr); if (map) { int cpu, i = 0; @@ -77,7 +77,7 @@ size_t cpu_map__fprintf(struct perf_cpu_map *map, FILE *fp) #undef BUFSIZE } -struct perf_cpu_map *cpu_map__empty_new(int nr) +struct perf_cpu_map *perf_cpu_map__empty_new(int nr) { struct perf_cpu_map *cpus = malloc(sizeof(*cpus) + sizeof(int) * nr); diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h index 77f85e9..3e06809 100644 --- a/tools/perf/util/cpumap.h +++ b/tools/perf/util/cpumap.h @@ -11,7 +11,7 @@ #include "perf.h" #include "util/debug.h" -struct perf_cpu_map *cpu_map__empty_new(int nr); +struct perf_cpu_map *perf_cpu_map__empty_new(int nr); struct perf_cpu_map *cpu_map__new_data(struct cpu_map_data *data); size_t cpu_map__snprint(struct perf_cpu_map *map, char *buf, size_t size); size_t cpu_map__snprint_mask(struct perf_cpu_map *map, char *buf, size_t size); @@ -49,11 +49,6 @@ static inline int cpu_map__id_to_cpu(int id) return id & 0xffff; } -static inline bool cpu_map__empty(const struct perf_cpu_map *map) -{ - return map ? map->map[0] == -1 : true; -} - int cpu__setup_cpunode_map(void); int cpu__max_node(void); diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index f440fdc..f433da8 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -1055,7 +1055,7 @@ static size_t mask_size(struct perf_cpu_map *map, int *max) void *cpu_map_data__alloc(struct perf_cpu_map *map, size_t *size, u16 *type, int *max) { size_t size_cpus, size_mask; - bool is_dummy = cpu_map__empty(map); + bool is_dummy = perf_cpu_map__empty(map); /* * Both array and mask data have variable size based diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index 15d1046..ba49b5e 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -386,7 +386,7 @@ static int perf_evlist__enable_event_thread(struct evlist *evlist, int perf_evlist__enable_event_idx(struct evlist *evlist, struct evsel *evsel, int idx) { - bool per_cpu_mmaps = !cpu_map__empty(evlist->core.cpus); + bool per_cpu_mmaps = !perf_cpu_map__empty(evlist->core.cpus); if (per_cpu_mmaps) return perf_evlist__enable_event_cpu(evlist, evsel, idx); @@ -693,7 +693,7 @@ static struct perf_mmap *perf_evlist__alloc_mmap(struct evlist *evlist, struct perf_mmap *map; evlist->nr_mmaps = perf_cpu_map__nr(evlist->core.cpus); - if (cpu_map__empty(evlist->core.cpus)) + if (perf_cpu_map__empty(evlist->core.cpus)) evlist->nr_mmaps = thread_map__nr(evlist->core.threads); map = zalloc(evlist->nr_mmaps * sizeof(struct perf_mmap)); if (!map) @@ -1018,7 +1018,7 @@ int perf_evlist__mmap_ex(struct evlist *evlist, unsigned int pages, return -ENOMEM; } - if (cpu_map__empty(cpus)) + if (perf_cpu_map__empty(cpus)) return perf_evlist__mmap_per_thread(evlist, &mp); return perf_evlist__mmap_per_cpu(evlist, &mp); diff --git a/tools/perf/util/record.c b/tools/perf/util/record.c index e59382d..51bbd07 100644 --- a/tools/perf/util/record.c +++ b/tools/perf/util/record.c @@ -275,7 +275,7 @@ bool perf_evlist__can_select_event(struct evlist *evlist, const char *str) evsel = perf_evlist__last(temp_evlist); - if (!evlist || cpu_map__empty(evlist->core.cpus)) { + if (!evlist || perf_cpu_map__empty(evlist->core.cpus)) { struct perf_cpu_map *cpus = perf_cpu_map__new(NULL); cpu = cpus ? cpus->map[0] : 0; diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index e4e4e3b..2715112 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -223,7 +223,7 @@ static int check_per_pkg(struct evsel *counter, if (!counter->per_pkg) return 0; - if (cpu_map__empty(cpus)) + if (perf_cpu_map__empty(cpus)) return 0; if (!mask) { ^ permalink raw reply related [flat|nested] 11+ messages in thread
* [PATCH 4/5] libperf: Move cpu_map__idx to perf_cpu_map__idx 2019-08-22 11:11 [PATCH 0/5] tools,libperf: Assorted fixes Jiri Olsa ` (2 preceding siblings ...) 2019-08-22 11:11 ` [PATCH 3/5] libperf: Move cpu_map__empty to perf_cpu_map__empty Jiri Olsa @ 2019-08-22 11:11 ` Jiri Olsa 2019-08-23 12:21 ` [tip: perf/core] libperf: Move perf's cpu_map__idx() to perf_cpu_map__idx() tip-bot2 for Jiri Olsa 2019-08-22 11:11 ` [PATCH 5/5] libperf: Add perf_thread_map__nr/perf_thread_map__pid functions Jiri Olsa 4 siblings, 1 reply; 11+ messages in thread From: Jiri Olsa @ 2019-08-22 11:11 UTC (permalink / raw) To: Arnaldo Carvalho de Melo Cc: lkml, Ingo Molnar, Namhyung Kim, Alexander Shishkin, Peter Zijlstra, Michael Petlan As an internal function that will be used by both perf and libperf, but is not exported at this point. Link: http://lkml.kernel.org/n/tip-wa4bngnci9395p6xk0nfmj9h@git.kernel.org Signed-off-by: Jiri Olsa <jolsa@kernel.org> --- tools/perf/lib/cpumap.c | 12 ++++++++++++ tools/perf/lib/include/internal/cpumap.h | 2 ++ tools/perf/util/cpumap.c | 14 +------------- tools/perf/util/cpumap.h | 1 - tools/perf/util/evlist.c | 2 +- 5 files changed, 16 insertions(+), 15 deletions(-) diff --git a/tools/perf/lib/cpumap.c b/tools/perf/lib/cpumap.c index 63f7df7e47ff..2834753576b2 100644 --- a/tools/perf/lib/cpumap.c +++ b/tools/perf/lib/cpumap.c @@ -242,3 +242,15 @@ bool perf_cpu_map__empty(const struct perf_cpu_map *map) { return map ? map->map[0] == -1 : true; } + +int perf_cpu_map__idx(struct perf_cpu_map *cpus, int cpu) +{ + int i; + + for (i = 0; i < cpus->nr; ++i) { + if (cpus->map[i] == cpu) + return i; + } + + return -1; +} diff --git a/tools/perf/lib/include/internal/cpumap.h b/tools/perf/lib/include/internal/cpumap.h index 3306319f7df8..840d4032587b 100644 --- a/tools/perf/lib/include/internal/cpumap.h +++ b/tools/perf/lib/include/internal/cpumap.h @@ -14,4 +14,6 @@ struct perf_cpu_map { #define MAX_NR_CPUS 2048 #endif +int perf_cpu_map__idx(struct perf_cpu_map *cpus, int cpu); + #endif /* __LIBPERF_INTERNAL_CPUMAP_H */ diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c index 4402e67445a4..8e6c2cbffedc 100644 --- a/tools/perf/util/cpumap.c +++ b/tools/perf/util/cpumap.c @@ -458,19 +458,7 @@ int cpu__setup_cpunode_map(void) bool cpu_map__has(struct perf_cpu_map *cpus, int cpu) { - return cpu_map__idx(cpus, cpu) != -1; -} - -int cpu_map__idx(struct perf_cpu_map *cpus, int cpu) -{ - int i; - - for (i = 0; i < cpus->nr; ++i) { - if (cpus->map[i] == cpu) - return i; - } - - return -1; + return perf_cpu_map__idx(cpus, cpu) != -1; } int cpu_map__cpu(struct perf_cpu_map *cpus, int idx) diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h index 3e068090612f..8dbedda7af45 100644 --- a/tools/perf/util/cpumap.h +++ b/tools/perf/util/cpumap.h @@ -62,5 +62,4 @@ int cpu_map__build_map(struct perf_cpu_map *cpus, struct perf_cpu_map **res, int cpu_map__cpu(struct perf_cpu_map *cpus, int idx); bool cpu_map__has(struct perf_cpu_map *cpus, int cpu); -int cpu_map__idx(struct perf_cpu_map *cpus, int cpu); #endif /* __PERF_CPUMAP_H */ diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index ba49b5ecffd0..8582560b59af 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -758,7 +758,7 @@ static int perf_evlist__mmap_per_evsel(struct evlist *evlist, int idx, if (evsel->system_wide && thread) continue; - cpu = cpu_map__idx(evsel->core.cpus, evlist_cpu); + cpu = perf_cpu_map__idx(evsel->core.cpus, evlist_cpu); if (cpu == -1) continue; -- 2.21.0 ^ permalink raw reply related [flat|nested] 11+ messages in thread
* [tip: perf/core] libperf: Move perf's cpu_map__idx() to perf_cpu_map__idx() 2019-08-22 11:11 ` [PATCH 4/5] libperf: Move cpu_map__idx to perf_cpu_map__idx Jiri Olsa @ 2019-08-23 12:21 ` tip-bot2 for Jiri Olsa 0 siblings, 0 replies; 11+ messages in thread From: tip-bot2 for Jiri Olsa @ 2019-08-23 12:21 UTC (permalink / raw) To: linux-tip-commits Cc: Jiri Olsa, Alexander Shishkin, Michael Petlan, Namhyung Kim, Peter Zijlstra, Arnaldo Carvalho de Melo, Ingo Molnar, linux-kernel The following commit has been merged into the perf/core branch of tip: Commit-ID: b4df75de3b3930703415aa053a269ae07c78d9b2 Gitweb: https://git.kernel.org/tip/b4df75de3b3930703415aa053a269ae07c78d9b2 Author: Jiri Olsa <jolsa@kernel.org> AuthorDate: Thu, 22 Aug 2019 13:11:40 +02:00 Committer: Arnaldo Carvalho de Melo <acme@redhat.com> CommitterDate: Thu, 22 Aug 2019 11:18:45 -03:00 libperf: Move perf's cpu_map__idx() to perf_cpu_map__idx() As an internal function that will be used by both perf and libperf, but is not exported at this point. Signed-off-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Michael Petlan <mpetlan@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20190822111141.25823-5-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> --- tools/perf/lib/cpumap.c | 12 ++++++++++++ tools/perf/lib/include/internal/cpumap.h | 2 ++ tools/perf/util/cpumap.c | 14 +------------- tools/perf/util/cpumap.h | 1 - tools/perf/util/evlist.c | 2 +- 5 files changed, 16 insertions(+), 15 deletions(-) diff --git a/tools/perf/lib/cpumap.c b/tools/perf/lib/cpumap.c index 63f7df7..2834753 100644 --- a/tools/perf/lib/cpumap.c +++ b/tools/perf/lib/cpumap.c @@ -242,3 +242,15 @@ bool perf_cpu_map__empty(const struct perf_cpu_map *map) { return map ? map->map[0] == -1 : true; } + +int perf_cpu_map__idx(struct perf_cpu_map *cpus, int cpu) +{ + int i; + + for (i = 0; i < cpus->nr; ++i) { + if (cpus->map[i] == cpu) + return i; + } + + return -1; +} diff --git a/tools/perf/lib/include/internal/cpumap.h b/tools/perf/lib/include/internal/cpumap.h index 3306319..840d403 100644 --- a/tools/perf/lib/include/internal/cpumap.h +++ b/tools/perf/lib/include/internal/cpumap.h @@ -14,4 +14,6 @@ struct perf_cpu_map { #define MAX_NR_CPUS 2048 #endif +int perf_cpu_map__idx(struct perf_cpu_map *cpus, int cpu); + #endif /* __LIBPERF_INTERNAL_CPUMAP_H */ diff --git a/tools/perf/util/cpumap.c b/tools/perf/util/cpumap.c index 4402e67..8e6c2cb 100644 --- a/tools/perf/util/cpumap.c +++ b/tools/perf/util/cpumap.c @@ -458,19 +458,7 @@ int cpu__setup_cpunode_map(void) bool cpu_map__has(struct perf_cpu_map *cpus, int cpu) { - return cpu_map__idx(cpus, cpu) != -1; -} - -int cpu_map__idx(struct perf_cpu_map *cpus, int cpu) -{ - int i; - - for (i = 0; i < cpus->nr; ++i) { - if (cpus->map[i] == cpu) - return i; - } - - return -1; + return perf_cpu_map__idx(cpus, cpu) != -1; } int cpu_map__cpu(struct perf_cpu_map *cpus, int idx) diff --git a/tools/perf/util/cpumap.h b/tools/perf/util/cpumap.h index 3e06809..8dbedda 100644 --- a/tools/perf/util/cpumap.h +++ b/tools/perf/util/cpumap.h @@ -62,5 +62,4 @@ int cpu_map__build_map(struct perf_cpu_map *cpus, struct perf_cpu_map **res, int cpu_map__cpu(struct perf_cpu_map *cpus, int idx); bool cpu_map__has(struct perf_cpu_map *cpus, int cpu); -int cpu_map__idx(struct perf_cpu_map *cpus, int cpu); #endif /* __PERF_CPUMAP_H */ diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index ba49b5e..8582560 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -758,7 +758,7 @@ static int perf_evlist__mmap_per_evsel(struct evlist *evlist, int idx, if (evsel->system_wide && thread) continue; - cpu = cpu_map__idx(evsel->core.cpus, evlist_cpu); + cpu = perf_cpu_map__idx(evsel->core.cpus, evlist_cpu); if (cpu == -1) continue; ^ permalink raw reply related [flat|nested] 11+ messages in thread
* [PATCH 5/5] libperf: Add perf_thread_map__nr/perf_thread_map__pid functions 2019-08-22 11:11 [PATCH 0/5] tools,libperf: Assorted fixes Jiri Olsa ` (3 preceding siblings ...) 2019-08-22 11:11 ` [PATCH 4/5] libperf: Move cpu_map__idx to perf_cpu_map__idx Jiri Olsa @ 2019-08-22 11:11 ` Jiri Olsa 2019-08-22 15:18 ` Arnaldo Carvalho de Melo 2019-08-23 12:29 ` [tip: perf/core] " tip-bot2 for Jiri Olsa 4 siblings, 2 replies; 11+ messages in thread From: Jiri Olsa @ 2019-08-22 11:11 UTC (permalink / raw) To: Arnaldo Carvalho de Melo Cc: lkml, Ingo Molnar, Namhyung Kim, Alexander Shishkin, Peter Zijlstra, Michael Petlan So it's part of libperf library as basic functions over struct perf_thread_map object. Link: http://lkml.kernel.org/n/tip-mjksprk4yct9ouez3j9mq2az@git.kernel.org Signed-off-by: Jiri Olsa <jolsa@kernel.org> --- tools/perf/builtin-ftrace.c | 2 +- tools/perf/builtin-script.c | 4 ++-- tools/perf/builtin-stat.c | 4 ++-- tools/perf/builtin-trace.c | 4 ++-- tools/perf/lib/include/perf/threadmap.h | 2 ++ tools/perf/lib/libperf.map | 2 ++ tools/perf/lib/threadmap.c | 10 ++++++++++ tools/perf/tests/thread-map.c | 6 +++--- tools/perf/util/auxtrace.c | 4 ++-- tools/perf/util/event.c | 8 ++++---- tools/perf/util/evlist.c | 12 ++++++------ tools/perf/util/evsel.c | 4 ++-- .../perf/util/scripting-engines/trace-event-python.c | 2 +- tools/perf/util/stat-display.c | 4 ++-- tools/perf/util/stat.c | 4 ++-- tools/perf/util/thread_map.c | 4 ++-- tools/perf/util/thread_map.h | 10 ---------- 17 files changed, 45 insertions(+), 41 deletions(-) diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c index 1367bb5046a7..565db782c1b9 100644 --- a/tools/perf/builtin-ftrace.c +++ b/tools/perf/builtin-ftrace.c @@ -158,7 +158,7 @@ static int set_tracing_pid(struct perf_ftrace *ftrace) if (target__has_cpu(&ftrace->target)) return 0; - for (i = 0; i < thread_map__nr(ftrace->evlist->core.threads); i++) { + for (i = 0; i < perf_thread_map__nr(ftrace->evlist->core.threads); i++) { scnprintf(buf, sizeof(buf), "%d", ftrace->evlist->core.threads->map[i]); if (append_tracing_file("set_ftrace_pid", buf) < 0) diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c index 1764efd16cd4..5d45be1d3885 100644 --- a/tools/perf/builtin-script.c +++ b/tools/perf/builtin-script.c @@ -1905,7 +1905,7 @@ static struct scripting_ops *scripting_ops; static void __process_stat(struct evsel *counter, u64 tstamp) { - int nthreads = thread_map__nr(counter->core.threads); + int nthreads = perf_thread_map__nr(counter->core.threads); int ncpus = perf_evsel__nr_cpus(counter); int cpu, thread; static int header_printed; @@ -1927,7 +1927,7 @@ static void __process_stat(struct evsel *counter, u64 tstamp) printf("%3d %8d %15" PRIu64 " %15" PRIu64 " %15" PRIu64 " %15" PRIu64 " %s\n", counter->core.cpus->map[cpu], - thread_map__pid(counter->core.threads, thread), + perf_thread_map__pid(counter->core.threads, thread), counts->val, counts->ena, counts->run, diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 90636a811b36..8a4f1a7d0cba 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -264,7 +264,7 @@ static int read_single_counter(struct evsel *counter, int cpu, */ static int read_counter(struct evsel *counter, struct timespec *rs) { - int nthreads = thread_map__nr(evsel_list->core.threads); + int nthreads = perf_thread_map__nr(evsel_list->core.threads); int ncpus, cpu, thread; if (target__has_cpu(&target) && !target__has_per_thread(&target)) @@ -1893,7 +1893,7 @@ int cmd_stat(int argc, const char **argv) thread_map__read_comms(evsel_list->core.threads); if (target.system_wide) { if (runtime_stat_new(&stat_config, - thread_map__nr(evsel_list->core.threads))) { + perf_thread_map__nr(evsel_list->core.threads))) { goto out; } } diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index bc44ed29e05a..de126258ca10 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -3188,7 +3188,7 @@ static int trace__set_filter_pids(struct trace *trace) err = bpf_map__set_filter_pids(trace->filter_pids.map, trace->filter_pids.nr, trace->filter_pids.entries); } - } else if (thread_map__pid(trace->evlist->core.threads, 0) == -1) { + } else if (perf_thread_map__pid(trace->evlist->core.threads, 0) == -1) { err = trace__set_filter_loop_pids(trace); } @@ -3417,7 +3417,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) evlist__enable(evlist); } - trace->multiple_threads = thread_map__pid(evlist->core.threads, 0) == -1 || + trace->multiple_threads = perf_thread_map__pid(evlist->core.threads, 0) == -1 || evlist->core.threads->nr > 1 || perf_evlist__first(evlist)->core.attr.inherit; diff --git a/tools/perf/lib/include/perf/threadmap.h b/tools/perf/lib/include/perf/threadmap.h index 456295273daa..a7c50de8d010 100644 --- a/tools/perf/lib/include/perf/threadmap.h +++ b/tools/perf/lib/include/perf/threadmap.h @@ -11,6 +11,8 @@ LIBPERF_API struct perf_thread_map *perf_thread_map__new_dummy(void); LIBPERF_API void perf_thread_map__set_pid(struct perf_thread_map *map, int thread, pid_t pid); LIBPERF_API char *perf_thread_map__comm(struct perf_thread_map *map, int thread); +LIBPERF_API int perf_thread_map__nr(struct perf_thread_map *threads); +LIBPERF_API pid_t perf_thread_map__pid(struct perf_thread_map *map, int thread); LIBPERF_API struct perf_thread_map *perf_thread_map__get(struct perf_thread_map *map); LIBPERF_API void perf_thread_map__put(struct perf_thread_map *map); diff --git a/tools/perf/lib/libperf.map b/tools/perf/lib/libperf.map index 3373dd51fcda..dc4d66363bc4 100644 --- a/tools/perf/lib/libperf.map +++ b/tools/perf/lib/libperf.map @@ -12,6 +12,8 @@ LIBPERF_0.0.1 { perf_thread_map__new_dummy; perf_thread_map__set_pid; perf_thread_map__comm; + perf_thread_map__nr; + perf_thread_map__pid; perf_thread_map__get; perf_thread_map__put; perf_evsel__new; diff --git a/tools/perf/lib/threadmap.c b/tools/perf/lib/threadmap.c index 4865b73e2586..e92c368b0a6c 100644 --- a/tools/perf/lib/threadmap.c +++ b/tools/perf/lib/threadmap.c @@ -79,3 +79,13 @@ void perf_thread_map__put(struct perf_thread_map *map) if (map && refcount_dec_and_test(&map->refcnt)) perf_thread_map__delete(map); } + +int perf_thread_map__nr(struct perf_thread_map *threads) +{ + return threads ? threads->nr : 1; +} + +pid_t perf_thread_map__pid(struct perf_thread_map *map, int thread) +{ + return map->map[thread].pid; +} diff --git a/tools/perf/tests/thread-map.c b/tools/perf/tests/thread-map.c index d61773cacf0b..d803eafedc60 100644 --- a/tools/perf/tests/thread-map.c +++ b/tools/perf/tests/thread-map.c @@ -26,7 +26,7 @@ int test__thread_map(struct test *test __maybe_unused, int subtest __maybe_unuse TEST_ASSERT_VAL("wrong nr", map->nr == 1); TEST_ASSERT_VAL("wrong pid", - thread_map__pid(map, 0) == getpid()); + perf_thread_map__pid(map, 0) == getpid()); TEST_ASSERT_VAL("wrong comm", perf_thread_map__comm(map, 0) && !strcmp(perf_thread_map__comm(map, 0), NAME)); @@ -41,7 +41,7 @@ int test__thread_map(struct test *test __maybe_unused, int subtest __maybe_unuse thread_map__read_comms(map); TEST_ASSERT_VAL("wrong nr", map->nr == 1); - TEST_ASSERT_VAL("wrong pid", thread_map__pid(map, 0) == -1); + TEST_ASSERT_VAL("wrong pid", perf_thread_map__pid(map, 0) == -1); TEST_ASSERT_VAL("wrong comm", perf_thread_map__comm(map, 0) && !strcmp(perf_thread_map__comm(map, 0), "dummy")); @@ -68,7 +68,7 @@ static int process_event(struct perf_tool *tool __maybe_unused, TEST_ASSERT_VAL("wrong nr", threads->nr == 1); TEST_ASSERT_VAL("wrong pid", - thread_map__pid(threads, 0) == getpid()); + perf_thread_map__pid(threads, 0) == getpid()); TEST_ASSERT_VAL("wrong comm", perf_thread_map__comm(threads, 0) && !strcmp(perf_thread_map__comm(threads, 0), NAME)); diff --git a/tools/perf/util/auxtrace.c b/tools/perf/util/auxtrace.c index 60428576426e..094e6ceb3cf2 100644 --- a/tools/perf/util/auxtrace.c +++ b/tools/perf/util/auxtrace.c @@ -132,12 +132,12 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp, if (per_cpu) { mp->cpu = evlist->core.cpus->map[idx]; if (evlist->core.threads) - mp->tid = thread_map__pid(evlist->core.threads, 0); + mp->tid = perf_thread_map__pid(evlist->core.threads, 0); else mp->tid = -1; } else { mp->cpu = -1; - mp->tid = thread_map__pid(evlist->core.threads, idx); + mp->tid = perf_thread_map__pid(evlist->core.threads, idx); } } diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index f433da85c45e..332edef8d394 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -647,7 +647,7 @@ int perf_event__synthesize_thread_map(struct perf_tool *tool, for (thread = 0; thread < threads->nr; ++thread) { if (__event__synthesize_thread(comm_event, mmap_event, fork_event, namespaces_event, - thread_map__pid(threads, thread), 0, + perf_thread_map__pid(threads, thread), 0, process, tool, machine, mmap_data)) { err = -1; @@ -658,12 +658,12 @@ int perf_event__synthesize_thread_map(struct perf_tool *tool, * comm.pid is set to thread group id by * perf_event__synthesize_comm */ - if ((int) comm_event->comm.pid != thread_map__pid(threads, thread)) { + if ((int) comm_event->comm.pid != perf_thread_map__pid(threads, thread)) { bool need_leader = true; /* is thread group leader in thread_map? */ for (j = 0; j < threads->nr; ++j) { - if ((int) comm_event->comm.pid == thread_map__pid(threads, j)) { + if ((int) comm_event->comm.pid == perf_thread_map__pid(threads, j)) { need_leader = false; break; } @@ -997,7 +997,7 @@ int perf_event__synthesize_thread_map2(struct perf_tool *tool, if (!comm) comm = (char *) ""; - entry->pid = thread_map__pid(threads, i); + entry->pid = perf_thread_map__pid(threads, i); strncpy((char *) &entry->comm, comm, sizeof(entry->comm)); } diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index 8582560b59af..23b56717d260 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -314,7 +314,7 @@ static int perf_evlist__nr_threads(struct evlist *evlist, if (evsel->system_wide) return 1; else - return thread_map__nr(evlist->core.threads); + return perf_thread_map__nr(evlist->core.threads); } void evlist__disable(struct evlist *evlist) @@ -397,7 +397,7 @@ int perf_evlist__enable_event_idx(struct evlist *evlist, int perf_evlist__alloc_pollfd(struct evlist *evlist) { int nr_cpus = perf_cpu_map__nr(evlist->core.cpus); - int nr_threads = thread_map__nr(evlist->core.threads); + int nr_threads = perf_thread_map__nr(evlist->core.threads); int nfds = 0; struct evsel *evsel; @@ -529,7 +529,7 @@ static void perf_evlist__set_sid_idx(struct evlist *evlist, else sid->cpu = -1; if (!evsel->system_wide && evlist->core.threads && thread >= 0) - sid->tid = thread_map__pid(evlist->core.threads, thread); + sid->tid = perf_thread_map__pid(evlist->core.threads, thread); else sid->tid = -1; } @@ -694,7 +694,7 @@ static struct perf_mmap *perf_evlist__alloc_mmap(struct evlist *evlist, evlist->nr_mmaps = perf_cpu_map__nr(evlist->core.cpus); if (perf_cpu_map__empty(evlist->core.cpus)) - evlist->nr_mmaps = thread_map__nr(evlist->core.threads); + evlist->nr_mmaps = perf_thread_map__nr(evlist->core.threads); map = zalloc(evlist->nr_mmaps * sizeof(struct perf_mmap)); if (!map) return NULL; @@ -808,7 +808,7 @@ static int perf_evlist__mmap_per_cpu(struct evlist *evlist, { int cpu, thread; int nr_cpus = perf_cpu_map__nr(evlist->core.cpus); - int nr_threads = thread_map__nr(evlist->core.threads); + int nr_threads = perf_thread_map__nr(evlist->core.threads); pr_debug2("perf event ring buffer mmapped per cpu\n"); for (cpu = 0; cpu < nr_cpus; cpu++) { @@ -836,7 +836,7 @@ static int perf_evlist__mmap_per_thread(struct evlist *evlist, struct mmap_params *mp) { int thread; - int nr_threads = thread_map__nr(evlist->core.threads); + int nr_threads = perf_thread_map__nr(evlist->core.threads); pr_debug2("perf event ring buffer mmapped per thread\n"); for (thread = 0; thread < nr_threads; thread++) { diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c index 0a33f7322ecc..45328a788e9f 100644 --- a/tools/perf/util/evsel.c +++ b/tools/perf/util/evsel.c @@ -1651,7 +1651,7 @@ static bool ignore_missing_thread(struct evsel *evsel, struct perf_thread_map *threads, int thread, int err) { - pid_t ignore_pid = thread_map__pid(threads, thread); + pid_t ignore_pid = perf_thread_map__pid(threads, thread); if (!evsel->ignore_missing_thread) return false; @@ -1814,7 +1814,7 @@ int evsel__open(struct evsel *evsel, struct perf_cpu_map *cpus, int fd, group_fd; if (!evsel->cgrp && !evsel->system_wide) - pid = thread_map__pid(threads, thread); + pid = perf_thread_map__pid(threads, thread); group_fd = get_group_fd(evsel, cpu, thread); retry_open: diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c index 32c17a727450..6801afaa84c4 100644 --- a/tools/perf/util/scripting-engines/trace-event-python.c +++ b/tools/perf/util/scripting-engines/trace-event-python.c @@ -1405,7 +1405,7 @@ static void python_process_stat(struct perf_stat_config *config, for (thread = 0; thread < threads->nr; thread++) { for (cpu = 0; cpu < cpus->nr; cpu++) { process_stat(counter, cpus->map[cpu], - thread_map__pid(threads, thread), tstamp, + perf_thread_map__pid(threads, thread), tstamp, perf_counts(counter->counts, cpu, thread)); } } diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c index 3df0e39ccd52..74e0f5ad1456 100644 --- a/tools/perf/util/stat-display.c +++ b/tools/perf/util/stat-display.c @@ -118,7 +118,7 @@ static void aggr_printout(struct perf_stat_config *config, config->csv_output ? 0 : 16, perf_thread_map__comm(evsel->core.threads, id), config->csv_output ? 0 : -8, - thread_map__pid(evsel->core.threads, id), + perf_thread_map__pid(evsel->core.threads, id), config->csv_sep); break; case AGGR_GLOBAL: @@ -744,7 +744,7 @@ static void print_aggr_thread(struct perf_stat_config *config, struct evsel *counter, char *prefix) { FILE *output = config->output; - int nthreads = thread_map__nr(counter->core.threads); + int nthreads = perf_thread_map__nr(counter->core.threads); int ncpus = perf_cpu_map__nr(counter->core.cpus); int thread, sorted_threads, id; struct perf_aggr_thread_value *buf; diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index 2715112290cf..7342389bc8e1 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -158,7 +158,7 @@ static void perf_evsel__free_prev_raw_counts(struct evsel *evsel) static int perf_evsel__alloc_stats(struct evsel *evsel, bool alloc_raw) { int ncpus = perf_evsel__nr_cpus(evsel); - int nthreads = thread_map__nr(evsel->core.threads); + int nthreads = perf_thread_map__nr(evsel->core.threads); if (perf_evsel__alloc_stat_priv(evsel) < 0 || perf_evsel__alloc_counts(evsel, ncpus, nthreads) < 0 || @@ -308,7 +308,7 @@ process_counter_values(struct perf_stat_config *config, struct evsel *evsel, static int process_counter_maps(struct perf_stat_config *config, struct evsel *counter) { - int nthreads = thread_map__nr(counter->core.threads); + int nthreads = perf_thread_map__nr(counter->core.threads); int ncpus = perf_evsel__nr_cpus(counter); int cpu, thread; diff --git a/tools/perf/util/thread_map.c b/tools/perf/util/thread_map.c index c58385ea05be..3e64525bf604 100644 --- a/tools/perf/util/thread_map.c +++ b/tools/perf/util/thread_map.c @@ -310,7 +310,7 @@ size_t thread_map__fprintf(struct perf_thread_map *threads, FILE *fp) size_t printed = fprintf(fp, "%d thread%s: ", threads->nr, threads->nr > 1 ? "s" : ""); for (i = 0; i < threads->nr; ++i) - printed += fprintf(fp, "%s%d", i ? ", " : "", thread_map__pid(threads, i)); + printed += fprintf(fp, "%s%d", i ? ", " : "", perf_thread_map__pid(threads, i)); return printed + fprintf(fp, "\n"); } @@ -341,7 +341,7 @@ static int get_comm(char **comm, pid_t pid) static void comm_init(struct perf_thread_map *map, int i) { - pid_t pid = thread_map__pid(map, i); + pid_t pid = perf_thread_map__pid(map, i); char *comm = NULL; /* dummy pid comm initialization */ diff --git a/tools/perf/util/thread_map.h b/tools/perf/util/thread_map.h index ba45c760be72..ca165fdf6cb0 100644 --- a/tools/perf/util/thread_map.h +++ b/tools/perf/util/thread_map.h @@ -25,16 +25,6 @@ struct perf_thread_map *thread_map__new_by_tid_str(const char *tid_str); size_t thread_map__fprintf(struct perf_thread_map *threads, FILE *fp); -static inline int thread_map__nr(struct perf_thread_map *threads) -{ - return threads ? threads->nr : 1; -} - -static inline pid_t thread_map__pid(struct perf_thread_map *map, int thread) -{ - return map->map[thread].pid; -} - void thread_map__read_comms(struct perf_thread_map *threads); bool thread_map__has(struct perf_thread_map *threads, pid_t pid); int thread_map__remove(struct perf_thread_map *threads, int idx); -- 2.21.0 ^ permalink raw reply related [flat|nested] 11+ messages in thread
* Re: [PATCH 5/5] libperf: Add perf_thread_map__nr/perf_thread_map__pid functions 2019-08-22 11:11 ` [PATCH 5/5] libperf: Add perf_thread_map__nr/perf_thread_map__pid functions Jiri Olsa @ 2019-08-22 15:18 ` Arnaldo Carvalho de Melo 2019-08-23 12:29 ` [tip: perf/core] " tip-bot2 for Jiri Olsa 1 sibling, 0 replies; 11+ messages in thread From: Arnaldo Carvalho de Melo @ 2019-08-22 15:18 UTC (permalink / raw) To: Jiri Olsa Cc: lkml, Ingo Molnar, Namhyung Kim, Alexander Shishkin, Peter Zijlstra, Michael Petlan Em Thu, Aug 22, 2019 at 01:11:41PM +0200, Jiri Olsa escreveu: > So it's part of libperf library as basic functions > over struct perf_thread_map object. Thanks, applied and the other ones as well. I thought this one wasn't applying but it was just a thinko on my part, - Arnaldo > Link: http://lkml.kernel.org/n/tip-mjksprk4yct9ouez3j9mq2az@git.kernel.org > Signed-off-by: Jiri Olsa <jolsa@kernel.org> > --- > tools/perf/builtin-ftrace.c | 2 +- > tools/perf/builtin-script.c | 4 ++-- > tools/perf/builtin-stat.c | 4 ++-- > tools/perf/builtin-trace.c | 4 ++-- > tools/perf/lib/include/perf/threadmap.h | 2 ++ > tools/perf/lib/libperf.map | 2 ++ > tools/perf/lib/threadmap.c | 10 ++++++++++ > tools/perf/tests/thread-map.c | 6 +++--- > tools/perf/util/auxtrace.c | 4 ++-- > tools/perf/util/event.c | 8 ++++---- > tools/perf/util/evlist.c | 12 ++++++------ > tools/perf/util/evsel.c | 4 ++-- > .../perf/util/scripting-engines/trace-event-python.c | 2 +- > tools/perf/util/stat-display.c | 4 ++-- > tools/perf/util/stat.c | 4 ++-- > tools/perf/util/thread_map.c | 4 ++-- > tools/perf/util/thread_map.h | 10 ---------- > 17 files changed, 45 insertions(+), 41 deletions(-) > > diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c > index 1367bb5046a7..565db782c1b9 100644 > --- a/tools/perf/builtin-ftrace.c > +++ b/tools/perf/builtin-ftrace.c > @@ -158,7 +158,7 @@ static int set_tracing_pid(struct perf_ftrace *ftrace) > if (target__has_cpu(&ftrace->target)) > return 0; > > - for (i = 0; i < thread_map__nr(ftrace->evlist->core.threads); i++) { > + for (i = 0; i < perf_thread_map__nr(ftrace->evlist->core.threads); i++) { > scnprintf(buf, sizeof(buf), "%d", > ftrace->evlist->core.threads->map[i]); > if (append_tracing_file("set_ftrace_pid", buf) < 0) > diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c > index 1764efd16cd4..5d45be1d3885 100644 > --- a/tools/perf/builtin-script.c > +++ b/tools/perf/builtin-script.c > @@ -1905,7 +1905,7 @@ static struct scripting_ops *scripting_ops; > > static void __process_stat(struct evsel *counter, u64 tstamp) > { > - int nthreads = thread_map__nr(counter->core.threads); > + int nthreads = perf_thread_map__nr(counter->core.threads); > int ncpus = perf_evsel__nr_cpus(counter); > int cpu, thread; > static int header_printed; > @@ -1927,7 +1927,7 @@ static void __process_stat(struct evsel *counter, u64 tstamp) > > printf("%3d %8d %15" PRIu64 " %15" PRIu64 " %15" PRIu64 " %15" PRIu64 " %s\n", > counter->core.cpus->map[cpu], > - thread_map__pid(counter->core.threads, thread), > + perf_thread_map__pid(counter->core.threads, thread), > counts->val, > counts->ena, > counts->run, > diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c > index 90636a811b36..8a4f1a7d0cba 100644 > --- a/tools/perf/builtin-stat.c > +++ b/tools/perf/builtin-stat.c > @@ -264,7 +264,7 @@ static int read_single_counter(struct evsel *counter, int cpu, > */ > static int read_counter(struct evsel *counter, struct timespec *rs) > { > - int nthreads = thread_map__nr(evsel_list->core.threads); > + int nthreads = perf_thread_map__nr(evsel_list->core.threads); > int ncpus, cpu, thread; > > if (target__has_cpu(&target) && !target__has_per_thread(&target)) > @@ -1893,7 +1893,7 @@ int cmd_stat(int argc, const char **argv) > thread_map__read_comms(evsel_list->core.threads); > if (target.system_wide) { > if (runtime_stat_new(&stat_config, > - thread_map__nr(evsel_list->core.threads))) { > + perf_thread_map__nr(evsel_list->core.threads))) { > goto out; > } > } > diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c > index bc44ed29e05a..de126258ca10 100644 > --- a/tools/perf/builtin-trace.c > +++ b/tools/perf/builtin-trace.c > @@ -3188,7 +3188,7 @@ static int trace__set_filter_pids(struct trace *trace) > err = bpf_map__set_filter_pids(trace->filter_pids.map, trace->filter_pids.nr, > trace->filter_pids.entries); > } > - } else if (thread_map__pid(trace->evlist->core.threads, 0) == -1) { > + } else if (perf_thread_map__pid(trace->evlist->core.threads, 0) == -1) { > err = trace__set_filter_loop_pids(trace); > } > > @@ -3417,7 +3417,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) > evlist__enable(evlist); > } > > - trace->multiple_threads = thread_map__pid(evlist->core.threads, 0) == -1 || > + trace->multiple_threads = perf_thread_map__pid(evlist->core.threads, 0) == -1 || > evlist->core.threads->nr > 1 || > perf_evlist__first(evlist)->core.attr.inherit; > > diff --git a/tools/perf/lib/include/perf/threadmap.h b/tools/perf/lib/include/perf/threadmap.h > index 456295273daa..a7c50de8d010 100644 > --- a/tools/perf/lib/include/perf/threadmap.h > +++ b/tools/perf/lib/include/perf/threadmap.h > @@ -11,6 +11,8 @@ LIBPERF_API struct perf_thread_map *perf_thread_map__new_dummy(void); > > LIBPERF_API void perf_thread_map__set_pid(struct perf_thread_map *map, int thread, pid_t pid); > LIBPERF_API char *perf_thread_map__comm(struct perf_thread_map *map, int thread); > +LIBPERF_API int perf_thread_map__nr(struct perf_thread_map *threads); > +LIBPERF_API pid_t perf_thread_map__pid(struct perf_thread_map *map, int thread); > > LIBPERF_API struct perf_thread_map *perf_thread_map__get(struct perf_thread_map *map); > LIBPERF_API void perf_thread_map__put(struct perf_thread_map *map); > diff --git a/tools/perf/lib/libperf.map b/tools/perf/lib/libperf.map > index 3373dd51fcda..dc4d66363bc4 100644 > --- a/tools/perf/lib/libperf.map > +++ b/tools/perf/lib/libperf.map > @@ -12,6 +12,8 @@ LIBPERF_0.0.1 { > perf_thread_map__new_dummy; > perf_thread_map__set_pid; > perf_thread_map__comm; > + perf_thread_map__nr; > + perf_thread_map__pid; > perf_thread_map__get; > perf_thread_map__put; > perf_evsel__new; > diff --git a/tools/perf/lib/threadmap.c b/tools/perf/lib/threadmap.c > index 4865b73e2586..e92c368b0a6c 100644 > --- a/tools/perf/lib/threadmap.c > +++ b/tools/perf/lib/threadmap.c > @@ -79,3 +79,13 @@ void perf_thread_map__put(struct perf_thread_map *map) > if (map && refcount_dec_and_test(&map->refcnt)) > perf_thread_map__delete(map); > } > + > +int perf_thread_map__nr(struct perf_thread_map *threads) > +{ > + return threads ? threads->nr : 1; > +} > + > +pid_t perf_thread_map__pid(struct perf_thread_map *map, int thread) > +{ > + return map->map[thread].pid; > +} > diff --git a/tools/perf/tests/thread-map.c b/tools/perf/tests/thread-map.c > index d61773cacf0b..d803eafedc60 100644 > --- a/tools/perf/tests/thread-map.c > +++ b/tools/perf/tests/thread-map.c > @@ -26,7 +26,7 @@ int test__thread_map(struct test *test __maybe_unused, int subtest __maybe_unuse > > TEST_ASSERT_VAL("wrong nr", map->nr == 1); > TEST_ASSERT_VAL("wrong pid", > - thread_map__pid(map, 0) == getpid()); > + perf_thread_map__pid(map, 0) == getpid()); > TEST_ASSERT_VAL("wrong comm", > perf_thread_map__comm(map, 0) && > !strcmp(perf_thread_map__comm(map, 0), NAME)); > @@ -41,7 +41,7 @@ int test__thread_map(struct test *test __maybe_unused, int subtest __maybe_unuse > thread_map__read_comms(map); > > TEST_ASSERT_VAL("wrong nr", map->nr == 1); > - TEST_ASSERT_VAL("wrong pid", thread_map__pid(map, 0) == -1); > + TEST_ASSERT_VAL("wrong pid", perf_thread_map__pid(map, 0) == -1); > TEST_ASSERT_VAL("wrong comm", > perf_thread_map__comm(map, 0) && > !strcmp(perf_thread_map__comm(map, 0), "dummy")); > @@ -68,7 +68,7 @@ static int process_event(struct perf_tool *tool __maybe_unused, > > TEST_ASSERT_VAL("wrong nr", threads->nr == 1); > TEST_ASSERT_VAL("wrong pid", > - thread_map__pid(threads, 0) == getpid()); > + perf_thread_map__pid(threads, 0) == getpid()); > TEST_ASSERT_VAL("wrong comm", > perf_thread_map__comm(threads, 0) && > !strcmp(perf_thread_map__comm(threads, 0), NAME)); > diff --git a/tools/perf/util/auxtrace.c b/tools/perf/util/auxtrace.c > index 60428576426e..094e6ceb3cf2 100644 > --- a/tools/perf/util/auxtrace.c > +++ b/tools/perf/util/auxtrace.c > @@ -132,12 +132,12 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp, > if (per_cpu) { > mp->cpu = evlist->core.cpus->map[idx]; > if (evlist->core.threads) > - mp->tid = thread_map__pid(evlist->core.threads, 0); > + mp->tid = perf_thread_map__pid(evlist->core.threads, 0); > else > mp->tid = -1; > } else { > mp->cpu = -1; > - mp->tid = thread_map__pid(evlist->core.threads, idx); > + mp->tid = perf_thread_map__pid(evlist->core.threads, idx); > } > } > > diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c > index f433da85c45e..332edef8d394 100644 > --- a/tools/perf/util/event.c > +++ b/tools/perf/util/event.c > @@ -647,7 +647,7 @@ int perf_event__synthesize_thread_map(struct perf_tool *tool, > for (thread = 0; thread < threads->nr; ++thread) { > if (__event__synthesize_thread(comm_event, mmap_event, > fork_event, namespaces_event, > - thread_map__pid(threads, thread), 0, > + perf_thread_map__pid(threads, thread), 0, > process, tool, machine, > mmap_data)) { > err = -1; > @@ -658,12 +658,12 @@ int perf_event__synthesize_thread_map(struct perf_tool *tool, > * comm.pid is set to thread group id by > * perf_event__synthesize_comm > */ > - if ((int) comm_event->comm.pid != thread_map__pid(threads, thread)) { > + if ((int) comm_event->comm.pid != perf_thread_map__pid(threads, thread)) { > bool need_leader = true; > > /* is thread group leader in thread_map? */ > for (j = 0; j < threads->nr; ++j) { > - if ((int) comm_event->comm.pid == thread_map__pid(threads, j)) { > + if ((int) comm_event->comm.pid == perf_thread_map__pid(threads, j)) { > need_leader = false; > break; > } > @@ -997,7 +997,7 @@ int perf_event__synthesize_thread_map2(struct perf_tool *tool, > if (!comm) > comm = (char *) ""; > > - entry->pid = thread_map__pid(threads, i); > + entry->pid = perf_thread_map__pid(threads, i); > strncpy((char *) &entry->comm, comm, sizeof(entry->comm)); > } > > diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c > index 8582560b59af..23b56717d260 100644 > --- a/tools/perf/util/evlist.c > +++ b/tools/perf/util/evlist.c > @@ -314,7 +314,7 @@ static int perf_evlist__nr_threads(struct evlist *evlist, > if (evsel->system_wide) > return 1; > else > - return thread_map__nr(evlist->core.threads); > + return perf_thread_map__nr(evlist->core.threads); > } > > void evlist__disable(struct evlist *evlist) > @@ -397,7 +397,7 @@ int perf_evlist__enable_event_idx(struct evlist *evlist, > int perf_evlist__alloc_pollfd(struct evlist *evlist) > { > int nr_cpus = perf_cpu_map__nr(evlist->core.cpus); > - int nr_threads = thread_map__nr(evlist->core.threads); > + int nr_threads = perf_thread_map__nr(evlist->core.threads); > int nfds = 0; > struct evsel *evsel; > > @@ -529,7 +529,7 @@ static void perf_evlist__set_sid_idx(struct evlist *evlist, > else > sid->cpu = -1; > if (!evsel->system_wide && evlist->core.threads && thread >= 0) > - sid->tid = thread_map__pid(evlist->core.threads, thread); > + sid->tid = perf_thread_map__pid(evlist->core.threads, thread); > else > sid->tid = -1; > } > @@ -694,7 +694,7 @@ static struct perf_mmap *perf_evlist__alloc_mmap(struct evlist *evlist, > > evlist->nr_mmaps = perf_cpu_map__nr(evlist->core.cpus); > if (perf_cpu_map__empty(evlist->core.cpus)) > - evlist->nr_mmaps = thread_map__nr(evlist->core.threads); > + evlist->nr_mmaps = perf_thread_map__nr(evlist->core.threads); > map = zalloc(evlist->nr_mmaps * sizeof(struct perf_mmap)); > if (!map) > return NULL; > @@ -808,7 +808,7 @@ static int perf_evlist__mmap_per_cpu(struct evlist *evlist, > { > int cpu, thread; > int nr_cpus = perf_cpu_map__nr(evlist->core.cpus); > - int nr_threads = thread_map__nr(evlist->core.threads); > + int nr_threads = perf_thread_map__nr(evlist->core.threads); > > pr_debug2("perf event ring buffer mmapped per cpu\n"); > for (cpu = 0; cpu < nr_cpus; cpu++) { > @@ -836,7 +836,7 @@ static int perf_evlist__mmap_per_thread(struct evlist *evlist, > struct mmap_params *mp) > { > int thread; > - int nr_threads = thread_map__nr(evlist->core.threads); > + int nr_threads = perf_thread_map__nr(evlist->core.threads); > > pr_debug2("perf event ring buffer mmapped per thread\n"); > for (thread = 0; thread < nr_threads; thread++) { > diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c > index 0a33f7322ecc..45328a788e9f 100644 > --- a/tools/perf/util/evsel.c > +++ b/tools/perf/util/evsel.c > @@ -1651,7 +1651,7 @@ static bool ignore_missing_thread(struct evsel *evsel, > struct perf_thread_map *threads, > int thread, int err) > { > - pid_t ignore_pid = thread_map__pid(threads, thread); > + pid_t ignore_pid = perf_thread_map__pid(threads, thread); > > if (!evsel->ignore_missing_thread) > return false; > @@ -1814,7 +1814,7 @@ int evsel__open(struct evsel *evsel, struct perf_cpu_map *cpus, > int fd, group_fd; > > if (!evsel->cgrp && !evsel->system_wide) > - pid = thread_map__pid(threads, thread); > + pid = perf_thread_map__pid(threads, thread); > > group_fd = get_group_fd(evsel, cpu, thread); > retry_open: > diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c > index 32c17a727450..6801afaa84c4 100644 > --- a/tools/perf/util/scripting-engines/trace-event-python.c > +++ b/tools/perf/util/scripting-engines/trace-event-python.c > @@ -1405,7 +1405,7 @@ static void python_process_stat(struct perf_stat_config *config, > for (thread = 0; thread < threads->nr; thread++) { > for (cpu = 0; cpu < cpus->nr; cpu++) { > process_stat(counter, cpus->map[cpu], > - thread_map__pid(threads, thread), tstamp, > + perf_thread_map__pid(threads, thread), tstamp, > perf_counts(counter->counts, cpu, thread)); > } > } > diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c > index 3df0e39ccd52..74e0f5ad1456 100644 > --- a/tools/perf/util/stat-display.c > +++ b/tools/perf/util/stat-display.c > @@ -118,7 +118,7 @@ static void aggr_printout(struct perf_stat_config *config, > config->csv_output ? 0 : 16, > perf_thread_map__comm(evsel->core.threads, id), > config->csv_output ? 0 : -8, > - thread_map__pid(evsel->core.threads, id), > + perf_thread_map__pid(evsel->core.threads, id), > config->csv_sep); > break; > case AGGR_GLOBAL: > @@ -744,7 +744,7 @@ static void print_aggr_thread(struct perf_stat_config *config, > struct evsel *counter, char *prefix) > { > FILE *output = config->output; > - int nthreads = thread_map__nr(counter->core.threads); > + int nthreads = perf_thread_map__nr(counter->core.threads); > int ncpus = perf_cpu_map__nr(counter->core.cpus); > int thread, sorted_threads, id; > struct perf_aggr_thread_value *buf; > diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c > index 2715112290cf..7342389bc8e1 100644 > --- a/tools/perf/util/stat.c > +++ b/tools/perf/util/stat.c > @@ -158,7 +158,7 @@ static void perf_evsel__free_prev_raw_counts(struct evsel *evsel) > static int perf_evsel__alloc_stats(struct evsel *evsel, bool alloc_raw) > { > int ncpus = perf_evsel__nr_cpus(evsel); > - int nthreads = thread_map__nr(evsel->core.threads); > + int nthreads = perf_thread_map__nr(evsel->core.threads); > > if (perf_evsel__alloc_stat_priv(evsel) < 0 || > perf_evsel__alloc_counts(evsel, ncpus, nthreads) < 0 || > @@ -308,7 +308,7 @@ process_counter_values(struct perf_stat_config *config, struct evsel *evsel, > static int process_counter_maps(struct perf_stat_config *config, > struct evsel *counter) > { > - int nthreads = thread_map__nr(counter->core.threads); > + int nthreads = perf_thread_map__nr(counter->core.threads); > int ncpus = perf_evsel__nr_cpus(counter); > int cpu, thread; > > diff --git a/tools/perf/util/thread_map.c b/tools/perf/util/thread_map.c > index c58385ea05be..3e64525bf604 100644 > --- a/tools/perf/util/thread_map.c > +++ b/tools/perf/util/thread_map.c > @@ -310,7 +310,7 @@ size_t thread_map__fprintf(struct perf_thread_map *threads, FILE *fp) > size_t printed = fprintf(fp, "%d thread%s: ", > threads->nr, threads->nr > 1 ? "s" : ""); > for (i = 0; i < threads->nr; ++i) > - printed += fprintf(fp, "%s%d", i ? ", " : "", thread_map__pid(threads, i)); > + printed += fprintf(fp, "%s%d", i ? ", " : "", perf_thread_map__pid(threads, i)); > > return printed + fprintf(fp, "\n"); > } > @@ -341,7 +341,7 @@ static int get_comm(char **comm, pid_t pid) > > static void comm_init(struct perf_thread_map *map, int i) > { > - pid_t pid = thread_map__pid(map, i); > + pid_t pid = perf_thread_map__pid(map, i); > char *comm = NULL; > > /* dummy pid comm initialization */ > diff --git a/tools/perf/util/thread_map.h b/tools/perf/util/thread_map.h > index ba45c760be72..ca165fdf6cb0 100644 > --- a/tools/perf/util/thread_map.h > +++ b/tools/perf/util/thread_map.h > @@ -25,16 +25,6 @@ struct perf_thread_map *thread_map__new_by_tid_str(const char *tid_str); > > size_t thread_map__fprintf(struct perf_thread_map *threads, FILE *fp); > > -static inline int thread_map__nr(struct perf_thread_map *threads) > -{ > - return threads ? threads->nr : 1; > -} > - > -static inline pid_t thread_map__pid(struct perf_thread_map *map, int thread) > -{ > - return map->map[thread].pid; > -} > - > void thread_map__read_comms(struct perf_thread_map *threads); > bool thread_map__has(struct perf_thread_map *threads, pid_t pid); > int thread_map__remove(struct perf_thread_map *threads, int idx); > -- > 2.21.0 -- - Arnaldo ^ permalink raw reply [flat|nested] 11+ messages in thread
* [tip: perf/core] libperf: Add perf_thread_map__nr/perf_thread_map__pid functions 2019-08-22 11:11 ` [PATCH 5/5] libperf: Add perf_thread_map__nr/perf_thread_map__pid functions Jiri Olsa 2019-08-22 15:18 ` Arnaldo Carvalho de Melo @ 2019-08-23 12:29 ` tip-bot2 for Jiri Olsa 1 sibling, 0 replies; 11+ messages in thread From: tip-bot2 for Jiri Olsa @ 2019-08-23 12:29 UTC (permalink / raw) To: linux-tip-commits Cc: Jiri Olsa, Alexander Shishkin, Michael Petlan, Namhyung Kim, Peter Zijlstra, Arnaldo Carvalho de Melo, Ingo Molnar, linux-kernel The following commit has been merged into the perf/core branch of tip: Commit-ID: a2f354e3abb853f9a40048829e1f839e8f7fada5 Gitweb: https://git.kernel.org/tip/a2f354e3abb853f9a40048829e1f839e8f7fada5 Author: Jiri Olsa <jolsa@kernel.org> AuthorDate: Thu, 22 Aug 2019 13:11:41 +02:00 Committer: Arnaldo Carvalho de Melo <acme@redhat.com> CommitterDate: Thu, 22 Aug 2019 17:16:57 -03:00 libperf: Add perf_thread_map__nr/perf_thread_map__pid functions So it's part of libperf library as basic functions operating on perf_thread_map objects. Signed-off-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Michael Petlan <mpetlan@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/20190822111141.25823-6-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> --- tools/perf/builtin-ftrace.c | 2 +- tools/perf/builtin-script.c | 4 +-- tools/perf/builtin-stat.c | 4 +-- tools/perf/builtin-trace.c | 4 +-- tools/perf/lib/include/perf/threadmap.h | 2 ++- tools/perf/lib/libperf.map | 2 ++- tools/perf/lib/threadmap.c | 10 ++++++++- tools/perf/tests/thread-map.c | 6 ++--- tools/perf/util/auxtrace.c | 4 +-- tools/perf/util/event.c | 8 +++--- tools/perf/util/evlist.c | 12 ++++----- tools/perf/util/evsel.c | 4 +-- tools/perf/util/scripting-engines/trace-event-python.c | 2 +- tools/perf/util/stat-display.c | 4 +-- tools/perf/util/stat.c | 4 +-- tools/perf/util/thread_map.c | 4 +-- tools/perf/util/thread_map.h | 10 +-------- 17 files changed, 45 insertions(+), 41 deletions(-) diff --git a/tools/perf/builtin-ftrace.c b/tools/perf/builtin-ftrace.c index 1367bb5..565db78 100644 --- a/tools/perf/builtin-ftrace.c +++ b/tools/perf/builtin-ftrace.c @@ -158,7 +158,7 @@ static int set_tracing_pid(struct perf_ftrace *ftrace) if (target__has_cpu(&ftrace->target)) return 0; - for (i = 0; i < thread_map__nr(ftrace->evlist->core.threads); i++) { + for (i = 0; i < perf_thread_map__nr(ftrace->evlist->core.threads); i++) { scnprintf(buf, sizeof(buf), "%d", ftrace->evlist->core.threads->map[i]); if (append_tracing_file("set_ftrace_pid", buf) < 0) diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c index e957b87..9b93dde 100644 --- a/tools/perf/builtin-script.c +++ b/tools/perf/builtin-script.c @@ -1906,7 +1906,7 @@ static struct scripting_ops *scripting_ops; static void __process_stat(struct evsel *counter, u64 tstamp) { - int nthreads = thread_map__nr(counter->core.threads); + int nthreads = perf_thread_map__nr(counter->core.threads); int ncpus = perf_evsel__nr_cpus(counter); int cpu, thread; static int header_printed; @@ -1928,7 +1928,7 @@ static void __process_stat(struct evsel *counter, u64 tstamp) printf("%3d %8d %15" PRIu64 " %15" PRIu64 " %15" PRIu64 " %15" PRIu64 " %s\n", counter->core.cpus->map[cpu], - thread_map__pid(counter->core.threads, thread), + perf_thread_map__pid(counter->core.threads, thread), counts->val, counts->ena, counts->run, diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 90636a8..8a4f1a7 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -264,7 +264,7 @@ static int read_single_counter(struct evsel *counter, int cpu, */ static int read_counter(struct evsel *counter, struct timespec *rs) { - int nthreads = thread_map__nr(evsel_list->core.threads); + int nthreads = perf_thread_map__nr(evsel_list->core.threads); int ncpus, cpu, thread; if (target__has_cpu(&target) && !target__has_per_thread(&target)) @@ -1893,7 +1893,7 @@ int cmd_stat(int argc, const char **argv) thread_map__read_comms(evsel_list->core.threads); if (target.system_wide) { if (runtime_stat_new(&stat_config, - thread_map__nr(evsel_list->core.threads))) { + perf_thread_map__nr(evsel_list->core.threads))) { goto out; } } diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c index bc44ed2..de12625 100644 --- a/tools/perf/builtin-trace.c +++ b/tools/perf/builtin-trace.c @@ -3188,7 +3188,7 @@ static int trace__set_filter_pids(struct trace *trace) err = bpf_map__set_filter_pids(trace->filter_pids.map, trace->filter_pids.nr, trace->filter_pids.entries); } - } else if (thread_map__pid(trace->evlist->core.threads, 0) == -1) { + } else if (perf_thread_map__pid(trace->evlist->core.threads, 0) == -1) { err = trace__set_filter_loop_pids(trace); } @@ -3417,7 +3417,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) evlist__enable(evlist); } - trace->multiple_threads = thread_map__pid(evlist->core.threads, 0) == -1 || + trace->multiple_threads = perf_thread_map__pid(evlist->core.threads, 0) == -1 || evlist->core.threads->nr > 1 || perf_evlist__first(evlist)->core.attr.inherit; diff --git a/tools/perf/lib/include/perf/threadmap.h b/tools/perf/lib/include/perf/threadmap.h index 4562952..a7c50de 100644 --- a/tools/perf/lib/include/perf/threadmap.h +++ b/tools/perf/lib/include/perf/threadmap.h @@ -11,6 +11,8 @@ LIBPERF_API struct perf_thread_map *perf_thread_map__new_dummy(void); LIBPERF_API void perf_thread_map__set_pid(struct perf_thread_map *map, int thread, pid_t pid); LIBPERF_API char *perf_thread_map__comm(struct perf_thread_map *map, int thread); +LIBPERF_API int perf_thread_map__nr(struct perf_thread_map *threads); +LIBPERF_API pid_t perf_thread_map__pid(struct perf_thread_map *map, int thread); LIBPERF_API struct perf_thread_map *perf_thread_map__get(struct perf_thread_map *map); LIBPERF_API void perf_thread_map__put(struct perf_thread_map *map); diff --git a/tools/perf/lib/libperf.map b/tools/perf/lib/libperf.map index 3373dd5..dc4d663 100644 --- a/tools/perf/lib/libperf.map +++ b/tools/perf/lib/libperf.map @@ -12,6 +12,8 @@ LIBPERF_0.0.1 { perf_thread_map__new_dummy; perf_thread_map__set_pid; perf_thread_map__comm; + perf_thread_map__nr; + perf_thread_map__pid; perf_thread_map__get; perf_thread_map__put; perf_evsel__new; diff --git a/tools/perf/lib/threadmap.c b/tools/perf/lib/threadmap.c index 4865b73..e92c368 100644 --- a/tools/perf/lib/threadmap.c +++ b/tools/perf/lib/threadmap.c @@ -79,3 +79,13 @@ void perf_thread_map__put(struct perf_thread_map *map) if (map && refcount_dec_and_test(&map->refcnt)) perf_thread_map__delete(map); } + +int perf_thread_map__nr(struct perf_thread_map *threads) +{ + return threads ? threads->nr : 1; +} + +pid_t perf_thread_map__pid(struct perf_thread_map *map, int thread) +{ + return map->map[thread].pid; +} diff --git a/tools/perf/tests/thread-map.c b/tools/perf/tests/thread-map.c index d61773c..d803eaf 100644 --- a/tools/perf/tests/thread-map.c +++ b/tools/perf/tests/thread-map.c @@ -26,7 +26,7 @@ int test__thread_map(struct test *test __maybe_unused, int subtest __maybe_unuse TEST_ASSERT_VAL("wrong nr", map->nr == 1); TEST_ASSERT_VAL("wrong pid", - thread_map__pid(map, 0) == getpid()); + perf_thread_map__pid(map, 0) == getpid()); TEST_ASSERT_VAL("wrong comm", perf_thread_map__comm(map, 0) && !strcmp(perf_thread_map__comm(map, 0), NAME)); @@ -41,7 +41,7 @@ int test__thread_map(struct test *test __maybe_unused, int subtest __maybe_unuse thread_map__read_comms(map); TEST_ASSERT_VAL("wrong nr", map->nr == 1); - TEST_ASSERT_VAL("wrong pid", thread_map__pid(map, 0) == -1); + TEST_ASSERT_VAL("wrong pid", perf_thread_map__pid(map, 0) == -1); TEST_ASSERT_VAL("wrong comm", perf_thread_map__comm(map, 0) && !strcmp(perf_thread_map__comm(map, 0), "dummy")); @@ -68,7 +68,7 @@ static int process_event(struct perf_tool *tool __maybe_unused, TEST_ASSERT_VAL("wrong nr", threads->nr == 1); TEST_ASSERT_VAL("wrong pid", - thread_map__pid(threads, 0) == getpid()); + perf_thread_map__pid(threads, 0) == getpid()); TEST_ASSERT_VAL("wrong comm", perf_thread_map__comm(threads, 0) && !strcmp(perf_thread_map__comm(threads, 0), NAME)); diff --git a/tools/perf/util/auxtrace.c b/tools/perf/util/auxtrace.c index 6042857..094e6ce 100644 --- a/tools/perf/util/auxtrace.c +++ b/tools/perf/util/auxtrace.c @@ -132,12 +132,12 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp, if (per_cpu) { mp->cpu = evlist->core.cpus->map[idx]; if (evlist->core.threads) - mp->tid = thread_map__pid(evlist->core.threads, 0); + mp->tid = perf_thread_map__pid(evlist->core.threads, 0); else mp->tid = -1; } else { mp->cpu = -1; - mp->tid = thread_map__pid(evlist->core.threads, idx); + mp->tid = perf_thread_map__pid(evlist->core.threads, idx); } } diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c index f433da8..332edef 100644 --- a/tools/perf/util/event.c +++ b/tools/perf/util/event.c @@ -647,7 +647,7 @@ int perf_event__synthesize_thread_map(struct perf_tool *tool, for (thread = 0; thread < threads->nr; ++thread) { if (__event__synthesize_thread(comm_event, mmap_event, fork_event, namespaces_event, - thread_map__pid(threads, thread), 0, + perf_thread_map__pid(threads, thread), 0, process, tool, machine, mmap_data)) { err = -1; @@ -658,12 +658,12 @@ int perf_event__synthesize_thread_map(struct perf_tool *tool, * comm.pid is set to thread group id by * perf_event__synthesize_comm */ - if ((int) comm_event->comm.pid != thread_map__pid(threads, thread)) { + if ((int) comm_event->comm.pid != perf_thread_map__pid(threads, thread)) { bool need_leader = true; /* is thread group leader in thread_map? */ for (j = 0; j < threads->nr; ++j) { - if ((int) comm_event->comm.pid == thread_map__pid(threads, j)) { + if ((int) comm_event->comm.pid == perf_thread_map__pid(threads, j)) { need_leader = false; break; } @@ -997,7 +997,7 @@ int perf_event__synthesize_thread_map2(struct perf_tool *tool, if (!comm) comm = (char *) ""; - entry->pid = thread_map__pid(threads, i); + entry->pid = perf_thread_map__pid(threads, i); strncpy((char *) &entry->comm, comm, sizeof(entry->comm)); } diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c index 68b7c94..ff41568 100644 --- a/tools/perf/util/evlist.c +++ b/tools/perf/util/evlist.c @@ -316,7 +316,7 @@ static int perf_evlist__nr_threads(struct evlist *evlist, if (evsel->system_wide) return 1; else - return thread_map__nr(evlist->core.threads); + return perf_thread_map__nr(evlist->core.threads); } void evlist__disable(struct evlist *evlist) @@ -399,7 +399,7 @@ int perf_evlist__enable_event_idx(struct evlist *evlist, int perf_evlist__alloc_pollfd(struct evlist *evlist) { int nr_cpus = perf_cpu_map__nr(evlist->core.cpus); - int nr_threads = thread_map__nr(evlist->core.threads); + int nr_threads = perf_thread_map__nr(evlist->core.threads); int nfds = 0; struct evsel *evsel; @@ -531,7 +531,7 @@ static void perf_evlist__set_sid_idx(struct evlist *evlist, else sid->cpu = -1; if (!evsel->system_wide && evlist->core.threads && thread >= 0) - sid->tid = thread_map__pid(evlist->core.threads, thread); + sid->tid = perf_thread_map__pid(evlist->core.threads, thread); else sid->tid = -1; } @@ -696,7 +696,7 @@ static struct perf_mmap *perf_evlist__alloc_mmap(struct evlist *evlist, evlist->nr_mmaps = perf_cpu_map__nr(evlist->core.cpus); if (perf_cpu_map__empty(evlist->core.cpus)) - evlist->nr_mmaps = thread_map__nr(evlist->core.threads); + evlist->nr_mmaps = perf_thread_map__nr(evlist->core.threads); map = zalloc(evlist->nr_mmaps * sizeof(struct perf_mmap)); if (!map) return NULL; @@ -810,7 +810,7 @@ static int perf_evlist__mmap_per_cpu(struct evlist *evlist, { int cpu, thread; int nr_cpus = perf_cpu_map__nr(evlist->core.cpus); - int nr_threads = thread_map__nr(evlist->core.threads); + int nr_threads = perf_thread_map__nr(evlist->core.threads); pr_debug2("perf event ring buffer mmapped per cpu\n"); for (cpu = 0; cpu < nr_cpus; cpu++) { @@ -838,7 +838,7 @@ static int perf_evlist__mmap_per_thread(struct evlist *evlist, struct mmap_params *mp) { int thread; - int nr_threads = thread_map__nr(evlist->core.threads); + int nr_threads = perf_thread_map__nr(evlist->core.threads); pr_debug2("perf event ring buffer mmapped per thread\n"); for (thread = 0; thread < nr_threads; thread++) { diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c index 7b43506..e983e72 100644 --- a/tools/perf/util/evsel.c +++ b/tools/perf/util/evsel.c @@ -1653,7 +1653,7 @@ static bool ignore_missing_thread(struct evsel *evsel, struct perf_thread_map *threads, int thread, int err) { - pid_t ignore_pid = thread_map__pid(threads, thread); + pid_t ignore_pid = perf_thread_map__pid(threads, thread); if (!evsel->ignore_missing_thread) return false; @@ -1816,7 +1816,7 @@ int evsel__open(struct evsel *evsel, struct perf_cpu_map *cpus, int fd, group_fd; if (!evsel->cgrp && !evsel->system_wide) - pid = thread_map__pid(threads, thread); + pid = perf_thread_map__pid(threads, thread); group_fd = get_group_fd(evsel, cpu, thread); retry_open: diff --git a/tools/perf/util/scripting-engines/trace-event-python.c b/tools/perf/util/scripting-engines/trace-event-python.c index 51771fc..78c8bc9 100644 --- a/tools/perf/util/scripting-engines/trace-event-python.c +++ b/tools/perf/util/scripting-engines/trace-event-python.c @@ -1406,7 +1406,7 @@ static void python_process_stat(struct perf_stat_config *config, for (thread = 0; thread < threads->nr; thread++) { for (cpu = 0; cpu < cpus->nr; cpu++) { process_stat(counter, cpus->map[cpu], - thread_map__pid(threads, thread), tstamp, + perf_thread_map__pid(threads, thread), tstamp, perf_counts(counter->counts, cpu, thread)); } } diff --git a/tools/perf/util/stat-display.c b/tools/perf/util/stat-display.c index 605a1fd..51d6781 100644 --- a/tools/perf/util/stat-display.c +++ b/tools/perf/util/stat-display.c @@ -119,7 +119,7 @@ static void aggr_printout(struct perf_stat_config *config, config->csv_output ? 0 : 16, perf_thread_map__comm(evsel->core.threads, id), config->csv_output ? 0 : -8, - thread_map__pid(evsel->core.threads, id), + perf_thread_map__pid(evsel->core.threads, id), config->csv_sep); break; case AGGR_GLOBAL: @@ -745,7 +745,7 @@ static void print_aggr_thread(struct perf_stat_config *config, struct evsel *counter, char *prefix) { FILE *output = config->output; - int nthreads = thread_map__nr(counter->core.threads); + int nthreads = perf_thread_map__nr(counter->core.threads); int ncpus = perf_cpu_map__nr(counter->core.cpus); int thread, sorted_threads, id; struct perf_aggr_thread_value *buf; diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index 1e6a25a..0cbfd1e 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -159,7 +159,7 @@ static void perf_evsel__free_prev_raw_counts(struct evsel *evsel) static int perf_evsel__alloc_stats(struct evsel *evsel, bool alloc_raw) { int ncpus = perf_evsel__nr_cpus(evsel); - int nthreads = thread_map__nr(evsel->core.threads); + int nthreads = perf_thread_map__nr(evsel->core.threads); if (perf_evsel__alloc_stat_priv(evsel) < 0 || perf_evsel__alloc_counts(evsel, ncpus, nthreads) < 0 || @@ -309,7 +309,7 @@ process_counter_values(struct perf_stat_config *config, struct evsel *evsel, static int process_counter_maps(struct perf_stat_config *config, struct evsel *counter) { - int nthreads = thread_map__nr(counter->core.threads); + int nthreads = perf_thread_map__nr(counter->core.threads); int ncpus = perf_evsel__nr_cpus(counter); int cpu, thread; diff --git a/tools/perf/util/thread_map.c b/tools/perf/util/thread_map.c index c58385e..3e64525 100644 --- a/tools/perf/util/thread_map.c +++ b/tools/perf/util/thread_map.c @@ -310,7 +310,7 @@ size_t thread_map__fprintf(struct perf_thread_map *threads, FILE *fp) size_t printed = fprintf(fp, "%d thread%s: ", threads->nr, threads->nr > 1 ? "s" : ""); for (i = 0; i < threads->nr; ++i) - printed += fprintf(fp, "%s%d", i ? ", " : "", thread_map__pid(threads, i)); + printed += fprintf(fp, "%s%d", i ? ", " : "", perf_thread_map__pid(threads, i)); return printed + fprintf(fp, "\n"); } @@ -341,7 +341,7 @@ static int get_comm(char **comm, pid_t pid) static void comm_init(struct perf_thread_map *map, int i) { - pid_t pid = thread_map__pid(map, i); + pid_t pid = perf_thread_map__pid(map, i); char *comm = NULL; /* dummy pid comm initialization */ diff --git a/tools/perf/util/thread_map.h b/tools/perf/util/thread_map.h index ba45c76..ca165fd 100644 --- a/tools/perf/util/thread_map.h +++ b/tools/perf/util/thread_map.h @@ -25,16 +25,6 @@ struct perf_thread_map *thread_map__new_by_tid_str(const char *tid_str); size_t thread_map__fprintf(struct perf_thread_map *threads, FILE *fp); -static inline int thread_map__nr(struct perf_thread_map *threads) -{ - return threads ? threads->nr : 1; -} - -static inline pid_t thread_map__pid(struct perf_thread_map *map, int thread) -{ - return map->map[thread].pid; -} - void thread_map__read_comms(struct perf_thread_map *threads); bool thread_map__has(struct perf_thread_map *threads, pid_t pid); int thread_map__remove(struct perf_thread_map *threads, int idx); ^ permalink raw reply related [flat|nested] 11+ messages in thread
end of thread, other threads:[~2019-08-23 12:29 UTC | newest] Thread overview: 11+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2019-08-22 11:11 [PATCH 0/5] tools,libperf: Assorted fixes Jiri Olsa 2019-08-22 11:11 ` [PATCH 1/5] tools: Add missing perf_event.h include Jiri Olsa 2019-08-22 14:13 ` Arnaldo Carvalho de Melo 2019-08-22 11:11 ` [PATCH 2/5] libperf: Use perf_cpu_map__nr instead of cpu_map__nr Jiri Olsa 2019-08-22 11:11 ` [PATCH 3/5] libperf: Move cpu_map__empty to perf_cpu_map__empty Jiri Olsa 2019-08-23 12:21 ` [tip: perf/core] libperf: Move perf's cpu_map__empty() to perf_cpu_map__empty() tip-bot2 for Jiri Olsa 2019-08-22 11:11 ` [PATCH 4/5] libperf: Move cpu_map__idx to perf_cpu_map__idx Jiri Olsa 2019-08-23 12:21 ` [tip: perf/core] libperf: Move perf's cpu_map__idx() to perf_cpu_map__idx() tip-bot2 for Jiri Olsa 2019-08-22 11:11 ` [PATCH 5/5] libperf: Add perf_thread_map__nr/perf_thread_map__pid functions Jiri Olsa 2019-08-22 15:18 ` Arnaldo Carvalho de Melo 2019-08-23 12:29 ` [tip: perf/core] " tip-bot2 for Jiri Olsa
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.