Merge tag 'perf-tools-for-v5.19-2022-05-28' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

Pull more perf tools updates from Arnaldo Carvalho de Melo:

 - Add BPF based off-CPU profiling

 - Improvements for system wide recording, specially for Intel PT

 - Improve DWARF unwinding on arm64

 - Support Arm CoreSight trace data disassembly in 'perf script' python

 - Fix build with new libbpf version, related to supporting older
   versions of distro released libbpf packages

 - Fix event syntax error caused by ExtSel in the JSON events infra

 - Use stdio interface if slang is not supported in 'perf c2c'

 - Add 'perf test' checking for perf stat CSV output

 - Sync the msr-index.h copy with the kernel sources

* tag 'perf-tools-for-v5.19-2022-05-28' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux: (38 commits)
  tools arch x86: Sync the msr-index.h copy with the kernel sources
  perf scripts python: Support Arm CoreSight trace data disassembly
  perf scripting python: Expose dso and map information
  perf jevents: Fix event syntax error caused by ExtSel
  perf tools arm64: Add support for VG register
  perf unwind arm64: Decouple Libunwind register names from Perf
  perf unwind: Use dynamic register set for DWARF unwind
  perf tools arm64: Copy perf_regs.h from the kernel
  perf unwind arm64: Use perf's copy of kernel headers
  perf c2c: Use stdio interface if slang is not supported
  perf test: Add a basic offcpu profiling test
  perf record: Add cgroup support for off-cpu profiling
  perf record: Handle argument change in sched_switch
  perf record: Implement basic filtering for off-cpu
  perf record: Enable off-cpu analysis with BPF
  perf report: Do not extend sample type of bpf-output event
  perf test: Add checking for perf stat CSV output.
  perf tools: Allow system-wide events to keep their own threads
  perf tools: Allow system-wide events to keep their own CPUs
  libperf evsel: Add comments for booleans
  ...

1 files changed, 56 insertions, 5 deletions

diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index 7f9f588e88c6..48af7d379d82 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -242,14 +242,20 @@ int __evlist__add_default(struct evlist *evlist, bool precise)
 	return 0;
 }
 
-int evlist__add_dummy(struct evlist *evlist)
+static struct evsel *evlist__dummy_event(struct evlist *evlist)
 {
 	struct perf_event_attr attr = {
 		.type	= PERF_TYPE_SOFTWARE,
 		.config = PERF_COUNT_SW_DUMMY,
 		.size	= sizeof(attr), /* to capture ABI version */
 	};
-	struct evsel *evsel = evsel__new_idx(&attr, evlist->core.nr_entries);
+
+	return evsel__new_idx(&attr, evlist->core.nr_entries);
+}
+
+int evlist__add_dummy(struct evlist *evlist)
+{
+	struct evsel *evsel = evlist__dummy_event(evlist);
 
 	if (evsel == NULL)
 		return -ENOMEM;
@@ -258,6 +264,51 @@ int evlist__add_dummy(struct evlist *evlist)
 	return 0;
 }
 
+static void evlist__add_on_all_cpus(struct evlist *evlist, struct evsel *evsel)
+{
+	evsel->core.system_wide = true;
+
+	/*
+	 * All CPUs.
+	 *
+	 * Note perf_event_open() does not accept CPUs that are not online, so
+	 * in fact this CPU list will include only all online CPUs.
+	 */
+	perf_cpu_map__put(evsel->core.own_cpus);
+	evsel->core.own_cpus = perf_cpu_map__new(NULL);
+	perf_cpu_map__put(evsel->core.cpus);
+	evsel->core.cpus = perf_cpu_map__get(evsel->core.own_cpus);
+
+	/* No threads */
+	perf_thread_map__put(evsel->core.threads);
+	evsel->core.threads = perf_thread_map__new_dummy();
+
+	evlist__add(evlist, evsel);
+}
+
+struct evsel *evlist__add_aux_dummy(struct evlist *evlist, bool system_wide)
+{
+	struct evsel *evsel = evlist__dummy_event(evlist);
+
+	if (!evsel)
+		return NULL;
+
+	evsel->core.attr.exclude_kernel = 1;
+	evsel->core.attr.exclude_guest = 1;
+	evsel->core.attr.exclude_hv = 1;
+	evsel->core.attr.freq = 0;
+	evsel->core.attr.sample_period = 1;
+	evsel->no_aux_samples = true;
+	evsel->name = strdup("dummy:u");
+
+	if (system_wide)
+		evlist__add_on_all_cpus(evlist, evsel);
+	else
+		evlist__add(evlist, evsel);
+
+	return evsel;
+}
+
 static int evlist__add_attrs(struct evlist *evlist, struct perf_event_attr *attrs, size_t nr_attrs)
 {
 	struct evsel *evsel, *n;
@@ -747,15 +798,15 @@ static struct mmap *evlist__alloc_mmap(struct evlist *evlist,
 
 static void
 perf_evlist__mmap_cb_idx(struct perf_evlist *_evlist,
-			 struct perf_evsel *_evsel __maybe_unused,
+			 struct perf_evsel *_evsel,
 			 struct perf_mmap_param *_mp,
 			 int idx)
 {
 	struct evlist *evlist = container_of(_evlist, struct evlist, core);
 	struct mmap_params *mp = container_of(_mp, struct mmap_params, core);
-	bool per_cpu = !perf_cpu_map__empty(_evlist->user_requested_cpus);
+	struct evsel *evsel = container_of(_evsel, struct evsel, core);
 
-	auxtrace_mmap_params__set_idx(&mp->auxtrace_mp, evlist, idx, per_cpu);
+	auxtrace_mmap_params__set_idx(&mp->auxtrace_mp, evlist, evsel, idx);
 }
 
 static struct perf_mmap*