Merge tag 'perf-core-for-mingo-4.16-20171206' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core

Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:

- Improve build messages for files needed by Intel-PT, originally copied
  from the kernel sources, that drifted from its original (Adrian Hunter)

- Allow computing 'perf stat' style metrics in 'perf script' (Andi Kleen)

- Fix 'perf report -D' output for user metadata events (Arnaldo Carvalho de Melo)

- Add feature test for pthread_barrier_t availability (Arnaldo Carvalho de Melo)

- Allow again using x86's asm.h when building for the 'bpf' clang target,
  making some 'perf test' LLVM/BPF entries work again (Arnaldo Carvalho de Melo)

- Use cpumaps in 'perf bench futex', eliminating some code duplication
  (Davidlohr Bueso)

- Improve PMU infrastructure to support amp64's ThunderX2 implementation
  defined core events (Ganapatrao Kulkarni)

- Add hint about how to add USDT probes for Node.js (Hansuk Hong)

- s/390 needs -fPIC to be incrementally linked or linked to shared
  libraries (Hendrik Brueckner)

- Use pthread_barrier to synch 'perf bench futex wake-parallel' waker
  threads (James Yang)

- Fix up build in hardened environments, such as Fedora 27 (Jiri Olsa)

- Add a tip about cacheline events in 'perf c2c' (Sangwon Hong)

- Set browser mode right before setup_browser(), because we may have
  errors printed before that, which were getting lost (Seokho Song)

- s390x doesn't support PERF_TYPE_BREAKPOINT, so disable 'perf test'
  cases 19 and 20 on s390x, that tests that feature (Thomas Richter)

- Fix unnecessary memory allocation for s390x 'perf annotate' objdump
  parsing, which could lead to thousands of needless entries in the
  instruction handling array (Thomas Richter)

- Fix objdump comment parsing for Intel mov dissassembly (Thomas Richter)

- Clarify usage of 'overwrite' and 'backward' in the evlist/mmap code,
  removing the 'overwrite' parameter from several functions as it was
  always used it as 'false' (Wang Nan)

- Fix 'perf record' backward recording, it wasn't doing what was
  expected: overwriting records when the ring buffer gets full (Wang Nan)

- Use more flexible pattern matching for CPU identification for perf
  vendor event's mapfile.csv, removing the need for a new perf binary
  for a sligthly different chip revision that shares the same set of
  counters (William Cohen)

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>

1 files changed, 95 insertions, 2 deletions

diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index ee7c7aaaae72..39d8b55f0db3 100644
--- a/tools/perf/builtin-script.c
+++ b/tools/perf/builtin-script.c
@@ -22,6 +22,7 @@
 #include "util/cpumap.h"
 #include "util/thread_map.h"
 #include "util/stat.h"
+#include "util/color.h"
 #include "util/string2.h"
 #include "util/thread-stack.h"
 #include "util/time-utils.h"
@@ -90,6 +91,7 @@ enum perf_output_field {
 	PERF_OUTPUT_SYNTH           = 1U << 25,
 	PERF_OUTPUT_PHYS_ADDR       = 1U << 26,
 	PERF_OUTPUT_UREGS	    = 1U << 27,
+	PERF_OUTPUT_METRIC	    = 1U << 28,
 };
 
 struct output_option {
@@ -124,6 +126,7 @@ struct output_option {
 	{.str = "brstackoff", .field = PERF_OUTPUT_BRSTACKOFF},
 	{.str = "synth", .field = PERF_OUTPUT_SYNTH},
 	{.str = "phys_addr", .field = PERF_OUTPUT_PHYS_ADDR},
+	{.str = "metric", .field = PERF_OUTPUT_METRIC},
 };
 
 enum {
@@ -215,12 +218,20 @@ struct perf_evsel_script {
        char *filename;
        FILE *fp;
        u64  samples;
+       /* For metric output */
+       u64  val;
+       int  gnum;
 };
 
+static inline struct perf_evsel_script *evsel_script(struct perf_evsel *evsel)
+{
+	return (struct perf_evsel_script *)evsel->priv;
+}
+
 static struct perf_evsel_script *perf_evsel_script__new(struct perf_evsel *evsel,
 							struct perf_data *data)
 {
-	struct perf_evsel_script *es = malloc(sizeof(*es));
+	struct perf_evsel_script *es = zalloc(sizeof(*es));
 
 	if (es != NULL) {
 		if (asprintf(&es->filename, "%s.%s.dump", data->file.path, perf_evsel__name(evsel)) < 0)
@@ -228,7 +239,6 @@ static struct perf_evsel_script *perf_evsel_script__new(struct perf_evsel *evsel
 		es->fp = fopen(es->filename, "w");
 		if (es->fp == NULL)
 			goto out_free_filename;
-		es->samples = 0;
 	}
 
 	return es;
@@ -1472,6 +1482,86 @@ static int data_src__fprintf(u64 data_src, FILE *fp)
 	return fprintf(fp, "%-*s", maxlen, out);
 }
 
+struct metric_ctx {
+	struct perf_sample	*sample;
+	struct thread		*thread;
+	struct perf_evsel	*evsel;
+	FILE 			*fp;
+};
+
+static void script_print_metric(void *ctx, const char *color,
+			        const char *fmt,
+			        const char *unit, double val)
+{
+	struct metric_ctx *mctx = ctx;
+
+	if (!fmt)
+		return;
+	perf_sample__fprintf_start(mctx->sample, mctx->thread, mctx->evsel,
+				   mctx->fp);
+	fputs("\tmetric: ", mctx->fp);
+	if (color)
+		color_fprintf(mctx->fp, color, fmt, val);
+	else
+		printf(fmt, val);
+	fprintf(mctx->fp, " %s\n", unit);
+}
+
+static void script_new_line(void *ctx)
+{
+	struct metric_ctx *mctx = ctx;
+
+	perf_sample__fprintf_start(mctx->sample, mctx->thread, mctx->evsel,
+				   mctx->fp);
+	fputs("\tmetric: ", mctx->fp);
+}
+
+static void perf_sample__fprint_metric(struct perf_script *script,
+				       struct thread *thread,
+				       struct perf_evsel *evsel,
+				       struct perf_sample *sample,
+				       FILE *fp)
+{
+	struct perf_stat_output_ctx ctx = {
+		.print_metric = script_print_metric,
+		.new_line = script_new_line,
+		.ctx = &(struct metric_ctx) {
+				.sample = sample,
+				.thread = thread,
+				.evsel  = evsel,
+				.fp     = fp,
+			 },
+		.force_header = false,
+	};
+	struct perf_evsel *ev2;
+	static bool init;
+	u64 val;
+
+	if (!init) {
+		perf_stat__init_shadow_stats();
+		init = true;
+	}
+	if (!evsel->stats)
+		perf_evlist__alloc_stats(script->session->evlist, false);
+	if (evsel_script(evsel->leader)->gnum++ == 0)
+		perf_stat__reset_shadow_stats();
+	val = sample->period * evsel->scale;
+	perf_stat__update_shadow_stats(evsel,
+				       val,
+				       sample->cpu);
+	evsel_script(evsel)->val = val;
+	if (evsel_script(evsel->leader)->gnum == evsel->leader->nr_members) {
+		for_each_group_member (ev2, evsel->leader) {
+			perf_stat__print_shadow_stats(ev2,
+						      evsel_script(ev2)->val,
+						      sample->cpu,
+						      &ctx,
+						      NULL);
+		}
+		evsel_script(evsel->leader)->gnum = 0;
+	}
+}
+
 static void process_event(struct perf_script *script,
 			  struct perf_sample *sample, struct perf_evsel *evsel,
 			  struct addr_location *al,
@@ -1559,6 +1649,9 @@ static void process_event(struct perf_script *script,
 	if (PRINT_FIELD(PHYS_ADDR))
 		fprintf(fp, "%16" PRIx64, sample->phys_addr);
 	fprintf(fp, "\n");
+
+	if (PRINT_FIELD(METRIC))
+		perf_sample__fprint_metric(script, thread, evsel, sample, fp);
 }
 
 static struct scripting_ops	*scripting_ops;