Merge tag 'perf-tools-for-v5.15-2021-09-04' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

Pull perf tool updates from Arnaldo Carvalho de Melo:
 "New features:

   - Improvements for the flamegraph python script, including:
       - Display perf.data header
       - Display PIDs of user stacks
       - Added option to change color scheme
       - Default to blue/green color scheme to improve accessibility
       - Correctly identify kernel stacks when debuginfo is available

   - Improvements for 'perf bench futex':
       - Add --mlockall parameter
       - Add --broadcast and --pi to the 'requeue' sub benchmark

   - Add support for PMU aliases.

   - Introduce an ARM Coresight ETE decoder.

   - Add a 'perf bench' entry for evlist open/close operations, to help
     quantify improvements with multithreading 'perf record'.

   - Allow reporting the [un]throttle PERF_RECORD_ meta event in 'perf
     script's python scripting.

   - Add a 'perf test' entry for PMU aliases.

   - Add a 'perf test' entry for 'perf record/perf report/perf script'
     pipe mode.

  Fixes:

   - perf script dlfilter (API for filtering via dynamically loaded
     shared object introduced in v5.14) fixes and a 'perf test' entry
     for it.

   - Fix get_current_dir_name() compilation on Android.

   - Fix issues with asciidoc and double dashes uses.

   - Fix memory leaks in the BTF handling code.

   - Fix leftover problems in the Documentation from the infrastructure
     originally lifted from the git codebase.

   - Fix *probe_vfs_getname.sh 'perf test' failures.

   - Handle fd gaps in 'perf test's test__dso_data_reopen().

   - Make sure to show disasembly warnings for 'perf annotate --stdio'.

   - Fix output from pipe to file and vice-versa in 'perf
     record/report/script'.

   - Correct 'perf data -h' output.

   - Fix wrong comm in system-wide mode with 'perf record --delay'.

   - Do not allow --for-each-cgroup without cpu in 'perf stat'

   - Make 'perf test --skip' work on shell tests.

   - Fix libperf's verbose printing.

  Misc improvements:

   - Preparatory patches for multithreading various 'perf record' phases
     (synthesizing, opening, recording, etc).

   - Add sparse context/locking annotations in compiler-types.h, also to
     help with the multithreading effort.

   - Optimize the generation of the arch specific erno tables used in
     'perf trace'.

   - Optimize libperf's perf_cpu_map__max().

   - Improve ARM's CoreSight warnings.

   - Report collisions in AUX records.

   - Improve warnings for the LLVM 'perf test' entry.

   - Improve the PMU events 'perf test' codebase.

   - perf test: Do not compare overheads in the zstd comp test

   - Better support annotation on ARM.

   - Update 'perf trace's cmd string table to decode sys_bpf() first
     arg.

  Vendor events:

   - Add JSON events and metrics for Intel's Ice Lake, Tiger Lake and
     Elhart Lake.

   - Update JSON eventsand metrics for Intel's Cascade Lake and Sky Lake
     servers.

  Hardware tracing:

   - Improvements for the ARM hardware tracing auxtrace support"

* tag 'perf-tools-for-v5.15-2021-09-04' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux: (130 commits)
  perf tests: Add test for PMU aliases
  perf pmu: Add PMU alias support
  perf session: Report collisions in AUX records
  perf script python: Allow reporting the [un]throttle PERF_RECORD_ meta event
  perf build: Report failure for testing feature libopencsd
  perf cs-etm: Show a warning for an unknown magic number
  perf cs-etm: Print the decoder name
  perf cs-etm: Create ETE decoder
  perf cs-etm: Update OpenCSD decoder for ETE
  perf cs-etm: Fix typo
  perf cs-etm: Save TRCDEVARCH register
  perf cs-etm: Refactor out ETMv4 header saving
  perf cs-etm: Initialise architecture based on TRCIDR1
  perf cs-etm: Refactor initialisation of decoder params.
  tools build: Fix feature detect clean for out of source builds
  perf evlist: Add evlist__for_each_entry_from() macro
  perf evsel: Handle precise_ip fallback in evsel__open_cpu()
  perf evsel: Move bpf_counter__install_pe() to success path in evsel__open_cpu()
  perf evsel: Move test_attr__open() to success path in evsel__open_cpu()
  perf evsel: Move ignore_missing_thread() to fallback code
  ...

1 files changed, 258 insertions, 0 deletions

diff --git a/tools/perf/bench/evlist-open-close.c b/tools/perf/bench/evlist-open-close.c
new file mode 100644
index 000000000000..83e9897c64a1
--- /dev/null
+++ b/tools/perf/bench/evlist-open-close.c
@@ -0,0 +1,258 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <inttypes.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <limits.h>
+#include "bench.h"
+#include "../util/debug.h"
+#include "../util/stat.h"
+#include "../util/evlist.h"
+#include "../util/evsel.h"
+#include "../util/strbuf.h"
+#include "../util/record.h"
+#include "../util/parse-events.h"
+#include "internal/threadmap.h"
+#include "internal/cpumap.h"
+#include <linux/perf_event.h>
+#include <linux/kernel.h>
+#include <linux/time64.h>
+#include <linux/string.h>
+#include <subcmd/parse-options.h>
+
+#define MMAP_FLUSH_DEFAULT 1
+
+static int iterations = 100;
+static int nr_events = 1;
+static const char *event_string = "dummy";
+
+static struct record_opts opts = {
+	.sample_time	     = true,
+	.mmap_pages	     = UINT_MAX,
+	.user_freq	     = UINT_MAX,
+	.user_interval	     = ULLONG_MAX,
+	.freq		     = 4000,
+	.target		     = {
+		.uses_mmap   = true,
+		.default_per_cpu = true,
+	},
+	.mmap_flush          = MMAP_FLUSH_DEFAULT,
+	.nr_threads_synthesize = 1,
+	.ctl_fd              = -1,
+	.ctl_fd_ack          = -1,
+};
+
+static const struct option options[] = {
+	OPT_STRING('e', "event", &event_string, "event", "event selector. use 'perf list' to list available events"),
+	OPT_INTEGER('n', "nr-events", &nr_events,
+		     "number of dummy events to create (default 1). If used with -e, it clones those events n times (1 = no change)"),
+	OPT_INTEGER('i', "iterations", &iterations, "Number of iterations used to compute average (default=100)"),
+	OPT_BOOLEAN('a', "all-cpus", &opts.target.system_wide, "system-wide collection from all CPUs"),
+	OPT_STRING('C', "cpu", &opts.target.cpu_list, "cpu", "list of cpus where to open events"),
+	OPT_STRING('p', "pid", &opts.target.pid, "pid", "record events on existing process id"),
+	OPT_STRING('t', "tid", &opts.target.tid, "tid", "record events on existing thread id"),
+	OPT_STRING('u', "uid", &opts.target.uid_str, "user", "user to profile"),
+	OPT_BOOLEAN(0, "per-thread", &opts.target.per_thread, "use per-thread mmaps"),
+	OPT_END()
+};
+
+static const char *const bench_usage[] = {
+	"perf bench internals evlist-open-close <options>",
+	NULL
+};
+
+static int evlist__count_evsel_fds(struct evlist *evlist)
+{
+	struct evsel *evsel;
+	int cnt = 0;
+
+	evlist__for_each_entry(evlist, evsel)
+		cnt += evsel->core.threads->nr * evsel->core.cpus->nr;
+
+	return cnt;
+}
+
+static struct evlist *bench__create_evlist(char *evstr)
+{
+	struct parse_events_error err = { .idx = 0, };
+	struct evlist *evlist = evlist__new();
+	int ret;
+
+	if (!evlist) {
+		pr_err("Not enough memory to create evlist\n");
+		return NULL;
+	}
+
+	ret = parse_events(evlist, evstr, &err);
+	if (ret) {
+		parse_events_print_error(&err, evstr);
+		pr_err("Run 'perf list' for a list of valid events\n");
+		ret = 1;
+		goto out_delete_evlist;
+	}
+
+	ret = evlist__create_maps(evlist, &opts.target);
+	if (ret < 0) {
+		pr_err("Not enough memory to create thread/cpu maps\n");
+		goto out_delete_evlist;
+	}
+
+	evlist__config(evlist, &opts, NULL);
+
+	return evlist;
+
+out_delete_evlist:
+	evlist__delete(evlist);
+	return NULL;
+}
+
+static int bench__do_evlist_open_close(struct evlist *evlist)
+{
+	char sbuf[STRERR_BUFSIZE];
+	int err = evlist__open(evlist);
+
+	if (err < 0) {
+		pr_err("evlist__open: %s\n", str_error_r(errno, sbuf, sizeof(sbuf)));
+		return err;
+	}
+
+	err = evlist__mmap(evlist, opts.mmap_pages);
+	if (err < 0) {
+		pr_err("evlist__mmap: %s\n", str_error_r(errno, sbuf, sizeof(sbuf)));
+		return err;
+	}
+
+	evlist__enable(evlist);
+	evlist__disable(evlist);
+	evlist__munmap(evlist);
+	evlist__close(evlist);
+
+	return 0;
+}
+
+static int bench_evlist_open_close__run(char *evstr)
+{
+	// used to print statistics only
+	struct evlist *evlist = bench__create_evlist(evstr);
+	double time_average, time_stddev;
+	struct timeval start, end, diff;
+	struct stats time_stats;
+	u64 runtime_us;
+	int i, err;
+
+	if (!evlist)
+		return -ENOMEM;
+
+	init_stats(&time_stats);
+
+	printf("  Number of cpus:\t%d\n", evlist->core.cpus->nr);
+	printf("  Number of threads:\t%d\n", evlist->core.threads->nr);
+	printf("  Number of events:\t%d (%d fds)\n",
+		evlist->core.nr_entries, evlist__count_evsel_fds(evlist));
+	printf("  Number of iterations:\t%d\n", iterations);
+
+	evlist__delete(evlist);
+
+	for (i = 0; i < iterations; i++) {
+		pr_debug("Started iteration %d\n", i);
+		evlist = bench__create_evlist(evstr);
+		if (!evlist)
+			return -ENOMEM;
+
+		gettimeofday(&start, NULL);
+		err = bench__do_evlist_open_close(evlist);
+		if (err) {
+			evlist__delete(evlist);
+			return err;
+		}
+
+		gettimeofday(&end, NULL);
+		timersub(&end, &start, &diff);
+		runtime_us = diff.tv_sec * USEC_PER_SEC + diff.tv_usec;
+		update_stats(&time_stats, runtime_us);
+
+		evlist__delete(evlist);
+		pr_debug("Iteration %d took:\t%" PRIu64 "us\n", i, runtime_us);
+	}
+
+	time_average = avg_stats(&time_stats);
+	time_stddev = stddev_stats(&time_stats);
+	printf("  Average open-close took: %.3f usec (+- %.3f usec)\n", time_average, time_stddev);
+
+	return 0;
+}
+
+static char *bench__repeat_event_string(const char *evstr, int n)
+{
+	char sbuf[STRERR_BUFSIZE];
+	struct strbuf buf;
+	int i, str_size = strlen(evstr),
+	    final_size = str_size * n + n,
+	    err = strbuf_init(&buf, final_size);
+
+	if (err) {
+		pr_err("strbuf_init: %s\n", str_error_r(err, sbuf, sizeof(sbuf)));
+		goto out_error;
+	}
+
+	for (i = 0; i < n; i++) {
+		err = strbuf_add(&buf, evstr, str_size);
+		if (err) {
+			pr_err("strbuf_add: %s\n", str_error_r(err, sbuf, sizeof(sbuf)));
+			goto out_error;
+		}
+
+		err = strbuf_addch(&buf, i == n-1 ? '\0' : ',');
+		if (err) {
+			pr_err("strbuf_addch: %s\n", str_error_r(err, sbuf, sizeof(sbuf)));
+			goto out_error;
+		}
+	}
+
+	return strbuf_detach(&buf, NULL);
+
+out_error:
+	strbuf_release(&buf);
+	return NULL;
+}
+
+
+int bench_evlist_open_close(int argc, const char **argv)
+{
+	char *evstr, errbuf[BUFSIZ];
+	int err;
+
+	argc = parse_options(argc, argv, options, bench_usage, 0);
+	if (argc) {
+		usage_with_options(bench_usage, options);
+		exit(EXIT_FAILURE);
+	}
+
+	err = target__validate(&opts.target);
+	if (err) {
+		target__strerror(&opts.target, err, errbuf, sizeof(errbuf));
+		pr_err("%s\n", errbuf);
+		goto out;
+	}
+
+	err = target__parse_uid(&opts.target);
+	if (err) {
+		target__strerror(&opts.target, err, errbuf, sizeof(errbuf));
+		pr_err("%s", errbuf);
+		goto out;
+	}
+
+	/* Enable ignoring missing threads when -u/-p option is defined. */
+	opts.ignore_missing_thread = opts.target.uid != UINT_MAX || opts.target.pid;
+
+	evstr = bench__repeat_event_string(event_string, nr_events);
+	if (!evstr) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	err = bench_evlist_open_close__run(evstr);
+
+	free(evstr);
+out:
+	return err;
+}