perf tools: Add optional support for libpfm4

Stephane Eranian · acmel · commit 709434907842 · 2020-05-29T16:51:38.000-03:00
This patch links perf with the libpfm4 library if it is available and LIBPFM4 is passed to the build. The libpfm4 library contains hardware event tables for all processors supported by perf_events. It is a helper library that helps convert from a symbolic event name to the event encoding required by the underlying kernel interface. This library is open-source and available from: http://perfmon2.sf.net. With this patch, it is possible to specify full hardware events by name. Hardware filters are also supported. Events must be specified via the --pfm-events and not -e option. Both options are active at the same time and it is possible to mix and match: $ perf stat --pfm-events inst_retired:any_p:c=1:i -e cycles .... One needs to explicitely ask for its inclusion by using the LIBPFM4 make command line option, ie its opt-in rather than opt-out of feature detection and build support. Signed-off-by: Stephane Eranian <eranian@google.com> Reviewed-by: Ian Rogers <irogers@google.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Alexey Budankov <alexey.budankov@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Andrii Nakryiko <andriin@fb.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Florian Fainelli <f.fainelli@gmail.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Igor Lubashev <ilubashe@akamai.com> Cc: Jin Yao <yao.jin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Jiwei Sun <jiwei.sun@windriver.com> Cc: John Garry <john.garry@huawei.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Leo Yan <leo.yan@linaro.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Martin KaFai Lau <kafai@fb.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Yonghong Song <yhs@fb.com> Cc: bpf@vger.kernel.org Cc: netdev@vger.kernel.org Cc: yuzhoujian <yuzhoujian@didichuxing.com> Link: http://lore.kernel.org/lkml/20200505182943.218248-2-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
diff --git a/tools/perf/Documentation/perf-record.txt b/tools/perf/Documentation/perf-record.txt
@@ -615,6 +615,17 @@ appended unit character - B/K/M/G
 	The number of threads to run when synthesizing events for existing processes.
 	By default, the number of threads equals 1.
 
+ifdef::HAVE_LIBPFM[]
+--pfm-events events::
+Select a PMU event using libpfm4 syntax (see http://perfmon2.sf.net)
+including support for event filters. For example '--pfm-events
+inst_retired:any_p:u:c=1:i'. More than one event can be passed to the
+option using the comma separator. Hardware events and generic hardware
+events cannot be mixed together. The latter must be used with the -e
+option. The -e option and this one can be mixed and matched.  Events
+can be grouped using the {} notation.
+endif::HAVE_LIBPFM[]
+
 SEE ALSO
 --------
 linkperf:perf-stat[1], linkperf:perf-list[1], linkperf:perf-intel-pt[1]
diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
@@ -71,6 +71,16 @@ report::
 --tid=<tid>::
         stat events on existing thread id (comma separated list)
 
+ifdef::HAVE_LIBPFM[]
+--pfm-events events::
+Select a PMU event using libpfm4 syntax (see http://perfmon2.sf.net)
+including support for event filters. For example '--pfm-events
+inst_retired:any_p:u:c=1:i'. More than one event can be passed to the
+option using the comma separator. Hardware events and generic hardware
+events cannot be mixed together. The latter must be used with the -e
+option. The -e option and this one can be mixed and matched.  Events
+can be grouped using the {} notation.
+endif::HAVE_LIBPFM[]
 
 -a::
 --all-cpus::
diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt
@@ -329,6 +329,17 @@ Default is to monitor all CPUS.
 	The known limitations include exception handing such as
 	setjmp/longjmp will have calls/returns not match.
 
+ifdef::HAVE_LIBPFM[]
+--pfm-events events::
+Select a PMU event using libpfm4 syntax (see http://perfmon2.sf.net)
+including support for event filters. For example '--pfm-events
+inst_retired:any_p:u:c=1:i'. More than one event can be passed to the
+option using the comma separator. Hardware events and generic hardware
+events cannot be mixed together. The latter must be used with the -e
+option. The -e option and this one can be mixed and matched.  Events
+can be grouped using the {} notation.
+endif::HAVE_LIBPFM[]
+
 INTERACTIVE PROMPTING KEYS
 --------------------------
 
diff --git a/tools/perf/Makefile.config b/tools/perf/Makefile.config
@@ -1022,6 +1022,19 @@ ifdef LIBCLANGLLVM
   endif
 endif
 
+ifdef LIBPFM4
+  $(call feature_check,libpfm4)
+  ifeq ($(feature-libpfm4), 1)
+    CFLAGS += -DHAVE_LIBPFM
+    EXTLIBS += -lpfm
+    ASCIIDOC_EXTRA = -aHAVE_LIBPFM=1
+    $(call detected,CONFIG_LIBPFM4)
+  else
+    msg := $(warning libpfm4 not found, disables libpfm4 support. Please install libpfm4-dev);
+    NO_LIBPFM4 := 1
+  endif
+endif
+
 # Among the variables below, these:
 #   perfexecdir
 #   perf_include_dir
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
@@ -121,6 +121,9 @@ include ../scripts/utilities.mak
 # Define NO_SYSCALL_TABLE=1 to disable the use of syscall id to/from name tables
 # generated from the kernel .tbl or unistd.h files and use, if available, libaudit
 # for doing the conversions to/from strings/id.
+#
+# Define LIBPFM4 to enable libpfm4 events extension.
+#
 
 # As per kernel Makefile, avoid funny character set dependencies
 unexport LC_ALL
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
@@ -45,6 +45,7 @@
 #include "util/units.h"
 #include "util/bpf-event.h"
 #include "util/util.h"
+#include "util/pfm.h"
 #include "asm/bug.h"
 #include "perf.h"
 
@@ -2555,6 +2556,11 @@ static struct option __record_options[] = {
 	OPT_UINTEGER(0, "num-thread-synthesize",
 		     &record.opts.nr_threads_synthesize,
 		     "number of threads to run for event synthesis"),
+#ifdef HAVE_LIBPFM
+	OPT_CALLBACK(0, "pfm-events", &record.evlist, "event",
+		"libpfm4 event selector. use 'perf list' to list available events",
+		parse_libpfm_events_option),
+#endif
 	OPT_END()
 };
 
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
@@ -66,6 +66,7 @@
 #include "util/time-utils.h"
 #include "util/top.h"
 #include "util/affinity.h"
+#include "util/pfm.h"
 #include "asm/bug.h"
 
 #include <linux/time64.h>
@@ -1012,6 +1013,11 @@ static struct option stat_options[] = {
 		    "Use with 'percore' event qualifier to show the event "
 		    "counts of one hardware thread by sum up total hardware "
 		    "threads of same physical core"),
+#ifdef HAVE_LIBPFM
+	OPT_CALLBACK(0, "pfm-events", &evsel_list, "event",
+		"libpfm4 event selector. use 'perf list' to list available events",
+		parse_libpfm_events_option),
+#endif
 	OPT_END()
 };
 
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
@@ -53,6 +53,7 @@
 
 #include "util/debug.h"
 #include "util/ordered-events.h"
+#include "util/pfm.h"
 
 #include <assert.h>
 #include <elf.h>
@@ -1575,6 +1576,11 @@ int cmd_top(int argc, const char **argv)
 		    "WARNING: should be used on grouped events."),
 	OPT_BOOLEAN(0, "stitch-lbr", &top.stitch_lbr,
 		    "Enable LBR callgraph stitching approach"),
+#ifdef HAVE_LIBPFM
+	OPT_CALLBACK(0, "pfm-events", &top.evlist, "event",
+		"libpfm4 event selector. use 'perf list' to list available events",
+		parse_libpfm_events_option),
+#endif
 	OPTS_EVSWITCH(&top.evswitch),
 	OPT_END()
 	};
diff --git a/tools/perf/tests/Build b/tools/perf/tests/Build
@@ -58,6 +58,7 @@ perf-y += time-utils-test.o
 perf-y += genelf.o
 perf-y += api-io.o
 perf-y += demangle-java-test.o
+perf-y += pfm.o
 
 $(OUTPUT)tests/llvm-src-base.c: tests/bpf-script-example.c tests/Build
 	$(call rule_mkdir)
diff --git a/tools/perf/tests/builtin-test.c b/tools/perf/tests/builtin-test.c
@@ -316,6 +316,15 @@ static struct test generic_tests[] = {
 		.desc = "Test jit_write_elf",
 		.func = test__jit_write_elf,
 	},
+	{
+		.desc = "Test libpfm4 support",
+		.func = test__pfm,
+		.subtest = {
+			.skip_if_fail	= true,
+			.get_nr		= test__pfm_subtest_get_nr,
+			.get_desc	= test__pfm_subtest_get_desc,
+		}
+	},
 	{
 		.desc = "Test api io",
 		.func = test__api_io,
diff --git a/tools/perf/tests/pfm.c b/tools/perf/tests/pfm.c
@@ -0,0 +1,203 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Test support for libpfm4 event encodings.
+ *
+ * Copyright 2020 Google LLC.
+ */
+#include "tests.h"
+#include "util/debug.h"
+#include "util/evlist.h"
+#include "util/pfm.h"
+
+#include <linux/kernel.h>
+
+#ifdef HAVE_LIBPFM
+static int test__pfm_events(void);
+static int test__pfm_group(void);
+#endif
+
+static const struct {
+	int (*func)(void);
+	const char *desc;
+} pfm_testcase_table[] = {
+#ifdef HAVE_LIBPFM
+	{
+		.func = test__pfm_events,
+		.desc = "test of individual --pfm-events",
+	},
+	{
+		.func = test__pfm_group,
+		.desc = "test groups of --pfm-events",
+	},
+#endif
+};
+
+#ifdef HAVE_LIBPFM
+static int count_pfm_events(struct perf_evlist *evlist)
+{
+	struct perf_evsel *evsel;
+	int count = 0;
+
+	perf_evlist__for_each_entry(evlist, evsel) {
+		count++;
+	}
+	return count;
+}
+
+static int test__pfm_events(void)
+{
+	struct evlist *evlist;
+	struct option opt;
+	size_t i;
+	const struct {
+		const char *events;
+		int nr_events;
+	} table[] = {
+		{
+			.events = "",
+			.nr_events = 0,
+		},
+		{
+			.events = "instructions",
+			.nr_events = 1,
+		},
+		{
+			.events = "instructions,cycles",
+			.nr_events = 2,
+		},
+		{
+			.events = "stereolab",
+			.nr_events = 0,
+		},
+		{
+			.events = "instructions,instructions",
+			.nr_events = 2,
+		},
+		{
+			.events = "stereolab,instructions",
+			.nr_events = 0,
+		},
+		{
+			.events = "instructions,stereolab",
+			.nr_events = 1,
+		},
+	};
+
+	for (i = 0; i < ARRAY_SIZE(table); i++) {
+		evlist = evlist__new();
+		if (evlist == NULL)
+			return -ENOMEM;
+
+		opt.value = evlist;
+		parse_libpfm_events_option(&opt,
+					table[i].events,
+					0);
+		TEST_ASSERT_EQUAL(table[i].events,
+				count_pfm_events(&evlist->core),
+				table[i].nr_events);
+		TEST_ASSERT_EQUAL(table[i].events,
+				evlist->nr_groups,
+				0);
+
+		evlist__delete(evlist);
+	}
+	return 0;
+}
+
+static int test__pfm_group(void)
+{
+	struct evlist *evlist;
+	struct option opt;
+	size_t i;
+	const struct {
+		const char *events;
+		int nr_events;
+		int nr_groups;
+	} table[] = {
+		{
+			.events = "{},",
+			.nr_events = 0,
+			.nr_groups = 0,
+		},
+		{
+			.events = "{instructions}",
+			.nr_events = 1,
+			.nr_groups = 1,
+		},
+		{
+			.events = "{instructions},{}",
+			.nr_events = 1,
+			.nr_groups = 1,
+		},
+		{
+			.events = "{},{instructions}",
+			.nr_events = 0,
+			.nr_groups = 0,
+		},
+		{
+			.events = "{instructions},{instructions}",
+			.nr_events = 2,
+			.nr_groups = 2,
+		},
+		{
+			.events = "{instructions,cycles},{instructions,cycles}",
+			.nr_events = 4,
+			.nr_groups = 2,
+		},
+		{
+			.events = "{stereolab}",
+			.nr_events = 0,
+			.nr_groups = 0,
+		},
+		{
+			.events =
+			"{instructions,cycles},{instructions,stereolab}",
+			.nr_events = 3,
+			.nr_groups = 1,
+		},
+	};
+
+	for (i = 0; i < ARRAY_SIZE(table); i++) {
+		evlist = evlist__new();
+		if (evlist == NULL)
+			return -ENOMEM;
+
+		opt.value = evlist;
+		parse_libpfm_events_option(&opt,
+					table[i].events,
+					0);
+		TEST_ASSERT_EQUAL(table[i].events,
+				count_pfm_events(&evlist->core),
+				table[i].nr_events);
+		TEST_ASSERT_EQUAL(table[i].events,
+				evlist->nr_groups,
+				table[i].nr_groups);
+
+		evlist__delete(evlist);
+	}
+	return 0;
+}
+#endif
+
+const char *test__pfm_subtest_get_desc(int i)
+{
+	if (i < 0 || i >= (int)ARRAY_SIZE(pfm_testcase_table))
+		return NULL;
+	return pfm_testcase_table[i].desc;
+}
+
+int test__pfm_subtest_get_nr(void)
+{
+	return (int)ARRAY_SIZE(pfm_testcase_table);
+}
+
+int test__pfm(struct test *test __maybe_unused, int i __maybe_unused)
+{
+#ifdef HAVE_LIBPFM
+	if (i < 0 || i >= (int)ARRAY_SIZE(pfm_testcase_table))
+		return TEST_FAIL;
+	return pfm_testcase_table[i].func();
+#else
+	return TEST_SKIP;
+#endif
+}
diff --git a/tools/perf/tests/tests.h b/tools/perf/tests/tests.h
@@ -118,6 +118,9 @@ int test__time_utils(struct test *t, int subtest);
 int test__jit_write_elf(struct test *test, int subtest);
 int test__api_io(struct test *test, int subtest);
 int test__demangle_java(struct test *test, int subtest);
+int test__pfm(struct test *test, int subtest);
+const char *test__pfm_subtest_get_desc(int subtest);
+int test__pfm_subtest_get_nr(void);
 
 bool test__bp_signal_is_supported(void);
 bool test__bp_account_is_supported(void);
diff --git a/tools/perf/util/Build b/tools/perf/util/Build
@@ -183,6 +183,8 @@ perf-$(CONFIG_LIBBPF) += bpf-event.o
 
 perf-$(CONFIG_CXX) += c++/
 
+perf-$(CONFIG_LIBPFM4) += pfm.o
+
 CFLAGS_config.o   += -DETC_PERFCONFIG="BUILD_STR($(ETC_PERFCONFIG_SQ))"
 CFLAGS_llvm-utils.o += -DPERF_INCLUDE_DIR="BUILD_STR($(perf_include_dir_SQ))"
 
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
@@ -2417,7 +2417,7 @@ bool evsel__fallback(struct evsel *evsel, int err, char *msg, size_t msgsize)
 
 		/* Is there already the separator in the name. */
 		if (strchr(name, '/') ||
-		    strchr(name, ':'))
+		    (strchr(name, ':') && !evsel->is_libpfm_event))
 			sep = "";
 
 		if (asprintf(&new_name, "%s%su", name, sep) < 0)
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
diff --git a/tools/perf/util/parse-events.h b/tools/perf/util/parse-events.h
diff --git a/tools/perf/util/pfm.c b/tools/perf/util/pfm.c
diff --git a/tools/perf/util/pfm.h b/tools/perf/util/pfm.h
