linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andi Kleen <andi@firstfloor.org>
To: acme@kernel.org
Cc: jolsa@kernel.org, linux-kernel@vger.kernel.org,
	eranian@google.com, namhyung@kernel.org, peterz@infradead.org,
	mingo@kernel.org, Andi Kleen <ak@linux.intel.com>
Subject: [PATCH 3/9] perf, tools, stat: Basic support for TopDown in perf stat
Date: Fri,  7 Aug 2015 18:06:19 -0700	[thread overview]
Message-ID: <1438995985-13631-4-git-send-email-andi@firstfloor.org> (raw)
In-Reply-To: <1438995985-13631-1-git-send-email-andi@firstfloor.org>

From: Andi Kleen <ak@linux.intel.com>

Add basic plumbing for TopDown in perf stat

Add a new --topdown options to enable events.
When --topdown is specified set up events for all topdown
events supported by the kernel.
Add topdown-* as a special case to the event parser, as is
needed for all events containing -.

The actual code to compute the metrics is in follow-on patches.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 tools/perf/Documentation/perf-stat.txt |   8 +++
 tools/perf/builtin-stat.c              | 124 ++++++++++++++++++++++++++++++++-
 tools/perf/util/parse-events.l         |   1 +
 3 files changed, 131 insertions(+), 2 deletions(-)

diff --git a/tools/perf/Documentation/perf-stat.txt b/tools/perf/Documentation/perf-stat.txt
index 47469ab..86c03e9 100644
--- a/tools/perf/Documentation/perf-stat.txt
+++ b/tools/perf/Documentation/perf-stat.txt
@@ -158,6 +158,14 @@ filter out the startup phase of the program, which is often very different.
 
 Print statistics of transactional execution if supported.
 
+--topdown::
+
+Print top down level 1 metrics if supported by the CPU. This allows to
+determine bottle necks in the CPU pipeline for CPU bound workloads,
+by breaking it down into frontend bound, backend bound, bad speculation
+and retiring.  Specifying the option multiple times shows metrics even
+if the don't cross a threshold.
+
 EXAMPLES
 --------
 
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 2590c75..a83f26f 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -93,6 +93,15 @@ static const char * transaction_limited_attrs = {
 	"}"
 };
 
+static const char * topdown_attrs[] = {
+	"topdown-total-slots",
+	"topdown-fetch-bubbles",
+	"topdown-slots-retired",
+	"topdown-recovery-bubbles",
+	"topdown-slots-issued",
+	NULL,
+};
+
 static struct perf_evlist	*evsel_list;
 
 static struct target target = {
@@ -105,6 +114,7 @@ static volatile pid_t		child_pid			= -1;
 static bool			null_run			=  false;
 static int			detailed_run			=  0;
 static bool			transaction_run;
+static int			topdown_run			= 0;
 static bool			big_num				=  true;
 static int			big_num_opt			=  -1;
 static const char		*csv_sep			= NULL;
@@ -735,7 +745,8 @@ static void printout(int id, int nr, struct perf_evsel *counter, double uval,
 				first_shadow_cpu(counter, id),
 				pm,
 				nl,
-				&os);
+				&os,
+				topdown_run);
 
 	if (!csv_output) {
 		print_noise(counter, noise);
@@ -1093,12 +1104,90 @@ static int perf_stat_init_aggr_mode(void)
 	return 0;
 }
 
+static void filter_events(const char **attr, char **str, bool use_group)
+{
+	int off = 0;
+	int i;
+	int len = 0;
+	char *s;
+
+	for (i = 0; attr[i]; i++) {
+		if (pmu_have_event("cpu", attr[i])) {
+			len += strlen(attr[i]) + 1;
+			attr[i - off] = attr[i];
+		} else
+			off++;
+	}
+	attr[i - off] = NULL;
+
+	*str = malloc(len + 1 + 2);
+	if (!*str)
+		return;
+	s = *str;
+	if (i - off == 0) {
+		*s = 0;
+		return;
+	}
+	if (use_group)
+		*s++ = '{';
+	for (i = 0; attr[i]; i++) {
+		strcpy(s, attr[i]);
+		s += strlen(s);
+		*s++ = ',';
+	}
+	if (use_group) {
+		s[-1] = '}';
+		*s = 0;
+	} else
+		s[-1] = 0;
+}
+
+/* Caller must free result */
+static char *sysctl_read(const char *fn)
+{
+	int n;
+	char *line = NULL;
+	size_t linelen = 0;
+	FILE *f = fopen(fn, "r");
+	if (!f)
+		return NULL;
+	n = getline(&line, &linelen, f);
+	fclose(f);
+	if (n > 0)
+		return line;
+	free(line);
+	return NULL;
+}
+
+/*
+ * Check whether we can use a group for top down.
+ * Without a group may get bad results.
+ */
+static bool check_group(bool *warn)
+{
+	char *v = sysctl_read("/proc/sys/kernel/nmi_watchdog");
+	int n;
+
+	*warn = false;
+	if (v) {
+		bool res = sscanf(v, "%d", &n) == 1 && n != 0;
+		free(v);
+		if (res) {
+			*warn = true;
+			return false;
+		}
+		return true;
+	}
+	return false; /* Don't know, so don't use group */
+}
+
 /*
  * Add default attributes, if there were no attributes specified or
  * if -d/--detailed, -d -d or -d -d -d is used:
  */
 static int add_default_attributes(void)
 {
+	int err;
 	struct perf_event_attr default_attrs[] = {
 
   { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK		},
@@ -1211,7 +1300,6 @@ static int add_default_attributes(void)
 		return 0;
 
 	if (transaction_run) {
-		int err;
 		if (pmu_have_event("cpu", "cycles-ct") &&
 		    pmu_have_event("cpu", "el-start"))
 			err = parse_events(evsel_list, transaction_attrs, NULL);
@@ -1224,6 +1312,36 @@ static int add_default_attributes(void)
 		return 0;
 	}
 
+	if (topdown_run) {
+		char *str = NULL;
+		bool warn;
+
+		filter_events(topdown_attrs, &str, check_group(&warn));
+		if (topdown_attrs[0] && str) {
+			if (warn)
+				fprintf(stderr,
+		"nmi_watchdog enabled with topdown. May give wrong results.\n"
+		"Disable with echo 0 > /proc/sys/kernel/nmi_watchdog\n");
+			err = parse_events(evsel_list, str, NULL);
+			if (err) {
+				fprintf(stderr,
+					"Cannot set up top down events %s: %d\n",
+					str, err);
+				free(str);
+				return -1;
+			}
+		} else {
+			fprintf(stderr, "System does not support topdown\n");
+			return -1;
+		}
+		free(str);
+		/*
+		 * Right now combining with the other attributes breaks group
+		 * semantics.
+		 */
+		return 0;
+	}
+
 	if (!evsel_list->nr_entries) {
 		if (perf_evlist__add_default_attrs(evsel_list, default_attrs) < 0)
 			return -1;
@@ -1260,6 +1378,8 @@ int cmd_stat(int argc, const char **argv, const char *prefix __maybe_unused)
 	const struct option options[] = {
 	OPT_BOOLEAN('T', "transaction", &transaction_run,
 		    "hardware transaction statistics"),
+	OPT_INCR(0, "topdown", &topdown_run,
+		    "measure topdown level 1 statistics"),
 	OPT_CALLBACK('e', "event", &evsel_list, "event",
 		     "event selector. use 'perf list' to list available events",
 		     parse_events_option),
diff --git a/tools/perf/util/parse-events.l b/tools/perf/util/parse-events.l
index f542750..a3b9903 100644
--- a/tools/perf/util/parse-events.l
+++ b/tools/perf/util/parse-events.l
@@ -239,6 +239,7 @@ cycles-ct					{ return str(yyscanner, PE_KERNEL_PMU_EVENT); }
 cycles-t					{ return str(yyscanner, PE_KERNEL_PMU_EVENT); }
 mem-loads					{ return str(yyscanner, PE_KERNEL_PMU_EVENT); }
 mem-stores					{ return str(yyscanner, PE_KERNEL_PMU_EVENT); }
+topdown-[a-z-]+					{ return str(yyscanner, PE_KERNEL_PMU_EVENT); }
 
 L1-dcache|l1-d|l1d|L1-data		|
 L1-icache|l1-i|l1i|L1-instruction	|
-- 
2.4.3


  parent reply	other threads:[~2015-08-08  1:06 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-08-08  1:06 Add top down metrics to " Andi Kleen
2015-08-08  1:06 ` [PATCH 1/9] perf, tools: Dont stop PMU parsing on alias parse error Andi Kleen
2015-08-11 13:07   ` Jiri Olsa
2015-08-11 13:14     ` Andi Kleen
2015-08-11 13:24       ` Jiri Olsa
2015-08-11 13:40         ` Andi Kleen
2015-08-11 14:39           ` Jiri Olsa
2015-08-11 16:59             ` Andi Kleen
2015-08-08  1:06 ` [PATCH 2/9] perf, tools, stat: Support up-scaling of events Andi Kleen
2015-08-11 13:25   ` Jiri Olsa
2015-08-11 13:38     ` Andi Kleen
2015-08-11 13:54       ` Jiri Olsa
2015-08-11 17:00         ` Andi Kleen
2015-08-11 17:13           ` Jiri Olsa
2015-08-11 17:17             ` Andi Kleen
2015-08-08  1:06 ` Andi Kleen [this message]
2015-08-08  1:06 ` [PATCH 4/9] perf, tools, stat: Add computation of TopDown formulas Andi Kleen
2015-08-08  1:06 ` [PATCH 5/9] x86, perf: Support sysfs files depending on SMT status Andi Kleen
2015-08-08  1:06 ` [PATCH 6/9] x86, perf: Add Top Down events to Intel Core Andi Kleen
2015-08-08  1:06 ` [PATCH 7/9] x86, perf: Add Top Down events to Intel Atom Andi Kleen
2015-08-08  1:06 ` [PATCH 8/9] perf, tools, stat: Add extra output of counter values with -v Andi Kleen
2015-08-08  1:06 ` [PATCH 9/9] perf, tools, stat: Force --per-core mode for .agg-per-core aliases Andi Kleen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1438995985-13631-4-git-send-email-andi@firstfloor.org \
    --to=andi@firstfloor.org \
    --cc=acme@kernel.org \
    --cc=ak@linux.intel.com \
    --cc=eranian@google.com \
    --cc=jolsa@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=namhyung@kernel.org \
    --cc=peterz@infradead.org \
    --subject='Re: [PATCH 3/9] perf, tools, stat: Basic support for TopDown in perf stat' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).