[PATCH 3/9] perf, tools, stat: Basic support for TopDown in perf stat
From: Andi Kleen
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
---
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 boolnull_run= false;
static int detailed_run= 0;
static booltransaction_run;
+static int topdown_run = 0;
static boolbig_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,
- );
+ ,
+ 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(, , 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", ) == 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:
[PATCH 3/9] perf, tools, stat: Basic support for TopDown in perf stat
From: Andi Kleen a...@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 a...@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 boolnull_run= false;
static int detailed_run= 0;
static booltransaction_run;
+static int topdown_run = 0;
static boolbig_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
