perf stat: Basic support for TopDown in perf stat

tools/perf/util/stat-shadow.c static void print_ll_cache_misses(int cpu, · line 424

Quoted 10 lines · Andi Kleen

tools/perf/util/stat-shadow.c +10 -0

+static double td_be_bound(int ctx, int cpu)
+{
+       double sum = (td_fe_bound(ctx, cpu) +
+                     td_bad_spec(ctx, cpu) +
+                     td_retiring(ctx, cpu));
+       if (sum == 0)
+               return 0;
+       return sanitize_val(1.0 - sum);
+}
+

Can you explain why we need the check on sum?

Quoted 13 lines · Nilay Vaish

tools/perf/util/stat-shadow.c static void print_ll_cache_misses(int cpu, · line 424

Quoted 10 lines · Andi Kleen

tools/perf/util/stat-shadow.c +10 -0

+static double td_be_bound(int ctx, int cpu)
+{
+       double sum = (td_fe_bound(ctx, cpu) +
+                     td_bad_spec(ctx, cpu) +
+                     td_retiring(ctx, cpu));
+       if (sum == 0)
+               return 0;
+       return sanitize_val(1.0 - sum);
+}
+

Can you explain why we need the check on sum?

You mean the if statement?

Otherwise if nothing was measured it would always report everything backend bound, which wouldn't be correct.

-Andi

--
[email protected] -- Speaking for myself only

Andi, I am talking about the if statement. I don't know why it would happen that nothing got measured. I am guessing you saw it happen. May be we can add a comment in the patch that it is possible that all counter values are zero and therefore we need that if statement.

From: Andi Kleen <[email protected]>

Add basic plumbing for TopDown in perf stat

TopDown is intended to replace the frontend cycles idle/ backend cycles idle metrics in standard perf stat output. These metrics are not reliable in many workloads, due to out of order effects.

This implements a new --topdown mode in perf stat (similar to --transaction) that measures the pipe line bottlenecks using standardized formulas. The measurement can be all done with 5 counters (one fixed counter)

The result are four metrics:
FrontendBound, BackendBound, BadSpeculation, Retiring

that describe the CPU pipeline behavior on a high level.

FrontendBound and BackendBound
BadSpeculation is a higher

The full top down methology has many hierarchical metrics. This implementation only supports level 1 which can be collected without multiplexing. A full implementation of top down on top of perf is available in pmu-tools toplev.
(http://github.com/andikleen/pmu-tools)

The current version works on Intel Core CPUs starting with Sandy Bridge, and Atom CPUs starting with Silvermont. In principle the generic metrics should be also implementable on other out of order CPUs.

TopDown level 1 uses a set of abstracted metrics which are generic to out of order CPU cores (although some CPUs may not implement all of them):

topdown-total-slots Available slots in the pipeline

topdown-slots-issued          Slots issued into the pipeline
topdown-slots-retired         Slots successfully retired
topdown-fetch-bubbles         Pipeline gaps in the frontend

topdown-recovery-bubbles Pipeline gaps during recovery
from misspeculation

These metrics then allow to compute four useful metrics:
FrontendBound, BackendBound, Retiring, BadSpeculation.

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.

v2: Use standard sysctl read function.
v3: Move x86 specific code to arch/
v4: Enable --metric-only implicitly for topdown.
v5: Add --single-thread option to not force per core mode
v6: Fix output order of topdown metrics
v7: Allow combining with -d
v8: Remove --single-thread again
v9: Rename functions, adding arch_ and topdown_.
v10: Expand man page and describe TopDown better Paste intro into commit description.
Print error when malloc fails.

Acked-by Jiri Olsa <[email protected]>

Signed-off-by Andi Kleen <[email protected]>

---
 tools/perf/Documentation/perf-stat.txt |  32 +++++++++
 tools/perf/arch/x86/util/Build         |   1 +
 tools/perf/arch/x86/util/group.c       |  27 ++++++++
 tools/perf/builtin-stat.c              | 119 ++++++++++++++++++++++++++++++++-
 tools/perf/util/group.h                |   7 ++
 tools/perf/util/parse-events.l         |   1 +
 6 files changed, 184 insertions(+), 3 deletions(-)
 create mode 100644 tools/perf/arch/x86/util/group.c
 create mode 100644 tools/perf/util/group.h

tools/perf/Documentation/perf-stat.txt +32 -0

index 04f23b404bbc..d96ccd4844df 100644
@@ -204,6 +204,38 @@ Aggregate counts per physical processor for system-wide mode measurements.
 --no-aggr::
 Do not aggregate counts across all monitored CPUs.
 
+--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 the cycles consumed down into frontend bound, backend bound,
+bad speculation and retiring.
+
+Frontend bound means that the CPU cannot fetch and decode instructions fast
+enough. Backend bound means that computation or memory access is the bottle
+neck. Bad Speculation means that the CPU wasted cycles due to branch
+mispredictions and similar issues. Retiring means that the CPU computed without
+an apparently bottleneck. The bottleneck is only the real bottleneck
+if the workload is actually bound by the CPU and not by something else.
+
+For best results it is usually a good idea to use it with interval
+mode like -I 1000, as the bottleneck of workloads can change often.
+
+The top down metrics are collected per core instead of per
+CPU thread. Per core mode is automatically enabled
+and -a (global monitoring) is needed, requiring root rights or
+perf.perf_event_paranoid=-1.
+
+Topdown uses the full Performance Monitoring Unit, and needs
+disabling of the NMI watchdog (as root):
+echo 0 > /proc/sys/kernel/nmi_watchdog
+for best results. Otherwise the bottlenecks may be inconsistent
+on workload with changing phases.
+
+This enables --metric-only, unless overriden with --no-metric-only.
+
+To interpret the results it is usually needed to know on which
+CPUs the workload runs on. If needed the CPUs can be forced using
+taskset.
 
 EXAMPLES
 --------

tools/perf/arch/x86/util/Build +1 -0

index 465970370f3e..4cd8a16b1b7b 100644
@@ -3,6 +3,7 @@ libperf-y += tsc.o
 libperf-y += pmu.o
 libperf-y += kvm-stat.o
 libperf-y += perf_regs.o
+libperf-y += group.o
 
 libperf-$(CONFIG_DWARF) += dwarf-regs.o
 libperf-$(CONFIG_BPF_PROLOGUE) += dwarf-regs.o

tools/perf/arch/x86/util/group.c new file +27 -0

new file mode 100644
index 000000000000..37f92aa39a5d
@@ -0,0 +1,27 @@
+#include <stdio.h>
+#include "api/fs/fs.h"
+#include "util/group.h"
+
+/*
+ * Check whether we can use a group for top down.
+ * Without a group may get bad results due to multiplexing.
+ */
+bool arch_topdown_check_group(bool *warn)
+{
+	int n;
+
+	if (sysctl__read_int("kernel/nmi_watchdog", &n) < 0)
+		return false;
+	if (n > 0) {
+		*warn = true;
+		return false;
+	}
+	return true;
+}
+
+void arch_topdown_group_warn(void)
+{
+	fprintf(stderr,
+		"nmi_watchdog enabled with topdown. May give wrong results.\n"
+		"Disable with echo 0 > /proc/sys/kernel/nmi_watchdog\n");
+}

tools/perf/builtin-stat.c +116 -3

index 715a1128daeb..dab184d86816 100644
@@ -59,10 +59,13 @@
 #include "util/thread.h"
 #include "util/thread_map.h"
 #include "util/counts.h"
+#include "util/group.h"
 #include "util/session.h"
 #include "util/tool.h"
+#include "util/group.h"
 #include "asm/bug.h"
 
+#include <api/fs/fs.h>
 #include <stdlib.h>
 #include <sys/prctl.h>
 #include <locale.h>

@@ -98,6 +101,15 @@ static const char * transaction_limited_attrs = {
 	"}"
 };
 
+static const char * topdown_attrs[] = {
+	"topdown-total-slots",
+	"topdown-slots-retired",
+	"topdown-recovery-bubbles",
+	"topdown-fetch-bubbles",
+	"topdown-slots-issued",
+	NULL,
+};
+
 static struct perf_evlist	*evsel_list;
 
 static struct target target = {

@@ -112,6 +124,7 @@ static volatile pid_t		child_pid			= -1;
 static bool			null_run			=  false;
 static int			detailed_run			=  0;
 static bool			transaction_run;
+static bool			topdown_run			= false;
 static bool			big_num				=  true;
 static int			big_num_opt			=  -1;
 static const char		*csv_sep			= NULL;

@@ -124,6 +137,7 @@ static unsigned int		initial_delay			= 0;
 static unsigned int		unit_width			= 4; /* strlen("unit") */
 static bool			forever				= false;
 static bool			metric_only			= false;
+static bool			force_metric_only		= false;
 static struct timespec		ref_time;
 static struct cpu_map		*aggr_map;
 static aggr_get_id_t		aggr_get_id;

@@ -1520,6 +1534,14 @@ static int stat__set_big_num(const struct option *opt __maybe_unused,
 	return 0;
 }
 
+static int enable_metric_only(const struct option *opt __maybe_unused,
+			      const char *s __maybe_unused, int unset)
+{
+	force_metric_only = true;
+	metric_only = !unset;
+	return 0;
+}
+
 static const struct option stat_options[] = {
 	OPT_BOOLEAN('T', "transaction", &transaction_run,
 		    "hardware transaction statistics"),

@@ -1578,8 +1600,10 @@ static const struct option stat_options[] = {
 		     "aggregate counts per thread", AGGR_THREAD),
 	OPT_UINTEGER('D', "delay", &initial_delay,
 		     "ms to wait before starting measurement after program start"),
-	OPT_BOOLEAN(0, "metric-only", &metric_only,
-			"Only print computed metrics. No raw values"),
+	OPT_CALLBACK_NOOPT(0, "metric-only", &metric_only, NULL,
+			"Only print computed metrics. No raw values", enable_metric_only),
+	OPT_BOOLEAN(0, "topdown", &topdown_run,
+			"measure topdown level 1 statistics"),
 	OPT_END()
 };
 

@@ -1772,12 +1796,62 @@ static int perf_stat_init_aggr_mode_file(struct perf_stat *st)
 	return 0;
 }
 
+static int topdown_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 -1;
+	s = *str;
+	if (i - off == 0) {
+		*s = 0;
+		return 0;
+	}
+	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;
+	return 0;
+}
+
+__weak bool arch_topdown_check_group(bool *warn)
+{
+	*warn = false;
+	return false;
+}
+
+__weak void arch_topdown_group_warn(void)
+{
+}
+
 /*
  * 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_attrs0[] = {
 
   { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK		},

@@ -1896,7 +1970,6 @@ 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);

@@ -1909,6 +1982,46 @@ static int add_default_attributes(void)
 		return 0;
 	}
 
+	if (topdown_run) {
+		char *str = NULL;
+		bool warn = false;
+
+		if (stat_config.aggr_mode != AGGR_GLOBAL &&
+		    stat_config.aggr_mode != AGGR_CORE) {
+			pr_err("top down event configuration requires --per-core mode\n");
+			return -1;
+		}
+		stat_config.aggr_mode = AGGR_CORE;
+		if (nr_cgroups || !target__has_cpu(&target)) {
+			pr_err("top down event configuration requires system-wide mode (-a)\n");
+			return -1;
+		}
+
+		if (!force_metric_only)
+			metric_only = true;
+		if (topdown_filter_events(topdown_attrs, &str,
+				arch_topdown_check_group(&warn)) < 0) {
+			pr_err("Out of memory\n");
+			return -1;
+		}
+		if (topdown_attrs[0] && str) {
+			if (warn)
+				arch_topdown_group_warn();
+			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);
+	}
+
 	if (!evsel_list->nr_entries) {
 		if (perf_evlist__add_default_attrs(evsel_list, default_attrs0) < 0)
 			return -1;

tools/perf/util/group.h new file +7 -0

new file mode 100644
index 000000000000..116debe7a995
@@ -0,0 +1,7 @@
+#ifndef GROUP_H
+#define GROUP_H 1
+
+bool arch_topdown_check_group(bool *warn);
+void arch_topdown_group_warn(void);
+
+#endif

tools/perf/util/parse-events.l +1 -0

index 1477fbc78993..744ebe3fa30f 100644
@@ -259,6 +259,7 @@ 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	|

Signature · 1 line

2.5.5

Which is a Broadwell-U, dual-core, 14 nm.

[root@jouet linux]# echo 0 > /proc/sys/kernel/nmi_watchdog [root@jouet linux]# perf stat --topdown -a usleep 1 System does not support topdown
[root@jouet linux]# cat /proc/sys/kernel/nmi_watchdog 0
[root@jouet linux]#

Please advise.

Quoted 9 lines

Which is a Broadwell-U, dual-core, 14 nm.

[root@jouet linux]# echo 0 > /proc/sys/kernel/nmi_watchdog [root@jouet linux]# perf stat --topdown -a usleep 1 System does not support topdown
[root@jouet linux]# cat /proc/sys/kernel/nmi_watchdog 0
[root@jouet linux]#

Quoted 1 line

Please advise.

You don't have the kernel patches?

They're in PeterZ's tree, but not yet in tip
https://git.kernel.org/cgit/linux/kernel/git/peterz/queue.git/log/?h=perf/core

Em Mon, May 30, 2016 at 09:04:02AM -0700, Andi Kleen escreveu:

Quoted 9 lines

Quoted 9 lines

Which is a Broadwell-U, dual-core, 14 nm.

[root@jouet linux]# echo 0 > /proc/sys/kernel/nmi_watchdog [root@jouet linux]# perf stat --topdown -a usleep 1 System does not support topdown
[root@jouet linux]# cat /proc/sys/kernel/nmi_watchdog 0
[root@jouet linux]#

Quoted 6 lines

Quoted 1 line

Please advise.

You don't have the kernel patches?

They're in PeterZ's tree, but not yet in tip
https://git.kernel.org/cgit/linux/kernel/git/peterz/queue.git/log/?h=perf/core

oops, will try applying those patches, thanks for claryfing.

The kernel patches are in tip/perf/core now, so I applied the patches and they seem to work, but they broke one 'perf test' entry:

# perf test -v 5
<SNIP>
running test 50 '4:0x6530160/name=numpmu/'
running test 51 'L1-dcache-misses/name=cachepmu/'
running test 0 'cpu/config=10,config1,config2=3,period=1000/u'
running test 1 'cpu/config=1,name=krava/u,cpu/config=2/u'
running test 2 'cpu/config=1,call-graph=fp,time,period=100000/,cpu/config=2,call-graph=no,time=0,period=2000/'
Invalid sysfs entry event=topdown-recovery-bubbles.scale
failed to parse event 'cpu/event=topdown-recovery-bubbles.scale/u', err 1
test child finished with 1
---- end ----
parse events tests: FAILED!
#

Em Mon, Jun 06, 2016 at 11:00:03AM -0300, Arnaldo Carvalho de Melo escreveu:

Quoted 16 lines

The kernel patches are in tip/perf/core now, so I applied the patches and they seem to work, but they broke one 'perf test' entry:

# perf test -v 5
<SNIP>
running test 50 '4:0x6530160/name=numpmu/'
running test 51 'L1-dcache-misses/name=cachepmu/'
running test 0 'cpu/config=10,config1,config2=3,period=1000/u'
running test 1 'cpu/config=1,name=krava/u,cpu/config=2/u'
running test 2 'cpu/config=1,call-graph=fp,time,period=100000/,cpu/config=2,call-graph=no,time=0,period=2000/'
Invalid sysfs entry event=topdown-recovery-bubbles.scale
failed to parse event 'cpu/event=topdown-recovery-bubbles.scale/u', err 1
test child finished with 1
---- end ----
parse events tests: FAILED!
#

Ok, the test failure is not due to this patchkit, the reason are the new topdown*.scale files in the kernel, was there a patch to handle that that is missing in my branch?

From: Andi Kleen <[email protected]>

tools/perf/builtin-stat.c +0 -0

index 715a1128daeb..dab184d86816 100644

tools/perf/builtin-stat.c +14 -0

+static int topdown_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++;
+       }

tools/perf/builtin-stat.c +30 -0

@@ -1909,6 +1982,46 @@ static int add_default_attributes(void)
                return 0;
        }
 
+       if (topdown_run) {
+               char *str = NULL;
+               bool warn = false;
+
+               if (stat_config.aggr_mode != AGGR_GLOBAL &&
+                   stat_config.aggr_mode != AGGR_CORE) {
+                       pr_err("top down event configuration requires --per-core mode\n");
+                       return -1;
+               }
+               stat_config.aggr_mode = AGGR_CORE;
+               if (nr_cgroups || !target__has_cpu(&target)) {
+                       pr_err("top down event configuration requires system-wide mode (-a)\n");
+                       return -1;
+               }
+
+               if (!force_metric_only)
+                       metric_only = true;
+               if (topdown_filter_events(topdown_attrs, &str,
+                               arch_topdown_check_group(&warn)) < 0) {
+                       pr_err("Out of memory\n");
+                       return -1;
+               }
+               if (topdown_attrs[0] && str) {
+                       if (warn)
+                               arch_topdown_group_warn();
+                       err = parse_events(evsel_list, str, NULL);
+                       if (err) {
+                               fprintf(stderr,
+                                       "Cannot set up top down events %s: %d\n",
+                                       str, err);

Quoted 7 lines · Andi Kleen

From: Andi Kleen <[email protected]>

tools/perf/builtin-stat.c +0 -0

index 715a1128daeb..dab184d86816 100644

tools/perf/builtin-stat.c line 66

Quoted 8 lines

tools/perf/builtin-stat.c +2 -0

@@ -59,10 +59,13 @@
 #include "util/thread.h"
 #include "util/thread_map.h"
 #include "util/counts.h"
+#include "util/group.h"
 #include "util/session.h"
 #include "util/tool.h"
+#include "util/group.h"
 #include "asm/bug.h"

You have included util/group.h twice. Is this intentional?

Continuing with my comment from above memory leak, if i == off, topdown_attrs[0] will be NULL. So we would enter the else portion here and return -1. But we never free the string we allocated in the
function topdown_filter_events(). I think we are leaking some memory
though seems only about 3 bytes.

Here's an updated patch addresses the duplicated issue and explicitly frees memory.

---

Add basic plumbing for TopDown in perf stat

TopDown is intended to replace the frontend cycles idle/ backend cycles idle metrics in standard perf stat output. These metrics are not reliable in many workloads, due to out of order effects.

This implements a new --topdown mode in perf stat (similar to --transaction) that measures the pipe line bottlenecks using standardized formulas. The measurement can be all done with 5 counters (one fixed counter)

The result are four metrics:
FrontendBound, BackendBound, BadSpeculation, Retiring

that describe the CPU pipeline behavior on a high level.

FrontendBound and BackendBound
BadSpeculation is a higher

The full top down methology has many hierarchical metrics. This implementation only supports level 1 which can be collected without multiplexing. A full implementation of top down on top of perf is available in pmu-tools toplev.
(http://github.com/andikleen/pmu-tools)

The current version works on Intel Core CPUs starting with Sandy Bridge, and Atom CPUs starting with Silvermont. In principle the generic metrics should be also implementable on other out of order CPUs.

TopDown level 1 uses a set of abstracted metrics which are generic to out of order CPU cores (although some CPUs may not implement all of them):

topdown-total-slots Available slots in the pipeline

topdown-slots-issued          Slots issued into the pipeline
topdown-slots-retired         Slots successfully retired
topdown-fetch-bubbles         Pipeline gaps in the frontend

topdown-recovery-bubbles Pipeline gaps during recovery
from misspeculation

These metrics then allow to compute four useful metrics:
FrontendBound, BackendBound, Retiring, BadSpeculation.

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.

v2: Use standard sysctl read function.
v3: Move x86 specific code to arch/
v4: Enable --metric-only implicitly for topdown.
v5: Add --single-thread option to not force per core mode
v6: Fix output order of topdown metrics
v7: Allow combining with -d
v8: Remove --single-thread again
v9: Rename functions, adding arch_ and topdown_.
v10: Expand man page and describe TopDown better Paste intro into commit description.
Print error when malloc fails.
v11:
Free memory before exit
Remove duplicate include.

Acked-by Jiri Olsa <[email protected]>

Signed-off-by Andi Kleen <[email protected]>

---
 tools/perf/Documentation/perf-stat.txt |  32 +++++++++
 tools/perf/arch/x86/util/Build         |   1 +
 tools/perf/arch/x86/util/group.c       |  27 ++++++++
 tools/perf/builtin-stat.c              | 119 ++++++++++++++++++++++++++++++++-
 tools/perf/util/group.h                |   7 ++
 tools/perf/util/parse-events.l         |   1 +
 6 files changed, 184 insertions(+), 3 deletions(-)
 create mode 100644 tools/perf/arch/x86/util/group.c
 create mode 100644 tools/perf/util/group.h

tools/perf/Documentation/perf-stat.txt +32 -0

index 04f23b404bbc..d96ccd4844df 100644
@@ -204,6 +204,38 @@ Aggregate counts per physical processor for system-wide mode measurements.
 --no-aggr::
 Do not aggregate counts across all monitored CPUs.
 
+--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 the cycles consumed down into frontend bound, backend bound,
+bad speculation and retiring.
+
+Frontend bound means that the CPU cannot fetch and decode instructions fast
+enough. Backend bound means that computation or memory access is the bottle
+neck. Bad Speculation means that the CPU wasted cycles due to branch
+mispredictions and similar issues. Retiring means that the CPU computed without
+an apparently bottleneck. The bottleneck is only the real bottleneck
+if the workload is actually bound by the CPU and not by something else.
+
+For best results it is usually a good idea to use it with interval
+mode like -I 1000, as the bottleneck of workloads can change often.
+
+The top down metrics are collected per core instead of per
+CPU thread. Per core mode is automatically enabled
+and -a (global monitoring) is needed, requiring root rights or
+perf.perf_event_paranoid=-1.
+
+Topdown uses the full Performance Monitoring Unit, and needs
+disabling of the NMI watchdog (as root):
+echo 0 > /proc/sys/kernel/nmi_watchdog
+for best results. Otherwise the bottlenecks may be inconsistent
+on workload with changing phases.
+
+This enables --metric-only, unless overriden with --no-metric-only.
+
+To interpret the results it is usually needed to know on which
+CPUs the workload runs on. If needed the CPUs can be forced using
+taskset.
 
 EXAMPLES
 --------

tools/perf/arch/x86/util/Build +1 -0

index 465970370f3e..4cd8a16b1b7b 100644
@@ -3,6 +3,7 @@ libperf-y += tsc.o
 libperf-y += pmu.o
 libperf-y += kvm-stat.o
 libperf-y += perf_regs.o
+libperf-y += group.o
 
 libperf-$(CONFIG_DWARF) += dwarf-regs.o
 libperf-$(CONFIG_BPF_PROLOGUE) += dwarf-regs.o

tools/perf/arch/x86/util/group.c new file +27 -0

new file mode 100644
index 000000000000..37f92aa39a5d
@@ -0,0 +1,27 @@
+#include <stdio.h>
+#include "api/fs/fs.h"
+#include "util/group.h"
+
+/*
+ * Check whether we can use a group for top down.
+ * Without a group may get bad results due to multiplexing.
+ */
+bool arch_topdown_check_group(bool *warn)
+{
+       int n;
+
+       if (sysctl__read_int("kernel/nmi_watchdog", &n) < 0)
+               return false;
+       if (n > 0) {
+               *warn = true;
+               return false;
+       }
+       return true;
+}
+
+void arch_topdown_group_warn(void)
+{
+       fprintf(stderr,
+               "nmi_watchdog enabled with topdown. May give wrong results.\n"
+               "Disable with echo 0 > /proc/sys/kernel/nmi_watchdog\n");
+}

tools/perf/builtin-stat.c +116 -3

index 715a1128daeb..0e7d65969028 100644
@@ -59,10 +59,12 @@
 #include "util/thread.h"
 #include "util/thread_map.h"
 #include "util/counts.h"
+#include "util/group.h"
 #include "util/session.h"
 #include "util/tool.h"
 #include "asm/bug.h"
 
+#include <api/fs/fs.h>
 #include <stdlib.h>
 #include <sys/prctl.h>
 #include <locale.h>

@@ -98,6 +100,15 @@ static const char * transaction_limited_attrs = {
        "}"
 };
 
+static const char * topdown_attrs[] = {
+       "topdown-total-slots",
+       "topdown-slots-retired",
+       "topdown-recovery-bubbles",
+       "topdown-fetch-bubbles",
+       "topdown-slots-issued",
+       NULL,
+};
+
 static struct perf_evlist      *evsel_list;
 
 static struct target target = {

@@ -112,6 +123,7 @@ static volatile pid_t               child_pid                       = -1;
 static bool                    null_run                        =  false;
 static int                     detailed_run                    =  0;
 static bool                    transaction_run;
+static bool                    topdown_run                     = false;
 static bool                    big_num                         =  true;
 static int                     big_num_opt                     =  -1;
 static const char              *csv_sep                        = NULL;

@@ -124,6 +136,7 @@ static unsigned int         initial_delay                   = 0;
 static unsigned int            unit_width                      = 4; /* strlen("unit") */
 static bool                    forever                         = false;
 static bool                    metric_only                     = false;
+static bool                    force_metric_only               = false;
 static struct timespec         ref_time;
 static struct cpu_map          *aggr_map;
 static aggr_get_id_t           aggr_get_id;

@@ -1520,6 +1533,14 @@ static int stat__set_big_num(const struct option *opt __maybe_unused,
        return 0;
 }
 
+static int enable_metric_only(const struct option *opt __maybe_unused,
+                             const char *s __maybe_unused, int unset)
+{
+       force_metric_only = true;
+       metric_only = !unset;
+       return 0;
+}
+
 static const struct option stat_options[] = {
        OPT_BOOLEAN('T', "transaction", &transaction_run,
                    "hardware transaction statistics"),

@@ -1578,8 +1599,10 @@ static const struct option stat_options[] = {
                     "aggregate counts per thread", AGGR_THREAD),
        OPT_UINTEGER('D', "delay", &initial_delay,
                     "ms to wait before starting measurement after program start"),
-       OPT_BOOLEAN(0, "metric-only", &metric_only,
-                       "Only print computed metrics. No raw values"),
+       OPT_CALLBACK_NOOPT(0, "metric-only", &metric_only, NULL,
+                       "Only print computed metrics. No raw values", enable_metric_only),
+       OPT_BOOLEAN(0, "topdown", &topdown_run,
+                       "measure topdown level 1 statistics"),
        OPT_END()
 };
 

@@ -1772,12 +1795,62 @@ static int perf_stat_init_aggr_mode_file(struct perf_stat *st)
        return 0;
 }
 
+static int topdown_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 -1;
+       s = *str;
+       if (i - off == 0) {
+               *s = 0;
+               return 0;
+       }
+       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;
+       return 0;
+}
+
+__weak bool arch_topdown_check_group(bool *warn)
+{
+       *warn = false;
+       return false;
+}
+
+__weak void arch_topdown_group_warn(void)
+{
+}
+
 /*
  * 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_attrs0[] = {
 
   { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK             },

@@ -1896,7 +1969,6 @@ 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);

@@ -1909,6 +1981,47 @@ static int add_default_attributes(void)
                return 0;
        }
 
+       if (topdown_run) {
+               char *str = NULL;
+               bool warn = false;
+
+               if (stat_config.aggr_mode != AGGR_GLOBAL &&
+                   stat_config.aggr_mode != AGGR_CORE) {
+                       pr_err("top down event configuration requires --per-core mode\n");
+                       return -1;
+               }
+               stat_config.aggr_mode = AGGR_CORE;
+               if (nr_cgroups || !target__has_cpu(&target)) {
+                       pr_err("top down event configuration requires system-wide mode (-a)\n");
+                       return -1;
+               }
+
+               if (!force_metric_only)
+                       metric_only = true;
+               if (topdown_filter_events(topdown_attrs, &str,
+                               arch_topdown_check_group(&warn)) < 0) {
+                       pr_err("Out of memory\n");
+                       return -1;
+               }
+               if (topdown_attrs[0] && str) {
+                       if (warn)
+                               arch_topdown_group_warn();
+                       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");
+                       free(str);
+                       return -1;
+               }
+               free(str);
+       }
+
        if (!evsel_list->nr_entries) {
                if (perf_evlist__add_default_attrs(evsel_list, default_attrs0) < 0)
                        return -1;

tools/perf/util/group.h new file +7 -0

new file mode 100644
index 000000000000..116debe7a995
@@ -0,0 +1,7 @@
+#ifndef GROUP_H
+#define GROUP_H 1
+
+bool arch_topdown_check_group(bool *warn);
+void arch_topdown_group_warn(void);
+
+#endif

tools/perf/util/parse-events.l +1 -1

index 1477fbc78993..744ebe3fa30f 100644
@@ -259,6 +259,7 @@ 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      |
--

2.5.5