Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit aa3a655b authored by Ingo Molnar's avatar Ingo Molnar
Browse files

Merge tag 'perf-core-for-mingo-20160606' of... 

Merge tag 'perf-core-for-mingo-20160606' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

 into perf/core

Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:

User visible changes:

- Tooling support for TopDown counters, recently added to the kernel (Andi Kleen)

- Show call graphs in 'perf script' when 1st event doesn't have it but some other has (He Kuang)

- Fix terminal cleanup when handling invalid .perfconfig files in 'perf top' (Taeung Song)

Build fixes:

- Respect CROSS_COMPILE for the linker in libapi (Lucas Stach)

Infrastructure changes:

- Fix perf_evlist__alloc_mmap() failure path (Wang Nan)

- Provide way to extract integer value from format_field (Arnaldo Carvalho de Melo)

Signed-off-by: default avatarArnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: default avatarIngo Molnar <mingo@kernel.org>
parents 616d1c1b 7db91f25

tools/lib/api/Makefile

+1 −0
Original line number Diff line number Diff line
@@ -10,6 +10,7 @@ endif 


CC = $(CROSS_COMPILE)gcc

AR = $(CROSS_COMPILE)ar

LD = $(CROSS_COMPILE)ld



MAKEFLAGS += --no-print-directory

tools/perf/Documentation/perf-stat.txt

+32 −0
Original line number Diff line number Diff line
@@ -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
Original line number Diff line number Diff line
@@ -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

0 → 100644
+27 −0
Original line number Diff line number Diff line 
#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-script.c

+13 −10
Original line number Diff line number Diff line
@@ -339,7 +339,7 @@ static void set_print_ip_opts(struct perf_event_attr *attr) 
 */

static int perf_session__check_output_opt(struct perf_session *session)

{

	int j;

	unsigned int j;

	struct perf_evsel *evsel;



	for (j = 0; j < PERF_TYPE_MAX; ++j) {
@@ -388,9 +388,10 @@ static int perf_session__check_output_opt(struct perf_session *session) 
		struct perf_event_attr *attr;



		j = PERF_TYPE_TRACEPOINT;

		evsel = perf_session__find_first_evtype(session, j);

		if (evsel == NULL)

			goto out;



		evlist__for_each(session->evlist, evsel) {

			if (evsel->attr.type != j)

				continue;



			attr = &evsel->attr;
@@ -399,6 +400,8 @@ static int perf_session__check_output_opt(struct perf_session *session) 
				output[j].fields |= PERF_OUTPUT_SYM;

				output[j].fields |= PERF_OUTPUT_DSO;

				set_print_ip_opts(attr);

				goto out;

			}

		}

	}