Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull perf changes from Ingo Molnar:
 "Kernel side changes:

   - Consolidate the PMU interrupt-disabled code amongst architectures
     (Vince Weaver)

   - misc fixes

  Tooling changes (new features, user visible changes):

   - Add support for pagefault tracing in 'trace', please see multiple
     examples in the changeset messages (Stanislav Fomichev).

   - Add pagefault statistics in 'trace' (Stanislav Fomichev)

   - Add header for columns in 'top' and 'report' TUI browsers (Jiri
     Olsa)

   - Add pagefault statistics in 'trace' (Stanislav Fomichev)

   - Add IO mode into timechart command (Stanislav Fomichev)

   - Fallback to syscalls:* when raw_syscalls:* is not available in the
     perl and python perf scripts.  (Daniel Bristot de Oliveira)

   - Add --repeat global option to 'perf bench' to be used in benchmarks
     such as the existing 'futex' one, that was modified to use it
     instead of a local option.  (Davidlohr Bueso)

   - Fix fd -> pathname resolution in 'trace', be it using /proc or a
     vfs_getname probe point.  (Arnaldo Carvalho de Melo)

   - Add suggestion of how to set perf_event_paranoid sysctl, to help
     non-root users trying tools like 'trace' to get a working
     environment.  (Arnaldo Carvalho de Melo)

   - Updates from trace-cmd for traceevent plugin_kvm plus args cleanup
     (Steven Rostedt, Jan Kiszka)

   - Support S/390 in 'perf kvm stat' (Alexander Yarygin)

  Tooling infrastructure changes:

   - Allow reserving a row for header purposes in the hists browser
     (Arnaldo Carvalho de Melo)

   - Various fixes and prep work related to supporting Intel PT (Adrian
     Hunter)

   - Introduce multiple debug variables control (Jiri Olsa)

   - Add callchain and additional sample information for python scripts
     (Joseph Schuchart)

   - More prep work to support Intel PT: (Adrian Hunter)
     - Polishing 'script' BTS output
     - 'inject' can specify --kallsym
     - VDSO is per machine, not a global var
     - Expose data addr lookup functions previously private to 'script'
     - Large mmap fixes in events processing

   - Include standard stringify macros in power pc code (Sukadev
     Bhattiprolu)

  Tooling cleanups:

   - Convert open coded equivalents to asprintf() (Andy Shevchenko)

   - Remove needless reassignments in 'trace' (Arnaldo Carvalho de Melo)

   - Cache the is_exit syscall test in 'trace) (Arnaldo Carvalho de
     Melo)

   - No need to reimplement err() in 'perf bench sched-messaging', drop
     barf().  (Davidlohr Bueso).

   - Remove ev_name argument from perf_evsel__hists_browse, can be
     obtained from the other parameters.  (Jiri Olsa)

  Tooling fixes:

   - Fix memory leak in the 'sched-messaging' perf bench test.
     (Davidlohr Bueso)

   - The -o and -n 'perf bench mem' options are mutually exclusive, emit
     error when both are specified.  (Davidlohr Bueso)

   - Fix scrollbar refresh row index in the ui browser, problem exposed
     now that headers will be added and will be allowed to be switched
     on/off.  (Jiri Olsa)

   - Handle the num array type in python properly (Sebastian Andrzej
     Siewior)

   - Fix wrong condition for allocation failure (Jiri Olsa)

   - Adjust callchain based on DWARF debug info on powerpc (Sukadev
     Bhattiprolu)

   - Fix a risk for doing free on uninitialized pointer in traceevent
     lib (Rickard Strandqvist)

   - Update attr test with PERF_FLAG_FD_CLOEXEC flag (Jiri Olsa)

   - Enable close-on-exec flag on perf file descriptor (Yann Droneaud)

   - Fix build on gcc 4.4.7 (Arnaldo Carvalho de Melo)

   - Event ordering fixes (Jiri Olsa)"

* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (123 commits)
  Revert "perf tools: Fix jump label always changing during tracing"
  perf tools: Fix perf usage string leftover
  perf: Check permission only for parent tracepoint event
  perf record: Store PERF_RECORD_FINISHED_ROUND only for nonempty rounds
  perf record: Always force PERF_RECORD_FINISHED_ROUND event
  perf inject: Add --kallsyms parameter
  perf tools: Expose 'addr' functions so they can be reused
  perf session: Fix accounting of ordered samples queue
  perf powerpc: Include util/util.h and remove stringify macros
  perf tools: Fix build on gcc 4.4.7
  perf tools: Add thread parameter to vdso__dso_findnew()
  perf tools: Add dso__type()
  perf tools: Separate the VDSO map name from the VDSO dso name
  perf tools: Add vdso__new()
  perf machine: Fix the lifetime of the VDSO temporary file
  perf tools: Group VDSO global variables into a structure
  perf session: Add ability to skip 4GiB or more
  perf session: Add ability to 'skip' a non-piped event stream
  perf tools: Pass machine to vdso__dso_findnew()
  perf tools: Add dso__data_size()
  ...

arch/arc/kernel/perf_event.c

@@ -99,10 +99,6 @@ static int arc_pmu_event_init(struct perf_event *event)
 	struct hw_perf_event *hwc = &event->hw;
 	int ret;
 
-	/* ARC 700 PMU does not support sampling events */
-	if (is_sampling_event(event))
-		return -ENOENT;
-
 	switch (event->attr.type) {
 	case PERF_TYPE_HARDWARE:
 		if (event->attr.config >= PERF_COUNT_HW_MAX)
@@ -298,6 +294,9 @@ static int arc_pmu_device_probe(struct platform_device *pdev)
 		.read		= arc_pmu_read,
 	};
 
+	/* ARC 700 PMU does not support sampling events */
+	arc_pmu->pmu.capabilities |= PERF_PMU_CAP_NO_INTERRUPT;
+
 	ret = perf_pmu_register(&arc_pmu->pmu, pdev->name, PERF_TYPE_RAW);
 
 	return ret;

arch/blackfin/kernel/perf_event.c

@@ -389,14 +389,6 @@ static int bfin_pmu_event_init(struct perf_event *event)
 	if (attr->exclude_hv || attr->exclude_idle)
 		return -EPERM;
 
-	/*
-	 * All of the on-chip counters are "limited", in that they have
-	 * no interrupts, and are therefore unable to do sampling without
-	 * further work and timer assistance.
-	 */
-	if (hwc->sample_period)
-		return -EINVAL;
-
 	ret = 0;
 	switch (attr->type) {
 	case PERF_TYPE_RAW:
@@ -490,6 +482,13 @@ static int __init bfin_pmu_init(void)
 {
 	int ret;
 
+	/*
+	 * All of the on-chip counters are "limited", in that they have
+	 * no interrupts, and are therefore unable to do sampling without
+	 * further work and timer assistance.
+	 */
+	pmu.capabilities |= PERF_PMU_CAP_NO_INTERRUPT;
+
 	ret = perf_pmu_register(&pmu, "cpu", PERF_TYPE_RAW);
 	if (!ret)
 		perf_cpu_notifier(bfin_pmu_notifier);

arch/metag/kernel/perf/perf_event.c

@@ -567,16 +567,6 @@ static int _hw_perf_event_init(struct perf_event *event)
 	if (mapping == -1)
 		return -EINVAL;
 
-	/*
-	 * Early cores have "limited" counters - they have no overflow
-	 * interrupts - and so are unable to do sampling without extra work
-	 * and timer assistance.
-	 */
-	if (metag_pmu->max_period == 0) {
-		if (hwc->sample_period)
-			return -EINVAL;
-	}
-
 	/*
 	 * Don't assign an index until the event is placed into the hardware.
 	 * -1 signifies that we're still deciding where to put it. On SMP
@@ -866,6 +856,15 @@ static int __init init_hw_perf_events(void)
 	pr_info("enabled with %s PMU driver, %d counters available\n",
 			metag_pmu->name, metag_pmu->max_events);
 
+	/*
+	 * Early cores have "limited" counters - they have no overflow
+	 * interrupts - and so are unable to do sampling without extra work
+	 * and timer assistance.
+	 */
+	if (metag_pmu->max_period == 0) {
+		metag_pmu->pmu.capabilities |= PERF_PMU_CAP_NO_INTERRUPT;
+	}
+
 	/* Initialise the active events and reservation mutex */
 	atomic_set(&metag_pmu->active_events, 0);
 	mutex_init(&metag_pmu->reserve_mutex);

arch/powerpc/perf/hv-24x7.c

@@ -387,8 +387,7 @@ static int h_24x7_event_init(struct perf_event *event)
 	    event->attr.exclude_hv     ||
 	    event->attr.exclude_idle   ||
 	    event->attr.exclude_host   ||
-	    event->attr.exclude_guest  ||
-	    is_sampling_event(event)) /* no sampling */
+	    event->attr.exclude_guest)
 		return -EINVAL;
 
 	/* no branch sampling */
@@ -513,6 +512,9 @@ static int hv_24x7_init(void)
 	if (!hv_page_cache)
 		return -ENOMEM;
 
+	/* sampling not supported */
+	h_24x7_pmu.capabilities |= PERF_PMU_CAP_NO_INTERRUPT;
+
 	r = perf_pmu_register(&h_24x7_pmu, h_24x7_pmu.name, -1);
 	if (r)
 		return r;

arch/powerpc/perf/hv-gpci.c

@@ -210,8 +210,7 @@ static int h_gpci_event_init(struct perf_event *event)
 	    event->attr.exclude_hv     ||
 	    event->attr.exclude_idle   ||
 	    event->attr.exclude_host   ||
-	    event->attr.exclude_guest  ||
-	    is_sampling_event(event)) /* no sampling */
+	    event->attr.exclude_guest)
 		return -EINVAL;
 
 	/* no branch sampling */
@@ -284,6 +283,9 @@ static int hv_gpci_init(void)
 		return -ENODEV;
 	}
 
+	/* sampling not supported */
+	h_gpci_pmu.capabilities |= PERF_PMU_CAP_NO_INTERRUPT;
+
 	r = perf_pmu_register(&h_gpci_pmu, h_gpci_pmu.name, -1);
 	if (r)
 		return r;

arch/s390/include/uapi/asm/Kbuild

@@ -16,6 +16,7 @@ header-y += ioctls.h
 header-y += ipcbuf.h
 header-y += kvm.h
 header-y += kvm_para.h
+header-y += kvm_perf.h
 header-y += kvm_virtio.h
 header-y += mman.h
 header-y += monwriter.h

arch/s390/include/uapi/asm/kvm_perf.h

+/*
+ * Definitions for perf-kvm on s390
+ *
+ * Copyright 2014 IBM Corp.
+ * Author(s): Alexander Yarygin <yarygin@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License (version 2 only)
+ * as published by the Free Software Foundation.
+ */
+
+#ifndef __LINUX_KVM_PERF_S390_H
+#define __LINUX_KVM_PERF_S390_H
+
+#include <asm/sie.h>
+
+#define DECODE_STR_LEN 40
+
+#define VCPU_ID "id"
+
+#define KVM_ENTRY_TRACE "kvm:kvm_s390_sie_enter"
+#define KVM_EXIT_TRACE "kvm:kvm_s390_sie_exit"
+#define KVM_EXIT_REASON "icptcode"
+
+#endif

arch/s390/kernel/perf_cpum_cf.c

@@ -411,12 +411,6 @@ static int cpumf_pmu_event_init(struct perf_event *event)
 	case PERF_TYPE_HARDWARE:
 	case PERF_TYPE_HW_CACHE:
 	case PERF_TYPE_RAW:
-		/* The CPU measurement counter facility does not have overflow
-		 * interrupts to do sampling.  Sampling must be provided by
-		 * external means, for example, by timers.
-		 */
-		if (is_sampling_event(event))
-			return -ENOENT;
 		err = __hw_perf_event_init(event);
 		break;
 	default:
@@ -681,6 +675,12 @@ static int __init cpumf_pmu_init(void)
 		goto out;
 	}
 
+	/* The CPU measurement counter facility does not have overflow
+	 * interrupts to do sampling.  Sampling must be provided by
+	 * external means, for example, by timers.
+	 */
+	cpumf_pmu.capabilities |= PERF_PMU_CAP_NO_INTERRUPT;
+
 	cpumf_pmu.attr_groups = cpumf_cf_event_group();
 	rc = perf_pmu_register(&cpumf_pmu, "cpum_cf", PERF_TYPE_RAW);
 	if (rc) {

arch/sh/kernel/perf_event.c

@@ -128,14 +128,6 @@ static int __hw_perf_event_init(struct perf_event *event)
 	if (!sh_pmu_initialized())
 		return -ENODEV;
 
-	/*
-	 * All of the on-chip counters are "limited", in that they have
-	 * no interrupts, and are therefore unable to do sampling without
-	 * further work and timer assistance.
-	 */
-	if (hwc->sample_period)
-		return -EINVAL;
-
 	/*
 	 * See if we need to reserve the counter.
 	 *
@@ -392,6 +384,13 @@ int register_sh_pmu(struct sh_pmu *_pmu)
 
 	pr_info("Performance Events: %s support registered\n", _pmu->name);
 
+	/*
+	 * All of the on-chip counters are "limited", in that they have
+	 * no interrupts, and are therefore unable to do sampling without
+	 * further work and timer assistance.
+	 */
+	pmu.capabilities |= PERF_PMU_CAP_NO_INTERRUPT;
+
 	WARN_ON(_pmu->num_events > MAX_HWEVENTS);
 
 	perf_pmu_register(&pmu, "cpu", PERF_TYPE_RAW);

arch/x86/include/uapi/asm/Kbuild

@@ -22,6 +22,7 @@ header-y += ipcbuf.h
 header-y += ist.h
 header-y += kvm.h
 header-y += kvm_para.h
+header-y += kvm_perf.h
 header-y += ldt.h
 header-y += mce.h
 header-y += mman.h

arch/x86/include/uapi/asm/kvm_perf.h

+#ifndef _ASM_X86_KVM_PERF_H
+#define _ASM_X86_KVM_PERF_H
+
+#include <asm/svm.h>
+#include <asm/vmx.h>
+#include <asm/kvm.h>
+
+#define DECODE_STR_LEN 20
+
+#define VCPU_ID "vcpu_id"
+
+#define KVM_ENTRY_TRACE "kvm:kvm_entry"
+#define KVM_EXIT_TRACE "kvm:kvm_exit"
+#define KVM_EXIT_REASON "exit_reason"
+
+#endif /* _ASM_X86_KVM_PERF_H */

arch/x86/kernel/cpu/perf_event_amd_uncore.c

@@ -294,31 +294,41 @@ static struct amd_uncore *amd_uncore_alloc(unsigned int cpu)
 			cpu_to_node(cpu));
 }
 
-static void amd_uncore_cpu_up_prepare(unsigned int cpu)
+static int amd_uncore_cpu_up_prepare(unsigned int cpu)
 {
-	struct amd_uncore *uncore;
+	struct amd_uncore *uncore_nb = NULL, *uncore_l2;
 
 	if (amd_uncore_nb) {
-		uncore = amd_uncore_alloc(cpu);
-		uncore->cpu = cpu;
-		uncore->num_counters = NUM_COUNTERS_NB;
-		uncore->rdpmc_base = RDPMC_BASE_NB;
-		uncore->msr_base = MSR_F15H_NB_PERF_CTL;
-		uncore->active_mask = &amd_nb_active_mask;
-		uncore->pmu = &amd_nb_pmu;
-		*per_cpu_ptr(amd_uncore_nb, cpu) = uncore;
+		uncore_nb = amd_uncore_alloc(cpu);
+		if (!uncore_nb)
+			goto fail;
+		uncore_nb->cpu = cpu;
+		uncore_nb->num_counters = NUM_COUNTERS_NB;
+		uncore_nb->rdpmc_base = RDPMC_BASE_NB;
+		uncore_nb->msr_base = MSR_F15H_NB_PERF_CTL;
+		uncore_nb->active_mask = &amd_nb_active_mask;
+		uncore_nb->pmu = &amd_nb_pmu;
+		*per_cpu_ptr(amd_uncore_nb, cpu) = uncore_nb;
 	}
 
 	if (amd_uncore_l2) {
-		uncore = amd_uncore_alloc(cpu);
-		uncore->cpu = cpu;
-		uncore->num_counters = NUM_COUNTERS_L2;
-		uncore->rdpmc_base = RDPMC_BASE_L2;
-		uncore->msr_base = MSR_F16H_L2I_PERF_CTL;
-		uncore->active_mask = &amd_l2_active_mask;
-		uncore->pmu = &amd_l2_pmu;
-		*per_cpu_ptr(amd_uncore_l2, cpu) = uncore;
+		uncore_l2 = amd_uncore_alloc(cpu);
+		if (!uncore_l2)
+			goto fail;
+		uncore_l2->cpu = cpu;
+		uncore_l2->num_counters = NUM_COUNTERS_L2;
+		uncore_l2->rdpmc_base = RDPMC_BASE_L2;
+		uncore_l2->msr_base = MSR_F16H_L2I_PERF_CTL;
+		uncore_l2->active_mask = &amd_l2_active_mask;
+		uncore_l2->pmu = &amd_l2_pmu;
+		*per_cpu_ptr(amd_uncore_l2, cpu) = uncore_l2;
 	}
+
+	return 0;
+
+fail:
+	kfree(uncore_nb);
+	return -ENOMEM;
 }
 
 static struct amd_uncore *
@@ -441,7 +451,7 @@ static void uncore_dead(unsigned int cpu, struct amd_uncore * __percpu *uncores)
 
 	if (!--uncore->refcnt)
 		kfree(uncore);
-	*per_cpu_ptr(amd_uncore_nb, cpu) = NULL;
+	*per_cpu_ptr(uncores, cpu) = NULL;
 }
 
 static void amd_uncore_cpu_dead(unsigned int cpu)
@@ -461,7 +471,8 @@ amd_uncore_cpu_notifier(struct notifier_block *self, unsigned long action,
 
 	switch (action & ~CPU_TASKS_FROZEN) {
 	case CPU_UP_PREPARE:
-		amd_uncore_cpu_up_prepare(cpu);
+		if (amd_uncore_cpu_up_prepare(cpu))
+			return notifier_from_errno(-ENOMEM);
 		break;
 
 	case CPU_STARTING:
@@ -501,20 +512,33 @@ static void __init init_cpu_already_online(void *dummy)
 	amd_uncore_cpu_online(cpu);
 }
 
+static void cleanup_cpu_online(void *dummy)
+{
+	unsigned int cpu = smp_processor_id();
+
+	amd_uncore_cpu_dead(cpu);
+}
+
 static int __init amd_uncore_init(void)
 {
-	unsigned int cpu;
+	unsigned int cpu, cpu2;
 	int ret = -ENODEV;
 
 	if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD)
-		return -ENODEV;
+		goto fail_nodev;
 
 	if (!cpu_has_topoext)
-		return -ENODEV;
+		goto fail_nodev;
 
 	if (cpu_has_perfctr_nb) {
 		amd_uncore_nb = alloc_percpu(struct amd_uncore *);
-		perf_pmu_register(&amd_nb_pmu, amd_nb_pmu.name, -1);
+		if (!amd_uncore_nb) {
+			ret = -ENOMEM;
+			goto fail_nb;
+		}
+		ret = perf_pmu_register(&amd_nb_pmu, amd_nb_pmu.name, -1);
+		if (ret)
+			goto fail_nb;
 
 		printk(KERN_INFO "perf: AMD NB counters detected\n");
 		ret = 0;
@@ -522,20 +546,28 @@ static int __init amd_uncore_init(void)
 
 	if (cpu_has_perfctr_l2) {
 		amd_uncore_l2 = alloc_percpu(struct amd_uncore *);
-		perf_pmu_register(&amd_l2_pmu, amd_l2_pmu.name, -1);
+		if (!amd_uncore_l2) {
+			ret = -ENOMEM;
+			goto fail_l2;
+		}
+		ret = perf_pmu_register(&amd_l2_pmu, amd_l2_pmu.name, -1);
+		if (ret)
+			goto fail_l2;
 
 		printk(KERN_INFO "perf: AMD L2I counters detected\n");
 		ret = 0;
 	}
 
 	if (ret)
-		return -ENODEV;
+		goto fail_nodev;
 
 	cpu_notifier_register_begin();
 
 	/* init cpus already online before registering for hotplug notifier */
 	for_each_online_cpu(cpu) {
-		amd_uncore_cpu_up_prepare(cpu);
+		ret = amd_uncore_cpu_up_prepare(cpu);
+		if (ret)
+			goto fail_online;
 		smp_call_function_single(cpu, init_cpu_already_online, NULL, 1);
 	}
 
@@ -543,5 +575,30 @@ static int __init amd_uncore_init(void)
 	cpu_notifier_register_done();
 
 	return 0;
+
+
+fail_online:
+	for_each_online_cpu(cpu2) {
+		if (cpu2 == cpu)
+			break;
+		smp_call_function_single(cpu, cleanup_cpu_online, NULL, 1);
+	}
+	cpu_notifier_register_done();
+
+	/* amd_uncore_nb/l2 should have been freed by cleanup_cpu_online */
+	amd_uncore_nb = amd_uncore_l2 = NULL;
+	if (cpu_has_perfctr_l2)
+		perf_pmu_unregister(&amd_l2_pmu);
+fail_l2:
+	if (cpu_has_perfctr_nb)
+		perf_pmu_unregister(&amd_nb_pmu);
+	if (amd_uncore_l2)
+		free_percpu(amd_uncore_l2);
+fail_nb:
+	if (amd_uncore_nb)
+		free_percpu(amd_uncore_nb);
+
+fail_nodev:
+	return ret;
 }
 device_initcall(amd_uncore_init);

arch/x86/kernel/cpu/perf_event_intel_uncore.c

@@ -2947,10 +2947,7 @@ nhmex_rbox_get_constraint(struct intel_uncore_box *box, struct perf_event *event
 		 * extra registers. If we failed to take an extra
 		 * register, try the alternative.
 		 */
-		if (idx % 2)
-			idx--;
-		else
-			idx++;
+		idx ^= 1;
 		if (idx != reg1->idx % 6) {
 			if (idx == 2)
 				config1 >>= 8;

kernel/events/core.c

@@ -5266,6 +5266,12 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
 
 		goto got_name;
 	} else {
+		if (vma->vm_ops && vma->vm_ops->name) {
+			name = (char *) vma->vm_ops->name(vma);
+			if (name)
+				goto cpy_name;
+		}
+
 		name = (char *)arch_vma_name(vma);
 		if (name)
 			goto cpy_name;
@@ -7804,7 +7810,7 @@ inherit_task_group(struct perf_event *event, struct task_struct *parent,
 /*
  * Initialize the perf_event context in task_struct
  */
-int perf_event_init_context(struct task_struct *child, int ctxn)
+static int perf_event_init_context(struct task_struct *child, int ctxn)
 {
 	struct perf_event_context *child_ctx, *parent_ctx;
 	struct perf_event_context *cloned_ctx;

kernel/trace/trace_event_perf.c

@@ -30,6 +30,18 @@ static int perf_trace_event_perm(struct ftrace_event_call *tp_event,
 			return ret;
 	}
 
+	/*
+	 * We checked and allowed to create parent,
+	 * allow children without checking.
+	 */
+	if (p_event->parent)
+		return 0;
+
+	/*
+	 * It's ok to check current process (owner) permissions in here,
+	 * because code below is called only via perf_event_open syscall.
+	 */
+
 	/* The ftrace function trace is allowed only for root. */
 	if (ftrace_event_is_function(tp_event)) {
 		if (perf_paranoid_tracepoint_raw() && !capable(CAP_SYS_ADMIN))

tools/lib/traceevent/event-parse.c

@@ -2395,7 +2395,7 @@ process_flags(struct event_format *event, struct print_arg *arg, char **tok)
 {
 	struct print_arg *field;
 	enum event_type type;
-	char *token;
+	char *token = NULL;
 
 	memset(arg, 0, sizeof(*arg));
 	arg->type = PRINT_FLAGS;
@@ -2448,7 +2448,7 @@ process_symbols(struct event_format *event, struct print_arg *arg, char **tok)
 {
 	struct print_arg *field;
 	enum event_type type;
-	char *token;
+	char *token = NULL;
 
 	memset(arg, 0, sizeof(*arg));
 	arg->type = PRINT_SYMBOL;
@@ -2487,7 +2487,7 @@ process_hex(struct event_format *event, struct print_arg *arg, char **tok)
 {
 	struct print_arg *field;
 	enum event_type type;
-	char *token;
+	char *token = NULL;
 
 	memset(arg, 0, sizeof(*arg));
 	arg->type = PRINT_HEX;

tools/lib/traceevent/plugin_cfg80211.c

@@ -5,8 +5,7 @@
 #include "event-parse.h"
 
 static unsigned long long
-process___le16_to_cpup(struct trace_seq *s,
-		       unsigned long long *args)
+process___le16_to_cpup(struct trace_seq *s, unsigned long long *args)
 {
 	uint16_t *val = (uint16_t *) (unsigned long) args[0];
 	return val ? (long long) le16toh(*val) : 0;

tools/lib/traceevent/plugin_jbd2.c

@@ -30,8 +30,7 @@
 #define MINOR(dev)	((unsigned int) ((dev) & MINORMASK))
 
 static unsigned long long
-process_jbd2_dev_to_name(struct trace_seq *s,
-			 unsigned long long *args)
+process_jbd2_dev_to_name(struct trace_seq *s, unsigned long long *args)
 {
 	unsigned int dev = args[0];
 
@@ -40,8 +39,7 @@ process_jbd2_dev_to_name(struct trace_seq *s,
 }
 
 static unsigned long long
-process_jiffies_to_msecs(struct trace_seq *s,
-			 unsigned long long *args)
+process_jiffies_to_msecs(struct trace_seq *s, unsigned long long *args)
 {
 	unsigned long long jiffies = args[0];
 

tools/lib/traceevent/plugin_kvm.c

@@ -240,25 +240,38 @@ static const char *find_exit_reason(unsigned isa, int val)
 	for (i = 0; strings[i].val >= 0; i++)
 		if (strings[i].val == val)
 			break;
-	if (strings[i].str)
-		return strings[i].str;
-	return "UNKNOWN";
+
+	return strings[i].str;
 }
 
-static int kvm_exit_handler(struct trace_seq *s, struct pevent_record *record,
-			    struct event_format *event, void *context)
+static int print_exit_reason(struct trace_seq *s, struct pevent_record *record,
+			     struct event_format *event, const char *field)
 {
 	unsigned long long isa;
 	unsigned long long val;
-	unsigned long long info1 = 0, info2 = 0;
+	const char *reason;
 
-	if (pevent_get_field_val(s, event, "exit_reason", record, &val, 1) < 0)
+	if (pevent_get_field_val(s, event, field, record, &val, 1) < 0)
 		return -1;
 
 	if (pevent_get_field_val(s, event, "isa", record, &isa, 0) < 0)
 		isa = 1;
 
-	trace_seq_printf(s, "reason %s", find_exit_reason(isa, val));
+	reason = find_exit_reason(isa, val);
+	if (reason)
+		trace_seq_printf(s, "reason %s", reason);
+	else
+		trace_seq_printf(s, "reason UNKNOWN (%llu)", val);
+	return 0;
+}
+
+static int kvm_exit_handler(struct trace_seq *s, struct pevent_record *record,
+			    struct event_format *event, void *context)
+{
+	unsigned long long info1 = 0, info2 = 0;
+
+	if (print_exit_reason(s, record, event, "exit_reason") < 0)
+		return -1;
 
 	pevent_print_num_field(s, " rip 0x%lx", event, "guest_rip", record, 1);
 
@@ -313,6 +326,29 @@ static int kvm_emulate_insn_handler(struct trace_seq *s,
 	return 0;
 }
 
+
+static int kvm_nested_vmexit_inject_handler(struct trace_seq *s, struct pevent_record *record,
+					    struct event_format *event, void *context)
+{
+	if (print_exit_reason(s, record, event, "exit_code") < 0)
+		return -1;
+
+	pevent_print_num_field(s, " info1 %llx", event, "exit_info1", record, 1);
+	pevent_print_num_field(s, " info2 %llx", event, "exit_info2", record, 1);
+	pevent_print_num_field(s, " int_info %llx", event, "exit_int_info", record, 1);
+	pevent_print_num_field(s, " int_info_err %llx", event, "exit_int_info_err", record, 1);
+
+	return 0;
+}
+
+static int kvm_nested_vmexit_handler(struct trace_seq *s, struct pevent_record *record,
+				     struct event_format *event, void *context)
+{
+	pevent_print_num_field(s, "rip %llx ", event, "rip", record, 1);
+
+	return kvm_nested_vmexit_inject_handler(s, record, event, context);
+}
+
 union kvm_mmu_page_role {
 	unsigned word;
 	struct {
@@ -409,6 +445,12 @@ int PEVENT_PLUGIN_LOADER(struct pevent *pevent)
 	pevent_register_event_handler(pevent, -1, "kvm", "kvm_emulate_insn",
 				      kvm_emulate_insn_handler, NULL);
 
+	pevent_register_event_handler(pevent, -1, "kvm", "kvm_nested_vmexit",
+				      kvm_nested_vmexit_handler, NULL);
+
+	pevent_register_event_handler(pevent, -1, "kvm", "kvm_nested_vmexit_inject",
+				      kvm_nested_vmexit_inject_handler, NULL);
+
 	pevent_register_event_handler(pevent, -1, "kvmmmu", "kvm_mmu_get_page",
 				      kvm_mmu_get_page_handler, NULL);
 
@@ -443,6 +485,12 @@ void PEVENT_PLUGIN_UNLOADER(struct pevent *pevent)
 	pevent_unregister_event_handler(pevent, -1, "kvm", "kvm_emulate_insn",
 					kvm_emulate_insn_handler, NULL);
 
+	pevent_unregister_event_handler(pevent, -1, "kvm", "kvm_nested_vmexit",
+					kvm_nested_vmexit_handler, NULL);
+
+	pevent_unregister_event_handler(pevent, -1, "kvm", "kvm_nested_vmexit_inject",
+					kvm_nested_vmexit_inject_handler, NULL);
+
 	pevent_unregister_event_handler(pevent, -1, "kvmmmu", "kvm_mmu_get_page",
 					kvm_mmu_get_page_handler, NULL);
 

tools/perf/Documentation/perf-bench.txt

@@ -16,6 +16,10 @@ This 'perf bench' command is a general framework for benchmark suites.
 
 COMMON OPTIONS
 --------------
+-r::
+--repeat=::
+Specify amount of times to repeat the run (default 10).
+
 -f::
 --format=::
 Specify format style.

tools/perf/Documentation/perf-inject.txt

@@ -41,6 +41,9 @@ OPTIONS
 	tasks slept. sched_switch contains a callchain where a task slept and
 	sched_stat contains a timeslice how long a task slept.
 
+--kallsyms=<file>::
+	kallsyms pathname
+
 SEE ALSO
 --------
 linkperf:perf-record[1], linkperf:perf-report[1], linkperf:perf-archive[1]

tools/perf/Documentation/perf-kvm.txt

@@ -51,9 +51,9 @@ There are a couple of variants of perf kvm:
   'perf kvm stat <command>' to run a command and gather performance counter
   statistics.
   Especially, perf 'kvm stat record/report' generates a statistical analysis
-  of KVM events. Currently, vmexit, mmio and ioport events are supported.
-  'perf kvm stat record <command>' records kvm events and the events between
-  start and end <command>.
+  of KVM events. Currently, vmexit, mmio (x86 only) and ioport (x86 only)
+  events are supported. 'perf kvm stat record <command>' records kvm events
+  and the events between start and end <command>.
   And this command produces a file which contains tracing results of kvm
   events.
 
@@ -103,8 +103,8 @@ STAT REPORT OPTIONS
        analyze events which occures on this vcpu. (default: all vcpus)
 
 --event=<value>::
-       event to be analyzed. Possible values: vmexit, mmio, ioport.
-       (default: vmexit)
+       event to be analyzed. Possible values: vmexit, mmio (x86 only),
+       ioport (x86 only). (default: vmexit)
 -k::
 --key=<value>::
        Sorting key. Possible values: sample (default, sort by samples
@@ -138,7 +138,8 @@ STAT LIVE OPTIONS
 
 
 --event=<value>::
-       event to be analyzed. Possible values: vmexit, mmio, ioport.
+       event to be analyzed. Possible values: vmexit,
+       mmio (x86 only), ioport (x86 only).
        (default: vmexit)
 
 -k::
@@ -147,7 +148,8 @@ STAT LIVE OPTIONS
        number), time (sort by average time).
 
 --duration=<value>::
-       Show events other than HLT that take longer than duration usecs.
+       Show events other than HLT (x86 only) or Wait state (s390 only)
+       that take longer than duration usecs.
 
 SEE ALSO
 --------

tools/perf/Documentation/perf-timechart.txt

@@ -15,10 +15,20 @@ DESCRIPTION
 There are two variants of perf timechart:
 
   'perf timechart record <command>' to record the system level events
-  of an arbitrary workload.
+  of an arbitrary workload. By default timechart records only scheduler
+  and CPU events (task switches, running times, CPU power states, etc),
+  but it's possible to record IO (disk, network) activity using -I argument.
 
   'perf timechart' to turn a trace into a Scalable Vector Graphics file,
-  that can be viewed with popular SVG viewers such as 'Inkscape'.
+  that can be viewed with popular SVG viewers such as 'Inkscape'. Depending
+  on the events in the perf.data file, timechart will contain scheduler/cpu
+  events or IO events.
+
+  In IO mode, every bar has two charts: upper and lower.
+  Upper bar shows incoming events (disk reads, ingress network packets).
+  Lower bar shows outgoing events (disk writes, egress network packets).
+  There are also poll bars which show how much time application spent
+  in poll/epoll/select syscalls.
 
 TIMECHART OPTIONS
 -----------------
@@ -54,6 +64,19 @@ TIMECHART OPTIONS
 	duration or tasks with given name. If number is given it's interpreted
 	as number of nanoseconds. If non-numeric string is given it's
 	interpreted as task name.
+--io-skip-eagain::
+	Don't draw EAGAIN IO events.
+--io-min-time=<nsecs>::
+	Draw small events as if they lasted min-time. Useful when you need
+	to see very small and fast IO. It's possible to specify ms or us
+	suffix to specify time in milliseconds or microseconds.
+	Default value is 1ms.
+--io-merge-dist=<nsecs>::
+	Merge events that are merge-dist nanoseconds apart.
+	Reduces number of figures on the SVG and makes it more render-friendly.
+	It's possible to specify ms or us suffix to specify time in
+	milliseconds or microseconds.
+	Default value is 1us.
 
 RECORD OPTIONS
 --------------
@@ -63,6 +86,9 @@ RECORD OPTIONS
 -T::
 --tasks-only::
         Record only tasks-related events
+-I::
+--io-only::
+        Record only io-related events
 -g::
 --callchain::
         Do call-graph (stack chain/backtrace) recording
@@ -87,6 +113,14 @@ Record system-wide timechart:
 
   $ perf timechart --highlight gcc
 
+Record system-wide IO events:
+
+  $ perf timechart record -I
+
+  then generate timechart:
+
+  $ perf timechart
+
 SEE ALSO
 --------
 linkperf:perf-record[1]

tools/perf/Documentation/perf-trace.txt

@@ -107,6 +107,52 @@ the thread executes on the designated CPUs. Default is to monitor all CPUs.
 	Show tool stats such as number of times fd->pathname was discovered thru
 	hooking the open syscall return + vfs_getname or via reading /proc/pid/fd, etc.
 
+-F=[all|min|maj]::
+--pf=[all|min|maj]::
+	Trace pagefaults. Optionally, you can specify whether you want minor,
+	major or all pagefaults. Default value is maj.
+
+--syscalls::
+	Trace system calls. This options is enabled by default.
+
+PAGEFAULTS
+----------
+
+When tracing pagefaults, the format of the trace is as follows:
+
+<min|maj>fault [<ip.symbol>+<ip.offset>] => <addr.dso@addr.offset> (<map type><addr level>).
+
+- min/maj indicates whether fault event is minor or major;
+- ip.symbol shows symbol for instruction pointer (the code that generated the
+  fault); if no debug symbols available, perf trace will print raw IP;
+- addr.dso shows DSO for the faulted address;
+- map type is either 'd' for non-executable maps or 'x' for executable maps;
+- addr level is either 'k' for kernel dso or '.' for user dso.
+
+For symbols resolution you may need to install debugging symbols.
+
+Please be aware that duration is currently always 0 and doesn't reflect actual
+time it took for fault to be handled!
+
+When --verbose specified, perf trace tries to print all available information
+for both IP and fault address in the form of dso@symbol+offset.
+
+EXAMPLES
+--------
+
+Trace only major pagefaults:
+
+ $ perf trace --no-syscalls -F
+
+Trace syscalls, major and minor pagefaults:
+
+ $ perf trace -F all
+
+  1416.547 ( 0.000 ms): python/20235 majfault [CRYPTO_push_info_+0x0] => /lib/x86_64-linux-gnu/libcrypto.so.1.0.0@0x61be0 (x.)
+
+  As you can see, there was major pagefault in python process, from
+  CRYPTO_push_info_ routine which faulted somewhere in libcrypto.so.
+
 SEE ALSO
 --------
 linkperf:perf-record[1], linkperf:perf-script[1]

tools/perf/Documentation/perf.txt

@@ -8,7 +8,15 @@ perf - Performance analysis tools for Linux
 SYNOPSIS
 --------
 [verse]
-'perf' [--version] [--help] COMMAND [ARGS]
+'perf' [--version] [--help] [OPTIONS] COMMAND [ARGS]
+
+OPTIONS
+-------
+--debug::
+	Setup debug variable (just verbose for now) in value
+	range (0, 10). Use like:
+	  --debug verbose   # sets verbose = 1
+	  --debug verbose=2 # sets verbose = 2
 
 DESCRIPTION
 -----------

tools/perf/MANIFEST

@@ -37,3 +37,6 @@ arch/x86/include/asm/kvm_host.h
 arch/x86/include/uapi/asm/svm.h
 arch/x86/include/uapi/asm/vmx.h
 arch/x86/include/uapi/asm/kvm.h
+arch/x86/include/uapi/asm/kvm_perf.h
+arch/s390/include/uapi/asm/sie.h
+arch/s390/include/uapi/asm/kvm_perf.h

tools/perf/Makefile.perf

@@ -295,11 +295,13 @@ LIB_H += util/intlist.h
 LIB_H += util/perf_regs.h
 LIB_H += util/unwind.h
 LIB_H += util/vdso.h
+LIB_H += util/tsc.h
 LIB_H += ui/helpline.h
 LIB_H += ui/progress.h
 LIB_H += ui/util.h
 LIB_H += ui/ui.h
 LIB_H += util/data.h
+LIB_H += util/kvm-stat.h
 
 LIB_OBJS += $(OUTPUT)util/abspath.o
 LIB_OBJS += $(OUTPUT)util/alias.o
@@ -373,6 +375,8 @@ LIB_OBJS += $(OUTPUT)util/stat.o
 LIB_OBJS += $(OUTPUT)util/record.o
 LIB_OBJS += $(OUTPUT)util/srcline.o
 LIB_OBJS += $(OUTPUT)util/data.o
+LIB_OBJS += $(OUTPUT)util/tsc.o
+LIB_OBJS += $(OUTPUT)util/cloexec.o
 
 LIB_OBJS += $(OUTPUT)ui/setup.o
 LIB_OBJS += $(OUTPUT)ui/helpline.o

tools/perf/arch/powerpc/Makefile

@@ -3,3 +3,4 @@ PERF_HAVE_DWARF_REGS := 1
 LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o
 endif
 LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/header.o
+LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/skip-callchain-idx.o

tools/perf/arch/powerpc/util/header.c

@@ -5,9 +5,7 @@
 #include <string.h>
 
 #include "../../util/header.h"
-
-#define __stringify_1(x)        #x
-#define __stringify(x)          __stringify_1(x)
+#include "../../util/util.h"
 
 #define mfspr(rn)       ({unsigned long rval; \
 			 asm volatile("mfspr %0," __stringify(rn) \

tools/perf/arch/powerpc/util/skip-callchain-idx.c

+/*
+ * Use DWARF Debug information to skip unnecessary callchain entries.
+ *
+ * Copyright (C) 2014 Sukadev Bhattiprolu, IBM Corporation.
+ * Copyright (C) 2014 Ulrich Weigand, IBM Corporation.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+#include <inttypes.h>
+#include <dwarf.h>
+#include <elfutils/libdwfl.h>
+
+#include "util/thread.h"
+#include "util/callchain.h"
+
+/*
+ * When saving the callchain on Power, the kernel conservatively saves
+ * excess entries in the callchain. A few of these entries are needed
+ * in some cases but not others. If the unnecessary entries are not
+ * ignored, we end up with duplicate arcs in the call-graphs. Use
+ * DWARF debug information to skip over any unnecessary callchain
+ * entries.
+ *
+ * See function header for arch_adjust_callchain() below for more details.
+ *
+ * The libdwfl code in this file is based on code from elfutils
+ * (libdwfl/argp-std.c, libdwfl/tests/addrcfi.c, etc).
+ */
+static char *debuginfo_path;
+
+static const Dwfl_Callbacks offline_callbacks = {
+	.debuginfo_path = &debuginfo_path,
+	.find_debuginfo = dwfl_standard_find_debuginfo,
+	.section_address = dwfl_offline_section_address,
+};
+
+
+/*
+ * Use the DWARF expression for the Call-frame-address and determine
+ * if return address is in LR and if a new frame was allocated.
+ */
+static int check_return_reg(int ra_regno, Dwarf_Frame *frame)
+{
+	Dwarf_Op ops_mem[2];
+	Dwarf_Op dummy;
+	Dwarf_Op *ops = &dummy;
+	size_t nops;
+	int result;
+
+	result = dwarf_frame_register(frame, ra_regno, ops_mem, &ops, &nops);
+	if (result < 0) {
+		pr_debug("dwarf_frame_register() %s\n", dwarf_errmsg(-1));
+		return -1;
+	}
+
+	/*
+	 * Check if return address is on the stack.
+	 */
+	if (nops != 0 || ops != NULL)
+		return 0;
+
+	/*
+	 * Return address is in LR. Check if a frame was allocated
+	 * but not-yet used.
+	 */
+	result = dwarf_frame_cfa(frame, &ops, &nops);
+	if (result < 0) {
+		pr_debug("dwarf_frame_cfa() returns %d, %s\n", result,
+					dwarf_errmsg(-1));
+		return -1;
+	}
+
+	/*
+	 * If call frame address is in r1, no new frame was allocated.
+	 */
+	if (nops == 1 && ops[0].atom == DW_OP_bregx && ops[0].number == 1 &&
+				ops[0].number2 == 0)
+		return 1;
+
+	/*
+	 * A new frame was allocated but has not yet been used.
+	 */
+	return 2;
+}
+
+/*
+ * Get the DWARF frame from the .eh_frame section.
+ */
+static Dwarf_Frame *get_eh_frame(Dwfl_Module *mod, Dwarf_Addr pc)
+{
+	int		result;
+	Dwarf_Addr	bias;
+	Dwarf_CFI	*cfi;
+	Dwarf_Frame	*frame;
+
+	cfi = dwfl_module_eh_cfi(mod, &bias);
+	if (!cfi) {
+		pr_debug("%s(): no CFI - %s\n", __func__, dwfl_errmsg(-1));
+		return NULL;
+	}
+
+	result = dwarf_cfi_addrframe(cfi, pc, &frame);
+	if (result) {
+		pr_debug("%s(): %s\n", __func__, dwfl_errmsg(-1));
+		return NULL;
+	}
+
+	return frame;
+}
+
+/*
+ * Get the DWARF frame from the .debug_frame section.
+ */
+static Dwarf_Frame *get_dwarf_frame(Dwfl_Module *mod, Dwarf_Addr pc)
+{
+	Dwarf_CFI       *cfi;
+	Dwarf_Addr      bias;
+	Dwarf_Frame     *frame;
+	int             result;
+
+	cfi = dwfl_module_dwarf_cfi(mod, &bias);
+	if (!cfi) {
+		pr_debug("%s(): no CFI - %s\n", __func__, dwfl_errmsg(-1));
+		return NULL;
+	}
+
+	result = dwarf_cfi_addrframe(cfi, pc, &frame);
+	if (result) {
+		pr_debug("%s(): %s\n", __func__, dwfl_errmsg(-1));
+		return NULL;
+	}
+
+	return frame;
+}
+
+/*
+ * Return:
+ *	0 if return address for the program counter @pc is on stack
+ *	1 if return address is in LR and no new stack frame was allocated
+ *	2 if return address is in LR and a new frame was allocated (but not
+ *		yet used)
+ *	-1 in case of errors
+ */
+static int check_return_addr(const char *exec_file, Dwarf_Addr pc)
+{
+	int		rc = -1;
+	Dwfl		*dwfl;
+	Dwfl_Module	*mod;
+	Dwarf_Frame	*frame;
+	int		ra_regno;
+	Dwarf_Addr	start = pc;
+	Dwarf_Addr	end = pc;
+	bool		signalp;
+
+	dwfl = dwfl_begin(&offline_callbacks);
+	if (!dwfl) {
+		pr_debug("dwfl_begin() failed: %s\n", dwarf_errmsg(-1));
+		return -1;
+	}
+
+	if (dwfl_report_offline(dwfl, "",  exec_file, -1) == NULL) {
+		pr_debug("dwfl_report_offline() failed %s\n", dwarf_errmsg(-1));
+		goto out;
+	}
+
+	mod = dwfl_addrmodule(dwfl, pc);
+	if (!mod) {
+		pr_debug("dwfl_addrmodule() failed, %s\n", dwarf_errmsg(-1));
+		goto out;
+	}
+
+	/*
+	 * To work with split debug info files (eg: glibc), check both
+	 * .eh_frame and .debug_frame sections of the ELF header.
+	 */
+	frame = get_eh_frame(mod, pc);
+	if (!frame) {
+		frame = get_dwarf_frame(mod, pc);
+		if (!frame)
+			goto out;
+	}
+
+	ra_regno = dwarf_frame_info(frame, &start, &end, &signalp);
+	if (ra_regno < 0) {
+		pr_debug("Return address register unavailable: %s\n",
+				dwarf_errmsg(-1));
+		goto out;
+	}
+
+	rc = check_return_reg(ra_regno, frame);
+
+out:
+	dwfl_end(dwfl);
+	return rc;
+}
+
+/*
+ * The callchain saved by the kernel always includes the link register (LR).
+ *
+ *	0:	PERF_CONTEXT_USER
+ *	1:	Program counter (Next instruction pointer)
+ *	2:	LR value
+ *	3:	Caller's caller
+ *	4:	...
+ *
+ * The value in LR is only needed when it holds a return address. If the
+ * return address is on the stack, we should ignore the LR value.
+ *
+ * Further, when the return address is in the LR, if a new frame was just
+ * allocated but the LR was not saved into it, then the LR contains the
+ * caller, slot 4: contains the caller's caller and the contents of slot 3:
+ * (chain->ips[3]) is undefined and must be ignored.
+ *
+ * Use DWARF debug information to determine if any entries need to be skipped.
+ *
+ * Return:
+ *	index:	of callchain entry that needs to be ignored (if any)
+ *	-1	if no entry needs to be ignored or in case of errors
+ */
+int arch_skip_callchain_idx(struct machine *machine, struct thread *thread,
+				struct ip_callchain *chain)
+{
+	struct addr_location al;
+	struct dso *dso = NULL;
+	int rc;
+	u64 ip;
+	u64 skip_slot = -1;
+
+	if (chain->nr < 3)
+		return skip_slot;
+
+	ip = chain->ips[2];
+
+	thread__find_addr_location(thread, machine, PERF_RECORD_MISC_USER,
+			MAP__FUNCTION, ip, &al);
+
+	if (al.map)
+		dso = al.map->dso;
+
+	if (!dso) {
+		pr_debug("%" PRIx64 " dso is NULL\n", ip);
+		return skip_slot;
+	}
+
+	rc = check_return_addr(dso->long_name, ip);
+
+	pr_debug("DSO %s, nr %" PRIx64 ", ip 0x%" PRIx64 "rc %d\n",
+				dso->long_name, chain->nr, ip, rc);
+
+	if (rc == 0) {
+		/*
+		 * Return address on stack. Ignore LR value in callchain
+		 */
+		skip_slot = 2;
+	} else if (rc == 2) {
+		/*
+		 * New frame allocated but return address still in LR.
+		 * Ignore the caller's caller entry in callchain.
+		 */
+		skip_slot = 3;
+	}
+	return skip_slot;
+}

tools/perf/arch/s390/Makefile

@@ -2,3 +2,6 @@ ifndef NO_DWARF
 PERF_HAVE_DWARF_REGS := 1
 LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/dwarf-regs.o
 endif
+LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/header.o
+HAVE_KVM_STAT_SUPPORT := 1
+LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/kvm-stat.o

tools/perf/arch/s390/util/header.c

+/*
+ * Implementation of get_cpuid().
+ *
+ * Copyright 2014 IBM Corp.
+ * Author(s): Alexander Yarygin <yarygin@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License (version 2 only)
+ * as published by the Free Software Foundation.
+ */
+
+#include <sys/types.h>
+#include <unistd.h>
+#include <stdio.h>
+#include <string.h>
+
+#include "../../util/header.h"
+
+int get_cpuid(char *buffer, size_t sz)
+{
+	const char *cpuid = "IBM/S390";
+
+	if (strlen(cpuid) + 1 > sz)
+		return -1;
+
+	strcpy(buffer, cpuid);
+	return 0;
+}

tools/perf/arch/s390/util/kvm-stat.c

+/*
+ * Arch specific functions for perf kvm stat.
+ *
+ * Copyright 2014 IBM Corp.
+ * Author(s): Alexander Yarygin <yarygin@linux.vnet.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License (version 2 only)
+ * as published by the Free Software Foundation.
+ */
+
+#include "../../util/kvm-stat.h"
+#include <asm/kvm_perf.h>
+
+define_exit_reasons_table(sie_exit_reasons, sie_intercept_code);
+define_exit_reasons_table(sie_icpt_insn_codes, icpt_insn_codes);
+define_exit_reasons_table(sie_sigp_order_codes, sigp_order_codes);
+define_exit_reasons_table(sie_diagnose_codes, diagnose_codes);
+define_exit_reasons_table(sie_icpt_prog_codes, icpt_prog_codes);
+
+static void event_icpt_insn_get_key(struct perf_evsel *evsel,
+				    struct perf_sample *sample,
+				    struct event_key *key)
+{
+	unsigned long insn;
+
+	insn = perf_evsel__intval(evsel, sample, "instruction");
+	key->key = icpt_insn_decoder(insn);
+	key->exit_reasons = sie_icpt_insn_codes;
+}
+
+static void event_sigp_get_key(struct perf_evsel *evsel,
+			       struct perf_sample *sample,
+			       struct event_key *key)
+{
+	key->key = perf_evsel__intval(evsel, sample, "order_code");
+	key->exit_reasons = sie_sigp_order_codes;
+}
+
+static void event_diag_get_key(struct perf_evsel *evsel,
+			       struct perf_sample *sample,
+			       struct event_key *key)
+{
+	key->key = perf_evsel__intval(evsel, sample, "code");
+	key->exit_reasons = sie_diagnose_codes;
+}
+
+static void event_icpt_prog_get_key(struct perf_evsel *evsel,
+				    struct perf_sample *sample,
+				    struct event_key *key)
+{
+	key->key = perf_evsel__intval(evsel, sample, "code");
+	key->exit_reasons = sie_icpt_prog_codes;
+}
+
+static struct child_event_ops child_events[] = {
+	{ .name = "kvm:kvm_s390_intercept_instruction",
+	  .get_key = event_icpt_insn_get_key },
+	{ .name = "kvm:kvm_s390_handle_sigp",
+	  .get_key = event_sigp_get_key },
+	{ .name = "kvm:kvm_s390_handle_diag",
+	  .get_key = event_diag_get_key },
+	{ .name = "kvm:kvm_s390_intercept_prog",
+	  .get_key = event_icpt_prog_get_key },
+	{ NULL, NULL },
+};
+
+static struct kvm_events_ops exit_events = {
+	.is_begin_event = exit_event_begin,
+	.is_end_event = exit_event_end,
+	.child_ops = child_events,
+	.decode_key = exit_event_decode_key,
+	.name = "VM-EXIT"
+};
+
+const char * const kvm_events_tp[] = {
+	"kvm:kvm_s390_sie_enter",
+	"kvm:kvm_s390_sie_exit",
+	"kvm:kvm_s390_intercept_instruction",
+	"kvm:kvm_s390_handle_sigp",
+	"kvm:kvm_s390_handle_diag",
+	"kvm:kvm_s390_intercept_prog",
+	NULL,
+};
+
+struct kvm_reg_events_ops kvm_reg_events_ops[] = {
+	{ .name = "vmexit", .ops = &exit_events },
+	{ NULL, NULL },
+};
+
+const char * const kvm_skip_events[] = {
+	"Wait state",
+	NULL,
+};
+
+int cpu_isa_init(struct perf_kvm_stat *kvm, const char *cpuid)
+{
+	if (strstr(cpuid, "IBM/S390")) {
+		kvm->exit_reasons = sie_exit_reasons;
+		kvm->exit_reasons_isa = "SIE";
+	} else
+		return -ENOTSUP;
+
+	return 0;
+}

tools/perf/arch/x86/Makefile

@@ -15,3 +15,5 @@ endif
 LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/header.o
 LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/tsc.o
 LIB_H += arch/$(ARCH)/util/tsc.h
+HAVE_KVM_STAT_SUPPORT := 1
+LIB_OBJS += $(OUTPUT)arch/$(ARCH)/util/kvm-stat.o

tools/perf/arch/x86/tests/dwarf-unwind.c

@@ -3,6 +3,7 @@
 #include "thread.h"
 #include "map.h"
 #include "event.h"
+#include "debug.h"
 #include "tests/tests.h"
 
 #define STACK_SIZE 8192

tools/perf/arch/x86/util/kvm-stat.c

+#include "../../util/kvm-stat.h"
+#include <asm/kvm_perf.h>
+
+define_exit_reasons_table(vmx_exit_reasons, VMX_EXIT_REASONS);
+define_exit_reasons_table(svm_exit_reasons, SVM_EXIT_REASONS);
+
+static struct kvm_events_ops exit_events = {
+	.is_begin_event = exit_event_begin,
+	.is_end_event = exit_event_end,
+	.decode_key = exit_event_decode_key,
+	.name = "VM-EXIT"
+};
+
+/*
+ * For the mmio events, we treat:
+ * the time of MMIO write: kvm_mmio(KVM_TRACE_MMIO_WRITE...) -> kvm_entry
+ * the time of MMIO read: kvm_exit -> kvm_mmio(KVM_TRACE_MMIO_READ...).
+ */
+static void mmio_event_get_key(struct perf_evsel *evsel, struct perf_sample *sample,
+			       struct event_key *key)
+{
+	key->key  = perf_evsel__intval(evsel, sample, "gpa");
+	key->info = perf_evsel__intval(evsel, sample, "type");
+}
+
+#define KVM_TRACE_MMIO_READ_UNSATISFIED 0
+#define KVM_TRACE_MMIO_READ 1
+#define KVM_TRACE_MMIO_WRITE 2
+
+static bool mmio_event_begin(struct perf_evsel *evsel,
+			     struct perf_sample *sample, struct event_key *key)
+{
+	/* MMIO read begin event in kernel. */
+	if (kvm_exit_event(evsel))
+		return true;
+
+	/* MMIO write begin event in kernel. */
+	if (!strcmp(evsel->name, "kvm:kvm_mmio") &&
+	    perf_evsel__intval(evsel, sample, "type") == KVM_TRACE_MMIO_WRITE) {
+		mmio_event_get_key(evsel, sample, key);
+		return true;
+	}
+
+	return false;
+}
+
+static bool mmio_event_end(struct perf_evsel *evsel, struct perf_sample *sample,
+			   struct event_key *key)
+{
+	/* MMIO write end event in kernel. */
+	if (kvm_entry_event(evsel))
+		return true;
+
+	/* MMIO read end event in kernel.*/
+	if (!strcmp(evsel->name, "kvm:kvm_mmio") &&
+	    perf_evsel__intval(evsel, sample, "type") == KVM_TRACE_MMIO_READ) {
+		mmio_event_get_key(evsel, sample, key);
+		return true;
+	}
+
+	return false;
+}
+
+static void mmio_event_decode_key(struct perf_kvm_stat *kvm __maybe_unused,
+				  struct event_key *key,
+				  char *decode)
+{
+	scnprintf(decode, DECODE_STR_LEN, "%#lx:%s",
+		  (unsigned long)key->key,
+		  key->info == KVM_TRACE_MMIO_WRITE ? "W" : "R");
+}
+
+static struct kvm_events_ops mmio_events = {
+	.is_begin_event = mmio_event_begin,
+	.is_end_event = mmio_event_end,
+	.decode_key = mmio_event_decode_key,
+	.name = "MMIO Access"
+};
+
+ /* The time of emulation pio access is from kvm_pio to kvm_entry. */
+static void ioport_event_get_key(struct perf_evsel *evsel,
+				 struct perf_sample *sample,
+				 struct event_key *key)
+{
+	key->key  = perf_evsel__intval(evsel, sample, "port");
+	key->info = perf_evsel__intval(evsel, sample, "rw");
+}
+
+static bool ioport_event_begin(struct perf_evsel *evsel,
+			       struct perf_sample *sample,
+			       struct event_key *key)
+{
+	if (!strcmp(evsel->name, "kvm:kvm_pio")) {
+		ioport_event_get_key(evsel, sample, key);
+		return true;
+	}
+
+	return false;
+}
+
+static bool ioport_event_end(struct perf_evsel *evsel,
+			     struct perf_sample *sample __maybe_unused,
+			     struct event_key *key __maybe_unused)
+{
+	return kvm_entry_event(evsel);
+}
+
+static void ioport_event_decode_key(struct perf_kvm_stat *kvm __maybe_unused,
+				    struct event_key *key,
+				    char *decode)
+{
+	scnprintf(decode, DECODE_STR_LEN, "%#llx:%s",
+		  (unsigned long long)key->key,
+		  key->info ? "POUT" : "PIN");
+}
+
+static struct kvm_events_ops ioport_events = {
+	.is_begin_event = ioport_event_begin,
+	.is_end_event = ioport_event_end,
+	.decode_key = ioport_event_decode_key,
+	.name = "IO Port Access"
+};
+
+const char * const kvm_events_tp[] = {
+	"kvm:kvm_entry",
+	"kvm:kvm_exit",
+	"kvm:kvm_mmio",
+	"kvm:kvm_pio",
+	NULL,
+};
+
+struct kvm_reg_events_ops kvm_reg_events_ops[] = {
+	{ .name = "vmexit", .ops = &exit_events },
+	{ .name = "mmio", .ops = &mmio_events },
+	{ .name = "ioport", .ops = &ioport_events },
+	{ NULL, NULL },
+};
+
+const char * const kvm_skip_events[] = {
+	"HLT",
+	NULL,
+};
+
+int cpu_isa_init(struct perf_kvm_stat *kvm, const char *cpuid)
+{
+	if (strstr(cpuid, "Intel")) {
+		kvm->exit_reasons = vmx_exit_reasons;
+		kvm->exit_reasons_isa = "VMX";
+	} else if (strstr(cpuid, "AMD")) {
+		kvm->exit_reasons = svm_exit_reasons;
+		kvm->exit_reasons_isa = "SVM";
+	} else
+		return -ENOTSUP;
+
+	return 0;
+}

tools/perf/arch/x86/util/tsc.c

@@ -6,29 +6,9 @@
 #include "../../perf.h"
 #include <linux/types.h>
 #include "../../util/debug.h"
+#include "../../util/tsc.h"
 #include "tsc.h"
 
-u64 perf_time_to_tsc(u64 ns, struct perf_tsc_conversion *tc)
-{
-	u64 t, quot, rem;
-
-	t = ns - tc->time_zero;
-	quot = t / tc->time_mult;
-	rem  = t % tc->time_mult;
-	return (quot << tc->time_shift) +
-	       (rem << tc->time_shift) / tc->time_mult;
-}
-
-u64 tsc_to_perf_time(u64 cyc, struct perf_tsc_conversion *tc)
-{
-	u64 quot, rem;
-
-	quot = cyc >> tc->time_shift;
-	rem  = cyc & ((1 << tc->time_shift) - 1);
-	return tc->time_zero + quot * tc->time_mult +
-	       ((rem * tc->time_mult) >> tc->time_shift);
-}
-
 int perf_read_tsc_conversion(const struct perf_event_mmap_page *pc,
 			     struct perf_tsc_conversion *tc)
 {
@@ -57,3 +37,12 @@ int perf_read_tsc_conversion(const struct perf_event_mmap_page *pc,
 
 	return 0;
 }
+
+u64 rdtsc(void)
+{
+	unsigned int low, high;
+
+	asm volatile("rdtsc" : "=a" (low), "=d" (high));
+
+	return low | ((u64)high) << 32;
+}

tools/perf/arch/x86/util/tsc.h

@@ -14,7 +14,4 @@ struct perf_event_mmap_page;
 int perf_read_tsc_conversion(const struct perf_event_mmap_page *pc,
 			     struct perf_tsc_conversion *tc);
 
-u64 perf_time_to_tsc(u64 ns, struct perf_tsc_conversion *tc);
-u64 tsc_to_perf_time(u64 cyc, struct perf_tsc_conversion *tc);
-
 #endif /* TOOLS_PERF_ARCH_X86_UTIL_TSC_H__ */

tools/perf/arch/x86/util/unwind-libunwind.c

@@ -3,6 +3,7 @@
 #include <libunwind.h>
 #include "perf_regs.h"
 #include "../../util/unwind.h"
+#include "../../util/debug.h"
 
 #ifdef HAVE_ARCH_X86_64_SUPPORT
 int libunwind__arch_reg_id(int regnum)

tools/perf/bench/bench.h

@@ -43,5 +43,6 @@ extern int bench_futex_requeue(int argc, const char **argv, const char *prefix);
 #define BENCH_FORMAT_UNKNOWN		-1
 
 extern int bench_format;
+extern unsigned int bench_repeat;
 
 #endif

tools/perf/bench/futex-requeue.c

@@ -29,13 +29,6 @@ static u_int32_t futex1 = 0, futex2 = 0;
  */
 static unsigned int nrequeue = 1;
 
-/*
- * There can be significant variance from run to run,
- * the more repeats, the more exact the overall avg and
- * the better idea of the futex latency.
- */
-static unsigned int repeat = 10;
-
 static pthread_t *worker;
 static bool done = 0, silent = 0;
 static pthread_mutex_t thread_lock;
@@ -46,7 +39,6 @@ static unsigned int ncpus, threads_starting, nthreads = 0;
 static const struct option options[] = {
 	OPT_UINTEGER('t', "threads",  &nthreads, "Specify amount of threads"),
 	OPT_UINTEGER('q', "nrequeue", &nrequeue, "Specify amount of threads to requeue at once"),
-	OPT_UINTEGER('r', "repeat",   &repeat,   "Specify amount of times to repeat the run"),
 	OPT_BOOLEAN( 's', "silent",   &silent,   "Silent mode: do not display data/details"),
 	OPT_END()
 };
@@ -146,7 +138,7 @@ int bench_futex_requeue(int argc, const char **argv,
 	pthread_cond_init(&thread_parent, NULL);
 	pthread_cond_init(&thread_worker, NULL);
 
-	for (j = 0; j < repeat && !done; j++) {
+	for (j = 0; j < bench_repeat && !done; j++) {
 		unsigned int nrequeued = 0;
 		struct timeval start, end, runtime;
 

tools/perf/bench/futex-wake.c

@@ -30,15 +30,8 @@ static u_int32_t futex1 = 0;
  */
 static unsigned int nwakes = 1;
 
-/*
- * There can be significant variance from run to run,
- * the more repeats, the more exact the overall avg and
- * the better idea of the futex latency.
- */
-static unsigned int repeat = 10;
-
 pthread_t *worker;
-static bool done = 0, silent = 0;
+static bool done = false, silent = false;
 static pthread_mutex_t thread_lock;
 static pthread_cond_t thread_parent, thread_worker;
 static struct stats waketime_stats, wakeup_stats;
@@ -47,7 +40,6 @@ static unsigned int ncpus, threads_starting, nthreads = 0;
 static const struct option options[] = {
 	OPT_UINTEGER('t', "threads", &nthreads, "Specify amount of threads"),
 	OPT_UINTEGER('w', "nwakes",  &nwakes,   "Specify amount of threads to wake at once"),
-	OPT_UINTEGER('r', "repeat",  &repeat,   "Specify amount of times to repeat the run"),
 	OPT_BOOLEAN( 's', "silent",  &silent,   "Silent mode: do not display data/details"),
 	OPT_END()
 };
@@ -149,7 +141,7 @@ int bench_futex_wake(int argc, const char **argv,
 	pthread_cond_init(&thread_parent, NULL);
 	pthread_cond_init(&thread_worker, NULL);
 
-	for (j = 0; j < repeat && !done; j++) {
+	for (j = 0; j < bench_repeat && !done; j++) {
 		unsigned int nwoken = 0;
 		struct timeval start, end, runtime;
 

tools/perf/bench/mem-memcpy.c

@@ -10,6 +10,7 @@
 #include "../util/util.h"
 #include "../util/parse-options.h"
 #include "../util/header.h"
+#include "../util/cloexec.h"
 #include "bench.h"
 #include "mem-memcpy-arch.h"
 
@@ -83,7 +84,8 @@ static struct perf_event_attr cycle_attr = {
 
 static void init_cycle(void)
 {
-	cycle_fd = sys_perf_event_open(&cycle_attr, getpid(), -1, -1, 0);
+	cycle_fd = sys_perf_event_open(&cycle_attr, getpid(), -1, -1,
+				       perf_event_open_cloexec_flag());
 
 	if (cycle_fd < 0 && errno == ENOSYS)
 		die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
@@ -189,6 +191,11 @@ int bench_mem_memcpy(int argc, const char **argv,
 	argc = parse_options(argc, argv, options,
 			     bench_mem_memcpy_usage, 0);
 
+	if (no_prefault && only_prefault) {
+		fprintf(stderr, "Invalid options: -o and -n are mutually exclusive\n");
+		return 1;
+	}
+
 	if (use_cycle)
 		init_cycle();
 

tools/perf/bench/mem-memset.c

@@ -10,6 +10,7 @@
 #include "../util/util.h"
 #include "../util/parse-options.h"
 #include "../util/header.h"
+#include "../util/cloexec.h"
 #include "bench.h"
 #include "mem-memset-arch.h"
 
@@ -83,7 +84,8 @@ static struct perf_event_attr cycle_attr = {
 
 static void init_cycle(void)
 {
-	cycle_fd = sys_perf_event_open(&cycle_attr, getpid(), -1, -1, 0);
+	cycle_fd = sys_perf_event_open(&cycle_attr, getpid(), -1, -1,
+				       perf_event_open_cloexec_flag());
 
 	if (cycle_fd < 0 && errno == ENOSYS)
 		die("No CONFIG_PERF_EVENTS=y kernel support configured?\n");
@@ -181,6 +183,11 @@ int bench_mem_memset(int argc, const char **argv,
 	argc = parse_options(argc, argv, options,
 			     bench_mem_memset_usage, 0);
 
+	if (no_prefault && only_prefault) {
+		fprintf(stderr, "Invalid options: -o and -n are mutually exclusive\n");
+		return 1;
+	}
+
 	if (use_cycle)
 		init_cycle();
 

tools/perf/bench/sched-messaging.c

@@ -28,6 +28,7 @@
 #include <sys/time.h>
 #include <sys/poll.h>
 #include <limits.h>
+#include <err.h>
 
 #define DATASIZE 100
 
@@ -50,12 +51,6 @@ struct receiver_context {
 	int wakefd;
 };
 
-static void barf(const char *msg)
-{
-	fprintf(stderr, "%s (error: %s)\n", msg, strerror(errno));
-	exit(1);
-}
-
 static void fdpair(int fds[2])
 {
 	if (use_pipes) {
@@ -66,7 +61,7 @@ static void fdpair(int fds[2])
 			return;
 	}
 
-	barf(use_pipes ? "pipe()" : "socketpair()");
+	err(EXIT_FAILURE, use_pipes ? "pipe()" : "socketpair()");
 }
 
 /* Block until we're ready to go */
@@ -77,11 +72,11 @@ static void ready(int ready_out, int wakefd)
 
 	/* Tell them we're ready. */
 	if (write(ready_out, &dummy, 1) != 1)
-		barf("CLIENT: ready write");
+		err(EXIT_FAILURE, "CLIENT: ready write");
 
 	/* Wait for "GO" signal */
 	if (poll(&pollfd, 1, -1) != 1)
-		barf("poll");
+		err(EXIT_FAILURE, "poll");
 }
 
 /* Sender sprays loops messages down each file descriptor */
@@ -101,7 +96,7 @@ static void *sender(struct sender_context *ctx)
 			ret = write(ctx->out_fds[j], data + done,
 				    sizeof(data)-done);
 			if (ret < 0)
-				barf("SENDER: write");
+				err(EXIT_FAILURE, "SENDER: write");
 			done += ret;
 			if (done < DATASIZE)
 				goto again;
@@ -131,7 +126,7 @@ static void *receiver(struct receiver_context* ctx)
 again:
 		ret = read(ctx->in_fds[0], data + done, DATASIZE - done);
 		if (ret < 0)
-			barf("SERVER: read");
+			err(EXIT_FAILURE, "SERVER: read");
 		done += ret;
 		if (done < DATASIZE)
 			goto again;
@@ -144,14 +139,14 @@ static pthread_t create_worker(void *ctx, void *(*func)(void *))
 {
 	pthread_attr_t attr;
 	pthread_t childid;
-	int err;
+	int ret;
 
 	if (!thread_mode) {
 		/* process mode */
 		/* Fork the receiver. */
 		switch (fork()) {
 		case -1:
-			barf("fork()");
+			err(EXIT_FAILURE, "fork()");
 			break;
 		case 0:
 			(*func) (ctx);
@@ -165,19 +160,17 @@ static pthread_t create_worker(void *ctx, void *(*func)(void *))
 	}
 
 	if (pthread_attr_init(&attr) != 0)
-		barf("pthread_attr_init:");
+		err(EXIT_FAILURE, "pthread_attr_init:");
 
 #ifndef __ia64__
 	if (pthread_attr_setstacksize(&attr, PTHREAD_STACK_MIN) != 0)
-		barf("pthread_attr_setstacksize");
+		err(EXIT_FAILURE, "pthread_attr_setstacksize");
 #endif
 
-	err = pthread_create(&childid, &attr, func, ctx);
-	if (err != 0) {
-		fprintf(stderr, "pthread_create failed: %s (%d)\n",
-			strerror(err), err);
-		exit(-1);
-	}
+	ret = pthread_create(&childid, &attr, func, ctx);
+	if (ret != 0)
+		err(EXIT_FAILURE, "pthread_create failed");
+
 	return childid;
 }
 
@@ -207,14 +200,14 @@ static unsigned int group(pthread_t *pth,
 			+ num_fds * sizeof(int));
 
 	if (!snd_ctx)
-		barf("malloc()");
+		err(EXIT_FAILURE, "malloc()");
 
 	for (i = 0; i < num_fds; i++) {
 		int fds[2];
 		struct receiver_context *ctx = malloc(sizeof(*ctx));
 
 		if (!ctx)
-			barf("malloc()");
+			err(EXIT_FAILURE, "malloc()");
 
 
 		/* Create the pipe between client and server */
@@ -281,7 +274,7 @@ int bench_sched_messaging(int argc, const char **argv,
 
 	pth_tab = malloc(num_fds * 2 * num_groups * sizeof(pthread_t));
 	if (!pth_tab)
-		barf("main:malloc()");
+		err(EXIT_FAILURE, "main:malloc()");
 
 	fdpair(readyfds);
 	fdpair(wakefds);
@@ -294,13 +287,13 @@ int bench_sched_messaging(int argc, const char **argv,
 	/* Wait for everyone to be ready */
 	for (i = 0; i < total_children; i++)
 		if (read(readyfds[0], &dummy, 1) != 1)
-			barf("Reading for readyfds");
+			err(EXIT_FAILURE, "Reading for readyfds");
 
 	gettimeofday(&start, NULL);
 
 	/* Kick them off */
 	if (write(wakefds[1], &dummy, 1) != 1)
-		barf("Writing to start them");
+		err(EXIT_FAILURE, "Writing to start them");
 
 	/* Reap them all */
 	for (i = 0; i < total_children; i++)
@@ -332,5 +325,7 @@ int bench_sched_messaging(int argc, const char **argv,
 		break;
 	}
 
+	free(pth_tab);
+
 	return 0;
 }

tools/perf/builtin-bench.c

@@ -104,9 +104,11 @@ static const char *bench_format_str;
 
 /* Output/formatting style, exported to benchmark modules: */
 int bench_format = BENCH_FORMAT_DEFAULT;
+unsigned int bench_repeat = 10; /* default number of times to repeat the run */
 
 static const struct option bench_options[] = {
 	OPT_STRING('f', "format", &bench_format_str, "default", "Specify format style"),
+	OPT_UINTEGER('r', "repeat",  &bench_repeat,   "Specify amount of times to repeat the run"),
 	OPT_END()
 };
 
@@ -226,6 +228,11 @@ int cmd_bench(int argc, const char **argv, const char *prefix __maybe_unused)
 		goto end;
 	}
 
+	if (bench_repeat == 0) {
+		printf("Invalid repeat option: Must specify a positive value\n");
+		goto end;
+	}
+
 	if (argc < 1) {
 		print_usage();
 		goto end;

tools/perf/builtin-buildid-cache.c

@@ -125,7 +125,8 @@ static int build_id_cache__kcore_existing(const char *from_dir, char *to_dir,
 	return ret;
 }
 
-static int build_id_cache__add_kcore(const char *filename, const char *debugdir)
+static int build_id_cache__add_kcore(const char *filename, const char *debugdir,
+				     bool force)
 {
 	char dir[32], sbuildid[BUILD_ID_SIZE * 2 + 1];
 	char from_dir[PATH_MAX], to_dir[PATH_MAX];
@@ -144,7 +145,8 @@ static int build_id_cache__add_kcore(const char *filename, const char *debugdir)
 	scnprintf(to_dir, sizeof(to_dir), "%s/[kernel.kcore]/%s",
 		  debugdir, sbuildid);
 
-	if (!build_id_cache__kcore_existing(from_dir, to_dir, sizeof(to_dir))) {
+	if (!force &&
+	    !build_id_cache__kcore_existing(from_dir, to_dir, sizeof(to_dir))) {
 		pr_debug("same kcore found in %s\n", to_dir);
 		return 0;
 	}
@@ -389,7 +391,7 @@ int cmd_buildid_cache(int argc, const char **argv,
 	}
 
 	if (kcore_filename &&
-	    build_id_cache__add_kcore(kcore_filename, debugdir))
+	    build_id_cache__add_kcore(kcore_filename, debugdir, force))
 		pr_warning("Couldn't add %s\n", kcore_filename);
 
 	return ret;

tools/perf/builtin-evlist.c

@@ -15,6 +15,7 @@
 #include "util/parse-options.h"
 #include "util/session.h"
 #include "util/data.h"
+#include "util/debug.h"
 
 static int __cmd_evlist(const char *file_name, struct perf_attr_details *details)
 {

tools/perf/builtin-help.c

@@ -11,6 +11,7 @@
 #include "util/parse-options.h"
 #include "util/run-command.h"
 #include "util/help.h"
+#include "util/debug.h"
 
 static struct man_viewer_list {
 	struct man_viewer_list *next;

tools/perf/builtin-inject.c

@@ -389,6 +389,9 @@ static int __cmd_inject(struct perf_inject *inject)
 	ret = perf_session__process_events(session, &inject->tool);
 
 	if (!file_out->is_pipe) {
+		if (inject->build_ids)
+			perf_header__set_feat(&session->header,
+					      HEADER_BUILD_ID);
 		session->header.data_size = inject->bytes_written;
 		perf_session__write_header(session, session->evlist, file_out->fd, true);
 	}
@@ -436,6 +439,8 @@ int cmd_inject(int argc, const char **argv, const char *prefix __maybe_unused)
 			    "where and how long tasks slept"),
 		OPT_INCR('v', "verbose", &verbose,
 			 "be more verbose (show build ids, etc)"),
+		OPT_STRING(0, "kallsyms", &symbol_conf.kallsyms_name, "file",
+			   "kallsyms pathname"),
 		OPT_END()
 	};
 	const char * const inject_usage[] = {

tools/perf/builtin-record.c

@@ -238,6 +238,7 @@ static struct perf_event_header finished_round_event = {
 
 static int record__mmap_read_all(struct record *rec)
 {
+	u64 bytes_written = rec->bytes_written;
 	int i;
 	int rc = 0;
 
@@ -250,7 +251,11 @@ static int record__mmap_read_all(struct record *rec)
 		}
 	}
 
-	if (perf_header__has_feat(&rec->session->header, HEADER_TRACING_DATA))
+	/*
+	 * Mark the round finished in case we wrote
+	 * at least one event.
+	 */
+	if (bytes_written != rec->bytes_written)
 		rc = record__write(rec, &finished_round_event, sizeof(finished_round_event));
 
 out:

tools/perf/builtin-sched.c

@@ -10,6 +10,7 @@
 #include "util/header.h"
 #include "util/session.h"
 #include "util/tool.h"
+#include "util/cloexec.h"
 
 #include "util/parse-options.h"
 #include "util/trace-event.h"
@@ -434,7 +435,8 @@ static int self_open_counters(void)
 	attr.type = PERF_TYPE_SOFTWARE;
 	attr.config = PERF_COUNT_SW_TASK_CLOCK;
 
-	fd = sys_perf_event_open(&attr, 0, -1, -1, 0);
+	fd = sys_perf_event_open(&attr, 0, -1, -1,
+				 perf_event_open_cloexec_flag());
 
 	if (fd < 0)
 		pr_err("Error: sys_perf_event_open() syscall returned "
@@ -935,8 +937,8 @@ static int latency_switch_event(struct perf_sched *sched,
 		return -1;
 	}
 
-	sched_out = machine__findnew_thread(machine, 0, prev_pid);
-	sched_in = machine__findnew_thread(machine, 0, next_pid);
+	sched_out = machine__findnew_thread(machine, -1, prev_pid);
+	sched_in = machine__findnew_thread(machine, -1, next_pid);
 
 	out_events = thread_atoms_search(&sched->atom_root, sched_out, &sched->cmp_pid);
 	if (!out_events) {
@@ -979,7 +981,7 @@ static int latency_runtime_event(struct perf_sched *sched,
 {
 	const u32 pid	   = perf_evsel__intval(evsel, sample, "pid");
 	const u64 runtime  = perf_evsel__intval(evsel, sample, "runtime");
-	struct thread *thread = machine__findnew_thread(machine, 0, pid);
+	struct thread *thread = machine__findnew_thread(machine, -1, pid);
 	struct work_atoms *atoms = thread_atoms_search(&sched->atom_root, thread, &sched->cmp_pid);
 	u64 timestamp = sample->time;
 	int cpu = sample->cpu;
@@ -1012,7 +1014,7 @@ static int latency_wakeup_event(struct perf_sched *sched,
 	struct thread *wakee;
 	u64 timestamp = sample->time;
 
-	wakee = machine__findnew_thread(machine, 0, pid);
+	wakee = machine__findnew_thread(machine, -1, pid);
 	atoms = thread_atoms_search(&sched->atom_root, wakee, &sched->cmp_pid);
 	if (!atoms) {
 		if (thread_atoms_insert(sched, wakee))
@@ -1072,7 +1074,7 @@ static int latency_migrate_task_event(struct perf_sched *sched,
 	if (sched->profile_cpu == -1)
 		return 0;
 
-	migrant = machine__findnew_thread(machine, 0, pid);
+	migrant = machine__findnew_thread(machine, -1, pid);
 	atoms = thread_atoms_search(&sched->atom_root, migrant, &sched->cmp_pid);
 	if (!atoms) {
 		if (thread_atoms_insert(sched, migrant))
@@ -1290,7 +1292,7 @@ static int map_switch_event(struct perf_sched *sched, struct perf_evsel *evsel,
 		return -1;
 	}
 
-	sched_in = machine__findnew_thread(machine, 0, next_pid);
+	sched_in = machine__findnew_thread(machine, -1, next_pid);
 
 	sched->curr_thread[this_cpu] = sched_in;
 

tools/perf/builtin-script.c

@@ -358,27 +358,6 @@ static void print_sample_start(struct perf_sample *sample,
 	}
 }
 
-static bool is_bts_event(struct perf_event_attr *attr)
-{
-	return ((attr->type == PERF_TYPE_HARDWARE) &&
-		(attr->config & PERF_COUNT_HW_BRANCH_INSTRUCTIONS) &&
-		(attr->sample_period == 1));
-}
-
-static bool sample_addr_correlates_sym(struct perf_event_attr *attr)
-{
-	if ((attr->type == PERF_TYPE_SOFTWARE) &&
-	    ((attr->config == PERF_COUNT_SW_PAGE_FAULTS) ||
-	     (attr->config == PERF_COUNT_SW_PAGE_FAULTS_MIN) ||
-	     (attr->config == PERF_COUNT_SW_PAGE_FAULTS_MAJ)))
-		return true;
-
-	if (is_bts_event(attr))
-		return true;
-
-	return false;
-}
-
 static void print_sample_addr(union perf_event *event,
 			  struct perf_sample *sample,
 			  struct machine *machine,
@@ -386,24 +365,13 @@ static void print_sample_addr(union perf_event *event,
 			  struct perf_event_attr *attr)
 {
 	struct addr_location al;
-	u8 cpumode = event->header.misc & PERF_RECORD_MISC_CPUMODE_MASK;
 
 	printf("%16" PRIx64, sample->addr);
 
 	if (!sample_addr_correlates_sym(attr))
 		return;
 
-	thread__find_addr_map(thread, machine, cpumode, MAP__FUNCTION,
-			      sample->addr, &al);
-	if (!al.map)
-		thread__find_addr_map(thread, machine, cpumode, MAP__VARIABLE,
-				      sample->addr, &al);
-
-	al.cpu = sample->cpu;
-	al.sym = NULL;
-
-	if (al.map)
-		al.sym = map__find_symbol(al.map, al.addr, NULL);
+	perf_event__preprocess_sample_addr(event, sample, machine, thread, &al);
 
 	if (PRINT_FIELD(SYM)) {
 		printf(" ");
@@ -427,25 +395,35 @@ static void print_sample_bts(union perf_event *event,
 			     struct addr_location *al)
 {
 	struct perf_event_attr *attr = &evsel->attr;
+	bool print_srcline_last = false;
 
 	/* print branch_from information */
 	if (PRINT_FIELD(IP)) {
-		if (!symbol_conf.use_callchain)
-			printf(" ");
-		else
+		unsigned int print_opts = output[attr->type].print_ip_opts;
+
+		if (symbol_conf.use_callchain && sample->callchain) {
 			printf("\n");
-		perf_evsel__print_ip(evsel, sample, al,
-				     output[attr->type].print_ip_opts,
+		} else {
+			printf(" ");
+			if (print_opts & PRINT_IP_OPT_SRCLINE) {
+				print_srcline_last = true;
+				print_opts &= ~PRINT_IP_OPT_SRCLINE;
+			}
+		}
+		perf_evsel__print_ip(evsel, sample, al, print_opts,
 				     PERF_MAX_STACK_DEPTH);
 	}
 
-	printf(" => ");
-
 	/* print branch_to information */
 	if (PRINT_FIELD(ADDR) ||
 	    ((evsel->attr.sample_type & PERF_SAMPLE_ADDR) &&
-	     !output[attr->type].user_set))
+	     !output[attr->type].user_set)) {
+		printf(" => ");
 		print_sample_addr(event, sample, al->machine, thread, attr);
+	}
+
+	if (print_srcline_last)
+		map__fprintf_srcline(al->map, al->addr, "\n  ", stdout);
 
 	printf("\n");
 }

tools/perf/builtin-stat.c

@@ -184,7 +184,7 @@ static void perf_evsel__reset_stat_priv(struct perf_evsel *evsel)
 static int perf_evsel__alloc_stat_priv(struct perf_evsel *evsel)
 {
 	evsel->priv = zalloc(sizeof(struct perf_stat));
-	if (evsel == NULL)
+	if (evsel->priv == NULL)
 		return -ENOMEM;
 	perf_evsel__reset_stat_priv(evsel);
 	return 0;

tools/perf/config/Makefile

@@ -48,6 +48,10 @@ ifneq ($(ARCH),$(filter $(ARCH),x86 arm))
   NO_LIBDW_DWARF_UNWIND := 1
 endif
 
+ifeq ($(ARCH),powerpc)
+  CFLAGS += -DHAVE_SKIP_CALLCHAIN_IDX
+endif
+
 ifeq ($(LIBUNWIND_LIBS),)
   NO_LIBUNWIND := 1
 else
@@ -160,6 +164,7 @@ CORE_FEATURE_TESTS =			\
 	backtrace			\
 	dwarf				\
 	fortify-source			\
+	sync-compare-and-swap		\
 	glibc				\
 	gtk2				\
 	gtk2-infobar			\
@@ -195,6 +200,7 @@ LIB_FEATURE_TESTS =			\
 VF_FEATURE_TESTS =			\
 	backtrace			\
 	fortify-source			\
+	sync-compare-and-swap		\
 	gtk2-infobar			\
 	libelf-getphdrnum		\
 	libelf-mmap			\
@@ -268,6 +274,10 @@ CFLAGS += -I$(LIB_INCLUDE)
 
 CFLAGS += -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE
 
+ifeq ($(feature-sync-compare-and-swap), 1)
+  CFLAGS += -DHAVE_SYNC_COMPARE_AND_SWAP_SUPPORT
+endif
+
 ifndef NO_BIONIC
   $(call feature_check,bionic)
   ifeq ($(feature-bionic), 1)
@@ -590,6 +600,10 @@ ifndef NO_LIBNUMA
   endif
 endif
 
+ifdef HAVE_KVM_STAT_SUPPORT
+    CFLAGS += -DHAVE_KVM_STAT_SUPPORT
+endif
+
 # Among the variables below, these:
 #   perfexecdir
 #   template_dir

tools/perf/config/feature-checks/Makefile

@@ -5,6 +5,7 @@ FILES=					\
 	test-bionic.bin			\
 	test-dwarf.bin			\
 	test-fortify-source.bin		\
+	test-sync-compare-and-swap.bin	\
 	test-glibc.bin			\
 	test-gtk2.bin			\
 	test-gtk2-infobar.bin		\
@@ -141,6 +142,9 @@ test-timerfd.bin:
 test-libdw-dwarf-unwind.bin:
 	$(BUILD)
 
+test-sync-compare-and-swap.bin:
+	$(BUILD) -Werror
+
 -include *.d
 
 ###############################

tools/perf/config/feature-checks/test-all.c

@@ -89,6 +89,10 @@
 # include "test-libdw-dwarf-unwind.c"
 #undef main
 
+#define main main_test_sync_compare_and_swap
+# include "test-sync-compare-and-swap.c"
+#undef main
+
 int main(int argc, char *argv[])
 {
 	main_test_libpython();
@@ -111,6 +115,7 @@ int main(int argc, char *argv[])
 	main_test_timerfd();
 	main_test_stackprotector_all();
 	main_test_libdw_dwarf_unwind();
+	main_test_sync_compare_and_swap(argc, argv);
 
 	return 0;
 }

tools/perf/config/feature-checks/test-sync-compare-and-swap.c

+#include <stdint.h>
+
+volatile uint64_t x;
+
+int main(int argc, char *argv[])
+{
+	uint64_t old, new = argc;
+
+	argv = argv;
+	do {
+		old = __sync_val_compare_and_swap(&x, 0, 0);
+	} while (!__sync_bool_compare_and_swap(&x, old, new));
+	return old == new;
+}

tools/perf/perf-sys.h

@@ -54,6 +54,7 @@
 #define mb()		asm volatile("bcr 15,0" ::: "memory")
 #define wmb()		asm volatile("bcr 15,0" ::: "memory")
 #define rmb()		asm volatile("bcr 15,0" ::: "memory")
+#define CPUINFO_PROC	"vendor_id"
 #endif
 
 #ifdef __sh__

tools/perf/perf.c

@@ -13,11 +13,12 @@
 #include "util/quote.h"
 #include "util/run-command.h"
 #include "util/parse-events.h"
+#include "util/debug.h"
 #include <api/fs/debugfs.h>
 #include <pthread.h>
 
 const char perf_usage_string[] =
-	"perf [--version] [--help] COMMAND [ARGS]";
+	"perf [--version] [--help] [OPTIONS] COMMAND [ARGS]";
 
 const char perf_more_info_string[] =
 	"See 'perf help COMMAND' for more information on a specific command.";
@@ -212,6 +213,16 @@ static int handle_options(const char ***argv, int *argc, int *envchanged)
 				printf("%s ", p->cmd);
 			}
 			exit(0);
+		} else if (!strcmp(cmd, "--debug")) {
+			if (*argc < 2) {
+				fprintf(stderr, "No variable specified for --debug.\n");
+				usage(perf_usage_string);
+			}
+			if (perf_debug_option((*argv)[1]))
+				usage(perf_usage_string);
+
+			(*argv)++;
+			(*argc)--;
 		} else {
 			fprintf(stderr, "Unknown option: %s\n", cmd);
 			usage(perf_usage_string);

tools/perf/scripts/perl/bin/failed-syscalls-record

 #!/bin/bash
-perf record -e raw_syscalls:sys_exit $@
+(perf record -e raw_syscalls:sys_exit $@ || \
+ perf record -e syscalls:sys_exit $@) 2> /dev/null

tools/perf/scripts/perl/failed-syscalls.pl

@@ -26,6 +26,11 @@ sub raw_syscalls::sys_exit
 	}
 }
 
+sub syscalls::sys_exit
+{
+	raw_syscalls::sys_exit(@_)
+}
+
 sub trace_end
 {
     printf("\nfailed syscalls by comm:\n\n");

tools/perf/scripts/python/Perf-Trace-Util/lib/Perf/Trace/Core.py

@@ -107,12 +107,13 @@ def taskState(state):
 
 class EventHeaders:
 	def __init__(self, common_cpu, common_secs, common_nsecs,
-		     common_pid, common_comm):
+		     common_pid, common_comm, common_callchain):
 		self.cpu = common_cpu
 		self.secs = common_secs
 		self.nsecs = common_nsecs
 		self.pid = common_pid
 		self.comm = common_comm
+		self.callchain = common_callchain
 
 	def ts(self):
 		return (self.secs * (10 ** 9)) + self.nsecs

tools/perf/scripts/python/bin/failed-syscalls-by-pid-record

 #!/bin/bash
-perf record -e raw_syscalls:sys_exit $@
+(perf record -e raw_syscalls:sys_exit $@ || \
+ perf record -e syscalls:sys_exit $@) 2> /dev/null

tools/perf/scripts/python/bin/sctop-record

 #!/bin/bash
-perf record -e raw_syscalls:sys_enter $@
+(perf record -e raw_syscalls:sys_enter $@ || \
+ perf record -e syscalls:sys_enter $@) 2> /dev/null

tools/perf/scripts/python/bin/syscall-counts-by-pid-record

 #!/bin/bash
-perf record -e raw_syscalls:sys_enter $@
+(perf record -e raw_syscalls:sys_enter $@ || \
+ perf record -e syscalls:sys_enter $@) 2> /dev/null

tools/perf/scripts/python/bin/syscall-counts-record

 #!/bin/bash
-perf record -e raw_syscalls:sys_enter $@
+(perf record -e raw_syscalls:sys_enter $@ || \
+ perf record -e syscalls:sys_enter $@) 2> /dev/null

tools/perf/scripts/python/check-perf-trace.py

@@ -27,7 +27,7 @@ def trace_end():
 
 def irq__softirq_entry(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	vec):
+	common_callchain, vec):
 		print_header(event_name, common_cpu, common_secs, common_nsecs,
 			common_pid, common_comm)
 
@@ -38,7 +38,7 @@ def irq__softirq_entry(event_name, context, common_cpu,
 
 def kmem__kmalloc(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	call_site, ptr, bytes_req, bytes_alloc,
+	common_callchain, call_site, ptr, bytes_req, bytes_alloc,
 	gfp_flags):
 		print_header(event_name, common_cpu, common_secs, common_nsecs,
 			common_pid, common_comm)

tools/perf/scripts/python/failed-syscalls-by-pid.py

@@ -39,7 +39,7 @@ def trace_end():
 
 def raw_syscalls__sys_exit(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	id, ret):
+	common_callchain, id, ret):
 	if (for_comm and common_comm != for_comm) or \
 	   (for_pid  and common_pid  != for_pid ):
 		return
@@ -50,6 +50,11 @@ def raw_syscalls__sys_exit(event_name, context, common_cpu,
 		except TypeError:
 			syscalls[common_comm][common_pid][id][ret] = 1
 
+def syscalls__sys_exit(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	id, ret):
+	raw_syscalls__sys_exit(**locals())
+
 def print_error_totals():
     if for_comm is not None:
 	    print "\nsyscall errors for %s:\n\n" % (for_comm),

tools/perf/scripts/python/futex-contention.py

@@ -21,7 +21,7 @@ thread_blocktime = {}
 lock_waits = {} # long-lived stats on (tid,lock) blockage elapsed time
 process_names = {} # long-lived pid-to-execname mapping
 
-def syscalls__sys_enter_futex(event, ctxt, cpu, s, ns, tid, comm,
+def syscalls__sys_enter_futex(event, ctxt, cpu, s, ns, tid, comm, callchain,
 			      nr, uaddr, op, val, utime, uaddr2, val3):
 	cmd = op & FUTEX_CMD_MASK
 	if cmd != FUTEX_WAIT:
@@ -31,7 +31,7 @@ def syscalls__sys_enter_futex(event, ctxt, cpu, s, ns, tid, comm,
 	thread_thislock[tid] = uaddr
 	thread_blocktime[tid] = nsecs(s, ns)
 
-def syscalls__sys_exit_futex(event, ctxt, cpu, s, ns, tid, comm,
+def syscalls__sys_exit_futex(event, ctxt, cpu, s, ns, tid, comm, callchain,
 			     nr, ret):
 	if thread_blocktime.has_key(tid):
 		elapsed = nsecs(s, ns) - thread_blocktime[tid]

tools/perf/scripts/python/net_dropmonitor.py

@@ -66,7 +66,7 @@ def trace_end():
 	print_drop_table()
 
 # called from perf, when it finds a correspoinding event
-def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm,
+def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm, callchain,
 		   skbaddr, location, protocol):
 	slocation = str(location)
 	try:

tools/perf/scripts/python/netdev-times.py

@@ -224,75 +224,75 @@ def trace_end():
 			(len(rx_skb_list), of_count_rx_skb_list)
 
 # called from perf, when it finds a correspoinding event
-def irq__softirq_entry(name, context, cpu, sec, nsec, pid, comm, vec):
+def irq__softirq_entry(name, context, cpu, sec, nsec, pid, comm, callchain, vec):
 	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
 		return
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
 	all_event_list.append(event_info)
 
-def irq__softirq_exit(name, context, cpu, sec, nsec, pid, comm, vec):
+def irq__softirq_exit(name, context, cpu, sec, nsec, pid, comm, callchain, vec):
 	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
 		return
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
 	all_event_list.append(event_info)
 
-def irq__softirq_raise(name, context, cpu, sec, nsec, pid, comm, vec):
+def irq__softirq_raise(name, context, cpu, sec, nsec, pid, comm, callchain, vec):
 	if symbol_str("irq__softirq_entry", "vec", vec) != "NET_RX":
 		return
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, vec)
 	all_event_list.append(event_info)
 
 def irq__irq_handler_entry(name, context, cpu, sec, nsec, pid, comm,
-			irq, irq_name):
+			callchain, irq, irq_name):
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
 			irq, irq_name)
 	all_event_list.append(event_info)
 
-def irq__irq_handler_exit(name, context, cpu, sec, nsec, pid, comm, irq, ret):
+def irq__irq_handler_exit(name, context, cpu, sec, nsec, pid, comm, callchain, irq, ret):
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm, irq, ret)
 	all_event_list.append(event_info)
 
-def napi__napi_poll(name, context, cpu, sec, nsec, pid, comm, napi, dev_name):
+def napi__napi_poll(name, context, cpu, sec, nsec, pid, comm, callchain, napi, dev_name):
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
 			napi, dev_name)
 	all_event_list.append(event_info)
 
-def net__netif_receive_skb(name, context, cpu, sec, nsec, pid, comm, skbaddr,
+def net__netif_receive_skb(name, context, cpu, sec, nsec, pid, comm, callchain, skbaddr,
 			skblen, dev_name):
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
 			skbaddr, skblen, dev_name)
 	all_event_list.append(event_info)
 
-def net__netif_rx(name, context, cpu, sec, nsec, pid, comm, skbaddr,
+def net__netif_rx(name, context, cpu, sec, nsec, pid, comm, callchain, skbaddr,
 			skblen, dev_name):
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
 			skbaddr, skblen, dev_name)
 	all_event_list.append(event_info)
 
-def net__net_dev_queue(name, context, cpu, sec, nsec, pid, comm,
+def net__net_dev_queue(name, context, cpu, sec, nsec, pid, comm, callchain,
 			skbaddr, skblen, dev_name):
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
 			skbaddr, skblen, dev_name)
 	all_event_list.append(event_info)
 
-def net__net_dev_xmit(name, context, cpu, sec, nsec, pid, comm,
+def net__net_dev_xmit(name, context, cpu, sec, nsec, pid, comm, callchain,
 			skbaddr, skblen, rc, dev_name):
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
 			skbaddr, skblen, rc ,dev_name)
 	all_event_list.append(event_info)
 
-def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm,
+def skb__kfree_skb(name, context, cpu, sec, nsec, pid, comm, callchain,
 			skbaddr, protocol, location):
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
 			skbaddr, protocol, location)
 	all_event_list.append(event_info)
 
-def skb__consume_skb(name, context, cpu, sec, nsec, pid, comm, skbaddr):
+def skb__consume_skb(name, context, cpu, sec, nsec, pid, comm, callchain, skbaddr):
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
 			skbaddr)
 	all_event_list.append(event_info)
 
-def skb__skb_copy_datagram_iovec(name, context, cpu, sec, nsec, pid, comm,
+def skb__skb_copy_datagram_iovec(name, context, cpu, sec, nsec, pid, comm, callchain,
 	skbaddr, skblen):
 	event_info = (name, context, cpu, nsecs(sec, nsec), pid, comm,
 			skbaddr, skblen)

tools/perf/scripts/python/sched-migration.py

@@ -369,93 +369,92 @@ def trace_end():
 
 def sched__sched_stat_runtime(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid, runtime, vruntime):
+	common_callchain, comm, pid, runtime, vruntime):
 	pass
 
 def sched__sched_stat_iowait(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid, delay):
+	common_callchain, comm, pid, delay):
 	pass
 
 def sched__sched_stat_sleep(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid, delay):
+	common_callchain, comm, pid, delay):
 	pass
 
 def sched__sched_stat_wait(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid, delay):
+	common_callchain, comm, pid, delay):
 	pass
 
 def sched__sched_process_fork(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	parent_comm, parent_pid, child_comm, child_pid):
+	common_callchain, parent_comm, parent_pid, child_comm, child_pid):
 	pass
 
 def sched__sched_process_wait(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid, prio):
+	common_callchain, comm, pid, prio):
 	pass
 
 def sched__sched_process_exit(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid, prio):
+	common_callchain, comm, pid, prio):
 	pass
 
 def sched__sched_process_free(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid, prio):
+	common_callchain, comm, pid, prio):
 	pass
 
 def sched__sched_migrate_task(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid, prio, orig_cpu,
+	common_callchain, comm, pid, prio, orig_cpu,
 	dest_cpu):
 	headers = EventHeaders(common_cpu, common_secs, common_nsecs,
-				common_pid, common_comm)
+				common_pid, common_comm, common_callchain)
 	parser.migrate(headers, pid, prio, orig_cpu, dest_cpu)
 
 def sched__sched_switch(event_name, context, common_cpu,
-	common_secs, common_nsecs, common_pid, common_comm,
+	common_secs, common_nsecs, common_pid, common_comm, common_callchain,
 	prev_comm, prev_pid, prev_prio, prev_state,
 	next_comm, next_pid, next_prio):
 
 	headers = EventHeaders(common_cpu, common_secs, common_nsecs,
-				common_pid, common_comm)
+				common_pid, common_comm, common_callchain)
 	parser.sched_switch(headers, prev_comm, prev_pid, prev_prio, prev_state,
 			 next_comm, next_pid, next_prio)
 
 def sched__sched_wakeup_new(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid, prio, success,
+	common_callchain, comm, pid, prio, success,
 	target_cpu):
 	headers = EventHeaders(common_cpu, common_secs, common_nsecs,
-				common_pid, common_comm)
+				common_pid, common_comm, common_callchain)
 	parser.wake_up(headers, comm, pid, success, target_cpu, 1)
 
 def sched__sched_wakeup(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid, prio, success,
+	common_callchain, comm, pid, prio, success,
 	target_cpu):
 	headers = EventHeaders(common_cpu, common_secs, common_nsecs,
-				common_pid, common_comm)
+				common_pid, common_comm, common_callchain)
 	parser.wake_up(headers, comm, pid, success, target_cpu, 0)
 
 def sched__sched_wait_task(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid, prio):
+	common_callchain, comm, pid, prio):
 	pass
 
 def sched__sched_kthread_stop_ret(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	ret):
+	common_callchain, ret):
 	pass
 
 def sched__sched_kthread_stop(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	comm, pid):
+	common_callchain, comm, pid):
 	pass
 
-def trace_unhandled(event_name, context, common_cpu, common_secs, common_nsecs,
-		common_pid, common_comm):
+def trace_unhandled(event_name, context, event_fields_dict):
 	pass

tools/perf/scripts/python/sctop.py

@@ -44,7 +44,7 @@ def trace_begin():
 
 def raw_syscalls__sys_enter(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	id, args):
+	common_callchain, id, args):
 	if for_comm is not None:
 		if common_comm != for_comm:
 			return
@@ -53,6 +53,11 @@ def raw_syscalls__sys_enter(event_name, context, common_cpu,
 	except TypeError:
 		syscalls[id] = 1
 
+def syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	id, args):
+	raw_syscalls__sys_enter(**locals())
+
 def print_syscall_totals(interval):
 	while 1:
 		clear_term()

tools/perf/scripts/python/syscall-counts-by-pid.py

@@ -38,7 +38,7 @@ def trace_end():
 
 def raw_syscalls__sys_enter(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	id, args):
+	common_callchain, id, args):
 
 	if (for_comm and common_comm != for_comm) or \
 	   (for_pid  and common_pid  != for_pid ):
@@ -48,6 +48,11 @@ def raw_syscalls__sys_enter(event_name, context, common_cpu,
 	except TypeError:
 		syscalls[common_comm][common_pid][id] = 1
 
+def syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	id, args):
+	raw_syscalls__sys_enter(**locals())
+
 def print_syscall_totals():
     if for_comm is not None:
 	    print "\nsyscall events for %s:\n\n" % (for_comm),

tools/perf/scripts/python/syscall-counts.py

@@ -35,7 +35,7 @@ def trace_end():
 
 def raw_syscalls__sys_enter(event_name, context, common_cpu,
 	common_secs, common_nsecs, common_pid, common_comm,
-	id, args):
+	common_callchain, id, args):
 	if for_comm is not None:
 		if common_comm != for_comm:
 			return
@@ -44,6 +44,11 @@ def raw_syscalls__sys_enter(event_name, context, common_cpu,
 	except TypeError:
 		syscalls[id] = 1
 
+def syscalls__sys_enter(event_name, context, common_cpu,
+	common_secs, common_nsecs, common_pid, common_comm,
+	id, args):
+	raw_syscalls__sys_enter(**locals())
+
 def print_syscall_totals():
     if for_comm is not None:
 	    print "\nsyscall events for %s:\n\n" % (for_comm),

tools/perf/tests/attr/base-record

 [event]
 fd=1
 group_fd=-1
-flags=0
+# 0 or PERF_FLAG_FD_CLOEXEC flag
+flags=0|8
 cpu=*
 type=0|1
 size=96

tools/perf/tests/attr/base-stat

 [event]
 fd=1
 group_fd=-1
-flags=0
+# 0 or PERF_FLAG_FD_CLOEXEC flag
+flags=0|8
 cpu=*
 type=0
 size=96

tools/perf/tests/bp_signal.c

@@ -25,6 +25,7 @@
 #include "tests.h"
 #include "debug.h"
 #include "perf.h"
+#include "cloexec.h"
 
 static int fd1;
 static int fd2;
@@ -78,7 +79,8 @@ static int bp_event(void *fn, int setup_signal)
 	pe.exclude_kernel = 1;
 	pe.exclude_hv = 1;
 
-	fd = sys_perf_event_open(&pe, 0, -1, -1, 0);
+	fd = sys_perf_event_open(&pe, 0, -1, -1,
+				 perf_event_open_cloexec_flag());
 	if (fd < 0) {
 		pr_debug("failed opening event %llx\n", pe.config);
 		return TEST_FAIL;

tools/perf/tests/bp_signal_overflow.c

@@ -24,6 +24,7 @@
 #include "tests.h"
 #include "debug.h"
 #include "perf.h"
+#include "cloexec.h"
 
 static int overflows;
 
@@ -91,7 +92,8 @@ int test__bp_signal_overflow(void)
 	pe.exclude_kernel = 1;
 	pe.exclude_hv = 1;
 
-	fd = sys_perf_event_open(&pe, 0, -1, -1, 0);
+	fd = sys_perf_event_open(&pe, 0, -1, -1,
+				 perf_event_open_cloexec_flag());
 	if (fd < 0) {
 		pr_debug("failed opening event %llx\n", pe.config);
 		return TEST_FAIL;

tools/perf/tests/dso-data.c

@@ -10,6 +10,7 @@
 #include "machine.h"
 #include "symbol.h"
 #include "tests.h"
+#include "debug.h"
 
 static char *test_file(int size)
 {

tools/perf/tests/evsel-roundtrip-name.c

@@ -2,6 +2,7 @@
 #include "evsel.h"
 #include "parse-events.h"
 #include "tests.h"
+#include "debug.h"
 
 static int perf_evsel__roundtrip_cache_name_test(void)
 {

tools/perf/tests/evsel-tp-sched.c

 #include <traceevent/event-parse.h>
 #include "evsel.h"
 #include "tests.h"
+#include "debug.h"
 
 static int perf_evsel__test_field(struct perf_evsel *evsel, const char *name,
 				  int size, bool should_be_signed)

tools/perf/tests/open-syscall-tp-fields.c

@@ -3,6 +3,7 @@
 #include "evsel.h"
 #include "thread_map.h"
 #include "tests.h"
+#include "debug.h"
 
 int test__syscall_open_tp_fields(void)
 {

tools/perf/tests/parse-events.c

@@ -5,6 +5,7 @@
 #include <api/fs/fs.h>
 #include <api/fs/debugfs.h>
 #include "tests.h"
+#include "debug.h"
 #include <linux/hw_breakpoint.h>
 
 #define PERF_TP_SAMPLE_TYPE (PERF_SAMPLE_RAW | PERF_SAMPLE_TIME | \

tools/perf/tests/parse-no-sample-id-all.c

@@ -7,6 +7,7 @@
 #include "evlist.h"
 #include "header.h"
 #include "util.h"
+#include "debug.h"
 
 static int process_event(struct perf_evlist **pevlist, union perf_event *event)
 {

tools/perf/tests/perf-time-to-tsc.c

@@ -8,10 +8,9 @@
 #include "evsel.h"
 #include "thread_map.h"
 #include "cpumap.h"
+#include "tsc.h"
 #include "tests.h"
 
-#include "../arch/x86/util/tsc.h"
-
 #define CHECK__(x) {				\
 	while ((x) < 0) {			\
 		pr_debug(#x " failed!\n");	\
@@ -26,15 +25,6 @@
 	}					\
 }
 
-static u64 rdtsc(void)
-{
-	unsigned int low, high;
-
-	asm volatile("rdtsc" : "=a" (low), "=d" (high));
-
-	return low | ((u64)high) << 32;
-}
-
 /**
  * test__perf_time_to_tsc - test converting perf time to TSC.
  *

tools/perf/tests/rdpmc.c

@@ -6,6 +6,7 @@
 #include "perf.h"
 #include "debug.h"
 #include "tests.h"
+#include "cloexec.h"
 
 #if defined(__x86_64__) || defined(__i386__)
 
@@ -104,7 +105,8 @@ static int __test__rdpmc(void)
 	sa.sa_sigaction = segfault_handler;
 	sigaction(SIGSEGV, &sa, NULL);
 
-	fd = sys_perf_event_open(&attr, 0, -1, -1, 0);
+	fd = sys_perf_event_open(&attr, 0, -1, -1,
+				 perf_event_open_cloexec_flag());
 	if (fd < 0) {
 		pr_err("Error: sys_perf_event_open() syscall returned "
 		       "with %d (%s)\n", fd, strerror(errno));

tools/perf/tests/sample-parsing.c

@@ -4,6 +4,7 @@
 #include "util.h"
 #include "event.h"
 #include "evsel.h"
+#include "debug.h"
 
 #include "tests.h"
 

tools/perf/tests/thread-mg-share.c

@@ -2,6 +2,7 @@
 #include "machine.h"
 #include "thread.h"
 #include "map.h"
+#include "debug.h"
 
 int test__thread_mg_share(void)
 {

tools/perf/ui/browser.c

@@ -150,7 +150,7 @@ unsigned int ui_browser__rb_tree_refresh(struct ui_browser *browser)
 	while (nd != NULL) {
 		ui_browser__gotorc(browser, row, 0);
 		browser->write(browser, nd, row);
-		if (++row == browser->height)
+		if (++row == browser->rows)
 			break;
 		nd = rb_next(nd);
 	}
@@ -166,7 +166,7 @@ bool ui_browser__is_current_entry(struct ui_browser *browser, unsigned row)
 void ui_browser__refresh_dimensions(struct ui_browser *browser)
 {
 	browser->width = SLtt_Screen_Cols - 1;
-	browser->height = SLtt_Screen_Rows - 2;
+	browser->height = browser->rows = SLtt_Screen_Rows - 2;
 	browser->y = 1;
 	browser->x = 0;
 }
@@ -250,7 +250,10 @@ int ui_browser__show(struct ui_browser *browser, const char *title,
 	int err;
 	va_list ap;
 
-	ui_browser__refresh_dimensions(browser);
+	if (browser->refresh_dimensions == NULL)
+		browser->refresh_dimensions = ui_browser__refresh_dimensions;
+
+	browser->refresh_dimensions(browser);
 
 	pthread_mutex_lock(&ui__lock);
 	__ui_browser__show_title(browser, title);
@@ -279,7 +282,7 @@ static void ui_browser__scrollbar_set(struct ui_browser *browser)
 {
 	int height = browser->height, h = 0, pct = 0,
 	    col = browser->width,
-	    row = browser->y - 1;
+	    row = 0;
 
 	if (browser->nr_entries > 1) {
 		pct = ((browser->index * (browser->height - 1)) /
@@ -367,7 +370,7 @@ int ui_browser__run(struct ui_browser *browser, int delay_secs)
 
 		if (key == K_RESIZE) {
 			ui__refresh_dimensions(false);
-			ui_browser__refresh_dimensions(browser);
+			browser->refresh_dimensions(browser);
 			__ui_browser__show_title(browser, browser->title);
 			ui_helpline__puts(browser->helpline);
 			continue;
@@ -389,7 +392,7 @@ int ui_browser__run(struct ui_browser *browser, int delay_secs)
 			if (browser->index == browser->nr_entries - 1)
 				break;
 			++browser->index;
-			if (browser->index == browser->top_idx + browser->height) {
+			if (browser->index == browser->top_idx + browser->rows) {
 				++browser->top_idx;
 				browser->seek(browser, +1, SEEK_CUR);
 			}
@@ -405,10 +408,10 @@ int ui_browser__run(struct ui_browser *browser, int delay_secs)
 			break;
 		case K_PGDN:
 		case ' ':
-			if (browser->top_idx + browser->height > browser->nr_entries - 1)
+			if (browser->top_idx + browser->rows > browser->nr_entries - 1)
 				break;
 
-			offset = browser->height;
+			offset = browser->rows;
 			if (browser->index + offset > browser->nr_entries - 1)
 				offset = browser->nr_entries - 1 - browser->index;
 			browser->index += offset;
@@ -419,10 +422,10 @@ int ui_browser__run(struct ui_browser *browser, int delay_secs)
 			if (browser->top_idx == 0)
 				break;
 
-			if (browser->top_idx < browser->height)
+			if (browser->top_idx < browser->rows)
 				offset = browser->top_idx;
 			else
-				offset = browser->height;
+				offset = browser->rows;
 
 			browser->index -= offset;
 			browser->top_idx -= offset;
@@ -432,7 +435,7 @@ int ui_browser__run(struct ui_browser *browser, int delay_secs)
 			ui_browser__reset_index(browser);
 			break;
 		case K_END:
-			offset = browser->height - 1;
+			offset = browser->rows - 1;
 			if (offset >= browser->nr_entries)
 				offset = browser->nr_entries - 1;
 
@@ -462,7 +465,7 @@ unsigned int ui_browser__list_head_refresh(struct ui_browser *browser)
 		if (!browser->filter || !browser->filter(browser, pos)) {
 			ui_browser__gotorc(browser, row, 0);
 			browser->write(browser, pos, row);
-			if (++row == browser->height)
+			if (++row == browser->rows)
 				break;
 		}
 	}
@@ -587,7 +590,7 @@ unsigned int ui_browser__argv_refresh(struct ui_browser *browser)
 		if (!browser->filter || !browser->filter(browser, *pos)) {
 			ui_browser__gotorc(browser, row, 0);
 			browser->write(browser, pos, row);
-			if (++row == browser->height)
+			if (++row == browser->rows)
 				break;
 		}
 
@@ -623,7 +626,7 @@ static void __ui_browser__line_arrow_up(struct ui_browser *browser,
 
 	SLsmg_set_char_set(1);
 
-	if (start < browser->top_idx + browser->height) {
+	if (start < browser->top_idx + browser->rows) {
 		row = start - browser->top_idx;
 		ui_browser__gotorc(browser, row, column);
 		SLsmg_write_char(SLSMG_LLCORN_CHAR);
@@ -633,7 +636,7 @@ static void __ui_browser__line_arrow_up(struct ui_browser *browser,
 		if (row-- == 0)
 			goto out;
 	} else
-		row = browser->height - 1;
+		row = browser->rows - 1;
 
 	if (end > browser->top_idx)
 		end_row = end - browser->top_idx;
@@ -675,8 +678,8 @@ static void __ui_browser__line_arrow_down(struct ui_browser *browser,
 	} else
 		row = 0;
 
-	if (end >= browser->top_idx + browser->height)
-		end_row = browser->height - 1;
+	if (end >= browser->top_idx + browser->rows)
+		end_row = browser->rows - 1;
 	else
 		end_row = end - browser->top_idx;
 
@@ -684,7 +687,7 @@ static void __ui_browser__line_arrow_down(struct ui_browser *browser,
 	SLsmg_draw_vline(end_row - row + 1);
 
 	ui_browser__gotorc(browser, end_row, column);
-	if (end < browser->top_idx + browser->height) {
+	if (end < browser->top_idx + browser->rows) {
 		SLsmg_write_char(SLSMG_LLCORN_CHAR);
 		ui_browser__gotorc(browser, end_row, column + 1);
 		SLsmg_write_char(SLSMG_HLINE_CHAR);

tools/perf/ui/browser.h

@@ -14,11 +14,12 @@
 struct ui_browser {
 	u64	      index, top_idx;
 	void	      *top, *entries;
-	u16	      y, x, width, height;
+	u16	      y, x, width, height, rows;
 	int	      current_color;
 	void	      *priv;
 	const char    *title;
 	char	      *helpline;
+	void 	      (*refresh_dimensions)(struct ui_browser *browser);
 	unsigned int  (*refresh)(struct ui_browser *browser);
 	void	      (*write)(struct ui_browser *browser, void *entry, int row);
 	void	      (*seek)(struct ui_browser *browser, off_t offset, int whence);

tools/perf/ui/stdio/hist.c

@@ -479,7 +479,7 @@ size_t hists__fprintf(struct hists *hists, bool show_header, int max_rows,
 
 		if (h->ms.map == NULL && verbose > 1) {
 			__map_groups__fprintf_maps(h->thread->mg,
-						   MAP__FUNCTION, verbose, fp);
+						   MAP__FUNCTION, fp);
 			fprintf(fp, "%.10s end\n", graph_dotted_line);
 		}
 	}

tools/perf/util/callchain.c

@@ -626,7 +626,7 @@ int sample__resolve_callchain(struct perf_sample *sample, struct symbol **parent
 
 int hist_entry__append_callchain(struct hist_entry *he, struct perf_sample *sample)
 {
-	if (!symbol_conf.use_callchain)
+	if (!symbol_conf.use_callchain || sample->callchain == NULL)
 		return 0;
 	return callchain_append(he->callchain, &callchain_cursor, sample->period);
 }

tools/perf/util/callchain.h

@@ -176,4 +176,17 @@ static inline void callchain_cursor_snapshot(struct callchain_cursor *dest,
 	dest->first = src->curr;
 	dest->nr -= src->pos;
 }
+
+#ifdef HAVE_SKIP_CALLCHAIN_IDX
+extern int arch_skip_callchain_idx(struct machine *machine,
+			struct thread *thread, struct ip_callchain *chain);
+#else
+static inline int arch_skip_callchain_idx(struct machine *machine __maybe_unused,
+			struct thread *thread __maybe_unused,
+			struct ip_callchain *chain __maybe_unused)
+{
+	return -1;
+}
+#endif
+
 #endif	/* __PERF_CALLCHAIN_H */

tools/perf/util/cloexec.c

+#include "util.h"
+#include "../perf.h"
+#include "cloexec.h"
+#include "asm/bug.h"
+
+static unsigned long flag = PERF_FLAG_FD_CLOEXEC;
+
+static int perf_flag_probe(void)
+{
+	/* use 'safest' configuration as used in perf_evsel__fallback() */
+	struct perf_event_attr attr = {
+		.type = PERF_COUNT_SW_CPU_CLOCK,
+		.config = PERF_COUNT_SW_CPU_CLOCK,
+	};
+	int fd;
+	int err;
+
+	/* check cloexec flag */
+	fd = sys_perf_event_open(&attr, 0, -1, -1,
+				 PERF_FLAG_FD_CLOEXEC);
+	err = errno;
+
+	if (fd >= 0) {
+		close(fd);
+		return 1;
+	}
+
+	WARN_ONCE(err != EINVAL,
+		  "perf_event_open(..., PERF_FLAG_FD_CLOEXEC) failed with unexpected error %d (%s)\n",
+		  err, strerror(err));
+
+	/* not supported, confirm error related to PERF_FLAG_FD_CLOEXEC */
+	fd = sys_perf_event_open(&attr, 0, -1, -1, 0);
+	err = errno;
+
+	if (WARN_ONCE(fd < 0,
+		      "perf_event_open(..., 0) failed unexpectedly with error %d (%s)\n",
+		      err, strerror(err)))
+		return -1;
+
+	close(fd);
+
+	return 0;
+}
+
+unsigned long perf_event_open_cloexec_flag(void)
+{
+	static bool probed;
+
+	if (!probed) {
+		if (perf_flag_probe() <= 0)
+			flag = 0;
+		probed = true;
+	}
+
+	return flag;
+}