Merge branch 'devel-stable' into for-linus

Conflicts:
	drivers/perf/arm_pmu.c

9 files changed, 1336 insertions, 5 deletions

diff --git a/drivers/Kconfig b/drivers/Kconfig
index 6e973b8e3a3b..3497485f5eab 100644
--- a/drivers/Kconfig
+++ b/drivers/Kconfig
@@ -176,6 +176,8 @@ source "drivers/powercap/Kconfig"
 
 source "drivers/mcb/Kconfig"
 
+source "drivers/perf/Kconfig"
+
 source "drivers/ras/Kconfig"
 
 source "drivers/thunderbolt/Kconfig"
diff --git a/drivers/Makefile b/drivers/Makefile
index b64b49f6e01b..f245f2291b8a 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -161,6 +161,7 @@ obj-$(CONFIG_NTB)		+= ntb/
 obj-$(CONFIG_FMC)		+= fmc/
 obj-$(CONFIG_POWERCAP)		+= powercap/
 obj-$(CONFIG_MCB)		+= mcb/
+obj-$(CONFIG_PERF_EVENTS)	+= perf/
 obj-$(CONFIG_RAS)		+= ras/
 obj-$(CONFIG_THUNDERBOLT)	+= thunderbolt/
 obj-$(CONFIG_CORESIGHT)		+= hwtracing/coresight/
diff --git a/drivers/cpuidle/cpuidle-calxeda.c b/drivers/cpuidle/cpuidle-calxeda.c
index c13feec89ea1..ea9728fde9b3 100644
--- a/drivers/cpuidle/cpuidle-calxeda.c
+++ b/drivers/cpuidle/cpuidle-calxeda.c
@@ -25,16 +25,21 @@
 #include <linux/init.h>
 #include <linux/mm.h>
 #include <linux/platform_device.h>
+#include <linux/psci.h>
+
 #include <asm/cpuidle.h>
 #include <asm/suspend.h>
-#include <asm/psci.h>
+
+#include <uapi/linux/psci.h>
+
+#define CALXEDA_IDLE_PARAM \
+	((0 << PSCI_0_2_POWER_STATE_ID_SHIFT) | \
+	 (0 << PSCI_0_2_POWER_STATE_AFFL_SHIFT) | \
+	 (PSCI_POWER_STATE_TYPE_POWER_DOWN << PSCI_0_2_POWER_STATE_TYPE_SHIFT))
 
 static int calxeda_idle_finish(unsigned long val)
 {
-	const struct psci_power_state ps = {
-		.type = PSCI_POWER_STATE_TYPE_POWER_DOWN,
-	};
-	return psci_ops.cpu_suspend(ps, __pa(cpu_resume));
+	return psci_ops.cpu_suspend(CALXEDA_IDLE_PARAM, __pa(cpu_resume));
 }
 
 static int calxeda_pwrdown_idle(struct cpuidle_device *dev,
diff --git a/drivers/firmware/Kconfig b/drivers/firmware/Kconfig
index 99c69a3205c4..d8de6a8dd4de 100644
--- a/drivers/firmware/Kconfig
+++ b/drivers/firmware/Kconfig
@@ -5,6 +5,9 @@
 
 menu "Firmware Drivers"
 
+config ARM_PSCI_FW
+	bool
+
 config EDD
 	tristate "BIOS Enhanced Disk Drive calls determine boot disk"
 	depends on X86
diff --git a/drivers/firmware/Makefile b/drivers/firmware/Makefile
index 4a4b897f9314..000830fc6707 100644
--- a/drivers/firmware/Makefile
+++ b/drivers/firmware/Makefile
@@ -1,6 +1,7 @@
 #
 # Makefile for the linux kernel.
 #
+obj-$(CONFIG_ARM_PSCI_FW)	+= psci.o
 obj-$(CONFIG_DMI)		+= dmi_scan.o
 obj-$(CONFIG_DMI_SYSFS)		+= dmi-sysfs.o
 obj-$(CONFIG_EDD)		+= edd.o
diff --git a/drivers/firmware/psci.c b/drivers/firmware/psci.c
new file mode 100644
index 000000000000..a6956007dd38
--- /dev/null
+++ b/drivers/firmware/psci.c
@@ -0,0 +1,382 @@
+/*
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * Copyright (C) 2015 ARM Limited
+ */
+
+#define pr_fmt(fmt) "psci: " fmt
+
+#include <linux/errno.h>
+#include <linux/linkage.h>
+#include <linux/of.h>
+#include <linux/pm.h>
+#include <linux/printk.h>
+#include <linux/psci.h>
+#include <linux/reboot.h>
+
+#include <uapi/linux/psci.h>
+
+#include <asm/cputype.h>
+#include <asm/system_misc.h>
+#include <asm/smp_plat.h>
+
+/*
+ * While a 64-bit OS can make calls with SMC32 calling conventions, for some
+ * calls it is necessary to use SMC64 to pass or return 64-bit values. For such
+ * calls PSCI_0_2_FN_NATIVE(x) will choose the appropriate (native-width)
+ * function ID.
+ */
+#ifdef CONFIG_64BIT
+#define PSCI_0_2_FN_NATIVE(name)	PSCI_0_2_FN64_##name
+#else
+#define PSCI_0_2_FN_NATIVE(name)	PSCI_0_2_FN_##name
+#endif
+
+/*
+ * The CPU any Trusted OS is resident on. The trusted OS may reject CPU_OFF
+ * calls to its resident CPU, so we must avoid issuing those. We never migrate
+ * a Trusted OS even if it claims to be capable of migration -- doing so will
+ * require cooperation with a Trusted OS driver.
+ */
+static int resident_cpu = -1;
+
+bool psci_tos_resident_on(int cpu)
+{
+	return cpu == resident_cpu;
+}
+
+struct psci_operations psci_ops;
+
+typedef unsigned long (psci_fn)(unsigned long, unsigned long,
+				unsigned long, unsigned long);
+asmlinkage psci_fn __invoke_psci_fn_hvc;
+asmlinkage psci_fn __invoke_psci_fn_smc;
+static psci_fn *invoke_psci_fn;
+
+enum psci_function {
+	PSCI_FN_CPU_SUSPEND,
+	PSCI_FN_CPU_ON,
+	PSCI_FN_CPU_OFF,
+	PSCI_FN_MIGRATE,
+	PSCI_FN_MAX,
+};
+
+static u32 psci_function_id[PSCI_FN_MAX];
+
+static int psci_to_linux_errno(int errno)
+{
+	switch (errno) {
+	case PSCI_RET_SUCCESS:
+		return 0;
+	case PSCI_RET_NOT_SUPPORTED:
+		return -EOPNOTSUPP;
+	case PSCI_RET_INVALID_PARAMS:
+		return -EINVAL;
+	case PSCI_RET_DENIED:
+		return -EPERM;
+	};
+
+	return -EINVAL;
+}
+
+static u32 psci_get_version(void)
+{
+	return invoke_psci_fn(PSCI_0_2_FN_PSCI_VERSION, 0, 0, 0);
+}
+
+static int psci_cpu_suspend(u32 state, unsigned long entry_point)
+{
+	int err;
+	u32 fn;
+
+	fn = psci_function_id[PSCI_FN_CPU_SUSPEND];
+	err = invoke_psci_fn(fn, state, entry_point, 0);
+	return psci_to_linux_errno(err);
+}
+
+static int psci_cpu_off(u32 state)
+{
+	int err;
+	u32 fn;
+
+	fn = psci_function_id[PSCI_FN_CPU_OFF];
+	err = invoke_psci_fn(fn, state, 0, 0);
+	return psci_to_linux_errno(err);
+}
+
+static int psci_cpu_on(unsigned long cpuid, unsigned long entry_point)
+{
+	int err;
+	u32 fn;
+
+	fn = psci_function_id[PSCI_FN_CPU_ON];
+	err = invoke_psci_fn(fn, cpuid, entry_point, 0);
+	return psci_to_linux_errno(err);
+}
+
+static int psci_migrate(unsigned long cpuid)
+{
+	int err;
+	u32 fn;
+
+	fn = psci_function_id[PSCI_FN_MIGRATE];
+	err = invoke_psci_fn(fn, cpuid, 0, 0);
+	return psci_to_linux_errno(err);
+}
+
+static int psci_affinity_info(unsigned long target_affinity,
+		unsigned long lowest_affinity_level)
+{
+	return invoke_psci_fn(PSCI_0_2_FN_NATIVE(AFFINITY_INFO),
+			      target_affinity, lowest_affinity_level, 0);
+}
+
+static int psci_migrate_info_type(void)
+{
+	return invoke_psci_fn(PSCI_0_2_FN_MIGRATE_INFO_TYPE, 0, 0, 0);
+}
+
+static unsigned long psci_migrate_info_up_cpu(void)
+{
+	return invoke_psci_fn(PSCI_0_2_FN_NATIVE(MIGRATE_INFO_UP_CPU),
+			      0, 0, 0);
+}
+
+static int get_set_conduit_method(struct device_node *np)
+{
+	const char *method;
+
+	pr_info("probing for conduit method from DT.\n");
+
+	if (of_property_read_string(np, "method", &method)) {
+		pr_warn("missing \"method\" property\n");
+		return -ENXIO;
+	}
+
+	if (!strcmp("hvc", method)) {
+		invoke_psci_fn = __invoke_psci_fn_hvc;
+	} else if (!strcmp("smc", method)) {
+		invoke_psci_fn = __invoke_psci_fn_smc;
+	} else {
+		pr_warn("invalid \"method\" property: %s\n", method);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static void psci_sys_reset(enum reboot_mode reboot_mode, const char *cmd)
+{
+	invoke_psci_fn(PSCI_0_2_FN_SYSTEM_RESET, 0, 0, 0);
+}
+
+static void psci_sys_poweroff(void)
+{
+	invoke_psci_fn(PSCI_0_2_FN_SYSTEM_OFF, 0, 0, 0);
+}
+
+/*
+ * Detect the presence of a resident Trusted OS which may cause CPU_OFF to
+ * return DENIED (which would be fatal).
+ */
+static void __init psci_init_migrate(void)
+{
+	unsigned long cpuid;
+	int type, cpu = -1;
+
+	type = psci_ops.migrate_info_type();
+
+	if (type == PSCI_0_2_TOS_MP) {
+		pr_info("Trusted OS migration not required\n");
+		return;
+	}
+
+	if (type == PSCI_RET_NOT_SUPPORTED) {
+		pr_info("MIGRATE_INFO_TYPE not supported.\n");
+		return;
+	}
+
+	if (type != PSCI_0_2_TOS_UP_MIGRATE &&
+	    type != PSCI_0_2_TOS_UP_NO_MIGRATE) {
+		pr_err("MIGRATE_INFO_TYPE returned unknown type (%d)\n", type);
+		return;
+	}
+
+	cpuid = psci_migrate_info_up_cpu();
+	if (cpuid & ~MPIDR_HWID_BITMASK) {
+		pr_warn("MIGRATE_INFO_UP_CPU reported invalid physical ID (0x%lx)\n",
+			cpuid);
+		return;
+	}
+
+	cpu = get_logical_index(cpuid);
+	resident_cpu = cpu >= 0 ? cpu : -1;
+
+	pr_info("Trusted OS resident on physical CPU 0x%lx\n", cpuid);
+}
+
+static void __init psci_0_2_set_functions(void)
+{
+	pr_info("Using standard PSCI v0.2 function IDs\n");
+	psci_function_id[PSCI_FN_CPU_SUSPEND] = PSCI_0_2_FN_NATIVE(CPU_SUSPEND);
+	psci_ops.cpu_suspend = psci_cpu_suspend;
+
+	psci_function_id[PSCI_FN_CPU_OFF] = PSCI_0_2_FN_CPU_OFF;
+	psci_ops.cpu_off = psci_cpu_off;
+
+	psci_function_id[PSCI_FN_CPU_ON] = PSCI_0_2_FN_NATIVE(CPU_ON);
+	psci_ops.cpu_on = psci_cpu_on;
+
+	psci_function_id[PSCI_FN_MIGRATE] = PSCI_0_2_FN_NATIVE(MIGRATE);
+	psci_ops.migrate = psci_migrate;
+
+	psci_ops.affinity_info = psci_affinity_info;
+
+	psci_ops.migrate_info_type = psci_migrate_info_type;
+
+	arm_pm_restart = psci_sys_reset;
+
+	pm_power_off = psci_sys_poweroff;
+}
+
+/*
+ * Probe function for PSCI firmware versions >= 0.2
+ */
+static int __init psci_probe(void)
+{
+	u32 ver = psci_get_version();
+
+	pr_info("PSCIv%d.%d detected in firmware.\n",
+			PSCI_VERSION_MAJOR(ver),
+			PSCI_VERSION_MINOR(ver));
+
+	if (PSCI_VERSION_MAJOR(ver) == 0 && PSCI_VERSION_MINOR(ver) < 2) {
+		pr_err("Conflicting PSCI version detected.\n");
+		return -EINVAL;
+	}
+
+	psci_0_2_set_functions();
+
+	psci_init_migrate();
+
+	return 0;
+}
+
+typedef int (*psci_initcall_t)(const struct device_node *);
+
+/*
+ * PSCI init function for PSCI versions >=0.2
+ *
+ * Probe based on PSCI PSCI_VERSION function
+ */
+static int __init psci_0_2_init(struct device_node *np)
+{
+	int err;
+
+	err = get_set_conduit_method(np);
+
+	if (err)
+		goto out_put_node;
+	/*
+	 * Starting with v0.2, the PSCI specification introduced a call
+	 * (PSCI_VERSION) that allows probing the firmware version, so
+	 * that PSCI function IDs and version specific initialization
+	 * can be carried out according to the specific version reported
+	 * by firmware
+	 */
+	err = psci_probe();
+
+out_put_node:
+	of_node_put(np);
+	return err;
+}
+
+/*
+ * PSCI < v0.2 get PSCI Function IDs via DT.
+ */
+static int __init psci_0_1_init(struct device_node *np)
+{
+	u32 id;
+	int err;
+
+	err = get_set_conduit_method(np);
+
+	if (err)
+		goto out_put_node;
+
+	pr_info("Using PSCI v0.1 Function IDs from DT\n");
+
+	if (!of_property_read_u32(np, "cpu_suspend", &id)) {
+		psci_function_id[PSCI_FN_CPU_SUSPEND] = id;
+		psci_ops.cpu_suspend = psci_cpu_suspend;
+	}
+
+	if (!of_property_read_u32(np, "cpu_off", &id)) {
+		psci_function_id[PSCI_FN_CPU_OFF] = id;
+		psci_ops.cpu_off = psci_cpu_off;
+	}
+
+	if (!of_property_read_u32(np, "cpu_on", &id)) {
+		psci_function_id[PSCI_FN_CPU_ON] = id;
+		psci_ops.cpu_on = psci_cpu_on;
+	}
+
+	if (!of_property_read_u32(np, "migrate", &id)) {
+		psci_function_id[PSCI_FN_MIGRATE] = id;
+		psci_ops.migrate = psci_migrate;
+	}
+
+out_put_node:
+	of_node_put(np);
+	return err;
+}
+
+static const struct of_device_id psci_of_match[] __initconst = {
+	{ .compatible = "arm,psci",	.data = psci_0_1_init},
+	{ .compatible = "arm,psci-0.2",	.data = psci_0_2_init},
+	{},
+};
+
+int __init psci_dt_init(void)
+{
+	struct device_node *np;
+	const struct of_device_id *matched_np;
+	psci_initcall_t init_fn;
+
+	np = of_find_matching_node_and_match(NULL, psci_of_match, &matched_np);
+
+	if (!np)
+		return -ENODEV;
+
+	init_fn = (psci_initcall_t)matched_np->data;
+	return init_fn(np);
+}
+
+#ifdef CONFIG_ACPI
+/*
+ * We use PSCI 0.2+ when ACPI is deployed on ARM64 and it's
+ * explicitly clarified in SBBR
+ */
+int __init psci_acpi_init(void)
+{
+	if (!acpi_psci_present()) {
+		pr_info("is not implemented in ACPI.\n");
+		return -EOPNOTSUPP;
+	}
+
+	pr_info("probing for conduit method from ACPI.\n");
+
+	if (acpi_psci_use_hvc())
+		invoke_psci_fn = __invoke_psci_fn_hvc;
+	else
+		invoke_psci_fn = __invoke_psci_fn_smc;
+
+	return psci_probe();
+}
+#endif
diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig
new file mode 100644
index 000000000000..d9de36ee165d
--- /dev/null
+++ b/drivers/perf/Kconfig
@@ -0,0 +1,15 @@
+#
+# Performance Monitor Drivers
+#
+
+menu "Performance monitor support"
+
+config ARM_PMU
+	depends on PERF_EVENTS && ARM
+	bool "ARM PMU framework"
+	default y
+	help
+	  Say y if you want to use CPU performance monitors on ARM-based
+	  systems.
+
+endmenu
diff --git a/drivers/perf/Makefile b/drivers/perf/Makefile
new file mode 100644
index 000000000000..acd2397ded94
--- /dev/null
+++ b/drivers/perf/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_ARM_PMU) += arm_pmu.o
diff --git a/drivers/perf/arm_pmu.c b/drivers/perf/arm_pmu.c
new file mode 100644
index 000000000000..2365a32a595e
--- /dev/null
+++ b/drivers/perf/arm_pmu.c
@@ -0,0 +1,921 @@
+#undef DEBUG
+
+/*
+ * ARM performance counter support.
+ *
+ * Copyright (C) 2009 picoChip Designs, Ltd., Jamie Iles
+ * Copyright (C) 2010 ARM Ltd., Will Deacon <will.deacon@arm.com>
+ *
+ * This code is based on the sparc64 perf event code, which is in turn based
+ * on the x86 code.
+ */
+#define pr_fmt(fmt) "hw perfevents: " fmt
+
+#include <linux/bitmap.h>
+#include <linux/cpumask.h>
+#include <linux/export.h>
+#include <linux/kernel.h>
+#include <linux/of_device.h>
+#include <linux/perf/arm_pmu.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/irq.h>
+#include <linux/irqdesc.h>
+
+#include <asm/cputype.h>
+#include <asm/irq_regs.h>
+
+static int
+armpmu_map_cache_event(const unsigned (*cache_map)
+				      [PERF_COUNT_HW_CACHE_MAX]
+				      [PERF_COUNT_HW_CACHE_OP_MAX]
+				      [PERF_COUNT_HW_CACHE_RESULT_MAX],
+		       u64 config)
+{
+	unsigned int cache_type, cache_op, cache_result, ret;
+
+	cache_type = (config >>  0) & 0xff;
+	if (cache_type >= PERF_COUNT_HW_CACHE_MAX)
+		return -EINVAL;
+
+	cache_op = (config >>  8) & 0xff;
+	if (cache_op >= PERF_COUNT_HW_CACHE_OP_MAX)
+		return -EINVAL;
+
+	cache_result = (config >> 16) & 0xff;
+	if (cache_result >= PERF_COUNT_HW_CACHE_RESULT_MAX)
+		return -EINVAL;
+
+	ret = (int)(*cache_map)[cache_type][cache_op][cache_result];
+
+	if (ret == CACHE_OP_UNSUPPORTED)
+		return -ENOENT;
+
+	return ret;
+}
+
+static int
+armpmu_map_hw_event(const unsigned (*event_map)[PERF_COUNT_HW_MAX], u64 config)
+{
+	int mapping;
+
+	if (config >= PERF_COUNT_HW_MAX)
+		return -EINVAL;
+
+	mapping = (*event_map)[config];
+	return mapping == HW_OP_UNSUPPORTED ? -ENOENT : mapping;
+}
+
+static int
+armpmu_map_raw_event(u32 raw_event_mask, u64 config)
+{
+	return (int)(config & raw_event_mask);
+}
+
+int
+armpmu_map_event(struct perf_event *event,
+		 const unsigned (*event_map)[PERF_COUNT_HW_MAX],
+		 const unsigned (*cache_map)
+				[PERF_COUNT_HW_CACHE_MAX]
+				[PERF_COUNT_HW_CACHE_OP_MAX]
+				[PERF_COUNT_HW_CACHE_RESULT_MAX],
+		 u32 raw_event_mask)
+{
+	u64 config = event->attr.config;
+	int type = event->attr.type;
+
+	if (type == event->pmu->type)
+		return armpmu_map_raw_event(raw_event_mask, config);
+
+	switch (type) {
+	case PERF_TYPE_HARDWARE:
+		return armpmu_map_hw_event(event_map, config);
+	case PERF_TYPE_HW_CACHE:
+		return armpmu_map_cache_event(cache_map, config);
+	case PERF_TYPE_RAW:
+		return armpmu_map_raw_event(raw_event_mask, config);
+	}
+
+	return -ENOENT;
+}
+
+int armpmu_event_set_period(struct perf_event *event)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
+	struct hw_perf_event *hwc = &event->hw;
+	s64 left = local64_read(&hwc->period_left);
+	s64 period = hwc->sample_period;
+	int ret = 0;
+
+	if (unlikely(left <= -period)) {
+		left = period;
+		local64_set(&hwc->period_left, left);
+		hwc->last_period = period;
+		ret = 1;
+	}
+
+	if (unlikely(left <= 0)) {
+		left += period;
+		local64_set(&hwc->period_left, left);
+		hwc->last_period = period;
+		ret = 1;
+	}
+
+	/*
+	 * Limit the maximum period to prevent the counter value
+	 * from overtaking the one we are about to program. In
+	 * effect we are reducing max_period to account for
+	 * interrupt latency (and we are being very conservative).
+	 */
+	if (left > (armpmu->max_period >> 1))
+		left = armpmu->max_period >> 1;
+
+	local64_set(&hwc->prev_count, (u64)-left);
+
+	armpmu->write_counter(event, (u64)(-left) & 0xffffffff);
+
+	perf_event_update_userpage(event);
+
+	return ret;
+}
+
+u64 armpmu_event_update(struct perf_event *event)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
+	struct hw_perf_event *hwc = &event->hw;
+	u64 delta, prev_raw_count, new_raw_count;
+
+again:
+	prev_raw_count = local64_read(&hwc->prev_count);
+	new_raw_count = armpmu->read_counter(event);
+
+	if (local64_cmpxchg(&hwc->prev_count, prev_raw_count,
+			     new_raw_count) != prev_raw_count)
+		goto again;
+
+	delta = (new_raw_count - prev_raw_count) & armpmu->max_period;
+
+	local64_add(delta, &event->count);
+	local64_sub(delta, &hwc->period_left);
+
+	return new_raw_count;
+}
+
+static void
+armpmu_read(struct perf_event *event)
+{
+	armpmu_event_update(event);
+}
+
+static void
+armpmu_stop(struct perf_event *event, int flags)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
+	struct hw_perf_event *hwc = &event->hw;
+
+	/*
+	 * ARM pmu always has to update the counter, so ignore
+	 * PERF_EF_UPDATE, see comments in armpmu_start().
+	 */
+	if (!(hwc->state & PERF_HES_STOPPED)) {
+		armpmu->disable(event);
+		armpmu_event_update(event);
+		hwc->state |= PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	}
+}
+
+static void armpmu_start(struct perf_event *event, int flags)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
+	struct hw_perf_event *hwc = &event->hw;
+
+	/*
+	 * ARM pmu always has to reprogram the period, so ignore
+	 * PERF_EF_RELOAD, see the comment below.
+	 */
+	if (flags & PERF_EF_RELOAD)
+		WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE));
+
+	hwc->state = 0;
+	/*
+	 * Set the period again. Some counters can't be stopped, so when we
+	 * were stopped we simply disabled the IRQ source and the counter
+	 * may have been left counting. If we don't do this step then we may
+	 * get an interrupt too soon or *way* too late if the overflow has
+	 * happened since disabling.
+	 */
+	armpmu_event_set_period(event);
+	armpmu->enable(event);
+}
+
+static void
+armpmu_del(struct perf_event *event, int flags)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
+	struct pmu_hw_events *hw_events = this_cpu_ptr(armpmu->hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	int idx = hwc->idx;
+
+	armpmu_stop(event, PERF_EF_UPDATE);
+	hw_events->events[idx] = NULL;
+	clear_bit(idx, hw_events->used_mask);
+	if (armpmu->clear_event_idx)
+		armpmu->clear_event_idx(hw_events, event);
+
+	perf_event_update_userpage(event);
+}
+
+static int
+armpmu_add(struct perf_event *event, int flags)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
+	struct pmu_hw_events *hw_events = this_cpu_ptr(armpmu->hw_events);
+	struct hw_perf_event *hwc = &event->hw;
+	int idx;
+	int err = 0;
+
+	/* An event following a process won't be stopped earlier */
+	if (!cpumask_test_cpu(smp_processor_id(), &armpmu->supported_cpus))
+		return -ENOENT;
+
+	perf_pmu_disable(event->pmu);
+
+	/* If we don't have a space for the counter then finish early. */
+	idx = armpmu->get_event_idx(hw_events, event);
+	if (idx < 0) {
+		err = idx;
+		goto out;
+	}
+
+	/*
+	 * If there is an event in the counter we are going to use then make
+	 * sure it is disabled.
+	 */
+	event->hw.idx = idx;
+	armpmu->disable(event);
+	hw_events->events[idx] = event;
+
+	hwc->state = PERF_HES_STOPPED | PERF_HES_UPTODATE;
+	if (flags & PERF_EF_START)
+		armpmu_start(event, PERF_EF_RELOAD);
+
+	/* Propagate our changes to the userspace mapping. */
+	perf_event_update_userpage(event);
+
+out:
+	perf_pmu_enable(event->pmu);
+	return err;
+}
+
+static int
+validate_event(struct pmu *pmu, struct pmu_hw_events *hw_events,
+			       struct perf_event *event)
+{
+	struct arm_pmu *armpmu;
+
+	if (is_software_event(event))
+		return 1;
+
+	/*
+	 * Reject groups spanning multiple HW PMUs (e.g. CPU + CCI). The
+	 * core perf code won't check that the pmu->ctx == leader->ctx
+	 * until after pmu->event_init(event).
+	 */
+	if (event->pmu != pmu)
+		return 0;
+
+	if (event->state < PERF_EVENT_STATE_OFF)
+		return 1;
+
+	if (event->state == PERF_EVENT_STATE_OFF && !event->attr.enable_on_exec)
+		return 1;
+
+	armpmu = to_arm_pmu(event->pmu);
+	return armpmu->get_event_idx(hw_events, event) >= 0;
+}
+
+static int
+validate_group(struct perf_event *event)
+{
+	struct perf_event *sibling, *leader = event->group_leader;
+	struct pmu_hw_events fake_pmu;
+
+	/*
+	 * Initialise the fake PMU. We only need to populate the
+	 * used_mask for the purposes of validation.
+	 */
+	memset(&fake_pmu.used_mask, 0, sizeof(fake_pmu.used_mask));
+
+	if (!validate_event(event->pmu, &fake_pmu, leader))
+		return -EINVAL;
+
+	list_for_each_entry(sibling, &leader->sibling_list, group_entry) {
+		if (!validate_event(event->pmu, &fake_pmu, sibling))
+			return -EINVAL;
+	}
+
+	if (!validate_event(event->pmu, &fake_pmu, event))
+		return -EINVAL;
+
+	return 0;
+}
+
+static irqreturn_t armpmu_dispatch_irq(int irq, void *dev)
+{
+	struct arm_pmu *armpmu;
+	struct platform_device *plat_device;
+	struct arm_pmu_platdata *plat;
+	int ret;
+	u64 start_clock, finish_clock;
+
+	/*
+	 * we request the IRQ with a (possibly percpu) struct arm_pmu**, but
+	 * the handlers expect a struct arm_pmu*. The percpu_irq framework will
+	 * do any necessary shifting, we just need to perform the first
+	 * dereference.
+	 */
+	armpmu = *(void **)dev;
+	plat_device = armpmu->plat_device;
+	plat = dev_get_platdata(&plat_device->dev);
+
+	start_clock = sched_clock();
+	if (plat && plat->handle_irq)
+		ret = plat->handle_irq(irq, armpmu, armpmu->handle_irq);
+	else
+		ret = armpmu->handle_irq(irq, armpmu);
+	finish_clock = sched_clock();
+
+	perf_sample_event_took(finish_clock - start_clock);
+	return ret;
+}
+
+static void
+armpmu_release_hardware(struct arm_pmu *armpmu)
+{
+	armpmu->free_irq(armpmu);
+}
+
+static int
+armpmu_reserve_hardware(struct arm_pmu *armpmu)
+{
+	int err = armpmu->request_irq(armpmu, armpmu_dispatch_irq);
+	if (err) {
+		armpmu_release_hardware(armpmu);
+		return err;
+	}
+
+	return 0;
+}
+
+static void
+hw_perf_event_destroy(struct perf_event *event)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
+	atomic_t *active_events	 = &armpmu->active_events;
+	struct mutex *pmu_reserve_mutex = &armpmu->reserve_mutex;
+
+	if (atomic_dec_and_mutex_lock(active_events, pmu_reserve_mutex)) {
+		armpmu_release_hardware(armpmu);
+		mutex_unlock(pmu_reserve_mutex);
+	}
+}
+
+static int
+event_requires_mode_exclusion(struct perf_event_attr *attr)
+{
+	return attr->exclude_idle || attr->exclude_user ||
+	       attr->exclude_kernel || attr->exclude_hv;
+}
+
+static int
+__hw_perf_event_init(struct perf_event *event)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
+	struct hw_perf_event *hwc = &event->hw;
+	int mapping;
+
+	mapping = armpmu->map_event(event);
+
+	if (mapping < 0) {
+		pr_debug("event %x:%llx not supported\n", event->attr.type,
+			 event->attr.config);
+		return mapping;
+	}
+
+	/*
+	 * We don't assign an index until we actually place the event onto
+	 * hardware. Use -1 to signify that we haven't decided where to put it
+	 * yet. For SMP systems, each core has it's own PMU so we can't do any
+	 * clever allocation or constraints checking at this point.
+	 */
+	hwc->idx		= -1;
+	hwc->config_base	= 0;
+	hwc->config		= 0;
+	hwc->event_base		= 0;
+
+	/*
+	 * Check whether we need to exclude the counter from certain modes.
+	 */
+	if ((!armpmu->set_event_filter ||
+	     armpmu->set_event_filter(hwc, &event->attr)) &&
+	     event_requires_mode_exclusion(&event->attr)) {
+		pr_debug("ARM performance counters do not support "
+			 "mode exclusion\n");
+		return -EOPNOTSUPP;
+	}
+
+	/*
+	 * Store the event encoding into the config_base field.
+	 */
+	hwc->config_base	    |= (unsigned long)mapping;
+
+	if (!is_sampling_event(event)) {
+		/*
+		 * For non-sampling runs, limit the sample_period to half
+		 * of the counter width. That way, the new counter value
+		 * is far less likely to overtake the previous one unless
+		 * you have some serious IRQ latency issues.
+		 */
+		hwc->sample_period  = armpmu->max_period >> 1;
+		hwc->last_period    = hwc->sample_period;
+		local64_set(&hwc->period_left, hwc->sample_period);
+	}
+
+	if (event->group_leader != event) {
+		if (validate_group(event) != 0)
+			return -EINVAL;
+	}
+
+	return 0;
+}
+
+static int armpmu_event_init(struct perf_event *event)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
+	int err = 0;
+	atomic_t *active_events = &armpmu->active_events;
+
+	/*
+	 * Reject CPU-affine events for CPUs that are of a different class to
+	 * that which this PMU handles. Process-following events (where
+	 * event->cpu == -1) can be migrated between CPUs, and thus we have to
+	 * reject them later (in armpmu_add) if they're scheduled on a
+	 * different class of CPU.
+	 */
+	if (event->cpu != -1 &&
+		!cpumask_test_cpu(event->cpu, &armpmu->supported_cpus))
+		return -ENOENT;
+
+	/* does not support taken branch sampling */
+	if (has_branch_stack(event))
+		return -EOPNOTSUPP;
+
+	if (armpmu->map_event(event) == -ENOENT)
+		return -ENOENT;
+
+	event->destroy = hw_perf_event_destroy;
+
+	if (!atomic_inc_not_zero(active_events)) {
+		mutex_lock(&armpmu->reserve_mutex);
+		if (atomic_read(active_events) == 0)
+			err = armpmu_reserve_hardware(armpmu);
+
+		if (!err)
+			atomic_inc(active_events);
+		mutex_unlock(&armpmu->reserve_mutex);
+	}
+
+	if (err)
+		return err;
+
+	err = __hw_perf_event_init(event);
+	if (err)
+		hw_perf_event_destroy(event);
+
+	return err;
+}
+
+static void armpmu_enable(struct pmu *pmu)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(pmu);
+	struct pmu_hw_events *hw_events = this_cpu_ptr(armpmu->hw_events);
+	int enabled = bitmap_weight(hw_events->used_mask, armpmu->num_events);
+
+	/* For task-bound events we may be called on other CPUs */
+	if (!cpumask_test_cpu(smp_processor_id(), &armpmu->supported_cpus))
+		return;
+
+	if (enabled)
+		armpmu->start(armpmu);
+}
+
+static void armpmu_disable(struct pmu *pmu)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(pmu);
+
+	/* For task-bound events we may be called on other CPUs */
+	if (!cpumask_test_cpu(smp_processor_id(), &armpmu->supported_cpus))
+		return;
+
+	armpmu->stop(armpmu);
+}
+
+/*
+ * In heterogeneous systems, events are specific to a particular
+ * microarchitecture, and aren't suitable for another. Thus, only match CPUs of
+ * the same microarchitecture.
+ */
+static int armpmu_filter_match(struct perf_event *event)
+{
+	struct arm_pmu *armpmu = to_arm_pmu(event->pmu);
+	unsigned int cpu = smp_processor_id();
+	return cpumask_test_cpu(cpu, &armpmu->supported_cpus);
+}
+
+static void armpmu_init(struct arm_pmu *armpmu)
+{
+	atomic_set(&armpmu->active_events, 0);
+	mutex_init(&armpmu->reserve_mutex);
+
+	armpmu->pmu = (struct pmu) {
+		.pmu_enable	= armpmu_enable,
+		.pmu_disable	= armpmu_disable,
+		.event_init	= armpmu_event_init,
+		.add		= armpmu_add,
+		.del		= armpmu_del,
+		.start		= armpmu_start,
+		.stop		= armpmu_stop,
+		.read		= armpmu_read,
+		.filter_match	= armpmu_filter_match,
+	};
+}
+
+int armpmu_register(struct arm_pmu *armpmu, int type)
+{
+	armpmu_init(armpmu);
+	pr_info("enabled with %s PMU driver, %d counters available\n",
+			armpmu->name, armpmu->num_events);
+	return perf_pmu_register(&armpmu->pmu, armpmu->name, type);
+}
+
+/* Set at runtime when we know what CPU type we are. */
+static struct arm_pmu *__oprofile_cpu_pmu;
+
+/*
+ * Despite the names, these two functions are CPU-specific and are used
+ * by the OProfile/perf code.
+ */
+const char *perf_pmu_name(void)
+{
+	if (!__oprofile_cpu_pmu)
+		return NULL;
+
+	return __oprofile_cpu_pmu->name;
+}
+EXPORT_SYMBOL_GPL(perf_pmu_name);
+
+int perf_num_counters(void)
+{
+	int max_events = 0;
+
+	if (__oprofile_cpu_pmu != NULL)
+		max_events = __oprofile_cpu_pmu->num_events;
+
+	return max_events;
+}
+EXPORT_SYMBOL_GPL(perf_num_counters);
+
+static void cpu_pmu_enable_percpu_irq(void *data)
+{
+	int irq = *(int *)data;
+
+	enable_percpu_irq(irq, IRQ_TYPE_NONE);
+}
+
+static void cpu_pmu_disable_percpu_irq(void *data)
+{
+	int irq = *(int *)data;
+
+	disable_percpu_irq(irq);
+}
+
+static void cpu_pmu_free_irq(struct arm_pmu *cpu_pmu)
+{
+	int i, irq, irqs;
+	struct platform_device *pmu_device = cpu_pmu->plat_device;
+	struct pmu_hw_events __percpu *hw_events = cpu_pmu->hw_events;
+
+	irqs = min(pmu_device->num_resources, num_possible_cpus());
+
+	irq = platform_get_irq(pmu_device, 0);
+	if (irq >= 0 && irq_is_percpu(irq)) {
+		on_each_cpu(cpu_pmu_disable_percpu_irq, &irq, 1);
+		free_percpu_irq(irq, &hw_events->percpu_pmu);
+	} else {
+		for (i = 0; i < irqs; ++i) {
+			int cpu = i;
+
+			if (cpu_pmu->irq_affinity)
+				cpu = cpu_pmu->irq_affinity[i];
+
+			if (!cpumask_test_and_clear_cpu(cpu, &cpu_pmu->active_irqs))
+				continue;
+			irq = platform_get_irq(pmu_device, i);
+			if (irq >= 0)
+				free_irq(irq, per_cpu_ptr(&hw_events->percpu_pmu, cpu));
+		}
+	}
+}
+
+static int cpu_pmu_request_irq(struct arm_pmu *cpu_pmu, irq_handler_t handler)
+{
+	int i, err, irq, irqs;
+	struct platform_device *pmu_device = cpu_pmu->plat_device;
+	struct pmu_hw_events __percpu *hw_events = cpu_pmu->hw_events;
+
+	if (!pmu_device)
+		return -ENODEV;
+
+	irqs = min(pmu_device->num_resources, num_possible_cpus());
+	if (irqs < 1) {
+		pr_warn_once("perf/ARM: No irqs for PMU defined, sampling events not supported\n");
+		return 0;
+	}
+
+	irq = platform_get_irq(pmu_device, 0);
+	if (irq >= 0 && irq_is_percpu(irq)) {
+		err = request_percpu_irq(irq, handler, "arm-pmu",
+					 &hw_events->percpu_pmu);
+		if (err) {
+			pr_err("unable to request IRQ%d for ARM PMU counters\n",
+				irq);
+			return err;
+		}
+		on_each_cpu(cpu_pmu_enable_percpu_irq, &irq, 1);
+	} else {
+		for (i = 0; i < irqs; ++i) {
+			int cpu = i;
+
+			err = 0;
+			irq = platform_get_irq(pmu_device, i);
+			if (irq < 0)
+				continue;
+
+			if (cpu_pmu->irq_affinity)
+				cpu = cpu_pmu->irq_affinity[i];
+
+			/*
+			 * If we have a single PMU interrupt that we can't shift,
+			 * assume that we're running on a uniprocessor machine and
+			 * continue. Otherwise, continue without this interrupt.
+			 */
+			if (irq_set_affinity(irq, cpumask_of(cpu)) && irqs > 1) {
+				pr_warn("unable to set irq affinity (irq=%d, cpu=%u)\n",
+					irq, cpu);
+				continue;
+			}
+
+			err = request_irq(irq, handler,
+					  IRQF_NOBALANCING | IRQF_NO_THREAD, "arm-pmu",
+					  per_cpu_ptr(&hw_events->percpu_pmu, cpu));
+			if (err) {
+				pr_err("unable to request IRQ%d for ARM PMU counters\n",
+					irq);
+				return err;
+			}
+
+			cpumask_set_cpu(cpu, &cpu_pmu->active_irqs);
+		}
+	}
+
+	return 0;
+}
+
+/*
+ * PMU hardware loses all context when a CPU goes offline.
+ * When a CPU is hotplugged back in, since some hardware registers are
+ * UNKNOWN at reset, the PMU must be explicitly reset to avoid reading
+ * junk values out of them.
+ */
+static int cpu_pmu_notify(struct notifier_block *b, unsigned long action,
+			  void *hcpu)
+{
+	int cpu = (unsigned long)hcpu;
+	struct arm_pmu *pmu = container_of(b, struct arm_pmu, hotplug_nb);
+
+	if ((action & ~CPU_TASKS_FROZEN) != CPU_STARTING)
+		return NOTIFY_DONE;
+
+	if (!cpumask_test_cpu(cpu, &pmu->supported_cpus))
+		return NOTIFY_DONE;
+
+	if (pmu->reset)
+		pmu->reset(pmu);
+	else
+		return NOTIFY_DONE;
+
+	return NOTIFY_OK;
+}
+
+static int cpu_pmu_init(struct arm_pmu *cpu_pmu)
+{
+	int err;
+	int cpu;
+	struct pmu_hw_events __percpu *cpu_hw_events;
+
+	cpu_hw_events = alloc_percpu(struct pmu_hw_events);
+	if (!cpu_hw_events)
+		return -ENOMEM;
+
+	cpu_pmu->hotplug_nb.notifier_call = cpu_pmu_notify;
+	err = register_cpu_notifier(&cpu_pmu->hotplug_nb);
+	if (err)
+		goto out_hw_events;
+
+	for_each_possible_cpu(cpu) {
+		struct pmu_hw_events *events = per_cpu_ptr(cpu_hw_events, cpu);
+		raw_spin_lock_init(&events->pmu_lock);
+		events->percpu_pmu = cpu_pmu;
+	}
+
+	cpu_pmu->hw_events	= cpu_hw_events;
+	cpu_pmu->request_irq	= cpu_pmu_request_irq;
+	cpu_pmu->free_irq	= cpu_pmu_free_irq;
+
+	/* Ensure the PMU has sane values out of reset. */
+	if (cpu_pmu->reset)
+		on_each_cpu_mask(&cpu_pmu->supported_cpus, cpu_pmu->reset,
+			 cpu_pmu, 1);
+
+	/* If no interrupts available, set the corresponding capability flag */
+	if (!platform_get_irq(cpu_pmu->plat_device, 0))
+		cpu_pmu->pmu.capabilities |= PERF_PMU_CAP_NO_INTERRUPT;
+
+	return 0;
+
+out_hw_events:
+	free_percpu(cpu_hw_events);
+	return err;
+}
+
+static void cpu_pmu_destroy(struct arm_pmu *cpu_pmu)
+{
+	unregister_cpu_notifier(&cpu_pmu->hotplug_nb);
+	free_percpu(cpu_pmu->hw_events);
+}
+
+/*
+ * CPU PMU identification and probing.
+ */
+static int probe_current_pmu(struct arm_pmu *pmu,
+			     const struct pmu_probe_info *info)
+{
+	int cpu = get_cpu();
+	unsigned int cpuid = read_cpuid_id();
+	int ret = -ENODEV;
+
+	pr_info("probing PMU on CPU %d\n", cpu);
+
+	for (; info->init != NULL; info++) {
+		if ((cpuid & info->mask) != info->cpuid)
+			continue;
+		ret = info->init(pmu);
+		break;
+	}
+
+	put_cpu();
+	return ret;
+}
+
+static int of_pmu_irq_cfg(struct arm_pmu *pmu)
+{
+	int *irqs, i = 0;
+	bool using_spi = false;
+	struct platform_device *pdev = pmu->plat_device;
+
+	irqs = kcalloc(pdev->num_resources, sizeof(*irqs), GFP_KERNEL);
+	if (!irqs)
+		return -ENOMEM;
+
+	do {
+		struct device_node *dn;
+		int cpu, irq;
+
+		/* See if we have an affinity entry */
+		dn = of_parse_phandle(pdev->dev.of_node, "interrupt-affinity", i);
+		if (!dn)
+			break;
+
+		/* Check the IRQ type and prohibit a mix of PPIs and SPIs */
+		irq = platform_get_irq(pdev, i);
+		if (irq >= 0) {
+			bool spi = !irq_is_percpu(irq);
+
+			if (i > 0 && spi != using_spi) {
+				pr_err("PPI/SPI IRQ type mismatch for %s!\n",
+					dn->name);
+				kfree(irqs);
+				return -EINVAL;
+			}
+
+			using_spi = spi;
+		}
+
+		/* Now look up the logical CPU number */
+		for_each_possible_cpu(cpu)
+			if (dn == of_cpu_device_node_get(cpu))
+				break;
+
+		if (cpu >= nr_cpu_ids) {
+			pr_warn("Failed to find logical CPU for %s\n",
+				dn->name);
+			of_node_put(dn);
+			cpumask_setall(&pmu->supported_cpus);
+			break;
+		}
+		of_node_put(dn);
+
+		/* For SPIs, we need to track the affinity per IRQ */
+		if (using_spi) {
+			if (i >= pdev->num_resources) {
+				of_node_put(dn);
+				break;
+			}
+
+			irqs[i] = cpu;
+		}
+
+		/* Keep track of the CPUs containing this PMU type */
+		cpumask_set_cpu(cpu, &pmu->supported_cpus);
+		of_node_put(dn);
+		i++;
+	} while (1);
+
+	/* If we didn't manage to parse anything, claim to support all CPUs */
+	if (cpumask_weight(&pmu->supported_cpus) == 0)
+		cpumask_setall(&pmu->supported_cpus);
+
+	/* If we matched up the IRQ affinities, use them to route the SPIs */
+	if (using_spi && i == pdev->num_resources)
+		pmu->irq_affinity = irqs;
+	else
+		kfree(irqs);
+
+	return 0;
+}
+
+int arm_pmu_device_probe(struct platform_device *pdev,
+			 const struct of_device_id *of_table,
+			 const struct pmu_probe_info *probe_table)
+{
+	const struct of_device_id *of_id;
+	const int (*init_fn)(struct arm_pmu *);
+	struct device_node *node = pdev->dev.of_node;
+	struct arm_pmu *pmu;
+	int ret = -ENODEV;
+
+	pmu = kzalloc(sizeof(struct arm_pmu), GFP_KERNEL);
+	if (!pmu) {
+		pr_info("failed to allocate PMU device!\n");
+		return -ENOMEM;
+	}
+
+	if (!__oprofile_cpu_pmu)
+		__oprofile_cpu_pmu = pmu;
+
+	pmu->plat_device = pdev;
+
+	if (node && (of_id = of_match_node(of_table, pdev->dev.of_node))) {
+		init_fn = of_id->data;
+
+		ret = of_pmu_irq_cfg(pmu);
+		if (!ret)
+			ret = init_fn(pmu);
+	} else {
+		ret = probe_current_pmu(pmu, probe_table);
+		cpumask_setall(&pmu->supported_cpus);
+	}
+
+	if (ret) {
+		pr_info("failed to probe PMU!\n");
+		goto out_free;
+	}
+
+	ret = cpu_pmu_init(pmu);
+	if (ret)
+		goto out_free;
+
+	ret = armpmu_register(pmu, -1);
+	if (ret)
+		goto out_destroy;
+
+	return 0;
+
+out_destroy:
+	cpu_pmu_destroy(pmu);
+out_free:
+	pr_info("failed to register PMU devices!\n");
+	kfree(pmu);
+	return ret;
+}