Merge branch 'for-next/perf' into for-next/core

* for-next/perf:
  drivers/perf: hisi: Update HiSilicon PMU maintainers
  arm_pmu: acpi: Add a representative platform device for TRBE
  arm_pmu: acpi: Refactor arm_spe_acpi_register_device()
  hw_breakpoint: fix single-stepping when using bpf_overflow_handler
  perf/imx_ddr: don't enable counter0 if none of 4 counters are used
  perf/imx_ddr: speed up overflow frequency of cycle
  drivers/perf: hisi: Schedule perf session according to locality
  perf/arm-dmc620: Fix dmc620_pmu_irqs_lock/cpu_hotplug_lock circular lock dependency
  perf/smmuv3: Add MODULE_ALIAS for module auto loading
  perf/smmuv3: Enable HiSilicon Erratum 162001900 quirk for HIP08/09
  perf: pmuv3: Remove comments from armv8pmu_[enable|disable]_event()
  perf/arm-cmn: Add CMN-700 r3 support
  perf/arm-cmn: Refactor HN-F event selector macros
  perf/arm-cmn: Remove spurious event aliases
  drivers/perf: Explicitly include correct DT includes
  perf: pmuv3: Add Cortex A520, A715, A720, X3 and X4 PMUs
  dt-bindings: arm: pmu: Add Cortex A520, A715, A720, X3, and X4
  perf/smmuv3: Remove build dependency on ACPI
  perf: xgene_pmu: Convert to devm_platform_ioremap_resource()
  driver/perf: Add identifier sysfs file for Yitian 710 DDR

Documentation/arch/arm64/silicon-errata.rst

@@ -195,6 +195,9 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | Hisilicon      | Hip08 SMMU PMCG | #162001800      | N/A                         |
 +----------------+-----------------+-----------------+-----------------------------+
+| Hisilicon      | Hip08 SMMU PMCG | #162001900      | N/A                         |
+|                | Hip09 SMMU PMCG |                 |                             |
++----------------+-----------------+-----------------+-----------------------------+
 +----------------+-----------------+-----------------+-----------------------------+
 | Qualcomm Tech. | Kryo/Falkor v1  | E1003           | QCOM_FALKOR_ERRATUM_1003    |
 +----------------+-----------------+-----------------+-----------------------------+

Documentation/devicetree/bindings/arm/pmu.yaml

@@ -49,9 +49,14 @@ properties:
           - arm,cortex-a77-pmu
           - arm,cortex-a78-pmu
           - arm,cortex-a510-pmu
+          - arm,cortex-a520-pmu
           - arm,cortex-a710-pmu
+          - arm,cortex-a715-pmu
+          - arm,cortex-a720-pmu
           - arm,cortex-x1-pmu
           - arm,cortex-x2-pmu
+          - arm,cortex-x3-pmu
+          - arm,cortex-x4-pmu
           - arm,neoverse-e1-pmu
           - arm,neoverse-n1-pmu
           - arm,neoverse-n2-pmu

MAINTAINERS

@@ -9306,7 +9306,7 @@ F:	drivers/crypto/hisilicon/hpre/hpre_crypto.c
 F:	drivers/crypto/hisilicon/hpre/hpre_main.c
 
 HISILICON HNS3 PMU DRIVER
-M:	Guangbin Huang <huangguangbin2@huawei.com>
+M:	Jijie Shao <shaojijie@huawei.com>
 S:	Supported
 F:	Documentation/admin-guide/perf/hns3-pmu.rst
 F:	drivers/perf/hisilicon/hns3_pmu.c
@@ -9344,7 +9344,7 @@ F:	Documentation/devicetree/bindings/net/hisilicon*.txt
 F:	drivers/net/ethernet/hisilicon/
 
 HISILICON PMU DRIVER
-M:	Shaokun Zhang <zhangshaokun@hisilicon.com>
+M:	Yicong Yang <yangyicong@hisilicon.com>
 M:	Jonathan Cameron <jonathan.cameron@huawei.com>
 S:	Supported
 W:	http://www.hisilicon.com

arch/arm/kernel/hw_breakpoint.c

@@ -626,7 +626,7 @@ int hw_breakpoint_arch_parse(struct perf_event *bp,
 	hw->address &= ~alignment_mask;
 	hw->ctrl.len <<= offset;
 
-	if (is_default_overflow_handler(bp)) {
+	if (uses_default_overflow_handler(bp)) {
 		/*
 		 * Mismatch breakpoints are required for single-stepping
 		 * breakpoints.
@@ -798,7 +798,7 @@ static void watchpoint_handler(unsigned long addr, unsigned int fsr,
 		 * Otherwise, insert a temporary mismatch breakpoint so that
 		 * we can single-step over the watchpoint trigger.
 		 */
-		if (!is_default_overflow_handler(wp))
+		if (!uses_default_overflow_handler(wp))
 			continue;
 step:
 		enable_single_step(wp, instruction_pointer(regs));
@@ -811,7 +811,7 @@ static void watchpoint_handler(unsigned long addr, unsigned int fsr,
 		info->trigger = addr;
 		pr_debug("watchpoint fired: address = 0x%x\n", info->trigger);
 		perf_bp_event(wp, regs);
-		if (is_default_overflow_handler(wp))
+		if (uses_default_overflow_handler(wp))
 			enable_single_step(wp, instruction_pointer(regs));
 	}
 
@@ -886,7 +886,7 @@ static void breakpoint_handler(unsigned long unknown, struct pt_regs *regs)
 			info->trigger = addr;
 			pr_debug("breakpoint fired: address = 0x%x\n", addr);
 			perf_bp_event(bp, regs);
-			if (is_default_overflow_handler(bp))
+			if (uses_default_overflow_handler(bp))
 				enable_single_step(bp, addr);
 			goto unlock;
 		}

arch/arm64/include/asm/acpi.h

@@ -42,6 +42,9 @@
 #define ACPI_MADT_GICC_SPE  (offsetof(struct acpi_madt_generic_interrupt, \
 	spe_interrupt) + sizeof(u16))
 
+#define ACPI_MADT_GICC_TRBE  (offsetof(struct acpi_madt_generic_interrupt, \
+	trbe_interrupt) + sizeof(u16))
+
 /* Basic configuration for ACPI */
 #ifdef	CONFIG_ACPI
 pgprot_t __acpi_get_mem_attribute(phys_addr_t addr);

arch/arm64/kernel/hw_breakpoint.c

@@ -654,7 +654,7 @@ static int breakpoint_handler(unsigned long unused, unsigned long esr,
 		perf_bp_event(bp, regs);
 
 		/* Do we need to handle the stepping? */
-		if (is_default_overflow_handler(bp))
+		if (uses_default_overflow_handler(bp))
 			step = 1;
 unlock:
 		rcu_read_unlock();
@@ -733,7 +733,7 @@ static u64 get_distance_from_watchpoint(unsigned long addr, u64 val,
 static int watchpoint_report(struct perf_event *wp, unsigned long addr,
 			     struct pt_regs *regs)
 {
-	int step = is_default_overflow_handler(wp);
+	int step = uses_default_overflow_handler(wp);
 	struct arch_hw_breakpoint *info = counter_arch_bp(wp);
 
 	info->trigger = addr;

drivers/acpi/arm64/iort.c

@@ -1711,7 +1711,10 @@ static void __init arm_smmu_v3_pmcg_init_resources(struct resource *res,
 static struct acpi_platform_list pmcg_plat_info[] __initdata = {
 	/* HiSilicon Hip08 Platform */
 	{"HISI  ", "HIP08   ", 0, ACPI_SIG_IORT, greater_than_or_equal,
-	 "Erratum #162001800", IORT_SMMU_V3_PMCG_HISI_HIP08},
+	 "Erratum #162001800, Erratum #162001900", IORT_SMMU_V3_PMCG_HISI_HIP08},
+	/* HiSilicon Hip09 Platform */
+	{"HISI  ", "HIP09   ", 0, ACPI_SIG_IORT, greater_than_or_equal,
+	 "Erratum #162001900", IORT_SMMU_V3_PMCG_HISI_HIP09},
 	{ }
 };
 

drivers/perf/Kconfig

@@ -92,7 +92,7 @@ config ARM_PMU_ACPI
 
 config ARM_SMMU_V3_PMU
 	 tristate "ARM SMMUv3 Performance Monitors Extension"
-	 depends on (ARM64 && ACPI) || (COMPILE_TEST && 64BIT)
+	 depends on ARM64 || (COMPILE_TEST && 64BIT)
 	 depends on GENERIC_MSI_IRQ
 	   help
 	   Provides support for the ARM SMMUv3 Performance Monitor Counter

drivers/perf/alibaba_uncore_drw_pmu.c

@@ -236,10 +236,37 @@ static const struct attribute_group ali_drw_pmu_cpumask_attr_group = {
 	.attrs = ali_drw_pmu_cpumask_attrs,
 };
 
+static ssize_t ali_drw_pmu_identifier_show(struct device *dev,
+					struct device_attribute *attr,
+					char *page)
+{
+	return sysfs_emit(page, "%s\n", "ali_drw_pmu");
+}
+
+static umode_t ali_drw_pmu_identifier_attr_visible(struct kobject *kobj,
+						struct attribute *attr, int n)
+{
+	return attr->mode;
+}
+
+static struct device_attribute ali_drw_pmu_identifier_attr =
+	__ATTR(identifier, 0444, ali_drw_pmu_identifier_show, NULL);
+
+static struct attribute *ali_drw_pmu_identifier_attrs[] = {
+	&ali_drw_pmu_identifier_attr.attr,
+	NULL
+};
+
+static const struct attribute_group ali_drw_pmu_identifier_attr_group = {
+	.attrs = ali_drw_pmu_identifier_attrs,
+	.is_visible = ali_drw_pmu_identifier_attr_visible
+};
+
 static const struct attribute_group *ali_drw_pmu_attr_groups[] = {
 	&ali_drw_pmu_events_attr_group,
 	&ali_drw_pmu_cpumask_attr_group,
 	&ali_drw_pmu_format_group,
+	&ali_drw_pmu_identifier_attr_group,
 	NULL,
 };
 

drivers/perf/amlogic/meson_ddr_pmu_core.c

@@ -9,8 +9,6 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/of.h>
-#include <linux/of_device.h>
-#include <linux/of_irq.h>
 #include <linux/perf_event.h>
 #include <linux/platform_device.h>
 #include <linux/printk.h>

drivers/perf/arm-cci.c

@@ -7,10 +7,7 @@
 #include <linux/io.h>
 #include <linux/interrupt.h>
 #include <linux/module.h>
-#include <linux/of_address.h>
-#include <linux/of_device.h>
-#include <linux/of_irq.h>
-#include <linux/of_platform.h>
+#include <linux/of.h>
 #include <linux/perf_event.h>
 #include <linux/platform_device.h>
 #include <linux/slab.h>

drivers/perf/arm_dmc620_pmu.c

@@ -66,8 +66,13 @@
 #define DMC620_PMU_COUNTERn_OFFSET(n) \
 	(DMC620_PMU_COUNTERS_BASE + 0x28 * (n))
 
-static LIST_HEAD(dmc620_pmu_irqs);
+/*
+ * dmc620_pmu_irqs_lock: protects dmc620_pmu_irqs list
+ * dmc620_pmu_node_lock: protects pmus_node lists in all dmc620_pmu instances
+ */
 static DEFINE_MUTEX(dmc620_pmu_irqs_lock);
+static DEFINE_MUTEX(dmc620_pmu_node_lock);
+static LIST_HEAD(dmc620_pmu_irqs);
 
 struct dmc620_pmu_irq {
 	struct hlist_node node;
@@ -475,9 +480,9 @@ static int dmc620_pmu_get_irq(struct dmc620_pmu *dmc620_pmu, int irq_num)
 		return PTR_ERR(irq);
 
 	dmc620_pmu->irq = irq;
-	mutex_lock(&dmc620_pmu_irqs_lock);
+	mutex_lock(&dmc620_pmu_node_lock);
 	list_add_rcu(&dmc620_pmu->pmus_node, &irq->pmus_node);
-	mutex_unlock(&dmc620_pmu_irqs_lock);
+	mutex_unlock(&dmc620_pmu_node_lock);
 
 	return 0;
 }
@@ -486,9 +491,11 @@ static void dmc620_pmu_put_irq(struct dmc620_pmu *dmc620_pmu)
 {
 	struct dmc620_pmu_irq *irq = dmc620_pmu->irq;
 
-	mutex_lock(&dmc620_pmu_irqs_lock);
+	mutex_lock(&dmc620_pmu_node_lock);
 	list_del_rcu(&dmc620_pmu->pmus_node);
+	mutex_unlock(&dmc620_pmu_node_lock);
 
+	mutex_lock(&dmc620_pmu_irqs_lock);
 	if (!refcount_dec_and_test(&irq->refcount)) {
 		mutex_unlock(&dmc620_pmu_irqs_lock);
 		return;
@@ -638,10 +645,10 @@ static int dmc620_pmu_cpu_teardown(unsigned int cpu,
 		return 0;
 
 	/* We're only reading, but this isn't the place to be involving RCU */
-	mutex_lock(&dmc620_pmu_irqs_lock);
+	mutex_lock(&dmc620_pmu_node_lock);
 	list_for_each_entry(dmc620_pmu, &irq->pmus_node, pmus_node)
 		perf_pmu_migrate_context(&dmc620_pmu->pmu, irq->cpu, target);
-	mutex_unlock(&dmc620_pmu_irqs_lock);
+	mutex_unlock(&dmc620_pmu_node_lock);
 
 	WARN_ON(irq_set_affinity(irq->irq_num, cpumask_of(target)));
 	irq->cpu = target;

drivers/perf/arm_dsu_pmu.c

@@ -20,7 +20,7 @@
 #include <linux/interrupt.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/of_device.h>
+#include <linux/of.h>
 #include <linux/perf_event.h>
 #include <linux/platform_device.h>
 #include <linux/spinlock.h>

drivers/perf/arm_pmu_acpi.c

@@ -69,6 +69,62 @@ static void arm_pmu_acpi_unregister_irq(int cpu)
 		acpi_unregister_gsi(gsi);
 }
 
+static int __maybe_unused
+arm_acpi_register_pmu_device(struct platform_device *pdev, u8 len,
+			     u16 (*parse_gsi)(struct acpi_madt_generic_interrupt *))
+{
+	int cpu, this_hetid, hetid, irq, ret;
+	u16 this_gsi = 0, gsi = 0;
+
+	/*
+	 * Ensure that platform device must have IORESOURCE_IRQ
+	 * resource to hold gsi interrupt.
+	 */
+	if (pdev->num_resources != 1)
+		return -ENXIO;
+
+	if (pdev->resource[0].flags != IORESOURCE_IRQ)
+		return -ENXIO;
+
+	/*
+	 * Sanity check all the GICC tables for the same interrupt
+	 * number. For now, only support homogeneous ACPI machines.
+	 */
+	for_each_possible_cpu(cpu) {
+		struct acpi_madt_generic_interrupt *gicc;
+
+		gicc = acpi_cpu_get_madt_gicc(cpu);
+		if (gicc->header.length < len)
+			return gsi ? -ENXIO : 0;
+
+		this_gsi = parse_gsi(gicc);
+		this_hetid = find_acpi_cpu_topology_hetero_id(cpu);
+		if (!gsi) {
+			hetid = this_hetid;
+			gsi = this_gsi;
+		} else if (hetid != this_hetid || gsi != this_gsi) {
+			pr_warn("ACPI: %s: must be homogeneous\n", pdev->name);
+			return -ENXIO;
+		}
+	}
+
+	if (!this_gsi)
+		return 0;
+
+	irq = acpi_register_gsi(NULL, gsi, ACPI_LEVEL_SENSITIVE, ACPI_ACTIVE_HIGH);
+	if (irq < 0) {
+		pr_warn("ACPI: %s Unable to register interrupt: %d\n", pdev->name, gsi);
+		return -ENXIO;
+	}
+
+	pdev->resource[0].start = irq;
+	ret = platform_device_register(pdev);
+	if (ret)
+		acpi_unregister_gsi(gsi);
+
+	return ret;
+}
+
 #if IS_ENABLED(CONFIG_ARM_SPE_PMU)
 static struct resource spe_resources[] = {
 	{
@@ -84,6 +140,11 @@ static struct platform_device spe_dev = {
 	.num_resources = ARRAY_SIZE(spe_resources)
 };
 
+static u16 arm_spe_parse_gsi(struct acpi_madt_generic_interrupt *gicc)
+{
+	return gicc->spe_interrupt;
+}
+
 /*
  * For lack of a better place, hook the normal PMU MADT walk
  * and create a SPE device if we detect a recent MADT with
@@ -91,53 +152,50 @@ static struct platform_device spe_dev = {
  */
 static void arm_spe_acpi_register_device(void)
 {
-	int cpu, hetid, irq, ret;
-	bool first = true;
-	u16 gsi = 0;
-
-	/*
-	 * Sanity check all the GICC tables for the same interrupt number.
-	 * For now, we only support homogeneous ACPI/SPE machines.
-	 */
-	for_each_possible_cpu(cpu) {
-		struct acpi_madt_generic_interrupt *gicc;
+	int ret = arm_acpi_register_pmu_device(&spe_dev, ACPI_MADT_GICC_SPE,
+					       arm_spe_parse_gsi);
+	if (ret)
+		pr_warn("ACPI: SPE: Unable to register device\n");
+}
+#else
+static inline void arm_spe_acpi_register_device(void)
+{
+}
+#endif /* CONFIG_ARM_SPE_PMU */
 
-		gicc = acpi_cpu_get_madt_gicc(cpu);
-		if (gicc->header.length < ACPI_MADT_GICC_SPE)
-			return;
-
-		if (first) {
-			gsi = gicc->spe_interrupt;
-			if (!gsi)
-				return;
-			hetid = find_acpi_cpu_topology_hetero_id(cpu);
-			first = false;
-		} else if ((gsi != gicc->spe_interrupt) ||
-			   (hetid != find_acpi_cpu_topology_hetero_id(cpu))) {
-			pr_warn("ACPI: SPE must be homogeneous\n");
-			return;
-		}
+#if IS_ENABLED(CONFIG_CORESIGHT_TRBE)
+static struct resource trbe_resources[] = {
+	{
+		/* irq */
+		.flags          = IORESOURCE_IRQ,
 	}
+};
 
-	irq = acpi_register_gsi(NULL, gsi, ACPI_LEVEL_SENSITIVE,
-				ACPI_ACTIVE_HIGH);
-	if (irq < 0) {
-		pr_warn("ACPI: SPE Unable to register interrupt: %d\n", gsi);
-		return;
-	}
+static struct platform_device trbe_dev = {
+	.name = ARMV8_TRBE_PDEV_NAME,
+	.id = -1,
+	.resource = trbe_resources,
+	.num_resources = ARRAY_SIZE(trbe_resources)
+};
 
-	spe_resources[0].start = irq;
-	ret = platform_device_register(&spe_dev);
-	if (ret < 0) {
-		pr_warn("ACPI: SPE: Unable to register device\n");
-		acpi_unregister_gsi(gsi);
-	}
+static u16 arm_trbe_parse_gsi(struct acpi_madt_generic_interrupt *gicc)
+{
+	return gicc->trbe_interrupt;
+}
+
+static void arm_trbe_acpi_register_device(void)
+{
+	int ret = arm_acpi_register_pmu_device(&trbe_dev, ACPI_MADT_GICC_TRBE,
+					       arm_trbe_parse_gsi);
+	if (ret)
+		pr_warn("ACPI: TRBE: Unable to register device\n");
 }
 #else
-static inline void arm_spe_acpi_register_device(void)
+static inline void arm_trbe_acpi_register_device(void)
 {
+
 }
-#endif /* CONFIG_ARM_SPE_PMU */
+#endif /* CONFIG_CORESIGHT_TRBE */
 
 static int arm_pmu_acpi_parse_irqs(void)
 {
@@ -374,6 +432,7 @@ static int arm_pmu_acpi_init(void)
 		return 0;
 
 	arm_spe_acpi_register_device();
+	arm_trbe_acpi_register_device();
 
 	return 0;
 }

drivers/perf/arm_pmu_platform.c

@@ -16,7 +16,6 @@
 #include <linux/irqdesc.h>
 #include <linux/kconfig.h>
 #include <linux/of.h>
-#include <linux/of_device.h>
 #include <linux/percpu.h>
 #include <linux/perf/arm_pmu.h>
 #include <linux/platform_device.h>

drivers/perf/arm_pmuv3.c

@@ -721,38 +721,15 @@ static void armv8pmu_enable_event(struct perf_event *event)
 	 * Enable counter and interrupt, and set the counter to count
 	 * the event that we're interested in.
 	 */
-
-	/*
-	 * Disable counter
-	 */
 	armv8pmu_disable_event_counter(event);
-
-	/*
-	 * Set event.
-	 */
 	armv8pmu_write_event_type(event);
-
-	/*
-	 * Enable interrupt for this counter
-	 */
 	armv8pmu_enable_event_irq(event);
-
-	/*
-	 * Enable counter
-	 */
 	armv8pmu_enable_event_counter(event);
 }
 
 static void armv8pmu_disable_event(struct perf_event *event)
 {
-	/*
-	 * Disable counter
-	 */
 	armv8pmu_disable_event_counter(event);
-
-	/*
-	 * Disable interrupt for this counter
-	 */
 	armv8pmu_disable_event_irq(event);
 }
 
@@ -1266,9 +1243,14 @@ PMUV3_INIT_SIMPLE(armv8_cortex_a76)
 PMUV3_INIT_SIMPLE(armv8_cortex_a77)
 PMUV3_INIT_SIMPLE(armv8_cortex_a78)
 PMUV3_INIT_SIMPLE(armv9_cortex_a510)
+PMUV3_INIT_SIMPLE(armv9_cortex_a520)
 PMUV3_INIT_SIMPLE(armv9_cortex_a710)
+PMUV3_INIT_SIMPLE(armv9_cortex_a715)
+PMUV3_INIT_SIMPLE(armv9_cortex_a720)
 PMUV3_INIT_SIMPLE(armv8_cortex_x1)
 PMUV3_INIT_SIMPLE(armv9_cortex_x2)
+PMUV3_INIT_SIMPLE(armv9_cortex_x3)
+PMUV3_INIT_SIMPLE(armv9_cortex_x4)
 PMUV3_INIT_SIMPLE(armv8_neoverse_e1)
 PMUV3_INIT_SIMPLE(armv8_neoverse_n1)
 PMUV3_INIT_SIMPLE(armv9_neoverse_n2)
@@ -1334,9 +1316,14 @@ static const struct of_device_id armv8_pmu_of_device_ids[] = {
 	{.compatible = "arm,cortex-a77-pmu",	.data = armv8_cortex_a77_pmu_init},
 	{.compatible = "arm,cortex-a78-pmu",	.data = armv8_cortex_a78_pmu_init},
 	{.compatible = "arm,cortex-a510-pmu",	.data = armv9_cortex_a510_pmu_init},
+	{.compatible = "arm,cortex-a520-pmu",	.data = armv9_cortex_a520_pmu_init},
 	{.compatible = "arm,cortex-a710-pmu",	.data = armv9_cortex_a710_pmu_init},
+	{.compatible = "arm,cortex-a715-pmu",	.data = armv9_cortex_a715_pmu_init},
+	{.compatible = "arm,cortex-a720-pmu",	.data = armv9_cortex_a720_pmu_init},
 	{.compatible = "arm,cortex-x1-pmu",	.data = armv8_cortex_x1_pmu_init},
 	{.compatible = "arm,cortex-x2-pmu",	.data = armv9_cortex_x2_pmu_init},
+	{.compatible = "arm,cortex-x3-pmu",	.data = armv9_cortex_x3_pmu_init},
+	{.compatible = "arm,cortex-x4-pmu",	.data = armv9_cortex_x4_pmu_init},
 	{.compatible = "arm,neoverse-e1-pmu",	.data = armv8_neoverse_e1_pmu_init},
 	{.compatible = "arm,neoverse-n1-pmu",	.data = armv8_neoverse_n1_pmu_init},
 	{.compatible = "arm,neoverse-n2-pmu",	.data = armv9_neoverse_n2_pmu_init},

drivers/perf/arm_smmuv3_pmu.c

@@ -115,6 +115,7 @@
 #define SMMU_PMCG_PA_SHIFT              12
 
 #define SMMU_PMCG_EVCNTR_RDONLY         BIT(0)
+#define SMMU_PMCG_HARDEN_DISABLE        BIT(1)
 
 static int cpuhp_state_num;
 
@@ -159,6 +160,20 @@ static inline void smmu_pmu_enable(struct pmu *pmu)
 	writel(SMMU_PMCG_CR_ENABLE, smmu_pmu->reg_base + SMMU_PMCG_CR);
 }
 
+static int smmu_pmu_apply_event_filter(struct smmu_pmu *smmu_pmu,
+				       struct perf_event *event, int idx);
+
+static inline void smmu_pmu_enable_quirk_hip08_09(struct pmu *pmu)
+{
+	struct smmu_pmu *smmu_pmu = to_smmu_pmu(pmu);
+	unsigned int idx;
+
+	for_each_set_bit(idx, smmu_pmu->used_counters, smmu_pmu->num_counters)
+		smmu_pmu_apply_event_filter(smmu_pmu, smmu_pmu->events[idx], idx);
+
+	smmu_pmu_enable(pmu);
+}
+
 static inline void smmu_pmu_disable(struct pmu *pmu)
 {
 	struct smmu_pmu *smmu_pmu = to_smmu_pmu(pmu);
@@ -167,6 +182,22 @@ static inline void smmu_pmu_disable(struct pmu *pmu)
 	writel(0, smmu_pmu->reg_base + SMMU_PMCG_IRQ_CTRL);
 }
 
+static inline void smmu_pmu_disable_quirk_hip08_09(struct pmu *pmu)
+{
+	struct smmu_pmu *smmu_pmu = to_smmu_pmu(pmu);
+	unsigned int idx;
+
+	/*
+	 * The global disable of PMU sometimes fail to stop the counting.
+	 * Harden this by writing an invalid event type to each used counter
+	 * to forcibly stop counting.
+	 */
+	for_each_set_bit(idx, smmu_pmu->used_counters, smmu_pmu->num_counters)
+		writel(0xffff, smmu_pmu->reg_base + SMMU_PMCG_EVTYPER(idx));
+
+	smmu_pmu_disable(pmu);
+}
+
 static inline void smmu_pmu_counter_set_value(struct smmu_pmu *smmu_pmu,
 					      u32 idx, u64 value)
 {
@@ -765,7 +796,10 @@ static void smmu_pmu_get_acpi_options(struct smmu_pmu *smmu_pmu)
 	switch (model) {
 	case IORT_SMMU_V3_PMCG_HISI_HIP08:
 		/* HiSilicon Erratum 162001800 */
-		smmu_pmu->options |= SMMU_PMCG_EVCNTR_RDONLY;
+		smmu_pmu->options |= SMMU_PMCG_EVCNTR_RDONLY | SMMU_PMCG_HARDEN_DISABLE;
+		break;
+	case IORT_SMMU_V3_PMCG_HISI_HIP09:
+		smmu_pmu->options |= SMMU_PMCG_HARDEN_DISABLE;
 		break;
 	}
 
@@ -890,6 +924,16 @@ static int smmu_pmu_probe(struct platform_device *pdev)
 	if (!dev->of_node)
 		smmu_pmu_get_acpi_options(smmu_pmu);
 
+	/*
+	 * For platforms suffer this quirk, the PMU disable sometimes fails to
+	 * stop the counters. This will leads to inaccurate or error counting.
+	 * Forcibly disable the counters with these quirk handler.
+	 */
+	if (smmu_pmu->options & SMMU_PMCG_HARDEN_DISABLE) {
+		smmu_pmu->pmu.pmu_enable = smmu_pmu_enable_quirk_hip08_09;
+		smmu_pmu->pmu.pmu_disable = smmu_pmu_disable_quirk_hip08_09;
+	}
+
 	/* Pick one CPU to be the preferred one to use */
 	smmu_pmu->on_cpu = raw_smp_processor_id();
 	WARN_ON(irq_set_affinity(smmu_pmu->irq, cpumask_of(smmu_pmu->on_cpu)));
@@ -984,6 +1028,7 @@ static void __exit arm_smmu_pmu_exit(void)
 
 module_exit(arm_smmu_pmu_exit);
 
+MODULE_ALIAS("platform:arm-smmu-v3-pmcg");
 MODULE_DESCRIPTION("PMU driver for ARM SMMUv3 Performance Monitors Extension");
 MODULE_AUTHOR("Neil Leeder <nleeder@codeaurora.org>");
 MODULE_AUTHOR("Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>");

drivers/perf/arm_spe_pmu.c

@@ -25,8 +25,7 @@
 #include <linux/kernel.h>
 #include <linux/list.h>
 #include <linux/module.h>
-#include <linux/of_address.h>
-#include <linux/of_device.h>
+#include <linux/of.h>
 #include <linux/perf_event.h>
 #include <linux/perf/arm_pmu.h>
 #include <linux/platform_device.h>

drivers/perf/fsl_imx8_ddr_perf.c

@@ -10,10 +10,9 @@
 #include <linux/io.h>
 #include <linux/module.h>
 #include <linux/of.h>
-#include <linux/of_address.h>
-#include <linux/of_device.h>
 #include <linux/of_irq.h>
 #include <linux/perf_event.h>
+#include <linux/platform_device.h>
 #include <linux/slab.h>
 
 #define COUNTER_CNTL		0x0
@@ -28,6 +27,8 @@
 #define CNTL_CLEAR_MASK		0xFFFFFFFD
 #define CNTL_OVER_MASK		0xFFFFFFFE
 
+#define CNTL_CP_SHIFT		16
+#define CNTL_CP_MASK		(0xFF << CNTL_CP_SHIFT)
 #define CNTL_CSV_SHIFT		24
 #define CNTL_CSV_MASK		(0xFFU << CNTL_CSV_SHIFT)
 
@@ -35,6 +36,8 @@
 #define EVENT_CYCLES_COUNTER	0
 #define NUM_COUNTERS		4
 
+/* For removing bias if cycle counter CNTL.CP is set to 0xf0 */
+#define CYCLES_COUNTER_MASK	0x0FFFFFFF
 #define AXI_MASKING_REVERT	0xffff0000	/* AXI_MASKING(MSB 16bits) + AXI_ID(LSB 16bits) */
 
 #define to_ddr_pmu(p)		container_of(p, struct ddr_pmu, pmu)
@@ -101,6 +104,7 @@ struct ddr_pmu {
 	const struct fsl_ddr_devtype_data *devtype_data;
 	int irq;
 	int id;
+	int active_counter;
 };
 
 static ssize_t ddr_perf_identifier_show(struct device *dev,
@@ -427,6 +431,17 @@ static void ddr_perf_counter_enable(struct ddr_pmu *pmu, int config,
 		writel(0, pmu->base + reg);
 		val = CNTL_EN | CNTL_CLEAR;
 		val |= FIELD_PREP(CNTL_CSV_MASK, config);
+
+		/*
+		 * On i.MX8MP we need to bias the cycle counter to overflow more often.
+		 * We do this by initializing bits [23:16] of the counter value via the
+		 * COUNTER_CTRL Counter Parameter (CP) field.
+		 */
+		if (pmu->devtype_data->quirks & DDR_CAP_AXI_ID_FILTER_ENHANCED) {
+			if (counter == EVENT_CYCLES_COUNTER)
+				val |= FIELD_PREP(CNTL_CP_MASK, 0xf0);
+		}
+
 		writel(val, pmu->base + reg);
 	} else {
 		/* Disable counter */
@@ -466,6 +481,12 @@ static void ddr_perf_event_update(struct perf_event *event)
 	int ret;
 
 	new_raw_count = ddr_perf_read_counter(pmu, counter);
+	/* Remove the bias applied in ddr_perf_counter_enable(). */
+	if (pmu->devtype_data->quirks & DDR_CAP_AXI_ID_FILTER_ENHANCED) {
+		if (counter == EVENT_CYCLES_COUNTER)
+			new_raw_count &= CYCLES_COUNTER_MASK;
+	}
+
 	local64_add(new_raw_count, &event->count);
 
 	/*
@@ -495,6 +516,10 @@ static void ddr_perf_event_start(struct perf_event *event, int flags)
 
 	ddr_perf_counter_enable(pmu, event->attr.config, counter, true);
 
+	if (!pmu->active_counter++)
+		ddr_perf_counter_enable(pmu, EVENT_CYCLES_ID,
+			EVENT_CYCLES_COUNTER, true);
+
 	hwc->state = 0;
 }
 
@@ -548,6 +573,10 @@ static void ddr_perf_event_stop(struct perf_event *event, int flags)
 	ddr_perf_counter_enable(pmu, event->attr.config, counter, false);
 	ddr_perf_event_update(event);
 
+	if (!--pmu->active_counter)
+		ddr_perf_counter_enable(pmu, EVENT_CYCLES_ID,
+			EVENT_CYCLES_COUNTER, false);
+
 	hwc->state |= PERF_HES_STOPPED;
 }
 
@@ -565,25 +594,10 @@ static void ddr_perf_event_del(struct perf_event *event, int flags)
 
 static void ddr_perf_pmu_enable(struct pmu *pmu)
 {
-	struct ddr_pmu *ddr_pmu = to_ddr_pmu(pmu);
-
-	/* enable cycle counter if cycle is not active event list */
-	if (ddr_pmu->events[EVENT_CYCLES_COUNTER] == NULL)
-		ddr_perf_counter_enable(ddr_pmu,
-				      EVENT_CYCLES_ID,
-				      EVENT_CYCLES_COUNTER,
-				      true);
 }
 
 static void ddr_perf_pmu_disable(struct pmu *pmu)
 {
-	struct ddr_pmu *ddr_pmu = to_ddr_pmu(pmu);
-
-	if (ddr_pmu->events[EVENT_CYCLES_COUNTER] == NULL)
-		ddr_perf_counter_enable(ddr_pmu,
-				      EVENT_CYCLES_ID,
-				      EVENT_CYCLES_COUNTER,
-				      false);
 }
 
 static int ddr_perf_init(struct ddr_pmu *pmu, void __iomem *base,

drivers/perf/fsl_imx9_ddr_perf.c

@@ -7,9 +7,7 @@
 #include <linux/io.h>
 #include <linux/module.h>
 #include <linux/of.h>
-#include <linux/of_address.h>
-#include <linux/of_device.h>
-#include <linux/of_irq.h>
+#include <linux/platform_device.h>
 #include <linux/perf_event.h>
 
 /* Performance monitor configuration */

drivers/perf/hisilicon/hisi_pcie_pmu.c

@@ -665,8 +665,8 @@ static int hisi_pcie_pmu_online_cpu(unsigned int cpu, struct hlist_node *node)
 	struct hisi_pcie_pmu *pcie_pmu = hlist_entry_safe(node, struct hisi_pcie_pmu, node);
 
 	if (pcie_pmu->on_cpu == -1) {
-		pcie_pmu->on_cpu = cpu;
-		WARN_ON(irq_set_affinity(pcie_pmu->irq, cpumask_of(cpu)));
+		pcie_pmu->on_cpu = cpumask_local_spread(0, dev_to_node(&pcie_pmu->pdev->dev));
+		WARN_ON(irq_set_affinity(pcie_pmu->irq, cpumask_of(pcie_pmu->on_cpu)));
 	}
 
 	return 0;
@@ -676,14 +676,23 @@ static int hisi_pcie_pmu_offline_cpu(unsigned int cpu, struct hlist_node *node)
 {
 	struct hisi_pcie_pmu *pcie_pmu = hlist_entry_safe(node, struct hisi_pcie_pmu, node);
 	unsigned int target;
+	cpumask_t mask;
+	int numa_node;
 
 	/* Nothing to do if this CPU doesn't own the PMU */
 	if (pcie_pmu->on_cpu != cpu)
 		return 0;
 
 	pcie_pmu->on_cpu = -1;
-	/* Choose a new CPU from all online cpus. */
-	target = cpumask_any_but(cpu_online_mask, cpu);
+
+	/* Choose a local CPU from all online cpus. */
+	numa_node = dev_to_node(&pcie_pmu->pdev->dev);
+	if (cpumask_and(&mask, cpumask_of_node(numa_node), cpu_online_mask) &&
+	    cpumask_andnot(&mask, &mask, cpumask_of(cpu)))
+		target = cpumask_any(&mask);
+	else
+		target = cpumask_any_but(cpu_online_mask, cpu);
+
 	if (target >= nr_cpu_ids) {
 		pci_err(pcie_pmu->pdev, "There is no CPU to set\n");
 		return 0;

drivers/perf/marvell_cn10k_ddr_pmu.c

@@ -8,11 +8,10 @@
 #include <linux/io.h>
 #include <linux/module.h>
 #include <linux/of.h>
-#include <linux/of_address.h>
-#include <linux/of_device.h>
 #include <linux/perf_event.h>
 #include <linux/hrtimer.h>
 #include <linux/acpi.h>
+#include <linux/platform_device.h>
 
 /* Performance Counters Operating Mode Control Registers */
 #define DDRC_PERF_CNT_OP_MODE_CTRL	0x8020

drivers/perf/marvell_cn10k_tad_pmu.c

@@ -6,10 +6,9 @@
 
 #define pr_fmt(fmt) "tad_pmu: " fmt
 
+#include <linux/io.h>
 #include <linux/module.h>
 #include <linux/of.h>
-#include <linux/of_address.h>
-#include <linux/of_device.h>
 #include <linux/cpuhotplug.h>
 #include <linux/perf_event.h>
 #include <linux/platform_device.h>

drivers/perf/xgene_pmu.c

@@ -1833,7 +1833,6 @@ static int xgene_pmu_probe(struct platform_device *pdev)
 	const struct xgene_pmu_data *dev_data;
 	const struct of_device_id *of_id;
 	struct xgene_pmu *xgene_pmu;
-	struct resource *res;
 	int irq, rc;
 	int version;
 
@@ -1883,8 +1882,7 @@ static int xgene_pmu_probe(struct platform_device *pdev)
 	xgene_pmu->version = version;
 	dev_info(&pdev->dev, "X-Gene PMU version %d\n", xgene_pmu->version);
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	xgene_pmu->pcppmu_csr = devm_ioremap_resource(&pdev->dev, res);
+	xgene_pmu->pcppmu_csr = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(xgene_pmu->pcppmu_csr)) {
 		dev_err(&pdev->dev, "ioremap failed for PCP PMU resource\n");
 		return PTR_ERR(xgene_pmu->pcppmu_csr);

include/linux/acpi_iort.h

@@ -21,6 +21,7 @@
  */
 #define IORT_SMMU_V3_PMCG_GENERIC        0x00000000 /* Generic SMMUv3 PMCG */
 #define IORT_SMMU_V3_PMCG_HISI_HIP08     0x00000001 /* HiSilicon HIP08 PMCG */
+#define IORT_SMMU_V3_PMCG_HISI_HIP09     0x00000002 /* HiSilicon HIP09 PMCG */
 
 int iort_register_domain_token(int trans_id, phys_addr_t base,
 			       struct fwnode_handle *fw_node);

include/linux/perf/arm_pmu.h

@@ -187,5 +187,6 @@ void armpmu_free_irq(int irq, int cpu);
 #endif /* CONFIG_ARM_PMU */
 
 #define ARMV8_SPE_PDEV_NAME "arm,spe-v1"
+#define ARMV8_TRBE_PDEV_NAME "arm,trbe"
 
 #endif /* __ARM_PMU_H__ */

include/linux/perf_event.h

@@ -1316,15 +1316,31 @@ extern int perf_event_output(struct perf_event *event,
 			     struct pt_regs *regs);
 
 static inline bool
-is_default_overflow_handler(struct perf_event *event)
+__is_default_overflow_handler(perf_overflow_handler_t overflow_handler)
 {
-	if (likely(event->overflow_handler == perf_event_output_forward))
+	if (likely(overflow_handler == perf_event_output_forward))
 		return true;
-	if (unlikely(event->overflow_handler == perf_event_output_backward))
+	if (unlikely(overflow_handler == perf_event_output_backward))
 		return true;
 	return false;
 }
 
+#define is_default_overflow_handler(event) \
+	__is_default_overflow_handler((event)->overflow_handler)
+
+#ifdef CONFIG_BPF_SYSCALL
+static inline bool uses_default_overflow_handler(struct perf_event *event)
+{
+	if (likely(is_default_overflow_handler(event)))
+		return true;
+
+	return __is_default_overflow_handler(event->orig_overflow_handler);
+}
+#else
+#define uses_default_overflow_handler(event) \
+	is_default_overflow_handler(event)
+#endif
+
 extern void
 perf_event_header__init_id(struct perf_event_header *header,
 			   struct perf_sample_data *data,