Merge tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux

Pull arm64 updates from Will Deacon: "The highlights are support for Arm's "Permission Overlay Extension" using memory protection keys, support for running as a protected guest on Android as well as perf support for a bunch of new interconnect PMUs. Summary: ACPI: - Enable PMCG erratum workaround for HiSilicon HIP10 and 11 platforms. - Ensure arm64-specific IORT header is covered by MAINTAINERS. CPU Errata: - Enable workaround for hardware access/dirty issue on Ampere-1A cores. Memory management: - Define PHYSMEM_END to fix a crash in the amdgpu driver. - Avoid tripping over invalid kernel mappings on the kexec() path. - Userspace support for the Permission Overlay Extension (POE) using protection keys. Perf and PMUs: - Add support for the "fixed instruction counter" extension in the CPU PMU architecture. - Extend and fix the event encodings for Apple's M1 CPU PMU. - Allow LSM hooks to decide on SPE permissions for physical profiling. - Add support for the CMN S3 and NI-700 PMUs. Confidential Computing: - Add support for booting an arm64 kernel as a protected guest under Android's "Protected KVM" (pKVM) hypervisor. Selftests: - Fix vector length issues in the SVE/SME sigreturn tests - Fix build warning in the ptrace tests. Timers: - Add support for PR_{G,S}ET_TSC so that 'rr' can deal with non-determinism arising from the architected counter. Miscellaneous: - Rework our IPI-based CPU stopping code to try NMIs if regular IPIs don't succeed. - Minor fixes and cleanups" * tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (94 commits) perf: arm-ni: Fix an NULL vs IS_ERR() bug arm64: hibernate: Fix warning for cast from restricted gfp_t arm64: esr: Define ESR_ELx_EC_* constants as UL arm64: pkeys: remove redundant WARN perf: arm_pmuv3: Use BR_RETIRED for HW branch event if enabled MAINTAINERS: List Arm interconnect PMUs as supported perf: Add driver for Arm NI-700 interconnect PMU dt-bindings/perf: Add Arm NI-700 PMU perf/arm-cmn: Improve format attr printing perf/arm-cmn: Clean up unnecessary NUMA_NO_NODE check arm64/mm: use lm_alias() with addresses passed to memblock_free() mm: arm64: document why pte is not advanced in contpte_ptep_set_access_flags() arm64: Expose the end of the linear map in PHYSMEM_END arm64: trans_pgd: mark PTEs entries as valid to avoid dead kexec() arm64/mm: Delete __init region from memblock.reserved perf/arm-cmn: Support CMN S3 dt-bindings: perf: arm-cmn: Add CMN S3 perf/arm-cmn: Refactor DTC PMU register access perf/arm-cmn: Make cycle counts less surprising perf/arm-cmn: Improve build-time assertion ...
author: Linus Torvalds <torvalds@linux-foundation.org> 2024-09-16 06:55:07 +0200
committer: Linus Torvalds <torvalds@linux-foundation.org> 2024-09-16 06:55:07 +0200
commit: 114143a595895c03fbefccfd8346fc51fb4908ed (patch)
tree: 161f753ad69a2aabe7cd7ac7cc493c3678dbd760 /drivers/perf
parent: 8617d7d6298f54dfef4038281863270b5864fe83 (diff)
parent: 75078ba2b38a38d94017bd334f71aaed205e30a4 (diff)
15 files changed, 1243 insertions, 369 deletions
diff --git a/drivers/perf/Kconfig b/drivers/perf/Kconfig
index aa9530b4064f..bab8ba64162f 100644
--- a/drivers/perf/Kconfig
+++ b/drivers/perf/Kconfig
@@ -48,6 +48,13 @@ config ARM_CMN
 	  Support for PMU events monitoring on the Arm CMN-600 Coherent Mesh
 	  Network interconnect.
 
+config ARM_NI
+	tristate "Arm NI-700 PMU support"
+	depends on ARM64 || COMPILE_TEST
+	help
+	  Support for PMU events monitoring on the Arm NI-700 Network-on-Chip
+	  interconnect and family.
+
 config ARM_PMU
 	depends on ARM || ARM64
 	bool "ARM PMU framework"
diff --git a/drivers/perf/Makefile b/drivers/perf/Makefile
index d43df81d52f7..8268f38e42c5 100644
--- a/drivers/perf/Makefile
+++ b/drivers/perf/Makefile
@@ -3,6 +3,7 @@ obj-$(CONFIG_ARM_CCI_PMU) += arm-cci.o
 obj-$(CONFIG_ARM_CCN) += arm-ccn.o
 obj-$(CONFIG_ARM_CMN) += arm-cmn.o
 obj-$(CONFIG_ARM_DSU_PMU) += arm_dsu_pmu.o
+obj-$(CONFIG_ARM_NI) += arm-ni.o
 obj-$(CONFIG_ARM_PMU) += arm_pmu.o arm_pmu_platform.o
 obj-$(CONFIG_ARM_PMU_ACPI) += arm_pmu_acpi.o
 obj-$(CONFIG_ARM_PMUV3) += arm_pmuv3.o
diff --git a/drivers/perf/alibaba_uncore_drw_pmu.c b/drivers/perf/alibaba_uncore_drw_pmu.c
index 38a2947ae813..c6ff1bc7d336 100644
--- a/drivers/perf/alibaba_uncore_drw_pmu.c
+++ b/drivers/perf/alibaba_uncore_drw_pmu.c
@@ -400,7 +400,7 @@ static irqreturn_t ali_drw_pmu_isr(int irq_num, void *data)
 			}
 
 			/* clear common counter intr status */
-			clr_status = FIELD_PREP(ALI_DRW_PMCOM_CNT_OV_INTR_MASK, 1);
+			clr_status = FIELD_PREP(ALI_DRW_PMCOM_CNT_OV_INTR_MASK, status);
 			writel(clr_status,
 			       drw_pmu->cfg_base + ALI_DRW_PMU_OV_INTR_CLR);
 		}
diff --git a/drivers/perf/apple_m1_cpu_pmu.c b/drivers/perf/apple_m1_cpu_pmu.c
index f322e5ca1114..1d4d01e1275e 100644
--- a/drivers/perf/apple_m1_cpu_pmu.c
+++ b/drivers/perf/apple_m1_cpu_pmu.c
@@ -47,46 +47,79 @@
  * implementations, we'll have to introduce per cpu-type tables.
  */
 enum m1_pmu_events {
-	M1_PMU_PERFCTR_UNKNOWN_01	= 0x01,
-	M1_PMU_PERFCTR_CPU_CYCLES	= 0x02,
-	M1_PMU_PERFCTR_INSTRUCTIONS	= 0x8c,
-	M1_PMU_PERFCTR_UNKNOWN_8d	= 0x8d,
-	M1_PMU_PERFCTR_UNKNOWN_8e	= 0x8e,
-	M1_PMU_PERFCTR_UNKNOWN_8f	= 0x8f,
-	M1_PMU_PERFCTR_UNKNOWN_90	= 0x90,
-	M1_PMU_PERFCTR_UNKNOWN_93	= 0x93,
-	M1_PMU_PERFCTR_UNKNOWN_94	= 0x94,
-	M1_PMU_PERFCTR_UNKNOWN_95	= 0x95,
-	M1_PMU_PERFCTR_UNKNOWN_96	= 0x96,
-	M1_PMU_PERFCTR_UNKNOWN_97	= 0x97,
-	M1_PMU_PERFCTR_UNKNOWN_98	= 0x98,
-	M1_PMU_PERFCTR_UNKNOWN_99	= 0x99,
-	M1_PMU_PERFCTR_UNKNOWN_9a	= 0x9a,
-	M1_PMU_PERFCTR_UNKNOWN_9b	= 0x9b,
-	M1_PMU_PERFCTR_UNKNOWN_9c	= 0x9c,
-	M1_PMU_PERFCTR_UNKNOWN_9f	= 0x9f,
-	M1_PMU_PERFCTR_UNKNOWN_bf	= 0xbf,
-	M1_PMU_PERFCTR_UNKNOWN_c0	= 0xc0,
-	M1_PMU_PERFCTR_UNKNOWN_c1	= 0xc1,
-	M1_PMU_PERFCTR_UNKNOWN_c4	= 0xc4,
-	M1_PMU_PERFCTR_UNKNOWN_c5	= 0xc5,
-	M1_PMU_PERFCTR_UNKNOWN_c6	= 0xc6,
-	M1_PMU_PERFCTR_UNKNOWN_c8	= 0xc8,
-	M1_PMU_PERFCTR_UNKNOWN_ca	= 0xca,
-	M1_PMU_PERFCTR_UNKNOWN_cb	= 0xcb,
-	M1_PMU_PERFCTR_UNKNOWN_f5	= 0xf5,
-	M1_PMU_PERFCTR_UNKNOWN_f6	= 0xf6,
-	M1_PMU_PERFCTR_UNKNOWN_f7	= 0xf7,
-	M1_PMU_PERFCTR_UNKNOWN_f8	= 0xf8,
-	M1_PMU_PERFCTR_UNKNOWN_fd	= 0xfd,
-	M1_PMU_PERFCTR_LAST		= M1_PMU_CFG_EVENT,
+	M1_PMU_PERFCTR_RETIRE_UOP				= 0x1,
+	M1_PMU_PERFCTR_CORE_ACTIVE_CYCLE			= 0x2,
+	M1_PMU_PERFCTR_L1I_TLB_FILL				= 0x4,
+	M1_PMU_PERFCTR_L1D_TLB_FILL				= 0x5,
+	M1_PMU_PERFCTR_MMU_TABLE_WALK_INSTRUCTION		= 0x7,
+	M1_PMU_PERFCTR_MMU_TABLE_WALK_DATA			= 0x8,
+	M1_PMU_PERFCTR_L2_TLB_MISS_INSTRUCTION			= 0xa,
+	M1_PMU_PERFCTR_L2_TLB_MISS_DATA				= 0xb,
+	M1_PMU_PERFCTR_MMU_VIRTUAL_MEMORY_FAULT_NONSPEC		= 0xd,
+	M1_PMU_PERFCTR_SCHEDULE_UOP				= 0x52,
+	M1_PMU_PERFCTR_INTERRUPT_PENDING			= 0x6c,
+	M1_PMU_PERFCTR_MAP_STALL_DISPATCH			= 0x70,
+	M1_PMU_PERFCTR_MAP_REWIND				= 0x75,
+	M1_PMU_PERFCTR_MAP_STALL				= 0x76,
+	M1_PMU_PERFCTR_MAP_INT_UOP				= 0x7c,
+	M1_PMU_PERFCTR_MAP_LDST_UOP				= 0x7d,
+	M1_PMU_PERFCTR_MAP_SIMD_UOP				= 0x7e,
+	M1_PMU_PERFCTR_FLUSH_RESTART_OTHER_NONSPEC		= 0x84,
+	M1_PMU_PERFCTR_INST_ALL					= 0x8c,
+	M1_PMU_PERFCTR_INST_BRANCH				= 0x8d,
+	M1_PMU_PERFCTR_INST_BRANCH_CALL				= 0x8e,
+	M1_PMU_PERFCTR_INST_BRANCH_RET				= 0x8f,
+	M1_PMU_PERFCTR_INST_BRANCH_TAKEN			= 0x90,
+	M1_PMU_PERFCTR_INST_BRANCH_INDIR			= 0x93,
+	M1_PMU_PERFCTR_INST_BRANCH_COND				= 0x94,
+	M1_PMU_PERFCTR_INST_INT_LD				= 0x95,
+	M1_PMU_PERFCTR_INST_INT_ST				= 0x96,
+	M1_PMU_PERFCTR_INST_INT_ALU				= 0x97,
+	M1_PMU_PERFCTR_INST_SIMD_LD				= 0x98,
+	M1_PMU_PERFCTR_INST_SIMD_ST				= 0x99,
+	M1_PMU_PERFCTR_INST_SIMD_ALU				= 0x9a,
+	M1_PMU_PERFCTR_INST_LDST				= 0x9b,
+	M1_PMU_PERFCTR_INST_BARRIER				= 0x9c,
+	M1_PMU_PERFCTR_UNKNOWN_9f				= 0x9f,
+	M1_PMU_PERFCTR_L1D_TLB_ACCESS				= 0xa0,
+	M1_PMU_PERFCTR_L1D_TLB_MISS				= 0xa1,
+	M1_PMU_PERFCTR_L1D_CACHE_MISS_ST			= 0xa2,
+	M1_PMU_PERFCTR_L1D_CACHE_MISS_LD			= 0xa3,
+	M1_PMU_PERFCTR_LD_UNIT_UOP				= 0xa6,
+	M1_PMU_PERFCTR_ST_UNIT_UOP				= 0xa7,
+	M1_PMU_PERFCTR_L1D_CACHE_WRITEBACK			= 0xa8,
+	M1_PMU_PERFCTR_LDST_X64_UOP				= 0xb1,
+	M1_PMU_PERFCTR_LDST_XPG_UOP				= 0xb2,
+	M1_PMU_PERFCTR_ATOMIC_OR_EXCLUSIVE_SUCC			= 0xb3,
+	M1_PMU_PERFCTR_ATOMIC_OR_EXCLUSIVE_FAIL			= 0xb4,
+	M1_PMU_PERFCTR_L1D_CACHE_MISS_LD_NONSPEC		= 0xbf,
+	M1_PMU_PERFCTR_L1D_CACHE_MISS_ST_NONSPEC		= 0xc0,
+	M1_PMU_PERFCTR_L1D_TLB_MISS_NONSPEC			= 0xc1,
+	M1_PMU_PERFCTR_ST_MEMORY_ORDER_VIOLATION_NONSPEC	= 0xc4,
+	M1_PMU_PERFCTR_BRANCH_COND_MISPRED_NONSPEC		= 0xc5,
+	M1_PMU_PERFCTR_BRANCH_INDIR_MISPRED_NONSPEC		= 0xc6,
+	M1_PMU_PERFCTR_BRANCH_RET_INDIR_MISPRED_NONSPEC		= 0xc8,
+	M1_PMU_PERFCTR_BRANCH_CALL_INDIR_MISPRED_NONSPEC	= 0xca,
+	M1_PMU_PERFCTR_BRANCH_MISPRED_NONSPEC			= 0xcb,
+	M1_PMU_PERFCTR_L1I_TLB_MISS_DEMAND			= 0xd4,
+	M1_PMU_PERFCTR_MAP_DISPATCH_BUBBLE			= 0xd6,
+	M1_PMU_PERFCTR_L1I_CACHE_MISS_DEMAND			= 0xdb,
+	M1_PMU_PERFCTR_FETCH_RESTART				= 0xde,
+	M1_PMU_PERFCTR_ST_NT_UOP				= 0xe5,
+	M1_PMU_PERFCTR_LD_NT_UOP				= 0xe6,
+	M1_PMU_PERFCTR_UNKNOWN_f5				= 0xf5,
+	M1_PMU_PERFCTR_UNKNOWN_f6				= 0xf6,
+	M1_PMU_PERFCTR_UNKNOWN_f7				= 0xf7,
+	M1_PMU_PERFCTR_UNKNOWN_f8				= 0xf8,
+	M1_PMU_PERFCTR_UNKNOWN_fd				= 0xfd,
+	M1_PMU_PERFCTR_LAST					= M1_PMU_CFG_EVENT,
 
 	/*
 	 * From this point onwards, these are not actual HW events,
 	 * but attributes that get stored in hw->config_base.
 	 */
-	M1_PMU_CFG_COUNT_USER		= BIT(8),
-	M1_PMU_CFG_COUNT_KERNEL		= BIT(9),
+	M1_PMU_CFG_COUNT_USER					= BIT(8),
+	M1_PMU_CFG_COUNT_KERNEL					= BIT(9),
 };
 
 /*
@@ -96,46 +129,45 @@ enum m1_pmu_events {
  * counters had strange affinities.
  */
 static const u16 m1_pmu_event_affinity[M1_PMU_PERFCTR_LAST + 1] = {
-	[0 ... M1_PMU_PERFCTR_LAST]	= ANY_BUT_0_1,
-	[M1_PMU_PERFCTR_UNKNOWN_01]	= BIT(7),
-	[M1_PMU_PERFCTR_CPU_CYCLES]	= ANY_BUT_0_1 | BIT(0),
-	[M1_PMU_PERFCTR_INSTRUCTIONS]	= BIT(7) | BIT(1),
-	[M1_PMU_PERFCTR_UNKNOWN_8d]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_8e]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_8f]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_90]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_93]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_94]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_95]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_96]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_97]	= BIT(7),
-	[M1_PMU_PERFCTR_UNKNOWN_98]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_99]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_9a]	= BIT(7),
-	[M1_PMU_PERFCTR_UNKNOWN_9b]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_9c]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_9f]	= BIT(7),
-	[M1_PMU_PERFCTR_UNKNOWN_bf]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_c0]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_c1]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_c4]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_c5]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_c6]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_c8]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_ca]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_cb]	= ONLY_5_6_7,
-	[M1_PMU_PERFCTR_UNKNOWN_f5]	= ONLY_2_4_6,
-	[M1_PMU_PERFCTR_UNKNOWN_f6]	= ONLY_2_4_6,
-	[M1_PMU_PERFCTR_UNKNOWN_f7]	= ONLY_2_4_6,
-	[M1_PMU_PERFCTR_UNKNOWN_f8]	= ONLY_2_TO_7,
-	[M1_PMU_PERFCTR_UNKNOWN_fd]	= ONLY_2_4_6,
+	[0 ... M1_PMU_PERFCTR_LAST]				= ANY_BUT_0_1,
+	[M1_PMU_PERFCTR_RETIRE_UOP]				= BIT(7),
+	[M1_PMU_PERFCTR_CORE_ACTIVE_CYCLE]			= ANY_BUT_0_1 | BIT(0),
+	[M1_PMU_PERFCTR_INST_ALL]				= BIT(7) | BIT(1),
+	[M1_PMU_PERFCTR_INST_BRANCH]				= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_INST_BRANCH_CALL]			= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_INST_BRANCH_RET]			= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_INST_BRANCH_TAKEN]			= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_INST_BRANCH_INDIR]			= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_INST_BRANCH_COND]			= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_INST_INT_LD]				= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_INST_INT_ST]				= BIT(7),
+	[M1_PMU_PERFCTR_INST_INT_ALU]				= BIT(7),
+	[M1_PMU_PERFCTR_INST_SIMD_LD]				= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_INST_SIMD_ST]				= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_INST_SIMD_ALU]				= BIT(7),
+	[M1_PMU_PERFCTR_INST_LDST]				= BIT(7),
+	[M1_PMU_PERFCTR_INST_BARRIER]				= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_UNKNOWN_9f]				= BIT(7),
+	[M1_PMU_PERFCTR_L1D_CACHE_MISS_LD_NONSPEC]		= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_L1D_CACHE_MISS_ST_NONSPEC]		= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_L1D_TLB_MISS_NONSPEC]			= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_ST_MEMORY_ORDER_VIOLATION_NONSPEC]	= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_BRANCH_COND_MISPRED_NONSPEC]		= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_BRANCH_INDIR_MISPRED_NONSPEC]		= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_BRANCH_RET_INDIR_MISPRED_NONSPEC]	= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_BRANCH_CALL_INDIR_MISPRED_NONSPEC]	= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_BRANCH_MISPRED_NONSPEC]			= ONLY_5_6_7,
+	[M1_PMU_PERFCTR_UNKNOWN_f5]				= ONLY_2_4_6,
+	[M1_PMU_PERFCTR_UNKNOWN_f6]				= ONLY_2_4_6,
+	[M1_PMU_PERFCTR_UNKNOWN_f7]				= ONLY_2_4_6,
+	[M1_PMU_PERFCTR_UNKNOWN_f8]				= ONLY_2_TO_7,
+	[M1_PMU_PERFCTR_UNKNOWN_fd]				= ONLY_2_4_6,
 };
 
 static const unsigned m1_pmu_perf_map[PERF_COUNT_HW_MAX] = {
 	PERF_MAP_ALL_UNSUPPORTED,
-	[PERF_COUNT_HW_CPU_CYCLES]	= M1_PMU_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	= M1_PMU_PERFCTR_INSTRUCTIONS,
-	/* No idea about the rest yet */
+	[PERF_COUNT_HW_CPU_CYCLES]		= M1_PMU_PERFCTR_CORE_ACTIVE_CYCLE,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= M1_PMU_PERFCTR_INST_ALL,
 };
 
 /* sysfs definitions */
@@ -154,8 +186,8 @@ static ssize_t m1_pmu_events_sysfs_show(struct device *dev,
 	PMU_EVENT_ATTR_ID(name, m1_pmu_events_sysfs_show, config)
 
 static struct attribute *m1_pmu_event_attrs[] = {
-	M1_PMU_EVENT_ATTR(cycles, M1_PMU_PERFCTR_CPU_CYCLES),
-	M1_PMU_EVENT_ATTR(instructions, M1_PMU_PERFCTR_INSTRUCTIONS),
+	M1_PMU_EVENT_ATTR(cycles, M1_PMU_PERFCTR_CORE_ACTIVE_CYCLE),
+	M1_PMU_EVENT_ATTR(instructions, M1_PMU_PERFCTR_INST_ALL),
 	NULL,
 };
 
@@ -400,7 +432,7 @@ static irqreturn_t m1_pmu_handle_irq(struct arm_pmu *cpu_pmu)
 
 	regs = get_irq_regs();
 
-	for (idx = 0; idx < cpu_pmu->num_events; idx++) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, M1_PMU_NR_COUNTERS) {
 		struct perf_event *event = cpuc->events[idx];
 		struct perf_sample_data data;
 
@@ -560,7 +592,7 @@ static int m1_pmu_init(struct arm_pmu *cpu_pmu, u32 flags)
 	cpu_pmu->reset		  = m1_pmu_reset;
 	cpu_pmu->set_event_filter = m1_pmu_set_event_filter;
 
-	cpu_pmu->num_events	  = M1_PMU_NR_COUNTERS;
+	bitmap_set(cpu_pmu->cntr_mask, 0, M1_PMU_NR_COUNTERS);
 	cpu_pmu->attr_groups[ARMPMU_ATTR_GROUP_EVENTS] = &m1_pmu_events_attr_group;
 	cpu_pmu->attr_groups[ARMPMU_ATTR_GROUP_FORMATS] = &m1_pmu_format_attr_group;
 	return 0;
diff --git a/drivers/perf/arm-cmn.c b/drivers/perf/arm-cmn.c
index c932d9d355cf..397a46410f7c 100644
--- a/drivers/perf/arm-cmn.c
+++ b/drivers/perf/arm-cmn.c
@@ -24,14 +24,6 @@
 #define CMN_NI_NODE_ID			GENMASK_ULL(31, 16)
 #define CMN_NI_LOGICAL_ID		GENMASK_ULL(47, 32)
 
-#define CMN_NODEID_DEVID(reg)		((reg) & 3)
-#define CMN_NODEID_EXT_DEVID(reg)	((reg) & 1)
-#define CMN_NODEID_PID(reg)		(((reg) >> 2) & 1)
-#define CMN_NODEID_EXT_PID(reg)		(((reg) >> 1) & 3)
-#define CMN_NODEID_1x1_PID(reg)		(((reg) >> 2) & 7)
-#define CMN_NODEID_X(reg, bits)		((reg) >> (3 + (bits)))
-#define CMN_NODEID_Y(reg, bits)		(((reg) >> 3) & ((1U << (bits)) - 1))
-
 #define CMN_CHILD_INFO			0x0080
 #define CMN_CI_CHILD_COUNT		GENMASK_ULL(15, 0)
 #define CMN_CI_CHILD_PTR_OFFSET		GENMASK_ULL(31, 16)
@@ -43,6 +35,9 @@
 #define CMN_MAX_XPS			(CMN_MAX_DIMENSION * CMN_MAX_DIMENSION)
 #define CMN_MAX_DTMS			(CMN_MAX_XPS + (CMN_MAX_DIMENSION - 1) * 4)
 
+/* Currently XPs are the node type we can have most of; others top out at 128 */
+#define CMN_MAX_NODES_PER_EVENT		CMN_MAX_XPS
+
 /* The CFG node has various info besides the discovery tree */
 #define CMN_CFGM_PERIPH_ID_01		0x0008
 #define CMN_CFGM_PID0_PART_0		GENMASK_ULL(7, 0)
@@ -50,24 +45,28 @@
 #define CMN_CFGM_PERIPH_ID_23		0x0010
 #define CMN_CFGM_PID2_REVISION		GENMASK_ULL(7, 4)
 
-#define CMN_CFGM_INFO_GLOBAL		0x900
+#define CMN_CFGM_INFO_GLOBAL		0x0900
 #define CMN_INFO_MULTIPLE_DTM_EN	BIT_ULL(63)
 #define CMN_INFO_RSP_VC_NUM		GENMASK_ULL(53, 52)
 #define CMN_INFO_DAT_VC_NUM		GENMASK_ULL(51, 50)
+#define CMN_INFO_DEVICE_ISO_ENABLE	BIT_ULL(44)
 
-#define CMN_CFGM_INFO_GLOBAL_1		0x908
+#define CMN_CFGM_INFO_GLOBAL_1		0x0908
 #define CMN_INFO_SNP_VC_NUM		GENMASK_ULL(3, 2)
 #define CMN_INFO_REQ_VC_NUM		GENMASK_ULL(1, 0)
 
 /* XPs also have some local topology info which has uses too */
 #define CMN_MXP__CONNECT_INFO(p)	(0x0008 + 8 * (p))
-#define CMN__CONNECT_INFO_DEVICE_TYPE	GENMASK_ULL(4, 0)
+#define CMN__CONNECT_INFO_DEVICE_TYPE	GENMASK_ULL(5, 0)
 
 #define CMN_MAX_PORTS			6
 #define CI700_CONNECT_INFO_P2_5_OFFSET	0x10
 
 /* PMU registers occupy the 3rd 4KB page of each node's region */
 #define CMN_PMU_OFFSET			0x2000
+/* ...except when they don't :( */
+#define CMN_S3_DTM_OFFSET		0xa000
+#define CMN_S3_PMU_OFFSET		0xd900
 
 /* For most nodes, this is all there is */
 #define CMN_PMU_EVENT_SEL		0x000
@@ -78,7 +77,8 @@
 /* Technically this is 4 bits wide on DNs, but we only use 2 there anyway */
 #define CMN__PMU_OCCUP1_ID		GENMASK_ULL(34, 32)
 
-/* HN-Ps are weird... */
+/* Some types are designed to coexist with another device in the same node */
+#define CMN_CCLA_PMU_EVENT_SEL		0x008
 #define CMN_HNP_PMU_EVENT_SEL		0x008
 
 /* DTMs live in the PMU space of XP registers */
@@ -123,27 +123,28 @@
 /* The DTC node is where the magic happens */
 #define CMN_DT_DTC_CTL			0x0a00
 #define CMN_DT_DTC_CTL_DT_EN		BIT(0)
+#define CMN_DT_DTC_CTL_CG_DISABLE	BIT(10)
 
 /* DTC counters are paired in 64-bit registers on a 16-byte stride. Yuck */
 #define _CMN_DT_CNT_REG(n)		((((n) / 2) * 4 + (n) % 2) * 4)
-#define CMN_DT_PMEVCNT(n)		(CMN_PMU_OFFSET + _CMN_DT_CNT_REG(n))
-#define CMN_DT_PMCCNTR			(CMN_PMU_OFFSET + 0x40)
+#define CMN_DT_PMEVCNT(dtc, n)		((dtc)->pmu_base + _CMN_DT_CNT_REG(n))
+#define CMN_DT_PMCCNTR(dtc)		((dtc)->pmu_base + 0x40)
 
-#define CMN_DT_PMEVCNTSR(n)		(CMN_PMU_OFFSET + 0x50 + _CMN_DT_CNT_REG(n))
-#define CMN_DT_PMCCNTRSR		(CMN_PMU_OFFSET + 0x90)
+#define CMN_DT_PMEVCNTSR(dtc, n)	((dtc)->pmu_base + 0x50 + _CMN_DT_CNT_REG(n))
+#define CMN_DT_PMCCNTRSR(dtc)		((dtc)->pmu_base + 0x90)
 
-#define CMN_DT_PMCR			(CMN_PMU_OFFSET + 0x100)
+#define CMN_DT_PMCR(dtc)		((dtc)->pmu_base + 0x100)
 #define CMN_DT_PMCR_PMU_EN		BIT(0)
 #define CMN_DT_PMCR_CNTR_RST		BIT(5)
 #define CMN_DT_PMCR_OVFL_INTR_EN	BIT(6)
 
-#define CMN_DT_PMOVSR			(CMN_PMU_OFFSET + 0x118)
-#define CMN_DT_PMOVSR_CLR		(CMN_PMU_OFFSET + 0x120)
+#define CMN_DT_PMOVSR(dtc)		((dtc)->pmu_base + 0x118)
+#define CMN_DT_PMOVSR_CLR(dtc)		((dtc)->pmu_base + 0x120)
 
-#define CMN_DT_PMSSR			(CMN_PMU_OFFSET + 0x128)
+#define CMN_DT_PMSSR(dtc)		((dtc)->pmu_base + 0x128)
 #define CMN_DT_PMSSR_SS_STATUS(n)	BIT(n)
 
-#define CMN_DT_PMSRR			(CMN_PMU_OFFSET + 0x130)
+#define CMN_DT_PMSRR(dtc)		((dtc)->pmu_base + 0x130)
 #define CMN_DT_PMSRR_SS_REQ		BIT(0)
 
 #define CMN_DT_NUM_COUNTERS		8
@@ -198,10 +199,11 @@ enum cmn_model {
 	CMN650 = 2,
 	CMN700 = 4,
 	CI700 = 8,
+	CMNS3 = 16,
 	/* ...and then we can use bitmap tricks for commonality */
 	CMN_ANY = -1,
 	NOT_CMN600 = -2,
-	CMN_650ON = CMN650 | CMN700,
+	CMN_650ON = CMN650 | CMN700 | CMNS3,
 };
 
 /* Actual part numbers and revision IDs defined by the hardware */
@@ -210,6 +212,7 @@ enum cmn_part {
 	PART_CMN650 = 0x436,
 	PART_CMN700 = 0x43c,
 	PART_CI700 = 0x43a,
+	PART_CMN_S3 = 0x43e,
 };
 
 /* CMN-600 r0px shouldn't exist in silicon, thankfully */
@@ -261,6 +264,7 @@ enum cmn_node_type {
 	CMN_TYPE_HNS = 0x200,
 	CMN_TYPE_HNS_MPAM_S,
 	CMN_TYPE_HNS_MPAM_NS,
+	CMN_TYPE_APB = 0x1000,
 	/* Not a real node type */
 	CMN_TYPE_WP = 0x7770
 };
@@ -280,8 +284,11 @@ struct arm_cmn_node {
 	u16 id, logid;
 	enum cmn_node_type type;
 
+	/* XP properties really, but replicated to children for convenience */
 	u8 dtm;
 	s8 dtc;
+	u8 portid_bits:4;
+	u8 deviceid_bits:4;
 	/* DN/HN-F/CXHA */
 	struct {
 		u8 val : 4;
@@ -307,8 +314,9 @@ struct arm_cmn_dtm {
 
 struct arm_cmn_dtc {
 	void __iomem *base;
+	void __iomem *pmu_base;
 	int irq;
-	int irq_friend;
+	s8 irq_friend;
 	bool cc_active;
 
 	struct perf_event *counters[CMN_DT_NUM_COUNTERS];
@@ -357,49 +365,33 @@ struct arm_cmn {
 static int arm_cmn_hp_state;
 
 struct arm_cmn_nodeid {
-	u8 x;
-	u8 y;
 	u8 port;
 	u8 dev;
 };
 
 static int arm_cmn_xyidbits(const struct arm_cmn *cmn)
 {
-	return fls((cmn->mesh_x - 1) | (cmn->mesh_y - 1) | 2);
+	return fls((cmn->mesh_x - 1) | (cmn->mesh_y - 1));
 }
 
-static struct arm_cmn_nodeid arm_cmn_nid(const struct arm_cmn *cmn, u16 id)
+static struct arm_cmn_nodeid arm_cmn_nid(const struct arm_cmn_node *dn)
 {
 	struct arm_cmn_nodeid nid;
 
-	if (cmn->num_xps == 1) {
-		nid.x = 0;
-		nid.y = 0;
-		nid.port = CMN_NODEID_1x1_PID(id);
-		nid.dev = CMN_NODEID_DEVID(id);
-	} else {
-		int bits = arm_cmn_xyidbits(cmn);
-
-		nid.x = CMN_NODEID_X(id, bits);
-		nid.y = CMN_NODEID_Y(id, bits);
-		if (cmn->ports_used & 0xc) {
-			nid.port = CMN_NODEID_EXT_PID(id);
-			nid.dev = CMN_NODEID_EXT_DEVID(id);
-		} else {
-			nid.port = CMN_NODEID_PID(id);
-			nid.dev = CMN_NODEID_DEVID(id);
-		}
-	}
+	nid.dev = dn->id & ((1U << dn->deviceid_bits) - 1);
+	nid.port = (dn->id >> dn->deviceid_bits) & ((1U << dn->portid_bits) - 1);
 	return nid;
 }
 
 static struct arm_cmn_node *arm_cmn_node_to_xp(const struct arm_cmn *cmn,
 					       const struct arm_cmn_node *dn)
 {
-	struct arm_cmn_nodeid nid = arm_cmn_nid(cmn, dn->id);
-	int xp_idx = cmn->mesh_x * nid.y + nid.x;
+	int id = dn->id >> (dn->portid_bits + dn->deviceid_bits);
+	int bits = arm_cmn_xyidbits(cmn);
+	int x = id >> bits;
+	int y = id & ((1U << bits) - 1);
 
-	return cmn->xps + xp_idx;
+	return cmn->xps + cmn->mesh_x * y + x;
 }
 static struct arm_cmn_node *arm_cmn_node(const struct arm_cmn *cmn,
 					 enum cmn_node_type type)
@@ -423,15 +415,27 @@ static enum cmn_model arm_cmn_model(const struct arm_cmn *cmn)
 		return CMN700;
 	case PART_CI700:
 		return CI700;
+	case PART_CMN_S3:
+		return CMNS3;
 	default:
 		return 0;
 	};
 }
 
+static int arm_cmn_pmu_offset(const struct arm_cmn *cmn, const struct arm_cmn_node *dn)
+{
+	if (cmn->part == PART_CMN_S3) {
+		if (dn->type == CMN_TYPE_XP)
+			return CMN_S3_DTM_OFFSET;
+		return CMN_S3_PMU_OFFSET;
+	}
+	return CMN_PMU_OFFSET;
+}
+
 static u32 arm_cmn_device_connect_info(const struct arm_cmn *cmn,
 				       const struct arm_cmn_node *xp, int port)
 {
-	int offset = CMN_MXP__CONNECT_INFO(port);
+	int offset = CMN_MXP__CONNECT_INFO(port) - arm_cmn_pmu_offset(cmn, xp);
 
 	if (port >= 2) {
 		if (cmn->part == PART_CMN600 || cmn->part == PART_CMN650)
@@ -444,7 +448,7 @@ static u32 arm_cmn_device_connect_info(const struct arm_cmn *cmn,
 			offset += CI700_CONNECT_INFO_P2_5_OFFSET;
 	}
 
-	return readl_relaxed(xp->pmu_base - CMN_PMU_OFFSET + offset);
+	return readl_relaxed(xp->pmu_base + offset);
 }
 
 static struct dentry *arm_cmn_debugfs;
@@ -478,20 +482,25 @@ static const char *arm_cmn_device_type(u8 type)
 		case 0x17: return "RN-F_C_E|";
 		case 0x18: return " RN-F_E |";
 		case 0x19: return "RN-F_E_E|";
+		case 0x1a: return "  HN-S  |";
+		case 0x1b: return "  LCN   |";
 		case 0x1c: return "  MTSX  |";
 		case 0x1d: return "  HN-V  |";
 		case 0x1e: return "  CCG   |";
+		case 0x20: return " RN-F_F |";
+		case 0x21: return "RN-F_F_E|";
+		case 0x22: return " SN-F_F |";
 		default:   return "  ????  |";
 	}
 }
 
-static void arm_cmn_show_logid(struct seq_file *s, int x, int y, int p, int d)
+static void arm_cmn_show_logid(struct seq_file *s, const struct arm_cmn_node *xp, int p, int d)
 {
 	struct arm_cmn *cmn = s->private;
 	struct arm_cmn_node *dn;
+	u16 id = xp->id | d | (p << xp->deviceid_bits);
 
 	for (dn = cmn->dns; dn->type; dn++) {
-		struct arm_cmn_nodeid nid = arm_cmn_nid(cmn, dn->id);
 		int pad = dn->logid < 10;
 
 		if (dn->type == CMN_TYPE_XP)
@@ -500,7 +509,7 @@ static void arm_cmn_show_logid(struct seq_file *s, int x, int y, int p, int d)
 		if (dn->type < CMN_TYPE_HNI)
 			continue;
 
-		if (nid.x != x || nid.y != y || nid.port != p || nid.dev != d)
+		if (dn->id != id)
 			continue;
 
 		seq_printf(s, " %*c#%-*d  |", pad + 1, ' ', 3 - pad, dn->logid);
@@ -521,6 +530,7 @@ static int arm_cmn_map_show(struct seq_file *s, void *data)
 	y = cmn->mesh_y;
 	while (y--) {
 		int xp_base = cmn->mesh_x * y;
+		struct arm_cmn_node *xp = cmn->xps + xp_base;
 		u8 port[CMN_MAX_PORTS][CMN_MAX_DIMENSION];
 
 		for (x = 0; x < cmn->mesh_x; x++)
@@ -528,16 +538,14 @@ static int arm_cmn_map_show(struct seq_file *s, void *data)
 
 		seq_printf(s, "\n%-2d   |", y);
 		for (x = 0; x < cmn->mesh_x; x++) {
-			struct arm_cmn_node *xp = cmn->xps + xp_base + x;
-
 			for (p = 0; p < CMN_MAX_PORTS; p++)
-				port[p][x] = arm_cmn_device_connect_info(cmn, xp, p);
+				port[p][x] = arm_cmn_device_connect_info(cmn, xp + x, p);
 			seq_printf(s, " XP #%-3d|", xp_base + x);
 		}
 
 		seq_puts(s, "\n     |");
 		for (x = 0; x < cmn->mesh_x; x++) {
-			s8 dtc = cmn->xps[xp_base + x].dtc;
+			s8 dtc = xp[x].dtc;
 
 			if (dtc < 0)
 				seq_puts(s, " DTC ?? |");
@@ -554,10 +562,10 @@ static int arm_cmn_map_show(struct seq_file *s, void *data)
 				seq_puts(s, arm_cmn_device_type(port[p][x]));
 			seq_puts(s, "\n    0|");
 			for (x = 0; x < cmn->mesh_x; x++)
-				arm_cmn_show_logid(s, x, y, p, 0);
+				arm_cmn_show_logid(s, xp + x, p, 0);
 			seq_puts(s, "\n    1|");
 			for (x = 0; x < cmn->mesh_x; x++)
-				arm_cmn_show_logid(s, x, y, p, 1);
+				arm_cmn_show_logid(s, xp + x, p, 1);
 		}
 		seq_puts(s, "\n-----+");
 	}
@@ -585,7 +593,7 @@ static void arm_cmn_debugfs_init(struct arm_cmn *cmn, int id) {}
 
 struct arm_cmn_hw_event {
 	struct arm_cmn_node *dn;
-	u64 dtm_idx[4];
+	u64 dtm_idx[DIV_ROUND_UP(CMN_MAX_NODES_PER_EVENT * 2, 64)];
 	s8 dtc_idx[CMN_MAX_DTCS];
 	u8 num_dns;
 	u8 dtm_offset;
@@ -599,6 +607,7 @@ struct arm_cmn_hw_event {
 	bool wide_sel;
 	enum cmn_filter_select filter_sel;
 };
+static_assert(sizeof(struct arm_cmn_hw_event) <= offsetof(struct hw_perf_event, target));
 
 #define for_each_hw_dn(hw, dn, i) \
 	for (i = 0, dn = hw->dn; i < hw->num_dns; i++, dn++)
@@ -609,7 +618,6 @@ struct arm_cmn_hw_event {
 
 static struct arm_cmn_hw_event *to_cmn_hw(struct perf_event *event)
 {
-	BUILD_BUG_ON(sizeof(struct arm_cmn_hw_event) > offsetof(struct hw_perf_event, target));
 	return (struct arm_cmn_hw_event *)&event->hw;
 }
 
@@ -790,8 +798,8 @@ static umode_t arm_cmn_event_attr_is_visible(struct kobject *kobj,
 	CMN_EVENT_ATTR(CMN_ANY, cxha_##_name, CMN_TYPE_CXHA, _event)
 #define CMN_EVENT_CCRA(_name, _event)				\
 	CMN_EVENT_ATTR(CMN_ANY, ccra_##_name, CMN_TYPE_CCRA, _event)
-#define CMN_EVENT_CCHA(_name, _event)				\
-	CMN_EVENT_ATTR(CMN_ANY, ccha_##_name, CMN_TYPE_CCHA, _event)
+#define CMN_EVENT_CCHA(_model, _name, _event)				\
+	CMN_EVENT_ATTR(_model, ccha_##_name, CMN_TYPE_CCHA, _event)
 #define CMN_EVENT_CCLA(_name, _event)				\
 	CMN_EVENT_ATTR(CMN_ANY, ccla_##_name, CMN_TYPE_CCLA, _event)
 #define CMN_EVENT_CCLA_RNI(_name, _event)				\
@@ -1149,42 +1157,43 @@ static struct attribute *arm_cmn_event_attrs[] = {
 	CMN_EVENT_CCRA(wdb_alloc,			0x59),
 	CMN_EVENT_CCRA(ssb_alloc,			0x5a),
 
-	CMN_EVENT_CCHA(rddatbyp,			0x61),
-	CMN_EVENT_CCHA(chirsp_up_stall,			0x62),
-	CMN_EVENT_CCHA(chidat_up_stall,			0x63),
-	CMN_EVENT_CCHA(snppcrd_link0_stall,		0x64),
-	CMN_EVENT_CCHA(snppcrd_link1_stall,		0x65),
-	CMN_EVENT_CCHA(snppcrd_link2_stall,		0x66),
-	CMN_EVENT_CCHA(reqtrk_occ,			0x67),
-	CMN_EVENT_CCHA(rdb_occ,				0x68),
-	CMN_EVENT_CCHA(rdbyp_occ,			0x69),
-	CMN_EVENT_CCHA(wdb_occ,				0x6a),
-	CMN_EVENT_CCHA(snptrk_occ,			0x6b),
-	CMN_EVENT_CCHA(sdb_occ,				0x6c),
-	CMN_EVENT_CCHA(snphaz_occ,			0x6d),
-	CMN_EVENT_CCHA(reqtrk_alloc,			0x6e),
-	CMN_EVENT_CCHA(rdb_alloc,			0x6f),
-	CMN_EVENT_CCHA(rdbyp_alloc,			0x70),
-	CMN_EVENT_CCHA(wdb_alloc,			0x71),
-	CMN_EVENT_CCHA(snptrk_alloc,			0x72),
-	CMN_EVENT_CCHA(sdb_alloc,			0x73),
-	CMN_EVENT_CCHA(snphaz_alloc,			0x74),
-	CMN_EVENT_CCHA(pb_rhu_req_occ,			0x75),
-	CMN_EVENT_CCHA(pb_rhu_req_alloc,		0x76),
-	CMN_EVENT_CCHA(pb_rhu_pcie_req_occ,		0x77),
-	CMN_EVENT_CCHA(pb_rhu_pcie_req_alloc,		0x78),
-	CMN_EVENT_CCHA(pb_pcie_wr_req_occ,		0x79),
-	CMN_EVENT_CCHA(pb_pcie_wr_req_alloc,		0x7a),
-	CMN_EVENT_CCHA(pb_pcie_reg_req_occ,		0x7b),
-	CMN_EVENT_CCHA(pb_pcie_reg_req_alloc,		0x7c),
-	CMN_EVENT_CCHA(pb_pcie_rsvd_req_occ,		0x7d),
-	CMN_EVENT_CCHA(pb_pcie_rsvd_req_alloc,		0x7e),
-	CMN_EVENT_CCHA(pb_rhu_dat_occ,			0x7f),
-	CMN_EVENT_CCHA(pb_rhu_dat_alloc,		0x80),
-	CMN_EVENT_CCHA(pb_rhu_pcie_dat_occ,		0x81),
-	CMN_EVENT_CCHA(pb_rhu_pcie_dat_alloc,		0x82),
-	CMN_EVENT_CCHA(pb_pcie_wr_dat_occ,		0x83),
-	CMN_EVENT_CCHA(pb_pcie_wr_dat_alloc,		0x84),
+	CMN_EVENT_CCHA(CMN_ANY, rddatbyp,		0x61),
+	CMN_EVENT_CCHA(CMN_ANY, chirsp_up_stall,	0x62),
+	CMN_EVENT_CCHA(CMN_ANY, chidat_up_stall,	0x63),
+	CMN_EVENT_CCHA(CMN_ANY, snppcrd_link0_stall,	0x64),
+	CMN_EVENT_CCHA(CMN_ANY, snppcrd_link1_stall,	0x65),
+	CMN_EVENT_CCHA(CMN_ANY, snppcrd_link2_stall,	0x66),
+	CMN_EVENT_CCHA(CMN_ANY, reqtrk_occ,		0x67),
+	CMN_EVENT_CCHA(CMN_ANY, rdb_occ,		0x68),
+	CMN_EVENT_CCHA(CMN_ANY, rdbyp_occ,		0x69),
+	CMN_EVENT_CCHA(CMN_ANY, wdb_occ,		0x6a),
+	CMN_EVENT_CCHA(CMN_ANY, snptrk_occ,		0x6b),
+	CMN_EVENT_CCHA(CMN_ANY, sdb_occ,		0x6c),
+	CMN_EVENT_CCHA(CMN_ANY, snphaz_occ,		0x6d),
+	CMN_EVENT_CCHA(CMN_ANY, reqtrk_alloc,		0x6e),
+	CMN_EVENT_CCHA(CMN_ANY, rdb_alloc,		0x6f),
+	CMN_EVENT_CCHA(CMN_ANY, rdbyp_alloc,		0x70),
+	CMN_EVENT_CCHA(CMN_ANY, wdb_alloc,		0x71),
+	CMN_EVENT_CCHA(CMN_ANY, snptrk_alloc,		0x72),
+	CMN_EVENT_CCHA(CMN_ANY, db_alloc,		0x73),
+	CMN_EVENT_CCHA(CMN_ANY, snphaz_alloc,		0x74),
+	CMN_EVENT_CCHA(CMN_ANY, pb_rhu_req_occ,		0x75),
+	CMN_EVENT_CCHA(CMN_ANY, pb_rhu_req_alloc,	0x76),
+	CMN_EVENT_CCHA(CMN_ANY, pb_rhu_pcie_req_occ,	0x77),
+	CMN_EVENT_CCHA(CMN_ANY, pb_rhu_pcie_req_alloc,	0x78),
+	CMN_EVENT_CCHA(CMN_ANY, pb_pcie_wr_req_occ,	0x79),
+	CMN_EVENT_CCHA(CMN_ANY, pb_pcie_wr_req_alloc,	0x7a),
+	CMN_EVENT_CCHA(CMN_ANY, pb_pcie_reg_req_occ,	0x7b),
+	CMN_EVENT_CCHA(CMN_ANY, pb_pcie_reg_req_alloc,	0x7c),
+	CMN_EVENT_CCHA(CMN_ANY, pb_pcie_rsvd_req_occ,	0x7d),
+	CMN_EVENT_CCHA(CMN_ANY, pb_pcie_rsvd_req_alloc,	0x7e),
+	CMN_EVENT_CCHA(CMN_ANY, pb_rhu_dat_occ,		0x7f),
+	CMN_EVENT_CCHA(CMN_ANY, pb_rhu_dat_alloc,	0x80),
+	CMN_EVENT_CCHA(CMN_ANY, pb_rhu_pcie_dat_occ,	0x81),
+	CMN_EVENT_CCHA(CMN_ANY, pb_rhu_pcie_dat_alloc,	0x82),
+	CMN_EVENT_CCHA(CMN_ANY, pb_pcie_wr_dat_occ,	0x83),
+	CMN_EVENT_CCHA(CMN_ANY, pb_pcie_wr_dat_alloc,	0x84),
+	CMN_EVENT_CCHA(CMNS3, chirsp1_up_stall,		0x85),
 
 	CMN_EVENT_CCLA(rx_cxs,				0x21),
 	CMN_EVENT_CCLA(tx_cxs,				0x22),
@@ -1271,15 +1280,11 @@ static ssize_t arm_cmn_format_show(struct device *dev,
 				   struct device_attribute *attr, char *buf)
 {
 	struct arm_cmn_format_attr *fmt = container_of(attr, typeof(*fmt), attr);
-	int lo = __ffs(fmt->field), hi = __fls(fmt->field);
-
-	if (lo == hi)
-		return sysfs_emit(buf, "config:%d\n", lo);
 
 	if (!fmt->config)
-		return sysfs_emit(buf, "config:%d-%d\n", lo, hi);
+		return sysfs_emit(buf, "config:%*pbl\n", 64, &fmt->field);
 
-	return sysfs_emit(buf, "config%d:%d-%d\n", fmt->config, lo, hi);
+	return sysfs_emit(buf, "config%d:%*pbl\n", fmt->config, 64, &fmt->field);
 }
 
 #define _CMN_FORMAT_ATTR(_name, _cfg, _fld)				\
@@ -1415,7 +1420,7 @@ static u32 arm_cmn_wp_config(struct perf_event *event, int wp_idx)
 static void arm_cmn_set_state(struct arm_cmn *cmn, u32 state)
 {
 	if (!cmn->state)
-		writel_relaxed(0, cmn->dtc[0].base + CMN_DT_PMCR);
+		writel_relaxed(0, CMN_DT_PMCR(&cmn->dtc[0]));
 	cmn->state |= state;
 }
 
@@ -1424,7 +1429,7 @@ static void arm_cmn_clear_state(struct arm_cmn *cmn, u32 state)
 	cmn->state &= ~state;
 	if (!cmn->state)
 		writel_relaxed(CMN_DT_PMCR_PMU_EN | CMN_DT_PMCR_OVFL_INTR_EN,
-			       cmn->dtc[0].base + CMN_DT_PMCR);
+			       CMN_DT_PMCR(&cmn->dtc[0]));
 }
 
 static void arm_cmn_pmu_enable(struct pmu *pmu)
@@ -1459,18 +1464,19 @@ static u64 arm_cmn_read_dtm(struct arm_cmn *cmn, struct arm_cmn_hw_event *hw,
 
 static u64 arm_cmn_read_cc(struct arm_cmn_dtc *dtc)
 {
-	u64 val = readq_relaxed(dtc->base + CMN_DT_PMCCNTR);
+	void __iomem *pmccntr = CMN_DT_PMCCNTR(dtc);
+	u64 val = readq_relaxed(pmccntr);
 
-	writeq_relaxed(CMN_CC_INIT, dtc->base + CMN_DT_PMCCNTR);
+	writeq_relaxed(CMN_CC_INIT, pmccntr);
 	return (val - CMN_CC_INIT) & ((CMN_CC_INIT << 1) - 1);
 }
 
 static u32 arm_cmn_read_counter(struct arm_cmn_dtc *dtc, int idx)
 {
-	u32 val, pmevcnt = CMN_DT_PMEVCNT(idx);
+	void __iomem *pmevcnt = CMN_DT_PMEVCNT(dtc, idx);
+	u32 val = readl_relaxed(pmevcnt);
 
-	val = readl_relaxed(dtc->base + pmevcnt);
-	writel_relaxed(CMN_COUNTER_INIT, dtc->base + pmevcnt);
+	writel_relaxed(CMN_COUNTER_INIT, pmevcnt);
 	return val - CMN_COUNTER_INIT;
 }
 
@@ -1481,7 +1487,7 @@ static void arm_cmn_init_counter(struct perf_event *event)
 	u64 count;
 
 	for_each_hw_dtc_idx(hw, i, idx) {
-		writel_relaxed(CMN_COUNTER_INIT, cmn->dtc[i].base + CMN_DT_PMEVCNT(idx));
+		writel_relaxed(CMN_COUNTER_INIT, CMN_DT_PMEVCNT(&cmn->dtc[i], idx));
 		cmn->dtc[i].counters[idx] = event;
 	}
 
@@ -1564,9 +1570,12 @@ static void arm_cmn_event_start(struct perf_event *event, int flags)
 	int i;
 
 	if (type == CMN_TYPE_DTC) {
-		i = hw->dtc_idx[0];
-		writeq_relaxed(CMN_CC_INIT, cmn->dtc[i].base + CMN_DT_PMCCNTR);
-		cmn->dtc[i].cc_active = true;
+		struct arm_cmn_dtc *dtc = cmn->dtc + hw->dtc_idx[0];
+
+		writel_relaxed(CMN_DT_DTC_CTL_DT_EN | CMN_DT_DTC_CTL_CG_DISABLE,
+			       dtc->base + CMN_DT_DTC_CTL);
+		writeq_relaxed(CMN_CC_INIT, CMN_DT_PMCCNTR(dtc));
+		dtc->cc_active = true;
 	} else if (type == CMN_TYPE_WP) {
 		u64 val = CMN_EVENT_WP_VAL(event);
 		u64 mask = CMN_EVENT_WP_MASK(event);
@@ -1595,8 +1604,10 @@ static void arm_cmn_event_stop(struct perf_event *event, int flags)
 	int i;
 
 	if (type == CMN_TYPE_DTC) {
-		i = hw->dtc_idx[0];
-		cmn->dtc[i].cc_active = false;
+		struct arm_cmn_dtc *dtc = cmn->dtc + hw->dtc_idx[0];
+
+		dtc->cc_active = false;
+		writel_relaxed(CMN_DT_DTC_CTL_DT_EN, dtc->base + CMN_DT_DTC_CTL);
 	} else if (type == CMN_TYPE_WP) {
 		for_each_hw_dn(hw, dn, i) {
 			void __iomem *base = dn->pmu_base + CMN_DTM_OFFSET(hw->dtm_offset);
@@ -1784,7 +1795,8 @@ static int arm_cmn_event_init(struct perf_event *event)
 		/* ...but the DTM may depend on which port we're watching */
 		if (cmn->multi_dtm)
 			hw->dtm_offset = CMN_EVENT_WP_DEV_SEL(event) / 2;
-	} else if (type == CMN_TYPE_XP && cmn->part == PART_CMN700) {
+	} else if (type == CMN_TYPE_XP &&
+		   (cmn->part == PART_CMN700 || cmn->part == PART_CMN_S3)) {
 		hw->wide_sel = true;
 	}
 
@@ -1815,10 +1827,7 @@ static int arm_cmn_event_init(struct perf_event *event)
 	}
 
 	if (!hw->num_dns) {
-		struct arm_cmn_nodeid nid = arm_cmn_nid(cmn, nodeid);
-
-		dev_dbg(cmn->dev, "invalid node 0x%x (%d,%d,%d,%d) type 0x%x\n",
-			nodeid, nid.x, nid.y, nid.port, nid.dev, type);
+		dev_dbg(cmn->dev, "invalid node 0x%x type 0x%x\n", nodeid, type);
 		return -EINVAL;
 	}
 
@@ -1921,7 +1930,7 @@ static int arm_cmn_event_add(struct perf_event *event, int flags)
 			arm_cmn_claim_wp_idx(dtm, event, d, wp_idx, i);
 			writel_relaxed(cfg, dtm->base + CMN_DTM_WPn_CONFIG(wp_idx));
 		} else {
-			struct arm_cmn_nodeid nid = arm_cmn_nid(cmn, dn->id);
+			struct arm_cmn_nodeid nid = arm_cmn_nid(dn);
 
 			if (cmn->multi_dtm)
 				nid.port %= 2;
@@ -2010,7 +2019,7 @@ static int arm_cmn_pmu_online_cpu(unsigned int cpu, struct hlist_node *cpuhp_nod
 
 	cmn = hlist_entry_safe(cpuhp_node, struct arm_cmn, cpuhp_node);
 	node = dev_to_node(cmn->dev);
-	if (node != NUMA_NO_NODE && cpu_to_node(cmn->cpu) != node && cpu_to_node(cpu) == node)
+	if (cpu_to_node(cmn->cpu) != node && cpu_to_node(cpu) == node)
 		arm_cmn_migrate(cmn, cpu);
 	return 0;
 }
@@ -2043,7 +2052,7 @@ static irqreturn_t arm_cmn_handle_irq(int irq, void *dev_id)
 	irqreturn_t ret = IRQ_NONE;
 
 	for (;;) {
-		u32 status = readl_relaxed(dtc->base + CMN_DT_PMOVSR);
+		u32 status = readl_relaxed(CMN_DT_PMOVSR(dtc));
 		u64 delta;
 		int i;
 
@@ -2065,7 +2074,7 @@ static irqreturn_t arm_cmn_handle_irq(int irq, void *dev_id)
 			}
 		}
 
-		writel_relaxed(status, dtc->base + CMN_DT_PMOVSR_CLR);
+		writel_relaxed(status, CMN_DT_PMOVSR_CLR(dtc));
 
 		if (!dtc->irq_friend)
 			return ret;
@@ -2119,15 +2128,16 @@ static int arm_cmn_init_dtc(struct arm_cmn *cmn, struct arm_cmn_node *dn, int id
 {
 	struct arm_cmn_dtc *dtc = cmn->dtc + idx;
 
-	dtc->base = dn->pmu_base - CMN_PMU_OFFSET;
+	dtc->pmu_base = dn->pmu_base;
+	dtc->base = dtc->pmu_base - arm_cmn_pmu_offset(cmn, dn);
 	dtc->irq = platform_get_irq(to_platform_device(cmn->dev), idx);
 	if (dtc->irq < 0)
 		return dtc->irq;
 
 	writel_relaxed(CMN_DT_DTC_CTL_DT_EN, dtc->base + CMN_DT_DTC_CTL);
-	writel_relaxed(CMN_DT_PMCR_PMU_EN | CMN_DT_PMCR_OVFL_INTR_EN, dtc->base + CMN_DT_PMCR);
-	writeq_relaxed(0, dtc->base + CMN_DT_PMCCNTR);
-	writel_relaxed(0x1ff, dtc->base + CMN_DT_PMOVSR_CLR);
+	writel_relaxed(CMN_DT_PMCR_PMU_EN | CMN_DT_PMCR_OVFL_INTR_EN, CMN_DT_PMCR(dtc));
+	writeq_relaxed(0, CMN_DT_PMCCNTR(dtc));
+	writel_relaxed(0x1ff, CMN_DT_PMOVSR_CLR(dtc));
 
 	return 0;
 }
@@ -2168,10 +2178,12 @@ static int arm_cmn_init_dtcs(struct arm_cmn *cmn)
 			continue;
 
 		xp = arm_cmn_node_to_xp(cmn, dn);
+		dn->portid_bits = xp->portid_bits;
+		dn->deviceid_bits = xp->deviceid_bits;
 		dn->dtc = xp->dtc;
 		dn->dtm = xp->dtm;
 		if (cmn->multi_dtm)
-			dn->dtm += arm_cmn_nid(cmn, dn->id).port / 2;
+			dn->dtm += arm_cmn_nid(dn).port / 2;
 
 		if (dn->type == CMN_TYPE_DTC) {
 			int err = arm_cmn_init_dtc(cmn, dn, dtc_idx++);
@@ -2213,7 +2225,7 @@ static void arm_cmn_init_node_info(struct arm_cmn *cmn, u32 offset, struct arm_c
 	node->id = FIELD_GET(CMN_NI_NODE_ID, reg);
 	node->logid = FIELD_GET(CMN_NI_LOGICAL_ID, reg);
 
-	node->pmu_base = cmn->base + offset + CMN_PMU_OFFSET;
+	node->pmu_base = cmn->base + offset + arm_cmn_pmu_offset(cmn, node);
 
 	if (node->type == CMN_TYPE_CFG)
 		level = 0;
@@ -2271,7 +2283,17 @@ static int arm_cmn_discover(struct arm_cmn *cmn, unsigned int rgn_offset)
 	reg = readl_relaxed(cfg_region + CMN_CFGM_PERIPH_ID_23);
 	cmn->rev = FIELD_GET(CMN_CFGM_PID2_REVISION, reg);
 
+	/*
+	 * With the device isolation feature, if firmware has neglected to enable
+	 * an XP port then we risk locking up if we try to access anything behind
+	 * it; however we also have no way to tell from Non-Secure whether any
+	 * given port is disabled or not, so the only way to win is not to play...
+	 */
 	reg = readq_relaxed(cfg_region + CMN_CFGM_INFO_GLOBAL);
+	if (reg & CMN_INFO_DEVICE_ISO_ENABLE) {
+		dev_err(cmn->dev, "Device isolation enabled, not continuing due to risk of lockup\n");
+		return -ENODEV;
+	}
 	cmn->multi_dtm = reg & CMN_INFO_MULTIPLE_DTM_EN;
 	cmn->rsp_vc_num = FIELD_GET(CMN_INFO_RSP_VC_NUM, reg);
 	cmn->dat_vc_num = FIELD_GET(CMN_INFO_DAT_VC_NUM, reg);
@@ -2341,18 +2363,27 @@ static int arm_cmn_discover(struct arm_cmn *cmn, unsigned int rgn_offset)
 		arm_cmn_init_dtm(dtm++, xp, 0);
 		/*
 		 * Keeping track of connected ports will let us filter out
-		 * unnecessary XP events easily. We can also reliably infer the
-		 * "extra device ports" configuration for the node ID format
-		 * from this, since in that case we will see at least one XP
-		 * with port 2 connected, for the HN-D.
+		 * unnecessary XP events easily, and also infer the per-XP
+		 * part of the node ID format.
 		 */
 		for (int p = 0; p < CMN_MAX_PORTS; p++)
 			if (arm_cmn_device_connect_info(cmn, xp, p))
 				xp_ports |= BIT(p);
 
-		if (cmn->multi_dtm && (xp_ports & 0xc))
+		if (cmn->num_xps == 1) {
+			xp->portid_bits = 3;
+			xp->deviceid_bits = 2;
+		} else if (xp_ports > 0x3) {
+			xp->portid_bits = 2;
+			xp->deviceid_bits = 1;
+		} else {
+			xp->portid_bits = 1;
+			xp->deviceid_bits = 2;
+		}
+
+		if (cmn->multi_dtm && (xp_ports > 0x3))
 			arm_cmn_init_dtm(dtm++, xp, 1);
-		if (cmn->multi_dtm && (xp_ports & 0x30))
+		if (cmn->multi_dtm && (xp_ports > 0xf))
 			arm_cmn_init_dtm(dtm++, xp, 2);
 
 		cmn->ports_used |= xp_ports;
@@ -2407,10 +2438,13 @@ static int arm_cmn_discover(struct arm_cmn *cmn, unsigned int rgn_offset)
 			case CMN_TYPE_CXHA:
 			case CMN_TYPE_CCRA:
 			case CMN_TYPE_CCHA:
-			case CMN_TYPE_CCLA:
 			case CMN_TYPE_HNS:
 				dn++;
 				break;
+			case CMN_TYPE_CCLA:
+				dn->pmu_base += CMN_CCLA_PMU_EVENT_SEL;
+				dn++;
+				break;
 			/* Nothing to see here */
 			case CMN_TYPE_MPAM_S:
 			case CMN_TYPE_MPAM_NS:
@@ -2418,6 +2452,7 @@ static int arm_cmn_discover(struct arm_cmn *cmn, unsigned int rgn_offset)
 			case CMN_TYPE_CXLA:
 			case CMN_TYPE_HNS_MPAM_S:
 			case CMN_TYPE_HNS_MPAM_NS:
+			case CMN_TYPE_APB:
 				break;
 			/*
 			 * Split "optimised" combination nodes into separate
@@ -2428,7 +2463,7 @@ static int arm_cmn_discover(struct arm_cmn *cmn, unsigned int rgn_offset)
 			case CMN_TYPE_HNP:
 			case CMN_TYPE_CCLA_RNI:
 				dn[1] = dn[0];
-				dn[0].pmu_base += CMN_HNP_PMU_EVENT_SEL;
+				dn[0].pmu_base += CMN_CCLA_PMU_EVENT_SEL;
 				dn[1].type = arm_cmn_subtype(dn->type);
 				dn += 2;
 				break;
@@ -2603,6 +2638,7 @@ static const struct of_device_id arm_cmn_of_match[] = {
 	{ .compatible = "arm,cmn-600", .data = (void *)PART_CMN600 },
 	{ .compatible = "arm,cmn-650" },
 	{ .compatible = "arm,cmn-700" },
+	{ .compatible = "arm,cmn-s3" },
 	{ .compatible = "arm,ci-700" },
 	{}
 };
diff --git a/drivers/perf/arm-ni.c b/drivers/perf/arm-ni.c
new file mode 100644
index 000000000000..90fcfe693439
--- /dev/null
+++ b/drivers/perf/arm-ni.c
@@ -0,0 +1,781 @@
+// SPDX-License-Identifier: GPL-2.0
+// Copyright (C) 2022-2024 Arm Limited
+// NI-700 Network-on-Chip PMU driver
+
+#include <linux/acpi.h>
+#include <linux/bitfield.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/io-64-nonatomic-lo-hi.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/perf_event.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+/* Common registers */
+#define NI_NODE_TYPE		0x000
+#define NI_NODE_TYPE_NODE_ID	GENMASK(31, 16)
+#define NI_NODE_TYPE_NODE_TYPE	GENMASK(15, 0)
+
+#define NI_CHILD_NODE_INFO	0x004
+#define NI_CHILD_PTR(n)		(0x008 + (n) * 4)
+
+#define NI700_PMUSELA		0x00c
+
+/* Config node */
+#define NI_PERIPHERAL_ID0	0xfe0
+#define NI_PIDR0_PART_7_0	GENMASK(7, 0)
+#define NI_PERIPHERAL_ID1	0xfe4
+#define NI_PIDR1_PART_11_8	GENMASK(3, 0)
+#define NI_PERIPHERAL_ID2	0xfe8
+#define NI_PIDR2_VERSION	GENMASK(7, 4)
+
+/* PMU node */
+#define NI_PMEVCNTR(n)		(0x008 + (n) * 8)
+#define NI_PMCCNTR_L		0x0f8
+#define NI_PMCCNTR_U		0x0fc
+#define NI_PMEVTYPER(n)		(0x400 + (n) * 4)
+#define NI_PMEVTYPER_NODE_TYPE	GENMASK(12, 9)
+#define NI_PMEVTYPER_NODE_ID	GENMASK(8, 0)
+#define NI_PMCNTENSET		0xc00
+#define NI_PMCNTENCLR		0xc20
+#define NI_PMINTENSET		0xc40
+#define NI_PMINTENCLR		0xc60
+#define NI_PMOVSCLR		0xc80
+#define NI_PMOVSSET		0xcc0
+#define NI_PMCFGR		0xe00
+#define NI_PMCR			0xe04
+#define NI_PMCR_RESET_CCNT	BIT(2)
+#define NI_PMCR_RESET_EVCNT	BIT(1)
+#define NI_PMCR_ENABLE		BIT(0)
+
+#define NI_NUM_COUNTERS		8
+#define NI_CCNT_IDX		31
+
+/* Event attributes */
+#define NI_CONFIG_TYPE		GENMASK_ULL(15, 0)
+#define NI_CONFIG_NODEID	GENMASK_ULL(31, 16)
+#define NI_CONFIG_EVENTID	GENMASK_ULL(47, 32)
+
+#define NI_EVENT_TYPE(event)	FIELD_GET(NI_CONFIG_TYPE, (event)->attr.config)
+#define NI_EVENT_NODEID(event)	FIELD_GET(NI_CONFIG_NODEID, (event)->attr.config)
+#define NI_EVENT_EVENTID(event)	FIELD_GET(NI_CONFIG_EVENTID, (event)->attr.config)
+
+enum ni_part {
+	PART_NI_700 = 0x43b,
+	PART_NI_710AE = 0x43d,
+};
+
+enum ni_node_type {
+	NI_GLOBAL,
+	NI_VOLTAGE,
+	NI_POWER,
+	NI_CLOCK,
+	NI_ASNI,
+	NI_AMNI,
+	NI_PMU,
+	NI_HSNI,
+	NI_HMNI,
+	NI_PMNI,
+};
+
+struct arm_ni_node {
+	void __iomem *base;
+	enum ni_node_type type;
+	u16 id;
+	u32 num_components;
+};
+
+struct arm_ni_unit {
+	void __iomem *pmusela;
+	enum ni_node_type type;
+	u16 id;
+	bool ns;
+	union {
+		__le64 pmusel;
+		u8 event[8];
+	};
+};
+
+struct arm_ni_cd {
+	void __iomem *pmu_base;
+	u16 id;
+	int num_units;
+	int irq;
+	int cpu;
+	struct hlist_node cpuhp_node;
+	struct pmu pmu;
+	struct arm_ni_unit *units;
+	struct perf_event *evcnt[NI_NUM_COUNTERS];
+	struct perf_event *ccnt;
+};
+
+struct arm_ni {
+	struct device *dev;
+	void __iomem *base;
+	enum ni_part part;
+	int id;
+	int num_cds;
+	struct arm_ni_cd cds[] __counted_by(num_cds);
+};
+
+#define cd_to_ni(cd) container_of((cd), struct arm_ni, cds[(cd)->id])
+#define pmu_to_cd(p) container_of((p), struct arm_ni_cd, pmu)
+
+#define cd_for_each_unit(cd, u) \
+	for (struct arm_ni_unit *u = cd->units; u < cd->units + cd->num_units; u++)
+
+static int arm_ni_hp_state;
+
+struct arm_ni_event_attr {
+	struct device_attribute attr;
+	enum ni_node_type type;
+};
+
+#define NI_EVENT_ATTR(_name, _type)					\
+	(&((struct arm_ni_event_attr[]) {{				\
+		.attr = __ATTR(_name, 0444, arm_ni_event_show, NULL),	\
+		.type = _type,						\
+	}})[0].attr.attr)
+
+static ssize_t arm_ni_event_show(struct device *dev,
+				 struct device_attribute *attr, char *buf)
+{
+	struct arm_ni_event_attr *eattr = container_of(attr, typeof(*eattr), attr);
+
+	if (eattr->type == NI_PMU)
+		return sysfs_emit(buf, "type=0x%x\n", eattr->type);
+
+	return sysfs_emit(buf, "type=0x%x,eventid=?,nodeid=?\n", eattr->type);
+}
+
+static umode_t arm_ni_event_attr_is_visible(struct kobject *kobj,
+					    struct attribute *attr, int unused)
+{
+	struct device *dev = kobj_to_dev(kobj);
+	struct arm_ni_cd *cd = pmu_to_cd(dev_get_drvdata(dev));
+	struct arm_ni_event_attr *eattr;
+
+	eattr = container_of(attr, typeof(*eattr), attr.attr);
+
+	cd_for_each_unit(cd, unit) {
+		if (unit->type == eattr->type && unit->ns)
+			return attr->mode;
+	}
+
+	return 0;
+}
+
+static struct attribute *arm_ni_event_attrs[] = {
+	NI_EVENT_ATTR(asni, NI_ASNI),
+	NI_EVENT_ATTR(amni, NI_AMNI),
+	NI_EVENT_ATTR(cycles, NI_PMU),
+	NI_EVENT_ATTR(hsni, NI_HSNI),
+	NI_EVENT_ATTR(hmni, NI_HMNI),
+	NI_EVENT_ATTR(pmni, NI_PMNI),
+	NULL
+};
+
+static const struct attribute_group arm_ni_event_attrs_group = {
+	.name = "events",
+	.attrs = arm_ni_event_attrs,
+	.is_visible = arm_ni_event_attr_is_visible,
+};
+
+struct arm_ni_format_attr {
+	struct device_attribute attr;
+	u64 field;
+};
+
+#define NI_FORMAT_ATTR(_name, _fld)					\
+	(&((struct arm_ni_format_attr[]) {{				\
+		.attr = __ATTR(_name, 0444, arm_ni_format_show, NULL),	\
+		.field = _fld,						\
+	}})[0].attr.attr)
+
+static ssize_t arm_ni_format_show(struct device *dev,
+				  struct device_attribute *attr, char *buf)
+{
+	struct arm_ni_format_attr *fmt = container_of(attr, typeof(*fmt), attr);
+
+	return sysfs_emit(buf, "config:%*pbl\n", 64, &fmt->field);
+}
+
+static struct attribute *arm_ni_format_attrs[] = {
+	NI_FORMAT_ATTR(type, NI_CONFIG_TYPE),
+	NI_FORMAT_ATTR(nodeid, NI_CONFIG_NODEID),
+	NI_FORMAT_ATTR(eventid, NI_CONFIG_EVENTID),
+	NULL
+};
+
+static const struct attribute_group arm_ni_format_attrs_group = {
+	.name = "format",
+	.attrs = arm_ni_format_attrs,
+};
+
+static ssize_t arm_ni_cpumask_show(struct device *dev,
+				   struct device_attribute *attr, char *buf)
+{
+	struct arm_ni_cd *cd = pmu_to_cd(dev_get_drvdata(dev));
+
+	return cpumap_print_to_pagebuf(true, buf, cpumask_of(cd->cpu));
+}
+
+static struct device_attribute arm_ni_cpumask_attr =
+		__ATTR(cpumask, 0444, arm_ni_cpumask_show, NULL);
+
+static ssize_t arm_ni_identifier_show(struct device *dev,
+				      struct device_attribute *attr, char *buf)
+{
+	struct arm_ni *ni = cd_to_ni(pmu_to_cd(dev_get_drvdata(dev)));
+	u32 reg = readl_relaxed(ni->base + NI_PERIPHERAL_ID2);
+	int version = FIELD_GET(NI_PIDR2_VERSION, reg);
+
+	return sysfs_emit(buf, "%03x%02x\n", ni->part, version);
+}
+
+static struct device_attribute arm_ni_identifier_attr =
+		__ATTR(identifier, 0444, arm_ni_identifier_show, NULL);
+
+static struct attribute *arm_ni_other_attrs[] = {
+	&arm_ni_cpumask_attr.attr,
+	&arm_ni_identifier_attr.attr,
+	NULL
+};
+
+static const struct attribute_group arm_ni_other_attr_group = {
+	.attrs = arm_ni_other_attrs,
+	NULL
+};
+
+static const struct attribute_group *arm_ni_attr_groups[] = {
+	&arm_ni_event_attrs_group,
+	&arm_ni_format_attrs_group,
+	&arm_ni_other_attr_group,
+	NULL
+};
+
+static void arm_ni_pmu_enable(struct pmu *pmu)
+{
+	writel_relaxed(NI_PMCR_ENABLE, pmu_to_cd(pmu)->pmu_base + NI_PMCR);
+}
+
+static void arm_ni_pmu_disable(struct pmu *pmu)
+{
+	writel_relaxed(0, pmu_to_cd(pmu)->pmu_base + NI_PMCR);
+}
+
+struct arm_ni_val {
+	unsigned int evcnt;
+	unsigned int ccnt;
+};
+
+static bool arm_ni_val_count_event(struct perf_event *evt, struct arm_ni_val *val)
+{
+	if (is_software_event(evt))
+		return true;
+
+	if (NI_EVENT_TYPE(evt) == NI_PMU) {
+		val->ccnt++;
+		return val->ccnt <= 1;
+	}
+
+	val->evcnt++;
+	return val->evcnt <= NI_NUM_COUNTERS;
+}
+
+static int arm_ni_validate_group(struct perf_event *event)
+{
+	struct perf_event *sibling, *leader = event->group_leader;
+	struct arm_ni_val val = { 0 };
+
+	if (leader == event)
+		return 0;
+
+	arm_ni_val_count_event(event, &val);
+	if (!arm_ni_val_count_event(leader, &val))
+		return -EINVAL;
+
+	for_each_sibling_event(sibling, leader) {
+		if (!arm_ni_val_count_event(sibling, &val))
+			return -EINVAL;
+	}
+	return 0;
+}
+
+static int arm_ni_event_init(struct perf_event *event)
+{
+	struct arm_ni_cd *cd = pmu_to_cd(event->pmu);
+
+	if (event->attr.type != event->pmu->type)
+		return -ENOENT;
+
+	if (is_sampling_event(event))
+		return -EINVAL;
+
+	event->cpu = cd->cpu;
+	if (NI_EVENT_TYPE(event) == NI_PMU)
+		return arm_ni_validate_group(event);
+
+	cd_for_each_unit(cd, unit) {
+		if (unit->type == NI_EVENT_TYPE(event) &&
+		    unit->id == NI_EVENT_NODEID(event) && unit->ns) {
+			event->hw.config_base = (unsigned long)unit;
+			return arm_ni_validate_group(event);
+		}
+	}
+	return -EINVAL;
+}
+
+static u64 arm_ni_read_ccnt(struct arm_ni_cd *cd)
+{
+	u64 l, u_old, u_new;
+	int retries = 3; /* 1st time unlucky, 2nd improbable, 3rd just broken */
+
+	u_new = readl_relaxed(cd->pmu_base + NI_PMCCNTR_U);
+	do {
+		u_old = u_new;
+		l = readl_relaxed(cd->pmu_base + NI_PMCCNTR_L);
+		u_new = readl_relaxed(cd->pmu_base + NI_PMCCNTR_U);
+	} while (u_new != u_old && --retries);
+	WARN_ON(!retries);
+
+	return (u_new << 32) | l;
+}
+
+static void arm_ni_event_read(struct perf_event *event)
+{
+	struct arm_ni_cd *cd = pmu_to_cd(event->pmu);
+	struct hw_perf_event *hw = &event->hw;
+	u64 count, prev;
+	bool ccnt = hw->idx == NI_CCNT_IDX;
+
+	do {
+		prev = local64_read(&hw->prev_count);
+		if (ccnt)
+			count = arm_ni_read_ccnt(cd);
+		else
+			count = readl_relaxed(cd->pmu_base + NI_PMEVCNTR(hw->idx));
+	} while (local64_cmpxchg(&hw->prev_count, prev, count) != prev);
+
+	count -= prev;
+	if (!ccnt)
+		count = (u32)count;
+	local64_add(count, &event->count);
+}
+
+static void arm_ni_event_start(struct perf_event *event, int flags)
+{
+	struct arm_ni_cd *cd = pmu_to_cd(event->pmu);
+
+	writel_relaxed(1U << event->hw.idx, cd->pmu_base + NI_PMCNTENSET);
+}
+
+static void arm_ni_event_stop(struct perf_event *event, int flags)
+{
+	struct arm_ni_cd *cd = pmu_to_cd(event->pmu);
+
+	writel_relaxed(1U << event->hw.idx, cd->pmu_base + NI_PMCNTENCLR);
+	if (flags & PERF_EF_UPDATE)
+		arm_ni_event_read(event);
+}
+
+static void arm_ni_init_ccnt(struct arm_ni_cd *cd)
+{
+	local64_set(&cd->ccnt->hw.prev_count, S64_MIN);
+	lo_hi_writeq_relaxed(S64_MIN, cd->pmu_base + NI_PMCCNTR_L);
+}
+
+static void arm_ni_init_evcnt(struct arm_ni_cd *cd, int idx)
+{
+	local64_set(&cd->evcnt[idx]->hw.prev_count, S32_MIN);
+	writel_relaxed(S32_MIN, cd->pmu_base + NI_PMEVCNTR(idx));
+}
+
+static int arm_ni_event_add(struct perf_event *event, int flags)
+{
+	struct arm_ni_cd *cd = pmu_to_cd(event->pmu);
+	struct hw_perf_event *hw = &event->hw;
+	struct arm_ni_unit *unit;
+	enum ni_node_type type = NI_EVENT_TYPE(event);
+	u32 reg;
+
+	if (type == NI_PMU) {
+		if (cd->ccnt)
+			return -ENOSPC;
+		hw->idx = NI_CCNT_IDX;
+		cd->ccnt = event;
+		arm_ni_init_ccnt(cd);
+	} else {
+		hw->idx = 0;
+		while (cd->evcnt[hw->idx]) {
+			if (++hw->idx == NI_NUM_COUNTERS)
+				return -ENOSPC;
+		}
+		cd->evcnt[hw->idx] = event;
+		unit = (void *)hw->config_base;
+		unit->event[hw->idx] = NI_EVENT_EVENTID(event);
+		arm_ni_init_evcnt(cd, hw->idx);
+		lo_hi_writeq_relaxed(le64_to_cpu(unit->pmusel), unit->pmusela);
+
+		reg = FIELD_PREP(NI_PMEVTYPER_NODE_TYPE, type) |
+		      FIELD_PREP(NI_PMEVTYPER_NODE_ID, NI_EVENT_NODEID(event));
+		writel_relaxed(reg, cd->pmu_base + NI_PMEVTYPER(hw->idx));
+	}
+	if (flags & PERF_EF_START)
+		arm_ni_event_start(event, 0);
+	return 0;
+}
+
+static void arm_ni_event_del(struct perf_event *event, int flags)
+{
+	struct arm_ni_cd *cd = pmu_to_cd(event->pmu);
+	struct hw_perf_event *hw = &event->hw;
+
+	arm_ni_event_stop(event, PERF_EF_UPDATE);
+
+	if (hw->idx == NI_CCNT_IDX)
+		cd->ccnt = NULL;
+	else
+		cd->evcnt[hw->idx] = NULL;
+}
+
+static irqreturn_t arm_ni_handle_irq(int irq, void *dev_id)
+{
+	struct arm_ni_cd *cd = dev_id;
+	irqreturn_t ret = IRQ_NONE;
+	u32 reg = readl_relaxed(cd->pmu_base + NI_PMOVSCLR);
+
+	if (reg & (1U << NI_CCNT_IDX)) {
+		ret = IRQ_HANDLED;
+		if (!(WARN_ON(!cd->ccnt))) {
+			arm_ni_event_read(cd->ccnt);
+			arm_ni_init_ccnt(cd);
+		}
+	}
+	for (int i = 0; i < NI_NUM_COUNTERS; i++) {
+		if (!(reg & (1U << i)))
+			continue;
+		ret = IRQ_HANDLED;
+		if (!(WARN_ON(!cd->evcnt[i]))) {
+			arm_ni_event_read(cd->evcnt[i]);
+			arm_ni_init_evcnt(cd, i);
+		}
+	}
+	writel_relaxed(reg, cd->pmu_base + NI_PMOVSCLR);
+	return ret;
+}
+
+static int arm_ni_init_cd(struct arm_ni *ni, struct arm_ni_node *node, u64 res_start)
+{
+	struct arm_ni_cd *cd = ni->cds + node->id;
+	const char *name;
+	int err;
+
+	cd->id = node->id;
+	cd->num_units = node->num_components;
+	cd->units = devm_kcalloc(ni->dev, cd->num_units, sizeof(*(cd->units)), GFP_KERNEL);
+	if (!cd->units)
+		return -ENOMEM;
+
+	for (int i = 0; i < cd->num_units; i++) {
+		u32 reg = readl_relaxed(node->base + NI_CHILD_PTR(i));
+		void __iomem *unit_base = ni->base + reg;
+		struct arm_ni_unit *unit = cd->units + i;
+
+		reg = readl_relaxed(unit_base + NI_NODE_TYPE);
+		unit->type = FIELD_GET(NI_NODE_TYPE_NODE_TYPE, reg);
+		unit->id = FIELD_GET(NI_NODE_TYPE_NODE_ID, reg);
+
+		switch (unit->type) {
+		case NI_PMU:
+			reg = readl_relaxed(unit_base + NI_PMCFGR);
+			if (!reg) {
+				dev_info(ni->dev, "No access to PMU %d\n", cd->id);
+				devm_kfree(ni->dev, cd->units);
+				return 0;
+			}
+			unit->ns = true;
+			cd->pmu_base = unit_base;
+			break;
+		case NI_ASNI:
+		case NI_AMNI:
+		case NI_HSNI:
+		case NI_HMNI:
+		case NI_PMNI:
+			unit->pmusela = unit_base + NI700_PMUSELA;
+			writel_relaxed(1, unit->pmusela);
+			if (readl_relaxed(unit->pmusela) != 1)
+				dev_info(ni->dev, "No access to node 0x%04x%04x\n", unit->id, unit->type);
+			else
+				unit->ns = true;
+			break;
+		default:
+			/*
+			 * e.g. FMU - thankfully bits 3:2 of FMU_ERR_FR0 are RES0 so
+			 * can't alias any of the leaf node types we're looking for.
+			 */
+			dev_dbg(ni->dev, "Mystery node 0x%04x%04x\n", unit->id, unit->type);
+			break;
+		}
+	}
+
+	res_start += cd->pmu_base - ni->base;
+	if (!devm_request_mem_region(ni->dev, res_start, SZ_4K, dev_name(ni->dev))) {
+		dev_err(ni->dev, "Failed to request PMU region 0x%llx\n", res_start);
+		return -EBUSY;
+	}
+
+	writel_relaxed(NI_PMCR_RESET_CCNT | NI_PMCR_RESET_EVCNT,
+		       cd->pmu_base + NI_PMCR);
+	writel_relaxed(U32_MAX, cd->pmu_base + NI_PMCNTENCLR);
+	writel_relaxed(U32_MAX, cd->pmu_base + NI_PMOVSCLR);
+	writel_relaxed(U32_MAX, cd->pmu_base + NI_PMINTENSET);
+
+	cd->irq = platform_get_irq(to_platform_device(ni->dev), cd->id);
+	if (cd->irq < 0)
+		return cd->irq;
+
+	err = devm_request_irq(ni->dev, cd->irq, arm_ni_handle_irq,
+			       IRQF_NOBALANCING | IRQF_NO_THREAD,
+			       dev_name(ni->dev), cd);
+	if (err)
+		return err;
+
+	cd->cpu = cpumask_local_spread(0, dev_to_node(ni->dev));
+	cd->pmu = (struct pmu) {
+		.module = THIS_MODULE,
+		.parent = ni->dev,
+		.attr_groups = arm_ni_attr_groups,
+		.capabilities = PERF_PMU_CAP_NO_EXCLUDE,
+		.task_ctx_nr = perf_invalid_context,
+		.pmu_enable = arm_ni_pmu_enable,
+		.pmu_disable = arm_ni_pmu_disable,
+		.event_init = arm_ni_event_init,
+		.add = arm_ni_event_add,
+		.del = arm_ni_event_del,
+		.start = arm_ni_event_start,
+		.stop = arm_ni_event_stop,
+		.read = arm_ni_event_read,
+	};
+
+	name = devm_kasprintf(ni->dev, GFP_KERNEL, "arm_ni_%d_cd_%d", ni->id, cd->id);
+	if (!name)
+		return -ENOMEM;
+
+	err = cpuhp_state_add_instance_nocalls(arm_ni_hp_state, &cd->cpuhp_node);
+	if (err)
+		return err;
+
+	err = perf_pmu_register(&cd->pmu, name, -1);
+	if (err)
+		cpuhp_state_remove_instance_nocalls(arm_ni_hp_state, &cd->cpuhp_node);
+
+	return err;
+}
+
+static void arm_ni_probe_domain(void __iomem *base, struct arm_ni_node *node)
+{
+	u32 reg = readl_relaxed(base + NI_NODE_TYPE);
+
+	node->base = base;
+	node->type = FIELD_GET(NI_NODE_TYPE_NODE_TYPE, reg);
+	node->id = FIELD_GET(NI_NODE_TYPE_NODE_ID, reg);
+	node->num_components = readl_relaxed(base + NI_CHILD_NODE_INFO);
+}
+
+static int arm_ni_probe(struct platform_device *pdev)
+{
+	struct arm_ni_node cfg, vd, pd, cd;
+	struct arm_ni *ni;
+	struct resource *res;
+	void __iomem *base;
+	static atomic_t id;
+	int num_cds;
+	u32 reg, part;
+
+	/*
+	 * We want to map the whole configuration space for ease of discovery,
+	 * but the PMU pages are the only ones for which we can honestly claim
+	 * exclusive ownership, so we'll request them explicitly once found.
+	 */
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	base = devm_ioremap(&pdev->dev, res->start, resource_size(res));
+	if (!base)
+		return -ENOMEM;
+
+	arm_ni_probe_domain(base, &cfg);
+	if (cfg.type != NI_GLOBAL)
+		return -ENODEV;
+
+	reg = readl_relaxed(cfg.base + NI_PERIPHERAL_ID0);
+	part = FIELD_GET(NI_PIDR0_PART_7_0, reg);
+	reg = readl_relaxed(cfg.base + NI_PERIPHERAL_ID1);
+	part |= FIELD_GET(NI_PIDR1_PART_11_8, reg) << 8;
+
+	switch (part) {
+	case PART_NI_700:
+	case PART_NI_710AE:
+		break;
+	default:
+		dev_WARN(&pdev->dev, "Unknown part number: 0x%03x, this may go badly\n", part);
+		break;
+	}
+
+	num_cds = 0;
+	for (int v = 0; v < cfg.num_components; v++) {
+		reg = readl_relaxed(cfg.base + NI_CHILD_PTR(v));
+		arm_ni_probe_domain(base + reg, &vd);
+		for (int p = 0; p < vd.num_components; p++) {
+			reg = readl_relaxed(vd.base + NI_CHILD_PTR(p));
+			arm_ni_probe_domain(base + reg, &pd);
+			num_cds += pd.num_components;
+		}
+	}
+
+	ni = devm_kzalloc(&pdev->dev, struct_size(ni, cds, num_cds), GFP_KERNEL);
+	if (!ni)
+		return -ENOMEM;
+
+	ni->dev = &pdev->dev;
+	ni->base = base;
+	ni->num_cds = num_cds;
+	ni->part = part;
+	ni->id = atomic_fetch_inc(&id);
+
+	for (int v = 0; v < cfg.num_components; v++) {
+		reg = readl_relaxed(cfg.base + NI_CHILD_PTR(v));
+		arm_ni_probe_domain(base + reg, &vd);
+		for (int p = 0; p < vd.num_components; p++) {
+			reg = readl_relaxed(vd.base + NI_CHILD_PTR(p));
+			arm_ni_probe_domain(base + reg, &pd);
+			for (int c = 0; c < pd.num_components; c++) {
+				int ret;
+
+				reg = readl_relaxed(pd.base + NI_CHILD_PTR(c));
+				arm_ni_probe_domain(base + reg, &cd);
+				ret = arm_ni_init_cd(ni, &cd, res->start);
+				if (ret)
+					return ret;
+			}
+		}
+	}
+
+	return 0;
+}
+
+static void arm_ni_remove(struct platform_device *pdev)
+{
+	struct arm_ni *ni = platform_get_drvdata(pdev);
+
+	for (int i = 0; i < ni->num_cds; i++) {
+		struct arm_ni_cd *cd = ni->cds + i;
+
+		if (!cd->pmu_base)
+			continue;
+
+		writel_relaxed(0, cd->pmu_base + NI_PMCR);
+		writel_relaxed(U32_MAX, cd->pmu_base + NI_PMINTENCLR);
+		perf_pmu_unregister(&cd->pmu);
+		cpuhp_state_remove_instance_nocalls(arm_ni_hp_state, &cd->cpuhp_node);
+	}
+}
+
+#ifdef CONFIG_OF
+static const struct of_device_id arm_ni_of_match[] = {
+	{ .compatible = "arm,ni-700" },
+	{}
+};
+MODULE_DEVICE_TABLE(of, arm_ni_of_match);
+#endif
+
+#ifdef CONFIG_ACPI
+static const struct acpi_device_id arm_ni_acpi_match[] = {
+	{ "ARMHCB70" },
+	{}
+};
+MODULE_DEVICE_TABLE(acpi, arm_ni_acpi_match);
+#endif
+
+static struct platform_driver arm_ni_driver = {
+	.driver = {
+		.name = "arm-ni",
+		.of_match_table = of_match_ptr(arm_ni_of_match),
+		.acpi_match_table = ACPI_PTR(arm_ni_acpi_match),
+	},
+	.probe = arm_ni_probe,
+	.remove = arm_ni_remove,
+};
+
+static void arm_ni_pmu_migrate(struct arm_ni_cd *cd, unsigned int cpu)
+{
+	perf_pmu_migrate_context(&cd->pmu, cd->cpu, cpu);
+	irq_set_affinity(cd->irq, cpumask_of(cpu));
+	cd->cpu = cpu;
+}
+
+static int arm_ni_pmu_online_cpu(unsigned int cpu, struct hlist_node *cpuhp_node)
+{
+	struct arm_ni_cd *cd;
+	int node;
+
+	cd = hlist_entry_safe(cpuhp_node, struct arm_ni_cd, cpuhp_node);
+	node = dev_to_node(cd_to_ni(cd)->dev);
+	if (cpu_to_node(cd->cpu) != node && cpu_to_node(cpu) == node)
+		arm_ni_pmu_migrate(cd, cpu);
+	return 0;
+}
+
+static int arm_ni_pmu_offline_cpu(unsigned int cpu, struct hlist_node *cpuhp_node)
+{
+	struct arm_ni_cd *cd;
+	unsigned int target;
+	int node;
+
+	cd = hlist_entry_safe(cpuhp_node, struct arm_ni_cd, cpuhp_node);
+	if (cpu != cd->cpu)
+		return 0;
+
+	node = dev_to_node(cd_to_ni(cd)->dev);
+	target = cpumask_any_and_but(cpumask_of_node(node), cpu_online_mask, cpu);
+	if (target >= nr_cpu_ids)
+		target = cpumask_any_but(cpu_online_mask, cpu);
+
+	if (target < nr_cpu_ids)
+		arm_ni_pmu_migrate(cd, target);
+	return 0;
+}
+
+static int __init arm_ni_init(void)
+{
+	int ret;
+
+	ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN,
+				      "perf/arm/ni:online",
+				      arm_ni_pmu_online_cpu,
+				      arm_ni_pmu_offline_cpu);
+	if (ret < 0)
+		return ret;
+
+	arm_ni_hp_state = ret;
+
+	ret = platform_driver_register(&arm_ni_driver);
+	if (ret)
+		cpuhp_remove_multi_state(arm_ni_hp_state);
+	return ret;
+}
+
+static void __exit arm_ni_exit(void)
+{
+	platform_driver_unregister(&arm_ni_driver);
+	cpuhp_remove_multi_state(arm_ni_hp_state);
+}
+
+module_init(arm_ni_init);
+module_exit(arm_ni_exit);
+
+MODULE_AUTHOR("Robin Murphy <robin.murphy@arm.com>");
+MODULE_DESCRIPTION("Arm NI-700 PMU driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/perf/arm_pmu.c b/drivers/perf/arm_pmu.c
index 8458fe2cebb4..398cce3d76fc 100644
--- a/drivers/perf/arm_pmu.c
+++ b/drivers/perf/arm_pmu.c
@@ -522,7 +522,7 @@ 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);
-	bool enabled = !bitmap_empty(hw_events->used_mask, armpmu->num_events);
+	bool enabled = !bitmap_empty(hw_events->used_mask, ARMPMU_MAX_HWEVENTS);
 
 	/* For task-bound events we may be called on other CPUs */
 	if (!cpumask_test_cpu(smp_processor_id(), &armpmu->supported_cpus))
@@ -742,7 +742,7 @@ static void cpu_pm_pmu_setup(struct arm_pmu *armpmu, unsigned long cmd)
 	struct perf_event *event;
 	int idx;
 
-	for (idx = 0; idx < armpmu->num_events; idx++) {
+	for_each_set_bit(idx, armpmu->cntr_mask, ARMPMU_MAX_HWEVENTS) {
 		event = hw_events->events[idx];
 		if (!event)
 			continue;
@@ -772,7 +772,7 @@ static int cpu_pm_pmu_notify(struct notifier_block *b, unsigned long cmd,
 {
 	struct arm_pmu *armpmu = container_of(b, struct arm_pmu, cpu_pm_nb);
 	struct pmu_hw_events *hw_events = this_cpu_ptr(armpmu->hw_events);
-	bool enabled = !bitmap_empty(hw_events->used_mask, armpmu->num_events);
+	bool enabled = !bitmap_empty(hw_events->used_mask, ARMPMU_MAX_HWEVENTS);
 
 	if (!cpumask_test_cpu(smp_processor_id(), &armpmu->supported_cpus))
 		return NOTIFY_DONE;
@@ -924,8 +924,9 @@ int armpmu_register(struct arm_pmu *pmu)
 	if (ret)
 		goto out_destroy;
 
-	pr_info("enabled with %s PMU driver, %d counters available%s\n",
-		pmu->name, pmu->num_events,
+	pr_info("enabled with %s PMU driver, %d (%*pb) counters available%s\n",
+		pmu->name, bitmap_weight(pmu->cntr_mask, ARMPMU_MAX_HWEVENTS),
+		ARMPMU_MAX_HWEVENTS, &pmu->cntr_mask,
 		has_nmi ? ", using NMIs" : "");
 
 	kvm_host_pmu_init(pmu);
diff --git a/drivers/perf/arm_pmu_platform.c b/drivers/perf/arm_pmu_platform.c
index 4b1a9a92ea11..118170a5cede 100644
--- a/drivers/perf/arm_pmu_platform.c
+++ b/drivers/perf/arm_pmu_platform.c
@@ -59,7 +59,7 @@ static int pmu_parse_percpu_irq(struct arm_pmu *pmu, int irq)
 
 static bool pmu_has_irq_affinity(struct device_node *node)
 {
-	return !!of_find_property(node, "interrupt-affinity", NULL);
+	return of_property_present(node, "interrupt-affinity");
 }
 
 static int pmu_parse_irq_affinity(struct device *dev, int i)
diff --git a/drivers/perf/arm_pmuv3.c b/drivers/perf/arm_pmuv3.c
index d246840797b6..0afe02f879b4 100644
--- a/drivers/perf/arm_pmuv3.c
+++ b/drivers/perf/arm_pmuv3.c
@@ -452,13 +452,6 @@ static const struct attribute_group armv8_pmuv3_caps_attr_group = {
 };
 
 /*
- * Perf Events' indices
- */
-#define	ARMV8_IDX_CYCLE_COUNTER	0
-#define	ARMV8_IDX_COUNTER0	1
-#define	ARMV8_IDX_CYCLE_COUNTER_USER	32
-
-/*
  * We unconditionally enable ARMv8.5-PMU long event counter support
  * (64-bit events) where supported. Indicate if this arm_pmu has long
  * event counter support.
@@ -489,19 +482,12 @@ static bool armv8pmu_event_is_chained(struct perf_event *event)
 	return !armv8pmu_event_has_user_read(event) &&
 	       armv8pmu_event_is_64bit(event) &&
 	       !armv8pmu_has_long_event(cpu_pmu) &&
-	       (idx != ARMV8_IDX_CYCLE_COUNTER);
+	       (idx < ARMV8_PMU_MAX_GENERAL_COUNTERS);
 }
 
 /*
  * ARMv8 low level PMU access
  */
-
-/*
- * Perf Event to low level counters mapping
- */
-#define	ARMV8_IDX_TO_COUNTER(x)	\
-	(((x) - ARMV8_IDX_COUNTER0) & ARMV8_PMU_COUNTER_MASK)
-
 static u64 armv8pmu_pmcr_read(void)
 {
 	return read_pmcr();
@@ -514,21 +500,19 @@ static void armv8pmu_pmcr_write(u64 val)
 	write_pmcr(val);
 }
 
-static int armv8pmu_has_overflowed(u32 pmovsr)
+static int armv8pmu_has_overflowed(u64 pmovsr)
 {
-	return pmovsr & ARMV8_PMU_OVERFLOWED_MASK;
+	return !!(pmovsr & ARMV8_PMU_OVERFLOWED_MASK);
 }
 
-static int armv8pmu_counter_has_overflowed(u32 pmnc, int idx)
+static int armv8pmu_counter_has_overflowed(u64 pmnc, int idx)
 {
-	return pmnc & BIT(ARMV8_IDX_TO_COUNTER(idx));
+	return !!(pmnc & BIT(idx));
 }
 
 static u64 armv8pmu_read_evcntr(int idx)
 {
-	u32 counter = ARMV8_IDX_TO_COUNTER(idx);
-
-	return read_pmevcntrn(counter);
+	return read_pmevcntrn(idx);
 }
 
 static u64 armv8pmu_read_hw_counter(struct perf_event *event)
@@ -557,7 +541,7 @@ static bool armv8pmu_event_needs_bias(struct perf_event *event)
 		return false;
 
 	if (armv8pmu_has_long_event(cpu_pmu) ||
-	    idx == ARMV8_IDX_CYCLE_COUNTER)
+	    idx >= ARMV8_PMU_MAX_GENERAL_COUNTERS)
 		return true;
 
 	return false;
@@ -585,8 +569,10 @@ static u64 armv8pmu_read_counter(struct perf_event *event)
 	int idx = hwc->idx;
 	u64 value;
 
-	if (idx == ARMV8_IDX_CYCLE_COUNTER)
+	if (idx == ARMV8_PMU_CYCLE_IDX)
 		value = read_pmccntr();
+	else if (idx == ARMV8_PMU_INSTR_IDX)
+		value = read_pmicntr();
 	else
 		value = armv8pmu_read_hw_counter(event);
 
@@ -595,9 +581,7 @@ static u64 armv8pmu_read_counter(struct perf_event *event)
 
 static void armv8pmu_write_evcntr(int idx, u64 value)
 {
-	u32 counter = ARMV8_IDX_TO_COUNTER(idx);
-
-	write_pmevcntrn(counter, value);
+	write_pmevcntrn(idx, value);
 }
 
 static void armv8pmu_write_hw_counter(struct perf_event *event,
@@ -620,15 +604,16 @@ static void armv8pmu_write_counter(struct perf_event *event, u64 value)
 
 	value = armv8pmu_bias_long_counter(event, value);
 
-	if (idx == ARMV8_IDX_CYCLE_COUNTER)
+	if (idx == ARMV8_PMU_CYCLE_IDX)
 		write_pmccntr(value);
+	else if (idx == ARMV8_PMU_INSTR_IDX)
+		write_pmicntr(value);
 	else
 		armv8pmu_write_hw_counter(event, value);
 }
 
 static void armv8pmu_write_evtype(int idx, unsigned long val)
 {
-	u32 counter = ARMV8_IDX_TO_COUNTER(idx);
 	unsigned long mask = ARMV8_PMU_EVTYPE_EVENT |
 			     ARMV8_PMU_INCLUDE_EL2 |
 			     ARMV8_PMU_EXCLUDE_EL0 |
@@ -638,7 +623,7 @@ static void armv8pmu_write_evtype(int idx, unsigned long val)
 		mask |= ARMV8_PMU_EVTYPE_TC | ARMV8_PMU_EVTYPE_TH;
 
 	val &= mask;
-	write_pmevtypern(counter, val);
+	write_pmevtypern(idx, val);
 }
 
 static void armv8pmu_write_event_type(struct perf_event *event)
@@ -658,24 +643,26 @@ static void armv8pmu_write_event_type(struct perf_event *event)
 		armv8pmu_write_evtype(idx - 1, hwc->config_base);
 		armv8pmu_write_evtype(idx, chain_evt);
 	} else {
-		if (idx == ARMV8_IDX_CYCLE_COUNTER)
+		if (idx == ARMV8_PMU_CYCLE_IDX)
 			write_pmccfiltr(hwc->config_base);
+		else if (idx == ARMV8_PMU_INSTR_IDX)
+			write_pmicfiltr(hwc->config_base);
 		else
 			armv8pmu_write_evtype(idx, hwc->config_base);
 	}
 }
 
-static u32 armv8pmu_event_cnten_mask(struct perf_event *event)
+static u64 armv8pmu_event_cnten_mask(struct perf_event *event)
 {
-	int counter = ARMV8_IDX_TO_COUNTER(event->hw.idx);
-	u32 mask = BIT(counter);
+	int counter = event->hw.idx;
+	u64 mask = BIT(counter);
 
 	if (armv8pmu_event_is_chained(event))
 		mask |= BIT(counter - 1);
 	return mask;
 }
 
-static void armv8pmu_enable_counter(u32 mask)
+static void armv8pmu_enable_counter(u64 mask)
 {
 	/*
 	 * Make sure event configuration register writes are visible before we
@@ -688,7 +675,7 @@ static void armv8pmu_enable_counter(u32 mask)
 static void armv8pmu_enable_event_counter(struct perf_event *event)
 {
 	struct perf_event_attr *attr = &event->attr;
-	u32 mask = armv8pmu_event_cnten_mask(event);
+	u64 mask = armv8pmu_event_cnten_mask(event);
 
 	kvm_set_pmu_events(mask, attr);
 
@@ -697,7 +684,7 @@ static void armv8pmu_enable_event_counter(struct perf_event *event)
 		armv8pmu_enable_counter(mask);
 }
 
-static void armv8pmu_disable_counter(u32 mask)
+static void armv8pmu_disable_counter(u64 mask)
 {
 	write_pmcntenclr(mask);
 	/*
@@ -710,7 +697,7 @@ static void armv8pmu_disable_counter(u32 mask)
 static void armv8pmu_disable_event_counter(struct perf_event *event)
 {
 	struct perf_event_attr *attr = &event->attr;
-	u32 mask = armv8pmu_event_cnten_mask(event);
+	u64 mask = armv8pmu_event_cnten_mask(event);
 
 	kvm_clr_pmu_events(mask);
 
@@ -719,18 +706,17 @@ static void armv8pmu_disable_event_counter(struct perf_event *event)
 		armv8pmu_disable_counter(mask);
 }
 
-static void armv8pmu_enable_intens(u32 mask)
+static void armv8pmu_enable_intens(u64 mask)
 {
 	write_pmintenset(mask);
 }
 
 static void armv8pmu_enable_event_irq(struct perf_event *event)
 {
-	u32 counter = ARMV8_IDX_TO_COUNTER(event->hw.idx);
-	armv8pmu_enable_intens(BIT(counter));
+	armv8pmu_enable_intens(BIT(event->hw.idx));
 }
 
-static void armv8pmu_disable_intens(u32 mask)
+static void armv8pmu_disable_intens(u64 mask)
 {
 	write_pmintenclr(mask);
 	isb();
@@ -741,13 +727,12 @@ static void armv8pmu_disable_intens(u32 mask)
 
 static void armv8pmu_disable_event_irq(struct perf_event *event)
 {
-	u32 counter = ARMV8_IDX_TO_COUNTER(event->hw.idx);
-	armv8pmu_disable_intens(BIT(counter));
+	armv8pmu_disable_intens(BIT(event->hw.idx));
 }
 
-static u32 armv8pmu_getreset_flags(void)
+static u64 armv8pmu_getreset_flags(void)
 {
-	u32 value;
+	u64 value;
 
 	/* Read */
 	value = read_pmovsclr();
@@ -786,9 +771,12 @@ static void armv8pmu_enable_user_access(struct arm_pmu *cpu_pmu)
 	struct pmu_hw_events *cpuc = this_cpu_ptr(cpu_pmu->hw_events);
 
 	/* Clear any unused counters to avoid leaking their contents */
-	for_each_clear_bit(i, cpuc->used_mask, cpu_pmu->num_events) {
-		if (i == ARMV8_IDX_CYCLE_COUNTER)
+	for_each_andnot_bit(i, cpu_pmu->cntr_mask, cpuc->used_mask,
+			    ARMPMU_MAX_HWEVENTS) {
+		if (i == ARMV8_PMU_CYCLE_IDX)
 			write_pmccntr(0);
+		else if (i == ARMV8_PMU_INSTR_IDX)
+			write_pmicntr(0);
 		else
 			armv8pmu_write_evcntr(i, 0);
 	}
@@ -842,7 +830,7 @@ static void armv8pmu_stop(struct arm_pmu *cpu_pmu)
 
 static irqreturn_t armv8pmu_handle_irq(struct arm_pmu *cpu_pmu)
 {
-	u32 pmovsr;
+	u64 pmovsr;
 	struct perf_sample_data data;
 	struct pmu_hw_events *cpuc = this_cpu_ptr(cpu_pmu->hw_events);
 	struct pt_regs *regs;
@@ -869,7 +857,7 @@ static irqreturn_t armv8pmu_handle_irq(struct arm_pmu *cpu_pmu)
 	 * to prevent skews in group events.
 	 */
 	armv8pmu_stop(cpu_pmu);
-	for (idx = 0; idx < cpu_pmu->num_events; ++idx) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, ARMPMU_MAX_HWEVENTS) {
 		struct perf_event *event = cpuc->events[idx];
 		struct hw_perf_event *hwc;
 
@@ -908,7 +896,7 @@ static int armv8pmu_get_single_idx(struct pmu_hw_events *cpuc,
 {
 	int idx;
 
-	for (idx = ARMV8_IDX_COUNTER0; idx < cpu_pmu->num_events; idx++) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, ARMV8_PMU_MAX_GENERAL_COUNTERS) {
 		if (!test_and_set_bit(idx, cpuc->used_mask))
 			return idx;
 	}
@@ -924,7 +912,9 @@ static int armv8pmu_get_chain_idx(struct pmu_hw_events *cpuc,
 	 * Chaining requires two consecutive event counters, where
 	 * the lower idx must be even.
 	 */
-	for (idx = ARMV8_IDX_COUNTER0 + 1; idx < cpu_pmu->num_events; idx += 2) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, ARMV8_PMU_MAX_GENERAL_COUNTERS) {
+		if (!(idx & 0x1))
+			continue;
 		if (!test_and_set_bit(idx, cpuc->used_mask)) {
 			/* Check if the preceding even counter is available */
 			if (!test_and_set_bit(idx - 1, cpuc->used_mask))
@@ -946,8 +936,8 @@ static int armv8pmu_get_event_idx(struct pmu_hw_events *cpuc,
 	/* Always prefer to place a cycle counter into the cycle counter. */
 	if ((evtype == ARMV8_PMUV3_PERFCTR_CPU_CYCLES) &&
 	    !armv8pmu_event_get_threshold(&event->attr)) {
-		if (!test_and_set_bit(ARMV8_IDX_CYCLE_COUNTER, cpuc->used_mask))
-			return ARMV8_IDX_CYCLE_COUNTER;
+		if (!test_and_set_bit(ARMV8_PMU_CYCLE_IDX, cpuc->used_mask))
+			return ARMV8_PMU_CYCLE_IDX;
 		else if (armv8pmu_event_is_64bit(event) &&
 			   armv8pmu_event_want_user_access(event) &&
 			   !armv8pmu_has_long_event(cpu_pmu))
@@ -955,6 +945,19 @@ static int armv8pmu_get_event_idx(struct pmu_hw_events *cpuc,
 	}
 
 	/*
+	 * Always prefer to place a instruction counter into the instruction counter,
+	 * but don't expose the instruction counter to userspace access as userspace
+	 * may not know how to handle it.
+	 */
+	if ((evtype == ARMV8_PMUV3_PERFCTR_INST_RETIRED) &&
+	    !armv8pmu_event_get_threshold(&event->attr) &&
+	    test_bit(ARMV8_PMU_INSTR_IDX, cpu_pmu->cntr_mask) &&
+	    !armv8pmu_event_want_user_access(event)) {
+		if (!test_and_set_bit(ARMV8_PMU_INSTR_IDX, cpuc->used_mask))
+			return ARMV8_PMU_INSTR_IDX;
+	}
+
+	/*
 	 * Otherwise use events counters
 	 */
 	if (armv8pmu_event_is_chained(event))
@@ -978,15 +981,7 @@ static int armv8pmu_user_event_idx(struct perf_event *event)
 	if (!sysctl_perf_user_access || !armv8pmu_event_has_user_read(event))
 		return 0;
 
-	/*
-	 * We remap the cycle counter index to 32 to
-	 * match the offset applied to the rest of
-	 * the counter indices.
-	 */
-	if (event->hw.idx == ARMV8_IDX_CYCLE_COUNTER)
-		return ARMV8_IDX_CYCLE_COUNTER_USER;
-
-	return event->hw.idx;
+	return event->hw.idx + 1;
 }
 
 /*
@@ -1061,14 +1056,16 @@ static int armv8pmu_set_event_filter(struct hw_perf_event *event,
 static void armv8pmu_reset(void *info)
 {
 	struct arm_pmu *cpu_pmu = (struct arm_pmu *)info;
-	u64 pmcr;
+	u64 pmcr, mask;
+
+	bitmap_to_arr64(&mask, cpu_pmu->cntr_mask, ARMPMU_MAX_HWEVENTS);
 
 	/* The counter and interrupt enable registers are unknown at reset. */
-	armv8pmu_disable_counter(U32_MAX);
-	armv8pmu_disable_intens(U32_MAX);
+	armv8pmu_disable_counter(mask);
+	armv8pmu_disable_intens(mask);
 
 	/* Clear the counters we flip at guest entry/exit */
-	kvm_clr_pmu_events(U32_MAX);
+	kvm_clr_pmu_events(mask);
 
 	/*
 	 * Initialize & Reset PMNC. Request overflow interrupt for
@@ -1089,14 +1086,14 @@ static int __armv8_pmuv3_map_event_id(struct arm_pmu *armpmu,
 	if (event->attr.type == PERF_TYPE_HARDWARE &&
 	    event->attr.config == PERF_COUNT_HW_BRANCH_INSTRUCTIONS) {
 
-		if (test_bit(ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED,
-			     armpmu->pmceid_bitmap))
-			return ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED;
-
 		if (test_bit(ARMV8_PMUV3_PERFCTR_BR_RETIRED,
 			     armpmu->pmceid_bitmap))
 			return ARMV8_PMUV3_PERFCTR_BR_RETIRED;
 
+		if (test_bit(ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED,
+			     armpmu->pmceid_bitmap))
+			return ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED;
+
 		return HW_OP_UNSUPPORTED;
 	}
 
@@ -1211,10 +1208,15 @@ static void __armv8pmu_probe_pmu(void *info)
 	probe->present = true;
 
 	/* Read the nb of CNTx counters supported from PMNC */
-	cpu_pmu->num_events = FIELD_GET(ARMV8_PMU_PMCR_N, armv8pmu_pmcr_read());
+	bitmap_set(cpu_pmu->cntr_mask,
+		   0, FIELD_GET(ARMV8_PMU_PMCR_N, armv8pmu_pmcr_read()));
 
 	/* Add the CPU cycles counter */
-	cpu_pmu->num_events += 1;
+	set_bit(ARMV8_PMU_CYCLE_IDX, cpu_pmu->cntr_mask);
+
+	/* Add the CPU instructions counter */
+	if (pmuv3_has_icntr())
+		set_bit(ARMV8_PMU_INSTR_IDX, cpu_pmu->cntr_mask);
 
 	pmceid[0] = pmceid_raw[0] = read_pmceid0();
 	pmceid[1] = pmceid_raw[1] = read_pmceid1();
diff --git a/drivers/perf/arm_spe_pmu.c b/drivers/perf/arm_spe_pmu.c
index 9100d82bfabc..3569050f9cf3 100644
--- a/drivers/perf/arm_spe_pmu.c
+++ b/drivers/perf/arm_spe_pmu.c
@@ -41,7 +41,7 @@
 
 /*
  * Cache if the event is allowed to trace Context information.
- * This allows us to perform the check, i.e, perfmon_capable(),
+ * This allows us to perform the check, i.e, perf_allow_kernel(),
  * in the context of the event owner, once, during the event_init().
  */
 #define SPE_PMU_HW_FLAGS_CX			0x00001
@@ -50,7 +50,7 @@ static_assert((PERF_EVENT_FLAG_ARCH & SPE_PMU_HW_FLAGS_CX) == SPE_PMU_HW_FLAGS_C
 
 static void set_spe_event_has_cx(struct perf_event *event)
 {
-	if (IS_ENABLED(CONFIG_PID_IN_CONTEXTIDR) && perfmon_capable())
+	if (IS_ENABLED(CONFIG_PID_IN_CONTEXTIDR) && !perf_allow_kernel(&event->attr))
 		event->hw.flags |= SPE_PMU_HW_FLAGS_CX;
 }
 
@@ -745,9 +745,8 @@ static int arm_spe_pmu_event_init(struct perf_event *event)
 
 	set_spe_event_has_cx(event);
 	reg = arm_spe_event_to_pmscr(event);
-	if (!perfmon_capable() &&
-	    (reg & (PMSCR_EL1_PA | PMSCR_EL1_PCT)))
-		return -EACCES;
+	if (reg & (PMSCR_EL1_PA | PMSCR_EL1_PCT))
+		return perf_allow_kernel(&event->attr);
 
 	return 0;
 }
diff --git a/drivers/perf/arm_v6_pmu.c b/drivers/perf/arm_v6_pmu.c
index 0bb685b4bac5..b09615bb2bb2 100644
--- a/drivers/perf/arm_v6_pmu.c
+++ b/drivers/perf/arm_v6_pmu.c
@@ -64,6 +64,7 @@ enum armv6_counters {
 	ARMV6_CYCLE_COUNTER = 0,
 	ARMV6_COUNTER0,
 	ARMV6_COUNTER1,
+	ARMV6_NUM_COUNTERS
 };
 
 /*
@@ -254,7 +255,7 @@ armv6pmu_handle_irq(struct arm_pmu *cpu_pmu)
 	 */
 	armv6_pmcr_write(pmcr);
 
-	for (idx = 0; idx < cpu_pmu->num_events; ++idx) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, ARMV6_NUM_COUNTERS) {
 		struct perf_event *event = cpuc->events[idx];
 		struct hw_perf_event *hwc;
 
@@ -391,7 +392,8 @@ static void armv6pmu_init(struct arm_pmu *cpu_pmu)
 	cpu_pmu->start		= armv6pmu_start;
 	cpu_pmu->stop		= armv6pmu_stop;
 	cpu_pmu->map_event	= armv6_map_event;
-	cpu_pmu->num_events	= 3;
+
+	bitmap_set(cpu_pmu->cntr_mask, 0, ARMV6_NUM_COUNTERS);
 }
 
 static int armv6_1136_pmu_init(struct arm_pmu *cpu_pmu)
diff --git a/drivers/perf/arm_v7_pmu.c b/drivers/perf/arm_v7_pmu.c
index 928ac3d626ed..420cadd108e7 100644
--- a/drivers/perf/arm_v7_pmu.c
+++ b/drivers/perf/arm_v7_pmu.c
@@ -649,25 +649,13 @@ static struct attribute_group armv7_pmuv2_events_attr_group = {
 /*
  * Perf Events' indices
  */
-#define	ARMV7_IDX_CYCLE_COUNTER	0
-#define	ARMV7_IDX_COUNTER0	1
-#define	ARMV7_IDX_COUNTER_LAST(cpu_pmu) \
-	(ARMV7_IDX_CYCLE_COUNTER + cpu_pmu->num_events - 1)
-
-#define	ARMV7_MAX_COUNTERS	32
-#define	ARMV7_COUNTER_MASK	(ARMV7_MAX_COUNTERS - 1)
-
+#define	ARMV7_IDX_CYCLE_COUNTER	31
+#define	ARMV7_IDX_COUNTER_MAX	31
 /*
  * ARMv7 low level PMNC access
  */
 
 /*
- * Perf Event to low level counters mapping
- */
-#define	ARMV7_IDX_TO_COUNTER(x)	\
-	(((x) - ARMV7_IDX_COUNTER0) & ARMV7_COUNTER_MASK)
-
-/*
  * Per-CPU PMNC: config reg
  */
 #define ARMV7_PMNC_E		(1 << 0) /* Enable all counters */
@@ -725,19 +713,17 @@ static inline int armv7_pmnc_has_overflowed(u32 pmnc)
 
 static inline int armv7_pmnc_counter_valid(struct arm_pmu *cpu_pmu, int idx)
 {
-	return idx >= ARMV7_IDX_CYCLE_COUNTER &&
-		idx <= ARMV7_IDX_COUNTER_LAST(cpu_pmu);
+	return test_bit(idx, cpu_pmu->cntr_mask);
 }
 
 static inline int armv7_pmnc_counter_has_overflowed(u32 pmnc, int idx)
 {
-	return pmnc & BIT(ARMV7_IDX_TO_COUNTER(idx));
+	return pmnc & BIT(idx);
 }
 
 static inline void armv7_pmnc_select_counter(int idx)
 {
-	u32 counter = ARMV7_IDX_TO_COUNTER(idx);
-	asm volatile("mcr p15, 0, %0, c9, c12, 5" : : "r" (counter));
+	asm volatile("mcr p15, 0, %0, c9, c12, 5" : : "r" (idx));
 	isb();
 }
 
@@ -787,29 +773,25 @@ static inline void armv7_pmnc_write_evtsel(int idx, u32 val)
 
 static inline void armv7_pmnc_enable_counter(int idx)
 {
-	u32 counter = ARMV7_IDX_TO_COUNTER(idx);
-	asm volatile("mcr p15, 0, %0, c9, c12, 1" : : "r" (BIT(counter)));
+	asm volatile("mcr p15, 0, %0, c9, c12, 1" : : "r" (BIT(idx)));
 }
 
 static inline void armv7_pmnc_disable_counter(int idx)
 {
-	u32 counter = ARMV7_IDX_TO_COUNTER(idx);
-	asm volatile("mcr p15, 0, %0, c9, c12, 2" : : "r" (BIT(counter)));
+	asm volatile("mcr p15, 0, %0, c9, c12, 2" : : "r" (BIT(idx)));
 }
 
 static inline void armv7_pmnc_enable_intens(int idx)
 {
-	u32 counter = ARMV7_IDX_TO_COUNTER(idx);
-	asm volatile("mcr p15, 0, %0, c9, c14, 1" : : "r" (BIT(counter)));
+	asm volatile("mcr p15, 0, %0, c9, c14, 1" : : "r" (BIT(idx)));
 }
 
 static inline void armv7_pmnc_disable_intens(int idx)
 {
-	u32 counter = ARMV7_IDX_TO_COUNTER(idx);
-	asm volatile("mcr p15, 0, %0, c9, c14, 2" : : "r" (BIT(counter)));
+	asm volatile("mcr p15, 0, %0, c9, c14, 2" : : "r" (BIT(idx)));
 	isb();
 	/* Clear the overflow flag in case an interrupt is pending. */
-	asm volatile("mcr p15, 0, %0, c9, c12, 3" : : "r" (BIT(counter)));
+	asm volatile("mcr p15, 0, %0, c9, c12, 3" : : "r" (BIT(idx)));
 	isb();
 }
 
@@ -853,15 +835,12 @@ static void armv7_pmnc_dump_regs(struct arm_pmu *cpu_pmu)
 	asm volatile("mrc p15, 0, %0, c9, c13, 0" : "=r" (val));
 	pr_info("CCNT  =0x%08x\n", val);
 
-	for (cnt = ARMV7_IDX_COUNTER0;
-			cnt <= ARMV7_IDX_COUNTER_LAST(cpu_pmu); cnt++) {
+	for_each_set_bit(cnt, cpu_pmu->cntr_mask, ARMV7_IDX_COUNTER_MAX) {
 		armv7_pmnc_select_counter(cnt);
 		asm volatile("mrc p15, 0, %0, c9, c13, 2" : "=r" (val));
-		pr_info("CNT[%d] count =0x%08x\n",
-			ARMV7_IDX_TO_COUNTER(cnt), val);
+		pr_info("CNT[%d] count =0x%08x\n", cnt, val);
 		asm volatile("mrc p15, 0, %0, c9, c13, 1" : "=r" (val));
-		pr_info("CNT[%d] evtsel=0x%08x\n",
-			ARMV7_IDX_TO_COUNTER(cnt), val);
+		pr_info("CNT[%d] evtsel=0x%08x\n", cnt, val);
 	}
 }
 #endif
@@ -958,7 +937,7 @@ static irqreturn_t armv7pmu_handle_irq(struct arm_pmu *cpu_pmu)
 	 */
 	regs = get_irq_regs();
 
-	for (idx = 0; idx < cpu_pmu->num_events; ++idx) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, ARMPMU_MAX_HWEVENTS) {
 		struct perf_event *event = cpuc->events[idx];
 		struct hw_perf_event *hwc;
 
@@ -1027,7 +1006,7 @@ static int armv7pmu_get_event_idx(struct pmu_hw_events *cpuc,
 	 * For anything other than a cycle counter, try and use
 	 * the events counters
 	 */
-	for (idx = ARMV7_IDX_COUNTER0; idx < cpu_pmu->num_events; ++idx) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, ARMV7_IDX_COUNTER_MAX) {
 		if (!test_and_set_bit(idx, cpuc->used_mask))
 			return idx;
 	}
@@ -1073,7 +1052,7 @@ static int armv7pmu_set_event_filter(struct hw_perf_event *event,
 static void armv7pmu_reset(void *info)
 {
 	struct arm_pmu *cpu_pmu = (struct arm_pmu *)info;
-	u32 idx, nb_cnt = cpu_pmu->num_events, val;
+	u32 idx, val;
 
 	if (cpu_pmu->secure_access) {
 		asm volatile("mrc p15, 0, %0, c1, c1, 1" : "=r" (val));
@@ -1082,7 +1061,7 @@ static void armv7pmu_reset(void *info)
 	}
 
 	/* The counter and interrupt enable registers are unknown at reset. */
-	for (idx = ARMV7_IDX_CYCLE_COUNTER; idx < nb_cnt; ++idx) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, ARMPMU_MAX_HWEVENTS) {
 		armv7_pmnc_disable_counter(idx);
 		armv7_pmnc_disable_intens(idx);
 	}
@@ -1161,20 +1140,22 @@ static void armv7pmu_init(struct arm_pmu *cpu_pmu)
 
 static void armv7_read_num_pmnc_events(void *info)
 {
-	int *nb_cnt = info;
+	int nb_cnt;
+	struct arm_pmu *cpu_pmu = info;
 
 	/* Read the nb of CNTx counters supported from PMNC */
-	*nb_cnt = (armv7_pmnc_read() >> ARMV7_PMNC_N_SHIFT) & ARMV7_PMNC_N_MASK;
+	nb_cnt = (armv7_pmnc_read() >> ARMV7_PMNC_N_SHIFT) & ARMV7_PMNC_N_MASK;
+	bitmap_set(cpu_pmu->cntr_mask, 0, nb_cnt);
 
 	/* Add the CPU cycles counter */
-	*nb_cnt += 1;
+	set_bit(ARMV7_IDX_CYCLE_COUNTER, cpu_pmu->cntr_mask);
 }
 
 static int armv7_probe_num_events(struct arm_pmu *arm_pmu)
 {
 	return smp_call_function_any(&arm_pmu->supported_cpus,
 				     armv7_read_num_pmnc_events,
-				     &arm_pmu->num_events, 1);
+				     arm_pmu, 1);
 }
 
 static int armv7_a8_pmu_init(struct arm_pmu *cpu_pmu)
@@ -1524,7 +1505,7 @@ static void krait_pmu_reset(void *info)
 {
 	u32 vval, fval;
 	struct arm_pmu *cpu_pmu = info;
-	u32 idx, nb_cnt = cpu_pmu->num_events;
+	u32 idx;
 
 	armv7pmu_reset(info);
 
@@ -1538,7 +1519,7 @@ static void krait_pmu_reset(void *info)
 	venum_post_pmresr(vval, fval);
 
 	/* Reset PMxEVNCTCR to sane default */
-	for (idx = ARMV7_IDX_CYCLE_COUNTER; idx < nb_cnt; ++idx) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, ARMV7_IDX_COUNTER_MAX) {
 		armv7_pmnc_select_counter(idx);
 		asm volatile("mcr p15, 0, %0, c9, c15, 0" : : "r" (0));
 	}
@@ -1562,7 +1543,7 @@ static int krait_event_to_bit(struct perf_event *event, unsigned int region,
 	 * Lower bits are reserved for use by the counters (see
 	 * armv7pmu_get_event_idx() for more info)
 	 */
-	bit += ARMV7_IDX_COUNTER_LAST(cpu_pmu) + 1;
+	bit += bitmap_weight(cpu_pmu->cntr_mask, ARMV7_IDX_COUNTER_MAX);
 
 	return bit;
 }
@@ -1845,7 +1826,7 @@ static void scorpion_pmu_reset(void *info)
 {
 	u32 vval, fval;
 	struct arm_pmu *cpu_pmu = info;
-	u32 idx, nb_cnt = cpu_pmu->num_events;
+	u32 idx;
 
 	armv7pmu_reset(info);
 
@@ -1860,7 +1841,7 @@ static void scorpion_pmu_reset(void *info)
 	venum_post_pmresr(vval, fval);
 
 	/* Reset PMxEVNCTCR to sane default */
-	for (idx = ARMV7_IDX_CYCLE_COUNTER; idx < nb_cnt; ++idx) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, ARMV7_IDX_COUNTER_MAX) {
 		armv7_pmnc_select_counter(idx);
 		asm volatile("mcr p15, 0, %0, c9, c15, 0" : : "r" (0));
 	}
@@ -1883,7 +1864,7 @@ static int scorpion_event_to_bit(struct perf_event *event, unsigned int region,
 	 * Lower bits are reserved for use by the counters (see
 	 * armv7pmu_get_event_idx() for more info)
 	 */
-	bit += ARMV7_IDX_COUNTER_LAST(cpu_pmu) + 1;
+	bit += bitmap_weight(cpu_pmu->cntr_mask, ARMV7_IDX_COUNTER_MAX);
 
 	return bit;
 }
diff --git a/drivers/perf/arm_xscale_pmu.c b/drivers/perf/arm_xscale_pmu.c
index 3d8b72d6b37f..638fea9b1263 100644
--- a/drivers/perf/arm_xscale_pmu.c
+++ b/drivers/perf/arm_xscale_pmu.c
@@ -53,6 +53,8 @@ enum xscale_counters {
 	XSCALE_COUNTER2,
 	XSCALE_COUNTER3,
 };
+#define XSCALE1_NUM_COUNTERS	3
+#define XSCALE2_NUM_COUNTERS	5
 
 static const unsigned xscale_perf_map[PERF_COUNT_HW_MAX] = {
 	PERF_MAP_ALL_UNSUPPORTED,
@@ -168,7 +170,7 @@ xscale1pmu_handle_irq(struct arm_pmu *cpu_pmu)
 
 	regs = get_irq_regs();
 
-	for (idx = 0; idx < cpu_pmu->num_events; ++idx) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, XSCALE1_NUM_COUNTERS) {
 		struct perf_event *event = cpuc->events[idx];
 		struct hw_perf_event *hwc;
 
@@ -364,7 +366,8 @@ static int xscale1pmu_init(struct arm_pmu *cpu_pmu)
 	cpu_pmu->start		= xscale1pmu_start;
 	cpu_pmu->stop		= xscale1pmu_stop;
 	cpu_pmu->map_event	= xscale_map_event;
-	cpu_pmu->num_events	= 3;
+
+	bitmap_set(cpu_pmu->cntr_mask, 0, XSCALE1_NUM_COUNTERS);
 
 	return 0;
 }
@@ -500,7 +503,7 @@ xscale2pmu_handle_irq(struct arm_pmu *cpu_pmu)
 
 	regs = get_irq_regs();
 
-	for (idx = 0; idx < cpu_pmu->num_events; ++idx) {
+	for_each_set_bit(idx, cpu_pmu->cntr_mask, XSCALE2_NUM_COUNTERS) {
 		struct perf_event *event = cpuc->events[idx];
 		struct hw_perf_event *hwc;
 
@@ -719,7 +722,8 @@ static int xscale2pmu_init(struct arm_pmu *cpu_pmu)
 	cpu_pmu->start		= xscale2pmu_start;
 	cpu_pmu->stop		= xscale2pmu_stop;
 	cpu_pmu->map_event	= xscale_map_event;
-	cpu_pmu->num_events	= 5;
+
+	bitmap_set(cpu_pmu->cntr_mask, 0, XSCALE2_NUM_COUNTERS);
 
 	return 0;
 }
diff --git a/drivers/perf/dwc_pcie_pmu.c b/drivers/perf/dwc_pcie_pmu.c
index c5e328f23841..4ca50f9b6dfe 100644
--- a/drivers/perf/dwc_pcie_pmu.c
+++ b/drivers/perf/dwc_pcie_pmu.c
@@ -107,6 +107,7 @@ struct dwc_pcie_vendor_id {
 
 static const struct dwc_pcie_vendor_id dwc_pcie_vendor_ids[] = {
 	{.vendor_id = PCI_VENDOR_ID_ALIBABA },
+	{.vendor_id = PCI_VENDOR_ID_QCOM },
 	{} /* terminator */
 };
 
@@ -556,10 +557,10 @@ static int dwc_pcie_register_dev(struct pci_dev *pdev)
 {
 	struct platform_device *plat_dev;
 	struct dwc_pcie_dev_info *dev_info;
-	u32 bdf;
+	u32 sbdf;
 
-	bdf = PCI_DEVID(pdev->bus->number, pdev->devfn);
-	plat_dev = platform_device_register_data(NULL, "dwc_pcie_pmu", bdf,
+	sbdf = (pci_domain_nr(pdev->bus) << 16) | PCI_DEVID(pdev->bus->number, pdev->devfn);
+	plat_dev = platform_device_register_data(NULL, "dwc_pcie_pmu", sbdf,
 						 pdev, sizeof(*pdev));
 
 	if (IS_ERR(plat_dev))
@@ -611,15 +612,15 @@ static int dwc_pcie_pmu_probe(struct platform_device *plat_dev)
 	struct pci_dev *pdev = plat_dev->dev.platform_data;
 	struct dwc_pcie_pmu *pcie_pmu;
 	char *name;
-	u32 bdf, val;
+	u32 sbdf, val;
 	u16 vsec;
 	int ret;
 
 	vsec = pci_find_vsec_capability(pdev, pdev->vendor,
 					DWC_PCIE_VSEC_RAS_DES_ID);
 	pci_read_config_dword(pdev, vsec + PCI_VNDR_HEADER, &val);
-	bdf = PCI_DEVID(pdev->bus->number, pdev->devfn);
-	name = devm_kasprintf(&plat_dev->dev, GFP_KERNEL, "dwc_rootport_%x", bdf);
+	sbdf = plat_dev->id;
+	name = devm_kasprintf(&plat_dev->dev, GFP_KERNEL, "dwc_rootport_%x", sbdf);
 	if (!name)
 		return -ENOMEM;
 
@@ -650,7 +651,7 @@ static int dwc_pcie_pmu_probe(struct platform_device *plat_dev)
 	ret = cpuhp_state_add_instance(dwc_pcie_pmu_hp_state,
 				       &pcie_pmu->cpuhp_node);
 	if (ret) {
-		pci_err(pdev, "Error %d registering hotplug @%x\n", ret, bdf);
+		pci_err(pdev, "Error %d registering hotplug @%x\n", ret, sbdf);
 		return ret;
 	}
 
@@ -663,7 +664,7 @@ static int dwc_pcie_pmu_probe(struct platform_device *plat_dev)
 
 	ret = perf_pmu_register(&pcie_pmu->pmu, name, -1);
 	if (ret) {
-		pci_err(pdev, "Error %d registering PMU @%x\n", ret, bdf);
+		pci_err(pdev, "Error %d registering PMU @%x\n", ret, sbdf);
 		return ret;
 	}
 	ret = devm_add_action_or_reset(&plat_dev->dev, dwc_pcie_unregister_pmu,
@@ -726,7 +727,6 @@ static struct platform_driver dwc_pcie_pmu_driver = {
 static int __init dwc_pcie_pmu_init(void)
 {
 	struct pci_dev *pdev = NULL;
-	bool found = false;
 	int ret;
 
 	for_each_pci_dev(pdev) {
@@ -738,11 +738,7 @@ static int __init dwc_pcie_pmu_init(void)
 			pci_dev_put(pdev);
 			return ret;
 		}
-
-		found = true;
 	}
-	if (!found)
-		return -ENODEV;
 
 	ret = cpuhp_setup_state_multi(CPUHP_AP_ONLINE_DYN,
 				      "perf/dwc_pcie_pmu:online",
diff --git a/drivers/perf/hisilicon/hisi_pcie_pmu.c b/drivers/perf/hisilicon/hisi_pcie_pmu.c
index f06027574a24..c5394d007b61 100644
--- a/drivers/perf/hisilicon/hisi_pcie_pmu.c
+++ b/drivers/perf/hisilicon/hisi_pcie_pmu.c
@@ -141,6 +141,22 @@ static ssize_t bus_show(struct device *dev, struct device_attribute *attr, char
 }
 static DEVICE_ATTR_RO(bus);
 
+static ssize_t bdf_min_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct hisi_pcie_pmu *pcie_pmu = to_pcie_pmu(dev_get_drvdata(dev));
+
+	return sysfs_emit(buf, "%#04x\n", pcie_pmu->bdf_min);
+}
+static DEVICE_ATTR_RO(bdf_min);
+
+static ssize_t bdf_max_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct hisi_pcie_pmu *pcie_pmu = to_pcie_pmu(dev_get_drvdata(dev));
+
+	return sysfs_emit(buf, "%#04x\n", pcie_pmu->bdf_max);
+}
+static DEVICE_ATTR_RO(bdf_max);
+
 static struct hisi_pcie_reg_pair
 hisi_pcie_parse_reg_value(struct hisi_pcie_pmu *pcie_pmu, u32 reg_off)
 {
@@ -208,7 +224,7 @@ static void hisi_pcie_pmu_writeq(struct hisi_pcie_pmu *pcie_pmu, u32 reg_offset,
 static u64 hisi_pcie_pmu_get_event_ctrl_val(struct perf_event *event)
 {
 	u64 port, trig_len, thr_len, len_mode;
-	u64 reg = HISI_PCIE_INIT_SET;
+	u64 reg = 0;
 
 	/* Config HISI_PCIE_EVENT_CTRL according to event. */
 	reg |= FIELD_PREP(HISI_PCIE_EVENT_M, hisi_pcie_get_real_event(event));
@@ -452,10 +468,24 @@ static void hisi_pcie_pmu_set_period(struct perf_event *event)
 	struct hisi_pcie_pmu *pcie_pmu = to_pcie_pmu(event->pmu);
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
+	u64 orig_cnt, cnt;
+
+	orig_cnt = hisi_pcie_pmu_read_counter(event);
 
 	local64_set(&hwc->prev_count, HISI_PCIE_INIT_VAL);
 	hisi_pcie_pmu_writeq(pcie_pmu, HISI_PCIE_CNT, idx, HISI_PCIE_INIT_VAL);
 	hisi_pcie_pmu_writeq(pcie_pmu, HISI_PCIE_EXT_CNT, idx, HISI_PCIE_INIT_VAL);
+
+	/*
+	 * The counter maybe unwritable if the target event is unsupported.
+	 * Check this by comparing the counts after setting the period. If
+	 * the counts stay unchanged after setting the period then update
+	 * the hwc->prev_count correctly. Otherwise the final counts user
+	 * get maybe totally wrong.
+	 */
+	cnt = hisi_pcie_pmu_read_counter(event);
+	if (orig_cnt == cnt)
+		local64_set(&hwc->prev_count, cnt);
 }
 
 static void hisi_pcie_pmu_enable_counter(struct hisi_pcie_pmu *pcie_pmu, struct hw_perf_event *hwc)
@@ -749,6 +779,8 @@ static const struct attribute_group hisi_pcie_pmu_format_group = {
 
 static struct attribute *hisi_pcie_pmu_bus_attrs[] = {
 	&dev_attr_bus.attr,
+	&dev_attr_bdf_max.attr,
+	&dev_attr_bdf_min.attr,
 	NULL
 };
author	Linus Torvalds <torvalds@linux-foundation.org>	2024-09-16 06:55:07 +0200
committer	Linus Torvalds <torvalds@linux-foundation.org>	2024-09-16 06:55:07 +0200
commit	114143a595895c03fbefccfd8346fc51fb4908ed (patch)
tree	161f753ad69a2aabe7cd7ac7cc493c3678dbd760 /drivers/perf
parent	8617d7d6298f54dfef4038281863270b5864fe83 (diff)
parent	75078ba2b38a38d94017bd334f71aaed205e30a4 (diff)