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

Pull arm64 updates from Will Deacon:

 - Support for including MTE tags in ELF coredumps

 - Instruction encoder updates, including fixes to 64-bit immediate
   generation and support for the LSE atomic instructions

 - Improvements to kselftests for MTE and fpsimd

 - Symbol aliasing and linker script cleanups

 - Reduce instruction cache maintenance performed for user mappings
   created using contiguous PTEs

 - Support for the new "asymmetric" MTE mode, where stores are checked
   asynchronously but loads are checked synchronously

 - Support for the latest pointer authentication algorithm ("QARMA3")

 - Support for the DDR PMU present in the Marvell CN10K platform

 - Support for the CPU PMU present in the Apple M1 platform

 - Use the RNDR instruction for arch_get_random_{int,long}()

 - Update our copy of the Arm optimised string routines for str{n}cmp()

 - Fix signal frame generation for CPUs which have foolishly elected to
   avoid building in support for the fpsimd instructions

 - Workaround for Marvell GICv3 erratum #38545

 - Clarification to our Documentation (booting reqs. and MTE prctl())

 - Miscellanous cleanups and minor fixes

* tag 'arm64-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux: (90 commits)
  docs: sysfs-devices-system-cpu: document "asymm" value for mte_tcf_preferred
  arm64/mte: Remove asymmetric mode from the prctl() interface
  arm64: Add cavium_erratum_23154_cpus missing sentinel
  perf/marvell: Fix !CONFIG_OF build for CN10K DDR PMU driver
  arm64: mm: Drop 'const' from conditional arm64_dma_phys_limit definition
  Documentation: vmcoreinfo: Fix htmldocs warning
  kasan: fix a missing header include of static_keys.h
  drivers/perf: Add Apple icestorm/firestorm CPU PMU driver
  drivers/perf: arm_pmu: Handle 47 bit counters
  arm64: perf: Consistently make all event numbers as 16-bits
  arm64: perf: Expose some Armv9 common events under sysfs
  perf/marvell: cn10k DDR perf event core ownership
  perf/marvell: cn10k DDR perfmon event overflow handling
  perf/marvell: CN10k DDR performance monitor support
  dt-bindings: perf: marvell: cn10k ddr performance monitor
  arm64: clean up tools Makefile
  perf/arm-cmn: Update watchpoint format
  perf/arm-cmn: Hide XP PUB events for CMN-600
  arm64: drop unused includes of <linux/personality.h>
  arm64: Do not defer reserve_crashkernel() for platforms with no DMA memory zones
  ...

1 files changed, 72 insertions, 22 deletions

diff --git a/drivers/irqchip/irq-apple-aic.c b/drivers/irqchip/irq-apple-aic.c
index 38091ebb9403..b40199c6625e 100644
--- a/drivers/irqchip/irq-apple-aic.c
+++ b/drivers/irqchip/irq-apple-aic.c
@@ -55,6 +55,7 @@
 #include <linux/limits.h>
 #include <linux/of_address.h>
 #include <linux/slab.h>
+#include <asm/apple_m1_pmu.h>
 #include <asm/exception.h>
 #include <asm/sysreg.h>
 #include <asm/virt.h>
@@ -109,16 +110,6 @@
  * Note: sysreg-based IPIs are not supported yet.
  */
 
-/* Core PMC control register */
-#define SYS_IMP_APL_PMCR0_EL1		sys_reg(3, 1, 15, 0, 0)
-#define PMCR0_IMODE			GENMASK(10, 8)
-#define PMCR0_IMODE_OFF			0
-#define PMCR0_IMODE_PMI			1
-#define PMCR0_IMODE_AIC			2
-#define PMCR0_IMODE_HALT		3
-#define PMCR0_IMODE_FIQ			4
-#define PMCR0_IACT			BIT(11)
-
 /* IPI request registers */
 #define SYS_IMP_APL_IPI_RR_LOCAL_EL1	sys_reg(3, 5, 15, 0, 0)
 #define SYS_IMP_APL_IPI_RR_GLOBAL_EL1	sys_reg(3, 5, 15, 0, 1)
@@ -155,7 +146,7 @@
 #define SYS_IMP_APL_UPMSR_EL1		sys_reg(3, 7, 15, 6, 4)
 #define UPMSR_IACT			BIT(0)
 
-#define AIC_NR_FIQ		4
+#define AIC_NR_FIQ		6
 #define AIC_NR_SWIPI		32
 
 /*
@@ -177,6 +168,9 @@ struct aic_irq_chip {
 	void __iomem *base;
 	struct irq_domain *hw_domain;
 	struct irq_domain *ipi_domain;
+	struct {
+		cpumask_t aff;
+	} *fiq_aff[AIC_NR_FIQ];
 	int nr_hw;
 };
 
@@ -412,16 +406,15 @@ static void __exception_irq_entry aic_handle_fiq(struct pt_regs *regs)
 						  aic_irqc->nr_hw + AIC_TMR_EL02_VIRT);
 	}
 
-	if ((read_sysreg_s(SYS_IMP_APL_PMCR0_EL1) & (PMCR0_IMODE | PMCR0_IACT)) ==
-			(FIELD_PREP(PMCR0_IMODE, PMCR0_IMODE_FIQ) | PMCR0_IACT)) {
-		/*
-		 * Not supported yet, let's figure out how to handle this when
-		 * we implement these proprietary performance counters. For now,
-		 * just mask it and move on.
-		 */
-		pr_err_ratelimited("PMC FIQ fired. Masking.\n");
-		sysreg_clear_set_s(SYS_IMP_APL_PMCR0_EL1, PMCR0_IMODE | PMCR0_IACT,
-				   FIELD_PREP(PMCR0_IMODE, PMCR0_IMODE_OFF));
+	if (read_sysreg_s(SYS_IMP_APL_PMCR0_EL1) & PMCR0_IACT) {
+		int irq;
+		if (cpumask_test_cpu(smp_processor_id(),
+				     &aic_irqc->fiq_aff[AIC_CPU_PMU_P]->aff))
+			irq = AIC_CPU_PMU_P;
+		else
+			irq = AIC_CPU_PMU_E;
+		generic_handle_domain_irq(aic_irqc->hw_domain,
+					  aic_irqc->nr_hw + irq);
 	}
 
 	if (FIELD_GET(UPMCR0_IMODE, read_sysreg_s(SYS_IMP_APL_UPMCR0_EL1)) == UPMCR0_IMODE_FIQ &&
@@ -461,7 +454,18 @@ static int aic_irq_domain_map(struct irq_domain *id, unsigned int irq,
 				    handle_fasteoi_irq, NULL, NULL);
 		irqd_set_single_target(irq_desc_get_irq_data(irq_to_desc(irq)));
 	} else {
-		irq_set_percpu_devid(irq);
+		int fiq = hw - ic->nr_hw;
+
+		switch (fiq) {
+		case AIC_CPU_PMU_P:
+		case AIC_CPU_PMU_E:
+			irq_set_percpu_devid_partition(irq, &ic->fiq_aff[fiq]->aff);
+			break;
+		default:
+			irq_set_percpu_devid(irq);
+			break;
+		}
+
 		irq_domain_set_info(id, irq, hw, &fiq_chip, id->host_data,
 				    handle_percpu_devid_irq, NULL, NULL);
 	}
@@ -793,12 +797,50 @@ static struct gic_kvm_info vgic_info __initdata = {
 	.no_hw_deactivation	= true,
 };
 
+static void build_fiq_affinity(struct aic_irq_chip *ic, struct device_node *aff)
+{
+	int i, n;
+	u32 fiq;
+
+	if (of_property_read_u32(aff, "apple,fiq-index", &fiq) ||
+	    WARN_ON(fiq >= AIC_NR_FIQ) || ic->fiq_aff[fiq])
+		return;
+
+	n = of_property_count_elems_of_size(aff, "cpus", sizeof(u32));
+	if (WARN_ON(n < 0))
+		return;
+
+	ic->fiq_aff[fiq] = kzalloc(sizeof(ic->fiq_aff[fiq]), GFP_KERNEL);
+	if (!ic->fiq_aff[fiq])
+		return;
+
+	for (i = 0; i < n; i++) {
+		struct device_node *cpu_node;
+		u32 cpu_phandle;
+		int cpu;
+
+		if (of_property_read_u32_index(aff, "cpus", i, &cpu_phandle))
+			continue;
+
+		cpu_node = of_find_node_by_phandle(cpu_phandle);
+		if (WARN_ON(!cpu_node))
+			continue;
+
+		cpu = of_cpu_node_to_id(cpu_node);
+		if (WARN_ON(cpu < 0))
+			continue;
+
+		cpumask_set_cpu(cpu, &ic->fiq_aff[fiq]->aff);
+	}
+}
+
 static int __init aic_of_ic_init(struct device_node *node, struct device_node *parent)
 {
 	int i;
 	void __iomem *regs;
 	u32 info;
 	struct aic_irq_chip *irqc;
+	struct device_node *affs;
 
 	regs = of_iomap(node, 0);
 	if (WARN_ON(!regs))
@@ -832,6 +874,14 @@ static int __init aic_of_ic_init(struct device_node *node, struct device_node *p
 		return -ENODEV;
 	}
 
+	affs = of_get_child_by_name(node, "affinities");
+	if (affs) {
+		struct device_node *chld;
+
+		for_each_child_of_node(affs, chld)
+			build_fiq_affinity(irqc, chld);
+	}
+
 	set_handle_irq(aic_handle_irq);
 	set_handle_fiq(aic_handle_fiq);