Merge tag 'irq-core-2023-02-20' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull irq updates from Thomas Gleixner:
 "Updates for the interrupt subsystem:

  Core:

   - Move the interrupt affinity spreading mechanism into lib/group_cpus
     so it can be used for similar spreading requirements, e.g. in the
     block multi-queue code

     This also contains a first usecase in the block multi-queue code
     which Jens asked to take along with the librarization

   - Improve irqdomain locking to close a number race conditions which
     can be observed with massive parallel device driver probing

   - Enforce and document the semantics of disable_irq() which cannot be
     invoked safely from non-sleepable context

   - Move the IPI multiplexing code from the Apple AIC driver into the
     core, so it can be reused by RISCV

  Drivers:

   - Plug OF node refcounting leaks in various drivers

   - Correctly mark level triggered interrupts in the Broadcom L2
     drivers

   - The usual small fixes and improvements

   - No new drivers for the record!"

* tag 'irq-core-2023-02-20' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (42 commits)
  irqchip/irq-bcm7120-l2: Set IRQ_LEVEL for level triggered interrupts
  irqchip/irq-brcmstb-l2: Set IRQ_LEVEL for level triggered interrupts
  irqdomain: Switch to per-domain locking
  irqchip/mvebu-odmi: Use irq_domain_create_hierarchy()
  irqchip/loongson-pch-msi: Use irq_domain_create_hierarchy()
  irqchip/gic-v3-mbi: Use irq_domain_create_hierarchy()
  irqchip/gic-v3-its: Use irq_domain_create_hierarchy()
  irqchip/gic-v2m: Use irq_domain_create_hierarchy()
  irqchip/alpine-msi: Use irq_domain_add_hierarchy()
  x86/uv: Use irq_domain_create_hierarchy()
  x86/ioapic: Use irq_domain_create_hierarchy()
  irqdomain: Clean up irq_domain_push/pop_irq()
  irqdomain: Drop leftover brackets
  irqdomain: Drop dead domain-name assignment
  irqdomain: Drop revmap mutex
  irqdomain: Fix domain registration race
  irqdomain: Fix mapping-creation race
  irqdomain: Refactor __irq_domain_alloc_irqs()
  irqdomain: Look for existing mapping only once
  irqdomain: Drop bogus fwspec-mapping error handling
  ...

Documentation/kernel-hacking/locking.rst

@@ -1277,11 +1277,11 @@ Manfred Spraul points out that you can still do this, even if the data
 is very occasionally accessed in user context or softirqs/tasklets. The
 irq handler doesn't use a lock, and all other accesses are done as so::
 
-        spin_lock(&lock);
+        mutex_lock(&lock);
         disable_irq(irq);
         ...
         enable_irq(irq);
-        spin_unlock(&lock);
+        mutex_unlock(&lock);
 
 The disable_irq() prevents the irq handler from running
 (and waits for it to finish if it's currently running on other CPUs).

Documentation/translations/it_IT/kernel-hacking/locking.rst

@@ -1307,11 +1307,11 @@ se i dati vengono occasionalmente utilizzati da un contesto utente o
 da un'interruzione software. Il gestore d'interruzione non utilizza alcun
 *lock*, e tutti gli altri accessi verranno fatti così::
 
-        spin_lock(&lock);
+        mutex_lock(&lock);
         disable_irq(irq);
         ...
         enable_irq(irq);
-        spin_unlock(&lock);
+        mutex_unlock(&lock);
 
 La funzione disable_irq() impedisce al gestore d'interruzioni
 d'essere eseguito (e aspetta che finisca nel caso fosse in esecuzione su

MAINTAINERS

@@ -10777,6 +10777,8 @@ L:	linux-kernel@vger.kernel.org
 S:	Maintained
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git irq/core
 F:	kernel/irq/
+F:	include/linux/group_cpus.h
+F:	lib/group_cpus.c
 
 IRQCHIP DRIVERS
 M:	Thomas Gleixner <tglx@linutronix.de>

arch/x86/kernel/apic/io_apic.c

@@ -2364,9 +2364,8 @@ static int mp_irqdomain_create(int ioapic)
 		return -ENODEV;
 	}
 
-	ip->irqdomain = irq_domain_create_linear(fn, hwirqs, cfg->ops,
-						 (void *)(long)ioapic);
-
+	ip->irqdomain = irq_domain_create_hierarchy(parent, 0, hwirqs, fn, cfg->ops,
+						    (void *)(long)ioapic);
 	if (!ip->irqdomain) {
 		/* Release fw handle if it was allocated above */
 		if (!cfg->dev)
@@ -2374,8 +2373,6 @@ static int mp_irqdomain_create(int ioapic)
 		return -ENOMEM;
 	}
 
-	ip->irqdomain->parent = parent;
-
 	if (cfg->type == IOAPIC_DOMAIN_LEGACY ||
 	    cfg->type == IOAPIC_DOMAIN_STRICT)
 		ioapic_dynirq_base = max(ioapic_dynirq_base,

arch/x86/platform/uv/uv_irq.c

@@ -166,10 +166,9 @@ static struct irq_domain *uv_get_irq_domain(void)
 	if (!fn)
 		goto out;
 
-	uv_domain = irq_domain_create_tree(fn, &uv_domain_ops, NULL);
-	if (uv_domain)
-		uv_domain->parent = x86_vector_domain;
-	else
+	uv_domain = irq_domain_create_hierarchy(x86_vector_domain, 0, 0, fn,
+						&uv_domain_ops, NULL);
+	if (!uv_domain)
 		irq_domain_free_fwnode(fn);
 out:
 	mutex_unlock(&uv_lock);

block/blk-mq-cpumap.c

@@ -10,66 +10,29 @@
 #include <linux/mm.h>
 #include <linux/smp.h>
 #include <linux/cpu.h>
+#include <linux/group_cpus.h>
 
 #include <linux/blk-mq.h>
 #include "blk.h"
 #include "blk-mq.h"
 
-static int queue_index(struct blk_mq_queue_map *qmap,
-		       unsigned int nr_queues, const int q)
-{
-	return qmap->queue_offset + (q % nr_queues);
-}
-
-static int get_first_sibling(unsigned int cpu)
-{
-	unsigned int ret;
-
-	ret = cpumask_first(topology_sibling_cpumask(cpu));
-	if (ret < nr_cpu_ids)
-		return ret;
-
-	return cpu;
-}
-
 void blk_mq_map_queues(struct blk_mq_queue_map *qmap)
 {
-	unsigned int *map = qmap->mq_map;
-	unsigned int nr_queues = qmap->nr_queues;
-	unsigned int cpu, first_sibling, q = 0;
-
-	for_each_possible_cpu(cpu)
-		map[cpu] = -1;
-
-	/*
-	 * Spread queues among present CPUs first for minimizing
-	 * count of dead queues which are mapped by all un-present CPUs
-	 */
-	for_each_present_cpu(cpu) {
-		if (q >= nr_queues)
-			break;
-		map[cpu] = queue_index(qmap, nr_queues, q++);
+	const struct cpumask *masks;
+	unsigned int queue, cpu;
+
+	masks = group_cpus_evenly(qmap->nr_queues);
+	if (!masks) {
+		for_each_possible_cpu(cpu)
+			qmap->mq_map[cpu] = qmap->queue_offset;
+		return;
 	}
 
-	for_each_possible_cpu(cpu) {
-		if (map[cpu] != -1)
-			continue;
-		/*
-		 * First do sequential mapping between CPUs and queues.
-		 * In case we still have CPUs to map, and we have some number of
-		 * threads per cores then map sibling threads to the same queue
-		 * for performance optimizations.
-		 */
-		if (q < nr_queues) {
-			map[cpu] = queue_index(qmap, nr_queues, q++);
-		} else {
-			first_sibling = get_first_sibling(cpu);
-			if (first_sibling == cpu)
-				map[cpu] = queue_index(qmap, nr_queues, q++);
-			else
-				map[cpu] = map[first_sibling];
-		}
+	for (queue = 0; queue < qmap->nr_queues; queue++) {
+		for_each_cpu(cpu, &masks[queue])
+			qmap->mq_map[cpu] = qmap->queue_offset + queue;
 	}
+	kfree(masks);
 }
 EXPORT_SYMBOL_GPL(blk_mq_map_queues);
 

drivers/irqchip/Kconfig

@@ -384,7 +384,7 @@ config LS_EXTIRQ
 
 config LS_SCFG_MSI
 	def_bool y if SOC_LS1021A || ARCH_LAYERSCAPE
-	depends on PCI && PCI_MSI
+	depends on PCI_MSI
 
 config PARTITION_PERCPU
 	bool
@@ -653,6 +653,7 @@ config APPLE_AIC
 	bool "Apple Interrupt Controller (AIC)"
 	depends on ARM64
 	depends on ARCH_APPLE || COMPILE_TEST
+	select GENERIC_IRQ_IPI_MUX
 	help
 	  Support for the Apple Interrupt Controller found on Apple Silicon SoCs,
 	  such as the M1.

drivers/irqchip/irq-alpine-msi.c

@@ -199,21 +199,20 @@ static int alpine_msix_init_domains(struct alpine_msix_data *priv,
 	}
 
 	gic_domain = irq_find_host(gic_node);
+	of_node_put(gic_node);
 	if (!gic_domain) {
 		pr_err("Failed to find the GIC domain\n");
 		return -ENXIO;
 	}
 
-	middle_domain = irq_domain_add_tree(NULL,
-					    &alpine_msix_middle_domain_ops,
-					    priv);
+	middle_domain = irq_domain_add_hierarchy(gic_domain, 0, 0, NULL,
+						 &alpine_msix_middle_domain_ops,
+						 priv);
 	if (!middle_domain) {
 		pr_err("Failed to create the MSIX middle domain\n");
 		return -ENOMEM;
 	}
 
-	middle_domain->parent = gic_domain;
-
 	msi_domain = pci_msi_create_irq_domain(of_node_to_fwnode(node),
 					       &alpine_msix_domain_info,
 					       middle_domain);

drivers/irqchip/irq-apple-aic.c

@@ -292,7 +292,6 @@ struct aic_irq_chip {
 	void __iomem *base;
 	void __iomem *event;
 	struct irq_domain *hw_domain;
-	struct irq_domain *ipi_domain;
 	struct {
 		cpumask_t aff;
 	} *fiq_aff[AIC_NR_FIQ];
@@ -307,9 +306,6 @@ struct aic_irq_chip {
 
 static DEFINE_PER_CPU(uint32_t, aic_fiq_unmasked);
 
-static DEFINE_PER_CPU(atomic_t, aic_vipi_flag);
-static DEFINE_PER_CPU(atomic_t, aic_vipi_enable);
-
 static struct aic_irq_chip *aic_irqc;
 
 static void aic_handle_ipi(struct pt_regs *regs);
@@ -751,98 +747,8 @@ static void aic_ipi_send_fast(int cpu)
 	isb();
 }
 
-static void aic_ipi_mask(struct irq_data *d)
-{
-	u32 irq_bit = BIT(irqd_to_hwirq(d));
-
-	/* No specific ordering requirements needed here. */
-	atomic_andnot(irq_bit, this_cpu_ptr(&aic_vipi_enable));
-}
-
-static void aic_ipi_unmask(struct irq_data *d)
-{
-	struct aic_irq_chip *ic = irq_data_get_irq_chip_data(d);
-	u32 irq_bit = BIT(irqd_to_hwirq(d));
-
-	atomic_or(irq_bit, this_cpu_ptr(&aic_vipi_enable));
-
-	/*
-	 * The atomic_or() above must complete before the atomic_read()
-	 * below to avoid racing aic_ipi_send_mask().
-	 */
-	smp_mb__after_atomic();
-
-	/*
-	 * If a pending vIPI was unmasked, raise a HW IPI to ourselves.
-	 * No barriers needed here since this is a self-IPI.
-	 */
-	if (atomic_read(this_cpu_ptr(&aic_vipi_flag)) & irq_bit) {
-		if (static_branch_likely(&use_fast_ipi))
-			aic_ipi_send_fast(smp_processor_id());
-		else
-			aic_ic_write(ic, AIC_IPI_SEND, AIC_IPI_SEND_CPU(smp_processor_id()));
-	}
-}
-
-static void aic_ipi_send_mask(struct irq_data *d, const struct cpumask *mask)
-{
-	struct aic_irq_chip *ic = irq_data_get_irq_chip_data(d);
-	u32 irq_bit = BIT(irqd_to_hwirq(d));
-	u32 send = 0;
-	int cpu;
-	unsigned long pending;
-
-	for_each_cpu(cpu, mask) {
-		/*
-		 * This sequence is the mirror of the one in aic_ipi_unmask();
-		 * see the comment there. Additionally, release semantics
-		 * ensure that the vIPI flag set is ordered after any shared
-		 * memory accesses that precede it. This therefore also pairs
-		 * with the atomic_fetch_andnot in aic_handle_ipi().
-		 */
-		pending = atomic_fetch_or_release(irq_bit, per_cpu_ptr(&aic_vipi_flag, cpu));
-
-		/*
-		 * The atomic_fetch_or_release() above must complete before the
-		 * atomic_read() below to avoid racing aic_ipi_unmask().
-		 */
-		smp_mb__after_atomic();
-
-		if (!(pending & irq_bit) &&
-		    (atomic_read(per_cpu_ptr(&aic_vipi_enable, cpu)) & irq_bit)) {
-			if (static_branch_likely(&use_fast_ipi))
-				aic_ipi_send_fast(cpu);
-			else
-				send |= AIC_IPI_SEND_CPU(cpu);
-		}
-	}
-
-	/*
-	 * The flag writes must complete before the physical IPI is issued
-	 * to another CPU. This is implied by the control dependency on
-	 * the result of atomic_read_acquire() above, which is itself
-	 * already ordered after the vIPI flag write.
-	 */
-	if (send)
-		aic_ic_write(ic, AIC_IPI_SEND, send);
-}
-
-static struct irq_chip ipi_chip = {
-	.name = "AIC-IPI",
-	.irq_mask = aic_ipi_mask,
-	.irq_unmask = aic_ipi_unmask,
-	.ipi_send_mask = aic_ipi_send_mask,
-};
-
-/*
- * IPI IRQ domain
- */
-
 static void aic_handle_ipi(struct pt_regs *regs)
 {
-	int i;
-	unsigned long enabled, firing;
-
 	/*
 	 * Ack the IPI. We need to order this after the AIC event read, but
 	 * that is enforced by normal MMIO ordering guarantees.
@@ -857,27 +763,7 @@ static void aic_handle_ipi(struct pt_regs *regs)
 		aic_ic_write(aic_irqc, AIC_IPI_ACK, AIC_IPI_OTHER);
 	}
 
-	/*
-	 * The mask read does not need to be ordered. Only we can change
-	 * our own mask anyway, so no races are possible here, as long as
-	 * we are properly in the interrupt handler (which is covered by
-	 * the barrier that is part of the top-level AIC handler's readl()).
-	 */
-	enabled = atomic_read(this_cpu_ptr(&aic_vipi_enable));
-
-	/*
-	 * Clear the IPIs we are about to handle. This pairs with the
-	 * atomic_fetch_or_release() in aic_ipi_send_mask(), and needs to be
-	 * ordered after the aic_ic_write() above (to avoid dropping vIPIs) and
-	 * before IPI handling code (to avoid races handling vIPIs before they
-	 * are signaled). The former is taken care of by the release semantics
-	 * of the write portion, while the latter is taken care of by the
-	 * acquire semantics of the read portion.
-	 */
-	firing = atomic_fetch_andnot(enabled, this_cpu_ptr(&aic_vipi_flag)) & enabled;
-
-	for_each_set_bit(i, &firing, AIC_NR_SWIPI)
-		generic_handle_domain_irq(aic_irqc->ipi_domain, i);
+	ipi_mux_process();
 
 	/*
 	 * No ordering needed here; at worst this just changes the timing of
@@ -887,55 +773,24 @@ static void aic_handle_ipi(struct pt_regs *regs)
 		aic_ic_write(aic_irqc, AIC_IPI_MASK_CLR, AIC_IPI_OTHER);
 }
 
-static int aic_ipi_alloc(struct irq_domain *d, unsigned int virq,
-			 unsigned int nr_irqs, void *args)
+static void aic_ipi_send_single(unsigned int cpu)
 {
-	int i;
-
-	for (i = 0; i < nr_irqs; i++) {
-		irq_set_percpu_devid(virq + i);
-		irq_domain_set_info(d, virq + i, i, &ipi_chip, d->host_data,
-				    handle_percpu_devid_irq, NULL, NULL);
-	}
-
-	return 0;
-}
-
-static void aic_ipi_free(struct irq_domain *d, unsigned int virq, unsigned int nr_irqs)
-{
-	/* Not freeing IPIs */
+	if (static_branch_likely(&use_fast_ipi))
+		aic_ipi_send_fast(cpu);
+	else
+		aic_ic_write(aic_irqc, AIC_IPI_SEND, AIC_IPI_SEND_CPU(cpu));
 }
 
-static const struct irq_domain_ops aic_ipi_domain_ops = {
-	.alloc = aic_ipi_alloc,
-	.free = aic_ipi_free,
-};
-
 static int __init aic_init_smp(struct aic_irq_chip *irqc, struct device_node *node)
 {
-	struct irq_domain *ipi_domain;
 	int base_ipi;
 
-	ipi_domain = irq_domain_create_linear(irqc->hw_domain->fwnode, AIC_NR_SWIPI,
-					      &aic_ipi_domain_ops, irqc);
-	if (WARN_ON(!ipi_domain))
-		return -ENODEV;
-
-	ipi_domain->flags |= IRQ_DOMAIN_FLAG_IPI_SINGLE;
-	irq_domain_update_bus_token(ipi_domain, DOMAIN_BUS_IPI);
-
-	base_ipi = __irq_domain_alloc_irqs(ipi_domain, -1, AIC_NR_SWIPI,
-					   NUMA_NO_NODE, NULL, false, NULL);
-
-	if (WARN_ON(!base_ipi)) {
-		irq_domain_remove(ipi_domain);
+	base_ipi = ipi_mux_create(AIC_NR_SWIPI, aic_ipi_send_single);
+	if (WARN_ON(base_ipi <= 0))
 		return -ENODEV;
-	}
 
 	set_smp_ipi_range(base_ipi, AIC_NR_SWIPI);
 
-	irqc->ipi_domain = ipi_domain;
-
 	return 0;
 }
 

drivers/irqchip/irq-armada-370-xp.c

@@ -454,8 +454,7 @@ static __init void armada_xp_ipi_init(struct device_node *node)
 		return;
 
 	irq_domain_update_bus_token(ipi_domain, DOMAIN_BUS_IPI);
-	base_ipi = __irq_domain_alloc_irqs(ipi_domain, -1, IPI_DOORBELL_END,
-					   NUMA_NO_NODE, NULL, false, NULL);
+	base_ipi = irq_domain_alloc_irqs(ipi_domain, IPI_DOORBELL_END, NUMA_NO_NODE, NULL);
 	if (WARN_ON(!base_ipi))
 		return;
 

drivers/irqchip/irq-aspeed-scu-ic.c

@@ -17,8 +17,9 @@
 
 #define ASPEED_SCU_IC_REG		0x018
 #define ASPEED_SCU_IC_SHIFT		0
-#define ASPEED_SCU_IC_ENABLE		GENMASK(6, ASPEED_SCU_IC_SHIFT)
+#define ASPEED_SCU_IC_ENABLE		GENMASK(15, ASPEED_SCU_IC_SHIFT)
 #define ASPEED_SCU_IC_NUM_IRQS		7
+#define ASPEED_SCU_IC_STATUS		GENMASK(28, 16)
 #define ASPEED_SCU_IC_STATUS_SHIFT	16
 
 #define ASPEED_AST2600_SCU_IC0_REG	0x560
@@ -155,6 +156,8 @@ static int aspeed_scu_ic_of_init_common(struct aspeed_scu_ic *scu_ic,
 		rc = PTR_ERR(scu_ic->scu);
 		goto err;
 	}
+	regmap_write_bits(scu_ic->scu, scu_ic->reg, ASPEED_SCU_IC_STATUS, ASPEED_SCU_IC_STATUS);
+	regmap_write_bits(scu_ic->scu, scu_ic->reg, ASPEED_SCU_IC_ENABLE, 0);
 
 	irq = irq_of_parse_and_map(node, 0);
 	if (!irq) {

drivers/irqchip/irq-bcm2836.c

@@ -268,10 +268,7 @@ static void __init bcm2836_arm_irqchip_smp_init(void)
 	ipi_domain->flags |= IRQ_DOMAIN_FLAG_IPI_SINGLE;
 	irq_domain_update_bus_token(ipi_domain, DOMAIN_BUS_IPI);
 
-	base_ipi = __irq_domain_alloc_irqs(ipi_domain, -1, BITS_PER_MBOX,
-					   NUMA_NO_NODE, NULL,
-					   false, NULL);
-
+	base_ipi = irq_domain_alloc_irqs(ipi_domain, BITS_PER_MBOX, NUMA_NO_NODE, NULL);
 	if (WARN_ON(!base_ipi))
 		return;
 

drivers/irqchip/irq-bcm7120-l2.c

@@ -279,7 +279,8 @@ static int __init bcm7120_l2_intc_probe(struct device_node *dn,
 		flags |= IRQ_GC_BE_IO;
 
 	ret = irq_alloc_domain_generic_chips(data->domain, IRQS_PER_WORD, 1,
-				dn->full_name, handle_level_irq, clr, 0, flags);
+				dn->full_name, handle_level_irq, clr,
+				IRQ_LEVEL, flags);
 	if (ret) {
 		pr_err("failed to allocate generic irq chip\n");
 		goto out_free_domain;

drivers/irqchip/irq-brcmstb-l2.c

@@ -161,6 +161,7 @@ static int __init brcmstb_l2_intc_of_init(struct device_node *np,
 					  *init_params)
 {
 	unsigned int clr = IRQ_NOREQUEST | IRQ_NOPROBE | IRQ_NOAUTOEN;
+	unsigned int set = 0;
 	struct brcmstb_l2_intc_data *data;
 	struct irq_chip_type *ct;
 	int ret;
@@ -208,9 +209,12 @@ static int __init brcmstb_l2_intc_of_init(struct device_node *np,
 	if (IS_ENABLED(CONFIG_MIPS) && IS_ENABLED(CONFIG_CPU_BIG_ENDIAN))
 		flags |= IRQ_GC_BE_IO;
 
+	if (init_params->handler == handle_level_irq)
+		set |= IRQ_LEVEL;
+
 	/* Allocate a single Generic IRQ chip for this node */
 	ret = irq_alloc_domain_generic_chips(data->domain, 32, 1,
-			np->full_name, init_params->handler, clr, 0, flags);
+			np->full_name, init_params->handler, clr, set, flags);
 	if (ret) {
 		pr_err("failed to allocate generic irq chip\n");
 		goto out_free_domain;

drivers/irqchip/irq-gic-v2m.c

@@ -287,15 +287,14 @@ static __init int gicv2m_allocate_domains(struct irq_domain *parent)
 	if (!v2m)
 		return 0;
 
-	inner_domain = irq_domain_create_tree(v2m->fwnode,
-					      &gicv2m_domain_ops, v2m);
+	inner_domain = irq_domain_create_hierarchy(parent, 0, 0, v2m->fwnode,
+						   &gicv2m_domain_ops, v2m);
 	if (!inner_domain) {
 		pr_err("Failed to create GICv2m domain\n");
 		return -ENOMEM;
 	}
 
 	irq_domain_update_bus_token(inner_domain, DOMAIN_BUS_NEXUS);
-	inner_domain->parent = parent;
 	pci_domain = pci_msi_create_irq_domain(v2m->fwnode,
 					       &gicv2m_msi_domain_info,
 					       inner_domain);

drivers/irqchip/irq-gic-v3-its.c

@@ -4909,18 +4909,19 @@ static int its_init_domain(struct fwnode_handle *handle, struct its_node *its)
 	if (!info)
 		return -ENOMEM;
 
-	inner_domain = irq_domain_create_tree(handle, &its_domain_ops, its);
+	info->ops = &its_msi_domain_ops;
+	info->data = its;
+
+	inner_domain = irq_domain_create_hierarchy(its_parent,
+						   its->msi_domain_flags, 0,
+						   handle, &its_domain_ops,
+						   info);
 	if (!inner_domain) {
 		kfree(info);
 		return -ENOMEM;
 	}
 
-	inner_domain->parent = its_parent;
 	irq_domain_update_bus_token(inner_domain, DOMAIN_BUS_NEXUS);
-	inner_domain->flags |= its->msi_domain_flags;
-	info->ops = &its_msi_domain_ops;
-	info->data = its;
-	inner_domain->host_data = info;
 
 	return 0;
 }

drivers/irqchip/irq-gic-v3-mbi.c

@@ -233,13 +233,12 @@ static int mbi_allocate_domains(struct irq_domain *parent)
 	struct irq_domain *nexus_domain, *pci_domain, *plat_domain;
 	int err;
 
-	nexus_domain = irq_domain_create_tree(parent->fwnode,
-					      &mbi_domain_ops, NULL);
+	nexus_domain = irq_domain_create_hierarchy(parent, 0, 0, parent->fwnode,
+						   &mbi_domain_ops, NULL);
 	if (!nexus_domain)
 		return -ENOMEM;
 
 	irq_domain_update_bus_token(nexus_domain, DOMAIN_BUS_NEXUS);
-	nexus_domain->parent = parent;
 
 	err = mbi_allocate_pci_domain(nexus_domain, &pci_domain);
 

drivers/irqchip/irq-gic-v3.c

@@ -1310,9 +1310,7 @@ static void __init gic_smp_init(void)
 				  gic_starting_cpu, NULL);
 
 	/* Register all 8 non-secure SGIs */
-	base_sgi = __irq_domain_alloc_irqs(gic_data.domain, -1, 8,
-					   NUMA_NO_NODE, &sgi_fwspec,
-					   false, NULL);
+	base_sgi = irq_domain_alloc_irqs(gic_data.domain, 8, NUMA_NO_NODE, &sgi_fwspec);
 	if (WARN_ON(base_sgi <= 0))
 		return;
 

drivers/irqchip/irq-gic-v4.c

@@ -139,9 +139,7 @@ static int its_alloc_vcpu_sgis(struct its_vpe *vpe, int idx)
 	if (!vpe->sgi_domain)
 		goto err;
 
-	sgi_base = __irq_domain_alloc_irqs(vpe->sgi_domain, -1, 16,
-					       NUMA_NO_NODE, vpe,
-					       false, NULL);
+	sgi_base = irq_domain_alloc_irqs(vpe->sgi_domain, 16, NUMA_NO_NODE, vpe);
 	if (sgi_base <= 0)
 		goto err;
 
@@ -176,9 +174,8 @@ int its_alloc_vcpu_irqs(struct its_vm *vm)
 		vm->vpes[i]->idai = true;
 	}
 
-	vpe_base_irq = __irq_domain_alloc_irqs(vm->domain, -1, vm->nr_vpes,
-					       NUMA_NO_NODE, vm,
-					       false, NULL);
+	vpe_base_irq = irq_domain_alloc_irqs(vm->domain, vm->nr_vpes,
+					     NUMA_NO_NODE, vm);
 	if (vpe_base_irq <= 0)
 		goto err;
 

drivers/irqchip/irq-gic.c

@@ -868,9 +868,7 @@ static __init void gic_smp_init(void)
 				  "irqchip/arm/gic:starting",
 				  gic_starting_cpu, NULL);
 
-	base_sgi = __irq_domain_alloc_irqs(gic_data[0].domain, -1, 8,
-					   NUMA_NO_NODE, &sgi_fwspec,
-					   false, NULL);
+	base_sgi = irq_domain_alloc_irqs(gic_data[0].domain, 8, NUMA_NO_NODE, &sgi_fwspec);
 	if (WARN_ON(base_sgi <= 0))
 		return;
 

drivers/irqchip/irq-loongson-liointc.c

@@ -55,6 +55,8 @@ struct liointc_priv {
 	struct liointc_handler_data	handler[LIOINTC_NUM_PARENT];
 	void __iomem			*core_isr[LIOINTC_NUM_CORES];
 	u8				map_cache[LIOINTC_CHIP_IRQ];
+	u32				int_pol;
+	u32				int_edge;
 	bool				has_lpc_irq_errata;
 };
 
@@ -138,6 +140,14 @@ static int liointc_set_type(struct irq_data *data, unsigned int type)
 	return 0;
 }
 
+static void liointc_suspend(struct irq_chip_generic *gc)
+{
+	struct liointc_priv *priv = gc->private;
+
+	priv->int_pol = readl(gc->reg_base + LIOINTC_REG_INTC_POL);
+	priv->int_edge = readl(gc->reg_base + LIOINTC_REG_INTC_EDGE);
+}
+
 static void liointc_resume(struct irq_chip_generic *gc)
 {
 	struct liointc_priv *priv = gc->private;
@@ -150,6 +160,8 @@ static void liointc_resume(struct irq_chip_generic *gc)
 	/* Restore map cache */
 	for (i = 0; i < LIOINTC_CHIP_IRQ; i++)
 		writeb(priv->map_cache[i], gc->reg_base + i);
+	writel(priv->int_pol, gc->reg_base + LIOINTC_REG_INTC_POL);
+	writel(priv->int_edge, gc->reg_base + LIOINTC_REG_INTC_EDGE);
 	/* Restore mask cache */
 	writel(gc->mask_cache, gc->reg_base + LIOINTC_REG_INTC_ENABLE);
 	irq_gc_unlock_irqrestore(gc, flags);
@@ -269,6 +281,7 @@ static int liointc_init(phys_addr_t addr, unsigned long size, int revision,
 	gc->private = priv;
 	gc->reg_base = base;
 	gc->domain = domain;
+	gc->suspend = liointc_suspend;
 	gc->resume = liointc_resume;
 
 	ct = gc->chip_types;

drivers/irqchip/irq-loongson-pch-msi.c

@@ -163,16 +163,15 @@ static int pch_msi_init_domains(struct pch_msi_data *priv,
 {
 	struct irq_domain *middle_domain, *msi_domain;
 
-	middle_domain = irq_domain_create_linear(domain_handle,
-						priv->num_irqs,
-						&pch_msi_middle_domain_ops,
-						priv);
+	middle_domain = irq_domain_create_hierarchy(parent, 0, priv->num_irqs,
+						    domain_handle,
+						    &pch_msi_middle_domain_ops,
+						    priv);
 	if (!middle_domain) {
 		pr_err("Failed to create the MSI middle domain\n");
 		return -ENOMEM;
 	}
 
-	middle_domain->parent = parent;
 	irq_domain_update_bus_token(middle_domain, DOMAIN_BUS_NEXUS);
 
 	msi_domain = pci_msi_create_irq_domain(domain_handle,

drivers/irqchip/irq-mvebu-gicp.c

@@ -221,6 +221,7 @@ static int mvebu_gicp_probe(struct platform_device *pdev)
 	}
 
 	parent_domain = irq_find_host(irq_parent_dn);
+	of_node_put(irq_parent_dn);
 	if (!parent_domain) {
 		dev_err(&pdev->dev, "failed to find parent IRQ domain\n");
 		return -ENODEV;

drivers/irqchip/irq-mvebu-odmi.c

@@ -161,7 +161,7 @@ static struct msi_domain_info odmi_msi_domain_info = {
 static int __init mvebu_odmi_init(struct device_node *node,
 				  struct device_node *parent)
 {
-	struct irq_domain *inner_domain, *plat_domain;
+	struct irq_domain *parent_domain, *inner_domain, *plat_domain;
 	int ret, i;
 
 	if (of_property_read_u32(node, "marvell,odmi-frames", &odmis_count))
@@ -197,16 +197,17 @@ static int __init mvebu_odmi_init(struct device_node *node,
 		}
 	}
 
-	inner_domain = irq_domain_create_linear(of_node_to_fwnode(node),
-						odmis_count * NODMIS_PER_FRAME,
-						&odmi_domain_ops, NULL);
+	parent_domain = irq_find_host(parent);
+
+	inner_domain = irq_domain_create_hierarchy(parent_domain, 0,
+						   odmis_count * NODMIS_PER_FRAME,
+						   of_node_to_fwnode(node),
+						   &odmi_domain_ops, NULL);
 	if (!inner_domain) {
 		ret = -ENOMEM;
 		goto err_unmap;
 	}
 
-	inner_domain->parent = irq_find_host(parent);
-
 	plat_domain = platform_msi_create_irq_domain(of_node_to_fwnode(node),
 						     &odmi_msi_domain_info,
 						     inner_domain);

drivers/irqchip/irq-ti-sci-intr.c

@@ -236,6 +236,7 @@ static int ti_sci_intr_irq_domain_probe(struct platform_device *pdev)
 	}
 
 	parent_domain = irq_find_host(parent_node);
+	of_node_put(parent_node);
 	if (!parent_domain) {
 		dev_err(dev, "Failed to find IRQ parent domain\n");
 		return -ENODEV;

drivers/irqchip/irqchip.c

@@ -38,8 +38,10 @@ int platform_irqchip_probe(struct platform_device *pdev)
 	struct device_node *par_np = of_irq_find_parent(np);
 	of_irq_init_cb_t irq_init_cb = of_device_get_match_data(&pdev->dev);
 
-	if (!irq_init_cb)
+	if (!irq_init_cb) {
+		of_node_put(par_np);
 		return -EINVAL;
+	}
 
 	if (par_np == np)
 		par_np = NULL;
@@ -52,8 +54,10 @@ int platform_irqchip_probe(struct platform_device *pdev)
 	 * interrupt controller. The actual initialization callback of this
 	 * interrupt controller can check for specific domains as necessary.
 	 */
-	if (par_np && !irq_find_matching_host(par_np, DOMAIN_BUS_ANY))
+	if (par_np && !irq_find_matching_host(par_np, DOMAIN_BUS_ANY)) {
+		of_node_put(par_np);
 		return -EPROBE_DEFER;
+	}
 
 	return irq_init_cb(np, par_np);
 }

include/linux/group_cpus.h

+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (C) 2016 Thomas Gleixner.
+ * Copyright (C) 2016-2017 Christoph Hellwig.
+ */
+
+#ifndef __LINUX_GROUP_CPUS_H
+#define __LINUX_GROUP_CPUS_H
+#include <linux/kernel.h>
+#include <linux/cpu.h>
+
+struct cpumask *group_cpus_evenly(unsigned int numgrps);
+
+#endif

include/linux/irq.h

@@ -1266,6 +1266,9 @@ int __ipi_send_mask(struct irq_desc *desc, const struct cpumask *dest);
 int ipi_send_single(unsigned int virq, unsigned int cpu);
 int ipi_send_mask(unsigned int virq, const struct cpumask *dest);
 
+void ipi_mux_process(void);
+int ipi_mux_create(unsigned int nr_ipi, void (*mux_send)(unsigned int cpu));
+
 #ifdef CONFIG_GENERIC_IRQ_MULTI_HANDLER
 /*
  * Registers a generic IRQ handling function as the top-level IRQ handler in

include/linux/irqdomain.h

@@ -125,6 +125,8 @@ struct irq_domain_chip_generic;
  *		core code.
  * @flags:	Per irq_domain flags
  * @mapcount:	The number of mapped interrupts
+ * @mutex:	Domain lock, hierarchical domains use root domain's lock
+ * @root:	Pointer to root domain, or containing structure if non-hierarchical
  *
  * Optional elements:
  * @fwnode:	Pointer to firmware node associated with the irq_domain. Pretty easy
@@ -143,7 +145,6 @@ struct irq_domain_chip_generic;
  * Revmap data, used internally by the irq domain code:
  * @revmap_size:	Size of the linear map table @revmap[]
  * @revmap_tree:	Radix map tree for hwirqs that don't fit in the linear map
- * @revmap_mutex:	Lock for the revmap
  * @revmap:		Linear table of irq_data pointers
  */
 struct irq_domain {
@@ -153,6 +154,8 @@ struct irq_domain {
 	void				*host_data;
 	unsigned int			flags;
 	unsigned int			mapcount;
+	struct mutex			mutex;
+	struct irq_domain		*root;
 
 	/* Optional data */
 	struct fwnode_handle		*fwnode;
@@ -171,7 +174,6 @@ struct irq_domain {
 	irq_hw_number_t			hwirq_max;
 	unsigned int			revmap_size;
 	struct radix_tree_root		revmap_tree;
-	struct mutex			revmap_mutex;
 	struct irq_data __rcu		*revmap[];
 };
 

kernel/irq/Kconfig

@@ -86,6 +86,11 @@ config GENERIC_IRQ_IPI
 	depends on SMP
 	select IRQ_DOMAIN_HIERARCHY
 
+# Generic IRQ IPI Mux support
+config GENERIC_IRQ_IPI_MUX
+	bool
+	depends on SMP
+
 # Generic MSI hierarchical interrupt domain support
 config GENERIC_MSI_IRQ
 	bool

kernel/irq/Makefile

@@ -15,6 +15,7 @@ obj-$(CONFIG_GENERIC_IRQ_MIGRATION) += cpuhotplug.o
 obj-$(CONFIG_PM_SLEEP) += pm.o
 obj-$(CONFIG_GENERIC_MSI_IRQ) += msi.o
 obj-$(CONFIG_GENERIC_IRQ_IPI) += ipi.o
+obj-$(CONFIG_GENERIC_IRQ_IPI_MUX) += ipi-mux.o
 obj-$(CONFIG_SMP) += affinity.o
 obj-$(CONFIG_GENERIC_IRQ_DEBUGFS) += debugfs.o
 obj-$(CONFIG_GENERIC_IRQ_MATRIX_ALLOCATOR) += matrix.o

kernel/irq/ipi-mux.c

+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Multiplex several virtual IPIs over a single HW IPI.
+ *
+ * Copyright The Asahi Linux Contributors
+ * Copyright (c) 2022 Ventana Micro Systems Inc.
+ */
+
+#define pr_fmt(fmt) "ipi-mux: " fmt
+#include <linux/cpu.h>
+#include <linux/init.h>
+#include <linux/irq.h>
+#include <linux/irqchip.h>
+#include <linux/irqchip/chained_irq.h>
+#include <linux/irqdomain.h>
+#include <linux/jump_label.h>
+#include <linux/percpu.h>
+#include <linux/smp.h>
+
+struct ipi_mux_cpu {
+	atomic_t			enable;
+	atomic_t			bits;
+};
+
+static struct ipi_mux_cpu __percpu *ipi_mux_pcpu;
+static struct irq_domain *ipi_mux_domain;
+static void (*ipi_mux_send)(unsigned int cpu);
+
+static void ipi_mux_mask(struct irq_data *d)
+{
+	struct ipi_mux_cpu *icpu = this_cpu_ptr(ipi_mux_pcpu);
+
+	atomic_andnot(BIT(irqd_to_hwirq(d)), &icpu->enable);
+}
+
+static void ipi_mux_unmask(struct irq_data *d)
+{
+	struct ipi_mux_cpu *icpu = this_cpu_ptr(ipi_mux_pcpu);
+	u32 ibit = BIT(irqd_to_hwirq(d));
+
+	atomic_or(ibit, &icpu->enable);
+
+	/*
+	 * The atomic_or() above must complete before the atomic_read()
+	 * below to avoid racing ipi_mux_send_mask().
+	 */
+	smp_mb__after_atomic();
+
+	/* If a pending IPI was unmasked, raise a parent IPI immediately. */
+	if (atomic_read(&icpu->bits) & ibit)
+		ipi_mux_send(smp_processor_id());
+}
+
+static void ipi_mux_send_mask(struct irq_data *d, const struct cpumask *mask)
+{
+	struct ipi_mux_cpu *icpu = this_cpu_ptr(ipi_mux_pcpu);
+	u32 ibit = BIT(irqd_to_hwirq(d));
+	unsigned long pending;
+	int cpu;
+
+	for_each_cpu(cpu, mask) {
+		icpu = per_cpu_ptr(ipi_mux_pcpu, cpu);
+
+		/*
+		 * This sequence is the mirror of the one in ipi_mux_unmask();
+		 * see the comment there. Additionally, release semantics
+		 * ensure that the vIPI flag set is ordered after any shared
+		 * memory accesses that precede it. This therefore also pairs
+		 * with the atomic_fetch_andnot in ipi_mux_process().
+		 */
+		pending = atomic_fetch_or_release(ibit, &icpu->bits);
+
+		/*
+		 * The atomic_fetch_or_release() above must complete
+		 * before the atomic_read() below to avoid racing with
+		 * ipi_mux_unmask().
+		 */
+		smp_mb__after_atomic();
+
+		/*
+		 * The flag writes must complete before the physical IPI is
+		 * issued to another CPU. This is implied by the control
+		 * dependency on the result of atomic_read() below, which is
+		 * itself already ordered after the vIPI flag write.
+		 */
+		if (!(pending & ibit) && (atomic_read(&icpu->enable) & ibit))
+			ipi_mux_send(cpu);
+	}
+}
+
+static const struct irq_chip ipi_mux_chip = {
+	.name		= "IPI Mux",
+	.irq_mask	= ipi_mux_mask,
+	.irq_unmask	= ipi_mux_unmask,
+	.ipi_send_mask	= ipi_mux_send_mask,
+};
+
+static int ipi_mux_domain_alloc(struct irq_domain *d, unsigned int virq,
+				unsigned int nr_irqs, void *arg)
+{
+	int i;
+
+	for (i = 0; i < nr_irqs; i++) {
+		irq_set_percpu_devid(virq + i);
+		irq_domain_set_info(d, virq + i, i, &ipi_mux_chip, NULL,
+				    handle_percpu_devid_irq, NULL, NULL);
+	}
+
+	return 0;
+}
+
+static const struct irq_domain_ops ipi_mux_domain_ops = {
+	.alloc		= ipi_mux_domain_alloc,
+	.free		= irq_domain_free_irqs_top,
+};
+
+/**
+ * ipi_mux_process - Process multiplexed virtual IPIs
+ */
+void ipi_mux_process(void)
+{
+	struct ipi_mux_cpu *icpu = this_cpu_ptr(ipi_mux_pcpu);
+	irq_hw_number_t hwirq;
+	unsigned long ipis;
+	unsigned int en;
+
+	/*
+	 * Reading enable mask does not need to be ordered as long as
+	 * this function is called from interrupt handler because only
+	 * the CPU itself can change it's own enable mask.
+	 */
+	en = atomic_read(&icpu->enable);
+
+	/*
+	 * Clear the IPIs we are about to handle. This pairs with the
+	 * atomic_fetch_or_release() in ipi_mux_send_mask().
+	 */
+	ipis = atomic_fetch_andnot(en, &icpu->bits) & en;
+
+	for_each_set_bit(hwirq, &ipis, BITS_PER_TYPE(int))
+		generic_handle_domain_irq(ipi_mux_domain, hwirq);
+}
+
+/**
+ * ipi_mux_create - Create virtual IPIs multiplexed on top of a single
+ * parent IPI.
+ * @nr_ipi:		number of virtual IPIs to create. This should
+ *			be <= BITS_PER_TYPE(int)
+ * @mux_send:		callback to trigger parent IPI for a particular CPU
+ *
+ * Returns first virq of the newly created virtual IPIs upon success
+ * or <=0 upon failure
+ */
+int ipi_mux_create(unsigned int nr_ipi, void (*mux_send)(unsigned int cpu))
+{
+	struct fwnode_handle *fwnode;
+	struct irq_domain *domain;
+	int rc;
+
+	if (ipi_mux_domain)
+		return -EEXIST;
+
+	if (BITS_PER_TYPE(int) < nr_ipi || !mux_send)
+		return -EINVAL;
+
+	ipi_mux_pcpu = alloc_percpu(typeof(*ipi_mux_pcpu));
+	if (!ipi_mux_pcpu)
+		return -ENOMEM;
+
+	fwnode = irq_domain_alloc_named_fwnode("IPI-Mux");
+	if (!fwnode) {
+		pr_err("unable to create IPI Mux fwnode\n");
+		rc = -ENOMEM;
+		goto fail_free_cpu;
+	}
+
+	domain = irq_domain_create_linear(fwnode, nr_ipi,
+					  &ipi_mux_domain_ops, NULL);
+	if (!domain) {
+		pr_err("unable to add IPI Mux domain\n");
+		rc = -ENOMEM;
+		goto fail_free_fwnode;
+	}
+
+	domain->flags |= IRQ_DOMAIN_FLAG_IPI_SINGLE;
+	irq_domain_update_bus_token(domain, DOMAIN_BUS_IPI);
+
+	rc = irq_domain_alloc_irqs(domain, nr_ipi, NUMA_NO_NODE, NULL);
+	if (rc <= 0) {
+		pr_err("unable to alloc IRQs from IPI Mux domain\n");
+		goto fail_free_domain;
+	}
+
+	ipi_mux_domain = domain;
+	ipi_mux_send = mux_send;
+
+	return rc;
+
+fail_free_domain:
+	irq_domain_remove(domain);
+fail_free_fwnode:
+	irq_domain_free_fwnode(fwnode);
+fail_free_cpu:
+	free_percpu(ipi_mux_pcpu);
+	return rc;
+}

kernel/irq/manage.c

@@ -723,10 +723,13 @@ EXPORT_SYMBOL(disable_irq_nosync);
  *	to complete before returning. If you use this function while
  *	holding a resource the IRQ handler may need you will deadlock.
  *
- *	This function may be called - with care - from IRQ context.
+ *	Can only be called from preemptible code as it might sleep when
+ *	an interrupt thread is associated to @irq.
+ *
  */
 void disable_irq(unsigned int irq)
 {
+	might_sleep();
 	if (!__disable_irq_nosync(irq))
 		synchronize_irq(irq);
 }

lib/Makefile

@@ -353,6 +353,8 @@ obj-$(CONFIG_SBITMAP) += sbitmap.o
 
 obj-$(CONFIG_PARMAN) += parman.o
 
+obj-y += group_cpus.o
+
 # GCC library routines
 obj-$(CONFIG_GENERIC_LIB_ASHLDI3) += ashldi3.o
 obj-$(CONFIG_GENERIC_LIB_ASHRDI3) += ashrdi3.o