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
  ...

Author: torvalds

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:	[email protected]
 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 <[email protected]>

arch/x86/kernel/apic/io_apic.c

@@ -2364,18 +2364,15 @@ 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)
 			irq_domain_free_fwnode(fn);
 		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);