Merge branch 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull irq updates from Thomas Gleixner:
 "The irq department delivers:

   - Expand the generic infrastructure handling the irq migration on CPU
     hotplug and convert X86 over to it. (Thomas Gleixner)

     Aside of consolidating code this is a preparatory change for:

   - Finalizing the affinity management for multi-queue devices. The
     main change here is to shut down interrupts which are affine to a
     outgoing CPU and reenabling them when the CPU comes online again.
     That avoids moving interrupts pointlessly around and breaking and
     reestablishing affinities for no value. (Christoph Hellwig)

     Note: This contains also the BLOCK-MQ and NVME changes which depend
     on the rework of the irq core infrastructure. Jens acked them and
     agreed that they should go with the irq changes.

   - Consolidation of irq domain code (Marc Zyngier)

   - State tracking consolidation in the core code (Jeffy Chen)

   - Add debug infrastructure for hierarchical irq domains (Thomas
     Gleixner)

   - Infrastructure enhancement for managing generic interrupt chips via
     devmem (Bartosz Golaszewski)

   - Constification work all over the place (Tobias Klauser)

   - Two new interrupt controller drivers for MVEBU (Thomas Petazzoni)

   - The usual set of fixes, updates and enhancements all over the
     place"

* 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (112 commits)
  irqchip/or1k-pic: Fix interrupt acknowledgement
  irqchip/irq-mvebu-gicp: Allocate enough memory for spi_bitmap
  irqchip/gic-v3: Fix out-of-bound access in gic_set_affinity
  nvme: Allocate queues for all possible CPUs
  blk-mq: Create hctx for each present CPU
  blk-mq: Include all present CPUs in the default queue mapping
  genirq: Avoid unnecessary low level irq function calls
  genirq: Set irq masked state when initializing irq_desc
  genirq/timings: Add infrastructure for estimating the next interrupt arrival time
  genirq/timings: Add infrastructure to track the interrupt timings
  genirq/debugfs: Remove pointless NULL pointer check
  irqchip/gic-v3-its: Don't assume GICv3 hardware supports 16bit INTID
  irqchip/gic-v3-its: Add ACPI NUMA node mapping
  irqchip/gic-v3-its-platform-msi: Make of_device_ids const
  irqchip/gic-v3-its: Make of_device_ids const
  irqchip/irq-mvebu-icu: Add new driver for Marvell ICU
  irqchip/irq-mvebu-gicp: Add new driver for Marvell GICP
  dt-bindings/interrupt-controller: Add DT binding for the Marvell ICU
  genirq/irqdomain: Remove auto-recursive hierarchy support
  irqchip/MSI: Use irq_domain_update_bus_token instead of an open coded access
  ...

1 files changed, 130 insertions, 18 deletions

diff --git a/drivers/irqchip/irq-armada-370-xp.c b/drivers/irqchip/irq-armada-370-xp.c
index eb0d4d41b156..b207b2c3aa55 100644
--- a/drivers/irqchip/irq-armada-370-xp.c
+++ b/drivers/irqchip/irq-armada-370-xp.c
@@ -34,25 +34,104 @@
 #include <asm/smp_plat.h>
 #include <asm/mach/irq.h>
 
-/* Interrupt Controller Registers Map */
-#define ARMADA_370_XP_INT_SET_MASK_OFFS		(0x48)
-#define ARMADA_370_XP_INT_CLEAR_MASK_OFFS	(0x4C)
-#define ARMADA_370_XP_INT_FABRIC_MASK_OFFS	(0x54)
-#define ARMADA_370_XP_INT_CAUSE_PERF(cpu)	(1 << cpu)
+/*
+ * Overall diagram of the Armada XP interrupt controller:
+ *
+ *    To CPU 0                 To CPU 1
+ *
+ *       /\                       /\
+ *       ||                       ||
+ * +---------------+     +---------------+
+ * |               |	 |               |
+ * |    per-CPU    |	 |    per-CPU    |
+ * |  mask/unmask  |	 |  mask/unmask  |
+ * |     CPU0      |	 |     CPU1      |
+ * |               |	 |               |
+ * +---------------+	 +---------------+
+ *        /\                       /\
+ *        ||                       ||
+ *        \\_______________________//
+ *                     ||
+ *            +-------------------+
+ *            |                   |
+ *            | Global interrupt  |
+ *            |    mask/unmask    |
+ *            |                   |
+ *            +-------------------+
+ *                     /\
+ *                     ||
+ *               interrupt from
+ *                   device
+ *
+ * The "global interrupt mask/unmask" is modified using the
+ * ARMADA_370_XP_INT_SET_ENABLE_OFFS and
+ * ARMADA_370_XP_INT_CLEAR_ENABLE_OFFS registers, which are relative
+ * to "main_int_base".
+ *
+ * The "per-CPU mask/unmask" is modified using the
+ * ARMADA_370_XP_INT_SET_MASK_OFFS and
+ * ARMADA_370_XP_INT_CLEAR_MASK_OFFS registers, which are relative to
+ * "per_cpu_int_base". This base address points to a special address,
+ * which automatically accesses the registers of the current CPU.
+ *
+ * The per-CPU mask/unmask can also be adjusted using the global
+ * per-interrupt ARMADA_370_XP_INT_SOURCE_CTL register, which we use
+ * to configure interrupt affinity.
+ *
+ * Due to this model, all interrupts need to be mask/unmasked at two
+ * different levels: at the global level and at the per-CPU level.
+ *
+ * This driver takes the following approach to deal with this:
+ *
+ *  - For global interrupts:
+ *
+ *    At ->map() time, a global interrupt is unmasked at the per-CPU
+ *    mask/unmask level. It is therefore unmasked at this level for
+ *    the current CPU, running the ->map() code. This allows to have
+ *    the interrupt unmasked at this level in non-SMP
+ *    configurations. In SMP configurations, the ->set_affinity()
+ *    callback is called, which using the
+ *    ARMADA_370_XP_INT_SOURCE_CTL() readjusts the per-CPU mask/unmask
+ *    for the interrupt.
+ *
+ *    The ->mask() and ->unmask() operations only mask/unmask the
+ *    interrupt at the "global" level.
+ *
+ *    So, a global interrupt is enabled at the per-CPU level as soon
+ *    as it is mapped. At run time, the masking/unmasking takes place
+ *    at the global level.
+ *
+ *  - For per-CPU interrupts
+ *
+ *    At ->map() time, a per-CPU interrupt is unmasked at the global
+ *    mask/unmask level.
+ *
+ *    The ->mask() and ->unmask() operations mask/unmask the interrupt
+ *    at the per-CPU level.
+ *
+ *    So, a per-CPU interrupt is enabled at the global level as soon
+ *    as it is mapped. At run time, the masking/unmasking takes place
+ *    at the per-CPU level.
+ */
 
+/* Registers relative to main_int_base */
 #define ARMADA_370_XP_INT_CONTROL		(0x00)
+#define ARMADA_370_XP_SW_TRIG_INT_OFFS		(0x04)
 #define ARMADA_370_XP_INT_SET_ENABLE_OFFS	(0x30)
 #define ARMADA_370_XP_INT_CLEAR_ENABLE_OFFS	(0x34)
 #define ARMADA_370_XP_INT_SOURCE_CTL(irq)	(0x100 + irq*4)
 #define ARMADA_370_XP_INT_SOURCE_CPU_MASK	0xF
 #define ARMADA_370_XP_INT_IRQ_FIQ_MASK(cpuid)	((BIT(0) | BIT(8)) << cpuid)
 
-#define ARMADA_370_XP_CPU_INTACK_OFFS		(0x44)
+/* Registers relative to per_cpu_int_base */
+#define ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS	(0x08)
+#define ARMADA_370_XP_IN_DRBEL_MSK_OFFS		(0x0c)
 #define ARMADA_375_PPI_CAUSE			(0x10)
-
-#define ARMADA_370_XP_SW_TRIG_INT_OFFS           (0x4)
-#define ARMADA_370_XP_IN_DRBEL_MSK_OFFS          (0xc)
-#define ARMADA_370_XP_IN_DRBEL_CAUSE_OFFS        (0x8)
+#define ARMADA_370_XP_CPU_INTACK_OFFS		(0x44)
+#define ARMADA_370_XP_INT_SET_MASK_OFFS		(0x48)
+#define ARMADA_370_XP_INT_CLEAR_MASK_OFFS	(0x4C)
+#define ARMADA_370_XP_INT_FABRIC_MASK_OFFS	(0x54)
+#define ARMADA_370_XP_INT_CAUSE_PERF(cpu)	(1 << cpu)
 
 #define ARMADA_370_XP_MAX_PER_CPU_IRQS		(28)
 
@@ -281,13 +360,11 @@ static int armada_370_xp_mpic_irq_map(struct irq_domain *h,
 		irq_set_percpu_devid(virq);
 		irq_set_chip_and_handler(virq, &armada_370_xp_irq_chip,
 					handle_percpu_devid_irq);
-
 	} else {
 		irq_set_chip_and_handler(virq, &armada_370_xp_irq_chip,
 					handle_level_irq);
 	}
 	irq_set_probe(virq);
-	irq_clear_status_flags(virq, IRQ_NOAUTOEN);
 
 	return 0;
 }
@@ -345,16 +422,40 @@ static void armada_mpic_send_doorbell(const struct cpumask *mask,
 		ARMADA_370_XP_SW_TRIG_INT_OFFS);
 }
 
+static void armada_xp_mpic_reenable_percpu(void)
+{
+	unsigned int irq;
+
+	/* Re-enable per-CPU interrupts that were enabled before suspend */
+	for (irq = 0; irq < ARMADA_370_XP_MAX_PER_CPU_IRQS; irq++) {
+		struct irq_data *data;
+		int virq;
+
+		virq = irq_linear_revmap(armada_370_xp_mpic_domain, irq);
+		if (virq == 0)
+			continue;
+
+		data = irq_get_irq_data(virq);
+
+		if (!irq_percpu_is_enabled(virq))
+			continue;
+
+		armada_370_xp_irq_unmask(data);
+	}
+}
+
 static int armada_xp_mpic_starting_cpu(unsigned int cpu)
 {
 	armada_xp_mpic_perf_init();
 	armada_xp_mpic_smp_cpu_init();
+	armada_xp_mpic_reenable_percpu();
 	return 0;
 }
 
 static int mpic_cascaded_starting_cpu(unsigned int cpu)
 {
 	armada_xp_mpic_perf_init();
+	armada_xp_mpic_reenable_percpu();
 	enable_percpu_irq(parent_irq, IRQ_TYPE_NONE);
 	return 0;
 }
@@ -502,16 +603,27 @@ static void armada_370_xp_mpic_resume(void)
 		if (virq == 0)
 			continue;
 
-		if (!is_percpu_irq(irq))
+		data = irq_get_irq_data(virq);
+
+		if (!is_percpu_irq(irq)) {
+			/* Non per-CPU interrupts */
 			writel(irq, per_cpu_int_base +
 			       ARMADA_370_XP_INT_CLEAR_MASK_OFFS);
-		else
+			if (!irqd_irq_disabled(data))
+				armada_370_xp_irq_unmask(data);
+		} else {
+			/* Per-CPU interrupts */
 			writel(irq, main_int_base +
 			       ARMADA_370_XP_INT_SET_ENABLE_OFFS);
 
-		data = irq_get_irq_data(virq);
-		if (!irqd_irq_disabled(data))
-			armada_370_xp_irq_unmask(data);
+			/*
+			 * Re-enable on the current CPU,
+			 * armada_xp_mpic_reenable_percpu() will take
+			 * care of secondary CPUs when they come up.
+			 */
+			if (irq_percpu_is_enabled(virq))
+				armada_370_xp_irq_unmask(data);
+		}
 	}
 
 	/* Reconfigure doorbells for IPIs and MSIs */
@@ -563,7 +675,7 @@ static int __init armada_370_xp_mpic_of_init(struct device_node *node,
 		irq_domain_add_linear(node, nr_irqs,
 				&armada_370_xp_mpic_irq_ops, NULL);
 	BUG_ON(!armada_370_xp_mpic_domain);
-	armada_370_xp_mpic_domain->bus_token = DOMAIN_BUS_WIRED;
+	irq_domain_update_bus_token(armada_370_xp_mpic_domain, DOMAIN_BUS_WIRED);
 
 	/* Setup for the boot CPU */
 	armada_xp_mpic_perf_init();