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 provides:

   - a major update to the auto affinity management code, which is used
     by multi-queue devices

   - move of the microblaze irq chip driver into the common driver code
     so it can be shared between microblaze, powerpc and MIPS

   - a series of updates to the ARM GICV3 interrupt controller

   - the usual pile of fixes and small improvements all over the place"

* 'irq-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (25 commits)
  powerpc/virtex: Use generic xilinx irqchip driver
  irqchip/xilinx: Try to fall back if xlnx,kind-of-intr not provided
  irqchip/xilinx: Add support for parent intc
  irqchip/xilinx: Rename get_irq to xintc_get_irq
  irqchip/xilinx: Restructure and use jump label api
  irqchip/xilinx: Clean up print messages
  microblaze/irqchip: Move intc driver to irqchip
  ARM: virt: Select ARM_GIC_V3_ITS
  ARM: gic-v3-its: Add 32bit support to GICv3 ITS
  irqchip/gic-v3-its: Specialise readq and writeq accesses
  irqchip/gic-v3-its: Specialise flush_dcache operation
  irqchip/gic-v3-its: Narrow down Entry Size when used as a divider
  irqchip/gic-v3-its: Change unsigned types for AArch32 compatibility
  irqchip/gic-v3: Use nops macro for Cavium ThunderX erratum 23154
  irqchip/gic-v3: Convert arm64 GIC accessors to {read,write}_sysreg_s
  genirq/msi: Drop artificial PCI dependency
  irqchip/bcm7038-l1: Implement irq_cpu_offline() callback
  genirq/affinity: Use default affinity mask for reserved vectors
  genirq/affinity: Take reserved vectors into account when spreading irqs
  PCI: Remove the irq_affinity mask from struct pci_dev
  ...

arch/arm/Kconfig

@@ -703,6 +703,7 @@ config ARCH_VIRT
 	select ARM_GIC
 	select ARM_GIC_V2M if PCI
 	select ARM_GIC_V3
+	select ARM_GIC_V3_ITS if PCI
 	select ARM_PSCI
 	select HAVE_ARM_ARCH_TIMER
 

arch/arm/include/asm/arch_gicv3.h

@@ -22,6 +22,7 @@
 
 #include <linux/io.h>
 #include <asm/barrier.h>
+#include <asm/cacheflush.h>
 #include <asm/cp15.h>
 
 #define ICC_EOIR1			__ACCESS_CP15(c12, 0, c12, 1)
@@ -230,19 +231,14 @@ static inline void gic_write_bpr1(u32 val)
  * AArch32, since the syndrome register doesn't provide any information for
  * them.
  * Consequently, the following IO helpers use 32bit accesses.
- *
- * There are only two registers that need 64bit accesses in this driver:
- * - GICD_IROUTERn, contain the affinity values associated to each interrupt.
- *   The upper-word (aff3) will always be 0, so there is no need for a lock.
- * - GICR_TYPER is an ID register and doesn't need atomicity.
  */
-static inline void gic_write_irouter(u64 val, volatile void __iomem *addr)
+static inline void __gic_writeq_nonatomic(u64 val, volatile void __iomem *addr)
 {
 	writel_relaxed((u32)val, addr);
 	writel_relaxed((u32)(val >> 32), addr + 4);
 }
 
-static inline u64 gic_read_typer(const volatile void __iomem *addr)
+static inline u64 __gic_readq_nonatomic(const volatile void __iomem *addr)
 {
 	u64 val;
 
@@ -251,5 +247,49 @@ static inline u64 gic_read_typer(const volatile void __iomem *addr)
 	return val;
 }
 
+#define gic_flush_dcache_to_poc(a,l)    __cpuc_flush_dcache_area((a), (l))
+
+/*
+ *  GICD_IROUTERn, contain the affinity values associated to each interrupt.
+ *  The upper-word (aff3) will always be 0, so there is no need for a lock.
+ */
+#define gic_write_irouter(v, c)		__gic_writeq_nonatomic(v, c)
+
+/*
+ * GICR_TYPER is an ID register and doesn't need atomicity.
+ */
+#define gic_read_typer(c)		__gic_readq_nonatomic(c)
+
+/*
+ * GITS_BASER - hi and lo bits may be accessed independently.
+ */
+#define gits_read_baser(c)		__gic_readq_nonatomic(c)
+#define gits_write_baser(v, c)		__gic_writeq_nonatomic(v, c)
+
+/*
+ * GICR_PENDBASER and GICR_PROPBASE are changed with LPIs disabled, so they
+ * won't be being used during any updates and can be changed non-atomically
+ */
+#define gicr_read_propbaser(c)		__gic_readq_nonatomic(c)
+#define gicr_write_propbaser(v, c)	__gic_writeq_nonatomic(v, c)
+#define gicr_read_pendbaser(c)		__gic_readq_nonatomic(c)
+#define gicr_write_pendbaser(v, c)	__gic_writeq_nonatomic(v, c)
+
+/*
+ * GITS_TYPER is an ID register and doesn't need atomicity.
+ */
+#define gits_read_typer(c)		__gic_readq_nonatomic(c)
+
+/*
+ * GITS_CBASER - hi and lo bits may be accessed independently.
+ */
+#define gits_read_cbaser(c)		__gic_readq_nonatomic(c)
+#define gits_write_cbaser(v, c)		__gic_writeq_nonatomic(v, c)
+
+/*
+ * GITS_CWRITER - hi and lo bits may be accessed independently.
+ */
+#define gits_write_cwriter(v, c)	__gic_writeq_nonatomic(v, c)
+
 #endif /* !__ASSEMBLY__ */
 #endif /* !__ASM_ARCH_GICV3_H */

arch/arm64/include/asm/arch_gicv3.h

@@ -79,19 +79,10 @@
 
 #include <linux/stringify.h>
 #include <asm/barrier.h>
+#include <asm/cacheflush.h>
 
-#define read_gicreg(r)							\
-	({								\
-		u64 reg;						\
-		asm volatile("mrs_s %0, " __stringify(r) : "=r" (reg));	\
-		reg;							\
-	})
-
-#define write_gicreg(v,r)						\
-	do {								\
-		u64 __val = (v);					\
-		asm volatile("msr_s " __stringify(r) ", %0" : : "r" (__val));\
-	} while (0)
+#define read_gicreg			read_sysreg_s
+#define write_gicreg			write_sysreg_s
 
 /*
  * Low-level accessors
@@ -102,13 +93,13 @@
 
 static inline void gic_write_eoir(u32 irq)
 {
-	asm volatile("msr_s " __stringify(ICC_EOIR1_EL1) ", %0" : : "r" ((u64)irq));
+	write_sysreg_s(irq, ICC_EOIR1_EL1);
 	isb();
 }
 
 static inline void gic_write_dir(u32 irq)
 {
-	asm volatile("msr_s " __stringify(ICC_DIR_EL1) ", %0" : : "r" ((u64)irq));
+	write_sysreg_s(irq, ICC_DIR_EL1);
 	isb();
 }
 
@@ -116,7 +107,7 @@ static inline u64 gic_read_iar_common(void)
 {
 	u64 irqstat;
 
-	asm volatile("mrs_s %0, " __stringify(ICC_IAR1_EL1) : "=r" (irqstat));
+	irqstat = read_sysreg_s(ICC_IAR1_EL1);
 	dsb(sy);
 	return irqstat;
 }
@@ -132,12 +123,9 @@ static inline u64 gic_read_iar_cavium_thunderx(void)
 {
 	u64 irqstat;
 
-	asm volatile(
-		"nop;nop;nop;nop\n\t"
-		"nop;nop;nop;nop\n\t"
-		"mrs_s %0, " __stringify(ICC_IAR1_EL1) "\n\t"
-		"nop;nop;nop;nop"
-		: "=r" (irqstat));
+	nops(8);
+	irqstat = read_sysreg_s(ICC_IAR1_EL1);
+	nops(4);
 	mb();
 
 	return irqstat;
@@ -145,37 +133,34 @@ static inline u64 gic_read_iar_cavium_thunderx(void)
 
 static inline void gic_write_pmr(u32 val)
 {
-	asm volatile("msr_s " __stringify(ICC_PMR_EL1) ", %0" : : "r" ((u64)val));
+	write_sysreg_s(val, ICC_PMR_EL1);
 }
 
 static inline void gic_write_ctlr(u32 val)
 {
-	asm volatile("msr_s " __stringify(ICC_CTLR_EL1) ", %0" : : "r" ((u64)val));
+	write_sysreg_s(val, ICC_CTLR_EL1);
 	isb();
 }
 
 static inline void gic_write_grpen1(u32 val)
 {
-	asm volatile("msr_s " __stringify(ICC_GRPEN1_EL1) ", %0" : : "r" ((u64)val));
+	write_sysreg_s(val, ICC_GRPEN1_EL1);
 	isb();
 }
 
 static inline void gic_write_sgi1r(u64 val)
 {
-	asm volatile("msr_s " __stringify(ICC_SGI1R_EL1) ", %0" : : "r" (val));
+	write_sysreg_s(val, ICC_SGI1R_EL1);
 }
 
 static inline u32 gic_read_sre(void)
 {
-	u64 val;
-
-	asm volatile("mrs_s %0, " __stringify(ICC_SRE_EL1) : "=r" (val));
-	return val;
+	return read_sysreg_s(ICC_SRE_EL1);
 }
 
 static inline void gic_write_sre(u32 val)
 {
-	asm volatile("msr_s " __stringify(ICC_SRE_EL1) ", %0" : : "r" ((u64)val));
+	write_sysreg_s(val, ICC_SRE_EL1);
 	isb();
 }
 
@@ -187,5 +172,21 @@ static inline void gic_write_bpr1(u32 val)
 #define gic_read_typer(c)		readq_relaxed(c)
 #define gic_write_irouter(v, c)		writeq_relaxed(v, c)
 
+#define gic_flush_dcache_to_poc(a,l)	__flush_dcache_area((a), (l))
+
+#define gits_read_baser(c)		readq_relaxed(c)
+#define gits_write_baser(v, c)		writeq_relaxed(v, c)
+
+#define gits_read_cbaser(c)		readq_relaxed(c)
+#define gits_write_cbaser(v, c)		writeq_relaxed(v, c)
+
+#define gits_write_cwriter(v, c)	writeq_relaxed(v, c)
+
+#define gicr_read_propbaser(c)		readq_relaxed(c)
+#define gicr_write_propbaser(v, c)	writeq_relaxed(v, c)
+
+#define gicr_write_pendbaser(v, c)	writeq_relaxed(v, c)
+#define gicr_read_pendbaser(c)		readq_relaxed(c)
+
 #endif /* __ASSEMBLY__ */
 #endif /* __ASM_ARCH_GICV3_H */

arch/microblaze/Kconfig

@@ -27,6 +27,7 @@ config MICROBLAZE
 	select HAVE_MEMBLOCK_NODE_MAP
 	select HAVE_OPROFILE
 	select IRQ_DOMAIN
+	select XILINX_INTC
 	select MODULES_USE_ELF_RELA
 	select OF
 	select OF_EARLY_FLATTREE

arch/microblaze/include/asm/irq.h

@@ -16,6 +16,6 @@ struct pt_regs;
 extern void do_IRQ(struct pt_regs *regs);
 
 /* should be defined in each interrupt controller driver */
-extern unsigned int get_irq(void);
+extern unsigned int xintc_get_irq(void);
 
 #endif /* _ASM_MICROBLAZE_IRQ_H */

arch/microblaze/kernel/Makefile

@@ -15,7 +15,7 @@ endif
 extra-y := head.o vmlinux.lds
 
 obj-y += dma.o exceptions.o \
-	hw_exception_handler.o intc.o irq.o \
+	hw_exception_handler.o irq.o \
 	platform.o process.o prom.o ptrace.o \
 	reset.o setup.o signal.o sys_microblaze.o timer.o traps.o unwind.o
 

arch/microblaze/kernel/irq.c

@@ -29,12 +29,12 @@ void __irq_entry do_IRQ(struct pt_regs *regs)
 	trace_hardirqs_off();
 
 	irq_enter();
-	irq = get_irq();
+	irq = xintc_get_irq();
 next_irq:
 	BUG_ON(!irq);
 	generic_handle_irq(irq);
 
-	irq = get_irq();
+	irq = xintc_get_irq();
 	if (irq != -1U) {
 		pr_debug("next irq: %d\n", irq);
 		++concurrent_irq;

arch/powerpc/include/asm/xilinx_intc.h

@@ -14,7 +14,7 @@
 #ifdef __KERNEL__
 
 extern void __init xilinx_intc_init_tree(void);
-extern unsigned int xilinx_intc_get_irq(void);
+extern unsigned int xintc_get_irq(void);
 
 #endif /* __KERNEL__ */
 #endif /* _ASM_POWERPC_XILINX_INTC_H */

arch/powerpc/platforms/40x/Kconfig

@@ -64,6 +64,7 @@ config XILINX_VIRTEX_GENERIC_BOARD
 	default n
 	select XILINX_VIRTEX_II_PRO
 	select XILINX_VIRTEX_4_FX
+	select XILINX_INTC
 	help
 	  This option enables generic support for Xilinx Virtex based boards.
 

arch/powerpc/platforms/40x/virtex.c

@@ -48,7 +48,7 @@ define_machine(virtex) {
 	.probe			= virtex_probe,
 	.setup_arch		= xilinx_pci_init,
 	.init_IRQ		= xilinx_intc_init_tree,
-	.get_irq		= xilinx_intc_get_irq,
+	.get_irq		= xintc_get_irq,
 	.restart		= ppc4xx_reset_system,
 	.calibrate_decr		= generic_calibrate_decr,
 };

arch/powerpc/platforms/44x/Kconfig

@@ -241,6 +241,7 @@ config XILINX_VIRTEX440_GENERIC_BOARD
 	depends on 44x
 	default n
 	select XILINX_VIRTEX_5_FXT
+	select XILINX_INTC
 	help
 	  This option enables generic support for Xilinx Virtex based boards
 	  that use a 440 based processor in the Virtex 5 FXT FPGA architecture.

arch/powerpc/platforms/44x/virtex.c

@@ -54,7 +54,7 @@ define_machine(virtex) {
 	.probe			= virtex_probe,
 	.setup_arch		= xilinx_pci_init,
 	.init_IRQ		= xilinx_intc_init_tree,
-	.get_irq		= xilinx_intc_get_irq,
+	.get_irq		= xintc_get_irq,
 	.calibrate_decr		= generic_calibrate_decr,
 	.restart		= ppc4xx_reset_system,
 };

arch/powerpc/sysdev/xilinx_intc.c

@@ -29,194 +29,7 @@
 #include <asm/processor.h>
 #include <asm/i8259.h>
 #include <asm/irq.h>
-
-/*
- * INTC Registers
- */
-#define XINTC_ISR	0	/* Interrupt Status */
-#define XINTC_IPR	4	/* Interrupt Pending */
-#define XINTC_IER	8	/* Interrupt Enable */
-#define XINTC_IAR	12	/* Interrupt Acknowledge */
-#define XINTC_SIE	16	/* Set Interrupt Enable bits */
-#define XINTC_CIE	20	/* Clear Interrupt Enable bits */
-#define XINTC_IVR	24	/* Interrupt Vector */
-#define XINTC_MER	28	/* Master Enable */
-
-static struct irq_domain *master_irqhost;
-
-#define XILINX_INTC_MAXIRQS	(32)
-
-/* The following table allows the interrupt type, edge or level,
- * to be cached after being read from the device tree until the interrupt
- * is mapped
- */
-static int xilinx_intc_typetable[XILINX_INTC_MAXIRQS];
-
-/* Map the interrupt type from the device tree to the interrupt types
- * used by the interrupt subsystem
- */
-static unsigned char xilinx_intc_map_senses[] = {
-	IRQ_TYPE_EDGE_RISING,
-	IRQ_TYPE_EDGE_FALLING,
-	IRQ_TYPE_LEVEL_HIGH,
-	IRQ_TYPE_LEVEL_LOW,
-};
-
-/*
- * The interrupt controller is setup such that it doesn't work well with
- * the level interrupt handler in the kernel because the handler acks the
- * interrupt before calling the application interrupt handler. To deal with
- * that, we use 2 different irq chips so that different functions can be
- * used for level and edge type interrupts.
- *
- * IRQ Chip common (across level and edge) operations
- */
-static void xilinx_intc_mask(struct irq_data *d)
-{
-	int irq = irqd_to_hwirq(d);
-	void * regs = irq_data_get_irq_chip_data(d);
-	pr_debug("mask: %d\n", irq);
-	out_be32(regs + XINTC_CIE, 1 << irq);
-}
-
-static int xilinx_intc_set_type(struct irq_data *d, unsigned int flow_type)
-{
-	return 0;
-}
-
-/*
- * IRQ Chip level operations
- */
-static void xilinx_intc_level_unmask(struct irq_data *d)
-{
-	int irq = irqd_to_hwirq(d);
-	void * regs = irq_data_get_irq_chip_data(d);
-	pr_debug("unmask: %d\n", irq);
-	out_be32(regs + XINTC_SIE, 1 << irq);
-
-	/* ack level irqs because they can't be acked during
-	 * ack function since the handle_level_irq function
-	 * acks the irq before calling the inerrupt handler
-	 */
-	out_be32(regs + XINTC_IAR, 1 << irq);
-}
-
-static struct irq_chip xilinx_intc_level_irqchip = {
-	.name = "Xilinx Level INTC",
-	.irq_mask = xilinx_intc_mask,
-	.irq_mask_ack = xilinx_intc_mask,
-	.irq_unmask = xilinx_intc_level_unmask,
-	.irq_set_type = xilinx_intc_set_type,
-};
-
-/*
- * IRQ Chip edge operations
- */
-static void xilinx_intc_edge_unmask(struct irq_data *d)
-{
-	int irq = irqd_to_hwirq(d);
-	void *regs = irq_data_get_irq_chip_data(d);
-	pr_debug("unmask: %d\n", irq);
-	out_be32(regs + XINTC_SIE, 1 << irq);
-}
-
-static void xilinx_intc_edge_ack(struct irq_data *d)
-{
-	int irq = irqd_to_hwirq(d);
-	void * regs = irq_data_get_irq_chip_data(d);
-	pr_debug("ack: %d\n", irq);
-	out_be32(regs + XINTC_IAR, 1 << irq);
-}
-
-static struct irq_chip xilinx_intc_edge_irqchip = {
-	.name = "Xilinx Edge  INTC",
-	.irq_mask = xilinx_intc_mask,
-	.irq_unmask = xilinx_intc_edge_unmask,
-	.irq_ack = xilinx_intc_edge_ack,
-	.irq_set_type = xilinx_intc_set_type,
-};
-
-/*
- * IRQ Host operations
- */
-
-/**
- * xilinx_intc_xlate - translate virq# from device tree interrupts property
- */
-static int xilinx_intc_xlate(struct irq_domain *h, struct device_node *ct,
-				const u32 *intspec, unsigned int intsize,
-				irq_hw_number_t *out_hwirq,
-				unsigned int *out_flags)
-{
-	if ((intsize < 2) || (intspec[0] >= XILINX_INTC_MAXIRQS))
-		return -EINVAL;
-
-	/* keep a copy of the interrupt type til the interrupt is mapped
-	 */
-	xilinx_intc_typetable[intspec[0]] = xilinx_intc_map_senses[intspec[1]];
-
-	/* Xilinx uses 2 interrupt entries, the 1st being the h/w
-	 * interrupt number, the 2nd being the interrupt type, edge or level
-	 */
-	*out_hwirq = intspec[0];
-	*out_flags = xilinx_intc_map_senses[intspec[1]];
-
-	return 0;
-}
-static int xilinx_intc_map(struct irq_domain *h, unsigned int virq,
-				  irq_hw_number_t irq)
-{
-	irq_set_chip_data(virq, h->host_data);
-
-	if (xilinx_intc_typetable[irq] == IRQ_TYPE_LEVEL_HIGH ||
-	    xilinx_intc_typetable[irq] == IRQ_TYPE_LEVEL_LOW) {
-		irq_set_chip_and_handler(virq, &xilinx_intc_level_irqchip,
-					 handle_level_irq);
-	} else {
-		irq_set_chip_and_handler(virq, &xilinx_intc_edge_irqchip,
-					 handle_edge_irq);
-	}
-	return 0;
-}
-
-static const struct irq_domain_ops xilinx_intc_ops = {
-	.map = xilinx_intc_map,
-	.xlate = xilinx_intc_xlate,
-};
-
-struct irq_domain * __init
-xilinx_intc_init(struct device_node *np)
-{
-	struct irq_domain * irq;
-	void * regs;
-
-	/* Find and map the intc registers */
-	regs = of_iomap(np, 0);
-	if (!regs) {
-		pr_err("xilinx_intc: could not map registers\n");
-		return NULL;
-	}
-
-	/* Setup interrupt controller */
-	out_be32(regs + XINTC_IER, 0); /* disable all irqs */
-	out_be32(regs + XINTC_IAR, ~(u32) 0); /* Acknowledge pending irqs */
-	out_be32(regs + XINTC_MER, 0x3UL); /* Turn on the Master Enable. */
-
-	/* Allocate and initialize an irq_domain structure. */
-	irq = irq_domain_add_linear(np, XILINX_INTC_MAXIRQS, &xilinx_intc_ops,
-				    regs);
-	if (!irq)
-		panic(__FILE__ ": Cannot allocate IRQ host\n");
-
-	return irq;
-}
-
-int xilinx_intc_get_irq(void)
-{
-	void * regs = master_irqhost->host_data;
-	pr_debug("get_irq:\n");
-	return irq_linear_revmap(master_irqhost, in_be32(regs + XINTC_IVR));
-}
+#include <linux/irqchip.h>
 
 #if defined(CONFIG_PPC_I8259)
 /*
@@ -265,31 +78,11 @@ static void __init xilinx_i8259_setup_cascade(void)
 static inline void xilinx_i8259_setup_cascade(void) { return; }
 #endif /* defined(CONFIG_PPC_I8259) */
 
-static const struct of_device_id xilinx_intc_match[] __initconst = {
-	{ .compatible = "xlnx,opb-intc-1.00.c", },
-	{ .compatible = "xlnx,xps-intc-1.00.a", },
-	{}
-};
-
 /*
  * Initialize master Xilinx interrupt controller
  */
 void __init xilinx_intc_init_tree(void)
 {
-	struct device_node *np;
-
-	/* find top level interrupt controller */
-	for_each_matching_node(np, xilinx_intc_match) {
-		if (!of_get_property(np, "interrupts", NULL))
-			break;
-	}
-	BUG_ON(!np);
-
-	master_irqhost = xilinx_intc_init(np);
-	BUG_ON(!master_irqhost);
-
-	irq_set_default_host(master_irqhost);
-	of_node_put(np);
-
+	irqchip_init();
 	xilinx_i8259_setup_cascade();
 }

drivers/irqchip/Kconfig

@@ -211,6 +211,10 @@ config XTENSA_MX
 	bool
 	select IRQ_DOMAIN
 
+config XILINX_INTC
+	bool
+	select IRQ_DOMAIN
+
 config IRQ_CROSSBAR
 	bool
 	help

drivers/irqchip/Makefile

@@ -52,6 +52,7 @@ obj-$(CONFIG_TB10X_IRQC)		+= irq-tb10x.o
 obj-$(CONFIG_TS4800_IRQ)		+= irq-ts4800.o
 obj-$(CONFIG_XTENSA)			+= irq-xtensa-pic.o
 obj-$(CONFIG_XTENSA_MX)			+= irq-xtensa-mx.o
+obj-$(CONFIG_XILINX_INTC)		+= irq-xilinx-intc.o
 obj-$(CONFIG_IRQ_CROSSBAR)		+= irq-crossbar.o
 obj-$(CONFIG_SOC_VF610)			+= irq-vf610-mscm-ir.o
 obj-$(CONFIG_BCM6345_L1_IRQ)		+= irq-bcm6345-l1.o

drivers/irqchip/irq-bcm7038-l1.c

@@ -215,6 +215,31 @@ static int bcm7038_l1_set_affinity(struct irq_data *d,
 	return 0;
 }
 
+static void bcm7038_l1_cpu_offline(struct irq_data *d)
+{
+	struct cpumask *mask = irq_data_get_affinity_mask(d);
+	int cpu = smp_processor_id();
+	cpumask_t new_affinity;
+
+	/* This CPU was not on the affinity mask */
+	if (!cpumask_test_cpu(cpu, mask))
+		return;
+
+	if (cpumask_weight(mask) > 1) {
+		/*
+		 * Multiple CPU affinity, remove this CPU from the affinity
+		 * mask
+		 */
+		cpumask_copy(&new_affinity, mask);
+		cpumask_clear_cpu(cpu, &new_affinity);
+	} else {
+		/* Only CPU, put on the lowest online CPU */
+		cpumask_clear(&new_affinity);
+		cpumask_set_cpu(cpumask_first(cpu_online_mask), &new_affinity);
+	}
+	irq_set_affinity_locked(d, &new_affinity, false);
+}
+
 static int __init bcm7038_l1_init_one(struct device_node *dn,
 				      unsigned int idx,
 				      struct bcm7038_l1_chip *intc)
@@ -266,6 +291,7 @@ static struct irq_chip bcm7038_l1_irq_chip = {
 	.irq_mask		= bcm7038_l1_mask,
 	.irq_unmask		= bcm7038_l1_unmask,
 	.irq_set_affinity	= bcm7038_l1_set_affinity,
+	.irq_cpu_offline	= bcm7038_l1_cpu_offline,
 };
 
 static int bcm7038_l1_map(struct irq_domain *d, unsigned int virq,

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

@@ -37,7 +37,6 @@
 #include <linux/irqchip.h>
 #include <linux/irqchip/arm-gic-v3.h>
 
-#include <asm/cacheflush.h>
 #include <asm/cputype.h>
 #include <asm/exception.h>
 
@@ -196,7 +195,7 @@ typedef struct its_collection *(*its_cmd_builder_t)(struct its_cmd_block *,
 
 static void its_encode_cmd(struct its_cmd_block *cmd, u8 cmd_nr)
 {
-	cmd->raw_cmd[0] &= ~0xffUL;
+	cmd->raw_cmd[0] &= ~0xffULL;
 	cmd->raw_cmd[0] |= cmd_nr;
 }
 
@@ -208,43 +207,43 @@ static void its_encode_devid(struct its_cmd_block *cmd, u32 devid)
 
 static void its_encode_event_id(struct its_cmd_block *cmd, u32 id)
 {
-	cmd->raw_cmd[1] &= ~0xffffffffUL;
+	cmd->raw_cmd[1] &= ~0xffffffffULL;
 	cmd->raw_cmd[1] |= id;
 }
 
 static void its_encode_phys_id(struct its_cmd_block *cmd, u32 phys_id)
 {
-	cmd->raw_cmd[1] &= 0xffffffffUL;
+	cmd->raw_cmd[1] &= 0xffffffffULL;
 	cmd->raw_cmd[1] |= ((u64)phys_id) << 32;
 }
 
 static void its_encode_size(struct its_cmd_block *cmd, u8 size)
 {
-	cmd->raw_cmd[1] &= ~0x1fUL;
+	cmd->raw_cmd[1] &= ~0x1fULL;
 	cmd->raw_cmd[1] |= size & 0x1f;
 }
 
 static void its_encode_itt(struct its_cmd_block *cmd, u64 itt_addr)
 {
-	cmd->raw_cmd[2] &= ~0xffffffffffffUL;
-	cmd->raw_cmd[2] |= itt_addr & 0xffffffffff00UL;
+	cmd->raw_cmd[2] &= ~0xffffffffffffULL;
+	cmd->raw_cmd[2] |= itt_addr & 0xffffffffff00ULL;
 }
 
 static void its_encode_valid(struct its_cmd_block *cmd, int valid)
 {
-	cmd->raw_cmd[2] &= ~(1UL << 63);
+	cmd->raw_cmd[2] &= ~(1ULL << 63);
 	cmd->raw_cmd[2] |= ((u64)!!valid) << 63;
 }
 
 static void its_encode_target(struct its_cmd_block *cmd, u64 target_addr)
 {
-	cmd->raw_cmd[2] &= ~(0xffffffffUL << 16);
-	cmd->raw_cmd[2] |= (target_addr & (0xffffffffUL << 16));
+	cmd->raw_cmd[2] &= ~(0xffffffffULL << 16);
+	cmd->raw_cmd[2] |= (target_addr & (0xffffffffULL << 16));
 }
 
 static void its_encode_collection(struct its_cmd_block *cmd, u16 col)
 {
-	cmd->raw_cmd[2] &= ~0xffffUL;
+	cmd->raw_cmd[2] &= ~0xffffULL;
 	cmd->raw_cmd[2] |= col;
 }
 
@@ -433,7 +432,7 @@ static void its_flush_cmd(struct its_node *its, struct its_cmd_block *cmd)
 	 * the ITS.
 	 */
 	if (its->flags & ITS_FLAGS_CMDQ_NEEDS_FLUSHING)
-		__flush_dcache_area(cmd, sizeof(*cmd));
+		gic_flush_dcache_to_poc(cmd, sizeof(*cmd));
 	else
 		dsb(ishst);
 }
@@ -602,7 +601,7 @@ static void lpi_set_config(struct irq_data *d, bool enable)
 	 * Humpf...
 	 */
 	if (gic_rdists->flags & RDIST_FLAGS_PROPBASE_NEEDS_FLUSHING)
-		__flush_dcache_area(cfg, sizeof(*cfg));
+		gic_flush_dcache_to_poc(cfg, sizeof(*cfg));
 	else
 		dsb(ishst);
 	its_send_inv(its_dev, id);
@@ -657,8 +656,8 @@ static void its_irq_compose_msi_msg(struct irq_data *d, struct msi_msg *msg)
 	its = its_dev->its;
 	addr = its->phys_base + GITS_TRANSLATER;
 
-	msg->address_lo		= addr & ((1UL << 32) - 1);
-	msg->address_hi		= addr >> 32;
+	msg->address_lo		= lower_32_bits(addr);
+	msg->address_hi		= upper_32_bits(addr);
 	msg->data		= its_get_event_id(d);
 
 	iommu_dma_map_msi_msg(d->irq, msg);
@@ -817,7 +816,7 @@ static int __init its_alloc_lpi_tables(void)
 	       LPI_PROPBASE_SZ);
 
 	/* Make sure the GIC will observe the written configuration */
-	__flush_dcache_area(page_address(gic_rdists->prop_page), LPI_PROPBASE_SZ);
+	gic_flush_dcache_to_poc(page_address(gic_rdists->prop_page), LPI_PROPBASE_SZ);
 
 	return 0;
 }
@@ -836,7 +835,7 @@ static u64 its_read_baser(struct its_node *its, struct its_baser *baser)
 {
 	u32 idx = baser - its->tables;
 
-	return readq_relaxed(its->base + GITS_BASER + (idx << 3));
+	return gits_read_baser(its->base + GITS_BASER + (idx << 3));
 }
 
 static void its_write_baser(struct its_node *its, struct its_baser *baser,
@@ -844,7 +843,7 @@ static void its_write_baser(struct its_node *its, struct its_baser *baser,
 {
 	u32 idx = baser - its->tables;
 
-	writeq_relaxed(val, its->base + GITS_BASER + (idx << 3));
+	gits_write_baser(val, its->base + GITS_BASER + (idx << 3));
 	baser->val = its_read_baser(its, baser);
 }
 
@@ -910,7 +909,7 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
 		shr = tmp & GITS_BASER_SHAREABILITY_MASK;
 		if (!shr) {
 			cache = GITS_BASER_nC;
-			__flush_dcache_area(base, PAGE_ORDER_TO_SIZE(order));
+			gic_flush_dcache_to_poc(base, PAGE_ORDER_TO_SIZE(order));
 		}
 		goto retry_baser;
 	}
@@ -935,9 +934,9 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
 	}
 
 	if (val != tmp) {
-		pr_err("ITS@%pa: %s doesn't stick: %lx %lx\n",
+		pr_err("ITS@%pa: %s doesn't stick: %llx %llx\n",
 		       &its->phys_base, its_base_type_string[type],
-		       (unsigned long) val, (unsigned long) tmp);
+		       val, tmp);
 		free_pages((unsigned long)base, order);
 		return -ENXIO;
 	}
@@ -948,7 +947,7 @@ static int its_setup_baser(struct its_node *its, struct its_baser *baser,
 	tmp = indirect ? GITS_LVL1_ENTRY_SIZE : esz;
 
 	pr_info("ITS@%pa: allocated %d %s @%lx (%s, esz %d, psz %dK, shr %d)\n",
-		&its->phys_base, (int)(PAGE_ORDER_TO_SIZE(order) / tmp),
+		&its->phys_base, (int)(PAGE_ORDER_TO_SIZE(order) / (int)tmp),
 		its_base_type_string[type],
 		(unsigned long)virt_to_phys(base),
 		indirect ? "indirect" : "flat", (int)esz,
@@ -983,7 +982,7 @@ static bool its_parse_baser_device(struct its_node *its, struct its_baser *baser
 			 * which is reported by ITS hardware times lvl1 table
 			 * entry size.
 			 */
-			ids -= ilog2(psz / esz);
+			ids -= ilog2(psz / (int)esz);
 			esz = GITS_LVL1_ENTRY_SIZE;
 		}
 	}
@@ -998,7 +997,7 @@ static bool its_parse_baser_device(struct its_node *its, struct its_baser *baser
 	new_order = max_t(u32, get_order(esz << ids), new_order);
 	if (new_order >= MAX_ORDER) {
 		new_order = MAX_ORDER - 1;
-		ids = ilog2(PAGE_ORDER_TO_SIZE(new_order) / esz);
+		ids = ilog2(PAGE_ORDER_TO_SIZE(new_order) / (int)esz);
 		pr_warn("ITS@%pa: Device Table too large, reduce ids %u->%u\n",
 			&its->phys_base, its->device_ids, ids);
 	}
@@ -1102,7 +1101,7 @@ static void its_cpu_init_lpis(void)
 		}
 
 		/* Make sure the GIC will observe the zero-ed page */
-		__flush_dcache_area(page_address(pend_page), LPI_PENDBASE_SZ);
+		gic_flush_dcache_to_poc(page_address(pend_page), LPI_PENDBASE_SZ);
 
 		paddr = page_to_phys(pend_page);
 		pr_info("CPU%d: using LPI pending table @%pa\n",
@@ -1126,8 +1125,8 @@ static void its_cpu_init_lpis(void)
 	       GICR_PROPBASER_WaWb |
 	       ((LPI_NRBITS - 1) & GICR_PROPBASER_IDBITS_MASK));
 
-	writeq_relaxed(val, rbase + GICR_PROPBASER);
-	tmp = readq_relaxed(rbase + GICR_PROPBASER);
+	gicr_write_propbaser(val, rbase + GICR_PROPBASER);
+	tmp = gicr_read_propbaser(rbase + GICR_PROPBASER);
 
 	if ((tmp ^ val) & GICR_PROPBASER_SHAREABILITY_MASK) {
 		if (!(tmp & GICR_PROPBASER_SHAREABILITY_MASK)) {
@@ -1139,7 +1138,7 @@ static void its_cpu_init_lpis(void)
 			val &= ~(GICR_PROPBASER_SHAREABILITY_MASK |
 				 GICR_PROPBASER_CACHEABILITY_MASK);
 			val |= GICR_PROPBASER_nC;
-			writeq_relaxed(val, rbase + GICR_PROPBASER);
+			gicr_write_propbaser(val, rbase + GICR_PROPBASER);
 		}
 		pr_info_once("GIC: using cache flushing for LPI property table\n");
 		gic_rdists->flags |= RDIST_FLAGS_PROPBASE_NEEDS_FLUSHING;
@@ -1150,8 +1149,8 @@ static void its_cpu_init_lpis(void)
 	       GICR_PENDBASER_InnerShareable |
 	       GICR_PENDBASER_WaWb);
 
-	writeq_relaxed(val, rbase + GICR_PENDBASER);
-	tmp = readq_relaxed(rbase + GICR_PENDBASER);
+	gicr_write_pendbaser(val, rbase + GICR_PENDBASER);
+	tmp = gicr_read_pendbaser(rbase + GICR_PENDBASER);
 
 	if (!(tmp & GICR_PENDBASER_SHAREABILITY_MASK)) {
 		/*
@@ -1161,7 +1160,7 @@ static void its_cpu_init_lpis(void)
 		val &= ~(GICR_PENDBASER_SHAREABILITY_MASK |
 			 GICR_PENDBASER_CACHEABILITY_MASK);
 		val |= GICR_PENDBASER_nC;
-		writeq_relaxed(val, rbase + GICR_PENDBASER);
+		gicr_write_pendbaser(val, rbase + GICR_PENDBASER);
 	}
 
 	/* Enable LPIs */
@@ -1287,13 +1286,13 @@ static bool its_alloc_device_table(struct its_node *its, u32 dev_id)
 
 		/* Flush Lvl2 table to PoC if hw doesn't support coherency */
 		if (!(baser->val & GITS_BASER_SHAREABILITY_MASK))
-			__flush_dcache_area(page_address(page), baser->psz);
+			gic_flush_dcache_to_poc(page_address(page), baser->psz);
 
 		table[idx] = cpu_to_le64(page_to_phys(page) | GITS_BASER_VALID);
 
 		/* Flush Lvl1 entry to PoC if hw doesn't support coherency */
 		if (!(baser->val & GITS_BASER_SHAREABILITY_MASK))
-			__flush_dcache_area(table + idx, GITS_LVL1_ENTRY_SIZE);
+			gic_flush_dcache_to_poc(table + idx, GITS_LVL1_ENTRY_SIZE);
 
 		/* Ensure updated table contents are visible to ITS hardware */
 		dsb(sy);
@@ -1340,7 +1339,7 @@ static struct its_device *its_create_device(struct its_node *its, u32 dev_id,
 		return NULL;
 	}
 
-	__flush_dcache_area(itt, sz);
+	gic_flush_dcache_to_poc(itt, sz);
 
 	dev->its = its;
 	dev->itt = itt;
@@ -1717,8 +1716,8 @@ static int __init its_probe_one(struct resource *res,
 		 (ITS_CMD_QUEUE_SZ / SZ_4K - 1)	|
 		 GITS_CBASER_VALID);
 
-	writeq_relaxed(baser, its->base + GITS_CBASER);
-	tmp = readq_relaxed(its->base + GITS_CBASER);
+	gits_write_cbaser(baser, its->base + GITS_CBASER);
+	tmp = gits_read_cbaser(its->base + GITS_CBASER);
 
 	if ((tmp ^ baser) & GITS_CBASER_SHAREABILITY_MASK) {
 		if (!(tmp & GITS_CBASER_SHAREABILITY_MASK)) {
@@ -1730,13 +1729,13 @@ static int __init its_probe_one(struct resource *res,
 			baser &= ~(GITS_CBASER_SHAREABILITY_MASK |
 				   GITS_CBASER_CACHEABILITY_MASK);
 			baser |= GITS_CBASER_nC;
-			writeq_relaxed(baser, its->base + GITS_CBASER);
+			gits_write_cbaser(baser, its->base + GITS_CBASER);
 		}
 		pr_info("ITS: using cache flushing for cmd queue\n");
 		its->flags |= ITS_FLAGS_CMDQ_NEEDS_FLUSHING;
 	}
 
-	writeq_relaxed(0, its->base + GITS_CWRITER);
+	gits_write_cwriter(0, its->base + GITS_CWRITER);
 	writel_relaxed(GITS_CTLR_ENABLE, its->base + GITS_CTLR);
 
 	err = its_init_domain(handle, its);

arch/microblaze/kernel/intc.c → drivers/irqchip/irq-xilinx-intc.c

@@ -12,11 +12,12 @@
 #include <linux/irqdomain.h>
 #include <linux/irq.h>
 #include <linux/irqchip.h>
+#include <linux/irqchip/chained_irq.h>
 #include <linux/of_address.h>
 #include <linux/io.h>
+#include <linux/jump_label.h>
 #include <linux/bug.h>
-
-static void __iomem *intc_baseaddr;
+#include <linux/of_irq.h>
 
 /* No one else should require these constants, so define them locally here. */
 #define ISR 0x00			/* Interrupt Status Register */
@@ -31,64 +32,67 @@ static void __iomem *intc_baseaddr;
 #define MER_ME (1<<0)
 #define MER_HIE (1<<1)
 
-static unsigned int (*read_fn)(void __iomem *);
-static void (*write_fn)(u32, void __iomem *);
+static DEFINE_STATIC_KEY_FALSE(xintc_is_be);
 
-static void intc_write32(u32 val, void __iomem *addr)
-{
-	iowrite32(val, addr);
-}
+struct xintc_irq_chip {
+	void		__iomem *base;
+	struct		irq_domain *root_domain;
+	u32		intr_mask;
+};
 
-static unsigned int intc_read32(void __iomem *addr)
-{
-	return ioread32(addr);
-}
+static struct xintc_irq_chip *xintc_irqc;
 
-static void intc_write32_be(u32 val, void __iomem *addr)
+static void xintc_write(int reg, u32 data)
 {
-	iowrite32be(val, addr);
+	if (static_branch_unlikely(&xintc_is_be))
+		iowrite32be(data, xintc_irqc->base + reg);
+	else
+		iowrite32(data, xintc_irqc->base + reg);
 }
 
-static unsigned int intc_read32_be(void __iomem *addr)
+static unsigned int xintc_read(int reg)
 {
-	return ioread32be(addr);
+	if (static_branch_unlikely(&xintc_is_be))
+		return ioread32be(xintc_irqc->base + reg);
+	else
+		return ioread32(xintc_irqc->base + reg);
 }
 
 static void intc_enable_or_unmask(struct irq_data *d)
 {
 	unsigned long mask = 1 << d->hwirq;
 
-	pr_debug("enable_or_unmask: %ld\n", d->hwirq);
+	pr_debug("irq-xilinx: enable_or_unmask: %ld\n", d->hwirq);
 
 	/* ack level irqs because they can't be acked during
 	 * ack function since the handle_level_irq function
 	 * acks the irq before calling the interrupt handler
 	 */
 	if (irqd_is_level_type(d))
-		write_fn(mask, intc_baseaddr + IAR);
+		xintc_write(IAR, mask);
 
-	write_fn(mask, intc_baseaddr + SIE);
+	xintc_write(SIE, mask);
 }
 
 static void intc_disable_or_mask(struct irq_data *d)
 {
-	pr_debug("disable: %ld\n", d->hwirq);
-	write_fn(1 << d->hwirq, intc_baseaddr + CIE);
+	pr_debug("irq-xilinx: disable: %ld\n", d->hwirq);
+	xintc_write(CIE, 1 << d->hwirq);
 }
 
 static void intc_ack(struct irq_data *d)
 {
-	pr_debug("ack: %ld\n", d->hwirq);
-	write_fn(1 << d->hwirq, intc_baseaddr + IAR);
+	pr_debug("irq-xilinx: ack: %ld\n", d->hwirq);
+	xintc_write(IAR, 1 << d->hwirq);
 }
 
 static void intc_mask_ack(struct irq_data *d)
 {
 	unsigned long mask = 1 << d->hwirq;
 
-	pr_debug("disable_and_ack: %ld\n", d->hwirq);
-	write_fn(mask, intc_baseaddr + CIE);
-	write_fn(mask, intc_baseaddr + IAR);
+	pr_debug("irq-xilinx: disable_and_ack: %ld\n", d->hwirq);
+	xintc_write(CIE, mask);
+	xintc_write(IAR, mask);
 }
 
 static struct irq_chip intc_dev = {
@@ -99,26 +103,22 @@ static struct irq_chip intc_dev = {
 	.irq_mask_ack = intc_mask_ack,
 };
 
-static struct irq_domain *root_domain;
-
-unsigned int get_irq(void)
+unsigned int xintc_get_irq(void)
 {
 	unsigned int hwirq, irq = -1;
 
-	hwirq = read_fn(intc_baseaddr + IVR);
+	hwirq = xintc_read(IVR);
 	if (hwirq != -1U)
-		irq = irq_find_mapping(root_domain, hwirq);
+		irq = irq_find_mapping(xintc_irqc->root_domain, hwirq);
 
-	pr_debug("get_irq: hwirq=%d, irq=%d\n", hwirq, irq);
+	pr_debug("irq-xilinx: hwirq=%d, irq=%d\n", hwirq, irq);
 
 	return irq;
 }
 
 static int xintc_map(struct irq_domain *d, unsigned int irq, irq_hw_number_t hw)
 {
-	u32 intr_mask = (u32)d->host_data;
-
-	if (intr_mask & (1 << hw)) {
+	if (xintc_irqc->intr_mask & (1 << hw)) {
 		irq_set_chip_and_handler_name(irq, &intc_dev,
 						handle_edge_irq, "edge");
 		irq_clear_status_flags(irq, IRQ_LEVEL);
@@ -135,62 +135,107 @@ static const struct irq_domain_ops xintc_irq_domain_ops = {
 	.map = xintc_map,
 };
 
+static void xil_intc_irq_handler(struct irq_desc *desc)
+{
+	struct irq_chip *chip = irq_desc_get_chip(desc);
+	u32 pending;
+
+	chained_irq_enter(chip, desc);
+	do {
+		pending = xintc_get_irq();
+		if (pending == -1U)
+			break;
+		generic_handle_irq(pending);
+	} while (true);
+	chained_irq_exit(chip, desc);
+}
+
 static int __init xilinx_intc_of_init(struct device_node *intc,
 					     struct device_node *parent)
 {
-	u32 nr_irq, intr_mask;
-	int ret;
+	u32 nr_irq;
+	int ret, irq;
+	struct xintc_irq_chip *irqc;
+
+	if (xintc_irqc) {
+		pr_err("irq-xilinx: Multiple instances aren't supported\n");
+		return -EINVAL;
+	}
+
+	irqc = kzalloc(sizeof(*irqc), GFP_KERNEL);
+	if (!irqc)
+		return -ENOMEM;
+
+	xintc_irqc = irqc;
 
-	intc_baseaddr = of_iomap(intc, 0);
-	BUG_ON(!intc_baseaddr);
+	irqc->base = of_iomap(intc, 0);
+	BUG_ON(!irqc->base);
 
 	ret = of_property_read_u32(intc, "xlnx,num-intr-inputs", &nr_irq);
 	if (ret < 0) {
-		pr_err("%s: unable to read xlnx,num-intr-inputs\n", __func__);
-		return ret;
+		pr_err("irq-xilinx: unable to read xlnx,num-intr-inputs\n");
+		goto err_alloc;
 	}
 
-	ret = of_property_read_u32(intc, "xlnx,kind-of-intr", &intr_mask);
+	ret = of_property_read_u32(intc, "xlnx,kind-of-intr", &irqc->intr_mask);
 	if (ret < 0) {
-		pr_err("%s: unable to read xlnx,kind-of-intr\n", __func__);
-		return ret;
+		pr_warn("irq-xilinx: unable to read xlnx,kind-of-intr\n");
+		irqc->intr_mask = 0;
 	}
 
-	if (intr_mask >> nr_irq)
-		pr_warn("%s: mismatch in kind-of-intr param\n", __func__);
+	if (irqc->intr_mask >> nr_irq)
+		pr_warn("irq-xilinx: mismatch in kind-of-intr param\n");
 
-	pr_info("%s: num_irq=%d, edge=0x%x\n",
-		intc->full_name, nr_irq, intr_mask);
+	pr_info("irq-xilinx: %s: num_irq=%d, edge=0x%x\n",
+		intc->full_name, nr_irq, irqc->intr_mask);
 
-	write_fn = intc_write32;
-	read_fn = intc_read32;
 
 	/*
 	 * Disable all external interrupts until they are
 	 * explicity requested.
 	 */
-	write_fn(0, intc_baseaddr + IER);
+	xintc_write(IER, 0);
 
 	/* Acknowledge any pending interrupts just in case. */
-	write_fn(0xffffffff, intc_baseaddr + IAR);
+	xintc_write(IAR, 0xffffffff);
 
 	/* Turn on the Master Enable. */
-	write_fn(MER_HIE | MER_ME, intc_baseaddr + MER);
-	if (!(read_fn(intc_baseaddr + MER) & (MER_HIE | MER_ME))) {
-		write_fn = intc_write32_be;
-		read_fn = intc_read32_be;
-		write_fn(MER_HIE | MER_ME, intc_baseaddr + MER);
+	xintc_write(MER, MER_HIE | MER_ME);
+	if (!(xintc_read(MER) & (MER_HIE | MER_ME))) {
+		static_branch_enable(&xintc_is_be);
+		xintc_write(MER, MER_HIE | MER_ME);
 	}
 
-	/* Yeah, okay, casting the intr_mask to a void* is butt-ugly, but I'm
-	 * lazy and Michal can clean it up to something nicer when he tests
-	 * and commits this patch.  ~~gcl */
-	root_domain = irq_domain_add_linear(intc, nr_irq, &xintc_irq_domain_ops,
-							(void *)intr_mask);
+	irqc->root_domain = irq_domain_add_linear(intc, nr_irq,
+						  &xintc_irq_domain_ops, irqc);
+	if (!irqc->root_domain) {
+		pr_err("irq-xilinx: Unable to create IRQ domain\n");
+		goto err_alloc;
+	}
 
-	irq_set_default_host(root_domain);
+	if (parent) {
+		irq = irq_of_parse_and_map(intc, 0);
+		if (irq) {
+			irq_set_chained_handler_and_data(irq,
+							 xil_intc_irq_handler,
+							 irqc);
+		} else {
+			pr_err("irq-xilinx: interrupts property not in DT\n");
+			ret = -EINVAL;
+			goto err_alloc;
+		}
+	} else {
+		irq_set_default_host(irqc->root_domain);
+	}
 
 	return 0;
+
+err_alloc:
+	xintc_irqc = NULL;
+	kfree(irqc);
+	return ret;
+
 }
 
-IRQCHIP_DECLARE(xilinx_intc, "xlnx,xps-intc-1.00.a", xilinx_intc_of_init);
+IRQCHIP_DECLARE(xilinx_intc_xps, "xlnx,xps-intc-1.00.a", xilinx_intc_of_init);
+IRQCHIP_DECLARE(xilinx_intc_opb, "xlnx,opb-intc-1.00.c", xilinx_intc_of_init);

drivers/pci/msi.c

@@ -551,14 +551,14 @@ static int populate_msi_sysfs(struct pci_dev *pdev)
 }
 
 static struct msi_desc *
-msi_setup_entry(struct pci_dev *dev, int nvec, bool affinity)
+msi_setup_entry(struct pci_dev *dev, int nvec, const struct irq_affinity *affd)
 {
 	struct cpumask *masks = NULL;
 	struct msi_desc *entry;
 	u16 control;
 
-	if (affinity) {
-		masks = irq_create_affinity_masks(dev->irq_affinity, nvec);
+	if (affd) {
+		masks = irq_create_affinity_masks(nvec, affd);
 		if (!masks)
 			pr_err("Unable to allocate affinity masks, ignoring\n");
 	}
@@ -618,7 +618,8 @@ static int msi_verify_entries(struct pci_dev *dev)
  * an error, and a positive return value indicates the number of interrupts
  * which could have been allocated.
  */
-static int msi_capability_init(struct pci_dev *dev, int nvec, bool affinity)
+static int msi_capability_init(struct pci_dev *dev, int nvec,
+			       const struct irq_affinity *affd)
 {
 	struct msi_desc *entry;
 	int ret;
@@ -626,7 +627,7 @@ static int msi_capability_init(struct pci_dev *dev, int nvec, bool affinity)
 
 	pci_msi_set_enable(dev, 0);	/* Disable MSI during set up */
 
-	entry = msi_setup_entry(dev, nvec, affinity);
+	entry = msi_setup_entry(dev, nvec, affd);
 	if (!entry)
 		return -ENOMEM;
 
@@ -690,14 +691,14 @@ static void __iomem *msix_map_region(struct pci_dev *dev, unsigned nr_entries)
 
 static int msix_setup_entries(struct pci_dev *dev, void __iomem *base,
 			      struct msix_entry *entries, int nvec,
-			      bool affinity)
+			      const struct irq_affinity *affd)
 {
 	struct cpumask *curmsk, *masks = NULL;
 	struct msi_desc *entry;
 	int ret, i;
 
-	if (affinity) {
-		masks = irq_create_affinity_masks(dev->irq_affinity, nvec);
+	if (affd) {
+		masks = irq_create_affinity_masks(nvec, affd);
 		if (!masks)
 			pr_err("Unable to allocate affinity masks, ignoring\n");
 	}
@@ -753,14 +754,14 @@ static void msix_program_entries(struct pci_dev *dev,
  * @dev: pointer to the pci_dev data structure of MSI-X device function
  * @entries: pointer to an array of struct msix_entry entries
  * @nvec: number of @entries
- * @affinity: flag to indicate cpu irq affinity mask should be set
+ * @affd: Optional pointer to enable automatic affinity assignement
  *
  * Setup the MSI-X capability structure of device function with a
  * single MSI-X irq. A return of zero indicates the successful setup of
  * requested MSI-X entries with allocated irqs or non-zero for otherwise.
  **/
 static int msix_capability_init(struct pci_dev *dev, struct msix_entry *entries,
-				int nvec, bool affinity)
+				int nvec, const struct irq_affinity *affd)
 {
 	int ret;
 	u16 control;
@@ -775,7 +776,7 @@ static int msix_capability_init(struct pci_dev *dev, struct msix_entry *entries,
 	if (!base)
 		return -ENOMEM;
 
-	ret = msix_setup_entries(dev, base, entries, nvec, affinity);
+	ret = msix_setup_entries(dev, base, entries, nvec, affd);
 	if (ret)
 		return ret;
 
@@ -956,7 +957,7 @@ int pci_msix_vec_count(struct pci_dev *dev)
 EXPORT_SYMBOL(pci_msix_vec_count);
 
 static int __pci_enable_msix(struct pci_dev *dev, struct msix_entry *entries,
-			     int nvec, bool affinity)
+			     int nvec, const struct irq_affinity *affd)
 {
 	int nr_entries;
 	int i, j;
@@ -988,7 +989,7 @@ static int __pci_enable_msix(struct pci_dev *dev, struct msix_entry *entries,
 		dev_info(&dev->dev, "can't enable MSI-X (MSI IRQ already assigned)\n");
 		return -EINVAL;
 	}
-	return msix_capability_init(dev, entries, nvec, affinity);
+	return msix_capability_init(dev, entries, nvec, affd);
 }
 
 /**
@@ -1008,7 +1009,7 @@ static int __pci_enable_msix(struct pci_dev *dev, struct msix_entry *entries,
  **/
 int pci_enable_msix(struct pci_dev *dev, struct msix_entry *entries, int nvec)
 {
-	return __pci_enable_msix(dev, entries, nvec, false);
+	return __pci_enable_msix(dev, entries, nvec, NULL);
 }
 EXPORT_SYMBOL(pci_enable_msix);
 
@@ -1059,9 +1060,8 @@ int pci_msi_enabled(void)
 EXPORT_SYMBOL(pci_msi_enabled);
 
 static int __pci_enable_msi_range(struct pci_dev *dev, int minvec, int maxvec,
-		unsigned int flags)
+				  const struct irq_affinity *affd)
 {
-	bool affinity = flags & PCI_IRQ_AFFINITY;
 	int nvec;
 	int rc;
 
@@ -1090,14 +1090,13 @@ static int __pci_enable_msi_range(struct pci_dev *dev, int minvec, int maxvec,
 		nvec = maxvec;
 
 	for (;;) {
-		if (affinity) {
-			nvec = irq_calc_affinity_vectors(dev->irq_affinity,
-					nvec);
+		if (affd) {
+			nvec = irq_calc_affinity_vectors(nvec, affd);
 			if (nvec < minvec)
 				return -ENOSPC;
 		}
 
-		rc = msi_capability_init(dev, nvec, affinity);
+		rc = msi_capability_init(dev, nvec, affd);
 		if (rc == 0)
 			return nvec;
 
@@ -1124,29 +1123,27 @@ static int __pci_enable_msi_range(struct pci_dev *dev, int minvec, int maxvec,
  **/
 int pci_enable_msi_range(struct pci_dev *dev, int minvec, int maxvec)
 {
-	return __pci_enable_msi_range(dev, minvec, maxvec, 0);
+	return __pci_enable_msi_range(dev, minvec, maxvec, NULL);
 }
 EXPORT_SYMBOL(pci_enable_msi_range);
 
 static int __pci_enable_msix_range(struct pci_dev *dev,
-		struct msix_entry *entries, int minvec, int maxvec,
-		unsigned int flags)
+				   struct msix_entry *entries, int minvec,
+				   int maxvec, const struct irq_affinity *affd)
 {
-	bool affinity = flags & PCI_IRQ_AFFINITY;
 	int rc, nvec = maxvec;
 
 	if (maxvec < minvec)
 		return -ERANGE;
 
 	for (;;) {
-		if (affinity) {
-			nvec = irq_calc_affinity_vectors(dev->irq_affinity,
-					nvec);
+		if (affd) {
+			nvec = irq_calc_affinity_vectors(nvec, affd);
 			if (nvec < minvec)
 				return -ENOSPC;
 		}
 
-		rc = __pci_enable_msix(dev, entries, nvec, affinity);
+		rc = __pci_enable_msix(dev, entries, nvec, affd);
 		if (rc == 0)
 			return nvec;
 
@@ -1177,16 +1174,17 @@ static int __pci_enable_msix_range(struct pci_dev *dev,
 int pci_enable_msix_range(struct pci_dev *dev, struct msix_entry *entries,
 		int minvec, int maxvec)
 {
-	return __pci_enable_msix_range(dev, entries, minvec, maxvec, 0);
+	return __pci_enable_msix_range(dev, entries, minvec, maxvec, NULL);
 }
 EXPORT_SYMBOL(pci_enable_msix_range);
 
 /**
- * pci_alloc_irq_vectors - allocate multiple IRQs for a device
+ * pci_alloc_irq_vectors_affinity - allocate multiple IRQs for a device
  * @dev:		PCI device to operate on
  * @min_vecs:		minimum number of vectors required (must be >= 1)
  * @max_vecs:		maximum (desired) number of vectors
  * @flags:		flags or quirks for the allocation
+ * @affd:		optional description of the affinity requirements
  *
  * Allocate up to @max_vecs interrupt vectors for @dev, using MSI-X or MSI
  * vectors if available, and fall back to a single legacy vector
@@ -1198,20 +1196,30 @@ EXPORT_SYMBOL(pci_enable_msix_range);
  * To get the Linux IRQ number used for a vector that can be passed to
  * request_irq() use the pci_irq_vector() helper.
  */
-int pci_alloc_irq_vectors(struct pci_dev *dev, unsigned int min_vecs,
-		unsigned int max_vecs, unsigned int flags)
+int pci_alloc_irq_vectors_affinity(struct pci_dev *dev, unsigned int min_vecs,
+				   unsigned int max_vecs, unsigned int flags,
+				   const struct irq_affinity *affd)
 {
+	static const struct irq_affinity msi_default_affd;
 	int vecs = -ENOSPC;
 
+	if (flags & PCI_IRQ_AFFINITY) {
+		if (!affd)
+			affd = &msi_default_affd;
+	} else {
+		if (WARN_ON(affd))
+			affd = NULL;
+	}
+
 	if (flags & PCI_IRQ_MSIX) {
 		vecs = __pci_enable_msix_range(dev, NULL, min_vecs, max_vecs,
-				flags);
+				affd);
 		if (vecs > 0)
 			return vecs;
 	}
 
 	if (flags & PCI_IRQ_MSI) {
-		vecs = __pci_enable_msi_range(dev, min_vecs, max_vecs, flags);
+		vecs = __pci_enable_msi_range(dev, min_vecs, max_vecs, affd);
 		if (vecs > 0)
 			return vecs;
 	}
@@ -1224,7 +1232,7 @@ int pci_alloc_irq_vectors(struct pci_dev *dev, unsigned int min_vecs,
 
 	return vecs;
 }
-EXPORT_SYMBOL(pci_alloc_irq_vectors);
+EXPORT_SYMBOL(pci_alloc_irq_vectors_affinity);
 
 /**
  * pci_free_irq_vectors - free previously allocated IRQs for a device

include/linux/interrupt.h

@@ -232,6 +232,18 @@ struct irq_affinity_notify {
 	void (*release)(struct kref *ref);
 };
 
+/**
+ * struct irq_affinity - Description for automatic irq affinity assignements
+ * @pre_vectors:	Don't apply affinity to @pre_vectors at beginning of
+ *			the MSI(-X) vector space
+ * @post_vectors:	Don't apply affinity to @post_vectors at end of
+ *			the MSI(-X) vector space
+ */
+struct irq_affinity {
+	int	pre_vectors;
+	int	post_vectors;
+};
+
 #if defined(CONFIG_SMP)
 
 extern cpumask_var_t irq_default_affinity;
@@ -278,8 +290,8 @@ extern int irq_set_affinity_hint(unsigned int irq, const struct cpumask *m);
 extern int
 irq_set_affinity_notifier(unsigned int irq, struct irq_affinity_notify *notify);
 
-struct cpumask *irq_create_affinity_masks(const struct cpumask *affinity, int nvec);
-int irq_calc_affinity_vectors(const struct cpumask *affinity, int maxvec);
+struct cpumask *irq_create_affinity_masks(int nvec, const struct irq_affinity *affd);
+int irq_calc_affinity_vectors(int maxvec, const struct irq_affinity *affd);
 
 #else /* CONFIG_SMP */
 
@@ -313,13 +325,13 @@ irq_set_affinity_notifier(unsigned int irq, struct irq_affinity_notify *notify)
 }
 
 static inline struct cpumask *
-irq_create_affinity_masks(const struct cpumask *affinity, int nvec)
+irq_create_affinity_masks(int nvec, const struct irq_affinity *affd)
 {
 	return NULL;
 }
 
 static inline int
-irq_calc_affinity_vectors(const struct cpumask *affinity, int maxvec)
+irq_calc_affinity_vectors(int maxvec, const struct irq_affinity *affd)
 {
 	return maxvec;
 }

include/linux/irqchip/arm-gic-v3.h

@@ -239,7 +239,7 @@
 #define GITS_TYPER_PTA			(1UL << 19)
 #define GITS_TYPER_HWCOLLCNT_SHIFT	24
 
-#define GITS_CBASER_VALID			(1UL << 63)
+#define GITS_CBASER_VALID			(1ULL << 63)
 #define GITS_CBASER_SHAREABILITY_SHIFT		(10)
 #define GITS_CBASER_INNER_CACHEABILITY_SHIFT	(59)
 #define GITS_CBASER_OUTER_CACHEABILITY_SHIFT	(53)
@@ -265,7 +265,7 @@
 
 #define GITS_BASER_NR_REGS		8
 
-#define GITS_BASER_VALID			(1UL << 63)
+#define GITS_BASER_VALID			(1ULL << 63)
 #define GITS_BASER_INDIRECT			(1ULL << 62)
 
 #define GITS_BASER_INNER_CACHEABILITY_SHIFT	(59)

include/linux/pci.h

@@ -244,6 +244,7 @@ struct pci_cap_saved_state {
 	struct pci_cap_saved_data cap;
 };
 
+struct irq_affinity;
 struct pcie_link_state;
 struct pci_vpd;
 struct pci_sriov;
@@ -332,7 +333,6 @@ struct pci_dev {
 	 * directly, use the values stored here. They might be different!
 	 */
 	unsigned int	irq;
-	struct cpumask	*irq_affinity;
 	struct resource resource[DEVICE_COUNT_RESOURCE]; /* I/O and memory regions + expansion ROMs */
 
 	bool match_driver;		/* Skip attaching driver */
@@ -1310,8 +1310,10 @@ static inline int pci_enable_msix_exact(struct pci_dev *dev,
 		return rc;
 	return 0;
 }
-int pci_alloc_irq_vectors(struct pci_dev *dev, unsigned int min_vecs,
-		unsigned int max_vecs, unsigned int flags);
+int pci_alloc_irq_vectors_affinity(struct pci_dev *dev, unsigned int min_vecs,
+				   unsigned int max_vecs, unsigned int flags,
+				   const struct irq_affinity *affd);
+
 void pci_free_irq_vectors(struct pci_dev *dev);
 int pci_irq_vector(struct pci_dev *dev, unsigned int nr);
 const struct cpumask *pci_irq_get_affinity(struct pci_dev *pdev, int vec);
@@ -1339,14 +1341,17 @@ static inline int pci_enable_msix_range(struct pci_dev *dev,
 static inline int pci_enable_msix_exact(struct pci_dev *dev,
 		      struct msix_entry *entries, int nvec)
 { return -ENOSYS; }
-static inline int pci_alloc_irq_vectors(struct pci_dev *dev,
-		unsigned int min_vecs, unsigned int max_vecs,
-		unsigned int flags)
+
+static inline int
+pci_alloc_irq_vectors_affinity(struct pci_dev *dev, unsigned int min_vecs,
+			       unsigned int max_vecs, unsigned int flags,
+			       const struct irq_affinity *aff_desc)
 {
 	if (min_vecs > 1)
 		return -EINVAL;
 	return 1;
 }
+
 static inline void pci_free_irq_vectors(struct pci_dev *dev)
 {
 }
@@ -1364,6 +1369,14 @@ static inline const struct cpumask *pci_irq_get_affinity(struct pci_dev *pdev,
 }
 #endif
 
+static inline int
+pci_alloc_irq_vectors(struct pci_dev *dev, unsigned int min_vecs,
+		      unsigned int max_vecs, unsigned int flags)
+{
+	return pci_alloc_irq_vectors_affinity(dev, min_vecs, max_vecs, flags,
+					      NULL);
+}
+
 #ifdef CONFIG_PCIEPORTBUS
 extern bool pcie_ports_disabled;
 extern bool pcie_ports_auto;

kernel/irq/affinity.c

@@ -51,16 +51,17 @@ static int get_nodes_in_cpumask(const struct cpumask *mask, nodemask_t *nodemsk)
 
 /**
  * irq_create_affinity_masks - Create affinity masks for multiqueue spreading
- * @affinity:		The affinity mask to spread. If NULL cpu_online_mask
- *			is used
- * @nvecs:		The number of vectors
+ * @nvecs:	The total number of vectors
+ * @affd:	Description of the affinity requirements
  *
  * Returns the masks pointer or NULL if allocation failed.
  */
-struct cpumask *irq_create_affinity_masks(const struct cpumask *affinity,
-					  int nvec)
+struct cpumask *
+irq_create_affinity_masks(int nvecs, const struct irq_affinity *affd)
 {
-	int n, nodes, vecs_per_node, cpus_per_vec, extra_vecs, curvec = 0;
+	int n, nodes, vecs_per_node, cpus_per_vec, extra_vecs, curvec;
+	int affv = nvecs - affd->pre_vectors - affd->post_vectors;
+	int last_affv = affv + affd->pre_vectors;
 	nodemask_t nodemsk = NODE_MASK_NONE;
 	struct cpumask *masks;
 	cpumask_var_t nmsk;
@@ -68,46 +69,47 @@ struct cpumask *irq_create_affinity_masks(const struct cpumask *affinity,
 	if (!zalloc_cpumask_var(&nmsk, GFP_KERNEL))
 		return NULL;
 
-	masks = kzalloc(nvec * sizeof(*masks), GFP_KERNEL);
+	masks = kcalloc(nvecs, sizeof(*masks), GFP_KERNEL);
 	if (!masks)
 		goto out;
 
+	/* Fill out vectors at the beginning that don't need affinity */
+	for (curvec = 0; curvec < affd->pre_vectors; curvec++)
+		cpumask_copy(masks + curvec, irq_default_affinity);
+
 	/* Stabilize the cpumasks */
 	get_online_cpus();
-	/* If the supplied affinity mask is NULL, use cpu online mask */
-	if (!affinity)
-		affinity = cpu_online_mask;
-
-	nodes = get_nodes_in_cpumask(affinity, &nodemsk);
+	nodes = get_nodes_in_cpumask(cpu_online_mask, &nodemsk);
 
 	/*
 	 * If the number of nodes in the mask is less than or equal the
 	 * number of vectors we just spread the vectors across the nodes.
 	 */
-	if (nvec <= nodes) {
+	if (affv <= nodes) {
 		for_each_node_mask(n, nodemsk) {
 			cpumask_copy(masks + curvec, cpumask_of_node(n));
-			if (++curvec == nvec)
+			if (++curvec == last_affv)
 				break;
 		}
-		goto outonl;
+		goto done;
 	}
 
 	/* Spread the vectors per node */
-	vecs_per_node = nvec / nodes;
+	vecs_per_node = affv / nodes;
 	/* Account for rounding errors */
-	extra_vecs = nvec - (nodes * vecs_per_node);
+	extra_vecs = affv - (nodes * vecs_per_node);
 
 	for_each_node_mask(n, nodemsk) {
 		int ncpus, v, vecs_to_assign = vecs_per_node;
 
 		/* Get the cpus on this node which are in the mask */
-		cpumask_and(nmsk, affinity, cpumask_of_node(n));
+		cpumask_and(nmsk, cpu_online_mask, cpumask_of_node(n));
 
 		/* Calculate the number of cpus per vector */
 		ncpus = cpumask_weight(nmsk);
 
-		for (v = 0; curvec < nvec && v < vecs_to_assign; curvec++, v++) {
+		for (v = 0; curvec < last_affv && v < vecs_to_assign;
+		     curvec++, v++) {
 			cpus_per_vec = ncpus / vecs_to_assign;
 
 			/* Account for extra vectors to compensate rounding errors */
@@ -119,36 +121,36 @@ struct cpumask *irq_create_affinity_masks(const struct cpumask *affinity,
 			irq_spread_init_one(masks + curvec, nmsk, cpus_per_vec);
 		}
 
-		if (curvec >= nvec)
+		if (curvec >= last_affv)
 			break;
 	}
 
-outonl:
+done:
 	put_online_cpus();
+
+	/* Fill out vectors at the end that don't need affinity */
+	for (; curvec < nvecs; curvec++)
+		cpumask_copy(masks + curvec, irq_default_affinity);
 out:
 	free_cpumask_var(nmsk);
 	return masks;
 }
 
 /**
- * irq_calc_affinity_vectors - Calculate to optimal number of vectors for a given affinity mask
- * @affinity:		The affinity mask to spread. If NULL cpu_online_mask
- *			is used
- * @maxvec:		The maximum number of vectors available
+ * irq_calc_affinity_vectors - Calculate the optimal number of vectors
+ * @maxvec:	The maximum number of vectors available
+ * @affd:	Description of the affinity requirements
  */
-int irq_calc_affinity_vectors(const struct cpumask *affinity, int maxvec)
+int irq_calc_affinity_vectors(int maxvec, const struct irq_affinity *affd)
 {
-	int cpus, ret;
+	int resv = affd->pre_vectors + affd->post_vectors;
+	int vecs = maxvec - resv;
+	int cpus;
 
 	/* Stabilize the cpumasks */
 	get_online_cpus();
-	/* If the supplied affinity mask is NULL, use cpu online mask */
-	if (!affinity)
-		affinity = cpu_online_mask;
-
-	cpus = cpumask_weight(affinity);
-	ret = (cpus < maxvec) ? cpus : maxvec;
-
+	cpus = cpumask_weight(cpu_online_mask);
 	put_online_cpus();
-	return ret;
+
+	return min(cpus, vecs) + resv;
 }

kernel/irq/msi.c

@@ -14,9 +14,7 @@
 #include <linux/irq.h>
 #include <linux/irqdomain.h>
 #include <linux/msi.h>
-
-/* Temparory solution for building, will be removed later */
-#include <linux/pci.h>
+#include <linux/slab.h>
 
 /**
  * alloc_msi_entry - Allocate an initialize msi_entry