ARM: perf_event: allow platform-specific interrupt handler

index 8ccea01..7544ce6 100644
@@ -12,11 +12,25 @@
 #ifndef __ARM_PMU_H__
 #define __ARM_PMU_H__
 
+#include <linux/interrupt.h>
+
 enum arm_pmu_type {
 	ARM_PMU_DEVICE_CPU	= 0,
 	ARM_NUM_PMU_DEVICES,
 };
 
+/*
+ * struct arm_pmu_platdata - ARM PMU platform data
+ *
+ * @handle_irq: an optional handler which will be called from the interrupt and
+ * passed the address of the low level handler, and can be used to implement
+ * any platform specific handling before or after calling it.
+ */
+struct arm_pmu_platdata {
+	irqreturn_t (*handle_irq)(int irq, void *dev,
+				  irq_handler_t pmu_handler);
+};
+
 #ifdef CONFIG_CPU_HAS_PMU
 
 /**

index 5efa264..3239511 100644
@@ -377,9 +377,18 @@ validate_group(struct perf_event *event)
 	return 0;
 }
 
+static irqreturn_t armpmu_platform_irq(int irq, void *dev)
+{
+	struct arm_pmu_platdata *plat = dev_get_platdata(&pmu_device->dev);
+
+	return plat->handle_irq(irq, dev, armpmu->handle_irq);
+}
+
 static int
 armpmu_reserve_hardware(void)
 {
+	struct arm_pmu_platdata *plat;
+	irq_handler_t handle_irq;
 	int i, err = -ENODEV, irq;
 
 	pmu_device = reserve_pmu(ARM_PMU_DEVICE_CPU);

@@ -390,6 +399,12 @@ armpmu_reserve_hardware(void)
 
 	init_pmu(ARM_PMU_DEVICE_CPU);
 
+	plat = dev_get_platdata(&pmu_device->dev);
+	if (plat && plat->handle_irq)
+		handle_irq = armpmu_platform_irq;
+	else
+		handle_irq = armpmu->handle_irq;
+
 	if (pmu_device->num_resources < 1) {
 		pr_err("no irqs for PMUs defined\n");
 		return -ENODEV;

@@ -400,7 +415,7 @@ armpmu_reserve_hardware(void)
 		if (irq < 0)
 			continue;
 
-		err = request_irq(irq, armpmu->handle_irq,
+		err = request_irq(irq, handle_irq,
 				  IRQF_DISABLED | IRQF_NOBALANCING,
 				  "armpmu", NULL);
 		if (err) {

index 706bc95..516126c 100644
@@ -12,12 +12,14 @@
 #include <linux/init.h>
 #include <linux/device.h>
 #include <linux/amba/bus.h>
+#include <linux/interrupt.h>
 #include <linux/irq.h>
 #include <linux/gpio.h>
 #include <linux/platform_device.h>
 #include <linux/io.h>
 
 #include <asm/mach/map.h>
+#include <asm/pmu.h>
 #include <mach/hardware.h>
 #include <mach/setup.h>
 #include <mach/devices.h>

@@ -26,10 +28,6 @@
 #include "devices-db8500.h"
 #include "ste-dma40-db8500.h"
 
-static struct platform_device *platform_devs[] __initdata = {
-	&u8500_dma40_device,
-};
-
 /* minimum static i/o mapping required to boot U8500 platforms */
 static struct map_desc u8500_uart_io_desc[] __initdata = {
 	__IO_DEV_DESC(U8500_UART0_BASE, SZ_4K),

@@ -91,6 +89,51 @@ void __init u8500_map_io(void)
 		iotable_init(u8500_v2_io_desc, ARRAY_SIZE(u8500_v2_io_desc));
 }
 
+static struct resource db8500_pmu_resources[] = {
+	[0] = {
+		.start		= IRQ_DB8500_PMU,
+		.end		= IRQ_DB8500_PMU,
+		.flags		= IORESOURCE_IRQ,
+	},
+};
+
+/*
+ * The PMU IRQ lines of two cores are wired together into a single interrupt.
+ * Bounce the interrupt to the other core if it's not ours.
+ */
+static irqreturn_t db8500_pmu_handler(int irq, void *dev, irq_handler_t handler)
+{
+	irqreturn_t ret = handler(irq, dev);
+	int other = !smp_processor_id();
+
+	if (ret == IRQ_NONE && cpu_online(other))
+		irq_set_affinity(irq, cpumask_of(other));
+
+	/*
+	 * We should be able to get away with the amount of IRQ_NONEs we give,
+	 * while still having the spurious IRQ detection code kick in if the
+	 * interrupt really starts hitting spuriously.
+	 */
+	return ret;
+}
+
+static struct arm_pmu_platdata db8500_pmu_platdata = {
+	.handle_irq		= db8500_pmu_handler,
+};
+
+static struct platform_device db8500_pmu_device = {
+	.name			= "arm-pmu",
+	.id			= ARM_PMU_DEVICE_CPU,
+	.num_resources		= ARRAY_SIZE(db8500_pmu_resources),
+	.resource		= db8500_pmu_resources,
+	.dev.platform_data	= &db8500_pmu_platdata,
+};
+
+static struct platform_device *platform_devs[] __initdata = {
+	&u8500_dma40_device,
+	&db8500_pmu_device,
+};
+
 static resource_size_t __initdata db8500_gpio_base[] = {
 	U8500_GPIOBANK0_BASE,
 	U8500_GPIOBANK1_BASE,

index 9def30b..512f55f 100644
@@ -145,7 +145,7 @@ gic_set_cpu(struct irq_data *d, const struct cpumask *mask_val, bool force)
 {
        void __iomem *reg = gic_dist_base(d) + GIC_DIST_TARGET + (gic_irq(d) & ~3);
        unsigned int shift = (d->irq % 4) * 8;
-       unsigned int cpu = cpumask_first(mask_val);
+       unsigned int cpu_map, cpu = cpumask_first(mask_val);
        u32 val;
        struct irq_desc *desc;
 

@@ -155,9 +155,19 @@ gic_set_cpu(struct irq_data *d, const struct cpumask *mask_val, bool force)
                spin_unlock(&irq_controller_lock);
                return -EINVAL;
        }
+
        d->node = cpu;
+
+       if (CHECK_IRQ_PER_CPU(desc->status)) {
+               cpu_map = 0;
+               for_each_cpu(cpu, mask_val)
+                       cpu_map |= 1 << (cpu + shift);
+       } else {
+               cpu_map = 1 << (cpu + shift);
+       }
+
        val = readl(reg) & ~(0xff << shift);
-       val |= 1 << (cpu + shift);
+       val |= cpu_map;
        writel(val, reg);
        spin_unlock(&irq_controller_lock);