io_apic.c

	apic_printk(APIC_QUIET, KERN_INFO
		    "...trying to set up timer as ExtINT IRQ...\n");

	init_8259A(0);
	make_8259A_irq(0);
	apic_write(APIC_LVT0, APIC_DM_EXTINT);

	unlock_ExtINT_logic();

	if (timer_irq_works()) {
		apic_printk(APIC_QUIET, KERN_INFO "..... works.\n");
		goto out;
	}
	local_irq_disable();
	apic_printk(APIC_QUIET, KERN_INFO "..... failed :(.\n");
	panic("IO-APIC + timer doesn't work!  Boot with apic=debug and send a "
		"report.  Then try booting with the 'noapic' option.\n");
out:
	local_irq_restore(flags);
}

/*
 * Traditionally ISA IRQ2 is the cascade IRQ, and is not available
 * to devices.  However there may be an I/O APIC pin available for
 * this interrupt regardless.  The pin may be left unconnected, but
 * typically it will be reused as an ExtINT cascade interrupt for
 * the master 8259A.  In the MPS case such a pin will normally be
 * reported as an ExtINT interrupt in the MP table.  With ACPI
 * there is no provision for ExtINT interrupts, and in the absence
 * of an override it would be treated as an ordinary ISA I/O APIC
 * interrupt, that is edge-triggered and unmasked by default.  We
 * used to do this, but it caused problems on some systems because
 * of the NMI watchdog and sometimes IRQ0 of the 8254 timer using
 * the same ExtINT cascade interrupt to drive the local APIC of the
 * bootstrap processor.  Therefore we refrain from routing IRQ2 to
 * the I/O APIC in all cases now.  No actual device should request
 * it anyway.  --macro
 */
#define PIC_IRQS	(1UL << PIC_CASCADE_IR)

void __init setup_IO_APIC(void)
{

	/*
	 * calling enable_IO_APIC() is moved to setup_local_APIC for BP
	 */
	io_apic_irqs = nr_legacy_irqs ? ~PIC_IRQS : ~0UL;

	apic_printk(APIC_VERBOSE, "ENABLING IO-APIC IRQs\n");
	/*
         * Set up IO-APIC IRQ routing.
         */
	x86_init.mpparse.setup_ioapic_ids();

	sync_Arb_IDs();
	setup_IO_APIC_irqs();
	init_IO_APIC_traps();
	if (nr_legacy_irqs)
		check_timer();
}

/*
 *      Called after all the initialization is done. If we didnt find any
 *      APIC bugs then we can allow the modify fast path
 */

static int __init io_apic_bug_finalize(void)
{
	if (sis_apic_bug == -1)
		sis_apic_bug = 0;
	return 0;
}

late_initcall(io_apic_bug_finalize);

struct sysfs_ioapic_data {
	struct sys_device dev;
	struct IO_APIC_route_entry entry[0];
};
static struct sysfs_ioapic_data * mp_ioapic_data[MAX_IO_APICS];

static int ioapic_suspend(struct sys_device *dev, pm_message_t state)
{
	struct IO_APIC_route_entry *entry;
	struct sysfs_ioapic_data *data;
	int i;

	data = container_of(dev, struct sysfs_ioapic_data, dev);
	entry = data->entry;
	for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ )
		*entry = ioapic_read_entry(dev->id, i);

	return 0;
}

static int ioapic_resume(struct sys_device *dev)
{
	struct IO_APIC_route_entry *entry;
	struct sysfs_ioapic_data *data;
	unsigned long flags;
	union IO_APIC_reg_00 reg_00;
	int i;

	data = container_of(dev, struct sysfs_ioapic_data, dev);
	entry = data->entry;

	spin_lock_irqsave(&ioapic_lock, flags);
	reg_00.raw = io_apic_read(dev->id, 0);
	if (reg_00.bits.ID != mp_ioapics[dev->id].apicid) {
		reg_00.bits.ID = mp_ioapics[dev->id].apicid;
		io_apic_write(dev->id, 0, reg_00.raw);
	}
	spin_unlock_irqrestore(&ioapic_lock, flags);
	for (i = 0; i < nr_ioapic_registers[dev->id]; i++)
		ioapic_write_entry(dev->id, i, entry[i]);

	return 0;
}

static struct sysdev_class ioapic_sysdev_class = {
	.name = "ioapic",
	.suspend = ioapic_suspend,
	.resume = ioapic_resume,
};

static int __init ioapic_init_sysfs(void)
{
	struct sys_device * dev;
	int i, size, error;

	error = sysdev_class_register(&ioapic_sysdev_class);
	if (error)
		return error;

	for (i = 0; i < nr_ioapics; i++ ) {
		size = sizeof(struct sys_device) + nr_ioapic_registers[i]
			* sizeof(struct IO_APIC_route_entry);
		mp_ioapic_data[i] = kzalloc(size, GFP_KERNEL);
		if (!mp_ioapic_data[i]) {
			printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i);
			continue;
		}
		dev = &mp_ioapic_data[i]->dev;
		dev->id = i;
		dev->cls = &ioapic_sysdev_class;
		error = sysdev_register(dev);
		if (error) {
			kfree(mp_ioapic_data[i]);
			mp_ioapic_data[i] = NULL;
			printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i);
			continue;
		}
	}

	return 0;
}

device_initcall(ioapic_init_sysfs);

/*
 * Dynamic irq allocate and deallocation
 */
unsigned int create_irq_nr(unsigned int irq_want, int node)
{
	/* Allocate an unused irq */
	unsigned int irq;
	unsigned int new;
	unsigned long flags;
	struct irq_cfg *cfg_new = NULL;
	struct irq_desc *desc_new = NULL;

	irq = 0;
	if (irq_want < nr_irqs_gsi)
		irq_want = nr_irqs_gsi;

	spin_lock_irqsave(&vector_lock, flags);
	for (new = irq_want; new < nr_irqs; new++) {
		desc_new = irq_to_desc_alloc_node(new, node);
		if (!desc_new) {
			printk(KERN_INFO "can not get irq_desc for %d\n", new);
			continue;
		}
		cfg_new = desc_new->chip_data;

		if (cfg_new->vector != 0)
			continue;

		desc_new = move_irq_desc(desc_new, node);
		cfg_new = desc_new->chip_data;

		if (__assign_irq_vector(new, cfg_new, apic->target_cpus()) == 0)
			irq = new;
		break;
	}
	spin_unlock_irqrestore(&vector_lock, flags);

	if (irq > 0) {
		dynamic_irq_init(irq);
		/* restore it, in case dynamic_irq_init clear it */
		if (desc_new)
			desc_new->chip_data = cfg_new;
	}
	return irq;
}

int create_irq(void)
{
	int node = cpu_to_node(boot_cpu_id);
	unsigned int irq_want;
	int irq;

	irq_want = nr_irqs_gsi;
	irq = create_irq_nr(irq_want, node);

	if (irq == 0)
		irq = -1;

	return irq;
}

void destroy_irq(unsigned int irq)
{
	unsigned long flags;
	struct irq_cfg *cfg;
	struct irq_desc *desc;

	/* store it, in case dynamic_irq_cleanup clear it */
	desc = irq_to_desc(irq);
	cfg = desc->chip_data;
	dynamic_irq_cleanup(irq);
	/* connect back irq_cfg */
	desc->chip_data = cfg;

	free_irte(irq);
	spin_lock_irqsave(&vector_lock, flags);
	__clear_irq_vector(irq, cfg);
	spin_unlock_irqrestore(&vector_lock, flags);
}

/*
 * MSI message composition
 */
#ifdef CONFIG_PCI_MSI
static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_msg *msg)
{
	struct irq_cfg *cfg;
	int err;
	unsigned dest;

	if (disable_apic)
		return -ENXIO;

	cfg = irq_cfg(irq);
	err = assign_irq_vector(irq, cfg, apic->target_cpus());
	if (err)
		return err;

	dest = apic->cpu_mask_to_apicid_and(cfg->domain, apic->target_cpus());

	if (irq_remapped(irq)) {
		struct irte irte;
		int ir_index;
		u16 sub_handle;

		ir_index = map_irq_to_irte_handle(irq, &sub_handle);
		BUG_ON(ir_index == -1);

		memset (&irte, 0, sizeof(irte));

		irte.present = 1;
		irte.dst_mode = apic->irq_dest_mode;
		irte.trigger_mode = 0; /* edge */
		irte.dlvry_mode = apic->irq_delivery_mode;
		irte.vector = cfg->vector;
		irte.dest_id = IRTE_DEST(dest);

		/* Set source-id of interrupt request */
		set_msi_sid(&irte, pdev);

		modify_irte(irq, &irte);

		msg->address_hi = MSI_ADDR_BASE_HI;
		msg->data = sub_handle;
		msg->address_lo = MSI_ADDR_BASE_LO | MSI_ADDR_IR_EXT_INT |
				  MSI_ADDR_IR_SHV |
				  MSI_ADDR_IR_INDEX1(ir_index) |
				  MSI_ADDR_IR_INDEX2(ir_index);
	} else {
		if (x2apic_enabled())
			msg->address_hi = MSI_ADDR_BASE_HI |
					  MSI_ADDR_EXT_DEST_ID(dest);
		else
			msg->address_hi = MSI_ADDR_BASE_HI;

		msg->address_lo =
			MSI_ADDR_BASE_LO |
			((apic->irq_dest_mode == 0) ?
				MSI_ADDR_DEST_MODE_PHYSICAL:
				MSI_ADDR_DEST_MODE_LOGICAL) |
			((apic->irq_delivery_mode != dest_LowestPrio) ?
				MSI_ADDR_REDIRECTION_CPU:
				MSI_ADDR_REDIRECTION_LOWPRI) |
			MSI_ADDR_DEST_ID(dest);

		msg->data =
			MSI_DATA_TRIGGER_EDGE |
			MSI_DATA_LEVEL_ASSERT |
			((apic->irq_delivery_mode != dest_LowestPrio) ?
				MSI_DATA_DELIVERY_FIXED:
				MSI_DATA_DELIVERY_LOWPRI) |
			MSI_DATA_VECTOR(cfg->vector);
	}
	return err;
}

#ifdef CONFIG_SMP
static int set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
{
	struct irq_desc *desc = irq_to_desc(irq);
	struct irq_cfg *cfg;
	struct msi_msg msg;
	unsigned int dest;

	dest = set_desc_affinity(desc, mask);
	if (dest == BAD_APICID)
		return -1;

	cfg = desc->chip_data;

	read_msi_msg_desc(desc, &msg);

	msg.data &= ~MSI_DATA_VECTOR_MASK;
	msg.data |= MSI_DATA_VECTOR(cfg->vector);
	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
	msg.address_lo |= MSI_ADDR_DEST_ID(dest);

	write_msi_msg_desc(desc, &msg);

	return 0;
}
#ifdef CONFIG_INTR_REMAP
/*
 * Migrate the MSI irq to another cpumask. This migration is
 * done in the process context using interrupt-remapping hardware.
 */
static int
ir_set_msi_irq_affinity(unsigned int irq, const struct cpumask *mask)
{
	struct irq_desc *desc = irq_to_desc(irq);
	struct irq_cfg *cfg = desc->chip_data;
	unsigned int dest;
	struct irte irte;

	if (get_irte(irq, &irte))
		return -1;

	dest = set_desc_affinity(desc, mask);
	if (dest == BAD_APICID)
		return -1;

	irte.vector = cfg->vector;
	irte.dest_id = IRTE_DEST(dest);

	/*
	 * atomically update the IRTE with the new destination and vector.
	 */
	modify_irte(irq, &irte);

	/*
	 * After this point, all the interrupts will start arriving
	 * at the new destination. So, time to cleanup the previous
	 * vector allocation.
	 */
	if (cfg->move_in_progress)
		send_cleanup_vector(cfg);

	return 0;
}

#endif
#endif /* CONFIG_SMP */

/*
 * IRQ Chip for MSI PCI/PCI-X/PCI-Express Devices,
 * which implement the MSI or MSI-X Capability Structure.
 */
static struct irq_chip msi_chip = {
	.name		= "PCI-MSI",
	.unmask		= unmask_msi_irq,
	.mask		= mask_msi_irq,
	.ack		= ack_apic_edge,
#ifdef CONFIG_SMP
	.set_affinity	= set_msi_irq_affinity,
#endif
	.retrigger	= ioapic_retrigger_irq,
};

static struct irq_chip msi_ir_chip = {
	.name		= "IR-PCI-MSI",
	.unmask		= unmask_msi_irq,
	.mask		= mask_msi_irq,
#ifdef CONFIG_INTR_REMAP
	.ack		= ir_ack_apic_edge,
#ifdef CONFIG_SMP
	.set_affinity	= ir_set_msi_irq_affinity,
#endif
#endif
	.retrigger	= ioapic_retrigger_irq,
};

/*
 * Map the PCI dev to the corresponding remapping hardware unit
 * and allocate 'nvec' consecutive interrupt-remapping table entries
 * in it.
 */
static int msi_alloc_irte(struct pci_dev *dev, int irq, int nvec)
{
	struct intel_iommu *iommu;
	int index;

	iommu = map_dev_to_ir(dev);
	if (!iommu) {
		printk(KERN_ERR
		       "Unable to map PCI %s to iommu\n", pci_name(dev));
		return -ENOENT;
	}

	index = alloc_irte(iommu, irq, nvec);
	if (index < 0) {
		printk(KERN_ERR
		       "Unable to allocate %d IRTE for PCI %s\n", nvec,
		       pci_name(dev));
		return -ENOSPC;
	}
	return index;
}

static int setup_msi_irq(struct pci_dev *dev, struct msi_desc *msidesc, int irq)
{
	int ret;
	struct msi_msg msg;

	ret = msi_compose_msg(dev, irq, &msg);
	if (ret < 0)
		return ret;

	set_irq_msi(irq, msidesc);
	write_msi_msg(irq, &msg);

	if (irq_remapped(irq)) {
		struct irq_desc *desc = irq_to_desc(irq);
		/*
		 * irq migration in process context
		 */
		desc->status |= IRQ_MOVE_PCNTXT;
		set_irq_chip_and_handler_name(irq, &msi_ir_chip, handle_edge_irq, "edge");
	} else
		set_irq_chip_and_handler_name(irq, &msi_chip, handle_edge_irq, "edge");

	dev_printk(KERN_DEBUG, &dev->dev, "irq %d for MSI/MSI-X\n", irq);

	return 0;
}

int arch_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
{
	unsigned int irq;
	int ret, sub_handle;
	struct msi_desc *msidesc;
	unsigned int irq_want;
	struct intel_iommu *iommu = NULL;
	int index = 0;
	int node;

	/* x86 doesn't support multiple MSI yet */
	if (type == PCI_CAP_ID_MSI && nvec > 1)
		return 1;

	node = dev_to_node(&dev->dev);
	irq_want = nr_irqs_gsi;
	sub_handle = 0;
	list_for_each_entry(msidesc, &dev->msi_list, list) {
		irq = create_irq_nr(irq_want, node);
		if (irq == 0)
			return -1;
		irq_want = irq + 1;
		if (!intr_remapping_enabled)
			goto no_ir;

		if (!sub_handle) {
			/*
			 * allocate the consecutive block of IRTE's
			 * for 'nvec'
			 */
			index = msi_alloc_irte(dev, irq, nvec);
			if (index < 0) {
				ret = index;
				goto error;
			}
		} else {
			iommu = map_dev_to_ir(dev);
			if (!iommu) {
				ret = -ENOENT;
				goto error;
			}
			/*
			 * setup the mapping between the irq and the IRTE
			 * base index, the sub_handle pointing to the
			 * appropriate interrupt remap table entry.
			 */
			set_irte_irq(irq, iommu, index, sub_handle);
		}
no_ir:
		ret = setup_msi_irq(dev, msidesc, irq);
		if (ret < 0)
			goto error;
		sub_handle++;
	}
	return 0;

error:
	destroy_irq(irq);
	return ret;
}

void arch_teardown_msi_irq(unsigned int irq)
{
	destroy_irq(irq);
}

#if defined (CONFIG_DMAR) || defined (CONFIG_INTR_REMAP)
#ifdef CONFIG_SMP
static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
{
	struct irq_desc *desc = irq_to_desc(irq);
	struct irq_cfg *cfg;
	struct msi_msg msg;
	unsigned int dest;

	dest = set_desc_affinity(desc, mask);
	if (dest == BAD_APICID)
		return -1;

	cfg = desc->chip_data;

	dmar_msi_read(irq, &msg);

	msg.data &= ~MSI_DATA_VECTOR_MASK;
	msg.data |= MSI_DATA_VECTOR(cfg->vector);
	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
	msg.address_lo |= MSI_ADDR_DEST_ID(dest);

	dmar_msi_write(irq, &msg);

	return 0;
}

#endif /* CONFIG_SMP */

static struct irq_chip dmar_msi_type = {
	.name = "DMAR_MSI",
	.unmask = dmar_msi_unmask,
	.mask = dmar_msi_mask,
	.ack = ack_apic_edge,
#ifdef CONFIG_SMP
	.set_affinity = dmar_msi_set_affinity,
#endif
	.retrigger = ioapic_retrigger_irq,
};

int arch_setup_dmar_msi(unsigned int irq)
{
	int ret;
	struct msi_msg msg;

	ret = msi_compose_msg(NULL, irq, &msg);
	if (ret < 0)
		return ret;
	dmar_msi_write(irq, &msg);
	set_irq_chip_and_handler_name(irq, &dmar_msi_type, handle_edge_irq,
		"edge");
	return 0;
}
#endif

#ifdef CONFIG_HPET_TIMER

#ifdef CONFIG_SMP
static int hpet_msi_set_affinity(unsigned int irq, const struct cpumask *mask)
{
	struct irq_desc *desc = irq_to_desc(irq);
	struct irq_cfg *cfg;
	struct msi_msg msg;
	unsigned int dest;

	dest = set_desc_affinity(desc, mask);
	if (dest == BAD_APICID)
		return -1;

	cfg = desc->chip_data;

	hpet_msi_read(irq, &msg);

	msg.data &= ~MSI_DATA_VECTOR_MASK;
	msg.data |= MSI_DATA_VECTOR(cfg->vector);
	msg.address_lo &= ~MSI_ADDR_DEST_ID_MASK;
	msg.address_lo |= MSI_ADDR_DEST_ID(dest);

	hpet_msi_write(irq, &msg);

	return 0;
}

#endif /* CONFIG_SMP */

static struct irq_chip hpet_msi_type = {
	.name = "HPET_MSI",
	.unmask = hpet_msi_unmask,
	.mask = hpet_msi_mask,
	.ack = ack_apic_edge,
#ifdef CONFIG_SMP
	.set_affinity = hpet_msi_set_affinity,
#endif
	.retrigger = ioapic_retrigger_irq,
};

int arch_setup_hpet_msi(unsigned int irq)
{
	int ret;
	struct msi_msg msg;
	struct irq_desc *desc = irq_to_desc(irq);

	ret = msi_compose_msg(NULL, irq, &msg);
	if (ret < 0)
		return ret;

	hpet_msi_write(irq, &msg);
	desc->status |= IRQ_MOVE_PCNTXT;
	set_irq_chip_and_handler_name(irq, &hpet_msi_type, handle_edge_irq,
		"edge");

	return 0;
}
#endif

#endif /* CONFIG_PCI_MSI */
/*
 * Hypertransport interrupt support
 */
#ifdef CONFIG_HT_IRQ

#ifdef CONFIG_SMP

static void target_ht_irq(unsigned int irq, unsigned int dest, u8 vector)
{
	struct ht_irq_msg msg;
	fetch_ht_irq_msg(irq, &msg);

	msg.address_lo &= ~(HT_IRQ_LOW_VECTOR_MASK | HT_IRQ_LOW_DEST_ID_MASK);
	msg.address_hi &= ~(HT_IRQ_HIGH_DEST_ID_MASK);

	msg.address_lo |= HT_IRQ_LOW_VECTOR(vector) | HT_IRQ_LOW_DEST_ID(dest);
	msg.address_hi |= HT_IRQ_HIGH_DEST_ID(dest);

	write_ht_irq_msg(irq, &msg);
}

static int set_ht_irq_affinity(unsigned int irq, const struct cpumask *mask)
{
	struct irq_desc *desc = irq_to_desc(irq);
	struct irq_cfg *cfg;
	unsigned int dest;

	dest = set_desc_affinity(desc, mask);
	if (dest == BAD_APICID)
		return -1;

	cfg = desc->chip_data;

	target_ht_irq(irq, dest, cfg->vector);

	return 0;
}

#endif

static struct irq_chip ht_irq_chip = {
	.name		= "PCI-HT",
	.mask		= mask_ht_irq,
	.unmask		= unmask_ht_irq,
	.ack		= ack_apic_edge,
#ifdef CONFIG_SMP
	.set_affinity	= set_ht_irq_affinity,
#endif
	.retrigger	= ioapic_retrigger_irq,
};

int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev)
{
	struct irq_cfg *cfg;
	int err;

	if (disable_apic)
		return -ENXIO;

	cfg = irq_cfg(irq);
	err = assign_irq_vector(irq, cfg, apic->target_cpus());
	if (!err) {
		struct ht_irq_msg msg;
		unsigned dest;

		dest = apic->cpu_mask_to_apicid_and(cfg->domain,
						    apic->target_cpus());

		msg.address_hi = HT_IRQ_HIGH_DEST_ID(dest);

		msg.address_lo =
			HT_IRQ_LOW_BASE |
			HT_IRQ_LOW_DEST_ID(dest) |
			HT_IRQ_LOW_VECTOR(cfg->vector) |
			((apic->irq_dest_mode == 0) ?
				HT_IRQ_LOW_DM_PHYSICAL :
				HT_IRQ_LOW_DM_LOGICAL) |
			HT_IRQ_LOW_RQEOI_EDGE |
			((apic->irq_delivery_mode != dest_LowestPrio) ?
				HT_IRQ_LOW_MT_FIXED :
				HT_IRQ_LOW_MT_ARBITRATED) |
			HT_IRQ_LOW_IRQ_MASKED;

		write_ht_irq_msg(irq, &msg);

		set_irq_chip_and_handler_name(irq, &ht_irq_chip,
					      handle_edge_irq, "edge");

		dev_printk(KERN_DEBUG, &dev->dev, "irq %d for HT\n", irq);
	}
	return err;
}
#endif /* CONFIG_HT_IRQ */

int __init io_apic_get_redir_entries (int ioapic)
{
	union IO_APIC_reg_01	reg_01;
	unsigned long flags;

	spin_lock_irqsave(&ioapic_lock, flags);
	reg_01.raw = io_apic_read(ioapic, 1);
	spin_unlock_irqrestore(&ioapic_lock, flags);

	return reg_01.bits.entries;
}

void __init probe_nr_irqs_gsi(void)
{
	int nr = 0;

	nr = acpi_probe_gsi();
	if (nr > nr_irqs_gsi) {
		nr_irqs_gsi = nr;
	} else {
		/* for acpi=off or acpi is not compiled in */
		int idx;

		nr = 0;
		for (idx = 0; idx < nr_ioapics; idx++)
			nr += io_apic_get_redir_entries(idx) + 1;

		if (nr > nr_irqs_gsi)
			nr_irqs_gsi = nr;
	}

	printk(KERN_DEBUG "nr_irqs_gsi: %d\n", nr_irqs_gsi);
}

#ifdef CONFIG_SPARSE_IRQ
int __init arch_probe_nr_irqs(void)
{
	int nr;

	if (nr_irqs > (NR_VECTORS * nr_cpu_ids))
		nr_irqs = NR_VECTORS * nr_cpu_ids;

	nr = nr_irqs_gsi + 8 * nr_cpu_ids;
#if defined(CONFIG_PCI_MSI) || defined(CONFIG_HT_IRQ)
	/*
	 * for MSI and HT dyn irq
	 */
	nr += nr_irqs_gsi * 16;
#endif
	if (nr < nr_irqs)
		nr_irqs = nr;

	return 0;
}
#endif

static int __io_apic_set_pci_routing(struct device *dev, int irq,
				struct io_apic_irq_attr *irq_attr)
{
	struct irq_desc *desc;
	struct irq_cfg *cfg;
	int node;
	int ioapic, pin;
	int trigger, polarity;

	ioapic = irq_attr->ioapic;
	if (!IO_APIC_IRQ(irq)) {
		apic_printk(APIC_QUIET,KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n",
			ioapic);
		return -EINVAL;
	}

	if (dev)
		node = dev_to_node(dev);
	else
		node = cpu_to_node(boot_cpu_id);

	desc = irq_to_desc_alloc_node(irq, node);
	if (!desc) {
		printk(KERN_INFO "can not get irq_desc %d\n", irq);
		return 0;
	}

	pin = irq_attr->ioapic_pin;
	trigger = irq_attr->trigger;
	polarity = irq_attr->polarity;

	/*
	 * IRQs < 16 are already in the irq_2_pin[] map
	 */
	if (irq >= nr_legacy_irqs) {
		cfg = desc->chip_data;
		if (add_pin_to_irq_node_nopanic(cfg, node, ioapic, pin)) {
			printk(KERN_INFO "can not add pin %d for irq %d\n",
				pin, irq);
			return 0;
		}
	}

	setup_IO_APIC_irq(ioapic, pin, irq, desc, trigger, polarity);

	return 0;
}

int io_apic_set_pci_routing(struct device *dev, int irq,
				struct io_apic_irq_attr *irq_attr)
{
	int ioapic, pin;
	/*
	 * Avoid pin reprogramming.  PRTs typically include entries
	 * with redundant pin->gsi mappings (but unique PCI devices);
	 * we only program the IOAPIC on the first.
	 */
	ioapic = irq_attr->ioapic;
	pin = irq_attr->ioapic_pin;
	if (test_bit(pin, mp_ioapic_routing[ioapic].pin_programmed)) {
		pr_debug("Pin %d-%d already programmed\n",
			 mp_ioapics[ioapic].apicid, pin);
		return 0;
	}
	set_bit(pin, mp_ioapic_routing[ioapic].pin_programmed);

	return __io_apic_set_pci_routing(dev, irq, irq_attr);
}

u8 __init io_apic_unique_id(u8 id)
{
#ifdef CONFIG_X86_32
	if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) &&
	    !APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
		return io_apic_get_unique_id(nr_ioapics, id);
	else
		return id;
#else
	int i;
	DECLARE_BITMAP(used, 256);

	bitmap_zero(used, 256);
	for (i = 0; i < nr_ioapics; i++) {
		struct mpc_ioapic *ia = &mp_ioapics[i];
		__set_bit(ia->apicid, used);
	}
	if (!test_bit(id, used))
		return id;
	return find_first_zero_bit(used, 256);
#endif
}

#ifdef CONFIG_X86_32
int __init io_apic_get_unique_id(int ioapic, int apic_id)
{
	union IO_APIC_reg_00 reg_00;
	static physid_mask_t apic_id_map = PHYSID_MASK_NONE;
	physid_mask_t tmp;
	unsigned long flags;
	int i = 0;

	/*
	 * The P4 platform supports up to 256 APIC IDs on two separate APIC
	 * buses (one for LAPICs, one for IOAPICs), where predecessors only
	 * supports up to 16 on one shared APIC bus.
	 *
	 * TBD: Expand LAPIC/IOAPIC support on P4-class systems to take full
	 *      advantage of new APIC bus architecture.
	 */

	if (physids_empty(apic_id_map))
		apic->ioapic_phys_id_map(&phys_cpu_present_map, &apic_id_map);

	spin_lock_irqsave(&ioapic_lock, flags);
	reg_00.raw = io_apic_read(ioapic, 0);
	spin_unlock_irqrestore(&ioapic_lock, flags);

	if (apic_id >= get_physical_broadcast()) {
		printk(KERN_WARNING "IOAPIC[%d]: Invalid apic_id %d, trying "
			"%d\n", ioapic, apic_id, reg_00.bits.ID);
		apic_id = reg_00.bits.ID;
	}

	/*
	 * Every APIC in a system must have a unique ID or we get lots of nice
	 * 'stuck on smp_invalidate_needed IPI wait' messages.
	 */
	if (apic->check_apicid_used(&apic_id_map, apic_id)) {

		for (i = 0; i < get_physical_broadcast(); i++) {
			if (!apic->check_apicid_used(&apic_id_map, i))
				break;
		}

		if (i == get_physical_broadcast())
			panic("Max apic_id exceeded!\n");

		printk(KERN_WARNING "IOAPIC[%d]: apic_id %d already used, "
			"trying %d\n", ioapic, apic_id, i);

		apic_id = i;
	}

	apic->apicid_to_cpu_present(apic_id, &tmp);
	physids_or(apic_id_map, apic_id_map, tmp);

	if (reg_00.bits.ID != apic_id) {
		reg_00.bits.ID = apic_id;

		spin_lock_irqsave(&ioapic_lock, flags);
		io_apic_write(ioapic, 0, reg_00.raw);
		reg_00.raw = io_apic_read(ioapic, 0);
		spin_unlock_irqrestore(&ioapic_lock, flags);

		/* Sanity check */
		if (reg_00.bits.ID != apic_id) {
			printk("IOAPIC[%d]: Unable to change apic_id!\n", ioapic);
			return -1;
		}
	}

	apic_printk(APIC_VERBOSE, KERN_INFO
			"IOAPIC[%d]: Assigned apic_id %d\n", ioapic, apic_id);

	return apic_id;
}
#endif

int __init io_apic_get_version(int ioapic)
{
	union IO_APIC_reg_01	reg_01;
	unsigned long flags;

	spin_lock_irqsave(&ioapic_lock, flags);
	reg_01.raw = io_apic_read(ioapic, 1);
	spin_unlock_irqrestore(&ioapic_lock, flags);

	return reg_01.bits.version;
}

int acpi_get_override_irq(int bus_irq, int *trigger, int *polarity)
{
	int i;

	if (skip_ioapic_setup)
		return -1;

	for (i = 0; i < mp_irq_entries; i++)
		if (mp_irqs[i].irqtype == mp_INT &&
		    mp_irqs[i].srcbusirq == bus_irq)
			break;
	if (i >= mp_irq_entries)
		return -1;

	*trigger = irq_trigger(i);
	*polarity = irq_polarity(i);
	return 0;
}

/*
 * This function currently is only a helper for the i386 smp boot process where
 * we need to reprogram the ioredtbls to cater for the cpus which have come online
 * so mask in all cases should simply be apic->target_cpus()
 */