io_apic.c

	 * IOAPIC RTE aswell as interrupt-remapping table entry).
	 * As this gets called during crash dump, keep this simple for now.
	 */
	if (ioapic_i8259.pin != -1 && !intr_remapping_enabled) {
		struct IO_APIC_route_entry entry;

		memset(&entry, 0, sizeof(entry));
		entry.mask            = 0; /* Enabled */
		entry.trigger         = 0; /* Edge */
		entry.irr             = 0;
		entry.polarity        = 0; /* High */
		entry.delivery_status = 0;
		entry.dest_mode       = 0; /* Physical */
		entry.delivery_mode   = dest_ExtINT; /* ExtInt */
		entry.vector          = 0;
		entry.dest            = read_apic_id();

		/*
		 * Add it to the IO-APIC irq-routing table:
		 */
		ioapic_write_entry(ioapic_i8259.apic, ioapic_i8259.pin, entry);
	}

	/*
	 * Use virtual wire A mode when interrupt remapping is enabled.
	 */
	if (cpu_has_apic || apic_from_smp_config())
		disconnect_bsp_APIC(!intr_remapping_enabled &&
				ioapic_i8259.pin != -1);
}

#ifdef CONFIG_X86_32
/*
 * function to set the IO-APIC physical IDs based on the
 * values stored in the MPC table.
 *
 * by Matt Domsch <Matt_Domsch@dell.com>  Tue Dec 21 12:25:05 CST 1999
 */

void __init setup_ioapic_ids_from_mpc(void)
{
	union IO_APIC_reg_00 reg_00;
	physid_mask_t phys_id_present_map;
	int apic_id;
	int i;
	unsigned char old_id;
	unsigned long flags;

	if (acpi_ioapic)
		return;
	/*
	 * Don't check I/O APIC IDs for xAPIC systems.  They have
	 * no meaning without the serial APIC bus.
	 */
	if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
		|| APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
		return;
	/*
	 * This is broken; anything with a real cpu count has to
	 * circumvent this idiocy regardless.
	 */
	apic->ioapic_phys_id_map(&phys_cpu_present_map, &phys_id_present_map);

	/*
	 * Set the IOAPIC ID to the value stored in the MPC table.
	 */
	for (apic_id = 0; apic_id < nr_ioapics; apic_id++) {

		/* Read the register 0 value */
		raw_spin_lock_irqsave(&ioapic_lock, flags);
		reg_00.raw = io_apic_read(apic_id, 0);
		raw_spin_unlock_irqrestore(&ioapic_lock, flags);

		old_id = mp_ioapics[apic_id].apicid;

		if (mp_ioapics[apic_id].apicid >= get_physical_broadcast()) {
			printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n",
				apic_id, mp_ioapics[apic_id].apicid);
			printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n",
				reg_00.bits.ID);
			mp_ioapics[apic_id].apicid = reg_00.bits.ID;
		}

		/*
		 * Sanity check, is the ID really free? 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(&phys_id_present_map,
					mp_ioapics[apic_id].apicid)) {
			printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n",
				apic_id, mp_ioapics[apic_id].apicid);
			for (i = 0; i < get_physical_broadcast(); i++)
				if (!physid_isset(i, phys_id_present_map))
					break;
			if (i >= get_physical_broadcast())
				panic("Max APIC ID exceeded!\n");
			printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n",
				i);
			physid_set(i, phys_id_present_map);
			mp_ioapics[apic_id].apicid = i;
		} else {
			physid_mask_t tmp;
			apic->apicid_to_cpu_present(mp_ioapics[apic_id].apicid, &tmp);
			apic_printk(APIC_VERBOSE, "Setting %d in the "
					"phys_id_present_map\n",
					mp_ioapics[apic_id].apicid);
			physids_or(phys_id_present_map, phys_id_present_map, tmp);
		}


		/*
		 * We need to adjust the IRQ routing table
		 * if the ID changed.
		 */
		if (old_id != mp_ioapics[apic_id].apicid)
			for (i = 0; i < mp_irq_entries; i++)
				if (mp_irqs[i].dstapic == old_id)
					mp_irqs[i].dstapic
						= mp_ioapics[apic_id].apicid;

		/*
		 * Read the right value from the MPC table and
		 * write it into the ID register.
		 */
		apic_printk(APIC_VERBOSE, KERN_INFO
			"...changing IO-APIC physical APIC ID to %d ...",
			mp_ioapics[apic_id].apicid);

		reg_00.bits.ID = mp_ioapics[apic_id].apicid;
		raw_spin_lock_irqsave(&ioapic_lock, flags);
		io_apic_write(apic_id, 0, reg_00.raw);
		raw_spin_unlock_irqrestore(&ioapic_lock, flags);

		/*
		 * Sanity check
		 */
		raw_spin_lock_irqsave(&ioapic_lock, flags);
		reg_00.raw = io_apic_read(apic_id, 0);
		raw_spin_unlock_irqrestore(&ioapic_lock, flags);
		if (reg_00.bits.ID != mp_ioapics[apic_id].apicid)
			printk("could not set ID!\n");
		else
			apic_printk(APIC_VERBOSE, " ok.\n");
	}
}
#endif

int no_timer_check __initdata;

static int __init notimercheck(char *s)
{
	no_timer_check = 1;
	return 1;
}
__setup("no_timer_check", notimercheck);

/*
 * There is a nasty bug in some older SMP boards, their mptable lies
 * about the timer IRQ. We do the following to work around the situation:
 *
 *	- timer IRQ defaults to IO-APIC IRQ
 *	- if this function detects that timer IRQs are defunct, then we fall
 *	  back to ISA timer IRQs
 */
static int __init timer_irq_works(void)
{
	unsigned long t1 = jiffies;
	unsigned long flags;

	if (no_timer_check)
		return 1;

	local_save_flags(flags);
	local_irq_enable();
	/* Let ten ticks pass... */
	mdelay((10 * 1000) / HZ);
	local_irq_restore(flags);

	/*
	 * Expect a few ticks at least, to be sure some possible
	 * glue logic does not lock up after one or two first
	 * ticks in a non-ExtINT mode.  Also the local APIC
	 * might have cached one ExtINT interrupt.  Finally, at
	 * least one tick may be lost due to delays.
	 */

	/* jiffies wrap? */
	if (time_after(jiffies, t1 + 4))
		return 1;
	return 0;
}

/*
 * In the SMP+IOAPIC case it might happen that there are an unspecified
 * number of pending IRQ events unhandled. These cases are very rare,
 * so we 'resend' these IRQs via IPIs, to the same CPU. It's much
 * better to do it this way as thus we do not have to be aware of
 * 'pending' interrupts in the IRQ path, except at this point.
 */
/*
 * Edge triggered needs to resend any interrupt
 * that was delayed but this is now handled in the device
 * independent code.
 */

/*
 * Starting up a edge-triggered IO-APIC interrupt is
 * nasty - we need to make sure that we get the edge.
 * If it is already asserted for some reason, we need
 * return 1 to indicate that is was pending.
 *
 * This is not complete - we should be able to fake
 * an edge even if it isn't on the 8259A...
 */

static unsigned int startup_ioapic_irq(struct irq_data *data)
{
	int was_pending = 0, irq = data->irq;
	unsigned long flags;

	raw_spin_lock_irqsave(&ioapic_lock, flags);
	if (irq < legacy_pic->nr_legacy_irqs) {
		legacy_pic->mask(irq);
		if (legacy_pic->irq_pending(irq))
			was_pending = 1;
	}
	__unmask_ioapic(data->chip_data);
	raw_spin_unlock_irqrestore(&ioapic_lock, flags);

	return was_pending;
}

static int ioapic_retrigger_irq(struct irq_data *data)
{
	struct irq_cfg *cfg = data->chip_data;
	unsigned long flags;

	raw_spin_lock_irqsave(&vector_lock, flags);
	apic->send_IPI_mask(cpumask_of(cpumask_first(cfg->domain)), cfg->vector);
	raw_spin_unlock_irqrestore(&vector_lock, flags);

	return 1;
}

/*
 * Level and edge triggered IO-APIC interrupts need different handling,
 * so we use two separate IRQ descriptors. Edge triggered IRQs can be
 * handled with the level-triggered descriptor, but that one has slightly
 * more overhead. Level-triggered interrupts cannot be handled with the
 * edge-triggered handler, without risking IRQ storms and other ugly
 * races.
 */

#ifdef CONFIG_SMP
void send_cleanup_vector(struct irq_cfg *cfg)
{
	cpumask_var_t cleanup_mask;

	if (unlikely(!alloc_cpumask_var(&cleanup_mask, GFP_ATOMIC))) {
		unsigned int i;
		for_each_cpu_and(i, cfg->old_domain, cpu_online_mask)
			apic->send_IPI_mask(cpumask_of(i), IRQ_MOVE_CLEANUP_VECTOR);
	} else {
		cpumask_and(cleanup_mask, cfg->old_domain, cpu_online_mask);
		apic->send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
		free_cpumask_var(cleanup_mask);
	}
	cfg->move_in_progress = 0;
}

static void __target_IO_APIC_irq(unsigned int irq, unsigned int dest, struct irq_cfg *cfg)
{
	int apic, pin;
	struct irq_pin_list *entry;
	u8 vector = cfg->vector;

	for_each_irq_pin(entry, cfg->irq_2_pin) {
		unsigned int reg;

		apic = entry->apic;
		pin = entry->pin;
		/*
		 * With interrupt-remapping, destination information comes
		 * from interrupt-remapping table entry.
		 */
		if (!irq_remapped(irq))
			io_apic_write(apic, 0x11 + pin*2, dest);
		reg = io_apic_read(apic, 0x10 + pin*2);
		reg &= ~IO_APIC_REDIR_VECTOR_MASK;
		reg |= vector;
		io_apic_modify(apic, 0x10 + pin*2, reg);
	}
}

/*
 * Either sets desc->affinity to a valid value, and returns
 * ->cpu_mask_to_apicid of that in dest_id, or returns -1 and
 * leaves desc->affinity untouched.
 */
unsigned int
set_desc_affinity(struct irq_desc *desc, const struct cpumask *mask,
		  unsigned int *dest_id)
{
	struct irq_cfg *cfg;
	unsigned int irq;

	if (!cpumask_intersects(mask, cpu_online_mask))
		return -1;

	irq = desc->irq;
	cfg = get_irq_desc_chip_data(desc);
	if (assign_irq_vector(irq, cfg, mask))
		return -1;

	cpumask_copy(desc->affinity, mask);

	*dest_id = apic->cpu_mask_to_apicid_and(desc->affinity, cfg->domain);
	return 0;
}

static int
set_ioapic_affinity_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
{
	struct irq_cfg *cfg;
	unsigned long flags;
	unsigned int dest;
	unsigned int irq;
	int ret = -1;

	irq = desc->irq;
	cfg = get_irq_desc_chip_data(desc);

	raw_spin_lock_irqsave(&ioapic_lock, flags);
	ret = set_desc_affinity(desc, mask, &dest);
	if (!ret) {
		/* Only the high 8 bits are valid. */
		dest = SET_APIC_LOGICAL_ID(dest);
		__target_IO_APIC_irq(irq, dest, cfg);
	}
	raw_spin_unlock_irqrestore(&ioapic_lock, flags);

	return ret;
}

static int
set_ioapic_affinity_irq(unsigned int irq, const struct cpumask *mask)
{
	struct irq_desc *desc;

	desc = irq_to_desc(irq);

	return set_ioapic_affinity_irq_desc(desc, mask);
}

#ifdef CONFIG_INTR_REMAP

/*
 * Migrate the IO-APIC irq in the presence of intr-remapping.
 *
 * For both level and edge triggered, irq migration is a simple atomic
 * update(of vector and cpu destination) of IRTE and flush the hardware cache.
 *
 * For level triggered, we eliminate the io-apic RTE modification (with the
 * updated vector information), by using a virtual vector (io-apic pin number).
 * Real vector that is used for interrupting cpu will be coming from
 * the interrupt-remapping table entry.
 */
static int
migrate_ioapic_irq_desc(struct irq_desc *desc, const struct cpumask *mask)
{
	struct irq_cfg *cfg;
	struct irte irte;
	unsigned int dest;
	unsigned int irq;
	int ret = -1;

	if (!cpumask_intersects(mask, cpu_online_mask))
		return ret;

	irq = desc->irq;
	if (get_irte(irq, &irte))
		return ret;

	cfg = get_irq_desc_chip_data(desc);
	if (assign_irq_vector(irq, cfg, mask))
		return ret;

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

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

	/*
	 * Modified the IRTE and flushes the Interrupt entry cache.
	 */
	modify_irte(irq, &irte);

	if (cfg->move_in_progress)
		send_cleanup_vector(cfg);

	cpumask_copy(desc->affinity, mask);

	return 0;
}

/*
 * Migrates the IRQ destination in the process context.
 */
static int set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
					    const struct cpumask *mask)
{
	return migrate_ioapic_irq_desc(desc, mask);
}
static int set_ir_ioapic_affinity_irq(unsigned int irq,
				       const struct cpumask *mask)
{
	struct irq_desc *desc = irq_to_desc(irq);

	return set_ir_ioapic_affinity_irq_desc(desc, mask);
}
#else
static inline int set_ir_ioapic_affinity_irq_desc(struct irq_desc *desc,
						   const struct cpumask *mask)
{
	return 0;
}
#endif

asmlinkage void smp_irq_move_cleanup_interrupt(void)
{
	unsigned vector, me;

	ack_APIC_irq();
	exit_idle();
	irq_enter();

	me = smp_processor_id();
	for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; vector++) {
		unsigned int irq;
		unsigned int irr;
		struct irq_desc *desc;
		struct irq_cfg *cfg;
		irq = __get_cpu_var(vector_irq)[vector];

		if (irq == -1)
			continue;

		desc = irq_to_desc(irq);
		if (!desc)
			continue;

		cfg = irq_cfg(irq);
		raw_spin_lock(&desc->lock);

		/*
		 * Check if the irq migration is in progress. If so, we
		 * haven't received the cleanup request yet for this irq.
		 */
		if (cfg->move_in_progress)
			goto unlock;

		if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
			goto unlock;

		irr = apic_read(APIC_IRR + (vector / 32 * 0x10));
		/*
		 * Check if the vector that needs to be cleanedup is
		 * registered at the cpu's IRR. If so, then this is not
		 * the best time to clean it up. Lets clean it up in the
		 * next attempt by sending another IRQ_MOVE_CLEANUP_VECTOR
		 * to myself.
		 */
		if (irr  & (1 << (vector % 32))) {
			apic->send_IPI_self(IRQ_MOVE_CLEANUP_VECTOR);
			goto unlock;
		}
		__get_cpu_var(vector_irq)[vector] = -1;
unlock:
		raw_spin_unlock(&desc->lock);
	}

	irq_exit();
}

static void __irq_complete_move(struct irq_cfg *cfg, unsigned vector)
{
	unsigned me;

	if (likely(!cfg->move_in_progress))
		return;

	me = smp_processor_id();

	if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain))
		send_cleanup_vector(cfg);
}

static void irq_complete_move(struct irq_cfg *cfg)
{
	__irq_complete_move(cfg, ~get_irq_regs()->orig_ax);
}

void irq_force_complete_move(int irq)
{
	struct irq_cfg *cfg = get_irq_chip_data(irq);

	if (!cfg)
		return;

	__irq_complete_move(cfg, cfg->vector);
}
#else
static inline void irq_complete_move(struct irq_cfg *cfg) { }
#endif

static void ack_apic_edge(struct irq_data *data)
{
	irq_complete_move(data->chip_data);
	move_native_irq(data->irq);
	ack_APIC_irq();
}

atomic_t irq_mis_count;

/*
 * IO-APIC versions below 0x20 don't support EOI register.
 * For the record, here is the information about various versions:
 *     0Xh     82489DX
 *     1Xh     I/OAPIC or I/O(x)APIC which are not PCI 2.2 Compliant
 *     2Xh     I/O(x)APIC which is PCI 2.2 Compliant
 *     30h-FFh Reserved
 *
 * Some of the Intel ICH Specs (ICH2 to ICH5) documents the io-apic
 * version as 0x2. This is an error with documentation and these ICH chips
 * use io-apic's of version 0x20.
 *
 * For IO-APIC's with EOI register, we use that to do an explicit EOI.
 * Otherwise, we simulate the EOI message manually by changing the trigger
 * mode to edge and then back to level, with RTE being masked during this.
*/
static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg)
{
	struct irq_pin_list *entry;
	unsigned long flags;

	raw_spin_lock_irqsave(&ioapic_lock, flags);
	for_each_irq_pin(entry, cfg->irq_2_pin) {
		if (mp_ioapics[entry->apic].apicver >= 0x20) {
			/*
			 * Intr-remapping uses pin number as the virtual vector
			 * in the RTE. Actual vector is programmed in
			 * intr-remapping table entry. Hence for the io-apic
			 * EOI we use the pin number.
			 */
			if (irq_remapped(irq))
				io_apic_eoi(entry->apic, entry->pin);
			else
				io_apic_eoi(entry->apic, cfg->vector);
		} else {
			__mask_and_edge_IO_APIC_irq(entry);
			__unmask_and_level_IO_APIC_irq(entry);
		}
	}
	raw_spin_unlock_irqrestore(&ioapic_lock, flags);
}

static void ack_apic_level(struct irq_data *data)
{
	struct irq_cfg *cfg = data->chip_data;
	int i, do_unmask_irq = 0, irq = data->irq;
	struct irq_desc *desc = irq_to_desc(irq);
	unsigned long v;

	irq_complete_move(cfg);
#ifdef CONFIG_GENERIC_PENDING_IRQ
	/* If we are moving the irq we need to mask it */
	if (unlikely(desc->status & IRQ_MOVE_PENDING)) {
		do_unmask_irq = 1;
		mask_ioapic(cfg);
	}
#endif

	/*
	 * It appears there is an erratum which affects at least version 0x11
	 * of I/O APIC (that's the 82093AA and cores integrated into various
	 * chipsets).  Under certain conditions a level-triggered interrupt is
	 * erroneously delivered as edge-triggered one but the respective IRR
	 * bit gets set nevertheless.  As a result the I/O unit expects an EOI
	 * message but it will never arrive and further interrupts are blocked
	 * from the source.  The exact reason is so far unknown, but the
	 * phenomenon was observed when two consecutive interrupt requests
	 * from a given source get delivered to the same CPU and the source is
	 * temporarily disabled in between.
	 *
	 * A workaround is to simulate an EOI message manually.  We achieve it
	 * by setting the trigger mode to edge and then to level when the edge
	 * trigger mode gets detected in the TMR of a local APIC for a
	 * level-triggered interrupt.  We mask the source for the time of the
	 * operation to prevent an edge-triggered interrupt escaping meanwhile.
	 * The idea is from Manfred Spraul.  --macro
	 *
	 * Also in the case when cpu goes offline, fixup_irqs() will forward
	 * any unhandled interrupt on the offlined cpu to the new cpu
	 * destination that is handling the corresponding interrupt. This
	 * interrupt forwarding is done via IPI's. Hence, in this case also
	 * level-triggered io-apic interrupt will be seen as an edge
	 * interrupt in the IRR. And we can't rely on the cpu's EOI
	 * to be broadcasted to the IO-APIC's which will clear the remoteIRR
	 * corresponding to the level-triggered interrupt. Hence on IO-APIC's
	 * supporting EOI register, we do an explicit EOI to clear the
	 * remote IRR and on IO-APIC's which don't have an EOI register,
	 * we use the above logic (mask+edge followed by unmask+level) from
	 * Manfred Spraul to clear the remote IRR.
	 */
	i = cfg->vector;
	v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1));

	/*
	 * We must acknowledge the irq before we move it or the acknowledge will
	 * not propagate properly.
	 */
	ack_APIC_irq();

	/*
	 * Tail end of clearing remote IRR bit (either by delivering the EOI
	 * message via io-apic EOI register write or simulating it using
	 * mask+edge followed by unnask+level logic) manually when the
	 * level triggered interrupt is seen as the edge triggered interrupt
	 * at the cpu.
	 */
	if (!(v & (1 << (i & 0x1f)))) {
		atomic_inc(&irq_mis_count);

		eoi_ioapic_irq(irq, cfg);
	}

	/* Now we can move and renable the irq */
	if (unlikely(do_unmask_irq)) {
		/* Only migrate the irq if the ack has been received.
		 *
		 * On rare occasions the broadcast level triggered ack gets
		 * delayed going to ioapics, and if we reprogram the
		 * vector while Remote IRR is still set the irq will never
		 * fire again.
		 *
		 * To prevent this scenario we read the Remote IRR bit
		 * of the ioapic.  This has two effects.
		 * - On any sane system the read of the ioapic will
		 *   flush writes (and acks) going to the ioapic from
		 *   this cpu.
		 * - We get to see if the ACK has actually been delivered.
		 *
		 * Based on failed experiments of reprogramming the
		 * ioapic entry from outside of irq context starting
		 * with masking the ioapic entry and then polling until
		 * Remote IRR was clear before reprogramming the
		 * ioapic I don't trust the Remote IRR bit to be
		 * completey accurate.
		 *
		 * However there appears to be no other way to plug
		 * this race, so if the Remote IRR bit is not
		 * accurate and is causing problems then it is a hardware bug
		 * and you can go talk to the chipset vendor about it.
		 */
		if (!io_apic_level_ack_pending(cfg))
			move_masked_irq(irq);
		unmask_ioapic(cfg);
	}
}

#ifdef CONFIG_INTR_REMAP
static void ir_ack_apic_edge(struct irq_data *data)
{
	ack_APIC_irq();
}

static void ir_ack_apic_level(struct irq_data *data)
{
	ack_APIC_irq();
	eoi_ioapic_irq(data->irq, data->chip_data);
}
#endif /* CONFIG_INTR_REMAP */

static struct irq_chip ioapic_chip __read_mostly = {
	.name		= "IO-APIC",
	.irq_startup	= startup_ioapic_irq,
	.irq_mask	= mask_ioapic_irq,
	.irq_unmask	= unmask_ioapic_irq,
	.irq_ack	= ack_apic_edge,
	.irq_eoi	= ack_apic_level,
#ifdef CONFIG_SMP
	.set_affinity	= set_ioapic_affinity_irq,
#endif
	.irq_retrigger	= ioapic_retrigger_irq,
};

static struct irq_chip ir_ioapic_chip __read_mostly = {
	.name		= "IR-IO-APIC",
	.irq_startup	= startup_ioapic_irq,
	.irq_mask	= mask_ioapic_irq,
	.irq_unmask	= unmask_ioapic_irq,
#ifdef CONFIG_INTR_REMAP
	.irq_ack	= ir_ack_apic_edge,
	.irq_eoi	= ir_ack_apic_level,
#ifdef CONFIG_SMP
	.set_affinity	= set_ir_ioapic_affinity_irq,
#endif
#endif
	.irq_retrigger	= ioapic_retrigger_irq,
};

static inline void init_IO_APIC_traps(void)
{
	int irq;
	struct irq_desc *desc;
	struct irq_cfg *cfg;

	/*
	 * NOTE! The local APIC isn't very good at handling
	 * multiple interrupts at the same interrupt level.
	 * As the interrupt level is determined by taking the
	 * vector number and shifting that right by 4, we
	 * want to spread these out a bit so that they don't
	 * all fall in the same interrupt level.
	 *
	 * Also, we've got to be careful not to trash gate
	 * 0x80, because int 0x80 is hm, kind of importantish. ;)
	 */
	for_each_irq_desc(irq, desc) {
		cfg = get_irq_desc_chip_data(desc);
		if (IO_APIC_IRQ(irq) && cfg && !cfg->vector) {
			/*
			 * Hmm.. We don't have an entry for this,
			 * so default to an old-fashioned 8259
			 * interrupt if we can..
			 */
			if (irq < legacy_pic->nr_legacy_irqs)
				legacy_pic->make_irq(irq);
			else
				/* Strange. Oh, well.. */
				desc->chip = &no_irq_chip;
		}
	}
}

/*
 * The local APIC irq-chip implementation:
 */

static void mask_lapic_irq(struct irq_data *data)
{
	unsigned long v;

	v = apic_read(APIC_LVT0);
	apic_write(APIC_LVT0, v | APIC_LVT_MASKED);
}

static void unmask_lapic_irq(struct irq_data *data)
{
	unsigned long v;

	v = apic_read(APIC_LVT0);
	apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED);
}

static void ack_lapic_irq(struct irq_data *data)
{
	ack_APIC_irq();
}

static struct irq_chip lapic_chip __read_mostly = {
	.name		= "local-APIC",
	.irq_mask	= mask_lapic_irq,
	.irq_unmask	= unmask_lapic_irq,
	.irq_ack	= ack_lapic_irq,
};

static void lapic_register_intr(int irq, struct irq_desc *desc)
{
	desc->status &= ~IRQ_LEVEL;
	set_irq_chip_and_handler_name(irq, &lapic_chip, handle_edge_irq,
				      "edge");
}

static void __init setup_nmi(void)
{
	/*
	 * Dirty trick to enable the NMI watchdog ...
	 * We put the 8259A master into AEOI mode and
	 * unmask on all local APICs LVT0 as NMI.
	 *
	 * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire')
	 * is from Maciej W. Rozycki - so we do not have to EOI from
	 * the NMI handler or the timer interrupt.
	 */
	apic_printk(APIC_VERBOSE, KERN_INFO "activating NMI Watchdog ...");

	enable_NMI_through_LVT0();

	apic_printk(APIC_VERBOSE, " done.\n");
}

/*
 * This looks a bit hackish but it's about the only one way of sending
 * a few INTA cycles to 8259As and any associated glue logic.  ICR does
 * not support the ExtINT mode, unfortunately.  We need to send these
 * cycles as some i82489DX-based boards have glue logic that keeps the
 * 8259A interrupt line asserted until INTA.  --macro
 */
static inline void __init unlock_ExtINT_logic(void)
{
	int apic, pin, i;
	struct IO_APIC_route_entry entry0, entry1;
	unsigned char save_control, save_freq_select;

	pin  = find_isa_irq_pin(8, mp_INT);
	if (pin == -1) {
		WARN_ON_ONCE(1);
		return;
	}
	apic = find_isa_irq_apic(8, mp_INT);
	if (apic == -1) {
		WARN_ON_ONCE(1);
		return;
	}

	entry0 = ioapic_read_entry(apic, pin);
	clear_IO_APIC_pin(apic, pin);

	memset(&entry1, 0, sizeof(entry1));

	entry1.dest_mode = 0;			/* physical delivery */
	entry1.mask = 0;			/* unmask IRQ now */
	entry1.dest = hard_smp_processor_id();
	entry1.delivery_mode = dest_ExtINT;
	entry1.polarity = entry0.polarity;
	entry1.trigger = 0;
	entry1.vector = 0;

	ioapic_write_entry(apic, pin, entry1);

	save_control = CMOS_READ(RTC_CONTROL);
	save_freq_select = CMOS_READ(RTC_FREQ_SELECT);
	CMOS_WRITE((save_freq_select & ~RTC_RATE_SELECT) | 0x6,
		   RTC_FREQ_SELECT);
	CMOS_WRITE(save_control | RTC_PIE, RTC_CONTROL);

	i = 100;
	while (i-- > 0) {
		mdelay(10);
		if ((CMOS_READ(RTC_INTR_FLAGS) & RTC_PF) == RTC_PF)
			i -= 10;
	}

	CMOS_WRITE(save_control, RTC_CONTROL);
	CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
	clear_IO_APIC_pin(apic, pin);

	ioapic_write_entry(apic, pin, entry0);
}

static int disable_timer_pin_1 __initdata;
/* Actually the next is obsolete, but keep it for paranoid reasons -AK */
static int __init disable_timer_pin_setup(char *arg)
{
	disable_timer_pin_1 = 1;
	return 0;
}
early_param("disable_timer_pin_1", disable_timer_pin_setup);

int timer_through_8259 __initdata;

/*
 * This code may look a bit paranoid, but it's supposed to cooperate with
 * a wide range of boards and BIOS bugs.  Fortunately only the timer IRQ
 * is so screwy.  Thanks to Brian Perkins for testing/hacking this beast
 * fanatically on his truly buggy board.
 *
 * FIXME: really need to revamp this for all platforms.
 */
static inline void __init check_timer(void)
{
	struct irq_desc *desc = irq_to_desc(0);
	struct irq_cfg *cfg = get_irq_desc_chip_data(desc);
	int node = cpu_to_node(0);
	int apic1, pin1, apic2, pin2;
	unsigned long flags;
	int no_pin1 = 0;

	local_irq_save(flags);

	/*
	 * get/set the timer IRQ vector:
	 */
	legacy_pic->mask(0);
	assign_irq_vector(0, cfg, apic->target_cpus());

	/*
	 * As IRQ0 is to be enabled in the 8259A, the virtual
	 * wire has to be disabled in the local APIC.  Also
	 * timer interrupts need to be acknowledged manually in
	 * the 8259A for the i82489DX when using the NMI
	 * watchdog as that APIC treats NMIs as level-triggered.
	 * The AEOI mode will finish them in the 8259A
	 * automatically.
	 */
	apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
	legacy_pic->init(1);
#ifdef CONFIG_X86_32
	{
		unsigned int ver;

		ver = apic_read(APIC_LVR);
		ver = GET_APIC_VERSION(ver);
		timer_ack = (nmi_watchdog == NMI_IO_APIC && !APIC_INTEGRATED(ver));
	}
#endif

	pin1  = find_isa_irq_pin(0, mp_INT);
	apic1 = find_isa_irq_apic(0, mp_INT);
	pin2  = ioapic_i8259.pin;
	apic2 = ioapic_i8259.apic;

	apic_printk(APIC_QUIET, KERN_INFO "..TIMER: vector=0x%02X "
		    "apic1=%d pin1=%d apic2=%d pin2=%d\n",
		    cfg->vector, apic1, pin1, apic2, pin2);

	/*
	 * Some BIOS writers are clueless and report the ExtINTA
	 * I/O APIC input from the cascaded 8259A as the timer
	 * interrupt input.  So just in case, if only one pin
	 * was found above, try it both directly and through the
	 * 8259A.
	 */
	if (pin1 == -1) {
		if (intr_remapping_enabled)
			panic("BIOS bug: timer not connected to IO-APIC");
		pin1 = pin2;
		apic1 = apic2;
		no_pin1 = 1;
	} else if (pin2 == -1) {
		pin2 = pin1;
		apic2 = apic1;
	}

	if (pin1 != -1) {
		/*
		 * Ok, does IRQ0 through the IOAPIC work?
		 */
		if (no_pin1) {
			add_pin_to_irq_node(cfg, node, apic1, pin1);
			setup_timer_IRQ0_pin(apic1, pin1, cfg->vector);
		} else {
			/* for edge trigger, setup_IO_APIC_irq already
			 * leave it unmasked.
			 * so only need to unmask if it is level-trigger
			 * do we really have level trigger timer?
			 */
			int idx;
			idx = find_irq_entry(apic1, pin1, mp_INT);
			if (idx != -1 && irq_trigger(idx))
				unmask_ioapic(cfg);
		}
		if (timer_irq_works()) {
			if (nmi_watchdog == NMI_IO_APIC) {
				setup_nmi();
				legacy_pic->unmask(0);
			}
			if (disable_timer_pin_1 > 0)
				clear_IO_APIC_pin(0, pin1);
			goto out;
		}
		if (intr_remapping_enabled)
			panic("timer doesn't work through Interrupt-remapped IO-APIC");
		local_irq_disable();
		clear_IO_APIC_pin(apic1, pin1);
		if (!no_pin1)
			apic_printk(APIC_QUIET, KERN_ERR "..MP-BIOS bug: "
				    "8254 timer not connected to IO-APIC\n");

		apic_printk(APIC_QUIET, KERN_INFO "...trying to set up timer "
			    "(IRQ0) through the 8259A ...\n");
		apic_printk(APIC_QUIET, KERN_INFO
			    "..... (found apic %d pin %d) ...\n", apic2, pin2);
		/*
		 * legacy devices should be connected to IO APIC #0
		 */
		replace_pin_at_irq_node(cfg, node, apic1, pin1, apic2, pin2);
		setup_timer_IRQ0_pin(apic2, pin2, cfg->vector);
		legacy_pic->unmask(0);
		if (timer_irq_works()) {
			apic_printk(APIC_QUIET, KERN_INFO "....... works.\n");
			timer_through_8259 = 1;
			if (nmi_watchdog == NMI_IO_APIC) {
				legacy_pic->mask(0);
				setup_nmi();
				legacy_pic->unmask(0);
			}
			goto out;