io_apic.c

自己根据lkd和情境分析

		spin_unlock_irqrestore(&ioapic_lock, flags);

		/*
		 * Sanity check
		 */
		spin_lock_irqsave(&ioapic_lock, flags);
		*(int *)&reg_00 = io_apic_read(apic, 0);
		spin_unlock_irqrestore(&ioapic_lock, flags);
		if (reg_00.ID != mp_ioapics[apic].mpc_apicid)
			panic("could not set ID!\n");
		else
			printk(" ok.\n");
	}
}

/*
 * 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 int t1 = jiffies;

	sti();
	/* Let ten ticks pass... */
	mdelay((10 * 1000) / HZ);

	/*
	 * 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.
	 */
	if (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.
 */
#define enable_edge_ioapic_irq unmask_IO_APIC_irq

static void disable_edge_ioapic_irq (unsigned int irq) { /* nothing */ }

/*
 * 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_edge_ioapic_irq(unsigned int irq)
{
	int was_pending = 0;
	unsigned long flags;

	spin_lock_irqsave(&ioapic_lock, flags);
	if (irq < 16) {
		disable_8259A_irq(irq);
		if (i8259A_irq_pending(irq))
			was_pending = 1;
	}
	__unmask_IO_APIC_irq(irq);
	spin_unlock_irqrestore(&ioapic_lock, flags);

	return was_pending;
}

#define shutdown_edge_ioapic_irq	disable_edge_ioapic_irq

/*
 * Once we have recorded IRQ_PENDING already, we can mask the
 * interrupt for real. This prevents IRQ storms from unhandled
 * devices.
 */
static void ack_edge_ioapic_irq(unsigned int irq)
{
	if ((irq_desc[irq].status & (IRQ_PENDING | IRQ_DISABLED))
					== (IRQ_PENDING | IRQ_DISABLED))
		mask_IO_APIC_irq(irq);
	ack_APIC_irq();
}

static void end_edge_ioapic_irq (unsigned int i) { /* nothing */ }


/*
 * Level triggered interrupts can just be masked,
 * and shutting down and starting up the interrupt
 * is the same as enabling and disabling them -- except
 * with a startup need to return a "was pending" value.
 *
 * Level triggered interrupts are special because we
 * do not touch any IO-APIC register while handling
 * them. We ack the APIC in the end-IRQ handler, not
 * in the start-IRQ-handler. Protection against reentrance
 * from the same interrupt is still provided, both by the
 * generic IRQ layer and by the fact that an unacked local
 * APIC does not accept IRQs.
 */
static unsigned int startup_level_ioapic_irq (unsigned int irq)
{
	unmask_IO_APIC_irq(irq);

	return 0; /* don't check for pending */
}

#define shutdown_level_ioapic_irq	mask_IO_APIC_irq
#define enable_level_ioapic_irq		unmask_IO_APIC_irq
#define disable_level_ioapic_irq	mask_IO_APIC_irq

static void end_level_ioapic_irq (unsigned int irq)
{
	unsigned long v;
	int i;

/*
 * 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
 */
	i = IO_APIC_VECTOR(irq);
	v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1));

	ack_APIC_irq();

	if (!(v & (1 << (i & 0x1f)))) {
#ifdef APIC_LOCKUP_DEBUG
		struct irq_pin_list *entry;
#endif

#ifdef APIC_MISMATCH_DEBUG
		atomic_inc(&irq_mis_count);
#endif
		spin_lock(&ioapic_lock);
		__mask_and_edge_IO_APIC_irq(irq);
#ifdef APIC_LOCKUP_DEBUG
		for (entry = irq_2_pin + irq;;) {
			unsigned int reg;

			if (entry->pin == -1)
				break;
			reg = io_apic_read(entry->apic, 0x10 + entry->pin * 2);
			if (reg & 0x00004000)
				printk(KERN_CRIT "Aieee!!!  Remote IRR"
					" still set after unlock!\n");
			if (!entry->next)
				break;
			entry = irq_2_pin + entry->next;
		}
#endif
		__unmask_and_level_IO_APIC_irq(irq);
		spin_unlock(&ioapic_lock);
	}
}

static void mask_and_ack_level_ioapic_irq (unsigned int irq) { /* nothing */ }

static void set_ioapic_affinity (unsigned int irq, unsigned long mask)
{
	unsigned long flags;
	/*
	 * Only the first 8 bits are valid.
	 */
	mask = mask << 24;

	spin_lock_irqsave(&ioapic_lock, flags);
	__DO_ACTION(1, = mask, )
	spin_unlock_irqrestore(&ioapic_lock, flags);
}

/*
 * 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.
 */

static struct hw_interrupt_type ioapic_edge_irq_type = {
	"IO-APIC-edge",
	startup_edge_ioapic_irq,
	shutdown_edge_ioapic_irq,
	enable_edge_ioapic_irq,
	disable_edge_ioapic_irq,
	ack_edge_ioapic_irq,
	end_edge_ioapic_irq,
	set_ioapic_affinity,
};

static struct hw_interrupt_type ioapic_level_irq_type = {
	"IO-APIC-level",
	startup_level_ioapic_irq,
	shutdown_level_ioapic_irq,
	enable_level_ioapic_irq,
	disable_level_ioapic_irq,
	mask_and_ack_level_ioapic_irq,
	end_level_ioapic_irq,
	set_ioapic_affinity,
};

static inline void init_IO_APIC_traps(void)
{
	int irq;

	/*
	 * 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 (irq = 0; irq < NR_IRQS ; irq++) {
		if (IO_APIC_IRQ(irq) && !IO_APIC_VECTOR(irq)) {
			/*
			 * Hmm.. We don't have an entry for this,
			 * so default to an old-fashioned 8259
			 * interrupt if we can..
			 */
			if (irq < 16)
				make_8259A_irq(irq);
			else
				/* Strange. Oh, well.. */
				irq_desc[irq].handler = &no_irq_type;
		}
	}
}

static void enable_lapic_irq (unsigned int irq)
{
	unsigned long v;

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

static void disable_lapic_irq (unsigned int irq)
{
	unsigned long v;

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

static void ack_lapic_irq (unsigned int irq)
{
	ack_APIC_irq();
}

static void end_lapic_irq (unsigned int i) { /* nothing */ }

static struct hw_interrupt_type lapic_irq_type = {
	"local-APIC-edge",
	NULL, /* startup_irq() not used for IRQ0 */
	NULL, /* shutdown_irq() not used for IRQ0 */
	enable_lapic_irq,
	disable_lapic_irq,
	ack_lapic_irq,
	end_lapic_irq
};

static void enable_NMI_through_LVT0 (void * dummy)
{
	unsigned int v, ver;

	ver = apic_read(APIC_LVR);
	ver = GET_APIC_VERSION(ver);
	v = APIC_DM_NMI;			/* unmask and set to NMI */
	if (!APIC_INTEGRATED(ver))		/* 82489DX */
		v |= APIC_LVT_LEVEL_TRIGGER;
	apic_write_around(APIC_LVT0, v);
}

static void 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.
	 */ 
	printk(KERN_INFO "activating NMI Watchdog ...");

	smp_call_function(enable_NMI_through_LVT0, NULL, 1, 1);
	enable_NMI_through_LVT0(NULL);

	printk(" 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 unlock_ExtINT_logic(void)
{
	int pin, i;
	struct IO_APIC_route_entry entry0, entry1;
	unsigned char save_control, save_freq_select;
	unsigned long flags;

	pin = find_isa_irq_pin(8, mp_INT);
	if (pin == -1)
		return;

	spin_lock_irqsave(&ioapic_lock, flags);
	*(((int *)&entry0) + 1) = io_apic_read(0, 0x11 + 2 * pin);
	*(((int *)&entry0) + 0) = io_apic_read(0, 0x10 + 2 * pin);
	spin_unlock_irqrestore(&ioapic_lock, flags);
	clear_IO_APIC_pin(0, pin);

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

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

	spin_lock_irqsave(&ioapic_lock, flags);
	io_apic_write(0, 0x11 + 2 * pin, *(((int *)&entry1) + 1));
	io_apic_write(0, 0x10 + 2 * pin, *(((int *)&entry1) + 0));
	spin_unlock_irqrestore(&ioapic_lock, flags);

	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(0, pin);

	spin_lock_irqsave(&ioapic_lock, flags);
	io_apic_write(0, 0x11 + 2 * pin, *(((int *)&entry0) + 1));
	io_apic_write(0, 0x10 + 2 * pin, *(((int *)&entry0) + 0));
	spin_unlock_irqrestore(&ioapic_lock, flags);
}

/*
 * 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.
 */
static inline void check_timer(void)
{
	extern int timer_ack;
	int pin1, pin2;
	int vector;

	/*
	 * get/set the timer IRQ vector:
	 */
	disable_8259A_irq(0);
	vector = assign_irq_vector(0);
	set_intr_gate(vector, interrupt[0]);

	/*
	 * Subtle, code in do_timer_interrupt() expects an AEOI
	 * mode for the 8259A whenever interrupts are routed
	 * through I/O APICs.  Also IRQ0 has to be enabled in
	 * the 8259A which implies the virtual wire has to be
	 * disabled in the local APIC.
	 */
	apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
	init_8259A(1);
	timer_ack = 1;
	enable_8259A_irq(0);

	pin1 = find_isa_irq_pin(0, mp_INT);
	pin2 = find_isa_irq_pin(0, mp_ExtINT);

	printk(KERN_INFO "..TIMER: vector=0x%02X pin1=%d pin2=%d\n", vector, pin1, pin2);

	if (pin1 != -1) {
		/*
		 * Ok, does IRQ0 through the IOAPIC work?
		 */
		unmask_IO_APIC_irq(0);
		if (timer_irq_works()) {
			if (nmi_watchdog == NMI_IO_APIC) {
				disable_8259A_irq(0);
				setup_nmi();
				enable_8259A_irq(0);
				check_nmi_watchdog();
			}
			return;
		}
		clear_IO_APIC_pin(0, pin1);
		printk(KERN_ERR "..MP-BIOS bug: 8254 timer not connected to IO-APIC\n");
	}

	printk(KERN_INFO "...trying to set up timer (IRQ0) through the 8259A ... ");
	if (pin2 != -1) {
		printk("\n..... (found pin %d) ...", pin2);
		/*
		 * legacy devices should be connected to IO APIC #0
		 */
		setup_ExtINT_IRQ0_pin(pin2, vector);
		if (timer_irq_works()) {
			printk("works.\n");
			if (nmi_watchdog == NMI_IO_APIC) {
				setup_nmi();
				check_nmi_watchdog();
			}
			return;
		}
		/*
		 * Cleanup, just in case ...
		 */
		clear_IO_APIC_pin(0, pin2);
	}
	printk(" failed.\n");

	if (nmi_watchdog) {
		printk(KERN_WARNING "timer doesnt work through the IO-APIC - disabling NMI Watchdog!\n");
		nmi_watchdog = 0;
	}

	printk(KERN_INFO "...trying to set up timer as Virtual Wire IRQ...");

	disable_8259A_irq(0);
	irq_desc[0].handler = &lapic_irq_type;
	apic_write_around(APIC_LVT0, APIC_DM_FIXED | vector);	/* Fixed mode */
	enable_8259A_irq(0);

	if (timer_irq_works()) {
		printk(" works.\n");
		return;
	}
	apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | vector);
	printk(" failed.\n");

	printk(KERN_INFO "...trying to set up timer as ExtINT IRQ...");

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

	unlock_ExtINT_logic();

	if (timer_irq_works()) {
		printk(" works.\n");
		return;
	}
	printk(" failed :(.\n");
	panic("IO-APIC + timer doesn't work! pester [e-mail]");
}

/*
 *
 * IRQ's that are handled by the old PIC in all cases:
 * - IRQ2 is the cascade IRQ, and cannot be a io-apic IRQ.
 *   Linux doesn't really care, as it's not actually used
 *   for any interrupt handling anyway.
 * - There used to be IRQ13 here as well, but all
 *   MPS-compliant must not use it for FPU coupling and we
 *   want to use exception 16 anyway.  And there are
 *   systems who connect it to an I/O APIC for other uses.
 *   Thus we don't mark it special any longer.
 *
 * Additionally, something is definitely wrong with irq9
 * on PIIX4 boards.
 */
#define PIC_IRQS	(1<<2)

void __init setup_IO_APIC(void)
{
	enable_IO_APIC();

	io_apic_irqs = ~PIC_IRQS;
	printk("ENABLING IO-APIC IRQs\n");

	/*
	 * Set up the IO-APIC IRQ routing table by parsing the MP-BIOS
	 * mptable:
	 */
	setup_ioapic_ids_from_mpc();
	sync_Arb_IDs();
	setup_IO_APIC_irqs();
	init_IO_APIC_traps();
	check_timer();
	print_IO_APIC();
}