irq.c

包括初始化、进程管理、进程通信、内存管理、设备管理、中断、文件系统、系统调用等精选源码分析。

/*
 *	linux/arch/i386/kernel/irq.c
 *
 *	Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar
 *
 * This file contains the code used by various IRQ handling routines:
 * asking for different IRQ's should be done through these routines
 * instead of just grabbing them. Thus setups with different IRQ numbers
 * shouldn't result in any weird surprises, and installing new handlers
 * should be easier.
 */

/*
 * (mostly architecture independent, will move to kernel/irq.c in 2.5.)
 *
 * IRQs are in fact implemented a bit like signal handlers for the kernel.
 * Naturally it's not a 1:1 relation, but there are similarities.
 */

#include <linux/config.h>
#include <linux/ptrace.h>
#include <linux/errno.h>
#include <linux/signal.h>
#include <linux/sched.h>
#include <linux/ioport.h>
#include <linux/interrupt.h>
#include <linux/timex.h>
#include <linux/malloc.h>
#include <linux/random.h>
#include <linux/smp_lock.h>
#include <linux/init.h>
#include <linux/kernel_stat.h>
#include <linux/irq.h>
#include <linux/proc_fs.h>

#include <asm/io.h>
#include <asm/smp.h>
#include <asm/system.h>
#include <asm/bitops.h>
#include <asm/uaccess.h>
#include <asm/pgalloc.h>
#include <asm/delay.h>
#include <asm/desc.h>
#include <asm/irq.h>



/*
 * Linux has a controller-independent x86 interrupt architecture.
 * every controller has a 'controller-template', that is used
 * by the main code to do the right thing. Each driver-visible
 * interrupt source is transparently wired to the apropriate
 * controller. Thus drivers need not be aware of the
 * interrupt-controller.
 *
 * Various interrupt controllers we handle: 8259 PIC, SMP IO-APIC,
 * PIIX4's internal 8259 PIC and SGI's Visual Workstation Cobalt (IO-)APIC.
 * (IO-APICs assumed to be messaging to Pentium local-APICs)
 *
 * the code is designed to be easily extended with new/different
 * interrupt controllers, without having to do assembly magic.
 */

/*
 * Controller mappings for all interrupt sources:
 */
irq_desc_t irq_desc[NR_IRQS] __cacheline_aligned =
	{ [0 ... NR_IRQS-1] = { 0, &no_irq_type, NULL, 0, SPIN_LOCK_UNLOCKED}};

static void register_irq_proc (unsigned int irq);

/*
 * Special irq handlers.
 */

void no_action(int cpl, void *dev_id, struct pt_regs *regs) { }

/*
 * Generic no controller code
 */

static void enable_none(unsigned int irq) { }
static unsigned int startup_none(unsigned int irq) { return 0; }
static void disable_none(unsigned int irq) { }
static void ack_none(unsigned int irq)
{
/*
 * 'what should we do if we get a hw irq event on an illegal vector'.
 * each architecture has to answer this themselves, it doesnt deserve
 * a generic callback i think.
 */
#if CONFIG_X86
	printk("unexpected IRQ trap at vector %02x\n", irq);
#ifdef CONFIG_X86_LOCAL_APIC
	/*
	 * Currently unexpected vectors happen only on SMP and APIC.
	 * We _must_ ack these because every local APIC has only N
	 * irq slots per priority level, and a 'hanging, unacked' IRQ
	 * holds up an irq slot - in excessive cases (when multiple
	 * unexpected vectors occur) that might lock up the APIC
	 * completely.
	 */
	ack_APIC_irq();
#endif
#endif
}

/* startup is the same as "enable", shutdown is same as "disable" */
#define shutdown_none	disable_none
#define end_none	enable_none

struct hw_interrupt_type no_irq_type = {
	"none",
	startup_none,
	shutdown_none,
	enable_none,
	disable_none,
	ack_none,
	end_none
};

volatile unsigned long irq_err_count;

/*
 * Generic, controller-independent functions:
 */

int get_irq_list(char *buf)
{
	int i, j;
	struct irqaction * action;
	char *p = buf;

	p += sprintf(p, "           ");
	for (j=0; j<smp_num_cpus; j++)
		p += sprintf(p, "CPU%d       ",j);
	*p++ = '\n';

	for (i = 0 ; i < NR_IRQS ; i++) {
		action = irq_desc[i].action;
		if (!action) 
			continue;
		p += sprintf(p, "%3d: ",i);
#ifndef CONFIG_SMP
		p += sprintf(p, "%10u ", kstat_irqs(i));
#else
		for (j = 0; j < smp_num_cpus; j++)
			p += sprintf(p, "%10u ",
				kstat.irqs[cpu_logical_map(j)][i]);
#endif
		p += sprintf(p, " %14s", irq_desc[i].handler->typename);
		p += sprintf(p, "  %s", action->name);

		for (action=action->next; action; action = action->next)
			p += sprintf(p, ", %s", action->name);
		*p++ = '\n';
	}
	p += sprintf(p, "NMI: ");
	for (j = 0; j < smp_num_cpus; j++)
		p += sprintf(p, "%10u ",
			nmi_count(cpu_logical_map(j)));
	p += sprintf(p, "\n");
#if CONFIG_SMP
	p += sprintf(p, "LOC: ");
	for (j = 0; j < smp_num_cpus; j++)
		p += sprintf(p, "%10u ",
			apic_timer_irqs[cpu_logical_map(j)]);
	p += sprintf(p, "\n");
#endif
	p += sprintf(p, "ERR: %10lu\n", irq_err_count);
	return p - buf;
}


/*
 * Global interrupt locks for SMP. Allow interrupts to come in on any
 * CPU, yet make cli/sti act globally to protect critical regions..
 */

#ifdef CONFIG_SMP
unsigned char global_irq_holder = NO_PROC_ID;
unsigned volatile long global_irq_lock; /* pendantic: long for set_bit --RR */

extern void show_stack(unsigned long* esp);

static void show(char * str)
{
	int i;
	int cpu = smp_processor_id();

	printk("\n%s, CPU %d:\n", str, cpu);
	printk("irq:  %d [",irqs_running());
	for(i=0;i < smp_num_cpus;i++)
		printk(" %d",local_irq_count(i));
	printk(" ]\nbh:   %d [",spin_is_locked(&global_bh_lock) ? 1 : 0);
	for(i=0;i < smp_num_cpus;i++)
		printk(" %d",local_bh_count(i));

	printk(" ]\nStack dumps:");
	for(i = 0; i < smp_num_cpus; i++) {
		unsigned long esp;
		if (i == cpu)
			continue;
		printk("\nCPU %d:",i);
		esp = init_tss[i].esp0;
		if (!esp) {
			/* tss->esp0 is set to NULL in cpu_init(),
			 * it's initialized when the cpu returns to user
			 * space. -- manfreds
			 */
			printk(" <unknown> ");
			continue;
		}
		esp &= ~(THREAD_SIZE-1);
		esp += sizeof(struct task_struct);
		show_stack((void*)esp);
 	}
	printk("\nCPU %d:",cpu);
	show_stack(NULL);
	printk("\n");
}
	
#define MAXCOUNT 100000000

/*
 * I had a lockup scenario where a tight loop doing
 * spin_unlock()/spin_lock() on CPU#1 was racing with
 * spin_lock() on CPU#0. CPU#0 should have noticed spin_unlock(), but
 * apparently the spin_unlock() information did not make it
 * through to CPU#0 ... nasty, is this by design, do we have to limit
 * 'memory update oscillation frequency' artificially like here?
 *
 * Such 'high frequency update' races can be avoided by careful design, but
 * some of our major constructs like spinlocks use similar techniques,
 * it would be nice to clarify this issue. Set this define to 0 if you
 * want to check whether your system freezes.  I suspect the delay done
 * by SYNC_OTHER_CORES() is in correlation with 'snooping latency', but
 * i thought that such things are guaranteed by design, since we use
 * the 'LOCK' prefix.
 */
#define SUSPECTED_CPU_OR_CHIPSET_BUG_WORKAROUND 0

#if SUSPECTED_CPU_OR_CHIPSET_BUG_WORKAROUND
# define SYNC_OTHER_CORES(x) udelay(x+1)
#else
/*
 * We have to allow irqs to arrive between __sti and __cli
 */
# define SYNC_OTHER_CORES(x) __asm__ __volatile__ ("nop")
#endif

static inline void wait_on_irq(int cpu)
{
	int count = MAXCOUNT;

	for (;;) {

		/*
		 * Wait until all interrupts are gone. Wait
		 * for bottom half handlers unless we're
		 * already executing in one..
		 */
		if (!irqs_running())
			if (local_bh_count(cpu) || !spin_is_locked(&global_bh_lock))
				break;

		/* Duh, we have to loop. Release the lock to avoid deadlocks */
		clear_bit(0,&global_irq_lock);

		for (;;) {
			if (!--count) {
				show("wait_on_irq");
				count = ~0;
			}
			__sti();
			SYNC_OTHER_CORES(cpu);
			__cli();
			if (irqs_running())
				continue;
			if (global_irq_lock)
				continue;
			if (!local_bh_count(cpu) && spin_is_locked(&global_bh_lock))
				continue;
			if (!test_and_set_bit(0,&global_irq_lock))
				break;
		}
	}
}

/*
 * This is called when we want to synchronize with
 * interrupts. We may for example tell a device to
 * stop sending interrupts: but to make sure there
 * are no interrupts that are executing on another
 * CPU we need to call this function.
 */
void synchronize_irq(void)
{
	if (irqs_running()) {
		/* Stupid approach */
		cli();
		sti();
	}
}

static inline void get_irqlock(int cpu)
{
	if (test_and_set_bit(0,&global_irq_lock)) {
		/* do we already hold the lock? */
		if ((unsigned char) cpu == global_irq_holder)
			return;
		/* Uhhuh.. Somebody else got it. Wait.. */
		do {
			do {
			} while (test_bit(0,&global_irq_lock));
		} while (test_and_set_bit(0,&global_irq_lock));		
	}
	/* 
	 * We also to make sure that nobody else is running
	 * in an interrupt context. 
	 */
	wait_on_irq(cpu);

	/*
	 * Ok, finally..
	 */
	global_irq_holder = cpu;
}

#define EFLAGS_IF_SHIFT 9

/*
 * A global "cli()" while in an interrupt context
 * turns into just a local cli(). Interrupts
 * should use spinlocks for the (very unlikely)
 * case that they ever want to protect against
 * each other.
 *
 * If we already have local interrupts disabled,
 * this will not turn a local disable into a
 * global one (problems with spinlocks: this makes
 * save_flags+cli+sti usable inside a spinlock).
 */
void __global_cli(void)
{
	unsigned int flags;

	__save_flags(flags);
	if (flags & (1 << EFLAGS_IF_SHIFT)) {
		int cpu = smp_processor_id();
		__cli();
		if (!local_irq_count(cpu))
			get_irqlock(cpu);
	}
}

void __global_sti(void)
{
	int cpu = smp_processor_id();

	if (!local_irq_count(cpu))
		release_irqlock(cpu);
	__sti();
}

/*
 * SMP flags value to restore to:
 * 0 - global cli
 * 1 - global sti
 * 2 - local cli
 * 3 - local sti
 */
unsigned long __global_save_flags(void)
{
	int retval;
	int local_enabled;
	unsigned long flags;
	int cpu = smp_processor_id();

	__save_flags(flags);
	local_enabled = (flags >> EFLAGS_IF_SHIFT) & 1;
	/* default to local */
	retval = 2 + local_enabled;

	/* check for global flags if we're not in an interrupt */
	if (!local_irq_count(cpu)) {
		if (local_enabled)
			retval = 1;
		if (global_irq_holder == cpu)
			retval = 0;
	}
	return retval;
}

void __global_restore_flags(unsigned long flags)
{
	switch (flags) {
	case 0:
		__global_cli();
		break;
	case 1:
		__global_sti();
		break;
	case 2:
		__cli();
		break;
	case 3:
		__sti();
		break;
	default:
		printk("global_restore_flags: %08lx (%08lx)\n",
			flags, (&flags)[-1]);
	}
}

#endif

/*
 * This should really return information about whether
 * we should do bottom half handling etc. Right now we
 * end up _always_ checking the bottom half, which is a
 * waste of time and is not what some drivers would
 * prefer.
 */
int handle_IRQ_event(unsigned int irq, struct pt_regs * regs, struct irqaction * action)
{
	int status;
	int cpu = smp_processor_id();

	irq_enter(cpu, irq);

	status = 1;	/* Force the "do bottom halves" bit */

	if (!(action->flags & SA_INTERRUPT))
		__sti();

	do {
		status |= action->flags;
		action->handler(irq, action->dev_id, regs);
		action = action->next;
	} while (action);
	if (status & SA_SAMPLE_RANDOM)
		add_interrupt_randomness(irq);
	__cli();

	irq_exit(cpu, irq);

	return status;
}

/*
 * Generic enable/disable code: this just calls
 * down into the PIC-specific version for the actual
 * hardware disable after having gotten the irq
 * controller lock. 
 */
 
/**
 *	disable_irq_nosync - disable an irq without waiting
 *	@irq: Interrupt to disable
 *
 *	Disable the selected interrupt line. Disables of an interrupt
 *	stack. Unlike disable_irq(), this function does not ensure existing
 *	instances of the IRQ handler have completed before returning.
 *
 *	This function may be called from IRQ context.
 */
 
void inline disable_irq_nosync(unsigned int irq)
{
	irq_desc_t *desc = irq_desc + irq;
	unsigned long flags;

	spin_lock_irqsave(&desc->lock, flags);
	if (!desc->depth++) {
		desc->status |= IRQ_DISABLED;
		desc->handler->disable(irq);
	}
	spin_unlock_irqrestore(&desc->lock, flags);
}

/**
 *	disable_irq - disable an irq and wait for completion
 *	@irq: Interrupt to disable
 *
 *	Disable the selected interrupt line. Disables of an interrupt
 *	stack. That is for two disables you need two enables. This
 *	function waits for any pending IRQ handlers for this interrupt
 *	to complete before returning. If you use this function while
 *	holding a resource the IRQ handler may need you will deadlock.
 *
 *	This function may be called - with care - from IRQ context.
 */
 
void disable_irq(unsigned int irq)
{
	disable_irq_nosync(irq);

	if (!local_irq_count(smp_processor_id())) {
		do {
			barrier();
		} while (irq_desc[irq].status & IRQ_INPROGRESS);
	}
}

/**
 *	enable_irq - enable interrupt handling on an irq
 *	@irq: Interrupt to enable
 *
 *	Re-enables the processing of interrupts on this IRQ line
 *	providing no disable_irq calls are now in effect.
 *
 *	This function may be called from IRQ context.
 */
 
void enable_irq(unsigned int irq)
{
	irq_desc_t *desc = irq_desc + irq;
	unsigned long flags;

	spin_lock_irqsave(&desc->lock, flags);
	switch (desc->depth) {
	case 1: {
		unsigned int status = desc->status & ~IRQ_DISABLED;
		desc->status = status;
		if ((status & (IRQ_PENDING | IRQ_REPLAY)) == IRQ_PENDING) {
			desc->status = status | IRQ_REPLAY;
			hw_resend_irq(desc->handler,irq);
		}
		desc->handler->enable(irq);
		/* fall-through */
	}
	default:
		desc->depth--;
		break;
	case 0:
		printk("enable_irq(%u) unbalanced from %p\n", irq,
		       __builtin_return_address(0));
	}
	spin_unlock_irqrestore(&desc->lock, flags);
}

/*
 * do_IRQ handles all normal device IRQ's (the special
 * SMP cross-CPU interrupts have their own specific
 * handlers).
 */
asmlinkage unsigned int do_IRQ(struct pt_regs regs)
{	
	/* 
	 * We ack quickly, we don't want the irq controller
	 * thinking we're snobs just because some other CPU has
	 * disabled global interrupts (we have already done the
	 * INT_ACK cycles, it's too late to try to pretend to the
	 * controller that we aren't taking the interrupt).
	 *
	 * 0 return value means that this irq is already being
	 * handled by some other CPU. (or is disabled)
	 */
	int irq = regs.orig_eax & 0xff; /* high bits used in ret_from_ code  */
	int cpu = smp_processor_id();
	irq_desc_t *desc = irq_desc + irq;
	struct irqaction * action;
	unsigned int status;

	kstat.irqs[cpu][irq]++;
	spin_lock(&desc->lock);
	desc->handler->ack(irq);
	/*
	   REPLAY is when Linux resends an IRQ that was dropped earlier
	   WAITING is used by probe to mark irqs that are being tested
	   */
	status = desc->status & ~(IRQ_REPLAY | IRQ_WAITING);
	status |= IRQ_PENDING; /* we _want_ to handle it */

	/*
	 * If the IRQ is disabled for whatever reason, we cannot
	 * use the action we have.
	 */
	action = NULL;
	if (!(status & (IRQ_DISABLED | IRQ_INPROGRESS))) {
		action = desc->action;
		status &= ~IRQ_PENDING; /* we commit to handling */
		status |= IRQ_INPROGRESS; /* we are handling it */
	}
	desc->status = status;