ipmi_watchdog.c

linux 内核源代码

		       msg->msg.cmd);
	}
	
	ipmi_free_recv_msg(msg);
}

static void ipmi_wdog_pretimeout_handler(void *handler_data)
{
	if (preaction_val != WDOG_PRETIMEOUT_NONE) {
		if (preop_val == WDOG_PREOP_PANIC) {
			if (atomic_inc_and_test(&preop_panic_excl))
				panic("Watchdog pre-timeout");
		} else if (preop_val == WDOG_PREOP_GIVE_DATA) {
			spin_lock(&ipmi_read_lock);
			data_to_read = 1;
			wake_up_interruptible(&read_q);
			kill_fasync(&fasync_q, SIGIO, POLL_IN);

			spin_unlock(&ipmi_read_lock);
		}
	}

	/* On some machines, the heartbeat will give
	   an error and not work unless we re-enable
	   the timer.   So do so. */
	pretimeout_since_last_heartbeat = 1;
}

static struct ipmi_user_hndl ipmi_hndlrs =
{
	.ipmi_recv_hndl           = ipmi_wdog_msg_handler,
	.ipmi_watchdog_pretimeout = ipmi_wdog_pretimeout_handler
};

static void ipmi_register_watchdog(int ipmi_intf)
{
	int rv = -EBUSY;

	if (watchdog_user)
		goto out;

	if ((ifnum_to_use >= 0) && (ifnum_to_use != ipmi_intf))
		goto out;

	watchdog_ifnum = ipmi_intf;

	rv = ipmi_create_user(ipmi_intf, &ipmi_hndlrs, NULL, &watchdog_user);
	if (rv < 0) {
		printk(KERN_CRIT PFX "Unable to register with ipmi\n");
		goto out;
	}

	ipmi_get_version(watchdog_user,
			 &ipmi_version_major,
			 &ipmi_version_minor);

	rv = misc_register(&ipmi_wdog_miscdev);
	if (rv < 0) {
		ipmi_destroy_user(watchdog_user);
		watchdog_user = NULL;
		printk(KERN_CRIT PFX "Unable to register misc device\n");
	}

#ifdef HAVE_DIE_NMI
	if (nmi_handler_registered) {
		int old_pretimeout = pretimeout;
		int old_timeout = timeout;
		int old_preop_val = preop_val;

		/* Set the pretimeout to go off in a second and give
		   ourselves plenty of time to stop the timer. */
		ipmi_watchdog_state = WDOG_TIMEOUT_RESET;
		preop_val = WDOG_PREOP_NONE; /* Make sure nothing happens */
		pretimeout = 99;
		timeout = 100;

		testing_nmi = 1;

		rv = ipmi_set_timeout(IPMI_SET_TIMEOUT_FORCE_HB);
		if (rv) {
			printk(KERN_WARNING PFX "Error starting timer to"
			       " test NMI: 0x%x.  The NMI pretimeout will"
			       " likely not work\n", rv);
			rv = 0;
			goto out_restore;
		}

		msleep(1500);

		if (testing_nmi != 2) {
			printk(KERN_WARNING PFX "IPMI NMI didn't seem to"
			       " occur.  The NMI pretimeout will"
			       " likely not work\n");
		}
	out_restore:
		testing_nmi = 0;
		preop_val = old_preop_val;
		pretimeout = old_pretimeout;
		timeout = old_timeout;
	}
#endif

 out:
	if ((start_now) && (rv == 0)) {
		/* Run from startup, so start the timer now. */
		start_now = 0; /* Disable this function after first startup. */
		ipmi_watchdog_state = action_val;
		ipmi_set_timeout(IPMI_SET_TIMEOUT_FORCE_HB);
		printk(KERN_INFO PFX "Starting now!\n");
	} else {
		/* Stop the timer now. */
		ipmi_watchdog_state = WDOG_TIMEOUT_NONE;
		ipmi_set_timeout(IPMI_SET_TIMEOUT_NO_HB);
	}
}

static void ipmi_unregister_watchdog(int ipmi_intf)
{
	int rv;

	if (!watchdog_user)
		goto out;

	if (watchdog_ifnum != ipmi_intf)
		goto out;

	/* Make sure no one can call us any more. */
	misc_deregister(&ipmi_wdog_miscdev);

	/* Wait to make sure the message makes it out.  The lower layer has
	   pointers to our buffers, we want to make sure they are done before
	   we release our memory. */
	while (atomic_read(&set_timeout_tofree))
		schedule_timeout_uninterruptible(1);

	/* Disconnect from IPMI. */
	rv = ipmi_destroy_user(watchdog_user);
	if (rv) {
		printk(KERN_WARNING PFX "error unlinking from IPMI: %d\n",
		       rv);
	}
	watchdog_user = NULL;

 out:
	return;
}

#ifdef HAVE_DIE_NMI
static int
ipmi_nmi(struct notifier_block *self, unsigned long val, void *data)
{
	struct die_args *args = data;

	if (val != DIE_NMI)
		return NOTIFY_OK;

	/* Hack, if it's a memory or I/O error, ignore it. */
	if (args->err & 0xc0)
		return NOTIFY_OK;

	/*
	 * If we get here, it's an NMI that's not a memory or I/O
	 * error.  We can't truly tell if it's from IPMI or not
	 * without sending a message, and sending a message is almost
	 * impossible because of locking.
	 */

	if (testing_nmi) {
		testing_nmi = 2;
		return NOTIFY_STOP;
	}

        /* If we are not expecting a timeout, ignore it. */
	if (ipmi_watchdog_state == WDOG_TIMEOUT_NONE)
		return NOTIFY_OK;

	if (preaction_val != WDOG_PRETIMEOUT_NMI)
		return NOTIFY_OK;

	/* If no one else handled the NMI, we assume it was the IPMI
           watchdog. */
	if (preop_val == WDOG_PREOP_PANIC) {
		/* On some machines, the heartbeat will give
		   an error and not work unless we re-enable
		   the timer.   So do so. */
		pretimeout_since_last_heartbeat = 1;
		if (atomic_inc_and_test(&preop_panic_excl))
			panic(PFX "pre-timeout");
	}

	return NOTIFY_STOP;
}

static struct notifier_block ipmi_nmi_handler = {
	.notifier_call = ipmi_nmi
};
#endif

static int wdog_reboot_handler(struct notifier_block *this,
			       unsigned long         code,
			       void                  *unused)
{
	static int reboot_event_handled = 0;

	if ((watchdog_user) && (!reboot_event_handled)) {
		/* Make sure we only do this once. */
		reboot_event_handled = 1;

		if (code == SYS_POWER_OFF || code == SYS_HALT) {
			/* Disable the WDT if we are shutting down. */
			ipmi_watchdog_state = WDOG_TIMEOUT_NONE;
			panic_halt_ipmi_set_timeout();
		} else if (ipmi_watchdog_state != WDOG_TIMEOUT_NONE) {
			/* Set a long timer to let the reboot happens, but
			   reboot if it hangs, but only if the watchdog
			   timer was already running. */
			timeout = 120;
			pretimeout = 0;
			ipmi_watchdog_state = WDOG_TIMEOUT_RESET;
			panic_halt_ipmi_set_timeout();
		}
	}
	return NOTIFY_OK;
}

static struct notifier_block wdog_reboot_notifier = {
	.notifier_call	= wdog_reboot_handler,
	.next		= NULL,
	.priority	= 0
};

static int wdog_panic_handler(struct notifier_block *this,
			      unsigned long         event,
			      void                  *unused)
{
	static int panic_event_handled = 0;

	/* On a panic, if we have a panic timeout, make sure to extend
	   the watchdog timer to a reasonable value to complete the
	   panic, if the watchdog timer is running.  Plus the
	   pretimeout is meaningless at panic time. */
	if (watchdog_user && !panic_event_handled &&
	    ipmi_watchdog_state != WDOG_TIMEOUT_NONE) {
		/* Make sure we do this only once. */
		panic_event_handled = 1;
	    
		timeout = 255;
		pretimeout = 0;
		panic_halt_ipmi_set_timeout();
	}

	return NOTIFY_OK;
}

static struct notifier_block wdog_panic_notifier = {
	.notifier_call	= wdog_panic_handler,
	.next		= NULL,
	.priority	= 150	/* priority: INT_MAX >= x >= 0 */
};


static void ipmi_new_smi(int if_num, struct device *device)
{
	ipmi_register_watchdog(if_num);
}

static void ipmi_smi_gone(int if_num)
{
	ipmi_unregister_watchdog(if_num);
}

static struct ipmi_smi_watcher smi_watcher =
{
	.owner    = THIS_MODULE,
	.new_smi  = ipmi_new_smi,
	.smi_gone = ipmi_smi_gone
};

static int action_op(const char *inval, char *outval)
{
	if (outval)
		strcpy(outval, action);

	if (!inval)
		return 0;

	if (strcmp(inval, "reset") == 0)
		action_val = WDOG_TIMEOUT_RESET;
	else if (strcmp(inval, "none") == 0)
		action_val = WDOG_TIMEOUT_NONE;
	else if (strcmp(inval, "power_cycle") == 0)
		action_val = WDOG_TIMEOUT_POWER_CYCLE;
	else if (strcmp(inval, "power_off") == 0)
		action_val = WDOG_TIMEOUT_POWER_DOWN;
	else
		return -EINVAL;
	strcpy(action, inval);
	return 0;
}

static int preaction_op(const char *inval, char *outval)
{
	if (outval)
		strcpy(outval, preaction);

	if (!inval)
		return 0;

	if (strcmp(inval, "pre_none") == 0)
		preaction_val = WDOG_PRETIMEOUT_NONE;
	else if (strcmp(inval, "pre_smi") == 0)
		preaction_val = WDOG_PRETIMEOUT_SMI;
#ifdef HAVE_DIE_NMI
	else if (strcmp(inval, "pre_nmi") == 0)
		preaction_val = WDOG_PRETIMEOUT_NMI;
#endif
	else if (strcmp(inval, "pre_int") == 0)
		preaction_val = WDOG_PRETIMEOUT_MSG_INT;
	else
		return -EINVAL;
	strcpy(preaction, inval);
	return 0;
}

static int preop_op(const char *inval, char *outval)
{
	if (outval)
		strcpy(outval, preop);

	if (!inval)
		return 0;

	if (strcmp(inval, "preop_none") == 0)
		preop_val = WDOG_PREOP_NONE;
	else if (strcmp(inval, "preop_panic") == 0)
		preop_val = WDOG_PREOP_PANIC;
	else if (strcmp(inval, "preop_give_data") == 0)
		preop_val = WDOG_PREOP_GIVE_DATA;
	else
		return -EINVAL;
	strcpy(preop, inval);
	return 0;
}

static void check_parms(void)
{
#ifdef HAVE_DIE_NMI
	int do_nmi = 0;
	int rv;

	if (preaction_val == WDOG_PRETIMEOUT_NMI) {
		do_nmi = 1;
		if (preop_val == WDOG_PREOP_GIVE_DATA) {
			printk(KERN_WARNING PFX "Pretimeout op is to give data"
			       " but NMI pretimeout is enabled, setting"
			       " pretimeout op to none\n");
			preop_op("preop_none", NULL);
			do_nmi = 0;
		}
	}
	if (do_nmi && !nmi_handler_registered) {
		rv = register_die_notifier(&ipmi_nmi_handler);
		if (rv) {
			printk(KERN_WARNING PFX
			       "Can't register nmi handler\n");
			return;
		} else
			nmi_handler_registered = 1;
	} else if (!do_nmi && nmi_handler_registered) {
		unregister_die_notifier(&ipmi_nmi_handler);
		nmi_handler_registered = 0;
	}
#endif
}

static int __init ipmi_wdog_init(void)
{
	int rv;

	if (action_op(action, NULL)) {
		action_op("reset", NULL);
		printk(KERN_INFO PFX "Unknown action '%s', defaulting to"
		       " reset\n", action);
	}

	if (preaction_op(preaction, NULL)) {
		preaction_op("pre_none", NULL);
		printk(KERN_INFO PFX "Unknown preaction '%s', defaulting to"
		       " none\n", preaction);
	}

	if (preop_op(preop, NULL)) {
		preop_op("preop_none", NULL);
		printk(KERN_INFO PFX "Unknown preop '%s', defaulting to"
		       " none\n", preop);
	}

	check_parms();

	register_reboot_notifier(&wdog_reboot_notifier);
	atomic_notifier_chain_register(&panic_notifier_list,
			&wdog_panic_notifier);

	rv = ipmi_smi_watcher_register(&smi_watcher);
	if (rv) {
#ifdef HAVE_DIE_NMI
		if (nmi_handler_registered)
			unregister_die_notifier(&ipmi_nmi_handler);
#endif
		atomic_notifier_chain_unregister(&panic_notifier_list,
						 &wdog_panic_notifier);
		unregister_reboot_notifier(&wdog_reboot_notifier);
		printk(KERN_WARNING PFX "can't register smi watcher\n");
		return rv;
	}

	printk(KERN_INFO PFX "driver initialized\n");

	return 0;
}

static void __exit ipmi_wdog_exit(void)
{
	ipmi_smi_watcher_unregister(&smi_watcher);
	ipmi_unregister_watchdog(watchdog_ifnum);

#ifdef HAVE_DIE_NMI
	if (nmi_handler_registered)
		unregister_die_notifier(&ipmi_nmi_handler);
#endif

	atomic_notifier_chain_unregister(&panic_notifier_list,
					 &wdog_panic_notifier);
	unregister_reboot_notifier(&wdog_reboot_notifier);
}
module_exit(ipmi_wdog_exit);
module_init(ipmi_wdog_init);
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Corey Minyard <[e-mail]>");
MODULE_DESCRIPTION("watchdog timer based upon the IPMI interface.");