ipmi_watchdog.c

linux-2.6.15.6

	.read    = ipmi_read,
	.poll    = ipmi_poll,
	.write   = ipmi_write,
	.ioctl   = ipmi_ioctl,
	.open    = ipmi_open,
	.release = ipmi_close,
	.fasync  = ipmi_fasync,
};

static struct miscdevice ipmi_wdog_miscdev = {
	.minor		= WATCHDOG_MINOR,
	.name		= "watchdog",
	.fops		= &ipmi_wdog_fops
};

static void ipmi_wdog_msg_handler(struct ipmi_recv_msg *msg,
				  void                 *handler_data)
{
	if (msg->msg.data[0] != 0) {
		printk(KERN_ERR PFX "response: Error %x on cmd %x\n",
		       msg->msg.data[0],
		       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;

	down_write(&register_sem);
	if (watchdog_user)
		goto out;

	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");
	}

 out:
	up_write(&register_sem);

	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");
	}
}

#ifdef HAVE_NMI_HANDLER
static int
ipmi_nmi(void *dev_id, struct pt_regs *regs, int cpu, int handled)
{
        /* If we are not expecting a timeout, ignore it. */
	if (ipmi_watchdog_state == WDOG_TIMEOUT_NONE)
		return NOTIFY_DONE;

	/* If no one else handled the NMI, we assume it was the IPMI
           watchdog. */
	if ((!handled) && (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_DONE;
}

static struct nmi_handler ipmi_nmi_handler =
{
	.link     = LIST_HEAD_INIT(ipmi_nmi_handler.link),
	.dev_name = "ipmi_watchdog",
	.dev_id   = NULL,
	.handler  = ipmi_nmi,
	.priority = 0, /* Call us last. */
};
int nmi_handler_registered;
#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_DOWN || code == SYS_HALT) {
			/* Disable the WDT if we are shutting down. */
			ipmi_watchdog_state = WDOG_TIMEOUT_NONE;
			panic_halt_ipmi_set_timeout();
		} else {
			/* Set a long timer to let the reboot happens, but
			   reboot if it hangs. */
			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 that the thing
	   reboots, even if it hangs during that panic. */
	if (watchdog_user && !panic_event_handled) {
		/* Make sure the panic doesn't hang, and make sure we
		   do this only once. */
		panic_event_handled = 1;
	    
		timeout = 255;
		pretimeout = 0;
		ipmi_watchdog_state = WDOG_TIMEOUT_RESET;
		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)
{
	ipmi_register_watchdog(if_num);
}

static void ipmi_smi_gone(int if_num)
{
	/* This can never be called, because once the watchdog is
	   registered, the interface can't go away until the watchdog
	   is unregistered. */
}

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_NMI_HANDLER
	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_NMI_HANDLER
	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;
		}
#ifdef CONFIG_X86_LOCAL_APIC
		if (nmi_watchdog == NMI_IO_APIC) {
			printk(KERN_WARNING PFX "nmi_watchdog is set to IO APIC"
			       " mode (value is %d), that is incompatible"
			       " with using NMI in the IPMI watchdog."
			       " Disabling IPMI nmi pretimeout.\n",
			       nmi_watchdog);
			preaction_val = WDOG_PRETIMEOUT_NONE;
			do_nmi = 0;
		}
#endif
	}
	if (do_nmi && !nmi_handler_registered) {
		rv = request_nmi(&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) {
		release_nmi(&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();

	rv = ipmi_smi_watcher_register(&smi_watcher);
	if (rv) {
#ifdef HAVE_NMI_HANDLER
		if (preaction_val == WDOG_PRETIMEOUT_NMI)
			release_nmi(&ipmi_nmi_handler);
#endif
		printk(KERN_WARNING PFX "can't register smi watcher\n");
		return rv;
	}

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

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

	return 0;
}

static __exit void ipmi_unregister_watchdog(void)
{
	int rv;

	down_write(&register_sem);

#ifdef HAVE_NMI_HANDLER
	if (nmi_handler_registered)
		release_nmi(&ipmi_nmi_handler);
#endif

	notifier_chain_unregister(&panic_notifier_list, &wdog_panic_notifier);
	unregister_reboot_notifier(&wdog_reboot_notifier);

	if (! watchdog_user)
		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:
	up_write(&register_sem);
}

static void __exit ipmi_wdog_exit(void)
{
	ipmi_smi_watcher_unregister(&smi_watcher);
	ipmi_unregister_watchdog();
}
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.");