scsi_error.c

来自「Linux Kernel 2.6.9 for OMAP1710」· C语言 代码 · 共 1,846 行 · 第 1/4 页

/**
 * scsi_eh_lock_done - done function for eh door lock request
 * @scmd:	SCSI command block for the door lock request
 *
 * Notes:
 * 	We completed the asynchronous door lock request, and it has either
 * 	locked the door or failed.  We must free the command structures
 * 	associated with this request.
 **/
static void scsi_eh_lock_done(struct scsi_cmnd *scmd)
{
	struct scsi_request *sreq = scmd->sc_request;

	scsi_release_request(sreq);
}


/**
 * scsi_eh_lock_door - Prevent medium removal for the specified device
 * @sdev:	SCSI device to prevent medium removal
 *
 * Locking:
 * 	We must be called from process context; scsi_allocate_request()
 * 	may sleep.
 *
 * Notes:
 * 	We queue up an asynchronous "ALLOW MEDIUM REMOVAL" request on the
 * 	head of the devices request queue, and continue.
 *
 * Bugs:
 * 	scsi_allocate_request() may sleep waiting for existing requests to
 * 	be processed.  However, since we haven't kicked off any request
 * 	processing for this host, this may deadlock.
 *
 *	If scsi_allocate_request() fails for what ever reason, we
 *	completely forget to lock the door.
 **/
static void scsi_eh_lock_door(struct scsi_device *sdev)
{
	struct scsi_request *sreq = scsi_allocate_request(sdev, GFP_KERNEL);

	if (unlikely(!sreq)) {
		printk(KERN_ERR "%s: request allocate failed,"
		       "prevent media removal cmd not sent\n", __FUNCTION__);
		return;
	}

	sreq->sr_cmnd[0] = ALLOW_MEDIUM_REMOVAL;
	sreq->sr_cmnd[1] = 0;
	sreq->sr_cmnd[2] = 0;
	sreq->sr_cmnd[3] = 0;
	sreq->sr_cmnd[4] = SCSI_REMOVAL_PREVENT;
	sreq->sr_cmnd[5] = 0;
	sreq->sr_data_direction = DMA_NONE;
	sreq->sr_bufflen = 0;
	sreq->sr_buffer = NULL;
	sreq->sr_allowed = 5;
	sreq->sr_done = scsi_eh_lock_done;
	sreq->sr_timeout_per_command = 10 * HZ;
	sreq->sr_cmd_len = COMMAND_SIZE(sreq->sr_cmnd[0]);

	scsi_insert_special_req(sreq, 1);
}


/**
 * scsi_restart_operations - restart io operations to the specified host.
 * @shost:	Host we are restarting.
 *
 * Notes:
 *    When we entered the error handler, we blocked all further i/o to
 *    this device.  we need to 'reverse' this process.
 **/
static void scsi_restart_operations(struct Scsi_Host *shost)
{
	struct scsi_device *sdev;

	/*
	 * If the door was locked, we need to insert a door lock request
	 * onto the head of the SCSI request queue for the device.  There
	 * is no point trying to lock the door of an off-line device.
	 */
	shost_for_each_device(sdev, shost) {
		if (scsi_device_online(sdev) && sdev->locked)
			scsi_eh_lock_door(sdev);
	}

	/*
	 * next free up anything directly waiting upon the host.  this
	 * will be requests for character device operations, and also for
	 * ioctls to queued block devices.
	 */
	SCSI_LOG_ERROR_RECOVERY(3, printk("%s: waking up host to restart\n",
					  __FUNCTION__));

	clear_bit(SHOST_RECOVERY, &shost->shost_state);

	wake_up(&shost->host_wait);

	/*
	 * finally we need to re-initiate requests that may be pending.  we will
	 * have had everything blocked while error handling is taking place, and
	 * now that error recovery is done, we will need to ensure that these
	 * requests are started.
	 */
	scsi_run_host_queues(shost);
}

/**
 * scsi_eh_ready_devs - check device ready state and recover if not.
 * @shost: 	host to be recovered.
 * @eh_done_q:	list_head for processed commands.
 *
 **/
static void scsi_eh_ready_devs(struct Scsi_Host *shost,
			       struct list_head *work_q,
			       struct list_head *done_q)
{
	if (!scsi_eh_stu(shost, work_q, done_q))
		if (!scsi_eh_bus_device_reset(shost, work_q, done_q))
			if (!scsi_eh_bus_reset(shost, work_q, done_q))
				if (!scsi_eh_host_reset(work_q, done_q))
					scsi_eh_offline_sdevs(work_q, done_q);
}

/**
 * scsi_eh_flush_done_q - finish processed commands or retry them.
 * @done_q:	list_head of processed commands.
 *
 **/
static void scsi_eh_flush_done_q(struct list_head *done_q)
{
	struct list_head *lh, *lh_sf;
	struct scsi_cmnd *scmd;

	list_for_each_safe(lh, lh_sf, done_q) {
		scmd = list_entry(lh, struct scsi_cmnd, eh_entry);
		list_del_init(lh);
		if (scsi_device_online(scmd->device) &&
		    !blk_noretry_request(scmd->request) &&
		    (++scmd->retries < scmd->allowed)) {
			SCSI_LOG_ERROR_RECOVERY(3, printk("%s: flush"
							  " retry cmd: %p\n",
							  current->comm,
							  scmd));
				scsi_queue_insert(scmd, SCSI_MLQUEUE_EH_RETRY);
		} else {
			if (!scmd->result)
				scmd->result |= (DRIVER_TIMEOUT << 24);
			SCSI_LOG_ERROR_RECOVERY(3, printk("%s: flush finish"
							" cmd: %p\n",
							current->comm, scmd));
			scsi_finish_command(scmd);
		}
	}
}

/**
 * scsi_unjam_host - Attempt to fix a host which has a cmd that failed.
 * @shost:	Host to unjam.
 *
 * Notes:
 *    When we come in here, we *know* that all commands on the bus have
 *    either completed, failed or timed out.  we also know that no further
 *    commands are being sent to the host, so things are relatively quiet
 *    and we have freedom to fiddle with things as we wish.
 *
 *    This is only the *default* implementation.  it is possible for
 *    individual drivers to supply their own version of this function, and
 *    if the maintainer wishes to do this, it is strongly suggested that
 *    this function be taken as a template and modified.  this function
 *    was designed to correctly handle problems for about 95% of the
 *    different cases out there, and it should always provide at least a
 *    reasonable amount of error recovery.
 *
 *    Any command marked 'failed' or 'timeout' must eventually have
 *    scsi_finish_cmd() called for it.  we do all of the retry stuff
 *    here, so when we restart the host after we return it should have an
 *    empty queue.
 **/
static void scsi_unjam_host(struct Scsi_Host *shost)
{
	unsigned long flags;
	LIST_HEAD(eh_work_q);
	LIST_HEAD(eh_done_q);

	spin_lock_irqsave(shost->host_lock, flags);
	list_splice_init(&shost->eh_cmd_q, &eh_work_q);
	spin_unlock_irqrestore(shost->host_lock, flags);

	SCSI_LOG_ERROR_RECOVERY(1, scsi_eh_prt_fail_stats(shost, &eh_work_q));

	if (!scsi_eh_get_sense(&eh_work_q, &eh_done_q))
		if (!scsi_eh_abort_cmds(&eh_work_q, &eh_done_q))
			scsi_eh_ready_devs(shost, &eh_work_q, &eh_done_q);

	scsi_eh_flush_done_q(&eh_done_q);
}

/**
 * scsi_error_handler - Handle errors/timeouts of SCSI cmds.
 * @data:	Host for which we are running.
 *
 * Notes:
 *    This is always run in the context of a kernel thread.  The idea is
 *    that we start this thing up when the kernel starts up (one per host
 *    that we detect), and it immediately goes to sleep and waits for some
 *    event (i.e. failure).  When this takes place, we have the job of
 *    trying to unjam the bus and restarting things.
 **/
int scsi_error_handler(void *data)
{
	struct Scsi_Host *shost = (struct Scsi_Host *) data;
	int rtn;
	DECLARE_MUTEX_LOCKED(sem);

	/*
	 *    Flush resources
	 */

	daemonize("scsi_eh_%d", shost->host_no);

	current->flags |= PF_NOFREEZE;

	shost->eh_wait = &sem;
	shost->ehandler = current;

	/*
	 * Wake up the thread that created us.
	 */
	SCSI_LOG_ERROR_RECOVERY(3, printk("Wake up parent of"
					  " scsi_eh_%d\n",shost->host_no));

	complete(shost->eh_notify);

	while (1) {
		/*
		 * If we get a signal, it means we are supposed to go
		 * away and die.  This typically happens if the user is
		 * trying to unload a module.
		 */
		SCSI_LOG_ERROR_RECOVERY(1, printk("Error handler"
						  " scsi_eh_%d"
						  " sleeping\n",shost->host_no));

		/*
		 * Note - we always use down_interruptible with the semaphore
		 * even if the module was loaded as part of the kernel.  The
		 * reason is that down() will cause this thread to be counted
		 * in the load average as a running process, and down
		 * interruptible doesn't.  Given that we need to allow this
		 * thread to die if the driver was loaded as a module, using
		 * semaphores isn't unreasonable.
		 */
		down_interruptible(&sem);
		if (shost->eh_kill)
			break;

		SCSI_LOG_ERROR_RECOVERY(1, printk("Error handler"
						  " scsi_eh_%d waking"
						  " up\n",shost->host_no));

		shost->eh_active = 1;

		/*
		 * We have a host that is failing for some reason.  Figure out
		 * what we need to do to get it up and online again (if we can).
		 * If we fail, we end up taking the thing offline.
		 */
		if (shost->hostt->eh_strategy_handler) 
			rtn = shost->hostt->eh_strategy_handler(shost);
		else
			scsi_unjam_host(shost);

		shost->eh_active = 0;

		/*
		 * Note - if the above fails completely, the action is to take
		 * individual devices offline and flush the queue of any
		 * outstanding requests that may have been pending.  When we
		 * restart, we restart any I/O to any other devices on the bus
		 * which are still online.
		 */
		scsi_restart_operations(shost);

	}

	SCSI_LOG_ERROR_RECOVERY(1, printk("Error handler scsi_eh_%d"
					  " exiting\n",shost->host_no));

	/*
	 * Make sure that nobody tries to wake us up again.
	 */
	shost->eh_wait = NULL;

	/*
	 * Knock this down too.  From this point on, the host is flying
	 * without a pilot.  If this is because the module is being unloaded,
	 * that's fine.  If the user sent a signal to this thing, we are
	 * potentially in real danger.
	 */
	shost->eh_active = 0;
	shost->ehandler = NULL;

	/*
	 * If anyone is waiting for us to exit (i.e. someone trying to unload
	 * a driver), then wake up that process to let them know we are on
	 * the way out the door.
	 */
	complete_and_exit(shost->eh_notify, 0);
	return 0;
}

/*
 * Function:    scsi_report_bus_reset()
 *
 * Purpose:     Utility function used by low-level drivers to report that
 *		they have observed a bus reset on the bus being handled.
 *
 * Arguments:   shost       - Host in question
 *		channel     - channel on which reset was observed.
 *
 * Returns:     Nothing
 *
 * Lock status: Host lock must be held.
 *
 * Notes:       This only needs to be called if the reset is one which
 *		originates from an unknown location.  Resets originated
 *		by the mid-level itself don't need to call this, but there
 *		should be no harm.
 *
 *		The main purpose of this is to make sure that a CHECK_CONDITION
 *		is properly treated.
 */
void scsi_report_bus_reset(struct Scsi_Host *shost, int channel)
{
	struct scsi_device *sdev;

	__shost_for_each_device(sdev, shost) {
		if (channel == sdev->channel) {
			sdev->was_reset = 1;
			sdev->expecting_cc_ua = 1;
		}
	}
}

/*
 * Function:    scsi_report_device_reset()
 *
 * Purpose:     Utility function used by low-level drivers to report that
 *		they have observed a device reset on the device being handled.
 *
 * Arguments:   shost       - Host in question
 *		channel     - channel on which reset was observed
 *		target	    - target on which reset was observed
 *
 * Returns:     Nothing
 *
 * Lock status: Host lock must be held
 *
 * Notes:       This only needs to be called if the reset is one which
 *		originates from an unknown location.  Resets originated
 *		by the mid-level itself don't need to call this, but there
 *		should be no harm.
 *
 *		The main purpose of this is to make sure that a CHECK_CONDITION
 *		is properly treated.
 */
void scsi_report_device_reset(struct Scsi_Host *shost, int channel, int target)
{
	struct scsi_device *sdev;

	__shost_for_each_device(sdev, shost) {
		if (channel == sdev->channel &&
		    target == sdev->id) {
			sdev->was_reset = 1;
			sdev->expecting_cc_ua = 1;
		}
	}
}

static void
scsi_reset_provider_done_command(struct scsi_cmnd *scmd)
{
}

/*
 * Function:	scsi_reset_provider
 *
 * Purpose:	Send requested reset to a bus or device at any phase.
 *
 * Arguments:	device	- device to send reset to
 *		flag - reset type (see scsi.h)
 *
 * Returns:	SUCCESS/FAILURE.
 *
 * Notes:	This is used by the SCSI Generic driver to provide
 *		Bus/Device reset capability.
 */
int
scsi_reset_provider(struct scsi_device *dev, int flag)
{
	struct scsi_cmnd *scmd = scsi_get_command(dev, GFP_KERNEL);
	struct request req;
	int rtn;

	scmd->request = &req;
	memset(&scmd->eh_timeout, 0, sizeof(scmd->eh_timeout));
	scmd->request->rq_status      	= RQ_SCSI_BUSY;
	scmd->state                   	= SCSI_STATE_INITIALIZING;
	scmd->owner	     		= SCSI_OWNER_MIDLEVEL;
    
	memset(&scmd->cmnd, '\0', sizeof(scmd->cmnd));
    
	scmd->scsi_done		= scsi_reset_provider_done_command;
	scmd->done			= NULL;
	scmd->buffer			= NULL;
	scmd->bufflen			= 0;
	scmd->request_buffer		= NULL;
	scmd->request_bufflen		= 0;
	scmd->internal_timeout		= NORMAL_TIMEOUT;
	scmd->abort_reason		= DID_ABORT;

	scmd->cmd_len			= 0;

	scmd->sc_data_direction		= DMA_BIDIRECTIONAL;
	scmd->sc_request		= NULL;
	scmd->sc_magic			= SCSI_CMND_MAGIC;

	init_timer(&scmd->eh_timeout);

	/*
	 * Sometimes the command can get back into the timer chain,
	 * so use the pid as an identifier.
	 */
	scmd->pid			= 0;

	switch (flag) {
	case SCSI_TRY_RESET_DEVICE:
		rtn = scsi_try_bus_device_reset(scmd);
		if (rtn == SUCCESS)
			break;
		/* FALLTHROUGH */
	case SCSI_TRY_RESET_BUS:
		rtn = scsi_try_bus_reset(scmd);
		if (rtn == SUCCESS)
			break;
		/* FALLTHROUGH */
	case SCSI_TRY_RESET_HOST:
		rtn = scsi_try_host_reset(scmd);
		break;
	default:
		rtn = FAILED;
	}

	scsi_delete_timer(scmd);
	scsi_next_command(scmd);
	return rtn;
}