diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c
index 2ffdc9e0532ddbc8aae4a87cd039dfb3b4c495c5..38c3a291efacd532cfdba99d7e7c291fc8de0395 100644
--- a/drivers/scsi/hosts.c
+++ b/drivers/scsi/hosts.c
@@ -263,6 +263,10 @@ static void scsi_host_dev_release(struct device *dev)
 		kthread_stop(shost->ehandler);
 	if (shost->work_q)
 		destroy_workqueue(shost->work_q);
+	if (shost->uspace_req_q) {
+		kfree(shost->uspace_req_q->queuedata);
+		scsi_free_queue(shost->uspace_req_q);
+	}
 
 	scsi_destroy_command_freelist(shost);
 	if (shost->bqt)
diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index 780d6dc92b42537ec8282634bf5c3f87e8198547..fafc00deaadea6fab42609d673b03aa6c8f4e7ee 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -156,8 +156,7 @@ static struct scsi_host_cmd_pool scsi_cmd_dma_pool = {
 
 static DEFINE_MUTEX(host_cmd_pool_mutex);
 
-static struct scsi_cmnd *__scsi_get_command(struct Scsi_Host *shost,
-					    gfp_t gfp_mask)
+struct scsi_cmnd *__scsi_get_command(struct Scsi_Host *shost, gfp_t gfp_mask)
 {
 	struct scsi_cmnd *cmd;
 
@@ -178,6 +177,7 @@ static struct scsi_cmnd *__scsi_get_command(struct Scsi_Host *shost,
 
 	return cmd;
 }
+EXPORT_SYMBOL_GPL(__scsi_get_command);
 
 /*
  * Function:	scsi_get_command()
@@ -214,9 +214,29 @@ struct scsi_cmnd *scsi_get_command(struct scsi_device *dev, gfp_t gfp_mask)
 		put_device(&dev->sdev_gendev);
 
 	return cmd;
-}				
+}
 EXPORT_SYMBOL(scsi_get_command);
 
+void __scsi_put_command(struct Scsi_Host *shost, struct scsi_cmnd *cmd,
+			struct device *dev)
+{
+	unsigned long flags;
+
+	/* changing locks here, don't need to restore the irq state */
+	spin_lock_irqsave(&shost->free_list_lock, flags);
+	if (unlikely(list_empty(&shost->free_list))) {
+		list_add(&cmd->list, &shost->free_list);
+		cmd = NULL;
+	}
+	spin_unlock_irqrestore(&shost->free_list_lock, flags);
+
+	if (likely(cmd != NULL))
+		kmem_cache_free(shost->cmd_pool->slab, cmd);
+
+	put_device(dev);
+}
+EXPORT_SYMBOL(__scsi_put_command);
+
 /*
  * Function:	scsi_put_command()
  *
@@ -231,26 +251,15 @@ EXPORT_SYMBOL(scsi_get_command);
 void scsi_put_command(struct scsi_cmnd *cmd)
 {
 	struct scsi_device *sdev = cmd->device;
-	struct Scsi_Host *shost = sdev->host;
 	unsigned long flags;
-	
+
 	/* serious error if the command hasn't come from a device list */
 	spin_lock_irqsave(&cmd->device->list_lock, flags);
 	BUG_ON(list_empty(&cmd->list));
 	list_del_init(&cmd->list);
-	spin_unlock(&cmd->device->list_lock);
-	/* changing locks here, don't need to restore the irq state */
-	spin_lock(&shost->free_list_lock);
-	if (unlikely(list_empty(&shost->free_list))) {
-		list_add(&cmd->list, &shost->free_list);
-		cmd = NULL;
-	}
-	spin_unlock_irqrestore(&shost->free_list_lock, flags);
+	spin_unlock_irqrestore(&cmd->device->list_lock, flags);
 
-	if (likely(cmd != NULL))
-		kmem_cache_free(shost->cmd_pool->slab, cmd);
-
-	put_device(&sdev->sdev_gendev);
+	__scsi_put_command(cmd->device->host, cmd, &sdev->sdev_gendev);
 }
 EXPORT_SYMBOL(scsi_put_command);
 
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 2f12f9f12fcb269f5d55a11d9048b6d808bf5f12..fb616c69151f69a3d6697658de3bbc87827729b6 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -704,7 +704,7 @@ static struct scsi_cmnd *scsi_end_request(struct scsi_cmnd *cmd, int uptodate,
 	return NULL;
 }
 
-static struct scatterlist *scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_mask)
+struct scatterlist *scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_mask)
 {
 	struct scsi_host_sg_pool *sgp;
 	struct scatterlist *sgl;
@@ -745,7 +745,9 @@ static struct scatterlist *scsi_alloc_sgtable(struct scsi_cmnd *cmd, gfp_t gfp_m
 	return sgl;
 }
 
-static void scsi_free_sgtable(struct scatterlist *sgl, int index)
+EXPORT_SYMBOL(scsi_alloc_sgtable);
+
+void scsi_free_sgtable(struct scatterlist *sgl, int index)
 {
 	struct scsi_host_sg_pool *sgp;
 
@@ -755,6 +757,8 @@ static void scsi_free_sgtable(struct scatterlist *sgl, int index)
 	mempool_free(sgl, sgp->pool);
 }
 
+EXPORT_SYMBOL(scsi_free_sgtable);
+
 /*
  * Function:    scsi_release_buffers()
  *
@@ -1567,29 +1571,40 @@ u64 scsi_calculate_bounce_limit(struct Scsi_Host *shost)
 }
 EXPORT_SYMBOL(scsi_calculate_bounce_limit);
 
-struct request_queue *scsi_alloc_queue(struct scsi_device *sdev)
+struct request_queue *__scsi_alloc_queue(struct Scsi_Host *shost,
+					 request_fn_proc *request_fn)
 {
-	struct Scsi_Host *shost = sdev->host;
 	struct request_queue *q;
 
-	q = blk_init_queue(scsi_request_fn, NULL);
+	q = blk_init_queue(request_fn, NULL);
 	if (!q)
 		return NULL;
 
-	blk_queue_prep_rq(q, scsi_prep_fn);
-
 	blk_queue_max_hw_segments(q, shost->sg_tablesize);
 	blk_queue_max_phys_segments(q, SCSI_MAX_PHYS_SEGMENTS);
 	blk_queue_max_sectors(q, shost->max_sectors);
 	blk_queue_bounce_limit(q, scsi_calculate_bounce_limit(shost));
 	blk_queue_segment_boundary(q, shost->dma_boundary);
-	blk_queue_issue_flush_fn(q, scsi_issue_flush_fn);
-	blk_queue_softirq_done(q, scsi_softirq_done);
 
 	if (!shost->use_clustering)
 		clear_bit(QUEUE_FLAG_CLUSTER, &q->queue_flags);
 	return q;
 }
+EXPORT_SYMBOL(__scsi_alloc_queue);
+
+struct request_queue *scsi_alloc_queue(struct scsi_device *sdev)
+{
+	struct request_queue *q;
+
+	q = __scsi_alloc_queue(sdev->host, scsi_request_fn);
+	if (!q)
+		return NULL;
+
+	blk_queue_prep_rq(q, scsi_prep_fn);
+	blk_queue_issue_flush_fn(q, scsi_issue_flush_fn);
+	blk_queue_softirq_done(q, scsi_softirq_done);
+	return q;
+}
 
 void scsi_free_queue(struct request_queue *q)
 {
diff --git a/include/scsi/scsi_cmnd.h b/include/scsi/scsi_cmnd.h
index be117f812deb95c8064d98c3e9125223b88160a2..d6948d0e8cdb47d2347f2f2e9c9b991e7788b804 100644
--- a/include/scsi/scsi_cmnd.h
+++ b/include/scsi/scsi_cmnd.h
@@ -8,6 +8,7 @@
 
 struct request;
 struct scatterlist;
+struct Scsi_Host;
 struct scsi_device;
 
 
@@ -72,6 +73,9 @@ struct scsi_cmnd {
 	unsigned short use_sg;	/* Number of pieces of scatter-gather */
 	unsigned short sglist_len;	/* size of malloc'd scatter-gather list */
 
+	/* offset in cmd we are at (for multi-transfer tgt cmds) */
+	unsigned offset;
+
 	unsigned underflow;	/* Return error if less than
 				   this amount is transferred */
 
@@ -119,7 +123,10 @@ struct scsi_cmnd {
 };
 
 extern struct scsi_cmnd *scsi_get_command(struct scsi_device *, gfp_t);
+extern struct scsi_cmnd *__scsi_get_command(struct Scsi_Host *, gfp_t);
 extern void scsi_put_command(struct scsi_cmnd *);
+extern void __scsi_put_command(struct Scsi_Host *, struct scsi_cmnd *,
+			       struct device *);
 extern void scsi_io_completion(struct scsi_cmnd *, unsigned int);
 extern void scsi_finish_command(struct scsi_cmnd *cmd);
 extern void scsi_req_abort_cmd(struct scsi_cmnd *cmd);
@@ -128,4 +135,7 @@ extern void *scsi_kmap_atomic_sg(struct scatterlist *sg, int sg_count,
 				 size_t *offset, size_t *len);
 extern void scsi_kunmap_atomic_sg(void *virt);
 
+extern struct scatterlist *scsi_alloc_sgtable(struct scsi_cmnd *, gfp_t);
+extern void scsi_free_sgtable(struct scatterlist *, int);
+
 #endif /* _SCSI_SCSI_CMND_H */
diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h
index 6d8945d71c65af1cc8cf847a86be4454123f57c2..7f1f411d07af04c47f1faf7ce6400b57c197fef2 100644
--- a/include/scsi/scsi_host.h
+++ b/include/scsi/scsi_host.h
@@ -7,6 +7,7 @@
 #include <linux/workqueue.h>
 #include <linux/mutex.h>
 
+struct request_queue;
 struct block_device;
 struct completion;
 struct module;
@@ -123,6 +124,39 @@ struct scsi_host_template {
 	int (* queuecommand)(struct scsi_cmnd *,
 			     void (*done)(struct scsi_cmnd *));
 
+	/*
+	 * The transfer functions are used to queue a scsi command to
+	 * the LLD. When the driver is finished processing the command
+	 * the done callback is invoked.
+	 *
+	 * return values: see queuecommand
+	 *
+	 * If the LLD accepts the cmd, it should set the result to an
+	 * appropriate value when completed before calling the done function.
+	 *
+	 * STATUS: REQUIRED FOR TARGET DRIVERS
+	 */
+	/* TODO: rename */
+	int (* transfer_response)(struct scsi_cmnd *,
+				  void (*done)(struct scsi_cmnd *));
+	/*
+	 * This is called to inform the LLD to transfer cmd->request_bufflen
+	 * bytes of the cmd at cmd->offset in the cmd. The cmd->use_sg
+	 * speciefies the number of scatterlist entried in the command
+	 * and cmd->request_buffer contains the scatterlist.
+	 *
+	 * If the command cannot be processed in one transfer_data call
+	 * becuase a scatterlist within the LLD's limits cannot be
+	 * created then transfer_data will be called multiple times.
+	 * It is initially called from process context, and later
+	 * calls are from the interrup context.
+	 */
+	int (* transfer_data)(struct scsi_cmnd *,
+			      void (*done)(struct scsi_cmnd *));
+
+	/* Used as callback for the completion of task management request. */
+	int (* tsk_mgmt_response)(u64 mid, int result);
+
 	/*
 	 * This is an error handling strategy routine.  You don't need to
 	 * define one of these if you don't want to - there is a default
@@ -589,6 +623,12 @@ struct Scsi_Host {
 	 */
 	unsigned int max_host_blocked;
 
+	/*
+	 * q used for scsi_tgt msgs, async events or any other requests that
+	 * need to be processed in userspace
+	 */
+	struct request_queue *uspace_req_q;
+
 	/* legacy crap */
 	unsigned long base;
 	unsigned long io_port;
@@ -687,6 +727,9 @@ extern void scsi_unblock_requests(struct Scsi_Host *);
 extern void scsi_block_requests(struct Scsi_Host *);
 
 struct class_container;
+
+extern struct request_queue *__scsi_alloc_queue(struct Scsi_Host *shost,
+						void (*) (struct request_queue *));
 /*
  * These two functions are used to allocate and free a pseudo device
  * which will connect to the host adapter itself rather than any