fix mismerge in ll_rw_blk.c

drivers/block/ll_rw_blk.c

@@ -284,6 +284,7 @@ static inline void rq_init(request_queue_t *q, struct request *rq)
 	rq->special = NULL;
 	rq->data_len = 0;
 	rq->data = NULL;
+	rq->nr_phys_segments = 0;
 	rq->sense = NULL;
 	rq->end_io = NULL;
 	rq->end_io_data = NULL;
@@ -2115,7 +2116,7 @@ EXPORT_SYMBOL(blk_insert_request);
 /**
  * blk_rq_map_user - map user data to a request, for REQ_BLOCK_PC usage
  * @q:		request queue where request should be inserted
- * @rw:		READ or WRITE data
+ * @rq:		request structure to fill
  * @ubuf:	the user buffer
  * @len:	length of user data
  *
@@ -2132,21 +2133,19 @@ EXPORT_SYMBOL(blk_insert_request);
  *    original bio must be passed back in to blk_rq_unmap_user() for proper
  *    unmapping.
  */
-struct request *blk_rq_map_user(request_queue_t *q, int rw, void __user *ubuf,
-				unsigned int len)
+int blk_rq_map_user(request_queue_t *q, struct request *rq, void __user *ubuf,
+		    unsigned int len)
 {
 	unsigned long uaddr;
-	struct request *rq;
 	struct bio *bio;
+	int reading;
 
 	if (len > (q->max_sectors << 9))
-		return ERR_PTR(-EINVAL);
-	if ((!len && ubuf) || (len && !ubuf))
-		return ERR_PTR(-EINVAL);
+		return -EINVAL;
+	if (!len || !ubuf)
+		return -EINVAL;
 
-	rq = blk_get_request(q, rw, __GFP_WAIT);
-	if (!rq)
-		return ERR_PTR(-ENOMEM);
+	reading = rq_data_dir(rq) == READ;
 
 	/*
 	 * if alignment requirement is satisfied, map in user pages for
@@ -2154,9 +2153,9 @@ struct request *blk_rq_map_user(request_queue_t *q, int rw, void __user *ubuf,
 	 */
 	uaddr = (unsigned long) ubuf;
 	if (!(uaddr & queue_dma_alignment(q)) && !(len & queue_dma_alignment(q)))
-		bio = bio_map_user(q, NULL, uaddr, len, rw == READ);
+		bio = bio_map_user(q, NULL, uaddr, len, reading);
 	else
-		bio = bio_copy_user(q, uaddr, len, rw == READ);
+		bio = bio_copy_user(q, uaddr, len, reading);
 
 	if (!IS_ERR(bio)) {
 		rq->bio = rq->biotail = bio;
@@ -2164,28 +2163,70 @@ struct request *blk_rq_map_user(request_queue_t *q, int rw, void __user *ubuf,
 
 		rq->buffer = rq->data = NULL;
 		rq->data_len = len;
-		return rq;
+		return 0;
 	}
 
 	/*
 	 * bio is the err-ptr
 	 */
-	blk_put_request(rq);
-	return (struct request *) bio;
+	return PTR_ERR(bio);
 }
 
 EXPORT_SYMBOL(blk_rq_map_user);
 
+/**
+ * blk_rq_map_user_iov - map user data to a request, for REQ_BLOCK_PC usage
+ * @q:		request queue where request should be inserted
+ * @rq:		request to map data to
+ * @iov:	pointer to the iovec
+ * @iov_count:	number of elements in the iovec
+ *
+ * Description:
+ *    Data will be mapped directly for zero copy io, if possible. Otherwise
+ *    a kernel bounce buffer is used.
+ *
+ *    A matching blk_rq_unmap_user() must be issued at the end of io, while
+ *    still in process context.
+ *
+ *    Note: The mapped bio may need to be bounced through blk_queue_bounce()
+ *    before being submitted to the device, as pages mapped may be out of
+ *    reach. It's the callers responsibility to make sure this happens. The
+ *    original bio must be passed back in to blk_rq_unmap_user() for proper
+ *    unmapping.
+ */
+int blk_rq_map_user_iov(request_queue_t *q, struct request *rq,
+			struct sg_iovec *iov, int iov_count)
+{
+	struct bio *bio;
+
+	if (!iov || iov_count <= 0)
+		return -EINVAL;
+
+	/* we don't allow misaligned data like bio_map_user() does.  If the
+	 * user is using sg, they're expected to know the alignment constraints
+	 * and respect them accordingly */
+	bio = bio_map_user_iov(q, NULL, iov, iov_count, rq_data_dir(rq)== READ);
+	if (IS_ERR(bio))
+		return PTR_ERR(bio);
+
+	rq->bio = rq->biotail = bio;
+	blk_rq_bio_prep(q, rq, bio);
+	rq->buffer = rq->data = NULL;
+	rq->data_len = bio->bi_size;
+	return 0;
+}
+
+EXPORT_SYMBOL(blk_rq_map_user_iov);
+
 /**
  * blk_rq_unmap_user - unmap a request with user data
- * @rq:		request to be unmapped
- * @bio:	bio for the request
+ * @bio:	bio to be unmapped
  * @ulen:	length of user buffer
  *
  * Description:
- *    Unmap a request previously mapped by blk_rq_map_user().
+ *    Unmap a bio previously mapped by blk_rq_map_user().
  */
-int blk_rq_unmap_user(struct request *rq, struct bio *bio, unsigned int ulen)
+int blk_rq_unmap_user(struct bio *bio, unsigned int ulen)
 {
 	int ret = 0;
 
@@ -2196,31 +2237,89 @@ int blk_rq_unmap_user(struct request *rq, struct bio *bio, unsigned int ulen)
 			ret = bio_uncopy_user(bio);
 	}
 
-	blk_put_request(rq);
-	return ret;
+	return 0;
 }
 
 EXPORT_SYMBOL(blk_rq_unmap_user);
 
+/**
+ * blk_rq_map_kern - map kernel data to a request, for REQ_BLOCK_PC usage
+ * @q:		request queue where request should be inserted
+ * @rq:		request to fill
+ * @kbuf:	the kernel buffer
+ * @len:	length of user data
+ * @gfp_mask:	memory allocation flags
+ */
+int blk_rq_map_kern(request_queue_t *q, struct request *rq, void *kbuf,
+		    unsigned int len, unsigned int gfp_mask)
+{
+	struct bio *bio;
+
+	if (len > (q->max_sectors << 9))
+		return -EINVAL;
+	if (!len || !kbuf)
+		return -EINVAL;
+
+	bio = bio_map_kern(q, kbuf, len, gfp_mask);
+	if (IS_ERR(bio))
+		return PTR_ERR(bio);
+
+	if (rq_data_dir(rq) == WRITE)
+		bio->bi_rw |= (1 << BIO_RW);
+
+	rq->bio = rq->biotail = bio;
+	blk_rq_bio_prep(q, rq, bio);
+
+	rq->buffer = rq->data = NULL;
+	rq->data_len = len;
+	return 0;
+}
+
+EXPORT_SYMBOL(blk_rq_map_kern);
+
+/**
+ * blk_execute_rq_nowait - insert a request into queue for execution
+ * @q:		queue to insert the request in
+ * @bd_disk:	matching gendisk
+ * @rq:		request to insert
+ * @at_head:    insert request at head or tail of queue
+ * @done:	I/O completion handler
+ *
+ * Description:
+ *    Insert a fully prepared request at the back of the io scheduler queue
+ *    for execution.  Don't wait for completion.
+ */
+void blk_execute_rq_nowait(request_queue_t *q, struct gendisk *bd_disk,
+			   struct request *rq, int at_head,
+			   void (*done)(struct request *))
+{
+	int where = at_head ? ELEVATOR_INSERT_FRONT : ELEVATOR_INSERT_BACK;
+
+	rq->rq_disk = bd_disk;
+	rq->flags |= REQ_NOMERGE;
+	rq->end_io = done;
+	elv_add_request(q, rq, where, 1);
+	generic_unplug_device(q);
+}
+
 /**
  * blk_execute_rq - insert a request into queue for execution
  * @q:		queue to insert the request in
  * @bd_disk:	matching gendisk
  * @rq:		request to insert
+ * @at_head:    insert request at head or tail of queue
  *
  * Description:
  *    Insert a fully prepared request at the back of the io scheduler queue
- *    for execution.
+ *    for execution and wait for completion.
  */
 int blk_execute_rq(request_queue_t *q, struct gendisk *bd_disk,
-		   struct request *rq)
+		   struct request *rq, int at_head)
 {
 	DECLARE_COMPLETION(wait);
 	char sense[SCSI_SENSE_BUFFERSIZE];
 	int err = 0;
 
-	rq->rq_disk = bd_disk;
-
 	/*
 	 * we need an extra reference to the request, so we can look at
 	 * it after io completion
@@ -2233,11 +2332,8 @@ int blk_execute_rq(request_queue_t *q, struct gendisk *bd_disk,
 		rq->sense_len = 0;
 	}
 
-	rq->flags |= REQ_NOMERGE;
 	rq->waiting = &wait;
-	rq->end_io = blk_end_sync_rq;
-	elv_add_request(q, rq, ELEVATOR_INSERT_BACK, 1);
-	generic_unplug_device(q);
+	blk_execute_rq_nowait(q, bd_disk, rq, at_head, blk_end_sync_rq);
 	wait_for_completion(&wait);
 	rq->waiting = NULL;
 
@@ -2277,6 +2373,44 @@ int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector)
 
 EXPORT_SYMBOL(blkdev_issue_flush);
 
+/**
+ * blkdev_scsi_issue_flush_fn - issue flush for SCSI devices
+ * @q:		device queue
+ * @disk:	gendisk
+ * @error_sector:	error offset
+ *
+ * Description:
+ *    Devices understanding the SCSI command set, can use this function as
+ *    a helper for issuing a cache flush. Note: driver is required to store
+ *    the error offset (in case of error flushing) in ->sector of struct
+ *    request.
+ */
+int blkdev_scsi_issue_flush_fn(request_queue_t *q, struct gendisk *disk,
+			       sector_t *error_sector)
+{
+	struct request *rq = blk_get_request(q, WRITE, __GFP_WAIT);
+	int ret;
+
+	rq->flags |= REQ_BLOCK_PC | REQ_SOFTBARRIER;
+	rq->sector = 0;
+	memset(rq->cmd, 0, sizeof(rq->cmd));
+	rq->cmd[0] = 0x35;
+	rq->cmd_len = 12;
+	rq->data = NULL;
+	rq->data_len = 0;
+	rq->timeout = 60 * HZ;
+
+	ret = blk_execute_rq(q, disk, rq, 0);
+
+	if (ret && error_sector)
+		*error_sector = rq->sector;
+
+	blk_put_request(rq);
+	return ret;
+}
+
+EXPORT_SYMBOL(blkdev_scsi_issue_flush_fn);
+
 static void drive_stat_acct(struct request *rq, int nr_sectors, int new_io)
 {
 	int rw = rq_data_dir(rq);

drivers/block/scsi_ioctl.c

@@ -216,7 +216,7 @@ static int sg_io(struct file *file, request_queue_t *q,
 		struct gendisk *bd_disk, struct sg_io_hdr *hdr)
 {
 	unsigned long start_time;
-	int reading, writing;
+	int writing = 0, ret = 0;
 	struct request *rq;
 	struct bio *bio;
 	char sense[SCSI_SENSE_BUFFERSIZE];
@@ -231,38 +231,48 @@ static int sg_io(struct file *file, request_queue_t *q,
 	if (verify_command(file, cmd))
 		return -EPERM;
 
-	/*
-	 * we'll do that later
-	 */
-	if (hdr->iovec_count)
-		return -EOPNOTSUPP;
-
 	if (hdr->dxfer_len > (q->max_sectors << 9))
 		return -EIO;
 
-	reading = writing = 0;
-	if (hdr->dxfer_len) {
+	if (hdr->dxfer_len)
 		switch (hdr->dxfer_direction) {
 		default:
 			return -EINVAL;
 		case SG_DXFER_TO_FROM_DEV:
-			reading = 1;
-			/* fall through */
 		case SG_DXFER_TO_DEV:
 			writing = 1;
 			break;
 		case SG_DXFER_FROM_DEV:
-			reading = 1;
 			break;
 		}
 
-		rq = blk_rq_map_user(q, writing ? WRITE : READ, hdr->dxferp,
-				     hdr->dxfer_len);
+	rq = blk_get_request(q, writing ? WRITE : READ, GFP_KERNEL);
+	if (!rq)
+		return -ENOMEM;
+
+	if (hdr->iovec_count) {
+		const int size = sizeof(struct sg_iovec) * hdr->iovec_count;
+		struct sg_iovec *iov;
+
+		iov = kmalloc(size, GFP_KERNEL);
+		if (!iov) {
+			ret = -ENOMEM;
+			goto out;
+		}
+
+		if (copy_from_user(iov, hdr->dxferp, size)) {
+			kfree(iov);
+			ret = -EFAULT;
+			goto out;
+		}
+
+		ret = blk_rq_map_user_iov(q, rq, iov, hdr->iovec_count);
+		kfree(iov);
+	} else if (hdr->dxfer_len)
+		ret = blk_rq_map_user(q, rq, hdr->dxferp, hdr->dxfer_len);
 
-		if (IS_ERR(rq))
-			return PTR_ERR(rq);
-	} else
-		rq = blk_get_request(q, READ, __GFP_WAIT);
+	if (ret)
+		goto out;
 
 	/*
 	 * fill in request structure
@@ -298,7 +308,7 @@ static int sg_io(struct file *file, request_queue_t *q,
 	 * (if he doesn't check that is his problem).
 	 * N.B. a non-zero SCSI status is _not_ necessarily an error.
 	 */
-	blk_execute_rq(q, bd_disk, rq);
+	blk_execute_rq(q, bd_disk, rq, 0);
 
 	/* write to all output members */
 	hdr->status = 0xff & rq->errors;
@@ -320,12 +330,14 @@ static int sg_io(struct file *file, request_queue_t *q,
 			hdr->sb_len_wr = len;
 	}
 
-	if (blk_rq_unmap_user(rq, bio, hdr->dxfer_len))
-		return -EFAULT;
+	if (blk_rq_unmap_user(bio, hdr->dxfer_len))
+		ret = -EFAULT;
 
 	/* may not have succeeded, but output values written to control
 	 * structure (struct sg_io_hdr).  */
-	return 0;
+out:
+	blk_put_request(rq);
+	return ret;
 }
 
 #define OMAX_SB_LEN 16          /* For backward compatibility */
@@ -408,7 +420,7 @@ static int sg_scsi_ioctl(struct file *file, request_queue_t *q,
 	rq->data_len = bytes;
 	rq->flags |= REQ_BLOCK_PC;
 
-	blk_execute_rq(q, bd_disk, rq);
+	blk_execute_rq(q, bd_disk, rq, 0);
 	err = rq->errors & 0xff;	/* only 8 bit SCSI status */
 	if (err) {
 		if (rq->sense_len && rq->sense) {
@@ -561,7 +573,7 @@ int scsi_cmd_ioctl(struct file *file, struct gendisk *bd_disk, unsigned int cmd,
 			rq->cmd[0] = GPCMD_START_STOP_UNIT;
 			rq->cmd[4] = 0x02 + (close != 0);
 			rq->cmd_len = 6;
-			err = blk_execute_rq(q, bd_disk, rq);
+			err = blk_execute_rq(q, bd_disk, rq, 0);
 			blk_put_request(rq);
 			break;
 		default:

drivers/cdrom/cdrom.c

@@ -2097,6 +2097,10 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf,
 	if (!q)
 		return -ENXIO;
 
+	rq = blk_get_request(q, READ, GFP_KERNEL);
+	if (!rq)
+		return -ENOMEM;
+
 	cdi->last_sense = 0;
 
 	while (nframes) {
@@ -2108,9 +2112,9 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf,
 
 		len = nr * CD_FRAMESIZE_RAW;
 
-		rq = blk_rq_map_user(q, READ, ubuf, len);
-		if (IS_ERR(rq))
-			return PTR_ERR(rq);
+		ret = blk_rq_map_user(q, rq, ubuf, len);
+		if (ret)
+			break;
 
 		memset(rq->cmd, 0, sizeof(rq->cmd));
 		rq->cmd[0] = GPCMD_READ_CD;
@@ -2132,13 +2136,13 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf,
 		if (rq->bio)
 			blk_queue_bounce(q, &rq->bio);
 
-		if (blk_execute_rq(q, cdi->disk, rq)) {
+		if (blk_execute_rq(q, cdi->disk, rq, 0)) {
 			struct request_sense *s = rq->sense;
 			ret = -EIO;
 			cdi->last_sense = s->sense_key;
 		}
 
-		if (blk_rq_unmap_user(rq, bio, len))
+		if (blk_rq_unmap_user(bio, len))
 			ret = -EFAULT;
 
 		if (ret)
@@ -2149,6 +2153,7 @@ static int cdrom_read_cdda_bpc(struct cdrom_device_info *cdi, __u8 __user *ubuf,
 		ubuf += len;
 	}
 
+	blk_put_request(rq);
 	return ret;
 }
 

drivers/ide/ide-disk.c

@@ -754,7 +754,7 @@ static int idedisk_issue_flush(request_queue_t *q, struct gendisk *disk,
 
 	idedisk_prepare_flush(q, rq);
 
-	ret = blk_execute_rq(q, disk, rq);
+	ret = blk_execute_rq(q, disk, rq, 0);
 
 	/*
 	 * if we failed and caller wants error offset, get it

fs/bio.c

@@ -25,6 +25,7 @@
 #include <linux/module.h>
 #include <linux/mempool.h>
 #include <linux/workqueue.h>
+#include <scsi/sg.h>		/* for struct sg_iovec */
 
 #define BIO_POOL_SIZE 256
 
@@ -546,22 +547,34 @@ struct bio *bio_copy_user(request_queue_t *q, unsigned long uaddr,
 	return ERR_PTR(ret);
 }
 
-static struct bio *__bio_map_user(request_queue_t *q, struct block_device *bdev,
-				  unsigned long uaddr, unsigned int len,
-				  int write_to_vm)
+static struct bio *__bio_map_user_iov(request_queue_t *q,
+				      struct block_device *bdev,
+				      struct sg_iovec *iov, int iov_count,
+				      int write_to_vm)
 {
-	unsigned long end = (uaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
-	unsigned long start = uaddr >> PAGE_SHIFT;
-	const int nr_pages = end - start;
-	int ret, offset, i;
+	int i, j;
+	int nr_pages = 0;
 	struct page **pages;
 	struct bio *bio;
+	int cur_page = 0;
+	int ret, offset;
 
-	/*
-	 * transfer and buffer must be aligned to at least hardsector
-	 * size for now, in the future we can relax this restriction
-	 */
-	if ((uaddr & queue_dma_alignment(q)) || (len & queue_dma_alignment(q)))
+	for (i = 0; i < iov_count; i++) {
+		unsigned long uaddr = (unsigned long)iov[i].iov_base;
+		unsigned long len = iov[i].iov_len;
+		unsigned long end = (uaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+		unsigned long start = uaddr >> PAGE_SHIFT;
+
+		nr_pages += end - start;
+		/*
+		 * transfer and buffer must be aligned to at least hardsector
+		 * size for now, in the future we can relax this restriction
+		 */
+		if ((uaddr & queue_dma_alignment(q)) || (len & queue_dma_alignment(q)))
+			return ERR_PTR(-EINVAL);
+	}
+
+	if (!nr_pages)
 		return ERR_PTR(-EINVAL);
 
 	bio = bio_alloc(GFP_KERNEL, nr_pages);
@@ -573,42 +586,54 @@ static struct bio *__bio_map_user(request_queue_t *q, struct block_device *bdev,
 	if (!pages)
 		goto out;
 
-	down_read(&current->mm->mmap_sem);
-	ret = get_user_pages(current, current->mm, uaddr, nr_pages,
-						write_to_vm, 0, pages, NULL);
-	up_read(&current->mm->mmap_sem);
-
-	if (ret < nr_pages)
-		goto out;
-
-	bio->bi_bdev = bdev;
-
-	offset = uaddr & ~PAGE_MASK;
-	for (i = 0; i < nr_pages; i++) {
-		unsigned int bytes = PAGE_SIZE - offset;
-
-		if (len <= 0)
-			break;
-
-		if (bytes > len)
-			bytes = len;
+	memset(pages, 0, nr_pages * sizeof(struct page *));
+
+	for (i = 0; i < iov_count; i++) {
+		unsigned long uaddr = (unsigned long)iov[i].iov_base;
+		unsigned long len = iov[i].iov_len;
+		unsigned long end = (uaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+		unsigned long start = uaddr >> PAGE_SHIFT;
+		const int local_nr_pages = end - start;
+		const int page_limit = cur_page + local_nr_pages;
+		
+		down_read(&current->mm->mmap_sem);
+		ret = get_user_pages(current, current->mm, uaddr,
+				     local_nr_pages,
+				     write_to_vm, 0, &pages[cur_page], NULL);
+		up_read(&current->mm->mmap_sem);
+
+		if (ret < local_nr_pages)
+			goto out_unmap;
+
+
+		offset = uaddr & ~PAGE_MASK;
+		for (j = cur_page; j < page_limit; j++) {
+			unsigned int bytes = PAGE_SIZE - offset;
+
+			if (len <= 0)
+				break;
+			
+			if (bytes > len)
+				bytes = len;
+
+			/*
+			 * sorry...
+			 */
+			if (__bio_add_page(q, bio, pages[j], bytes, offset) < bytes)
+				break;
+
+			len -= bytes;
+			offset = 0;
+		}
 
+		cur_page = j;
 		/*
-		 * sorry...
+		 * release the pages we didn't map into the bio, if any
 		 */
-		if (__bio_add_page(q, bio, pages[i], bytes, offset) < bytes)
-			break;
-
-		len -= bytes;
-		offset = 0;
+		while (j < page_limit)
+			page_cache_release(pages[j++]);
 	}
 
-	/*
-	 * release the pages we didn't map into the bio, if any
-	 */
-	while (i < nr_pages)
-		page_cache_release(pages[i++]);
-
 	kfree(pages);
 
 	/*
@@ -617,9 +642,17 @@ static struct bio *__bio_map_user(request_queue_t *q, struct block_device *bdev,
 	if (!write_to_vm)
 		bio->bi_rw |= (1 << BIO_RW);
 
+	bio->bi_bdev = bdev;
 	bio->bi_flags |= (1 << BIO_USER_MAPPED);
 	return bio;
-out:
+
+ out_unmap:
+	for (i = 0; i < nr_pages; i++) {
+		if(!pages[i])
+			break;
+		page_cache_release(pages[i]);
+	}
+ out:
 	kfree(pages);
 	bio_put(bio);
 	return ERR_PTR(ret);
@@ -638,10 +671,34 @@ static struct bio *__bio_map_user(request_queue_t *q, struct block_device *bdev,
  */
 struct bio *bio_map_user(request_queue_t *q, struct block_device *bdev,
 			 unsigned long uaddr, unsigned int len, int write_to_vm)
+{
+	struct sg_iovec iov;
+
+	iov.iov_base = (__user void *)uaddr;
+	iov.iov_len = len;
+
+	return bio_map_user_iov(q, bdev, &iov, 1, write_to_vm);
+}
+
+/**
+ *	bio_map_user_iov - map user sg_iovec table into bio
+ *	@q: the request_queue_t for the bio
+ *	@bdev: destination block device
+ *	@iov:	the iovec.
+ *	@iov_count: number of elements in the iovec
+ *	@write_to_vm: bool indicating writing to pages or not
+ *
+ *	Map the user space address into a bio suitable for io to a block
+ *	device. Returns an error pointer in case of error.
+ */
+struct bio *bio_map_user_iov(request_queue_t *q, struct block_device *bdev,
+			     struct sg_iovec *iov, int iov_count,
+			     int write_to_vm)
 {
 	struct bio *bio;
+	int len = 0, i;
 
-	bio = __bio_map_user(q, bdev, uaddr, len, write_to_vm);
+	bio = __bio_map_user_iov(q, bdev, iov, iov_count, write_to_vm);
 
 	if (IS_ERR(bio))
 		return bio;
@@ -654,6 +711,9 @@ struct bio *bio_map_user(request_queue_t *q, struct block_device *bdev,
 	 */
 	bio_get(bio);
 
+	for (i = 0; i < iov_count; i++)
+		len += iov[i].iov_len;
+
 	if (bio->bi_size == len)
 		return bio;
 
@@ -698,6 +758,82 @@ void bio_unmap_user(struct bio *bio)
 	bio_put(bio);
 }
 
+static int bio_map_kern_endio(struct bio *bio, unsigned int bytes_done, int err)
+{
+	if (bio->bi_size)
+		return 1;
+
+	bio_put(bio);
+	return 0;
+}
+
+
+static struct bio *__bio_map_kern(request_queue_t *q, void *data,
+				  unsigned int len, unsigned int gfp_mask)
+{
+	unsigned long kaddr = (unsigned long)data;
+	unsigned long end = (kaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	unsigned long start = kaddr >> PAGE_SHIFT;
+	const int nr_pages = end - start;
+	int offset, i;
+	struct bio *bio;
+
+	bio = bio_alloc(gfp_mask, nr_pages);
+	if (!bio)
+		return ERR_PTR(-ENOMEM);
+
+	offset = offset_in_page(kaddr);
+	for (i = 0; i < nr_pages; i++) {
+		unsigned int bytes = PAGE_SIZE - offset;
+
+		if (len <= 0)
+			break;
+
+		if (bytes > len)
+			bytes = len;
+
+		if (__bio_add_page(q, bio, virt_to_page(data), bytes,
+				   offset) < bytes)
+			break;
+
+		data += bytes;
+		len -= bytes;
+		offset = 0;
+	}
+
+	bio->bi_end_io = bio_map_kern_endio;
+	return bio;
+}
+
+/**
+ *	bio_map_kern	-	map kernel address into bio
+ *	@q: the request_queue_t for the bio
+ *	@data: pointer to buffer to map
+ *	@len: length in bytes
+ *	@gfp_mask: allocation flags for bio allocation
+ *
+ *	Map the kernel address into a bio suitable for io to a block
+ *	device. Returns an error pointer in case of error.
+ */
+struct bio *bio_map_kern(request_queue_t *q, void *data, unsigned int len,
+			 unsigned int gfp_mask)
+{
+	struct bio *bio;
+
+	bio = __bio_map_kern(q, data, len, gfp_mask);
+	if (IS_ERR(bio))
+		return bio;
+
+	if (bio->bi_size == len)
+		return bio;
+
+	/*
+	 * Don't support partial mappings.
+	 */
+	bio_put(bio);
+	return ERR_PTR(-EINVAL);
+}
+
 /*
  * bio_set_pages_dirty() and bio_check_pages_dirty() are support functions
  * for performing direct-IO in BIOs.
@@ -1085,6 +1221,7 @@ EXPORT_SYMBOL(bio_add_page);
 EXPORT_SYMBOL(bio_get_nr_vecs);
 EXPORT_SYMBOL(bio_map_user);
 EXPORT_SYMBOL(bio_unmap_user);
+EXPORT_SYMBOL(bio_map_kern);
 EXPORT_SYMBOL(bio_pair_release);
 EXPORT_SYMBOL(bio_split);
 EXPORT_SYMBOL(bio_split_pool);

include/linux/bio.h

@@ -295,7 +295,13 @@ extern int bio_add_page(struct bio *, struct page *, unsigned int,unsigned int);
 extern int bio_get_nr_vecs(struct block_device *);
 extern struct bio *bio_map_user(struct request_queue *, struct block_device *,
 				unsigned long, unsigned int, int);
+struct sg_iovec;
+extern struct bio *bio_map_user_iov(struct request_queue *,
+				    struct block_device *,
+				    struct sg_iovec *, int, int);
 extern void bio_unmap_user(struct bio *);
+extern struct bio *bio_map_kern(struct request_queue *, void *, unsigned int,
+				unsigned int);
 extern void bio_set_pages_dirty(struct bio *bio);
 extern void bio_check_pages_dirty(struct bio *bio);
 extern struct bio *bio_copy_user(struct request_queue *, unsigned long, unsigned int, int);

include/linux/blkdev.h

@@ -563,10 +563,12 @@ extern void blk_sync_queue(struct request_queue *q);
 extern void __blk_stop_queue(request_queue_t *q);
 extern void blk_run_queue(request_queue_t *);
 extern void blk_queue_activity_fn(request_queue_t *, activity_fn *, void *);
-extern struct request *blk_rq_map_user(request_queue_t *, int, void __user *, unsigned int);
-extern int blk_rq_unmap_user(struct request *, struct bio *, unsigned int);
-extern int blk_execute_rq(request_queue_t *, struct gendisk *, struct request *);
-
+extern int blk_rq_map_user(request_queue_t *, struct request *, void __user *, unsigned int);
+extern int blk_rq_unmap_user(struct bio *, unsigned int);
+extern int blk_rq_map_kern(request_queue_t *, struct request *, void *, unsigned int, unsigned int);
+extern int blk_rq_map_user_iov(request_queue_t *, struct request *, struct sg_iovec *, int);
+extern int blk_execute_rq(request_queue_t *, struct gendisk *,
+			  struct request *, int);
 static inline request_queue_t *bdev_get_queue(struct block_device *bdev)
 {
 	return bdev->bd_disk->queue;