X-Git-Url: http://pilppa.org/gitweb/gitweb.cgi?a=blobdiff_plain;f=drivers%2Fmd%2Fdm.c;h=11a98df298ece4db805682edc80d9860635f5cc4;hb=c02929c278f2bca68635e4c2daa00b7825d71061;hp=aeb63a3ac3308f68b9e40be73a7db75048d76108;hpb=8757b7764f13e336f3c0eb1f634440d4ee4c3a67;p=linux-2.6-omap-h63xx.git

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index aeb63a3ac33..11a98df298e 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -68,10 +68,12 @@ union map_info *dm_get_mapinfo(struct bio *bio)
 #define DMF_FROZEN 2
 #define DMF_FREEING 3
 #define DMF_DELETING 4
+#define DMF_NOFLUSH_SUSPENDING 5
 
 struct mapped_device {
 	struct rw_semaphore io_lock;
 	struct semaphore suspend_lock;
+	spinlock_t pushback_lock;
 	rwlock_t map_lock;
 	atomic_t holders;
 	atomic_t open_count;
@@ -89,7 +91,8 @@ struct mapped_device {
 	 */
 	atomic_t pending;
 	wait_queue_head_t wait;
- 	struct bio_list deferred;
+	struct bio_list deferred;
+	struct bio_list pushback;
 
 	/*
 	 * The current mapping.
@@ -102,6 +105,8 @@ struct mapped_device {
 	mempool_t *io_pool;
 	mempool_t *tio_pool;
 
+	struct bio_set *bs;
+
 	/*
 	 * Event handling.
 	 */
@@ -119,19 +124,13 @@ struct mapped_device {
 };
 
 #define MIN_IOS 256
-static kmem_cache_t *_io_cache;
-static kmem_cache_t *_tio_cache;
-
-static struct bio_set *dm_set;
+static struct kmem_cache *_io_cache;
+static struct kmem_cache *_tio_cache;
 
 static int __init local_init(void)
 {
 	int r;
 
-	dm_set = bioset_create(16, 16, 4);
-	if (!dm_set)
-		return -ENOMEM;
-
 	/* allocate a slab for the dm_ios */
 	_io_cache = kmem_cache_create("dm_io",
 				      sizeof(struct dm_io), 0, 0, NULL, NULL);
@@ -165,8 +164,6 @@ static void local_exit(void)
 	kmem_cache_destroy(_tio_cache);
 	kmem_cache_destroy(_io_cache);
 
-	bioset_free(dm_set);
-
 	if (unregister_blkdev(_major, _name) < 0)
 		DMERR("unregister_blkdev failed");
 
@@ -450,23 +447,50 @@ int dm_set_geometry(struct mapped_device *md, struct hd_geometry *geo)
  *   you this clearly demarcated crap.
  *---------------------------------------------------------------*/
 
+static int __noflush_suspending(struct mapped_device *md)
+{
+	return test_bit(DMF_NOFLUSH_SUSPENDING, &md->flags);
+}
+
 /*
  * Decrements the number of outstanding ios that a bio has been
  * cloned into, completing the original io if necc.
  */
 static void dec_pending(struct dm_io *io, int error)
 {
-	if (error)
+	unsigned long flags;
+
+	/* Push-back supersedes any I/O errors */
+	if (error && !(io->error > 0 && __noflush_suspending(io->md)))
 		io->error = error;
 
 	if (atomic_dec_and_test(&io->io_count)) {
+		if (io->error == DM_ENDIO_REQUEUE) {
+			/*
+			 * Target requested pushing back the I/O.
+			 * This must be handled before the sleeper on
+			 * suspend queue merges the pushback list.
+			 */
+			spin_lock_irqsave(&io->md->pushback_lock, flags);
+			if (__noflush_suspending(io->md))
+				bio_list_add(&io->md->pushback, io->bio);
+			else
+				/* noflush suspend was interrupted. */
+				io->error = -EIO;
+			spin_unlock_irqrestore(&io->md->pushback_lock, flags);
+		}
+
 		if (end_io_acct(io))
 			/* nudge anyone waiting on suspend queue */
 			wake_up(&io->md->wait);
 
-		blk_add_trace_bio(io->md->queue, io->bio, BLK_TA_COMPLETE);
+		if (io->error != DM_ENDIO_REQUEUE) {
+			blk_add_trace_bio(io->md->queue, io->bio,
+					  BLK_TA_COMPLETE);
+
+			bio_endio(io->bio, io->bio->bi_size, io->error);
+		}
 
-		bio_endio(io->bio, io->bio->bi_size, io->error);
 		free_io(io->md, io);
 	}
 }
@@ -475,7 +499,7 @@ static int clone_endio(struct bio *bio, unsigned int done, int error)
 {
 	int r = 0;
 	struct target_io *tio = bio->bi_private;
-	struct dm_io *io = tio->io;
+	struct mapped_device *md = tio->io->md;
 	dm_endio_fn endio = tio->ti->type->end_io;
 
 	if (bio->bi_size)
@@ -486,17 +510,30 @@ static int clone_endio(struct bio *bio, unsigned int done, int error)
 
 	if (endio) {
 		r = endio(tio->ti, bio, error, &tio->info);
-		if (r < 0)
+		if (r < 0 || r == DM_ENDIO_REQUEUE)
+			/*
+			 * error and requeue request are handled
+			 * in dec_pending().
+			 */
 			error = r;
-
-		else if (r > 0)
-			/* the target wants another shot at the io */
+		else if (r == DM_ENDIO_INCOMPLETE)
+			/* The target will handle the io */
 			return 1;
+		else if (r) {
+			DMWARN("unimplemented target endio return value: %d", r);
+			BUG();
+		}
 	}
 
-	free_tio(io->md, tio);
-	dec_pending(io, error);
+	dec_pending(tio->io, error);
+
+	/*
+	 * Store md for cleanup instead of tio which is about to get freed.
+	 */
+	bio->bi_private = md->bs;
+
 	bio_put(bio);
+	free_tio(md, tio);
 	return r;
 }
 
@@ -525,6 +562,7 @@ static void __map_bio(struct dm_target *ti, struct bio *clone,
 {
 	int r;
 	sector_t sector;
+	struct mapped_device *md;
 
 	/*
 	 * Sanity checks.
@@ -542,7 +580,7 @@ static void __map_bio(struct dm_target *ti, struct bio *clone,
 	atomic_inc(&tio->io->io_count);
 	sector = clone->bi_sector;
 	r = ti->type->map(ti, clone, &tio->info);
-	if (r > 0) {
+	if (r == DM_MAPIO_REMAPPED) {
 		/* the bio has been remapped so dispatch it */
 
 		blk_add_trace_remap(bdev_get_queue(clone->bi_bdev), clone,
@@ -550,14 +588,19 @@ static void __map_bio(struct dm_target *ti, struct bio *clone,
 				    clone->bi_sector);
 
 		generic_make_request(clone);
-	}
-
-	else if (r < 0) {
-		/* error the io and bail out */
-		struct dm_io *io = tio->io;
-		free_tio(tio->io->md, tio);
-		dec_pending(io, r);
+	} else if (r < 0 || r == DM_MAPIO_REQUEUE) {
+		/* error the io and bail out, or requeue it if needed */
+		md = tio->io->md;
+		dec_pending(tio->io, r);
+		/*
+		 * Store bio_set for cleanup.
+		 */
+		clone->bi_private = md->bs;
 		bio_put(clone);
+		free_tio(md, tio);
+	} else if (r) {
+		DMWARN("unimplemented target map return value: %d", r);
+		BUG();
 	}
 }
 
@@ -573,7 +616,9 @@ struct clone_info {
 
 static void dm_bio_destructor(struct bio *bio)
 {
-	bio_free(bio, dm_set);
+	struct bio_set *bs = bio->bi_private;
+
+	bio_free(bio, bs);
 }
 
 /*
@@ -581,12 +626,12 @@ static void dm_bio_destructor(struct bio *bio)
  */
 static struct bio *split_bvec(struct bio *bio, sector_t sector,
 			      unsigned short idx, unsigned int offset,
-			      unsigned int len)
+			      unsigned int len, struct bio_set *bs)
 {
 	struct bio *clone;
 	struct bio_vec *bv = bio->bi_io_vec + idx;
 
-	clone = bio_alloc_bioset(GFP_NOIO, 1, dm_set);
+	clone = bio_alloc_bioset(GFP_NOIO, 1, bs);
 	clone->bi_destructor = dm_bio_destructor;
 	*clone->bi_io_vec = *bv;
 
@@ -606,11 +651,13 @@ static struct bio *split_bvec(struct bio *bio, sector_t sector,
  */
 static struct bio *clone_bio(struct bio *bio, sector_t sector,
 			     unsigned short idx, unsigned short bv_count,
-			     unsigned int len)
+			     unsigned int len, struct bio_set *bs)
 {
 	struct bio *clone;
 
-	clone = bio_clone(bio, GFP_NOIO);
+	clone = bio_alloc_bioset(GFP_NOIO, bio->bi_max_vecs, bs);
+	__bio_clone(clone, bio);
+	clone->bi_destructor = dm_bio_destructor;
 	clone->bi_sector = sector;
 	clone->bi_idx = idx;
 	clone->bi_vcnt = idx + bv_count;
@@ -641,7 +688,8 @@ static void __clone_and_map(struct clone_info *ci)
 		 * the remaining io with a single clone.
 		 */
 		clone = clone_bio(bio, ci->sector, ci->idx,
-				  bio->bi_vcnt - ci->idx, ci->sector_count);
+				  bio->bi_vcnt - ci->idx, ci->sector_count,
+				  ci->md->bs);
 		__map_bio(ti, clone, tio);
 		ci->sector_count = 0;
 
@@ -664,7 +712,8 @@ static void __clone_and_map(struct clone_info *ci)
 			len += bv_len;
 		}
 
-		clone = clone_bio(bio, ci->sector, ci->idx, i - ci->idx, len);
+		clone = clone_bio(bio, ci->sector, ci->idx, i - ci->idx, len,
+				  ci->md->bs);
 		__map_bio(ti, clone, tio);
 
 		ci->sector += len;
@@ -693,7 +742,8 @@ static void __clone_and_map(struct clone_info *ci)
 			len = min(remaining, max);
 
 			clone = split_bvec(bio, ci->sector, ci->idx,
-					   bv->bv_offset + offset, len);
+					   bv->bv_offset + offset, len,
+					   ci->md->bs);
 
 			__map_bio(ti, clone, tio);
 
@@ -936,6 +986,7 @@ static struct mapped_device *alloc_dev(int minor)
 	memset(md, 0, sizeof(*md));
 	init_rwsem(&md->io_lock);
 	init_MUTEX(&md->suspend_lock);
+	spin_lock_init(&md->pushback_lock);
 	rwlock_init(&md->map_lock);
 	atomic_set(&md->holders, 1);
 	atomic_set(&md->open_count, 0);
@@ -954,13 +1005,17 @@ static struct mapped_device *alloc_dev(int minor)
 	md->queue->issue_flush_fn = dm_flush_all;
 
 	md->io_pool = mempool_create_slab_pool(MIN_IOS, _io_cache);
- 	if (!md->io_pool)
- 		goto bad2;
+	if (!md->io_pool)
+		goto bad2;
 
 	md->tio_pool = mempool_create_slab_pool(MIN_IOS, _tio_cache);
 	if (!md->tio_pool)
 		goto bad3;
 
+	md->bs = bioset_create(16, 16);
+	if (!md->bs)
+		goto bad_no_bioset;
+
 	md->disk = alloc_disk(1);
 	if (!md->disk)
 		goto bad4;
@@ -988,6 +1043,8 @@ static struct mapped_device *alloc_dev(int minor)
 	return md;
 
  bad4:
+	bioset_free(md->bs);
+ bad_no_bioset:
 	mempool_destroy(md->tio_pool);
  bad3:
 	mempool_destroy(md->io_pool);
@@ -1012,6 +1069,7 @@ static void free_dev(struct mapped_device *md)
 	}
 	mempool_destroy(md->tio_pool);
 	mempool_destroy(md->io_pool);
+	bioset_free(md->bs);
 	del_gendisk(md->disk);
 	free_minor(minor);
 
@@ -1058,7 +1116,8 @@ static int __bind(struct mapped_device *md, struct dm_table *t)
 	if (size != get_capacity(md->disk))
 		memset(&md->geometry, 0, sizeof(md->geometry));
 
-	__set_size(md, size);
+	if (md->suspended_bdev)
+		__set_size(md, size);
 	if (size == 0)
 		return 0;
 
@@ -1206,6 +1265,11 @@ int dm_swap_table(struct mapped_device *md, struct dm_table *table)
 	if (!dm_suspended(md))
 		goto out;
 
+	/* without bdev, the device size cannot be changed */
+	if (!md->suspended_bdev)
+		if (get_capacity(md->disk) != dm_table_get_size(table))
+			goto out;
+
 	__unbind(md);
 	r = __bind(md, table);
 
@@ -1256,32 +1320,48 @@ static void unlock_fs(struct mapped_device *md)
  * dm_bind_table, dm_suspend must be called to flush any in
  * flight bios and ensure that any further io gets deferred.
  */
-int dm_suspend(struct mapped_device *md, int do_lockfs)
+int dm_suspend(struct mapped_device *md, unsigned suspend_flags)
 {
 	struct dm_table *map = NULL;
+	unsigned long flags;
 	DECLARE_WAITQUEUE(wait, current);
 	struct bio *def;
 	int r = -EINVAL;
+	int do_lockfs = suspend_flags & DM_SUSPEND_LOCKFS_FLAG ? 1 : 0;
+	int noflush = suspend_flags & DM_SUSPEND_NOFLUSH_FLAG ? 1 : 0;
 
 	down(&md->suspend_lock);
 
 	if (dm_suspended(md))
-		goto out;
+		goto out_unlock;
 
 	map = dm_get_table(md);
 
+	/*
+	 * DMF_NOFLUSH_SUSPENDING must be set before presuspend.
+	 * This flag is cleared before dm_suspend returns.
+	 */
+	if (noflush)
+		set_bit(DMF_NOFLUSH_SUSPENDING, &md->flags);
+
 	/* This does not get reverted if there's an error later. */
 	dm_table_presuspend_targets(map);
 
-	md->suspended_bdev = bdget_disk(md->disk, 0);
-	if (!md->suspended_bdev) {
-		DMWARN("bdget failed in dm_suspend");
-		r = -ENOMEM;
-		goto out;
+	/* bdget() can stall if the pending I/Os are not flushed */
+	if (!noflush) {
+		md->suspended_bdev = bdget_disk(md->disk, 0);
+		if (!md->suspended_bdev) {
+			DMWARN("bdget failed in dm_suspend");
+			r = -ENOMEM;
+			goto flush_and_out;
+		}
 	}
 
-	/* Flush I/O to the device. */
-	if (do_lockfs) {
+	/*
+	 * Flush I/O to the device.
+	 * noflush supersedes do_lockfs, because lock_fs() needs to flush I/Os.
+	 */
+	if (do_lockfs && !noflush) {
 		r = lock_fs(md);
 		if (r)
 			goto out;
@@ -1317,6 +1397,14 @@ int dm_suspend(struct mapped_device *md, int do_lockfs)
 	down_write(&md->io_lock);
 	remove_wait_queue(&md->wait, &wait);
 
+	if (noflush) {
+		spin_lock_irqsave(&md->pushback_lock, flags);
+		clear_bit(DMF_NOFLUSH_SUSPENDING, &md->flags);
+		bio_list_merge_head(&md->deferred, &md->pushback);
+		bio_list_init(&md->pushback);
+		spin_unlock_irqrestore(&md->pushback_lock, flags);
+	}
+
 	/* were we interrupted ? */
 	r = -EINTR;
 	if (atomic_read(&md->pending)) {
@@ -1325,7 +1413,7 @@ int dm_suspend(struct mapped_device *md, int do_lockfs)
 		__flush_deferred_io(md, def);
 		up_write(&md->io_lock);
 		unlock_fs(md);
-		goto out;
+		goto out; /* pushback list is already flushed, so skip flush */
 	}
 	up_write(&md->io_lock);
 
@@ -1335,6 +1423,25 @@ int dm_suspend(struct mapped_device *md, int do_lockfs)
 
 	r = 0;
 
+flush_and_out:
+	if (r && noflush) {
+		/*
+		 * Because there may be already I/Os in the pushback list,
+		 * flush them before return.
+		 */
+		down_write(&md->io_lock);
+
+		spin_lock_irqsave(&md->pushback_lock, flags);
+		clear_bit(DMF_NOFLUSH_SUSPENDING, &md->flags);
+		bio_list_merge_head(&md->deferred, &md->pushback);
+		bio_list_init(&md->pushback);
+		spin_unlock_irqrestore(&md->pushback_lock, flags);
+
+		def = bio_list_get(&md->deferred);
+		__flush_deferred_io(md, def);
+		up_write(&md->io_lock);
+	}
+
 out:
 	if (r && md->suspended_bdev) {
 		bdput(md->suspended_bdev);
@@ -1342,6 +1449,8 @@ out:
 	}
 
 	dm_table_put(map);
+
+out_unlock:
 	up(&md->suspend_lock);
 	return r;
 }
@@ -1373,8 +1482,10 @@ int dm_resume(struct mapped_device *md)
 
 	unlock_fs(md);
 
-	bdput(md->suspended_bdev);
-	md->suspended_bdev = NULL;
+	if (md->suspended_bdev) {
+		bdput(md->suspended_bdev);
+		md->suspended_bdev = NULL;
+	}
 
 	clear_bit(DMF_SUSPENDED, &md->flags);
 
@@ -1419,6 +1530,17 @@ int dm_suspended(struct mapped_device *md)
 	return test_bit(DMF_SUSPENDED, &md->flags);
 }
 
+int dm_noflush_suspending(struct dm_target *ti)
+{
+	struct mapped_device *md = dm_table_get_md(ti->table);
+	int r = __noflush_suspending(md);
+
+	dm_put(md);
+
+	return r;
+}
+EXPORT_SYMBOL_GPL(dm_noflush_suspending);
+
 static struct block_device_operations dm_blk_dops = {
 	.open = dm_blk_open,
 	.release = dm_blk_close,