block: convert blkdev_issue_flush() to use empty barriers

Then we can get rid of ->issue_flush_fn() and all the driver private
implementations of that.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

block/ll_rw_blk.c

@@ -304,23 +304,6 @@ int blk_queue_ordered(struct request_queue *q, unsigned ordered,
 
 EXPORT_SYMBOL(blk_queue_ordered);
 
-/**
- * blk_queue_issue_flush_fn - set function for issuing a flush
- * @q:     the request queue
- * @iff:   the function to be called issuing the flush
- *
- * Description:
- *   If a driver supports issuing a flush command, the support is notified
- *   to the block layer by defining it through this call.
- *
- **/
-void blk_queue_issue_flush_fn(struct request_queue *q, issue_flush_fn *iff)
-{
-	q->issue_flush_fn = iff;
-}
-
-EXPORT_SYMBOL(blk_queue_issue_flush_fn);
-
 /*
  * Cache flushing for ordered writes handling
  */
@@ -2666,6 +2649,14 @@ int blk_execute_rq(struct request_queue *q, struct gendisk *bd_disk,
 
 EXPORT_SYMBOL(blk_execute_rq);
 
+static void bio_end_empty_barrier(struct bio *bio, int err)
+{
+	if (err)
+		clear_bit(BIO_UPTODATE, &bio->bi_flags);
+
+	complete(bio->bi_private);
+}
+
 /**
  * blkdev_issue_flush - queue a flush
  * @bdev:	blockdev to issue flush for
@@ -2678,7 +2669,10 @@ EXPORT_SYMBOL(blk_execute_rq);
  */
 int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector)
 {
+	DECLARE_COMPLETION_ONSTACK(wait);
 	struct request_queue *q;
+	struct bio *bio;
+	int ret;
 
 	if (bdev->bd_disk == NULL)
 		return -ENXIO;
@@ -2686,10 +2680,32 @@ int blkdev_issue_flush(struct block_device *bdev, sector_t *error_sector)
 	q = bdev_get_queue(bdev);
 	if (!q)
 		return -ENXIO;
-	if (!q->issue_flush_fn)
-		return -EOPNOTSUPP;
 
-	return q->issue_flush_fn(q, bdev->bd_disk, error_sector);
+	bio = bio_alloc(GFP_KERNEL, 0);
+	if (!bio)
+		return -ENOMEM;
+
+	bio->bi_end_io = bio_end_empty_barrier;
+	bio->bi_private = &wait;
+	bio->bi_bdev = bdev;
+	submit_bio(1 << BIO_RW_BARRIER, bio);
+
+	wait_for_completion(&wait);
+
+	/*
+	 * The driver must store the error location in ->bi_sector, if
+	 * it supports it. For non-stacked drivers, this should be copied
+	 * from rq->sector.
+	 */
+	if (error_sector)
+		*error_sector = bio->bi_sector;
+
+	ret = 0;
+	if (!bio_flagged(bio, BIO_UPTODATE))
+		ret = -EIO;
+
+	bio_put(bio);
+	return ret;
 }
 
 EXPORT_SYMBOL(blkdev_issue_flush);

drivers/block/ps3disk.c

@@ -414,26 +414,6 @@ static void ps3disk_prepare_flush(struct request_queue *q, struct request *req)
 	req->cmd_type = REQ_TYPE_FLUSH;
 }
 
-static int ps3disk_issue_flush(struct request_queue *q, struct gendisk *gendisk,
-			       sector_t *sector)
-{
-	struct ps3_storage_device *dev = q->queuedata;
-	struct request *req;
-	int res;
-
-	dev_dbg(&dev->sbd.core, "%s:%u\n", __func__, __LINE__);
-
-	req = blk_get_request(q, WRITE, __GFP_WAIT);
-	ps3disk_prepare_flush(q, req);
-	res = blk_execute_rq(q, gendisk, req, 0);
-	if (res)
-		dev_err(&dev->sbd.core, "%s:%u: flush request failed %d\n",
-			__func__, __LINE__, res);
-	blk_put_request(req);
-	return res;
-}
-
-
 static unsigned long ps3disk_mask;
 
 static DEFINE_MUTEX(ps3disk_mask_mutex);
@@ -506,7 +486,6 @@ static int __devinit ps3disk_probe(struct ps3_system_bus_device *_dev)
 	blk_queue_dma_alignment(queue, dev->blk_size-1);
 	blk_queue_hardsect_size(queue, dev->blk_size);
 
-	blk_queue_issue_flush_fn(queue, ps3disk_issue_flush);
 	blk_queue_ordered(queue, QUEUE_ORDERED_DRAIN_FLUSH,
 			  ps3disk_prepare_flush);
 

drivers/ide/ide-disk.c

@@ -716,32 +716,6 @@ static void idedisk_prepare_flush(struct request_queue *q, struct request *rq)
 	rq->buffer = rq->cmd;
 }
 
-static int idedisk_issue_flush(struct request_queue *q, struct gendisk *disk,
-			       sector_t *error_sector)
-{
-	ide_drive_t *drive = q->queuedata;
-	struct request *rq;
-	int ret;
-
-	if (!drive->wcache)
-		return 0;
-
-	rq = blk_get_request(q, WRITE, __GFP_WAIT);
-
-	idedisk_prepare_flush(q, rq);
-
-	ret = blk_execute_rq(q, disk, rq, 0);
-
-	/*
-	 * if we failed and caller wants error offset, get it
-	 */
-	if (ret && error_sector)
-		*error_sector = ide_get_error_location(drive, rq->cmd);
-
-	blk_put_request(rq);
-	return ret;
-}
-
 /*
  * This is tightly woven into the driver->do_special can not touch.
  * DON'T do it again until a total personality rewrite is committed.
@@ -781,7 +755,6 @@ static void update_ordered(ide_drive_t *drive)
 	struct hd_driveid *id = drive->id;
 	unsigned ordered = QUEUE_ORDERED_NONE;
 	prepare_flush_fn *prep_fn = NULL;
-	issue_flush_fn *issue_fn = NULL;
 
 	if (drive->wcache) {
 		unsigned long long capacity;
@@ -805,13 +778,11 @@ static void update_ordered(ide_drive_t *drive)
 		if (barrier) {
 			ordered = QUEUE_ORDERED_DRAIN_FLUSH;
 			prep_fn = idedisk_prepare_flush;
-			issue_fn = idedisk_issue_flush;
 		}
 	} else
 		ordered = QUEUE_ORDERED_DRAIN;
 
 	blk_queue_ordered(drive->queue, ordered, prep_fn);
-	blk_queue_issue_flush_fn(drive->queue, issue_fn);
 }
 
 static int write_cache(ide_drive_t *drive, int arg)

drivers/md/dm-table.c

@@ -999,33 +999,6 @@ void dm_table_unplug_all(struct dm_table *t)
 	}
 }
 
-int dm_table_flush_all(struct dm_table *t)
-{
-	struct list_head *d, *devices = dm_table_get_devices(t);
-	int ret = 0;
-	unsigned i;
-
-	for (i = 0; i < t->num_targets; i++)
-		if (t->targets[i].type->flush)
-			t->targets[i].type->flush(&t->targets[i]);
-
-	for (d = devices->next; d != devices; d = d->next) {
-		struct dm_dev *dd = list_entry(d, struct dm_dev, list);
-		struct request_queue *q = bdev_get_queue(dd->bdev);
-		int err;
-
-		if (!q->issue_flush_fn)
-			err = -EOPNOTSUPP;
-		else
-			err = q->issue_flush_fn(q, dd->bdev->bd_disk, NULL);
-
-		if (!ret)
-			ret = err;
-	}
-
-	return ret;
-}
-
 struct mapped_device *dm_table_get_md(struct dm_table *t)
 {
 	dm_get(t->md);
@@ -1043,4 +1016,3 @@ EXPORT_SYMBOL(dm_table_get_md);
 EXPORT_SYMBOL(dm_table_put);
 EXPORT_SYMBOL(dm_table_get);
 EXPORT_SYMBOL(dm_table_unplug_all);
-EXPORT_SYMBOL(dm_table_flush_all);

drivers/md/dm.c

@@ -840,21 +840,6 @@ static int dm_request(struct request_queue *q, struct bio *bio)
 	return 0;
 }
 
-static int dm_flush_all(struct request_queue *q, struct gendisk *disk,
-			sector_t *error_sector)
-{
-	struct mapped_device *md = q->queuedata;
-	struct dm_table *map = dm_get_table(md);
-	int ret = -ENXIO;
-
-	if (map) {
-		ret = dm_table_flush_all(map);
-		dm_table_put(map);
-	}
-
-	return ret;
-}
-
 static void dm_unplug_all(struct request_queue *q)
 {
 	struct mapped_device *md = q->queuedata;
@@ -1003,7 +988,6 @@ static struct mapped_device *alloc_dev(int minor)
 	blk_queue_make_request(md->queue, dm_request);
 	blk_queue_bounce_limit(md->queue, BLK_BOUNCE_ANY);
 	md->queue->unplug_fn = dm_unplug_all;
-	md->queue->issue_flush_fn = dm_flush_all;
 
 	md->io_pool = mempool_create_slab_pool(MIN_IOS, _io_cache);
 	if (!md->io_pool)

drivers/md/dm.h

@@ -111,7 +111,6 @@ void dm_table_postsuspend_targets(struct dm_table *t);
 int dm_table_resume_targets(struct dm_table *t);
 int dm_table_any_congested(struct dm_table *t, int bdi_bits);
 void dm_table_unplug_all(struct dm_table *t);
-int dm_table_flush_all(struct dm_table *t);
 
 /*-----------------------------------------------------------------
  * A registry of target types.

drivers/md/linear.c

@@ -92,25 +92,6 @@ static void linear_unplug(struct request_queue *q)
 	}
 }
 
-static int linear_issue_flush(struct request_queue *q, struct gendisk *disk,
-			      sector_t *error_sector)
-{
-	mddev_t *mddev = q->queuedata;
-	linear_conf_t *conf = mddev_to_conf(mddev);
-	int i, ret = 0;
-
-	for (i=0; i < mddev->raid_disks && ret == 0; i++) {
-		struct block_device *bdev = conf->disks[i].rdev->bdev;
-		struct request_queue *r_queue = bdev_get_queue(bdev);
-
-		if (!r_queue->issue_flush_fn)
-			ret = -EOPNOTSUPP;
-		else
-			ret = r_queue->issue_flush_fn(r_queue, bdev->bd_disk, error_sector);
-	}
-	return ret;
-}
-
 static int linear_congested(void *data, int bits)
 {
 	mddev_t *mddev = data;
@@ -279,7 +260,6 @@ static int linear_run (mddev_t *mddev)
 
 	blk_queue_merge_bvec(mddev->queue, linear_mergeable_bvec);
 	mddev->queue->unplug_fn = linear_unplug;
-	mddev->queue->issue_flush_fn = linear_issue_flush;
 	mddev->queue->backing_dev_info.congested_fn = linear_congested;
 	mddev->queue->backing_dev_info.congested_data = mddev;
 	return 0;

drivers/md/md.c

@@ -3463,7 +3463,6 @@ static int do_md_stop(mddev_t * mddev, int mode)
 			mddev->pers->stop(mddev);
 			mddev->queue->merge_bvec_fn = NULL;
 			mddev->queue->unplug_fn = NULL;
-			mddev->queue->issue_flush_fn = NULL;
 			mddev->queue->backing_dev_info.congested_fn = NULL;
 			if (mddev->pers->sync_request)
 				sysfs_remove_group(&mddev->kobj, &md_redundancy_group);

drivers/md/multipath.c

@@ -194,35 +194,6 @@ static void multipath_status (struct seq_file *seq, mddev_t *mddev)
 	seq_printf (seq, "]");
 }
 
-static int multipath_issue_flush(struct request_queue *q, struct gendisk *disk,
-				 sector_t *error_sector)
-{
-	mddev_t *mddev = q->queuedata;
-	multipath_conf_t *conf = mddev_to_conf(mddev);
-	int i, ret = 0;
-
-	rcu_read_lock();
-	for (i=0; i<mddev->raid_disks && ret == 0; i++) {
-		mdk_rdev_t *rdev = rcu_dereference(conf->multipaths[i].rdev);
-		if (rdev && !test_bit(Faulty, &rdev->flags)) {
-			struct block_device *bdev = rdev->bdev;
-			struct request_queue *r_queue = bdev_get_queue(bdev);
-
-			if (!r_queue->issue_flush_fn)
-				ret = -EOPNOTSUPP;
-			else {
-				atomic_inc(&rdev->nr_pending);
-				rcu_read_unlock();
-				ret = r_queue->issue_flush_fn(r_queue, bdev->bd_disk,
-							      error_sector);
-				rdev_dec_pending(rdev, mddev);
-				rcu_read_lock();
-			}
-		}
-	}
-	rcu_read_unlock();
-	return ret;
-}
 static int multipath_congested(void *data, int bits)
 {
 	mddev_t *mddev = data;
@@ -527,7 +498,6 @@ static int multipath_run (mddev_t *mddev)
 	mddev->array_size = mddev->size;
 
 	mddev->queue->unplug_fn = multipath_unplug;
-	mddev->queue->issue_flush_fn = multipath_issue_flush;
 	mddev->queue->backing_dev_info.congested_fn = multipath_congested;
 	mddev->queue->backing_dev_info.congested_data = mddev;
 

drivers/md/raid0.c

@@ -40,26 +40,6 @@ static void raid0_unplug(struct request_queue *q)
 	}
 }
 
-static int raid0_issue_flush(struct request_queue *q, struct gendisk *disk,
-			     sector_t *error_sector)
-{
-	mddev_t *mddev = q->queuedata;
-	raid0_conf_t *conf = mddev_to_conf(mddev);
-	mdk_rdev_t **devlist = conf->strip_zone[0].dev;
-	int i, ret = 0;
-
-	for (i=0; i<mddev->raid_disks && ret == 0; i++) {
-		struct block_device *bdev = devlist[i]->bdev;
-		struct request_queue *r_queue = bdev_get_queue(bdev);
-
-		if (!r_queue->issue_flush_fn)
-			ret = -EOPNOTSUPP;
-		else
-			ret = r_queue->issue_flush_fn(r_queue, bdev->bd_disk, error_sector);
-	}
-	return ret;
-}
-
 static int raid0_congested(void *data, int bits)
 {
 	mddev_t *mddev = data;
@@ -250,7 +230,6 @@ static int create_strip_zones (mddev_t *mddev)
 
 	mddev->queue->unplug_fn = raid0_unplug;
 
-	mddev->queue->issue_flush_fn = raid0_issue_flush;
 	mddev->queue->backing_dev_info.congested_fn = raid0_congested;
 	mddev->queue->backing_dev_info.congested_data = mddev;
 

drivers/md/raid1.c

@@ -567,36 +567,6 @@ static void raid1_unplug(struct request_queue *q)
 	md_wakeup_thread(mddev->thread);
 }
 
-static int raid1_issue_flush(struct request_queue *q, struct gendisk *disk,
-			     sector_t *error_sector)
-{
-	mddev_t *mddev = q->queuedata;
-	conf_t *conf = mddev_to_conf(mddev);
-	int i, ret = 0;
-
-	rcu_read_lock();
-	for (i=0; i<mddev->raid_disks && ret == 0; i++) {
-		mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev);
-		if (rdev && !test_bit(Faulty, &rdev->flags)) {
-			struct block_device *bdev = rdev->bdev;
-			struct request_queue *r_queue = bdev_get_queue(bdev);
-
-			if (!r_queue->issue_flush_fn)
-				ret = -EOPNOTSUPP;
-			else {
-				atomic_inc(&rdev->nr_pending);
-				rcu_read_unlock();
-				ret = r_queue->issue_flush_fn(r_queue, bdev->bd_disk,
-							      error_sector);
-				rdev_dec_pending(rdev, mddev);
-				rcu_read_lock();
-			}
-		}
-	}
-	rcu_read_unlock();
-	return ret;
-}
-
 static int raid1_congested(void *data, int bits)
 {
 	mddev_t *mddev = data;
@@ -1997,7 +1967,6 @@ static int run(mddev_t *mddev)
 	mddev->array_size = mddev->size;
 
 	mddev->queue->unplug_fn = raid1_unplug;
-	mddev->queue->issue_flush_fn = raid1_issue_flush;
 	mddev->queue->backing_dev_info.congested_fn = raid1_congested;
 	mddev->queue->backing_dev_info.congested_data = mddev;
 

drivers/md/raid10.c

@@ -611,36 +611,6 @@ static void raid10_unplug(struct request_queue *q)
 	md_wakeup_thread(mddev->thread);
 }
 
-static int raid10_issue_flush(struct request_queue *q, struct gendisk *disk,
-			     sector_t *error_sector)
-{
-	mddev_t *mddev = q->queuedata;
-	conf_t *conf = mddev_to_conf(mddev);
-	int i, ret = 0;
-
-	rcu_read_lock();
-	for (i=0; i<mddev->raid_disks && ret == 0; i++) {
-		mdk_rdev_t *rdev = rcu_dereference(conf->mirrors[i].rdev);
-		if (rdev && !test_bit(Faulty, &rdev->flags)) {
-			struct block_device *bdev = rdev->bdev;
-			struct request_queue *r_queue = bdev_get_queue(bdev);
-
-			if (!r_queue->issue_flush_fn)
-				ret = -EOPNOTSUPP;
-			else {
-				atomic_inc(&rdev->nr_pending);
-				rcu_read_unlock();
-				ret = r_queue->issue_flush_fn(r_queue, bdev->bd_disk,
-							      error_sector);
-				rdev_dec_pending(rdev, mddev);
-				rcu_read_lock();
-			}
-		}
-	}
-	rcu_read_unlock();
-	return ret;
-}
-
 static int raid10_congested(void *data, int bits)
 {
 	mddev_t *mddev = data;
@@ -2118,7 +2088,6 @@ static int run(mddev_t *mddev)
 	mddev->resync_max_sectors = size << conf->chunk_shift;
 
 	mddev->queue->unplug_fn = raid10_unplug;
-	mddev->queue->issue_flush_fn = raid10_issue_flush;
 	mddev->queue->backing_dev_info.congested_fn = raid10_congested;
 	mddev->queue->backing_dev_info.congested_data = mddev;
 

drivers/md/raid5.c

@@ -3204,36 +3204,6 @@ static void raid5_unplug_device(struct request_queue *q)
 	unplug_slaves(mddev);
 }
 
-static int raid5_issue_flush(struct request_queue *q, struct gendisk *disk,
-			     sector_t *error_sector)
-{
-	mddev_t *mddev = q->queuedata;
-	raid5_conf_t *conf = mddev_to_conf(mddev);
-	int i, ret = 0;
-
-	rcu_read_lock();
-	for (i=0; i<mddev->raid_disks && ret == 0; i++) {
-		mdk_rdev_t *rdev = rcu_dereference(conf->disks[i].rdev);
-		if (rdev && !test_bit(Faulty, &rdev->flags)) {
-			struct block_device *bdev = rdev->bdev;
-			struct request_queue *r_queue = bdev_get_queue(bdev);
-
-			if (!r_queue->issue_flush_fn)
-				ret = -EOPNOTSUPP;
-			else {
-				atomic_inc(&rdev->nr_pending);
-				rcu_read_unlock();
-				ret = r_queue->issue_flush_fn(r_queue, bdev->bd_disk,
-							      error_sector);
-				rdev_dec_pending(rdev, mddev);
-				rcu_read_lock();
-			}
-		}
-	}
-	rcu_read_unlock();
-	return ret;
-}
-
 static int raid5_congested(void *data, int bits)
 {
 	mddev_t *mddev = data;
@@ -4263,7 +4233,6 @@ static int run(mddev_t *mddev)
 		       mdname(mddev));
 
 	mddev->queue->unplug_fn = raid5_unplug_device;
-	mddev->queue->issue_flush_fn = raid5_issue_flush;
 	mddev->queue->backing_dev_info.congested_data = mddev;
 	mddev->queue->backing_dev_info.congested_fn = raid5_congested;
 

drivers/message/i2o/i2o_block.c

@@ -148,29 +148,6 @@ static int i2o_block_device_flush(struct i2o_device *dev)
 	return i2o_msg_post_wait(dev->iop, msg, 60);
 };
 
-/**
- *	i2o_block_issue_flush - device-flush interface for block-layer
- *	@queue: the request queue of the device which should be flushed
- *	@disk: gendisk
- *	@error_sector: error offset
- *
- *	Helper function to provide flush functionality to block-layer.
- *
- *	Returns 0 on success or negative error code on failure.
- */
-
-static int i2o_block_issue_flush(struct request_queue * queue, struct gendisk *disk,
-				 sector_t * error_sector)
-{
-	struct i2o_block_device *i2o_blk_dev = queue->queuedata;
-	int rc = -ENODEV;
-
-	if (likely(i2o_blk_dev))
-		rc = i2o_block_device_flush(i2o_blk_dev->i2o_dev);
-
-	return rc;
-}
-
 /**
  *	i2o_block_device_mount - Mount (load) the media of device dev
  *	@dev: I2O device which should receive the mount request
@@ -1009,7 +986,6 @@ static struct i2o_block_device *i2o_block_device_alloc(void)
 	}
 
 	blk_queue_prep_rq(queue, i2o_block_prep_req_fn);
-	blk_queue_issue_flush_fn(queue, i2o_block_issue_flush);
 
 	gd->major = I2O_MAJOR;
 	gd->queue = queue;

drivers/scsi/sd.c

@@ -826,27 +826,6 @@ static int sd_sync_cache(struct scsi_disk *sdkp)
 	return 0;
 }
 
-static int sd_issue_flush(struct request_queue *q, struct gendisk *disk,
-			  sector_t *error_sector)
-{
-	int ret = 0;
-	struct scsi_device *sdp = q->queuedata;
-	struct scsi_disk *sdkp;
-
-	if (sdp->sdev_state != SDEV_RUNNING)
-		return -ENXIO;
-
-	sdkp = scsi_disk_get_from_dev(&sdp->sdev_gendev);
-
-	if (!sdkp)
-               return -ENODEV;
-
-	if (sdkp->WCE)
-		ret = sd_sync_cache(sdkp);
-	scsi_disk_put(sdkp);
-	return ret;
-}
-
 static void sd_prepare_flush(struct request_queue *q, struct request *rq)
 {
 	memset(rq->cmd, 0, sizeof(rq->cmd));
@@ -1697,7 +1676,6 @@ static int sd_probe(struct device *dev)
 	sd_revalidate_disk(gd);
 
 	blk_queue_prep_rq(sdp->request_queue, sd_prep_fn);
-	blk_queue_issue_flush_fn(sdp->request_queue, sd_issue_flush);
 
 	gd->driverfs_dev = &sdp->sdev_gendev;
 	gd->flags = GENHD_FL_DRIVERFS;

include/linux/blkdev.h

@@ -330,7 +330,6 @@ typedef void (unplug_fn) (struct request_queue *);
 
 struct bio_vec;
 typedef int (merge_bvec_fn) (struct request_queue *, struct bio *, struct bio_vec *);
-typedef int (issue_flush_fn) (struct request_queue *, struct gendisk *, sector_t *);
 typedef void (prepare_flush_fn) (struct request_queue *, struct request *);
 typedef void (softirq_done_fn)(struct request *);
 
@@ -368,7 +367,6 @@ struct request_queue
 	prep_rq_fn		*prep_rq_fn;
 	unplug_fn		*unplug_fn;
 	merge_bvec_fn		*merge_bvec_fn;
-	issue_flush_fn		*issue_flush_fn;
 	prepare_flush_fn	*prepare_flush_fn;
 	softirq_done_fn		*softirq_done_fn;
 
@@ -770,7 +768,6 @@ extern void blk_queue_dma_alignment(struct request_queue *, int);
 extern void blk_queue_softirq_done(struct request_queue *, softirq_done_fn *);
 extern struct backing_dev_info *blk_get_backing_dev_info(struct block_device *bdev);
 extern int blk_queue_ordered(struct request_queue *, unsigned, prepare_flush_fn *);
-extern void blk_queue_issue_flush_fn(struct request_queue *, issue_flush_fn *);
 extern int blk_do_ordered(struct request_queue *, struct request **);
 extern unsigned blk_ordered_cur_seq(struct request_queue *);
 extern unsigned blk_ordered_req_seq(struct request *);