block: move the bdi from the request_queue to the gendisk

The backing device information only makes sense for file system I/O,
and thus belongs into the gendisk and not the lower level request_queue
structure.  Move it there.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Link: https://lore.kernel.org/r/20210809141744.1203023-5-hch@lst.de
Signed-off-by: Jens Axboe <axboe@kernel.dk>

block/bfq-iosched.c

@@ -5266,7 +5266,7 @@ bfq_set_next_ioprio_data(struct bfq_queue *bfqq, struct bfq_io_cq *bic)
 	switch (ioprio_class) {
 	default:
 		pr_err("bdi %s: bfq: bad prio class %d\n",
-				bdi_dev_name(bfqq->bfqd->queue->backing_dev_info),
+			bdi_dev_name(queue_to_disk(bfqq->bfqd->queue)->bdi),
 			ioprio_class);
 		fallthrough;
 	case IOPRIO_CLASS_NONE:

block/blk-cgroup.c

@@ -489,10 +489,9 @@ static int blkcg_reset_stats(struct cgroup_subsys_state *css,
 
 const char *blkg_dev_name(struct blkcg_gq *blkg)
 {
-	/* some drivers (floppy) instantiate a queue w/o disk registered */
-	if (blkg->q->backing_dev_info->dev)
-		return bdi_dev_name(blkg->q->backing_dev_info);
+	if (!queue_has_disk(blkg->q) || !queue_to_disk(blkg->q)->bdi->dev)
 		return NULL;
+	return bdi_dev_name(queue_to_disk(blkg->q)->bdi);
 }
 
 /**

block/blk-core.c

@@ -14,7 +14,6 @@
  */
 #include <linux/kernel.h>
 #include <linux/module.h>
-#include <linux/backing-dev.h>
 #include <linux/bio.h>
 #include <linux/blkdev.h>
 #include <linux/blk-mq.h>
@@ -531,13 +530,9 @@ struct request_queue *blk_alloc_queue(int node_id)
 	if (ret)
 		goto fail_id;
 
-	q->backing_dev_info = bdi_alloc(node_id);
-	if (!q->backing_dev_info)
-		goto fail_split;
-
 	q->stats = blk_alloc_queue_stats();
 	if (!q->stats)
-		goto fail_stats;
+		goto fail_split;
 
 	q->node = node_id;
 
@@ -567,7 +562,7 @@ struct request_queue *blk_alloc_queue(int node_id)
 	if (percpu_ref_init(&q->q_usage_counter,
 				blk_queue_usage_counter_release,
 				PERCPU_REF_INIT_ATOMIC, GFP_KERNEL))
-		goto fail_bdi;
+		goto fail_stats;
 
 	if (blkcg_init_queue(q))
 		goto fail_ref;
@@ -580,10 +575,8 @@ struct request_queue *blk_alloc_queue(int node_id)
 
 fail_ref:
 	percpu_ref_exit(&q->q_usage_counter);
-fail_bdi:
-	blk_free_queue_stats(q->stats);
 fail_stats:
-	bdi_put(q->backing_dev_info);
+	blk_free_queue_stats(q->stats);
 fail_split:
 	bioset_exit(&q->bio_split);
 fail_id:

block/blk-mq.c

@@ -525,7 +525,7 @@ void blk_mq_free_request(struct request *rq)
 		__blk_mq_dec_active_requests(hctx);
 
 	if (unlikely(laptop_mode && !blk_rq_is_passthrough(rq)))
-		laptop_io_completion(q->backing_dev_info);
+		laptop_io_completion(queue_to_disk(q)->bdi);
 
 	rq_qos_done(q, rq);
 

block/blk-settings.c

@@ -8,6 +8,7 @@
 #include <linux/bio.h>
 #include <linux/blkdev.h>
 #include <linux/pagemap.h>
+#include <linux/backing-dev-defs.h>
 #include <linux/gcd.h>
 #include <linux/lcm.h>
 #include <linux/jiffies.h>
@@ -140,7 +141,9 @@ void blk_queue_max_hw_sectors(struct request_queue *q, unsigned int max_hw_secto
 				 limits->logical_block_size >> SECTOR_SHIFT);
 	limits->max_sectors = max_sectors;
 
-	q->backing_dev_info->io_pages = max_sectors >> (PAGE_SHIFT - 9);
+	if (!queue_has_disk(q))
+		return;
+	queue_to_disk(q)->bdi->io_pages = max_sectors >> (PAGE_SHIFT - 9);
 }
 EXPORT_SYMBOL(blk_queue_max_hw_sectors);
 
@@ -388,10 +391,9 @@ void disk_update_readahead(struct gendisk *disk)
 	 * For read-ahead of large files to be effective, we need to read ahead
 	 * at least twice the optimal I/O size.
 	 */
-	q->backing_dev_info->ra_pages =
+	disk->bdi->ra_pages =
 		max(queue_io_opt(q) * 2 / PAGE_SIZE, VM_READAHEAD_PAGES);
-	q->backing_dev_info->io_pages =
-		queue_max_sectors(q) >> (PAGE_SHIFT - 9);
+	disk->bdi->io_pages = queue_max_sectors(q) >> (PAGE_SHIFT - 9);
 }
 EXPORT_SYMBOL_GPL(disk_update_readahead);
 
@@ -473,7 +475,9 @@ EXPORT_SYMBOL(blk_limits_io_opt);
 void blk_queue_io_opt(struct request_queue *q, unsigned int opt)
 {
 	blk_limits_io_opt(&q->limits, opt);
-	q->backing_dev_info->ra_pages =
+	if (!queue_has_disk(q))
+		return;
+	queue_to_disk(q)->bdi->ra_pages =
 		max(queue_io_opt(q) * 2 / PAGE_SIZE, VM_READAHEAD_PAGES);
 }
 EXPORT_SYMBOL(blk_queue_io_opt);

block/blk-sysfs.c

@@ -88,9 +88,11 @@ queue_requests_store(struct request_queue *q, const char *page, size_t count)
 
 static ssize_t queue_ra_show(struct request_queue *q, char *page)
 {
-	unsigned long ra_kb = q->backing_dev_info->ra_pages <<
-					(PAGE_SHIFT - 10);
+	unsigned long ra_kb;
 
+	if (!queue_has_disk(q))
+		return -EINVAL;
+	ra_kb = queue_to_disk(q)->bdi->ra_pages << (PAGE_SHIFT - 10);
 	return queue_var_show(ra_kb, page);
 }
 
@@ -98,13 +100,14 @@ static ssize_t
 queue_ra_store(struct request_queue *q, const char *page, size_t count)
 {
 	unsigned long ra_kb;
-	ssize_t ret = queue_var_store(&ra_kb, page, count);
+	ssize_t ret;
 
+	if (!queue_has_disk(q))
+		return -EINVAL;
+	ret = queue_var_store(&ra_kb, page, count);
 	if (ret < 0)
 		return ret;
-
-	q->backing_dev_info->ra_pages = ra_kb >> (PAGE_SHIFT - 10);
-
+	queue_to_disk(q)->bdi->ra_pages = ra_kb >> (PAGE_SHIFT - 10);
 	return ret;
 }
 
@@ -251,7 +254,9 @@ queue_max_sectors_store(struct request_queue *q, const char *page, size_t count)
 
 	spin_lock_irq(&q->queue_lock);
 	q->limits.max_sectors = max_sectors_kb << 1;
-	q->backing_dev_info->io_pages = max_sectors_kb >> (PAGE_SHIFT - 10);
+	if (queue_has_disk(q))
+		queue_to_disk(q)->bdi->io_pages =
+			max_sectors_kb >> (PAGE_SHIFT - 10);
 	spin_unlock_irq(&q->queue_lock);
 
 	return ret;
@@ -766,13 +771,6 @@ static void blk_exit_queue(struct request_queue *q)
 	 * e.g. blkcg_print_blkgs() to crash.
 	 */
 	blkcg_exit_queue(q);
-
-	/*
-	 * Since the cgroup code may dereference the @q->backing_dev_info
-	 * pointer, only decrease its reference count after having removed the
-	 * association with the block cgroup controller.
-	 */
-	bdi_put(q->backing_dev_info);
 }
 
 /**

block/blk-wbt.c

@@ -97,7 +97,7 @@ static void wb_timestamp(struct rq_wb *rwb, unsigned long *var)
  */
 static bool wb_recent_wait(struct rq_wb *rwb)
 {
-	struct bdi_writeback *wb = &rwb->rqos.q->backing_dev_info->wb;
+	struct bdi_writeback *wb = &queue_to_disk(rwb->rqos.q)->bdi->wb;
 
 	return time_before(jiffies, wb->dirty_sleep + HZ);
 }
@@ -234,7 +234,7 @@ enum {
 
 static int latency_exceeded(struct rq_wb *rwb, struct blk_rq_stat *stat)
 {
-	struct backing_dev_info *bdi = rwb->rqos.q->backing_dev_info;
+	struct backing_dev_info *bdi = queue_to_disk(rwb->rqos.q)->bdi;
 	struct rq_depth *rqd = &rwb->rq_depth;
 	u64 thislat;
 
@@ -287,7 +287,7 @@ static int latency_exceeded(struct rq_wb *rwb, struct blk_rq_stat *stat)
 
 static void rwb_trace_step(struct rq_wb *rwb, const char *msg)
 {
-	struct backing_dev_info *bdi = rwb->rqos.q->backing_dev_info;
+	struct backing_dev_info *bdi = queue_to_disk(rwb->rqos.q)->bdi;
 	struct rq_depth *rqd = &rwb->rq_depth;
 
 	trace_wbt_step(bdi, msg, rqd->scale_step, rwb->cur_win_nsec,
@@ -359,8 +359,8 @@ static void wb_timer_fn(struct blk_stat_callback *cb)
 
 	status = latency_exceeded(rwb, cb->stat);
 
-	trace_wbt_timer(rwb->rqos.q->backing_dev_info, status, rqd->scale_step,
-			inflight);
+	trace_wbt_timer(queue_to_disk(rwb->rqos.q)->bdi, status,
+			rqd->scale_step, inflight);
 
 	/*
 	 * If we exceeded the latency target, step down. If we did not,

block/genhd.c

@@ -466,9 +466,8 @@ static void register_disk(struct device *parent, struct gendisk *disk,
 	dev_set_uevent_suppress(ddev, 0);
 	disk_uevent(disk, KOBJ_ADD);
 
-	if (disk->queue->backing_dev_info->dev) {
-		err = sysfs_create_link(&ddev->kobj,
-			  &disk->queue->backing_dev_info->dev->kobj,
+	if (disk->bdi->dev) {
+		err = sysfs_create_link(&ddev->kobj, &disk->bdi->dev->kobj,
 					"bdi");
 		WARN_ON(err);
 	}
@@ -540,15 +539,14 @@ void device_add_disk(struct device *parent, struct gendisk *disk,
 		disk->flags |= GENHD_FL_SUPPRESS_PARTITION_INFO;
 		disk->flags |= GENHD_FL_NO_PART_SCAN;
 	} else {
-		struct backing_dev_info *bdi = disk->queue->backing_dev_info;
 		struct device *dev = disk_to_dev(disk);
 
 		/* Register BDI before referencing it from bdev */
 		dev->devt = MKDEV(disk->major, disk->first_minor);
-		ret = bdi_register(bdi, "%u:%u",
+		ret = bdi_register(disk->bdi, "%u:%u",
 				   disk->major, disk->first_minor);
 		WARN_ON(ret);
-		bdi_set_owner(bdi, dev);
+		bdi_set_owner(disk->bdi, dev);
 		bdev_add(disk->part0, dev->devt);
 	}
 	register_disk(parent, disk, groups);
@@ -615,7 +613,7 @@ void del_gendisk(struct gendisk *disk)
 		 * Unregister bdi before releasing device numbers (as they can
 		 * get reused and we'd get clashes in sysfs).
 		 */
-		bdi_unregister(disk->queue->backing_dev_info);
+		bdi_unregister(disk->bdi);
 	}
 
 	blk_unregister_queue(disk);
@@ -1088,6 +1086,7 @@ static void disk_release(struct device *dev)
 
 	might_sleep();
 
+	bdi_put(disk->bdi);
 	if (MAJOR(dev->devt) == BLOCK_EXT_MAJOR)
 		blk_free_ext_minor(MINOR(dev->devt));
 	disk_release_events(disk);
@@ -1268,9 +1267,13 @@ struct gendisk *__alloc_disk_node(int minors, int node_id)
 	if (!disk)
 		return NULL;
 
+	disk->bdi = bdi_alloc(node_id);
+	if (!disk->bdi)
+		goto out_free_disk;
+
 	disk->part0 = bdev_alloc(disk, 0);
 	if (!disk->part0)
-		goto out_free_disk;
+		goto out_free_bdi;
 
 	disk->node_id = node_id;
 	mutex_init(&disk->open_mutex);
@@ -1292,6 +1295,8 @@ struct gendisk *__alloc_disk_node(int minors, int node_id)
 out_destroy_part_tbl:
 	xa_destroy(&disk->part_tbl);
 	iput(disk->part0->bd_inode);
+out_free_bdi:
+	bdi_put(disk->bdi);
 out_free_disk:
 	kfree(disk);
 	return NULL;

drivers/block/drbd/drbd_req.c

@@ -905,13 +905,12 @@ static bool drbd_may_do_local_read(struct drbd_device *device, sector_t sector,
 static bool remote_due_to_read_balancing(struct drbd_device *device, sector_t sector,
 		enum drbd_read_balancing rbm)
 {
-	struct backing_dev_info *bdi;
 	int stripe_shift;
 
 	switch (rbm) {
 	case RB_CONGESTED_REMOTE:
-		bdi = device->ldev->backing_bdev->bd_disk->queue->backing_dev_info;
-		return bdi_read_congested(bdi);
+		return bdi_read_congested(
+			device->ldev->backing_bdev->bd_disk->bdi);
 	case RB_LEAST_PENDING:
 		return atomic_read(&device->local_cnt) >
 			atomic_read(&device->ap_pending_cnt) + atomic_read(&device->rs_pending_cnt);

drivers/block/pktcdvd.c

@@ -1183,10 +1183,8 @@ try_next_bio:
 	wakeup = (pd->write_congestion_on > 0
 	 		&& pd->bio_queue_size <= pd->write_congestion_off);
 	spin_unlock(&pd->lock);
-	if (wakeup) {
-		clear_bdi_congested(pd->disk->queue->backing_dev_info,
-					BLK_RW_ASYNC);
-	}
+	if (wakeup)
+		clear_bdi_congested(pd->disk->bdi, BLK_RW_ASYNC);
 
 	pkt->sleep_time = max(PACKET_WAIT_TIME, 1);
 	pkt_set_state(pkt, PACKET_WAITING_STATE);
@@ -2366,7 +2364,7 @@ static void pkt_make_request_write(struct request_queue *q, struct bio *bio)
 	spin_lock(&pd->lock);
 	if (pd->write_congestion_on > 0
 	    && pd->bio_queue_size >= pd->write_congestion_on) {
-		set_bdi_congested(q->backing_dev_info, BLK_RW_ASYNC);
+		set_bdi_congested(bio->bi_bdev->bd_disk->bdi, BLK_RW_ASYNC);
 		do {
 			spin_unlock(&pd->lock);
 			congestion_wait(BLK_RW_ASYNC, HZ);

fs/block_dev.c

@@ -1087,7 +1087,7 @@ static int blkdev_get_whole(struct block_device *bdev, fmode_t mode)
 	if (!bdev->bd_openers) {
 		set_init_blocksize(bdev);
 		if (bdev->bd_bdi == &noop_backing_dev_info)
-			bdev->bd_bdi = bdi_get(disk->queue->backing_dev_info);
+			bdev->bd_bdi = bdi_get(disk->bdi);
 	}
 	if (test_bit(GD_NEED_PART_SCAN, &disk->state))
 		bdev_disk_changed(disk, false);
@@ -1122,7 +1122,7 @@ static int blkdev_get_part(struct block_device *part, fmode_t mode)
 	disk->open_partitions++;
 	set_init_blocksize(part);
 	if (part->bd_bdi == &noop_backing_dev_info)
-		part->bd_bdi = bdi_get(disk->queue->backing_dev_info);
+		part->bd_bdi = bdi_get(disk->bdi);
 done:
 	part->bd_openers++;
 	return 0;

fs/fat/fatent.c

@@ -5,6 +5,7 @@
 
 #include <linux/blkdev.h>
 #include <linux/sched/signal.h>
+#include <linux/backing-dev-defs.h>
 #include "fat.h"
 
 struct fatent_operations {

include/linux/blkdev.h

@@ -11,7 +11,6 @@
 #include <linux/minmax.h>
 #include <linux/timer.h>
 #include <linux/workqueue.h>
-#include <linux/backing-dev-defs.h>
 #include <linux/wait.h>
 #include <linux/mempool.h>
 #include <linux/pfn.h>
@@ -398,8 +397,6 @@ struct request_queue {
 	struct blk_mq_hw_ctx	**queue_hw_ctx;
 	unsigned int		nr_hw_queues;
 
-	struct backing_dev_info	*backing_dev_info;
-
 	/*
 	 * The queue owner gets to use this for whatever they like.
 	 * ll_rw_blk doesn't touch it.

include/linux/genhd.h

@@ -158,6 +158,7 @@ struct gendisk {
 	struct mutex open_mutex;	/* open/close mutex */
 	unsigned open_partitions;	/* number of open partitions */
 
+	struct backing_dev_info	*bdi;
 	struct kobject *slave_dir;
 #ifdef CONFIG_BLOCK_HOLDER_DEPRECATED
 	struct list_head slave_bdevs;