path: root/drivers/md/dm-zone.c

// SPDX-License-Identifier: GPL-2.0
/*
 * Copyright (C) 2021 Western Digital Corporation or its affiliates.
 */

#include <linux/blkdev.h>
#include <linux/mm.h>
#include <linux/sched/mm.h>
#include <linux/slab.h>

#include "dm-core.h"

#define DM_MSG_PREFIX "zone"

#define DM_ZONE_INVALID_WP_OFST		UINT_MAX

/*
 * For internal zone reports bypassing the top BIO submission path.
 */
static int dm_blk_do_report_zones(struct mapped_device *md, struct dm_table *t,
				  sector_t sector, unsigned int nr_zones,
				  report_zones_cb cb, void *data)
{
	struct gendisk *disk = md->disk;
	int ret;
	struct dm_report_zones_args args = {
		.next_sector = sector,
		.orig_data = data,
		.orig_cb = cb,
	};

	do {
		struct dm_target *tgt;

		tgt = dm_table_find_target(t, args.next_sector);
		if (WARN_ON_ONCE(!tgt->type->report_zones))
			return -EIO;

		args.tgt = tgt;
		ret = tgt->type->report_zones(tgt, &args,
					      nr_zones - args.zone_idx);
		if (ret < 0)
			return ret;
	} while (args.zone_idx < nr_zones &&
		 args.next_sector < get_capacity(disk));

	return args.zone_idx;
}

/*
 * User facing dm device block device report zone operation. This calls the
 * report_zones operation for each target of a device table. This operation is
 * generally implemented by targets using dm_report_zones().
 */
int dm_blk_report_zones(struct gendisk *disk, sector_t sector,
			unsigned int nr_zones, report_zones_cb cb, void *data)
{
	struct mapped_device *md = disk->private_data;
	struct dm_table *map;
	int srcu_idx, ret;

	if (dm_suspended_md(md))
		return -EAGAIN;

	map = dm_get_live_table(md, &srcu_idx);
	if (!map)
		return -EIO;

	ret = dm_blk_do_report_zones(md, map, sector, nr_zones, cb, data);

	dm_put_live_table(md, srcu_idx);

	return ret;
}

static int dm_report_zones_cb(struct blk_zone *zone, unsigned int idx,
			      void *data)
{
	struct dm_report_zones_args *args = data;
	sector_t sector_diff = args->tgt->begin - args->start;

	/*
	 * Ignore zones beyond the target range.
	 */
	if (zone->start >= args->start + args->tgt->len)
		return 0;

	/*
	 * Remap the start sector and write pointer position of the zone
	 * to match its position in the target range.
	 */
	zone->start += sector_diff;
	if (zone->type != BLK_ZONE_TYPE_CONVENTIONAL) {
		if (zone->cond == BLK_ZONE_COND_FULL)
			zone->wp = zone->start + zone->len;
		else if (zone->cond == BLK_ZONE_COND_EMPTY)
			zone->wp = zone->start;
		else
			zone->wp += sector_diff;
	}

	args->next_sector = zone->start + zone->len;
	return args->orig_cb(zone, args->zone_idx++, args->orig_data);
}

/*
 * Helper for drivers of zoned targets to implement struct target_type
 * report_zones operation.
 */
int dm_report_zones(struct block_device *bdev, sector_t start, sector_t sector,
		    struct dm_report_zones_args *args, unsigned int nr_zones)
{
	/*
	 * Set the target mapping start sector first so that
	 * dm_report_zones_cb() can correctly remap zone information.
	 */
	args->start = start;

	return blkdev_report_zones(bdev, sector, nr_zones,
				   dm_report_zones_cb, args);
}
EXPORT_SYMBOL_GPL(dm_report_zones);

bool dm_is_zone_write(struct mapped_device *md, struct bio *bio)
{
	struct request_queue *q = md->queue;

	if (!blk_queue_is_zoned(q))
		return false;

	switch (bio_op(bio)) {
	case REQ_OP_WRITE_ZEROES:
	case REQ_OP_WRITE:
		return !op_is_flush(bio->bi_opf) && bio_sectors(bio);
	default:
		return false;
	}
}

void dm_cleanup_zoned_dev(struct mapped_device *md)
{
	struct request_queue *q = md->queue;

	if (q) {
		kfree(q->conv_zones_bitmap);
		q->conv_zones_bitmap = NULL;
		kfree(q->seq_zones_wlock);
		q->seq_zones_wlock = NULL;
	}

	kvfree(md->zwp_offset);
	md->zwp_offset = NULL;
	md->nr_zones = 0;
}

static unsigned int dm_get_zone_wp_offset(struct blk_zone *zone)
{
	switch (zone->cond) {
	case BLK_ZONE_COND_IMP_OPEN:
	case BLK_ZONE_COND_EXP_OPEN:
	case BLK_ZONE_COND_CLOSED:
		return zone->wp - zone->start;
	case BLK_ZONE_COND_FULL:
		return zone->len;
	case BLK_ZONE_COND_EMPTY:
	case BLK_ZONE_COND_NOT_WP:
	case BLK_ZONE_COND_OFFLINE:
	case BLK_ZONE_COND_READONLY:
	default:
		/*
		 * Conventional, offline and read-only zones do not have a valid
		 * write pointer. Use 0 as for an empty zone.
		 */
		return 0;
	}
}

static int dm_zone_revalidate_cb(struct blk_zone *zone, unsigned int idx,
				 void *data)
{
	struct mapped_device *md = data;
	struct request_queue *q = md->queue;

	switch (zone->type) {
	case BLK_ZONE_TYPE_CONVENTIONAL:
		if (!q->conv_zones_bitmap) {
			q->conv_zones_bitmap =
				kcalloc(BITS_TO_LONGS(q->nr_zones),
					sizeof(unsigned long), GFP_NOIO);
			if (!q->conv_zones_bitmap)
				return -ENOMEM;
		}
		set_bit(idx, q->conv_zones_bitmap);
		break;
	case BLK_ZONE_TYPE_SEQWRITE_REQ:
	case BLK_ZONE_TYPE_SEQWRITE_PREF:
		if (!q->seq_zones_wlock) {
			q->seq_zones_wlock =
				kcalloc(BITS_TO_LONGS(q->nr_zones),
					sizeof(unsigned long), GFP_NOIO);
			if (!q->seq_zones_wlock)
				return -ENOMEM;
		}
		if (!md->zwp_offset) {
			md->zwp_offset =
				kvcalloc(q->nr_zones, sizeof(unsigned int),
					 GFP_KERNEL);
			if (!md->zwp_offset)
				return -ENOMEM;
		}
		md->zwp_offset[idx] = dm_get_zone_wp_offset(zone);

		break;
	default:
		DMERR("Invalid zone type 0x%x at sectors %llu",
		      (int)zone->type, zone->start);
		return -ENODEV;
	}

	return 0;
}

/*
 * Revalidate the zones of a mapped device to initialize resource necessary
 * for zone append emulation. Note that we cannot simply use the block layer
 * blk_revalidate_disk_zones() function here as the mapped device is suspended
 * (this is called from __bind() context).
 */
static int dm_revalidate_zones(struct mapped_device *md, struct dm_table *t)
{
	struct request_queue *q = md->queue;
	unsigned int noio_flag;
	int ret;

	/*
	 * Check if something changed. If yes, cleanup the current resources
	 * and reallocate everything.
	 */
	if (!q->nr_zones || q->nr_zones != md->nr_zones)
		dm_cleanup_zoned_dev(md);
	if (md->nr_zones)
		return 0;

	/*
	 * Scan all zones to initialize everything. Ensure that all vmalloc
	 * operations in this context are done as if GFP_NOIO was specified.
	 */
	noio_flag = memalloc_noio_save();
	ret = dm_blk_do_report_zones(md, t, 0, q->nr_zones,
				     dm_zone_revalidate_cb, md);
	memalloc_noio_restore(noio_flag);
	if (ret < 0)
		goto err;
	if (ret != q->nr_zones) {
		ret = -EIO;
		goto err;
	}

	md->nr_zones = q->nr_zones;

	return 0;

err:
	DMERR("Revalidate zones failed %d", ret);
	dm_cleanup_zoned_dev(md);
	return ret;
}

static int device_not_zone_append_capable(struct dm_target *ti,
					  struct dm_dev *dev, sector_t start,
					  sector_t len, void *data)
{
	return !blk_queue_is_zoned(bdev_get_queue(dev->bdev));
}

static bool dm_table_supports_zone_append(struct dm_table *t)
{
	struct dm_target *ti;
	unsigned int i;

	for (i = 0; i < dm_table_get_num_targets(t); i++) {
		ti = dm_table_get_target(t, i);

		if (ti->emulate_zone_append)
			return false;

		if (!ti->type->iterate_devices ||
		    ti->type->iterate_devices(ti, device_not_zone_append_capable, NULL))
			return false;
	}

	return tru