[PATCH RFC v3 for-6.8/block 00/17] block: don't access bd_inode directly from other modules

@@ -89,6 +89,25 @@ void invalidate_bdev(struct block_device *bdev)
 }
 EXPORT_SYMBOL(invalidate_bdev);
 
+/**
+ * invalidate_bdev_pages - Invalidate clean unused buffers and pagecache.
+ * @bdev: the block device which holds the cache to invalidate
+ * @start: the offset 'from' which to invalidate
+ * @end: the offset 'to' which to invalidate (inclusive)
+ *
+ * This function removes pages that are clean, unmapped and unlocked,
+ * as well as shadow entries. It will not block on IO activity.
+ *
+ * If you want to remove all the pages of one block device, regardless of
+ * their use and writeback state, use truncate_bdev_range().
+ */
+void invalidate_bdev_range(struct block_device *bdev, pgoff_t start,
+			   pgoff_t end)
+{
+	invalidate_mapping_pages(bdev->bd_inode->i_mapping, start, end);
+}
+EXPORT_SYMBOL_GPL(invalidate_bdev_range);
+
 /*
  * Drop all buffers & page cache for given bdev range. This function bails
  * with error if bdev has other exclusive owner (such as filesystem).

@@ -121,6 +140,7 @@ int truncate_bdev_range(struct block_device *bdev, blk_mode_t mode,
 					     lstart >> PAGE_SHIFT,
 					     lend >> PAGE_SHIFT);
 }
+EXPORT_SYMBOL_GPL(truncate_bdev_range);
 
 static void set_init_blocksize(struct block_device *bdev)
 {

@@ -1102,3 +1122,131 @@ void bdev_statx_dioalign(struct inode *inode, struct kstat *stat)
 
 	blkdev_put_no_open(bdev);
 }
+
+/**
+ * bdev_read_folio - Read into block device page cache.
+ * @bdev: the block device which holds the cache to read.
+ * @pos: the offset that allocated folio will contain.
+ *
+ * Read one page into the block device page cache. If it succeeds, the folio
+ * returned will contain @pos;
+ *
+ * Return: Uptodate folio on success, ERR_PTR() on failure.
+ */
+struct folio *bdev_read_folio(struct block_device *bdev, loff_t pos)
+{
+	return mapping_read_folio_gfp(bdev->bd_inode->i_mapping,
+				      pos >> PAGE_SHIFT, GFP_KERNEL);
+}
+EXPORT_SYMBOL_GPL(bdev_read_folio);
+
+/**
+ * bdev_get_folio - Find and get a reference to a folio.
+ * @bdev: the block device which holds the address_space to search.
+ * @pos: the offset the returned folio will contain.
+ * @fgp_flags: %FGP flags modify how the folio is returned.
+ * @gfp: Memory allocation flags to use if %FGP_CREAT is specified.
+ *
+ * Looks up the page cache entry at @bdev->bd_inode->i_mapping from @pos. If
+ * this function returns a folio, it is returned with an increased refcount.
+ *
+ * Return: The found folio or an ERR_PTR() otherwise.
+ */
+struct folio *bdev_get_folio(struct block_device *bdev, loff_t pos,
+			     fgf_t fgp_flags, gfp_t gfp)
+{
+	return __filemap_get_folio(bdev->bd_inode->i_mapping, pos >> PAGE_SHIFT,
+				   fgp_flags, gfp);
+}
+EXPORT_SYMBOL_GPL(bdev_get_folio);
+
+/**
+ * bdev_wb_err_check - Has block device writeback error occurred?
+ * @bdev: the block device to check.
+ * @since: Previously-sampled @bdev->bd_inode->i_mapping->wb_err.
+ *
+ * Grab @bdev->bd_inode->i_mapping->wb_err, and see if it has changed @since
+ * the given value was sampled.
+ *
+ * Return: The latest error or 0 if it hasn't changed.
+ */
+int bdev_wb_err_check(struct block_device *bdev, errseq_t since)
+{
+	return errseq_check(&bdev->bd_inode->i_mapping->wb_err, since);
+}
+EXPORT_SYMBOL_GPL(bdev_wb_err_check);
+
+/**
+ * bdev_wb_err_check_and_advance() - Check block device writeback error and
+ * advance to current value.
+ * @bdev: the block device to check;
+ * @since: Pointer to previously-sampled @bdev->bd_inode->i_mapping->wb_err to
+ * check against and advance.
+ *
+ * Grab @bdev->bd_inode->i_mapping->wb_err, and see whether it matches the
+ * value that @since points to. If it does, then just return 0; If it doesn't,
+ * then the value has changed. Set the "seen" flag, and try to swap it into
+ * place as the new eseq value. Then, set that value as the new @since value,
+ * and return whatever the error portion is set to.
+ *
+ * Return: Negative errno if one has been stored, or 0 if no new error has
+ * occurred.
+ */
+int bdev_wb_err_check_and_advance(struct block_device *bdev, errseq_t *since)
+{
+	return errseq_check_and_advance(&bdev->bd_inode->i_mapping->wb_err,
+					since);
+}
+EXPORT_SYMBOL_GPL(bdev_wb_err_check_and_advance);
+
+/**
+ * bdev_balance_dirty_pages_ratelimited - balance dirty memory state.
+ * @bdev: the block device which was dirtied.
+ *
+ * Check the system's dirty state and will initiate writeback if needed.
+ */
+void bdev_balance_dirty_pages_ratelimited(struct block_device *bdev)
+{
+	return balance_dirty_pages_ratelimited(bdev->bd_inode->i_mapping);
+}
+EXPORT_SYMBOL_GPL(bdev_balance_dirty_pages_ratelimited);
+
+/**
+ * bdev_async_readahead - readahead for marked block device pages
+ * @bdev: the block device to read.
+ * @ra: file_ra_state which holds the readahead state.
+ * @file: Used by the filesystem for authentication.
+ * @index: Index of first page to be read.
+ * @req_count: Total number of pages being read by the caller.
+ *
+ * Read multiple pages into the block device page cache. The readahead logic may
+ * decide to piggyback more pages onto the read request if access patterns
+ * suggest it will improve performance.
+ */
+void bdev_sync_readahead(struct block_device *bdev, struct file_ra_state *ra,
+			 struct file *file, pgoff_t index,
+			 unsigned long req_count)
+{
+	struct file_ra_state tmp_ra = {};
+
+	if (!ra) {
+		ra = &tmp_ra;
+		file_ra_state_init(ra, bdev->bd_inode->i_mapping);
+	}
+	page_cache_sync_readahead(bdev->bd_inode->i_mapping, ra, file, index,
+				  req_count);
+}
+EXPORT_SYMBOL_GPL(bdev_sync_readahead);
+
+/**
+ * bdev_attach_wb - associate an block device with its wb
+ * @bdev: block device of interest
+ *
+ * If @bdev->bd_inode doesn't have its wb, associate it with the wb matching the
+ * %current.
+ */
+void bdev_attach_wb(struct block_device *bdev)
+{
+	inode_attach_wb(bdev->bd_inode, NULL);
+}
+EXPORT_SYMBOL_GPL(bdev_attach_wb);

@@ -465,8 +465,6 @@ extern struct device_attribute dev_attr_events_poll_msecs;
 extern struct attribute_group blk_trace_attr_group;
 
 blk_mode_t file_to_blk_mode(struct file *file);
-int truncate_bdev_range(struct block_device *bdev, blk_mode_t mode,
-		loff_t lstart, loff_t lend);
 long blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg);
 long compat_blkdev_ioctl(struct file *file, unsigned cmd, unsigned long arg);
 

@@ -24,6 +24,7 @@
 #include <linux/sbitmap.h>
 #include <linux/uuid.h>
 #include <linux/xarray.h>
+#include <linux/pagemap.h>
 
 struct module;
 struct request_queue;

@@ -1475,6 +1476,22 @@ struct block_device *blkdev_get_no_open(dev_t dev);
 void blkdev_put_no_open(struct block_device *bdev);
 
 struct block_device *I_BDEV(struct inode *inode);
+void invalidate_bdev_range(struct block_device *bdev, pgoff_t start,
+			   pgoff_t end);
+int truncate_bdev_range(struct block_device *bdev, blk_mode_t mode,
+		loff_t lstart, loff_t lend);
+struct folio *bdev_read_folio(struct block_device *bdev, loff_t pos);
+struct folio *bdev_get_folio(struct block_device *bdev, loff_t pos,
+			     fgf_t fgp_flags, gfp_t gfp);
+int bdev_wb_err_check(struct block_device *bdev, errseq_t since);
+int bdev_wb_err_check_and_advance(struct block_device *bdev, errseq_t *since);
+void bdev_balance_dirty_pages_ratelimited(struct block_device *bdev);
+void bdev_sync_readahead(struct block_device *bdev, struct file_ra_state *ra,
+			 struct file *file, pgoff_t index,
+			 unsigned long req_count);
+void bdev_attach_wb(struct block_device *bdev);
+void bdev_associated_mapping(struct block_device *bdev,
+			     struct address_space *mapping);
 
 #ifdef CONFIG_BLOCK
 void invalidate_bdev(struct block_device *bdev);

@@ -104,8 +104,7 @@ static void xen_update_blkif_status(struct xen_blkif *blkif)
 		xenbus_dev_error(blkif->be->dev, err, "block flush");
 		return;
 	}
-	invalidate_inode_pages2(
-			blkif->vbd.bdev_handle->bdev->bd_inode->i_mapping);
+	invalidate_bdev(blkif->vbd.bdev_handle->bdev);
 
 	for (i = 0; i < blkif->nr_rings; i++) {
 		ring = &blkif->rings[i];

@@ -168,14 +168,13 @@ static const char *read_super(struct cache_sb *sb, struct block_device *bdev,
 {
 	const char *err;
 	struct cache_sb_disk *s;
-	struct page *page;
+	struct folio *folio;
 	unsigned int i;
 
-	page = read_cache_page_gfp(bdev->bd_inode->i_mapping,
-				   SB_OFFSET >> PAGE_SHIFT, GFP_KERNEL);
-	if (IS_ERR(page))
+	folio = bdev_read_folio(bdev, SB_OFFSET);
+	if (IS_ERR(folio))
 		return "IO error";
-	s = page_address(page) + offset_in_page(SB_OFFSET);
+	s = folio_address(folio) + offset_in_folio(folio, SB_OFFSET);
 
 	sb->offset		= le64_to_cpu(s->offset);
 	sb->version		= le64_to_cpu(s->version);

@@ -272,7 +271,7 @@ static const char *read_super(struct cache_sb *sb, struct block_device *bdev,
 	*res = s;
 	return NULL;
 err:
-	put_page(page);
+	folio_put(folio);
 	return err;
 }
 

@@ -46,40 +46,34 @@ struct block2mtd_dev {
 /* Static info about the MTD, used in cleanup_module */
 static LIST_HEAD(blkmtd_device_list);
 
-
-static struct page *page_read(struct address_space *mapping, pgoff_t index)
-{
-	return read_mapping_page(mapping, index, NULL);
-}
-
 /* erase a specified part of the device */
 static int _block2mtd_erase(struct block2mtd_dev *dev, loff_t to, size_t len)
 {
-	struct address_space *mapping =
-				dev->bdev_handle->bdev->bd_inode->i_mapping;
-	struct page *page;
+	struct block_device *bdev = dev->bdev_handle->bdev;
+	struct folio *folio;
 	pgoff_t index = to >> PAGE_SHIFT;	// page index
 	int pages = len >> PAGE_SHIFT;
 	u_long *p;
 	u_long *max;
 
 	while (pages) {
-		page = page_read(mapping, index);
-		if (IS_ERR(page))
-			return PTR_ERR(page);
+		folio = bdev_read_folio(bdev, index << PAGE_SHIFT);
+		if (IS_ERR(folio))
+			return PTR_ERR(folio);
 
-		max = page_address(page) + PAGE_SIZE;
-		for (p=page_address(page); p<max; p++)
+		max = folio_address(folio) + folio_size(folio);
+		for (p = folio_address(folio); p < max; p++)
 			if (*p != -1UL) {
-				lock_page(page);
-				memset(page_address(page), 0xff, PAGE_SIZE);
-				set_page_dirty(page);
-				unlock_page(page);
-				balance_dirty_pages_ratelimited(mapping);
+				folio_lock(folio);
+				memset(folio_address(folio), 0xff,
+				       folio_size(folio));
+				folio_mark_dirty(folio);
+				folio_unlock(folio);
+				bdev_balance_dirty_pages_ratelimited(bdev);
 				break;
 			}
 
-		put_page(page);
+		folio_put(folio);
 		pages--;
 		index++;
 	}

@@ -106,9 +100,7 @@ static int block2mtd_read(struct mtd_info *mtd, loff_t from, size_t len,
 		size_t *retlen, u_char *buf)
 {
 	struct block2mtd_dev *dev = mtd->priv;
-	struct address_space *mapping =
-				dev->bdev_handle->bdev->bd_inode->i_mapping;
-	struct page *page;
+	struct folio *folio;
 	pgoff_t index = from >> PAGE_SHIFT;
 	int offset = from & (PAGE_SIZE-1);
 	int cpylen;

@@ -120,12 +112,13 @@ static int block2mtd_read(struct mtd_info *mtd, loff_t from, size_t len,
 			cpylen = len;	// this page
 		len = len - cpylen;
 
-		page = page_read(mapping, index);
-		if (IS_ERR(page))
-			return PTR_ERR(page);
+		folio = bdev_read_folio(dev->bdev_handle->bdev,
+					index << PAGE_SHIFT);
+		if (IS_ERR(folio))
+			return PTR_ERR(folio);
 
-		memcpy(buf, page_address(page) + offset, cpylen);
-		put_page(page);
+		memcpy(buf, folio_address(folio) + offset, cpylen);
+		folio_put(folio);
 
 		if (retlen)
 			*retlen += cpylen;

@@ -141,9 +134,8 @@ static int block2mtd_read(struct mtd_info *mtd, loff_t from, size_t len,
 static int _block2mtd_write(struct block2mtd_dev *dev, const u_char *buf,
 		loff_t to, size_t len, size_t *retlen)
 {
-	struct page *page;
-	struct address_space *mapping =
-				dev->bdev_handle->bdev->bd_inode->i_mapping;
+	struct block_device *bdev = dev->bdev_handle->bdev;
+	struct folio *folio;
 	pgoff_t index = to >> PAGE_SHIFT;	// page index
 	int offset = to & ~PAGE_MASK;	// page offset
 	int cpylen;

@@ -155,18 +147,18 @@ static int _block2mtd_write(struct block2mtd_dev *dev, const u_char *buf,
 			cpylen = len;			// this page
 		len = len - cpylen;
 
-		page = page_read(mapping, index);
-		if (IS_ERR(page))
-			return PTR_ERR(page);
+		folio = bdev_read_folio(bdev, index << PAGE_SHIFT);
+		if (IS_ERR(folio))
+			return PTR_ERR(folio);
 
-		if (memcmp(page_address(page)+offset, buf, cpylen)) {
-			lock_page(page);
-			memcpy(page_address(page) + offset, buf, cpylen);
-			set_page_dirty(page);
-			unlock_page(page);
-			balance_dirty_pages_ratelimited(mapping);
+		if (memcmp(folio_address(folio) + offset, buf, cpylen)) {
+			folio_lock(folio);
+			memcpy(folio_address(folio) + offset, buf, cpylen);
+			folio_mark_dirty(folio);
+			folio_unlock(folio);
+			bdev_balance_dirty_pages_ratelimited(bdev);
 		}
-		put_page(page);
+		folio_put(folio);
 
 		if (retlen)
 			*retlen += cpylen;

@@ -211,8 +203,7 @@ static void block2mtd_free_device(struct block2mtd_dev *dev)
 	kfree(dev->mtd.name);
 
 	if (dev->bdev_handle) {
-		invalidate_mapping_pages(
-			dev->bdev_handle->bdev->bd_inode->i_mapping, 0, -1);
+		invalidate_bdev(dev->bdev_handle->bdev);
 		bdev_release(dev->bdev_handle);
 	}
 

@@ -295,7 +286,7 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 		goto err_free_block2mtd;
 	}
 
-	if ((long)bdev->bd_inode->i_size % erase_size) {
+	if (bdev_nr_bytes(bdev) % erase_size) {
 		pr_err("erasesize must be a divisor of device size\n");
 		goto err_free_block2mtd;
 	}

@@ -313,7 +304,7 @@ static struct block2mtd_dev *add_device(char *devname, int erase_size,
 
 	dev->mtd.name = name;
 
-	dev->mtd.size = bdev->bd_inode->i_size & PAGE_MASK;
+	dev->mtd.size = bdev_nr_bytes(bdev) & PAGE_MASK;
 	dev->mtd.erasesize = erase_size;
 	dev->mtd.writesize = 1;
 	dev->mtd.writebufsize = PAGE_SIZE;

@@ -221,8 +221,9 @@ dasd_format(struct dasd_block *block, struct format_data_t *fdata)
 	 * enabling the device later.
 	 */
 	if (fdata->start_unit == 0) {
-		block->gdp->part0->bd_inode->i_blkbits =
-			blksize_bits(fdata->blksize);
+		rc = set_blocksize(block->gdp->part0, fdata->blksize);
+		if (rc)
+			return rc;
 	}
 
 	rc = base->discipline->format_device(base, fdata, 1);

@@ -32,11 +32,9 @@
  */
 unsigned char *scsi_bios_ptable(struct block_device *dev)
 {
-	struct address_space *mapping = bdev_whole(dev)->bd_inode->i_mapping;
 	unsigned char *res = NULL;
-	struct folio *folio;
+	struct folio *folio = bdev_read_folio(bdev_whole(dev), 0);
 
-	folio = read_mapping_folio(mapping, 0, NULL);
 	if (IS_ERR(folio))
 		return NULL;
 

@@ -516,11 +516,6 @@ static inline unsigned fract_exp_two(unsigned x, unsigned fract_bits)
 void bch2_bio_map(struct bio *bio, void *base, size_t);
 int bch2_bio_alloc_pages(struct bio *, size_t, gfp_t);
 
-static inline sector_t bdev_sectors(struct block_device *bdev)
-{
-	return bdev->bd_inode->i_size >> 9;
-}
-
 #define closure_bio_submit(bio, cl)					\
 do {									\
 	closure_get(cl);						\

@@ -1122,6 +1122,7 @@ void bio_add_folio_nofail(struct bio *bio, struct folio *folio, size_t len,
 	WARN_ON_ONCE(off > UINT_MAX);
 	__bio_add_page(bio, &folio->page, len, off);
 }
+EXPORT_SYMBOL_GPL(bio_add_folio_nofail);
 
 /**
  * bio_add_folio - Attempt to add part of a folio to a bio.

@@ -3620,28 +3620,24 @@ ALLOW_ERROR_INJECTION(open_ctree, ERRNO);
 static void btrfs_end_super_write(struct bio *bio)
 {
 	struct btrfs_device *device = bio->bi_private;
-	struct bio_vec *bvec;
-	struct bvec_iter_all iter_all;
-	struct page *page;
-
-	bio_for_each_segment_all(bvec, bio, iter_all) {
-		page = bvec->bv_page;
+	struct folio_iter fi;
 
+	bio_for_each_folio_all(fi, bio) {
 		if (bio->bi_status) {
 			btrfs_warn_rl_in_rcu(device->fs_info,
 				"lost page write due to IO error on %s (%d)",
 				btrfs_dev_name(device),
 				blk_status_to_errno(bio->bi_status));
-			ClearPageUptodate(page);
-			SetPageError(page);
+			folio_clear_uptodate(fi.folio);
+			folio_set_error(fi.folio);
 			btrfs_dev_stat_inc_and_print(device,
 						     BTRFS_DEV_STAT_WRITE_ERRS);
 		} else {
-			SetPageUptodate(page);
+			folio_mark_uptodate(fi.folio);
 		}
 
-		put_page(page);
-		unlock_page(page);
+		folio_put(fi.folio);
+		folio_unlock(fi.folio);
 	}
 
 	bio_put(bio);

@@ -3651,9 +3647,9 @@ struct btrfs_super_block *btrfs_read_dev_one_super(struct block_device *bdev,
 						   int copy_num, bool drop_cache)
 {
 	struct btrfs_super_block *super;
-	struct page *page;
+	struct folio *folio;
 	u64 bytenr, bytenr_orig;
-	struct address_space *mapping = bdev->bd_inode->i_mapping;
+	unsigned int nofs_flag;
 	int ret;
 
 	bytenr_orig = btrfs_sb_offset(copy_num);

@@ -3674,16 +3670,17 @@ struct btrfs_super_block *btrfs_read_dev_one_super(struct block_device *bdev,
 		 * Drop the page of the primary superblock, so later read will
 		 * always read from the device.
 		 */
-		invalidate_inode_pages2_range(mapping,
-				bytenr >> PAGE_SHIFT,
+		invalidate_bdev_range(bdev, bytenr >> PAGE_SHIFT,
 				(bytenr + BTRFS_SUPER_INFO_SIZE) >> PAGE_SHIFT);
 	}
 
-	page = read_cache_page_gfp(mapping, bytenr >> PAGE_SHIFT, GFP_NOFS);
-	if (IS_ERR(page))
-		return ERR_CAST(page);
+	nofs_flag = memalloc_nofs_save();
+	folio = bdev_read_folio(bdev, bytenr);
+	memalloc_nofs_restore(nofs_flag);
+	if (IS_ERR(folio))
+		return ERR_CAST(folio);
 
-	super = page_address(page);
+	super = folio_address(folio);
 	if (btrfs_super_magic(super) != BTRFS_MAGIC) {
 		btrfs_release_disk_super(super);
 		return ERR_PTR(-ENODATA);

@@ -3740,7 +3737,6 @@ static int write_dev_supers(struct btrfs_device *device,
 			    struct btrfs_super_block *sb, int max_mirrors)
 {
 	struct btrfs_fs_info *fs_info = device->fs_info;
-	struct address_space *mapping = device->bdev->bd_inode->i_mapping;
 	SHASH_DESC_ON_STACK(shash, fs_info->csum_shash);
 	int i;
 	int errors = 0;

@@ -3753,7 +3749,7 @@ static int write_dev_supers(struct btrfs_device *device,
 	shash->tfm = fs_info->csum_shash;
 
 	for (i = 0; i < max_mirrors; i++) {
-		struct page *page;
+		struct folio *folio;
 		struct bio *bio;
 		struct btrfs_super_block *disk_super;
 

@@ -3778,9 +3774,10 @@ static int write_dev_supers(struct btrfs_device *device,
 				    BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE,
 				    sb->csum);
 
-		page = find_or_create_page(mapping, bytenr >> PAGE_SHIFT,
-					   GFP_NOFS);
-		if (!page) {
+		folio = bdev_get_folio(device->bdev, bytenr,
+				       FGP_LOCK|FGP_ACCESSED|FGP_CREAT,
+				       GFP_NOFS);
+		if (IS_ERR(folio)) {
 			btrfs_err(device->fs_info,
 			    "couldn't get super block page for bytenr %llu",
 			    bytenr);

@@ -3789,9 +3786,9 @@ static int write_dev_supers(struct btrfs_device *device,
 		}
 
 		/* Bump the refcount for wait_dev_supers() */
-		get_page(page);
+		folio_get(folio);
 
-		disk_super = page_address(page);
+		disk_super = folio_address(folio);
 		memcpy(disk_super, sb, BTRFS_SUPER_INFO_SIZE);
 
 		/*

@@ -3805,8 +3802,8 @@ static int write_dev_supers(struct btrfs_device *device,
 		bio->bi_iter.bi_sector = bytenr >> SECTOR_SHIFT;
 		bio->bi_private = device;
 		bio->bi_end_io = btrfs_end_super_write;
-		__bio_add_page(bio, page, BTRFS_SUPER_INFO_SIZE,
-			       offset_in_page(bytenr));
+		bio_add_folio_nofail(bio, folio, BTRFS_SUPER_INFO_SIZE,
+				     offset_in_folio(folio, bytenr));
 
 		/*
 		 * We FUA only the first super block.  The others we allow to

@@ -3842,7 +3839,7 @@ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors)
 		max_mirrors = BTRFS_SUPER_MIRROR_MAX;
 
 	for (i = 0; i < max_mirrors; i++) {
-		struct page *page;
+		struct folio *folio;
 
 		ret = btrfs_sb_log_location(device, i, READ, &bytenr);
 		if (ret == -ENOENT) {

@@ -3857,27 +3854,23 @@ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors)
 		    device->commit_total_bytes)
 			break;
 
-		page = find_get_page(device->bdev->bd_inode->i_mapping,
-				     bytenr >> PAGE_SHIFT);
-		if (!page) {
+		folio = bdev_get_folio(device->bdev, bytenr, 0, 0);
+		if (!IS_ERR(folio)) {
 			errors++;
 			if (i == 0)
 				primary_failed = true;
 			continue;
 		}
 		/* Page is submitted locked and unlocked once the IO completes */
-		wait_on_page_locked(page);
-		if (PageError(page)) {
+		folio_wait_locked(folio);
+		if (folio_test_error(folio)) {
 			errors++;
 			if (i == 0)
 				primary_failed = true;
 		}
 
-		/* Drop our reference */
-		put_page(page);
-
-		/* Drop the reference from the writing run */
-		put_page(page);
+		/* Drop our reference and the reference from the writing run */
+		folio_put_refs(folio, 2);
 	}
 
 	/* log error, force error return */

@@ -1230,16 +1230,16 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 
 void btrfs_release_disk_super(struct btrfs_super_block *super)
 {
-	struct page *page = virt_to_page(super);
+	struct folio *folio = virt_to_folio(super);
 
-	put_page(page);
+	folio_put(folio);
 }
 
 static struct btrfs_super_block *btrfs_read_disk_super(struct block_device *bdev,
 						       u64 bytenr, u64 bytenr_orig)
 {
 	struct btrfs_super_block *disk_super;
-	struct page *page;
+	struct folio *folio;
 	void *p;
 	pgoff_t index;
 

@@ -1257,15 +1257,14 @@ static struct btrfs_super_block *btrfs_read_disk_super(struct block_device *bdev
 		return ERR_PTR(-EINVAL);
 
 	/* pull in the page with our super */
-	page = read_cache_page_gfp(bdev->bd_inode->i_mapping, index, GFP_KERNEL);
+	folio = bdev_read_folio(bdev, index);
+	if (IS_ERR(folio))
+		return ERR_CAST(folio);
 
-	if (IS_ERR(page))
-		return ERR_CAST(page);
-
-	p = page_address(page);
+	p = folio_address(folio);
 
 	/* align our pointer to the offset of the super block */
-	disk_super = p + offset_in_page(bytenr);
+	disk_super = p + offset_in_folio(folio, bytenr);
 
 	if (btrfs_super_bytenr(disk_super) != bytenr_orig ||
 	    btrfs_super_magic(disk_super) != BTRFS_MAGIC) {

@@ -120,8 +120,6 @@ static int sb_write_pointer(struct block_device *bdev, struct blk_zone *zones,
 		return -ENOENT;
 	} else if (full[0] && full[1]) {
 		/* Compare two super blocks */
-		struct address_space *mapping = bdev->bd_inode->i_mapping;
-		struct page *page[BTRFS_NR_SB_LOG_ZONES];
 		struct btrfs_super_block *super[BTRFS_NR_SB_LOG_ZONES];
 		int i;
 

@@ -129,15 +127,18 @@ static int sb_write_pointer(struct block_device *bdev, struct blk_zone *zones,
 			u64 zone_end = (zones[i].start + zones[i].capacity) << SECTOR_SHIFT;
 			u64 bytenr = ALIGN_DOWN(zone_end, BTRFS_SUPER_INFO_SIZE) -
 						BTRFS_SUPER_INFO_SIZE;
+			struct folio *folio;
+			unsigned int nofs_flag;
 
-			page[i] = read_cache_page_gfp(mapping,
-					bytenr >> PAGE_SHIFT, GFP_NOFS);
-			if (IS_ERR(page[i])) {
+			nofs_flag = memalloc_nofs_save();
+			folio = bdev_read_folio(bdev, bytenr);
+			memalloc_nofs_restore(nofs_flag);
+			if (IS_ERR(folio)) {
 				if (i == 1)
 					btrfs_release_disk_super(super[0]);
-				return PTR_ERR(page[i]);
+				return PTR_ERR(folio);
 			}
-			super[i] = page_address(page[i]);
+			super[i] = folio_address(folio);
 		}
 
 		if (btrfs_super_generation(super[0]) >

@@ -183,9 +183,6 @@ static int next_buffer;
 static void *cramfs_blkdev_read(struct super_block *sb, unsigned int offset,
 				unsigned int len)
 {
-	struct address_space *mapping = sb->s_bdev->bd_inode->i_mapping;
-	struct file_ra_state ra = {};
-	struct page *pages[BLKS_PER_BUF];
 	unsigned i, blocknr, buffer;
 	unsigned long devsize;
 	char *data;

@@ -214,37 +211,30 @@ static void *cramfs_blkdev_read(struct super_block *sb, unsigned int offset,
 	devsize = bdev_nr_bytes(sb->s_bdev) >> PAGE_SHIFT;
 
 	/* Ok, read in BLKS_PER_BUF pages completely first. */
-	file_ra_state_init(&ra, mapping);
-	page_cache_sync_readahead(mapping, &ra, NULL, blocknr, BLKS_PER_BUF);
-
-	for (i = 0; i < BLKS_PER_BUF; i++) {
-		struct page *page = NULL;
-
-		if (blocknr + i < devsize) {
-			page = read_mapping_page(mapping, blocknr + i, NULL);
-			/* synchronous error? */
-			if (IS_ERR(page))
-				page = NULL;
-		}
-		pages[i] = page;
-	}
+	bdev_sync_readahead(sb->s_bdev, NULL, NULL, blocknr, BLKS_PER_BUF);
 
 	buffer = next_buffer;
 	next_buffer = NEXT_BUFFER(buffer);
 	buffer_blocknr[buffer] = blocknr;
 	buffer_dev[buffer] = sb;
-
 	data = read_buffers[buffer];
+
 	for (i = 0; i < BLKS_PER_BUF; i++) {
-		struct page *page = pages[i];
+		struct folio *folio = NULL;
+
+		if (blocknr + i < devsize)
+			folio = bdev_read_folio(sb->s_bdev,
+						(blocknr + i) << PAGE_SHIFT);
 
-		if (page) {
-			memcpy_from_page(data, page, 0, PAGE_SIZE);
-			put_page(page);
-		} else
+		if (IS_ERR_OR_NULL(folio)) {
 			memset(data, 0, PAGE_SIZE);
+		} else {
+			memcpy_from_folio(data, folio, 0, PAGE_SIZE);
+			folio_put(folio);
+		}
 		data += PAGE_SIZE;
 	}
+
 	return read_buffers[buffer] + offset;
 }
 

@@ -32,8 +32,8 @@ void erofs_put_metabuf(struct erofs_buf *buf)
 void *erofs_bread(struct erofs_buf *buf, erofs_blk_t blkaddr,
 		  enum erofs_kmap_type type)
 {
-	struct inode *inode = buf->inode;
-	erofs_off_t offset = (erofs_off_t)blkaddr << inode->i_blkbits;
+	u8 blkszbits = buf->inode ? buf->inode->i_blkbits : buf->blkszbits;
+	erofs_off_t offset = (erofs_off_t)blkaddr << blkszbits;
 	pgoff_t index = offset >> PAGE_SHIFT;
 	struct page *page = buf->page;
 	struct folio *folio;

@@ -43,7 +43,9 @@ void *erofs_bread(struct erofs_buf *buf, erofs_blk_t blkaddr,
 		erofs_put_metabuf(buf);
 
 		nofs_flag = memalloc_nofs_save();
-		folio = read_cache_folio(inode->i_mapping, index, NULL, NULL);
+		folio = buf->inode ?
+			read_mapping_folio(buf->inode->i_mapping, index, NULL) :
+			bdev_read_folio(buf->bdev, offset);
 		memalloc_nofs_restore(nofs_flag);
 		if (IS_ERR(folio))
 			return folio;

@@ -67,10 +69,14 @@ void *erofs_bread(struct erofs_buf *buf, erofs_blk_t blkaddr,
 
 void erofs_init_metabuf(struct erofs_buf *buf, struct super_block *sb)
 {
-	if (erofs_is_fscache_mode(sb))
+	if (erofs_is_fscache_mode(sb)) {
 		buf->inode = EROFS_SB(sb)->s_fscache->inode;
-	else
-		buf->inode = sb->s_bdev->bd_inode;
+		buf->bdev = NULL;
+	} else {
+		buf->inode = NULL;
+		buf->bdev = sb->s_bdev;
+		buf->blkszbits = EROFS_SB(sb)->blkszbits;
+	}
 }
 
 void *erofs_read_metabuf(struct erofs_buf *buf, struct super_block *sb,

@@ -224,8 +224,10 @@ enum erofs_kmap_type {
 
 struct erofs_buf {
 	struct inode *inode;
+	struct block_device *bdev;
 	struct page *page;
 	void *base;
+	u8 blkszbits;
 	enum erofs_kmap_type kmap_type;
 };
 #define __EROFS_BUF_INITIALIZER	((struct erofs_buf){ .page = NULL })