Message-Id: <20231221085712.1766333-2-yukuai1@huaweicloud.com>
Date: Thu, 21 Dec 2023 16:56:56 +0800
From: Yu Kuai <yukuai1@...weicloud.com>
To: axboe@...nel.dk,
	roger.pau@...rix.com,
	colyli@...e.de,
	kent.overstreet@...il.com,
	joern@...ybastard.org,
	miquel.raynal@...tlin.com,
	richard@....at,
	vigneshr@...com,
	sth@...ux.ibm.com,
	hoeppner@...ux.ibm.com,
	hca@...ux.ibm.com,
	gor@...ux.ibm.com,
	agordeev@...ux.ibm.com,
	jejb@...ux.ibm.com,
	martin.petersen@...cle.com,
	clm@...com,
	josef@...icpanda.com,
	dsterba@...e.com,
	viro@...iv.linux.org.uk,
	brauner@...nel.org,
	nico@...xnic.net,
	xiang@...nel.org,
	chao@...nel.org,
	tytso@....edu,
	adilger.kernel@...ger.ca,
	jack@...e.com,
	konishi.ryusuke@...il.com,
	willy@...radead.org,
	akpm@...ux-foundation.org,
	hare@...e.de,
	p.raghav@...sung.com
Cc: linux-block@...r.kernel.org,
	linux-kernel@...r.kernel.org,
	xen-devel@...ts.xenproject.org,
	linux-bcache@...r.kernel.org,
	linux-mtd@...ts.infradead.org,
	linux-s390@...r.kernel.org,
	linux-scsi@...r.kernel.org,
	linux-bcachefs@...r.kernel.org,
	linux-btrfs@...r.kernel.org,
	linux-fsdevel@...r.kernel.org,
	linux-erofs@...ts.ozlabs.org,
	linux-ext4@...r.kernel.org,
	linux-nilfs@...r.kernel.org,
	yukuai3@...wei.com,
	yukuai1@...weicloud.com,
	yi.zhang@...wei.com,
	yangerkun@...wei.com
Subject: [PATCH RFC v3 for-6.8/block 01/17] block: add some bdev apis

From: Yu Kuai <yukuai3@...wei.com>

Those apis will be used for other modules, so that bd_inode won't be
accessed directly from other modules.

Signed-off-by: Yu Kuai <yukuai3@...wei.com>
---
 block/bdev.c           | 148 +++++++++++++++++++++++++++++++++++++++++
 block/blk.h            |   2 -
 include/linux/blkdev.h |  17 +++++
 3 files changed, 165 insertions(+), 2 deletions(-)

diff --git a/block/bdev.c b/block/bdev.c
index 750aec178b6a..6204621c6db6 100644
--- a/block/bdev.c
+++ b/block/bdev.c
@@ -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);
diff --git a/block/blk.h b/block/blk.h
index 1ef920f72e0f..ce5fcd927eaa 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -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);
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index bc236e77d85e..ccac7d32bb86 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -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);
-- 
2.39.2

Powered by blists - more mailing lists