for-5.16/inode-sync-2021-10-29

-----BEGIN PGP SIGNATURE-----
 
 iQJEBAABCAAuFiEEwPw5LcreJtl1+l5K99NY+ylx4KYFAmF8MEkQHGF4Ym9lQGtl
 cm5lbC5kawAKCRD301j7KXHgpkWyEACBp3TltQu/jvyFlCzuOQJqpIqVw6ZeRn9h
 0cYZaYsRzNBTzIOKogpmhT3lWYOMxIbFMq6RyzLCPaQz6juEP+tmQIdLdPMxC5ON
 XdzItF0bMaLzoW0IRK21/aF1s/7UFcr1OLT0BT8F0umeQQXcEOOSim4kZuK9u6mS
 4pOvh61yXeB7UZxDOpMqH3aVlwrLjIr51j0ECGx/Qz1OZtXREQSeptlRUKEKVTXB
 uYPCB9FLL6ZWFyiDAuaiO4Gi//dhpoOe7Yich9m0tbtfei8gl74TqgzeaCBu+gFj
 aRyfwhyvFcm69MJqPGmRBDVxtXVC6ofjd4G6PSG8R/cAuAgPFywL/s0ETmjUJBvY
 HqnExUnMcr8FUHGIfYHmX7EWCAtD+FbpUSnCgWH2ulUhziKFR/LLE/ZYayPbhrgL
 aA89BYpeDS/POc94KXJJON/Ux612vGwhJxVsngYBEboYNeiP7YwsaQapU9RsKp0o
 YTlhz8zFuToUPEh6BQLYuOZek5AsEue5o7525Aj0vdjpxH/qH6JhjE790c7yWhL+
 hbxlTAAdqdVO2Xxrr3qdMXBUI3wnFKKu8Z6+oqi7ujQRKJZmLnXYn4ZkNRs6C858
 3NEW0mySPHxNRCZrt2M7zWmoq/eZtcJIzPy4JMW3xkQgqgdImuT1z7PrgRDw6/h8
 GB382CO2AQ==
 =AKpp
 -----END PGP SIGNATURE-----

Merge tag 'for-5.16/inode-sync-2021-10-29' of git://git.kernel.dk/linux-block

Pull block inode sync updates from Jens Axboe:
 "This contains improvements to how bdev inode syncing is handled,
  unifying the API"

* tag 'for-5.16/inode-sync-2021-10-29' of git://git.kernel.dk/linux-block:
  block: simplify the block device syncing code
  ntfs3: use sync_blockdev_nowait
  fat: use sync_blockdev_nowait
  btrfs: use sync_blockdev
  xen-blkback: use sync_blockdev
  block: remove __sync_blockdev
  fs: remove __sync_filesystem

block/bdev.c

@@ -185,14 +185,13 @@ int sb_min_blocksize(struct super_block *sb, int size)
 
 EXPORT_SYMBOL(sb_min_blocksize);
 
-int __sync_blockdev(struct block_device *bdev, int wait)
+int sync_blockdev_nowait(struct block_device *bdev)
 {
 	if (!bdev)
 		return 0;
-	if (!wait)
-		return filemap_flush(bdev->bd_inode->i_mapping);
-	return filemap_write_and_wait(bdev->bd_inode->i_mapping);
+	return filemap_flush(bdev->bd_inode->i_mapping);
 }
+EXPORT_SYMBOL_GPL(sync_blockdev_nowait);
 
 /*
  * Write out and wait upon all the dirty data associated with a block
@@ -200,7 +199,9 @@ int __sync_blockdev(struct block_device *bdev, int wait)
  */
 int sync_blockdev(struct block_device *bdev)
 {
-	return __sync_blockdev(bdev, 1);
+	if (!bdev)
+		return 0;
+	return filemap_write_and_wait(bdev->bd_inode->i_mapping);
 }
 EXPORT_SYMBOL(sync_blockdev);
 
@@ -1020,7 +1021,7 @@ int __invalidate_device(struct block_device *bdev, bool kill_dirty)
 }
 EXPORT_SYMBOL(__invalidate_device);
 
-void iterate_bdevs(void (*func)(struct block_device *, void *), void *arg)
+void sync_bdevs(bool wait)
 {
 	struct inode *inode, *old_inode = NULL;
 
@@ -1051,8 +1052,19 @@ void iterate_bdevs(void (*func)(struct block_device *, void *), void *arg)
 		bdev = I_BDEV(inode);
 
 		mutex_lock(&bdev->bd_disk->open_mutex);
-		if (bdev->bd_openers)
-			func(bdev, arg);
+		if (!bdev->bd_openers) {
+			; /* skip */
+		} else if (wait) {
+			/*
+			 * We keep the error status of individual mapping so
+			 * that applications can catch the writeback error using
+			 * fsync(2). See filemap_fdatawait_keep_errors() for
+			 * details.
+			 */
+			filemap_fdatawait_keep_errors(inode->i_mapping);
+		} else {
+			filemap_fdatawrite(inode->i_mapping);
+		}
 		mutex_unlock(&bdev->bd_disk->open_mutex);
 
 		spin_lock(&blockdev_superblock->s_inode_list_lock);

drivers/block/xen-blkback/xenbus.c

@@ -98,7 +98,7 @@ static void xen_update_blkif_status(struct xen_blkif *blkif)
 		return;
 	}
 
-	err = filemap_write_and_wait(blkif->vbd.bdev->bd_inode->i_mapping);
+	err = sync_blockdev(blkif->vbd.bdev);
 	if (err) {
 		xenbus_dev_error(blkif->be->dev, err, "block flush");
 		return;

fs/btrfs/volumes.c

@@ -508,7 +508,7 @@ btrfs_get_bdev_and_sb(const char *device_path, fmode_t flags, void *holder,
 	}
 
 	if (flush)
-		filemap_write_and_wait((*bdev)->bd_inode->i_mapping);
+		sync_blockdev(*bdev);
 	ret = set_blocksize(*bdev, BTRFS_BDEV_BLOCKSIZE);
 	if (ret) {
 		blkdev_put(*bdev, flags);

fs/fat/inode.c

@@ -1940,10 +1940,8 @@ int fat_flush_inodes(struct super_block *sb, struct inode *i1, struct inode *i2)
 		ret = writeback_inode(i1);
 	if (!ret && i2)
 		ret = writeback_inode(i2);
-	if (!ret) {
-		struct address_space *mapping = sb->s_bdev->bd_inode->i_mapping;
-		ret = filemap_flush(mapping);
-	}
+	if (!ret)
+		ret = sync_blockdev_nowait(sb->s_bdev);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(fat_flush_inodes);

fs/internal.h

@@ -23,22 +23,11 @@ struct pipe_inode_info;
 #ifdef CONFIG_BLOCK
 extern void __init bdev_cache_init(void);
 
-extern int __sync_blockdev(struct block_device *bdev, int wait);
-void iterate_bdevs(void (*)(struct block_device *, void *), void *);
 void emergency_thaw_bdev(struct super_block *sb);
 #else
 static inline void bdev_cache_init(void)
 {
 }
-
-static inline int __sync_blockdev(struct block_device *bdev, int wait)
-{
-	return 0;
-}
-static inline void iterate_bdevs(void (*f)(struct block_device *, void *),
-		void *arg)
-{
-}
 static inline int emergency_thaw_bdev(struct super_block *sb)
 {
 	return 0;

fs/ntfs3/inode.c

@@ -1046,7 +1046,7 @@ int ntfs_flush_inodes(struct super_block *sb, struct inode *i1,
 	if (!ret && i2)
 		ret = writeback_inode(i2);
 	if (!ret)
-		ret = filemap_flush(sb->s_bdev->bd_inode->i_mapping);
+		ret = sync_blockdev_nowait(sb->s_bdev);
 	return ret;
 }
 

fs/sync.c

@@ -3,6 +3,7 @@
  * High-level sync()-related operations
  */
 
+#include <linux/blkdev.h>
 #include <linux/kernel.h>
 #include <linux/file.h>
 #include <linux/fs.h>
@@ -21,25 +22,6 @@
 #define VALID_FLAGS (SYNC_FILE_RANGE_WAIT_BEFORE|SYNC_FILE_RANGE_WRITE| \
 			SYNC_FILE_RANGE_WAIT_AFTER)
 
-/*
- * Do the filesystem syncing work. For simple filesystems
- * writeback_inodes_sb(sb) just dirties buffers with inodes so we have to
- * submit IO for these buffers via __sync_blockdev(). This also speeds up the
- * wait == 1 case since in that case write_inode() functions do
- * sync_dirty_buffer() and thus effectively write one block at a time.
- */
-static int __sync_filesystem(struct super_block *sb, int wait)
-{
-	if (wait)
-		sync_inodes_sb(sb);
-	else
-		writeback_inodes_sb(sb, WB_REASON_SYNC);
-
-	if (sb->s_op->sync_fs)
-		sb->s_op->sync_fs(sb, wait);
-	return __sync_blockdev(sb->s_bdev, wait);
-}
-
 /*
  * Write out and wait upon all dirty data associated with this
  * superblock.  Filesystem data as well as the underlying block
@@ -61,10 +43,25 @@ int sync_filesystem(struct super_block *sb)
 	if (sb_rdonly(sb))
 		return 0;
 
-	ret = __sync_filesystem(sb, 0);
+	/*
+	 * Do the filesystem syncing work.  For simple filesystems
+	 * writeback_inodes_sb(sb) just dirties buffers with inodes so we have
+	 * to submit I/O for these buffers via sync_blockdev().  This also
+	 * speeds up the wait == 1 case since in that case write_inode()
+	 * methods call sync_dirty_buffer() and thus effectively write one block
+	 * at a time.
+	 */
+	writeback_inodes_sb(sb, WB_REASON_SYNC);
+	if (sb->s_op->sync_fs)
+		sb->s_op->sync_fs(sb, 0);
+	ret = sync_blockdev_nowait(sb->s_bdev);
 	if (ret < 0)
 		return ret;
-	return __sync_filesystem(sb, 1);
+
+	sync_inodes_sb(sb);
+	if (sb->s_op->sync_fs)
+		sb->s_op->sync_fs(sb, 1);
+	return sync_blockdev(sb->s_bdev);
 }
 EXPORT_SYMBOL(sync_filesystem);
 
@@ -81,21 +78,6 @@ static void sync_fs_one_sb(struct super_block *sb, void *arg)
 		sb->s_op->sync_fs(sb, *(int *)arg);
 }
 
-static void fdatawrite_one_bdev(struct block_device *bdev, void *arg)
-{
-	filemap_fdatawrite(bdev->bd_inode->i_mapping);
-}
-
-static void fdatawait_one_bdev(struct block_device *bdev, void *arg)
-{
-	/*
-	 * We keep the error status of individual mapping so that
-	 * applications can catch the writeback error using fsync(2).
-	 * See filemap_fdatawait_keep_errors() for details.
-	 */
-	filemap_fdatawait_keep_errors(bdev->bd_inode->i_mapping);
-}
-
 /*
  * Sync everything. We start by waking flusher threads so that most of
  * writeback runs on all devices in parallel. Then we sync all inodes reliably
@@ -114,8 +96,8 @@ void ksys_sync(void)
 	iterate_supers(sync_inodes_one_sb, NULL);
 	iterate_supers(sync_fs_one_sb, &nowait);
 	iterate_supers(sync_fs_one_sb, &wait);
-	iterate_bdevs(fdatawrite_one_bdev, NULL);
-	iterate_bdevs(fdatawait_one_bdev, NULL);
+	sync_bdevs(false);
+	sync_bdevs(true);
 	if (unlikely(laptop_mode))
 		laptop_sync_completion();
 }
@@ -136,10 +118,10 @@ static void do_sync_work(struct work_struct *work)
 	 */
 	iterate_supers(sync_inodes_one_sb, &nowait);
 	iterate_supers(sync_fs_one_sb, &nowait);
-	iterate_bdevs(fdatawrite_one_bdev, NULL);
+	sync_bdevs(false);
 	iterate_supers(sync_inodes_one_sb, &nowait);
 	iterate_supers(sync_fs_one_sb, &nowait);
-	iterate_bdevs(fdatawrite_one_bdev, NULL);
+	sync_bdevs(false);
 	printk("Emergency Sync complete\n");
 	kfree(work);
 }

include/linux/blkdev.h

@@ -1304,6 +1304,8 @@ int truncate_bdev_range(struct block_device *bdev, fmode_t mode, loff_t lstart,
 #ifdef CONFIG_BLOCK
 void invalidate_bdev(struct block_device *bdev);
 int sync_blockdev(struct block_device *bdev);
+int sync_blockdev_nowait(struct block_device *bdev);
+void sync_bdevs(bool wait);
 #else
 static inline void invalidate_bdev(struct block_device *bdev)
 {
@@ -1312,6 +1314,13 @@ static inline int sync_blockdev(struct block_device *bdev)
 {
 	return 0;
 }
+static inline int sync_blockdev_nowait(struct block_device *bdev)
+{
+	return 0;
+}
+static inline void sync_bdevs(bool wait)
+{
+}
 #endif
 int fsync_bdev(struct block_device *bdev);