[v4,2/9] vfs: support STATX_DIOALIGN on block devices

Commit Message

Eric Biggers July 22, 2022, 7:12 a.m. UTC

From: Eric Biggers <[e-mail]>

Add support for STATX_DIOALIGN to block devices, so that direct I/O
alignment restrictions are exposed to userspace in a generic way.

Note that this breaks the tradition of stat operating only on the block
device node, not the block device itself.  However, it was felt that
doing this is preferable, in order to make the interface useful and
avoid needing separate interfaces for regular files and block devices.

Signed-off-by: Eric Biggers <[e-mail]>
---
 block/bdev.c           | 25 +++++++++++++++++++++++++
 fs/stat.c              | 12 ++++++++++++
 include/linux/blkdev.h |  4 ++++
 3 files changed, 41 insertions(+)

Comments

Christoph Hellwig July 22, 2022, 8:10 a.m. UTC | #1

Looks good:

Reviewed-by: Christoph Hellwig <[e-mail]>

Martin K. Petersen July 22, 2022, 5:32 p.m. UTC | #2

Eric,

> Add support for STATX_DIOALIGN to block devices, so that direct I/O
> alignment restrictions are exposed to userspace in a generic way.

Reviewed-by: Martin K. Petersen <[e-mail]>

Patch

diff --git a/block/bdev.c b/block/bdev.c
index 5fe06c1f2def41..cee0951e27a82a 100644
--- a/block/bdev.c
+++ b/block/bdev.c
@@ -26,6 +26,7 @@ 
 #include <linux/namei.h>
 #include <linux/part_stat.h>
 #include <linux/uaccess.h>
+#include <linux/stat.h>
 #include "../fs/internal.h"
 #include "blk.h"
 
@@ -1071,3 +1072,27 @@  void sync_bdevs(bool wait)
 	spin_unlock(&blockdev_superblock->s_inode_list_lock);
 	iput(old_inode);
 }
+
+/*
+ * Handle STATX_DIOALIGN for block devices.
+ *
+ * Note that the inode passed to this is the inode of a block device node file,
+ * not the block device's internal inode.  Therefore it is *not* valid to use
+ * I_BDEV() here; the block device has to be looked up by i_rdev instead.
+ */
+void bdev_statx_dioalign(struct inode *inode, struct kstat *stat)
+{
+	struct block_device *bdev;
+	unsigned int lbs;
+
+	bdev = blkdev_get_no_open(inode->i_rdev);
+	if (!bdev)
+		return;
+
+	lbs = bdev_logical_block_size(bdev);
+	stat->dio_mem_align = lbs;
+	stat->dio_offset_align = lbs;
+	stat->result_mask |= STATX_DIOALIGN;
+
+	blkdev_put_no_open(bdev);
+}
diff --git a/fs/stat.c b/fs/stat.c
index a7930d74448304..ef50573c72a269 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -5,6 +5,7 @@ 
  *  Copyright (C) 1991, 1992  Linus Torvalds
  */
 
+#include <linux/blkdev.h>
 #include <linux/export.h>
 #include <linux/mm.h>
 #include <linux/errno.h>
@@ -230,11 +231,22 @@  static int vfs_statx(int dfd, struct filename *filename, int flags,
 		goto out;
 
 	error = vfs_getattr(&path, stat, request_mask, flags);
+
 	stat->mnt_id = real_mount(path.mnt)->mnt_id;
 	stat->result_mask |= STATX_MNT_ID;
+
 	if (path.mnt->mnt_root == path.dentry)
 		stat->attributes |= STATX_ATTR_MOUNT_ROOT;
 	stat->attributes_mask |= STATX_ATTR_MOUNT_ROOT;
+
+	/* Handle STATX_DIOALIGN for block devices. */
+	if (request_mask & STATX_DIOALIGN) {
+		struct inode *inode = d_backing_inode(path.dentry);
+
+		if (S_ISBLK(inode->i_mode))
+			bdev_statx_dioalign(inode, stat);
+	}
+
 	path_put(&path);
 	if (retry_estale(error, lookup_flags)) {
 		lookup_flags |= LOOKUP_REVAL;
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 2f7b43444c5f8d..d75151bd43b541 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -1538,6 +1538,7 @@  int sync_blockdev(struct block_device *bdev);
 int sync_blockdev_range(struct block_device *bdev, loff_t lstart, loff_t lend);
 int sync_blockdev_nowait(struct block_device *bdev);
 void sync_bdevs(bool wait);
+void bdev_statx_dioalign(struct inode *inode, struct kstat *stat);
 void printk_all_partitions(void);
 #else
 static inline void invalidate_bdev(struct block_device *bdev)
@@ -1554,6 +1555,9 @@  static inline int sync_blockdev_nowait(struct block_device *bdev)
 static inline void sync_bdevs(bool wait)
 {
 }
+static inline void bdev_statx_dioalign(struct inode *inode, struct kstat *stat)
+{
+}
 static inline void printk_all_partitions(void)
 {
 }