VFS: add FMODE_CAN_ODIRECT file flag

Currently various places test if direct IO is possible on a file by
checking for the existence of the direct_IO address space operation.
This is a poor choice, as the direct_IO operation may not be used - it is
only used if the generic_file_*_iter functions are called for direct IO
and some filesystems - particularly NFS - don't do this.

Instead, introduce a new f_mode flag: FMODE_CAN_ODIRECT and change the
various places to check this (avoiding pointer dereferences).
do_dentry_open() will set this flag if ->direct_IO is present, so
filesystems do not need to be changed.

NFS *is* changed, to set the flag explicitly and discard the direct_IO
entry in the address_space_operations for files.

Other filesystems which currently use noop_direct_IO could usefully be
changed to set this flag instead.

Link: https://lkml.kernel.org/r/164859778128.29473.15189737957277399416.stgit@noble.brown
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: NeilBrown <neilb@suse.de>
Tested-by: David Howells <dhowells@redhat.com>
Tested-by: Geert Uytterhoeven <geert+renesas@glider.be>
Cc: Hugh Dickins <hughd@google.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Trond Myklebust <trond.myklebust@hammerspace.com>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

drivers/block/loop.c

@@ -186,8 +186,8 @@ static void __loop_update_dio(struct loop_device *lo, bool dio)
 	 */
 	if (dio) {
 		if (queue_logical_block_size(lo->lo_queue) >= sb_bsize &&
-				!(lo->lo_offset & dio_align) &&
+		    !(lo->lo_offset & dio_align) &&
-				mapping->a_ops->direct_IO)
+		    (file->f_mode & FMODE_CAN_ODIRECT))
 			use_dio = true;
 		else
 			use_dio = false;

fs/fcntl.c

@@ -56,11 +56,10 @@ static int setfl(int fd, struct file * filp, unsigned long arg)
 		   arg |= O_NONBLOCK;
 
 	/* Pipe packetized mode is controlled by O_DIRECT flag */
-	if (!S_ISFIFO(inode->i_mode) && (arg & O_DIRECT)) {
+	if (!S_ISFIFO(inode->i_mode) &&
-		if (!filp->f_mapping || !filp->f_mapping->a_ops ||
+	    (arg & O_DIRECT) &&
-			!filp->f_mapping->a_ops->direct_IO)
+	    !(filp->f_mode & FMODE_CAN_ODIRECT))
-				return -EINVAL;
+		return -EINVAL;
-	}
 
 	if (filp->f_op->check_flags)
 		error = filp->f_op->check_flags(arg);

fs/nfs/file.c

@@ -69,6 +69,8 @@ nfs_file_open(struct inode *inode, struct file *filp)
 		return res;
 
 	res = nfs_open(inode, filp);
+	if (res == 0)
+		filp->f_mode |= FMODE_CAN_ODIRECT;
 	return res;
 }
 
@@ -536,7 +538,6 @@ const struct address_space_operations nfs_file_aops = {
 	.write_end = nfs_write_end,
 	.invalidate_folio = nfs_invalidate_folio,
 	.releasepage = nfs_release_page,
-	.direct_IO = nfs_direct_IO,
 #ifdef CONFIG_MIGRATION
 	.migratepage = nfs_migrate_page,
 #endif

fs/open.c

@@ -834,16 +834,15 @@ static int do_dentry_open(struct file *f,
 	if ((f->f_mode & FMODE_WRITE) &&
 	     likely(f->f_op->write || f->f_op->write_iter))
 		f->f_mode |= FMODE_CAN_WRITE;
+	if (f->f_mapping->a_ops && f->f_mapping->a_ops->direct_IO)
+		f->f_mode |= FMODE_CAN_ODIRECT;
 
 	f->f_flags &= ~(O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC);
 
 	file_ra_state_init(&f->f_ra, f->f_mapping->host->i_mapping);
 
-	/* NB: we're sure to have correct a_ops only after f_op->open */
+	if ((f->f_flags & O_DIRECT) && !(f->f_mode & FMODE_CAN_ODIRECT))
-	if (f->f_flags & O_DIRECT) {
+		return -EINVAL;
-		if (!f->f_mapping->a_ops || !f->f_mapping->a_ops->direct_IO)
-			return -EINVAL;
-	}
 
 	/*
 	 * XXX: Huge page cache doesn't support writing yet. Drop all page

fs/overlayfs/file.c

@@ -82,11 +82,8 @@ static int ovl_change_flags(struct file *file, unsigned int flags)
 	if (((flags ^ file->f_flags) & O_APPEND) && IS_APPEND(inode))
 		return -EPERM;
 
-	if (flags & O_DIRECT) {
+	if ((flags & O_DIRECT) && !(file->f_mode & FMODE_CAN_ODIRECT))
-		if (!file->f_mapping->a_ops ||
+		return -EINVAL;
-		    !file->f_mapping->a_ops->direct_IO)
-			return -EINVAL;
-	}
 
 	if (file->f_op->check_flags) {
 		err = file->f_op->check_flags(flags);
@@ -306,8 +303,7 @@ static ssize_t ovl_read_iter(struct kiocb *iocb, struct iov_iter *iter)
 
 	ret = -EINVAL;
 	if (iocb->ki_flags & IOCB_DIRECT &&
-	    (!real.file->f_mapping->a_ops ||
+	    !(real.file->f_mode & FMODE_CAN_ODIRECT))
-	     !real.file->f_mapping->a_ops->direct_IO))
 		goto out_fdput;
 
 	old_cred = ovl_override_creds(file_inode(file)->i_sb);
@@ -367,8 +363,7 @@ static ssize_t ovl_write_iter(struct kiocb *iocb, struct iov_iter *iter)
 
 	ret = -EINVAL;
 	if (iocb->ki_flags & IOCB_DIRECT &&
-	    (!real.file->f_mapping->a_ops ||
+	    !(real.file->f_mode & FMODE_CAN_ODIRECT))
-	     !real.file->f_mapping->a_ops->direct_IO))
 		goto out_fdput;
 
 	if (!ovl_should_sync(OVL_FS(inode->i_sb)))

include/linux/fs.h

@@ -162,6 +162,9 @@ typedef int (dio_iodone_t)(struct kiocb *iocb, loff_t offset,
 /* File is stream-like */
 #define FMODE_STREAM		((__force fmode_t)0x200000)
 
+/* File supports DIRECT IO */
+#define	FMODE_CAN_ODIRECT	((__force fmode_t)0x400000)
+
 /* File was opened by fanotify and shouldn't generate fanotify events */
 #define FMODE_NONOTIFY		((__force fmode_t)0x4000000)