diff mbox

[14/18] orangefs: implement direct_IO for the read case

Message ID 20171212183424.26406-15-martin@martinbrandenburg.com (mailing list archive)
State New, archived
Headers show

Commit Message

Martin Brandenburg Dec. 12, 2017, 6:34 p.m. UTC 
From: Martin Brandenburg <martin@omnibond.com>

Since orangefs_file_read_iter now calls generic_file_read_iter, O_DIRECT
now goes through direct_IO in the read case.  In the write case,
orangefs_file_write_iter never calls direct_IO, but handles the direct
write manually.

Signed-off-by: Martin Brandenburg <martin@omnibond.com>
---
 fs/orangefs/file.c | 21 ++++++++-------------
 1 file changed, 8 insertions(+), 13 deletions(-)
diff mbox

Patch

diff --git a/fs/orangefs/file.c b/fs/orangefs/file.c
index 1c30e9abb8cb..0223a0351d11 100644
--- a/fs/orangefs/file.c
+++ b/fs/orangefs/file.c
@@ -787,22 +787,17 @@  static int orangefs_releasepage(struct page *page, gfp_t foo)
 	return 0;
 }
 
-/*
- * Having a direct_IO entry point in the address_space_operations
- * struct causes the kernel to allows us to use O_DIRECT on
- * open. Nothing will ever call this thing, but in the future we
- * will need to be able to use O_DIRECT on open in order to support
- * AIO. Modeled after NFS, they do this too.
- */
-
 static ssize_t orangefs_direct_IO(struct kiocb *iocb,
 				  struct iov_iter *iter)
 {
-	gossip_debug(GOSSIP_INODE_DEBUG,
-		     "orangefs_direct_IO: %pD\n",
-		     iocb->ki_filp);
-
-	return -EINVAL;
+	struct file *file = iocb->ki_filp;
+	loff_t pos = *(&iocb->ki_pos);
+	/*
+	 * This cannot happen until write_iter becomes
+	 * generic_file_write_iter.
+	 */
+	BUG_ON(iov_iter_rw(iter) != READ);
+	return do_readv_writev(ORANGEFS_IO_READ, file, &pos, iter);
 }
 
 /** ORANGEFS2 implementation of address space operations */