Ocfs2: Correctly treat O_DIRECT write in holes or outside 'i_size'.

PATCH New Tristan Ye 2 participants 2 messages Aug 19, 2010 OCFS2

Share & tools

Tristan Ye

author

2010-08-19 01:14 UTC #0 lore raw

PATCH 1/1 Ocfs2: Correctly treat O_DIRECT write in holes or outside 'i_size'.

Currently, O_DIRECT writes in a hole or outside the 'i_size' was not actually honored, while using a buffered io instead.

It hurts/cheats the 'O_DIRECT' semantics somehow, users may be confusing about the data lost after a crash when performing the odirect append writes.

A straightforward way is to flush the pagecache like O_SYNC did to guarantee the real touch to disk before writes get returned.

Signed-off-by Tristan Ye <[email protected]>

---
 fs/ocfs2/file.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

fs/ocfs2/file.c +1 -1

index 2b10b36..97b41f8 100644
@@ -2321,7 +2321,7 @@ out_dio:
 	BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_DIRECT));
 
 	if (((file->f_flags & O_DSYNC) && !direct_io) || IS_SYNC(inode) ||
-	    ((file->f_flags & O_DIRECT) && has_refcount)) {
+	    ((file->f_flags & O_DIRECT) && !direct_io)) {
 		ret = filemap_fdatawrite_range(file->f_mapping, pos,
 					       pos + count - 1);
 		if (ret < 0)

Signature · 1 line

1.5.5

Tao Ma

2010-08-19 02:40 UTC #1 lore raw

PATCH 1/1 Ocfs2: Correctly treat O_DIRECT write in holes or outside 'i_size'.

Quoted 8 lines · Tristan Ye

Currently, O_DIRECT writes in a hole or outside the 'i_size' was not actually honored, while using a buffered io instead.
It hurts/cheats the 'O_DIRECT' semantics somehow, users may be confusing about the data lost after a crash when performing the odirect append writes.
A straightforward way is to flush the pagecache like O_SYNC did to guarantee the real touch to disk before writes get returned.

well, actually it is a regression bug.
So tristan, could you please describe the commit that trigger it and the real problem that I describe in the bugzilla. Also please mention the bug no.
From this commit log, it seems that we are bad from the very beginning
which isn't true. ;)

Regards,
Tao

Quoted 19 lines

Signed-off-by Tristan Ye<[email protected]>

---
  fs/ocfs2/file.c |    2 +-
  1 files changed, 1 insertions(+), 1 deletions(-)

fs/ocfs2/file.c +1 -1

index 2b10b36..97b41f8 100644
@@ -2321,7 +2321,7 @@ out_dio:
  	BUG_ON(ret == -EIOCBQUEUED&&  !(file->f_flags&  O_DIRECT));
 
  	if (((file->f_flags&  O_DSYNC)&&  !direct_io) || IS_SYNC(inode) ||
-	    ((file->f_flags&  O_DIRECT)&&  has_refcount)) {
+	    ((file->f_flags&  O_DIRECT)&&  !direct_io)) {
  		ret = filemap_fdatawrite_range(file->f_mapping, pos,
  					       pos + count - 1);
  		if (ret<  0)