diff mbox series

[1/2] fuse: disable atomic_o_trunc if no_open is enabled

Message ID 20210812054618.26057-2-jefflexu@linux.alibaba.com (mailing list archive)
State New, archived
Headers show
Series virtiofs: miscellaneous fixes | expand

Commit Message

Jingbo Xu Aug. 12, 2021, 5:46 a.m. UTC 
From: Liu Bo <bo.liu@linux.alibaba.com>

When 'no_open' is used by virtiofsd, guest kernel won't send OPEN request
any more.  However, with atomic_o_trunc, SETATTR request is also omitted in
OPEN(O_TRUNC) so that the backend file is not truncated.  With a following
GETATTR, inode size on guest side is updated to be same with that on host
side, the end result is that O_TRUNC semantic is broken.

This disables atomic_o_trunc as well if with no_open.

Reviewed-by: Peng Tao <tao.peng@linux.alibaba.com>
Signed-off-by: Liu Bo <bo.liu@linux.alibaba.com>
Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com>
---
 fs/fuse/file.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

Comments

Miklos Szeredi Sept. 7, 2021, 8:34 a.m. UTC | #1 
On Thu, 12 Aug 2021 at 07:46, Jeffle Xu <jefflexu@linux.alibaba.com> wrote:
>
> From: Liu Bo <bo.liu@linux.alibaba.com>
>
> When 'no_open' is used by virtiofsd, guest kernel won't send OPEN request
> any more.  However, with atomic_o_trunc, SETATTR request is also omitted in
> OPEN(O_TRUNC) so that the backend file is not truncated.  With a following
> GETATTR, inode size on guest side is updated to be same with that on host
> side, the end result is that O_TRUNC semantic is broken.
>
> This disables atomic_o_trunc as well if with no_open.

I don't quite get it why one would want to enable atomic_o_trunc with
no_open in the first place?

Thanks,
Miklos
Jingbo Xu Sept. 8, 2021, 9:41 a.m. UTC | #2 
On 9/7/21 4:34 PM, Miklos Szeredi wrote:
> On Thu, 12 Aug 2021 at 07:46, Jeffle Xu <jefflexu@linux.alibaba.com> wrote:
>>
>> From: Liu Bo <bo.liu@linux.alibaba.com>
>>
>> When 'no_open' is used by virtiofsd, guest kernel won't send OPEN request
>> any more.  However, with atomic_o_trunc, SETATTR request is also omitted in
>> OPEN(O_TRUNC) so that the backend file is not truncated.  With a following
>> GETATTR, inode size on guest side is updated to be same with that on host
>> side, the end result is that O_TRUNC semantic is broken.
>>
>> This disables atomic_o_trunc as well if with no_open.
> 
> I don't quite get it why one would want to enable atomic_o_trunc with
> no_open in the first place?

Oops..We didn't realize that it could also be worked around by fuse
daemon side. Please ignore this.
diff mbox series

Patch

diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index b494ff08f08c..1231128f8dd6 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -151,10 +151,16 @@  struct fuse_file *fuse_file_open(struct fuse_mount *fm, u64 nodeid,
 			fuse_file_free(ff);
 			return ERR_PTR(err);
 		} else {
-			if (isdir)
+			if (isdir) {
 				fc->no_opendir = 1;
-			else
+			} else {
 				fc->no_open = 1;
+				/*
+				 * In case of no_open, disable atomic_o_trunc as
+				 * well.
+				 */
+				fc->atomic_o_trunc = 0;
+			}
 		}
 	}