2019-04-09 03:15:51

by Hou Tao

Subject: [PATCH] fat: issue flush after the writeback of FAT

fsync() needs to make sure the data & meta-data of file are persistent
after the return of fsync(), even when a power-failure occurs later.
In the case of fat-fs, the FAT belongs to the meta-data of file,
so we need to issue a flush after the writeback of FAT instead before.

Also bail out early when any stage of fsync fails.

Signed-off-by: Hou Tao <[email protected]>
---
fs/fat/file.c | 11 ++++++++---
1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/fs/fat/file.c b/fs/fat/file.c
index b3bed32946b1..0e3ed79fcc3f 100644
--- a/fs/fat/file.c
+++ b/fs/fat/file.c
@@ -193,12 +193,17 @@ static int fat_file_release(struct inode *inode, struct file *filp)
int fat_file_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
{
struct inode *inode = filp->f_mapping->host;
- int res, err;
+ int err;
+
+ err = __generic_file_fsync(filp, start, end, datasync);
+ if (err)
+ return err;

- res = generic_file_fsync(filp, start, end, datasync);
err = sync_mapping_buffers(MSDOS_SB(inode->i_sb)->fat_inode->i_mapping);
+ if (err)
+ return err;

- return res ? res : err;
+ return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
}


--
2.16.2.dirty

2019-04-09 03:42:33

by Darrick J. Wong

Subject: Re: [PATCH] fat: issue flush after the writeback of FAT

On Tue, Apr 09, 2019 at 11:01:58AM +0800, Hou Tao wrote:
> fsync() needs to make sure the data & meta-data of file are persistent
> after the return of fsync(), even when a power-failure occurs later.
> In the case of fat-fs, the FAT belongs to the meta-data of file,
> so we need to issue a flush after the writeback of FAT instead before.
>
> Also bail out early when any stage of fsync fails.
>
> Signed-off-by: Hou Tao <[email protected]>
> ---
> fs/fat/file.c | 11 ++++++++---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/fs/fat/file.c b/fs/fat/file.c
> index b3bed32946b1..0e3ed79fcc3f 100644
> --- a/fs/fat/file.c
> +++ b/fs/fat/file.c
> @@ -193,12 +193,17 @@ static int fat_file_release(struct inode *inode, struct file *filp)
> int fat_file_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
> {
> struct inode *inode = filp->f_mapping->host;
> - int res, err;
> + int err;
> +
> + err = __generic_file_fsync(filp, start, end, datasync);
> + if (err)
> + return err;
>
> - res = generic_file_fsync(filp, start, end, datasync);
> err = sync_mapping_buffers(MSDOS_SB(inode->i_sb)->fat_inode->i_mapping);

Huh. I would've thought that flushing the FAT would also be required
at the end of a WB_SYNC_ALL (aka data integrity) writepages call?

The patch itself seems good, though.

--D

> + if (err)
> + return err;
>
> - return res ? res : err;
> + return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
> }
>
>
> --
> 2.16.2.dirty
>

2019-04-09 05:25:36

by OGAWA Hirofumi

Subject: Re: [PATCH] fat: issue flush after the writeback of FAT

"Darrick J. Wong" <[email protected]> writes:

>> + err = __generic_file_fsync(filp, start, end, datasync);
>> + if (err)
>> + return err;
>>
>> - res = generic_file_fsync(filp, start, end, datasync);
>> err = sync_mapping_buffers(MSDOS_SB(inode->i_sb)->fat_inode->i_mapping);
>
> Huh. I would've thought that flushing the FAT would also be required
> at the end of a WB_SYNC_ALL (aka data integrity) writepages call?

In fatfs implement, FAT area is flushed by sync_mapping_buffers(fat_inode).
(FAT buffer is dirtied only by using bh->b_assoc_map to fat_inode)

Thanks.
--
OGAWA Hirofumi <[email protected]>

2019-04-09 05:28:19

by OGAWA Hirofumi

Subject: Re: [PATCH] fat: issue flush after the writeback of FAT

Hou Tao <[email protected]> writes:

> fsync() needs to make sure the data & meta-data of file are persistent
> after the return of fsync(), even when a power-failure occurs later.
> In the case of fat-fs, the FAT belongs to the meta-data of file,
> so we need to issue a flush after the writeback of FAT instead before.
>
> Also bail out early when any stage of fsync fails.
>
> Signed-off-by: Hou Tao <[email protected]>

Looks good.

Acked-by: OGAWA Hirofumi <[email protected]>

> ---
> fs/fat/file.c | 11 ++++++++---
> 1 file changed, 8 insertions(+), 3 deletions(-)
>
> diff --git a/fs/fat/file.c b/fs/fat/file.c
> index b3bed32946b1..0e3ed79fcc3f 100644
> --- a/fs/fat/file.c
> +++ b/fs/fat/file.c
> @@ -193,12 +193,17 @@ static int fat_file_release(struct inode *inode, struct file *filp)
> int fat_file_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
> {
> struct inode *inode = filp->f_mapping->host;
> - int res, err;
> + int err;
> +
> + err = __generic_file_fsync(filp, start, end, datasync);
> + if (err)
> + return err;
>
> - res = generic_file_fsync(filp, start, end, datasync);
> err = sync_mapping_buffers(MSDOS_SB(inode->i_sb)->fat_inode->i_mapping);
> + if (err)
> + return err;
>
> - return res ? res : err;
> + return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
> }

--
OGAWA Hirofumi <[email protected]>