2017-04-26 07:03:54

by Heloise

[permalink] [raw]
Subject: [PATCH] fs:affs:fix sb_bread() return value

When the function sb_bread() fails, the return value should
be -EIO, fix it.

Signed-off-by: Heloise <[email protected]>
---
fs/affs/super.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/affs/super.c b/fs/affs/super.c
index c2c27a8..df1f4fe 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -419,7 +419,7 @@ static int affs_fill_super(struct super_block *sb, void *data, int silent)
boot_bh = sb_bread(sb, 0);
if (!boot_bh) {
pr_err("Cannot read boot block\n");
- return -EINVAL;
+ return -EIO;
}
memcpy(sig, boot_bh->b_data, 4);
brelse(boot_bh);
--
2.1.0



2017-04-26 07:46:36

by Jan Kara

[permalink] [raw]
Subject: Re: [PATCH] fs:affs:fix sb_bread() return value

On Tue 25-04-17 23:59:14, Heloise wrote:
> When the function sb_bread() fails, the return value should
> be -EIO, fix it.
>
> Signed-off-by: Heloise <[email protected]>

Well, in the mount(2) manpage we have:

EINVAL source had an invalid superblock. Or, a remount (MS_REMOUNT) was
attempted, but source was not already mounted on target. Or, a move
(MS_MOVE) was attempted, but source was not a mount point, or was '/'.

And EIO is not documented as possible error for mount(2). Looking at e.g.
ext4 it also returns EINVAL when superblock cannot be read so I guess it is
an expected return value in this case.

Honza

> ---
> fs/affs/super.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/affs/super.c b/fs/affs/super.c
> index c2c27a8..df1f4fe 100644
> --- a/fs/affs/super.c
> +++ b/fs/affs/super.c
> @@ -419,7 +419,7 @@ static int affs_fill_super(struct super_block *sb, void *data, int silent)
> boot_bh = sb_bread(sb, 0);
> if (!boot_bh) {
> pr_err("Cannot read boot block\n");
> - return -EINVAL;
> + return -EIO;
> }
> memcpy(sig, boot_bh->b_data, 4);
> brelse(boot_bh);
> --
> 2.1.0
>
>
--
Jan Kara <[email protected]>
SUSE Labs, CR