2007-08-19 18:05:44

by Evgeniy Dushistov

Subject: [PATCH] ufs fix sun state

Different types of ufs hold state in different places,
to hide complexity of this, there is ufs_get_fs_state,
it returns state according to "UFS_SB(sb)->s_flags",
but during mount ufs_get_fs_state is called,
before setting s_flags, this cause message for
ufs types like sun ufs: "fs need fsck",
and remount in readonly state.

This patch depend on other patches that now in -mm branch.


Signed-off-by: Evgeniy Dushistov <[email protected]>

---

Index: linux-2.6.23-rc3-git1/fs/ufs/super.c
===================================================================
--- linux-2.6.23-rc3-git1.orig/fs/ufs/super.c
+++ linux-2.6.23-rc3-git1/fs/ufs/super.c
@@ -934,7 +934,7 @@ magic_found:
flags |= UFS_ST_SUN;
}

-
+ sbi->s_flags = flags;/*after that line some functions use s_flags*/
ufs_print_super_stuff(sb, usb1, usb2, usb3);

/*
@@ -1065,8 +1065,6 @@ magic_found:
UFS_MOUNT_UFSTYPE_44BSD)
uspi->s_maxsymlinklen =
fs32_to_cpu(sb, usb3->fs_un2.fs_44.fs_maxsymlinklen);
-
- sbi->s_flags = flags;

inode = iget(sb, UFS_ROOTINO);
if (!inode || is_bad_inode(inode))

--
/Evgeniy

2007-09-05 01:14:22

by Satyam Sharma

Subject: [PATCH -mm] ufs: Fix mount check in ufs_fill_super()

Hi Evgeniy,


On Sun, 19 Aug 2007, Evgeniy Dushistov wrote:
>
> Different types of ufs hold state in different places,
> to hide complexity of this, there is ufs_get_fs_state,
> it returns state according to "UFS_SB(sb)->s_flags",
> but during mount ufs_get_fs_state is called,
> before setting s_flags, this cause message for
> ufs types like sun ufs: "fs need fsck",
> and remount in readonly state.

I noticed another strange thing with that same compound condition in
ufs_fill_super(). In the present code, if (flags & UFS_ST_MASK) ==
UFS_ST_44BSD, then the whole ufs_get_fs_state() == UFS_FSOK check gets
completely skipped (!)

Is this intentional? If so, I'd recommend plonking a comment in there.
But doesn't look that way to me, because ufs_get_fs_state() does handle
the UFS_ST_44BSD case also ... Does the patch below look correct to you?

Satyam


[PATCH -mm] ufs: Fix mount check in ufs_fill_super()

The current code skips the check to verify whether the filesystem was
previously cleanly unmounted, if (flags & UFS_ST_MASK) == UFS_ST_44BSD
or UFS_ST_OLD. This looks like an inadvertent bug that slipped in due
to parantheses in the compound conditional to me, especially given that
ufs_get_fs_state() handles the UFS_ST_44BSD case perfectly well. So,
let's fix the compound condition appropriately.

Signed-off-by: Satyam Sharma <[email protected]>

---

fs/ufs/super.c | 15 ++++++++-------
1 file changed, 8 insertions(+), 7 deletions(-)

--- linux-2.6.23-rc4-mm1/fs/ufs/super.c~fix 2007-09-05 06:35:12.000000000 +0530
+++ linux-2.6.23-rc4-mm1/fs/ufs/super.c 2007-09-05 06:40:05.000000000 +0530
@@ -934,19 +934,20 @@ magic_found:
flags |= UFS_ST_SUN;
}

- sbi->s_flags = flags;/*after that line some functions use s_flags*/
+ /* Set sbi->s_flags here, used by ufs_get_fs_state() below */
+ sbi->s_flags = flags;
ufs_print_super_stuff(sb, usb1, usb2, usb3);

/*
* Check, if file system was correctly unmounted.
* If not, make it read only.
*/
- if (((flags & UFS_ST_MASK) == UFS_ST_44BSD) ||
- ((flags & UFS_ST_MASK) == UFS_ST_OLD) ||
- (((flags & UFS_ST_MASK) == UFS_ST_SUN ||
- (flags & UFS_ST_MASK) == UFS_ST_SUNOS ||
- (flags & UFS_ST_MASK) == UFS_ST_SUNx86) &&
- (ufs_get_fs_state(sb, usb1, usb3) == (UFS_FSOK - fs32_to_cpu(sb, usb1->fs_time))))) {
+ if ((((flags & UFS_ST_MASK) == UFS_ST_44BSD) ||
+ ((flags & UFS_ST_MASK) == UFS_ST_OLD) ||
+ ((flags & UFS_ST_MASK) == UFS_ST_SUN) ||
+ ((flags & UFS_ST_MASK) == UFS_ST_SUNOS) ||
+ ((flags & UFS_ST_MASK) == UFS_ST_SUNx86)) &&
+ (ufs_get_fs_state(sb, usb1, usb3) == (UFS_FSOK - fs32_to_cpu(sb, usb1->fs_time)))) {
switch(usb1->fs_clean) {
case UFS_FSCLEAN:
UFSD("fs is clean\n");