The ioctls for the generic freeze feature are below.
o Freeze the filesystem
  int ioctl(int fd, int FIFREEZE, arg)
    fd: The file descriptor of the mountpoint
    FIFREEZE: request code for the freeze
    arg: Ignored
    Return value: 0 if the operation succeeds. Otherwise, -1

o Unfreeze the filesystem
  int ioctl(int fd, int FITHAW, arg)
    fd: The file descriptor of the mountpoint
    FITHAW: request code for unfreeze
    arg: Ignored
    Return value: 0 if the operation succeeds. Otherwise, -1

Signed-off-by: Takashi Sato <t-sato@yk.jp.nec.com>
Signed-off-by: Masayuki Hamaguchi <m-hamaguchi@ys.jp.nec.com>
---
 fs/block_dev.c              |    2 +
 fs/buffer.c                 |   34 ++++++++++++++++++++++++++++++-
 fs/ioctl.c                  |   47 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/buffer_head.h |    2 -
 include/linux/fs.h          |    7 ++++++
 5 files changed, 90 insertions(+), 2 deletions(-)

diff -uprN -X linux-2.6.27-rc5.org/Documentation/dontdiff linux-2.6.27-rc5.org/fs/block_dev.c linux-2.6.27-rc5-freeze/fs
/block_dev.c
--- linux-2.6.27-rc5.org/fs/block_dev.c	2008-08-29 07:52:02.000000000 +0900
+++ linux-2.6.27-rc5-freeze/fs/block_dev.c	2008-09-05 20:00:29.000000000 +0900
@@ -285,6 +285,8 @@ static void init_once(void *foo)
 	INIT_LIST_HEAD(&bdev->bd_holder_list);
 #endif
 	inode_init_once(&ei->vfs_inode);
+	/* Initialize mutex for freeze. */
+	mutex_init(&bdev->bd_fsfreeze_mutex);
 }
 
 static inline void __bd_forget(struct inode *inode)
diff -uprN -X linux-2.6.27-rc5.org/Documentation/dontdiff linux-2.6.27-rc5.org/fs/buffer.c linux-2.6.27-rc5-freeze/fs/bu
ffer.c
--- linux-2.6.27-rc5.org/fs/buffer.c	2008-08-29 07:52:02.000000000 +0900
+++ linux-2.6.27-rc5-freeze/fs/buffer.c	2008-09-05 20:23:13.000000000 +0900
@@ -196,11 +196,25 @@ int fsync_bdev(struct block_device *bdev
  * happen on bdev until thaw_bdev() is called.
  * If a superblock is found on this device, we take the s_umount semaphore
  * on it to make sure ...
[ message continues ]

Note that we still have duplication with the bd_mount_sem.  I think
you should look into getting rid of it and instead do a check of

Why do you add a return value here if we always return 0 anyway?

--

[ message continues ]

Hi,


The Linux kernel has already had the name of "freezer" in the part of
power-management.  So I named the above mutex "fsfreeze"
(filesystem freeze) to distinguish it from the existent "freezer".

In the original implementation,
while the filesystem is frozen, subsequent mounts wait for unfreeze
with the semaphore (bd_mount_sem).
But if we replace the semphore with the check of the freeze_count,
subsequent mounts will abort.
I think the original behavior shouldn't be changed, so the existing bd_mount_sem

I forgot to remove the unneeded return value in above old patch.
But I need to implement a return value in the new patch
because thaw_bdev() needs to return an IO error which occurs
in unlockfs().
Eric pointed it out.

Cheers, Takashi

--

[ message continues ]