ext4: Convert ext4 to new truncate calling convention

Mostly trivial conversion. We fix a bug that IS_IMMUTABLE and IS_APPEND
files could not be truncated during failed writes as we change the code.

Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/ext4/file.c  |    1 -
 fs/ext4/inode.c |   24 ++++++++++++++++++++----
 2 files changed, 20 insertions(+), 5 deletions(-)

  Ted, can you please merge this patch? Thanks.

On Wed 25-05-11 10:43:08, Jan Kara wrote:
> Mostly trivial conversion. We fix a bug that IS_IMMUTABLE and IS_APPEND
> files could not be truncated during failed writes as we change the code.
> 
> Signed-off-by: Jan Kara <jack@suse.cz>
  I've just noticed that Ted has changed the code this week as well and
removed vmtruncate() from ->setattr(). So Ted, please ignore this patch and
I'll send you another patch based on current ext4 state in your tree.

								Honza
> ---
>  fs/ext4/file.c  |    1 -
>  fs/ext4/inode.c |   24 ++++++++++++++++++++----
>  2 files changed, 20 insertions(+), 5 deletions(-)
> 
>   Ted, can you please merge this patch? Thanks.
> 
> diff --git a/fs/ext4/file.c b/fs/ext4/file.c
> index 7b80d54..2c09723 100644
> --- a/fs/ext4/file.c
> +++ b/fs/ext4/file.c
> @@ -272,7 +272,6 @@ const struct file_operations ext4_file_operations = {
>  };
>  
>  const struct inode_operations ext4_file_inode_operations = {
> -	.truncate	= ext4_truncate,
>  	.setattr	= ext4_setattr,
>  	.getattr	= ext4_getattr,
>  #ifdef CONFIG_EXT4_FS_XATTR
> diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
> index 4eca654..0aa52b4 100644
> --- a/fs/ext4/inode.c
> +++ b/fs/ext4/inode.c
> @@ -3513,7 +3513,7 @@ retry:
>  			loff_t end = offset + iov_length(iov, nr_segs);
>  
>  			if (end > isize)
> -				vmtruncate(inode, isize);
> +				ext4_truncate_failed_write(inode);
>  		}
>  	}
>  	if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
> @@ -4380,8 +4380,6 @@ static void ext4_free_branches(handle_t *handle, struct inode *inode,
>  
>  int ext4_can_truncate(struct inode *inode)
>  {
> -	if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
> -		return 0;
>  	if (S_ISREG(inode->i_mode))
>  		return 1;
>  	if (S_ISDIR(inode->i_mode))
> @@ -5236,6 +5234,24 @@ int ext4_write_inode(struct inode *inode, struct writeback_control *wbc)
>  	return err;
>  }
>  
> +static int ext4_setsize(struct inode *inode, loff_t newsize)
> +{
> +	loff_t oldsize;
> +	int rc;
> +
> +	rc = inode_newsize_ok(inode, newsize);
> +	if (rc)
> +		return rc;
> +	if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
> +		return -EPERM;
> +	oldsize = inode->i_size;
> +	i_size_write(inode, newsize);
> +	truncate_pagecache(inode, oldsize, newsize);
> +	ext4_truncate(inode);
> +
> +	return 0;
> +}
> +
>  /*
>   * ext4_setattr()
>   *
> @@ -5353,7 +5369,7 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr)
>  
>  	if ((attr->ia_valid & ATTR_SIZE) &&
>  	    attr->ia_size != i_size_read(inode))
> -		rc = vmtruncate(inode, attr->ia_size);
> +		rc = ext4_setsize(inode, attr->ia_size);
>  
>  	if (!rc) {
>  		setattr_copy(inode, attr);
> -- 
> 1.7.1
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On Wed, May 25, 2011 at 10:43:08AM +0200, Jan Kara wrote:
>  int ext4_can_truncate(struct inode *inode)
>  {
> -	if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
> -		return 0;

I agree this change is correct, but I noticed that at the VFS level,
the ftruncate() path doesn't seem to check for IS_APPEND() and
IS_IMMUTABLE().  Am I missing something?  Fixing that should be a
separate patch, of course, but it does seem to be a problem.

	 	   	       	       - Ted
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

On Wed 25-05-11 09:45:36, Ted Tso wrote:
> On Wed, May 25, 2011 at 10:43:08AM +0200, Jan Kara wrote:
> >  int ext4_can_truncate(struct inode *inode)
> >  {
> > -	if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
> > -		return 0;
> 
> I agree this change is correct, but I noticed that at the VFS level,
> the ftruncate() path doesn't seem to check for IS_APPEND() and
> IS_IMMUTABLE().  Am I missing something?  Fixing that should be a
> separate patch, of course, but it does seem to be a problem.
  It does check it - IMMUTABLE file cannot be open for writing so cannot
be ftruncated, APPEND file is checked in do_sys_ftruncate() AFAICS.


								Honza