On Fri, Aug 21, 2009 at 04:06:59PM +0200, Jan Kara wrote:

> >   Hi,
> >
> > > I also have commented a possible bug in existing ext2 code, marked with XXX.
> >   Looks good, except:
> >
> > > +int ext2_setsize(struct inode *inode, loff_t newsize)
> >   This could be static.
> >
> > > @@ -1459,8 +1540,15 @@ int ext2_setattr(struct dentry *dentry,
> > >  		if (error)
> > >  			return error;
> > >  	}
> > > -	error = inode_setattr(inode, iattr);
> > > +	if (iattr->ia_valid & ATTR_SIZE) {
> > > +		error = ext2_setsize(inode, iattr->ia_size);
> > > +		if (error)
> > > +			return error;
> > > +	}
> > > +	generic_setattr(inode, iattr);
> >   Here, we should store the error code I suppose...
>   Ah, I was confused. generic_setattr() returns void. But then remove
> the check !error from:
>   if (!error && (iattr->ia_valid & ATTR_MODE))
> which just follows the generic_setattr(). That's what made me think
> generic_setattr() returns something :)

Yep, good suggestion.

Cc: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

fs/ext2/inode.c

@@ -1548,7 +1548,7 @@ int ext2_setattr(struct dentry *dentry, struct iattr *iattr)
 			return error;
 	}
 	generic_setattr(inode, iattr);
-	if (!error && (iattr->ia_valid & ATTR_MODE))
+	if (iattr->ia_valid & ATTR_MODE)
 		error = ext2_acl_chmod(inode);
 	mark_inode_dirty(inode);