[PATCH v2] f2fs: correct removexattr behavior for null valued extended attribute

[PATCH v2] f2fs: correct removexattr behavior for null valued extended attribute

[Daeho Jeong]

__vfs_removexattr() transfers "NULL" value to the setxattr handler of
the f2fs filesystem in order to remove the extended attribute. But,
__f2fs_setxattr() just ignores the removal request when the value of
the extended attribute is already NULL. We have to remove the extended
attribute itself even if the value of that is already NULL.

We can reporduce this bug with the below:

1. touch file
2. setfattr -n "user.foo" file
3. setfattr -x "user.foo" file
4. getfattr -d file
> user.foo

Signed-off-by: Daeho Jeong <daeho.jeong@samsung.com>
Signed-off-by: Youngjin Gil <youngjin.gil@samsung.com>
Tested-by: Hobin Woo <hobin.woo@samsung.com>
Tested-by: Chao Yu <yuchao0@huawei.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
---
 fs/f2fs/xattr.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
index ec8961e..2776618 100644
--- a/fs/f2fs/xattr.c
+++ b/fs/f2fs/xattr.c
@@ -598,7 +598,7 @@ static int __f2fs_setxattr(struct inode *inode, int index,
 			goto exit;
 		}
 
-		if (f2fs_xattr_value_same(here, value, size))
+		if (value && f2fs_xattr_value_same(here, value, size))
 			goto exit;
 	} else if ((flags & XATTR_REPLACE)) {
 		error = -ENODATA;
-- 
1.9.1


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot

Re: [PATCH v2] f2fs: correct removexattr behavior for null valued extended attribute

[Chao Yu]

Hi Jaegeuk,

Forgot to merge this patch? ;)

On 2018/1/10 10:24, Daeho Jeong wrote:
> __vfs_removexattr() transfers "NULL" value to the setxattr handler of
> the f2fs filesystem in order to remove the extended attribute. But,
> __f2fs_setxattr() just ignores the removal request when the value of
> the extended attribute is already NULL. We have to remove the extended
> attribute itself even if the value of that is already NULL.
> 
> We can reporduce this bug with the below:
> 
> 1. touch file
> 2. setfattr -n "user.foo" file
> 3. setfattr -x "user.foo" file
> 4. getfattr -d file
>> user.foo
> 
> Signed-off-by: Daeho Jeong <daeho.jeong@samsung.com>
> Signed-off-by: Youngjin Gil <youngjin.gil@samsung.com>
> Tested-by: Hobin Woo <hobin.woo@samsung.com>
> Tested-by: Chao Yu <yuchao0@huawei.com>
> Reviewed-by: Chao Yu <yuchao0@huawei.com>
> ---
>  fs/f2fs/xattr.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
> index ec8961e..2776618 100644
> --- a/fs/f2fs/xattr.c
> +++ b/fs/f2fs/xattr.c
> @@ -598,7 +598,7 @@ static int __f2fs_setxattr(struct inode *inode, int index,
>  			goto exit;
>  		}
>  
> -		if (f2fs_xattr_value_same(here, value, size))
> +		if (value && f2fs_xattr_value_same(here, value, size))
>  			goto exit;
>  	} else if ((flags & XATTR_REPLACE)) {
>  		error = -ENODATA;
> 


------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot

Re: [PATCH v2] f2fs: correct removexattr behavior for null valued extended attribute

[Jaegeuk Kim]

On 01/17, Chao Yu wrote:
> Hi Jaegeuk,
> 
> Forgot to merge this patch? ;)

Weird. I didn't get this patch before.
Is this a full patch?

> 
> On 2018/1/10 10:24, Daeho Jeong wrote:
> > __vfs_removexattr() transfers "NULL" value to the setxattr handler of
> > the f2fs filesystem in order to remove the extended attribute. But,
> > __f2fs_setxattr() just ignores the removal request when the value of
> > the extended attribute is already NULL. We have to remove the extended
> > attribute itself even if the value of that is already NULL.
> > 
> > We can reporduce this bug with the below:
> > 
> > 1. touch file
> > 2. setfattr -n "user.foo" file
> > 3. setfattr -x "user.foo" file
> > 4. getfattr -d file
> >> user.foo
> > 
> > Signed-off-by: Daeho Jeong <daeho.jeong@samsung.com>
> > Signed-off-by: Youngjin Gil <youngjin.gil@samsung.com>
> > Tested-by: Hobin Woo <hobin.woo@samsung.com>
> > Tested-by: Chao Yu <yuchao0@huawei.com>
> > Reviewed-by: Chao Yu <yuchao0@huawei.com>
> > ---
> >  fs/f2fs/xattr.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
> > index ec8961e..2776618 100644
> > --- a/fs/f2fs/xattr.c
> > +++ b/fs/f2fs/xattr.c
> > @@ -598,7 +598,7 @@ static int __f2fs_setxattr(struct inode *inode, int index,
> >  			goto exit;
> >  		}
> >  
> > -		if (f2fs_xattr_value_same(here, value, size))
> > +		if (value && f2fs_xattr_value_same(here, value, size))
> >  			goto exit;
> >  	} else if ((flags & XATTR_REPLACE)) {
> >  		error = -ENODATA;
> > 

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot

Re: [PATCH v2] f2fs: correct removexattr behavior for null valued extended attribute

[Chao Yu]

On 2018/1/20 12:24, Jaegeuk Kim wrote:
> On 01/17, Chao Yu wrote:
>> Hi Jaegeuk,
>>
>> Forgot to merge this patch? ;)
> 
> Weird. I didn't get this patch before.

I'm sure it has been sent to f2fs mailing list, may be it's been junked?

Anyway, I just resent it for Daeho Jeong, please check it.

> Is this a full patch?
> 
>>
>> On 2018/1/10 10:24, Daeho Jeong wrote:
>>> __vfs_removexattr() transfers "NULL" value to the setxattr handler of
>>> the f2fs filesystem in order to remove the extended attribute. But,
>>> __f2fs_setxattr() just ignores the removal request when the value of
>>> the extended attribute is already NULL. We have to remove the extended
>>> attribute itself even if the value of that is already NULL.
>>>
>>> We can reporduce this bug with the below:
>>>
>>> 1. touch file
>>> 2. setfattr -n "user.foo" file
>>> 3. setfattr -x "user.foo" file
>>> 4. getfattr -d file
>>>> user.foo
>>>
>>> Signed-off-by: Daeho Jeong <daeho.jeong@samsung.com>
>>> Signed-off-by: Youngjin Gil <youngjin.gil@samsung.com>
>>> Tested-by: Hobin Woo <hobin.woo@samsung.com>
>>> Tested-by: Chao Yu <yuchao0@huawei.com>
>>> Reviewed-by: Chao Yu <yuchao0@huawei.com>
>>> ---
>>>  fs/f2fs/xattr.c | 2 +-
>>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
>>> index ec8961e..2776618 100644
>>> --- a/fs/f2fs/xattr.c
>>> +++ b/fs/f2fs/xattr.c
>>> @@ -598,7 +598,7 @@ static int __f2fs_setxattr(struct inode *inode, int index,
>>>  			goto exit;
>>>  		}
>>>  
>>> -		if (f2fs_xattr_value_same(here, value, size))
>>> +		if (value && f2fs_xattr_value_same(here, value, size))
>>>  			goto exit;
>>>  	} else if ((flags & XATTR_REPLACE)) {
>>>  		error = -ENODATA;
>>>
> 
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> _______________________________________________
> Linux-f2fs-devel mailing list
> Linux-f2fs-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
> 

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot