Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1595106 > unrolled thread

[PATCH] f2fs: don't allow rename unencrypted file to encrypted directory

Back to article view | Back to linux.kernel

Contents

  [PATCH] f2fs: don't allow rename unencrypted file to encrypted directory Chao Yu <yuchao0@huawei.com> - 2017-03-08 13:20 +0100
    Re: [f2fs-dev] [PATCH] f2fs: don't allow rename unencrypted file to  encrypted directory Kinglong Mee <kinglongmee@gmail.com> - 2017-03-08 16:30 +0100
      Re: [f2fs-dev] [PATCH] f2fs: don't allow rename unencrypted file to  encrypted directory Chao Yu <yuchao0@huawei.com> - 2017-03-09 02:40 +0100
    Re: [PATCH] f2fs: don't allow rename unencrypted file to encrypted  directory kbuild test robot <lkp@intel.com> - 2017-03-10 23:00 +0100

#1595106 — [PATCH] f2fs: don't allow rename unencrypted file to encrypted directory

In commit d9cdc9033181 ("ext4 crypto: enforce context consistency") we
declared that:

2) All files or directories in a directory must be protected using the
    same key as their containing directory.

But in f2fs_cross_rename there is a vulnerability that allow to cross
rename unencrypted file into encrypted directory, it needs to be refused.

Signed-off-by: Chao Yu <yuchao0@huawei.com>
---
 fs/f2fs/namei.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
index 25c073f6c7d4..8de684b84cb9 100644
--- a/fs/f2fs/namei.c
+++ b/fs/f2fs/namei.c
@@ -855,6 +855,10 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 			!fscrypt_has_encryption_key(new_dir)))
 		return -ENOKEY;
 
+	if (f2fs_encrypted_inode(old_dir) && !f2fs_encrypted_inode(new_inode) ||
+		f2fs_encrypted_inode(new_dir) && !f2fs_encrypted_inode(old_inode))
+		return -EPERM;
+
 	if ((f2fs_encrypted_inode(old_dir) || f2fs_encrypted_inode(new_dir)) &&
 			(old_dir != new_dir) &&
 			(!fscrypt_has_permitted_context(new_dir, old_inode) ||
-- 
2.8.2.295.g3f1c1d0

[toc] | [next] | [standalone]

#1595274 — Re: [f2fs-dev] [PATCH] f2fs: don't allow rename unencrypted file to encrypted directory

On 3/8/2017 20:08, Chao Yu wrote:
> In commit d9cdc9033181 ("ext4 crypto: enforce context consistency") we
> declared that:
> 
> 2) All files or directories in a directory must be protected using the
>     same key as their containing directory.
> 
> But in f2fs_cross_rename there is a vulnerability that allow to cross
> rename unencrypted file into encrypted directory, it needs to be refused.

fscrypt_has_permitted_context has do the checking as this patch,

168         /* no restrictions if the parent directory is not encrypted */
169         if (!parent->i_sb->s_cop->is_encrypted(parent))
170                 return 1;
171         /* if the child directory is not encrypted, this is always a problem */
172         if (!parent->i_sb->s_cop->is_encrypted(child))
173                 return 0;

So, the cross rename unencrypted file into encrypted directory is permitted right now. 

I have a encrypted directory "ncry",  "new" is unencrypted file.

[root@nfstestnic f2fs]# renameat2 -x encry/hello new
Operation not permitted
[root@nfstestnic f2fs]# renameat2 -x encry/hello new
Operation not permitted

How do you test it? 

thanks,
Kinglong Mee
> 
> Signed-off-by: Chao Yu <yuchao0@huawei.com>
> ---
>  fs/f2fs/namei.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
> index 25c073f6c7d4..8de684b84cb9 100644
> --- a/fs/f2fs/namei.c
> +++ b/fs/f2fs/namei.c
> @@ -855,6 +855,10 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
>  			!fscrypt_has_encryption_key(new_dir)))
>  		return -ENOKEY;
>  
> +	if (f2fs_encrypted_inode(old_dir) && !f2fs_encrypted_inode(new_inode) ||
> +		f2fs_encrypted_inode(new_dir) && !f2fs_encrypted_inode(old_inode))
> +		return -EPERM;
> +
>  	if ((f2fs_encrypted_inode(old_dir) || f2fs_encrypted_inode(new_dir)) &&
>  			(old_dir != new_dir) &&
>  			(!fscrypt_has_permitted_context(new_dir, old_inode) ||
> 

[toc] | [prev] | [next] | [standalone]

#1595625 — Re: [f2fs-dev] [PATCH] f2fs: don't allow rename unencrypted file to encrypted directory

On 2017/3/8 21:35, Kinglong Mee wrote:
> On 3/8/2017 20:08, Chao Yu wrote:
>> In commit d9cdc9033181 ("ext4 crypto: enforce context consistency") we
>> declared that:
>>
>> 2) All files or directories in a directory must be protected using the
>>     same key as their containing directory.
>>
>> But in f2fs_cross_rename there is a vulnerability that allow to cross
>> rename unencrypted file into encrypted directory, it needs to be refused.
> 
> fscrypt_has_permitted_context has do the checking as this patch,
> 
> 168         /* no restrictions if the parent directory is not encrypted */
> 169         if (!parent->i_sb->s_cop->is_encrypted(parent))
> 170                 return 1;
> 171         /* if the child directory is not encrypted, this is always a problem */
> 172         if (!parent->i_sb->s_cop->is_encrypted(child))
> 173                 return 0;
> 
> So, the cross rename unencrypted file into encrypted directory is permitted right now. 
> 
> I have a encrypted directory "ncry",  "new" is unencrypted file.
> 
> [root@nfstestnic f2fs]# renameat2 -x encry/hello new
> Operation not permitted
> [root@nfstestnic f2fs]# renameat2 -x encry/hello new
> Operation not permitted

I've tried this, the same result as yours, so this patch is wrong, please ignore
it, sorry about the noise.

Thanks,

> 
> How do you test it? 
> 
> thanks,
> Kinglong Mee
>>
>> Signed-off-by: Chao Yu <yuchao0@huawei.com>
>> ---
>>  fs/f2fs/namei.c | 4 ++++
>>  1 file changed, 4 insertions(+)
>>
>> diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
>> index 25c073f6c7d4..8de684b84cb9 100644
>> --- a/fs/f2fs/namei.c
>> +++ b/fs/f2fs/namei.c
>> @@ -855,6 +855,10 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
>>  			!fscrypt_has_encryption_key(new_dir)))
>>  		return -ENOKEY;
>>  
>> +	if (f2fs_encrypted_inode(old_dir) && !f2fs_encrypted_inode(new_inode) ||
>> +		f2fs_encrypted_inode(new_dir) && !f2fs_encrypted_inode(old_inode))
>> +		return -EPERM;
>> +
>>  	if ((f2fs_encrypted_inode(old_dir) || f2fs_encrypted_inode(new_dir)) &&
>>  			(old_dir != new_dir) &&
>>  			(!fscrypt_has_permitted_context(new_dir, old_inode) ||
>>
> 
> .
> 

[toc] | [prev] | [next] | [standalone]

#1598132 — Re: [PATCH] f2fs: don't allow rename unencrypted file to encrypted directory

[Multipart message — attachments visible in raw view] — view raw

Hi Chao,

[auto build test WARNING on f2fs/dev]
[also build test WARNING on v4.11-rc1 next-20170309]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url:    https://github.com/0day-ci/linux/commits/Chao-Yu/f2fs-don-t-allow-rename-unencrypted-file-to-encrypted-directory/20170311-052705
base:   https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git dev
config: i386-randconfig-x005-201710 (attached as .config)
compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901
reproduce:
        # save the attached .config to linux build tree
        make ARCH=i386 

All warnings (new ones prefixed by >>):

   fs/f2fs/namei.c: In function 'f2fs_cross_rename':
>> fs/f2fs/namei.c:858:36: warning: suggest parentheses around '&&' within '||' [-Wparentheses]
     if (f2fs_encrypted_inode(old_dir) && !f2fs_encrypted_inode(new_inode) ||
         ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

vim +858 fs/f2fs/namei.c

   842		struct f2fs_sb_info *sbi = F2FS_I_SB(old_dir);
   843		struct inode *old_inode = d_inode(old_dentry);
   844		struct inode *new_inode = d_inode(new_dentry);
   845		struct page *old_dir_page, *new_dir_page;
   846		struct page *old_page, *new_page;
   847		struct f2fs_dir_entry *old_dir_entry = NULL, *new_dir_entry = NULL;
   848		struct f2fs_dir_entry *old_entry, *new_entry;
   849		int old_nlink = 0, new_nlink = 0;
   850		int err = -ENOENT;
   851	
   852		if ((f2fs_encrypted_inode(old_dir) &&
   853				!fscrypt_has_encryption_key(old_dir)) ||
   854				(f2fs_encrypted_inode(new_dir) &&
   855				!fscrypt_has_encryption_key(new_dir)))
   856			return -ENOKEY;
   857	
 > 858		if (f2fs_encrypted_inode(old_dir) && !f2fs_encrypted_inode(new_inode) ||
   859			f2fs_encrypted_inode(new_dir) && !f2fs_encrypted_inode(old_inode))
   860			return -EPERM;
   861	
   862		if ((f2fs_encrypted_inode(old_dir) || f2fs_encrypted_inode(new_dir)) &&
   863				(old_dir != new_dir) &&
   864				(!fscrypt_has_permitted_context(new_dir, old_inode) ||
   865				 !fscrypt_has_permitted_context(old_dir, new_inode)))
   866			return -EPERM;

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web