Re: [PATCH V2] ext4: add casefolded feature check before setup encrypted info

Re: [PATCH V2] ext4: add casefolded feature check before setup encrypted info

[Lizhi Xu]

On Thu, 30 May 2024 20:11:33 -0700, Eric Biggers wrote:
> > Due to the current file system not supporting the casefolded feature, only 
> > i_crypt_info was initialized when creating encrypted information, without actually
> > setting the sighash. Therefore, when creating an inode, if the system does not 
> > support the casefolded feature, encrypted information will not be created.
> > 
> > Reported-by: syzbot+340581ba9dceb7e06fb3@syzkaller.appspotmail.com
> > Signed-off-by: Lizhi Xu <lizhi.xu@windriver.com>
> > ---
> >  fs/ext4/ialloc.c | 3 ++-
> >  1 file changed, 2 insertions(+), 1 deletion(-)
> > 
> > diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
> > index e9bbb1da2d0a..47b75589fdf4 100644
> > --- a/fs/ext4/ialloc.c
> > +++ b/fs/ext4/ialloc.c
> > @@ -983,7 +983,8 @@ struct inode *__ext4_new_inode(struct mnt_idmap *idmap,
> >  		ei->i_projid = make_kprojid(&init_user_ns, EXT4_DEF_PROJID);
> >  
> >  	if (!(i_flags & EXT4_EA_INODE_FL)) {
> > -		err = fscrypt_prepare_new_inode(dir, inode, &encrypt);
> > +		if (ext4_has_feature_casefold(inode->i_sb))
> > +			err = fscrypt_prepare_new_inode(dir, inode, &encrypt);
> >  		if (err)
> >  			goto out;
> 
> No, this is not correct at all.  This just disables encryption on filesystems
> with the casefold feature.
If filesystems not support casefold feature, Why do I need to setup encrypted
information when creating a directory? Can encrypted information not include *hash?
> 
> As I said before, please also use the correct mailing lists.
Added.

Lizhi

Re: [PATCH V2] ext4: add casefolded feature check before setup encrypted info

[Eric Biggers]

On Fri, May 31, 2024 at 11:30:44AM +0800, 'Lizhi Xu' via syzkaller-bugs wrote:
> On Thu, 30 May 2024 20:11:33 -0700, Eric Biggers wrote:
> > > Due to the current file system not supporting the casefolded feature, only 
> > > i_crypt_info was initialized when creating encrypted information, without actually
> > > setting the sighash. Therefore, when creating an inode, if the system does not 
> > > support the casefolded feature, encrypted information will not be created.
> > > 
> > > Reported-by: syzbot+340581ba9dceb7e06fb3@syzkaller.appspotmail.com
> > > Signed-off-by: Lizhi Xu <lizhi.xu@windriver.com>
> > > ---
> > >  fs/ext4/ialloc.c | 3 ++-
> > >  1 file changed, 2 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
> > > index e9bbb1da2d0a..47b75589fdf4 100644
> > > --- a/fs/ext4/ialloc.c
> > > +++ b/fs/ext4/ialloc.c
> > > @@ -983,7 +983,8 @@ struct inode *__ext4_new_inode(struct mnt_idmap *idmap,
> > >  		ei->i_projid = make_kprojid(&init_user_ns, EXT4_DEF_PROJID);
> > >  
> > >  	if (!(i_flags & EXT4_EA_INODE_FL)) {
> > > -		err = fscrypt_prepare_new_inode(dir, inode, &encrypt);
> > > +		if (ext4_has_feature_casefold(inode->i_sb))
> > > +			err = fscrypt_prepare_new_inode(dir, inode, &encrypt);
> > >  		if (err)
> > >  			goto out;
> > 
> > No, this is not correct at all.  This just disables encryption on filesystems
> > with the casefold feature.
> If filesystems not support casefold feature, Why do I need to setup encrypted
> information when creating a directory? Can encrypted information not include *hash?

Encryption is a separate feature.  It is supported both with and without
casefold.

- Eric

[PATCH V3] ext4: check hash version and filesystem casefolded consistent

[Lizhi Xu]

When mounting the ext4 filesystem, if the hash version and casefolded are not
consistent, exit the mounting.

Reported-by: syzbot+340581ba9dceb7e06fb3@syzkaller.appspotmail.com
Signed-off-by: Lizhi Xu <lizhi.xu@windriver.com>
---
 fs/ext4/super.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index c682fb927b64..c0036e3922c2 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -5262,6 +5262,9 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb)
 		goto failed_mount;
 
 	ext4_hash_info_init(sb);
+	if (es->s_def_hash_version == DX_HASH_SIPHASH && 
+	    !ext4_has_feature_casefold(sb))
+		goto failed_mount;
 
 	err = ext4_handle_clustersize(sb);
 	if (err)
-- 
2.43.0

Re: [PATCH V3] ext4: check hash version and filesystem casefolded consistent

[Dan Carpenter]

Hi Lizhi,

kernel test robot noticed the following build warnings:

https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Lizhi-Xu/ext4-check-hash-version-and-filesystem-casefolded-consistent/20240531-170046
base:   https://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4.git dev
patch link:    https://lore.kernel.org/r/20240531085647.2918240-1-lizhi.xu%40windriver.com
patch subject: [PATCH V3] ext4: check hash version and filesystem casefolded consistent
config: i386-randconfig-141-20240601 (https://download.01.org/0day-ci/archive/20240602/202406020752.Ii2MU4KP-lkp@intel.com/config)
compiler: clang version 18.1.5 (https://github.com/llvm/llvm-project 617a15a9eac96088ae5e9134248d8236e34b91b1)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
| Closes: https://lore.kernel.org/r/202406020752.Ii2MU4KP-lkp@intel.com/

smatch warnings:
fs/ext4/super.c:5287 __ext4_fill_super() warn: missing error code 'err'

vim +/err +5287 fs/ext4/super.c

d4fab7b28e2f5d7 Theodore Ts'o           2023-04-27  5280  	err = ext4_block_group_meta_init(sb, silent);
d4fab7b28e2f5d7 Theodore Ts'o           2023-04-27  5281  	if (err)
0d1ee42f27d30ee Alexandre Ratchov       2006-10-11  5282  		goto failed_mount;
0b8e58a140cae2b Andreas Dilger          2009-06-03  5283  
db9345d9e6f075e Jason Yan               2023-03-23  5284  	ext4_hash_info_init(sb);
66b3f078839bbdb Lizhi Xu                2024-05-31  5285  	if (es->s_def_hash_version == DX_HASH_SIPHASH && 
66b3f078839bbdb Lizhi Xu                2024-05-31  5286  	    !ext4_has_feature_casefold(sb))
66b3f078839bbdb Lizhi Xu                2024-05-31 @5287  		goto failed_mount;


Should this be an error path?  err = something?

ac27a0ec112a089 Dave Kleikamp           2006-10-11  5288  
d4fab7b28e2f5d7 Theodore Ts'o           2023-04-27  5289  	err = ext4_handle_clustersize(sb);
d4fab7b28e2f5d7 Theodore Ts'o           2023-04-27  5290  	if (err)
281b59959707dfa Theodore Ts'o           2011-09-09  5291  		goto failed_mount;
960fd856fdc3b08 Theodore Ts'o           2013-07-05  5292  
d4fab7b28e2f5d7 Theodore Ts'o           2023-04-27  5293  	err = ext4_check_geometry(sb, es);
d4fab7b28e2f5d7 Theodore Ts'o           2023-04-27  5294  	if (err)
bfe0a5f47ada40d Theodore Ts'o           2018-06-17  5295  		goto failed_mount;
bfe0a5f47ada40d Theodore Ts'o           2018-06-17  5296  

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

Re: [PATCH V3] ext4: check hash version and filesystem casefolded consistent

[Lizhi Xu]

On Tue, 4 Jun 2024 12:27:06 +0300, Dan Carpenter wrote: 
> kernel test robot noticed the following build warnings:
> 
> https://git-scm.com/docs/git-format-patch#_base_tree_information]
> 
> url:    https://github.com/intel-lab-lkp/linux/commits/Lizhi-Xu/ext4-check-hash-version-and-filesystem-casefolded-consistent/20240531-170046
> base:   https://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4.git dev
> patch link:    https://lore.kernel.org/r/20240531085647.2918240-1-lizhi.xu%40windriver.com
> patch subject: [PATCH V3] ext4: check hash version and filesystem casefolded consistent
> config: i386-randconfig-141-20240601 (https://download.01.org/0day-ci/archive/20240602/202406020752.Ii2MU4KP-lkp@intel.com/config)
> compiler: clang version 18.1.5 (https://github.com/llvm/llvm-project 617a15a9eac96088ae5e9134248d8236e34b91b1)
> 
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@intel.com>
> | Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
> | Closes: https://lore.kernel.org/r/202406020752.Ii2MU4KP-lkp@intel.com/
> 
> smatch warnings:
> fs/ext4/super.c:5287 __ext4_fill_super() warn: missing error code 'err'
> 
> vim +/err +5287 fs/ext4/super.c
> 
> d4fab7b28e2f5d7 Theodore Ts'o           2023-04-27  5280  	err = ext4_block_group_meta_init(sb, silent);
> d4fab7b28e2f5d7 Theodore Ts'o           2023-04-27  5281  	if (err)
> 0d1ee42f27d30ee Alexandre Ratchov       2006-10-11  5282  		goto failed_mount;
> 0b8e58a140cae2b Andreas Dilger          2009-06-03  5283
> db9345d9e6f075e Jason Yan               2023-03-23  5284  	ext4_hash_info_init(sb);
> 66b3f078839bbdb Lizhi Xu                2024-05-31  5285  	if (es->s_def_hash_version == DX_HASH_SIPHASH &&
> 66b3f078839bbdb Lizhi Xu                2024-05-31  5286  	    !ext4_has_feature_casefold(sb))
> 66b3f078839bbdb Lizhi Xu                2024-05-31 @5287  		goto failed_mount;
This warning has been resolved in the following patch: 
https://lore.kernel.org/all/20240601113749.473058-1-lizhi.xu@windriver.com/

Lizhi