diff mbox series

evm: Change vfs_getxattr() with __vfs_getxattr() in evm_calc_hmac_or_hash()

Message ID 20240307122240.3560688-1-roberto.sassu@huaweicloud.com (mailing list archive)
State Handled Elsewhere
Headers show
Series evm: Change vfs_getxattr() with __vfs_getxattr() in evm_calc_hmac_or_hash() | expand

Commit Message

Roberto Sassu March 7, 2024, 12:22 p.m. UTC
From: Roberto Sassu <roberto.sassu@huawei.com>

Use __vfs_getxattr() instead of vfs_getxattr(), in preparation for
deprecating using the vfs_ interfaces for retrieving fscaps.

__vfs_getxattr() is only used for debugging purposes, to check if kernel
space and user space see the same xattr value.

Cc: stable@vger.kernel.org # 5.14.x
Cc: linux-fsdevel@vger.kernel.org
Cc: Christian Brauner <brauner@kernel.org>
Cc: Seth Forshee (DigitalOcean) <sforshee@kernel.org>
Fixes: 907a399de7b0 ("evm: Check xattr size discrepancy between kernel and user")
Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
---
 security/integrity/evm/evm_crypto.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Comments

Seth Forshee March 7, 2024, 2:31 p.m. UTC | #1
On Thu, Mar 07, 2024 at 01:22:39PM +0100, Roberto Sassu wrote:
> From: Roberto Sassu <roberto.sassu@huawei.com>
> 
> Use __vfs_getxattr() instead of vfs_getxattr(), in preparation for
> deprecating using the vfs_ interfaces for retrieving fscaps.
> 
> __vfs_getxattr() is only used for debugging purposes, to check if kernel
> space and user space see the same xattr value.

__vfs_getxattr() won't give you the value as seen by userspace though.
Userspace goes through vfs_getxattr() -> xattr_getsecurity() ->
cap_inode_getsecurity(), which does the conversion to the value
userspace sees. __vfs_getxattr() just gives the raw disk data.

I'm also currently working on changes to my fscaps series that will make
it so that __vfs_getxattr() also cannot be used to read fscaps xattrs.
I'll fix this and other code in EVM which will be broken by that change
as part of the next version too.

> 
> Cc: stable@vger.kernel.org # 5.14.x
> Cc: linux-fsdevel@vger.kernel.org
> Cc: Christian Brauner <brauner@kernel.org>
> Cc: Seth Forshee (DigitalOcean) <sforshee@kernel.org>
> Fixes: 907a399de7b0 ("evm: Check xattr size discrepancy between kernel and user")
> Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
> ---
>  security/integrity/evm/evm_crypto.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/security/integrity/evm/evm_crypto.c b/security/integrity/evm/evm_crypto.c
> index b1ffd4cc0b44..168d98c63513 100644
> --- a/security/integrity/evm/evm_crypto.c
> +++ b/security/integrity/evm/evm_crypto.c
> @@ -278,8 +278,8 @@ static int evm_calc_hmac_or_hash(struct dentry *dentry,
>  		if (size < 0)
>  			continue;
>  
> -		user_space_size = vfs_getxattr(&nop_mnt_idmap, dentry,
> -					       xattr->name, NULL, 0);
> +		user_space_size = __vfs_getxattr(dentry, inode, xattr->name,
> +						 NULL, 0);
>  		if (user_space_size != size)
>  			pr_debug("file %s: xattr %s size mismatch (kernel: %d, user: %d)\n",
>  				 dentry->d_name.name, xattr->name, size,
> -- 
> 2.34.1
>
Roberto Sassu March 7, 2024, 2:36 p.m. UTC | #2
On Thu, 2024-03-07 at 08:31 -0600, Seth Forshee wrote:
> On Thu, Mar 07, 2024 at 01:22:39PM +0100, Roberto Sassu wrote:
> > From: Roberto Sassu <roberto.sassu@huawei.com>
> > 
> > Use __vfs_getxattr() instead of vfs_getxattr(), in preparation for
> > deprecating using the vfs_ interfaces for retrieving fscaps.
> > 
> > __vfs_getxattr() is only used for debugging purposes, to check if kernel
> > space and user space see the same xattr value.
> 
> __vfs_getxattr() won't give you the value as seen by userspace though.
> Userspace goes through vfs_getxattr() -> xattr_getsecurity() ->
> cap_inode_getsecurity(), which does the conversion to the value
> userspace sees. __vfs_getxattr() just gives the raw disk data.
> 
> I'm also currently working on changes to my fscaps series that will make
> it so that __vfs_getxattr() also cannot be used to read fscaps xattrs.
> I'll fix this and other code in EVM which will be broken by that change
> as part of the next version too.

You are right, thank you!

Roberto

> > 
> > Cc: stable@vger.kernel.org # 5.14.x
> > Cc: linux-fsdevel@vger.kernel.org
> > Cc: Christian Brauner <brauner@kernel.org>
> > Cc: Seth Forshee (DigitalOcean) <sforshee@kernel.org>
> > Fixes: 907a399de7b0 ("evm: Check xattr size discrepancy between kernel and user")
> > Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
> > ---
> >  security/integrity/evm/evm_crypto.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/security/integrity/evm/evm_crypto.c b/security/integrity/evm/evm_crypto.c
> > index b1ffd4cc0b44..168d98c63513 100644
> > --- a/security/integrity/evm/evm_crypto.c
> > +++ b/security/integrity/evm/evm_crypto.c
> > @@ -278,8 +278,8 @@ static int evm_calc_hmac_or_hash(struct dentry *dentry,
> >  		if (size < 0)
> >  			continue;
> >  
> > -		user_space_size = vfs_getxattr(&nop_mnt_idmap, dentry,
> > -					       xattr->name, NULL, 0);
> > +		user_space_size = __vfs_getxattr(dentry, inode, xattr->name,
> > +						 NULL, 0);
> >  		if (user_space_size != size)
> >  			pr_debug("file %s: xattr %s size mismatch (kernel: %d, user: %d)\n",
> >  				 dentry->d_name.name, xattr->name, size,
> > -- 
> > 2.34.1
> >
Roberto Sassu March 7, 2024, 3:02 p.m. UTC | #3
On Thu, 2024-03-07 at 15:36 +0100, Roberto Sassu wrote:
> On Thu, 2024-03-07 at 08:31 -0600, Seth Forshee wrote:
> > On Thu, Mar 07, 2024 at 01:22:39PM +0100, Roberto Sassu wrote:
> > > From: Roberto Sassu <roberto.sassu@huawei.com>
> > > 
> > > Use __vfs_getxattr() instead of vfs_getxattr(), in preparation for
> > > deprecating using the vfs_ interfaces for retrieving fscaps.
> > > 
> > > __vfs_getxattr() is only used for debugging purposes, to check if kernel
> > > space and user space see the same xattr value.
> > 
> > __vfs_getxattr() won't give you the value as seen by userspace though.
> > Userspace goes through vfs_getxattr() -> xattr_getsecurity() ->
> > cap_inode_getsecurity(), which does the conversion to the value
> > userspace sees. __vfs_getxattr() just gives the raw disk data.
> > 
> > I'm also currently working on changes to my fscaps series that will make
> > it so that __vfs_getxattr() also cannot be used to read fscaps xattrs.
> > I'll fix this and other code in EVM which will be broken by that change
> > as part of the next version too.
> 
> You are right, thank you!

(Apologies, I should have been more careful).

Roberto

> Roberto
> 
> > > 
> > > Cc: stable@vger.kernel.org # 5.14.x
> > > Cc: linux-fsdevel@vger.kernel.org
> > > Cc: Christian Brauner <brauner@kernel.org>
> > > Cc: Seth Forshee (DigitalOcean) <sforshee@kernel.org>
> > > Fixes: 907a399de7b0 ("evm: Check xattr size discrepancy between kernel and user")
> > > Signed-off-by: Roberto Sassu <roberto.sassu@huawei.com>
> > > ---
> > >  security/integrity/evm/evm_crypto.c | 4 ++--
> > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/security/integrity/evm/evm_crypto.c b/security/integrity/evm/evm_crypto.c
> > > index b1ffd4cc0b44..168d98c63513 100644
> > > --- a/security/integrity/evm/evm_crypto.c
> > > +++ b/security/integrity/evm/evm_crypto.c
> > > @@ -278,8 +278,8 @@ static int evm_calc_hmac_or_hash(struct dentry *dentry,
> > >  		if (size < 0)
> > >  			continue;
> > >  
> > > -		user_space_size = vfs_getxattr(&nop_mnt_idmap, dentry,
> > > -					       xattr->name, NULL, 0);
> > > +		user_space_size = __vfs_getxattr(dentry, inode, xattr->name,
> > > +						 NULL, 0);
> > >  		if (user_space_size != size)
> > >  			pr_debug("file %s: xattr %s size mismatch (kernel: %d, user: %d)\n",
> > >  				 dentry->d_name.name, xattr->name, size,
> > > -- 
> > > 2.34.1
> > >
diff mbox series

Patch

diff --git a/security/integrity/evm/evm_crypto.c b/security/integrity/evm/evm_crypto.c
index b1ffd4cc0b44..168d98c63513 100644
--- a/security/integrity/evm/evm_crypto.c
+++ b/security/integrity/evm/evm_crypto.c
@@ -278,8 +278,8 @@  static int evm_calc_hmac_or_hash(struct dentry *dentry,
 		if (size < 0)
 			continue;
 
-		user_space_size = vfs_getxattr(&nop_mnt_idmap, dentry,
-					       xattr->name, NULL, 0);
+		user_space_size = __vfs_getxattr(dentry, inode, xattr->name,
+						 NULL, 0);
 		if (user_space_size != size)
 			pr_debug("file %s: xattr %s size mismatch (kernel: %d, user: %d)\n",
 				 dentry->d_name.name, xattr->name, size,