Re: [RFC PATCH 1/2] fs/xattr: add *at family syscalls

[Christian Brauner]

On Tue, Aug 30, 2022 at 05:28:39PM +0200, Christian Göttsche wrote:
> Add the four syscalls setxattrat(), getxattrat(), listxattrat() and
> removexattrat() to enable extended attribute operations via file
> descriptors.  This can be used from userspace to avoid race conditions,
> especially on security related extended attributes, like SELinux labels
> ("security.selinux") via setfiles(8).
> 
> Use the do_{name}at() pattern from fs/open.c.
> Use a single flag parameter for extended attribute flags (currently
> XATTR_CREATE and XATTR_REPLACE) and *at() flags to not exceed six
> syscall arguments in setxattrat().
> 
> Previous discussion ("f*xattr: allow O_PATH descriptors"): https://lore.kernel.org/all/20220607153139.35588-1-cgzones@xxxxxxxxxxxxxx/
> 
> Signed-off-by: Christian Göttsche <cgzones@xxxxxxxxxxxxxx>
> ---

Having a way to operate via file descriptors on xattrs does make a lot
of sense to me in general. We've got code that changes ownership
recursively including using llistxattr(), getxattr(), and setxattr() any
xattrs that store ownership information and being able to passing in an
fd directly would be quite neat...

>  fs/xattr.c | 108 +++++++++++++++++++++++++++++++++++++++++------------
>  1 file changed, 85 insertions(+), 23 deletions(-)
> 
> diff --git a/fs/xattr.c b/fs/xattr.c
> index a1f4998bc6be..a4738e28be8c 100644
> --- a/fs/xattr.c
> +++ b/fs/xattr.c
> @@ -27,6 +27,8 @@
>  
>  #include "internal.h"
>  
> +#define XATTR__FLAGS (XATTR_CREATE | XATTR_REPLACE)
> +
>  static const char *
>  strcmp_prefix(const char *a, const char *a_prefix)
>  {
> @@ -559,7 +561,7 @@ int setxattr_copy(const char __user *name, struct xattr_ctx *ctx)
>  {
>  	int error;
>  
> -	if (ctx->flags & ~(XATTR_CREATE|XATTR_REPLACE))
> +	if (ctx->flags & ~XATTR__FLAGS)
>  		return -EINVAL;
>  
>  	error = strncpy_from_user(ctx->kname->name, name,
> @@ -626,21 +628,31 @@ setxattr(struct user_namespace *mnt_userns, struct dentry *d,
>  	return error;
>  }
>  
> -static int path_setxattr(const char __user *pathname,
> +static int do_setxattrat(int dfd, const char __user *pathname,
>  			 const char __user *name, const void __user *value,
> -			 size_t size, int flags, unsigned int lookup_flags)
> +			 size_t size, int flags)
>  {
>  	struct path path;
>  	int error;
> +	int lookup_flags;
> +
> +	/* AT_ and XATTR_ flags must not overlap. */
> +	BUILD_BUG_ON(((AT_SYMLINK_NOFOLLOW | AT_EMPTY_PATH) & XATTR__FLAGS) != 0);
> +
> +	if ((flags & ~(AT_SYMLINK_NOFOLLOW | AT_EMPTY_PATH | XATTR__FLAGS)) != 0)
> +		return -EINVAL;

It's a bit tasteless that we end up mixing AT_* and XATTR_* flags for sure.

>  
> +	lookup_flags = (flags & AT_SYMLINK_NOFOLLOW) ? 0 : LOOKUP_FOLLOW;
> +	if (flags & AT_EMPTY_PATH)
> +		lookup_flags |= LOOKUP_EMPTY;
>  retry:
> -	error = user_path_at(AT_FDCWD, pathname, lookup_flags, &path);
> +	error = user_path_at(dfd, pathname, lookup_flags, &path);
>  	if (error)
>  		return error;
>  	error = mnt_want_write(path.mnt);
>  	if (!error) {
>  		error = setxattr(mnt_user_ns(path.mnt), path.dentry, name,
> -				 value, size, flags);
> +				 value, size, flags & XATTR__FLAGS);
>  		mnt_drop_write(path.mnt);
>  	}
>  	path_put(&path);
> @@ -651,18 +663,25 @@ static int path_setxattr(const char __user *pathname,
>  	return error;
>  }
>  
> +SYSCALL_DEFINE6(setxattrat, int, dfd, const char __user *, pathname,
> +		const char __user *, name, const void __user *, value,
> +		size_t, size, int, flags)
> +{
> +	return do_setxattrat(dfd, pathname, name, value, size, flags);
> +}

I'm not sure we need setxattrat()? Yes, it doesn't have a "pathname"
argument but imho it's not that big of a deal to first perform the
lookup via openat*() and then call fsetxattr() and have fsetxattr()
recognize AT_* flags. It's not that the xattr system calls are
lightweight in the first place.

But maybe the consistency is nicer. I have no strong feelings here.