Re: [PATCH v2] LSM: SafeSetID: fix UID printed instead of GID

[Paul Moore]

On Wed, May 3, 2023 at 2:44 AM Alexander Mikhalitsyn
<aleksandr.mikhalitsyn@xxxxxxxxxxxxx> wrote:
>
> pr_warn message clearly says that GID should be printed,
> but we have UID there. Let's fix that.
>
> Found accidentaly during the work on isolated user namespaces.
>
> Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@xxxxxxxxxxxxx>
> ---
> v2: __kuid_val -> __kgid_val
> ---
>  security/safesetid/lsm.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

I'm assuming you're going to pick this up Micah?

Reviewed-by: Paul Moore <paul@xxxxxxxxxxxxxx>

> diff --git a/security/safesetid/lsm.c b/security/safesetid/lsm.c
> index e806739f7868..5be5894aa0ea 100644
> --- a/security/safesetid/lsm.c
> +++ b/security/safesetid/lsm.c
> @@ -131,7 +131,7 @@ static int safesetid_security_capable(const struct cred *cred,
>                  * set*gid() (e.g. setting up userns gid mappings).
>                  */
>                 pr_warn("Operation requires CAP_SETGID, which is not available to GID %u for operations besides approved set*gid transitions\n",
> -                       __kuid_val(cred->uid));
> +                       __kgid_val(cred->gid));
>                 return -EPERM;
>         default:
>                 /* Error, the only capabilities were checking for is CAP_SETUID/GID */
> --
> 2.34.1

-- 
paul-moore.com