Re: [RFC PATCH 1/3] lsm: cleanup the size counters in security_getselfattr()

From: Casey Schaufler <casey@schaufler-ca.com>
To: Paul Moore <paul@paul-moore.com>, linux-security-module@vger.kernel.org
Cc: selinux@vger.kernel.org,
	"John Johansen" <john.johansen@canonical.com>,
	"Mickaël Salaün" <mic@digikod.net>,
	"Casey Schaufler" <casey@schaufler-ca.com>
Subject: Re: [RFC PATCH 1/3] lsm: cleanup the size counters in security_getselfattr()
Date: Tue, 24 Oct 2023 15:23:07 -0700	[thread overview]
Message-ID: <8fcaab11-6340-4056-b9e0-4650be05b270@schaufler-ca.com> (raw)
In-Reply-To: <20231024213525.361332-5-paul@paul-moore.com>

On 10/24/2023 2:35 PM, Paul Moore wrote:
> Zero out all of the size counters in the -E2BIG case (buffer too
> small) to help make the current code a bit more robust in the face of
> future code changes.

I don't see how this change would have the described effect.
What it looks like it would do is change the return from -E2BIG
to 0, which would not have the desired result.

>
> Signed-off-by: Paul Moore <paul@paul-moore.com>
> ---
>  security/security.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/security/security.c b/security/security.c
> index 988483fcf153..9c63acded4ee 100644
> --- a/security/security.c
> +++ b/security/security.c
> @@ -3951,8 +3951,9 @@ int security_getselfattr(unsigned int attr, struct lsm_ctx __user *uctx,
>  			continue;
>  		}
>  		if (rc == -E2BIG) {
> -			toobig = true;
> +			rc = 0;
>  			left = 0;
> +			toobig = true;
>  		} else if (rc < 0)
>  			return rc;
>  		else