[apparmor] [PATCH] apparmor: Fix profile conflict logic

[John Johansen]

On 01/11/2018 01:07 PM, Matthew Garrett wrote:
> The intended behaviour in apparmor profile matching is to flag a
> conflict if two profiles match equally well. However, right now a
> conflict is generated if another profile has the same match length even
> if that profile doesn't actually match. Fix the logic so we only
> generate a conflict if the profiles match.
> 
> Signed-off-by: Matthew Garrett <mjg59 at google.com>

Acked-by: John Johansne <john.johansen at caonical.com>

I'll get a pull request together asap

> ---
>  security/apparmor/domain.c | 9 +++++----
>  1 file changed, 5 insertions(+), 4 deletions(-)
> 
> diff --git a/security/apparmor/domain.c b/security/apparmor/domain.c
> index 04ba9d0718ea..6a54d2ffa840 100644
> --- a/security/apparmor/domain.c
> +++ b/security/apparmor/domain.c
> @@ -330,10 +330,7 @@ static struct aa_profile *__attach_match(const char *name,
>  			continue;
>  
>  		if (profile->xmatch) {
> -			if (profile->xmatch_len == len) {
> -				conflict = true;
> -				continue;
> -			} else if (profile->xmatch_len > len) {
> +			if (profile->xmatch_len >= len) {
>  				unsigned int state;
>  				u32 perm;
>  
> @@ -342,6 +339,10 @@ static struct aa_profile *__attach_match(const char *name,
>  				perm = dfa_user_allow(profile->xmatch, state);
>  				/* any accepting state means a valid match. */
>  				if (perm & MAY_EXEC) {
> +					if (profile->xmatch_len == len) {
> +						conflict = true;
> +						continue;
> +					}
>  					candidate = profile;
>  					len = profile->xmatch_len;
>  					conflict = false;
>