linux-kernel - Re: [PATCH RFC 7/15] Audit: Call only the first of the audit rule hooks

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <36f0c22554337b4d5551eabee2d86479@paul-moore.com>
Date: Tue, 14 Oct 2025 19:12:51 -0400
From: Paul Moore <paul@...l-moore.com>
To: Casey Schaufler <casey@...aufler-ca.com>, casey@...aufler-ca.com, eparis@...hat.com, linux-security-module@...r.kernel.org, audit@...r.kernel.org
Cc: jmorris@...ei.org, serge@...lyn.com, keescook@...omium.org, john.johansen@...onical.com, penguin-kernel@...ove.sakura.ne.jp, stephen.smalley.work@...il.com, linux-kernel@...r.kernel.org, selinux@...r.kernel.org
Subject: Re: [PATCH RFC 7/15] Audit: Call only the first of the audit rule  hooks

On Jun 21, 2025 Casey Schaufler <casey@...aufler-ca.com> wrote:
> 
> The audit system is not (yet) capable for distinguishing
> between audit rules specified for multiple security modules.
> Call only the first registered of the audit rule hooks.
> The order of registration, which can be specified with the
> lsm= boot parameter, is hence an important consideration.
> 
> Signed-off-by: Casey Schaufler <casey@...aufler-ca.com>
> ---
>  security/security.c | 30 ++++++++++++++++++++++++++----
>  1 file changed, 26 insertions(+), 4 deletions(-)
> 
> diff --git a/security/security.c b/security/security.c
> index 2286285f8aea..93d4ac39fe9f 100644
> --- a/security/security.c
> +++ b/security/security.c
> @@ -5056,7 +5056,13 @@ void security_key_post_create_or_update(struct key *keyring, struct key *key,
>  int security_audit_rule_init(u32 field, u32 op, char *rulestr, void **lsmrule,
>  			     gfp_t gfp)
>  {
> -	return call_int_hook(audit_rule_init, field, op, rulestr, lsmrule, gfp);
> +	struct lsm_static_call *scall;
> +
> +	lsm_for_each_hook(scall, audit_rule_init) {
> +		return scall->hl->hook.audit_rule_init(field, op, rulestr,
> +						       lsmrule, gfp);
> +	}
> +	return LSM_RET_DEFAULT(audit_rule_init);
>  }

Similar to the comments in patch 5/15, what would it look like to do the
enforcement of this callback restriction at LSM registration time?

>  /**
> @@ -5070,7 +5076,12 @@ int security_audit_rule_init(u32 field, u32 op, char *rulestr, void **lsmrule,
>   */
>  int security_audit_rule_known(struct audit_krule *krule)
>  {
> -	return call_int_hook(audit_rule_known, krule);
> +	struct lsm_static_call *scall;
> +
> +	lsm_for_each_hook(scall, audit_rule_known) {
> +		return scall->hl->hook.audit_rule_known(krule);
> +	}
> +	return LSM_RET_DEFAULT(audit_rule_known);
>  }
>  
>  /**
> @@ -5082,7 +5093,12 @@ int security_audit_rule_known(struct audit_krule *krule)
>   */
>  void security_audit_rule_free(void *lsmrule)
>  {
> -	call_void_hook(audit_rule_free, lsmrule);
> +	struct lsm_static_call *scall;
> +
> +	lsm_for_each_hook(scall, audit_rule_free) {
> +		scall->hl->hook.audit_rule_free(lsmrule);
> +		return;
> +	}
>  }
>  
>  /**
> @@ -5101,7 +5117,13 @@ void security_audit_rule_free(void *lsmrule)
>  int security_audit_rule_match(struct lsm_prop *prop, u32 field, u32 op,
>  			      void *lsmrule)
>  {
> -	return call_int_hook(audit_rule_match, prop, field, op, lsmrule);
> +	struct lsm_static_call *scall;
> +
> +	lsm_for_each_hook(scall, audit_rule_match) {
> +		return scall->hl->hook.audit_rule_match(prop, field, op,
> +							lsmrule);
> +	}
> +	return LSM_RET_DEFAULT(audit_rule_match);
>  }
>  #endif /* CONFIG_AUDIT */
>  
> -- 
> 2.47.0

--
paul-moore.com