Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <cover.1529003588.git.rgb@redhat.com> <65299efddc04e19e6ad4b06fbde0bb3db2643cb5.1529003588.git.rgb@redhat.com>
In-Reply-To: <65299efddc04e19e6ad4b06fbde0bb3db2643cb5.1529003588.git.rgb@redhat.com>
From:   Paul Moore <paul@paul-moore.com>
Date:   Thu, 28 Jun 2018 18:25:25 -0400
Message-ID: <CAHC9VhTffdnaymGk0LiCqXQfDK_7TW-_wxVTNuT9deS5SwsNBA@mail.gmail.com>
Subject: Re: [RFC PATCH ghak59 V1 5/6] audit: move EOE record after kill_trees
 for exit/free
To:     rgb@redhat.com
Cc:     linux-audit@redhat.com, linux-kernel@vger.kernel.org,
        Eric Paris <eparis@parisplace.org>, sgrubb@redhat.com,
        aviro@redhat.com
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Jun 14, 2018 at 4:23 PM Richard Guy Briggs <rgb@redhat.com> wrote:
> The EOE record was being issued prior to the pruning of the killed_tree
> list.
>
> Move the EOE record creation out of audit_log_exit() and into its
> callers __audit_free() and __audit_syscall_exit() so that
> audit_kill_trees() can be called prior to the EOE record creation and
> any purged trees CONFIG_CHANGE records included in the syscall record
> event.
>
> See: https://github.com/linux-audit/audit-kernel/issues/50
> See: https://github.com/linux-audit/audit-kernel/issues/59
> Signed-off-by: Richard Guy Briggs <rgb@redhat.com>
> ---
>  kernel/auditsc.c | 24 +++++++++++++++++-------
>  1 file changed, 17 insertions(+), 7 deletions(-)

See my comments in 4/6.  Assuming we are able to shuffle the ordering
of audit_log_exit() and audit_kill_trees() this patch would no longer
be needed, yes?

> diff --git a/kernel/auditsc.c b/kernel/auditsc.c
> index 2590c9e..d56aead 100644
> --- a/kernel/auditsc.c
> +++ b/kernel/auditsc.c
> @@ -1460,10 +1460,6 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts
>
>         audit_log_proctitle(tsk, context);
>
> -       /* Send end of event record to help user space know we are finished */
> -       ab = audit_log_start(context, GFP_KERNEL, AUDIT_EOE);
> -       if (ab)
> -               audit_log_end(ab);
>         if (call_panic)
>                 audit_panic("error converting sid to string");
>  }
> @@ -1491,6 +1487,14 @@ void __audit_free(struct task_struct *tsk)
>                 audit_log_exit(context, tsk);
>         if (!list_empty(&context->killed_trees))
>                 audit_kill_trees(context);
> +       if (context->in_syscall && context->current_state == AUDIT_RECORD_CONTEXT) {
> +               struct audit_buffer *ab;
> +
> +               /* Send end of event record to help user space know we are finished */
> +               ab = audit_log_start(context, GFP_KERNEL, AUDIT_EOE);
> +               if (ab)
> +                       audit_log_end(ab);
> +       }
>
>         audit_free_context(context);
>  }
> @@ -1572,13 +1576,19 @@ void __audit_syscall_exit(int success, long return_code)
>
>         if (context->in_syscall && context->current_state == AUDIT_RECORD_CONTEXT)
>                 audit_log_exit(context, current);
> +       if (!list_empty(&context->killed_trees))
> +               audit_kill_trees(context);
> +       if (context->in_syscall && context->current_state == AUDIT_RECORD_CONTEXT) {
> +               struct audit_buffer *ab;
>
> +               /* Send end of event record to help user space know we are finished */
> +               ab = audit_log_start(context, GFP_KERNEL, AUDIT_EOE);
> +               if (ab)
> +                       audit_log_end(ab);
> +       }
>         context->in_syscall = 0;
>         context->prio = context->state == AUDIT_RECORD_CONTEXT ? ~0ULL : 0;
>
> -       if (!list_empty(&context->killed_trees))
> -               audit_kill_trees(context);
> -
>         audit_free_names(context);
>         unroll_tree_refs(context, NULL, 0);
>         audit_free_aux(context);
> --
> 1.8.3.1
>


-- 
paul moore
www.paul-moore.com