[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20180510212657.qnwm3jeqjjke77yz@madcap2.tricolour.ca>
Date: Thu, 10 May 2018 17:26:57 -0400
From: Richard Guy Briggs <rgb@...hat.com>
To: Paul Moore <paul@...l-moore.com>
Cc: Linux-Audit Mailing List <linux-audit@...hat.com>,
LKML <linux-kernel@...r.kernel.org>,
Linux NetDev Upstream Mailing List <netdev@...r.kernel.org>,
Netfilter Devel List <netfilter-devel@...r.kernel.org>,
Linux Security Module list
<linux-security-module@...r.kernel.org>,
Integrity Measurement Architecture
<linux-integrity@...r.kernel.org>,
SElinux list <selinux@...ho.nsa.gov>,
Eric Paris <eparis@...hat.com>,
Steve Grubb <sgrubb@...hat.com>,
Ingo Molnar <mingo@...hat.com>,
David Howells <dhowells@...hat.com>
Subject: Re: [PATCH ghak81 RFC V1 5/5] audit: collect audit task parameters
On 2018-05-09 11:46, Paul Moore wrote:
> On Fri, May 4, 2018 at 4:54 PM, Richard Guy Briggs <rgb@...hat.com> wrote:
> > The audit-related parameters in struct task_struct should ideally be
> > collected together and accessed through a standard audit API.
> >
> > Collect the existing loginuid, sessionid and audit_context together in a
> > new struct audit_task_info pointer called "audit" in struct task_struct.
> >
> > Use kmem_cache to manage this pool of memory.
> > Un-inline audit_free() to be able to always recover that memory.
> >
> > See: https://github.com/linux-audit/audit-kernel/issues/81
> >
> > Signed-off-by: Richard Guy Briggs <rgb@...hat.com>
> > ---
> > MAINTAINERS | 2 +-
> > include/linux/audit.h | 8 ++++----
> > include/linux/audit_task.h | 31 +++++++++++++++++++++++++++++++
> > include/linux/sched.h | 6 ++----
> > init/init_task.c | 8 ++++++--
> > kernel/auditsc.c | 4 ++--
> > 6 files changed, 46 insertions(+), 13 deletions(-)
> > create mode 100644 include/linux/audit_task.h
> >
> > diff --git a/MAINTAINERS b/MAINTAINERS
> > index 0a1410d..8c7992d 100644
> > --- a/MAINTAINERS
> > +++ b/MAINTAINERS
> > @@ -2510,7 +2510,7 @@ L: linux-audit@...hat.com (moderated for non-subscribers)
> > W: https://github.com/linux-audit
> > T: git git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/audit.git
> > S: Supported
> > -F: include/linux/audit.h
> > +F: include/linux/audit*.h
> > F: include/uapi/linux/audit.h
> > F: kernel/audit*
> >
> > diff --git a/include/linux/audit.h b/include/linux/audit.h
> > index dba0d45..1324969 100644
> > --- a/include/linux/audit.h
> > +++ b/include/linux/audit.h
> > @@ -237,11 +237,11 @@ extern void __audit_inode_child(struct inode *parent,
> >
> > static inline void audit_set_context(struct task_struct *task, struct audit_context *ctx)
> > {
> > - task->audit_context = ctx;
> > + task->audit.ctx = ctx;
> > }
> > static inline struct audit_context *audit_context(struct task_struct *task)
> > {
> > - return task->audit_context;
> > + return task->audit.ctx;
> > }
> > static inline bool audit_dummy_context(void)
> > {
> > @@ -330,12 +330,12 @@ extern int auditsc_get_stamp(struct audit_context *ctx,
> >
> > static inline kuid_t audit_get_loginuid(struct task_struct *tsk)
> > {
> > - return tsk->loginuid;
> > + return tsk->audit.loginuid;
> > }
> >
> > static inline unsigned int audit_get_sessionid(struct task_struct *tsk)
> > {
> > - return tsk->sessionid;
> > + return tsk->audit.sessionid;
> > }
> >
> > extern void __audit_ipc_obj(struct kern_ipc_perm *ipcp);
> > diff --git a/include/linux/audit_task.h b/include/linux/audit_task.h
> > new file mode 100644
> > index 0000000..d4b3a20
> > --- /dev/null
> > +++ b/include/linux/audit_task.h
> > @@ -0,0 +1,31 @@
> > +/* SPDX-License-Identifier: GPL-2.0 */
> > +/* audit_task.h -- definition of audit_task_info structure
> > + *
> > + * Copyright 2018 Red Hat Inc., Raleigh, North Carolina.
> > + * All Rights Reserved.
> > + *
> > + * This program is free software; you can redistribute it and/or modify
> > + * it under the terms of the GNU General Public License as published by
> > + * the Free Software Foundation; either version 2 of the License, or
> > + * (at your option) any later version.
> > + *
> > + * This program is distributed in the hope that it will be useful,
> > + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> > + * GNU General Public License for more details.
> > + *
> > + * Written by Richard Guy Briggs <rgb@...hat.com>
> > + *
> > + */
> > +
> > +#ifndef _LINUX_AUDIT_TASK_H_
> > +#define _LINUX_AUDIT_TASK_H_
> > +
> > +struct audit_context;
> > +struct audit_task_info {
> > + kuid_t loginuid;
> > + unsigned int sessionid;
> > + struct audit_context *ctx;
> > +};
> > +
> > +#endif
> > diff --git a/include/linux/sched.h b/include/linux/sched.h
> > index b3d697f..b58eca0 100644
> > --- a/include/linux/sched.h
> > +++ b/include/linux/sched.h
> > @@ -27,9 +27,9 @@
> > #include <linux/signal_types.h>
> > #include <linux/mm_types_task.h>
> > #include <linux/task_io_accounting.h>
> > +#include <linux/audit_task.h>
> >
> > /* task_struct member predeclarations (sorted alphabetically): */
> > -struct audit_context;
> > struct backing_dev_info;
> > struct bio_list;
> > struct blk_plug;
> > @@ -832,10 +832,8 @@ struct task_struct {
> >
> > struct callback_head *task_works;
> >
> > - struct audit_context *audit_context;
> > #ifdef CONFIG_AUDITSYSCALL
> > - kuid_t loginuid;
> > - unsigned int sessionid;
> > + struct audit_task_info audit;
> > #endif
>
> Considering that the audit_context pointer is now in the
> audit_task_info struct, should the audit_task_info struct be placed
> outside the CONFIG_AUDITSYSCALL protections? Or rather, shouldn't the
> CONFIG_AUDITSYSCALL protections be moved inside audit_task_info or
> removed entirely?
Well, I wondered about that anyways. audit_context is only meaningful
in CONFIG_AUDIT_SYSCALL, and loginuid and sessionid were already there,
so the whole thing should be inside, but given that CONFIG_AUDIT_SYSCALL
is forced on when CONFIG_AUDIT is set I don't see that it matters.
Perhaps CONFIG_AUDIT_SYSCALL should be ripped out completely and the
code flattenned to the CONFIG_AUDIT case.
I see your point though, moving CONFIG_AUDIT_SYSCALL protections to
within the audit_task_info struct definition makes more sense than this
above.
> > diff --git a/init/init_task.c b/init/init_task.c
> > index c788f91..d33260d 100644
> > --- a/init/init_task.c
> > +++ b/init/init_task.c
> > @@ -9,6 +9,7 @@
> > #include <linux/init.h>
> > #include <linux/fs.h>
> > #include <linux/mm.h>
> > +#include <linux/audit.h>
> >
> > #include <asm/pgtable.h>
> > #include <linux/uaccess.h>
> > @@ -118,8 +119,11 @@ struct task_struct init_task
> > .thread_group = LIST_HEAD_INIT(init_task.thread_group),
> > .thread_node = LIST_HEAD_INIT(init_signals.thread_head),
> > #ifdef CONFIG_AUDITSYSCALL
> > - .loginuid = INVALID_UID,
> > - .sessionid = AUDIT_SID_UNSET,
> > + .audit = {
> > + .loginuid = INVALID_UID,
> > + .sessionid = AUDIT_SID_UNSET,
> > + .ctx = NULL,
> > + },
> > #endif
> > #ifdef CONFIG_PERF_EVENTS
> > .perf_event_mutex = __MUTEX_INITIALIZER(init_task.perf_event_mutex),
> > diff --git a/kernel/auditsc.c b/kernel/auditsc.c
> > index f294e4a..b5d8bff 100644
> > --- a/kernel/auditsc.c
> > +++ b/kernel/auditsc.c
> > @@ -2068,8 +2068,8 @@ int audit_set_loginuid(kuid_t loginuid)
> > sessionid = (unsigned int)atomic_inc_return(&session_id);
> > }
> >
> > - task->sessionid = sessionid;
> > - task->loginuid = loginuid;
> > + task->audit.sessionid = sessionid;
> > + task->audit.loginuid = loginuid;
> > out:
> > audit_log_set_loginuid(oldloginuid, loginuid, oldsessionid, sessionid, rc);
> > return rc;
> > --
> > 1.8.3.1
>
> --
> paul moore
> www.paul-moore.com
- RGB
--
Richard Guy Briggs <rgb@...hat.com>
Sr. S/W Engineer, Kernel Security, Base Operating Systems
Remote, Ottawa, Red Hat Canada
IRC: rgb, SunRaycer
Voice: +1.647.777.2635, Internal: (81) 32635
Powered by blists - more mailing lists