Skip to contentSkip to conversation

Reworked patch for labels on user space messages

PATCH In review Al Viro 4 participants 9 messages May 1, 2006 – May 3, 2006

Subsystems (3)
lore
Share & tools
lore

Al Viro

author
#0 lore raw

PATCH 11/14 Reworked patch for labels on user space messages

From: Steve Grubb <[email protected]>
Date: Mon Apr 3 09:08:13 2006 -0400

The below patch should be applied after the inode and ipc sid patches. This patch is a reworking of Tim's patch that has been updated to match the inode and ipc patches since its similar.

[updated:

Quoted 2 lines

Stephen Smalley also wanted to change a variable from isec to tsec in the
user sid patch. ]

---

 include/linux/netlink.h    |    1 +
 include/linux/selinux.h    |   16 ++++++++++++++++
 kernel/audit.c             |   22 +++++++++++++++++++---
 net/netlink/af_netlink.c   |    2 ++
 security/selinux/exports.c |   11 +++++++++++
 5 files changed, 49 insertions(+), 3 deletions(-)

e7c3497013a7e5496ce3d5fd3c73b5cf5af7a56e

include/linux/netlink.h +1 -0
index f8f3d1c..87b8a57 100644
@@ -143,6 +143,7 @@ struct netlink_skb_parms
 	__u32			dst_group;
 	kernel_cap_t		eff_cap;
 	__u32			loginuid;	/* Login (audit) uid */
+	__u32			sid;		/* SELinux security id */
 };
 
 #define NETLINK_CB(skb)		(*(struct netlink_skb_parms*)&((skb)->cb))
include/linux/selinux.h +16 -0
index 413d667..4047bcd 100644
@@ -5,6 +5,7 @@
  *
  * Copyright (C) 2005 Red Hat, Inc., James Morris <[email protected]>
  * Copyright (C) 2006 Trusted Computer Solutions, Inc. <[email protected]>
+ * Copyright (C) 2006 IBM Corporation, Timothy R. Chavez <[email protected]>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2,
@@ -108,6 +109,16 @@ void selinux_get_inode_sid(const struct 
  */
 void selinux_get_ipc_sid(const struct kern_ipc_perm *ipcp, u32 *sid);
 
+/**
+ *     selinux_get_task_sid - return the SID of task
+ *     @tsk: the task whose SID will be returned
+ *     @sid: pointer to security context ID to be filled in.
+ *
+ *     Returns nothing
+ */
+void selinux_get_task_sid(struct task_struct *tsk, u32 *sid);
+
+
 #else
 
 static inline int selinux_audit_rule_init(u32 field, u32 op,
@@ -156,6 +167,11 @@ static inline void selinux_get_ipc_sid(c
 	*sid = 0;
 }
 
+static inline void selinux_get_task_sid(struct task_struct *tsk, u32 *sid)
+{
+	*sid = 0;
+}
+
 #endif	/* CONFIG_SECURITY_SELINUX */
 
 #endif /* _LINUX_SELINUX_H */
kernel/audit.c +19 -3
index 9060be7..7ec9cca 100644
@@ -390,7 +390,7 @@ static int audit_netlink_ok(kernel_cap_t
 
 static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 {
-	u32			uid, pid, seq;
+	u32			uid, pid, seq, sid;
 	void			*data;
 	struct audit_status	*status_get, status_set;
 	int			err;
@@ -416,6 +416,7 @@ static int audit_receive_msg(struct sk_b
 	pid  = NETLINK_CREDS(skb)->pid;
 	uid  = NETLINK_CREDS(skb)->uid;
 	loginuid = NETLINK_CB(skb).loginuid;
+	sid  = NETLINK_CB(skb).sid;
 	seq  = nlh->nlmsg_seq;
 	data = NLMSG_DATA(nlh);
 
@@ -468,8 +469,23 @@ static int audit_receive_msg(struct sk_b
 			ab = audit_log_start(NULL, GFP_KERNEL, msg_type);
 			if (ab) {
 				audit_log_format(ab,
-						 "user pid=%d uid=%u auid=%u msg='%.1024s'",
-						 pid, uid, loginuid, (char *)data);
+						 "user pid=%d uid=%u auid=%u",
+						 pid, uid, loginuid);
+				if (sid) {
+					char *ctx = NULL;
+					u32 len;
+					if (selinux_ctxid_to_string(
+							sid, &ctx, &len)) {
+						audit_log_format(ab, 
+							" subj=%u", sid);
+						/* Maybe call audit_panic? */
+					} else
+						audit_log_format(ab, 
+							" subj=%s", ctx);
+					kfree(ctx);
+				}
+				audit_log_format(ab, " msg='%.1024s'",
+					 (char *)data);
 				audit_set_pid(ab, pid);
 				audit_log_end(ab);
 			}
net/netlink/af_netlink.c +2 -0
index 2a233ff..09fbc4b 100644
@@ -56,6 +56,7 @@ #include <linux/bitops.h>
 #include <linux/mm.h>
 #include <linux/types.h>
 #include <linux/audit.h>
+#include <linux/selinux.h>
 
 #include <net/sock.h>
 #include <net/scm.h>
@@ -1157,6 +1158,7 @@ static int netlink_sendmsg(struct kiocb 
 	NETLINK_CB(skb).dst_pid = dst_pid;
 	NETLINK_CB(skb).dst_group = dst_group;
 	NETLINK_CB(skb).loginuid = audit_get_loginuid(current->audit_context);
+	selinux_get_task_sid(current, &(NETLINK_CB(skb).sid));
 	memcpy(NETLINK_CREDS(skb), &siocb->scm->creds, sizeof(struct ucred));
 
 	/* What can I do? Netlink is asynchronous, so that
security/selinux/exports.c +11 -0
index 7357cf2..ae4c73e 100644
@@ -5,6 +5,7 @@
  *
  * Copyright (C) 2005 Red Hat, Inc., James Morris <[email protected]>
  * Copyright (C) 2006 Trusted Computer Solutions, Inc. <[email protected]>
+ * Copyright (C) 2006 IBM Corporation, Timothy R. Chavez <[email protected]>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2,
@@ -61,3 +62,13 @@ void selinux_get_ipc_sid(const struct ke
 	*sid = 0;
 }
 
+void selinux_get_task_sid(struct task_struct *tsk, u32 *sid)
+{
+	if (selinux_enabled) {
+		struct task_security_struct *tsec = tsk->security;
+		*sid = tsec->sid;
+		return;
+	}
+	*sid = 0;
+}
+
Signature · 1 line
1.3.0.g0080f
#1 lore raw

Something seems to be wrong in selinux_get_task_sid. I am getting thousands of these and can't boot the kernel.

May  3 08:51:53 jonsmirl kernel: Code: 00 00 c3 83 3d 60 c3 32 c0 00
74 09 8b 40 24 8b 40 08 89 02 c3 c7 02 00 00 00 00 c3 83 3d 60 c3 32
c0 00 74 0c 8b 80 90 04 00 00 <8b> 40 08 89 02 c3 c7 02 00 00 00 00 c3
83 3d 60 c3 32 c0 00 74
May  3 08:51:53 jonsmirl kernel: EIP: [<c01b3918>]
selinux_get_task_sid+0xf/0x1c SS:ESP 0068:e9b5cd9c
May  3 08:51:53 jonsmirl kernel:  <1>BUG: unable to handle kernel NULL
pointer dereference at virtual address 00000008
May  3 08:51:53 jonsmirl kernel:  printing eip:
May  3 08:51:53 jonsmirl kernel: c01b3918
May  3 08:51:53 jonsmirl kernel: *pde = 00000000
May  3 08:51:53 jonsmirl kernel: Oops: 0000 [#25]
May  3 08:51:53 jonsmirl kernel: SMP
May  3 08:51:53 jonsmirl kernel: Modules linked in: af_packet

xt_length ipt_ttl xt_tcpmss ipt_TCPMSS iptable_mangle xt_multiport xt_limit ipt_tos nfsd exportfs lockd sunrpc ipv6 autofs4 snd_usb_audio snd_seq_oss snd_seq_midi_event snd_seq snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd_page_alloc snd_usb_lib snd_rawmidi snd_seq_device snd_hwdep snd soundcore ip_conntrack_netbios_ns ipt_REJECT xt_state ip_conntrack nfnetlink xt_tcpudp iptable_filter ip_tables x_tables nls_iso8859_1 nls_cp437 vfat fat evdev usbhid video thermal processor fan button battery ac lp parport_pc parport floppy ohci1394 ieee1394 uhci_hcd ehci_hcd e1000 i2c_i801 i2c_core i82875p_edac hw_random edac_mc rtc usbcore unix

May  3 08:51:53 jonsmirl kernel: CPU:    0
May  3 08:51:53 jonsmirl kernel: EIP:    0060:[<c01b3918>]    Not tainted VLI
May  3 08:51:53 jonsmirl kernel: EFLAGS: 00010202   (2.6.17-rc3 #122)
May  3 08:51:53 jonsmirl kernel: EIP is at selinux_get_task_sid+0xf/0x1c
May  3 08:51:53 jonsmirl kernel: eax: 00000000   ebx: e944e000   ecx:
00000000   edx: f7bb9428
May  3 08:51:53 jonsmirl kernel: esi: e944eebc   edi: f7bb9408   ebp:
f7bb93d8   esp: e944ed9c
May  3 08:51:53 jonsmirl kernel: ds: 007b   es: 007b   ss: 0068
May  3 08:51:53 jonsmirl kernel: Process gdm-binary (pid: 3159,
threadinfo=e944e000 task=f7d9e580)
May  3 08:51:53 jonsmirl kernel: Stack: <0>c02622c8 00000004 e944ee46
e944ef3c c17fa8ac 00000000 00000000 e944ee60
May  3 08:51:53 jonsmirl kernel:        00000c57 00000000 00000000
00000000 00000000 c029fda0 e9b50904 00000078
May  3 08:51:53 jonsmirl kernel:        e944ef3c c024db1c 00000078
c01627af e944ee58 00000000 00000001 ffffffff
May  3 08:51:53 jonsmirl kernel: Call Trace:
May  3 08:51:53 jonsmirl kernel:  <c02622c8>
netlink_sendmsg+0x19f/0x280   <c024db1c> sock_sendmsg+0xd4/0xef
May  3 08:51:53 jonsmirl kernel:  <c01627af> __d_lookup+0x96/0xd5  <c0127b5c> autoremove_wake_function+0x0/0x35
May  3 08:51:53 jonsmirl kernel:  <c015bfb5>
__link_path_walk+0xbab/0xce8   <c0159ed6> do_lookup+0x4f/0x135
May  3 08:51:53 jonsmirl kernel:  <c017cbbf>
proc_pid_readlink+0x102/0x10c   <c024ebda> sys_sendto+0x116/0x140
May  3 08:51:53 jonsmirl kernel:  <c013a014> __alloc_pages+0x55/0x26c <c013ffee> __handle_mm_fault+0x168/0x6d9
May  3 08:51:53 jonsmirl kernel:  <c024f5db>

sys_socketcall+0x17b/0x261 <c0102813> sysenter_past_esp+0x54/0x75

Quoted 173 lines · Al Viro

From: Steve Grubb <[email protected]>
Date: Mon Apr 3 09:08:13 2006 -0400

The below patch should be applied after the inode and ipc sid patches. This patch is a reworking of Tim's patch that has been updated to match the inode and ipc patches since its similar.

[updated:

Quoted 2 lines

Stephen Smalley also wanted to change a variable from isec to tsec in the
user sid patch. ]

---

 include/linux/netlink.h    |    1 +
 include/linux/selinux.h    |   16 ++++++++++++++++
 kernel/audit.c             |   22 +++++++++++++++++++---
 net/netlink/af_netlink.c   |    2 ++
 security/selinux/exports.c |   11 +++++++++++
 5 files changed, 49 insertions(+), 3 deletions(-)

e7c3497013a7e5496ce3d5fd3c73b5cf5af7a56e

include/linux/netlink.h +1 -0
index f8f3d1c..87b8a57 100644
@@ -143,6 +143,7 @@ struct netlink_skb_parms
        __u32                   dst_group;
        kernel_cap_t            eff_cap;
        __u32                   loginuid;       /* Login (audit) uid */
+       __u32                   sid;            /* SELinux security id */
 };
 
 #define NETLINK_CB(skb)                (*(struct netlink_skb_parms*)&((skb)->cb))
include/linux/selinux.h +16 -0
index 413d667..4047bcd 100644
@@ -5,6 +5,7 @@
  *
  * Copyright (C) 2005 Red Hat, Inc., James Morris <[email protected]>
  * Copyright (C) 2006 Trusted Computer Solutions, Inc. <[email protected]>
+ * Copyright (C) 2006 IBM Corporation, Timothy R. Chavez <[email protected]>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2,
@@ -108,6 +109,16 @@ void selinux_get_inode_sid(const struct
  */
 void selinux_get_ipc_sid(const struct kern_ipc_perm *ipcp, u32 *sid);
 
+/**
+ *     selinux_get_task_sid - return the SID of task
+ *     @tsk: the task whose SID will be returned
+ *     @sid: pointer to security context ID to be filled in.
+ *
+ *     Returns nothing
+ */
+void selinux_get_task_sid(struct task_struct *tsk, u32 *sid);
+
+
 #else
 
 static inline int selinux_audit_rule_init(u32 field, u32 op,
@@ -156,6 +167,11 @@ static inline void selinux_get_ipc_sid(c
        *sid = 0;
 }
 
+static inline void selinux_get_task_sid(struct task_struct *tsk, u32 *sid)
+{
+       *sid = 0;
+}
+
 #endif /* CONFIG_SECURITY_SELINUX */
 
 #endif /* _LINUX_SELINUX_H */
kernel/audit.c +19 -3
index 9060be7..7ec9cca 100644
@@ -390,7 +390,7 @@ static int audit_netlink_ok(kernel_cap_t
 
 static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 {
-       u32                     uid, pid, seq;
+       u32                     uid, pid, seq, sid;
        void                    *data;
        struct audit_status     *status_get, status_set;
        int                     err;
@@ -416,6 +416,7 @@ static int audit_receive_msg(struct sk_b
        pid  = NETLINK_CREDS(skb)->pid;
        uid  = NETLINK_CREDS(skb)->uid;
        loginuid = NETLINK_CB(skb).loginuid;
+       sid  = NETLINK_CB(skb).sid;
        seq  = nlh->nlmsg_seq;
        data = NLMSG_DATA(nlh);
 
@@ -468,8 +469,23 @@ static int audit_receive_msg(struct sk_b
                        ab = audit_log_start(NULL, GFP_KERNEL, msg_type);
                        if (ab) {
                                audit_log_format(ab,
-                                                "user pid=%d uid=%u auid=%u msg='%.1024s'",
-                                                pid, uid, loginuid, (char *)data);
+                                                "user pid=%d uid=%u auid=%u",
+                                                pid, uid, loginuid);
+                               if (sid) {
+                                       char *ctx = NULL;
+                                       u32 len;
+                                       if (selinux_ctxid_to_string(
+                                                       sid, &ctx, &len)) {
+                                               audit_log_format(ab,
+                                                       " subj=%u", sid);
+                                               /* Maybe call audit_panic? */
+                                       } else
+                                               audit_log_format(ab,
+                                                       " subj=%s", ctx);
+                                       kfree(ctx);
+                               }
+                               audit_log_format(ab, " msg='%.1024s'",
+                                        (char *)data);
                                audit_set_pid(ab, pid);
                                audit_log_end(ab);
                        }
net/netlink/af_netlink.c +2 -0
index 2a233ff..09fbc4b 100644
@@ -56,6 +56,7 @@ #include <linux/bitops.h>
 #include <linux/mm.h>
 #include <linux/types.h>
 #include <linux/audit.h>
+#include <linux/selinux.h>
 
 #include <net/sock.h>
 #include <net/scm.h>
@@ -1157,6 +1158,7 @@ static int netlink_sendmsg(struct kiocb
        NETLINK_CB(skb).dst_pid = dst_pid;
        NETLINK_CB(skb).dst_group = dst_group;
        NETLINK_CB(skb).loginuid = audit_get_loginuid(current->audit_context);
+       selinux_get_task_sid(current, &(NETLINK_CB(skb).sid));
        memcpy(NETLINK_CREDS(skb), &siocb->scm->creds, sizeof(struct ucred));
 
        /* What can I do? Netlink is asynchronous, so that
security/selinux/exports.c +11 -1
index 7357cf2..ae4c73e 100644
@@ -5,6 +5,7 @@
  *
  * Copyright (C) 2005 Red Hat, Inc., James Morris <[email protected]>
  * Copyright (C) 2006 Trusted Computer Solutions, Inc. <[email protected]>
+ * Copyright (C) 2006 IBM Corporation, Timothy R. Chavez <[email protected]>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2,
@@ -61,3 +62,13 @@ void selinux_get_ipc_sid(const struct ke
        *sid = 0;
 }
 
+void selinux_get_task_sid(struct task_struct *tsk, u32 *sid)
+{
+       if (selinux_enabled) {
+               struct task_security_struct *tsec = tsk->security;
+               *sid = tsec->sid;
+               return;
+       }
+       *sid = 0;
+}
+
--

1.3.0.g0080f

Quoted 6 lines

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

--
Jon Smirl
[email protected]

author
#2 lore raw
Quoted 2 lines · Jon Smirl

Something seems to be wrong in selinux_get_task_sid. I am getting thousands of these and can't boot the kernel.

It's actually in security/selinux/hooks.c::selinux_disable() and gets triggered if you have selinux enabled and explicitly disable afterwards. Stephen Smalley had done a fix yesterday, basically adding

	selinux_enabled = 0;
after
        selinux_disabled = 1;

in there. selinux_get_task_sid() happens to step on that in visible way and nobody had caught that while this stuff was sitting in -mm ;-/

The only question I have about that patch: what would happen if we do not have CONFIG_SECURITY_SELINUX_BOOTPARAM? In that case selinux_enabled is defined to 1, so...

#3 lore raw
Quoted 10 lines · Al Viro
Quoted 2 lines · Jon Smirl

Something seems to be wrong in selinux_get_task_sid. I am getting thousands of these and can't boot the kernel.

It's actually in security/selinux/hooks.c::selinux_disable() and gets triggered if you have selinux enabled and explicitly disable afterwards. Stephen Smalley had done a fix yesterday, basically adding

        selinux_enabled = 0;
after
        selinux_disabled = 1;
Quoted 6 lines

in there. selinux_get_task_sid() happens to step on that in visible way and nobody had caught that while this stuff was sitting in -mm ;-/

The only question I have about that patch: what would happen if we do not have CONFIG_SECURITY_SELINUX_BOOTPARAM? In that case selinux_enabled is defined to 1, so...

I have these config options set:

CONFIG_SECURITY_SELINUX=y
CONFIG_SECURITY_SELINUX_BOOTPARAM=y
CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE=1
CONFIG_SECURITY_SELINUX_DISABLE=y
CONFIG_SECURITY_SELINUX_DEVELOP=y
CONFIG_SECURITY_SELINUX_AVC_STATS=y
CONFIG_SECURITY_SELINUX_CHECKREQPROT_VALUE=1

SELinux needs to be built in or FC5 won't run.

--
Jon Smirl
[email protected]

#5 lore raw
In reply to Al Viro
Quoted 10 lines · Al Viro
Quoted 2 lines · Jon Smirl

Something seems to be wrong in selinux_get_task_sid. I am getting thousands of these and can't boot the kernel.

It's actually in security/selinux/hooks.c::selinux_disable() and gets triggered if you have selinux enabled and explicitly disable afterwards. Stephen Smalley had done a fix yesterday, basically adding

	selinux_enabled = 0;
after
        selinux_disabled = 1;
Quoted 6 lines

in there. selinux_get_task_sid() happens to step on that in visible way and nobody had caught that while this stuff was sitting in -mm ;-/

The only question I have about that patch: what would happen if we do not have CONFIG_SECURITY_SELINUX_BOOTPARAM? In that case selinux_enabled is defined to 1, so...

Good point. Ok, take two.

[patch 1/1] selinux: Clear selinux_enabled flag upon runtime disable.

Clear selinux_enabled flag upon runtime disable of SELinux by userspace, and make sure it is defined even if selinux= boot parameter support is not enabled in configuration.

---

 security/selinux/hooks.c            |    3 +++
 security/selinux/include/security.h |    5 -----
 2 files changed, 3 insertions(+), 5 deletions(-)
security/selinux/hooks.c +3 -0
diff -X /home/sds/dontdiff -rup linux-2.6.17-rc3-mm1/security/selinux/hooks.c linux-2.6.17-rc3-mm1-x2/security/selinux/hooks.c
@@ -101,6 +101,8 @@ static int __init selinux_enabled_setup(
 	return 1;
 }
 __setup("selinux=", selinux_enabled_setup);
+#else
+int selinux_enabled = 1;
 #endif
 
 /* Original (dummy) security module. */
@@ -4535,6 +4537,7 @@ int selinux_disable(void)
 	printk(KERN_INFO "SELinux:  Disabled at runtime.\n");
 
 	selinux_disabled = 1;
+	selinux_enabled = 0;
 
 	/* Reset security_ops to the secondary module, dummy or capability. */
 	security_ops = secondary_ops;
security/selinux/include/security.h +0 -5
diff -X /home/sds/dontdiff -rup linux-2.6.17-rc3-mm1/security/selinux/include/security.h linux-2.6.17-rc3-mm1-x2/security/selinux/include/security.h
@@ -29,12 +29,7 @@
 #define POLICYDB_VERSION_MIN   POLICYDB_VERSION_BASE
 #define POLICYDB_VERSION_MAX   POLICYDB_VERSION_AVTAB
 
-#ifdef CONFIG_SECURITY_SELINUX_BOOTPARAM
 extern int selinux_enabled;
-#else
-#define selinux_enabled 1
-#endif
-
 extern int selinux_mls_enabled;
 
 int security_load_policy(void * data, size_t len);
Signature · 2 lines
Stephen Smalley
National Security Agency
author
#4 lore raw
In reply to Jon Smirl
Quoted 28 lines · Jon Smirl
Quoted 10 lines · Al Viro
Quoted 2 lines · Jon Smirl

Something seems to be wrong in selinux_get_task_sid. I am getting thousands of these and can't boot the kernel.

It's actually in security/selinux/hooks.c::selinux_disable() and gets triggered if you have selinux enabled and explicitly disable afterwards. Stephen Smalley had done a fix yesterday, basically adding

       selinux_enabled = 0;
after
       selinux_disabled = 1;
Quoted 6 lines

in there. selinux_get_task_sid() happens to step on that in visible way and nobody had caught that while this stuff was sitting in -mm ;-/

The only question I have about that patch: what would happen if we do not have CONFIG_SECURITY_SELINUX_BOOTPARAM? In that case selinux_enabled is defined to 1, so...

I have these config options set:

CONFIG_SECURITY_SELINUX=y
CONFIG_SECURITY_SELINUX_BOOTPARAM=y
CONFIG_SECURITY_SELINUX_BOOTPARAM_VALUE=1
CONFIG_SECURITY_SELINUX_DISABLE=y
CONFIG_SECURITY_SELINUX_DEVELOP=y
CONFIG_SECURITY_SELINUX_AVC_STATS=y
CONFIG_SECURITY_SELINUX_CHECKREQPROT_VALUE=1

Quoted 1 line

SELinux needs to be built in or FC5 won't run.

Then add the assignment above (selinux_enabled = 0;) in selinux_disable() and see if that fixes your problem.

#7 lore raw
In reply to Stephen Smalley

I can boot again with this patch...

Quoted 73 lines · Stephen Smalley
Quoted 10 lines · Al Viro
Quoted 2 lines · Jon Smirl

Something seems to be wrong in selinux_get_task_sid. I am getting thousands of these and can't boot the kernel.

It's actually in security/selinux/hooks.c::selinux_disable() and gets triggered if you have selinux enabled and explicitly disable afterwards. Stephen Smalley had done a fix yesterday, basically adding

      selinux_enabled = 0;
after
        selinux_disabled = 1;
Quoted 6 lines

in there. selinux_get_task_sid() happens to step on that in visible way and nobody had caught that while this stuff was sitting in -mm ;-/

The only question I have about that patch: what would happen if we do not have CONFIG_SECURITY_SELINUX_BOOTPARAM? In that case selinux_enabled is defined to 1, so...

Good point. Ok, take two.

[patch 1/1] selinux: Clear selinux_enabled flag upon runtime disable.

Clear selinux_enabled flag upon runtime disable of SELinux by userspace, and make sure it is defined even if selinux= boot parameter support is not enabled in configuration.

---

 security/selinux/hooks.c            |    3 +++
 security/selinux/include/security.h |    5 -----
 2 files changed, 3 insertions(+), 5 deletions(-)

diff -X /home/sds/dontdiff -rup linux-2.6.17-rc3-mm1/security/selinux/hooks.c linux-2.6.17-rc3-mm1-x2/security/selinux/hooks.c

security/selinux/hooks.c 2006-05-03 10:26:43.000000000 -0400 renamed from security/selinux/hooks.c 2006-05-02 09:08:02.000000000 -0400 +3 -0
@@ -101,6 +101,8 @@ static int __init selinux_enabled_setup(
        return 1;
 }
 __setup("selinux=", selinux_enabled_setup);
+#else
+int selinux_enabled = 1;
 #endif
 
 /* Original (dummy) security module. */
@@ -4535,6 +4537,7 @@ int selinux_disable(void)
        printk(KERN_INFO "SELinux:  Disabled at runtime.\n");
 
        selinux_disabled = 1;
+       selinux_enabled = 0;
 
        /* Reset security_ops to the secondary module, dummy or capability. */
        security_ops = secondary_ops;

diff -X /home/sds/dontdiff -rup linux-2.6.17-rc3-mm1/security/selinux/include/security.h linux-2.6.17-rc3-mm1-x2/security/selinux/include/security.h

security/selinux/include/security.h 2006-05-03 10:25:39.000000000 -0400 renamed from security/selinux/include/security.h 2006-03-20 00:53:29.000000000 -0500 +0 -6
@@ -29,12 +29,7 @@
 #define POLICYDB_VERSION_MIN   POLICYDB_VERSION_BASE
 #define POLICYDB_VERSION_MAX   POLICYDB_VERSION_AVTAB
 
-#ifdef CONFIG_SECURITY_SELINUX_BOOTPARAM
 extern int selinux_enabled;
-#else
-#define selinux_enabled 1
-#endif
-
 extern int selinux_mls_enabled;
 
 int security_load_policy(void * data, size_t len);
 
 
--
Quoted 4 lines

Stephen Smalley
National Security Agency

--
Jon Smirl
[email protected]

author Acked-by
#8 lore raw
In reply to Stephen Smalley
Quoted 20 lines · Stephen Smalley
Quoted 10 lines · Al Viro
Quoted 2 lines · Jon Smirl

Something seems to be wrong in selinux_get_task_sid. I am getting thousands of these and can't boot the kernel.

It's actually in security/selinux/hooks.c::selinux_disable() and gets triggered if you have selinux enabled and explicitly disable afterwards. Stephen Smalley had done a fix yesterday, basically adding

	selinux_enabled = 0;
after
        selinux_disabled = 1;
Quoted 6 lines

in there. selinux_get_task_sid() happens to step on that in visible way and nobody had caught that while this stuff was sitting in -mm ;-/

The only question I have about that patch: what would happen if we do not have CONFIG_SECURITY_SELINUX_BOOTPARAM? In that case selinux_enabled is defined to 1, so...

Good point. Ok, take two.

Quoted 7 lines

[patch 1/1] selinux: Clear selinux_enabled flag upon runtime disable.

Clear selinux_enabled flag upon runtime disable of SELinux by userspace, and make sure it is defined even if selinux= boot parameter support is not enabled in configuration.