summary refs log tree commit diff
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2017-11-23 20:48:26 -1000
committerLinus Torvalds <torvalds@linux-foundation.org>2017-11-23 20:48:26 -1000
commit26064dea2dc65c57d44ad37e645ebe47f1c51828 (patch)
treed327f210735f8da497bd11c0a2f6a5a60134e9d2
parent5a787756b809888e8925d722862167f1229b58f7 (diff)
parentfeb3c766a3ab32d233aaff7db13afd9ba5bc142d (diff)
downloadlinux-26064dea2dc65c57d44ad37e645ebe47f1c51828.tar.gz
Merge tag 'apparmor-pr-2017-11-21' of git://git.kernel.org/pub/scm/linux/kernel/git/jj/linux-apparmor
Pull apparmor updates from John Johansen:
 "No features this time, just minor cleanups and bug fixes.

  Cleanups:
   - fix spelling mistake: "resoure" -> "resource"
   - remove unused redundant variable stop
   - Fix bool initialization/comparison

  Bug Fixes:
   - initialized returned struct aa_perms
   - fix leak of null profile name if profile allocation fails
   - ensure that undecidable profile attachments fail
   - fix profile attachment for special unconfined profiles
   - fix locking when creating a new complain profile.
   - fix possible recursive lock warning in __aa_create_ns"

* tag 'apparmor-pr-2017-11-21' of git://git.kernel.org/pub/scm/linux/kernel/git/jj/linux-apparmor:
  apparmor: fix possible recursive lock warning in __aa_create_ns
  apparmor: fix locking when creating a new complain profile.
  apparmor: fix profile attachment for special unconfined profiles
  apparmor: ensure that undecidable profile attachments fail
  apparmor: fix leak of null profile name if profile allocation fails
  apparmor: remove unused redundant variable stop
  apparmor: Fix bool initialization/comparison
  apparmor: initialized returned struct aa_perms
  apparmor: fix spelling mistake: "resoure" -> "resource"
-rw-r--r--security/apparmor/apparmorfs.c22
-rw-r--r--security/apparmor/domain.c68
-rw-r--r--security/apparmor/file.c8
-rw-r--r--security/apparmor/label.c2
-rw-r--r--security/apparmor/lib.c16
-rw-r--r--security/apparmor/lsm.c8
-rw-r--r--security/apparmor/mount.c13
-rw-r--r--security/apparmor/policy.c11
-rw-r--r--security/apparmor/policy_ns.c8
-rw-r--r--security/apparmor/policy_unpack.c2
-rw-r--r--security/apparmor/resource.c4
11 files changed, 91 insertions, 71 deletions
diff --git a/security/apparmor/apparmorfs.c b/security/apparmor/apparmorfs.c
index caaf51dda648..8542e9a55e1b 100644
--- a/security/apparmor/apparmorfs.c
+++ b/security/apparmor/apparmorfs.c
@@ -533,7 +533,7 @@ static ssize_t ns_revision_read(struct file *file, char __user *buf,
 	long last_read;
 	int avail;
 
-	mutex_lock(&rev->ns->lock);
+	mutex_lock_nested(&rev->ns->lock, rev->ns->level);
 	last_read = rev->last_read;
 	if (last_read == rev->ns->revision) {
 		mutex_unlock(&rev->ns->lock);
@@ -543,7 +543,7 @@ static ssize_t ns_revision_read(struct file *file, char __user *buf,
 					     last_read !=
 					     READ_ONCE(rev->ns->revision)))
 			return -ERESTARTSYS;
-		mutex_lock(&rev->ns->lock);
+		mutex_lock_nested(&rev->ns->lock, rev->ns->level);
 	}
 
 	avail = sprintf(buffer, "%ld\n", rev->ns->revision);
@@ -577,7 +577,7 @@ static unsigned int ns_revision_poll(struct file *file, poll_table *pt)
 	unsigned int mask = 0;
 
 	if (rev) {
-		mutex_lock(&rev->ns->lock);
+		mutex_lock_nested(&rev->ns->lock, rev->ns->level);
 		poll_wait(file, &rev->ns->wait, pt);
 		if (rev->last_read < rev->ns->revision)
 			mask |= POLLIN | POLLRDNORM;
@@ -1643,7 +1643,7 @@ static int ns_mkdir_op(struct inode *dir, struct dentry *dentry, umode_t mode)
 	 */
 	inode_unlock(dir);
 	error = simple_pin_fs(&aafs_ops, &aafs_mnt, &aafs_count);
-	mutex_lock(&parent->lock);
+	mutex_lock_nested(&parent->lock, parent->level);
 	inode_lock_nested(dir, I_MUTEX_PARENT);
 	if (error)
 		goto out;
@@ -1692,7 +1692,7 @@ static int ns_rmdir_op(struct inode *dir, struct dentry *dentry)
 	inode_unlock(dir);
 	inode_unlock(dentry->d_inode);
 
-	mutex_lock(&parent->lock);
+	mutex_lock_nested(&parent->lock, parent->level);
 	ns = aa_get_ns(__aa_findn_ns(&parent->sub_ns, dentry->d_name.name,
 				     dentry->d_name.len));
 	if (!ns) {
@@ -1747,7 +1747,7 @@ void __aafs_ns_rmdir(struct aa_ns *ns)
 		__aafs_profile_rmdir(child);
 
 	list_for_each_entry(sub, &ns->sub_ns, base.list) {
-		mutex_lock(&sub->lock);
+		mutex_lock_nested(&sub->lock, sub->level);
 		__aafs_ns_rmdir(sub);
 		mutex_unlock(&sub->lock);
 	}
@@ -1877,7 +1877,7 @@ int __aafs_ns_mkdir(struct aa_ns *ns, struct dentry *parent, const char *name,
 
 	/* subnamespaces */
 	list_for_each_entry(sub, &ns->sub_ns, base.list) {
-		mutex_lock(&sub->lock);
+		mutex_lock_nested(&sub->lock, sub->level);
 		error = __aafs_ns_mkdir(sub, ns_subns_dir(ns), NULL, NULL);
 		mutex_unlock(&sub->lock);
 		if (error)
@@ -1921,7 +1921,7 @@ static struct aa_ns *__next_ns(struct aa_ns *root, struct aa_ns *ns)
 	/* is next namespace a child */
 	if (!list_empty(&ns->sub_ns)) {
 		next = list_first_entry(&ns->sub_ns, typeof(*ns), base.list);
-		mutex_lock(&next->lock);
+		mutex_lock_nested(&next->lock, next->level);
 		return next;
 	}
 
@@ -1931,7 +1931,7 @@ static struct aa_ns *__next_ns(struct aa_ns *root, struct aa_ns *ns)
 		mutex_unlock(&ns->lock);
 		next = list_next_entry(ns, base.list);
 		if (!list_entry_is_head(next, &parent->sub_ns, base.list)) {
-			mutex_lock(&next->lock);
+			mutex_lock_nested(&next->lock, next->level);
 			return next;
 		}
 		ns = parent;
@@ -2039,7 +2039,7 @@ static void *p_start(struct seq_file *f, loff_t *pos)
 	f->private = root;
 
 	/* find the first profile */
-	mutex_lock(&root->lock);
+	mutex_lock_nested(&root->lock, root->level);
 	profile = __first_profile(root, root);
 
 	/* skip to position */
@@ -2491,7 +2491,7 @@ static int __init aa_create_aafs(void)
 	ns_subrevision(root_ns) = dent;
 
 	/* policy tree referenced by magic policy symlink */
-	mutex_lock(&root_ns->lock);
+	mutex_lock_nested(&root_ns->lock, root_ns->level);
 	error = __aafs_ns_mkdir(root_ns, aafs_mnt->mnt_root, ".policy",
 				aafs_mnt->mnt_root);
 	mutex_unlock(&root_ns->lock);
diff --git a/security/apparmor/domain.c b/security/apparmor/domain.c
index dd754b7850a8..04ba9d0718ea 100644
--- a/security/apparmor/domain.c
+++ b/security/apparmor/domain.c
@@ -305,6 +305,7 @@ static int change_profile_perms(struct aa_profile *profile,
  * __attach_match_ - find an attachment match
  * @name - to match against  (NOT NULL)
  * @head - profile list to walk  (NOT NULL)
+ * @info - info message if there was an error (NOT NULL)
  *
  * Do a linear search on the profiles in the list.  There is a matching
  * preference where an exact match is preferred over a name which uses
@@ -316,28 +317,46 @@ static int change_profile_perms(struct aa_profile *profile,
  * Returns: profile or NULL if no match found
  */
 static struct aa_profile *__attach_match(const char *name,
-					 struct list_head *head)
+					 struct list_head *head,
+					 const char **info)
 {
 	int len = 0;
+	bool conflict = false;
 	struct aa_profile *profile, *candidate = NULL;
 
 	list_for_each_entry_rcu(profile, head, base.list) {
-		if (profile->label.flags & FLAG_NULL)
+		if (profile->label.flags & FLAG_NULL &&
+		    &profile->label == ns_unconfined(profile->ns))
 			continue;
-		if (profile->xmatch && profile->xmatch_len > len) {
-			unsigned int state = aa_dfa_match(profile->xmatch,
-							  DFA_START, name);
-			u32 perm = dfa_user_allow(profile->xmatch, state);
-			/* any accepting state means a valid match. */
-			if (perm & MAY_EXEC) {
-				candidate = profile;
-				len = profile->xmatch_len;
+
+		if (profile->xmatch) {
+			if (profile->xmatch_len == len) {
+				conflict = true;
+				continue;
+			} else if (profile->xmatch_len > len) {
+				unsigned int state;
+				u32 perm;
+
+				state = aa_dfa_match(profile->xmatch,
+						     DFA_START, name);
+				perm = dfa_user_allow(profile->xmatch, state);
+				/* any accepting state means a valid match. */
+				if (perm & MAY_EXEC) {
+					candidate = profile;
+					len = profile->xmatch_len;
+					conflict = false;
+				}
 			}
 		} else if (!strcmp(profile->base.name, name))
 			/* exact non-re match, no more searching required */
 			return profile;
 	}
 
+	if (conflict) {
+		*info = "conflicting profile attachments";
+		return NULL;
+	}
+
 	return candidate;
 }
 
@@ -346,16 +365,17 @@ static struct aa_profile *__attach_match(const char *name,
  * @ns: the current namespace  (NOT NULL)
  * @list: list to search  (NOT NULL)
  * @name: the executable name to match against  (NOT NULL)
+ * @info: info message if there was an error
  *
  * Returns: label or NULL if no match found
  */
 static struct aa_label *find_attach(struct aa_ns *ns, struct list_head *list,
-				    const char *name)
+				    const char *name, const char **info)
 {
 	struct aa_profile *profile;
 
 	rcu_read_lock();
-	profile = aa_get_profile(__attach_match(name, list));
+	profile = aa_get_profile(__attach_match(name, list, info));
 	rcu_read_unlock();
 
 	return profile ? &profile->label : NULL;
@@ -448,11 +468,11 @@ static struct aa_label *x_to_label(struct aa_profile *profile,
 		if (xindex & AA_X_CHILD)
 			/* released by caller */
 			new = find_attach(ns, &profile->base.profiles,
-						name);
+					  name, info);
 		else
 			/* released by caller */
 			new = find_attach(ns, &ns->base.profiles,
-						name);
+					  name, info);
 		*lookupname = name;
 		break;
 	}
@@ -516,7 +536,7 @@ static struct aa_label *profile_transition(struct aa_profile *profile,
 
 	if (profile_unconfined(profile)) {
 		new = find_attach(profile->ns, &profile->ns->base.profiles,
-				  name);
+				  name, &info);
 		if (new) {
 			AA_DEBUG("unconfined attached to new label");
 			return new;
@@ -541,9 +561,21 @@ static struct aa_label *profile_transition(struct aa_profile *profile,
 		}
 	} else if (COMPLAIN_MODE(profile)) {
 		/* no exec permission - learning mode */
-		struct aa_profile *new_profile = aa_new_null_profile(profile,
-							      false, name,
-							      GFP_ATOMIC);
+		struct aa_profile *new_profile = NULL;
+		char *n = kstrdup(name, GFP_ATOMIC);
+
+		if (n) {
+			/* name is ptr into buffer */
+			long pos = name - buffer;
+			/* break per cpu buffer hold */
+			put_buffers(buffer);
+			new_profile = aa_new_null_profile(profile, false, n,
+							  GFP_KERNEL);
+			get_buffers(buffer);
+			name = buffer + pos;
+			strcpy((char *)name, n);
+			kfree(n);
+		}
 		if (!new_profile) {
 			error = -ENOMEM;
 			info = "could not create null profile";
diff --git a/security/apparmor/file.c b/security/apparmor/file.c
index 3382518b87fa..e79bf44396a3 100644
--- a/security/apparmor/file.c
+++ b/security/apparmor/file.c
@@ -226,18 +226,12 @@ static u32 map_old_perms(u32 old)
 struct aa_perms aa_compute_fperms(struct aa_dfa *dfa, unsigned int state,
 				  struct path_cond *cond)
 {
-	struct aa_perms perms;
-
 	/* FIXME: change over to new dfa format
 	 * currently file perms are encoded in the dfa, new format
 	 * splits the permissions from the dfa.  This mapping can be
 	 * done at profile load
 	 */
-	perms.deny = 0;
-	perms.kill = perms.stop = 0;
-	perms.complain = perms.cond = 0;
-	perms.hide = 0;
-	perms.prompt = 0;
+	struct aa_perms perms = { };
 
 	if (uid_eq(current_fsuid(), cond->uid)) {
 		perms.allow = map_old_perms(dfa_user_allow(dfa, state));
diff --git a/security/apparmor/label.c b/security/apparmor/label.c
index ad28e03a6f30..324fe5c60f87 100644
--- a/security/apparmor/label.c
+++ b/security/apparmor/label.c
@@ -2115,7 +2115,7 @@ void __aa_labelset_update_subtree(struct aa_ns *ns)
 	__labelset_update(ns);
 
 	list_for_each_entry(child, &ns->sub_ns, base.list) {
-		mutex_lock(&child->lock);
+		mutex_lock_nested(&child->lock, child->level);
 		__aa_labelset_update_subtree(child);
 		mutex_unlock(&child->lock);
 	}
diff --git a/security/apparmor/lib.c b/security/apparmor/lib.c
index 08ca26bcca77..4d5e98e49d5e 100644
--- a/security/apparmor/lib.c
+++ b/security/apparmor/lib.c
@@ -317,14 +317,11 @@ static u32 map_other(u32 x)
 void aa_compute_perms(struct aa_dfa *dfa, unsigned int state,
 		      struct aa_perms *perms)
 {
-	perms->deny = 0;
-	perms->kill = perms->stop = 0;
-	perms->complain = perms->cond = 0;
-	perms->hide = 0;
-	perms->prompt = 0;
-	perms->allow = dfa_user_allow(dfa, state);
-	perms->audit = dfa_user_audit(dfa, state);
-	perms->quiet = dfa_user_quiet(dfa, state);
+	*perms = (struct aa_perms) {
+		.allow = dfa_user_allow(dfa, state),
+		.audit = dfa_user_audit(dfa, state),
+		.quiet = dfa_user_quiet(dfa, state),
+	};
 
 	/* for v5 perm mapping in the policydb, the other set is used
 	 * to extend the general perm set
@@ -426,7 +423,6 @@ int aa_check_perms(struct aa_profile *profile, struct aa_perms *perms,
 		   void (*cb)(struct audit_buffer *, void *))
 {
 	int type, error;
-	bool stop = false;
 	u32 denied = request & (~perms->allow | perms->deny);
 
 	if (likely(!denied)) {
@@ -447,8 +443,6 @@ int aa_check_perms(struct aa_profile *profile, struct aa_perms *perms,
 		else
 			type = AUDIT_APPARMOR_DENIED;
 
-		if (denied & perms->stop)
-			stop = true;
 		if (denied == (denied & perms->hide))
 			error = -ENOENT;
 
diff --git a/security/apparmor/lsm.c b/security/apparmor/lsm.c
index 17893fde4487..9a65eeaf7dfa 100644
--- a/security/apparmor/lsm.c
+++ b/security/apparmor/lsm.c
@@ -846,7 +846,7 @@ module_param_call(audit, param_set_audit, param_get_audit,
 /* Determines if audit header is included in audited messages.  This
  * provides more context if the audit daemon is not running
  */
-bool aa_g_audit_header = 1;
+bool aa_g_audit_header = true;
 module_param_named(audit_header, aa_g_audit_header, aabool,
 		   S_IRUSR | S_IWUSR);
 
@@ -871,7 +871,7 @@ module_param_named(path_max, aa_g_path_max, aauint, S_IRUSR);
  * DEPRECATED: read only as strict checking of load is always done now
  * that none root users (user namespaces) can load policy.
  */
-bool aa_g_paranoid_load = 1;
+bool aa_g_paranoid_load = true;
 module_param_named(paranoid_load, aa_g_paranoid_load, aabool, S_IRUGO);
 
 /* Boot time disable flag */
@@ -1119,7 +1119,7 @@ static int __init apparmor_init(void)
 
 	if (!apparmor_enabled || !security_module_enable("apparmor")) {
 		aa_info_message("AppArmor disabled by boot time parameter");
-		apparmor_enabled = 0;
+		apparmor_enabled = false;
 		return 0;
 	}
 
@@ -1175,7 +1175,7 @@ alloc_out:
 	aa_destroy_aafs();
 	aa_teardown_dfa_engine();
 
-	apparmor_enabled = 0;
+	apparmor_enabled = false;
 	return error;
 }
 
diff --git a/security/apparmor/mount.c b/security/apparmor/mount.c
index 82a64b58041d..ed9b4d0f9f7e 100644
--- a/security/apparmor/mount.c
+++ b/security/apparmor/mount.c
@@ -216,13 +216,12 @@ static unsigned int match_mnt_flags(struct aa_dfa *dfa, unsigned int state,
 static struct aa_perms compute_mnt_perms(struct aa_dfa *dfa,
 					   unsigned int state)
 {
-	struct aa_perms perms;
-
-	perms.kill = 0;
-	perms.allow = dfa_user_allow(dfa, state);
-	perms.audit = dfa_user_audit(dfa, state);
-	perms.quiet = dfa_user_quiet(dfa, state);
-	perms.xindex = dfa_user_xindex(dfa, state);
+	struct aa_perms perms = {
+		.allow = dfa_user_allow(dfa, state),
+		.audit = dfa_user_audit(dfa, state),
+		.quiet = dfa_user_quiet(dfa, state),
+		.xindex = dfa_user_xindex(dfa, state),
+	};
 
 	return perms;
 }
diff --git a/security/apparmor/policy.c b/security/apparmor/policy.c
index 4243b0c3f0e4..b0b58848c248 100644
--- a/security/apparmor/policy.c
+++ b/security/apparmor/policy.c
@@ -502,7 +502,7 @@ struct aa_profile *aa_new_null_profile(struct aa_profile *parent, bool hat,
 {
 	struct aa_profile *p, *profile;
 	const char *bname;
-	char *name;
+	char *name = NULL;
 
 	AA_BUG(!parent);
 
@@ -545,7 +545,7 @@ name:
 	profile->file.dfa = aa_get_dfa(nulldfa);
 	profile->policy.dfa = aa_get_dfa(nulldfa);
 
-	mutex_lock(&profile->ns->lock);
+	mutex_lock_nested(&profile->ns->lock, profile->ns->level);
 	p = __find_child(&parent->base.profiles, bname);
 	if (p) {
 		aa_free_profile(profile);
@@ -562,6 +562,7 @@ out:
 	return profile;
 
 fail:
+	kfree(name);
 	aa_free_profile(profile);
 	return NULL;
 }
@@ -905,7 +906,7 @@ ssize_t aa_replace_profiles(struct aa_ns *policy_ns, struct aa_label *label,
 	} else
 		ns = aa_get_ns(policy_ns ? policy_ns : labels_ns(label));
 
-	mutex_lock(&ns->lock);
+	mutex_lock_nested(&ns->lock, ns->level);
 	/* check for duplicate rawdata blobs: space and file dedup */
 	list_for_each_entry(rawdata_ent, &ns->rawdata_list, list) {
 		if (aa_rawdata_eq(rawdata_ent, udata)) {
@@ -1116,13 +1117,13 @@ ssize_t aa_remove_profiles(struct aa_ns *policy_ns, struct aa_label *subj,
 
 	if (!name) {
 		/* remove namespace - can only happen if fqname[0] == ':' */
-		mutex_lock(&ns->parent->lock);
+		mutex_lock_nested(&ns->parent->lock, ns->level);
 		__aa_remove_ns(ns);
 		__aa_bump_ns_revision(ns);
 		mutex_unlock(&ns->parent->lock);
 	} else {
 		/* remove profile */
-		mutex_lock(&ns->lock);
+		mutex_lock_nested(&ns->lock, ns->level);
 		profile = aa_get_profile(__lookup_profile(&ns->base, name));
 		if (!profile) {
 			error = -ENOENT;
diff --git a/security/apparmor/policy_ns.c b/security/apparmor/policy_ns.c
index 62a3589c62ab..b1e629cba70b 100644
--- a/security/apparmor/policy_ns.c
+++ b/security/apparmor/policy_ns.c
@@ -256,7 +256,8 @@ static struct aa_ns *__aa_create_ns(struct aa_ns *parent, const char *name,
 	ns = alloc_ns(parent->base.hname, name);
 	if (!ns)
 		return NULL;
-	mutex_lock(&ns->lock);
+	ns->level = parent->level + 1;
+	mutex_lock_nested(&ns->lock, ns->level);
 	error = __aafs_ns_mkdir(ns, ns_subns_dir(parent), name, dir);
 	if (error) {
 		AA_ERROR("Failed to create interface for ns %s\n",
@@ -266,7 +267,6 @@ static struct aa_ns *__aa_create_ns(struct aa_ns *parent, const char *name,
 		return ERR_PTR(error);
 	}
 	ns->parent = aa_get_ns(parent);
-	ns->level = parent->level + 1;
 	list_add_rcu(&ns->base.list, &parent->sub_ns);
 	/* add list ref */
 	aa_get_ns(ns);
@@ -313,7 +313,7 @@ struct aa_ns *aa_prepare_ns(struct aa_ns *parent, const char *name)
 {
 	struct aa_ns *ns;
 
-	mutex_lock(&parent->lock);
+	mutex_lock_nested(&parent->lock, parent->level);
 	/* try and find the specified ns and if it doesn't exist create it */
 	/* released by caller */
 	ns = aa_get_ns(__aa_find_ns(&parent->sub_ns, name));
@@ -336,7 +336,7 @@ static void destroy_ns(struct aa_ns *ns)
 	if (!ns)
 		return;
 
-	mutex_lock(&ns->lock);
+	mutex_lock_nested(&ns->lock, ns->level);
 	/* release all profiles in this namespace */
 	__aa_profile_list_release(&ns->base.profiles);
 
diff --git a/security/apparmor/policy_unpack.c b/security/apparmor/policy_unpack.c
index 4ede87c30f8b..59a1a25b7d43 100644
--- a/security/apparmor/policy_unpack.c
+++ b/security/apparmor/policy_unpack.c
@@ -157,7 +157,7 @@ static void do_loaddata_free(struct work_struct *work)
 	struct aa_ns *ns = aa_get_ns(d->ns);
 
 	if (ns) {
-		mutex_lock(&ns->lock);
+		mutex_lock_nested(&ns->lock, ns->level);
 		__aa_fs_remove_rawdata(d);
 		mutex_unlock(&ns->lock);
 		aa_put_ns(ns);
diff --git a/security/apparmor/resource.c b/security/apparmor/resource.c
index d8bc842594ed..cf4d234febe9 100644
--- a/security/apparmor/resource.c
+++ b/security/apparmor/resource.c
@@ -47,7 +47,7 @@ static void audit_cb(struct audit_buffer *ab, void *va)
 /**
  * audit_resource - audit setting resource limit
  * @profile: profile being enforced  (NOT NULL)
- * @resoure: rlimit being auditing
+ * @resource: rlimit being auditing
  * @value: value being set
  * @error: error value
  *
@@ -128,7 +128,7 @@ int aa_task_setrlimit(struct aa_label *label, struct task_struct *task,
 		error = fn_for_each(label, profile,
 				audit_resource(profile, resource,
 					       new_rlim->rlim_max, peer,
-					       "cap_sys_resoure", -EACCES));
+					       "cap_sys_resource", -EACCES));
 	else
 		error = fn_for_each_confined(label, profile,
 				profile_setrlimit(profile, resource, new_rlim));