Implement file posix capabilities

Implement file posix capabilities.  This allows programs to be given a
subset of root's powers regardless of who runs them, without having to use
setuid and giving the binary all of root's powers.

This version works with Kaigai Kohei's userspace tools, found at
http://www.kaigai.gr.jp/index.php.  For more information on how to use this
patch, Chris Friedhoff has posted a nice page at
http://www.friedhoff.org/fscaps.html.

Changelog:
	Nov 27:
	Incorporate fixes from Andrew Morton
	(security-introduce-file-caps-tweaks and
	security-introduce-file-caps-warning-fix)
	Fix Kconfig dependency.
	Fix change signaling behavior when file caps are not compiled in.

	Nov 13:
	Integrate comments from Alexey: Remove CONFIG_ ifdef from
	capability.h, and use %zd for printing a size_t.

	Nov 13:
	Fix endianness warnings by sparse as suggested by Alexey
	Dobriyan.

	Nov 09:
	Address warnings of unused variables at cap_bprm_set_security
	when file capabilities are disabled, and simultaneously clean
	up the code a little, by pulling the new code into a helper
	function.

	Nov 08:
	For pointers to required userspace tools and how to use
	them, see http://www.friedhoff.org/fscaps.html.

	Nov 07:
	Fix the calculation of the highest bit checked in
	check_cap_sanity().

	Nov 07:
	Allow file caps to be enabled without CONFIG_SECURITY, since
	capabilities are the default.
	Hook cap_task_setscheduler when !CONFIG_SECURITY.
	Move capable(TASK_KILL) to end of cap_task_kill to reduce
	audit messages.

	Nov 05:
	Add secondary calls in selinux/hooks.c to task_setioprio and
	task_setscheduler so that selinux and capabilities with file
	cap support can be stacked.

	Sep 05:
	As Seth Arnold points out, uid checks are out of place
	for capability code.

	Sep 01:
	Define task_setscheduler, task_setioprio, cap_task_kill, and
	task_setnice to make sure a user cannot affect a process in which
	they called a program with some fscaps.

	One remaining question is the note under task_setscheduler: are we
	ok with CAP_SYS_NICE being sufficient to confine a process to a
	cpuset?

	It is a semantic change, as without fsccaps, attach_task doesn't
	allow CAP_SYS_NICE to override the uid equivalence check.  But since
	it uses security_task_setscheduler, which elsewhere is used where
	CAP_SYS_NICE can be used to override the uid equivalence check,
	fixing it might be tough.

	     task_setscheduler
		 note: this also controls cpuset:attach_task.  Are we ok with
		     CAP_SYS_NICE being used to confine to a cpuset?
	     task_setioprio
	     task_setnice
		 sys_setpriority uses this (through set_one_prio) for another
		 process.  Need same checks as setrlimit

	Aug 21:
	Updated secureexec implementation to reflect the fact that
	euid and uid might be the same and nonzero, but the process
	might still have elevated caps.

	Aug 15:
	Handle endianness of xattrs.
	Enforce capability version match between kernel and disk.
	Enforce that no bits beyond the known max capability are
	set, else return -EPERM.
	With this extra processing, it may be worth reconsidering
	doing all the work at bprm_set_security rather than
	d_instantiate.

	Aug 10:
	Always call getxattr at bprm_set_security, rather than
	caching it at d_instantiate.

[morgan@kernel.org: file-caps clean up for linux/capability.h]
[bunk@kernel.org: unexport cap_inode_killpriv]
Signed-off-by: Serge E. Hallyn <serue@us.ibm.com>
Cc: Stephen Smalley <sds@tycho.nsa.gov>
Cc: James Morris <jmorris@namei.org>
Cc: Chris Wright <chrisw@sous-sol.org>
Cc: Andrew Morgan <morgan@kernel.org>
Signed-off-by: Andrew Morgan <morgan@kernel.org>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

6 files changed, 313 insertions, 43 deletions

diff --git a/security/Kconfig b/security/Kconfig
index a94ee94cf491..8086e61058e3 100644
--- a/security/Kconfig
+++ b/security/Kconfig
@@ -80,6 +80,16 @@ config SECURITY_CAPABILITIES
           This enables the "default" Linux capabilities functionality.
           If you are unsure how to answer this question, answer Y.
 
+config SECURITY_FILE_CAPABILITIES
+        bool "File POSIX Capabilities (EXPERIMENTAL)"
+        depends on (SECURITY=n || SECURITY_CAPABILITIES!=n) && EXPERIMENTAL
+        default n
+        help
+          This enables filesystem capabilities, allowing you to give
+          binaries a subset of root's powers without using setuid 0.
+
+          If in doubt, answer N.
+
 config SECURITY_ROOTPLUG
         bool "Root Plug Support"
         depends on USB=y && SECURITY
diff --git a/security/capability.c b/security/capability.c
index fda6a14cb24d..9e99f36a8b5c 100644
--- a/security/capability.c
+++ b/security/capability.c
@@ -37,7 +37,13 @@ static struct security_operations capability_ops = {
 
         .inode_setxattr =               cap_inode_setxattr,
         .inode_removexattr =            cap_inode_removexattr,
+        .inode_need_killpriv =          cap_inode_need_killpriv,
+        .inode_killpriv =               cap_inode_killpriv,
 
+        .task_kill =                    cap_task_kill,
+        .task_setscheduler =            cap_task_setscheduler,
+        .task_setioprio =               cap_task_setioprio,
+        .task_setnice =                 cap_task_setnice,
         .task_post_setuid =             cap_task_post_setuid,
         .task_reparent_to_init =        cap_task_reparent_to_init,
 
diff --git a/security/commoncap.c b/security/commoncap.c
index 0f8a2ce3f3a6..afca6dd4ae69 100644
--- a/security/commoncap.c
+++ b/security/commoncap.c
@@ -22,6 +22,7 @@
 #include <linux/ptrace.h>
 #include <linux/xattr.h>
 #include <linux/hugetlb.h>
+#include <linux/mount.h>
 
 int cap_netlink_send(struct sock *sk, struct sk_buff *skb)
 {
@@ -108,14 +109,130 @@ void cap_capset_set (struct task_struct *target, kernel_cap_t *effective,
         target->cap_permitted = *permitted;
 }
 
+static inline void bprm_clear_caps(struct linux_binprm *bprm)
+{
+        cap_clear(bprm->cap_inheritable);
+        cap_clear(bprm->cap_permitted);
+        bprm->cap_effective = false;
+}
+
+#ifdef CONFIG_SECURITY_FILE_CAPABILITIES
+
+int cap_inode_need_killpriv(struct dentry *dentry)
+{
+        struct inode *inode = dentry->d_inode;
+        int error;
+
+        if (!inode->i_op || !inode->i_op->getxattr)
+               return 0;
+
+        error = inode->i_op->getxattr(dentry, XATTR_NAME_CAPS, NULL, 0);
+        if (error <= 0)
+                return 0;
+        return 1;
+}
+
+int cap_inode_killpriv(struct dentry *dentry)
+{
+        struct inode *inode = dentry->d_inode;
+
+        if (!inode->i_op || !inode->i_op->removexattr)
+               return 0;
+
+        return inode->i_op->removexattr(dentry, XATTR_NAME_CAPS);
+}
+
+static inline int cap_from_disk(__le32 *caps, struct linux_binprm *bprm,
+                                int size)
+{
+        __u32 magic_etc;
+
+        if (size != XATTR_CAPS_SZ)
+                return -EINVAL;
+
+        magic_etc = le32_to_cpu(caps[0]);
+
+        switch ((magic_etc & VFS_CAP_REVISION_MASK)) {
+        case VFS_CAP_REVISION:
+                if (magic_etc & VFS_CAP_FLAGS_EFFECTIVE)
+                        bprm->cap_effective = true;
+                else
+                        bprm->cap_effective = false;
+                bprm->cap_permitted = to_cap_t( le32_to_cpu(caps[1]) );
+                bprm->cap_inheritable = to_cap_t( le32_to_cpu(caps[2]) );
+                return 0;
+        default:
+                return -EINVAL;
+        }
+}
+
+/* Locate any VFS capabilities: */
+static int get_file_caps(struct linux_binprm *bprm)
+{
+        struct dentry *dentry;
+        int rc = 0;
+        __le32 v1caps[XATTR_CAPS_SZ];
+        struct inode *inode;
+
+        if (bprm->file->f_vfsmnt->mnt_flags & MNT_NOSUID) {
+                bprm_clear_caps(bprm);
+                return 0;
+        }
+
+        dentry = dget(bprm->file->f_dentry);
+        inode = dentry->d_inode;
+        if (!inode->i_op || !inode->i_op->getxattr)
+                goto out;
+
+        rc = inode->i_op->getxattr(dentry, XATTR_NAME_CAPS, &v1caps,
+                                                        XATTR_CAPS_SZ);
+        if (rc == -ENODATA || rc == -EOPNOTSUPP) {
+                /* no data, that's ok */
+                rc = 0;
+                goto out;
+        }
+        if (rc < 0)
+                goto out;
+
+        rc = cap_from_disk(v1caps, bprm, rc);
+        if (rc)
+                printk(KERN_NOTICE "%s: cap_from_disk returned %d for %s\n",
+                        __FUNCTION__, rc, bprm->filename);
+
+out:
+        dput(dentry);
+        if (rc)
+                bprm_clear_caps(bprm);
+
+        return rc;
+}
+
+#else
+int cap_inode_need_killpriv(struct dentry *dentry)
+{
+        return 0;
+}
+
+int cap_inode_killpriv(struct dentry *dentry)
+{
+        return 0;
+}
+
+static inline int get_file_caps(struct linux_binprm *bprm)
+{
+        bprm_clear_caps(bprm);
+        return 0;
+}
+#endif
+
 int cap_bprm_set_security (struct linux_binprm *bprm)
 {
-        /* Copied from fs/exec.c:prepare_binprm. */
+        int ret;
 
-        /* We don't have VFS support for capabilities yet */
-        cap_clear (bprm->cap_inheritable);
-        cap_clear (bprm->cap_permitted);
-        cap_clear (bprm->cap_effective);
+        ret = get_file_caps(bprm);
+        if (ret)
+                printk(KERN_NOTICE "%s: get_file_caps returned %d for %s\n",
+                        __FUNCTION__, ret, bprm->filename);
 
         /*  To support inheritance of root-permissions and suid-root
          *  executables under compatibility mode, we raise all three
@@ -131,9 +248,10 @@ int cap_bprm_set_security (struct linux_binprm *bprm)
                         cap_set_full (bprm->cap_permitted);
                 }
                 if (bprm->e_uid == 0)
-                        cap_set_full (bprm->cap_effective);
+                        bprm->cap_effective = true;
         }
-        return 0;
+
+        return ret;
 }
 
 void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe)
@@ -149,6 +267,7 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe)
         if (bprm->e_uid != current->uid || bprm->e_gid != current->gid ||
             !cap_issubset (new_permitted, current->cap_permitted)) {
                 set_dumpable(current->mm, suid_dumpable);
+                current->pdeath_signal = 0;
 
                 if (unsafe & ~LSM_UNSAFE_PTRACE_CAP) {
                         if (!capable(CAP_SETUID)) {
@@ -170,8 +289,8 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe)
          * capability rules */
         if (!is_init(current)) {
                 current->cap_permitted = new_permitted;
-                current->cap_effective =
-                    cap_intersect (new_permitted, bprm->cap_effective);
+                current->cap_effective = bprm->cap_effective ?
+                                new_permitted : 0;
         }
 
         /* AUD: Audit candidate if current->cap_effective is set */
@@ -181,11 +300,15 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe)
 
 int cap_bprm_secureexec (struct linux_binprm *bprm)
 {
-        /* If/when this module is enhanced to incorporate capability
-           bits on files, the test below should be extended to also perform a 
-           test between the old and new capability sets.  For now,
-           it simply preserves the legacy decision algorithm used by
-           the old userland. */
+        if (current->uid != 0) {
+                if (bprm->cap_effective)
+                        return 1;
+                if (!cap_isclear(bprm->cap_permitted))
+                        return 1;
+                if (!cap_isclear(bprm->cap_inheritable))
+                        return 1;
+        }
+
         return (current->euid != current->uid ||
                 current->egid != current->gid);
 }
@@ -193,7 +316,11 @@ int cap_bprm_secureexec (struct linux_binprm *bprm)
 int cap_inode_setxattr(struct dentry *dentry, char *name, void *value,
                        size_t size, int flags)
 {
-        if (!strncmp(name, XATTR_SECURITY_PREFIX,
+        if (!strcmp(name, XATTR_NAME_CAPS)) {
+                if (!capable(CAP_SETFCAP))
+                        return -EPERM;
+                return 0;
+        } else if (!strncmp(name, XATTR_SECURITY_PREFIX,
                      sizeof(XATTR_SECURITY_PREFIX) - 1)  &&
             !capable(CAP_SYS_ADMIN))
                 return -EPERM;
@@ -202,7 +329,11 @@ int cap_inode_setxattr(struct dentry *dentry, char *name, void *value,
 
 int cap_inode_removexattr(struct dentry *dentry, char *name)
 {
-        if (!strncmp(name, XATTR_SECURITY_PREFIX,
+        if (!strcmp(name, XATTR_NAME_CAPS)) {
+                if (!capable(CAP_SETFCAP))
+                        return -EPERM;
+                return 0;
+        } else if (!strncmp(name, XATTR_SECURITY_PREFIX,
                      sizeof(XATTR_SECURITY_PREFIX) - 1)  &&
             !capable(CAP_SYS_ADMIN))
                 return -EPERM;
@@ -299,6 +430,83 @@ int cap_task_post_setuid (uid_t old_ruid, uid_t old_euid, uid_t old_suid,
         return 0;
 }
 
+#ifdef CONFIG_SECURITY_FILE_CAPABILITIES
+/*
+ * Rationale: code calling task_setscheduler, task_setioprio, and
+ * task_setnice, assumes that
+ *   . if capable(cap_sys_nice), then those actions should be allowed
+ *   . if not capable(cap_sys_nice), but acting on your own processes,
+ *      then those actions should be allowed
+ * This is insufficient now since you can call code without suid, but
+ * yet with increased caps.
+ * So we check for increased caps on the target process.
+ */
+static inline int cap_safe_nice(struct task_struct *p)
+{
+        if (!cap_issubset(p->cap_permitted, current->cap_permitted) &&
+            !__capable(current, CAP_SYS_NICE))
+                return -EPERM;
+        return 0;
+}
+
+int cap_task_setscheduler (struct task_struct *p, int policy,
+                           struct sched_param *lp)
+{
+        return cap_safe_nice(p);
+}
+
+int cap_task_setioprio (struct task_struct *p, int ioprio)
+{
+        return cap_safe_nice(p);
+}
+
+int cap_task_setnice (struct task_struct *p, int nice)
+{
+        return cap_safe_nice(p);
+}
+
+int cap_task_kill(struct task_struct *p, struct siginfo *info,
+                                int sig, u32 secid)
+{
+        if (info != SEND_SIG_NOINFO && (is_si_special(info) || SI_FROMKERNEL(info)))
+                return 0;
+
+        if (secid)
+                /*
+                 * Signal sent as a particular user.
+                 * Capabilities are ignored.  May be wrong, but it's the
+                 * only thing we can do at the moment.
+                 * Used only by usb drivers?
+                 */
+                return 0;
+        if (cap_issubset(p->cap_permitted, current->cap_permitted))
+                return 0;
+        if (capable(CAP_KILL))
+                return 0;
+
+        return -EPERM;
+}
+#else
+int cap_task_setscheduler (struct task_struct *p, int policy,
+                           struct sched_param *lp)
+{
+        return 0;
+}
+int cap_task_setioprio (struct task_struct *p, int ioprio)
+{
+        return 0;
+}
+int cap_task_setnice (struct task_struct *p, int nice)
+{
+        return 0;
+}
+int cap_task_kill(struct task_struct *p, struct siginfo *info,
+                                int sig, u32 secid)
+{
+        return 0;
+}
+#endif
+
 void cap_task_reparent_to_init (struct task_struct *p)
 {
         p->cap_effective = CAP_INIT_EFF_SET;
@@ -336,6 +544,10 @@ EXPORT_SYMBOL(cap_bprm_secureexec);
 EXPORT_SYMBOL(cap_inode_setxattr);
 EXPORT_SYMBOL(cap_inode_removexattr);
 EXPORT_SYMBOL(cap_task_post_setuid);
+EXPORT_SYMBOL(cap_task_kill);
+EXPORT_SYMBOL(cap_task_setscheduler);
+EXPORT_SYMBOL(cap_task_setioprio);
+EXPORT_SYMBOL(cap_task_setnice);
 EXPORT_SYMBOL(cap_task_reparent_to_init);
 EXPORT_SYMBOL(cap_syslog);
 EXPORT_SYMBOL(cap_vm_enough_memory);
diff --git a/security/dummy.c b/security/dummy.c
index 4129dcf3daec..c77dec822385 100644
--- a/security/dummy.c
+++ b/security/dummy.c
@@ -376,6 +376,16 @@ static int dummy_inode_removexattr (struct dentry *dentry, char *name)
         return 0;
 }
 
+static int dummy_inode_need_killpriv(struct dentry *dentry)
+{
+        return 0;
+}
+
+static int dummy_inode_killpriv(struct dentry *dentry)
+{
+        return 0;
+}
+
 static int dummy_inode_getsecurity(const struct inode *inode, const char *name, void *buffer, size_t size, int err)
 {
         return -EOPNOTSUPP;
@@ -1022,6 +1032,8 @@ void security_fixup_ops (struct security_operations *ops)
         set_to_dummy_if_null(ops, inode_getxattr);
         set_to_dummy_if_null(ops, inode_listxattr);
         set_to_dummy_if_null(ops, inode_removexattr);
+        set_to_dummy_if_null(ops, inode_need_killpriv);
+        set_to_dummy_if_null(ops, inode_killpriv);
         set_to_dummy_if_null(ops, inode_xattr_getsuffix);
         set_to_dummy_if_null(ops, inode_getsecurity);
         set_to_dummy_if_null(ops, inode_setsecurity);
diff --git a/security/security.c b/security/security.c
index 5b1c034815a8..2e1b35dd2550 100644
--- a/security/security.c
+++ b/security/security.c
@@ -518,6 +518,16 @@ int security_inode_removexattr(struct dentry *dentry, char *name)
         return security_ops->inode_removexattr(dentry, name);
 }
 
+int security_inode_need_killpriv(struct dentry *dentry)
+{
+        return security_ops->inode_need_killpriv(dentry);
+}
+
+int security_inode_killpriv(struct dentry *dentry)
+{
+        return security_ops->inode_killpriv(dentry);
+}
+
 const char *security_inode_xattr_getsuffix(void)
 {
         return security_ops->inode_xattr_getsuffix();
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 83a535b7bc60..221def6a0b1d 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -2297,6 +2297,25 @@ static int selinux_inode_getattr(struct vfsmount *mnt, struct dentry *dentry)
         return dentry_has_perm(current, mnt, dentry, FILE__GETATTR);
 }
 
+static int selinux_inode_setotherxattr(struct dentry *dentry, char *name)
+{
+        if (!strncmp(name, XATTR_SECURITY_PREFIX,
+                     sizeof XATTR_SECURITY_PREFIX - 1)) {
+                if (!strcmp(name, XATTR_NAME_CAPS)) {
+                        if (!capable(CAP_SETFCAP))
+                                return -EPERM;
+                } else if (!capable(CAP_SYS_ADMIN)) {
+                        /* A different attribute in the security namespace.
+                           Restrict to administrator. */
+                        return -EPERM;
+                }
+        }
+
+        /* Not an attribute we recognize, so just check the
+           ordinary setattr permission. */
+        return dentry_has_perm(current, NULL, dentry, FILE__SETATTR);
+}
+
 static int selinux_inode_setxattr(struct dentry *dentry, char *name, void *value, size_t size, int flags)
 {
         struct task_security_struct *tsec = current->security;
@@ -2307,19 +2326,8 @@ static int selinux_inode_setxattr(struct dentry *dentry, char *name, void *value
         u32 newsid;
         int rc = 0;
 
-        if (strcmp(name, XATTR_NAME_SELINUX)) {
-                if (!strncmp(name, XATTR_SECURITY_PREFIX,
-                             sizeof XATTR_SECURITY_PREFIX - 1) &&
-                    !capable(CAP_SYS_ADMIN)) {
-                        /* A different attribute in the security namespace.
-                           Restrict to administrator. */
-                        return -EPERM;
-                }
-
-                /* Not an attribute we recognize, so just check the
-                   ordinary setattr permission. */
-                return dentry_has_perm(current, NULL, dentry, FILE__SETATTR);
-        }
+        if (strcmp(name, XATTR_NAME_SELINUX))
+                return selinux_inode_setotherxattr(dentry, name);
 
         sbsec = inode->i_sb->s_security;
         if (sbsec->behavior == SECURITY_FS_USE_MNTPOINT)
@@ -2393,20 +2401,8 @@ static int selinux_inode_listxattr (struct dentry *dentry)
 
 static int selinux_inode_removexattr (struct dentry *dentry, char *name)
 {
-        if (strcmp(name, XATTR_NAME_SELINUX)) {
-                if (!strncmp(name, XATTR_SECURITY_PREFIX,
-                             sizeof XATTR_SECURITY_PREFIX - 1) &&
-                    !capable(CAP_SYS_ADMIN)) {
-                        /* A different attribute in the security namespace.
-                           Restrict to administrator. */
-                        return -EPERM;
-                }
-
-                /* Not an attribute we recognize, so just check the
-                   ordinary setattr permission. Might want a separate
-                   permission for removexattr. */
-                return dentry_has_perm(current, NULL, dentry, FILE__SETATTR);
-        }
+        if (strcmp(name, XATTR_NAME_SELINUX))
+                return selinux_inode_setotherxattr(dentry, name);
 
         /* No one is allowed to remove a SELinux security label.
            You can change the label, but all data must be labeled. */
@@ -2464,6 +2460,16 @@ static int selinux_inode_listsecurity(struct inode *inode, char *buffer, size_t
         return len;
 }
 
+static int selinux_inode_need_killpriv(struct dentry *dentry)
+{
+        return secondary_ops->inode_need_killpriv(dentry);
+}
+
+static int selinux_inode_killpriv(struct dentry *dentry)
+{
+        return secondary_ops->inode_killpriv(dentry);
+}
+
 /* file security operations */
 
 static int selinux_revalidate_file_permission(struct file *file, int mask)
@@ -2882,6 +2888,12 @@ static int selinux_task_setnice(struct task_struct *p, int nice)
 
 static int selinux_task_setioprio(struct task_struct *p, int ioprio)
 {
+        int rc;
+
+        rc = secondary_ops->task_setioprio(p, ioprio);
+        if (rc)
+                return rc;
+
         return task_has_perm(current, p, PROCESS__SETSCHED);
 }
 
@@ -2911,6 +2923,12 @@ static int selinux_task_setrlimit(unsigned int resource, struct rlimit *new_rlim
 
 static int selinux_task_setscheduler(struct task_struct *p, int policy, struct sched_param *lp)
 {
+        int rc;
+
+        rc = secondary_ops->task_setscheduler(p, policy, lp);
+        if (rc)
+                return rc;
+
         return task_has_perm(current, p, PROCESS__SETSCHED);
 }
 
@@ -4830,6 +4848,8 @@ static struct security_operations selinux_ops = {
         .inode_getsecurity =            selinux_inode_getsecurity,
         .inode_setsecurity =            selinux_inode_setsecurity,
         .inode_listsecurity =           selinux_inode_listsecurity,
+        .inode_need_killpriv =          selinux_inode_need_killpriv,
+        .inode_killpriv =               selinux_inode_killpriv,
 
         .file_permission =              selinux_file_permission,
         .file_alloc_security =          selinux_file_alloc_security,