[PATCH 3.2 101/106] ptrace: use fsuid, fsgid, effective creds for fs access checks

[Ben Hutchings <ben@decadent.org.uk>]

3.2.93-rc1 review patch.  If anyone has any objections, please let me know.

------------------

From: Jann Horn <jann@thejh.net>

commit caaee6234d05a58c5b4d05e7bf766131b810a657 upstream.

By checking the effective credentials instead of the real UID / permitted
capabilities, ensure that the calling process actually intended to use its
credentials.

To ensure that all ptrace checks use the correct caller credentials (e.g.
in case out-of-tree code or newly added code omits the PTRACE_MODE_*CREDS
flag), use two new flags and require one of them to be set.

The problem was that when a privileged task had temporarily dropped its
privileges, e.g.  by calling setreuid(0, user_uid), with the intent to
perform following syscalls with the credentials of a user, it still passed
ptrace access checks that the user would not be able to pass.

While an attacker should not be able to convince the privileged task to
perform a ptrace() syscall, this is a problem because the ptrace access
check is reused for things in procfs.

In particular, the following somewhat interesting procfs entries only rely
on ptrace access checks:

 /proc/$pid/stat - uses the check for determining whether pointers
     should be visible, useful for bypassing ASLR
 /proc/$pid/maps - also useful for bypassing ASLR
 /proc/$pid/cwd - useful for gaining access to restricted
     directories that contain files with lax permissions, e.g. in
     this scenario:
     lrwxrwxrwx root root /proc/13020/cwd -> /root/foobar
     drwx------ root root /root
     drwxr-xr-x root root /root/foobar
     -rw-r--r-- root root /root/foobar/secret

Therefore, on a system where a root-owned mode 6755 binary changes its
effective credentials as described and then dumps a user-specified file,
this could be used by an attacker to reveal the memory layout of root's
processes or reveal the contents of files he is not allowed to access
(through /proc/$pid/cwd).

[akpm@linux-foundation.org: fix warning]
Signed-off-by: Jann Horn <jann@thejh.net>
Acked-by: Kees Cook <keescook@chromium.org>
Cc: Casey Schaufler <casey@schaufler-ca.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Morris <james.l.morris@oracle.com>
Cc: "Serge E. Hallyn" <serge.hallyn@ubuntu.com>
Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Willy Tarreau <w@1wt.eu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[bwh: Backported to 3.2:
 - Drop changes to kcmp, procfs map_files, procfs has_pid_permissions()
 - Keep using uid_t, gid_t and == operator for IDs
 - Adjust context]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -380,7 +380,7 @@ static int do_task_stat(struct seq_file
 
 	state = *get_task_state(task);
 	vsize = eip = esp = 0;
-	permitted = ptrace_may_access(task, PTRACE_MODE_READ);
+	permitted = ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS);
 	mm = get_task_mm(task);
 	if (mm) {
 		vsize = task_vsize(mm);
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -216,7 +216,7 @@ static struct mm_struct *mm_access(struc
 
 struct mm_struct *mm_for_maps(struct task_struct *task)
 {
-	return mm_access(task, PTRACE_MODE_READ);
+	return mm_access(task, PTRACE_MODE_READ_FSCREDS);
 }
 
 static int proc_pid_cmdline(struct task_struct *task, char * buffer)
@@ -288,7 +288,7 @@ static int proc_pid_wchan(struct task_st
 	wchan = get_wchan(task);
 
 	if (lookup_symbol_name(wchan, symname) < 0)
-		if (!ptrace_may_access(task, PTRACE_MODE_READ))
+		if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS))
 			return 0;
 		else
 			return sprintf(buffer, "%lu", wchan);
@@ -302,7 +302,7 @@ static int lock_trace(struct task_struct
 	int err = mutex_lock_killable(&task->signal->cred_guard_mutex);
 	if (err)
 		return err;
-	if (!ptrace_may_access(task, PTRACE_MODE_ATTACH)) {
+	if (!ptrace_may_access(task, PTRACE_MODE_ATTACH_FSCREDS)) {
 		mutex_unlock(&task->signal->cred_guard_mutex);
 		return -EPERM;
 	}
@@ -544,7 +544,7 @@ static int proc_fd_access_allowed(struct
 	 */
 	task = get_proc_task(inode);
 	if (task) {
-		allowed = ptrace_may_access(task, PTRACE_MODE_READ);
+		allowed = ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS);
 		put_task_struct(task);
 	}
 	return allowed;
@@ -769,7 +769,7 @@ static int mem_open(struct inode* inode,
 	if (!task)
 		return -ESRCH;
 
-	mm = mm_access(task, PTRACE_MODE_ATTACH);
+	mm = mm_access(task, PTRACE_MODE_ATTACH | PTRACE_MODE_FSCREDS);
 	put_task_struct(task);
 
 	if (IS_ERR(mm))
@@ -2627,7 +2627,7 @@ static int do_io_accounting(struct task_
 	if (result)
 		return result;
 
-	if (!ptrace_may_access(task, PTRACE_MODE_READ)) {
+	if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS)) {
 		result = -EACCES;
 		goto out_unlock;
 	}
--- a/fs/proc/namespaces.c
+++ b/fs/proc/namespaces.c
@@ -91,7 +91,7 @@ static int proc_ns_dir_readdir(struct fi
 		goto out_no_task;
 
 	ret = -EPERM;
-	if (!ptrace_may_access(task, PTRACE_MODE_READ))
+	if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS))
 		goto out;
 
 	ret = 0;
@@ -154,7 +154,7 @@ static struct dentry *proc_ns_dir_lookup
 		goto out_no_task;
 
 	error = ERR_PTR(-EPERM);
-	if (!ptrace_may_access(task, PTRACE_MODE_READ))
+	if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS))
 		goto out;
 
 	last = &ns_entries[ARRAY_SIZE(ns_entries) - 1];
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -130,9 +130,31 @@ extern void __ptrace_unlink(struct task_
 extern void exit_ptrace(struct task_struct *tracer);
 #define PTRACE_MODE_READ   1
 #define PTRACE_MODE_ATTACH 2
+#define PTRACE_MODE_FSCREDS 0x08
+#define PTRACE_MODE_REALCREDS 0x10
+
+/* shorthands for READ/ATTACH and FSCREDS/REALCREDS combinations */
+#define PTRACE_MODE_READ_FSCREDS (PTRACE_MODE_READ | PTRACE_MODE_FSCREDS)
+#define PTRACE_MODE_READ_REALCREDS (PTRACE_MODE_READ | PTRACE_MODE_REALCREDS)
+#define PTRACE_MODE_ATTACH_FSCREDS (PTRACE_MODE_ATTACH | PTRACE_MODE_FSCREDS)
+#define PTRACE_MODE_ATTACH_REALCREDS (PTRACE_MODE_ATTACH | PTRACE_MODE_REALCREDS)
+
 /* Returns 0 on success, -errno on denial. */
 extern int __ptrace_may_access(struct task_struct *task, unsigned int mode);
-/* Returns true on success, false on denial. */
+/**
+ * ptrace_may_access - check whether the caller is permitted to access
+ * a target task.
+ * @task: target task
+ * @mode: selects type of access and caller credentials
+ *
+ * Returns true on success, false on denial.
+ *
+ * One of the flags PTRACE_MODE_FSCREDS and PTRACE_MODE_REALCREDS must
+ * be set in @mode to specify whether the access was requested through
+ * a filesystem syscall (should use effective capabilities and fsuid
+ * of the caller) or through an explicit syscall such as
+ * process_vm_writev or ptrace (and should use the real credentials).
+ */
 extern bool ptrace_may_access(struct task_struct *task, unsigned int mode);
 
 static inline int ptrace_reparented(struct task_struct *child)
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -3004,7 +3004,7 @@ find_lively_task_by_vpid(pid_t vpid)
 
 	/* Reuse ptrace permission checks for now. */
 	err = -EACCES;
-	if (!ptrace_may_access(task, PTRACE_MODE_READ))
+	if (!ptrace_may_access(task, PTRACE_MODE_READ_REALCREDS))
 		goto errout;
 
 	return task;
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -2627,7 +2627,7 @@ SYSCALL_DEFINE3(get_robust_list, int, pi
 	}
 
 	ret = -EPERM;
-	if (!ptrace_may_access(p, PTRACE_MODE_READ))
+	if (!ptrace_may_access(p, PTRACE_MODE_READ_REALCREDS))
 		goto err_unlock;
 
 	head = p->robust_list;
--- a/kernel/futex_compat.c
+++ b/kernel/futex_compat.c
@@ -154,7 +154,7 @@ compat_sys_get_robust_list(int pid, comp
 	}
 
 	ret = -EPERM;
-	if (!ptrace_may_access(p, PTRACE_MODE_READ))
+	if (!ptrace_may_access(p, PTRACE_MODE_READ_REALCREDS))
 		goto err_unlock;
 
 	head = p->compat_robust_list;
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -220,6 +220,14 @@ int ptrace_check_attach(struct task_stru
 int __ptrace_may_access(struct task_struct *task, unsigned int mode)
 {
 	const struct cred *cred = current_cred(), *tcred;
+	int dumpable = 0;
+	uid_t caller_uid;
+	gid_t caller_gid;
+
+	if (!(mode & PTRACE_MODE_FSCREDS) == !(mode & PTRACE_MODE_REALCREDS)) {
+		WARN(1, "denying ptrace access check without PTRACE_MODE_*CREDS\n");
+		return -EPERM;
+	}
 
 	/* May we inspect the given task?
 	 * This check is used both for attaching with ptrace
@@ -229,19 +237,34 @@ int __ptrace_may_access(struct task_stru
 	 * because setting up the necessary parent/child relationship
 	 * or halting the specified task is impossible.
 	 */
-	int dumpable = 0;
+
 	/* Don't let security modules deny introspection */
 	if (same_thread_group(task, current))
 		return 0;
 	rcu_read_lock();
+	if (mode & PTRACE_MODE_FSCREDS) {
+		caller_uid = cred->fsuid;
+		caller_gid = cred->fsgid;
+	} else {
+		/*
+		 * Using the euid would make more sense here, but something
+		 * in userland might rely on the old behavior, and this
+		 * shouldn't be a security problem since
+		 * PTRACE_MODE_REALCREDS implies that the caller explicitly
+		 * used a syscall that requests access to another process
+		 * (and not a filesystem syscall to procfs).
+		 */
+		caller_uid = cred->uid;
+		caller_gid = cred->gid;
+	}
 	tcred = __task_cred(task);
 	if (cred->user->user_ns == tcred->user->user_ns &&
-	    (cred->uid == tcred->euid &&
-	     cred->uid == tcred->suid &&
-	     cred->uid == tcred->uid  &&
-	     cred->gid == tcred->egid &&
-	     cred->gid == tcred->sgid &&
-	     cred->gid == tcred->gid))
+	    (caller_uid == tcred->euid &&
+	     caller_uid == tcred->suid &&
+	     caller_uid == tcred->uid  &&
+	     caller_gid == tcred->egid &&
+	     caller_gid == tcred->sgid &&
+	     caller_gid == tcred->gid))
 		goto ok;
 	if (ns_capable(tcred->user->user_ns, CAP_SYS_PTRACE))
 		goto ok;
@@ -308,7 +331,7 @@ static int ptrace_attach(struct task_str
 		goto out;
 
 	task_lock(task);
-	retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH);
+	retval = __ptrace_may_access(task, PTRACE_MODE_ATTACH_REALCREDS);
 	task_unlock(task);
 	if (retval)
 		goto unlock_creds;
--- a/mm/process_vm_access.c
+++ b/mm/process_vm_access.c
@@ -299,7 +299,7 @@ static ssize_t process_vm_rw_core(pid_t
 	}
 
 	task_lock(task);
-	if (__ptrace_may_access(task, PTRACE_MODE_ATTACH)) {
+	if (__ptrace_may_access(task, PTRACE_MODE_ATTACH_REALCREDS)) {
 		task_unlock(task);
 		rc = -EPERM;
 		goto put_task_struct;
--- a/security/commoncap.c
+++ b/security/commoncap.c
@@ -141,12 +141,17 @@ int cap_ptrace_access_check(struct task_
 {
 	int ret = 0;
 	const struct cred *cred, *child_cred;
+	const kernel_cap_t *caller_caps;
 
 	rcu_read_lock();
 	cred = current_cred();
 	child_cred = __task_cred(child);
+	if (mode & PTRACE_MODE_FSCREDS)
+		caller_caps = &cred->cap_effective;
+	else
+		caller_caps = &cred->cap_permitted;
 	if (cred->user->user_ns == child_cred->user->user_ns &&
-	    cap_issubset(child_cred->cap_permitted, cred->cap_permitted))
+	    cap_issubset(child_cred->cap_permitted, *caller_caps))
 		goto out;
 	if (ns_capable(child_cred->user->user_ns, CAP_SYS_PTRACE))
 		goto out;