ptrace: use fsuid, fsgid, effective creds for fs access checks

commit caaee623 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>
[bwh: Backported to 3.16:
 - Update mm_access() calls in fs/proc/task_{,no}mmu.c too
 - Adjust context]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>

fs/proc/array.c

@@ -391,7 +391,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns,
 
 	state = *get_task_state(task);
 	vsize = eip = esp = 0;
-	permitted = ptrace_may_access(task, PTRACE_MODE_READ | PTRACE_MODE_NOAUDIT);
+	permitted = ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS | PTRACE_MODE_NOAUDIT);
 	mm = get_task_mm(task);
 	if (mm) {
 		vsize = task_vsize(mm);

fs/proc/base.c

@@ -207,7 +207,7 @@ static int proc_pid_cmdline(struct task_struct *task, char *buffer)
 
 static int proc_pid_auxv(struct task_struct *task, char *buffer)
 {
-	struct mm_struct *mm = mm_access(task, PTRACE_MODE_READ);
+	struct mm_struct *mm = mm_access(task, PTRACE_MODE_READ_FSCREDS);
 	int res = PTR_ERR(mm);
 	if (mm && !IS_ERR(mm)) {
 		unsigned int nwords = 0;
@@ -236,7 +236,8 @@ static int proc_pid_wchan(struct task_struct *task, char *buffer)
 
 	wchan = get_wchan(task);
 
-	if (wchan && ptrace_may_access(task, PTRACE_MODE_READ) && !lookup_symbol_name(wchan, symname))
+	if (wchan && ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS)
+			&& !lookup_symbol_name(wchan, symname))
 		return sprintf(buffer, "%s", symname);
 	else
 		return sprintf(buffer, "0");
@@ -248,7 +249,7 @@ static int lock_trace(struct task_struct *task)
 	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;
 	}
@@ -522,7 +523,7 @@ static int proc_fd_access_allowed(struct inode *inode)
 	 */
 	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;
@@ -557,7 +558,7 @@ static bool has_pid_permissions(struct pid_namespace *pid,
 		return true;
 	if (in_group_p(pid->pid_gid))
 		return true;
-	return ptrace_may_access(task, PTRACE_MODE_READ);
+	return ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS);
 }
 
 
@@ -672,7 +673,7 @@ static int __mem_open(struct inode *inode, struct file *file, unsigned int mode)
 	if (!task)
 		return -ESRCH;
 
-	mm = mm_access(task, mode);
+	mm = mm_access(task, mode | PTRACE_MODE_FSCREDS);
 	put_task_struct(task);
 
 	if (IS_ERR(mm))
@@ -1729,7 +1730,7 @@ static int map_files_d_revalidate(struct dentry *dentry, unsigned int flags)
 	if (!task)
 		goto out_notask;
 
-	mm = mm_access(task, PTRACE_MODE_READ);
+	mm = mm_access(task, PTRACE_MODE_READ_FSCREDS);
 	if (IS_ERR_OR_NULL(mm))
 		goto out;
 
@@ -1864,7 +1865,7 @@ static struct dentry *proc_map_files_lookup(struct inode *dir,
 		goto out;
 
 	result = -EACCES;
-	if (!ptrace_may_access(task, PTRACE_MODE_READ))
+	if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS))
 		goto out_put_task;
 
 	result = -ENOENT;
@@ -1921,7 +1922,7 @@ proc_map_files_readdir(struct file *file, struct dir_context *ctx)
 		goto out;
 
 	ret = -EACCES;
-	if (!ptrace_may_access(task, PTRACE_MODE_READ))
+	if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS))
 		goto out_put_task;
 
 	ret = 0;
@@ -2400,7 +2401,7 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole)
 	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;
 	}

fs/proc/namespaces.c

@@ -119,7 +119,7 @@ static void *proc_ns_follow_link(struct dentry *dentry, struct nameidata *nd)
 	if (!task)
 		goto out;
 
-	if (!ptrace_may_access(task, PTRACE_MODE_READ))
+	if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS))
 		goto out_put_task;
 
 	ns_path.dentry = proc_ns_get_dentry(sb, task, ei->ns.ns_ops);
@@ -152,7 +152,7 @@ static int proc_ns_readlink(struct dentry *dentry, char __user *buffer, int bufl
 	if (!task)
 		goto out;
 
-	if (!ptrace_may_access(task, PTRACE_MODE_READ))
+	if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS))
 		goto out_put_task;
 
 	res = -ENOENT;

fs/proc/task_mmu.c

@@ -165,7 +165,7 @@ static void *m_start(struct seq_file *m, loff_t *pos)
 	if (!priv->task)
 		return ERR_PTR(-ESRCH);
 
-	mm = mm_access(priv->task, PTRACE_MODE_READ);
+	mm = mm_access(priv->task, PTRACE_MODE_READ_FSCREDS);
 	if (!mm || IS_ERR(mm))
 		return mm;
 	down_read(&mm->mmap_sem);
@@ -1188,7 +1188,7 @@ static ssize_t pagemap_read(struct file *file, char __user *buf,
 	if (!pm.buffer)
 		goto out_task;
 
-	mm = mm_access(task, PTRACE_MODE_READ);
+	mm = mm_access(task, PTRACE_MODE_READ_FSCREDS);
 	ret = PTR_ERR(mm);
 	if (!mm || IS_ERR(mm))
 		goto out_free;

fs/proc/task_nommu.c

@@ -216,7 +216,7 @@ static void *m_start(struct seq_file *m, loff_t *pos)
 	if (!priv->task)
 		return ERR_PTR(-ESRCH);
 
-	mm = mm_access(priv->task, PTRACE_MODE_READ);
+	mm = mm_access(priv->task, PTRACE_MODE_READ_FSCREDS);
 	if (!mm || IS_ERR(mm)) {
 		put_task_struct(priv->task);
 		priv->task = NULL;

include/linux/ptrace.h

@@ -56,7 +56,29 @@ extern void exit_ptrace(struct task_struct *tracer);
 #define PTRACE_MODE_READ	0x01
 #define PTRACE_MODE_ATTACH	0x02
 #define PTRACE_MODE_NOAUDIT	0x04
-/* Returns true on success, false on denial. */
+#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)
+
+/**
+ * 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)

kernel/events/core.c

@@ -3247,7 +3247,7 @@ 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;

kernel/futex.c

@@ -2797,7 +2797,7 @@ SYSCALL_DEFINE3(get_robust_list, int, pid,
 	}
 
 	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;

kernel/futex_compat.c

@@ -155,7 +155,7 @@ COMPAT_SYSCALL_DEFINE3(get_robust_list, int, pid,
 	}
 
 	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;

kernel/kcmp.c

@@ -122,8 +122,8 @@ SYSCALL_DEFINE5(kcmp, pid_t, pid1, pid_t, pid2, int, type,
 			&task2->signal->cred_guard_mutex);
 	if (ret)
 		goto err;
-	if (!ptrace_may_access(task1, PTRACE_MODE_READ) ||
-	    !ptrace_may_access(task2, PTRACE_MODE_READ)) {
+	if (!ptrace_may_access(task1, PTRACE_MODE_READ_REALCREDS) ||
+	    !ptrace_may_access(task2, PTRACE_MODE_READ_REALCREDS)) {
 		ret = -EPERM;
 		goto err_unlock;
 	}

kernel/ptrace.c

@@ -231,6 +231,14 @@ static int ptrace_has_cap(struct user_namespace *ns, unsigned int mode)
 static int __ptrace_may_access(struct task_struct *task, unsigned int mode)
 {
 	const struct cred *cred = current_cred(), *tcred;
+	int dumpable = 0;
+	kuid_t caller_uid;
+	kgid_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
@@ -240,18 +248,33 @@ static int __ptrace_may_access(struct task_struct *task, unsigned int mode)
 	 * 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 (uid_eq(cred->uid, tcred->euid) &&
-	    uid_eq(cred->uid, tcred->suid) &&
-	    uid_eq(cred->uid, tcred->uid)  &&
-	    gid_eq(cred->gid, tcred->egid) &&
-	    gid_eq(cred->gid, tcred->sgid) &&
-	    gid_eq(cred->gid, tcred->gid))
+	if (uid_eq(caller_uid, tcred->euid) &&
+	    uid_eq(caller_uid, tcred->suid) &&
+	    uid_eq(caller_uid, tcred->uid)  &&
+	    gid_eq(caller_gid, tcred->egid) &&
+	    gid_eq(caller_gid, tcred->sgid) &&
+	    gid_eq(caller_gid, tcred->gid))
 		goto ok;
 	if (ptrace_has_cap(tcred->user_ns, mode))
 		goto ok;
@@ -318,7 +341,7 @@ static int ptrace_attach(struct task_struct *task, long request,
 		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;

mm/process_vm_access.c

@@ -197,7 +197,7 @@ static ssize_t process_vm_rw_core(pid_t pid, struct iov_iter *iter,
 		goto free_proc_pages;
 	}
 
-	mm = mm_access(task, PTRACE_MODE_ATTACH);
+	mm = mm_access(task, PTRACE_MODE_ATTACH_REALCREDS);
 	if (!mm || IS_ERR(mm)) {
 		rc = IS_ERR(mm) ? PTR_ERR(mm) : -ESRCH;
 		/*

security/commoncap.c

@@ -142,12 +142,17 @@ int cap_ptrace_access_check(struct task_struct *child, unsigned int mode)
 {
 	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_ns == child_cred->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_ns, CAP_SYS_PTRACE))
 		goto out;