1 files changed, 66 insertions, 27 deletions

diff --git a/kernel/seccomp.c b/kernel/seccomp.c
index 3ee59ce0a323..53a7d1512dd7 100644
--- a/kernel/seccomp.c
+++ b/kernel/seccomp.c
@@ -38,7 +38,7 @@
 #include <linux/filter.h>
 #include <linux/pid.h>
 #include <linux/ptrace.h>
-#include <linux/security.h>
+#include <linux/capability.h>
 #include <linux/tracehook.h>
 #include <linux/uaccess.h>
 #include <linux/anon_inodes.h>
@@ -196,6 +196,10 @@ struct seccomp_filter {
  */
 static void populate_seccomp_data(struct seccomp_data *sd)
 {
+	/*
+	 * Instead of using current_pt_reg(), we're already doing the work
+	 * to safely fetch "current", so just use "task" everywhere below.
+	 */
 	struct task_struct *task = current;
 	struct pt_regs *regs = task_pt_regs(task);
 	unsigned long args[6];
@@ -554,8 +558,7 @@ static struct seccomp_filter *seccomp_prepare_filter(struct sock_fprog *fprog)
 	 * behavior of privileged children.
 	 */
 	if (!task_no_new_privs(current) &&
-	    security_capable(current_cred(), current_user_ns(),
-				     CAP_SYS_ADMIN, CAP_OPT_NOAUDIT) != 0)
+			!ns_capable_noaudit(current_user_ns(), CAP_SYS_ADMIN))
 		return ERR_PTR(-EACCES);
 
 	/* Allocate a new seccomp_filter */
@@ -910,7 +913,7 @@ out:
 	if (flags & SECCOMP_USER_NOTIF_FLAG_CONTINUE)
 		return 0;
 
-	syscall_set_return_value(current, task_pt_regs(current),
+	syscall_set_return_value(current, current_pt_regs(),
 				 err, ret);
 	return -1;
 }
@@ -943,13 +946,13 @@ static int __seccomp_filter(int this_syscall, const struct seccomp_data *sd,
 		/* Set low-order bits as an errno, capped at MAX_ERRNO. */
 		if (data > MAX_ERRNO)
 			data = MAX_ERRNO;
-		syscall_set_return_value(current, task_pt_regs(current),
+		syscall_set_return_value(current, current_pt_regs(),
 					 -data, 0);
 		goto skip;
 
 	case SECCOMP_RET_TRAP:
 		/* Show the handler the original registers. */
-		syscall_rollback(current, task_pt_regs(current));
+		syscall_rollback(current, current_pt_regs());
 		/* Let the filter pass back 16 bits of data. */
 		seccomp_send_sigsys(this_syscall, data);
 		goto skip;
@@ -962,7 +965,7 @@ static int __seccomp_filter(int this_syscall, const struct seccomp_data *sd,
 		/* ENOSYS these calls if there is no tracer attached. */
 		if (!ptrace_event_enabled(current, PTRACE_EVENT_SECCOMP)) {
 			syscall_set_return_value(current,
-						 task_pt_regs(current),
+						 current_pt_regs(),
 						 -ENOSYS, 0);
 			goto skip;
 		}
@@ -982,7 +985,7 @@ static int __seccomp_filter(int this_syscall, const struct seccomp_data *sd,
 		if (fatal_signal_pending(current))
 			goto skip;
 		/* Check if the tracer forced the syscall to be skipped. */
-		this_syscall = syscall_get_nr(current, task_pt_regs(current));
+		this_syscall = syscall_get_nr(current, current_pt_regs());
 		if (this_syscall < 0)
 			goto skip;
 
@@ -1020,20 +1023,20 @@ static int __seccomp_filter(int this_syscall, const struct seccomp_data *sd,
 	default:
 		seccomp_log(this_syscall, SIGSYS, action, true);
 		/* Dump core only if this is the last remaining thread. */
-		if (action == SECCOMP_RET_KILL_PROCESS ||
+		if (action != SECCOMP_RET_KILL_THREAD ||
 		    get_nr_threads(current) == 1) {
 			kernel_siginfo_t info;
 
 			/* Show the original registers in the dump. */
-			syscall_rollback(current, task_pt_regs(current));
+			syscall_rollback(current, current_pt_regs());
 			/* Trigger a manual coredump since do_exit skips it. */
 			seccomp_init_siginfo(&info, this_syscall, data);
 			do_coredump(&info);
 		}
-		if (action == SECCOMP_RET_KILL_PROCESS)
-			do_group_exit(SIGSYS);
-		else
+		if (action == SECCOMP_RET_KILL_THREAD)
 			do_exit(SIGSYS);
+		else
+			do_group_exit(SIGSYS);
 	}
 
 	unreachable();
@@ -1060,7 +1063,7 @@ int __secure_computing(const struct seccomp_data *sd)
 		return 0;
 
 	this_syscall = sd ? sd->nr :
-		syscall_get_nr(current, task_pt_regs(current));
+		syscall_get_nr(current, current_pt_regs());
 
 	switch (mode) {
 	case SECCOMP_MODE_STRICT:
@@ -1109,13 +1112,18 @@ out:
 }
 
 #ifdef CONFIG_SECCOMP_FILTER
-static int seccomp_notify_release(struct inode *inode, struct file *file)
+static void seccomp_notify_free(struct seccomp_filter *filter)
+{
+	kfree(filter->notif);
+	filter->notif = NULL;
+}
+
+static void seccomp_notify_detach(struct seccomp_filter *filter)
 {
-	struct seccomp_filter *filter = file->private_data;
 	struct seccomp_knotif *knotif;
 
 	if (!filter)
-		return 0;
+		return;
 
 	mutex_lock(&filter->notify_lock);
 
@@ -1139,9 +1147,15 @@ static int seccomp_notify_release(struct inode *inode, struct file *file)
 		complete(&knotif->ready);
 	}
 
-	kfree(filter->notif);
-	filter->notif = NULL;
+	seccomp_notify_free(filter);
 	mutex_unlock(&filter->notify_lock);
+}
+
+static int seccomp_notify_release(struct inode *inode, struct file *file)
+{
+	struct seccomp_filter *filter = file->private_data;
+
+	seccomp_notify_detach(filter);
 	__put_seccomp_filter(filter);
 	return 0;
 }
@@ -1461,13 +1475,7 @@ static const struct file_operations seccomp_notify_ops = {
 
 static struct file *init_listener(struct seccomp_filter *filter)
 {
-	struct file *ret = ERR_PTR(-EBUSY);
-	struct seccomp_filter *cur;
-
-	for (cur = current->seccomp.filter; cur; cur = cur->prev) {
-		if (cur->notif)
-			goto out;
-	}
+	struct file *ret;
 
 	ret = ERR_PTR(-ENOMEM);
 	filter->notif = kzalloc(sizeof(*(filter->notif)), GFP_KERNEL);
@@ -1488,11 +1496,36 @@ static struct file *init_listener(struct seccomp_filter *filter)
 
 out_notif:
 	if (IS_ERR(ret))
-		kfree(filter->notif);
+		seccomp_notify_free(filter);
 out:
 	return ret;
 }
 
+/*
+ * Does @new_child have a listener while an ancestor also has a listener?
+ * If so, we'll want to reject this filter.
+ * This only has to be tested for the current process, even in the TSYNC case,
+ * because TSYNC installs @child with the same parent on all threads.
+ * Note that @new_child is not hooked up to its parent at this point yet, so
+ * we use current->seccomp.filter.
+ */
+static bool has_duplicate_listener(struct seccomp_filter *new_child)
+{
+	struct seccomp_filter *cur;
+
+	/* must be protected against concurrent TSYNC */
+	lockdep_assert_held(&current->sighand->siglock);
+
+	if (!new_child->notif)
+		return false;
+	for (cur = current->seccomp.filter; cur; cur = cur->prev) {
+		if (cur->notif)
+			return true;
+	}
+
+	return false;
+}
+
 /**
  * seccomp_set_mode_filter: internal function for setting seccomp filter
  * @flags:  flags to change filter behavior
@@ -1564,6 +1597,11 @@ static long seccomp_set_mode_filter(unsigned int flags,
 	if (!seccomp_may_assign_mode(seccomp_mode))
 		goto out;
 
+	if (has_duplicate_listener(prepared)) {
+		ret = -EBUSY;
+		goto out;
+	}
+
 	ret = seccomp_attach_filter(flags, prepared);
 	if (ret)
 		goto out;
@@ -1581,6 +1619,7 @@ out_put_fd:
 			listener_f->private_data = NULL;
 			fput(listener_f);
 			put_unused_fd(listener);
+			seccomp_notify_detach(prepared);
 		} else {
 			fd_install(listener, listener_f);
 			ret = listener;