[PATCH 3.16 17/19] vfs: Commit to never having exectuables on proc and sysfs.

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

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

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

From: "Eric W. Biederman" <ebiederm@xmission.com>

commit 22f6b4d34fcf039c63a94e7670e0da24f8575a5a upstream.

Today proc and sysfs do not contain any executable files.  Several
applications today mount proc or sysfs without noexec and nosuid and
then depend on there being no exectuables files on proc or sysfs.
Having any executable files show on proc or sysfs would cause
a user space visible regression, and most likely security problems.

Therefore commit to never allowing executables on proc and sysfs by
adding a new flag to mark them as filesystems without executables and
enforce that flag.

Test the flag where MNT_NOEXEC is tested today, so that the only user
visible effect will be that exectuables will be treated as if the
execute bit is cleared.

The filesystems proc and sysfs do not currently incoporate any
executable files so this does not result in any user visible effects.

This makes it unnecessary to vet changes to proc and sysfs tightly for
adding exectuable files or changes to chattr that would modify
existing files, as no matter what the individual file say they will
not be treated as exectuable files by the vfs.

Not having to vet changes to closely is important as without this we
are only one proc_create call (or another goof up in the
implementation of notify_change) from having problematic executables
on proc.  Those mistakes are all too easy to make and would create
a situation where there are security issues or the assumptions of
some program having to be broken (and cause userspace regressions).

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
[bwh: Backported to 3.16: we don't have super_block::s_iflags; use
 file_system_type::fs_flags instead]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -98,6 +98,12 @@ static inline void put_binfmt(struct lin
 	module_put(fmt->module);
 }
 
+bool path_noexec(const struct path *path)
+{
+	return (path->mnt->mnt_flags & MNT_NOEXEC) ||
+	       (path->mnt->mnt_sb->s_type->fs_flags & FS_NOEXEC);
+}
+
 #ifdef CONFIG_USELIB
 /*
  * Note that a shared library must be both readable and executable due to
@@ -132,7 +138,7 @@ SYSCALL_DEFINE1(uselib, const char __use
 		goto exit;
 
 	error = -EACCES;
-	if (file->f_path.mnt->mnt_flags & MNT_NOEXEC)
+	if (path_noexec(&file->f_path))
 		goto exit;
 
 	fsnotify_open(file);
@@ -773,7 +779,7 @@ static struct file *do_open_exec(struct
 	if (!S_ISREG(file_inode(file)->i_mode))
 		goto exit;
 
-	if (file->f_path.mnt->mnt_flags & MNT_NOEXEC)
+	if (path_noexec(&file->f_path))
 		goto exit;
 
 	fsnotify_open(file);
--- a/fs/open.c
+++ b/fs/open.c
@@ -359,7 +359,7 @@ retry:
 		 * with the "noexec" flag.
 		 */
 		res = -EACCES;
-		if (path.mnt->mnt_flags & MNT_NOEXEC)
+		if (path_noexec(&path))
 			goto out_path_release;
 	}
 
--- a/fs/proc/root.c
+++ b/fs/proc/root.c
@@ -161,7 +161,7 @@ static struct file_system_type proc_fs_t
 	.name		= "proc",
 	.mount		= proc_mount,
 	.kill_sb	= proc_kill_sb,
-	.fs_flags	= FS_USERNS_VISIBLE | FS_USERNS_MOUNT,
+	.fs_flags	= FS_USERNS_VISIBLE | FS_USERNS_MOUNT | FS_NOEXEC,
 };
 
 void __init proc_root_init(void)
--- a/fs/sysfs/mount.c
+++ b/fs/sysfs/mount.c
@@ -40,6 +40,7 @@ static struct dentry *sysfs_mount(struct
 				SYSFS_MAGIC, &new_sb, ns);
 	if (IS_ERR(root) || !new_sb)
 		kobj_ns_drop(KOBJ_NS_TYPE_NET, ns);
+
 	return root;
 }
 
@@ -55,7 +56,7 @@ static struct file_system_type sysfs_fs_
 	.name		= "sysfs",
 	.mount		= sysfs_mount,
 	.kill_sb	= sysfs_kill_sb,
-	.fs_flags	= FS_USERNS_VISIBLE | FS_USERNS_MOUNT,
+	.fs_flags	= FS_USERNS_VISIBLE | FS_USERNS_MOUNT | FS_NOEXEC,
 };
 
 int __init sysfs_init(void)
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1767,6 +1767,7 @@ struct file_system_type {
 #define FS_USERNS_MOUNT		8	/* Can be mounted by userns root */
 #define FS_USERNS_DEV_MOUNT	16 /* A userns mount does not imply MNT_NODEV */
 #define FS_USERNS_VISIBLE	32	/* FS must already be visible */
+#define FS_NOEXEC		64	/* Ignore executables on this fs */
 #define FS_RENAME_DOES_D_MOVE	32768	/* FS will handle d_move() during rename() internally. */
 	struct dentry *(*mount) (struct file_system_type *, int,
 		       const char *, void *);
@@ -2782,4 +2783,6 @@ static inline bool dir_relax(struct inod
 	return !IS_DEADDIR(inode);
 }
 
+extern bool path_noexec(const struct path *path);
+
 #endif /* _LINUX_FS_H */
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -1646,8 +1646,7 @@ static int prctl_set_mm_exe_file(struct
 	 * overall picture.
 	 */
 	err = -EACCES;
-	if (!S_ISREG(inode->i_mode)	||
-	    exe.file->f_path.mnt->mnt_flags & MNT_NOEXEC)
+	if (!S_ISREG(inode->i_mode) || path_noexec(&exe.file->f_path))
 		goto exit;
 
 	err = inode_permission(inode, MAY_EXEC);
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1250,7 +1250,7 @@ unsigned long do_mmap_pgoff(struct file
 	 *  mounted, in which case we dont add PROT_EXEC.)
 	 */
 	if ((prot & PROT_READ) && (current->personality & READ_IMPLIES_EXEC))
-		if (!(file && (file->f_path.mnt->mnt_flags & MNT_NOEXEC)))
+		if (!(file && path_noexec(&file->f_path)))
 			prot |= PROT_EXEC;
 
 	if (!len)
@@ -1322,7 +1322,7 @@ unsigned long do_mmap_pgoff(struct file
 		case MAP_PRIVATE:
 			if (!(file->f_mode & FMODE_READ))
 				return -EACCES;
-			if (file->f_path.mnt->mnt_flags & MNT_NOEXEC) {
+			if (path_noexec(&file->f_path)) {
 				if (vm_flags & VM_EXEC)
 					return -EPERM;
 				vm_flags &= ~VM_MAYEXEC;
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -1043,7 +1043,7 @@ static int validate_mmap_request(struct
 
 		/* handle executable mappings and implied executable
 		 * mappings */
-		if (file->f_path.mnt->mnt_flags & MNT_NOEXEC) {
+		if (path_noexec(&file->f_path)) {
 			if (prot & PROT_EXEC)
 				return -EPERM;
 		} else if ((prot & PROT_READ) && !(prot & PROT_EXEC)) {
--- a/security/security.c
+++ b/security/security.c
@@ -728,7 +728,7 @@ static inline unsigned long mmap_prot(st
 	 * ditto if it's not on noexec mount, except that on !MMU we need
 	 * BDI_CAP_EXEC_MMAP (== VM_MAYEXEC) in this case
 	 */
-	if (!(file->f_path.mnt->mnt_flags & MNT_NOEXEC)) {
+	if (!path_noexec(&file->f_path)) {
 #ifndef CONFIG_MMU
 		unsigned long caps = 0;
 		struct address_space *mapping = file->f_mapping;