ovl: check the capability before cred overridden

We found that it return success when we set IMMUTABLE_FL flag to a file in
docker even though the docker didn't have the capability
CAP_LINUX_IMMUTABLE.

The commit d1d04ef8 ("ovl: stack file ops") and dab5ca8f ("ovl: add
lsattr/chattr support") implemented chattr operations on a regular overlay
file. ovl_real_ioctl() overridden the current process's subjective
credentials with ofs->creator_cred which have the capability
CAP_LINUX_IMMUTABLE so that it will return success in
vfs_ioctl()->cap_capable().

Fix this by checking the capability before cred overridden. And here we
only care about APPEND_FL and IMMUTABLE_FL, so get these information from
inode.

[SzM: move check and call to underlying fs inside inode locked region to
prevent two such calls from racing with each other]
Signed-off-by: Jiufei Xue <jiufei.xue@linux.alibaba.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>

fs/overlayfs/file.c

@@ -11,6 +11,7 @@
 #include <linux/mount.h>
 #include <linux/xattr.h>
 #include <linux/uio.h>
+#include <linux/uaccess.h>
 #include "overlayfs.h"
 
 static char ovl_whatisit(struct inode *inode, struct inode *realinode)
@@ -408,34 +409,76 @@ static long ovl_real_ioctl(struct file *file, unsigned int cmd,
 	return ret;
 }
 
-static long ovl_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+static unsigned int ovl_get_inode_flags(struct inode *inode)
+{
+	unsigned int flags = READ_ONCE(inode->i_flags);
+	unsigned int ovl_iflags = 0;
+
+	if (flags & S_SYNC)
+		ovl_iflags |= FS_SYNC_FL;
+	if (flags & S_APPEND)
+		ovl_iflags |= FS_APPEND_FL;
+	if (flags & S_IMMUTABLE)
+		ovl_iflags |= FS_IMMUTABLE_FL;
+	if (flags & S_NOATIME)
+		ovl_iflags |= FS_NOATIME_FL;
+
+	return ovl_iflags;
+}
+
+static long ovl_ioctl_set_flags(struct file *file, unsigned long arg)
 {
 	long ret;
 	struct inode *inode = file_inode(file);
+	unsigned int flags;
+	unsigned int old_flags;
 
-	switch (cmd) {
-	case FS_IOC_GETFLAGS:
-		ret = ovl_real_ioctl(file, cmd, arg);
-		break;
-
-	case FS_IOC_SETFLAGS:
 	if (!inode_owner_or_capable(inode))
 		return -EACCES;
 
+	if (get_user(flags, (int __user *) arg))
+		return -EFAULT;
+
 	ret = mnt_want_write_file(file);
 	if (ret)
 		return ret;
 
+	inode_lock(inode);
+
+	/* Check the capability before cred override */
+	ret = -EPERM;
+	old_flags = ovl_get_inode_flags(inode);
+	if (((flags ^ old_flags) & (FS_APPEND_FL | FS_IMMUTABLE_FL)) &&
+	    !capable(CAP_LINUX_IMMUTABLE))
+		goto unlock;
+
 	ret = ovl_maybe_copy_up(file_dentry(file), O_WRONLY);
-		if (!ret) {
-			ret = ovl_real_ioctl(file, cmd, arg);
+	if (ret)
+		goto unlock;
+
+	ret = ovl_real_ioctl(file, FS_IOC_SETFLAGS, arg);
 
-			inode_lock(inode);
 	ovl_copyflags(ovl_inode_real(inode), inode);
+unlock:
 	inode_unlock(inode);
-		}
 
 	mnt_drop_write_file(file);
+
+	return ret;
+
+}
+
+static long ovl_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	long ret;
+
+	switch (cmd) {
+	case FS_IOC_GETFLAGS:
+		ret = ovl_real_ioctl(file, cmd, arg);
+		break;
+
+	case FS_IOC_SETFLAGS:
+		ret = ovl_ioctl_set_flags(file, arg);
 		break;
 
 	default: