Jordi Pujol <jordipujolp@xxxxxxxxx> writes: > Hello, > > The attached patch solves the problem, > > Thanks, > > Jordi Pujol > > Live never ending Tale > GNU/Linux Live forever! > http://livenet.selfip.com > > overlayfs v10: > - Test for readonly using the vfsmount superblock > from the main filesystem. > - Ignore WRITE request for the real filesystems. > > Signed-off-by: Jordi Pujol <jordipujolp@xxxxxxxxx> > > --- a/fs/overlayfs/inode.c > +++ b/fs/overlayfs/inode.c 2011-06-05 10:06:25.323823937 +0200 > @@ -54,6 +54,9 @@ int ovl_permission(struct inode *inode, > bool is_upper; > int err; > > + if (mask & MAY_WRITE && IS_RDONLY(inode)) > + return -EROFS; > + > if (S_ISDIR(inode->i_mode)) { > oe = inode->i_private; > } else if (flags & IPERM_FLAG_RCU) { > @@ -102,6 +105,8 @@ int ovl_permission(struct inode *inode, > err = -EACCES; > if (IS_IMMUTABLE(realinode)) > goto out_dput; > + > + mask &= ~MAY_WRITE; Well, this is not correct. We *do* want to check for write permission. The above would simply ignore the access control. What are the permission bits on the file in question? Thanks, Miklos -- To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
