On Thu, Apr 28, 2022 at 01:59:01AM +0000, Al Viro wrote:
> On Tue, Apr 26, 2022 at 07:11:27PM +0800, Yang Xu wrote:
> > Add a dedicated helper to handle the setgid bit when creating a new file
> > in a setgid directory. This is a preparatory patch for moving setgid
> > stripping into the vfs. The patch contains no functional changes.
> >
> > Currently the setgid stripping logic is open-coded directly in
> > inode_init_owner() and the individual filesystems are responsible for
> > handling setgid inheritance. Since this has proven to be brittle as
> > evidenced by old issues we uncovered over the last months (see [1] to
> > [3] below) we will try to move this logic into the vfs.
>
> First of all, inode_init_owner() is (and always had been) an optional helper.
> Filesystems are *NOT* required to call it, so putting any common functionality
> in there had always been a mistake.
>
> That goes for inode_fsuid_set() and inode_fsgid_set() calls as well.
> Consider e.g. this:
> struct inode *ext2_new_inode(struct inode *dir, umode_t mode,
> const struct qstr *qstr)
> {
> ...
> if (test_opt(sb, GRPID)) {
> inode->i_mode = mode;
> inode->i_uid = current_fsuid();
> inode->i_gid = dir->i_gid;
> } else
> inode_init_owner(&init_user_ns, inode, dir, mode);
>
> Here we have an explicit mount option, selecting the way S_ISGID on directories
> is handled. Mount ext2 with -o grpid and see for yourself - no inode_init_owner()
> calls there.
>
> The same goes for ext4 - that code is copied there unchanged.
>
> What's more, I'm not sure that Jann's fix made any sense in the first place.
> After all, the file being created here is empty; exec on it won't give you
> anything - it'll simply fail. And modifying that file ought to strip SGID,
> or we have much more interesting problems.
>
> What am I missing here?
BTW, xfs has grpid option as well:
if (dir && !(dir->i_mode & S_ISGID) && xfs_has_grpid(mp)) {
inode_fsuid_set(inode, mnt_userns);
inode->i_gid = dir->i_gid;
inode->i_mode = mode;
} else {
inode_init_owner(mnt_userns, inode, dir, mode);
}
We could lift that stuff into VFS, but it would require lifting that flag
(BSD vs. SysV behaviour wrt GID - BSD *always* inherits GID from parent
and ignores SGID on directories) into generic superblock. Otherwise we'd
be breaking existing behaviour for ext* and xfs...
