On Thu, 11 Jan 2024 22:01:32 +0100
Christian Brauner [off-list ref] wrote:

What I'm pointing out in the current logic is that the caller is
taxed twice:

(1) Once when the VFS has done inode_permission(MAY_EXEC, "xfs")
(2) And again when you call lookup_one_len() in eventfs_start_creating()
    _because_ the permission check in lookup_one_len() is the exact
    same permission check again that the vfs has done
    inode_permission(MAY_EXEC, "xfs").

As I described in: https://lore.kernel.org/all/20240110133154.6e18feb9@gandalf.local.home/ (local)

The eventfs files below "events" doesn't need the .permissions callback at
all. It's only there because the "events" inode uses it.

The .permissions call for eventfs has:

static int eventfs_permission(struct mnt_idmap *idmap,
			      struct inode *inode, int mask)
{
	set_top_events_ownership(inode);
	return generic_permission(idmap, inode, mask);
}

Where the "set_top_events_ownership() is a nop for everything but the
"events" directory.

I guess I could have two ops:

static const struct inode_operations eventfs_root_dir_inode_operations = {
	.lookup		= eventfs_root_lookup,
	.setattr	= eventfs_set_attr,
	.getattr	= eventfs_get_attr,
	.permission	= eventfs_permission,
};

static const struct inode_operations eventfs_dir_inode_operations = {
	.lookup		= eventfs_root_lookup,
	.setattr	= eventfs_set_attr,
	.getattr	= eventfs_get_attr,
};

And use the second one for all dentries below the root, but I figured it's
not that big of a deal if I called the permissions on all. Perhaps I should
do it with two?

Anyway, the issue is with "events" directory and remounting, because like
the tracefs system, the inode and dentry for "evnets" is created at boot
up, before the mount happens. The VFS layer is going to check the
permissions of its inode and dentry, which will be incorrect if the mount
was mounted with a "gid" option.


-- Steve