Re: [PATCH 6/9] mnt: simplify ns_common_init() handling

[Christian Brauner]

On Wed, Sep 17, 2025 at 06:45:11PM +0200, Jan Kara wrote:
> On Wed 17-09-25 12:28:05, Christian Brauner wrote:
> > Assign the reserved MNT_NS_ANON_INO sentinel to anonymous mount
> > namespaces and cleanup the initial mount ns allocation. This is just a
> > preparatory patch and the ns->inum check in ns_common_init() will be
> > dropped in the next patch.
> > 
> > Signed-off-by: Christian Brauner <brauner@xxxxxxxxxx>
> 
> ...
> > ---
> >  fs/namespace.c    | 7 ++++---
> >  kernel/nscommon.c | 2 +-
> >  2 files changed, 5 insertions(+), 4 deletions(-)
> > 
> > diff --git a/fs/namespace.c b/fs/namespace.c
> > index c8251545d57e..09e4ecd44972 100644
> > --- a/fs/namespace.c
> > +++ b/fs/namespace.c
> > @@ -4104,6 +4104,8 @@ static struct mnt_namespace *alloc_mnt_ns(struct user_namespace *user_ns, bool a
> >  		return ERR_PTR(-ENOMEM);
> >  	}
> >  
> > +	if (anon)
> > +		new_ns->ns.inum = MNT_NS_ANON_INO;
> >  	ret = ns_common_init(&new_ns->ns, &mntns_operations, !anon);
> >  	if (ret) {
> >  		kfree(new_ns);
> > @@ -6020,10 +6022,9 @@ static void __init init_mount_tree(void)
> >  	if (IS_ERR(mnt))
> >  		panic("Can't create rootfs");
> >  
> > -	ns = alloc_mnt_ns(&init_user_ns, true);
> > +	ns = alloc_mnt_ns(&init_user_ns, false);
> >  	if (IS_ERR(ns))
> >  		panic("Can't allocate initial namespace");
> > -	ns->ns.inum = PROC_MNT_INIT_INO;
> >  	m = real_mount(mnt);
> >  	ns->root = m;
> >  	ns->nr_mounts = 1;
> > @@ -6037,7 +6038,7 @@ static void __init init_mount_tree(void)
> >  	set_fs_pwd(current->fs, &root);
> >  	set_fs_root(current->fs, &root);
> >  
> > -	ns_tree_add(ns);
> > +	ns_tree_add_raw(ns);
> 
> But we don't have ns->ns_id set by anything now? Or am I missing something?

It is set in alloc_mnt_ns() via ns_tree_gen_id(). :)
Unless I'm missing something.
But we still need to add PROC_MNT_INIT_INTO until the final conversion.
I'm fixing that in-tree.