Re: [PATCH 01/13] fs/super.c: don't drop ->s_user_ns until we free struct super_block itself

On Sun, Feb 04, 2024 at 02:17:27AM +0000, Al Viro wrote:

Avoids fun races in RCU pathwalk...  Same goes for freeing LSM shite
hanging off super_block's arse.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---

Hah, I once had the same patch for the userns bit because I was
wondering about that,

Reviewed-by: Christian Brauner <brauner@kernel.org>

(Independent of whether or not this is pretty the s_user_ns should
probably be a separate type so it can't be confused with other
namespaces when checking permissions. Maybe I should respin my series
for that if I find the time.)