Hi, This new patch replaces the path_is_under() check with current_chrooted() as it is done with user namespaces. Indeed, it is much more simple to check the current root instead of limiting access to a subset of files. The chroot system call is currently limited to be used by processes with the CAP_SYS_CHROOT capability. This protects against malicious procesess willing to trick SUID-like binaries. The following patch allows unprivileged users to safely use chroot(2), which may be complementary to the use of user namespaces. This patch is a follow-up of a previous one sent by Andy Lutomirski some time ago: https://lore.kernel.org/lkml/0e2f0f54e19bff53a3739ecfddb4ffa9a6dbde4d.1327858005.git.luto@xxxxxxxxxxxxxx/ This patch can be applied on top of v5.12-rc2 . I would really appreciate constructive reviews. Previous version: https://lore.kernel.org/r/20210310181857.401675-1-mic@xxxxxxxxxxx Regards, Mickaël Salaün (1): fs: Allow no_new_privs tasks to call chroot(2) fs/open.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) base-commit: a38fd8748464831584a19438cbb3082b5a2dab15 -- 2.30.2