Re: [PATCH 0/8] ucounts: RLIMIT_NPROC fixes

[Shuah Khan]

On 2/10/22 7:01 PM, Eric W. Biederman wrote:
> Michal Koutný recently found some bugs in the enforcement of
> RLIMIT_NPROC in the recent ucount rlimit implementation.
>
> I saw some additional bugs and some cleaner ways to fix the problem so
> instead of starting with his fixes these are my own.
>
> I am aiming to send the first 5 of these to Linus once they have been
> reviewed.  Two more are fixes in principle but I don't think do anything
> in practice.  The last one is just a cleanup to prevent future
> divergence of RLIMIT_NPROC logic.
>
> Eric W. Biederman (8):
>        ucounts: Fix RLIMIT_NPROC regression
>        ucounts: Fix set_cred_ucounts
>        ucounts: Fix and simplify RLIMIT_NPROC handling during setuid()+execve
>        ucounts: Only except the root user in init_user_ns from RLIMIT_NPROC

Should this be "Only exempt"?
 
>        ucounts: Handle wrapping in is_ucounts_overlimit
>        ucounts: Handle inc_rlimit_ucounts wrapping in fork
>        rlimit: For RLIMIT_NPROC test the child not the parent for capabilites
>        ucounts: Use the same code to enforce RLIMIT_NPROC in fork and exec
>
>   fs/exec.c                    | 12 +++++-------
>   include/linux/sched.h        |  2 +-
>   include/linux/sched/signal.h |  2 ++
>   kernel/cred.c                | 24 +++++++++++-------------
>   kernel/fork.c                | 32 ++++++++++++++++++++++++--------
>   kernel/sys.c                 | 14 --------------
>   kernel/ucount.c              |  3 ++-
>   kernel/user_namespace.c      |  2 ++
>   8 files changed, 47 insertions(+), 44 deletions(-)
>
> Eric

Do we need updates to selftests - Michal's patch series included changes to
selftests/exec

thanks,
-- Shuah