| From: |
| Kees Cook <keescook-AT-chromium.org> |
| To: |
| "Eric W. Biederman" <ebiederm-AT-xmission.com> |
| Subject: |
| [PATCH 0/2] exit: Introduce __WCHILDSIGINFO for waitid |
| Date: |
| Mon, 14 Feb 2022 13:38:21 -0800 |
| Message-ID: |
| <20220214213823.3297816-1-keescook@chromium.org> |
| Cc: |
| Kees Cook <keescook-AT-chromium.org>, Christian Brauner <brauner-AT-kernel.org>, Andy Lutomirski <luto-AT-amacapital.net>, Robert Święcki <robert-AT-swiecki.net>, Jann Horn <jannh-AT-google.com>, Oleg Nesterov <oleg-AT-redhat.com>, Thomas Gleixner <tglx-AT-linutronix.de>, Ingo Molnar <mingo-AT-redhat.com>, linux-kernel-AT-vger.kernel.org, linux-api-AT-vger.kernel.org, linux-hardening-AT-vger.kernel.org |
| Archive-link: |
| Article |
Hi,
Okay, here's a working version of this. Is adding 48 bytes into task
struct worth it? Can this be improved, and is the non-signal-exit logic
for __WCHILDSIGINFO sane?
Other thoughts?
-Kees
Kees Cook (2):
exit: Introduce __WCHILDSIGINFO for waitid
selftests/seccomp: Check for waitid() behavior
include/linux/sched.h | 1 +
include/uapi/linux/wait.h | 1 +
kernel/exit.c | 23 +++-
kernel/signal.c | 4 +
tools/testing/selftests/seccomp/seccomp_bpf.c | 130 ++++++++++++++++++
5 files changed, 155 insertions(+), 4 deletions(-)
--
2.30.2
