linux-kernel - Re: [PATCH 1/2] signal: don't allow sending any signals to PF_IO

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <m1eeg9bxyi.fsf@fess.ebiederm.org>
Date:   Sat, 20 Mar 2021 11:18:29 -0500
From:   ebiederm@...ssion.com (Eric W. Biederman)
To:     Jens Axboe <axboe@...nel.dk>
Cc:     io-uring@...r.kernel.org, torvalds@...ux-foundation.org,
        linux-kernel@...r.kernel.org, oleg@...hat.com,
        Stefan Metzmacher <metze@...ba.org>
Subject: Re: [PATCH 1/2] signal: don't allow sending any signals to PF_IO_WORKER threads

Jens Axboe <axboe@...nel.dk> writes:

> They don't take signals individually, and even if they share signals with
> the parent task, don't allow them to be delivered through the worker
> thread.

This is silly I know, but why do we care?

The creds should be reasonably in-sync with the rest of the threads.

There are other threads that will receive the signal, especially when
you worry about group_send_sig_info.  Which signal sending code paths
are actually a problem.

> Reported-by: Stefan Metzmacher <metze@...ba.org>
> Signed-off-by: Jens Axboe <axboe@...nel.dk>
> ---
>  kernel/signal.c | 3 +++
>  1 file changed, 3 insertions(+)
>
> diff --git a/kernel/signal.c b/kernel/signal.c
> index ba4d1ef39a9e..730ecd3d6faf 100644
> --- a/kernel/signal.c
> +++ b/kernel/signal.c
> @@ -833,6 +833,9 @@ static int check_kill_permission(int sig, struct kernel_siginfo *info,
>  
>  	if (!valid_signal(sig))
>  		return -EINVAL;
> +	/* PF_IO_WORKER threads don't take any signals */
> +	if (t->flags & PF_IO_WORKER)
> +		return -EPERM;
>  
>  	if (!si_fromuser(info))
>  		return 0;