linux-kernel - Re: [PATCH] exec: do not call sched_mm_cid_after

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: <aUvYMRmkWXUuuWXW@ndev>
Date: Wed, 24 Dec 2025 20:10:49 +0800
From: Jinchao Wang <wangjinchao600@...il.com>
To: mathieu.desnoyers@...icios.com,
	Alexander Viro <viro@...iv.linux.org.uk>,
	Christian Brauner <brauner@...nel.org>, Jan Kara <jack@...e.cz>,
	Kees Cook <kees@...nel.org>, linux-fsdevel@...r.kernel.org,
	linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] exec: do not call sched_mm_cid_after_execve() on exec
 fail

On Thu, Dec 18, 2025 at 11:23:23AM +0800, Jinchao Wang wrote:

Hi, mathieu

Please review this patch for mm_cid.

The syzbot test failure occurs because the syz-repro breaks the system
state, causing all user commands to fail. Actually the syz-rero can only
trigger a panic or error.

The broken system can be recovered by resetting binfmt_misc with:

echo 0 > /proc/sys/fs/binfmt_misc/status


Thanks.

Cc: mathieu.desnoyers@...icios.com
> sched_mm_cid_after_execve() is called from the failure path
> of bprm_execve(). At that point exec has not completed successfully,
> so updating the mm CID state is incorrect and can trigger a panic,
> as reported by syzbot.
> 
> Remove the call from the exec failure path.
> 
> Cc: stable@...r.kernel.org
> Reported-by: syzbot+9ca2c6e6b098bf5ae60a@...kaller.appspotmail.com
> Signed-off-by: Jinchao Wang <wangjinchao600@...il.com>
> ---
>  fs/exec.c | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/fs/exec.c b/fs/exec.c
> index 9d5ebc9d15b0..9044a75d26ab 100644
> --- a/fs/exec.c
> +++ b/fs/exec.c
> @@ -1773,7 +1773,6 @@ static int bprm_execve(struct linux_binprm *bprm)
>  	if (bprm->point_of_no_return && !fatal_signal_pending(current))
>  		force_fatal_sig(SIGSEGV);
>  
> -	sched_mm_cid_after_execve(current);
>  	rseq_force_update();
>  	current->in_execve = 0;
>  
> -- 
> 2.43.0
>