linux-kernel - Re: [PATCH v2 6/6] module: Introduce hash-based integrity checking

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] [day] [month] [year] [list]

Message-ID: <8e5b171d-78fa-4cba-8217-1a661d23785b@archlinux.org>
Date: Thu, 23 Jan 2025 00:28:40 +0100
From: kpcyrd <kpcyrd@...hlinux.org>
To: Thomas Weißschuh <linux@...ssschuh.net>,
 Masahiro Yamada <masahiroy@...nel.org>, Nathan Chancellor
 <nathan@...nel.org>, Nicolas Schier <nicolas@...sle.eu>,
 Arnd Bergmann <arnd@...db.de>, Luis Chamberlain <mcgrof@...nel.org>,
 Petr Pavlu <petr.pavlu@...e.com>, Sami Tolvanen <samitolvanen@...gle.com>,
 Daniel Gomez <da.gomez@...sung.com>, Paul Moore <paul@...l-moore.com>,
 James Morris <jmorris@...ei.org>, "Serge E. Hallyn" <serge@...lyn.com>,
 Jonathan Corbet <corbet@....net>
Cc: Fabian Grünbichler <f.gruenbichler@...xmox.com>,
 Arnout Engelen <arnout@...t.net>, Mattia Rizzolo <mattia@...reri.org>,
 linux-kbuild@...r.kernel.org, linux-kernel@...r.kernel.org,
 linux-arch@...r.kernel.org, linux-modules@...r.kernel.org,
 linux-security-module@...r.kernel.org, linux-doc@...r.kernel.org
Subject: Re: [PATCH v2 6/6] module: Introduce hash-based integrity checking

Hi!

Thanks for reaching out, also your work on this is much appreciated and 
followed with great interest. <3

On 1/20/25 6:44 PM, Thomas Weißschuh wrote:
> diff --git a/kernel/module/main.c b/kernel/module/main.c
> index effe1db02973d4f60ff6cbc0d3b5241a3576fa3e..094ace81d795711b56d12a2abc75ea35449c8300 100644
> --- a/kernel/module/main.c
> +++ b/kernel/module/main.c
> @@ -3218,6 +3218,12 @@ static int module_integrity_check(struct load_info *info, int flags)
>   {
>   	int err = 0;
>   
> +	if (IS_ENABLED(CONFIG_MODULE_HASHES)) {
> +		err = module_hash_check(info, flags);
> +		if (!err)
> +			return 0;
> +	}
> +
>   	if (IS_ENABLED(CONFIG_MODULE_SIG))
>   		err = module_sig_check(info, flags);
>   

 From how I'm reading this (please let me know if I'm wrong):

## !CONFIG_MODULE_HASHES && !CONFIG_MODULE_SIG

No special checks, CAP_SYS_MODULE only.

## !CONFIG_MODULE_HASHES && CONFIG_MODULE_SIG

No change from how things work today:

- if the module signature verifies the module is permitted
- else, if sig_enforce=1, the module is rejected
- else, if lockdown mode is enabled, the module is rejected
- else, the module is permitted

## CONFIG_MODULE_HASHES && CONFIG_MODULE_SIG

This configuration is the one relevant for Arch Linux:

- if the module is in the set of allowed module_hashes it is permitted
- else, if the module signature verifies, the module is permitted
- else, if sig_enforce=1, the module is rejected
- else, if lockdown mode is enabled, the module is rejected
- else, the module is permitted

## CONFIG_MODULE_HASHES && !CONFIG_MODULE_SIG

This one is new:

- if the module is in the set of allowed module_hashes it is permitted
- else, if lockdown mode is enabled, the module is rejected
- else, the module is permitted

---

This all seems reasonable to me, maybe the check for 
is_module_sig_enforced() could be moved from kernel/module/signing.c to 
kernel/module/main.c, otherwise `sig_enforce=1` would not have any 
effect for a `CONFIG_MODULE_HASHES && !CONFIG_MODULE_SIG` kernel.

cheers,
kpcyrd