From: Prarit Bhargava <prarit@redhat.com>
To: "Alessandro Carminati (Red Hat)" <alessandro.carminati@gmail.com>,
linux-modules@vger.kernel.org
Cc: linux-kernel@vger.kernel.org,
Luis Chamberlain <mcgrof@kernel.org>,
Jonathan Corbet <corbet@lwn.net>,
linux-doc@vger.kernel.org
Subject: Re: [RFC PATCH 0/2] Enhancing Boot Speed and Security with Delayed Module Signature Verification
Date: Wed, 8 Nov 2023 10:32:13 -0500 [thread overview]
Message-ID: <99645b01-f297-48ed-81f2-b6ce186085d1@redhat.com> (raw)
In-Reply-To: <20230914112739.112729-1-alessandro.carminati@gmail.com>
On 9/14/23 07:27, Alessandro Carminati (Red Hat) wrote:
> This patch sets up a new feature to the Linux kernel to have the ability,
> while module signature checking is enabled, to delay the moment where
> these signatures are effectively checked. The feature is structure into
> two main key points, the feature can be enabled by a new command line
> kernel argument, while in delay mode, the kernel waits until the
> userspace communicates to start checking signature modules.
> This operation can be done by writing a value in a securityfs file,
> which works the same as /sys/kernel/security/lockdown.
>
> Patch 1/2: Modules: Introduce boot-time module signature flexibility
> The first patch in this set fundamentally alters the kernel's behavior
> at boot time by implementing a delayed module signature verification
> mechanism. It introduces a new boot-time kernel argument that allows
> users to request this delay. By doing so, we aim to capitalize on the
> cryptographic checks already performed on the kernel and initrd images
> during the secure boot process. As a result, we can significantly
> improve the boot speed without compromising system security.
>
> Patch 2/2: docs: Update kernel-parameters.txt for signature verification
> enhancement
> The second patch is just to update the kernel parameters list
> documentation.
>
> Background and Motivation
> In certain contexts, boot speed becomes crucial. This patch follows the
> recognition that security checks can at times be redundant. Therefore,
> it proves valuable to skip those checks that have already been validated.
>
> In a typical Secure Boot startup with an initrd, the bootloader is
> responsible for verifying artifacts before relinquishing control. In a
> verified initrd image, it is reasonable to assume that its content is
> also secure. Consequently, verifying module signatures may be deemed
> unnecessary.
> This patch introduces a feature to skip signature verification during
> the initrd boot phase.
>
I think this is fine to do. There is some risk for users who may use
this without realizing what they're actually doing and then would end up
creating a security hole. But there are far worse ways you can do that
with access to kernel paramaters.
P.
> Alessandro Carminati (Red Hat) (2):
> Modules: Introduce boot-time module signature flexibility
> docs: Update kernel-parameters.txt for signature verification
> enhancement
>
> .../admin-guide/kernel-parameters.txt | 9 +++
> include/linux/module.h | 4 ++
> kernel/module/main.c | 14 +++--
> kernel/module/signing.c | 56 +++++++++++++++++++
> 4 files changed, 77 insertions(+), 6 deletions(-)
>
next prev parent reply other threads:[~2023-11-08 15:32 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-14 11:27 [RFC PATCH 0/2] Enhancing Boot Speed and Security with Delayed Module Signature Verification Alessandro Carminati (Red Hat)
2023-09-14 11:27 ` [RFC PATCH 1/2] Modules: Introduce boot-time module signature flexibility Alessandro Carminati (Red Hat)
2023-11-16 17:35 ` Luis Chamberlain
2023-11-17 13:56 ` Alessandro Carminati
2023-11-17 18:33 ` Luis Chamberlain
2023-11-20 19:43 ` Luca Boccassi
2023-09-14 11:27 ` [RFC PATCH 2/2] docs: Update kernel-parameters.txt for signature verification enhancement Alessandro Carminati (Red Hat)
2023-09-15 15:59 ` Randy Dunlap
2023-11-08 15:33 ` Prarit Bhargava
2023-11-09 10:40 ` Alessandro Carminati
2023-11-08 15:32 ` Prarit Bhargava [this message]
2023-11-09 10:51 ` [RFC PATCH 0/2] Enhancing Boot Speed and Security with Delayed Module Signature Verification Alessandro Carminati
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=99645b01-f297-48ed-81f2-b6ce186085d1@redhat.com \
--to=prarit@redhat.com \
--cc=alessandro.carminati@gmail.com \
--cc=corbet@lwn.net \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-modules@vger.kernel.org \
--cc=mcgrof@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox