linux-kernel - Re: [PATCH] LSM: Infrastructure management of the sock

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: <0a2b8372-11f6-56a3-7d8e-41e93d1bf691@schaufler-ca.com>
Date:   Wed, 31 May 2023 14:36:27 -0700
From:   Casey Schaufler <casey@...aufler-ca.com>
To:     Paul Moore <paul@...l-moore.com>
Cc:     "GONG, Ruiqi" <gongruiqi@...weicloud.com>,
        John Johansen <john.johansen@...onical.com>,
        James Morris <jmorris@...ei.org>,
        "Serge E . Hallyn" <serge@...lyn.com>,
        Stephen Smalley <stephen.smalley.work@...il.com>,
        Eric Paris <eparis@...isplace.org>,
        linux-kernel@...r.kernel.org, apparmor@...ts.ubuntu.com,
        linux-security-module@...r.kernel.org, selinux@...r.kernel.org,
        Kees Cook <keescook@...omium.org>,
        Wang Weiyang <wangweiyang2@...wei.com>,
        Xiu Jianfeng <xiujianfeng@...wei.com>, gongruiqi1@...wei.com,
        Casey Schaufler <casey@...aufler-ca.com>
Subject: Re: [PATCH] LSM: Infrastructure management of the sock

On 5/31/2023 2:10 PM, Paul Moore wrote:
> On Wed, May 31, 2023 at 10:00 AM Casey Schaufler <casey@...aufler-ca.com> wrote:
>> On 5/31/2023 4:05 AM, GONG, Ruiqi wrote:
>>> As the security infrastructure has taken over the management of multiple
>>> *_security blobs that are accessed by multiple security modules, and
>>> sk->sk_security shares the same situation, move its management out of
>>> individual security modules and into the security infrastructure as
>>> well. The infrastructure does the memory allocation, and each relavant
>>> module uses its own share.
>> Do you have a reason to make this change? The LSM infrastructure
>> manages other security blobs to enable multiple concurrently active
>> LSMs to use the blob. If only one LSM on a system can use the
>> socket blob there's no reason to move the management.
> I think an argument could be made for consistent handling of security
> blobs, but with the LSM stacking work in development the argument for
> merging this patch needs to be a lot stronger than just "consistency".

I'm betting that someone has an out-of-tree LSM that uses a socket blob,
and that the intended use case includes stacking with one of the "major"
LSMs. I would encourage that someone to propose that LSM for upstream.