linux-kernel - Re: [PATCH 3/3] hwrng: Don't default to HW_RANDOM when UML

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: <d6fd6707808ec4429071e5473c9c2dca@manjaro.org>
Date: Sun, 09 Feb 2025 11:13:38 +0100
From: Dragan Simic <dsimic@...jaro.org>
To: Herbert Xu <herbert@...dor.apana.org.au>
Cc: linux-crypto@...r.kernel.org, linux-kernel@...r.kernel.org,
 olivia@...enic.com, didi.debian@...ow.org, heiko@...ech.de
Subject: Re: [PATCH 3/3] hwrng: Don't default to HW_RANDOM when UML_RANDOM is
 the trigger

On 2025-02-09 10:54, Herbert Xu wrote:
> On Sun, Feb 09, 2025 at 10:37:52AM +0100, Dragan Simic wrote:
>> 
>> Could you, please, clarify why we need(ed) the defaults at all?
>> Also, I'm a bit puzzled about what would be the defaults that are
>> actually needed?  Are you actually referring to what I proposed
>> in my earlier response? [1]  I'd appreciate a clarification.
> 
> In general there shouldn't be any default.  The only exception
> would be perhaps for embedded boards where the RNG is always
> present given the dependencies.  But in that case the default
> should be conditional on the dependency (or perhaps the whole
> config should become unconditional).
> 
> The current defaults are mostly there for historical reasons.

Thanks for the clarification.

I think it would be the best to get rid of the current defaults
in the drivers/char/hw_random/Kconfig file entirely, together with
updating as many of the affected arch/*/configs/*_defconfig files
as possible to select the relevant HW_RANDOM_* config option(s)
whenever that's a sane thing to do.

Such an approach should be both proper and least disruptive.