linux-kernel - Re: [PATCH] crypto: qcom-rng: Add hwrng support

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]

Date:   Tue, 5 Sep 2023 10:33:45 +0200
From:   Konrad Dybcio <konrad.dybcio@...aro.org>
To:     Om Prakash Singh <quic_omprsing@...cinc.com>,
        Bjorn Andersson <quic_bjorande@...cinc.com>
Cc:     neil.armstrong@...aro.org, agross@...nel.org, andersson@...nel.org,
        conor+dt@...nel.org, davem@...emloft.net,
        devicetree@...r.kernel.org, herbert@...dor.apana.org.au,
        krzysztof.kozlowski+dt@...aro.org, linux-arm-msm@...r.kernel.org,
        linux-crypto@...r.kernel.org, linux-kernel@...r.kernel.org,
        marijn.suijten@...ainline.org, robh+dt@...nel.org, vkoul@...nel.org
Subject: Re: [PATCH] crypto: qcom-rng: Add hwrng support

On 5.09.2023 04:50, Om Prakash Singh wrote:
> 
> 
> On 9/1/2023 8:16 PM, Bjorn Andersson wrote:
>> On Fri, Sep 01, 2023 at 06:45:02PM +0530, Om Prakash Singh wrote:
>>> This is follow patch on top of [1]
>>
>> This information does not add value to the git history, if you need to
>> inform the maintainer that the patch should be applied after some
>> in-flight dependency then state so after the "---" line below.
>>
>> But, this patch strictly conflicts with [1], so the statement won't make
>> sense if this is merged.
>>
>>> to add hwrng support for newer platform with trng capability.
>>
>> Please rewrite this so that it's clear that the problem you're trying to
>> solve with this patch (i.e. the problem description) is that newer
>> platforms has trng. Describe how this relates to the existing driver
>> (e.g. same/similar hardware interface). State that you purposefully kept
>> the crypto interface in place for the new hardware as well (so that it's
>> clear that this isn't an accident or oversight).
>>
>>>
>>> [1] https://lore.kernel.org/lkml/20230824-topic-sm8550-rng-v2-4-dfcafbb16a3e@linaro.org/
>>>
>>> Signed-off-by: Om Prakash Singh <quic_omprsing@...cinc.com>
>>> ---
[...]

>>
>> Can you please confirm that it's appropriate to name this "trng" without
>> the "-ee" suffix. Should all trng instances (v2 and v3) skip
>> initialization?
> All trng supported platform needs to skip initialzation.
> we don't need to have both "trng-ee" and "trng".
> If "trng-ee" is prefer we shold update it in patch [1] it itself,
Looking back at ba3ab6371cdd ("crypto: qcom-rng - Add support for prng-ee"),
it was solved in a way that we would stray from today - nowadays
we'd call it qcom,msm8996-prng or something.

The -ee part was only there to discern parts that were initialized
by other software.

Since you said that all TRNGs need that, I'm also for dropping "-ee".

Konrad