Received-SPF: softfail (google.com: domain of transitioning linux-crypto-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) client-ip=23.128.96.19;
Date:   Sun, 3 Apr 2022 20:58:32 +0200
From:   LABBE Corentin <clabbe@baylibre.com>
To:     Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Cc:     heiko@sntech.de, herbert@gondor.apana.org.au, krzk+dt@kernel.org,
        robh+dt@kernel.org, devicetree@vger.kernel.org,
        linux-arm-kernel@lists.infradead.org, linux-crypto@vger.kernel.org,
        linux-kernel@vger.kernel.org, linux-rockchip@lists.infradead.org
Subject: Re: [PATCH v4 28/33] dt-bindings: crypto: rockchip: convert to new
 driver bindings
Message-ID: <YknuWEOPq+I6jq9X@Red>
References: <20220401201804.2867154-1-clabbe@baylibre.com>
 <20220401201804.2867154-29-clabbe@baylibre.com>
 <30305936-4b69-e1ce-44c2-0d1d113b460e@linaro.org>
 <d092ede9-df92-9b34-d2d0-5da043e11679@linaro.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <d092ede9-df92-9b34-d2d0-5da043e11679@linaro.org>
Precedence: bulk

Le Sat, Apr 02, 2022 at 02:07:26PM +0200, Krzysztof Kozlowski a ?crit :
> On 02/04/2022 13:53, Krzysztof Kozlowski wrote:
> > On 01/04/2022 22:17, Corentin Labbe wrote:
> >> The latest addition to the rockchip crypto driver need to update the
> >> driver bindings.
> >>
> >> Signed-off-by: Corentin Labbe <clabbe@baylibre.com>
> >> ---
> >>  .../crypto/rockchip,rk3288-crypto.yaml        | 68 +++++++++++++++++--
> >>  1 file changed, 63 insertions(+), 5 deletions(-)
> >>
> >> diff --git a/Documentation/devicetree/bindings/crypto/rockchip,rk3288-crypto.yaml b/Documentation/devicetree/bindings/crypto/rockchip,rk3288-crypto.yaml
> >> index 66db671118c3..e6c00bc8bebf 100644
> >> --- a/Documentation/devicetree/bindings/crypto/rockchip,rk3288-crypto.yaml
> >> +++ b/Documentation/devicetree/bindings/crypto/rockchip,rk3288-crypto.yaml
> >> @@ -11,8 +11,18 @@ maintainers:
> >>  
> >>  properties:
> >>    compatible:
> >> -    enum:
> >> -      - rockchip,rk3288-crypto
> >> +    oneOf:
> >> +      - description: crypto IP present on RK3288 SoCs
> >> +        items:
> >> +          - const: rockchip,rk3288-crypto
> >> +      - description: crypto IP present on RK3328 SoCs
> > 
> > These two comments are not helping, so this should be just enum.
> > 
> >> +        items:
> >> +          - const: rockchip,rk3328-crypto
> >> +      - description: crypto IPs present on RK3399. crypto0 is the first IP with
> >> +                     RSA support, crypto1 is the second IP without RSA.
> > 
> > The second part of this comment is helpful, first not. You have chosen
> > enum in your first patch, so just extend it with comments. Additionally
> > indexing does not scale. What if next generation reverses it and crypto0
> > does not have RSA and crypto1 has?
> 
> Actually let me re-think this. Is programming model (registers?) same
> between crypto0 and crypto1? If yes, this should be same compatible and
> add a dedicated property "rockchip,rsa"?
> 
> I looked at your driver and you modeled it as main and sub devices. I
> wonder why - are there some dependencies? It would be helpful to have
> such information here in commit msg as well. Your commit #26 says that
> only difference is the RSA.
> 

Hello

There is no dependency, my only problem is that only one of 2 instance need to register crypto algos.
The only perfect way is to have a list_head of devices, but I found this a bit complex/overkill.
I understand my current way is not ideal, I will probably try this other way. In that case, yes problably the 2 node need to have the same compatible (and only a future rockchip,rsa will permit to distinct where RSA is).

Regards