[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <6711034a.050a0220.1ed172.7fa9@mx.google.com>
Date: Thu, 17 Oct 2024 14:29:58 +0200
From: Christian Marangi <ansuelsmth@...il.com>
To: Krzysztof Kozlowski <krzk@...nel.org>
Cc: Herbert Xu <herbert@...dor.apana.org.au>,
"David S. Miller" <davem@...emloft.net>,
Rob Herring <robh@...nel.org>,
Krzysztof Kozlowski <krzk+dt@...nel.org>,
Conor Dooley <conor+dt@...nel.org>,
Antoine Tenart <atenart@...nel.org>,
Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...hat.com>, Will Deacon <will@...nel.org>,
Waiman Long <longman@...hat.com>, Boqun Feng <boqun.feng@...il.com>,
Richard van Schagen <vschagen@...oud.com>,
linux-crypto@...r.kernel.org, devicetree@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH v2 2/3] dt-bindings: crypto: Add Inside Secure
SafeXcel EIP-93 crypto engine
On Thu, Oct 17, 2024 at 10:23:54AM +0200, Krzysztof Kozlowski wrote:
> On Thu, Oct 17, 2024 at 02:43:18AM +0200, Christian Marangi wrote:
> +
> > +description: |
> > + The Inside Secure SafeXcel EIP-93 is a cryptographic engine IP block
> > + integrated in varios devices with very different and generic name from
> > + PKTE to simply vendor+EIP93. The real IP under the hood is actually
> > + developed by Inside Secure and given to license to vendors.
> > +
> > + The IP block is sold with different model based on what feature are
> > + needed and are identified with the final letter. Each letter correspond
> > + to a specific set of feature and multiple letter reflect the sum of the
> > + feature set.
>
> You write it is licensed to vendors, so are you sure these could be
> used alone, without vendor customizations/hookups etc? I think you
> should have a dedicated, SoC-specific compatible in the front. I am not
> sure if this was discussed already, though.
Yes in v1, Rob asked some info about the compatible that was
mediatek,mtk-eip93 or airoha,mtk-eip93.
The thing is that from what I checked from different documentation
around, the register map and how the thing works doesn't change across
vendors, what I have seen is at max specific register added that are
outside of the crypto module usage (example debug register for BUS
bandwidth)
Any suggestion on what could be a good compatible? Honestly I'm more
tempted of using this similar to how is done by the new model EIP197.
> > +
> > + EIP-93 models:
> > + - EIP-93i: (basic) DES/Triple DES, AES, PRNG, IPsec ESP, SRTP, SHA1
> > + - EIP-93ie: i + SHA224/256, AES-192/256
> > + - EIP-93is: i + SSL/DTLS/DTLS, MD5, ARC4
> > + - EIP-93ies: i + e + s
> > + - EIP-93iw: i + AES-XCB-MAC, AES-CCM
> > +
> > +properties:
> > + compatible:
> > + enum:
> > + - inside-secure,safexcel-eip93i
> > + - inside-secure,safexcel-eip93ie
> > + - inside-secure,safexcel-eip93is
> > + - inside-secure,safexcel-eip93ies
> > + - inside-secure,safexcel-eip93iw
> > +
> > + reg:
> > + maxItems: 1
> > +
> > + interrupts:
> > + maxItems: 1
> > +
> > +required:
> > + - compatible
> > + - reg
> > + - interrupts
> > +
> > +additionalProperties: false
> > +
> > +examples:
> > + - |
> > + #include <dt-bindings/interrupt-controller/arm-gic.h>
> > +
> > + crypto@...04000 {
> > + compatible = "inside-secure,safexcel-eip93ies";
> > + reg = <0x1fb70000 0x1000>;
>
> Looks like not matching unit address.
>
Yes sorry a typo, will fix in v3.
--
Ansuel
Powered by blists - more mailing lists