linux-kernel - Re: [PATCH v13 02/12] pkcs7: Allow the signing algo to calculate the digest itself

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: <1501916.1768998695@warthog.procyon.org.uk>
Date: Wed, 21 Jan 2026 12:31:35 +0000
From: David Howells <dhowells@...hat.com>
To: Jarkko Sakkinen <jarkko@...nel.org>
Cc: dhowells@...hat.com, Lukas Wunner <lukas@...ner.de>,
    Ignat Korchagin <ignat@...udflare.com>,
    Herbert Xu <herbert@...dor.apana.org.au>,
    Eric Biggers <ebiggers@...nel.org>,
    Luis Chamberlain <mcgrof@...nel.org>,
    Petr Pavlu <petr.pavlu@...e.com>, Daniel Gomez <da.gomez@...nel.org>,
    Sami Tolvanen <samitolvanen@...gle.com>,
    "Jason A . Donenfeld" <Jason@...c4.com>,
    Ard Biesheuvel <ardb@...nel.org>,
    Stephan Mueller <smueller@...onox.de>, linux-crypto@...r.kernel.org,
    keyrings@...r.kernel.org, linux-modules@...r.kernel.org,
    linux-kernel@...r.kernel.org
Subject: Re: [PATCH v13 02/12] pkcs7: Allow the signing algo to calculate the digest itself

Jarkko Sakkinen <jarkko@...nel.org> wrote:

> I'd use the wording you used already in commit message, which
> factors more descriptive than what you have here. E.g., name
> it "external_digest".

ML-DSA uses "external" to mean that the caller does the
digestion/hashing/XOF-ing/whatever Eric wants to call it, but the caller also
has to put other stuff into the digest/hash/XOF/thing that then gets passed to
ML-DSA if it does this.

For added confusion, the NIST FIPS tests seem to consider what this patch does
as 'external' but an "external mu" as 'internal':

	"tgId": 1,
	"testType": "AFT",
	"parameterSet": "ML-DSA-44",
	"signatureInterface": "external",
	"preHash": "pure",

vs:

	"tgId": 7,
	"testType": "AFT",
	"parameterSet": "ML-DSA-44",
	"signatureInterface": "internal",
	"externalMu": true,

I haven't come up with a better name that particularly describes this.  Maybe
use "no_prehash" or "algo_takes_hash" or "algo_takes_data"?

Maybe better than using a true/false value, use an enum?

	enum public_key_hash {
		ALGO_SIGNS_HASH, /* RSA, ECDSA, ... */
		ALGO_SIGNS_DATA, /* MLDSA, ... */
	};

David