[PATCH 1/3] crypto: public_key: Remove check for valid hash_algo for ML-DSA keys

[Stefan Berger]

On 4/6/26 12:53 PM, Eric Biggers wrote:
> On Sun, Apr 05, 2026 at 07:12:22PM -0400, Stefan Berger wrote:
>> Remove the check for the hash_algo since ML-DSA is only used in pure mode
>> and there is no relevance of a hash_algo for the input data.
>>
>> Cc: David Howells <dhowells at redhat.com>
>> Cc: Lukas Wunner <lukas at wunner.de>
>> Cc: Ignat Korchagin <ignat at linux.win>
>> Cc: keyrings at vger.kernel.org
>> Cc: linux-crypto at vger.kernel.org
>> Signed-off-by: Stefan Berger <stefanb at linux.ibm.com>
>> ---
>>   crypto/asymmetric_keys/public_key.c | 5 -----
>>   1 file changed, 5 deletions(-)
>>
>> diff --git a/crypto/asymmetric_keys/public_key.c b/crypto/asymmetric_keys/public_key.c
>> index 09a0b83d5d77..df6918a77ab8 100644
>> --- a/crypto/asymmetric_keys/public_key.c
>> +++ b/crypto/asymmetric_keys/public_key.c
>> @@ -147,11 +147,6 @@ software_key_determine_akcipher(const struct public_key *pkey,
>>   		   strcmp(pkey->pkey_algo, "mldsa87") == 0) {
>>   		if (strcmp(encoding, "raw") != 0)
>>   			return -EINVAL;
>> -		if (!hash_algo)
>> -			return -EINVAL;
>> -		if (strcmp(hash_algo, "none") != 0 &&
>> -		    strcmp(hash_algo, "sha512") != 0)
>> -			return -EINVAL;
> 
> Does this broaden which hash algorithms are accepted for CMS signatures
> that use ML-DSA and contain signed attributes?

Right... dropping this patch and using the "none" route now.

> 
> - Eric
>