Open Dhruva21 opened 1 month ago
Hi @Dhruva21, liboqs doesn't yet implement ML-DSA (final standard). See https://github.com/open-quantum-safe/liboqs/pull/1919 which tracks the integration.
Appendix D.3: in the final version of ML-DSA, all of the bits of 𝑐 ̃ are used in the generation of 𝑐 (Algorithm 29),...
Hi @bhess , thank you.
If my understanding is correct, https://github.com/open-quantum-safe/liboqs/pull/1919 --> this change is for tracking the integration of final implementation of ML-DSA (final standard) correct? Once all TODO's are done the code will be merged?
Reason for the above question is, I checked out to this branch bhe-fips204-final, still I see few ACVP vectors in ML-DSA-44, ML-DSA-65 and ML-DSA-87 are failing..
And can I know where can I find the vectors online or how to generate those?
Thanks,
If my understanding is correct, open-quantum-safe/liboqs#1919 --> this change is for tracking the integration of final implementation of ML-DSA (final standard) correct? Once all TODO's are done the code will be merged?
Correct.
Reason for the above question is, I checked out to this branch bhe-fips204-final, still I see few ACVP vectors in ML-DSA-44, ML-DSA-65 and ML-DSA-87 are failing..
Yes, this is expected and related to the first two Todos in https://github.com/open-quantum-safe/liboqs/pull/1919
And can I know where can I find the vectors online or how to generate those?
They are available here: https://github.com/usnistgov/ACVP-Server/tree/master/gen-val/json-files
Hi @bhess ,
Is the liboqs source code has changes related to fips204 spec? I see the first two todo's are merged..,
Thanks,
I have built liboqs-python wrapper module which internally uses liboqs library. I was able to try sample python scripts present in examples directory.
I wrote some sample script to test the ACVP vectors (signature verification) present in liboqs source code: https://github.com/open-quantum-safe/liboqs/tree/main/tests/ACVP_Vectors/ML-DSA-sigVer-FIPS204. I see there are three failures for ML-DSA-65 and three failures for ML-DSA-87 as shown below. . --> Pass x --> Fail
Is there anything that needs to be updated from my side or there is some issue with test vectors I'm currently using?