[cryptotest] Parser for NIST SHA2/3, SHAKE test vectors

Signed-off-by: Ryan Torok <[email protected]>

sw/host/cryptotest/testvectors/data/BUILD

@@ -50,3 +50,57 @@ genrule(
         ("ecdsa_secp384r1_sha3_512_test.json", "wycheproof_ecdsa_p384_sha3_512"),
     ]
 ]
+
+[
+    run_binary(
+        name = "nist_cavp_{}_{}_{}_json".format(
+            src_repo,
+            algorithm.lower(),
+            msg_type.lower(),
+        ),
+        srcs = [
+            "@nist_cavp_{}//:{}{}.rsp".format(src_repo, algorithm, msg_type),
+            "//sw/host/cryptotest/testvectors/data/schemas:hash_schema.json",
+        ],
+        outs = [":nist_{}_{}.json".format(
+            algorithm.lower(),
+            msg_type.lower(),
+        )],
+        args = [
+            "--src",
+            "$(location @nist_cavp_{}//:{}{}.rsp)".format(src_repo, algorithm, msg_type),
+            "--dst",
+            "$(location :nist_{}_{}.json)".format(
+                algorithm.lower(),
+                msg_type.lower(),
+            ),
+            "--schema",
+            "$(location //sw/host/cryptotest/testvectors/data/schemas:hash_schema.json)",
+            "--algorithm",
+            algorithm,
+        ],
+        tool = "//sw/host/cryptotest/testvectors/parsers:nist_cavp_hash_parser",
+    )
+    for algorithm, src_repo, extra_msg_types in [
+        ("SHA256", "sha2_fips_180_4", []),
+        ("SHA384", "sha2_fips_180_4", []),
+        ("SHA512", "sha2_fips_180_4", []),
+        ("SHA3_256", "sha3_fips_202", []),
+        ("SHA3_384", "sha3_fips_202", []),
+        ("SHA3_512", "sha3_fips_202", []),
+        (
+            "SHAKE128",
+            "shake_fips_202",
+            ["VariableOut"],
+        ),
+        (
+            "SHAKE256",
+            "shake_fips_202",
+            ["VariableOut"],
+        ),
+    ]
+    for msg_type in [
+        "ShortMsg",
+        "LongMsg",
+    ] + extra_msg_types
+]

sw/host/cryptotest/testvectors/parsers/BUILD

@@ -41,6 +41,15 @@ py_binary(
     ],
 )
 
+py_binary(
+    name = "nist_cavp_hash_parser",
+    srcs = ["nist_cavp_hash_parser.py"],
+    deps = [
+        ":cryptotest_util",
+        requirement("jsonschema"),
+    ],
+)
+
 py_binary(
     name = "nist_cavp_hmac_parser",
     srcs = ["nist_cavp_hmac_parser.py"],

sw/host/cryptotest/testvectors/parsers/nist_cavp_hash_parser.py

@@ -0,0 +1,107 @@
+#!/usr/bin/env python3
+# Copyright lowRISC contributors.
+# Licensed under the Apache License, Version 2.0, see LICENSE for details.
+# SPDX-License-Identifier: Apache-2.0
+
+"""Parser for converting NIST CAVP Hash Function test vectors to JSON.
+
+"""
+
+import argparse
+import sys
+import json
+import jsonschema
+
+from cryptotest_util import parse_rsp, str_to_byte_array
+
+# Map hash function names as formatted in the test vectors to their
+# JSON schema names
+HASH_FUNCTION_NAME_MAPPING = {
+    "SHA256": "sha-256",
+    "SHA384": "sha-384",
+    "SHA512": "sha-512",
+    "SHA3_256": "sha3-256",
+    "SHA3_384": "sha3-384",
+    "SHA3_512": "sha3-512",
+    "SHAKE128": "shake-128",
+    "SHAKE256": "shake-256",
+}
+
+
+def parse_testcases(args) -> None:
+    raw_testcases = parse_rsp(args.src)
+    test_cases = list()
+    algorithm = HASH_FUNCTION_NAME_MAPPING[args.algorithm]
+    # The test vectors for the SHA functions use "MD" as the key for
+    # the message digest. SHAKE functions use "Output".
+    if algorithm.startswith("shake"):
+        digest_key = "Output"
+    else:
+        digest_key = "MD"
+    for section_name in raw_testcases.keys():
+        count = 1
+        for test_vec in raw_testcases[section_name]:
+            # The test vector includes a single placeholder zero byte if its
+            # intended message length is 0, so we need to ignore that byte.
+            if "Len" in test_vec and int(test_vec["Len"]) == 0:
+                message = []
+            else:
+                message = str_to_byte_array(test_vec["Msg"])
+            if "COUNT" in test_vec:
+                test_case_id = int(test_vec["COUNT"])
+            else:
+                test_case_id = count
+            test_case = {
+                "vendor": "nist",
+                "test_case_id": test_case_id,
+                "algorithm": algorithm,
+                "message": message,
+                "digest": str_to_byte_array(test_vec[digest_key]),
+                # All NIST hash test vectors are expected to have the
+                # right message digest
+                "result": True,
+            }
+
+            test_cases.append(test_case)
+            count += 1
+
+    json_filename = f"{args.dst}"
+    with open(json_filename, "w") as file:
+        json.dump(test_cases, file, indent=4)
+
+    # Validate generated JSON
+    with open(args.schema) as schema_file:
+        schema = json.load(schema_file)
+    jsonschema.validate(test_cases, schema)
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser(
+        description="Parsing utility for NIST CAVP Hash Function test vectors.")
+
+    parser.add_argument(
+        "--src",
+        help="Source file to import."
+    )
+    parser.add_argument(
+        "--dst",
+        help="Destination of the output file."
+    )
+    parser.add_argument(
+        "--schema",
+        type = str,
+        help = "Test vector schema file"
+    )
+    parser.add_argument(
+        "--algorithm",
+        type = str,
+        help = "Hash algorithm the test vectors are testing"
+    )
+    args = parser.parse_args()
+    parse_testcases(args)
+
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())