openbsd/src/regress/lib, branch libressl-v3.7.1 A mirror of https://github.com/libressl/openbsd.git https://git.lua4.win/openbsd/atom?h=libressl-v3.7.1 2023-03-15T04:26:23+00:00 Include tests with negative values in BN_mod_exp* regress. 2023-03-15T04:26:23+00:00 jsing 2023-03-15T04:26:23+00:00 urn:sha1:ff8b9955b55cdb95151225a1cdf20fe993f2085e This currently fails. bn_mod_exp_zero: rename result into got 2023-03-15T00:41:04+00:00 tb 2023-03-15T00:41:04+00:00 urn:sha1:95dc15ad075d5f5b9ccf5e7f5e62d226a729a40d Add OPENSSL_cleanup() calls to some regress. 2023-03-11T14:27:38+00:00 jsing 2023-03-11T14:27:38+00:00 urn:sha1:2e5da2b49d0dfd3201a55a5bde6359c5a9a1c4f5 This gets us some minimal test coverage. Add regress coverage for BN_{add,sub,mul,div,mod}_word(). 2023-03-11T14:04:21+00:00 jsing 2023-03-11T14:04:21+00:00 urn:sha1:fa84d7922034e108e9bcd911f395501553a0a874 This also provides some indirect coverage for BN_hex2bn(), BN_bn2hex() and BN_get_word(). Two of these tests are currently failing and will be fixed shortly. Mark test table as static const. 2023-03-11T14:02:26+00:00 jsing 2023-03-11T14:02:26+00:00 urn:sha1:2a954232ec5e12d1b069305badf158243cde4e1c Use BN_free() instead of BN_clear_free() 2023-03-08T16:51:42+00:00 tb 2023-03-08T16:51:42+00:00 urn:sha1:6f9fe457a470084647ce109884cb6191d847ff48 reduce number of tests in bn_rand_interval. 2023-03-08T06:44:45+00:00 tb 2023-03-08T06:44:45+00:00 urn:sha1:61553ad84a963afb2cb5178598c44ed09d4a4f08 This is only testing basic functionality anyway, so 10000 tests are more than enough. bn_isqrt: reduce number of tests to 100. 2023-03-08T06:28:08+00:00 tb 2023-03-08T06:28:08+00:00 urn:sha1:d00b93818b38166ff03a7f9dcf6993c5acd8edbe The runtime is roughly quadratic in N_TESTS. While it only takes 1-2s on modern machines, this test takes a long time on slow machines. A reduction of runtime by a factor of ~16 is significant. Process up to four test vector files concurrently. 2023-03-08T06:12:52+00:00 jsing 2023-03-08T06:12:52+00:00 urn:sha1:6718367e12207d6e0c2eb4f4ea75f16c46619449 This avoids having a slow down when processing test vector files that only have a single group. Note that the processing of test vector files is in turn going to be rate limited by the number of concurrent test groups, which means we do not need variable limits for vectors. Reduces a Wycheproof regress run down to ~8 seconds on an Apple M1. Run test groups concurrently. 2023-03-08T05:41:08+00:00 jsing 2023-03-08T05:41:08+00:00 urn:sha1:442f7bc80303382d3992bb55ef733a468c1662f2 Add a basic test coordinator, that allows for Wycheproof test groups to be run concurrently. This can be further improved (especially for vectors that have limited test groups), however it already reduces the regress duration by about half on an Apple M1.