diff options
Diffstat (limited to 'src/test')
-rw-r--r-- | src/test/serfloat_tests.cpp | 113 |
1 files changed, 69 insertions, 44 deletions
diff --git a/src/test/serfloat_tests.cpp b/src/test/serfloat_tests.cpp index d9b29b60ec..304541074f 100644 --- a/src/test/serfloat_tests.cpp +++ b/src/test/serfloat_tests.cpp @@ -37,6 +37,7 @@ uint64_t TestDouble(double f) { } // namespace BOOST_AUTO_TEST_CASE(double_serfloat_tests) { + // Test specific values against their expected encoding. BOOST_CHECK_EQUAL(TestDouble(0.0), 0U); BOOST_CHECK_EQUAL(TestDouble(-0.0), 0x8000000000000000); BOOST_CHECK_EQUAL(TestDouble(std::numeric_limits<double>::infinity()), 0x7ff0000000000000U); @@ -46,52 +47,76 @@ BOOST_AUTO_TEST_CASE(double_serfloat_tests) { BOOST_CHECK_EQUAL(TestDouble(2.0), 0x4000000000000000ULL); BOOST_CHECK_EQUAL(TestDouble(4.0), 0x4010000000000000ULL); BOOST_CHECK_EQUAL(TestDouble(785.066650390625), 0x4088888880000000ULL); + BOOST_CHECK_EQUAL(TestDouble(3.7243058682384174), 0x400dcb60e0031440); + BOOST_CHECK_EQUAL(TestDouble(91.64070592566159), 0x4056e901536d447a); + BOOST_CHECK_EQUAL(TestDouble(-98.63087668642575), 0xc058a860489c007a); + BOOST_CHECK_EQUAL(TestDouble(4.908737756962054), 0x4013a28c268b2b70); + BOOST_CHECK_EQUAL(TestDouble(77.9247330021754), 0x40537b2ed3547804); + BOOST_CHECK_EQUAL(TestDouble(40.24732825357566), 0x40441fa873c43dfc); + BOOST_CHECK_EQUAL(TestDouble(71.39395607929222), 0x4051d936938f27b6); + BOOST_CHECK_EQUAL(TestDouble(58.80100710817612), 0x404d668766a2bd70); + BOOST_CHECK_EQUAL(TestDouble(-30.10665786964975), 0xc03e1b4dee1e01b8); + BOOST_CHECK_EQUAL(TestDouble(60.15231509068704), 0x404e137f0f969814); + BOOST_CHECK_EQUAL(TestDouble(-48.15848711335961), 0xc04814494e445bc6); + BOOST_CHECK_EQUAL(TestDouble(26.68450101125353), 0x403aaf3b755169b0); + BOOST_CHECK_EQUAL(TestDouble(-65.72071986604303), 0xc0506e2046378ede); + BOOST_CHECK_EQUAL(TestDouble(17.95575825512381), 0x4031f4ac92b0a388); + BOOST_CHECK_EQUAL(TestDouble(-35.27171863226279), 0xc041a2c7ad17a42a); + BOOST_CHECK_EQUAL(TestDouble(-8.58810329425124), 0xc0212d1bdffef538); + BOOST_CHECK_EQUAL(TestDouble(88.51393044338977), 0x405620e43c83b1c8); + BOOST_CHECK_EQUAL(TestDouble(48.07224932612732), 0x4048093f77466ffc); + BOOST_CHECK_EQUAL(TestDouble(9.867348871395659e+117), 0x586f4daeb2459b9f); + BOOST_CHECK_EQUAL(TestDouble(-1.5166424385129721e+206), 0xeabe3bbc484bd458); + BOOST_CHECK_EQUAL(TestDouble(-8.585156555624594e-275), 0x8707c76eee012429); + BOOST_CHECK_EQUAL(TestDouble(2.2794371091628822e+113), 0x5777b2184458f4ee); + BOOST_CHECK_EQUAL(TestDouble(-1.1290476594131867e+163), 0xe1c91893d3488bb0); + BOOST_CHECK_EQUAL(TestDouble(9.143848423979275e-246), 0x0d0ff76e5f2620a3); + BOOST_CHECK_EQUAL(TestDouble(-2.8366718125941117e+81), 0xd0d7ec7e754b394a); + BOOST_CHECK_EQUAL(TestDouble(-1.2754409481684012e+229), 0xef80d32f8ec55342); + BOOST_CHECK_EQUAL(TestDouble(6.000577060053642e-186), 0x197a1be7c8209b6a); + BOOST_CHECK_EQUAL(TestDouble(2.0839423284378986e-302), 0x014c94f8689cb0a5); + BOOST_CHECK_EQUAL(TestDouble(-1.422140051483753e+259), 0xf5bd99271d04bb35); + BOOST_CHECK_EQUAL(TestDouble(-1.0593973991188853e+46), 0xc97db0cdb72d1046); + BOOST_CHECK_EQUAL(TestDouble(2.62945125875249e+190), 0x67779b36366c993b); + BOOST_CHECK_EQUAL(TestDouble(-2.920377657275094e+115), 0xd7e7b7b45908e23b); + BOOST_CHECK_EQUAL(TestDouble(9.790289014855851e-118), 0x27a3c031cc428bcc); + BOOST_CHECK_EQUAL(TestDouble(-4.629317182034961e-114), 0xa866ccf0b753705a); + BOOST_CHECK_EQUAL(TestDouble(-1.7674605603846528e+279), 0xf9e8ed383ffc3e25); + BOOST_CHECK_EQUAL(TestDouble(2.5308171727712605e+120), 0x58ef5cd55f0ec997); + BOOST_CHECK_EQUAL(TestDouble(-1.05034156412799e+54), 0xcb25eea1b9350fa0); - if (std::numeric_limits<double>::is_iec559) { - BOOST_CHECK_EQUAL(sizeof(double), 8U); - BOOST_CHECK_EQUAL(sizeof(uint64_t), 8U); - // Test extreme values - TestDouble(std::numeric_limits<double>::min()); - TestDouble(-std::numeric_limits<double>::min()); - TestDouble(std::numeric_limits<double>::max()); - TestDouble(-std::numeric_limits<double>::max()); - TestDouble(std::numeric_limits<double>::lowest()); - TestDouble(-std::numeric_limits<double>::lowest()); - TestDouble(std::numeric_limits<double>::quiet_NaN()); - TestDouble(-std::numeric_limits<double>::quiet_NaN()); - TestDouble(std::numeric_limits<double>::signaling_NaN()); - TestDouble(-std::numeric_limits<double>::signaling_NaN()); - TestDouble(std::numeric_limits<double>::denorm_min()); - TestDouble(-std::numeric_limits<double>::denorm_min()); - // On IEC559-compatible systems, construct doubles to test from the encoding. - for (int j = 0; j < 1000; ++j) { - // Iterate over 9 specific bits exhaustively; the others are chosen randomly. - // These specific bits are the sign bit, and the 2 top and bottom bits of - // exponent and mantissa in the IEEE754 binary64 format. - for (int x = 0; x < 512; ++x) { - uint64_t v = InsecureRandBits(64); - v &= ~(uint64_t{1} << 0); - if (x & 1) v |= (uint64_t{1} << 0); - v &= ~(uint64_t{1} << 1); - if (x & 2) v |= (uint64_t{1} << 1); - v &= ~(uint64_t{1} << 50); - if (x & 4) v |= (uint64_t{1} << 50); - v &= ~(uint64_t{1} << 51); - if (x & 8) v |= (uint64_t{1} << 51); - v &= ~(uint64_t{1} << 52); - if (x & 16) v |= (uint64_t{1} << 52); - v &= ~(uint64_t{1} << 53); - if (x & 32) v |= (uint64_t{1} << 53); - v &= ~(uint64_t{1} << 61); - if (x & 64) v |= (uint64_t{1} << 61); - v &= ~(uint64_t{1} << 62); - if (x & 128) v |= (uint64_t{1} << 62); - v &= ~(uint64_t{1} << 63); - if (x & 256) v |= (uint64_t{1} << 63); - double f; - memcpy(&f, &v, 8); - TestDouble(f); + // Test extreme values + BOOST_CHECK_EQUAL(TestDouble(std::numeric_limits<double>::min()), 0x10000000000000); + BOOST_CHECK_EQUAL(TestDouble(-std::numeric_limits<double>::min()), 0x8010000000000000); + BOOST_CHECK_EQUAL(TestDouble(std::numeric_limits<double>::max()), 0x7fefffffffffffff); + BOOST_CHECK_EQUAL(TestDouble(-std::numeric_limits<double>::max()), 0xffefffffffffffff); + BOOST_CHECK_EQUAL(TestDouble(std::numeric_limits<double>::lowest()), 0xffefffffffffffff); + BOOST_CHECK_EQUAL(TestDouble(-std::numeric_limits<double>::lowest()), 0x7fefffffffffffff); + BOOST_CHECK_EQUAL(TestDouble(std::numeric_limits<double>::denorm_min()), 0x1); + BOOST_CHECK_EQUAL(TestDouble(-std::numeric_limits<double>::denorm_min()), 0x8000000000000001); + // Note that all NaNs are encoded the same way. + BOOST_CHECK_EQUAL(TestDouble(std::numeric_limits<double>::quiet_NaN()), 0x7ff8000000000000); + BOOST_CHECK_EQUAL(TestDouble(-std::numeric_limits<double>::quiet_NaN()), 0x7ff8000000000000); + BOOST_CHECK_EQUAL(TestDouble(std::numeric_limits<double>::signaling_NaN()), 0x7ff8000000000000); + BOOST_CHECK_EQUAL(TestDouble(-std::numeric_limits<double>::signaling_NaN()), 0x7ff8000000000000); + + // Construct doubles to test from the encoding. + static_assert(sizeof(double) == 8); + static_assert(sizeof(uint64_t) == 8); + for (int j = 0; j < 1000; ++j) { + // Iterate over 9 specific bits exhaustively; the others are chosen randomly. + // These specific bits are the sign bit, and the 2 top and bottom bits of + // exponent and mantissa in the IEEE754 binary64 format. + for (int x = 0; x < 512; ++x) { + uint64_t v = InsecureRandBits(64); + int x_pos = 0; + for (int v_pos : {0, 1, 50, 51, 52, 53, 61, 62, 63}) { + v &= ~(uint64_t{1} << v_pos); + if ((x >> (x_pos++)) & 1) v |= (uint64_t{1} << v_pos); } + double f; + memcpy(&f, &v, 8); + TestDouble(f); } } } |