47 Commits

Author	SHA1	Message	Date
Christopher Kormanyos	aac40fcf0e	Reduce numbers of test cases	2021-08-05 17:12:24 +02:00
Christopher Kormanyos	ecb27a6156	Tune again df arithmetic tests	2021-08-05 16:47:36 +02:00
Christopher Kormanyos	54997589e9	Add a bit to a tolerance	2021-08-05 16:19:12 +02:00
Christopher Kormanyos	8052c2312c	Remove warnings and reduce a test range	2021-08-05 14:58:01 +02:00
Christopher Kormanyos	700a252d9a	Re-tune exp10 limits in rng strings	2021-08-05 13:58:35 +02:00
Christopher Kormanyos	293a7b23ea	Correct and improve df arith tests	2021-08-05 13:51:35 +02:00
Christopher Kormanyos	a7c3d2c5df	Restore CI	2021-08-05 13:41:27 +02:00
Christopher Kormanyos	e27ccc829c	Improve df arithmetic tests and limits	2021-08-05 13:00:05 +02:00
Christopher Kormanyos	d47162ef18	Merge branch 'develop' into gsoc2021_double_float_janek	2021-08-05 08:24:18 +02:00
Christopher Kormanyos	c5ffc42fde	Nightly CI and more comments in tests	2021-08-05 08:02:46 +02:00
Janek Kozicki	659af61946	These error values for + - * / √ were tested 2e7 times and are passing.	2021-08-04 23:10:26 +02:00
Christopher Kormanyos	17b66653e0	Simplify homegrown arithmetic tests	2021-08-04 10:47:00 +02:00
Christopher Kormanyos	53ecc28a52	Refine arith tests and add msvc14.2 and MacOS	2021-08-02 10:26:37 +02:00
Christopher Kormanyos	191215a045	Reduce test lengths and try add quadmath	2021-08-01 20:20:38 +02:00
Christopher Kormanyos	347477534d	Try new arithmetics and deal with ldbl exp10	2021-08-01 20:04:34 +02:00
Christopher Kormanyos	d98667d94a	Try for quite a few more boost-like tests	2021-07-28 20:04:10 +02:00
Christopher Kormanyos	2a92a884bb	Re-try Boost's arithmetic tests	2021-07-28 18:18:28 +02:00
Syed Fahad	2a91e292da	handle inf and nan	2021-07-28 19:54:50 +05:30
Christopher Kormanyos	4f157cbc06	Tweak the test parameters	2021-07-28 11:03:59 +02:00
Christopher Kormanyos	c9295d0300	Add (now unused) stuff to cascade dbl_flt type	2021-07-28 10:52:56 +02:00
Christopher Kormanyos	ba68d06f7a	Try rudimentary eval_sqrt (optimize later)	2021-07-27 20:10:48 +02:00
Christopher Kormanyos	bd45fa7ff2	Cleanup tests a bit	2021-07-27 19:49:03 +02:00
Christopher Kormanyos	00c3b7919a	Reduce test cases when addr sanitize detected	2021-07-26 19:53:47 +02:00
Christopher Kormanyos	cf5b72fa53	Boost-ified draft ok but GCC long double broke	2021-07-26 19:42:38 +02:00
Christopher Kormanyos	b614311e61	Try simplify tests but long double broken on GCC	2021-07-26 19:25:21 +02:00
Christopher Kormanyos	a157a5adc3	Restore yet again non-Boost-ified standalone	2021-07-26 14:59:31 +02:00
Christopher Kormanyos	5d79883096	Re-try Boost-ification draft	2021-07-26 14:23:19 +02:00
Christopher Kormanyos	6e31db76a7	Revert to standalone cpp_double_float for retry	2021-07-25 22:30:25 +02:00
Christopher Kormanyos	e6bcbe7387	Very preliminary backend but no 128 yet	2021-07-25 22:20:05 +02:00
Christopher Kormanyos	852f169692	Put back that least bit of tolerance	2021-07-24 22:05:43 +02:00
Christopher Kormanyos	8f0514e9bb	Merge branch 'develop' into gsoc2021_double_float_chris_modify_and_test	2021-07-24 20:31:07 +02:00
Christopher Kormanyos	8a060d0bd6	Cleanup and 1 more bit and increase test cases	2021-07-24 20:28:11 +02:00
Janek Kozicki	cf8fa4cce4	almost complete double<float128> support in tests.	2021-07-24 18:25:16 +02:00
Christopher Kormanyos	3497de7d34	Rework and simplify unary add/sub	2021-07-24 17:10:26 +02:00
Christopher Kormanyos	1245b3e564	Find out what is up with CI	2021-07-24 11:51:10 +02:00
Christopher Kormanyos	e57cff3149	Tweak the tests	2021-07-24 11:42:31 +02:00
Christopher Kormanyos	1e628c5017	Try in main dev branch to go green on CI	2021-07-24 11:39:27 +02:00
Christopher Kormanyos	56043776b4	draft of numeric_limits and pass arithmetic tests	2021-07-24 11:29:46 +02:00
Janek Kozicki	cebca3b1df	printing double<float128> wasn't working. But I'm not sure if that's the correct fix.	2021-07-22 19:20:04 +02:00
Janek Kozicki	c96eac3fb5	a little more meaningful message on CI: apple-gcc-clang-native (c++17, g++)	2021-07-19 22:28:24 +02:00
Janek Kozicki	98bef0c0ca	use struct instead	2021-07-19 22:28:24 +02:00
Janek Kozicki	74cf9e47e6	float128 seems to work in test/test_cpp_double_float_arithmetic.cpp	2021-07-19 22:28:24 +02:00
Christopher Kormanyos	14a2bd9775	remove warn from CI and go green as well	2021-07-18 23:04:23 +02:00
Syed Fahad	22bb1b04fa	fix g++ compilation problems	2021-07-16 23:09:56 +05:30
Syed Fahad	803b96723a	implement robust tests for correctness of arithmetic operators ... fix critical bug in split() function update std::numeric_limits specialization	2021-07-16 22:07:30 +05:30
Syed Fahad	8bb92af6e4	prettify test for arithmetic operators ... fix bug in conversion to decimal string	2021-07-15 16:08:36 +05:30
Christopher Kormanyos	bfdc8de445	Add constructor test CI and suggest test stuff	2021-07-05 19:50:52 +02:00