0d7a6bacc2constepxr-cpp_int: Disable advanced constexpr support for gcc 5.
jzmaddock
2019-09-04 11:50:06 +0100
3b7a901461constexpr-cpp_int: Fix up some more functions, add tests.
jzmaddock
2019-09-03 19:56:28 +0100
66412c3309constexpr-cpp_int: fix more functions to be be constexpr safe. [CI SKIP]
John Maddock
2019-09-03 15:08:55 +0100
4c4c400d33constexpr-cpp_int: Fix up more functions to be valid constexpr ones. Adjust numeric_limits to allow for fixed preciison cpp_int's with allocators which are not constexpr. [CI SKIP]
jzmaddock
2019-09-03 10:10:25 +0100
f2407da68aconstexpr-cpp_int: add missing newline at end of file. [CI SKIP]
jzmaddock
2019-09-02 13:13:21 +0100
787cbc34baconstexpr-cpp_int: disable support for gcc-5, it's not ready for advanced constexpr stuff.
jzmaddock
2019-09-02 12:46:22 +0100
1585622af3Correct use of <toolset>gcc-mingw:<cxxflags>-Wa,-mbig-obj
jzmaddock
2019-08-21 18:49:46 +0100
0c8f83df41constexpr: Add constexpr support to float128. Add tests for float128 constexpr. Fix existing code to support constexpr-ness in more contexts via std::is_constant_evaluated or equivalent. Change enable_if to enable_if_c in number.hpp in the (possibly vainglorious) hope of fixing some clang-apple failures.
jzmaddock
2019-08-20 17:57:04 +0100
ecd7967ccfAdded assembler option for big-obj to avoid file too big error.
pabristow
2019-08-20 17:50:16 +0100
35c9dd1ce3consexpr: add missing test file.
jzmaddock
2019-08-19 19:27:16 +0100
0c2b16b31bconstexpr: Add float128 constexpr support and tests.
jzmaddock
2019-08-19 19:22:34 +0100
7b0b6c3820Fix non-constexpr function
jzmaddock
2019-08-19 11:47:12 +0100
8007dcd864constexpr: add test case. [CI SKIP]
jzmaddock
2019-08-19 09:00:00 +0100
e7f03e0f94Fix up non-constexpr funcions
jzmaddock
2019-08-18 22:03:35 +0100
1488d12131constexpr: begin testing C++14 constexpr support.
jzmaddock
2019-08-18 15:16:27 +0100
7d62a03128constexpr: Add C++14 constexpr support to the front end. Begin testing with arithmetic_backend.
jzmaddock
2019-08-17 20:06:45 +0100
97d5ff01e7[CI SKIP] Change to revert to default filesystem::root; and instead use a change to the jamfile for location of the serials txt files. All tests and examples pass, doc rebuilt, inspect all OK.
pabristow
2019-08-16 09:44:56 +0100
30be1891c2cpp_int serial txt files moved to a new sub-folder /serial_txts and include files boost-no-inspect to avoid complaints from inspect.exe. I:\boost\libs\multiprecision\test\test_cpp_int_deserial.cpp uses boost::filesystem::path root("./serial_txts");
pabristow
2019-08-15 09:56:34 +0100
86a7142d96Edits to satisfy the picky inspect.exe program, now passing local tests using MSVC 14.2, except for six cpp_int*serial*.txt testdata files missing licence.
pabristow
2019-08-14 18:02:10 +0100
4336139112Add multiprecision.css custom style sheet, and a few new links, rebuild locally with no inspect complaints.
pabristow
2019-08-13 15:09:54 +0100
41dc0de098When the exponent is smaller than the smallest permitted, the result is zero in generic_interconvert. [CI SKIP]
jzmaddock
2019-06-29 18:53:15 +0100
Merge pull request #122 from boostorg/hexadecimal
Nick
2019-04-12 09:30:06 -0400
c06262912fFix for variable precision interconversions When a generic conversion between number<>'s is performed, we need to guard the precision of both types, not just the destination type.
jzmaddock
2019-04-06 18:13:42 +0100
jzmaddock