jzmaddock
ddd51d0e1c
Update error rates for Apple M1 machines.
...
Fixes https://github.com/boostorg/math/issues/724
2021-12-12 16:56:38 +00:00
Matt Borland
181831c0ab
Constexpr hypot ( #721 )
2021-11-27 16:33:11 -07:00
jzmaddock
40605d0b78
Correct MSVC-14.2 CI failure.
2021-11-06 17:03:48 +00:00
jzmaddock
32a4b92b43
Update history and regenerate docs (part 2).
2021-11-05 19:21:25 +00:00
jzmaddock
f5cdfc57a2
Revert changes in include/boost/math/quadrature/detail/tanh_sinh_detail.hpp
2021-11-05 18:23:09 +00:00
jzmaddock
e000cf5dd1
Update history and regenerate docs.
2021-11-05 17:53:09 +00:00
jzmaddock
7108ccc1a3
Merge pull request #716 from boostorg/cubic_roots_tlc
...
Remove unicode from comments and loosen up error tolerance.
2021-11-03 19:55:05 +00:00
jzmaddock
535fcc237a
Merge pull request #714 from boostorg/tanh_sinh_performance
...
Improve tanh_sinh.
2021-11-03 19:54:44 +00:00
Nick Thompson
15c680c93a
Remove unicode from comments and loosen up error tolerance.
2021-11-02 22:31:03 -07:00
jzmaddock
afaa03a090
Merge pull request #713 from boostorg/issue710
...
Update limits used in erf/erfc approximations.
2021-11-02 19:04:41 +00:00
jzmaddock
08ff95dd14
Convert file to ASCII.
2021-11-01 19:24:31 +00:00
jzmaddock
d8078c71a3
Fix copyright declaration.
2021-11-01 18:32:07 +00:00
jzmaddock
075ff0dc6c
Better estimate error due to truncation at endpoints.
2021-10-28 12:05:12 +01:00
Nick
923ed19a07
Cubic roots ( #703 )
2021-10-26 20:54:29 -07:00
jzmaddock
035188f974
tanh_sinh: detect and allow non-finite values at end points.
2021-10-26 16:29:15 +01:00
jzmaddock
11134978df
tanh_sinh: improve thrash detection.
...
Significantly reduces the number of iterations in many cases.
2021-10-25 18:29:14 +01:00
jzmaddock
18cb92bcde
Update limits used in erf/erfc approximations.
...
Fixes https://github.com/boostorg/math/issues/710 .
2021-10-24 19:10:36 +01:00
jzmaddock
67f451f0b3
Merge pull request #687 from boostorg/s390x
...
Fix low hanging fruit test failures on s390x.
2021-10-24 13:03:22 +01:00
jzmaddock
8d09154f79
Merge pull request #711 from boostorg/tanhsinh_limits_fix
...
TanhSinh: Correct m_inital_row_length in double precision case.
2021-10-24 13:02:51 +01:00
jzmaddock
5b61efb66d
Merge pull request #712 from boostorg/quaternion_cpp_dec_float_fix
...
Correct test for cpp_dec_float division.
2021-10-24 13:00:16 +01:00
jzmaddock
3c8cfa46d1
Correct inspection report issue (missing newline).
2021-10-23 19:44:47 +01:00
jzmaddock
6f496bf93b
Correct test for cpp_dec_float division.
2021-10-23 17:20:15 +01:00
jzmaddock
67b85a4a87
TanhSinh: Correct m_inital_row_length in double precision case.
...
Remove dead header.
2021-10-23 12:23:30 +01:00
Nick
b0d1e4f760
Disallow integral return types in quadrature. ( #704 )
2021-10-11 08:18:15 -07:00
Matt Borland
7339acd01a
constexpr copysign ( #701 )
...
* constexpr copysign
* Fix for mingw test failure
2021-10-10 10:38:26 -07:00
Matt Borland
8e8f6ec4be
constexpr remainder ( #700 )
2021-10-06 10:42:02 -07:00
Matt Borland
6d5aeb08ea
constexpr fmod ( #699 )
...
* constexpr fmod
* Fix for old clang versions
2021-10-02 08:07:14 -07:00
Oleg Alexandrov
85676085ea
Expand the catmull-rom spline doc ( #698 )
...
* Expand the document of Catmull-Rom spline
It is not clear from the current documentation how the curve parameter is relates to how to interpolate between any two of the input points, which is, at the end of the day, one reason for using spline interpolation. Here a blurb is added to the doc, following https://github.com/boostorg/math/issues/211 .
* Remove unneeded newline in catmull_rom doc
2021-09-29 11:10:15 -07:00
Matt Borland
392a1c016d
constexpr round ( #697 )
2021-09-27 09:01:47 -07:00
Matt Borland
3e89a0dabd
constexpr modf ( #696 )
2021-09-24 11:09:25 -07:00
jzmaddock
bbe18dfb49
Merge pull request #685 from mscastanho/fix-nanq
...
Fix cstdfloat_math_incl_test on ppc64le with ieee128
2021-09-21 19:01:22 +01:00
Matt Borland
5d6236fb08
Implement, test, and document trunc ( #695 )
2021-09-21 09:20:43 -07:00
Matt Borland
991fcff33d
constexpr floor and ceil ( #694 )
2021-09-20 09:51:59 -07:00
Matt Borland
4ad861217d
constexpr scalbn and scalbln ( #693 )
2021-09-18 10:12:47 -07:00
jzmaddock
4fa0e9d260
Merge pull request #692 from mborland/logb
...
Constexpr logb and ilogb
2021-09-18 09:47:37 +01:00
Matt Borland
c0383edcea
Constexpr logb and ilogb
2021-09-17 21:05:07 +03:00
Matt Borland
3b75b350a4
constexpr div ( #691 )
2021-09-15 11:51:06 -07:00
jzmaddock
e592fae531
Merge pull request #690 from mborland/constexpr_abs_overloads
...
Add abs specializations to ccmath
2021-09-12 16:40:52 +01:00
jzmaddock
ee919fa3e2
Merge pull request #688 from mborland/constexpr_ldexp
...
constexpr ldexp
2021-09-12 16:40:13 +01:00
Matt Borland
5f48b50848
Add abs specializations to ccmath
2021-09-11 21:44:18 +03:00
Matt Borland
5190e5934e
Simplify implementation
2021-09-11 20:59:17 +03:00
Matt Borland
c77b5a8727
Remove testing using __builtin_constant_p
2021-09-07 20:32:25 +03:00
Matt Borland
c0e7a1fa56
Fix docs
2021-09-06 21:47:29 +03:00
Matt Borland
a8cdcd85b1
Remove unneeded includes
2021-09-06 21:40:04 +03:00
Matt Borland
000aaae2d1
Add testing
2021-09-06 21:39:51 +03:00
Matt Borland
0a69c93a31
Initial commit
2021-09-06 21:39:35 +03:00
Matt Borland
0bbea8d8f9
constexpr frexp ( #686 )
2021-09-06 11:27:27 -07:00
jzmaddock
455b88c98e
Merge pull request #684 from boostorg/no_atomic_bernoulli
...
Provide graceful failure for Bernoulli code on platforms with no atomics.
2021-09-06 16:57:54 +01:00
jzmaddock
13cb3bdacd
Fix low hanging fruit test failures on s390x.
...
This adds some extra instrumentation to bessel_j0.hpp, everything else are fixes for the test programs.
The advantage of testing on this platform is that it has a true 128-bit long double which is a good test of our assumptions in code.
2021-09-04 10:46:26 +01:00
jzmaddock
ed9a0bae1f
Fix include path in tests.
2021-09-03 10:56:44 +01:00
