ckormanyos
73e26ca581
Merge branch 'cstdfloat' of https://github.com/boostorg/math into cstdfloat
2014-02-18 23:08:02 +01:00
ckormanyos
455c4bf3a9
In <boost/detail/cstdfloat_complex_std.hpp>, do some trivial clean-up and genreal efficiency improvements.
2014-02-18 23:07:11 +01:00
jzmaddock
287d66d124
Merge branch 'develop' of https://github.com/boostorg/math into develop
2014-02-18 18:30:53 +00:00
jzmaddock
d8c74aa1e3
Fix some (possibly hypothetical) error handling issues.
2014-02-18 18:30:24 +00:00
jzmaddock
c00cc0b889
Fix some (possibly hypothetical) error handling issues.
2014-02-18 18:29:42 +00:00
PABristow
a6871cd686
update trac #9672
2014-02-18 17:37:30 +00:00
PABristow
55bc21684b
Merge branch 'develop' of https://github.com/boostorg/math into develop
2014-02-18 17:29:16 +00:00
PABristow
bf40c60296
Random variate can be infinite.
2014-02-18 17:21:11 +00:00
PABristow
da499e8cc1
Adjusted tests and comments for random variate infinite.
2014-02-18 17:18:00 +00:00
PABristow
3cd3dbd39c
Adjusted tests as random variate can now be infinite.
2014-02-18 17:15:11 +00:00
jzmaddock
31937ff2ba
Merge branch 'cstdfloat' of https://github.com/boostorg/math into cstdfloat
2014-02-17 12:05:08 +00:00
ckormanyos
8392a52517
In <boost/cstdfloat.hpp>, ensure that the rendition of <complex> *also* works with other UDT, such as multiprecision.
2014-02-16 22:06:29 +01:00
jzmaddock
0028c59d7c
Fix for tgamma(x) where x < 0 and denorm.
2014-02-16 16:24:59 +00:00
ckormanyos
3391f44fab
In <boost/cstdfloat.hpp>, implement granular files.
2014-02-16 13:41:00 +01:00
ckormanyos
3b4ba962ab
In test_cstdfloat.cpp, improve the sanity check for the floating-point types.
2014-02-15 23:05:43 +01:00
ckormanyos
47385add82
In test_cstdfloat.cpp, the file now compiles and runs. Significant test depth is still missing.
2014-02-15 22:20:30 +01:00
ckormanyos
38829228f4
In <boost/cstdfloat.hpp>, move the C++ standard library support into separate detail files.
2014-02-15 22:06:29 +01:00
jzmaddock
7a823466ca
More fixes from Rocco Romeo: do the correct thing near a small negative integer, and handle denormalized inputs correctly.
2014-02-15 19:19:44 +00:00
jzmaddock
942288b128
Add missing using declaration.
2014-02-15 17:51:10 +00:00
jzmaddock
742ecfd751
Fix zeta instances used.
2014-02-15 17:51:09 +00:00
jzmaddock
e9dcbd97f3
Fix expected error rates on Mingw-x64
2014-02-15 17:51:09 +00:00
jzmaddock
0d85f89449
Add missing using declaration.
2014-02-15 17:46:20 +00:00
jzmaddock
2e10391ada
Merge branch 'cstdfloat' of https://github.com/boostorg/math into cstdfloat
2014-02-15 17:35:21 +00:00
jzmaddock
7cccb8ea2f
Fix zeta instances used.
2014-02-15 17:34:54 +00:00
jzmaddock
960eaddf2a
Fix expected error rates on Mingw-x64
2014-02-15 17:28:07 +00:00
jzmaddock
5a78d8d584
Add some more missing return statements.
2014-02-15 17:27:13 +00:00
ckormanyos
8acdc26d4e
Merge branch 'cstdfloat' of https://github.com/boostorg/math into cstdfloat
2014-02-15 15:15:46 +01:00
ckormanyos
4ca0ca17f7
Clean up <boost/cstdfloat.hpp> and do further work on test file.
2014-02-15 15:13:33 +01:00
jzmaddock
118a93e3a6
Add tests for cstdfloat.hpp defined types.
...
Fix Math code to not use lexical_cast for constants for boost::floatmax_t when that type is __float128.
Fix constant usage so it doesn't break when the suffix is "Q".
2014-02-15 11:14:00 +00:00
jzmaddock
b78281ec40
Merge branch 'develop' of https://github.com/boostorg/math into develop
2014-02-15 11:09:07 +00:00
jzmaddock
69dc556802
Add missing return statements.
2014-02-15 11:08:55 +00:00
ckormanyos
c8c2a76ebe
Add a *very* preliminary test file for cstdfloat. I don't even know if it compiles, let alone runs.
2014-02-15 00:14:53 +01:00
ckormanyos
54cd321d91
In <boost/cstdfloat.hpp>, do trivial cleanup.
2014-02-14 20:41:09 +01:00
ckormanyos
2c6d94b30e
In <boost/cstdfloat.hpp>, improve comments, correct typo in limits, and use boost::math::constants once again.
2014-02-14 15:22:05 +01:00
ckormanyos
26de94b06b
In <boost/cstdfloat.hpp>, re-work a conceptual error in complex<float128_t>. Temporarily remove the use of boost::math::constants.
2014-02-13 20:51:25 +01:00
ckormanyos
143870feb7
Merge branch 'develop' of https://github.com/boostorg/math into develop
2014-02-13 19:11:34 +01:00
ckormanyos
8dc420ac61
In <boost/cstdfloat.hpp>, corrected a conceptual error in complex<float128_t>.
2014-02-13 19:05:52 +01:00
jzmaddock
65675a57e1
Add abs overload for __float128.
2014-02-13 17:48:00 +00:00
jzmaddock
2c0eee7cd0
Quash some VC12 warnings.
2014-02-13 16:57:45 +00:00
jzmaddock
43f0f18591
Fix spurious overflows in tgamma/lgamma. Update tests to verify.
...
Thanks to Rocco Romeo for these.
2014-02-13 09:42:46 +00:00
ckormanyos
7e084a1ef2
In <boost/cstdfloat.h>, finish (moslty) support for *optional* comlpex<float128_t>.
2014-02-12 21:43:05 +01:00
ckormanyos
f1679a3629
Merge branch 'develop' of https://github.com/boostorg/math into develop
2014-02-12 01:01:06 +01:00
ckormanyos
f9b8cd6bed
In <boost/cstdfloat.hpp>, patch exp() function for __float128. Add most of <complex> for __float128. Still needs some wrinkle ironing.
2014-02-12 01:00:28 +01:00
jzmaddock
f9178d6f48
Fix some signed/unsigned inconsistencies (and warnings).
...
See https://svn.boost.org/trac/boost/ticket/9646 .
2014-02-11 13:20:25 +00:00
jzmaddock
91ea175da2
Remove register keyword: it's deprecated in C++11.
...
Fixes https://svn.boost.org/trac/boost/ticket/9639 .
2014-02-11 13:09:58 +00:00
PABristow
6bbb7114c2
commit change to ctsdfloat
2014-02-11 12:16:28 +00:00
jzmaddock
02462eb6a4
Merge branch 'develop' of https://github.com/boostorg/math into develop
2014-02-10 16:51:39 +00:00
jzmaddock
b0c9af5e61
Another attempt to accurately detect when we're in FTZ mode.
...
Fixes test failures for Win32 executables run on Win64.
2014-02-10 16:51:24 +00:00
ckormanyos
5eb0901b7b
In <boost/cstdfloat.hpp> add a preprocessor enable-switch for quadmath support.
2014-02-09 13:02:13 +01:00
ckormanyos
f44c883445
In <boost/cstdfloat.hpp>, GCC quadmath now has full I/O stream support. ICC quadmath I/O stream support is not yet implemented.
2014-02-08 21:36:07 +01:00
