mirror of
https://github.com/Mercury-Language/mercury.git
synced 2026-04-20 11:54:02 +00:00
886d2ae474
Estimated hours taken: 20
Branches: main
Make it easier to use shared libraries on x86 with
`mmc --make'.
There is now a third kind of object file, `.lpic_o'.
These files are compiled with `--pic-reg' but not with
CFLAGS_FOR_PIC, so they can be linked with shared Mercury
libraries.
On x86, executables which are linked with shared Mercury
libraries now depend on $(main_module.lpic_os), not
$(main_module.os).
This doesn't work with Mmake because ml doesn't know
which libraries are Mercury libraries, so it can't
link with the static versions of those libraries if
MERCURY_LINKAGE is set to "static".
configure.in:
bindist/bindist.configure.in:
bindist/bindist.build_vars.in:
Work out whether `.lpic_o' files are needed.
compiler/modules.m:
Add `.lpic_o' to the list of grade or architecture
dependent files.
NEWS:
README.Linux:
compiler/options.m:
doc/user_guide.texi:
Document MERCURY_LINKAGE, LINKAGE, --linkage,
--mercury-linkage and -R.
compiler/options_file.m:
compiler/make.program_target.m:
Handle LINKAGE and MERCURY_LINKAGE variables.
Allow LIBGRADES, LINKAGE and MERCURY_LINKAGE to be target-specific.
scripts/mmc.in:
Set up the default linkage using the MERCURY_LINKAGE
variable.
compiler/compile_target_code.m:
Build `.lpic_o' files.
Work out which type of object files to link with.
When linking statically with Mercury libraries,
find the absolute pathname for the `.a' file
for each Mercury library, and pass that to ml,
rather than just using `-lname'.
Pass `-R' options to ml for each `-R' option to mmc.
compiler/make.module_target.m:
compiler/make.program_target.m:
compiler/mercury_compile.m:
Specify which type of object files to build.
compiler/make.program_target.m:
compiler/make.module_target.m:
Make sure all generated object files are cleaned up.
compiler/prog_io.m:
Add a better message for files which can't be found.
compiler/make.util.m:
Add `.lpic_o' to the list of extensions.
compiler/Mmakefile:
profiler/Mmakefile:
deep_profiler/Mmakefile:
Pass `--linkage shared' to mmc (`--shared' is in MLFLAGS).
tests/Mmakefile:
tests/mmc_make/Mmakefile:
tests/mmc_make/Mercury.options:
tests/mmc_make/complex_test.{m,exp}:
tests/mmc_make/hello.{m,exp}:
Test `mmc --make'.
tests/lib/complex*.m:
The complex numbers library from the extras distribution,
for use in the mmc_make tests.
78 lines
2.1 KiB
Mathematica
78 lines
2.1 KiB
Mathematica
%---------------------------------------------------------------------------%
|
|
% Copyright (C) 1997-1998,2001 The University of Melbourne.
|
|
% This file may only be copied under the terms of the GNU Library General
|
|
% Public License - see the file COPYING.LIB in the Mercury distribution.
|
|
%---------------------------------------------------------------------------%
|
|
%
|
|
% File: imag.m.
|
|
% Main author: fjh.
|
|
% Stability: medium.
|
|
%
|
|
% Imaginary numbers.
|
|
%
|
|
% There are several reasons for supporting a separate type for imaginary
|
|
% numbers rather than just treating them as a special case of complex
|
|
% numbers. It is sometimes more convenient, and can be slightly more
|
|
% efficient. But perhaps the most important reason is to get correct
|
|
% handling of infinity and not-a-number on platforms that support IEEE
|
|
% floating point.
|
|
%
|
|
% Note that the overloaded versions of the binary operators which
|
|
% provide mixed type arithmetic are defined in different modules.
|
|
%
|
|
% See also:
|
|
% float.m, imag_float.m, float_imag.m,
|
|
% complex.m, imag_complex.m, complex_imag.m.
|
|
%
|
|
%---------------------------------------------------------------------------%
|
|
|
|
:- module complex_numbers:imag.
|
|
:- interface.
|
|
:- import_module float.
|
|
|
|
:- type imag ---> im(float).
|
|
|
|
:- func i = imag. % i = sqrt(-1)
|
|
:- func j = imag. % another name for `i'
|
|
|
|
% addition
|
|
:- func imag + imag = imag.
|
|
:- mode in + in = uo is det.
|
|
|
|
% subtraction
|
|
:- func imag - imag = imag.
|
|
:- mode in - in = uo is det.
|
|
|
|
% multiplication
|
|
:- func imag * imag = float.
|
|
:- mode in * in = uo is det.
|
|
|
|
% division
|
|
:- func imag / imag = float.
|
|
:- mode in / in = uo is det.
|
|
|
|
% unary plus
|
|
:- func + imag = imag.
|
|
:- mode + in = uo is det.
|
|
|
|
% unary minus
|
|
:- func - imag = imag.
|
|
:- mode - in = uo is det.
|
|
|
|
%---------------------------------------------------------------------------%
|
|
|
|
:- implementation.
|
|
|
|
i = im(1.0).
|
|
j = i.
|
|
|
|
+im(X) = im(X + 0.0).
|
|
-im(X) = im(-X).
|
|
im(X) + im(Y) = im(X + Y).
|
|
im(X) - im(Y) = im(X - Y).
|
|
im(X) * im(Y) = 0.0 - X * Y.
|
|
im(X) / im(Y) = X / Y.
|
|
|
|
%---------------------------------------------------------------------------%
|
|
%---------------------------------------------------------------------------%
|