[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: native Win32 guile 1.7.0
From: |
Kevin Ryde |
Subject: |
Re: native Win32 guile 1.7.0 |
Date: |
Sat, 31 May 2003 10:20:56 +1000 |
User-agent: |
Gnus/5.090019 (Oort Gnus v0.19) Emacs/21.2 (gnu/linux) |
stefan <address@hidden> writes:
>
> With the last changes in 'numbers.h' all of the errors in the error.log I
> posted earlier disappear, i.e. using _copysign, _isnan and _finite solved
> the problems in the 'numbers.test'.
I'd suggest doing those in a system-neutral style too. _copysign is a
function I take it. In which case add autoconf tests and use
(untested),
/* mingw offers _copysign rather than copysign */
#if ! HAVE_COPYSIGN && HAVE__COPYSIGN
#define copysign _copysign
#define HAVE_COPYSIGN 1
#endif
Again, good for adapting automatically to the future. (Autoconf is
about testing for features, not system types.)
> * I do use a special 'guile.cache' file as cache file for the configure
> script. The file defines some tests to be done and chached. I think
> this is normal for a cross-build.
In principle hopefully not :-). I needed to force
ac_cv_sys_restartable_syscalls=no when I tried a cross compile a while
ago though.
And an EXEEXT_FOR_BUILD is needed for gen-scmconfig. I've been
meaning to propose some code I'd used for gmp in that area.
> * The 'fileblocks.o' must be removed in configure before configure'ing.
> This is due to the macro AC_STRUCT_ST_BLOCKS.
It's always annoying when autoconf does more than wanted. Maybe
AC_CHECK_MEMBER should be used instead.