Building GCC

[Angelo Graziosi <angelo dot graziosi at alice dot it>]

I have seen that when I build, on Cygwin (1.5), configuring as:

----------------------
${gcc_dir}/configure --prefix="${prefix_dir}" \
        --exec-prefix="${eprefix_dir}" \
	--sysconfdir="${sysconf_dir}" \
	--libdir="${lib_dir}" \
	--libexecdir="${libexec_dir}" \
	--mandir="${man_dir}" \
	--infodir="${info_dir}" \
	--program-suffix="${suffix}" \
	--enable-languages=c,c++,fortran \
	--enable-bootstrap \
	--enable-libgomp \
	--enable-libssp \
	--enable-threads \
	--enable-version-specific-runtime-libs \
	--enable-nls \
	--enable-checking=release \
	--disable-sjlj-exceptions \
	--disable-shared \
	--disable-win32-registry \
	--with-arch=i686 \
	--with-tune=generic \
	--with-dwarf2 \
	--with-system-zlib \
        --without-included-gettext \
	--without-x || return 1
----------------------

in the build log there are messages like these:

------------------------
[...]
checking for gettimeofday... (cached) yes
yes
checking for realpath... no
i686-pc-cygwin

checking for unistd.h... checking for sysmp... checking for decimal
floating point... dpd
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

checking whether byte ordering is bigendian... yes
yes
checking minix/config.h usability... no
[...]
ranlib libdecnumber.a
no
checking valgrind.h presence... yes
make[3]: Leaving directory `/tmp/build/libdecnumber'
checking for stdlib.h... (cached) yes
updating cache ../config.cache
no
checking for valgrind.h... no

configure: WARNING: decimal float is not supported for this target
checking for unistd.h... (cached) yes
configure: WARNING: fixed-point is not supported for this target
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

[...]
checking for /tmp/build/./gcc/xgcc -B/tmp/build/./gcc/
-B/usr/local/gfortran/i686-pc-cygwin/bin/
-B/usr/local/gfortran/i686-pc-cygwin/lib/ -isystem
/usr/local/gfortran/i686-pc-cygwin/include -isystem
/usr/local/gfortran/i686-pc-cygwin/sys-include option to accept ANSI
C... none needed
checking how to run the C preprocessor... cpp-4
checking whether decimal floating point is supported... no
checking whether fixed-point is supported... no
[...]
checking target system type... yes
checking for stddef.h... yes
i686-pc-cygwin
checking for stdlib.h... (cached) yes
checking for decimal floating point... dpd
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

[...]
------------------------

If I instead add to previous configure:

+       --enable-decimal-float=bid
+       --disable-fixed-point

then, there are NO warnings

---------------------------
[...]
checking for strsignal... checking for decimal floating point... bid
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

[...]
checking how to run the C preprocessor... cpp-4
checking whether decimal floating point is supported... yes
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

checking whether fixed-point is supported... no
[...]
---------------------------

and many files are compiled with '-DENABLE_DECIMAL_BID_FORMAT'.

So, is there an optimal choice for configuring?

What it the best? DPD or BID?


Thanks,
Angelo.