There is another way to do this , make gcc accept the -m32 -m64 options , as gcc is normally a "script" when the script has an -m32 option it calls our current gcc and when our script has a -m64 option it calls x86_64-w64-mingw32-gcc , same for all the other necessary build tools. This way ./configure should work with the ABI option as well
Jason ================================================================================= Hi I properly installed MinGW64 and built MPIR-2.5.0 and MPIR-2.5.1 , there were two trivial corrections that were needed but otherwise all was fine but I've only tried a few combinations so far.MinGW64 and 32 both support fat builds , although the 64bit fat build may break WIN64 exception?/epilog?/prolog? rules , but it appears to work. The main problem people seem to be having is that they are building MPIR the wrong way. The cpu detection is unique to MPIR/GMP and autotools knows nothing about it , nor does autotools recognize sandybridge or nehalem or k10 The correct way is to tell configure the toolchain ie /configure CC=x86_64-w64-mingw32-gcc.exe NM=x86_64-w64-mingw32-nm.exe AR=x86_64-w64-mingw32-ar.exe CXX=x86_64-w64-mingw32-g++.exe RANLIB=x86_64-w64-mingw32-ranlib.exe shared builds should present no problem either , but I'm still having a few hiccups with C++ I issue a 2.5.2-rc0 later , and run a full set of tests on sandybridge , I do have code that will test all cpus , but the runtime is probably a week :) thanks Jason -- You received this message because you are subscribed to the Google Groups "mpir-devel" group. To post to this group, send email to mpir-devel@googlegroups.com. To unsubscribe from this group, send email to mpir-devel+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/mpir-devel?hl=en.
