On Mar  9 07:02, Tony Mechelynck wrote:
> On 08/03/09 10:53, Corinna Vinschen wrote:
> > The -mno-cygwin option was always just a hack.  You could have put this
> > hack into the Linux i686 compiler as well, but why would you?  Same for
> > Cygwin.  The -mno-cygwin option will be removed.  To build a Mingw
> > binary, which is logically a cross-build, you will need a
> > cross-compiler.  With the move to the new Cygwin 1.7 release and the
> > move to the latest gcc-4.x release, the -mno-cygwin option will be
> > replaced by a mingw cross-compiler which will become part of the distro.
> >
> > Does that make sense?
> >
> >
> > Corinna
> >
> 
> Yes, it does. Then IIUC the Make_cyg.mak can remain, with a different 
> compiler name (which is already a variable anyway IIUC), possibly a 
> different linker name (since we will need to use a cross-linker IIUC) 
> removing the -mno-cygwin argument to the compiler and the -mwindows 
> argument to the linker, and no other changes than these (which are 
> rather minor) IIUC.

Looks like YUC.  Make_cyg.mak appears to use $(CC) throughout for
compile and link stage anyway.  So it should be sufficient (grain/salt)
to call

  CC=i686-pc-mingw32-gcc make -f Make_cyg.mak

after the -mno-cygwin option has been removed.


Corinna

-- 
Corinna Vinschen
Cygwin Project Co-Leader
Red Hat

--~--~---------~--~----~------------~-------~--~----~
You received this message from the "vim_dev" maillist.
For more information, visit http://www.vim.org/maillist.php
-~----------~----~----~----~------~----~------~--~---

Raspunde prin e-mail lui