This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: IA64 cross compilation broken
- To: mark at codesourcery dot com
- Subject: Re: IA64 cross compilation broken
- From: Jim Wilson <wilson at cygnus dot com>
- Date: Wed, 2 Aug 2000 13:57:02 -0700
- Cc: gcc-bugs at gcc dot gnu dot org
In article <20000802133204Y.mitchell@codesourcery.com> you write:
>I can't seem to build a x86->ia64 cross compiler. I get:
>
> ../../gcc/libgcc2.h:99: no data type for mode `TI'
> ../../gcc/libgcc2.h:100: no data type for mode `TI'
I believe all cross compilation from 32-bit hosts to 64-bit targets is broken.
This is a consequence of two patches:
1) Kenner's patch to make the DImode routines in libgcc2.c do TImode
arithmetic for 64-bit targets (Apr 15).
1) John Anglin's change to disable TImode for 32-bit hosts (May 17).
This isn't an immediate problem for me, as I never build cross compilers to
IA-64 anymore, and I haven't had time to worry about it yet.
The IA-64 port doesn't have any 128-bit types, so we don't actually need any
TImode functions in libgcc2.c. But then the Alpha doesn't either as far as
I know. Maybe Kenner needed them for Ada?
At Red Hat, we have a configure hack that sets HOST_WIDE_INT to "long long" if
the host is a 32-bit machine, and the target is a 64-bit machine. That makes
the cross compilers work at the expense of getting bigger and slower cc1
binaries. I don't know if any other changes are required other than the
configure hack. Oh, it does require that you use gcc to compile the cross
compiler.
Jim