hello all..
i am attempting to compile glibc for my platform (4xx core powerpc) and
as this core doesnt have an fpu, am compiling with -msoft-float.. but it
dont work.. during compile i receive this error:
../sysdeps/powerpc/s_isnan.c:27: Invalid `asm' statement:
../sysdeps/powerpc/s_isnan.c:27: fixed or forbidden register 32 (0) was
spilled for class FLOAT_REGS.
(this occurs on an attempted assemble of a fpu instruction)
my system specs are:
- compiling with gcc 2.95.2 native on powermac g4
- glibc 2.1.3
- kernel(powermac)=2.2.15pre14
- kernel(4xx)=2.3.42 + patches
- glibc configure options: configure --target=powerpc-linux
--host=powerpc-linux
--enable-add-ons=crypt, linuxthreads --prefix=/usr/local/ppc403
--without-fp
- CFLAGS extras: -msoft-float -DNDEBUG=1 -O2
does anyone have a clue as to what is going on?? will i have to add some
sort of software emulated floating point support in the kernel (use 8xx
emu)?? or am i just missing a compiler option??
is it worth using this late revision of glibc, or am i better off to use
an earlier/simpler revision??
cheers for any insights!
ben
** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/
