glibc-2.1 on PowerPC cannot disable floating-point
Geoff Keating
geoffk@ozemail.com.au
Thu Apr 1 00:00:00 GMT 1999
> Date: Fri, 12 Feb 1999 10:05:15 -0500
> From: Zack Weinberg <zack@rabi.columbia.edu>
> It should suffice to create a sysdeps/powerpc/fpu directory and move
> all the floating-point-related files in sysdeps/powerpc to that
> directory. I think this is the list:
>
> Versions bits/fenv.h bits/mathdef.h bits/mathinline.h
> e_sqrt.c e_sqrtf.c fclrexcpt.c fe_nomask.c
> fegetenv.c fegetround.c feholdexcpt.c fenv_const.c
> fenv_libc.h fesetenv.c fesetround.c feupdateenv.c
> fgetexcptflg.c fpu_control.h fraiseexcpt.c fsetexcptflg.c
> ftestexcept.c machine-gmon.h s_copysign.S s_copysignf.S
> s_fabs.S s_fabsf.S s_fdim.c s_fdimf.c
> s_fmax.S s_fmaxf.S s_fmin.S s_fminf.S
> s_isnan.c s_isnanf.S s_llrint.c s_llrintf.c
> s_llround.c s_llroundf.c s_lrint.c s_lrintf.S
> s_lround.c s_lroundf.c s_rint.c s_rintf.c
> t_sqrt.c test-arith.c test-arithf.c w_sqrt.c
> w_sqrtf.c
>
> You'll also need to take ieee754 out of powerpc/Implies and add it to
> powerpc/fpu/Implies. Dist and Makefile may need to be adjusted.
Versions and machine-gmon.h are only partially FPU-dependent, and
you've missed the setjmp code which is also partially FPU-dependent.
If this is really important to people, I can move it up on my priority
list, but right now I'm a little busy. It also doesn't help that I
have no way to test the result, since I don't have any machines
without a FPU.
I don't understand what people want this for. What systems do you
have that are so large that they can run glibc, but so small they
can't run a FPU emulator???
--
Geoffrey Keating <geoffk@ozemail.com.au>
More information about the Libc-alpha
mailing list