FloatMathPlugin for Croquet

Andreas Raab andreas.raab at gmx.de
Mon Mar 27 04:18:26 UTC 2006


John M McIntosh wrote:
> LIkely you need to share your compiler options for x87 and ppc.

No compiler options on my end; I'm doing it at runtime via:

/* default fpu control word:
    _RC_NEAR: round to nearest
    _PC_53 :  double precision arithmetic (instead of extended)
    _EM_XXX: silent operations (no signals please)
*/
#define FPU_DEFAULT (_RC_NEAR + _PC_53 + _EM_INVALID + _EM_ZERODIVIDE + 
_EM_OVERFLOW + _EM_UNDERFLOW + _EM_INEXACT + _EM_DENORMAL)


and then (before doing much else):

   /* set default fpu control word */
   _control87(FPU_DEFAULT, _MCW_EM | _MCW_RC | _MCW_PC | _MCW_IC);


Cheers,
   - Andreas



More information about the Vm-dev mailing list