Re: [Lightning] About jit_roundr_d_i

[Paulo César Pereira de Andrade]

Em 11 de setembro de 2010 11:00, Paulo César Pereira de Andrade

>> I am a bit wary of introducing new opcodes since that would have to be
>> done for PowerPC and SPARC too.

  It could be some kind of extension. For example, not offering sin/cos/tan/etc
when easily available looks wrong. It could still do something "weird" on
other platforms, like:
#define jit_sqrt_d(rd, rs) \
 (jit_prepare(0), \
  jit_prepare_d(1), \
    jit_pusharg_d(rs), \
  jit_finish(sqrt), \
  jit_retval_d(rd))

but things can get ugly if that invalidates JIT_FPR(n) register contents.

Paulo