altivec.h

linux下编程用 编译软件

#define vec_sr __builtin_vec_sr
#define vec_sra __builtin_vec_sra
#define vec_srl __builtin_vec_srl
#define vec_sro __builtin_vec_sro
#define vec_st __builtin_vec_st
#define vec_ste __builtin_vec_ste
#define vec_stl __builtin_vec_stl
#define vec_stvebx __builtin_vec_stvebx
#define vec_stvehx __builtin_vec_stvehx
#define vec_stvewx __builtin_vec_stvewx
#define vec_sub __builtin_vec_sub
#define vec_subs __builtin_vec_subs
#define vec_sum __builtin_vec_sum
#define vec_unpackh __builtin_vec_unpackh
#define vec_unpackl __builtin_vec_unpackl
#define vec_vaddubm __builtin_vec_vaddubm
#define vec_vaddubs __builtin_vec_vaddubs
#define vec_vadduhm __builtin_vec_vadduhm
#define vec_vadduhs __builtin_vec_vadduhs
#define vec_vadduwm __builtin_vec_vadduwm
#define vec_vadduws __builtin_vec_vadduws
#define vec_vcmpequb __builtin_vec_vcmpequb
#define vec_vcmpequh __builtin_vec_vcmpequh
#define vec_vcmpequw __builtin_vec_vcmpequw
#define vec_vmaxub __builtin_vec_vmaxub
#define vec_vmaxuh __builtin_vec_vmaxuh
#define vec_vmaxuw __builtin_vec_vmaxuw
#define vec_vminub __builtin_vec_vminub
#define vec_vminuh __builtin_vec_vminuh
#define vec_vminuw __builtin_vec_vminuw
#define vec_vmrghb __builtin_vec_vmrghb
#define vec_vmrghh __builtin_vec_vmrghh
#define vec_vmrghw __builtin_vec_vmrghw
#define vec_vmrglb __builtin_vec_vmrglb
#define vec_vmrglh __builtin_vec_vmrglh
#define vec_vmrglw __builtin_vec_vmrglw
#define vec_vpkuhum __builtin_vec_vpkuhum
#define vec_vpkuwum __builtin_vec_vpkuwum
#define vec_vrlb __builtin_vec_vrlb
#define vec_vrlh __builtin_vec_vrlh
#define vec_vrlw __builtin_vec_vrlw
#define vec_vslb __builtin_vec_vslb
#define vec_vslh __builtin_vec_vslh
#define vec_vslw __builtin_vec_vslw
#define vec_vspltb __builtin_vec_vspltb
#define vec_vsplth __builtin_vec_vsplth
#define vec_vspltw __builtin_vec_vspltw
#define vec_vsrab __builtin_vec_vsrab
#define vec_vsrah __builtin_vec_vsrah
#define vec_vsraw __builtin_vec_vsraw
#define vec_vsrb __builtin_vec_vsrb
#define vec_vsrh __builtin_vec_vsrh
#define vec_vsrw __builtin_vec_vsrw
#define vec_vsububs __builtin_vec_vsububs
#define vec_vsububm __builtin_vec_vsububm
#define vec_vsubuhm __builtin_vec_vsubuhm
#define vec_vsubuhs __builtin_vec_vsubuhs
#define vec_vsubuwm __builtin_vec_vsubuwm
#define vec_vsubuws __builtin_vec_vsubuws
#define vec_xor __builtin_vec_xor

/* Predicates.
   For C++, we use templates in order to allow non-parenthesized arguments.
   For C, instead, we use macros since non-parenthesized arguments were
   not allowed even in older GCC implementation of AltiVec.

   In the future, we may add more magic to the back-end, so that no
   one- or two-argument macros are used.  */

#ifdef __cplusplus__
#define __altivec_unary_pred(NAME, CALL) \
template <class T> int NAME (T a1) { return CALL; }

#define __altivec_scalar_pred(NAME, CALL) \
template <class T, class U> int NAME (T a1, U a2) { return CALL; }

/* Given the vec_step of a type, return the corresponding bool type.  */
template <int STEP> class __altivec_bool_ret { };
template <> class __altivec_bool_ret <4> {
  typedef __vector __bool int __ret;
};
template <> class __altivec_bool_ret <8> {
  typedef __vector __bool short __ret;
};
template <> class __altivec_bool_ret <16> {
  typedef __vector __bool char __ret;
};

/* Be very liberal in the pairs we accept.  Mistakes such as passing
   a `vector char' and `vector short' will be caught by the middle-end,
   while any attempt to detect them here would produce hard to understand
   error messages involving the implementation details of AltiVec.  */
#define __altivec_binary_pred(NAME, CALL) \
template <class T, class U> \
typename __altivec_bool_ret <vec_step (T)>::__ret \
NAME (T a1, U a2) \
{ \
  return CALL; \
}

__altivec_binary_pred(vec_cmplt,
  __builtin_vec_cmpgt (a2, a1))
__altivec_binary_pred(vec_cmple,
  __builtin_altivec_cmpge (a2, a1))

__altivec_scalar_pred(vec_all_in,
  __builtin_altivec_vcmpbfp_p (__CR6_EQ, a1, a2))
__altivec_scalar_pred(vec_any_out,
  __builtin_altivec_vcmpbfp_p (__CR6_EQ_REV, a1, a2))

__altivec_unary_pred(vec_all_nan,
  __builtin_altivec_vcmpeqfp_p (__CR6_EQ, a1, a1))
__altivec_unary_pred(vec_any_nan,
  __builtin_altivec_vcmpeqfp_p (__CR6_LT_REV, a1, a1))

__altivec_unary_pred(vec_all_numeric,
  __builtin_altivec_vcmpeqfp_p (__CR6_LT, a1, a1))
__altivec_unary_pred(vec_any_numeric,
  __builtin_altivec_vcmpeqfp_p (__CR6_EQ_REV, a1, a1))

__altivec_scalar_pred(vec_all_eq,
  __builtin_vec_vcmpeq_p (__CR6_LT, a1, a2))
__altivec_scalar_pred(vec_all_ne,
  __builtin_vec_vcmpeq_p (__CR6_EQ, a1, a2))
__altivec_scalar_pred(vec_any_eq,
  __builtin_vec_vcmpeq_p (__CR6_EQ_REV, a1, a2))
__altivec_scalar_pred(vec_any_ne,
  __builtin_vec_vcmpeq_p (__CR6_LT_REV, a1, a2))

__altivec_scalar_pred(vec_all_gt,
  __builtin_vec_vcmpgt_p (__CR6_LT, a1, a2))
__altivec_scalar_pred(vec_all_lt,
  __builtin_vec_vcmpgt_p (__CR6_LT, a2, a1))
__altivec_scalar_pred(vec_any_gt,
  __builtin_vec_vcmpgt_p (__CR6_EQ_REV, a1, a2))
__altivec_scalar_pred(vec_any_lt,
  __builtin_vec_vcmpgt_p (__CR6_EQ_REV, a2, a1))

__altivec_scalar_pred(vec_all_ngt,
  __builtin_altivec_vcmpgtfp_p (__CR6_EQ, a1, a2))
__altivec_scalar_pred(vec_all_nlt,
  __builtin_altivec_vcmpgtfp_p (__CR6_EQ, a2, a1))
__altivec_scalar_pred(vec_any_ngt,
  __builtin_altivec_vcmpgtfp_p (__CR6_LT_REV, a1, a2))
__altivec_scalar_pred(vec_any_nlt,
  __builtin_altivec_vcmpgtfp_p (__CR6_LT_REV, a2, a1))

/* __builtin_vec_vcmpge_p is vcmpgefp for floating-point vector types,
   while for integer types it is converted to __builtin_vec_vcmpgt_p,
   with inverted args and condition code.  */
__altivec_scalar_pred(vec_all_le,
  __builtin_vec_vcmpge_p (__CR6_LT, a2, a1))
__altivec_scalar_pred(vec_all_ge,
  __builtin_vec_vcmpge_p (__CR6_LT, a1, a2))
__altivec_scalar_pred(vec_any_le,
  __builtin_vec_vcmpge_p (__CR6_EQ_REV, a2, a1))
__altivec_scalar_pred(vec_any_ge,
  __builtin_vec_vcmpge_p (__CR6_EQ_REV, a1, a2))

__altivec_scalar_pred(vec_all_nge,
  __builtin_altivec_vcmpgefp_p (__CR6_EQ, a1, a2))
__altivec_scalar_pred(vec_all_nle,
  __builtin_altivec_vcmpgefp_p (__CR6_EQ, a2, a1))
__altivec_scalar_pred(vec_any_nge,
  __builtin_altivec_vcmpgefp_p (__CR6_LT_REV, a1, a2))
__altivec_scalar_pred(vec_any_nle,
  __builtin_altivec_vcmpgefp_p (__CR6_LT_REV, a2, a1))

#undef __altivec_scalar_pred
#undef __altivec_unary_pred
#undef __altivec_binary_pred
#else
#define vec_cmplt(a1, a2) __builtin_vec_cmpgt ((a2), (a1))
#define vec_cmple(a1, a2) __builtin_altivec_vcmpgefp ((a2), (a1))

#define vec_all_in(a1, a2) __builtin_altivec_vcmpbfp_p (__CR6_EQ, (a1), (a2))
#define vec_any_out(a1, a2) __builtin_altivec_vcmpbfp_p (__CR6_EQ_REV, (a1), (a2))

#define vec_all_nan(a1) __builtin_altivec_vcmpeqfp_p (__CR6_EQ, (a1), (a1))
#define vec_any_nan(a1) __builtin_altivec_vcmpeqfp_p (__CR6_LT_REV, (a1), (a1))

#define vec_all_numeric(a1) __builtin_altivec_vcmpeqfp_p (__CR6_LT, (a1), (a1))
#define vec_any_numeric(a1) __builtin_altivec_vcmpeqfp_p (__CR6_EQ_REV, (a1), (a1))

#define vec_all_eq(a1, a2) __builtin_vec_vcmpeq_p (__CR6_LT, (a1), (a2))
#define vec_all_ne(a1, a2) __builtin_vec_vcmpeq_p (__CR6_EQ, (a1), (a2))
#define vec_any_eq(a1, a2) __builtin_vec_vcmpeq_p (__CR6_EQ_REV, (a1), (a2))
#define vec_any_ne(a1, a2) __builtin_vec_vcmpeq_p (__CR6_LT_REV, (a1), (a2))

#define vec_all_gt(a1, a2) __builtin_vec_vcmpgt_p (__CR6_LT, (a1), (a2))
#define vec_all_lt(a1, a2) __builtin_vec_vcmpgt_p (__CR6_LT, (a2), (a1))
#define vec_any_gt(a1, a2) __builtin_vec_vcmpgt_p (__CR6_EQ_REV, (a1), (a2))
#define vec_any_lt(a1, a2) __builtin_vec_vcmpgt_p (__CR6_EQ_REV, (a2), (a1))

#define vec_all_ngt(a1, a2) __builtin_altivec_vcmpgtfp_p (__CR6_EQ, (a1), (a2))
#define vec_all_nlt(a1, a2) __builtin_altivec_vcmpgtfp_p (__CR6_EQ, (a2), (a1))
#define vec_any_ngt(a1, a2) __builtin_altivec_vcmpgtfp_p (__CR6_LT_REV, (a1), (a2))
#define vec_any_nlt(a1, a2) __builtin_altivec_vcmpgtfp_p (__CR6_LT_REV, (a2), (a1))

/* __builtin_vec_vcmpge_p is vcmpgefp for floating-point vector types,
   while for integer types it is converted to __builtin_vec_vcmpgt_p,
   with inverted args and condition code.  */
#define vec_all_le(a1, a2) __builtin_vec_vcmpge_p (__CR6_LT, (a2), (a1))
#define vec_all_ge(a1, a2) __builtin_vec_vcmpge_p (__CR6_LT, (a1), (a2))
#define vec_any_le(a1, a2) __builtin_vec_vcmpge_p (__CR6_EQ_REV, (a2), (a1))
#define vec_any_ge(a1, a2) __builtin_vec_vcmpge_p (__CR6_EQ_REV, (a1), (a2))

#define vec_all_nge(a1, a2) __builtin_altivec_vcmpgefp_p (__CR6_EQ, (a1), (a2))
#define vec_all_nle(a1, a2) __builtin_altivec_vcmpgefp_p (__CR6_EQ, (a2), (a1))
#define vec_any_nge(a1, a2) __builtin_altivec_vcmpgefp_p (__CR6_LT_REV, (a1), (a2))
#define vec_any_nle(a1, a2) __builtin_altivec_vcmpgefp_p (__CR6_LT_REV, (a2), (a1))
#endif

/* These do not accept vectors, so they do not have a __builtin_vec_*
   counterpart.  */
#define vec_dss(x) __builtin_altivec_dss((x))
#define vec_dssall() __builtin_altivec_dssall ()
#define vec_mfvscr() ((__vector unsigned short) __builtin_altivec_mfvscr ())
#define vec_splat_s8(x) __builtin_altivec_vspltisb ((x))
#define vec_splat_s16(x) __builtin_altivec_vspltish ((x))
#define vec_splat_s32(x) __builtin_altivec_vspltisw ((x))
#define vec_splat_u8(x) ((__vector unsigned char) vec_splat_s8 ((x)))
#define vec_splat_u16(x) ((__vector unsigned short) vec_splat_s16 ((x)))
#define vec_splat_u32(x) ((__vector unsigned int) vec_splat_s32 ((x)))

/* This also accepts a type for its parameter, so it is not enough
   to #define vec_step to __builtin_vec_step.  */
#define vec_step(x) __builtin_vec_step (* (__typeof__ (x) *) 0)

#endif /* _ALTIVEC_H */