Headers/ppc_wrappers/xmmintrin.h

198     _mm_move_ss(__m128 __A, __m128 __B) {
201   return (vec_sel((__v4sf)__A, (__v4sf)__B, __mask));
224     _mm_add_ss(__m128 __A, __m128 __B) {
233   __b = vec_splat(__B, 0);
239   __A[0] = __A[0] + __B[0];
246     _mm_sub_ss(__m128 __A, __m128 __B) {
255   __b = vec_splat(__B, 0);
261   __A[0] = __A[0] - __B[0];
268     _mm_mul_ss(__m128 __A, __m128 __B) {
277   __b = vec_splat(__B, 0);
283   __A[0] = __A[0] * __B[0];
290     _mm_div_ss(__m128 __A, __m128 __B) {
299   __b = vec_splat(__B, 0);
305   __A[0] = __A[0] / __B[0];
329     _mm_add_ps(__m128 __A, __m128 __B) {
330   return (__m128)((__v4sf)__A + (__v4sf)__B);
335     _mm_sub_ps(__m128 __A, __m128 __B) {
336   return (__m128)((__v4sf)__A - (__v4sf)__B);
341     _mm_mul_ps(__m128 __A, __m128 __B) {
342   return (__m128)((__v4sf)__A * (__v4sf)__B);
347     _mm_div_ps(__m128 __A, __m128 __B) {
348   return (__m128)((__v4sf)__A / (__v4sf)__B);
403     _mm_min_ss(__m128 __A, __m128 __B) {
411   __b = vec_splat((__v4sf)__B, 0);
420     _mm_max_ss(__m128 __A, __m128 __B) {
428   __b = vec_splat(__B, 0);
437     _mm_min_ps(__m128 __A, __m128 __B) {
438   __vector __bool int __m = vec_cmpgt((__v4sf)__B, (__v4sf)__A);
439   return vec_sel(__B, __A, __m);
444     _mm_max_ps(__m128 __A, __m128 __B) {
445   __vector __bool int __m = vec_cmpgt((__v4sf)__A, (__v4sf)__B);
446   return vec_sel(__B, __A, __m);
452     _mm_and_ps(__m128 __A, __m128 __B) {
453   return ((__m128)vec_and((__v4sf)__A, (__v4sf)__B));
454   //  return __builtin_ia32_andps (__A, __B);
459     _mm_andnot_ps(__m128 __A, __m128 __B) {
460   return ((__m128)vec_andc((__v4sf)__B, (__v4sf)__A));
465     _mm_or_ps(__m128 __A, __m128 __B) {
466   return ((__m128)vec_or((__v4sf)__A, (__v4sf)__B));
471     _mm_xor_ps(__m128 __A, __m128 __B) {
472   return ((__m128)vec_xor((__v4sf)__A, (__v4sf)__B));
480     _mm_cmpeq_ps(__m128 __A, __m128 __B) {
481   return ((__m128)vec_cmpeq((__v4sf)__A, (__v4sf)__B));
486     _mm_cmplt_ps(__m128 __A, __m128 __B) {
487   return ((__m128)vec_cmplt((__v4sf)__A, (__v4sf)__B));
492     _mm_cmple_ps(__m128 __A, __m128 __B) {
493   return ((__m128)vec_cmple((__v4sf)__A, (__v4sf)__B));
498     _mm_cmpgt_ps(__m128 __A, __m128 __B) {
499   return ((__m128)vec_cmpgt((__v4sf)__A, (__v4sf)__B));
504     _mm_cmpge_ps(__m128 __A, __m128 __B) {
505   return ((__m128)vec_cmpge((__v4sf)__A, (__v4sf)__B));
510     _mm_cmpneq_ps(__m128 __A, __m128 __B) {
511   __v4sf __temp = (__v4sf)vec_cmpeq((__v4sf)__A, (__v4sf)__B);
517     _mm_cmpnlt_ps(__m128 __A, __m128 __B) {
518   return ((__m128)vec_cmpge((__v4sf)__A, (__v4sf)__B));
523     _mm_cmpnle_ps(__m128 __A, __m128 __B) {
524   return ((__m128)vec_cmpgt((__v4sf)__A, (__v4sf)__B));
529     _mm_cmpngt_ps(__m128 __A, __m128 __B) {
530   return ((__m128)vec_cmple((__v4sf)__A, (__v4sf)__B));
535     _mm_cmpnge_ps(__m128 __A, __m128 __B) {
536   return ((__m128)vec_cmplt((__v4sf)__A, (__v4sf)__B));
541     _mm_cmpord_ps(__m128 __A, __m128 __B) {
548   __b = (__vector unsigned int)vec_abs((__v4sf)__B);
556     _mm_cmpunord_ps(__m128 __A, __m128 __B) {
563   __b = (__vector unsigned int)vec_abs((__v4sf)__B);
574     _mm_cmpeq_ss(__m128 __A, __m128 __B) {
582   __b = vec_splat((__v4sf)__B, 0);
591     _mm_cmplt_ss(__m128 __A, __m128 __B) {
599   __b = vec_splat((__v4sf)__B, 0);
608     _mm_cmple_ss(__m128 __A, __m128 __B) {
616   __b = vec_splat((__v4sf)__B, 0);
625     _mm_cmpgt_ss(__m128 __A, __m128 __B) {
633   __b = vec_splat((__v4sf)__B, 0);
642     _mm_cmpge_ss(__m128 __A, __m128 __B) {
650   __b = vec_splat((__v4sf)__B, 0);
659     _mm_cmpneq_ss(__m128 __A, __m128 __B) {
667   __b = vec_splat((__v4sf)__B, 0);
677     _mm_cmpnlt_ss(__m128 __A, __m128 __B) {
685   __b = vec_splat((__v4sf)__B, 0);
694     _mm_cmpnle_ss(__m128 __A, __m128 __B) {
702   __b = vec_splat((__v4sf)__B, 0);
711     _mm_cmpngt_ss(__m128 __A, __m128 __B) {
719   __b = vec_splat((__v4sf)__B, 0);
728     _mm_cmpnge_ss(__m128 __A, __m128 __B) {
736   __b = vec_splat((__v4sf)__B, 0);
745     _mm_cmpord_ss(__m128 __A, __m128 __B) {
753   __b = (__vector unsigned int)vec_abs((__v4sf)__B);
764     _mm_cmpunord_ss(__m128 __A, __m128 __B) {
772   __b = (__vector unsigned int)vec_abs((__v4sf)__B);
785     _mm_comieq_ss(__m128 __A, __m128 __B) {
786   return (__A[0] == __B[0]);
791     _mm_comilt_ss(__m128 __A, __m128 __B) {
792   return (__A[0] < __B[0]);
797     _mm_comile_ss(__m128 __A, __m128 __B) {
798   return (__A[0] <= __B[0]);
803     _mm_comigt_ss(__m128 __A, __m128 __B) {
804   return (__A[0] > __B[0]);
809     _mm_comige_ss(__m128 __A, __m128 __B) {
810   return (__A[0] >= __B[0]);
815     _mm_comineq_ss(__m128 __A, __m128 __B) {
816   return (__A[0] != __B[0]);
829     _mm_ucomieq_ss(__m128 __A, __m128 __B) {
830   return (__A[0] == __B[0]);
835     _mm_ucomilt_ss(__m128 __A, __m128 __B) {
836   return (__A[0] < __B[0]);
841     _mm_ucomile_ss(__m128 __A, __m128 __B) {
842   return (__A[0] <= __B[0]);
847     _mm_ucomigt_ss(__m128 __A, __m128 __B) {
848   return (__A[0] > __B[0]);
853     _mm_ucomige_ss(__m128 __A, __m128 __B) {
854   return (__A[0] >= __B[0]);
859     _mm_ucomineq_ss(__m128 __A, __m128 __B) {
860   return (__A[0] != __B[0]);
1033     _mm_cvtsi32_ss(__m128 __A, int __B) {
1034   float __temp = __B;
1042     _mm_cvt_si2ss(__m128 __A, int __B) {
1043   return _mm_cvtsi32_ss(__A, __B);
1050     _mm_cvtsi64_ss(__m128 __A, long long __B) {
1051   float __temp = __B;
1060     _mm_cvtsi64x_ss(__m128 __A, long long __B) {
1061   return _mm_cvtsi64_ss(__A, __B);
1068     _mm_cvtpi32_ps(__m128 __A, __m64 __B) {
1072   __vm1 = (__vector signed int)(__vector unsigned long long){__B, __B};
1082     _mm_cvt_pi2ps(__m128 __A, __m64 __B) {
1083   return _mm_cvtpi32_ps(__A, __B);
1168     _mm_cvtpi32x2_ps(__m64 __A, __m64 __B) {
1172   __vi4 = (__vector signed int)(__vector unsigned long long){__A, __B};
1213     _mm_shuffle_ps(__m128 __A, __m128 __B, int const __mask) {
1231   return vec_perm((__v4sf)__A, (__v4sf)__B, (__vector unsigned char)__t);
1237     _mm_unpackhi_ps(__m128 __A, __m128 __B) {
1238   return (__m128)vec_vmrglw((__v4sf)__A, (__v4sf)__B);
1244     _mm_unpacklo_ps(__m128 __A, __m128 __B) {
1245   return (__m128)vec_vmrghw((__v4sf)__A, (__v4sf)__B);
1272     _mm_movehl_ps(__m128 __A, __m128 __B) {
1273   return (__m128)vec_mergel((__vector unsigned long long)__B,
1280     _mm_movelh_ps(__m128 __A, __m128 __B) {
1282                             (__vector unsigned long long)__B);
1391     _mm_max_pi16(__m64 __A, __m64 __B) {
1397   __b = (__vector signed short)vec_splats(__B);
1405   __m2.as_m64 = __B;
1422     _m_pmaxsw(__m64 __A, __m64 __B) {
1423   return _mm_max_pi16(__A, __B);
1429     _mm_max_pu8(__m64 __A, __m64 __B) {
1435   __b = (__vector unsigned char)vec_splats(__B);
1444   __m2.as_m64 = __B;
1458     _m_pmaxub(__m64 __A, __m64 __B) {
1459   return _mm_max_pu8(__A, __B);
1465     _mm_min_pi16(__m64 __A, __m64 __B) {
1471   __b = (__vector signed short)vec_splats(__B);
1479   __m2.as_m64 = __B;
1496     _m_pminsw(__m64 __A, __m64 __B) {
1497   return _mm_min_pi16(__A, __B);
1503     _mm_min_pu8(__m64 __A, __m64 __B) {
1509   __b = (__vector unsigned char)vec_splats(__B);
1518   __m2.as_m64 = __B;
1532     _m_pminub(__m64 __A, __m64 __B) {
1533   return _mm_min_pu8(__A, __B);
1572     _mm_mulhi_pu16(__m64 __A, __m64 __B) {
1587   __b = (__vector unsigned short)vec_splats(__B);
1598     _m_pmulhuw(__m64 __A, __m64 __B) {
1599   return _mm_mulhi_pu16(__A, __B);
1669     _mm_avg_pu8(__m64 __A, __m64 __B) {
1673   __b = (__vector unsigned char)vec_splats(__B);
1680     _m_pavgb(__m64 __A, __m64 __B) {
1681   return _mm_avg_pu8(__A, __B);
1687     _mm_avg_pu16(__m64 __A, __m64 __B) {
1691   __b = (__vector unsigned short)vec_splats(__B);
1698     _m_pavgw(__m64 __A, __m64 __B) {
1699   return _mm_avg_pu16(__A, __B);
1707     _mm_sad_pu8(__m64 __A, __m64 __B) {
1715   __b = (__vector unsigned char)(__vector unsigned long long){0UL, __B};
1731     _m_psadbw(__m64 __A, __m64 __B) {
1732   return _mm_sad_pu8(__A, __B);