Headers/ppc_wrappers/emmintrin.h

17    Since X86 SSE2 intrinsics mainly handles __m128i and __m128d type,
61 typedef double __m128d __attribute__((__vector_size__(16), __may_alias__));
73 extern __inline __m128d
76   return __extension__(__m128d){__F, 0.0};
80 extern __inline __m128d
83   return __extension__(__m128d){__F, __F};
86 extern __inline __m128d
93 extern __inline __m128d
96   return __extension__(__m128d){__X, __W};
100 extern __inline __m128d
103   return __extension__(__m128d){__W, __X};
107 extern __inline __m128d
110   __m128d __Y = __Y;
115 extern __inline __m128d
118   return (__m128d)vec_splats(0);
122 extern __inline __m128d
124     _mm_move_sd(__m128d __A, __m128d __B) {
127   return (__m128d)__result;
131 extern __inline __m128d
134   return ((__m128d)vec_ld(0, (__v16qu *)__P));
138 extern __inline __m128d
145 extern __inline __m128d
152 extern __inline __m128d
158 extern __inline __m128d
165 extern __inline __m128d
169   return (__m128d)vec_xxpermdi(__tmp, __tmp, 2);
175     _mm_store_pd(double *__P, __m128d __A) {
182     _mm_storeu_pd(double *__P, __m128d __A) {
189     _mm_store_sd(double *__P, __m128d __A) {
195     _mm_cvtsd_f64(__m128d __A) {
201     _mm_storel_pd(double *__P, __m128d __A) {
208     _mm_storeh_pd(double *__P, __m128d __A) {
215     _mm_store1_pd(double *__P, __m128d __A) {
221     _mm_store_pd1(double *__P, __m128d __A) {
228     _mm_storer_pd(double *__P, __m128d __A) {
246 extern __inline __m128d
248     _mm_add_pd(__m128d __A, __m128d __B) {
249   return (__m128d)((__v2df)__A + (__v2df)__B);
255 extern __inline __m128d
257     _mm_add_sd(__m128d __A, __m128d __B) {
262 extern __inline __m128d
264     _mm_sub_pd(__m128d __A, __m128d __B) {
265   return (__m128d)((__v2df)__A - (__v2df)__B);
268 extern __inline __m128d
270     _mm_sub_sd(__m128d __A, __m128d __B) {
275 extern __inline __m128d
277     _mm_mul_pd(__m128d __A, __m128d __B) {
278   return (__m128d)((__v2df)__A * (__v2df)__B);
281 extern __inline __m128d
283     _mm_mul_sd(__m128d __A, __m128d __B) {
288 extern __inline __m128d
290     _mm_div_pd(__m128d __A, __m128d __B) {
291   return (__m128d)((__v2df)__A / (__v2df)__B);
294 extern __inline __m128d
296     _mm_div_sd(__m128d __A, __m128d __B) {
301 extern __inline __m128d
303     _mm_sqrt_pd(__m128d __A) {
308 extern __inline __m128d
310     _mm_sqrt_sd(__m128d __A, __m128d __B) {
313   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
316 extern __inline __m128d
318     _mm_min_pd(__m128d __A, __m128d __B) {
322 extern __inline __m128d
324     _mm_min_sd(__m128d __A, __m128d __B) {
329   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
332 extern __inline __m128d
334     _mm_max_pd(__m128d __A, __m128d __B) {
338 extern __inline __m128d
340     _mm_max_sd(__m128d __A, __m128d __B) {
345   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
348 extern __inline __m128d
350     _mm_cmpeq_pd(__m128d __A, __m128d __B) {
351   return ((__m128d)vec_cmpeq((__v2df)__A, (__v2df)__B));
354 extern __inline __m128d
356     _mm_cmplt_pd(__m128d __A, __m128d __B) {
357   return ((__m128d)vec_cmplt((__v2df)__A, (__v2df)__B));
360 extern __inline __m128d
362     _mm_cmple_pd(__m128d __A, __m128d __B) {
363   return ((__m128d)vec_cmple((__v2df)__A, (__v2df)__B));
366 extern __inline __m128d
368     _mm_cmpgt_pd(__m128d __A, __m128d __B) {
369   return ((__m128d)vec_cmpgt((__v2df)__A, (__v2df)__B));
372 extern __inline __m128d
374     _mm_cmpge_pd(__m128d __A, __m128d __B) {
375   return ((__m128d)vec_cmpge((__v2df)__A, (__v2df)__B));
378 extern __inline __m128d
380     _mm_cmpneq_pd(__m128d __A, __m128d __B) {
382   return ((__m128d)vec_nor(__temp, __temp));
385 extern __inline __m128d
387     _mm_cmpnlt_pd(__m128d __A, __m128d __B) {
388   return ((__m128d)vec_cmpge((__v2df)__A, (__v2df)__B));
391 extern __inline __m128d
393     _mm_cmpnle_pd(__m128d __A, __m128d __B) {
394   return ((__m128d)vec_cmpgt((__v2df)__A, (__v2df)__B));
397 extern __inline __m128d
399     _mm_cmpngt_pd(__m128d __A, __m128d __B) {
400   return ((__m128d)vec_cmple((__v2df)__A, (__v2df)__B));
403 extern __inline __m128d
405     _mm_cmpnge_pd(__m128d __A, __m128d __B) {
406   return ((__m128d)vec_cmplt((__v2df)__A, (__v2df)__B));
409 extern __inline __m128d
411     _mm_cmpord_pd(__m128d __A, __m128d __B) {
417   return ((__m128d)vec_and(__c, __d));
420 extern __inline __m128d
422     _mm_cmpunord_pd(__m128d __A, __m128d __B) {
431   return ((__m128d)vec_orc(__c, __d));
440   return ((__m128d)vec_or(__c, __d));
444 extern __inline __m128d
446     _mm_cmpeq_sd(__m128d __A, __m128d __B) {
457   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
460 extern __inline __m128d
462     _mm_cmplt_sd(__m128d __A, __m128d __B) {
467   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
470 extern __inline __m128d
472     _mm_cmple_sd(__m128d __A, __m128d __B) {
477   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
480 extern __inline __m128d
482     _mm_cmpgt_sd(__m128d __A, __m128d __B) {
487   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
490 extern __inline __m128d
492     _mm_cmpge_sd(__m128d __A, __m128d __B) {
497   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
500 extern __inline __m128d
502     _mm_cmpneq_sd(__m128d __A, __m128d __B) {
508   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
511 extern __inline __m128d
513     _mm_cmpnlt_sd(__m128d __A, __m128d __B) {
519   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
522 extern __inline __m128d
524     _mm_cmpnle_sd(__m128d __A, __m128d __B) {
530   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
533 extern __inline __m128d
535     _mm_cmpngt_sd(__m128d __A, __m128d __B) {
541   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
544 extern __inline __m128d
546     _mm_cmpnge_sd(__m128d __A, __m128d __B) {
552   return (__m128d)_mm_setr_pd(__c[0], __A[1]);
555 extern __inline __m128d
557     _mm_cmpord_sd(__m128d __A, __m128d __B) {
560   return (__m128d)_mm_setr_pd(__r[0], ((__v2df)__A)[1]);
563 extern __inline __m128d
565     _mm_cmpunord_sd(__m128d __A, __m128d __B) {
568   return (__m128d)_mm_setr_pd(__r[0], __A[1]);
580     _mm_comieq_sd(__m128d __A, __m128d __B) {
586     _mm_comilt_sd(__m128d __A, __m128d __B) {
592     _mm_comile_sd(__m128d __A, __m128d __B) {
598     _mm_comigt_sd(__m128d __A, __m128d __B) {
604     _mm_comige_sd(__m128d __A, __m128d __B) {
610     _mm_comineq_sd(__m128d __A, __m128d __B) {
616     _mm_ucomieq_sd(__m128d __A, __m128d __B) {
622     _mm_ucomilt_sd(__m128d __A, __m128d __B) {
628     _mm_ucomile_sd(__m128d __A, __m128d __B) {
634     _mm_ucomigt_sd(__m128d __A, __m128d __B) {
640     _mm_ucomige_sd(__m128d __A, __m128d __B) {
646     _mm_ucomineq_sd(__m128d __A, __m128d __B) {
822 extern __inline __m128d
830   return (__m128d)vec_ctf(__val, 0);
842     _mm_cvtpd_epi32(__m128d __A) {
871     _mm_cvtpd_pi32(__m128d __A) {
879     _mm_cvtpd_ps(__m128d __A) {
906     _mm_cvttpd_epi32(__m128d __A) {
936     _mm_cvttpd_pi32(__m128d __A) {
949 extern __inline __m128d
959   return (__m128d)__result;
983 extern __inline __m128d
988   return (__m128d)vec_doubleh((__v4sf)__A);
1010   return (__m128d)__result;
1016     _mm_cvtsd_si32(__m128d __A) {
1025     _mm_cvtsd_si64(__m128d __A) {
1035     _mm_cvtsd_si64x(__m128d __A) {
1041     _mm_cvttsd_si32(__m128d __A) {
1050     _mm_cvttsd_si64(__m128d __A) {
1059     _mm_cvttsd_si64x(__m128d __A) {
1065     _mm_cvtsd_ss(__m128 __A, __m128d __B) {
1085 extern __inline __m128d
1087     _mm_cvtsi32_sd(__m128d __A, int __B) {
1091   return (__m128d)__result;
1095 extern __inline __m128d
1097     _mm_cvtsi64_sd(__m128d __A, long long __B) {
1101   return (__m128d)__result;
1105 extern __inline __m128d
1107     _mm_cvtsi64x_sd(__m128d __A, long long __B) {
1111 extern __inline __m128d
1113     _mm_cvtss_sd(__m128d __A, __m128 __B) {
1120   return (__m128d)vec_mergel(__res, (__v2df)__A);
1124   return (__m128d)__res;
1128 extern __inline __m128d
1130     _mm_shuffle_pd(__m128d __A, __m128d __B, const int __mask) {
1153 extern __inline __m128d
1155     _mm_unpackhi_pd(__m128d __A, __m128d __B) {
1156   return (__m128d)vec_mergel((__v2df)__A, (__v2df)__B);
1159 extern __inline __m128d
1161     _mm_unpacklo_pd(__m128d __A, __m128d __B) {
1162   return (__m128d)vec_mergeh((__v2df)__A, (__v2df)__B);
1165 extern __inline __m128d
1167     _mm_loadh_pd(__m128d __A, double const *__B) {
1170   return (__m128d)__result;
1173 extern __inline __m128d
1175     _mm_loadl_pd(__m128d __A, double const *__B) {
1178   return (__m128d)__result;
1187     _mm_movemask_pd(__m128d __A) {
1812 extern __inline __m128d
1814     _mm_and_pd(__m128d __A, __m128d __B) {
1818 extern __inline __m128d
1820     _mm_andnot_pd(__m128d __A, __m128d __B) {
1824 extern __inline __m128d
1826     _mm_or_pd(__m128d __A, __m128d __B) {
1830 extern __inline __m128d
1832     _mm_xor_pd(__m128d __A, __m128d __B) {
2179     _mm_stream_pd(double *__A, __m128d __B) {
2182   *(__m128d *)__A = __B;
2229     _mm_castpd_ps(__m128d __A) {
2235     _mm_castpd_si128(__m128d __A) {
2239 extern __inline __m128d
2242   return (__m128d)__A;
2257 extern __inline __m128d
2260   return (__m128d)__A;