Lines Matching refs:__B
124 _mm_move_sd(__m128d __A, __m128d __B) {
126 __result[0] = ((__v2df)__B)[0];
248 _mm_add_pd(__m128d __A, __m128d __B) {
249 return (__m128d)((__v2df)__A + (__v2df)__B);
257 _mm_add_sd(__m128d __A, __m128d __B) {
258 __A[0] = __A[0] + __B[0];
264 _mm_sub_pd(__m128d __A, __m128d __B) {
265 return (__m128d)((__v2df)__A - (__v2df)__B);
270 _mm_sub_sd(__m128d __A, __m128d __B) {
271 __A[0] = __A[0] - __B[0];
277 _mm_mul_pd(__m128d __A, __m128d __B) {
278 return (__m128d)((__v2df)__A * (__v2df)__B);
283 _mm_mul_sd(__m128d __A, __m128d __B) {
284 __A[0] = __A[0] * __B[0];
290 _mm_div_pd(__m128d __A, __m128d __B) {
291 return (__m128d)((__v2df)__A / (__v2df)__B);
296 _mm_div_sd(__m128d __A, __m128d __B) {
297 __A[0] = __A[0] / __B[0];
310 _mm_sqrt_sd(__m128d __A, __m128d __B) {
312 __c = vec_sqrt((__v2df)_mm_set1_pd(__B[0]));
318 _mm_min_pd(__m128d __A, __m128d __B) {
319 return (vec_min(__A, __B));
324 _mm_min_sd(__m128d __A, __m128d __B) {
327 __b = vec_splats(__B[0]);
334 _mm_max_pd(__m128d __A, __m128d __B) {
335 return (vec_max(__A, __B));
340 _mm_max_sd(__m128d __A, __m128d __B) {
343 __b = vec_splats(__B[0]);
350 _mm_cmpeq_pd(__m128d __A, __m128d __B) {
351 return ((__m128d)vec_cmpeq((__v2df)__A, (__v2df)__B));
356 _mm_cmplt_pd(__m128d __A, __m128d __B) {
357 return ((__m128d)vec_cmplt((__v2df)__A, (__v2df)__B));
362 _mm_cmple_pd(__m128d __A, __m128d __B) {
363 return ((__m128d)vec_cmple((__v2df)__A, (__v2df)__B));
368 _mm_cmpgt_pd(__m128d __A, __m128d __B) {
369 return ((__m128d)vec_cmpgt((__v2df)__A, (__v2df)__B));
374 _mm_cmpge_pd(__m128d __A, __m128d __B) {
375 return ((__m128d)vec_cmpge((__v2df)__A, (__v2df)__B));
380 _mm_cmpneq_pd(__m128d __A, __m128d __B) {
381 __v2df __temp = (__v2df)vec_cmpeq((__v2df)__A, (__v2df)__B);
387 _mm_cmpnlt_pd(__m128d __A, __m128d __B) {
388 return ((__m128d)vec_cmpge((__v2df)__A, (__v2df)__B));
393 _mm_cmpnle_pd(__m128d __A, __m128d __B) {
394 return ((__m128d)vec_cmpgt((__v2df)__A, (__v2df)__B));
399 _mm_cmpngt_pd(__m128d __A, __m128d __B) {
400 return ((__m128d)vec_cmple((__v2df)__A, (__v2df)__B));
405 _mm_cmpnge_pd(__m128d __A, __m128d __B) {
406 return ((__m128d)vec_cmplt((__v2df)__A, (__v2df)__B));
411 _mm_cmpord_pd(__m128d __A, __m128d __B) {
415 __d = (__v2du)vec_cmpeq(__B, __B);
422 _mm_cmpunord_pd(__m128d __A, __m128d __B) {
427 __d = (__v2du)vec_cmpeq((__v2df)__B, (__v2df)__B);
436 __d = (__v2du)vec_cmpeq((__v2df)__B, (__v2df)__B);
446 _mm_cmpeq_sd(__m128d __A, __m128d __B) {
453 __b = vec_splats(__B[0]);
462 _mm_cmplt_sd(__m128d __A, __m128d __B) {
465 __b = vec_splats(__B[0]);
472 _mm_cmple_sd(__m128d __A, __m128d __B) {
475 __b = vec_splats(__B[0]);
482 _mm_cmpgt_sd(__m128d __A, __m128d __B) {
485 __b = vec_splats(__B[0]);
492 _mm_cmpge_sd(__m128d __A, __m128d __B) {
495 __b = vec_splats(__B[0]);
502 _mm_cmpneq_sd(__m128d __A, __m128d __B) {
505 __b = vec_splats(__B[0]);
513 _mm_cmpnlt_sd(__m128d __A, __m128d __B) {
516 __b = vec_splats(__B[0]);
524 _mm_cmpnle_sd(__m128d __A, __m128d __B) {
527 __b = vec_splats(__B[0]);
535 _mm_cmpngt_sd(__m128d __A, __m128d __B) {
538 __b = vec_splats(__B[0]);
546 _mm_cmpnge_sd(__m128d __A, __m128d __B) {
549 __b = vec_splats(__B[0]);
557 _mm_cmpord_sd(__m128d __A, __m128d __B) {
559 __r = (__v2df)_mm_cmpord_pd(vec_splats(__A[0]), vec_splats(__B[0]));
565 _mm_cmpunord_sd(__m128d __A, __m128d __B) {
567 __r = _mm_cmpunord_pd(vec_splats(__A[0]), vec_splats(__B[0]));
580 _mm_comieq_sd(__m128d __A, __m128d __B) {
581 return (__A[0] == __B[0]);
586 _mm_comilt_sd(__m128d __A, __m128d __B) {
587 return (__A[0] < __B[0]);
592 _mm_comile_sd(__m128d __A, __m128d __B) {
593 return (__A[0] <= __B[0]);
598 _mm_comigt_sd(__m128d __A, __m128d __B) {
599 return (__A[0] > __B[0]);
604 _mm_comige_sd(__m128d __A, __m128d __B) {
605 return (__A[0] >= __B[0]);
610 _mm_comineq_sd(__m128d __A, __m128d __B) {
611 return (__A[0] != __B[0]);
616 _mm_ucomieq_sd(__m128d __A, __m128d __B) {
617 return (__A[0] == __B[0]);
622 _mm_ucomilt_sd(__m128d __A, __m128d __B) {
623 return (__A[0] < __B[0]);
628 _mm_ucomile_sd(__m128d __A, __m128d __B) {
629 return (__A[0] <= __B[0]);
634 _mm_ucomigt_sd(__m128d __A, __m128d __B) {
635 return (__A[0] > __B[0]);
640 _mm_ucomige_sd(__m128d __A, __m128d __B) {
641 return (__A[0] >= __B[0]);
646 _mm_ucomineq_sd(__m128d __A, __m128d __B) {
647 return (__A[0] != __B[0]);
772 _mm_store_si128(__m128i *__P, __m128i __B) {
773 vec_st((__v16qu)__B, 0, (__v16qu *)__P);
778 _mm_storeu_si128(__m128i_u *__P, __m128i __B) {
779 *__P = __B;
784 _mm_storel_epi64(__m128i_u *__P, __m128i __B) {
785 *(long long *)__P = ((__v2di)__B)[0];
790 _mm_movepi64_pi64(__m128i_u __B) {
791 return (__m64)((__v2di)__B)[0];
1065 _mm_cvtsd_ss(__m128 __A, __m128d __B) {
1071 __v2df __temp_b = vec_splat((__v2df)__B, 0);
1080 __result[0] = ((__v2df)__B)[0];
1087 _mm_cvtsi32_sd(__m128d __A, int __B) {
1089 double __db = __B;
1097 _mm_cvtsi64_sd(__m128d __A, long long __B) {
1099 double __db = __B;
1107 _mm_cvtsi64x_sd(__m128d __A, long long __B) {
1108 return _mm_cvtsi64_sd(__A, __B);
1113 _mm_cvtss_sd(__m128d __A, __m128 __B) {
1116 __v4sf __temp = vec_splat((__v4sf)__B, 0);
1123 __res[0] = ((__v4sf)__B)[0];
1130 _mm_shuffle_pd(__m128d __A, __m128d __B, const int __mask) {
1135 __result = vec_mergeh(__A, __B);
1138 __result = vec_xxpermdi(__B, __A, 2);
1140 __result = vec_xxpermdi(__B, __A, 1);
1143 __result = vec_xxpermdi(__A, __B, 2);
1145 __result = vec_xxpermdi(__A, __B, 1);
1148 __result = vec_mergel(__A, __B);
1155 _mm_unpackhi_pd(__m128d __A, __m128d __B) {
1156 return (__m128d)vec_mergel((__v2df)__A, (__v2df)__B);
1161 _mm_unpacklo_pd(__m128d __A, __m128d __B) {
1162 return (__m128d)vec_mergeh((__v2df)__A, (__v2df)__B);
1167 _mm_loadh_pd(__m128d __A, double const *__B) {
1169 __result[1] = *__B;
1175 _mm_loadl_pd(__m128d __A, double const *__B) {
1177 __result[0] = *__B;
1214 _mm_packs_epi16(__m128i __A, __m128i __B) {
1215 return (__m128i)vec_packs((__v8hi)__A, (__v8hi)__B);
1220 _mm_packs_epi32(__m128i __A, __m128i __B) {
1221 return (__m128i)vec_packs((__v4si)__A, (__v4si)__B);
1226 _mm_packus_epi16(__m128i __A, __m128i __B) {
1227 return (__m128i)vec_packsu((__v8hi)__A, (__v8hi)__B);
1232 _mm_unpackhi_epi8(__m128i __A, __m128i __B) {
1233 return (__m128i)vec_mergel((__v16qu)__A, (__v16qu)__B);
1238 _mm_unpackhi_epi16(__m128i __A, __m128i __B) {
1239 return (__m128i)vec_mergel((__v8hu)__A, (__v8hu)__B);
1244 _mm_unpackhi_epi32(__m128i __A, __m128i __B) {
1245 return (__m128i)vec_mergel((__v4su)__A, (__v4su)__B);
1250 _mm_unpackhi_epi64(__m128i __A, __m128i __B) {
1251 return (__m128i)vec_mergel((__vector long long)__A, (__vector long long)__B);
1256 _mm_unpacklo_epi8(__m128i __A, __m128i __B) {
1257 return (__m128i)vec_mergeh((__v16qu)__A, (__v16qu)__B);
1262 _mm_unpacklo_epi16(__m128i __A, __m128i __B) {
1263 return (__m128i)vec_mergeh((__v8hi)__A, (__v8hi)__B);
1268 _mm_unpacklo_epi32(__m128i __A, __m128i __B) {
1269 return (__m128i)vec_mergeh((__v4si)__A, (__v4si)__B);
1274 _mm_unpacklo_epi64(__m128i __A, __m128i __B) {
1275 return (__m128i)vec_mergeh((__vector long long)__A, (__vector long long)__B);
1280 _mm_add_epi8(__m128i __A, __m128i __B) {
1281 return (__m128i)((__v16qu)__A + (__v16qu)__B);
1286 _mm_add_epi16(__m128i __A, __m128i __B) {
1287 return (__m128i)((__v8hu)__A + (__v8hu)__B);
1292 _mm_add_epi32(__m128i __A, __m128i __B) {
1293 return (__m128i)((__v4su)__A + (__v4su)__B);
1298 _mm_add_epi64(__m128i __A, __m128i __B) {
1299 return (__m128i)((__v2du)__A + (__v2du)__B);
1304 _mm_adds_epi8(__m128i __A, __m128i __B) {
1305 return (__m128i)vec_adds((__v16qi)__A, (__v16qi)__B);
1310 _mm_adds_epi16(__m128i __A, __m128i __B) {
1311 return (__m128i)vec_adds((__v8hi)__A, (__v8hi)__B);
1316 _mm_adds_epu8(__m128i __A, __m128i __B) {
1317 return (__m128i)vec_adds((__v16qu)__A, (__v16qu)__B);
1322 _mm_adds_epu16(__m128i __A, __m128i __B) {
1323 return (__m128i)vec_adds((__v8hu)__A, (__v8hu)__B);
1328 _mm_sub_epi8(__m128i __A, __m128i __B) {
1329 return (__m128i)((__v16qu)__A - (__v16qu)__B);
1334 _mm_sub_epi16(__m128i __A, __m128i __B) {
1335 return (__m128i)((__v8hu)__A - (__v8hu)__B);
1340 _mm_sub_epi32(__m128i __A, __m128i __B) {
1341 return (__m128i)((__v4su)__A - (__v4su)__B);
1346 _mm_sub_epi64(__m128i __A, __m128i __B) {
1347 return (__m128i)((__v2du)__A - (__v2du)__B);
1352 _mm_subs_epi8(__m128i __A, __m128i __B) {
1353 return (__m128i)vec_subs((__v16qi)__A, (__v16qi)__B);
1358 _mm_subs_epi16(__m128i __A, __m128i __B) {
1359 return (__m128i)vec_subs((__v8hi)__A, (__v8hi)__B);
1364 _mm_subs_epu8(__m128i __A, __m128i __B) {
1365 return (__m128i)vec_subs((__v16qu)__A, (__v16qu)__B);
1370 _mm_subs_epu16(__m128i __A, __m128i __B) {
1371 return (__m128i)vec_subs((__v8hu)__A, (__v8hu)__B);
1376 _mm_madd_epi16(__m128i __A, __m128i __B) {
1379 return (__m128i)vec_vmsumshm((__v8hi)__A, (__v8hi)__B, __zero);
1384 _mm_mulhi_epi16(__m128i __A, __m128i __B) {
1397 __w0 = vec_vmulesh((__v8hi)__A, (__v8hi)__B);
1398 __w1 = vec_vmulosh((__v8hi)__A, (__v8hi)__B);
1404 _mm_mullo_epi16(__m128i __A, __m128i __B) {
1405 return (__m128i)((__v8hi)__A * (__v8hi)__B);
1410 _mm_mul_su32(__m64 __A, __m64 __B) {
1412 unsigned int __b = __B;
1420 _mm_mul_epu32(__m128i __A, __m128i __B) {
1426 __asm__("vmulouw %0,%1,%2" : "=v"(__result) : "v"(__A), "v"(__B) :);
1429 __asm__("vmuleuw %0,%1,%2" : "=v"(__result) : "v"(__A), "v"(__B) :);
1433 return (__m128i)vec_mule((__v4su)__A, (__v4su)__B);
1440 _mm_slli_epi16(__m128i __A, int __B) {
1444 if (__B >= 0 && __B < 16) {
1445 if (__builtin_constant_p(__B))
1446 __lshift = (__v8hu)vec_splat_s16(__B);
1448 __lshift = vec_splats((unsigned short)__B);
1458 _mm_slli_epi32(__m128i __A, int __B) {
1462 if (__B >= 0 && __B < 32) {
1463 if (__builtin_constant_p(__B) && __B < 16)
1464 __lshift = (__v4su)vec_splat_s32(__B);
1466 __lshift = vec_splats((unsigned int)__B);
1477 _mm_slli_epi64(__m128i __A, int __B) {
1481 if (__B >= 0 && __B < 64) {
1482 if (__builtin_constant_p(__B) && __B < 16)
1483 __lshift = (__v2du)vec_splat_s32(__B);
1485 __lshift = (__v2du)vec_splats((unsigned int)__B);
1496 _mm_srai_epi16(__m128i __A, int __B) {
1500 if (__B < 16) {
1501 if (__builtin_constant_p(__B))
1502 __rshift = (__v8hu)vec_splat_s16(__B);
1504 __rshift = vec_splats((unsigned short)__B);
1513 _mm_srai_epi32(__m128i __A, int __B) {
1517 if (__B < 32) {
1518 if (__builtin_constant_p(__B)) {
1519 if (__B < 16)
1520 __rshift = (__v4su)vec_splat_s32(__B);
1522 __rshift = (__v4su)vec_splats((unsigned int)__B);
1524 __rshift = vec_splats((unsigned int)__B);
1601 _mm_srli_epi16(__m128i __A, int __B) {
1605 if (__B < 16) {
1606 if (__builtin_constant_p(__B))
1607 __rshift = (__v8hu)vec_splat_s16(__B);
1609 __rshift = vec_splats((unsigned short)__B);
1619 _mm_srli_epi32(__m128i __A, int __B) {
1623 if (__B < 32) {
1624 if (__builtin_constant_p(__B)) {
1625 if (__B < 16)
1626 __rshift = (__v4su)vec_splat_s32(__B);
1628 __rshift = (__v4su)vec_splats((unsigned int)__B);
1630 __rshift = vec_splats((unsigned int)__B);
1641 _mm_srli_epi64(__m128i __A, int __B) {
1645 if (__B < 64) {
1646 if (__builtin_constant_p(__B)) {
1647 if (__B < 16)
1648 __rshift = (__v2du)vec_splat_s32(__B);
1650 __rshift = (__v2du)vec_splats((unsigned long long)__B);
1652 __rshift = (__v2du)vec_splats((unsigned int)__B);
1663 _mm_sll_epi16(__m128i __A, __m128i __B) {
1670 __lshift = vec_splat((__v8hu)__B, 0);
1672 __lshift = vec_splat((__v8hu)__B, 3);
1683 _mm_sll_epi32(__m128i __A, __m128i __B) {
1689 __lshift = vec_splat((__v4su)__B, 0);
1691 __lshift = vec_splat((__v4su)__B, 1);
1703 _mm_sll_epi64(__m128i __A, __m128i __B) {
1709 __lshift = vec_splat((__v2du)__B, 0);
1720 _mm_sra_epi16(__m128i __A, __m128i __B) {
1726 __rshift = vec_splat((__v8hu)__B, 0);
1728 __rshift = vec_splat((__v8hu)__B, 3);
1738 _mm_sra_epi32(__m128i __A, __m128i __B) {
1744 __rshift = vec_splat((__v4su)__B, 0);
1746 __rshift = vec_splat((__v4su)__B, 1);
1756 _mm_srl_epi16(__m128i __A, __m128i __B) {
1763 __rshift = vec_splat((__v8hu)__B, 0);
1765 __rshift = vec_splat((__v8hu)__B, 3);
1776 _mm_srl_epi32(__m128i __A, __m128i __B) {
1783 __rshift = vec_splat((__v4su)__B, 0);
1785 __rshift = vec_splat((__v4su)__B, 1);
1797 _mm_srl_epi64(__m128i __A, __m128i __B) {
1803 __rshift = vec_splat((__v2du)__B, 0);
1814 _mm_and_pd(__m128d __A, __m128d __B) {
1815 return (vec_and((__v2df)__A, (__v2df)__B));
1820 _mm_andnot_pd(__m128d __A, __m128d __B) {
1821 return (vec_andc((__v2df)__B, (__v2df)__A));
1826 _mm_or_pd(__m128d __A, __m128d __B) {
1827 return (vec_or((__v2df)__A, (__v2df)__B));
1832 _mm_xor_pd(__m128d __A, __m128d __B) {
1833 return (vec_xor((__v2df)__A, (__v2df)__B));
1838 _mm_and_si128(__m128i __A, __m128i __B) {
1839 return (__m128i)vec_and((__v2di)__A, (__v2di)__B);
1844 _mm_andnot_si128(__m128i __A, __m128i __B) {
1845 return (__m128i)vec_andc((__v2di)__B, (__v2di)__A);
1850 _mm_or_si128(__m128i __A, __m128i __B) {
1851 return (__m128i)vec_or((__v2di)__A, (__v2di)__B);
1856 _mm_xor_si128(__m128i __A, __m128i __B) {
1857 return (__m128i)vec_xor((__v2di)__A, (__v2di)__B);
1862 _mm_cmpeq_epi8(__m128i __A, __m128i __B) {
1863 return (__m128i)vec_cmpeq((__v16qi)__A, (__v16qi)__B);
1868 _mm_cmpeq_epi16(__m128i __A, __m128i __B) {
1869 return (__m128i)vec_cmpeq((__v8hi)__A, (__v8hi)__B);
1874 _mm_cmpeq_epi32(__m128i __A, __m128i __B) {
1875 return (__m128i)vec_cmpeq((__v4si)__A, (__v4si)__B);
1880 _mm_cmplt_epi8(__m128i __A, __m128i __B) {
1881 return (__m128i)vec_cmplt((__v16qi)__A, (__v16qi)__B);
1886 _mm_cmplt_epi16(__m128i __A, __m128i __B) {
1887 return (__m128i)vec_cmplt((__v8hi)__A, (__v8hi)__B);
1892 _mm_cmplt_epi32(__m128i __A, __m128i __B) {
1893 return (__m128i)vec_cmplt((__v4si)__A, (__v4si)__B);
1898 _mm_cmpgt_epi8(__m128i __A, __m128i __B) {
1899 return (__m128i)vec_cmpgt((__v16qi)__A, (__v16qi)__B);
1904 _mm_cmpgt_epi16(__m128i __A, __m128i __B) {
1905 return (__m128i)vec_cmpgt((__v8hi)__A, (__v8hi)__B);
1910 _mm_cmpgt_epi32(__m128i __A, __m128i __B) {
1911 return (__m128i)vec_cmpgt((__v4si)__A, (__v4si)__B);
1932 _mm_max_epi16(__m128i __A, __m128i __B) {
1933 return (__m128i)vec_max((__v8hi)__A, (__v8hi)__B);
1938 _mm_max_epu8(__m128i __A, __m128i __B) {
1939 return (__m128i)vec_max((__v16qu)__A, (__v16qu)__B);
1944 _mm_min_epi16(__m128i __A, __m128i __B) {
1945 return (__m128i)vec_min((__v8hi)__A, (__v8hi)__B);
1950 _mm_min_epu8(__m128i __A, __m128i __B) {
1951 return (__m128i)vec_min((__v16qu)__A, (__v16qu)__B);
1984 _mm_mulhi_epu16(__m128i __A, __m128i __B) {
1996 __w0 = vec_vmuleuh((__v8hu)__A, (__v8hu)__B);
1997 __w1 = vec_vmulouh((__v8hu)__A, (__v8hu)__B);
2092 _mm_maskmoveu_si128(__m128i __A, __m128i __B, char *__C) {
2098 __mask = (__v16qu)vec_cmpgt((__v16qu)__B, (__v16qu)__hibit);
2105 _mm_avg_epu8(__m128i __A, __m128i __B) {
2106 return (__m128i)vec_avg((__v16qu)__A, (__v16qu)__B);
2111 _mm_avg_epu16(__m128i __A, __m128i __B) {
2112 return (__m128i)vec_avg((__v8hu)__A, (__v8hu)__B);
2117 _mm_sad_epu8(__m128i __A, __m128i __B) {
2125 __b = (__v16qu)__B;
2155 _mm_stream_si32(int *__A, int __B) {
2158 *__A = __B;
2163 _mm_stream_si64(long long int *__A, long long int __B) {
2166 *__A = __B;
2171 _mm_stream_si128(__m128i *__A, __m128i __B) {
2174 *__A = __B;
2179 _mm_stream_pd(double *__A, __m128d __B) {
2182 *(__m128d *)__A = __B;