54                  __target__("sse2,no-evex512"), __min_vector_width__(128)))
68 ///    A 128-bit vector of [2 x double] containing one of the source operands.
70 ///    A 128-bit vector of [2 x double] containing one of the source operands.
71 /// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
80 /// Adds two 128-bit vectors of [2 x double].
87 ///    A 128-bit vector of [2 x double] containing one of the source operands.
89 ///    A 128-bit vector of [2 x double] containing one of the source operands.
90 /// \returns A 128-bit vector of [2 x double] containing the sums of both
108 ///    A 128-bit vector of [2 x double] containing the minuend.
110 ///    A 128-bit vector of [2 x double] containing the subtrahend.
111 /// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
120 /// Subtracts two 128-bit vectors of [2 x double].
127 ///    A 128-bit vector of [2 x double] containing the minuend.
129 ///    A 128-bit vector of [2 x double] containing the subtrahend.
130 /// \returns A 128-bit vector of [2 x double] containing the differences between
147 ///    A 128-bit vector of [2 x double] containing one of the source operands.
149 ///    A 128-bit vector of [2 x double] containing one of the source operands.
150 /// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
159 /// Multiplies two 128-bit vectors of [2 x double].
166 ///    A 128-bit vector of [2 x double] containing one of the operands.
168 ///    A 128-bit vector of [2 x double] containing one of the operands.
169 /// \returns A 128-bit vector of [2 x double] containing the products of both
187 ///    A 128-bit vector of [2 x double] containing the dividend.
189 ///    A 128-bit vector of [2 x double] containing divisor.
190 /// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
199 /// Performs an element-by-element division of two 128-bit vectors of
207 ///    A 128-bit vector of [2 x double] containing the dividend.
209 ///    A 128-bit vector of [2 x double] containing the divisor.
210 /// \returns A 128-bit vector of [2 x double] containing the quotients of both
227 ///    A 128-bit vector of [2 x double] containing one of the operands. The
231 ///    A 128-bit vector of [2 x double] containing one of the operands. The
233 /// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
234 ///    square root of the lower 64 bits of operand \a __b, and whose upper 64
235 ///    bits are copied from the upper 64 bits of operand \a __a.
242 /// Calculates the square root of the each of two values stored in a
243 ///    128-bit vector of [2 x double].
250 ///    A 128-bit vector of [2 x double].
251 /// \returns A 128-bit vector of [2 x double] containing the square roots of the
262 ///    If either value in a comparison is NaN, returns the value from \a __b.
269 ///    A 128-bit vector of [2 x double] containing one of the operands. The
272 ///    A 128-bit vector of [2 x double] containing one of the operands. The
274 /// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
282 /// Performs element-by-element comparison of the two 128-bit vectors of
283 ///    [2 x double] and returns a vector containing the lesser of each pair of
286 ///    If either value in a comparison is NaN, returns the value from \a __b.
293 ///    A 128-bit vector of [2 x double] containing one of the operands.
295 ///    A 128-bit vector of [2 x double] containing one of the operands.
296 /// \returns A 128-bit vector of [2 x double] containing the minimum values
308 ///    If either value in a comparison is NaN, returns the value from \a __b.
315 ///    A 128-bit vector of [2 x double] containing one of the operands. The
318 ///    A 128-bit vector of [2 x double] containing one of the operands. The
320 /// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
328 /// Performs element-by-element comparison of the two 128-bit vectors of
329 ///    [2 x double] and returns a vector containing the greater of each pair
332 ///    If either value in a comparison is NaN, returns the value from \a __b.
339 ///    A 128-bit vector of [2 x double] containing one of the operands.
341 ///    A 128-bit vector of [2 x double] containing one of the operands.
342 /// \returns A 128-bit vector of [2 x double] containing the maximum values
349 /// Performs a bitwise AND of two 128-bit vectors of [2 x double].
356 ///    A 128-bit vector of [2 x double] containing one of the source operands.
358 ///    A 128-bit vector of [2 x double] containing one of the source operands.
359 /// \returns A 128-bit vector of [2 x double] containing the bitwise AND of the
366 /// Performs a bitwise AND of two 128-bit vectors of [2 x double], using
374 ///    A 128-bit vector of [2 x double] containing the left source operand. The
377 ///    A 128-bit vector of [2 x double] containing the right source operand.
378 /// \returns A 128-bit vector of [2 x double] containing the bitwise AND of the
386 /// Performs a bitwise OR of two 128-bit vectors of [2 x double].
393 ///    A 128-bit vector of [2 x double] containing one of the source operands.
395 ///    A 128-bit vector of [2 x double] containing one of the source operands.
396 /// \returns A 128-bit vector of [2 x double] containing the bitwise OR of the
403 /// Performs a bitwise XOR of two 128-bit vectors of [2 x double].
410 ///    A 128-bit vector of [2 x double] containing one of the source operands.
412 ///    A 128-bit vector of [2 x double] containing one of the source operands.
413 /// \returns A 128-bit vector of [2 x double] containing the bitwise XOR of the
421 ///    128-bit vectors of [2 x double] for equality.
424 ///    If either value in a comparison is NaN, returns false.
431 ///    A 128-bit vector of [2 x double].
433 ///    A 128-bit vector of [2 x double].
434 /// \returns A 128-bit vector containing the comparison results.
441 ///    128-bit vectors of [2 x double] to determine if the values in the first
445 ///    If either value in a comparison is NaN, returns false.
452 ///    A 128-bit vector of [2 x double].
454 ///    A 128-bit vector of [2 x double].
455 /// \returns A 128-bit vector containing the comparison results.
462 ///    128-bit vectors of [2 x double] to determine if the values in the first
466 ///    If either value in a comparison is NaN, returns false.
473 ///    A 128-bit vector of [2 x double].
475 ///    A 128-bit vector of [2 x double].
476 /// \returns A 128-bit vector containing the comparison results.
483 ///    128-bit vectors of [2 x double] to determine if the values in the first
487 ///    If either value in a comparison is NaN, returns false.
494 ///    A 128-bit vector of [2 x double].
496 ///    A 128-bit vector of [2 x double].
497 /// \returns A 128-bit vector containing the comparison results.
504 ///    128-bit vectors of [2 x double] to determine if the values in the first
508 ///    If either value in a comparison is NaN, returns false.
515 ///    A 128-bit vector of [2 x double].
517 ///    A 128-bit vector of [2 x double].
518 /// \returns A 128-bit vector containing the comparison results.
525 ///    128-bit vectors of [2 x double] to determine if the values in the first
528 ///    A pair of double-precision values are ordered with respect to each
529 ///    other if neither value is a NaN. Each comparison returns 0x0 for false,
537 ///    A 128-bit vector of [2 x double].
539 ///    A 128-bit vector of [2 x double].
540 /// \returns A 128-bit vector containing the comparison results.
547 ///    128-bit vectors of [2 x double] to determine if the values in the first
550 ///    A pair of double-precision values are unordered with respect to each
560 ///    A 128-bit vector of [2 x double].
562 ///    A 128-bit vector of [2 x double].
563 /// \returns A 128-bit vector containing the comparison results.
570 ///    128-bit vectors of [2 x double] to determine if the values in the first
574 ///    If either value in a comparison is NaN, returns true.
581 ///    A 128-bit vector of [2 x double].
583 ///    A 128-bit vector of [2 x double].
584 /// \returns A 128-bit vector containing the comparison results.
591 ///    128-bit vectors of [2 x double] to determine if the values in the first
595 ///    If either value in a comparison is NaN, returns true.
602 ///    A 128-bit vector of [2 x double].
604 ///    A 128-bit vector of [2 x double].
605 /// \returns A 128-bit vector containing the comparison results.
612 ///    128-bit vectors of [2 x double] to determine if the values in the first
616 ///    If either value in a comparison is NaN, returns true.
623 ///    A 128-bit vector of [2 x double].
625 ///    A 128-bit vector of [2 x double].
626 /// \returns A 128-bit vector containing the comparison results.
633 ///    128-bit vectors of [2 x double] to determine if the values in the first
637 ///    If either value in a comparison is NaN, returns true.
644 ///    A 128-bit vector of [2 x double].
646 ///    A 128-bit vector of [2 x double].
647 /// \returns A 128-bit vector containing the comparison results.
654 ///    128-bit vectors of [2 x double] to determine if the values in the first
658 ///    If either value in a comparison is NaN, returns true.
665 ///    A 128-bit vector of [2 x double].
667 ///    A 128-bit vector of [2 x double].
668 /// \returns A 128-bit vector containing the comparison results.
675 ///    the two 128-bit floating-point vectors of [2 x double] for equality.
678 ///    If either value in a comparison is NaN, returns false.
685 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
686 ///    compared to the lower double-precision value of \a __b.
688 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
689 ///    compared to the lower double-precision value of \a __a.
690 /// \returns A 128-bit vector. The lower 64 bits contains the comparison
691 ///    results. The upper 64 bits are copied from the upper 64 bits of \a __a.
698 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
703 ///    If either value in a comparison is NaN, returns false.
710 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
711 ///    compared to the lower double-precision value of \a __b.
713 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
714 ///    compared to the lower double-precision value of \a __a.
715 /// \returns A 128-bit vector. The lower 64 bits contains the comparison
716 ///    results. The upper 64 bits are copied from the upper 64 bits of \a __a.
723 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
728 ///    If either value in a comparison is NaN, returns false.
735 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
736 ///    compared to the lower double-precision value of \a __b.
738 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
739 ///    compared to the lower double-precision value of \a __a.
740 /// \returns A 128-bit vector. The lower 64 bits contains the comparison
741 ///    results. The upper 64 bits are copied from the upper 64 bits of \a __a.
748 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
753 ///    If either value in a comparison is NaN, returns false.
760 ///     A 128-bit vector of [2 x double]. The lower double-precision value is
761 ///     compared to the lower double-precision value of \a __b.
763 ///     A 128-bit vector of [2 x double]. The lower double-precision value is
764 ///     compared to the lower double-precision value of \a __a.
765 /// \returns A 128-bit vector. The lower 64 bits contains the comparison
766 ///     results. The upper 64 bits are copied from the upper 64 bits of \a __a.
774 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
779 ///    If either value in a comparison is NaN, returns false.
786 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
787 ///    compared to the lower double-precision value of \a __b.
789 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
790 ///    compared to the lower double-precision value of \a __a.
791 /// \returns A 128-bit vector. The lower 64 bits contains the comparison
792 ///    results. The upper 64 bits are copied from the upper 64 bits of \a __a.
800 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
804 ///    The comparison returns 0x0 for false, 0xFFFFFFFFFFFFFFFF for true. A pair
806 ///    neither value is a NaN.
813 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
814 ///    compared to the lower double-precision value of \a __b.
816 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
817 ///    compared to the lower double-precision value of \a __a.
818 /// \returns A 128-bit vector. The lower 64 bits contains the comparison
819 ///    results. The upper 64 bits are copied from the upper 64 bits of \a __a.
826 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
830 ///    The comparison returns 0x0 for false, 0xFFFFFFFFFFFFFFFF for true. A pair
840 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
841 ///    compared to the lower double-precision value of \a __b.
843 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
844 ///    compared to the lower double-precision value of \a __a.
845 /// \returns A 128-bit vector. The lower 64 bits contains the comparison
846 ///    results. The upper 64 bits are copied from the upper 64 bits of \a __a.
853 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
858 ///    If either value in a comparison is NaN, returns true.
865 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
866 ///    compared to the lower double-precision value of \a __b.
868 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
869 ///    compared to the lower double-precision value of \a __a.
870 /// \returns A 128-bit vector. The lower 64 bits contains the comparison
871 ///    results. The upper 64 bits are copied from the upper 64 bits of \a __a.
878 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
883 ///    If either value in a comparison is NaN, returns true.
890 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
891 ///    compared to the lower double-precision value of \a __b.
893 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
894 ///    compared to the lower double-precision value of \a __a.
895 /// \returns A 128-bit vector. The lower 64 bits contains the comparison
896 ///    results. The upper 64 bits are copied from the upper 64 bits of \a __a.
903 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
908 ///    If either value in a comparison is NaN, returns true.
915 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
916 ///    compared to the lower double-precision value of \a __b.
918 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
919 ///    compared to the lower double-precision value of \a __a.
920 /// \returns  A 128-bit vector. The lower 64 bits contains the comparison
921 ///    results. The upper 64 bits are copied from the upper 64 bits of \a __a.
928 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
933 ///    If either value in a comparison is NaN, returns true.
940 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
941 ///    compared to the lower double-precision value of \a __b.
943 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
944 ///    compared to the lower double-precision value of \a __a.
945 /// \returns A 128-bit vector. The lower 64 bits contains the comparison
946 ///    results. The upper 64 bits are copied from the upper 64 bits of \a __a.
954 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
959 ///    If either value in a comparison is NaN, returns true.
966 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
967 ///    compared to the lower double-precision value of \a __b.
969 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
970 ///    compared to the lower double-precision value of \a __a.
971 /// \returns A 128-bit vector. The lower 64 bits contains the comparison
972 ///    results. The upper 64 bits are copied from the upper 64 bits of \a __a.
980 ///    the two 128-bit floating-point vectors of [2 x double] for equality.
982 ///    The comparison returns 0 for false, 1 for true. If either value in a
990 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
991 ///    compared to the lower double-precision value of \a __b.
993 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
994 ///    compared to the lower double-precision value of \a __a.
1002 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
1006 ///    The comparison returns 0 for false, 1 for true. If either value in a
1014 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1015 ///    compared to the lower double-precision value of \a __b.
1017 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1018 ///    compared to the lower double-precision value of \a __a.
1026 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
1030 ///    The comparison returns 0 for false, 1 for true. If either value in a
1038 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1039 ///    compared to the lower double-precision value of \a __b.
1041 ///     A 128-bit vector of [2 x double]. The lower double-precision value is
1042 ///     compared to the lower double-precision value of \a __a.
1050 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
1054 ///    The comparison returns 0 for false, 1 for true. If either value in a
1062 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1063 ///    compared to the lower double-precision value of \a __b.
1065 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1066 ///    compared to the lower double-precision value of \a __a.
1074 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
1078 ///    The comparison returns 0 for false, 1 for true. If either value in a
1086 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1087 ///    compared to the lower double-precision value of \a __b.
1089 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1090 ///    compared to the lower double-precision value of \a __a.
1098 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
1102 ///    The comparison returns 0 for false, 1 for true. If either value in a
1110 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1111 ///    compared to the lower double-precision value of \a __b.
1113 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1114 ///    compared to the lower double-precision value of \a __a.
1122 ///    the two 128-bit floating-point vectors of [2 x double] for equality.
1124 ///    The comparison returns 0 for false, 1 for true. If either value in a
1132 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1133 ///    compared to the lower double-precision value of \a __b.
1135 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1136 ///    compared to the lower double-precision value of \a __a.
1144 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
1148 ///    The comparison returns 0 for false, 1 for true. If either value in a
1156 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1157 ///    compared to the lower double-precision value of \a __b.
1159 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1160 ///    compared to the lower double-precision value of \a __a.
1168 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
1172 ///    The comparison returns 0 for false, 1 for true. If either value in a
1180 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1181 ///    compared to the lower double-precision value of \a __b.
1183 ///     A 128-bit vector of [2 x double]. The lower double-precision value is
1184 ///     compared to the lower double-precision value of \a __a.
1192 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
1196 ///    The comparison returns 0 for false, 1 for true. If either value in a
1204 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1205 ///    compared to the lower double-precision value of \a __b.
1207 ///     A 128-bit vector of [2 x double]. The lower double-precision value is
1208 ///     compared to the lower double-precision value of \a __a.
1216 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
1220 ///    The comparison returns 0 for false, 1 for true. If either value in a
1228 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1229 ///    compared to the lower double-precision value of \a __b.
1231 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1232 ///    compared to the lower double-precision value of \a __a.
1240 ///    the two 128-bit floating-point vectors of [2 x double] to determine if
1244 ///    The comparison returns 0 for false, 1 for true. If either value in a
1252 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1253 ///    compared to the lower double-precision value of \a __b.
1255 ///    A 128-bit vector of [2 x double]. The lower double-precision value is
1256 ///    compared to the lower double-precision value of \a __a.
1263 /// Converts the two double-precision floating-point elements of a
1264 ///    128-bit vector of [2 x double] into two single-precision floating-point
1265 ///    values, returned in the lower 64 bits of a 128-bit vector of [4 x float].
1273 ///    A 128-bit vector of [2 x double].
1274 /// \returns A 128-bit vector of [4 x float] whose lower 64 bits contain the
1280 /// Converts the lower two single-precision floating-point elements of a
1281 ///    128-bit vector of [4 x float] into two double-precision floating-point
1282 ///    values, returned in a 128-bit vector of [2 x double]. The upper two
1290 ///    A 128-bit vector of [4 x float]. The lower two single-precision
1293 /// \returns A 128-bit vector of [2 x double] containing the converted values.
1299 /// Converts the lower two integer elements of a 128-bit vector of
1300 ///    [4 x i32] into two double-precision floating-point values, returned in a
1301 ///    128-bit vector of [2 x double].
1310 ///    A 128-bit integer vector of [4 x i32]. The lower two integer elements are
1314 /// \returns A 128-bit vector of [2 x double] containing the converted values.
1320 /// Converts the two double-precision floating-point elements of a
1321 ///    128-bit vector of [2 x double] into two signed 32-bit integer values,
1322 ///    returned in the lower 64 bits of a 128-bit vector of [4 x i32]. The upper
1325 ///    If a converted value does not fit in a 32-bit integer, raises a
1334 ///    A 128-bit vector of [2 x double].
1335 /// \returns A 128-bit vector of [4 x i32] whose lower 64 bits contain the
1341 /// Converts the low-order element of a 128-bit vector of [2 x double]
1342 ///    into a 32-bit signed integer value.
1344 ///    If the converted value does not fit in a 32-bit integer, raises a
1353 ///    A 128-bit vector of [2 x double]. The lower 64 bits are used in the
1355 /// \returns A 32-bit signed integer containing the converted value.
1360 /// Converts the lower double-precision floating-point element of a
1361 ///    128-bit vector of [2 x double], in the second parameter, into a
1362 ///    single-precision floating-point value, returned in the lower 32 bits of a
1363 ///    128-bit vector of [4 x float]. The upper 96 bits of the result vector are
1371 ///    A 128-bit vector of [4 x float]. The upper 96 bits of this parameter are
1374 ///    A 128-bit vector of [2 x double]. The lower double-precision
1376 /// \returns A 128-bit vector of [4 x float]. The lower 32 bits contain the
1384 /// Converts a 32-bit signed integer value, in the second parameter, into
1385 ///    a double-precision floating-point value, returned in the lower 64 bits of
1386 ///    a 128-bit vector of [2 x double]. The upper 64 bits of the result vector
1394 ///    A 128-bit vector of [2 x double]. The upper 64 bits of this parameter are
1397 ///    A 32-bit signed integer containing the value to be converted.
1398 /// \returns A 128-bit vector of [2 x double]. The lower 64 bits contain the
1407 /// Converts the lower single-precision floating-point element of a
1408 ///    128-bit vector of [4 x float], in the second parameter, into a
1410 ///    a 128-bit vector of [2 x double]. The upper 64 bits of the result vector
1418 ///    A 128-bit vector of [2 x double]. The upper 64 bits of this parameter are
1421 ///    A 128-bit vector of [4 x float]. The lower single-precision
1423 /// \returns A 128-bit vector of [2 x double]. The lower 64 bits contain the
1432 /// Converts the two double-precision floating-point elements of a
1433 ///    128-bit vector of [2 x double] into two signed truncated (rounded
1435 ///    of a 128-bit vector of [4 x i32].
1437 ///    If a converted value does not fit in a 32-bit integer, raises a
1447 ///    A 128-bit vector of [2 x double].
1448 /// \returns A 128-bit vector of [4 x i32] whose lower 64 bits contain the
1454 /// Converts the low-order element of a [2 x double] vector into a 32-bit
1457 ///    If the converted value does not fit in a 32-bit integer, raises a
1467 ///    A 128-bit vector of [2 x double]. The lower 64 bits are used in the
1469 /// \returns A 32-bit signed integer containing the converted value.
1474 /// Converts the two double-precision floating-point elements of a
1475 ///    128-bit vector of [2 x double] into two signed 32-bit integer values,
1476 ///    returned in a 64-bit vector of [2 x i32].
1478 ///    If a converted value does not fit in a 32-bit integer, raises a
1487 ///    A 128-bit vector of [2 x double].
1488 /// \returns A 64-bit vector of [2 x i32] containing the converted values.
1493 /// Converts the two double-precision floating-point elements of a
1494 ///    128-bit vector of [2 x double] into two signed truncated (rounded toward
1495 ///    zero) 32-bit integer values, returned in a 64-bit vector of [2 x i32].
1497 ///    If a converted value does not fit in a 32-bit integer, raises a
1506 ///    A 128-bit vector of [2 x double].
1507 /// \returns A 64-bit vector of [2 x i32] containing the converted values.
1512 /// Converts the two signed 32-bit integer elements of a 64-bit vector of
1513 ///    [2 x i32] into two double-precision floating-point values, returned in a
1514 ///    128-bit vector of [2 x double].
1521 ///    A 64-bit vector of [2 x i32].
1522 /// \returns A 128-bit vector of [2 x double] containing the converted values.
1527 /// Returns the low-order element of a 128-bit vector of [2 x double] as
1528 ///    a double-precision floating-point value.
1535 ///    A 128-bit vector of [2 x double]. The lower 64 bits are returned.
1536 /// \returns A double-precision floating-point value copied from the lower 64
1537 ///    bits of \a __a.
1542 /// Loads a 128-bit floating-point vector of [2 x double] from an aligned
1550 ///    A pointer to a 128-bit memory location. The address of the memory
1552 /// \returns A 128-bit vector of [2 x double] containing the loaded values.
1557 /// Loads a double-precision floating-point value from a specified memory
1558 ///    location and duplicates it to both vector elements of a 128-bit vector of
1566 ///    A pointer to a memory location containing a double-precision value.
1567 /// \returns A 128-bit vector of [2 x double] containing the loaded and
1580 ///    memory location into a 128-bit vector of [2 x double].
1586 /// with the \c VMOVAPD, resulting in only a \c VPERMILPD instruction.
1589 ///    A 16-byte aligned pointer to an array of double-precision values to be
1591 /// \returns A 128-bit vector of [2 x double] containing the reversed loaded
1598 /// Loads a 128-bit floating-point vector of [2 x double] from an
1606 ///    A pointer to a 128-bit memory location. The address of the memory
1608 /// \returns A 128-bit vector of [2 x double] containing the loaded values.
1616 /// Loads a 64-bit integer value to the low element of a 128-bit integer
1624 ///    A pointer to a 64-bit memory location. The address of the memory
1626 /// \returns A 128-bit vector of [2 x i64] containing the loaded value.
1635 /// Loads a 32-bit integer value to the low element of a 128-bit integer
1643 ///    A pointer to a 32-bit memory location. The address of the memory
1645 /// \returns A 128-bit vector of [4 x i32] containing the loaded value.
1654 /// Loads a 16-bit integer value to the low element of a 128-bit integer
1659 /// This intrinsic does not correspond to a specific instruction.
1662 ///    A pointer to a 16-bit memory location. The address of the memory
1664 /// \returns A 128-bit vector of [8 x i16] containing the loaded value.
1673 /// Loads a 64-bit double-precision value to the low element of a
1674 ///    128-bit integer vector and clears the upper element.
1681 ///    A pointer to a memory location containing a double-precision value.
1683 /// \returns A 128-bit vector of [2 x double] containing the loaded value.
1692 /// Loads a double-precision value into the high-order bits of a 128-bit
1701 ///    A 128-bit vector of [2 x double]. \n
1704 ///    A pointer to a 64-bit memory location containing a double-precision
1708 /// \returns A 128-bit vector of [2 x double] containing the moved values.
1718 /// Loads a double-precision value into the low-order bits of a 128-bit
1727 ///    A 128-bit vector of [2 x double]. \n
1730 ///    A pointer to a 64-bit memory location containing a double-precision
1734 /// \returns A 128-bit vector of [2 x double] containing the moved values.
1744 /// Constructs a 128-bit floating-point vector of [2 x double] with
1753 /// \returns A 128-bit floating-point vector of [2 x double] with unspecified
1759 /// Constructs a 128-bit floating-point vector of [2 x double]. The lower
1768 ///    A double-precision floating-point value used to initialize the lower 64
1770 /// \returns An initialized 128-bit floating-point vector of [2 x double]. The
1777 /// Constructs a 128-bit floating-point vector of [2 x double], with each
1786 ///    A double-precision floating-point value used to initialize each vector
1788 /// \returns An initialized 128-bit floating-point vector of [2 x double].
1793 /// Constructs a 128-bit floating-point vector of [2 x double], with each
1802 ///    A double-precision floating-point value used to initialize each vector
1804 /// \returns An initialized 128-bit floating-point vector of [2 x double].
1809 /// Constructs a 128-bit floating-point vector of [2 x double]
1817 ///    A double-precision floating-point value used to initialize the upper 64
1820 ///    A double-precision floating-point value used to initialize the lower 64
1822 /// \returns An initialized 128-bit floating-point vector of [2 x double].
1828 /// Constructs a 128-bit floating-point vector of [2 x double],
1837 ///    A double-precision floating-point value used to initialize the lower 64
1840 ///    A double-precision floating-point value used to initialize the upper 64
1842 /// \returns An initialized 128-bit floating-point vector of [2 x double].
1848 /// Constructs a 128-bit floating-point vector of [2 x double]
1855 /// \returns An initialized 128-bit floating-point vector of [2 x double] with
1861 /// Constructs a 128-bit floating-point vector of [2 x double]. The lower
1870 ///    A 128-bit vector of [2 x double]. The upper 64 bits are written to the
1873 ///    A 128-bit vector of [2 x double]. The lower 64 bits are written to the
1875 /// \returns A 128-bit vector of [2 x double] containing the moved values.
1882 /// Stores the lower 64 bits of a 128-bit vector of [2 x double] to a
1890 ///    A pointer to a 64-bit memory location.
1892 ///    A 128-bit vector of [2 x double] containing the value to be stored.
1901 /// Moves packed double-precision values from a 128-bit vector of
1902 ///    [2 x double] to a memory location.
1909 ///    A pointer to an aligned memory location that can store two
1912 ///    A packed 128-bit vector of [2 x double] containing the values to be
1919 /// Moves the lower 64 bits of a 128-bit vector of [2 x double] twice to
1920 ///    the upper and lower 64 bits of a memory location.
1928 ///    A pointer to a memory location that can store two double-precision
1931 ///    A 128-bit vector of [2 x double] whose lower 64 bits are copied to each
1932 ///    of the values in \a __dp.
1939 /// Moves the lower 64 bits of a 128-bit vector of [2 x double] twice to
1940 ///    the upper and lower 64 bits of a memory location.
1948 ///    A pointer to a memory location that can store two double-precision
1951 ///    A 128-bit vector of [2 x double] whose lower 64 bits are copied to each
1952 ///    of the values in \a __dp.
1958 /// Stores a 128-bit vector of [2 x double] into an unaligned memory
1966 ///    A pointer to a 128-bit memory location. The address of the memory
1969 ///    A 128-bit vector of [2 x double] containing the values to be stored.
1978 /// Stores two double-precision values, in reverse order, from a 128-bit
1979 ///    vector of [2 x double] to a 16-byte aligned memory location.
1983 /// This intrinsic corresponds to a shuffling instruction followed by a
1987 ///    A pointer to a 16-byte aligned memory location that can store two
1990 ///    A 128-bit vector of [2 x double] containing the values to be reversed and
1998 /// Stores the upper 64 bits of a 128-bit vector of [2 x double] to a
2006 ///    A pointer to a 64-bit memory location.
2008 ///    A 128-bit vector of [2 x double] containing the value to be stored.
2017 /// Stores the lower 64 bits of a 128-bit vector of [2 x double] to a
2025 ///    A pointer to a 64-bit memory location.
2027 ///    A 128-bit vector of [2 x double] containing the value to be stored.
2036 /// Adds the corresponding elements of two 128-bit vectors of [16 x i8],
2037 ///    saving the lower 8 bits of each sum in the corresponding element of a
2038 ///    128-bit result vector of [16 x i8].
2047 ///    A 128-bit vector of [16 x i8].
2049 ///    A 128-bit vector of [16 x i8].
2050 /// \returns A 128-bit vector of [16 x i8] containing the sums of both
2057 /// Adds the corresponding elements of two 128-bit vectors of [8 x i16],
2058 ///    saving the lower 16 bits of each sum in the corresponding element of a
2059 ///    128-bit result vector of [8 x i16].
2068 ///    A 128-bit vector of [8 x i16].
2070 ///    A 128-bit vector of [8 x i16].
2071 /// \returns A 128-bit vector of [8 x i16] containing the sums of both
2078 /// Adds the corresponding elements of two 128-bit vectors of [4 x i32],
2079 ///    saving the lower 32 bits of each sum in the corresponding element of a
2080 ///    128-bit result vector of [4 x i32].
2089 ///    A 128-bit vector of [4 x i32].
2091 ///    A 128-bit vector of [4 x i32].
2092 /// \returns A 128-bit vector of [4 x i32] containing the sums of both
2107 ///    A 64-bit integer.
2109 ///    A 64-bit integer.
2110 /// \returns A 64-bit integer containing the sum of both parameters.
2116 /// Adds the corresponding elements of two 128-bit vectors of [2 x i64],
2117 ///    saving the lower 64 bits of each sum in the corresponding element of a
2118 ///    128-bit result vector of [2 x i64].
2127 ///    A 128-bit vector of [2 x i64].
2129 ///    A 128-bit vector of [2 x i64].
2130 /// \returns A 128-bit vector of [2 x i64] containing the sums of both
2137 /// Adds, with saturation, the corresponding elements of two 128-bit
2139 ///    of a 128-bit result vector of [16 x i8].
2149 ///    A 128-bit signed [16 x i8] vector.
2151 ///    A 128-bit signed [16 x i8] vector.
2152 /// \returns A 128-bit signed [16 x i8] vector containing the saturated sums of
2159 /// Adds, with saturation, the corresponding elements of two 128-bit
2161 ///    of a 128-bit result vector of [8 x i16].
2171 ///    A 128-bit signed [8 x i16] vector.
2173 ///    A 128-bit signed [8 x i16] vector.
2174 /// \returns A 128-bit signed [8 x i16] vector containing the saturated sums of
2181 /// Adds, with saturation, the corresponding elements of two 128-bit
2183 ///    of a 128-bit result vector of [16 x i8].
2193 ///    A 128-bit unsigned [16 x i8] vector.
2195 ///    A 128-bit unsigned [16 x i8] vector.
2196 /// \returns A 128-bit unsigned [16 x i8] vector containing the saturated sums
2203 /// Adds, with saturation, the corresponding elements of two 128-bit
2205 ///    of a 128-bit result vector of [8 x i16].
2215 ///    A 128-bit unsigned [8 x i16] vector.
2217 ///    A 128-bit unsigned [8 x i16] vector.
2218 /// \returns A 128-bit unsigned [8 x i16] vector containing the saturated sums
2226 ///    128-bit unsigned [16 x i8] vectors, saving each result in the
2227 ///    corresponding element of a 128-bit result vector of [16 x i8].
2234 ///    A 128-bit unsigned [16 x i8] vector.
2236 ///    A 128-bit unsigned [16 x i8] vector.
2237 /// \returns A 128-bit unsigned [16 x i8] vector containing the rounded
2245 ///    128-bit unsigned [8 x i16] vectors, saving each result in the
2246 ///    corresponding element of a 128-bit result vector of [8 x i16].
2253 ///    A 128-bit unsigned [8 x i16] vector.
2255 ///    A 128-bit unsigned [8 x i16] vector.
2256 /// \returns A 128-bit unsigned [8 x i16] vector containing the rounded
2263 /// Multiplies the corresponding elements of two 128-bit signed [8 x i16]
2265 ///    adds the consecutive pairs of 32-bit products to form a 128-bit signed
2268 ///    For example, bits [15:0] of both parameters are multiplied producing a
2270 ///    a 32-bit product, and the sum of those two products becomes bits [31:0]
2278 ///    A 128-bit signed [8 x i16] vector.
2280 ///    A 128-bit signed [8 x i16] vector.
2281 /// \returns A 128-bit signed [4 x i32] vector containing the sums of products
2288 /// Compares corresponding elements of two 128-bit signed [8 x i16]
2290 ///    corresponding element of a 128-bit result vector of [8 x i16].
2297 ///    A 128-bit signed [8 x i16] vector.
2299 ///    A 128-bit signed [8 x i16] vector.
2300 /// \returns A 128-bit signed [8 x i16] vector containing the greater value of
2307 /// Compares corresponding elements of two 128-bit unsigned [16 x i8]
2309 ///    corresponding element of a 128-bit result vector of [16 x i8].
2316 ///    A 128-bit unsigned [16 x i8] vector.
2318 ///    A 128-bit unsigned [16 x i8] vector.
2319 /// \returns A 128-bit unsigned [16 x i8] vector containing the greater value of
2326 /// Compares corresponding elements of two 128-bit signed [8 x i16]
2328 ///    corresponding element of a 128-bit result vector of [8 x i16].
2335 ///    A 128-bit signed [8 x i16] vector.
2337 ///    A 128-bit signed [8 x i16] vector.
2338 /// \returns A 128-bit signed [8 x i16] vector containing the smaller value of
2345 /// Compares corresponding elements of two 128-bit unsigned [16 x i8]
2347 ///    corresponding element of a 128-bit result vector of [16 x i8].
2354 ///    A 128-bit unsigned [16 x i8] vector.
2356 ///    A 128-bit unsigned [16 x i8] vector.
2357 /// \returns A 128-bit unsigned [16 x i8] vector containing the smaller value of
2366 ///    corresponding element of a 128-bit signed [8 x i16] result vector.
2373 ///    A 128-bit signed [8 x i16] vector.
2375 ///    A 128-bit signed [8 x i16] vector.
2376 /// \returns A 128-bit signed [8 x i16] vector containing the upper 16 bits of
2385 ///    corresponding element of a 128-bit unsigned [8 x i16] result vector.
2392 ///    A 128-bit unsigned [8 x i16] vector.
2394 ///    A 128-bit unsigned [8 x i16] vector.
2395 /// \returns A 128-bit unsigned [8 x i16] vector containing the upper 16 bits
2404 ///    corresponding element of a 128-bit signed [8 x i16] result vector.
2411 ///    A 128-bit signed [8 x i16] vector.
2413 ///    A 128-bit signed [8 x i16] vector.
2414 /// \returns A 128-bit signed [8 x i16] vector containing the lower 16 bits of
2430 ///    A 64-bit integer containing one of the source operands.
2432 ///    A 64-bit integer containing one of the source operands.
2433 /// \returns A 64-bit integer vector containing the product of both operands.
2441 ///    the 64-bit products in the corresponding elements of a [2 x i64] vector.
2448 ///    A [2 x i64] vector containing one of the source operands.
2450 ///    A [2 x i64] vector containing one of the source operands.
2451 /// \returns A [2 x i64] vector containing the product of both operands.
2458 ///    values in two 128-bit vectors. Sums the first 8 absolute differences, and
2460 ///    unsigned 16-bit integer sums into the upper and lower elements of a
2468 ///    A 128-bit integer vector containing one of the source operands.
2470 ///    A 128-bit integer vector containing one of the source operands.
2471 /// \returns A [2 x i64] vector containing the sums of the sets of absolute
2485 ///    A 128-bit integer vector containing the minuends.
2487 ///    A 128-bit integer vector containing the subtrahends.
2488 /// \returns A 128-bit integer vector containing the differences of the values
2502 ///    A 128-bit integer vector containing the minuends.
2504 ///    A 128-bit integer vector containing the subtrahends.
2505 /// \returns A 128-bit integer vector containing the differences of the values
2519 ///    A 128-bit integer vector containing the minuends.
2521 ///    A 128-bit integer vector containing the subtrahends.
2522 /// \returns A 128-bit integer vector containing the differences of the values
2537 ///    A 64-bit integer vector containing the minuend.
2539 ///    A 64-bit integer vector containing the subtrahend.
2540 /// \returns A 64-bit integer vector containing the difference of the values in
2554 ///    A 128-bit integer vector containing the minuends.
2556 ///    A 128-bit integer vector containing the subtrahends.
2557 /// \returns A 128-bit integer vector containing the differences of the values
2576 ///    A 128-bit integer vector containing the minuends.
2578 ///    A 128-bit integer vector containing the subtrahends.
2579 /// \returns A 128-bit integer vector containing the differences of the values
2598 ///    A 128-bit integer vector containing the minuends.
2600 ///    A 128-bit integer vector containing the subtrahends.
2601 /// \returns A 128-bit integer vector containing the differences of the values
2619 ///    A 128-bit integer vector containing the minuends.
2621 ///    A 128-bit integer vector containing the subtrahends.
2622 /// \returns A 128-bit integer vector containing the unsigned integer
2640 ///    A 128-bit integer vector containing the minuends.
2642 ///    A 128-bit integer vector containing the subtrahends.
2643 /// \returns A 128-bit integer vector containing the unsigned integer
2650 /// Performs a bitwise AND of two 128-bit integer vectors.
2657 ///    A 128-bit integer vector containing one of the source operands.
2659 ///    A 128-bit integer vector containing one of the source operands.
2660 /// \returns A 128-bit integer vector containing the bitwise AND of the values
2667 /// Performs a bitwise AND of two 128-bit integer vectors, using the
2675 ///    A 128-bit vector containing the left source operand. The one's complement
2678 ///    A 128-bit vector containing the right source operand.
2679 /// \returns A 128-bit integer vector containing the bitwise AND of the one's
2685 /// Performs a bitwise OR of two 128-bit integer vectors.
2692 ///    A 128-bit integer vector containing one of the source operands.
2694 ///    A 128-bit integer vector containing one of the source operands.
2695 /// \returns A 128-bit integer vector containing the bitwise OR of the values
2702 /// Performs a bitwise exclusive OR of two 128-bit integer vectors.
2709 ///    A 128-bit integer vector containing one of the source operands.
2711 ///    A 128-bit integer vector containing one of the source operands.
2712 /// \returns A 128-bit integer vector containing the bitwise exclusive OR of the
2719 /// Left-shifts the 128-bit integer vector operand by the specified
2725 /// __m128i _mm_slli_si128(__m128i a, const int imm);
2730 /// \param a
2731 ///    A 128-bit integer vector containing the source operand.
2734 ///    \a a.
2735 /// \returns A 128-bit integer vector containing the left-shifted value.
2736 #define _mm_slli_si128(a, imm)                                                 \  argument
2737   ((__m128i)__builtin_ia32_pslldqi128_byteshift((__v2di)(__m128i)(a),          \
2740 #define _mm_bslli_si128(a, imm)                                                \  argument
2741   ((__m128i)__builtin_ia32_pslldqi128_byteshift((__v2di)(__m128i)(a),          \
2744 /// Left-shifts each 16-bit value in the 128-bit integer vector operand
2752 ///    A 128-bit integer vector containing the source operand.
2755 ///    in operand \a __a.
2756 /// \returns A 128-bit integer vector containing the left-shifted values.
2762 /// Left-shifts each 16-bit value in the 128-bit integer vector operand
2770 ///    A 128-bit integer vector containing the source operand.
2772 ///    A 128-bit integer vector in which bits [63:0] specify the number of bits
2773 ///    to left-shift each value in operand \a __a.
2774 /// \returns A 128-bit integer vector containing the left-shifted values.
2780 /// Left-shifts each 32-bit value in the 128-bit integer vector operand
2788 ///    A 128-bit integer vector containing the source operand.
2791 ///    in operand \a __a.
2792 /// \returns A 128-bit integer vector containing the left-shifted values.
2798 /// Left-shifts each 32-bit value in the 128-bit integer vector operand
2806 ///    A 128-bit integer vector containing the source operand.
2808 ///    A 128-bit integer vector in which bits [63:0] specify the number of bits
2809 ///    to left-shift each value in operand \a __a.
2810 /// \returns A 128-bit integer vector containing the left-shifted values.
2816 /// Left-shifts each 64-bit value in the 128-bit integer vector operand
2824 ///    A 128-bit integer vector containing the source operand.
2827 ///    in operand \a __a.
2828 /// \returns A 128-bit integer vector containing the left-shifted values.
2834 /// Left-shifts each 64-bit value in the 128-bit integer vector operand
2842 ///    A 128-bit integer vector containing the source operand.
2844 ///    A 128-bit integer vector in which bits [63:0] specify the number of bits
2845 ///    to left-shift each value in operand \a __a.
2846 /// \returns A 128-bit integer vector containing the left-shifted values.
2852 /// Right-shifts each 16-bit value in the 128-bit integer vector operand
2861 ///    A 128-bit integer vector containing the source operand.
2864 ///    in operand \a __a.
2865 /// \returns A 128-bit integer vector containing the right-shifted values.
2871 /// Right-shifts each 16-bit value in the 128-bit integer vector operand
2880 ///    A 128-bit integer vector containing the source operand.
2882 ///    A 128-bit integer vector in which bits [63:0] specify the number of bits
2883 ///    to right-shift each value in operand \a __a.
2884 /// \returns A 128-bit integer vector containing the right-shifted values.
2890 /// Right-shifts each 32-bit value in the 128-bit integer vector operand
2899 ///    A 128-bit integer vector containing the source operand.
2902 ///    in operand \a __a.
2903 /// \returns A 128-bit integer vector containing the right-shifted values.
2909 /// Right-shifts each 32-bit value in the 128-bit integer vector operand
2918 ///    A 128-bit integer vector containing the source operand.
2920 ///    A 128-bit integer vector in which bits [63:0] specify the number of bits
2921 ///    to right-shift each value in operand \a __a.
2922 /// \returns A 128-bit integer vector containing the right-shifted values.
2928 /// Right-shifts the 128-bit integer vector operand by the specified
2934 /// __m128i _mm_srli_si128(__m128i a, const int imm);
2939 /// \param a
2940 ///    A 128-bit integer vector containing the source operand.
2943 ///    \a a.
2944 /// \returns A 128-bit integer vector containing the right-shifted value.
2945 #define _mm_srli_si128(a, imm)                                                 \  argument
2946   ((__m128i)__builtin_ia32_psrldqi128_byteshift((__v2di)(__m128i)(a),          \
2949 #define _mm_bsrli_si128(a, imm)                                                \  argument
2950   ((__m128i)__builtin_ia32_psrldqi128_byteshift((__v2di)(__m128i)(a),          \
2953 /// Right-shifts each of 16-bit values in the 128-bit integer vector
2961 ///    A 128-bit integer vector containing the source operand.
2964 ///    in operand \a __a.
2965 /// \returns A 128-bit integer vector containing the right-shifted values.
2971 /// Right-shifts each of 16-bit values in the 128-bit integer vector
2979 ///    A 128-bit integer vector containing the source operand.
2981 ///    A 128-bit integer vector in which bits [63:0] specify the number of bits
2982 ///    to right-shift each value in operand \a __a.
2983 /// \returns A 128-bit integer vector containing the right-shifted values.
2989 /// Right-shifts each of 32-bit values in the 128-bit integer vector
2997 ///    A 128-bit integer vector containing the source operand.
3000 ///    in operand \a __a.
3001 /// \returns A 128-bit integer vector containing the right-shifted values.
3007 /// Right-shifts each of 32-bit values in the 128-bit integer vector
3015 ///    A 128-bit integer vector containing the source operand.
3017 ///    A 128-bit integer vector in which bits [63:0] specify the number of bits
3018 ///    to right-shift each value in operand \a __a.
3019 /// \returns A 128-bit integer vector containing the right-shifted values.
3025 /// Right-shifts each of 64-bit values in the 128-bit integer vector
3033 ///    A 128-bit integer vector containing the source operand.
3036 ///    in operand \a __a.
3037 /// \returns A 128-bit integer vector containing the right-shifted values.
3043 /// Right-shifts each of 64-bit values in the 128-bit integer vector
3051 ///    A 128-bit integer vector containing the source operand.
3053 ///    A 128-bit integer vector in which bits [63:0] specify the number of bits
3054 ///    to right-shift each value in operand \a __a.
3055 /// \returns A 128-bit integer vector containing the right-shifted values.
3061 /// Compares each of the corresponding 8-bit values of the 128-bit
3071 ///    A 128-bit integer vector.
3073 ///    A 128-bit integer vector.
3074 /// \returns A 128-bit integer vector containing the comparison results.
3080 /// Compares each of the corresponding 16-bit values of the 128-bit
3090 ///    A 128-bit integer vector.
3092 ///    A 128-bit integer vector.
3093 /// \returns A 128-bit integer vector containing the comparison results.
3099 /// Compares each of the corresponding 32-bit values of the 128-bit
3109 ///    A 128-bit integer vector.
3111 ///    A 128-bit integer vector.
3112 /// \returns A 128-bit integer vector containing the comparison results.
3118 /// Compares each of the corresponding signed 8-bit values of the 128-bit
3129 ///    A 128-bit integer vector.
3131 ///    A 128-bit integer vector.
3132 /// \returns A 128-bit integer vector containing the comparison results.
3135   /* This function always performs a signed comparison, but __v16qi is a char  in _mm_cmpgt_epi8()
3141 ///    128-bit integer vectors to determine if the values in the first operand
3151 ///    A 128-bit integer vector.
3153 ///    A 128-bit integer vector.
3154 /// \returns A 128-bit integer vector containing the comparison results.
3161 ///    128-bit integer vectors to determine if the values in the first operand
3171 ///    A 128-bit integer vector.
3173 ///    A 128-bit integer vector.
3174 /// \returns A 128-bit integer vector containing the comparison results.
3180 /// Compares each of the corresponding signed 8-bit values of the 128-bit
3191 ///    A 128-bit integer vector.
3193 ///    A 128-bit integer vector.
3194 /// \returns A 128-bit integer vector containing the comparison results.
3201 ///    128-bit integer vectors to determine if the values in the first operand
3211 ///    A 128-bit integer vector.
3213 ///    A 128-bit integer vector.
3214 /// \returns A 128-bit integer vector containing the comparison results.
3221 ///    128-bit integer vectors to determine if the values in the first operand
3231 ///    A 128-bit integer vector.
3233 ///    A 128-bit integer vector.
3234 /// \returns A 128-bit integer vector containing the comparison results.
3241 /// Converts a 64-bit signed integer value from the second operand into a
3242 ///    double-precision value and returns it in the lower element of a [2 x
3251 ///    A 128-bit vector of [2 x double]. The upper 64 bits of this operand are
3254 ///    A 64-bit signed integer operand containing the value to be converted.
3255 /// \returns A 128-bit vector of [2 x double] whose lower 64 bits contain the
3264 /// Converts the first (lower) element of a vector of [2 x double] into a
3267 ///    If the converted value does not fit in a 64-bit integer, raises a
3276 ///    A 128-bit vector of [2 x double]. The lower 64 bits are used in the
3278 /// \returns A 64-bit signed integer containing the converted value.
3283 /// Converts the first (lower) element of a vector of [2 x double] into a
3286 ///    If a converted value does not fit in a 64-bit integer, raises a
3296 ///    A 128-bit vector of [2 x double]. The lower 64 bits are used in the
3298 /// \returns A 64-bit signed integer containing the converted value.
3304 /// Converts a vector of [4 x i32] into a vector of [4 x float].
3311 ///    A 128-bit integer vector.
3312 /// \returns A 128-bit vector of [4 x float] containing the converted values.
3317 /// Converts a vector of [4 x float] into a vector of [4 x i32].
3319 ///    If a converted value does not fit in a 32-bit integer, raises a
3328 ///    A 128-bit vector of [4 x float].
3329 /// \returns A 128-bit integer vector of [4 x i32] containing the converted
3335 /// Converts a vector of [4 x float] into four signed truncated (rounded toward
3336 ///    zero) 32-bit integers, returned in a vector of [4 x i32].
3338 ///    If a converted value does not fit in a 32-bit integer, raises a
3348 ///    A 128-bit vector of [4 x float].
3349 /// \returns A 128-bit vector of [4 x i32] containing the converted values.
3354 /// Returns a vector of [4 x i32] where the lowest element is the input
3362 ///    A 32-bit signed integer operand.
3363 /// \returns A 128-bit vector of [4 x i32].
3368 /// Returns a vector of [2 x i64] where the lower element is the input
3377 ///    A 64-bit signed integer operand containing the value to be converted.
3378 /// \returns A 128-bit vector of [2 x i64] containing the converted value.
3383 /// Moves the least significant 32 bits of a vector of [4 x i32] to a
3391 ///    A vector of [4 x i32]. The least significant 32 bits are moved to the
3393 /// \returns A 32-bit signed integer containing the moved value.
3399 /// Moves the least significant 64 bits of a vector of [2 x i64] to a
3407 ///    A vector of [2 x i64]. The least significant 64 bits are moved to the
3409 /// \returns A 64-bit signed integer containing the moved value.
3414 /// Moves packed integer values from an aligned 128-bit memory location
3415 ///    to elements in a 128-bit integer vector.
3422 ///    An aligned pointer to a memory location containing integer values.
3423 /// \returns A 128-bit integer vector containing the moved values.
3429 /// Moves packed integer values from an unaligned 128-bit memory location
3430 ///    to elements in a 128-bit integer vector.
3437 ///    A pointer to a memory location containing integer values.
3438 /// \returns A 128-bit integer vector containing the moved values.
3447 /// Returns a vector of [2 x i64] where the lower element is taken from
3455 ///    A 128-bit vector of [2 x i64]. Bits [63:0] are written to bits [63:0] of
3457 /// \returns A 128-bit vector of [2 x i64]. The lower order bits contain the
3468 /// Generates a 128-bit vector of [4 x i32] with unspecified content.
3476 /// \returns A 128-bit vector of [4 x i32] with unspecified content.
3481 /// Initializes both 64-bit values in a 128-bit vector of [2 x i64] with
3486 /// This intrinsic is a utility function and does not correspond to a specific
3490 ///    A 64-bit integer value used to initialize the upper 64 bits of the
3493 ///    A 64-bit integer value used to initialize the lower 64 bits of the
3495 /// \returns An initialized 128-bit vector of [2 x i64] containing the values
3502 /// Initializes both 64-bit values in a 128-bit vector of [2 x i64] with
3507 /// This intrinsic is a utility function and does not correspond to a specific
3511 ///    A 64-bit integer value used to initialize the upper 64 bits of the
3514 ///    A 64-bit integer value used to initialize the lower 64 bits of the
3516 /// \returns An initialized 128-bit vector of [2 x i64] containing the values
3523 /// Initializes the 32-bit values in a 128-bit vector of [4 x i32] with
3528 /// This intrinsic is a utility function and does not correspond to a specific
3532 ///    A 32-bit integer value used to initialize bits [127:96] of the
3535 ///    A 32-bit integer value used to initialize bits [95:64] of the destination
3538 ///    A 32-bit integer value used to initialize bits [63:32] of the destination
3541 ///    A 32-bit integer value used to initialize bits [31:0] of the destination
3543 /// \returns An initialized 128-bit vector of [4 x i32] containing the values
3550 /// Initializes the 16-bit values in a 128-bit vector of [8 x i16] with
3555 /// This intrinsic is a utility function and does not correspond to a specific
3559 ///    A 16-bit integer value used to initialize bits [127:112] of the
3562 ///    A 16-bit integer value used to initialize bits [111:96] of the
3565 ///    A 16-bit integer value used to initialize bits [95:80] of the destination
3568 ///    A 16-bit integer value used to initialize bits [79:64] of the destination
3571 ///    A 16-bit integer value used to initialize bits [63:48] of the destination
3574 ///    A 16-bit integer value used to initialize bits [47:32] of the destination
3577 ///    A 16-bit integer value used to initialize bits [31:16] of the destination
3580 ///    A 16-bit integer value used to initialize bits [15:0] of the destination
3582 /// \returns An initialized 128-bit vector of [8 x i16] containing the values
3591 /// Initializes the 8-bit values in a 128-bit vector of [16 x i8] with
3596 /// This intrinsic is a utility function and does not correspond to a specific
3631 /// \returns An initialized 128-bit vector of [16 x i8] containing the values
3642 /// Initializes both values in a 128-bit integer vector with the
3647 /// This intrinsic is a utility function and does not correspond to a specific
3653 /// \returns An initialized 128-bit integer vector of [2 x i64] with both
3659 /// Initializes both values in a 128-bit vector of [2 x i64] with the
3664 /// This intrinsic is a utility function and does not correspond to a specific
3668 ///    A 64-bit value used to initialize the elements of the destination integer
3670 /// \returns An initialized 128-bit vector of [2 x i64] with all elements
3676 /// Initializes all values in a 128-bit vector of [4 x i32] with the
3681 /// This intrinsic is a utility function and does not correspond to a specific
3685 ///    A 32-bit value used to initialize the elements of the destination integer
3687 /// \returns An initialized 128-bit vector of [4 x i32] with all elements
3693 /// Initializes all values in a 128-bit vector of [8 x i16] with the
3698 /// This intrinsic is a utility function and does not correspond to a specific
3702 ///    A 16-bit value used to initialize the elements of the destination integer
3704 /// \returns An initialized 128-bit vector of [8 x i16] with all elements
3710 /// Initializes all values in a 128-bit vector of [16 x i8] with the
3715 /// This intrinsic is a utility function and does not correspond to a specific
3721 /// \returns An initialized 128-bit vector of [16 x i8] with all elements
3728 /// Constructs a 128-bit integer vector, initialized in reverse order
3733 /// This intrinsic does not correspond to a specific instruction.
3736 ///    A 64-bit integral value used to initialize the lower 64 bits of the
3739 ///    A 64-bit integral value used to initialize the upper 64 bits of the
3741 /// \returns An initialized 128-bit integer vector.
3747 /// Constructs a 128-bit integer vector, initialized in reverse order
3752 /// This intrinsic is a utility function and does not correspond to a specific
3756 ///    A 32-bit integral value used to initialize bits [31:0] of the result.
3758 ///    A 32-bit integral value used to initialize bits [63:32] of the result.
3760 ///    A 32-bit integral value used to initialize bits [95:64] of the result.
3762 ///    A 32-bit integral value used to initialize bits [127:96] of the result.
3763 /// \returns An initialized 128-bit integer vector.
3770 /// Constructs a 128-bit integer vector, initialized in reverse order
3775 /// This intrinsic is a utility function and does not correspond to a specific
3779 ///    A 16-bit integral value used to initialize bits [15:0] of the result.
3781 ///    A 16-bit integral value used to initialize bits [31:16] of the result.
3783 ///    A 16-bit integral value used to initialize bits [47:32] of the result.
3785 ///    A 16-bit integral value used to initialize bits [63:48] of the result.
3787 ///    A 16-bit integral value used to initialize bits [79:64] of the result.
3789 ///    A 16-bit integral value used to initialize bits [95:80] of the result.
3791 ///    A 16-bit integral value used to initialize bits [111:96] of the result.
3793 ///    A 16-bit integral value used to initialize bits [127:112] of the result.
3794 /// \returns An initialized 128-bit integer vector.
3801 /// Constructs a 128-bit integer vector, initialized in reverse order
3806 /// This intrinsic is a utility function and does not correspond to a specific
3841 /// \returns An initialized 128-bit integer vector.
3850 /// Creates a 128-bit integer vector initialized to zero.
3856 /// \returns An initialized 128-bit integer vector with all elements set to
3862 /// Stores a 128-bit integer vector to a memory location aligned on a
3863 ///    128-bit boundary.
3870 ///    A pointer to an aligned memory location that will receive the integer
3873 ///    A 128-bit integer vector containing the values to be moved.
3879 /// Stores a 128-bit integer vector to an unaligned memory location.
3886 ///    A pointer to a memory location that will receive the integer values.
3888 ///    A 128-bit integer vector containing the values to be moved.
3897 /// Stores a 64-bit integer value from the low element of a 128-bit integer
3905 ///    A pointer to a 64-bit memory location. The address of the memory
3908 ///    A 128-bit integer vector containing the value to be stored.
3917 /// Stores a 32-bit integer value from the low element of a 128-bit integer
3925 ///    A pointer to a 32-bit memory location. The address of the memory
3928 ///    A 128-bit integer vector containing the value to be stored.
3937 /// Stores a 16-bit integer value from the low element of a 128-bit integer
3942 /// This intrinsic does not correspond to a specific instruction.
3945 ///    A pointer to a 16-bit memory location. The address of the memory
3948 ///    A 128-bit integer vector containing the value to be stored.
3958 ///    specified unaligned memory location. When a mask bit is 1, the
3971 ///    A 128-bit integer vector containing the values to be moved.
3973 ///    A 128-bit integer vector containing the mask. The most significant bit of
3976 ///    A pointer to an unaligned 128-bit memory location where the specified
3984 /// Stores the lower 64 bits of a 128-bit integer vector of [2 x i64] to
3985 ///    a memory location.
3992 ///    A pointer to a 64-bit memory location that will receive the lower 64 bits
3995 ///    A 128-bit integer vector of [2 x i64]. The lower 64 bits contain the
4005 /// Stores a 128-bit floating point vector of [2 x double] to a 128-bit
4016 ///    A pointer to the 128-bit aligned memory location used to store the value.
4018 ///    A vector of [2 x double] containing the 64-bit values to be stored.
4024 /// Stores a 128-bit integer vector to a 128-bit aligned memory location.
4034 ///    A pointer to the 128-bit aligned memory location used to store the value.
4036 ///    A 128-bit integer vector containing the values to be stored.
4042 /// Stores a 32-bit integer value in the specified memory location.
4052 ///    A pointer to the 32-bit memory location used to store the value.
4054 ///    A 32-bit integer containing the value to be stored.
4062 /// Stores a 64-bit integer value in the specified memory location.
4072 ///    A pointer to the 64-bit memory location used to store the value.
4074 ///    A 64-bit integer containing the value to be stored.
4086 /// The cache line containing \a __p is flushed and invalidated from all
4094 ///    A pointer to the memory location used to identify the cache line to be
4124 /// Converts, with saturation, 16-bit signed integers from both 128-bit integer
4136 ///   A 128-bit integer vector of [8 x i16]. The converted [8 x i8] values are
4139 ///   A 128-bit integer vector of [8 x i16]. The converted [8 x i8] values are
4141 /// \returns A 128-bit vector of [16 x i8] containing the converted values.
4147 /// Converts, with saturation, 32-bit signed integers from both 128-bit integer
4159 ///    A 128-bit integer vector of [4 x i32]. The converted [4 x i16] values
4162 ///    A 128-bit integer vector of [4 x i32]. The converted [4 x i16] values
4164 /// \returns A 128-bit vector of [8 x i16] containing the converted values.
4170 /// Converts, with saturation, 16-bit signed integers from both 128-bit integer
4182 ///    A 128-bit integer vector of [8 x i16]. The converted [8 x i8] values are
4185 ///    A 128-bit integer vector of [8 x i16]. The converted [8 x i8] values are
4187 /// \returns A 128-bit vector of [16 x i8] containing the converted values.
4193 /// Extracts 16 bits from a 128-bit integer vector of [8 x i16], using
4194 ///    the immediate-value parameter as a selector.
4199 /// __m128i _mm_extract_epi16(__m128i a, const int imm);
4204 /// \param a
4205 ///    A 128-bit integer vector.
4207 ///    An immediate value. Bits [2:0] selects values from \a a to be assigned
4209 ///    000: assign values from bits [15:0] of \a a. \n
4210 ///    001: assign values from bits [31:16] of \a a. \n
4211 ///    010: assign values from bits [47:32] of \a a. \n
4212 ///    011: assign values from bits [63:48] of \a a. \n
4213 ///    100: assign values from bits [79:64] of \a a. \n
4214 ///    101: assign values from bits [95:80] of \a a. \n
4215 ///    110: assign values from bits [111:96] of \a a. \n
4216 ///    111: assign values from bits [127:112] of \a a.
4217 /// \returns An integer, whose lower 16 bits are selected from the 128-bit
4219 #define _mm_extract_epi16(a, imm)                                              \  argument
4220   ((int)(unsigned short)__builtin_ia32_vec_ext_v8hi((__v8hi)(__m128i)(a),      \
4223 /// Constructs a 128-bit integer vector by first making a copy of the
4224 ///    128-bit integer vector parameter, and then inserting the lower 16 bits
4231 /// __m128i _mm_insert_epi16(__m128i a, int b, const int imm);
4236 /// \param a
4237 ///    A 128-bit integer vector of [8 x i16]. This vector is copied to the
4239 ///    the lower 16 bits of \a b.
4242 ///    result beginning at an offset specified by \a imm.
4245 ///    lower 16 bits of \a b are written.
4246 /// \returns A 128-bit integer vector containing the constructed values.
4247 #define _mm_insert_epi16(a, b, imm)                                            \  argument
4248   ((__m128i)__builtin_ia32_vec_set_v8hi((__v8hi)(__m128i)(a), (int)(b),        \
4252 ///    element in a 128-bit integer vector of [16 x i8] to create a 16-bit mask
4260 ///    A 128-bit integer vector containing the values with bits to be extracted.
4261 /// \returns The most significant bits from each 8-bit element in \a __a,
4267 /// Constructs a 128-bit integer vector by shuffling four 32-bit
4268 ///    elements of a 128-bit integer vector parameter, using the immediate-value
4269 ///    parameter as a specifier.
4274 /// __m128i _mm_shuffle_epi32(__m128i a, const int imm);
4279 /// \param a
4280 ///    A 128-bit integer vector containing the values to be copied.
4283 ///    copy from a. The destinations within the 128-bit destination are assigned
4290 ///    00: assign values from bits [31:0] of \a a. \n
4291 ///    01: assign values from bits [63:32] of \a a. \n
4292 ///    10: assign values from bits [95:64] of \a a. \n
4293 ///    11: assign values from bits [127:96] of \a a. \n
4294 ///    Note: To generate a mask, you can use the \c _MM_SHUFFLE macro.
4297 /// \returns A 128-bit integer vector containing the shuffled values.
4298 #define _mm_shuffle_epi32(a, imm)                                              \  argument
4299   ((__m128i)__builtin_ia32_pshufd((__v4si)(__m128i)(a), (int)(imm)))
4301 /// Constructs a 128-bit integer vector by shuffling four lower 16-bit
4302 ///    elements of a 128-bit integer vector of [8 x i16], using the immediate
4303 ///    value parameter as a specifier.
4308 /// __m128i _mm_shufflelo_epi16(__m128i a, const int imm);
4313 /// \param a
4314 ///    A 128-bit integer vector of [8 x i16]. Bits [127:64] are copied to bits
4317 ///    An 8-bit immediate value specifying which elements to copy from \a a. \n
4323 ///    00: assign values from bits [15:0] of \a a. \n
4324 ///    01: assign values from bits [31:16] of \a a. \n
4325 ///    10: assign values from bits [47:32] of \a a. \n
4326 ///    11: assign values from bits [63:48] of \a a. \n
4327 ///    Note: To generate a mask, you can use the \c _MM_SHUFFLE macro.
4330 /// \returns A 128-bit integer vector containing the shuffled values.
4331 #define _mm_shufflelo_epi16(a, imm)                                            \  argument
4332   ((__m128i)__builtin_ia32_pshuflw((__v8hi)(__m128i)(a), (int)(imm)))
4334 /// Constructs a 128-bit integer vector by shuffling four upper 16-bit
4335 ///    elements of a 128-bit integer vector of [8 x i16], using the immediate
4336 ///    value parameter as a specifier.
4341 /// __m128i _mm_shufflehi_epi16(__m128i a, const int imm);
4346 /// \param a
4347 ///    A 128-bit integer vector of [8 x i16]. Bits [63:0] are copied to bits
4350 ///    An 8-bit immediate value specifying which elements to copy from \a a. \n
4356 ///    00: assign values from bits [79:64] of \a a. \n
4357 ///    01: assign values from bits [95:80] of \a a. \n
4358 ///    10: assign values from bits [111:96] of \a a. \n
4359 ///    11: assign values from bits [127:112] of \a a. \n
4360 ///    Note: To generate a mask, you can use the \c _MM_SHUFFLE macro.
4363 /// \returns A 128-bit integer vector containing the shuffled values.
4364 #define _mm_shufflehi_epi16(a, imm)                                            \  argument
4365   ((__m128i)__builtin_ia32_pshufhw((__v8hi)(__m128i)(a), (int)(imm)))
4367 /// Unpacks the high-order (index 8-15) values from two 128-bit vectors
4368 ///    of [16 x i8] and interleaves them into a 128-bit vector of [16 x i8].
4376 ///    A 128-bit vector of [16 x i8].
4386 ///    A 128-bit vector of [16 x i8]. \n
4395 /// \returns A 128-bit vector of [16 x i8] containing the interleaved values.
4403 /// Unpacks the high-order (index 4-7) values from two 128-bit vectors of
4404 ///    [8 x i16] and interleaves them into a 128-bit vector of [8 x i16].
4412 ///    A 128-bit vector of [8 x i16].
4418 ///    A 128-bit vector of [8 x i16].
4423 /// \returns A 128-bit vector of [8 x i16] containing the interleaved values.
4430 /// Unpacks the high-order (index 2,3) values from two 128-bit vectors of
4431 ///    [4 x i32] and interleaves them into a 128-bit vector of [4 x i32].
4439 ///    A 128-bit vector of [4 x i32]. \n
4443 ///    A 128-bit vector of [4 x i32]. \n
4446 /// \returns A 128-bit vector of [4 x i32] containing the interleaved values.
4453 /// Unpacks the high-order 64-bit elements from two 128-bit vectors of
4454 ///    [2 x i64] and interleaves them into a 128-bit vector of [2 x i64].
4462 ///    A 128-bit vector of [2 x i64]. \n
4465 ///    A 128-bit vector of [2 x i64]. \n
4467 /// \returns A 128-bit vector of [2 x i64] containing the interleaved values.
4473 /// Unpacks the low-order (index 0-7) values from two 128-bit vectors of
4474 ///    [16 x i8] and interleaves them into a 128-bit vector of [16 x i8].
4482 ///    A 128-bit vector of [16 x i8]. \n
4492 ///    A 128-bit vector of [16 x i8].
4501 /// \returns A 128-bit vector of [16 x i8] containing the interleaved values.
4509 /// Unpacks the low-order (index 0-3) values from each of the two 128-bit
4510 ///    vectors of [8 x i16] and interleaves them into a 128-bit vector of
4519 ///    A 128-bit vector of [8 x i16].
4525 ///    A 128-bit vector of [8 x i16].
4530 /// \returns A 128-bit vector of [8 x i16] containing the interleaved values.
4537 /// Unpacks the low-order (index 0,1) values from two 128-bit vectors of
4538 ///    [4 x i32] and interleaves them into a 128-bit vector of [4 x i32].
4546 ///    A 128-bit vector of [4 x i32]. \n
4550 ///    A 128-bit vector of [4 x i32]. \n
4553 /// \returns A 128-bit vector of [4 x i32] containing the interleaved values.
4560 /// Unpacks the low-order 64-bit elements from two 128-bit vectors of
4561 ///    [2 x i64] and interleaves them into a 128-bit vector of [2 x i64].
4569 ///    A 128-bit vector of [2 x i64]. \n
4572 ///    A 128-bit vector of [2 x i64]. \n
4574 /// \returns A 128-bit vector of [2 x i64] containing the interleaved values.
4580 /// Returns the lower 64 bits of a 128-bit integer vector as a 64-bit
4588 ///    A 128-bit integer vector operand. The lower 64 bits are moved to the
4590 /// \returns A 64-bit integer containing the lower 64 bits of the parameter.
4595 /// Moves the 64-bit operand to a 128-bit integer vector, zeroing the
4603 ///    A 64-bit value.
4604 /// \returns A 128-bit integer vector. The lower 64 bits contain the value from
4610 /// Moves the lower 64 bits of a 128-bit integer vector to a 128-bit
4618 ///    A 128-bit integer vector operand. The lower 64 bits are moved to the
4620 /// \returns A 128-bit integer vector. The lower 64 bits contain the value from
4626 /// Unpacks the high-order 64-bit elements from two 128-bit vectors of
4627 ///    [2 x double] and interleaves them into a 128-bit vector of [2 x
4635 ///    A 128-bit vector of [2 x double]. \n
4638 ///    A 128-bit vector of [2 x double]. \n
4640 /// \returns A 128-bit vector of [2 x double] containing the interleaved values.
4646 /// Unpacks the low-order 64-bit elements from two 128-bit vectors
4647 ///    of [2 x double] and interleaves them into a 128-bit vector of [2 x
4655 ///    A 128-bit vector of [2 x double]. \n
4658 ///    A 128-bit vector of [2 x double]. \n
4660 /// \returns A 128-bit vector of [2 x double] containing the interleaved values.
4666 /// Extracts the sign bits of the double-precision values in the 128-bit
4675 ///    A 128-bit vector of [2 x double] containing the values with sign bits to
4677 /// \returns The sign bits from each of the double-precision elements in \a __a,
4683 /// Constructs a 128-bit floating-point vector of [2 x double] from two
4684 ///    128-bit vector parameters of [2 x double], using the immediate-value
4685 ///     parameter as a specifier.
4690 /// __m128d _mm_shuffle_pd(__m128d a, __m128d b, const int i);
4695 /// \param a
4696 ///    A 128-bit vector of [2 x double].
4698 ///    A 128-bit vector of [2 x double].
4701 ///    elements to copy from \a a and \a b: \n
4702 ///    Bit[0] = 0: lower element of \a a copied to lower element of result. \n
4703 ///    Bit[0] = 1: upper element of \a a copied to lower element of result. \n
4704 ///    Bit[1] = 0: lower element of \a b copied to upper element of result. \n
4705 ///    Bit[1] = 1: upper element of \a b copied to upper element of result. \n
4706 ///    Note: To generate a mask, you can use the \c _MM_SHUFFLE2 macro.
4707 ///    <c>_MM_SHUFFLE2(b1, b0)</c> can create a 2-bit mask of the form
4709 /// \returns A 128-bit vector of [2 x double] containing the shuffled values.
4710 #define _mm_shuffle_pd(a, b, i)                                                \  argument
4711   ((__m128d)__builtin_ia32_shufpd((__v2df)(__m128d)(a), (__v2df)(__m128d)(b),  \
4714 /// Casts a 128-bit floating-point vector of [2 x double] into a 128-bit
4722 ///    A 128-bit floating-point vector of [2 x double].
4723 /// \returns A 128-bit floating-point vector of [4 x float] containing the same
4729 /// Casts a 128-bit floating-point vector of [2 x double] into a 128-bit
4737 ///    A 128-bit floating-point vector of [2 x double].
4738 /// \returns A 128-bit integer vector containing the same bitwise pattern as the
4744 /// Casts a 128-bit floating-point vector of [4 x float] into a 128-bit
4752 ///    A 128-bit floating-point vector of [4 x float].
4753 /// \returns A 128-bit floating-point vector of [2 x double] containing the same
4759 /// Casts a 128-bit floating-point vector of [4 x float] into a 128-bit
4767 ///    A 128-bit floating-point vector of [4 x float].
4768 /// \returns A 128-bit integer vector containing the same bitwise pattern as the
4774 /// Casts a 128-bit integer vector into a 128-bit floating-point vector
4782 ///    A 128-bit integer vector.
4783 /// \returns A 128-bit floating-point vector of [4 x float] containing the same
4789 /// Casts a 128-bit integer vector into a 128-bit floating-point vector
4797 ///    A 128-bit integer vector.
4798 /// \returns A 128-bit floating-point vector of [2 x double] containing the same
4805 ///    128-bit vectors of [2 x double], using the operation specified by the
4809 ///    If either value in a comparison is NaN, comparisons that are ordered
4815 /// __m128d _mm_cmp_pd(__m128d a, __m128d b, const int c);
4820 /// \param a
4821 ///    A 128-bit vector of [2 x double].
4823 ///    A 128-bit vector of [2 x double].
4835 /// \returns A 128-bit vector of [2 x double] containing the comparison results.
4836 #define _mm_cmp_pd(a, b, c)                                                    \  argument
4837   ((__m128d)__builtin_ia32_cmppd((__v2df)(__m128d)(a), (__v2df)(__m128d)(b),   \
4841 ///    two 128-bit vectors of [2 x double], using the operation specified by the
4845 ///    If either value in a comparison is NaN, comparisons that are ordered
4851 /// __m128d _mm_cmp_sd(__m128d a, __m128d b, const int c);
4856 /// \param a
4857 ///    A 128-bit vector of [2 x double].
4859 ///    A 128-bit vector of [2 x double].
4871 /// \returns A 128-bit vector of [2 x double] containing the comparison results.
4872 #define _mm_cmp_sd(a, b, c)                                                    \  argument
4873   ((__m128d)__builtin_ia32_cmpsd((__v2df)(__m128d)(a), (__v2df)(__m128d)(b),   \
4880 /// Indicates that a spin loop is being executed for the purposes of