Lines Matching refs:__p1

433 __ai uint8x16_t vabdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
435   __ret = (uint8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
439 __ai uint8x16_t vabdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
441   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
447 __ai uint8x16_t __noswap_vabdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
449   __ret = (uint8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
455 __ai uint32x4_t vabdq_u32(uint32x4_t __p0, uint32x4_t __p1) {
457   __ret = (uint32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
461 __ai uint32x4_t vabdq_u32(uint32x4_t __p0, uint32x4_t __p1) {
463   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
469 __ai uint32x4_t __noswap_vabdq_u32(uint32x4_t __p0, uint32x4_t __p1) {
471   __ret = (uint32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
477 __ai uint16x8_t vabdq_u16(uint16x8_t __p0, uint16x8_t __p1) {
479   __ret = (uint16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
483 __ai uint16x8_t vabdq_u16(uint16x8_t __p0, uint16x8_t __p1) {
485   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
491 __ai uint16x8_t __noswap_vabdq_u16(uint16x8_t __p0, uint16x8_t __p1) {
493   __ret = (uint16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
499 __ai int8x16_t vabdq_s8(int8x16_t __p0, int8x16_t __p1) {
501   __ret = (int8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
505 __ai int8x16_t vabdq_s8(int8x16_t __p0, int8x16_t __p1) {
507   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
513 __ai int8x16_t __noswap_vabdq_s8(int8x16_t __p0, int8x16_t __p1) {
515   __ret = (int8x16_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
521 __ai float32x4_t vabdq_f32(float32x4_t __p0, float32x4_t __p1) {
523   __ret = (float32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
527 __ai float32x4_t vabdq_f32(float32x4_t __p0, float32x4_t __p1) {
529   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
538 __ai int32x4_t vabdq_s32(int32x4_t __p0, int32x4_t __p1) {
540   __ret = (int32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
544 __ai int32x4_t vabdq_s32(int32x4_t __p0, int32x4_t __p1) {
546   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
552 __ai int32x4_t __noswap_vabdq_s32(int32x4_t __p0, int32x4_t __p1) {
554   __ret = (int32x4_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
560 __ai int16x8_t vabdq_s16(int16x8_t __p0, int16x8_t __p1) {
562   __ret = (int16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
566 __ai int16x8_t vabdq_s16(int16x8_t __p0, int16x8_t __p1) {
568   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
574 __ai int16x8_t __noswap_vabdq_s16(int16x8_t __p0, int16x8_t __p1) {
576   __ret = (int16x8_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
582 __ai uint8x8_t vabd_u8(uint8x8_t __p0, uint8x8_t __p1) {
584   __ret = (uint8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
588 __ai uint8x8_t vabd_u8(uint8x8_t __p0, uint8x8_t __p1) {
590   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
596 __ai uint8x8_t __noswap_vabd_u8(uint8x8_t __p0, uint8x8_t __p1) {
598   __ret = (uint8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
604 __ai uint32x2_t vabd_u32(uint32x2_t __p0, uint32x2_t __p1) {
606   __ret = (uint32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
610 __ai uint32x2_t vabd_u32(uint32x2_t __p0, uint32x2_t __p1) {
612   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
618 __ai uint32x2_t __noswap_vabd_u32(uint32x2_t __p0, uint32x2_t __p1) {
620   __ret = (uint32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
626 __ai uint16x4_t vabd_u16(uint16x4_t __p0, uint16x4_t __p1) {
628   __ret = (uint16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
632 __ai uint16x4_t vabd_u16(uint16x4_t __p0, uint16x4_t __p1) {
634   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
640 __ai uint16x4_t __noswap_vabd_u16(uint16x4_t __p0, uint16x4_t __p1) {
642   __ret = (uint16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
648 __ai int8x8_t vabd_s8(int8x8_t __p0, int8x8_t __p1) {
650   __ret = (int8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
654 __ai int8x8_t vabd_s8(int8x8_t __p0, int8x8_t __p1) {
656   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
662 __ai int8x8_t __noswap_vabd_s8(int8x8_t __p0, int8x8_t __p1) {
664   __ret = (int8x8_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
670 __ai float32x2_t vabd_f32(float32x2_t __p0, float32x2_t __p1) {
672   __ret = (float32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
676 __ai float32x2_t vabd_f32(float32x2_t __p0, float32x2_t __p1) {
678   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
687 __ai int32x2_t vabd_s32(int32x2_t __p0, int32x2_t __p1) {
689   __ret = (int32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
693 __ai int32x2_t vabd_s32(int32x2_t __p0, int32x2_t __p1) {
695   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
701 __ai int32x2_t __noswap_vabd_s32(int32x2_t __p0, int32x2_t __p1) {
703   __ret = (int32x2_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
709 __ai int16x4_t vabd_s16(int16x4_t __p0, int16x4_t __p1) {
711   __ret = (int16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
715 __ai int16x4_t vabd_s16(int16x4_t __p0, int16x4_t __p1) {
717   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
723 __ai int16x4_t __noswap_vabd_s16(int16x4_t __p0, int16x4_t __p1) {
725   __ret = (int16x4_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
859 __ai uint8x16_t vaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
861   __ret = __p0 + __p1;
865 __ai uint8x16_t vaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
867   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
876 __ai uint32x4_t vaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
878   __ret = __p0 + __p1;
882 __ai uint32x4_t vaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
884   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
893 __ai uint64x2_t vaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
895   __ret = __p0 + __p1;
899 __ai uint64x2_t vaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
901   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
910 __ai uint16x8_t vaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
912   __ret = __p0 + __p1;
916 __ai uint16x8_t vaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
918   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
927 __ai int8x16_t vaddq_s8(int8x16_t __p0, int8x16_t __p1) {
929   __ret = __p0 + __p1;
933 __ai int8x16_t vaddq_s8(int8x16_t __p0, int8x16_t __p1) {
935   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
944 __ai float32x4_t vaddq_f32(float32x4_t __p0, float32x4_t __p1) {
946   __ret = __p0 + __p1;
950 __ai float32x4_t vaddq_f32(float32x4_t __p0, float32x4_t __p1) {
952   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
961 __ai int32x4_t vaddq_s32(int32x4_t __p0, int32x4_t __p1) {
963   __ret = __p0 + __p1;
967 __ai int32x4_t vaddq_s32(int32x4_t __p0, int32x4_t __p1) {
969   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
978 __ai int64x2_t vaddq_s64(int64x2_t __p0, int64x2_t __p1) {
980   __ret = __p0 + __p1;
984 __ai int64x2_t vaddq_s64(int64x2_t __p0, int64x2_t __p1) {
986   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
995 __ai int16x8_t vaddq_s16(int16x8_t __p0, int16x8_t __p1) {
997   __ret = __p0 + __p1;
1001 __ai int16x8_t vaddq_s16(int16x8_t __p0, int16x8_t __p1) {
1003   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1012 __ai uint8x8_t vadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
1014   __ret = __p0 + __p1;
1018 __ai uint8x8_t vadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
1020   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1029 __ai uint32x2_t vadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
1031   __ret = __p0 + __p1;
1035 __ai uint32x2_t vadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
1037   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1046 __ai uint64x1_t vadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
1048   __ret = __p0 + __p1;
1052 __ai uint64x1_t vadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
1054   __ret = __p0 + __p1;
1060 __ai uint16x4_t vadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
1062   __ret = __p0 + __p1;
1066 __ai uint16x4_t vadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
1068   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1077 __ai int8x8_t vadd_s8(int8x8_t __p0, int8x8_t __p1) {
1079   __ret = __p0 + __p1;
1083 __ai int8x8_t vadd_s8(int8x8_t __p0, int8x8_t __p1) {
1085   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1094 __ai float32x2_t vadd_f32(float32x2_t __p0, float32x2_t __p1) {
1096   __ret = __p0 + __p1;
1100 __ai float32x2_t vadd_f32(float32x2_t __p0, float32x2_t __p1) {
1102   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1111 __ai int32x2_t vadd_s32(int32x2_t __p0, int32x2_t __p1) {
1113   __ret = __p0 + __p1;
1117 __ai int32x2_t vadd_s32(int32x2_t __p0, int32x2_t __p1) {
1119   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1128 __ai int64x1_t vadd_s64(int64x1_t __p0, int64x1_t __p1) {
1130   __ret = __p0 + __p1;
1134 __ai int64x1_t vadd_s64(int64x1_t __p0, int64x1_t __p1) {
1136   __ret = __p0 + __p1;
1142 __ai int16x4_t vadd_s16(int16x4_t __p0, int16x4_t __p1) {
1144   __ret = __p0 + __p1;
1148 __ai int16x4_t vadd_s16(int16x4_t __p0, int16x4_t __p1) {
1150   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1159 __ai uint16x4_t vaddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
1161   __ret = (uint16x4_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
1165 __ai uint16x4_t vaddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
1167   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1173 __ai uint16x4_t __noswap_vaddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
1175   __ret = (uint16x4_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
1181 __ai uint32x2_t vaddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
1183   __ret = (uint32x2_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
1187 __ai uint32x2_t vaddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
1189   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1195 __ai uint32x2_t __noswap_vaddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
1197   __ret = (uint32x2_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
1203 __ai uint8x8_t vaddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
1205   __ret = (uint8x8_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
1209 __ai uint8x8_t vaddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
1211   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1217 __ai uint8x8_t __noswap_vaddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
1219   __ret = (uint8x8_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
1225 __ai int16x4_t vaddhn_s32(int32x4_t __p0, int32x4_t __p1) {
1227   __ret = (int16x4_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
1231 __ai int16x4_t vaddhn_s32(int32x4_t __p0, int32x4_t __p1) {
1233   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1239 __ai int16x4_t __noswap_vaddhn_s32(int32x4_t __p0, int32x4_t __p1) {
1241   __ret = (int16x4_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
1247 __ai int32x2_t vaddhn_s64(int64x2_t __p0, int64x2_t __p1) {
1249   __ret = (int32x2_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
1253 __ai int32x2_t vaddhn_s64(int64x2_t __p0, int64x2_t __p1) {
1255   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1261 __ai int32x2_t __noswap_vaddhn_s64(int64x2_t __p0, int64x2_t __p1) {
1263   __ret = (int32x2_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
1269 __ai int8x8_t vaddhn_s16(int16x8_t __p0, int16x8_t __p1) {
1271   __ret = (int8x8_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
1275 __ai int8x8_t vaddhn_s16(int16x8_t __p0, int16x8_t __p1) {
1277   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1283 __ai int8x8_t __noswap_vaddhn_s16(int16x8_t __p0, int16x8_t __p1) {
1285   __ret = (int8x8_t) __builtin_neon_vaddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
1291 __ai uint8x16_t vandq_u8(uint8x16_t __p0, uint8x16_t __p1) {
1293   __ret = __p0 & __p1;
1297 __ai uint8x16_t vandq_u8(uint8x16_t __p0, uint8x16_t __p1) {
1299   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1308 __ai uint32x4_t vandq_u32(uint32x4_t __p0, uint32x4_t __p1) {
1310   __ret = __p0 & __p1;
1314 __ai uint32x4_t vandq_u32(uint32x4_t __p0, uint32x4_t __p1) {
1316   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1325 __ai uint64x2_t vandq_u64(uint64x2_t __p0, uint64x2_t __p1) {
1327   __ret = __p0 & __p1;
1331 __ai uint64x2_t vandq_u64(uint64x2_t __p0, uint64x2_t __p1) {
1333   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1342 __ai uint16x8_t vandq_u16(uint16x8_t __p0, uint16x8_t __p1) {
1344   __ret = __p0 & __p1;
1348 __ai uint16x8_t vandq_u16(uint16x8_t __p0, uint16x8_t __p1) {
1350   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1359 __ai int8x16_t vandq_s8(int8x16_t __p0, int8x16_t __p1) {
1361   __ret = __p0 & __p1;
1365 __ai int8x16_t vandq_s8(int8x16_t __p0, int8x16_t __p1) {
1367   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1376 __ai int32x4_t vandq_s32(int32x4_t __p0, int32x4_t __p1) {
1378   __ret = __p0 & __p1;
1382 __ai int32x4_t vandq_s32(int32x4_t __p0, int32x4_t __p1) {
1384   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1393 __ai int64x2_t vandq_s64(int64x2_t __p0, int64x2_t __p1) {
1395   __ret = __p0 & __p1;
1399 __ai int64x2_t vandq_s64(int64x2_t __p0, int64x2_t __p1) {
1401   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1410 __ai int16x8_t vandq_s16(int16x8_t __p0, int16x8_t __p1) {
1412   __ret = __p0 & __p1;
1416 __ai int16x8_t vandq_s16(int16x8_t __p0, int16x8_t __p1) {
1418   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1427 __ai uint8x8_t vand_u8(uint8x8_t __p0, uint8x8_t __p1) {
1429   __ret = __p0 & __p1;
1433 __ai uint8x8_t vand_u8(uint8x8_t __p0, uint8x8_t __p1) {
1435   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1444 __ai uint32x2_t vand_u32(uint32x2_t __p0, uint32x2_t __p1) {
1446   __ret = __p0 & __p1;
1450 __ai uint32x2_t vand_u32(uint32x2_t __p0, uint32x2_t __p1) {
1452   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1461 __ai uint64x1_t vand_u64(uint64x1_t __p0, uint64x1_t __p1) {
1463   __ret = __p0 & __p1;
1467 __ai uint64x1_t vand_u64(uint64x1_t __p0, uint64x1_t __p1) {
1469   __ret = __p0 & __p1;
1475 __ai uint16x4_t vand_u16(uint16x4_t __p0, uint16x4_t __p1) {
1477   __ret = __p0 & __p1;
1481 __ai uint16x4_t vand_u16(uint16x4_t __p0, uint16x4_t __p1) {
1483   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1492 __ai int8x8_t vand_s8(int8x8_t __p0, int8x8_t __p1) {
1494   __ret = __p0 & __p1;
1498 __ai int8x8_t vand_s8(int8x8_t __p0, int8x8_t __p1) {
1500   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1509 __ai int32x2_t vand_s32(int32x2_t __p0, int32x2_t __p1) {
1511   __ret = __p0 & __p1;
1515 __ai int32x2_t vand_s32(int32x2_t __p0, int32x2_t __p1) {
1517   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1526 __ai int64x1_t vand_s64(int64x1_t __p0, int64x1_t __p1) {
1528   __ret = __p0 & __p1;
1532 __ai int64x1_t vand_s64(int64x1_t __p0, int64x1_t __p1) {
1534   __ret = __p0 & __p1;
1540 __ai int16x4_t vand_s16(int16x4_t __p0, int16x4_t __p1) {
1542   __ret = __p0 & __p1;
1546 __ai int16x4_t vand_s16(int16x4_t __p0, int16x4_t __p1) {
1548   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1557 __ai uint8x16_t vbicq_u8(uint8x16_t __p0, uint8x16_t __p1) {
1559   __ret = __p0 & ~__p1;
1563 __ai uint8x16_t vbicq_u8(uint8x16_t __p0, uint8x16_t __p1) {
1565   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1574 __ai uint32x4_t vbicq_u32(uint32x4_t __p0, uint32x4_t __p1) {
1576   __ret = __p0 & ~__p1;
1580 __ai uint32x4_t vbicq_u32(uint32x4_t __p0, uint32x4_t __p1) {
1582   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1591 __ai uint64x2_t vbicq_u64(uint64x2_t __p0, uint64x2_t __p1) {
1593   __ret = __p0 & ~__p1;
1597 __ai uint64x2_t vbicq_u64(uint64x2_t __p0, uint64x2_t __p1) {
1599   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1608 __ai uint16x8_t vbicq_u16(uint16x8_t __p0, uint16x8_t __p1) {
1610   __ret = __p0 & ~__p1;
1614 __ai uint16x8_t vbicq_u16(uint16x8_t __p0, uint16x8_t __p1) {
1616   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1625 __ai int8x16_t vbicq_s8(int8x16_t __p0, int8x16_t __p1) {
1627   __ret = __p0 & ~__p1;
1631 __ai int8x16_t vbicq_s8(int8x16_t __p0, int8x16_t __p1) {
1633   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1642 __ai int32x4_t vbicq_s32(int32x4_t __p0, int32x4_t __p1) {
1644   __ret = __p0 & ~__p1;
1648 __ai int32x4_t vbicq_s32(int32x4_t __p0, int32x4_t __p1) {
1650   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1659 __ai int64x2_t vbicq_s64(int64x2_t __p0, int64x2_t __p1) {
1661   __ret = __p0 & ~__p1;
1665 __ai int64x2_t vbicq_s64(int64x2_t __p0, int64x2_t __p1) {
1667   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1676 __ai int16x8_t vbicq_s16(int16x8_t __p0, int16x8_t __p1) {
1678   __ret = __p0 & ~__p1;
1682 __ai int16x8_t vbicq_s16(int16x8_t __p0, int16x8_t __p1) {
1684   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1693 __ai uint8x8_t vbic_u8(uint8x8_t __p0, uint8x8_t __p1) {
1695   __ret = __p0 & ~__p1;
1699 __ai uint8x8_t vbic_u8(uint8x8_t __p0, uint8x8_t __p1) {
1701   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1710 __ai uint32x2_t vbic_u32(uint32x2_t __p0, uint32x2_t __p1) {
1712   __ret = __p0 & ~__p1;
1716 __ai uint32x2_t vbic_u32(uint32x2_t __p0, uint32x2_t __p1) {
1718   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1727 __ai uint64x1_t vbic_u64(uint64x1_t __p0, uint64x1_t __p1) {
1729   __ret = __p0 & ~__p1;
1733 __ai uint64x1_t vbic_u64(uint64x1_t __p0, uint64x1_t __p1) {
1735   __ret = __p0 & ~__p1;
1741 __ai uint16x4_t vbic_u16(uint16x4_t __p0, uint16x4_t __p1) {
1743   __ret = __p0 & ~__p1;
1747 __ai uint16x4_t vbic_u16(uint16x4_t __p0, uint16x4_t __p1) {
1749   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1758 __ai int8x8_t vbic_s8(int8x8_t __p0, int8x8_t __p1) {
1760   __ret = __p0 & ~__p1;
1764 __ai int8x8_t vbic_s8(int8x8_t __p0, int8x8_t __p1) {
1766   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1775 __ai int32x2_t vbic_s32(int32x2_t __p0, int32x2_t __p1) {
1777   __ret = __p0 & ~__p1;
1781 __ai int32x2_t vbic_s32(int32x2_t __p0, int32x2_t __p1) {
1783   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1792 __ai int64x1_t vbic_s64(int64x1_t __p0, int64x1_t __p1) {
1794   __ret = __p0 & ~__p1;
1798 __ai int64x1_t vbic_s64(int64x1_t __p0, int64x1_t __p1) {
1800   __ret = __p0 & ~__p1;
1806 __ai int16x4_t vbic_s16(int16x4_t __p0, int16x4_t __p1) {
1808   __ret = __p0 & ~__p1;
1812 __ai int16x4_t vbic_s16(int16x4_t __p0, int16x4_t __p1) {
1814   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1823 __ai poly8x8_t vbsl_p8(uint8x8_t __p0, poly8x8_t __p1, poly8x8_t __p2) {
1825   __ret = (poly8x8_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 4);
1829 __ai poly8x8_t vbsl_p8(uint8x8_t __p0, poly8x8_t __p1, poly8x8_t __p2) {
1831   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1841 __ai poly16x4_t vbsl_p16(uint16x4_t __p0, poly16x4_t __p1, poly16x4_t __p2) {
1843   __ret = (poly16x4_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 5);
1847 __ai poly16x4_t vbsl_p16(uint16x4_t __p0, poly16x4_t __p1, poly16x4_t __p2) {
1849   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1859 __ai poly8x16_t vbslq_p8(uint8x16_t __p0, poly8x16_t __p1, poly8x16_t __p2) {
1861   __ret = (poly8x16_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 36);
1865 __ai poly8x16_t vbslq_p8(uint8x16_t __p0, poly8x16_t __p1, poly8x16_t __p2) {
1867   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1877 __ai poly16x8_t vbslq_p16(uint16x8_t __p0, poly16x8_t __p1, poly16x8_t __p2) {
1879   __ret = (poly16x8_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 37);
1883 __ai poly16x8_t vbslq_p16(uint16x8_t __p0, poly16x8_t __p1, poly16x8_t __p2) {
1885   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1895 __ai uint8x16_t vbslq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
1897   __ret = (uint8x16_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 48);
1901 __ai uint8x16_t vbslq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
1903   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1913 __ai uint32x4_t vbslq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
1915   __ret = (uint32x4_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
1919 __ai uint32x4_t vbslq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
1921   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
1931 __ai uint64x2_t vbslq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
1933   __ret = (uint64x2_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 51);
1937 __ai uint64x2_t vbslq_u64(uint64x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
1939   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
1949 __ai uint16x8_t vbslq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
1951   __ret = (uint16x8_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 49);
1955 __ai uint16x8_t vbslq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
1957   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
1967 __ai int8x16_t vbslq_s8(uint8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
1969   __ret = (int8x16_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 32);
1973 __ai int8x16_t vbslq_s8(uint8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
1975   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
1985 __ai float32x4_t vbslq_f32(uint32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
1987   __ret = (float32x4_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
1991 __ai float32x4_t vbslq_f32(uint32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
1993   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2003 __ai int32x4_t vbslq_s32(uint32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
2005   __ret = (int32x4_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 34);
2009 __ai int32x4_t vbslq_s32(uint32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
2011   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2021 __ai int64x2_t vbslq_s64(uint64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
2023   __ret = (int64x2_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 35);
2027 __ai int64x2_t vbslq_s64(uint64x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
2029   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2039 __ai int16x8_t vbslq_s16(uint16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
2041   __ret = (int16x8_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 33);
2045 __ai int16x8_t vbslq_s16(uint16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
2047   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2057 __ai uint8x8_t vbsl_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
2059   __ret = (uint8x8_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 16);
2063 __ai uint8x8_t vbsl_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
2065   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2075 __ai uint32x2_t vbsl_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
2077   __ret = (uint32x2_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 18);
2081 __ai uint32x2_t vbsl_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
2083   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2093 __ai uint64x1_t vbsl_u64(uint64x1_t __p0, uint64x1_t __p1, uint64x1_t __p2) {
2095   __ret = (uint64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 19);
2099 __ai uint64x1_t vbsl_u64(uint64x1_t __p0, uint64x1_t __p1, uint64x1_t __p2) {
2101   __ret = (uint64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 19);
2107 __ai uint16x4_t vbsl_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
2109   __ret = (uint16x4_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 17);
2113 __ai uint16x4_t vbsl_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
2115   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2125 __ai int8x8_t vbsl_s8(uint8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
2127   __ret = (int8x8_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 0);
2131 __ai int8x8_t vbsl_s8(uint8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
2133   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2143 __ai float32x2_t vbsl_f32(uint32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
2145   __ret = (float32x2_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
2149 __ai float32x2_t vbsl_f32(uint32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
2151   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2161 __ai int32x2_t vbsl_s32(uint32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
2163   __ret = (int32x2_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 2);
2167 __ai int32x2_t vbsl_s32(uint32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
2169   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2179 __ai int64x1_t vbsl_s64(uint64x1_t __p0, int64x1_t __p1, int64x1_t __p2) {
2181   __ret = (int64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 3);
2185 __ai int64x1_t vbsl_s64(uint64x1_t __p0, int64x1_t __p1, int64x1_t __p2) {
2187   __ret = (int64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 3);
2193 __ai int16x4_t vbsl_s16(uint16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
2195   __ret = (int16x4_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 1);
2199 __ai int16x4_t vbsl_s16(uint16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
2201   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2211 __ai uint32x4_t vcageq_f32(float32x4_t __p0, float32x4_t __p1) {
2213   __ret = (uint32x4_t) __builtin_neon_vcageq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
2217 __ai uint32x4_t vcageq_f32(float32x4_t __p0, float32x4_t __p1) {
2219   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2228 __ai uint32x2_t vcage_f32(float32x2_t __p0, float32x2_t __p1) {
2230   __ret = (uint32x2_t) __builtin_neon_vcage_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
2234 __ai uint32x2_t vcage_f32(float32x2_t __p0, float32x2_t __p1) {
2236   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2245 __ai uint32x4_t vcagtq_f32(float32x4_t __p0, float32x4_t __p1) {
2247   __ret = (uint32x4_t) __builtin_neon_vcagtq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
2251 __ai uint32x4_t vcagtq_f32(float32x4_t __p0, float32x4_t __p1) {
2253   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2262 __ai uint32x2_t vcagt_f32(float32x2_t __p0, float32x2_t __p1) {
2264   __ret = (uint32x2_t) __builtin_neon_vcagt_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
2268 __ai uint32x2_t vcagt_f32(float32x2_t __p0, float32x2_t __p1) {
2270   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2279 __ai uint32x4_t vcaleq_f32(float32x4_t __p0, float32x4_t __p1) {
2281   __ret = (uint32x4_t) __builtin_neon_vcaleq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
2285 __ai uint32x4_t vcaleq_f32(float32x4_t __p0, float32x4_t __p1) {
2287   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2296 __ai uint32x2_t vcale_f32(float32x2_t __p0, float32x2_t __p1) {
2298   __ret = (uint32x2_t) __builtin_neon_vcale_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
2302 __ai uint32x2_t vcale_f32(float32x2_t __p0, float32x2_t __p1) {
2304   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2313 __ai uint32x4_t vcaltq_f32(float32x4_t __p0, float32x4_t __p1) {
2315   __ret = (uint32x4_t) __builtin_neon_vcaltq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
2319 __ai uint32x4_t vcaltq_f32(float32x4_t __p0, float32x4_t __p1) {
2321   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2330 __ai uint32x2_t vcalt_f32(float32x2_t __p0, float32x2_t __p1) {
2332   __ret = (uint32x2_t) __builtin_neon_vcalt_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
2336 __ai uint32x2_t vcalt_f32(float32x2_t __p0, float32x2_t __p1) {
2338   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2347 __ai uint8x8_t vceq_p8(poly8x8_t __p0, poly8x8_t __p1) {
2349   __ret = (uint8x8_t)(__p0 == __p1);
2353 __ai uint8x8_t vceq_p8(poly8x8_t __p0, poly8x8_t __p1) {
2355   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2364 __ai uint8x16_t vceqq_p8(poly8x16_t __p0, poly8x16_t __p1) {
2366   __ret = (uint8x16_t)(__p0 == __p1);
2370 __ai uint8x16_t vceqq_p8(poly8x16_t __p0, poly8x16_t __p1) {
2372   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2381 __ai uint8x16_t vceqq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2383   __ret = (uint8x16_t)(__p0 == __p1);
2387 __ai uint8x16_t vceqq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2389   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2398 __ai uint32x4_t vceqq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2400   __ret = (uint32x4_t)(__p0 == __p1);
2404 __ai uint32x4_t vceqq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2406   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2415 __ai uint16x8_t vceqq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2417   __ret = (uint16x8_t)(__p0 == __p1);
2421 __ai uint16x8_t vceqq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2423   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2432 __ai uint8x16_t vceqq_s8(int8x16_t __p0, int8x16_t __p1) {
2434   __ret = (uint8x16_t)(__p0 == __p1);
2438 __ai uint8x16_t vceqq_s8(int8x16_t __p0, int8x16_t __p1) {
2440   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2449 __ai uint32x4_t vceqq_f32(float32x4_t __p0, float32x4_t __p1) {
2451   __ret = (uint32x4_t)(__p0 == __p1);
2455 __ai uint32x4_t vceqq_f32(float32x4_t __p0, float32x4_t __p1) {
2457   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2466 __ai uint32x4_t vceqq_s32(int32x4_t __p0, int32x4_t __p1) {
2468   __ret = (uint32x4_t)(__p0 == __p1);
2472 __ai uint32x4_t vceqq_s32(int32x4_t __p0, int32x4_t __p1) {
2474   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2483 __ai uint16x8_t vceqq_s16(int16x8_t __p0, int16x8_t __p1) {
2485   __ret = (uint16x8_t)(__p0 == __p1);
2489 __ai uint16x8_t vceqq_s16(int16x8_t __p0, int16x8_t __p1) {
2491   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2500 __ai uint8x8_t vceq_u8(uint8x8_t __p0, uint8x8_t __p1) {
2502   __ret = (uint8x8_t)(__p0 == __p1);
2506 __ai uint8x8_t vceq_u8(uint8x8_t __p0, uint8x8_t __p1) {
2508   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2517 __ai uint32x2_t vceq_u32(uint32x2_t __p0, uint32x2_t __p1) {
2519   __ret = (uint32x2_t)(__p0 == __p1);
2523 __ai uint32x2_t vceq_u32(uint32x2_t __p0, uint32x2_t __p1) {
2525   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2534 __ai uint16x4_t vceq_u16(uint16x4_t __p0, uint16x4_t __p1) {
2536   __ret = (uint16x4_t)(__p0 == __p1);
2540 __ai uint16x4_t vceq_u16(uint16x4_t __p0, uint16x4_t __p1) {
2542   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2551 __ai uint8x8_t vceq_s8(int8x8_t __p0, int8x8_t __p1) {
2553   __ret = (uint8x8_t)(__p0 == __p1);
2557 __ai uint8x8_t vceq_s8(int8x8_t __p0, int8x8_t __p1) {
2559   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2568 __ai uint32x2_t vceq_f32(float32x2_t __p0, float32x2_t __p1) {
2570   __ret = (uint32x2_t)(__p0 == __p1);
2574 __ai uint32x2_t vceq_f32(float32x2_t __p0, float32x2_t __p1) {
2576   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2585 __ai uint32x2_t vceq_s32(int32x2_t __p0, int32x2_t __p1) {
2587   __ret = (uint32x2_t)(__p0 == __p1);
2591 __ai uint32x2_t vceq_s32(int32x2_t __p0, int32x2_t __p1) {
2593   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2602 __ai uint16x4_t vceq_s16(int16x4_t __p0, int16x4_t __p1) {
2604   __ret = (uint16x4_t)(__p0 == __p1);
2608 __ai uint16x4_t vceq_s16(int16x4_t __p0, int16x4_t __p1) {
2610   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2619 __ai uint8x16_t vcgeq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2621   __ret = (uint8x16_t)(__p0 >= __p1);
2625 __ai uint8x16_t vcgeq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2627   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2636 __ai uint32x4_t vcgeq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2638   __ret = (uint32x4_t)(__p0 >= __p1);
2642 __ai uint32x4_t vcgeq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2644   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2653 __ai uint16x8_t vcgeq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2655   __ret = (uint16x8_t)(__p0 >= __p1);
2659 __ai uint16x8_t vcgeq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2661   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2670 __ai uint8x16_t vcgeq_s8(int8x16_t __p0, int8x16_t __p1) {
2672   __ret = (uint8x16_t)(__p0 >= __p1);
2676 __ai uint8x16_t vcgeq_s8(int8x16_t __p0, int8x16_t __p1) {
2678   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2687 __ai uint32x4_t vcgeq_f32(float32x4_t __p0, float32x4_t __p1) {
2689   __ret = (uint32x4_t)(__p0 >= __p1);
2693 __ai uint32x4_t vcgeq_f32(float32x4_t __p0, float32x4_t __p1) {
2695   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2704 __ai uint32x4_t vcgeq_s32(int32x4_t __p0, int32x4_t __p1) {
2706   __ret = (uint32x4_t)(__p0 >= __p1);
2710 __ai uint32x4_t vcgeq_s32(int32x4_t __p0, int32x4_t __p1) {
2712   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2721 __ai uint16x8_t vcgeq_s16(int16x8_t __p0, int16x8_t __p1) {
2723   __ret = (uint16x8_t)(__p0 >= __p1);
2727 __ai uint16x8_t vcgeq_s16(int16x8_t __p0, int16x8_t __p1) {
2729   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2738 __ai uint8x8_t vcge_u8(uint8x8_t __p0, uint8x8_t __p1) {
2740   __ret = (uint8x8_t)(__p0 >= __p1);
2744 __ai uint8x8_t vcge_u8(uint8x8_t __p0, uint8x8_t __p1) {
2746   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2755 __ai uint32x2_t vcge_u32(uint32x2_t __p0, uint32x2_t __p1) {
2757   __ret = (uint32x2_t)(__p0 >= __p1);
2761 __ai uint32x2_t vcge_u32(uint32x2_t __p0, uint32x2_t __p1) {
2763   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2772 __ai uint16x4_t vcge_u16(uint16x4_t __p0, uint16x4_t __p1) {
2774   __ret = (uint16x4_t)(__p0 >= __p1);
2778 __ai uint16x4_t vcge_u16(uint16x4_t __p0, uint16x4_t __p1) {
2780   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2789 __ai uint8x8_t vcge_s8(int8x8_t __p0, int8x8_t __p1) {
2791   __ret = (uint8x8_t)(__p0 >= __p1);
2795 __ai uint8x8_t vcge_s8(int8x8_t __p0, int8x8_t __p1) {
2797   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2806 __ai uint32x2_t vcge_f32(float32x2_t __p0, float32x2_t __p1) {
2808   __ret = (uint32x2_t)(__p0 >= __p1);
2812 __ai uint32x2_t vcge_f32(float32x2_t __p0, float32x2_t __p1) {
2814   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2823 __ai uint32x2_t vcge_s32(int32x2_t __p0, int32x2_t __p1) {
2825   __ret = (uint32x2_t)(__p0 >= __p1);
2829 __ai uint32x2_t vcge_s32(int32x2_t __p0, int32x2_t __p1) {
2831   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
2840 __ai uint16x4_t vcge_s16(int16x4_t __p0, int16x4_t __p1) {
2842   __ret = (uint16x4_t)(__p0 >= __p1);
2846 __ai uint16x4_t vcge_s16(int16x4_t __p0, int16x4_t __p1) {
2848   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2857 __ai uint8x16_t vcgtq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2859   __ret = (uint8x16_t)(__p0 > __p1);
2863 __ai uint8x16_t vcgtq_u8(uint8x16_t __p0, uint8x16_t __p1) {
2865   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2874 __ai uint32x4_t vcgtq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2876   __ret = (uint32x4_t)(__p0 > __p1);
2880 __ai uint32x4_t vcgtq_u32(uint32x4_t __p0, uint32x4_t __p1) {
2882   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2891 __ai uint16x8_t vcgtq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2893   __ret = (uint16x8_t)(__p0 > __p1);
2897 __ai uint16x8_t vcgtq_u16(uint16x8_t __p0, uint16x8_t __p1) {
2899   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2908 __ai uint8x16_t vcgtq_s8(int8x16_t __p0, int8x16_t __p1) {
2910   __ret = (uint8x16_t)(__p0 > __p1);
2914 __ai uint8x16_t vcgtq_s8(int8x16_t __p0, int8x16_t __p1) {
2916   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
2925 __ai uint32x4_t vcgtq_f32(float32x4_t __p0, float32x4_t __p1) {
2927   __ret = (uint32x4_t)(__p0 > __p1);
2931 __ai uint32x4_t vcgtq_f32(float32x4_t __p0, float32x4_t __p1) {
2933   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2942 __ai uint32x4_t vcgtq_s32(int32x4_t __p0, int32x4_t __p1) {
2944   __ret = (uint32x4_t)(__p0 > __p1);
2948 __ai uint32x4_t vcgtq_s32(int32x4_t __p0, int32x4_t __p1) {
2950   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
2959 __ai uint16x8_t vcgtq_s16(int16x8_t __p0, int16x8_t __p1) {
2961   __ret = (uint16x8_t)(__p0 > __p1);
2965 __ai uint16x8_t vcgtq_s16(int16x8_t __p0, int16x8_t __p1) {
2967   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2976 __ai uint8x8_t vcgt_u8(uint8x8_t __p0, uint8x8_t __p1) {
2978   __ret = (uint8x8_t)(__p0 > __p1);
2982 __ai uint8x8_t vcgt_u8(uint8x8_t __p0, uint8x8_t __p1) {
2984   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
2993 __ai uint32x2_t vcgt_u32(uint32x2_t __p0, uint32x2_t __p1) {
2995   __ret = (uint32x2_t)(__p0 > __p1);
2999 __ai uint32x2_t vcgt_u32(uint32x2_t __p0, uint32x2_t __p1) {
3001   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3010 __ai uint16x4_t vcgt_u16(uint16x4_t __p0, uint16x4_t __p1) {
3012   __ret = (uint16x4_t)(__p0 > __p1);
3016 __ai uint16x4_t vcgt_u16(uint16x4_t __p0, uint16x4_t __p1) {
3018   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3027 __ai uint8x8_t vcgt_s8(int8x8_t __p0, int8x8_t __p1) {
3029   __ret = (uint8x8_t)(__p0 > __p1);
3033 __ai uint8x8_t vcgt_s8(int8x8_t __p0, int8x8_t __p1) {
3035   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3044 __ai uint32x2_t vcgt_f32(float32x2_t __p0, float32x2_t __p1) {
3046   __ret = (uint32x2_t)(__p0 > __p1);
3050 __ai uint32x2_t vcgt_f32(float32x2_t __p0, float32x2_t __p1) {
3052   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3061 __ai uint32x2_t vcgt_s32(int32x2_t __p0, int32x2_t __p1) {
3063   __ret = (uint32x2_t)(__p0 > __p1);
3067 __ai uint32x2_t vcgt_s32(int32x2_t __p0, int32x2_t __p1) {
3069   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3078 __ai uint16x4_t vcgt_s16(int16x4_t __p0, int16x4_t __p1) {
3080   __ret = (uint16x4_t)(__p0 > __p1);
3084 __ai uint16x4_t vcgt_s16(int16x4_t __p0, int16x4_t __p1) {
3086   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3095 __ai uint8x16_t vcleq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3097   __ret = (uint8x16_t)(__p0 <= __p1);
3101 __ai uint8x16_t vcleq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3103   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3112 __ai uint32x4_t vcleq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3114   __ret = (uint32x4_t)(__p0 <= __p1);
3118 __ai uint32x4_t vcleq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3120   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3129 __ai uint16x8_t vcleq_u16(uint16x8_t __p0, uint16x8_t __p1) {
3131   __ret = (uint16x8_t)(__p0 <= __p1);
3135 __ai uint16x8_t vcleq_u16(uint16x8_t __p0, uint16x8_t __p1) {
3137   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3146 __ai uint8x16_t vcleq_s8(int8x16_t __p0, int8x16_t __p1) {
3148   __ret = (uint8x16_t)(__p0 <= __p1);
3152 __ai uint8x16_t vcleq_s8(int8x16_t __p0, int8x16_t __p1) {
3154   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3163 __ai uint32x4_t vcleq_f32(float32x4_t __p0, float32x4_t __p1) {
3165   __ret = (uint32x4_t)(__p0 <= __p1);
3169 __ai uint32x4_t vcleq_f32(float32x4_t __p0, float32x4_t __p1) {
3171   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3180 __ai uint32x4_t vcleq_s32(int32x4_t __p0, int32x4_t __p1) {
3182   __ret = (uint32x4_t)(__p0 <= __p1);
3186 __ai uint32x4_t vcleq_s32(int32x4_t __p0, int32x4_t __p1) {
3188   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3197 __ai uint16x8_t vcleq_s16(int16x8_t __p0, int16x8_t __p1) {
3199   __ret = (uint16x8_t)(__p0 <= __p1);
3203 __ai uint16x8_t vcleq_s16(int16x8_t __p0, int16x8_t __p1) {
3205   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3214 __ai uint8x8_t vcle_u8(uint8x8_t __p0, uint8x8_t __p1) {
3216   __ret = (uint8x8_t)(__p0 <= __p1);
3220 __ai uint8x8_t vcle_u8(uint8x8_t __p0, uint8x8_t __p1) {
3222   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3231 __ai uint32x2_t vcle_u32(uint32x2_t __p0, uint32x2_t __p1) {
3233   __ret = (uint32x2_t)(__p0 <= __p1);
3237 __ai uint32x2_t vcle_u32(uint32x2_t __p0, uint32x2_t __p1) {
3239   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3248 __ai uint16x4_t vcle_u16(uint16x4_t __p0, uint16x4_t __p1) {
3250   __ret = (uint16x4_t)(__p0 <= __p1);
3254 __ai uint16x4_t vcle_u16(uint16x4_t __p0, uint16x4_t __p1) {
3256   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3265 __ai uint8x8_t vcle_s8(int8x8_t __p0, int8x8_t __p1) {
3267   __ret = (uint8x8_t)(__p0 <= __p1);
3271 __ai uint8x8_t vcle_s8(int8x8_t __p0, int8x8_t __p1) {
3273   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3282 __ai uint32x2_t vcle_f32(float32x2_t __p0, float32x2_t __p1) {
3284   __ret = (uint32x2_t)(__p0 <= __p1);
3288 __ai uint32x2_t vcle_f32(float32x2_t __p0, float32x2_t __p1) {
3290   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3299 __ai uint32x2_t vcle_s32(int32x2_t __p0, int32x2_t __p1) {
3301   __ret = (uint32x2_t)(__p0 <= __p1);
3305 __ai uint32x2_t vcle_s32(int32x2_t __p0, int32x2_t __p1) {
3307   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3316 __ai uint16x4_t vcle_s16(int16x4_t __p0, int16x4_t __p1) {
3318   __ret = (uint16x4_t)(__p0 <= __p1);
3322 __ai uint16x4_t vcle_s16(int16x4_t __p0, int16x4_t __p1) {
3324   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3429 __ai uint8x16_t vcltq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3431   __ret = (uint8x16_t)(__p0 < __p1);
3435 __ai uint8x16_t vcltq_u8(uint8x16_t __p0, uint8x16_t __p1) {
3437   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3446 __ai uint32x4_t vcltq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3448   __ret = (uint32x4_t)(__p0 < __p1);
3452 __ai uint32x4_t vcltq_u32(uint32x4_t __p0, uint32x4_t __p1) {
3454   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3463 __ai uint16x8_t vcltq_u16(uint16x8_t __p0, uint16x8_t __p1) {
3465   __ret = (uint16x8_t)(__p0 < __p1);
3469 __ai uint16x8_t vcltq_u16(uint16x8_t __p0, uint16x8_t __p1) {
3471   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3480 __ai uint8x16_t vcltq_s8(int8x16_t __p0, int8x16_t __p1) {
3482   __ret = (uint8x16_t)(__p0 < __p1);
3486 __ai uint8x16_t vcltq_s8(int8x16_t __p0, int8x16_t __p1) {
3488   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
3497 __ai uint32x4_t vcltq_f32(float32x4_t __p0, float32x4_t __p1) {
3499   __ret = (uint32x4_t)(__p0 < __p1);
3503 __ai uint32x4_t vcltq_f32(float32x4_t __p0, float32x4_t __p1) {
3505   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3514 __ai uint32x4_t vcltq_s32(int32x4_t __p0, int32x4_t __p1) {
3516   __ret = (uint32x4_t)(__p0 < __p1);
3520 __ai uint32x4_t vcltq_s32(int32x4_t __p0, int32x4_t __p1) {
3522   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3531 __ai uint16x8_t vcltq_s16(int16x8_t __p0, int16x8_t __p1) {
3533   __ret = (uint16x8_t)(__p0 < __p1);
3537 __ai uint16x8_t vcltq_s16(int16x8_t __p0, int16x8_t __p1) {
3539   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3548 __ai uint8x8_t vclt_u8(uint8x8_t __p0, uint8x8_t __p1) {
3550   __ret = (uint8x8_t)(__p0 < __p1);
3554 __ai uint8x8_t vclt_u8(uint8x8_t __p0, uint8x8_t __p1) {
3556   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3565 __ai uint32x2_t vclt_u32(uint32x2_t __p0, uint32x2_t __p1) {
3567   __ret = (uint32x2_t)(__p0 < __p1);
3571 __ai uint32x2_t vclt_u32(uint32x2_t __p0, uint32x2_t __p1) {
3573   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3582 __ai uint16x4_t vclt_u16(uint16x4_t __p0, uint16x4_t __p1) {
3584   __ret = (uint16x4_t)(__p0 < __p1);
3588 __ai uint16x4_t vclt_u16(uint16x4_t __p0, uint16x4_t __p1) {
3590   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3599 __ai uint8x8_t vclt_s8(int8x8_t __p0, int8x8_t __p1) {
3601   __ret = (uint8x8_t)(__p0 < __p1);
3605 __ai uint8x8_t vclt_s8(int8x8_t __p0, int8x8_t __p1) {
3607   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3616 __ai uint32x2_t vclt_f32(float32x2_t __p0, float32x2_t __p1) {
3618   __ret = (uint32x2_t)(__p0 < __p1);
3622 __ai uint32x2_t vclt_f32(float32x2_t __p0, float32x2_t __p1) {
3624   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3633 __ai uint32x2_t vclt_s32(int32x2_t __p0, int32x2_t __p1) {
3635   __ret = (uint32x2_t)(__p0 < __p1);
3639 __ai uint32x2_t vclt_s32(int32x2_t __p0, int32x2_t __p1) {
3641   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
3650 __ai uint16x4_t vclt_s16(int16x4_t __p0, int16x4_t __p1) {
3652   __ret = (uint16x4_t)(__p0 < __p1);
3656 __ai uint16x4_t vclt_s16(int16x4_t __p0, int16x4_t __p1) {
3658   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3955 __ai poly8x16_t vcombine_p8(poly8x8_t __p0, poly8x8_t __p1) {
3957   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
3961 __ai poly8x16_t vcombine_p8(poly8x8_t __p0, poly8x8_t __p1) {
3963   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
3972 __ai poly16x8_t vcombine_p16(poly16x4_t __p0, poly16x4_t __p1) {
3974   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
3978 __ai poly16x8_t vcombine_p16(poly16x4_t __p0, poly16x4_t __p1) {
3980   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
3989 __ai uint8x16_t vcombine_u8(uint8x8_t __p0, uint8x8_t __p1) {
3991   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
3995 __ai uint8x16_t vcombine_u8(uint8x8_t __p0, uint8x8_t __p1) {
3997   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4003 __ai uint8x16_t __noswap_vcombine_u8(uint8x8_t __p0, uint8x8_t __p1) {
4005   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
4011 __ai uint32x4_t vcombine_u32(uint32x2_t __p0, uint32x2_t __p1) {
4013   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3);
4017 __ai uint32x4_t vcombine_u32(uint32x2_t __p0, uint32x2_t __p1) {
4019   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4025 __ai uint32x4_t __noswap_vcombine_u32(uint32x2_t __p0, uint32x2_t __p1) {
4027   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3);
4033 __ai uint64x2_t vcombine_u64(uint64x1_t __p0, uint64x1_t __p1) {
4035   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
4039 __ai uint64x2_t vcombine_u64(uint64x1_t __p0, uint64x1_t __p1) {
4041   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
4048 __ai uint16x8_t vcombine_u16(uint16x4_t __p0, uint16x4_t __p1) {
4050   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
4054 __ai uint16x8_t vcombine_u16(uint16x4_t __p0, uint16x4_t __p1) {
4056   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4062 __ai uint16x8_t __noswap_vcombine_u16(uint16x4_t __p0, uint16x4_t __p1) {
4064   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
4070 __ai int8x16_t vcombine_s8(int8x8_t __p0, int8x8_t __p1) {
4072   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
4076 __ai int8x16_t vcombine_s8(int8x8_t __p0, int8x8_t __p1) {
4078   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
4084 __ai int8x16_t __noswap_vcombine_s8(int8x8_t __p0, int8x8_t __p1) {
4086   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15);
4092 __ai float32x4_t vcombine_f32(float32x2_t __p0, float32x2_t __p1) {
4094   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3);
4098 __ai float32x4_t vcombine_f32(float32x2_t __p0, float32x2_t __p1) {
4100   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4106 __ai float32x4_t __noswap_vcombine_f32(float32x2_t __p0, float32x2_t __p1) {
4108   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3);
4114 __ai float16x8_t vcombine_f16(float16x4_t __p0, float16x4_t __p1) {
4116   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
4120 __ai float16x8_t vcombine_f16(float16x4_t __p0, float16x4_t __p1) {
4122   float16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4128 __ai float16x8_t __noswap_vcombine_f16(float16x4_t __p0, float16x4_t __p1) {
4130   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
4136 __ai int32x4_t vcombine_s32(int32x2_t __p0, int32x2_t __p1) {
4138   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3);
4142 __ai int32x4_t vcombine_s32(int32x2_t __p0, int32x2_t __p1) {
4144   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
4150 __ai int32x4_t __noswap_vcombine_s32(int32x2_t __p0, int32x2_t __p1) {
4152   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3);
4158 __ai int64x2_t vcombine_s64(int64x1_t __p0, int64x1_t __p1) {
4160   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
4164 __ai int64x2_t vcombine_s64(int64x1_t __p0, int64x1_t __p1) {
4166   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
4173 __ai int16x8_t vcombine_s16(int16x4_t __p0, int16x4_t __p1) {
4175   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
4179 __ai int16x8_t vcombine_s16(int16x4_t __p0, int16x4_t __p1) {
4181   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
4187 __ai int16x8_t __noswap_vcombine_s16(int16x4_t __p0, int16x4_t __p1) {
4189   __ret = __builtin_shufflevector(__p0, __p1, 0, 1, 2, 3, 4, 5, 6, 7);
4469 #define vcvtq_n_f32_u32(__p0, __p1) __extension__ ({ \
4472   __ret = (float32x4_t) __builtin_neon_vcvtq_n_f32_v((int8x16_t)__s0, __p1, 50); \
4476 #define vcvtq_n_f32_u32(__p0, __p1) __extension__ ({ \
4480   __ret = (float32x4_t) __builtin_neon_vcvtq_n_f32_v((int8x16_t)__rev0, __p1, 50); \
4487 #define vcvtq_n_f32_s32(__p0, __p1) __extension__ ({ \
4490   __ret = (float32x4_t) __builtin_neon_vcvtq_n_f32_v((int8x16_t)__s0, __p1, 34); \
4494 #define vcvtq_n_f32_s32(__p0, __p1) __extension__ ({ \
4498   __ret = (float32x4_t) __builtin_neon_vcvtq_n_f32_v((int8x16_t)__rev0, __p1, 34); \
4505 #define vcvt_n_f32_u32(__p0, __p1) __extension__ ({ \
4508   __ret = (float32x2_t) __builtin_neon_vcvt_n_f32_v((int8x8_t)__s0, __p1, 18); \
4512 #define vcvt_n_f32_u32(__p0, __p1) __extension__ ({ \
4516   __ret = (float32x2_t) __builtin_neon_vcvt_n_f32_v((int8x8_t)__rev0, __p1, 18); \
4523 #define vcvt_n_f32_s32(__p0, __p1) __extension__ ({ \
4526   __ret = (float32x2_t) __builtin_neon_vcvt_n_f32_v((int8x8_t)__s0, __p1, 2); \
4530 #define vcvt_n_f32_s32(__p0, __p1) __extension__ ({ \
4534   __ret = (float32x2_t) __builtin_neon_vcvt_n_f32_v((int8x8_t)__rev0, __p1, 2); \
4541 #define vcvtq_n_s32_f32(__p0, __p1) __extension__ ({ \
4544   __ret = (int32x4_t) __builtin_neon_vcvtq_n_s32_v((int8x16_t)__s0, __p1, 34); \
4548 #define vcvtq_n_s32_f32(__p0, __p1) __extension__ ({ \
4552   __ret = (int32x4_t) __builtin_neon_vcvtq_n_s32_v((int8x16_t)__rev0, __p1, 34); \
4559 #define vcvt_n_s32_f32(__p0, __p1) __extension__ ({ \
4562   __ret = (int32x2_t) __builtin_neon_vcvt_n_s32_v((int8x8_t)__s0, __p1, 2); \
4566 #define vcvt_n_s32_f32(__p0, __p1) __extension__ ({ \
4570   __ret = (int32x2_t) __builtin_neon_vcvt_n_s32_v((int8x8_t)__rev0, __p1, 2); \
4577 #define vcvtq_n_u32_f32(__p0, __p1) __extension__ ({ \
4580   __ret = (uint32x4_t) __builtin_neon_vcvtq_n_u32_v((int8x16_t)__s0, __p1, 50); \
4584 #define vcvtq_n_u32_f32(__p0, __p1) __extension__ ({ \
4588   __ret = (uint32x4_t) __builtin_neon_vcvtq_n_u32_v((int8x16_t)__rev0, __p1, 50); \
4595 #define vcvt_n_u32_f32(__p0, __p1) __extension__ ({ \
4598   __ret = (uint32x2_t) __builtin_neon_vcvt_n_u32_v((int8x8_t)__s0, __p1, 18); \
4602 #define vcvt_n_u32_f32(__p0, __p1) __extension__ ({ \
4606   __ret = (uint32x2_t) __builtin_neon_vcvt_n_u32_v((int8x8_t)__rev0, __p1, 18); \
4677 #define vdup_lane_p8(__p0, __p1) __extension__ ({ \
4680   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4684 #define vdup_lane_p8(__p0, __p1) __extension__ ({ \
4688   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4695 #define vdup_lane_p16(__p0, __p1) __extension__ ({ \
4698   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
4702 #define vdup_lane_p16(__p0, __p1) __extension__ ({ \
4706   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
4713 #define vdupq_lane_p8(__p0, __p1) __extension__ ({ \
4716   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4720 #define vdupq_lane_p8(__p0, __p1) __extension__ ({ \
4724   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4731 #define vdupq_lane_p16(__p0, __p1) __extension__ ({ \
4734   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4738 #define vdupq_lane_p16(__p0, __p1) __extension__ ({ \
4742   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4749 #define vdupq_lane_u8(__p0, __p1) __extension__ ({ \
4752   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4756 #define vdupq_lane_u8(__p0, __p1) __extension__ ({ \
4760   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4767 #define vdupq_lane_u32(__p0, __p1) __extension__ ({ \
4770   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
4774 #define vdupq_lane_u32(__p0, __p1) __extension__ ({ \
4778   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
4785 #define vdupq_lane_u64(__p0, __p1) __extension__ ({ \
4788   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
4792 #define vdupq_lane_u64(__p0, __p1) __extension__ ({ \
4795   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
4802 #define vdupq_lane_u16(__p0, __p1) __extension__ ({ \
4805   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4809 #define vdupq_lane_u16(__p0, __p1) __extension__ ({ \
4813   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4820 #define vdupq_lane_s8(__p0, __p1) __extension__ ({ \
4823   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4827 #define vdupq_lane_s8(__p0, __p1) __extension__ ({ \
4831   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4838 #define vdupq_lane_f32(__p0, __p1) __extension__ ({ \
4841   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
4845 #define vdupq_lane_f32(__p0, __p1) __extension__ ({ \
4849   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
4856 #define vdupq_lane_s32(__p0, __p1) __extension__ ({ \
4859   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
4863 #define vdupq_lane_s32(__p0, __p1) __extension__ ({ \
4867   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
4874 #define vdupq_lane_s64(__p0, __p1) __extension__ ({ \
4877   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
4881 #define vdupq_lane_s64(__p0, __p1) __extension__ ({ \
4884   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
4891 #define vdupq_lane_s16(__p0, __p1) __extension__ ({ \
4894   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4898 #define vdupq_lane_s16(__p0, __p1) __extension__ ({ \
4902   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4909 #define vdup_lane_u8(__p0, __p1) __extension__ ({ \
4912   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4916 #define vdup_lane_u8(__p0, __p1) __extension__ ({ \
4920   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4927 #define vdup_lane_u32(__p0, __p1) __extension__ ({ \
4930   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
4934 #define vdup_lane_u32(__p0, __p1) __extension__ ({ \
4938   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
4945 #define vdup_lane_u64(__p0, __p1) __extension__ ({ \
4948   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
4952 #define vdup_lane_u64(__p0, __p1) __extension__ ({ \
4955   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
4961 #define vdup_lane_u16(__p0, __p1) __extension__ ({ \
4964   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
4968 #define vdup_lane_u16(__p0, __p1) __extension__ ({ \
4972   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
4979 #define vdup_lane_s8(__p0, __p1) __extension__ ({ \
4982   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4986 #define vdup_lane_s8(__p0, __p1) __extension__ ({ \
4990   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
4997 #define vdup_lane_f32(__p0, __p1) __extension__ ({ \
5000   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
5004 #define vdup_lane_f32(__p0, __p1) __extension__ ({ \
5008   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
5015 #define vdup_lane_s32(__p0, __p1) __extension__ ({ \
5018   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
5022 #define vdup_lane_s32(__p0, __p1) __extension__ ({ \
5026   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
5033 #define vdup_lane_s64(__p0, __p1) __extension__ ({ \
5036   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
5040 #define vdup_lane_s64(__p0, __p1) __extension__ ({ \
5043   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
5049 #define vdup_lane_s16(__p0, __p1) __extension__ ({ \
5052   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
5056 #define vdup_lane_s16(__p0, __p1) __extension__ ({ \
5060   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
5429 __ai uint8x16_t veorq_u8(uint8x16_t __p0, uint8x16_t __p1) {
5431   __ret = __p0 ^ __p1;
5435 __ai uint8x16_t veorq_u8(uint8x16_t __p0, uint8x16_t __p1) {
5437   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5446 __ai uint32x4_t veorq_u32(uint32x4_t __p0, uint32x4_t __p1) {
5448   __ret = __p0 ^ __p1;
5452 __ai uint32x4_t veorq_u32(uint32x4_t __p0, uint32x4_t __p1) {
5454   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5463 __ai uint64x2_t veorq_u64(uint64x2_t __p0, uint64x2_t __p1) {
5465   __ret = __p0 ^ __p1;
5469 __ai uint64x2_t veorq_u64(uint64x2_t __p0, uint64x2_t __p1) {
5471   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5480 __ai uint16x8_t veorq_u16(uint16x8_t __p0, uint16x8_t __p1) {
5482   __ret = __p0 ^ __p1;
5486 __ai uint16x8_t veorq_u16(uint16x8_t __p0, uint16x8_t __p1) {
5488   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
5497 __ai int8x16_t veorq_s8(int8x16_t __p0, int8x16_t __p1) {
5499   __ret = __p0 ^ __p1;
5503 __ai int8x16_t veorq_s8(int8x16_t __p0, int8x16_t __p1) {
5505   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
5514 __ai int32x4_t veorq_s32(int32x4_t __p0, int32x4_t __p1) {
5516   __ret = __p0 ^ __p1;
5520 __ai int32x4_t veorq_s32(int32x4_t __p0, int32x4_t __p1) {
5522   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5531 __ai int64x2_t veorq_s64(int64x2_t __p0, int64x2_t __p1) {
5533   __ret = __p0 ^ __p1;
5537 __ai int64x2_t veorq_s64(int64x2_t __p0, int64x2_t __p1) {
5539   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5548 __ai int16x8_t veorq_s16(int16x8_t __p0, int16x8_t __p1) {
5550   __ret = __p0 ^ __p1;
5554 __ai int16x8_t veorq_s16(int16x8_t __p0, int16x8_t __p1) {
5556   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
5565 __ai uint8x8_t veor_u8(uint8x8_t __p0, uint8x8_t __p1) {
5567   __ret = __p0 ^ __p1;
5571 __ai uint8x8_t veor_u8(uint8x8_t __p0, uint8x8_t __p1) {
5573   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
5582 __ai uint32x2_t veor_u32(uint32x2_t __p0, uint32x2_t __p1) {
5584   __ret = __p0 ^ __p1;
5588 __ai uint32x2_t veor_u32(uint32x2_t __p0, uint32x2_t __p1) {
5590   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5599 __ai uint64x1_t veor_u64(uint64x1_t __p0, uint64x1_t __p1) {
5601   __ret = __p0 ^ __p1;
5605 __ai uint64x1_t veor_u64(uint64x1_t __p0, uint64x1_t __p1) {
5607   __ret = __p0 ^ __p1;
5613 __ai uint16x4_t veor_u16(uint16x4_t __p0, uint16x4_t __p1) {
5615   __ret = __p0 ^ __p1;
5619 __ai uint16x4_t veor_u16(uint16x4_t __p0, uint16x4_t __p1) {
5621   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5630 __ai int8x8_t veor_s8(int8x8_t __p0, int8x8_t __p1) {
5632   __ret = __p0 ^ __p1;
5636 __ai int8x8_t veor_s8(int8x8_t __p0, int8x8_t __p1) {
5638   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
5647 __ai int32x2_t veor_s32(int32x2_t __p0, int32x2_t __p1) {
5649   __ret = __p0 ^ __p1;
5653 __ai int32x2_t veor_s32(int32x2_t __p0, int32x2_t __p1) {
5655   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
5664 __ai int64x1_t veor_s64(int64x1_t __p0, int64x1_t __p1) {
5666   __ret = __p0 ^ __p1;
5670 __ai int64x1_t veor_s64(int64x1_t __p0, int64x1_t __p1) {
5672   __ret = __p0 ^ __p1;
5678 __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
5680   __ret = __p0 ^ __p1;
5684 __ai int16x4_t veor_s16(int16x4_t __p0, int16x4_t __p1) {
5686   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
5695 #define vext_p8(__p0, __p1, __p2) __extension__ ({ \
5697   poly8x8_t __s1 = __p1; \
5703 #define vext_p8(__p0, __p1, __p2) __extension__ ({ \
5705   poly8x8_t __s1 = __p1; \
5716 #define vext_p16(__p0, __p1, __p2) __extension__ ({ \
5718   poly16x4_t __s1 = __p1; \
5724 #define vext_p16(__p0, __p1, __p2) __extension__ ({ \
5726   poly16x4_t __s1 = __p1; \
5737 #define vextq_p8(__p0, __p1, __p2) __extension__ ({ \
5739   poly8x16_t __s1 = __p1; \
5745 #define vextq_p8(__p0, __p1, __p2) __extension__ ({ \
5747   poly8x16_t __s1 = __p1; \
5758 #define vextq_p16(__p0, __p1, __p2) __extension__ ({ \
5760   poly16x8_t __s1 = __p1; \
5766 #define vextq_p16(__p0, __p1, __p2) __extension__ ({ \
5768   poly16x8_t __s1 = __p1; \
5779 #define vextq_u8(__p0, __p1, __p2) __extension__ ({ \
5781   uint8x16_t __s1 = __p1; \
5787 #define vextq_u8(__p0, __p1, __p2) __extension__ ({ \
5789   uint8x16_t __s1 = __p1; \
5800 #define vextq_u32(__p0, __p1, __p2) __extension__ ({ \
5802   uint32x4_t __s1 = __p1; \
5808 #define vextq_u32(__p0, __p1, __p2) __extension__ ({ \
5810   uint32x4_t __s1 = __p1; \
5821 #define vextq_u64(__p0, __p1, __p2) __extension__ ({ \
5823   uint64x2_t __s1 = __p1; \
5829 #define vextq_u64(__p0, __p1, __p2) __extension__ ({ \
5831   uint64x2_t __s1 = __p1; \
5842 #define vextq_u16(__p0, __p1, __p2) __extension__ ({ \
5844   uint16x8_t __s1 = __p1; \
5850 #define vextq_u16(__p0, __p1, __p2) __extension__ ({ \
5852   uint16x8_t __s1 = __p1; \
5863 #define vextq_s8(__p0, __p1, __p2) __extension__ ({ \
5865   int8x16_t __s1 = __p1; \
5871 #define vextq_s8(__p0, __p1, __p2) __extension__ ({ \
5873   int8x16_t __s1 = __p1; \
5884 #define vextq_f32(__p0, __p1, __p2) __extension__ ({ \
5886   float32x4_t __s1 = __p1; \
5892 #define vextq_f32(__p0, __p1, __p2) __extension__ ({ \
5894   float32x4_t __s1 = __p1; \
5905 #define vextq_s32(__p0, __p1, __p2) __extension__ ({ \
5907   int32x4_t __s1 = __p1; \
5913 #define vextq_s32(__p0, __p1, __p2) __extension__ ({ \
5915   int32x4_t __s1 = __p1; \
5926 #define vextq_s64(__p0, __p1, __p2) __extension__ ({ \
5928   int64x2_t __s1 = __p1; \
5934 #define vextq_s64(__p0, __p1, __p2) __extension__ ({ \
5936   int64x2_t __s1 = __p1; \
5947 #define vextq_s16(__p0, __p1, __p2) __extension__ ({ \
5949   int16x8_t __s1 = __p1; \
5955 #define vextq_s16(__p0, __p1, __p2) __extension__ ({ \
5957   int16x8_t __s1 = __p1; \
5968 #define vext_u8(__p0, __p1, __p2) __extension__ ({ \
5970   uint8x8_t __s1 = __p1; \
5976 #define vext_u8(__p0, __p1, __p2) __extension__ ({ \
5978   uint8x8_t __s1 = __p1; \
5989 #define vext_u32(__p0, __p1, __p2) __extension__ ({ \
5991   uint32x2_t __s1 = __p1; \
5997 #define vext_u32(__p0, __p1, __p2) __extension__ ({ \
5999   uint32x2_t __s1 = __p1; \
6010 #define vext_u64(__p0, __p1, __p2) __extension__ ({ \
6012   uint64x1_t __s1 = __p1; \
6018 #define vext_u64(__p0, __p1, __p2) __extension__ ({ \
6020   uint64x1_t __s1 = __p1; \
6028 #define vext_u16(__p0, __p1, __p2) __extension__ ({ \
6030   uint16x4_t __s1 = __p1; \
6036 #define vext_u16(__p0, __p1, __p2) __extension__ ({ \
6038   uint16x4_t __s1 = __p1; \
6049 #define vext_s8(__p0, __p1, __p2) __extension__ ({ \
6051   int8x8_t __s1 = __p1; \
6057 #define vext_s8(__p0, __p1, __p2) __extension__ ({ \
6059   int8x8_t __s1 = __p1; \
6070 #define vext_f32(__p0, __p1, __p2) __extension__ ({ \
6072   float32x2_t __s1 = __p1; \
6078 #define vext_f32(__p0, __p1, __p2) __extension__ ({ \
6080   float32x2_t __s1 = __p1; \
6091 #define vext_s32(__p0, __p1, __p2) __extension__ ({ \
6093   int32x2_t __s1 = __p1; \
6099 #define vext_s32(__p0, __p1, __p2) __extension__ ({ \
6101   int32x2_t __s1 = __p1; \
6112 #define vext_s64(__p0, __p1, __p2) __extension__ ({ \
6114   int64x1_t __s1 = __p1; \
6120 #define vext_s64(__p0, __p1, __p2) __extension__ ({ \
6122   int64x1_t __s1 = __p1; \
6130 #define vext_s16(__p0, __p1, __p2) __extension__ ({ \
6132   int16x4_t __s1 = __p1; \
6138 #define vext_s16(__p0, __p1, __p2) __extension__ ({ \
6140   int16x4_t __s1 = __p1; \
6151 __ai float32x4_t vfmaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
6153   __ret = (float32x4_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
6157 __ai float32x4_t vfmaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
6159   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
6166 __ai float32x4_t __noswap_vfmaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
6168   __ret = (float32x4_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
6174 __ai float32x2_t vfma_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
6176   __ret = (float32x2_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
6180 __ai float32x2_t vfma_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
6182   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
6189 __ai float32x2_t __noswap_vfma_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
6191   __ret = (float32x2_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
6432 #define vget_lane_p8(__p0, __p1) __extension__ ({ \
6435   __ret = (poly8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
6439 #define vget_lane_p8(__p0, __p1) __extension__ ({ \
6443   __ret = (poly8_t) __builtin_neon_vget_lane_i8((int8x8_t)__rev0, __p1); \
6446 #define __noswap_vget_lane_p8(__p0, __p1) __extension__ ({ \
6449   __ret = (poly8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
6455 #define vget_lane_p16(__p0, __p1) __extension__ ({ \
6458   __ret = (poly16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
6462 #define vget_lane_p16(__p0, __p1) __extension__ ({ \
6466   __ret = (poly16_t) __builtin_neon_vget_lane_i16((int8x8_t)__rev0, __p1); \
6469 #define __noswap_vget_lane_p16(__p0, __p1) __extension__ ({ \
6472   __ret = (poly16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
6478 #define vgetq_lane_p8(__p0, __p1) __extension__ ({ \
6481   __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
6485 #define vgetq_lane_p8(__p0, __p1) __extension__ ({ \
6489   __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__rev0, __p1); \
6492 #define __noswap_vgetq_lane_p8(__p0, __p1) __extension__ ({ \
6495   __ret = (poly8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
6501 #define vgetq_lane_p16(__p0, __p1) __extension__ ({ \
6504   __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
6508 #define vgetq_lane_p16(__p0, __p1) __extension__ ({ \
6512   __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__rev0, __p1); \
6515 #define __noswap_vgetq_lane_p16(__p0, __p1) __extension__ ({ \
6518   __ret = (poly16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
6524 #define vgetq_lane_u8(__p0, __p1) __extension__ ({ \
6527   __ret = (uint8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
6531 #define vgetq_lane_u8(__p0, __p1) __extension__ ({ \
6535   __ret = (uint8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__rev0, __p1); \
6538 #define __noswap_vgetq_lane_u8(__p0, __p1) __extension__ ({ \
6541   __ret = (uint8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
6547 #define vgetq_lane_u32(__p0, __p1) __extension__ ({ \
6550   __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__s0, __p1); \
6554 #define vgetq_lane_u32(__p0, __p1) __extension__ ({ \
6558   __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__rev0, __p1); \
6561 #define __noswap_vgetq_lane_u32(__p0, __p1) __extension__ ({ \
6564   __ret = (uint32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__s0, __p1); \
6570 #define vgetq_lane_u64(__p0, __p1) __extension__ ({ \
6573   __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
6577 #define vgetq_lane_u64(__p0, __p1) __extension__ ({ \
6581   __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__rev0, __p1); \
6584 #define __noswap_vgetq_lane_u64(__p0, __p1) __extension__ ({ \
6587   __ret = (uint64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
6593 #define vgetq_lane_u16(__p0, __p1) __extension__ ({ \
6596   __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
6600 #define vgetq_lane_u16(__p0, __p1) __extension__ ({ \
6604   __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__rev0, __p1); \
6607 #define __noswap_vgetq_lane_u16(__p0, __p1) __extension__ ({ \
6610   __ret = (uint16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
6616 #define vgetq_lane_s8(__p0, __p1) __extension__ ({ \
6619   __ret = (int8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
6623 #define vgetq_lane_s8(__p0, __p1) __extension__ ({ \
6627   __ret = (int8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__rev0, __p1); \
6630 #define __noswap_vgetq_lane_s8(__p0, __p1) __extension__ ({ \
6633   __ret = (int8_t) __builtin_neon_vgetq_lane_i8((int8x16_t)__s0, __p1); \
6639 #define vgetq_lane_f32(__p0, __p1) __extension__ ({ \
6642   __ret = (float32_t) __builtin_neon_vgetq_lane_f32((int8x16_t)__s0, __p1); \
6646 #define vgetq_lane_f32(__p0, __p1) __extension__ ({ \
6650   __ret = (float32_t) __builtin_neon_vgetq_lane_f32((int8x16_t)__rev0, __p1); \
6653 #define __noswap_vgetq_lane_f32(__p0, __p1) __extension__ ({ \
6656   __ret = (float32_t) __builtin_neon_vgetq_lane_f32((int8x16_t)__s0, __p1); \
6662 #define vgetq_lane_s32(__p0, __p1) __extension__ ({ \
6665   __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__s0, __p1); \
6669 #define vgetq_lane_s32(__p0, __p1) __extension__ ({ \
6673   __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__rev0, __p1); \
6676 #define __noswap_vgetq_lane_s32(__p0, __p1) __extension__ ({ \
6679   __ret = (int32_t) __builtin_neon_vgetq_lane_i32((int8x16_t)__s0, __p1); \
6685 #define vgetq_lane_s64(__p0, __p1) __extension__ ({ \
6688   __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
6692 #define vgetq_lane_s64(__p0, __p1) __extension__ ({ \
6696   __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__rev0, __p1); \
6699 #define __noswap_vgetq_lane_s64(__p0, __p1) __extension__ ({ \
6702   __ret = (int64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
6708 #define vgetq_lane_s16(__p0, __p1) __extension__ ({ \
6711   __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
6715 #define vgetq_lane_s16(__p0, __p1) __extension__ ({ \
6719   __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__rev0, __p1); \
6722 #define __noswap_vgetq_lane_s16(__p0, __p1) __extension__ ({ \
6725   __ret = (int16_t) __builtin_neon_vgetq_lane_i16((int8x16_t)__s0, __p1); \
6731 #define vget_lane_u8(__p0, __p1) __extension__ ({ \
6734   __ret = (uint8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
6738 #define vget_lane_u8(__p0, __p1) __extension__ ({ \
6742   __ret = (uint8_t) __builtin_neon_vget_lane_i8((int8x8_t)__rev0, __p1); \
6745 #define __noswap_vget_lane_u8(__p0, __p1) __extension__ ({ \
6748   __ret = (uint8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
6754 #define vget_lane_u32(__p0, __p1) __extension__ ({ \
6757   __ret = (uint32_t) __builtin_neon_vget_lane_i32((int8x8_t)__s0, __p1); \
6761 #define vget_lane_u32(__p0, __p1) __extension__ ({ \
6765   __ret = (uint32_t) __builtin_neon_vget_lane_i32((int8x8_t)__rev0, __p1); \
6768 #define __noswap_vget_lane_u32(__p0, __p1) __extension__ ({ \
6771   __ret = (uint32_t) __builtin_neon_vget_lane_i32((int8x8_t)__s0, __p1); \
6777 #define vget_lane_u64(__p0, __p1) __extension__ ({ \
6780   __ret = (uint64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
6784 #define vget_lane_u64(__p0, __p1) __extension__ ({ \
6787   __ret = (uint64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
6790 #define __noswap_vget_lane_u64(__p0, __p1) __extension__ ({ \
6793   __ret = (uint64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
6799 #define vget_lane_u16(__p0, __p1) __extension__ ({ \
6802   __ret = (uint16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
6806 #define vget_lane_u16(__p0, __p1) __extension__ ({ \
6810   __ret = (uint16_t) __builtin_neon_vget_lane_i16((int8x8_t)__rev0, __p1); \
6813 #define __noswap_vget_lane_u16(__p0, __p1) __extension__ ({ \
6816   __ret = (uint16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
6822 #define vget_lane_s8(__p0, __p1) __extension__ ({ \
6825   __ret = (int8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
6829 #define vget_lane_s8(__p0, __p1) __extension__ ({ \
6833   __ret = (int8_t) __builtin_neon_vget_lane_i8((int8x8_t)__rev0, __p1); \
6836 #define __noswap_vget_lane_s8(__p0, __p1) __extension__ ({ \
6839   __ret = (int8_t) __builtin_neon_vget_lane_i8((int8x8_t)__s0, __p1); \
6845 #define vget_lane_f32(__p0, __p1) __extension__ ({ \
6848   __ret = (float32_t) __builtin_neon_vget_lane_f32((int8x8_t)__s0, __p1); \
6852 #define vget_lane_f32(__p0, __p1) __extension__ ({ \
6856   __ret = (float32_t) __builtin_neon_vget_lane_f32((int8x8_t)__rev0, __p1); \
6859 #define __noswap_vget_lane_f32(__p0, __p1) __extension__ ({ \
6862   __ret = (float32_t) __builtin_neon_vget_lane_f32((int8x8_t)__s0, __p1); \
6868 #define vget_lane_s32(__p0, __p1) __extension__ ({ \
6871   __ret = (int32_t) __builtin_neon_vget_lane_i32((int8x8_t)__s0, __p1); \
6875 #define vget_lane_s32(__p0, __p1) __extension__ ({ \
6879   __ret = (int32_t) __builtin_neon_vget_lane_i32((int8x8_t)__rev0, __p1); \
6882 #define __noswap_vget_lane_s32(__p0, __p1) __extension__ ({ \
6885   __ret = (int32_t) __builtin_neon_vget_lane_i32((int8x8_t)__s0, __p1); \
6891 #define vget_lane_s64(__p0, __p1) __extension__ ({ \
6894   __ret = (int64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
6898 #define vget_lane_s64(__p0, __p1) __extension__ ({ \
6901   __ret = (int64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
6904 #define __noswap_vget_lane_s64(__p0, __p1) __extension__ ({ \
6907   __ret = (int64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
6913 #define vget_lane_s16(__p0, __p1) __extension__ ({ \
6916   __ret = (int16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
6920 #define vget_lane_s16(__p0, __p1) __extension__ ({ \
6924   __ret = (int16_t) __builtin_neon_vget_lane_i16((int8x8_t)__rev0, __p1); \
6927 #define __noswap_vget_lane_s16(__p0, __p1) __extension__ ({ \
6930   __ret = (int16_t) __builtin_neon_vget_lane_i16((int8x8_t)__s0, __p1); \
7126 __ai uint8x16_t vhaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
7128   __ret = (uint8x16_t) __builtin_neon_vhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
7132 __ai uint8x16_t vhaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
7134   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
7143 __ai uint32x4_t vhaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
7145   __ret = (uint32x4_t) __builtin_neon_vhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
7149 __ai uint32x4_t vhaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
7151   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
7160 __ai uint16x8_t vhaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
7162   __ret = (uint16x8_t) __builtin_neon_vhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
7166 __ai uint16x8_t vhaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
7168   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
7177 __ai int8x16_t vhaddq_s8(int8x16_t __p0, int8x16_t __p1) {
7179   __ret = (int8x16_t) __builtin_neon_vhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
7183 __ai int8x16_t vhaddq_s8(int8x16_t __p0, int8x16_t __p1) {
7185   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
7194 __ai int32x4_t vhaddq_s32(int32x4_t __p0, int32x4_t __p1) {
7196   __ret = (int32x4_t) __builtin_neon_vhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
7200 __ai int32x4_t vhaddq_s32(int32x4_t __p0, int32x4_t __p1) {
7202   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
7211 __ai int16x8_t vhaddq_s16(int16x8_t __p0, int16x8_t __p1) {
7213   __ret = (int16x8_t) __builtin_neon_vhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
7217 __ai int16x8_t vhaddq_s16(int16x8_t __p0, int16x8_t __p1) {
7219   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
7228 __ai uint8x8_t vhadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
7230   __ret = (uint8x8_t) __builtin_neon_vhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
7234 __ai uint8x8_t vhadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
7236   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
7245 __ai uint32x2_t vhadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
7247   __ret = (uint32x2_t) __builtin_neon_vhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
7251 __ai uint32x2_t vhadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
7253   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
7262 __ai uint16x4_t vhadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
7264   __ret = (uint16x4_t) __builtin_neon_vhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
7268 __ai uint16x4_t vhadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
7270   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
7279 __ai int8x8_t vhadd_s8(int8x8_t __p0, int8x8_t __p1) {
7281   __ret = (int8x8_t) __builtin_neon_vhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
7285 __ai int8x8_t vhadd_s8(int8x8_t __p0, int8x8_t __p1) {
7287   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
7296 __ai int32x2_t vhadd_s32(int32x2_t __p0, int32x2_t __p1) {
7298   __ret = (int32x2_t) __builtin_neon_vhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
7302 __ai int32x2_t vhadd_s32(int32x2_t __p0, int32x2_t __p1) {
7304   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
7313 __ai int16x4_t vhadd_s16(int16x4_t __p0, int16x4_t __p1) {
7315   __ret = (int16x4_t) __builtin_neon_vhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
7319 __ai int16x4_t vhadd_s16(int16x4_t __p0, int16x4_t __p1) {
7321   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
7330 __ai uint8x16_t vhsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
7332   __ret = (uint8x16_t) __builtin_neon_vhsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
7336 __ai uint8x16_t vhsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
7338   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
7347 __ai uint32x4_t vhsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
7349   __ret = (uint32x4_t) __builtin_neon_vhsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
7353 __ai uint32x4_t vhsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
7355   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
7364 __ai uint16x8_t vhsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
7366   __ret = (uint16x8_t) __builtin_neon_vhsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
7370 __ai uint16x8_t vhsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
7372   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
7381 __ai int8x16_t vhsubq_s8(int8x16_t __p0, int8x16_t __p1) {
7383   __ret = (int8x16_t) __builtin_neon_vhsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
7387 __ai int8x16_t vhsubq_s8(int8x16_t __p0, int8x16_t __p1) {
7389   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
7398 __ai int32x4_t vhsubq_s32(int32x4_t __p0, int32x4_t __p1) {
7400   __ret = (int32x4_t) __builtin_neon_vhsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
7404 __ai int32x4_t vhsubq_s32(int32x4_t __p0, int32x4_t __p1) {
7406   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
7415 __ai int16x8_t vhsubq_s16(int16x8_t __p0, int16x8_t __p1) {
7417   __ret = (int16x8_t) __builtin_neon_vhsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
7421 __ai int16x8_t vhsubq_s16(int16x8_t __p0, int16x8_t __p1) {
7423   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
7432 __ai uint8x8_t vhsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
7434   __ret = (uint8x8_t) __builtin_neon_vhsub_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
7438 __ai uint8x8_t vhsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
7440   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
7449 __ai uint32x2_t vhsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
7451   __ret = (uint32x2_t) __builtin_neon_vhsub_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
7455 __ai uint32x2_t vhsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
7457   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
7466 __ai uint16x4_t vhsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
7468   __ret = (uint16x4_t) __builtin_neon_vhsub_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
7472 __ai uint16x4_t vhsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
7474   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
7483 __ai int8x8_t vhsub_s8(int8x8_t __p0, int8x8_t __p1) {
7485   __ret = (int8x8_t) __builtin_neon_vhsub_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
7489 __ai int8x8_t vhsub_s8(int8x8_t __p0, int8x8_t __p1) {
7491   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
7500 __ai int32x2_t vhsub_s32(int32x2_t __p0, int32x2_t __p1) {
7502   __ret = (int32x2_t) __builtin_neon_vhsub_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
7506 __ai int32x2_t vhsub_s32(int32x2_t __p0, int32x2_t __p1) {
7508   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
7517 __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
7519   __ret = (int16x4_t) __builtin_neon_vhsub_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
7523 __ai int16x4_t vhsub_s16(int16x4_t __p0, int16x4_t __p1) {
7525   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
8250 #define vld1_lane_p8(__p0, __p1, __p2) __extension__ ({ \
8251   poly8x8_t __s1 = __p1; \
8257 #define vld1_lane_p8(__p0, __p1, __p2) __extension__ ({ \
8258   poly8x8_t __s1 = __p1; \
8268 #define vld1_lane_p16(__p0, __p1, __p2) __extension__ ({ \
8269   poly16x4_t __s1 = __p1; \
8275 #define vld1_lane_p16(__p0, __p1, __p2) __extension__ ({ \
8276   poly16x4_t __s1 = __p1; \
8286 #define vld1q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
8287   poly8x16_t __s1 = __p1; \
8293 #define vld1q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
8294   poly8x16_t __s1 = __p1; \
8304 #define vld1q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
8305   poly16x8_t __s1 = __p1; \
8311 #define vld1q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
8312   poly16x8_t __s1 = __p1; \
8322 #define vld1q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
8323   uint8x16_t __s1 = __p1; \
8329 #define vld1q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
8330   uint8x16_t __s1 = __p1; \
8340 #define vld1q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
8341   uint32x4_t __s1 = __p1; \
8347 #define vld1q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
8348   uint32x4_t __s1 = __p1; \
8358 #define vld1q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
8359   uint64x2_t __s1 = __p1; \
8365 #define vld1q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
8366   uint64x2_t __s1 = __p1; \
8376 #define vld1q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
8377   uint16x8_t __s1 = __p1; \
8383 #define vld1q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
8384   uint16x8_t __s1 = __p1; \
8394 #define vld1q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
8395   int8x16_t __s1 = __p1; \
8401 #define vld1q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
8402   int8x16_t __s1 = __p1; \
8412 #define vld1q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
8413   float32x4_t __s1 = __p1; \
8419 #define vld1q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
8420   float32x4_t __s1 = __p1; \
8430 #define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
8431   float16x8_t __s1 = __p1; \
8437 #define vld1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
8438   float16x8_t __s1 = __p1; \
8448 #define vld1q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
8449   int32x4_t __s1 = __p1; \
8455 #define vld1q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
8456   int32x4_t __s1 = __p1; \
8466 #define vld1q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
8467   int64x2_t __s1 = __p1; \
8473 #define vld1q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
8474   int64x2_t __s1 = __p1; \
8484 #define vld1q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
8485   int16x8_t __s1 = __p1; \
8491 #define vld1q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
8492   int16x8_t __s1 = __p1; \
8502 #define vld1_lane_u8(__p0, __p1, __p2) __extension__ ({ \
8503   uint8x8_t __s1 = __p1; \
8509 #define vld1_lane_u8(__p0, __p1, __p2) __extension__ ({ \
8510   uint8x8_t __s1 = __p1; \
8520 #define vld1_lane_u32(__p0, __p1, __p2) __extension__ ({ \
8521   uint32x2_t __s1 = __p1; \
8527 #define vld1_lane_u32(__p0, __p1, __p2) __extension__ ({ \
8528   uint32x2_t __s1 = __p1; \
8538 #define vld1_lane_u64(__p0, __p1, __p2) __extension__ ({ \
8539   uint64x1_t __s1 = __p1; \
8545 #define vld1_lane_u64(__p0, __p1, __p2) __extension__ ({ \
8546   uint64x1_t __s1 = __p1; \
8554 #define vld1_lane_u16(__p0, __p1, __p2) __extension__ ({ \
8555   uint16x4_t __s1 = __p1; \
8561 #define vld1_lane_u16(__p0, __p1, __p2) __extension__ ({ \
8562   uint16x4_t __s1 = __p1; \
8572 #define vld1_lane_s8(__p0, __p1, __p2) __extension__ ({ \
8573   int8x8_t __s1 = __p1; \
8579 #define vld1_lane_s8(__p0, __p1, __p2) __extension__ ({ \
8580   int8x8_t __s1 = __p1; \
8590 #define vld1_lane_f32(__p0, __p1, __p2) __extension__ ({ \
8591   float32x2_t __s1 = __p1; \
8597 #define vld1_lane_f32(__p0, __p1, __p2) __extension__ ({ \
8598   float32x2_t __s1 = __p1; \
8608 #define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
8609   float16x4_t __s1 = __p1; \
8615 #define vld1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
8616   float16x4_t __s1 = __p1; \
8626 #define vld1_lane_s32(__p0, __p1, __p2) __extension__ ({ \
8627   int32x2_t __s1 = __p1; \
8633 #define vld1_lane_s32(__p0, __p1, __p2) __extension__ ({ \
8634   int32x2_t __s1 = __p1; \
8644 #define vld1_lane_s64(__p0, __p1, __p2) __extension__ ({ \
8645   int64x1_t __s1 = __p1; \
8651 #define vld1_lane_s64(__p0, __p1, __p2) __extension__ ({ \
8652   int64x1_t __s1 = __p1; \
8660 #define vld1_lane_s16(__p0, __p1, __p2) __extension__ ({ \
8661   int16x4_t __s1 = __p1; \
8667 #define vld1_lane_s16(__p0, __p1, __p2) __extension__ ({ \
8668   int16x4_t __s1 = __p1; \
9244 #define vld2_lane_p8(__p0, __p1, __p2) __extension__ ({ \
9245   poly8x8x2_t __s1 = __p1; \
9251 #define vld2_lane_p8(__p0, __p1, __p2) __extension__ ({ \
9252   poly8x8x2_t __s1 = __p1; \
9266 #define vld2_lane_p16(__p0, __p1, __p2) __extension__ ({ \
9267   poly16x4x2_t __s1 = __p1; \
9273 #define vld2_lane_p16(__p0, __p1, __p2) __extension__ ({ \
9274   poly16x4x2_t __s1 = __p1; \
9288 #define vld2q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
9289   poly16x8x2_t __s1 = __p1; \
9295 #define vld2q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
9296   poly16x8x2_t __s1 = __p1; \
9310 #define vld2q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
9311   uint32x4x2_t __s1 = __p1; \
9317 #define vld2q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
9318   uint32x4x2_t __s1 = __p1; \
9332 #define vld2q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
9333   uint16x8x2_t __s1 = __p1; \
9339 #define vld2q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
9340   uint16x8x2_t __s1 = __p1; \
9354 #define vld2q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
9355   float32x4x2_t __s1 = __p1; \
9361 #define vld2q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
9362   float32x4x2_t __s1 = __p1; \
9376 #define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
9377   float16x8x2_t __s1 = __p1; \
9383 #define vld2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
9384   float16x8x2_t __s1 = __p1; \
9398 #define vld2q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
9399   int32x4x2_t __s1 = __p1; \
9405 #define vld2q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
9406   int32x4x2_t __s1 = __p1; \
9420 #define vld2q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
9421   int16x8x2_t __s1 = __p1; \
9427 #define vld2q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
9428   int16x8x2_t __s1 = __p1; \
9442 #define vld2_lane_u8(__p0, __p1, __p2) __extension__ ({ \
9443   uint8x8x2_t __s1 = __p1; \
9449 #define vld2_lane_u8(__p0, __p1, __p2) __extension__ ({ \
9450   uint8x8x2_t __s1 = __p1; \
9464 #define vld2_lane_u32(__p0, __p1, __p2) __extension__ ({ \
9465   uint32x2x2_t __s1 = __p1; \
9471 #define vld2_lane_u32(__p0, __p1, __p2) __extension__ ({ \
9472   uint32x2x2_t __s1 = __p1; \
9486 #define vld2_lane_u16(__p0, __p1, __p2) __extension__ ({ \
9487   uint16x4x2_t __s1 = __p1; \
9493 #define vld2_lane_u16(__p0, __p1, __p2) __extension__ ({ \
9494   uint16x4x2_t __s1 = __p1; \
9508 #define vld2_lane_s8(__p0, __p1, __p2) __extension__ ({ \
9509   int8x8x2_t __s1 = __p1; \
9515 #define vld2_lane_s8(__p0, __p1, __p2) __extension__ ({ \
9516   int8x8x2_t __s1 = __p1; \
9530 #define vld2_lane_f32(__p0, __p1, __p2) __extension__ ({ \
9531   float32x2x2_t __s1 = __p1; \
9537 #define vld2_lane_f32(__p0, __p1, __p2) __extension__ ({ \
9538   float32x2x2_t __s1 = __p1; \
9552 #define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
9553   float16x4x2_t __s1 = __p1; \
9559 #define vld2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
9560   float16x4x2_t __s1 = __p1; \
9574 #define vld2_lane_s32(__p0, __p1, __p2) __extension__ ({ \
9575   int32x2x2_t __s1 = __p1; \
9581 #define vld2_lane_s32(__p0, __p1, __p2) __extension__ ({ \
9582   int32x2x2_t __s1 = __p1; \
9596 #define vld2_lane_s16(__p0, __p1, __p2) __extension__ ({ \
9597   int16x4x2_t __s1 = __p1; \
9603 #define vld2_lane_s16(__p0, __p1, __p2) __extension__ ({ \
9604   int16x4x2_t __s1 = __p1; \
10214 #define vld3_lane_p8(__p0, __p1, __p2) __extension__ ({ \
10215   poly8x8x3_t __s1 = __p1; \
10221 #define vld3_lane_p8(__p0, __p1, __p2) __extension__ ({ \
10222   poly8x8x3_t __s1 = __p1; \
10238 #define vld3_lane_p16(__p0, __p1, __p2) __extension__ ({ \
10239   poly16x4x3_t __s1 = __p1; \
10245 #define vld3_lane_p16(__p0, __p1, __p2) __extension__ ({ \
10246   poly16x4x3_t __s1 = __p1; \
10262 #define vld3q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
10263   poly16x8x3_t __s1 = __p1; \
10269 #define vld3q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
10270   poly16x8x3_t __s1 = __p1; \
10286 #define vld3q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
10287   uint32x4x3_t __s1 = __p1; \
10293 #define vld3q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
10294   uint32x4x3_t __s1 = __p1; \
10310 #define vld3q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
10311   uint16x8x3_t __s1 = __p1; \
10317 #define vld3q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
10318   uint16x8x3_t __s1 = __p1; \
10334 #define vld3q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
10335   float32x4x3_t __s1 = __p1; \
10341 #define vld3q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
10342   float32x4x3_t __s1 = __p1; \
10358 #define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
10359   float16x8x3_t __s1 = __p1; \
10365 #define vld3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
10366   float16x8x3_t __s1 = __p1; \
10382 #define vld3q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
10383   int32x4x3_t __s1 = __p1; \
10389 #define vld3q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
10390   int32x4x3_t __s1 = __p1; \
10406 #define vld3q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
10407   int16x8x3_t __s1 = __p1; \
10413 #define vld3q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
10414   int16x8x3_t __s1 = __p1; \
10430 #define vld3_lane_u8(__p0, __p1, __p2) __extension__ ({ \
10431   uint8x8x3_t __s1 = __p1; \
10437 #define vld3_lane_u8(__p0, __p1, __p2) __extension__ ({ \
10438   uint8x8x3_t __s1 = __p1; \
10454 #define vld3_lane_u32(__p0, __p1, __p2) __extension__ ({ \
10455   uint32x2x3_t __s1 = __p1; \
10461 #define vld3_lane_u32(__p0, __p1, __p2) __extension__ ({ \
10462   uint32x2x3_t __s1 = __p1; \
10478 #define vld3_lane_u16(__p0, __p1, __p2) __extension__ ({ \
10479   uint16x4x3_t __s1 = __p1; \
10485 #define vld3_lane_u16(__p0, __p1, __p2) __extension__ ({ \
10486   uint16x4x3_t __s1 = __p1; \
10502 #define vld3_lane_s8(__p0, __p1, __p2) __extension__ ({ \
10503   int8x8x3_t __s1 = __p1; \
10509 #define vld3_lane_s8(__p0, __p1, __p2) __extension__ ({ \
10510   int8x8x3_t __s1 = __p1; \
10526 #define vld3_lane_f32(__p0, __p1, __p2) __extension__ ({ \
10527   float32x2x3_t __s1 = __p1; \
10533 #define vld3_lane_f32(__p0, __p1, __p2) __extension__ ({ \
10534   float32x2x3_t __s1 = __p1; \
10550 #define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
10551   float16x4x3_t __s1 = __p1; \
10557 #define vld3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
10558   float16x4x3_t __s1 = __p1; \
10574 #define vld3_lane_s32(__p0, __p1, __p2) __extension__ ({ \
10575   int32x2x3_t __s1 = __p1; \
10581 #define vld3_lane_s32(__p0, __p1, __p2) __extension__ ({ \
10582   int32x2x3_t __s1 = __p1; \
10598 #define vld3_lane_s16(__p0, __p1, __p2) __extension__ ({ \
10599   int16x4x3_t __s1 = __p1; \
10605 #define vld3_lane_s16(__p0, __p1, __p2) __extension__ ({ \
10606   int16x4x3_t __s1 = __p1; \
11248 #define vld4_lane_p8(__p0, __p1, __p2) __extension__ ({ \
11249   poly8x8x4_t __s1 = __p1; \
11255 #define vld4_lane_p8(__p0, __p1, __p2) __extension__ ({ \
11256   poly8x8x4_t __s1 = __p1; \
11274 #define vld4_lane_p16(__p0, __p1, __p2) __extension__ ({ \
11275   poly16x4x4_t __s1 = __p1; \
11281 #define vld4_lane_p16(__p0, __p1, __p2) __extension__ ({ \
11282   poly16x4x4_t __s1 = __p1; \
11300 #define vld4q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
11301   poly16x8x4_t __s1 = __p1; \
11307 #define vld4q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
11308   poly16x8x4_t __s1 = __p1; \
11326 #define vld4q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
11327   uint32x4x4_t __s1 = __p1; \
11333 #define vld4q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
11334   uint32x4x4_t __s1 = __p1; \
11352 #define vld4q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
11353   uint16x8x4_t __s1 = __p1; \
11359 #define vld4q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
11360   uint16x8x4_t __s1 = __p1; \
11378 #define vld4q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
11379   float32x4x4_t __s1 = __p1; \
11385 #define vld4q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
11386   float32x4x4_t __s1 = __p1; \
11404 #define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
11405   float16x8x4_t __s1 = __p1; \
11411 #define vld4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
11412   float16x8x4_t __s1 = __p1; \
11430 #define vld4q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
11431   int32x4x4_t __s1 = __p1; \
11437 #define vld4q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
11438   int32x4x4_t __s1 = __p1; \
11456 #define vld4q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
11457   int16x8x4_t __s1 = __p1; \
11463 #define vld4q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
11464   int16x8x4_t __s1 = __p1; \
11482 #define vld4_lane_u8(__p0, __p1, __p2) __extension__ ({ \
11483   uint8x8x4_t __s1 = __p1; \
11489 #define vld4_lane_u8(__p0, __p1, __p2) __extension__ ({ \
11490   uint8x8x4_t __s1 = __p1; \
11508 #define vld4_lane_u32(__p0, __p1, __p2) __extension__ ({ \
11509   uint32x2x4_t __s1 = __p1; \
11515 #define vld4_lane_u32(__p0, __p1, __p2) __extension__ ({ \
11516   uint32x2x4_t __s1 = __p1; \
11534 #define vld4_lane_u16(__p0, __p1, __p2) __extension__ ({ \
11535   uint16x4x4_t __s1 = __p1; \
11541 #define vld4_lane_u16(__p0, __p1, __p2) __extension__ ({ \
11542   uint16x4x4_t __s1 = __p1; \
11560 #define vld4_lane_s8(__p0, __p1, __p2) __extension__ ({ \
11561   int8x8x4_t __s1 = __p1; \
11567 #define vld4_lane_s8(__p0, __p1, __p2) __extension__ ({ \
11568   int8x8x4_t __s1 = __p1; \
11586 #define vld4_lane_f32(__p0, __p1, __p2) __extension__ ({ \
11587   float32x2x4_t __s1 = __p1; \
11593 #define vld4_lane_f32(__p0, __p1, __p2) __extension__ ({ \
11594   float32x2x4_t __s1 = __p1; \
11612 #define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
11613   float16x4x4_t __s1 = __p1; \
11619 #define vld4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
11620   float16x4x4_t __s1 = __p1; \
11638 #define vld4_lane_s32(__p0, __p1, __p2) __extension__ ({ \
11639   int32x2x4_t __s1 = __p1; \
11645 #define vld4_lane_s32(__p0, __p1, __p2) __extension__ ({ \
11646   int32x2x4_t __s1 = __p1; \
11664 #define vld4_lane_s16(__p0, __p1, __p2) __extension__ ({ \
11665   int16x4x4_t __s1 = __p1; \
11671 #define vld4_lane_s16(__p0, __p1, __p2) __extension__ ({ \
11672   int16x4x4_t __s1 = __p1; \
11690 __ai uint8x16_t vmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
11692   __ret = (uint8x16_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
11696 __ai uint8x16_t vmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
11698   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
11707 __ai uint32x4_t vmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
11709   __ret = (uint32x4_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
11713 __ai uint32x4_t vmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
11715   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
11724 __ai uint16x8_t vmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
11726   __ret = (uint16x8_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
11730 __ai uint16x8_t vmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
11732   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
11741 __ai int8x16_t vmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
11743   __ret = (int8x16_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
11747 __ai int8x16_t vmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
11749   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
11758 __ai float32x4_t vmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
11760   __ret = (float32x4_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
11764 __ai float32x4_t vmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
11766   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
11775 __ai int32x4_t vmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
11777   __ret = (int32x4_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
11781 __ai int32x4_t vmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
11783   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
11792 __ai int16x8_t vmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
11794   __ret = (int16x8_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
11798 __ai int16x8_t vmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
11800   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
11809 __ai uint8x8_t vmax_u8(uint8x8_t __p0, uint8x8_t __p1) {
11811   __ret = (uint8x8_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
11815 __ai uint8x8_t vmax_u8(uint8x8_t __p0, uint8x8_t __p1) {
11817   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
11826 __ai uint32x2_t vmax_u32(uint32x2_t __p0, uint32x2_t __p1) {
11828   __ret = (uint32x2_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
11832 __ai uint32x2_t vmax_u32(uint32x2_t __p0, uint32x2_t __p1) {
11834   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
11843 __ai uint16x4_t vmax_u16(uint16x4_t __p0, uint16x4_t __p1) {
11845   __ret = (uint16x4_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
11849 __ai uint16x4_t vmax_u16(uint16x4_t __p0, uint16x4_t __p1) {
11851   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
11860 __ai int8x8_t vmax_s8(int8x8_t __p0, int8x8_t __p1) {
11862   __ret = (int8x8_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
11866 __ai int8x8_t vmax_s8(int8x8_t __p0, int8x8_t __p1) {
11868   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
11877 __ai float32x2_t vmax_f32(float32x2_t __p0, float32x2_t __p1) {
11879   __ret = (float32x2_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
11883 __ai float32x2_t vmax_f32(float32x2_t __p0, float32x2_t __p1) {
11885   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
11894 __ai int32x2_t vmax_s32(int32x2_t __p0, int32x2_t __p1) {
11896   __ret = (int32x2_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
11900 __ai int32x2_t vmax_s32(int32x2_t __p0, int32x2_t __p1) {
11902   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
11911 __ai int16x4_t vmax_s16(int16x4_t __p0, int16x4_t __p1) {
11913   __ret = (int16x4_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
11917 __ai int16x4_t vmax_s16(int16x4_t __p0, int16x4_t __p1) {
11919   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
11928 __ai uint8x16_t vminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
11930   __ret = (uint8x16_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
11934 __ai uint8x16_t vminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
11936   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
11945 __ai uint32x4_t vminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
11947   __ret = (uint32x4_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
11951 __ai uint32x4_t vminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
11953   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
11962 __ai uint16x8_t vminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
11964   __ret = (uint16x8_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
11968 __ai uint16x8_t vminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
11970   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
11979 __ai int8x16_t vminq_s8(int8x16_t __p0, int8x16_t __p1) {
11981   __ret = (int8x16_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
11985 __ai int8x16_t vminq_s8(int8x16_t __p0, int8x16_t __p1) {
11987   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
11996 __ai float32x4_t vminq_f32(float32x4_t __p0, float32x4_t __p1) {
11998   __ret = (float32x4_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
12002 __ai float32x4_t vminq_f32(float32x4_t __p0, float32x4_t __p1) {
12004   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12013 __ai int32x4_t vminq_s32(int32x4_t __p0, int32x4_t __p1) {
12015   __ret = (int32x4_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
12019 __ai int32x4_t vminq_s32(int32x4_t __p0, int32x4_t __p1) {
12021   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12030 __ai int16x8_t vminq_s16(int16x8_t __p0, int16x8_t __p1) {
12032   __ret = (int16x8_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
12036 __ai int16x8_t vminq_s16(int16x8_t __p0, int16x8_t __p1) {
12038   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12047 __ai uint8x8_t vmin_u8(uint8x8_t __p0, uint8x8_t __p1) {
12049   __ret = (uint8x8_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
12053 __ai uint8x8_t vmin_u8(uint8x8_t __p0, uint8x8_t __p1) {
12055   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12064 __ai uint32x2_t vmin_u32(uint32x2_t __p0, uint32x2_t __p1) {
12066   __ret = (uint32x2_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
12070 __ai uint32x2_t vmin_u32(uint32x2_t __p0, uint32x2_t __p1) {
12072   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
12081 __ai uint16x4_t vmin_u16(uint16x4_t __p0, uint16x4_t __p1) {
12083   __ret = (uint16x4_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
12087 __ai uint16x4_t vmin_u16(uint16x4_t __p0, uint16x4_t __p1) {
12089   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12098 __ai int8x8_t vmin_s8(int8x8_t __p0, int8x8_t __p1) {
12100   __ret = (int8x8_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
12104 __ai int8x8_t vmin_s8(int8x8_t __p0, int8x8_t __p1) {
12106   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12115 __ai float32x2_t vmin_f32(float32x2_t __p0, float32x2_t __p1) {
12117   __ret = (float32x2_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
12121 __ai float32x2_t vmin_f32(float32x2_t __p0, float32x2_t __p1) {
12123   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
12132 __ai int32x2_t vmin_s32(int32x2_t __p0, int32x2_t __p1) {
12134   __ret = (int32x2_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
12138 __ai int32x2_t vmin_s32(int32x2_t __p0, int32x2_t __p1) {
12140   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
12149 __ai int16x4_t vmin_s16(int16x4_t __p0, int16x4_t __p1) {
12151   __ret = (int16x4_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
12155 __ai int16x4_t vmin_s16(int16x4_t __p0, int16x4_t __p1) {
12157   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12166 __ai uint8x16_t vmlaq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
12168   __ret = __p0 + __p1 * __p2;
12172 __ai uint8x16_t vmlaq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
12174   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
12184 __ai uint32x4_t vmlaq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
12186   __ret = __p0 + __p1 * __p2;
12190 __ai uint32x4_t vmlaq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
12192   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12202 __ai uint16x8_t vmlaq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
12204   __ret = __p0 + __p1 * __p2;
12208 __ai uint16x8_t vmlaq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
12210   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12220 __ai int8x16_t vmlaq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
12222   __ret = __p0 + __p1 * __p2;
12226 __ai int8x16_t vmlaq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
12228   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
12238 __ai float32x4_t vmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
12240   __ret = __p0 + __p1 * __p2;
12244 __ai float32x4_t vmlaq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
12246   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12256 __ai int32x4_t vmlaq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
12258   __ret = __p0 + __p1 * __p2;
12262 __ai int32x4_t vmlaq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
12264   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12274 __ai int16x8_t vmlaq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
12276   __ret = __p0 + __p1 * __p2;
12280 __ai int16x8_t vmlaq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
12282   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12292 __ai uint8x8_t vmla_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
12294   __ret = __p0 + __p1 * __p2;
12298 __ai uint8x8_t vmla_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
12300   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12310 __ai uint32x2_t vmla_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
12312   __ret = __p0 + __p1 * __p2;
12316 __ai uint32x2_t vmla_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
12318   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
12328 __ai uint16x4_t vmla_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
12330   __ret = __p0 + __p1 * __p2;
12334 __ai uint16x4_t vmla_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
12336   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12346 __ai int8x8_t vmla_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
12348   __ret = __p0 + __p1 * __p2;
12352 __ai int8x8_t vmla_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
12354   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12364 __ai float32x2_t vmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
12366   __ret = __p0 + __p1 * __p2;
12370 __ai float32x2_t vmla_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
12372   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
12382 __ai int32x2_t vmla_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
12384   __ret = __p0 + __p1 * __p2;
12388 __ai int32x2_t vmla_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
12390   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
12400 __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
12402   __ret = __p0 + __p1 * __p2;
12406 __ai int16x4_t vmla_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
12408   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12418 #define vmlaq_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
12420   uint32x4_t __s1 = __p1; \
12427 #define vmlaq_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
12429   uint32x4_t __s1 = __p1; \
12442 #define vmlaq_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
12444   uint16x8_t __s1 = __p1; \
12451 #define vmlaq_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
12453   uint16x8_t __s1 = __p1; \
12466 #define vmlaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
12468   float32x4_t __s1 = __p1; \
12475 #define vmlaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
12477   float32x4_t __s1 = __p1; \
12490 #define vmlaq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
12492   int32x4_t __s1 = __p1; \
12499 #define vmlaq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
12501   int32x4_t __s1 = __p1; \
12514 #define vmlaq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
12516   int16x8_t __s1 = __p1; \
12523 #define vmlaq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
12525   int16x8_t __s1 = __p1; \
12538 #define vmla_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
12540   uint32x2_t __s1 = __p1; \
12547 #define vmla_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
12549   uint32x2_t __s1 = __p1; \
12562 #define vmla_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
12564   uint16x4_t __s1 = __p1; \
12571 #define vmla_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
12573   uint16x4_t __s1 = __p1; \
12586 #define vmla_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
12588   float32x2_t __s1 = __p1; \
12595 #define vmla_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
12597   float32x2_t __s1 = __p1; \
12610 #define vmla_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
12612   int32x2_t __s1 = __p1; \
12619 #define vmla_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
12621   int32x2_t __s1 = __p1; \
12634 #define vmla_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
12636   int16x4_t __s1 = __p1; \
12643 #define vmla_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
12645   int16x4_t __s1 = __p1; \
12658 __ai uint32x4_t vmlaq_n_u32(uint32x4_t __p0, uint32x4_t __p1, uint32_t __p2) {
12660   __ret = __p0 + __p1 * (uint32x4_t) {__p2, __p2, __p2, __p2};
12664 __ai uint32x4_t vmlaq_n_u32(uint32x4_t __p0, uint32x4_t __p1, uint32_t __p2) {
12666   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12675 __ai uint16x8_t vmlaq_n_u16(uint16x8_t __p0, uint16x8_t __p1, uint16_t __p2) {
12677   __ret = __p0 + __p1 * (uint16x8_t) {__p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2};
12681 __ai uint16x8_t vmlaq_n_u16(uint16x8_t __p0, uint16x8_t __p1, uint16_t __p2) {
12683   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12692 __ai float32x4_t vmlaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
12694   __ret = __p0 + __p1 * (float32x4_t) {__p2, __p2, __p2, __p2};
12698 __ai float32x4_t vmlaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
12700   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12709 __ai int32x4_t vmlaq_n_s32(int32x4_t __p0, int32x4_t __p1, int32_t __p2) {
12711   __ret = __p0 + __p1 * (int32x4_t) {__p2, __p2, __p2, __p2};
12715 __ai int32x4_t vmlaq_n_s32(int32x4_t __p0, int32x4_t __p1, int32_t __p2) {
12717   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12726 __ai int16x8_t vmlaq_n_s16(int16x8_t __p0, int16x8_t __p1, int16_t __p2) {
12728   __ret = __p0 + __p1 * (int16x8_t) {__p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2};
12732 __ai int16x8_t vmlaq_n_s16(int16x8_t __p0, int16x8_t __p1, int16_t __p2) {
12734   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12743 __ai uint32x2_t vmla_n_u32(uint32x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
12745   __ret = __p0 + __p1 * (uint32x2_t) {__p2, __p2};
12749 __ai uint32x2_t vmla_n_u32(uint32x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
12751   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
12760 __ai uint16x4_t vmla_n_u16(uint16x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
12762   __ret = __p0 + __p1 * (uint16x4_t) {__p2, __p2, __p2, __p2};
12766 __ai uint16x4_t vmla_n_u16(uint16x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
12768   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12777 __ai float32x2_t vmla_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
12779   __ret = __p0 + __p1 * (float32x2_t) {__p2, __p2};
12783 __ai float32x2_t vmla_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
12785   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
12794 __ai int32x2_t vmla_n_s32(int32x2_t __p0, int32x2_t __p1, int32_t __p2) {
12796   __ret = __p0 + __p1 * (int32x2_t) {__p2, __p2};
12800 __ai int32x2_t vmla_n_s32(int32x2_t __p0, int32x2_t __p1, int32_t __p2) {
12802   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
12811 __ai int16x4_t vmla_n_s16(int16x4_t __p0, int16x4_t __p1, int16_t __p2) {
12813   __ret = __p0 + __p1 * (int16x4_t) {__p2, __p2, __p2, __p2};
12817 __ai int16x4_t vmla_n_s16(int16x4_t __p0, int16x4_t __p1, int16_t __p2) {
12819   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12828 __ai uint8x16_t vmlsq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
12830   __ret = __p0 - __p1 * __p2;
12834 __ai uint8x16_t vmlsq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
12836   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
12846 __ai uint32x4_t vmlsq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
12848   __ret = __p0 - __p1 * __p2;
12852 __ai uint32x4_t vmlsq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
12854   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12864 __ai uint16x8_t vmlsq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
12866   __ret = __p0 - __p1 * __p2;
12870 __ai uint16x8_t vmlsq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
12872   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12882 __ai int8x16_t vmlsq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
12884   __ret = __p0 - __p1 * __p2;
12888 __ai int8x16_t vmlsq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
12890   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
12900 __ai float32x4_t vmlsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
12902   __ret = __p0 - __p1 * __p2;
12906 __ai float32x4_t vmlsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
12908   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12918 __ai int32x4_t vmlsq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
12920   __ret = __p0 - __p1 * __p2;
12924 __ai int32x4_t vmlsq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
12926   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
12936 __ai int16x8_t vmlsq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
12938   __ret = __p0 - __p1 * __p2;
12942 __ai int16x8_t vmlsq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
12944   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12954 __ai uint8x8_t vmls_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
12956   __ret = __p0 - __p1 * __p2;
12960 __ai uint8x8_t vmls_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
12962   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
12972 __ai uint32x2_t vmls_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
12974   __ret = __p0 - __p1 * __p2;
12978 __ai uint32x2_t vmls_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
12980   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
12990 __ai uint16x4_t vmls_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
12992   __ret = __p0 - __p1 * __p2;
12996 __ai uint16x4_t vmls_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
12998   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
13008 __ai int8x8_t vmls_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
13010   __ret = __p0 - __p1 * __p2;
13014 __ai int8x8_t vmls_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
13016   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
13026 __ai float32x2_t vmls_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
13028   __ret = __p0 - __p1 * __p2;
13032 __ai float32x2_t vmls_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
13034   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
13044 __ai int32x2_t vmls_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
13046   __ret = __p0 - __p1 * __p2;
13050 __ai int32x2_t vmls_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
13052   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
13062 __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
13064   __ret = __p0 - __p1 * __p2;
13068 __ai int16x4_t vmls_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
13070   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
13080 #define vmlsq_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
13082   uint32x4_t __s1 = __p1; \
13089 #define vmlsq_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
13091   uint32x4_t __s1 = __p1; \
13104 #define vmlsq_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
13106   uint16x8_t __s1 = __p1; \
13113 #define vmlsq_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
13115   uint16x8_t __s1 = __p1; \
13128 #define vmlsq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
13130   float32x4_t __s1 = __p1; \
13137 #define vmlsq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
13139   float32x4_t __s1 = __p1; \
13152 #define vmlsq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
13154   int32x4_t __s1 = __p1; \
13161 #define vmlsq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
13163   int32x4_t __s1 = __p1; \
13176 #define vmlsq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
13178   int16x8_t __s1 = __p1; \
13185 #define vmlsq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
13187   int16x8_t __s1 = __p1; \
13200 #define vmls_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
13202   uint32x2_t __s1 = __p1; \
13209 #define vmls_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
13211   uint32x2_t __s1 = __p1; \
13224 #define vmls_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
13226   uint16x4_t __s1 = __p1; \
13233 #define vmls_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
13235   uint16x4_t __s1 = __p1; \
13248 #define vmls_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
13250   float32x2_t __s1 = __p1; \
13257 #define vmls_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
13259   float32x2_t __s1 = __p1; \
13272 #define vmls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
13274   int32x2_t __s1 = __p1; \
13281 #define vmls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
13283   int32x2_t __s1 = __p1; \
13296 #define vmls_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
13298   int16x4_t __s1 = __p1; \
13305 #define vmls_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
13307   int16x4_t __s1 = __p1; \
13320 __ai uint32x4_t vmlsq_n_u32(uint32x4_t __p0, uint32x4_t __p1, uint32_t __p2) {
13322   __ret = __p0 - __p1 * (uint32x4_t) {__p2, __p2, __p2, __p2};
13326 __ai uint32x4_t vmlsq_n_u32(uint32x4_t __p0, uint32x4_t __p1, uint32_t __p2) {
13328   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
13337 __ai uint16x8_t vmlsq_n_u16(uint16x8_t __p0, uint16x8_t __p1, uint16_t __p2) {
13339   __ret = __p0 - __p1 * (uint16x8_t) {__p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2};
13343 __ai uint16x8_t vmlsq_n_u16(uint16x8_t __p0, uint16x8_t __p1, uint16_t __p2) {
13345   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
13354 __ai float32x4_t vmlsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
13356   __ret = __p0 - __p1 * (float32x4_t) {__p2, __p2, __p2, __p2};
13360 __ai float32x4_t vmlsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
13362   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
13371 __ai int32x4_t vmlsq_n_s32(int32x4_t __p0, int32x4_t __p1, int32_t __p2) {
13373   __ret = __p0 - __p1 * (int32x4_t) {__p2, __p2, __p2, __p2};
13377 __ai int32x4_t vmlsq_n_s32(int32x4_t __p0, int32x4_t __p1, int32_t __p2) {
13379   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
13388 __ai int16x8_t vmlsq_n_s16(int16x8_t __p0, int16x8_t __p1, int16_t __p2) {
13390   __ret = __p0 - __p1 * (int16x8_t) {__p2, __p2, __p2, __p2, __p2, __p2, __p2, __p2};
13394 __ai int16x8_t vmlsq_n_s16(int16x8_t __p0, int16x8_t __p1, int16_t __p2) {
13396   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
13405 __ai uint32x2_t vmls_n_u32(uint32x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
13407   __ret = __p0 - __p1 * (uint32x2_t) {__p2, __p2};
13411 __ai uint32x2_t vmls_n_u32(uint32x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
13413   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
13422 __ai uint16x4_t vmls_n_u16(uint16x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
13424   __ret = __p0 - __p1 * (uint16x4_t) {__p2, __p2, __p2, __p2};
13428 __ai uint16x4_t vmls_n_u16(uint16x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
13430   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
13439 __ai float32x2_t vmls_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
13441   __ret = __p0 - __p1 * (float32x2_t) {__p2, __p2};
13445 __ai float32x2_t vmls_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
13447   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
13456 __ai int32x2_t vmls_n_s32(int32x2_t __p0, int32x2_t __p1, int32_t __p2) {
13458   __ret = __p0 - __p1 * (int32x2_t) {__p2, __p2};
13462 __ai int32x2_t vmls_n_s32(int32x2_t __p0, int32x2_t __p1, int32_t __p2) {
13464   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
13473 __ai int16x4_t vmls_n_s16(int16x4_t __p0, int16x4_t __p1, int16_t __p2) {
13475   __ret = __p0 - __p1 * (int16x4_t) {__p2, __p2, __p2, __p2};
13479 __ai int16x4_t vmls_n_s16(int16x4_t __p0, int16x4_t __p1, int16_t __p2) {
13481   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14104 __ai uint8x16_t vmulq_u8(uint8x16_t __p0, uint8x16_t __p1) {
14106   __ret = __p0 * __p1;
14110 __ai uint8x16_t vmulq_u8(uint8x16_t __p0, uint8x16_t __p1) {
14112   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
14121 __ai uint32x4_t vmulq_u32(uint32x4_t __p0, uint32x4_t __p1) {
14123   __ret = __p0 * __p1;
14127 __ai uint32x4_t vmulq_u32(uint32x4_t __p0, uint32x4_t __p1) {
14129   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14138 __ai uint16x8_t vmulq_u16(uint16x8_t __p0, uint16x8_t __p1) {
14140   __ret = __p0 * __p1;
14144 __ai uint16x8_t vmulq_u16(uint16x8_t __p0, uint16x8_t __p1) {
14146   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14155 __ai int8x16_t vmulq_s8(int8x16_t __p0, int8x16_t __p1) {
14157   __ret = __p0 * __p1;
14161 __ai int8x16_t vmulq_s8(int8x16_t __p0, int8x16_t __p1) {
14163   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
14172 __ai float32x4_t vmulq_f32(float32x4_t __p0, float32x4_t __p1) {
14174   __ret = __p0 * __p1;
14178 __ai float32x4_t vmulq_f32(float32x4_t __p0, float32x4_t __p1) {
14180   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14189 __ai int32x4_t vmulq_s32(int32x4_t __p0, int32x4_t __p1) {
14191   __ret = __p0 * __p1;
14195 __ai int32x4_t vmulq_s32(int32x4_t __p0, int32x4_t __p1) {
14197   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14206 __ai int16x8_t vmulq_s16(int16x8_t __p0, int16x8_t __p1) {
14208   __ret = __p0 * __p1;
14212 __ai int16x8_t vmulq_s16(int16x8_t __p0, int16x8_t __p1) {
14214   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14223 __ai uint8x8_t vmul_u8(uint8x8_t __p0, uint8x8_t __p1) {
14225   __ret = __p0 * __p1;
14229 __ai uint8x8_t vmul_u8(uint8x8_t __p0, uint8x8_t __p1) {
14231   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14240 __ai uint32x2_t vmul_u32(uint32x2_t __p0, uint32x2_t __p1) {
14242   __ret = __p0 * __p1;
14246 __ai uint32x2_t vmul_u32(uint32x2_t __p0, uint32x2_t __p1) {
14248   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
14257 __ai uint16x4_t vmul_u16(uint16x4_t __p0, uint16x4_t __p1) {
14259   __ret = __p0 * __p1;
14263 __ai uint16x4_t vmul_u16(uint16x4_t __p0, uint16x4_t __p1) {
14265   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14274 __ai int8x8_t vmul_s8(int8x8_t __p0, int8x8_t __p1) {
14276   __ret = __p0 * __p1;
14280 __ai int8x8_t vmul_s8(int8x8_t __p0, int8x8_t __p1) {
14282   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14291 __ai float32x2_t vmul_f32(float32x2_t __p0, float32x2_t __p1) {
14293   __ret = __p0 * __p1;
14297 __ai float32x2_t vmul_f32(float32x2_t __p0, float32x2_t __p1) {
14299   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
14308 __ai int32x2_t vmul_s32(int32x2_t __p0, int32x2_t __p1) {
14310   __ret = __p0 * __p1;
14314 __ai int32x2_t vmul_s32(int32x2_t __p0, int32x2_t __p1) {
14316   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
14325 __ai int16x4_t vmul_s16(int16x4_t __p0, int16x4_t __p1) {
14327   __ret = __p0 * __p1;
14331 __ai int16x4_t vmul_s16(int16x4_t __p0, int16x4_t __p1) {
14333   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14342 __ai poly8x8_t vmul_p8(poly8x8_t __p0, poly8x8_t __p1) {
14344   __ret = (poly8x8_t) __builtin_neon_vmul_v((int8x8_t)__p0, (int8x8_t)__p1, 4);
14348 __ai poly8x8_t vmul_p8(poly8x8_t __p0, poly8x8_t __p1) {
14350   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14359 __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
14361   __ret = (poly8x16_t) __builtin_neon_vmulq_v((int8x16_t)__p0, (int8x16_t)__p1, 36);
14365 __ai poly8x16_t vmulq_p8(poly8x16_t __p0, poly8x16_t __p1) {
14367   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
14376 #define vmulq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
14378   uint32x2_t __s1 = __p1; \
14384 #define vmulq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
14386   uint32x2_t __s1 = __p1; \
14397 #define vmulq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
14399   uint16x4_t __s1 = __p1; \
14405 #define vmulq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
14407   uint16x4_t __s1 = __p1; \
14418 #define vmulq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
14420   float32x2_t __s1 = __p1; \
14426 #define vmulq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
14428   float32x2_t __s1 = __p1; \
14439 #define vmulq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
14441   int32x2_t __s1 = __p1; \
14447 #define vmulq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
14449   int32x2_t __s1 = __p1; \
14460 #define vmulq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
14462   int16x4_t __s1 = __p1; \
14468 #define vmulq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
14470   int16x4_t __s1 = __p1; \
14481 #define vmul_lane_u32(__p0, __p1, __p2) __extension__ ({ \
14483   uint32x2_t __s1 = __p1; \
14489 #define vmul_lane_u32(__p0, __p1, __p2) __extension__ ({ \
14491   uint32x2_t __s1 = __p1; \
14502 #define vmul_lane_u16(__p0, __p1, __p2) __extension__ ({ \
14504   uint16x4_t __s1 = __p1; \
14510 #define vmul_lane_u16(__p0, __p1, __p2) __extension__ ({ \
14512   uint16x4_t __s1 = __p1; \
14523 #define vmul_lane_f32(__p0, __p1, __p2) __extension__ ({ \
14525   float32x2_t __s1 = __p1; \
14531 #define vmul_lane_f32(__p0, __p1, __p2) __extension__ ({ \
14533   float32x2_t __s1 = __p1; \
14544 #define vmul_lane_s32(__p0, __p1, __p2) __extension__ ({ \
14546   int32x2_t __s1 = __p1; \
14552 #define vmul_lane_s32(__p0, __p1, __p2) __extension__ ({ \
14554   int32x2_t __s1 = __p1; \
14565 #define vmul_lane_s16(__p0, __p1, __p2) __extension__ ({ \
14567   int16x4_t __s1 = __p1; \
14573 #define vmul_lane_s16(__p0, __p1, __p2) __extension__ ({ \
14575   int16x4_t __s1 = __p1; \
14586 __ai uint32x4_t vmulq_n_u32(uint32x4_t __p0, uint32_t __p1) {
14588   __ret = __p0 * (uint32x4_t) {__p1, __p1, __p1, __p1};
14592 __ai uint32x4_t vmulq_n_u32(uint32x4_t __p0, uint32_t __p1) {
14595   __ret = __rev0 * (uint32x4_t) {__p1, __p1, __p1, __p1};
14602 __ai uint16x8_t vmulq_n_u16(uint16x8_t __p0, uint16_t __p1) {
14604   __ret = __p0 * (uint16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1};
14608 __ai uint16x8_t vmulq_n_u16(uint16x8_t __p0, uint16_t __p1) {
14611   __ret = __rev0 * (uint16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1};
14618 __ai float32x4_t vmulq_n_f32(float32x4_t __p0, float32_t __p1) {
14620   __ret = __p0 * (float32x4_t) {__p1, __p1, __p1, __p1};
14624 __ai float32x4_t vmulq_n_f32(float32x4_t __p0, float32_t __p1) {
14627   __ret = __rev0 * (float32x4_t) {__p1, __p1, __p1, __p1};
14634 __ai int32x4_t vmulq_n_s32(int32x4_t __p0, int32_t __p1) {
14636   __ret = __p0 * (int32x4_t) {__p1, __p1, __p1, __p1};
14640 __ai int32x4_t vmulq_n_s32(int32x4_t __p0, int32_t __p1) {
14643   __ret = __rev0 * (int32x4_t) {__p1, __p1, __p1, __p1};
14650 __ai int16x8_t vmulq_n_s16(int16x8_t __p0, int16_t __p1) {
14652   __ret = __p0 * (int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1};
14656 __ai int16x8_t vmulq_n_s16(int16x8_t __p0, int16_t __p1) {
14659   __ret = __rev0 * (int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1};
14666 __ai uint32x2_t vmul_n_u32(uint32x2_t __p0, uint32_t __p1) {
14668   __ret = __p0 * (uint32x2_t) {__p1, __p1};
14672 __ai uint32x2_t vmul_n_u32(uint32x2_t __p0, uint32_t __p1) {
14675   __ret = __rev0 * (uint32x2_t) {__p1, __p1};
14682 __ai uint16x4_t vmul_n_u16(uint16x4_t __p0, uint16_t __p1) {
14684   __ret = __p0 * (uint16x4_t) {__p1, __p1, __p1, __p1};
14688 __ai uint16x4_t vmul_n_u16(uint16x4_t __p0, uint16_t __p1) {
14691   __ret = __rev0 * (uint16x4_t) {__p1, __p1, __p1, __p1};
14698 __ai float32x2_t vmul_n_f32(float32x2_t __p0, float32_t __p1) {
14700   __ret = __p0 * (float32x2_t) {__p1, __p1};
14704 __ai float32x2_t vmul_n_f32(float32x2_t __p0, float32_t __p1) {
14707   __ret = __rev0 * (float32x2_t) {__p1, __p1};
14714 __ai int32x2_t vmul_n_s32(int32x2_t __p0, int32_t __p1) {
14716   __ret = __p0 * (int32x2_t) {__p1, __p1};
14720 __ai int32x2_t vmul_n_s32(int32x2_t __p0, int32_t __p1) {
14723   __ret = __rev0 * (int32x2_t) {__p1, __p1};
14730 __ai int16x4_t vmul_n_s16(int16x4_t __p0, int16_t __p1) {
14732   __ret = __p0 * (int16x4_t) {__p1, __p1, __p1, __p1};
14736 __ai int16x4_t vmul_n_s16(int16x4_t __p0, int16_t __p1) {
14739   __ret = __rev0 * (int16x4_t) {__p1, __p1, __p1, __p1};
14746 __ai poly16x8_t vmull_p8(poly8x8_t __p0, poly8x8_t __p1) {
14748   __ret = (poly16x8_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 37);
14752 __ai poly16x8_t vmull_p8(poly8x8_t __p0, poly8x8_t __p1) {
14754   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14760 __ai poly16x8_t __noswap_vmull_p8(poly8x8_t __p0, poly8x8_t __p1) {
14762   __ret = (poly16x8_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 37);
14768 __ai uint16x8_t vmull_u8(uint8x8_t __p0, uint8x8_t __p1) {
14770   __ret = (uint16x8_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 49);
14774 __ai uint16x8_t vmull_u8(uint8x8_t __p0, uint8x8_t __p1) {
14776   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14782 __ai uint16x8_t __noswap_vmull_u8(uint8x8_t __p0, uint8x8_t __p1) {
14784   __ret = (uint16x8_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 49);
14790 __ai uint64x2_t vmull_u32(uint32x2_t __p0, uint32x2_t __p1) {
14792   __ret = (uint64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 51);
14796 __ai uint64x2_t vmull_u32(uint32x2_t __p0, uint32x2_t __p1) {
14798   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
14804 __ai uint64x2_t __noswap_vmull_u32(uint32x2_t __p0, uint32x2_t __p1) {
14806   __ret = (uint64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 51);
14812 __ai uint32x4_t vmull_u16(uint16x4_t __p0, uint16x4_t __p1) {
14814   __ret = (uint32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 50);
14818 __ai uint32x4_t vmull_u16(uint16x4_t __p0, uint16x4_t __p1) {
14820   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14826 __ai uint32x4_t __noswap_vmull_u16(uint16x4_t __p0, uint16x4_t __p1) {
14828   __ret = (uint32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 50);
14834 __ai int16x8_t vmull_s8(int8x8_t __p0, int8x8_t __p1) {
14836   __ret = (int16x8_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 33);
14840 __ai int16x8_t vmull_s8(int8x8_t __p0, int8x8_t __p1) {
14842   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
14848 __ai int16x8_t __noswap_vmull_s8(int8x8_t __p0, int8x8_t __p1) {
14850   __ret = (int16x8_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 33);
14856 __ai int64x2_t vmull_s32(int32x2_t __p0, int32x2_t __p1) {
14858   __ret = (int64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 35);
14862 __ai int64x2_t vmull_s32(int32x2_t __p0, int32x2_t __p1) {
14864   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
14870 __ai int64x2_t __noswap_vmull_s32(int32x2_t __p0, int32x2_t __p1) {
14872   __ret = (int64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 35);
14878 __ai int32x4_t vmull_s16(int16x4_t __p0, int16x4_t __p1) {
14880   __ret = (int32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 34);
14884 __ai int32x4_t vmull_s16(int16x4_t __p0, int16x4_t __p1) {
14886   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
14892 __ai int32x4_t __noswap_vmull_s16(int16x4_t __p0, int16x4_t __p1) {
14894   __ret = (int32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)__p1, 34);
14900 #define vmull_lane_u32(__p0, __p1, __p2) __extension__ ({ \
14902   uint32x2_t __s1 = __p1; \
14908 #define vmull_lane_u32(__p0, __p1, __p2) __extension__ ({ \
14910   uint32x2_t __s1 = __p1; \
14921 #define vmull_lane_u16(__p0, __p1, __p2) __extension__ ({ \
14923   uint16x4_t __s1 = __p1; \
14929 #define vmull_lane_u16(__p0, __p1, __p2) __extension__ ({ \
14931   uint16x4_t __s1 = __p1; \
14942 #define vmull_lane_s32(__p0, __p1, __p2) __extension__ ({ \
14944   int32x2_t __s1 = __p1; \
14950 #define vmull_lane_s32(__p0, __p1, __p2) __extension__ ({ \
14952   int32x2_t __s1 = __p1; \
14963 #define vmull_lane_s16(__p0, __p1, __p2) __extension__ ({ \
14965   int16x4_t __s1 = __p1; \
14971 #define vmull_lane_s16(__p0, __p1, __p2) __extension__ ({ \
14973   int16x4_t __s1 = __p1; \
14984 __ai uint64x2_t vmull_n_u32(uint32x2_t __p0, uint32_t __p1) {
14986   __ret = (uint64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(uint32x2_t) {__p1, __p1}, 51);
14990 __ai uint64x2_t vmull_n_u32(uint32x2_t __p0, uint32_t __p1) {
14993   __ret = (uint64x2_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)(uint32x2_t) {__p1, __p1}, 51);
14997 __ai uint64x2_t __noswap_vmull_n_u32(uint32x2_t __p0, uint32_t __p1) {
14999   __ret = (uint64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(uint32x2_t) {__p1, __p1}, 51);
15005 __ai uint32x4_t vmull_n_u16(uint16x4_t __p0, uint16_t __p1) {
15007   __ret = (uint32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(uint16x4_t) {__p1, __p1, __p1, __p1}, 50);
15011 __ai uint32x4_t vmull_n_u16(uint16x4_t __p0, uint16_t __p1) {
15014   __ret = (uint32x4_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)(uint16x4_t) {__p1, __p1, __p1, __p1}, 50);
15018 __ai uint32x4_t __noswap_vmull_n_u16(uint16x4_t __p0, uint16_t __p1) {
15020   __ret = (uint32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(uint16x4_t) {__p1, __p1, __p1, __p1}, 50);
15026 __ai int64x2_t vmull_n_s32(int32x2_t __p0, int32_t __p1) {
15028   __ret = (int64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
15032 __ai int64x2_t vmull_n_s32(int32x2_t __p0, int32_t __p1) {
15035   __ret = (int64x2_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
15039 __ai int64x2_t __noswap_vmull_n_s32(int32x2_t __p0, int32_t __p1) {
15041   __ret = (int64x2_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
15047 __ai int32x4_t vmull_n_s16(int16x4_t __p0, int16_t __p1) {
15049   __ret = (int32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
15053 __ai int32x4_t vmull_n_s16(int16x4_t __p0, int16_t __p1) {
15056   __ret = (int32x4_t) __builtin_neon_vmull_v((int8x8_t)__rev0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
15060 __ai int32x4_t __noswap_vmull_n_s16(int16x4_t __p0, int16_t __p1) {
15062   __ret = (int32x4_t) __builtin_neon_vmull_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
15420 __ai uint8x16_t vornq_u8(uint8x16_t __p0, uint8x16_t __p1) {
15422   __ret = __p0 | ~__p1;
15426 __ai uint8x16_t vornq_u8(uint8x16_t __p0, uint8x16_t __p1) {
15428   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
15437 __ai uint32x4_t vornq_u32(uint32x4_t __p0, uint32x4_t __p1) {
15439   __ret = __p0 | ~__p1;
15443 __ai uint32x4_t vornq_u32(uint32x4_t __p0, uint32x4_t __p1) {
15445   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15454 __ai uint64x2_t vornq_u64(uint64x2_t __p0, uint64x2_t __p1) {
15456   __ret = __p0 | ~__p1;
15460 __ai uint64x2_t vornq_u64(uint64x2_t __p0, uint64x2_t __p1) {
15462   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15471 __ai uint16x8_t vornq_u16(uint16x8_t __p0, uint16x8_t __p1) {
15473   __ret = __p0 | ~__p1;
15477 __ai uint16x8_t vornq_u16(uint16x8_t __p0, uint16x8_t __p1) {
15479   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
15488 __ai int8x16_t vornq_s8(int8x16_t __p0, int8x16_t __p1) {
15490   __ret = __p0 | ~__p1;
15494 __ai int8x16_t vornq_s8(int8x16_t __p0, int8x16_t __p1) {
15496   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
15505 __ai int32x4_t vornq_s32(int32x4_t __p0, int32x4_t __p1) {
15507   __ret = __p0 | ~__p1;
15511 __ai int32x4_t vornq_s32(int32x4_t __p0, int32x4_t __p1) {
15513   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15522 __ai int64x2_t vornq_s64(int64x2_t __p0, int64x2_t __p1) {
15524   __ret = __p0 | ~__p1;
15528 __ai int64x2_t vornq_s64(int64x2_t __p0, int64x2_t __p1) {
15530   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15539 __ai int16x8_t vornq_s16(int16x8_t __p0, int16x8_t __p1) {
15541   __ret = __p0 | ~__p1;
15545 __ai int16x8_t vornq_s16(int16x8_t __p0, int16x8_t __p1) {
15547   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
15556 __ai uint8x8_t vorn_u8(uint8x8_t __p0, uint8x8_t __p1) {
15558   __ret = __p0 | ~__p1;
15562 __ai uint8x8_t vorn_u8(uint8x8_t __p0, uint8x8_t __p1) {
15564   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
15573 __ai uint32x2_t vorn_u32(uint32x2_t __p0, uint32x2_t __p1) {
15575   __ret = __p0 | ~__p1;
15579 __ai uint32x2_t vorn_u32(uint32x2_t __p0, uint32x2_t __p1) {
15581   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15590 __ai uint64x1_t vorn_u64(uint64x1_t __p0, uint64x1_t __p1) {
15592   __ret = __p0 | ~__p1;
15596 __ai uint64x1_t vorn_u64(uint64x1_t __p0, uint64x1_t __p1) {
15598   __ret = __p0 | ~__p1;
15604 __ai uint16x4_t vorn_u16(uint16x4_t __p0, uint16x4_t __p1) {
15606   __ret = __p0 | ~__p1;
15610 __ai uint16x4_t vorn_u16(uint16x4_t __p0, uint16x4_t __p1) {
15612   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15621 __ai int8x8_t vorn_s8(int8x8_t __p0, int8x8_t __p1) {
15623   __ret = __p0 | ~__p1;
15627 __ai int8x8_t vorn_s8(int8x8_t __p0, int8x8_t __p1) {
15629   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
15638 __ai int32x2_t vorn_s32(int32x2_t __p0, int32x2_t __p1) {
15640   __ret = __p0 | ~__p1;
15644 __ai int32x2_t vorn_s32(int32x2_t __p0, int32x2_t __p1) {
15646   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15655 __ai int64x1_t vorn_s64(int64x1_t __p0, int64x1_t __p1) {
15657   __ret = __p0 | ~__p1;
15661 __ai int64x1_t vorn_s64(int64x1_t __p0, int64x1_t __p1) {
15663   __ret = __p0 | ~__p1;
15669 __ai int16x4_t vorn_s16(int16x4_t __p0, int16x4_t __p1) {
15671   __ret = __p0 | ~__p1;
15675 __ai int16x4_t vorn_s16(int16x4_t __p0, int16x4_t __p1) {
15677   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15686 __ai uint8x16_t vorrq_u8(uint8x16_t __p0, uint8x16_t __p1) {
15688   __ret = __p0 | __p1;
15692 __ai uint8x16_t vorrq_u8(uint8x16_t __p0, uint8x16_t __p1) {
15694   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
15703 __ai uint32x4_t vorrq_u32(uint32x4_t __p0, uint32x4_t __p1) {
15705   __ret = __p0 | __p1;
15709 __ai uint32x4_t vorrq_u32(uint32x4_t __p0, uint32x4_t __p1) {
15711   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15720 __ai uint64x2_t vorrq_u64(uint64x2_t __p0, uint64x2_t __p1) {
15722   __ret = __p0 | __p1;
15726 __ai uint64x2_t vorrq_u64(uint64x2_t __p0, uint64x2_t __p1) {
15728   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15737 __ai uint16x8_t vorrq_u16(uint16x8_t __p0, uint16x8_t __p1) {
15739   __ret = __p0 | __p1;
15743 __ai uint16x8_t vorrq_u16(uint16x8_t __p0, uint16x8_t __p1) {
15745   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
15754 __ai int8x16_t vorrq_s8(int8x16_t __p0, int8x16_t __p1) {
15756   __ret = __p0 | __p1;
15760 __ai int8x16_t vorrq_s8(int8x16_t __p0, int8x16_t __p1) {
15762   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
15771 __ai int32x4_t vorrq_s32(int32x4_t __p0, int32x4_t __p1) {
15773   __ret = __p0 | __p1;
15777 __ai int32x4_t vorrq_s32(int32x4_t __p0, int32x4_t __p1) {
15779   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15788 __ai int64x2_t vorrq_s64(int64x2_t __p0, int64x2_t __p1) {
15790   __ret = __p0 | __p1;
15794 __ai int64x2_t vorrq_s64(int64x2_t __p0, int64x2_t __p1) {
15796   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15805 __ai int16x8_t vorrq_s16(int16x8_t __p0, int16x8_t __p1) {
15807   __ret = __p0 | __p1;
15811 __ai int16x8_t vorrq_s16(int16x8_t __p0, int16x8_t __p1) {
15813   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
15822 __ai uint8x8_t vorr_u8(uint8x8_t __p0, uint8x8_t __p1) {
15824   __ret = __p0 | __p1;
15828 __ai uint8x8_t vorr_u8(uint8x8_t __p0, uint8x8_t __p1) {
15830   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
15839 __ai uint32x2_t vorr_u32(uint32x2_t __p0, uint32x2_t __p1) {
15841   __ret = __p0 | __p1;
15845 __ai uint32x2_t vorr_u32(uint32x2_t __p0, uint32x2_t __p1) {
15847   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15856 __ai uint64x1_t vorr_u64(uint64x1_t __p0, uint64x1_t __p1) {
15858   __ret = __p0 | __p1;
15862 __ai uint64x1_t vorr_u64(uint64x1_t __p0, uint64x1_t __p1) {
15864   __ret = __p0 | __p1;
15870 __ai uint16x4_t vorr_u16(uint16x4_t __p0, uint16x4_t __p1) {
15872   __ret = __p0 | __p1;
15876 __ai uint16x4_t vorr_u16(uint16x4_t __p0, uint16x4_t __p1) {
15878   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15887 __ai int8x8_t vorr_s8(int8x8_t __p0, int8x8_t __p1) {
15889   __ret = __p0 | __p1;
15893 __ai int8x8_t vorr_s8(int8x8_t __p0, int8x8_t __p1) {
15895   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
15904 __ai int32x2_t vorr_s32(int32x2_t __p0, int32x2_t __p1) {
15906   __ret = __p0 | __p1;
15910 __ai int32x2_t vorr_s32(int32x2_t __p0, int32x2_t __p1) {
15912   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
15921 __ai int64x1_t vorr_s64(int64x1_t __p0, int64x1_t __p1) {
15923   __ret = __p0 | __p1;
15927 __ai int64x1_t vorr_s64(int64x1_t __p0, int64x1_t __p1) {
15929   __ret = __p0 | __p1;
15935 __ai int16x4_t vorr_s16(int16x4_t __p0, int16x4_t __p1) {
15937   __ret = __p0 | __p1;
15941 __ai int16x4_t vorr_s16(int16x4_t __p0, int16x4_t __p1) {
15943   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15952 __ai uint16x8_t vpadalq_u8(uint16x8_t __p0, uint8x16_t __p1) {
15954   __ret = (uint16x8_t) __builtin_neon_vpadalq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
15958 __ai uint16x8_t vpadalq_u8(uint16x8_t __p0, uint8x16_t __p1) {
15960   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
15969 __ai uint64x2_t vpadalq_u32(uint64x2_t __p0, uint32x4_t __p1) {
15971   __ret = (uint64x2_t) __builtin_neon_vpadalq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
15975 __ai uint64x2_t vpadalq_u32(uint64x2_t __p0, uint32x4_t __p1) {
15977   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
15986 __ai uint32x4_t vpadalq_u16(uint32x4_t __p0, uint16x8_t __p1) {
15988   __ret = (uint32x4_t) __builtin_neon_vpadalq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
15992 __ai uint32x4_t vpadalq_u16(uint32x4_t __p0, uint16x8_t __p1) {
15994   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16003 __ai int16x8_t vpadalq_s8(int16x8_t __p0, int8x16_t __p1) {
16005   __ret = (int16x8_t) __builtin_neon_vpadalq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
16009 __ai int16x8_t vpadalq_s8(int16x8_t __p0, int8x16_t __p1) {
16011   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
16020 __ai int64x2_t vpadalq_s32(int64x2_t __p0, int32x4_t __p1) {
16022   __ret = (int64x2_t) __builtin_neon_vpadalq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
16026 __ai int64x2_t vpadalq_s32(int64x2_t __p0, int32x4_t __p1) {
16028   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16037 __ai int32x4_t vpadalq_s16(int32x4_t __p0, int16x8_t __p1) {
16039   __ret = (int32x4_t) __builtin_neon_vpadalq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
16043 __ai int32x4_t vpadalq_s16(int32x4_t __p0, int16x8_t __p1) {
16045   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16054 __ai uint16x4_t vpadal_u8(uint16x4_t __p0, uint8x8_t __p1) {
16056   __ret = (uint16x4_t) __builtin_neon_vpadal_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
16060 __ai uint16x4_t vpadal_u8(uint16x4_t __p0, uint8x8_t __p1) {
16062   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16071 __ai uint64x1_t vpadal_u32(uint64x1_t __p0, uint32x2_t __p1) {
16073   __ret = (uint64x1_t) __builtin_neon_vpadal_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
16077 __ai uint64x1_t vpadal_u32(uint64x1_t __p0, uint32x2_t __p1) {
16078   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16086 __ai uint32x2_t vpadal_u16(uint32x2_t __p0, uint16x4_t __p1) {
16088   __ret = (uint32x2_t) __builtin_neon_vpadal_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
16092 __ai uint32x2_t vpadal_u16(uint32x2_t __p0, uint16x4_t __p1) {
16094   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16103 __ai int16x4_t vpadal_s8(int16x4_t __p0, int8x8_t __p1) {
16105   __ret = (int16x4_t) __builtin_neon_vpadal_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
16109 __ai int16x4_t vpadal_s8(int16x4_t __p0, int8x8_t __p1) {
16111   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16120 __ai int64x1_t vpadal_s32(int64x1_t __p0, int32x2_t __p1) {
16122   __ret = (int64x1_t) __builtin_neon_vpadal_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
16126 __ai int64x1_t vpadal_s32(int64x1_t __p0, int32x2_t __p1) {
16127   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16135 __ai int32x2_t vpadal_s16(int32x2_t __p0, int16x4_t __p1) {
16137   __ret = (int32x2_t) __builtin_neon_vpadal_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
16141 __ai int32x2_t vpadal_s16(int32x2_t __p0, int16x4_t __p1) {
16143   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16152 __ai uint8x8_t vpadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
16154   __ret = (uint8x8_t) __builtin_neon_vpadd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
16158 __ai uint8x8_t vpadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
16160   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16169 __ai uint32x2_t vpadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
16171   __ret = (uint32x2_t) __builtin_neon_vpadd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
16175 __ai uint32x2_t vpadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
16177   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16186 __ai uint16x4_t vpadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
16188   __ret = (uint16x4_t) __builtin_neon_vpadd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
16192 __ai uint16x4_t vpadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
16194   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16203 __ai int8x8_t vpadd_s8(int8x8_t __p0, int8x8_t __p1) {
16205   __ret = (int8x8_t) __builtin_neon_vpadd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
16209 __ai int8x8_t vpadd_s8(int8x8_t __p0, int8x8_t __p1) {
16211   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16220 __ai float32x2_t vpadd_f32(float32x2_t __p0, float32x2_t __p1) {
16222   __ret = (float32x2_t) __builtin_neon_vpadd_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
16226 __ai float32x2_t vpadd_f32(float32x2_t __p0, float32x2_t __p1) {
16228   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16237 __ai int32x2_t vpadd_s32(int32x2_t __p0, int32x2_t __p1) {
16239   __ret = (int32x2_t) __builtin_neon_vpadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
16243 __ai int32x2_t vpadd_s32(int32x2_t __p0, int32x2_t __p1) {
16245   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16254 __ai int16x4_t vpadd_s16(int16x4_t __p0, int16x4_t __p1) {
16256   __ret = (int16x4_t) __builtin_neon_vpadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
16260 __ai int16x4_t vpadd_s16(int16x4_t __p0, int16x4_t __p1) {
16262   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16461 __ai uint8x8_t vpmax_u8(uint8x8_t __p0, uint8x8_t __p1) {
16463   __ret = (uint8x8_t) __builtin_neon_vpmax_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
16467 __ai uint8x8_t vpmax_u8(uint8x8_t __p0, uint8x8_t __p1) {
16469   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16478 __ai uint32x2_t vpmax_u32(uint32x2_t __p0, uint32x2_t __p1) {
16480   __ret = (uint32x2_t) __builtin_neon_vpmax_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
16484 __ai uint32x2_t vpmax_u32(uint32x2_t __p0, uint32x2_t __p1) {
16486   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16495 __ai uint16x4_t vpmax_u16(uint16x4_t __p0, uint16x4_t __p1) {
16497   __ret = (uint16x4_t) __builtin_neon_vpmax_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
16501 __ai uint16x4_t vpmax_u16(uint16x4_t __p0, uint16x4_t __p1) {
16503   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16512 __ai int8x8_t vpmax_s8(int8x8_t __p0, int8x8_t __p1) {
16514   __ret = (int8x8_t) __builtin_neon_vpmax_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
16518 __ai int8x8_t vpmax_s8(int8x8_t __p0, int8x8_t __p1) {
16520   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16529 __ai float32x2_t vpmax_f32(float32x2_t __p0, float32x2_t __p1) {
16531   __ret = (float32x2_t) __builtin_neon_vpmax_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
16535 __ai float32x2_t vpmax_f32(float32x2_t __p0, float32x2_t __p1) {
16537   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16546 __ai int32x2_t vpmax_s32(int32x2_t __p0, int32x2_t __p1) {
16548   __ret = (int32x2_t) __builtin_neon_vpmax_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
16552 __ai int32x2_t vpmax_s32(int32x2_t __p0, int32x2_t __p1) {
16554   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16563 __ai int16x4_t vpmax_s16(int16x4_t __p0, int16x4_t __p1) {
16565   __ret = (int16x4_t) __builtin_neon_vpmax_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
16569 __ai int16x4_t vpmax_s16(int16x4_t __p0, int16x4_t __p1) {
16571   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16580 __ai uint8x8_t vpmin_u8(uint8x8_t __p0, uint8x8_t __p1) {
16582   __ret = (uint8x8_t) __builtin_neon_vpmin_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
16586 __ai uint8x8_t vpmin_u8(uint8x8_t __p0, uint8x8_t __p1) {
16588   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16597 __ai uint32x2_t vpmin_u32(uint32x2_t __p0, uint32x2_t __p1) {
16599   __ret = (uint32x2_t) __builtin_neon_vpmin_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
16603 __ai uint32x2_t vpmin_u32(uint32x2_t __p0, uint32x2_t __p1) {
16605   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16614 __ai uint16x4_t vpmin_u16(uint16x4_t __p0, uint16x4_t __p1) {
16616   __ret = (uint16x4_t) __builtin_neon_vpmin_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
16620 __ai uint16x4_t vpmin_u16(uint16x4_t __p0, uint16x4_t __p1) {
16622   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16631 __ai int8x8_t vpmin_s8(int8x8_t __p0, int8x8_t __p1) {
16633   __ret = (int8x8_t) __builtin_neon_vpmin_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
16637 __ai int8x8_t vpmin_s8(int8x8_t __p0, int8x8_t __p1) {
16639   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16648 __ai float32x2_t vpmin_f32(float32x2_t __p0, float32x2_t __p1) {
16650   __ret = (float32x2_t) __builtin_neon_vpmin_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
16654 __ai float32x2_t vpmin_f32(float32x2_t __p0, float32x2_t __p1) {
16656   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16665 __ai int32x2_t vpmin_s32(int32x2_t __p0, int32x2_t __p1) {
16667   __ret = (int32x2_t) __builtin_neon_vpmin_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
16671 __ai int32x2_t vpmin_s32(int32x2_t __p0, int32x2_t __p1) {
16673   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16682 __ai int16x4_t vpmin_s16(int16x4_t __p0, int16x4_t __p1) {
16684   __ret = (int16x4_t) __builtin_neon_vpmin_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
16688 __ai int16x4_t vpmin_s16(int16x4_t __p0, int16x4_t __p1) {
16690   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16795 __ai uint8x16_t vqaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
16797   __ret = (uint8x16_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
16801 __ai uint8x16_t vqaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
16803   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
16812 __ai uint32x4_t vqaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
16814   __ret = (uint32x4_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
16818 __ai uint32x4_t vqaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
16820   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16829 __ai uint64x2_t vqaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
16831   __ret = (uint64x2_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
16835 __ai uint64x2_t vqaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
16837   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16846 __ai uint16x8_t vqaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
16848   __ret = (uint16x8_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
16852 __ai uint16x8_t vqaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
16854   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16863 __ai int8x16_t vqaddq_s8(int8x16_t __p0, int8x16_t __p1) {
16865   __ret = (int8x16_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
16869 __ai int8x16_t vqaddq_s8(int8x16_t __p0, int8x16_t __p1) {
16871   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
16880 __ai int32x4_t vqaddq_s32(int32x4_t __p0, int32x4_t __p1) {
16882   __ret = (int32x4_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
16886 __ai int32x4_t vqaddq_s32(int32x4_t __p0, int32x4_t __p1) {
16888   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16894 __ai int32x4_t __noswap_vqaddq_s32(int32x4_t __p0, int32x4_t __p1) {
16896   __ret = (int32x4_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
16902 __ai int64x2_t vqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
16904   __ret = (int64x2_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
16908 __ai int64x2_t vqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
16910   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16919 __ai int16x8_t vqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
16921   __ret = (int16x8_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
16925 __ai int16x8_t vqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
16927   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16933 __ai int16x8_t __noswap_vqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
16935   __ret = (int16x8_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
16941 __ai uint8x8_t vqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
16943   __ret = (uint8x8_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
16947 __ai uint8x8_t vqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
16949   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16958 __ai uint32x2_t vqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
16960   __ret = (uint32x2_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
16964 __ai uint32x2_t vqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
16966   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16975 __ai uint64x1_t vqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
16977   __ret = (uint64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
16981 __ai uint64x1_t vqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
16983   __ret = (uint64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
16989 __ai uint16x4_t vqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
16991   __ret = (uint16x4_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
16995 __ai uint16x4_t vqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
16997   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17006 __ai int8x8_t vqadd_s8(int8x8_t __p0, int8x8_t __p1) {
17008   __ret = (int8x8_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
17012 __ai int8x8_t vqadd_s8(int8x8_t __p0, int8x8_t __p1) {
17014   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
17023 __ai int32x2_t vqadd_s32(int32x2_t __p0, int32x2_t __p1) {
17025   __ret = (int32x2_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
17029 __ai int32x2_t vqadd_s32(int32x2_t __p0, int32x2_t __p1) {
17031   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17037 __ai int32x2_t __noswap_vqadd_s32(int32x2_t __p0, int32x2_t __p1) {
17039   __ret = (int32x2_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
17045 __ai int64x1_t vqadd_s64(int64x1_t __p0, int64x1_t __p1) {
17047   __ret = (int64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
17051 __ai int64x1_t vqadd_s64(int64x1_t __p0, int64x1_t __p1) {
17053   __ret = (int64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
17059 __ai int16x4_t vqadd_s16(int16x4_t __p0, int16x4_t __p1) {
17061   __ret = (int16x4_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
17065 __ai int16x4_t vqadd_s16(int16x4_t __p0, int16x4_t __p1) {
17067   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17073 __ai int16x4_t __noswap_vqadd_s16(int16x4_t __p0, int16x4_t __p1) {
17075   __ret = (int16x4_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
17081 __ai int64x2_t vqdmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17083   __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 35);
17087 __ai int64x2_t vqdmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17089   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17096 __ai int64x2_t __noswap_vqdmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17098   __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 35);
17104 __ai int32x4_t vqdmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17106   __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 34);
17110 __ai int32x4_t vqdmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17112   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17119 __ai int32x4_t __noswap_vqdmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17121   __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 34);
17127 #define vqdmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
17129   int32x2_t __s1 = __p1; \
17136 #define vqdmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
17138   int32x2_t __s1 = __p1; \
17151 #define vqdmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
17153   int16x4_t __s1 = __p1; \
17160 #define vqdmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
17162   int16x4_t __s1 = __p1; \
17175 __ai int64x2_t vqdmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17177   __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
17181 __ai int64x2_t vqdmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17183   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17189 __ai int64x2_t __noswap_vqdmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17191   __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
17197 __ai int32x4_t vqdmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17199   __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
17203 __ai int32x4_t vqdmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17205   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17211 __ai int32x4_t __noswap_vqdmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17213   __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
17219 __ai int64x2_t vqdmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17221   __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 35);
17225 __ai int64x2_t vqdmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17227   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17234 __ai int64x2_t __noswap_vqdmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17236   __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 35);
17242 __ai int32x4_t vqdmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17244   __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 34);
17248 __ai int32x4_t vqdmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17250   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17257 __ai int32x4_t __noswap_vqdmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17259   __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 34);
17265 #define vqdmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
17267   int32x2_t __s1 = __p1; \
17274 #define vqdmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
17276   int32x2_t __s1 = __p1; \
17289 #define vqdmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
17291   int16x4_t __s1 = __p1; \
17298 #define vqdmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
17300   int16x4_t __s1 = __p1; \
17313 __ai int64x2_t vqdmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17315   __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
17319 __ai int64x2_t vqdmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17321   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17327 __ai int64x2_t __noswap_vqdmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17329   __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
17335 __ai int32x4_t vqdmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17337   __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
17341 __ai int32x4_t vqdmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17343   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17349 __ai int32x4_t __noswap_vqdmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17351   __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
17357 __ai int32x4_t vqdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
17359   __ret = (int32x4_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
17363 __ai int32x4_t vqdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
17365   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17371 __ai int32x4_t __noswap_vqdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
17373   __ret = (int32x4_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
17379 __ai int16x8_t vqdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
17381   __ret = (int16x8_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
17385 __ai int16x8_t vqdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
17387   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
17393 __ai int16x8_t __noswap_vqdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
17395   __ret = (int16x8_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
17401 __ai int32x2_t vqdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
17403   __ret = (int32x2_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
17407 __ai int32x2_t vqdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
17409   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17415 __ai int32x2_t __noswap_vqdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
17417   __ret = (int32x2_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
17423 __ai int16x4_t vqdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
17425   __ret = (int16x4_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
17429 __ai int16x4_t vqdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
17431   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17437 __ai int16x4_t __noswap_vqdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
17439   __ret = (int16x4_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
17445 #define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17447   int32x2_t __s1 = __p1; \
17453 #define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17455   int32x2_t __s1 = __p1; \
17466 #define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17468   int16x4_t __s1 = __p1; \
17474 #define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17476   int16x4_t __s1 = __p1; \
17487 #define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17489   int32x2_t __s1 = __p1; \
17495 #define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17497   int32x2_t __s1 = __p1; \
17508 #define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17510   int16x4_t __s1 = __p1; \
17516 #define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17518   int16x4_t __s1 = __p1; \
17529 __ai int32x4_t vqdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
17531   __ret = (int32x4_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
17535 __ai int32x4_t vqdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
17538   __ret = (int32x4_t) __builtin_neon_vqdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
17545 __ai int16x8_t vqdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
17547   __ret = (int16x8_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
17551 __ai int16x8_t vqdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
17554   __ret = (int16x8_t) __builtin_neon_vqdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
17561 __ai int32x2_t vqdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
17563   __ret = (int32x2_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
17567 __ai int32x2_t vqdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
17570   __ret = (int32x2_t) __builtin_neon_vqdmulh_v((int8x8_t)__rev0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
17577 __ai int16x4_t vqdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
17579   __ret = (int16x4_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
17583 __ai int16x4_t vqdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
17586   __ret = (int16x4_t) __builtin_neon_vqdmulh_v((int8x8_t)__rev0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
17593 __ai int64x2_t vqdmull_s32(int32x2_t __p0, int32x2_t __p1) {
17595   __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)__p1, 35);
17599 __ai int64x2_t vqdmull_s32(int32x2_t __p0, int32x2_t __p1) {
17601   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17607 __ai int64x2_t __noswap_vqdmull_s32(int32x2_t __p0, int32x2_t __p1) {
17609   __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)__p1, 35);
17615 __ai int32x4_t vqdmull_s16(int16x4_t __p0, int16x4_t __p1) {
17617   __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)__p1, 34);
17621 __ai int32x4_t vqdmull_s16(int16x4_t __p0, int16x4_t __p1) {
17623   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17629 __ai int32x4_t __noswap_vqdmull_s16(int16x4_t __p0, int16x4_t __p1) {
17631   __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)__p1, 34);
17637 #define vqdmull_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17639   int32x2_t __s1 = __p1; \
17645 #define vqdmull_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17647   int32x2_t __s1 = __p1; \
17658 #define vqdmull_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17660   int16x4_t __s1 = __p1; \
17666 #define vqdmull_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17668   int16x4_t __s1 = __p1; \
17679 __ai int64x2_t vqdmull_n_s32(int32x2_t __p0, int32_t __p1) {
17681   __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
17685 __ai int64x2_t vqdmull_n_s32(int32x2_t __p0, int32_t __p1) {
17688   __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__rev0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
17692 __ai int64x2_t __noswap_vqdmull_n_s32(int32x2_t __p0, int32_t __p1) {
17694   __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
17700 __ai int32x4_t vqdmull_n_s16(int16x4_t __p0, int16_t __p1) {
17702   __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
17706 __ai int32x4_t vqdmull_n_s16(int16x4_t __p0, int16_t __p1) {
17709   __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__rev0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
17713 __ai int32x4_t __noswap_vqdmull_n_s16(int16x4_t __p0, int16_t __p1) {
17715   __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
18006 __ai int32x4_t vqrdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
18008   __ret = (int32x4_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
18012 __ai int32x4_t vqrdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
18014   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18020 __ai int32x4_t __noswap_vqrdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
18022   __ret = (int32x4_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
18028 __ai int16x8_t vqrdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
18030   __ret = (int16x8_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
18034 __ai int16x8_t vqrdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
18036   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18042 __ai int16x8_t __noswap_vqrdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
18044   __ret = (int16x8_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
18050 __ai int32x2_t vqrdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
18052   __ret = (int32x2_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
18056 __ai int32x2_t vqrdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
18058   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18064 __ai int32x2_t __noswap_vqrdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
18066   __ret = (int32x2_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
18072 __ai int16x4_t vqrdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
18074   __ret = (int16x4_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
18078 __ai int16x4_t vqrdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
18080   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18086 __ai int16x4_t __noswap_vqrdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
18088   __ret = (int16x4_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
18094 #define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
18096   int32x2_t __s1 = __p1; \
18102 #define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
18104   int32x2_t __s1 = __p1; \
18115 #define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
18117   int16x4_t __s1 = __p1; \
18123 #define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
18125   int16x4_t __s1 = __p1; \
18136 #define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
18138   int32x2_t __s1 = __p1; \
18144 #define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
18146   int32x2_t __s1 = __p1; \
18157 #define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
18159   int16x4_t __s1 = __p1; \
18165 #define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
18167   int16x4_t __s1 = __p1; \
18178 __ai int32x4_t vqrdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
18180   __ret = (int32x4_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
18184 __ai int32x4_t vqrdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
18187   __ret = (int32x4_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
18194 __ai int16x8_t vqrdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
18196   __ret = (int16x8_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
18200 __ai int16x8_t vqrdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
18203   __ret = (int16x8_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
18210 __ai int32x2_t vqrdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
18212   __ret = (int32x2_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
18216 __ai int32x2_t vqrdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
18219   __ret = (int32x2_t) __builtin_neon_vqrdmulh_v((int8x8_t)__rev0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
18226 __ai int16x4_t vqrdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
18228   __ret = (int16x4_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
18232 __ai int16x4_t vqrdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
18235   __ret = (int16x4_t) __builtin_neon_vqrdmulh_v((int8x8_t)__rev0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
18242 __ai uint8x16_t vqrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
18244   __ret = (uint8x16_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
18248 __ai uint8x16_t vqrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
18250   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18259 __ai uint32x4_t vqrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
18261   __ret = (uint32x4_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
18265 __ai uint32x4_t vqrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
18267   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18276 __ai uint64x2_t vqrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
18278   __ret = (uint64x2_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
18282 __ai uint64x2_t vqrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
18284   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18293 __ai uint16x8_t vqrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
18295   __ret = (uint16x8_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
18299 __ai uint16x8_t vqrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
18301   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18310 __ai int8x16_t vqrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
18312   __ret = (int8x16_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
18316 __ai int8x16_t vqrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
18318   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18327 __ai int32x4_t vqrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
18329   __ret = (int32x4_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
18333 __ai int32x4_t vqrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
18335   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18344 __ai int64x2_t vqrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
18346   __ret = (int64x2_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
18350 __ai int64x2_t vqrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
18352   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18361 __ai int16x8_t vqrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
18363   __ret = (int16x8_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
18367 __ai int16x8_t vqrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
18369   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18378 __ai uint8x8_t vqrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
18380   __ret = (uint8x8_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
18384 __ai uint8x8_t vqrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
18386   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18395 __ai uint32x2_t vqrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
18397   __ret = (uint32x2_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
18401 __ai uint32x2_t vqrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
18403   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18412 __ai uint64x1_t vqrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
18414   __ret = (uint64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
18418 __ai uint64x1_t vqrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
18420   __ret = (uint64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
18426 __ai uint16x4_t vqrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
18428   __ret = (uint16x4_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
18432 __ai uint16x4_t vqrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
18434   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18443 __ai int8x8_t vqrshl_s8(int8x8_t __p0, int8x8_t __p1) {
18445   __ret = (int8x8_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
18449 __ai int8x8_t vqrshl_s8(int8x8_t __p0, int8x8_t __p1) {
18451   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18460 __ai int32x2_t vqrshl_s32(int32x2_t __p0, int32x2_t __p1) {
18462   __ret = (int32x2_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
18466 __ai int32x2_t vqrshl_s32(int32x2_t __p0, int32x2_t __p1) {
18468   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18477 __ai int64x1_t vqrshl_s64(int64x1_t __p0, int64x1_t __p1) {
18479   __ret = (int64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
18483 __ai int64x1_t vqrshl_s64(int64x1_t __p0, int64x1_t __p1) {
18485   __ret = (int64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
18491 __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
18493   __ret = (int16x4_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
18497 __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
18499   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18508 #define vqrshrn_n_u32(__p0, __p1) __extension__ ({ \
18511   __ret = (uint16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 17); \
18515 #define vqrshrn_n_u32(__p0, __p1) __extension__ ({ \
18519   __ret = (uint16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 17); \
18523 #define __noswap_vqrshrn_n_u32(__p0, __p1) __extension__ ({ \
18526   __ret = (uint16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 17); \
18532 #define vqrshrn_n_u64(__p0, __p1) __extension__ ({ \
18535   __ret = (uint32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 18); \
18539 #define vqrshrn_n_u64(__p0, __p1) __extension__ ({ \
18543   __ret = (uint32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 18); \
18547 #define __noswap_vqrshrn_n_u64(__p0, __p1) __extension__ ({ \
18550   __ret = (uint32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 18); \
18556 #define vqrshrn_n_u16(__p0, __p1) __extension__ ({ \
18559   __ret = (uint8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 16); \
18563 #define vqrshrn_n_u16(__p0, __p1) __extension__ ({ \
18567   __ret = (uint8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 16); \
18571 #define __noswap_vqrshrn_n_u16(__p0, __p1) __extension__ ({ \
18574   __ret = (uint8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 16); \
18580 #define vqrshrn_n_s32(__p0, __p1) __extension__ ({ \
18583   __ret = (int16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 1); \
18587 #define vqrshrn_n_s32(__p0, __p1) __extension__ ({ \
18591   __ret = (int16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 1); \
18595 #define __noswap_vqrshrn_n_s32(__p0, __p1) __extension__ ({ \
18598   __ret = (int16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 1); \
18604 #define vqrshrn_n_s64(__p0, __p1) __extension__ ({ \
18607   __ret = (int32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 2); \
18611 #define vqrshrn_n_s64(__p0, __p1) __extension__ ({ \
18615   __ret = (int32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 2); \
18619 #define __noswap_vqrshrn_n_s64(__p0, __p1) __extension__ ({ \
18622   __ret = (int32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 2); \
18628 #define vqrshrn_n_s16(__p0, __p1) __extension__ ({ \
18631   __ret = (int8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 0); \
18635 #define vqrshrn_n_s16(__p0, __p1) __extension__ ({ \
18639   __ret = (int8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 0); \
18643 #define __noswap_vqrshrn_n_s16(__p0, __p1) __extension__ ({ \
18646   __ret = (int8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 0); \
18652 #define vqrshrun_n_s32(__p0, __p1) __extension__ ({ \
18655   __ret = (uint16x4_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 17); \
18659 #define vqrshrun_n_s32(__p0, __p1) __extension__ ({ \
18663   __ret = (uint16x4_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__rev0, __p1, 17); \
18667 #define __noswap_vqrshrun_n_s32(__p0, __p1) __extension__ ({ \
18670   __ret = (uint16x4_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 17); \
18676 #define vqrshrun_n_s64(__p0, __p1) __extension__ ({ \
18679   __ret = (uint32x2_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 18); \
18683 #define vqrshrun_n_s64(__p0, __p1) __extension__ ({ \
18687   __ret = (uint32x2_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__rev0, __p1, 18); \
18691 #define __noswap_vqrshrun_n_s64(__p0, __p1) __extension__ ({ \
18694   __ret = (uint32x2_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 18); \
18700 #define vqrshrun_n_s16(__p0, __p1) __extension__ ({ \
18703   __ret = (uint8x8_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 16); \
18707 #define vqrshrun_n_s16(__p0, __p1) __extension__ ({ \
18711   __ret = (uint8x8_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__rev0, __p1, 16); \
18715 #define __noswap_vqrshrun_n_s16(__p0, __p1) __extension__ ({ \
18718   __ret = (uint8x8_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 16); \
18724 __ai uint8x16_t vqshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
18726   __ret = (uint8x16_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
18730 __ai uint8x16_t vqshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
18732   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18741 __ai uint32x4_t vqshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
18743   __ret = (uint32x4_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
18747 __ai uint32x4_t vqshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
18749   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18758 __ai uint64x2_t vqshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
18760   __ret = (uint64x2_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
18764 __ai uint64x2_t vqshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
18766   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18775 __ai uint16x8_t vqshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
18777   __ret = (uint16x8_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
18781 __ai uint16x8_t vqshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
18783   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18792 __ai int8x16_t vqshlq_s8(int8x16_t __p0, int8x16_t __p1) {
18794   __ret = (int8x16_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
18798 __ai int8x16_t vqshlq_s8(int8x16_t __p0, int8x16_t __p1) {
18800   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18809 __ai int32x4_t vqshlq_s32(int32x4_t __p0, int32x4_t __p1) {
18811   __ret = (int32x4_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
18815 __ai int32x4_t vqshlq_s32(int32x4_t __p0, int32x4_t __p1) {
18817   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18826 __ai int64x2_t vqshlq_s64(int64x2_t __p0, int64x2_t __p1) {
18828   __ret = (int64x2_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
18832 __ai int64x2_t vqshlq_s64(int64x2_t __p0, int64x2_t __p1) {
18834   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18843 __ai int16x8_t vqshlq_s16(int16x8_t __p0, int16x8_t __p1) {
18845   __ret = (int16x8_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
18849 __ai int16x8_t vqshlq_s16(int16x8_t __p0, int16x8_t __p1) {
18851   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18860 __ai uint8x8_t vqshl_u8(uint8x8_t __p0, int8x8_t __p1) {
18862   __ret = (uint8x8_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
18866 __ai uint8x8_t vqshl_u8(uint8x8_t __p0, int8x8_t __p1) {
18868   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18877 __ai uint32x2_t vqshl_u32(uint32x2_t __p0, int32x2_t __p1) {
18879   __ret = (uint32x2_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
18883 __ai uint32x2_t vqshl_u32(uint32x2_t __p0, int32x2_t __p1) {
18885   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18894 __ai uint64x1_t vqshl_u64(uint64x1_t __p0, int64x1_t __p1) {
18896   __ret = (uint64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
18900 __ai uint64x1_t vqshl_u64(uint64x1_t __p0, int64x1_t __p1) {
18902   __ret = (uint64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
18908 __ai uint16x4_t vqshl_u16(uint16x4_t __p0, int16x4_t __p1) {
18910   __ret = (uint16x4_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
18914 __ai uint16x4_t vqshl_u16(uint16x4_t __p0, int16x4_t __p1) {
18916   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18925 __ai int8x8_t vqshl_s8(int8x8_t __p0, int8x8_t __p1) {
18927   __ret = (int8x8_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
18931 __ai int8x8_t vqshl_s8(int8x8_t __p0, int8x8_t __p1) {
18933   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18942 __ai int32x2_t vqshl_s32(int32x2_t __p0, int32x2_t __p1) {
18944   __ret = (int32x2_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
18948 __ai int32x2_t vqshl_s32(int32x2_t __p0, int32x2_t __p1) {
18950   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18959 __ai int64x1_t vqshl_s64(int64x1_t __p0, int64x1_t __p1) {
18961   __ret = (int64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
18965 __ai int64x1_t vqshl_s64(int64x1_t __p0, int64x1_t __p1) {
18967   __ret = (int64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
18973 __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
18975   __ret = (int16x4_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
18979 __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
18981   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18990 #define vqshlq_n_u8(__p0, __p1) __extension__ ({ \
18993   __ret = (uint8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 48); \
18997 #define vqshlq_n_u8(__p0, __p1) __extension__ ({ \
19001   __ret = (uint8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 48); \
19008 #define vqshlq_n_u32(__p0, __p1) __extension__ ({ \
19011   __ret = (uint32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 50); \
19015 #define vqshlq_n_u32(__p0, __p1) __extension__ ({ \
19019   __ret = (uint32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 50); \
19026 #define vqshlq_n_u64(__p0, __p1) __extension__ ({ \
19029   __ret = (uint64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 51); \
19033 #define vqshlq_n_u64(__p0, __p1) __extension__ ({ \
19037   __ret = (uint64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 51); \
19044 #define vqshlq_n_u16(__p0, __p1) __extension__ ({ \
19047   __ret = (uint16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 49); \
19051 #define vqshlq_n_u16(__p0, __p1) __extension__ ({ \
19055   __ret = (uint16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 49); \
19062 #define vqshlq_n_s8(__p0, __p1) __extension__ ({ \
19065   __ret = (int8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 32); \
19069 #define vqshlq_n_s8(__p0, __p1) __extension__ ({ \
19073   __ret = (int8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 32); \
19080 #define vqshlq_n_s32(__p0, __p1) __extension__ ({ \
19083   __ret = (int32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 34); \
19087 #define vqshlq_n_s32(__p0, __p1) __extension__ ({ \
19091   __ret = (int32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 34); \
19098 #define vqshlq_n_s64(__p0, __p1) __extension__ ({ \
19101   __ret = (int64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 35); \
19105 #define vqshlq_n_s64(__p0, __p1) __extension__ ({ \
19109   __ret = (int64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 35); \
19116 #define vqshlq_n_s16(__p0, __p1) __extension__ ({ \
19119   __ret = (int16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 33); \
19123 #define vqshlq_n_s16(__p0, __p1) __extension__ ({ \
19127   __ret = (int16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 33); \
19134 #define vqshl_n_u8(__p0, __p1) __extension__ ({ \
19137   __ret = (uint8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 16); \
19141 #define vqshl_n_u8(__p0, __p1) __extension__ ({ \
19145   __ret = (uint8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 16); \
19152 #define vqshl_n_u32(__p0, __p1) __extension__ ({ \
19155   __ret = (uint32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 18); \
19159 #define vqshl_n_u32(__p0, __p1) __extension__ ({ \
19163   __ret = (uint32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 18); \
19170 #define vqshl_n_u64(__p0, __p1) __extension__ ({ \
19173   __ret = (uint64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 19); \
19177 #define vqshl_n_u64(__p0, __p1) __extension__ ({ \
19180   __ret = (uint64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 19); \
19186 #define vqshl_n_u16(__p0, __p1) __extension__ ({ \
19189   __ret = (uint16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 17); \
19193 #define vqshl_n_u16(__p0, __p1) __extension__ ({ \
19197   __ret = (uint16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 17); \
19204 #define vqshl_n_s8(__p0, __p1) __extension__ ({ \
19207   __ret = (int8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 0); \
19211 #define vqshl_n_s8(__p0, __p1) __extension__ ({ \
19215   __ret = (int8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 0); \
19222 #define vqshl_n_s32(__p0, __p1) __extension__ ({ \
19225   __ret = (int32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 2); \
19229 #define vqshl_n_s32(__p0, __p1) __extension__ ({ \
19233   __ret = (int32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 2); \
19240 #define vqshl_n_s64(__p0, __p1) __extension__ ({ \
19243   __ret = (int64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 3); \
19247 #define vqshl_n_s64(__p0, __p1) __extension__ ({ \
19250   __ret = (int64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 3); \
19256 #define vqshl_n_s16(__p0, __p1) __extension__ ({ \
19259   __ret = (int16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 1); \
19263 #define vqshl_n_s16(__p0, __p1) __extension__ ({ \
19267   __ret = (int16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 1); \
19274 #define vqshluq_n_s8(__p0, __p1) __extension__ ({ \
19277   __ret = (uint8x16_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 48); \
19281 #define vqshluq_n_s8(__p0, __p1) __extension__ ({ \
19285   __ret = (uint8x16_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 48); \
19292 #define vqshluq_n_s32(__p0, __p1) __extension__ ({ \
19295   __ret = (uint32x4_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 50); \
19299 #define vqshluq_n_s32(__p0, __p1) __extension__ ({ \
19303   __ret = (uint32x4_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 50); \
19310 #define vqshluq_n_s64(__p0, __p1) __extension__ ({ \
19313   __ret = (uint64x2_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 51); \
19317 #define vqshluq_n_s64(__p0, __p1) __extension__ ({ \
19321   __ret = (uint64x2_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 51); \
19328 #define vqshluq_n_s16(__p0, __p1) __extension__ ({ \
19331   __ret = (uint16x8_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 49); \
19335 #define vqshluq_n_s16(__p0, __p1) __extension__ ({ \
19339   __ret = (uint16x8_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 49); \
19346 #define vqshlu_n_s8(__p0, __p1) __extension__ ({ \
19349   __ret = (uint8x8_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 16); \
19353 #define vqshlu_n_s8(__p0, __p1) __extension__ ({ \
19357   __ret = (uint8x8_t) __builtin_neon_vqshlu_n_v((int8x8_t)__rev0, __p1, 16); \
19364 #define vqshlu_n_s32(__p0, __p1) __extension__ ({ \
19367   __ret = (uint32x2_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 18); \
19371 #define vqshlu_n_s32(__p0, __p1) __extension__ ({ \
19375   __ret = (uint32x2_t) __builtin_neon_vqshlu_n_v((int8x8_t)__rev0, __p1, 18); \
19382 #define vqshlu_n_s64(__p0, __p1) __extension__ ({ \
19385   __ret = (uint64x1_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 19); \
19389 #define vqshlu_n_s64(__p0, __p1) __extension__ ({ \
19392   __ret = (uint64x1_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 19); \
19398 #define vqshlu_n_s16(__p0, __p1) __extension__ ({ \
19401   __ret = (uint16x4_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 17); \
19405 #define vqshlu_n_s16(__p0, __p1) __extension__ ({ \
19409   __ret = (uint16x4_t) __builtin_neon_vqshlu_n_v((int8x8_t)__rev0, __p1, 17); \
19416 #define vqshrn_n_u32(__p0, __p1) __extension__ ({ \
19419   __ret = (uint16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 17); \
19423 #define vqshrn_n_u32(__p0, __p1) __extension__ ({ \
19427   __ret = (uint16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 17); \
19431 #define __noswap_vqshrn_n_u32(__p0, __p1) __extension__ ({ \
19434   __ret = (uint16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 17); \
19440 #define vqshrn_n_u64(__p0, __p1) __extension__ ({ \
19443   __ret = (uint32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 18); \
19447 #define vqshrn_n_u64(__p0, __p1) __extension__ ({ \
19451   __ret = (uint32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 18); \
19455 #define __noswap_vqshrn_n_u64(__p0, __p1) __extension__ ({ \
19458   __ret = (uint32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 18); \
19464 #define vqshrn_n_u16(__p0, __p1) __extension__ ({ \
19467   __ret = (uint8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 16); \
19471 #define vqshrn_n_u16(__p0, __p1) __extension__ ({ \
19475   __ret = (uint8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 16); \
19479 #define __noswap_vqshrn_n_u16(__p0, __p1) __extension__ ({ \
19482   __ret = (uint8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 16); \
19488 #define vqshrn_n_s32(__p0, __p1) __extension__ ({ \
19491   __ret = (int16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 1); \
19495 #define vqshrn_n_s32(__p0, __p1) __extension__ ({ \
19499   __ret = (int16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 1); \
19503 #define __noswap_vqshrn_n_s32(__p0, __p1) __extension__ ({ \
19506   __ret = (int16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 1); \
19512 #define vqshrn_n_s64(__p0, __p1) __extension__ ({ \
19515   __ret = (int32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 2); \
19519 #define vqshrn_n_s64(__p0, __p1) __extension__ ({ \
19523   __ret = (int32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 2); \
19527 #define __noswap_vqshrn_n_s64(__p0, __p1) __extension__ ({ \
19530   __ret = (int32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 2); \
19536 #define vqshrn_n_s16(__p0, __p1) __extension__ ({ \
19539   __ret = (int8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 0); \
19543 #define vqshrn_n_s16(__p0, __p1) __extension__ ({ \
19547   __ret = (int8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 0); \
19551 #define __noswap_vqshrn_n_s16(__p0, __p1) __extension__ ({ \
19554   __ret = (int8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 0); \
19560 #define vqshrun_n_s32(__p0, __p1) __extension__ ({ \
19563   __ret = (uint16x4_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 17); \
19567 #define vqshrun_n_s32(__p0, __p1) __extension__ ({ \
19571   __ret = (uint16x4_t) __builtin_neon_vqshrun_n_v((int8x16_t)__rev0, __p1, 17); \
19575 #define __noswap_vqshrun_n_s32(__p0, __p1) __extension__ ({ \
19578   __ret = (uint16x4_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 17); \
19584 #define vqshrun_n_s64(__p0, __p1) __extension__ ({ \
19587   __ret = (uint32x2_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 18); \
19591 #define vqshrun_n_s64(__p0, __p1) __extension__ ({ \
19595   __ret = (uint32x2_t) __builtin_neon_vqshrun_n_v((int8x16_t)__rev0, __p1, 18); \
19599 #define __noswap_vqshrun_n_s64(__p0, __p1) __extension__ ({ \
19602   __ret = (uint32x2_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 18); \
19608 #define vqshrun_n_s16(__p0, __p1) __extension__ ({ \
19611   __ret = (uint8x8_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 16); \
19615 #define vqshrun_n_s16(__p0, __p1) __extension__ ({ \
19619   __ret = (uint8x8_t) __builtin_neon_vqshrun_n_v((int8x16_t)__rev0, __p1, 16); \
19623 #define __noswap_vqshrun_n_s16(__p0, __p1) __extension__ ({ \
19626   __ret = (uint8x8_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 16); \
19632 __ai uint8x16_t vqsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
19634   __ret = (uint8x16_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
19638 __ai uint8x16_t vqsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
19640   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
19649 __ai uint32x4_t vqsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
19651   __ret = (uint32x4_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
19655 __ai uint32x4_t vqsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
19657   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19666 __ai uint64x2_t vqsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
19668   __ret = (uint64x2_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
19672 __ai uint64x2_t vqsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
19674   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19683 __ai uint16x8_t vqsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
19685   __ret = (uint16x8_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
19689 __ai uint16x8_t vqsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
19691   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19700 __ai int8x16_t vqsubq_s8(int8x16_t __p0, int8x16_t __p1) {
19702   __ret = (int8x16_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
19706 __ai int8x16_t vqsubq_s8(int8x16_t __p0, int8x16_t __p1) {
19708   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
19717 __ai int32x4_t vqsubq_s32(int32x4_t __p0, int32x4_t __p1) {
19719   __ret = (int32x4_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
19723 __ai int32x4_t vqsubq_s32(int32x4_t __p0, int32x4_t __p1) {
19725   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19731 __ai int32x4_t __noswap_vqsubq_s32(int32x4_t __p0, int32x4_t __p1) {
19733   __ret = (int32x4_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
19739 __ai int64x2_t vqsubq_s64(int64x2_t __p0, int64x2_t __p1) {
19741   __ret = (int64x2_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
19745 __ai int64x2_t vqsubq_s64(int64x2_t __p0, int64x2_t __p1) {
19747   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19756 __ai int16x8_t vqsubq_s16(int16x8_t __p0, int16x8_t __p1) {
19758   __ret = (int16x8_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
19762 __ai int16x8_t vqsubq_s16(int16x8_t __p0, int16x8_t __p1) {
19764   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19770 __ai int16x8_t __noswap_vqsubq_s16(int16x8_t __p0, int16x8_t __p1) {
19772   __ret = (int16x8_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
19778 __ai uint8x8_t vqsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
19780   __ret = (uint8x8_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
19784 __ai uint8x8_t vqsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
19786   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19795 __ai uint32x2_t vqsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
19797   __ret = (uint32x2_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
19801 __ai uint32x2_t vqsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
19803   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19812 __ai uint64x1_t vqsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
19814   __ret = (uint64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
19818 __ai uint64x1_t vqsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
19820   __ret = (uint64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
19826 __ai uint16x4_t vqsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
19828   __ret = (uint16x4_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
19832 __ai uint16x4_t vqsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
19834   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19843 __ai int8x8_t vqsub_s8(int8x8_t __p0, int8x8_t __p1) {
19845   __ret = (int8x8_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
19849 __ai int8x8_t vqsub_s8(int8x8_t __p0, int8x8_t __p1) {
19851   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19860 __ai int32x2_t vqsub_s32(int32x2_t __p0, int32x2_t __p1) {
19862   __ret = (int32x2_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
19866 __ai int32x2_t vqsub_s32(int32x2_t __p0, int32x2_t __p1) {
19868   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19874 __ai int32x2_t __noswap_vqsub_s32(int32x2_t __p0, int32x2_t __p1) {
19876   __ret = (int32x2_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
19882 __ai int64x1_t vqsub_s64(int64x1_t __p0, int64x1_t __p1) {
19884   __ret = (int64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
19888 __ai int64x1_t vqsub_s64(int64x1_t __p0, int64x1_t __p1) {
19890   __ret = (int64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
19896 __ai int16x4_t vqsub_s16(int16x4_t __p0, int16x4_t __p1) {
19898   __ret = (int16x4_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
19902 __ai int16x4_t vqsub_s16(int16x4_t __p0, int16x4_t __p1) {
19904   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19910 __ai int16x4_t __noswap_vqsub_s16(int16x4_t __p0, int16x4_t __p1) {
19912   __ret = (int16x4_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
19918 __ai uint16x4_t vraddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
19920   __ret = (uint16x4_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
19924 __ai uint16x4_t vraddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
19926   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19932 __ai uint16x4_t __noswap_vraddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
19934   __ret = (uint16x4_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
19940 __ai uint32x2_t vraddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
19942   __ret = (uint32x2_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
19946 __ai uint32x2_t vraddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
19948   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19954 __ai uint32x2_t __noswap_vraddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
19956   __ret = (uint32x2_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
19962 __ai uint8x8_t vraddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
19964   __ret = (uint8x8_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
19968 __ai uint8x8_t vraddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
19970   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19976 __ai uint8x8_t __noswap_vraddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
19978   __ret = (uint8x8_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
19984 __ai int16x4_t vraddhn_s32(int32x4_t __p0, int32x4_t __p1) {
19986   __ret = (int16x4_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
19990 __ai int16x4_t vraddhn_s32(int32x4_t __p0, int32x4_t __p1) {
19992   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19998 __ai int16x4_t __noswap_vraddhn_s32(int32x4_t __p0, int32x4_t __p1) {
20000   __ret = (int16x4_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
20006 __ai int32x2_t vraddhn_s64(int64x2_t __p0, int64x2_t __p1) {
20008   __ret = (int32x2_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
20012 __ai int32x2_t vraddhn_s64(int64x2_t __p0, int64x2_t __p1) {
20014   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20020 __ai int32x2_t __noswap_vraddhn_s64(int64x2_t __p0, int64x2_t __p1) {
20022   __ret = (int32x2_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
20028 __ai int8x8_t vraddhn_s16(int16x8_t __p0, int16x8_t __p1) {
20030   __ret = (int8x8_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
20034 __ai int8x8_t vraddhn_s16(int16x8_t __p0, int16x8_t __p1) {
20036   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20042 __ai int8x8_t __noswap_vraddhn_s16(int16x8_t __p0, int16x8_t __p1) {
20044   __ret = (int8x8_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
20114 __ai float32x4_t vrecpsq_f32(float32x4_t __p0, float32x4_t __p1) {
20116   __ret = (float32x4_t) __builtin_neon_vrecpsq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
20120 __ai float32x4_t vrecpsq_f32(float32x4_t __p0, float32x4_t __p1) {
20122   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20131 __ai float32x2_t vrecps_f32(float32x2_t __p0, float32x2_t __p1) {
20133   __ret = (float32x2_t) __builtin_neon_vrecps_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
20137 __ai float32x2_t vrecps_f32(float32x2_t __p0, float32x2_t __p1) {
20139   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20724 __ai uint8x16_t vrhaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
20726   __ret = (uint8x16_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
20730 __ai uint8x16_t vrhaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
20732   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20741 __ai uint32x4_t vrhaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
20743   __ret = (uint32x4_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
20747 __ai uint32x4_t vrhaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
20749   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20758 __ai uint16x8_t vrhaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
20760   __ret = (uint16x8_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
20764 __ai uint16x8_t vrhaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
20766   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20775 __ai int8x16_t vrhaddq_s8(int8x16_t __p0, int8x16_t __p1) {
20777   __ret = (int8x16_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
20781 __ai int8x16_t vrhaddq_s8(int8x16_t __p0, int8x16_t __p1) {
20783   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20792 __ai int32x4_t vrhaddq_s32(int32x4_t __p0, int32x4_t __p1) {
20794   __ret = (int32x4_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
20798 __ai int32x4_t vrhaddq_s32(int32x4_t __p0, int32x4_t __p1) {
20800   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20809 __ai int16x8_t vrhaddq_s16(int16x8_t __p0, int16x8_t __p1) {
20811   __ret = (int16x8_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
20815 __ai int16x8_t vrhaddq_s16(int16x8_t __p0, int16x8_t __p1) {
20817   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20826 __ai uint8x8_t vrhadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
20828   __ret = (uint8x8_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
20832 __ai uint8x8_t vrhadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
20834   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20843 __ai uint32x2_t vrhadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
20845   __ret = (uint32x2_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
20849 __ai uint32x2_t vrhadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
20851   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20860 __ai uint16x4_t vrhadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
20862   __ret = (uint16x4_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
20866 __ai uint16x4_t vrhadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
20868   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20877 __ai int8x8_t vrhadd_s8(int8x8_t __p0, int8x8_t __p1) {
20879   __ret = (int8x8_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
20883 __ai int8x8_t vrhadd_s8(int8x8_t __p0, int8x8_t __p1) {
20885   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20894 __ai int32x2_t vrhadd_s32(int32x2_t __p0, int32x2_t __p1) {
20896   __ret = (int32x2_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
20900 __ai int32x2_t vrhadd_s32(int32x2_t __p0, int32x2_t __p1) {
20902   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20911 __ai int16x4_t vrhadd_s16(int16x4_t __p0, int16x4_t __p1) {
20913   __ret = (int16x4_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
20917 __ai int16x4_t vrhadd_s16(int16x4_t __p0, int16x4_t __p1) {
20919   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20928 __ai uint8x16_t vrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
20930   __ret = (uint8x16_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
20934 __ai uint8x16_t vrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
20936   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20945 __ai uint32x4_t vrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
20947   __ret = (uint32x4_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
20951 __ai uint32x4_t vrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
20953   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20962 __ai uint64x2_t vrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
20964   __ret = (uint64x2_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
20968 __ai uint64x2_t vrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
20970   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20979 __ai uint16x8_t vrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
20981   __ret = (uint16x8_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
20985 __ai uint16x8_t vrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
20987   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20996 __ai int8x16_t vrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
20998   __ret = (int8x16_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
21002 __ai int8x16_t vrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
21004   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
21013 __ai int32x4_t vrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
21015   __ret = (int32x4_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
21019 __ai int32x4_t vrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
21021   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21030 __ai int64x2_t vrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
21032   __ret = (int64x2_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
21036 __ai int64x2_t vrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
21038   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21047 __ai int16x8_t vrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
21049   __ret = (int16x8_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
21053 __ai int16x8_t vrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
21055   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21064 __ai uint8x8_t vrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
21066   __ret = (uint8x8_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
21070 __ai uint8x8_t vrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
21072   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21081 __ai uint32x2_t vrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
21083   __ret = (uint32x2_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
21087 __ai uint32x2_t vrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
21089   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21098 __ai uint64x1_t vrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
21100   __ret = (uint64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
21104 __ai uint64x1_t vrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
21106   __ret = (uint64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
21112 __ai uint16x4_t vrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
21114   __ret = (uint16x4_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
21118 __ai uint16x4_t vrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
21120   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21129 __ai int8x8_t vrshl_s8(int8x8_t __p0, int8x8_t __p1) {
21131   __ret = (int8x8_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
21135 __ai int8x8_t vrshl_s8(int8x8_t __p0, int8x8_t __p1) {
21137   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21146 __ai int32x2_t vrshl_s32(int32x2_t __p0, int32x2_t __p1) {
21148   __ret = (int32x2_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
21152 __ai int32x2_t vrshl_s32(int32x2_t __p0, int32x2_t __p1) {
21154   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21163 __ai int64x1_t vrshl_s64(int64x1_t __p0, int64x1_t __p1) {
21165   __ret = (int64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
21169 __ai int64x1_t vrshl_s64(int64x1_t __p0, int64x1_t __p1) {
21171   __ret = (int64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
21177 __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
21179   __ret = (int16x4_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
21183 __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
21185   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21194 #define vrshrq_n_u8(__p0, __p1) __extension__ ({ \
21197   __ret = (uint8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 48); \
21201 #define vrshrq_n_u8(__p0, __p1) __extension__ ({ \
21205   __ret = (uint8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 48); \
21212 #define vrshrq_n_u32(__p0, __p1) __extension__ ({ \
21215   __ret = (uint32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 50); \
21219 #define vrshrq_n_u32(__p0, __p1) __extension__ ({ \
21223   __ret = (uint32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 50); \
21230 #define vrshrq_n_u64(__p0, __p1) __extension__ ({ \
21233   __ret = (uint64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 51); \
21237 #define vrshrq_n_u64(__p0, __p1) __extension__ ({ \
21241   __ret = (uint64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 51); \
21248 #define vrshrq_n_u16(__p0, __p1) __extension__ ({ \
21251   __ret = (uint16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 49); \
21255 #define vrshrq_n_u16(__p0, __p1) __extension__ ({ \
21259   __ret = (uint16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 49); \
21266 #define vrshrq_n_s8(__p0, __p1) __extension__ ({ \
21269   __ret = (int8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 32); \
21273 #define vrshrq_n_s8(__p0, __p1) __extension__ ({ \
21277   __ret = (int8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 32); \
21284 #define vrshrq_n_s32(__p0, __p1) __extension__ ({ \
21287   __ret = (int32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 34); \
21291 #define vrshrq_n_s32(__p0, __p1) __extension__ ({ \
21295   __ret = (int32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 34); \
21302 #define vrshrq_n_s64(__p0, __p1) __extension__ ({ \
21305   __ret = (int64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 35); \
21309 #define vrshrq_n_s64(__p0, __p1) __extension__ ({ \
21313   __ret = (int64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 35); \
21320 #define vrshrq_n_s16(__p0, __p1) __extension__ ({ \
21323   __ret = (int16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 33); \
21327 #define vrshrq_n_s16(__p0, __p1) __extension__ ({ \
21331   __ret = (int16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 33); \
21338 #define vrshr_n_u8(__p0, __p1) __extension__ ({ \
21341   __ret = (uint8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 16); \
21345 #define vrshr_n_u8(__p0, __p1) __extension__ ({ \
21349   __ret = (uint8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 16); \
21356 #define vrshr_n_u32(__p0, __p1) __extension__ ({ \
21359   __ret = (uint32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 18); \
21363 #define vrshr_n_u32(__p0, __p1) __extension__ ({ \
21367   __ret = (uint32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 18); \
21374 #define vrshr_n_u64(__p0, __p1) __extension__ ({ \
21377   __ret = (uint64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 19); \
21381 #define vrshr_n_u64(__p0, __p1) __extension__ ({ \
21384   __ret = (uint64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 19); \
21390 #define vrshr_n_u16(__p0, __p1) __extension__ ({ \
21393   __ret = (uint16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 17); \
21397 #define vrshr_n_u16(__p0, __p1) __extension__ ({ \
21401   __ret = (uint16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 17); \
21408 #define vrshr_n_s8(__p0, __p1) __extension__ ({ \
21411   __ret = (int8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 0); \
21415 #define vrshr_n_s8(__p0, __p1) __extension__ ({ \
21419   __ret = (int8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 0); \
21426 #define vrshr_n_s32(__p0, __p1) __extension__ ({ \
21429   __ret = (int32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 2); \
21433 #define vrshr_n_s32(__p0, __p1) __extension__ ({ \
21437   __ret = (int32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 2); \
21444 #define vrshr_n_s64(__p0, __p1) __extension__ ({ \
21447   __ret = (int64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 3); \
21451 #define vrshr_n_s64(__p0, __p1) __extension__ ({ \
21454   __ret = (int64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 3); \
21460 #define vrshr_n_s16(__p0, __p1) __extension__ ({ \
21463   __ret = (int16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 1); \
21467 #define vrshr_n_s16(__p0, __p1) __extension__ ({ \
21471   __ret = (int16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 1); \
21478 #define vrshrn_n_u32(__p0, __p1) __extension__ ({ \
21481   __ret = (uint16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 17); \
21485 #define vrshrn_n_u32(__p0, __p1) __extension__ ({ \
21489   __ret = (uint16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 17); \
21493 #define __noswap_vrshrn_n_u32(__p0, __p1) __extension__ ({ \
21496   __ret = (uint16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 17); \
21502 #define vrshrn_n_u64(__p0, __p1) __extension__ ({ \
21505   __ret = (uint32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 18); \
21509 #define vrshrn_n_u64(__p0, __p1) __extension__ ({ \
21513   __ret = (uint32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 18); \
21517 #define __noswap_vrshrn_n_u64(__p0, __p1) __extension__ ({ \
21520   __ret = (uint32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 18); \
21526 #define vrshrn_n_u16(__p0, __p1) __extension__ ({ \
21529   __ret = (uint8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 16); \
21533 #define vrshrn_n_u16(__p0, __p1) __extension__ ({ \
21537   __ret = (uint8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 16); \
21541 #define __noswap_vrshrn_n_u16(__p0, __p1) __extension__ ({ \
21544   __ret = (uint8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 16); \
21550 #define vrshrn_n_s32(__p0, __p1) __extension__ ({ \
21553   __ret = (int16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 1); \
21557 #define vrshrn_n_s32(__p0, __p1) __extension__ ({ \
21561   __ret = (int16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 1); \
21565 #define __noswap_vrshrn_n_s32(__p0, __p1) __extension__ ({ \
21568   __ret = (int16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 1); \
21574 #define vrshrn_n_s64(__p0, __p1) __extension__ ({ \
21577   __ret = (int32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 2); \
21581 #define vrshrn_n_s64(__p0, __p1) __extension__ ({ \
21585   __ret = (int32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 2); \
21589 #define __noswap_vrshrn_n_s64(__p0, __p1) __extension__ ({ \
21592   __ret = (int32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 2); \
21598 #define vrshrn_n_s16(__p0, __p1) __extension__ ({ \
21601   __ret = (int8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 0); \
21605 #define vrshrn_n_s16(__p0, __p1) __extension__ ({ \
21609   __ret = (int8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 0); \
21613 #define __noswap_vrshrn_n_s16(__p0, __p1) __extension__ ({ \
21616   __ret = (int8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 0); \
21686 __ai float32x4_t vrsqrtsq_f32(float32x4_t __p0, float32x4_t __p1) {
21688   __ret = (float32x4_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
21692 __ai float32x4_t vrsqrtsq_f32(float32x4_t __p0, float32x4_t __p1) {
21694   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21703 __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
21705   __ret = (float32x2_t) __builtin_neon_vrsqrts_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
21709 __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
21711   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21720 #define vrsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
21722   uint8x16_t __s1 = __p1; \
21728 #define vrsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
21730   uint8x16_t __s1 = __p1; \
21741 #define vrsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
21743   uint32x4_t __s1 = __p1; \
21749 #define vrsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
21751   uint32x4_t __s1 = __p1; \
21762 #define vrsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
21764   uint64x2_t __s1 = __p1; \
21770 #define vrsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
21772   uint64x2_t __s1 = __p1; \
21783 #define vrsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
21785   uint16x8_t __s1 = __p1; \
21791 #define vrsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
21793   uint16x8_t __s1 = __p1; \
21804 #define vrsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
21806   int8x16_t __s1 = __p1; \
21812 #define vrsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
21814   int8x16_t __s1 = __p1; \
21825 #define vrsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
21827   int32x4_t __s1 = __p1; \
21833 #define vrsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
21835   int32x4_t __s1 = __p1; \
21846 #define vrsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
21848   int64x2_t __s1 = __p1; \
21854 #define vrsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
21856   int64x2_t __s1 = __p1; \
21867 #define vrsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
21869   int16x8_t __s1 = __p1; \
21875 #define vrsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
21877   int16x8_t __s1 = __p1; \
21888 #define vrsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
21890   uint8x8_t __s1 = __p1; \
21896 #define vrsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
21898   uint8x8_t __s1 = __p1; \
21909 #define vrsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
21911   uint32x2_t __s1 = __p1; \
21917 #define vrsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
21919   uint32x2_t __s1 = __p1; \
21930 #define vrsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
21932   uint64x1_t __s1 = __p1; \
21938 #define vrsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
21940   uint64x1_t __s1 = __p1; \
21948 #define vrsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
21950   uint16x4_t __s1 = __p1; \
21956 #define vrsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
21958   uint16x4_t __s1 = __p1; \
21969 #define vrsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
21971   int8x8_t __s1 = __p1; \
21977 #define vrsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
21979   int8x8_t __s1 = __p1; \
21990 #define vrsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
21992   int32x2_t __s1 = __p1; \
21998 #define vrsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
22000   int32x2_t __s1 = __p1; \
22011 #define vrsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
22013   int64x1_t __s1 = __p1; \
22019 #define vrsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
22021   int64x1_t __s1 = __p1; \
22029 #define vrsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
22031   int16x4_t __s1 = __p1; \
22037 #define vrsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
22039   int16x4_t __s1 = __p1; \
22050 __ai uint16x4_t vrsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
22052   __ret = (uint16x4_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
22056 __ai uint16x4_t vrsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
22058   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22064 __ai uint16x4_t __noswap_vrsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
22066   __ret = (uint16x4_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
22072 __ai uint32x2_t vrsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
22074   __ret = (uint32x2_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
22078 __ai uint32x2_t vrsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
22080   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22086 __ai uint32x2_t __noswap_vrsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
22088   __ret = (uint32x2_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
22094 __ai uint8x8_t vrsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
22096   __ret = (uint8x8_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
22100 __ai uint8x8_t vrsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
22102   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22108 __ai uint8x8_t __noswap_vrsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
22110   __ret = (uint8x8_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
22116 __ai int16x4_t vrsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
22118   __ret = (int16x4_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
22122 __ai int16x4_t vrsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
22124   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22130 __ai int16x4_t __noswap_vrsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
22132   __ret = (int16x4_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
22138 __ai int32x2_t vrsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
22140   __ret = (int32x2_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
22144 __ai int32x2_t vrsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
22146   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22152 __ai int32x2_t __noswap_vrsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
22154   __ret = (int32x2_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
22160 __ai int8x8_t vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
22162   __ret = (int8x8_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
22166 __ai int8x8_t vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
22168   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22174 __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
22176   __ret = (int8x8_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
22182 #define vset_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22184   poly8x8_t __s1 = __p1; \
22190 #define vset_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22192   poly8x8_t __s1 = __p1; \
22199 #define __noswap_vset_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22201   poly8x8_t __s1 = __p1; \
22209 #define vset_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22211   poly16x4_t __s1 = __p1; \
22217 #define vset_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22219   poly16x4_t __s1 = __p1; \
22226 #define __noswap_vset_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22228   poly16x4_t __s1 = __p1; \
22236 #define vsetq_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22238   poly8x16_t __s1 = __p1; \
22244 #define vsetq_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22246   poly8x16_t __s1 = __p1; \
22253 #define __noswap_vsetq_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22255   poly8x16_t __s1 = __p1; \
22263 #define vsetq_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22265   poly16x8_t __s1 = __p1; \
22271 #define vsetq_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22273   poly16x8_t __s1 = __p1; \
22280 #define __noswap_vsetq_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22282   poly16x8_t __s1 = __p1; \
22290 #define vsetq_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22292   uint8x16_t __s1 = __p1; \
22298 #define vsetq_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22300   uint8x16_t __s1 = __p1; \
22307 #define __noswap_vsetq_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22309   uint8x16_t __s1 = __p1; \
22317 #define vsetq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22319   uint32x4_t __s1 = __p1; \
22325 #define vsetq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22327   uint32x4_t __s1 = __p1; \
22334 #define __noswap_vsetq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22336   uint32x4_t __s1 = __p1; \
22344 #define vsetq_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22346   uint64x2_t __s1 = __p1; \
22352 #define vsetq_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22354   uint64x2_t __s1 = __p1; \
22361 #define __noswap_vsetq_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22363   uint64x2_t __s1 = __p1; \
22371 #define vsetq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22373   uint16x8_t __s1 = __p1; \
22379 #define vsetq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22381   uint16x8_t __s1 = __p1; \
22388 #define __noswap_vsetq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22390   uint16x8_t __s1 = __p1; \
22398 #define vsetq_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22400   int8x16_t __s1 = __p1; \
22406 #define vsetq_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22408   int8x16_t __s1 = __p1; \
22415 #define __noswap_vsetq_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22417   int8x16_t __s1 = __p1; \
22425 #define vsetq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22427   float32x4_t __s1 = __p1; \
22433 #define vsetq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22435   float32x4_t __s1 = __p1; \
22442 #define __noswap_vsetq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22444   float32x4_t __s1 = __p1; \
22452 #define vsetq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22454   int32x4_t __s1 = __p1; \
22460 #define vsetq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22462   int32x4_t __s1 = __p1; \
22469 #define __noswap_vsetq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22471   int32x4_t __s1 = __p1; \
22479 #define vsetq_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22481   int64x2_t __s1 = __p1; \
22487 #define vsetq_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22489   int64x2_t __s1 = __p1; \
22496 #define __noswap_vsetq_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22498   int64x2_t __s1 = __p1; \
22506 #define vsetq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22508   int16x8_t __s1 = __p1; \
22514 #define vsetq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22516   int16x8_t __s1 = __p1; \
22523 #define __noswap_vsetq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22525   int16x8_t __s1 = __p1; \
22533 #define vset_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22535   uint8x8_t __s1 = __p1; \
22541 #define vset_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22543   uint8x8_t __s1 = __p1; \
22550 #define __noswap_vset_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22552   uint8x8_t __s1 = __p1; \
22560 #define vset_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22562   uint32x2_t __s1 = __p1; \
22568 #define vset_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22570   uint32x2_t __s1 = __p1; \
22577 #define __noswap_vset_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22579   uint32x2_t __s1 = __p1; \
22587 #define vset_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22589   uint64x1_t __s1 = __p1; \
22595 #define vset_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22597   uint64x1_t __s1 = __p1; \
22602 #define __noswap_vset_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22604   uint64x1_t __s1 = __p1; \
22612 #define vset_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22614   uint16x4_t __s1 = __p1; \
22620 #define vset_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22622   uint16x4_t __s1 = __p1; \
22629 #define __noswap_vset_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22631   uint16x4_t __s1 = __p1; \
22639 #define vset_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22641   int8x8_t __s1 = __p1; \
22647 #define vset_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22649   int8x8_t __s1 = __p1; \
22656 #define __noswap_vset_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22658   int8x8_t __s1 = __p1; \
22666 #define vset_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22668   float32x2_t __s1 = __p1; \
22674 #define vset_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22676   float32x2_t __s1 = __p1; \
22683 #define __noswap_vset_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22685   float32x2_t __s1 = __p1; \
22693 #define vset_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22695   int32x2_t __s1 = __p1; \
22701 #define vset_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22703   int32x2_t __s1 = __p1; \
22710 #define __noswap_vset_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22712   int32x2_t __s1 = __p1; \
22720 #define vset_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22722   int64x1_t __s1 = __p1; \
22728 #define vset_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22730   int64x1_t __s1 = __p1; \
22735 #define __noswap_vset_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22737   int64x1_t __s1 = __p1; \
22745 #define vset_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22747   int16x4_t __s1 = __p1; \
22753 #define vset_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22755   int16x4_t __s1 = __p1; \
22762 #define __noswap_vset_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22764   int16x4_t __s1 = __p1; \
22772 __ai uint8x16_t vshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
22774   __ret = (uint8x16_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
22778 __ai uint8x16_t vshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
22780   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
22789 __ai uint32x4_t vshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
22791   __ret = (uint32x4_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
22795 __ai uint32x4_t vshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
22797   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22806 __ai uint64x2_t vshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
22808   __ret = (uint64x2_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
22812 __ai uint64x2_t vshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
22814   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22823 __ai uint16x8_t vshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
22825   __ret = (uint16x8_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
22829 __ai uint16x8_t vshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
22831   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22840 __ai int8x16_t vshlq_s8(int8x16_t __p0, int8x16_t __p1) {
22842   __ret = (int8x16_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
22846 __ai int8x16_t vshlq_s8(int8x16_t __p0, int8x16_t __p1) {
22848   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
22857 __ai int32x4_t vshlq_s32(int32x4_t __p0, int32x4_t __p1) {
22859   __ret = (int32x4_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
22863 __ai int32x4_t vshlq_s32(int32x4_t __p0, int32x4_t __p1) {
22865   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22874 __ai int64x2_t vshlq_s64(int64x2_t __p0, int64x2_t __p1) {
22876   __ret = (int64x2_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
22880 __ai int64x2_t vshlq_s64(int64x2_t __p0, int64x2_t __p1) {
22882   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22891 __ai int16x8_t vshlq_s16(int16x8_t __p0, int16x8_t __p1) {
22893   __ret = (int16x8_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
22897 __ai int16x8_t vshlq_s16(int16x8_t __p0, int16x8_t __p1) {
22899   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22908 __ai uint8x8_t vshl_u8(uint8x8_t __p0, int8x8_t __p1) {
22910   __ret = (uint8x8_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
22914 __ai uint8x8_t vshl_u8(uint8x8_t __p0, int8x8_t __p1) {
22916   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22925 __ai uint32x2_t vshl_u32(uint32x2_t __p0, int32x2_t __p1) {
22927   __ret = (uint32x2_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
22931 __ai uint32x2_t vshl_u32(uint32x2_t __p0, int32x2_t __p1) {
22933   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22942 __ai uint64x1_t vshl_u64(uint64x1_t __p0, int64x1_t __p1) {
22944   __ret = (uint64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
22948 __ai uint64x1_t vshl_u64(uint64x1_t __p0, int64x1_t __p1) {
22950   __ret = (uint64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
22956 __ai uint16x4_t vshl_u16(uint16x4_t __p0, int16x4_t __p1) {
22958   __ret = (uint16x4_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
22962 __ai uint16x4_t vshl_u16(uint16x4_t __p0, int16x4_t __p1) {
22964   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22973 __ai int8x8_t vshl_s8(int8x8_t __p0, int8x8_t __p1) {
22975   __ret = (int8x8_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
22979 __ai int8x8_t vshl_s8(int8x8_t __p0, int8x8_t __p1) {
22981   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22990 __ai int32x2_t vshl_s32(int32x2_t __p0, int32x2_t __p1) {
22992   __ret = (int32x2_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
22996 __ai int32x2_t vshl_s32(int32x2_t __p0, int32x2_t __p1) {
22998   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
23007 __ai int64x1_t vshl_s64(int64x1_t __p0, int64x1_t __p1) {
23009   __ret = (int64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
23013 __ai int64x1_t vshl_s64(int64x1_t __p0, int64x1_t __p1) {
23015   __ret = (int64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
23021 __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
23023   __ret = (int16x4_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
23027 __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
23029   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
23038 #define vshlq_n_u8(__p0, __p1) __extension__ ({ \
23041   __ret = (uint8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 48); \
23045 #define vshlq_n_u8(__p0, __p1) __extension__ ({ \
23049   __ret = (uint8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 48); \
23056 #define vshlq_n_u32(__p0, __p1) __extension__ ({ \
23059   __ret = (uint32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 50); \
23063 #define vshlq_n_u32(__p0, __p1) __extension__ ({ \
23067   __ret = (uint32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 50); \
23074 #define vshlq_n_u64(__p0, __p1) __extension__ ({ \
23077   __ret = (uint64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 51); \
23081 #define vshlq_n_u64(__p0, __p1) __extension__ ({ \
23085   __ret = (uint64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 51); \
23092 #define vshlq_n_u16(__p0, __p1) __extension__ ({ \
23095   __ret = (uint16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 49); \
23099 #define vshlq_n_u16(__p0, __p1) __extension__ ({ \
23103   __ret = (uint16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 49); \
23110 #define vshlq_n_s8(__p0, __p1) __extension__ ({ \
23113   __ret = (int8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 32); \
23117 #define vshlq_n_s8(__p0, __p1) __extension__ ({ \
23121   __ret = (int8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 32); \
23128 #define vshlq_n_s32(__p0, __p1) __extension__ ({ \
23131   __ret = (int32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 34); \
23135 #define vshlq_n_s32(__p0, __p1) __extension__ ({ \
23139   __ret = (int32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 34); \
23146 #define vshlq_n_s64(__p0, __p1) __extension__ ({ \
23149   __ret = (int64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 35); \
23153 #define vshlq_n_s64(__p0, __p1) __extension__ ({ \
23157   __ret = (int64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 35); \
23164 #define vshlq_n_s16(__p0, __p1) __extension__ ({ \
23167   __ret = (int16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 33); \
23171 #define vshlq_n_s16(__p0, __p1) __extension__ ({ \
23175   __ret = (int16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 33); \
23182 #define vshl_n_u8(__p0, __p1) __extension__ ({ \
23185   __ret = (uint8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 16); \
23189 #define vshl_n_u8(__p0, __p1) __extension__ ({ \
23193   __ret = (uint8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 16); \
23200 #define vshl_n_u32(__p0, __p1) __extension__ ({ \
23203   __ret = (uint32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 18); \
23207 #define vshl_n_u32(__p0, __p1) __extension__ ({ \
23211   __ret = (uint32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 18); \
23218 #define vshl_n_u64(__p0, __p1) __extension__ ({ \
23221   __ret = (uint64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 19); \
23225 #define vshl_n_u64(__p0, __p1) __extension__ ({ \
23228   __ret = (uint64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 19); \
23234 #define vshl_n_u16(__p0, __p1) __extension__ ({ \
23237   __ret = (uint16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 17); \
23241 #define vshl_n_u16(__p0, __p1) __extension__ ({ \
23245   __ret = (uint16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 17); \
23252 #define vshl_n_s8(__p0, __p1) __extension__ ({ \
23255   __ret = (int8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 0); \
23259 #define vshl_n_s8(__p0, __p1) __extension__ ({ \
23263   __ret = (int8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 0); \
23270 #define vshl_n_s32(__p0, __p1) __extension__ ({ \
23273   __ret = (int32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 2); \
23277 #define vshl_n_s32(__p0, __p1) __extension__ ({ \
23281   __ret = (int32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 2); \
23288 #define vshl_n_s64(__p0, __p1) __extension__ ({ \
23291   __ret = (int64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 3); \
23295 #define vshl_n_s64(__p0, __p1) __extension__ ({ \
23298   __ret = (int64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 3); \
23304 #define vshl_n_s16(__p0, __p1) __extension__ ({ \
23307   __ret = (int16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 1); \
23311 #define vshl_n_s16(__p0, __p1) __extension__ ({ \
23315   __ret = (int16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 1); \
23322 #define vshll_n_u8(__p0, __p1) __extension__ ({ \
23325   __ret = (uint16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 49); \
23329 #define vshll_n_u8(__p0, __p1) __extension__ ({ \
23333   __ret = (uint16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 49); \
23337 #define __noswap_vshll_n_u8(__p0, __p1) __extension__ ({ \
23340   __ret = (uint16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 49); \
23346 #define vshll_n_u32(__p0, __p1) __extension__ ({ \
23349   __ret = (uint64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 51); \
23353 #define vshll_n_u32(__p0, __p1) __extension__ ({ \
23357   __ret = (uint64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 51); \
23361 #define __noswap_vshll_n_u32(__p0, __p1) __extension__ ({ \
23364   __ret = (uint64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 51); \
23370 #define vshll_n_u16(__p0, __p1) __extension__ ({ \
23373   __ret = (uint32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 50); \
23377 #define vshll_n_u16(__p0, __p1) __extension__ ({ \
23381   __ret = (uint32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 50); \
23385 #define __noswap_vshll_n_u16(__p0, __p1) __extension__ ({ \
23388   __ret = (uint32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 50); \
23394 #define vshll_n_s8(__p0, __p1) __extension__ ({ \
23397   __ret = (int16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 33); \
23401 #define vshll_n_s8(__p0, __p1) __extension__ ({ \
23405   __ret = (int16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 33); \
23409 #define __noswap_vshll_n_s8(__p0, __p1) __extension__ ({ \
23412   __ret = (int16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 33); \
23418 #define vshll_n_s32(__p0, __p1) __extension__ ({ \
23421   __ret = (int64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 35); \
23425 #define vshll_n_s32(__p0, __p1) __extension__ ({ \
23429   __ret = (int64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 35); \
23433 #define __noswap_vshll_n_s32(__p0, __p1) __extension__ ({ \
23436   __ret = (int64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 35); \
23442 #define vshll_n_s16(__p0, __p1) __extension__ ({ \
23445   __ret = (int32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 34); \
23449 #define vshll_n_s16(__p0, __p1) __extension__ ({ \
23453   __ret = (int32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 34); \
23457 #define __noswap_vshll_n_s16(__p0, __p1) __extension__ ({ \
23460   __ret = (int32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 34); \
23466 #define vshrq_n_u8(__p0, __p1) __extension__ ({ \
23469   __ret = (uint8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 48); \
23473 #define vshrq_n_u8(__p0, __p1) __extension__ ({ \
23477   __ret = (uint8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 48); \
23484 #define vshrq_n_u32(__p0, __p1) __extension__ ({ \
23487   __ret = (uint32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 50); \
23491 #define vshrq_n_u32(__p0, __p1) __extension__ ({ \
23495   __ret = (uint32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 50); \
23502 #define vshrq_n_u64(__p0, __p1) __extension__ ({ \
23505   __ret = (uint64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 51); \
23509 #define vshrq_n_u64(__p0, __p1) __extension__ ({ \
23513   __ret = (uint64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 51); \
23520 #define vshrq_n_u16(__p0, __p1) __extension__ ({ \
23523   __ret = (uint16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 49); \
23527 #define vshrq_n_u16(__p0, __p1) __extension__ ({ \
23531   __ret = (uint16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 49); \
23538 #define vshrq_n_s8(__p0, __p1) __extension__ ({ \
23541   __ret = (int8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 32); \
23545 #define vshrq_n_s8(__p0, __p1) __extension__ ({ \
23549   __ret = (int8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 32); \
23556 #define vshrq_n_s32(__p0, __p1) __extension__ ({ \
23559   __ret = (int32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 34); \
23563 #define vshrq_n_s32(__p0, __p1) __extension__ ({ \
23567   __ret = (int32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 34); \
23574 #define vshrq_n_s64(__p0, __p1) __extension__ ({ \
23577   __ret = (int64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 35); \
23581 #define vshrq_n_s64(__p0, __p1) __extension__ ({ \
23585   __ret = (int64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 35); \
23592 #define vshrq_n_s16(__p0, __p1) __extension__ ({ \
23595   __ret = (int16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 33); \
23599 #define vshrq_n_s16(__p0, __p1) __extension__ ({ \
23603   __ret = (int16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 33); \
23610 #define vshr_n_u8(__p0, __p1) __extension__ ({ \
23613   __ret = (uint8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 16); \
23617 #define vshr_n_u8(__p0, __p1) __extension__ ({ \
23621   __ret = (uint8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 16); \
23628 #define vshr_n_u32(__p0, __p1) __extension__ ({ \
23631   __ret = (uint32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 18); \
23635 #define vshr_n_u32(__p0, __p1) __extension__ ({ \
23639   __ret = (uint32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 18); \
23646 #define vshr_n_u64(__p0, __p1) __extension__ ({ \
23649   __ret = (uint64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 19); \
23653 #define vshr_n_u64(__p0, __p1) __extension__ ({ \
23656   __ret = (uint64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 19); \
23662 #define vshr_n_u16(__p0, __p1) __extension__ ({ \
23665   __ret = (uint16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 17); \
23669 #define vshr_n_u16(__p0, __p1) __extension__ ({ \
23673   __ret = (uint16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 17); \
23680 #define vshr_n_s8(__p0, __p1) __extension__ ({ \
23683   __ret = (int8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 0); \
23687 #define vshr_n_s8(__p0, __p1) __extension__ ({ \
23691   __ret = (int8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 0); \
23698 #define vshr_n_s32(__p0, __p1) __extension__ ({ \
23701   __ret = (int32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 2); \
23705 #define vshr_n_s32(__p0, __p1) __extension__ ({ \
23709   __ret = (int32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 2); \
23716 #define vshr_n_s64(__p0, __p1) __extension__ ({ \
23719   __ret = (int64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 3); \
23723 #define vshr_n_s64(__p0, __p1) __extension__ ({ \
23726   __ret = (int64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 3); \
23732 #define vshr_n_s16(__p0, __p1) __extension__ ({ \
23735   __ret = (int16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 1); \
23739 #define vshr_n_s16(__p0, __p1) __extension__ ({ \
23743   __ret = (int16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 1); \
23750 #define vshrn_n_u32(__p0, __p1) __extension__ ({ \
23753   __ret = (uint16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 17); \
23757 #define vshrn_n_u32(__p0, __p1) __extension__ ({ \
23761   __ret = (uint16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 17); \
23765 #define __noswap_vshrn_n_u32(__p0, __p1) __extension__ ({ \
23768   __ret = (uint16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 17); \
23774 #define vshrn_n_u64(__p0, __p1) __extension__ ({ \
23777   __ret = (uint32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 18); \
23781 #define vshrn_n_u64(__p0, __p1) __extension__ ({ \
23785   __ret = (uint32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 18); \
23789 #define __noswap_vshrn_n_u64(__p0, __p1) __extension__ ({ \
23792   __ret = (uint32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 18); \
23798 #define vshrn_n_u16(__p0, __p1) __extension__ ({ \
23801   __ret = (uint8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 16); \
23805 #define vshrn_n_u16(__p0, __p1) __extension__ ({ \
23809   __ret = (uint8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 16); \
23813 #define __noswap_vshrn_n_u16(__p0, __p1) __extension__ ({ \
23816   __ret = (uint8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 16); \
23822 #define vshrn_n_s32(__p0, __p1) __extension__ ({ \
23825   __ret = (int16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 1); \
23829 #define vshrn_n_s32(__p0, __p1) __extension__ ({ \
23833   __ret = (int16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 1); \
23837 #define __noswap_vshrn_n_s32(__p0, __p1) __extension__ ({ \
23840   __ret = (int16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 1); \
23846 #define vshrn_n_s64(__p0, __p1) __extension__ ({ \
23849   __ret = (int32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 2); \
23853 #define vshrn_n_s64(__p0, __p1) __extension__ ({ \
23857   __ret = (int32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 2); \
23861 #define __noswap_vshrn_n_s64(__p0, __p1) __extension__ ({ \
23864   __ret = (int32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 2); \
23870 #define vshrn_n_s16(__p0, __p1) __extension__ ({ \
23873   __ret = (int8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 0); \
23877 #define vshrn_n_s16(__p0, __p1) __extension__ ({ \
23881   __ret = (int8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 0); \
23885 #define __noswap_vshrn_n_s16(__p0, __p1) __extension__ ({ \
23888   __ret = (int8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 0); \
23894 #define vsli_n_p8(__p0, __p1, __p2) __extension__ ({ \
23896   poly8x8_t __s1 = __p1; \
23902 #define vsli_n_p8(__p0, __p1, __p2) __extension__ ({ \
23904   poly8x8_t __s1 = __p1; \
23915 #define vsli_n_p16(__p0, __p1, __p2) __extension__ ({ \
23917   poly16x4_t __s1 = __p1; \
23923 #define vsli_n_p16(__p0, __p1, __p2) __extension__ ({ \
23925   poly16x4_t __s1 = __p1; \
23936 #define vsliq_n_p8(__p0, __p1, __p2) __extension__ ({ \
23938   poly8x16_t __s1 = __p1; \
23944 #define vsliq_n_p8(__p0, __p1, __p2) __extension__ ({ \
23946   poly8x16_t __s1 = __p1; \
23957 #define vsliq_n_p16(__p0, __p1, __p2) __extension__ ({ \
23959   poly16x8_t __s1 = __p1; \
23965 #define vsliq_n_p16(__p0, __p1, __p2) __extension__ ({ \
23967   poly16x8_t __s1 = __p1; \
23978 #define vsliq_n_u8(__p0, __p1, __p2) __extension__ ({ \
23980   uint8x16_t __s1 = __p1; \
23986 #define vsliq_n_u8(__p0, __p1, __p2) __extension__ ({ \
23988   uint8x16_t __s1 = __p1; \
23999 #define vsliq_n_u32(__p0, __p1, __p2) __extension__ ({ \
24001   uint32x4_t __s1 = __p1; \
24007 #define vsliq_n_u32(__p0, __p1, __p2) __extension__ ({ \
24009   uint32x4_t __s1 = __p1; \
24020 #define vsliq_n_u64(__p0, __p1, __p2) __extension__ ({ \
24022   uint64x2_t __s1 = __p1; \
24028 #define vsliq_n_u64(__p0, __p1, __p2) __extension__ ({ \
24030   uint64x2_t __s1 = __p1; \
24041 #define vsliq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24043   uint16x8_t __s1 = __p1; \
24049 #define vsliq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24051   uint16x8_t __s1 = __p1; \
24062 #define vsliq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24064   int8x16_t __s1 = __p1; \
24070 #define vsliq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24072   int8x16_t __s1 = __p1; \
24083 #define vsliq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24085   int32x4_t __s1 = __p1; \
24091 #define vsliq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24093   int32x4_t __s1 = __p1; \
24104 #define vsliq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24106   int64x2_t __s1 = __p1; \
24112 #define vsliq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24114   int64x2_t __s1 = __p1; \
24125 #define vsliq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24127   int16x8_t __s1 = __p1; \
24133 #define vsliq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24135   int16x8_t __s1 = __p1; \
24146 #define vsli_n_u8(__p0, __p1, __p2) __extension__ ({ \
24148   uint8x8_t __s1 = __p1; \
24154 #define vsli_n_u8(__p0, __p1, __p2) __extension__ ({ \
24156   uint8x8_t __s1 = __p1; \
24167 #define vsli_n_u32(__p0, __p1, __p2) __extension__ ({ \
24169   uint32x2_t __s1 = __p1; \
24175 #define vsli_n_u32(__p0, __p1, __p2) __extension__ ({ \
24177   uint32x2_t __s1 = __p1; \
24188 #define vsli_n_u64(__p0, __p1, __p2) __extension__ ({ \
24190   uint64x1_t __s1 = __p1; \
24196 #define vsli_n_u64(__p0, __p1, __p2) __extension__ ({ \
24198   uint64x1_t __s1 = __p1; \
24206 #define vsli_n_u16(__p0, __p1, __p2) __extension__ ({ \
24208   uint16x4_t __s1 = __p1; \
24214 #define vsli_n_u16(__p0, __p1, __p2) __extension__ ({ \
24216   uint16x4_t __s1 = __p1; \
24227 #define vsli_n_s8(__p0, __p1, __p2) __extension__ ({ \
24229   int8x8_t __s1 = __p1; \
24235 #define vsli_n_s8(__p0, __p1, __p2) __extension__ ({ \
24237   int8x8_t __s1 = __p1; \
24248 #define vsli_n_s32(__p0, __p1, __p2) __extension__ ({ \
24250   int32x2_t __s1 = __p1; \
24256 #define vsli_n_s32(__p0, __p1, __p2) __extension__ ({ \
24258   int32x2_t __s1 = __p1; \
24269 #define vsli_n_s64(__p0, __p1, __p2) __extension__ ({ \
24271   int64x1_t __s1 = __p1; \
24277 #define vsli_n_s64(__p0, __p1, __p2) __extension__ ({ \
24279   int64x1_t __s1 = __p1; \
24287 #define vsli_n_s16(__p0, __p1, __p2) __extension__ ({ \
24289   int16x4_t __s1 = __p1; \
24295 #define vsli_n_s16(__p0, __p1, __p2) __extension__ ({ \
24297   int16x4_t __s1 = __p1; \
24308 #define vsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
24310   uint8x16_t __s1 = __p1; \
24316 #define vsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
24318   uint8x16_t __s1 = __p1; \
24329 #define vsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
24331   uint32x4_t __s1 = __p1; \
24337 #define vsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
24339   uint32x4_t __s1 = __p1; \
24350 #define vsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
24352   uint64x2_t __s1 = __p1; \
24358 #define vsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
24360   uint64x2_t __s1 = __p1; \
24371 #define vsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24373   uint16x8_t __s1 = __p1; \
24379 #define vsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24381   uint16x8_t __s1 = __p1; \
24392 #define vsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24394   int8x16_t __s1 = __p1; \
24400 #define vsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24402   int8x16_t __s1 = __p1; \
24413 #define vsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24415   int32x4_t __s1 = __p1; \
24421 #define vsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24423   int32x4_t __s1 = __p1; \
24434 #define vsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24436   int64x2_t __s1 = __p1; \
24442 #define vsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24444   int64x2_t __s1 = __p1; \
24455 #define vsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24457   int16x8_t __s1 = __p1; \
24463 #define vsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24465   int16x8_t __s1 = __p1; \
24476 #define vsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
24478   uint8x8_t __s1 = __p1; \
24484 #define vsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
24486   uint8x8_t __s1 = __p1; \
24497 #define vsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
24499   uint32x2_t __s1 = __p1; \
24505 #define vsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
24507   uint32x2_t __s1 = __p1; \
24518 #define vsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
24520   uint64x1_t __s1 = __p1; \
24526 #define vsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
24528   uint64x1_t __s1 = __p1; \
24536 #define vsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
24538   uint16x4_t __s1 = __p1; \
24544 #define vsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
24546   uint16x4_t __s1 = __p1; \
24557 #define vsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
24559   int8x8_t __s1 = __p1; \
24565 #define vsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
24567   int8x8_t __s1 = __p1; \
24578 #define vsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
24580   int32x2_t __s1 = __p1; \
24586 #define vsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
24588   int32x2_t __s1 = __p1; \
24599 #define vsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
24601   int64x1_t __s1 = __p1; \
24607 #define vsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
24609   int64x1_t __s1 = __p1; \
24617 #define vsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
24619   int16x4_t __s1 = __p1; \
24625 #define vsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
24627   int16x4_t __s1 = __p1; \
24638 #define vsri_n_p8(__p0, __p1, __p2) __extension__ ({ \
24640   poly8x8_t __s1 = __p1; \
24646 #define vsri_n_p8(__p0, __p1, __p2) __extension__ ({ \
24648   poly8x8_t __s1 = __p1; \
24659 #define vsri_n_p16(__p0, __p1, __p2) __extension__ ({ \
24661   poly16x4_t __s1 = __p1; \
24667 #define vsri_n_p16(__p0, __p1, __p2) __extension__ ({ \
24669   poly16x4_t __s1 = __p1; \
24680 #define vsriq_n_p8(__p0, __p1, __p2) __extension__ ({ \
24682   poly8x16_t __s1 = __p1; \
24688 #define vsriq_n_p8(__p0, __p1, __p2) __extension__ ({ \
24690   poly8x16_t __s1 = __p1; \
24701 #define vsriq_n_p16(__p0, __p1, __p2) __extension__ ({ \
24703   poly16x8_t __s1 = __p1; \
24709 #define vsriq_n_p16(__p0, __p1, __p2) __extension__ ({ \
24711   poly16x8_t __s1 = __p1; \
24722 #define vsriq_n_u8(__p0, __p1, __p2) __extension__ ({ \
24724   uint8x16_t __s1 = __p1; \
24730 #define vsriq_n_u8(__p0, __p1, __p2) __extension__ ({ \
24732   uint8x16_t __s1 = __p1; \
24743 #define vsriq_n_u32(__p0, __p1, __p2) __extension__ ({ \
24745   uint32x4_t __s1 = __p1; \
24751 #define vsriq_n_u32(__p0, __p1, __p2) __extension__ ({ \
24753   uint32x4_t __s1 = __p1; \
24764 #define vsriq_n_u64(__p0, __p1, __p2) __extension__ ({ \
24766   uint64x2_t __s1 = __p1; \
24772 #define vsriq_n_u64(__p0, __p1, __p2) __extension__ ({ \
24774   uint64x2_t __s1 = __p1; \
24785 #define vsriq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24787   uint16x8_t __s1 = __p1; \
24793 #define vsriq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24795   uint16x8_t __s1 = __p1; \
24806 #define vsriq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24808   int8x16_t __s1 = __p1; \
24814 #define vsriq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24816   int8x16_t __s1 = __p1; \
24827 #define vsriq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24829   int32x4_t __s1 = __p1; \
24835 #define vsriq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24837   int32x4_t __s1 = __p1; \
24848 #define vsriq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24850   int64x2_t __s1 = __p1; \
24856 #define vsriq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24858   int64x2_t __s1 = __p1; \
24869 #define vsriq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24871   int16x8_t __s1 = __p1; \
24877 #define vsriq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24879   int16x8_t __s1 = __p1; \
24890 #define vsri_n_u8(__p0, __p1, __p2) __extension__ ({ \
24892   uint8x8_t __s1 = __p1; \
24898 #define vsri_n_u8(__p0, __p1, __p2) __extension__ ({ \
24900   uint8x8_t __s1 = __p1; \
24911 #define vsri_n_u32(__p0, __p1, __p2) __extension__ ({ \
24913   uint32x2_t __s1 = __p1; \
24919 #define vsri_n_u32(__p0, __p1, __p2) __extension__ ({ \
24921   uint32x2_t __s1 = __p1; \
24932 #define vsri_n_u64(__p0, __p1, __p2) __extension__ ({ \
24934   uint64x1_t __s1 = __p1; \
24940 #define vsri_n_u64(__p0, __p1, __p2) __extension__ ({ \
24942   uint64x1_t __s1 = __p1; \
24950 #define vsri_n_u16(__p0, __p1, __p2) __extension__ ({ \
24952   uint16x4_t __s1 = __p1; \
24958 #define vsri_n_u16(__p0, __p1, __p2) __extension__ ({ \
24960   uint16x4_t __s1 = __p1; \
24971 #define vsri_n_s8(__p0, __p1, __p2) __extension__ ({ \
24973   int8x8_t __s1 = __p1; \
24979 #define vsri_n_s8(__p0, __p1, __p2) __extension__ ({ \
24981   int8x8_t __s1 = __p1; \
24992 #define vsri_n_s32(__p0, __p1, __p2) __extension__ ({ \
24994   int32x2_t __s1 = __p1; \
25000 #define vsri_n_s32(__p0, __p1, __p2) __extension__ ({ \
25002   int32x2_t __s1 = __p1; \
25013 #define vsri_n_s64(__p0, __p1, __p2) __extension__ ({ \
25015   int64x1_t __s1 = __p1; \
25021 #define vsri_n_s64(__p0, __p1, __p2) __extension__ ({ \
25023   int64x1_t __s1 = __p1; \
25031 #define vsri_n_s16(__p0, __p1, __p2) __extension__ ({ \
25033   int16x4_t __s1 = __p1; \
25039 #define vsri_n_s16(__p0, __p1, __p2) __extension__ ({ \
25041   int16x4_t __s1 = __p1; \
25052 #define vst1_p8(__p0, __p1) __extension__ ({ \
25053   poly8x8_t __s1 = __p1; \
25057 #define vst1_p8(__p0, __p1) __extension__ ({ \
25058   poly8x8_t __s1 = __p1; \
25065 #define vst1_p16(__p0, __p1) __extension__ ({ \
25066   poly16x4_t __s1 = __p1; \
25070 #define vst1_p16(__p0, __p1) __extension__ ({ \
25071   poly16x4_t __s1 = __p1; \
25078 #define vst1q_p8(__p0, __p1) __extension__ ({ \
25079   poly8x16_t __s1 = __p1; \
25083 #define vst1q_p8(__p0, __p1) __extension__ ({ \
25084   poly8x16_t __s1 = __p1; \
25091 #define vst1q_p16(__p0, __p1) __extension__ ({ \
25092   poly16x8_t __s1 = __p1; \
25096 #define vst1q_p16(__p0, __p1) __extension__ ({ \
25097   poly16x8_t __s1 = __p1; \
25104 #define vst1q_u8(__p0, __p1) __extension__ ({ \
25105   uint8x16_t __s1 = __p1; \
25109 #define vst1q_u8(__p0, __p1) __extension__ ({ \
25110   uint8x16_t __s1 = __p1; \
25117 #define vst1q_u32(__p0, __p1) __extension__ ({ \
25118   uint32x4_t __s1 = __p1; \
25122 #define vst1q_u32(__p0, __p1) __extension__ ({ \
25123   uint32x4_t __s1 = __p1; \
25130 #define vst1q_u64(__p0, __p1) __extension__ ({ \
25131   uint64x2_t __s1 = __p1; \
25135 #define vst1q_u64(__p0, __p1) __extension__ ({ \
25136   uint64x2_t __s1 = __p1; \
25143 #define vst1q_u16(__p0, __p1) __extension__ ({ \
25144   uint16x8_t __s1 = __p1; \
25148 #define vst1q_u16(__p0, __p1) __extension__ ({ \
25149   uint16x8_t __s1 = __p1; \
25156 #define vst1q_s8(__p0, __p1) __extension__ ({ \
25157   int8x16_t __s1 = __p1; \
25161 #define vst1q_s8(__p0, __p1) __extension__ ({ \
25162   int8x16_t __s1 = __p1; \
25169 #define vst1q_f32(__p0, __p1) __extension__ ({ \
25170   float32x4_t __s1 = __p1; \
25174 #define vst1q_f32(__p0, __p1) __extension__ ({ \
25175   float32x4_t __s1 = __p1; \
25182 #define vst1q_f16(__p0, __p1) __extension__ ({ \
25183   float16x8_t __s1 = __p1; \
25187 #define vst1q_f16(__p0, __p1) __extension__ ({ \
25188   float16x8_t __s1 = __p1; \
25195 #define vst1q_s32(__p0, __p1) __extension__ ({ \
25196   int32x4_t __s1 = __p1; \
25200 #define vst1q_s32(__p0, __p1) __extension__ ({ \
25201   int32x4_t __s1 = __p1; \
25208 #define vst1q_s64(__p0, __p1) __extension__ ({ \
25209   int64x2_t __s1 = __p1; \
25213 #define vst1q_s64(__p0, __p1) __extension__ ({ \
25214   int64x2_t __s1 = __p1; \
25221 #define vst1q_s16(__p0, __p1) __extension__ ({ \
25222   int16x8_t __s1 = __p1; \
25226 #define vst1q_s16(__p0, __p1) __extension__ ({ \
25227   int16x8_t __s1 = __p1; \
25234 #define vst1_u8(__p0, __p1) __extension__ ({ \
25235   uint8x8_t __s1 = __p1; \
25239 #define vst1_u8(__p0, __p1) __extension__ ({ \
25240   uint8x8_t __s1 = __p1; \
25247 #define vst1_u32(__p0, __p1) __extension__ ({ \
25248   uint32x2_t __s1 = __p1; \
25252 #define vst1_u32(__p0, __p1) __extension__ ({ \
25253   uint32x2_t __s1 = __p1; \
25260 #define vst1_u64(__p0, __p1) __extension__ ({ \
25261   uint64x1_t __s1 = __p1; \
25265 #define vst1_u64(__p0, __p1) __extension__ ({ \
25266   uint64x1_t __s1 = __p1; \
25272 #define vst1_u16(__p0, __p1) __extension__ ({ \
25273   uint16x4_t __s1 = __p1; \
25277 #define vst1_u16(__p0, __p1) __extension__ ({ \
25278   uint16x4_t __s1 = __p1; \
25285 #define vst1_s8(__p0, __p1) __extension__ ({ \
25286   int8x8_t __s1 = __p1; \
25290 #define vst1_s8(__p0, __p1) __extension__ ({ \
25291   int8x8_t __s1 = __p1; \
25298 #define vst1_f32(__p0, __p1) __extension__ ({ \
25299   float32x2_t __s1 = __p1; \
25303 #define vst1_f32(__p0, __p1) __extension__ ({ \
25304   float32x2_t __s1 = __p1; \
25311 #define vst1_f16(__p0, __p1) __extension__ ({ \
25312   float16x4_t __s1 = __p1; \
25316 #define vst1_f16(__p0, __p1) __extension__ ({ \
25317   float16x4_t __s1 = __p1; \
25324 #define vst1_s32(__p0, __p1) __extension__ ({ \
25325   int32x2_t __s1 = __p1; \
25329 #define vst1_s32(__p0, __p1) __extension__ ({ \
25330   int32x2_t __s1 = __p1; \
25337 #define vst1_s64(__p0, __p1) __extension__ ({ \
25338   int64x1_t __s1 = __p1; \
25342 #define vst1_s64(__p0, __p1) __extension__ ({ \
25343   int64x1_t __s1 = __p1; \
25349 #define vst1_s16(__p0, __p1) __extension__ ({ \
25350   int16x4_t __s1 = __p1; \
25354 #define vst1_s16(__p0, __p1) __extension__ ({ \
25355   int16x4_t __s1 = __p1; \
25362 #define vst1_lane_p8(__p0, __p1, __p2) __extension__ ({ \
25363   poly8x8_t __s1 = __p1; \
25367 #define vst1_lane_p8(__p0, __p1, __p2) __extension__ ({ \
25368   poly8x8_t __s1 = __p1; \
25375 #define vst1_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25376   poly16x4_t __s1 = __p1; \
25380 #define vst1_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25381   poly16x4_t __s1 = __p1; \
25388 #define vst1q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
25389   poly8x16_t __s1 = __p1; \
25393 #define vst1q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
25394   poly8x16_t __s1 = __p1; \
25401 #define vst1q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25402   poly16x8_t __s1 = __p1; \
25406 #define vst1q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25407   poly16x8_t __s1 = __p1; \
25414 #define vst1q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
25415   uint8x16_t __s1 = __p1; \
25419 #define vst1q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
25420   uint8x16_t __s1 = __p1; \
25427 #define vst1q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
25428   uint32x4_t __s1 = __p1; \
25432 #define vst1q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
25433   uint32x4_t __s1 = __p1; \
25440 #define vst1q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
25441   uint64x2_t __s1 = __p1; \
25445 #define vst1q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
25446   uint64x2_t __s1 = __p1; \
25453 #define vst1q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
25454   uint16x8_t __s1 = __p1; \
25458 #define vst1q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
25459   uint16x8_t __s1 = __p1; \
25466 #define vst1q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
25467   int8x16_t __s1 = __p1; \
25471 #define vst1q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
25472   int8x16_t __s1 = __p1; \
25479 #define vst1q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
25480   float32x4_t __s1 = __p1; \
25484 #define vst1q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
25485   float32x4_t __s1 = __p1; \
25492 #define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
25493   float16x8_t __s1 = __p1; \
25497 #define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
25498   float16x8_t __s1 = __p1; \
25505 #define vst1q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
25506   int32x4_t __s1 = __p1; \
25510 #define vst1q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
25511   int32x4_t __s1 = __p1; \
25518 #define vst1q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
25519   int64x2_t __s1 = __p1; \
25523 #define vst1q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
25524   int64x2_t __s1 = __p1; \
25531 #define vst1q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
25532   int16x8_t __s1 = __p1; \
25536 #define vst1q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
25537   int16x8_t __s1 = __p1; \
25544 #define vst1_lane_u8(__p0, __p1, __p2) __extension__ ({ \
25545   uint8x8_t __s1 = __p1; \
25549 #define vst1_lane_u8(__p0, __p1, __p2) __extension__ ({ \
25550   uint8x8_t __s1 = __p1; \
25557 #define vst1_lane_u32(__p0, __p1, __p2) __extension__ ({ \
25558   uint32x2_t __s1 = __p1; \
25562 #define vst1_lane_u32(__p0, __p1, __p2) __extension__ ({ \
25563   uint32x2_t __s1 = __p1; \
25570 #define vst1_lane_u64(__p0, __p1, __p2) __extension__ ({ \
25571   uint64x1_t __s1 = __p1; \
25575 #define vst1_lane_u64(__p0, __p1, __p2) __extension__ ({ \
25576   uint64x1_t __s1 = __p1; \
25582 #define vst1_lane_u16(__p0, __p1, __p2) __extension__ ({ \
25583   uint16x4_t __s1 = __p1; \
25587 #define vst1_lane_u16(__p0, __p1, __p2) __extension__ ({ \
25588   uint16x4_t __s1 = __p1; \
25595 #define vst1_lane_s8(__p0, __p1, __p2) __extension__ ({ \
25596   int8x8_t __s1 = __p1; \
25600 #define vst1_lane_s8(__p0, __p1, __p2) __extension__ ({ \
25601   int8x8_t __s1 = __p1; \
25608 #define vst1_lane_f32(__p0, __p1, __p2) __extension__ ({ \
25609   float32x2_t __s1 = __p1; \
25613 #define vst1_lane_f32(__p0, __p1, __p2) __extension__ ({ \
25614   float32x2_t __s1 = __p1; \
25621 #define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
25622   float16x4_t __s1 = __p1; \
25626 #define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
25627   float16x4_t __s1 = __p1; \
25634 #define vst1_lane_s32(__p0, __p1, __p2) __extension__ ({ \
25635   int32x2_t __s1 = __p1; \
25639 #define vst1_lane_s32(__p0, __p1, __p2) __extension__ ({ \
25640   int32x2_t __s1 = __p1; \
25647 #define vst1_lane_s64(__p0, __p1, __p2) __extension__ ({ \
25648   int64x1_t __s1 = __p1; \
25652 #define vst1_lane_s64(__p0, __p1, __p2) __extension__ ({ \
25653   int64x1_t __s1 = __p1; \
25659 #define vst1_lane_s16(__p0, __p1, __p2) __extension__ ({ \
25660   int16x4_t __s1 = __p1; \
25664 #define vst1_lane_s16(__p0, __p1, __p2) __extension__ ({ \
25665   int16x4_t __s1 = __p1; \
25672 #define vst2_p8(__p0, __p1) __extension__ ({ \
25673   poly8x8x2_t __s1 = __p1; \
25677 #define vst2_p8(__p0, __p1) __extension__ ({ \
25678   poly8x8x2_t __s1 = __p1; \
25687 #define vst2_p16(__p0, __p1) __extension__ ({ \
25688   poly16x4x2_t __s1 = __p1; \
25692 #define vst2_p16(__p0, __p1) __extension__ ({ \
25693   poly16x4x2_t __s1 = __p1; \
25702 #define vst2q_p8(__p0, __p1) __extension__ ({ \
25703   poly8x16x2_t __s1 = __p1; \
25707 #define vst2q_p8(__p0, __p1) __extension__ ({ \
25708   poly8x16x2_t __s1 = __p1; \
25717 #define vst2q_p16(__p0, __p1) __extension__ ({ \
25718   poly16x8x2_t __s1 = __p1; \
25722 #define vst2q_p16(__p0, __p1) __extension__ ({ \
25723   poly16x8x2_t __s1 = __p1; \
25732 #define vst2q_u8(__p0, __p1) __extension__ ({ \
25733   uint8x16x2_t __s1 = __p1; \
25737 #define vst2q_u8(__p0, __p1) __extension__ ({ \
25738   uint8x16x2_t __s1 = __p1; \
25747 #define vst2q_u32(__p0, __p1) __extension__ ({ \
25748   uint32x4x2_t __s1 = __p1; \
25752 #define vst2q_u32(__p0, __p1) __extension__ ({ \
25753   uint32x4x2_t __s1 = __p1; \
25762 #define vst2q_u16(__p0, __p1) __extension__ ({ \
25763   uint16x8x2_t __s1 = __p1; \
25767 #define vst2q_u16(__p0, __p1) __extension__ ({ \
25768   uint16x8x2_t __s1 = __p1; \
25777 #define vst2q_s8(__p0, __p1) __extension__ ({ \
25778   int8x16x2_t __s1 = __p1; \
25782 #define vst2q_s8(__p0, __p1) __extension__ ({ \
25783   int8x16x2_t __s1 = __p1; \
25792 #define vst2q_f32(__p0, __p1) __extension__ ({ \
25793   float32x4x2_t __s1 = __p1; \
25797 #define vst2q_f32(__p0, __p1) __extension__ ({ \
25798   float32x4x2_t __s1 = __p1; \
25807 #define vst2q_f16(__p0, __p1) __extension__ ({ \
25808   float16x8x2_t __s1 = __p1; \
25812 #define vst2q_f16(__p0, __p1) __extension__ ({ \
25813   float16x8x2_t __s1 = __p1; \
25822 #define vst2q_s32(__p0, __p1) __extension__ ({ \
25823   int32x4x2_t __s1 = __p1; \
25827 #define vst2q_s32(__p0, __p1) __extension__ ({ \
25828   int32x4x2_t __s1 = __p1; \
25837 #define vst2q_s16(__p0, __p1) __extension__ ({ \
25838   int16x8x2_t __s1 = __p1; \
25842 #define vst2q_s16(__p0, __p1) __extension__ ({ \
25843   int16x8x2_t __s1 = __p1; \
25852 #define vst2_u8(__p0, __p1) __extension__ ({ \
25853   uint8x8x2_t __s1 = __p1; \
25857 #define vst2_u8(__p0, __p1) __extension__ ({ \
25858   uint8x8x2_t __s1 = __p1; \
25867 #define vst2_u32(__p0, __p1) __extension__ ({ \
25868   uint32x2x2_t __s1 = __p1; \
25872 #define vst2_u32(__p0, __p1) __extension__ ({ \
25873   uint32x2x2_t __s1 = __p1; \
25882 #define vst2_u64(__p0, __p1) __extension__ ({ \
25883   uint64x1x2_t __s1 = __p1; \
25887 #define vst2_u64(__p0, __p1) __extension__ ({ \
25888   uint64x1x2_t __s1 = __p1; \
25894 #define vst2_u16(__p0, __p1) __extension__ ({ \
25895   uint16x4x2_t __s1 = __p1; \
25899 #define vst2_u16(__p0, __p1) __extension__ ({ \
25900   uint16x4x2_t __s1 = __p1; \
25909 #define vst2_s8(__p0, __p1) __extension__ ({ \
25910   int8x8x2_t __s1 = __p1; \
25914 #define vst2_s8(__p0, __p1) __extension__ ({ \
25915   int8x8x2_t __s1 = __p1; \
25924 #define vst2_f32(__p0, __p1) __extension__ ({ \
25925   float32x2x2_t __s1 = __p1; \
25929 #define vst2_f32(__p0, __p1) __extension__ ({ \
25930   float32x2x2_t __s1 = __p1; \
25939 #define vst2_f16(__p0, __p1) __extension__ ({ \
25940   float16x4x2_t __s1 = __p1; \
25944 #define vst2_f16(__p0, __p1) __extension__ ({ \
25945   float16x4x2_t __s1 = __p1; \
25954 #define vst2_s32(__p0, __p1) __extension__ ({ \
25955   int32x2x2_t __s1 = __p1; \
25959 #define vst2_s32(__p0, __p1) __extension__ ({ \
25960   int32x2x2_t __s1 = __p1; \
25969 #define vst2_s64(__p0, __p1) __extension__ ({ \
25970   int64x1x2_t __s1 = __p1; \
25974 #define vst2_s64(__p0, __p1) __extension__ ({ \
25975   int64x1x2_t __s1 = __p1; \
25981 #define vst2_s16(__p0, __p1) __extension__ ({ \
25982   int16x4x2_t __s1 = __p1; \
25986 #define vst2_s16(__p0, __p1) __extension__ ({ \
25987   int16x4x2_t __s1 = __p1; \
25996 #define vst2_lane_p8(__p0, __p1, __p2) __extension__ ({ \
25997   poly8x8x2_t __s1 = __p1; \
26001 #define vst2_lane_p8(__p0, __p1, __p2) __extension__ ({ \
26002   poly8x8x2_t __s1 = __p1; \
26011 #define vst2_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26012   poly16x4x2_t __s1 = __p1; \
26016 #define vst2_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26017   poly16x4x2_t __s1 = __p1; \
26026 #define vst2q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26027   poly16x8x2_t __s1 = __p1; \
26031 #define vst2q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26032   poly16x8x2_t __s1 = __p1; \
26041 #define vst2q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26042   uint32x4x2_t __s1 = __p1; \
26046 #define vst2q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26047   uint32x4x2_t __s1 = __p1; \
26056 #define vst2q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26057   uint16x8x2_t __s1 = __p1; \
26061 #define vst2q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26062   uint16x8x2_t __s1 = __p1; \
26071 #define vst2q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26072   float32x4x2_t __s1 = __p1; \
26076 #define vst2q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26077   float32x4x2_t __s1 = __p1; \
26086 #define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26087   float16x8x2_t __s1 = __p1; \
26091 #define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26092   float16x8x2_t __s1 = __p1; \
26101 #define vst2q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26102   int32x4x2_t __s1 = __p1; \
26106 #define vst2q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26107   int32x4x2_t __s1 = __p1; \
26116 #define vst2q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26117   int16x8x2_t __s1 = __p1; \
26121 #define vst2q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26122   int16x8x2_t __s1 = __p1; \
26131 #define vst2_lane_u8(__p0, __p1, __p2) __extension__ ({ \
26132   uint8x8x2_t __s1 = __p1; \
26136 #define vst2_lane_u8(__p0, __p1, __p2) __extension__ ({ \
26137   uint8x8x2_t __s1 = __p1; \
26146 #define vst2_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26147   uint32x2x2_t __s1 = __p1; \
26151 #define vst2_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26152   uint32x2x2_t __s1 = __p1; \
26161 #define vst2_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26162   uint16x4x2_t __s1 = __p1; \
26166 #define vst2_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26167   uint16x4x2_t __s1 = __p1; \
26176 #define vst2_lane_s8(__p0, __p1, __p2) __extension__ ({ \
26177   int8x8x2_t __s1 = __p1; \
26181 #define vst2_lane_s8(__p0, __p1, __p2) __extension__ ({ \
26182   int8x8x2_t __s1 = __p1; \
26191 #define vst2_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26192   float32x2x2_t __s1 = __p1; \
26196 #define vst2_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26197   float32x2x2_t __s1 = __p1; \
26206 #define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26207   float16x4x2_t __s1 = __p1; \
26211 #define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26212   float16x4x2_t __s1 = __p1; \
26221 #define vst2_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26222   int32x2x2_t __s1 = __p1; \
26226 #define vst2_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26227   int32x2x2_t __s1 = __p1; \
26236 #define vst2_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26237   int16x4x2_t __s1 = __p1; \
26241 #define vst2_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26242   int16x4x2_t __s1 = __p1; \
26251 #define vst3_p8(__p0, __p1) __extension__ ({ \
26252   poly8x8x3_t __s1 = __p1; \
26256 #define vst3_p8(__p0, __p1) __extension__ ({ \
26257   poly8x8x3_t __s1 = __p1; \
26267 #define vst3_p16(__p0, __p1) __extension__ ({ \
26268   poly16x4x3_t __s1 = __p1; \
26272 #define vst3_p16(__p0, __p1) __extension__ ({ \
26273   poly16x4x3_t __s1 = __p1; \
26283 #define vst3q_p8(__p0, __p1) __extension__ ({ \
26284   poly8x16x3_t __s1 = __p1; \
26288 #define vst3q_p8(__p0, __p1) __extension__ ({ \
26289   poly8x16x3_t __s1 = __p1; \
26299 #define vst3q_p16(__p0, __p1) __extension__ ({ \
26300   poly16x8x3_t __s1 = __p1; \
26304 #define vst3q_p16(__p0, __p1) __extension__ ({ \
26305   poly16x8x3_t __s1 = __p1; \
26315 #define vst3q_u8(__p0, __p1) __extension__ ({ \
26316   uint8x16x3_t __s1 = __p1; \
26320 #define vst3q_u8(__p0, __p1) __extension__ ({ \
26321   uint8x16x3_t __s1 = __p1; \
26331 #define vst3q_u32(__p0, __p1) __extension__ ({ \
26332   uint32x4x3_t __s1 = __p1; \
26336 #define vst3q_u32(__p0, __p1) __extension__ ({ \
26337   uint32x4x3_t __s1 = __p1; \
26347 #define vst3q_u16(__p0, __p1) __extension__ ({ \
26348   uint16x8x3_t __s1 = __p1; \
26352 #define vst3q_u16(__p0, __p1) __extension__ ({ \
26353   uint16x8x3_t __s1 = __p1; \
26363 #define vst3q_s8(__p0, __p1) __extension__ ({ \
26364   int8x16x3_t __s1 = __p1; \
26368 #define vst3q_s8(__p0, __p1) __extension__ ({ \
26369   int8x16x3_t __s1 = __p1; \
26379 #define vst3q_f32(__p0, __p1) __extension__ ({ \
26380   float32x4x3_t __s1 = __p1; \
26384 #define vst3q_f32(__p0, __p1) __extension__ ({ \
26385   float32x4x3_t __s1 = __p1; \
26395 #define vst3q_f16(__p0, __p1) __extension__ ({ \
26396   float16x8x3_t __s1 = __p1; \
26400 #define vst3q_f16(__p0, __p1) __extension__ ({ \
26401   float16x8x3_t __s1 = __p1; \
26411 #define vst3q_s32(__p0, __p1) __extension__ ({ \
26412   int32x4x3_t __s1 = __p1; \
26416 #define vst3q_s32(__p0, __p1) __extension__ ({ \
26417   int32x4x3_t __s1 = __p1; \
26427 #define vst3q_s16(__p0, __p1) __extension__ ({ \
26428   int16x8x3_t __s1 = __p1; \
26432 #define vst3q_s16(__p0, __p1) __extension__ ({ \
26433   int16x8x3_t __s1 = __p1; \
26443 #define vst3_u8(__p0, __p1) __extension__ ({ \
26444   uint8x8x3_t __s1 = __p1; \
26448 #define vst3_u8(__p0, __p1) __extension__ ({ \
26449   uint8x8x3_t __s1 = __p1; \
26459 #define vst3_u32(__p0, __p1) __extension__ ({ \
26460   uint32x2x3_t __s1 = __p1; \
26464 #define vst3_u32(__p0, __p1) __extension__ ({ \
26465   uint32x2x3_t __s1 = __p1; \
26475 #define vst3_u64(__p0, __p1) __extension__ ({ \
26476   uint64x1x3_t __s1 = __p1; \
26480 #define vst3_u64(__p0, __p1) __extension__ ({ \
26481   uint64x1x3_t __s1 = __p1; \
26487 #define vst3_u16(__p0, __p1) __extension__ ({ \
26488   uint16x4x3_t __s1 = __p1; \
26492 #define vst3_u16(__p0, __p1) __extension__ ({ \
26493   uint16x4x3_t __s1 = __p1; \
26503 #define vst3_s8(__p0, __p1) __extension__ ({ \
26504   int8x8x3_t __s1 = __p1; \
26508 #define vst3_s8(__p0, __p1) __extension__ ({ \
26509   int8x8x3_t __s1 = __p1; \
26519 #define vst3_f32(__p0, __p1) __extension__ ({ \
26520   float32x2x3_t __s1 = __p1; \
26524 #define vst3_f32(__p0, __p1) __extension__ ({ \
26525   float32x2x3_t __s1 = __p1; \
26535 #define vst3_f16(__p0, __p1) __extension__ ({ \
26536   float16x4x3_t __s1 = __p1; \
26540 #define vst3_f16(__p0, __p1) __extension__ ({ \
26541   float16x4x3_t __s1 = __p1; \
26551 #define vst3_s32(__p0, __p1) __extension__ ({ \
26552   int32x2x3_t __s1 = __p1; \
26556 #define vst3_s32(__p0, __p1) __extension__ ({ \
26557   int32x2x3_t __s1 = __p1; \
26567 #define vst3_s64(__p0, __p1) __extension__ ({ \
26568   int64x1x3_t __s1 = __p1; \
26572 #define vst3_s64(__p0, __p1) __extension__ ({ \
26573   int64x1x3_t __s1 = __p1; \
26579 #define vst3_s16(__p0, __p1) __extension__ ({ \
26580   int16x4x3_t __s1 = __p1; \
26584 #define vst3_s16(__p0, __p1) __extension__ ({ \
26585   int16x4x3_t __s1 = __p1; \
26595 #define vst3_lane_p8(__p0, __p1, __p2) __extension__ ({ \
26596   poly8x8x3_t __s1 = __p1; \
26600 #define vst3_lane_p8(__p0, __p1, __p2) __extension__ ({ \
26601   poly8x8x3_t __s1 = __p1; \
26611 #define vst3_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26612   poly16x4x3_t __s1 = __p1; \
26616 #define vst3_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26617   poly16x4x3_t __s1 = __p1; \
26627 #define vst3q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26628   poly16x8x3_t __s1 = __p1; \
26632 #define vst3q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26633   poly16x8x3_t __s1 = __p1; \
26643 #define vst3q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26644   uint32x4x3_t __s1 = __p1; \
26648 #define vst3q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26649   uint32x4x3_t __s1 = __p1; \
26659 #define vst3q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26660   uint16x8x3_t __s1 = __p1; \
26664 #define vst3q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26665   uint16x8x3_t __s1 = __p1; \
26675 #define vst3q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26676   float32x4x3_t __s1 = __p1; \
26680 #define vst3q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26681   float32x4x3_t __s1 = __p1; \
26691 #define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26692   float16x8x3_t __s1 = __p1; \
26696 #define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26697   float16x8x3_t __s1 = __p1; \
26707 #define vst3q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26708   int32x4x3_t __s1 = __p1; \
26712 #define vst3q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26713   int32x4x3_t __s1 = __p1; \
26723 #define vst3q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26724   int16x8x3_t __s1 = __p1; \
26728 #define vst3q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26729   int16x8x3_t __s1 = __p1; \
26739 #define vst3_lane_u8(__p0, __p1, __p2) __extension__ ({ \
26740   uint8x8x3_t __s1 = __p1; \
26744 #define vst3_lane_u8(__p0, __p1, __p2) __extension__ ({ \
26745   uint8x8x3_t __s1 = __p1; \
26755 #define vst3_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26756   uint32x2x3_t __s1 = __p1; \
26760 #define vst3_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26761   uint32x2x3_t __s1 = __p1; \
26771 #define vst3_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26772   uint16x4x3_t __s1 = __p1; \
26776 #define vst3_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26777   uint16x4x3_t __s1 = __p1; \
26787 #define vst3_lane_s8(__p0, __p1, __p2) __extension__ ({ \
26788   int8x8x3_t __s1 = __p1; \
26792 #define vst3_lane_s8(__p0, __p1, __p2) __extension__ ({ \
26793   int8x8x3_t __s1 = __p1; \
26803 #define vst3_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26804   float32x2x3_t __s1 = __p1; \
26808 #define vst3_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26809   float32x2x3_t __s1 = __p1; \
26819 #define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26820   float16x4x3_t __s1 = __p1; \
26824 #define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26825   float16x4x3_t __s1 = __p1; \
26835 #define vst3_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26836   int32x2x3_t __s1 = __p1; \
26840 #define vst3_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26841   int32x2x3_t __s1 = __p1; \
26851 #define vst3_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26852   int16x4x3_t __s1 = __p1; \
26856 #define vst3_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26857   int16x4x3_t __s1 = __p1; \
26867 #define vst4_p8(__p0, __p1) __extension__ ({ \
26868   poly8x8x4_t __s1 = __p1; \
26872 #define vst4_p8(__p0, __p1) __extension__ ({ \
26873   poly8x8x4_t __s1 = __p1; \
26884 #define vst4_p16(__p0, __p1) __extension__ ({ \
26885   poly16x4x4_t __s1 = __p1; \
26889 #define vst4_p16(__p0, __p1) __extension__ ({ \
26890   poly16x4x4_t __s1 = __p1; \
26901 #define vst4q_p8(__p0, __p1) __extension__ ({ \
26902   poly8x16x4_t __s1 = __p1; \
26906 #define vst4q_p8(__p0, __p1) __extension__ ({ \
26907   poly8x16x4_t __s1 = __p1; \
26918 #define vst4q_p16(__p0, __p1) __extension__ ({ \
26919   poly16x8x4_t __s1 = __p1; \
26923 #define vst4q_p16(__p0, __p1) __extension__ ({ \
26924   poly16x8x4_t __s1 = __p1; \
26935 #define vst4q_u8(__p0, __p1) __extension__ ({ \
26936   uint8x16x4_t __s1 = __p1; \
26940 #define vst4q_u8(__p0, __p1) __extension__ ({ \
26941   uint8x16x4_t __s1 = __p1; \
26952 #define vst4q_u32(__p0, __p1) __extension__ ({ \
26953   uint32x4x4_t __s1 = __p1; \
26957 #define vst4q_u32(__p0, __p1) __extension__ ({ \
26958   uint32x4x4_t __s1 = __p1; \
26969 #define vst4q_u16(__p0, __p1) __extension__ ({ \
26970   uint16x8x4_t __s1 = __p1; \
26974 #define vst4q_u16(__p0, __p1) __extension__ ({ \
26975   uint16x8x4_t __s1 = __p1; \
26986 #define vst4q_s8(__p0, __p1) __extension__ ({ \
26987   int8x16x4_t __s1 = __p1; \
26991 #define vst4q_s8(__p0, __p1) __extension__ ({ \
26992   int8x16x4_t __s1 = __p1; \
27003 #define vst4q_f32(__p0, __p1) __extension__ ({ \
27004   float32x4x4_t __s1 = __p1; \
27008 #define vst4q_f32(__p0, __p1) __extension__ ({ \
27009   float32x4x4_t __s1 = __p1; \
27020 #define vst4q_f16(__p0, __p1) __extension__ ({ \
27021   float16x8x4_t __s1 = __p1; \
27025 #define vst4q_f16(__p0, __p1) __extension__ ({ \
27026   float16x8x4_t __s1 = __p1; \
27037 #define vst4q_s32(__p0, __p1) __extension__ ({ \
27038   int32x4x4_t __s1 = __p1; \
27042 #define vst4q_s32(__p0, __p1) __extension__ ({ \
27043   int32x4x4_t __s1 = __p1; \
27054 #define vst4q_s16(__p0, __p1) __extension__ ({ \
27055   int16x8x4_t __s1 = __p1; \
27059 #define vst4q_s16(__p0, __p1) __extension__ ({ \
27060   int16x8x4_t __s1 = __p1; \
27071 #define vst4_u8(__p0, __p1) __extension__ ({ \
27072   uint8x8x4_t __s1 = __p1; \
27076 #define vst4_u8(__p0, __p1) __extension__ ({ \
27077   uint8x8x4_t __s1 = __p1; \
27088 #define vst4_u32(__p0, __p1) __extension__ ({ \
27089   uint32x2x4_t __s1 = __p1; \
27093 #define vst4_u32(__p0, __p1) __extension__ ({ \
27094   uint32x2x4_t __s1 = __p1; \
27105 #define vst4_u64(__p0, __p1) __extension__ ({ \
27106   uint64x1x4_t __s1 = __p1; \
27110 #define vst4_u64(__p0, __p1) __extension__ ({ \
27111   uint64x1x4_t __s1 = __p1; \
27117 #define vst4_u16(__p0, __p1) __extension__ ({ \
27118   uint16x4x4_t __s1 = __p1; \
27122 #define vst4_u16(__p0, __p1) __extension__ ({ \
27123   uint16x4x4_t __s1 = __p1; \
27134 #define vst4_s8(__p0, __p1) __extension__ ({ \
27135   int8x8x4_t __s1 = __p1; \
27139 #define vst4_s8(__p0, __p1) __extension__ ({ \
27140   int8x8x4_t __s1 = __p1; \
27151 #define vst4_f32(__p0, __p1) __extension__ ({ \
27152   float32x2x4_t __s1 = __p1; \
27156 #define vst4_f32(__p0, __p1) __extension__ ({ \
27157   float32x2x4_t __s1 = __p1; \
27168 #define vst4_f16(__p0, __p1) __extension__ ({ \
27169   float16x4x4_t __s1 = __p1; \
27173 #define vst4_f16(__p0, __p1) __extension__ ({ \
27174   float16x4x4_t __s1 = __p1; \
27185 #define vst4_s32(__p0, __p1) __extension__ ({ \
27186   int32x2x4_t __s1 = __p1; \
27190 #define vst4_s32(__p0, __p1) __extension__ ({ \
27191   int32x2x4_t __s1 = __p1; \
27202 #define vst4_s64(__p0, __p1) __extension__ ({ \
27203   int64x1x4_t __s1 = __p1; \
27207 #define vst4_s64(__p0, __p1) __extension__ ({ \
27208   int64x1x4_t __s1 = __p1; \
27214 #define vst4_s16(__p0, __p1) __extension__ ({ \
27215   int16x4x4_t __s1 = __p1; \
27219 #define vst4_s16(__p0, __p1) __extension__ ({ \
27220   int16x4x4_t __s1 = __p1; \
27231 #define vst4_lane_p8(__p0, __p1, __p2) __extension__ ({ \
27232   poly8x8x4_t __s1 = __p1; \
27236 #define vst4_lane_p8(__p0, __p1, __p2) __extension__ ({ \
27237   poly8x8x4_t __s1 = __p1; \
27248 #define vst4_lane_p16(__p0, __p1, __p2) __extension__ ({ \
27249   poly16x4x4_t __s1 = __p1; \
27253 #define vst4_lane_p16(__p0, __p1, __p2) __extension__ ({ \
27254   poly16x4x4_t __s1 = __p1; \
27265 #define vst4q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
27266   poly16x8x4_t __s1 = __p1; \
27270 #define vst4q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
27271   poly16x8x4_t __s1 = __p1; \
27282 #define vst4q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
27283   uint32x4x4_t __s1 = __p1; \
27287 #define vst4q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
27288   uint32x4x4_t __s1 = __p1; \
27299 #define vst4q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
27300   uint16x8x4_t __s1 = __p1; \
27304 #define vst4q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
27305   uint16x8x4_t __s1 = __p1; \
27316 #define vst4q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
27317   float32x4x4_t __s1 = __p1; \
27321 #define vst4q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
27322   float32x4x4_t __s1 = __p1; \
27333 #define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
27334   float16x8x4_t __s1 = __p1; \
27338 #define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
27339   float16x8x4_t __s1 = __p1; \
27350 #define vst4q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
27351   int32x4x4_t __s1 = __p1; \
27355 #define vst4q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
27356   int32x4x4_t __s1 = __p1; \
27367 #define vst4q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
27368   int16x8x4_t __s1 = __p1; \
27372 #define vst4q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
27373   int16x8x4_t __s1 = __p1; \
27384 #define vst4_lane_u8(__p0, __p1, __p2) __extension__ ({ \
27385   uint8x8x4_t __s1 = __p1; \
27389 #define vst4_lane_u8(__p0, __p1, __p2) __extension__ ({ \
27390   uint8x8x4_t __s1 = __p1; \
27401 #define vst4_lane_u32(__p0, __p1, __p2) __extension__ ({ \
27402   uint32x2x4_t __s1 = __p1; \
27406 #define vst4_lane_u32(__p0, __p1, __p2) __extension__ ({ \
27407   uint32x2x4_t __s1 = __p1; \
27418 #define vst4_lane_u16(__p0, __p1, __p2) __extension__ ({ \
27419   uint16x4x4_t __s1 = __p1; \
27423 #define vst4_lane_u16(__p0, __p1, __p2) __extension__ ({ \
27424   uint16x4x4_t __s1 = __p1; \
27435 #define vst4_lane_s8(__p0, __p1, __p2) __extension__ ({ \
27436   int8x8x4_t __s1 = __p1; \
27440 #define vst4_lane_s8(__p0, __p1, __p2) __extension__ ({ \
27441   int8x8x4_t __s1 = __p1; \
27452 #define vst4_lane_f32(__p0, __p1, __p2) __extension__ ({ \
27453   float32x2x4_t __s1 = __p1; \
27457 #define vst4_lane_f32(__p0, __p1, __p2) __extension__ ({ \
27458   float32x2x4_t __s1 = __p1; \
27469 #define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
27470   float16x4x4_t __s1 = __p1; \
27474 #define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
27475   float16x4x4_t __s1 = __p1; \
27486 #define vst4_lane_s32(__p0, __p1, __p2) __extension__ ({ \
27487   int32x2x4_t __s1 = __p1; \
27491 #define vst4_lane_s32(__p0, __p1, __p2) __extension__ ({ \
27492   int32x2x4_t __s1 = __p1; \
27503 #define vst4_lane_s16(__p0, __p1, __p2) __extension__ ({ \
27504   int16x4x4_t __s1 = __p1; \
27508 #define vst4_lane_s16(__p0, __p1, __p2) __extension__ ({ \
27509   int16x4x4_t __s1 = __p1; \
27520 __ai uint8x16_t vsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
27522   __ret = __p0 - __p1;
27526 __ai uint8x16_t vsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
27528   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
27537 __ai uint32x4_t vsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
27539   __ret = __p0 - __p1;
27543 __ai uint32x4_t vsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
27545   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27554 __ai uint64x2_t vsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
27556   __ret = __p0 - __p1;
27560 __ai uint64x2_t vsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
27562   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27571 __ai uint16x8_t vsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
27573   __ret = __p0 - __p1;
27577 __ai uint16x8_t vsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
27579   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27588 __ai int8x16_t vsubq_s8(int8x16_t __p0, int8x16_t __p1) {
27590   __ret = __p0 - __p1;
27594 __ai int8x16_t vsubq_s8(int8x16_t __p0, int8x16_t __p1) {
27596   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
27605 __ai float32x4_t vsubq_f32(float32x4_t __p0, float32x4_t __p1) {
27607   __ret = __p0 - __p1;
27611 __ai float32x4_t vsubq_f32(float32x4_t __p0, float32x4_t __p1) {
27613   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27622 __ai int32x4_t vsubq_s32(int32x4_t __p0, int32x4_t __p1) {
27624   __ret = __p0 - __p1;
27628 __ai int32x4_t vsubq_s32(int32x4_t __p0, int32x4_t __p1) {
27630   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27639 __ai int64x2_t vsubq_s64(int64x2_t __p0, int64x2_t __p1) {
27641   __ret = __p0 - __p1;
27645 __ai int64x2_t vsubq_s64(int64x2_t __p0, int64x2_t __p1) {
27647   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27656 __ai int16x8_t vsubq_s16(int16x8_t __p0, int16x8_t __p1) {
27658   __ret = __p0 - __p1;
27662 __ai int16x8_t vsubq_s16(int16x8_t __p0, int16x8_t __p1) {
27664   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27673 __ai uint8x8_t vsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
27675   __ret = __p0 - __p1;
27679 __ai uint8x8_t vsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
27681   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27690 __ai uint32x2_t vsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
27692   __ret = __p0 - __p1;
27696 __ai uint32x2_t vsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
27698   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27707 __ai uint64x1_t vsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
27709   __ret = __p0 - __p1;
27713 __ai uint64x1_t vsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
27715   __ret = __p0 - __p1;
27721 __ai uint16x4_t vsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
27723   __ret = __p0 - __p1;
27727 __ai uint16x4_t vsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
27729   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27738 __ai int8x8_t vsub_s8(int8x8_t __p0, int8x8_t __p1) {
27740   __ret = __p0 - __p1;
27744 __ai int8x8_t vsub_s8(int8x8_t __p0, int8x8_t __p1) {
27746   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27755 __ai float32x2_t vsub_f32(float32x2_t __p0, float32x2_t __p1) {
27757   __ret = __p0 - __p1;
27761 __ai float32x2_t vsub_f32(float32x2_t __p0, float32x2_t __p1) {
27763   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27772 __ai int32x2_t vsub_s32(int32x2_t __p0, int32x2_t __p1) {
27774   __ret = __p0 - __p1;
27778 __ai int32x2_t vsub_s32(int32x2_t __p0, int32x2_t __p1) {
27780   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27789 __ai int64x1_t vsub_s64(int64x1_t __p0, int64x1_t __p1) {
27791   __ret = __p0 - __p1;
27795 __ai int64x1_t vsub_s64(int64x1_t __p0, int64x1_t __p1) {
27797   __ret = __p0 - __p1;
27803 __ai int16x4_t vsub_s16(int16x4_t __p0, int16x4_t __p1) {
27805   __ret = __p0 - __p1;
27809 __ai int16x4_t vsub_s16(int16x4_t __p0, int16x4_t __p1) {
27811   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27820 __ai uint16x4_t vsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
27822   __ret = (uint16x4_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
27826 __ai uint16x4_t vsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
27828   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27834 __ai uint16x4_t __noswap_vsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
27836   __ret = (uint16x4_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
27842 __ai uint32x2_t vsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
27844   __ret = (uint32x2_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
27848 __ai uint32x2_t vsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
27850   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27856 __ai uint32x2_t __noswap_vsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
27858   __ret = (uint32x2_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
27864 __ai uint8x8_t vsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
27866   __ret = (uint8x8_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
27870 __ai uint8x8_t vsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
27872   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27878 __ai uint8x8_t __noswap_vsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
27880   __ret = (uint8x8_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
27886 __ai int16x4_t vsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
27888   __ret = (int16x4_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
27892 __ai int16x4_t vsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
27894   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27900 __ai int16x4_t __noswap_vsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
27902   __ret = (int16x4_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
27908 __ai int32x2_t vsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
27910   __ret = (int32x2_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
27914 __ai int32x2_t vsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
27916   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27922 __ai int32x2_t __noswap_vsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
27924   __ret = (int32x2_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
27930 __ai int8x8_t vsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
27932   __ret = (int8x8_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
27936 __ai int8x8_t vsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
27938   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27944 __ai int8x8_t __noswap_vsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
27946   __ret = (int8x8_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
27952 __ai uint16x8_t vsubl_u8(uint8x8_t __p0, uint8x8_t __p1) {
27954   __ret = vmovl_u8(__p0) - vmovl_u8(__p1);
27958 __ai uint16x8_t vsubl_u8(uint8x8_t __p0, uint8x8_t __p1) {
27960   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27969 __ai uint64x2_t vsubl_u32(uint32x2_t __p0, uint32x2_t __p1) {
27971   __ret = vmovl_u32(__p0) - vmovl_u32(__p1);
27975 __ai uint64x2_t vsubl_u32(uint32x2_t __p0, uint32x2_t __p1) {
27977   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27986 __ai uint32x4_t vsubl_u16(uint16x4_t __p0, uint16x4_t __p1) {
27988   __ret = vmovl_u16(__p0) - vmovl_u16(__p1);
27992 __ai uint32x4_t vsubl_u16(uint16x4_t __p0, uint16x4_t __p1) {
27994   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28003 __ai int16x8_t vsubl_s8(int8x8_t __p0, int8x8_t __p1) {
28005   __ret = vmovl_s8(__p0) - vmovl_s8(__p1);
28009 __ai int16x8_t vsubl_s8(int8x8_t __p0, int8x8_t __p1) {
28011   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28020 __ai int64x2_t vsubl_s32(int32x2_t __p0, int32x2_t __p1) {
28022   __ret = vmovl_s32(__p0) - vmovl_s32(__p1);
28026 __ai int64x2_t vsubl_s32(int32x2_t __p0, int32x2_t __p1) {
28028   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
28037 __ai int32x4_t vsubl_s16(int16x4_t __p0, int16x4_t __p1) {
28039   __ret = vmovl_s16(__p0) - vmovl_s16(__p1);
28043 __ai int32x4_t vsubl_s16(int16x4_t __p0, int16x4_t __p1) {
28045   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28054 __ai uint16x8_t vsubw_u8(uint16x8_t __p0, uint8x8_t __p1) {
28056   __ret = __p0 - vmovl_u8(__p1);
28060 __ai uint16x8_t vsubw_u8(uint16x8_t __p0, uint8x8_t __p1) {
28062   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28071 __ai uint64x2_t vsubw_u32(uint64x2_t __p0, uint32x2_t __p1) {
28073   __ret = __p0 - vmovl_u32(__p1);
28077 __ai uint64x2_t vsubw_u32(uint64x2_t __p0, uint32x2_t __p1) {
28079   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
28088 __ai uint32x4_t vsubw_u16(uint32x4_t __p0, uint16x4_t __p1) {
28090   __ret = __p0 - vmovl_u16(__p1);
28094 __ai uint32x4_t vsubw_u16(uint32x4_t __p0, uint16x4_t __p1) {
28096   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28105 __ai int16x8_t vsubw_s8(int16x8_t __p0, int8x8_t __p1) {
28107   __ret = __p0 - vmovl_s8(__p1);
28111 __ai int16x8_t vsubw_s8(int16x8_t __p0, int8x8_t __p1) {
28113   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28122 __ai int64x2_t vsubw_s32(int64x2_t __p0, int32x2_t __p1) {
28124   __ret = __p0 - vmovl_s32(__p1);
28128 __ai int64x2_t vsubw_s32(int64x2_t __p0, int32x2_t __p1) {
28130   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
28139 __ai int32x4_t vsubw_s16(int32x4_t __p0, int16x4_t __p1) {
28141   __ret = __p0 - vmovl_s16(__p1);
28145 __ai int32x4_t vsubw_s16(int32x4_t __p0, int16x4_t __p1) {
28147   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28156 __ai poly8x8_t vtbl1_p8(poly8x8_t __p0, uint8x8_t __p1) {
28158   __ret = (poly8x8_t) __builtin_neon_vtbl1_v((int8x8_t)__p0, (int8x8_t)__p1, 4);
28162 __ai poly8x8_t vtbl1_p8(poly8x8_t __p0, uint8x8_t __p1) {
28164   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28173 __ai uint8x8_t vtbl1_u8(uint8x8_t __p0, uint8x8_t __p1) {
28175   __ret = (uint8x8_t) __builtin_neon_vtbl1_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
28179 __ai uint8x8_t vtbl1_u8(uint8x8_t __p0, uint8x8_t __p1) {
28181   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28190 __ai int8x8_t vtbl1_s8(int8x8_t __p0, int8x8_t __p1) {
28192   __ret = (int8x8_t) __builtin_neon_vtbl1_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
28196 __ai int8x8_t vtbl1_s8(int8x8_t __p0, int8x8_t __p1) {
28198   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28207 __ai poly8x8_t vtbl2_p8(poly8x8x2_t __p0, uint8x8_t __p1) {
28209   __ret = (poly8x8_t) __builtin_neon_vtbl2_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p1, 4);
28213 __ai poly8x8_t vtbl2_p8(poly8x8x2_t __p0, uint8x8_t __p1) {
28217   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28226 __ai uint8x8_t vtbl2_u8(uint8x8x2_t __p0, uint8x8_t __p1) {
28228   __ret = (uint8x8_t) __builtin_neon_vtbl2_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p1, 16);
28232 __ai uint8x8_t vtbl2_u8(uint8x8x2_t __p0, uint8x8_t __p1) {
28236   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28245 __ai int8x8_t vtbl2_s8(int8x8x2_t __p0, int8x8_t __p1) {
28247   __ret = (int8x8_t) __builtin_neon_vtbl2_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p1, 0);
28251 __ai int8x8_t vtbl2_s8(int8x8x2_t __p0, int8x8_t __p1) {
28255   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28264 __ai poly8x8_t vtbl3_p8(poly8x8x3_t __p0, uint8x8_t __p1) {
28266   __ret = (poly8x8_t) __builtin_neon_vtbl3_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p0.val[2], (int8x8_t)__p1, 4);
28270 __ai poly8x8_t vtbl3_p8(poly8x8x3_t __p0, uint8x8_t __p1) {
28275   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28284 __ai uint8x8_t vtbl3_u8(uint8x8x3_t __p0, uint8x8_t __p1) {
28286   __ret = (uint8x8_t) __builtin_neon_vtbl3_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p0.val[2], (int8x8_t)__p1, 16);
28290 __ai uint8x8_t vtbl3_u8(uint8x8x3_t __p0, uint8x8_t __p1) {
28295   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28304 __ai int8x8_t vtbl3_s8(int8x8x3_t __p0, int8x8_t __p1) {
28306   __ret = (int8x8_t) __builtin_neon_vtbl3_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p0.val[2], (int8x8_t)__p1, 0);
28310 __ai int8x8_t vtbl3_s8(int8x8x3_t __p0, int8x8_t __p1) {
28315   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28324 __ai poly8x8_t vtbl4_p8(poly8x8x4_t __p0, uint8x8_t __p1) {
28326   __ret = (poly8x8_t) __builtin_neon_vtbl4_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p0.val[2], (int8x8_t)__p0.val[3], (int8x8_t)__p1, 4);
28330 __ai poly8x8_t vtbl4_p8(poly8x8x4_t __p0, uint8x8_t __p1) {
28336   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28345 __ai uint8x8_t vtbl4_u8(uint8x8x4_t __p0, uint8x8_t __p1) {
28347   __ret = (uint8x8_t) __builtin_neon_vtbl4_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p0.val[2], (int8x8_t)__p0.val[3], (int8x8_t)__p1, 16);
28351 __ai uint8x8_t vtbl4_u8(uint8x8x4_t __p0, uint8x8_t __p1) {
28357   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28366 __ai int8x8_t vtbl4_s8(int8x8x4_t __p0, int8x8_t __p1) {
28368   __ret = (int8x8_t) __builtin_neon_vtbl4_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p0.val[2], (int8x8_t)__p0.val[3], (int8x8_t)__p1, 0);
28372 __ai int8x8_t vtbl4_s8(int8x8x4_t __p0, int8x8_t __p1) {
28378   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28387 __ai poly8x8_t vtbx1_p8(poly8x8_t __p0, poly8x8_t __p1, uint8x8_t __p2) {
28389   __ret = (poly8x8_t) __builtin_neon_vtbx1_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 4);
28393 __ai poly8x8_t vtbx1_p8(poly8x8_t __p0, poly8x8_t __p1, uint8x8_t __p2) {
28395   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28405 __ai uint8x8_t vtbx1_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
28407   __ret = (uint8x8_t) __builtin_neon_vtbx1_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 16);
28411 __ai uint8x8_t vtbx1_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
28413   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28423 __ai int8x8_t vtbx1_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
28425   __ret = (int8x8_t) __builtin_neon_vtbx1_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 0);
28429 __ai int8x8_t vtbx1_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
28431   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28441 __ai poly8x8_t vtbx2_p8(poly8x8_t __p0, poly8x8x2_t __p1, uint8x8_t __p2) {
28443   __ret = (poly8x8_t) __builtin_neon_vtbx2_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p2, 4);
28447 __ai poly8x8_t vtbx2_p8(poly8x8_t __p0, poly8x8x2_t __p1, uint8x8_t __p2) {
28450   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28451   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28461 __ai uint8x8_t vtbx2_u8(uint8x8_t __p0, uint8x8x2_t __p1, uint8x8_t __p2) {
28463   __ret = (uint8x8_t) __builtin_neon_vtbx2_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p2, 16);
28467 __ai uint8x8_t vtbx2_u8(uint8x8_t __p0, uint8x8x2_t __p1, uint8x8_t __p2) {
28470   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28471   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28481 __ai int8x8_t vtbx2_s8(int8x8_t __p0, int8x8x2_t __p1, int8x8_t __p2) {
28483   __ret = (int8x8_t) __builtin_neon_vtbx2_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p2, 0);
28487 __ai int8x8_t vtbx2_s8(int8x8_t __p0, int8x8x2_t __p1, int8x8_t __p2) {
28490   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28491   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28501 __ai poly8x8_t vtbx3_p8(poly8x8_t __p0, poly8x8x3_t __p1, uint8x8_t __p2) {
28503   __ret = (poly8x8_t) __builtin_neon_vtbx3_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p1.val[2], (int8x8_t)__p2, 4);
28507 __ai poly8x8_t vtbx3_p8(poly8x8_t __p0, poly8x8x3_t __p1, uint8x8_t __p2) {
28510   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28511   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28512   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28522 __ai uint8x8_t vtbx3_u8(uint8x8_t __p0, uint8x8x3_t __p1, uint8x8_t __p2) {
28524   __ret = (uint8x8_t) __builtin_neon_vtbx3_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p1.val[2], (int8x8_t)__p2, 16);
28528 __ai uint8x8_t vtbx3_u8(uint8x8_t __p0, uint8x8x3_t __p1, uint8x8_t __p2) {
28531   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28532   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28533   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28543 __ai int8x8_t vtbx3_s8(int8x8_t __p0, int8x8x3_t __p1, int8x8_t __p2) {
28545   __ret = (int8x8_t) __builtin_neon_vtbx3_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p1.val[2], (int8x8_t)__p2, 0);
28549 __ai int8x8_t vtbx3_s8(int8x8_t __p0, int8x8x3_t __p1, int8x8_t __p2) {
28552   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28553   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28554   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28564 __ai poly8x8_t vtbx4_p8(poly8x8_t __p0, poly8x8x4_t __p1, uint8x8_t __p2) {
28566   __ret = (poly8x8_t) __builtin_neon_vtbx4_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p1.val[2], (int8x8_t)__p1.val[3], (int8x8_t)__p2, 4);
28570 __ai poly8x8_t vtbx4_p8(poly8x8_t __p0, poly8x8x4_t __p1, uint8x8_t __p2) {
28573   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28574   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28575   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28576   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
28586 __ai uint8x8_t vtbx4_u8(uint8x8_t __p0, uint8x8x4_t __p1, uint8x8_t __p2) {
28588   __ret = (uint8x8_t) __builtin_neon_vtbx4_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p1.val[2], (int8x8_t)__p1.val[3], (int8x8_t)__p2, 16);
28592 __ai uint8x8_t vtbx4_u8(uint8x8_t __p0, uint8x8x4_t __p1, uint8x8_t __p2) {
28595   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28596   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28597   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28598   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
28608 __ai int8x8_t vtbx4_s8(int8x8_t __p0, int8x8x4_t __p1, int8x8_t __p2) {
28610   __ret = (int8x8_t) __builtin_neon_vtbx4_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p1.val[2], (int8x8_t)__p1.val[3], (int8x8_t)__p2, 0);
28614 __ai int8x8_t vtbx4_s8(int8x8_t __p0, int8x8x4_t __p1, int8x8_t __p2) {
28617   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28618   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28619   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28620   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
28630 __ai poly8x8x2_t vtrn_p8(poly8x8_t __p0, poly8x8_t __p1) {
28632   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 4);
28636 __ai poly8x8x2_t vtrn_p8(poly8x8_t __p0, poly8x8_t __p1) {
28638   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28649 __ai poly16x4x2_t vtrn_p16(poly16x4_t __p0, poly16x4_t __p1) {
28651   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 5);
28655 __ai poly16x4x2_t vtrn_p16(poly16x4_t __p0, poly16x4_t __p1) {
28657   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28668 __ai poly8x16x2_t vtrnq_p8(poly8x16_t __p0, poly8x16_t __p1) {
28670   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 36);
28674 __ai poly8x16x2_t vtrnq_p8(poly8x16_t __p0, poly8x16_t __p1) {
28676   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
28687 __ai poly16x8x2_t vtrnq_p16(poly16x8_t __p0, poly16x8_t __p1) {
28689   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 37);
28693 __ai poly16x8x2_t vtrnq_p16(poly16x8_t __p0, poly16x8_t __p1) {
28695   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28706 __ai uint8x16x2_t vtrnq_u8(uint8x16_t __p0, uint8x16_t __p1) {
28708   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 48);
28712 __ai uint8x16x2_t vtrnq_u8(uint8x16_t __p0, uint8x16_t __p1) {
28714   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
28725 __ai uint32x4x2_t vtrnq_u32(uint32x4_t __p0, uint32x4_t __p1) {
28727   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 50);
28731 __ai uint32x4x2_t vtrnq_u32(uint32x4_t __p0, uint32x4_t __p1) {
28733   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28744 __ai uint16x8x2_t vtrnq_u16(uint16x8_t __p0, uint16x8_t __p1) {
28746   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 49);
28750 __ai uint16x8x2_t vtrnq_u16(uint16x8_t __p0, uint16x8_t __p1) {
28752   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28763 __ai int8x16x2_t vtrnq_s8(int8x16_t __p0, int8x16_t __p1) {
28765   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 32);
28769 __ai int8x16x2_t vtrnq_s8(int8x16_t __p0, int8x16_t __p1) {
28771   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
28782 __ai float32x4x2_t vtrnq_f32(float32x4_t __p0, float32x4_t __p1) {
28784   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 41);
28788 __ai float32x4x2_t vtrnq_f32(float32x4_t __p0, float32x4_t __p1) {
28790   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28801 __ai int32x4x2_t vtrnq_s32(int32x4_t __p0, int32x4_t __p1) {
28803   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 34);
28807 __ai int32x4x2_t vtrnq_s32(int32x4_t __p0, int32x4_t __p1) {
28809   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28820 __ai int16x8x2_t vtrnq_s16(int16x8_t __p0, int16x8_t __p1) {
28822   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 33);
28826 __ai int16x8x2_t vtrnq_s16(int16x8_t __p0, int16x8_t __p1) {
28828   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28839 __ai uint8x8x2_t vtrn_u8(uint8x8_t __p0, uint8x8_t __p1) {
28841   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 16);
28845 __ai uint8x8x2_t vtrn_u8(uint8x8_t __p0, uint8x8_t __p1) {
28847   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28858 __ai uint32x2x2_t vtrn_u32(uint32x2_t __p0, uint32x2_t __p1) {
28860   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 18);
28864 __ai uint32x2x2_t vtrn_u32(uint32x2_t __p0, uint32x2_t __p1) {
28866   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
28877 __ai uint16x4x2_t vtrn_u16(uint16x4_t __p0, uint16x4_t __p1) {
28879   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 17);
28883 __ai uint16x4x2_t vtrn_u16(uint16x4_t __p0, uint16x4_t __p1) {
28885   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28896 __ai int8x8x2_t vtrn_s8(int8x8_t __p0, int8x8_t __p1) {
28898   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 0);
28902 __ai int8x8x2_t vtrn_s8(int8x8_t __p0, int8x8_t __p1) {
28904   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28915 __ai float32x2x2_t vtrn_f32(float32x2_t __p0, float32x2_t __p1) {
28917   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 9);
28921 __ai float32x2x2_t vtrn_f32(float32x2_t __p0, float32x2_t __p1) {
28923   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
28934 __ai int32x2x2_t vtrn_s32(int32x2_t __p0, int32x2_t __p1) {
28936   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 2);
28940 __ai int32x2x2_t vtrn_s32(int32x2_t __p0, int32x2_t __p1) {
28942   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
28953 __ai int16x4x2_t vtrn_s16(int16x4_t __p0, int16x4_t __p1) {
28955   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 1);
28959 __ai int16x4x2_t vtrn_s16(int16x4_t __p0, int16x4_t __p1) {
28961   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28972 __ai uint8x8_t vtst_p8(poly8x8_t __p0, poly8x8_t __p1) {
28974   __ret = (uint8x8_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
28978 __ai uint8x8_t vtst_p8(poly8x8_t __p0, poly8x8_t __p1) {
28980   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28989 __ai uint16x4_t vtst_p16(poly16x4_t __p0, poly16x4_t __p1) {
28991   __ret = (uint16x4_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
28995 __ai uint16x4_t vtst_p16(poly16x4_t __p0, poly16x4_t __p1) {
28997   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29006 __ai uint8x16_t vtstq_p8(poly8x16_t __p0, poly8x16_t __p1) {
29008   __ret = (uint8x16_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
29012 __ai uint8x16_t vtstq_p8(poly8x16_t __p0, poly8x16_t __p1) {
29014   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29023 __ai uint16x8_t vtstq_p16(poly16x8_t __p0, poly16x8_t __p1) {
29025   __ret = (uint16x8_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
29029 __ai uint16x8_t vtstq_p16(poly16x8_t __p0, poly16x8_t __p1) {
29031   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29040 __ai uint8x16_t vtstq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29042   __ret = (uint8x16_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
29046 __ai uint8x16_t vtstq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29048   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29057 __ai uint32x4_t vtstq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29059   __ret = (uint32x4_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
29063 __ai uint32x4_t vtstq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29065   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29074 __ai uint16x8_t vtstq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29076   __ret = (uint16x8_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
29080 __ai uint16x8_t vtstq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29082   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29091 __ai uint8x16_t vtstq_s8(int8x16_t __p0, int8x16_t __p1) {
29093   __ret = (uint8x16_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
29097 __ai uint8x16_t vtstq_s8(int8x16_t __p0, int8x16_t __p1) {
29099   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29108 __ai uint32x4_t vtstq_s32(int32x4_t __p0, int32x4_t __p1) {
29110   __ret = (uint32x4_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
29114 __ai uint32x4_t vtstq_s32(int32x4_t __p0, int32x4_t __p1) {
29116   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29125 __ai uint16x8_t vtstq_s16(int16x8_t __p0, int16x8_t __p1) {
29127   __ret = (uint16x8_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
29131 __ai uint16x8_t vtstq_s16(int16x8_t __p0, int16x8_t __p1) {
29133   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29142 __ai uint8x8_t vtst_u8(uint8x8_t __p0, uint8x8_t __p1) {
29144   __ret = (uint8x8_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
29148 __ai uint8x8_t vtst_u8(uint8x8_t __p0, uint8x8_t __p1) {
29150   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29159 __ai uint32x2_t vtst_u32(uint32x2_t __p0, uint32x2_t __p1) {
29161   __ret = (uint32x2_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
29165 __ai uint32x2_t vtst_u32(uint32x2_t __p0, uint32x2_t __p1) {
29167   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29176 __ai uint16x4_t vtst_u16(uint16x4_t __p0, uint16x4_t __p1) {
29178   __ret = (uint16x4_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
29182 __ai uint16x4_t vtst_u16(uint16x4_t __p0, uint16x4_t __p1) {
29184   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29193 __ai uint8x8_t vtst_s8(int8x8_t __p0, int8x8_t __p1) {
29195   __ret = (uint8x8_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
29199 __ai uint8x8_t vtst_s8(int8x8_t __p0, int8x8_t __p1) {
29201   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29210 __ai uint32x2_t vtst_s32(int32x2_t __p0, int32x2_t __p1) {
29212   __ret = (uint32x2_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
29216 __ai uint32x2_t vtst_s32(int32x2_t __p0, int32x2_t __p1) {
29218   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29227 __ai uint16x4_t vtst_s16(int16x4_t __p0, int16x4_t __p1) {
29229   __ret = (uint16x4_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
29233 __ai uint16x4_t vtst_s16(int16x4_t __p0, int16x4_t __p1) {
29235   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29244 __ai poly8x8x2_t vuzp_p8(poly8x8_t __p0, poly8x8_t __p1) {
29246   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 4);
29250 __ai poly8x8x2_t vuzp_p8(poly8x8_t __p0, poly8x8_t __p1) {
29252   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29263 __ai poly16x4x2_t vuzp_p16(poly16x4_t __p0, poly16x4_t __p1) {
29265   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 5);
29269 __ai poly16x4x2_t vuzp_p16(poly16x4_t __p0, poly16x4_t __p1) {
29271   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29282 __ai poly8x16x2_t vuzpq_p8(poly8x16_t __p0, poly8x16_t __p1) {
29284   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 36);
29288 __ai poly8x16x2_t vuzpq_p8(poly8x16_t __p0, poly8x16_t __p1) {
29290   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29301 __ai poly16x8x2_t vuzpq_p16(poly16x8_t __p0, poly16x8_t __p1) {
29303   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 37);
29307 __ai poly16x8x2_t vuzpq_p16(poly16x8_t __p0, poly16x8_t __p1) {
29309   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29320 __ai uint8x16x2_t vuzpq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29322   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 48);
29326 __ai uint8x16x2_t vuzpq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29328   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29339 __ai uint32x4x2_t vuzpq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29341   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 50);
29345 __ai uint32x4x2_t vuzpq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29347   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29358 __ai uint16x8x2_t vuzpq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29360   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 49);
29364 __ai uint16x8x2_t vuzpq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29366   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29377 __ai int8x16x2_t vuzpq_s8(int8x16_t __p0, int8x16_t __p1) {
29379   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 32);
29383 __ai int8x16x2_t vuzpq_s8(int8x16_t __p0, int8x16_t __p1) {
29385   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29396 __ai float32x4x2_t vuzpq_f32(float32x4_t __p0, float32x4_t __p1) {
29398   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 41);
29402 __ai float32x4x2_t vuzpq_f32(float32x4_t __p0, float32x4_t __p1) {
29404   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29415 __ai int32x4x2_t vuzpq_s32(int32x4_t __p0, int32x4_t __p1) {
29417   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 34);
29421 __ai int32x4x2_t vuzpq_s32(int32x4_t __p0, int32x4_t __p1) {
29423   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29434 __ai int16x8x2_t vuzpq_s16(int16x8_t __p0, int16x8_t __p1) {
29436   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 33);
29440 __ai int16x8x2_t vuzpq_s16(int16x8_t __p0, int16x8_t __p1) {
29442   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29453 __ai uint8x8x2_t vuzp_u8(uint8x8_t __p0, uint8x8_t __p1) {
29455   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 16);
29459 __ai uint8x8x2_t vuzp_u8(uint8x8_t __p0, uint8x8_t __p1) {
29461   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29472 __ai uint32x2x2_t vuzp_u32(uint32x2_t __p0, uint32x2_t __p1) {
29474   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 18);
29478 __ai uint32x2x2_t vuzp_u32(uint32x2_t __p0, uint32x2_t __p1) {
29480   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29491 __ai uint16x4x2_t vuzp_u16(uint16x4_t __p0, uint16x4_t __p1) {
29493   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 17);
29497 __ai uint16x4x2_t vuzp_u16(uint16x4_t __p0, uint16x4_t __p1) {
29499   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29510 __ai int8x8x2_t vuzp_s8(int8x8_t __p0, int8x8_t __p1) {
29512   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 0);
29516 __ai int8x8x2_t vuzp_s8(int8x8_t __p0, int8x8_t __p1) {
29518   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29529 __ai float32x2x2_t vuzp_f32(float32x2_t __p0, float32x2_t __p1) {
29531   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 9);
29535 __ai float32x2x2_t vuzp_f32(float32x2_t __p0, float32x2_t __p1) {
29537   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29548 __ai int32x2x2_t vuzp_s32(int32x2_t __p0, int32x2_t __p1) {
29550   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 2);
29554 __ai int32x2x2_t vuzp_s32(int32x2_t __p0, int32x2_t __p1) {
29556   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29567 __ai int16x4x2_t vuzp_s16(int16x4_t __p0, int16x4_t __p1) {
29569   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 1);
29573 __ai int16x4x2_t vuzp_s16(int16x4_t __p0, int16x4_t __p1) {
29575   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29586 __ai poly8x8x2_t vzip_p8(poly8x8_t __p0, poly8x8_t __p1) {
29588   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 4);
29592 __ai poly8x8x2_t vzip_p8(poly8x8_t __p0, poly8x8_t __p1) {
29594   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29605 __ai poly16x4x2_t vzip_p16(poly16x4_t __p0, poly16x4_t __p1) {
29607   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 5);
29611 __ai poly16x4x2_t vzip_p16(poly16x4_t __p0, poly16x4_t __p1) {
29613   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29624 __ai poly8x16x2_t vzipq_p8(poly8x16_t __p0, poly8x16_t __p1) {
29626   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 36);
29630 __ai poly8x16x2_t vzipq_p8(poly8x16_t __p0, poly8x16_t __p1) {
29632   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29643 __ai poly16x8x2_t vzipq_p16(poly16x8_t __p0, poly16x8_t __p1) {
29645   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 37);
29649 __ai poly16x8x2_t vzipq_p16(poly16x8_t __p0, poly16x8_t __p1) {
29651   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29662 __ai uint8x16x2_t vzipq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29664   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 48);
29668 __ai uint8x16x2_t vzipq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29670   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29681 __ai uint32x4x2_t vzipq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29683   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 50);
29687 __ai uint32x4x2_t vzipq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29689   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29700 __ai uint16x8x2_t vzipq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29702   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 49);
29706 __ai uint16x8x2_t vzipq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29708   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29719 __ai int8x16x2_t vzipq_s8(int8x16_t __p0, int8x16_t __p1) {
29721   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 32);
29725 __ai int8x16x2_t vzipq_s8(int8x16_t __p0, int8x16_t __p1) {
29727   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29738 __ai float32x4x2_t vzipq_f32(float32x4_t __p0, float32x4_t __p1) {
29740   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 41);
29744 __ai float32x4x2_t vzipq_f32(float32x4_t __p0, float32x4_t __p1) {
29746   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29757 __ai int32x4x2_t vzipq_s32(int32x4_t __p0, int32x4_t __p1) {
29759   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 34);
29763 __ai int32x4x2_t vzipq_s32(int32x4_t __p0, int32x4_t __p1) {
29765   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29776 __ai int16x8x2_t vzipq_s16(int16x8_t __p0, int16x8_t __p1) {
29778   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 33);
29782 __ai int16x8x2_t vzipq_s16(int16x8_t __p0, int16x8_t __p1) {
29784   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29795 __ai uint8x8x2_t vzip_u8(uint8x8_t __p0, uint8x8_t __p1) {
29797   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 16);
29801 __ai uint8x8x2_t vzip_u8(uint8x8_t __p0, uint8x8_t __p1) {
29803   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29814 __ai uint32x2x2_t vzip_u32(uint32x2_t __p0, uint32x2_t __p1) {
29816   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 18);
29820 __ai uint32x2x2_t vzip_u32(uint32x2_t __p0, uint32x2_t __p1) {
29822   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29833 __ai uint16x4x2_t vzip_u16(uint16x4_t __p0, uint16x4_t __p1) {
29835   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 17);
29839 __ai uint16x4x2_t vzip_u16(uint16x4_t __p0, uint16x4_t __p1) {
29841   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29852 __ai int8x8x2_t vzip_s8(int8x8_t __p0, int8x8_t __p1) {
29854   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 0);
29858 __ai int8x8x2_t vzip_s8(int8x8_t __p0, int8x8_t __p1) {
29860   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29871 __ai float32x2x2_t vzip_f32(float32x2_t __p0, float32x2_t __p1) {
29873   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 9);
29877 __ai float32x2x2_t vzip_f32(float32x2_t __p0, float32x2_t __p1) {
29879   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29890 __ai int32x2x2_t vzip_s32(int32x2_t __p0, int32x2_t __p1) {
29892   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 2);
29896 __ai int32x2x2_t vzip_s32(int32x2_t __p0, int32x2_t __p1) {
29898   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29909 __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
29911   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 1);
29915 __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
29917   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
34079 __ai float32x4_t vmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
34081   __ret = (float32x4_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
34085 __ai float32x4_t vmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
34087   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
34096 __ai float32x2_t vmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
34098   __ret = (float32x2_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
34102 __ai float32x2_t vmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
34104   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
34113 __ai float32x4_t vminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
34115   __ret = (float32x4_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
34119 __ai float32x4_t vminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
34121   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
34130 __ai float32x2_t vminnm_f32(float32x2_t __p0, float32x2_t __p1) {
34132   __ret = (float32x2_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
34136 __ai float32x2_t vminnm_f32(float32x2_t __p0, float32x2_t __p1) {
34138   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40123 __ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
40125   __ret = (float64x2_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
40129 __ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
40131   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40140 __ai float64x1_t vmaxnm_f64(float64x1_t __p0, float64x1_t __p1) {
40142   __ret = (float64x1_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
40146 __ai float64x1_t vmaxnm_f64(float64x1_t __p0, float64x1_t __p1) {
40148   __ret = (float64x1_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
40154 __ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
40156   __ret = (float64x2_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
40160 __ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
40162   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40171 __ai float64x1_t vminnm_f64(float64x1_t __p0, float64x1_t __p1) {
40173   __ret = (float64x1_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
40177 __ai float64x1_t vminnm_f64(float64x1_t __p0, float64x1_t __p1) {
40179   __ret = (float64x1_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
40187 __ai uint8x16_t vaesdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
40189   __ret = (uint8x16_t) __builtin_neon_vaesdq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
40193 __ai uint8x16_t vaesdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
40195   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
40204 __ai uint8x16_t vaeseq_u8(uint8x16_t __p0, uint8x16_t __p1) {
40206   __ret = (uint8x16_t) __builtin_neon_vaeseq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
40210 __ai uint8x16_t vaeseq_u8(uint8x16_t __p0, uint8x16_t __p1) {
40212   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
40253 __ai uint32x4_t vsha1cq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
40255   __ret = (uint32x4_t) __builtin_neon_vsha1cq_u32((int8x16_t)__p0, __p1, (int8x16_t)__p2);
40259 __ai uint32x4_t vsha1cq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
40263   __ret = (uint32x4_t) __builtin_neon_vsha1cq_u32((int8x16_t)__rev0, __p1, (int8x16_t)__rev2);
40284 __ai uint32x4_t vsha1mq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
40286   __ret = (uint32x4_t) __builtin_neon_vsha1mq_u32((int8x16_t)__p0, __p1, (int8x16_t)__p2);
40290 __ai uint32x4_t vsha1mq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
40294   __ret = (uint32x4_t) __builtin_neon_vsha1mq_u32((int8x16_t)__rev0, __p1, (int8x16_t)__rev2);
40301 __ai uint32x4_t vsha1pq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
40303   __ret = (uint32x4_t) __builtin_neon_vsha1pq_u32((int8x16_t)__p0, __p1, (int8x16_t)__p2);
40307 __ai uint32x4_t vsha1pq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
40311   __ret = (uint32x4_t) __builtin_neon_vsha1pq_u32((int8x16_t)__rev0, __p1, (int8x16_t)__rev2);
40318 __ai uint32x4_t vsha1su0q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
40320   __ret = (uint32x4_t) __builtin_neon_vsha1su0q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
40324 __ai uint32x4_t vsha1su0q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
40326   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40336 __ai uint32x4_t vsha1su1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
40338   __ret = (uint32x4_t) __builtin_neon_vsha1su1q_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
40342 __ai uint32x4_t vsha1su1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
40344   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40353 __ai uint32x4_t vsha256hq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
40355   __ret = (uint32x4_t) __builtin_neon_vsha256hq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
40359 __ai uint32x4_t vsha256hq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
40361   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40371 __ai uint32x4_t vsha256h2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
40373   __ret = (uint32x4_t) __builtin_neon_vsha256h2q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
40377 __ai uint32x4_t vsha256h2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
40379   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40389 __ai uint32x4_t vsha256su0q_u32(uint32x4_t __p0, uint32x4_t __p1) {
40391   __ret = (uint32x4_t) __builtin_neon_vsha256su0q_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
40395 __ai uint32x4_t vsha256su0q_u32(uint32x4_t __p0, uint32x4_t __p1) {
40397   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40406 __ai uint32x4_t vsha256su1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
40408   __ret = (uint32x4_t) __builtin_neon_vsha256su1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
40412 __ai uint32x4_t vsha256su1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
40414   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40426 __ai int32x4_t vqrdmlahq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
40428   __ret = vqaddq_s32(__p0, vqrdmulhq_s32(__p1, __p2));
40432 __ai int32x4_t vqrdmlahq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
40434   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40444 __ai int16x8_t vqrdmlahq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
40446   __ret = vqaddq_s16(__p0, vqrdmulhq_s16(__p1, __p2));
40450 __ai int16x8_t vqrdmlahq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
40452   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
40462 __ai int32x2_t vqrdmlah_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
40464   __ret = vqadd_s32(__p0, vqrdmulh_s32(__p1, __p2));
40468 __ai int32x2_t vqrdmlah_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
40470   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40480 __ai int16x4_t vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
40482   __ret = vqadd_s16(__p0, vqrdmulh_s16(__p1, __p2));
40486 __ai int16x4_t vqrdmlah_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
40488   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40498 #define vqrdmlahq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40500   int32x4_t __s1 = __p1; \
40507 #define vqrdmlahq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40509   int32x4_t __s1 = __p1; \
40522 #define vqrdmlahq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40524   int16x8_t __s1 = __p1; \
40531 #define vqrdmlahq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40533   int16x8_t __s1 = __p1; \
40546 #define vqrdmlah_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40548   int32x2_t __s1 = __p1; \
40555 #define vqrdmlah_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40557   int32x2_t __s1 = __p1; \
40570 #define vqrdmlah_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40572   int16x4_t __s1 = __p1; \
40579 #define vqrdmlah_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40581   int16x4_t __s1 = __p1; \
40594 __ai int32x4_t vqrdmlshq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
40596   __ret = vqsubq_s32(__p0, vqrdmulhq_s32(__p1, __p2));
40600 __ai int32x4_t vqrdmlshq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
40602   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40612 __ai int16x8_t vqrdmlshq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
40614   __ret = vqsubq_s16(__p0, vqrdmulhq_s16(__p1, __p2));
40618 __ai int16x8_t vqrdmlshq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
40620   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
40630 __ai int32x2_t vqrdmlsh_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
40632   __ret = vqsub_s32(__p0, vqrdmulh_s32(__p1, __p2));
40636 __ai int32x2_t vqrdmlsh_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
40638   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40648 __ai int16x4_t vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
40650   __ret = vqsub_s16(__p0, vqrdmulh_s16(__p1, __p2));
40654 __ai int16x4_t vqrdmlsh_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
40656   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40666 #define vqrdmlshq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40668   int32x4_t __s1 = __p1; \
40675 #define vqrdmlshq_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40677   int32x4_t __s1 = __p1; \
40690 #define vqrdmlshq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40692   int16x8_t __s1 = __p1; \
40699 #define vqrdmlshq_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40701   int16x8_t __s1 = __p1; \
40714 #define vqrdmlsh_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40716   int32x2_t __s1 = __p1; \
40723 #define vqrdmlsh_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40725   int32x2_t __s1 = __p1; \
40738 #define vqrdmlsh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40740   int16x4_t __s1 = __p1; \
40747 #define vqrdmlsh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40749   int16x4_t __s1 = __p1; \
40764 #define vqrdmlahq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40766   int32x4_t __s1 = __p1; \
40773 #define vqrdmlahq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40775   int32x4_t __s1 = __p1; \
40788 #define vqrdmlahq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40790   int16x8_t __s1 = __p1; \
40797 #define vqrdmlahq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40799   int16x8_t __s1 = __p1; \
40812 #define vqrdmlah_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40814   int32x2_t __s1 = __p1; \
40821 #define vqrdmlah_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40823   int32x2_t __s1 = __p1; \
40836 #define vqrdmlah_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40838   int16x4_t __s1 = __p1; \
40845 #define vqrdmlah_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40847   int16x4_t __s1 = __p1; \
40860 #define vqrdmlshq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40862   int32x4_t __s1 = __p1; \
40869 #define vqrdmlshq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40871   int32x4_t __s1 = __p1; \
40884 #define vqrdmlshq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40886   int16x8_t __s1 = __p1; \
40893 #define vqrdmlshq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40895   int16x8_t __s1 = __p1; \
40908 #define vqrdmlsh_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40910   int32x2_t __s1 = __p1; \
40917 #define vqrdmlsh_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
40919   int32x2_t __s1 = __p1; \
40932 #define vqrdmlsh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40934   int16x4_t __s1 = __p1; \
40941 #define vqrdmlsh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
40943   int16x4_t __s1 = __p1; \
40958 __ai float64x2_t vabdq_f64(float64x2_t __p0, float64x2_t __p1) {
40960   __ret = (float64x2_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
40964 __ai float64x2_t vabdq_f64(float64x2_t __p0, float64x2_t __p1) {
40966   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40975 __ai float64x1_t vabd_f64(float64x1_t __p0, float64x1_t __p1) {
40977   __ret = (float64x1_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
40981 __ai float64x1_t vabd_f64(float64x1_t __p0, float64x1_t __p1) {
40983   __ret = (float64x1_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
40989 __ai float64_t vabdd_f64(float64_t __p0, float64_t __p1) {
40991   __ret = (float64_t) __builtin_neon_vabdd_f64(__p0, __p1);
40995 __ai float64_t vabdd_f64(float64_t __p0, float64_t __p1) {
40997   __ret = (float64_t) __builtin_neon_vabdd_f64(__p0, __p1);
41003 __ai float32_t vabds_f32(float32_t __p0, float32_t __p1) {
41005   __ret = (float32_t) __builtin_neon_vabds_f32(__p0, __p1);
41009 __ai float32_t vabds_f32(float32_t __p0, float32_t __p1) {
41011   __ret = (float32_t) __builtin_neon_vabds_f32(__p0, __p1);
41091 __ai float64x2_t vaddq_f64(float64x2_t __p0, float64x2_t __p1) {
41093   __ret = __p0 + __p1;
41097 __ai float64x2_t vaddq_f64(float64x2_t __p0, float64x2_t __p1) {
41099   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41108 __ai float64x1_t vadd_f64(float64x1_t __p0, float64x1_t __p1) {
41110   __ret = __p0 + __p1;
41114 __ai float64x1_t vadd_f64(float64x1_t __p0, float64x1_t __p1) {
41116   __ret = __p0 + __p1;
41122 __ai uint64_t vaddd_u64(uint64_t __p0, uint64_t __p1) {
41124   __ret = (uint64_t) __builtin_neon_vaddd_u64(__p0, __p1);
41128 __ai uint64_t vaddd_u64(uint64_t __p0, uint64_t __p1) {
41130   __ret = (uint64_t) __builtin_neon_vaddd_u64(__p0, __p1);
41136 __ai int64_t vaddd_s64(int64_t __p0, int64_t __p1) {
41138   __ret = (int64_t) __builtin_neon_vaddd_s64(__p0, __p1);
41142 __ai int64_t vaddd_s64(int64_t __p0, int64_t __p1) {
41144   __ret = (int64_t) __builtin_neon_vaddd_s64(__p0, __p1);
41150 __ai uint16x8_t vaddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
41152   __ret = vcombine_u16(__p0, vaddhn_u32(__p1, __p2));
41156 __ai uint16x8_t vaddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
41158   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41168 __ai uint32x4_t vaddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
41170   __ret = vcombine_u32(__p0, vaddhn_u64(__p1, __p2));
41174 __ai uint32x4_t vaddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
41176   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41186 __ai uint8x16_t vaddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
41188   __ret = vcombine_u8(__p0, vaddhn_u16(__p1, __p2));
41192 __ai uint8x16_t vaddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
41194   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
41204 __ai int16x8_t vaddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
41206   __ret = vcombine_s16(__p0, vaddhn_s32(__p1, __p2));
41210 __ai int16x8_t vaddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
41212   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
41222 __ai int32x4_t vaddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
41224   __ret = vcombine_s32(__p0, vaddhn_s64(__p1, __p2));
41228 __ai int32x4_t vaddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
41230   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41240 __ai int8x16_t vaddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
41242   __ret = vcombine_s8(__p0, vaddhn_s16(__p1, __p2));
41246 __ai int8x16_t vaddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
41248   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
41693 __ai poly64x1_t vbsl_p64(uint64x1_t __p0, poly64x1_t __p1, poly64x1_t __p2) {
41695   __ret = (poly64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 6);
41699 __ai poly64x1_t vbsl_p64(uint64x1_t __p0, poly64x1_t __p1, poly64x1_t __p2) {
41701   __ret = (poly64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 6);
41707 __ai poly64x2_t vbslq_p64(uint64x2_t __p0, poly64x2_t __p1, poly64x2_t __p2) {
41709   __ret = (poly64x2_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 38);
41713 __ai poly64x2_t vbslq_p64(uint64x2_t __p0, poly64x2_t __p1, poly64x2_t __p2) {
41715   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41725 __ai float64x2_t vbslq_f64(uint64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
41727   __ret = (float64x2_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
41731 __ai float64x2_t vbslq_f64(uint64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
41733   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41743 __ai float64x1_t vbsl_f64(uint64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
41745   __ret = (float64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
41749 __ai float64x1_t vbsl_f64(uint64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
41751   __ret = (float64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
41757 __ai uint64x2_t vcageq_f64(float64x2_t __p0, float64x2_t __p1) {
41759   __ret = (uint64x2_t) __builtin_neon_vcageq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
41763 __ai uint64x2_t vcageq_f64(float64x2_t __p0, float64x2_t __p1) {
41765   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41774 __ai uint64x1_t vcage_f64(float64x1_t __p0, float64x1_t __p1) {
41776   __ret = (uint64x1_t) __builtin_neon_vcage_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
41780 __ai uint64x1_t vcage_f64(float64x1_t __p0, float64x1_t __p1) {
41782   __ret = (uint64x1_t) __builtin_neon_vcage_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
41788 __ai uint64_t vcaged_f64(float64_t __p0, float64_t __p1) {
41790   __ret = (uint64_t) __builtin_neon_vcaged_f64(__p0, __p1);
41794 __ai uint64_t vcaged_f64(float64_t __p0, float64_t __p1) {
41796   __ret = (uint64_t) __builtin_neon_vcaged_f64(__p0, __p1);
41802 __ai uint32_t vcages_f32(float32_t __p0, float32_t __p1) {
41804   __ret = (uint32_t) __builtin_neon_vcages_f32(__p0, __p1);
41808 __ai uint32_t vcages_f32(float32_t __p0, float32_t __p1) {
41810   __ret = (uint32_t) __builtin_neon_vcages_f32(__p0, __p1);
41816 __ai uint64x2_t vcagtq_f64(float64x2_t __p0, float64x2_t __p1) {
41818   __ret = (uint64x2_t) __builtin_neon_vcagtq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
41822 __ai uint64x2_t vcagtq_f64(float64x2_t __p0, float64x2_t __p1) {
41824   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41833 __ai uint64x1_t vcagt_f64(float64x1_t __p0, float64x1_t __p1) {
41835   __ret = (uint64x1_t) __builtin_neon_vcagt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
41839 __ai uint64x1_t vcagt_f64(float64x1_t __p0, float64x1_t __p1) {
41841   __ret = (uint64x1_t) __builtin_neon_vcagt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
41847 __ai uint64_t vcagtd_f64(float64_t __p0, float64_t __p1) {
41849   __ret = (uint64_t) __builtin_neon_vcagtd_f64(__p0, __p1);
41853 __ai uint64_t vcagtd_f64(float64_t __p0, float64_t __p1) {
41855   __ret = (uint64_t) __builtin_neon_vcagtd_f64(__p0, __p1);
41861 __ai uint32_t vcagts_f32(float32_t __p0, float32_t __p1) {
41863   __ret = (uint32_t) __builtin_neon_vcagts_f32(__p0, __p1);
41867 __ai uint32_t vcagts_f32(float32_t __p0, float32_t __p1) {
41869   __ret = (uint32_t) __builtin_neon_vcagts_f32(__p0, __p1);
41875 __ai uint64x2_t vcaleq_f64(float64x2_t __p0, float64x2_t __p1) {
41877   __ret = (uint64x2_t) __builtin_neon_vcaleq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
41881 __ai uint64x2_t vcaleq_f64(float64x2_t __p0, float64x2_t __p1) {
41883   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41892 __ai uint64x1_t vcale_f64(float64x1_t __p0, float64x1_t __p1) {
41894   __ret = (uint64x1_t) __builtin_neon_vcale_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
41898 __ai uint64x1_t vcale_f64(float64x1_t __p0, float64x1_t __p1) {
41900   __ret = (uint64x1_t) __builtin_neon_vcale_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
41906 __ai uint64_t vcaled_f64(float64_t __p0, float64_t __p1) {
41908   __ret = (uint64_t) __builtin_neon_vcaled_f64(__p0, __p1);
41912 __ai uint64_t vcaled_f64(float64_t __p0, float64_t __p1) {
41914   __ret = (uint64_t) __builtin_neon_vcaled_f64(__p0, __p1);
41920 __ai uint32_t vcales_f32(float32_t __p0, float32_t __p1) {
41922   __ret = (uint32_t) __builtin_neon_vcales_f32(__p0, __p1);
41926 __ai uint32_t vcales_f32(float32_t __p0, float32_t __p1) {
41928   __ret = (uint32_t) __builtin_neon_vcales_f32(__p0, __p1);
41934 __ai uint64x2_t vcaltq_f64(float64x2_t __p0, float64x2_t __p1) {
41936   __ret = (uint64x2_t) __builtin_neon_vcaltq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
41940 __ai uint64x2_t vcaltq_f64(float64x2_t __p0, float64x2_t __p1) {
41942   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41951 __ai uint64x1_t vcalt_f64(float64x1_t __p0, float64x1_t __p1) {
41953   __ret = (uint64x1_t) __builtin_neon_vcalt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
41957 __ai uint64x1_t vcalt_f64(float64x1_t __p0, float64x1_t __p1) {
41959   __ret = (uint64x1_t) __builtin_neon_vcalt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
41965 __ai uint64_t vcaltd_f64(float64_t __p0, float64_t __p1) {
41967   __ret = (uint64_t) __builtin_neon_vcaltd_f64(__p0, __p1);
41971 __ai uint64_t vcaltd_f64(float64_t __p0, float64_t __p1) {
41973   __ret = (uint64_t) __builtin_neon_vcaltd_f64(__p0, __p1);
41979 __ai uint32_t vcalts_f32(float32_t __p0, float32_t __p1) {
41981   __ret = (uint32_t) __builtin_neon_vcalts_f32(__p0, __p1);
41985 __ai uint32_t vcalts_f32(float32_t __p0, float32_t __p1) {
41987   __ret = (uint32_t) __builtin_neon_vcalts_f32(__p0, __p1);
41993 __ai uint64x1_t vceq_p64(poly64x1_t __p0, poly64x1_t __p1) {
41995   __ret = (uint64x1_t)(__p0 == __p1);
41999 __ai uint64x1_t vceq_p64(poly64x1_t __p0, poly64x1_t __p1) {
42001   __ret = (uint64x1_t)(__p0 == __p1);
42007 __ai uint64x2_t vceqq_p64(poly64x2_t __p0, poly64x2_t __p1) {
42009   __ret = (uint64x2_t)(__p0 == __p1);
42013 __ai uint64x2_t vceqq_p64(poly64x2_t __p0, poly64x2_t __p1) {
42015   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42024 __ai uint64x2_t vceqq_u64(uint64x2_t __p0, uint64x2_t __p1) {
42026   __ret = (uint64x2_t)(__p0 == __p1);
42030 __ai uint64x2_t vceqq_u64(uint64x2_t __p0, uint64x2_t __p1) {
42032   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42041 __ai uint64x2_t vceqq_f64(float64x2_t __p0, float64x2_t __p1) {
42043   __ret = (uint64x2_t)(__p0 == __p1);
42047 __ai uint64x2_t vceqq_f64(float64x2_t __p0, float64x2_t __p1) {
42049   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42058 __ai uint64x2_t vceqq_s64(int64x2_t __p0, int64x2_t __p1) {
42060   __ret = (uint64x2_t)(__p0 == __p1);
42064 __ai uint64x2_t vceqq_s64(int64x2_t __p0, int64x2_t __p1) {
42066   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42075 __ai uint64x1_t vceq_u64(uint64x1_t __p0, uint64x1_t __p1) {
42077   __ret = (uint64x1_t)(__p0 == __p1);
42081 __ai uint64x1_t vceq_u64(uint64x1_t __p0, uint64x1_t __p1) {
42083   __ret = (uint64x1_t)(__p0 == __p1);
42089 __ai uint64x1_t vceq_f64(float64x1_t __p0, float64x1_t __p1) {
42091   __ret = (uint64x1_t)(__p0 == __p1);
42095 __ai uint64x1_t vceq_f64(float64x1_t __p0, float64x1_t __p1) {
42097   __ret = (uint64x1_t)(__p0 == __p1);
42103 __ai uint64x1_t vceq_s64(int64x1_t __p0, int64x1_t __p1) {
42105   __ret = (uint64x1_t)(__p0 == __p1);
42109 __ai uint64x1_t vceq_s64(int64x1_t __p0, int64x1_t __p1) {
42111   __ret = (uint64x1_t)(__p0 == __p1);
42117 __ai uint64_t vceqd_u64(uint64_t __p0, uint64_t __p1) {
42119   __ret = (uint64_t) __builtin_neon_vceqd_u64(__p0, __p1);
42123 __ai uint64_t vceqd_u64(uint64_t __p0, uint64_t __p1) {
42125   __ret = (uint64_t) __builtin_neon_vceqd_u64(__p0, __p1);
42131 __ai int64_t vceqd_s64(int64_t __p0, int64_t __p1) {
42133   __ret = (int64_t) __builtin_neon_vceqd_s64(__p0, __p1);
42137 __ai int64_t vceqd_s64(int64_t __p0, int64_t __p1) {
42139   __ret = (int64_t) __builtin_neon_vceqd_s64(__p0, __p1);
42145 __ai uint64_t vceqd_f64(float64_t __p0, float64_t __p1) {
42147   __ret = (uint64_t) __builtin_neon_vceqd_f64(__p0, __p1);
42151 __ai uint64_t vceqd_f64(float64_t __p0, float64_t __p1) {
42153   __ret = (uint64_t) __builtin_neon_vceqd_f64(__p0, __p1);
42159 __ai uint32_t vceqs_f32(float32_t __p0, float32_t __p1) {
42161   __ret = (uint32_t) __builtin_neon_vceqs_f32(__p0, __p1);
42165 __ai uint32_t vceqs_f32(float32_t __p0, float32_t __p1) {
42167   __ret = (uint32_t) __builtin_neon_vceqs_f32(__p0, __p1);
42637 __ai uint64x2_t vcgeq_u64(uint64x2_t __p0, uint64x2_t __p1) {
42639   __ret = (uint64x2_t)(__p0 >= __p1);
42643 __ai uint64x2_t vcgeq_u64(uint64x2_t __p0, uint64x2_t __p1) {
42645   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42654 __ai uint64x2_t vcgeq_f64(float64x2_t __p0, float64x2_t __p1) {
42656   __ret = (uint64x2_t)(__p0 >= __p1);
42660 __ai uint64x2_t vcgeq_f64(float64x2_t __p0, float64x2_t __p1) {
42662   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42671 __ai uint64x2_t vcgeq_s64(int64x2_t __p0, int64x2_t __p1) {
42673   __ret = (uint64x2_t)(__p0 >= __p1);
42677 __ai uint64x2_t vcgeq_s64(int64x2_t __p0, int64x2_t __p1) {
42679   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42688 __ai uint64x1_t vcge_u64(uint64x1_t __p0, uint64x1_t __p1) {
42690   __ret = (uint64x1_t)(__p0 >= __p1);
42694 __ai uint64x1_t vcge_u64(uint64x1_t __p0, uint64x1_t __p1) {
42696   __ret = (uint64x1_t)(__p0 >= __p1);
42702 __ai uint64x1_t vcge_f64(float64x1_t __p0, float64x1_t __p1) {
42704   __ret = (uint64x1_t)(__p0 >= __p1);
42708 __ai uint64x1_t vcge_f64(float64x1_t __p0, float64x1_t __p1) {
42710   __ret = (uint64x1_t)(__p0 >= __p1);
42716 __ai uint64x1_t vcge_s64(int64x1_t __p0, int64x1_t __p1) {
42718   __ret = (uint64x1_t)(__p0 >= __p1);
42722 __ai uint64x1_t vcge_s64(int64x1_t __p0, int64x1_t __p1) {
42724   __ret = (uint64x1_t)(__p0 >= __p1);
42730 __ai int64_t vcged_s64(int64_t __p0, int64_t __p1) {
42732   __ret = (int64_t) __builtin_neon_vcged_s64(__p0, __p1);
42736 __ai int64_t vcged_s64(int64_t __p0, int64_t __p1) {
42738   __ret = (int64_t) __builtin_neon_vcged_s64(__p0, __p1);
42744 __ai uint64_t vcged_u64(uint64_t __p0, uint64_t __p1) {
42746   __ret = (uint64_t) __builtin_neon_vcged_u64(__p0, __p1);
42750 __ai uint64_t vcged_u64(uint64_t __p0, uint64_t __p1) {
42752   __ret = (uint64_t) __builtin_neon_vcged_u64(__p0, __p1);
42758 __ai uint64_t vcged_f64(float64_t __p0, float64_t __p1) {
42760   __ret = (uint64_t) __builtin_neon_vcged_f64(__p0, __p1);
42764 __ai uint64_t vcged_f64(float64_t __p0, float64_t __p1) {
42766   __ret = (uint64_t) __builtin_neon_vcged_f64(__p0, __p1);
42772 __ai uint32_t vcges_f32(float32_t __p0, float32_t __p1) {
42774   __ret = (uint32_t) __builtin_neon_vcges_f32(__p0, __p1);
42778 __ai uint32_t vcges_f32(float32_t __p0, float32_t __p1) {
42780   __ret = (uint32_t) __builtin_neon_vcges_f32(__p0, __p1);
43016 __ai uint64x2_t vcgtq_u64(uint64x2_t __p0, uint64x2_t __p1) {
43018   __ret = (uint64x2_t)(__p0 > __p1);
43022 __ai uint64x2_t vcgtq_u64(uint64x2_t __p0, uint64x2_t __p1) {
43024   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43033 __ai uint64x2_t vcgtq_f64(float64x2_t __p0, float64x2_t __p1) {
43035   __ret = (uint64x2_t)(__p0 > __p1);
43039 __ai uint64x2_t vcgtq_f64(float64x2_t __p0, float64x2_t __p1) {
43041   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43050 __ai uint64x2_t vcgtq_s64(int64x2_t __p0, int64x2_t __p1) {
43052   __ret = (uint64x2_t)(__p0 > __p1);
43056 __ai uint64x2_t vcgtq_s64(int64x2_t __p0, int64x2_t __p1) {
43058   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43067 __ai uint64x1_t vcgt_u64(uint64x1_t __p0, uint64x1_t __p1) {
43069   __ret = (uint64x1_t)(__p0 > __p1);
43073 __ai uint64x1_t vcgt_u64(uint64x1_t __p0, uint64x1_t __p1) {
43075   __ret = (uint64x1_t)(__p0 > __p1);
43081 __ai uint64x1_t vcgt_f64(float64x1_t __p0, float64x1_t __p1) {
43083   __ret = (uint64x1_t)(__p0 > __p1);
43087 __ai uint64x1_t vcgt_f64(float64x1_t __p0, float64x1_t __p1) {
43089   __ret = (uint64x1_t)(__p0 > __p1);
43095 __ai uint64x1_t vcgt_s64(int64x1_t __p0, int64x1_t __p1) {
43097   __ret = (uint64x1_t)(__p0 > __p1);
43101 __ai uint64x1_t vcgt_s64(int64x1_t __p0, int64x1_t __p1) {
43103   __ret = (uint64x1_t)(__p0 > __p1);
43109 __ai int64_t vcgtd_s64(int64_t __p0, int64_t __p1) {
43111   __ret = (int64_t) __builtin_neon_vcgtd_s64(__p0, __p1);
43115 __ai int64_t vcgtd_s64(int64_t __p0, int64_t __p1) {
43117   __ret = (int64_t) __builtin_neon_vcgtd_s64(__p0, __p1);
43123 __ai uint64_t vcgtd_u64(uint64_t __p0, uint64_t __p1) {
43125   __ret = (uint64_t) __builtin_neon_vcgtd_u64(__p0, __p1);
43129 __ai uint64_t vcgtd_u64(uint64_t __p0, uint64_t __p1) {
43131   __ret = (uint64_t) __builtin_neon_vcgtd_u64(__p0, __p1);
43137 __ai uint64_t vcgtd_f64(float64_t __p0, float64_t __p1) {
43139   __ret = (uint64_t) __builtin_neon_vcgtd_f64(__p0, __p1);
43143 __ai uint64_t vcgtd_f64(float64_t __p0, float64_t __p1) {
43145   __ret = (uint64_t) __builtin_neon_vcgtd_f64(__p0, __p1);
43151 __ai uint32_t vcgts_f32(float32_t __p0, float32_t __p1) {
43153   __ret = (uint32_t) __builtin_neon_vcgts_f32(__p0, __p1);
43157 __ai uint32_t vcgts_f32(float32_t __p0, float32_t __p1) {
43159   __ret = (uint32_t) __builtin_neon_vcgts_f32(__p0, __p1);
43395 __ai uint64x2_t vcleq_u64(uint64x2_t __p0, uint64x2_t __p1) {
43397   __ret = (uint64x2_t)(__p0 <= __p1);
43401 __ai uint64x2_t vcleq_u64(uint64x2_t __p0, uint64x2_t __p1) {
43403   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43412 __ai uint64x2_t vcleq_f64(float64x2_t __p0, float64x2_t __p1) {
43414   __ret = (uint64x2_t)(__p0 <= __p1);
43418 __ai uint64x2_t vcleq_f64(float64x2_t __p0, float64x2_t __p1) {
43420   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43429 __ai uint64x2_t vcleq_s64(int64x2_t __p0, int64x2_t __p1) {
43431   __ret = (uint64x2_t)(__p0 <= __p1);
43435 __ai uint64x2_t vcleq_s64(int64x2_t __p0, int64x2_t __p1) {
43437   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43446 __ai uint64x1_t vcle_u64(uint64x1_t __p0, uint64x1_t __p1) {
43448   __ret = (uint64x1_t)(__p0 <= __p1);
43452 __ai uint64x1_t vcle_u64(uint64x1_t __p0, uint64x1_t __p1) {
43454   __ret = (uint64x1_t)(__p0 <= __p1);
43460 __ai uint64x1_t vcle_f64(float64x1_t __p0, float64x1_t __p1) {
43462   __ret = (uint64x1_t)(__p0 <= __p1);
43466 __ai uint64x1_t vcle_f64(float64x1_t __p0, float64x1_t __p1) {
43468   __ret = (uint64x1_t)(__p0 <= __p1);
43474 __ai uint64x1_t vcle_s64(int64x1_t __p0, int64x1_t __p1) {
43476   __ret = (uint64x1_t)(__p0 <= __p1);
43480 __ai uint64x1_t vcle_s64(int64x1_t __p0, int64x1_t __p1) {
43482   __ret = (uint64x1_t)(__p0 <= __p1);
43488 __ai uint64_t vcled_u64(uint64_t __p0, uint64_t __p1) {
43490   __ret = (uint64_t) __builtin_neon_vcled_u64(__p0, __p1);
43494 __ai uint64_t vcled_u64(uint64_t __p0, uint64_t __p1) {
43496   __ret = (uint64_t) __builtin_neon_vcled_u64(__p0, __p1);
43502 __ai int64_t vcled_s64(int64_t __p0, int64_t __p1) {
43504   __ret = (int64_t) __builtin_neon_vcled_s64(__p0, __p1);
43508 __ai int64_t vcled_s64(int64_t __p0, int64_t __p1) {
43510   __ret = (int64_t) __builtin_neon_vcled_s64(__p0, __p1);
43516 __ai uint64_t vcled_f64(float64_t __p0, float64_t __p1) {
43518   __ret = (uint64_t) __builtin_neon_vcled_f64(__p0, __p1);
43522 __ai uint64_t vcled_f64(float64_t __p0, float64_t __p1) {
43524   __ret = (uint64_t) __builtin_neon_vcled_f64(__p0, __p1);
43530 __ai uint32_t vcles_f32(float32_t __p0, float32_t __p1) {
43532   __ret = (uint32_t) __builtin_neon_vcles_f32(__p0, __p1);
43536 __ai uint32_t vcles_f32(float32_t __p0, float32_t __p1) {
43538   __ret = (uint32_t) __builtin_neon_vcles_f32(__p0, __p1);
43774 __ai uint64x2_t vcltq_u64(uint64x2_t __p0, uint64x2_t __p1) {
43776   __ret = (uint64x2_t)(__p0 < __p1);
43780 __ai uint64x2_t vcltq_u64(uint64x2_t __p0, uint64x2_t __p1) {
43782   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43791 __ai uint64x2_t vcltq_f64(float64x2_t __p0, float64x2_t __p1) {
43793   __ret = (uint64x2_t)(__p0 < __p1);
43797 __ai uint64x2_t vcltq_f64(float64x2_t __p0, float64x2_t __p1) {
43799   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43808 __ai uint64x2_t vcltq_s64(int64x2_t __p0, int64x2_t __p1) {
43810   __ret = (uint64x2_t)(__p0 < __p1);
43814 __ai uint64x2_t vcltq_s64(int64x2_t __p0, int64x2_t __p1) {
43816   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
43825 __ai uint64x1_t vclt_u64(uint64x1_t __p0, uint64x1_t __p1) {
43827   __ret = (uint64x1_t)(__p0 < __p1);
43831 __ai uint64x1_t vclt_u64(uint64x1_t __p0, uint64x1_t __p1) {
43833   __ret = (uint64x1_t)(__p0 < __p1);
43839 __ai uint64x1_t vclt_f64(float64x1_t __p0, float64x1_t __p1) {
43841   __ret = (uint64x1_t)(__p0 < __p1);
43845 __ai uint64x1_t vclt_f64(float64x1_t __p0, float64x1_t __p1) {
43847   __ret = (uint64x1_t)(__p0 < __p1);
43853 __ai uint64x1_t vclt_s64(int64x1_t __p0, int64x1_t __p1) {
43855   __ret = (uint64x1_t)(__p0 < __p1);
43859 __ai uint64x1_t vclt_s64(int64x1_t __p0, int64x1_t __p1) {
43861   __ret = (uint64x1_t)(__p0 < __p1);
43867 __ai uint64_t vcltd_u64(uint64_t __p0, uint64_t __p1) {
43869   __ret = (uint64_t) __builtin_neon_vcltd_u64(__p0, __p1);
43873 __ai uint64_t vcltd_u64(uint64_t __p0, uint64_t __p1) {
43875   __ret = (uint64_t) __builtin_neon_vcltd_u64(__p0, __p1);
43881 __ai int64_t vcltd_s64(int64_t __p0, int64_t __p1) {
43883   __ret = (int64_t) __builtin_neon_vcltd_s64(__p0, __p1);
43887 __ai int64_t vcltd_s64(int64_t __p0, int64_t __p1) {
43889   __ret = (int64_t) __builtin_neon_vcltd_s64(__p0, __p1);
43895 __ai uint64_t vcltd_f64(float64_t __p0, float64_t __p1) {
43897   __ret = (uint64_t) __builtin_neon_vcltd_f64(__p0, __p1);
43901 __ai uint64_t vcltd_f64(float64_t __p0, float64_t __p1) {
43903   __ret = (uint64_t) __builtin_neon_vcltd_f64(__p0, __p1);
43909 __ai uint32_t vclts_f32(float32_t __p0, float32_t __p1) {
43911   __ret = (uint32_t) __builtin_neon_vclts_f32(__p0, __p1);
43915 __ai uint32_t vclts_f32(float32_t __p0, float32_t __p1) {
43917   __ret = (uint32_t) __builtin_neon_vclts_f32(__p0, __p1);
44153 __ai poly64x2_t vcombine_p64(poly64x1_t __p0, poly64x1_t __p1) {
44155   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
44159 __ai poly64x2_t vcombine_p64(poly64x1_t __p0, poly64x1_t __p1) {
44161   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
44168 __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
44170   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
44174 __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
44176   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
45281 __ai float16x8_t vcvt_high_f16_f32(float16x4_t __p0, float32x4_t __p1) {
45283   __ret = vcombine_f16(__p0, vcvt_f16_f32(__p1));
45287 __ai float16x8_t vcvt_high_f16_f32(float16x4_t __p0, float32x4_t __p1) {
45289   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
45314 __ai float32x4_t vcvt_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
45316   __ret = vcombine_f32(__p0, vcvt_f32_f64(__p1));
45320 __ai float32x4_t vcvt_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
45322   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
45347 #define vcvts_n_f32_u32(__p0, __p1) __extension__ ({ \
45350   __ret = (float32_t) __builtin_neon_vcvts_n_f32_u32(__s0, __p1); \
45354 #define vcvts_n_f32_u32(__p0, __p1) __extension__ ({ \
45357   __ret = (float32_t) __builtin_neon_vcvts_n_f32_u32(__s0, __p1); \
45363 #define vcvts_n_f32_s32(__p0, __p1) __extension__ ({ \
45366   __ret = (float32_t) __builtin_neon_vcvts_n_f32_s32(__s0, __p1); \
45370 #define vcvts_n_f32_s32(__p0, __p1) __extension__ ({ \
45373   __ret = (float32_t) __builtin_neon_vcvts_n_f32_s32(__s0, __p1); \
45379 #define vcvtq_n_f64_u64(__p0, __p1) __extension__ ({ \
45382   __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__s0, __p1, 51); \
45386 #define vcvtq_n_f64_u64(__p0, __p1) __extension__ ({ \
45390   __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__rev0, __p1, 51); \
45397 #define vcvtq_n_f64_s64(__p0, __p1) __extension__ ({ \
45400   __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__s0, __p1, 35); \
45404 #define vcvtq_n_f64_s64(__p0, __p1) __extension__ ({ \
45408   __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__rev0, __p1, 35); \
45415 #define vcvt_n_f64_u64(__p0, __p1) __extension__ ({ \
45418   __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 19); \
45422 #define vcvt_n_f64_u64(__p0, __p1) __extension__ ({ \
45425   __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 19); \
45431 #define vcvt_n_f64_s64(__p0, __p1) __extension__ ({ \
45434   __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 3); \
45438 #define vcvt_n_f64_s64(__p0, __p1) __extension__ ({ \
45441   __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 3); \
45447 #define vcvtd_n_f64_u64(__p0, __p1) __extension__ ({ \
45450   __ret = (float64_t) __builtin_neon_vcvtd_n_f64_u64(__s0, __p1); \
45454 #define vcvtd_n_f64_u64(__p0, __p1) __extension__ ({ \
45457   __ret = (float64_t) __builtin_neon_vcvtd_n_f64_u64(__s0, __p1); \
45463 #define vcvtd_n_f64_s64(__p0, __p1) __extension__ ({ \
45466   __ret = (float64_t) __builtin_neon_vcvtd_n_f64_s64(__s0, __p1); \
45470 #define vcvtd_n_f64_s64(__p0, __p1) __extension__ ({ \
45473   __ret = (float64_t) __builtin_neon_vcvtd_n_f64_s64(__s0, __p1); \
45479 #define vcvts_n_s32_f32(__p0, __p1) __extension__ ({ \
45482   __ret = (int32_t) __builtin_neon_vcvts_n_s32_f32(__s0, __p1); \
45486 #define vcvts_n_s32_f32(__p0, __p1) __extension__ ({ \
45489   __ret = (int32_t) __builtin_neon_vcvts_n_s32_f32(__s0, __p1); \
45495 #define vcvtq_n_s64_f64(__p0, __p1) __extension__ ({ \
45498   __ret = (int64x2_t) __builtin_neon_vcvtq_n_s64_v((int8x16_t)__s0, __p1, 35); \
45502 #define vcvtq_n_s64_f64(__p0, __p1) __extension__ ({ \
45506   __ret = (int64x2_t) __builtin_neon_vcvtq_n_s64_v((int8x16_t)__rev0, __p1, 35); \
45513 #define vcvt_n_s64_f64(__p0, __p1) __extension__ ({ \
45516   __ret = (int64x1_t) __builtin_neon_vcvt_n_s64_v((int8x8_t)__s0, __p1, 3); \
45520 #define vcvt_n_s64_f64(__p0, __p1) __extension__ ({ \
45523   __ret = (int64x1_t) __builtin_neon_vcvt_n_s64_v((int8x8_t)__s0, __p1, 3); \
45529 #define vcvtd_n_s64_f64(__p0, __p1) __extension__ ({ \
45532   __ret = (int64_t) __builtin_neon_vcvtd_n_s64_f64(__s0, __p1); \
45536 #define vcvtd_n_s64_f64(__p0, __p1) __extension__ ({ \
45539   __ret = (int64_t) __builtin_neon_vcvtd_n_s64_f64(__s0, __p1); \
45545 #define vcvts_n_u32_f32(__p0, __p1) __extension__ ({ \
45548   __ret = (uint32_t) __builtin_neon_vcvts_n_u32_f32(__s0, __p1); \
45552 #define vcvts_n_u32_f32(__p0, __p1) __extension__ ({ \
45555   __ret = (uint32_t) __builtin_neon_vcvts_n_u32_f32(__s0, __p1); \
45561 #define vcvtq_n_u64_f64(__p0, __p1) __extension__ ({ \
45564   __ret = (uint64x2_t) __builtin_neon_vcvtq_n_u64_v((int8x16_t)__s0, __p1, 51); \
45568 #define vcvtq_n_u64_f64(__p0, __p1) __extension__ ({ \
45572   __ret = (uint64x2_t) __builtin_neon_vcvtq_n_u64_v((int8x16_t)__rev0, __p1, 51); \
45579 #define vcvt_n_u64_f64(__p0, __p1) __extension__ ({ \
45582   __ret = (uint64x1_t) __builtin_neon_vcvt_n_u64_v((int8x8_t)__s0, __p1, 19); \
45586 #define vcvt_n_u64_f64(__p0, __p1) __extension__ ({ \
45589   __ret = (uint64x1_t) __builtin_neon_vcvt_n_u64_v((int8x8_t)__s0, __p1, 19); \
45595 #define vcvtd_n_u64_f64(__p0, __p1) __extension__ ({ \
45598   __ret = (uint64_t) __builtin_neon_vcvtd_n_u64_f64(__s0, __p1); \
45602 #define vcvtd_n_u64_f64(__p0, __p1) __extension__ ({ \
45605   __ret = (uint64_t) __builtin_neon_vcvtd_n_u64_f64(__s0, __p1); \
45986 __ai float32x4_t vcvtx_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
45988   __ret = vcombine_f32(__p0, vcvtx_f32_f64(__p1));
45992 __ai float32x4_t vcvtx_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
45994   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
46003 __ai float64x2_t vdivq_f64(float64x2_t __p0, float64x2_t __p1) {
46005   __ret = __p0 / __p1;
46009 __ai float64x2_t vdivq_f64(float64x2_t __p0, float64x2_t __p1) {
46011   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
46020 __ai float32x4_t vdivq_f32(float32x4_t __p0, float32x4_t __p1) {
46022   __ret = __p0 / __p1;
46026 __ai float32x4_t vdivq_f32(float32x4_t __p0, float32x4_t __p1) {
46028   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46037 __ai float64x1_t vdiv_f64(float64x1_t __p0, float64x1_t __p1) {
46039   __ret = __p0 / __p1;
46043 __ai float64x1_t vdiv_f64(float64x1_t __p0, float64x1_t __p1) {
46045   __ret = __p0 / __p1;
46051 __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
46053   __ret = __p0 / __p1;
46057 __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
46059   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
46068 #define vdupb_lane_p8(__p0, __p1) __extension__ ({ \
46071   __ret = (poly8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
46075 #define vdupb_lane_p8(__p0, __p1) __extension__ ({ \
46079   __ret = (poly8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
46085 #define vduph_lane_p16(__p0, __p1) __extension__ ({ \
46088   __ret = (poly16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__s0, __p1); \
46092 #define vduph_lane_p16(__p0, __p1) __extension__ ({ \
46096   __ret = (poly16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__rev0, __p1); \
46102 #define vdupb_lane_u8(__p0, __p1) __extension__ ({ \
46105   __ret = (uint8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
46109 #define vdupb_lane_u8(__p0, __p1) __extension__ ({ \
46113   __ret = (uint8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
46119 #define vdups_lane_u32(__p0, __p1) __extension__ ({ \
46122   __ret = (uint32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__s0, __p1); \
46126 #define vdups_lane_u32(__p0, __p1) __extension__ ({ \
46130   __ret = (uint32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__rev0, __p1); \
46136 #define vdupd_lane_u64(__p0, __p1) __extension__ ({ \
46139   __ret = (uint64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
46143 #define vdupd_lane_u64(__p0, __p1) __extension__ ({ \
46146   __ret = (uint64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
46152 #define vduph_lane_u16(__p0, __p1) __extension__ ({ \
46155   __ret = (uint16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__s0, __p1); \
46159 #define vduph_lane_u16(__p0, __p1) __extension__ ({ \
46163   __ret = (uint16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__rev0, __p1); \
46169 #define vdupb_lane_s8(__p0, __p1) __extension__ ({ \
46172   __ret = (int8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
46176 #define vdupb_lane_s8(__p0, __p1) __extension__ ({ \
46180   __ret = (int8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
46186 #define vdupd_lane_f64(__p0, __p1) __extension__ ({ \
46189   __ret = (float64_t) __builtin_neon_vdupd_lane_f64((int8x8_t)__s0, __p1); \
46193 #define vdupd_lane_f64(__p0, __p1) __extension__ ({ \
46196   __ret = (float64_t) __builtin_neon_vdupd_lane_f64((int8x8_t)__s0, __p1); \
46202 #define vdups_lane_f32(__p0, __p1) __extension__ ({ \
46205   __ret = (float32_t) __builtin_neon_vdups_lane_f32((int8x8_t)__s0, __p1); \
46209 #define vdups_lane_f32(__p0, __p1) __extension__ ({ \
46213   __ret = (float32_t) __builtin_neon_vdups_lane_f32((int8x8_t)__rev0, __p1); \
46219 #define vdups_lane_s32(__p0, __p1) __extension__ ({ \
46222   __ret = (int32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__s0, __p1); \
46226 #define vdups_lane_s32(__p0, __p1) __extension__ ({ \
46230   __ret = (int32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__rev0, __p1); \
46236 #define vdupd_lane_s64(__p0, __p1) __extension__ ({ \
46239   __ret = (int64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
46243 #define vdupd_lane_s64(__p0, __p1) __extension__ ({ \
46246   __ret = (int64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
46252 #define vduph_lane_s16(__p0, __p1) __extension__ ({ \
46255   __ret = (int16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__s0, __p1); \
46259 #define vduph_lane_s16(__p0, __p1) __extension__ ({ \
46263   __ret = (int16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__rev0, __p1); \
46269 #define vdup_lane_p64(__p0, __p1) __extension__ ({ \
46272   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
46276 #define vdup_lane_p64(__p0, __p1) __extension__ ({ \
46279   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
46285 #define vdupq_lane_p64(__p0, __p1) __extension__ ({ \
46288   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
46292 #define vdupq_lane_p64(__p0, __p1) __extension__ ({ \
46295   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
46302 #define vdupq_lane_f64(__p0, __p1) __extension__ ({ \
46305   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
46309 #define vdupq_lane_f64(__p0, __p1) __extension__ ({ \
46312   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
46319 #define vdupq_lane_f16(__p0, __p1) __extension__ ({ \
46322   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46326 #define vdupq_lane_f16(__p0, __p1) __extension__ ({ \
46330   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46337 #define vdup_lane_f64(__p0, __p1) __extension__ ({ \
46340   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
46344 #define vdup_lane_f64(__p0, __p1) __extension__ ({ \
46347   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
46353 #define vdup_lane_f16(__p0, __p1) __extension__ ({ \
46356   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
46360 #define vdup_lane_f16(__p0, __p1) __extension__ ({ \
46364   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
46371 #define vdupb_laneq_p8(__p0, __p1) __extension__ ({ \
46374   __ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
46378 #define vdupb_laneq_p8(__p0, __p1) __extension__ ({ \
46382   __ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
46388 #define vduph_laneq_p16(__p0, __p1) __extension__ ({ \
46391   __ret = (poly16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__s0, __p1); \
46395 #define vduph_laneq_p16(__p0, __p1) __extension__ ({ \
46399   __ret = (poly16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__rev0, __p1); \
46405 #define vdupb_laneq_u8(__p0, __p1) __extension__ ({ \
46408   __ret = (uint8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
46412 #define vdupb_laneq_u8(__p0, __p1) __extension__ ({ \
46416   __ret = (uint8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
46422 #define vdups_laneq_u32(__p0, __p1) __extension__ ({ \
46425   __ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__s0, __p1); \
46429 #define vdups_laneq_u32(__p0, __p1) __extension__ ({ \
46433   __ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__rev0, __p1); \
46439 #define vdupd_laneq_u64(__p0, __p1) __extension__ ({ \
46442   __ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__s0, __p1); \
46446 #define vdupd_laneq_u64(__p0, __p1) __extension__ ({ \
46450   __ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__rev0, __p1); \
46456 #define vduph_laneq_u16(__p0, __p1) __extension__ ({ \
46459   __ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__s0, __p1); \
46463 #define vduph_laneq_u16(__p0, __p1) __extension__ ({ \
46467   __ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__rev0, __p1); \
46473 #define vdupb_laneq_s8(__p0, __p1) __extension__ ({ \
46476   __ret = (int8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
46480 #define vdupb_laneq_s8(__p0, __p1) __extension__ ({ \
46484   __ret = (int8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
46490 #define vdupd_laneq_f64(__p0, __p1) __extension__ ({ \
46493   __ret = (float64_t) __builtin_neon_vdupd_laneq_f64((int8x16_t)__s0, __p1); \
46497 #define vdupd_laneq_f64(__p0, __p1) __extension__ ({ \
46501   __ret = (float64_t) __builtin_neon_vdupd_laneq_f64((int8x16_t)__rev0, __p1); \
46507 #define vdups_laneq_f32(__p0, __p1) __extension__ ({ \
46510   __ret = (float32_t) __builtin_neon_vdups_laneq_f32((int8x16_t)__s0, __p1); \
46514 #define vdups_laneq_f32(__p0, __p1) __extension__ ({ \
46518   __ret = (float32_t) __builtin_neon_vdups_laneq_f32((int8x16_t)__rev0, __p1); \
46524 #define vdups_laneq_s32(__p0, __p1) __extension__ ({ \
46527   __ret = (int32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__s0, __p1); \
46531 #define vdups_laneq_s32(__p0, __p1) __extension__ ({ \
46535   __ret = (int32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__rev0, __p1); \
46541 #define vdupd_laneq_s64(__p0, __p1) __extension__ ({ \
46544   __ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__s0, __p1); \
46548 #define vdupd_laneq_s64(__p0, __p1) __extension__ ({ \
46552   __ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__rev0, __p1); \
46558 #define vduph_laneq_s16(__p0, __p1) __extension__ ({ \
46561   __ret = (int16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__s0, __p1); \
46565 #define vduph_laneq_s16(__p0, __p1) __extension__ ({ \
46569   __ret = (int16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__rev0, __p1); \
46575 #define vdup_laneq_p8(__p0, __p1) __extension__ ({ \
46578   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46582 #define vdup_laneq_p8(__p0, __p1) __extension__ ({ \
46586   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46593 #define vdup_laneq_p64(__p0, __p1) __extension__ ({ \
46596   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
46600 #define vdup_laneq_p64(__p0, __p1) __extension__ ({ \
46604   __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
46610 #define vdup_laneq_p16(__p0, __p1) __extension__ ({ \
46613   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
46617 #define vdup_laneq_p16(__p0, __p1) __extension__ ({ \
46621   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
46628 #define vdupq_laneq_p8(__p0, __p1) __extension__ ({ \
46631   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46635 #define vdupq_laneq_p8(__p0, __p1) __extension__ ({ \
46639   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46646 #define vdupq_laneq_p64(__p0, __p1) __extension__ ({ \
46649   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
46653 #define vdupq_laneq_p64(__p0, __p1) __extension__ ({ \
46657   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
46664 #define vdupq_laneq_p16(__p0, __p1) __extension__ ({ \
46667   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46671 #define vdupq_laneq_p16(__p0, __p1) __extension__ ({ \
46675   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46682 #define vdupq_laneq_u8(__p0, __p1) __extension__ ({ \
46685   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46689 #define vdupq_laneq_u8(__p0, __p1) __extension__ ({ \
46693   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46700 #define vdupq_laneq_u32(__p0, __p1) __extension__ ({ \
46703   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
46707 #define vdupq_laneq_u32(__p0, __p1) __extension__ ({ \
46711   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
46718 #define vdupq_laneq_u64(__p0, __p1) __extension__ ({ \
46721   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
46725 #define vdupq_laneq_u64(__p0, __p1) __extension__ ({ \
46729   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
46736 #define vdupq_laneq_u16(__p0, __p1) __extension__ ({ \
46739   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46743 #define vdupq_laneq_u16(__p0, __p1) __extension__ ({ \
46747   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46754 #define vdupq_laneq_s8(__p0, __p1) __extension__ ({ \
46757   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46761 #define vdupq_laneq_s8(__p0, __p1) __extension__ ({ \
46765   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46772 #define vdupq_laneq_f64(__p0, __p1) __extension__ ({ \
46775   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
46779 #define vdupq_laneq_f64(__p0, __p1) __extension__ ({ \
46783   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
46790 #define vdupq_laneq_f32(__p0, __p1) __extension__ ({ \
46793   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
46797 #define vdupq_laneq_f32(__p0, __p1) __extension__ ({ \
46801   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
46808 #define vdupq_laneq_f16(__p0, __p1) __extension__ ({ \
46811   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46815 #define vdupq_laneq_f16(__p0, __p1) __extension__ ({ \
46819   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46826 #define vdupq_laneq_s32(__p0, __p1) __extension__ ({ \
46829   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
46833 #define vdupq_laneq_s32(__p0, __p1) __extension__ ({ \
46837   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
46844 #define vdupq_laneq_s64(__p0, __p1) __extension__ ({ \
46847   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
46851 #define vdupq_laneq_s64(__p0, __p1) __extension__ ({ \
46855   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
46862 #define vdupq_laneq_s16(__p0, __p1) __extension__ ({ \
46865   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46869 #define vdupq_laneq_s16(__p0, __p1) __extension__ ({ \
46873   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46880 #define vdup_laneq_u8(__p0, __p1) __extension__ ({ \
46883   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46887 #define vdup_laneq_u8(__p0, __p1) __extension__ ({ \
46891   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46898 #define vdup_laneq_u32(__p0, __p1) __extension__ ({ \
46901   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
46905 #define vdup_laneq_u32(__p0, __p1) __extension__ ({ \
46909   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
46916 #define vdup_laneq_u64(__p0, __p1) __extension__ ({ \
46919   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
46923 #define vdup_laneq_u64(__p0, __p1) __extension__ ({ \
46927   __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
46933 #define vdup_laneq_u16(__p0, __p1) __extension__ ({ \
46936   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
46940 #define vdup_laneq_u16(__p0, __p1) __extension__ ({ \
46944   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
46951 #define vdup_laneq_s8(__p0, __p1) __extension__ ({ \
46954   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46958 #define vdup_laneq_s8(__p0, __p1) __extension__ ({ \
46962   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
46969 #define vdup_laneq_f64(__p0, __p1) __extension__ ({ \
46972   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
46976 #define vdup_laneq_f64(__p0, __p1) __extension__ ({ \
46980   __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
46986 #define vdup_laneq_f32(__p0, __p1) __extension__ ({ \
46989   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
46993 #define vdup_laneq_f32(__p0, __p1) __extension__ ({ \
46997   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
47004 #define vdup_laneq_f16(__p0, __p1) __extension__ ({ \
47007   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
47011 #define vdup_laneq_f16(__p0, __p1) __extension__ ({ \
47015   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
47022 #define vdup_laneq_s32(__p0, __p1) __extension__ ({ \
47025   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
47029 #define vdup_laneq_s32(__p0, __p1) __extension__ ({ \
47033   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
47040 #define vdup_laneq_s64(__p0, __p1) __extension__ ({ \
47043   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
47047 #define vdup_laneq_s64(__p0, __p1) __extension__ ({ \
47051   __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
47057 #define vdup_laneq_s16(__p0, __p1) __extension__ ({ \
47060   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
47064 #define vdup_laneq_s16(__p0, __p1) __extension__ ({ \
47068   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
47133 #define vext_p64(__p0, __p1, __p2) __extension__ ({ \
47135   poly64x1_t __s1 = __p1; \
47141 #define vext_p64(__p0, __p1, __p2) __extension__ ({ \
47143   poly64x1_t __s1 = __p1; \
47151 #define vextq_p64(__p0, __p1, __p2) __extension__ ({ \
47153   poly64x2_t __s1 = __p1; \
47159 #define vextq_p64(__p0, __p1, __p2) __extension__ ({ \
47161   poly64x2_t __s1 = __p1; \
47172 #define vextq_f64(__p0, __p1, __p2) __extension__ ({ \
47174   float64x2_t __s1 = __p1; \
47180 #define vextq_f64(__p0, __p1, __p2) __extension__ ({ \
47182   float64x2_t __s1 = __p1; \
47193 #define vext_f64(__p0, __p1, __p2) __extension__ ({ \
47195   float64x1_t __s1 = __p1; \
47201 #define vext_f64(__p0, __p1, __p2) __extension__ ({ \
47203   float64x1_t __s1 = __p1; \
47211 __ai float64x2_t vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
47213   __ret = (float64x2_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
47217 __ai float64x2_t vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
47219   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47226 __ai float64x2_t __noswap_vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
47228   __ret = (float64x2_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
47234 __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
47236   __ret = (float64x1_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
47240 __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
47242   __ret = (float64x1_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
47248 #define vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47250   float64_t __s1 = __p1; \
47257 #define vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47259   float64_t __s1 = __p1; \
47265 #define __noswap_vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47267   float64_t __s1 = __p1; \
47276 #define vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47278   float32_t __s1 = __p1; \
47285 #define vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47287   float32_t __s1 = __p1; \
47294 #define __noswap_vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47296   float32_t __s1 = __p1; \
47305 #define vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47307   float64x2_t __s1 = __p1; \
47314 #define vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47316   float64x2_t __s1 = __p1; \
47325 #define __noswap_vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47327   float64x2_t __s1 = __p1; \
47336 #define vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47338   float32x4_t __s1 = __p1; \
47345 #define vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47347   float32x4_t __s1 = __p1; \
47357 #define __noswap_vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47359   float32x4_t __s1 = __p1; \
47368 #define vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47370   float64x1_t __s1 = __p1; \
47377 #define vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47379   float64x1_t __s1 = __p1; \
47385 #define __noswap_vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47387   float64x1_t __s1 = __p1; \
47396 #define vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47398   float32x2_t __s1 = __p1; \
47405 #define vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47407   float32x2_t __s1 = __p1; \
47417 #define __noswap_vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47419   float32x2_t __s1 = __p1; \
47428 #define vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47430   float64_t __s1 = __p1; \
47437 #define vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47439   float64_t __s1 = __p1; \
47446 #define __noswap_vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47448   float64_t __s1 = __p1; \
47457 #define vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47459   float32_t __s1 = __p1; \
47466 #define vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47468   float32_t __s1 = __p1; \
47475 #define __noswap_vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47477   float32_t __s1 = __p1; \
47486 #define vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47488   float64x2_t __s1 = __p1; \
47495 #define vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47497   float64x2_t __s1 = __p1; \
47507 #define __noswap_vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47509   float64x2_t __s1 = __p1; \
47518 #define vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47520   float32x4_t __s1 = __p1; \
47527 #define vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47529   float32x4_t __s1 = __p1; \
47539 #define __noswap_vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47541   float32x4_t __s1 = __p1; \
47550 #define vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47552   float64x1_t __s1 = __p1; \
47559 #define vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47561   float64x1_t __s1 = __p1; \
47568 #define __noswap_vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
47570   float64x1_t __s1 = __p1; \
47579 #define vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47581   float32x2_t __s1 = __p1; \
47588 #define vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47590   float32x2_t __s1 = __p1; \
47600 #define __noswap_vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
47602   float32x2_t __s1 = __p1; \
47611 __ai float64x2_t vfmaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
47613   __ret = vfmaq_f64(__p0, __p1, (float64x2_t) {__p2, __p2});
47617 __ai float64x2_t vfmaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
47619   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47628 __ai float32x4_t vfmaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
47630   __ret = vfmaq_f32(__p0, __p1, (float32x4_t) {__p2, __p2, __p2, __p2});
47634 __ai float32x4_t vfmaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
47636   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
47645 __ai float32x2_t vfma_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
47647   __ret = vfma_f32(__p0, __p1, (float32x2_t) {__p2, __p2});
47651 __ai float32x2_t vfma_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
47653   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47662 __ai float64x2_t vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
47664   __ret = (float64x2_t) __builtin_neon_vfmsq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
47668 __ai float64x2_t vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
47670   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47677 __ai float64x2_t __noswap_vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
47679   __ret = (float64x2_t) __builtin_neon_vfmsq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
47685 __ai float32x4_t vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
47687   __ret = (float32x4_t) __builtin_neon_vfmsq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
47691 __ai float32x4_t vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
47693   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
47700 __ai float32x4_t __noswap_vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
47702   __ret = (float32x4_t) __builtin_neon_vfmsq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
47708 __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
47710   __ret = (float64x1_t) __builtin_neon_vfms_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
47714 __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
47716   __ret = (float64x1_t) __builtin_neon_vfms_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
47722 __ai float32x2_t vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
47724   __ret = (float32x2_t) __builtin_neon_vfms_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
47728 __ai float32x2_t vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
47730   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
47737 __ai float32x2_t __noswap_vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
47739   __ret = (float32x2_t) __builtin_neon_vfms_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
48012 __ai float64x2_t vfmsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
48014   __ret = vfmsq_f64(__p0, __p1, (float64x2_t) {__p2, __p2});
48018 __ai float64x2_t vfmsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
48020   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48029 __ai float32x4_t vfmsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
48031   __ret = vfmsq_f32(__p0, __p1, (float32x4_t) {__p2, __p2, __p2, __p2});
48035 __ai float32x4_t vfmsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
48037   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
48046 __ai float32x2_t vfms_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
48048   __ret = vfms_f32(__p0, __p1, (float32x2_t) {__p2, __p2});
48052 __ai float32x2_t vfms_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
48054   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
48098 #define vget_lane_p64(__p0, __p1) __extension__ ({ \
48101   __ret = (poly64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
48105 #define vget_lane_p64(__p0, __p1) __extension__ ({ \
48108   __ret = (poly64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
48111 #define __noswap_vget_lane_p64(__p0, __p1) __extension__ ({ \
48114   __ret = (poly64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
48120 #define vgetq_lane_p64(__p0, __p1) __extension__ ({ \
48123   __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
48127 #define vgetq_lane_p64(__p0, __p1) __extension__ ({ \
48131   __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__rev0, __p1); \
48134 #define __noswap_vgetq_lane_p64(__p0, __p1) __extension__ ({ \
48137   __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
48143 #define vgetq_lane_f64(__p0, __p1) __extension__ ({ \
48146   __ret = (float64_t) __builtin_neon_vgetq_lane_f64((int8x16_t)__s0, __p1); \
48150 #define vgetq_lane_f64(__p0, __p1) __extension__ ({ \
48154   __ret = (float64_t) __builtin_neon_vgetq_lane_f64((int8x16_t)__rev0, __p1); \
48157 #define __noswap_vgetq_lane_f64(__p0, __p1) __extension__ ({ \
48160   __ret = (float64_t) __builtin_neon_vgetq_lane_f64((int8x16_t)__s0, __p1); \
48166 #define vget_lane_f64(__p0, __p1) __extension__ ({ \
48169   __ret = (float64_t) __builtin_neon_vget_lane_f64((int8x8_t)__s0, __p1); \
48173 #define vget_lane_f64(__p0, __p1) __extension__ ({ \
48176   __ret = (float64_t) __builtin_neon_vget_lane_f64((int8x8_t)__s0, __p1); \
48179 #define __noswap_vget_lane_f64(__p0, __p1) __extension__ ({ \
48182   __ret = (float64_t) __builtin_neon_vget_lane_f64((int8x8_t)__s0, __p1); \
48334 #define vld1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
48335   poly64x1_t __s1 = __p1; \
48341 #define vld1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
48342   poly64x1_t __s1 = __p1; \
48350 #define vld1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
48351   poly64x2_t __s1 = __p1; \
48357 #define vld1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
48358   poly64x2_t __s1 = __p1; \
48368 #define vld1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
48369   float64x2_t __s1 = __p1; \
48375 #define vld1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
48376   float64x2_t __s1 = __p1; \
48386 #define vld1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
48387   float64x1_t __s1 = __p1; \
48393 #define vld1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
48394   float64x1_t __s1 = __p1; \
50228 #define vld2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50229   poly64x1x2_t __s1 = __p1; \
50235 #define vld2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50236   poly64x1x2_t __s1 = __p1; \
50244 #define vld2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
50245   poly8x16x2_t __s1 = __p1; \
50251 #define vld2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
50252   poly8x16x2_t __s1 = __p1; \
50266 #define vld2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50267   poly64x2x2_t __s1 = __p1; \
50273 #define vld2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50274   poly64x2x2_t __s1 = __p1; \
50288 #define vld2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
50289   uint8x16x2_t __s1 = __p1; \
50295 #define vld2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
50296   uint8x16x2_t __s1 = __p1; \
50310 #define vld2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
50311   uint64x2x2_t __s1 = __p1; \
50317 #define vld2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
50318   uint64x2x2_t __s1 = __p1; \
50332 #define vld2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
50333   int8x16x2_t __s1 = __p1; \
50339 #define vld2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
50340   int8x16x2_t __s1 = __p1; \
50354 #define vld2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
50355   float64x2x2_t __s1 = __p1; \
50361 #define vld2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
50362   float64x2x2_t __s1 = __p1; \
50376 #define vld2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
50377   int64x2x2_t __s1 = __p1; \
50383 #define vld2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
50384   int64x2x2_t __s1 = __p1; \
50398 #define vld2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
50399   uint64x1x2_t __s1 = __p1; \
50405 #define vld2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
50406   uint64x1x2_t __s1 = __p1; \
50414 #define vld2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
50415   float64x1x2_t __s1 = __p1; \
50421 #define vld2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
50422   float64x1x2_t __s1 = __p1; \
50430 #define vld2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
50431   int64x1x2_t __s1 = __p1; \
50437 #define vld2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
50438   int64x1x2_t __s1 = __p1; \
50826 #define vld3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50827   poly64x1x3_t __s1 = __p1; \
50833 #define vld3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50834   poly64x1x3_t __s1 = __p1; \
50842 #define vld3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
50843   poly8x16x3_t __s1 = __p1; \
50849 #define vld3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
50850   poly8x16x3_t __s1 = __p1; \
50866 #define vld3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50867   poly64x2x3_t __s1 = __p1; \
50873 #define vld3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50874   poly64x2x3_t __s1 = __p1; \
50890 #define vld3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
50891   uint8x16x3_t __s1 = __p1; \
50897 #define vld3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
50898   uint8x16x3_t __s1 = __p1; \
50914 #define vld3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
50915   uint64x2x3_t __s1 = __p1; \
50921 #define vld3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
50922   uint64x2x3_t __s1 = __p1; \
50938 #define vld3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
50939   int8x16x3_t __s1 = __p1; \
50945 #define vld3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
50946   int8x16x3_t __s1 = __p1; \
50962 #define vld3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
50963   float64x2x3_t __s1 = __p1; \
50969 #define vld3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
50970   float64x2x3_t __s1 = __p1; \
50986 #define vld3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
50987   int64x2x3_t __s1 = __p1; \
50993 #define vld3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
50994   int64x2x3_t __s1 = __p1; \
51010 #define vld3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
51011   uint64x1x3_t __s1 = __p1; \
51017 #define vld3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
51018   uint64x1x3_t __s1 = __p1; \
51026 #define vld3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
51027   float64x1x3_t __s1 = __p1; \
51033 #define vld3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
51034   float64x1x3_t __s1 = __p1; \
51042 #define vld3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
51043   int64x1x3_t __s1 = __p1; \
51049 #define vld3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
51050   int64x1x3_t __s1 = __p1; \
51456 #define vld4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
51457   poly64x1x4_t __s1 = __p1; \
51463 #define vld4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
51464   poly64x1x4_t __s1 = __p1; \
51472 #define vld4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
51473   poly8x16x4_t __s1 = __p1; \
51479 #define vld4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
51480   poly8x16x4_t __s1 = __p1; \
51498 #define vld4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
51499   poly64x2x4_t __s1 = __p1; \
51505 #define vld4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
51506   poly64x2x4_t __s1 = __p1; \
51524 #define vld4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
51525   uint8x16x4_t __s1 = __p1; \
51531 #define vld4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
51532   uint8x16x4_t __s1 = __p1; \
51550 #define vld4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
51551   uint64x2x4_t __s1 = __p1; \
51557 #define vld4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
51558   uint64x2x4_t __s1 = __p1; \
51576 #define vld4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
51577   int8x16x4_t __s1 = __p1; \
51583 #define vld4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
51584   int8x16x4_t __s1 = __p1; \
51602 #define vld4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
51603   float64x2x4_t __s1 = __p1; \
51609 #define vld4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
51610   float64x2x4_t __s1 = __p1; \
51628 #define vld4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
51629   int64x2x4_t __s1 = __p1; \
51635 #define vld4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
51636   int64x2x4_t __s1 = __p1; \
51654 #define vld4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
51655   uint64x1x4_t __s1 = __p1; \
51661 #define vld4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
51662   uint64x1x4_t __s1 = __p1; \
51670 #define vld4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
51671   float64x1x4_t __s1 = __p1; \
51677 #define vld4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
51678   float64x1x4_t __s1 = __p1; \
51686 #define vld4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
51687   int64x1x4_t __s1 = __p1; \
51693 #define vld4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
51694   int64x1x4_t __s1 = __p1; \
51716 __ai float64x2_t vmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
51718   __ret = (float64x2_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
51722 __ai float64x2_t vmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
51724   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51733 __ai float64x1_t vmax_f64(float64x1_t __p0, float64x1_t __p1) {
51735   __ret = (float64x1_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
51739 __ai float64x1_t vmax_f64(float64x1_t __p0, float64x1_t __p1) {
51741   __ret = (float64x1_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
52017 __ai float64x2_t vminq_f64(float64x2_t __p0, float64x2_t __p1) {
52019   __ret = (float64x2_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
52023 __ai float64x2_t vminq_f64(float64x2_t __p0, float64x2_t __p1) {
52025   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
52034 __ai float64x1_t vmin_f64(float64x1_t __p0, float64x1_t __p1) {
52036   __ret = (float64x1_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
52040 __ai float64x1_t vmin_f64(float64x1_t __p0, float64x1_t __p1) {
52042   __ret = (float64x1_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
52318 __ai float64x2_t vmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
52320   __ret = __p0 + __p1 * __p2;
52324 __ai float64x2_t vmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
52326   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
52336 __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
52338   __ret = __p0 + __p1 * __p2;
52342 __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
52344   __ret = __p0 + __p1 * __p2;
52350 #define vmlaq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52352   uint32x4_t __s1 = __p1; \
52359 #define vmlaq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52361   uint32x4_t __s1 = __p1; \
52374 #define vmlaq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52376   uint16x8_t __s1 = __p1; \
52383 #define vmlaq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52385   uint16x8_t __s1 = __p1; \
52398 #define vmlaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52400   float32x4_t __s1 = __p1; \
52407 #define vmlaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52409   float32x4_t __s1 = __p1; \
52422 #define vmlaq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52424   int32x4_t __s1 = __p1; \
52431 #define vmlaq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52433   int32x4_t __s1 = __p1; \
52446 #define vmlaq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52448   int16x8_t __s1 = __p1; \
52455 #define vmlaq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52457   int16x8_t __s1 = __p1; \
52470 #define vmla_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52472   uint32x2_t __s1 = __p1; \
52479 #define vmla_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52481   uint32x2_t __s1 = __p1; \
52494 #define vmla_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52496   uint16x4_t __s1 = __p1; \
52503 #define vmla_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52505   uint16x4_t __s1 = __p1; \
52518 #define vmla_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52520   float32x2_t __s1 = __p1; \
52527 #define vmla_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52529   float32x2_t __s1 = __p1; \
52542 #define vmla_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52544   int32x2_t __s1 = __p1; \
52551 #define vmla_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52553   int32x2_t __s1 = __p1; \
52566 #define vmla_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52568   int16x4_t __s1 = __p1; \
52575 #define vmla_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52577   int16x4_t __s1 = __p1; \
52590 __ai float64x2_t vmlaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
52592   __ret = __p0 + __p1 * (float64x2_t) {__p2, __p2};
52596 __ai float64x2_t vmlaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
52598   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
52607 #define vmlal_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52609   uint32x4_t __s1 = __p1; \
52616 #define vmlal_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52618   uint32x4_t __s1 = __p1; \
52631 #define vmlal_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52633   uint16x8_t __s1 = __p1; \
52640 #define vmlal_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52642   uint16x8_t __s1 = __p1; \
52655 #define vmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52657   int32x4_t __s1 = __p1; \
52664 #define vmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52666   int32x4_t __s1 = __p1; \
52679 #define vmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52681   int16x8_t __s1 = __p1; \
52688 #define vmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52690   int16x8_t __s1 = __p1; \
52703 #define vmlal_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52705   uint32x4_t __s1 = __p1; \
52712 #define vmlal_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52714   uint32x4_t __s1 = __p1; \
52727 #define vmlal_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52729   uint16x8_t __s1 = __p1; \
52736 #define vmlal_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52738   uint16x8_t __s1 = __p1; \
52751 #define vmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52753   int32x4_t __s1 = __p1; \
52760 #define vmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52762   int32x4_t __s1 = __p1; \
52775 #define vmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52777   int16x8_t __s1 = __p1; \
52784 #define vmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52786   int16x8_t __s1 = __p1; \
52799 #define vmlal_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52801   uint32x2_t __s1 = __p1; \
52808 #define vmlal_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52810   uint32x2_t __s1 = __p1; \
52823 #define vmlal_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52825   uint16x4_t __s1 = __p1; \
52832 #define vmlal_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52834   uint16x4_t __s1 = __p1; \
52847 #define vmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52849   int32x2_t __s1 = __p1; \
52856 #define vmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52858   int32x2_t __s1 = __p1; \
52871 #define vmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52873   int16x4_t __s1 = __p1; \
52880 #define vmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52882   int16x4_t __s1 = __p1; \
52895 __ai float64x2_t vmlsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
52897   __ret = __p0 - __p1 * __p2;
52901 __ai float64x2_t vmlsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
52903   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
52913 __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
52915   __ret = __p0 - __p1 * __p2;
52919 __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
52921   __ret = __p0 - __p1 * __p2;
52927 #define vmlsq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52929   uint32x4_t __s1 = __p1; \
52936 #define vmlsq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52938   uint32x4_t __s1 = __p1; \
52951 #define vmlsq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52953   uint16x8_t __s1 = __p1; \
52960 #define vmlsq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52962   uint16x8_t __s1 = __p1; \
52975 #define vmlsq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52977   float32x4_t __s1 = __p1; \
52984 #define vmlsq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52986   float32x4_t __s1 = __p1; \
52999 #define vmlsq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
53001   int32x4_t __s1 = __p1; \
53008 #define vmlsq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
53010   int32x4_t __s1 = __p1; \
53023 #define vmlsq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
53025   int16x8_t __s1 = __p1; \
53032 #define vmlsq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
53034   int16x8_t __s1 = __p1; \
53047 #define vmls_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
53049   uint32x2_t __s1 = __p1; \
53056 #define vmls_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
53058   uint32x2_t __s1 = __p1; \
53071 #define vmls_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
53073   uint16x4_t __s1 = __p1; \
53080 #define vmls_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
53082   uint16x4_t __s1 = __p1; \
53095 #define vmls_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
53097   float32x2_t __s1 = __p1; \
53104 #define vmls_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
53106   float32x2_t __s1 = __p1; \
53119 #define vmls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
53121   int32x2_t __s1 = __p1; \
53128 #define vmls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
53130   int32x2_t __s1 = __p1; \
53143 #define vmls_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
53145   int16x4_t __s1 = __p1; \
53152 #define vmls_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
53154   int16x4_t __s1 = __p1; \
53167 __ai float64x2_t vmlsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
53169   __ret = __p0 - __p1 * (float64x2_t) {__p2, __p2};
53173 __ai float64x2_t vmlsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
53175   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53184 #define vmlsl_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
53186   uint32x4_t __s1 = __p1; \
53193 #define vmlsl_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
53195   uint32x4_t __s1 = __p1; \
53208 #define vmlsl_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
53210   uint16x8_t __s1 = __p1; \
53217 #define vmlsl_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
53219   uint16x8_t __s1 = __p1; \
53232 #define vmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
53234   int32x4_t __s1 = __p1; \
53241 #define vmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
53243   int32x4_t __s1 = __p1; \
53256 #define vmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
53258   int16x8_t __s1 = __p1; \
53265 #define vmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
53267   int16x8_t __s1 = __p1; \
53280 #define vmlsl_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
53282   uint32x4_t __s1 = __p1; \
53289 #define vmlsl_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
53291   uint32x4_t __s1 = __p1; \
53304 #define vmlsl_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
53306   uint16x8_t __s1 = __p1; \
53313 #define vmlsl_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
53315   uint16x8_t __s1 = __p1; \
53328 #define vmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
53330   int32x4_t __s1 = __p1; \
53337 #define vmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
53339   int32x4_t __s1 = __p1; \
53352 #define vmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
53354   int16x8_t __s1 = __p1; \
53361 #define vmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
53363   int16x8_t __s1 = __p1; \
53376 #define vmlsl_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
53378   uint32x2_t __s1 = __p1; \
53385 #define vmlsl_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
53387   uint32x2_t __s1 = __p1; \
53400 #define vmlsl_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
53402   uint16x4_t __s1 = __p1; \
53409 #define vmlsl_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
53411   uint16x4_t __s1 = __p1; \
53424 #define vmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
53426   int32x2_t __s1 = __p1; \
53433 #define vmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
53435   int32x2_t __s1 = __p1; \
53448 #define vmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
53450   int16x4_t __s1 = __p1; \
53457 #define vmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
53459   int16x4_t __s1 = __p1; \
53674 __ai uint16x8_t vmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
53676   __ret = vcombine_u16(__p0, vmovn_u32(__p1));
53680 __ai uint16x8_t vmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
53682   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
53691 __ai uint32x4_t vmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
53693   __ret = vcombine_u32(__p0, vmovn_u64(__p1));
53697 __ai uint32x4_t vmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
53699   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53708 __ai uint8x16_t vmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
53710   __ret = vcombine_u8(__p0, vmovn_u16(__p1));
53714 __ai uint8x16_t vmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
53716   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
53725 __ai int16x8_t vmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
53727   __ret = vcombine_s16(__p0, vmovn_s32(__p1));
53731 __ai int16x8_t vmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
53733   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
53742 __ai int32x4_t vmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
53744   __ret = vcombine_s32(__p0, vmovn_s64(__p1));
53748 __ai int32x4_t vmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
53750   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53759 __ai int8x16_t vmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
53761   __ret = vcombine_s8(__p0, vmovn_s16(__p1));
53765 __ai int8x16_t vmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
53767   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
53776 __ai float64x2_t vmulq_f64(float64x2_t __p0, float64x2_t __p1) {
53778   __ret = __p0 * __p1;
53782 __ai float64x2_t vmulq_f64(float64x2_t __p0, float64x2_t __p1) {
53784   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53793 __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
53795   __ret = __p0 * __p1;
53799 __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
53801   __ret = __p0 * __p1;
53844 #define vmul_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53846   float64x1_t __s1 = __p1; \
53852 #define vmul_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53854   float64x1_t __s1 = __p1; \
53862 #define vmulq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53864   float64x1_t __s1 = __p1; \
53870 #define vmulq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53872   float64x1_t __s1 = __p1; \
53920 #define vmul_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
53922   float64x2_t __s1 = __p1; \
53928 #define vmul_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
53930   float64x2_t __s1 = __p1; \
53939 #define vmulq_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
53941   uint32x4_t __s1 = __p1; \
53947 #define vmulq_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
53949   uint32x4_t __s1 = __p1; \
53960 #define vmulq_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
53962   uint16x8_t __s1 = __p1; \
53968 #define vmulq_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
53970   uint16x8_t __s1 = __p1; \
53981 #define vmulq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
53983   float64x2_t __s1 = __p1; \
53989 #define vmulq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
53991   float64x2_t __s1 = __p1; \
54002 #define vmulq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
54004   float32x4_t __s1 = __p1; \
54010 #define vmulq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
54012   float32x4_t __s1 = __p1; \
54023 #define vmulq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
54025   int32x4_t __s1 = __p1; \
54031 #define vmulq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
54033   int32x4_t __s1 = __p1; \
54044 #define vmulq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
54046   int16x8_t __s1 = __p1; \
54052 #define vmulq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
54054   int16x8_t __s1 = __p1; \
54065 #define vmul_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
54067   uint32x4_t __s1 = __p1; \
54073 #define vmul_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
54075   uint32x4_t __s1 = __p1; \
54086 #define vmul_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
54088   uint16x8_t __s1 = __p1; \
54094 #define vmul_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
54096   uint16x8_t __s1 = __p1; \
54107 #define vmul_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
54109   float32x4_t __s1 = __p1; \
54115 #define vmul_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
54117   float32x4_t __s1 = __p1; \
54128 #define vmul_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
54130   int32x4_t __s1 = __p1; \
54136 #define vmul_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
54138   int32x4_t __s1 = __p1; \
54149 #define vmul_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
54151   int16x8_t __s1 = __p1; \
54157 #define vmul_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
54159   int16x8_t __s1 = __p1; \
54170 __ai float64x1_t vmul_n_f64(float64x1_t __p0, float64_t __p1) {
54172   __ret = (float64x1_t) __builtin_neon_vmul_n_f64((int8x8_t)__p0, __p1);
54176 __ai float64x1_t vmul_n_f64(float64x1_t __p0, float64_t __p1) {
54178   __ret = (float64x1_t) __builtin_neon_vmul_n_f64((int8x8_t)__p0, __p1);
54184 __ai float64x2_t vmulq_n_f64(float64x2_t __p0, float64_t __p1) {
54186   __ret = __p0 * (float64x2_t) {__p1, __p1};
54190 __ai float64x2_t vmulq_n_f64(float64x2_t __p0, float64_t __p1) {
54193   __ret = __rev0 * (float64x2_t) {__p1, __p1};
54200 __ai poly128_t vmull_p64(poly64_t __p0, poly64_t __p1) {
54202   __ret = (poly128_t) __builtin_neon_vmull_p64(__p0, __p1);
54206 __ai poly128_t vmull_p64(poly64_t __p0, poly64_t __p1) {
54208   __ret = (poly128_t) __builtin_neon_vmull_p64(__p0, __p1);
54211 __ai poly128_t __noswap_vmull_p64(poly64_t __p0, poly64_t __p1) {
54213   __ret = (poly128_t) __builtin_neon_vmull_p64(__p0, __p1);
54219 __ai poly16x8_t vmull_high_p8(poly8x16_t __p0, poly8x16_t __p1) {
54221   __ret = vmull_p8(vget_high_p8(__p0), vget_high_p8(__p1));
54225 __ai poly16x8_t vmull_high_p8(poly8x16_t __p0, poly8x16_t __p1) {
54227   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54236 __ai uint16x8_t vmull_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
54238   __ret = vmull_u8(vget_high_u8(__p0), vget_high_u8(__p1));
54242 __ai uint16x8_t vmull_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
54244   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54253 __ai uint64x2_t vmull_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
54255   __ret = vmull_u32(vget_high_u32(__p0), vget_high_u32(__p1));
54259 __ai uint64x2_t vmull_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
54261   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54270 __ai uint32x4_t vmull_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
54272   __ret = vmull_u16(vget_high_u16(__p0), vget_high_u16(__p1));
54276 __ai uint32x4_t vmull_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
54278   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
54287 __ai int16x8_t vmull_high_s8(int8x16_t __p0, int8x16_t __p1) {
54289   __ret = vmull_s8(vget_high_s8(__p0), vget_high_s8(__p1));
54293 __ai int16x8_t vmull_high_s8(int8x16_t __p0, int8x16_t __p1) {
54295   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54304 __ai int64x2_t vmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
54306   __ret = vmull_s32(vget_high_s32(__p0), vget_high_s32(__p1));
54310 __ai int64x2_t vmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
54312   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54321 __ai int32x4_t vmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
54323   __ret = vmull_s16(vget_high_s16(__p0), vget_high_s16(__p1));
54327 __ai int32x4_t vmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
54329   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
54338 __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
54340   __ret = vmull_p64((poly64_t)(vget_high_p64(__p0)), (poly64_t)(vget_high_p64(__p1)));
54344 __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
54346   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54354 #define vmull_high_lane_u32(__p0, __p1, __p2) __extension__ ({ \
54356   uint32x2_t __s1 = __p1; \
54362 #define vmull_high_lane_u32(__p0, __p1, __p2) __extension__ ({ \
54364   uint32x2_t __s1 = __p1; \
54375 #define vmull_high_lane_u16(__p0, __p1, __p2) __extension__ ({ \
54377   uint16x4_t __s1 = __p1; \
54383 #define vmull_high_lane_u16(__p0, __p1, __p2) __extension__ ({ \
54385   uint16x4_t __s1 = __p1; \
54396 #define vmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
54398   int32x2_t __s1 = __p1; \
54404 #define vmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
54406   int32x2_t __s1 = __p1; \
54417 #define vmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
54419   int16x4_t __s1 = __p1; \
54425 #define vmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
54427   int16x4_t __s1 = __p1; \
54438 #define vmull_high_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
54440   uint32x4_t __s1 = __p1; \
54446 #define vmull_high_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
54448   uint32x4_t __s1 = __p1; \
54459 #define vmull_high_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
54461   uint16x8_t __s1 = __p1; \
54467 #define vmull_high_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
54469   uint16x8_t __s1 = __p1; \
54480 #define vmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
54482   int32x4_t __s1 = __p1; \
54488 #define vmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
54490   int32x4_t __s1 = __p1; \
54501 #define vmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
54503   int16x8_t __s1 = __p1; \
54509 #define vmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
54511   int16x8_t __s1 = __p1; \
54522 __ai uint64x2_t vmull_high_n_u32(uint32x4_t __p0, uint32_t __p1) {
54524   __ret = vmull_n_u32(vget_high_u32(__p0), __p1);
54528 __ai uint64x2_t vmull_high_n_u32(uint32x4_t __p0, uint32_t __p1) {
54531   __ret = __noswap_vmull_n_u32(__noswap_vget_high_u32(__rev0), __p1);
54538 __ai uint32x4_t vmull_high_n_u16(uint16x8_t __p0, uint16_t __p1) {
54540   __ret = vmull_n_u16(vget_high_u16(__p0), __p1);
54544 __ai uint32x4_t vmull_high_n_u16(uint16x8_t __p0, uint16_t __p1) {
54547   __ret = __noswap_vmull_n_u16(__noswap_vget_high_u16(__rev0), __p1);
54554 __ai int64x2_t vmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
54556   __ret = vmull_n_s32(vget_high_s32(__p0), __p1);
54560 __ai int64x2_t vmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
54563   __ret = __noswap_vmull_n_s32(__noswap_vget_high_s32(__rev0), __p1);
54570 __ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
54572   __ret = vmull_n_s16(vget_high_s16(__p0), __p1);
54576 __ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
54579   __ret = __noswap_vmull_n_s16(__noswap_vget_high_s16(__rev0), __p1);
54586 #define vmull_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
54588   uint32x4_t __s1 = __p1; \
54594 #define vmull_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
54596   uint32x4_t __s1 = __p1; \
54607 #define vmull_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
54609   uint16x8_t __s1 = __p1; \
54615 #define vmull_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
54617   uint16x8_t __s1 = __p1; \
54628 #define vmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
54630   int32x4_t __s1 = __p1; \
54636 #define vmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
54638   int32x4_t __s1 = __p1; \
54649 #define vmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
54651   int16x8_t __s1 = __p1; \
54657 #define vmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
54659   int16x8_t __s1 = __p1; \
54670 __ai float64x2_t vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
54672   __ret = (float64x2_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
54676 __ai float64x2_t vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
54678   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54684 __ai float64x2_t __noswap_vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
54686   __ret = (float64x2_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
54692 __ai float32x4_t vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
54694   __ret = (float32x4_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
54698 __ai float32x4_t vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
54700   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54706 __ai float32x4_t __noswap_vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
54708   __ret = (float32x4_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
54714 __ai float64x1_t vmulx_f64(float64x1_t __p0, float64x1_t __p1) {
54716   __ret = (float64x1_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
54720 __ai float64x1_t vmulx_f64(float64x1_t __p0, float64x1_t __p1) {
54722   __ret = (float64x1_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
54728 __ai float32x2_t vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
54730   __ret = (float32x2_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
54734 __ai float32x2_t vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
54736   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54742 __ai float32x2_t __noswap_vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
54744   __ret = (float32x2_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
54750 __ai float64_t vmulxd_f64(float64_t __p0, float64_t __p1) {
54752   __ret = (float64_t) __builtin_neon_vmulxd_f64(__p0, __p1);
54756 __ai float64_t vmulxd_f64(float64_t __p0, float64_t __p1) {
54758   __ret = (float64_t) __builtin_neon_vmulxd_f64(__p0, __p1);
54761 __ai float64_t __noswap_vmulxd_f64(float64_t __p0, float64_t __p1) {
54763   __ret = (float64_t) __builtin_neon_vmulxd_f64(__p0, __p1);
54769 __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
54771   __ret = (float32_t) __builtin_neon_vmulxs_f32(__p0, __p1);
54775 __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
54777   __ret = (float32_t) __builtin_neon_vmulxs_f32(__p0, __p1);
54780 __ai float32_t __noswap_vmulxs_f32(float32_t __p0, float32_t __p1) {
54782   __ret = (float32_t) __builtin_neon_vmulxs_f32(__p0, __p1);
54825 #define vmulxq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54827   float64x1_t __s1 = __p1; \
54833 #define vmulxq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
54835   float64x1_t __s1 = __p1; \
54845 #define vmulxq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
54847   float32x2_t __s1 = __p1; \
54853 #define vmulxq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
54855   float32x2_t __s1 = __p1; \
54866 #define vmulx_lane_f32(__p0, __p1, __p2) __extension__ ({ \
54868   float32x2_t __s1 = __p1; \
54874 #define vmulx_lane_f32(__p0, __p1, __p2) __extension__ ({ \
54876   float32x2_t __s1 = __p1; \
54925 #define vmulxq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
54927   float64x2_t __s1 = __p1; \
54933 #define vmulxq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
54935   float64x2_t __s1 = __p1; \
54946 #define vmulxq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
54948   float32x4_t __s1 = __p1; \
54954 #define vmulxq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
54956   float32x4_t __s1 = __p1; \
54967 #define vmulx_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
54969   float32x4_t __s1 = __p1; \
54975 #define vmulx_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
54977   float32x4_t __s1 = __p1; \
55062 __ai uint8x16_t vpaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
55064   __ret = (uint8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
55068 __ai uint8x16_t vpaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
55070   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
55079 __ai uint32x4_t vpaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
55081   __ret = (uint32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
55085 __ai uint32x4_t vpaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
55087   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55096 __ai uint64x2_t vpaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
55098   __ret = (uint64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
55102 __ai uint64x2_t vpaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
55104   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55113 __ai uint16x8_t vpaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
55115   __ret = (uint16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
55119 __ai uint16x8_t vpaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
55121   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
55130 __ai int8x16_t vpaddq_s8(int8x16_t __p0, int8x16_t __p1) {
55132   __ret = (int8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
55136 __ai int8x16_t vpaddq_s8(int8x16_t __p0, int8x16_t __p1) {
55138   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
55147 __ai float64x2_t vpaddq_f64(float64x2_t __p0, float64x2_t __p1) {
55149   __ret = (float64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
55153 __ai float64x2_t vpaddq_f64(float64x2_t __p0, float64x2_t __p1) {
55155   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55164 __ai float32x4_t vpaddq_f32(float32x4_t __p0, float32x4_t __p1) {
55166   __ret = (float32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
55170 __ai float32x4_t vpaddq_f32(float32x4_t __p0, float32x4_t __p1) {
55172   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55181 __ai int32x4_t vpaddq_s32(int32x4_t __p0, int32x4_t __p1) {
55183   __ret = (int32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
55187 __ai int32x4_t vpaddq_s32(int32x4_t __p0, int32x4_t __p1) {
55189   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55198 __ai int64x2_t vpaddq_s64(int64x2_t __p0, int64x2_t __p1) {
55200   __ret = (int64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
55204 __ai int64x2_t vpaddq_s64(int64x2_t __p0, int64x2_t __p1) {
55206   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55215 __ai int16x8_t vpaddq_s16(int16x8_t __p0, int16x8_t __p1) {
55217   __ret = (int16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
55221 __ai int16x8_t vpaddq_s16(int16x8_t __p0, int16x8_t __p1) {
55223   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
55292 __ai uint8x16_t vpmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
55294   __ret = (uint8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
55298 __ai uint8x16_t vpmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
55300   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
55309 __ai uint32x4_t vpmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
55311   __ret = (uint32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
55315 __ai uint32x4_t vpmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
55317   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55326 __ai uint16x8_t vpmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
55328   __ret = (uint16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
55332 __ai uint16x8_t vpmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
55334   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
55343 __ai int8x16_t vpmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
55345   __ret = (int8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
55349 __ai int8x16_t vpmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
55351   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
55360 __ai float64x2_t vpmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
55362   __ret = (float64x2_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
55366 __ai float64x2_t vpmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
55368   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55377 __ai float32x4_t vpmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
55379   __ret = (float32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
55383 __ai float32x4_t vpmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
55385   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55394 __ai int32x4_t vpmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
55396   __ret = (int32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
55400 __ai int32x4_t vpmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
55402   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55411 __ai int16x8_t vpmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
55413   __ret = (int16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
55417 __ai int16x8_t vpmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
55419   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
55458 __ai float64x2_t vpmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
55460   __ret = (float64x2_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
55464 __ai float64x2_t vpmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
55466   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55475 __ai float32x4_t vpmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
55477   __ret = (float32x4_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
55481 __ai float32x4_t vpmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
55483   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55492 __ai float32x2_t vpmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
55494   __ret = (float32x2_t) __builtin_neon_vpmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
55498 __ai float32x2_t vpmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
55500   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55539 __ai uint8x16_t vpminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
55541   __ret = (uint8x16_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
55545 __ai uint8x16_t vpminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
55547   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
55556 __ai uint32x4_t vpminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
55558   __ret = (uint32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
55562 __ai uint32x4_t vpminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
55564   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55573 __ai uint16x8_t vpminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
55575   __ret = (uint16x8_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
55579 __ai uint16x8_t vpminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
55581   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
55590 __ai int8x16_t vpminq_s8(int8x16_t __p0, int8x16_t __p1) {
55592   __ret = (int8x16_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
55596 __ai int8x16_t vpminq_s8(int8x16_t __p0, int8x16_t __p1) {
55598   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
55607 __ai float64x2_t vpminq_f64(float64x2_t __p0, float64x2_t __p1) {
55609   __ret = (float64x2_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
55613 __ai float64x2_t vpminq_f64(float64x2_t __p0, float64x2_t __p1) {
55615   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55624 __ai float32x4_t vpminq_f32(float32x4_t __p0, float32x4_t __p1) {
55626   __ret = (float32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
55630 __ai float32x4_t vpminq_f32(float32x4_t __p0, float32x4_t __p1) {
55632   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55641 __ai int32x4_t vpminq_s32(int32x4_t __p0, int32x4_t __p1) {
55643   __ret = (int32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
55647 __ai int32x4_t vpminq_s32(int32x4_t __p0, int32x4_t __p1) {
55649   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55658 __ai int16x8_t vpminq_s16(int16x8_t __p0, int16x8_t __p1) {
55660   __ret = (int16x8_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
55664 __ai int16x8_t vpminq_s16(int16x8_t __p0, int16x8_t __p1) {
55666   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
55705 __ai float64x2_t vpminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
55707   __ret = (float64x2_t) __builtin_neon_vpminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
55711 __ai float64x2_t vpminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
55713   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55722 __ai float32x4_t vpminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
55724   __ret = (float32x4_t) __builtin_neon_vpminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
55728 __ai float32x4_t vpminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
55730   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55739 __ai float32x2_t vpminnm_f32(float32x2_t __p0, float32x2_t __p1) {
55741   __ret = (float32x2_t) __builtin_neon_vpminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
55745 __ai float32x2_t vpminnm_f32(float32x2_t __p0, float32x2_t __p1) {
55747   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
55872 __ai uint8_t vqaddb_u8(uint8_t __p0, uint8_t __p1) {
55874   __ret = (uint8_t) __builtin_neon_vqaddb_u8(__p0, __p1);
55878 __ai uint8_t vqaddb_u8(uint8_t __p0, uint8_t __p1) {
55880   __ret = (uint8_t) __builtin_neon_vqaddb_u8(__p0, __p1);
55886 __ai uint32_t vqadds_u32(uint32_t __p0, uint32_t __p1) {
55888   __ret = (uint32_t) __builtin_neon_vqadds_u32(__p0, __p1);
55892 __ai uint32_t vqadds_u32(uint32_t __p0, uint32_t __p1) {
55894   __ret = (uint32_t) __builtin_neon_vqadds_u32(__p0, __p1);
55900 __ai uint64_t vqaddd_u64(uint64_t __p0, uint64_t __p1) {
55902   __ret = (uint64_t) __builtin_neon_vqaddd_u64(__p0, __p1);
55906 __ai uint64_t vqaddd_u64(uint64_t __p0, uint64_t __p1) {
55908   __ret = (uint64_t) __builtin_neon_vqaddd_u64(__p0, __p1);
55914 __ai uint16_t vqaddh_u16(uint16_t __p0, uint16_t __p1) {
55916   __ret = (uint16_t) __builtin_neon_vqaddh_u16(__p0, __p1);
55920 __ai uint16_t vqaddh_u16(uint16_t __p0, uint16_t __p1) {
55922   __ret = (uint16_t) __builtin_neon_vqaddh_u16(__p0, __p1);
55928 __ai int8_t vqaddb_s8(int8_t __p0, int8_t __p1) {
55930   __ret = (int8_t) __builtin_neon_vqaddb_s8(__p0, __p1);
55934 __ai int8_t vqaddb_s8(int8_t __p0, int8_t __p1) {
55936   __ret = (int8_t) __builtin_neon_vqaddb_s8(__p0, __p1);
55942 __ai int32_t vqadds_s32(int32_t __p0, int32_t __p1) {
55944   __ret = (int32_t) __builtin_neon_vqadds_s32(__p0, __p1);
55948 __ai int32_t vqadds_s32(int32_t __p0, int32_t __p1) {
55950   __ret = (int32_t) __builtin_neon_vqadds_s32(__p0, __p1);
55953 __ai int32_t __noswap_vqadds_s32(int32_t __p0, int32_t __p1) {
55955   __ret = (int32_t) __builtin_neon_vqadds_s32(__p0, __p1);
55961 __ai int64_t vqaddd_s64(int64_t __p0, int64_t __p1) {
55963   __ret = (int64_t) __builtin_neon_vqaddd_s64(__p0, __p1);
55967 __ai int64_t vqaddd_s64(int64_t __p0, int64_t __p1) {
55969   __ret = (int64_t) __builtin_neon_vqaddd_s64(__p0, __p1);
55975 __ai int16_t vqaddh_s16(int16_t __p0, int16_t __p1) {
55977   __ret = (int16_t) __builtin_neon_vqaddh_s16(__p0, __p1);
55981 __ai int16_t vqaddh_s16(int16_t __p0, int16_t __p1) {
55983   __ret = (int16_t) __builtin_neon_vqaddh_s16(__p0, __p1);
55986 __ai int16_t __noswap_vqaddh_s16(int16_t __p0, int16_t __p1) {
55988   __ret = (int16_t) __builtin_neon_vqaddh_s16(__p0, __p1);
55994 __ai int64_t vqdmlals_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
55996   __ret = (int64_t) __builtin_neon_vqdmlals_s32(__p0, __p1, __p2);
56000 __ai int64_t vqdmlals_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
56002   __ret = (int64_t) __builtin_neon_vqdmlals_s32(__p0, __p1, __p2);
56008 __ai int32_t vqdmlalh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
56010   __ret = (int32_t) __builtin_neon_vqdmlalh_s16(__p0, __p1, __p2);
56014 __ai int32_t vqdmlalh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
56016   __ret = (int32_t) __builtin_neon_vqdmlalh_s16(__p0, __p1, __p2);
56022 __ai int64x2_t vqdmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
56024   __ret = vqdmlal_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
56028 __ai int64x2_t vqdmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
56030   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56040 __ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
56042   __ret = vqdmlal_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
56046 __ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
56048   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56058 #define vqdmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56060   int32x4_t __s1 = __p1; \
56067 #define vqdmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56069   int32x4_t __s1 = __p1; \
56082 #define vqdmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56084   int16x8_t __s1 = __p1; \
56091 #define vqdmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56093   int16x8_t __s1 = __p1; \
56106 #define vqdmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56108   int32x4_t __s1 = __p1; \
56115 #define vqdmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56117   int32x4_t __s1 = __p1; \
56130 #define vqdmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56132   int16x8_t __s1 = __p1; \
56139 #define vqdmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56141   int16x8_t __s1 = __p1; \
56154 __ai int64x2_t vqdmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
56156   __ret = vqdmlal_n_s32(__p0, vget_high_s32(__p1), __p2);
56160 __ai int64x2_t vqdmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
56162   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56171 __ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
56173   __ret = vqdmlal_n_s16(__p0, vget_high_s16(__p1), __p2);
56177 __ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
56179   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56188 #define vqdmlals_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56190   int32_t __s1 = __p1; \
56197 #define vqdmlals_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56199   int32_t __s1 = __p1; \
56209 #define vqdmlalh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56211   int16_t __s1 = __p1; \
56218 #define vqdmlalh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56220   int16_t __s1 = __p1; \
56230 #define vqdmlals_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56232   int32_t __s1 = __p1; \
56239 #define vqdmlals_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56241   int32_t __s1 = __p1; \
56251 #define vqdmlalh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56253   int16_t __s1 = __p1; \
56260 #define vqdmlalh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56262   int16_t __s1 = __p1; \
56272 #define vqdmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56274   int32x2_t __s1 = __p1; \
56281 #define vqdmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56283   int32x2_t __s1 = __p1; \
56296 #define vqdmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56298   int16x4_t __s1 = __p1; \
56305 #define vqdmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56307   int16x4_t __s1 = __p1; \
56320 __ai int64_t vqdmlsls_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
56322   __ret = (int64_t) __builtin_neon_vqdmlsls_s32(__p0, __p1, __p2);
56326 __ai int64_t vqdmlsls_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
56328   __ret = (int64_t) __builtin_neon_vqdmlsls_s32(__p0, __p1, __p2);
56334 __ai int32_t vqdmlslh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
56336   __ret = (int32_t) __builtin_neon_vqdmlslh_s16(__p0, __p1, __p2);
56340 __ai int32_t vqdmlslh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
56342   __ret = (int32_t) __builtin_neon_vqdmlslh_s16(__p0, __p1, __p2);
56348 __ai int64x2_t vqdmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
56350   __ret = vqdmlsl_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
56354 __ai int64x2_t vqdmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
56356   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56366 __ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
56368   __ret = vqdmlsl_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
56372 __ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
56374   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56384 #define vqdmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56386   int32x4_t __s1 = __p1; \
56393 #define vqdmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56395   int32x4_t __s1 = __p1; \
56408 #define vqdmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56410   int16x8_t __s1 = __p1; \
56417 #define vqdmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56419   int16x8_t __s1 = __p1; \
56432 #define vqdmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56434   int32x4_t __s1 = __p1; \
56441 #define vqdmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56443   int32x4_t __s1 = __p1; \
56456 #define vqdmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56458   int16x8_t __s1 = __p1; \
56465 #define vqdmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56467   int16x8_t __s1 = __p1; \
56480 __ai int64x2_t vqdmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
56482   __ret = vqdmlsl_n_s32(__p0, vget_high_s32(__p1), __p2);
56486 __ai int64x2_t vqdmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
56488   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56497 __ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
56499   __ret = vqdmlsl_n_s16(__p0, vget_high_s16(__p1), __p2);
56503 __ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
56505   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56514 #define vqdmlsls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56516   int32_t __s1 = __p1; \
56523 #define vqdmlsls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56525   int32_t __s1 = __p1; \
56535 #define vqdmlslh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56537   int16_t __s1 = __p1; \
56544 #define vqdmlslh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56546   int16_t __s1 = __p1; \
56556 #define vqdmlsls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56558   int32_t __s1 = __p1; \
56565 #define vqdmlsls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56567   int32_t __s1 = __p1; \
56577 #define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56579   int16_t __s1 = __p1; \
56586 #define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56588   int16_t __s1 = __p1; \
56598 #define vqdmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56600   int32x2_t __s1 = __p1; \
56607 #define vqdmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
56609   int32x2_t __s1 = __p1; \
56622 #define vqdmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56624   int16x4_t __s1 = __p1; \
56631 #define vqdmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
56633   int16x4_t __s1 = __p1; \
56646 __ai int32_t vqdmulhs_s32(int32_t __p0, int32_t __p1) {
56648   __ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
56652 __ai int32_t vqdmulhs_s32(int32_t __p0, int32_t __p1) {
56654   __ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
56657 __ai int32_t __noswap_vqdmulhs_s32(int32_t __p0, int32_t __p1) {
56659   __ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
56665 __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
56667   __ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
56671 __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
56673   __ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
56676 __ai int16_t __noswap_vqdmulhh_s16(int16_t __p0, int16_t __p1) {
56678   __ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
56760 #define vqdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56762   int32x4_t __s1 = __p1; \
56768 #define vqdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56770   int32x4_t __s1 = __p1; \
56781 #define vqdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56783   int16x8_t __s1 = __p1; \
56789 #define vqdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56791   int16x8_t __s1 = __p1; \
56802 #define vqdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56804   int32x4_t __s1 = __p1; \
56810 #define vqdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56812   int32x4_t __s1 = __p1; \
56823 #define vqdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56825   int16x8_t __s1 = __p1; \
56831 #define vqdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56833   int16x8_t __s1 = __p1; \
56844 __ai int64_t vqdmulls_s32(int32_t __p0, int32_t __p1) {
56846   __ret = (int64_t) __builtin_neon_vqdmulls_s32(__p0, __p1);
56850 __ai int64_t vqdmulls_s32(int32_t __p0, int32_t __p1) {
56852   __ret = (int64_t) __builtin_neon_vqdmulls_s32(__p0, __p1);
56855 __ai int64_t __noswap_vqdmulls_s32(int32_t __p0, int32_t __p1) {
56857   __ret = (int64_t) __builtin_neon_vqdmulls_s32(__p0, __p1);
56863 __ai int32_t vqdmullh_s16(int16_t __p0, int16_t __p1) {
56865   __ret = (int32_t) __builtin_neon_vqdmullh_s16(__p0, __p1);
56869 __ai int32_t vqdmullh_s16(int16_t __p0, int16_t __p1) {
56871   __ret = (int32_t) __builtin_neon_vqdmullh_s16(__p0, __p1);
56874 __ai int32_t __noswap_vqdmullh_s16(int16_t __p0, int16_t __p1) {
56876   __ret = (int32_t) __builtin_neon_vqdmullh_s16(__p0, __p1);
56882 __ai int64x2_t vqdmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
56884   __ret = vqdmull_s32(vget_high_s32(__p0), vget_high_s32(__p1));
56888 __ai int64x2_t vqdmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
56890   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56899 __ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
56901   __ret = vqdmull_s16(vget_high_s16(__p0), vget_high_s16(__p1));
56905 __ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
56907   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56916 #define vqdmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
56918   int32x2_t __s1 = __p1; \
56924 #define vqdmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
56926   int32x2_t __s1 = __p1; \
56937 #define vqdmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
56939   int16x4_t __s1 = __p1; \
56945 #define vqdmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
56947   int16x4_t __s1 = __p1; \
56958 #define vqdmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56960   int32x4_t __s1 = __p1; \
56966 #define vqdmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56968   int32x4_t __s1 = __p1; \
56979 #define vqdmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56981   int16x8_t __s1 = __p1; \
56987 #define vqdmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56989   int16x8_t __s1 = __p1; \
57000 __ai int64x2_t vqdmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
57002   __ret = vqdmull_n_s32(vget_high_s32(__p0), __p1);
57006 __ai int64x2_t vqdmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
57009   __ret = __noswap_vqdmull_n_s32(__noswap_vget_high_s32(__rev0), __p1);
57016 __ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
57018   __ret = vqdmull_n_s16(vget_high_s16(__p0), __p1);
57022 __ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
57025   __ret = __noswap_vqdmull_n_s16(__noswap_vget_high_s16(__rev0), __p1);
57108 #define vqdmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
57110   int32x4_t __s1 = __p1; \
57116 #define vqdmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
57118   int32x4_t __s1 = __p1; \
57129 #define vqdmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
57131   int16x8_t __s1 = __p1; \
57137 #define vqdmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
57139   int16x8_t __s1 = __p1; \
57234 __ai uint16x8_t vqmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
57236   __ret = vcombine_u16(__p0, vqmovn_u32(__p1));
57240 __ai uint16x8_t vqmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
57242   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57251 __ai uint32x4_t vqmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
57253   __ret = vcombine_u32(__p0, vqmovn_u64(__p1));
57257 __ai uint32x4_t vqmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
57259   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57268 __ai uint8x16_t vqmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
57270   __ret = vcombine_u8(__p0, vqmovn_u16(__p1));
57274 __ai uint8x16_t vqmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
57276   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57285 __ai int16x8_t vqmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
57287   __ret = vcombine_s16(__p0, vqmovn_s32(__p1));
57291 __ai int16x8_t vqmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
57293   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57302 __ai int32x4_t vqmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
57304   __ret = vcombine_s32(__p0, vqmovn_s64(__p1));
57308 __ai int32x4_t vqmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
57310   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57319 __ai int8x16_t vqmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
57321   __ret = vcombine_s8(__p0, vqmovn_s16(__p1));
57325 __ai int8x16_t vqmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
57327   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57378 __ai int16x8_t vqmovun_high_s32(int16x4_t __p0, int32x4_t __p1) {
57380   __ret = vcombine_u16((uint16x4_t)(__p0), vqmovun_s32(__p1));
57384 __ai int16x8_t vqmovun_high_s32(int16x4_t __p0, int32x4_t __p1) {
57386   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
57395 __ai int32x4_t vqmovun_high_s64(int32x2_t __p0, int64x2_t __p1) {
57397   __ret = vcombine_u32((uint32x2_t)(__p0), vqmovun_s64(__p1));
57401 __ai int32x4_t vqmovun_high_s64(int32x2_t __p0, int64x2_t __p1) {
57403   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
57412 __ai int8x16_t vqmovun_high_s16(int8x8_t __p0, int16x8_t __p1) {
57414   __ret = vcombine_u8((uint8x8_t)(__p0), vqmovun_s16(__p1));
57418 __ai int8x16_t vqmovun_high_s16(int8x8_t __p0, int16x8_t __p1) {
57420   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57515 __ai int32_t vqrdmulhs_s32(int32_t __p0, int32_t __p1) {
57517   __ret = (int32_t) __builtin_neon_vqrdmulhs_s32(__p0, __p1);
57521 __ai int32_t vqrdmulhs_s32(int32_t __p0, int32_t __p1) {
57523   __ret = (int32_t) __builtin_neon_vqrdmulhs_s32(__p0, __p1);
57526 __ai int32_t __noswap_vqrdmulhs_s32(int32_t __p0, int32_t __p1) {
57528   __ret = (int32_t) __builtin_neon_vqrdmulhs_s32(__p0, __p1);
57534 __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
57536   __ret = (int16_t) __builtin_neon_vqrdmulhh_s16(__p0, __p1);
57540 __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
57542   __ret = (int16_t) __builtin_neon_vqrdmulhh_s16(__p0, __p1);
57545 __ai int16_t __noswap_vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
57547   __ret = (int16_t) __builtin_neon_vqrdmulhh_s16(__p0, __p1);
57629 #define vqrdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
57631   int32x4_t __s1 = __p1; \
57637 #define vqrdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
57639   int32x4_t __s1 = __p1; \
57650 #define vqrdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
57652   int16x8_t __s1 = __p1; \
57658 #define vqrdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
57660   int16x8_t __s1 = __p1; \
57671 #define vqrdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
57673   int32x4_t __s1 = __p1; \
57679 #define vqrdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
57681   int32x4_t __s1 = __p1; \
57692 #define vqrdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
57694   int16x8_t __s1 = __p1; \
57700 #define vqrdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
57702   int16x8_t __s1 = __p1; \
57713 __ai uint8_t vqrshlb_u8(uint8_t __p0, uint8_t __p1) {
57715   __ret = (uint8_t) __builtin_neon_vqrshlb_u8(__p0, __p1);
57719 __ai uint8_t vqrshlb_u8(uint8_t __p0, uint8_t __p1) {
57721   __ret = (uint8_t) __builtin_neon_vqrshlb_u8(__p0, __p1);
57727 __ai uint32_t vqrshls_u32(uint32_t __p0, uint32_t __p1) {
57729   __ret = (uint32_t) __builtin_neon_vqrshls_u32(__p0, __p1);
57733 __ai uint32_t vqrshls_u32(uint32_t __p0, uint32_t __p1) {
57735   __ret = (uint32_t) __builtin_neon_vqrshls_u32(__p0, __p1);
57741 __ai uint64_t vqrshld_u64(uint64_t __p0, uint64_t __p1) {
57743   __ret = (uint64_t) __builtin_neon_vqrshld_u64(__p0, __p1);
57747 __ai uint64_t vqrshld_u64(uint64_t __p0, uint64_t __p1) {
57749   __ret = (uint64_t) __builtin_neon_vqrshld_u64(__p0, __p1);
57755 __ai uint16_t vqrshlh_u16(uint16_t __p0, uint16_t __p1) {
57757   __ret = (uint16_t) __builtin_neon_vqrshlh_u16(__p0, __p1);
57761 __ai uint16_t vqrshlh_u16(uint16_t __p0, uint16_t __p1) {
57763   __ret = (uint16_t) __builtin_neon_vqrshlh_u16(__p0, __p1);
57769 __ai int8_t vqrshlb_s8(int8_t __p0, int8_t __p1) {
57771   __ret = (int8_t) __builtin_neon_vqrshlb_s8(__p0, __p1);
57775 __ai int8_t vqrshlb_s8(int8_t __p0, int8_t __p1) {
57777   __ret = (int8_t) __builtin_neon_vqrshlb_s8(__p0, __p1);
57783 __ai int32_t vqrshls_s32(int32_t __p0, int32_t __p1) {
57785   __ret = (int32_t) __builtin_neon_vqrshls_s32(__p0, __p1);
57789 __ai int32_t vqrshls_s32(int32_t __p0, int32_t __p1) {
57791   __ret = (int32_t) __builtin_neon_vqrshls_s32(__p0, __p1);
57797 __ai int64_t vqrshld_s64(int64_t __p0, int64_t __p1) {
57799   __ret = (int64_t) __builtin_neon_vqrshld_s64(__p0, __p1);
57803 __ai int64_t vqrshld_s64(int64_t __p0, int64_t __p1) {
57805   __ret = (int64_t) __builtin_neon_vqrshld_s64(__p0, __p1);
57811 __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
57813   __ret = (int16_t) __builtin_neon_vqrshlh_s16(__p0, __p1);
57817 __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
57819   __ret = (int16_t) __builtin_neon_vqrshlh_s16(__p0, __p1);
57951 #define vqrshrns_n_u32(__p0, __p1) __extension__ ({ \
57954   __ret = (uint16_t) __builtin_neon_vqrshrns_n_u32(__s0, __p1); \
57958 #define vqrshrns_n_u32(__p0, __p1) __extension__ ({ \
57961   __ret = (uint16_t) __builtin_neon_vqrshrns_n_u32(__s0, __p1); \
57967 #define vqrshrnd_n_u64(__p0, __p1) __extension__ ({ \
57970   __ret = (uint32_t) __builtin_neon_vqrshrnd_n_u64(__s0, __p1); \
57974 #define vqrshrnd_n_u64(__p0, __p1) __extension__ ({ \
57977   __ret = (uint32_t) __builtin_neon_vqrshrnd_n_u64(__s0, __p1); \
57983 #define vqrshrnh_n_u16(__p0, __p1) __extension__ ({ \
57986   __ret = (uint8_t) __builtin_neon_vqrshrnh_n_u16(__s0, __p1); \
57990 #define vqrshrnh_n_u16(__p0, __p1) __extension__ ({ \
57993   __ret = (uint8_t) __builtin_neon_vqrshrnh_n_u16(__s0, __p1); \
57999 #define vqrshrns_n_s32(__p0, __p1) __extension__ ({ \
58002   __ret = (int16_t) __builtin_neon_vqrshrns_n_s32(__s0, __p1); \
58006 #define vqrshrns_n_s32(__p0, __p1) __extension__ ({ \
58009   __ret = (int16_t) __builtin_neon_vqrshrns_n_s32(__s0, __p1); \
58015 #define vqrshrnd_n_s64(__p0, __p1) __extension__ ({ \
58018   __ret = (int32_t) __builtin_neon_vqrshrnd_n_s64(__s0, __p1); \
58022 #define vqrshrnd_n_s64(__p0, __p1) __extension__ ({ \
58025   __ret = (int32_t) __builtin_neon_vqrshrnd_n_s64(__s0, __p1); \
58031 #define vqrshrnh_n_s16(__p0, __p1) __extension__ ({ \
58034   __ret = (int8_t) __builtin_neon_vqrshrnh_n_s16(__s0, __p1); \
58038 #define vqrshrnh_n_s16(__p0, __p1) __extension__ ({ \
58041   __ret = (int8_t) __builtin_neon_vqrshrnh_n_s16(__s0, __p1); \
58110 #define vqrshruns_n_s32(__p0, __p1) __extension__ ({ \
58113   __ret = (int16_t) __builtin_neon_vqrshruns_n_s32(__s0, __p1); \
58117 #define vqrshruns_n_s32(__p0, __p1) __extension__ ({ \
58120   __ret = (int16_t) __builtin_neon_vqrshruns_n_s32(__s0, __p1); \
58126 #define vqrshrund_n_s64(__p0, __p1) __extension__ ({ \
58129   __ret = (int32_t) __builtin_neon_vqrshrund_n_s64(__s0, __p1); \
58133 #define vqrshrund_n_s64(__p0, __p1) __extension__ ({ \
58136   __ret = (int32_t) __builtin_neon_vqrshrund_n_s64(__s0, __p1); \
58142 #define vqrshrunh_n_s16(__p0, __p1) __extension__ ({ \
58145   __ret = (int8_t) __builtin_neon_vqrshrunh_n_s16(__s0, __p1); \
58149 #define vqrshrunh_n_s16(__p0, __p1) __extension__ ({ \
58152   __ret = (int8_t) __builtin_neon_vqrshrunh_n_s16(__s0, __p1); \
58158 __ai uint8_t vqshlb_u8(uint8_t __p0, uint8_t __p1) {
58160   __ret = (uint8_t) __builtin_neon_vqshlb_u8(__p0, __p1);
58164 __ai uint8_t vqshlb_u8(uint8_t __p0, uint8_t __p1) {
58166   __ret = (uint8_t) __builtin_neon_vqshlb_u8(__p0, __p1);
58172 __ai uint32_t vqshls_u32(uint32_t __p0, uint32_t __p1) {
58174   __ret = (uint32_t) __builtin_neon_vqshls_u32(__p0, __p1);
58178 __ai uint32_t vqshls_u32(uint32_t __p0, uint32_t __p1) {
58180   __ret = (uint32_t) __builtin_neon_vqshls_u32(__p0, __p1);
58186 __ai uint64_t vqshld_u64(uint64_t __p0, uint64_t __p1) {
58188   __ret = (uint64_t) __builtin_neon_vqshld_u64(__p0, __p1);
58192 __ai uint64_t vqshld_u64(uint64_t __p0, uint64_t __p1) {
58194   __ret = (uint64_t) __builtin_neon_vqshld_u64(__p0, __p1);
58200 __ai uint16_t vqshlh_u16(uint16_t __p0, uint16_t __p1) {
58202   __ret = (uint16_t) __builtin_neon_vqshlh_u16(__p0, __p1);
58206 __ai uint16_t vqshlh_u16(uint16_t __p0, uint16_t __p1) {
58208   __ret = (uint16_t) __builtin_neon_vqshlh_u16(__p0, __p1);
58214 __ai int8_t vqshlb_s8(int8_t __p0, int8_t __p1) {
58216   __ret = (int8_t) __builtin_neon_vqshlb_s8(__p0, __p1);
58220 __ai int8_t vqshlb_s8(int8_t __p0, int8_t __p1) {
58222   __ret = (int8_t) __builtin_neon_vqshlb_s8(__p0, __p1);
58228 __ai int32_t vqshls_s32(int32_t __p0, int32_t __p1) {
58230   __ret = (int32_t) __builtin_neon_vqshls_s32(__p0, __p1);
58234 __ai int32_t vqshls_s32(int32_t __p0, int32_t __p1) {
58236   __ret = (int32_t) __builtin_neon_vqshls_s32(__p0, __p1);
58242 __ai int64_t vqshld_s64(int64_t __p0, int64_t __p1) {
58244   __ret = (int64_t) __builtin_neon_vqshld_s64(__p0, __p1);
58248 __ai int64_t vqshld_s64(int64_t __p0, int64_t __p1) {
58250   __ret = (int64_t) __builtin_neon_vqshld_s64(__p0, __p1);
58256 __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
58258   __ret = (int16_t) __builtin_neon_vqshlh_s16(__p0, __p1);
58262 __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
58264   __ret = (int16_t) __builtin_neon_vqshlh_s16(__p0, __p1);
58270 #define vqshlb_n_u8(__p0, __p1) __extension__ ({ \
58273   __ret = (uint8_t) __builtin_neon_vqshlb_n_u8(__s0, __p1); \
58277 #define vqshlb_n_u8(__p0, __p1) __extension__ ({ \
58280   __ret = (uint8_t) __builtin_neon_vqshlb_n_u8(__s0, __p1); \
58286 #define vqshls_n_u32(__p0, __p1) __extension__ ({ \
58289   __ret = (uint32_t) __builtin_neon_vqshls_n_u32(__s0, __p1); \
58293 #define vqshls_n_u32(__p0, __p1) __extension__ ({ \
58296   __ret = (uint32_t) __builtin_neon_vqshls_n_u32(__s0, __p1); \
58302 #define vqshld_n_u64(__p0, __p1) __extension__ ({ \
58305   __ret = (uint64_t) __builtin_neon_vqshld_n_u64(__s0, __p1); \
58309 #define vqshld_n_u64(__p0, __p1) __extension__ ({ \
58312   __ret = (uint64_t) __builtin_neon_vqshld_n_u64(__s0, __p1); \
58318 #define vqshlh_n_u16(__p0, __p1) __extension__ ({ \
58321   __ret = (uint16_t) __builtin_neon_vqshlh_n_u16(__s0, __p1); \
58325 #define vqshlh_n_u16(__p0, __p1) __extension__ ({ \
58328   __ret = (uint16_t) __builtin_neon_vqshlh_n_u16(__s0, __p1); \
58334 #define vqshlb_n_s8(__p0, __p1) __extension__ ({ \
58337   __ret = (int8_t) __builtin_neon_vqshlb_n_s8(__s0, __p1); \
58341 #define vqshlb_n_s8(__p0, __p1) __extension__ ({ \
58344   __ret = (int8_t) __builtin_neon_vqshlb_n_s8(__s0, __p1); \
58350 #define vqshls_n_s32(__p0, __p1) __extension__ ({ \
58353   __ret = (int32_t) __builtin_neon_vqshls_n_s32(__s0, __p1); \
58357 #define vqshls_n_s32(__p0, __p1) __extension__ ({ \
58360   __ret = (int32_t) __builtin_neon_vqshls_n_s32(__s0, __p1); \
58366 #define vqshld_n_s64(__p0, __p1) __extension__ ({ \
58369   __ret = (int64_t) __builtin_neon_vqshld_n_s64(__s0, __p1); \
58373 #define vqshld_n_s64(__p0, __p1) __extension__ ({ \
58376   __ret = (int64_t) __builtin_neon_vqshld_n_s64(__s0, __p1); \
58382 #define vqshlh_n_s16(__p0, __p1) __extension__ ({ \
58385   __ret = (int16_t) __builtin_neon_vqshlh_n_s16(__s0, __p1); \
58389 #define vqshlh_n_s16(__p0, __p1) __extension__ ({ \
58392   __ret = (int16_t) __builtin_neon_vqshlh_n_s16(__s0, __p1); \
58398 #define vqshlub_n_s8(__p0, __p1) __extension__ ({ \
58401   __ret = (int8_t) __builtin_neon_vqshlub_n_s8(__s0, __p1); \
58405 #define vqshlub_n_s8(__p0, __p1) __extension__ ({ \
58408   __ret = (int8_t) __builtin_neon_vqshlub_n_s8(__s0, __p1); \
58414 #define vqshlus_n_s32(__p0, __p1) __extension__ ({ \
58417   __ret = (int32_t) __builtin_neon_vqshlus_n_s32(__s0, __p1); \
58421 #define vqshlus_n_s32(__p0, __p1) __extension__ ({ \
58424   __ret = (int32_t) __builtin_neon_vqshlus_n_s32(__s0, __p1); \
58430 #define vqshlud_n_s64(__p0, __p1) __extension__ ({ \
58433   __ret = (int64_t) __builtin_neon_vqshlud_n_s64(__s0, __p1); \
58437 #define vqshlud_n_s64(__p0, __p1) __extension__ ({ \
58440   __ret = (int64_t) __builtin_neon_vqshlud_n_s64(__s0, __p1); \
58446 #define vqshluh_n_s16(__p0, __p1) __extension__ ({ \
58449   __ret = (int16_t) __builtin_neon_vqshluh_n_s16(__s0, __p1); \
58453 #define vqshluh_n_s16(__p0, __p1) __extension__ ({ \
58456   __ret = (int16_t) __builtin_neon_vqshluh_n_s16(__s0, __p1); \
58588 #define vqshrns_n_u32(__p0, __p1) __extension__ ({ \
58591   __ret = (uint16_t) __builtin_neon_vqshrns_n_u32(__s0, __p1); \
58595 #define vqshrns_n_u32(__p0, __p1) __extension__ ({ \
58598   __ret = (uint16_t) __builtin_neon_vqshrns_n_u32(__s0, __p1); \
58604 #define vqshrnd_n_u64(__p0, __p1) __extension__ ({ \
58607   __ret = (uint32_t) __builtin_neon_vqshrnd_n_u64(__s0, __p1); \
58611 #define vqshrnd_n_u64(__p0, __p1) __extension__ ({ \
58614   __ret = (uint32_t) __builtin_neon_vqshrnd_n_u64(__s0, __p1); \
58620 #define vqshrnh_n_u16(__p0, __p1) __extension__ ({ \
58623   __ret = (uint8_t) __builtin_neon_vqshrnh_n_u16(__s0, __p1); \
58627 #define vqshrnh_n_u16(__p0, __p1) __extension__ ({ \
58630   __ret = (uint8_t) __builtin_neon_vqshrnh_n_u16(__s0, __p1); \
58636 #define vqshrns_n_s32(__p0, __p1) __extension__ ({ \
58639   __ret = (int16_t) __builtin_neon_vqshrns_n_s32(__s0, __p1); \
58643 #define vqshrns_n_s32(__p0, __p1) __extension__ ({ \
58646   __ret = (int16_t) __builtin_neon_vqshrns_n_s32(__s0, __p1); \
58652 #define vqshrnd_n_s64(__p0, __p1) __extension__ ({ \
58655   __ret = (int32_t) __builtin_neon_vqshrnd_n_s64(__s0, __p1); \
58659 #define vqshrnd_n_s64(__p0, __p1) __extension__ ({ \
58662   __ret = (int32_t) __builtin_neon_vqshrnd_n_s64(__s0, __p1); \
58668 #define vqshrnh_n_s16(__p0, __p1) __extension__ ({ \
58671   __ret = (int8_t) __builtin_neon_vqshrnh_n_s16(__s0, __p1); \
58675 #define vqshrnh_n_s16(__p0, __p1) __extension__ ({ \
58678   __ret = (int8_t) __builtin_neon_vqshrnh_n_s16(__s0, __p1); \
58747 #define vqshruns_n_s32(__p0, __p1) __extension__ ({ \
58750   __ret = (int16_t) __builtin_neon_vqshruns_n_s32(__s0, __p1); \
58754 #define vqshruns_n_s32(__p0, __p1) __extension__ ({ \
58757   __ret = (int16_t) __builtin_neon_vqshruns_n_s32(__s0, __p1); \
58763 #define vqshrund_n_s64(__p0, __p1) __extension__ ({ \
58766   __ret = (int32_t) __builtin_neon_vqshrund_n_s64(__s0, __p1); \
58770 #define vqshrund_n_s64(__p0, __p1) __extension__ ({ \
58773   __ret = (int32_t) __builtin_neon_vqshrund_n_s64(__s0, __p1); \
58779 #define vqshrunh_n_s16(__p0, __p1) __extension__ ({ \
58782   __ret = (int8_t) __builtin_neon_vqshrunh_n_s16(__s0, __p1); \
58786 #define vqshrunh_n_s16(__p0, __p1) __extension__ ({ \
58789   __ret = (int8_t) __builtin_neon_vqshrunh_n_s16(__s0, __p1); \
58795 __ai uint8_t vqsubb_u8(uint8_t __p0, uint8_t __p1) {
58797   __ret = (uint8_t) __builtin_neon_vqsubb_u8(__p0, __p1);
58801 __ai uint8_t vqsubb_u8(uint8_t __p0, uint8_t __p1) {
58803   __ret = (uint8_t) __builtin_neon_vqsubb_u8(__p0, __p1);
58809 __ai uint32_t vqsubs_u32(uint32_t __p0, uint32_t __p1) {
58811   __ret = (uint32_t) __builtin_neon_vqsubs_u32(__p0, __p1);
58815 __ai uint32_t vqsubs_u32(uint32_t __p0, uint32_t __p1) {
58817   __ret = (uint32_t) __builtin_neon_vqsubs_u32(__p0, __p1);
58823 __ai uint64_t vqsubd_u64(uint64_t __p0, uint64_t __p1) {
58825   __ret = (uint64_t) __builtin_neon_vqsubd_u64(__p0, __p1);
58829 __ai uint64_t vqsubd_u64(uint64_t __p0, uint64_t __p1) {
58831   __ret = (uint64_t) __builtin_neon_vqsubd_u64(__p0, __p1);
58837 __ai uint16_t vqsubh_u16(uint16_t __p0, uint16_t __p1) {
58839   __ret = (uint16_t) __builtin_neon_vqsubh_u16(__p0, __p1);
58843 __ai uint16_t vqsubh_u16(uint16_t __p0, uint16_t __p1) {
58845   __ret = (uint16_t) __builtin_neon_vqsubh_u16(__p0, __p1);
58851 __ai int8_t vqsubb_s8(int8_t __p0, int8_t __p1) {
58853   __ret = (int8_t) __builtin_neon_vqsubb_s8(__p0, __p1);
58857 __ai int8_t vqsubb_s8(int8_t __p0, int8_t __p1) {
58859   __ret = (int8_t) __builtin_neon_vqsubb_s8(__p0, __p1);
58865 __ai int32_t vqsubs_s32(int32_t __p0, int32_t __p1) {
58867   __ret = (int32_t) __builtin_neon_vqsubs_s32(__p0, __p1);
58871 __ai int32_t vqsubs_s32(int32_t __p0, int32_t __p1) {
58873   __ret = (int32_t) __builtin_neon_vqsubs_s32(__p0, __p1);
58876 __ai int32_t __noswap_vqsubs_s32(int32_t __p0, int32_t __p1) {
58878   __ret = (int32_t) __builtin_neon_vqsubs_s32(__p0, __p1);
58884 __ai int64_t vqsubd_s64(int64_t __p0, int64_t __p1) {
58886   __ret = (int64_t) __builtin_neon_vqsubd_s64(__p0, __p1);
58890 __ai int64_t vqsubd_s64(int64_t __p0, int64_t __p1) {
58892   __ret = (int64_t) __builtin_neon_vqsubd_s64(__p0, __p1);
58898 __ai int16_t vqsubh_s16(int16_t __p0, int16_t __p1) {
58900   __ret = (int16_t) __builtin_neon_vqsubh_s16(__p0, __p1);
58904 __ai int16_t vqsubh_s16(int16_t __p0, int16_t __p1) {
58906   __ret = (int16_t) __builtin_neon_vqsubh_s16(__p0, __p1);
58909 __ai int16_t __noswap_vqsubh_s16(int16_t __p0, int16_t __p1) {
58911   __ret = (int16_t) __builtin_neon_vqsubh_s16(__p0, __p1);
58917 __ai poly8x8_t vqtbl1_p8(poly8x16_t __p0, uint8x8_t __p1) {
58919   __ret = (poly8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 4);
58923 __ai poly8x8_t vqtbl1_p8(poly8x16_t __p0, uint8x8_t __p1) {
58925   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58934 __ai poly8x16_t vqtbl1q_p8(poly8x16_t __p0, uint8x16_t __p1) {
58936   __ret = (poly8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 36);
58940 __ai poly8x16_t vqtbl1q_p8(poly8x16_t __p0, uint8x16_t __p1) {
58942   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58951 __ai uint8x16_t vqtbl1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
58953   __ret = (uint8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
58957 __ai uint8x16_t vqtbl1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
58959   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58968 __ai int8x16_t vqtbl1q_s8(int8x16_t __p0, int8x16_t __p1) {
58970   __ret = (int8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
58974 __ai int8x16_t vqtbl1q_s8(int8x16_t __p0, int8x16_t __p1) {
58976   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58985 __ai uint8x8_t vqtbl1_u8(uint8x16_t __p0, uint8x8_t __p1) {
58987   __ret = (uint8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 16);
58991 __ai uint8x8_t vqtbl1_u8(uint8x16_t __p0, uint8x8_t __p1) {
58993   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59002 __ai int8x8_t vqtbl1_s8(int8x16_t __p0, int8x8_t __p1) {
59004   __ret = (int8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 0);
59008 __ai int8x8_t vqtbl1_s8(int8x16_t __p0, int8x8_t __p1) {
59010   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59019 __ai poly8x8_t vqtbl2_p8(poly8x16x2_t __p0, uint8x8_t __p1) {
59021   __ret = (poly8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 4);
59025 __ai poly8x8_t vqtbl2_p8(poly8x16x2_t __p0, uint8x8_t __p1) {
59029   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59038 __ai poly8x16_t vqtbl2q_p8(poly8x16x2_t __p0, uint8x16_t __p1) {
59040   __ret = (poly8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 36);
59044 __ai poly8x16_t vqtbl2q_p8(poly8x16x2_t __p0, uint8x16_t __p1) {
59048   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59057 __ai uint8x16_t vqtbl2q_u8(uint8x16x2_t __p0, uint8x16_t __p1) {
59059   __ret = (uint8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 48);
59063 __ai uint8x16_t vqtbl2q_u8(uint8x16x2_t __p0, uint8x16_t __p1) {
59067   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59076 __ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, int8x16_t __p1) {
59078   __ret = (int8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 32);
59082 __ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, int8x16_t __p1) {
59086   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59095 __ai uint8x8_t vqtbl2_u8(uint8x16x2_t __p0, uint8x8_t __p1) {
59097   __ret = (uint8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 16);
59101 __ai uint8x8_t vqtbl2_u8(uint8x16x2_t __p0, uint8x8_t __p1) {
59105   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59114 __ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, int8x8_t __p1) {
59116   __ret = (int8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 0);
59120 __ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, int8x8_t __p1) {
59124   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59133 __ai poly8x8_t vqtbl3_p8(poly8x16x3_t __p0, uint8x8_t __p1) {
59135   __ret = (poly8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x8_t)__p1, 4);
59139 __ai poly8x8_t vqtbl3_p8(poly8x16x3_t __p0, uint8x8_t __p1) {
59144   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59153 __ai poly8x16_t vqtbl3q_p8(poly8x16x3_t __p0, uint8x16_t __p1) {
59155   __ret = (poly8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p1, 36);
59159 __ai poly8x16_t vqtbl3q_p8(poly8x16x3_t __p0, uint8x16_t __p1) {
59164   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59173 __ai uint8x16_t vqtbl3q_u8(uint8x16x3_t __p0, uint8x16_t __p1) {
59175   __ret = (uint8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p1, 48);
59179 __ai uint8x16_t vqtbl3q_u8(uint8x16x3_t __p0, uint8x16_t __p1) {
59184   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59193 __ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, int8x16_t __p1) {
59195   __ret = (int8x16_t) __builtin_neon_vqtbl3q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p1, 32);
59199 __ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, int8x16_t __p1) {
59204   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59213 __ai uint8x8_t vqtbl3_u8(uint8x16x3_t __p0, uint8x8_t __p1) {
59215   __ret = (uint8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x8_t)__p1, 16);
59219 __ai uint8x8_t vqtbl3_u8(uint8x16x3_t __p0, uint8x8_t __p1) {
59224   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59233 __ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, int8x8_t __p1) {
59235   __ret = (int8x8_t) __builtin_neon_vqtbl3_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x8_t)__p1, 0);
59239 __ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, int8x8_t __p1) {
59244   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59253 __ai poly8x8_t vqtbl4_p8(poly8x16x4_t __p0, uint8x8_t __p1) {
59255   __ret = (poly8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x8_t)__p1, 4);
59259 __ai poly8x8_t vqtbl4_p8(poly8x16x4_t __p0, uint8x8_t __p1) {
59265   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59274 __ai poly8x16_t vqtbl4q_p8(poly8x16x4_t __p0, uint8x16_t __p1) {
59276   __ret = (poly8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x16_t)__p1, 36);
59280 __ai poly8x16_t vqtbl4q_p8(poly8x16x4_t __p0, uint8x16_t __p1) {
59286   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59295 __ai uint8x16_t vqtbl4q_u8(uint8x16x4_t __p0, uint8x16_t __p1) {
59297   __ret = (uint8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x16_t)__p1, 48);
59301 __ai uint8x16_t vqtbl4q_u8(uint8x16x4_t __p0, uint8x16_t __p1) {
59307   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59316 __ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, int8x16_t __p1) {
59318   __ret = (int8x16_t) __builtin_neon_vqtbl4q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x16_t)__p1, 32);
59322 __ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, int8x16_t __p1) {
59328   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59337 __ai uint8x8_t vqtbl4_u8(uint8x16x4_t __p0, uint8x8_t __p1) {
59339   __ret = (uint8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x8_t)__p1, 16);
59343 __ai uint8x8_t vqtbl4_u8(uint8x16x4_t __p0, uint8x8_t __p1) {
59349   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59358 __ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, int8x8_t __p1) {
59360   __ret = (int8x8_t) __builtin_neon_vqtbl4_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p0.val[2], (int8x16_t)__p0.val[3], (int8x8_t)__p1, 0);
59364 __ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, int8x8_t __p1) {
59370   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59379 __ai poly8x8_t vqtbx1_p8(poly8x8_t __p0, poly8x16_t __p1, uint8x8_t __p2) {
59381   __ret = (poly8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 4);
59385 __ai poly8x8_t vqtbx1_p8(poly8x8_t __p0, poly8x16_t __p1, uint8x8_t __p2) {
59387   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59397 __ai poly8x16_t vqtbx1q_p8(poly8x16_t __p0, poly8x16_t __p1, uint8x16_t __p2) {
59399   __ret = (poly8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 36);
59403 __ai poly8x16_t vqtbx1q_p8(poly8x16_t __p0, poly8x16_t __p1, uint8x16_t __p2) {
59405   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59415 __ai uint8x16_t vqtbx1q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
59417   __ret = (uint8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 48);
59421 __ai uint8x16_t vqtbx1q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
59423   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59433 __ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
59435   __ret = (int8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 32);
59439 __ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
59441   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59451 __ai uint8x8_t vqtbx1_u8(uint8x8_t __p0, uint8x16_t __p1, uint8x8_t __p2) {
59453   __ret = (uint8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 16);
59457 __ai uint8x8_t vqtbx1_u8(uint8x8_t __p0, uint8x16_t __p1, uint8x8_t __p2) {
59459   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59469 __ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, int8x8_t __p2) {
59471   __ret = (int8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 0);
59475 __ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, int8x8_t __p2) {
59477   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59487 __ai poly8x8_t vqtbx2_p8(poly8x8_t __p0, poly8x16x2_t __p1, uint8x8_t __p2) {
59489   __ret = (poly8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 4);
59493 __ai poly8x8_t vqtbx2_p8(poly8x8_t __p0, poly8x16x2_t __p1, uint8x8_t __p2) {
59496   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59497   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59507 __ai poly8x16_t vqtbx2q_p8(poly8x16_t __p0, poly8x16x2_t __p1, uint8x16_t __p2) {
59509   __ret = (poly8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 36);
59513 __ai poly8x16_t vqtbx2q_p8(poly8x16_t __p0, poly8x16x2_t __p1, uint8x16_t __p2) {
59516   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59517   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59527 __ai uint8x16_t vqtbx2q_u8(uint8x16_t __p0, uint8x16x2_t __p1, uint8x16_t __p2) {
59529   __ret = (uint8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 48);
59533 __ai uint8x16_t vqtbx2q_u8(uint8x16_t __p0, uint8x16x2_t __p1, uint8x16_t __p2) {
59536   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59537   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59547 __ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, int8x16_t __p2) {
59549   __ret = (int8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 32);
59553 __ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, int8x16_t __p2) {
59556   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59557   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59567 __ai uint8x8_t vqtbx2_u8(uint8x8_t __p0, uint8x16x2_t __p1, uint8x8_t __p2) {
59569   __ret = (uint8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 16);
59573 __ai uint8x8_t vqtbx2_u8(uint8x8_t __p0, uint8x16x2_t __p1, uint8x8_t __p2) {
59576   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59577   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59587 __ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, int8x8_t __p2) {
59589   __ret = (int8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 0);
59593 __ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, int8x8_t __p2) {
59596   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59597   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59607 __ai poly8x8_t vqtbx3_p8(poly8x8_t __p0, poly8x16x3_t __p1, uint8x8_t __p2) {
59609   __ret = (poly8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x8_t)__p2, 4);
59613 __ai poly8x8_t vqtbx3_p8(poly8x8_t __p0, poly8x16x3_t __p1, uint8x8_t __p2) {
59616   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59617   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59618   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59628 __ai poly8x16_t vqtbx3q_p8(poly8x16_t __p0, poly8x16x3_t __p1, uint8x16_t __p2) {
59630   __ret = (poly8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p2, 36);
59634 __ai poly8x16_t vqtbx3q_p8(poly8x16_t __p0, poly8x16x3_t __p1, uint8x16_t __p2) {
59637   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59638   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59639   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59649 __ai uint8x16_t vqtbx3q_u8(uint8x16_t __p0, uint8x16x3_t __p1, uint8x16_t __p2) {
59651   __ret = (uint8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p2, 48);
59655 __ai uint8x16_t vqtbx3q_u8(uint8x16_t __p0, uint8x16x3_t __p1, uint8x16_t __p2) {
59658   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59659   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59660   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59670 __ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, int8x16_t __p2) {
59672   __ret = (int8x16_t) __builtin_neon_vqtbx3q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p2, 32);
59676 __ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, int8x16_t __p2) {
59679   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59680   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59681   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59691 __ai uint8x8_t vqtbx3_u8(uint8x8_t __p0, uint8x16x3_t __p1, uint8x8_t __p2) {
59693   __ret = (uint8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x8_t)__p2, 16);
59697 __ai uint8x8_t vqtbx3_u8(uint8x8_t __p0, uint8x16x3_t __p1, uint8x8_t __p2) {
59700   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59701   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59702   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59712 __ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, int8x8_t __p2) {
59714   __ret = (int8x8_t) __builtin_neon_vqtbx3_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x8_t)__p2, 0);
59718 __ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, int8x8_t __p2) {
59721   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59722   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59723   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59733 __ai poly8x8_t vqtbx4_p8(poly8x8_t __p0, poly8x16x4_t __p1, uint8x8_t __p2) {
59735   __ret = (poly8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x8_t)__p2, 4);
59739 __ai poly8x8_t vqtbx4_p8(poly8x8_t __p0, poly8x16x4_t __p1, uint8x8_t __p2) {
59742   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59743   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59744   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59745   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59755 __ai poly8x16_t vqtbx4q_p8(poly8x16_t __p0, poly8x16x4_t __p1, uint8x16_t __p2) {
59757   __ret = (poly8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x16_t)__p2, 36);
59761 __ai poly8x16_t vqtbx4q_p8(poly8x16_t __p0, poly8x16x4_t __p1, uint8x16_t __p2) {
59764   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59765   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59766   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59767   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59777 __ai uint8x16_t vqtbx4q_u8(uint8x16_t __p0, uint8x16x4_t __p1, uint8x16_t __p2) {
59779   __ret = (uint8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x16_t)__p2, 48);
59783 __ai uint8x16_t vqtbx4q_u8(uint8x16_t __p0, uint8x16x4_t __p1, uint8x16_t __p2) {
59786   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59787   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59788   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59789   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59799 __ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, int8x16_t __p2) {
59801   __ret = (int8x16_t) __builtin_neon_vqtbx4q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x16_t)__p2, 32);
59805 __ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, int8x16_t __p2) {
59808   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59809   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59810   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59811   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59821 __ai uint8x8_t vqtbx4_u8(uint8x8_t __p0, uint8x16x4_t __p1, uint8x8_t __p2) {
59823   __ret = (uint8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x8_t)__p2, 16);
59827 __ai uint8x8_t vqtbx4_u8(uint8x8_t __p0, uint8x16x4_t __p1, uint8x8_t __p2) {
59830   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59831   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59832   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59833   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59843 __ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, int8x8_t __p2) {
59845   __ret = (int8x8_t) __builtin_neon_vqtbx4_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p1.val[2], (int8x16_t)__p1.val[3], (int8x8_t)__p2, 0);
59849 __ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, int8x8_t __p2) {
59852   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59853   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59854   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59855   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
59865 __ai uint16x8_t vraddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
59867   __ret = vcombine_u16(__p0, vraddhn_u32(__p1, __p2));
59871 __ai uint16x8_t vraddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
59873   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59883 __ai uint32x4_t vraddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
59885   __ret = vcombine_u32(__p0, vraddhn_u64(__p1, __p2));
59889 __ai uint32x4_t vraddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
59891   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
59901 __ai uint8x16_t vraddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
59903   __ret = vcombine_u8(__p0, vraddhn_u16(__p1, __p2));
59907 __ai uint8x16_t vraddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
59909   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59919 __ai int16x8_t vraddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
59921   __ret = vcombine_s16(__p0, vraddhn_s32(__p1, __p2));
59925 __ai int16x8_t vraddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
59927   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59937 __ai int32x4_t vraddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
59939   __ret = vcombine_s32(__p0, vraddhn_s64(__p1, __p2));
59943 __ai int32x4_t vraddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
59945   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
59955 __ai int8x16_t vraddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
59957   __ret = vcombine_s8(__p0, vraddhn_s16(__p1, __p2));
59961 __ai int8x16_t vraddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
59963   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
60127 __ai float64x2_t vrecpsq_f64(float64x2_t __p0, float64x2_t __p1) {
60129   __ret = (float64x2_t) __builtin_neon_vrecpsq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
60133 __ai float64x2_t vrecpsq_f64(float64x2_t __p0, float64x2_t __p1) {
60135   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
60144 __ai float64x1_t vrecps_f64(float64x1_t __p0, float64x1_t __p1) {
60146   __ret = (float64x1_t) __builtin_neon_vrecps_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
60150 __ai float64x1_t vrecps_f64(float64x1_t __p0, float64x1_t __p1) {
60152   __ret = (float64x1_t) __builtin_neon_vrecps_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
60158 __ai float64_t vrecpsd_f64(float64_t __p0, float64_t __p1) {
60160   __ret = (float64_t) __builtin_neon_vrecpsd_f64(__p0, __p1);
60164 __ai float64_t vrecpsd_f64(float64_t __p0, float64_t __p1) {
60166   __ret = (float64_t) __builtin_neon_vrecpsd_f64(__p0, __p1);
60172 __ai float32_t vrecpss_f32(float32_t __p0, float32_t __p1) {
60174   __ret = (float32_t) __builtin_neon_vrecpss_f32(__p0, __p1);
60178 __ai float32_t vrecpss_f32(float32_t __p0, float32_t __p1) {
60180   __ret = (float32_t) __builtin_neon_vrecpss_f32(__p0, __p1);
60214 __ai uint64_t vrshld_u64(uint64_t __p0, uint64_t __p1) {
60216   __ret = (uint64_t) __builtin_neon_vrshld_u64(__p0, __p1);
60220 __ai uint64_t vrshld_u64(uint64_t __p0, uint64_t __p1) {
60222   __ret = (uint64_t) __builtin_neon_vrshld_u64(__p0, __p1);
60228 __ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
60230   __ret = (int64_t) __builtin_neon_vrshld_s64(__p0, __p1);
60234 __ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
60236   __ret = (int64_t) __builtin_neon_vrshld_s64(__p0, __p1);
60242 #define vrshrd_n_u64(__p0, __p1) __extension__ ({ \
60245   __ret = (uint64_t) __builtin_neon_vrshrd_n_u64(__s0, __p1); \
60249 #define vrshrd_n_u64(__p0, __p1) __extension__ ({ \
60252   __ret = (uint64_t) __builtin_neon_vrshrd_n_u64(__s0, __p1); \
60258 #define vrshrd_n_s64(__p0, __p1) __extension__ ({ \
60261   __ret = (int64_t) __builtin_neon_vrshrd_n_s64(__s0, __p1); \
60265 #define vrshrd_n_s64(__p0, __p1) __extension__ ({ \
60268   __ret = (int64_t) __builtin_neon_vrshrd_n_s64(__s0, __p1); \
60458 __ai float64x2_t vrsqrtsq_f64(float64x2_t __p0, float64x2_t __p1) {
60460   __ret = (float64x2_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
60464 __ai float64x2_t vrsqrtsq_f64(float64x2_t __p0, float64x2_t __p1) {
60466   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
60475 __ai float64x1_t vrsqrts_f64(float64x1_t __p0, float64x1_t __p1) {
60477   __ret = (float64x1_t) __builtin_neon_vrsqrts_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
60481 __ai float64x1_t vrsqrts_f64(float64x1_t __p0, float64x1_t __p1) {
60483   __ret = (float64x1_t) __builtin_neon_vrsqrts_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
60489 __ai float64_t vrsqrtsd_f64(float64_t __p0, float64_t __p1) {
60491   __ret = (float64_t) __builtin_neon_vrsqrtsd_f64(__p0, __p1);
60495 __ai float64_t vrsqrtsd_f64(float64_t __p0, float64_t __p1) {
60497   __ret = (float64_t) __builtin_neon_vrsqrtsd_f64(__p0, __p1);
60503 __ai float32_t vrsqrtss_f32(float32_t __p0, float32_t __p1) {
60505   __ret = (float32_t) __builtin_neon_vrsqrtss_f32(__p0, __p1);
60509 __ai float32_t vrsqrtss_f32(float32_t __p0, float32_t __p1) {
60511   __ret = (float32_t) __builtin_neon_vrsqrtss_f32(__p0, __p1);
60517 #define vrsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
60519   uint64_t __s1 = __p1; \
60525 #define vrsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
60527   uint64_t __s1 = __p1; \
60535 #define vrsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
60537   int64_t __s1 = __p1; \
60543 #define vrsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
60545   int64_t __s1 = __p1; \
60553 __ai uint16x8_t vrsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
60555   __ret = vcombine_u16(__p0, vrsubhn_u32(__p1, __p2));
60559 __ai uint16x8_t vrsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
60561   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
60571 __ai uint32x4_t vrsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
60573   __ret = vcombine_u32(__p0, vrsubhn_u64(__p1, __p2));
60577 __ai uint32x4_t vrsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
60579   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
60589 __ai uint8x16_t vrsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
60591   __ret = vcombine_u8(__p0, vrsubhn_u16(__p1, __p2));
60595 __ai uint8x16_t vrsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
60597   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
60607 __ai int16x8_t vrsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
60609   __ret = vcombine_s16(__p0, vrsubhn_s32(__p1, __p2));
60613 __ai int16x8_t vrsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
60615   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
60625 __ai int32x4_t vrsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
60627   __ret = vcombine_s32(__p0, vrsubhn_s64(__p1, __p2));
60631 __ai int32x4_t vrsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
60633   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
60643 __ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
60645   __ret = vcombine_s8(__p0, vrsubhn_s16(__p1, __p2));
60649 __ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
60651   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
60661 #define vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60663   poly64x1_t __s1 = __p1; \
60669 #define vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60671   poly64x1_t __s1 = __p1; \
60676 #define __noswap_vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60678   poly64x1_t __s1 = __p1; \
60686 #define vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60688   poly64x2_t __s1 = __p1; \
60694 #define vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60696   poly64x2_t __s1 = __p1; \
60703 #define __noswap_vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60705   poly64x2_t __s1 = __p1; \
60713 #define vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60715   float64x2_t __s1 = __p1; \
60721 #define vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60723   float64x2_t __s1 = __p1; \
60730 #define __noswap_vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60732   float64x2_t __s1 = __p1; \
60740 #define vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60742   float64x1_t __s1 = __p1; \
60748 #define vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60750   float64x1_t __s1 = __p1; \
60755 #define __noswap_vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60757   float64x1_t __s1 = __p1; \
60765 __ai uint64_t vshld_u64(uint64_t __p0, uint64_t __p1) {
60767   __ret = (uint64_t) __builtin_neon_vshld_u64(__p0, __p1);
60771 __ai uint64_t vshld_u64(uint64_t __p0, uint64_t __p1) {
60773   __ret = (uint64_t) __builtin_neon_vshld_u64(__p0, __p1);
60779 __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
60781   __ret = (int64_t) __builtin_neon_vshld_s64(__p0, __p1);
60785 __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
60787   __ret = (int64_t) __builtin_neon_vshld_s64(__p0, __p1);
60793 #define vshld_n_u64(__p0, __p1) __extension__ ({ \
60796   __ret = (uint64_t) __builtin_neon_vshld_n_u64(__s0, __p1); \
60800 #define vshld_n_u64(__p0, __p1) __extension__ ({ \
60803   __ret = (uint64_t) __builtin_neon_vshld_n_u64(__s0, __p1); \
60809 #define vshld_n_s64(__p0, __p1) __extension__ ({ \
60812   __ret = (int64_t) __builtin_neon_vshld_n_s64(__s0, __p1); \
60816 #define vshld_n_s64(__p0, __p1) __extension__ ({ \
60819   __ret = (int64_t) __builtin_neon_vshld_n_s64(__s0, __p1); \
60933 #define vshrd_n_u64(__p0, __p1) __extension__ ({ \
60936   __ret = (uint64_t) __builtin_neon_vshrd_n_u64(__s0, __p1); \
60940 #define vshrd_n_u64(__p0, __p1) __extension__ ({ \
60943   __ret = (uint64_t) __builtin_neon_vshrd_n_u64(__s0, __p1); \
60949 #define vshrd_n_s64(__p0, __p1) __extension__ ({ \
60952   __ret = (int64_t) __builtin_neon_vshrd_n_s64(__s0, __p1); \
60956 #define vshrd_n_s64(__p0, __p1) __extension__ ({ \
60959   __ret = (int64_t) __builtin_neon_vshrd_n_s64(__s0, __p1); \
61091 #define vslid_n_u64(__p0, __p1, __p2) __extension__ ({ \
61093   uint64_t __s1 = __p1; \
61099 #define vslid_n_u64(__p0, __p1, __p2) __extension__ ({ \
61101   uint64_t __s1 = __p1; \
61109 #define vslid_n_s64(__p0, __p1, __p2) __extension__ ({ \
61111   int64_t __s1 = __p1; \
61117 #define vslid_n_s64(__p0, __p1, __p2) __extension__ ({ \
61119   int64_t __s1 = __p1; \
61127 #define vsli_n_p64(__p0, __p1, __p2) __extension__ ({ \
61129   poly64x1_t __s1 = __p1; \
61135 #define vsli_n_p64(__p0, __p1, __p2) __extension__ ({ \
61137   poly64x1_t __s1 = __p1; \
61145 #define vsliq_n_p64(__p0, __p1, __p2) __extension__ ({ \
61147   poly64x2_t __s1 = __p1; \
61153 #define vsliq_n_p64(__p0, __p1, __p2) __extension__ ({ \
61155   poly64x2_t __s1 = __p1; \
61166 __ai uint8_t vsqaddb_u8(uint8_t __p0, uint8_t __p1) {
61168   __ret = (uint8_t) __builtin_neon_vsqaddb_u8(__p0, __p1);
61172 __ai uint8_t vsqaddb_u8(uint8_t __p0, uint8_t __p1) {
61174   __ret = (uint8_t) __builtin_neon_vsqaddb_u8(__p0, __p1);
61180 __ai uint32_t vsqadds_u32(uint32_t __p0, uint32_t __p1) {
61182   __ret = (uint32_t) __builtin_neon_vsqadds_u32(__p0, __p1);
61186 __ai uint32_t vsqadds_u32(uint32_t __p0, uint32_t __p1) {
61188   __ret = (uint32_t) __builtin_neon_vsqadds_u32(__p0, __p1);
61194 __ai uint64_t vsqaddd_u64(uint64_t __p0, uint64_t __p1) {
61196   __ret = (uint64_t) __builtin_neon_vsqaddd_u64(__p0, __p1);
61200 __ai uint64_t vsqaddd_u64(uint64_t __p0, uint64_t __p1) {
61202   __ret = (uint64_t) __builtin_neon_vsqaddd_u64(__p0, __p1);
61208 __ai uint16_t vsqaddh_u16(uint16_t __p0, uint16_t __p1) {
61210   __ret = (uint16_t) __builtin_neon_vsqaddh_u16(__p0, __p1);
61214 __ai uint16_t vsqaddh_u16(uint16_t __p0, uint16_t __p1) {
61216   __ret = (uint16_t) __builtin_neon_vsqaddh_u16(__p0, __p1);
61222 __ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
61224   __ret = (uint8x16_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
61228 __ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
61230   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
61239 __ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
61241   __ret = (uint32x4_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
61245 __ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
61247   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
61256 __ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
61258   __ret = (uint64x2_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
61262 __ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
61264   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
61273 __ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
61275   __ret = (uint16x8_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
61279 __ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
61281   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61290 __ai uint8x8_t vsqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
61292   __ret = (uint8x8_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
61296 __ai uint8x8_t vsqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
61298   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
61307 __ai uint32x2_t vsqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
61309   __ret = (uint32x2_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
61313 __ai uint32x2_t vsqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
61315   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
61324 __ai uint64x1_t vsqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
61326   __ret = (uint64x1_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
61330 __ai uint64x1_t vsqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
61332   __ret = (uint64x1_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
61338 __ai uint16x4_t vsqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
61340   __ret = (uint16x4_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
61344 __ai uint16x4_t vsqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
61346   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
61417 #define vsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
61419   uint64_t __s1 = __p1; \
61425 #define vsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
61427   uint64_t __s1 = __p1; \
61435 #define vsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
61437   int64_t __s1 = __p1; \
61443 #define vsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
61445   int64_t __s1 = __p1; \
61453 #define vsrid_n_u64(__p0, __p1, __p2) __extension__ ({ \
61455   uint64_t __s1 = __p1; \
61461 #define vsrid_n_u64(__p0, __p1, __p2) __extension__ ({ \
61463   uint64_t __s1 = __p1; \
61471 #define vsrid_n_s64(__p0, __p1, __p2) __extension__ ({ \
61473   int64_t __s1 = __p1; \
61479 #define vsrid_n_s64(__p0, __p1, __p2) __extension__ ({ \
61481   int64_t __s1 = __p1; \
61489 #define vsri_n_p64(__p0, __p1, __p2) __extension__ ({ \
61491   poly64x1_t __s1 = __p1; \
61497 #define vsri_n_p64(__p0, __p1, __p2) __extension__ ({ \
61499   poly64x1_t __s1 = __p1; \
61507 #define vsriq_n_p64(__p0, __p1, __p2) __extension__ ({ \
61509   poly64x2_t __s1 = __p1; \
61515 #define vsriq_n_p64(__p0, __p1, __p2) __extension__ ({ \
61517   poly64x2_t __s1 = __p1; \
61528 #define vst1_p64(__p0, __p1) __extension__ ({ \
61529   poly64x1_t __s1 = __p1; \
61533 #define vst1_p64(__p0, __p1) __extension__ ({ \
61534   poly64x1_t __s1 = __p1; \
61540 #define vst1q_p64(__p0, __p1) __extension__ ({ \
61541   poly64x2_t __s1 = __p1; \
61545 #define vst1q_p64(__p0, __p1) __extension__ ({ \
61546   poly64x2_t __s1 = __p1; \
61553 #define vst1q_f64(__p0, __p1) __extension__ ({ \
61554   float64x2_t __s1 = __p1; \
61558 #define vst1q_f64(__p0, __p1) __extension__ ({ \
61559   float64x2_t __s1 = __p1; \
61566 #define vst1_f64(__p0, __p1) __extension__ ({ \
61567   float64x1_t __s1 = __p1; \
61571 #define vst1_f64(__p0, __p1) __extension__ ({ \
61572   float64x1_t __s1 = __p1; \
61578 #define vst1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
61579   poly64x1_t __s1 = __p1; \
61583 #define vst1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
61584   poly64x1_t __s1 = __p1; \
61590 #define vst1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
61591   poly64x2_t __s1 = __p1; \
61595 #define vst1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
61596   poly64x2_t __s1 = __p1; \
61603 #define vst1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
61604   float64x2_t __s1 = __p1; \
61608 #define vst1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
61609   float64x2_t __s1 = __p1; \
61616 #define vst1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
61617   float64x1_t __s1 = __p1; \
61621 #define vst1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
61622   float64x1_t __s1 = __p1; \
61628 #define vst1_p8_x2(__p0, __p1) __extension__ ({ \
61629   poly8x8x2_t __s1 = __p1; \
61633 #define vst1_p8_x2(__p0, __p1) __extension__ ({ \
61634   poly8x8x2_t __s1 = __p1; \
61643 #define vst1_p64_x2(__p0, __p1) __extension__ ({ \
61644   poly64x1x2_t __s1 = __p1; \
61648 #define vst1_p64_x2(__p0, __p1) __extension__ ({ \
61649   poly64x1x2_t __s1 = __p1; \
61655 #define vst1_p16_x2(__p0, __p1) __extension__ ({ \
61656   poly16x4x2_t __s1 = __p1; \
61660 #define vst1_p16_x2(__p0, __p1) __extension__ ({ \
61661   poly16x4x2_t __s1 = __p1; \
61670 #define vst1q_p8_x2(__p0, __p1) __extension__ ({ \
61671   poly8x16x2_t __s1 = __p1; \
61675 #define vst1q_p8_x2(__p0, __p1) __extension__ ({ \
61676   poly8x16x2_t __s1 = __p1; \
61685 #define vst1q_p64_x2(__p0, __p1) __extension__ ({ \
61686   poly64x2x2_t __s1 = __p1; \
61690 #define vst1q_p64_x2(__p0, __p1) __extension__ ({ \
61691   poly64x2x2_t __s1 = __p1; \
61700 #define vst1q_p16_x2(__p0, __p1) __extension__ ({ \
61701   poly16x8x2_t __s1 = __p1; \
61705 #define vst1q_p16_x2(__p0, __p1) __extension__ ({ \
61706   poly16x8x2_t __s1 = __p1; \
61715 #define vst1q_u8_x2(__p0, __p1) __extension__ ({ \
61716   uint8x16x2_t __s1 = __p1; \
61720 #define vst1q_u8_x2(__p0, __p1) __extension__ ({ \
61721   uint8x16x2_t __s1 = __p1; \
61730 #define vst1q_u32_x2(__p0, __p1) __extension__ ({ \
61731   uint32x4x2_t __s1 = __p1; \
61735 #define vst1q_u32_x2(__p0, __p1) __extension__ ({ \
61736   uint32x4x2_t __s1 = __p1; \
61745 #define vst1q_u64_x2(__p0, __p1) __extension__ ({ \
61746   uint64x2x2_t __s1 = __p1; \
61750 #define vst1q_u64_x2(__p0, __p1) __extension__ ({ \
61751   uint64x2x2_t __s1 = __p1; \
61760 #define vst1q_u16_x2(__p0, __p1) __extension__ ({ \
61761   uint16x8x2_t __s1 = __p1; \
61765 #define vst1q_u16_x2(__p0, __p1) __extension__ ({ \
61766   uint16x8x2_t __s1 = __p1; \
61775 #define vst1q_s8_x2(__p0, __p1) __extension__ ({ \
61776   int8x16x2_t __s1 = __p1; \
61780 #define vst1q_s8_x2(__p0, __p1) __extension__ ({ \
61781   int8x16x2_t __s1 = __p1; \
61790 #define vst1q_f64_x2(__p0, __p1) __extension__ ({ \
61791   float64x2x2_t __s1 = __p1; \
61795 #define vst1q_f64_x2(__p0, __p1) __extension__ ({ \
61796   float64x2x2_t __s1 = __p1; \
61805 #define vst1q_f32_x2(__p0, __p1) __extension__ ({ \
61806   float32x4x2_t __s1 = __p1; \
61810 #define vst1q_f32_x2(__p0, __p1) __extension__ ({ \
61811   float32x4x2_t __s1 = __p1; \
61820 #define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
61821   float16x8x2_t __s1 = __p1; \
61825 #define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
61826   float16x8x2_t __s1 = __p1; \
61835 #define vst1q_s32_x2(__p0, __p1) __extension__ ({ \
61836   int32x4x2_t __s1 = __p1; \
61840 #define vst1q_s32_x2(__p0, __p1) __extension__ ({ \
61841   int32x4x2_t __s1 = __p1; \
61850 #define vst1q_s64_x2(__p0, __p1) __extension__ ({ \
61851   int64x2x2_t __s1 = __p1; \
61855 #define vst1q_s64_x2(__p0, __p1) __extension__ ({ \
61856   int64x2x2_t __s1 = __p1; \
61865 #define vst1q_s16_x2(__p0, __p1) __extension__ ({ \
61866   int16x8x2_t __s1 = __p1; \
61870 #define vst1q_s16_x2(__p0, __p1) __extension__ ({ \
61871   int16x8x2_t __s1 = __p1; \
61880 #define vst1_u8_x2(__p0, __p1) __extension__ ({ \
61881   uint8x8x2_t __s1 = __p1; \
61885 #define vst1_u8_x2(__p0, __p1) __extension__ ({ \
61886   uint8x8x2_t __s1 = __p1; \
61895 #define vst1_u32_x2(__p0, __p1) __extension__ ({ \
61896   uint32x2x2_t __s1 = __p1; \
61900 #define vst1_u32_x2(__p0, __p1) __extension__ ({ \
61901   uint32x2x2_t __s1 = __p1; \
61910 #define vst1_u64_x2(__p0, __p1) __extension__ ({ \
61911   uint64x1x2_t __s1 = __p1; \
61915 #define vst1_u64_x2(__p0, __p1) __extension__ ({ \
61916   uint64x1x2_t __s1 = __p1; \
61922 #define vst1_u16_x2(__p0, __p1) __extension__ ({ \
61923   uint16x4x2_t __s1 = __p1; \
61927 #define vst1_u16_x2(__p0, __p1) __extension__ ({ \
61928   uint16x4x2_t __s1 = __p1; \
61937 #define vst1_s8_x2(__p0, __p1) __extension__ ({ \
61938   int8x8x2_t __s1 = __p1; \
61942 #define vst1_s8_x2(__p0, __p1) __extension__ ({ \
61943   int8x8x2_t __s1 = __p1; \
61952 #define vst1_f64_x2(__p0, __p1) __extension__ ({ \
61953   float64x1x2_t __s1 = __p1; \
61957 #define vst1_f64_x2(__p0, __p1) __extension__ ({ \
61958   float64x1x2_t __s1 = __p1; \
61964 #define vst1_f32_x2(__p0, __p1) __extension__ ({ \
61965   float32x2x2_t __s1 = __p1; \
61969 #define vst1_f32_x2(__p0, __p1) __extension__ ({ \
61970   float32x2x2_t __s1 = __p1; \
61979 #define vst1_f16_x2(__p0, __p1) __extension__ ({ \
61980   float16x4x2_t __s1 = __p1; \
61984 #define vst1_f16_x2(__p0, __p1) __extension__ ({ \
61985   float16x4x2_t __s1 = __p1; \
61994 #define vst1_s32_x2(__p0, __p1) __extension__ ({ \
61995   int32x2x2_t __s1 = __p1; \
61999 #define vst1_s32_x2(__p0, __p1) __extension__ ({ \
62000   int32x2x2_t __s1 = __p1; \
62009 #define vst1_s64_x2(__p0, __p1) __extension__ ({ \
62010   int64x1x2_t __s1 = __p1; \
62014 #define vst1_s64_x2(__p0, __p1) __extension__ ({ \
62015   int64x1x2_t __s1 = __p1; \
62021 #define vst1_s16_x2(__p0, __p1) __extension__ ({ \
62022   int16x4x2_t __s1 = __p1; \
62026 #define vst1_s16_x2(__p0, __p1) __extension__ ({ \
62027   int16x4x2_t __s1 = __p1; \
62036 #define vst1_p8_x3(__p0, __p1) __extension__ ({ \
62037   poly8x8x3_t __s1 = __p1; \
62041 #define vst1_p8_x3(__p0, __p1) __extension__ ({ \
62042   poly8x8x3_t __s1 = __p1; \
62052 #define vst1_p64_x3(__p0, __p1) __extension__ ({ \
62053   poly64x1x3_t __s1 = __p1; \
62057 #define vst1_p64_x3(__p0, __p1) __extension__ ({ \
62058   poly64x1x3_t __s1 = __p1; \
62064 #define vst1_p16_x3(__p0, __p1) __extension__ ({ \
62065   poly16x4x3_t __s1 = __p1; \
62069 #define vst1_p16_x3(__p0, __p1) __extension__ ({ \
62070   poly16x4x3_t __s1 = __p1; \
62080 #define vst1q_p8_x3(__p0, __p1) __extension__ ({ \
62081   poly8x16x3_t __s1 = __p1; \
62085 #define vst1q_p8_x3(__p0, __p1) __extension__ ({ \
62086   poly8x16x3_t __s1 = __p1; \
62096 #define vst1q_p64_x3(__p0, __p1) __extension__ ({ \
62097   poly64x2x3_t __s1 = __p1; \
62101 #define vst1q_p64_x3(__p0, __p1) __extension__ ({ \
62102   poly64x2x3_t __s1 = __p1; \
62112 #define vst1q_p16_x3(__p0, __p1) __extension__ ({ \
62113   poly16x8x3_t __s1 = __p1; \
62117 #define vst1q_p16_x3(__p0, __p1) __extension__ ({ \
62118   poly16x8x3_t __s1 = __p1; \
62128 #define vst1q_u8_x3(__p0, __p1) __extension__ ({ \
62129   uint8x16x3_t __s1 = __p1; \
62133 #define vst1q_u8_x3(__p0, __p1) __extension__ ({ \
62134   uint8x16x3_t __s1 = __p1; \
62144 #define vst1q_u32_x3(__p0, __p1) __extension__ ({ \
62145   uint32x4x3_t __s1 = __p1; \
62149 #define vst1q_u32_x3(__p0, __p1) __extension__ ({ \
62150   uint32x4x3_t __s1 = __p1; \
62160 #define vst1q_u64_x3(__p0, __p1) __extension__ ({ \
62161   uint64x2x3_t __s1 = __p1; \
62165 #define vst1q_u64_x3(__p0, __p1) __extension__ ({ \
62166   uint64x2x3_t __s1 = __p1; \
62176 #define vst1q_u16_x3(__p0, __p1) __extension__ ({ \
62177   uint16x8x3_t __s1 = __p1; \
62181 #define vst1q_u16_x3(__p0, __p1) __extension__ ({ \
62182   uint16x8x3_t __s1 = __p1; \
62192 #define vst1q_s8_x3(__p0, __p1) __extension__ ({ \
62193   int8x16x3_t __s1 = __p1; \
62197 #define vst1q_s8_x3(__p0, __p1) __extension__ ({ \
62198   int8x16x3_t __s1 = __p1; \
62208 #define vst1q_f64_x3(__p0, __p1) __extension__ ({ \
62209   float64x2x3_t __s1 = __p1; \
62213 #define vst1q_f64_x3(__p0, __p1) __extension__ ({ \
62214   float64x2x3_t __s1 = __p1; \
62224 #define vst1q_f32_x3(__p0, __p1) __extension__ ({ \
62225   float32x4x3_t __s1 = __p1; \
62229 #define vst1q_f32_x3(__p0, __p1) __extension__ ({ \
62230   float32x4x3_t __s1 = __p1; \
62240 #define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
62241   float16x8x3_t __s1 = __p1; \
62245 #define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
62246   float16x8x3_t __s1 = __p1; \
62256 #define vst1q_s32_x3(__p0, __p1) __extension__ ({ \
62257   int32x4x3_t __s1 = __p1; \
62261 #define vst1q_s32_x3(__p0, __p1) __extension__ ({ \
62262   int32x4x3_t __s1 = __p1; \
62272 #define vst1q_s64_x3(__p0, __p1) __extension__ ({ \
62273   int64x2x3_t __s1 = __p1; \
62277 #define vst1q_s64_x3(__p0, __p1) __extension__ ({ \
62278   int64x2x3_t __s1 = __p1; \
62288 #define vst1q_s16_x3(__p0, __p1) __extension__ ({ \
62289   int16x8x3_t __s1 = __p1; \
62293 #define vst1q_s16_x3(__p0, __p1) __extension__ ({ \
62294   int16x8x3_t __s1 = __p1; \
62304 #define vst1_u8_x3(__p0, __p1) __extension__ ({ \
62305   uint8x8x3_t __s1 = __p1; \
62309 #define vst1_u8_x3(__p0, __p1) __extension__ ({ \
62310   uint8x8x3_t __s1 = __p1; \
62320 #define vst1_u32_x3(__p0, __p1) __extension__ ({ \
62321   uint32x2x3_t __s1 = __p1; \
62325 #define vst1_u32_x3(__p0, __p1) __extension__ ({ \
62326   uint32x2x3_t __s1 = __p1; \
62336 #define vst1_u64_x3(__p0, __p1) __extension__ ({ \
62337   uint64x1x3_t __s1 = __p1; \
62341 #define vst1_u64_x3(__p0, __p1) __extension__ ({ \
62342   uint64x1x3_t __s1 = __p1; \
62348 #define vst1_u16_x3(__p0, __p1) __extension__ ({ \
62349   uint16x4x3_t __s1 = __p1; \
62353 #define vst1_u16_x3(__p0, __p1) __extension__ ({ \
62354   uint16x4x3_t __s1 = __p1; \
62364 #define vst1_s8_x3(__p0, __p1) __extension__ ({ \
62365   int8x8x3_t __s1 = __p1; \
62369 #define vst1_s8_x3(__p0, __p1) __extension__ ({ \
62370   int8x8x3_t __s1 = __p1; \
62380 #define vst1_f64_x3(__p0, __p1) __extension__ ({ \
62381   float64x1x3_t __s1 = __p1; \
62385 #define vst1_f64_x3(__p0, __p1) __extension__ ({ \
62386   float64x1x3_t __s1 = __p1; \
62392 #define vst1_f32_x3(__p0, __p1) __extension__ ({ \
62393   float32x2x3_t __s1 = __p1; \
62397 #define vst1_f32_x3(__p0, __p1) __extension__ ({ \
62398   float32x2x3_t __s1 = __p1; \
62408 #define vst1_f16_x3(__p0, __p1) __extension__ ({ \
62409   float16x4x3_t __s1 = __p1; \
62413 #define vst1_f16_x3(__p0, __p1) __extension__ ({ \
62414   float16x4x3_t __s1 = __p1; \
62424 #define vst1_s32_x3(__p0, __p1) __extension__ ({ \
62425   int32x2x3_t __s1 = __p1; \
62429 #define vst1_s32_x3(__p0, __p1) __extension__ ({ \
62430   int32x2x3_t __s1 = __p1; \
62440 #define vst1_s64_x3(__p0, __p1) __extension__ ({ \
62441   int64x1x3_t __s1 = __p1; \
62445 #define vst1_s64_x3(__p0, __p1) __extension__ ({ \
62446   int64x1x3_t __s1 = __p1; \
62452 #define vst1_s16_x3(__p0, __p1) __extension__ ({ \
62453   int16x4x3_t __s1 = __p1; \
62457 #define vst1_s16_x3(__p0, __p1) __extension__ ({ \
62458   int16x4x3_t __s1 = __p1; \
62468 #define vst1_p8_x4(__p0, __p1) __extension__ ({ \
62469   poly8x8x4_t __s1 = __p1; \
62473 #define vst1_p8_x4(__p0, __p1) __extension__ ({ \
62474   poly8x8x4_t __s1 = __p1; \
62485 #define vst1_p64_x4(__p0, __p1) __extension__ ({ \
62486   poly64x1x4_t __s1 = __p1; \
62490 #define vst1_p64_x4(__p0, __p1) __extension__ ({ \
62491   poly64x1x4_t __s1 = __p1; \
62497 #define vst1_p16_x4(__p0, __p1) __extension__ ({ \
62498   poly16x4x4_t __s1 = __p1; \
62502 #define vst1_p16_x4(__p0, __p1) __extension__ ({ \
62503   poly16x4x4_t __s1 = __p1; \
62514 #define vst1q_p8_x4(__p0, __p1) __extension__ ({ \
62515   poly8x16x4_t __s1 = __p1; \
62519 #define vst1q_p8_x4(__p0, __p1) __extension__ ({ \
62520   poly8x16x4_t __s1 = __p1; \
62531 #define vst1q_p64_x4(__p0, __p1) __extension__ ({ \
62532   poly64x2x4_t __s1 = __p1; \
62536 #define vst1q_p64_x4(__p0, __p1) __extension__ ({ \
62537   poly64x2x4_t __s1 = __p1; \
62548 #define vst1q_p16_x4(__p0, __p1) __extension__ ({ \
62549   poly16x8x4_t __s1 = __p1; \
62553 #define vst1q_p16_x4(__p0, __p1) __extension__ ({ \
62554   poly16x8x4_t __s1 = __p1; \
62565 #define vst1q_u8_x4(__p0, __p1) __extension__ ({ \
62566   uint8x16x4_t __s1 = __p1; \
62570 #define vst1q_u8_x4(__p0, __p1) __extension__ ({ \
62571   uint8x16x4_t __s1 = __p1; \
62582 #define vst1q_u32_x4(__p0, __p1) __extension__ ({ \
62583   uint32x4x4_t __s1 = __p1; \
62587 #define vst1q_u32_x4(__p0, __p1) __extension__ ({ \
62588   uint32x4x4_t __s1 = __p1; \
62599 #define vst1q_u64_x4(__p0, __p1) __extension__ ({ \
62600   uint64x2x4_t __s1 = __p1; \
62604 #define vst1q_u64_x4(__p0, __p1) __extension__ ({ \
62605   uint64x2x4_t __s1 = __p1; \
62616 #define vst1q_u16_x4(__p0, __p1) __extension__ ({ \
62617   uint16x8x4_t __s1 = __p1; \
62621 #define vst1q_u16_x4(__p0, __p1) __extension__ ({ \
62622   uint16x8x4_t __s1 = __p1; \
62633 #define vst1q_s8_x4(__p0, __p1) __extension__ ({ \
62634   int8x16x4_t __s1 = __p1; \
62638 #define vst1q_s8_x4(__p0, __p1) __extension__ ({ \
62639   int8x16x4_t __s1 = __p1; \
62650 #define vst1q_f64_x4(__p0, __p1) __extension__ ({ \
62651   float64x2x4_t __s1 = __p1; \
62655 #define vst1q_f64_x4(__p0, __p1) __extension__ ({ \
62656   float64x2x4_t __s1 = __p1; \
62667 #define vst1q_f32_x4(__p0, __p1) __extension__ ({ \
62668   float32x4x4_t __s1 = __p1; \
62672 #define vst1q_f32_x4(__p0, __p1) __extension__ ({ \
62673   float32x4x4_t __s1 = __p1; \
62684 #define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
62685   float16x8x4_t __s1 = __p1; \
62689 #define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
62690   float16x8x4_t __s1 = __p1; \
62701 #define vst1q_s32_x4(__p0, __p1) __extension__ ({ \
62702   int32x4x4_t __s1 = __p1; \
62706 #define vst1q_s32_x4(__p0, __p1) __extension__ ({ \
62707   int32x4x4_t __s1 = __p1; \
62718 #define vst1q_s64_x4(__p0, __p1) __extension__ ({ \
62719   int64x2x4_t __s1 = __p1; \
62723 #define vst1q_s64_x4(__p0, __p1) __extension__ ({ \
62724   int64x2x4_t __s1 = __p1; \
62735 #define vst1q_s16_x4(__p0, __p1) __extension__ ({ \
62736   int16x8x4_t __s1 = __p1; \
62740 #define vst1q_s16_x4(__p0, __p1) __extension__ ({ \
62741   int16x8x4_t __s1 = __p1; \
62752 #define vst1_u8_x4(__p0, __p1) __extension__ ({ \
62753   uint8x8x4_t __s1 = __p1; \
62757 #define vst1_u8_x4(__p0, __p1) __extension__ ({ \
62758   uint8x8x4_t __s1 = __p1; \
62769 #define vst1_u32_x4(__p0, __p1) __extension__ ({ \
62770   uint32x2x4_t __s1 = __p1; \
62774 #define vst1_u32_x4(__p0, __p1) __extension__ ({ \
62775   uint32x2x4_t __s1 = __p1; \
62786 #define vst1_u64_x4(__p0, __p1) __extension__ ({ \
62787   uint64x1x4_t __s1 = __p1; \
62791 #define vst1_u64_x4(__p0, __p1) __extension__ ({ \
62792   uint64x1x4_t __s1 = __p1; \
62798 #define vst1_u16_x4(__p0, __p1) __extension__ ({ \
62799   uint16x4x4_t __s1 = __p1; \
62803 #define vst1_u16_x4(__p0, __p1) __extension__ ({ \
62804   uint16x4x4_t __s1 = __p1; \
62815 #define vst1_s8_x4(__p0, __p1) __extension__ ({ \
62816   int8x8x4_t __s1 = __p1; \
62820 #define vst1_s8_x4(__p0, __p1) __extension__ ({ \
62821   int8x8x4_t __s1 = __p1; \
62832 #define vst1_f64_x4(__p0, __p1) __extension__ ({ \
62833   float64x1x4_t __s1 = __p1; \
62837 #define vst1_f64_x4(__p0, __p1) __extension__ ({ \
62838   float64x1x4_t __s1 = __p1; \
62844 #define vst1_f32_x4(__p0, __p1) __extension__ ({ \
62845   float32x2x4_t __s1 = __p1; \
62849 #define vst1_f32_x4(__p0, __p1) __extension__ ({ \
62850   float32x2x4_t __s1 = __p1; \
62861 #define vst1_f16_x4(__p0, __p1) __extension__ ({ \
62862   float16x4x4_t __s1 = __p1; \
62866 #define vst1_f16_x4(__p0, __p1) __extension__ ({ \
62867   float16x4x4_t __s1 = __p1; \
62878 #define vst1_s32_x4(__p0, __p1) __extension__ ({ \
62879   int32x2x4_t __s1 = __p1; \
62883 #define vst1_s32_x4(__p0, __p1) __extension__ ({ \
62884   int32x2x4_t __s1 = __p1; \
62895 #define vst1_s64_x4(__p0, __p1) __extension__ ({ \
62896   int64x1x4_t __s1 = __p1; \
62900 #define vst1_s64_x4(__p0, __p1) __extension__ ({ \
62901   int64x1x4_t __s1 = __p1; \
62907 #define vst1_s16_x4(__p0, __p1) __extension__ ({ \
62908   int16x4x4_t __s1 = __p1; \
62912 #define vst1_s16_x4(__p0, __p1) __extension__ ({ \
62913   int16x4x4_t __s1 = __p1; \
62924 #define vst2_p64(__p0, __p1) __extension__ ({ \
62925   poly64x1x2_t __s1 = __p1; \
62929 #define vst2_p64(__p0, __p1) __extension__ ({ \
62930   poly64x1x2_t __s1 = __p1; \
62936 #define vst2q_p64(__p0, __p1) __extension__ ({ \
62937   poly64x2x2_t __s1 = __p1; \
62941 #define vst2q_p64(__p0, __p1) __extension__ ({ \
62942   poly64x2x2_t __s1 = __p1; \
62951 #define vst2q_u64(__p0, __p1) __extension__ ({ \
62952   uint64x2x2_t __s1 = __p1; \
62956 #define vst2q_u64(__p0, __p1) __extension__ ({ \
62957   uint64x2x2_t __s1 = __p1; \
62966 #define vst2q_f64(__p0, __p1) __extension__ ({ \
62967   float64x2x2_t __s1 = __p1; \
62971 #define vst2q_f64(__p0, __p1) __extension__ ({ \
62972   float64x2x2_t __s1 = __p1; \
62981 #define vst2q_s64(__p0, __p1) __extension__ ({ \
62982   int64x2x2_t __s1 = __p1; \
62986 #define vst2q_s64(__p0, __p1) __extension__ ({ \
62987   int64x2x2_t __s1 = __p1; \
62996 #define vst2_f64(__p0, __p1) __extension__ ({ \
62997   float64x1x2_t __s1 = __p1; \
63001 #define vst2_f64(__p0, __p1) __extension__ ({ \
63002   float64x1x2_t __s1 = __p1; \
63008 #define vst2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63009   poly64x1x2_t __s1 = __p1; \
63013 #define vst2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63014   poly64x1x2_t __s1 = __p1; \
63020 #define vst2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63021   poly8x16x2_t __s1 = __p1; \
63025 #define vst2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63026   poly8x16x2_t __s1 = __p1; \
63035 #define vst2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63036   poly64x2x2_t __s1 = __p1; \
63040 #define vst2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63041   poly64x2x2_t __s1 = __p1; \
63050 #define vst2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63051   uint8x16x2_t __s1 = __p1; \
63055 #define vst2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63056   uint8x16x2_t __s1 = __p1; \
63065 #define vst2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63066   uint64x2x2_t __s1 = __p1; \
63070 #define vst2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63071   uint64x2x2_t __s1 = __p1; \
63080 #define vst2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63081   int8x16x2_t __s1 = __p1; \
63085 #define vst2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63086   int8x16x2_t __s1 = __p1; \
63095 #define vst2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63096   float64x2x2_t __s1 = __p1; \
63100 #define vst2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63101   float64x2x2_t __s1 = __p1; \
63110 #define vst2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63111   int64x2x2_t __s1 = __p1; \
63115 #define vst2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63116   int64x2x2_t __s1 = __p1; \
63125 #define vst2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63126   uint64x1x2_t __s1 = __p1; \
63130 #define vst2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63131   uint64x1x2_t __s1 = __p1; \
63137 #define vst2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63138   float64x1x2_t __s1 = __p1; \
63142 #define vst2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63143   float64x1x2_t __s1 = __p1; \
63149 #define vst2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63150   int64x1x2_t __s1 = __p1; \
63154 #define vst2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63155   int64x1x2_t __s1 = __p1; \
63161 #define vst3_p64(__p0, __p1) __extension__ ({ \
63162   poly64x1x3_t __s1 = __p1; \
63166 #define vst3_p64(__p0, __p1) __extension__ ({ \
63167   poly64x1x3_t __s1 = __p1; \
63173 #define vst3q_p64(__p0, __p1) __extension__ ({ \
63174   poly64x2x3_t __s1 = __p1; \
63178 #define vst3q_p64(__p0, __p1) __extension__ ({ \
63179   poly64x2x3_t __s1 = __p1; \
63189 #define vst3q_u64(__p0, __p1) __extension__ ({ \
63190   uint64x2x3_t __s1 = __p1; \
63194 #define vst3q_u64(__p0, __p1) __extension__ ({ \
63195   uint64x2x3_t __s1 = __p1; \
63205 #define vst3q_f64(__p0, __p1) __extension__ ({ \
63206   float64x2x3_t __s1 = __p1; \
63210 #define vst3q_f64(__p0, __p1) __extension__ ({ \
63211   float64x2x3_t __s1 = __p1; \
63221 #define vst3q_s64(__p0, __p1) __extension__ ({ \
63222   int64x2x3_t __s1 = __p1; \
63226 #define vst3q_s64(__p0, __p1) __extension__ ({ \
63227   int64x2x3_t __s1 = __p1; \
63237 #define vst3_f64(__p0, __p1) __extension__ ({ \
63238   float64x1x3_t __s1 = __p1; \
63242 #define vst3_f64(__p0, __p1) __extension__ ({ \
63243   float64x1x3_t __s1 = __p1; \
63249 #define vst3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63250   poly64x1x3_t __s1 = __p1; \
63254 #define vst3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63255   poly64x1x3_t __s1 = __p1; \
63261 #define vst3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63262   poly8x16x3_t __s1 = __p1; \
63266 #define vst3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63267   poly8x16x3_t __s1 = __p1; \
63277 #define vst3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63278   poly64x2x3_t __s1 = __p1; \
63282 #define vst3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63283   poly64x2x3_t __s1 = __p1; \
63293 #define vst3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63294   uint8x16x3_t __s1 = __p1; \
63298 #define vst3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63299   uint8x16x3_t __s1 = __p1; \
63309 #define vst3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63310   uint64x2x3_t __s1 = __p1; \
63314 #define vst3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63315   uint64x2x3_t __s1 = __p1; \
63325 #define vst3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63326   int8x16x3_t __s1 = __p1; \
63330 #define vst3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63331   int8x16x3_t __s1 = __p1; \
63341 #define vst3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63342   float64x2x3_t __s1 = __p1; \
63346 #define vst3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63347   float64x2x3_t __s1 = __p1; \
63357 #define vst3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63358   int64x2x3_t __s1 = __p1; \
63362 #define vst3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63363   int64x2x3_t __s1 = __p1; \
63373 #define vst3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63374   uint64x1x3_t __s1 = __p1; \
63378 #define vst3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63379   uint64x1x3_t __s1 = __p1; \
63385 #define vst3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63386   float64x1x3_t __s1 = __p1; \
63390 #define vst3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63391   float64x1x3_t __s1 = __p1; \
63397 #define vst3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63398   int64x1x3_t __s1 = __p1; \
63402 #define vst3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63403   int64x1x3_t __s1 = __p1; \
63409 #define vst4_p64(__p0, __p1) __extension__ ({ \
63410   poly64x1x4_t __s1 = __p1; \
63414 #define vst4_p64(__p0, __p1) __extension__ ({ \
63415   poly64x1x4_t __s1 = __p1; \
63421 #define vst4q_p64(__p0, __p1) __extension__ ({ \
63422   poly64x2x4_t __s1 = __p1; \
63426 #define vst4q_p64(__p0, __p1) __extension__ ({ \
63427   poly64x2x4_t __s1 = __p1; \
63438 #define vst4q_u64(__p0, __p1) __extension__ ({ \
63439   uint64x2x4_t __s1 = __p1; \
63443 #define vst4q_u64(__p0, __p1) __extension__ ({ \
63444   uint64x2x4_t __s1 = __p1; \
63455 #define vst4q_f64(__p0, __p1) __extension__ ({ \
63456   float64x2x4_t __s1 = __p1; \
63460 #define vst4q_f64(__p0, __p1) __extension__ ({ \
63461   float64x2x4_t __s1 = __p1; \
63472 #define vst4q_s64(__p0, __p1) __extension__ ({ \
63473   int64x2x4_t __s1 = __p1; \
63477 #define vst4q_s64(__p0, __p1) __extension__ ({ \
63478   int64x2x4_t __s1 = __p1; \
63489 #define vst4_f64(__p0, __p1) __extension__ ({ \
63490   float64x1x4_t __s1 = __p1; \
63494 #define vst4_f64(__p0, __p1) __extension__ ({ \
63495   float64x1x4_t __s1 = __p1; \
63501 #define vst4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63502   poly64x1x4_t __s1 = __p1; \
63506 #define vst4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63507   poly64x1x4_t __s1 = __p1; \
63513 #define vst4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63514   poly8x16x4_t __s1 = __p1; \
63518 #define vst4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
63519   poly8x16x4_t __s1 = __p1; \
63530 #define vst4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63531   poly64x2x4_t __s1 = __p1; \
63535 #define vst4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63536   poly64x2x4_t __s1 = __p1; \
63547 #define vst4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63548   uint8x16x4_t __s1 = __p1; \
63552 #define vst4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63553   uint8x16x4_t __s1 = __p1; \
63564 #define vst4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63565   uint64x2x4_t __s1 = __p1; \
63569 #define vst4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63570   uint64x2x4_t __s1 = __p1; \
63581 #define vst4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63582   int8x16x4_t __s1 = __p1; \
63586 #define vst4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63587   int8x16x4_t __s1 = __p1; \
63598 #define vst4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63599   float64x2x4_t __s1 = __p1; \
63603 #define vst4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63604   float64x2x4_t __s1 = __p1; \
63615 #define vst4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63616   int64x2x4_t __s1 = __p1; \
63620 #define vst4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63621   int64x2x4_t __s1 = __p1; \
63632 #define vst4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63633   uint64x1x4_t __s1 = __p1; \
63637 #define vst4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63638   uint64x1x4_t __s1 = __p1; \
63644 #define vst4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63645   float64x1x4_t __s1 = __p1; \
63649 #define vst4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63650   float64x1x4_t __s1 = __p1; \
63656 #define vst4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63657   int64x1x4_t __s1 = __p1; \
63661 #define vst4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63662   int64x1x4_t __s1 = __p1; \
63668 #define vstrq_p128(__p0, __p1) __extension__ ({ \
63669   poly128_t __s1 = __p1; \
63673 #define vstrq_p128(__p0, __p1) __extension__ ({ \
63674   poly128_t __s1 = __p1; \
63680 __ai uint64_t vsubd_u64(uint64_t __p0, uint64_t __p1) {
63682   __ret = (uint64_t) __builtin_neon_vsubd_u64(__p0, __p1);
63686 __ai uint64_t vsubd_u64(uint64_t __p0, uint64_t __p1) {
63688   __ret = (uint64_t) __builtin_neon_vsubd_u64(__p0, __p1);
63694 __ai int64_t vsubd_s64(int64_t __p0, int64_t __p1) {
63696   __ret = (int64_t) __builtin_neon_vsubd_s64(__p0, __p1);
63700 __ai int64_t vsubd_s64(int64_t __p0, int64_t __p1) {
63702   __ret = (int64_t) __builtin_neon_vsubd_s64(__p0, __p1);
63708 __ai float64x2_t vsubq_f64(float64x2_t __p0, float64x2_t __p1) {
63710   __ret = __p0 - __p1;
63714 __ai float64x2_t vsubq_f64(float64x2_t __p0, float64x2_t __p1) {
63716   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63725 __ai float64x1_t vsub_f64(float64x1_t __p0, float64x1_t __p1) {
63727   __ret = __p0 - __p1;
63731 __ai float64x1_t vsub_f64(float64x1_t __p0, float64x1_t __p1) {
63733   __ret = __p0 - __p1;
63739 __ai uint16x8_t vsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
63741   __ret = vcombine_u16(__p0, vsubhn_u32(__p1, __p2));
63745 __ai uint16x8_t vsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
63747   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63757 __ai uint32x4_t vsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
63759   __ret = vcombine_u32(__p0, vsubhn_u64(__p1, __p2));
63763 __ai uint32x4_t vsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
63765   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63775 __ai uint8x16_t vsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
63777   __ret = vcombine_u8(__p0, vsubhn_u16(__p1, __p2));
63781 __ai uint8x16_t vsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
63783   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63793 __ai int16x8_t vsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
63795   __ret = vcombine_s16(__p0, vsubhn_s32(__p1, __p2));
63799 __ai int16x8_t vsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
63801   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63811 __ai int32x4_t vsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
63813   __ret = vcombine_s32(__p0, vsubhn_s64(__p1, __p2));
63817 __ai int32x4_t vsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
63819   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63829 __ai int8x16_t vsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
63831   __ret = vcombine_s8(__p0, vsubhn_s16(__p1, __p2));
63835 __ai int8x16_t vsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
63837   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63847 __ai uint16x8_t vsubl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
63849   __ret = vmovl_high_u8(__p0) - vmovl_high_u8(__p1);
63853 __ai uint16x8_t vsubl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
63855   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63864 __ai uint64x2_t vsubl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
63866   __ret = vmovl_high_u32(__p0) - vmovl_high_u32(__p1);
63870 __ai uint64x2_t vsubl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
63872   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63881 __ai uint32x4_t vsubl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
63883   __ret = vmovl_high_u16(__p0) - vmovl_high_u16(__p1);
63887 __ai uint32x4_t vsubl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
63889   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63898 __ai int16x8_t vsubl_high_s8(int8x16_t __p0, int8x16_t __p1) {
63900   __ret = vmovl_high_s8(__p0) - vmovl_high_s8(__p1);
63904 __ai int16x8_t vsubl_high_s8(int8x16_t __p0, int8x16_t __p1) {
63906   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63915 __ai int64x2_t vsubl_high_s32(int32x4_t __p0, int32x4_t __p1) {
63917   __ret = vmovl_high_s32(__p0) - vmovl_high_s32(__p1);
63921 __ai int64x2_t vsubl_high_s32(int32x4_t __p0, int32x4_t __p1) {
63923   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63932 __ai int32x4_t vsubl_high_s16(int16x8_t __p0, int16x8_t __p1) {
63934   __ret = vmovl_high_s16(__p0) - vmovl_high_s16(__p1);
63938 __ai int32x4_t vsubl_high_s16(int16x8_t __p0, int16x8_t __p1) {
63940   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63949 __ai uint16x8_t vsubw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
63951   __ret = __p0 - vmovl_high_u8(__p1);
63955 __ai uint16x8_t vsubw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
63957   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63966 __ai uint64x2_t vsubw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
63968   __ret = __p0 - vmovl_high_u32(__p1);
63972 __ai uint64x2_t vsubw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
63974   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63983 __ai uint32x4_t vsubw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
63985   __ret = __p0 - vmovl_high_u16(__p1);
63989 __ai uint32x4_t vsubw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
63991   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64000 __ai int16x8_t vsubw_high_s8(int16x8_t __p0, int8x16_t __p1) {
64002   __ret = __p0 - vmovl_high_s8(__p1);
64006 __ai int16x8_t vsubw_high_s8(int16x8_t __p0, int8x16_t __p1) {
64008   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64017 __ai int64x2_t vsubw_high_s32(int64x2_t __p0, int32x4_t __p1) {
64019   __ret = __p0 - vmovl_high_s32(__p1);
64023 __ai int64x2_t vsubw_high_s32(int64x2_t __p0, int32x4_t __p1) {
64025   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64034 __ai int32x4_t vsubw_high_s16(int32x4_t __p0, int16x8_t __p1) {
64036   __ret = __p0 - vmovl_high_s16(__p1);
64040 __ai int32x4_t vsubw_high_s16(int32x4_t __p0, int16x8_t __p1) {
64042   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64051 __ai poly8x8_t vtrn1_p8(poly8x8_t __p0, poly8x8_t __p1) {
64053   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64057 __ai poly8x8_t vtrn1_p8(poly8x8_t __p0, poly8x8_t __p1) {
64059   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64068 __ai poly16x4_t vtrn1_p16(poly16x4_t __p0, poly16x4_t __p1) {
64070   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64074 __ai poly16x4_t vtrn1_p16(poly16x4_t __p0, poly16x4_t __p1) {
64076   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64085 __ai poly8x16_t vtrn1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64087   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
64091 __ai poly8x16_t vtrn1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64093   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64102 __ai poly64x2_t vtrn1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64104   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64108 __ai poly64x2_t vtrn1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64110   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64119 __ai poly16x8_t vtrn1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64121   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64125 __ai poly16x8_t vtrn1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64127   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64136 __ai uint8x16_t vtrn1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64138   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
64142 __ai uint8x16_t vtrn1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64144   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64153 __ai uint32x4_t vtrn1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64155   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64159 __ai uint32x4_t vtrn1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64161   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64170 __ai uint64x2_t vtrn1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64172   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64176 __ai uint64x2_t vtrn1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64178   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64187 __ai uint16x8_t vtrn1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64189   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64193 __ai uint16x8_t vtrn1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64195   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64204 __ai int8x16_t vtrn1q_s8(int8x16_t __p0, int8x16_t __p1) {
64206   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
64210 __ai int8x16_t vtrn1q_s8(int8x16_t __p0, int8x16_t __p1) {
64212   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64221 __ai float64x2_t vtrn1q_f64(float64x2_t __p0, float64x2_t __p1) {
64223   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64227 __ai float64x2_t vtrn1q_f64(float64x2_t __p0, float64x2_t __p1) {
64229   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64238 __ai float32x4_t vtrn1q_f32(float32x4_t __p0, float32x4_t __p1) {
64240   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64244 __ai float32x4_t vtrn1q_f32(float32x4_t __p0, float32x4_t __p1) {
64246   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64255 __ai int32x4_t vtrn1q_s32(int32x4_t __p0, int32x4_t __p1) {
64257   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64261 __ai int32x4_t vtrn1q_s32(int32x4_t __p0, int32x4_t __p1) {
64263   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64272 __ai int64x2_t vtrn1q_s64(int64x2_t __p0, int64x2_t __p1) {
64274   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64278 __ai int64x2_t vtrn1q_s64(int64x2_t __p0, int64x2_t __p1) {
64280   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64289 __ai int16x8_t vtrn1q_s16(int16x8_t __p0, int16x8_t __p1) {
64291   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64295 __ai int16x8_t vtrn1q_s16(int16x8_t __p0, int16x8_t __p1) {
64297   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64306 __ai uint8x8_t vtrn1_u8(uint8x8_t __p0, uint8x8_t __p1) {
64308   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64312 __ai uint8x8_t vtrn1_u8(uint8x8_t __p0, uint8x8_t __p1) {
64314   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64323 __ai uint32x2_t vtrn1_u32(uint32x2_t __p0, uint32x2_t __p1) {
64325   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64329 __ai uint32x2_t vtrn1_u32(uint32x2_t __p0, uint32x2_t __p1) {
64331   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64340 __ai uint16x4_t vtrn1_u16(uint16x4_t __p0, uint16x4_t __p1) {
64342   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64346 __ai uint16x4_t vtrn1_u16(uint16x4_t __p0, uint16x4_t __p1) {
64348   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64357 __ai int8x8_t vtrn1_s8(int8x8_t __p0, int8x8_t __p1) {
64359   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
64363 __ai int8x8_t vtrn1_s8(int8x8_t __p0, int8x8_t __p1) {
64365   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64374 __ai float32x2_t vtrn1_f32(float32x2_t __p0, float32x2_t __p1) {
64376   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64380 __ai float32x2_t vtrn1_f32(float32x2_t __p0, float32x2_t __p1) {
64382   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64391 __ai int32x2_t vtrn1_s32(int32x2_t __p0, int32x2_t __p1) {
64393   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64397 __ai int32x2_t vtrn1_s32(int32x2_t __p0, int32x2_t __p1) {
64399   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64408 __ai int16x4_t vtrn1_s16(int16x4_t __p0, int16x4_t __p1) {
64410   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
64414 __ai int16x4_t vtrn1_s16(int16x4_t __p0, int16x4_t __p1) {
64416   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64425 __ai poly8x8_t vtrn2_p8(poly8x8_t __p0, poly8x8_t __p1) {
64427   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64431 __ai poly8x8_t vtrn2_p8(poly8x8_t __p0, poly8x8_t __p1) {
64433   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64442 __ai poly16x4_t vtrn2_p16(poly16x4_t __p0, poly16x4_t __p1) {
64444   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64448 __ai poly16x4_t vtrn2_p16(poly16x4_t __p0, poly16x4_t __p1) {
64450   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64459 __ai poly8x16_t vtrn2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64461   __ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
64465 __ai poly8x16_t vtrn2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64467   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64476 __ai poly64x2_t vtrn2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64478   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64482 __ai poly64x2_t vtrn2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64484   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64493 __ai poly16x8_t vtrn2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64495   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64499 __ai poly16x8_t vtrn2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64501   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64510 __ai uint8x16_t vtrn2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64512   __ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
64516 __ai uint8x16_t vtrn2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64518   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64527 __ai uint32x4_t vtrn2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64529   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64533 __ai uint32x4_t vtrn2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64535   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64544 __ai uint64x2_t vtrn2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64546   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64550 __ai uint64x2_t vtrn2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64552   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64561 __ai uint16x8_t vtrn2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64563   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64567 __ai uint16x8_t vtrn2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64569   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64578 __ai int8x16_t vtrn2q_s8(int8x16_t __p0, int8x16_t __p1) {
64580   __ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
64584 __ai int8x16_t vtrn2q_s8(int8x16_t __p0, int8x16_t __p1) {
64586   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64595 __ai float64x2_t vtrn2q_f64(float64x2_t __p0, float64x2_t __p1) {
64597   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64601 __ai float64x2_t vtrn2q_f64(float64x2_t __p0, float64x2_t __p1) {
64603   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64612 __ai float32x4_t vtrn2q_f32(float32x4_t __p0, float32x4_t __p1) {
64614   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64618 __ai float32x4_t vtrn2q_f32(float32x4_t __p0, float32x4_t __p1) {
64620   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64629 __ai int32x4_t vtrn2q_s32(int32x4_t __p0, int32x4_t __p1) {
64631   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64635 __ai int32x4_t vtrn2q_s32(int32x4_t __p0, int32x4_t __p1) {
64637   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64646 __ai int64x2_t vtrn2q_s64(int64x2_t __p0, int64x2_t __p1) {
64648   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64652 __ai int64x2_t vtrn2q_s64(int64x2_t __p0, int64x2_t __p1) {
64654   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64663 __ai int16x8_t vtrn2q_s16(int16x8_t __p0, int16x8_t __p1) {
64665   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64669 __ai int16x8_t vtrn2q_s16(int16x8_t __p0, int16x8_t __p1) {
64671   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64680 __ai uint8x8_t vtrn2_u8(uint8x8_t __p0, uint8x8_t __p1) {
64682   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64686 __ai uint8x8_t vtrn2_u8(uint8x8_t __p0, uint8x8_t __p1) {
64688   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64697 __ai uint32x2_t vtrn2_u32(uint32x2_t __p0, uint32x2_t __p1) {
64699   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64703 __ai uint32x2_t vtrn2_u32(uint32x2_t __p0, uint32x2_t __p1) {
64705   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64714 __ai uint16x4_t vtrn2_u16(uint16x4_t __p0, uint16x4_t __p1) {
64716   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64720 __ai uint16x4_t vtrn2_u16(uint16x4_t __p0, uint16x4_t __p1) {
64722   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64731 __ai int8x8_t vtrn2_s8(int8x8_t __p0, int8x8_t __p1) {
64733   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64737 __ai int8x8_t vtrn2_s8(int8x8_t __p0, int8x8_t __p1) {
64739   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64748 __ai float32x2_t vtrn2_f32(float32x2_t __p0, float32x2_t __p1) {
64750   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64754 __ai float32x2_t vtrn2_f32(float32x2_t __p0, float32x2_t __p1) {
64756   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64765 __ai int32x2_t vtrn2_s32(int32x2_t __p0, int32x2_t __p1) {
64767   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64771 __ai int32x2_t vtrn2_s32(int32x2_t __p0, int32x2_t __p1) {
64773   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64782 __ai int16x4_t vtrn2_s16(int16x4_t __p0, int16x4_t __p1) {
64784   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64788 __ai int16x4_t vtrn2_s16(int16x4_t __p0, int16x4_t __p1) {
64790   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64799 __ai uint64x1_t vtst_p64(poly64x1_t __p0, poly64x1_t __p1) {
64801   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64805 __ai uint64x1_t vtst_p64(poly64x1_t __p0, poly64x1_t __p1) {
64807   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64813 __ai uint64x2_t vtstq_p64(poly64x2_t __p0, poly64x2_t __p1) {
64815   __ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
64819 __ai uint64x2_t vtstq_p64(poly64x2_t __p0, poly64x2_t __p1) {
64821   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64830 __ai uint64x2_t vtstq_u64(uint64x2_t __p0, uint64x2_t __p1) {
64832   __ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
64836 __ai uint64x2_t vtstq_u64(uint64x2_t __p0, uint64x2_t __p1) {
64838   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64847 __ai uint64x2_t vtstq_s64(int64x2_t __p0, int64x2_t __p1) {
64849   __ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
64853 __ai uint64x2_t vtstq_s64(int64x2_t __p0, int64x2_t __p1) {
64855   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64864 __ai uint64x1_t vtst_u64(uint64x1_t __p0, uint64x1_t __p1) {
64866   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64870 __ai uint64x1_t vtst_u64(uint64x1_t __p0, uint64x1_t __p1) {
64872   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64878 __ai uint64x1_t vtst_s64(int64x1_t __p0, int64x1_t __p1) {
64880   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64884 __ai uint64x1_t vtst_s64(int64x1_t __p0, int64x1_t __p1) {
64886   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64892 __ai uint64_t vtstd_u64(uint64_t __p0, uint64_t __p1) {
64894   __ret = (uint64_t) __builtin_neon_vtstd_u64(__p0, __p1);
64898 __ai uint64_t vtstd_u64(uint64_t __p0, uint64_t __p1) {
64900   __ret = (uint64_t) __builtin_neon_vtstd_u64(__p0, __p1);
64906 __ai int64_t vtstd_s64(int64_t __p0, int64_t __p1) {
64908   __ret = (int64_t) __builtin_neon_vtstd_s64(__p0, __p1);
64912 __ai int64_t vtstd_s64(int64_t __p0, int64_t __p1) {
64914   __ret = (int64_t) __builtin_neon_vtstd_s64(__p0, __p1);
64920 __ai int8_t vuqaddb_s8(int8_t __p0, int8_t __p1) {
64922   __ret = (int8_t) __builtin_neon_vuqaddb_s8(__p0, __p1);
64926 __ai int8_t vuqaddb_s8(int8_t __p0, int8_t __p1) {
64928   __ret = (int8_t) __builtin_neon_vuqaddb_s8(__p0, __p1);
64934 __ai int32_t vuqadds_s32(int32_t __p0, int32_t __p1) {
64936   __ret = (int32_t) __builtin_neon_vuqadds_s32(__p0, __p1);
64940 __ai int32_t vuqadds_s32(int32_t __p0, int32_t __p1) {
64942   __ret = (int32_t) __builtin_neon_vuqadds_s32(__p0, __p1);
64948 __ai int64_t vuqaddd_s64(int64_t __p0, int64_t __p1) {
64950   __ret = (int64_t) __builtin_neon_vuqaddd_s64(__p0, __p1);
64954 __ai int64_t vuqaddd_s64(int64_t __p0, int64_t __p1) {
64956   __ret = (int64_t) __builtin_neon_vuqaddd_s64(__p0, __p1);
64962 __ai int16_t vuqaddh_s16(int16_t __p0, int16_t __p1) {
64964   __ret = (int16_t) __builtin_neon_vuqaddh_s16(__p0, __p1);
64968 __ai int16_t vuqaddh_s16(int16_t __p0, int16_t __p1) {
64970   __ret = (int16_t) __builtin_neon_vuqaddh_s16(__p0, __p1);
64976 __ai int8x16_t vuqaddq_s8(int8x16_t __p0, int8x16_t __p1) {
64978   __ret = (int8x16_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
64982 __ai int8x16_t vuqaddq_s8(int8x16_t __p0, int8x16_t __p1) {
64984   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64993 __ai int32x4_t vuqaddq_s32(int32x4_t __p0, int32x4_t __p1) {
64995   __ret = (int32x4_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
64999 __ai int32x4_t vuqaddq_s32(int32x4_t __p0, int32x4_t __p1) {
65001   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65010 __ai int64x2_t vuqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
65012   __ret = (int64x2_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
65016 __ai int64x2_t vuqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
65018   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65027 __ai int16x8_t vuqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
65029   __ret = (int16x8_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
65033 __ai int16x8_t vuqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
65035   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65044 __ai int8x8_t vuqadd_s8(int8x8_t __p0, int8x8_t __p1) {
65046   __ret = (int8x8_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
65050 __ai int8x8_t vuqadd_s8(int8x8_t __p0, int8x8_t __p1) {
65052   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65061 __ai int32x2_t vuqadd_s32(int32x2_t __p0, int32x2_t __p1) {
65063   __ret = (int32x2_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
65067 __ai int32x2_t vuqadd_s32(int32x2_t __p0, int32x2_t __p1) {
65069   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65078 __ai int64x1_t vuqadd_s64(int64x1_t __p0, int64x1_t __p1) {
65080   __ret = (int64x1_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
65084 __ai int64x1_t vuqadd_s64(int64x1_t __p0, int64x1_t __p1) {
65086   __ret = (int64x1_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
65092 __ai int16x4_t vuqadd_s16(int16x4_t __p0, int16x4_t __p1) {
65094   __ret = (int16x4_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
65098 __ai int16x4_t vuqadd_s16(int16x4_t __p0, int16x4_t __p1) {
65100   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65109 __ai poly8x8_t vuzp1_p8(poly8x8_t __p0, poly8x8_t __p1) {
65111   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
65115 __ai poly8x8_t vuzp1_p8(poly8x8_t __p0, poly8x8_t __p1) {
65117   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65126 __ai poly16x4_t vuzp1_p16(poly16x4_t __p0, poly16x4_t __p1) {
65128   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
65132 __ai poly16x4_t vuzp1_p16(poly16x4_t __p0, poly16x4_t __p1) {
65134   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65143 __ai poly8x16_t vuzp1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65145   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
65149 __ai poly8x16_t vuzp1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65151   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65160 __ai poly64x2_t vuzp1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65162   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65166 __ai poly64x2_t vuzp1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65168   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65177 __ai poly16x8_t vuzp1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65179   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
65183 __ai poly16x8_t vuzp1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65185   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65194 __ai uint8x16_t vuzp1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65196   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
65200 __ai uint8x16_t vuzp1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65202   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65211 __ai uint32x4_t vuzp1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65213   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
65217 __ai uint32x4_t vuzp1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65219   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65228 __ai uint64x2_t vuzp1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65230   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65234 __ai uint64x2_t vuzp1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65236   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65245 __ai uint16x8_t vuzp1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65247   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
65251 __ai uint16x8_t vuzp1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65253   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65262 __ai int8x16_t vuzp1q_s8(int8x16_t __p0, int8x16_t __p1) {
65264   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
65268 __ai int8x16_t vuzp1q_s8(int8x16_t __p0, int8x16_t __p1) {
65270   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65279 __ai float64x2_t vuzp1q_f64(float64x2_t __p0, float64x2_t __p1) {
65281   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65285 __ai float64x2_t vuzp1q_f64(float64x2_t __p0, float64x2_t __p1) {
65287   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65296 __ai float32x4_t vuzp1q_f32(float32x4_t __p0, float32x4_t __p1) {
65298   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
65302 __ai float32x4_t vuzp1q_f32(float32x4_t __p0, float32x4_t __p1) {
65304   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65313 __ai int32x4_t vuzp1q_s32(int32x4_t __p0, int32x4_t __p1) {
65315   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
65319 __ai int32x4_t vuzp1q_s32(int32x4_t __p0, int32x4_t __p1) {
65321   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65330 __ai int64x2_t vuzp1q_s64(int64x2_t __p0, int64x2_t __p1) {
65332   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65336 __ai int64x2_t vuzp1q_s64(int64x2_t __p0, int64x2_t __p1) {
65338   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65347 __ai int16x8_t vuzp1q_s16(int16x8_t __p0, int16x8_t __p1) {
65349   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
65353 __ai int16x8_t vuzp1q_s16(int16x8_t __p0, int16x8_t __p1) {
65355   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65364 __ai uint8x8_t vuzp1_u8(uint8x8_t __p0, uint8x8_t __p1) {
65366   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
65370 __ai uint8x8_t vuzp1_u8(uint8x8_t __p0, uint8x8_t __p1) {
65372   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65381 __ai uint32x2_t vuzp1_u32(uint32x2_t __p0, uint32x2_t __p1) {
65383   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65387 __ai uint32x2_t vuzp1_u32(uint32x2_t __p0, uint32x2_t __p1) {
65389   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65398 __ai uint16x4_t vuzp1_u16(uint16x4_t __p0, uint16x4_t __p1) {
65400   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
65404 __ai uint16x4_t vuzp1_u16(uint16x4_t __p0, uint16x4_t __p1) {
65406   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65415 __ai int8x8_t vuzp1_s8(int8x8_t __p0, int8x8_t __p1) {
65417   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
65421 __ai int8x8_t vuzp1_s8(int8x8_t __p0, int8x8_t __p1) {
65423   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65432 __ai float32x2_t vuzp1_f32(float32x2_t __p0, float32x2_t __p1) {
65434   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65438 __ai float32x2_t vuzp1_f32(float32x2_t __p0, float32x2_t __p1) {
65440   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65449 __ai int32x2_t vuzp1_s32(int32x2_t __p0, int32x2_t __p1) {
65451   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65455 __ai int32x2_t vuzp1_s32(int32x2_t __p0, int32x2_t __p1) {
65457   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65466 __ai int16x4_t vuzp1_s16(int16x4_t __p0, int16x4_t __p1) {
65468   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
65472 __ai int16x4_t vuzp1_s16(int16x4_t __p0, int16x4_t __p1) {
65474   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65483 __ai poly8x8_t vuzp2_p8(poly8x8_t __p0, poly8x8_t __p1) {
65485   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65489 __ai poly8x8_t vuzp2_p8(poly8x8_t __p0, poly8x8_t __p1) {
65491   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65500 __ai poly16x4_t vuzp2_p16(poly16x4_t __p0, poly16x4_t __p1) {
65502   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65506 __ai poly16x4_t vuzp2_p16(poly16x4_t __p0, poly16x4_t __p1) {
65508   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65517 __ai poly8x16_t vuzp2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65519   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
65523 __ai poly8x16_t vuzp2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65525   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65534 __ai poly64x2_t vuzp2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65536   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65540 __ai poly64x2_t vuzp2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65542   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65551 __ai poly16x8_t vuzp2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65553   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65557 __ai poly16x8_t vuzp2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65559   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65568 __ai uint8x16_t vuzp2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65570   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
65574 __ai uint8x16_t vuzp2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65576   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65585 __ai uint32x4_t vuzp2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65587   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65591 __ai uint32x4_t vuzp2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65593   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65602 __ai uint64x2_t vuzp2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65604   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65608 __ai uint64x2_t vuzp2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65610   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65619 __ai uint16x8_t vuzp2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65621   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65625 __ai uint16x8_t vuzp2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65627   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65636 __ai int8x16_t vuzp2q_s8(int8x16_t __p0, int8x16_t __p1) {
65638   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
65642 __ai int8x16_t vuzp2q_s8(int8x16_t __p0, int8x16_t __p1) {
65644   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65653 __ai float64x2_t vuzp2q_f64(float64x2_t __p0, float64x2_t __p1) {
65655   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65659 __ai float64x2_t vuzp2q_f64(float64x2_t __p0, float64x2_t __p1) {
65661   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65670 __ai float32x4_t vuzp2q_f32(float32x4_t __p0, float32x4_t __p1) {
65672   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65676 __ai float32x4_t vuzp2q_f32(float32x4_t __p0, float32x4_t __p1) {
65678   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65687 __ai int32x4_t vuzp2q_s32(int32x4_t __p0, int32x4_t __p1) {
65689   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65693 __ai int32x4_t vuzp2q_s32(int32x4_t __p0, int32x4_t __p1) {
65695   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65704 __ai int64x2_t vuzp2q_s64(int64x2_t __p0, int64x2_t __p1) {
65706   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65710 __ai int64x2_t vuzp2q_s64(int64x2_t __p0, int64x2_t __p1) {
65712   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65721 __ai int16x8_t vuzp2q_s16(int16x8_t __p0, int16x8_t __p1) {
65723   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65727 __ai int16x8_t vuzp2q_s16(int16x8_t __p0, int16x8_t __p1) {
65729   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65738 __ai uint8x8_t vuzp2_u8(uint8x8_t __p0, uint8x8_t __p1) {
65740   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65744 __ai uint8x8_t vuzp2_u8(uint8x8_t __p0, uint8x8_t __p1) {
65746   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65755 __ai uint32x2_t vuzp2_u32(uint32x2_t __p0, uint32x2_t __p1) {
65757   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65761 __ai uint32x2_t vuzp2_u32(uint32x2_t __p0, uint32x2_t __p1) {
65763   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65772 __ai uint16x4_t vuzp2_u16(uint16x4_t __p0, uint16x4_t __p1) {
65774   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65778 __ai uint16x4_t vuzp2_u16(uint16x4_t __p0, uint16x4_t __p1) {
65780   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65789 __ai int8x8_t vuzp2_s8(int8x8_t __p0, int8x8_t __p1) {
65791   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65795 __ai int8x8_t vuzp2_s8(int8x8_t __p0, int8x8_t __p1) {
65797   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65806 __ai float32x2_t vuzp2_f32(float32x2_t __p0, float32x2_t __p1) {
65808   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65812 __ai float32x2_t vuzp2_f32(float32x2_t __p0, float32x2_t __p1) {
65814   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65823 __ai int32x2_t vuzp2_s32(int32x2_t __p0, int32x2_t __p1) {
65825   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65829 __ai int32x2_t vuzp2_s32(int32x2_t __p0, int32x2_t __p1) {
65831   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65840 __ai int16x4_t vuzp2_s16(int16x4_t __p0, int16x4_t __p1) {
65842   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65846 __ai int16x4_t vuzp2_s16(int16x4_t __p0, int16x4_t __p1) {
65848   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65857 __ai poly8x8_t vzip1_p8(poly8x8_t __p0, poly8x8_t __p1) {
65859   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
65863 __ai poly8x8_t vzip1_p8(poly8x8_t __p0, poly8x8_t __p1) {
65865   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65874 __ai poly16x4_t vzip1_p16(poly16x4_t __p0, poly16x4_t __p1) {
65876   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
65880 __ai poly16x4_t vzip1_p16(poly16x4_t __p0, poly16x4_t __p1) {
65882   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65891 __ai poly8x16_t vzip1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65893   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
65897 __ai poly8x16_t vzip1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65899   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65908 __ai poly64x2_t vzip1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65910   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65914 __ai poly64x2_t vzip1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65916   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65925 __ai poly16x8_t vzip1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65927   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
65931 __ai poly16x8_t vzip1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65933   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65942 __ai uint8x16_t vzip1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65944   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
65948 __ai uint8x16_t vzip1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65950   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65959 __ai uint32x4_t vzip1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65961   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
65965 __ai uint32x4_t vzip1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65967   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65976 __ai uint64x2_t vzip1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65978   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65982 __ai uint64x2_t vzip1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65984   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65993 __ai uint16x8_t vzip1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65995   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
65999 __ai uint16x8_t vzip1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
66001   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66010 __ai int8x16_t vzip1q_s8(int8x16_t __p0, int8x16_t __p1) {
66012   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
66016 __ai int8x16_t vzip1q_s8(int8x16_t __p0, int8x16_t __p1) {
66018   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66027 __ai float64x2_t vzip1q_f64(float64x2_t __p0, float64x2_t __p1) {
66029   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
66033 __ai float64x2_t vzip1q_f64(float64x2_t __p0, float64x2_t __p1) {
66035   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66044 __ai float32x4_t vzip1q_f32(float32x4_t __p0, float32x4_t __p1) {
66046   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
66050 __ai float32x4_t vzip1q_f32(float32x4_t __p0, float32x4_t __p1) {
66052   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66061 __ai int32x4_t vzip1q_s32(int32x4_t __p0, int32x4_t __p1) {
66063   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
66067 __ai int32x4_t vzip1q_s32(int32x4_t __p0, int32x4_t __p1) {
66069   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66078 __ai int64x2_t vzip1q_s64(int64x2_t __p0, int64x2_t __p1) {
66080   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
66084 __ai int64x2_t vzip1q_s64(int64x2_t __p0, int64x2_t __p1) {
66086   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66095 __ai int16x8_t vzip1q_s16(int16x8_t __p0, int16x8_t __p1) {
66097   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
66101 __ai int16x8_t vzip1q_s16(int16x8_t __p0, int16x8_t __p1) {
66103   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66112 __ai uint8x8_t vzip1_u8(uint8x8_t __p0, uint8x8_t __p1) {
66114   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
66118 __ai uint8x8_t vzip1_u8(uint8x8_t __p0, uint8x8_t __p1) {
66120   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66129 __ai uint32x2_t vzip1_u32(uint32x2_t __p0, uint32x2_t __p1) {
66131   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
66135 __ai uint32x2_t vzip1_u32(uint32x2_t __p0, uint32x2_t __p1) {
66137   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66146 __ai uint16x4_t vzip1_u16(uint16x4_t __p0, uint16x4_t __p1) {
66148   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
66152 __ai uint16x4_t vzip1_u16(uint16x4_t __p0, uint16x4_t __p1) {
66154   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66163 __ai int8x8_t vzip1_s8(int8x8_t __p0, int8x8_t __p1) {
66165   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
66169 __ai int8x8_t vzip1_s8(int8x8_t __p0, int8x8_t __p1) {
66171   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66180 __ai float32x2_t vzip1_f32(float32x2_t __p0, float32x2_t __p1) {
66182   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
66186 __ai float32x2_t vzip1_f32(float32x2_t __p0, float32x2_t __p1) {
66188   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66197 __ai int32x2_t vzip1_s32(int32x2_t __p0, int32x2_t __p1) {
66199   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
66203 __ai int32x2_t vzip1_s32(int32x2_t __p0, int32x2_t __p1) {
66205   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66214 __ai int16x4_t vzip1_s16(int16x4_t __p0, int16x4_t __p1) {
66216   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
66220 __ai int16x4_t vzip1_s16(int16x4_t __p0, int16x4_t __p1) {
66222   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66231 __ai poly8x8_t vzip2_p8(poly8x8_t __p0, poly8x8_t __p1) {
66233   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
66237 __ai poly8x8_t vzip2_p8(poly8x8_t __p0, poly8x8_t __p1) {
66239   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66248 __ai poly16x4_t vzip2_p16(poly16x4_t __p0, poly16x4_t __p1) {
66250   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
66254 __ai poly16x4_t vzip2_p16(poly16x4_t __p0, poly16x4_t __p1) {
66256   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66265 __ai poly8x16_t vzip2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
66267   __ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
66271 __ai poly8x16_t vzip2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
66273   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66282 __ai poly64x2_t vzip2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
66284   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66288 __ai poly64x2_t vzip2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
66290   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66299 __ai poly16x8_t vzip2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
66301   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
66305 __ai poly16x8_t vzip2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
66307   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66316 __ai uint8x16_t vzip2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
66318   __ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
66322 __ai uint8x16_t vzip2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
66324   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66333 __ai uint32x4_t vzip2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
66335   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
66339 __ai uint32x4_t vzip2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
66341   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66350 __ai uint64x2_t vzip2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
66352   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66356 __ai uint64x2_t vzip2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
66358   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66367 __ai uint16x8_t vzip2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
66369   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
66373 __ai uint16x8_t vzip2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
66375   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66384 __ai int8x16_t vzip2q_s8(int8x16_t __p0, int8x16_t __p1) {
66386   __ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
66390 __ai int8x16_t vzip2q_s8(int8x16_t __p0, int8x16_t __p1) {
66392   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66401 __ai float64x2_t vzip2q_f64(float64x2_t __p0, float64x2_t __p1) {
66403   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66407 __ai float64x2_t vzip2q_f64(float64x2_t __p0, float64x2_t __p1) {
66409   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66418 __ai float32x4_t vzip2q_f32(float32x4_t __p0, float32x4_t __p1) {
66420   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
66424 __ai float32x4_t vzip2q_f32(float32x4_t __p0, float32x4_t __p1) {
66426   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66435 __ai int32x4_t vzip2q_s32(int32x4_t __p0, int32x4_t __p1) {
66437   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
66441 __ai int32x4_t vzip2q_s32(int32x4_t __p0, int32x4_t __p1) {
66443   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66452 __ai int64x2_t vzip2q_s64(int64x2_t __p0, int64x2_t __p1) {
66454   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66458 __ai int64x2_t vzip2q_s64(int64x2_t __p0, int64x2_t __p1) {
66460   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66469 __ai int16x8_t vzip2q_s16(int16x8_t __p0, int16x8_t __p1) {
66471   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
66475 __ai int16x8_t vzip2q_s16(int16x8_t __p0, int16x8_t __p1) {
66477   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66486 __ai uint8x8_t vzip2_u8(uint8x8_t __p0, uint8x8_t __p1) {
66488   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
66492 __ai uint8x8_t vzip2_u8(uint8x8_t __p0, uint8x8_t __p1) {
66494   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66503 __ai uint32x2_t vzip2_u32(uint32x2_t __p0, uint32x2_t __p1) {
66505   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66509 __ai uint32x2_t vzip2_u32(uint32x2_t __p0, uint32x2_t __p1) {
66511   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66520 __ai uint16x4_t vzip2_u16(uint16x4_t __p0, uint16x4_t __p1) {
66522   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
66526 __ai uint16x4_t vzip2_u16(uint16x4_t __p0, uint16x4_t __p1) {
66528   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66537 __ai int8x8_t vzip2_s8(int8x8_t __p0, int8x8_t __p1) {
66539   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
66543 __ai int8x8_t vzip2_s8(int8x8_t __p0, int8x8_t __p1) {
66545   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66554 __ai float32x2_t vzip2_f32(float32x2_t __p0, float32x2_t __p1) {
66556   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66560 __ai float32x2_t vzip2_f32(float32x2_t __p0, float32x2_t __p1) {
66562   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66571 __ai int32x2_t vzip2_s32(int32x2_t __p0, int32x2_t __p1) {
66573   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66577 __ai int32x2_t vzip2_s32(int32x2_t __p0, int32x2_t __p1) {
66579   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66588 __ai int16x4_t vzip2_s16(int16x4_t __p0, int16x4_t __p1) {
66590   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
66594 __ai int16x4_t vzip2_s16(int16x4_t __p0, int16x4_t __p1) {
66596   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66606 __ai uint8x16_t vabaq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
66608   __ret = __p0 + vabdq_u8(__p1, __p2);
66612 __ai uint8x16_t vabaq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
66614   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66624 __ai uint32x4_t vabaq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
66626   __ret = __p0 + vabdq_u32(__p1, __p2);
66630 __ai uint32x4_t vabaq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
66632   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66642 __ai uint16x8_t vabaq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
66644   __ret = __p0 + vabdq_u16(__p1, __p2);
66648 __ai uint16x8_t vabaq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
66650   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66660 __ai int8x16_t vabaq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
66662   __ret = __p0 + vabdq_s8(__p1, __p2);
66666 __ai int8x16_t vabaq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
66668   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66678 __ai int32x4_t vabaq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
66680   __ret = __p0 + vabdq_s32(__p1, __p2);
66684 __ai int32x4_t vabaq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
66686   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66696 __ai int16x8_t vabaq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
66698   __ret = __p0 + vabdq_s16(__p1, __p2);
66702 __ai int16x8_t vabaq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
66704   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66714 __ai uint8x8_t vaba_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
66716   __ret = __p0 + vabd_u8(__p1, __p2);
66720 __ai uint8x8_t vaba_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
66722   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66732 __ai uint32x2_t vaba_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
66734   __ret = __p0 + vabd_u32(__p1, __p2);
66738 __ai uint32x2_t vaba_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
66740   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66750 __ai uint16x4_t vaba_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
66752   __ret = __p0 + vabd_u16(__p1, __p2);
66756 __ai uint16x4_t vaba_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
66758   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66768 __ai int8x8_t vaba_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
66770   __ret = __p0 + vabd_s8(__p1, __p2);
66774 __ai int8x8_t vaba_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
66776   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66786 __ai int32x2_t vaba_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
66788   __ret = __p0 + vabd_s32(__p1, __p2);
66792 __ai int32x2_t vaba_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
66794   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66804 __ai int16x4_t vaba_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
66806   __ret = __p0 + vabd_s16(__p1, __p2);
66810 __ai int16x4_t vaba_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
66812   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66822 __ai uint16x8_t vabdl_u8(uint8x8_t __p0, uint8x8_t __p1) {
66824   __ret = (uint16x8_t)(vmovl_u8((uint8x8_t)(vabd_u8(__p0, __p1))));
66828 __ai uint16x8_t vabdl_u8(uint8x8_t __p0, uint8x8_t __p1) {
66830   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66836 __ai uint16x8_t __noswap_vabdl_u8(uint8x8_t __p0, uint8x8_t __p1) {
66838   __ret = (uint16x8_t)(__noswap_vmovl_u8((uint8x8_t)(__noswap_vabd_u8(__p0, __p1))));
66844 __ai uint64x2_t vabdl_u32(uint32x2_t __p0, uint32x2_t __p1) {
66846   __ret = (uint64x2_t)(vmovl_u32((uint32x2_t)(vabd_u32(__p0, __p1))));
66850 __ai uint64x2_t vabdl_u32(uint32x2_t __p0, uint32x2_t __p1) {
66852   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66858 __ai uint64x2_t __noswap_vabdl_u32(uint32x2_t __p0, uint32x2_t __p1) {
66860   __ret = (uint64x2_t)(__noswap_vmovl_u32((uint32x2_t)(__noswap_vabd_u32(__p0, __p1))));
66866 __ai uint32x4_t vabdl_u16(uint16x4_t __p0, uint16x4_t __p1) {
66868   __ret = (uint32x4_t)(vmovl_u16((uint16x4_t)(vabd_u16(__p0, __p1))));
66872 __ai uint32x4_t vabdl_u16(uint16x4_t __p0, uint16x4_t __p1) {
66874   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66880 __ai uint32x4_t __noswap_vabdl_u16(uint16x4_t __p0, uint16x4_t __p1) {
66882   __ret = (uint32x4_t)(__noswap_vmovl_u16((uint16x4_t)(__noswap_vabd_u16(__p0, __p1))));
66888 __ai int16x8_t vabdl_s8(int8x8_t __p0, int8x8_t __p1) {
66890   __ret = (int16x8_t)(vmovl_u8((uint8x8_t)(vabd_s8(__p0, __p1))));
66894 __ai int16x8_t vabdl_s8(int8x8_t __p0, int8x8_t __p1) {
66896   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66902 __ai int16x8_t __noswap_vabdl_s8(int8x8_t __p0, int8x8_t __p1) {
66904   __ret = (int16x8_t)(__noswap_vmovl_u8((uint8x8_t)(__noswap_vabd_s8(__p0, __p1))));
66910 __ai int64x2_t vabdl_s32(int32x2_t __p0, int32x2_t __p1) {
66912   __ret = (int64x2_t)(vmovl_u32((uint32x2_t)(vabd_s32(__p0, __p1))));
66916 __ai int64x2_t vabdl_s32(int32x2_t __p0, int32x2_t __p1) {
66918   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66924 __ai int64x2_t __noswap_vabdl_s32(int32x2_t __p0, int32x2_t __p1) {
66926   __ret = (int64x2_t)(__noswap_vmovl_u32((uint32x2_t)(__noswap_vabd_s32(__p0, __p1))));
66932 __ai int32x4_t vabdl_s16(int16x4_t __p0, int16x4_t __p1) {
66934   __ret = (int32x4_t)(vmovl_u16((uint16x4_t)(vabd_s16(__p0, __p1))));
66938 __ai int32x4_t vabdl_s16(int16x4_t __p0, int16x4_t __p1) {
66940   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66946 __ai int32x4_t __noswap_vabdl_s16(int16x4_t __p0, int16x4_t __p1) {
66948   __ret = (int32x4_t)(__noswap_vmovl_u16((uint16x4_t)(__noswap_vabd_s16(__p0, __p1))));
66954 __ai uint16x8_t vaddl_u8(uint8x8_t __p0, uint8x8_t __p1) {
66956   __ret = vmovl_u8(__p0) + vmovl_u8(__p1);
66960 __ai uint16x8_t vaddl_u8(uint8x8_t __p0, uint8x8_t __p1) {
66962   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66971 __ai uint64x2_t vaddl_u32(uint32x2_t __p0, uint32x2_t __p1) {
66973   __ret = vmovl_u32(__p0) + vmovl_u32(__p1);
66977 __ai uint64x2_t vaddl_u32(uint32x2_t __p0, uint32x2_t __p1) {
66979   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66988 __ai uint32x4_t vaddl_u16(uint16x4_t __p0, uint16x4_t __p1) {
66990   __ret = vmovl_u16(__p0) + vmovl_u16(__p1);
66994 __ai uint32x4_t vaddl_u16(uint16x4_t __p0, uint16x4_t __p1) {
66996   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67005 __ai int16x8_t vaddl_s8(int8x8_t __p0, int8x8_t __p1) {
67007   __ret = vmovl_s8(__p0) + vmovl_s8(__p1);
67011 __ai int16x8_t vaddl_s8(int8x8_t __p0, int8x8_t __p1) {
67013   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67022 __ai int64x2_t vaddl_s32(int32x2_t __p0, int32x2_t __p1) {
67024   __ret = vmovl_s32(__p0) + vmovl_s32(__p1);
67028 __ai int64x2_t vaddl_s32(int32x2_t __p0, int32x2_t __p1) {
67030   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67039 __ai int32x4_t vaddl_s16(int16x4_t __p0, int16x4_t __p1) {
67041   __ret = vmovl_s16(__p0) + vmovl_s16(__p1);
67045 __ai int32x4_t vaddl_s16(int16x4_t __p0, int16x4_t __p1) {
67047   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67056 __ai uint16x8_t vaddw_u8(uint16x8_t __p0, uint8x8_t __p1) {
67058   __ret = __p0 + vmovl_u8(__p1);
67062 __ai uint16x8_t vaddw_u8(uint16x8_t __p0, uint8x8_t __p1) {
67064   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67073 __ai uint64x2_t vaddw_u32(uint64x2_t __p0, uint32x2_t __p1) {
67075   __ret = __p0 + vmovl_u32(__p1);
67079 __ai uint64x2_t vaddw_u32(uint64x2_t __p0, uint32x2_t __p1) {
67081   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67090 __ai uint32x4_t vaddw_u16(uint32x4_t __p0, uint16x4_t __p1) {
67092   __ret = __p0 + vmovl_u16(__p1);
67096 __ai uint32x4_t vaddw_u16(uint32x4_t __p0, uint16x4_t __p1) {
67098   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67107 __ai int16x8_t vaddw_s8(int16x8_t __p0, int8x8_t __p1) {
67109   __ret = __p0 + vmovl_s8(__p1);
67113 __ai int16x8_t vaddw_s8(int16x8_t __p0, int8x8_t __p1) {
67115   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67124 __ai int64x2_t vaddw_s32(int64x2_t __p0, int32x2_t __p1) {
67126   __ret = __p0 + vmovl_s32(__p1);
67130 __ai int64x2_t vaddw_s32(int64x2_t __p0, int32x2_t __p1) {
67132   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67141 __ai int32x4_t vaddw_s16(int32x4_t __p0, int16x4_t __p1) {
67143   __ret = __p0 + vmovl_s16(__p1);
67147 __ai int32x4_t vaddw_s16(int32x4_t __p0, int16x4_t __p1) {
67149   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67200 __ai uint16x8_t vmlal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
67202   __ret = __p0 + vmull_u8(__p1, __p2);
67206 __ai uint16x8_t vmlal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
67208   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67215 __ai uint16x8_t __noswap_vmlal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
67217   __ret = __p0 + __noswap_vmull_u8(__p1, __p2);
67223 __ai uint64x2_t vmlal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
67225   __ret = __p0 + vmull_u32(__p1, __p2);
67229 __ai uint64x2_t vmlal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
67231   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67238 __ai uint64x2_t __noswap_vmlal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
67240   __ret = __p0 + __noswap_vmull_u32(__p1, __p2);
67246 __ai uint32x4_t vmlal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
67248   __ret = __p0 + vmull_u16(__p1, __p2);
67252 __ai uint32x4_t vmlal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
67254   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67261 __ai uint32x4_t __noswap_vmlal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
67263   __ret = __p0 + __noswap_vmull_u16(__p1, __p2);
67269 __ai int16x8_t vmlal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67271   __ret = __p0 + vmull_s8(__p1, __p2);
67275 __ai int16x8_t vmlal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67277   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67284 __ai int16x8_t __noswap_vmlal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67286   __ret = __p0 + __noswap_vmull_s8(__p1, __p2);
67292 __ai int64x2_t vmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67294   __ret = __p0 + vmull_s32(__p1, __p2);
67298 __ai int64x2_t vmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67300   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67307 __ai int64x2_t __noswap_vmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67309   __ret = __p0 + __noswap_vmull_s32(__p1, __p2);
67315 __ai int32x4_t vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67317   __ret = __p0 + vmull_s16(__p1, __p2);
67321 __ai int32x4_t vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67323   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67330 __ai int32x4_t __noswap_vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67332   __ret = __p0 + __noswap_vmull_s16(__p1, __p2);
67338 #define vmlal_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
67340   uint32x2_t __s1 = __p1; \
67347 #define vmlal_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
67349   uint32x2_t __s1 = __p1; \
67362 #define vmlal_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
67364   uint16x4_t __s1 = __p1; \
67371 #define vmlal_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
67373   uint16x4_t __s1 = __p1; \
67386 #define vmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
67388   int32x2_t __s1 = __p1; \
67395 #define vmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
67397   int32x2_t __s1 = __p1; \
67410 #define vmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
67412   int16x4_t __s1 = __p1; \
67419 #define vmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
67421   int16x4_t __s1 = __p1; \
67434 __ai uint64x2_t vmlal_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
67436   __ret = __p0 + vmull_u32(__p1, (uint32x2_t) {__p2, __p2});
67440 __ai uint64x2_t vmlal_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
67442   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67448 __ai uint64x2_t __noswap_vmlal_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
67450   __ret = __p0 + __noswap_vmull_u32(__p1, (uint32x2_t) {__p2, __p2});
67456 __ai uint32x4_t vmlal_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
67458   __ret = __p0 + vmull_u16(__p1, (uint16x4_t) {__p2, __p2, __p2, __p2});
67462 __ai uint32x4_t vmlal_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
67464   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67470 __ai uint32x4_t __noswap_vmlal_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
67472   __ret = __p0 + __noswap_vmull_u16(__p1, (uint16x4_t) {__p2, __p2, __p2, __p2});
67478 __ai int64x2_t vmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
67480   __ret = __p0 + vmull_s32(__p1, (int32x2_t) {__p2, __p2});
67484 __ai int64x2_t vmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
67486   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67492 __ai int64x2_t __noswap_vmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
67494   __ret = __p0 + __noswap_vmull_s32(__p1, (int32x2_t) {__p2, __p2});
67500 __ai int32x4_t vmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
67502   __ret = __p0 + vmull_s16(__p1, (int16x4_t) {__p2, __p2, __p2, __p2});
67506 __ai int32x4_t vmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
67508   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67514 __ai int32x4_t __noswap_vmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
67516   __ret = __p0 + __noswap_vmull_s16(__p1, (int16x4_t) {__p2, __p2, __p2, __p2});
67522 __ai uint16x8_t vmlsl_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
67524   __ret = __p0 - vmull_u8(__p1, __p2);
67528 __ai uint16x8_t vmlsl_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
67530   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67537 __ai uint16x8_t __noswap_vmlsl_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
67539   __ret = __p0 - __noswap_vmull_u8(__p1, __p2);
67545 __ai uint64x2_t vmlsl_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
67547   __ret = __p0 - vmull_u32(__p1, __p2);
67551 __ai uint64x2_t vmlsl_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
67553   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67560 __ai uint64x2_t __noswap_vmlsl_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
67562   __ret = __p0 - __noswap_vmull_u32(__p1, __p2);
67568 __ai uint32x4_t vmlsl_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
67570   __ret = __p0 - vmull_u16(__p1, __p2);
67574 __ai uint32x4_t vmlsl_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
67576   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67583 __ai uint32x4_t __noswap_vmlsl_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
67585   __ret = __p0 - __noswap_vmull_u16(__p1, __p2);
67591 __ai int16x8_t vmlsl_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67593   __ret = __p0 - vmull_s8(__p1, __p2);
67597 __ai int16x8_t vmlsl_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67599   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67606 __ai int16x8_t __noswap_vmlsl_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67608   __ret = __p0 - __noswap_vmull_s8(__p1, __p2);
67614 __ai int64x2_t vmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67616   __ret = __p0 - vmull_s32(__p1, __p2);
67620 __ai int64x2_t vmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67622   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67629 __ai int64x2_t __noswap_vmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67631   __ret = __p0 - __noswap_vmull_s32(__p1, __p2);
67637 __ai int32x4_t vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67639   __ret = __p0 - vmull_s16(__p1, __p2);
67643 __ai int32x4_t vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67645   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67652 __ai int32x4_t __noswap_vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67654   __ret = __p0 - __noswap_vmull_s16(__p1, __p2);
67660 #define vmlsl_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
67662   uint32x2_t __s1 = __p1; \
67669 #define vmlsl_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
67671   uint32x2_t __s1 = __p1; \
67684 #define vmlsl_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
67686   uint16x4_t __s1 = __p1; \
67693 #define vmlsl_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
67695   uint16x4_t __s1 = __p1; \
67708 #define vmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
67710   int32x2_t __s1 = __p1; \
67717 #define vmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
67719   int32x2_t __s1 = __p1; \
67732 #define vmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
67734   int16x4_t __s1 = __p1; \
67741 #define vmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
67743   int16x4_t __s1 = __p1; \
67756 __ai uint64x2_t vmlsl_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
67758   __ret = __p0 - vmull_u32(__p1, (uint32x2_t) {__p2, __p2});
67762 __ai uint64x2_t vmlsl_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
67764   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67770 __ai uint64x2_t __noswap_vmlsl_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
67772   __ret = __p0 - __noswap_vmull_u32(__p1, (uint32x2_t) {__p2, __p2});
67778 __ai uint32x4_t vmlsl_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
67780   __ret = __p0 - vmull_u16(__p1, (uint16x4_t) {__p2, __p2, __p2, __p2});
67784 __ai uint32x4_t vmlsl_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
67786   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67792 __ai uint32x4_t __noswap_vmlsl_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
67794   __ret = __p0 - __noswap_vmull_u16(__p1, (uint16x4_t) {__p2, __p2, __p2, __p2});
67800 __ai int64x2_t vmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
67802   __ret = __p0 - vmull_s32(__p1, (int32x2_t) {__p2, __p2});
67806 __ai int64x2_t vmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
67808   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67814 __ai int64x2_t __noswap_vmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
67816   __ret = __p0 - __noswap_vmull_s32(__p1, (int32x2_t) {__p2, __p2});
67822 __ai int32x4_t vmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
67824   __ret = __p0 - vmull_s16(__p1, (int16x4_t) {__p2, __p2, __p2, __p2});
67828 __ai int32x4_t vmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
67830   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67836 __ai int32x4_t __noswap_vmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
67838   __ret = __p0 - __noswap_vmull_s16(__p1, (int16x4_t) {__p2, __p2, __p2, __p2});
67897 __ai int32_t vqrdmlahs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
67899   __ret = vqadds_s32(__p0, vqrdmulhs_s32(__p1, __p2));
67903 __ai int32_t vqrdmlahs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
67905   __ret = __noswap_vqadds_s32(__p0, __noswap_vqrdmulhs_s32(__p1, __p2));
67911 __ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
67913   __ret = vqaddh_s16(__p0, vqrdmulhh_s16(__p1, __p2));
67917 __ai int16_t vqrdmlahh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
67919   __ret = __noswap_vqaddh_s16(__p0, __noswap_vqrdmulhh_s16(__p1, __p2));
68009 __ai int32_t vqrdmlshs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
68011   __ret = vqsubs_s32(__p0, vqrdmulhs_s32(__p1, __p2));
68015 __ai int32_t vqrdmlshs_s32(int32_t __p0, int32_t __p1, int32_t __p2) {
68017   __ret = __noswap_vqsubs_s32(__p0, __noswap_vqrdmulhs_s32(__p1, __p2));
68023 __ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
68025   __ret = vqsubh_s16(__p0, vqrdmulhh_s16(__p1, __p2));
68029 __ai int16_t vqrdmlshh_s16(int16_t __p0, int16_t __p1, int16_t __p2) {
68031   __ret = __noswap_vqsubh_s16(__p0, __noswap_vqrdmulhh_s16(__p1, __p2));
68123 __ai uint16x8_t vabdl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
68125   __ret = vabdl_u8(vget_high_u8(__p0), vget_high_u8(__p1));
68129 __ai uint16x8_t vabdl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
68131   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68140 __ai uint64x2_t vabdl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
68142   __ret = vabdl_u32(vget_high_u32(__p0), vget_high_u32(__p1));
68146 __ai uint64x2_t vabdl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
68148   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68157 __ai uint32x4_t vabdl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
68159   __ret = vabdl_u16(vget_high_u16(__p0), vget_high_u16(__p1));
68163 __ai uint32x4_t vabdl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
68165   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68174 __ai int16x8_t vabdl_high_s8(int8x16_t __p0, int8x16_t __p1) {
68176   __ret = vabdl_s8(vget_high_s8(__p0), vget_high_s8(__p1));
68180 __ai int16x8_t vabdl_high_s8(int8x16_t __p0, int8x16_t __p1) {
68182   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68191 __ai int64x2_t vabdl_high_s32(int32x4_t __p0, int32x4_t __p1) {
68193   __ret = vabdl_s32(vget_high_s32(__p0), vget_high_s32(__p1));
68197 __ai int64x2_t vabdl_high_s32(int32x4_t __p0, int32x4_t __p1) {
68199   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68208 __ai int32x4_t vabdl_high_s16(int16x8_t __p0, int16x8_t __p1) {
68210   __ret = vabdl_s16(vget_high_s16(__p0), vget_high_s16(__p1));
68214 __ai int32x4_t vabdl_high_s16(int16x8_t __p0, int16x8_t __p1) {
68216   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68225 __ai uint16x8_t vaddl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
68227   __ret = vmovl_high_u8(__p0) + vmovl_high_u8(__p1);
68231 __ai uint16x8_t vaddl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
68233   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68242 __ai uint64x2_t vaddl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
68244   __ret = vmovl_high_u32(__p0) + vmovl_high_u32(__p1);
68248 __ai uint64x2_t vaddl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
68250   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68259 __ai uint32x4_t vaddl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
68261   __ret = vmovl_high_u16(__p0) + vmovl_high_u16(__p1);
68265 __ai uint32x4_t vaddl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
68267   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68276 __ai int16x8_t vaddl_high_s8(int8x16_t __p0, int8x16_t __p1) {
68278   __ret = vmovl_high_s8(__p0) + vmovl_high_s8(__p1);
68282 __ai int16x8_t vaddl_high_s8(int8x16_t __p0, int8x16_t __p1) {
68284   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68293 __ai int64x2_t vaddl_high_s32(int32x4_t __p0, int32x4_t __p1) {
68295   __ret = vmovl_high_s32(__p0) + vmovl_high_s32(__p1);
68299 __ai int64x2_t vaddl_high_s32(int32x4_t __p0, int32x4_t __p1) {
68301   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68310 __ai int32x4_t vaddl_high_s16(int16x8_t __p0, int16x8_t __p1) {
68312   __ret = vmovl_high_s16(__p0) + vmovl_high_s16(__p1);
68316 __ai int32x4_t vaddl_high_s16(int16x8_t __p0, int16x8_t __p1) {
68318   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68327 __ai uint16x8_t vaddw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
68329   __ret = __p0 + vmovl_high_u8(__p1);
68333 __ai uint16x8_t vaddw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
68335   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68344 __ai uint64x2_t vaddw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
68346   __ret = __p0 + vmovl_high_u32(__p1);
68350 __ai uint64x2_t vaddw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
68352   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68361 __ai uint32x4_t vaddw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
68363   __ret = __p0 + vmovl_high_u16(__p1);
68367 __ai uint32x4_t vaddw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
68369   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68378 __ai int16x8_t vaddw_high_s8(int16x8_t __p0, int8x16_t __p1) {
68380   __ret = __p0 + vmovl_high_s8(__p1);
68384 __ai int16x8_t vaddw_high_s8(int16x8_t __p0, int8x16_t __p1) {
68386   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68395 __ai int64x2_t vaddw_high_s32(int64x2_t __p0, int32x4_t __p1) {
68397   __ret = __p0 + vmovl_high_s32(__p1);
68401 __ai int64x2_t vaddw_high_s32(int64x2_t __p0, int32x4_t __p1) {
68403   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68412 __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
68414   __ret = __p0 + vmovl_high_s16(__p1);
68418 __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
68420   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68585 __ai uint16x8_t vmlal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
68587   __ret = vmlal_u8(__p0, vget_high_u8(__p1), vget_high_u8(__p2));
68591 __ai uint16x8_t vmlal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
68593   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68603 __ai uint64x2_t vmlal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
68605   __ret = vmlal_u32(__p0, vget_high_u32(__p1), vget_high_u32(__p2));
68609 __ai uint64x2_t vmlal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
68611   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68621 __ai uint32x4_t vmlal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
68623   __ret = vmlal_u16(__p0, vget_high_u16(__p1), vget_high_u16(__p2));
68627 __ai uint32x4_t vmlal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
68629   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68639 __ai int16x8_t vmlal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
68641   __ret = vmlal_s8(__p0, vget_high_s8(__p1), vget_high_s8(__p2));
68645 __ai int16x8_t vmlal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
68647   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68657 __ai int64x2_t vmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
68659   __ret = vmlal_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
68663 __ai int64x2_t vmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
68665   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68675 __ai int32x4_t vmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
68677   __ret = vmlal_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
68681 __ai int32x4_t vmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
68683   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68693 __ai uint64x2_t vmlal_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2) {
68695   __ret = vmlal_n_u32(__p0, vget_high_u32(__p1), __p2);
68699 __ai uint64x2_t vmlal_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2) {
68701   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68710 __ai uint32x4_t vmlal_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2) {
68712   __ret = vmlal_n_u16(__p0, vget_high_u16(__p1), __p2);
68716 __ai uint32x4_t vmlal_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2) {
68718   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68727 __ai int64x2_t vmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
68729   __ret = vmlal_n_s32(__p0, vget_high_s32(__p1), __p2);
68733 __ai int64x2_t vmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
68735   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68744 __ai int32x4_t vmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
68746   __ret = vmlal_n_s16(__p0, vget_high_s16(__p1), __p2);
68750 __ai int32x4_t vmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
68752   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68761 __ai uint16x8_t vmlsl_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
68763   __ret = vmlsl_u8(__p0, vget_high_u8(__p1), vget_high_u8(__p2));
68767 __ai uint16x8_t vmlsl_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
68769   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68779 __ai uint64x2_t vmlsl_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
68781   __ret = vmlsl_u32(__p0, vget_high_u32(__p1), vget_high_u32(__p2));
68785 __ai uint64x2_t vmlsl_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
68787   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68797 __ai uint32x4_t vmlsl_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
68799   __ret = vmlsl_u16(__p0, vget_high_u16(__p1), vget_high_u16(__p2));
68803 __ai uint32x4_t vmlsl_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
68805   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68815 __ai int16x8_t vmlsl_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
68817   __ret = vmlsl_s8(__p0, vget_high_s8(__p1), vget_high_s8(__p2));
68821 __ai int16x8_t vmlsl_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
68823   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68833 __ai int64x2_t vmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
68835   __ret = vmlsl_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
68839 __ai int64x2_t vmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
68841   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68851 __ai int32x4_t vmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
68853   __ret = vmlsl_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
68857 __ai int32x4_t vmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
68859   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68869 __ai uint64x2_t vmlsl_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2) {
68871   __ret = vmlsl_n_u32(__p0, vget_high_u32(__p1), __p2);
68875 __ai uint64x2_t vmlsl_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2) {
68877   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68886 __ai uint32x4_t vmlsl_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2) {
68888   __ret = vmlsl_n_u16(__p0, vget_high_u16(__p1), __p2);
68892 __ai uint32x4_t vmlsl_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2) {
68894   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68903 __ai int64x2_t vmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
68905   __ret = vmlsl_n_s32(__p0, vget_high_s32(__p1), __p2);
68909 __ai int64x2_t vmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
68911   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68920 __ai int32x4_t vmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
68922   __ret = vmlsl_n_s16(__p0, vget_high_s16(__p1), __p2);
68926 __ai int32x4_t vmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
68928   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68987 __ai uint16x8_t vabal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
68989   __ret = __p0 + vabdl_u8(__p1, __p2);
68993 __ai uint16x8_t vabal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
68995   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
69002 __ai uint16x8_t __noswap_vabal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
69004   __ret = __p0 + __noswap_vabdl_u8(__p1, __p2);
69010 __ai uint64x2_t vabal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
69012   __ret = __p0 + vabdl_u32(__p1, __p2);
69016 __ai uint64x2_t vabal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
69018   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
69025 __ai uint64x2_t __noswap_vabal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
69027   __ret = __p0 + __noswap_vabdl_u32(__p1, __p2);
69033 __ai uint32x4_t vabal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
69035   __ret = __p0 + vabdl_u16(__p1, __p2);
69039 __ai uint32x4_t vabal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
69041   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
69048 __ai uint32x4_t __noswap_vabal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
69050   __ret = __p0 + __noswap_vabdl_u16(__p1, __p2);
69056 __ai int16x8_t vabal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
69058   __ret = __p0 + vabdl_s8(__p1, __p2);
69062 __ai int16x8_t vabal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
69064   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
69071 __ai int16x8_t __noswap_vabal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
69073   __ret = __p0 + __noswap_vabdl_s8(__p1, __p2);
69079 __ai int64x2_t vabal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
69081   __ret = __p0 + vabdl_s32(__p1, __p2);
69085 __ai int64x2_t vabal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
69087   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
69094 __ai int64x2_t __noswap_vabal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
69096   __ret = __p0 + __noswap_vabdl_s32(__p1, __p2);
69102 __ai int32x4_t vabal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
69104   __ret = __p0 + vabdl_s16(__p1, __p2);
69108 __ai int32x4_t vabal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
69110   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
69117 __ai int32x4_t __noswap_vabal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
69119   __ret = __p0 + __noswap_vabdl_s16(__p1, __p2);
69126 __ai uint16x8_t vabal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
69128   __ret = vabal_u8(__p0, vget_high_u8(__p1), vget_high_u8(__p2));
69132 __ai uint16x8_t vabal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
69134   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
69144 __ai uint64x2_t vabal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
69146   __ret = vabal_u32(__p0, vget_high_u32(__p1), vget_high_u32(__p2));
69150 __ai uint64x2_t vabal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
69152   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
69162 __ai uint32x4_t vabal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
69164   __ret = vabal_u16(__p0, vget_high_u16(__p1), vget_high_u16(__p2));
69168 __ai uint32x4_t vabal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
69170   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
69180 __ai int16x8_t vabal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
69182   __ret = vabal_s8(__p0, vget_high_s8(__p1), vget_high_s8(__p2));
69186 __ai int16x8_t vabal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
69188   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
69198 __ai int64x2_t vabal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
69200   __ret = vabal_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
69204 __ai int64x2_t vabal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
69206   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
69216 __ai int32x4_t vabal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
69218   __ret = vabal_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
69222 __ai int32x4_t vabal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
69224   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);