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);
16897 __ai int64x2_t vqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
16899   __ret = (int64x2_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
16903 __ai int64x2_t vqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
16905   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16914 __ai int16x8_t vqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
16916   __ret = (int16x8_t) __builtin_neon_vqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
16920 __ai int16x8_t vqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
16922   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16931 __ai uint8x8_t vqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
16933   __ret = (uint8x8_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
16937 __ai uint8x8_t vqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
16939   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
16948 __ai uint32x2_t vqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
16950   __ret = (uint32x2_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
16954 __ai uint32x2_t vqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
16956   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
16965 __ai uint64x1_t vqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
16967   __ret = (uint64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
16971 __ai uint64x1_t vqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
16973   __ret = (uint64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
16979 __ai uint16x4_t vqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
16981   __ret = (uint16x4_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
16985 __ai uint16x4_t vqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
16987   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
16996 __ai int8x8_t vqadd_s8(int8x8_t __p0, int8x8_t __p1) {
16998   __ret = (int8x8_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
17002 __ai int8x8_t vqadd_s8(int8x8_t __p0, int8x8_t __p1) {
17004   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
17013 __ai int32x2_t vqadd_s32(int32x2_t __p0, int32x2_t __p1) {
17015   __ret = (int32x2_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
17019 __ai int32x2_t vqadd_s32(int32x2_t __p0, int32x2_t __p1) {
17021   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17030 __ai int64x1_t vqadd_s64(int64x1_t __p0, int64x1_t __p1) {
17032   __ret = (int64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
17036 __ai int64x1_t vqadd_s64(int64x1_t __p0, int64x1_t __p1) {
17038   __ret = (int64x1_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
17044 __ai int16x4_t vqadd_s16(int16x4_t __p0, int16x4_t __p1) {
17046   __ret = (int16x4_t) __builtin_neon_vqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
17050 __ai int16x4_t vqadd_s16(int16x4_t __p0, int16x4_t __p1) {
17052   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17061 __ai int64x2_t vqdmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17063   __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 35);
17067 __ai int64x2_t vqdmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17069   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17076 __ai int64x2_t __noswap_vqdmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17078   __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 35);
17084 __ai int32x4_t vqdmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17086   __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 34);
17090 __ai int32x4_t vqdmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17092   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17099 __ai int32x4_t __noswap_vqdmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17101   __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 34);
17107 #define vqdmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
17109   int32x2_t __s1 = __p1; \
17116 #define vqdmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
17118   int32x2_t __s1 = __p1; \
17131 #define vqdmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
17133   int16x4_t __s1 = __p1; \
17140 #define vqdmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
17142   int16x4_t __s1 = __p1; \
17155 __ai int64x2_t vqdmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17157   __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
17161 __ai int64x2_t vqdmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17163   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17169 __ai int64x2_t __noswap_vqdmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17171   __ret = (int64x2_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
17177 __ai int32x4_t vqdmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17179   __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
17183 __ai int32x4_t vqdmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17185   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17191 __ai int32x4_t __noswap_vqdmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17193   __ret = (int32x4_t) __builtin_neon_vqdmlal_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
17199 __ai int64x2_t vqdmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17201   __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 35);
17205 __ai int64x2_t vqdmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17207   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17214 __ai int64x2_t __noswap_vqdmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
17216   __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 35);
17222 __ai int32x4_t vqdmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17224   __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 34);
17228 __ai int32x4_t vqdmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17230   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17237 __ai int32x4_t __noswap_vqdmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
17239   __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 34);
17245 #define vqdmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
17247   int32x2_t __s1 = __p1; \
17254 #define vqdmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
17256   int32x2_t __s1 = __p1; \
17269 #define vqdmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
17271   int16x4_t __s1 = __p1; \
17278 #define vqdmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
17280   int16x4_t __s1 = __p1; \
17293 __ai int64x2_t vqdmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17295   __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
17299 __ai int64x2_t vqdmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17301   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17307 __ai int64x2_t __noswap_vqdmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
17309   __ret = (int64x2_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int32x2_t) {__p2, __p2}, 35);
17315 __ai int32x4_t vqdmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17317   __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
17321 __ai int32x4_t vqdmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17323   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17329 __ai int32x4_t __noswap_vqdmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
17331   __ret = (int32x4_t) __builtin_neon_vqdmlsl_v((int8x16_t)__p0, (int8x8_t)__p1, (int8x8_t)(int16x4_t) {__p2, __p2, __p2, __p2}, 34);
17337 __ai int32x4_t vqdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
17339   __ret = (int32x4_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
17343 __ai int32x4_t vqdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
17345   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17351 __ai int32x4_t __noswap_vqdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
17353   __ret = (int32x4_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
17359 __ai int16x8_t vqdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
17361   __ret = (int16x8_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
17365 __ai int16x8_t vqdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
17367   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
17373 __ai int16x8_t __noswap_vqdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
17375   __ret = (int16x8_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
17381 __ai int32x2_t vqdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
17383   __ret = (int32x2_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
17387 __ai int32x2_t vqdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
17389   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17395 __ai int32x2_t __noswap_vqdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
17397   __ret = (int32x2_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
17403 __ai int16x4_t vqdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
17405   __ret = (int16x4_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
17409 __ai int16x4_t vqdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
17411   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17417 __ai int16x4_t __noswap_vqdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
17419   __ret = (int16x4_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
17425 #define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17427   int32x2_t __s1 = __p1; \
17433 #define vqdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17435   int32x2_t __s1 = __p1; \
17446 #define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17448   int16x4_t __s1 = __p1; \
17454 #define vqdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17456   int16x4_t __s1 = __p1; \
17467 #define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17469   int32x2_t __s1 = __p1; \
17475 #define vqdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17477   int32x2_t __s1 = __p1; \
17488 #define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17490   int16x4_t __s1 = __p1; \
17496 #define vqdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17498   int16x4_t __s1 = __p1; \
17509 __ai int32x4_t vqdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
17511   __ret = (int32x4_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
17515 __ai int32x4_t vqdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
17518   __ret = (int32x4_t) __builtin_neon_vqdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
17525 __ai int16x8_t vqdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
17527   __ret = (int16x8_t) __builtin_neon_vqdmulhq_v((int8x16_t)__p0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
17531 __ai int16x8_t vqdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
17534   __ret = (int16x8_t) __builtin_neon_vqdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
17541 __ai int32x2_t vqdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
17543   __ret = (int32x2_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
17547 __ai int32x2_t vqdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
17550   __ret = (int32x2_t) __builtin_neon_vqdmulh_v((int8x8_t)__rev0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
17557 __ai int16x4_t vqdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
17559   __ret = (int16x4_t) __builtin_neon_vqdmulh_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
17563 __ai int16x4_t vqdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
17566   __ret = (int16x4_t) __builtin_neon_vqdmulh_v((int8x8_t)__rev0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
17573 __ai int64x2_t vqdmull_s32(int32x2_t __p0, int32x2_t __p1) {
17575   __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)__p1, 35);
17579 __ai int64x2_t vqdmull_s32(int32x2_t __p0, int32x2_t __p1) {
17581   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
17587 __ai int64x2_t __noswap_vqdmull_s32(int32x2_t __p0, int32x2_t __p1) {
17589   __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)__p1, 35);
17595 __ai int32x4_t vqdmull_s16(int16x4_t __p0, int16x4_t __p1) {
17597   __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)__p1, 34);
17601 __ai int32x4_t vqdmull_s16(int16x4_t __p0, int16x4_t __p1) {
17603   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
17609 __ai int32x4_t __noswap_vqdmull_s16(int16x4_t __p0, int16x4_t __p1) {
17611   __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)__p1, 34);
17617 #define vqdmull_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17619   int32x2_t __s1 = __p1; \
17625 #define vqdmull_lane_s32(__p0, __p1, __p2) __extension__ ({ \
17627   int32x2_t __s1 = __p1; \
17638 #define vqdmull_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17640   int16x4_t __s1 = __p1; \
17646 #define vqdmull_lane_s16(__p0, __p1, __p2) __extension__ ({ \
17648   int16x4_t __s1 = __p1; \
17659 __ai int64x2_t vqdmull_n_s32(int32x2_t __p0, int32_t __p1) {
17661   __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
17665 __ai int64x2_t vqdmull_n_s32(int32x2_t __p0, int32_t __p1) {
17668   __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__rev0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
17672 __ai int64x2_t __noswap_vqdmull_n_s32(int32x2_t __p0, int32_t __p1) {
17674   __ret = (int64x2_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 35);
17680 __ai int32x4_t vqdmull_n_s16(int16x4_t __p0, int16_t __p1) {
17682   __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
17686 __ai int32x4_t vqdmull_n_s16(int16x4_t __p0, int16_t __p1) {
17689   __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__rev0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
17693 __ai int32x4_t __noswap_vqdmull_n_s16(int16x4_t __p0, int16_t __p1) {
17695   __ret = (int32x4_t) __builtin_neon_vqdmull_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 34);
17986 __ai int32x4_t vqrdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
17988   __ret = (int32x4_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
17992 __ai int32x4_t vqrdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
17994   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18000 __ai int32x4_t __noswap_vqrdmulhq_s32(int32x4_t __p0, int32x4_t __p1) {
18002   __ret = (int32x4_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
18008 __ai int16x8_t vqrdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
18010   __ret = (int16x8_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
18014 __ai int16x8_t vqrdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
18016   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18022 __ai int16x8_t __noswap_vqrdmulhq_s16(int16x8_t __p0, int16x8_t __p1) {
18024   __ret = (int16x8_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
18030 __ai int32x2_t vqrdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
18032   __ret = (int32x2_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
18036 __ai int32x2_t vqrdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
18038   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18044 __ai int32x2_t __noswap_vqrdmulh_s32(int32x2_t __p0, int32x2_t __p1) {
18046   __ret = (int32x2_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
18052 __ai int16x4_t vqrdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
18054   __ret = (int16x4_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
18058 __ai int16x4_t vqrdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
18060   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18066 __ai int16x4_t __noswap_vqrdmulh_s16(int16x4_t __p0, int16x4_t __p1) {
18068   __ret = (int16x4_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
18074 #define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
18076   int32x2_t __s1 = __p1; \
18082 #define vqrdmulhq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
18084   int32x2_t __s1 = __p1; \
18095 #define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
18097   int16x4_t __s1 = __p1; \
18103 #define vqrdmulhq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
18105   int16x4_t __s1 = __p1; \
18116 #define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
18118   int32x2_t __s1 = __p1; \
18124 #define vqrdmulh_lane_s32(__p0, __p1, __p2) __extension__ ({ \
18126   int32x2_t __s1 = __p1; \
18137 #define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
18139   int16x4_t __s1 = __p1; \
18145 #define vqrdmulh_lane_s16(__p0, __p1, __p2) __extension__ ({ \
18147   int16x4_t __s1 = __p1; \
18158 __ai int32x4_t vqrdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
18160   __ret = (int32x4_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
18164 __ai int32x4_t vqrdmulhq_n_s32(int32x4_t __p0, int32_t __p1) {
18167   __ret = (int32x4_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int32x4_t) {__p1, __p1, __p1, __p1}, 34);
18174 __ai int16x8_t vqrdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
18176   __ret = (int16x8_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__p0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
18180 __ai int16x8_t vqrdmulhq_n_s16(int16x8_t __p0, int16_t __p1) {
18183   __ret = (int16x8_t) __builtin_neon_vqrdmulhq_v((int8x16_t)__rev0, (int8x16_t)(int16x8_t) {__p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1}, 33);
18190 __ai int32x2_t vqrdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
18192   __ret = (int32x2_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
18196 __ai int32x2_t vqrdmulh_n_s32(int32x2_t __p0, int32_t __p1) {
18199   __ret = (int32x2_t) __builtin_neon_vqrdmulh_v((int8x8_t)__rev0, (int8x8_t)(int32x2_t) {__p1, __p1}, 2);
18206 __ai int16x4_t vqrdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
18208   __ret = (int16x4_t) __builtin_neon_vqrdmulh_v((int8x8_t)__p0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
18212 __ai int16x4_t vqrdmulh_n_s16(int16x4_t __p0, int16_t __p1) {
18215   __ret = (int16x4_t) __builtin_neon_vqrdmulh_v((int8x8_t)__rev0, (int8x8_t)(int16x4_t) {__p1, __p1, __p1, __p1}, 1);
18222 __ai uint8x16_t vqrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
18224   __ret = (uint8x16_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
18228 __ai uint8x16_t vqrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
18230   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18239 __ai uint32x4_t vqrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
18241   __ret = (uint32x4_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
18245 __ai uint32x4_t vqrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
18247   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18256 __ai uint64x2_t vqrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
18258   __ret = (uint64x2_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
18262 __ai uint64x2_t vqrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
18264   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18273 __ai uint16x8_t vqrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
18275   __ret = (uint16x8_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
18279 __ai uint16x8_t vqrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
18281   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18290 __ai int8x16_t vqrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
18292   __ret = (int8x16_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
18296 __ai int8x16_t vqrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
18298   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18307 __ai int32x4_t vqrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
18309   __ret = (int32x4_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
18313 __ai int32x4_t vqrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
18315   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18324 __ai int64x2_t vqrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
18326   __ret = (int64x2_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
18330 __ai int64x2_t vqrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
18332   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18341 __ai int16x8_t vqrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
18343   __ret = (int16x8_t) __builtin_neon_vqrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
18347 __ai int16x8_t vqrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
18349   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18358 __ai uint8x8_t vqrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
18360   __ret = (uint8x8_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
18364 __ai uint8x8_t vqrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
18366   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18375 __ai uint32x2_t vqrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
18377   __ret = (uint32x2_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
18381 __ai uint32x2_t vqrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
18383   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18392 __ai uint64x1_t vqrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
18394   __ret = (uint64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
18398 __ai uint64x1_t vqrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
18400   __ret = (uint64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
18406 __ai uint16x4_t vqrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
18408   __ret = (uint16x4_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
18412 __ai uint16x4_t vqrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
18414   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18423 __ai int8x8_t vqrshl_s8(int8x8_t __p0, int8x8_t __p1) {
18425   __ret = (int8x8_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
18429 __ai int8x8_t vqrshl_s8(int8x8_t __p0, int8x8_t __p1) {
18431   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18440 __ai int32x2_t vqrshl_s32(int32x2_t __p0, int32x2_t __p1) {
18442   __ret = (int32x2_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
18446 __ai int32x2_t vqrshl_s32(int32x2_t __p0, int32x2_t __p1) {
18448   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18457 __ai int64x1_t vqrshl_s64(int64x1_t __p0, int64x1_t __p1) {
18459   __ret = (int64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
18463 __ai int64x1_t vqrshl_s64(int64x1_t __p0, int64x1_t __p1) {
18465   __ret = (int64x1_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
18471 __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
18473   __ret = (int16x4_t) __builtin_neon_vqrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
18477 __ai int16x4_t vqrshl_s16(int16x4_t __p0, int16x4_t __p1) {
18479   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18488 #define vqrshrn_n_u32(__p0, __p1) __extension__ ({ \
18491   __ret = (uint16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 17); \
18495 #define vqrshrn_n_u32(__p0, __p1) __extension__ ({ \
18499   __ret = (uint16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 17); \
18503 #define __noswap_vqrshrn_n_u32(__p0, __p1) __extension__ ({ \
18506   __ret = (uint16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 17); \
18512 #define vqrshrn_n_u64(__p0, __p1) __extension__ ({ \
18515   __ret = (uint32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 18); \
18519 #define vqrshrn_n_u64(__p0, __p1) __extension__ ({ \
18523   __ret = (uint32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 18); \
18527 #define __noswap_vqrshrn_n_u64(__p0, __p1) __extension__ ({ \
18530   __ret = (uint32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 18); \
18536 #define vqrshrn_n_u16(__p0, __p1) __extension__ ({ \
18539   __ret = (uint8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 16); \
18543 #define vqrshrn_n_u16(__p0, __p1) __extension__ ({ \
18547   __ret = (uint8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 16); \
18551 #define __noswap_vqrshrn_n_u16(__p0, __p1) __extension__ ({ \
18554   __ret = (uint8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 16); \
18560 #define vqrshrn_n_s32(__p0, __p1) __extension__ ({ \
18563   __ret = (int16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 1); \
18567 #define vqrshrn_n_s32(__p0, __p1) __extension__ ({ \
18571   __ret = (int16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 1); \
18575 #define __noswap_vqrshrn_n_s32(__p0, __p1) __extension__ ({ \
18578   __ret = (int16x4_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 1); \
18584 #define vqrshrn_n_s64(__p0, __p1) __extension__ ({ \
18587   __ret = (int32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 2); \
18591 #define vqrshrn_n_s64(__p0, __p1) __extension__ ({ \
18595   __ret = (int32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 2); \
18599 #define __noswap_vqrshrn_n_s64(__p0, __p1) __extension__ ({ \
18602   __ret = (int32x2_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 2); \
18608 #define vqrshrn_n_s16(__p0, __p1) __extension__ ({ \
18611   __ret = (int8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 0); \
18615 #define vqrshrn_n_s16(__p0, __p1) __extension__ ({ \
18619   __ret = (int8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__rev0, __p1, 0); \
18623 #define __noswap_vqrshrn_n_s16(__p0, __p1) __extension__ ({ \
18626   __ret = (int8x8_t) __builtin_neon_vqrshrn_n_v((int8x16_t)__s0, __p1, 0); \
18632 #define vqrshrun_n_s32(__p0, __p1) __extension__ ({ \
18635   __ret = (uint16x4_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 17); \
18639 #define vqrshrun_n_s32(__p0, __p1) __extension__ ({ \
18643   __ret = (uint16x4_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__rev0, __p1, 17); \
18647 #define __noswap_vqrshrun_n_s32(__p0, __p1) __extension__ ({ \
18650   __ret = (uint16x4_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 17); \
18656 #define vqrshrun_n_s64(__p0, __p1) __extension__ ({ \
18659   __ret = (uint32x2_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 18); \
18663 #define vqrshrun_n_s64(__p0, __p1) __extension__ ({ \
18667   __ret = (uint32x2_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__rev0, __p1, 18); \
18671 #define __noswap_vqrshrun_n_s64(__p0, __p1) __extension__ ({ \
18674   __ret = (uint32x2_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 18); \
18680 #define vqrshrun_n_s16(__p0, __p1) __extension__ ({ \
18683   __ret = (uint8x8_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 16); \
18687 #define vqrshrun_n_s16(__p0, __p1) __extension__ ({ \
18691   __ret = (uint8x8_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__rev0, __p1, 16); \
18695 #define __noswap_vqrshrun_n_s16(__p0, __p1) __extension__ ({ \
18698   __ret = (uint8x8_t) __builtin_neon_vqrshrun_n_v((int8x16_t)__s0, __p1, 16); \
18704 __ai uint8x16_t vqshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
18706   __ret = (uint8x16_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
18710 __ai uint8x16_t vqshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
18712   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18721 __ai uint32x4_t vqshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
18723   __ret = (uint32x4_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
18727 __ai uint32x4_t vqshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
18729   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18738 __ai uint64x2_t vqshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
18740   __ret = (uint64x2_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
18744 __ai uint64x2_t vqshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
18746   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18755 __ai uint16x8_t vqshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
18757   __ret = (uint16x8_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
18761 __ai uint16x8_t vqshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
18763   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18772 __ai int8x16_t vqshlq_s8(int8x16_t __p0, int8x16_t __p1) {
18774   __ret = (int8x16_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
18778 __ai int8x16_t vqshlq_s8(int8x16_t __p0, int8x16_t __p1) {
18780   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
18789 __ai int32x4_t vqshlq_s32(int32x4_t __p0, int32x4_t __p1) {
18791   __ret = (int32x4_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
18795 __ai int32x4_t vqshlq_s32(int32x4_t __p0, int32x4_t __p1) {
18797   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18806 __ai int64x2_t vqshlq_s64(int64x2_t __p0, int64x2_t __p1) {
18808   __ret = (int64x2_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
18812 __ai int64x2_t vqshlq_s64(int64x2_t __p0, int64x2_t __p1) {
18814   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18823 __ai int16x8_t vqshlq_s16(int16x8_t __p0, int16x8_t __p1) {
18825   __ret = (int16x8_t) __builtin_neon_vqshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
18829 __ai int16x8_t vqshlq_s16(int16x8_t __p0, int16x8_t __p1) {
18831   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18840 __ai uint8x8_t vqshl_u8(uint8x8_t __p0, int8x8_t __p1) {
18842   __ret = (uint8x8_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
18846 __ai uint8x8_t vqshl_u8(uint8x8_t __p0, int8x8_t __p1) {
18848   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18857 __ai uint32x2_t vqshl_u32(uint32x2_t __p0, int32x2_t __p1) {
18859   __ret = (uint32x2_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
18863 __ai uint32x2_t vqshl_u32(uint32x2_t __p0, int32x2_t __p1) {
18865   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18874 __ai uint64x1_t vqshl_u64(uint64x1_t __p0, int64x1_t __p1) {
18876   __ret = (uint64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
18880 __ai uint64x1_t vqshl_u64(uint64x1_t __p0, int64x1_t __p1) {
18882   __ret = (uint64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
18888 __ai uint16x4_t vqshl_u16(uint16x4_t __p0, int16x4_t __p1) {
18890   __ret = (uint16x4_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
18894 __ai uint16x4_t vqshl_u16(uint16x4_t __p0, int16x4_t __p1) {
18896   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18905 __ai int8x8_t vqshl_s8(int8x8_t __p0, int8x8_t __p1) {
18907   __ret = (int8x8_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
18911 __ai int8x8_t vqshl_s8(int8x8_t __p0, int8x8_t __p1) {
18913   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
18922 __ai int32x2_t vqshl_s32(int32x2_t __p0, int32x2_t __p1) {
18924   __ret = (int32x2_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
18928 __ai int32x2_t vqshl_s32(int32x2_t __p0, int32x2_t __p1) {
18930   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
18939 __ai int64x1_t vqshl_s64(int64x1_t __p0, int64x1_t __p1) {
18941   __ret = (int64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
18945 __ai int64x1_t vqshl_s64(int64x1_t __p0, int64x1_t __p1) {
18947   __ret = (int64x1_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
18953 __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
18955   __ret = (int16x4_t) __builtin_neon_vqshl_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
18959 __ai int16x4_t vqshl_s16(int16x4_t __p0, int16x4_t __p1) {
18961   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
18970 #define vqshlq_n_u8(__p0, __p1) __extension__ ({ \
18973   __ret = (uint8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 48); \
18977 #define vqshlq_n_u8(__p0, __p1) __extension__ ({ \
18981   __ret = (uint8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 48); \
18988 #define vqshlq_n_u32(__p0, __p1) __extension__ ({ \
18991   __ret = (uint32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 50); \
18995 #define vqshlq_n_u32(__p0, __p1) __extension__ ({ \
18999   __ret = (uint32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 50); \
19006 #define vqshlq_n_u64(__p0, __p1) __extension__ ({ \
19009   __ret = (uint64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 51); \
19013 #define vqshlq_n_u64(__p0, __p1) __extension__ ({ \
19017   __ret = (uint64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 51); \
19024 #define vqshlq_n_u16(__p0, __p1) __extension__ ({ \
19027   __ret = (uint16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 49); \
19031 #define vqshlq_n_u16(__p0, __p1) __extension__ ({ \
19035   __ret = (uint16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 49); \
19042 #define vqshlq_n_s8(__p0, __p1) __extension__ ({ \
19045   __ret = (int8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 32); \
19049 #define vqshlq_n_s8(__p0, __p1) __extension__ ({ \
19053   __ret = (int8x16_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 32); \
19060 #define vqshlq_n_s32(__p0, __p1) __extension__ ({ \
19063   __ret = (int32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 34); \
19067 #define vqshlq_n_s32(__p0, __p1) __extension__ ({ \
19071   __ret = (int32x4_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 34); \
19078 #define vqshlq_n_s64(__p0, __p1) __extension__ ({ \
19081   __ret = (int64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 35); \
19085 #define vqshlq_n_s64(__p0, __p1) __extension__ ({ \
19089   __ret = (int64x2_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 35); \
19096 #define vqshlq_n_s16(__p0, __p1) __extension__ ({ \
19099   __ret = (int16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__s0, __p1, 33); \
19103 #define vqshlq_n_s16(__p0, __p1) __extension__ ({ \
19107   __ret = (int16x8_t) __builtin_neon_vqshlq_n_v((int8x16_t)__rev0, __p1, 33); \
19114 #define vqshl_n_u8(__p0, __p1) __extension__ ({ \
19117   __ret = (uint8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 16); \
19121 #define vqshl_n_u8(__p0, __p1) __extension__ ({ \
19125   __ret = (uint8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 16); \
19132 #define vqshl_n_u32(__p0, __p1) __extension__ ({ \
19135   __ret = (uint32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 18); \
19139 #define vqshl_n_u32(__p0, __p1) __extension__ ({ \
19143   __ret = (uint32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 18); \
19150 #define vqshl_n_u64(__p0, __p1) __extension__ ({ \
19153   __ret = (uint64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 19); \
19157 #define vqshl_n_u64(__p0, __p1) __extension__ ({ \
19160   __ret = (uint64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 19); \
19166 #define vqshl_n_u16(__p0, __p1) __extension__ ({ \
19169   __ret = (uint16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 17); \
19173 #define vqshl_n_u16(__p0, __p1) __extension__ ({ \
19177   __ret = (uint16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 17); \
19184 #define vqshl_n_s8(__p0, __p1) __extension__ ({ \
19187   __ret = (int8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 0); \
19191 #define vqshl_n_s8(__p0, __p1) __extension__ ({ \
19195   __ret = (int8x8_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 0); \
19202 #define vqshl_n_s32(__p0, __p1) __extension__ ({ \
19205   __ret = (int32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 2); \
19209 #define vqshl_n_s32(__p0, __p1) __extension__ ({ \
19213   __ret = (int32x2_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 2); \
19220 #define vqshl_n_s64(__p0, __p1) __extension__ ({ \
19223   __ret = (int64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 3); \
19227 #define vqshl_n_s64(__p0, __p1) __extension__ ({ \
19230   __ret = (int64x1_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 3); \
19236 #define vqshl_n_s16(__p0, __p1) __extension__ ({ \
19239   __ret = (int16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__s0, __p1, 1); \
19243 #define vqshl_n_s16(__p0, __p1) __extension__ ({ \
19247   __ret = (int16x4_t) __builtin_neon_vqshl_n_v((int8x8_t)__rev0, __p1, 1); \
19254 #define vqshluq_n_s8(__p0, __p1) __extension__ ({ \
19257   __ret = (uint8x16_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 48); \
19261 #define vqshluq_n_s8(__p0, __p1) __extension__ ({ \
19265   __ret = (uint8x16_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 48); \
19272 #define vqshluq_n_s32(__p0, __p1) __extension__ ({ \
19275   __ret = (uint32x4_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 50); \
19279 #define vqshluq_n_s32(__p0, __p1) __extension__ ({ \
19283   __ret = (uint32x4_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 50); \
19290 #define vqshluq_n_s64(__p0, __p1) __extension__ ({ \
19293   __ret = (uint64x2_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 51); \
19297 #define vqshluq_n_s64(__p0, __p1) __extension__ ({ \
19301   __ret = (uint64x2_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 51); \
19308 #define vqshluq_n_s16(__p0, __p1) __extension__ ({ \
19311   __ret = (uint16x8_t) __builtin_neon_vqshluq_n_v((int8x16_t)__s0, __p1, 49); \
19315 #define vqshluq_n_s16(__p0, __p1) __extension__ ({ \
19319   __ret = (uint16x8_t) __builtin_neon_vqshluq_n_v((int8x16_t)__rev0, __p1, 49); \
19326 #define vqshlu_n_s8(__p0, __p1) __extension__ ({ \
19329   __ret = (uint8x8_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 16); \
19333 #define vqshlu_n_s8(__p0, __p1) __extension__ ({ \
19337   __ret = (uint8x8_t) __builtin_neon_vqshlu_n_v((int8x8_t)__rev0, __p1, 16); \
19344 #define vqshlu_n_s32(__p0, __p1) __extension__ ({ \
19347   __ret = (uint32x2_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 18); \
19351 #define vqshlu_n_s32(__p0, __p1) __extension__ ({ \
19355   __ret = (uint32x2_t) __builtin_neon_vqshlu_n_v((int8x8_t)__rev0, __p1, 18); \
19362 #define vqshlu_n_s64(__p0, __p1) __extension__ ({ \
19365   __ret = (uint64x1_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 19); \
19369 #define vqshlu_n_s64(__p0, __p1) __extension__ ({ \
19372   __ret = (uint64x1_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 19); \
19378 #define vqshlu_n_s16(__p0, __p1) __extension__ ({ \
19381   __ret = (uint16x4_t) __builtin_neon_vqshlu_n_v((int8x8_t)__s0, __p1, 17); \
19385 #define vqshlu_n_s16(__p0, __p1) __extension__ ({ \
19389   __ret = (uint16x4_t) __builtin_neon_vqshlu_n_v((int8x8_t)__rev0, __p1, 17); \
19396 #define vqshrn_n_u32(__p0, __p1) __extension__ ({ \
19399   __ret = (uint16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 17); \
19403 #define vqshrn_n_u32(__p0, __p1) __extension__ ({ \
19407   __ret = (uint16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 17); \
19411 #define __noswap_vqshrn_n_u32(__p0, __p1) __extension__ ({ \
19414   __ret = (uint16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 17); \
19420 #define vqshrn_n_u64(__p0, __p1) __extension__ ({ \
19423   __ret = (uint32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 18); \
19427 #define vqshrn_n_u64(__p0, __p1) __extension__ ({ \
19431   __ret = (uint32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 18); \
19435 #define __noswap_vqshrn_n_u64(__p0, __p1) __extension__ ({ \
19438   __ret = (uint32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 18); \
19444 #define vqshrn_n_u16(__p0, __p1) __extension__ ({ \
19447   __ret = (uint8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 16); \
19451 #define vqshrn_n_u16(__p0, __p1) __extension__ ({ \
19455   __ret = (uint8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 16); \
19459 #define __noswap_vqshrn_n_u16(__p0, __p1) __extension__ ({ \
19462   __ret = (uint8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 16); \
19468 #define vqshrn_n_s32(__p0, __p1) __extension__ ({ \
19471   __ret = (int16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 1); \
19475 #define vqshrn_n_s32(__p0, __p1) __extension__ ({ \
19479   __ret = (int16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 1); \
19483 #define __noswap_vqshrn_n_s32(__p0, __p1) __extension__ ({ \
19486   __ret = (int16x4_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 1); \
19492 #define vqshrn_n_s64(__p0, __p1) __extension__ ({ \
19495   __ret = (int32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 2); \
19499 #define vqshrn_n_s64(__p0, __p1) __extension__ ({ \
19503   __ret = (int32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 2); \
19507 #define __noswap_vqshrn_n_s64(__p0, __p1) __extension__ ({ \
19510   __ret = (int32x2_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 2); \
19516 #define vqshrn_n_s16(__p0, __p1) __extension__ ({ \
19519   __ret = (int8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 0); \
19523 #define vqshrn_n_s16(__p0, __p1) __extension__ ({ \
19527   __ret = (int8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__rev0, __p1, 0); \
19531 #define __noswap_vqshrn_n_s16(__p0, __p1) __extension__ ({ \
19534   __ret = (int8x8_t) __builtin_neon_vqshrn_n_v((int8x16_t)__s0, __p1, 0); \
19540 #define vqshrun_n_s32(__p0, __p1) __extension__ ({ \
19543   __ret = (uint16x4_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 17); \
19547 #define vqshrun_n_s32(__p0, __p1) __extension__ ({ \
19551   __ret = (uint16x4_t) __builtin_neon_vqshrun_n_v((int8x16_t)__rev0, __p1, 17); \
19555 #define __noswap_vqshrun_n_s32(__p0, __p1) __extension__ ({ \
19558   __ret = (uint16x4_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 17); \
19564 #define vqshrun_n_s64(__p0, __p1) __extension__ ({ \
19567   __ret = (uint32x2_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 18); \
19571 #define vqshrun_n_s64(__p0, __p1) __extension__ ({ \
19575   __ret = (uint32x2_t) __builtin_neon_vqshrun_n_v((int8x16_t)__rev0, __p1, 18); \
19579 #define __noswap_vqshrun_n_s64(__p0, __p1) __extension__ ({ \
19582   __ret = (uint32x2_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 18); \
19588 #define vqshrun_n_s16(__p0, __p1) __extension__ ({ \
19591   __ret = (uint8x8_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 16); \
19595 #define vqshrun_n_s16(__p0, __p1) __extension__ ({ \
19599   __ret = (uint8x8_t) __builtin_neon_vqshrun_n_v((int8x16_t)__rev0, __p1, 16); \
19603 #define __noswap_vqshrun_n_s16(__p0, __p1) __extension__ ({ \
19606   __ret = (uint8x8_t) __builtin_neon_vqshrun_n_v((int8x16_t)__s0, __p1, 16); \
19612 __ai uint8x16_t vqsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
19614   __ret = (uint8x16_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
19618 __ai uint8x16_t vqsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
19620   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
19629 __ai uint32x4_t vqsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
19631   __ret = (uint32x4_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
19635 __ai uint32x4_t vqsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
19637   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19646 __ai uint64x2_t vqsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
19648   __ret = (uint64x2_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
19652 __ai uint64x2_t vqsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
19654   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19663 __ai uint16x8_t vqsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
19665   __ret = (uint16x8_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
19669 __ai uint16x8_t vqsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
19671   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19680 __ai int8x16_t vqsubq_s8(int8x16_t __p0, int8x16_t __p1) {
19682   __ret = (int8x16_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
19686 __ai int8x16_t vqsubq_s8(int8x16_t __p0, int8x16_t __p1) {
19688   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
19697 __ai int32x4_t vqsubq_s32(int32x4_t __p0, int32x4_t __p1) {
19699   __ret = (int32x4_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
19703 __ai int32x4_t vqsubq_s32(int32x4_t __p0, int32x4_t __p1) {
19705   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19714 __ai int64x2_t vqsubq_s64(int64x2_t __p0, int64x2_t __p1) {
19716   __ret = (int64x2_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
19720 __ai int64x2_t vqsubq_s64(int64x2_t __p0, int64x2_t __p1) {
19722   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19731 __ai int16x8_t vqsubq_s16(int16x8_t __p0, int16x8_t __p1) {
19733   __ret = (int16x8_t) __builtin_neon_vqsubq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
19737 __ai int16x8_t vqsubq_s16(int16x8_t __p0, int16x8_t __p1) {
19739   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19748 __ai uint8x8_t vqsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
19750   __ret = (uint8x8_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
19754 __ai uint8x8_t vqsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
19756   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19765 __ai uint32x2_t vqsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
19767   __ret = (uint32x2_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
19771 __ai uint32x2_t vqsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
19773   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19782 __ai uint64x1_t vqsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
19784   __ret = (uint64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
19788 __ai uint64x1_t vqsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
19790   __ret = (uint64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
19796 __ai uint16x4_t vqsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
19798   __ret = (uint16x4_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
19802 __ai uint16x4_t vqsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
19804   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19813 __ai int8x8_t vqsub_s8(int8x8_t __p0, int8x8_t __p1) {
19815   __ret = (int8x8_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
19819 __ai int8x8_t vqsub_s8(int8x8_t __p0, int8x8_t __p1) {
19821   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19830 __ai int32x2_t vqsub_s32(int32x2_t __p0, int32x2_t __p1) {
19832   __ret = (int32x2_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
19836 __ai int32x2_t vqsub_s32(int32x2_t __p0, int32x2_t __p1) {
19838   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19847 __ai int64x1_t vqsub_s64(int64x1_t __p0, int64x1_t __p1) {
19849   __ret = (int64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
19853 __ai int64x1_t vqsub_s64(int64x1_t __p0, int64x1_t __p1) {
19855   __ret = (int64x1_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
19861 __ai int16x4_t vqsub_s16(int16x4_t __p0, int16x4_t __p1) {
19863   __ret = (int16x4_t) __builtin_neon_vqsub_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
19867 __ai int16x4_t vqsub_s16(int16x4_t __p0, int16x4_t __p1) {
19869   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19878 __ai uint16x4_t vraddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
19880   __ret = (uint16x4_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
19884 __ai uint16x4_t vraddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
19886   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19892 __ai uint16x4_t __noswap_vraddhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
19894   __ret = (uint16x4_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
19900 __ai uint32x2_t vraddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
19902   __ret = (uint32x2_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
19906 __ai uint32x2_t vraddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
19908   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19914 __ai uint32x2_t __noswap_vraddhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
19916   __ret = (uint32x2_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
19922 __ai uint8x8_t vraddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
19924   __ret = (uint8x8_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
19928 __ai uint8x8_t vraddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
19930   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
19936 __ai uint8x8_t __noswap_vraddhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
19938   __ret = (uint8x8_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
19944 __ai int16x4_t vraddhn_s32(int32x4_t __p0, int32x4_t __p1) {
19946   __ret = (int16x4_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
19950 __ai int16x4_t vraddhn_s32(int32x4_t __p0, int32x4_t __p1) {
19952   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
19958 __ai int16x4_t __noswap_vraddhn_s32(int32x4_t __p0, int32x4_t __p1) {
19960   __ret = (int16x4_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
19966 __ai int32x2_t vraddhn_s64(int64x2_t __p0, int64x2_t __p1) {
19968   __ret = (int32x2_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
19972 __ai int32x2_t vraddhn_s64(int64x2_t __p0, int64x2_t __p1) {
19974   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
19980 __ai int32x2_t __noswap_vraddhn_s64(int64x2_t __p0, int64x2_t __p1) {
19982   __ret = (int32x2_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
19988 __ai int8x8_t vraddhn_s16(int16x8_t __p0, int16x8_t __p1) {
19990   __ret = (int8x8_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
19994 __ai int8x8_t vraddhn_s16(int16x8_t __p0, int16x8_t __p1) {
19996   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20002 __ai int8x8_t __noswap_vraddhn_s16(int16x8_t __p0, int16x8_t __p1) {
20004   __ret = (int8x8_t) __builtin_neon_vraddhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
20074 __ai float32x4_t vrecpsq_f32(float32x4_t __p0, float32x4_t __p1) {
20076   __ret = (float32x4_t) __builtin_neon_vrecpsq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
20080 __ai float32x4_t vrecpsq_f32(float32x4_t __p0, float32x4_t __p1) {
20082   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20091 __ai float32x2_t vrecps_f32(float32x2_t __p0, float32x2_t __p1) {
20093   __ret = (float32x2_t) __builtin_neon_vrecps_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
20097 __ai float32x2_t vrecps_f32(float32x2_t __p0, float32x2_t __p1) {
20099   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20684 __ai uint8x16_t vrhaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
20686   __ret = (uint8x16_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
20690 __ai uint8x16_t vrhaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
20692   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20701 __ai uint32x4_t vrhaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
20703   __ret = (uint32x4_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
20707 __ai uint32x4_t vrhaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
20709   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20718 __ai uint16x8_t vrhaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
20720   __ret = (uint16x8_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
20724 __ai uint16x8_t vrhaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
20726   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20735 __ai int8x16_t vrhaddq_s8(int8x16_t __p0, int8x16_t __p1) {
20737   __ret = (int8x16_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
20741 __ai int8x16_t vrhaddq_s8(int8x16_t __p0, int8x16_t __p1) {
20743   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20752 __ai int32x4_t vrhaddq_s32(int32x4_t __p0, int32x4_t __p1) {
20754   __ret = (int32x4_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
20758 __ai int32x4_t vrhaddq_s32(int32x4_t __p0, int32x4_t __p1) {
20760   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20769 __ai int16x8_t vrhaddq_s16(int16x8_t __p0, int16x8_t __p1) {
20771   __ret = (int16x8_t) __builtin_neon_vrhaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
20775 __ai int16x8_t vrhaddq_s16(int16x8_t __p0, int16x8_t __p1) {
20777   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20786 __ai uint8x8_t vrhadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
20788   __ret = (uint8x8_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
20792 __ai uint8x8_t vrhadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
20794   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20803 __ai uint32x2_t vrhadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
20805   __ret = (uint32x2_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
20809 __ai uint32x2_t vrhadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
20811   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20820 __ai uint16x4_t vrhadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
20822   __ret = (uint16x4_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
20826 __ai uint16x4_t vrhadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
20828   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20837 __ai int8x8_t vrhadd_s8(int8x8_t __p0, int8x8_t __p1) {
20839   __ret = (int8x8_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
20843 __ai int8x8_t vrhadd_s8(int8x8_t __p0, int8x8_t __p1) {
20845   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20854 __ai int32x2_t vrhadd_s32(int32x2_t __p0, int32x2_t __p1) {
20856   __ret = (int32x2_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
20860 __ai int32x2_t vrhadd_s32(int32x2_t __p0, int32x2_t __p1) {
20862   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20871 __ai int16x4_t vrhadd_s16(int16x4_t __p0, int16x4_t __p1) {
20873   __ret = (int16x4_t) __builtin_neon_vrhadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
20877 __ai int16x4_t vrhadd_s16(int16x4_t __p0, int16x4_t __p1) {
20879   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20888 __ai uint8x16_t vrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
20890   __ret = (uint8x16_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
20894 __ai uint8x16_t vrshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
20896   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20905 __ai uint32x4_t vrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
20907   __ret = (uint32x4_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
20911 __ai uint32x4_t vrshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
20913   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20922 __ai uint64x2_t vrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
20924   __ret = (uint64x2_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
20928 __ai uint64x2_t vrshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
20930   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
20939 __ai uint16x8_t vrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
20941   __ret = (uint16x8_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
20945 __ai uint16x8_t vrshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
20947   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
20956 __ai int8x16_t vrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
20958   __ret = (int8x16_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
20962 __ai int8x16_t vrshlq_s8(int8x16_t __p0, int8x16_t __p1) {
20964   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
20973 __ai int32x4_t vrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
20975   __ret = (int32x4_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
20979 __ai int32x4_t vrshlq_s32(int32x4_t __p0, int32x4_t __p1) {
20981   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
20990 __ai int64x2_t vrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
20992   __ret = (int64x2_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
20996 __ai int64x2_t vrshlq_s64(int64x2_t __p0, int64x2_t __p1) {
20998   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21007 __ai int16x8_t vrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
21009   __ret = (int16x8_t) __builtin_neon_vrshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
21013 __ai int16x8_t vrshlq_s16(int16x8_t __p0, int16x8_t __p1) {
21015   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21024 __ai uint8x8_t vrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
21026   __ret = (uint8x8_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
21030 __ai uint8x8_t vrshl_u8(uint8x8_t __p0, int8x8_t __p1) {
21032   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21041 __ai uint32x2_t vrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
21043   __ret = (uint32x2_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
21047 __ai uint32x2_t vrshl_u32(uint32x2_t __p0, int32x2_t __p1) {
21049   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21058 __ai uint64x1_t vrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
21060   __ret = (uint64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
21064 __ai uint64x1_t vrshl_u64(uint64x1_t __p0, int64x1_t __p1) {
21066   __ret = (uint64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
21072 __ai uint16x4_t vrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
21074   __ret = (uint16x4_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
21078 __ai uint16x4_t vrshl_u16(uint16x4_t __p0, int16x4_t __p1) {
21080   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21089 __ai int8x8_t vrshl_s8(int8x8_t __p0, int8x8_t __p1) {
21091   __ret = (int8x8_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
21095 __ai int8x8_t vrshl_s8(int8x8_t __p0, int8x8_t __p1) {
21097   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
21106 __ai int32x2_t vrshl_s32(int32x2_t __p0, int32x2_t __p1) {
21108   __ret = (int32x2_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
21112 __ai int32x2_t vrshl_s32(int32x2_t __p0, int32x2_t __p1) {
21114   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21123 __ai int64x1_t vrshl_s64(int64x1_t __p0, int64x1_t __p1) {
21125   __ret = (int64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
21129 __ai int64x1_t vrshl_s64(int64x1_t __p0, int64x1_t __p1) {
21131   __ret = (int64x1_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
21137 __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
21139   __ret = (int16x4_t) __builtin_neon_vrshl_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
21143 __ai int16x4_t vrshl_s16(int16x4_t __p0, int16x4_t __p1) {
21145   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21154 #define vrshrq_n_u8(__p0, __p1) __extension__ ({ \
21157   __ret = (uint8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 48); \
21161 #define vrshrq_n_u8(__p0, __p1) __extension__ ({ \
21165   __ret = (uint8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 48); \
21172 #define vrshrq_n_u32(__p0, __p1) __extension__ ({ \
21175   __ret = (uint32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 50); \
21179 #define vrshrq_n_u32(__p0, __p1) __extension__ ({ \
21183   __ret = (uint32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 50); \
21190 #define vrshrq_n_u64(__p0, __p1) __extension__ ({ \
21193   __ret = (uint64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 51); \
21197 #define vrshrq_n_u64(__p0, __p1) __extension__ ({ \
21201   __ret = (uint64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 51); \
21208 #define vrshrq_n_u16(__p0, __p1) __extension__ ({ \
21211   __ret = (uint16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 49); \
21215 #define vrshrq_n_u16(__p0, __p1) __extension__ ({ \
21219   __ret = (uint16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 49); \
21226 #define vrshrq_n_s8(__p0, __p1) __extension__ ({ \
21229   __ret = (int8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 32); \
21233 #define vrshrq_n_s8(__p0, __p1) __extension__ ({ \
21237   __ret = (int8x16_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 32); \
21244 #define vrshrq_n_s32(__p0, __p1) __extension__ ({ \
21247   __ret = (int32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 34); \
21251 #define vrshrq_n_s32(__p0, __p1) __extension__ ({ \
21255   __ret = (int32x4_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 34); \
21262 #define vrshrq_n_s64(__p0, __p1) __extension__ ({ \
21265   __ret = (int64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 35); \
21269 #define vrshrq_n_s64(__p0, __p1) __extension__ ({ \
21273   __ret = (int64x2_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 35); \
21280 #define vrshrq_n_s16(__p0, __p1) __extension__ ({ \
21283   __ret = (int16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__s0, __p1, 33); \
21287 #define vrshrq_n_s16(__p0, __p1) __extension__ ({ \
21291   __ret = (int16x8_t) __builtin_neon_vrshrq_n_v((int8x16_t)__rev0, __p1, 33); \
21298 #define vrshr_n_u8(__p0, __p1) __extension__ ({ \
21301   __ret = (uint8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 16); \
21305 #define vrshr_n_u8(__p0, __p1) __extension__ ({ \
21309   __ret = (uint8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 16); \
21316 #define vrshr_n_u32(__p0, __p1) __extension__ ({ \
21319   __ret = (uint32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 18); \
21323 #define vrshr_n_u32(__p0, __p1) __extension__ ({ \
21327   __ret = (uint32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 18); \
21334 #define vrshr_n_u64(__p0, __p1) __extension__ ({ \
21337   __ret = (uint64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 19); \
21341 #define vrshr_n_u64(__p0, __p1) __extension__ ({ \
21344   __ret = (uint64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 19); \
21350 #define vrshr_n_u16(__p0, __p1) __extension__ ({ \
21353   __ret = (uint16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 17); \
21357 #define vrshr_n_u16(__p0, __p1) __extension__ ({ \
21361   __ret = (uint16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 17); \
21368 #define vrshr_n_s8(__p0, __p1) __extension__ ({ \
21371   __ret = (int8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 0); \
21375 #define vrshr_n_s8(__p0, __p1) __extension__ ({ \
21379   __ret = (int8x8_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 0); \
21386 #define vrshr_n_s32(__p0, __p1) __extension__ ({ \
21389   __ret = (int32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 2); \
21393 #define vrshr_n_s32(__p0, __p1) __extension__ ({ \
21397   __ret = (int32x2_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 2); \
21404 #define vrshr_n_s64(__p0, __p1) __extension__ ({ \
21407   __ret = (int64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 3); \
21411 #define vrshr_n_s64(__p0, __p1) __extension__ ({ \
21414   __ret = (int64x1_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 3); \
21420 #define vrshr_n_s16(__p0, __p1) __extension__ ({ \
21423   __ret = (int16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__s0, __p1, 1); \
21427 #define vrshr_n_s16(__p0, __p1) __extension__ ({ \
21431   __ret = (int16x4_t) __builtin_neon_vrshr_n_v((int8x8_t)__rev0, __p1, 1); \
21438 #define vrshrn_n_u32(__p0, __p1) __extension__ ({ \
21441   __ret = (uint16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 17); \
21445 #define vrshrn_n_u32(__p0, __p1) __extension__ ({ \
21449   __ret = (uint16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 17); \
21453 #define __noswap_vrshrn_n_u32(__p0, __p1) __extension__ ({ \
21456   __ret = (uint16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 17); \
21462 #define vrshrn_n_u64(__p0, __p1) __extension__ ({ \
21465   __ret = (uint32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 18); \
21469 #define vrshrn_n_u64(__p0, __p1) __extension__ ({ \
21473   __ret = (uint32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 18); \
21477 #define __noswap_vrshrn_n_u64(__p0, __p1) __extension__ ({ \
21480   __ret = (uint32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 18); \
21486 #define vrshrn_n_u16(__p0, __p1) __extension__ ({ \
21489   __ret = (uint8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 16); \
21493 #define vrshrn_n_u16(__p0, __p1) __extension__ ({ \
21497   __ret = (uint8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 16); \
21501 #define __noswap_vrshrn_n_u16(__p0, __p1) __extension__ ({ \
21504   __ret = (uint8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 16); \
21510 #define vrshrn_n_s32(__p0, __p1) __extension__ ({ \
21513   __ret = (int16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 1); \
21517 #define vrshrn_n_s32(__p0, __p1) __extension__ ({ \
21521   __ret = (int16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 1); \
21525 #define __noswap_vrshrn_n_s32(__p0, __p1) __extension__ ({ \
21528   __ret = (int16x4_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 1); \
21534 #define vrshrn_n_s64(__p0, __p1) __extension__ ({ \
21537   __ret = (int32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 2); \
21541 #define vrshrn_n_s64(__p0, __p1) __extension__ ({ \
21545   __ret = (int32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 2); \
21549 #define __noswap_vrshrn_n_s64(__p0, __p1) __extension__ ({ \
21552   __ret = (int32x2_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 2); \
21558 #define vrshrn_n_s16(__p0, __p1) __extension__ ({ \
21561   __ret = (int8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 0); \
21565 #define vrshrn_n_s16(__p0, __p1) __extension__ ({ \
21569   __ret = (int8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__rev0, __p1, 0); \
21573 #define __noswap_vrshrn_n_s16(__p0, __p1) __extension__ ({ \
21576   __ret = (int8x8_t) __builtin_neon_vrshrn_n_v((int8x16_t)__s0, __p1, 0); \
21646 __ai float32x4_t vrsqrtsq_f32(float32x4_t __p0, float32x4_t __p1) {
21648   __ret = (float32x4_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
21652 __ai float32x4_t vrsqrtsq_f32(float32x4_t __p0, float32x4_t __p1) {
21654   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
21663 __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
21665   __ret = (float32x2_t) __builtin_neon_vrsqrts_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
21669 __ai float32x2_t vrsqrts_f32(float32x2_t __p0, float32x2_t __p1) {
21671   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
21680 #define vrsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
21682   uint8x16_t __s1 = __p1; \
21688 #define vrsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
21690   uint8x16_t __s1 = __p1; \
21701 #define vrsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
21703   uint32x4_t __s1 = __p1; \
21709 #define vrsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
21711   uint32x4_t __s1 = __p1; \
21722 #define vrsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
21724   uint64x2_t __s1 = __p1; \
21730 #define vrsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
21732   uint64x2_t __s1 = __p1; \
21743 #define vrsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
21745   uint16x8_t __s1 = __p1; \
21751 #define vrsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
21753   uint16x8_t __s1 = __p1; \
21764 #define vrsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
21766   int8x16_t __s1 = __p1; \
21772 #define vrsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
21774   int8x16_t __s1 = __p1; \
21785 #define vrsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
21787   int32x4_t __s1 = __p1; \
21793 #define vrsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
21795   int32x4_t __s1 = __p1; \
21806 #define vrsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
21808   int64x2_t __s1 = __p1; \
21814 #define vrsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
21816   int64x2_t __s1 = __p1; \
21827 #define vrsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
21829   int16x8_t __s1 = __p1; \
21835 #define vrsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
21837   int16x8_t __s1 = __p1; \
21848 #define vrsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
21850   uint8x8_t __s1 = __p1; \
21856 #define vrsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
21858   uint8x8_t __s1 = __p1; \
21869 #define vrsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
21871   uint32x2_t __s1 = __p1; \
21877 #define vrsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
21879   uint32x2_t __s1 = __p1; \
21890 #define vrsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
21892   uint64x1_t __s1 = __p1; \
21898 #define vrsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
21900   uint64x1_t __s1 = __p1; \
21908 #define vrsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
21910   uint16x4_t __s1 = __p1; \
21916 #define vrsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
21918   uint16x4_t __s1 = __p1; \
21929 #define vrsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
21931   int8x8_t __s1 = __p1; \
21937 #define vrsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
21939   int8x8_t __s1 = __p1; \
21950 #define vrsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
21952   int32x2_t __s1 = __p1; \
21958 #define vrsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
21960   int32x2_t __s1 = __p1; \
21971 #define vrsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
21973   int64x1_t __s1 = __p1; \
21979 #define vrsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
21981   int64x1_t __s1 = __p1; \
21989 #define vrsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
21991   int16x4_t __s1 = __p1; \
21997 #define vrsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
21999   int16x4_t __s1 = __p1; \
22010 __ai uint16x4_t vrsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
22012   __ret = (uint16x4_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
22016 __ai uint16x4_t vrsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
22018   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22024 __ai uint16x4_t __noswap_vrsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
22026   __ret = (uint16x4_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
22032 __ai uint32x2_t vrsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
22034   __ret = (uint32x2_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
22038 __ai uint32x2_t vrsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
22040   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22046 __ai uint32x2_t __noswap_vrsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
22048   __ret = (uint32x2_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
22054 __ai uint8x8_t vrsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
22056   __ret = (uint8x8_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
22060 __ai uint8x8_t vrsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
22062   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22068 __ai uint8x8_t __noswap_vrsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
22070   __ret = (uint8x8_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
22076 __ai int16x4_t vrsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
22078   __ret = (int16x4_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
22082 __ai int16x4_t vrsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
22084   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22090 __ai int16x4_t __noswap_vrsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
22092   __ret = (int16x4_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
22098 __ai int32x2_t vrsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
22100   __ret = (int32x2_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
22104 __ai int32x2_t vrsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
22106   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22112 __ai int32x2_t __noswap_vrsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
22114   __ret = (int32x2_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
22120 __ai int8x8_t vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
22122   __ret = (int8x8_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
22126 __ai int8x8_t vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
22128   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22134 __ai int8x8_t __noswap_vrsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
22136   __ret = (int8x8_t) __builtin_neon_vrsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
22142 #define vset_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22144   poly8x8_t __s1 = __p1; \
22150 #define vset_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22152   poly8x8_t __s1 = __p1; \
22159 #define __noswap_vset_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22161   poly8x8_t __s1 = __p1; \
22169 #define vset_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22171   poly16x4_t __s1 = __p1; \
22177 #define vset_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22179   poly16x4_t __s1 = __p1; \
22186 #define __noswap_vset_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22188   poly16x4_t __s1 = __p1; \
22196 #define vsetq_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22198   poly8x16_t __s1 = __p1; \
22204 #define vsetq_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22206   poly8x16_t __s1 = __p1; \
22213 #define __noswap_vsetq_lane_p8(__p0, __p1, __p2) __extension__ ({ \
22215   poly8x16_t __s1 = __p1; \
22223 #define vsetq_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22225   poly16x8_t __s1 = __p1; \
22231 #define vsetq_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22233   poly16x8_t __s1 = __p1; \
22240 #define __noswap_vsetq_lane_p16(__p0, __p1, __p2) __extension__ ({ \
22242   poly16x8_t __s1 = __p1; \
22250 #define vsetq_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22252   uint8x16_t __s1 = __p1; \
22258 #define vsetq_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22260   uint8x16_t __s1 = __p1; \
22267 #define __noswap_vsetq_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22269   uint8x16_t __s1 = __p1; \
22277 #define vsetq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22279   uint32x4_t __s1 = __p1; \
22285 #define vsetq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22287   uint32x4_t __s1 = __p1; \
22294 #define __noswap_vsetq_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22296   uint32x4_t __s1 = __p1; \
22304 #define vsetq_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22306   uint64x2_t __s1 = __p1; \
22312 #define vsetq_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22314   uint64x2_t __s1 = __p1; \
22321 #define __noswap_vsetq_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22323   uint64x2_t __s1 = __p1; \
22331 #define vsetq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22333   uint16x8_t __s1 = __p1; \
22339 #define vsetq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22341   uint16x8_t __s1 = __p1; \
22348 #define __noswap_vsetq_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22350   uint16x8_t __s1 = __p1; \
22358 #define vsetq_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22360   int8x16_t __s1 = __p1; \
22366 #define vsetq_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22368   int8x16_t __s1 = __p1; \
22375 #define __noswap_vsetq_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22377   int8x16_t __s1 = __p1; \
22385 #define vsetq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22387   float32x4_t __s1 = __p1; \
22393 #define vsetq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22395   float32x4_t __s1 = __p1; \
22402 #define __noswap_vsetq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22404   float32x4_t __s1 = __p1; \
22412 #define vsetq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22414   int32x4_t __s1 = __p1; \
22420 #define vsetq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22422   int32x4_t __s1 = __p1; \
22429 #define __noswap_vsetq_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22431   int32x4_t __s1 = __p1; \
22439 #define vsetq_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22441   int64x2_t __s1 = __p1; \
22447 #define vsetq_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22449   int64x2_t __s1 = __p1; \
22456 #define __noswap_vsetq_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22458   int64x2_t __s1 = __p1; \
22466 #define vsetq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22468   int16x8_t __s1 = __p1; \
22474 #define vsetq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22476   int16x8_t __s1 = __p1; \
22483 #define __noswap_vsetq_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22485   int16x8_t __s1 = __p1; \
22493 #define vset_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22495   uint8x8_t __s1 = __p1; \
22501 #define vset_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22503   uint8x8_t __s1 = __p1; \
22510 #define __noswap_vset_lane_u8(__p0, __p1, __p2) __extension__ ({ \
22512   uint8x8_t __s1 = __p1; \
22520 #define vset_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22522   uint32x2_t __s1 = __p1; \
22528 #define vset_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22530   uint32x2_t __s1 = __p1; \
22537 #define __noswap_vset_lane_u32(__p0, __p1, __p2) __extension__ ({ \
22539   uint32x2_t __s1 = __p1; \
22547 #define vset_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22549   uint64x1_t __s1 = __p1; \
22555 #define vset_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22557   uint64x1_t __s1 = __p1; \
22562 #define __noswap_vset_lane_u64(__p0, __p1, __p2) __extension__ ({ \
22564   uint64x1_t __s1 = __p1; \
22572 #define vset_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22574   uint16x4_t __s1 = __p1; \
22580 #define vset_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22582   uint16x4_t __s1 = __p1; \
22589 #define __noswap_vset_lane_u16(__p0, __p1, __p2) __extension__ ({ \
22591   uint16x4_t __s1 = __p1; \
22599 #define vset_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22601   int8x8_t __s1 = __p1; \
22607 #define vset_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22609   int8x8_t __s1 = __p1; \
22616 #define __noswap_vset_lane_s8(__p0, __p1, __p2) __extension__ ({ \
22618   int8x8_t __s1 = __p1; \
22626 #define vset_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22628   float32x2_t __s1 = __p1; \
22634 #define vset_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22636   float32x2_t __s1 = __p1; \
22643 #define __noswap_vset_lane_f32(__p0, __p1, __p2) __extension__ ({ \
22645   float32x2_t __s1 = __p1; \
22653 #define vset_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22655   int32x2_t __s1 = __p1; \
22661 #define vset_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22663   int32x2_t __s1 = __p1; \
22670 #define __noswap_vset_lane_s32(__p0, __p1, __p2) __extension__ ({ \
22672   int32x2_t __s1 = __p1; \
22680 #define vset_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22682   int64x1_t __s1 = __p1; \
22688 #define vset_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22690   int64x1_t __s1 = __p1; \
22695 #define __noswap_vset_lane_s64(__p0, __p1, __p2) __extension__ ({ \
22697   int64x1_t __s1 = __p1; \
22705 #define vset_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22707   int16x4_t __s1 = __p1; \
22713 #define vset_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22715   int16x4_t __s1 = __p1; \
22722 #define __noswap_vset_lane_s16(__p0, __p1, __p2) __extension__ ({ \
22724   int16x4_t __s1 = __p1; \
22732 __ai uint8x16_t vshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
22734   __ret = (uint8x16_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
22738 __ai uint8x16_t vshlq_u8(uint8x16_t __p0, int8x16_t __p1) {
22740   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
22749 __ai uint32x4_t vshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
22751   __ret = (uint32x4_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
22755 __ai uint32x4_t vshlq_u32(uint32x4_t __p0, int32x4_t __p1) {
22757   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22766 __ai uint64x2_t vshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
22768   __ret = (uint64x2_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
22772 __ai uint64x2_t vshlq_u64(uint64x2_t __p0, int64x2_t __p1) {
22774   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22783 __ai uint16x8_t vshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
22785   __ret = (uint16x8_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
22789 __ai uint16x8_t vshlq_u16(uint16x8_t __p0, int16x8_t __p1) {
22791   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22800 __ai int8x16_t vshlq_s8(int8x16_t __p0, int8x16_t __p1) {
22802   __ret = (int8x16_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
22806 __ai int8x16_t vshlq_s8(int8x16_t __p0, int8x16_t __p1) {
22808   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
22817 __ai int32x4_t vshlq_s32(int32x4_t __p0, int32x4_t __p1) {
22819   __ret = (int32x4_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
22823 __ai int32x4_t vshlq_s32(int32x4_t __p0, int32x4_t __p1) {
22825   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22834 __ai int64x2_t vshlq_s64(int64x2_t __p0, int64x2_t __p1) {
22836   __ret = (int64x2_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
22840 __ai int64x2_t vshlq_s64(int64x2_t __p0, int64x2_t __p1) {
22842   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22851 __ai int16x8_t vshlq_s16(int16x8_t __p0, int16x8_t __p1) {
22853   __ret = (int16x8_t) __builtin_neon_vshlq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
22857 __ai int16x8_t vshlq_s16(int16x8_t __p0, int16x8_t __p1) {
22859   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22868 __ai uint8x8_t vshl_u8(uint8x8_t __p0, int8x8_t __p1) {
22870   __ret = (uint8x8_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
22874 __ai uint8x8_t vshl_u8(uint8x8_t __p0, int8x8_t __p1) {
22876   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22885 __ai uint32x2_t vshl_u32(uint32x2_t __p0, int32x2_t __p1) {
22887   __ret = (uint32x2_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
22891 __ai uint32x2_t vshl_u32(uint32x2_t __p0, int32x2_t __p1) {
22893   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22902 __ai uint64x1_t vshl_u64(uint64x1_t __p0, int64x1_t __p1) {
22904   __ret = (uint64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
22908 __ai uint64x1_t vshl_u64(uint64x1_t __p0, int64x1_t __p1) {
22910   __ret = (uint64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
22916 __ai uint16x4_t vshl_u16(uint16x4_t __p0, int16x4_t __p1) {
22918   __ret = (uint16x4_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
22922 __ai uint16x4_t vshl_u16(uint16x4_t __p0, int16x4_t __p1) {
22924   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22933 __ai int8x8_t vshl_s8(int8x8_t __p0, int8x8_t __p1) {
22935   __ret = (int8x8_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
22939 __ai int8x8_t vshl_s8(int8x8_t __p0, int8x8_t __p1) {
22941   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
22950 __ai int32x2_t vshl_s32(int32x2_t __p0, int32x2_t __p1) {
22952   __ret = (int32x2_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
22956 __ai int32x2_t vshl_s32(int32x2_t __p0, int32x2_t __p1) {
22958   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
22967 __ai int64x1_t vshl_s64(int64x1_t __p0, int64x1_t __p1) {
22969   __ret = (int64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
22973 __ai int64x1_t vshl_s64(int64x1_t __p0, int64x1_t __p1) {
22975   __ret = (int64x1_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
22981 __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
22983   __ret = (int16x4_t) __builtin_neon_vshl_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
22987 __ai int16x4_t vshl_s16(int16x4_t __p0, int16x4_t __p1) {
22989   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
22998 #define vshlq_n_u8(__p0, __p1) __extension__ ({ \
23001   __ret = (uint8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 48); \
23005 #define vshlq_n_u8(__p0, __p1) __extension__ ({ \
23009   __ret = (uint8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 48); \
23016 #define vshlq_n_u32(__p0, __p1) __extension__ ({ \
23019   __ret = (uint32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 50); \
23023 #define vshlq_n_u32(__p0, __p1) __extension__ ({ \
23027   __ret = (uint32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 50); \
23034 #define vshlq_n_u64(__p0, __p1) __extension__ ({ \
23037   __ret = (uint64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 51); \
23041 #define vshlq_n_u64(__p0, __p1) __extension__ ({ \
23045   __ret = (uint64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 51); \
23052 #define vshlq_n_u16(__p0, __p1) __extension__ ({ \
23055   __ret = (uint16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 49); \
23059 #define vshlq_n_u16(__p0, __p1) __extension__ ({ \
23063   __ret = (uint16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 49); \
23070 #define vshlq_n_s8(__p0, __p1) __extension__ ({ \
23073   __ret = (int8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 32); \
23077 #define vshlq_n_s8(__p0, __p1) __extension__ ({ \
23081   __ret = (int8x16_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 32); \
23088 #define vshlq_n_s32(__p0, __p1) __extension__ ({ \
23091   __ret = (int32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 34); \
23095 #define vshlq_n_s32(__p0, __p1) __extension__ ({ \
23099   __ret = (int32x4_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 34); \
23106 #define vshlq_n_s64(__p0, __p1) __extension__ ({ \
23109   __ret = (int64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 35); \
23113 #define vshlq_n_s64(__p0, __p1) __extension__ ({ \
23117   __ret = (int64x2_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 35); \
23124 #define vshlq_n_s16(__p0, __p1) __extension__ ({ \
23127   __ret = (int16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__s0, __p1, 33); \
23131 #define vshlq_n_s16(__p0, __p1) __extension__ ({ \
23135   __ret = (int16x8_t) __builtin_neon_vshlq_n_v((int8x16_t)__rev0, __p1, 33); \
23142 #define vshl_n_u8(__p0, __p1) __extension__ ({ \
23145   __ret = (uint8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 16); \
23149 #define vshl_n_u8(__p0, __p1) __extension__ ({ \
23153   __ret = (uint8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 16); \
23160 #define vshl_n_u32(__p0, __p1) __extension__ ({ \
23163   __ret = (uint32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 18); \
23167 #define vshl_n_u32(__p0, __p1) __extension__ ({ \
23171   __ret = (uint32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 18); \
23178 #define vshl_n_u64(__p0, __p1) __extension__ ({ \
23181   __ret = (uint64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 19); \
23185 #define vshl_n_u64(__p0, __p1) __extension__ ({ \
23188   __ret = (uint64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 19); \
23194 #define vshl_n_u16(__p0, __p1) __extension__ ({ \
23197   __ret = (uint16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 17); \
23201 #define vshl_n_u16(__p0, __p1) __extension__ ({ \
23205   __ret = (uint16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 17); \
23212 #define vshl_n_s8(__p0, __p1) __extension__ ({ \
23215   __ret = (int8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 0); \
23219 #define vshl_n_s8(__p0, __p1) __extension__ ({ \
23223   __ret = (int8x8_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 0); \
23230 #define vshl_n_s32(__p0, __p1) __extension__ ({ \
23233   __ret = (int32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 2); \
23237 #define vshl_n_s32(__p0, __p1) __extension__ ({ \
23241   __ret = (int32x2_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 2); \
23248 #define vshl_n_s64(__p0, __p1) __extension__ ({ \
23251   __ret = (int64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 3); \
23255 #define vshl_n_s64(__p0, __p1) __extension__ ({ \
23258   __ret = (int64x1_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 3); \
23264 #define vshl_n_s16(__p0, __p1) __extension__ ({ \
23267   __ret = (int16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__s0, __p1, 1); \
23271 #define vshl_n_s16(__p0, __p1) __extension__ ({ \
23275   __ret = (int16x4_t) __builtin_neon_vshl_n_v((int8x8_t)__rev0, __p1, 1); \
23282 #define vshll_n_u8(__p0, __p1) __extension__ ({ \
23285   __ret = (uint16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 49); \
23289 #define vshll_n_u8(__p0, __p1) __extension__ ({ \
23293   __ret = (uint16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 49); \
23297 #define __noswap_vshll_n_u8(__p0, __p1) __extension__ ({ \
23300   __ret = (uint16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 49); \
23306 #define vshll_n_u32(__p0, __p1) __extension__ ({ \
23309   __ret = (uint64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 51); \
23313 #define vshll_n_u32(__p0, __p1) __extension__ ({ \
23317   __ret = (uint64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 51); \
23321 #define __noswap_vshll_n_u32(__p0, __p1) __extension__ ({ \
23324   __ret = (uint64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 51); \
23330 #define vshll_n_u16(__p0, __p1) __extension__ ({ \
23333   __ret = (uint32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 50); \
23337 #define vshll_n_u16(__p0, __p1) __extension__ ({ \
23341   __ret = (uint32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 50); \
23345 #define __noswap_vshll_n_u16(__p0, __p1) __extension__ ({ \
23348   __ret = (uint32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 50); \
23354 #define vshll_n_s8(__p0, __p1) __extension__ ({ \
23357   __ret = (int16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 33); \
23361 #define vshll_n_s8(__p0, __p1) __extension__ ({ \
23365   __ret = (int16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 33); \
23369 #define __noswap_vshll_n_s8(__p0, __p1) __extension__ ({ \
23372   __ret = (int16x8_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 33); \
23378 #define vshll_n_s32(__p0, __p1) __extension__ ({ \
23381   __ret = (int64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 35); \
23385 #define vshll_n_s32(__p0, __p1) __extension__ ({ \
23389   __ret = (int64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 35); \
23393 #define __noswap_vshll_n_s32(__p0, __p1) __extension__ ({ \
23396   __ret = (int64x2_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 35); \
23402 #define vshll_n_s16(__p0, __p1) __extension__ ({ \
23405   __ret = (int32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 34); \
23409 #define vshll_n_s16(__p0, __p1) __extension__ ({ \
23413   __ret = (int32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__rev0, __p1, 34); \
23417 #define __noswap_vshll_n_s16(__p0, __p1) __extension__ ({ \
23420   __ret = (int32x4_t) __builtin_neon_vshll_n_v((int8x8_t)__s0, __p1, 34); \
23426 #define vshrq_n_u8(__p0, __p1) __extension__ ({ \
23429   __ret = (uint8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 48); \
23433 #define vshrq_n_u8(__p0, __p1) __extension__ ({ \
23437   __ret = (uint8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 48); \
23444 #define vshrq_n_u32(__p0, __p1) __extension__ ({ \
23447   __ret = (uint32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 50); \
23451 #define vshrq_n_u32(__p0, __p1) __extension__ ({ \
23455   __ret = (uint32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 50); \
23462 #define vshrq_n_u64(__p0, __p1) __extension__ ({ \
23465   __ret = (uint64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 51); \
23469 #define vshrq_n_u64(__p0, __p1) __extension__ ({ \
23473   __ret = (uint64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 51); \
23480 #define vshrq_n_u16(__p0, __p1) __extension__ ({ \
23483   __ret = (uint16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 49); \
23487 #define vshrq_n_u16(__p0, __p1) __extension__ ({ \
23491   __ret = (uint16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 49); \
23498 #define vshrq_n_s8(__p0, __p1) __extension__ ({ \
23501   __ret = (int8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 32); \
23505 #define vshrq_n_s8(__p0, __p1) __extension__ ({ \
23509   __ret = (int8x16_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 32); \
23516 #define vshrq_n_s32(__p0, __p1) __extension__ ({ \
23519   __ret = (int32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 34); \
23523 #define vshrq_n_s32(__p0, __p1) __extension__ ({ \
23527   __ret = (int32x4_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 34); \
23534 #define vshrq_n_s64(__p0, __p1) __extension__ ({ \
23537   __ret = (int64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 35); \
23541 #define vshrq_n_s64(__p0, __p1) __extension__ ({ \
23545   __ret = (int64x2_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 35); \
23552 #define vshrq_n_s16(__p0, __p1) __extension__ ({ \
23555   __ret = (int16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__s0, __p1, 33); \
23559 #define vshrq_n_s16(__p0, __p1) __extension__ ({ \
23563   __ret = (int16x8_t) __builtin_neon_vshrq_n_v((int8x16_t)__rev0, __p1, 33); \
23570 #define vshr_n_u8(__p0, __p1) __extension__ ({ \
23573   __ret = (uint8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 16); \
23577 #define vshr_n_u8(__p0, __p1) __extension__ ({ \
23581   __ret = (uint8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 16); \
23588 #define vshr_n_u32(__p0, __p1) __extension__ ({ \
23591   __ret = (uint32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 18); \
23595 #define vshr_n_u32(__p0, __p1) __extension__ ({ \
23599   __ret = (uint32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 18); \
23606 #define vshr_n_u64(__p0, __p1) __extension__ ({ \
23609   __ret = (uint64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 19); \
23613 #define vshr_n_u64(__p0, __p1) __extension__ ({ \
23616   __ret = (uint64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 19); \
23622 #define vshr_n_u16(__p0, __p1) __extension__ ({ \
23625   __ret = (uint16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 17); \
23629 #define vshr_n_u16(__p0, __p1) __extension__ ({ \
23633   __ret = (uint16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 17); \
23640 #define vshr_n_s8(__p0, __p1) __extension__ ({ \
23643   __ret = (int8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 0); \
23647 #define vshr_n_s8(__p0, __p1) __extension__ ({ \
23651   __ret = (int8x8_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 0); \
23658 #define vshr_n_s32(__p0, __p1) __extension__ ({ \
23661   __ret = (int32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 2); \
23665 #define vshr_n_s32(__p0, __p1) __extension__ ({ \
23669   __ret = (int32x2_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 2); \
23676 #define vshr_n_s64(__p0, __p1) __extension__ ({ \
23679   __ret = (int64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 3); \
23683 #define vshr_n_s64(__p0, __p1) __extension__ ({ \
23686   __ret = (int64x1_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 3); \
23692 #define vshr_n_s16(__p0, __p1) __extension__ ({ \
23695   __ret = (int16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__s0, __p1, 1); \
23699 #define vshr_n_s16(__p0, __p1) __extension__ ({ \
23703   __ret = (int16x4_t) __builtin_neon_vshr_n_v((int8x8_t)__rev0, __p1, 1); \
23710 #define vshrn_n_u32(__p0, __p1) __extension__ ({ \
23713   __ret = (uint16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 17); \
23717 #define vshrn_n_u32(__p0, __p1) __extension__ ({ \
23721   __ret = (uint16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 17); \
23725 #define __noswap_vshrn_n_u32(__p0, __p1) __extension__ ({ \
23728   __ret = (uint16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 17); \
23734 #define vshrn_n_u64(__p0, __p1) __extension__ ({ \
23737   __ret = (uint32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 18); \
23741 #define vshrn_n_u64(__p0, __p1) __extension__ ({ \
23745   __ret = (uint32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 18); \
23749 #define __noswap_vshrn_n_u64(__p0, __p1) __extension__ ({ \
23752   __ret = (uint32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 18); \
23758 #define vshrn_n_u16(__p0, __p1) __extension__ ({ \
23761   __ret = (uint8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 16); \
23765 #define vshrn_n_u16(__p0, __p1) __extension__ ({ \
23769   __ret = (uint8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 16); \
23773 #define __noswap_vshrn_n_u16(__p0, __p1) __extension__ ({ \
23776   __ret = (uint8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 16); \
23782 #define vshrn_n_s32(__p0, __p1) __extension__ ({ \
23785   __ret = (int16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 1); \
23789 #define vshrn_n_s32(__p0, __p1) __extension__ ({ \
23793   __ret = (int16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 1); \
23797 #define __noswap_vshrn_n_s32(__p0, __p1) __extension__ ({ \
23800   __ret = (int16x4_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 1); \
23806 #define vshrn_n_s64(__p0, __p1) __extension__ ({ \
23809   __ret = (int32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 2); \
23813 #define vshrn_n_s64(__p0, __p1) __extension__ ({ \
23817   __ret = (int32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 2); \
23821 #define __noswap_vshrn_n_s64(__p0, __p1) __extension__ ({ \
23824   __ret = (int32x2_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 2); \
23830 #define vshrn_n_s16(__p0, __p1) __extension__ ({ \
23833   __ret = (int8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 0); \
23837 #define vshrn_n_s16(__p0, __p1) __extension__ ({ \
23841   __ret = (int8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__rev0, __p1, 0); \
23845 #define __noswap_vshrn_n_s16(__p0, __p1) __extension__ ({ \
23848   __ret = (int8x8_t) __builtin_neon_vshrn_n_v((int8x16_t)__s0, __p1, 0); \
23854 #define vsli_n_p8(__p0, __p1, __p2) __extension__ ({ \
23856   poly8x8_t __s1 = __p1; \
23862 #define vsli_n_p8(__p0, __p1, __p2) __extension__ ({ \
23864   poly8x8_t __s1 = __p1; \
23875 #define vsli_n_p16(__p0, __p1, __p2) __extension__ ({ \
23877   poly16x4_t __s1 = __p1; \
23883 #define vsli_n_p16(__p0, __p1, __p2) __extension__ ({ \
23885   poly16x4_t __s1 = __p1; \
23896 #define vsliq_n_p8(__p0, __p1, __p2) __extension__ ({ \
23898   poly8x16_t __s1 = __p1; \
23904 #define vsliq_n_p8(__p0, __p1, __p2) __extension__ ({ \
23906   poly8x16_t __s1 = __p1; \
23917 #define vsliq_n_p16(__p0, __p1, __p2) __extension__ ({ \
23919   poly16x8_t __s1 = __p1; \
23925 #define vsliq_n_p16(__p0, __p1, __p2) __extension__ ({ \
23927   poly16x8_t __s1 = __p1; \
23938 #define vsliq_n_u8(__p0, __p1, __p2) __extension__ ({ \
23940   uint8x16_t __s1 = __p1; \
23946 #define vsliq_n_u8(__p0, __p1, __p2) __extension__ ({ \
23948   uint8x16_t __s1 = __p1; \
23959 #define vsliq_n_u32(__p0, __p1, __p2) __extension__ ({ \
23961   uint32x4_t __s1 = __p1; \
23967 #define vsliq_n_u32(__p0, __p1, __p2) __extension__ ({ \
23969   uint32x4_t __s1 = __p1; \
23980 #define vsliq_n_u64(__p0, __p1, __p2) __extension__ ({ \
23982   uint64x2_t __s1 = __p1; \
23988 #define vsliq_n_u64(__p0, __p1, __p2) __extension__ ({ \
23990   uint64x2_t __s1 = __p1; \
24001 #define vsliq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24003   uint16x8_t __s1 = __p1; \
24009 #define vsliq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24011   uint16x8_t __s1 = __p1; \
24022 #define vsliq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24024   int8x16_t __s1 = __p1; \
24030 #define vsliq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24032   int8x16_t __s1 = __p1; \
24043 #define vsliq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24045   int32x4_t __s1 = __p1; \
24051 #define vsliq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24053   int32x4_t __s1 = __p1; \
24064 #define vsliq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24066   int64x2_t __s1 = __p1; \
24072 #define vsliq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24074   int64x2_t __s1 = __p1; \
24085 #define vsliq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24087   int16x8_t __s1 = __p1; \
24093 #define vsliq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24095   int16x8_t __s1 = __p1; \
24106 #define vsli_n_u8(__p0, __p1, __p2) __extension__ ({ \
24108   uint8x8_t __s1 = __p1; \
24114 #define vsli_n_u8(__p0, __p1, __p2) __extension__ ({ \
24116   uint8x8_t __s1 = __p1; \
24127 #define vsli_n_u32(__p0, __p1, __p2) __extension__ ({ \
24129   uint32x2_t __s1 = __p1; \
24135 #define vsli_n_u32(__p0, __p1, __p2) __extension__ ({ \
24137   uint32x2_t __s1 = __p1; \
24148 #define vsli_n_u64(__p0, __p1, __p2) __extension__ ({ \
24150   uint64x1_t __s1 = __p1; \
24156 #define vsli_n_u64(__p0, __p1, __p2) __extension__ ({ \
24158   uint64x1_t __s1 = __p1; \
24166 #define vsli_n_u16(__p0, __p1, __p2) __extension__ ({ \
24168   uint16x4_t __s1 = __p1; \
24174 #define vsli_n_u16(__p0, __p1, __p2) __extension__ ({ \
24176   uint16x4_t __s1 = __p1; \
24187 #define vsli_n_s8(__p0, __p1, __p2) __extension__ ({ \
24189   int8x8_t __s1 = __p1; \
24195 #define vsli_n_s8(__p0, __p1, __p2) __extension__ ({ \
24197   int8x8_t __s1 = __p1; \
24208 #define vsli_n_s32(__p0, __p1, __p2) __extension__ ({ \
24210   int32x2_t __s1 = __p1; \
24216 #define vsli_n_s32(__p0, __p1, __p2) __extension__ ({ \
24218   int32x2_t __s1 = __p1; \
24229 #define vsli_n_s64(__p0, __p1, __p2) __extension__ ({ \
24231   int64x1_t __s1 = __p1; \
24237 #define vsli_n_s64(__p0, __p1, __p2) __extension__ ({ \
24239   int64x1_t __s1 = __p1; \
24247 #define vsli_n_s16(__p0, __p1, __p2) __extension__ ({ \
24249   int16x4_t __s1 = __p1; \
24255 #define vsli_n_s16(__p0, __p1, __p2) __extension__ ({ \
24257   int16x4_t __s1 = __p1; \
24268 #define vsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
24270   uint8x16_t __s1 = __p1; \
24276 #define vsraq_n_u8(__p0, __p1, __p2) __extension__ ({ \
24278   uint8x16_t __s1 = __p1; \
24289 #define vsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
24291   uint32x4_t __s1 = __p1; \
24297 #define vsraq_n_u32(__p0, __p1, __p2) __extension__ ({ \
24299   uint32x4_t __s1 = __p1; \
24310 #define vsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
24312   uint64x2_t __s1 = __p1; \
24318 #define vsraq_n_u64(__p0, __p1, __p2) __extension__ ({ \
24320   uint64x2_t __s1 = __p1; \
24331 #define vsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24333   uint16x8_t __s1 = __p1; \
24339 #define vsraq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24341   uint16x8_t __s1 = __p1; \
24352 #define vsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24354   int8x16_t __s1 = __p1; \
24360 #define vsraq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24362   int8x16_t __s1 = __p1; \
24373 #define vsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24375   int32x4_t __s1 = __p1; \
24381 #define vsraq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24383   int32x4_t __s1 = __p1; \
24394 #define vsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24396   int64x2_t __s1 = __p1; \
24402 #define vsraq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24404   int64x2_t __s1 = __p1; \
24415 #define vsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24417   int16x8_t __s1 = __p1; \
24423 #define vsraq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24425   int16x8_t __s1 = __p1; \
24436 #define vsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
24438   uint8x8_t __s1 = __p1; \
24444 #define vsra_n_u8(__p0, __p1, __p2) __extension__ ({ \
24446   uint8x8_t __s1 = __p1; \
24457 #define vsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
24459   uint32x2_t __s1 = __p1; \
24465 #define vsra_n_u32(__p0, __p1, __p2) __extension__ ({ \
24467   uint32x2_t __s1 = __p1; \
24478 #define vsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
24480   uint64x1_t __s1 = __p1; \
24486 #define vsra_n_u64(__p0, __p1, __p2) __extension__ ({ \
24488   uint64x1_t __s1 = __p1; \
24496 #define vsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
24498   uint16x4_t __s1 = __p1; \
24504 #define vsra_n_u16(__p0, __p1, __p2) __extension__ ({ \
24506   uint16x4_t __s1 = __p1; \
24517 #define vsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
24519   int8x8_t __s1 = __p1; \
24525 #define vsra_n_s8(__p0, __p1, __p2) __extension__ ({ \
24527   int8x8_t __s1 = __p1; \
24538 #define vsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
24540   int32x2_t __s1 = __p1; \
24546 #define vsra_n_s32(__p0, __p1, __p2) __extension__ ({ \
24548   int32x2_t __s1 = __p1; \
24559 #define vsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
24561   int64x1_t __s1 = __p1; \
24567 #define vsra_n_s64(__p0, __p1, __p2) __extension__ ({ \
24569   int64x1_t __s1 = __p1; \
24577 #define vsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
24579   int16x4_t __s1 = __p1; \
24585 #define vsra_n_s16(__p0, __p1, __p2) __extension__ ({ \
24587   int16x4_t __s1 = __p1; \
24598 #define vsri_n_p8(__p0, __p1, __p2) __extension__ ({ \
24600   poly8x8_t __s1 = __p1; \
24606 #define vsri_n_p8(__p0, __p1, __p2) __extension__ ({ \
24608   poly8x8_t __s1 = __p1; \
24619 #define vsri_n_p16(__p0, __p1, __p2) __extension__ ({ \
24621   poly16x4_t __s1 = __p1; \
24627 #define vsri_n_p16(__p0, __p1, __p2) __extension__ ({ \
24629   poly16x4_t __s1 = __p1; \
24640 #define vsriq_n_p8(__p0, __p1, __p2) __extension__ ({ \
24642   poly8x16_t __s1 = __p1; \
24648 #define vsriq_n_p8(__p0, __p1, __p2) __extension__ ({ \
24650   poly8x16_t __s1 = __p1; \
24661 #define vsriq_n_p16(__p0, __p1, __p2) __extension__ ({ \
24663   poly16x8_t __s1 = __p1; \
24669 #define vsriq_n_p16(__p0, __p1, __p2) __extension__ ({ \
24671   poly16x8_t __s1 = __p1; \
24682 #define vsriq_n_u8(__p0, __p1, __p2) __extension__ ({ \
24684   uint8x16_t __s1 = __p1; \
24690 #define vsriq_n_u8(__p0, __p1, __p2) __extension__ ({ \
24692   uint8x16_t __s1 = __p1; \
24703 #define vsriq_n_u32(__p0, __p1, __p2) __extension__ ({ \
24705   uint32x4_t __s1 = __p1; \
24711 #define vsriq_n_u32(__p0, __p1, __p2) __extension__ ({ \
24713   uint32x4_t __s1 = __p1; \
24724 #define vsriq_n_u64(__p0, __p1, __p2) __extension__ ({ \
24726   uint64x2_t __s1 = __p1; \
24732 #define vsriq_n_u64(__p0, __p1, __p2) __extension__ ({ \
24734   uint64x2_t __s1 = __p1; \
24745 #define vsriq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24747   uint16x8_t __s1 = __p1; \
24753 #define vsriq_n_u16(__p0, __p1, __p2) __extension__ ({ \
24755   uint16x8_t __s1 = __p1; \
24766 #define vsriq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24768   int8x16_t __s1 = __p1; \
24774 #define vsriq_n_s8(__p0, __p1, __p2) __extension__ ({ \
24776   int8x16_t __s1 = __p1; \
24787 #define vsriq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24789   int32x4_t __s1 = __p1; \
24795 #define vsriq_n_s32(__p0, __p1, __p2) __extension__ ({ \
24797   int32x4_t __s1 = __p1; \
24808 #define vsriq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24810   int64x2_t __s1 = __p1; \
24816 #define vsriq_n_s64(__p0, __p1, __p2) __extension__ ({ \
24818   int64x2_t __s1 = __p1; \
24829 #define vsriq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24831   int16x8_t __s1 = __p1; \
24837 #define vsriq_n_s16(__p0, __p1, __p2) __extension__ ({ \
24839   int16x8_t __s1 = __p1; \
24850 #define vsri_n_u8(__p0, __p1, __p2) __extension__ ({ \
24852   uint8x8_t __s1 = __p1; \
24858 #define vsri_n_u8(__p0, __p1, __p2) __extension__ ({ \
24860   uint8x8_t __s1 = __p1; \
24871 #define vsri_n_u32(__p0, __p1, __p2) __extension__ ({ \
24873   uint32x2_t __s1 = __p1; \
24879 #define vsri_n_u32(__p0, __p1, __p2) __extension__ ({ \
24881   uint32x2_t __s1 = __p1; \
24892 #define vsri_n_u64(__p0, __p1, __p2) __extension__ ({ \
24894   uint64x1_t __s1 = __p1; \
24900 #define vsri_n_u64(__p0, __p1, __p2) __extension__ ({ \
24902   uint64x1_t __s1 = __p1; \
24910 #define vsri_n_u16(__p0, __p1, __p2) __extension__ ({ \
24912   uint16x4_t __s1 = __p1; \
24918 #define vsri_n_u16(__p0, __p1, __p2) __extension__ ({ \
24920   uint16x4_t __s1 = __p1; \
24931 #define vsri_n_s8(__p0, __p1, __p2) __extension__ ({ \
24933   int8x8_t __s1 = __p1; \
24939 #define vsri_n_s8(__p0, __p1, __p2) __extension__ ({ \
24941   int8x8_t __s1 = __p1; \
24952 #define vsri_n_s32(__p0, __p1, __p2) __extension__ ({ \
24954   int32x2_t __s1 = __p1; \
24960 #define vsri_n_s32(__p0, __p1, __p2) __extension__ ({ \
24962   int32x2_t __s1 = __p1; \
24973 #define vsri_n_s64(__p0, __p1, __p2) __extension__ ({ \
24975   int64x1_t __s1 = __p1; \
24981 #define vsri_n_s64(__p0, __p1, __p2) __extension__ ({ \
24983   int64x1_t __s1 = __p1; \
24991 #define vsri_n_s16(__p0, __p1, __p2) __extension__ ({ \
24993   int16x4_t __s1 = __p1; \
24999 #define vsri_n_s16(__p0, __p1, __p2) __extension__ ({ \
25001   int16x4_t __s1 = __p1; \
25012 #define vst1_p8(__p0, __p1) __extension__ ({ \
25013   poly8x8_t __s1 = __p1; \
25017 #define vst1_p8(__p0, __p1) __extension__ ({ \
25018   poly8x8_t __s1 = __p1; \
25025 #define vst1_p16(__p0, __p1) __extension__ ({ \
25026   poly16x4_t __s1 = __p1; \
25030 #define vst1_p16(__p0, __p1) __extension__ ({ \
25031   poly16x4_t __s1 = __p1; \
25038 #define vst1q_p8(__p0, __p1) __extension__ ({ \
25039   poly8x16_t __s1 = __p1; \
25043 #define vst1q_p8(__p0, __p1) __extension__ ({ \
25044   poly8x16_t __s1 = __p1; \
25051 #define vst1q_p16(__p0, __p1) __extension__ ({ \
25052   poly16x8_t __s1 = __p1; \
25056 #define vst1q_p16(__p0, __p1) __extension__ ({ \
25057   poly16x8_t __s1 = __p1; \
25064 #define vst1q_u8(__p0, __p1) __extension__ ({ \
25065   uint8x16_t __s1 = __p1; \
25069 #define vst1q_u8(__p0, __p1) __extension__ ({ \
25070   uint8x16_t __s1 = __p1; \
25077 #define vst1q_u32(__p0, __p1) __extension__ ({ \
25078   uint32x4_t __s1 = __p1; \
25082 #define vst1q_u32(__p0, __p1) __extension__ ({ \
25083   uint32x4_t __s1 = __p1; \
25090 #define vst1q_u64(__p0, __p1) __extension__ ({ \
25091   uint64x2_t __s1 = __p1; \
25095 #define vst1q_u64(__p0, __p1) __extension__ ({ \
25096   uint64x2_t __s1 = __p1; \
25103 #define vst1q_u16(__p0, __p1) __extension__ ({ \
25104   uint16x8_t __s1 = __p1; \
25108 #define vst1q_u16(__p0, __p1) __extension__ ({ \
25109   uint16x8_t __s1 = __p1; \
25116 #define vst1q_s8(__p0, __p1) __extension__ ({ \
25117   int8x16_t __s1 = __p1; \
25121 #define vst1q_s8(__p0, __p1) __extension__ ({ \
25122   int8x16_t __s1 = __p1; \
25129 #define vst1q_f32(__p0, __p1) __extension__ ({ \
25130   float32x4_t __s1 = __p1; \
25134 #define vst1q_f32(__p0, __p1) __extension__ ({ \
25135   float32x4_t __s1 = __p1; \
25142 #define vst1q_f16(__p0, __p1) __extension__ ({ \
25143   float16x8_t __s1 = __p1; \
25147 #define vst1q_f16(__p0, __p1) __extension__ ({ \
25148   float16x8_t __s1 = __p1; \
25155 #define vst1q_s32(__p0, __p1) __extension__ ({ \
25156   int32x4_t __s1 = __p1; \
25160 #define vst1q_s32(__p0, __p1) __extension__ ({ \
25161   int32x4_t __s1 = __p1; \
25168 #define vst1q_s64(__p0, __p1) __extension__ ({ \
25169   int64x2_t __s1 = __p1; \
25173 #define vst1q_s64(__p0, __p1) __extension__ ({ \
25174   int64x2_t __s1 = __p1; \
25181 #define vst1q_s16(__p0, __p1) __extension__ ({ \
25182   int16x8_t __s1 = __p1; \
25186 #define vst1q_s16(__p0, __p1) __extension__ ({ \
25187   int16x8_t __s1 = __p1; \
25194 #define vst1_u8(__p0, __p1) __extension__ ({ \
25195   uint8x8_t __s1 = __p1; \
25199 #define vst1_u8(__p0, __p1) __extension__ ({ \
25200   uint8x8_t __s1 = __p1; \
25207 #define vst1_u32(__p0, __p1) __extension__ ({ \
25208   uint32x2_t __s1 = __p1; \
25212 #define vst1_u32(__p0, __p1) __extension__ ({ \
25213   uint32x2_t __s1 = __p1; \
25220 #define vst1_u64(__p0, __p1) __extension__ ({ \
25221   uint64x1_t __s1 = __p1; \
25225 #define vst1_u64(__p0, __p1) __extension__ ({ \
25226   uint64x1_t __s1 = __p1; \
25232 #define vst1_u16(__p0, __p1) __extension__ ({ \
25233   uint16x4_t __s1 = __p1; \
25237 #define vst1_u16(__p0, __p1) __extension__ ({ \
25238   uint16x4_t __s1 = __p1; \
25245 #define vst1_s8(__p0, __p1) __extension__ ({ \
25246   int8x8_t __s1 = __p1; \
25250 #define vst1_s8(__p0, __p1) __extension__ ({ \
25251   int8x8_t __s1 = __p1; \
25258 #define vst1_f32(__p0, __p1) __extension__ ({ \
25259   float32x2_t __s1 = __p1; \
25263 #define vst1_f32(__p0, __p1) __extension__ ({ \
25264   float32x2_t __s1 = __p1; \
25271 #define vst1_f16(__p0, __p1) __extension__ ({ \
25272   float16x4_t __s1 = __p1; \
25276 #define vst1_f16(__p0, __p1) __extension__ ({ \
25277   float16x4_t __s1 = __p1; \
25284 #define vst1_s32(__p0, __p1) __extension__ ({ \
25285   int32x2_t __s1 = __p1; \
25289 #define vst1_s32(__p0, __p1) __extension__ ({ \
25290   int32x2_t __s1 = __p1; \
25297 #define vst1_s64(__p0, __p1) __extension__ ({ \
25298   int64x1_t __s1 = __p1; \
25302 #define vst1_s64(__p0, __p1) __extension__ ({ \
25303   int64x1_t __s1 = __p1; \
25309 #define vst1_s16(__p0, __p1) __extension__ ({ \
25310   int16x4_t __s1 = __p1; \
25314 #define vst1_s16(__p0, __p1) __extension__ ({ \
25315   int16x4_t __s1 = __p1; \
25322 #define vst1_lane_p8(__p0, __p1, __p2) __extension__ ({ \
25323   poly8x8_t __s1 = __p1; \
25327 #define vst1_lane_p8(__p0, __p1, __p2) __extension__ ({ \
25328   poly8x8_t __s1 = __p1; \
25335 #define vst1_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25336   poly16x4_t __s1 = __p1; \
25340 #define vst1_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25341   poly16x4_t __s1 = __p1; \
25348 #define vst1q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
25349   poly8x16_t __s1 = __p1; \
25353 #define vst1q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
25354   poly8x16_t __s1 = __p1; \
25361 #define vst1q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25362   poly16x8_t __s1 = __p1; \
25366 #define vst1q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25367   poly16x8_t __s1 = __p1; \
25374 #define vst1q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
25375   uint8x16_t __s1 = __p1; \
25379 #define vst1q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
25380   uint8x16_t __s1 = __p1; \
25387 #define vst1q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
25388   uint32x4_t __s1 = __p1; \
25392 #define vst1q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
25393   uint32x4_t __s1 = __p1; \
25400 #define vst1q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
25401   uint64x2_t __s1 = __p1; \
25405 #define vst1q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
25406   uint64x2_t __s1 = __p1; \
25413 #define vst1q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
25414   uint16x8_t __s1 = __p1; \
25418 #define vst1q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
25419   uint16x8_t __s1 = __p1; \
25426 #define vst1q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
25427   int8x16_t __s1 = __p1; \
25431 #define vst1q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
25432   int8x16_t __s1 = __p1; \
25439 #define vst1q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
25440   float32x4_t __s1 = __p1; \
25444 #define vst1q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
25445   float32x4_t __s1 = __p1; \
25452 #define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
25453   float16x8_t __s1 = __p1; \
25457 #define vst1q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
25458   float16x8_t __s1 = __p1; \
25465 #define vst1q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
25466   int32x4_t __s1 = __p1; \
25470 #define vst1q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
25471   int32x4_t __s1 = __p1; \
25478 #define vst1q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
25479   int64x2_t __s1 = __p1; \
25483 #define vst1q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
25484   int64x2_t __s1 = __p1; \
25491 #define vst1q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
25492   int16x8_t __s1 = __p1; \
25496 #define vst1q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
25497   int16x8_t __s1 = __p1; \
25504 #define vst1_lane_u8(__p0, __p1, __p2) __extension__ ({ \
25505   uint8x8_t __s1 = __p1; \
25509 #define vst1_lane_u8(__p0, __p1, __p2) __extension__ ({ \
25510   uint8x8_t __s1 = __p1; \
25517 #define vst1_lane_u32(__p0, __p1, __p2) __extension__ ({ \
25518   uint32x2_t __s1 = __p1; \
25522 #define vst1_lane_u32(__p0, __p1, __p2) __extension__ ({ \
25523   uint32x2_t __s1 = __p1; \
25530 #define vst1_lane_u64(__p0, __p1, __p2) __extension__ ({ \
25531   uint64x1_t __s1 = __p1; \
25535 #define vst1_lane_u64(__p0, __p1, __p2) __extension__ ({ \
25536   uint64x1_t __s1 = __p1; \
25542 #define vst1_lane_u16(__p0, __p1, __p2) __extension__ ({ \
25543   uint16x4_t __s1 = __p1; \
25547 #define vst1_lane_u16(__p0, __p1, __p2) __extension__ ({ \
25548   uint16x4_t __s1 = __p1; \
25555 #define vst1_lane_s8(__p0, __p1, __p2) __extension__ ({ \
25556   int8x8_t __s1 = __p1; \
25560 #define vst1_lane_s8(__p0, __p1, __p2) __extension__ ({ \
25561   int8x8_t __s1 = __p1; \
25568 #define vst1_lane_f32(__p0, __p1, __p2) __extension__ ({ \
25569   float32x2_t __s1 = __p1; \
25573 #define vst1_lane_f32(__p0, __p1, __p2) __extension__ ({ \
25574   float32x2_t __s1 = __p1; \
25581 #define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
25582   float16x4_t __s1 = __p1; \
25586 #define vst1_lane_f16(__p0, __p1, __p2) __extension__ ({ \
25587   float16x4_t __s1 = __p1; \
25594 #define vst1_lane_s32(__p0, __p1, __p2) __extension__ ({ \
25595   int32x2_t __s1 = __p1; \
25599 #define vst1_lane_s32(__p0, __p1, __p2) __extension__ ({ \
25600   int32x2_t __s1 = __p1; \
25607 #define vst1_lane_s64(__p0, __p1, __p2) __extension__ ({ \
25608   int64x1_t __s1 = __p1; \
25612 #define vst1_lane_s64(__p0, __p1, __p2) __extension__ ({ \
25613   int64x1_t __s1 = __p1; \
25619 #define vst1_lane_s16(__p0, __p1, __p2) __extension__ ({ \
25620   int16x4_t __s1 = __p1; \
25624 #define vst1_lane_s16(__p0, __p1, __p2) __extension__ ({ \
25625   int16x4_t __s1 = __p1; \
25632 #define vst2_p8(__p0, __p1) __extension__ ({ \
25633   poly8x8x2_t __s1 = __p1; \
25637 #define vst2_p8(__p0, __p1) __extension__ ({ \
25638   poly8x8x2_t __s1 = __p1; \
25647 #define vst2_p16(__p0, __p1) __extension__ ({ \
25648   poly16x4x2_t __s1 = __p1; \
25652 #define vst2_p16(__p0, __p1) __extension__ ({ \
25653   poly16x4x2_t __s1 = __p1; \
25662 #define vst2q_p8(__p0, __p1) __extension__ ({ \
25663   poly8x16x2_t __s1 = __p1; \
25667 #define vst2q_p8(__p0, __p1) __extension__ ({ \
25668   poly8x16x2_t __s1 = __p1; \
25677 #define vst2q_p16(__p0, __p1) __extension__ ({ \
25678   poly16x8x2_t __s1 = __p1; \
25682 #define vst2q_p16(__p0, __p1) __extension__ ({ \
25683   poly16x8x2_t __s1 = __p1; \
25692 #define vst2q_u8(__p0, __p1) __extension__ ({ \
25693   uint8x16x2_t __s1 = __p1; \
25697 #define vst2q_u8(__p0, __p1) __extension__ ({ \
25698   uint8x16x2_t __s1 = __p1; \
25707 #define vst2q_u32(__p0, __p1) __extension__ ({ \
25708   uint32x4x2_t __s1 = __p1; \
25712 #define vst2q_u32(__p0, __p1) __extension__ ({ \
25713   uint32x4x2_t __s1 = __p1; \
25722 #define vst2q_u16(__p0, __p1) __extension__ ({ \
25723   uint16x8x2_t __s1 = __p1; \
25727 #define vst2q_u16(__p0, __p1) __extension__ ({ \
25728   uint16x8x2_t __s1 = __p1; \
25737 #define vst2q_s8(__p0, __p1) __extension__ ({ \
25738   int8x16x2_t __s1 = __p1; \
25742 #define vst2q_s8(__p0, __p1) __extension__ ({ \
25743   int8x16x2_t __s1 = __p1; \
25752 #define vst2q_f32(__p0, __p1) __extension__ ({ \
25753   float32x4x2_t __s1 = __p1; \
25757 #define vst2q_f32(__p0, __p1) __extension__ ({ \
25758   float32x4x2_t __s1 = __p1; \
25767 #define vst2q_f16(__p0, __p1) __extension__ ({ \
25768   float16x8x2_t __s1 = __p1; \
25772 #define vst2q_f16(__p0, __p1) __extension__ ({ \
25773   float16x8x2_t __s1 = __p1; \
25782 #define vst2q_s32(__p0, __p1) __extension__ ({ \
25783   int32x4x2_t __s1 = __p1; \
25787 #define vst2q_s32(__p0, __p1) __extension__ ({ \
25788   int32x4x2_t __s1 = __p1; \
25797 #define vst2q_s16(__p0, __p1) __extension__ ({ \
25798   int16x8x2_t __s1 = __p1; \
25802 #define vst2q_s16(__p0, __p1) __extension__ ({ \
25803   int16x8x2_t __s1 = __p1; \
25812 #define vst2_u8(__p0, __p1) __extension__ ({ \
25813   uint8x8x2_t __s1 = __p1; \
25817 #define vst2_u8(__p0, __p1) __extension__ ({ \
25818   uint8x8x2_t __s1 = __p1; \
25827 #define vst2_u32(__p0, __p1) __extension__ ({ \
25828   uint32x2x2_t __s1 = __p1; \
25832 #define vst2_u32(__p0, __p1) __extension__ ({ \
25833   uint32x2x2_t __s1 = __p1; \
25842 #define vst2_u64(__p0, __p1) __extension__ ({ \
25843   uint64x1x2_t __s1 = __p1; \
25847 #define vst2_u64(__p0, __p1) __extension__ ({ \
25848   uint64x1x2_t __s1 = __p1; \
25854 #define vst2_u16(__p0, __p1) __extension__ ({ \
25855   uint16x4x2_t __s1 = __p1; \
25859 #define vst2_u16(__p0, __p1) __extension__ ({ \
25860   uint16x4x2_t __s1 = __p1; \
25869 #define vst2_s8(__p0, __p1) __extension__ ({ \
25870   int8x8x2_t __s1 = __p1; \
25874 #define vst2_s8(__p0, __p1) __extension__ ({ \
25875   int8x8x2_t __s1 = __p1; \
25884 #define vst2_f32(__p0, __p1) __extension__ ({ \
25885   float32x2x2_t __s1 = __p1; \
25889 #define vst2_f32(__p0, __p1) __extension__ ({ \
25890   float32x2x2_t __s1 = __p1; \
25899 #define vst2_f16(__p0, __p1) __extension__ ({ \
25900   float16x4x2_t __s1 = __p1; \
25904 #define vst2_f16(__p0, __p1) __extension__ ({ \
25905   float16x4x2_t __s1 = __p1; \
25914 #define vst2_s32(__p0, __p1) __extension__ ({ \
25915   int32x2x2_t __s1 = __p1; \
25919 #define vst2_s32(__p0, __p1) __extension__ ({ \
25920   int32x2x2_t __s1 = __p1; \
25929 #define vst2_s64(__p0, __p1) __extension__ ({ \
25930   int64x1x2_t __s1 = __p1; \
25934 #define vst2_s64(__p0, __p1) __extension__ ({ \
25935   int64x1x2_t __s1 = __p1; \
25941 #define vst2_s16(__p0, __p1) __extension__ ({ \
25942   int16x4x2_t __s1 = __p1; \
25946 #define vst2_s16(__p0, __p1) __extension__ ({ \
25947   int16x4x2_t __s1 = __p1; \
25956 #define vst2_lane_p8(__p0, __p1, __p2) __extension__ ({ \
25957   poly8x8x2_t __s1 = __p1; \
25961 #define vst2_lane_p8(__p0, __p1, __p2) __extension__ ({ \
25962   poly8x8x2_t __s1 = __p1; \
25971 #define vst2_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25972   poly16x4x2_t __s1 = __p1; \
25976 #define vst2_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25977   poly16x4x2_t __s1 = __p1; \
25986 #define vst2q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25987   poly16x8x2_t __s1 = __p1; \
25991 #define vst2q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
25992   poly16x8x2_t __s1 = __p1; \
26001 #define vst2q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26002   uint32x4x2_t __s1 = __p1; \
26006 #define vst2q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26007   uint32x4x2_t __s1 = __p1; \
26016 #define vst2q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26017   uint16x8x2_t __s1 = __p1; \
26021 #define vst2q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26022   uint16x8x2_t __s1 = __p1; \
26031 #define vst2q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26032   float32x4x2_t __s1 = __p1; \
26036 #define vst2q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26037   float32x4x2_t __s1 = __p1; \
26046 #define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26047   float16x8x2_t __s1 = __p1; \
26051 #define vst2q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26052   float16x8x2_t __s1 = __p1; \
26061 #define vst2q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26062   int32x4x2_t __s1 = __p1; \
26066 #define vst2q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26067   int32x4x2_t __s1 = __p1; \
26076 #define vst2q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26077   int16x8x2_t __s1 = __p1; \
26081 #define vst2q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26082   int16x8x2_t __s1 = __p1; \
26091 #define vst2_lane_u8(__p0, __p1, __p2) __extension__ ({ \
26092   uint8x8x2_t __s1 = __p1; \
26096 #define vst2_lane_u8(__p0, __p1, __p2) __extension__ ({ \
26097   uint8x8x2_t __s1 = __p1; \
26106 #define vst2_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26107   uint32x2x2_t __s1 = __p1; \
26111 #define vst2_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26112   uint32x2x2_t __s1 = __p1; \
26121 #define vst2_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26122   uint16x4x2_t __s1 = __p1; \
26126 #define vst2_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26127   uint16x4x2_t __s1 = __p1; \
26136 #define vst2_lane_s8(__p0, __p1, __p2) __extension__ ({ \
26137   int8x8x2_t __s1 = __p1; \
26141 #define vst2_lane_s8(__p0, __p1, __p2) __extension__ ({ \
26142   int8x8x2_t __s1 = __p1; \
26151 #define vst2_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26152   float32x2x2_t __s1 = __p1; \
26156 #define vst2_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26157   float32x2x2_t __s1 = __p1; \
26166 #define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26167   float16x4x2_t __s1 = __p1; \
26171 #define vst2_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26172   float16x4x2_t __s1 = __p1; \
26181 #define vst2_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26182   int32x2x2_t __s1 = __p1; \
26186 #define vst2_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26187   int32x2x2_t __s1 = __p1; \
26196 #define vst2_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26197   int16x4x2_t __s1 = __p1; \
26201 #define vst2_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26202   int16x4x2_t __s1 = __p1; \
26211 #define vst3_p8(__p0, __p1) __extension__ ({ \
26212   poly8x8x3_t __s1 = __p1; \
26216 #define vst3_p8(__p0, __p1) __extension__ ({ \
26217   poly8x8x3_t __s1 = __p1; \
26227 #define vst3_p16(__p0, __p1) __extension__ ({ \
26228   poly16x4x3_t __s1 = __p1; \
26232 #define vst3_p16(__p0, __p1) __extension__ ({ \
26233   poly16x4x3_t __s1 = __p1; \
26243 #define vst3q_p8(__p0, __p1) __extension__ ({ \
26244   poly8x16x3_t __s1 = __p1; \
26248 #define vst3q_p8(__p0, __p1) __extension__ ({ \
26249   poly8x16x3_t __s1 = __p1; \
26259 #define vst3q_p16(__p0, __p1) __extension__ ({ \
26260   poly16x8x3_t __s1 = __p1; \
26264 #define vst3q_p16(__p0, __p1) __extension__ ({ \
26265   poly16x8x3_t __s1 = __p1; \
26275 #define vst3q_u8(__p0, __p1) __extension__ ({ \
26276   uint8x16x3_t __s1 = __p1; \
26280 #define vst3q_u8(__p0, __p1) __extension__ ({ \
26281   uint8x16x3_t __s1 = __p1; \
26291 #define vst3q_u32(__p0, __p1) __extension__ ({ \
26292   uint32x4x3_t __s1 = __p1; \
26296 #define vst3q_u32(__p0, __p1) __extension__ ({ \
26297   uint32x4x3_t __s1 = __p1; \
26307 #define vst3q_u16(__p0, __p1) __extension__ ({ \
26308   uint16x8x3_t __s1 = __p1; \
26312 #define vst3q_u16(__p0, __p1) __extension__ ({ \
26313   uint16x8x3_t __s1 = __p1; \
26323 #define vst3q_s8(__p0, __p1) __extension__ ({ \
26324   int8x16x3_t __s1 = __p1; \
26328 #define vst3q_s8(__p0, __p1) __extension__ ({ \
26329   int8x16x3_t __s1 = __p1; \
26339 #define vst3q_f32(__p0, __p1) __extension__ ({ \
26340   float32x4x3_t __s1 = __p1; \
26344 #define vst3q_f32(__p0, __p1) __extension__ ({ \
26345   float32x4x3_t __s1 = __p1; \
26355 #define vst3q_f16(__p0, __p1) __extension__ ({ \
26356   float16x8x3_t __s1 = __p1; \
26360 #define vst3q_f16(__p0, __p1) __extension__ ({ \
26361   float16x8x3_t __s1 = __p1; \
26371 #define vst3q_s32(__p0, __p1) __extension__ ({ \
26372   int32x4x3_t __s1 = __p1; \
26376 #define vst3q_s32(__p0, __p1) __extension__ ({ \
26377   int32x4x3_t __s1 = __p1; \
26387 #define vst3q_s16(__p0, __p1) __extension__ ({ \
26388   int16x8x3_t __s1 = __p1; \
26392 #define vst3q_s16(__p0, __p1) __extension__ ({ \
26393   int16x8x3_t __s1 = __p1; \
26403 #define vst3_u8(__p0, __p1) __extension__ ({ \
26404   uint8x8x3_t __s1 = __p1; \
26408 #define vst3_u8(__p0, __p1) __extension__ ({ \
26409   uint8x8x3_t __s1 = __p1; \
26419 #define vst3_u32(__p0, __p1) __extension__ ({ \
26420   uint32x2x3_t __s1 = __p1; \
26424 #define vst3_u32(__p0, __p1) __extension__ ({ \
26425   uint32x2x3_t __s1 = __p1; \
26435 #define vst3_u64(__p0, __p1) __extension__ ({ \
26436   uint64x1x3_t __s1 = __p1; \
26440 #define vst3_u64(__p0, __p1) __extension__ ({ \
26441   uint64x1x3_t __s1 = __p1; \
26447 #define vst3_u16(__p0, __p1) __extension__ ({ \
26448   uint16x4x3_t __s1 = __p1; \
26452 #define vst3_u16(__p0, __p1) __extension__ ({ \
26453   uint16x4x3_t __s1 = __p1; \
26463 #define vst3_s8(__p0, __p1) __extension__ ({ \
26464   int8x8x3_t __s1 = __p1; \
26468 #define vst3_s8(__p0, __p1) __extension__ ({ \
26469   int8x8x3_t __s1 = __p1; \
26479 #define vst3_f32(__p0, __p1) __extension__ ({ \
26480   float32x2x3_t __s1 = __p1; \
26484 #define vst3_f32(__p0, __p1) __extension__ ({ \
26485   float32x2x3_t __s1 = __p1; \
26495 #define vst3_f16(__p0, __p1) __extension__ ({ \
26496   float16x4x3_t __s1 = __p1; \
26500 #define vst3_f16(__p0, __p1) __extension__ ({ \
26501   float16x4x3_t __s1 = __p1; \
26511 #define vst3_s32(__p0, __p1) __extension__ ({ \
26512   int32x2x3_t __s1 = __p1; \
26516 #define vst3_s32(__p0, __p1) __extension__ ({ \
26517   int32x2x3_t __s1 = __p1; \
26527 #define vst3_s64(__p0, __p1) __extension__ ({ \
26528   int64x1x3_t __s1 = __p1; \
26532 #define vst3_s64(__p0, __p1) __extension__ ({ \
26533   int64x1x3_t __s1 = __p1; \
26539 #define vst3_s16(__p0, __p1) __extension__ ({ \
26540   int16x4x3_t __s1 = __p1; \
26544 #define vst3_s16(__p0, __p1) __extension__ ({ \
26545   int16x4x3_t __s1 = __p1; \
26555 #define vst3_lane_p8(__p0, __p1, __p2) __extension__ ({ \
26556   poly8x8x3_t __s1 = __p1; \
26560 #define vst3_lane_p8(__p0, __p1, __p2) __extension__ ({ \
26561   poly8x8x3_t __s1 = __p1; \
26571 #define vst3_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26572   poly16x4x3_t __s1 = __p1; \
26576 #define vst3_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26577   poly16x4x3_t __s1 = __p1; \
26587 #define vst3q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26588   poly16x8x3_t __s1 = __p1; \
26592 #define vst3q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
26593   poly16x8x3_t __s1 = __p1; \
26603 #define vst3q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26604   uint32x4x3_t __s1 = __p1; \
26608 #define vst3q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26609   uint32x4x3_t __s1 = __p1; \
26619 #define vst3q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26620   uint16x8x3_t __s1 = __p1; \
26624 #define vst3q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26625   uint16x8x3_t __s1 = __p1; \
26635 #define vst3q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26636   float32x4x3_t __s1 = __p1; \
26640 #define vst3q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26641   float32x4x3_t __s1 = __p1; \
26651 #define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26652   float16x8x3_t __s1 = __p1; \
26656 #define vst3q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26657   float16x8x3_t __s1 = __p1; \
26667 #define vst3q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26668   int32x4x3_t __s1 = __p1; \
26672 #define vst3q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26673   int32x4x3_t __s1 = __p1; \
26683 #define vst3q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26684   int16x8x3_t __s1 = __p1; \
26688 #define vst3q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26689   int16x8x3_t __s1 = __p1; \
26699 #define vst3_lane_u8(__p0, __p1, __p2) __extension__ ({ \
26700   uint8x8x3_t __s1 = __p1; \
26704 #define vst3_lane_u8(__p0, __p1, __p2) __extension__ ({ \
26705   uint8x8x3_t __s1 = __p1; \
26715 #define vst3_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26716   uint32x2x3_t __s1 = __p1; \
26720 #define vst3_lane_u32(__p0, __p1, __p2) __extension__ ({ \
26721   uint32x2x3_t __s1 = __p1; \
26731 #define vst3_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26732   uint16x4x3_t __s1 = __p1; \
26736 #define vst3_lane_u16(__p0, __p1, __p2) __extension__ ({ \
26737   uint16x4x3_t __s1 = __p1; \
26747 #define vst3_lane_s8(__p0, __p1, __p2) __extension__ ({ \
26748   int8x8x3_t __s1 = __p1; \
26752 #define vst3_lane_s8(__p0, __p1, __p2) __extension__ ({ \
26753   int8x8x3_t __s1 = __p1; \
26763 #define vst3_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26764   float32x2x3_t __s1 = __p1; \
26768 #define vst3_lane_f32(__p0, __p1, __p2) __extension__ ({ \
26769   float32x2x3_t __s1 = __p1; \
26779 #define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26780   float16x4x3_t __s1 = __p1; \
26784 #define vst3_lane_f16(__p0, __p1, __p2) __extension__ ({ \
26785   float16x4x3_t __s1 = __p1; \
26795 #define vst3_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26796   int32x2x3_t __s1 = __p1; \
26800 #define vst3_lane_s32(__p0, __p1, __p2) __extension__ ({ \
26801   int32x2x3_t __s1 = __p1; \
26811 #define vst3_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26812   int16x4x3_t __s1 = __p1; \
26816 #define vst3_lane_s16(__p0, __p1, __p2) __extension__ ({ \
26817   int16x4x3_t __s1 = __p1; \
26827 #define vst4_p8(__p0, __p1) __extension__ ({ \
26828   poly8x8x4_t __s1 = __p1; \
26832 #define vst4_p8(__p0, __p1) __extension__ ({ \
26833   poly8x8x4_t __s1 = __p1; \
26844 #define vst4_p16(__p0, __p1) __extension__ ({ \
26845   poly16x4x4_t __s1 = __p1; \
26849 #define vst4_p16(__p0, __p1) __extension__ ({ \
26850   poly16x4x4_t __s1 = __p1; \
26861 #define vst4q_p8(__p0, __p1) __extension__ ({ \
26862   poly8x16x4_t __s1 = __p1; \
26866 #define vst4q_p8(__p0, __p1) __extension__ ({ \
26867   poly8x16x4_t __s1 = __p1; \
26878 #define vst4q_p16(__p0, __p1) __extension__ ({ \
26879   poly16x8x4_t __s1 = __p1; \
26883 #define vst4q_p16(__p0, __p1) __extension__ ({ \
26884   poly16x8x4_t __s1 = __p1; \
26895 #define vst4q_u8(__p0, __p1) __extension__ ({ \
26896   uint8x16x4_t __s1 = __p1; \
26900 #define vst4q_u8(__p0, __p1) __extension__ ({ \
26901   uint8x16x4_t __s1 = __p1; \
26912 #define vst4q_u32(__p0, __p1) __extension__ ({ \
26913   uint32x4x4_t __s1 = __p1; \
26917 #define vst4q_u32(__p0, __p1) __extension__ ({ \
26918   uint32x4x4_t __s1 = __p1; \
26929 #define vst4q_u16(__p0, __p1) __extension__ ({ \
26930   uint16x8x4_t __s1 = __p1; \
26934 #define vst4q_u16(__p0, __p1) __extension__ ({ \
26935   uint16x8x4_t __s1 = __p1; \
26946 #define vst4q_s8(__p0, __p1) __extension__ ({ \
26947   int8x16x4_t __s1 = __p1; \
26951 #define vst4q_s8(__p0, __p1) __extension__ ({ \
26952   int8x16x4_t __s1 = __p1; \
26963 #define vst4q_f32(__p0, __p1) __extension__ ({ \
26964   float32x4x4_t __s1 = __p1; \
26968 #define vst4q_f32(__p0, __p1) __extension__ ({ \
26969   float32x4x4_t __s1 = __p1; \
26980 #define vst4q_f16(__p0, __p1) __extension__ ({ \
26981   float16x8x4_t __s1 = __p1; \
26985 #define vst4q_f16(__p0, __p1) __extension__ ({ \
26986   float16x8x4_t __s1 = __p1; \
26997 #define vst4q_s32(__p0, __p1) __extension__ ({ \
26998   int32x4x4_t __s1 = __p1; \
27002 #define vst4q_s32(__p0, __p1) __extension__ ({ \
27003   int32x4x4_t __s1 = __p1; \
27014 #define vst4q_s16(__p0, __p1) __extension__ ({ \
27015   int16x8x4_t __s1 = __p1; \
27019 #define vst4q_s16(__p0, __p1) __extension__ ({ \
27020   int16x8x4_t __s1 = __p1; \
27031 #define vst4_u8(__p0, __p1) __extension__ ({ \
27032   uint8x8x4_t __s1 = __p1; \
27036 #define vst4_u8(__p0, __p1) __extension__ ({ \
27037   uint8x8x4_t __s1 = __p1; \
27048 #define vst4_u32(__p0, __p1) __extension__ ({ \
27049   uint32x2x4_t __s1 = __p1; \
27053 #define vst4_u32(__p0, __p1) __extension__ ({ \
27054   uint32x2x4_t __s1 = __p1; \
27065 #define vst4_u64(__p0, __p1) __extension__ ({ \
27066   uint64x1x4_t __s1 = __p1; \
27070 #define vst4_u64(__p0, __p1) __extension__ ({ \
27071   uint64x1x4_t __s1 = __p1; \
27077 #define vst4_u16(__p0, __p1) __extension__ ({ \
27078   uint16x4x4_t __s1 = __p1; \
27082 #define vst4_u16(__p0, __p1) __extension__ ({ \
27083   uint16x4x4_t __s1 = __p1; \
27094 #define vst4_s8(__p0, __p1) __extension__ ({ \
27095   int8x8x4_t __s1 = __p1; \
27099 #define vst4_s8(__p0, __p1) __extension__ ({ \
27100   int8x8x4_t __s1 = __p1; \
27111 #define vst4_f32(__p0, __p1) __extension__ ({ \
27112   float32x2x4_t __s1 = __p1; \
27116 #define vst4_f32(__p0, __p1) __extension__ ({ \
27117   float32x2x4_t __s1 = __p1; \
27128 #define vst4_f16(__p0, __p1) __extension__ ({ \
27129   float16x4x4_t __s1 = __p1; \
27133 #define vst4_f16(__p0, __p1) __extension__ ({ \
27134   float16x4x4_t __s1 = __p1; \
27145 #define vst4_s32(__p0, __p1) __extension__ ({ \
27146   int32x2x4_t __s1 = __p1; \
27150 #define vst4_s32(__p0, __p1) __extension__ ({ \
27151   int32x2x4_t __s1 = __p1; \
27162 #define vst4_s64(__p0, __p1) __extension__ ({ \
27163   int64x1x4_t __s1 = __p1; \
27167 #define vst4_s64(__p0, __p1) __extension__ ({ \
27168   int64x1x4_t __s1 = __p1; \
27174 #define vst4_s16(__p0, __p1) __extension__ ({ \
27175   int16x4x4_t __s1 = __p1; \
27179 #define vst4_s16(__p0, __p1) __extension__ ({ \
27180   int16x4x4_t __s1 = __p1; \
27191 #define vst4_lane_p8(__p0, __p1, __p2) __extension__ ({ \
27192   poly8x8x4_t __s1 = __p1; \
27196 #define vst4_lane_p8(__p0, __p1, __p2) __extension__ ({ \
27197   poly8x8x4_t __s1 = __p1; \
27208 #define vst4_lane_p16(__p0, __p1, __p2) __extension__ ({ \
27209   poly16x4x4_t __s1 = __p1; \
27213 #define vst4_lane_p16(__p0, __p1, __p2) __extension__ ({ \
27214   poly16x4x4_t __s1 = __p1; \
27225 #define vst4q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
27226   poly16x8x4_t __s1 = __p1; \
27230 #define vst4q_lane_p16(__p0, __p1, __p2) __extension__ ({ \
27231   poly16x8x4_t __s1 = __p1; \
27242 #define vst4q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
27243   uint32x4x4_t __s1 = __p1; \
27247 #define vst4q_lane_u32(__p0, __p1, __p2) __extension__ ({ \
27248   uint32x4x4_t __s1 = __p1; \
27259 #define vst4q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
27260   uint16x8x4_t __s1 = __p1; \
27264 #define vst4q_lane_u16(__p0, __p1, __p2) __extension__ ({ \
27265   uint16x8x4_t __s1 = __p1; \
27276 #define vst4q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
27277   float32x4x4_t __s1 = __p1; \
27281 #define vst4q_lane_f32(__p0, __p1, __p2) __extension__ ({ \
27282   float32x4x4_t __s1 = __p1; \
27293 #define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
27294   float16x8x4_t __s1 = __p1; \
27298 #define vst4q_lane_f16(__p0, __p1, __p2) __extension__ ({ \
27299   float16x8x4_t __s1 = __p1; \
27310 #define vst4q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
27311   int32x4x4_t __s1 = __p1; \
27315 #define vst4q_lane_s32(__p0, __p1, __p2) __extension__ ({ \
27316   int32x4x4_t __s1 = __p1; \
27327 #define vst4q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
27328   int16x8x4_t __s1 = __p1; \
27332 #define vst4q_lane_s16(__p0, __p1, __p2) __extension__ ({ \
27333   int16x8x4_t __s1 = __p1; \
27344 #define vst4_lane_u8(__p0, __p1, __p2) __extension__ ({ \
27345   uint8x8x4_t __s1 = __p1; \
27349 #define vst4_lane_u8(__p0, __p1, __p2) __extension__ ({ \
27350   uint8x8x4_t __s1 = __p1; \
27361 #define vst4_lane_u32(__p0, __p1, __p2) __extension__ ({ \
27362   uint32x2x4_t __s1 = __p1; \
27366 #define vst4_lane_u32(__p0, __p1, __p2) __extension__ ({ \
27367   uint32x2x4_t __s1 = __p1; \
27378 #define vst4_lane_u16(__p0, __p1, __p2) __extension__ ({ \
27379   uint16x4x4_t __s1 = __p1; \
27383 #define vst4_lane_u16(__p0, __p1, __p2) __extension__ ({ \
27384   uint16x4x4_t __s1 = __p1; \
27395 #define vst4_lane_s8(__p0, __p1, __p2) __extension__ ({ \
27396   int8x8x4_t __s1 = __p1; \
27400 #define vst4_lane_s8(__p0, __p1, __p2) __extension__ ({ \
27401   int8x8x4_t __s1 = __p1; \
27412 #define vst4_lane_f32(__p0, __p1, __p2) __extension__ ({ \
27413   float32x2x4_t __s1 = __p1; \
27417 #define vst4_lane_f32(__p0, __p1, __p2) __extension__ ({ \
27418   float32x2x4_t __s1 = __p1; \
27429 #define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
27430   float16x4x4_t __s1 = __p1; \
27434 #define vst4_lane_f16(__p0, __p1, __p2) __extension__ ({ \
27435   float16x4x4_t __s1 = __p1; \
27446 #define vst4_lane_s32(__p0, __p1, __p2) __extension__ ({ \
27447   int32x2x4_t __s1 = __p1; \
27451 #define vst4_lane_s32(__p0, __p1, __p2) __extension__ ({ \
27452   int32x2x4_t __s1 = __p1; \
27463 #define vst4_lane_s16(__p0, __p1, __p2) __extension__ ({ \
27464   int16x4x4_t __s1 = __p1; \
27468 #define vst4_lane_s16(__p0, __p1, __p2) __extension__ ({ \
27469   int16x4x4_t __s1 = __p1; \
27480 __ai uint8x16_t vsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
27482   __ret = __p0 - __p1;
27486 __ai uint8x16_t vsubq_u8(uint8x16_t __p0, uint8x16_t __p1) {
27488   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
27497 __ai uint32x4_t vsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
27499   __ret = __p0 - __p1;
27503 __ai uint32x4_t vsubq_u32(uint32x4_t __p0, uint32x4_t __p1) {
27505   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27514 __ai uint64x2_t vsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
27516   __ret = __p0 - __p1;
27520 __ai uint64x2_t vsubq_u64(uint64x2_t __p0, uint64x2_t __p1) {
27522   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27531 __ai uint16x8_t vsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
27533   __ret = __p0 - __p1;
27537 __ai uint16x8_t vsubq_u16(uint16x8_t __p0, uint16x8_t __p1) {
27539   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27548 __ai int8x16_t vsubq_s8(int8x16_t __p0, int8x16_t __p1) {
27550   __ret = __p0 - __p1;
27554 __ai int8x16_t vsubq_s8(int8x16_t __p0, int8x16_t __p1) {
27556   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
27565 __ai float32x4_t vsubq_f32(float32x4_t __p0, float32x4_t __p1) {
27567   __ret = __p0 - __p1;
27571 __ai float32x4_t vsubq_f32(float32x4_t __p0, float32x4_t __p1) {
27573   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27582 __ai int32x4_t vsubq_s32(int32x4_t __p0, int32x4_t __p1) {
27584   __ret = __p0 - __p1;
27588 __ai int32x4_t vsubq_s32(int32x4_t __p0, int32x4_t __p1) {
27590   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27599 __ai int64x2_t vsubq_s64(int64x2_t __p0, int64x2_t __p1) {
27601   __ret = __p0 - __p1;
27605 __ai int64x2_t vsubq_s64(int64x2_t __p0, int64x2_t __p1) {
27607   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27616 __ai int16x8_t vsubq_s16(int16x8_t __p0, int16x8_t __p1) {
27618   __ret = __p0 - __p1;
27622 __ai int16x8_t vsubq_s16(int16x8_t __p0, int16x8_t __p1) {
27624   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27633 __ai uint8x8_t vsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
27635   __ret = __p0 - __p1;
27639 __ai uint8x8_t vsub_u8(uint8x8_t __p0, uint8x8_t __p1) {
27641   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27650 __ai uint32x2_t vsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
27652   __ret = __p0 - __p1;
27656 __ai uint32x2_t vsub_u32(uint32x2_t __p0, uint32x2_t __p1) {
27658   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27667 __ai uint64x1_t vsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
27669   __ret = __p0 - __p1;
27673 __ai uint64x1_t vsub_u64(uint64x1_t __p0, uint64x1_t __p1) {
27675   __ret = __p0 - __p1;
27681 __ai uint16x4_t vsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
27683   __ret = __p0 - __p1;
27687 __ai uint16x4_t vsub_u16(uint16x4_t __p0, uint16x4_t __p1) {
27689   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27698 __ai int8x8_t vsub_s8(int8x8_t __p0, int8x8_t __p1) {
27700   __ret = __p0 - __p1;
27704 __ai int8x8_t vsub_s8(int8x8_t __p0, int8x8_t __p1) {
27706   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27715 __ai float32x2_t vsub_f32(float32x2_t __p0, float32x2_t __p1) {
27717   __ret = __p0 - __p1;
27721 __ai float32x2_t vsub_f32(float32x2_t __p0, float32x2_t __p1) {
27723   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27732 __ai int32x2_t vsub_s32(int32x2_t __p0, int32x2_t __p1) {
27734   __ret = __p0 - __p1;
27738 __ai int32x2_t vsub_s32(int32x2_t __p0, int32x2_t __p1) {
27740   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27749 __ai int64x1_t vsub_s64(int64x1_t __p0, int64x1_t __p1) {
27751   __ret = __p0 - __p1;
27755 __ai int64x1_t vsub_s64(int64x1_t __p0, int64x1_t __p1) {
27757   __ret = __p0 - __p1;
27763 __ai int16x4_t vsub_s16(int16x4_t __p0, int16x4_t __p1) {
27765   __ret = __p0 - __p1;
27769 __ai int16x4_t vsub_s16(int16x4_t __p0, int16x4_t __p1) {
27771   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27780 __ai uint16x4_t vsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
27782   __ret = (uint16x4_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
27786 __ai uint16x4_t vsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
27788   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27794 __ai uint16x4_t __noswap_vsubhn_u32(uint32x4_t __p0, uint32x4_t __p1) {
27796   __ret = (uint16x4_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 17);
27802 __ai uint32x2_t vsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
27804   __ret = (uint32x2_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
27808 __ai uint32x2_t vsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
27810   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27816 __ai uint32x2_t __noswap_vsubhn_u64(uint64x2_t __p0, uint64x2_t __p1) {
27818   __ret = (uint32x2_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 18);
27824 __ai uint8x8_t vsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
27826   __ret = (uint8x8_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
27830 __ai uint8x8_t vsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
27832   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27838 __ai uint8x8_t __noswap_vsubhn_u16(uint16x8_t __p0, uint16x8_t __p1) {
27840   __ret = (uint8x8_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 16);
27846 __ai int16x4_t vsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
27848   __ret = (int16x4_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
27852 __ai int16x4_t vsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
27854   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27860 __ai int16x4_t __noswap_vsubhn_s32(int32x4_t __p0, int32x4_t __p1) {
27862   __ret = (int16x4_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 1);
27868 __ai int32x2_t vsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
27870   __ret = (int32x2_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
27874 __ai int32x2_t vsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
27876   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27882 __ai int32x2_t __noswap_vsubhn_s64(int64x2_t __p0, int64x2_t __p1) {
27884   __ret = (int32x2_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 2);
27890 __ai int8x8_t vsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
27892   __ret = (int8x8_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
27896 __ai int8x8_t vsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
27898   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27904 __ai int8x8_t __noswap_vsubhn_s16(int16x8_t __p0, int16x8_t __p1) {
27906   __ret = (int8x8_t) __builtin_neon_vsubhn_v((int8x16_t)__p0, (int8x16_t)__p1, 0);
27912 __ai uint16x8_t vsubl_u8(uint8x8_t __p0, uint8x8_t __p1) {
27914   __ret = vmovl_u8(__p0) - vmovl_u8(__p1);
27918 __ai uint16x8_t vsubl_u8(uint8x8_t __p0, uint8x8_t __p1) {
27920   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27929 __ai uint64x2_t vsubl_u32(uint32x2_t __p0, uint32x2_t __p1) {
27931   __ret = vmovl_u32(__p0) - vmovl_u32(__p1);
27935 __ai uint64x2_t vsubl_u32(uint32x2_t __p0, uint32x2_t __p1) {
27937   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27946 __ai uint32x4_t vsubl_u16(uint16x4_t __p0, uint16x4_t __p1) {
27948   __ret = vmovl_u16(__p0) - vmovl_u16(__p1);
27952 __ai uint32x4_t vsubl_u16(uint16x4_t __p0, uint16x4_t __p1) {
27954   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
27963 __ai int16x8_t vsubl_s8(int8x8_t __p0, int8x8_t __p1) {
27965   __ret = vmovl_s8(__p0) - vmovl_s8(__p1);
27969 __ai int16x8_t vsubl_s8(int8x8_t __p0, int8x8_t __p1) {
27971   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
27980 __ai int64x2_t vsubl_s32(int32x2_t __p0, int32x2_t __p1) {
27982   __ret = vmovl_s32(__p0) - vmovl_s32(__p1);
27986 __ai int64x2_t vsubl_s32(int32x2_t __p0, int32x2_t __p1) {
27988   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
27997 __ai int32x4_t vsubl_s16(int16x4_t __p0, int16x4_t __p1) {
27999   __ret = vmovl_s16(__p0) - vmovl_s16(__p1);
28003 __ai int32x4_t vsubl_s16(int16x4_t __p0, int16x4_t __p1) {
28005   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28014 __ai uint16x8_t vsubw_u8(uint16x8_t __p0, uint8x8_t __p1) {
28016   __ret = __p0 - vmovl_u8(__p1);
28020 __ai uint16x8_t vsubw_u8(uint16x8_t __p0, uint8x8_t __p1) {
28022   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28031 __ai uint64x2_t vsubw_u32(uint64x2_t __p0, uint32x2_t __p1) {
28033   __ret = __p0 - vmovl_u32(__p1);
28037 __ai uint64x2_t vsubw_u32(uint64x2_t __p0, uint32x2_t __p1) {
28039   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
28048 __ai uint32x4_t vsubw_u16(uint32x4_t __p0, uint16x4_t __p1) {
28050   __ret = __p0 - vmovl_u16(__p1);
28054 __ai uint32x4_t vsubw_u16(uint32x4_t __p0, uint16x4_t __p1) {
28056   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28065 __ai int16x8_t vsubw_s8(int16x8_t __p0, int8x8_t __p1) {
28067   __ret = __p0 - vmovl_s8(__p1);
28071 __ai int16x8_t vsubw_s8(int16x8_t __p0, int8x8_t __p1) {
28073   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28082 __ai int64x2_t vsubw_s32(int64x2_t __p0, int32x2_t __p1) {
28084   __ret = __p0 - vmovl_s32(__p1);
28088 __ai int64x2_t vsubw_s32(int64x2_t __p0, int32x2_t __p1) {
28090   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
28099 __ai int32x4_t vsubw_s16(int32x4_t __p0, int16x4_t __p1) {
28101   __ret = __p0 - vmovl_s16(__p1);
28105 __ai int32x4_t vsubw_s16(int32x4_t __p0, int16x4_t __p1) {
28107   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28116 __ai poly8x8_t vtbl1_p8(poly8x8_t __p0, uint8x8_t __p1) {
28118   __ret = (poly8x8_t) __builtin_neon_vtbl1_v((int8x8_t)__p0, (int8x8_t)__p1, 4);
28122 __ai poly8x8_t vtbl1_p8(poly8x8_t __p0, uint8x8_t __p1) {
28124   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28133 __ai uint8x8_t vtbl1_u8(uint8x8_t __p0, uint8x8_t __p1) {
28135   __ret = (uint8x8_t) __builtin_neon_vtbl1_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
28139 __ai uint8x8_t vtbl1_u8(uint8x8_t __p0, uint8x8_t __p1) {
28141   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28150 __ai int8x8_t vtbl1_s8(int8x8_t __p0, int8x8_t __p1) {
28152   __ret = (int8x8_t) __builtin_neon_vtbl1_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
28156 __ai int8x8_t vtbl1_s8(int8x8_t __p0, int8x8_t __p1) {
28158   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28167 __ai poly8x8_t vtbl2_p8(poly8x8x2_t __p0, uint8x8_t __p1) {
28169   __ret = (poly8x8_t) __builtin_neon_vtbl2_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p1, 4);
28173 __ai poly8x8_t vtbl2_p8(poly8x8x2_t __p0, uint8x8_t __p1) {
28177   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28186 __ai uint8x8_t vtbl2_u8(uint8x8x2_t __p0, uint8x8_t __p1) {
28188   __ret = (uint8x8_t) __builtin_neon_vtbl2_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p1, 16);
28192 __ai uint8x8_t vtbl2_u8(uint8x8x2_t __p0, uint8x8_t __p1) {
28196   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28205 __ai int8x8_t vtbl2_s8(int8x8x2_t __p0, int8x8_t __p1) {
28207   __ret = (int8x8_t) __builtin_neon_vtbl2_v((int8x8_t)__p0.val[0], (int8x8_t)__p0.val[1], (int8x8_t)__p1, 0);
28211 __ai int8x8_t vtbl2_s8(int8x8x2_t __p0, int8x8_t __p1) {
28215   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28224 __ai poly8x8_t vtbl3_p8(poly8x8x3_t __p0, uint8x8_t __p1) {
28226   __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);
28230 __ai poly8x8_t vtbl3_p8(poly8x8x3_t __p0, uint8x8_t __p1) {
28235   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28244 __ai uint8x8_t vtbl3_u8(uint8x8x3_t __p0, uint8x8_t __p1) {
28246   __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);
28250 __ai uint8x8_t vtbl3_u8(uint8x8x3_t __p0, uint8x8_t __p1) {
28255   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28264 __ai int8x8_t vtbl3_s8(int8x8x3_t __p0, int8x8_t __p1) {
28266   __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);
28270 __ai int8x8_t vtbl3_s8(int8x8x3_t __p0, int8x8_t __p1) {
28275   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28284 __ai poly8x8_t vtbl4_p8(poly8x8x4_t __p0, uint8x8_t __p1) {
28286   __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);
28290 __ai poly8x8_t vtbl4_p8(poly8x8x4_t __p0, uint8x8_t __p1) {
28296   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28305 __ai uint8x8_t vtbl4_u8(uint8x8x4_t __p0, uint8x8_t __p1) {
28307   __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);
28311 __ai uint8x8_t vtbl4_u8(uint8x8x4_t __p0, uint8x8_t __p1) {
28317   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28326 __ai int8x8_t vtbl4_s8(int8x8x4_t __p0, int8x8_t __p1) {
28328   __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);
28332 __ai int8x8_t vtbl4_s8(int8x8x4_t __p0, int8x8_t __p1) {
28338   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28347 __ai poly8x8_t vtbx1_p8(poly8x8_t __p0, poly8x8_t __p1, uint8x8_t __p2) {
28349   __ret = (poly8x8_t) __builtin_neon_vtbx1_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 4);
28353 __ai poly8x8_t vtbx1_p8(poly8x8_t __p0, poly8x8_t __p1, uint8x8_t __p2) {
28355   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28365 __ai uint8x8_t vtbx1_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
28367   __ret = (uint8x8_t) __builtin_neon_vtbx1_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 16);
28371 __ai uint8x8_t vtbx1_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
28373   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28383 __ai int8x8_t vtbx1_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
28385   __ret = (int8x8_t) __builtin_neon_vtbx1_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 0);
28389 __ai int8x8_t vtbx1_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
28391   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28401 __ai poly8x8_t vtbx2_p8(poly8x8_t __p0, poly8x8x2_t __p1, uint8x8_t __p2) {
28403   __ret = (poly8x8_t) __builtin_neon_vtbx2_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p2, 4);
28407 __ai poly8x8_t vtbx2_p8(poly8x8_t __p0, poly8x8x2_t __p1, uint8x8_t __p2) {
28410   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28411   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28421 __ai uint8x8_t vtbx2_u8(uint8x8_t __p0, uint8x8x2_t __p1, uint8x8_t __p2) {
28423   __ret = (uint8x8_t) __builtin_neon_vtbx2_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p2, 16);
28427 __ai uint8x8_t vtbx2_u8(uint8x8_t __p0, uint8x8x2_t __p1, uint8x8_t __p2) {
28430   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28431   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28441 __ai int8x8_t vtbx2_s8(int8x8_t __p0, int8x8x2_t __p1, int8x8_t __p2) {
28443   __ret = (int8x8_t) __builtin_neon_vtbx2_v((int8x8_t)__p0, (int8x8_t)__p1.val[0], (int8x8_t)__p1.val[1], (int8x8_t)__p2, 0);
28447 __ai int8x8_t vtbx2_s8(int8x8_t __p0, int8x8x2_t __p1, int8x8_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 poly8x8_t vtbx3_p8(poly8x8_t __p0, poly8x8x3_t __p1, uint8x8_t __p2) {
28463   __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);
28467 __ai poly8x8_t vtbx3_p8(poly8x8_t __p0, poly8x8x3_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);
28472   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28482 __ai uint8x8_t vtbx3_u8(uint8x8_t __p0, uint8x8x3_t __p1, uint8x8_t __p2) {
28484   __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);
28488 __ai uint8x8_t vtbx3_u8(uint8x8_t __p0, uint8x8x3_t __p1, uint8x8_t __p2) {
28491   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28492   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28493   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28503 __ai int8x8_t vtbx3_s8(int8x8_t __p0, int8x8x3_t __p1, int8x8_t __p2) {
28505   __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);
28509 __ai int8x8_t vtbx3_s8(int8x8_t __p0, int8x8x3_t __p1, int8x8_t __p2) {
28512   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28513   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28514   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28524 __ai poly8x8_t vtbx4_p8(poly8x8_t __p0, poly8x8x4_t __p1, uint8x8_t __p2) {
28526   __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);
28530 __ai poly8x8_t vtbx4_p8(poly8x8_t __p0, poly8x8x4_t __p1, uint8x8_t __p2) {
28533   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28534   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28535   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28536   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
28546 __ai uint8x8_t vtbx4_u8(uint8x8_t __p0, uint8x8x4_t __p1, uint8x8_t __p2) {
28548   __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);
28552 __ai uint8x8_t vtbx4_u8(uint8x8_t __p0, uint8x8x4_t __p1, uint8x8_t __p2) {
28555   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28556   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28557   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28558   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
28568 __ai int8x8_t vtbx4_s8(int8x8_t __p0, int8x8x4_t __p1, int8x8_t __p2) {
28570   __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);
28574 __ai int8x8_t vtbx4_s8(int8x8_t __p0, int8x8x4_t __p1, int8x8_t __p2) {
28577   __rev1.val[0] = __builtin_shufflevector(__p1.val[0], __p1.val[0], 7, 6, 5, 4, 3, 2, 1, 0);
28578   __rev1.val[1] = __builtin_shufflevector(__p1.val[1], __p1.val[1], 7, 6, 5, 4, 3, 2, 1, 0);
28579   __rev1.val[2] = __builtin_shufflevector(__p1.val[2], __p1.val[2], 7, 6, 5, 4, 3, 2, 1, 0);
28580   __rev1.val[3] = __builtin_shufflevector(__p1.val[3], __p1.val[3], 7, 6, 5, 4, 3, 2, 1, 0);
28590 __ai poly8x8x2_t vtrn_p8(poly8x8_t __p0, poly8x8_t __p1) {
28592   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 4);
28596 __ai poly8x8x2_t vtrn_p8(poly8x8_t __p0, poly8x8_t __p1) {
28598   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28609 __ai poly16x4x2_t vtrn_p16(poly16x4_t __p0, poly16x4_t __p1) {
28611   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 5);
28615 __ai poly16x4x2_t vtrn_p16(poly16x4_t __p0, poly16x4_t __p1) {
28617   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28628 __ai poly8x16x2_t vtrnq_p8(poly8x16_t __p0, poly8x16_t __p1) {
28630   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 36);
28634 __ai poly8x16x2_t vtrnq_p8(poly8x16_t __p0, poly8x16_t __p1) {
28636   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
28647 __ai poly16x8x2_t vtrnq_p16(poly16x8_t __p0, poly16x8_t __p1) {
28649   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 37);
28653 __ai poly16x8x2_t vtrnq_p16(poly16x8_t __p0, poly16x8_t __p1) {
28655   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28666 __ai uint8x16x2_t vtrnq_u8(uint8x16_t __p0, uint8x16_t __p1) {
28668   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 48);
28672 __ai uint8x16x2_t vtrnq_u8(uint8x16_t __p0, uint8x16_t __p1) {
28674   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
28685 __ai uint32x4x2_t vtrnq_u32(uint32x4_t __p0, uint32x4_t __p1) {
28687   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 50);
28691 __ai uint32x4x2_t vtrnq_u32(uint32x4_t __p0, uint32x4_t __p1) {
28693   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28704 __ai uint16x8x2_t vtrnq_u16(uint16x8_t __p0, uint16x8_t __p1) {
28706   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 49);
28710 __ai uint16x8x2_t vtrnq_u16(uint16x8_t __p0, uint16x8_t __p1) {
28712   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28723 __ai int8x16x2_t vtrnq_s8(int8x16_t __p0, int8x16_t __p1) {
28725   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 32);
28729 __ai int8x16x2_t vtrnq_s8(int8x16_t __p0, int8x16_t __p1) {
28731   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
28742 __ai float32x4x2_t vtrnq_f32(float32x4_t __p0, float32x4_t __p1) {
28744   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 41);
28748 __ai float32x4x2_t vtrnq_f32(float32x4_t __p0, float32x4_t __p1) {
28750   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28761 __ai int32x4x2_t vtrnq_s32(int32x4_t __p0, int32x4_t __p1) {
28763   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 34);
28767 __ai int32x4x2_t vtrnq_s32(int32x4_t __p0, int32x4_t __p1) {
28769   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28780 __ai int16x8x2_t vtrnq_s16(int16x8_t __p0, int16x8_t __p1) {
28782   __builtin_neon_vtrnq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 33);
28786 __ai int16x8x2_t vtrnq_s16(int16x8_t __p0, int16x8_t __p1) {
28788   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28799 __ai uint8x8x2_t vtrn_u8(uint8x8_t __p0, uint8x8_t __p1) {
28801   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 16);
28805 __ai uint8x8x2_t vtrn_u8(uint8x8_t __p0, uint8x8_t __p1) {
28807   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28818 __ai uint32x2x2_t vtrn_u32(uint32x2_t __p0, uint32x2_t __p1) {
28820   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 18);
28824 __ai uint32x2x2_t vtrn_u32(uint32x2_t __p0, uint32x2_t __p1) {
28826   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
28837 __ai uint16x4x2_t vtrn_u16(uint16x4_t __p0, uint16x4_t __p1) {
28839   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 17);
28843 __ai uint16x4x2_t vtrn_u16(uint16x4_t __p0, uint16x4_t __p1) {
28845   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28856 __ai int8x8x2_t vtrn_s8(int8x8_t __p0, int8x8_t __p1) {
28858   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 0);
28862 __ai int8x8x2_t vtrn_s8(int8x8_t __p0, int8x8_t __p1) {
28864   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28875 __ai float32x2x2_t vtrn_f32(float32x2_t __p0, float32x2_t __p1) {
28877   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 9);
28881 __ai float32x2x2_t vtrn_f32(float32x2_t __p0, float32x2_t __p1) {
28883   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
28894 __ai int32x2x2_t vtrn_s32(int32x2_t __p0, int32x2_t __p1) {
28896   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 2);
28900 __ai int32x2x2_t vtrn_s32(int32x2_t __p0, int32x2_t __p1) {
28902   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
28913 __ai int16x4x2_t vtrn_s16(int16x4_t __p0, int16x4_t __p1) {
28915   __builtin_neon_vtrn_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 1);
28919 __ai int16x4x2_t vtrn_s16(int16x4_t __p0, int16x4_t __p1) {
28921   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28932 __ai uint8x8_t vtst_p8(poly8x8_t __p0, poly8x8_t __p1) {
28934   __ret = (uint8x8_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
28938 __ai uint8x8_t vtst_p8(poly8x8_t __p0, poly8x8_t __p1) {
28940   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
28949 __ai uint16x4_t vtst_p16(poly16x4_t __p0, poly16x4_t __p1) {
28951   __ret = (uint16x4_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
28955 __ai uint16x4_t vtst_p16(poly16x4_t __p0, poly16x4_t __p1) {
28957   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
28966 __ai uint8x16_t vtstq_p8(poly8x16_t __p0, poly8x16_t __p1) {
28968   __ret = (uint8x16_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
28972 __ai uint8x16_t vtstq_p8(poly8x16_t __p0, poly8x16_t __p1) {
28974   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
28983 __ai uint16x8_t vtstq_p16(poly16x8_t __p0, poly16x8_t __p1) {
28985   __ret = (uint16x8_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
28989 __ai uint16x8_t vtstq_p16(poly16x8_t __p0, poly16x8_t __p1) {
28991   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29000 __ai uint8x16_t vtstq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29002   __ret = (uint8x16_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
29006 __ai uint8x16_t vtstq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29008   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29017 __ai uint32x4_t vtstq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29019   __ret = (uint32x4_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
29023 __ai uint32x4_t vtstq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29025   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29034 __ai uint16x8_t vtstq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29036   __ret = (uint16x8_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
29040 __ai uint16x8_t vtstq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29042   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29051 __ai uint8x16_t vtstq_s8(int8x16_t __p0, int8x16_t __p1) {
29053   __ret = (uint8x16_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
29057 __ai uint8x16_t vtstq_s8(int8x16_t __p0, int8x16_t __p1) {
29059   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29068 __ai uint32x4_t vtstq_s32(int32x4_t __p0, int32x4_t __p1) {
29070   __ret = (uint32x4_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
29074 __ai uint32x4_t vtstq_s32(int32x4_t __p0, int32x4_t __p1) {
29076   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29085 __ai uint16x8_t vtstq_s16(int16x8_t __p0, int16x8_t __p1) {
29087   __ret = (uint16x8_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
29091 __ai uint16x8_t vtstq_s16(int16x8_t __p0, int16x8_t __p1) {
29093   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29102 __ai uint8x8_t vtst_u8(uint8x8_t __p0, uint8x8_t __p1) {
29104   __ret = (uint8x8_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
29108 __ai uint8x8_t vtst_u8(uint8x8_t __p0, uint8x8_t __p1) {
29110   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29119 __ai uint32x2_t vtst_u32(uint32x2_t __p0, uint32x2_t __p1) {
29121   __ret = (uint32x2_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
29125 __ai uint32x2_t vtst_u32(uint32x2_t __p0, uint32x2_t __p1) {
29127   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29136 __ai uint16x4_t vtst_u16(uint16x4_t __p0, uint16x4_t __p1) {
29138   __ret = (uint16x4_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
29142 __ai uint16x4_t vtst_u16(uint16x4_t __p0, uint16x4_t __p1) {
29144   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29153 __ai uint8x8_t vtst_s8(int8x8_t __p0, int8x8_t __p1) {
29155   __ret = (uint8x8_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
29159 __ai uint8x8_t vtst_s8(int8x8_t __p0, int8x8_t __p1) {
29161   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29170 __ai uint32x2_t vtst_s32(int32x2_t __p0, int32x2_t __p1) {
29172   __ret = (uint32x2_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
29176 __ai uint32x2_t vtst_s32(int32x2_t __p0, int32x2_t __p1) {
29178   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29187 __ai uint16x4_t vtst_s16(int16x4_t __p0, int16x4_t __p1) {
29189   __ret = (uint16x4_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
29193 __ai uint16x4_t vtst_s16(int16x4_t __p0, int16x4_t __p1) {
29195   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29204 __ai poly8x8x2_t vuzp_p8(poly8x8_t __p0, poly8x8_t __p1) {
29206   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 4);
29210 __ai poly8x8x2_t vuzp_p8(poly8x8_t __p0, poly8x8_t __p1) {
29212   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29223 __ai poly16x4x2_t vuzp_p16(poly16x4_t __p0, poly16x4_t __p1) {
29225   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 5);
29229 __ai poly16x4x2_t vuzp_p16(poly16x4_t __p0, poly16x4_t __p1) {
29231   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29242 __ai poly8x16x2_t vuzpq_p8(poly8x16_t __p0, poly8x16_t __p1) {
29244   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 36);
29248 __ai poly8x16x2_t vuzpq_p8(poly8x16_t __p0, poly8x16_t __p1) {
29250   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29261 __ai poly16x8x2_t vuzpq_p16(poly16x8_t __p0, poly16x8_t __p1) {
29263   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 37);
29267 __ai poly16x8x2_t vuzpq_p16(poly16x8_t __p0, poly16x8_t __p1) {
29269   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29280 __ai uint8x16x2_t vuzpq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29282   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 48);
29286 __ai uint8x16x2_t vuzpq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29288   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29299 __ai uint32x4x2_t vuzpq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29301   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 50);
29305 __ai uint32x4x2_t vuzpq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29307   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29318 __ai uint16x8x2_t vuzpq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29320   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 49);
29324 __ai uint16x8x2_t vuzpq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29326   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29337 __ai int8x16x2_t vuzpq_s8(int8x16_t __p0, int8x16_t __p1) {
29339   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 32);
29343 __ai int8x16x2_t vuzpq_s8(int8x16_t __p0, int8x16_t __p1) {
29345   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29356 __ai float32x4x2_t vuzpq_f32(float32x4_t __p0, float32x4_t __p1) {
29358   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 41);
29362 __ai float32x4x2_t vuzpq_f32(float32x4_t __p0, float32x4_t __p1) {
29364   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29375 __ai int32x4x2_t vuzpq_s32(int32x4_t __p0, int32x4_t __p1) {
29377   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 34);
29381 __ai int32x4x2_t vuzpq_s32(int32x4_t __p0, int32x4_t __p1) {
29383   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29394 __ai int16x8x2_t vuzpq_s16(int16x8_t __p0, int16x8_t __p1) {
29396   __builtin_neon_vuzpq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 33);
29400 __ai int16x8x2_t vuzpq_s16(int16x8_t __p0, int16x8_t __p1) {
29402   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29413 __ai uint8x8x2_t vuzp_u8(uint8x8_t __p0, uint8x8_t __p1) {
29415   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 16);
29419 __ai uint8x8x2_t vuzp_u8(uint8x8_t __p0, uint8x8_t __p1) {
29421   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29432 __ai uint32x2x2_t vuzp_u32(uint32x2_t __p0, uint32x2_t __p1) {
29434   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 18);
29438 __ai uint32x2x2_t vuzp_u32(uint32x2_t __p0, uint32x2_t __p1) {
29440   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29451 __ai uint16x4x2_t vuzp_u16(uint16x4_t __p0, uint16x4_t __p1) {
29453   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 17);
29457 __ai uint16x4x2_t vuzp_u16(uint16x4_t __p0, uint16x4_t __p1) {
29459   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29470 __ai int8x8x2_t vuzp_s8(int8x8_t __p0, int8x8_t __p1) {
29472   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 0);
29476 __ai int8x8x2_t vuzp_s8(int8x8_t __p0, int8x8_t __p1) {
29478   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29489 __ai float32x2x2_t vuzp_f32(float32x2_t __p0, float32x2_t __p1) {
29491   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 9);
29495 __ai float32x2x2_t vuzp_f32(float32x2_t __p0, float32x2_t __p1) {
29497   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29508 __ai int32x2x2_t vuzp_s32(int32x2_t __p0, int32x2_t __p1) {
29510   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 2);
29514 __ai int32x2x2_t vuzp_s32(int32x2_t __p0, int32x2_t __p1) {
29516   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29527 __ai int16x4x2_t vuzp_s16(int16x4_t __p0, int16x4_t __p1) {
29529   __builtin_neon_vuzp_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 1);
29533 __ai int16x4x2_t vuzp_s16(int16x4_t __p0, int16x4_t __p1) {
29535   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29546 __ai poly8x8x2_t vzip_p8(poly8x8_t __p0, poly8x8_t __p1) {
29548   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 4);
29552 __ai poly8x8x2_t vzip_p8(poly8x8_t __p0, poly8x8_t __p1) {
29554   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29565 __ai poly16x4x2_t vzip_p16(poly16x4_t __p0, poly16x4_t __p1) {
29567   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 5);
29571 __ai poly16x4x2_t vzip_p16(poly16x4_t __p0, poly16x4_t __p1) {
29573   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29584 __ai poly8x16x2_t vzipq_p8(poly8x16_t __p0, poly8x16_t __p1) {
29586   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 36);
29590 __ai poly8x16x2_t vzipq_p8(poly8x16_t __p0, poly8x16_t __p1) {
29592   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29603 __ai poly16x8x2_t vzipq_p16(poly16x8_t __p0, poly16x8_t __p1) {
29605   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 37);
29609 __ai poly16x8x2_t vzipq_p16(poly16x8_t __p0, poly16x8_t __p1) {
29611   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29622 __ai uint8x16x2_t vzipq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29624   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 48);
29628 __ai uint8x16x2_t vzipq_u8(uint8x16_t __p0, uint8x16_t __p1) {
29630   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29641 __ai uint32x4x2_t vzipq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29643   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 50);
29647 __ai uint32x4x2_t vzipq_u32(uint32x4_t __p0, uint32x4_t __p1) {
29649   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29660 __ai uint16x8x2_t vzipq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29662   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 49);
29666 __ai uint16x8x2_t vzipq_u16(uint16x8_t __p0, uint16x8_t __p1) {
29668   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29679 __ai int8x16x2_t vzipq_s8(int8x16_t __p0, int8x16_t __p1) {
29681   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 32);
29685 __ai int8x16x2_t vzipq_s8(int8x16_t __p0, int8x16_t __p1) {
29687   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
29698 __ai float32x4x2_t vzipq_f32(float32x4_t __p0, float32x4_t __p1) {
29700   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 41);
29704 __ai float32x4x2_t vzipq_f32(float32x4_t __p0, float32x4_t __p1) {
29706   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29717 __ai int32x4x2_t vzipq_s32(int32x4_t __p0, int32x4_t __p1) {
29719   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 34);
29723 __ai int32x4x2_t vzipq_s32(int32x4_t __p0, int32x4_t __p1) {
29725   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29736 __ai int16x8x2_t vzipq_s16(int16x8_t __p0, int16x8_t __p1) {
29738   __builtin_neon_vzipq_v(&__ret, (int8x16_t)__p0, (int8x16_t)__p1, 33);
29742 __ai int16x8x2_t vzipq_s16(int16x8_t __p0, int16x8_t __p1) {
29744   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29755 __ai uint8x8x2_t vzip_u8(uint8x8_t __p0, uint8x8_t __p1) {
29757   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 16);
29761 __ai uint8x8x2_t vzip_u8(uint8x8_t __p0, uint8x8_t __p1) {
29763   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29774 __ai uint32x2x2_t vzip_u32(uint32x2_t __p0, uint32x2_t __p1) {
29776   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 18);
29780 __ai uint32x2x2_t vzip_u32(uint32x2_t __p0, uint32x2_t __p1) {
29782   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29793 __ai uint16x4x2_t vzip_u16(uint16x4_t __p0, uint16x4_t __p1) {
29795   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 17);
29799 __ai uint16x4x2_t vzip_u16(uint16x4_t __p0, uint16x4_t __p1) {
29801   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
29812 __ai int8x8x2_t vzip_s8(int8x8_t __p0, int8x8_t __p1) {
29814   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 0);
29818 __ai int8x8x2_t vzip_s8(int8x8_t __p0, int8x8_t __p1) {
29820   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
29831 __ai float32x2x2_t vzip_f32(float32x2_t __p0, float32x2_t __p1) {
29833   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 9);
29837 __ai float32x2x2_t vzip_f32(float32x2_t __p0, float32x2_t __p1) {
29839   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29850 __ai int32x2x2_t vzip_s32(int32x2_t __p0, int32x2_t __p1) {
29852   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 2);
29856 __ai int32x2x2_t vzip_s32(int32x2_t __p0, int32x2_t __p1) {
29858   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
29869 __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
29871   __builtin_neon_vzip_v(&__ret, (int8x8_t)__p0, (int8x8_t)__p1, 1);
29875 __ai int16x4x2_t vzip_s16(int16x4_t __p0, int16x4_t __p1) {
29877   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
39575 __ai uint8x16_t vaesdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
39577   __ret = (uint8x16_t) __builtin_neon_vaesdq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
39581 __ai uint8x16_t vaesdq_u8(uint8x16_t __p0, uint8x16_t __p1) {
39583   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
39592 __ai uint8x16_t vaeseq_u8(uint8x16_t __p0, uint8x16_t __p1) {
39594   __ret = (uint8x16_t) __builtin_neon_vaeseq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
39598 __ai uint8x16_t vaeseq_u8(uint8x16_t __p0, uint8x16_t __p1) {
39600   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
39641 __ai uint32x4_t vsha1cq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
39643   __ret = (uint32x4_t) __builtin_neon_vsha1cq_u32((int8x16_t)__p0, __p1, (int8x16_t)__p2);
39647 __ai uint32x4_t vsha1cq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
39651   __ret = (uint32x4_t) __builtin_neon_vsha1cq_u32((int8x16_t)__rev0, __p1, (int8x16_t)__rev2);
39672 __ai uint32x4_t vsha1mq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
39674   __ret = (uint32x4_t) __builtin_neon_vsha1mq_u32((int8x16_t)__p0, __p1, (int8x16_t)__p2);
39678 __ai uint32x4_t vsha1mq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
39682   __ret = (uint32x4_t) __builtin_neon_vsha1mq_u32((int8x16_t)__rev0, __p1, (int8x16_t)__rev2);
39689 __ai uint32x4_t vsha1pq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
39691   __ret = (uint32x4_t) __builtin_neon_vsha1pq_u32((int8x16_t)__p0, __p1, (int8x16_t)__p2);
39695 __ai uint32x4_t vsha1pq_u32(uint32x4_t __p0, uint32_t __p1, uint32x4_t __p2) {
39699   __ret = (uint32x4_t) __builtin_neon_vsha1pq_u32((int8x16_t)__rev0, __p1, (int8x16_t)__rev2);
39706 __ai uint32x4_t vsha1su0q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
39708   __ret = (uint32x4_t) __builtin_neon_vsha1su0q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
39712 __ai uint32x4_t vsha1su0q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
39714   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
39724 __ai uint32x4_t vsha1su1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
39726   __ret = (uint32x4_t) __builtin_neon_vsha1su1q_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
39730 __ai uint32x4_t vsha1su1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
39732   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
39741 __ai uint32x4_t vsha256hq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
39743   __ret = (uint32x4_t) __builtin_neon_vsha256hq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
39747 __ai uint32x4_t vsha256hq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
39749   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
39759 __ai uint32x4_t vsha256h2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
39761   __ret = (uint32x4_t) __builtin_neon_vsha256h2q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
39765 __ai uint32x4_t vsha256h2q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
39767   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
39777 __ai uint32x4_t vsha256su0q_u32(uint32x4_t __p0, uint32x4_t __p1) {
39779   __ret = (uint32x4_t) __builtin_neon_vsha256su0q_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
39783 __ai uint32x4_t vsha256su0q_u32(uint32x4_t __p0, uint32x4_t __p1) {
39785   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
39794 __ai uint32x4_t vsha256su1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
39796   __ret = (uint32x4_t) __builtin_neon_vsha256su1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 50);
39800 __ai uint32x4_t vsha256su1q_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
39802   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
39814 __ai float64x2_t vabdq_f64(float64x2_t __p0, float64x2_t __p1) {
39816   __ret = (float64x2_t) __builtin_neon_vabdq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
39820 __ai float64x2_t vabdq_f64(float64x2_t __p0, float64x2_t __p1) {
39822   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
39831 __ai float64x1_t vabd_f64(float64x1_t __p0, float64x1_t __p1) {
39833   __ret = (float64x1_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
39837 __ai float64x1_t vabd_f64(float64x1_t __p0, float64x1_t __p1) {
39839   __ret = (float64x1_t) __builtin_neon_vabd_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
39845 __ai float64_t vabdd_f64(float64_t __p0, float64_t __p1) {
39847   __ret = (float64_t) __builtin_neon_vabdd_f64(__p0, __p1);
39851 __ai float64_t vabdd_f64(float64_t __p0, float64_t __p1) {
39853   __ret = (float64_t) __builtin_neon_vabdd_f64(__p0, __p1);
39859 __ai float32_t vabds_f32(float32_t __p0, float32_t __p1) {
39861   __ret = (float32_t) __builtin_neon_vabds_f32(__p0, __p1);
39865 __ai float32_t vabds_f32(float32_t __p0, float32_t __p1) {
39867   __ret = (float32_t) __builtin_neon_vabds_f32(__p0, __p1);
39947 __ai float64x2_t vaddq_f64(float64x2_t __p0, float64x2_t __p1) {
39949   __ret = __p0 + __p1;
39953 __ai float64x2_t vaddq_f64(float64x2_t __p0, float64x2_t __p1) {
39955   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
39964 __ai float64x1_t vadd_f64(float64x1_t __p0, float64x1_t __p1) {
39966   __ret = __p0 + __p1;
39970 __ai float64x1_t vadd_f64(float64x1_t __p0, float64x1_t __p1) {
39972   __ret = __p0 + __p1;
39978 __ai uint64_t vaddd_u64(uint64_t __p0, uint64_t __p1) {
39980   __ret = (uint64_t) __builtin_neon_vaddd_u64(__p0, __p1);
39984 __ai uint64_t vaddd_u64(uint64_t __p0, uint64_t __p1) {
39986   __ret = (uint64_t) __builtin_neon_vaddd_u64(__p0, __p1);
39992 __ai int64_t vaddd_s64(int64_t __p0, int64_t __p1) {
39994   __ret = (int64_t) __builtin_neon_vaddd_s64(__p0, __p1);
39998 __ai int64_t vaddd_s64(int64_t __p0, int64_t __p1) {
40000   __ret = (int64_t) __builtin_neon_vaddd_s64(__p0, __p1);
40006 __ai uint16x8_t vaddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
40008   __ret = vcombine_u16(__p0, vaddhn_u32(__p1, __p2));
40012 __ai uint16x8_t vaddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
40014   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40024 __ai uint32x4_t vaddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
40026   __ret = vcombine_u32(__p0, vaddhn_u64(__p1, __p2));
40030 __ai uint32x4_t vaddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
40032   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40042 __ai uint8x16_t vaddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
40044   __ret = vcombine_u8(__p0, vaddhn_u16(__p1, __p2));
40048 __ai uint8x16_t vaddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
40050   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
40060 __ai int16x8_t vaddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
40062   __ret = vcombine_s16(__p0, vaddhn_s32(__p1, __p2));
40066 __ai int16x8_t vaddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
40068   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
40078 __ai int32x4_t vaddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
40080   __ret = vcombine_s32(__p0, vaddhn_s64(__p1, __p2));
40084 __ai int32x4_t vaddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
40086   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40096 __ai int8x16_t vaddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
40098   __ret = vcombine_s8(__p0, vaddhn_s16(__p1, __p2));
40102 __ai int8x16_t vaddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
40104   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
40549 __ai poly64x1_t vbsl_p64(uint64x1_t __p0, poly64x1_t __p1, poly64x1_t __p2) {
40551   __ret = (poly64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 6);
40555 __ai poly64x1_t vbsl_p64(uint64x1_t __p0, poly64x1_t __p1, poly64x1_t __p2) {
40557   __ret = (poly64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 6);
40563 __ai poly64x2_t vbslq_p64(uint64x2_t __p0, poly64x2_t __p1, poly64x2_t __p2) {
40565   __ret = (poly64x2_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 38);
40569 __ai poly64x2_t vbslq_p64(uint64x2_t __p0, poly64x2_t __p1, poly64x2_t __p2) {
40571   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40581 __ai float64x2_t vbslq_f64(uint64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
40583   __ret = (float64x2_t) __builtin_neon_vbslq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
40587 __ai float64x2_t vbslq_f64(uint64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
40589   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40599 __ai float64x1_t vbsl_f64(uint64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
40601   __ret = (float64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
40605 __ai float64x1_t vbsl_f64(uint64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
40607   __ret = (float64x1_t) __builtin_neon_vbsl_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
40613 __ai uint64x2_t vcageq_f64(float64x2_t __p0, float64x2_t __p1) {
40615   __ret = (uint64x2_t) __builtin_neon_vcageq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
40619 __ai uint64x2_t vcageq_f64(float64x2_t __p0, float64x2_t __p1) {
40621   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40630 __ai uint64x1_t vcage_f64(float64x1_t __p0, float64x1_t __p1) {
40632   __ret = (uint64x1_t) __builtin_neon_vcage_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
40636 __ai uint64x1_t vcage_f64(float64x1_t __p0, float64x1_t __p1) {
40638   __ret = (uint64x1_t) __builtin_neon_vcage_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
40644 __ai uint64_t vcaged_f64(float64_t __p0, float64_t __p1) {
40646   __ret = (uint64_t) __builtin_neon_vcaged_f64(__p0, __p1);
40650 __ai uint64_t vcaged_f64(float64_t __p0, float64_t __p1) {
40652   __ret = (uint64_t) __builtin_neon_vcaged_f64(__p0, __p1);
40658 __ai uint32_t vcages_f32(float32_t __p0, float32_t __p1) {
40660   __ret = (uint32_t) __builtin_neon_vcages_f32(__p0, __p1);
40664 __ai uint32_t vcages_f32(float32_t __p0, float32_t __p1) {
40666   __ret = (uint32_t) __builtin_neon_vcages_f32(__p0, __p1);
40672 __ai uint64x2_t vcagtq_f64(float64x2_t __p0, float64x2_t __p1) {
40674   __ret = (uint64x2_t) __builtin_neon_vcagtq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
40678 __ai uint64x2_t vcagtq_f64(float64x2_t __p0, float64x2_t __p1) {
40680   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40689 __ai uint64x1_t vcagt_f64(float64x1_t __p0, float64x1_t __p1) {
40691   __ret = (uint64x1_t) __builtin_neon_vcagt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
40695 __ai uint64x1_t vcagt_f64(float64x1_t __p0, float64x1_t __p1) {
40697   __ret = (uint64x1_t) __builtin_neon_vcagt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
40703 __ai uint64_t vcagtd_f64(float64_t __p0, float64_t __p1) {
40705   __ret = (uint64_t) __builtin_neon_vcagtd_f64(__p0, __p1);
40709 __ai uint64_t vcagtd_f64(float64_t __p0, float64_t __p1) {
40711   __ret = (uint64_t) __builtin_neon_vcagtd_f64(__p0, __p1);
40717 __ai uint32_t vcagts_f32(float32_t __p0, float32_t __p1) {
40719   __ret = (uint32_t) __builtin_neon_vcagts_f32(__p0, __p1);
40723 __ai uint32_t vcagts_f32(float32_t __p0, float32_t __p1) {
40725   __ret = (uint32_t) __builtin_neon_vcagts_f32(__p0, __p1);
40731 __ai uint64x2_t vcaleq_f64(float64x2_t __p0, float64x2_t __p1) {
40733   __ret = (uint64x2_t) __builtin_neon_vcaleq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
40737 __ai uint64x2_t vcaleq_f64(float64x2_t __p0, float64x2_t __p1) {
40739   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40748 __ai uint64x1_t vcale_f64(float64x1_t __p0, float64x1_t __p1) {
40750   __ret = (uint64x1_t) __builtin_neon_vcale_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
40754 __ai uint64x1_t vcale_f64(float64x1_t __p0, float64x1_t __p1) {
40756   __ret = (uint64x1_t) __builtin_neon_vcale_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
40762 __ai uint64_t vcaled_f64(float64_t __p0, float64_t __p1) {
40764   __ret = (uint64_t) __builtin_neon_vcaled_f64(__p0, __p1);
40768 __ai uint64_t vcaled_f64(float64_t __p0, float64_t __p1) {
40770   __ret = (uint64_t) __builtin_neon_vcaled_f64(__p0, __p1);
40776 __ai uint32_t vcales_f32(float32_t __p0, float32_t __p1) {
40778   __ret = (uint32_t) __builtin_neon_vcales_f32(__p0, __p1);
40782 __ai uint32_t vcales_f32(float32_t __p0, float32_t __p1) {
40784   __ret = (uint32_t) __builtin_neon_vcales_f32(__p0, __p1);
40790 __ai uint64x2_t vcaltq_f64(float64x2_t __p0, float64x2_t __p1) {
40792   __ret = (uint64x2_t) __builtin_neon_vcaltq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
40796 __ai uint64x2_t vcaltq_f64(float64x2_t __p0, float64x2_t __p1) {
40798   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40807 __ai uint64x1_t vcalt_f64(float64x1_t __p0, float64x1_t __p1) {
40809   __ret = (uint64x1_t) __builtin_neon_vcalt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
40813 __ai uint64x1_t vcalt_f64(float64x1_t __p0, float64x1_t __p1) {
40815   __ret = (uint64x1_t) __builtin_neon_vcalt_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
40821 __ai uint64_t vcaltd_f64(float64_t __p0, float64_t __p1) {
40823   __ret = (uint64_t) __builtin_neon_vcaltd_f64(__p0, __p1);
40827 __ai uint64_t vcaltd_f64(float64_t __p0, float64_t __p1) {
40829   __ret = (uint64_t) __builtin_neon_vcaltd_f64(__p0, __p1);
40835 __ai uint32_t vcalts_f32(float32_t __p0, float32_t __p1) {
40837   __ret = (uint32_t) __builtin_neon_vcalts_f32(__p0, __p1);
40841 __ai uint32_t vcalts_f32(float32_t __p0, float32_t __p1) {
40843   __ret = (uint32_t) __builtin_neon_vcalts_f32(__p0, __p1);
40849 __ai uint64x1_t vceq_p64(poly64x1_t __p0, poly64x1_t __p1) {
40851   __ret = (uint64x1_t)(__p0 == __p1);
40855 __ai uint64x1_t vceq_p64(poly64x1_t __p0, poly64x1_t __p1) {
40857   __ret = (uint64x1_t)(__p0 == __p1);
40863 __ai uint64x2_t vceqq_p64(poly64x2_t __p0, poly64x2_t __p1) {
40865   __ret = (uint64x2_t)(__p0 == __p1);
40869 __ai uint64x2_t vceqq_p64(poly64x2_t __p0, poly64x2_t __p1) {
40871   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40880 __ai uint64x2_t vceqq_u64(uint64x2_t __p0, uint64x2_t __p1) {
40882   __ret = (uint64x2_t)(__p0 == __p1);
40886 __ai uint64x2_t vceqq_u64(uint64x2_t __p0, uint64x2_t __p1) {
40888   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40897 __ai uint64x2_t vceqq_f64(float64x2_t __p0, float64x2_t __p1) {
40899   __ret = (uint64x2_t)(__p0 == __p1);
40903 __ai uint64x2_t vceqq_f64(float64x2_t __p0, float64x2_t __p1) {
40905   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40914 __ai uint64x2_t vceqq_s64(int64x2_t __p0, int64x2_t __p1) {
40916   __ret = (uint64x2_t)(__p0 == __p1);
40920 __ai uint64x2_t vceqq_s64(int64x2_t __p0, int64x2_t __p1) {
40922   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
40931 __ai uint64x1_t vceq_u64(uint64x1_t __p0, uint64x1_t __p1) {
40933   __ret = (uint64x1_t)(__p0 == __p1);
40937 __ai uint64x1_t vceq_u64(uint64x1_t __p0, uint64x1_t __p1) {
40939   __ret = (uint64x1_t)(__p0 == __p1);
40945 __ai uint64x1_t vceq_f64(float64x1_t __p0, float64x1_t __p1) {
40947   __ret = (uint64x1_t)(__p0 == __p1);
40951 __ai uint64x1_t vceq_f64(float64x1_t __p0, float64x1_t __p1) {
40953   __ret = (uint64x1_t)(__p0 == __p1);
40959 __ai uint64x1_t vceq_s64(int64x1_t __p0, int64x1_t __p1) {
40961   __ret = (uint64x1_t)(__p0 == __p1);
40965 __ai uint64x1_t vceq_s64(int64x1_t __p0, int64x1_t __p1) {
40967   __ret = (uint64x1_t)(__p0 == __p1);
40973 __ai uint64_t vceqd_u64(uint64_t __p0, uint64_t __p1) {
40975   __ret = (uint64_t) __builtin_neon_vceqd_u64(__p0, __p1);
40979 __ai uint64_t vceqd_u64(uint64_t __p0, uint64_t __p1) {
40981   __ret = (uint64_t) __builtin_neon_vceqd_u64(__p0, __p1);
40987 __ai int64_t vceqd_s64(int64_t __p0, int64_t __p1) {
40989   __ret = (int64_t) __builtin_neon_vceqd_s64(__p0, __p1);
40993 __ai int64_t vceqd_s64(int64_t __p0, int64_t __p1) {
40995   __ret = (int64_t) __builtin_neon_vceqd_s64(__p0, __p1);
41001 __ai uint64_t vceqd_f64(float64_t __p0, float64_t __p1) {
41003   __ret = (uint64_t) __builtin_neon_vceqd_f64(__p0, __p1);
41007 __ai uint64_t vceqd_f64(float64_t __p0, float64_t __p1) {
41009   __ret = (uint64_t) __builtin_neon_vceqd_f64(__p0, __p1);
41015 __ai uint32_t vceqs_f32(float32_t __p0, float32_t __p1) {
41017   __ret = (uint32_t) __builtin_neon_vceqs_f32(__p0, __p1);
41021 __ai uint32_t vceqs_f32(float32_t __p0, float32_t __p1) {
41023   __ret = (uint32_t) __builtin_neon_vceqs_f32(__p0, __p1);
41493 __ai uint64x2_t vcgeq_u64(uint64x2_t __p0, uint64x2_t __p1) {
41495   __ret = (uint64x2_t)(__p0 >= __p1);
41499 __ai uint64x2_t vcgeq_u64(uint64x2_t __p0, uint64x2_t __p1) {
41501   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41510 __ai uint64x2_t vcgeq_f64(float64x2_t __p0, float64x2_t __p1) {
41512   __ret = (uint64x2_t)(__p0 >= __p1);
41516 __ai uint64x2_t vcgeq_f64(float64x2_t __p0, float64x2_t __p1) {
41518   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41527 __ai uint64x2_t vcgeq_s64(int64x2_t __p0, int64x2_t __p1) {
41529   __ret = (uint64x2_t)(__p0 >= __p1);
41533 __ai uint64x2_t vcgeq_s64(int64x2_t __p0, int64x2_t __p1) {
41535   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41544 __ai uint64x1_t vcge_u64(uint64x1_t __p0, uint64x1_t __p1) {
41546   __ret = (uint64x1_t)(__p0 >= __p1);
41550 __ai uint64x1_t vcge_u64(uint64x1_t __p0, uint64x1_t __p1) {
41552   __ret = (uint64x1_t)(__p0 >= __p1);
41558 __ai uint64x1_t vcge_f64(float64x1_t __p0, float64x1_t __p1) {
41560   __ret = (uint64x1_t)(__p0 >= __p1);
41564 __ai uint64x1_t vcge_f64(float64x1_t __p0, float64x1_t __p1) {
41566   __ret = (uint64x1_t)(__p0 >= __p1);
41572 __ai uint64x1_t vcge_s64(int64x1_t __p0, int64x1_t __p1) {
41574   __ret = (uint64x1_t)(__p0 >= __p1);
41578 __ai uint64x1_t vcge_s64(int64x1_t __p0, int64x1_t __p1) {
41580   __ret = (uint64x1_t)(__p0 >= __p1);
41586 __ai int64_t vcged_s64(int64_t __p0, int64_t __p1) {
41588   __ret = (int64_t) __builtin_neon_vcged_s64(__p0, __p1);
41592 __ai int64_t vcged_s64(int64_t __p0, int64_t __p1) {
41594   __ret = (int64_t) __builtin_neon_vcged_s64(__p0, __p1);
41600 __ai uint64_t vcged_u64(uint64_t __p0, uint64_t __p1) {
41602   __ret = (uint64_t) __builtin_neon_vcged_u64(__p0, __p1);
41606 __ai uint64_t vcged_u64(uint64_t __p0, uint64_t __p1) {
41608   __ret = (uint64_t) __builtin_neon_vcged_u64(__p0, __p1);
41614 __ai uint64_t vcged_f64(float64_t __p0, float64_t __p1) {
41616   __ret = (uint64_t) __builtin_neon_vcged_f64(__p0, __p1);
41620 __ai uint64_t vcged_f64(float64_t __p0, float64_t __p1) {
41622   __ret = (uint64_t) __builtin_neon_vcged_f64(__p0, __p1);
41628 __ai uint32_t vcges_f32(float32_t __p0, float32_t __p1) {
41630   __ret = (uint32_t) __builtin_neon_vcges_f32(__p0, __p1);
41634 __ai uint32_t vcges_f32(float32_t __p0, float32_t __p1) {
41636   __ret = (uint32_t) __builtin_neon_vcges_f32(__p0, __p1);
41872 __ai uint64x2_t vcgtq_u64(uint64x2_t __p0, uint64x2_t __p1) {
41874   __ret = (uint64x2_t)(__p0 > __p1);
41878 __ai uint64x2_t vcgtq_u64(uint64x2_t __p0, uint64x2_t __p1) {
41880   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41889 __ai uint64x2_t vcgtq_f64(float64x2_t __p0, float64x2_t __p1) {
41891   __ret = (uint64x2_t)(__p0 > __p1);
41895 __ai uint64x2_t vcgtq_f64(float64x2_t __p0, float64x2_t __p1) {
41897   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41906 __ai uint64x2_t vcgtq_s64(int64x2_t __p0, int64x2_t __p1) {
41908   __ret = (uint64x2_t)(__p0 > __p1);
41912 __ai uint64x2_t vcgtq_s64(int64x2_t __p0, int64x2_t __p1) {
41914   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
41923 __ai uint64x1_t vcgt_u64(uint64x1_t __p0, uint64x1_t __p1) {
41925   __ret = (uint64x1_t)(__p0 > __p1);
41929 __ai uint64x1_t vcgt_u64(uint64x1_t __p0, uint64x1_t __p1) {
41931   __ret = (uint64x1_t)(__p0 > __p1);
41937 __ai uint64x1_t vcgt_f64(float64x1_t __p0, float64x1_t __p1) {
41939   __ret = (uint64x1_t)(__p0 > __p1);
41943 __ai uint64x1_t vcgt_f64(float64x1_t __p0, float64x1_t __p1) {
41945   __ret = (uint64x1_t)(__p0 > __p1);
41951 __ai uint64x1_t vcgt_s64(int64x1_t __p0, int64x1_t __p1) {
41953   __ret = (uint64x1_t)(__p0 > __p1);
41957 __ai uint64x1_t vcgt_s64(int64x1_t __p0, int64x1_t __p1) {
41959   __ret = (uint64x1_t)(__p0 > __p1);
41965 __ai int64_t vcgtd_s64(int64_t __p0, int64_t __p1) {
41967   __ret = (int64_t) __builtin_neon_vcgtd_s64(__p0, __p1);
41971 __ai int64_t vcgtd_s64(int64_t __p0, int64_t __p1) {
41973   __ret = (int64_t) __builtin_neon_vcgtd_s64(__p0, __p1);
41979 __ai uint64_t vcgtd_u64(uint64_t __p0, uint64_t __p1) {
41981   __ret = (uint64_t) __builtin_neon_vcgtd_u64(__p0, __p1);
41985 __ai uint64_t vcgtd_u64(uint64_t __p0, uint64_t __p1) {
41987   __ret = (uint64_t) __builtin_neon_vcgtd_u64(__p0, __p1);
41993 __ai uint64_t vcgtd_f64(float64_t __p0, float64_t __p1) {
41995   __ret = (uint64_t) __builtin_neon_vcgtd_f64(__p0, __p1);
41999 __ai uint64_t vcgtd_f64(float64_t __p0, float64_t __p1) {
42001   __ret = (uint64_t) __builtin_neon_vcgtd_f64(__p0, __p1);
42007 __ai uint32_t vcgts_f32(float32_t __p0, float32_t __p1) {
42009   __ret = (uint32_t) __builtin_neon_vcgts_f32(__p0, __p1);
42013 __ai uint32_t vcgts_f32(float32_t __p0, float32_t __p1) {
42015   __ret = (uint32_t) __builtin_neon_vcgts_f32(__p0, __p1);
42251 __ai uint64x2_t vcleq_u64(uint64x2_t __p0, uint64x2_t __p1) {
42253   __ret = (uint64x2_t)(__p0 <= __p1);
42257 __ai uint64x2_t vcleq_u64(uint64x2_t __p0, uint64x2_t __p1) {
42259   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42268 __ai uint64x2_t vcleq_f64(float64x2_t __p0, float64x2_t __p1) {
42270   __ret = (uint64x2_t)(__p0 <= __p1);
42274 __ai uint64x2_t vcleq_f64(float64x2_t __p0, float64x2_t __p1) {
42276   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42285 __ai uint64x2_t vcleq_s64(int64x2_t __p0, int64x2_t __p1) {
42287   __ret = (uint64x2_t)(__p0 <= __p1);
42291 __ai uint64x2_t vcleq_s64(int64x2_t __p0, int64x2_t __p1) {
42293   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42302 __ai uint64x1_t vcle_u64(uint64x1_t __p0, uint64x1_t __p1) {
42304   __ret = (uint64x1_t)(__p0 <= __p1);
42308 __ai uint64x1_t vcle_u64(uint64x1_t __p0, uint64x1_t __p1) {
42310   __ret = (uint64x1_t)(__p0 <= __p1);
42316 __ai uint64x1_t vcle_f64(float64x1_t __p0, float64x1_t __p1) {
42318   __ret = (uint64x1_t)(__p0 <= __p1);
42322 __ai uint64x1_t vcle_f64(float64x1_t __p0, float64x1_t __p1) {
42324   __ret = (uint64x1_t)(__p0 <= __p1);
42330 __ai uint64x1_t vcle_s64(int64x1_t __p0, int64x1_t __p1) {
42332   __ret = (uint64x1_t)(__p0 <= __p1);
42336 __ai uint64x1_t vcle_s64(int64x1_t __p0, int64x1_t __p1) {
42338   __ret = (uint64x1_t)(__p0 <= __p1);
42344 __ai uint64_t vcled_u64(uint64_t __p0, uint64_t __p1) {
42346   __ret = (uint64_t) __builtin_neon_vcled_u64(__p0, __p1);
42350 __ai uint64_t vcled_u64(uint64_t __p0, uint64_t __p1) {
42352   __ret = (uint64_t) __builtin_neon_vcled_u64(__p0, __p1);
42358 __ai int64_t vcled_s64(int64_t __p0, int64_t __p1) {
42360   __ret = (int64_t) __builtin_neon_vcled_s64(__p0, __p1);
42364 __ai int64_t vcled_s64(int64_t __p0, int64_t __p1) {
42366   __ret = (int64_t) __builtin_neon_vcled_s64(__p0, __p1);
42372 __ai uint64_t vcled_f64(float64_t __p0, float64_t __p1) {
42374   __ret = (uint64_t) __builtin_neon_vcled_f64(__p0, __p1);
42378 __ai uint64_t vcled_f64(float64_t __p0, float64_t __p1) {
42380   __ret = (uint64_t) __builtin_neon_vcled_f64(__p0, __p1);
42386 __ai uint32_t vcles_f32(float32_t __p0, float32_t __p1) {
42388   __ret = (uint32_t) __builtin_neon_vcles_f32(__p0, __p1);
42392 __ai uint32_t vcles_f32(float32_t __p0, float32_t __p1) {
42394   __ret = (uint32_t) __builtin_neon_vcles_f32(__p0, __p1);
42630 __ai uint64x2_t vcltq_u64(uint64x2_t __p0, uint64x2_t __p1) {
42632   __ret = (uint64x2_t)(__p0 < __p1);
42636 __ai uint64x2_t vcltq_u64(uint64x2_t __p0, uint64x2_t __p1) {
42638   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42647 __ai uint64x2_t vcltq_f64(float64x2_t __p0, float64x2_t __p1) {
42649   __ret = (uint64x2_t)(__p0 < __p1);
42653 __ai uint64x2_t vcltq_f64(float64x2_t __p0, float64x2_t __p1) {
42655   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42664 __ai uint64x2_t vcltq_s64(int64x2_t __p0, int64x2_t __p1) {
42666   __ret = (uint64x2_t)(__p0 < __p1);
42670 __ai uint64x2_t vcltq_s64(int64x2_t __p0, int64x2_t __p1) {
42672   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
42681 __ai uint64x1_t vclt_u64(uint64x1_t __p0, uint64x1_t __p1) {
42683   __ret = (uint64x1_t)(__p0 < __p1);
42687 __ai uint64x1_t vclt_u64(uint64x1_t __p0, uint64x1_t __p1) {
42689   __ret = (uint64x1_t)(__p0 < __p1);
42695 __ai uint64x1_t vclt_f64(float64x1_t __p0, float64x1_t __p1) {
42697   __ret = (uint64x1_t)(__p0 < __p1);
42701 __ai uint64x1_t vclt_f64(float64x1_t __p0, float64x1_t __p1) {
42703   __ret = (uint64x1_t)(__p0 < __p1);
42709 __ai uint64x1_t vclt_s64(int64x1_t __p0, int64x1_t __p1) {
42711   __ret = (uint64x1_t)(__p0 < __p1);
42715 __ai uint64x1_t vclt_s64(int64x1_t __p0, int64x1_t __p1) {
42717   __ret = (uint64x1_t)(__p0 < __p1);
42723 __ai uint64_t vcltd_u64(uint64_t __p0, uint64_t __p1) {
42725   __ret = (uint64_t) __builtin_neon_vcltd_u64(__p0, __p1);
42729 __ai uint64_t vcltd_u64(uint64_t __p0, uint64_t __p1) {
42731   __ret = (uint64_t) __builtin_neon_vcltd_u64(__p0, __p1);
42737 __ai int64_t vcltd_s64(int64_t __p0, int64_t __p1) {
42739   __ret = (int64_t) __builtin_neon_vcltd_s64(__p0, __p1);
42743 __ai int64_t vcltd_s64(int64_t __p0, int64_t __p1) {
42745   __ret = (int64_t) __builtin_neon_vcltd_s64(__p0, __p1);
42751 __ai uint64_t vcltd_f64(float64_t __p0, float64_t __p1) {
42753   __ret = (uint64_t) __builtin_neon_vcltd_f64(__p0, __p1);
42757 __ai uint64_t vcltd_f64(float64_t __p0, float64_t __p1) {
42759   __ret = (uint64_t) __builtin_neon_vcltd_f64(__p0, __p1);
42765 __ai uint32_t vclts_f32(float32_t __p0, float32_t __p1) {
42767   __ret = (uint32_t) __builtin_neon_vclts_f32(__p0, __p1);
42771 __ai uint32_t vclts_f32(float32_t __p0, float32_t __p1) {
42773   __ret = (uint32_t) __builtin_neon_vclts_f32(__p0, __p1);
43009 __ai poly64x2_t vcombine_p64(poly64x1_t __p0, poly64x1_t __p1) {
43011   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
43015 __ai poly64x2_t vcombine_p64(poly64x1_t __p0, poly64x1_t __p1) {
43017   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
43024 __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
43026   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
43030 __ai float64x2_t vcombine_f64(float64x1_t __p0, float64x1_t __p1) {
43032   __ret = __builtin_shufflevector(__p0, __p1, 0, 1);
44137 __ai float16x8_t vcvt_high_f16_f32(float16x4_t __p0, float32x4_t __p1) {
44139   __ret = vcombine_f16(__p0, vcvt_f16_f32(__p1));
44143 __ai float16x8_t vcvt_high_f16_f32(float16x4_t __p0, float32x4_t __p1) {
44145   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44170 __ai float32x4_t vcvt_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
44172   __ret = vcombine_f32(__p0, vcvt_f32_f64(__p1));
44176 __ai float32x4_t vcvt_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
44178   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
44203 #define vcvts_n_f32_u32(__p0, __p1) __extension__ ({ \
44206   __ret = (float32_t) __builtin_neon_vcvts_n_f32_u32(__s0, __p1); \
44210 #define vcvts_n_f32_u32(__p0, __p1) __extension__ ({ \
44213   __ret = (float32_t) __builtin_neon_vcvts_n_f32_u32(__s0, __p1); \
44219 #define vcvts_n_f32_s32(__p0, __p1) __extension__ ({ \
44222   __ret = (float32_t) __builtin_neon_vcvts_n_f32_s32(__s0, __p1); \
44226 #define vcvts_n_f32_s32(__p0, __p1) __extension__ ({ \
44229   __ret = (float32_t) __builtin_neon_vcvts_n_f32_s32(__s0, __p1); \
44235 #define vcvtq_n_f64_u64(__p0, __p1) __extension__ ({ \
44238   __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__s0, __p1, 51); \
44242 #define vcvtq_n_f64_u64(__p0, __p1) __extension__ ({ \
44246   __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__rev0, __p1, 51); \
44253 #define vcvtq_n_f64_s64(__p0, __p1) __extension__ ({ \
44256   __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__s0, __p1, 35); \
44260 #define vcvtq_n_f64_s64(__p0, __p1) __extension__ ({ \
44264   __ret = (float64x2_t) __builtin_neon_vcvtq_n_f64_v((int8x16_t)__rev0, __p1, 35); \
44271 #define vcvt_n_f64_u64(__p0, __p1) __extension__ ({ \
44274   __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 19); \
44278 #define vcvt_n_f64_u64(__p0, __p1) __extension__ ({ \
44281   __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 19); \
44287 #define vcvt_n_f64_s64(__p0, __p1) __extension__ ({ \
44290   __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 3); \
44294 #define vcvt_n_f64_s64(__p0, __p1) __extension__ ({ \
44297   __ret = (float64x1_t) __builtin_neon_vcvt_n_f64_v((int8x8_t)__s0, __p1, 3); \
44303 #define vcvtd_n_f64_u64(__p0, __p1) __extension__ ({ \
44306   __ret = (float64_t) __builtin_neon_vcvtd_n_f64_u64(__s0, __p1); \
44310 #define vcvtd_n_f64_u64(__p0, __p1) __extension__ ({ \
44313   __ret = (float64_t) __builtin_neon_vcvtd_n_f64_u64(__s0, __p1); \
44319 #define vcvtd_n_f64_s64(__p0, __p1) __extension__ ({ \
44322   __ret = (float64_t) __builtin_neon_vcvtd_n_f64_s64(__s0, __p1); \
44326 #define vcvtd_n_f64_s64(__p0, __p1) __extension__ ({ \
44329   __ret = (float64_t) __builtin_neon_vcvtd_n_f64_s64(__s0, __p1); \
44335 #define vcvts_n_s32_f32(__p0, __p1) __extension__ ({ \
44338   __ret = (int32_t) __builtin_neon_vcvts_n_s32_f32(__s0, __p1); \
44342 #define vcvts_n_s32_f32(__p0, __p1) __extension__ ({ \
44345   __ret = (int32_t) __builtin_neon_vcvts_n_s32_f32(__s0, __p1); \
44351 #define vcvtq_n_s64_f64(__p0, __p1) __extension__ ({ \
44354   __ret = (int64x2_t) __builtin_neon_vcvtq_n_s64_v((int8x16_t)__s0, __p1, 35); \
44358 #define vcvtq_n_s64_f64(__p0, __p1) __extension__ ({ \
44362   __ret = (int64x2_t) __builtin_neon_vcvtq_n_s64_v((int8x16_t)__rev0, __p1, 35); \
44369 #define vcvt_n_s64_f64(__p0, __p1) __extension__ ({ \
44372   __ret = (int64x1_t) __builtin_neon_vcvt_n_s64_v((int8x8_t)__s0, __p1, 3); \
44376 #define vcvt_n_s64_f64(__p0, __p1) __extension__ ({ \
44379   __ret = (int64x1_t) __builtin_neon_vcvt_n_s64_v((int8x8_t)__s0, __p1, 3); \
44385 #define vcvtd_n_s64_f64(__p0, __p1) __extension__ ({ \
44388   __ret = (int64_t) __builtin_neon_vcvtd_n_s64_f64(__s0, __p1); \
44392 #define vcvtd_n_s64_f64(__p0, __p1) __extension__ ({ \
44395   __ret = (int64_t) __builtin_neon_vcvtd_n_s64_f64(__s0, __p1); \
44401 #define vcvts_n_u32_f32(__p0, __p1) __extension__ ({ \
44404   __ret = (uint32_t) __builtin_neon_vcvts_n_u32_f32(__s0, __p1); \
44408 #define vcvts_n_u32_f32(__p0, __p1) __extension__ ({ \
44411   __ret = (uint32_t) __builtin_neon_vcvts_n_u32_f32(__s0, __p1); \
44417 #define vcvtq_n_u64_f64(__p0, __p1) __extension__ ({ \
44420   __ret = (uint64x2_t) __builtin_neon_vcvtq_n_u64_v((int8x16_t)__s0, __p1, 51); \
44424 #define vcvtq_n_u64_f64(__p0, __p1) __extension__ ({ \
44428   __ret = (uint64x2_t) __builtin_neon_vcvtq_n_u64_v((int8x16_t)__rev0, __p1, 51); \
44435 #define vcvt_n_u64_f64(__p0, __p1) __extension__ ({ \
44438   __ret = (uint64x1_t) __builtin_neon_vcvt_n_u64_v((int8x8_t)__s0, __p1, 19); \
44442 #define vcvt_n_u64_f64(__p0, __p1) __extension__ ({ \
44445   __ret = (uint64x1_t) __builtin_neon_vcvt_n_u64_v((int8x8_t)__s0, __p1, 19); \
44451 #define vcvtd_n_u64_f64(__p0, __p1) __extension__ ({ \
44454   __ret = (uint64_t) __builtin_neon_vcvtd_n_u64_f64(__s0, __p1); \
44458 #define vcvtd_n_u64_f64(__p0, __p1) __extension__ ({ \
44461   __ret = (uint64_t) __builtin_neon_vcvtd_n_u64_f64(__s0, __p1); \
44842 __ai float32x4_t vcvtx_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
44844   __ret = vcombine_f32(__p0, vcvtx_f32_f64(__p1));
44848 __ai float32x4_t vcvtx_high_f32_f64(float32x2_t __p0, float64x2_t __p1) {
44850   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
44859 __ai float64x2_t vdivq_f64(float64x2_t __p0, float64x2_t __p1) {
44861   __ret = __p0 / __p1;
44865 __ai float64x2_t vdivq_f64(float64x2_t __p0, float64x2_t __p1) {
44867   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
44876 __ai float32x4_t vdivq_f32(float32x4_t __p0, float32x4_t __p1) {
44878   __ret = __p0 / __p1;
44882 __ai float32x4_t vdivq_f32(float32x4_t __p0, float32x4_t __p1) {
44884   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
44893 __ai float64x1_t vdiv_f64(float64x1_t __p0, float64x1_t __p1) {
44895   __ret = __p0 / __p1;
44899 __ai float64x1_t vdiv_f64(float64x1_t __p0, float64x1_t __p1) {
44901   __ret = __p0 / __p1;
44907 __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
44909   __ret = __p0 / __p1;
44913 __ai float32x2_t vdiv_f32(float32x2_t __p0, float32x2_t __p1) {
44915   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
44924 #define vdupb_lane_p8(__p0, __p1) __extension__ ({ \
44927   __ret = (poly8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
44931 #define vdupb_lane_p8(__p0, __p1) __extension__ ({ \
44935   __ret = (poly8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
44941 #define vduph_lane_p16(__p0, __p1) __extension__ ({ \
44944   __ret = (poly16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__s0, __p1); \
44948 #define vduph_lane_p16(__p0, __p1) __extension__ ({ \
44952   __ret = (poly16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__rev0, __p1); \
44958 #define vdupb_lane_u8(__p0, __p1) __extension__ ({ \
44961   __ret = (uint8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
44965 #define vdupb_lane_u8(__p0, __p1) __extension__ ({ \
44969   __ret = (uint8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
44975 #define vdups_lane_u32(__p0, __p1) __extension__ ({ \
44978   __ret = (uint32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__s0, __p1); \
44982 #define vdups_lane_u32(__p0, __p1) __extension__ ({ \
44986   __ret = (uint32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__rev0, __p1); \
44992 #define vdupd_lane_u64(__p0, __p1) __extension__ ({ \
44995   __ret = (uint64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
44999 #define vdupd_lane_u64(__p0, __p1) __extension__ ({ \
45002   __ret = (uint64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
45008 #define vduph_lane_u16(__p0, __p1) __extension__ ({ \
45011   __ret = (uint16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__s0, __p1); \
45015 #define vduph_lane_u16(__p0, __p1) __extension__ ({ \
45019   __ret = (uint16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__rev0, __p1); \
45025 #define vdupb_lane_s8(__p0, __p1) __extension__ ({ \
45028   __ret = (int8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__s0, __p1); \
45032 #define vdupb_lane_s8(__p0, __p1) __extension__ ({ \
45036   __ret = (int8_t) __builtin_neon_vdupb_lane_i8((int8x8_t)__rev0, __p1); \
45042 #define vdupd_lane_f64(__p0, __p1) __extension__ ({ \
45045   __ret = (float64_t) __builtin_neon_vdupd_lane_f64((int8x8_t)__s0, __p1); \
45049 #define vdupd_lane_f64(__p0, __p1) __extension__ ({ \
45052   __ret = (float64_t) __builtin_neon_vdupd_lane_f64((int8x8_t)__s0, __p1); \
45058 #define vdups_lane_f32(__p0, __p1) __extension__ ({ \
45061   __ret = (float32_t) __builtin_neon_vdups_lane_f32((int8x8_t)__s0, __p1); \
45065 #define vdups_lane_f32(__p0, __p1) __extension__ ({ \
45069   __ret = (float32_t) __builtin_neon_vdups_lane_f32((int8x8_t)__rev0, __p1); \
45075 #define vdups_lane_s32(__p0, __p1) __extension__ ({ \
45078   __ret = (int32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__s0, __p1); \
45082 #define vdups_lane_s32(__p0, __p1) __extension__ ({ \
45086   __ret = (int32_t) __builtin_neon_vdups_lane_i32((int8x8_t)__rev0, __p1); \
45092 #define vdupd_lane_s64(__p0, __p1) __extension__ ({ \
45095   __ret = (int64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
45099 #define vdupd_lane_s64(__p0, __p1) __extension__ ({ \
45102   __ret = (int64_t) __builtin_neon_vdupd_lane_i64((int8x8_t)__s0, __p1); \
45108 #define vduph_lane_s16(__p0, __p1) __extension__ ({ \
45111   __ret = (int16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__s0, __p1); \
45115 #define vduph_lane_s16(__p0, __p1) __extension__ ({ \
45119   __ret = (int16_t) __builtin_neon_vduph_lane_i16((int8x8_t)__rev0, __p1); \
45125 #define vdup_lane_p64(__p0, __p1) __extension__ ({ \
45128   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
45132 #define vdup_lane_p64(__p0, __p1) __extension__ ({ \
45135   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
45141 #define vdupq_lane_p64(__p0, __p1) __extension__ ({ \
45144   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
45148 #define vdupq_lane_p64(__p0, __p1) __extension__ ({ \
45151   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
45158 #define vdupq_lane_f64(__p0, __p1) __extension__ ({ \
45161   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
45165 #define vdupq_lane_f64(__p0, __p1) __extension__ ({ \
45168   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
45175 #define vdupq_lane_f16(__p0, __p1) __extension__ ({ \
45178   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45182 #define vdupq_lane_f16(__p0, __p1) __extension__ ({ \
45186   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45193 #define vdup_lane_f64(__p0, __p1) __extension__ ({ \
45196   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
45200 #define vdup_lane_f64(__p0, __p1) __extension__ ({ \
45203   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
45209 #define vdup_lane_f16(__p0, __p1) __extension__ ({ \
45212   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
45216 #define vdup_lane_f16(__p0, __p1) __extension__ ({ \
45220   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
45227 #define vdupb_laneq_p8(__p0, __p1) __extension__ ({ \
45230   __ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
45234 #define vdupb_laneq_p8(__p0, __p1) __extension__ ({ \
45238   __ret = (poly8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
45244 #define vduph_laneq_p16(__p0, __p1) __extension__ ({ \
45247   __ret = (poly16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__s0, __p1); \
45251 #define vduph_laneq_p16(__p0, __p1) __extension__ ({ \
45255   __ret = (poly16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__rev0, __p1); \
45261 #define vdupb_laneq_u8(__p0, __p1) __extension__ ({ \
45264   __ret = (uint8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
45268 #define vdupb_laneq_u8(__p0, __p1) __extension__ ({ \
45272   __ret = (uint8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
45278 #define vdups_laneq_u32(__p0, __p1) __extension__ ({ \
45281   __ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__s0, __p1); \
45285 #define vdups_laneq_u32(__p0, __p1) __extension__ ({ \
45289   __ret = (uint32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__rev0, __p1); \
45295 #define vdupd_laneq_u64(__p0, __p1) __extension__ ({ \
45298   __ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__s0, __p1); \
45302 #define vdupd_laneq_u64(__p0, __p1) __extension__ ({ \
45306   __ret = (uint64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__rev0, __p1); \
45312 #define vduph_laneq_u16(__p0, __p1) __extension__ ({ \
45315   __ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__s0, __p1); \
45319 #define vduph_laneq_u16(__p0, __p1) __extension__ ({ \
45323   __ret = (uint16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__rev0, __p1); \
45329 #define vdupb_laneq_s8(__p0, __p1) __extension__ ({ \
45332   __ret = (int8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__s0, __p1); \
45336 #define vdupb_laneq_s8(__p0, __p1) __extension__ ({ \
45340   __ret = (int8_t) __builtin_neon_vdupb_laneq_i8((int8x16_t)__rev0, __p1); \
45346 #define vdupd_laneq_f64(__p0, __p1) __extension__ ({ \
45349   __ret = (float64_t) __builtin_neon_vdupd_laneq_f64((int8x16_t)__s0, __p1); \
45353 #define vdupd_laneq_f64(__p0, __p1) __extension__ ({ \
45357   __ret = (float64_t) __builtin_neon_vdupd_laneq_f64((int8x16_t)__rev0, __p1); \
45363 #define vdups_laneq_f32(__p0, __p1) __extension__ ({ \
45366   __ret = (float32_t) __builtin_neon_vdups_laneq_f32((int8x16_t)__s0, __p1); \
45370 #define vdups_laneq_f32(__p0, __p1) __extension__ ({ \
45374   __ret = (float32_t) __builtin_neon_vdups_laneq_f32((int8x16_t)__rev0, __p1); \
45380 #define vdups_laneq_s32(__p0, __p1) __extension__ ({ \
45383   __ret = (int32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__s0, __p1); \
45387 #define vdups_laneq_s32(__p0, __p1) __extension__ ({ \
45391   __ret = (int32_t) __builtin_neon_vdups_laneq_i32((int8x16_t)__rev0, __p1); \
45397 #define vdupd_laneq_s64(__p0, __p1) __extension__ ({ \
45400   __ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__s0, __p1); \
45404 #define vdupd_laneq_s64(__p0, __p1) __extension__ ({ \
45408   __ret = (int64_t) __builtin_neon_vdupd_laneq_i64((int8x16_t)__rev0, __p1); \
45414 #define vduph_laneq_s16(__p0, __p1) __extension__ ({ \
45417   __ret = (int16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__s0, __p1); \
45421 #define vduph_laneq_s16(__p0, __p1) __extension__ ({ \
45425   __ret = (int16_t) __builtin_neon_vduph_laneq_i16((int8x16_t)__rev0, __p1); \
45431 #define vdup_laneq_p8(__p0, __p1) __extension__ ({ \
45434   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45438 #define vdup_laneq_p8(__p0, __p1) __extension__ ({ \
45442   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45449 #define vdup_laneq_p64(__p0, __p1) __extension__ ({ \
45452   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
45456 #define vdup_laneq_p64(__p0, __p1) __extension__ ({ \
45460   __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
45466 #define vdup_laneq_p16(__p0, __p1) __extension__ ({ \
45469   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
45473 #define vdup_laneq_p16(__p0, __p1) __extension__ ({ \
45477   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
45484 #define vdupq_laneq_p8(__p0, __p1) __extension__ ({ \
45487   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45491 #define vdupq_laneq_p8(__p0, __p1) __extension__ ({ \
45495   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45502 #define vdupq_laneq_p64(__p0, __p1) __extension__ ({ \
45505   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
45509 #define vdupq_laneq_p64(__p0, __p1) __extension__ ({ \
45513   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
45520 #define vdupq_laneq_p16(__p0, __p1) __extension__ ({ \
45523   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45527 #define vdupq_laneq_p16(__p0, __p1) __extension__ ({ \
45531   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45538 #define vdupq_laneq_u8(__p0, __p1) __extension__ ({ \
45541   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45545 #define vdupq_laneq_u8(__p0, __p1) __extension__ ({ \
45549   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45556 #define vdupq_laneq_u32(__p0, __p1) __extension__ ({ \
45559   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
45563 #define vdupq_laneq_u32(__p0, __p1) __extension__ ({ \
45567   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
45574 #define vdupq_laneq_u64(__p0, __p1) __extension__ ({ \
45577   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
45581 #define vdupq_laneq_u64(__p0, __p1) __extension__ ({ \
45585   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
45592 #define vdupq_laneq_u16(__p0, __p1) __extension__ ({ \
45595   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45599 #define vdupq_laneq_u16(__p0, __p1) __extension__ ({ \
45603   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45610 #define vdupq_laneq_s8(__p0, __p1) __extension__ ({ \
45613   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45617 #define vdupq_laneq_s8(__p0, __p1) __extension__ ({ \
45621   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45628 #define vdupq_laneq_f64(__p0, __p1) __extension__ ({ \
45631   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
45635 #define vdupq_laneq_f64(__p0, __p1) __extension__ ({ \
45639   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
45646 #define vdupq_laneq_f32(__p0, __p1) __extension__ ({ \
45649   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
45653 #define vdupq_laneq_f32(__p0, __p1) __extension__ ({ \
45657   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
45664 #define vdupq_laneq_f16(__p0, __p1) __extension__ ({ \
45667   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45671 #define vdupq_laneq_f16(__p0, __p1) __extension__ ({ \
45675   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45682 #define vdupq_laneq_s32(__p0, __p1) __extension__ ({ \
45685   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
45689 #define vdupq_laneq_s32(__p0, __p1) __extension__ ({ \
45693   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
45700 #define vdupq_laneq_s64(__p0, __p1) __extension__ ({ \
45703   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
45707 #define vdupq_laneq_s64(__p0, __p1) __extension__ ({ \
45711   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
45718 #define vdupq_laneq_s16(__p0, __p1) __extension__ ({ \
45721   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45725 #define vdupq_laneq_s16(__p0, __p1) __extension__ ({ \
45729   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45736 #define vdup_laneq_u8(__p0, __p1) __extension__ ({ \
45739   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45743 #define vdup_laneq_u8(__p0, __p1) __extension__ ({ \
45747   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45754 #define vdup_laneq_u32(__p0, __p1) __extension__ ({ \
45757   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
45761 #define vdup_laneq_u32(__p0, __p1) __extension__ ({ \
45765   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
45772 #define vdup_laneq_u64(__p0, __p1) __extension__ ({ \
45775   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
45779 #define vdup_laneq_u64(__p0, __p1) __extension__ ({ \
45783   __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
45789 #define vdup_laneq_u16(__p0, __p1) __extension__ ({ \
45792   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
45796 #define vdup_laneq_u16(__p0, __p1) __extension__ ({ \
45800   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
45807 #define vdup_laneq_s8(__p0, __p1) __extension__ ({ \
45810   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45814 #define vdup_laneq_s8(__p0, __p1) __extension__ ({ \
45818   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1, __p1, __p1, __p1, __p1); \
45825 #define vdup_laneq_f64(__p0, __p1) __extension__ ({ \
45828   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
45832 #define vdup_laneq_f64(__p0, __p1) __extension__ ({ \
45836   __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
45842 #define vdup_laneq_f32(__p0, __p1) __extension__ ({ \
45845   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
45849 #define vdup_laneq_f32(__p0, __p1) __extension__ ({ \
45853   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
45860 #define vdup_laneq_f16(__p0, __p1) __extension__ ({ \
45863   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
45867 #define vdup_laneq_f16(__p0, __p1) __extension__ ({ \
45871   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
45878 #define vdup_laneq_s32(__p0, __p1) __extension__ ({ \
45881   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1); \
45885 #define vdup_laneq_s32(__p0, __p1) __extension__ ({ \
45889   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1); \
45896 #define vdup_laneq_s64(__p0, __p1) __extension__ ({ \
45899   __ret = __builtin_shufflevector(__s0, __s0, __p1); \
45903 #define vdup_laneq_s64(__p0, __p1) __extension__ ({ \
45907   __ret = __builtin_shufflevector(__rev0, __rev0, __p1); \
45913 #define vdup_laneq_s16(__p0, __p1) __extension__ ({ \
45916   __ret = __builtin_shufflevector(__s0, __s0, __p1, __p1, __p1, __p1); \
45920 #define vdup_laneq_s16(__p0, __p1) __extension__ ({ \
45924   __ret = __builtin_shufflevector(__rev0, __rev0, __p1, __p1, __p1, __p1); \
45989 #define vext_p64(__p0, __p1, __p2) __extension__ ({ \
45991   poly64x1_t __s1 = __p1; \
45997 #define vext_p64(__p0, __p1, __p2) __extension__ ({ \
45999   poly64x1_t __s1 = __p1; \
46007 #define vextq_p64(__p0, __p1, __p2) __extension__ ({ \
46009   poly64x2_t __s1 = __p1; \
46015 #define vextq_p64(__p0, __p1, __p2) __extension__ ({ \
46017   poly64x2_t __s1 = __p1; \
46028 #define vextq_f64(__p0, __p1, __p2) __extension__ ({ \
46030   float64x2_t __s1 = __p1; \
46036 #define vextq_f64(__p0, __p1, __p2) __extension__ ({ \
46038   float64x2_t __s1 = __p1; \
46049 #define vext_f64(__p0, __p1, __p2) __extension__ ({ \
46051   float64x1_t __s1 = __p1; \
46057 #define vext_f64(__p0, __p1, __p2) __extension__ ({ \
46059   float64x1_t __s1 = __p1; \
46067 __ai float64x2_t vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
46069   __ret = (float64x2_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
46073 __ai float64x2_t vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
46075   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
46082 __ai float64x2_t __noswap_vfmaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
46084   __ret = (float64x2_t) __builtin_neon_vfmaq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
46090 __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
46092   __ret = (float64x1_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
46096 __ai float64x1_t vfma_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
46098   __ret = (float64x1_t) __builtin_neon_vfma_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
46104 #define vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46106   float64_t __s1 = __p1; \
46113 #define vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46115   float64_t __s1 = __p1; \
46121 #define __noswap_vfmad_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46123   float64_t __s1 = __p1; \
46132 #define vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46134   float32_t __s1 = __p1; \
46141 #define vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46143   float32_t __s1 = __p1; \
46150 #define __noswap_vfmas_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46152   float32_t __s1 = __p1; \
46161 #define vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46163   float64x2_t __s1 = __p1; \
46170 #define vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46172   float64x2_t __s1 = __p1; \
46181 #define __noswap_vfmaq_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46183   float64x2_t __s1 = __p1; \
46192 #define vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46194   float32x4_t __s1 = __p1; \
46201 #define vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46203   float32x4_t __s1 = __p1; \
46213 #define __noswap_vfmaq_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46215   float32x4_t __s1 = __p1; \
46224 #define vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46226   float64x1_t __s1 = __p1; \
46233 #define vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46235   float64x1_t __s1 = __p1; \
46241 #define __noswap_vfma_lane_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46243   float64x1_t __s1 = __p1; \
46252 #define vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46254   float32x2_t __s1 = __p1; \
46261 #define vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46263   float32x2_t __s1 = __p1; \
46273 #define __noswap_vfma_lane_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46275   float32x2_t __s1 = __p1; \
46284 #define vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46286   float64_t __s1 = __p1; \
46293 #define vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46295   float64_t __s1 = __p1; \
46302 #define __noswap_vfmad_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46304   float64_t __s1 = __p1; \
46313 #define vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46315   float32_t __s1 = __p1; \
46322 #define vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46324   float32_t __s1 = __p1; \
46331 #define __noswap_vfmas_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46333   float32_t __s1 = __p1; \
46342 #define vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46344   float64x2_t __s1 = __p1; \
46351 #define vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46353   float64x2_t __s1 = __p1; \
46363 #define __noswap_vfmaq_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46365   float64x2_t __s1 = __p1; \
46374 #define vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46376   float32x4_t __s1 = __p1; \
46383 #define vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46385   float32x4_t __s1 = __p1; \
46395 #define __noswap_vfmaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46397   float32x4_t __s1 = __p1; \
46406 #define vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46408   float64x1_t __s1 = __p1; \
46415 #define vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46417   float64x1_t __s1 = __p1; \
46424 #define __noswap_vfma_laneq_f64(__p0, __p1, __p2, __p3) __extension__ ({ \
46426   float64x1_t __s1 = __p1; \
46435 #define vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46437   float32x2_t __s1 = __p1; \
46444 #define vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46446   float32x2_t __s1 = __p1; \
46456 #define __noswap_vfma_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
46458   float32x2_t __s1 = __p1; \
46467 __ai float64x2_t vfmaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
46469   __ret = vfmaq_f64(__p0, __p1, (float64x2_t) {__p2, __p2});
46473 __ai float64x2_t vfmaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
46475   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
46484 __ai float32x4_t vfmaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
46486   __ret = vfmaq_f32(__p0, __p1, (float32x4_t) {__p2, __p2, __p2, __p2});
46490 __ai float32x4_t vfmaq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
46492   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46501 __ai float32x2_t vfma_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
46503   __ret = vfma_f32(__p0, __p1, (float32x2_t) {__p2, __p2});
46507 __ai float32x2_t vfma_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
46509   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
46518 __ai float64x2_t vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
46520   __ret = (float64x2_t) __builtin_neon_vfmsq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
46524 __ai float64x2_t vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
46526   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
46533 __ai float64x2_t __noswap_vfmsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
46535   __ret = (float64x2_t) __builtin_neon_vfmsq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 42);
46541 __ai float32x4_t vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
46543   __ret = (float32x4_t) __builtin_neon_vfmsq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
46547 __ai float32x4_t vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
46549   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46556 __ai float32x4_t __noswap_vfmsq_f32(float32x4_t __p0, float32x4_t __p1, float32x4_t __p2) {
46558   __ret = (float32x4_t) __builtin_neon_vfmsq_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 41);
46564 __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
46566   __ret = (float64x1_t) __builtin_neon_vfms_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
46570 __ai float64x1_t vfms_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
46572   __ret = (float64x1_t) __builtin_neon_vfms_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 10);
46578 __ai float32x2_t vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
46580   __ret = (float32x2_t) __builtin_neon_vfms_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
46584 __ai float32x2_t vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
46586   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
46593 __ai float32x2_t __noswap_vfms_f32(float32x2_t __p0, float32x2_t __p1, float32x2_t __p2) {
46595   __ret = (float32x2_t) __builtin_neon_vfms_v((int8x8_t)__p0, (int8x8_t)__p1, (int8x8_t)__p2, 9);
46868 __ai float64x2_t vfmsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
46870   __ret = vfmsq_f64(__p0, __p1, (float64x2_t) {__p2, __p2});
46874 __ai float64x2_t vfmsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
46876   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
46885 __ai float32x4_t vfmsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
46887   __ret = vfmsq_f32(__p0, __p1, (float32x4_t) {__p2, __p2, __p2, __p2});
46891 __ai float32x4_t vfmsq_n_f32(float32x4_t __p0, float32x4_t __p1, float32_t __p2) {
46893   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
46902 __ai float32x2_t vfms_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
46904   __ret = vfms_f32(__p0, __p1, (float32x2_t) {__p2, __p2});
46908 __ai float32x2_t vfms_n_f32(float32x2_t __p0, float32x2_t __p1, float32_t __p2) {
46910   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
46954 #define vget_lane_p64(__p0, __p1) __extension__ ({ \
46957   __ret = (poly64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
46961 #define vget_lane_p64(__p0, __p1) __extension__ ({ \
46964   __ret = (poly64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
46967 #define __noswap_vget_lane_p64(__p0, __p1) __extension__ ({ \
46970   __ret = (poly64_t) __builtin_neon_vget_lane_i64((int8x8_t)__s0, __p1); \
46976 #define vgetq_lane_p64(__p0, __p1) __extension__ ({ \
46979   __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
46983 #define vgetq_lane_p64(__p0, __p1) __extension__ ({ \
46987   __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__rev0, __p1); \
46990 #define __noswap_vgetq_lane_p64(__p0, __p1) __extension__ ({ \
46993   __ret = (poly64_t) __builtin_neon_vgetq_lane_i64((int8x16_t)__s0, __p1); \
46999 #define vgetq_lane_f64(__p0, __p1) __extension__ ({ \
47002   __ret = (float64_t) __builtin_neon_vgetq_lane_f64((int8x16_t)__s0, __p1); \
47006 #define vgetq_lane_f64(__p0, __p1) __extension__ ({ \
47010   __ret = (float64_t) __builtin_neon_vgetq_lane_f64((int8x16_t)__rev0, __p1); \
47013 #define __noswap_vgetq_lane_f64(__p0, __p1) __extension__ ({ \
47016   __ret = (float64_t) __builtin_neon_vgetq_lane_f64((int8x16_t)__s0, __p1); \
47022 #define vget_lane_f64(__p0, __p1) __extension__ ({ \
47025   __ret = (float64_t) __builtin_neon_vget_lane_f64((int8x8_t)__s0, __p1); \
47029 #define vget_lane_f64(__p0, __p1) __extension__ ({ \
47032   __ret = (float64_t) __builtin_neon_vget_lane_f64((int8x8_t)__s0, __p1); \
47035 #define __noswap_vget_lane_f64(__p0, __p1) __extension__ ({ \
47038   __ret = (float64_t) __builtin_neon_vget_lane_f64((int8x8_t)__s0, __p1); \
47232 #define vld1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
47233   poly64x1_t __s1 = __p1; \
47239 #define vld1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
47240   poly64x1_t __s1 = __p1; \
47248 #define vld1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
47249   poly64x2_t __s1 = __p1; \
47255 #define vld1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
47256   poly64x2_t __s1 = __p1; \
47266 #define vld1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
47267   float64x2_t __s1 = __p1; \
47273 #define vld1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
47274   float64x2_t __s1 = __p1; \
47284 #define vld1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
47285   float64x1_t __s1 = __p1; \
47291 #define vld1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
47292   float64x1_t __s1 = __p1; \
49126 #define vld2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
49127   poly64x1x2_t __s1 = __p1; \
49133 #define vld2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
49134   poly64x1x2_t __s1 = __p1; \
49142 #define vld2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
49143   poly8x16x2_t __s1 = __p1; \
49149 #define vld2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
49150   poly8x16x2_t __s1 = __p1; \
49164 #define vld2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
49165   poly64x2x2_t __s1 = __p1; \
49171 #define vld2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
49172   poly64x2x2_t __s1 = __p1; \
49186 #define vld2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
49187   uint8x16x2_t __s1 = __p1; \
49193 #define vld2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
49194   uint8x16x2_t __s1 = __p1; \
49208 #define vld2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
49209   uint64x2x2_t __s1 = __p1; \
49215 #define vld2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
49216   uint64x2x2_t __s1 = __p1; \
49230 #define vld2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
49231   int8x16x2_t __s1 = __p1; \
49237 #define vld2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
49238   int8x16x2_t __s1 = __p1; \
49252 #define vld2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
49253   float64x2x2_t __s1 = __p1; \
49259 #define vld2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
49260   float64x2x2_t __s1 = __p1; \
49274 #define vld2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
49275   int64x2x2_t __s1 = __p1; \
49281 #define vld2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
49282   int64x2x2_t __s1 = __p1; \
49296 #define vld2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
49297   uint64x1x2_t __s1 = __p1; \
49303 #define vld2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
49304   uint64x1x2_t __s1 = __p1; \
49312 #define vld2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
49313   float64x1x2_t __s1 = __p1; \
49319 #define vld2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
49320   float64x1x2_t __s1 = __p1; \
49328 #define vld2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
49329   int64x1x2_t __s1 = __p1; \
49335 #define vld2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
49336   int64x1x2_t __s1 = __p1; \
49724 #define vld3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
49725   poly64x1x3_t __s1 = __p1; \
49731 #define vld3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
49732   poly64x1x3_t __s1 = __p1; \
49740 #define vld3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
49741   poly8x16x3_t __s1 = __p1; \
49747 #define vld3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
49748   poly8x16x3_t __s1 = __p1; \
49764 #define vld3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
49765   poly64x2x3_t __s1 = __p1; \
49771 #define vld3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
49772   poly64x2x3_t __s1 = __p1; \
49788 #define vld3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
49789   uint8x16x3_t __s1 = __p1; \
49795 #define vld3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
49796   uint8x16x3_t __s1 = __p1; \
49812 #define vld3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
49813   uint64x2x3_t __s1 = __p1; \
49819 #define vld3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
49820   uint64x2x3_t __s1 = __p1; \
49836 #define vld3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
49837   int8x16x3_t __s1 = __p1; \
49843 #define vld3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
49844   int8x16x3_t __s1 = __p1; \
49860 #define vld3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
49861   float64x2x3_t __s1 = __p1; \
49867 #define vld3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
49868   float64x2x3_t __s1 = __p1; \
49884 #define vld3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
49885   int64x2x3_t __s1 = __p1; \
49891 #define vld3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
49892   int64x2x3_t __s1 = __p1; \
49908 #define vld3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
49909   uint64x1x3_t __s1 = __p1; \
49915 #define vld3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
49916   uint64x1x3_t __s1 = __p1; \
49924 #define vld3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
49925   float64x1x3_t __s1 = __p1; \
49931 #define vld3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
49932   float64x1x3_t __s1 = __p1; \
49940 #define vld3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
49941   int64x1x3_t __s1 = __p1; \
49947 #define vld3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
49948   int64x1x3_t __s1 = __p1; \
50354 #define vld4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50355   poly64x1x4_t __s1 = __p1; \
50361 #define vld4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50362   poly64x1x4_t __s1 = __p1; \
50370 #define vld4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
50371   poly8x16x4_t __s1 = __p1; \
50377 #define vld4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
50378   poly8x16x4_t __s1 = __p1; \
50396 #define vld4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50397   poly64x2x4_t __s1 = __p1; \
50403 #define vld4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
50404   poly64x2x4_t __s1 = __p1; \
50422 #define vld4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
50423   uint8x16x4_t __s1 = __p1; \
50429 #define vld4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
50430   uint8x16x4_t __s1 = __p1; \
50448 #define vld4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
50449   uint64x2x4_t __s1 = __p1; \
50455 #define vld4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
50456   uint64x2x4_t __s1 = __p1; \
50474 #define vld4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
50475   int8x16x4_t __s1 = __p1; \
50481 #define vld4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
50482   int8x16x4_t __s1 = __p1; \
50500 #define vld4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
50501   float64x2x4_t __s1 = __p1; \
50507 #define vld4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
50508   float64x2x4_t __s1 = __p1; \
50526 #define vld4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
50527   int64x2x4_t __s1 = __p1; \
50533 #define vld4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
50534   int64x2x4_t __s1 = __p1; \
50552 #define vld4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
50553   uint64x1x4_t __s1 = __p1; \
50559 #define vld4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
50560   uint64x1x4_t __s1 = __p1; \
50568 #define vld4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
50569   float64x1x4_t __s1 = __p1; \
50575 #define vld4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
50576   float64x1x4_t __s1 = __p1; \
50584 #define vld4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
50585   int64x1x4_t __s1 = __p1; \
50591 #define vld4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
50592   int64x1x4_t __s1 = __p1; \
50614 __ai float64x2_t vmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
50616   __ret = (float64x2_t) __builtin_neon_vmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
50620 __ai float64x2_t vmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
50622   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
50631 __ai float64x1_t vmax_f64(float64x1_t __p0, float64x1_t __p1) {
50633   __ret = (float64x1_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
50637 __ai float64x1_t vmax_f64(float64x1_t __p0, float64x1_t __p1) {
50639   __ret = (float64x1_t) __builtin_neon_vmax_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
50645 __ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
50647   __ret = (float64x2_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
50651 __ai float64x2_t vmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
50653   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
50662 __ai float32x4_t vmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
50664   __ret = (float32x4_t) __builtin_neon_vmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
50668 __ai float32x4_t vmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
50670   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
50679 __ai float64x1_t vmaxnm_f64(float64x1_t __p0, float64x1_t __p1) {
50681   __ret = (float64x1_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
50685 __ai float64x1_t vmaxnm_f64(float64x1_t __p0, float64x1_t __p1) {
50687   __ret = (float64x1_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
50693 __ai float32x2_t vmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
50695   __ret = (float32x2_t) __builtin_neon_vmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
50699 __ai float32x2_t vmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
50701   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
50980 __ai float64x2_t vminq_f64(float64x2_t __p0, float64x2_t __p1) {
50982   __ret = (float64x2_t) __builtin_neon_vminq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
50986 __ai float64x2_t vminq_f64(float64x2_t __p0, float64x2_t __p1) {
50988   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
50997 __ai float64x1_t vmin_f64(float64x1_t __p0, float64x1_t __p1) {
50999   __ret = (float64x1_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
51003 __ai float64x1_t vmin_f64(float64x1_t __p0, float64x1_t __p1) {
51005   __ret = (float64x1_t) __builtin_neon_vmin_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
51011 __ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
51013   __ret = (float64x2_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
51017 __ai float64x2_t vminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
51019   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51028 __ai float32x4_t vminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
51030   __ret = (float32x4_t) __builtin_neon_vminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
51034 __ai float32x4_t vminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
51036   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
51045 __ai float64x1_t vminnm_f64(float64x1_t __p0, float64x1_t __p1) {
51047   __ret = (float64x1_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
51051 __ai float64x1_t vminnm_f64(float64x1_t __p0, float64x1_t __p1) {
51053   __ret = (float64x1_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
51059 __ai float32x2_t vminnm_f32(float32x2_t __p0, float32x2_t __p1) {
51061   __ret = (float32x2_t) __builtin_neon_vminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
51065 __ai float32x2_t vminnm_f32(float32x2_t __p0, float32x2_t __p1) {
51067   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51346 __ai float64x2_t vmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
51348   __ret = __p0 + __p1 * __p2;
51352 __ai float64x2_t vmlaq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
51354   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51364 __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
51366   __ret = __p0 + __p1 * __p2;
51370 __ai float64x1_t vmla_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
51372   __ret = __p0 + __p1 * __p2;
51378 #define vmlaq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51380   uint32x4_t __s1 = __p1; \
51387 #define vmlaq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51389   uint32x4_t __s1 = __p1; \
51402 #define vmlaq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51404   uint16x8_t __s1 = __p1; \
51411 #define vmlaq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51413   uint16x8_t __s1 = __p1; \
51426 #define vmlaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
51428   float32x4_t __s1 = __p1; \
51435 #define vmlaq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
51437   float32x4_t __s1 = __p1; \
51450 #define vmlaq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
51452   int32x4_t __s1 = __p1; \
51459 #define vmlaq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
51461   int32x4_t __s1 = __p1; \
51474 #define vmlaq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
51476   int16x8_t __s1 = __p1; \
51483 #define vmlaq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
51485   int16x8_t __s1 = __p1; \
51498 #define vmla_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51500   uint32x2_t __s1 = __p1; \
51507 #define vmla_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51509   uint32x2_t __s1 = __p1; \
51522 #define vmla_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51524   uint16x4_t __s1 = __p1; \
51531 #define vmla_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51533   uint16x4_t __s1 = __p1; \
51546 #define vmla_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
51548   float32x2_t __s1 = __p1; \
51555 #define vmla_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
51557   float32x2_t __s1 = __p1; \
51570 #define vmla_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
51572   int32x2_t __s1 = __p1; \
51579 #define vmla_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
51581   int32x2_t __s1 = __p1; \
51594 #define vmla_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
51596   int16x4_t __s1 = __p1; \
51603 #define vmla_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
51605   int16x4_t __s1 = __p1; \
51618 __ai float64x2_t vmlaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
51620   __ret = __p0 + __p1 * (float64x2_t) {__p2, __p2};
51624 __ai float64x2_t vmlaq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
51626   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51635 #define vmlal_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51637   uint32x4_t __s1 = __p1; \
51644 #define vmlal_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51646   uint32x4_t __s1 = __p1; \
51659 #define vmlal_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51661   uint16x8_t __s1 = __p1; \
51668 #define vmlal_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51670   uint16x8_t __s1 = __p1; \
51683 #define vmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
51685   int32x4_t __s1 = __p1; \
51692 #define vmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
51694   int32x4_t __s1 = __p1; \
51707 #define vmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
51709   int16x8_t __s1 = __p1; \
51716 #define vmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
51718   int16x8_t __s1 = __p1; \
51731 #define vmlal_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51733   uint32x4_t __s1 = __p1; \
51740 #define vmlal_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51742   uint32x4_t __s1 = __p1; \
51755 #define vmlal_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51757   uint16x8_t __s1 = __p1; \
51764 #define vmlal_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51766   uint16x8_t __s1 = __p1; \
51779 #define vmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
51781   int32x4_t __s1 = __p1; \
51788 #define vmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
51790   int32x4_t __s1 = __p1; \
51803 #define vmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
51805   int16x8_t __s1 = __p1; \
51812 #define vmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
51814   int16x8_t __s1 = __p1; \
51827 #define vmlal_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51829   uint32x2_t __s1 = __p1; \
51836 #define vmlal_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51838   uint32x2_t __s1 = __p1; \
51851 #define vmlal_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51853   uint16x4_t __s1 = __p1; \
51860 #define vmlal_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51862   uint16x4_t __s1 = __p1; \
51875 #define vmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
51877   int32x2_t __s1 = __p1; \
51884 #define vmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
51886   int32x2_t __s1 = __p1; \
51899 #define vmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
51901   int16x4_t __s1 = __p1; \
51908 #define vmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
51910   int16x4_t __s1 = __p1; \
51923 __ai float64x2_t vmlsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
51925   __ret = __p0 - __p1 * __p2;
51929 __ai float64x2_t vmlsq_f64(float64x2_t __p0, float64x2_t __p1, float64x2_t __p2) {
51931   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
51941 __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
51943   __ret = __p0 - __p1 * __p2;
51947 __ai float64x1_t vmls_f64(float64x1_t __p0, float64x1_t __p1, float64x1_t __p2) {
51949   __ret = __p0 - __p1 * __p2;
51955 #define vmlsq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51957   uint32x4_t __s1 = __p1; \
51964 #define vmlsq_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
51966   uint32x4_t __s1 = __p1; \
51979 #define vmlsq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51981   uint16x8_t __s1 = __p1; \
51988 #define vmlsq_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
51990   uint16x8_t __s1 = __p1; \
52003 #define vmlsq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52005   float32x4_t __s1 = __p1; \
52012 #define vmlsq_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52014   float32x4_t __s1 = __p1; \
52027 #define vmlsq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52029   int32x4_t __s1 = __p1; \
52036 #define vmlsq_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52038   int32x4_t __s1 = __p1; \
52051 #define vmlsq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52053   int16x8_t __s1 = __p1; \
52060 #define vmlsq_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52062   int16x8_t __s1 = __p1; \
52075 #define vmls_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52077   uint32x2_t __s1 = __p1; \
52084 #define vmls_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52086   uint32x2_t __s1 = __p1; \
52099 #define vmls_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52101   uint16x4_t __s1 = __p1; \
52108 #define vmls_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52110   uint16x4_t __s1 = __p1; \
52123 #define vmls_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52125   float32x2_t __s1 = __p1; \
52132 #define vmls_laneq_f32(__p0, __p1, __p2, __p3) __extension__ ({ \
52134   float32x2_t __s1 = __p1; \
52147 #define vmls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52149   int32x2_t __s1 = __p1; \
52156 #define vmls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52158   int32x2_t __s1 = __p1; \
52171 #define vmls_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52173   int16x4_t __s1 = __p1; \
52180 #define vmls_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52182   int16x4_t __s1 = __p1; \
52195 __ai float64x2_t vmlsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
52197   __ret = __p0 - __p1 * (float64x2_t) {__p2, __p2};
52201 __ai float64x2_t vmlsq_n_f64(float64x2_t __p0, float64x2_t __p1, float64_t __p2) {
52203   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
52212 #define vmlsl_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52214   uint32x4_t __s1 = __p1; \
52221 #define vmlsl_high_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52223   uint32x4_t __s1 = __p1; \
52236 #define vmlsl_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52238   uint16x8_t __s1 = __p1; \
52245 #define vmlsl_high_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52247   uint16x8_t __s1 = __p1; \
52260 #define vmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52262   int32x4_t __s1 = __p1; \
52269 #define vmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52271   int32x4_t __s1 = __p1; \
52284 #define vmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52286   int16x8_t __s1 = __p1; \
52293 #define vmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52295   int16x8_t __s1 = __p1; \
52308 #define vmlsl_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52310   uint32x4_t __s1 = __p1; \
52317 #define vmlsl_high_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52319   uint32x4_t __s1 = __p1; \
52332 #define vmlsl_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52334   uint16x8_t __s1 = __p1; \
52341 #define vmlsl_high_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52343   uint16x8_t __s1 = __p1; \
52356 #define vmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52358   int32x4_t __s1 = __p1; \
52365 #define vmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52367   int32x4_t __s1 = __p1; \
52380 #define vmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52382   int16x8_t __s1 = __p1; \
52389 #define vmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52391   int16x8_t __s1 = __p1; \
52404 #define vmlsl_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52406   uint32x2_t __s1 = __p1; \
52413 #define vmlsl_laneq_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
52415   uint32x2_t __s1 = __p1; \
52428 #define vmlsl_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52430   uint16x4_t __s1 = __p1; \
52437 #define vmlsl_laneq_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
52439   uint16x4_t __s1 = __p1; \
52452 #define vmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52454   int32x2_t __s1 = __p1; \
52461 #define vmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
52463   int32x2_t __s1 = __p1; \
52476 #define vmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52478   int16x4_t __s1 = __p1; \
52485 #define vmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
52487   int16x4_t __s1 = __p1; \
52673 __ai uint16x8_t vmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
52675   __ret = vcombine_u16(__p0, vmovn_u32(__p1));
52679 __ai uint16x8_t vmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
52681   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
52690 __ai uint32x4_t vmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
52692   __ret = vcombine_u32(__p0, vmovn_u64(__p1));
52696 __ai uint32x4_t vmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
52698   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
52707 __ai uint8x16_t vmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
52709   __ret = vcombine_u8(__p0, vmovn_u16(__p1));
52713 __ai uint8x16_t vmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
52715   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
52724 __ai int16x8_t vmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
52726   __ret = vcombine_s16(__p0, vmovn_s32(__p1));
52730 __ai int16x8_t vmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
52732   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
52741 __ai int32x4_t vmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
52743   __ret = vcombine_s32(__p0, vmovn_s64(__p1));
52747 __ai int32x4_t vmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
52749   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
52758 __ai int8x16_t vmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
52760   __ret = vcombine_s8(__p0, vmovn_s16(__p1));
52764 __ai int8x16_t vmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
52766   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
52775 __ai float64x2_t vmulq_f64(float64x2_t __p0, float64x2_t __p1) {
52777   __ret = __p0 * __p1;
52781 __ai float64x2_t vmulq_f64(float64x2_t __p0, float64x2_t __p1) {
52783   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
52792 __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
52794   __ret = __p0 * __p1;
52798 __ai float64x1_t vmul_f64(float64x1_t __p0, float64x1_t __p1) {
52800   __ret = __p0 * __p1;
52843 #define vmul_lane_f64(__p0, __p1, __p2) __extension__ ({ \
52845   float64x1_t __s1 = __p1; \
52851 #define vmul_lane_f64(__p0, __p1, __p2) __extension__ ({ \
52853   float64x1_t __s1 = __p1; \
52861 #define vmulq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
52863   float64x1_t __s1 = __p1; \
52869 #define vmulq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
52871   float64x1_t __s1 = __p1; \
52919 #define vmul_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
52921   float64x2_t __s1 = __p1; \
52927 #define vmul_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
52929   float64x2_t __s1 = __p1; \
52938 #define vmulq_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
52940   uint32x4_t __s1 = __p1; \
52946 #define vmulq_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
52948   uint32x4_t __s1 = __p1; \
52959 #define vmulq_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
52961   uint16x8_t __s1 = __p1; \
52967 #define vmulq_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
52969   uint16x8_t __s1 = __p1; \
52980 #define vmulq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
52982   float64x2_t __s1 = __p1; \
52988 #define vmulq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
52990   float64x2_t __s1 = __p1; \
53001 #define vmulq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
53003   float32x4_t __s1 = __p1; \
53009 #define vmulq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
53011   float32x4_t __s1 = __p1; \
53022 #define vmulq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
53024   int32x4_t __s1 = __p1; \
53030 #define vmulq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
53032   int32x4_t __s1 = __p1; \
53043 #define vmulq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
53045   int16x8_t __s1 = __p1; \
53051 #define vmulq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
53053   int16x8_t __s1 = __p1; \
53064 #define vmul_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
53066   uint32x4_t __s1 = __p1; \
53072 #define vmul_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
53074   uint32x4_t __s1 = __p1; \
53085 #define vmul_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
53087   uint16x8_t __s1 = __p1; \
53093 #define vmul_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
53095   uint16x8_t __s1 = __p1; \
53106 #define vmul_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
53108   float32x4_t __s1 = __p1; \
53114 #define vmul_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
53116   float32x4_t __s1 = __p1; \
53127 #define vmul_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
53129   int32x4_t __s1 = __p1; \
53135 #define vmul_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
53137   int32x4_t __s1 = __p1; \
53148 #define vmul_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
53150   int16x8_t __s1 = __p1; \
53156 #define vmul_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
53158   int16x8_t __s1 = __p1; \
53169 __ai float64x1_t vmul_n_f64(float64x1_t __p0, float64_t __p1) {
53171   __ret = (float64x1_t) __builtin_neon_vmul_n_f64((int8x8_t)__p0, __p1);
53175 __ai float64x1_t vmul_n_f64(float64x1_t __p0, float64_t __p1) {
53177   __ret = (float64x1_t) __builtin_neon_vmul_n_f64((int8x8_t)__p0, __p1);
53183 __ai float64x2_t vmulq_n_f64(float64x2_t __p0, float64_t __p1) {
53185   __ret = __p0 * (float64x2_t) {__p1, __p1};
53189 __ai float64x2_t vmulq_n_f64(float64x2_t __p0, float64_t __p1) {
53192   __ret = __rev0 * (float64x2_t) {__p1, __p1};
53199 __ai poly128_t vmull_p64(poly64_t __p0, poly64_t __p1) {
53201   __ret = (poly128_t) __builtin_neon_vmull_p64(__p0, __p1);
53205 __ai poly128_t vmull_p64(poly64_t __p0, poly64_t __p1) {
53207   __ret = (poly128_t) __builtin_neon_vmull_p64(__p0, __p1);
53210 __ai poly128_t __noswap_vmull_p64(poly64_t __p0, poly64_t __p1) {
53212   __ret = (poly128_t) __builtin_neon_vmull_p64(__p0, __p1);
53218 __ai poly16x8_t vmull_high_p8(poly8x16_t __p0, poly8x16_t __p1) {
53220   __ret = vmull_p8(vget_high_p8(__p0), vget_high_p8(__p1));
53224 __ai poly16x8_t vmull_high_p8(poly8x16_t __p0, poly8x16_t __p1) {
53226   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53235 __ai uint16x8_t vmull_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
53237   __ret = vmull_u8(vget_high_u8(__p0), vget_high_u8(__p1));
53241 __ai uint16x8_t vmull_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
53243   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53252 __ai uint64x2_t vmull_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
53254   __ret = vmull_u32(vget_high_u32(__p0), vget_high_u32(__p1));
53258 __ai uint64x2_t vmull_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
53260   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
53269 __ai uint32x4_t vmull_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
53271   __ret = vmull_u16(vget_high_u16(__p0), vget_high_u16(__p1));
53275 __ai uint32x4_t vmull_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
53277   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
53286 __ai int16x8_t vmull_high_s8(int8x16_t __p0, int8x16_t __p1) {
53288   __ret = vmull_s8(vget_high_s8(__p0), vget_high_s8(__p1));
53292 __ai int16x8_t vmull_high_s8(int8x16_t __p0, int8x16_t __p1) {
53294   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
53303 __ai int64x2_t vmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
53305   __ret = vmull_s32(vget_high_s32(__p0), vget_high_s32(__p1));
53309 __ai int64x2_t vmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
53311   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
53320 __ai int32x4_t vmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
53322   __ret = vmull_s16(vget_high_s16(__p0), vget_high_s16(__p1));
53326 __ai int32x4_t vmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
53328   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
53337 __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
53339   __ret = vmull_p64((poly64_t)(vget_high_p64(__p0)), (poly64_t)(vget_high_p64(__p1)));
53343 __ai poly128_t vmull_high_p64(poly64x2_t __p0, poly64x2_t __p1) {
53345   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53353 #define vmull_high_lane_u32(__p0, __p1, __p2) __extension__ ({ \
53355   uint32x2_t __s1 = __p1; \
53361 #define vmull_high_lane_u32(__p0, __p1, __p2) __extension__ ({ \
53363   uint32x2_t __s1 = __p1; \
53374 #define vmull_high_lane_u16(__p0, __p1, __p2) __extension__ ({ \
53376   uint16x4_t __s1 = __p1; \
53382 #define vmull_high_lane_u16(__p0, __p1, __p2) __extension__ ({ \
53384   uint16x4_t __s1 = __p1; \
53395 #define vmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
53397   int32x2_t __s1 = __p1; \
53403 #define vmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
53405   int32x2_t __s1 = __p1; \
53416 #define vmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
53418   int16x4_t __s1 = __p1; \
53424 #define vmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
53426   int16x4_t __s1 = __p1; \
53437 #define vmull_high_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
53439   uint32x4_t __s1 = __p1; \
53445 #define vmull_high_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
53447   uint32x4_t __s1 = __p1; \
53458 #define vmull_high_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
53460   uint16x8_t __s1 = __p1; \
53466 #define vmull_high_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
53468   uint16x8_t __s1 = __p1; \
53479 #define vmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
53481   int32x4_t __s1 = __p1; \
53487 #define vmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
53489   int32x4_t __s1 = __p1; \
53500 #define vmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
53502   int16x8_t __s1 = __p1; \
53508 #define vmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
53510   int16x8_t __s1 = __p1; \
53521 __ai uint64x2_t vmull_high_n_u32(uint32x4_t __p0, uint32_t __p1) {
53523   __ret = vmull_n_u32(vget_high_u32(__p0), __p1);
53527 __ai uint64x2_t vmull_high_n_u32(uint32x4_t __p0, uint32_t __p1) {
53530   __ret = __noswap_vmull_n_u32(__noswap_vget_high_u32(__rev0), __p1);
53537 __ai uint32x4_t vmull_high_n_u16(uint16x8_t __p0, uint16_t __p1) {
53539   __ret = vmull_n_u16(vget_high_u16(__p0), __p1);
53543 __ai uint32x4_t vmull_high_n_u16(uint16x8_t __p0, uint16_t __p1) {
53546   __ret = __noswap_vmull_n_u16(__noswap_vget_high_u16(__rev0), __p1);
53553 __ai int64x2_t vmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
53555   __ret = vmull_n_s32(vget_high_s32(__p0), __p1);
53559 __ai int64x2_t vmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
53562   __ret = __noswap_vmull_n_s32(__noswap_vget_high_s32(__rev0), __p1);
53569 __ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
53571   __ret = vmull_n_s16(vget_high_s16(__p0), __p1);
53575 __ai int32x4_t vmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
53578   __ret = __noswap_vmull_n_s16(__noswap_vget_high_s16(__rev0), __p1);
53585 #define vmull_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
53587   uint32x4_t __s1 = __p1; \
53593 #define vmull_laneq_u32(__p0, __p1, __p2) __extension__ ({ \
53595   uint32x4_t __s1 = __p1; \
53606 #define vmull_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
53608   uint16x8_t __s1 = __p1; \
53614 #define vmull_laneq_u16(__p0, __p1, __p2) __extension__ ({ \
53616   uint16x8_t __s1 = __p1; \
53627 #define vmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
53629   int32x4_t __s1 = __p1; \
53635 #define vmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
53637   int32x4_t __s1 = __p1; \
53648 #define vmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
53650   int16x8_t __s1 = __p1; \
53656 #define vmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
53658   int16x8_t __s1 = __p1; \
53669 __ai float64x2_t vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
53671   __ret = (float64x2_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
53675 __ai float64x2_t vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
53677   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53683 __ai float64x2_t __noswap_vmulxq_f64(float64x2_t __p0, float64x2_t __p1) {
53685   __ret = (float64x2_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
53691 __ai float32x4_t vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
53693   __ret = (float32x4_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
53697 __ai float32x4_t vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
53699   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
53705 __ai float32x4_t __noswap_vmulxq_f32(float32x4_t __p0, float32x4_t __p1) {
53707   __ret = (float32x4_t) __builtin_neon_vmulxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
53713 __ai float64x1_t vmulx_f64(float64x1_t __p0, float64x1_t __p1) {
53715   __ret = (float64x1_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
53719 __ai float64x1_t vmulx_f64(float64x1_t __p0, float64x1_t __p1) {
53721   __ret = (float64x1_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
53727 __ai float32x2_t vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
53729   __ret = (float32x2_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
53733 __ai float32x2_t vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
53735   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
53741 __ai float32x2_t __noswap_vmulx_f32(float32x2_t __p0, float32x2_t __p1) {
53743   __ret = (float32x2_t) __builtin_neon_vmulx_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
53749 __ai float64_t vmulxd_f64(float64_t __p0, float64_t __p1) {
53751   __ret = (float64_t) __builtin_neon_vmulxd_f64(__p0, __p1);
53755 __ai float64_t vmulxd_f64(float64_t __p0, float64_t __p1) {
53757   __ret = (float64_t) __builtin_neon_vmulxd_f64(__p0, __p1);
53760 __ai float64_t __noswap_vmulxd_f64(float64_t __p0, float64_t __p1) {
53762   __ret = (float64_t) __builtin_neon_vmulxd_f64(__p0, __p1);
53768 __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
53770   __ret = (float32_t) __builtin_neon_vmulxs_f32(__p0, __p1);
53774 __ai float32_t vmulxs_f32(float32_t __p0, float32_t __p1) {
53776   __ret = (float32_t) __builtin_neon_vmulxs_f32(__p0, __p1);
53779 __ai float32_t __noswap_vmulxs_f32(float32_t __p0, float32_t __p1) {
53781   __ret = (float32_t) __builtin_neon_vmulxs_f32(__p0, __p1);
53824 #define vmulxq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53826   float64x1_t __s1 = __p1; \
53832 #define vmulxq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
53834   float64x1_t __s1 = __p1; \
53844 #define vmulxq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
53846   float32x2_t __s1 = __p1; \
53852 #define vmulxq_lane_f32(__p0, __p1, __p2) __extension__ ({ \
53854   float32x2_t __s1 = __p1; \
53865 #define vmulx_lane_f32(__p0, __p1, __p2) __extension__ ({ \
53867   float32x2_t __s1 = __p1; \
53873 #define vmulx_lane_f32(__p0, __p1, __p2) __extension__ ({ \
53875   float32x2_t __s1 = __p1; \
53924 #define vmulxq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
53926   float64x2_t __s1 = __p1; \
53932 #define vmulxq_laneq_f64(__p0, __p1, __p2) __extension__ ({ \
53934   float64x2_t __s1 = __p1; \
53945 #define vmulxq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
53947   float32x4_t __s1 = __p1; \
53953 #define vmulxq_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
53955   float32x4_t __s1 = __p1; \
53966 #define vmulx_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
53968   float32x4_t __s1 = __p1; \
53974 #define vmulx_laneq_f32(__p0, __p1, __p2) __extension__ ({ \
53976   float32x4_t __s1 = __p1; \
54061 __ai uint8x16_t vpaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
54063   __ret = (uint8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
54067 __ai uint8x16_t vpaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
54069   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54078 __ai uint32x4_t vpaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
54080   __ret = (uint32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
54084 __ai uint32x4_t vpaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
54086   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54095 __ai uint64x2_t vpaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
54097   __ret = (uint64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
54101 __ai uint64x2_t vpaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
54103   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54112 __ai uint16x8_t vpaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
54114   __ret = (uint16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
54118 __ai uint16x8_t vpaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
54120   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
54129 __ai int8x16_t vpaddq_s8(int8x16_t __p0, int8x16_t __p1) {
54131   __ret = (int8x16_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
54135 __ai int8x16_t vpaddq_s8(int8x16_t __p0, int8x16_t __p1) {
54137   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54146 __ai float64x2_t vpaddq_f64(float64x2_t __p0, float64x2_t __p1) {
54148   __ret = (float64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
54152 __ai float64x2_t vpaddq_f64(float64x2_t __p0, float64x2_t __p1) {
54154   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54163 __ai float32x4_t vpaddq_f32(float32x4_t __p0, float32x4_t __p1) {
54165   __ret = (float32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
54169 __ai float32x4_t vpaddq_f32(float32x4_t __p0, float32x4_t __p1) {
54171   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54180 __ai int32x4_t vpaddq_s32(int32x4_t __p0, int32x4_t __p1) {
54182   __ret = (int32x4_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
54186 __ai int32x4_t vpaddq_s32(int32x4_t __p0, int32x4_t __p1) {
54188   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54197 __ai int64x2_t vpaddq_s64(int64x2_t __p0, int64x2_t __p1) {
54199   __ret = (int64x2_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
54203 __ai int64x2_t vpaddq_s64(int64x2_t __p0, int64x2_t __p1) {
54205   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54214 __ai int16x8_t vpaddq_s16(int16x8_t __p0, int16x8_t __p1) {
54216   __ret = (int16x8_t) __builtin_neon_vpaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
54220 __ai int16x8_t vpaddq_s16(int16x8_t __p0, int16x8_t __p1) {
54222   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
54291 __ai uint8x16_t vpmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
54293   __ret = (uint8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
54297 __ai uint8x16_t vpmaxq_u8(uint8x16_t __p0, uint8x16_t __p1) {
54299   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54308 __ai uint32x4_t vpmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
54310   __ret = (uint32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
54314 __ai uint32x4_t vpmaxq_u32(uint32x4_t __p0, uint32x4_t __p1) {
54316   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54325 __ai uint16x8_t vpmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
54327   __ret = (uint16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
54331 __ai uint16x8_t vpmaxq_u16(uint16x8_t __p0, uint16x8_t __p1) {
54333   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
54342 __ai int8x16_t vpmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
54344   __ret = (int8x16_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
54348 __ai int8x16_t vpmaxq_s8(int8x16_t __p0, int8x16_t __p1) {
54350   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54359 __ai float64x2_t vpmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
54361   __ret = (float64x2_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
54365 __ai float64x2_t vpmaxq_f64(float64x2_t __p0, float64x2_t __p1) {
54367   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54376 __ai float32x4_t vpmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
54378   __ret = (float32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
54382 __ai float32x4_t vpmaxq_f32(float32x4_t __p0, float32x4_t __p1) {
54384   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54393 __ai int32x4_t vpmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
54395   __ret = (int32x4_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
54399 __ai int32x4_t vpmaxq_s32(int32x4_t __p0, int32x4_t __p1) {
54401   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54410 __ai int16x8_t vpmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
54412   __ret = (int16x8_t) __builtin_neon_vpmaxq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
54416 __ai int16x8_t vpmaxq_s16(int16x8_t __p0, int16x8_t __p1) {
54418   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
54457 __ai float64x2_t vpmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
54459   __ret = (float64x2_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
54463 __ai float64x2_t vpmaxnmq_f64(float64x2_t __p0, float64x2_t __p1) {
54465   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54474 __ai float32x4_t vpmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
54476   __ret = (float32x4_t) __builtin_neon_vpmaxnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
54480 __ai float32x4_t vpmaxnmq_f32(float32x4_t __p0, float32x4_t __p1) {
54482   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54491 __ai float32x2_t vpmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
54493   __ret = (float32x2_t) __builtin_neon_vpmaxnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
54497 __ai float32x2_t vpmaxnm_f32(float32x2_t __p0, float32x2_t __p1) {
54499   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54538 __ai uint8x16_t vpminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
54540   __ret = (uint8x16_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
54544 __ai uint8x16_t vpminq_u8(uint8x16_t __p0, uint8x16_t __p1) {
54546   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54555 __ai uint32x4_t vpminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
54557   __ret = (uint32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
54561 __ai uint32x4_t vpminq_u32(uint32x4_t __p0, uint32x4_t __p1) {
54563   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54572 __ai uint16x8_t vpminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
54574   __ret = (uint16x8_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
54578 __ai uint16x8_t vpminq_u16(uint16x8_t __p0, uint16x8_t __p1) {
54580   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
54589 __ai int8x16_t vpminq_s8(int8x16_t __p0, int8x16_t __p1) {
54591   __ret = (int8x16_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
54595 __ai int8x16_t vpminq_s8(int8x16_t __p0, int8x16_t __p1) {
54597   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
54606 __ai float64x2_t vpminq_f64(float64x2_t __p0, float64x2_t __p1) {
54608   __ret = (float64x2_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
54612 __ai float64x2_t vpminq_f64(float64x2_t __p0, float64x2_t __p1) {
54614   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54623 __ai float32x4_t vpminq_f32(float32x4_t __p0, float32x4_t __p1) {
54625   __ret = (float32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
54629 __ai float32x4_t vpminq_f32(float32x4_t __p0, float32x4_t __p1) {
54631   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54640 __ai int32x4_t vpminq_s32(int32x4_t __p0, int32x4_t __p1) {
54642   __ret = (int32x4_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
54646 __ai int32x4_t vpminq_s32(int32x4_t __p0, int32x4_t __p1) {
54648   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54657 __ai int16x8_t vpminq_s16(int16x8_t __p0, int16x8_t __p1) {
54659   __ret = (int16x8_t) __builtin_neon_vpminq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
54663 __ai int16x8_t vpminq_s16(int16x8_t __p0, int16x8_t __p1) {
54665   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
54704 __ai float64x2_t vpminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
54706   __ret = (float64x2_t) __builtin_neon_vpminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
54710 __ai float64x2_t vpminnmq_f64(float64x2_t __p0, float64x2_t __p1) {
54712   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54721 __ai float32x4_t vpminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
54723   __ret = (float32x4_t) __builtin_neon_vpminnmq_v((int8x16_t)__p0, (int8x16_t)__p1, 41);
54727 __ai float32x4_t vpminnmq_f32(float32x4_t __p0, float32x4_t __p1) {
54729   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
54738 __ai float32x2_t vpminnm_f32(float32x2_t __p0, float32x2_t __p1) {
54740   __ret = (float32x2_t) __builtin_neon_vpminnm_v((int8x8_t)__p0, (int8x8_t)__p1, 9);
54744 __ai float32x2_t vpminnm_f32(float32x2_t __p0, float32x2_t __p1) {
54746   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
54871 __ai uint8_t vqaddb_u8(uint8_t __p0, uint8_t __p1) {
54873   __ret = (uint8_t) __builtin_neon_vqaddb_u8(__p0, __p1);
54877 __ai uint8_t vqaddb_u8(uint8_t __p0, uint8_t __p1) {
54879   __ret = (uint8_t) __builtin_neon_vqaddb_u8(__p0, __p1);
54885 __ai uint32_t vqadds_u32(uint32_t __p0, uint32_t __p1) {
54887   __ret = (uint32_t) __builtin_neon_vqadds_u32(__p0, __p1);
54891 __ai uint32_t vqadds_u32(uint32_t __p0, uint32_t __p1) {
54893   __ret = (uint32_t) __builtin_neon_vqadds_u32(__p0, __p1);
54899 __ai uint64_t vqaddd_u64(uint64_t __p0, uint64_t __p1) {
54901   __ret = (uint64_t) __builtin_neon_vqaddd_u64(__p0, __p1);
54905 __ai uint64_t vqaddd_u64(uint64_t __p0, uint64_t __p1) {
54907   __ret = (uint64_t) __builtin_neon_vqaddd_u64(__p0, __p1);
54913 __ai uint16_t vqaddh_u16(uint16_t __p0, uint16_t __p1) {
54915   __ret = (uint16_t) __builtin_neon_vqaddh_u16(__p0, __p1);
54919 __ai uint16_t vqaddh_u16(uint16_t __p0, uint16_t __p1) {
54921   __ret = (uint16_t) __builtin_neon_vqaddh_u16(__p0, __p1);
54927 __ai int8_t vqaddb_s8(int8_t __p0, int8_t __p1) {
54929   __ret = (int8_t) __builtin_neon_vqaddb_s8(__p0, __p1);
54933 __ai int8_t vqaddb_s8(int8_t __p0, int8_t __p1) {
54935   __ret = (int8_t) __builtin_neon_vqaddb_s8(__p0, __p1);
54941 __ai int32_t vqadds_s32(int32_t __p0, int32_t __p1) {
54943   __ret = (int32_t) __builtin_neon_vqadds_s32(__p0, __p1);
54947 __ai int32_t vqadds_s32(int32_t __p0, int32_t __p1) {
54949   __ret = (int32_t) __builtin_neon_vqadds_s32(__p0, __p1);
54955 __ai int64_t vqaddd_s64(int64_t __p0, int64_t __p1) {
54957   __ret = (int64_t) __builtin_neon_vqaddd_s64(__p0, __p1);
54961 __ai int64_t vqaddd_s64(int64_t __p0, int64_t __p1) {
54963   __ret = (int64_t) __builtin_neon_vqaddd_s64(__p0, __p1);
54969 __ai int16_t vqaddh_s16(int16_t __p0, int16_t __p1) {
54971   __ret = (int16_t) __builtin_neon_vqaddh_s16(__p0, __p1);
54975 __ai int16_t vqaddh_s16(int16_t __p0, int16_t __p1) {
54977   __ret = (int16_t) __builtin_neon_vqaddh_s16(__p0, __p1);
54983 __ai int64_t vqdmlals_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
54985   __ret = (int64_t) __builtin_neon_vqdmlals_s32(__p0, __p1, __p2);
54989 __ai int64_t vqdmlals_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
54991   __ret = (int64_t) __builtin_neon_vqdmlals_s32(__p0, __p1, __p2);
54997 __ai int32_t vqdmlalh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
54999   __ret = (int32_t) __builtin_neon_vqdmlalh_s16(__p0, __p1, __p2);
55003 __ai int32_t vqdmlalh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
55005   __ret = (int32_t) __builtin_neon_vqdmlalh_s16(__p0, __p1, __p2);
55011 __ai int64x2_t vqdmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
55013   __ret = vqdmlal_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
55017 __ai int64x2_t vqdmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
55019   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55029 __ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
55031   __ret = vqdmlal_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
55035 __ai int32x4_t vqdmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
55037   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
55047 #define vqdmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55049   int32x4_t __s1 = __p1; \
55056 #define vqdmlal_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55058   int32x4_t __s1 = __p1; \
55071 #define vqdmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55073   int16x8_t __s1 = __p1; \
55080 #define vqdmlal_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55082   int16x8_t __s1 = __p1; \
55095 #define vqdmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55097   int32x4_t __s1 = __p1; \
55104 #define vqdmlal_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55106   int32x4_t __s1 = __p1; \
55119 #define vqdmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55121   int16x8_t __s1 = __p1; \
55128 #define vqdmlal_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55130   int16x8_t __s1 = __p1; \
55143 __ai int64x2_t vqdmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
55145   __ret = vqdmlal_n_s32(__p0, vget_high_s32(__p1), __p2);
55149 __ai int64x2_t vqdmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
55151   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55160 __ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
55162   __ret = vqdmlal_n_s16(__p0, vget_high_s16(__p1), __p2);
55166 __ai int32x4_t vqdmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
55168   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
55177 #define vqdmlals_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55179   int32_t __s1 = __p1; \
55186 #define vqdmlals_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55188   int32_t __s1 = __p1; \
55198 #define vqdmlalh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55200   int16_t __s1 = __p1; \
55207 #define vqdmlalh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55209   int16_t __s1 = __p1; \
55219 #define vqdmlals_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55221   int32_t __s1 = __p1; \
55228 #define vqdmlals_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55230   int32_t __s1 = __p1; \
55240 #define vqdmlalh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55242   int16_t __s1 = __p1; \
55249 #define vqdmlalh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55251   int16_t __s1 = __p1; \
55261 #define vqdmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55263   int32x2_t __s1 = __p1; \
55270 #define vqdmlal_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55272   int32x2_t __s1 = __p1; \
55285 #define vqdmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55287   int16x4_t __s1 = __p1; \
55294 #define vqdmlal_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55296   int16x4_t __s1 = __p1; \
55309 __ai int64_t vqdmlsls_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
55311   __ret = (int64_t) __builtin_neon_vqdmlsls_s32(__p0, __p1, __p2);
55315 __ai int64_t vqdmlsls_s32(int64_t __p0, int32_t __p1, int32_t __p2) {
55317   __ret = (int64_t) __builtin_neon_vqdmlsls_s32(__p0, __p1, __p2);
55323 __ai int32_t vqdmlslh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
55325   __ret = (int32_t) __builtin_neon_vqdmlslh_s16(__p0, __p1, __p2);
55329 __ai int32_t vqdmlslh_s16(int32_t __p0, int16_t __p1, int16_t __p2) {
55331   __ret = (int32_t) __builtin_neon_vqdmlslh_s16(__p0, __p1, __p2);
55337 __ai int64x2_t vqdmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
55339   __ret = vqdmlsl_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
55343 __ai int64x2_t vqdmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
55345   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55355 __ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
55357   __ret = vqdmlsl_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
55361 __ai int32x4_t vqdmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
55363   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
55373 #define vqdmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55375   int32x4_t __s1 = __p1; \
55382 #define vqdmlsl_high_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55384   int32x4_t __s1 = __p1; \
55397 #define vqdmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55399   int16x8_t __s1 = __p1; \
55406 #define vqdmlsl_high_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55408   int16x8_t __s1 = __p1; \
55421 #define vqdmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55423   int32x4_t __s1 = __p1; \
55430 #define vqdmlsl_high_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55432   int32x4_t __s1 = __p1; \
55445 #define vqdmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55447   int16x8_t __s1 = __p1; \
55454 #define vqdmlsl_high_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55456   int16x8_t __s1 = __p1; \
55469 __ai int64x2_t vqdmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
55471   __ret = vqdmlsl_n_s32(__p0, vget_high_s32(__p1), __p2);
55475 __ai int64x2_t vqdmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
55477   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55486 __ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
55488   __ret = vqdmlsl_n_s16(__p0, vget_high_s16(__p1), __p2);
55492 __ai int32x4_t vqdmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
55494   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
55503 #define vqdmlsls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55505   int32_t __s1 = __p1; \
55512 #define vqdmlsls_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55514   int32_t __s1 = __p1; \
55524 #define vqdmlslh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55526   int16_t __s1 = __p1; \
55533 #define vqdmlslh_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55535   int16_t __s1 = __p1; \
55545 #define vqdmlsls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55547   int32_t __s1 = __p1; \
55554 #define vqdmlsls_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55556   int32_t __s1 = __p1; \
55566 #define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55568   int16_t __s1 = __p1; \
55575 #define vqdmlslh_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55577   int16_t __s1 = __p1; \
55587 #define vqdmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55589   int32x2_t __s1 = __p1; \
55596 #define vqdmlsl_laneq_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
55598   int32x2_t __s1 = __p1; \
55611 #define vqdmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55613   int16x4_t __s1 = __p1; \
55620 #define vqdmlsl_laneq_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
55622   int16x4_t __s1 = __p1; \
55635 __ai int32_t vqdmulhs_s32(int32_t __p0, int32_t __p1) {
55637   __ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
55641 __ai int32_t vqdmulhs_s32(int32_t __p0, int32_t __p1) {
55643   __ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
55646 __ai int32_t __noswap_vqdmulhs_s32(int32_t __p0, int32_t __p1) {
55648   __ret = (int32_t) __builtin_neon_vqdmulhs_s32(__p0, __p1);
55654 __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
55656   __ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
55660 __ai int16_t vqdmulhh_s16(int16_t __p0, int16_t __p1) {
55662   __ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
55665 __ai int16_t __noswap_vqdmulhh_s16(int16_t __p0, int16_t __p1) {
55667   __ret = (int16_t) __builtin_neon_vqdmulhh_s16(__p0, __p1);
55749 #define vqdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
55751   int32x4_t __s1 = __p1; \
55757 #define vqdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
55759   int32x4_t __s1 = __p1; \
55770 #define vqdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
55772   int16x8_t __s1 = __p1; \
55778 #define vqdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
55780   int16x8_t __s1 = __p1; \
55791 #define vqdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
55793   int32x4_t __s1 = __p1; \
55799 #define vqdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
55801   int32x4_t __s1 = __p1; \
55812 #define vqdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
55814   int16x8_t __s1 = __p1; \
55820 #define vqdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
55822   int16x8_t __s1 = __p1; \
55833 __ai int64_t vqdmulls_s32(int32_t __p0, int32_t __p1) {
55835   __ret = (int64_t) __builtin_neon_vqdmulls_s32(__p0, __p1);
55839 __ai int64_t vqdmulls_s32(int32_t __p0, int32_t __p1) {
55841   __ret = (int64_t) __builtin_neon_vqdmulls_s32(__p0, __p1);
55844 __ai int64_t __noswap_vqdmulls_s32(int32_t __p0, int32_t __p1) {
55846   __ret = (int64_t) __builtin_neon_vqdmulls_s32(__p0, __p1);
55852 __ai int32_t vqdmullh_s16(int16_t __p0, int16_t __p1) {
55854   __ret = (int32_t) __builtin_neon_vqdmullh_s16(__p0, __p1);
55858 __ai int32_t vqdmullh_s16(int16_t __p0, int16_t __p1) {
55860   __ret = (int32_t) __builtin_neon_vqdmullh_s16(__p0, __p1);
55863 __ai int32_t __noswap_vqdmullh_s16(int16_t __p0, int16_t __p1) {
55865   __ret = (int32_t) __builtin_neon_vqdmullh_s16(__p0, __p1);
55871 __ai int64x2_t vqdmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
55873   __ret = vqdmull_s32(vget_high_s32(__p0), vget_high_s32(__p1));
55877 __ai int64x2_t vqdmull_high_s32(int32x4_t __p0, int32x4_t __p1) {
55879   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
55888 __ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
55890   __ret = vqdmull_s16(vget_high_s16(__p0), vget_high_s16(__p1));
55894 __ai int32x4_t vqdmull_high_s16(int16x8_t __p0, int16x8_t __p1) {
55896   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
55905 #define vqdmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
55907   int32x2_t __s1 = __p1; \
55913 #define vqdmull_high_lane_s32(__p0, __p1, __p2) __extension__ ({ \
55915   int32x2_t __s1 = __p1; \
55926 #define vqdmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
55928   int16x4_t __s1 = __p1; \
55934 #define vqdmull_high_lane_s16(__p0, __p1, __p2) __extension__ ({ \
55936   int16x4_t __s1 = __p1; \
55947 #define vqdmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
55949   int32x4_t __s1 = __p1; \
55955 #define vqdmull_high_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
55957   int32x4_t __s1 = __p1; \
55968 #define vqdmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
55970   int16x8_t __s1 = __p1; \
55976 #define vqdmull_high_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
55978   int16x8_t __s1 = __p1; \
55989 __ai int64x2_t vqdmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
55991   __ret = vqdmull_n_s32(vget_high_s32(__p0), __p1);
55995 __ai int64x2_t vqdmull_high_n_s32(int32x4_t __p0, int32_t __p1) {
55998   __ret = __noswap_vqdmull_n_s32(__noswap_vget_high_s32(__rev0), __p1);
56005 __ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
56007   __ret = vqdmull_n_s16(vget_high_s16(__p0), __p1);
56011 __ai int32x4_t vqdmull_high_n_s16(int16x8_t __p0, int16_t __p1) {
56014   __ret = __noswap_vqdmull_n_s16(__noswap_vget_high_s16(__rev0), __p1);
56097 #define vqdmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56099   int32x4_t __s1 = __p1; \
56105 #define vqdmull_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56107   int32x4_t __s1 = __p1; \
56118 #define vqdmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56120   int16x8_t __s1 = __p1; \
56126 #define vqdmull_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56128   int16x8_t __s1 = __p1; \
56223 __ai uint16x8_t vqmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
56225   __ret = vcombine_u16(__p0, vqmovn_u32(__p1));
56229 __ai uint16x8_t vqmovn_high_u32(uint16x4_t __p0, uint32x4_t __p1) {
56231   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56240 __ai uint32x4_t vqmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
56242   __ret = vcombine_u32(__p0, vqmovn_u64(__p1));
56246 __ai uint32x4_t vqmovn_high_u64(uint32x2_t __p0, uint64x2_t __p1) {
56248   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
56257 __ai uint8x16_t vqmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
56259   __ret = vcombine_u8(__p0, vqmovn_u16(__p1));
56263 __ai uint8x16_t vqmovn_high_u16(uint8x8_t __p0, uint16x8_t __p1) {
56265   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56274 __ai int16x8_t vqmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
56276   __ret = vcombine_s16(__p0, vqmovn_s32(__p1));
56280 __ai int16x8_t vqmovn_high_s32(int16x4_t __p0, int32x4_t __p1) {
56282   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56291 __ai int32x4_t vqmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
56293   __ret = vcombine_s32(__p0, vqmovn_s64(__p1));
56297 __ai int32x4_t vqmovn_high_s64(int32x2_t __p0, int64x2_t __p1) {
56299   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
56308 __ai int8x16_t vqmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
56310   __ret = vcombine_s8(__p0, vqmovn_s16(__p1));
56314 __ai int8x16_t vqmovn_high_s16(int8x8_t __p0, int16x8_t __p1) {
56316   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56367 __ai int16x8_t vqmovun_high_s32(int16x4_t __p0, int32x4_t __p1) {
56369   __ret = vcombine_u16((uint16x4_t)(__p0), vqmovun_s32(__p1));
56373 __ai int16x8_t vqmovun_high_s32(int16x4_t __p0, int32x4_t __p1) {
56375   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
56384 __ai int32x4_t vqmovun_high_s64(int32x2_t __p0, int64x2_t __p1) {
56386   __ret = vcombine_u32((uint32x2_t)(__p0), vqmovun_s64(__p1));
56390 __ai int32x4_t vqmovun_high_s64(int32x2_t __p0, int64x2_t __p1) {
56392   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
56401 __ai int8x16_t vqmovun_high_s16(int8x8_t __p0, int16x8_t __p1) {
56403   __ret = vcombine_u8((uint8x8_t)(__p0), vqmovun_s16(__p1));
56407 __ai int8x16_t vqmovun_high_s16(int8x8_t __p0, int16x8_t __p1) {
56409   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
56504 __ai int32_t vqrdmulhs_s32(int32_t __p0, int32_t __p1) {
56506   __ret = (int32_t) __builtin_neon_vqrdmulhs_s32(__p0, __p1);
56510 __ai int32_t vqrdmulhs_s32(int32_t __p0, int32_t __p1) {
56512   __ret = (int32_t) __builtin_neon_vqrdmulhs_s32(__p0, __p1);
56515 __ai int32_t __noswap_vqrdmulhs_s32(int32_t __p0, int32_t __p1) {
56517   __ret = (int32_t) __builtin_neon_vqrdmulhs_s32(__p0, __p1);
56523 __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
56525   __ret = (int16_t) __builtin_neon_vqrdmulhh_s16(__p0, __p1);
56529 __ai int16_t vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
56531   __ret = (int16_t) __builtin_neon_vqrdmulhh_s16(__p0, __p1);
56534 __ai int16_t __noswap_vqrdmulhh_s16(int16_t __p0, int16_t __p1) {
56536   __ret = (int16_t) __builtin_neon_vqrdmulhh_s16(__p0, __p1);
56618 #define vqrdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56620   int32x4_t __s1 = __p1; \
56626 #define vqrdmulhq_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56628   int32x4_t __s1 = __p1; \
56639 #define vqrdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56641   int16x8_t __s1 = __p1; \
56647 #define vqrdmulhq_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56649   int16x8_t __s1 = __p1; \
56660 #define vqrdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56662   int32x4_t __s1 = __p1; \
56668 #define vqrdmulh_laneq_s32(__p0, __p1, __p2) __extension__ ({ \
56670   int32x4_t __s1 = __p1; \
56681 #define vqrdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56683   int16x8_t __s1 = __p1; \
56689 #define vqrdmulh_laneq_s16(__p0, __p1, __p2) __extension__ ({ \
56691   int16x8_t __s1 = __p1; \
56702 __ai uint8_t vqrshlb_u8(uint8_t __p0, uint8_t __p1) {
56704   __ret = (uint8_t) __builtin_neon_vqrshlb_u8(__p0, __p1);
56708 __ai uint8_t vqrshlb_u8(uint8_t __p0, uint8_t __p1) {
56710   __ret = (uint8_t) __builtin_neon_vqrshlb_u8(__p0, __p1);
56716 __ai uint32_t vqrshls_u32(uint32_t __p0, uint32_t __p1) {
56718   __ret = (uint32_t) __builtin_neon_vqrshls_u32(__p0, __p1);
56722 __ai uint32_t vqrshls_u32(uint32_t __p0, uint32_t __p1) {
56724   __ret = (uint32_t) __builtin_neon_vqrshls_u32(__p0, __p1);
56730 __ai uint64_t vqrshld_u64(uint64_t __p0, uint64_t __p1) {
56732   __ret = (uint64_t) __builtin_neon_vqrshld_u64(__p0, __p1);
56736 __ai uint64_t vqrshld_u64(uint64_t __p0, uint64_t __p1) {
56738   __ret = (uint64_t) __builtin_neon_vqrshld_u64(__p0, __p1);
56744 __ai uint16_t vqrshlh_u16(uint16_t __p0, uint16_t __p1) {
56746   __ret = (uint16_t) __builtin_neon_vqrshlh_u16(__p0, __p1);
56750 __ai uint16_t vqrshlh_u16(uint16_t __p0, uint16_t __p1) {
56752   __ret = (uint16_t) __builtin_neon_vqrshlh_u16(__p0, __p1);
56758 __ai int8_t vqrshlb_s8(int8_t __p0, int8_t __p1) {
56760   __ret = (int8_t) __builtin_neon_vqrshlb_s8(__p0, __p1);
56764 __ai int8_t vqrshlb_s8(int8_t __p0, int8_t __p1) {
56766   __ret = (int8_t) __builtin_neon_vqrshlb_s8(__p0, __p1);
56772 __ai int32_t vqrshls_s32(int32_t __p0, int32_t __p1) {
56774   __ret = (int32_t) __builtin_neon_vqrshls_s32(__p0, __p1);
56778 __ai int32_t vqrshls_s32(int32_t __p0, int32_t __p1) {
56780   __ret = (int32_t) __builtin_neon_vqrshls_s32(__p0, __p1);
56786 __ai int64_t vqrshld_s64(int64_t __p0, int64_t __p1) {
56788   __ret = (int64_t) __builtin_neon_vqrshld_s64(__p0, __p1);
56792 __ai int64_t vqrshld_s64(int64_t __p0, int64_t __p1) {
56794   __ret = (int64_t) __builtin_neon_vqrshld_s64(__p0, __p1);
56800 __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
56802   __ret = (int16_t) __builtin_neon_vqrshlh_s16(__p0, __p1);
56806 __ai int16_t vqrshlh_s16(int16_t __p0, int16_t __p1) {
56808   __ret = (int16_t) __builtin_neon_vqrshlh_s16(__p0, __p1);
56940 #define vqrshrns_n_u32(__p0, __p1) __extension__ ({ \
56943   __ret = (uint16_t) __builtin_neon_vqrshrns_n_u32(__s0, __p1); \
56947 #define vqrshrns_n_u32(__p0, __p1) __extension__ ({ \
56950   __ret = (uint16_t) __builtin_neon_vqrshrns_n_u32(__s0, __p1); \
56956 #define vqrshrnd_n_u64(__p0, __p1) __extension__ ({ \
56959   __ret = (uint32_t) __builtin_neon_vqrshrnd_n_u64(__s0, __p1); \
56963 #define vqrshrnd_n_u64(__p0, __p1) __extension__ ({ \
56966   __ret = (uint32_t) __builtin_neon_vqrshrnd_n_u64(__s0, __p1); \
56972 #define vqrshrnh_n_u16(__p0, __p1) __extension__ ({ \
56975   __ret = (uint8_t) __builtin_neon_vqrshrnh_n_u16(__s0, __p1); \
56979 #define vqrshrnh_n_u16(__p0, __p1) __extension__ ({ \
56982   __ret = (uint8_t) __builtin_neon_vqrshrnh_n_u16(__s0, __p1); \
56988 #define vqrshrns_n_s32(__p0, __p1) __extension__ ({ \
56991   __ret = (int16_t) __builtin_neon_vqrshrns_n_s32(__s0, __p1); \
56995 #define vqrshrns_n_s32(__p0, __p1) __extension__ ({ \
56998   __ret = (int16_t) __builtin_neon_vqrshrns_n_s32(__s0, __p1); \
57004 #define vqrshrnd_n_s64(__p0, __p1) __extension__ ({ \
57007   __ret = (int32_t) __builtin_neon_vqrshrnd_n_s64(__s0, __p1); \
57011 #define vqrshrnd_n_s64(__p0, __p1) __extension__ ({ \
57014   __ret = (int32_t) __builtin_neon_vqrshrnd_n_s64(__s0, __p1); \
57020 #define vqrshrnh_n_s16(__p0, __p1) __extension__ ({ \
57023   __ret = (int8_t) __builtin_neon_vqrshrnh_n_s16(__s0, __p1); \
57027 #define vqrshrnh_n_s16(__p0, __p1) __extension__ ({ \
57030   __ret = (int8_t) __builtin_neon_vqrshrnh_n_s16(__s0, __p1); \
57099 #define vqrshruns_n_s32(__p0, __p1) __extension__ ({ \
57102   __ret = (int16_t) __builtin_neon_vqrshruns_n_s32(__s0, __p1); \
57106 #define vqrshruns_n_s32(__p0, __p1) __extension__ ({ \
57109   __ret = (int16_t) __builtin_neon_vqrshruns_n_s32(__s0, __p1); \
57115 #define vqrshrund_n_s64(__p0, __p1) __extension__ ({ \
57118   __ret = (int32_t) __builtin_neon_vqrshrund_n_s64(__s0, __p1); \
57122 #define vqrshrund_n_s64(__p0, __p1) __extension__ ({ \
57125   __ret = (int32_t) __builtin_neon_vqrshrund_n_s64(__s0, __p1); \
57131 #define vqrshrunh_n_s16(__p0, __p1) __extension__ ({ \
57134   __ret = (int8_t) __builtin_neon_vqrshrunh_n_s16(__s0, __p1); \
57138 #define vqrshrunh_n_s16(__p0, __p1) __extension__ ({ \
57141   __ret = (int8_t) __builtin_neon_vqrshrunh_n_s16(__s0, __p1); \
57147 __ai uint8_t vqshlb_u8(uint8_t __p0, uint8_t __p1) {
57149   __ret = (uint8_t) __builtin_neon_vqshlb_u8(__p0, __p1);
57153 __ai uint8_t vqshlb_u8(uint8_t __p0, uint8_t __p1) {
57155   __ret = (uint8_t) __builtin_neon_vqshlb_u8(__p0, __p1);
57161 __ai uint32_t vqshls_u32(uint32_t __p0, uint32_t __p1) {
57163   __ret = (uint32_t) __builtin_neon_vqshls_u32(__p0, __p1);
57167 __ai uint32_t vqshls_u32(uint32_t __p0, uint32_t __p1) {
57169   __ret = (uint32_t) __builtin_neon_vqshls_u32(__p0, __p1);
57175 __ai uint64_t vqshld_u64(uint64_t __p0, uint64_t __p1) {
57177   __ret = (uint64_t) __builtin_neon_vqshld_u64(__p0, __p1);
57181 __ai uint64_t vqshld_u64(uint64_t __p0, uint64_t __p1) {
57183   __ret = (uint64_t) __builtin_neon_vqshld_u64(__p0, __p1);
57189 __ai uint16_t vqshlh_u16(uint16_t __p0, uint16_t __p1) {
57191   __ret = (uint16_t) __builtin_neon_vqshlh_u16(__p0, __p1);
57195 __ai uint16_t vqshlh_u16(uint16_t __p0, uint16_t __p1) {
57197   __ret = (uint16_t) __builtin_neon_vqshlh_u16(__p0, __p1);
57203 __ai int8_t vqshlb_s8(int8_t __p0, int8_t __p1) {
57205   __ret = (int8_t) __builtin_neon_vqshlb_s8(__p0, __p1);
57209 __ai int8_t vqshlb_s8(int8_t __p0, int8_t __p1) {
57211   __ret = (int8_t) __builtin_neon_vqshlb_s8(__p0, __p1);
57217 __ai int32_t vqshls_s32(int32_t __p0, int32_t __p1) {
57219   __ret = (int32_t) __builtin_neon_vqshls_s32(__p0, __p1);
57223 __ai int32_t vqshls_s32(int32_t __p0, int32_t __p1) {
57225   __ret = (int32_t) __builtin_neon_vqshls_s32(__p0, __p1);
57231 __ai int64_t vqshld_s64(int64_t __p0, int64_t __p1) {
57233   __ret = (int64_t) __builtin_neon_vqshld_s64(__p0, __p1);
57237 __ai int64_t vqshld_s64(int64_t __p0, int64_t __p1) {
57239   __ret = (int64_t) __builtin_neon_vqshld_s64(__p0, __p1);
57245 __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
57247   __ret = (int16_t) __builtin_neon_vqshlh_s16(__p0, __p1);
57251 __ai int16_t vqshlh_s16(int16_t __p0, int16_t __p1) {
57253   __ret = (int16_t) __builtin_neon_vqshlh_s16(__p0, __p1);
57259 #define vqshlb_n_u8(__p0, __p1) __extension__ ({ \
57262   __ret = (uint8_t) __builtin_neon_vqshlb_n_u8(__s0, __p1); \
57266 #define vqshlb_n_u8(__p0, __p1) __extension__ ({ \
57269   __ret = (uint8_t) __builtin_neon_vqshlb_n_u8(__s0, __p1); \
57275 #define vqshls_n_u32(__p0, __p1) __extension__ ({ \
57278   __ret = (uint32_t) __builtin_neon_vqshls_n_u32(__s0, __p1); \
57282 #define vqshls_n_u32(__p0, __p1) __extension__ ({ \
57285   __ret = (uint32_t) __builtin_neon_vqshls_n_u32(__s0, __p1); \
57291 #define vqshld_n_u64(__p0, __p1) __extension__ ({ \
57294   __ret = (uint64_t) __builtin_neon_vqshld_n_u64(__s0, __p1); \
57298 #define vqshld_n_u64(__p0, __p1) __extension__ ({ \
57301   __ret = (uint64_t) __builtin_neon_vqshld_n_u64(__s0, __p1); \
57307 #define vqshlh_n_u16(__p0, __p1) __extension__ ({ \
57310   __ret = (uint16_t) __builtin_neon_vqshlh_n_u16(__s0, __p1); \
57314 #define vqshlh_n_u16(__p0, __p1) __extension__ ({ \
57317   __ret = (uint16_t) __builtin_neon_vqshlh_n_u16(__s0, __p1); \
57323 #define vqshlb_n_s8(__p0, __p1) __extension__ ({ \
57326   __ret = (int8_t) __builtin_neon_vqshlb_n_s8(__s0, __p1); \
57330 #define vqshlb_n_s8(__p0, __p1) __extension__ ({ \
57333   __ret = (int8_t) __builtin_neon_vqshlb_n_s8(__s0, __p1); \
57339 #define vqshls_n_s32(__p0, __p1) __extension__ ({ \
57342   __ret = (int32_t) __builtin_neon_vqshls_n_s32(__s0, __p1); \
57346 #define vqshls_n_s32(__p0, __p1) __extension__ ({ \
57349   __ret = (int32_t) __builtin_neon_vqshls_n_s32(__s0, __p1); \
57355 #define vqshld_n_s64(__p0, __p1) __extension__ ({ \
57358   __ret = (int64_t) __builtin_neon_vqshld_n_s64(__s0, __p1); \
57362 #define vqshld_n_s64(__p0, __p1) __extension__ ({ \
57365   __ret = (int64_t) __builtin_neon_vqshld_n_s64(__s0, __p1); \
57371 #define vqshlh_n_s16(__p0, __p1) __extension__ ({ \
57374   __ret = (int16_t) __builtin_neon_vqshlh_n_s16(__s0, __p1); \
57378 #define vqshlh_n_s16(__p0, __p1) __extension__ ({ \
57381   __ret = (int16_t) __builtin_neon_vqshlh_n_s16(__s0, __p1); \
57387 #define vqshlub_n_s8(__p0, __p1) __extension__ ({ \
57390   __ret = (int8_t) __builtin_neon_vqshlub_n_s8(__s0, __p1); \
57394 #define vqshlub_n_s8(__p0, __p1) __extension__ ({ \
57397   __ret = (int8_t) __builtin_neon_vqshlub_n_s8(__s0, __p1); \
57403 #define vqshlus_n_s32(__p0, __p1) __extension__ ({ \
57406   __ret = (int32_t) __builtin_neon_vqshlus_n_s32(__s0, __p1); \
57410 #define vqshlus_n_s32(__p0, __p1) __extension__ ({ \
57413   __ret = (int32_t) __builtin_neon_vqshlus_n_s32(__s0, __p1); \
57419 #define vqshlud_n_s64(__p0, __p1) __extension__ ({ \
57422   __ret = (int64_t) __builtin_neon_vqshlud_n_s64(__s0, __p1); \
57426 #define vqshlud_n_s64(__p0, __p1) __extension__ ({ \
57429   __ret = (int64_t) __builtin_neon_vqshlud_n_s64(__s0, __p1); \
57435 #define vqshluh_n_s16(__p0, __p1) __extension__ ({ \
57438   __ret = (int16_t) __builtin_neon_vqshluh_n_s16(__s0, __p1); \
57442 #define vqshluh_n_s16(__p0, __p1) __extension__ ({ \
57445   __ret = (int16_t) __builtin_neon_vqshluh_n_s16(__s0, __p1); \
57577 #define vqshrns_n_u32(__p0, __p1) __extension__ ({ \
57580   __ret = (uint16_t) __builtin_neon_vqshrns_n_u32(__s0, __p1); \
57584 #define vqshrns_n_u32(__p0, __p1) __extension__ ({ \
57587   __ret = (uint16_t) __builtin_neon_vqshrns_n_u32(__s0, __p1); \
57593 #define vqshrnd_n_u64(__p0, __p1) __extension__ ({ \
57596   __ret = (uint32_t) __builtin_neon_vqshrnd_n_u64(__s0, __p1); \
57600 #define vqshrnd_n_u64(__p0, __p1) __extension__ ({ \
57603   __ret = (uint32_t) __builtin_neon_vqshrnd_n_u64(__s0, __p1); \
57609 #define vqshrnh_n_u16(__p0, __p1) __extension__ ({ \
57612   __ret = (uint8_t) __builtin_neon_vqshrnh_n_u16(__s0, __p1); \
57616 #define vqshrnh_n_u16(__p0, __p1) __extension__ ({ \
57619   __ret = (uint8_t) __builtin_neon_vqshrnh_n_u16(__s0, __p1); \
57625 #define vqshrns_n_s32(__p0, __p1) __extension__ ({ \
57628   __ret = (int16_t) __builtin_neon_vqshrns_n_s32(__s0, __p1); \
57632 #define vqshrns_n_s32(__p0, __p1) __extension__ ({ \
57635   __ret = (int16_t) __builtin_neon_vqshrns_n_s32(__s0, __p1); \
57641 #define vqshrnd_n_s64(__p0, __p1) __extension__ ({ \
57644   __ret = (int32_t) __builtin_neon_vqshrnd_n_s64(__s0, __p1); \
57648 #define vqshrnd_n_s64(__p0, __p1) __extension__ ({ \
57651   __ret = (int32_t) __builtin_neon_vqshrnd_n_s64(__s0, __p1); \
57657 #define vqshrnh_n_s16(__p0, __p1) __extension__ ({ \
57660   __ret = (int8_t) __builtin_neon_vqshrnh_n_s16(__s0, __p1); \
57664 #define vqshrnh_n_s16(__p0, __p1) __extension__ ({ \
57667   __ret = (int8_t) __builtin_neon_vqshrnh_n_s16(__s0, __p1); \
57736 #define vqshruns_n_s32(__p0, __p1) __extension__ ({ \
57739   __ret = (int16_t) __builtin_neon_vqshruns_n_s32(__s0, __p1); \
57743 #define vqshruns_n_s32(__p0, __p1) __extension__ ({ \
57746   __ret = (int16_t) __builtin_neon_vqshruns_n_s32(__s0, __p1); \
57752 #define vqshrund_n_s64(__p0, __p1) __extension__ ({ \
57755   __ret = (int32_t) __builtin_neon_vqshrund_n_s64(__s0, __p1); \
57759 #define vqshrund_n_s64(__p0, __p1) __extension__ ({ \
57762   __ret = (int32_t) __builtin_neon_vqshrund_n_s64(__s0, __p1); \
57768 #define vqshrunh_n_s16(__p0, __p1) __extension__ ({ \
57771   __ret = (int8_t) __builtin_neon_vqshrunh_n_s16(__s0, __p1); \
57775 #define vqshrunh_n_s16(__p0, __p1) __extension__ ({ \
57778   __ret = (int8_t) __builtin_neon_vqshrunh_n_s16(__s0, __p1); \
57784 __ai uint8_t vqsubb_u8(uint8_t __p0, uint8_t __p1) {
57786   __ret = (uint8_t) __builtin_neon_vqsubb_u8(__p0, __p1);
57790 __ai uint8_t vqsubb_u8(uint8_t __p0, uint8_t __p1) {
57792   __ret = (uint8_t) __builtin_neon_vqsubb_u8(__p0, __p1);
57798 __ai uint32_t vqsubs_u32(uint32_t __p0, uint32_t __p1) {
57800   __ret = (uint32_t) __builtin_neon_vqsubs_u32(__p0, __p1);
57804 __ai uint32_t vqsubs_u32(uint32_t __p0, uint32_t __p1) {
57806   __ret = (uint32_t) __builtin_neon_vqsubs_u32(__p0, __p1);
57812 __ai uint64_t vqsubd_u64(uint64_t __p0, uint64_t __p1) {
57814   __ret = (uint64_t) __builtin_neon_vqsubd_u64(__p0, __p1);
57818 __ai uint64_t vqsubd_u64(uint64_t __p0, uint64_t __p1) {
57820   __ret = (uint64_t) __builtin_neon_vqsubd_u64(__p0, __p1);
57826 __ai uint16_t vqsubh_u16(uint16_t __p0, uint16_t __p1) {
57828   __ret = (uint16_t) __builtin_neon_vqsubh_u16(__p0, __p1);
57832 __ai uint16_t vqsubh_u16(uint16_t __p0, uint16_t __p1) {
57834   __ret = (uint16_t) __builtin_neon_vqsubh_u16(__p0, __p1);
57840 __ai int8_t vqsubb_s8(int8_t __p0, int8_t __p1) {
57842   __ret = (int8_t) __builtin_neon_vqsubb_s8(__p0, __p1);
57846 __ai int8_t vqsubb_s8(int8_t __p0, int8_t __p1) {
57848   __ret = (int8_t) __builtin_neon_vqsubb_s8(__p0, __p1);
57854 __ai int32_t vqsubs_s32(int32_t __p0, int32_t __p1) {
57856   __ret = (int32_t) __builtin_neon_vqsubs_s32(__p0, __p1);
57860 __ai int32_t vqsubs_s32(int32_t __p0, int32_t __p1) {
57862   __ret = (int32_t) __builtin_neon_vqsubs_s32(__p0, __p1);
57868 __ai int64_t vqsubd_s64(int64_t __p0, int64_t __p1) {
57870   __ret = (int64_t) __builtin_neon_vqsubd_s64(__p0, __p1);
57874 __ai int64_t vqsubd_s64(int64_t __p0, int64_t __p1) {
57876   __ret = (int64_t) __builtin_neon_vqsubd_s64(__p0, __p1);
57882 __ai int16_t vqsubh_s16(int16_t __p0, int16_t __p1) {
57884   __ret = (int16_t) __builtin_neon_vqsubh_s16(__p0, __p1);
57888 __ai int16_t vqsubh_s16(int16_t __p0, int16_t __p1) {
57890   __ret = (int16_t) __builtin_neon_vqsubh_s16(__p0, __p1);
57896 __ai poly8x8_t vqtbl1_p8(poly8x16_t __p0, uint8x8_t __p1) {
57898   __ret = (poly8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 4);
57902 __ai poly8x8_t vqtbl1_p8(poly8x16_t __p0, uint8x8_t __p1) {
57904   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57913 __ai poly8x16_t vqtbl1q_p8(poly8x16_t __p0, uint8x16_t __p1) {
57915   __ret = (poly8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 36);
57919 __ai poly8x16_t vqtbl1q_p8(poly8x16_t __p0, uint8x16_t __p1) {
57921   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57930 __ai uint8x16_t vqtbl1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
57932   __ret = (uint8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
57936 __ai uint8x16_t vqtbl1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
57938   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57947 __ai int8x16_t vqtbl1q_s8(int8x16_t __p0, int8x16_t __p1) {
57949   __ret = (int8x16_t) __builtin_neon_vqtbl1q_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
57953 __ai int8x16_t vqtbl1q_s8(int8x16_t __p0, int8x16_t __p1) {
57955   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
57964 __ai uint8x8_t vqtbl1_u8(uint8x16_t __p0, uint8x8_t __p1) {
57966   __ret = (uint8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 16);
57970 __ai uint8x8_t vqtbl1_u8(uint8x16_t __p0, uint8x8_t __p1) {
57972   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57981 __ai int8x8_t vqtbl1_s8(int8x16_t __p0, int8x8_t __p1) {
57983   __ret = (int8x8_t) __builtin_neon_vqtbl1_v((int8x16_t)__p0, (int8x8_t)__p1, 0);
57987 __ai int8x8_t vqtbl1_s8(int8x16_t __p0, int8x8_t __p1) {
57989   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
57998 __ai poly8x8_t vqtbl2_p8(poly8x16x2_t __p0, uint8x8_t __p1) {
58000   __ret = (poly8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 4);
58004 __ai poly8x8_t vqtbl2_p8(poly8x16x2_t __p0, uint8x8_t __p1) {
58008   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58017 __ai poly8x16_t vqtbl2q_p8(poly8x16x2_t __p0, uint8x16_t __p1) {
58019   __ret = (poly8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 36);
58023 __ai poly8x16_t vqtbl2q_p8(poly8x16x2_t __p0, uint8x16_t __p1) {
58027   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58036 __ai uint8x16_t vqtbl2q_u8(uint8x16x2_t __p0, uint8x16_t __p1) {
58038   __ret = (uint8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 48);
58042 __ai uint8x16_t vqtbl2q_u8(uint8x16x2_t __p0, uint8x16_t __p1) {
58046   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58055 __ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, int8x16_t __p1) {
58057   __ret = (int8x16_t) __builtin_neon_vqtbl2q_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x16_t)__p1, 32);
58061 __ai int8x16_t vqtbl2q_s8(int8x16x2_t __p0, int8x16_t __p1) {
58065   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58074 __ai uint8x8_t vqtbl2_u8(uint8x16x2_t __p0, uint8x8_t __p1) {
58076   __ret = (uint8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 16);
58080 __ai uint8x8_t vqtbl2_u8(uint8x16x2_t __p0, uint8x8_t __p1) {
58084   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58093 __ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, int8x8_t __p1) {
58095   __ret = (int8x8_t) __builtin_neon_vqtbl2_v((int8x16_t)__p0.val[0], (int8x16_t)__p0.val[1], (int8x8_t)__p1, 0);
58099 __ai int8x8_t vqtbl2_s8(int8x16x2_t __p0, int8x8_t __p1) {
58103   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58112 __ai poly8x8_t vqtbl3_p8(poly8x16x3_t __p0, uint8x8_t __p1) {
58114   __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);
58118 __ai poly8x8_t vqtbl3_p8(poly8x16x3_t __p0, uint8x8_t __p1) {
58123   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58132 __ai poly8x16_t vqtbl3q_p8(poly8x16x3_t __p0, uint8x16_t __p1) {
58134   __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);
58138 __ai poly8x16_t vqtbl3q_p8(poly8x16x3_t __p0, uint8x16_t __p1) {
58143   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58152 __ai uint8x16_t vqtbl3q_u8(uint8x16x3_t __p0, uint8x16_t __p1) {
58154   __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);
58158 __ai uint8x16_t vqtbl3q_u8(uint8x16x3_t __p0, uint8x16_t __p1) {
58163   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58172 __ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, int8x16_t __p1) {
58174   __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);
58178 __ai int8x16_t vqtbl3q_s8(int8x16x3_t __p0, int8x16_t __p1) {
58183   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58192 __ai uint8x8_t vqtbl3_u8(uint8x16x3_t __p0, uint8x8_t __p1) {
58194   __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);
58198 __ai uint8x8_t vqtbl3_u8(uint8x16x3_t __p0, uint8x8_t __p1) {
58203   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58212 __ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, int8x8_t __p1) {
58214   __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);
58218 __ai int8x8_t vqtbl3_s8(int8x16x3_t __p0, int8x8_t __p1) {
58223   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58232 __ai poly8x8_t vqtbl4_p8(poly8x16x4_t __p0, uint8x8_t __p1) {
58234   __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);
58238 __ai poly8x8_t vqtbl4_p8(poly8x16x4_t __p0, uint8x8_t __p1) {
58244   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58253 __ai poly8x16_t vqtbl4q_p8(poly8x16x4_t __p0, uint8x16_t __p1) {
58255   __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);
58259 __ai poly8x16_t vqtbl4q_p8(poly8x16x4_t __p0, uint8x16_t __p1) {
58265   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58274 __ai uint8x16_t vqtbl4q_u8(uint8x16x4_t __p0, uint8x16_t __p1) {
58276   __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);
58280 __ai uint8x16_t vqtbl4q_u8(uint8x16x4_t __p0, uint8x16_t __p1) {
58286   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58295 __ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, int8x16_t __p1) {
58297   __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);
58301 __ai int8x16_t vqtbl4q_s8(int8x16x4_t __p0, int8x16_t __p1) {
58307   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58316 __ai uint8x8_t vqtbl4_u8(uint8x16x4_t __p0, uint8x8_t __p1) {
58318   __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);
58322 __ai uint8x8_t vqtbl4_u8(uint8x16x4_t __p0, uint8x8_t __p1) {
58328   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58337 __ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, int8x8_t __p1) {
58339   __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);
58343 __ai int8x8_t vqtbl4_s8(int8x16x4_t __p0, int8x8_t __p1) {
58349   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58358 __ai poly8x8_t vqtbx1_p8(poly8x8_t __p0, poly8x16_t __p1, uint8x8_t __p2) {
58360   __ret = (poly8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 4);
58364 __ai poly8x8_t vqtbx1_p8(poly8x8_t __p0, poly8x16_t __p1, uint8x8_t __p2) {
58366   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58376 __ai poly8x16_t vqtbx1q_p8(poly8x16_t __p0, poly8x16_t __p1, uint8x16_t __p2) {
58378   __ret = (poly8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 36);
58382 __ai poly8x16_t vqtbx1q_p8(poly8x16_t __p0, poly8x16_t __p1, uint8x16_t __p2) {
58384   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58394 __ai uint8x16_t vqtbx1q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
58396   __ret = (uint8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 48);
58400 __ai uint8x16_t vqtbx1q_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
58402   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58412 __ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
58414   __ret = (int8x16_t) __builtin_neon_vqtbx1q_v((int8x16_t)__p0, (int8x16_t)__p1, (int8x16_t)__p2, 32);
58418 __ai int8x16_t vqtbx1q_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
58420   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58430 __ai uint8x8_t vqtbx1_u8(uint8x8_t __p0, uint8x16_t __p1, uint8x8_t __p2) {
58432   __ret = (uint8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 16);
58436 __ai uint8x8_t vqtbx1_u8(uint8x8_t __p0, uint8x16_t __p1, uint8x8_t __p2) {
58438   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58448 __ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, int8x8_t __p2) {
58450   __ret = (int8x8_t) __builtin_neon_vqtbx1_v((int8x8_t)__p0, (int8x16_t)__p1, (int8x8_t)__p2, 0);
58454 __ai int8x8_t vqtbx1_s8(int8x8_t __p0, int8x16_t __p1, int8x8_t __p2) {
58456   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
58466 __ai poly8x8_t vqtbx2_p8(poly8x8_t __p0, poly8x16x2_t __p1, uint8x8_t __p2) {
58468   __ret = (poly8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 4);
58472 __ai poly8x8_t vqtbx2_p8(poly8x8_t __p0, poly8x16x2_t __p1, uint8x8_t __p2) {
58475   __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);
58476   __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);
58486 __ai poly8x16_t vqtbx2q_p8(poly8x16_t __p0, poly8x16x2_t __p1, uint8x16_t __p2) {
58488   __ret = (poly8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 36);
58492 __ai poly8x16_t vqtbx2q_p8(poly8x16_t __p0, poly8x16x2_t __p1, uint8x16_t __p2) {
58495   __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);
58496   __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);
58506 __ai uint8x16_t vqtbx2q_u8(uint8x16_t __p0, uint8x16x2_t __p1, uint8x16_t __p2) {
58508   __ret = (uint8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 48);
58512 __ai uint8x16_t vqtbx2q_u8(uint8x16_t __p0, uint8x16x2_t __p1, uint8x16_t __p2) {
58515   __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);
58516   __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);
58526 __ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, int8x16_t __p2) {
58528   __ret = (int8x16_t) __builtin_neon_vqtbx2q_v((int8x16_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x16_t)__p2, 32);
58532 __ai int8x16_t vqtbx2q_s8(int8x16_t __p0, int8x16x2_t __p1, int8x16_t __p2) {
58535   __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);
58536   __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);
58546 __ai uint8x8_t vqtbx2_u8(uint8x8_t __p0, uint8x16x2_t __p1, uint8x8_t __p2) {
58548   __ret = (uint8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 16);
58552 __ai uint8x8_t vqtbx2_u8(uint8x8_t __p0, uint8x16x2_t __p1, uint8x8_t __p2) {
58555   __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);
58556   __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);
58566 __ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, int8x8_t __p2) {
58568   __ret = (int8x8_t) __builtin_neon_vqtbx2_v((int8x8_t)__p0, (int8x16_t)__p1.val[0], (int8x16_t)__p1.val[1], (int8x8_t)__p2, 0);
58572 __ai int8x8_t vqtbx2_s8(int8x8_t __p0, int8x16x2_t __p1, int8x8_t __p2) {
58575   __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);
58576   __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);
58586 __ai poly8x8_t vqtbx3_p8(poly8x8_t __p0, poly8x16x3_t __p1, uint8x8_t __p2) {
58588   __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);
58592 __ai poly8x8_t vqtbx3_p8(poly8x8_t __p0, poly8x16x3_t __p1, uint8x8_t __p2) {
58595   __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);
58596   __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);
58597   __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);
58607 __ai poly8x16_t vqtbx3q_p8(poly8x16_t __p0, poly8x16x3_t __p1, uint8x16_t __p2) {
58609   __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);
58613 __ai poly8x16_t vqtbx3q_p8(poly8x16_t __p0, poly8x16x3_t __p1, uint8x16_t __p2) {
58616   __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);
58617   __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);
58618   __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);
58628 __ai uint8x16_t vqtbx3q_u8(uint8x16_t __p0, uint8x16x3_t __p1, uint8x16_t __p2) {
58630   __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);
58634 __ai uint8x16_t vqtbx3q_u8(uint8x16_t __p0, uint8x16x3_t __p1, uint8x16_t __p2) {
58637   __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);
58638   __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);
58639   __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);
58649 __ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, int8x16_t __p2) {
58651   __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);
58655 __ai int8x16_t vqtbx3q_s8(int8x16_t __p0, int8x16x3_t __p1, int8x16_t __p2) {
58658   __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);
58659   __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);
58660   __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);
58670 __ai uint8x8_t vqtbx3_u8(uint8x8_t __p0, uint8x16x3_t __p1, uint8x8_t __p2) {
58672   __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);
58676 __ai uint8x8_t vqtbx3_u8(uint8x8_t __p0, uint8x16x3_t __p1, uint8x8_t __p2) {
58679   __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);
58680   __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);
58681   __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);
58691 __ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, int8x8_t __p2) {
58693   __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);
58697 __ai int8x8_t vqtbx3_s8(int8x8_t __p0, int8x16x3_t __p1, int8x8_t __p2) {
58700   __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);
58701   __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);
58702   __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);
58712 __ai poly8x8_t vqtbx4_p8(poly8x8_t __p0, poly8x16x4_t __p1, uint8x8_t __p2) {
58714   __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);
58718 __ai poly8x8_t vqtbx4_p8(poly8x8_t __p0, poly8x16x4_t __p1, uint8x8_t __p2) {
58721   __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);
58722   __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);
58723   __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);
58724   __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);
58734 __ai poly8x16_t vqtbx4q_p8(poly8x16_t __p0, poly8x16x4_t __p1, uint8x16_t __p2) {
58736   __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);
58740 __ai poly8x16_t vqtbx4q_p8(poly8x16_t __p0, poly8x16x4_t __p1, uint8x16_t __p2) {
58743   __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);
58744   __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);
58745   __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);
58746   __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);
58756 __ai uint8x16_t vqtbx4q_u8(uint8x16_t __p0, uint8x16x4_t __p1, uint8x16_t __p2) {
58758   __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);
58762 __ai uint8x16_t vqtbx4q_u8(uint8x16_t __p0, uint8x16x4_t __p1, uint8x16_t __p2) {
58765   __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);
58766   __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);
58767   __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);
58768   __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);
58778 __ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, int8x16_t __p2) {
58780   __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);
58784 __ai int8x16_t vqtbx4q_s8(int8x16_t __p0, int8x16x4_t __p1, int8x16_t __p2) {
58787   __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);
58788   __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);
58789   __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);
58790   __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);
58800 __ai uint8x8_t vqtbx4_u8(uint8x8_t __p0, uint8x16x4_t __p1, uint8x8_t __p2) {
58802   __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);
58806 __ai uint8x8_t vqtbx4_u8(uint8x8_t __p0, uint8x16x4_t __p1, uint8x8_t __p2) {
58809   __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);
58810   __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);
58811   __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);
58812   __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);
58822 __ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, int8x8_t __p2) {
58824   __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);
58828 __ai int8x8_t vqtbx4_s8(int8x8_t __p0, int8x16x4_t __p1, int8x8_t __p2) {
58831   __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);
58832   __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);
58833   __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);
58834   __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);
58844 __ai uint16x8_t vraddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
58846   __ret = vcombine_u16(__p0, vraddhn_u32(__p1, __p2));
58850 __ai uint16x8_t vraddhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
58852   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58862 __ai uint32x4_t vraddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
58864   __ret = vcombine_u32(__p0, vraddhn_u64(__p1, __p2));
58868 __ai uint32x4_t vraddhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
58870   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58880 __ai uint8x16_t vraddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
58882   __ret = vcombine_u8(__p0, vraddhn_u16(__p1, __p2));
58886 __ai uint8x16_t vraddhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
58888   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
58898 __ai int16x8_t vraddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
58900   __ret = vcombine_s16(__p0, vraddhn_s32(__p1, __p2));
58904 __ai int16x8_t vraddhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
58906   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
58916 __ai int32x4_t vraddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
58918   __ret = vcombine_s32(__p0, vraddhn_s64(__p1, __p2));
58922 __ai int32x4_t vraddhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
58924   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
58934 __ai int8x16_t vraddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
58936   __ret = vcombine_s8(__p0, vraddhn_s16(__p1, __p2));
58940 __ai int8x16_t vraddhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
58942   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
59106 __ai float64x2_t vrecpsq_f64(float64x2_t __p0, float64x2_t __p1) {
59108   __ret = (float64x2_t) __builtin_neon_vrecpsq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
59112 __ai float64x2_t vrecpsq_f64(float64x2_t __p0, float64x2_t __p1) {
59114   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
59123 __ai float64x1_t vrecps_f64(float64x1_t __p0, float64x1_t __p1) {
59125   __ret = (float64x1_t) __builtin_neon_vrecps_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
59129 __ai float64x1_t vrecps_f64(float64x1_t __p0, float64x1_t __p1) {
59131   __ret = (float64x1_t) __builtin_neon_vrecps_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
59137 __ai float64_t vrecpsd_f64(float64_t __p0, float64_t __p1) {
59139   __ret = (float64_t) __builtin_neon_vrecpsd_f64(__p0, __p1);
59143 __ai float64_t vrecpsd_f64(float64_t __p0, float64_t __p1) {
59145   __ret = (float64_t) __builtin_neon_vrecpsd_f64(__p0, __p1);
59151 __ai float32_t vrecpss_f32(float32_t __p0, float32_t __p1) {
59153   __ret = (float32_t) __builtin_neon_vrecpss_f32(__p0, __p1);
59157 __ai float32_t vrecpss_f32(float32_t __p0, float32_t __p1) {
59159   __ret = (float32_t) __builtin_neon_vrecpss_f32(__p0, __p1);
59627 __ai uint64_t vrshld_u64(uint64_t __p0, uint64_t __p1) {
59629   __ret = (uint64_t) __builtin_neon_vrshld_u64(__p0, __p1);
59633 __ai uint64_t vrshld_u64(uint64_t __p0, uint64_t __p1) {
59635   __ret = (uint64_t) __builtin_neon_vrshld_u64(__p0, __p1);
59641 __ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
59643   __ret = (int64_t) __builtin_neon_vrshld_s64(__p0, __p1);
59647 __ai int64_t vrshld_s64(int64_t __p0, int64_t __p1) {
59649   __ret = (int64_t) __builtin_neon_vrshld_s64(__p0, __p1);
59655 #define vrshrd_n_u64(__p0, __p1) __extension__ ({ \
59658   __ret = (uint64_t) __builtin_neon_vrshrd_n_u64(__s0, __p1); \
59662 #define vrshrd_n_u64(__p0, __p1) __extension__ ({ \
59665   __ret = (uint64_t) __builtin_neon_vrshrd_n_u64(__s0, __p1); \
59671 #define vrshrd_n_s64(__p0, __p1) __extension__ ({ \
59674   __ret = (int64_t) __builtin_neon_vrshrd_n_s64(__s0, __p1); \
59678 #define vrshrd_n_s64(__p0, __p1) __extension__ ({ \
59681   __ret = (int64_t) __builtin_neon_vrshrd_n_s64(__s0, __p1); \
59871 __ai float64x2_t vrsqrtsq_f64(float64x2_t __p0, float64x2_t __p1) {
59873   __ret = (float64x2_t) __builtin_neon_vrsqrtsq_v((int8x16_t)__p0, (int8x16_t)__p1, 42);
59877 __ai float64x2_t vrsqrtsq_f64(float64x2_t __p0, float64x2_t __p1) {
59879   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
59888 __ai float64x1_t vrsqrts_f64(float64x1_t __p0, float64x1_t __p1) {
59890   __ret = (float64x1_t) __builtin_neon_vrsqrts_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
59894 __ai float64x1_t vrsqrts_f64(float64x1_t __p0, float64x1_t __p1) {
59896   __ret = (float64x1_t) __builtin_neon_vrsqrts_v((int8x8_t)__p0, (int8x8_t)__p1, 10);
59902 __ai float64_t vrsqrtsd_f64(float64_t __p0, float64_t __p1) {
59904   __ret = (float64_t) __builtin_neon_vrsqrtsd_f64(__p0, __p1);
59908 __ai float64_t vrsqrtsd_f64(float64_t __p0, float64_t __p1) {
59910   __ret = (float64_t) __builtin_neon_vrsqrtsd_f64(__p0, __p1);
59916 __ai float32_t vrsqrtss_f32(float32_t __p0, float32_t __p1) {
59918   __ret = (float32_t) __builtin_neon_vrsqrtss_f32(__p0, __p1);
59922 __ai float32_t vrsqrtss_f32(float32_t __p0, float32_t __p1) {
59924   __ret = (float32_t) __builtin_neon_vrsqrtss_f32(__p0, __p1);
59930 #define vrsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
59932   uint64_t __s1 = __p1; \
59938 #define vrsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
59940   uint64_t __s1 = __p1; \
59948 #define vrsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
59950   int64_t __s1 = __p1; \
59956 #define vrsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
59958   int64_t __s1 = __p1; \
59966 __ai uint16x8_t vrsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
59968   __ret = vcombine_u16(__p0, vrsubhn_u32(__p1, __p2));
59972 __ai uint16x8_t vrsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
59974   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
59984 __ai uint32x4_t vrsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
59986   __ret = vcombine_u32(__p0, vrsubhn_u64(__p1, __p2));
59990 __ai uint32x4_t vrsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
59992   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
60002 __ai uint8x16_t vrsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
60004   __ret = vcombine_u8(__p0, vrsubhn_u16(__p1, __p2));
60008 __ai uint8x16_t vrsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
60010   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
60020 __ai int16x8_t vrsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
60022   __ret = vcombine_s16(__p0, vrsubhn_s32(__p1, __p2));
60026 __ai int16x8_t vrsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
60028   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
60038 __ai int32x4_t vrsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
60040   __ret = vcombine_s32(__p0, vrsubhn_s64(__p1, __p2));
60044 __ai int32x4_t vrsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
60046   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
60056 __ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
60058   __ret = vcombine_s8(__p0, vrsubhn_s16(__p1, __p2));
60062 __ai int8x16_t vrsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
60064   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
60126 #define vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60128   poly64x1_t __s1 = __p1; \
60134 #define vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60136   poly64x1_t __s1 = __p1; \
60141 #define __noswap_vset_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60143   poly64x1_t __s1 = __p1; \
60151 #define vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60153   poly64x2_t __s1 = __p1; \
60159 #define vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60161   poly64x2_t __s1 = __p1; \
60168 #define __noswap_vsetq_lane_p64(__p0, __p1, __p2) __extension__ ({ \
60170   poly64x2_t __s1 = __p1; \
60178 #define vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60180   float64x2_t __s1 = __p1; \
60186 #define vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60188   float64x2_t __s1 = __p1; \
60195 #define __noswap_vsetq_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60197   float64x2_t __s1 = __p1; \
60205 #define vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60207   float64x1_t __s1 = __p1; \
60213 #define vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60215   float64x1_t __s1 = __p1; \
60220 #define __noswap_vset_lane_f64(__p0, __p1, __p2) __extension__ ({ \
60222   float64x1_t __s1 = __p1; \
60230 __ai uint64_t vshld_u64(uint64_t __p0, uint64_t __p1) {
60232   __ret = (uint64_t) __builtin_neon_vshld_u64(__p0, __p1);
60236 __ai uint64_t vshld_u64(uint64_t __p0, uint64_t __p1) {
60238   __ret = (uint64_t) __builtin_neon_vshld_u64(__p0, __p1);
60244 __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
60246   __ret = (int64_t) __builtin_neon_vshld_s64(__p0, __p1);
60250 __ai int64_t vshld_s64(int64_t __p0, int64_t __p1) {
60252   __ret = (int64_t) __builtin_neon_vshld_s64(__p0, __p1);
60258 #define vshld_n_u64(__p0, __p1) __extension__ ({ \
60261   __ret = (uint64_t) __builtin_neon_vshld_n_u64(__s0, __p1); \
60265 #define vshld_n_u64(__p0, __p1) __extension__ ({ \
60268   __ret = (uint64_t) __builtin_neon_vshld_n_u64(__s0, __p1); \
60274 #define vshld_n_s64(__p0, __p1) __extension__ ({ \
60277   __ret = (int64_t) __builtin_neon_vshld_n_s64(__s0, __p1); \
60281 #define vshld_n_s64(__p0, __p1) __extension__ ({ \
60284   __ret = (int64_t) __builtin_neon_vshld_n_s64(__s0, __p1); \
60398 #define vshrd_n_u64(__p0, __p1) __extension__ ({ \
60401   __ret = (uint64_t) __builtin_neon_vshrd_n_u64(__s0, __p1); \
60405 #define vshrd_n_u64(__p0, __p1) __extension__ ({ \
60408   __ret = (uint64_t) __builtin_neon_vshrd_n_u64(__s0, __p1); \
60414 #define vshrd_n_s64(__p0, __p1) __extension__ ({ \
60417   __ret = (int64_t) __builtin_neon_vshrd_n_s64(__s0, __p1); \
60421 #define vshrd_n_s64(__p0, __p1) __extension__ ({ \
60424   __ret = (int64_t) __builtin_neon_vshrd_n_s64(__s0, __p1); \
60556 #define vslid_n_u64(__p0, __p1, __p2) __extension__ ({ \
60558   uint64_t __s1 = __p1; \
60564 #define vslid_n_u64(__p0, __p1, __p2) __extension__ ({ \
60566   uint64_t __s1 = __p1; \
60574 #define vslid_n_s64(__p0, __p1, __p2) __extension__ ({ \
60576   int64_t __s1 = __p1; \
60582 #define vslid_n_s64(__p0, __p1, __p2) __extension__ ({ \
60584   int64_t __s1 = __p1; \
60592 #define vsli_n_p64(__p0, __p1, __p2) __extension__ ({ \
60594   poly64x1_t __s1 = __p1; \
60600 #define vsli_n_p64(__p0, __p1, __p2) __extension__ ({ \
60602   poly64x1_t __s1 = __p1; \
60610 #define vsliq_n_p64(__p0, __p1, __p2) __extension__ ({ \
60612   poly64x2_t __s1 = __p1; \
60618 #define vsliq_n_p64(__p0, __p1, __p2) __extension__ ({ \
60620   poly64x2_t __s1 = __p1; \
60631 __ai uint8_t vsqaddb_u8(uint8_t __p0, uint8_t __p1) {
60633   __ret = (uint8_t) __builtin_neon_vsqaddb_u8(__p0, __p1);
60637 __ai uint8_t vsqaddb_u8(uint8_t __p0, uint8_t __p1) {
60639   __ret = (uint8_t) __builtin_neon_vsqaddb_u8(__p0, __p1);
60645 __ai uint32_t vsqadds_u32(uint32_t __p0, uint32_t __p1) {
60647   __ret = (uint32_t) __builtin_neon_vsqadds_u32(__p0, __p1);
60651 __ai uint32_t vsqadds_u32(uint32_t __p0, uint32_t __p1) {
60653   __ret = (uint32_t) __builtin_neon_vsqadds_u32(__p0, __p1);
60659 __ai uint64_t vsqaddd_u64(uint64_t __p0, uint64_t __p1) {
60661   __ret = (uint64_t) __builtin_neon_vsqaddd_u64(__p0, __p1);
60665 __ai uint64_t vsqaddd_u64(uint64_t __p0, uint64_t __p1) {
60667   __ret = (uint64_t) __builtin_neon_vsqaddd_u64(__p0, __p1);
60673 __ai uint16_t vsqaddh_u16(uint16_t __p0, uint16_t __p1) {
60675   __ret = (uint16_t) __builtin_neon_vsqaddh_u16(__p0, __p1);
60679 __ai uint16_t vsqaddh_u16(uint16_t __p0, uint16_t __p1) {
60681   __ret = (uint16_t) __builtin_neon_vsqaddh_u16(__p0, __p1);
60687 __ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
60689   __ret = (uint8x16_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 48);
60693 __ai uint8x16_t vsqaddq_u8(uint8x16_t __p0, uint8x16_t __p1) {
60695   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
60704 __ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
60706   __ret = (uint32x4_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 50);
60710 __ai uint32x4_t vsqaddq_u32(uint32x4_t __p0, uint32x4_t __p1) {
60712   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
60721 __ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
60723   __ret = (uint64x2_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
60727 __ai uint64x2_t vsqaddq_u64(uint64x2_t __p0, uint64x2_t __p1) {
60729   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
60738 __ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
60740   __ret = (uint16x8_t) __builtin_neon_vsqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 49);
60744 __ai uint16x8_t vsqaddq_u16(uint16x8_t __p0, uint16x8_t __p1) {
60746   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
60755 __ai uint8x8_t vsqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
60757   __ret = (uint8x8_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 16);
60761 __ai uint8x8_t vsqadd_u8(uint8x8_t __p0, uint8x8_t __p1) {
60763   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
60772 __ai uint32x2_t vsqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
60774   __ret = (uint32x2_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 18);
60778 __ai uint32x2_t vsqadd_u32(uint32x2_t __p0, uint32x2_t __p1) {
60780   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
60789 __ai uint64x1_t vsqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
60791   __ret = (uint64x1_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
60795 __ai uint64x1_t vsqadd_u64(uint64x1_t __p0, uint64x1_t __p1) {
60797   __ret = (uint64x1_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
60803 __ai uint16x4_t vsqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
60805   __ret = (uint16x4_t) __builtin_neon_vsqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 17);
60809 __ai uint16x4_t vsqadd_u16(uint16x4_t __p0, uint16x4_t __p1) {
60811   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
60882 #define vsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
60884   uint64_t __s1 = __p1; \
60890 #define vsrad_n_u64(__p0, __p1, __p2) __extension__ ({ \
60892   uint64_t __s1 = __p1; \
60900 #define vsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
60902   int64_t __s1 = __p1; \
60908 #define vsrad_n_s64(__p0, __p1, __p2) __extension__ ({ \
60910   int64_t __s1 = __p1; \
60918 #define vsrid_n_u64(__p0, __p1, __p2) __extension__ ({ \
60920   uint64_t __s1 = __p1; \
60926 #define vsrid_n_u64(__p0, __p1, __p2) __extension__ ({ \
60928   uint64_t __s1 = __p1; \
60936 #define vsrid_n_s64(__p0, __p1, __p2) __extension__ ({ \
60938   int64_t __s1 = __p1; \
60944 #define vsrid_n_s64(__p0, __p1, __p2) __extension__ ({ \
60946   int64_t __s1 = __p1; \
60954 #define vsri_n_p64(__p0, __p1, __p2) __extension__ ({ \
60956   poly64x1_t __s1 = __p1; \
60962 #define vsri_n_p64(__p0, __p1, __p2) __extension__ ({ \
60964   poly64x1_t __s1 = __p1; \
60972 #define vsriq_n_p64(__p0, __p1, __p2) __extension__ ({ \
60974   poly64x2_t __s1 = __p1; \
60980 #define vsriq_n_p64(__p0, __p1, __p2) __extension__ ({ \
60982   poly64x2_t __s1 = __p1; \
60993 #define vst1_p64(__p0, __p1) __extension__ ({ \
60994   poly64x1_t __s1 = __p1; \
60998 #define vst1_p64(__p0, __p1) __extension__ ({ \
60999   poly64x1_t __s1 = __p1; \
61005 #define vst1q_p64(__p0, __p1) __extension__ ({ \
61006   poly64x2_t __s1 = __p1; \
61010 #define vst1q_p64(__p0, __p1) __extension__ ({ \
61011   poly64x2_t __s1 = __p1; \
61018 #define vst1q_f64(__p0, __p1) __extension__ ({ \
61019   float64x2_t __s1 = __p1; \
61023 #define vst1q_f64(__p0, __p1) __extension__ ({ \
61024   float64x2_t __s1 = __p1; \
61031 #define vst1_f64(__p0, __p1) __extension__ ({ \
61032   float64x1_t __s1 = __p1; \
61036 #define vst1_f64(__p0, __p1) __extension__ ({ \
61037   float64x1_t __s1 = __p1; \
61043 #define vst1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
61044   poly64x1_t __s1 = __p1; \
61048 #define vst1_lane_p64(__p0, __p1, __p2) __extension__ ({ \
61049   poly64x1_t __s1 = __p1; \
61055 #define vst1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
61056   poly64x2_t __s1 = __p1; \
61060 #define vst1q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
61061   poly64x2_t __s1 = __p1; \
61068 #define vst1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
61069   float64x2_t __s1 = __p1; \
61073 #define vst1q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
61074   float64x2_t __s1 = __p1; \
61081 #define vst1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
61082   float64x1_t __s1 = __p1; \
61086 #define vst1_lane_f64(__p0, __p1, __p2) __extension__ ({ \
61087   float64x1_t __s1 = __p1; \
61093 #define vst1_p8_x2(__p0, __p1) __extension__ ({ \
61094   poly8x8x2_t __s1 = __p1; \
61098 #define vst1_p8_x2(__p0, __p1) __extension__ ({ \
61099   poly8x8x2_t __s1 = __p1; \
61108 #define vst1_p64_x2(__p0, __p1) __extension__ ({ \
61109   poly64x1x2_t __s1 = __p1; \
61113 #define vst1_p64_x2(__p0, __p1) __extension__ ({ \
61114   poly64x1x2_t __s1 = __p1; \
61120 #define vst1_p16_x2(__p0, __p1) __extension__ ({ \
61121   poly16x4x2_t __s1 = __p1; \
61125 #define vst1_p16_x2(__p0, __p1) __extension__ ({ \
61126   poly16x4x2_t __s1 = __p1; \
61135 #define vst1q_p8_x2(__p0, __p1) __extension__ ({ \
61136   poly8x16x2_t __s1 = __p1; \
61140 #define vst1q_p8_x2(__p0, __p1) __extension__ ({ \
61141   poly8x16x2_t __s1 = __p1; \
61150 #define vst1q_p64_x2(__p0, __p1) __extension__ ({ \
61151   poly64x2x2_t __s1 = __p1; \
61155 #define vst1q_p64_x2(__p0, __p1) __extension__ ({ \
61156   poly64x2x2_t __s1 = __p1; \
61165 #define vst1q_p16_x2(__p0, __p1) __extension__ ({ \
61166   poly16x8x2_t __s1 = __p1; \
61170 #define vst1q_p16_x2(__p0, __p1) __extension__ ({ \
61171   poly16x8x2_t __s1 = __p1; \
61180 #define vst1q_u8_x2(__p0, __p1) __extension__ ({ \
61181   uint8x16x2_t __s1 = __p1; \
61185 #define vst1q_u8_x2(__p0, __p1) __extension__ ({ \
61186   uint8x16x2_t __s1 = __p1; \
61195 #define vst1q_u32_x2(__p0, __p1) __extension__ ({ \
61196   uint32x4x2_t __s1 = __p1; \
61200 #define vst1q_u32_x2(__p0, __p1) __extension__ ({ \
61201   uint32x4x2_t __s1 = __p1; \
61210 #define vst1q_u64_x2(__p0, __p1) __extension__ ({ \
61211   uint64x2x2_t __s1 = __p1; \
61215 #define vst1q_u64_x2(__p0, __p1) __extension__ ({ \
61216   uint64x2x2_t __s1 = __p1; \
61225 #define vst1q_u16_x2(__p0, __p1) __extension__ ({ \
61226   uint16x8x2_t __s1 = __p1; \
61230 #define vst1q_u16_x2(__p0, __p1) __extension__ ({ \
61231   uint16x8x2_t __s1 = __p1; \
61240 #define vst1q_s8_x2(__p0, __p1) __extension__ ({ \
61241   int8x16x2_t __s1 = __p1; \
61245 #define vst1q_s8_x2(__p0, __p1) __extension__ ({ \
61246   int8x16x2_t __s1 = __p1; \
61255 #define vst1q_f64_x2(__p0, __p1) __extension__ ({ \
61256   float64x2x2_t __s1 = __p1; \
61260 #define vst1q_f64_x2(__p0, __p1) __extension__ ({ \
61261   float64x2x2_t __s1 = __p1; \
61270 #define vst1q_f32_x2(__p0, __p1) __extension__ ({ \
61271   float32x4x2_t __s1 = __p1; \
61275 #define vst1q_f32_x2(__p0, __p1) __extension__ ({ \
61276   float32x4x2_t __s1 = __p1; \
61285 #define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
61286   float16x8x2_t __s1 = __p1; \
61290 #define vst1q_f16_x2(__p0, __p1) __extension__ ({ \
61291   float16x8x2_t __s1 = __p1; \
61300 #define vst1q_s32_x2(__p0, __p1) __extension__ ({ \
61301   int32x4x2_t __s1 = __p1; \
61305 #define vst1q_s32_x2(__p0, __p1) __extension__ ({ \
61306   int32x4x2_t __s1 = __p1; \
61315 #define vst1q_s64_x2(__p0, __p1) __extension__ ({ \
61316   int64x2x2_t __s1 = __p1; \
61320 #define vst1q_s64_x2(__p0, __p1) __extension__ ({ \
61321   int64x2x2_t __s1 = __p1; \
61330 #define vst1q_s16_x2(__p0, __p1) __extension__ ({ \
61331   int16x8x2_t __s1 = __p1; \
61335 #define vst1q_s16_x2(__p0, __p1) __extension__ ({ \
61336   int16x8x2_t __s1 = __p1; \
61345 #define vst1_u8_x2(__p0, __p1) __extension__ ({ \
61346   uint8x8x2_t __s1 = __p1; \
61350 #define vst1_u8_x2(__p0, __p1) __extension__ ({ \
61351   uint8x8x2_t __s1 = __p1; \
61360 #define vst1_u32_x2(__p0, __p1) __extension__ ({ \
61361   uint32x2x2_t __s1 = __p1; \
61365 #define vst1_u32_x2(__p0, __p1) __extension__ ({ \
61366   uint32x2x2_t __s1 = __p1; \
61375 #define vst1_u64_x2(__p0, __p1) __extension__ ({ \
61376   uint64x1x2_t __s1 = __p1; \
61380 #define vst1_u64_x2(__p0, __p1) __extension__ ({ \
61381   uint64x1x2_t __s1 = __p1; \
61387 #define vst1_u16_x2(__p0, __p1) __extension__ ({ \
61388   uint16x4x2_t __s1 = __p1; \
61392 #define vst1_u16_x2(__p0, __p1) __extension__ ({ \
61393   uint16x4x2_t __s1 = __p1; \
61402 #define vst1_s8_x2(__p0, __p1) __extension__ ({ \
61403   int8x8x2_t __s1 = __p1; \
61407 #define vst1_s8_x2(__p0, __p1) __extension__ ({ \
61408   int8x8x2_t __s1 = __p1; \
61417 #define vst1_f64_x2(__p0, __p1) __extension__ ({ \
61418   float64x1x2_t __s1 = __p1; \
61422 #define vst1_f64_x2(__p0, __p1) __extension__ ({ \
61423   float64x1x2_t __s1 = __p1; \
61429 #define vst1_f32_x2(__p0, __p1) __extension__ ({ \
61430   float32x2x2_t __s1 = __p1; \
61434 #define vst1_f32_x2(__p0, __p1) __extension__ ({ \
61435   float32x2x2_t __s1 = __p1; \
61444 #define vst1_f16_x2(__p0, __p1) __extension__ ({ \
61445   float16x4x2_t __s1 = __p1; \
61449 #define vst1_f16_x2(__p0, __p1) __extension__ ({ \
61450   float16x4x2_t __s1 = __p1; \
61459 #define vst1_s32_x2(__p0, __p1) __extension__ ({ \
61460   int32x2x2_t __s1 = __p1; \
61464 #define vst1_s32_x2(__p0, __p1) __extension__ ({ \
61465   int32x2x2_t __s1 = __p1; \
61474 #define vst1_s64_x2(__p0, __p1) __extension__ ({ \
61475   int64x1x2_t __s1 = __p1; \
61479 #define vst1_s64_x2(__p0, __p1) __extension__ ({ \
61480   int64x1x2_t __s1 = __p1; \
61486 #define vst1_s16_x2(__p0, __p1) __extension__ ({ \
61487   int16x4x2_t __s1 = __p1; \
61491 #define vst1_s16_x2(__p0, __p1) __extension__ ({ \
61492   int16x4x2_t __s1 = __p1; \
61501 #define vst1_p8_x3(__p0, __p1) __extension__ ({ \
61502   poly8x8x3_t __s1 = __p1; \
61506 #define vst1_p8_x3(__p0, __p1) __extension__ ({ \
61507   poly8x8x3_t __s1 = __p1; \
61517 #define vst1_p64_x3(__p0, __p1) __extension__ ({ \
61518   poly64x1x3_t __s1 = __p1; \
61522 #define vst1_p64_x3(__p0, __p1) __extension__ ({ \
61523   poly64x1x3_t __s1 = __p1; \
61529 #define vst1_p16_x3(__p0, __p1) __extension__ ({ \
61530   poly16x4x3_t __s1 = __p1; \
61534 #define vst1_p16_x3(__p0, __p1) __extension__ ({ \
61535   poly16x4x3_t __s1 = __p1; \
61545 #define vst1q_p8_x3(__p0, __p1) __extension__ ({ \
61546   poly8x16x3_t __s1 = __p1; \
61550 #define vst1q_p8_x3(__p0, __p1) __extension__ ({ \
61551   poly8x16x3_t __s1 = __p1; \
61561 #define vst1q_p64_x3(__p0, __p1) __extension__ ({ \
61562   poly64x2x3_t __s1 = __p1; \
61566 #define vst1q_p64_x3(__p0, __p1) __extension__ ({ \
61567   poly64x2x3_t __s1 = __p1; \
61577 #define vst1q_p16_x3(__p0, __p1) __extension__ ({ \
61578   poly16x8x3_t __s1 = __p1; \
61582 #define vst1q_p16_x3(__p0, __p1) __extension__ ({ \
61583   poly16x8x3_t __s1 = __p1; \
61593 #define vst1q_u8_x3(__p0, __p1) __extension__ ({ \
61594   uint8x16x3_t __s1 = __p1; \
61598 #define vst1q_u8_x3(__p0, __p1) __extension__ ({ \
61599   uint8x16x3_t __s1 = __p1; \
61609 #define vst1q_u32_x3(__p0, __p1) __extension__ ({ \
61610   uint32x4x3_t __s1 = __p1; \
61614 #define vst1q_u32_x3(__p0, __p1) __extension__ ({ \
61615   uint32x4x3_t __s1 = __p1; \
61625 #define vst1q_u64_x3(__p0, __p1) __extension__ ({ \
61626   uint64x2x3_t __s1 = __p1; \
61630 #define vst1q_u64_x3(__p0, __p1) __extension__ ({ \
61631   uint64x2x3_t __s1 = __p1; \
61641 #define vst1q_u16_x3(__p0, __p1) __extension__ ({ \
61642   uint16x8x3_t __s1 = __p1; \
61646 #define vst1q_u16_x3(__p0, __p1) __extension__ ({ \
61647   uint16x8x3_t __s1 = __p1; \
61657 #define vst1q_s8_x3(__p0, __p1) __extension__ ({ \
61658   int8x16x3_t __s1 = __p1; \
61662 #define vst1q_s8_x3(__p0, __p1) __extension__ ({ \
61663   int8x16x3_t __s1 = __p1; \
61673 #define vst1q_f64_x3(__p0, __p1) __extension__ ({ \
61674   float64x2x3_t __s1 = __p1; \
61678 #define vst1q_f64_x3(__p0, __p1) __extension__ ({ \
61679   float64x2x3_t __s1 = __p1; \
61689 #define vst1q_f32_x3(__p0, __p1) __extension__ ({ \
61690   float32x4x3_t __s1 = __p1; \
61694 #define vst1q_f32_x3(__p0, __p1) __extension__ ({ \
61695   float32x4x3_t __s1 = __p1; \
61705 #define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
61706   float16x8x3_t __s1 = __p1; \
61710 #define vst1q_f16_x3(__p0, __p1) __extension__ ({ \
61711   float16x8x3_t __s1 = __p1; \
61721 #define vst1q_s32_x3(__p0, __p1) __extension__ ({ \
61722   int32x4x3_t __s1 = __p1; \
61726 #define vst1q_s32_x3(__p0, __p1) __extension__ ({ \
61727   int32x4x3_t __s1 = __p1; \
61737 #define vst1q_s64_x3(__p0, __p1) __extension__ ({ \
61738   int64x2x3_t __s1 = __p1; \
61742 #define vst1q_s64_x3(__p0, __p1) __extension__ ({ \
61743   int64x2x3_t __s1 = __p1; \
61753 #define vst1q_s16_x3(__p0, __p1) __extension__ ({ \
61754   int16x8x3_t __s1 = __p1; \
61758 #define vst1q_s16_x3(__p0, __p1) __extension__ ({ \
61759   int16x8x3_t __s1 = __p1; \
61769 #define vst1_u8_x3(__p0, __p1) __extension__ ({ \
61770   uint8x8x3_t __s1 = __p1; \
61774 #define vst1_u8_x3(__p0, __p1) __extension__ ({ \
61775   uint8x8x3_t __s1 = __p1; \
61785 #define vst1_u32_x3(__p0, __p1) __extension__ ({ \
61786   uint32x2x3_t __s1 = __p1; \
61790 #define vst1_u32_x3(__p0, __p1) __extension__ ({ \
61791   uint32x2x3_t __s1 = __p1; \
61801 #define vst1_u64_x3(__p0, __p1) __extension__ ({ \
61802   uint64x1x3_t __s1 = __p1; \
61806 #define vst1_u64_x3(__p0, __p1) __extension__ ({ \
61807   uint64x1x3_t __s1 = __p1; \
61813 #define vst1_u16_x3(__p0, __p1) __extension__ ({ \
61814   uint16x4x3_t __s1 = __p1; \
61818 #define vst1_u16_x3(__p0, __p1) __extension__ ({ \
61819   uint16x4x3_t __s1 = __p1; \
61829 #define vst1_s8_x3(__p0, __p1) __extension__ ({ \
61830   int8x8x3_t __s1 = __p1; \
61834 #define vst1_s8_x3(__p0, __p1) __extension__ ({ \
61835   int8x8x3_t __s1 = __p1; \
61845 #define vst1_f64_x3(__p0, __p1) __extension__ ({ \
61846   float64x1x3_t __s1 = __p1; \
61850 #define vst1_f64_x3(__p0, __p1) __extension__ ({ \
61851   float64x1x3_t __s1 = __p1; \
61857 #define vst1_f32_x3(__p0, __p1) __extension__ ({ \
61858   float32x2x3_t __s1 = __p1; \
61862 #define vst1_f32_x3(__p0, __p1) __extension__ ({ \
61863   float32x2x3_t __s1 = __p1; \
61873 #define vst1_f16_x3(__p0, __p1) __extension__ ({ \
61874   float16x4x3_t __s1 = __p1; \
61878 #define vst1_f16_x3(__p0, __p1) __extension__ ({ \
61879   float16x4x3_t __s1 = __p1; \
61889 #define vst1_s32_x3(__p0, __p1) __extension__ ({ \
61890   int32x2x3_t __s1 = __p1; \
61894 #define vst1_s32_x3(__p0, __p1) __extension__ ({ \
61895   int32x2x3_t __s1 = __p1; \
61905 #define vst1_s64_x3(__p0, __p1) __extension__ ({ \
61906   int64x1x3_t __s1 = __p1; \
61910 #define vst1_s64_x3(__p0, __p1) __extension__ ({ \
61911   int64x1x3_t __s1 = __p1; \
61917 #define vst1_s16_x3(__p0, __p1) __extension__ ({ \
61918   int16x4x3_t __s1 = __p1; \
61922 #define vst1_s16_x3(__p0, __p1) __extension__ ({ \
61923   int16x4x3_t __s1 = __p1; \
61933 #define vst1_p8_x4(__p0, __p1) __extension__ ({ \
61934   poly8x8x4_t __s1 = __p1; \
61938 #define vst1_p8_x4(__p0, __p1) __extension__ ({ \
61939   poly8x8x4_t __s1 = __p1; \
61950 #define vst1_p64_x4(__p0, __p1) __extension__ ({ \
61951   poly64x1x4_t __s1 = __p1; \
61955 #define vst1_p64_x4(__p0, __p1) __extension__ ({ \
61956   poly64x1x4_t __s1 = __p1; \
61962 #define vst1_p16_x4(__p0, __p1) __extension__ ({ \
61963   poly16x4x4_t __s1 = __p1; \
61967 #define vst1_p16_x4(__p0, __p1) __extension__ ({ \
61968   poly16x4x4_t __s1 = __p1; \
61979 #define vst1q_p8_x4(__p0, __p1) __extension__ ({ \
61980   poly8x16x4_t __s1 = __p1; \
61984 #define vst1q_p8_x4(__p0, __p1) __extension__ ({ \
61985   poly8x16x4_t __s1 = __p1; \
61996 #define vst1q_p64_x4(__p0, __p1) __extension__ ({ \
61997   poly64x2x4_t __s1 = __p1; \
62001 #define vst1q_p64_x4(__p0, __p1) __extension__ ({ \
62002   poly64x2x4_t __s1 = __p1; \
62013 #define vst1q_p16_x4(__p0, __p1) __extension__ ({ \
62014   poly16x8x4_t __s1 = __p1; \
62018 #define vst1q_p16_x4(__p0, __p1) __extension__ ({ \
62019   poly16x8x4_t __s1 = __p1; \
62030 #define vst1q_u8_x4(__p0, __p1) __extension__ ({ \
62031   uint8x16x4_t __s1 = __p1; \
62035 #define vst1q_u8_x4(__p0, __p1) __extension__ ({ \
62036   uint8x16x4_t __s1 = __p1; \
62047 #define vst1q_u32_x4(__p0, __p1) __extension__ ({ \
62048   uint32x4x4_t __s1 = __p1; \
62052 #define vst1q_u32_x4(__p0, __p1) __extension__ ({ \
62053   uint32x4x4_t __s1 = __p1; \
62064 #define vst1q_u64_x4(__p0, __p1) __extension__ ({ \
62065   uint64x2x4_t __s1 = __p1; \
62069 #define vst1q_u64_x4(__p0, __p1) __extension__ ({ \
62070   uint64x2x4_t __s1 = __p1; \
62081 #define vst1q_u16_x4(__p0, __p1) __extension__ ({ \
62082   uint16x8x4_t __s1 = __p1; \
62086 #define vst1q_u16_x4(__p0, __p1) __extension__ ({ \
62087   uint16x8x4_t __s1 = __p1; \
62098 #define vst1q_s8_x4(__p0, __p1) __extension__ ({ \
62099   int8x16x4_t __s1 = __p1; \
62103 #define vst1q_s8_x4(__p0, __p1) __extension__ ({ \
62104   int8x16x4_t __s1 = __p1; \
62115 #define vst1q_f64_x4(__p0, __p1) __extension__ ({ \
62116   float64x2x4_t __s1 = __p1; \
62120 #define vst1q_f64_x4(__p0, __p1) __extension__ ({ \
62121   float64x2x4_t __s1 = __p1; \
62132 #define vst1q_f32_x4(__p0, __p1) __extension__ ({ \
62133   float32x4x4_t __s1 = __p1; \
62137 #define vst1q_f32_x4(__p0, __p1) __extension__ ({ \
62138   float32x4x4_t __s1 = __p1; \
62149 #define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
62150   float16x8x4_t __s1 = __p1; \
62154 #define vst1q_f16_x4(__p0, __p1) __extension__ ({ \
62155   float16x8x4_t __s1 = __p1; \
62166 #define vst1q_s32_x4(__p0, __p1) __extension__ ({ \
62167   int32x4x4_t __s1 = __p1; \
62171 #define vst1q_s32_x4(__p0, __p1) __extension__ ({ \
62172   int32x4x4_t __s1 = __p1; \
62183 #define vst1q_s64_x4(__p0, __p1) __extension__ ({ \
62184   int64x2x4_t __s1 = __p1; \
62188 #define vst1q_s64_x4(__p0, __p1) __extension__ ({ \
62189   int64x2x4_t __s1 = __p1; \
62200 #define vst1q_s16_x4(__p0, __p1) __extension__ ({ \
62201   int16x8x4_t __s1 = __p1; \
62205 #define vst1q_s16_x4(__p0, __p1) __extension__ ({ \
62206   int16x8x4_t __s1 = __p1; \
62217 #define vst1_u8_x4(__p0, __p1) __extension__ ({ \
62218   uint8x8x4_t __s1 = __p1; \
62222 #define vst1_u8_x4(__p0, __p1) __extension__ ({ \
62223   uint8x8x4_t __s1 = __p1; \
62234 #define vst1_u32_x4(__p0, __p1) __extension__ ({ \
62235   uint32x2x4_t __s1 = __p1; \
62239 #define vst1_u32_x4(__p0, __p1) __extension__ ({ \
62240   uint32x2x4_t __s1 = __p1; \
62251 #define vst1_u64_x4(__p0, __p1) __extension__ ({ \
62252   uint64x1x4_t __s1 = __p1; \
62256 #define vst1_u64_x4(__p0, __p1) __extension__ ({ \
62257   uint64x1x4_t __s1 = __p1; \
62263 #define vst1_u16_x4(__p0, __p1) __extension__ ({ \
62264   uint16x4x4_t __s1 = __p1; \
62268 #define vst1_u16_x4(__p0, __p1) __extension__ ({ \
62269   uint16x4x4_t __s1 = __p1; \
62280 #define vst1_s8_x4(__p0, __p1) __extension__ ({ \
62281   int8x8x4_t __s1 = __p1; \
62285 #define vst1_s8_x4(__p0, __p1) __extension__ ({ \
62286   int8x8x4_t __s1 = __p1; \
62297 #define vst1_f64_x4(__p0, __p1) __extension__ ({ \
62298   float64x1x4_t __s1 = __p1; \
62302 #define vst1_f64_x4(__p0, __p1) __extension__ ({ \
62303   float64x1x4_t __s1 = __p1; \
62309 #define vst1_f32_x4(__p0, __p1) __extension__ ({ \
62310   float32x2x4_t __s1 = __p1; \
62314 #define vst1_f32_x4(__p0, __p1) __extension__ ({ \
62315   float32x2x4_t __s1 = __p1; \
62326 #define vst1_f16_x4(__p0, __p1) __extension__ ({ \
62327   float16x4x4_t __s1 = __p1; \
62331 #define vst1_f16_x4(__p0, __p1) __extension__ ({ \
62332   float16x4x4_t __s1 = __p1; \
62343 #define vst1_s32_x4(__p0, __p1) __extension__ ({ \
62344   int32x2x4_t __s1 = __p1; \
62348 #define vst1_s32_x4(__p0, __p1) __extension__ ({ \
62349   int32x2x4_t __s1 = __p1; \
62360 #define vst1_s64_x4(__p0, __p1) __extension__ ({ \
62361   int64x1x4_t __s1 = __p1; \
62365 #define vst1_s64_x4(__p0, __p1) __extension__ ({ \
62366   int64x1x4_t __s1 = __p1; \
62372 #define vst1_s16_x4(__p0, __p1) __extension__ ({ \
62373   int16x4x4_t __s1 = __p1; \
62377 #define vst1_s16_x4(__p0, __p1) __extension__ ({ \
62378   int16x4x4_t __s1 = __p1; \
62389 #define vst2_p64(__p0, __p1) __extension__ ({ \
62390   poly64x1x2_t __s1 = __p1; \
62394 #define vst2_p64(__p0, __p1) __extension__ ({ \
62395   poly64x1x2_t __s1 = __p1; \
62401 #define vst2q_p64(__p0, __p1) __extension__ ({ \
62402   poly64x2x2_t __s1 = __p1; \
62406 #define vst2q_p64(__p0, __p1) __extension__ ({ \
62407   poly64x2x2_t __s1 = __p1; \
62416 #define vst2q_u64(__p0, __p1) __extension__ ({ \
62417   uint64x2x2_t __s1 = __p1; \
62421 #define vst2q_u64(__p0, __p1) __extension__ ({ \
62422   uint64x2x2_t __s1 = __p1; \
62431 #define vst2q_f64(__p0, __p1) __extension__ ({ \
62432   float64x2x2_t __s1 = __p1; \
62436 #define vst2q_f64(__p0, __p1) __extension__ ({ \
62437   float64x2x2_t __s1 = __p1; \
62446 #define vst2q_s64(__p0, __p1) __extension__ ({ \
62447   int64x2x2_t __s1 = __p1; \
62451 #define vst2q_s64(__p0, __p1) __extension__ ({ \
62452   int64x2x2_t __s1 = __p1; \
62461 #define vst2_f64(__p0, __p1) __extension__ ({ \
62462   float64x1x2_t __s1 = __p1; \
62466 #define vst2_f64(__p0, __p1) __extension__ ({ \
62467   float64x1x2_t __s1 = __p1; \
62473 #define vst2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62474   poly64x1x2_t __s1 = __p1; \
62478 #define vst2_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62479   poly64x1x2_t __s1 = __p1; \
62485 #define vst2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
62486   poly8x16x2_t __s1 = __p1; \
62490 #define vst2q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
62491   poly8x16x2_t __s1 = __p1; \
62500 #define vst2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62501   poly64x2x2_t __s1 = __p1; \
62505 #define vst2q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62506   poly64x2x2_t __s1 = __p1; \
62515 #define vst2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
62516   uint8x16x2_t __s1 = __p1; \
62520 #define vst2q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
62521   uint8x16x2_t __s1 = __p1; \
62530 #define vst2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
62531   uint64x2x2_t __s1 = __p1; \
62535 #define vst2q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
62536   uint64x2x2_t __s1 = __p1; \
62545 #define vst2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
62546   int8x16x2_t __s1 = __p1; \
62550 #define vst2q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
62551   int8x16x2_t __s1 = __p1; \
62560 #define vst2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62561   float64x2x2_t __s1 = __p1; \
62565 #define vst2q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62566   float64x2x2_t __s1 = __p1; \
62575 #define vst2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
62576   int64x2x2_t __s1 = __p1; \
62580 #define vst2q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
62581   int64x2x2_t __s1 = __p1; \
62590 #define vst2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
62591   uint64x1x2_t __s1 = __p1; \
62595 #define vst2_lane_u64(__p0, __p1, __p2) __extension__ ({ \
62596   uint64x1x2_t __s1 = __p1; \
62602 #define vst2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62603   float64x1x2_t __s1 = __p1; \
62607 #define vst2_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62608   float64x1x2_t __s1 = __p1; \
62614 #define vst2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
62615   int64x1x2_t __s1 = __p1; \
62619 #define vst2_lane_s64(__p0, __p1, __p2) __extension__ ({ \
62620   int64x1x2_t __s1 = __p1; \
62626 #define vst3_p64(__p0, __p1) __extension__ ({ \
62627   poly64x1x3_t __s1 = __p1; \
62631 #define vst3_p64(__p0, __p1) __extension__ ({ \
62632   poly64x1x3_t __s1 = __p1; \
62638 #define vst3q_p64(__p0, __p1) __extension__ ({ \
62639   poly64x2x3_t __s1 = __p1; \
62643 #define vst3q_p64(__p0, __p1) __extension__ ({ \
62644   poly64x2x3_t __s1 = __p1; \
62654 #define vst3q_u64(__p0, __p1) __extension__ ({ \
62655   uint64x2x3_t __s1 = __p1; \
62659 #define vst3q_u64(__p0, __p1) __extension__ ({ \
62660   uint64x2x3_t __s1 = __p1; \
62670 #define vst3q_f64(__p0, __p1) __extension__ ({ \
62671   float64x2x3_t __s1 = __p1; \
62675 #define vst3q_f64(__p0, __p1) __extension__ ({ \
62676   float64x2x3_t __s1 = __p1; \
62686 #define vst3q_s64(__p0, __p1) __extension__ ({ \
62687   int64x2x3_t __s1 = __p1; \
62691 #define vst3q_s64(__p0, __p1) __extension__ ({ \
62692   int64x2x3_t __s1 = __p1; \
62702 #define vst3_f64(__p0, __p1) __extension__ ({ \
62703   float64x1x3_t __s1 = __p1; \
62707 #define vst3_f64(__p0, __p1) __extension__ ({ \
62708   float64x1x3_t __s1 = __p1; \
62714 #define vst3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62715   poly64x1x3_t __s1 = __p1; \
62719 #define vst3_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62720   poly64x1x3_t __s1 = __p1; \
62726 #define vst3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
62727   poly8x16x3_t __s1 = __p1; \
62731 #define vst3q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
62732   poly8x16x3_t __s1 = __p1; \
62742 #define vst3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62743   poly64x2x3_t __s1 = __p1; \
62747 #define vst3q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62748   poly64x2x3_t __s1 = __p1; \
62758 #define vst3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
62759   uint8x16x3_t __s1 = __p1; \
62763 #define vst3q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
62764   uint8x16x3_t __s1 = __p1; \
62774 #define vst3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
62775   uint64x2x3_t __s1 = __p1; \
62779 #define vst3q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
62780   uint64x2x3_t __s1 = __p1; \
62790 #define vst3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
62791   int8x16x3_t __s1 = __p1; \
62795 #define vst3q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
62796   int8x16x3_t __s1 = __p1; \
62806 #define vst3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62807   float64x2x3_t __s1 = __p1; \
62811 #define vst3q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62812   float64x2x3_t __s1 = __p1; \
62822 #define vst3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
62823   int64x2x3_t __s1 = __p1; \
62827 #define vst3q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
62828   int64x2x3_t __s1 = __p1; \
62838 #define vst3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
62839   uint64x1x3_t __s1 = __p1; \
62843 #define vst3_lane_u64(__p0, __p1, __p2) __extension__ ({ \
62844   uint64x1x3_t __s1 = __p1; \
62850 #define vst3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62851   float64x1x3_t __s1 = __p1; \
62855 #define vst3_lane_f64(__p0, __p1, __p2) __extension__ ({ \
62856   float64x1x3_t __s1 = __p1; \
62862 #define vst3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
62863   int64x1x3_t __s1 = __p1; \
62867 #define vst3_lane_s64(__p0, __p1, __p2) __extension__ ({ \
62868   int64x1x3_t __s1 = __p1; \
62874 #define vst4_p64(__p0, __p1) __extension__ ({ \
62875   poly64x1x4_t __s1 = __p1; \
62879 #define vst4_p64(__p0, __p1) __extension__ ({ \
62880   poly64x1x4_t __s1 = __p1; \
62886 #define vst4q_p64(__p0, __p1) __extension__ ({ \
62887   poly64x2x4_t __s1 = __p1; \
62891 #define vst4q_p64(__p0, __p1) __extension__ ({ \
62892   poly64x2x4_t __s1 = __p1; \
62903 #define vst4q_u64(__p0, __p1) __extension__ ({ \
62904   uint64x2x4_t __s1 = __p1; \
62908 #define vst4q_u64(__p0, __p1) __extension__ ({ \
62909   uint64x2x4_t __s1 = __p1; \
62920 #define vst4q_f64(__p0, __p1) __extension__ ({ \
62921   float64x2x4_t __s1 = __p1; \
62925 #define vst4q_f64(__p0, __p1) __extension__ ({ \
62926   float64x2x4_t __s1 = __p1; \
62937 #define vst4q_s64(__p0, __p1) __extension__ ({ \
62938   int64x2x4_t __s1 = __p1; \
62942 #define vst4q_s64(__p0, __p1) __extension__ ({ \
62943   int64x2x4_t __s1 = __p1; \
62954 #define vst4_f64(__p0, __p1) __extension__ ({ \
62955   float64x1x4_t __s1 = __p1; \
62959 #define vst4_f64(__p0, __p1) __extension__ ({ \
62960   float64x1x4_t __s1 = __p1; \
62966 #define vst4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62967   poly64x1x4_t __s1 = __p1; \
62971 #define vst4_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62972   poly64x1x4_t __s1 = __p1; \
62978 #define vst4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
62979   poly8x16x4_t __s1 = __p1; \
62983 #define vst4q_lane_p8(__p0, __p1, __p2) __extension__ ({ \
62984   poly8x16x4_t __s1 = __p1; \
62995 #define vst4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
62996   poly64x2x4_t __s1 = __p1; \
63000 #define vst4q_lane_p64(__p0, __p1, __p2) __extension__ ({ \
63001   poly64x2x4_t __s1 = __p1; \
63012 #define vst4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63013   uint8x16x4_t __s1 = __p1; \
63017 #define vst4q_lane_u8(__p0, __p1, __p2) __extension__ ({ \
63018   uint8x16x4_t __s1 = __p1; \
63029 #define vst4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63030   uint64x2x4_t __s1 = __p1; \
63034 #define vst4q_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63035   uint64x2x4_t __s1 = __p1; \
63046 #define vst4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63047   int8x16x4_t __s1 = __p1; \
63051 #define vst4q_lane_s8(__p0, __p1, __p2) __extension__ ({ \
63052   int8x16x4_t __s1 = __p1; \
63063 #define vst4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63064   float64x2x4_t __s1 = __p1; \
63068 #define vst4q_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63069   float64x2x4_t __s1 = __p1; \
63080 #define vst4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63081   int64x2x4_t __s1 = __p1; \
63085 #define vst4q_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63086   int64x2x4_t __s1 = __p1; \
63097 #define vst4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63098   uint64x1x4_t __s1 = __p1; \
63102 #define vst4_lane_u64(__p0, __p1, __p2) __extension__ ({ \
63103   uint64x1x4_t __s1 = __p1; \
63109 #define vst4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63110   float64x1x4_t __s1 = __p1; \
63114 #define vst4_lane_f64(__p0, __p1, __p2) __extension__ ({ \
63115   float64x1x4_t __s1 = __p1; \
63121 #define vst4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63122   int64x1x4_t __s1 = __p1; \
63126 #define vst4_lane_s64(__p0, __p1, __p2) __extension__ ({ \
63127   int64x1x4_t __s1 = __p1; \
63133 #define vstrq_p128(__p0, __p1) __extension__ ({ \
63134   poly128_t __s1 = __p1; \
63138 #define vstrq_p128(__p0, __p1) __extension__ ({ \
63139   poly128_t __s1 = __p1; \
63145 __ai uint64_t vsubd_u64(uint64_t __p0, uint64_t __p1) {
63147   __ret = (uint64_t) __builtin_neon_vsubd_u64(__p0, __p1);
63151 __ai uint64_t vsubd_u64(uint64_t __p0, uint64_t __p1) {
63153   __ret = (uint64_t) __builtin_neon_vsubd_u64(__p0, __p1);
63159 __ai int64_t vsubd_s64(int64_t __p0, int64_t __p1) {
63161   __ret = (int64_t) __builtin_neon_vsubd_s64(__p0, __p1);
63165 __ai int64_t vsubd_s64(int64_t __p0, int64_t __p1) {
63167   __ret = (int64_t) __builtin_neon_vsubd_s64(__p0, __p1);
63173 __ai float64x2_t vsubq_f64(float64x2_t __p0, float64x2_t __p1) {
63175   __ret = __p0 - __p1;
63179 __ai float64x2_t vsubq_f64(float64x2_t __p0, float64x2_t __p1) {
63181   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63190 __ai float64x1_t vsub_f64(float64x1_t __p0, float64x1_t __p1) {
63192   __ret = __p0 - __p1;
63196 __ai float64x1_t vsub_f64(float64x1_t __p0, float64x1_t __p1) {
63198   __ret = __p0 - __p1;
63204 __ai uint16x8_t vsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
63206   __ret = vcombine_u16(__p0, vsubhn_u32(__p1, __p2));
63210 __ai uint16x8_t vsubhn_high_u32(uint16x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
63212   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63222 __ai uint32x4_t vsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
63224   __ret = vcombine_u32(__p0, vsubhn_u64(__p1, __p2));
63228 __ai uint32x4_t vsubhn_high_u64(uint32x2_t __p0, uint64x2_t __p1, uint64x2_t __p2) {
63230   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63240 __ai uint8x16_t vsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
63242   __ret = vcombine_u8(__p0, vsubhn_u16(__p1, __p2));
63246 __ai uint8x16_t vsubhn_high_u16(uint8x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
63248   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63258 __ai int16x8_t vsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
63260   __ret = vcombine_s16(__p0, vsubhn_s32(__p1, __p2));
63264 __ai int16x8_t vsubhn_high_s32(int16x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
63266   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63276 __ai int32x4_t vsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
63278   __ret = vcombine_s32(__p0, vsubhn_s64(__p1, __p2));
63282 __ai int32x4_t vsubhn_high_s64(int32x2_t __p0, int64x2_t __p1, int64x2_t __p2) {
63284   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63294 __ai int8x16_t vsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
63296   __ret = vcombine_s8(__p0, vsubhn_s16(__p1, __p2));
63300 __ai int8x16_t vsubhn_high_s16(int8x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
63302   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63312 __ai uint16x8_t vsubl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
63314   __ret = vmovl_high_u8(__p0) - vmovl_high_u8(__p1);
63318 __ai uint16x8_t vsubl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
63320   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63329 __ai uint64x2_t vsubl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
63331   __ret = vmovl_high_u32(__p0) - vmovl_high_u32(__p1);
63335 __ai uint64x2_t vsubl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
63337   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63346 __ai uint32x4_t vsubl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
63348   __ret = vmovl_high_u16(__p0) - vmovl_high_u16(__p1);
63352 __ai uint32x4_t vsubl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
63354   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63363 __ai int16x8_t vsubl_high_s8(int8x16_t __p0, int8x16_t __p1) {
63365   __ret = vmovl_high_s8(__p0) - vmovl_high_s8(__p1);
63369 __ai int16x8_t vsubl_high_s8(int8x16_t __p0, int8x16_t __p1) {
63371   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63380 __ai int64x2_t vsubl_high_s32(int32x4_t __p0, int32x4_t __p1) {
63382   __ret = vmovl_high_s32(__p0) - vmovl_high_s32(__p1);
63386 __ai int64x2_t vsubl_high_s32(int32x4_t __p0, int32x4_t __p1) {
63388   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63397 __ai int32x4_t vsubl_high_s16(int16x8_t __p0, int16x8_t __p1) {
63399   __ret = vmovl_high_s16(__p0) - vmovl_high_s16(__p1);
63403 __ai int32x4_t vsubl_high_s16(int16x8_t __p0, int16x8_t __p1) {
63405   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63414 __ai uint16x8_t vsubw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
63416   __ret = __p0 - vmovl_high_u8(__p1);
63420 __ai uint16x8_t vsubw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
63422   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63431 __ai uint64x2_t vsubw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
63433   __ret = __p0 - vmovl_high_u32(__p1);
63437 __ai uint64x2_t vsubw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
63439   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63448 __ai uint32x4_t vsubw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
63450   __ret = __p0 - vmovl_high_u16(__p1);
63454 __ai uint32x4_t vsubw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
63456   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63465 __ai int16x8_t vsubw_high_s8(int16x8_t __p0, int8x16_t __p1) {
63467   __ret = __p0 - vmovl_high_s8(__p1);
63471 __ai int16x8_t vsubw_high_s8(int16x8_t __p0, int8x16_t __p1) {
63473   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63482 __ai int64x2_t vsubw_high_s32(int64x2_t __p0, int32x4_t __p1) {
63484   __ret = __p0 - vmovl_high_s32(__p1);
63488 __ai int64x2_t vsubw_high_s32(int64x2_t __p0, int32x4_t __p1) {
63490   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63499 __ai int32x4_t vsubw_high_s16(int32x4_t __p0, int16x8_t __p1) {
63501   __ret = __p0 - vmovl_high_s16(__p1);
63505 __ai int32x4_t vsubw_high_s16(int32x4_t __p0, int16x8_t __p1) {
63507   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63516 __ai poly8x8_t vtrn1_p8(poly8x8_t __p0, poly8x8_t __p1) {
63518   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
63522 __ai poly8x8_t vtrn1_p8(poly8x8_t __p0, poly8x8_t __p1) {
63524   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63533 __ai poly16x4_t vtrn1_p16(poly16x4_t __p0, poly16x4_t __p1) {
63535   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
63539 __ai poly16x4_t vtrn1_p16(poly16x4_t __p0, poly16x4_t __p1) {
63541   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63550 __ai poly8x16_t vtrn1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
63552   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
63556 __ai poly8x16_t vtrn1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
63558   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63567 __ai poly64x2_t vtrn1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
63569   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
63573 __ai poly64x2_t vtrn1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
63575   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63584 __ai poly16x8_t vtrn1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
63586   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
63590 __ai poly16x8_t vtrn1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
63592   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63601 __ai uint8x16_t vtrn1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
63603   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
63607 __ai uint8x16_t vtrn1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
63609   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63618 __ai uint32x4_t vtrn1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
63620   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
63624 __ai uint32x4_t vtrn1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
63626   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63635 __ai uint64x2_t vtrn1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
63637   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
63641 __ai uint64x2_t vtrn1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
63643   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63652 __ai uint16x8_t vtrn1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
63654   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
63658 __ai uint16x8_t vtrn1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
63660   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63669 __ai int8x16_t vtrn1q_s8(int8x16_t __p0, int8x16_t __p1) {
63671   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 2, 18, 4, 20, 6, 22, 8, 24, 10, 26, 12, 28, 14, 30);
63675 __ai int8x16_t vtrn1q_s8(int8x16_t __p0, int8x16_t __p1) {
63677   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63686 __ai float64x2_t vtrn1q_f64(float64x2_t __p0, float64x2_t __p1) {
63688   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
63692 __ai float64x2_t vtrn1q_f64(float64x2_t __p0, float64x2_t __p1) {
63694   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63703 __ai float32x4_t vtrn1q_f32(float32x4_t __p0, float32x4_t __p1) {
63705   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
63709 __ai float32x4_t vtrn1q_f32(float32x4_t __p0, float32x4_t __p1) {
63711   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63720 __ai int32x4_t vtrn1q_s32(int32x4_t __p0, int32x4_t __p1) {
63722   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
63726 __ai int32x4_t vtrn1q_s32(int32x4_t __p0, int32x4_t __p1) {
63728   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63737 __ai int64x2_t vtrn1q_s64(int64x2_t __p0, int64x2_t __p1) {
63739   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
63743 __ai int64x2_t vtrn1q_s64(int64x2_t __p0, int64x2_t __p1) {
63745   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63754 __ai int16x8_t vtrn1q_s16(int16x8_t __p0, int16x8_t __p1) {
63756   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
63760 __ai int16x8_t vtrn1q_s16(int16x8_t __p0, int16x8_t __p1) {
63762   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63771 __ai uint8x8_t vtrn1_u8(uint8x8_t __p0, uint8x8_t __p1) {
63773   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
63777 __ai uint8x8_t vtrn1_u8(uint8x8_t __p0, uint8x8_t __p1) {
63779   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63788 __ai uint32x2_t vtrn1_u32(uint32x2_t __p0, uint32x2_t __p1) {
63790   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
63794 __ai uint32x2_t vtrn1_u32(uint32x2_t __p0, uint32x2_t __p1) {
63796   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63805 __ai uint16x4_t vtrn1_u16(uint16x4_t __p0, uint16x4_t __p1) {
63807   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
63811 __ai uint16x4_t vtrn1_u16(uint16x4_t __p0, uint16x4_t __p1) {
63813   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63822 __ai int8x8_t vtrn1_s8(int8x8_t __p0, int8x8_t __p1) {
63824   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 2, 10, 4, 12, 6, 14);
63828 __ai int8x8_t vtrn1_s8(int8x8_t __p0, int8x8_t __p1) {
63830   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63839 __ai float32x2_t vtrn1_f32(float32x2_t __p0, float32x2_t __p1) {
63841   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
63845 __ai float32x2_t vtrn1_f32(float32x2_t __p0, float32x2_t __p1) {
63847   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63856 __ai int32x2_t vtrn1_s32(int32x2_t __p0, int32x2_t __p1) {
63858   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
63862 __ai int32x2_t vtrn1_s32(int32x2_t __p0, int32x2_t __p1) {
63864   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63873 __ai int16x4_t vtrn1_s16(int16x4_t __p0, int16x4_t __p1) {
63875   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 2, 6);
63879 __ai int16x4_t vtrn1_s16(int16x4_t __p0, int16x4_t __p1) {
63881   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63890 __ai poly8x8_t vtrn2_p8(poly8x8_t __p0, poly8x8_t __p1) {
63892   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
63896 __ai poly8x8_t vtrn2_p8(poly8x8_t __p0, poly8x8_t __p1) {
63898   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63907 __ai poly16x4_t vtrn2_p16(poly16x4_t __p0, poly16x4_t __p1) {
63909   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
63913 __ai poly16x4_t vtrn2_p16(poly16x4_t __p0, poly16x4_t __p1) {
63915   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
63924 __ai poly8x16_t vtrn2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
63926   __ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
63930 __ai poly8x16_t vtrn2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
63932   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63941 __ai poly64x2_t vtrn2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
63943   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
63947 __ai poly64x2_t vtrn2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
63949   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
63958 __ai poly16x8_t vtrn2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
63960   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
63964 __ai poly16x8_t vtrn2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
63966   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
63975 __ai uint8x16_t vtrn2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
63977   __ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
63981 __ai uint8x16_t vtrn2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
63983   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
63992 __ai uint32x4_t vtrn2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
63994   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
63998 __ai uint32x4_t vtrn2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64000   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64009 __ai uint64x2_t vtrn2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64011   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64015 __ai uint64x2_t vtrn2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64017   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64026 __ai uint16x8_t vtrn2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64028   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64032 __ai uint16x8_t vtrn2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64034   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64043 __ai int8x16_t vtrn2q_s8(int8x16_t __p0, int8x16_t __p1) {
64045   __ret = __builtin_shufflevector(__p0, __p1, 1, 17, 3, 19, 5, 21, 7, 23, 9, 25, 11, 27, 13, 29, 15, 31);
64049 __ai int8x16_t vtrn2q_s8(int8x16_t __p0, int8x16_t __p1) {
64051   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64060 __ai float64x2_t vtrn2q_f64(float64x2_t __p0, float64x2_t __p1) {
64062   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64066 __ai float64x2_t vtrn2q_f64(float64x2_t __p0, float64x2_t __p1) {
64068   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64077 __ai float32x4_t vtrn2q_f32(float32x4_t __p0, float32x4_t __p1) {
64079   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64083 __ai float32x4_t vtrn2q_f32(float32x4_t __p0, float32x4_t __p1) {
64085   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64094 __ai int32x4_t vtrn2q_s32(int32x4_t __p0, int32x4_t __p1) {
64096   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64100 __ai int32x4_t vtrn2q_s32(int32x4_t __p0, int32x4_t __p1) {
64102   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64111 __ai int64x2_t vtrn2q_s64(int64x2_t __p0, int64x2_t __p1) {
64113   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64117 __ai int64x2_t vtrn2q_s64(int64x2_t __p0, int64x2_t __p1) {
64119   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64128 __ai int16x8_t vtrn2q_s16(int16x8_t __p0, int16x8_t __p1) {
64130   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64134 __ai int16x8_t vtrn2q_s16(int16x8_t __p0, int16x8_t __p1) {
64136   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64145 __ai uint8x8_t vtrn2_u8(uint8x8_t __p0, uint8x8_t __p1) {
64147   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64151 __ai uint8x8_t vtrn2_u8(uint8x8_t __p0, uint8x8_t __p1) {
64153   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64162 __ai uint32x2_t vtrn2_u32(uint32x2_t __p0, uint32x2_t __p1) {
64164   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64168 __ai uint32x2_t vtrn2_u32(uint32x2_t __p0, uint32x2_t __p1) {
64170   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64179 __ai uint16x4_t vtrn2_u16(uint16x4_t __p0, uint16x4_t __p1) {
64181   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64185 __ai uint16x4_t vtrn2_u16(uint16x4_t __p0, uint16x4_t __p1) {
64187   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64196 __ai int8x8_t vtrn2_s8(int8x8_t __p0, int8x8_t __p1) {
64198   __ret = __builtin_shufflevector(__p0, __p1, 1, 9, 3, 11, 5, 13, 7, 15);
64202 __ai int8x8_t vtrn2_s8(int8x8_t __p0, int8x8_t __p1) {
64204   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64213 __ai float32x2_t vtrn2_f32(float32x2_t __p0, float32x2_t __p1) {
64215   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64219 __ai float32x2_t vtrn2_f32(float32x2_t __p0, float32x2_t __p1) {
64221   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64230 __ai int32x2_t vtrn2_s32(int32x2_t __p0, int32x2_t __p1) {
64232   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
64236 __ai int32x2_t vtrn2_s32(int32x2_t __p0, int32x2_t __p1) {
64238   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64247 __ai int16x4_t vtrn2_s16(int16x4_t __p0, int16x4_t __p1) {
64249   __ret = __builtin_shufflevector(__p0, __p1, 1, 5, 3, 7);
64253 __ai int16x4_t vtrn2_s16(int16x4_t __p0, int16x4_t __p1) {
64255   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64264 __ai uint64x1_t vtst_p64(poly64x1_t __p0, poly64x1_t __p1) {
64266   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64270 __ai uint64x1_t vtst_p64(poly64x1_t __p0, poly64x1_t __p1) {
64272   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64278 __ai uint64x2_t vtstq_p64(poly64x2_t __p0, poly64x2_t __p1) {
64280   __ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
64284 __ai uint64x2_t vtstq_p64(poly64x2_t __p0, poly64x2_t __p1) {
64286   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64295 __ai uint64x2_t vtstq_u64(uint64x2_t __p0, uint64x2_t __p1) {
64297   __ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
64301 __ai uint64x2_t vtstq_u64(uint64x2_t __p0, uint64x2_t __p1) {
64303   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64312 __ai uint64x2_t vtstq_s64(int64x2_t __p0, int64x2_t __p1) {
64314   __ret = (uint64x2_t) __builtin_neon_vtstq_v((int8x16_t)__p0, (int8x16_t)__p1, 51);
64318 __ai uint64x2_t vtstq_s64(int64x2_t __p0, int64x2_t __p1) {
64320   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64329 __ai uint64x1_t vtst_u64(uint64x1_t __p0, uint64x1_t __p1) {
64331   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64335 __ai uint64x1_t vtst_u64(uint64x1_t __p0, uint64x1_t __p1) {
64337   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64343 __ai uint64x1_t vtst_s64(int64x1_t __p0, int64x1_t __p1) {
64345   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64349 __ai uint64x1_t vtst_s64(int64x1_t __p0, int64x1_t __p1) {
64351   __ret = (uint64x1_t) __builtin_neon_vtst_v((int8x8_t)__p0, (int8x8_t)__p1, 19);
64357 __ai uint64_t vtstd_u64(uint64_t __p0, uint64_t __p1) {
64359   __ret = (uint64_t) __builtin_neon_vtstd_u64(__p0, __p1);
64363 __ai uint64_t vtstd_u64(uint64_t __p0, uint64_t __p1) {
64365   __ret = (uint64_t) __builtin_neon_vtstd_u64(__p0, __p1);
64371 __ai int64_t vtstd_s64(int64_t __p0, int64_t __p1) {
64373   __ret = (int64_t) __builtin_neon_vtstd_s64(__p0, __p1);
64377 __ai int64_t vtstd_s64(int64_t __p0, int64_t __p1) {
64379   __ret = (int64_t) __builtin_neon_vtstd_s64(__p0, __p1);
64385 __ai int8_t vuqaddb_s8(int8_t __p0, int8_t __p1) {
64387   __ret = (int8_t) __builtin_neon_vuqaddb_s8(__p0, __p1);
64391 __ai int8_t vuqaddb_s8(int8_t __p0, int8_t __p1) {
64393   __ret = (int8_t) __builtin_neon_vuqaddb_s8(__p0, __p1);
64399 __ai int32_t vuqadds_s32(int32_t __p0, int32_t __p1) {
64401   __ret = (int32_t) __builtin_neon_vuqadds_s32(__p0, __p1);
64405 __ai int32_t vuqadds_s32(int32_t __p0, int32_t __p1) {
64407   __ret = (int32_t) __builtin_neon_vuqadds_s32(__p0, __p1);
64413 __ai int64_t vuqaddd_s64(int64_t __p0, int64_t __p1) {
64415   __ret = (int64_t) __builtin_neon_vuqaddd_s64(__p0, __p1);
64419 __ai int64_t vuqaddd_s64(int64_t __p0, int64_t __p1) {
64421   __ret = (int64_t) __builtin_neon_vuqaddd_s64(__p0, __p1);
64427 __ai int16_t vuqaddh_s16(int16_t __p0, int16_t __p1) {
64429   __ret = (int16_t) __builtin_neon_vuqaddh_s16(__p0, __p1);
64433 __ai int16_t vuqaddh_s16(int16_t __p0, int16_t __p1) {
64435   __ret = (int16_t) __builtin_neon_vuqaddh_s16(__p0, __p1);
64441 __ai int8x16_t vuqaddq_s8(int8x16_t __p0, int8x16_t __p1) {
64443   __ret = (int8x16_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 32);
64447 __ai int8x16_t vuqaddq_s8(int8x16_t __p0, int8x16_t __p1) {
64449   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64458 __ai int32x4_t vuqaddq_s32(int32x4_t __p0, int32x4_t __p1) {
64460   __ret = (int32x4_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 34);
64464 __ai int32x4_t vuqaddq_s32(int32x4_t __p0, int32x4_t __p1) {
64466   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64475 __ai int64x2_t vuqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
64477   __ret = (int64x2_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 35);
64481 __ai int64x2_t vuqaddq_s64(int64x2_t __p0, int64x2_t __p1) {
64483   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64492 __ai int16x8_t vuqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
64494   __ret = (int16x8_t) __builtin_neon_vuqaddq_v((int8x16_t)__p0, (int8x16_t)__p1, 33);
64498 __ai int16x8_t vuqaddq_s16(int16x8_t __p0, int16x8_t __p1) {
64500   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64509 __ai int8x8_t vuqadd_s8(int8x8_t __p0, int8x8_t __p1) {
64511   __ret = (int8x8_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 0);
64515 __ai int8x8_t vuqadd_s8(int8x8_t __p0, int8x8_t __p1) {
64517   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64526 __ai int32x2_t vuqadd_s32(int32x2_t __p0, int32x2_t __p1) {
64528   __ret = (int32x2_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 2);
64532 __ai int32x2_t vuqadd_s32(int32x2_t __p0, int32x2_t __p1) {
64534   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64543 __ai int64x1_t vuqadd_s64(int64x1_t __p0, int64x1_t __p1) {
64545   __ret = (int64x1_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
64549 __ai int64x1_t vuqadd_s64(int64x1_t __p0, int64x1_t __p1) {
64551   __ret = (int64x1_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 3);
64557 __ai int16x4_t vuqadd_s16(int16x4_t __p0, int16x4_t __p1) {
64559   __ret = (int16x4_t) __builtin_neon_vuqadd_v((int8x8_t)__p0, (int8x8_t)__p1, 1);
64563 __ai int16x4_t vuqadd_s16(int16x4_t __p0, int16x4_t __p1) {
64565   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64574 __ai poly8x8_t vuzp1_p8(poly8x8_t __p0, poly8x8_t __p1) {
64576   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
64580 __ai poly8x8_t vuzp1_p8(poly8x8_t __p0, poly8x8_t __p1) {
64582   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64591 __ai poly16x4_t vuzp1_p16(poly16x4_t __p0, poly16x4_t __p1) {
64593   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
64597 __ai poly16x4_t vuzp1_p16(poly16x4_t __p0, poly16x4_t __p1) {
64599   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64608 __ai poly8x16_t vuzp1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64610   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
64614 __ai poly8x16_t vuzp1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64616   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64625 __ai poly64x2_t vuzp1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64627   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64631 __ai poly64x2_t vuzp1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
64633   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64642 __ai poly16x8_t vuzp1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64644   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
64648 __ai poly16x8_t vuzp1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
64650   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64659 __ai uint8x16_t vuzp1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64661   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
64665 __ai uint8x16_t vuzp1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
64667   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64676 __ai uint32x4_t vuzp1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64678   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
64682 __ai uint32x4_t vuzp1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
64684   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64693 __ai uint64x2_t vuzp1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64695   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64699 __ai uint64x2_t vuzp1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
64701   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64710 __ai uint16x8_t vuzp1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64712   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
64716 __ai uint16x8_t vuzp1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
64718   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64727 __ai int8x16_t vuzp1q_s8(int8x16_t __p0, int8x16_t __p1) {
64729   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30);
64733 __ai int8x16_t vuzp1q_s8(int8x16_t __p0, int8x16_t __p1) {
64735   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64744 __ai float64x2_t vuzp1q_f64(float64x2_t __p0, float64x2_t __p1) {
64746   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64750 __ai float64x2_t vuzp1q_f64(float64x2_t __p0, float64x2_t __p1) {
64752   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64761 __ai float32x4_t vuzp1q_f32(float32x4_t __p0, float32x4_t __p1) {
64763   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
64767 __ai float32x4_t vuzp1q_f32(float32x4_t __p0, float32x4_t __p1) {
64769   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64778 __ai int32x4_t vuzp1q_s32(int32x4_t __p0, int32x4_t __p1) {
64780   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
64784 __ai int32x4_t vuzp1q_s32(int32x4_t __p0, int32x4_t __p1) {
64786   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64795 __ai int64x2_t vuzp1q_s64(int64x2_t __p0, int64x2_t __p1) {
64797   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64801 __ai int64x2_t vuzp1q_s64(int64x2_t __p0, int64x2_t __p1) {
64803   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64812 __ai int16x8_t vuzp1q_s16(int16x8_t __p0, int16x8_t __p1) {
64814   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
64818 __ai int16x8_t vuzp1q_s16(int16x8_t __p0, int16x8_t __p1) {
64820   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64829 __ai uint8x8_t vuzp1_u8(uint8x8_t __p0, uint8x8_t __p1) {
64831   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
64835 __ai uint8x8_t vuzp1_u8(uint8x8_t __p0, uint8x8_t __p1) {
64837   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64846 __ai uint32x2_t vuzp1_u32(uint32x2_t __p0, uint32x2_t __p1) {
64848   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64852 __ai uint32x2_t vuzp1_u32(uint32x2_t __p0, uint32x2_t __p1) {
64854   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64863 __ai uint16x4_t vuzp1_u16(uint16x4_t __p0, uint16x4_t __p1) {
64865   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
64869 __ai uint16x4_t vuzp1_u16(uint16x4_t __p0, uint16x4_t __p1) {
64871   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64880 __ai int8x8_t vuzp1_s8(int8x8_t __p0, int8x8_t __p1) {
64882   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6, 8, 10, 12, 14);
64886 __ai int8x8_t vuzp1_s8(int8x8_t __p0, int8x8_t __p1) {
64888   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64897 __ai float32x2_t vuzp1_f32(float32x2_t __p0, float32x2_t __p1) {
64899   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64903 __ai float32x2_t vuzp1_f32(float32x2_t __p0, float32x2_t __p1) {
64905   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64914 __ai int32x2_t vuzp1_s32(int32x2_t __p0, int32x2_t __p1) {
64916   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
64920 __ai int32x2_t vuzp1_s32(int32x2_t __p0, int32x2_t __p1) {
64922   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
64931 __ai int16x4_t vuzp1_s16(int16x4_t __p0, int16x4_t __p1) {
64933   __ret = __builtin_shufflevector(__p0, __p1, 0, 2, 4, 6);
64937 __ai int16x4_t vuzp1_s16(int16x4_t __p0, int16x4_t __p1) {
64939   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64948 __ai poly8x8_t vuzp2_p8(poly8x8_t __p0, poly8x8_t __p1) {
64950   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
64954 __ai poly8x8_t vuzp2_p8(poly8x8_t __p0, poly8x8_t __p1) {
64956   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
64965 __ai poly16x4_t vuzp2_p16(poly16x4_t __p0, poly16x4_t __p1) {
64967   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
64971 __ai poly16x4_t vuzp2_p16(poly16x4_t __p0, poly16x4_t __p1) {
64973   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
64982 __ai poly8x16_t vuzp2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64984   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
64988 __ai poly8x16_t vuzp2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
64990   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
64999 __ai poly64x2_t vuzp2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65001   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65005 __ai poly64x2_t vuzp2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65007   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65016 __ai poly16x8_t vuzp2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65018   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65022 __ai poly16x8_t vuzp2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65024   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65033 __ai uint8x16_t vuzp2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65035   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
65039 __ai uint8x16_t vuzp2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65041   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65050 __ai uint32x4_t vuzp2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65052   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65056 __ai uint32x4_t vuzp2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65058   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65067 __ai uint64x2_t vuzp2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65069   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65073 __ai uint64x2_t vuzp2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65075   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65084 __ai uint16x8_t vuzp2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65086   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65090 __ai uint16x8_t vuzp2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65092   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65101 __ai int8x16_t vuzp2q_s8(int8x16_t __p0, int8x16_t __p1) {
65103   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31);
65107 __ai int8x16_t vuzp2q_s8(int8x16_t __p0, int8x16_t __p1) {
65109   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65118 __ai float64x2_t vuzp2q_f64(float64x2_t __p0, float64x2_t __p1) {
65120   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65124 __ai float64x2_t vuzp2q_f64(float64x2_t __p0, float64x2_t __p1) {
65126   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65135 __ai float32x4_t vuzp2q_f32(float32x4_t __p0, float32x4_t __p1) {
65137   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65141 __ai float32x4_t vuzp2q_f32(float32x4_t __p0, float32x4_t __p1) {
65143   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65152 __ai int32x4_t vuzp2q_s32(int32x4_t __p0, int32x4_t __p1) {
65154   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65158 __ai int32x4_t vuzp2q_s32(int32x4_t __p0, int32x4_t __p1) {
65160   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65169 __ai int64x2_t vuzp2q_s64(int64x2_t __p0, int64x2_t __p1) {
65171   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65175 __ai int64x2_t vuzp2q_s64(int64x2_t __p0, int64x2_t __p1) {
65177   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65186 __ai int16x8_t vuzp2q_s16(int16x8_t __p0, int16x8_t __p1) {
65188   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65192 __ai int16x8_t vuzp2q_s16(int16x8_t __p0, int16x8_t __p1) {
65194   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65203 __ai uint8x8_t vuzp2_u8(uint8x8_t __p0, uint8x8_t __p1) {
65205   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65209 __ai uint8x8_t vuzp2_u8(uint8x8_t __p0, uint8x8_t __p1) {
65211   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65220 __ai uint32x2_t vuzp2_u32(uint32x2_t __p0, uint32x2_t __p1) {
65222   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65226 __ai uint32x2_t vuzp2_u32(uint32x2_t __p0, uint32x2_t __p1) {
65228   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65237 __ai uint16x4_t vuzp2_u16(uint16x4_t __p0, uint16x4_t __p1) {
65239   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65243 __ai uint16x4_t vuzp2_u16(uint16x4_t __p0, uint16x4_t __p1) {
65245   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65254 __ai int8x8_t vuzp2_s8(int8x8_t __p0, int8x8_t __p1) {
65256   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7, 9, 11, 13, 15);
65260 __ai int8x8_t vuzp2_s8(int8x8_t __p0, int8x8_t __p1) {
65262   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65271 __ai float32x2_t vuzp2_f32(float32x2_t __p0, float32x2_t __p1) {
65273   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65277 __ai float32x2_t vuzp2_f32(float32x2_t __p0, float32x2_t __p1) {
65279   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65288 __ai int32x2_t vuzp2_s32(int32x2_t __p0, int32x2_t __p1) {
65290   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65294 __ai int32x2_t vuzp2_s32(int32x2_t __p0, int32x2_t __p1) {
65296   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65305 __ai int16x4_t vuzp2_s16(int16x4_t __p0, int16x4_t __p1) {
65307   __ret = __builtin_shufflevector(__p0, __p1, 1, 3, 5, 7);
65311 __ai int16x4_t vuzp2_s16(int16x4_t __p0, int16x4_t __p1) {
65313   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65322 __ai poly8x8_t vzip1_p8(poly8x8_t __p0, poly8x8_t __p1) {
65324   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
65328 __ai poly8x8_t vzip1_p8(poly8x8_t __p0, poly8x8_t __p1) {
65330   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65339 __ai poly16x4_t vzip1_p16(poly16x4_t __p0, poly16x4_t __p1) {
65341   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
65345 __ai poly16x4_t vzip1_p16(poly16x4_t __p0, poly16x4_t __p1) {
65347   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65356 __ai poly8x16_t vzip1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65358   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
65362 __ai poly8x16_t vzip1q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65364   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65373 __ai poly64x2_t vzip1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65375   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65379 __ai poly64x2_t vzip1q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65381   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65390 __ai poly16x8_t vzip1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65392   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
65396 __ai poly16x8_t vzip1q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65398   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65407 __ai uint8x16_t vzip1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65409   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
65413 __ai uint8x16_t vzip1q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65415   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65424 __ai uint32x4_t vzip1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65426   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
65430 __ai uint32x4_t vzip1q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65432   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65441 __ai uint64x2_t vzip1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65443   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65447 __ai uint64x2_t vzip1q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65449   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65458 __ai uint16x8_t vzip1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65460   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
65464 __ai uint16x8_t vzip1q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65466   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65475 __ai int8x16_t vzip1q_s8(int8x16_t __p0, int8x16_t __p1) {
65477   __ret = __builtin_shufflevector(__p0, __p1, 0, 16, 1, 17, 2, 18, 3, 19, 4, 20, 5, 21, 6, 22, 7, 23);
65481 __ai int8x16_t vzip1q_s8(int8x16_t __p0, int8x16_t __p1) {
65483   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65492 __ai float64x2_t vzip1q_f64(float64x2_t __p0, float64x2_t __p1) {
65494   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65498 __ai float64x2_t vzip1q_f64(float64x2_t __p0, float64x2_t __p1) {
65500   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65509 __ai float32x4_t vzip1q_f32(float32x4_t __p0, float32x4_t __p1) {
65511   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
65515 __ai float32x4_t vzip1q_f32(float32x4_t __p0, float32x4_t __p1) {
65517   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65526 __ai int32x4_t vzip1q_s32(int32x4_t __p0, int32x4_t __p1) {
65528   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
65532 __ai int32x4_t vzip1q_s32(int32x4_t __p0, int32x4_t __p1) {
65534   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65543 __ai int64x2_t vzip1q_s64(int64x2_t __p0, int64x2_t __p1) {
65545   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65549 __ai int64x2_t vzip1q_s64(int64x2_t __p0, int64x2_t __p1) {
65551   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65560 __ai int16x8_t vzip1q_s16(int16x8_t __p0, int16x8_t __p1) {
65562   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
65566 __ai int16x8_t vzip1q_s16(int16x8_t __p0, int16x8_t __p1) {
65568   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65577 __ai uint8x8_t vzip1_u8(uint8x8_t __p0, uint8x8_t __p1) {
65579   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
65583 __ai uint8x8_t vzip1_u8(uint8x8_t __p0, uint8x8_t __p1) {
65585   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65594 __ai uint32x2_t vzip1_u32(uint32x2_t __p0, uint32x2_t __p1) {
65596   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65600 __ai uint32x2_t vzip1_u32(uint32x2_t __p0, uint32x2_t __p1) {
65602   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65611 __ai uint16x4_t vzip1_u16(uint16x4_t __p0, uint16x4_t __p1) {
65613   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
65617 __ai uint16x4_t vzip1_u16(uint16x4_t __p0, uint16x4_t __p1) {
65619   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65628 __ai int8x8_t vzip1_s8(int8x8_t __p0, int8x8_t __p1) {
65630   __ret = __builtin_shufflevector(__p0, __p1, 0, 8, 1, 9, 2, 10, 3, 11);
65634 __ai int8x8_t vzip1_s8(int8x8_t __p0, int8x8_t __p1) {
65636   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65645 __ai float32x2_t vzip1_f32(float32x2_t __p0, float32x2_t __p1) {
65647   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65651 __ai float32x2_t vzip1_f32(float32x2_t __p0, float32x2_t __p1) {
65653   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65662 __ai int32x2_t vzip1_s32(int32x2_t __p0, int32x2_t __p1) {
65664   __ret = __builtin_shufflevector(__p0, __p1, 0, 2);
65668 __ai int32x2_t vzip1_s32(int32x2_t __p0, int32x2_t __p1) {
65670   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65679 __ai int16x4_t vzip1_s16(int16x4_t __p0, int16x4_t __p1) {
65681   __ret = __builtin_shufflevector(__p0, __p1, 0, 4, 1, 5);
65685 __ai int16x4_t vzip1_s16(int16x4_t __p0, int16x4_t __p1) {
65687   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65696 __ai poly8x8_t vzip2_p8(poly8x8_t __p0, poly8x8_t __p1) {
65698   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
65702 __ai poly8x8_t vzip2_p8(poly8x8_t __p0, poly8x8_t __p1) {
65704   poly8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65713 __ai poly16x4_t vzip2_p16(poly16x4_t __p0, poly16x4_t __p1) {
65715   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
65719 __ai poly16x4_t vzip2_p16(poly16x4_t __p0, poly16x4_t __p1) {
65721   poly16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65730 __ai poly8x16_t vzip2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65732   __ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
65736 __ai poly8x16_t vzip2q_p8(poly8x16_t __p0, poly8x16_t __p1) {
65738   poly8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65747 __ai poly64x2_t vzip2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65749   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65753 __ai poly64x2_t vzip2q_p64(poly64x2_t __p0, poly64x2_t __p1) {
65755   poly64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65764 __ai poly16x8_t vzip2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65766   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
65770 __ai poly16x8_t vzip2q_p16(poly16x8_t __p0, poly16x8_t __p1) {
65772   poly16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65781 __ai uint8x16_t vzip2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65783   __ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
65787 __ai uint8x16_t vzip2q_u8(uint8x16_t __p0, uint8x16_t __p1) {
65789   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65798 __ai uint32x4_t vzip2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65800   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
65804 __ai uint32x4_t vzip2q_u32(uint32x4_t __p0, uint32x4_t __p1) {
65806   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65815 __ai uint64x2_t vzip2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65817   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65821 __ai uint64x2_t vzip2q_u64(uint64x2_t __p0, uint64x2_t __p1) {
65823   uint64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65832 __ai uint16x8_t vzip2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65834   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
65838 __ai uint16x8_t vzip2q_u16(uint16x8_t __p0, uint16x8_t __p1) {
65840   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65849 __ai int8x16_t vzip2q_s8(int8x16_t __p0, int8x16_t __p1) {
65851   __ret = __builtin_shufflevector(__p0, __p1, 8, 24, 9, 25, 10, 26, 11, 27, 12, 28, 13, 29, 14, 30, 15, 31);
65855 __ai int8x16_t vzip2q_s8(int8x16_t __p0, int8x16_t __p1) {
65857   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
65866 __ai float64x2_t vzip2q_f64(float64x2_t __p0, float64x2_t __p1) {
65868   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65872 __ai float64x2_t vzip2q_f64(float64x2_t __p0, float64x2_t __p1) {
65874   float64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65883 __ai float32x4_t vzip2q_f32(float32x4_t __p0, float32x4_t __p1) {
65885   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
65889 __ai float32x4_t vzip2q_f32(float32x4_t __p0, float32x4_t __p1) {
65891   float32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65900 __ai int32x4_t vzip2q_s32(int32x4_t __p0, int32x4_t __p1) {
65902   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
65906 __ai int32x4_t vzip2q_s32(int32x4_t __p0, int32x4_t __p1) {
65908   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
65917 __ai int64x2_t vzip2q_s64(int64x2_t __p0, int64x2_t __p1) {
65919   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65923 __ai int64x2_t vzip2q_s64(int64x2_t __p0, int64x2_t __p1) {
65925   int64x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65934 __ai int16x8_t vzip2q_s16(int16x8_t __p0, int16x8_t __p1) {
65936   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
65940 __ai int16x8_t vzip2q_s16(int16x8_t __p0, int16x8_t __p1) {
65942   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65951 __ai uint8x8_t vzip2_u8(uint8x8_t __p0, uint8x8_t __p1) {
65953   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
65957 __ai uint8x8_t vzip2_u8(uint8x8_t __p0, uint8x8_t __p1) {
65959   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
65968 __ai uint32x2_t vzip2_u32(uint32x2_t __p0, uint32x2_t __p1) {
65970   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
65974 __ai uint32x2_t vzip2_u32(uint32x2_t __p0, uint32x2_t __p1) {
65976   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
65985 __ai uint16x4_t vzip2_u16(uint16x4_t __p0, uint16x4_t __p1) {
65987   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
65991 __ai uint16x4_t vzip2_u16(uint16x4_t __p0, uint16x4_t __p1) {
65993   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66002 __ai int8x8_t vzip2_s8(int8x8_t __p0, int8x8_t __p1) {
66004   __ret = __builtin_shufflevector(__p0, __p1, 4, 12, 5, 13, 6, 14, 7, 15);
66008 __ai int8x8_t vzip2_s8(int8x8_t __p0, int8x8_t __p1) {
66010   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66019 __ai float32x2_t vzip2_f32(float32x2_t __p0, float32x2_t __p1) {
66021   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66025 __ai float32x2_t vzip2_f32(float32x2_t __p0, float32x2_t __p1) {
66027   float32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66036 __ai int32x2_t vzip2_s32(int32x2_t __p0, int32x2_t __p1) {
66038   __ret = __builtin_shufflevector(__p0, __p1, 1, 3);
66042 __ai int32x2_t vzip2_s32(int32x2_t __p0, int32x2_t __p1) {
66044   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66053 __ai int16x4_t vzip2_s16(int16x4_t __p0, int16x4_t __p1) {
66055   __ret = __builtin_shufflevector(__p0, __p1, 2, 6, 3, 7);
66059 __ai int16x4_t vzip2_s16(int16x4_t __p0, int16x4_t __p1) {
66061   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66071 __ai uint8x16_t vabaq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
66073   __ret = __p0 + vabdq_u8(__p1, __p2);
66077 __ai uint8x16_t vabaq_u8(uint8x16_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
66079   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66089 __ai uint32x4_t vabaq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
66091   __ret = __p0 + vabdq_u32(__p1, __p2);
66095 __ai uint32x4_t vabaq_u32(uint32x4_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
66097   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66107 __ai uint16x8_t vabaq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
66109   __ret = __p0 + vabdq_u16(__p1, __p2);
66113 __ai uint16x8_t vabaq_u16(uint16x8_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
66115   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66125 __ai int8x16_t vabaq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
66127   __ret = __p0 + vabdq_s8(__p1, __p2);
66131 __ai int8x16_t vabaq_s8(int8x16_t __p0, int8x16_t __p1, int8x16_t __p2) {
66133   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
66143 __ai int32x4_t vabaq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
66145   __ret = __p0 + vabdq_s32(__p1, __p2);
66149 __ai int32x4_t vabaq_s32(int32x4_t __p0, int32x4_t __p1, int32x4_t __p2) {
66151   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66161 __ai int16x8_t vabaq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
66163   __ret = __p0 + vabdq_s16(__p1, __p2);
66167 __ai int16x8_t vabaq_s16(int16x8_t __p0, int16x8_t __p1, int16x8_t __p2) {
66169   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66179 __ai uint8x8_t vaba_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
66181   __ret = __p0 + vabd_u8(__p1, __p2);
66185 __ai uint8x8_t vaba_u8(uint8x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
66187   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66197 __ai uint32x2_t vaba_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
66199   __ret = __p0 + vabd_u32(__p1, __p2);
66203 __ai uint32x2_t vaba_u32(uint32x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
66205   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66215 __ai uint16x4_t vaba_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
66217   __ret = __p0 + vabd_u16(__p1, __p2);
66221 __ai uint16x4_t vaba_u16(uint16x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
66223   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66233 __ai int8x8_t vaba_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
66235   __ret = __p0 + vabd_s8(__p1, __p2);
66239 __ai int8x8_t vaba_s8(int8x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
66241   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66251 __ai int32x2_t vaba_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
66253   __ret = __p0 + vabd_s32(__p1, __p2);
66257 __ai int32x2_t vaba_s32(int32x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
66259   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66269 __ai int16x4_t vaba_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
66271   __ret = __p0 + vabd_s16(__p1, __p2);
66275 __ai int16x4_t vaba_s16(int16x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
66277   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66287 __ai uint16x8_t vabdl_u8(uint8x8_t __p0, uint8x8_t __p1) {
66289   __ret = (uint16x8_t)(vmovl_u8((uint8x8_t)(vabd_u8(__p0, __p1))));
66293 __ai uint16x8_t vabdl_u8(uint8x8_t __p0, uint8x8_t __p1) {
66295   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66301 __ai uint16x8_t __noswap_vabdl_u8(uint8x8_t __p0, uint8x8_t __p1) {
66303   __ret = (uint16x8_t)(__noswap_vmovl_u8((uint8x8_t)(__noswap_vabd_u8(__p0, __p1))));
66309 __ai uint64x2_t vabdl_u32(uint32x2_t __p0, uint32x2_t __p1) {
66311   __ret = (uint64x2_t)(vmovl_u32((uint32x2_t)(vabd_u32(__p0, __p1))));
66315 __ai uint64x2_t vabdl_u32(uint32x2_t __p0, uint32x2_t __p1) {
66317   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66323 __ai uint64x2_t __noswap_vabdl_u32(uint32x2_t __p0, uint32x2_t __p1) {
66325   __ret = (uint64x2_t)(__noswap_vmovl_u32((uint32x2_t)(__noswap_vabd_u32(__p0, __p1))));
66331 __ai uint32x4_t vabdl_u16(uint16x4_t __p0, uint16x4_t __p1) {
66333   __ret = (uint32x4_t)(vmovl_u16((uint16x4_t)(vabd_u16(__p0, __p1))));
66337 __ai uint32x4_t vabdl_u16(uint16x4_t __p0, uint16x4_t __p1) {
66339   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66345 __ai uint32x4_t __noswap_vabdl_u16(uint16x4_t __p0, uint16x4_t __p1) {
66347   __ret = (uint32x4_t)(__noswap_vmovl_u16((uint16x4_t)(__noswap_vabd_u16(__p0, __p1))));
66353 __ai int16x8_t vabdl_s8(int8x8_t __p0, int8x8_t __p1) {
66355   __ret = (int16x8_t)(vmovl_u8((uint8x8_t)(vabd_s8(__p0, __p1))));
66359 __ai int16x8_t vabdl_s8(int8x8_t __p0, int8x8_t __p1) {
66361   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66367 __ai int16x8_t __noswap_vabdl_s8(int8x8_t __p0, int8x8_t __p1) {
66369   __ret = (int16x8_t)(__noswap_vmovl_u8((uint8x8_t)(__noswap_vabd_s8(__p0, __p1))));
66375 __ai int64x2_t vabdl_s32(int32x2_t __p0, int32x2_t __p1) {
66377   __ret = (int64x2_t)(vmovl_u32((uint32x2_t)(vabd_s32(__p0, __p1))));
66381 __ai int64x2_t vabdl_s32(int32x2_t __p0, int32x2_t __p1) {
66383   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66389 __ai int64x2_t __noswap_vabdl_s32(int32x2_t __p0, int32x2_t __p1) {
66391   __ret = (int64x2_t)(__noswap_vmovl_u32((uint32x2_t)(__noswap_vabd_s32(__p0, __p1))));
66397 __ai int32x4_t vabdl_s16(int16x4_t __p0, int16x4_t __p1) {
66399   __ret = (int32x4_t)(vmovl_u16((uint16x4_t)(vabd_s16(__p0, __p1))));
66403 __ai int32x4_t vabdl_s16(int16x4_t __p0, int16x4_t __p1) {
66405   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66411 __ai int32x4_t __noswap_vabdl_s16(int16x4_t __p0, int16x4_t __p1) {
66413   __ret = (int32x4_t)(__noswap_vmovl_u16((uint16x4_t)(__noswap_vabd_s16(__p0, __p1))));
66419 __ai uint16x8_t vaddl_u8(uint8x8_t __p0, uint8x8_t __p1) {
66421   __ret = vmovl_u8(__p0) + vmovl_u8(__p1);
66425 __ai uint16x8_t vaddl_u8(uint8x8_t __p0, uint8x8_t __p1) {
66427   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66436 __ai uint64x2_t vaddl_u32(uint32x2_t __p0, uint32x2_t __p1) {
66438   __ret = vmovl_u32(__p0) + vmovl_u32(__p1);
66442 __ai uint64x2_t vaddl_u32(uint32x2_t __p0, uint32x2_t __p1) {
66444   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66453 __ai uint32x4_t vaddl_u16(uint16x4_t __p0, uint16x4_t __p1) {
66455   __ret = vmovl_u16(__p0) + vmovl_u16(__p1);
66459 __ai uint32x4_t vaddl_u16(uint16x4_t __p0, uint16x4_t __p1) {
66461   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66470 __ai int16x8_t vaddl_s8(int8x8_t __p0, int8x8_t __p1) {
66472   __ret = vmovl_s8(__p0) + vmovl_s8(__p1);
66476 __ai int16x8_t vaddl_s8(int8x8_t __p0, int8x8_t __p1) {
66478   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66487 __ai int64x2_t vaddl_s32(int32x2_t __p0, int32x2_t __p1) {
66489   __ret = vmovl_s32(__p0) + vmovl_s32(__p1);
66493 __ai int64x2_t vaddl_s32(int32x2_t __p0, int32x2_t __p1) {
66495   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66504 __ai int32x4_t vaddl_s16(int16x4_t __p0, int16x4_t __p1) {
66506   __ret = vmovl_s16(__p0) + vmovl_s16(__p1);
66510 __ai int32x4_t vaddl_s16(int16x4_t __p0, int16x4_t __p1) {
66512   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66521 __ai uint16x8_t vaddw_u8(uint16x8_t __p0, uint8x8_t __p1) {
66523   __ret = __p0 + vmovl_u8(__p1);
66527 __ai uint16x8_t vaddw_u8(uint16x8_t __p0, uint8x8_t __p1) {
66529   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66538 __ai uint64x2_t vaddw_u32(uint64x2_t __p0, uint32x2_t __p1) {
66540   __ret = __p0 + vmovl_u32(__p1);
66544 __ai uint64x2_t vaddw_u32(uint64x2_t __p0, uint32x2_t __p1) {
66546   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66555 __ai uint32x4_t vaddw_u16(uint32x4_t __p0, uint16x4_t __p1) {
66557   __ret = __p0 + vmovl_u16(__p1);
66561 __ai uint32x4_t vaddw_u16(uint32x4_t __p0, uint16x4_t __p1) {
66563   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66572 __ai int16x8_t vaddw_s8(int16x8_t __p0, int8x8_t __p1) {
66574   __ret = __p0 + vmovl_s8(__p1);
66578 __ai int16x8_t vaddw_s8(int16x8_t __p0, int8x8_t __p1) {
66580   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66589 __ai int64x2_t vaddw_s32(int64x2_t __p0, int32x2_t __p1) {
66591   __ret = __p0 + vmovl_s32(__p1);
66595 __ai int64x2_t vaddw_s32(int64x2_t __p0, int32x2_t __p1) {
66597   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66606 __ai int32x4_t vaddw_s16(int32x4_t __p0, int16x4_t __p1) {
66608   __ret = __p0 + vmovl_s16(__p1);
66612 __ai int32x4_t vaddw_s16(int32x4_t __p0, int16x4_t __p1) {
66614   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66623 __ai uint16x8_t vmlal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
66625   __ret = __p0 + vmull_u8(__p1, __p2);
66629 __ai uint16x8_t vmlal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
66631   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66638 __ai uint16x8_t __noswap_vmlal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
66640   __ret = __p0 + __noswap_vmull_u8(__p1, __p2);
66646 __ai uint64x2_t vmlal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
66648   __ret = __p0 + vmull_u32(__p1, __p2);
66652 __ai uint64x2_t vmlal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
66654   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66661 __ai uint64x2_t __noswap_vmlal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
66663   __ret = __p0 + __noswap_vmull_u32(__p1, __p2);
66669 __ai uint32x4_t vmlal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
66671   __ret = __p0 + vmull_u16(__p1, __p2);
66675 __ai uint32x4_t vmlal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
66677   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66684 __ai uint32x4_t __noswap_vmlal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
66686   __ret = __p0 + __noswap_vmull_u16(__p1, __p2);
66692 __ai int16x8_t vmlal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
66694   __ret = __p0 + vmull_s8(__p1, __p2);
66698 __ai int16x8_t vmlal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
66700   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66707 __ai int16x8_t __noswap_vmlal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
66709   __ret = __p0 + __noswap_vmull_s8(__p1, __p2);
66715 __ai int64x2_t vmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
66717   __ret = __p0 + vmull_s32(__p1, __p2);
66721 __ai int64x2_t vmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
66723   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66730 __ai int64x2_t __noswap_vmlal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
66732   __ret = __p0 + __noswap_vmull_s32(__p1, __p2);
66738 __ai int32x4_t vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
66740   __ret = __p0 + vmull_s16(__p1, __p2);
66744 __ai int32x4_t vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
66746   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66753 __ai int32x4_t __noswap_vmlal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
66755   __ret = __p0 + __noswap_vmull_s16(__p1, __p2);
66761 #define vmlal_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
66763   uint32x2_t __s1 = __p1; \
66770 #define vmlal_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
66772   uint32x2_t __s1 = __p1; \
66785 #define vmlal_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
66787   uint16x4_t __s1 = __p1; \
66794 #define vmlal_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
66796   uint16x4_t __s1 = __p1; \
66809 #define vmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
66811   int32x2_t __s1 = __p1; \
66818 #define vmlal_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
66820   int32x2_t __s1 = __p1; \
66833 #define vmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
66835   int16x4_t __s1 = __p1; \
66842 #define vmlal_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
66844   int16x4_t __s1 = __p1; \
66857 __ai uint64x2_t vmlal_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
66859   __ret = __p0 + vmull_u32(__p1, (uint32x2_t) {__p2, __p2});
66863 __ai uint64x2_t vmlal_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
66865   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66871 __ai uint64x2_t __noswap_vmlal_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
66873   __ret = __p0 + __noswap_vmull_u32(__p1, (uint32x2_t) {__p2, __p2});
66879 __ai uint32x4_t vmlal_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
66881   __ret = __p0 + vmull_u16(__p1, (uint16x4_t) {__p2, __p2, __p2, __p2});
66885 __ai uint32x4_t vmlal_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
66887   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66893 __ai uint32x4_t __noswap_vmlal_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
66895   __ret = __p0 + __noswap_vmull_u16(__p1, (uint16x4_t) {__p2, __p2, __p2, __p2});
66901 __ai int64x2_t vmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
66903   __ret = __p0 + vmull_s32(__p1, (int32x2_t) {__p2, __p2});
66907 __ai int64x2_t vmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
66909   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66915 __ai int64x2_t __noswap_vmlal_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
66917   __ret = __p0 + __noswap_vmull_s32(__p1, (int32x2_t) {__p2, __p2});
66923 __ai int32x4_t vmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
66925   __ret = __p0 + vmull_s16(__p1, (int16x4_t) {__p2, __p2, __p2, __p2});
66929 __ai int32x4_t vmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
66931   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
66937 __ai int32x4_t __noswap_vmlal_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
66939   __ret = __p0 + __noswap_vmull_s16(__p1, (int16x4_t) {__p2, __p2, __p2, __p2});
66945 __ai uint16x8_t vmlsl_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
66947   __ret = __p0 - vmull_u8(__p1, __p2);
66951 __ai uint16x8_t vmlsl_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
66953   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
66960 __ai uint16x8_t __noswap_vmlsl_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
66962   __ret = __p0 - __noswap_vmull_u8(__p1, __p2);
66968 __ai uint64x2_t vmlsl_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
66970   __ret = __p0 - vmull_u32(__p1, __p2);
66974 __ai uint64x2_t vmlsl_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
66976   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
66983 __ai uint64x2_t __noswap_vmlsl_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
66985   __ret = __p0 - __noswap_vmull_u32(__p1, __p2);
66991 __ai uint32x4_t vmlsl_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
66993   __ret = __p0 - vmull_u16(__p1, __p2);
66997 __ai uint32x4_t vmlsl_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
66999   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67006 __ai uint32x4_t __noswap_vmlsl_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
67008   __ret = __p0 - __noswap_vmull_u16(__p1, __p2);
67014 __ai int16x8_t vmlsl_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67016   __ret = __p0 - vmull_s8(__p1, __p2);
67020 __ai int16x8_t vmlsl_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67022   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67029 __ai int16x8_t __noswap_vmlsl_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
67031   __ret = __p0 - __noswap_vmull_s8(__p1, __p2);
67037 __ai int64x2_t vmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67039   __ret = __p0 - vmull_s32(__p1, __p2);
67043 __ai int64x2_t vmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67045   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67052 __ai int64x2_t __noswap_vmlsl_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
67054   __ret = __p0 - __noswap_vmull_s32(__p1, __p2);
67060 __ai int32x4_t vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67062   __ret = __p0 - vmull_s16(__p1, __p2);
67066 __ai int32x4_t vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67068   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67075 __ai int32x4_t __noswap_vmlsl_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
67077   __ret = __p0 - __noswap_vmull_s16(__p1, __p2);
67083 #define vmlsl_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
67085   uint32x2_t __s1 = __p1; \
67092 #define vmlsl_lane_u32(__p0, __p1, __p2, __p3) __extension__ ({ \
67094   uint32x2_t __s1 = __p1; \
67107 #define vmlsl_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
67109   uint16x4_t __s1 = __p1; \
67116 #define vmlsl_lane_u16(__p0, __p1, __p2, __p3) __extension__ ({ \
67118   uint16x4_t __s1 = __p1; \
67131 #define vmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
67133   int32x2_t __s1 = __p1; \
67140 #define vmlsl_lane_s32(__p0, __p1, __p2, __p3) __extension__ ({ \
67142   int32x2_t __s1 = __p1; \
67155 #define vmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
67157   int16x4_t __s1 = __p1; \
67164 #define vmlsl_lane_s16(__p0, __p1, __p2, __p3) __extension__ ({ \
67166   int16x4_t __s1 = __p1; \
67179 __ai uint64x2_t vmlsl_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
67181   __ret = __p0 - vmull_u32(__p1, (uint32x2_t) {__p2, __p2});
67185 __ai uint64x2_t vmlsl_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
67187   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67193 __ai uint64x2_t __noswap_vmlsl_n_u32(uint64x2_t __p0, uint32x2_t __p1, uint32_t __p2) {
67195   __ret = __p0 - __noswap_vmull_u32(__p1, (uint32x2_t) {__p2, __p2});
67201 __ai uint32x4_t vmlsl_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
67203   __ret = __p0 - vmull_u16(__p1, (uint16x4_t) {__p2, __p2, __p2, __p2});
67207 __ai uint32x4_t vmlsl_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
67209   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67215 __ai uint32x4_t __noswap_vmlsl_n_u16(uint32x4_t __p0, uint16x4_t __p1, uint16_t __p2) {
67217   __ret = __p0 - __noswap_vmull_u16(__p1, (uint16x4_t) {__p2, __p2, __p2, __p2});
67223 __ai int64x2_t vmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
67225   __ret = __p0 - vmull_s32(__p1, (int32x2_t) {__p2, __p2});
67229 __ai int64x2_t vmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
67231   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
67237 __ai int64x2_t __noswap_vmlsl_n_s32(int64x2_t __p0, int32x2_t __p1, int32_t __p2) {
67239   __ret = __p0 - __noswap_vmull_s32(__p1, (int32x2_t) {__p2, __p2});
67245 __ai int32x4_t vmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
67247   __ret = __p0 - vmull_s16(__p1, (int16x4_t) {__p2, __p2, __p2, __p2});
67251 __ai int32x4_t vmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
67253   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67259 __ai int32x4_t __noswap_vmlsl_n_s16(int32x4_t __p0, int16x4_t __p1, int16_t __p2) {
67261   __ret = __p0 - __noswap_vmull_s16(__p1, (int16x4_t) {__p2, __p2, __p2, __p2});
67268 __ai uint16x8_t vabdl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
67270   __ret = vabdl_u8(vget_high_u8(__p0), vget_high_u8(__p1));
67274 __ai uint16x8_t vabdl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
67276   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67285 __ai uint64x2_t vabdl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
67287   __ret = vabdl_u32(vget_high_u32(__p0), vget_high_u32(__p1));
67291 __ai uint64x2_t vabdl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
67293   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67302 __ai uint32x4_t vabdl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
67304   __ret = vabdl_u16(vget_high_u16(__p0), vget_high_u16(__p1));
67308 __ai uint32x4_t vabdl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
67310   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67319 __ai int16x8_t vabdl_high_s8(int8x16_t __p0, int8x16_t __p1) {
67321   __ret = vabdl_s8(vget_high_s8(__p0), vget_high_s8(__p1));
67325 __ai int16x8_t vabdl_high_s8(int8x16_t __p0, int8x16_t __p1) {
67327   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67336 __ai int64x2_t vabdl_high_s32(int32x4_t __p0, int32x4_t __p1) {
67338   __ret = vabdl_s32(vget_high_s32(__p0), vget_high_s32(__p1));
67342 __ai int64x2_t vabdl_high_s32(int32x4_t __p0, int32x4_t __p1) {
67344   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67353 __ai int32x4_t vabdl_high_s16(int16x8_t __p0, int16x8_t __p1) {
67355   __ret = vabdl_s16(vget_high_s16(__p0), vget_high_s16(__p1));
67359 __ai int32x4_t vabdl_high_s16(int16x8_t __p0, int16x8_t __p1) {
67361   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67370 __ai uint16x8_t vaddl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
67372   __ret = vmovl_high_u8(__p0) + vmovl_high_u8(__p1);
67376 __ai uint16x8_t vaddl_high_u8(uint8x16_t __p0, uint8x16_t __p1) {
67378   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67387 __ai uint64x2_t vaddl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
67389   __ret = vmovl_high_u32(__p0) + vmovl_high_u32(__p1);
67393 __ai uint64x2_t vaddl_high_u32(uint32x4_t __p0, uint32x4_t __p1) {
67395   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67404 __ai uint32x4_t vaddl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
67406   __ret = vmovl_high_u16(__p0) + vmovl_high_u16(__p1);
67410 __ai uint32x4_t vaddl_high_u16(uint16x8_t __p0, uint16x8_t __p1) {
67412   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67421 __ai int16x8_t vaddl_high_s8(int8x16_t __p0, int8x16_t __p1) {
67423   __ret = vmovl_high_s8(__p0) + vmovl_high_s8(__p1);
67427 __ai int16x8_t vaddl_high_s8(int8x16_t __p0, int8x16_t __p1) {
67429   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67438 __ai int64x2_t vaddl_high_s32(int32x4_t __p0, int32x4_t __p1) {
67440   __ret = vmovl_high_s32(__p0) + vmovl_high_s32(__p1);
67444 __ai int64x2_t vaddl_high_s32(int32x4_t __p0, int32x4_t __p1) {
67446   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67455 __ai int32x4_t vaddl_high_s16(int16x8_t __p0, int16x8_t __p1) {
67457   __ret = vmovl_high_s16(__p0) + vmovl_high_s16(__p1);
67461 __ai int32x4_t vaddl_high_s16(int16x8_t __p0, int16x8_t __p1) {
67463   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67472 __ai uint16x8_t vaddw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
67474   __ret = __p0 + vmovl_high_u8(__p1);
67478 __ai uint16x8_t vaddw_high_u8(uint16x8_t __p0, uint8x16_t __p1) {
67480   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67489 __ai uint64x2_t vaddw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
67491   __ret = __p0 + vmovl_high_u32(__p1);
67495 __ai uint64x2_t vaddw_high_u32(uint64x2_t __p0, uint32x4_t __p1) {
67497   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67506 __ai uint32x4_t vaddw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
67508   __ret = __p0 + vmovl_high_u16(__p1);
67512 __ai uint32x4_t vaddw_high_u16(uint32x4_t __p0, uint16x8_t __p1) {
67514   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67523 __ai int16x8_t vaddw_high_s8(int16x8_t __p0, int8x16_t __p1) {
67525   __ret = __p0 + vmovl_high_s8(__p1);
67529 __ai int16x8_t vaddw_high_s8(int16x8_t __p0, int8x16_t __p1) {
67531   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67540 __ai int64x2_t vaddw_high_s32(int64x2_t __p0, int32x4_t __p1) {
67542   __ret = __p0 + vmovl_high_s32(__p1);
67546 __ai int64x2_t vaddw_high_s32(int64x2_t __p0, int32x4_t __p1) {
67548   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67557 __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
67559   __ret = __p0 + vmovl_high_s16(__p1);
67563 __ai int32x4_t vaddw_high_s16(int32x4_t __p0, int16x8_t __p1) {
67565   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67730 __ai uint16x8_t vmlal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
67732   __ret = vmlal_u8(__p0, vget_high_u8(__p1), vget_high_u8(__p2));
67736 __ai uint16x8_t vmlal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
67738   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67748 __ai uint64x2_t vmlal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
67750   __ret = vmlal_u32(__p0, vget_high_u32(__p1), vget_high_u32(__p2));
67754 __ai uint64x2_t vmlal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
67756   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67766 __ai uint32x4_t vmlal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
67768   __ret = vmlal_u16(__p0, vget_high_u16(__p1), vget_high_u16(__p2));
67772 __ai uint32x4_t vmlal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
67774   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67784 __ai int16x8_t vmlal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
67786   __ret = vmlal_s8(__p0, vget_high_s8(__p1), vget_high_s8(__p2));
67790 __ai int16x8_t vmlal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
67792   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67802 __ai int64x2_t vmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
67804   __ret = vmlal_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
67808 __ai int64x2_t vmlal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
67810   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67820 __ai int32x4_t vmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
67822   __ret = vmlal_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
67826 __ai int32x4_t vmlal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
67828   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67838 __ai uint64x2_t vmlal_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2) {
67840   __ret = vmlal_n_u32(__p0, vget_high_u32(__p1), __p2);
67844 __ai uint64x2_t vmlal_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2) {
67846   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67855 __ai uint32x4_t vmlal_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2) {
67857   __ret = vmlal_n_u16(__p0, vget_high_u16(__p1), __p2);
67861 __ai uint32x4_t vmlal_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2) {
67863   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67872 __ai int64x2_t vmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
67874   __ret = vmlal_n_s32(__p0, vget_high_s32(__p1), __p2);
67878 __ai int64x2_t vmlal_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
67880   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67889 __ai int32x4_t vmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
67891   __ret = vmlal_n_s16(__p0, vget_high_s16(__p1), __p2);
67895 __ai int32x4_t vmlal_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
67897   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67906 __ai uint16x8_t vmlsl_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
67908   __ret = vmlsl_u8(__p0, vget_high_u8(__p1), vget_high_u8(__p2));
67912 __ai uint16x8_t vmlsl_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
67914   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67924 __ai uint64x2_t vmlsl_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
67926   __ret = vmlsl_u32(__p0, vget_high_u32(__p1), vget_high_u32(__p2));
67930 __ai uint64x2_t vmlsl_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
67932   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67942 __ai uint32x4_t vmlsl_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
67944   __ret = vmlsl_u16(__p0, vget_high_u16(__p1), vget_high_u16(__p2));
67948 __ai uint32x4_t vmlsl_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
67950   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
67960 __ai int16x8_t vmlsl_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
67962   __ret = vmlsl_s8(__p0, vget_high_s8(__p1), vget_high_s8(__p2));
67966 __ai int16x8_t vmlsl_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
67968   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
67978 __ai int64x2_t vmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
67980   __ret = vmlsl_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
67984 __ai int64x2_t vmlsl_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
67986   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
67996 __ai int32x4_t vmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
67998   __ret = vmlsl_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
68002 __ai int32x4_t vmlsl_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
68004   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68014 __ai uint64x2_t vmlsl_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2) {
68016   __ret = vmlsl_n_u32(__p0, vget_high_u32(__p1), __p2);
68020 __ai uint64x2_t vmlsl_high_n_u32(uint64x2_t __p0, uint32x4_t __p1, uint32_t __p2) {
68022   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68031 __ai uint32x4_t vmlsl_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2) {
68033   __ret = vmlsl_n_u16(__p0, vget_high_u16(__p1), __p2);
68037 __ai uint32x4_t vmlsl_high_n_u16(uint32x4_t __p0, uint16x8_t __p1, uint16_t __p2) {
68039   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68048 __ai int64x2_t vmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
68050   __ret = vmlsl_n_s32(__p0, vget_high_s32(__p1), __p2);
68054 __ai int64x2_t vmlsl_high_n_s32(int64x2_t __p0, int32x4_t __p1, int32_t __p2) {
68056   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68065 __ai int32x4_t vmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
68067   __ret = vmlsl_n_s16(__p0, vget_high_s16(__p1), __p2);
68071 __ai int32x4_t vmlsl_high_n_s16(int32x4_t __p0, int16x8_t __p1, int16_t __p2) {
68073   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68132 __ai uint16x8_t vabal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
68134   __ret = __p0 + vabdl_u8(__p1, __p2);
68138 __ai uint16x8_t vabal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
68140   uint8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68147 __ai uint16x8_t __noswap_vabal_u8(uint16x8_t __p0, uint8x8_t __p1, uint8x8_t __p2) {
68149   __ret = __p0 + __noswap_vabdl_u8(__p1, __p2);
68155 __ai uint64x2_t vabal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
68157   __ret = __p0 + vabdl_u32(__p1, __p2);
68161 __ai uint64x2_t vabal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
68163   uint32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
68170 __ai uint64x2_t __noswap_vabal_u32(uint64x2_t __p0, uint32x2_t __p1, uint32x2_t __p2) {
68172   __ret = __p0 + __noswap_vabdl_u32(__p1, __p2);
68178 __ai uint32x4_t vabal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
68180   __ret = __p0 + vabdl_u16(__p1, __p2);
68184 __ai uint32x4_t vabal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
68186   uint16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68193 __ai uint32x4_t __noswap_vabal_u16(uint32x4_t __p0, uint16x4_t __p1, uint16x4_t __p2) {
68195   __ret = __p0 + __noswap_vabdl_u16(__p1, __p2);
68201 __ai int16x8_t vabal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
68203   __ret = __p0 + vabdl_s8(__p1, __p2);
68207 __ai int16x8_t vabal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
68209   int8x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68216 __ai int16x8_t __noswap_vabal_s8(int16x8_t __p0, int8x8_t __p1, int8x8_t __p2) {
68218   __ret = __p0 + __noswap_vabdl_s8(__p1, __p2);
68224 __ai int64x2_t vabal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
68226   __ret = __p0 + vabdl_s32(__p1, __p2);
68230 __ai int64x2_t vabal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
68232   int32x2_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 1, 0);
68239 __ai int64x2_t __noswap_vabal_s32(int64x2_t __p0, int32x2_t __p1, int32x2_t __p2) {
68241   __ret = __p0 + __noswap_vabdl_s32(__p1, __p2);
68247 __ai int32x4_t vabal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
68249   __ret = __p0 + vabdl_s16(__p1, __p2);
68253 __ai int32x4_t vabal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
68255   int16x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68262 __ai int32x4_t __noswap_vabal_s16(int32x4_t __p0, int16x4_t __p1, int16x4_t __p2) {
68264   __ret = __p0 + __noswap_vabdl_s16(__p1, __p2);
68271 __ai uint16x8_t vabal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
68273   __ret = vabal_u8(__p0, vget_high_u8(__p1), vget_high_u8(__p2));
68277 __ai uint16x8_t vabal_high_u8(uint16x8_t __p0, uint8x16_t __p1, uint8x16_t __p2) {
68279   uint8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68289 __ai uint64x2_t vabal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
68291   __ret = vabal_u32(__p0, vget_high_u32(__p1), vget_high_u32(__p2));
68295 __ai uint64x2_t vabal_high_u32(uint64x2_t __p0, uint32x4_t __p1, uint32x4_t __p2) {
68297   uint32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68307 __ai uint32x4_t vabal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
68309   __ret = vabal_u16(__p0, vget_high_u16(__p1), vget_high_u16(__p2));
68313 __ai uint32x4_t vabal_high_u16(uint32x4_t __p0, uint16x8_t __p1, uint16x8_t __p2) {
68315   uint16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);
68325 __ai int16x8_t vabal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
68327   __ret = vabal_s8(__p0, vget_high_s8(__p1), vget_high_s8(__p2));
68331 __ai int16x8_t vabal_high_s8(int16x8_t __p0, int8x16_t __p1, int8x16_t __p2) {
68333   int8x16_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 15, 14, 13, 12, 11, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0);
68343 __ai int64x2_t vabal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
68345   __ret = vabal_s32(__p0, vget_high_s32(__p1), vget_high_s32(__p2));
68349 __ai int64x2_t vabal_high_s32(int64x2_t __p0, int32x4_t __p1, int32x4_t __p2) {
68351   int32x4_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 3, 2, 1, 0);
68361 __ai int32x4_t vabal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
68363   __ret = vabal_s16(__p0, vget_high_s16(__p1), vget_high_s16(__p2));
68367 __ai int32x4_t vabal_high_s16(int32x4_t __p0, int16x8_t __p1, int16x8_t __p2) {
68369   int16x8_t __rev1;  __rev1 = __builtin_shufflevector(__p1, __p1, 7, 6, 5, 4, 3, 2, 1, 0);