Untitled :: RISC-V Ratified Specifications Library

Vector Loads and Stores Intrinsics

Vector Unit-Stride Load Intrinsics

vfloat16mf4_t __riscv_vle16_tu(vfloat16mf4_t vd, const _Float16 *rs1,
                               size_t vl);
vfloat16mf2_t __riscv_vle16_tu(vfloat16mf2_t vd, const _Float16 *rs1,
                               size_t vl);
vfloat16m1_t __riscv_vle16_tu(vfloat16m1_t vd, const _Float16 *rs1, size_t vl);
vfloat16m2_t __riscv_vle16_tu(vfloat16m2_t vd, const _Float16 *rs1, size_t vl);
vfloat16m4_t __riscv_vle16_tu(vfloat16m4_t vd, const _Float16 *rs1, size_t vl);
vfloat16m8_t __riscv_vle16_tu(vfloat16m8_t vd, const _Float16 *rs1, size_t vl);
vfloat32mf2_t __riscv_vle32_tu(vfloat32mf2_t vd, const float *rs1, size_t vl);
vfloat32m1_t __riscv_vle32_tu(vfloat32m1_t vd, const float *rs1, size_t vl);
vfloat32m2_t __riscv_vle32_tu(vfloat32m2_t vd, const float *rs1, size_t vl);
vfloat32m4_t __riscv_vle32_tu(vfloat32m4_t vd, const float *rs1, size_t vl);
vfloat32m8_t __riscv_vle32_tu(vfloat32m8_t vd, const float *rs1, size_t vl);
vfloat64m1_t __riscv_vle64_tu(vfloat64m1_t vd, const double *rs1, size_t vl);
vfloat64m2_t __riscv_vle64_tu(vfloat64m2_t vd, const double *rs1, size_t vl);
vfloat64m4_t __riscv_vle64_tu(vfloat64m4_t vd, const double *rs1, size_t vl);
vfloat64m8_t __riscv_vle64_tu(vfloat64m8_t vd, const double *rs1, size_t vl);
vint8mf8_t __riscv_vle8_tu(vint8mf8_t vd, const int8_t *rs1, size_t vl);
vint8mf4_t __riscv_vle8_tu(vint8mf4_t vd, const int8_t *rs1, size_t vl);
vint8mf2_t __riscv_vle8_tu(vint8mf2_t vd, const int8_t *rs1, size_t vl);
vint8m1_t __riscv_vle8_tu(vint8m1_t vd, const int8_t *rs1, size_t vl);
vint8m2_t __riscv_vle8_tu(vint8m2_t vd, const int8_t *rs1, size_t vl);
vint8m4_t __riscv_vle8_tu(vint8m4_t vd, const int8_t *rs1, size_t vl);
vint8m8_t __riscv_vle8_tu(vint8m8_t vd, const int8_t *rs1, size_t vl);
vint16mf4_t __riscv_vle16_tu(vint16mf4_t vd, const int16_t *rs1, size_t vl);
vint16mf2_t __riscv_vle16_tu(vint16mf2_t vd, const int16_t *rs1, size_t vl);
vint16m1_t __riscv_vle16_tu(vint16m1_t vd, const int16_t *rs1, size_t vl);
vint16m2_t __riscv_vle16_tu(vint16m2_t vd, const int16_t *rs1, size_t vl);
vint16m4_t __riscv_vle16_tu(vint16m4_t vd, const int16_t *rs1, size_t vl);
vint16m8_t __riscv_vle16_tu(vint16m8_t vd, const int16_t *rs1, size_t vl);
vint32mf2_t __riscv_vle32_tu(vint32mf2_t vd, const int32_t *rs1, size_t vl);
vint32m1_t __riscv_vle32_tu(vint32m1_t vd, const int32_t *rs1, size_t vl);
vint32m2_t __riscv_vle32_tu(vint32m2_t vd, const int32_t *rs1, size_t vl);
vint32m4_t __riscv_vle32_tu(vint32m4_t vd, const int32_t *rs1, size_t vl);
vint32m8_t __riscv_vle32_tu(vint32m8_t vd, const int32_t *rs1, size_t vl);
vint64m1_t __riscv_vle64_tu(vint64m1_t vd, const int64_t *rs1, size_t vl);
vint64m2_t __riscv_vle64_tu(vint64m2_t vd, const int64_t *rs1, size_t vl);
vint64m4_t __riscv_vle64_tu(vint64m4_t vd, const int64_t *rs1, size_t vl);
vint64m8_t __riscv_vle64_tu(vint64m8_t vd, const int64_t *rs1, size_t vl);
vuint8mf8_t __riscv_vle8_tu(vuint8mf8_t vd, const uint8_t *rs1, size_t vl);
vuint8mf4_t __riscv_vle8_tu(vuint8mf4_t vd, const uint8_t *rs1, size_t vl);
vuint8mf2_t __riscv_vle8_tu(vuint8mf2_t vd, const uint8_t *rs1, size_t vl);
vuint8m1_t __riscv_vle8_tu(vuint8m1_t vd, const uint8_t *rs1, size_t vl);
vuint8m2_t __riscv_vle8_tu(vuint8m2_t vd, const uint8_t *rs1, size_t vl);
vuint8m4_t __riscv_vle8_tu(vuint8m4_t vd, const uint8_t *rs1, size_t vl);
vuint8m8_t __riscv_vle8_tu(vuint8m8_t vd, const uint8_t *rs1, size_t vl);
vuint16mf4_t __riscv_vle16_tu(vuint16mf4_t vd, const uint16_t *rs1, size_t vl);
vuint16mf2_t __riscv_vle16_tu(vuint16mf2_t vd, const uint16_t *rs1, size_t vl);
vuint16m1_t __riscv_vle16_tu(vuint16m1_t vd, const uint16_t *rs1, size_t vl);
vuint16m2_t __riscv_vle16_tu(vuint16m2_t vd, const uint16_t *rs1, size_t vl);
vuint16m4_t __riscv_vle16_tu(vuint16m4_t vd, const uint16_t *rs1, size_t vl);
vuint16m8_t __riscv_vle16_tu(vuint16m8_t vd, const uint16_t *rs1, size_t vl);
vuint32mf2_t __riscv_vle32_tu(vuint32mf2_t vd, const uint32_t *rs1, size_t vl);
vuint32m1_t __riscv_vle32_tu(vuint32m1_t vd, const uint32_t *rs1, size_t vl);
vuint32m2_t __riscv_vle32_tu(vuint32m2_t vd, const uint32_t *rs1, size_t vl);
vuint32m4_t __riscv_vle32_tu(vuint32m4_t vd, const uint32_t *rs1, size_t vl);
vuint32m8_t __riscv_vle32_tu(vuint32m8_t vd, const uint32_t *rs1, size_t vl);
vuint64m1_t __riscv_vle64_tu(vuint64m1_t vd, const uint64_t *rs1, size_t vl);
vuint64m2_t __riscv_vle64_tu(vuint64m2_t vd, const uint64_t *rs1, size_t vl);
vuint64m4_t __riscv_vle64_tu(vuint64m4_t vd, const uint64_t *rs1, size_t vl);
vuint64m8_t __riscv_vle64_tu(vuint64m8_t vd, const uint64_t *rs1, size_t vl);
// masked functions
vfloat16mf4_t __riscv_vle16_tum(vbool64_t vm, vfloat16mf4_t vd,
                                const _Float16 *rs1, size_t vl);
vfloat16mf2_t __riscv_vle16_tum(vbool32_t vm, vfloat16mf2_t vd,
                                const _Float16 *rs1, size_t vl);
vfloat16m1_t __riscv_vle16_tum(vbool16_t vm, vfloat16m1_t vd,
                               const _Float16 *rs1, size_t vl);
vfloat16m2_t __riscv_vle16_tum(vbool8_t vm, vfloat16m2_t vd,
                               const _Float16 *rs1, size_t vl);
vfloat16m4_t __riscv_vle16_tum(vbool4_t vm, vfloat16m4_t vd,
                               const _Float16 *rs1, size_t vl);
vfloat16m8_t __riscv_vle16_tum(vbool2_t vm, vfloat16m8_t vd,
                               const _Float16 *rs1, size_t vl);
vfloat32mf2_t __riscv_vle32_tum(vbool64_t vm, vfloat32mf2_t vd,
                                const float *rs1, size_t vl);
vfloat32m1_t __riscv_vle32_tum(vbool32_t vm, vfloat32m1_t vd, const float *rs1,
                               size_t vl);
vfloat32m2_t __riscv_vle32_tum(vbool16_t vm, vfloat32m2_t vd, const float *rs1,
                               size_t vl);
vfloat32m4_t __riscv_vle32_tum(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                               size_t vl);
vfloat32m8_t __riscv_vle32_tum(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                               size_t vl);
vfloat64m1_t __riscv_vle64_tum(vbool64_t vm, vfloat64m1_t vd, const double *rs1,
                               size_t vl);
vfloat64m2_t __riscv_vle64_tum(vbool32_t vm, vfloat64m2_t vd, const double *rs1,
                               size_t vl);
vfloat64m4_t __riscv_vle64_tum(vbool16_t vm, vfloat64m4_t vd, const double *rs1,
                               size_t vl);
vfloat64m8_t __riscv_vle64_tum(vbool8_t vm, vfloat64m8_t vd, const double *rs1,
                               size_t vl);
vint8mf8_t __riscv_vle8_tum(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                            size_t vl);
vint8mf4_t __riscv_vle8_tum(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                            size_t vl);
vint8mf2_t __riscv_vle8_tum(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                            size_t vl);
vint8m1_t __riscv_vle8_tum(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                           size_t vl);
vint8m2_t __riscv_vle8_tum(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                           size_t vl);
vint8m4_t __riscv_vle8_tum(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                           size_t vl);
vint8m8_t __riscv_vle8_tum(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                           size_t vl);
vint16mf4_t __riscv_vle16_tum(vbool64_t vm, vint16mf4_t vd, const int16_t *rs1,
                              size_t vl);
vint16mf2_t __riscv_vle16_tum(vbool32_t vm, vint16mf2_t vd, const int16_t *rs1,
                              size_t vl);
vint16m1_t __riscv_vle16_tum(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                             size_t vl);
vint16m2_t __riscv_vle16_tum(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                             size_t vl);
vint16m4_t __riscv_vle16_tum(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                             size_t vl);
vint16m8_t __riscv_vle16_tum(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                             size_t vl);
vint32mf2_t __riscv_vle32_tum(vbool64_t vm, vint32mf2_t vd, const int32_t *rs1,
                              size_t vl);
vint32m1_t __riscv_vle32_tum(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                             size_t vl);
vint32m2_t __riscv_vle32_tum(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                             size_t vl);
vint32m4_t __riscv_vle32_tum(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                             size_t vl);
vint32m8_t __riscv_vle32_tum(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                             size_t vl);
vint64m1_t __riscv_vle64_tum(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                             size_t vl);
vint64m2_t __riscv_vle64_tum(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                             size_t vl);
vint64m4_t __riscv_vle64_tum(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                             size_t vl);
vint64m8_t __riscv_vle64_tum(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                             size_t vl);
vuint8mf8_t __riscv_vle8_tum(vbool64_t vm, vuint8mf8_t vd, const uint8_t *rs1,
                             size_t vl);
vuint8mf4_t __riscv_vle8_tum(vbool32_t vm, vuint8mf4_t vd, const uint8_t *rs1,
                             size_t vl);
vuint8mf2_t __riscv_vle8_tum(vbool16_t vm, vuint8mf2_t vd, const uint8_t *rs1,
                             size_t vl);
vuint8m1_t __riscv_vle8_tum(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                            size_t vl);
vuint8m2_t __riscv_vle8_tum(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                            size_t vl);
vuint8m4_t __riscv_vle8_tum(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                            size_t vl);
vuint8m8_t __riscv_vle8_tum(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                            size_t vl);
vuint16mf4_t __riscv_vle16_tum(vbool64_t vm, vuint16mf4_t vd,
                               const uint16_t *rs1, size_t vl);
vuint16mf2_t __riscv_vle16_tum(vbool32_t vm, vuint16mf2_t vd,
                               const uint16_t *rs1, size_t vl);
vuint16m1_t __riscv_vle16_tum(vbool16_t vm, vuint16m1_t vd, const uint16_t *rs1,
                              size_t vl);
vuint16m2_t __riscv_vle16_tum(vbool8_t vm, vuint16m2_t vd, const uint16_t *rs1,
                              size_t vl);
vuint16m4_t __riscv_vle16_tum(vbool4_t vm, vuint16m4_t vd, const uint16_t *rs1,
                              size_t vl);
vuint16m8_t __riscv_vle16_tum(vbool2_t vm, vuint16m8_t vd, const uint16_t *rs1,
                              size_t vl);
vuint32mf2_t __riscv_vle32_tum(vbool64_t vm, vuint32mf2_t vd,
                               const uint32_t *rs1, size_t vl);
vuint32m1_t __riscv_vle32_tum(vbool32_t vm, vuint32m1_t vd, const uint32_t *rs1,
                              size_t vl);
vuint32m2_t __riscv_vle32_tum(vbool16_t vm, vuint32m2_t vd, const uint32_t *rs1,
                              size_t vl);
vuint32m4_t __riscv_vle32_tum(vbool8_t vm, vuint32m4_t vd, const uint32_t *rs1,
                              size_t vl);
vuint32m8_t __riscv_vle32_tum(vbool4_t vm, vuint32m8_t vd, const uint32_t *rs1,
                              size_t vl);
vuint64m1_t __riscv_vle64_tum(vbool64_t vm, vuint64m1_t vd, const uint64_t *rs1,
                              size_t vl);
vuint64m2_t __riscv_vle64_tum(vbool32_t vm, vuint64m2_t vd, const uint64_t *rs1,
                              size_t vl);
vuint64m4_t __riscv_vle64_tum(vbool16_t vm, vuint64m4_t vd, const uint64_t *rs1,
                              size_t vl);
vuint64m8_t __riscv_vle64_tum(vbool8_t vm, vuint64m8_t vd, const uint64_t *rs1,
                              size_t vl);
// masked functions
vfloat16mf4_t __riscv_vle16_tumu(vbool64_t vm, vfloat16mf4_t vd,
                                 const _Float16 *rs1, size_t vl);
vfloat16mf2_t __riscv_vle16_tumu(vbool32_t vm, vfloat16mf2_t vd,
                                 const _Float16 *rs1, size_t vl);
vfloat16m1_t __riscv_vle16_tumu(vbool16_t vm, vfloat16m1_t vd,
                                const _Float16 *rs1, size_t vl);
vfloat16m2_t __riscv_vle16_tumu(vbool8_t vm, vfloat16m2_t vd,
                                const _Float16 *rs1, size_t vl);
vfloat16m4_t __riscv_vle16_tumu(vbool4_t vm, vfloat16m4_t vd,
                                const _Float16 *rs1, size_t vl);
vfloat16m8_t __riscv_vle16_tumu(vbool2_t vm, vfloat16m8_t vd,
                                const _Float16 *rs1, size_t vl);
vfloat32mf2_t __riscv_vle32_tumu(vbool64_t vm, vfloat32mf2_t vd,
                                 const float *rs1, size_t vl);
vfloat32m1_t __riscv_vle32_tumu(vbool32_t vm, vfloat32m1_t vd, const float *rs1,
                                size_t vl);
vfloat32m2_t __riscv_vle32_tumu(vbool16_t vm, vfloat32m2_t vd, const float *rs1,
                                size_t vl);
vfloat32m4_t __riscv_vle32_tumu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                size_t vl);
vfloat32m8_t __riscv_vle32_tumu(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                size_t vl);
vfloat64m1_t __riscv_vle64_tumu(vbool64_t vm, vfloat64m1_t vd,
                                const double *rs1, size_t vl);
vfloat64m2_t __riscv_vle64_tumu(vbool32_t vm, vfloat64m2_t vd,
                                const double *rs1, size_t vl);
vfloat64m4_t __riscv_vle64_tumu(vbool16_t vm, vfloat64m4_t vd,
                                const double *rs1, size_t vl);
vfloat64m8_t __riscv_vle64_tumu(vbool8_t vm, vfloat64m8_t vd, const double *rs1,
                                size_t vl);
vint8mf8_t __riscv_vle8_tumu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                             size_t vl);
vint8mf4_t __riscv_vle8_tumu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                             size_t vl);
vint8mf2_t __riscv_vle8_tumu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                             size_t vl);
vint8m1_t __riscv_vle8_tumu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                            size_t vl);
vint8m2_t __riscv_vle8_tumu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                            size_t vl);
vint8m4_t __riscv_vle8_tumu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                            size_t vl);
vint8m8_t __riscv_vle8_tumu(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                            size_t vl);
vint16mf4_t __riscv_vle16_tumu(vbool64_t vm, vint16mf4_t vd, const int16_t *rs1,
                               size_t vl);
vint16mf2_t __riscv_vle16_tumu(vbool32_t vm, vint16mf2_t vd, const int16_t *rs1,
                               size_t vl);
vint16m1_t __riscv_vle16_tumu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                              size_t vl);
vint16m2_t __riscv_vle16_tumu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                              size_t vl);
vint16m4_t __riscv_vle16_tumu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                              size_t vl);
vint16m8_t __riscv_vle16_tumu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                              size_t vl);
vint32mf2_t __riscv_vle32_tumu(vbool64_t vm, vint32mf2_t vd, const int32_t *rs1,
                               size_t vl);
vint32m1_t __riscv_vle32_tumu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                              size_t vl);
vint32m2_t __riscv_vle32_tumu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                              size_t vl);
vint32m4_t __riscv_vle32_tumu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                              size_t vl);
vint32m8_t __riscv_vle32_tumu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                              size_t vl);
vint64m1_t __riscv_vle64_tumu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                              size_t vl);
vint64m2_t __riscv_vle64_tumu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                              size_t vl);
vint64m4_t __riscv_vle64_tumu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                              size_t vl);
vint64m8_t __riscv_vle64_tumu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                              size_t vl);
vuint8mf8_t __riscv_vle8_tumu(vbool64_t vm, vuint8mf8_t vd, const uint8_t *rs1,
                              size_t vl);
vuint8mf4_t __riscv_vle8_tumu(vbool32_t vm, vuint8mf4_t vd, const uint8_t *rs1,
                              size_t vl);
vuint8mf2_t __riscv_vle8_tumu(vbool16_t vm, vuint8mf2_t vd, const uint8_t *rs1,
                              size_t vl);
vuint8m1_t __riscv_vle8_tumu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                             size_t vl);
vuint8m2_t __riscv_vle8_tumu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                             size_t vl);
vuint8m4_t __riscv_vle8_tumu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                             size_t vl);
vuint8m8_t __riscv_vle8_tumu(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                             size_t vl);
vuint16mf4_t __riscv_vle16_tumu(vbool64_t vm, vuint16mf4_t vd,
                                const uint16_t *rs1, size_t vl);
vuint16mf2_t __riscv_vle16_tumu(vbool32_t vm, vuint16mf2_t vd,
                                const uint16_t *rs1, size_t vl);
vuint16m1_t __riscv_vle16_tumu(vbool16_t vm, vuint16m1_t vd,
                               const uint16_t *rs1, size_t vl);
vuint16m2_t __riscv_vle16_tumu(vbool8_t vm, vuint16m2_t vd, const uint16_t *rs1,
                               size_t vl);
vuint16m4_t __riscv_vle16_tumu(vbool4_t vm, vuint16m4_t vd, const uint16_t *rs1,
                               size_t vl);
vuint16m8_t __riscv_vle16_tumu(vbool2_t vm, vuint16m8_t vd, const uint16_t *rs1,
                               size_t vl);
vuint32mf2_t __riscv_vle32_tumu(vbool64_t vm, vuint32mf2_t vd,
                                const uint32_t *rs1, size_t vl);
vuint32m1_t __riscv_vle32_tumu(vbool32_t vm, vuint32m1_t vd,
                               const uint32_t *rs1, size_t vl);
vuint32m2_t __riscv_vle32_tumu(vbool16_t vm, vuint32m2_t vd,
                               const uint32_t *rs1, size_t vl);
vuint32m4_t __riscv_vle32_tumu(vbool8_t vm, vuint32m4_t vd, const uint32_t *rs1,
                               size_t vl);
vuint32m8_t __riscv_vle32_tumu(vbool4_t vm, vuint32m8_t vd, const uint32_t *rs1,
                               size_t vl);
vuint64m1_t __riscv_vle64_tumu(vbool64_t vm, vuint64m1_t vd,
                               const uint64_t *rs1, size_t vl);
vuint64m2_t __riscv_vle64_tumu(vbool32_t vm, vuint64m2_t vd,
                               const uint64_t *rs1, size_t vl);
vuint64m4_t __riscv_vle64_tumu(vbool16_t vm, vuint64m4_t vd,
                               const uint64_t *rs1, size_t vl);
vuint64m8_t __riscv_vle64_tumu(vbool8_t vm, vuint64m8_t vd, const uint64_t *rs1,
                               size_t vl);
// masked functions
vfloat16mf4_t __riscv_vle16_mu(vbool64_t vm, vfloat16mf4_t vd,
                               const _Float16 *rs1, size_t vl);
vfloat16mf2_t __riscv_vle16_mu(vbool32_t vm, vfloat16mf2_t vd,
                               const _Float16 *rs1, size_t vl);
vfloat16m1_t __riscv_vle16_mu(vbool16_t vm, vfloat16m1_t vd,
                              const _Float16 *rs1, size_t vl);
vfloat16m2_t __riscv_vle16_mu(vbool8_t vm, vfloat16m2_t vd, const _Float16 *rs1,
                              size_t vl);
vfloat16m4_t __riscv_vle16_mu(vbool4_t vm, vfloat16m4_t vd, const _Float16 *rs1,
                              size_t vl);
vfloat16m8_t __riscv_vle16_mu(vbool2_t vm, vfloat16m8_t vd, const _Float16 *rs1,
                              size_t vl);
vfloat32mf2_t __riscv_vle32_mu(vbool64_t vm, vfloat32mf2_t vd, const float *rs1,
                               size_t vl);
vfloat32m1_t __riscv_vle32_mu(vbool32_t vm, vfloat32m1_t vd, const float *rs1,
                              size_t vl);
vfloat32m2_t __riscv_vle32_mu(vbool16_t vm, vfloat32m2_t vd, const float *rs1,
                              size_t vl);
vfloat32m4_t __riscv_vle32_mu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                              size_t vl);
vfloat32m8_t __riscv_vle32_mu(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                              size_t vl);
vfloat64m1_t __riscv_vle64_mu(vbool64_t vm, vfloat64m1_t vd, const double *rs1,
                              size_t vl);
vfloat64m2_t __riscv_vle64_mu(vbool32_t vm, vfloat64m2_t vd, const double *rs1,
                              size_t vl);
vfloat64m4_t __riscv_vle64_mu(vbool16_t vm, vfloat64m4_t vd, const double *rs1,
                              size_t vl);
vfloat64m8_t __riscv_vle64_mu(vbool8_t vm, vfloat64m8_t vd, const double *rs1,
                              size_t vl);
vint8mf8_t __riscv_vle8_mu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                           size_t vl);
vint8mf4_t __riscv_vle8_mu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                           size_t vl);
vint8mf2_t __riscv_vle8_mu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                           size_t vl);
vint8m1_t __riscv_vle8_mu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                          size_t vl);
vint8m2_t __riscv_vle8_mu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                          size_t vl);
vint8m4_t __riscv_vle8_mu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                          size_t vl);
vint8m8_t __riscv_vle8_mu(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                          size_t vl);
vint16mf4_t __riscv_vle16_mu(vbool64_t vm, vint16mf4_t vd, const int16_t *rs1,
                             size_t vl);
vint16mf2_t __riscv_vle16_mu(vbool32_t vm, vint16mf2_t vd, const int16_t *rs1,
                             size_t vl);
vint16m1_t __riscv_vle16_mu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                            size_t vl);
vint16m2_t __riscv_vle16_mu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                            size_t vl);
vint16m4_t __riscv_vle16_mu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                            size_t vl);
vint16m8_t __riscv_vle16_mu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                            size_t vl);
vint32mf2_t __riscv_vle32_mu(vbool64_t vm, vint32mf2_t vd, const int32_t *rs1,
                             size_t vl);
vint32m1_t __riscv_vle32_mu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                            size_t vl);
vint32m2_t __riscv_vle32_mu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                            size_t vl);
vint32m4_t __riscv_vle32_mu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                            size_t vl);
vint32m8_t __riscv_vle32_mu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                            size_t vl);
vint64m1_t __riscv_vle64_mu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                            size_t vl);
vint64m2_t __riscv_vle64_mu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                            size_t vl);
vint64m4_t __riscv_vle64_mu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                            size_t vl);
vint64m8_t __riscv_vle64_mu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                            size_t vl);
vuint8mf8_t __riscv_vle8_mu(vbool64_t vm, vuint8mf8_t vd, const uint8_t *rs1,
                            size_t vl);
vuint8mf4_t __riscv_vle8_mu(vbool32_t vm, vuint8mf4_t vd, const uint8_t *rs1,
                            size_t vl);
vuint8mf2_t __riscv_vle8_mu(vbool16_t vm, vuint8mf2_t vd, const uint8_t *rs1,
                            size_t vl);
vuint8m1_t __riscv_vle8_mu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                           size_t vl);
vuint8m2_t __riscv_vle8_mu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                           size_t vl);
vuint8m4_t __riscv_vle8_mu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                           size_t vl);
vuint8m8_t __riscv_vle8_mu(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                           size_t vl);
vuint16mf4_t __riscv_vle16_mu(vbool64_t vm, vuint16mf4_t vd,
                              const uint16_t *rs1, size_t vl);
vuint16mf2_t __riscv_vle16_mu(vbool32_t vm, vuint16mf2_t vd,
                              const uint16_t *rs1, size_t vl);
vuint16m1_t __riscv_vle16_mu(vbool16_t vm, vuint16m1_t vd, const uint16_t *rs1,
                             size_t vl);
vuint16m2_t __riscv_vle16_mu(vbool8_t vm, vuint16m2_t vd, const uint16_t *rs1,
                             size_t vl);
vuint16m4_t __riscv_vle16_mu(vbool4_t vm, vuint16m4_t vd, const uint16_t *rs1,
                             size_t vl);
vuint16m8_t __riscv_vle16_mu(vbool2_t vm, vuint16m8_t vd, const uint16_t *rs1,
                             size_t vl);
vuint32mf2_t __riscv_vle32_mu(vbool64_t vm, vuint32mf2_t vd,
                              const uint32_t *rs1, size_t vl);
vuint32m1_t __riscv_vle32_mu(vbool32_t vm, vuint32m1_t vd, const uint32_t *rs1,
                             size_t vl);
vuint32m2_t __riscv_vle32_mu(vbool16_t vm, vuint32m2_t vd, const uint32_t *rs1,
                             size_t vl);
vuint32m4_t __riscv_vle32_mu(vbool8_t vm, vuint32m4_t vd, const uint32_t *rs1,
                             size_t vl);
vuint32m8_t __riscv_vle32_mu(vbool4_t vm, vuint32m8_t vd, const uint32_t *rs1,
                             size_t vl);
vuint64m1_t __riscv_vle64_mu(vbool64_t vm, vuint64m1_t vd, const uint64_t *rs1,
                             size_t vl);
vuint64m2_t __riscv_vle64_mu(vbool32_t vm, vuint64m2_t vd, const uint64_t *rs1,
                             size_t vl);
vuint64m4_t __riscv_vle64_mu(vbool16_t vm, vuint64m4_t vd, const uint64_t *rs1,
                             size_t vl);
vuint64m8_t __riscv_vle64_mu(vbool8_t vm, vuint64m8_t vd, const uint64_t *rs1,
                             size_t vl);

Vector Unit-Stride Store Intrinsics

Intrinsics here don’t have a policy variant.

Vector Mask Load/Store Intrinsics

Intrinsics here don’t have a policy variant.

Vector Strided Load Intrinsics

vfloat16mf4_t __riscv_vlse16_tu(vfloat16mf4_t vd, const _Float16 *rs1,
                                ptrdiff_t rs2, size_t vl);
vfloat16mf2_t __riscv_vlse16_tu(vfloat16mf2_t vd, const _Float16 *rs1,
                                ptrdiff_t rs2, size_t vl);
vfloat16m1_t __riscv_vlse16_tu(vfloat16m1_t vd, const _Float16 *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat16m2_t __riscv_vlse16_tu(vfloat16m2_t vd, const _Float16 *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat16m4_t __riscv_vlse16_tu(vfloat16m4_t vd, const _Float16 *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat16m8_t __riscv_vlse16_tu(vfloat16m8_t vd, const _Float16 *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat32mf2_t __riscv_vlse32_tu(vfloat32mf2_t vd, const float *rs1,
                                ptrdiff_t rs2, size_t vl);
vfloat32m1_t __riscv_vlse32_tu(vfloat32m1_t vd, const float *rs1, ptrdiff_t rs2,
                               size_t vl);
vfloat32m2_t __riscv_vlse32_tu(vfloat32m2_t vd, const float *rs1, ptrdiff_t rs2,
                               size_t vl);
vfloat32m4_t __riscv_vlse32_tu(vfloat32m4_t vd, const float *rs1, ptrdiff_t rs2,
                               size_t vl);
vfloat32m8_t __riscv_vlse32_tu(vfloat32m8_t vd, const float *rs1, ptrdiff_t rs2,
                               size_t vl);
vfloat64m1_t __riscv_vlse64_tu(vfloat64m1_t vd, const double *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat64m2_t __riscv_vlse64_tu(vfloat64m2_t vd, const double *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat64m4_t __riscv_vlse64_tu(vfloat64m4_t vd, const double *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat64m8_t __riscv_vlse64_tu(vfloat64m8_t vd, const double *rs1,
                               ptrdiff_t rs2, size_t vl);
vint8mf8_t __riscv_vlse8_tu(vint8mf8_t vd, const int8_t *rs1, ptrdiff_t rs2,
                            size_t vl);
vint8mf4_t __riscv_vlse8_tu(vint8mf4_t vd, const int8_t *rs1, ptrdiff_t rs2,
                            size_t vl);
vint8mf2_t __riscv_vlse8_tu(vint8mf2_t vd, const int8_t *rs1, ptrdiff_t rs2,
                            size_t vl);
vint8m1_t __riscv_vlse8_tu(vint8m1_t vd, const int8_t *rs1, ptrdiff_t rs2,
                           size_t vl);
vint8m2_t __riscv_vlse8_tu(vint8m2_t vd, const int8_t *rs1, ptrdiff_t rs2,
                           size_t vl);
vint8m4_t __riscv_vlse8_tu(vint8m4_t vd, const int8_t *rs1, ptrdiff_t rs2,
                           size_t vl);
vint8m8_t __riscv_vlse8_tu(vint8m8_t vd, const int8_t *rs1, ptrdiff_t rs2,
                           size_t vl);
vint16mf4_t __riscv_vlse16_tu(vint16mf4_t vd, const int16_t *rs1, ptrdiff_t rs2,
                              size_t vl);
vint16mf2_t __riscv_vlse16_tu(vint16mf2_t vd, const int16_t *rs1, ptrdiff_t rs2,
                              size_t vl);
vint16m1_t __riscv_vlse16_tu(vint16m1_t vd, const int16_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vint16m2_t __riscv_vlse16_tu(vint16m2_t vd, const int16_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vint16m4_t __riscv_vlse16_tu(vint16m4_t vd, const int16_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vint16m8_t __riscv_vlse16_tu(vint16m8_t vd, const int16_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vint32mf2_t __riscv_vlse32_tu(vint32mf2_t vd, const int32_t *rs1, ptrdiff_t rs2,
                              size_t vl);
vint32m1_t __riscv_vlse32_tu(vint32m1_t vd, const int32_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vint32m2_t __riscv_vlse32_tu(vint32m2_t vd, const int32_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vint32m4_t __riscv_vlse32_tu(vint32m4_t vd, const int32_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vint32m8_t __riscv_vlse32_tu(vint32m8_t vd, const int32_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vint64m1_t __riscv_vlse64_tu(vint64m1_t vd, const int64_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vint64m2_t __riscv_vlse64_tu(vint64m2_t vd, const int64_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vint64m4_t __riscv_vlse64_tu(vint64m4_t vd, const int64_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vint64m8_t __riscv_vlse64_tu(vint64m8_t vd, const int64_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vuint8mf8_t __riscv_vlse8_tu(vuint8mf8_t vd, const uint8_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vuint8mf4_t __riscv_vlse8_tu(vuint8mf4_t vd, const uint8_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vuint8mf2_t __riscv_vlse8_tu(vuint8mf2_t vd, const uint8_t *rs1, ptrdiff_t rs2,
                             size_t vl);
vuint8m1_t __riscv_vlse8_tu(vuint8m1_t vd, const uint8_t *rs1, ptrdiff_t rs2,
                            size_t vl);
vuint8m2_t __riscv_vlse8_tu(vuint8m2_t vd, const uint8_t *rs1, ptrdiff_t rs2,
                            size_t vl);
vuint8m4_t __riscv_vlse8_tu(vuint8m4_t vd, const uint8_t *rs1, ptrdiff_t rs2,
                            size_t vl);
vuint8m8_t __riscv_vlse8_tu(vuint8m8_t vd, const uint8_t *rs1, ptrdiff_t rs2,
                            size_t vl);
vuint16mf4_t __riscv_vlse16_tu(vuint16mf4_t vd, const uint16_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint16mf2_t __riscv_vlse16_tu(vuint16mf2_t vd, const uint16_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint16m1_t __riscv_vlse16_tu(vuint16m1_t vd, const uint16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint16m2_t __riscv_vlse16_tu(vuint16m2_t vd, const uint16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint16m4_t __riscv_vlse16_tu(vuint16m4_t vd, const uint16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint16m8_t __riscv_vlse16_tu(vuint16m8_t vd, const uint16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint32mf2_t __riscv_vlse32_tu(vuint32mf2_t vd, const uint32_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint32m1_t __riscv_vlse32_tu(vuint32m1_t vd, const uint32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint32m2_t __riscv_vlse32_tu(vuint32m2_t vd, const uint32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint32m4_t __riscv_vlse32_tu(vuint32m4_t vd, const uint32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint32m8_t __riscv_vlse32_tu(vuint32m8_t vd, const uint32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint64m1_t __riscv_vlse64_tu(vuint64m1_t vd, const uint64_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint64m2_t __riscv_vlse64_tu(vuint64m2_t vd, const uint64_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint64m4_t __riscv_vlse64_tu(vuint64m4_t vd, const uint64_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint64m8_t __riscv_vlse64_tu(vuint64m8_t vd, const uint64_t *rs1,
                              ptrdiff_t rs2, size_t vl);
// masked functions
vfloat16mf4_t __riscv_vlse16_tum(vbool64_t vm, vfloat16mf4_t vd,
                                 const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16mf2_t __riscv_vlse16_tum(vbool32_t vm, vfloat16mf2_t vd,
                                 const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16m1_t __riscv_vlse16_tum(vbool16_t vm, vfloat16m1_t vd,
                                const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16m2_t __riscv_vlse16_tum(vbool8_t vm, vfloat16m2_t vd,
                                const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16m4_t __riscv_vlse16_tum(vbool4_t vm, vfloat16m4_t vd,
                                const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16m8_t __riscv_vlse16_tum(vbool2_t vm, vfloat16m8_t vd,
                                const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat32mf2_t __riscv_vlse32_tum(vbool64_t vm, vfloat32mf2_t vd,
                                 const float *rs1, ptrdiff_t rs2, size_t vl);
vfloat32m1_t __riscv_vlse32_tum(vbool32_t vm, vfloat32m1_t vd, const float *rs1,
                                ptrdiff_t rs2, size_t vl);
vfloat32m2_t __riscv_vlse32_tum(vbool16_t vm, vfloat32m2_t vd, const float *rs1,
                                ptrdiff_t rs2, size_t vl);
vfloat32m4_t __riscv_vlse32_tum(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                ptrdiff_t rs2, size_t vl);
vfloat32m8_t __riscv_vlse32_tum(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                ptrdiff_t rs2, size_t vl);
vfloat64m1_t __riscv_vlse64_tum(vbool64_t vm, vfloat64m1_t vd,
                                const double *rs1, ptrdiff_t rs2, size_t vl);
vfloat64m2_t __riscv_vlse64_tum(vbool32_t vm, vfloat64m2_t vd,
                                const double *rs1, ptrdiff_t rs2, size_t vl);
vfloat64m4_t __riscv_vlse64_tum(vbool16_t vm, vfloat64m4_t vd,
                                const double *rs1, ptrdiff_t rs2, size_t vl);
vfloat64m8_t __riscv_vlse64_tum(vbool8_t vm, vfloat64m8_t vd, const double *rs1,
                                ptrdiff_t rs2, size_t vl);
vint8mf8_t __riscv_vlse8_tum(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint8mf4_t __riscv_vlse8_tum(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint8mf2_t __riscv_vlse8_tum(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint8m1_t __riscv_vlse8_tum(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                            ptrdiff_t rs2, size_t vl);
vint8m2_t __riscv_vlse8_tum(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                            ptrdiff_t rs2, size_t vl);
vint8m4_t __riscv_vlse8_tum(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                            ptrdiff_t rs2, size_t vl);
vint8m8_t __riscv_vlse8_tum(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                            ptrdiff_t rs2, size_t vl);
vint16mf4_t __riscv_vlse16_tum(vbool64_t vm, vint16mf4_t vd, const int16_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint16mf2_t __riscv_vlse16_tum(vbool32_t vm, vint16mf2_t vd, const int16_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint16m1_t __riscv_vlse16_tum(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint16m2_t __riscv_vlse16_tum(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint16m4_t __riscv_vlse16_tum(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint16m8_t __riscv_vlse16_tum(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint32mf2_t __riscv_vlse32_tum(vbool64_t vm, vint32mf2_t vd, const int32_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint32m1_t __riscv_vlse32_tum(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint32m2_t __riscv_vlse32_tum(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint32m4_t __riscv_vlse32_tum(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint32m8_t __riscv_vlse32_tum(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint64m1_t __riscv_vlse64_tum(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint64m2_t __riscv_vlse64_tum(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint64m4_t __riscv_vlse64_tum(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint64m8_t __riscv_vlse64_tum(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint8mf8_t __riscv_vlse8_tum(vbool64_t vm, vuint8mf8_t vd, const uint8_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint8mf4_t __riscv_vlse8_tum(vbool32_t vm, vuint8mf4_t vd, const uint8_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint8mf2_t __riscv_vlse8_tum(vbool16_t vm, vuint8mf2_t vd, const uint8_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint8m1_t __riscv_vlse8_tum(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vuint8m2_t __riscv_vlse8_tum(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vuint8m4_t __riscv_vlse8_tum(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vuint8m8_t __riscv_vlse8_tum(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vuint16mf4_t __riscv_vlse16_tum(vbool64_t vm, vuint16mf4_t vd,
                                const uint16_t *rs1, ptrdiff_t rs2, size_t vl);
vuint16mf2_t __riscv_vlse16_tum(vbool32_t vm, vuint16mf2_t vd,
                                const uint16_t *rs1, ptrdiff_t rs2, size_t vl);
vuint16m1_t __riscv_vlse16_tum(vbool16_t vm, vuint16m1_t vd,
                               const uint16_t *rs1, ptrdiff_t rs2, size_t vl);
vuint16m2_t __riscv_vlse16_tum(vbool8_t vm, vuint16m2_t vd, const uint16_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint16m4_t __riscv_vlse16_tum(vbool4_t vm, vuint16m4_t vd, const uint16_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint16m8_t __riscv_vlse16_tum(vbool2_t vm, vuint16m8_t vd, const uint16_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint32mf2_t __riscv_vlse32_tum(vbool64_t vm, vuint32mf2_t vd,
                                const uint32_t *rs1, ptrdiff_t rs2, size_t vl);
vuint32m1_t __riscv_vlse32_tum(vbool32_t vm, vuint32m1_t vd,
                               const uint32_t *rs1, ptrdiff_t rs2, size_t vl);
vuint32m2_t __riscv_vlse32_tum(vbool16_t vm, vuint32m2_t vd,
                               const uint32_t *rs1, ptrdiff_t rs2, size_t vl);
vuint32m4_t __riscv_vlse32_tum(vbool8_t vm, vuint32m4_t vd, const uint32_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint32m8_t __riscv_vlse32_tum(vbool4_t vm, vuint32m8_t vd, const uint32_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint64m1_t __riscv_vlse64_tum(vbool64_t vm, vuint64m1_t vd,
                               const uint64_t *rs1, ptrdiff_t rs2, size_t vl);
vuint64m2_t __riscv_vlse64_tum(vbool32_t vm, vuint64m2_t vd,
                               const uint64_t *rs1, ptrdiff_t rs2, size_t vl);
vuint64m4_t __riscv_vlse64_tum(vbool16_t vm, vuint64m4_t vd,
                               const uint64_t *rs1, ptrdiff_t rs2, size_t vl);
vuint64m8_t __riscv_vlse64_tum(vbool8_t vm, vuint64m8_t vd, const uint64_t *rs1,
                               ptrdiff_t rs2, size_t vl);
// masked functions
vfloat16mf4_t __riscv_vlse16_tumu(vbool64_t vm, vfloat16mf4_t vd,
                                  const _Float16 *rs1, ptrdiff_t rs2,
                                  size_t vl);
vfloat16mf2_t __riscv_vlse16_tumu(vbool32_t vm, vfloat16mf2_t vd,
                                  const _Float16 *rs1, ptrdiff_t rs2,
                                  size_t vl);
vfloat16m1_t __riscv_vlse16_tumu(vbool16_t vm, vfloat16m1_t vd,
                                 const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16m2_t __riscv_vlse16_tumu(vbool8_t vm, vfloat16m2_t vd,
                                 const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16m4_t __riscv_vlse16_tumu(vbool4_t vm, vfloat16m4_t vd,
                                 const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16m8_t __riscv_vlse16_tumu(vbool2_t vm, vfloat16m8_t vd,
                                 const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat32mf2_t __riscv_vlse32_tumu(vbool64_t vm, vfloat32mf2_t vd,
                                  const float *rs1, ptrdiff_t rs2, size_t vl);
vfloat32m1_t __riscv_vlse32_tumu(vbool32_t vm, vfloat32m1_t vd,
                                 const float *rs1, ptrdiff_t rs2, size_t vl);
vfloat32m2_t __riscv_vlse32_tumu(vbool16_t vm, vfloat32m2_t vd,
                                 const float *rs1, ptrdiff_t rs2, size_t vl);
vfloat32m4_t __riscv_vlse32_tumu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                 ptrdiff_t rs2, size_t vl);
vfloat32m8_t __riscv_vlse32_tumu(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                 ptrdiff_t rs2, size_t vl);
vfloat64m1_t __riscv_vlse64_tumu(vbool64_t vm, vfloat64m1_t vd,
                                 const double *rs1, ptrdiff_t rs2, size_t vl);
vfloat64m2_t __riscv_vlse64_tumu(vbool32_t vm, vfloat64m2_t vd,
                                 const double *rs1, ptrdiff_t rs2, size_t vl);
vfloat64m4_t __riscv_vlse64_tumu(vbool16_t vm, vfloat64m4_t vd,
                                 const double *rs1, ptrdiff_t rs2, size_t vl);
vfloat64m8_t __riscv_vlse64_tumu(vbool8_t vm, vfloat64m8_t vd,
                                 const double *rs1, ptrdiff_t rs2, size_t vl);
vint8mf8_t __riscv_vlse8_tumu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint8mf4_t __riscv_vlse8_tumu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint8mf2_t __riscv_vlse8_tumu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint8m1_t __riscv_vlse8_tumu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint8m2_t __riscv_vlse8_tumu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint8m4_t __riscv_vlse8_tumu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint8m8_t __riscv_vlse8_tumu(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint16mf4_t __riscv_vlse16_tumu(vbool64_t vm, vint16mf4_t vd,
                                const int16_t *rs1, ptrdiff_t rs2, size_t vl);
vint16mf2_t __riscv_vlse16_tumu(vbool32_t vm, vint16mf2_t vd,
                                const int16_t *rs1, ptrdiff_t rs2, size_t vl);
vint16m1_t __riscv_vlse16_tumu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint16m2_t __riscv_vlse16_tumu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint16m4_t __riscv_vlse16_tumu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint16m8_t __riscv_vlse16_tumu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint32mf2_t __riscv_vlse32_tumu(vbool64_t vm, vint32mf2_t vd,
                                const int32_t *rs1, ptrdiff_t rs2, size_t vl);
vint32m1_t __riscv_vlse32_tumu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint32m2_t __riscv_vlse32_tumu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint32m4_t __riscv_vlse32_tumu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint32m8_t __riscv_vlse32_tumu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint64m1_t __riscv_vlse64_tumu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint64m2_t __riscv_vlse64_tumu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint64m4_t __riscv_vlse64_tumu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vint64m8_t __riscv_vlse64_tumu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint8mf8_t __riscv_vlse8_tumu(vbool64_t vm, vuint8mf8_t vd, const uint8_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint8mf4_t __riscv_vlse8_tumu(vbool32_t vm, vuint8mf4_t vd, const uint8_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint8mf2_t __riscv_vlse8_tumu(vbool16_t vm, vuint8mf2_t vd, const uint8_t *rs1,
                               ptrdiff_t rs2, size_t vl);
vuint8m1_t __riscv_vlse8_tumu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint8m2_t __riscv_vlse8_tumu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint8m4_t __riscv_vlse8_tumu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint8m8_t __riscv_vlse8_tumu(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint16mf4_t __riscv_vlse16_tumu(vbool64_t vm, vuint16mf4_t vd,
                                 const uint16_t *rs1, ptrdiff_t rs2, size_t vl);
vuint16mf2_t __riscv_vlse16_tumu(vbool32_t vm, vuint16mf2_t vd,
                                 const uint16_t *rs1, ptrdiff_t rs2, size_t vl);
vuint16m1_t __riscv_vlse16_tumu(vbool16_t vm, vuint16m1_t vd,
                                const uint16_t *rs1, ptrdiff_t rs2, size_t vl);
vuint16m2_t __riscv_vlse16_tumu(vbool8_t vm, vuint16m2_t vd,
                                const uint16_t *rs1, ptrdiff_t rs2, size_t vl);
vuint16m4_t __riscv_vlse16_tumu(vbool4_t vm, vuint16m4_t vd,
                                const uint16_t *rs1, ptrdiff_t rs2, size_t vl);
vuint16m8_t __riscv_vlse16_tumu(vbool2_t vm, vuint16m8_t vd,
                                const uint16_t *rs1, ptrdiff_t rs2, size_t vl);
vuint32mf2_t __riscv_vlse32_tumu(vbool64_t vm, vuint32mf2_t vd,
                                 const uint32_t *rs1, ptrdiff_t rs2, size_t vl);
vuint32m1_t __riscv_vlse32_tumu(vbool32_t vm, vuint32m1_t vd,
                                const uint32_t *rs1, ptrdiff_t rs2, size_t vl);
vuint32m2_t __riscv_vlse32_tumu(vbool16_t vm, vuint32m2_t vd,
                                const uint32_t *rs1, ptrdiff_t rs2, size_t vl);
vuint32m4_t __riscv_vlse32_tumu(vbool8_t vm, vuint32m4_t vd,
                                const uint32_t *rs1, ptrdiff_t rs2, size_t vl);
vuint32m8_t __riscv_vlse32_tumu(vbool4_t vm, vuint32m8_t vd,
                                const uint32_t *rs1, ptrdiff_t rs2, size_t vl);
vuint64m1_t __riscv_vlse64_tumu(vbool64_t vm, vuint64m1_t vd,
                                const uint64_t *rs1, ptrdiff_t rs2, size_t vl);
vuint64m2_t __riscv_vlse64_tumu(vbool32_t vm, vuint64m2_t vd,
                                const uint64_t *rs1, ptrdiff_t rs2, size_t vl);
vuint64m4_t __riscv_vlse64_tumu(vbool16_t vm, vuint64m4_t vd,
                                const uint64_t *rs1, ptrdiff_t rs2, size_t vl);
vuint64m8_t __riscv_vlse64_tumu(vbool8_t vm, vuint64m8_t vd,
                                const uint64_t *rs1, ptrdiff_t rs2, size_t vl);
// masked functions
vfloat16mf4_t __riscv_vlse16_mu(vbool64_t vm, vfloat16mf4_t vd,
                                const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16mf2_t __riscv_vlse16_mu(vbool32_t vm, vfloat16mf2_t vd,
                                const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16m1_t __riscv_vlse16_mu(vbool16_t vm, vfloat16m1_t vd,
                               const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16m2_t __riscv_vlse16_mu(vbool8_t vm, vfloat16m2_t vd,
                               const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16m4_t __riscv_vlse16_mu(vbool4_t vm, vfloat16m4_t vd,
                               const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat16m8_t __riscv_vlse16_mu(vbool2_t vm, vfloat16m8_t vd,
                               const _Float16 *rs1, ptrdiff_t rs2, size_t vl);
vfloat32mf2_t __riscv_vlse32_mu(vbool64_t vm, vfloat32mf2_t vd,
                                const float *rs1, ptrdiff_t rs2, size_t vl);
vfloat32m1_t __riscv_vlse32_mu(vbool32_t vm, vfloat32m1_t vd, const float *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat32m2_t __riscv_vlse32_mu(vbool16_t vm, vfloat32m2_t vd, const float *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat32m4_t __riscv_vlse32_mu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat32m8_t __riscv_vlse32_mu(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat64m1_t __riscv_vlse64_mu(vbool64_t vm, vfloat64m1_t vd, const double *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat64m2_t __riscv_vlse64_mu(vbool32_t vm, vfloat64m2_t vd, const double *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat64m4_t __riscv_vlse64_mu(vbool16_t vm, vfloat64m4_t vd, const double *rs1,
                               ptrdiff_t rs2, size_t vl);
vfloat64m8_t __riscv_vlse64_mu(vbool8_t vm, vfloat64m8_t vd, const double *rs1,
                               ptrdiff_t rs2, size_t vl);
vint8mf8_t __riscv_vlse8_mu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                            ptrdiff_t rs2, size_t vl);
vint8mf4_t __riscv_vlse8_mu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                            ptrdiff_t rs2, size_t vl);
vint8mf2_t __riscv_vlse8_mu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                            ptrdiff_t rs2, size_t vl);
vint8m1_t __riscv_vlse8_mu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                           ptrdiff_t rs2, size_t vl);
vint8m2_t __riscv_vlse8_mu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                           ptrdiff_t rs2, size_t vl);
vint8m4_t __riscv_vlse8_mu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                           ptrdiff_t rs2, size_t vl);
vint8m8_t __riscv_vlse8_mu(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                           ptrdiff_t rs2, size_t vl);
vint16mf4_t __riscv_vlse16_mu(vbool64_t vm, vint16mf4_t vd, const int16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint16mf2_t __riscv_vlse16_mu(vbool32_t vm, vint16mf2_t vd, const int16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint16m1_t __riscv_vlse16_mu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint16m2_t __riscv_vlse16_mu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint16m4_t __riscv_vlse16_mu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint16m8_t __riscv_vlse16_mu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint32mf2_t __riscv_vlse32_mu(vbool64_t vm, vint32mf2_t vd, const int32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vint32m1_t __riscv_vlse32_mu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint32m2_t __riscv_vlse32_mu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint32m4_t __riscv_vlse32_mu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint32m8_t __riscv_vlse32_mu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint64m1_t __riscv_vlse64_mu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint64m2_t __riscv_vlse64_mu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint64m4_t __riscv_vlse64_mu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vint64m8_t __riscv_vlse64_mu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vuint8mf8_t __riscv_vlse8_mu(vbool64_t vm, vuint8mf8_t vd, const uint8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vuint8mf4_t __riscv_vlse8_mu(vbool32_t vm, vuint8mf4_t vd, const uint8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vuint8mf2_t __riscv_vlse8_mu(vbool16_t vm, vuint8mf2_t vd, const uint8_t *rs1,
                             ptrdiff_t rs2, size_t vl);
vuint8m1_t __riscv_vlse8_mu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                            ptrdiff_t rs2, size_t vl);
vuint8m2_t __riscv_vlse8_mu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                            ptrdiff_t rs2, size_t vl);
vuint8m4_t __riscv_vlse8_mu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                            ptrdiff_t rs2, size_t vl);
vuint8m8_t __riscv_vlse8_mu(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                            ptrdiff_t rs2, size_t vl);
vuint16mf4_t __riscv_vlse16_mu(vbool64_t vm, vuint16mf4_t vd,
                               const uint16_t *rs1, ptrdiff_t rs2, size_t vl);
vuint16mf2_t __riscv_vlse16_mu(vbool32_t vm, vuint16mf2_t vd,
                               const uint16_t *rs1, ptrdiff_t rs2, size_t vl);
vuint16m1_t __riscv_vlse16_mu(vbool16_t vm, vuint16m1_t vd, const uint16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint16m2_t __riscv_vlse16_mu(vbool8_t vm, vuint16m2_t vd, const uint16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint16m4_t __riscv_vlse16_mu(vbool4_t vm, vuint16m4_t vd, const uint16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint16m8_t __riscv_vlse16_mu(vbool2_t vm, vuint16m8_t vd, const uint16_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint32mf2_t __riscv_vlse32_mu(vbool64_t vm, vuint32mf2_t vd,
                               const uint32_t *rs1, ptrdiff_t rs2, size_t vl);
vuint32m1_t __riscv_vlse32_mu(vbool32_t vm, vuint32m1_t vd, const uint32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint32m2_t __riscv_vlse32_mu(vbool16_t vm, vuint32m2_t vd, const uint32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint32m4_t __riscv_vlse32_mu(vbool8_t vm, vuint32m4_t vd, const uint32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint32m8_t __riscv_vlse32_mu(vbool4_t vm, vuint32m8_t vd, const uint32_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint64m1_t __riscv_vlse64_mu(vbool64_t vm, vuint64m1_t vd, const uint64_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint64m2_t __riscv_vlse64_mu(vbool32_t vm, vuint64m2_t vd, const uint64_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint64m4_t __riscv_vlse64_mu(vbool16_t vm, vuint64m4_t vd, const uint64_t *rs1,
                              ptrdiff_t rs2, size_t vl);
vuint64m8_t __riscv_vlse64_mu(vbool8_t vm, vuint64m8_t vd, const uint64_t *rs1,
                              ptrdiff_t rs2, size_t vl);

Vector Strided Store Intrinsics

Intrinsics here don’t have a policy variant.

Vector Indexed Load Intrinsics

vfloat16mf4_t __riscv_vloxei8_tu(vfloat16mf4_t vd, const _Float16 *rs1,
                                 vuint8mf8_t rs2, size_t vl);
vfloat16mf2_t __riscv_vloxei8_tu(vfloat16mf2_t vd, const _Float16 *rs1,
                                 vuint8mf4_t rs2, size_t vl);
vfloat16m1_t __riscv_vloxei8_tu(vfloat16m1_t vd, const _Float16 *rs1,
                                vuint8mf2_t rs2, size_t vl);
vfloat16m2_t __riscv_vloxei8_tu(vfloat16m2_t vd, const _Float16 *rs1,
                                vuint8m1_t rs2, size_t vl);
vfloat16m4_t __riscv_vloxei8_tu(vfloat16m4_t vd, const _Float16 *rs1,
                                vuint8m2_t rs2, size_t vl);
vfloat16m8_t __riscv_vloxei8_tu(vfloat16m8_t vd, const _Float16 *rs1,
                                vuint8m4_t rs2, size_t vl);
vfloat16mf4_t __riscv_vloxei16_tu(vfloat16mf4_t vd, const _Float16 *rs1,
                                  vuint16mf4_t rs2, size_t vl);
vfloat16mf2_t __riscv_vloxei16_tu(vfloat16mf2_t vd, const _Float16 *rs1,
                                  vuint16mf2_t rs2, size_t vl);
vfloat16m1_t __riscv_vloxei16_tu(vfloat16m1_t vd, const _Float16 *rs1,
                                 vuint16m1_t rs2, size_t vl);
vfloat16m2_t __riscv_vloxei16_tu(vfloat16m2_t vd, const _Float16 *rs1,
                                 vuint16m2_t rs2, size_t vl);
vfloat16m4_t __riscv_vloxei16_tu(vfloat16m4_t vd, const _Float16 *rs1,
                                 vuint16m4_t rs2, size_t vl);
vfloat16m8_t __riscv_vloxei16_tu(vfloat16m8_t vd, const _Float16 *rs1,
                                 vuint16m8_t rs2, size_t vl);
vfloat16mf4_t __riscv_vloxei32_tu(vfloat16mf4_t vd, const _Float16 *rs1,
                                  vuint32mf2_t rs2, size_t vl);
vfloat16mf2_t __riscv_vloxei32_tu(vfloat16mf2_t vd, const _Float16 *rs1,
                                  vuint32m1_t rs2, size_t vl);
vfloat16m1_t __riscv_vloxei32_tu(vfloat16m1_t vd, const _Float16 *rs1,
                                 vuint32m2_t rs2, size_t vl);
vfloat16m2_t __riscv_vloxei32_tu(vfloat16m2_t vd, const _Float16 *rs1,
                                 vuint32m4_t rs2, size_t vl);
vfloat16m4_t __riscv_vloxei32_tu(vfloat16m4_t vd, const _Float16 *rs1,
                                 vuint32m8_t rs2, size_t vl);
vfloat16mf4_t __riscv_vloxei64_tu(vfloat16mf4_t vd, const _Float16 *rs1,
                                  vuint64m1_t rs2, size_t vl);
vfloat16mf2_t __riscv_vloxei64_tu(vfloat16mf2_t vd, const _Float16 *rs1,
                                  vuint64m2_t rs2, size_t vl);
vfloat16m1_t __riscv_vloxei64_tu(vfloat16m1_t vd, const _Float16 *rs1,
                                 vuint64m4_t rs2, size_t vl);
vfloat16m2_t __riscv_vloxei64_tu(vfloat16m2_t vd, const _Float16 *rs1,
                                 vuint64m8_t rs2, size_t vl);
vfloat32mf2_t __riscv_vloxei8_tu(vfloat32mf2_t vd, const float *rs1,
                                 vuint8mf8_t rs2, size_t vl);
vfloat32m1_t __riscv_vloxei8_tu(vfloat32m1_t vd, const float *rs1,
                                vuint8mf4_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei8_tu(vfloat32m2_t vd, const float *rs1,
                                vuint8mf2_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei8_tu(vfloat32m4_t vd, const float *rs1,
                                vuint8m1_t rs2, size_t vl);
vfloat32m8_t __riscv_vloxei8_tu(vfloat32m8_t vd, const float *rs1,
                                vuint8m2_t rs2, size_t vl);
vfloat32mf2_t __riscv_vloxei16_tu(vfloat32mf2_t vd, const float *rs1,
                                  vuint16mf4_t rs2, size_t vl);
vfloat32m1_t __riscv_vloxei16_tu(vfloat32m1_t vd, const float *rs1,
                                 vuint16mf2_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei16_tu(vfloat32m2_t vd, const float *rs1,
                                 vuint16m1_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei16_tu(vfloat32m4_t vd, const float *rs1,
                                 vuint16m2_t rs2, size_t vl);
vfloat32m8_t __riscv_vloxei16_tu(vfloat32m8_t vd, const float *rs1,
                                 vuint16m4_t rs2, size_t vl);
vfloat32mf2_t __riscv_vloxei32_tu(vfloat32mf2_t vd, const float *rs1,
                                  vuint32mf2_t rs2, size_t vl);
vfloat32m1_t __riscv_vloxei32_tu(vfloat32m1_t vd, const float *rs1,
                                 vuint32m1_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei32_tu(vfloat32m2_t vd, const float *rs1,
                                 vuint32m2_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei32_tu(vfloat32m4_t vd, const float *rs1,
                                 vuint32m4_t rs2, size_t vl);
vfloat32m8_t __riscv_vloxei32_tu(vfloat32m8_t vd, const float *rs1,
                                 vuint32m8_t rs2, size_t vl);
vfloat32mf2_t __riscv_vloxei64_tu(vfloat32mf2_t vd, const float *rs1,
                                  vuint64m1_t rs2, size_t vl);
vfloat32m1_t __riscv_vloxei64_tu(vfloat32m1_t vd, const float *rs1,
                                 vuint64m2_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei64_tu(vfloat32m2_t vd, const float *rs1,
                                 vuint64m4_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei64_tu(vfloat32m4_t vd, const float *rs1,
                                 vuint64m8_t rs2, size_t vl);
vfloat64m1_t __riscv_vloxei8_tu(vfloat64m1_t vd, const double *rs1,
                                vuint8mf8_t rs2, size_t vl);
vfloat64m2_t __riscv_vloxei8_tu(vfloat64m2_t vd, const double *rs1,
                                vuint8mf4_t rs2, size_t vl);
vfloat64m4_t __riscv_vloxei8_tu(vfloat64m4_t vd, const double *rs1,
                                vuint8mf2_t rs2, size_t vl);
vfloat64m8_t __riscv_vloxei8_tu(vfloat64m8_t vd, const double *rs1,
                                vuint8m1_t rs2, size_t vl);
vfloat64m1_t __riscv_vloxei16_tu(vfloat64m1_t vd, const double *rs1,
                                 vuint16mf4_t rs2, size_t vl);
vfloat64m2_t __riscv_vloxei16_tu(vfloat64m2_t vd, const double *rs1,
                                 vuint16mf2_t rs2, size_t vl);
vfloat64m4_t __riscv_vloxei16_tu(vfloat64m4_t vd, const double *rs1,
                                 vuint16m1_t rs2, size_t vl);
vfloat64m8_t __riscv_vloxei16_tu(vfloat64m8_t vd, const double *rs1,
                                 vuint16m2_t rs2, size_t vl);
vfloat64m1_t __riscv_vloxei32_tu(vfloat64m1_t vd, const double *rs1,
                                 vuint32mf2_t rs2, size_t vl);
vfloat64m2_t __riscv_vloxei32_tu(vfloat64m2_t vd, const double *rs1,
                                 vuint32m1_t rs2, size_t vl);
vfloat64m4_t __riscv_vloxei32_tu(vfloat64m4_t vd, const double *rs1,
                                 vuint32m2_t rs2, size_t vl);
vfloat64m8_t __riscv_vloxei32_tu(vfloat64m8_t vd, const double *rs1,
                                 vuint32m4_t rs2, size_t vl);
vfloat64m1_t __riscv_vloxei64_tu(vfloat64m1_t vd, const double *rs1,
                                 vuint64m1_t rs2, size_t vl);
vfloat64m2_t __riscv_vloxei64_tu(vfloat64m2_t vd, const double *rs1,
                                 vuint64m2_t rs2, size_t vl);
vfloat64m4_t __riscv_vloxei64_tu(vfloat64m4_t vd, const double *rs1,
                                 vuint64m4_t rs2, size_t vl);
vfloat64m8_t __riscv_vloxei64_tu(vfloat64m8_t vd, const double *rs1,
                                 vuint64m8_t rs2, size_t vl);
vfloat16mf4_t __riscv_vluxei8_tu(vfloat16mf4_t vd, const _Float16 *rs1,
                                 vuint8mf8_t rs2, size_t vl);
vfloat16mf2_t __riscv_vluxei8_tu(vfloat16mf2_t vd, const _Float16 *rs1,
                                 vuint8mf4_t rs2, size_t vl);
vfloat16m1_t __riscv_vluxei8_tu(vfloat16m1_t vd, const _Float16 *rs1,
                                vuint8mf2_t rs2, size_t vl);
vfloat16m2_t __riscv_vluxei8_tu(vfloat16m2_t vd, const _Float16 *rs1,
                                vuint8m1_t rs2, size_t vl);
vfloat16m4_t __riscv_vluxei8_tu(vfloat16m4_t vd, const _Float16 *rs1,
                                vuint8m2_t rs2, size_t vl);
vfloat16m8_t __riscv_vluxei8_tu(vfloat16m8_t vd, const _Float16 *rs1,
                                vuint8m4_t rs2, size_t vl);
vfloat16mf4_t __riscv_vluxei16_tu(vfloat16mf4_t vd, const _Float16 *rs1,
                                  vuint16mf4_t rs2, size_t vl);
vfloat16mf2_t __riscv_vluxei16_tu(vfloat16mf2_t vd, const _Float16 *rs1,
                                  vuint16mf2_t rs2, size_t vl);
vfloat16m1_t __riscv_vluxei16_tu(vfloat16m1_t vd, const _Float16 *rs1,
                                 vuint16m1_t rs2, size_t vl);
vfloat16m2_t __riscv_vluxei16_tu(vfloat16m2_t vd, const _Float16 *rs1,
                                 vuint16m2_t rs2, size_t vl);
vfloat16m4_t __riscv_vluxei16_tu(vfloat16m4_t vd, const _Float16 *rs1,
                                 vuint16m4_t rs2, size_t vl);
vfloat16m8_t __riscv_vluxei16_tu(vfloat16m8_t vd, const _Float16 *rs1,
                                 vuint16m8_t rs2, size_t vl);
vfloat16mf4_t __riscv_vluxei32_tu(vfloat16mf4_t vd, const _Float16 *rs1,
                                  vuint32mf2_t rs2, size_t vl);
vfloat16mf2_t __riscv_vluxei32_tu(vfloat16mf2_t vd, const _Float16 *rs1,
                                  vuint32m1_t rs2, size_t vl);
vfloat16m1_t __riscv_vluxei32_tu(vfloat16m1_t vd, const _Float16 *rs1,
                                 vuint32m2_t rs2, size_t vl);
vfloat16m2_t __riscv_vluxei32_tu(vfloat16m2_t vd, const _Float16 *rs1,
                                 vuint32m4_t rs2, size_t vl);
vfloat16m4_t __riscv_vluxei32_tu(vfloat16m4_t vd, const _Float16 *rs1,
                                 vuint32m8_t rs2, size_t vl);
vfloat16mf4_t __riscv_vluxei64_tu(vfloat16mf4_t vd, const _Float16 *rs1,
                                  vuint64m1_t rs2, size_t vl);
vfloat16mf2_t __riscv_vluxei64_tu(vfloat16mf2_t vd, const _Float16 *rs1,
                                  vuint64m2_t rs2, size_t vl);
vfloat16m1_t __riscv_vluxei64_tu(vfloat16m1_t vd, const _Float16 *rs1,
                                 vuint64m4_t rs2, size_t vl);
vfloat16m2_t __riscv_vluxei64_tu(vfloat16m2_t vd, const _Float16 *rs1,
                                 vuint64m8_t rs2, size_t vl);
vfloat32mf2_t __riscv_vluxei8_tu(vfloat32mf2_t vd, const float *rs1,
                                 vuint8mf8_t rs2, size_t vl);
vfloat32m1_t __riscv_vluxei8_tu(vfloat32m1_t vd, const float *rs1,
                                vuint8mf4_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei8_tu(vfloat32m2_t vd, const float *rs1,
                                vuint8mf2_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei8_tu(vfloat32m4_t vd, const float *rs1,
                                vuint8m1_t rs2, size_t vl);
vfloat32m8_t __riscv_vluxei8_tu(vfloat32m8_t vd, const float *rs1,
                                vuint8m2_t rs2, size_t vl);
vfloat32mf2_t __riscv_vluxei16_tu(vfloat32mf2_t vd, const float *rs1,
                                  vuint16mf4_t rs2, size_t vl);
vfloat32m1_t __riscv_vluxei16_tu(vfloat32m1_t vd, const float *rs1,
                                 vuint16mf2_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei16_tu(vfloat32m2_t vd, const float *rs1,
                                 vuint16m1_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei16_tu(vfloat32m4_t vd, const float *rs1,
                                 vuint16m2_t rs2, size_t vl);
vfloat32m8_t __riscv_vluxei16_tu(vfloat32m8_t vd, const float *rs1,
                                 vuint16m4_t rs2, size_t vl);
vfloat32mf2_t __riscv_vluxei32_tu(vfloat32mf2_t vd, const float *rs1,
                                  vuint32mf2_t rs2, size_t vl);
vfloat32m1_t __riscv_vluxei32_tu(vfloat32m1_t vd, const float *rs1,
                                 vuint32m1_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei32_tu(vfloat32m2_t vd, const float *rs1,
                                 vuint32m2_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei32_tu(vfloat32m4_t vd, const float *rs1,
                                 vuint32m4_t rs2, size_t vl);
vfloat32m8_t __riscv_vluxei32_tu(vfloat32m8_t vd, const float *rs1,
                                 vuint32m8_t rs2, size_t vl);
vfloat32mf2_t __riscv_vluxei64_tu(vfloat32mf2_t vd, const float *rs1,
                                  vuint64m1_t rs2, size_t vl);
vfloat32m1_t __riscv_vluxei64_tu(vfloat32m1_t vd, const float *rs1,
                                 vuint64m2_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei64_tu(vfloat32m2_t vd, const float *rs1,
                                 vuint64m4_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei64_tu(vfloat32m4_t vd, const float *rs1,
                                 vuint64m8_t rs2, size_t vl);
vfloat64m1_t __riscv_vluxei8_tu(vfloat64m1_t vd, const double *rs1,
                                vuint8mf8_t rs2, size_t vl);
vfloat64m2_t __riscv_vluxei8_tu(vfloat64m2_t vd, const double *rs1,
                                vuint8mf4_t rs2, size_t vl);
vfloat64m4_t __riscv_vluxei8_tu(vfloat64m4_t vd, const double *rs1,
                                vuint8mf2_t rs2, size_t vl);
vfloat64m8_t __riscv_vluxei8_tu(vfloat64m8_t vd, const double *rs1,
                                vuint8m1_t rs2, size_t vl);
vfloat64m1_t __riscv_vluxei16_tu(vfloat64m1_t vd, const double *rs1,
                                 vuint16mf4_t rs2, size_t vl);
vfloat64m2_t __riscv_vluxei16_tu(vfloat64m2_t vd, const double *rs1,
                                 vuint16mf2_t rs2, size_t vl);
vfloat64m4_t __riscv_vluxei16_tu(vfloat64m4_t vd, const double *rs1,
                                 vuint16m1_t rs2, size_t vl);
vfloat64m8_t __riscv_vluxei16_tu(vfloat64m8_t vd, const double *rs1,
                                 vuint16m2_t rs2, size_t vl);
vfloat64m1_t __riscv_vluxei32_tu(vfloat64m1_t vd, const double *rs1,
                                 vuint32mf2_t rs2, size_t vl);
vfloat64m2_t __riscv_vluxei32_tu(vfloat64m2_t vd, const double *rs1,
                                 vuint32m1_t rs2, size_t vl);
vfloat64m4_t __riscv_vluxei32_tu(vfloat64m4_t vd, const double *rs1,
                                 vuint32m2_t rs2, size_t vl);
vfloat64m8_t __riscv_vluxei32_tu(vfloat64m8_t vd, const double *rs1,
                                 vuint32m4_t rs2, size_t vl);
vfloat64m1_t __riscv_vluxei64_tu(vfloat64m1_t vd, const double *rs1,
                                 vuint64m1_t rs2, size_t vl);
vfloat64m2_t __riscv_vluxei64_tu(vfloat64m2_t vd, const double *rs1,
                                 vuint64m2_t rs2, size_t vl);
vfloat64m4_t __riscv_vluxei64_tu(vfloat64m4_t vd, const double *rs1,
                                 vuint64m4_t rs2, size_t vl);
vfloat64m8_t __riscv_vluxei64_tu(vfloat64m8_t vd, const double *rs1,
                                 vuint64m8_t rs2, size_t vl);
vint8mf8_t __riscv_vloxei8_tu(vint8mf8_t vd, const int8_t *rs1, vuint8mf8_t rs2,
                              size_t vl);
vint8mf4_t __riscv_vloxei8_tu(vint8mf4_t vd, const int8_t *rs1, vuint8mf4_t rs2,
                              size_t vl);
vint8mf2_t __riscv_vloxei8_tu(vint8mf2_t vd, const int8_t *rs1, vuint8mf2_t rs2,
                              size_t vl);
vint8m1_t __riscv_vloxei8_tu(vint8m1_t vd, const int8_t *rs1, vuint8m1_t rs2,
                             size_t vl);
vint8m2_t __riscv_vloxei8_tu(vint8m2_t vd, const int8_t *rs1, vuint8m2_t rs2,
                             size_t vl);
vint8m4_t __riscv_vloxei8_tu(vint8m4_t vd, const int8_t *rs1, vuint8m4_t rs2,
                             size_t vl);
vint8m8_t __riscv_vloxei8_tu(vint8m8_t vd, const int8_t *rs1, vuint8m8_t rs2,
                             size_t vl);
vint8mf8_t __riscv_vloxei16_tu(vint8mf8_t vd, const int8_t *rs1,
                               vuint16mf4_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei16_tu(vint8mf4_t vd, const int8_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei16_tu(vint8mf2_t vd, const int8_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint8m1_t __riscv_vloxei16_tu(vint8m1_t vd, const int8_t *rs1, vuint16m2_t rs2,
                              size_t vl);
vint8m2_t __riscv_vloxei16_tu(vint8m2_t vd, const int8_t *rs1, vuint16m4_t rs2,
                              size_t vl);
vint8m4_t __riscv_vloxei16_tu(vint8m4_t vd, const int8_t *rs1, vuint16m8_t rs2,
                              size_t vl);
vint8mf8_t __riscv_vloxei32_tu(vint8mf8_t vd, const int8_t *rs1,
                               vuint32mf2_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei32_tu(vint8mf4_t vd, const int8_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei32_tu(vint8mf2_t vd, const int8_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint8m1_t __riscv_vloxei32_tu(vint8m1_t vd, const int8_t *rs1, vuint32m4_t rs2,
                              size_t vl);
vint8m2_t __riscv_vloxei32_tu(vint8m2_t vd, const int8_t *rs1, vuint32m8_t rs2,
                              size_t vl);
vint8mf8_t __riscv_vloxei64_tu(vint8mf8_t vd, const int8_t *rs1,
                               vuint64m1_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei64_tu(vint8mf4_t vd, const int8_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei64_tu(vint8mf2_t vd, const int8_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint8m1_t __riscv_vloxei64_tu(vint8m1_t vd, const int8_t *rs1, vuint64m8_t rs2,
                              size_t vl);
vint16mf4_t __riscv_vloxei8_tu(vint16mf4_t vd, const int16_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint16mf2_t __riscv_vloxei8_tu(vint16mf2_t vd, const int16_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vint16m1_t __riscv_vloxei8_tu(vint16m1_t vd, const int16_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint16m2_t __riscv_vloxei8_tu(vint16m2_t vd, const int16_t *rs1, vuint8m1_t rs2,
                              size_t vl);
vint16m4_t __riscv_vloxei8_tu(vint16m4_t vd, const int16_t *rs1, vuint8m2_t rs2,
                              size_t vl);
vint16m8_t __riscv_vloxei8_tu(vint16m8_t vd, const int16_t *rs1, vuint8m4_t rs2,
                              size_t vl);
vint16mf4_t __riscv_vloxei16_tu(vint16mf4_t vd, const int16_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vint16mf2_t __riscv_vloxei16_tu(vint16mf2_t vd, const int16_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vint16m1_t __riscv_vloxei16_tu(vint16m1_t vd, const int16_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint16m2_t __riscv_vloxei16_tu(vint16m2_t vd, const int16_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint16m4_t __riscv_vloxei16_tu(vint16m4_t vd, const int16_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vint16m8_t __riscv_vloxei16_tu(vint16m8_t vd, const int16_t *rs1,
                               vuint16m8_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei32_tu(vint16mf4_t vd, const int16_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vint16mf2_t __riscv_vloxei32_tu(vint16mf2_t vd, const int16_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vint16m1_t __riscv_vloxei32_tu(vint16m1_t vd, const int16_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint16m2_t __riscv_vloxei32_tu(vint16m2_t vd, const int16_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint16m4_t __riscv_vloxei32_tu(vint16m4_t vd, const int16_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei64_tu(vint16mf4_t vd, const int16_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vint16mf2_t __riscv_vloxei64_tu(vint16mf2_t vd, const int16_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vint16m1_t __riscv_vloxei64_tu(vint16m1_t vd, const int16_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint16m2_t __riscv_vloxei64_tu(vint16m2_t vd, const int16_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei8_tu(vint32mf2_t vd, const int32_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint32m1_t __riscv_vloxei8_tu(vint32m1_t vd, const int32_t *rs1,
                              vuint8mf4_t rs2, size_t vl);
vint32m2_t __riscv_vloxei8_tu(vint32m2_t vd, const int32_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint32m4_t __riscv_vloxei8_tu(vint32m4_t vd, const int32_t *rs1, vuint8m1_t rs2,
                              size_t vl);
vint32m8_t __riscv_vloxei8_tu(vint32m8_t vd, const int32_t *rs1, vuint8m2_t rs2,
                              size_t vl);
vint32mf2_t __riscv_vloxei16_tu(vint32mf2_t vd, const int32_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vint32m1_t __riscv_vloxei16_tu(vint32m1_t vd, const int32_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint32m2_t __riscv_vloxei16_tu(vint32m2_t vd, const int32_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint32m4_t __riscv_vloxei16_tu(vint32m4_t vd, const int32_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint32m8_t __riscv_vloxei16_tu(vint32m8_t vd, const int32_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei32_tu(vint32mf2_t vd, const int32_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vint32m1_t __riscv_vloxei32_tu(vint32m1_t vd, const int32_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint32m2_t __riscv_vloxei32_tu(vint32m2_t vd, const int32_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint32m4_t __riscv_vloxei32_tu(vint32m4_t vd, const int32_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint32m8_t __riscv_vloxei32_tu(vint32m8_t vd, const int32_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei64_tu(vint32mf2_t vd, const int32_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vint32m1_t __riscv_vloxei64_tu(vint32m1_t vd, const int32_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint32m2_t __riscv_vloxei64_tu(vint32m2_t vd, const int32_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint32m4_t __riscv_vloxei64_tu(vint32m4_t vd, const int32_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint64m1_t __riscv_vloxei8_tu(vint64m1_t vd, const int64_t *rs1,
                              vuint8mf8_t rs2, size_t vl);
vint64m2_t __riscv_vloxei8_tu(vint64m2_t vd, const int64_t *rs1,
                              vuint8mf4_t rs2, size_t vl);
vint64m4_t __riscv_vloxei8_tu(vint64m4_t vd, const int64_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint64m8_t __riscv_vloxei8_tu(vint64m8_t vd, const int64_t *rs1, vuint8m1_t rs2,
                              size_t vl);
vint64m1_t __riscv_vloxei16_tu(vint64m1_t vd, const int64_t *rs1,
                               vuint16mf4_t rs2, size_t vl);
vint64m2_t __riscv_vloxei16_tu(vint64m2_t vd, const int64_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint64m4_t __riscv_vloxei16_tu(vint64m4_t vd, const int64_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint64m8_t __riscv_vloxei16_tu(vint64m8_t vd, const int64_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint64m1_t __riscv_vloxei32_tu(vint64m1_t vd, const int64_t *rs1,
                               vuint32mf2_t rs2, size_t vl);
vint64m2_t __riscv_vloxei32_tu(vint64m2_t vd, const int64_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint64m4_t __riscv_vloxei32_tu(vint64m4_t vd, const int64_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint64m8_t __riscv_vloxei32_tu(vint64m8_t vd, const int64_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint64m1_t __riscv_vloxei64_tu(vint64m1_t vd, const int64_t *rs1,
                               vuint64m1_t rs2, size_t vl);
vint64m2_t __riscv_vloxei64_tu(vint64m2_t vd, const int64_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint64m4_t __riscv_vloxei64_tu(vint64m4_t vd, const int64_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint64m8_t __riscv_vloxei64_tu(vint64m8_t vd, const int64_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei8_tu(vint8mf8_t vd, const int8_t *rs1, vuint8mf8_t rs2,
                              size_t vl);
vint8mf4_t __riscv_vluxei8_tu(vint8mf4_t vd, const int8_t *rs1, vuint8mf4_t rs2,
                              size_t vl);
vint8mf2_t __riscv_vluxei8_tu(vint8mf2_t vd, const int8_t *rs1, vuint8mf2_t rs2,
                              size_t vl);
vint8m1_t __riscv_vluxei8_tu(vint8m1_t vd, const int8_t *rs1, vuint8m1_t rs2,
                             size_t vl);
vint8m2_t __riscv_vluxei8_tu(vint8m2_t vd, const int8_t *rs1, vuint8m2_t rs2,
                             size_t vl);
vint8m4_t __riscv_vluxei8_tu(vint8m4_t vd, const int8_t *rs1, vuint8m4_t rs2,
                             size_t vl);
vint8m8_t __riscv_vluxei8_tu(vint8m8_t vd, const int8_t *rs1, vuint8m8_t rs2,
                             size_t vl);
vint8mf8_t __riscv_vluxei16_tu(vint8mf8_t vd, const int8_t *rs1,
                               vuint16mf4_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei16_tu(vint8mf4_t vd, const int8_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei16_tu(vint8mf2_t vd, const int8_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint8m1_t __riscv_vluxei16_tu(vint8m1_t vd, const int8_t *rs1, vuint16m2_t rs2,
                              size_t vl);
vint8m2_t __riscv_vluxei16_tu(vint8m2_t vd, const int8_t *rs1, vuint16m4_t rs2,
                              size_t vl);
vint8m4_t __riscv_vluxei16_tu(vint8m4_t vd, const int8_t *rs1, vuint16m8_t rs2,
                              size_t vl);
vint8mf8_t __riscv_vluxei32_tu(vint8mf8_t vd, const int8_t *rs1,
                               vuint32mf2_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei32_tu(vint8mf4_t vd, const int8_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei32_tu(vint8mf2_t vd, const int8_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint8m1_t __riscv_vluxei32_tu(vint8m1_t vd, const int8_t *rs1, vuint32m4_t rs2,
                              size_t vl);
vint8m2_t __riscv_vluxei32_tu(vint8m2_t vd, const int8_t *rs1, vuint32m8_t rs2,
                              size_t vl);
vint8mf8_t __riscv_vluxei64_tu(vint8mf8_t vd, const int8_t *rs1,
                               vuint64m1_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei64_tu(vint8mf4_t vd, const int8_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei64_tu(vint8mf2_t vd, const int8_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint8m1_t __riscv_vluxei64_tu(vint8m1_t vd, const int8_t *rs1, vuint64m8_t rs2,
                              size_t vl);
vint16mf4_t __riscv_vluxei8_tu(vint16mf4_t vd, const int16_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint16mf2_t __riscv_vluxei8_tu(vint16mf2_t vd, const int16_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vint16m1_t __riscv_vluxei8_tu(vint16m1_t vd, const int16_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint16m2_t __riscv_vluxei8_tu(vint16m2_t vd, const int16_t *rs1, vuint8m1_t rs2,
                              size_t vl);
vint16m4_t __riscv_vluxei8_tu(vint16m4_t vd, const int16_t *rs1, vuint8m2_t rs2,
                              size_t vl);
vint16m8_t __riscv_vluxei8_tu(vint16m8_t vd, const int16_t *rs1, vuint8m4_t rs2,
                              size_t vl);
vint16mf4_t __riscv_vluxei16_tu(vint16mf4_t vd, const int16_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vint16mf2_t __riscv_vluxei16_tu(vint16mf2_t vd, const int16_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vint16m1_t __riscv_vluxei16_tu(vint16m1_t vd, const int16_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint16m2_t __riscv_vluxei16_tu(vint16m2_t vd, const int16_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint16m4_t __riscv_vluxei16_tu(vint16m4_t vd, const int16_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vint16m8_t __riscv_vluxei16_tu(vint16m8_t vd, const int16_t *rs1,
                               vuint16m8_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei32_tu(vint16mf4_t vd, const int16_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vint16mf2_t __riscv_vluxei32_tu(vint16mf2_t vd, const int16_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vint16m1_t __riscv_vluxei32_tu(vint16m1_t vd, const int16_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint16m2_t __riscv_vluxei32_tu(vint16m2_t vd, const int16_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint16m4_t __riscv_vluxei32_tu(vint16m4_t vd, const int16_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei64_tu(vint16mf4_t vd, const int16_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vint16mf2_t __riscv_vluxei64_tu(vint16mf2_t vd, const int16_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vint16m1_t __riscv_vluxei64_tu(vint16m1_t vd, const int16_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint16m2_t __riscv_vluxei64_tu(vint16m2_t vd, const int16_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei8_tu(vint32mf2_t vd, const int32_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint32m1_t __riscv_vluxei8_tu(vint32m1_t vd, const int32_t *rs1,
                              vuint8mf4_t rs2, size_t vl);
vint32m2_t __riscv_vluxei8_tu(vint32m2_t vd, const int32_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint32m4_t __riscv_vluxei8_tu(vint32m4_t vd, const int32_t *rs1, vuint8m1_t rs2,
                              size_t vl);
vint32m8_t __riscv_vluxei8_tu(vint32m8_t vd, const int32_t *rs1, vuint8m2_t rs2,
                              size_t vl);
vint32mf2_t __riscv_vluxei16_tu(vint32mf2_t vd, const int32_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vint32m1_t __riscv_vluxei16_tu(vint32m1_t vd, const int32_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint32m2_t __riscv_vluxei16_tu(vint32m2_t vd, const int32_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint32m4_t __riscv_vluxei16_tu(vint32m4_t vd, const int32_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint32m8_t __riscv_vluxei16_tu(vint32m8_t vd, const int32_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei32_tu(vint32mf2_t vd, const int32_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vint32m1_t __riscv_vluxei32_tu(vint32m1_t vd, const int32_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint32m2_t __riscv_vluxei32_tu(vint32m2_t vd, const int32_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint32m4_t __riscv_vluxei32_tu(vint32m4_t vd, const int32_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint32m8_t __riscv_vluxei32_tu(vint32m8_t vd, const int32_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei64_tu(vint32mf2_t vd, const int32_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vint32m1_t __riscv_vluxei64_tu(vint32m1_t vd, const int32_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint32m2_t __riscv_vluxei64_tu(vint32m2_t vd, const int32_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint32m4_t __riscv_vluxei64_tu(vint32m4_t vd, const int32_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint64m1_t __riscv_vluxei8_tu(vint64m1_t vd, const int64_t *rs1,
                              vuint8mf8_t rs2, size_t vl);
vint64m2_t __riscv_vluxei8_tu(vint64m2_t vd, const int64_t *rs1,
                              vuint8mf4_t rs2, size_t vl);
vint64m4_t __riscv_vluxei8_tu(vint64m4_t vd, const int64_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint64m8_t __riscv_vluxei8_tu(vint64m8_t vd, const int64_t *rs1, vuint8m1_t rs2,
                              size_t vl);
vint64m1_t __riscv_vluxei16_tu(vint64m1_t vd, const int64_t *rs1,
                               vuint16mf4_t rs2, size_t vl);
vint64m2_t __riscv_vluxei16_tu(vint64m2_t vd, const int64_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint64m4_t __riscv_vluxei16_tu(vint64m4_t vd, const int64_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint64m8_t __riscv_vluxei16_tu(vint64m8_t vd, const int64_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint64m1_t __riscv_vluxei32_tu(vint64m1_t vd, const int64_t *rs1,
                               vuint32mf2_t rs2, size_t vl);
vint64m2_t __riscv_vluxei32_tu(vint64m2_t vd, const int64_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint64m4_t __riscv_vluxei32_tu(vint64m4_t vd, const int64_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint64m8_t __riscv_vluxei32_tu(vint64m8_t vd, const int64_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint64m1_t __riscv_vluxei64_tu(vint64m1_t vd, const int64_t *rs1,
                               vuint64m1_t rs2, size_t vl);
vint64m2_t __riscv_vluxei64_tu(vint64m2_t vd, const int64_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint64m4_t __riscv_vluxei64_tu(vint64m4_t vd, const int64_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint64m8_t __riscv_vluxei64_tu(vint64m8_t vd, const int64_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei8_tu(vuint8mf8_t vd, const uint8_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vuint8mf4_t __riscv_vloxei8_tu(vuint8mf4_t vd, const uint8_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vuint8mf2_t __riscv_vloxei8_tu(vuint8mf2_t vd, const uint8_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vuint8m1_t __riscv_vloxei8_tu(vuint8m1_t vd, const uint8_t *rs1, vuint8m1_t rs2,
                              size_t vl);
vuint8m2_t __riscv_vloxei8_tu(vuint8m2_t vd, const uint8_t *rs1, vuint8m2_t rs2,
                              size_t vl);
vuint8m4_t __riscv_vloxei8_tu(vuint8m4_t vd, const uint8_t *rs1, vuint8m4_t rs2,
                              size_t vl);
vuint8m8_t __riscv_vloxei8_tu(vuint8m8_t vd, const uint8_t *rs1, vuint8m8_t rs2,
                              size_t vl);
vuint8mf8_t __riscv_vloxei16_tu(vuint8mf8_t vd, const uint8_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vuint8mf4_t __riscv_vloxei16_tu(vuint8mf4_t vd, const uint8_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vuint8mf2_t __riscv_vloxei16_tu(vuint8mf2_t vd, const uint8_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vuint8m1_t __riscv_vloxei16_tu(vuint8m1_t vd, const uint8_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vuint8m2_t __riscv_vloxei16_tu(vuint8m2_t vd, const uint8_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vuint8m4_t __riscv_vloxei16_tu(vuint8m4_t vd, const uint8_t *rs1,
                               vuint16m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei32_tu(vuint8mf8_t vd, const uint8_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vuint8mf4_t __riscv_vloxei32_tu(vuint8mf4_t vd, const uint8_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vuint8mf2_t __riscv_vloxei32_tu(vuint8mf2_t vd, const uint8_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vuint8m1_t __riscv_vloxei32_tu(vuint8m1_t vd, const uint8_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vuint8m2_t __riscv_vloxei32_tu(vuint8m2_t vd, const uint8_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei64_tu(vuint8mf8_t vd, const uint8_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vuint8mf4_t __riscv_vloxei64_tu(vuint8mf4_t vd, const uint8_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vuint8mf2_t __riscv_vloxei64_tu(vuint8mf2_t vd, const uint8_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vuint8m1_t __riscv_vloxei64_tu(vuint8m1_t vd, const uint8_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vloxei8_tu(vuint16mf4_t vd, const uint16_t *rs1,
                                vuint8mf8_t rs2, size_t vl);
vuint16mf2_t __riscv_vloxei8_tu(vuint16mf2_t vd, const uint16_t *rs1,
                                vuint8mf4_t rs2, size_t vl);
vuint16m1_t __riscv_vloxei8_tu(vuint16m1_t vd, const uint16_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vuint16m2_t __riscv_vloxei8_tu(vuint16m2_t vd, const uint16_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint16m4_t __riscv_vloxei8_tu(vuint16m4_t vd, const uint16_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vuint16m8_t __riscv_vloxei8_tu(vuint16m8_t vd, const uint16_t *rs1,
                               vuint8m4_t rs2, size_t vl);
vuint16mf4_t __riscv_vloxei16_tu(vuint16mf4_t vd, const uint16_t *rs1,
                                 vuint16mf4_t rs2, size_t vl);
vuint16mf2_t __riscv_vloxei16_tu(vuint16mf2_t vd, const uint16_t *rs1,
                                 vuint16mf2_t rs2, size_t vl);
vuint16m1_t __riscv_vloxei16_tu(vuint16m1_t vd, const uint16_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vuint16m2_t __riscv_vloxei16_tu(vuint16m2_t vd, const uint16_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vuint16m4_t __riscv_vloxei16_tu(vuint16m4_t vd, const uint16_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vuint16m8_t __riscv_vloxei16_tu(vuint16m8_t vd, const uint16_t *rs1,
                                vuint16m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vloxei32_tu(vuint16mf4_t vd, const uint16_t *rs1,
                                 vuint32mf2_t rs2, size_t vl);
vuint16mf2_t __riscv_vloxei32_tu(vuint16mf2_t vd, const uint16_t *rs1,
                                 vuint32m1_t rs2, size_t vl);
vuint16m1_t __riscv_vloxei32_tu(vuint16m1_t vd, const uint16_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vuint16m2_t __riscv_vloxei32_tu(vuint16m2_t vd, const uint16_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vuint16m4_t __riscv_vloxei32_tu(vuint16m4_t vd, const uint16_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vloxei64_tu(vuint16mf4_t vd, const uint16_t *rs1,
                                 vuint64m1_t rs2, size_t vl);
vuint16mf2_t __riscv_vloxei64_tu(vuint16mf2_t vd, const uint16_t *rs1,
                                 vuint64m2_t rs2, size_t vl);
vuint16m1_t __riscv_vloxei64_tu(vuint16m1_t vd, const uint16_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vuint16m2_t __riscv_vloxei64_tu(vuint16m2_t vd, const uint16_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vuint32mf2_t __riscv_vloxei8_tu(vuint32mf2_t vd, const uint32_t *rs1,
                                vuint8mf8_t rs2, size_t vl);
vuint32m1_t __riscv_vloxei8_tu(vuint32m1_t vd, const uint32_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vuint32m2_t __riscv_vloxei8_tu(vuint32m2_t vd, const uint32_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vuint32m4_t __riscv_vloxei8_tu(vuint32m4_t vd, const uint32_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint32m8_t __riscv_vloxei8_tu(vuint32m8_t vd, const uint32_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vuint32mf2_t __riscv_vloxei16_tu(vuint32mf2_t vd, const uint32_t *rs1,
                                 vuint16mf4_t rs2, size_t vl);
vuint32m1_t __riscv_vloxei16_tu(vuint32m1_t vd, const uint32_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vuint32m2_t __riscv_vloxei16_tu(vuint32m2_t vd, const uint32_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vuint32m4_t __riscv_vloxei16_tu(vuint32m4_t vd, const uint32_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vuint32m8_t __riscv_vloxei16_tu(vuint32m8_t vd, const uint32_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vuint32mf2_t __riscv_vloxei32_tu(vuint32mf2_t vd, const uint32_t *rs1,
                                 vuint32mf2_t rs2, size_t vl);
vuint32m1_t __riscv_vloxei32_tu(vuint32m1_t vd, const uint32_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vuint32m2_t __riscv_vloxei32_tu(vuint32m2_t vd, const uint32_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vuint32m4_t __riscv_vloxei32_tu(vuint32m4_t vd, const uint32_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vuint32m8_t __riscv_vloxei32_tu(vuint32m8_t vd, const uint32_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vuint32mf2_t __riscv_vloxei64_tu(vuint32mf2_t vd, const uint32_t *rs1,
                                 vuint64m1_t rs2, size_t vl);
vuint32m1_t __riscv_vloxei64_tu(vuint32m1_t vd, const uint32_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vuint32m2_t __riscv_vloxei64_tu(vuint32m2_t vd, const uint32_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vuint32m4_t __riscv_vloxei64_tu(vuint32m4_t vd, const uint32_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vuint64m1_t __riscv_vloxei8_tu(vuint64m1_t vd, const uint64_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vuint64m2_t __riscv_vloxei8_tu(vuint64m2_t vd, const uint64_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vuint64m4_t __riscv_vloxei8_tu(vuint64m4_t vd, const uint64_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vuint64m8_t __riscv_vloxei8_tu(vuint64m8_t vd, const uint64_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint64m1_t __riscv_vloxei16_tu(vuint64m1_t vd, const uint64_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vuint64m2_t __riscv_vloxei16_tu(vuint64m2_t vd, const uint64_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vuint64m4_t __riscv_vloxei16_tu(vuint64m4_t vd, const uint64_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vuint64m8_t __riscv_vloxei16_tu(vuint64m8_t vd, const uint64_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vuint64m1_t __riscv_vloxei32_tu(vuint64m1_t vd, const uint64_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vuint64m2_t __riscv_vloxei32_tu(vuint64m2_t vd, const uint64_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vuint64m4_t __riscv_vloxei32_tu(vuint64m4_t vd, const uint64_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vuint64m8_t __riscv_vloxei32_tu(vuint64m8_t vd, const uint64_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vuint64m1_t __riscv_vloxei64_tu(vuint64m1_t vd, const uint64_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vuint64m2_t __riscv_vloxei64_tu(vuint64m2_t vd, const uint64_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vuint64m4_t __riscv_vloxei64_tu(vuint64m4_t vd, const uint64_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vuint64m8_t __riscv_vloxei64_tu(vuint64m8_t vd, const uint64_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei8_tu(vuint8mf8_t vd, const uint8_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vuint8mf4_t __riscv_vluxei8_tu(vuint8mf4_t vd, const uint8_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vuint8mf2_t __riscv_vluxei8_tu(vuint8mf2_t vd, const uint8_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vuint8m1_t __riscv_vluxei8_tu(vuint8m1_t vd, const uint8_t *rs1, vuint8m1_t rs2,
                              size_t vl);
vuint8m2_t __riscv_vluxei8_tu(vuint8m2_t vd, const uint8_t *rs1, vuint8m2_t rs2,
                              size_t vl);
vuint8m4_t __riscv_vluxei8_tu(vuint8m4_t vd, const uint8_t *rs1, vuint8m4_t rs2,
                              size_t vl);
vuint8m8_t __riscv_vluxei8_tu(vuint8m8_t vd, const uint8_t *rs1, vuint8m8_t rs2,
                              size_t vl);
vuint8mf8_t __riscv_vluxei16_tu(vuint8mf8_t vd, const uint8_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vuint8mf4_t __riscv_vluxei16_tu(vuint8mf4_t vd, const uint8_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vuint8mf2_t __riscv_vluxei16_tu(vuint8mf2_t vd, const uint8_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vuint8m1_t __riscv_vluxei16_tu(vuint8m1_t vd, const uint8_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vuint8m2_t __riscv_vluxei16_tu(vuint8m2_t vd, const uint8_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vuint8m4_t __riscv_vluxei16_tu(vuint8m4_t vd, const uint8_t *rs1,
                               vuint16m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei32_tu(vuint8mf8_t vd, const uint8_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vuint8mf4_t __riscv_vluxei32_tu(vuint8mf4_t vd, const uint8_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vuint8mf2_t __riscv_vluxei32_tu(vuint8mf2_t vd, const uint8_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vuint8m1_t __riscv_vluxei32_tu(vuint8m1_t vd, const uint8_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vuint8m2_t __riscv_vluxei32_tu(vuint8m2_t vd, const uint8_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei64_tu(vuint8mf8_t vd, const uint8_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vuint8mf4_t __riscv_vluxei64_tu(vuint8mf4_t vd, const uint8_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vuint8mf2_t __riscv_vluxei64_tu(vuint8mf2_t vd, const uint8_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vuint8m1_t __riscv_vluxei64_tu(vuint8m1_t vd, const uint8_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vluxei8_tu(vuint16mf4_t vd, const uint16_t *rs1,
                                vuint8mf8_t rs2, size_t vl);
vuint16mf2_t __riscv_vluxei8_tu(vuint16mf2_t vd, const uint16_t *rs1,
                                vuint8mf4_t rs2, size_t vl);
vuint16m1_t __riscv_vluxei8_tu(vuint16m1_t vd, const uint16_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vuint16m2_t __riscv_vluxei8_tu(vuint16m2_t vd, const uint16_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint16m4_t __riscv_vluxei8_tu(vuint16m4_t vd, const uint16_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vuint16m8_t __riscv_vluxei8_tu(vuint16m8_t vd, const uint16_t *rs1,
                               vuint8m4_t rs2, size_t vl);
vuint16mf4_t __riscv_vluxei16_tu(vuint16mf4_t vd, const uint16_t *rs1,
                                 vuint16mf4_t rs2, size_t vl);
vuint16mf2_t __riscv_vluxei16_tu(vuint16mf2_t vd, const uint16_t *rs1,
                                 vuint16mf2_t rs2, size_t vl);
vuint16m1_t __riscv_vluxei16_tu(vuint16m1_t vd, const uint16_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vuint16m2_t __riscv_vluxei16_tu(vuint16m2_t vd, const uint16_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vuint16m4_t __riscv_vluxei16_tu(vuint16m4_t vd, const uint16_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vuint16m8_t __riscv_vluxei16_tu(vuint16m8_t vd, const uint16_t *rs1,
                                vuint16m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vluxei32_tu(vuint16mf4_t vd, const uint16_t *rs1,
                                 vuint32mf2_t rs2, size_t vl);
vuint16mf2_t __riscv_vluxei32_tu(vuint16mf2_t vd, const uint16_t *rs1,
                                 vuint32m1_t rs2, size_t vl);
vuint16m1_t __riscv_vluxei32_tu(vuint16m1_t vd, const uint16_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vuint16m2_t __riscv_vluxei32_tu(vuint16m2_t vd, const uint16_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vuint16m4_t __riscv_vluxei32_tu(vuint16m4_t vd, const uint16_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vluxei64_tu(vuint16mf4_t vd, const uint16_t *rs1,
                                 vuint64m1_t rs2, size_t vl);
vuint16mf2_t __riscv_vluxei64_tu(vuint16mf2_t vd, const uint16_t *rs1,
                                 vuint64m2_t rs2, size_t vl);
vuint16m1_t __riscv_vluxei64_tu(vuint16m1_t vd, const uint16_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vuint16m2_t __riscv_vluxei64_tu(vuint16m2_t vd, const uint16_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vuint32mf2_t __riscv_vluxei8_tu(vuint32mf2_t vd, const uint32_t *rs1,
                                vuint8mf8_t rs2, size_t vl);
vuint32m1_t __riscv_vluxei8_tu(vuint32m1_t vd, const uint32_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vuint32m2_t __riscv_vluxei8_tu(vuint32m2_t vd, const uint32_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vuint32m4_t __riscv_vluxei8_tu(vuint32m4_t vd, const uint32_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint32m8_t __riscv_vluxei8_tu(vuint32m8_t vd, const uint32_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vuint32mf2_t __riscv_vluxei16_tu(vuint32mf2_t vd, const uint32_t *rs1,
                                 vuint16mf4_t rs2, size_t vl);
vuint32m1_t __riscv_vluxei16_tu(vuint32m1_t vd, const uint32_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vuint32m2_t __riscv_vluxei16_tu(vuint32m2_t vd, const uint32_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vuint32m4_t __riscv_vluxei16_tu(vuint32m4_t vd, const uint32_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vuint32m8_t __riscv_vluxei16_tu(vuint32m8_t vd, const uint32_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vuint32mf2_t __riscv_vluxei32_tu(vuint32mf2_t vd, const uint32_t *rs1,
                                 vuint32mf2_t rs2, size_t vl);
vuint32m1_t __riscv_vluxei32_tu(vuint32m1_t vd, const uint32_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vuint32m2_t __riscv_vluxei32_tu(vuint32m2_t vd, const uint32_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vuint32m4_t __riscv_vluxei32_tu(vuint32m4_t vd, const uint32_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vuint32m8_t __riscv_vluxei32_tu(vuint32m8_t vd, const uint32_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vuint32mf2_t __riscv_vluxei64_tu(vuint32mf2_t vd, const uint32_t *rs1,
                                 vuint64m1_t rs2, size_t vl);
vuint32m1_t __riscv_vluxei64_tu(vuint32m1_t vd, const uint32_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vuint32m2_t __riscv_vluxei64_tu(vuint32m2_t vd, const uint32_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vuint32m4_t __riscv_vluxei64_tu(vuint32m4_t vd, const uint32_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vuint64m1_t __riscv_vluxei8_tu(vuint64m1_t vd, const uint64_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vuint64m2_t __riscv_vluxei8_tu(vuint64m2_t vd, const uint64_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vuint64m4_t __riscv_vluxei8_tu(vuint64m4_t vd, const uint64_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vuint64m8_t __riscv_vluxei8_tu(vuint64m8_t vd, const uint64_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint64m1_t __riscv_vluxei16_tu(vuint64m1_t vd, const uint64_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vuint64m2_t __riscv_vluxei16_tu(vuint64m2_t vd, const uint64_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vuint64m4_t __riscv_vluxei16_tu(vuint64m4_t vd, const uint64_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vuint64m8_t __riscv_vluxei16_tu(vuint64m8_t vd, const uint64_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vuint64m1_t __riscv_vluxei32_tu(vuint64m1_t vd, const uint64_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vuint64m2_t __riscv_vluxei32_tu(vuint64m2_t vd, const uint64_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vuint64m4_t __riscv_vluxei32_tu(vuint64m4_t vd, const uint64_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vuint64m8_t __riscv_vluxei32_tu(vuint64m8_t vd, const uint64_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vuint64m1_t __riscv_vluxei64_tu(vuint64m1_t vd, const uint64_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vuint64m2_t __riscv_vluxei64_tu(vuint64m2_t vd, const uint64_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vuint64m4_t __riscv_vluxei64_tu(vuint64m4_t vd, const uint64_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vuint64m8_t __riscv_vluxei64_tu(vuint64m8_t vd, const uint64_t *rs1,
                                vuint64m8_t rs2, size_t vl);
// masked functions
vfloat16mf4_t __riscv_vloxei8_tum(vbool64_t vm, vfloat16mf4_t vd,
                                  const _Float16 *rs1, vuint8mf8_t rs2,
                                  size_t vl);
vfloat16mf2_t __riscv_vloxei8_tum(vbool32_t vm, vfloat16mf2_t vd,
                                  const _Float16 *rs1, vuint8mf4_t rs2,
                                  size_t vl);
vfloat16m1_t __riscv_vloxei8_tum(vbool16_t vm, vfloat16m1_t vd,
                                 const _Float16 *rs1, vuint8mf2_t rs2,
                                 size_t vl);
vfloat16m2_t __riscv_vloxei8_tum(vbool8_t vm, vfloat16m2_t vd,
                                 const _Float16 *rs1, vuint8m1_t rs2,
                                 size_t vl);
vfloat16m4_t __riscv_vloxei8_tum(vbool4_t vm, vfloat16m4_t vd,
                                 const _Float16 *rs1, vuint8m2_t rs2,
                                 size_t vl);
vfloat16m8_t __riscv_vloxei8_tum(vbool2_t vm, vfloat16m8_t vd,
                                 const _Float16 *rs1, vuint8m4_t rs2,
                                 size_t vl);
vfloat16mf4_t __riscv_vloxei16_tum(vbool64_t vm, vfloat16mf4_t vd,
                                   const _Float16 *rs1, vuint16mf4_t rs2,
                                   size_t vl);
vfloat16mf2_t __riscv_vloxei16_tum(vbool32_t vm, vfloat16mf2_t vd,
                                   const _Float16 *rs1, vuint16mf2_t rs2,
                                   size_t vl);
vfloat16m1_t __riscv_vloxei16_tum(vbool16_t vm, vfloat16m1_t vd,
                                  const _Float16 *rs1, vuint16m1_t rs2,
                                  size_t vl);
vfloat16m2_t __riscv_vloxei16_tum(vbool8_t vm, vfloat16m2_t vd,
                                  const _Float16 *rs1, vuint16m2_t rs2,
                                  size_t vl);
vfloat16m4_t __riscv_vloxei16_tum(vbool4_t vm, vfloat16m4_t vd,
                                  const _Float16 *rs1, vuint16m4_t rs2,
                                  size_t vl);
vfloat16m8_t __riscv_vloxei16_tum(vbool2_t vm, vfloat16m8_t vd,
                                  const _Float16 *rs1, vuint16m8_t rs2,
                                  size_t vl);
vfloat16mf4_t __riscv_vloxei32_tum(vbool64_t vm, vfloat16mf4_t vd,
                                   const _Float16 *rs1, vuint32mf2_t rs2,
                                   size_t vl);
vfloat16mf2_t __riscv_vloxei32_tum(vbool32_t vm, vfloat16mf2_t vd,
                                   const _Float16 *rs1, vuint32m1_t rs2,
                                   size_t vl);
vfloat16m1_t __riscv_vloxei32_tum(vbool16_t vm, vfloat16m1_t vd,
                                  const _Float16 *rs1, vuint32m2_t rs2,
                                  size_t vl);
vfloat16m2_t __riscv_vloxei32_tum(vbool8_t vm, vfloat16m2_t vd,
                                  const _Float16 *rs1, vuint32m4_t rs2,
                                  size_t vl);
vfloat16m4_t __riscv_vloxei32_tum(vbool4_t vm, vfloat16m4_t vd,
                                  const _Float16 *rs1, vuint32m8_t rs2,
                                  size_t vl);
vfloat16mf4_t __riscv_vloxei64_tum(vbool64_t vm, vfloat16mf4_t vd,
                                   const _Float16 *rs1, vuint64m1_t rs2,
                                   size_t vl);
vfloat16mf2_t __riscv_vloxei64_tum(vbool32_t vm, vfloat16mf2_t vd,
                                   const _Float16 *rs1, vuint64m2_t rs2,
                                   size_t vl);
vfloat16m1_t __riscv_vloxei64_tum(vbool16_t vm, vfloat16m1_t vd,
                                  const _Float16 *rs1, vuint64m4_t rs2,
                                  size_t vl);
vfloat16m2_t __riscv_vloxei64_tum(vbool8_t vm, vfloat16m2_t vd,
                                  const _Float16 *rs1, vuint64m8_t rs2,
                                  size_t vl);
vfloat32mf2_t __riscv_vloxei8_tum(vbool64_t vm, vfloat32mf2_t vd,
                                  const float *rs1, vuint8mf8_t rs2, size_t vl);
vfloat32m1_t __riscv_vloxei8_tum(vbool32_t vm, vfloat32m1_t vd,
                                 const float *rs1, vuint8mf4_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei8_tum(vbool16_t vm, vfloat32m2_t vd,
                                 const float *rs1, vuint8mf2_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei8_tum(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                 vuint8m1_t rs2, size_t vl);
vfloat32m8_t __riscv_vloxei8_tum(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                 vuint8m2_t rs2, size_t vl);
vfloat32mf2_t __riscv_vloxei16_tum(vbool64_t vm, vfloat32mf2_t vd,
                                   const float *rs1, vuint16mf4_t rs2,
                                   size_t vl);
vfloat32m1_t __riscv_vloxei16_tum(vbool32_t vm, vfloat32m1_t vd,
                                  const float *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vfloat32m2_t __riscv_vloxei16_tum(vbool16_t vm, vfloat32m2_t vd,
                                  const float *rs1, vuint16m1_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei16_tum(vbool8_t vm, vfloat32m4_t vd,
                                  const float *rs1, vuint16m2_t rs2, size_t vl);
vfloat32m8_t __riscv_vloxei16_tum(vbool4_t vm, vfloat32m8_t vd,
                                  const float *rs1, vuint16m4_t rs2, size_t vl);
vfloat32mf2_t __riscv_vloxei32_tum(vbool64_t vm, vfloat32mf2_t vd,
                                   const float *rs1, vuint32mf2_t rs2,
                                   size_t vl);
vfloat32m1_t __riscv_vloxei32_tum(vbool32_t vm, vfloat32m1_t vd,
                                  const float *rs1, vuint32m1_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei32_tum(vbool16_t vm, vfloat32m2_t vd,
                                  const float *rs1, vuint32m2_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei32_tum(vbool8_t vm, vfloat32m4_t vd,
                                  const float *rs1, vuint32m4_t rs2, size_t vl);
vfloat32m8_t __riscv_vloxei32_tum(vbool4_t vm, vfloat32m8_t vd,
                                  const float *rs1, vuint32m8_t rs2, size_t vl);
vfloat32mf2_t __riscv_vloxei64_tum(vbool64_t vm, vfloat32mf2_t vd,
                                   const float *rs1, vuint64m1_t rs2,
                                   size_t vl);
vfloat32m1_t __riscv_vloxei64_tum(vbool32_t vm, vfloat32m1_t vd,
                                  const float *rs1, vuint64m2_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei64_tum(vbool16_t vm, vfloat32m2_t vd,
                                  const float *rs1, vuint64m4_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei64_tum(vbool8_t vm, vfloat32m4_t vd,
                                  const float *rs1, vuint64m8_t rs2, size_t vl);
vfloat64m1_t __riscv_vloxei8_tum(vbool64_t vm, vfloat64m1_t vd,
                                 const double *rs1, vuint8mf8_t rs2, size_t vl);
vfloat64m2_t __riscv_vloxei8_tum(vbool32_t vm, vfloat64m2_t vd,
                                 const double *rs1, vuint8mf4_t rs2, size_t vl);
vfloat64m4_t __riscv_vloxei8_tum(vbool16_t vm, vfloat64m4_t vd,
                                 const double *rs1, vuint8mf2_t rs2, size_t vl);
vfloat64m8_t __riscv_vloxei8_tum(vbool8_t vm, vfloat64m8_t vd,
                                 const double *rs1, vuint8m1_t rs2, size_t vl);
vfloat64m1_t __riscv_vloxei16_tum(vbool64_t vm, vfloat64m1_t vd,
                                  const double *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vfloat64m2_t __riscv_vloxei16_tum(vbool32_t vm, vfloat64m2_t vd,
                                  const double *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vfloat64m4_t __riscv_vloxei16_tum(vbool16_t vm, vfloat64m4_t vd,
                                  const double *rs1, vuint16m1_t rs2,
                                  size_t vl);
vfloat64m8_t __riscv_vloxei16_tum(vbool8_t vm, vfloat64m8_t vd,
                                  const double *rs1, vuint16m2_t rs2,
                                  size_t vl);
vfloat64m1_t __riscv_vloxei32_tum(vbool64_t vm, vfloat64m1_t vd,
                                  const double *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vfloat64m2_t __riscv_vloxei32_tum(vbool32_t vm, vfloat64m2_t vd,
                                  const double *rs1, vuint32m1_t rs2,
                                  size_t vl);
vfloat64m4_t __riscv_vloxei32_tum(vbool16_t vm, vfloat64m4_t vd,
                                  const double *rs1, vuint32m2_t rs2,
                                  size_t vl);
vfloat64m8_t __riscv_vloxei32_tum(vbool8_t vm, vfloat64m8_t vd,
                                  const double *rs1, vuint32m4_t rs2,
                                  size_t vl);
vfloat64m1_t __riscv_vloxei64_tum(vbool64_t vm, vfloat64m1_t vd,
                                  const double *rs1, vuint64m1_t rs2,
                                  size_t vl);
vfloat64m2_t __riscv_vloxei64_tum(vbool32_t vm, vfloat64m2_t vd,
                                  const double *rs1, vuint64m2_t rs2,
                                  size_t vl);
vfloat64m4_t __riscv_vloxei64_tum(vbool16_t vm, vfloat64m4_t vd,
                                  const double *rs1, vuint64m4_t rs2,
                                  size_t vl);
vfloat64m8_t __riscv_vloxei64_tum(vbool8_t vm, vfloat64m8_t vd,
                                  const double *rs1, vuint64m8_t rs2,
                                  size_t vl);
vfloat16mf4_t __riscv_vluxei8_tum(vbool64_t vm, vfloat16mf4_t vd,
                                  const _Float16 *rs1, vuint8mf8_t rs2,
                                  size_t vl);
vfloat16mf2_t __riscv_vluxei8_tum(vbool32_t vm, vfloat16mf2_t vd,
                                  const _Float16 *rs1, vuint8mf4_t rs2,
                                  size_t vl);
vfloat16m1_t __riscv_vluxei8_tum(vbool16_t vm, vfloat16m1_t vd,
                                 const _Float16 *rs1, vuint8mf2_t rs2,
                                 size_t vl);
vfloat16m2_t __riscv_vluxei8_tum(vbool8_t vm, vfloat16m2_t vd,
                                 const _Float16 *rs1, vuint8m1_t rs2,
                                 size_t vl);
vfloat16m4_t __riscv_vluxei8_tum(vbool4_t vm, vfloat16m4_t vd,
                                 const _Float16 *rs1, vuint8m2_t rs2,
                                 size_t vl);
vfloat16m8_t __riscv_vluxei8_tum(vbool2_t vm, vfloat16m8_t vd,
                                 const _Float16 *rs1, vuint8m4_t rs2,
                                 size_t vl);
vfloat16mf4_t __riscv_vluxei16_tum(vbool64_t vm, vfloat16mf4_t vd,
                                   const _Float16 *rs1, vuint16mf4_t rs2,
                                   size_t vl);
vfloat16mf2_t __riscv_vluxei16_tum(vbool32_t vm, vfloat16mf2_t vd,
                                   const _Float16 *rs1, vuint16mf2_t rs2,
                                   size_t vl);
vfloat16m1_t __riscv_vluxei16_tum(vbool16_t vm, vfloat16m1_t vd,
                                  const _Float16 *rs1, vuint16m1_t rs2,
                                  size_t vl);
vfloat16m2_t __riscv_vluxei16_tum(vbool8_t vm, vfloat16m2_t vd,
                                  const _Float16 *rs1, vuint16m2_t rs2,
                                  size_t vl);
vfloat16m4_t __riscv_vluxei16_tum(vbool4_t vm, vfloat16m4_t vd,
                                  const _Float16 *rs1, vuint16m4_t rs2,
                                  size_t vl);
vfloat16m8_t __riscv_vluxei16_tum(vbool2_t vm, vfloat16m8_t vd,
                                  const _Float16 *rs1, vuint16m8_t rs2,
                                  size_t vl);
vfloat16mf4_t __riscv_vluxei32_tum(vbool64_t vm, vfloat16mf4_t vd,
                                   const _Float16 *rs1, vuint32mf2_t rs2,
                                   size_t vl);
vfloat16mf2_t __riscv_vluxei32_tum(vbool32_t vm, vfloat16mf2_t vd,
                                   const _Float16 *rs1, vuint32m1_t rs2,
                                   size_t vl);
vfloat16m1_t __riscv_vluxei32_tum(vbool16_t vm, vfloat16m1_t vd,
                                  const _Float16 *rs1, vuint32m2_t rs2,
                                  size_t vl);
vfloat16m2_t __riscv_vluxei32_tum(vbool8_t vm, vfloat16m2_t vd,
                                  const _Float16 *rs1, vuint32m4_t rs2,
                                  size_t vl);
vfloat16m4_t __riscv_vluxei32_tum(vbool4_t vm, vfloat16m4_t vd,
                                  const _Float16 *rs1, vuint32m8_t rs2,
                                  size_t vl);
vfloat16mf4_t __riscv_vluxei64_tum(vbool64_t vm, vfloat16mf4_t vd,
                                   const _Float16 *rs1, vuint64m1_t rs2,
                                   size_t vl);
vfloat16mf2_t __riscv_vluxei64_tum(vbool32_t vm, vfloat16mf2_t vd,
                                   const _Float16 *rs1, vuint64m2_t rs2,
                                   size_t vl);
vfloat16m1_t __riscv_vluxei64_tum(vbool16_t vm, vfloat16m1_t vd,
                                  const _Float16 *rs1, vuint64m4_t rs2,
                                  size_t vl);
vfloat16m2_t __riscv_vluxei64_tum(vbool8_t vm, vfloat16m2_t vd,
                                  const _Float16 *rs1, vuint64m8_t rs2,
                                  size_t vl);
vfloat32mf2_t __riscv_vluxei8_tum(vbool64_t vm, vfloat32mf2_t vd,
                                  const float *rs1, vuint8mf8_t rs2, size_t vl);
vfloat32m1_t __riscv_vluxei8_tum(vbool32_t vm, vfloat32m1_t vd,
                                 const float *rs1, vuint8mf4_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei8_tum(vbool16_t vm, vfloat32m2_t vd,
                                 const float *rs1, vuint8mf2_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei8_tum(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                 vuint8m1_t rs2, size_t vl);
vfloat32m8_t __riscv_vluxei8_tum(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                 vuint8m2_t rs2, size_t vl);
vfloat32mf2_t __riscv_vluxei16_tum(vbool64_t vm, vfloat32mf2_t vd,
                                   const float *rs1, vuint16mf4_t rs2,
                                   size_t vl);
vfloat32m1_t __riscv_vluxei16_tum(vbool32_t vm, vfloat32m1_t vd,
                                  const float *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vfloat32m2_t __riscv_vluxei16_tum(vbool16_t vm, vfloat32m2_t vd,
                                  const float *rs1, vuint16m1_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei16_tum(vbool8_t vm, vfloat32m4_t vd,
                                  const float *rs1, vuint16m2_t rs2, size_t vl);
vfloat32m8_t __riscv_vluxei16_tum(vbool4_t vm, vfloat32m8_t vd,
                                  const float *rs1, vuint16m4_t rs2, size_t vl);
vfloat32mf2_t __riscv_vluxei32_tum(vbool64_t vm, vfloat32mf2_t vd,
                                   const float *rs1, vuint32mf2_t rs2,
                                   size_t vl);
vfloat32m1_t __riscv_vluxei32_tum(vbool32_t vm, vfloat32m1_t vd,
                                  const float *rs1, vuint32m1_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei32_tum(vbool16_t vm, vfloat32m2_t vd,
                                  const float *rs1, vuint32m2_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei32_tum(vbool8_t vm, vfloat32m4_t vd,
                                  const float *rs1, vuint32m4_t rs2, size_t vl);
vfloat32m8_t __riscv_vluxei32_tum(vbool4_t vm, vfloat32m8_t vd,
                                  const float *rs1, vuint32m8_t rs2, size_t vl);
vfloat32mf2_t __riscv_vluxei64_tum(vbool64_t vm, vfloat32mf2_t vd,
                                   const float *rs1, vuint64m1_t rs2,
                                   size_t vl);
vfloat32m1_t __riscv_vluxei64_tum(vbool32_t vm, vfloat32m1_t vd,
                                  const float *rs1, vuint64m2_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei64_tum(vbool16_t vm, vfloat32m2_t vd,
                                  const float *rs1, vuint64m4_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei64_tum(vbool8_t vm, vfloat32m4_t vd,
                                  const float *rs1, vuint64m8_t rs2, size_t vl);
vfloat64m1_t __riscv_vluxei8_tum(vbool64_t vm, vfloat64m1_t vd,
                                 const double *rs1, vuint8mf8_t rs2, size_t vl);
vfloat64m2_t __riscv_vluxei8_tum(vbool32_t vm, vfloat64m2_t vd,
                                 const double *rs1, vuint8mf4_t rs2, size_t vl);
vfloat64m4_t __riscv_vluxei8_tum(vbool16_t vm, vfloat64m4_t vd,
                                 const double *rs1, vuint8mf2_t rs2, size_t vl);
vfloat64m8_t __riscv_vluxei8_tum(vbool8_t vm, vfloat64m8_t vd,
                                 const double *rs1, vuint8m1_t rs2, size_t vl);
vfloat64m1_t __riscv_vluxei16_tum(vbool64_t vm, vfloat64m1_t vd,
                                  const double *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vfloat64m2_t __riscv_vluxei16_tum(vbool32_t vm, vfloat64m2_t vd,
                                  const double *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vfloat64m4_t __riscv_vluxei16_tum(vbool16_t vm, vfloat64m4_t vd,
                                  const double *rs1, vuint16m1_t rs2,
                                  size_t vl);
vfloat64m8_t __riscv_vluxei16_tum(vbool8_t vm, vfloat64m8_t vd,
                                  const double *rs1, vuint16m2_t rs2,
                                  size_t vl);
vfloat64m1_t __riscv_vluxei32_tum(vbool64_t vm, vfloat64m1_t vd,
                                  const double *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vfloat64m2_t __riscv_vluxei32_tum(vbool32_t vm, vfloat64m2_t vd,
                                  const double *rs1, vuint32m1_t rs2,
                                  size_t vl);
vfloat64m4_t __riscv_vluxei32_tum(vbool16_t vm, vfloat64m4_t vd,
                                  const double *rs1, vuint32m2_t rs2,
                                  size_t vl);
vfloat64m8_t __riscv_vluxei32_tum(vbool8_t vm, vfloat64m8_t vd,
                                  const double *rs1, vuint32m4_t rs2,
                                  size_t vl);
vfloat64m1_t __riscv_vluxei64_tum(vbool64_t vm, vfloat64m1_t vd,
                                  const double *rs1, vuint64m1_t rs2,
                                  size_t vl);
vfloat64m2_t __riscv_vluxei64_tum(vbool32_t vm, vfloat64m2_t vd,
                                  const double *rs1, vuint64m2_t rs2,
                                  size_t vl);
vfloat64m4_t __riscv_vluxei64_tum(vbool16_t vm, vfloat64m4_t vd,
                                  const double *rs1, vuint64m4_t rs2,
                                  size_t vl);
vfloat64m8_t __riscv_vluxei64_tum(vbool8_t vm, vfloat64m8_t vd,
                                  const double *rs1, vuint64m8_t rs2,
                                  size_t vl);
vint8mf8_t __riscv_vloxei8_tum(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei8_tum(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei8_tum(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vint8m1_t __riscv_vloxei8_tum(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                              vuint8m1_t rs2, size_t vl);
vint8m2_t __riscv_vloxei8_tum(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                              vuint8m2_t rs2, size_t vl);
vint8m4_t __riscv_vloxei8_tum(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                              vuint8m4_t rs2, size_t vl);
vint8m8_t __riscv_vloxei8_tum(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                              vuint8m8_t rs2, size_t vl);
vint8mf8_t __riscv_vloxei16_tum(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei16_tum(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei16_tum(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vint8m1_t __riscv_vloxei16_tum(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint8m2_t __riscv_vloxei16_tum(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vint8m4_t __riscv_vloxei16_tum(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                               vuint16m8_t rs2, size_t vl);
vint8mf8_t __riscv_vloxei32_tum(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei32_tum(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei32_tum(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vint8m1_t __riscv_vloxei32_tum(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint8m2_t __riscv_vloxei32_tum(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vint8mf8_t __riscv_vloxei64_tum(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei64_tum(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei64_tum(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vint8m1_t __riscv_vloxei64_tum(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei8_tum(vbool64_t vm, vint16mf4_t vd,
                                const int16_t *rs1, vuint8mf8_t rs2, size_t vl);
vint16mf2_t __riscv_vloxei8_tum(vbool32_t vm, vint16mf2_t vd,
                                const int16_t *rs1, vuint8mf4_t rs2, size_t vl);
vint16m1_t __riscv_vloxei8_tum(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vint16m2_t __riscv_vloxei8_tum(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vint16m4_t __riscv_vloxei8_tum(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vint16m8_t __riscv_vloxei8_tum(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                               vuint8m4_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei16_tum(vbool64_t vm, vint16mf4_t vd,
                                 const int16_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vint16mf2_t __riscv_vloxei16_tum(vbool32_t vm, vint16mf2_t vd,
                                 const int16_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vint16m1_t __riscv_vloxei16_tum(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vint16m2_t __riscv_vloxei16_tum(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vint16m4_t __riscv_vloxei16_tum(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vint16m8_t __riscv_vloxei16_tum(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                                vuint16m8_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei32_tum(vbool64_t vm, vint16mf4_t vd,
                                 const int16_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vint16mf2_t __riscv_vloxei32_tum(vbool32_t vm, vint16mf2_t vd,
                                 const int16_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vint16m1_t __riscv_vloxei32_tum(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vint16m2_t __riscv_vloxei32_tum(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vint16m4_t __riscv_vloxei32_tum(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei64_tum(vbool64_t vm, vint16mf4_t vd,
                                 const int16_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vint16mf2_t __riscv_vloxei64_tum(vbool32_t vm, vint16mf2_t vd,
                                 const int16_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vint16m1_t __riscv_vloxei64_tum(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vint16m2_t __riscv_vloxei64_tum(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei8_tum(vbool64_t vm, vint32mf2_t vd,
                                const int32_t *rs1, vuint8mf8_t rs2, size_t vl);
vint32m1_t __riscv_vloxei8_tum(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vint32m2_t __riscv_vloxei8_tum(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vint32m4_t __riscv_vloxei8_tum(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vint32m8_t __riscv_vloxei8_tum(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei16_tum(vbool64_t vm, vint32mf2_t vd,
                                 const int32_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vint32m1_t __riscv_vloxei16_tum(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vint32m2_t __riscv_vloxei16_tum(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vint32m4_t __riscv_vloxei16_tum(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vint32m8_t __riscv_vloxei16_tum(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei32_tum(vbool64_t vm, vint32mf2_t vd,
                                 const int32_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vint32m1_t __riscv_vloxei32_tum(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vint32m2_t __riscv_vloxei32_tum(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vint32m4_t __riscv_vloxei32_tum(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vint32m8_t __riscv_vloxei32_tum(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei64_tum(vbool64_t vm, vint32mf2_t vd,
                                 const int32_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vint32m1_t __riscv_vloxei64_tum(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vint32m2_t __riscv_vloxei64_tum(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vint32m4_t __riscv_vloxei64_tum(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vint64m1_t __riscv_vloxei8_tum(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint64m2_t __riscv_vloxei8_tum(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vint64m4_t __riscv_vloxei8_tum(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vint64m8_t __riscv_vloxei8_tum(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vint64m1_t __riscv_vloxei16_tum(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vint64m2_t __riscv_vloxei16_tum(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vint64m4_t __riscv_vloxei16_tum(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vint64m8_t __riscv_vloxei16_tum(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vint64m1_t __riscv_vloxei32_tum(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vint64m2_t __riscv_vloxei32_tum(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vint64m4_t __riscv_vloxei32_tum(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vint64m8_t __riscv_vloxei32_tum(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vint64m1_t __riscv_vloxei64_tum(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vint64m2_t __riscv_vloxei64_tum(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vint64m4_t __riscv_vloxei64_tum(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vint64m8_t __riscv_vloxei64_tum(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei8_tum(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei8_tum(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei8_tum(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vint8m1_t __riscv_vluxei8_tum(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                              vuint8m1_t rs2, size_t vl);
vint8m2_t __riscv_vluxei8_tum(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                              vuint8m2_t rs2, size_t vl);
vint8m4_t __riscv_vluxei8_tum(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                              vuint8m4_t rs2, size_t vl);
vint8m8_t __riscv_vluxei8_tum(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                              vuint8m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei16_tum(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei16_tum(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei16_tum(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vint8m1_t __riscv_vluxei16_tum(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint8m2_t __riscv_vluxei16_tum(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vint8m4_t __riscv_vluxei16_tum(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                               vuint16m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei32_tum(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei32_tum(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei32_tum(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vint8m1_t __riscv_vluxei32_tum(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint8m2_t __riscv_vluxei32_tum(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei64_tum(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei64_tum(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei64_tum(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vint8m1_t __riscv_vluxei64_tum(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei8_tum(vbool64_t vm, vint16mf4_t vd,
                                const int16_t *rs1, vuint8mf8_t rs2, size_t vl);
vint16mf2_t __riscv_vluxei8_tum(vbool32_t vm, vint16mf2_t vd,
                                const int16_t *rs1, vuint8mf4_t rs2, size_t vl);
vint16m1_t __riscv_vluxei8_tum(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vint16m2_t __riscv_vluxei8_tum(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vint16m4_t __riscv_vluxei8_tum(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vint16m8_t __riscv_vluxei8_tum(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                               vuint8m4_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei16_tum(vbool64_t vm, vint16mf4_t vd,
                                 const int16_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vint16mf2_t __riscv_vluxei16_tum(vbool32_t vm, vint16mf2_t vd,
                                 const int16_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vint16m1_t __riscv_vluxei16_tum(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vint16m2_t __riscv_vluxei16_tum(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vint16m4_t __riscv_vluxei16_tum(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vint16m8_t __riscv_vluxei16_tum(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                                vuint16m8_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei32_tum(vbool64_t vm, vint16mf4_t vd,
                                 const int16_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vint16mf2_t __riscv_vluxei32_tum(vbool32_t vm, vint16mf2_t vd,
                                 const int16_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vint16m1_t __riscv_vluxei32_tum(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vint16m2_t __riscv_vluxei32_tum(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vint16m4_t __riscv_vluxei32_tum(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei64_tum(vbool64_t vm, vint16mf4_t vd,
                                 const int16_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vint16mf2_t __riscv_vluxei64_tum(vbool32_t vm, vint16mf2_t vd,
                                 const int16_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vint16m1_t __riscv_vluxei64_tum(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vint16m2_t __riscv_vluxei64_tum(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei8_tum(vbool64_t vm, vint32mf2_t vd,
                                const int32_t *rs1, vuint8mf8_t rs2, size_t vl);
vint32m1_t __riscv_vluxei8_tum(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vint32m2_t __riscv_vluxei8_tum(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vint32m4_t __riscv_vluxei8_tum(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vint32m8_t __riscv_vluxei8_tum(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei16_tum(vbool64_t vm, vint32mf2_t vd,
                                 const int32_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vint32m1_t __riscv_vluxei16_tum(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vint32m2_t __riscv_vluxei16_tum(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vint32m4_t __riscv_vluxei16_tum(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vint32m8_t __riscv_vluxei16_tum(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei32_tum(vbool64_t vm, vint32mf2_t vd,
                                 const int32_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vint32m1_t __riscv_vluxei32_tum(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vint32m2_t __riscv_vluxei32_tum(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vint32m4_t __riscv_vluxei32_tum(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vint32m8_t __riscv_vluxei32_tum(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei64_tum(vbool64_t vm, vint32mf2_t vd,
                                 const int32_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vint32m1_t __riscv_vluxei64_tum(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vint32m2_t __riscv_vluxei64_tum(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vint32m4_t __riscv_vluxei64_tum(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vint64m1_t __riscv_vluxei8_tum(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint64m2_t __riscv_vluxei8_tum(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vint64m4_t __riscv_vluxei8_tum(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vint64m8_t __riscv_vluxei8_tum(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vint64m1_t __riscv_vluxei16_tum(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                                vuint16mf4_t rs2, size_t vl);
vint64m2_t __riscv_vluxei16_tum(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                                vuint16mf2_t rs2, size_t vl);
vint64m4_t __riscv_vluxei16_tum(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                                vuint16m1_t rs2, size_t vl);
vint64m8_t __riscv_vluxei16_tum(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vint64m1_t __riscv_vluxei32_tum(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                                vuint32mf2_t rs2, size_t vl);
vint64m2_t __riscv_vluxei32_tum(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                                vuint32m1_t rs2, size_t vl);
vint64m4_t __riscv_vluxei32_tum(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                                vuint32m2_t rs2, size_t vl);
vint64m8_t __riscv_vluxei32_tum(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vint64m1_t __riscv_vluxei64_tum(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                                vuint64m1_t rs2, size_t vl);
vint64m2_t __riscv_vluxei64_tum(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                                vuint64m2_t rs2, size_t vl);
vint64m4_t __riscv_vluxei64_tum(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                                vuint64m4_t rs2, size_t vl);
vint64m8_t __riscv_vluxei64_tum(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei8_tum(vbool64_t vm, vuint8mf8_t vd,
                                const uint8_t *rs1, vuint8mf8_t rs2, size_t vl);
vuint8mf4_t __riscv_vloxei8_tum(vbool32_t vm, vuint8mf4_t vd,
                                const uint8_t *rs1, vuint8mf4_t rs2, size_t vl);
vuint8mf2_t __riscv_vloxei8_tum(vbool16_t vm, vuint8mf2_t vd,
                                const uint8_t *rs1, vuint8mf2_t rs2, size_t vl);
vuint8m1_t __riscv_vloxei8_tum(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint8m2_t __riscv_vloxei8_tum(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vuint8m4_t __riscv_vloxei8_tum(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                               vuint8m4_t rs2, size_t vl);
vuint8m8_t __riscv_vloxei8_tum(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                               vuint8m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei16_tum(vbool64_t vm, vuint8mf8_t vd,
                                 const uint8_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vuint8mf4_t __riscv_vloxei16_tum(vbool32_t vm, vuint8mf4_t vd,
                                 const uint8_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vuint8mf2_t __riscv_vloxei16_tum(vbool16_t vm, vuint8mf2_t vd,
                                 const uint8_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vuint8m1_t __riscv_vloxei16_tum(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vuint8m2_t __riscv_vloxei16_tum(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vuint8m4_t __riscv_vloxei16_tum(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                                vuint16m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei32_tum(vbool64_t vm, vuint8mf8_t vd,
                                 const uint8_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vuint8mf4_t __riscv_vloxei32_tum(vbool32_t vm, vuint8mf4_t vd,
                                 const uint8_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vuint8mf2_t __riscv_vloxei32_tum(vbool16_t vm, vuint8mf2_t vd,
                                 const uint8_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vuint8m1_t __riscv_vloxei32_tum(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vuint8m2_t __riscv_vloxei32_tum(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei64_tum(vbool64_t vm, vuint8mf8_t vd,
                                 const uint8_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vuint8mf4_t __riscv_vloxei64_tum(vbool32_t vm, vuint8mf4_t vd,
                                 const uint8_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vuint8mf2_t __riscv_vloxei64_tum(vbool16_t vm, vuint8mf2_t vd,
                                 const uint8_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vuint8m1_t __riscv_vloxei64_tum(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vloxei8_tum(vbool64_t vm, vuint16mf4_t vd,
                                 const uint16_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vuint16mf2_t __riscv_vloxei8_tum(vbool32_t vm, vuint16mf2_t vd,
                                 const uint16_t *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vuint16m1_t __riscv_vloxei8_tum(vbool16_t vm, vuint16m1_t vd,
                                const uint16_t *rs1, vuint8mf2_t rs2,
                                size_t vl);
vuint16m2_t __riscv_vloxei8_tum(vbool8_t vm, vuint16m2_t vd,
                                const uint16_t *rs1, vuint8m1_t rs2, size_t vl);
vuint16m4_t __riscv_vloxei8_tum(vbool4_t vm, vuint16m4_t vd,
                                const uint16_t *rs1, vuint8m2_t rs2, size_t vl);
vuint16m8_t __riscv_vloxei8_tum(vbool2_t vm, vuint16m8_t vd,
                                const uint16_t *rs1, vuint8m4_t rs2, size_t vl);
vuint16mf4_t __riscv_vloxei16_tum(vbool64_t vm, vuint16mf4_t vd,
                                  const uint16_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vuint16mf2_t __riscv_vloxei16_tum(vbool32_t vm, vuint16mf2_t vd,
                                  const uint16_t *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vuint16m1_t __riscv_vloxei16_tum(vbool16_t vm, vuint16m1_t vd,
                                 const uint16_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vuint16m2_t __riscv_vloxei16_tum(vbool8_t vm, vuint16m2_t vd,
                                 const uint16_t *rs1, vuint16m2_t rs2,
                                 size_t vl);
vuint16m4_t __riscv_vloxei16_tum(vbool4_t vm, vuint16m4_t vd,
                                 const uint16_t *rs1, vuint16m4_t rs2,
                                 size_t vl);
vuint16m8_t __riscv_vloxei16_tum(vbool2_t vm, vuint16m8_t vd,
                                 const uint16_t *rs1, vuint16m8_t rs2,
                                 size_t vl);
vuint16mf4_t __riscv_vloxei32_tum(vbool64_t vm, vuint16mf4_t vd,
                                  const uint16_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vuint16mf2_t __riscv_vloxei32_tum(vbool32_t vm, vuint16mf2_t vd,
                                  const uint16_t *rs1, vuint32m1_t rs2,
                                  size_t vl);
vuint16m1_t __riscv_vloxei32_tum(vbool16_t vm, vuint16m1_t vd,
                                 const uint16_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vuint16m2_t __riscv_vloxei32_tum(vbool8_t vm, vuint16m2_t vd,
                                 const uint16_t *rs1, vuint32m4_t rs2,
                                 size_t vl);
vuint16m4_t __riscv_vloxei32_tum(vbool4_t vm, vuint16m4_t vd,
                                 const uint16_t *rs1, vuint32m8_t rs2,
                                 size_t vl);
vuint16mf4_t __riscv_vloxei64_tum(vbool64_t vm, vuint16mf4_t vd,
                                  const uint16_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vuint16mf2_t __riscv_vloxei64_tum(vbool32_t vm, vuint16mf2_t vd,
                                  const uint16_t *rs1, vuint64m2_t rs2,
                                  size_t vl);
vuint16m1_t __riscv_vloxei64_tum(vbool16_t vm, vuint16m1_t vd,
                                 const uint16_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vuint16m2_t __riscv_vloxei64_tum(vbool8_t vm, vuint16m2_t vd,
                                 const uint16_t *rs1, vuint64m8_t rs2,
                                 size_t vl);
vuint32mf2_t __riscv_vloxei8_tum(vbool64_t vm, vuint32mf2_t vd,
                                 const uint32_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vuint32m1_t __riscv_vloxei8_tum(vbool32_t vm, vuint32m1_t vd,
                                const uint32_t *rs1, vuint8mf4_t rs2,
                                size_t vl);
vuint32m2_t __riscv_vloxei8_tum(vbool16_t vm, vuint32m2_t vd,
                                const uint32_t *rs1, vuint8mf2_t rs2,
                                size_t vl);
vuint32m4_t __riscv_vloxei8_tum(vbool8_t vm, vuint32m4_t vd,
                                const uint32_t *rs1, vuint8m1_t rs2, size_t vl);
vuint32m8_t __riscv_vloxei8_tum(vbool4_t vm, vuint32m8_t vd,
                                const uint32_t *rs1, vuint8m2_t rs2, size_t vl);
vuint32mf2_t __riscv_vloxei16_tum(vbool64_t vm, vuint32mf2_t vd,
                                  const uint32_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vuint32m1_t __riscv_vloxei16_tum(vbool32_t vm, vuint32m1_t vd,
                                 const uint32_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vuint32m2_t __riscv_vloxei16_tum(vbool16_t vm, vuint32m2_t vd,
                                 const uint32_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vuint32m4_t __riscv_vloxei16_tum(vbool8_t vm, vuint32m4_t vd,
                                 const uint32_t *rs1, vuint16m2_t rs2,
                                 size_t vl);
vuint32m8_t __riscv_vloxei16_tum(vbool4_t vm, vuint32m8_t vd,
                                 const uint32_t *rs1, vuint16m4_t rs2,
                                 size_t vl);
vuint32mf2_t __riscv_vloxei32_tum(vbool64_t vm, vuint32mf2_t vd,
                                  const uint32_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vuint32m1_t __riscv_vloxei32_tum(vbool32_t vm, vuint32m1_t vd,
                                 const uint32_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vuint32m2_t __riscv_vloxei32_tum(vbool16_t vm, vuint32m2_t vd,
                                 const uint32_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vuint32m4_t __riscv_vloxei32_tum(vbool8_t vm, vuint32m4_t vd,
                                 const uint32_t *rs1, vuint32m4_t rs2,
                                 size_t vl);
vuint32m8_t __riscv_vloxei32_tum(vbool4_t vm, vuint32m8_t vd,
                                 const uint32_t *rs1, vuint32m8_t rs2,
                                 size_t vl);
vuint32mf2_t __riscv_vloxei64_tum(vbool64_t vm, vuint32mf2_t vd,
                                  const uint32_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vuint32m1_t __riscv_vloxei64_tum(vbool32_t vm, vuint32m1_t vd,
                                 const uint32_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vuint32m2_t __riscv_vloxei64_tum(vbool16_t vm, vuint32m2_t vd,
                                 const uint32_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vuint32m4_t __riscv_vloxei64_tum(vbool8_t vm, vuint32m4_t vd,
                                 const uint32_t *rs1, vuint64m8_t rs2,
                                 size_t vl);
vuint64m1_t __riscv_vloxei8_tum(vbool64_t vm, vuint64m1_t vd,
                                const uint64_t *rs1, vuint8mf8_t rs2,
                                size_t vl);
vuint64m2_t __riscv_vloxei8_tum(vbool32_t vm, vuint64m2_t vd,
                                const uint64_t *rs1, vuint8mf4_t rs2,
                                size_t vl);
vuint64m4_t __riscv_vloxei8_tum(vbool16_t vm, vuint64m4_t vd,
                                const uint64_t *rs1, vuint8mf2_t rs2,
                                size_t vl);
vuint64m8_t __riscv_vloxei8_tum(vbool8_t vm, vuint64m8_t vd,
                                const uint64_t *rs1, vuint8m1_t rs2, size_t vl);
vuint64m1_t __riscv_vloxei16_tum(vbool64_t vm, vuint64m1_t vd,
                                 const uint64_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vuint64m2_t __riscv_vloxei16_tum(vbool32_t vm, vuint64m2_t vd,
                                 const uint64_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vuint64m4_t __riscv_vloxei16_tum(vbool16_t vm, vuint64m4_t vd,
                                 const uint64_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vuint64m8_t __riscv_vloxei16_tum(vbool8_t vm, vuint64m8_t vd,
                                 const uint64_t *rs1, vuint16m2_t rs2,
                                 size_t vl);
vuint64m1_t __riscv_vloxei32_tum(vbool64_t vm, vuint64m1_t vd,
                                 const uint64_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vuint64m2_t __riscv_vloxei32_tum(vbool32_t vm, vuint64m2_t vd,
                                 const uint64_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vuint64m4_t __riscv_vloxei32_tum(vbool16_t vm, vuint64m4_t vd,
                                 const uint64_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vuint64m8_t __riscv_vloxei32_tum(vbool8_t vm, vuint64m8_t vd,
                                 const uint64_t *rs1, vuint32m4_t rs2,
                                 size_t vl);
vuint64m1_t __riscv_vloxei64_tum(vbool64_t vm, vuint64m1_t vd,
                                 const uint64_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vuint64m2_t __riscv_vloxei64_tum(vbool32_t vm, vuint64m2_t vd,
                                 const uint64_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vuint64m4_t __riscv_vloxei64_tum(vbool16_t vm, vuint64m4_t vd,
                                 const uint64_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vuint64m8_t __riscv_vloxei64_tum(vbool8_t vm, vuint64m8_t vd,
                                 const uint64_t *rs1, vuint64m8_t rs2,
                                 size_t vl);
vuint8mf8_t __riscv_vluxei8_tum(vbool64_t vm, vuint8mf8_t vd,
                                const uint8_t *rs1, vuint8mf8_t rs2, size_t vl);
vuint8mf4_t __riscv_vluxei8_tum(vbool32_t vm, vuint8mf4_t vd,
                                const uint8_t *rs1, vuint8mf4_t rs2, size_t vl);
vuint8mf2_t __riscv_vluxei8_tum(vbool16_t vm, vuint8mf2_t vd,
                                const uint8_t *rs1, vuint8mf2_t rs2, size_t vl);
vuint8m1_t __riscv_vluxei8_tum(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint8m2_t __riscv_vluxei8_tum(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vuint8m4_t __riscv_vluxei8_tum(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                               vuint8m4_t rs2, size_t vl);
vuint8m8_t __riscv_vluxei8_tum(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                               vuint8m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei16_tum(vbool64_t vm, vuint8mf8_t vd,
                                 const uint8_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vuint8mf4_t __riscv_vluxei16_tum(vbool32_t vm, vuint8mf4_t vd,
                                 const uint8_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vuint8mf2_t __riscv_vluxei16_tum(vbool16_t vm, vuint8mf2_t vd,
                                 const uint8_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vuint8m1_t __riscv_vluxei16_tum(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vuint8m2_t __riscv_vluxei16_tum(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vuint8m4_t __riscv_vluxei16_tum(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                                vuint16m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei32_tum(vbool64_t vm, vuint8mf8_t vd,
                                 const uint8_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vuint8mf4_t __riscv_vluxei32_tum(vbool32_t vm, vuint8mf4_t vd,
                                 const uint8_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vuint8mf2_t __riscv_vluxei32_tum(vbool16_t vm, vuint8mf2_t vd,
                                 const uint8_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vuint8m1_t __riscv_vluxei32_tum(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vuint8m2_t __riscv_vluxei32_tum(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei64_tum(vbool64_t vm, vuint8mf8_t vd,
                                 const uint8_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vuint8mf4_t __riscv_vluxei64_tum(vbool32_t vm, vuint8mf4_t vd,
                                 const uint8_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vuint8mf2_t __riscv_vluxei64_tum(vbool16_t vm, vuint8mf2_t vd,
                                 const uint8_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vuint8m1_t __riscv_vluxei64_tum(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vluxei8_tum(vbool64_t vm, vuint16mf4_t vd,
                                 const uint16_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vuint16mf2_t __riscv_vluxei8_tum(vbool32_t vm, vuint16mf2_t vd,
                                 const uint16_t *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vuint16m1_t __riscv_vluxei8_tum(vbool16_t vm, vuint16m1_t vd,
                                const uint16_t *rs1, vuint8mf2_t rs2,
                                size_t vl);
vuint16m2_t __riscv_vluxei8_tum(vbool8_t vm, vuint16m2_t vd,
                                const uint16_t *rs1, vuint8m1_t rs2, size_t vl);
vuint16m4_t __riscv_vluxei8_tum(vbool4_t vm, vuint16m4_t vd,
                                const uint16_t *rs1, vuint8m2_t rs2, size_t vl);
vuint16m8_t __riscv_vluxei8_tum(vbool2_t vm, vuint16m8_t vd,
                                const uint16_t *rs1, vuint8m4_t rs2, size_t vl);
vuint16mf4_t __riscv_vluxei16_tum(vbool64_t vm, vuint16mf4_t vd,
                                  const uint16_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vuint16mf2_t __riscv_vluxei16_tum(vbool32_t vm, vuint16mf2_t vd,
                                  const uint16_t *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vuint16m1_t __riscv_vluxei16_tum(vbool16_t vm, vuint16m1_t vd,
                                 const uint16_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vuint16m2_t __riscv_vluxei16_tum(vbool8_t vm, vuint16m2_t vd,
                                 const uint16_t *rs1, vuint16m2_t rs2,
                                 size_t vl);
vuint16m4_t __riscv_vluxei16_tum(vbool4_t vm, vuint16m4_t vd,
                                 const uint16_t *rs1, vuint16m4_t rs2,
                                 size_t vl);
vuint16m8_t __riscv_vluxei16_tum(vbool2_t vm, vuint16m8_t vd,
                                 const uint16_t *rs1, vuint16m8_t rs2,
                                 size_t vl);
vuint16mf4_t __riscv_vluxei32_tum(vbool64_t vm, vuint16mf4_t vd,
                                  const uint16_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vuint16mf2_t __riscv_vluxei32_tum(vbool32_t vm, vuint16mf2_t vd,
                                  const uint16_t *rs1, vuint32m1_t rs2,
                                  size_t vl);
vuint16m1_t __riscv_vluxei32_tum(vbool16_t vm, vuint16m1_t vd,
                                 const uint16_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vuint16m2_t __riscv_vluxei32_tum(vbool8_t vm, vuint16m2_t vd,
                                 const uint16_t *rs1, vuint32m4_t rs2,
                                 size_t vl);
vuint16m4_t __riscv_vluxei32_tum(vbool4_t vm, vuint16m4_t vd,
                                 const uint16_t *rs1, vuint32m8_t rs2,
                                 size_t vl);
vuint16mf4_t __riscv_vluxei64_tum(vbool64_t vm, vuint16mf4_t vd,
                                  const uint16_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vuint16mf2_t __riscv_vluxei64_tum(vbool32_t vm, vuint16mf2_t vd,
                                  const uint16_t *rs1, vuint64m2_t rs2,
                                  size_t vl);
vuint16m1_t __riscv_vluxei64_tum(vbool16_t vm, vuint16m1_t vd,
                                 const uint16_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vuint16m2_t __riscv_vluxei64_tum(vbool8_t vm, vuint16m2_t vd,
                                 const uint16_t *rs1, vuint64m8_t rs2,
                                 size_t vl);
vuint32mf2_t __riscv_vluxei8_tum(vbool64_t vm, vuint32mf2_t vd,
                                 const uint32_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vuint32m1_t __riscv_vluxei8_tum(vbool32_t vm, vuint32m1_t vd,
                                const uint32_t *rs1, vuint8mf4_t rs2,
                                size_t vl);
vuint32m2_t __riscv_vluxei8_tum(vbool16_t vm, vuint32m2_t vd,
                                const uint32_t *rs1, vuint8mf2_t rs2,
                                size_t vl);
vuint32m4_t __riscv_vluxei8_tum(vbool8_t vm, vuint32m4_t vd,
                                const uint32_t *rs1, vuint8m1_t rs2, size_t vl);
vuint32m8_t __riscv_vluxei8_tum(vbool4_t vm, vuint32m8_t vd,
                                const uint32_t *rs1, vuint8m2_t rs2, size_t vl);
vuint32mf2_t __riscv_vluxei16_tum(vbool64_t vm, vuint32mf2_t vd,
                                  const uint32_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vuint32m1_t __riscv_vluxei16_tum(vbool32_t vm, vuint32m1_t vd,
                                 const uint32_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vuint32m2_t __riscv_vluxei16_tum(vbool16_t vm, vuint32m2_t vd,
                                 const uint32_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vuint32m4_t __riscv_vluxei16_tum(vbool8_t vm, vuint32m4_t vd,
                                 const uint32_t *rs1, vuint16m2_t rs2,
                                 size_t vl);
vuint32m8_t __riscv_vluxei16_tum(vbool4_t vm, vuint32m8_t vd,
                                 const uint32_t *rs1, vuint16m4_t rs2,
                                 size_t vl);
vuint32mf2_t __riscv_vluxei32_tum(vbool64_t vm, vuint32mf2_t vd,
                                  const uint32_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vuint32m1_t __riscv_vluxei32_tum(vbool32_t vm, vuint32m1_t vd,
                                 const uint32_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vuint32m2_t __riscv_vluxei32_tum(vbool16_t vm, vuint32m2_t vd,
                                 const uint32_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vuint32m4_t __riscv_vluxei32_tum(vbool8_t vm, vuint32m4_t vd,
                                 const uint32_t *rs1, vuint32m4_t rs2,
                                 size_t vl);
vuint32m8_t __riscv_vluxei32_tum(vbool4_t vm, vuint32m8_t vd,
                                 const uint32_t *rs1, vuint32m8_t rs2,
                                 size_t vl);
vuint32mf2_t __riscv_vluxei64_tum(vbool64_t vm, vuint32mf2_t vd,
                                  const uint32_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vuint32m1_t __riscv_vluxei64_tum(vbool32_t vm, vuint32m1_t vd,
                                 const uint32_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vuint32m2_t __riscv_vluxei64_tum(vbool16_t vm, vuint32m2_t vd,
                                 const uint32_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vuint32m4_t __riscv_vluxei64_tum(vbool8_t vm, vuint32m4_t vd,
                                 const uint32_t *rs1, vuint64m8_t rs2,
                                 size_t vl);
vuint64m1_t __riscv_vluxei8_tum(vbool64_t vm, vuint64m1_t vd,
                                const uint64_t *rs1, vuint8mf8_t rs2,
                                size_t vl);
vuint64m2_t __riscv_vluxei8_tum(vbool32_t vm, vuint64m2_t vd,
                                const uint64_t *rs1, vuint8mf4_t rs2,
                                size_t vl);
vuint64m4_t __riscv_vluxei8_tum(vbool16_t vm, vuint64m4_t vd,
                                const uint64_t *rs1, vuint8mf2_t rs2,
                                size_t vl);
vuint64m8_t __riscv_vluxei8_tum(vbool8_t vm, vuint64m8_t vd,
                                const uint64_t *rs1, vuint8m1_t rs2, size_t vl);
vuint64m1_t __riscv_vluxei16_tum(vbool64_t vm, vuint64m1_t vd,
                                 const uint64_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vuint64m2_t __riscv_vluxei16_tum(vbool32_t vm, vuint64m2_t vd,
                                 const uint64_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vuint64m4_t __riscv_vluxei16_tum(vbool16_t vm, vuint64m4_t vd,
                                 const uint64_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vuint64m8_t __riscv_vluxei16_tum(vbool8_t vm, vuint64m8_t vd,
                                 const uint64_t *rs1, vuint16m2_t rs2,
                                 size_t vl);
vuint64m1_t __riscv_vluxei32_tum(vbool64_t vm, vuint64m1_t vd,
                                 const uint64_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vuint64m2_t __riscv_vluxei32_tum(vbool32_t vm, vuint64m2_t vd,
                                 const uint64_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vuint64m4_t __riscv_vluxei32_tum(vbool16_t vm, vuint64m4_t vd,
                                 const uint64_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vuint64m8_t __riscv_vluxei32_tum(vbool8_t vm, vuint64m8_t vd,
                                 const uint64_t *rs1, vuint32m4_t rs2,
                                 size_t vl);
vuint64m1_t __riscv_vluxei64_tum(vbool64_t vm, vuint64m1_t vd,
                                 const uint64_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vuint64m2_t __riscv_vluxei64_tum(vbool32_t vm, vuint64m2_t vd,
                                 const uint64_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vuint64m4_t __riscv_vluxei64_tum(vbool16_t vm, vuint64m4_t vd,
                                 const uint64_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vuint64m8_t __riscv_vluxei64_tum(vbool8_t vm, vuint64m8_t vd,
                                 const uint64_t *rs1, vuint64m8_t rs2,
                                 size_t vl);
// masked functions
vfloat16mf4_t __riscv_vloxei8_tumu(vbool64_t vm, vfloat16mf4_t vd,
                                   const _Float16 *rs1, vuint8mf8_t rs2,
                                   size_t vl);
vfloat16mf2_t __riscv_vloxei8_tumu(vbool32_t vm, vfloat16mf2_t vd,
                                   const _Float16 *rs1, vuint8mf4_t rs2,
                                   size_t vl);
vfloat16m1_t __riscv_vloxei8_tumu(vbool16_t vm, vfloat16m1_t vd,
                                  const _Float16 *rs1, vuint8mf2_t rs2,
                                  size_t vl);
vfloat16m2_t __riscv_vloxei8_tumu(vbool8_t vm, vfloat16m2_t vd,
                                  const _Float16 *rs1, vuint8m1_t rs2,
                                  size_t vl);
vfloat16m4_t __riscv_vloxei8_tumu(vbool4_t vm, vfloat16m4_t vd,
                                  const _Float16 *rs1, vuint8m2_t rs2,
                                  size_t vl);
vfloat16m8_t __riscv_vloxei8_tumu(vbool2_t vm, vfloat16m8_t vd,
                                  const _Float16 *rs1, vuint8m4_t rs2,
                                  size_t vl);
vfloat16mf4_t __riscv_vloxei16_tumu(vbool64_t vm, vfloat16mf4_t vd,
                                    const _Float16 *rs1, vuint16mf4_t rs2,
                                    size_t vl);
vfloat16mf2_t __riscv_vloxei16_tumu(vbool32_t vm, vfloat16mf2_t vd,
                                    const _Float16 *rs1, vuint16mf2_t rs2,
                                    size_t vl);
vfloat16m1_t __riscv_vloxei16_tumu(vbool16_t vm, vfloat16m1_t vd,
                                   const _Float16 *rs1, vuint16m1_t rs2,
                                   size_t vl);
vfloat16m2_t __riscv_vloxei16_tumu(vbool8_t vm, vfloat16m2_t vd,
                                   const _Float16 *rs1, vuint16m2_t rs2,
                                   size_t vl);
vfloat16m4_t __riscv_vloxei16_tumu(vbool4_t vm, vfloat16m4_t vd,
                                   const _Float16 *rs1, vuint16m4_t rs2,
                                   size_t vl);
vfloat16m8_t __riscv_vloxei16_tumu(vbool2_t vm, vfloat16m8_t vd,
                                   const _Float16 *rs1, vuint16m8_t rs2,
                                   size_t vl);
vfloat16mf4_t __riscv_vloxei32_tumu(vbool64_t vm, vfloat16mf4_t vd,
                                    const _Float16 *rs1, vuint32mf2_t rs2,
                                    size_t vl);
vfloat16mf2_t __riscv_vloxei32_tumu(vbool32_t vm, vfloat16mf2_t vd,
                                    const _Float16 *rs1, vuint32m1_t rs2,
                                    size_t vl);
vfloat16m1_t __riscv_vloxei32_tumu(vbool16_t vm, vfloat16m1_t vd,
                                   const _Float16 *rs1, vuint32m2_t rs2,
                                   size_t vl);
vfloat16m2_t __riscv_vloxei32_tumu(vbool8_t vm, vfloat16m2_t vd,
                                   const _Float16 *rs1, vuint32m4_t rs2,
                                   size_t vl);
vfloat16m4_t __riscv_vloxei32_tumu(vbool4_t vm, vfloat16m4_t vd,
                                   const _Float16 *rs1, vuint32m8_t rs2,
                                   size_t vl);
vfloat16mf4_t __riscv_vloxei64_tumu(vbool64_t vm, vfloat16mf4_t vd,
                                    const _Float16 *rs1, vuint64m1_t rs2,
                                    size_t vl);
vfloat16mf2_t __riscv_vloxei64_tumu(vbool32_t vm, vfloat16mf2_t vd,
                                    const _Float16 *rs1, vuint64m2_t rs2,
                                    size_t vl);
vfloat16m1_t __riscv_vloxei64_tumu(vbool16_t vm, vfloat16m1_t vd,
                                   const _Float16 *rs1, vuint64m4_t rs2,
                                   size_t vl);
vfloat16m2_t __riscv_vloxei64_tumu(vbool8_t vm, vfloat16m2_t vd,
                                   const _Float16 *rs1, vuint64m8_t rs2,
                                   size_t vl);
vfloat32mf2_t __riscv_vloxei8_tumu(vbool64_t vm, vfloat32mf2_t vd,
                                   const float *rs1, vuint8mf8_t rs2,
                                   size_t vl);
vfloat32m1_t __riscv_vloxei8_tumu(vbool32_t vm, vfloat32m1_t vd,
                                  const float *rs1, vuint8mf4_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei8_tumu(vbool16_t vm, vfloat32m2_t vd,
                                  const float *rs1, vuint8mf2_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei8_tumu(vbool8_t vm, vfloat32m4_t vd,
                                  const float *rs1, vuint8m1_t rs2, size_t vl);
vfloat32m8_t __riscv_vloxei8_tumu(vbool4_t vm, vfloat32m8_t vd,
                                  const float *rs1, vuint8m2_t rs2, size_t vl);
vfloat32mf2_t __riscv_vloxei16_tumu(vbool64_t vm, vfloat32mf2_t vd,
                                    const float *rs1, vuint16mf4_t rs2,
                                    size_t vl);
vfloat32m1_t __riscv_vloxei16_tumu(vbool32_t vm, vfloat32m1_t vd,
                                   const float *rs1, vuint16mf2_t rs2,
                                   size_t vl);
vfloat32m2_t __riscv_vloxei16_tumu(vbool16_t vm, vfloat32m2_t vd,
                                   const float *rs1, vuint16m1_t rs2,
                                   size_t vl);
vfloat32m4_t __riscv_vloxei16_tumu(vbool8_t vm, vfloat32m4_t vd,
                                   const float *rs1, vuint16m2_t rs2,
                                   size_t vl);
vfloat32m8_t __riscv_vloxei16_tumu(vbool4_t vm, vfloat32m8_t vd,
                                   const float *rs1, vuint16m4_t rs2,
                                   size_t vl);
vfloat32mf2_t __riscv_vloxei32_tumu(vbool64_t vm, vfloat32mf2_t vd,
                                    const float *rs1, vuint32mf2_t rs2,
                                    size_t vl);
vfloat32m1_t __riscv_vloxei32_tumu(vbool32_t vm, vfloat32m1_t vd,
                                   const float *rs1, vuint32m1_t rs2,
                                   size_t vl);
vfloat32m2_t __riscv_vloxei32_tumu(vbool16_t vm, vfloat32m2_t vd,
                                   const float *rs1, vuint32m2_t rs2,
                                   size_t vl);
vfloat32m4_t __riscv_vloxei32_tumu(vbool8_t vm, vfloat32m4_t vd,
                                   const float *rs1, vuint32m4_t rs2,
                                   size_t vl);
vfloat32m8_t __riscv_vloxei32_tumu(vbool4_t vm, vfloat32m8_t vd,
                                   const float *rs1, vuint32m8_t rs2,
                                   size_t vl);
vfloat32mf2_t __riscv_vloxei64_tumu(vbool64_t vm, vfloat32mf2_t vd,
                                    const float *rs1, vuint64m1_t rs2,
                                    size_t vl);
vfloat32m1_t __riscv_vloxei64_tumu(vbool32_t vm, vfloat32m1_t vd,
                                   const float *rs1, vuint64m2_t rs2,
                                   size_t vl);
vfloat32m2_t __riscv_vloxei64_tumu(vbool16_t vm, vfloat32m2_t vd,
                                   const float *rs1, vuint64m4_t rs2,
                                   size_t vl);
vfloat32m4_t __riscv_vloxei64_tumu(vbool8_t vm, vfloat32m4_t vd,
                                   const float *rs1, vuint64m8_t rs2,
                                   size_t vl);
vfloat64m1_t __riscv_vloxei8_tumu(vbool64_t vm, vfloat64m1_t vd,
                                  const double *rs1, vuint8mf8_t rs2,
                                  size_t vl);
vfloat64m2_t __riscv_vloxei8_tumu(vbool32_t vm, vfloat64m2_t vd,
                                  const double *rs1, vuint8mf4_t rs2,
                                  size_t vl);
vfloat64m4_t __riscv_vloxei8_tumu(vbool16_t vm, vfloat64m4_t vd,
                                  const double *rs1, vuint8mf2_t rs2,
                                  size_t vl);
vfloat64m8_t __riscv_vloxei8_tumu(vbool8_t vm, vfloat64m8_t vd,
                                  const double *rs1, vuint8m1_t rs2, size_t vl);
vfloat64m1_t __riscv_vloxei16_tumu(vbool64_t vm, vfloat64m1_t vd,
                                   const double *rs1, vuint16mf4_t rs2,
                                   size_t vl);
vfloat64m2_t __riscv_vloxei16_tumu(vbool32_t vm, vfloat64m2_t vd,
                                   const double *rs1, vuint16mf2_t rs2,
                                   size_t vl);
vfloat64m4_t __riscv_vloxei16_tumu(vbool16_t vm, vfloat64m4_t vd,
                                   const double *rs1, vuint16m1_t rs2,
                                   size_t vl);
vfloat64m8_t __riscv_vloxei16_tumu(vbool8_t vm, vfloat64m8_t vd,
                                   const double *rs1, vuint16m2_t rs2,
                                   size_t vl);
vfloat64m1_t __riscv_vloxei32_tumu(vbool64_t vm, vfloat64m1_t vd,
                                   const double *rs1, vuint32mf2_t rs2,
                                   size_t vl);
vfloat64m2_t __riscv_vloxei32_tumu(vbool32_t vm, vfloat64m2_t vd,
                                   const double *rs1, vuint32m1_t rs2,
                                   size_t vl);
vfloat64m4_t __riscv_vloxei32_tumu(vbool16_t vm, vfloat64m4_t vd,
                                   const double *rs1, vuint32m2_t rs2,
                                   size_t vl);
vfloat64m8_t __riscv_vloxei32_tumu(vbool8_t vm, vfloat64m8_t vd,
                                   const double *rs1, vuint32m4_t rs2,
                                   size_t vl);
vfloat64m1_t __riscv_vloxei64_tumu(vbool64_t vm, vfloat64m1_t vd,
                                   const double *rs1, vuint64m1_t rs2,
                                   size_t vl);
vfloat64m2_t __riscv_vloxei64_tumu(vbool32_t vm, vfloat64m2_t vd,
                                   const double *rs1, vuint64m2_t rs2,
                                   size_t vl);
vfloat64m4_t __riscv_vloxei64_tumu(vbool16_t vm, vfloat64m4_t vd,
                                   const double *rs1, vuint64m4_t rs2,
                                   size_t vl);
vfloat64m8_t __riscv_vloxei64_tumu(vbool8_t vm, vfloat64m8_t vd,
                                   const double *rs1, vuint64m8_t rs2,
                                   size_t vl);
vfloat16mf4_t __riscv_vluxei8_tumu(vbool64_t vm, vfloat16mf4_t vd,
                                   const _Float16 *rs1, vuint8mf8_t rs2,
                                   size_t vl);
vfloat16mf2_t __riscv_vluxei8_tumu(vbool32_t vm, vfloat16mf2_t vd,
                                   const _Float16 *rs1, vuint8mf4_t rs2,
                                   size_t vl);
vfloat16m1_t __riscv_vluxei8_tumu(vbool16_t vm, vfloat16m1_t vd,
                                  const _Float16 *rs1, vuint8mf2_t rs2,
                                  size_t vl);
vfloat16m2_t __riscv_vluxei8_tumu(vbool8_t vm, vfloat16m2_t vd,
                                  const _Float16 *rs1, vuint8m1_t rs2,
                                  size_t vl);
vfloat16m4_t __riscv_vluxei8_tumu(vbool4_t vm, vfloat16m4_t vd,
                                  const _Float16 *rs1, vuint8m2_t rs2,
                                  size_t vl);
vfloat16m8_t __riscv_vluxei8_tumu(vbool2_t vm, vfloat16m8_t vd,
                                  const _Float16 *rs1, vuint8m4_t rs2,
                                  size_t vl);
vfloat16mf4_t __riscv_vluxei16_tumu(vbool64_t vm, vfloat16mf4_t vd,
                                    const _Float16 *rs1, vuint16mf4_t rs2,
                                    size_t vl);
vfloat16mf2_t __riscv_vluxei16_tumu(vbool32_t vm, vfloat16mf2_t vd,
                                    const _Float16 *rs1, vuint16mf2_t rs2,
                                    size_t vl);
vfloat16m1_t __riscv_vluxei16_tumu(vbool16_t vm, vfloat16m1_t vd,
                                   const _Float16 *rs1, vuint16m1_t rs2,
                                   size_t vl);
vfloat16m2_t __riscv_vluxei16_tumu(vbool8_t vm, vfloat16m2_t vd,
                                   const _Float16 *rs1, vuint16m2_t rs2,
                                   size_t vl);
vfloat16m4_t __riscv_vluxei16_tumu(vbool4_t vm, vfloat16m4_t vd,
                                   const _Float16 *rs1, vuint16m4_t rs2,
                                   size_t vl);
vfloat16m8_t __riscv_vluxei16_tumu(vbool2_t vm, vfloat16m8_t vd,
                                   const _Float16 *rs1, vuint16m8_t rs2,
                                   size_t vl);
vfloat16mf4_t __riscv_vluxei32_tumu(vbool64_t vm, vfloat16mf4_t vd,
                                    const _Float16 *rs1, vuint32mf2_t rs2,
                                    size_t vl);
vfloat16mf2_t __riscv_vluxei32_tumu(vbool32_t vm, vfloat16mf2_t vd,
                                    const _Float16 *rs1, vuint32m1_t rs2,
                                    size_t vl);
vfloat16m1_t __riscv_vluxei32_tumu(vbool16_t vm, vfloat16m1_t vd,
                                   const _Float16 *rs1, vuint32m2_t rs2,
                                   size_t vl);
vfloat16m2_t __riscv_vluxei32_tumu(vbool8_t vm, vfloat16m2_t vd,
                                   const _Float16 *rs1, vuint32m4_t rs2,
                                   size_t vl);
vfloat16m4_t __riscv_vluxei32_tumu(vbool4_t vm, vfloat16m4_t vd,
                                   const _Float16 *rs1, vuint32m8_t rs2,
                                   size_t vl);
vfloat16mf4_t __riscv_vluxei64_tumu(vbool64_t vm, vfloat16mf4_t vd,
                                    const _Float16 *rs1, vuint64m1_t rs2,
                                    size_t vl);
vfloat16mf2_t __riscv_vluxei64_tumu(vbool32_t vm, vfloat16mf2_t vd,
                                    const _Float16 *rs1, vuint64m2_t rs2,
                                    size_t vl);
vfloat16m1_t __riscv_vluxei64_tumu(vbool16_t vm, vfloat16m1_t vd,
                                   const _Float16 *rs1, vuint64m4_t rs2,
                                   size_t vl);
vfloat16m2_t __riscv_vluxei64_tumu(vbool8_t vm, vfloat16m2_t vd,
                                   const _Float16 *rs1, vuint64m8_t rs2,
                                   size_t vl);
vfloat32mf2_t __riscv_vluxei8_tumu(vbool64_t vm, vfloat32mf2_t vd,
                                   const float *rs1, vuint8mf8_t rs2,
                                   size_t vl);
vfloat32m1_t __riscv_vluxei8_tumu(vbool32_t vm, vfloat32m1_t vd,
                                  const float *rs1, vuint8mf4_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei8_tumu(vbool16_t vm, vfloat32m2_t vd,
                                  const float *rs1, vuint8mf2_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei8_tumu(vbool8_t vm, vfloat32m4_t vd,
                                  const float *rs1, vuint8m1_t rs2, size_t vl);
vfloat32m8_t __riscv_vluxei8_tumu(vbool4_t vm, vfloat32m8_t vd,
                                  const float *rs1, vuint8m2_t rs2, size_t vl);
vfloat32mf2_t __riscv_vluxei16_tumu(vbool64_t vm, vfloat32mf2_t vd,
                                    const float *rs1, vuint16mf4_t rs2,
                                    size_t vl);
vfloat32m1_t __riscv_vluxei16_tumu(vbool32_t vm, vfloat32m1_t vd,
                                   const float *rs1, vuint16mf2_t rs2,
                                   size_t vl);
vfloat32m2_t __riscv_vluxei16_tumu(vbool16_t vm, vfloat32m2_t vd,
                                   const float *rs1, vuint16m1_t rs2,
                                   size_t vl);
vfloat32m4_t __riscv_vluxei16_tumu(vbool8_t vm, vfloat32m4_t vd,
                                   const float *rs1, vuint16m2_t rs2,
                                   size_t vl);
vfloat32m8_t __riscv_vluxei16_tumu(vbool4_t vm, vfloat32m8_t vd,
                                   const float *rs1, vuint16m4_t rs2,
                                   size_t vl);
vfloat32mf2_t __riscv_vluxei32_tumu(vbool64_t vm, vfloat32mf2_t vd,
                                    const float *rs1, vuint32mf2_t rs2,
                                    size_t vl);
vfloat32m1_t __riscv_vluxei32_tumu(vbool32_t vm, vfloat32m1_t vd,
                                   const float *rs1, vuint32m1_t rs2,
                                   size_t vl);
vfloat32m2_t __riscv_vluxei32_tumu(vbool16_t vm, vfloat32m2_t vd,
                                   const float *rs1, vuint32m2_t rs2,
                                   size_t vl);
vfloat32m4_t __riscv_vluxei32_tumu(vbool8_t vm, vfloat32m4_t vd,
                                   const float *rs1, vuint32m4_t rs2,
                                   size_t vl);
vfloat32m8_t __riscv_vluxei32_tumu(vbool4_t vm, vfloat32m8_t vd,
                                   const float *rs1, vuint32m8_t rs2,
                                   size_t vl);
vfloat32mf2_t __riscv_vluxei64_tumu(vbool64_t vm, vfloat32mf2_t vd,
                                    const float *rs1, vuint64m1_t rs2,
                                    size_t vl);
vfloat32m1_t __riscv_vluxei64_tumu(vbool32_t vm, vfloat32m1_t vd,
                                   const float *rs1, vuint64m2_t rs2,
                                   size_t vl);
vfloat32m2_t __riscv_vluxei64_tumu(vbool16_t vm, vfloat32m2_t vd,
                                   const float *rs1, vuint64m4_t rs2,
                                   size_t vl);
vfloat32m4_t __riscv_vluxei64_tumu(vbool8_t vm, vfloat32m4_t vd,
                                   const float *rs1, vuint64m8_t rs2,
                                   size_t vl);
vfloat64m1_t __riscv_vluxei8_tumu(vbool64_t vm, vfloat64m1_t vd,
                                  const double *rs1, vuint8mf8_t rs2,
                                  size_t vl);
vfloat64m2_t __riscv_vluxei8_tumu(vbool32_t vm, vfloat64m2_t vd,
                                  const double *rs1, vuint8mf4_t rs2,
                                  size_t vl);
vfloat64m4_t __riscv_vluxei8_tumu(vbool16_t vm, vfloat64m4_t vd,
                                  const double *rs1, vuint8mf2_t rs2,
                                  size_t vl);
vfloat64m8_t __riscv_vluxei8_tumu(vbool8_t vm, vfloat64m8_t vd,
                                  const double *rs1, vuint8m1_t rs2, size_t vl);
vfloat64m1_t __riscv_vluxei16_tumu(vbool64_t vm, vfloat64m1_t vd,
                                   const double *rs1, vuint16mf4_t rs2,
                                   size_t vl);
vfloat64m2_t __riscv_vluxei16_tumu(vbool32_t vm, vfloat64m2_t vd,
                                   const double *rs1, vuint16mf2_t rs2,
                                   size_t vl);
vfloat64m4_t __riscv_vluxei16_tumu(vbool16_t vm, vfloat64m4_t vd,
                                   const double *rs1, vuint16m1_t rs2,
                                   size_t vl);
vfloat64m8_t __riscv_vluxei16_tumu(vbool8_t vm, vfloat64m8_t vd,
                                   const double *rs1, vuint16m2_t rs2,
                                   size_t vl);
vfloat64m1_t __riscv_vluxei32_tumu(vbool64_t vm, vfloat64m1_t vd,
                                   const double *rs1, vuint32mf2_t rs2,
                                   size_t vl);
vfloat64m2_t __riscv_vluxei32_tumu(vbool32_t vm, vfloat64m2_t vd,
                                   const double *rs1, vuint32m1_t rs2,
                                   size_t vl);
vfloat64m4_t __riscv_vluxei32_tumu(vbool16_t vm, vfloat64m4_t vd,
                                   const double *rs1, vuint32m2_t rs2,
                                   size_t vl);
vfloat64m8_t __riscv_vluxei32_tumu(vbool8_t vm, vfloat64m8_t vd,
                                   const double *rs1, vuint32m4_t rs2,
                                   size_t vl);
vfloat64m1_t __riscv_vluxei64_tumu(vbool64_t vm, vfloat64m1_t vd,
                                   const double *rs1, vuint64m1_t rs2,
                                   size_t vl);
vfloat64m2_t __riscv_vluxei64_tumu(vbool32_t vm, vfloat64m2_t vd,
                                   const double *rs1, vuint64m2_t rs2,
                                   size_t vl);
vfloat64m4_t __riscv_vluxei64_tumu(vbool16_t vm, vfloat64m4_t vd,
                                   const double *rs1, vuint64m4_t rs2,
                                   size_t vl);
vfloat64m8_t __riscv_vluxei64_tumu(vbool8_t vm, vfloat64m8_t vd,
                                   const double *rs1, vuint64m8_t rs2,
                                   size_t vl);
vint8mf8_t __riscv_vloxei8_tumu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                vuint8mf8_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei8_tumu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                vuint8mf4_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei8_tumu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                vuint8mf2_t rs2, size_t vl);
vint8m1_t __riscv_vloxei8_tumu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vint8m2_t __riscv_vloxei8_tumu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vint8m4_t __riscv_vloxei8_tumu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                               vuint8m4_t rs2, size_t vl);
vint8m8_t __riscv_vloxei8_tumu(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                               vuint8m8_t rs2, size_t vl);
vint8mf8_t __riscv_vloxei16_tumu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                 vuint16mf4_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei16_tumu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                 vuint16mf2_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei16_tumu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                 vuint16m1_t rs2, size_t vl);
vint8m1_t __riscv_vloxei16_tumu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vint8m2_t __riscv_vloxei16_tumu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vint8m4_t __riscv_vloxei16_tumu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                                vuint16m8_t rs2, size_t vl);
vint8mf8_t __riscv_vloxei32_tumu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                 vuint32mf2_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei32_tumu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                 vuint32m1_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei32_tumu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                 vuint32m2_t rs2, size_t vl);
vint8m1_t __riscv_vloxei32_tumu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vint8m2_t __riscv_vloxei32_tumu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vint8mf8_t __riscv_vloxei64_tumu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                 vuint64m1_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei64_tumu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                 vuint64m2_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei64_tumu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                 vuint64m4_t rs2, size_t vl);
vint8m1_t __riscv_vloxei64_tumu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei8_tumu(vbool64_t vm, vint16mf4_t vd,
                                 const int16_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vint16mf2_t __riscv_vloxei8_tumu(vbool32_t vm, vint16mf2_t vd,
                                 const int16_t *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vint16m1_t __riscv_vloxei8_tumu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                                vuint8mf2_t rs2, size_t vl);
vint16m2_t __riscv_vloxei8_tumu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                vuint8m1_t rs2, size_t vl);
vint16m4_t __riscv_vloxei8_tumu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                                vuint8m2_t rs2, size_t vl);
vint16m8_t __riscv_vloxei8_tumu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                                vuint8m4_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei16_tumu(vbool64_t vm, vint16mf4_t vd,
                                  const int16_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vint16mf2_t __riscv_vloxei16_tumu(vbool32_t vm, vint16mf2_t vd,
                                  const int16_t *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vint16m1_t __riscv_vloxei16_tumu(vbool16_t vm, vint16m1_t vd,
                                 const int16_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vint16m2_t __riscv_vloxei16_tumu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                 vuint16m2_t rs2, size_t vl);
vint16m4_t __riscv_vloxei16_tumu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                                 vuint16m4_t rs2, size_t vl);
vint16m8_t __riscv_vloxei16_tumu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                                 vuint16m8_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei32_tumu(vbool64_t vm, vint16mf4_t vd,
                                  const int16_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vint16mf2_t __riscv_vloxei32_tumu(vbool32_t vm, vint16mf2_t vd,
                                  const int16_t *rs1, vuint32m1_t rs2,
                                  size_t vl);
vint16m1_t __riscv_vloxei32_tumu(vbool16_t vm, vint16m1_t vd,
                                 const int16_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vint16m2_t __riscv_vloxei32_tumu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                 vuint32m4_t rs2, size_t vl);
vint16m4_t __riscv_vloxei32_tumu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                                 vuint32m8_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei64_tumu(vbool64_t vm, vint16mf4_t vd,
                                  const int16_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vint16mf2_t __riscv_vloxei64_tumu(vbool32_t vm, vint16mf2_t vd,
                                  const int16_t *rs1, vuint64m2_t rs2,
                                  size_t vl);
vint16m1_t __riscv_vloxei64_tumu(vbool16_t vm, vint16m1_t vd,
                                 const int16_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vint16m2_t __riscv_vloxei64_tumu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                 vuint64m8_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei8_tumu(vbool64_t vm, vint32mf2_t vd,
                                 const int32_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vint32m1_t __riscv_vloxei8_tumu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                                vuint8mf4_t rs2, size_t vl);
vint32m2_t __riscv_vloxei8_tumu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                                vuint8mf2_t rs2, size_t vl);
vint32m4_t __riscv_vloxei8_tumu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                vuint8m1_t rs2, size_t vl);
vint32m8_t __riscv_vloxei8_tumu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                                vuint8m2_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei16_tumu(vbool64_t vm, vint32mf2_t vd,
                                  const int32_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vint32m1_t __riscv_vloxei16_tumu(vbool32_t vm, vint32m1_t vd,
                                 const int32_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vint32m2_t __riscv_vloxei16_tumu(vbool16_t vm, vint32m2_t vd,
                                 const int32_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vint32m4_t __riscv_vloxei16_tumu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                 vuint16m2_t rs2, size_t vl);
vint32m8_t __riscv_vloxei16_tumu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                                 vuint16m4_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei32_tumu(vbool64_t vm, vint32mf2_t vd,
                                  const int32_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vint32m1_t __riscv_vloxei32_tumu(vbool32_t vm, vint32m1_t vd,
                                 const int32_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vint32m2_t __riscv_vloxei32_tumu(vbool16_t vm, vint32m2_t vd,
                                 const int32_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vint32m4_t __riscv_vloxei32_tumu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                 vuint32m4_t rs2, size_t vl);
vint32m8_t __riscv_vloxei32_tumu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                                 vuint32m8_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei64_tumu(vbool64_t vm, vint32mf2_t vd,
                                  const int32_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vint32m1_t __riscv_vloxei64_tumu(vbool32_t vm, vint32m1_t vd,
                                 const int32_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vint32m2_t __riscv_vloxei64_tumu(vbool16_t vm, vint32m2_t vd,
                                 const int32_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vint32m4_t __riscv_vloxei64_tumu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                 vuint64m8_t rs2, size_t vl);
vint64m1_t __riscv_vloxei8_tumu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                                vuint8mf8_t rs2, size_t vl);
vint64m2_t __riscv_vloxei8_tumu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                                vuint8mf4_t rs2, size_t vl);
vint64m4_t __riscv_vloxei8_tumu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                                vuint8mf2_t rs2, size_t vl);
vint64m8_t __riscv_vloxei8_tumu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                vuint8m1_t rs2, size_t vl);
vint64m1_t __riscv_vloxei16_tumu(vbool64_t vm, vint64m1_t vd,
                                 const int64_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vint64m2_t __riscv_vloxei16_tumu(vbool32_t vm, vint64m2_t vd,
                                 const int64_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vint64m4_t __riscv_vloxei16_tumu(vbool16_t vm, vint64m4_t vd,
                                 const int64_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vint64m8_t __riscv_vloxei16_tumu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                 vuint16m2_t rs2, size_t vl);
vint64m1_t __riscv_vloxei32_tumu(vbool64_t vm, vint64m1_t vd,
                                 const int64_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vint64m2_t __riscv_vloxei32_tumu(vbool32_t vm, vint64m2_t vd,
                                 const int64_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vint64m4_t __riscv_vloxei32_tumu(vbool16_t vm, vint64m4_t vd,
                                 const int64_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vint64m8_t __riscv_vloxei32_tumu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                 vuint32m4_t rs2, size_t vl);
vint64m1_t __riscv_vloxei64_tumu(vbool64_t vm, vint64m1_t vd,
                                 const int64_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vint64m2_t __riscv_vloxei64_tumu(vbool32_t vm, vint64m2_t vd,
                                 const int64_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vint64m4_t __riscv_vloxei64_tumu(vbool16_t vm, vint64m4_t vd,
                                 const int64_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vint64m8_t __riscv_vloxei64_tumu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                 vuint64m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei8_tumu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                vuint8mf8_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei8_tumu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                vuint8mf4_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei8_tumu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                vuint8mf2_t rs2, size_t vl);
vint8m1_t __riscv_vluxei8_tumu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vint8m2_t __riscv_vluxei8_tumu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vint8m4_t __riscv_vluxei8_tumu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                               vuint8m4_t rs2, size_t vl);
vint8m8_t __riscv_vluxei8_tumu(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                               vuint8m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei16_tumu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                 vuint16mf4_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei16_tumu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                 vuint16mf2_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei16_tumu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                 vuint16m1_t rs2, size_t vl);
vint8m1_t __riscv_vluxei16_tumu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                                vuint16m2_t rs2, size_t vl);
vint8m2_t __riscv_vluxei16_tumu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                                vuint16m4_t rs2, size_t vl);
vint8m4_t __riscv_vluxei16_tumu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                                vuint16m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei32_tumu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                 vuint32mf2_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei32_tumu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                 vuint32m1_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei32_tumu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                 vuint32m2_t rs2, size_t vl);
vint8m1_t __riscv_vluxei32_tumu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                                vuint32m4_t rs2, size_t vl);
vint8m2_t __riscv_vluxei32_tumu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                                vuint32m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei64_tumu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                                 vuint64m1_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei64_tumu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                                 vuint64m2_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei64_tumu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                                 vuint64m4_t rs2, size_t vl);
vint8m1_t __riscv_vluxei64_tumu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                                vuint64m8_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei8_tumu(vbool64_t vm, vint16mf4_t vd,
                                 const int16_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vint16mf2_t __riscv_vluxei8_tumu(vbool32_t vm, vint16mf2_t vd,
                                 const int16_t *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vint16m1_t __riscv_vluxei8_tumu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                                vuint8mf2_t rs2, size_t vl);
vint16m2_t __riscv_vluxei8_tumu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                vuint8m1_t rs2, size_t vl);
vint16m4_t __riscv_vluxei8_tumu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                                vuint8m2_t rs2, size_t vl);
vint16m8_t __riscv_vluxei8_tumu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                                vuint8m4_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei16_tumu(vbool64_t vm, vint16mf4_t vd,
                                  const int16_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vint16mf2_t __riscv_vluxei16_tumu(vbool32_t vm, vint16mf2_t vd,
                                  const int16_t *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vint16m1_t __riscv_vluxei16_tumu(vbool16_t vm, vint16m1_t vd,
                                 const int16_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vint16m2_t __riscv_vluxei16_tumu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                 vuint16m2_t rs2, size_t vl);
vint16m4_t __riscv_vluxei16_tumu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                                 vuint16m4_t rs2, size_t vl);
vint16m8_t __riscv_vluxei16_tumu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                                 vuint16m8_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei32_tumu(vbool64_t vm, vint16mf4_t vd,
                                  const int16_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vint16mf2_t __riscv_vluxei32_tumu(vbool32_t vm, vint16mf2_t vd,
                                  const int16_t *rs1, vuint32m1_t rs2,
                                  size_t vl);
vint16m1_t __riscv_vluxei32_tumu(vbool16_t vm, vint16m1_t vd,
                                 const int16_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vint16m2_t __riscv_vluxei32_tumu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                 vuint32m4_t rs2, size_t vl);
vint16m4_t __riscv_vluxei32_tumu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                                 vuint32m8_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei64_tumu(vbool64_t vm, vint16mf4_t vd,
                                  const int16_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vint16mf2_t __riscv_vluxei64_tumu(vbool32_t vm, vint16mf2_t vd,
                                  const int16_t *rs1, vuint64m2_t rs2,
                                  size_t vl);
vint16m1_t __riscv_vluxei64_tumu(vbool16_t vm, vint16m1_t vd,
                                 const int16_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vint16m2_t __riscv_vluxei64_tumu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                 vuint64m8_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei8_tumu(vbool64_t vm, vint32mf2_t vd,
                                 const int32_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vint32m1_t __riscv_vluxei8_tumu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                                vuint8mf4_t rs2, size_t vl);
vint32m2_t __riscv_vluxei8_tumu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                                vuint8mf2_t rs2, size_t vl);
vint32m4_t __riscv_vluxei8_tumu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                vuint8m1_t rs2, size_t vl);
vint32m8_t __riscv_vluxei8_tumu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                                vuint8m2_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei16_tumu(vbool64_t vm, vint32mf2_t vd,
                                  const int32_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vint32m1_t __riscv_vluxei16_tumu(vbool32_t vm, vint32m1_t vd,
                                 const int32_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vint32m2_t __riscv_vluxei16_tumu(vbool16_t vm, vint32m2_t vd,
                                 const int32_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vint32m4_t __riscv_vluxei16_tumu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                 vuint16m2_t rs2, size_t vl);
vint32m8_t __riscv_vluxei16_tumu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                                 vuint16m4_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei32_tumu(vbool64_t vm, vint32mf2_t vd,
                                  const int32_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vint32m1_t __riscv_vluxei32_tumu(vbool32_t vm, vint32m1_t vd,
                                 const int32_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vint32m2_t __riscv_vluxei32_tumu(vbool16_t vm, vint32m2_t vd,
                                 const int32_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vint32m4_t __riscv_vluxei32_tumu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                 vuint32m4_t rs2, size_t vl);
vint32m8_t __riscv_vluxei32_tumu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                                 vuint32m8_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei64_tumu(vbool64_t vm, vint32mf2_t vd,
                                  const int32_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vint32m1_t __riscv_vluxei64_tumu(vbool32_t vm, vint32m1_t vd,
                                 const int32_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vint32m2_t __riscv_vluxei64_tumu(vbool16_t vm, vint32m2_t vd,
                                 const int32_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vint32m4_t __riscv_vluxei64_tumu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                 vuint64m8_t rs2, size_t vl);
vint64m1_t __riscv_vluxei8_tumu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                                vuint8mf8_t rs2, size_t vl);
vint64m2_t __riscv_vluxei8_tumu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                                vuint8mf4_t rs2, size_t vl);
vint64m4_t __riscv_vluxei8_tumu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                                vuint8mf2_t rs2, size_t vl);
vint64m8_t __riscv_vluxei8_tumu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                vuint8m1_t rs2, size_t vl);
vint64m1_t __riscv_vluxei16_tumu(vbool64_t vm, vint64m1_t vd,
                                 const int64_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vint64m2_t __riscv_vluxei16_tumu(vbool32_t vm, vint64m2_t vd,
                                 const int64_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vint64m4_t __riscv_vluxei16_tumu(vbool16_t vm, vint64m4_t vd,
                                 const int64_t *rs1, vuint16m1_t rs2,
                                 size_t vl);
vint64m8_t __riscv_vluxei16_tumu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                 vuint16m2_t rs2, size_t vl);
vint64m1_t __riscv_vluxei32_tumu(vbool64_t vm, vint64m1_t vd,
                                 const int64_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vint64m2_t __riscv_vluxei32_tumu(vbool32_t vm, vint64m2_t vd,
                                 const int64_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vint64m4_t __riscv_vluxei32_tumu(vbool16_t vm, vint64m4_t vd,
                                 const int64_t *rs1, vuint32m2_t rs2,
                                 size_t vl);
vint64m8_t __riscv_vluxei32_tumu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                 vuint32m4_t rs2, size_t vl);
vint64m1_t __riscv_vluxei64_tumu(vbool64_t vm, vint64m1_t vd,
                                 const int64_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vint64m2_t __riscv_vluxei64_tumu(vbool32_t vm, vint64m2_t vd,
                                 const int64_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vint64m4_t __riscv_vluxei64_tumu(vbool16_t vm, vint64m4_t vd,
                                 const int64_t *rs1, vuint64m4_t rs2,
                                 size_t vl);
vint64m8_t __riscv_vluxei64_tumu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                 vuint64m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei8_tumu(vbool64_t vm, vuint8mf8_t vd,
                                 const uint8_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vuint8mf4_t __riscv_vloxei8_tumu(vbool32_t vm, vuint8mf4_t vd,
                                 const uint8_t *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vuint8mf2_t __riscv_vloxei8_tumu(vbool16_t vm, vuint8mf2_t vd,
                                 const uint8_t *rs1, vuint8mf2_t rs2,
                                 size_t vl);
vuint8m1_t __riscv_vloxei8_tumu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                vuint8m1_t rs2, size_t vl);
vuint8m2_t __riscv_vloxei8_tumu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                                vuint8m2_t rs2, size_t vl);
vuint8m4_t __riscv_vloxei8_tumu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                                vuint8m4_t rs2, size_t vl);
vuint8m8_t __riscv_vloxei8_tumu(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                                vuint8m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei16_tumu(vbool64_t vm, vuint8mf8_t vd,
                                  const uint8_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vuint8mf4_t __riscv_vloxei16_tumu(vbool32_t vm, vuint8mf4_t vd,
                                  const uint8_t *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vuint8mf2_t __riscv_vloxei16_tumu(vbool16_t vm, vuint8mf2_t vd,
                                  const uint8_t *rs1, vuint16m1_t rs2,
                                  size_t vl);
vuint8m1_t __riscv_vloxei16_tumu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                 vuint16m2_t rs2, size_t vl);
vuint8m2_t __riscv_vloxei16_tumu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                                 vuint16m4_t rs2, size_t vl);
vuint8m4_t __riscv_vloxei16_tumu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                                 vuint16m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei32_tumu(vbool64_t vm, vuint8mf8_t vd,
                                  const uint8_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vuint8mf4_t __riscv_vloxei32_tumu(vbool32_t vm, vuint8mf4_t vd,
                                  const uint8_t *rs1, vuint32m1_t rs2,
                                  size_t vl);
vuint8mf2_t __riscv_vloxei32_tumu(vbool16_t vm, vuint8mf2_t vd,
                                  const uint8_t *rs1, vuint32m2_t rs2,
                                  size_t vl);
vuint8m1_t __riscv_vloxei32_tumu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                 vuint32m4_t rs2, size_t vl);
vuint8m2_t __riscv_vloxei32_tumu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                                 vuint32m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei64_tumu(vbool64_t vm, vuint8mf8_t vd,
                                  const uint8_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vuint8mf4_t __riscv_vloxei64_tumu(vbool32_t vm, vuint8mf4_t vd,
                                  const uint8_t *rs1, vuint64m2_t rs2,
                                  size_t vl);
vuint8mf2_t __riscv_vloxei64_tumu(vbool16_t vm, vuint8mf2_t vd,
                                  const uint8_t *rs1, vuint64m4_t rs2,
                                  size_t vl);
vuint8m1_t __riscv_vloxei64_tumu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                 vuint64m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vloxei8_tumu(vbool64_t vm, vuint16mf4_t vd,
                                  const uint16_t *rs1, vuint8mf8_t rs2,
                                  size_t vl);
vuint16mf2_t __riscv_vloxei8_tumu(vbool32_t vm, vuint16mf2_t vd,
                                  const uint16_t *rs1, vuint8mf4_t rs2,
                                  size_t vl);
vuint16m1_t __riscv_vloxei8_tumu(vbool16_t vm, vuint16m1_t vd,
                                 const uint16_t *rs1, vuint8mf2_t rs2,
                                 size_t vl);
vuint16m2_t __riscv_vloxei8_tumu(vbool8_t vm, vuint16m2_t vd,
                                 const uint16_t *rs1, vuint8m1_t rs2,
                                 size_t vl);
vuint16m4_t __riscv_vloxei8_tumu(vbool4_t vm, vuint16m4_t vd,
                                 const uint16_t *rs1, vuint8m2_t rs2,
                                 size_t vl);
vuint16m8_t __riscv_vloxei8_tumu(vbool2_t vm, vuint16m8_t vd,
                                 const uint16_t *rs1, vuint8m4_t rs2,
                                 size_t vl);
vuint16mf4_t __riscv_vloxei16_tumu(vbool64_t vm, vuint16mf4_t vd,
                                   const uint16_t *rs1, vuint16mf4_t rs2,
                                   size_t vl);
vuint16mf2_t __riscv_vloxei16_tumu(vbool32_t vm, vuint16mf2_t vd,
                                   const uint16_t *rs1, vuint16mf2_t rs2,
                                   size_t vl);
vuint16m1_t __riscv_vloxei16_tumu(vbool16_t vm, vuint16m1_t vd,
                                  const uint16_t *rs1, vuint16m1_t rs2,
                                  size_t vl);
vuint16m2_t __riscv_vloxei16_tumu(vbool8_t vm, vuint16m2_t vd,
                                  const uint16_t *rs1, vuint16m2_t rs2,
                                  size_t vl);
vuint16m4_t __riscv_vloxei16_tumu(vbool4_t vm, vuint16m4_t vd,
                                  const uint16_t *rs1, vuint16m4_t rs2,
                                  size_t vl);
vuint16m8_t __riscv_vloxei16_tumu(vbool2_t vm, vuint16m8_t vd,
                                  const uint16_t *rs1, vuint16m8_t rs2,
                                  size_t vl);
vuint16mf4_t __riscv_vloxei32_tumu(vbool64_t vm, vuint16mf4_t vd,
                                   const uint16_t *rs1, vuint32mf2_t rs2,
                                   size_t vl);
vuint16mf2_t __riscv_vloxei32_tumu(vbool32_t vm, vuint16mf2_t vd,
                                   const uint16_t *rs1, vuint32m1_t rs2,
                                   size_t vl);
vuint16m1_t __riscv_vloxei32_tumu(vbool16_t vm, vuint16m1_t vd,
                                  const uint16_t *rs1, vuint32m2_t rs2,
                                  size_t vl);
vuint16m2_t __riscv_vloxei32_tumu(vbool8_t vm, vuint16m2_t vd,
                                  const uint16_t *rs1, vuint32m4_t rs2,
                                  size_t vl);
vuint16m4_t __riscv_vloxei32_tumu(vbool4_t vm, vuint16m4_t vd,
                                  const uint16_t *rs1, vuint32m8_t rs2,
                                  size_t vl);
vuint16mf4_t __riscv_vloxei64_tumu(vbool64_t vm, vuint16mf4_t vd,
                                   const uint16_t *rs1, vuint64m1_t rs2,
                                   size_t vl);
vuint16mf2_t __riscv_vloxei64_tumu(vbool32_t vm, vuint16mf2_t vd,
                                   const uint16_t *rs1, vuint64m2_t rs2,
                                   size_t vl);
vuint16m1_t __riscv_vloxei64_tumu(vbool16_t vm, vuint16m1_t vd,
                                  const uint16_t *rs1, vuint64m4_t rs2,
                                  size_t vl);
vuint16m2_t __riscv_vloxei64_tumu(vbool8_t vm, vuint16m2_t vd,
                                  const uint16_t *rs1, vuint64m8_t rs2,
                                  size_t vl);
vuint32mf2_t __riscv_vloxei8_tumu(vbool64_t vm, vuint32mf2_t vd,
                                  const uint32_t *rs1, vuint8mf8_t rs2,
                                  size_t vl);
vuint32m1_t __riscv_vloxei8_tumu(vbool32_t vm, vuint32m1_t vd,
                                 const uint32_t *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vuint32m2_t __riscv_vloxei8_tumu(vbool16_t vm, vuint32m2_t vd,
                                 const uint32_t *rs1, vuint8mf2_t rs2,
                                 size_t vl);
vuint32m4_t __riscv_vloxei8_tumu(vbool8_t vm, vuint32m4_t vd,
                                 const uint32_t *rs1, vuint8m1_t rs2,
                                 size_t vl);
vuint32m8_t __riscv_vloxei8_tumu(vbool4_t vm, vuint32m8_t vd,
                                 const uint32_t *rs1, vuint8m2_t rs2,
                                 size_t vl);
vuint32mf2_t __riscv_vloxei16_tumu(vbool64_t vm, vuint32mf2_t vd,
                                   const uint32_t *rs1, vuint16mf4_t rs2,
                                   size_t vl);
vuint32m1_t __riscv_vloxei16_tumu(vbool32_t vm, vuint32m1_t vd,
                                  const uint32_t *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vuint32m2_t __riscv_vloxei16_tumu(vbool16_t vm, vuint32m2_t vd,
                                  const uint32_t *rs1, vuint16m1_t rs2,
                                  size_t vl);
vuint32m4_t __riscv_vloxei16_tumu(vbool8_t vm, vuint32m4_t vd,
                                  const uint32_t *rs1, vuint16m2_t rs2,
                                  size_t vl);
vuint32m8_t __riscv_vloxei16_tumu(vbool4_t vm, vuint32m8_t vd,
                                  const uint32_t *rs1, vuint16m4_t rs2,
                                  size_t vl);
vuint32mf2_t __riscv_vloxei32_tumu(vbool64_t vm, vuint32mf2_t vd,
                                   const uint32_t *rs1, vuint32mf2_t rs2,
                                   size_t vl);
vuint32m1_t __riscv_vloxei32_tumu(vbool32_t vm, vuint32m1_t vd,
                                  const uint32_t *rs1, vuint32m1_t rs2,
                                  size_t vl);
vuint32m2_t __riscv_vloxei32_tumu(vbool16_t vm, vuint32m2_t vd,
                                  const uint32_t *rs1, vuint32m2_t rs2,
                                  size_t vl);
vuint32m4_t __riscv_vloxei32_tumu(vbool8_t vm, vuint32m4_t vd,
                                  const uint32_t *rs1, vuint32m4_t rs2,
                                  size_t vl);
vuint32m8_t __riscv_vloxei32_tumu(vbool4_t vm, vuint32m8_t vd,
                                  const uint32_t *rs1, vuint32m8_t rs2,
                                  size_t vl);
vuint32mf2_t __riscv_vloxei64_tumu(vbool64_t vm, vuint32mf2_t vd,
                                   const uint32_t *rs1, vuint64m1_t rs2,
                                   size_t vl);
vuint32m1_t __riscv_vloxei64_tumu(vbool32_t vm, vuint32m1_t vd,
                                  const uint32_t *rs1, vuint64m2_t rs2,
                                  size_t vl);
vuint32m2_t __riscv_vloxei64_tumu(vbool16_t vm, vuint32m2_t vd,
                                  const uint32_t *rs1, vuint64m4_t rs2,
                                  size_t vl);
vuint32m4_t __riscv_vloxei64_tumu(vbool8_t vm, vuint32m4_t vd,
                                  const uint32_t *rs1, vuint64m8_t rs2,
                                  size_t vl);
vuint64m1_t __riscv_vloxei8_tumu(vbool64_t vm, vuint64m1_t vd,
                                 const uint64_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vuint64m2_t __riscv_vloxei8_tumu(vbool32_t vm, vuint64m2_t vd,
                                 const uint64_t *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vuint64m4_t __riscv_vloxei8_tumu(vbool16_t vm, vuint64m4_t vd,
                                 const uint64_t *rs1, vuint8mf2_t rs2,
                                 size_t vl);
vuint64m8_t __riscv_vloxei8_tumu(vbool8_t vm, vuint64m8_t vd,
                                 const uint64_t *rs1, vuint8m1_t rs2,
                                 size_t vl);
vuint64m1_t __riscv_vloxei16_tumu(vbool64_t vm, vuint64m1_t vd,
                                  const uint64_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vuint64m2_t __riscv_vloxei16_tumu(vbool32_t vm, vuint64m2_t vd,
                                  const uint64_t *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vuint64m4_t __riscv_vloxei16_tumu(vbool16_t vm, vuint64m4_t vd,
                                  const uint64_t *rs1, vuint16m1_t rs2,
                                  size_t vl);
vuint64m8_t __riscv_vloxei16_tumu(vbool8_t vm, vuint64m8_t vd,
                                  const uint64_t *rs1, vuint16m2_t rs2,
                                  size_t vl);
vuint64m1_t __riscv_vloxei32_tumu(vbool64_t vm, vuint64m1_t vd,
                                  const uint64_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vuint64m2_t __riscv_vloxei32_tumu(vbool32_t vm, vuint64m2_t vd,
                                  const uint64_t *rs1, vuint32m1_t rs2,
                                  size_t vl);
vuint64m4_t __riscv_vloxei32_tumu(vbool16_t vm, vuint64m4_t vd,
                                  const uint64_t *rs1, vuint32m2_t rs2,
                                  size_t vl);
vuint64m8_t __riscv_vloxei32_tumu(vbool8_t vm, vuint64m8_t vd,
                                  const uint64_t *rs1, vuint32m4_t rs2,
                                  size_t vl);
vuint64m1_t __riscv_vloxei64_tumu(vbool64_t vm, vuint64m1_t vd,
                                  const uint64_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vuint64m2_t __riscv_vloxei64_tumu(vbool32_t vm, vuint64m2_t vd,
                                  const uint64_t *rs1, vuint64m2_t rs2,
                                  size_t vl);
vuint64m4_t __riscv_vloxei64_tumu(vbool16_t vm, vuint64m4_t vd,
                                  const uint64_t *rs1, vuint64m4_t rs2,
                                  size_t vl);
vuint64m8_t __riscv_vloxei64_tumu(vbool8_t vm, vuint64m8_t vd,
                                  const uint64_t *rs1, vuint64m8_t rs2,
                                  size_t vl);
vuint8mf8_t __riscv_vluxei8_tumu(vbool64_t vm, vuint8mf8_t vd,
                                 const uint8_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vuint8mf4_t __riscv_vluxei8_tumu(vbool32_t vm, vuint8mf4_t vd,
                                 const uint8_t *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vuint8mf2_t __riscv_vluxei8_tumu(vbool16_t vm, vuint8mf2_t vd,
                                 const uint8_t *rs1, vuint8mf2_t rs2,
                                 size_t vl);
vuint8m1_t __riscv_vluxei8_tumu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                vuint8m1_t rs2, size_t vl);
vuint8m2_t __riscv_vluxei8_tumu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                                vuint8m2_t rs2, size_t vl);
vuint8m4_t __riscv_vluxei8_tumu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                                vuint8m4_t rs2, size_t vl);
vuint8m8_t __riscv_vluxei8_tumu(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                                vuint8m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei16_tumu(vbool64_t vm, vuint8mf8_t vd,
                                  const uint8_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vuint8mf4_t __riscv_vluxei16_tumu(vbool32_t vm, vuint8mf4_t vd,
                                  const uint8_t *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vuint8mf2_t __riscv_vluxei16_tumu(vbool16_t vm, vuint8mf2_t vd,
                                  const uint8_t *rs1, vuint16m1_t rs2,
                                  size_t vl);
vuint8m1_t __riscv_vluxei16_tumu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                 vuint16m2_t rs2, size_t vl);
vuint8m2_t __riscv_vluxei16_tumu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                                 vuint16m4_t rs2, size_t vl);
vuint8m4_t __riscv_vluxei16_tumu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                                 vuint16m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei32_tumu(vbool64_t vm, vuint8mf8_t vd,
                                  const uint8_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vuint8mf4_t __riscv_vluxei32_tumu(vbool32_t vm, vuint8mf4_t vd,
                                  const uint8_t *rs1, vuint32m1_t rs2,
                                  size_t vl);
vuint8mf2_t __riscv_vluxei32_tumu(vbool16_t vm, vuint8mf2_t vd,
                                  const uint8_t *rs1, vuint32m2_t rs2,
                                  size_t vl);
vuint8m1_t __riscv_vluxei32_tumu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                 vuint32m4_t rs2, size_t vl);
vuint8m2_t __riscv_vluxei32_tumu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                                 vuint32m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei64_tumu(vbool64_t vm, vuint8mf8_t vd,
                                  const uint8_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vuint8mf4_t __riscv_vluxei64_tumu(vbool32_t vm, vuint8mf4_t vd,
                                  const uint8_t *rs1, vuint64m2_t rs2,
                                  size_t vl);
vuint8mf2_t __riscv_vluxei64_tumu(vbool16_t vm, vuint8mf2_t vd,
                                  const uint8_t *rs1, vuint64m4_t rs2,
                                  size_t vl);
vuint8m1_t __riscv_vluxei64_tumu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                                 vuint64m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vluxei8_tumu(vbool64_t vm, vuint16mf4_t vd,
                                  const uint16_t *rs1, vuint8mf8_t rs2,
                                  size_t vl);
vuint16mf2_t __riscv_vluxei8_tumu(vbool32_t vm, vuint16mf2_t vd,
                                  const uint16_t *rs1, vuint8mf4_t rs2,
                                  size_t vl);
vuint16m1_t __riscv_vluxei8_tumu(vbool16_t vm, vuint16m1_t vd,
                                 const uint16_t *rs1, vuint8mf2_t rs2,
                                 size_t vl);
vuint16m2_t __riscv_vluxei8_tumu(vbool8_t vm, vuint16m2_t vd,
                                 const uint16_t *rs1, vuint8m1_t rs2,
                                 size_t vl);
vuint16m4_t __riscv_vluxei8_tumu(vbool4_t vm, vuint16m4_t vd,
                                 const uint16_t *rs1, vuint8m2_t rs2,
                                 size_t vl);
vuint16m8_t __riscv_vluxei8_tumu(vbool2_t vm, vuint16m8_t vd,
                                 const uint16_t *rs1, vuint8m4_t rs2,
                                 size_t vl);
vuint16mf4_t __riscv_vluxei16_tumu(vbool64_t vm, vuint16mf4_t vd,
                                   const uint16_t *rs1, vuint16mf4_t rs2,
                                   size_t vl);
vuint16mf2_t __riscv_vluxei16_tumu(vbool32_t vm, vuint16mf2_t vd,
                                   const uint16_t *rs1, vuint16mf2_t rs2,
                                   size_t vl);
vuint16m1_t __riscv_vluxei16_tumu(vbool16_t vm, vuint16m1_t vd,
                                  const uint16_t *rs1, vuint16m1_t rs2,
                                  size_t vl);
vuint16m2_t __riscv_vluxei16_tumu(vbool8_t vm, vuint16m2_t vd,
                                  const uint16_t *rs1, vuint16m2_t rs2,
                                  size_t vl);
vuint16m4_t __riscv_vluxei16_tumu(vbool4_t vm, vuint16m4_t vd,
                                  const uint16_t *rs1, vuint16m4_t rs2,
                                  size_t vl);
vuint16m8_t __riscv_vluxei16_tumu(vbool2_t vm, vuint16m8_t vd,
                                  const uint16_t *rs1, vuint16m8_t rs2,
                                  size_t vl);
vuint16mf4_t __riscv_vluxei32_tumu(vbool64_t vm, vuint16mf4_t vd,
                                   const uint16_t *rs1, vuint32mf2_t rs2,
                                   size_t vl);
vuint16mf2_t __riscv_vluxei32_tumu(vbool32_t vm, vuint16mf2_t vd,
                                   const uint16_t *rs1, vuint32m1_t rs2,
                                   size_t vl);
vuint16m1_t __riscv_vluxei32_tumu(vbool16_t vm, vuint16m1_t vd,
                                  const uint16_t *rs1, vuint32m2_t rs2,
                                  size_t vl);
vuint16m2_t __riscv_vluxei32_tumu(vbool8_t vm, vuint16m2_t vd,
                                  const uint16_t *rs1, vuint32m4_t rs2,
                                  size_t vl);
vuint16m4_t __riscv_vluxei32_tumu(vbool4_t vm, vuint16m4_t vd,
                                  const uint16_t *rs1, vuint32m8_t rs2,
                                  size_t vl);
vuint16mf4_t __riscv_vluxei64_tumu(vbool64_t vm, vuint16mf4_t vd,
                                   const uint16_t *rs1, vuint64m1_t rs2,
                                   size_t vl);
vuint16mf2_t __riscv_vluxei64_tumu(vbool32_t vm, vuint16mf2_t vd,
                                   const uint16_t *rs1, vuint64m2_t rs2,
                                   size_t vl);
vuint16m1_t __riscv_vluxei64_tumu(vbool16_t vm, vuint16m1_t vd,
                                  const uint16_t *rs1, vuint64m4_t rs2,
                                  size_t vl);
vuint16m2_t __riscv_vluxei64_tumu(vbool8_t vm, vuint16m2_t vd,
                                  const uint16_t *rs1, vuint64m8_t rs2,
                                  size_t vl);
vuint32mf2_t __riscv_vluxei8_tumu(vbool64_t vm, vuint32mf2_t vd,
                                  const uint32_t *rs1, vuint8mf8_t rs2,
                                  size_t vl);
vuint32m1_t __riscv_vluxei8_tumu(vbool32_t vm, vuint32m1_t vd,
                                 const uint32_t *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vuint32m2_t __riscv_vluxei8_tumu(vbool16_t vm, vuint32m2_t vd,
                                 const uint32_t *rs1, vuint8mf2_t rs2,
                                 size_t vl);
vuint32m4_t __riscv_vluxei8_tumu(vbool8_t vm, vuint32m4_t vd,
                                 const uint32_t *rs1, vuint8m1_t rs2,
                                 size_t vl);
vuint32m8_t __riscv_vluxei8_tumu(vbool4_t vm, vuint32m8_t vd,
                                 const uint32_t *rs1, vuint8m2_t rs2,
                                 size_t vl);
vuint32mf2_t __riscv_vluxei16_tumu(vbool64_t vm, vuint32mf2_t vd,
                                   const uint32_t *rs1, vuint16mf4_t rs2,
                                   size_t vl);
vuint32m1_t __riscv_vluxei16_tumu(vbool32_t vm, vuint32m1_t vd,
                                  const uint32_t *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vuint32m2_t __riscv_vluxei16_tumu(vbool16_t vm, vuint32m2_t vd,
                                  const uint32_t *rs1, vuint16m1_t rs2,
                                  size_t vl);
vuint32m4_t __riscv_vluxei16_tumu(vbool8_t vm, vuint32m4_t vd,
                                  const uint32_t *rs1, vuint16m2_t rs2,
                                  size_t vl);
vuint32m8_t __riscv_vluxei16_tumu(vbool4_t vm, vuint32m8_t vd,
                                  const uint32_t *rs1, vuint16m4_t rs2,
                                  size_t vl);
vuint32mf2_t __riscv_vluxei32_tumu(vbool64_t vm, vuint32mf2_t vd,
                                   const uint32_t *rs1, vuint32mf2_t rs2,
                                   size_t vl);
vuint32m1_t __riscv_vluxei32_tumu(vbool32_t vm, vuint32m1_t vd,
                                  const uint32_t *rs1, vuint32m1_t rs2,
                                  size_t vl);
vuint32m2_t __riscv_vluxei32_tumu(vbool16_t vm, vuint32m2_t vd,
                                  const uint32_t *rs1, vuint32m2_t rs2,
                                  size_t vl);
vuint32m4_t __riscv_vluxei32_tumu(vbool8_t vm, vuint32m4_t vd,
                                  const uint32_t *rs1, vuint32m4_t rs2,
                                  size_t vl);
vuint32m8_t __riscv_vluxei32_tumu(vbool4_t vm, vuint32m8_t vd,
                                  const uint32_t *rs1, vuint32m8_t rs2,
                                  size_t vl);
vuint32mf2_t __riscv_vluxei64_tumu(vbool64_t vm, vuint32mf2_t vd,
                                   const uint32_t *rs1, vuint64m1_t rs2,
                                   size_t vl);
vuint32m1_t __riscv_vluxei64_tumu(vbool32_t vm, vuint32m1_t vd,
                                  const uint32_t *rs1, vuint64m2_t rs2,
                                  size_t vl);
vuint32m2_t __riscv_vluxei64_tumu(vbool16_t vm, vuint32m2_t vd,
                                  const uint32_t *rs1, vuint64m4_t rs2,
                                  size_t vl);
vuint32m4_t __riscv_vluxei64_tumu(vbool8_t vm, vuint32m4_t vd,
                                  const uint32_t *rs1, vuint64m8_t rs2,
                                  size_t vl);
vuint64m1_t __riscv_vluxei8_tumu(vbool64_t vm, vuint64m1_t vd,
                                 const uint64_t *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vuint64m2_t __riscv_vluxei8_tumu(vbool32_t vm, vuint64m2_t vd,
                                 const uint64_t *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vuint64m4_t __riscv_vluxei8_tumu(vbool16_t vm, vuint64m4_t vd,
                                 const uint64_t *rs1, vuint8mf2_t rs2,
                                 size_t vl);
vuint64m8_t __riscv_vluxei8_tumu(vbool8_t vm, vuint64m8_t vd,
                                 const uint64_t *rs1, vuint8m1_t rs2,
                                 size_t vl);
vuint64m1_t __riscv_vluxei16_tumu(vbool64_t vm, vuint64m1_t vd,
                                  const uint64_t *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vuint64m2_t __riscv_vluxei16_tumu(vbool32_t vm, vuint64m2_t vd,
                                  const uint64_t *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vuint64m4_t __riscv_vluxei16_tumu(vbool16_t vm, vuint64m4_t vd,
                                  const uint64_t *rs1, vuint16m1_t rs2,
                                  size_t vl);
vuint64m8_t __riscv_vluxei16_tumu(vbool8_t vm, vuint64m8_t vd,
                                  const uint64_t *rs1, vuint16m2_t rs2,
                                  size_t vl);
vuint64m1_t __riscv_vluxei32_tumu(vbool64_t vm, vuint64m1_t vd,
                                  const uint64_t *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vuint64m2_t __riscv_vluxei32_tumu(vbool32_t vm, vuint64m2_t vd,
                                  const uint64_t *rs1, vuint32m1_t rs2,
                                  size_t vl);
vuint64m4_t __riscv_vluxei32_tumu(vbool16_t vm, vuint64m4_t vd,
                                  const uint64_t *rs1, vuint32m2_t rs2,
                                  size_t vl);
vuint64m8_t __riscv_vluxei32_tumu(vbool8_t vm, vuint64m8_t vd,
                                  const uint64_t *rs1, vuint32m4_t rs2,
                                  size_t vl);
vuint64m1_t __riscv_vluxei64_tumu(vbool64_t vm, vuint64m1_t vd,
                                  const uint64_t *rs1, vuint64m1_t rs2,
                                  size_t vl);
vuint64m2_t __riscv_vluxei64_tumu(vbool32_t vm, vuint64m2_t vd,
                                  const uint64_t *rs1, vuint64m2_t rs2,
                                  size_t vl);
vuint64m4_t __riscv_vluxei64_tumu(vbool16_t vm, vuint64m4_t vd,
                                  const uint64_t *rs1, vuint64m4_t rs2,
                                  size_t vl);
vuint64m8_t __riscv_vluxei64_tumu(vbool8_t vm, vuint64m8_t vd,
                                  const uint64_t *rs1, vuint64m8_t rs2,
                                  size_t vl);
// masked functions
vfloat16mf4_t __riscv_vloxei8_mu(vbool64_t vm, vfloat16mf4_t vd,
                                 const _Float16 *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vfloat16mf2_t __riscv_vloxei8_mu(vbool32_t vm, vfloat16mf2_t vd,
                                 const _Float16 *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vfloat16m1_t __riscv_vloxei8_mu(vbool16_t vm, vfloat16m1_t vd,
                                const _Float16 *rs1, vuint8mf2_t rs2,
                                size_t vl);
vfloat16m2_t __riscv_vloxei8_mu(vbool8_t vm, vfloat16m2_t vd,
                                const _Float16 *rs1, vuint8m1_t rs2, size_t vl);
vfloat16m4_t __riscv_vloxei8_mu(vbool4_t vm, vfloat16m4_t vd,
                                const _Float16 *rs1, vuint8m2_t rs2, size_t vl);
vfloat16m8_t __riscv_vloxei8_mu(vbool2_t vm, vfloat16m8_t vd,
                                const _Float16 *rs1, vuint8m4_t rs2, size_t vl);
vfloat16mf4_t __riscv_vloxei16_mu(vbool64_t vm, vfloat16mf4_t vd,
                                  const _Float16 *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vfloat16mf2_t __riscv_vloxei16_mu(vbool32_t vm, vfloat16mf2_t vd,
                                  const _Float16 *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vfloat16m1_t __riscv_vloxei16_mu(vbool16_t vm, vfloat16m1_t vd,
                                 const _Float16 *rs1, vuint16m1_t rs2,
                                 size_t vl);
vfloat16m2_t __riscv_vloxei16_mu(vbool8_t vm, vfloat16m2_t vd,
                                 const _Float16 *rs1, vuint16m2_t rs2,
                                 size_t vl);
vfloat16m4_t __riscv_vloxei16_mu(vbool4_t vm, vfloat16m4_t vd,
                                 const _Float16 *rs1, vuint16m4_t rs2,
                                 size_t vl);
vfloat16m8_t __riscv_vloxei16_mu(vbool2_t vm, vfloat16m8_t vd,
                                 const _Float16 *rs1, vuint16m8_t rs2,
                                 size_t vl);
vfloat16mf4_t __riscv_vloxei32_mu(vbool64_t vm, vfloat16mf4_t vd,
                                  const _Float16 *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vfloat16mf2_t __riscv_vloxei32_mu(vbool32_t vm, vfloat16mf2_t vd,
                                  const _Float16 *rs1, vuint32m1_t rs2,
                                  size_t vl);
vfloat16m1_t __riscv_vloxei32_mu(vbool16_t vm, vfloat16m1_t vd,
                                 const _Float16 *rs1, vuint32m2_t rs2,
                                 size_t vl);
vfloat16m2_t __riscv_vloxei32_mu(vbool8_t vm, vfloat16m2_t vd,
                                 const _Float16 *rs1, vuint32m4_t rs2,
                                 size_t vl);
vfloat16m4_t __riscv_vloxei32_mu(vbool4_t vm, vfloat16m4_t vd,
                                 const _Float16 *rs1, vuint32m8_t rs2,
                                 size_t vl);
vfloat16mf4_t __riscv_vloxei64_mu(vbool64_t vm, vfloat16mf4_t vd,
                                  const _Float16 *rs1, vuint64m1_t rs2,
                                  size_t vl);
vfloat16mf2_t __riscv_vloxei64_mu(vbool32_t vm, vfloat16mf2_t vd,
                                  const _Float16 *rs1, vuint64m2_t rs2,
                                  size_t vl);
vfloat16m1_t __riscv_vloxei64_mu(vbool16_t vm, vfloat16m1_t vd,
                                 const _Float16 *rs1, vuint64m4_t rs2,
                                 size_t vl);
vfloat16m2_t __riscv_vloxei64_mu(vbool8_t vm, vfloat16m2_t vd,
                                 const _Float16 *rs1, vuint64m8_t rs2,
                                 size_t vl);
vfloat32mf2_t __riscv_vloxei8_mu(vbool64_t vm, vfloat32mf2_t vd,
                                 const float *rs1, vuint8mf8_t rs2, size_t vl);
vfloat32m1_t __riscv_vloxei8_mu(vbool32_t vm, vfloat32m1_t vd, const float *rs1,
                                vuint8mf4_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei8_mu(vbool16_t vm, vfloat32m2_t vd, const float *rs1,
                                vuint8mf2_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei8_mu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                vuint8m1_t rs2, size_t vl);
vfloat32m8_t __riscv_vloxei8_mu(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                vuint8m2_t rs2, size_t vl);
vfloat32mf2_t __riscv_vloxei16_mu(vbool64_t vm, vfloat32mf2_t vd,
                                  const float *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vfloat32m1_t __riscv_vloxei16_mu(vbool32_t vm, vfloat32m1_t vd,
                                 const float *rs1, vuint16mf2_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei16_mu(vbool16_t vm, vfloat32m2_t vd,
                                 const float *rs1, vuint16m1_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei16_mu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                 vuint16m2_t rs2, size_t vl);
vfloat32m8_t __riscv_vloxei16_mu(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                 vuint16m4_t rs2, size_t vl);
vfloat32mf2_t __riscv_vloxei32_mu(vbool64_t vm, vfloat32mf2_t vd,
                                  const float *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vfloat32m1_t __riscv_vloxei32_mu(vbool32_t vm, vfloat32m1_t vd,
                                 const float *rs1, vuint32m1_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei32_mu(vbool16_t vm, vfloat32m2_t vd,
                                 const float *rs1, vuint32m2_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei32_mu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                 vuint32m4_t rs2, size_t vl);
vfloat32m8_t __riscv_vloxei32_mu(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                 vuint32m8_t rs2, size_t vl);
vfloat32mf2_t __riscv_vloxei64_mu(vbool64_t vm, vfloat32mf2_t vd,
                                  const float *rs1, vuint64m1_t rs2, size_t vl);
vfloat32m1_t __riscv_vloxei64_mu(vbool32_t vm, vfloat32m1_t vd,
                                 const float *rs1, vuint64m2_t rs2, size_t vl);
vfloat32m2_t __riscv_vloxei64_mu(vbool16_t vm, vfloat32m2_t vd,
                                 const float *rs1, vuint64m4_t rs2, size_t vl);
vfloat32m4_t __riscv_vloxei64_mu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                 vuint64m8_t rs2, size_t vl);
vfloat64m1_t __riscv_vloxei8_mu(vbool64_t vm, vfloat64m1_t vd,
                                const double *rs1, vuint8mf8_t rs2, size_t vl);
vfloat64m2_t __riscv_vloxei8_mu(vbool32_t vm, vfloat64m2_t vd,
                                const double *rs1, vuint8mf4_t rs2, size_t vl);
vfloat64m4_t __riscv_vloxei8_mu(vbool16_t vm, vfloat64m4_t vd,
                                const double *rs1, vuint8mf2_t rs2, size_t vl);
vfloat64m8_t __riscv_vloxei8_mu(vbool8_t vm, vfloat64m8_t vd, const double *rs1,
                                vuint8m1_t rs2, size_t vl);
vfloat64m1_t __riscv_vloxei16_mu(vbool64_t vm, vfloat64m1_t vd,
                                 const double *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vfloat64m2_t __riscv_vloxei16_mu(vbool32_t vm, vfloat64m2_t vd,
                                 const double *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vfloat64m4_t __riscv_vloxei16_mu(vbool16_t vm, vfloat64m4_t vd,
                                 const double *rs1, vuint16m1_t rs2, size_t vl);
vfloat64m8_t __riscv_vloxei16_mu(vbool8_t vm, vfloat64m8_t vd,
                                 const double *rs1, vuint16m2_t rs2, size_t vl);
vfloat64m1_t __riscv_vloxei32_mu(vbool64_t vm, vfloat64m1_t vd,
                                 const double *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vfloat64m2_t __riscv_vloxei32_mu(vbool32_t vm, vfloat64m2_t vd,
                                 const double *rs1, vuint32m1_t rs2, size_t vl);
vfloat64m4_t __riscv_vloxei32_mu(vbool16_t vm, vfloat64m4_t vd,
                                 const double *rs1, vuint32m2_t rs2, size_t vl);
vfloat64m8_t __riscv_vloxei32_mu(vbool8_t vm, vfloat64m8_t vd,
                                 const double *rs1, vuint32m4_t rs2, size_t vl);
vfloat64m1_t __riscv_vloxei64_mu(vbool64_t vm, vfloat64m1_t vd,
                                 const double *rs1, vuint64m1_t rs2, size_t vl);
vfloat64m2_t __riscv_vloxei64_mu(vbool32_t vm, vfloat64m2_t vd,
                                 const double *rs1, vuint64m2_t rs2, size_t vl);
vfloat64m4_t __riscv_vloxei64_mu(vbool16_t vm, vfloat64m4_t vd,
                                 const double *rs1, vuint64m4_t rs2, size_t vl);
vfloat64m8_t __riscv_vloxei64_mu(vbool8_t vm, vfloat64m8_t vd,
                                 const double *rs1, vuint64m8_t rs2, size_t vl);
vfloat16mf4_t __riscv_vluxei8_mu(vbool64_t vm, vfloat16mf4_t vd,
                                 const _Float16 *rs1, vuint8mf8_t rs2,
                                 size_t vl);
vfloat16mf2_t __riscv_vluxei8_mu(vbool32_t vm, vfloat16mf2_t vd,
                                 const _Float16 *rs1, vuint8mf4_t rs2,
                                 size_t vl);
vfloat16m1_t __riscv_vluxei8_mu(vbool16_t vm, vfloat16m1_t vd,
                                const _Float16 *rs1, vuint8mf2_t rs2,
                                size_t vl);
vfloat16m2_t __riscv_vluxei8_mu(vbool8_t vm, vfloat16m2_t vd,
                                const _Float16 *rs1, vuint8m1_t rs2, size_t vl);
vfloat16m4_t __riscv_vluxei8_mu(vbool4_t vm, vfloat16m4_t vd,
                                const _Float16 *rs1, vuint8m2_t rs2, size_t vl);
vfloat16m8_t __riscv_vluxei8_mu(vbool2_t vm, vfloat16m8_t vd,
                                const _Float16 *rs1, vuint8m4_t rs2, size_t vl);
vfloat16mf4_t __riscv_vluxei16_mu(vbool64_t vm, vfloat16mf4_t vd,
                                  const _Float16 *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vfloat16mf2_t __riscv_vluxei16_mu(vbool32_t vm, vfloat16mf2_t vd,
                                  const _Float16 *rs1, vuint16mf2_t rs2,
                                  size_t vl);
vfloat16m1_t __riscv_vluxei16_mu(vbool16_t vm, vfloat16m1_t vd,
                                 const _Float16 *rs1, vuint16m1_t rs2,
                                 size_t vl);
vfloat16m2_t __riscv_vluxei16_mu(vbool8_t vm, vfloat16m2_t vd,
                                 const _Float16 *rs1, vuint16m2_t rs2,
                                 size_t vl);
vfloat16m4_t __riscv_vluxei16_mu(vbool4_t vm, vfloat16m4_t vd,
                                 const _Float16 *rs1, vuint16m4_t rs2,
                                 size_t vl);
vfloat16m8_t __riscv_vluxei16_mu(vbool2_t vm, vfloat16m8_t vd,
                                 const _Float16 *rs1, vuint16m8_t rs2,
                                 size_t vl);
vfloat16mf4_t __riscv_vluxei32_mu(vbool64_t vm, vfloat16mf4_t vd,
                                  const _Float16 *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vfloat16mf2_t __riscv_vluxei32_mu(vbool32_t vm, vfloat16mf2_t vd,
                                  const _Float16 *rs1, vuint32m1_t rs2,
                                  size_t vl);
vfloat16m1_t __riscv_vluxei32_mu(vbool16_t vm, vfloat16m1_t vd,
                                 const _Float16 *rs1, vuint32m2_t rs2,
                                 size_t vl);
vfloat16m2_t __riscv_vluxei32_mu(vbool8_t vm, vfloat16m2_t vd,
                                 const _Float16 *rs1, vuint32m4_t rs2,
                                 size_t vl);
vfloat16m4_t __riscv_vluxei32_mu(vbool4_t vm, vfloat16m4_t vd,
                                 const _Float16 *rs1, vuint32m8_t rs2,
                                 size_t vl);
vfloat16mf4_t __riscv_vluxei64_mu(vbool64_t vm, vfloat16mf4_t vd,
                                  const _Float16 *rs1, vuint64m1_t rs2,
                                  size_t vl);
vfloat16mf2_t __riscv_vluxei64_mu(vbool32_t vm, vfloat16mf2_t vd,
                                  const _Float16 *rs1, vuint64m2_t rs2,
                                  size_t vl);
vfloat16m1_t __riscv_vluxei64_mu(vbool16_t vm, vfloat16m1_t vd,
                                 const _Float16 *rs1, vuint64m4_t rs2,
                                 size_t vl);
vfloat16m2_t __riscv_vluxei64_mu(vbool8_t vm, vfloat16m2_t vd,
                                 const _Float16 *rs1, vuint64m8_t rs2,
                                 size_t vl);
vfloat32mf2_t __riscv_vluxei8_mu(vbool64_t vm, vfloat32mf2_t vd,
                                 const float *rs1, vuint8mf8_t rs2, size_t vl);
vfloat32m1_t __riscv_vluxei8_mu(vbool32_t vm, vfloat32m1_t vd, const float *rs1,
                                vuint8mf4_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei8_mu(vbool16_t vm, vfloat32m2_t vd, const float *rs1,
                                vuint8mf2_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei8_mu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                vuint8m1_t rs2, size_t vl);
vfloat32m8_t __riscv_vluxei8_mu(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                vuint8m2_t rs2, size_t vl);
vfloat32mf2_t __riscv_vluxei16_mu(vbool64_t vm, vfloat32mf2_t vd,
                                  const float *rs1, vuint16mf4_t rs2,
                                  size_t vl);
vfloat32m1_t __riscv_vluxei16_mu(vbool32_t vm, vfloat32m1_t vd,
                                 const float *rs1, vuint16mf2_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei16_mu(vbool16_t vm, vfloat32m2_t vd,
                                 const float *rs1, vuint16m1_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei16_mu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                 vuint16m2_t rs2, size_t vl);
vfloat32m8_t __riscv_vluxei16_mu(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                 vuint16m4_t rs2, size_t vl);
vfloat32mf2_t __riscv_vluxei32_mu(vbool64_t vm, vfloat32mf2_t vd,
                                  const float *rs1, vuint32mf2_t rs2,
                                  size_t vl);
vfloat32m1_t __riscv_vluxei32_mu(vbool32_t vm, vfloat32m1_t vd,
                                 const float *rs1, vuint32m1_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei32_mu(vbool16_t vm, vfloat32m2_t vd,
                                 const float *rs1, vuint32m2_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei32_mu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                 vuint32m4_t rs2, size_t vl);
vfloat32m8_t __riscv_vluxei32_mu(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                 vuint32m8_t rs2, size_t vl);
vfloat32mf2_t __riscv_vluxei64_mu(vbool64_t vm, vfloat32mf2_t vd,
                                  const float *rs1, vuint64m1_t rs2, size_t vl);
vfloat32m1_t __riscv_vluxei64_mu(vbool32_t vm, vfloat32m1_t vd,
                                 const float *rs1, vuint64m2_t rs2, size_t vl);
vfloat32m2_t __riscv_vluxei64_mu(vbool16_t vm, vfloat32m2_t vd,
                                 const float *rs1, vuint64m4_t rs2, size_t vl);
vfloat32m4_t __riscv_vluxei64_mu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                 vuint64m8_t rs2, size_t vl);
vfloat64m1_t __riscv_vluxei8_mu(vbool64_t vm, vfloat64m1_t vd,
                                const double *rs1, vuint8mf8_t rs2, size_t vl);
vfloat64m2_t __riscv_vluxei8_mu(vbool32_t vm, vfloat64m2_t vd,
                                const double *rs1, vuint8mf4_t rs2, size_t vl);
vfloat64m4_t __riscv_vluxei8_mu(vbool16_t vm, vfloat64m4_t vd,
                                const double *rs1, vuint8mf2_t rs2, size_t vl);
vfloat64m8_t __riscv_vluxei8_mu(vbool8_t vm, vfloat64m8_t vd, const double *rs1,
                                vuint8m1_t rs2, size_t vl);
vfloat64m1_t __riscv_vluxei16_mu(vbool64_t vm, vfloat64m1_t vd,
                                 const double *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vfloat64m2_t __riscv_vluxei16_mu(vbool32_t vm, vfloat64m2_t vd,
                                 const double *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vfloat64m4_t __riscv_vluxei16_mu(vbool16_t vm, vfloat64m4_t vd,
                                 const double *rs1, vuint16m1_t rs2, size_t vl);
vfloat64m8_t __riscv_vluxei16_mu(vbool8_t vm, vfloat64m8_t vd,
                                 const double *rs1, vuint16m2_t rs2, size_t vl);
vfloat64m1_t __riscv_vluxei32_mu(vbool64_t vm, vfloat64m1_t vd,
                                 const double *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vfloat64m2_t __riscv_vluxei32_mu(vbool32_t vm, vfloat64m2_t vd,
                                 const double *rs1, vuint32m1_t rs2, size_t vl);
vfloat64m4_t __riscv_vluxei32_mu(vbool16_t vm, vfloat64m4_t vd,
                                 const double *rs1, vuint32m2_t rs2, size_t vl);
vfloat64m8_t __riscv_vluxei32_mu(vbool8_t vm, vfloat64m8_t vd,
                                 const double *rs1, vuint32m4_t rs2, size_t vl);
vfloat64m1_t __riscv_vluxei64_mu(vbool64_t vm, vfloat64m1_t vd,
                                 const double *rs1, vuint64m1_t rs2, size_t vl);
vfloat64m2_t __riscv_vluxei64_mu(vbool32_t vm, vfloat64m2_t vd,
                                 const double *rs1, vuint64m2_t rs2, size_t vl);
vfloat64m4_t __riscv_vluxei64_mu(vbool16_t vm, vfloat64m4_t vd,
                                 const double *rs1, vuint64m4_t rs2, size_t vl);
vfloat64m8_t __riscv_vluxei64_mu(vbool8_t vm, vfloat64m8_t vd,
                                 const double *rs1, vuint64m8_t rs2, size_t vl);
vint8mf8_t __riscv_vloxei8_mu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                              vuint8mf8_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei8_mu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                              vuint8mf4_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei8_mu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint8m1_t __riscv_vloxei8_mu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                             vuint8m1_t rs2, size_t vl);
vint8m2_t __riscv_vloxei8_mu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                             vuint8m2_t rs2, size_t vl);
vint8m4_t __riscv_vloxei8_mu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                             vuint8m4_t rs2, size_t vl);
vint8m8_t __riscv_vloxei8_mu(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                             vuint8m8_t rs2, size_t vl);
vint8mf8_t __riscv_vloxei16_mu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                               vuint16mf4_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei16_mu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei16_mu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint8m1_t __riscv_vloxei16_mu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                              vuint16m2_t rs2, size_t vl);
vint8m2_t __riscv_vloxei16_mu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                              vuint16m4_t rs2, size_t vl);
vint8m4_t __riscv_vloxei16_mu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                              vuint16m8_t rs2, size_t vl);
vint8mf8_t __riscv_vloxei32_mu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                               vuint32mf2_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei32_mu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei32_mu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint8m1_t __riscv_vloxei32_mu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                              vuint32m4_t rs2, size_t vl);
vint8m2_t __riscv_vloxei32_mu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                              vuint32m8_t rs2, size_t vl);
vint8mf8_t __riscv_vloxei64_mu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                               vuint64m1_t rs2, size_t vl);
vint8mf4_t __riscv_vloxei64_mu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint8mf2_t __riscv_vloxei64_mu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint8m1_t __riscv_vloxei64_mu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                              vuint64m8_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei8_mu(vbool64_t vm, vint16mf4_t vd, const int16_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint16mf2_t __riscv_vloxei8_mu(vbool32_t vm, vint16mf2_t vd, const int16_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vint16m1_t __riscv_vloxei8_mu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint16m2_t __riscv_vloxei8_mu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                              vuint8m1_t rs2, size_t vl);
vint16m4_t __riscv_vloxei8_mu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                              vuint8m2_t rs2, size_t vl);
vint16m8_t __riscv_vloxei8_mu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                              vuint8m4_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei16_mu(vbool64_t vm, vint16mf4_t vd,
                                const int16_t *rs1, vuint16mf4_t rs2,
                                size_t vl);
vint16mf2_t __riscv_vloxei16_mu(vbool32_t vm, vint16mf2_t vd,
                                const int16_t *rs1, vuint16mf2_t rs2,
                                size_t vl);
vint16m1_t __riscv_vloxei16_mu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint16m2_t __riscv_vloxei16_mu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint16m4_t __riscv_vloxei16_mu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vint16m8_t __riscv_vloxei16_mu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                               vuint16m8_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei32_mu(vbool64_t vm, vint16mf4_t vd,
                                const int16_t *rs1, vuint32mf2_t rs2,
                                size_t vl);
vint16mf2_t __riscv_vloxei32_mu(vbool32_t vm, vint16mf2_t vd,
                                const int16_t *rs1, vuint32m1_t rs2, size_t vl);
vint16m1_t __riscv_vloxei32_mu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint16m2_t __riscv_vloxei32_mu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint16m4_t __riscv_vloxei32_mu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vint16mf4_t __riscv_vloxei64_mu(vbool64_t vm, vint16mf4_t vd,
                                const int16_t *rs1, vuint64m1_t rs2, size_t vl);
vint16mf2_t __riscv_vloxei64_mu(vbool32_t vm, vint16mf2_t vd,
                                const int16_t *rs1, vuint64m2_t rs2, size_t vl);
vint16m1_t __riscv_vloxei64_mu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint16m2_t __riscv_vloxei64_mu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei8_mu(vbool64_t vm, vint32mf2_t vd, const int32_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint32m1_t __riscv_vloxei8_mu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                              vuint8mf4_t rs2, size_t vl);
vint32m2_t __riscv_vloxei8_mu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint32m4_t __riscv_vloxei8_mu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                              vuint8m1_t rs2, size_t vl);
vint32m8_t __riscv_vloxei8_mu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                              vuint8m2_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei16_mu(vbool64_t vm, vint32mf2_t vd,
                                const int32_t *rs1, vuint16mf4_t rs2,
                                size_t vl);
vint32m1_t __riscv_vloxei16_mu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint32m2_t __riscv_vloxei16_mu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint32m4_t __riscv_vloxei16_mu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint32m8_t __riscv_vloxei16_mu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei32_mu(vbool64_t vm, vint32mf2_t vd,
                                const int32_t *rs1, vuint32mf2_t rs2,
                                size_t vl);
vint32m1_t __riscv_vloxei32_mu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint32m2_t __riscv_vloxei32_mu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint32m4_t __riscv_vloxei32_mu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint32m8_t __riscv_vloxei32_mu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vint32mf2_t __riscv_vloxei64_mu(vbool64_t vm, vint32mf2_t vd,
                                const int32_t *rs1, vuint64m1_t rs2, size_t vl);
vint32m1_t __riscv_vloxei64_mu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint32m2_t __riscv_vloxei64_mu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint32m4_t __riscv_vloxei64_mu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint64m1_t __riscv_vloxei8_mu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                              vuint8mf8_t rs2, size_t vl);
vint64m2_t __riscv_vloxei8_mu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                              vuint8mf4_t rs2, size_t vl);
vint64m4_t __riscv_vloxei8_mu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint64m8_t __riscv_vloxei8_mu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                              vuint8m1_t rs2, size_t vl);
vint64m1_t __riscv_vloxei16_mu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                               vuint16mf4_t rs2, size_t vl);
vint64m2_t __riscv_vloxei16_mu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint64m4_t __riscv_vloxei16_mu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint64m8_t __riscv_vloxei16_mu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint64m1_t __riscv_vloxei32_mu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                               vuint32mf2_t rs2, size_t vl);
vint64m2_t __riscv_vloxei32_mu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint64m4_t __riscv_vloxei32_mu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint64m8_t __riscv_vloxei32_mu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint64m1_t __riscv_vloxei64_mu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                               vuint64m1_t rs2, size_t vl);
vint64m2_t __riscv_vloxei64_mu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint64m4_t __riscv_vloxei64_mu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint64m8_t __riscv_vloxei64_mu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei8_mu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                              vuint8mf8_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei8_mu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                              vuint8mf4_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei8_mu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint8m1_t __riscv_vluxei8_mu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                             vuint8m1_t rs2, size_t vl);
vint8m2_t __riscv_vluxei8_mu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                             vuint8m2_t rs2, size_t vl);
vint8m4_t __riscv_vluxei8_mu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                             vuint8m4_t rs2, size_t vl);
vint8m8_t __riscv_vluxei8_mu(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                             vuint8m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei16_mu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                               vuint16mf4_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei16_mu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei16_mu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint8m1_t __riscv_vluxei16_mu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                              vuint16m2_t rs2, size_t vl);
vint8m2_t __riscv_vluxei16_mu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                              vuint16m4_t rs2, size_t vl);
vint8m4_t __riscv_vluxei16_mu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                              vuint16m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei32_mu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                               vuint32mf2_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei32_mu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei32_mu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint8m1_t __riscv_vluxei32_mu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                              vuint32m4_t rs2, size_t vl);
vint8m2_t __riscv_vluxei32_mu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                              vuint32m8_t rs2, size_t vl);
vint8mf8_t __riscv_vluxei64_mu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                               vuint64m1_t rs2, size_t vl);
vint8mf4_t __riscv_vluxei64_mu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint8mf2_t __riscv_vluxei64_mu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint8m1_t __riscv_vluxei64_mu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                              vuint64m8_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei8_mu(vbool64_t vm, vint16mf4_t vd, const int16_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint16mf2_t __riscv_vluxei8_mu(vbool32_t vm, vint16mf2_t vd, const int16_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vint16m1_t __riscv_vluxei8_mu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint16m2_t __riscv_vluxei8_mu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                              vuint8m1_t rs2, size_t vl);
vint16m4_t __riscv_vluxei8_mu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                              vuint8m2_t rs2, size_t vl);
vint16m8_t __riscv_vluxei8_mu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                              vuint8m4_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei16_mu(vbool64_t vm, vint16mf4_t vd,
                                const int16_t *rs1, vuint16mf4_t rs2,
                                size_t vl);
vint16mf2_t __riscv_vluxei16_mu(vbool32_t vm, vint16mf2_t vd,
                                const int16_t *rs1, vuint16mf2_t rs2,
                                size_t vl);
vint16m1_t __riscv_vluxei16_mu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint16m2_t __riscv_vluxei16_mu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint16m4_t __riscv_vluxei16_mu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vint16m8_t __riscv_vluxei16_mu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                               vuint16m8_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei32_mu(vbool64_t vm, vint16mf4_t vd,
                                const int16_t *rs1, vuint32mf2_t rs2,
                                size_t vl);
vint16mf2_t __riscv_vluxei32_mu(vbool32_t vm, vint16mf2_t vd,
                                const int16_t *rs1, vuint32m1_t rs2, size_t vl);
vint16m1_t __riscv_vluxei32_mu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint16m2_t __riscv_vluxei32_mu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint16m4_t __riscv_vluxei32_mu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vint16mf4_t __riscv_vluxei64_mu(vbool64_t vm, vint16mf4_t vd,
                                const int16_t *rs1, vuint64m1_t rs2, size_t vl);
vint16mf2_t __riscv_vluxei64_mu(vbool32_t vm, vint16mf2_t vd,
                                const int16_t *rs1, vuint64m2_t rs2, size_t vl);
vint16m1_t __riscv_vluxei64_mu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint16m2_t __riscv_vluxei64_mu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei8_mu(vbool64_t vm, vint32mf2_t vd, const int32_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vint32m1_t __riscv_vluxei8_mu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                              vuint8mf4_t rs2, size_t vl);
vint32m2_t __riscv_vluxei8_mu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint32m4_t __riscv_vluxei8_mu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                              vuint8m1_t rs2, size_t vl);
vint32m8_t __riscv_vluxei8_mu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                              vuint8m2_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei16_mu(vbool64_t vm, vint32mf2_t vd,
                                const int32_t *rs1, vuint16mf4_t rs2,
                                size_t vl);
vint32m1_t __riscv_vluxei16_mu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint32m2_t __riscv_vluxei16_mu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint32m4_t __riscv_vluxei16_mu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint32m8_t __riscv_vluxei16_mu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei32_mu(vbool64_t vm, vint32mf2_t vd,
                                const int32_t *rs1, vuint32mf2_t rs2,
                                size_t vl);
vint32m1_t __riscv_vluxei32_mu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint32m2_t __riscv_vluxei32_mu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint32m4_t __riscv_vluxei32_mu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint32m8_t __riscv_vluxei32_mu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vint32mf2_t __riscv_vluxei64_mu(vbool64_t vm, vint32mf2_t vd,
                                const int32_t *rs1, vuint64m1_t rs2, size_t vl);
vint32m1_t __riscv_vluxei64_mu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint32m2_t __riscv_vluxei64_mu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint32m4_t __riscv_vluxei64_mu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vint64m1_t __riscv_vluxei8_mu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                              vuint8mf8_t rs2, size_t vl);
vint64m2_t __riscv_vluxei8_mu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                              vuint8mf4_t rs2, size_t vl);
vint64m4_t __riscv_vluxei8_mu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                              vuint8mf2_t rs2, size_t vl);
vint64m8_t __riscv_vluxei8_mu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                              vuint8m1_t rs2, size_t vl);
vint64m1_t __riscv_vluxei16_mu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                               vuint16mf4_t rs2, size_t vl);
vint64m2_t __riscv_vluxei16_mu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                               vuint16mf2_t rs2, size_t vl);
vint64m4_t __riscv_vluxei16_mu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                               vuint16m1_t rs2, size_t vl);
vint64m8_t __riscv_vluxei16_mu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vint64m1_t __riscv_vluxei32_mu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                               vuint32mf2_t rs2, size_t vl);
vint64m2_t __riscv_vluxei32_mu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                               vuint32m1_t rs2, size_t vl);
vint64m4_t __riscv_vluxei32_mu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                               vuint32m2_t rs2, size_t vl);
vint64m8_t __riscv_vluxei32_mu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vint64m1_t __riscv_vluxei64_mu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                               vuint64m1_t rs2, size_t vl);
vint64m2_t __riscv_vluxei64_mu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                               vuint64m2_t rs2, size_t vl);
vint64m4_t __riscv_vluxei64_mu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                               vuint64m4_t rs2, size_t vl);
vint64m8_t __riscv_vluxei64_mu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei8_mu(vbool64_t vm, vuint8mf8_t vd, const uint8_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vuint8mf4_t __riscv_vloxei8_mu(vbool32_t vm, vuint8mf4_t vd, const uint8_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vuint8mf2_t __riscv_vloxei8_mu(vbool16_t vm, vuint8mf2_t vd, const uint8_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vuint8m1_t __riscv_vloxei8_mu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                              vuint8m1_t rs2, size_t vl);
vuint8m2_t __riscv_vloxei8_mu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                              vuint8m2_t rs2, size_t vl);
vuint8m4_t __riscv_vloxei8_mu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                              vuint8m4_t rs2, size_t vl);
vuint8m8_t __riscv_vloxei8_mu(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                              vuint8m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei16_mu(vbool64_t vm, vuint8mf8_t vd,
                                const uint8_t *rs1, vuint16mf4_t rs2,
                                size_t vl);
vuint8mf4_t __riscv_vloxei16_mu(vbool32_t vm, vuint8mf4_t vd,
                                const uint8_t *rs1, vuint16mf2_t rs2,
                                size_t vl);
vuint8mf2_t __riscv_vloxei16_mu(vbool16_t vm, vuint8mf2_t vd,
                                const uint8_t *rs1, vuint16m1_t rs2, size_t vl);
vuint8m1_t __riscv_vloxei16_mu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vuint8m2_t __riscv_vloxei16_mu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vuint8m4_t __riscv_vloxei16_mu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                               vuint16m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei32_mu(vbool64_t vm, vuint8mf8_t vd,
                                const uint8_t *rs1, vuint32mf2_t rs2,
                                size_t vl);
vuint8mf4_t __riscv_vloxei32_mu(vbool32_t vm, vuint8mf4_t vd,
                                const uint8_t *rs1, vuint32m1_t rs2, size_t vl);
vuint8mf2_t __riscv_vloxei32_mu(vbool16_t vm, vuint8mf2_t vd,
                                const uint8_t *rs1, vuint32m2_t rs2, size_t vl);
vuint8m1_t __riscv_vloxei32_mu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vuint8m2_t __riscv_vloxei32_mu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vloxei64_mu(vbool64_t vm, vuint8mf8_t vd,
                                const uint8_t *rs1, vuint64m1_t rs2, size_t vl);
vuint8mf4_t __riscv_vloxei64_mu(vbool32_t vm, vuint8mf4_t vd,
                                const uint8_t *rs1, vuint64m2_t rs2, size_t vl);
vuint8mf2_t __riscv_vloxei64_mu(vbool16_t vm, vuint8mf2_t vd,
                                const uint8_t *rs1, vuint64m4_t rs2, size_t vl);
vuint8m1_t __riscv_vloxei64_mu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vloxei8_mu(vbool64_t vm, vuint16mf4_t vd,
                                const uint16_t *rs1, vuint8mf8_t rs2,
                                size_t vl);
vuint16mf2_t __riscv_vloxei8_mu(vbool32_t vm, vuint16mf2_t vd,
                                const uint16_t *rs1, vuint8mf4_t rs2,
                                size_t vl);
vuint16m1_t __riscv_vloxei8_mu(vbool16_t vm, vuint16m1_t vd,
                               const uint16_t *rs1, vuint8mf2_t rs2, size_t vl);
vuint16m2_t __riscv_vloxei8_mu(vbool8_t vm, vuint16m2_t vd, const uint16_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint16m4_t __riscv_vloxei8_mu(vbool4_t vm, vuint16m4_t vd, const uint16_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vuint16m8_t __riscv_vloxei8_mu(vbool2_t vm, vuint16m8_t vd, const uint16_t *rs1,
                               vuint8m4_t rs2, size_t vl);
vuint16mf4_t __riscv_vloxei16_mu(vbool64_t vm, vuint16mf4_t vd,
                                 const uint16_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vuint16mf2_t __riscv_vloxei16_mu(vbool32_t vm, vuint16mf2_t vd,
                                 const uint16_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vuint16m1_t __riscv_vloxei16_mu(vbool16_t vm, vuint16m1_t vd,
                                const uint16_t *rs1, vuint16m1_t rs2,
                                size_t vl);
vuint16m2_t __riscv_vloxei16_mu(vbool8_t vm, vuint16m2_t vd,
                                const uint16_t *rs1, vuint16m2_t rs2,
                                size_t vl);
vuint16m4_t __riscv_vloxei16_mu(vbool4_t vm, vuint16m4_t vd,
                                const uint16_t *rs1, vuint16m4_t rs2,
                                size_t vl);
vuint16m8_t __riscv_vloxei16_mu(vbool2_t vm, vuint16m8_t vd,
                                const uint16_t *rs1, vuint16m8_t rs2,
                                size_t vl);
vuint16mf4_t __riscv_vloxei32_mu(vbool64_t vm, vuint16mf4_t vd,
                                 const uint16_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vuint16mf2_t __riscv_vloxei32_mu(vbool32_t vm, vuint16mf2_t vd,
                                 const uint16_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vuint16m1_t __riscv_vloxei32_mu(vbool16_t vm, vuint16m1_t vd,
                                const uint16_t *rs1, vuint32m2_t rs2,
                                size_t vl);
vuint16m2_t __riscv_vloxei32_mu(vbool8_t vm, vuint16m2_t vd,
                                const uint16_t *rs1, vuint32m4_t rs2,
                                size_t vl);
vuint16m4_t __riscv_vloxei32_mu(vbool4_t vm, vuint16m4_t vd,
                                const uint16_t *rs1, vuint32m8_t rs2,
                                size_t vl);
vuint16mf4_t __riscv_vloxei64_mu(vbool64_t vm, vuint16mf4_t vd,
                                 const uint16_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vuint16mf2_t __riscv_vloxei64_mu(vbool32_t vm, vuint16mf2_t vd,
                                 const uint16_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vuint16m1_t __riscv_vloxei64_mu(vbool16_t vm, vuint16m1_t vd,
                                const uint16_t *rs1, vuint64m4_t rs2,
                                size_t vl);
vuint16m2_t __riscv_vloxei64_mu(vbool8_t vm, vuint16m2_t vd,
                                const uint16_t *rs1, vuint64m8_t rs2,
                                size_t vl);
vuint32mf2_t __riscv_vloxei8_mu(vbool64_t vm, vuint32mf2_t vd,
                                const uint32_t *rs1, vuint8mf8_t rs2,
                                size_t vl);
vuint32m1_t __riscv_vloxei8_mu(vbool32_t vm, vuint32m1_t vd,
                               const uint32_t *rs1, vuint8mf4_t rs2, size_t vl);
vuint32m2_t __riscv_vloxei8_mu(vbool16_t vm, vuint32m2_t vd,
                               const uint32_t *rs1, vuint8mf2_t rs2, size_t vl);
vuint32m4_t __riscv_vloxei8_mu(vbool8_t vm, vuint32m4_t vd, const uint32_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint32m8_t __riscv_vloxei8_mu(vbool4_t vm, vuint32m8_t vd, const uint32_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vuint32mf2_t __riscv_vloxei16_mu(vbool64_t vm, vuint32mf2_t vd,
                                 const uint32_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vuint32m1_t __riscv_vloxei16_mu(vbool32_t vm, vuint32m1_t vd,
                                const uint32_t *rs1, vuint16mf2_t rs2,
                                size_t vl);
vuint32m2_t __riscv_vloxei16_mu(vbool16_t vm, vuint32m2_t vd,
                                const uint32_t *rs1, vuint16m1_t rs2,
                                size_t vl);
vuint32m4_t __riscv_vloxei16_mu(vbool8_t vm, vuint32m4_t vd,
                                const uint32_t *rs1, vuint16m2_t rs2,
                                size_t vl);
vuint32m8_t __riscv_vloxei16_mu(vbool4_t vm, vuint32m8_t vd,
                                const uint32_t *rs1, vuint16m4_t rs2,
                                size_t vl);
vuint32mf2_t __riscv_vloxei32_mu(vbool64_t vm, vuint32mf2_t vd,
                                 const uint32_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vuint32m1_t __riscv_vloxei32_mu(vbool32_t vm, vuint32m1_t vd,
                                const uint32_t *rs1, vuint32m1_t rs2,
                                size_t vl);
vuint32m2_t __riscv_vloxei32_mu(vbool16_t vm, vuint32m2_t vd,
                                const uint32_t *rs1, vuint32m2_t rs2,
                                size_t vl);
vuint32m4_t __riscv_vloxei32_mu(vbool8_t vm, vuint32m4_t vd,
                                const uint32_t *rs1, vuint32m4_t rs2,
                                size_t vl);
vuint32m8_t __riscv_vloxei32_mu(vbool4_t vm, vuint32m8_t vd,
                                const uint32_t *rs1, vuint32m8_t rs2,
                                size_t vl);
vuint32mf2_t __riscv_vloxei64_mu(vbool64_t vm, vuint32mf2_t vd,
                                 const uint32_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vuint32m1_t __riscv_vloxei64_mu(vbool32_t vm, vuint32m1_t vd,
                                const uint32_t *rs1, vuint64m2_t rs2,
                                size_t vl);
vuint32m2_t __riscv_vloxei64_mu(vbool16_t vm, vuint32m2_t vd,
                                const uint32_t *rs1, vuint64m4_t rs2,
                                size_t vl);
vuint32m4_t __riscv_vloxei64_mu(vbool8_t vm, vuint32m4_t vd,
                                const uint32_t *rs1, vuint64m8_t rs2,
                                size_t vl);
vuint64m1_t __riscv_vloxei8_mu(vbool64_t vm, vuint64m1_t vd,
                               const uint64_t *rs1, vuint8mf8_t rs2, size_t vl);
vuint64m2_t __riscv_vloxei8_mu(vbool32_t vm, vuint64m2_t vd,
                               const uint64_t *rs1, vuint8mf4_t rs2, size_t vl);
vuint64m4_t __riscv_vloxei8_mu(vbool16_t vm, vuint64m4_t vd,
                               const uint64_t *rs1, vuint8mf2_t rs2, size_t vl);
vuint64m8_t __riscv_vloxei8_mu(vbool8_t vm, vuint64m8_t vd, const uint64_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint64m1_t __riscv_vloxei16_mu(vbool64_t vm, vuint64m1_t vd,
                                const uint64_t *rs1, vuint16mf4_t rs2,
                                size_t vl);
vuint64m2_t __riscv_vloxei16_mu(vbool32_t vm, vuint64m2_t vd,
                                const uint64_t *rs1, vuint16mf2_t rs2,
                                size_t vl);
vuint64m4_t __riscv_vloxei16_mu(vbool16_t vm, vuint64m4_t vd,
                                const uint64_t *rs1, vuint16m1_t rs2,
                                size_t vl);
vuint64m8_t __riscv_vloxei16_mu(vbool8_t vm, vuint64m8_t vd,
                                const uint64_t *rs1, vuint16m2_t rs2,
                                size_t vl);
vuint64m1_t __riscv_vloxei32_mu(vbool64_t vm, vuint64m1_t vd,
                                const uint64_t *rs1, vuint32mf2_t rs2,
                                size_t vl);
vuint64m2_t __riscv_vloxei32_mu(vbool32_t vm, vuint64m2_t vd,
                                const uint64_t *rs1, vuint32m1_t rs2,
                                size_t vl);
vuint64m4_t __riscv_vloxei32_mu(vbool16_t vm, vuint64m4_t vd,
                                const uint64_t *rs1, vuint32m2_t rs2,
                                size_t vl);
vuint64m8_t __riscv_vloxei32_mu(vbool8_t vm, vuint64m8_t vd,
                                const uint64_t *rs1, vuint32m4_t rs2,
                                size_t vl);
vuint64m1_t __riscv_vloxei64_mu(vbool64_t vm, vuint64m1_t vd,
                                const uint64_t *rs1, vuint64m1_t rs2,
                                size_t vl);
vuint64m2_t __riscv_vloxei64_mu(vbool32_t vm, vuint64m2_t vd,
                                const uint64_t *rs1, vuint64m2_t rs2,
                                size_t vl);
vuint64m4_t __riscv_vloxei64_mu(vbool16_t vm, vuint64m4_t vd,
                                const uint64_t *rs1, vuint64m4_t rs2,
                                size_t vl);
vuint64m8_t __riscv_vloxei64_mu(vbool8_t vm, vuint64m8_t vd,
                                const uint64_t *rs1, vuint64m8_t rs2,
                                size_t vl);
vuint8mf8_t __riscv_vluxei8_mu(vbool64_t vm, vuint8mf8_t vd, const uint8_t *rs1,
                               vuint8mf8_t rs2, size_t vl);
vuint8mf4_t __riscv_vluxei8_mu(vbool32_t vm, vuint8mf4_t vd, const uint8_t *rs1,
                               vuint8mf4_t rs2, size_t vl);
vuint8mf2_t __riscv_vluxei8_mu(vbool16_t vm, vuint8mf2_t vd, const uint8_t *rs1,
                               vuint8mf2_t rs2, size_t vl);
vuint8m1_t __riscv_vluxei8_mu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                              vuint8m1_t rs2, size_t vl);
vuint8m2_t __riscv_vluxei8_mu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                              vuint8m2_t rs2, size_t vl);
vuint8m4_t __riscv_vluxei8_mu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                              vuint8m4_t rs2, size_t vl);
vuint8m8_t __riscv_vluxei8_mu(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                              vuint8m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei16_mu(vbool64_t vm, vuint8mf8_t vd,
                                const uint8_t *rs1, vuint16mf4_t rs2,
                                size_t vl);
vuint8mf4_t __riscv_vluxei16_mu(vbool32_t vm, vuint8mf4_t vd,
                                const uint8_t *rs1, vuint16mf2_t rs2,
                                size_t vl);
vuint8mf2_t __riscv_vluxei16_mu(vbool16_t vm, vuint8mf2_t vd,
                                const uint8_t *rs1, vuint16m1_t rs2, size_t vl);
vuint8m1_t __riscv_vluxei16_mu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                               vuint16m2_t rs2, size_t vl);
vuint8m2_t __riscv_vluxei16_mu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                               vuint16m4_t rs2, size_t vl);
vuint8m4_t __riscv_vluxei16_mu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                               vuint16m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei32_mu(vbool64_t vm, vuint8mf8_t vd,
                                const uint8_t *rs1, vuint32mf2_t rs2,
                                size_t vl);
vuint8mf4_t __riscv_vluxei32_mu(vbool32_t vm, vuint8mf4_t vd,
                                const uint8_t *rs1, vuint32m1_t rs2, size_t vl);
vuint8mf2_t __riscv_vluxei32_mu(vbool16_t vm, vuint8mf2_t vd,
                                const uint8_t *rs1, vuint32m2_t rs2, size_t vl);
vuint8m1_t __riscv_vluxei32_mu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                               vuint32m4_t rs2, size_t vl);
vuint8m2_t __riscv_vluxei32_mu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                               vuint32m8_t rs2, size_t vl);
vuint8mf8_t __riscv_vluxei64_mu(vbool64_t vm, vuint8mf8_t vd,
                                const uint8_t *rs1, vuint64m1_t rs2, size_t vl);
vuint8mf4_t __riscv_vluxei64_mu(vbool32_t vm, vuint8mf4_t vd,
                                const uint8_t *rs1, vuint64m2_t rs2, size_t vl);
vuint8mf2_t __riscv_vluxei64_mu(vbool16_t vm, vuint8mf2_t vd,
                                const uint8_t *rs1, vuint64m4_t rs2, size_t vl);
vuint8m1_t __riscv_vluxei64_mu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                               vuint64m8_t rs2, size_t vl);
vuint16mf4_t __riscv_vluxei8_mu(vbool64_t vm, vuint16mf4_t vd,
                                const uint16_t *rs1, vuint8mf8_t rs2,
                                size_t vl);
vuint16mf2_t __riscv_vluxei8_mu(vbool32_t vm, vuint16mf2_t vd,
                                const uint16_t *rs1, vuint8mf4_t rs2,
                                size_t vl);
vuint16m1_t __riscv_vluxei8_mu(vbool16_t vm, vuint16m1_t vd,
                               const uint16_t *rs1, vuint8mf2_t rs2, size_t vl);
vuint16m2_t __riscv_vluxei8_mu(vbool8_t vm, vuint16m2_t vd, const uint16_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint16m4_t __riscv_vluxei8_mu(vbool4_t vm, vuint16m4_t vd, const uint16_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vuint16m8_t __riscv_vluxei8_mu(vbool2_t vm, vuint16m8_t vd, const uint16_t *rs1,
                               vuint8m4_t rs2, size_t vl);
vuint16mf4_t __riscv_vluxei16_mu(vbool64_t vm, vuint16mf4_t vd,
                                 const uint16_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vuint16mf2_t __riscv_vluxei16_mu(vbool32_t vm, vuint16mf2_t vd,
                                 const uint16_t *rs1, vuint16mf2_t rs2,
                                 size_t vl);
vuint16m1_t __riscv_vluxei16_mu(vbool16_t vm, vuint16m1_t vd,
                                const uint16_t *rs1, vuint16m1_t rs2,
                                size_t vl);
vuint16m2_t __riscv_vluxei16_mu(vbool8_t vm, vuint16m2_t vd,
                                const uint16_t *rs1, vuint16m2_t rs2,
                                size_t vl);
vuint16m4_t __riscv_vluxei16_mu(vbool4_t vm, vuint16m4_t vd,
                                const uint16_t *rs1, vuint16m4_t rs2,
                                size_t vl);
vuint16m8_t __riscv_vluxei16_mu(vbool2_t vm, vuint16m8_t vd,
                                const uint16_t *rs1, vuint16m8_t rs2,
                                size_t vl);
vuint16mf4_t __riscv_vluxei32_mu(vbool64_t vm, vuint16mf4_t vd,
                                 const uint16_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vuint16mf2_t __riscv_vluxei32_mu(vbool32_t vm, vuint16mf2_t vd,
                                 const uint16_t *rs1, vuint32m1_t rs2,
                                 size_t vl);
vuint16m1_t __riscv_vluxei32_mu(vbool16_t vm, vuint16m1_t vd,
                                const uint16_t *rs1, vuint32m2_t rs2,
                                size_t vl);
vuint16m2_t __riscv_vluxei32_mu(vbool8_t vm, vuint16m2_t vd,
                                const uint16_t *rs1, vuint32m4_t rs2,
                                size_t vl);
vuint16m4_t __riscv_vluxei32_mu(vbool4_t vm, vuint16m4_t vd,
                                const uint16_t *rs1, vuint32m8_t rs2,
                                size_t vl);
vuint16mf4_t __riscv_vluxei64_mu(vbool64_t vm, vuint16mf4_t vd,
                                 const uint16_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vuint16mf2_t __riscv_vluxei64_mu(vbool32_t vm, vuint16mf2_t vd,
                                 const uint16_t *rs1, vuint64m2_t rs2,
                                 size_t vl);
vuint16m1_t __riscv_vluxei64_mu(vbool16_t vm, vuint16m1_t vd,
                                const uint16_t *rs1, vuint64m4_t rs2,
                                size_t vl);
vuint16m2_t __riscv_vluxei64_mu(vbool8_t vm, vuint16m2_t vd,
                                const uint16_t *rs1, vuint64m8_t rs2,
                                size_t vl);
vuint32mf2_t __riscv_vluxei8_mu(vbool64_t vm, vuint32mf2_t vd,
                                const uint32_t *rs1, vuint8mf8_t rs2,
                                size_t vl);
vuint32m1_t __riscv_vluxei8_mu(vbool32_t vm, vuint32m1_t vd,
                               const uint32_t *rs1, vuint8mf4_t rs2, size_t vl);
vuint32m2_t __riscv_vluxei8_mu(vbool16_t vm, vuint32m2_t vd,
                               const uint32_t *rs1, vuint8mf2_t rs2, size_t vl);
vuint32m4_t __riscv_vluxei8_mu(vbool8_t vm, vuint32m4_t vd, const uint32_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint32m8_t __riscv_vluxei8_mu(vbool4_t vm, vuint32m8_t vd, const uint32_t *rs1,
                               vuint8m2_t rs2, size_t vl);
vuint32mf2_t __riscv_vluxei16_mu(vbool64_t vm, vuint32mf2_t vd,
                                 const uint32_t *rs1, vuint16mf4_t rs2,
                                 size_t vl);
vuint32m1_t __riscv_vluxei16_mu(vbool32_t vm, vuint32m1_t vd,
                                const uint32_t *rs1, vuint16mf2_t rs2,
                                size_t vl);
vuint32m2_t __riscv_vluxei16_mu(vbool16_t vm, vuint32m2_t vd,
                                const uint32_t *rs1, vuint16m1_t rs2,
                                size_t vl);
vuint32m4_t __riscv_vluxei16_mu(vbool8_t vm, vuint32m4_t vd,
                                const uint32_t *rs1, vuint16m2_t rs2,
                                size_t vl);
vuint32m8_t __riscv_vluxei16_mu(vbool4_t vm, vuint32m8_t vd,
                                const uint32_t *rs1, vuint16m4_t rs2,
                                size_t vl);
vuint32mf2_t __riscv_vluxei32_mu(vbool64_t vm, vuint32mf2_t vd,
                                 const uint32_t *rs1, vuint32mf2_t rs2,
                                 size_t vl);
vuint32m1_t __riscv_vluxei32_mu(vbool32_t vm, vuint32m1_t vd,
                                const uint32_t *rs1, vuint32m1_t rs2,
                                size_t vl);
vuint32m2_t __riscv_vluxei32_mu(vbool16_t vm, vuint32m2_t vd,
                                const uint32_t *rs1, vuint32m2_t rs2,
                                size_t vl);
vuint32m4_t __riscv_vluxei32_mu(vbool8_t vm, vuint32m4_t vd,
                                const uint32_t *rs1, vuint32m4_t rs2,
                                size_t vl);
vuint32m8_t __riscv_vluxei32_mu(vbool4_t vm, vuint32m8_t vd,
                                const uint32_t *rs1, vuint32m8_t rs2,
                                size_t vl);
vuint32mf2_t __riscv_vluxei64_mu(vbool64_t vm, vuint32mf2_t vd,
                                 const uint32_t *rs1, vuint64m1_t rs2,
                                 size_t vl);
vuint32m1_t __riscv_vluxei64_mu(vbool32_t vm, vuint32m1_t vd,
                                const uint32_t *rs1, vuint64m2_t rs2,
                                size_t vl);
vuint32m2_t __riscv_vluxei64_mu(vbool16_t vm, vuint32m2_t vd,
                                const uint32_t *rs1, vuint64m4_t rs2,
                                size_t vl);
vuint32m4_t __riscv_vluxei64_mu(vbool8_t vm, vuint32m4_t vd,
                                const uint32_t *rs1, vuint64m8_t rs2,
                                size_t vl);
vuint64m1_t __riscv_vluxei8_mu(vbool64_t vm, vuint64m1_t vd,
                               const uint64_t *rs1, vuint8mf8_t rs2, size_t vl);
vuint64m2_t __riscv_vluxei8_mu(vbool32_t vm, vuint64m2_t vd,
                               const uint64_t *rs1, vuint8mf4_t rs2, size_t vl);
vuint64m4_t __riscv_vluxei8_mu(vbool16_t vm, vuint64m4_t vd,
                               const uint64_t *rs1, vuint8mf2_t rs2, size_t vl);
vuint64m8_t __riscv_vluxei8_mu(vbool8_t vm, vuint64m8_t vd, const uint64_t *rs1,
                               vuint8m1_t rs2, size_t vl);
vuint64m1_t __riscv_vluxei16_mu(vbool64_t vm, vuint64m1_t vd,
                                const uint64_t *rs1, vuint16mf4_t rs2,
                                size_t vl);
vuint64m2_t __riscv_vluxei16_mu(vbool32_t vm, vuint64m2_t vd,
                                const uint64_t *rs1, vuint16mf2_t rs2,
                                size_t vl);
vuint64m4_t __riscv_vluxei16_mu(vbool16_t vm, vuint64m4_t vd,
                                const uint64_t *rs1, vuint16m1_t rs2,
                                size_t vl);
vuint64m8_t __riscv_vluxei16_mu(vbool8_t vm, vuint64m8_t vd,
                                const uint64_t *rs1, vuint16m2_t rs2,
                                size_t vl);
vuint64m1_t __riscv_vluxei32_mu(vbool64_t vm, vuint64m1_t vd,
                                const uint64_t *rs1, vuint32mf2_t rs2,
                                size_t vl);
vuint64m2_t __riscv_vluxei32_mu(vbool32_t vm, vuint64m2_t vd,
                                const uint64_t *rs1, vuint32m1_t rs2,
                                size_t vl);
vuint64m4_t __riscv_vluxei32_mu(vbool16_t vm, vuint64m4_t vd,
                                const uint64_t *rs1, vuint32m2_t rs2,
                                size_t vl);
vuint64m8_t __riscv_vluxei32_mu(vbool8_t vm, vuint64m8_t vd,
                                const uint64_t *rs1, vuint32m4_t rs2,
                                size_t vl);
vuint64m1_t __riscv_vluxei64_mu(vbool64_t vm, vuint64m1_t vd,
                                const uint64_t *rs1, vuint64m1_t rs2,
                                size_t vl);
vuint64m2_t __riscv_vluxei64_mu(vbool32_t vm, vuint64m2_t vd,
                                const uint64_t *rs1, vuint64m2_t rs2,
                                size_t vl);
vuint64m4_t __riscv_vluxei64_mu(vbool16_t vm, vuint64m4_t vd,
                                const uint64_t *rs1, vuint64m4_t rs2,
                                size_t vl);
vuint64m8_t __riscv_vluxei64_mu(vbool8_t vm, vuint64m8_t vd,
                                const uint64_t *rs1, vuint64m8_t rs2,
                                size_t vl);

Vector Indexed Store Intrinsics

Intrinsics here don’t have a policy variant.

Unit-stride Fault-Only-First Loads Intrinsics

vfloat16mf4_t __riscv_vle16ff_tu(vfloat16mf4_t vd, const _Float16 *rs1,
                                 size_t *new_vl, size_t vl);
vfloat16mf2_t __riscv_vle16ff_tu(vfloat16mf2_t vd, const _Float16 *rs1,
                                 size_t *new_vl, size_t vl);
vfloat16m1_t __riscv_vle16ff_tu(vfloat16m1_t vd, const _Float16 *rs1,
                                size_t *new_vl, size_t vl);
vfloat16m2_t __riscv_vle16ff_tu(vfloat16m2_t vd, const _Float16 *rs1,
                                size_t *new_vl, size_t vl);
vfloat16m4_t __riscv_vle16ff_tu(vfloat16m4_t vd, const _Float16 *rs1,
                                size_t *new_vl, size_t vl);
vfloat16m8_t __riscv_vle16ff_tu(vfloat16m8_t vd, const _Float16 *rs1,
                                size_t *new_vl, size_t vl);
vfloat32mf2_t __riscv_vle32ff_tu(vfloat32mf2_t vd, const float *rs1,
                                 size_t *new_vl, size_t vl);
vfloat32m1_t __riscv_vle32ff_tu(vfloat32m1_t vd, const float *rs1,
                                size_t *new_vl, size_t vl);
vfloat32m2_t __riscv_vle32ff_tu(vfloat32m2_t vd, const float *rs1,
                                size_t *new_vl, size_t vl);
vfloat32m4_t __riscv_vle32ff_tu(vfloat32m4_t vd, const float *rs1,
                                size_t *new_vl, size_t vl);
vfloat32m8_t __riscv_vle32ff_tu(vfloat32m8_t vd, const float *rs1,
                                size_t *new_vl, size_t vl);
vfloat64m1_t __riscv_vle64ff_tu(vfloat64m1_t vd, const double *rs1,
                                size_t *new_vl, size_t vl);
vfloat64m2_t __riscv_vle64ff_tu(vfloat64m2_t vd, const double *rs1,
                                size_t *new_vl, size_t vl);
vfloat64m4_t __riscv_vle64ff_tu(vfloat64m4_t vd, const double *rs1,
                                size_t *new_vl, size_t vl);
vfloat64m8_t __riscv_vle64ff_tu(vfloat64m8_t vd, const double *rs1,
                                size_t *new_vl, size_t vl);
vint8mf8_t __riscv_vle8ff_tu(vint8mf8_t vd, const int8_t *rs1, size_t *new_vl,
                             size_t vl);
vint8mf4_t __riscv_vle8ff_tu(vint8mf4_t vd, const int8_t *rs1, size_t *new_vl,
                             size_t vl);
vint8mf2_t __riscv_vle8ff_tu(vint8mf2_t vd, const int8_t *rs1, size_t *new_vl,
                             size_t vl);
vint8m1_t __riscv_vle8ff_tu(vint8m1_t vd, const int8_t *rs1, size_t *new_vl,
                            size_t vl);
vint8m2_t __riscv_vle8ff_tu(vint8m2_t vd, const int8_t *rs1, size_t *new_vl,
                            size_t vl);
vint8m4_t __riscv_vle8ff_tu(vint8m4_t vd, const int8_t *rs1, size_t *new_vl,
                            size_t vl);
vint8m8_t __riscv_vle8ff_tu(vint8m8_t vd, const int8_t *rs1, size_t *new_vl,
                            size_t vl);
vint16mf4_t __riscv_vle16ff_tu(vint16mf4_t vd, const int16_t *rs1,
                               size_t *new_vl, size_t vl);
vint16mf2_t __riscv_vle16ff_tu(vint16mf2_t vd, const int16_t *rs1,
                               size_t *new_vl, size_t vl);
vint16m1_t __riscv_vle16ff_tu(vint16m1_t vd, const int16_t *rs1, size_t *new_vl,
                              size_t vl);
vint16m2_t __riscv_vle16ff_tu(vint16m2_t vd, const int16_t *rs1, size_t *new_vl,
                              size_t vl);
vint16m4_t __riscv_vle16ff_tu(vint16m4_t vd, const int16_t *rs1, size_t *new_vl,
                              size_t vl);
vint16m8_t __riscv_vle16ff_tu(vint16m8_t vd, const int16_t *rs1, size_t *new_vl,
                              size_t vl);
vint32mf2_t __riscv_vle32ff_tu(vint32mf2_t vd, const int32_t *rs1,
                               size_t *new_vl, size_t vl);
vint32m1_t __riscv_vle32ff_tu(vint32m1_t vd, const int32_t *rs1, size_t *new_vl,
                              size_t vl);
vint32m2_t __riscv_vle32ff_tu(vint32m2_t vd, const int32_t *rs1, size_t *new_vl,
                              size_t vl);
vint32m4_t __riscv_vle32ff_tu(vint32m4_t vd, const int32_t *rs1, size_t *new_vl,
                              size_t vl);
vint32m8_t __riscv_vle32ff_tu(vint32m8_t vd, const int32_t *rs1, size_t *new_vl,
                              size_t vl);
vint64m1_t __riscv_vle64ff_tu(vint64m1_t vd, const int64_t *rs1, size_t *new_vl,
                              size_t vl);
vint64m2_t __riscv_vle64ff_tu(vint64m2_t vd, const int64_t *rs1, size_t *new_vl,
                              size_t vl);
vint64m4_t __riscv_vle64ff_tu(vint64m4_t vd, const int64_t *rs1, size_t *new_vl,
                              size_t vl);
vint64m8_t __riscv_vle64ff_tu(vint64m8_t vd, const int64_t *rs1, size_t *new_vl,
                              size_t vl);
vuint8mf8_t __riscv_vle8ff_tu(vuint8mf8_t vd, const uint8_t *rs1,
                              size_t *new_vl, size_t vl);
vuint8mf4_t __riscv_vle8ff_tu(vuint8mf4_t vd, const uint8_t *rs1,
                              size_t *new_vl, size_t vl);
vuint8mf2_t __riscv_vle8ff_tu(vuint8mf2_t vd, const uint8_t *rs1,
                              size_t *new_vl, size_t vl);
vuint8m1_t __riscv_vle8ff_tu(vuint8m1_t vd, const uint8_t *rs1, size_t *new_vl,
                             size_t vl);
vuint8m2_t __riscv_vle8ff_tu(vuint8m2_t vd, const uint8_t *rs1, size_t *new_vl,
                             size_t vl);
vuint8m4_t __riscv_vle8ff_tu(vuint8m4_t vd, const uint8_t *rs1, size_t *new_vl,
                             size_t vl);
vuint8m8_t __riscv_vle8ff_tu(vuint8m8_t vd, const uint8_t *rs1, size_t *new_vl,
                             size_t vl);
vuint16mf4_t __riscv_vle16ff_tu(vuint16mf4_t vd, const uint16_t *rs1,
                                size_t *new_vl, size_t vl);
vuint16mf2_t __riscv_vle16ff_tu(vuint16mf2_t vd, const uint16_t *rs1,
                                size_t *new_vl, size_t vl);
vuint16m1_t __riscv_vle16ff_tu(vuint16m1_t vd, const uint16_t *rs1,
                               size_t *new_vl, size_t vl);
vuint16m2_t __riscv_vle16ff_tu(vuint16m2_t vd, const uint16_t *rs1,
                               size_t *new_vl, size_t vl);
vuint16m4_t __riscv_vle16ff_tu(vuint16m4_t vd, const uint16_t *rs1,
                               size_t *new_vl, size_t vl);
vuint16m8_t __riscv_vle16ff_tu(vuint16m8_t vd, const uint16_t *rs1,
                               size_t *new_vl, size_t vl);
vuint32mf2_t __riscv_vle32ff_tu(vuint32mf2_t vd, const uint32_t *rs1,
                                size_t *new_vl, size_t vl);
vuint32m1_t __riscv_vle32ff_tu(vuint32m1_t vd, const uint32_t *rs1,
                               size_t *new_vl, size_t vl);
vuint32m2_t __riscv_vle32ff_tu(vuint32m2_t vd, const uint32_t *rs1,
                               size_t *new_vl, size_t vl);
vuint32m4_t __riscv_vle32ff_tu(vuint32m4_t vd, const uint32_t *rs1,
                               size_t *new_vl, size_t vl);
vuint32m8_t __riscv_vle32ff_tu(vuint32m8_t vd, const uint32_t *rs1,
                               size_t *new_vl, size_t vl);
vuint64m1_t __riscv_vle64ff_tu(vuint64m1_t vd, const uint64_t *rs1,
                               size_t *new_vl, size_t vl);
vuint64m2_t __riscv_vle64ff_tu(vuint64m2_t vd, const uint64_t *rs1,
                               size_t *new_vl, size_t vl);
vuint64m4_t __riscv_vle64ff_tu(vuint64m4_t vd, const uint64_t *rs1,
                               size_t *new_vl, size_t vl);
vuint64m8_t __riscv_vle64ff_tu(vuint64m8_t vd, const uint64_t *rs1,
                               size_t *new_vl, size_t vl);
// masked functions
vfloat16mf4_t __riscv_vle16ff_tum(vbool64_t vm, vfloat16mf4_t vd,
                                  const _Float16 *rs1, size_t *new_vl,
                                  size_t vl);
vfloat16mf2_t __riscv_vle16ff_tum(vbool32_t vm, vfloat16mf2_t vd,
                                  const _Float16 *rs1, size_t *new_vl,
                                  size_t vl);
vfloat16m1_t __riscv_vle16ff_tum(vbool16_t vm, vfloat16m1_t vd,
                                 const _Float16 *rs1, size_t *new_vl,
                                 size_t vl);
vfloat16m2_t __riscv_vle16ff_tum(vbool8_t vm, vfloat16m2_t vd,
                                 const _Float16 *rs1, size_t *new_vl,
                                 size_t vl);
vfloat16m4_t __riscv_vle16ff_tum(vbool4_t vm, vfloat16m4_t vd,
                                 const _Float16 *rs1, size_t *new_vl,
                                 size_t vl);
vfloat16m8_t __riscv_vle16ff_tum(vbool2_t vm, vfloat16m8_t vd,
                                 const _Float16 *rs1, size_t *new_vl,
                                 size_t vl);
vfloat32mf2_t __riscv_vle32ff_tum(vbool64_t vm, vfloat32mf2_t vd,
                                  const float *rs1, size_t *new_vl, size_t vl);
vfloat32m1_t __riscv_vle32ff_tum(vbool32_t vm, vfloat32m1_t vd,
                                 const float *rs1, size_t *new_vl, size_t vl);
vfloat32m2_t __riscv_vle32ff_tum(vbool16_t vm, vfloat32m2_t vd,
                                 const float *rs1, size_t *new_vl, size_t vl);
vfloat32m4_t __riscv_vle32ff_tum(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                 size_t *new_vl, size_t vl);
vfloat32m8_t __riscv_vle32ff_tum(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                 size_t *new_vl, size_t vl);
vfloat64m1_t __riscv_vle64ff_tum(vbool64_t vm, vfloat64m1_t vd,
                                 const double *rs1, size_t *new_vl, size_t vl);
vfloat64m2_t __riscv_vle64ff_tum(vbool32_t vm, vfloat64m2_t vd,
                                 const double *rs1, size_t *new_vl, size_t vl);
vfloat64m4_t __riscv_vle64ff_tum(vbool16_t vm, vfloat64m4_t vd,
                                 const double *rs1, size_t *new_vl, size_t vl);
vfloat64m8_t __riscv_vle64ff_tum(vbool8_t vm, vfloat64m8_t vd,
                                 const double *rs1, size_t *new_vl, size_t vl);
vint8mf8_t __riscv_vle8ff_tum(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                              size_t *new_vl, size_t vl);
vint8mf4_t __riscv_vle8ff_tum(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                              size_t *new_vl, size_t vl);
vint8mf2_t __riscv_vle8ff_tum(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                              size_t *new_vl, size_t vl);
vint8m1_t __riscv_vle8ff_tum(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                             size_t *new_vl, size_t vl);
vint8m2_t __riscv_vle8ff_tum(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                             size_t *new_vl, size_t vl);
vint8m4_t __riscv_vle8ff_tum(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                             size_t *new_vl, size_t vl);
vint8m8_t __riscv_vle8ff_tum(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                             size_t *new_vl, size_t vl);
vint16mf4_t __riscv_vle16ff_tum(vbool64_t vm, vint16mf4_t vd,
                                const int16_t *rs1, size_t *new_vl, size_t vl);
vint16mf2_t __riscv_vle16ff_tum(vbool32_t vm, vint16mf2_t vd,
                                const int16_t *rs1, size_t *new_vl, size_t vl);
vint16m1_t __riscv_vle16ff_tum(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                               size_t *new_vl, size_t vl);
vint16m2_t __riscv_vle16ff_tum(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                               size_t *new_vl, size_t vl);
vint16m4_t __riscv_vle16ff_tum(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                               size_t *new_vl, size_t vl);
vint16m8_t __riscv_vle16ff_tum(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                               size_t *new_vl, size_t vl);
vint32mf2_t __riscv_vle32ff_tum(vbool64_t vm, vint32mf2_t vd,
                                const int32_t *rs1, size_t *new_vl, size_t vl);
vint32m1_t __riscv_vle32ff_tum(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                               size_t *new_vl, size_t vl);
vint32m2_t __riscv_vle32ff_tum(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                               size_t *new_vl, size_t vl);
vint32m4_t __riscv_vle32ff_tum(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                               size_t *new_vl, size_t vl);
vint32m8_t __riscv_vle32ff_tum(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                               size_t *new_vl, size_t vl);
vint64m1_t __riscv_vle64ff_tum(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                               size_t *new_vl, size_t vl);
vint64m2_t __riscv_vle64ff_tum(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                               size_t *new_vl, size_t vl);
vint64m4_t __riscv_vle64ff_tum(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                               size_t *new_vl, size_t vl);
vint64m8_t __riscv_vle64ff_tum(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                               size_t *new_vl, size_t vl);
vuint8mf8_t __riscv_vle8ff_tum(vbool64_t vm, vuint8mf8_t vd, const uint8_t *rs1,
                               size_t *new_vl, size_t vl);
vuint8mf4_t __riscv_vle8ff_tum(vbool32_t vm, vuint8mf4_t vd, const uint8_t *rs1,
                               size_t *new_vl, size_t vl);
vuint8mf2_t __riscv_vle8ff_tum(vbool16_t vm, vuint8mf2_t vd, const uint8_t *rs1,
                               size_t *new_vl, size_t vl);
vuint8m1_t __riscv_vle8ff_tum(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                              size_t *new_vl, size_t vl);
vuint8m2_t __riscv_vle8ff_tum(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                              size_t *new_vl, size_t vl);
vuint8m4_t __riscv_vle8ff_tum(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                              size_t *new_vl, size_t vl);
vuint8m8_t __riscv_vle8ff_tum(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                              size_t *new_vl, size_t vl);
vuint16mf4_t __riscv_vle16ff_tum(vbool64_t vm, vuint16mf4_t vd,
                                 const uint16_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint16mf2_t __riscv_vle16ff_tum(vbool32_t vm, vuint16mf2_t vd,
                                 const uint16_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint16m1_t __riscv_vle16ff_tum(vbool16_t vm, vuint16m1_t vd,
                                const uint16_t *rs1, size_t *new_vl, size_t vl);
vuint16m2_t __riscv_vle16ff_tum(vbool8_t vm, vuint16m2_t vd,
                                const uint16_t *rs1, size_t *new_vl, size_t vl);
vuint16m4_t __riscv_vle16ff_tum(vbool4_t vm, vuint16m4_t vd,
                                const uint16_t *rs1, size_t *new_vl, size_t vl);
vuint16m8_t __riscv_vle16ff_tum(vbool2_t vm, vuint16m8_t vd,
                                const uint16_t *rs1, size_t *new_vl, size_t vl);
vuint32mf2_t __riscv_vle32ff_tum(vbool64_t vm, vuint32mf2_t vd,
                                 const uint32_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint32m1_t __riscv_vle32ff_tum(vbool32_t vm, vuint32m1_t vd,
                                const uint32_t *rs1, size_t *new_vl, size_t vl);
vuint32m2_t __riscv_vle32ff_tum(vbool16_t vm, vuint32m2_t vd,
                                const uint32_t *rs1, size_t *new_vl, size_t vl);
vuint32m4_t __riscv_vle32ff_tum(vbool8_t vm, vuint32m4_t vd,
                                const uint32_t *rs1, size_t *new_vl, size_t vl);
vuint32m8_t __riscv_vle32ff_tum(vbool4_t vm, vuint32m8_t vd,
                                const uint32_t *rs1, size_t *new_vl, size_t vl);
vuint64m1_t __riscv_vle64ff_tum(vbool64_t vm, vuint64m1_t vd,
                                const uint64_t *rs1, size_t *new_vl, size_t vl);
vuint64m2_t __riscv_vle64ff_tum(vbool32_t vm, vuint64m2_t vd,
                                const uint64_t *rs1, size_t *new_vl, size_t vl);
vuint64m4_t __riscv_vle64ff_tum(vbool16_t vm, vuint64m4_t vd,
                                const uint64_t *rs1, size_t *new_vl, size_t vl);
vuint64m8_t __riscv_vle64ff_tum(vbool8_t vm, vuint64m8_t vd,
                                const uint64_t *rs1, size_t *new_vl, size_t vl);
// masked functions
vfloat16mf4_t __riscv_vle16ff_tumu(vbool64_t vm, vfloat16mf4_t vd,
                                   const _Float16 *rs1, size_t *new_vl,
                                   size_t vl);
vfloat16mf2_t __riscv_vle16ff_tumu(vbool32_t vm, vfloat16mf2_t vd,
                                   const _Float16 *rs1, size_t *new_vl,
                                   size_t vl);
vfloat16m1_t __riscv_vle16ff_tumu(vbool16_t vm, vfloat16m1_t vd,
                                  const _Float16 *rs1, size_t *new_vl,
                                  size_t vl);
vfloat16m2_t __riscv_vle16ff_tumu(vbool8_t vm, vfloat16m2_t vd,
                                  const _Float16 *rs1, size_t *new_vl,
                                  size_t vl);
vfloat16m4_t __riscv_vle16ff_tumu(vbool4_t vm, vfloat16m4_t vd,
                                  const _Float16 *rs1, size_t *new_vl,
                                  size_t vl);
vfloat16m8_t __riscv_vle16ff_tumu(vbool2_t vm, vfloat16m8_t vd,
                                  const _Float16 *rs1, size_t *new_vl,
                                  size_t vl);
vfloat32mf2_t __riscv_vle32ff_tumu(vbool64_t vm, vfloat32mf2_t vd,
                                   const float *rs1, size_t *new_vl, size_t vl);
vfloat32m1_t __riscv_vle32ff_tumu(vbool32_t vm, vfloat32m1_t vd,
                                  const float *rs1, size_t *new_vl, size_t vl);
vfloat32m2_t __riscv_vle32ff_tumu(vbool16_t vm, vfloat32m2_t vd,
                                  const float *rs1, size_t *new_vl, size_t vl);
vfloat32m4_t __riscv_vle32ff_tumu(vbool8_t vm, vfloat32m4_t vd,
                                  const float *rs1, size_t *new_vl, size_t vl);
vfloat32m8_t __riscv_vle32ff_tumu(vbool4_t vm, vfloat32m8_t vd,
                                  const float *rs1, size_t *new_vl, size_t vl);
vfloat64m1_t __riscv_vle64ff_tumu(vbool64_t vm, vfloat64m1_t vd,
                                  const double *rs1, size_t *new_vl, size_t vl);
vfloat64m2_t __riscv_vle64ff_tumu(vbool32_t vm, vfloat64m2_t vd,
                                  const double *rs1, size_t *new_vl, size_t vl);
vfloat64m4_t __riscv_vle64ff_tumu(vbool16_t vm, vfloat64m4_t vd,
                                  const double *rs1, size_t *new_vl, size_t vl);
vfloat64m8_t __riscv_vle64ff_tumu(vbool8_t vm, vfloat64m8_t vd,
                                  const double *rs1, size_t *new_vl, size_t vl);
vint8mf8_t __riscv_vle8ff_tumu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                               size_t *new_vl, size_t vl);
vint8mf4_t __riscv_vle8ff_tumu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                               size_t *new_vl, size_t vl);
vint8mf2_t __riscv_vle8ff_tumu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                               size_t *new_vl, size_t vl);
vint8m1_t __riscv_vle8ff_tumu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                              size_t *new_vl, size_t vl);
vint8m2_t __riscv_vle8ff_tumu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                              size_t *new_vl, size_t vl);
vint8m4_t __riscv_vle8ff_tumu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                              size_t *new_vl, size_t vl);
vint8m8_t __riscv_vle8ff_tumu(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                              size_t *new_vl, size_t vl);
vint16mf4_t __riscv_vle16ff_tumu(vbool64_t vm, vint16mf4_t vd,
                                 const int16_t *rs1, size_t *new_vl, size_t vl);
vint16mf2_t __riscv_vle16ff_tumu(vbool32_t vm, vint16mf2_t vd,
                                 const int16_t *rs1, size_t *new_vl, size_t vl);
vint16m1_t __riscv_vle16ff_tumu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                                size_t *new_vl, size_t vl);
vint16m2_t __riscv_vle16ff_tumu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                                size_t *new_vl, size_t vl);
vint16m4_t __riscv_vle16ff_tumu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                                size_t *new_vl, size_t vl);
vint16m8_t __riscv_vle16ff_tumu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                                size_t *new_vl, size_t vl);
vint32mf2_t __riscv_vle32ff_tumu(vbool64_t vm, vint32mf2_t vd,
                                 const int32_t *rs1, size_t *new_vl, size_t vl);
vint32m1_t __riscv_vle32ff_tumu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                                size_t *new_vl, size_t vl);
vint32m2_t __riscv_vle32ff_tumu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                                size_t *new_vl, size_t vl);
vint32m4_t __riscv_vle32ff_tumu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                                size_t *new_vl, size_t vl);
vint32m8_t __riscv_vle32ff_tumu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                                size_t *new_vl, size_t vl);
vint64m1_t __riscv_vle64ff_tumu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                                size_t *new_vl, size_t vl);
vint64m2_t __riscv_vle64ff_tumu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                                size_t *new_vl, size_t vl);
vint64m4_t __riscv_vle64ff_tumu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                                size_t *new_vl, size_t vl);
vint64m8_t __riscv_vle64ff_tumu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                                size_t *new_vl, size_t vl);
vuint8mf8_t __riscv_vle8ff_tumu(vbool64_t vm, vuint8mf8_t vd,
                                const uint8_t *rs1, size_t *new_vl, size_t vl);
vuint8mf4_t __riscv_vle8ff_tumu(vbool32_t vm, vuint8mf4_t vd,
                                const uint8_t *rs1, size_t *new_vl, size_t vl);
vuint8mf2_t __riscv_vle8ff_tumu(vbool16_t vm, vuint8mf2_t vd,
                                const uint8_t *rs1, size_t *new_vl, size_t vl);
vuint8m1_t __riscv_vle8ff_tumu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                               size_t *new_vl, size_t vl);
vuint8m2_t __riscv_vle8ff_tumu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                               size_t *new_vl, size_t vl);
vuint8m4_t __riscv_vle8ff_tumu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                               size_t *new_vl, size_t vl);
vuint8m8_t __riscv_vle8ff_tumu(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                               size_t *new_vl, size_t vl);
vuint16mf4_t __riscv_vle16ff_tumu(vbool64_t vm, vuint16mf4_t vd,
                                  const uint16_t *rs1, size_t *new_vl,
                                  size_t vl);
vuint16mf2_t __riscv_vle16ff_tumu(vbool32_t vm, vuint16mf2_t vd,
                                  const uint16_t *rs1, size_t *new_vl,
                                  size_t vl);
vuint16m1_t __riscv_vle16ff_tumu(vbool16_t vm, vuint16m1_t vd,
                                 const uint16_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint16m2_t __riscv_vle16ff_tumu(vbool8_t vm, vuint16m2_t vd,
                                 const uint16_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint16m4_t __riscv_vle16ff_tumu(vbool4_t vm, vuint16m4_t vd,
                                 const uint16_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint16m8_t __riscv_vle16ff_tumu(vbool2_t vm, vuint16m8_t vd,
                                 const uint16_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint32mf2_t __riscv_vle32ff_tumu(vbool64_t vm, vuint32mf2_t vd,
                                  const uint32_t *rs1, size_t *new_vl,
                                  size_t vl);
vuint32m1_t __riscv_vle32ff_tumu(vbool32_t vm, vuint32m1_t vd,
                                 const uint32_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint32m2_t __riscv_vle32ff_tumu(vbool16_t vm, vuint32m2_t vd,
                                 const uint32_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint32m4_t __riscv_vle32ff_tumu(vbool8_t vm, vuint32m4_t vd,
                                 const uint32_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint32m8_t __riscv_vle32ff_tumu(vbool4_t vm, vuint32m8_t vd,
                                 const uint32_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint64m1_t __riscv_vle64ff_tumu(vbool64_t vm, vuint64m1_t vd,
                                 const uint64_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint64m2_t __riscv_vle64ff_tumu(vbool32_t vm, vuint64m2_t vd,
                                 const uint64_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint64m4_t __riscv_vle64ff_tumu(vbool16_t vm, vuint64m4_t vd,
                                 const uint64_t *rs1, size_t *new_vl,
                                 size_t vl);
vuint64m8_t __riscv_vle64ff_tumu(vbool8_t vm, vuint64m8_t vd,
                                 const uint64_t *rs1, size_t *new_vl,
                                 size_t vl);
// masked functions
vfloat16mf4_t __riscv_vle16ff_mu(vbool64_t vm, vfloat16mf4_t vd,
                                 const _Float16 *rs1, size_t *new_vl,
                                 size_t vl);
vfloat16mf2_t __riscv_vle16ff_mu(vbool32_t vm, vfloat16mf2_t vd,
                                 const _Float16 *rs1, size_t *new_vl,
                                 size_t vl);
vfloat16m1_t __riscv_vle16ff_mu(vbool16_t vm, vfloat16m1_t vd,
                                const _Float16 *rs1, size_t *new_vl, size_t vl);
vfloat16m2_t __riscv_vle16ff_mu(vbool8_t vm, vfloat16m2_t vd,
                                const _Float16 *rs1, size_t *new_vl, size_t vl);
vfloat16m4_t __riscv_vle16ff_mu(vbool4_t vm, vfloat16m4_t vd,
                                const _Float16 *rs1, size_t *new_vl, size_t vl);
vfloat16m8_t __riscv_vle16ff_mu(vbool2_t vm, vfloat16m8_t vd,
                                const _Float16 *rs1, size_t *new_vl, size_t vl);
vfloat32mf2_t __riscv_vle32ff_mu(vbool64_t vm, vfloat32mf2_t vd,
                                 const float *rs1, size_t *new_vl, size_t vl);
vfloat32m1_t __riscv_vle32ff_mu(vbool32_t vm, vfloat32m1_t vd, const float *rs1,
                                size_t *new_vl, size_t vl);
vfloat32m2_t __riscv_vle32ff_mu(vbool16_t vm, vfloat32m2_t vd, const float *rs1,
                                size_t *new_vl, size_t vl);
vfloat32m4_t __riscv_vle32ff_mu(vbool8_t vm, vfloat32m4_t vd, const float *rs1,
                                size_t *new_vl, size_t vl);
vfloat32m8_t __riscv_vle32ff_mu(vbool4_t vm, vfloat32m8_t vd, const float *rs1,
                                size_t *new_vl, size_t vl);
vfloat64m1_t __riscv_vle64ff_mu(vbool64_t vm, vfloat64m1_t vd,
                                const double *rs1, size_t *new_vl, size_t vl);
vfloat64m2_t __riscv_vle64ff_mu(vbool32_t vm, vfloat64m2_t vd,
                                const double *rs1, size_t *new_vl, size_t vl);
vfloat64m4_t __riscv_vle64ff_mu(vbool16_t vm, vfloat64m4_t vd,
                                const double *rs1, size_t *new_vl, size_t vl);
vfloat64m8_t __riscv_vle64ff_mu(vbool8_t vm, vfloat64m8_t vd, const double *rs1,
                                size_t *new_vl, size_t vl);
vint8mf8_t __riscv_vle8ff_mu(vbool64_t vm, vint8mf8_t vd, const int8_t *rs1,
                             size_t *new_vl, size_t vl);
vint8mf4_t __riscv_vle8ff_mu(vbool32_t vm, vint8mf4_t vd, const int8_t *rs1,
                             size_t *new_vl, size_t vl);
vint8mf2_t __riscv_vle8ff_mu(vbool16_t vm, vint8mf2_t vd, const int8_t *rs1,
                             size_t *new_vl, size_t vl);
vint8m1_t __riscv_vle8ff_mu(vbool8_t vm, vint8m1_t vd, const int8_t *rs1,
                            size_t *new_vl, size_t vl);
vint8m2_t __riscv_vle8ff_mu(vbool4_t vm, vint8m2_t vd, const int8_t *rs1,
                            size_t *new_vl, size_t vl);
vint8m4_t __riscv_vle8ff_mu(vbool2_t vm, vint8m4_t vd, const int8_t *rs1,
                            size_t *new_vl, size_t vl);
vint8m8_t __riscv_vle8ff_mu(vbool1_t vm, vint8m8_t vd, const int8_t *rs1,
                            size_t *new_vl, size_t vl);
vint16mf4_t __riscv_vle16ff_mu(vbool64_t vm, vint16mf4_t vd, const int16_t *rs1,
                               size_t *new_vl, size_t vl);
vint16mf2_t __riscv_vle16ff_mu(vbool32_t vm, vint16mf2_t vd, const int16_t *rs1,
                               size_t *new_vl, size_t vl);
vint16m1_t __riscv_vle16ff_mu(vbool16_t vm, vint16m1_t vd, const int16_t *rs1,
                              size_t *new_vl, size_t vl);
vint16m2_t __riscv_vle16ff_mu(vbool8_t vm, vint16m2_t vd, const int16_t *rs1,
                              size_t *new_vl, size_t vl);
vint16m4_t __riscv_vle16ff_mu(vbool4_t vm, vint16m4_t vd, const int16_t *rs1,
                              size_t *new_vl, size_t vl);
vint16m8_t __riscv_vle16ff_mu(vbool2_t vm, vint16m8_t vd, const int16_t *rs1,
                              size_t *new_vl, size_t vl);
vint32mf2_t __riscv_vle32ff_mu(vbool64_t vm, vint32mf2_t vd, const int32_t *rs1,
                               size_t *new_vl, size_t vl);
vint32m1_t __riscv_vle32ff_mu(vbool32_t vm, vint32m1_t vd, const int32_t *rs1,
                              size_t *new_vl, size_t vl);
vint32m2_t __riscv_vle32ff_mu(vbool16_t vm, vint32m2_t vd, const int32_t *rs1,
                              size_t *new_vl, size_t vl);
vint32m4_t __riscv_vle32ff_mu(vbool8_t vm, vint32m4_t vd, const int32_t *rs1,
                              size_t *new_vl, size_t vl);
vint32m8_t __riscv_vle32ff_mu(vbool4_t vm, vint32m8_t vd, const int32_t *rs1,
                              size_t *new_vl, size_t vl);
vint64m1_t __riscv_vle64ff_mu(vbool64_t vm, vint64m1_t vd, const int64_t *rs1,
                              size_t *new_vl, size_t vl);
vint64m2_t __riscv_vle64ff_mu(vbool32_t vm, vint64m2_t vd, const int64_t *rs1,
                              size_t *new_vl, size_t vl);
vint64m4_t __riscv_vle64ff_mu(vbool16_t vm, vint64m4_t vd, const int64_t *rs1,
                              size_t *new_vl, size_t vl);
vint64m8_t __riscv_vle64ff_mu(vbool8_t vm, vint64m8_t vd, const int64_t *rs1,
                              size_t *new_vl, size_t vl);
vuint8mf8_t __riscv_vle8ff_mu(vbool64_t vm, vuint8mf8_t vd, const uint8_t *rs1,
                              size_t *new_vl, size_t vl);
vuint8mf4_t __riscv_vle8ff_mu(vbool32_t vm, vuint8mf4_t vd, const uint8_t *rs1,
                              size_t *new_vl, size_t vl);
vuint8mf2_t __riscv_vle8ff_mu(vbool16_t vm, vuint8mf2_t vd, const uint8_t *rs1,
                              size_t *new_vl, size_t vl);
vuint8m1_t __riscv_vle8ff_mu(vbool8_t vm, vuint8m1_t vd, const uint8_t *rs1,
                             size_t *new_vl, size_t vl);
vuint8m2_t __riscv_vle8ff_mu(vbool4_t vm, vuint8m2_t vd, const uint8_t *rs1,
                             size_t *new_vl, size_t vl);
vuint8m4_t __riscv_vle8ff_mu(vbool2_t vm, vuint8m4_t vd, const uint8_t *rs1,
                             size_t *new_vl, size_t vl);
vuint8m8_t __riscv_vle8ff_mu(vbool1_t vm, vuint8m8_t vd, const uint8_t *rs1,
                             size_t *new_vl, size_t vl);
vuint16mf4_t __riscv_vle16ff_mu(vbool64_t vm, vuint16mf4_t vd,
                                const uint16_t *rs1, size_t *new_vl, size_t vl);
vuint16mf2_t __riscv_vle16ff_mu(vbool32_t vm, vuint16mf2_t vd,
                                const uint16_t *rs1, size_t *new_vl, size_t vl);
vuint16m1_t __riscv_vle16ff_mu(vbool16_t vm, vuint16m1_t vd,
                               const uint16_t *rs1, size_t *new_vl, size_t vl);
vuint16m2_t __riscv_vle16ff_mu(vbool8_t vm, vuint16m2_t vd, const uint16_t *rs1,
                               size_t *new_vl, size_t vl);
vuint16m4_t __riscv_vle16ff_mu(vbool4_t vm, vuint16m4_t vd, const uint16_t *rs1,
                               size_t *new_vl, size_t vl);
vuint16m8_t __riscv_vle16ff_mu(vbool2_t vm, vuint16m8_t vd, const uint16_t *rs1,
                               size_t *new_vl, size_t vl);
vuint32mf2_t __riscv_vle32ff_mu(vbool64_t vm, vuint32mf2_t vd,
                                const uint32_t *rs1, size_t *new_vl, size_t vl);
vuint32m1_t __riscv_vle32ff_mu(vbool32_t vm, vuint32m1_t vd,
                               const uint32_t *rs1, size_t *new_vl, size_t vl);
vuint32m2_t __riscv_vle32ff_mu(vbool16_t vm, vuint32m2_t vd,
                               const uint32_t *rs1, size_t *new_vl, size_t vl);
vuint32m4_t __riscv_vle32ff_mu(vbool8_t vm, vuint32m4_t vd, const uint32_t *rs1,
                               size_t *new_vl, size_t vl);
vuint32m8_t __riscv_vle32ff_mu(vbool4_t vm, vuint32m8_t vd, const uint32_t *rs1,
                               size_t *new_vl, size_t vl);
vuint64m1_t __riscv_vle64ff_mu(vbool64_t vm, vuint64m1_t vd,
                               const uint64_t *rs1, size_t *new_vl, size_t vl);
vuint64m2_t __riscv_vle64ff_mu(vbool32_t vm, vuint64m2_t vd,
                               const uint64_t *rs1, size_t *new_vl, size_t vl);
vuint64m4_t __riscv_vle64ff_mu(vbool16_t vm, vuint64m4_t vd,
                               const uint64_t *rs1, size_t *new_vl, size_t vl);
vuint64m8_t __riscv_vle64ff_mu(vbool8_t vm, vuint64m8_t vd, const uint64_t *rs1,
                               size_t *new_vl, size_t vl);