Untitled :: RISC-V Ratified Specifications Library

Vector Mask Intrinsics

Vector Mask-Register Logical

Intrinsics here don’t have a policy variant.

Vector count population in mask `vcpop.m`

Intrinsics here don’t have a policy variant.

`vfirst` find-first-set mask bit

Intrinsics here don’t have a policy variant.

`vmsbf.m` set-before-first mask bit

// masked functions
vbool1_t __riscv_vmsbf_m_b1_mu(vbool1_t vm, vbool1_t vd, vbool1_t vs2,
                               size_t vl);
vbool2_t __riscv_vmsbf_m_b2_mu(vbool2_t vm, vbool2_t vd, vbool2_t vs2,
                               size_t vl);
vbool4_t __riscv_vmsbf_m_b4_mu(vbool4_t vm, vbool4_t vd, vbool4_t vs2,
                               size_t vl);
vbool8_t __riscv_vmsbf_m_b8_mu(vbool8_t vm, vbool8_t vd, vbool8_t vs2,
                               size_t vl);
vbool16_t __riscv_vmsbf_m_b16_mu(vbool16_t vm, vbool16_t vd, vbool16_t vs2,
                                 size_t vl);
vbool32_t __riscv_vmsbf_m_b32_mu(vbool32_t vm, vbool32_t vd, vbool32_t vs2,
                                 size_t vl);
vbool64_t __riscv_vmsbf_m_b64_mu(vbool64_t vm, vbool64_t vd, vbool64_t vs2,
                                 size_t vl);

`vmsif.m` set-including-first mask bit

// masked functions
vbool1_t __riscv_vmsif_m_b1_mu(vbool1_t vm, vbool1_t vd, vbool1_t vs2,
                               size_t vl);
vbool2_t __riscv_vmsif_m_b2_mu(vbool2_t vm, vbool2_t vd, vbool2_t vs2,
                               size_t vl);
vbool4_t __riscv_vmsif_m_b4_mu(vbool4_t vm, vbool4_t vd, vbool4_t vs2,
                               size_t vl);
vbool8_t __riscv_vmsif_m_b8_mu(vbool8_t vm, vbool8_t vd, vbool8_t vs2,
                               size_t vl);
vbool16_t __riscv_vmsif_m_b16_mu(vbool16_t vm, vbool16_t vd, vbool16_t vs2,
                                 size_t vl);
vbool32_t __riscv_vmsif_m_b32_mu(vbool32_t vm, vbool32_t vd, vbool32_t vs2,
                                 size_t vl);
vbool64_t __riscv_vmsif_m_b64_mu(vbool64_t vm, vbool64_t vd, vbool64_t vs2,
                                 size_t vl);

`vmsof.m` set-only-first mask bit

// masked functions
vbool1_t __riscv_vmsof_m_b1_mu(vbool1_t vm, vbool1_t vd, vbool1_t vs2,
                               size_t vl);
vbool2_t __riscv_vmsof_m_b2_mu(vbool2_t vm, vbool2_t vd, vbool2_t vs2,
                               size_t vl);
vbool4_t __riscv_vmsof_m_b4_mu(vbool4_t vm, vbool4_t vd, vbool4_t vs2,
                               size_t vl);
vbool8_t __riscv_vmsof_m_b8_mu(vbool8_t vm, vbool8_t vd, vbool8_t vs2,
                               size_t vl);
vbool16_t __riscv_vmsof_m_b16_mu(vbool16_t vm, vbool16_t vd, vbool16_t vs2,
                                 size_t vl);
vbool32_t __riscv_vmsof_m_b32_mu(vbool32_t vm, vbool32_t vd, vbool32_t vs2,
                                 size_t vl);
vbool64_t __riscv_vmsof_m_b64_mu(vbool64_t vm, vbool64_t vd, vbool64_t vs2,
                                 size_t vl);

Vector Iota Intrinsics

vuint8mf8_t __riscv_viota_m_u8mf8_tu(vuint8mf8_t vd, vbool64_t vs2, size_t vl);
vuint8mf4_t __riscv_viota_m_u8mf4_tu(vuint8mf4_t vd, vbool32_t vs2, size_t vl);
vuint8mf2_t __riscv_viota_m_u8mf2_tu(vuint8mf2_t vd, vbool16_t vs2, size_t vl);
vuint8m1_t __riscv_viota_m_u8m1_tu(vuint8m1_t vd, vbool8_t vs2, size_t vl);
vuint8m2_t __riscv_viota_m_u8m2_tu(vuint8m2_t vd, vbool4_t vs2, size_t vl);
vuint8m4_t __riscv_viota_m_u8m4_tu(vuint8m4_t vd, vbool2_t vs2, size_t vl);
vuint8m8_t __riscv_viota_m_u8m8_tu(vuint8m8_t vd, vbool1_t vs2, size_t vl);
vuint16mf4_t __riscv_viota_m_u16mf4_tu(vuint16mf4_t vd, vbool64_t vs2,
                                       size_t vl);
vuint16mf2_t __riscv_viota_m_u16mf2_tu(vuint16mf2_t vd, vbool32_t vs2,
                                       size_t vl);
vuint16m1_t __riscv_viota_m_u16m1_tu(vuint16m1_t vd, vbool16_t vs2, size_t vl);
vuint16m2_t __riscv_viota_m_u16m2_tu(vuint16m2_t vd, vbool8_t vs2, size_t vl);
vuint16m4_t __riscv_viota_m_u16m4_tu(vuint16m4_t vd, vbool4_t vs2, size_t vl);
vuint16m8_t __riscv_viota_m_u16m8_tu(vuint16m8_t vd, vbool2_t vs2, size_t vl);
vuint32mf2_t __riscv_viota_m_u32mf2_tu(vuint32mf2_t vd, vbool64_t vs2,
                                       size_t vl);
vuint32m1_t __riscv_viota_m_u32m1_tu(vuint32m1_t vd, vbool32_t vs2, size_t vl);
vuint32m2_t __riscv_viota_m_u32m2_tu(vuint32m2_t vd, vbool16_t vs2, size_t vl);
vuint32m4_t __riscv_viota_m_u32m4_tu(vuint32m4_t vd, vbool8_t vs2, size_t vl);
vuint32m8_t __riscv_viota_m_u32m8_tu(vuint32m8_t vd, vbool4_t vs2, size_t vl);
vuint64m1_t __riscv_viota_m_u64m1_tu(vuint64m1_t vd, vbool64_t vs2, size_t vl);
vuint64m2_t __riscv_viota_m_u64m2_tu(vuint64m2_t vd, vbool32_t vs2, size_t vl);
vuint64m4_t __riscv_viota_m_u64m4_tu(vuint64m4_t vd, vbool16_t vs2, size_t vl);
vuint64m8_t __riscv_viota_m_u64m8_tu(vuint64m8_t vd, vbool8_t vs2, size_t vl);
// masked functions
vuint8mf8_t __riscv_viota_m_u8mf8_tum(vbool64_t vm, vuint8mf8_t vd,
                                      vbool64_t vs2, size_t vl);
vuint8mf4_t __riscv_viota_m_u8mf4_tum(vbool32_t vm, vuint8mf4_t vd,
                                      vbool32_t vs2, size_t vl);
vuint8mf2_t __riscv_viota_m_u8mf2_tum(vbool16_t vm, vuint8mf2_t vd,
                                      vbool16_t vs2, size_t vl);
vuint8m1_t __riscv_viota_m_u8m1_tum(vbool8_t vm, vuint8m1_t vd, vbool8_t vs2,
                                    size_t vl);
vuint8m2_t __riscv_viota_m_u8m2_tum(vbool4_t vm, vuint8m2_t vd, vbool4_t vs2,
                                    size_t vl);
vuint8m4_t __riscv_viota_m_u8m4_tum(vbool2_t vm, vuint8m4_t vd, vbool2_t vs2,
                                    size_t vl);
vuint8m8_t __riscv_viota_m_u8m8_tum(vbool1_t vm, vuint8m8_t vd, vbool1_t vs2,
                                    size_t vl);
vuint16mf4_t __riscv_viota_m_u16mf4_tum(vbool64_t vm, vuint16mf4_t vd,
                                        vbool64_t vs2, size_t vl);
vuint16mf2_t __riscv_viota_m_u16mf2_tum(vbool32_t vm, vuint16mf2_t vd,
                                        vbool32_t vs2, size_t vl);
vuint16m1_t __riscv_viota_m_u16m1_tum(vbool16_t vm, vuint16m1_t vd,
                                      vbool16_t vs2, size_t vl);
vuint16m2_t __riscv_viota_m_u16m2_tum(vbool8_t vm, vuint16m2_t vd, vbool8_t vs2,
                                      size_t vl);
vuint16m4_t __riscv_viota_m_u16m4_tum(vbool4_t vm, vuint16m4_t vd, vbool4_t vs2,
                                      size_t vl);
vuint16m8_t __riscv_viota_m_u16m8_tum(vbool2_t vm, vuint16m8_t vd, vbool2_t vs2,
                                      size_t vl);
vuint32mf2_t __riscv_viota_m_u32mf2_tum(vbool64_t vm, vuint32mf2_t vd,
                                        vbool64_t vs2, size_t vl);
vuint32m1_t __riscv_viota_m_u32m1_tum(vbool32_t vm, vuint32m1_t vd,
                                      vbool32_t vs2, size_t vl);
vuint32m2_t __riscv_viota_m_u32m2_tum(vbool16_t vm, vuint32m2_t vd,
                                      vbool16_t vs2, size_t vl);
vuint32m4_t __riscv_viota_m_u32m4_tum(vbool8_t vm, vuint32m4_t vd, vbool8_t vs2,
                                      size_t vl);
vuint32m8_t __riscv_viota_m_u32m8_tum(vbool4_t vm, vuint32m8_t vd, vbool4_t vs2,
                                      size_t vl);
vuint64m1_t __riscv_viota_m_u64m1_tum(vbool64_t vm, vuint64m1_t vd,
                                      vbool64_t vs2, size_t vl);
vuint64m2_t __riscv_viota_m_u64m2_tum(vbool32_t vm, vuint64m2_t vd,
                                      vbool32_t vs2, size_t vl);
vuint64m4_t __riscv_viota_m_u64m4_tum(vbool16_t vm, vuint64m4_t vd,
                                      vbool16_t vs2, size_t vl);
vuint64m8_t __riscv_viota_m_u64m8_tum(vbool8_t vm, vuint64m8_t vd, vbool8_t vs2,
                                      size_t vl);
// masked functions
vuint8mf8_t __riscv_viota_m_u8mf8_tumu(vbool64_t vm, vuint8mf8_t vd,
                                       vbool64_t vs2, size_t vl);
vuint8mf4_t __riscv_viota_m_u8mf4_tumu(vbool32_t vm, vuint8mf4_t vd,
                                       vbool32_t vs2, size_t vl);
vuint8mf2_t __riscv_viota_m_u8mf2_tumu(vbool16_t vm, vuint8mf2_t vd,
                                       vbool16_t vs2, size_t vl);
vuint8m1_t __riscv_viota_m_u8m1_tumu(vbool8_t vm, vuint8m1_t vd, vbool8_t vs2,
                                     size_t vl);
vuint8m2_t __riscv_viota_m_u8m2_tumu(vbool4_t vm, vuint8m2_t vd, vbool4_t vs2,
                                     size_t vl);
vuint8m4_t __riscv_viota_m_u8m4_tumu(vbool2_t vm, vuint8m4_t vd, vbool2_t vs2,
                                     size_t vl);
vuint8m8_t __riscv_viota_m_u8m8_tumu(vbool1_t vm, vuint8m8_t vd, vbool1_t vs2,
                                     size_t vl);
vuint16mf4_t __riscv_viota_m_u16mf4_tumu(vbool64_t vm, vuint16mf4_t vd,
                                         vbool64_t vs2, size_t vl);
vuint16mf2_t __riscv_viota_m_u16mf2_tumu(vbool32_t vm, vuint16mf2_t vd,
                                         vbool32_t vs2, size_t vl);
vuint16m1_t __riscv_viota_m_u16m1_tumu(vbool16_t vm, vuint16m1_t vd,
                                       vbool16_t vs2, size_t vl);
vuint16m2_t __riscv_viota_m_u16m2_tumu(vbool8_t vm, vuint16m2_t vd,
                                       vbool8_t vs2, size_t vl);
vuint16m4_t __riscv_viota_m_u16m4_tumu(vbool4_t vm, vuint16m4_t vd,
                                       vbool4_t vs2, size_t vl);
vuint16m8_t __riscv_viota_m_u16m8_tumu(vbool2_t vm, vuint16m8_t vd,
                                       vbool2_t vs2, size_t vl);
vuint32mf2_t __riscv_viota_m_u32mf2_tumu(vbool64_t vm, vuint32mf2_t vd,
                                         vbool64_t vs2, size_t vl);
vuint32m1_t __riscv_viota_m_u32m1_tumu(vbool32_t vm, vuint32m1_t vd,
                                       vbool32_t vs2, size_t vl);
vuint32m2_t __riscv_viota_m_u32m2_tumu(vbool16_t vm, vuint32m2_t vd,
                                       vbool16_t vs2, size_t vl);
vuint32m4_t __riscv_viota_m_u32m4_tumu(vbool8_t vm, vuint32m4_t vd,
                                       vbool8_t vs2, size_t vl);
vuint32m8_t __riscv_viota_m_u32m8_tumu(vbool4_t vm, vuint32m8_t vd,
                                       vbool4_t vs2, size_t vl);
vuint64m1_t __riscv_viota_m_u64m1_tumu(vbool64_t vm, vuint64m1_t vd,
                                       vbool64_t vs2, size_t vl);
vuint64m2_t __riscv_viota_m_u64m2_tumu(vbool32_t vm, vuint64m2_t vd,
                                       vbool32_t vs2, size_t vl);
vuint64m4_t __riscv_viota_m_u64m4_tumu(vbool16_t vm, vuint64m4_t vd,
                                       vbool16_t vs2, size_t vl);
vuint64m8_t __riscv_viota_m_u64m8_tumu(vbool8_t vm, vuint64m8_t vd,
                                       vbool8_t vs2, size_t vl);
// masked functions
vuint8mf8_t __riscv_viota_m_u8mf8_mu(vbool64_t vm, vuint8mf8_t vd,
                                     vbool64_t vs2, size_t vl);
vuint8mf4_t __riscv_viota_m_u8mf4_mu(vbool32_t vm, vuint8mf4_t vd,
                                     vbool32_t vs2, size_t vl);
vuint8mf2_t __riscv_viota_m_u8mf2_mu(vbool16_t vm, vuint8mf2_t vd,
                                     vbool16_t vs2, size_t vl);
vuint8m1_t __riscv_viota_m_u8m1_mu(vbool8_t vm, vuint8m1_t vd, vbool8_t vs2,
                                   size_t vl);
vuint8m2_t __riscv_viota_m_u8m2_mu(vbool4_t vm, vuint8m2_t vd, vbool4_t vs2,
                                   size_t vl);
vuint8m4_t __riscv_viota_m_u8m4_mu(vbool2_t vm, vuint8m4_t vd, vbool2_t vs2,
                                   size_t vl);
vuint8m8_t __riscv_viota_m_u8m8_mu(vbool1_t vm, vuint8m8_t vd, vbool1_t vs2,
                                   size_t vl);
vuint16mf4_t __riscv_viota_m_u16mf4_mu(vbool64_t vm, vuint16mf4_t vd,
                                       vbool64_t vs2, size_t vl);
vuint16mf2_t __riscv_viota_m_u16mf2_mu(vbool32_t vm, vuint16mf2_t vd,
                                       vbool32_t vs2, size_t vl);
vuint16m1_t __riscv_viota_m_u16m1_mu(vbool16_t vm, vuint16m1_t vd,
                                     vbool16_t vs2, size_t vl);
vuint16m2_t __riscv_viota_m_u16m2_mu(vbool8_t vm, vuint16m2_t vd, vbool8_t vs2,
                                     size_t vl);
vuint16m4_t __riscv_viota_m_u16m4_mu(vbool4_t vm, vuint16m4_t vd, vbool4_t vs2,
                                     size_t vl);
vuint16m8_t __riscv_viota_m_u16m8_mu(vbool2_t vm, vuint16m8_t vd, vbool2_t vs2,
                                     size_t vl);
vuint32mf2_t __riscv_viota_m_u32mf2_mu(vbool64_t vm, vuint32mf2_t vd,
                                       vbool64_t vs2, size_t vl);
vuint32m1_t __riscv_viota_m_u32m1_mu(vbool32_t vm, vuint32m1_t vd,
                                     vbool32_t vs2, size_t vl);
vuint32m2_t __riscv_viota_m_u32m2_mu(vbool16_t vm, vuint32m2_t vd,
                                     vbool16_t vs2, size_t vl);
vuint32m4_t __riscv_viota_m_u32m4_mu(vbool8_t vm, vuint32m4_t vd, vbool8_t vs2,
                                     size_t vl);
vuint32m8_t __riscv_viota_m_u32m8_mu(vbool4_t vm, vuint32m8_t vd, vbool4_t vs2,
                                     size_t vl);
vuint64m1_t __riscv_viota_m_u64m1_mu(vbool64_t vm, vuint64m1_t vd,
                                     vbool64_t vs2, size_t vl);
vuint64m2_t __riscv_viota_m_u64m2_mu(vbool32_t vm, vuint64m2_t vd,
                                     vbool32_t vs2, size_t vl);
vuint64m4_t __riscv_viota_m_u64m4_mu(vbool16_t vm, vuint64m4_t vd,
                                     vbool16_t vs2, size_t vl);
vuint64m8_t __riscv_viota_m_u64m8_mu(vbool8_t vm, vuint64m8_t vd, vbool8_t vs2,
                                     size_t vl);

Vector Element Index Intrinsics

vuint8mf8_t __riscv_vid_v_u8mf8_tu(vuint8mf8_t vd, size_t vl);
vuint8mf4_t __riscv_vid_v_u8mf4_tu(vuint8mf4_t vd, size_t vl);
vuint8mf2_t __riscv_vid_v_u8mf2_tu(vuint8mf2_t vd, size_t vl);
vuint8m1_t __riscv_vid_v_u8m1_tu(vuint8m1_t vd, size_t vl);
vuint8m2_t __riscv_vid_v_u8m2_tu(vuint8m2_t vd, size_t vl);
vuint8m4_t __riscv_vid_v_u8m4_tu(vuint8m4_t vd, size_t vl);
vuint8m8_t __riscv_vid_v_u8m8_tu(vuint8m8_t vd, size_t vl);
vuint16mf4_t __riscv_vid_v_u16mf4_tu(vuint16mf4_t vd, size_t vl);
vuint16mf2_t __riscv_vid_v_u16mf2_tu(vuint16mf2_t vd, size_t vl);
vuint16m1_t __riscv_vid_v_u16m1_tu(vuint16m1_t vd, size_t vl);
vuint16m2_t __riscv_vid_v_u16m2_tu(vuint16m2_t vd, size_t vl);
vuint16m4_t __riscv_vid_v_u16m4_tu(vuint16m4_t vd, size_t vl);
vuint16m8_t __riscv_vid_v_u16m8_tu(vuint16m8_t vd, size_t vl);
vuint32mf2_t __riscv_vid_v_u32mf2_tu(vuint32mf2_t vd, size_t vl);
vuint32m1_t __riscv_vid_v_u32m1_tu(vuint32m1_t vd, size_t vl);
vuint32m2_t __riscv_vid_v_u32m2_tu(vuint32m2_t vd, size_t vl);
vuint32m4_t __riscv_vid_v_u32m4_tu(vuint32m4_t vd, size_t vl);
vuint32m8_t __riscv_vid_v_u32m8_tu(vuint32m8_t vd, size_t vl);
vuint64m1_t __riscv_vid_v_u64m1_tu(vuint64m1_t vd, size_t vl);
vuint64m2_t __riscv_vid_v_u64m2_tu(vuint64m2_t vd, size_t vl);
vuint64m4_t __riscv_vid_v_u64m4_tu(vuint64m4_t vd, size_t vl);
vuint64m8_t __riscv_vid_v_u64m8_tu(vuint64m8_t vd, size_t vl);
// masked functions
vuint8mf8_t __riscv_vid_v_u8mf8_tum(vbool64_t vm, vuint8mf8_t vd, size_t vl);
vuint8mf4_t __riscv_vid_v_u8mf4_tum(vbool32_t vm, vuint8mf4_t vd, size_t vl);
vuint8mf2_t __riscv_vid_v_u8mf2_tum(vbool16_t vm, vuint8mf2_t vd, size_t vl);
vuint8m1_t __riscv_vid_v_u8m1_tum(vbool8_t vm, vuint8m1_t vd, size_t vl);
vuint8m2_t __riscv_vid_v_u8m2_tum(vbool4_t vm, vuint8m2_t vd, size_t vl);
vuint8m4_t __riscv_vid_v_u8m4_tum(vbool2_t vm, vuint8m4_t vd, size_t vl);
vuint8m8_t __riscv_vid_v_u8m8_tum(vbool1_t vm, vuint8m8_t vd, size_t vl);
vuint16mf4_t __riscv_vid_v_u16mf4_tum(vbool64_t vm, vuint16mf4_t vd, size_t vl);
vuint16mf2_t __riscv_vid_v_u16mf2_tum(vbool32_t vm, vuint16mf2_t vd, size_t vl);
vuint16m1_t __riscv_vid_v_u16m1_tum(vbool16_t vm, vuint16m1_t vd, size_t vl);
vuint16m2_t __riscv_vid_v_u16m2_tum(vbool8_t vm, vuint16m2_t vd, size_t vl);
vuint16m4_t __riscv_vid_v_u16m4_tum(vbool4_t vm, vuint16m4_t vd, size_t vl);
vuint16m8_t __riscv_vid_v_u16m8_tum(vbool2_t vm, vuint16m8_t vd, size_t vl);
vuint32mf2_t __riscv_vid_v_u32mf2_tum(vbool64_t vm, vuint32mf2_t vd, size_t vl);
vuint32m1_t __riscv_vid_v_u32m1_tum(vbool32_t vm, vuint32m1_t vd, size_t vl);
vuint32m2_t __riscv_vid_v_u32m2_tum(vbool16_t vm, vuint32m2_t vd, size_t vl);
vuint32m4_t __riscv_vid_v_u32m4_tum(vbool8_t vm, vuint32m4_t vd, size_t vl);
vuint32m8_t __riscv_vid_v_u32m8_tum(vbool4_t vm, vuint32m8_t vd, size_t vl);
vuint64m1_t __riscv_vid_v_u64m1_tum(vbool64_t vm, vuint64m1_t vd, size_t vl);
vuint64m2_t __riscv_vid_v_u64m2_tum(vbool32_t vm, vuint64m2_t vd, size_t vl);
vuint64m4_t __riscv_vid_v_u64m4_tum(vbool16_t vm, vuint64m4_t vd, size_t vl);
vuint64m8_t __riscv_vid_v_u64m8_tum(vbool8_t vm, vuint64m8_t vd, size_t vl);
// masked functions
vuint8mf8_t __riscv_vid_v_u8mf8_tumu(vbool64_t vm, vuint8mf8_t vd, size_t vl);
vuint8mf4_t __riscv_vid_v_u8mf4_tumu(vbool32_t vm, vuint8mf4_t vd, size_t vl);
vuint8mf2_t __riscv_vid_v_u8mf2_tumu(vbool16_t vm, vuint8mf2_t vd, size_t vl);
vuint8m1_t __riscv_vid_v_u8m1_tumu(vbool8_t vm, vuint8m1_t vd, size_t vl);
vuint8m2_t __riscv_vid_v_u8m2_tumu(vbool4_t vm, vuint8m2_t vd, size_t vl);
vuint8m4_t __riscv_vid_v_u8m4_tumu(vbool2_t vm, vuint8m4_t vd, size_t vl);
vuint8m8_t __riscv_vid_v_u8m8_tumu(vbool1_t vm, vuint8m8_t vd, size_t vl);
vuint16mf4_t __riscv_vid_v_u16mf4_tumu(vbool64_t vm, vuint16mf4_t vd,
                                       size_t vl);
vuint16mf2_t __riscv_vid_v_u16mf2_tumu(vbool32_t vm, vuint16mf2_t vd,
                                       size_t vl);
vuint16m1_t __riscv_vid_v_u16m1_tumu(vbool16_t vm, vuint16m1_t vd, size_t vl);
vuint16m2_t __riscv_vid_v_u16m2_tumu(vbool8_t vm, vuint16m2_t vd, size_t vl);
vuint16m4_t __riscv_vid_v_u16m4_tumu(vbool4_t vm, vuint16m4_t vd, size_t vl);
vuint16m8_t __riscv_vid_v_u16m8_tumu(vbool2_t vm, vuint16m8_t vd, size_t vl);
vuint32mf2_t __riscv_vid_v_u32mf2_tumu(vbool64_t vm, vuint32mf2_t vd,
                                       size_t vl);
vuint32m1_t __riscv_vid_v_u32m1_tumu(vbool32_t vm, vuint32m1_t vd, size_t vl);
vuint32m2_t __riscv_vid_v_u32m2_tumu(vbool16_t vm, vuint32m2_t vd, size_t vl);
vuint32m4_t __riscv_vid_v_u32m4_tumu(vbool8_t vm, vuint32m4_t vd, size_t vl);
vuint32m8_t __riscv_vid_v_u32m8_tumu(vbool4_t vm, vuint32m8_t vd, size_t vl);
vuint64m1_t __riscv_vid_v_u64m1_tumu(vbool64_t vm, vuint64m1_t vd, size_t vl);
vuint64m2_t __riscv_vid_v_u64m2_tumu(vbool32_t vm, vuint64m2_t vd, size_t vl);
vuint64m4_t __riscv_vid_v_u64m4_tumu(vbool16_t vm, vuint64m4_t vd, size_t vl);
vuint64m8_t __riscv_vid_v_u64m8_tumu(vbool8_t vm, vuint64m8_t vd, size_t vl);
// masked functions
vuint8mf8_t __riscv_vid_v_u8mf8_mu(vbool64_t vm, vuint8mf8_t vd, size_t vl);
vuint8mf4_t __riscv_vid_v_u8mf4_mu(vbool32_t vm, vuint8mf4_t vd, size_t vl);
vuint8mf2_t __riscv_vid_v_u8mf2_mu(vbool16_t vm, vuint8mf2_t vd, size_t vl);
vuint8m1_t __riscv_vid_v_u8m1_mu(vbool8_t vm, vuint8m1_t vd, size_t vl);
vuint8m2_t __riscv_vid_v_u8m2_mu(vbool4_t vm, vuint8m2_t vd, size_t vl);
vuint8m4_t __riscv_vid_v_u8m4_mu(vbool2_t vm, vuint8m4_t vd, size_t vl);
vuint8m8_t __riscv_vid_v_u8m8_mu(vbool1_t vm, vuint8m8_t vd, size_t vl);
vuint16mf4_t __riscv_vid_v_u16mf4_mu(vbool64_t vm, vuint16mf4_t vd, size_t vl);
vuint16mf2_t __riscv_vid_v_u16mf2_mu(vbool32_t vm, vuint16mf2_t vd, size_t vl);
vuint16m1_t __riscv_vid_v_u16m1_mu(vbool16_t vm, vuint16m1_t vd, size_t vl);
vuint16m2_t __riscv_vid_v_u16m2_mu(vbool8_t vm, vuint16m2_t vd, size_t vl);
vuint16m4_t __riscv_vid_v_u16m4_mu(vbool4_t vm, vuint16m4_t vd, size_t vl);
vuint16m8_t __riscv_vid_v_u16m8_mu(vbool2_t vm, vuint16m8_t vd, size_t vl);
vuint32mf2_t __riscv_vid_v_u32mf2_mu(vbool64_t vm, vuint32mf2_t vd, size_t vl);
vuint32m1_t __riscv_vid_v_u32m1_mu(vbool32_t vm, vuint32m1_t vd, size_t vl);
vuint32m2_t __riscv_vid_v_u32m2_mu(vbool16_t vm, vuint32m2_t vd, size_t vl);
vuint32m4_t __riscv_vid_v_u32m4_mu(vbool8_t vm, vuint32m4_t vd, size_t vl);
vuint32m8_t __riscv_vid_v_u32m8_mu(vbool4_t vm, vuint32m8_t vd, size_t vl);
vuint64m1_t __riscv_vid_v_u64m1_mu(vbool64_t vm, vuint64m1_t vd, size_t vl);
vuint64m2_t __riscv_vid_v_u64m2_mu(vbool32_t vm, vuint64m2_t vd, size_t vl);
vuint64m4_t __riscv_vid_v_u64m4_mu(vbool16_t vm, vuint64m4_t vd, size_t vl);
vuint64m8_t __riscv_vid_v_u64m8_mu(vbool8_t vm, vuint64m8_t vd, size_t vl);