The instructions have the same function as RVV1.0. Overall there are only general differences between XTheadVector and RVV1.0.
Signed-off-by: Huang Tao <eric.hu...@linux.alibaba.com> --- target/riscv/helper.h | 13 +++ .../riscv/insn_trans/trans_xtheadvector.c.inc | 99 +++++++++++++++++-- target/riscv/xtheadvector_helper.c | 26 +++++ 3 files changed, 132 insertions(+), 6 deletions(-) diff --git a/target/riscv/helper.h b/target/riscv/helper.h index 77251af8c9..d3170ba91f 100644 --- a/target/riscv/helper.h +++ b/target/riscv/helper.h @@ -1625,3 +1625,16 @@ DEF_HELPER_6(th_vsra_vx_b, void, ptr, ptr, tl, ptr, env, i32) DEF_HELPER_6(th_vsra_vx_h, void, ptr, ptr, tl, ptr, env, i32) DEF_HELPER_6(th_vsra_vx_w, void, ptr, ptr, tl, ptr, env, i32) DEF_HELPER_6(th_vsra_vx_d, void, ptr, ptr, tl, ptr, env, i32) + +DEF_HELPER_6(th_vnsrl_vv_b, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(th_vnsrl_vv_h, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(th_vnsrl_vv_w, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(th_vnsra_vv_b, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(th_vnsra_vv_h, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(th_vnsra_vv_w, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(th_vnsrl_vx_b, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(th_vnsrl_vx_h, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(th_vnsrl_vx_w, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(th_vnsra_vx_b, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(th_vnsra_vx_h, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(th_vnsra_vx_w, void, ptr, ptr, tl, ptr, env, i32) diff --git a/target/riscv/insn_trans/trans_xtheadvector.c.inc b/target/riscv/insn_trans/trans_xtheadvector.c.inc index d72320699c..68810ff0ec 100644 --- a/target/riscv/insn_trans/trans_xtheadvector.c.inc +++ b/target/riscv/insn_trans/trans_xtheadvector.c.inc @@ -1377,18 +1377,105 @@ GEN_OPIVI_GVEC_TRANS_TH(th_vsll_vi, IMM_TRUNC_SEW, th_vsll_vx, shli) GEN_OPIVI_GVEC_TRANS_TH(th_vsrl_vi, IMM_TRUNC_SEW, th_vsrl_vx, shri) GEN_OPIVI_GVEC_TRANS_TH(th_vsra_vi, IMM_TRUNC_SEW, th_vsra_vx, sari) +/* Vector Narrowing Integer Right Shift Instructions */ +static bool opivv_narrow_check_th(DisasContext *s, arg_rmrr *a) +{ + return (require_xtheadvector(s) && + vext_check_isa_ill(s) && + th_check_overlap_mask(s, a->rd, a->vm, false) && + th_check_reg(s, a->rd, false) && + th_check_reg(s, a->rs2, true) && + th_check_reg(s, a->rs1, false) && + th_check_overlap_group(a->rd, 1 << s->lmul, a->rs2, + 2 << s->lmul) && + (s->lmul < 0x3) && (s->sew < 0x3)); +} + +/* OPIVV with NARROW */ +#define GEN_OPIVV_NARROW_TRANS_TH(NAME) \ +static bool trans_##NAME(DisasContext *s, arg_rmrr *a) \ +{ \ + if (opivv_narrow_check_th(s, a)) { \ + uint32_t data = 0; \ + static gen_helper_gvec_4_ptr * const fns[3] = { \ + gen_helper_##NAME##_b, \ + gen_helper_##NAME##_h, \ + gen_helper_##NAME##_w, \ + }; \ + \ + data = FIELD_DP32(data, VDATA_TH, MLEN, s->mlen); \ + data = FIELD_DP32(data, VDATA_TH, VM, a->vm); \ + data = FIELD_DP32(data, VDATA_TH, LMUL, s->lmul); \ + tcg_gen_gvec_4_ptr(vreg_ofs(s, a->rd), \ + vreg_ofs(s, 0), \ + vreg_ofs(s, a->rs1), \ + vreg_ofs(s, a->rs2), tcg_env, \ + s->cfg_ptr->vlenb, \ + s->cfg_ptr->vlenb, data, \ + fns[s->sew]); \ + finalize_rvv_inst(s); \ + return true; \ + } \ + return false; \ +} +GEN_OPIVV_NARROW_TRANS_TH(th_vnsra_vv) +GEN_OPIVV_NARROW_TRANS_TH(th_vnsrl_vv) + +static bool opivx_narrow_check_th(DisasContext *s, arg_rmrr *a) +{ + return (require_xtheadvector(s) && + vext_check_isa_ill(s) && + th_check_overlap_mask(s, a->rd, a->vm, false) && + th_check_reg(s, a->rd, false) && + th_check_reg(s, a->rs2, true) && + th_check_overlap_group(a->rd, 1 << s->lmul, a->rs2, + 2 << s->lmul) && + (s->lmul < 0x3) && (s->sew < 0x3)); +} + +/* OPIVX with NARROW */ +#define GEN_OPIVX_NARROW_TRANS_TH(NAME) \ +static bool trans_##NAME(DisasContext *s, arg_rmrr *a) \ +{ \ + if (opivx_narrow_check_th(s, a)) { \ + static gen_helper_opivx * const fns[3] = { \ + gen_helper_##NAME##_b, \ + gen_helper_##NAME##_h, \ + gen_helper_##NAME##_w, \ + }; \ + return opivx_trans_th(a->rd, a->rs1, a->rs2, a->vm, fns[s->sew], s);\ + } \ + return false; \ +} + +GEN_OPIVX_NARROW_TRANS_TH(th_vnsra_vx) +GEN_OPIVX_NARROW_TRANS_TH(th_vnsrl_vx) + +/* OPIVI with NARROW */ +#define GEN_OPIVI_NARROW_TRANS_TH(NAME, ZX, OPIVX) \ +static bool trans_##NAME(DisasContext *s, arg_rmrr *a) \ +{ \ + if (opivx_narrow_check_th(s, a)) { \ + static gen_helper_opivx * const fns[3] = { \ + gen_helper_##OPIVX##_b, \ + gen_helper_##OPIVX##_h, \ + gen_helper_##OPIVX##_w, \ + }; \ + return opivi_trans_th(a->rd, a->rs1, a->rs2, a->vm, \ + fns[s->sew], s, ZX); \ + } \ + return false; \ +} + +GEN_OPIVI_NARROW_TRANS_TH(th_vnsra_vi, IMM_ZX, th_vnsra_vx) +GEN_OPIVI_NARROW_TRANS_TH(th_vnsrl_vi, IMM_ZX, th_vnsrl_vx) + #define TH_TRANS_STUB(NAME) \ static bool trans_##NAME(DisasContext *s, arg_##NAME *a) \ { \ return require_xtheadvector(s); \ } -TH_TRANS_STUB(th_vnsrl_vv) -TH_TRANS_STUB(th_vnsrl_vx) -TH_TRANS_STUB(th_vnsrl_vi) -TH_TRANS_STUB(th_vnsra_vv) -TH_TRANS_STUB(th_vnsra_vx) -TH_TRANS_STUB(th_vnsra_vi) TH_TRANS_STUB(th_vmseq_vv) TH_TRANS_STUB(th_vmseq_vx) TH_TRANS_STUB(th_vmseq_vi) diff --git a/target/riscv/xtheadvector_helper.c b/target/riscv/xtheadvector_helper.c index d3f10ad873..f4bd80349d 100644 --- a/target/riscv/xtheadvector_helper.c +++ b/target/riscv/xtheadvector_helper.c @@ -1389,3 +1389,29 @@ GEN_TH_SHIFT_VX(th_vsra_vx_w, int32_t, int32_t, H4, H4, TH_SRL, 0x1f, clearl_th) GEN_TH_SHIFT_VX(th_vsra_vx_d, int64_t, int64_t, H8, H8, TH_SRL, 0x3f, clearq_th) + +/* Vector Narrowing Integer Right Shift Instructions */ +GEN_TH_SHIFT_VV(th_vnsrl_vv_b, uint8_t, uint16_t, H1, H2, TH_SRL, + 0xf, clearb_th) +GEN_TH_SHIFT_VV(th_vnsrl_vv_h, uint16_t, uint32_t, H2, H4, TH_SRL, + 0x1f, clearh_th) +GEN_TH_SHIFT_VV(th_vnsrl_vv_w, uint32_t, uint64_t, H4, H8, TH_SRL, + 0x3f, clearl_th) +GEN_TH_SHIFT_VV(th_vnsra_vv_b, uint8_t, int16_t, H1, H2, TH_SRL, + 0xf, clearb_th) +GEN_TH_SHIFT_VV(th_vnsra_vv_h, uint16_t, int32_t, H2, H4, TH_SRL, + 0x1f, clearh_th) +GEN_TH_SHIFT_VV(th_vnsra_vv_w, uint32_t, int64_t, H4, H8, TH_SRL, + 0x3f, clearl_th) +GEN_TH_SHIFT_VX(th_vnsrl_vx_b, uint8_t, uint16_t, H1, H2, TH_SRL, + 0xf, clearb_th) +GEN_TH_SHIFT_VX(th_vnsrl_vx_h, uint16_t, uint32_t, H2, H4, TH_SRL, + 0x1f, clearh_th) +GEN_TH_SHIFT_VX(th_vnsrl_vx_w, uint32_t, uint64_t, H4, H8, TH_SRL, + 0x3f, clearl_th) +GEN_TH_SHIFT_VX(th_vnsra_vx_b, int8_t, int16_t, H1, H2, TH_SRL, + 0xf, clearb_th) +GEN_TH_SHIFT_VX(th_vnsra_vx_h, int16_t, int32_t, H2, H4, TH_SRL, + 0x1f, clearh_th) +GEN_TH_SHIFT_VX(th_vnsra_vx_w, int32_t, int64_t, H4, H8, TH_SRL, + 0x3f, clearl_th) -- 2.44.0