[v3,09/60] target/riscv: vector single-width integer add and subtract

Message ID	20200309082042.12967-10-zhiwei_liu@c-sky.com
State	New
Headers	show Return-Path: <qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org> From: LIU Zhiwei <zhiwei_liu@c-sky.com> To: richard.henderson@linaro.org, alistair23@gmail.com, chihmin.chao@sifive.com, palmer@dabbelt.com Subject: [PATCH v3 09/60] target/riscv: vector single-width integer add and subtract Date: Mon, 9 Mar 2020 16:19:51 +0800 Message-Id: <20200309082042.12967-10-zhiwei_liu@c-sky.com> In-Reply-To: <20200309082042.12967-1-zhiwei_liu@c-sky.com> References: <20200309082042.12967-1-zhiwei_liu@c-sky.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Precedence: list Cc: guoren@linux.alibaba.com, qemu-riscv@nongnu.org, qemu-devel@nongnu.org, wxy194768@alibaba-inc.com, wenmeng_zhang@c-sky.com, LIU Zhiwei <zhiwei_liu@c-sky.com> Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Sender: "Qemu-devel" <qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org>
Series	target/riscv: support vector extension v0.7.1 \| expand [v3,00/60] target/riscv: support vector extension v0.7.1 [v3,01/60] target/riscv: add vector extension field in CPURISCVState [v3,02/60] target/riscv: implementation-defined constant parameters [v3,03/60] target/riscv: support vector extension csr [v3,04/60] target/riscv: add vector configure instruction [v3,05/60] target/riscv: add vector stride load and store instructions [v3,06/60] target/riscv: add vector index load and store instructions [v3,07/60] target/riscv: add fault-only-first unit stride load [v3,08/60] target/riscv: add vector amo operations [v3,09/60] target/riscv: vector single-width integer add and subtract [v3,10/60] target/riscv: vector widening integer add and subtract [v3,11/60] target/riscv: vector integer add-with-carry / subtract-with-borrow instructions [v3,12/60] target/riscv: vector bitwise logical instructions [v3,13/60] target/riscv: vector single-width bit shift instructions [v3,14/60] target/riscv: vector narrowing integer right shift instructions [v3,15/60] target/riscv: vector integer comparison instructions [v3,17/60] target/riscv: vector single-width integer multiply instructions [v3,18/60] target/riscv: vector integer divide instructions [v3,19/60] target/riscv: vector widening integer multiply instructions [v3,20/60] target/riscv: vector single-width integer multiply-add instructions [v3,21/60] target/riscv: vector widening integer multiply-add instructions [v3,22/60] target/riscv: vector integer merge and move instructions [v3,23/60] target/riscv: vector single-width saturating add and subtract [v3,24/60] target/riscv: vector single-width averaging add and subtract [v3,25/60] target/riscv: vector single-width fractional multiply with rounding and saturation [v3,26/60] target/riscv: vector widening saturating scaled multiply-add [v3,27/60] target/riscv: vector single-width scaling shift instructions [v3,28/60] target/riscv: vector narrowing fixed-point clip instructions [v3,29/60] target/riscv: vector single-width floating-point add/subtract instructions [v3,30/60] target/riscv: vector widening floating-point add/subtract instructions [v3,31/60] target/riscv: vector single-width floating-point multiply/divide instructions [v3,32/60] target/riscv: vector widening floating-point multiply [v3,33/60] target/riscv: vector single-width floating-point fused multiply-add instructions [v3,34/60] target/riscv: vector widening floating-point fused multiply-add instructions [v3,35/60] target/riscv: vector floating-point square-root instruction [v3,36/60] target/riscv: vector floating-point min/max instructions [v3,37/60] target/riscv: vector floating-point sign-injection instructions [v3,38/60] target/riscv: vector floating-point compare instructions [v3,39/60] target/riscv: vector floating-point classify instructions [v3,40/60] target/riscv: vector floating-point merge instructions [v3,41/60] target/riscv: vector floating-point/integer type-convert instructions [v3,42/60] target/riscv: widening floating-point/integer type-convert instructions [v3,43/60] target/riscv: narrowing floating-point/integer type-convert instructions [v3,45/60] target/riscv: vector wideing integer reduction instructions [v3,48/60] target/riscv: vector mask-register logical instructions

diff --git a/target/riscv/helper.h b/target/riscv/helper.h index 70a4b05f75..e73701d4bb 100644 --- a/target/riscv/helper.h +++ b/target/riscv/helper.h @@ -269,3 +269,24 @@ DEF_HELPER_6(vamominw_v_w, void, ptr, ptr, tl, ptr, env, i32) DEF_HELPER_6(vamomaxw_v_w, void, ptr, ptr, tl, ptr, env, i32) DEF_HELPER_6(vamominuw_v_w, void, ptr, ptr, tl, ptr, env, i32) DEF_HELPER_6(vamomaxuw_v_w, void, ptr, ptr, tl, ptr, env, i32) + +DEF_HELPER_6(vadd_vv_b, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(vadd_vv_h, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(vadd_vv_w, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(vadd_vv_d, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(vsub_vv_b, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(vsub_vv_h, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(vsub_vv_w, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(vsub_vv_d, void, ptr, ptr, ptr, ptr, env, i32) +DEF_HELPER_6(vadd_vx_b, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(vadd_vx_h, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(vadd_vx_w, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(vadd_vx_d, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(vsub_vx_b, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(vsub_vx_h, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(vsub_vx_w, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(vsub_vx_d, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(vrsub_vx_b, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(vrsub_vx_h, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(vrsub_vx_w, void, ptr, ptr, tl, ptr, env, i32) +DEF_HELPER_6(vrsub_vx_d, void, ptr, ptr, tl, ptr, env, i32) diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode index 1330703720..d1034a0e61 100644 --- a/target/riscv/insn32.decode +++ b/target/riscv/insn32.decode @@ -44,6 +44,7 @@ &u imm rd &shift shamt rs1 rd &atomic aq rl rs2 rs1 rd +&rmrr vm rd rs1 rs2 &rwdvm vm wd rd rs1 rs2 &r2nfvm vm rd rs1 nf &rnfvm vm rd rs1 rs2 nf @@ -68,6 +69,7 @@ @r2 ....... ..... ..... ... ..... ....... %rs1 %rd @r2_nfvm ... ... vm:1 ..... ..... ... ..... ....... &r2nfvm %nf %rs1 %rd @r_nfvm ... ... vm:1 ..... ..... ... ..... ....... &rnfvm %nf %rs2 %rs1 %rd +@r_vm ...... vm:1 ..... ..... ... ..... ....... &rmrr %rs2 %rs1 %rd @r_wdvm ..... wd:1 vm:1 ..... ..... ... ..... ....... &rwdvm %rs2 %rs1 %rd @r2_zimm . zimm:11 ..... ... ..... ....... %rs1 %rd @@ -275,5 +277,13 @@ vamominuw_v 11000 . . ..... ..... 110 ..... 0101111 @r_wdvm vamomaxuw_v 11100 . . ..... ..... 110 ..... 0101111 @r_wdvm # *** new major opcode OP-V *** +vadd_vv 000000 . ..... ..... 000 ..... 1010111 @r_vm +vadd_vx 000000 . ..... ..... 100 ..... 1010111 @r_vm +vadd_vi 000000 . ..... ..... 011 ..... 1010111 @r_vm +vsub_vv 000010 . ..... ..... 000 ..... 1010111 @r_vm +vsub_vx 000010 . ..... ..... 100 ..... 1010111 @r_vm +vrsub_vx 000011 . ..... ..... 100 ..... 1010111 @r_vm +vrsub_vi 000011 . ..... ..... 011 ..... 1010111 @r_vm + vsetvli 0 ........... ..... 111 ..... 1010111 @r2_zimm vsetvl 1000000 ..... ..... 111 ..... 1010111 @r diff --git a/target/riscv/insn_trans/trans_rvv.inc.c b/target/riscv/insn_trans/trans_rvv.inc.c index 3c677160c5..00c7ec976f 100644 --- a/target/riscv/insn_trans/trans_rvv.inc.c +++ b/target/riscv/insn_trans/trans_rvv.inc.c @@ -730,3 +730,223 @@ GEN_VEXT_TRANS(vamomaxd_v, 15, rwdvm, amo_op, amo_check) GEN_VEXT_TRANS(vamominud_v, 16, rwdvm, amo_op, amo_check) GEN_VEXT_TRANS(vamomaxud_v, 17, rwdvm, amo_op, amo_check) #endif + +/* + *** Vector Integer Arithmetic Instructions + */ +#define MAXSZ(s) (s->vlen >> (3 - s->lmul)) + +static bool opivv_check(DisasContext *s, arg_rmrr *a) +{ + return (vext_check_isa_ill(s, RVV) && + vext_check_overlap_mask(s, a->rd, a->vm, false) && + vext_check_reg(s, a->rd, false) && + vext_check_reg(s, a->rs2, false) && + vext_check_reg(s, a->rs1, false)); +} + +/* OPIVV with GVEC IR */ +#define GEN_OPIVV_GVEC_TRANS(NAME, GVSUF) \ +static bool trans_##NAME(DisasContext *s, arg_rmrr *a) \ +{ \ + if (!opivv_check(s, a)) { \ + return false; \ + } \ + \ + if (a->vm && s->vl_eq_vlmax) { \ + tcg_gen_gvec_##GVSUF(8 << s->sew, vreg_ofs(s, a->rd), \ + vreg_ofs(s, a->rs2), vreg_ofs(s, a->rs1), \ + MAXSZ(s), MAXSZ(s)); \ + } else { \ + uint32_t data = 0; \ + static gen_helper_gvec_4_ptr * const fns[4] = { \ + gen_helper_##NAME##_b, gen_helper_##NAME##_h, \ + gen_helper_##NAME##_w, gen_helper_##NAME##_d, \ + }; \ + \ + data = FIELD_DP32(data, VDATA, MLEN, s->mlen); \ + data = FIELD_DP32(data, VDATA, VM, a->vm); \ + data = FIELD_DP32(data, VDATA, 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), \ + cpu_env, 0, s->vlen / 8, data, fns[s->sew]); \ + } \ + return true; \ +} +GEN_OPIVV_GVEC_TRANS(vadd_vv, add) +GEN_OPIVV_GVEC_TRANS(vsub_vv, sub) + +typedef void (*gen_helper_opivx)(TCGv_ptr, TCGv_ptr, TCGv, TCGv_ptr, + TCGv_env, TCGv_i32); + +static bool opivx_trans(uint32_t vd, uint32_t rs1, uint32_t vs2, + uint32_t data, gen_helper_opivx fn, DisasContext *s) +{ + TCGv_ptr dest, src2, mask; + TCGv src1; + TCGv_i32 desc; + + dest = tcg_temp_new_ptr(); + mask = tcg_temp_new_ptr(); + src2 = tcg_temp_new_ptr(); + src1 = tcg_temp_new(); + gen_get_gpr(src1, rs1); + desc = tcg_const_i32(simd_desc(0, s->vlen / 8, data)); + + tcg_gen_addi_ptr(dest, cpu_env, vreg_ofs(s, vd)); + tcg_gen_addi_ptr(src2, cpu_env, vreg_ofs(s, vs2)); + tcg_gen_addi_ptr(mask, cpu_env, vreg_ofs(s, 0)); + + fn(dest, mask, src1, src2, cpu_env, desc); + + tcg_temp_free_ptr(dest); + tcg_temp_free_ptr(mask); + tcg_temp_free_ptr(src2); + tcg_temp_free(src1); + tcg_temp_free_i32(desc); + return true; +} + +static bool opivx_check(DisasContext *s, arg_rmrr *a) +{ + return (vext_check_isa_ill(s, RVV) && + vext_check_overlap_mask(s, a->rd, a->vm, false) && + vext_check_reg(s, a->rd, false) && + vext_check_reg(s, a->rs2, false)); +} +/* OPIVX with GVEC IR */ +#define GEN_OPIVX_GVEC_TRANS(NAME, GVSUF) \ +static bool trans_##NAME(DisasContext *s, arg_rmrr *a) \ +{ \ + if (!opivx_check(s, a)) { \ + return false; \ + } \ + \ + if (a->vm && s->vl_eq_vlmax) { \ + TCGv_i64 src1 = tcg_temp_new_i64(); \ + TCGv tmp = tcg_temp_new(); \ + gen_get_gpr(tmp, a->rs1); \ + tcg_gen_ext_tl_i64(src1, tmp); \ + tcg_gen_gvec_##GVSUF(8 << s->sew, vreg_ofs(s, a->rd), \ + vreg_ofs(s, a->rs2), src1, MAXSZ(s), MAXSZ(s)); \ + tcg_temp_free_i64(src1); \ + tcg_temp_free(tmp); \ + return true; \ + } else { \ + uint32_t data = 0; \ + static gen_helper_opivx const fns[4] = { \ + gen_helper_##NAME##_b, gen_helper_##NAME##_h, \ + gen_helper_##NAME##_w, gen_helper_##NAME##_d, \ + }; \ + \ + data = FIELD_DP32(data, VDATA, MLEN, s->mlen); \ + data = FIELD_DP32(data, VDATA, VM, a->vm); \ + data = FIELD_DP32(data, VDATA, LMUL, s->lmul); \ + return opivx_trans(a->rd, a->rs1, a->rs2, data, fns[s->sew], s); \ + } \ + return true; \ +} +GEN_OPIVX_GVEC_TRANS(vadd_vx, adds) +GEN_OPIVX_GVEC_TRANS(vsub_vx, subs) + +/* OPIVX without GVEC IR */ +#define GEN_OPIVX_TRANS(NAME, CHECK) \ +static bool trans_##NAME(DisasContext *s, arg_rmrr *a) \ +{ \ + if (CHECK(s, a)) { \ + uint32_t data = 0; \ + static gen_helper_opivx const fns[4] = { \ + gen_helper_##NAME##_b, gen_helper_##NAME##_h, \ + gen_helper_##NAME##_w, gen_helper_##NAME##_d, \ + }; \ + \ + data = FIELD_DP32(data, VDATA, MLEN, s->mlen); \ + data = FIELD_DP32(data, VDATA, VM, a->vm); \ + data = FIELD_DP32(data, VDATA, LMUL, s->lmul); \ + return opivx_trans(a->rd, a->rs1, a->rs2, data, fns[s->sew], s); \ + } \ + return false; \ +} + +GEN_OPIVX_TRANS(vrsub_vx, opivx_check) + +static bool opivi_trans(uint32_t vd, uint32_t imm, uint32_t vs2, + uint32_t data, gen_helper_opivx fn, DisasContext *s, int zx) +{ + TCGv_ptr dest, src2, mask; + TCGv src1; + TCGv_i32 desc; + + dest = tcg_temp_new_ptr(); + mask = tcg_temp_new_ptr(); + src2 = tcg_temp_new_ptr(); + if (zx) { + src1 = tcg_const_tl(imm); + } else { + src1 = tcg_const_tl(sextract64(imm, 0, 5)); + } + desc = tcg_const_i32(simd_desc(0, s->vlen / 8, data)); + + tcg_gen_addi_ptr(dest, cpu_env, vreg_ofs(s, vd)); + tcg_gen_addi_ptr(src2, cpu_env, vreg_ofs(s, vs2)); + tcg_gen_addi_ptr(mask, cpu_env, vreg_ofs(s, 0)); + + fn(dest, mask, src1, src2, cpu_env, desc); + + tcg_temp_free_ptr(dest); + tcg_temp_free_ptr(mask); + tcg_temp_free_ptr(src2); + tcg_temp_free(src1); + tcg_temp_free_i32(desc); + return true; +} + +/* OPIVI with GVEC IR */ +#define GEN_OPIVI_GVEC_TRANS(NAME, ZX, OPIVX, GVSUF) \ +static bool trans_##NAME(DisasContext *s, arg_rmrr *a) \ +{ \ + if (!opivx_check(s, a)) { \ + return false; \ + } \ + \ + if (a->vm && s->vl_eq_vlmax) { \ + tcg_gen_gvec_##GVSUF(8 << s->sew, vreg_ofs(s, a->rd), \ + vreg_ofs(s, a->rs2), sextract64(a->rs1, 0, 5), \ + MAXSZ(s), MAXSZ(s)); \ + return true; \ + } else { \ + uint32_t data = 0; \ + static gen_helper_opivx const fns[4] = { \ + gen_helper_##OPIVX##_b, gen_helper_##OPIVX##_h, \ + gen_helper_##OPIVX##_w, gen_helper_##OPIVX##_d, \ + }; \ + \ + data = FIELD_DP32(data, VDATA, MLEN, s->mlen); \ + data = FIELD_DP32(data, VDATA, VM, a->vm); \ + data = FIELD_DP32(data, VDATA, LMUL, s->lmul); \ + return opivi_trans(a->rd, a->rs1, a->rs2, data, \ + fns[s->sew], s, ZX); \ + } \ + return true; \ +} +GEN_OPIVI_GVEC_TRANS(vadd_vi, 0, vadd_vx, addi) + +/* OPIVI without GVEC IR */ +#define GEN_OPIVI_TRANS(NAME, ZX, OPIVX, CHECK) \ +static bool trans_##NAME(DisasContext *s, arg_rmrr *a) \ +{ \ + if (CHECK(s, a)) { \ + uint32_t data = 0; \ + static gen_helper_opivx const fns[4] = { \ + gen_helper_##OPIVX##_b, gen_helper_##OPIVX##_h, \ + gen_helper_##OPIVX##_w, gen_helper_##OPIVX##_d, \ + }; \ + data = FIELD_DP32(data, VDATA, MLEN, s->mlen); \ + data = FIELD_DP32(data, VDATA, VM, a->vm); \ + data = FIELD_DP32(data, VDATA, LMUL, s->lmul); \ + return opivi_trans(a->rd, a->rs1, a->rs2, data, \ + fns[s->sew], s, ZX); \ + } \ + return false; \ +} +GEN_OPIVI_TRANS(vrsub_vi, 0, vrsub_vx, opivx_check) diff --git a/target/riscv/vector_helper.c b/target/riscv/vector_helper.c index f9b409b169..abdf3b82a8 100644 --- a/target/riscv/vector_helper.c +++ b/target/riscv/vector_helper.c @@ -828,3 +828,125 @@ GEN_VEXT_AMO(vamominw_v_w, int32_t, int32_t, idx_w, clearl) GEN_VEXT_AMO(vamomaxw_v_w, int32_t, int32_t, idx_w, clearl) GEN_VEXT_AMO(vamominuw_v_w, uint32_t, uint32_t, idx_w, clearl) GEN_VEXT_AMO(vamomaxuw_v_w, uint32_t, uint32_t, idx_w, clearl) + +/* + *** Vector Integer Arithmetic Instructions + */ + +/* expand macro args before macro */ +#define RVVCALL(macro, ...) macro(__VA_ARGS__) + +/* (TD, T1, T2, TX1, TX2) */ +#define OP_SSS_B int8_t, int8_t, int8_t, int8_t, int8_t +#define OP_SSS_H int16_t, int16_t, int16_t, int16_t, int16_t +#define OP_SSS_W int32_t, int32_t, int32_t, int32_t, int32_t +#define OP_SSS_D int64_t, int64_t, int64_t, int64_t, int64_t + +/* operation of two vector elements */ +#define OPIVV2(NAME, TD, T1, T2, TX1, TX2, HD, HS1, HS2, OP) \ +static void do_##NAME(void *vd, void *vs1, void *vs2, int i) \ +{ \ + TX1 s1 = *((T1 *)vs1 + HS1(i)); \ + TX2 s2 = *((T2 *)vs2 + HS2(i)); \ + *((TD *)vd + HD(i)) = OP(s2, s1); \ +} +#define DO_SUB(N, M) (N - M) +#define DO_RSUB(N, M) (M - N) + +RVVCALL(OPIVV2, vadd_vv_b, OP_SSS_B, H1, H1, H1, DO_ADD) +RVVCALL(OPIVV2, vadd_vv_h, OP_SSS_H, H2, H2, H2, DO_ADD) +RVVCALL(OPIVV2, vadd_vv_w, OP_SSS_W, H4, H4, H4, DO_ADD) +RVVCALL(OPIVV2, vadd_vv_d, OP_SSS_D, H8, H8, H8, DO_ADD) +RVVCALL(OPIVV2, vsub_vv_b, OP_SSS_B, H1, H1, H1, DO_SUB) +RVVCALL(OPIVV2, vsub_vv_h, OP_SSS_H, H2, H2, H2, DO_SUB) +RVVCALL(OPIVV2, vsub_vv_w, OP_SSS_W, H4, H4, H4, DO_SUB) +RVVCALL(OPIVV2, vsub_vv_d, OP_SSS_D, H8, H8, H8, DO_SUB) + +/* generate the helpers for OPIVV */ +#define GEN_VEXT_VV(NAME, ESZ, DSZ, CLEAR_FN) \ +void HELPER(NAME)(void *vd, void *v0, void *vs1, \ + void *vs2, CPURISCVState *env, uint32_t desc) \ +{ \ + uint32_t vlmax = vext_maxsz(desc) / ESZ; \ + uint32_t mlen = vext_mlen(desc); \ + uint32_t vm = vext_vm(desc); \ + uint32_t vl = env->vl; \ + uint32_t i; \ + for (i = 0; i < vl; i++) { \ + if (!vm && !vext_elem_mask(v0, mlen, i)) { \ + continue; \ + } \ + do_##NAME(vd, vs1, vs2, i); \ + } \ + if (i != 0) { \ + CLEAR_FN(vd, vl, vl * DSZ, vlmax * DSZ); \ + } \ +} + +GEN_VEXT_VV(vadd_vv_b, 1, 1, clearb) +GEN_VEXT_VV(vadd_vv_h, 2, 2, clearh) +GEN_VEXT_VV(vadd_vv_w, 4, 4, clearl) +GEN_VEXT_VV(vadd_vv_d, 8, 8, clearq) +GEN_VEXT_VV(vsub_vv_b, 1, 1, clearb) +GEN_VEXT_VV(vsub_vv_h, 2, 2, clearh) +GEN_VEXT_VV(vsub_vv_w, 4, 4, clearl) +GEN_VEXT_VV(vsub_vv_d, 8, 8, clearq) + +/* + * If XLEN < SEW, the value from the x register is sign-extended to SEW bits. + * So (target_long)s1 is need. (T1)(target_long)s1 gives the real operator type. + * (TX1)(T1)(target_long)s1 expands the operator type of widen operations + * or narrow operations + */ +#define OPIVX2(NAME, TD, T1, T2, TX1, TX2, HD, HS2, OP) \ +static void do_##NAME(void *vd, target_ulong s1, void *vs2, int i) \ +{ \ + TX2 s2 = *((T2 *)vs2 + HS2(i)); \ + *((TD *)vd + HD(i)) = OP(s2, (TX1)(T1)(target_long)s1); \ +} +RVVCALL(OPIVX2, vadd_vx_b, OP_SSS_B, H1, H1, DO_ADD) +RVVCALL(OPIVX2, vadd_vx_h, OP_SSS_H, H2, H2, DO_ADD) +RVVCALL(OPIVX2, vadd_vx_w, OP_SSS_W, H4, H4, DO_ADD) +RVVCALL(OPIVX2, vadd_vx_d, OP_SSS_D, H8, H8, DO_ADD) +RVVCALL(OPIVX2, vsub_vx_b, OP_SSS_B, H1, H1, DO_SUB) +RVVCALL(OPIVX2, vsub_vx_h, OP_SSS_H, H2, H2, DO_SUB) +RVVCALL(OPIVX2, vsub_vx_w, OP_SSS_W, H4, H4, DO_SUB) +RVVCALL(OPIVX2, vsub_vx_d, OP_SSS_D, H8, H8, DO_SUB) +RVVCALL(OPIVX2, vrsub_vx_b, OP_SSS_B, H1, H1, DO_RSUB) +RVVCALL(OPIVX2, vrsub_vx_h, OP_SSS_H, H2, H2, DO_RSUB) +RVVCALL(OPIVX2, vrsub_vx_w, OP_SSS_W, H4, H4, DO_RSUB) +RVVCALL(OPIVX2, vrsub_vx_d, OP_SSS_D, H8, H8, DO_RSUB) + +/* generate the helpers for instructions with one vector and one sclar */ +#define GEN_VEXT_VX(NAME, ESZ, DSZ, CLEAR_FN) \ +void HELPER(NAME)(void *vd, void *v0, target_ulong s1, \ + void *vs2, CPURISCVState *env, uint32_t desc) \ +{ \ + uint32_t vlmax = vext_maxsz(desc) / ESZ; \ + uint32_t mlen = vext_mlen(desc); \ + uint32_t vm = vext_vm(desc); \ + uint32_t vl = env->vl; \ + uint32_t i; \ + \ + for (i = 0; i < vl; i++) { \ + if (!vm && !vext_elem_mask(v0, mlen, i)) { \ + continue; \ + } \ + do_##NAME(vd, s1, vs2, i); \ + } \ + if (i != 0) { \ + CLEAR_FN(vd, vl, vl * DSZ, vlmax * DSZ); \ + } \ +} +GEN_VEXT_VX(vadd_vx_b, 1, 1, clearb) +GEN_VEXT_VX(vadd_vx_h, 2, 2, clearh) +GEN_VEXT_VX(vadd_vx_w, 4, 4, clearl) +GEN_VEXT_VX(vadd_vx_d, 8, 8, clearq) +GEN_VEXT_VX(vsub_vx_b, 1, 1, clearb) +GEN_VEXT_VX(vsub_vx_h, 2, 2, clearh) +GEN_VEXT_VX(vsub_vx_w, 4, 4, clearl) +GEN_VEXT_VX(vsub_vx_d, 8, 8, clearq) +GEN_VEXT_VX(vrsub_vx_b, 1, 1, clearb) +GEN_VEXT_VX(vrsub_vx_h, 2, 2, clearh) +GEN_VEXT_VX(vrsub_vx_w, 4, 4, clearl) +GEN_VEXT_VX(vrsub_vx_d, 8, 8, clearq)

[v3,09/60] target/riscv: vector single-width integer add and subtract

Commit Message

Patch