From eab84139855dac258c8d89ad736f6649e3edc76a Mon Sep 17 00:00:00 2001 From: Peter Maydell Date: Mon, 28 Jun 2021 14:58:23 +0100 Subject: target/arm: Implement MVE logical immediate insns Implement the MVE logical-immediate insns (VMOV, VMVN, VORR and VBIC). These have essentially the same encoding as their Neon equivalents, and we implement the decode in the same way. Signed-off-by: Peter Maydell Reviewed-by: Richard Henderson Message-id: 20210628135835.6690-7-peter.maydell@linaro.org --- target/arm/helper-mve.h | 4 ++++ target/arm/mve.decode | 17 ++++++++++++++++ target/arm/mve_helper.c | 24 ++++++++++++++++++++++ target/arm/translate-mve.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 95 insertions(+) diff --git a/target/arm/helper-mve.h b/target/arm/helper-mve.h index 4bbb9b3..5248dbe 100644 --- a/target/arm/helper-mve.h +++ b/target/arm/helper-mve.h @@ -355,3 +355,7 @@ DEF_HELPER_FLAGS_3(mve_vaddvsh, TCG_CALL_NO_WG, i32, env, ptr, i32) DEF_HELPER_FLAGS_3(mve_vaddvuh, TCG_CALL_NO_WG, i32, env, ptr, i32) DEF_HELPER_FLAGS_3(mve_vaddvsw, TCG_CALL_NO_WG, i32, env, ptr, i32) DEF_HELPER_FLAGS_3(mve_vaddvuw, TCG_CALL_NO_WG, i32, env, ptr, i32) + +DEF_HELPER_FLAGS_3(mve_vmovi, TCG_CALL_NO_WG, void, env, ptr, i64) +DEF_HELPER_FLAGS_3(mve_vandi, TCG_CALL_NO_WG, void, env, ptr, i64) +DEF_HELPER_FLAGS_3(mve_vorri, TCG_CALL_NO_WG, void, env, ptr, i64) diff --git a/target/arm/mve.decode b/target/arm/mve.decode index d9ece7b..caeb016 100644 --- a/target/arm/mve.decode +++ b/target/arm/mve.decode @@ -26,10 +26,14 @@ # VQDMULL has size in bit 28: 0 for 16 bit, 1 for 32 bit %size_28 28:1 !function=plus_1 +# 1imm format immediate +%imm_28_16_0 28:1 16:3 0:4 + &vldr_vstr rn qd imm p a w size l u &1op qd qm size &2op qd qm qn size &2scalar qd qn rm size +&1imm qd imm cmode op @vldr_vstr ....... . . . . l:1 rn:4 ... ...... imm:7 &vldr_vstr qd=%qd u=0 # Note that both Rn and Qd are 3 bits only (no D bit) @@ -41,6 +45,7 @@ @2op_nosz .... .... .... .... .... .... .... .... &2op qd=%qd qm=%qm qn=%qn size=0 @2op_sz28 .... .... .... .... .... .... .... .... &2op qd=%qd qm=%qm qn=%qn \ size=%size_28 +@1imm .... .... .... .... .... cmode:4 .. op:1 . .... &1imm qd=%qd imm=%imm_28_16_0 # The _rev suffix indicates that Vn and Vm are reversed. This is # the case for shifts. In the Arm ARM these insns are documented @@ -258,3 +263,15 @@ VADDV 111 u:1 1110 1111 size:2 01 ... 0 1111 0 0 a:1 0 qm:3 0 rda=%rd # Predicate operations %mask_22_13 22:1 13:3 VPST 1111 1110 0 . 11 000 1 ... 0 1111 0100 1101 mask=%mask_22_13 + +# Logical immediate operations (1 reg and modified-immediate) + +# The cmode/op bits here decode VORR/VBIC/VMOV/VMVN, but +# not in a way we can conveniently represent in decodetree without +# a lot of repetition: +# VORR: op=0, (cmode & 1) && cmode < 12 +# VBIC: op=1, (cmode & 1) && cmode < 12 +# VMOV: everything else +# So we have a single decode line and check the cmode/op in the +# trans function. +Vimm_1r 111 . 1111 1 . 00 0 ... ... 0 .... 0 1 . 1 .... @1imm diff --git a/target/arm/mve_helper.c b/target/arm/mve_helper.c index 85a552f..e6ced14 100644 --- a/target/arm/mve_helper.c +++ b/target/arm/mve_helper.c @@ -323,6 +323,30 @@ DO_1OP(vnegw, 4, int32_t, DO_NEG) DO_1OP(vfnegh, 8, uint64_t, DO_FNEGH) DO_1OP(vfnegs, 8, uint64_t, DO_FNEGS) +/* + * 1 operand immediates: Vda is destination and possibly also one source. + * All these insns work at 64-bit widths. + */ +#define DO_1OP_IMM(OP, FN) \ + void HELPER(mve_##OP)(CPUARMState *env, void *vda, uint64_t imm) \ + { \ + uint64_t *da = vda; \ + uint16_t mask = mve_element_mask(env); \ + unsigned e; \ + for (e = 0; e < 16 / 8; e++, mask >>= 8) { \ + mergemask(&da[H8(e)], FN(da[H8(e)], imm), mask); \ + } \ + mve_advance_vpt(env); \ + } + +#define DO_MOVI(N, I) (I) +#define DO_ANDI(N, I) ((N) & (I)) +#define DO_ORRI(N, I) ((N) | (I)) + +DO_1OP_IMM(vmovi, DO_MOVI) +DO_1OP_IMM(vandi, DO_ANDI) +DO_1OP_IMM(vorri, DO_ORRI) + #define DO_2OP(OP, ESIZE, TYPE, FN) \ void HELPER(glue(mve_, OP))(CPUARMState *env, \ void *vd, void *vn, void *vm) \ diff --git a/target/arm/translate-mve.c b/target/arm/translate-mve.c index e9a5442..f435a1c 100644 --- a/target/arm/translate-mve.c +++ b/target/arm/translate-mve.c @@ -34,6 +34,7 @@ typedef void MVEGenTwoOpFn(TCGv_ptr, TCGv_ptr, TCGv_ptr, TCGv_ptr); typedef void MVEGenTwoOpScalarFn(TCGv_ptr, TCGv_ptr, TCGv_ptr, TCGv_i32); typedef void MVEGenDualAccOpFn(TCGv_i64, TCGv_ptr, TCGv_ptr, TCGv_ptr, TCGv_i64); typedef void MVEGenVADDVFn(TCGv_i32, TCGv_ptr, TCGv_ptr, TCGv_i32); +typedef void MVEGenOneOpImmFn(TCGv_ptr, TCGv_ptr, TCGv_i64); /* Return the offset of a Qn register (same semantics as aa32_vfp_qreg()) */ static inline long mve_qreg_offset(unsigned reg) @@ -787,3 +788,52 @@ static bool trans_VADDV(DisasContext *s, arg_VADDV *a) mve_update_eci(s); return true; } + +static bool do_1imm(DisasContext *s, arg_1imm *a, MVEGenOneOpImmFn *fn) +{ + TCGv_ptr qd; + uint64_t imm; + + if (!dc_isar_feature(aa32_mve, s) || + !mve_check_qreg_bank(s, a->qd) || + !fn) { + return false; + } + if (!mve_eci_check(s) || !vfp_access_check(s)) { + return true; + } + + imm = asimd_imm_const(a->imm, a->cmode, a->op); + + qd = mve_qreg_ptr(a->qd); + fn(cpu_env, qd, tcg_constant_i64(imm)); + tcg_temp_free_ptr(qd); + mve_update_eci(s); + return true; +} + +static bool trans_Vimm_1r(DisasContext *s, arg_1imm *a) +{ + /* Handle decode of cmode/op here between VORR/VBIC/VMOV */ + MVEGenOneOpImmFn *fn; + + if ((a->cmode & 1) && a->cmode < 12) { + if (a->op) { + /* + * For op=1, the immediate will be inverted by asimd_imm_const(), + * so the VBIC becomes a logical AND operation. + */ + fn = gen_helper_mve_vandi; + } else { + fn = gen_helper_mve_vorri; + } + } else { + /* There is one unallocated cmode/op combination in this space */ + if (a->cmode == 15 && a->op == 1) { + return false; + } + /* asimd_imm_const() sorts out VMVNI vs VMOVI for us */ + fn = gen_helper_mve_vmovi; + } + return do_1imm(s, a, fn); +} -- cgit v1.1