From nobody Mon Feb 9 15:37:04 2026 Delivered-To: importer@patchew.org Received-SPF: pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) client-ip=208.118.235.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Authentication-Results: mx.zohomail.com; dkim=fail; spf=pass (zoho.com: domain of gnu.org designates 208.118.235.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=fail(p=none dis=none) header.from=linaro.org Return-Path: Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) by mx.zohomail.com with SMTPS id 1518893618954296.94788864331906; Sat, 17 Feb 2018 10:53:38 -0800 (PST) Received: from localhost ([::1]:48300 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1en7cA-0007uW-6j for importer@patchew.org; Sat, 17 Feb 2018 13:53:38 -0500 Received: from eggs.gnu.org ([2001:4830:134:3::10]:40111) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1en79e-0000la-9F for qemu-devel@nongnu.org; Sat, 17 Feb 2018 13:24:12 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1en79c-0001m2-Cn for qemu-devel@nongnu.org; Sat, 17 Feb 2018 13:24:10 -0500 Received: from mail-pl0-x244.google.com ([2607:f8b0:400e:c01::244]:44554) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1en79c-0001ld-4z for qemu-devel@nongnu.org; Sat, 17 Feb 2018 13:24:08 -0500 Received: by mail-pl0-x244.google.com with SMTP id w21so3426672plp.11 for ; Sat, 17 Feb 2018 10:24:08 -0800 (PST) Received: from cloudburst.twiddle.net ([50.0.192.64]) by smtp.gmail.com with ESMTPSA id h15sm13466712pfi.56.2018.02.17.10.24.05 (version=TLS1_2 cipher=ECDHE-RSA-CHACHA20-POLY1305 bits=256/256); Sat, 17 Feb 2018 10:24:05 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=GpiGhGCBLCpMu5HsxIdaTUuHdNgL0asf15r6AUQljAo=; b=DvkOv3ISU3xNKZSB91aWS50R8HNAYKHeuoVbRx0BFkQpMtYeUy5W3licpKHAexmX8b KkOj+w7TQBgDYH2Kz0hbvG/IdUS16gRdo8SmVXTjBE90cmAk2hdE+DtvY8a0HZWtFqD1 hP/Fh3xMPGWBVM6MtRHOJPKCbbz2VrEBSiaZE= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=GpiGhGCBLCpMu5HsxIdaTUuHdNgL0asf15r6AUQljAo=; b=WdNZEOcMYcHLJbWd129XejzZ1/GNUL6EXBvq7MtsCT+Gng86z60ctq/Att/YxA3DWv p+U1z+V1IiFrocbugqUJsDX9amLQdm+XzEz2dJR5auZVxeBzu2tIBOgeprpddQKDvZuO M3Kgx2wem0gaDjokWRxAOzFQdDtqtcrFFNlCVRpujsct2Z1SLU9q0vpW/79yBIfICEE/ TDQJ0wQte6gQADzWdBVFlw2wFrXH1ocOUbfp59BReYBmMVnzp+0AaiZUZxp/NrWOa1E8 Im8G61mj2jgk7KnnrASbn6dWm3Hs9YigSZHWjr9L63LCAbRD7gm2BEr+HDgdAIwHP+oi 34TQ== X-Gm-Message-State: APf1xPDo66ybsU2LsaOlcLpYDw6Usc41sRZgK5q6SDwhHkh42RFBBZ4a vlk8YjX9Mfa77NtCk9+LnlmVXQsBkrw= X-Google-Smtp-Source: AH8x226ZlKmyrAwKDVy9lOzic56OFhvYE9pIj/LMrW/f6Xwq+sHJbeQfipbm37jxIRum4pTIxOasQA== X-Received: by 2002:a17:902:7808:: with SMTP id p8-v6mr9620755pll.161.1518891846870; Sat, 17 Feb 2018 10:24:06 -0800 (PST) From: Richard Henderson To: qemu-devel@nongnu.org Date: Sat, 17 Feb 2018 10:22:41 -0800 Message-Id: <20180217182323.25885-26-richard.henderson@linaro.org> X-Mailer: git-send-email 2.14.3 In-Reply-To: <20180217182323.25885-1-richard.henderson@linaro.org> References: <20180217182323.25885-1-richard.henderson@linaro.org> X-detected-operating-system: by eggs.gnu.org: Genre and OS details not recognized. X-Received-From: 2607:f8b0:400e:c01::244 Subject: [Qemu-devel] [PATCH v2 25/67] target/arm: Implement SVE Integer Wide Immediate - Predicated Group X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: qemu-arm@nongnu.org Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: fail (Header signature does not verify) X-ZohoMail: RDKM_2 RSF_0 Z_629925259 SPT_0 Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Signed-off-by: Richard Henderson Reviewed-by: Peter Maydell --- target/arm/helper-sve.h | 10 +++++ target/arm/sve_helper.c | 108 +++++++++++++++++++++++++++++++++++++++++= ++++ target/arm/translate-sve.c | 92 ++++++++++++++++++++++++++++++++++++++ target/arm/sve.decode | 17 +++++++ 4 files changed, 227 insertions(+) diff --git a/target/arm/helper-sve.h b/target/arm/helper-sve.h index 2831e1643b..79493ab647 100644 --- a/target/arm/helper-sve.h +++ b/target/arm/helper-sve.h @@ -404,6 +404,16 @@ DEF_HELPER_FLAGS_4(sve_uqaddi_s, TCG_CALL_NO_RWG, void= , ptr, ptr, s64, i32) DEF_HELPER_FLAGS_4(sve_uqaddi_d, TCG_CALL_NO_RWG, void, ptr, ptr, i64, i32) DEF_HELPER_FLAGS_4(sve_uqsubi_d, TCG_CALL_NO_RWG, void, ptr, ptr, i64, i32) =20 +DEF_HELPER_FLAGS_5(sve_cpy_m_b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i64,= i32) +DEF_HELPER_FLAGS_5(sve_cpy_m_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i64,= i32) +DEF_HELPER_FLAGS_5(sve_cpy_m_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i64,= i32) +DEF_HELPER_FLAGS_5(sve_cpy_m_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i64,= i32) + +DEF_HELPER_FLAGS_4(sve_cpy_z_b, TCG_CALL_NO_RWG, void, ptr, ptr, i64, i32) +DEF_HELPER_FLAGS_4(sve_cpy_z_h, TCG_CALL_NO_RWG, void, ptr, ptr, i64, i32) +DEF_HELPER_FLAGS_4(sve_cpy_z_s, TCG_CALL_NO_RWG, void, ptr, ptr, i64, i32) +DEF_HELPER_FLAGS_4(sve_cpy_z_d, TCG_CALL_NO_RWG, void, ptr, ptr, i64, i32) + DEF_HELPER_FLAGS_5(sve_and_pppp, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr= , i32) DEF_HELPER_FLAGS_5(sve_bic_pppp, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr= , i32) DEF_HELPER_FLAGS_5(sve_eor_pppp, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr= , i32) diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index cfda16d520..6a95d1ec48 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -1361,3 +1361,111 @@ void HELPER(sve_uqsubi_d)(void *d, void *a, uint64_= t b, uint32_t desc) *(uint64_t *)(d + i) =3D (ai < b ? 0 : ai - b); } } + +/* Two operand predicated copy immediate with merge. All valid immediates + * can fit within 17 signed bits in the simd_data field. + */ +void HELPER(sve_cpy_m_b)(void *vd, void *vn, void *vg, + uint64_t mm, uint32_t desc) +{ + intptr_t i, opr_sz =3D simd_oprsz(desc) / 8; + uint64_t *d =3D vd, *n =3D vn; + uint8_t *pg =3D vg; + + mm =3D (mm & 0xff) * (-1ull / 0xff); + for (i =3D 0; i < opr_sz; i +=3D 1) { + uint64_t nn =3D n[i]; + uint64_t pp =3D expand_pred_b(pg[H1(i)]); + d[i] =3D (mm & pp) | (nn & ~pp); + } +} + +void HELPER(sve_cpy_m_h)(void *vd, void *vn, void *vg, + uint64_t mm, uint32_t desc) +{ + intptr_t i, opr_sz =3D simd_oprsz(desc) / 8; + uint64_t *d =3D vd, *n =3D vn; + uint8_t *pg =3D vg; + + mm =3D (mm & 0xffff) * (-1ull / 0xffff); + for (i =3D 0; i < opr_sz; i +=3D 1) { + uint64_t nn =3D n[i]; + uint64_t pp =3D expand_pred_h(pg[H1(i)]); + d[i] =3D (mm & pp) | (nn & ~pp); + } +} + +void HELPER(sve_cpy_m_s)(void *vd, void *vn, void *vg, + uint64_t mm, uint32_t desc) +{ + intptr_t i, opr_sz =3D simd_oprsz(desc) / 8; + uint64_t *d =3D vd, *n =3D vn; + uint8_t *pg =3D vg; + + mm =3D deposit64(mm, 32, 32, mm); + for (i =3D 0; i < opr_sz; i +=3D 1) { + uint64_t nn =3D n[i]; + uint64_t pp =3D expand_pred_s(pg[H1(i)]); + d[i] =3D (mm & pp) | (nn & ~pp); + } +} + +void HELPER(sve_cpy_m_d)(void *vd, void *vn, void *vg, + uint64_t mm, uint32_t desc) +{ + intptr_t i, opr_sz =3D simd_oprsz(desc) / 8; + uint64_t *d =3D vd, *n =3D vn; + uint8_t *pg =3D vg; + + for (i =3D 0; i < opr_sz; i +=3D 1) { + uint64_t nn =3D n[i]; + d[i] =3D (pg[H1(i)] & 1 ? mm : nn); + } +} + +void HELPER(sve_cpy_z_b)(void *vd, void *vg, uint64_t val, uint32_t desc) +{ + intptr_t i, opr_sz =3D simd_oprsz(desc) / 8; + uint64_t *d =3D vd; + uint8_t *pg =3D vg; + + val =3D (val & 0xff) * (-1ull / 0xff); + for (i =3D 0; i < opr_sz; i +=3D 1) { + d[i] =3D val & expand_pred_b(pg[H1(i)]); + } +} + +void HELPER(sve_cpy_z_h)(void *vd, void *vg, uint64_t val, uint32_t desc) +{ + intptr_t i, opr_sz =3D simd_oprsz(desc) / 8; + uint64_t *d =3D vd; + uint8_t *pg =3D vg; + + val =3D (val & 0xffff) * (-1ull / 0xffff); + for (i =3D 0; i < opr_sz; i +=3D 1) { + d[i] =3D val & expand_pred_h(pg[H1(i)]); + } +} + +void HELPER(sve_cpy_z_s)(void *vd, void *vg, uint64_t val, uint32_t desc) +{ + intptr_t i, opr_sz =3D simd_oprsz(desc) / 8; + uint64_t *d =3D vd; + uint8_t *pg =3D vg; + + val =3D deposit64(val, 32, 32, val); + for (i =3D 0; i < opr_sz; i +=3D 1) { + d[i] =3D val & expand_pred_s(pg[H1(i)]); + } +} + +void HELPER(sve_cpy_z_d)(void *vd, void *vg, uint64_t val, uint32_t desc) +{ + intptr_t i, opr_sz =3D simd_oprsz(desc) / 8; + uint64_t *d =3D vd; + uint8_t *pg =3D vg; + + for (i =3D 0; i < opr_sz; i +=3D 1) { + d[i] =3D (pg[H1(i)] & 1 ? val : 0); + } +} diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index 21b1e4df85..dd085b084b 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -68,6 +68,12 @@ static inline int plus1(int x) return x + 1; } =20 +/* The SH bit is in bit 8. Extract the low 8 and shift. */ +static inline int expand_imm_sh8s(int x) +{ + return (int8_t)x << (x & 0x100 ? 8 : 0); +} + /* * Include the generated decoder. */ @@ -1698,6 +1704,92 @@ static void trans_DUPM(DisasContext *s, arg_DUPM *a,= uint32_t insn) do_dupi_z(s, a->rd, imm); } =20 +/* + *** SVE Integer Wide Immediate - Predicated Group + */ + +/* Implement all merging copies. This is used for CPY (immediate), + * FCPY, CPY (scalar), CPY (SIMD&FP scalar). + */ +static void do_cpy_m(DisasContext *s, int esz, int rd, int rn, int pg, + TCGv_i64 val) +{ + typedef void gen_cpy(TCGv_ptr, TCGv_ptr, TCGv_ptr, TCGv_i64, TCGv_i32); + static gen_cpy * const fns[4] =3D { + gen_helper_sve_cpy_m_b, gen_helper_sve_cpy_m_h, + gen_helper_sve_cpy_m_s, gen_helper_sve_cpy_m_d, + }; + unsigned vsz =3D vec_full_reg_size(s); + TCGv_i32 desc =3D tcg_const_i32(simd_desc(vsz, vsz, 0)); + TCGv_ptr t_zd =3D tcg_temp_new_ptr(); + TCGv_ptr t_zn =3D tcg_temp_new_ptr(); + TCGv_ptr t_pg =3D tcg_temp_new_ptr(); + + tcg_gen_addi_ptr(t_zd, cpu_env, vec_full_reg_offset(s, rd)); + tcg_gen_addi_ptr(t_zn, cpu_env, vec_full_reg_offset(s, rn)); + tcg_gen_addi_ptr(t_pg, cpu_env, pred_full_reg_offset(s, pg)); + + fns[esz](t_zd, t_zn, t_pg, val, desc); + + tcg_temp_free_ptr(t_zd); + tcg_temp_free_ptr(t_zn); + tcg_temp_free_ptr(t_pg); + tcg_temp_free_i32(desc); +} + +static void trans_FCPY(DisasContext *s, arg_FCPY *a, uint32_t insn) +{ + uint64_t imm; + TCGv_i64 t_imm; + + if (a->esz =3D=3D 0) { + unallocated_encoding(s); + return; + } + + /* Decode the VFP immediate. */ + imm =3D vfp_expand_imm(a->esz, a->imm); + + t_imm =3D tcg_const_i64(imm); + do_cpy_m(s, a->esz, a->rd, a->rn, a->pg, t_imm); + tcg_temp_free_i64(t_imm); +} + +static void trans_CPY_m_i(DisasContext *s, arg_rpri_esz *a, uint32_t insn) +{ + TCGv_i64 t_imm; + + if (a->esz =3D=3D 0 && extract32(insn, 13, 1)) { + unallocated_encoding(s); + return; + } + + t_imm =3D tcg_const_i64(a->imm); + do_cpy_m(s, a->esz, a->rd, a->rn, a->pg, t_imm); + tcg_temp_free_i64(t_imm); +} + +static void trans_CPY_z_i(DisasContext *s, arg_CPY_z_i *a, uint32_t insn) +{ + static gen_helper_gvec_2i * const fns[4] =3D { + gen_helper_sve_cpy_z_b, gen_helper_sve_cpy_z_h, + gen_helper_sve_cpy_z_s, gen_helper_sve_cpy_z_d, + }; + unsigned vsz =3D vec_full_reg_size(s); + TCGv_i64 t_imm; + + if (a->esz =3D=3D 0 && extract32(insn, 13, 1)) { + unallocated_encoding(s); + return; + } + + t_imm =3D tcg_const_i64(a->imm); + tcg_gen_gvec_2i_ool(vec_full_reg_offset(s, a->rd), + pred_full_reg_offset(s, a->pg), + t_imm, vsz, vsz, 0, fns[a->esz]); + tcg_temp_free_i64(t_imm); +} + /* *** SVE Memory - 32-bit Gather and Unsized Contiguous Group */ diff --git a/target/arm/sve.decode b/target/arm/sve.decode index 0990d135f4..e6e10a4f84 100644 --- a/target/arm/sve.decode +++ b/target/arm/sve.decode @@ -39,6 +39,9 @@ %tszimm16_shr 22:2 16:5 !function=3Dtszimm_shr %tszimm16_shl 22:2 16:5 !function=3Dtszimm_shl =20 +# Signed 8-bit immediate, optionally shifted left by 8. +%sh8_i8s 5:9 !function=3Dexpand_imm_sh8s + # Either a copy of rd (at bit 0), or a different source # as propagated via the MOVPRFX instruction. %reg_movprfx 0:5 @@ -113,6 +116,11 @@ @rd_rn_tszimm ........ .. ... ... ...... rn:5 rd:5 \ &rri_esz esz=3D%tszimm16_esz =20 +# Two register operand, one immediate operand, with 4-bit predicate. +# User must fill in imm. +@rdn_pg4 ........ esz:2 .. pg:4 ... ........ rd:5 \ + &rpri_esz rn=3D%reg_movprfx + # Two register operand, one encoded bitmask. @rdn_dbm ........ .. .... dbm:13 rd:5 \ &rr_dbm rn=3D%reg_movprfx @@ -346,6 +354,15 @@ AND_zzi 00000101 10 0000 ............. ..... @rdn_dbm # SVE broadcast bitmask immediate DUPM 00000101 11 0000 dbm:13 rd:5 =20 +### SVE Integer Wide Immediate - Predicated Group + +# SVE copy floating-point immediate (predicated) +FCPY 00000101 .. 01 .... 110 imm:8 ..... @rdn_pg4 + +# SVE copy integer immediate (predicated) +CPY_m_i 00000101 .. 01 .... 01 . ........ ..... @rdn_pg4 imm=3D%sh8_i8s +CPY_z_i 00000101 .. 01 .... 00 . ........ ..... @rdn_pg4 imm=3D%sh8_i8s + ### SVE Predicate Logical Operations Group =20 # SVE predicate logical operations --=20 2.14.3