From nobody Tue Feb 10 20:30:27 2026 Delivered-To: importer@patchew.org Authentication-Results: mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass(p=none dis=none) header.from=linaro.org ARC-Seal: i=1; a=rsa-sha256; t=1589388332; cv=none; d=zohomail.com; s=zohoarc; b=EyQEGcC1H+0wboDBLd0YJGkQzzQ7Xzmsx+3vD/UcPLdj+LFmFwQU9PKoHNgcJ0NO23rGxBoYR58nlyLGggkbieMHELcwy70ij3PUF+NFqLYiy6yugY8TRlK6iSWHHfb9ZpOpUSclr1jpz+b8En1Dty/K045terXDzBKoGAlKcQ0= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1589388332; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=sMsSJIbayaVcN7fAKLPcCoZeyuYV2bazR+WjJ8tzbTs=; b=fdUmzTupEX5sHYxSdgkTJI3sBjV24v5fsYdKjap+cMyjOAl+7ZNJBkGO4sxp3vU3S6lIfy5OVxi/dUB0CM9WsVez5ryFsw2gh/dNBBLmAfcXp5pIiC25Q6Y8wb2gEkCBhV9anlhzZ1yuBZJLq6faPmiIcf28zSCCDNcYaE+cSa4= ARC-Authentication-Results: i=1; mx.zohomail.com; dkim=pass; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=pass header.from= (p=none dis=none) header.from= Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1589388332877289.0430165451829; Wed, 13 May 2020 09:45:32 -0700 (PDT) Received: from localhost ([::1]:33744 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jYuVf-0004uG-HT for importer@patchew.org; Wed, 13 May 2020 12:45:31 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:48052) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1jYuJj-0003nf-Pf for qemu-devel@nongnu.org; Wed, 13 May 2020 12:33:11 -0400 Received: from mail-pl1-x641.google.com ([2607:f8b0:4864:20::641]:42066) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1jYuJh-00023s-IS for qemu-devel@nongnu.org; Wed, 13 May 2020 12:33:11 -0400 Received: by mail-pl1-x641.google.com with SMTP id k19so38675pll.9 for ; Wed, 13 May 2020 09:33:08 -0700 (PDT) Received: from localhost.localdomain (174-21-143-238.tukw.qwest.net. [174.21.143.238]) by smtp.gmail.com with ESMTPSA id b11sm158025pgq.50.2020.05.13.09.33.05 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 13 May 2020 09:33:06 -0700 (PDT) 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 :mime-version:content-transfer-encoding; bh=sMsSJIbayaVcN7fAKLPcCoZeyuYV2bazR+WjJ8tzbTs=; b=fM4EGvFTSPXqTYDA3eCrVoULRLcsBmO9xM4iM9/ZWDPOqk+mcs9WqtH0vX5LljQ9pq fi7LHt8yirMEsi9kVwcG6L626Pba4kSoBUIVBOJG9YTgSAgt/Xb+qJahrKT6ONBTW8qF l22YvLHH2bvFDeYxPGgDsxc8mRfioVtUxxv04xDCyyAwXfFS/6rFlNSKEylLrQ4hc1i4 WBgifZFXt99bB5WCHz+1813PXALfz8NPyLde5bs4un3cbVRPYAhLD6WU4fLDX4Xren/5 sLpO8Rqb7Lk80syXqy+eQpSbV9M4SzlhvvS4NtNheyCXb0eIPrqhuYBxOK/hkbo4HFwU b5Aw== 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:mime-version:content-transfer-encoding; bh=sMsSJIbayaVcN7fAKLPcCoZeyuYV2bazR+WjJ8tzbTs=; b=dbFj/tSbvXpE+izaxbqCUeWw5EEzxrimLWC6fFSWtmzlN6pyKARQOuszqFRd73FQEK DlcuCp28VSCztV5bpOC8PFXOKwX/jZnBl1dpDPRciZYRYx3ZTUZgJoBq4Mdt3jKPkxLK tpAUx/t3vzVx5549zixF8Hr4tFI41CmFKnKUmgRJUiwF2FqfXkNa+zRZo31eV81MTUF1 VFOL8xkibqCX/cZQN5m8nQDs4dTJu9FGYV2KJoIeVNy3BCbveRyb62ufqd0lTjSxpKvx JLYeQMjDZ3nvBadN8oSPtLblIEKgeikc13B+nk3v1AL9QxbSVhSVLOxMRyYDTSYuQFL4 vYbA== X-Gm-Message-State: AGi0PuYjBAQaRoNXMNcjE20E1tDU7MjDJGhQlJJK+oZf6ex98Jk/f6pg jlko3sYYvCHoa+Rb1oLxGNbyCHDrXtY= X-Google-Smtp-Source: APiQypJcgsfyNPYt2rH8tM4Sb3rfBU2T3j1mmUS5sp0l9EompnFhTJpMNkrE9g+BECOoDx8eoH+mIQ== X-Received: by 2002:a17:90a:930c:: with SMTP id p12mr38892146pjo.64.1589387586951; Wed, 13 May 2020 09:33:06 -0700 (PDT) From: Richard Henderson To: qemu-devel@nongnu.org Subject: [PATCH v4 15/16] target/arm: Vectorize SABD/UABD Date: Wed, 13 May 2020 09:32:44 -0700 Message-Id: <20200513163245.17915-16-richard.henderson@linaro.org> X-Mailer: git-send-email 2.20.1 In-Reply-To: <20200513163245.17915-1-richard.henderson@linaro.org> References: <20200513163245.17915-1-richard.henderson@linaro.org> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Received-SPF: pass client-ip=2607:f8b0:4864:20::641; envelope-from=richard.henderson@linaro.org; helo=mail-pl1-x641.google.com X-detected-operating-system: by eggs.gnu.org: No matching host in p0f cache. That's all we know. X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001 autolearn=_AUTOLEARN X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: peter.maydell@linaro.org Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @linaro.org) Content-Type: text/plain; charset="utf-8" Include 64-bit element size in preparation for SVE2. Reviewed-by: Peter Maydell Signed-off-by: Richard Henderson --- target/arm/helper.h | 10 +++ target/arm/translate.h | 5 ++ target/arm/translate-a64.c | 8 ++- target/arm/translate.c | 133 ++++++++++++++++++++++++++++++++++++- target/arm/vec_helper.c | 24 +++++++ 5 files changed, 176 insertions(+), 4 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index aed3050965..4678d3a6f4 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -731,6 +731,16 @@ DEF_HELPER_FLAGS_3(gvec_sli_h, TCG_CALL_NO_RWG, void, = ptr, ptr, i32) DEF_HELPER_FLAGS_3(gvec_sli_s, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_3(gvec_sli_d, TCG_CALL_NO_RWG, void, ptr, ptr, i32) =20 +DEF_HELPER_FLAGS_4(gvec_sabd_b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_sabd_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_sabd_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_sabd_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) + +DEF_HELPER_FLAGS_4(gvec_uabd_b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_uabd_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_uabd_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(gvec_uabd_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) + #ifdef TARGET_AARCH64 #include "helper-a64.h" #include "helper-sve.h" diff --git a/target/arm/translate.h b/target/arm/translate.h index aea8a9759d..bbfe3d1393 100644 --- a/target/arm/translate.h +++ b/target/arm/translate.h @@ -337,6 +337,11 @@ void gen_gvec_sqrdmlah_qc(unsigned vece, uint32_t rd_o= fs, uint32_t rn_ofs, void gen_gvec_sqrdmlsh_qc(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_s= z); =20 +void gen_gvec_sabd(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); +void gen_gvec_uabd(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); + /* * Forward to the isar_feature_* tests given a DisasContext pointer. */ diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index 4577df3cf4..54b06553a6 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -11190,6 +11190,13 @@ static void disas_simd_3same_int(DisasContext *s, = uint32_t insn) gen_gvec_fn3(s, is_q, rd, rn, rm, tcg_gen_gvec_smin, size); } return; + case 0xe: /* SABD, UABD */ + if (u) { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_uabd, size); + } else { + gen_gvec_fn3(s, is_q, rd, rn, rm, gen_gvec_sabd, size); + } + return; case 0x10: /* ADD, SUB */ if (u) { gen_gvec_fn3(s, is_q, rd, rn, rm, tcg_gen_gvec_sub, size); @@ -11322,7 +11329,6 @@ static void disas_simd_3same_int(DisasContext *s, u= int32_t insn) genenvfn =3D fns[size][u]; break; } - case 0xe: /* SABD, UABD */ case 0xf: /* SABA, UABA */ { static NeonGenTwoOpFn * const fns[3][2] =3D { diff --git a/target/arm/translate.c b/target/arm/translate.c index 21529a9b8f..d288721c23 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -5102,6 +5102,126 @@ void gen_gvec_sqsub_qc(unsigned vece, uint32_t rd_o= fs, uint32_t rn_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); } =20 +static void gen_sabd_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) +{ + TCGv_i32 t =3D tcg_temp_new_i32(); + + tcg_gen_sub_i32(t, a, b); + tcg_gen_sub_i32(d, b, a); + tcg_gen_movcond_i32(TCG_COND_LT, d, a, b, d, t); + tcg_temp_free_i32(t); +} + +static void gen_sabd_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) +{ + TCGv_i64 t =3D tcg_temp_new_i64(); + + tcg_gen_sub_i64(t, a, b); + tcg_gen_sub_i64(d, b, a); + tcg_gen_movcond_i64(TCG_COND_LT, d, a, b, d, t); + tcg_temp_free_i64(t); +} + +static void gen_sabd_vec(unsigned vece, TCGv_vec d, TCGv_vec a, TCGv_vec b) +{ + TCGv_vec t =3D tcg_temp_new_vec_matching(d); + + tcg_gen_smin_vec(vece, t, a, b); + tcg_gen_smax_vec(vece, d, a, b); + tcg_gen_sub_vec(vece, d, d, t); + tcg_temp_free_vec(t); +} + +void gen_gvec_sabd(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] =3D { + INDEX_op_sub_vec, INDEX_op_smin_vec, INDEX_op_smax_vec, 0 + }; + static const GVecGen3 ops[4] =3D { + { .fniv =3D gen_sabd_vec, + .fno =3D gen_helper_gvec_sabd_b, + .opt_opc =3D vecop_list, + .vece =3D MO_8 }, + { .fniv =3D gen_sabd_vec, + .fno =3D gen_helper_gvec_sabd_h, + .opt_opc =3D vecop_list, + .vece =3D MO_16 }, + { .fni4 =3D gen_sabd_i32, + .fniv =3D gen_sabd_vec, + .fno =3D gen_helper_gvec_sabd_s, + .opt_opc =3D vecop_list, + .vece =3D MO_32 }, + { .fni8 =3D gen_sabd_i64, + .fniv =3D gen_sabd_vec, + .fno =3D gen_helper_gvec_sabd_d, + .prefer_i64 =3D TCG_TARGET_REG_BITS =3D=3D 64, + .opt_opc =3D vecop_list, + .vece =3D MO_64 }, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} + +static void gen_uabd_i32(TCGv_i32 d, TCGv_i32 a, TCGv_i32 b) +{ + TCGv_i32 t =3D tcg_temp_new_i32(); + + tcg_gen_sub_i32(t, a, b); + tcg_gen_sub_i32(d, b, a); + tcg_gen_movcond_i32(TCG_COND_LTU, d, a, b, d, t); + tcg_temp_free_i32(t); +} + +static void gen_uabd_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b) +{ + TCGv_i64 t =3D tcg_temp_new_i64(); + + tcg_gen_sub_i64(t, a, b); + tcg_gen_sub_i64(d, b, a); + tcg_gen_movcond_i64(TCG_COND_LTU, d, a, b, d, t); + tcg_temp_free_i64(t); +} + +static void gen_uabd_vec(unsigned vece, TCGv_vec d, TCGv_vec a, TCGv_vec b) +{ + TCGv_vec t =3D tcg_temp_new_vec_matching(d); + + tcg_gen_umin_vec(vece, t, a, b); + tcg_gen_umax_vec(vece, d, a, b); + tcg_gen_sub_vec(vece, d, d, t); + tcg_temp_free_vec(t); +} + +void gen_gvec_uabd(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] =3D { + INDEX_op_sub_vec, INDEX_op_umin_vec, INDEX_op_umax_vec, 0 + }; + static const GVecGen3 ops[4] =3D { + { .fniv =3D gen_uabd_vec, + .fno =3D gen_helper_gvec_uabd_b, + .opt_opc =3D vecop_list, + .vece =3D MO_8 }, + { .fniv =3D gen_uabd_vec, + .fno =3D gen_helper_gvec_uabd_h, + .opt_opc =3D vecop_list, + .vece =3D MO_16 }, + { .fni4 =3D gen_uabd_i32, + .fniv =3D gen_uabd_vec, + .fno =3D gen_helper_gvec_uabd_s, + .opt_opc =3D vecop_list, + .vece =3D MO_32 }, + { .fni8 =3D gen_uabd_i64, + .fniv =3D gen_uabd_vec, + .fno =3D gen_helper_gvec_uabd_d, + .prefer_i64 =3D TCG_TARGET_REG_BITS =3D=3D 64, + .opt_opc =3D vecop_list, + .vece =3D MO_64 }, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &ops[vece]); +} + /* Translate a NEON data processing instruction. Return nonzero if the instruction is invalid. We process data in a mixture of 32-bit and 64-bit chunks. @@ -5236,6 +5356,16 @@ static int disas_neon_data_insn(DisasContext *s, uin= t32_t insn) } return 1; =20 + case NEON_3R_VABD: + if (u) { + gen_gvec_uabd(size, rd_ofs, rn_ofs, rm_ofs, + vec_size, vec_size); + } else { + gen_gvec_sabd(size, rd_ofs, rn_ofs, rm_ofs, + vec_size, vec_size); + } + return 0; + case NEON_3R_VADD_VSUB: case NEON_3R_LOGIC: case NEON_3R_VMAX: @@ -5380,9 +5510,6 @@ static int disas_neon_data_insn(DisasContext *s, uint= 32_t insn) case NEON_3R_VQRSHL: GEN_NEON_INTEGER_OP_ENV(qrshl); break; - case NEON_3R_VABD: - GEN_NEON_INTEGER_OP(abd); - break; case NEON_3R_VABA: GEN_NEON_INTEGER_OP(abd); tcg_temp_free_i32(tmp2); diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index a483841add..a4972d02fc 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -1407,3 +1407,27 @@ DO_CMP0(gvec_cgt0_h, int16_t, >) DO_CMP0(gvec_cge0_h, int16_t, >=3D) =20 #undef DO_CMP0 + +#define DO_ABD(NAME, TYPE) \ +void HELPER(NAME)(void *vd, void *vn, void *vm, uint32_t desc) \ +{ \ + intptr_t i, opr_sz =3D simd_oprsz(desc); \ + TYPE *d =3D vd, *n =3D vn, *m =3D vm; \ + \ + for (i =3D 0; i < opr_sz / sizeof(TYPE); ++i) { \ + d[i] =3D n[i] < m[i] ? m[i] - n[i] : n[i] - m[i]; \ + } \ + clear_tail(d, opr_sz, simd_maxsz(desc)); \ +} + +DO_ABD(gvec_sabd_b, int8_t) +DO_ABD(gvec_sabd_h, int16_t) +DO_ABD(gvec_sabd_s, int32_t) +DO_ABD(gvec_sabd_d, int64_t) + +DO_ABD(gvec_uabd_b, uint8_t) +DO_ABD(gvec_uabd_h, uint16_t) +DO_ABD(gvec_uabd_s, uint32_t) +DO_ABD(gvec_uabd_d, uint64_t) + +#undef DO_ABD --=20 2.20.1