From nobody Sun May 19 23:34:11 2024 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=redhat.com ARC-Seal: i=1; a=rsa-sha256; t=1661446416; cv=none; d=zohomail.com; s=zohoarc; b=AbLjPWzQZeWHTdV098Sh9Gq1Cv9hj8MoBnSXKDpVC0K+eZKjDHmFVQBM5elh8UcvvI8C1R9x3BQaE71Z8BPkoHeN0403Owc6EGCNI6aQPPJTuPrxM+nHnkAPznc9jbRb7bPJTpTF2a0SNVd78WUiGEj2OLK1fRkEOHzrfeGOyE8= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1661446416; h=Content-Type:Content-Transfer-Encoding:Date:From:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:Sender:Subject:To; bh=4F7JyGUN6Ik/gJc/PLLg+kIf9SE5uPDEWBtsr9MyWFY=; b=hl5oYbt+YoXLbX05dqWgxsJKFGtnzA2eFj1HXnI/9B/t212onJfoILTJ6lrDNS/yGhRdQlxCdHcaPsU2pNQ0n5RknlX6ADz1XrZ0+v5MNyaZKapiBwpUDExIzO3JJSyTpou5fmf1bjppPf6G9ucIjEFH41fnJroF3ufEIN/zje4= 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) Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1661446416735185.0666059877043; Thu, 25 Aug 2022 09:53:36 -0700 (PDT) Received: from localhost ([::1]:58948 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1oRG6o-0007Pl-23 for importer@patchew.org; Thu, 25 Aug 2022 12:53:34 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:47598) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1oRG20-000689-8V for qemu-devel@nongnu.org; Thu, 25 Aug 2022 12:48:36 -0400 Received: from us-smtp-delivery-124.mimecast.com ([170.10.129.124]:55259) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1oRG1x-0002Oc-5q for qemu-devel@nongnu.org; Thu, 25 Aug 2022 12:48:34 -0400 Received: from mail-wm1-f71.google.com (mail-wm1-f71.google.com [209.85.128.71]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_128_GCM_SHA256) id us-mta-103-JBxU6RUbM12ng0IKT0MFyw-1; Thu, 25 Aug 2022 12:48:30 -0400 Received: by mail-wm1-f71.google.com with SMTP id a17-20020a05600c349100b003a545125f6eso2684019wmq.4 for ; Thu, 25 Aug 2022 09:48:30 -0700 (PDT) Received: from goa-sendmail ([93.56.170.129]) by smtp.gmail.com with ESMTPSA id b5-20020a05600c4e0500b003a331c6bffdsm5663561wmq.47.2022.08.25.09.48.27 for (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 25 Aug 2022 09:48:27 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1661446112; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding; bh=4F7JyGUN6Ik/gJc/PLLg+kIf9SE5uPDEWBtsr9MyWFY=; b=ZAbFE64PhdA1EFC+LQIcVjw+eDGDY5J7fUoM8ye/NlfuvNgfPlPETvkwVVYZ/gpkxHIwmQ L+Gqx0Y0GbR847UK7cW8WLNrvRmBCQ3x61Z5ece+Q1nFXuE5PX2Gq6jQfQf3gCYA9PUSNO i47w2K7vw3vapvd4XsKfKupImLGneH4= X-MC-Unique: JBxU6RUbM12ng0IKT0MFyw-1 X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding:mime-version:message-id:date:subject:to :from:x-gm-message-state:from:to:cc; bh=4F7JyGUN6Ik/gJc/PLLg+kIf9SE5uPDEWBtsr9MyWFY=; b=vNTZ6qfDSwqndnCe6zpdD/B348JMRRCd9+kwRQJOfgVyP6qMjXe1BfLRZoM2Ax13hN HdrFwyKC1feH32nPGaEaGYTZzMOOm720RBLSi+EO6oqECt12A7T/JFdAJLCWQAQrGqYI BEFi8lGghe4VNgVXXC7kYaNKj02V98vn6cK1IKGcu3PWp+LyLsdDdIV20sgtWDcnaZxs o5TNcS7jU5P/nsY2Jymu3WaDPUhOBAxRnSMegymBCBTWpLUrnEHT+gu3q2ElSMXH3S8B AjnVOTkK8CWUBqlNFFag7RcW8QZFEfGR9JL2k0OjWRBzgLWrZ79BaujTyVTqMMsK/JGL EcTA== X-Gm-Message-State: ACgBeo1+9wQKCD+4L7fNO4LjnCd7+AYNibQIHi0KCJrZG/FJIpyc8a51 Prbm/2d88NxjR4X1Dox1e3+zTRQ4usddLACDkgY6zF9MjEqSlqsM8OgdK1/HXrtCvY7sbcqLeDd 2RVVM6VSo8HnxEAiBK9IqKw+oZnzdt5O5MXmTfQmftSZXxrNr4tLEQbbU6QAQVP6jvhU= X-Received: by 2002:a05:6000:1c0c:b0:225:569c:dd2c with SMTP id ba12-20020a0560001c0c00b00225569cdd2cmr2750275wrb.628.1661446108805; Thu, 25 Aug 2022 09:48:28 -0700 (PDT) X-Google-Smtp-Source: AA6agR5/mPqI9ITlQVDF9L6Bs6phC+LNTDQcqB6Wy87M329TVORCsOQZeqEnvpDvGVPJYZodkWrEVA== X-Received: by 2002:a05:6000:1c0c:b0:225:569c:dd2c with SMTP id ba12-20020a0560001c0c00b00225569cdd2cmr2750257wrb.628.1661446108268; Thu, 25 Aug 2022 09:48:28 -0700 (PDT) From: Paolo Bonzini To: qemu-devel@nongnu.org Subject: [PATCH] i386: DPPS rounding fix Date: Thu, 25 Aug 2022 18:48:27 +0200 Message-Id: <20220825164827.392942-1-pbonzini@redhat.com> X-Mailer: git-send-email 2.37.1 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=170.10.129.124; envelope-from=pbonzini@redhat.com; helo=us-smtp-delivery-124.mimecast.com X-Spam_score_int: -27 X-Spam_score: -2.8 X-Spam_bar: -- X-Spam_report: (-2.8 / 5.0 requ) BAYES_00=-1.9, DKIMWL_WL_HIGH=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_LOW=-0.7, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" X-ZohoMail-DKIM: pass (identity @redhat.com) X-ZM-MESSAGEID: 1661446419196100001 Content-Type: text/plain; charset="utf-8" The DPPS (Dot Product) instruction is defined to first sum pairs of intermediate results, then sum those values to get the final result. i.e. (A+B)+(C+D) We incrementally sum the results, i.e. ((A+B)+C)+D, which can result in incorrect rouding. For consistency, also change the variable names to the ones used in the Intel SDM and implement DPPD following the manual. Based on a patch by Paul Brook . Signed-off-by: Paolo Bonzini Reviewed-by: Richard Henderson --- target/i386/ops_sse.h | 67 ++++++++++++++++++++++--------------------- 1 file changed, 35 insertions(+), 32 deletions(-) diff --git a/target/i386/ops_sse.h b/target/i386/ops_sse.h index 2524db4c25..b12b271fcd 100644 --- a/target/i386/ops_sse.h +++ b/target/i386/ops_sse.h @@ -1943,56 +1943,59 @@ SSE_HELPER_I(helper_pblendw, W, 8, FBLENDP) =20 void glue(helper_dpps, SUFFIX)(CPUX86State *env, Reg *d, Reg *s, uint32_t = mask) { - float32 iresult =3D float32_zero; + float32 prod1, prod2, temp2, temp3, temp4; =20 + /* + * We must evaluate (A+B)+(C+D), not ((A+B)+C)+D + * to correctly round the intermediate results + */ if (mask & (1 << 4)) { - iresult =3D float32_add(iresult, - float32_mul(d->ZMM_S(0), s->ZMM_S(0), - &env->sse_status), - &env->sse_status); + prod1 =3D float32_mul(d->ZMM_S(0), s->ZMM_S(0), &env->sse_status); + } else { + prod1 =3D float32_zero; } if (mask & (1 << 5)) { - iresult =3D float32_add(iresult, - float32_mul(d->ZMM_S(1), s->ZMM_S(1), - &env->sse_status), - &env->sse_status); + prod2 =3D float32_mul(d->ZMM_S(1), s->ZMM_S(1), &env->sse_status); + } else { + prod2 =3D float32_zero; } + temp2 =3D float32_add(prod1, prod2, &env->sse_status); if (mask & (1 << 6)) { - iresult =3D float32_add(iresult, - float32_mul(d->ZMM_S(2), s->ZMM_S(2), - &env->sse_status), - &env->sse_status); + prod1 =3D float32_mul(d->ZMM_S(2), s->ZMM_S(2), &env->sse_status); + } else { + prod1 =3D float32_zero; } if (mask & (1 << 7)) { - iresult =3D float32_add(iresult, - float32_mul(d->ZMM_S(3), s->ZMM_S(3), - &env->sse_status), - &env->sse_status); + prod2 =3D float32_mul(d->ZMM_S(3), s->ZMM_S(3), &env->sse_status); + } else { + prod2 =3D float32_zero; } - d->ZMM_S(0) =3D (mask & (1 << 0)) ? iresult : float32_zero; - d->ZMM_S(1) =3D (mask & (1 << 1)) ? iresult : float32_zero; - d->ZMM_S(2) =3D (mask & (1 << 2)) ? iresult : float32_zero; - d->ZMM_S(3) =3D (mask & (1 << 3)) ? iresult : float32_zero; + temp3 =3D float32_add(prod1, prod2, &env->sse_status); + temp4 =3D float32_add(temp2, temp3, &env->sse_status); + + d->ZMM_S(0) =3D (mask & (1 << 0)) ? temp4 : float32_zero; + d->ZMM_S(1) =3D (mask & (1 << 1)) ? temp4 : float32_zero; + d->ZMM_S(2) =3D (mask & (1 << 2)) ? temp4 : float32_zero; + d->ZMM_S(3) =3D (mask & (1 << 3)) ? temp4 : float32_zero; } =20 void glue(helper_dppd, SUFFIX)(CPUX86State *env, Reg *d, Reg *s, uint32_t = mask) { - float64 iresult =3D float64_zero; + float64 prod1, prod2, temp2; =20 if (mask & (1 << 4)) { - iresult =3D float64_add(iresult, - float64_mul(d->ZMM_D(0), s->ZMM_D(0), - &env->sse_status), - &env->sse_status); + prod1 =3D float64_mul(d->ZMM_D(0), s->ZMM_D(0), &env->sse_status); + } else { + prod1 =3D float64_zero; } if (mask & (1 << 5)) { - iresult =3D float64_add(iresult, - float64_mul(d->ZMM_D(1), s->ZMM_D(1), - &env->sse_status), - &env->sse_status); + prod2 =3D float64_mul(d->ZMM_D(1), s->ZMM_D(1), &env->sse_status); + } else { + prod2 =3D float64_zero; } - d->ZMM_D(0) =3D (mask & (1 << 0)) ? iresult : float64_zero; - d->ZMM_D(1) =3D (mask & (1 << 1)) ? iresult : float64_zero; + temp2 =3D float64_add(prod1, prod2, &env->sse_status); + d->ZMM_D(0) =3D (mask & (1 << 0)) ? temp2 : float64_zero; + d->ZMM_D(1) =3D (mask & (1 << 1)) ? temp2 : float64_zero; } =20 void glue(helper_mpsadbw, SUFFIX)(CPUX86State *env, Reg *d, Reg *s, --=20 2.37.1