From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-10.1 required=3.0 tests=BAYES_00,DKIM_INVALID, DKIM_SIGNED,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER, MAILING_LIST_MULTI,NICE_REPLY_A,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id C7AF4C433B4 for ; Tue, 18 May 2021 14:51:07 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 717BA61074 for ; Tue, 18 May 2021 14:51:07 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 717BA61074 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=linaro.org Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Received: from localhost ([::1]:36416 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1lj13q-00086O-FZ for qemu-devel@archiver.kernel.org; Tue, 18 May 2021 10:51:06 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53876) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1lj0yb-00036X-S7 for qemu-devel@nongnu.org; Tue, 18 May 2021 10:45:44 -0400 Received: from mail-vk1-xa2f.google.com ([2607:f8b0:4864:20::a2f]:39482) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1lj0yI-00031G-LJ for qemu-devel@nongnu.org; Tue, 18 May 2021 10:45:35 -0400 Received: by mail-vk1-xa2f.google.com with SMTP id k22so82387vko.6 for ; Tue, 18 May 2021 07:45:21 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; h=subject:to:cc:references:from:message-id:date:user-agent :mime-version:in-reply-to:content-language:content-transfer-encoding; bh=da1X5FDtZQFaLk6YknN3fhzzydUuvdyIdH0SwN1kKAQ=; b=nerQXeJaZvqvPeX5A5IaVcnS9Sm0q1YKLrva0qUaA8foIkSn+U85sub19F4w9xUcLb RjedmwJPam29h1LhD34c8EJ7mIHw54g4VkbCSO7wbjmL1Td3fybRgcdAnDXGDbDa93BO p4KlzUXpg23qCCh5GRhkFYa6QlxDRTbOHKBc2Ej8/iHwRhZKKx9f7OIj6ftl8vPgs6Jc JuShjZ8VjT1PhtEMVJWmKqBixg/DoxihwUL0OQlgdVd0afbMaPs4OLlRjqhW/aNcwNLr 78kkT8jSr7BJ2RCPLMNHpuXX5nKh17HJErmbwBDcSMWZ6NP5gH6fxgpDJSO2pazzMo6M +7vw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:subject:to:cc:references:from:message-id:date :user-agent:mime-version:in-reply-to:content-language :content-transfer-encoding; bh=da1X5FDtZQFaLk6YknN3fhzzydUuvdyIdH0SwN1kKAQ=; b=QLjQIe+3Rrfv7U/184XHnTqbo4LA7YKy5uVsIq4GifbtoBM+5qNwOJ6lO1/sj8DnWx WZg6aTZC6sGUTOGXaPjf8MRGhFZ8o2L1KCZ8X3Yu8d/XHTEIWwHNoMsrcMVnC9m3Y5h7 WrdlxlCCPmS3126+DU/GWq4VFWLHyPiyUrxX/Xwkmxzs/e6Xeds0t4y6TEDR/HHoKRgE k4Uf9N20WhB1BQbWZEnbd3M2fZ6k/Gk9fClhnOZaO/WMUfP5/PCAtmm8QxvJc6+XNPNN G17V8N5SzJrho0G/avfzoRVDWa+YWblmAxDpTfDIG2cM/3ruexFSN8ivGaoxJ2npz1Pl le2A== X-Gm-Message-State: AOAM531saa5Pjz9hMu1NLdOqpmypnVc3kz9Ox7JaMZum41UvtUQIMZ4+ S7t0YRuc8vmf/PSlp5Kx1gdgEg== X-Google-Smtp-Source: ABdhPJynNyah/DBGgOyIB/mHSiOsV7CHALnjlLliQaG2FPN3jVDgqmxA7KfKXqOlQW/WpjBG21sxaw== X-Received: by 2002:a1f:2b14:: with SMTP id r20mr6345684vkr.9.1621349121229; Tue, 18 May 2021 07:45:21 -0700 (PDT) Received: from [172.16.22.144] ([45.235.253.15]) by smtp.gmail.com with ESMTPSA id o72sm1961830vka.30.2021.05.18.07.45.20 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Tue, 18 May 2021 07:45:20 -0700 (PDT) Subject: Re: [PATCH v1 08/11] target/arm: Implement bfloat16 matrix multiply accumulate To: Peter Maydell References: <20210416235928.1631788-1-richard.henderson@linaro.org> <20210416235928.1631788-9-richard.henderson@linaro.org> From: Richard Henderson Message-ID: <40bfafe9-ba14-f718-3240-102766f17811@linaro.org> Date: Tue, 18 May 2021 09:45:18 -0500 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.8.1 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8; format=flowed Content-Language: en-US Content-Transfer-Encoding: 7bit Received-SPF: pass client-ip=2607:f8b0:4864:20::a2f; envelope-from=richard.henderson@linaro.org; helo=mail-vk1-xa2f.google.com 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, NICE_REPLY_A=-0.001, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=unavailable autolearn_force=no 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: qemu-arm , QEMU Developers Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" On 5/18/21 7:37 AM, Peter Maydell wrote: > On Sat, 17 Apr 2021 at 01:00, Richard Henderson > wrote: >> >> This is BFMMLA for both AArch64 AdvSIMD and SVE, >> and VMMLA.BF16 for AArch32 NEON. >> >> Signed-off-by: Richard Henderson > >> +void HELPER(gvec_bfmmla)(void *vd, void *vn, void *vm, void *va, uint32_t desc) >> +{ >> + intptr_t s, opr_sz = simd_oprsz(desc); >> + float32 *d = vd, *a = va; >> + uint32_t *n = vn, *m = vm; >> + >> + for (s = 0; s < opr_sz / 4; s += 4) { >> + float32 sum00, sum01, sum10, sum11; >> + >> + /* >> + * Process the entire segment at once, writing back the >> + * results only after we've consumed all of the inputs. >> + * >> + * Key to indicies by column: > > "indices" > >> + * i j i k j k >> + */ >> + sum00 = a[s + H4(0 + 0)]; >> + sum00 = bfdotadd(sum00, n[s + H4(0 + 0)], m[s + H4(0 + 0)]); >> + sum00 = bfdotadd(sum00, n[s + H4(0 + 1)], m[s + H4(0 + 1)]); > > I can't make these indices match up with the arm arm pseudocode ones, > which index by "4*i + 2*k + 0" and "4*i + 2*k + 1", not "2*i + k"; > are we hiding a division by 2 somewhere? Yes. We're passing BFloat16 pairs via uint32_t[] to bfdotadd(). r~