From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm1-x332.google.com (mail-wm1-x332.google.com [IPv6:2a00:1450:4864:20::332]) by sourceware.org (Postfix) with ESMTPS id B22F43858D28 for ; Wed, 12 Apr 2023 09:42:54 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org B22F43858D28 Authentication-Results: sourceware.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=gmail.com Received: by mail-wm1-x332.google.com with SMTP id l10-20020a05600c1d0a00b003f04bd3691eso16762838wms.5 for ; Wed, 12 Apr 2023 02:42:54 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20221208; t=1681292573; h=content-transfer-encoding:in-reply-to:from:references:cc:to :content-language:subject:user-agent:mime-version:date:message-id :from:to:cc:subject:date:message-id:reply-to; bh=9d0Uve9kKFaaEUoTlSnqJidXf6OPccXEbFPOQwyOLJI=; b=VlbJ0WBu1pNCId9qk/6YmrVNAdBna8XAP/ZgDaVMbKNQkPF19xqEkiP2gJlTB8wXmv qgnHU6kLlruV9hLVEpYF5i+fPG6Y5A9oZg4zaL8CEsSDvrDnG+sab1qvjSiQcSX+FHOB 0lzKdSaCSj2hlfYNveQcUFtGA3SY1T/fm7Zk16rpnDvxWbBOxLEoDcvYI+GruL5te11E /3uogA2DrnB1/zhRzaJ1IKvWqlaS5q4TyYkhmTUzq9JLQ3IsqavcUofUa0WvlCWSPfPJ h2fP1jJmoRo0YSSFPY9/ipUER33mZYYpt+P1VrIE3Ml9DeCyqgotZknDkiLIYswqBuGc qX2g== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1681292573; h=content-transfer-encoding:in-reply-to:from:references:cc:to :content-language:subject:user-agent:mime-version:date:message-id :x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=9d0Uve9kKFaaEUoTlSnqJidXf6OPccXEbFPOQwyOLJI=; b=CtcrETbqRRjLT7fVSYpj38T18E3LXM8pa6Ze0y9bCcj3F0M4+odf4tbhyn7TxxzyAV hCw9T03NdXEqoulTnnmn+t3oKLMlkMT0YJA38ZJwS36//l3bXoSFZu0oz6IY09d+5lTc AZkrG+PmbLYGGcYqa7fAI9njLjJgKWFE7ArUiN0gjbQV49WV90pOFo0u7z3+9eMTwhKt OtXOI6VmMeGroSWqXKcVqjgOegKqIkmxYUG3kR/ZoGC9CihtkK2EoVNrkSCzXTRDW81Q RCVHL1zl+ubxN4XxyV7vjfOv+O+7xpwdbq7bqtYl8lI5Y6uipabi9K72aKmKdHqmZiBR M6+A== X-Gm-Message-State: AAQBX9eBvKtxdRWi1BDIoZFsVH0hTE43P/PGgQn72zkvPZ4ZIb87sdgn oli8wSIh+Et9T+iRGb7XNQc= X-Google-Smtp-Source: AKy350am9ShLcUM96KGQOS4jJ3xmOi9zzN06lpXHAKPu+xgaFnjx8PaJa/6w+Qd1mvLNEAhaMbSHJw== X-Received: by 2002:a7b:c8cf:0:b0:3f0:a06a:7593 with SMTP id f15-20020a7bc8cf000000b003f0a06a7593mr224223wml.11.1681292573399; Wed, 12 Apr 2023 02:42:53 -0700 (PDT) Received: from [192.168.1.23] (ip-046-223-202-066.um13.pools.vodafone-ip.de. [46.223.202.66]) by smtp.gmail.com with ESMTPSA id y5-20020a1c4b05000000b003ef5f77901dsm1715217wma.45.2023.04.12.02.42.52 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Wed, 12 Apr 2023 02:42:52 -0700 (PDT) Message-ID: Date: Wed, 12 Apr 2023 11:42:51 +0200 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.9.1 Subject: Re: [PATCH] VECT: Add WHILE_LEN pattern for decrement IV support for auto-vectorization Content-Language: en-US To: Richard Biener , "juzhe.zhong@rivai.ai" Cc: "richard.sandiford" , gcc-patches , jeffreyalaw , linkw@linux.ibm.com, stefansf@linux.ibm.com, krebbel@linux.ibm.com References: <20230407014741.139387-1-juzhe.zhong@rivai.ai> <63723855B0BF2130+2023041120125573846623@rivai.ai> <139DA38AFC9CA5B5+2023041216004591287739@rivai.ai> From: Robin Dapp In-Reply-To: Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=0.4 required=5.0 tests=BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,FREEMAIL_FROM,NICE_REPLY_A,RCVD_IN_BARRACUDACENTRAL,RCVD_IN_DNSWL_NONE,SPF_HELO_NONE,SPF_PASS,TXREP autolearn=no autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org List-Id: >> I think we can CC IBM folks to see whether we can make WHILE_LEN works >> for both IBM and RVV ? > > I've CCed them. Adding WHILE_LEN support to rs6000/s390x would be > mainly the "easy" way to get len-masked (epilog) loop support. I've > figured actually implementing WHILE_ULT for AVX512 in the backend > results in some code generation challenges so I'm going to play > (again) with open-coding it as outlined above in the vectorizer itself > so followup passes (mostly IVOPTs) can do a better job. I'm with Ventana now but haven't updated my affiliation yet. CC'ing Stefan and Andreas fyi.