From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by sourceware.org (Postfix) with ESMTPS id C855638582B4 for ; Mon, 27 Jun 2022 16:02:51 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.1 sourceware.org C855638582B4 Received: from mail-qk1-f200.google.com (mail-qk1-f200.google.com [209.85.222.200]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-554-O06VCvzzPLyh6fgZCrxoDw-1; Mon, 27 Jun 2022 12:02:50 -0400 X-MC-Unique: O06VCvzzPLyh6fgZCrxoDw-1 Received: by mail-qk1-f200.google.com with SMTP id bl27-20020a05620a1a9b00b0069994eeb30cso10480829qkb.11 for ; Mon, 27 Jun 2022 09:02:50 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:message-id:date:mime-version:user-agent:subject :content-language:to:cc:references:from:organization:in-reply-to :content-transfer-encoding; bh=tazyMlOjTxb7iXV7JKwgTlVZ35R7laY5u0tZBFl/MHU=; b=N7g+BurVvBGmT/uODfbwDG3cgt89FOuKdUtQ8YfJsB5jOrVwhnn7qH5R86uLDJJk47 4SRJ4XSUj5Sk4Yn+Fc3BDXq8Op9coVD1vmx/9nObcRwvukedZRPfcC8o1L8WoGQlw5Cq xxVm7RPYyLDhYv5Q/i1L6JzVje+2EYvywYCnBC3wYhLntX3mcNlSlNXgYAVIrYtadze3 5PNgooiZqRCf/090Wi4dlYdmSLxdX28kAf0TdGnd7/Vj4a0tuf8SN6TrATGzgMP0eARv M4eLFpkO4/wo+okwdF/RGb77w3X6n4CmHx9+d0LihUMTgkn4duF+Zk2VujY5u462ZxnH WXSQ== X-Gm-Message-State: AJIora/WC2eyXnvJy6BoJiqt4OuYYo6U5oIs4GdLmHLKrYKjYqeZ5m4Q kG1Z4j81k1uw/tF4LB7HgkX0rwfIYQ5AmamfMpci3UtoOSw5oEbE94806DRGldAAz5PzJ0jX3SO IXpxzmSeF+sbsHY6Tc1Tp X-Received: by 2002:a05:620a:16d2:b0:6ae:ff30:1161 with SMTP id a18-20020a05620a16d200b006aeff301161mr8439076qkn.540.1656345769793; Mon, 27 Jun 2022 09:02:49 -0700 (PDT) X-Google-Smtp-Source: AGRyM1uXcCL08CLqBPMZLlrLVyKLepUjCEDqdlG/E5MZuBGI7/dDzPKRvpl+Sl6RhH2kZIYv/jqXJg== X-Received: by 2002:a05:620a:16d2:b0:6ae:ff30:1161 with SMTP id a18-20020a05620a16d200b006aeff301161mr8439040qkn.540.1656345769479; Mon, 27 Jun 2022 09:02:49 -0700 (PDT) Received: from [192.168.0.241] (135-23-175-80.cpe.pppoe.ca. [135.23.175.80]) by smtp.gmail.com with ESMTPSA id w20-20020a05620a445400b006af37b005bdsm1580544qkp.128.2022.06.27.09.02.48 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Mon, 27 Jun 2022 09:02:48 -0700 (PDT) Message-ID: <58f45a1d-dbf6-65ca-8436-881249678630@redhat.com> Date: Mon, 27 Jun 2022 12:02:47 -0400 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.9.0 Subject: Re: [PATCH] aarch64: Optimize string functions with shrn instruction To: danilak@google.com, libc-alpha@sourceware.org Cc: Szabolcs.Nagy@arm.com References: <20220622072946.3123916-1-danilak@google.com> From: Carlos O'Donell Organization: Red Hat In-Reply-To: <20220622072946.3123916-1-danilak@google.com> X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Language: en-US Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=-10.1 required=5.0 tests=BAYES_00, DKIMWL_WL_HIGH, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, NICE_REPLY_A, RCVD_IN_DNSWL_NONE, SPF_HELO_NONE, SPF_NONE, TXREP, T_SCC_BODY_TEXT_LINE autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org X-BeenThere: libc-alpha@sourceware.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Libc-alpha mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 27 Jun 2022 16:02:53 -0000 On 6/22/22 03:29, Danila Kutenin via Libc-alpha wrote: > We found that string functions were using AND+ADDP > to find the nibble/syndrome mask but there is an easier > opportunity through `SHRN dst, src, 4` and has same > latency on all SIMD ARMv8 targets as ADDP. There are also > possible gaps for memcmp but that's for another patch. > > We see 10-20% savings for small-mid size cases (<=128) > which are primary cases for general workloads. > > Signed-off-by: Danila Kutenin Thank you very much for working through this. This patch came up for review in Monday patch queue review. Please drop the "Signed-off-by" and submit the patch again. We only accept "Signed-off-by" in the cases where the submitter is not assigning copyright. In your case the Google blanket copyright assignment is in place (I verified) and so you are assigning to the FSF. Thank you for taking this extra step. I appreciate that it is a complicated extra step. We are working to keep the two submission processes clear (copyright assignment vs. no copyright assignment) and easy to follow. -- Cheers, Carlos.