From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 71A11440E3; Tue, 28 May 2024 16:59:41 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 1E155402E8; Tue, 28 May 2024 16:59:41 +0200 (CEST) Received: from mail-pf1-f178.google.com (mail-pf1-f178.google.com [209.85.210.178]) by mails.dpdk.org (Postfix) with ESMTP id A81DB402E4 for ; Tue, 28 May 2024 16:59:39 +0200 (CEST) Received: by mail-pf1-f178.google.com with SMTP id d2e1a72fcca58-6f8ea2df4b3so707838b3a.2 for ; Tue, 28 May 2024 07:59:39 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=networkplumber-org.20230601.gappssmtp.com; s=20230601; t=1716908379; x=1717513179; darn=dpdk.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:subject:cc:to:from:date:from:to:cc:subject:date :message-id:reply-to; bh=qbb3liC/+wdMGVqJcR2IrBLIcXL8xBZmIuyJ1UYiuW8=; b=S6Tk20YZQQ+D+yY9tNOHpk1JkKHFRWRqApWlle3MQBrFz827+PhW9pB4AgRGFx832c K95ohLYwpunn5+t9+ND83xtHcWVEHvWY5W4KiCJX8pwe4B/lg6wJ1Eg4LdXyQteV9pci OmYuVUaPlJtqdm63ic76gpE2O+O4cOvN4zX40elm7U2UJF2MbamtACxSCNpvneIYdnbd mb2EQHeKNfpF8G+eM9WPa9nidSNpffLrWp7Eza7CcIZHzuNgDRP/1sNgzNLLC3gxxxJK 27ukHkHZN9sC61ZTLq/LoolavYq8pc6hALv9O5LinGN1PS1rngEvZKNIMELDlouBTRh+ nGeA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1716908379; x=1717513179; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:subject:cc:to:from:date:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=qbb3liC/+wdMGVqJcR2IrBLIcXL8xBZmIuyJ1UYiuW8=; b=Bw1eBL2kGUxBP+bSOdOEi3KH6zeKGHeZoNadExCB7xoVCkTe2r7Bl51mOZtgdyXAiz 3J55G7mtu0hRvF6axrVCQhMRNAk22m4yBhI6D9a81OIneAqDPU7ILT/r8uPx+0RBrhjw uR1QuU5uCaI41wOumHZBAnrtWfcgNW6Pp4LASYYu4f404rg1s+enLeKTbVJVDnO6Ss4H K4jZuESa4Lj81KfMA2NAJQOJvfUDHIPex5bO2xwQMTZUDNYUnCNoc5NGIWyZgodAMa9/ F0nN7P479b3wnXT9CJsS+humDfDVm9p6pHHpYzpmi0R7cza3T1fRF0MZAZYCzsrPV2FP bpKw== X-Forwarded-Encrypted: i=1; AJvYcCWFoi86HlWQXm4GGbSOBJyJ4WoEqnh7cQDrHj3b+6/qqVtnAwtCIinNVLC4p8LzS0WtIYLBpbZe4jnpsBo= X-Gm-Message-State: AOJu0YzhnpYZct+0sLKl8cCP/2Ig/DadUO28JNJRVcFp9v2WvGzGEy+Q zi01zmSciJa7Ilw9BwHJct6ydDN26QaebuFqg5RyPCL7PpgDEEnSfJDldxRSP+8= X-Google-Smtp-Source: AGHT+IGc/2IxtutKu7bh6uTSQaVNl4QgrEGyb6p+S1YLAw7hZBXvWyRJycYX0VA3WI+3urdSP1+Ozw== X-Received: by 2002:a05:6a00:e0c:b0:6f4:3b76:ebf7 with SMTP id d2e1a72fcca58-6f8f37060e6mr12954337b3a.20.1716908378711; Tue, 28 May 2024 07:59:38 -0700 (PDT) Received: from hermes.local (204-195-96-226.wavecable.com. [204.195.96.226]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-6822092f45fsm6519326a12.9.2024.05.28.07.59.37 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 28 May 2024 07:59:38 -0700 (PDT) Date: Tue, 28 May 2024 07:59:36 -0700 From: Stephen Hemminger To: Mattias =?UTF-8?B?UsO2bm5ibG9t?= Cc: Mattias =?UTF-8?B?UsO2bm5ibG9t?= , dev@dpdk.org, Morten =?UTF-8?B?QnLDuHJ1cA==?= Subject: Re: [RFC v2] eal: provide option to use compiler memcpy instead of RTE Message-ID: <20240528075936.2110c31c@hermes.local> In-Reply-To: <738e376c-c5b6-44dc-ad51-00f40d2ea6b5@lysator.liu.se> References: <20240527111151.188607-1-mattias.ronnblom@ericsson.com> <20240528074354.190779-1-mattias.ronnblom@ericsson.com> <738e376c-c5b6-44dc-ad51-00f40d2ea6b5@lysator.liu.se> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org On Tue, 28 May 2024 10:19:15 +0200 Mattias R=C3=B6nnblom wrote: > > =20 >=20 > I've tested this patch some with DSW micro benchmarks, and the result is= =20 > a 2.5% reduction of the DSW+testapp overhead with cc/libc memcpy. GCC 11.= 4. >=20 > We've also run characteristic test suite of a large, real world app.=20 > Here, we saw no effect. GCC 10.5. >=20 > x86_64 in both cases (Skylake and Raptor Lake). >=20 > Last time we did the same, there were a noticeable performance=20 > degradation in both the above cases. >=20 > This is not a lot of data points, but I think it we should consider=20 > making the custom RTE memcpy() implementations optional in the next=20 > release, and if no-one complains, remove the implementations in the next= =20 > release. Lets go farther. 1. Announce that rte_memcpy will be marked deprecated in 24.11 release 2. In 24.11 do a global replace of rte_memcpy on the tree. And mark rte_memcpy as deprecated. 3. In 25.11 it can go away.