From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Hemminger Subject: Re: [PATCH v2 0/5] Optimize memcpy for AVX512 platforms Date: Mon, 18 Jan 2016 12:06:29 -0800 Message-ID: <20160118120629.5ed7bcd9@xeon-e3> References: <1452752002-107586-1-git-send-email-zhihong.wang@intel.com> <1453086314-30158-1-git-send-email-zhihong.wang@intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: dev@dpdk.org To: Zhihong Wang Return-path: Received: from mail-pa0-f54.google.com (mail-pa0-f54.google.com [209.85.220.54]) by dpdk.org (Postfix) with ESMTP id 88D3E8D97 for ; Mon, 18 Jan 2016 21:06:22 +0100 (CET) Received: by mail-pa0-f54.google.com with SMTP id ho8so177832031pac.2 for ; Mon, 18 Jan 2016 12:06:22 -0800 (PST) In-Reply-To: <1453086314-30158-1-git-send-email-zhihong.wang@intel.com> List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" On Sun, 17 Jan 2016 22:05:09 -0500 Zhihong Wang wrote: > This patch set optimizes DPDK memcpy for AVX512 platforms, to make full > utilization of hardware resources and deliver high performance. > > In current DPDK, memcpy holds a large proportion of execution time in > libs like Vhost, especially for large packets, and this patch can bring > considerable benefits. > > The implementation is based on the current DPDK memcpy framework, some > background introduction can be found in these threads: > http://dpdk.org/ml/archives/dev/2014-November/008158.html > http://dpdk.org/ml/archives/dev/2015-January/011800.html > > Code changes are: > > 1. Read CPUID to check if AVX512 is supported by CPU > > 2. Predefine AVX512 macro if AVX512 is enabled by compiler > > 3. Implement AVX512 memcpy and choose the right implementation based on > predefined macros > > 4. Decide alignment unit for memcpy perf test based on predefined macros Cool, I like it. How much impact does this have on VHOST?