From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Monjalon Subject: Re: [PATCH 0/6] DPDK PMD for new QLogic FastLinQ QL4xxxx 25G/40G CNAs Date: Sat, 20 Feb 2016 12:49:38 -0800 (PST) Message-ID: <2521824.WtpN8U54QY@xps13> References: <1455982831-21682-1-git-send-email-harish.patil@qlogic.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7Bit Cc: dev@dpdk.org To: Harish Patil Return-path: Received: from mail-wm0-f46.google.com (mail-wm0-f46.google.com [74.125.82.46]) by dpdk.org (Postfix) with ESMTP id EC888C5FE for ; Sat, 20 Feb 2016 21:49:39 +0100 (CET) Received: by mail-wm0-f46.google.com with SMTP id g62so114648829wme.0 for ; Sat, 20 Feb 2016 12:49:39 -0800 (PST) In-Reply-To: <1455982831-21682-1-git-send-email-harish.patil@qlogic.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" 2016-02-20 07:40, Harish Patil: > This patch set introduces DPDK poll mode driver for new QLogic FastLinQ QL4xxxx > 25G/40G capable family of CNAs as well as their SR-IOV Virtual Functions (VF). > > The overall PMD driver design includes a common module called ecore that deals > with the low level HW and a upper layer portion that provides the glue logic. > > Specifically, the ecore module contains all of the common logic, > e.g. initialization, cleanup, infrastructure for interrupt handling, link > management, slowpath etc. as well as protocol agnostic features and supplying > an abstraction layer for other modules. > > The higher layer implements DPDK exported APIs/driver entry points by > interfacing with the common module for configuration/status and also the > fastpath routines. A new driver is always a good news :) Please could you share some performance numbers? What is the status about the integration of this driver in other environments? The layer named ecore seems to be what is named a base driver in other DPDK drivers. Maybe you could consider renaming the directory to base/ for consistency? About the format of the patches, I think it's better to split driver imports in several pieces to ease reviewing and later reference for specific bugs. What do you think of introducing basic features one by one?