From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Monjalon Subject: Re: [PATCH] app/testpmd: do not enable Rx offloads by default Date: Fri, 26 Jan 2018 09:35:47 +0100 Message-ID: <4518250.zEhSojeenl@xps> References: <1516695081-178919-1-git-send-email-motih@mellanox.com> <6183373.cbWzoGE1aT@xps> <6A0DE07E22DDAD4C9103DF62FEBC09093B721D05@shsmsx102.ccr.corp.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7Bit Cc: Moti Haimovsky , dev@dpdk.org, "shahafs@mellanox.com" , "Yigit, Ferruh" To: "Lu, Wenzhuo" Return-path: Received: from out3-smtp.messagingengine.com (out3-smtp.messagingengine.com [66.111.4.27]) by dpdk.org (Postfix) with ESMTP id 6D9741B300 for ; Fri, 26 Jan 2018 09:36:33 +0100 (CET) In-Reply-To: <6A0DE07E22DDAD4C9103DF62FEBC09093B721D05@shsmsx102.ccr.corp.intel.com> List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" 26/01/2018 09:06, Lu, Wenzhuo: > From: Thomas Monjalon [mailto:thomas@monjalon.net] > > 26/01/2018 08:30, Lu, Wenzhuo: > > > From: Thomas Monjalon [mailto:thomas@monjalon.net] > > > > 25/01/2018 02:11, Lu, Wenzhuo: > > > > > > --- a/app/test-pmd/testpmd.c > > > > > > +++ b/app/test-pmd/testpmd.c > > > > > > @@ -305,9 +305,7 @@ struct fwd_engine * fwd_engines[] = { > > > > > > */ > > > > > > struct rte_eth_rxmode rx_mode = { > > > > > > .max_rx_pkt_len = ETHER_MAX_LEN, /**< Default maximum > > frame > > > > > > length. */ > > > > > > - .offloads = (DEV_RX_OFFLOAD_VLAN_FILTER | > > > > > > - DEV_RX_OFFLOAD_VLAN_STRIP | > > > > > > - DEV_RX_OFFLOAD_CRC_STRIP), > > > > > > + .offloads = 0, > > > > > > > > > > Change the default behavior may trigger other problems. I think TX > > > > > offload > > > > could be a good reference. Get the capability and check what's > > > > supported first, then ignore the not supported functions with > > > > printing a warning but not block anything... > > > > > > > > I agree that we should check the capabilities before requesting an offload. > > > > But I disagree on another point: we should not enable an offload if > > > > the user did not request it explicitly. It makes things unclear. > > > > This is a testing tool, it should be close to the ethdev API behavior. > > > > > > > > Why these offload flags are silently enabled? > > > > > > I don't think it's silently. It's a global configuration. In this case, testpmd is > > the user, it does request it explicitly. If it's not so explicit, maybe we can > > consider moving all the configuration to a specific configure file. > > > Talking about why it's enabled by default. Hard to figure out the history. > > To my opinion, the original designer wants to simulate the common case. > > > > Please do not justify a design mistake by history. > > > > This is a test tool, so we don't care about the common case. > > A test tool should not try to guess the best configuration. > > Only the user should decide the configuration to apply, and the default > > should be empty, as the API is. > > I see the divergence. You think testpmd is a tool, so it should be white paper as the tool users may suppose it is. > I think testpmd is an example to let the APP developers know how to use DPDK. So any pre-configuration is acceptable. Ah ah ah :) No Examples are in examples/ directory.