From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Dumitrescu, Cristian" Subject: Re: [v2 22/23] Packet Framework IPv4 pipeline sample app Date: Mon, 9 Jun 2014 13:25:44 +0000 Message-ID: <3EB4FA525960D640B5BDFFD6A3D891261B1C2415@IRSMSX102.ger.corp.intel.com> References: <1401905319-8882-1-git-send-email-cristian.dumitrescu@intel.com> <1401905319-8882-23-git-send-email-cristian.dumitrescu@intel.com> <53957A4E.6090401@6wind.com> <3EB4FA525960D640B5BDFFD6A3D891261B1C2354@IRSMSX102.ger.corp.intel.com> <5395A4F6.7050609@6wind.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Cc: "dev-VfR2kkLFssw@public.gmane.org" To: Olivier MATZ Return-path: In-Reply-To: <5395A4F6.7050609-pdR9zngts4EAvxtiuMwx3w@public.gmane.org> Content-Language: en-US List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces-VfR2kkLFssw@public.gmane.org Sender: "dev" Hi Olivier, A few notes on using pktmbuf here: 1. As the name implies, pktmbuf should be used for packets and ctrlmbuf sho= uld be used for control messages :). IMHO using pktmbuf to control messages= is a confusing workaround. 2. Pktmbuf has a lot of overhead that is not needed in order to send short = messages between cores. Pktmbuf has a lot of pointers and other fields that= do not make sense for messages. I don't think we want people to say DPDK i= s difficult to use because e.g. sending 2 bytes from core A to core B requi= res initializing a bunch of pointers and other fields that do not make sens= e. 3. Once we start using pktmbuf to send messages, it is likely that other pe= ople will follow this example, and they might do it incorrectly. I don't th= ink we want to see emails on this list from people asking e.g: i) Why does my app segfaults, when all I want to do is send 2 bytes from c= ore A to core B? ii) Why does my app segfaults when core A writes a message to a NIC TX que= ue? :) Using an app dependent structure requires duplicating the work to create/fr= ee the pool of such structures, and alloc/free mechanism. And then some peo= ple will ask why are we not using ctrlmbuf, as long as ctrlmbuf exists in D= PDK. I think that, as long as we have ctrlmbuf and pktmbuf in DPDK, we should fo= llow the existing model. We should not look for workarounds that we know we= plan to change anyway, we should look for the right solution. We both agre= e we need to refresh pktmbuf and ctrlmbuf, but my point is we should not do= changes as long as we don't know what the agreed solution will look like? Thanks, Cristian -----Original Message----- From: Olivier MATZ [mailto:olivier.matz-pdR9zngts4EAvxtiuMwx3w@public.gmane.org] = Sent: Monday, June 9, 2014 1:14 PM To: Dumitrescu, Cristian Cc: dev-VfR2kkLFssw@public.gmane.org Subject: Re: [dpdk-dev] [v2 22/23] Packet Framework IPv4 pipeline sample app Hi Christian, > We need a message type defined for message passing between cores, and > pktmbuf is definitely not the right approach. Could you please explain why a pktmbuf is not the right approach? As proposed in http://dpdk.org/ml/archives/dev/2014-May/002759.html I think the control mbuf could be replaced by a packet mbuf or an application private structure. Regards, Olivier -------------------------------------------------------------- Intel Shannon Limited Registered in Ireland Registered Office: Collinstown Industrial Park, Leixlip, County Kildare Registered Number: 308263 Business address: Dromore House, East Park, Shannon, Co. Clare This e-mail and any attachments may contain confidential material for the s= ole use of the intended recipient(s). Any review or distribution by others = is strictly prohibited. If you are not the intended recipient, please conta= ct the sender and delete all copies.