From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-5.8 required=3.0 tests=HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,MENTIONS_GIT_HOSTING,SPF_HELO_NONE,SPF_PASS autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id A9334C48BD3 for ; Wed, 26 Jun 2019 16:36:40 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 7863C20665 for ; Wed, 26 Jun 2019 16:36:40 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726347AbfFZQgj convert rfc822-to-8bit (ORCPT ); Wed, 26 Jun 2019 12:36:39 -0400 Received: from mx1.redhat.com ([209.132.183.28]:59474 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726014AbfFZQgj (ORCPT ); Wed, 26 Jun 2019 12:36:39 -0400 Received: from smtp.corp.redhat.com (int-mx07.intmail.prod.int.phx2.redhat.com [10.5.11.22]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id B64033082E0F; Wed, 26 Jun 2019 16:36:21 +0000 (UTC) Received: from carbon (ovpn-200-45.brq.redhat.com [10.40.200.45]) by smtp.corp.redhat.com (Postfix) with ESMTP id 434961001B04; Wed, 26 Jun 2019 16:36:03 +0000 (UTC) Date: Wed, 26 Jun 2019 18:36:01 +0200 From: Jesper Dangaard Brouer To: Toke =?UTF-8?B?SMO4aWxhbmQtSsO4cmdlbnNlbg==?= Cc: "Machulsky\, Zorik" , "Jubran\, Samih" , "davem\@davemloft.net" , "netdev\@vger.kernel.org" , "Woodhouse\, David" , "Matushevsky\, Alexander" , "Bshara\, Saeed" , "Wilson\, Matt" , "Liguori\, Anthony" , "Bshara\, Nafea" , "Tzalik\, Guy" , "Belgazal\, Netanel" , "Saidi\, Ali" , "Herrenschmidt\, Benjamin" , "Kiyanovski\, Arthur" , Daniel Borkmann , Ilias Apalodimas , Alexei Starovoitov , Jakub Kicinski , "xdp-newbies\@vger.kernel.org" , brouer@redhat.com Subject: Re: XDP multi-buffer incl. jumbo-frames (Was: [RFC V1 net-next 1/1] net: ena: implement XDP drop support) Message-ID: <20190626183601.1bd43d00@carbon> In-Reply-To: <87ef3gbcpz.fsf@toke.dk> References: <20190623070649.18447-1-sameehj@amazon.com> <20190623070649.18447-2-sameehj@amazon.com> <20190623162133.6b7f24e1@carbon> <20190626103829.5360ef2d@carbon> <87a7e4d0nj.fsf@toke.dk> <20190626164059.4a9511cf@carbon> <87ef3gbcpz.fsf@toke.dk> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8BIT X-Scanned-By: MIMEDefang 2.84 on 10.5.11.22 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.46]); Wed, 26 Jun 2019 16:36:38 +0000 (UTC) Sender: netdev-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: netdev@vger.kernel.org On Wed, 26 Jun 2019 17:14:32 +0200 Toke Høiland-Jørgensen wrote: > Jesper Dangaard Brouer writes: > > > On Wed, 26 Jun 2019 13:52:16 +0200 > > Toke Høiland-Jørgensen wrote: > > > >> Jesper Dangaard Brouer writes: > >> > >> > On Tue, 25 Jun 2019 03:19:22 +0000 > >> > "Machulsky, Zorik" wrote: > >> > > >> >> On 6/23/19, 7:21 AM, "Jesper Dangaard Brouer" wrote: > >> >> > >> >> On Sun, 23 Jun 2019 10:06:49 +0300 wrote: > >> >> > >> >> > This commit implements the basic functionality of drop/pass logic in the > >> >> > ena driver. > >> >> > >> >> Usually we require a driver to implement all the XDP return codes, > >> >> before we accept it. But as Daniel and I discussed with Zorik during > >> >> NetConf[1], we are going to make an exception and accept the driver > >> >> if you also implement XDP_TX. > >> >> > >> >> As we trust that Zorik/Amazon will follow and implement XDP_REDIRECT > >> >> later, given he/you wants AF_XDP support which requires XDP_REDIRECT. > >> >> > >> >> Jesper, thanks for your comments and very helpful discussion during > >> >> NetConf! That's the plan, as we agreed. From our side I would like to > >> >> reiterate again the importance of multi-buffer support by xdp frame. > >> >> We would really prefer not to see our MTU shrinking because of xdp > >> >> support. > >> > > >> > Okay we really need to make a serious attempt to find a way to support > >> > multi-buffer packets with XDP. With the important criteria of not > >> > hurting performance of the single-buffer per packet design. > >> > > >> > I've created a design document[2], that I will update based on our > >> > discussions: [2] https://github.com/xdp-project/xdp-project/blob/master/areas/core/xdp-multi-buffer01-design.org > >> > > >> > The use-case that really convinced me was Eric's packet header-split. > >> > > >> > > >> > Lets refresh: Why XDP don't have multi-buffer support: > >> > > >> > XDP is designed for maximum performance, which is why certain driver-level > >> > use-cases were not supported, like multi-buffer packets (like jumbo-frames). > >> > As it e.g. complicated the driver RX-loop and memory model handling. > >> > > >> > The single buffer per packet design, is also tied into eBPF Direct-Access > >> > (DA) to packet data, which can only be allowed if the packet memory is in > >> > contiguous memory. This DA feature is essential for XDP performance. > >> > > >> > > >> > One way forward is to define that XDP only get access to the first > >> > packet buffer, and it cannot see subsequent buffers. For XDP_TX and > >> > XDP_REDIRECT to work then XDP still need to carry pointers (plus > >> > len+offset) to the other buffers, which is 16 bytes per extra buffer. > >> > >> Yeah, I think this would be reasonable. As long as we can have a > >> metadata field with the full length + still give XDP programs the > >> ability to truncate the packet (i.e., discard the subsequent pages) > > > > You touch upon some interesting complications already: > > > > 1. It is valuable for XDP bpf_prog to know "full" length? > > (if so, then we need to extend xdp ctx with info) > > > > But if we need to know the full length, when the first-buffer is > > processed. Then realize that this affect the drivers RX-loop, because > > then we need to "collect" all the buffers before we can know the > > length (although some HW provide this in first descriptor). > > > > We likely have to change drivers RX-loop anyhow, as XDP_TX and > > XDP_REDIRECT will also need to "collect" all buffers before the packet > > can be forwarded. (Although this could potentially happen later in > > driver loop when it meet/find the End-Of-Packet descriptor bit). > > A few more points (mostly thinking out loud here): > > - In any case we probably need to loop through the subsequent > descriptors in all cases, right? (i.e., if we run XDP on first > segment, and that returns DROP, the rest that are part of the packet > still need to be discarded). So we may as well delay XDP execution > until we have the whole packet? For the XDP_DROP case, drivers usually have way to discard remaining buffers/segments, to handle error cases. But it heavily depend on the driver, how tricky/convoluted this code is... Generally I would say it would make sense to delay XDP execution until all buffers/segments are "collected". It would be the clean approach, but will likely require refactoring of driver level code. > - Will this allow us to run XDP on hardware-assembled GRO super-packets? Big YES. This is usually called LRO or TSO packets. And yes, I also want to support this use-case, which is also listed in [2]. If we go down this road, this use-case is also important. (Especially related to my alloc SKBs outside drivers[3], this is a hardware offload we must support). [2] https://github.com/xdp-project/xdp-project/blob/master/areas/core/xdp-multi-buffer01-design.org [3] http://vger.kernel.org/netconf2019_files/xdp-metadata-discussion.pdf -- Best regards, Jesper Dangaard Brouer MSc.CS, Principal Kernel Engineer at Red Hat LinkedIn: http://www.linkedin.com/in/brouer