From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from out1-smtp.messagingengine.com (out1-smtp.messagingengine.com [66.111.4.25]) by dpdk.org (Postfix) with ESMTP id DF0EA1E862 for ; Tue, 12 Jun 2018 15:44:09 +0200 (CEST) Received: from compute1.internal (compute1.nyi.internal [10.202.2.41]) by mailout.nyi.internal (Postfix) with ESMTP id 74E61212AD; Tue, 12 Jun 2018 09:44:09 -0400 (EDT) Received: from mailfrontend2 ([10.202.2.163]) by compute1.internal (MEProxy); Tue, 12 Jun 2018 09:44:09 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=monjalon.net; h= cc:content-transfer-encoding:content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to:x-me-sender :x-me-sender:x-sasl-enc; s=mesmtp; bh=BcJIuTbHnQNpmUIUbk5idOiNJN qUHuVuPimTZEuz21w=; b=D9DDF4Ho3N4KT8OHDMtCuhy+LnXUcM5McpdJkV88M7 Z+yBQirxNFIF4QnGeEy99w5spEIdsCUcpEXOjC5nxZveBoMtyMSNJkWsOWHWcw+Q WRJSTk0wtkJe+uo/QiEmATwBAgDTAA9/K2FmgcBkhHwGNp/EQr0UluEWTtlBbra9 0= DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d= messagingengine.com; h=cc:content-transfer-encoding:content-type :date:from:in-reply-to:message-id:mime-version:references :subject:to:x-me-sender:x-me-sender:x-sasl-enc; s=fm3; bh=BcJIuT bHnQNpmUIUbk5idOiNJNqUHuVuPimTZEuz21w=; b=N9ZEUrRRQywCegVQkver7i hg/IVD/JJnnnynl97POvENuuUjspXeFmZHz/Z/Vh237QKhoxfiKV/TvgNLFTuKvX A4or84jx3oSWF1ZhBdlaUyF4YFWWDWxfwirmQXWVuRN+RELKJL03pvop7Zo8jRKo VGU3atYje8Ps5bg20fHXv7iyZgnazplNsb7tWrW1Ka/D4GtpHZ+jxEOElyl10bcQ XREh2gRLt7Jdmyagcz5nnsjTYfZzoVkVWOAhMI4GS5xSggEZUlwQodBrOwkDY3/c e0vuMpk1Gc6Pj83sApd371fHOKRjEDkZsI+F35lg5zENFiJ8vcJX9cDC8+ytx5lw == X-ME-Proxy: X-ME-Proxy: X-ME-Proxy: X-ME-Proxy: X-ME-Proxy: X-ME-Proxy: X-ME-Sender: Received: from xps.localnet (184.203.134.77.rev.sfr.net [77.134.203.184]) by mail.messagingengine.com (Postfix) with ESMTPA id 9E5291035D; Tue, 12 Jun 2018 09:44:08 -0400 (EDT) From: Thomas Monjalon To: "Wiles, Keith" Cc: dev@dpdk.org, Ophir Munk , Pascal Mazon , Olga Shern Date: Tue, 12 Jun 2018 15:44:07 +0200 Message-ID: <2475252.9oVMbiDvHd@xps> In-Reply-To: <2FAE2DE7-DB89-42F8-9010-E522187248AD@intel.com> References: <1528733172-24747-1-git-send-email-ophirmu@mellanox.com> <3694221.zYt2n70HGy@xps> <2FAE2DE7-DB89-42F8-9010-E522187248AD@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="UTF-8" Subject: Re: [dpdk-dev] [PATCH v1] net/tap: explain how to compile eBPF C file X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Jun 2018 13:44:10 -0000 12/06/2018 15:33, Wiles, Keith: >=20 > > On Jun 12, 2018, at 7:58 AM, Thomas Monjalon wrot= e: > >=20 > > 12/06/2018 14:36, Wiles, Keith: > >>=20 > >>> On Jun 12, 2018, at 7:26 AM, Thomas Monjalon wr= ote: > >>>=20 > >>> 11/06/2018 18:35, Wiles, Keith: > >>>>=20 > >>>>> On Jun 11, 2018, at 11:06 AM, Ophir Munk wro= te: > >>>>>=20 > >>>>> This commit explains how to manually compile the C source file > >>>>> tap_bpf_program.c into an ELF file using the clang compiler. > >>>>> The code in tap_bpf_program.c requires definitions found in iproute2 > >>>>> source code. This commit suggests cloning the iproute2 git tree and > >>>>> include its path in the clang command. It also adds inclusion of fi= le > >>>>> bpf_api.h (required for eBPF definitions) which is located in iprou= te2 > >>>>> source tree. For more details refer to TAP documentation. > >>>>> This commit is related to commits [1] and [2]. > >>>>=20 > >>>> Normally I would have suggested that eBPF be disable in the TAP driv= er as it requires external code and programs, but that ship has sailed. > >>>=20 > >>> The external programs are required only to generate new instructions, > >>> changing the behaviour of the BPF program. > >>> Currently, the instructions for RSS behaviour are provided. > >>>=20 > >>>> I would like to see building the tap_bpf_program.o as a target in th= e Makefile, this way the developer can just run the =E2=80=98make bpf_progr= am=E2=80=99 target and it would be simpler and less error prone. > >=20 > > As explained in the documentation, for now there is a dependency on ipr= oute2 > > for the compilation of this BPF program. > > So we cannot make it as simple as a "make command". > > Probably that we can rework it to change the dependency. > > I heard there are some good BPF libraries available now? >=20 > Well the dependence of iproute2 is really no different then requiring say= libnuma, they just have to pull the code first to type the =E2=80=98make b= pf_program=E2=80=99 right? The iproute2 dependency is different because it is not a library. The .h file is never packaged. So we need to download the sources and set -I to this directory. > If that is the case then a make target make sense to me. If iproute2 is n= ot found then an error, right? > >>> For this to happen, we need to improve the tools. > >>=20 > >> In what way do we need to improve the tools and which tools are we tal= king about. Building the .o file below appears to be a simple set of comman= d lines. I have a question in my original email about what tool. > >=20 > > The .o file is only the an intermediate file. > > The next step (numbered as 5 in this patch) is to extract the section > > of BPF instructions to be uploaded in the kernel. > > This step must be done by a "tool". Ophir did it by hacking tc, > > but it is not upstreamed yet. > > There could be other ways (possibly easier) to achieve the same result. >=20 > Please change the doc to reflect the tool is not upstreamed yet and the d= eveloper needs to figure out how to extract the data from the binary. >=20 > I used objdump -j l3_l4 -s tap_bpf_program.o and got a hex dump of the l3= _l4 section >=20 > 0000 bf160000 00000000 61681000 00000000 > ... >=20 > Someone schooled in the art of Python coding should be able to convert th= at output to a =E2=80=98C=E2=80=99 data array. :-) >=20 > >=20 > >>> It is a work in progress. > >=20 > > Contributions are welcome. > >=20 > >>> This is a very first step to use Linux BPF with DPDK. > >>> If there are more interests, we should really streamline its usage > >>> for all parts of DPDK which runs on top of some kernel code. > >>=20 > >> streamlining other parts of DPDK would be nice, but we are now talking= about the tap/eBPF patch. >=20 > Regards, > Keith >=20 >=20