From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dpdk.org (dpdk.org [92.243.14.124]) by inbox.dpdk.org (Postfix) with ESMTP id 404A9A04AC for ; Tue, 1 Sep 2020 17:11:03 +0200 (CEST) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 3931B1C0BF; Tue, 1 Sep 2020 17:11:03 +0200 (CEST) Received: from mga04.intel.com (mga04.intel.com [192.55.52.120]) by dpdk.org (Postfix) with ESMTP id 2CBC01C0BF; Tue, 1 Sep 2020 17:11:01 +0200 (CEST) IronPort-SDR: XKz2MUT+GjNlyovtg8ZEmKMXwd4PRx0GlYN1HxxS8TUzchaKISza5X11QXMn7lE8UZHp8PCRC2 t+ujyPSBpemw== X-IronPort-AV: E=McAfee;i="6000,8403,9730"; a="154585234" X-IronPort-AV: E=Sophos;i="5.76,379,1592895600"; d="scan'208";a="154585234" X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga005.jf.intel.com ([10.7.209.41]) by fmsmga104.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 01 Sep 2020 08:10:59 -0700 IronPort-SDR: 1PlYnug3mZnnoP32tjNF9oq+PPg7WEio9L58mascKfTX4jVJ+auhhgLsJuaIMXAbIY1y3mm+ZU wgQAgkRwAQmw== X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.76,379,1592895600"; d="scan'208";a="477236295" Received: from irsmsx604.ger.corp.intel.com ([163.33.146.137]) by orsmga005.jf.intel.com with ESMTP; 01 Sep 2020 08:10:58 -0700 Received: from irsmsx604.ger.corp.intel.com (163.33.146.137) by IRSMSX604.ger.corp.intel.com (163.33.146.137) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.1713.5; Tue, 1 Sep 2020 16:10:46 +0100 Received: from irsmsx604.ger.corp.intel.com ([163.33.146.137]) by IRSMSX604.ger.corp.intel.com ([163.33.146.137]) with mapi id 15.01.1713.004; Tue, 1 Sep 2020 16:10:46 +0100 From: "Stokes, Ian" To: "Pai G, Sunil" , "Richardson, Bruce" , Christian Ehrhardt CC: Luca Boccassi , "stable@dpdk.org" , dev Thread-Topic: [dpdk-dev] 19.11.4 patches review and test Thread-Index: AQHWdYslssOruqKJsESXeWFDlimf6alTeI0AgABDfwCAAARJgIAACagAgAAtNwA= Date: Tue, 1 Sep 2020 15:10:46 +0000 Message-ID: References: <20200818181222.8462-1-bluca@debian.org> <20200901124747.GB1047@bricha3-MOBL.ger.corp.intel.com> In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: dlp-reaction: no-action dlp-product: dlpe-windows dlp-version: 11.5.1.3 x-originating-ip: [163.33.253.164] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-stable] [dpdk-dev] 19.11.4 patches review and test X-BeenThere: stable@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches for DPDK stable branches List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: stable-bounces@dpdk.org Sender: "stable" > Hi, >=20 > Yes , OVS was using pkg-config even before these patches were rolled out. > But it always used to pick the DPDK shared libs by default for OVS even o= n using > the -Bstatic/-Bshared flags. > These patches from Bruce simplify the process from DPDK side without havi= ng > the user to specify them. > Moreover, with these patches , the problem of shared DPDK libs always bei= ng > picked instead of static was not seen any more with a bit of changes from= the > OVS side as well. > http://patchwork.ozlabs.org/project/openvswitch/patch/20200707141126.7141 > 4-1-sunil.pai.g@intel.com/ . > The patches for ovs-master are ready as well and will them out soon. >=20 > Thanks and Regards, > Pai G, Sunil > Sunil Hi Folks, >From the OVS side there was a question WRT why we couldn't use meson in our= Travis tests to build DPDK for OVS supporting 19.11.x, in this case it was= related to the changes described by Sunil above and hence why the backport= was requested.=20 It doesn't have a huge impact on OVS, we can still use 19.11.4 if the patch= es are removed, however I suspect these changes are also on DPDK main branc= h at this point and will be part of 20.11? So eventually when we move to 20= .11 the same issues may be encountered, maybe that preferable if we are cha= nge LTS versions. >From the OVS side we can look to upstream meson support in the dpdk-latest = branch and then integrate it when 20.11 is released, but it would be nice t= o support it before then if possible. Regards Ian >=20 > > -----Original Message----- > > From: Bruce Richardson > > Sent: Tuesday, September 1, 2020 6:18 PM > > To: Christian Ehrhardt > > Cc: Luca Boccassi ; stable@dpdk.org; dev > > ; Pai G, Sunil ; Stokes, Ian > > > > Subject: Re: [dpdk-dev] 19.11.4 patches review and test > > > > On Tue, Sep 01, 2020 at 02:32:26PM +0200, Christian Ehrhardt wrote: > > > On Tue, Sep 1, 2020 at 10:30 AM Luca Boccassi wrot= e: > > > > > > > > On Tue, 2020-08-18 at 19:12 +0100, Luca Boccassi wrote: > > > > > Hi all, > > > > > > > > > > Here is a list of patches targeted for stable release 19.11.4. > > > > > > > > > > The planned date for the final release is August 31st. > > > > > > > > > > Please help with testing and validation of your use cases and > > > > > report any issues/results with reply-all to this mail. For the > > > > > final release the fixes and reported validations will be added > > > > > to the release > > notes. > > > > > > > > > > A release candidate tarball can be found at: > > > > > > > > > > https://dpdk.org/browse/dpdk-stable/tag/?id=3Dv19.11.4-rc1 > > > > > > > > > > These patches are located at branch 19.11 of dpdk-stable repo: > > > > > https://dpdk.org/browse/dpdk-stable/ > > > > > > > > > > Thanks. > > > > > > > > > > Luca Boccassi > > > > > > > > Microsoft's regression tests are still running, delaying until > > > > Thursday the 3rd. Apologies for any inconvenience. > > > > > > Due to report on OVS failing to build I happened to find that > > > 19.11.4 has massively changed linking. > > > =3D> https://paste.ubuntu.com/p/znCRR4gpjP/ > > > > > > This was meant to be helpful for sure and I assume is around: > > > 48f7fd27f6 build/pkg-config: prevent overlinking > > > 2d1535d592 build/pkg-config: improve static linking flags > > > 9fb13a12c1 build/pkg-config: output drivers first for static build > > > 59b108d824 build/pkg-config: move pkg-config file creation > > > aea915e944 devtools: test static linkage with pkg-config > > > > > > But overlinking has effectively become underlinking now > > > https://launchpadlibrarian.net/495845224/buildlog_ubuntu-groovy-amd64= . > > > openvswitch_2.13.1-0ubuntu2~ppa1_BUILDING.txt.gz > > > > > > /usr/bin/ld: /usr/lib/gcc/x86_64-linux-gnu/10/../../../x86_64-linux- > > gnu/librte_pmd_ring.a(net_ring_rte_eth_ring.c.o): > > > in function `rte_eth_from_rings': > > > (.text+0x91c): undefined reference to `rte_vdev_init' > > > /usr/bin/ld: /usr/lib/gcc/x86_64-linux-gnu/10/../../../x86_64-linux- > > gnu/librte_pmd_ring.a(net_ring_rte_eth_ring.c.o): > > > in function `vdrvinitfn_pmd_ring_drv': > > > (.text.startup+0x28): undefined reference to `rte_vdev_register' > > > collect2: error: ld returned 1 exit status > > > > > > Also as you can see in the pastebin above, CFlags and Libs massively > > > shrunk and likely too much so. > > > > > > Given that this should be a stable release I'd ask to back out those > > > changes so that a test build creates the same .pc file as before - > > > would that make sense? > > > > > > > > Adding some OVS folks on CC. > > > > The request for backporting these actually came from the OVS side, > > since they were deemed necessary to enable OVS to switch to use DPDK > > pkg-config rather than the older approach using a makefile-based > > build. However, it appears you guys were already building OVS using > > the pkg-config file, and since this causes issues where there were none= , > backing it out seems a prudent choice. > > > > Ian, Sunil, any comments here. > > > > Regards > > /Bruce