From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28]) by dpdk.org (Postfix) with ESMTP id 3A0D758EC for ; Wed, 28 Nov 2018 13:40:15 +0100 (CET) Received: from smtp.corp.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 8FD11C058CAF; Wed, 28 Nov 2018 12:40:14 +0000 (UTC) Received: from [10.36.116.109] (ovpn-116-109.ams2.redhat.com [10.36.116.109]) by smtp.corp.redhat.com (Postfix) with ESMTPS id 8417864426; Wed, 28 Nov 2018 12:40:11 +0000 (UTC) From: "Eelco Chaudron" To: "Thomas Monjalon" Cc: dev@dpdk.org, "David Marchand" , cristian.dumitrescu@intel.com Date: Wed, 28 Nov 2018 13:40:09 +0100 Message-ID: In-Reply-To: <3306361.jWiN20gDd9@xps> References: <154333204294.44971.12989297399338053044.stgit@dbuild> <3306361.jWiN20gDd9@xps> MIME-Version: 1.0 Content-Type: text/plain; format=flowed; markup=markdown X-Scanned-By: MIMEDefang 2.79 on 10.5.11.11 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.32]); Wed, 28 Nov 2018 12:40:14 +0000 (UTC) Subject: Re: [dpdk-dev] [PATCH 3/3] lib/librte_meter: update abi to include new rfc4115 function 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: Wed, 28 Nov 2018 12:40:15 -0000 On 28 Nov 2018, at 11:09, Thomas Monjalon wrote: > 28/11/2018 10:27, Eelco Chaudron: >> On 28 Nov 2018, at 9:38, David Marchand wrote: >>> On Tue, Nov 27, 2018 at 4:22 PM Eelco Chaudron >>> wrote: >>>> --- a/lib/librte_meter/Makefile >>>> +++ b/lib/librte_meter/Makefile >>>> -LIBABIVER := 2 >>>> +LIBABIVER := 3 >>> >>> As far as I understood the policy around the EXPERIMENTAL section, >>> you >>> don't need to bump the library version. >> >> Thought I would add the new function as none experimental, i.e. next >> version, but checkpatch did not allow me to do this. >> >> Tried to find info on what the right process was, as these functions >> are >> just another meter implementation using similar existing APIs. If >> anyone >> has any background on this please point me to it. > > It is documented here: > http://doc.dpdk.org/guides/contributing/versioning.html > > The case for "similar API" is not handled specifically so far. > So you need to introduce it as experimental. Thanks for the clarification, will update the APIs with __rte_experimental in the next iteration. >> I changed the library version as an existing data structure changed >> (which in theory should not change the location of the data), but the >> ABI check popped warnings so I decided to increase the version. > > It deserves to analyze why the ABI check raises a warning. > If it really needs to bump the ABI version, you should justify it > in the commit message, and explain what changed in the ABI section > of the release notes, plus update the version in the release notes. Will look at it more closely and update it for the next iteration.