From mboxrd@z Thu Jan  1 00:00:00 1970
From: Olivier MATZ <olivier.matz-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>
Subject: Re: [PATCH 2/2] Remove RTE_MBUF_REFCNT references
Date: Wed, 18 Feb 2015 12:01:27 +0100
Message-ID: <54E47107.3020103@6wind.com>
References: <1424102913-18944-1-git-send-email-sergio.gonzalez.monroy@intel.com>
 <1424102913-18944-3-git-send-email-sergio.gonzalez.monroy@intel.com>
 <54E45888.7070603@6wind.com> <20150218093548.GA14884@bricha3-MOBL3>
 <2601191342CEEE43887BDE71AB977258213EF5E4@irsmsx105.ger.corp.intel.com>
 <20150218100003.GA14728@bricha3-MOBL3> <54E46612.7050809@6wind.com>
 <20150218102253.GA6804@bricha3-MOBL3> <54E46A8C.9010105@6wind.com>
 <2601191342CEEE43887BDE71AB977258213EF67E@irsmsx105.ger.corp.intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "dev-VfR2kkLFssw@public.gmane.org" <dev-VfR2kkLFssw@public.gmane.org>
To: "Ananyev, Konstantin" <konstantin.ananyev-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
 "Richardson, Bruce" <bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
Return-path: <dev-bounces-VfR2kkLFssw@public.gmane.org>
In-Reply-To: <2601191342CEEE43887BDE71AB977258213EF67E-pww93C2UFcwu0RiL9chJVbfspsVTdybXVpNB7YpNyf8@public.gmane.org>
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request-VfR2kkLFssw@public.gmane.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev-VfR2kkLFssw@public.gmane.org>
List-Help: <mailto:dev-request-VfR2kkLFssw@public.gmane.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request-VfR2kkLFssw@public.gmane.org?subject=subscribe>
Errors-To: dev-bounces-VfR2kkLFssw@public.gmane.org
Sender: "dev" <dev-bounces-VfR2kkLFssw@public.gmane.org>

Hi Konstantin,

On 02/18/2015 11:47 AM, Ananyev, Konstantin wrote:
>>> How was this managed before, since refcnt field seems to be necessary in order
>>> to effectively manage indirect mbufs? Is this just the case that this is something
>>> that never worked and that needs to be solved, or is it something that was
>>> working that this patch will now break?
>>
>> This is something that never worked before: refcounts are not compatible
>> with reserving private data in mbufs. This patch does not change the
>> issue, it is still there.
>>
>> Before the patch, an application that wanted to reserve a private
>> data could disable refcounts at compile-time.
>> After the patch, the solution is just to avoid using refcounts.
>
> I'd say avoid using mbuf_attach/detach.
> refcnt itself has nothing to do with that.
> I finally understood what you  are talking about ...
> About private data - I suppose it is a matter of another patch.
> I still think it would be better to reserve private data space before mbuf, not after
> (at mbuf pool initialisation time).
> Then *BADDR* macros could be unaffected.

Indeed that could be a good idea.


Regards,
Olivier