From mboxrd@z Thu Jan  1 00:00:00 1970
From: Cyril Chemparathy <cchemparathy@ezchip.com>
Subject: Re: [PATCH v2 08/12] mempool: allow config override on
 element alignment
Date: Tue, 23 Jun 2015 13:43:04 -0700
Message-ID: <20150623134304.55c7233d@cchemparathy-ubuntu>
References: <1434999524-26528-1-git-send-email-cchemparathy@ezchip.com>
 <1434999524-26528-9-git-send-email-cchemparathy@ezchip.com>
 <2601191342CEEE43887BDE71AB97725836A1CDB8@irsmsx105.ger.corp.intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Cc: "dev@dpdk.org" <dev@dpdk.org>
To: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>
Return-path: <dev-bounces@dpdk.org>
Received: from emea01-am1-obe.outbound.protection.outlook.com
 (mail-am1on0075.outbound.protection.outlook.com [157.56.112.75])
 by dpdk.org (Postfix) with ESMTP id 3E7C9C6BA
 for <dev@dpdk.org>; Tue, 23 Jun 2015 22:43:23 +0200 (CEST)
In-Reply-To: <2601191342CEEE43887BDE71AB97725836A1CDB8@irsmsx105.ger.corp.intel.com>
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
Errors-To: dev-bounces@dpdk.org
Sender: "dev" <dev-bounces@dpdk.org>

On Tue, 23 Jun 2015 00:31:06 +0000
"Ananyev, Konstantin" <konstantin.ananyev@intel.com> wrote:

> > +#define RTE_MEMPOOL_ALIGN_MASK	(RTE_MEMPOOL_ALIGN - 1)  
> 
> I am probably a bit late with my comments, but why not make it a
> runtime decision then? I know we can't add a new parameter to
> mempool_xmem_create() without ABI breakage, but we can make some
> global variable for now, that could be setup at init time or
> something similar. 

But then, a global variable that is modified by an application _is_ a
part of the ABI, and a bad one at that.

I agree with the desire to make it runtime configurable, but I think we
should do so in the right spirit, with the appropriate interfaces, and
when we're open to changing the ABI accordingly.

Thanks
-- Cyril.