From mboxrd@z Thu Jan  1 00:00:00 1970
From: Pavan Nikhilesh Bhagavatula <pbhagavatula@caviumnetworks.com>
Subject: Re: [PATCH v4 2/3] eal: add u64 bit variant for
	reciprocal
Date: Wed, 6 Sep 2017 10:02:45 +0530
Message-ID: <20170906043244.GA27242@PBHAGAVATULA-LT>
References: <1504608532-18598-1-git-send-email-pbhagavatula@caviumnetworks.com>
 <1504608532-18598-2-git-send-email-pbhagavatula@caviumnetworks.com>
 <20170905102901.0d9db417@xeon-e3>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: dev@dpdk.org
To: Stephen Hemminger <stephen@networkplumber.org>
Return-path: <dev-bounces@dpdk.org>
Received: from NAM01-BY2-obe.outbound.protection.outlook.com
 (mail-by2nam01on0063.outbound.protection.outlook.com [104.47.34.63])
 by dpdk.org (Postfix) with ESMTP id CBB883DC
 for <dev@dpdk.org>; Wed,  6 Sep 2017 06:33:18 +0200 (CEST)
Content-Disposition: inline
In-Reply-To: <20170905102901.0d9db417@xeon-e3>
List-Id: DPDK patches and discussions <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, Sep 05, 2017 at 10:29:01AM -0700, Stephen Hemminger wrote:
> On Tue,  5 Sep 2017 16:18:51 +0530
> Pavan Nikhilesh <pbhagavatula@caviumnetworks.com> wrote:
>
> > +/**
> > + * Unsigned 32-bit divisor structure.
> > + */
> > +struct rte_reciprocal_u32 {
> >  	uint32_t m;
> >  	uint8_t sh1, sh2;
> > -};
> > +} __rte_cache_aligned;
> > +
> > +/**
> > + * Unsigned 64-bit divisor structure.
> > + */
> > +struct rte_reciprocal_u64 {
> > +	uint64_t m;
> > +	uint8_t sh1;
> > +} __rte_cache_aligned;
>
> I understand you want to squeeze every cycle out but it is not
> required that each of these structures always be cache aligned.
>
> They maybe embedded in other structures and having the structure
> padded so that these elements are cache aligned would take up
> more space and make cache performance worse.
>
> Better off to not put attributes on the structure definitions, and instead
> let usages of this feature align where appropriate.
>

Agreed, will remove cache alignment in the next version (v6).

Thanks,
Pavan.