From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Hunt, David" <david.hunt@intel.com>
Subject: Re: [PATCH v8 1/3] mempool: support external mempool
 operations
Date: Tue, 7 Jun 2016 10:25:48 +0100
Message-ID: <5756931C.70805@intel.com>
References: <1464874043-67467-1-git-send-email-david.hunt@intel.com>
 <1464965906-108927-1-git-send-email-david.hunt@intel.com>
 <1464965906-108927-2-git-send-email-david.hunt@intel.com>
 <DB5PR0401MB20545CC8494FB0965C338EE2905C0@DB5PR0401MB2054.eurprd04.prod.outlook.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "olivier.matz@6wind.com" <olivier.matz@6wind.com>,
 "viktorin@rehivetech.com" <viktorin@rehivetech.com>,
 "jerin.jacob@caviumnetworks.com" <jerin.jacob@caviumnetworks.com>
To: Shreyansh Jain <shreyansh.jain@nxp.com>, "dev@dpdk.org" <dev@dpdk.org>
Return-path: <dev-bounces@dpdk.org>
Received: from mga03.intel.com (mga03.intel.com [134.134.136.65])
 by dpdk.org (Postfix) with ESMTP id 28A6C95FD
 for <dev@dpdk.org>; Tue,  7 Jun 2016 11:25:51 +0200 (CEST)
In-Reply-To: <DB5PR0401MB20545CC8494FB0965C338EE2905C0@DB5PR0401MB2054.eurprd04.prod.outlook.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>

Hi Shreyansh,

On 6/6/2016 3:38 PM, Shreyansh Jain wrote:
> Hi,
>
> (Apologies for overly-eager email sent on this thread earlier. Will be more careful in future).
>
> This is more of a question/clarification than a comment. (And I have taken only some snippets from original mail to keep it cleaner)
>
> <snip>
>> +MEMPOOL_REGISTER_OPS(ops_mp_mc);
>> +MEMPOOL_REGISTER_OPS(ops_sp_sc);
>> +MEMPOOL_REGISTER_OPS(ops_mp_sc);
>> +MEMPOOL_REGISTER_OPS(ops_sp_mc);
> <snip>
>
>  From the above what I understand is that multiple packet pool handlers can be created.
>
> I have a use-case where application has multiple pools but only the packet pool is hardware backed. Using the hardware for general buffer requirements would prove costly.
>  From what I understand from the patch, selection of the pool is based on the flags below.

The flags are only used to select one of the default handlers for 
backward compatibility through
the rte_mempool_create call. If you wish to use a mempool handler that 
is not one of the
defaults, (i.e. a new hardware handler), you would use the 
rte_create_mempool_empty
followed by the rte_mempool_set_ops_byname call.
So, for the external handlers, you create and empty mempool, then set 
the operations (ops)
for that particular mempool.

> <snip>
>> +	/*
>> +	 * Since we have 4 combinations of the SP/SC/MP/MC examine the flags to
>> +	 * set the correct index into the table of ops structs.
>> +	 */
>> +	if (flags & (MEMPOOL_F_SP_PUT | MEMPOOL_F_SC_GET))
>> +		rte_mempool_set_ops_byname(mp, "ring_sp_sc");
>> +	else if (flags & MEMPOOL_F_SP_PUT)
>> +		rte_mempool_set_ops_byname(mp, "ring_sp_mc");
>> +	else if (flags & MEMPOOL_F_SC_GET)
>> +		rte_mempool_set_ops_byname(mp, "ring_mp_sc");
>> +	else
>> +		rte_mempool_set_ops_byname(mp, "ring_mp_mc");
>> +
> Is there any way I can achieve the above use case of multiple pools which can be selected by an application - something like a run-time toggle/flag?
>
> -
> Shreyansh

Yes, you can create multiple pools, some using the default handlers, and 
some using external handlers.
There is an example of this in the autotests (app/test/test_mempool.c). 
This test creates multiple
mempools, of which one is a custom malloc based mempool handler. The 
test puts and gets mbufs
to/from each mempool all in the same application.

Regards,
Dave.