From mboxrd@z Thu Jan  1 00:00:00 1970
From: Bruce Richardson <bruce.richardson@intel.com>
Subject: Re: ixgbe_recv_scattered_pkts_vec split_flags question
Date: Tue, 21 Jul 2015 16:17:14 +0100
Message-ID: <20150721151714.GA5712@bricha3-MOBL3>
References: <55AE60C7.8050406@linaro.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: "dev@dpdk.org" <dev@dpdk.org>
To: Zoltan Kiss <zoltan.kiss@linaro.org>
Return-path: <dev-bounces@dpdk.org>
Received: from mga09.intel.com (mga09.intel.com [134.134.136.24])
 by dpdk.org (Postfix) with ESMTP id 1989C5A44
 for <dev@dpdk.org>; Tue, 21 Jul 2015 17:17:22 +0200 (CEST)
Content-Disposition: inline
In-Reply-To: <55AE60C7.8050406@linaro.org>
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, Jul 21, 2015 at 04:09:59PM +0100, Zoltan Kiss wrote:
> Hi,
> 
> I have a question regarding split_flags in this question. It's defined as an
> array of 1 byte unsigned ints:
> 
> uint8_t split_flags[RTE_IXGBE_VPMD_RX_BURST] = {0};
> 
> RTE_IXGBE_VPMD_RX_BURST is 32, so it will be 32 bytes. Then we cast it into
> a pointer for 4 byte values, and check the first 4 elements of that array
> 
> const uint32_t *split_fl32 = (uint32_t *)split_flags;
> if (rxq->pkt_first_seg == NULL &&
> 		split_fl32[0] == 0 && split_fl32[1] == 0 &&
> 		split_fl32[2] == 0 && split_fl32[3] == 0)
> 
> So we only check the first half of this 32 byte array. But
> _recv_raw_pkts_vec() seems to use the whole array. Is this a bug or a
> feature? Or am I mistaken in the math somewhere?
> 
> Regards,
> 
> Zoltan Kiss

Yes, you are right, this does look like a bug. The uint32_t's should probably be
uint64_t's. :-(

/Bruce