From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dpdk.org (dpdk.org [92.243.14.124]) by inbox.dpdk.org (Postfix) with ESMTP id 0D788A0471 for ; Tue, 13 Aug 2019 09:25:02 +0200 (CEST) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 8FCBB1BDFD; Tue, 13 Aug 2019 09:25:00 +0200 (CEST) Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by dpdk.org (Postfix) with ESMTP id E6FDCDE3; Tue, 13 Aug 2019 09:24:58 +0200 (CEST) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga008.jf.intel.com ([10.7.209.65]) by orsmga102.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 13 Aug 2019 00:24:57 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.64,380,1559545200"; d="scan'208";a="170316067" Received: from fmsmsx105.amr.corp.intel.com ([10.18.124.203]) by orsmga008.jf.intel.com with ESMTP; 13 Aug 2019 00:24:57 -0700 Received: from fmsmsx156.amr.corp.intel.com (10.18.116.74) by FMSMSX105.amr.corp.intel.com (10.18.124.203) with Microsoft SMTP Server (TLS) id 14.3.439.0; Tue, 13 Aug 2019 00:24:57 -0700 Received: from shsmsx151.ccr.corp.intel.com (10.239.6.50) by fmsmsx156.amr.corp.intel.com (10.18.116.74) with Microsoft SMTP Server (TLS) id 14.3.439.0; Tue, 13 Aug 2019 00:24:56 -0700 Received: from shsmsx105.ccr.corp.intel.com ([169.254.11.15]) by SHSMSX151.ccr.corp.intel.com ([169.254.3.250]) with mapi id 14.03.0439.000; Tue, 13 Aug 2019 15:24:54 +0800 From: "Zhang, Xiao" To: "Ye, Xiaolong" CC: "dev@dpdk.org" , "Xing, Beilei" , "stable@dpdk.org" Thread-Topic: [dpdk-dev] [v5] net/i40e: fix vf runtime queues rss config Thread-Index: AQHVUXjktgs9GFsJq0KLDSwuPaH0vab4F4WAgACQVeA= Date: Tue, 13 Aug 2019 07:24:53 +0000 Message-ID: References: <1565631812-3136-1-git-send-email-xiao.zhang@intel.com> <1565692851-36962-1-git-send-email-xiao.zhang@intel.com> <20190813062815.GA10508@intel.com> In-Reply-To: <20190813062815.GA10508@intel.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.239.127.40] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [v5] net/i40e: fix vf runtime queues rss config X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" > -----Original Message----- > From: Ye, Xiaolong > Sent: Tuesday, August 13, 2019 2:28 PM > To: Zhang, Xiao > Cc: dev@dpdk.org; Xing, Beilei ; stable@dpdk.org > Subject: Re: [dpdk-dev] [v5] net/i40e: fix vf runtime queues rss config >=20 > Hi, Xiao >=20 > On 08/13, Xiao Zhang wrote: > >I40evf queue can not work properly with kernel pf driver. Eg. when > >configure 8 queues pair, only 4 queues can receive packets, and half > >packets will be lost if using 2 queues pair. > >This issue is caused by misconfiguration of look up table, use aq > >command to setup the lut to make it work properly. >=20 > So the original code of lookup table configuration is problematic? Can we= just > remove them? The original code does not work with device X722 VF. For other devices usin= g i40evf, the original code works. The commit message may missed this information. And the new code only workable for devices capable with AQ command, so we c= an not remove the original code. >=20 > Thanks, > Xiaolong >=20 > > > >Fixes: cea7a51c1750 ("i40evf: support RSS") > >Cc: stable@dpdk.org > > > >Signed-off-by: Xiao Zhang > >--- > >v5 fix compile issue > >v4 move local variable definition to the begin of the function > >v3 move LUT configuration in to i40evf_configure_rss > >v2 change for loop format to avoid build patch issue > >--- > > drivers/net/i40e/i40e_ethdev_vf.c | 32 > >++++++++++++++++++++++++++------ > > 1 file changed, 26 insertions(+), 6 deletions(-) > > > >diff --git a/drivers/net/i40e/i40e_ethdev_vf.c > >b/drivers/net/i40e/i40e_ethdev_vf.c > >index 308fb98..c77b30c 100644 > >--- a/drivers/net/i40e/i40e_ethdev_vf.c > >+++ b/drivers/net/i40e/i40e_ethdev_vf.c > >@@ -2598,7 +2598,10 @@ i40evf_config_rss(struct i40e_vf *vf) > > struct i40e_hw *hw =3D I40E_VF_TO_HW(vf); > > struct rte_eth_rss_conf rss_conf; > > uint32_t i, j, lut =3D 0, nb_q =3D (I40E_VFQF_HLUT_MAX_INDEX + 1) * 4; > >+ uint32_t rss_lut_size =3D (I40E_VFQF_HLUT1_MAX_INDEX + 1) * 4; > > uint16_t num; > >+ uint8_t *lut_info; > >+ int ret; > > > > if (vf->dev_data->dev_conf.rxmode.mq_mode !=3D ETH_MQ_RX_RSS) { > > i40evf_disable_rss(vf); > >@@ -2608,12 +2611,29 @@ i40evf_config_rss(struct i40e_vf *vf) > > > > num =3D RTE_MIN(vf->dev_data->nb_rx_queues, > I40E_MAX_QP_NUM_PER_VF); > > /* Fill out the look up table */ > >- for (i =3D 0, j =3D 0; i < nb_q; i++, j++) { > >- if (j >=3D num) > >- j =3D 0; > >- lut =3D (lut << 8) | j; > >- if ((i & 3) =3D=3D 3) > >- I40E_WRITE_REG(hw, I40E_VFQF_HLUT(i >> 2), lut); > >+ if (!(vf->flags & I40E_FLAG_RSS_AQ_CAPABLE)) { > >+ for (i =3D 0, j =3D 0; i < nb_q; i++, j++) { > >+ if (j >=3D num) > >+ j =3D 0; > >+ lut =3D (lut << 8) | j; > >+ if ((i & 3) =3D=3D 3) > >+ I40E_WRITE_REG(hw, I40E_VFQF_HLUT(i >> 2), > lut); > >+ } > >+ } else { > >+ lut_info =3D rte_zmalloc("i40e_rss_lut", rss_lut_size, 0); > >+ if (!lut_info) { > >+ PMD_DRV_LOG(ERR, "No memory can be allocated"); > >+ return -ENOMEM; > >+ } > >+ > >+ for (i =3D 0; i < rss_lut_size; i++) > >+ lut_info[i] =3D i % vf->num_queue_pairs; > >+ > >+ ret =3D i40evf_set_rss_lut(&vf->vsi, lut_info, > >+ rss_lut_size); > >+ rte_free(lut_info); > >+ if (ret) > >+ return ret; > > } > > > > rss_conf =3D vf->dev_data->dev_conf.rx_adv_conf.rss_conf; > >-- > >2.7.4 > >