From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by smtp.lore.kernel.org (Postfix) with ESMTP id 74984C79F9F for ; Thu, 10 Sep 2026 13:56:41 +0000 (UTC) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 9ED6142ED2; Thu, 10 Sep 2026 15:53:02 +0200 (CEST) Received: from inva020.nxp.com (inva020.nxp.com [92.121.34.13]) by mails.dpdk.org (Postfix) with ESMTP id B6FFC42E7D for ; Thu, 10 Sep 2026 15:52:48 +0200 (CEST) Received: from inva020.nxp.com (localhost [127.0.0.1]) by inva020.eu-rdc02.nxp.com (Postfix) with ESMTP id 9912A1A0043; Thu, 10 Sep 2026 15:52:48 +0200 (CEST) Received: from aprdc01srsp001v.ap-rdc01.nxp.com (aprdc01srsp001v.ap-rdc01.nxp.com [165.114.16.16]) by inva020.eu-rdc02.nxp.com (Postfix) with ESMTP id 61B3B1A0072; Thu, 10 Sep 2026 15:52:48 +0200 (CEST) Received: from lsv031405.swis.in-blr01.nxp.com (lsv031405.swis.in-blr01.nxp.com [92.120.147.93]) by aprdc01srsp001v.ap-rdc01.nxp.com (Postfix) with ESMTP id 5A9CC18000B1; Thu, 10 Sep 2026 21:52:47 +0800 (+08) From: Prashant Gupta To: stephen@networkplumber.org, dev@dpdk.org Cc: Jun Yang Subject: [PATCH v2 40/47] net/dpaa2: support Rx mempool per traffic class Date: Thu, 10 Sep 2026 19:21:51 +0530 Message-ID: <20260910135158.2181141-41-prashant.gupta_3@nxp.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20260910135158.2181141-1-prashant.gupta_3@nxp.com> References: <20260903135353.3358303-1-prashant.gupta_3@nxp.com> <20260910135158.2181141-1-prashant.gupta_3@nxp.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Virus-Scanned: ClamAV using ClamSMTP X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org From: Jun Yang Support configuring RX mempool per TC. This patch isolates mempools by traffic class to avoid individual TC's performance being impacted by lack of memory. Signed-off-by: Jun Yang --- drivers/net/dpaa2/dpaa2_ethdev.c | 119 ++++++++++++++++++++++++++++++- drivers/net/dpaa2/dpaa2_ethdev.h | 2 +- 2 files changed, 119 insertions(+), 2 deletions(-) diff --git a/drivers/net/dpaa2/dpaa2_ethdev.c b/drivers/net/dpaa2/dpaa2_ethdev.c index 812e3f529f..e44c073f82 100644 --- a/drivers/net/dpaa2/dpaa2_ethdev.c +++ b/drivers/net/dpaa2/dpaa2_ethdev.c @@ -546,6 +546,101 @@ dpaa2_update_flow_rss_dist(struct rte_eth_dev *eth_dev, return ret; } +static int +dpaa2_attach_bp_list(struct dpaa2_dev_priv *priv, + struct fsl_mc_io *dpni, void *blist, uint8_t tc_id) +{ + /* Function to attach a DPNI with a buffer pool list. Buffer pool list + * handle is passed in blist. + */ + int32_t retcode; + struct dpni_pools_cfg *bpool_cfg = &priv->pools_cfg; + struct dpaa2_bp_list *bp_list = blist; + struct dpni_buffer_layout layout; + int tot_size, out_min_hdr_room, in_min_hdr_room; + uint8_t bp_idx; + struct rte_mempool *mp = bp_list->mp; + + if (priv->flow_profile.mempool[tc_id] != mp) { + if (!priv->flow_profile.mempool[tc_id]) { + bp_idx = bpool_cfg->num_dpbp; + bpool_cfg->num_dpbp++; + priv->flow_profile.bp_idx[tc_id] = bp_idx; + } else { + bp_idx = priv->flow_profile.bp_idx[tc_id]; + } + priv->flow_profile.mempool[tc_id] = mp; + } else { + bp_idx = priv->flow_profile.bp_idx[tc_id]; + } + + /* ... rx buffer layout . + * Check alignment for buffer layouts first + */ + + /* ... rx buffer layout ... */ + if (priv->tx_conf_type == DPAA2_TX_DYNAMIC_CONF) { + /** Additional headroom layout for IPSec with TX configure + * dynamic enabled. + */ + in_min_hdr_room = DPAA2_RX_MIN_FD_OFFSET + + DPAA2_SEC_SIMPLE_FD_IB_MIN; + out_min_hdr_room = DPAA2_DYN_TX_MIN_FD_OFFSET + + DPAA2_SEC_SIMPLE_FD_OB_MIN; + tot_size = RTE_MAX(in_min_hdr_room, out_min_hdr_room); + if (tot_size < RTE_PKTMBUF_HEADROOM) + tot_size = RTE_PKTMBUF_HEADROOM; + } else { + tot_size = RTE_PKTMBUF_HEADROOM; + } + tot_size = RTE_ALIGN_CEIL(tot_size, DPAA2_PACKET_LAYOUT_ALIGN); + + memset(&layout, 0, sizeof(struct dpni_buffer_layout)); + layout.options = DPNI_BUF_LAYOUT_OPT_DATA_HEAD_ROOM | + DPNI_BUF_LAYOUT_OPT_FRAME_STATUS | + DPNI_BUF_LAYOUT_OPT_PARSER_RESULT | + DPNI_BUF_LAYOUT_OPT_DATA_ALIGN | + DPNI_BUF_LAYOUT_OPT_TIMESTAMP | + DPNI_BUF_LAYOUT_OPT_PRIVATE_DATA_SIZE; + + layout.pass_timestamp = true; + layout.pass_frame_status = 1; + layout.private_data_size = DPAA2_FD_PTA_SIZE; + layout.pass_parser_result = 1; + layout.data_align = DPAA2_PACKET_LAYOUT_ALIGN; + layout.data_head_room = tot_size - DPAA2_FD_PTA_SIZE - + DPAA2_MBUF_HW_ANNOTATION; + retcode = dpni_set_buffer_layout(dpni, CMD_PRI_LOW, priv->token, + DPNI_QUEUE_RX, &layout); + if (retcode) { + DPAA2_PMD_ERR("Error configuring buffer pool Rx layout (%d)", + retcode); + return retcode; + } + + /*Attach buffer pool to the network interface as described by the user*/ + bpool_cfg->pools[bp_idx].dpbp_id = bp_list->buf_pool.dpbp_node->dpbp_id; + bpool_cfg->pools[bp_idx].backup_pool = 0; + bpool_cfg->pools[bp_idx].buffer_size = RTE_ALIGN_CEIL(bp_list->buf_pool.size, + DPAA2_PACKET_LAYOUT_ALIGN); + /* priority_mask is a bitmask of the traffic classes served by this + * buffer pool, not a traffic class index. MC treats a zero mask as + * 0xff (all traffic classes). + */ + bpool_cfg->pools[bp_idx].priority_mask = RTE_BIT32(tc_id); + + retcode = dpni_set_pools(dpni, CMD_PRI_LOW, priv->token, bpool_cfg); + if (retcode) { + DPAA2_PMD_ERR("Error(%d) configuring pools[%d](id=%d) on %s.", + retcode, bp_idx, bpool_cfg->pools[bp_idx].dpbp_id, + priv->eth_dev->data->name); + return retcode; + } + + priv->bp_list = bp_list; + return 0; +} + static int dpaa2_vlan_filter_set(struct rte_eth_dev *dev, uint16_t vlan_id, int on) { @@ -1321,7 +1416,8 @@ dpaa2_dev_rx_queue_setup(struct rte_eth_dev *dev, struct dpni_taildrop taildrop; uint8_t qopt = 0; uint16_t flow_id; - int ret; + uint32_t bpid; + int ret, ops_idx; PMD_INIT_FUNC_TRACE(); @@ -1346,6 +1442,27 @@ dpaa2_dev_rx_queue_setup(struct rte_eth_dev *dev, dev->data->rx_queues[rx_queue_id] = dpaa2_q; return 0; } + + ops_idx = rte_dpaa2_mpool_get_ops_idx(); + if (ops_idx != mb_pool->ops_index) { + DPAA2_PMD_ERR("MP(%s)'s ops index(%d) != %d", + mb_pool->name, mb_pool->ops_index, ops_idx); + return -EINVAL; + } + + if (!priv->bp_list || priv->bp_list->mp != mb_pool) { + if (rte_eal_process_type() != RTE_PROC_PRIMARY) { + ret = rte_dpaa2_bpid_info_init(mb_pool); + if (ret) + return ret; + } + bpid = mempool_to_bpid(mb_pool); + ret = dpaa2_attach_bp_list(priv, dpni, + rte_dpaa2_bpid_info[bpid].bp_list, + dpaa2_q->tc_index); + if (ret) + return ret; + } cfg = rte_zmalloc(NULL, sizeof(struct dpni_queue), 0); if (!cfg) return -ENOMEM; diff --git a/drivers/net/dpaa2/dpaa2_ethdev.h b/drivers/net/dpaa2/dpaa2_ethdev.h index 312738931f..0934e04198 100644 --- a/drivers/net/dpaa2/dpaa2_ethdev.h +++ b/drivers/net/dpaa2/dpaa2_ethdev.h @@ -405,7 +405,7 @@ struct dpaa2_dev_priv { /** RXQs in same TC share same cgid.*/ uint8_t cgid_in_use[MAX_TCS]; rte_spinlock_t meter_lock; - + struct dpni_pools_cfg pools_cfg; /* Current hash distribution size per RX TC, written by * dpaa2_setup_flow_dist_size() and read by reta_query / reta_update. * Zero means "use default" (= nb_rx_queues clamped to dist_queues). -- 2.43.0