From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 413A6A034F; Mon, 7 Jun 2021 20:07:10 +0200 (CEST) Received: from [217.70.189.124] (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 80BB441168; Mon, 7 Jun 2021 20:04:42 +0200 (CEST) Received: from mx0b-0016f401.pphosted.com (mx0b-0016f401.pphosted.com [67.231.156.173]) by mails.dpdk.org (Postfix) with ESMTP id 0138041168 for ; Mon, 7 Jun 2021 20:04:40 +0200 (CEST) Received: from pps.filterd (m0045851.ppops.net [127.0.0.1]) by mx0b-0016f401.pphosted.com (8.16.0.43/8.16.0.43) with SMTP id 157I1bsW017517 for ; Mon, 7 Jun 2021 11:04:40 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=marvell.com; h=from : to : cc : subject : date : message-id : in-reply-to : references : mime-version : content-type; s=pfpt0220; bh=UfYd7zSHvabxLlHEeCLM21+L4ZDQ6hjYArBGerDuFkg=; b=N0ocwtg6m4s18+o2lZP1aMvNA2fpVm0h/wWXqrMrQ+ZWYjlJM0cgTt/JFl1BruPA6+0g th2LT6bviZRWEtvnYqacerG21m9Ce42dpbD2eux5YH7QpbXW/3fyEHPW4Z90AAOYchn2 Stv/OWDKDA+XN5E5fgFYHyZLakdQGOVp+Uy7rb8BEtKi+x8MBTOQHFtTczmzzfPsFhl5 zfbbMfiPH+EgDTpo9tekXD2xyJsANvX29bmBFMduhYjHP70cWnNwC4n94TjRblA5iIN9 9jwfYd58cPscYkhL1irtUvyAMUFFBpr2HYF+at7vdyE1j+6PEBuwpRvFxjBPO4IBpbEs ug== Received: from dc5-exch02.marvell.com ([199.233.59.182]) by mx0b-0016f401.pphosted.com with ESMTP id 391ecv2eh3-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-SHA384 bits=256 verify=NOT) for ; Mon, 07 Jun 2021 11:04:40 -0700 Received: from DC5-EXCH01.marvell.com (10.69.176.38) by DC5-EXCH02.marvell.com (10.69.176.39) with Microsoft SMTP Server (TLS) id 15.0.1497.2; Mon, 7 Jun 2021 11:04:38 -0700 Received: from maili.marvell.com (10.69.176.80) by DC5-EXCH01.marvell.com (10.69.176.38) with Microsoft SMTP Server id 15.0.1497.2 via Frontend Transport; Mon, 7 Jun 2021 11:04:38 -0700 Received: from hyd1588t430.marvell.com (unknown [10.29.52.204]) by maili.marvell.com (Postfix) with ESMTP id 5374D3F703F; Mon, 7 Jun 2021 11:04:35 -0700 (PDT) From: Nithin Dabilpuram To: CC: , , , , , , , Nithin Dabilpuram Date: Mon, 7 Jun 2021 23:29:12 +0530 Message-ID: <20210607175943.31690-32-ndabilpuram@marvell.com> X-Mailer: git-send-email 2.8.4 In-Reply-To: <20210607175943.31690-1-ndabilpuram@marvell.com> References: <20210306153404.10781-1-ndabilpuram@marvell.com> <20210607175943.31690-1-ndabilpuram@marvell.com> MIME-Version: 1.0 Content-Type: text/plain X-Proofpoint-GUID: 0yXqcnr3B0oeJT-DujM8ftcKSqGKWVUm X-Proofpoint-ORIG-GUID: 0yXqcnr3B0oeJT-DujM8ftcKSqGKWVUm X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10434:6.0.391, 18.0.761 definitions=2021-06-07_14:2021-06-04, 2021-06-07 signatures=0 Subject: [dpdk-dev] [PATCH v2 31/62] net/cnxk: add device start and stop operations 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 Sender: "dev" Add device start and stop operation callbacks for CN9K and CN10K. Device stop is common for both platforms while device start as some platform dependent portion where the platform specific offload flags are recomputed and the right Rx/Tx burst function is chosen. Signed-off-by: Nithin Dabilpuram --- doc/guides/nics/cnxk.rst | 84 ++++++++++++++++++++++++++ drivers/net/cnxk/cn10k_ethdev.c | 124 +++++++++++++++++++++++++++++++++++++++ drivers/net/cnxk/cn9k_ethdev.c | 127 ++++++++++++++++++++++++++++++++++++++++ drivers/net/cnxk/cnxk_ethdev.c | 90 ++++++++++++++++++++++++++++ drivers/net/cnxk/cnxk_ethdev.h | 2 + drivers/net/cnxk/cnxk_link.c | 11 ++++ 6 files changed, 438 insertions(+) diff --git a/doc/guides/nics/cnxk.rst b/doc/guides/nics/cnxk.rst index 555730d..42aa7a5 100644 --- a/doc/guides/nics/cnxk.rst +++ b/doc/guides/nics/cnxk.rst @@ -39,6 +39,58 @@ Driver compilation and testing Refer to the document :ref:`compiling and testing a PMD for a NIC ` for details. +#. Running testpmd: + + Follow instructions available in the document + :ref:`compiling and testing a PMD for a NIC ` + to run testpmd. + + Example output: + + .. code-block:: console + + .//app/dpdk-testpmd -c 0xc -a 0002:02:00.0 -- --portmask=0x1 --nb-cores=1 --port-topology=loop --rxq=1 --txq=1 + EAL: Detected 4 lcore(s) + EAL: Detected 1 NUMA nodes + EAL: Multi-process socket /var/run/dpdk/rte/mp_socket + EAL: Selected IOVA mode 'VA' + EAL: No available hugepages reported in hugepages-16777216kB + EAL: No available hugepages reported in hugepages-2048kB + EAL: Probing VFIO support... + EAL: VFIO support initialized + EAL: using IOMMU type 1 (Type 1) + [ 2003.202721] vfio-pci 0002:02:00.0: vfio_cap_init: hiding cap 0x14@0x98 + EAL: Probe PCI driver: net_cn10k (177d:a063) device: 0002:02:00.0 (socket 0) + PMD: RoC Model: cn10k + EAL: No legacy callbacks, legacy socket not created + testpmd: create a new mbuf pool : n=155456, size=2176, socket=0 + testpmd: preferred mempool ops selected: cn10k_mempool_ops + Configuring Port 0 (socket 0) + PMD: Port 0: Link Up - speed 25000 Mbps - full-duplex + + Port 0: link state change event + Port 0: 96:D4:99:72:A5:BF + Checking link statuses... + Done + No commandline core given, start packet forwarding + io packet forwarding - ports=1 - cores=1 - streams=1 - NUMA support enabled, MP allocation mode: native + Logical Core 3 (socket 0) forwards packets on 1 streams: + RX P=0/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00 + + io packet forwarding packets/burst=32 + nb forwarding cores=1 - nb forwarding ports=1 + port 0: RX queue number: 1 Tx queue number: 1 + Rx offloads=0x0 Tx offloads=0x10000 + RX queue: 0 + RX desc=4096 - RX free threshold=0 + RX threshold registers: pthresh=0 hthresh=0 wthresh=0 + RX Offloads=0x0 + TX queue: 0 + TX desc=512 - TX free threshold=0 + TX threshold registers: pthresh=0 hthresh=0 wthresh=0 + TX offloads=0x0 - TX RS bit threshold=0 + Press enter to exit + Runtime Config Options ---------------------- @@ -132,3 +184,35 @@ Runtime Config Options Above devarg parameters are configurable per device, user needs to pass the parameters to all the PCIe devices if application requires to configure on all the ethdev ports. + +Limitations +----------- + +``mempool_cnxk`` external mempool handler dependency +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +The OCTEON CN9K/CN10K SoC family NIC has inbuilt HW assisted external mempool manager. +``net_cnxk`` pmd only works with ``mempool_cnxk`` mempool handler +as it is performance wise most effective way for packet allocation and Tx buffer +recycling on OCTEON TX2 SoC platform. + +CRC stripping +~~~~~~~~~~~~~ + +The OCTEON CN9K/CN10K SoC family NICs strip the CRC for every packet being received by +the host interface irrespective of the offload configuration. + +Debugging Options +----------------- + +.. _table_cnxk_ethdev_debug_options: + +.. table:: cnxk ethdev debug options + + +---+------------+-------------------------------------------------------+ + | # | Component | EAL log command | + +===+============+=======================================================+ + | 1 | NIX | --log-level='pmd\.net.cnxk,8' | + +---+------------+-------------------------------------------------------+ + | 2 | NPC | --log-level='pmd\.net.cnxk\.flow,8' | + +---+------------+-------------------------------------------------------+ diff --git a/drivers/net/cnxk/cn10k_ethdev.c b/drivers/net/cnxk/cn10k_ethdev.c index 9df30ae..fa7b835 100644 --- a/drivers/net/cnxk/cn10k_ethdev.c +++ b/drivers/net/cnxk/cn10k_ethdev.c @@ -5,6 +5,98 @@ #include "cn10k_rx.h" #include "cn10k_tx.h" +static uint16_t +nix_rx_offload_flags(struct rte_eth_dev *eth_dev) +{ + struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev); + struct rte_eth_dev_data *data = eth_dev->data; + struct rte_eth_conf *conf = &data->dev_conf; + struct rte_eth_rxmode *rxmode = &conf->rxmode; + uint16_t flags = 0; + + if (rxmode->mq_mode == ETH_MQ_RX_RSS && + (dev->rx_offloads & DEV_RX_OFFLOAD_RSS_HASH)) + flags |= NIX_RX_OFFLOAD_RSS_F; + + if (dev->rx_offloads & + (DEV_RX_OFFLOAD_TCP_CKSUM | DEV_RX_OFFLOAD_UDP_CKSUM)) + flags |= NIX_RX_OFFLOAD_CHECKSUM_F; + + if (dev->rx_offloads & + (DEV_RX_OFFLOAD_IPV4_CKSUM | DEV_RX_OFFLOAD_OUTER_IPV4_CKSUM)) + flags |= NIX_RX_OFFLOAD_CHECKSUM_F; + + if (dev->rx_offloads & DEV_RX_OFFLOAD_SCATTER) + flags |= NIX_RX_MULTI_SEG_F; + + if (!dev->ptype_disable) + flags |= NIX_RX_OFFLOAD_PTYPE_F; + + return flags; +} + +static uint16_t +nix_tx_offload_flags(struct rte_eth_dev *eth_dev) +{ + struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev); + uint64_t conf = dev->tx_offloads; + uint16_t flags = 0; + + /* Fastpath is dependent on these enums */ + RTE_BUILD_BUG_ON(PKT_TX_TCP_CKSUM != (1ULL << 52)); + RTE_BUILD_BUG_ON(PKT_TX_SCTP_CKSUM != (2ULL << 52)); + RTE_BUILD_BUG_ON(PKT_TX_UDP_CKSUM != (3ULL << 52)); + RTE_BUILD_BUG_ON(PKT_TX_IP_CKSUM != (1ULL << 54)); + RTE_BUILD_BUG_ON(PKT_TX_IPV4 != (1ULL << 55)); + RTE_BUILD_BUG_ON(PKT_TX_OUTER_IP_CKSUM != (1ULL << 58)); + RTE_BUILD_BUG_ON(PKT_TX_OUTER_IPV4 != (1ULL << 59)); + RTE_BUILD_BUG_ON(PKT_TX_OUTER_IPV6 != (1ULL << 60)); + RTE_BUILD_BUG_ON(PKT_TX_OUTER_UDP_CKSUM != (1ULL << 41)); + RTE_BUILD_BUG_ON(RTE_MBUF_L2_LEN_BITS != 7); + RTE_BUILD_BUG_ON(RTE_MBUF_L3_LEN_BITS != 9); + RTE_BUILD_BUG_ON(RTE_MBUF_OUTL2_LEN_BITS != 7); + RTE_BUILD_BUG_ON(RTE_MBUF_OUTL3_LEN_BITS != 9); + RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, data_off) != + offsetof(struct rte_mbuf, buf_iova) + 8); + RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, ol_flags) != + offsetof(struct rte_mbuf, buf_iova) + 16); + RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, pkt_len) != + offsetof(struct rte_mbuf, ol_flags) + 12); + RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, tx_offload) != + offsetof(struct rte_mbuf, pool) + 2 * sizeof(void *)); + + if (conf & DEV_TX_OFFLOAD_VLAN_INSERT || + conf & DEV_TX_OFFLOAD_QINQ_INSERT) + flags |= NIX_TX_OFFLOAD_VLAN_QINQ_F; + + if (conf & DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM || + conf & DEV_TX_OFFLOAD_OUTER_UDP_CKSUM) + flags |= NIX_TX_OFFLOAD_OL3_OL4_CSUM_F; + + if (conf & DEV_TX_OFFLOAD_IPV4_CKSUM || + conf & DEV_TX_OFFLOAD_TCP_CKSUM || + conf & DEV_TX_OFFLOAD_UDP_CKSUM || conf & DEV_TX_OFFLOAD_SCTP_CKSUM) + flags |= NIX_TX_OFFLOAD_L3_L4_CSUM_F; + + if (!(conf & DEV_TX_OFFLOAD_MBUF_FAST_FREE)) + flags |= NIX_TX_OFFLOAD_MBUF_NOFF_F; + + if (conf & DEV_TX_OFFLOAD_MULTI_SEGS) + flags |= NIX_TX_MULTI_SEG_F; + + /* Enable Inner checksum for TSO */ + if (conf & DEV_TX_OFFLOAD_TCP_TSO) + flags |= (NIX_TX_OFFLOAD_TSO_F | NIX_TX_OFFLOAD_L3_L4_CSUM_F); + + /* Enable Inner and Outer checksum for Tunnel TSO */ + if (conf & (DEV_TX_OFFLOAD_VXLAN_TNL_TSO | + DEV_TX_OFFLOAD_GENEVE_TNL_TSO | DEV_TX_OFFLOAD_GRE_TNL_TSO)) + flags |= (NIX_TX_OFFLOAD_TSO_F | NIX_TX_OFFLOAD_OL3_OL4_CSUM_F | + NIX_TX_OFFLOAD_L3_L4_CSUM_F); + + return flags; +} + static int cn10k_nix_ptypes_set(struct rte_eth_dev *eth_dev, uint32_t ptype_mask) { @@ -18,6 +110,7 @@ cn10k_nix_ptypes_set(struct rte_eth_dev *eth_dev, uint32_t ptype_mask) dev->ptype_disable = 1; } + cn10k_eth_set_rx_function(eth_dev); return 0; } @@ -162,6 +255,10 @@ cn10k_nix_configure(struct rte_eth_dev *eth_dev) if (rc) return rc; + /* Update offload flags */ + dev->rx_offload_flags = nix_rx_offload_flags(eth_dev); + dev->tx_offload_flags = nix_tx_offload_flags(eth_dev); + plt_nix_dbg("Configured port%d platform specific rx_offload_flags=%x" " tx_offload_flags=0x%x", eth_dev->data->port_id, dev->rx_offload_flags, @@ -169,6 +266,28 @@ cn10k_nix_configure(struct rte_eth_dev *eth_dev) return 0; } +static int +cn10k_nix_dev_start(struct rte_eth_dev *eth_dev) +{ + struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev); + int rc; + + /* Common eth dev start */ + rc = cnxk_nix_dev_start(eth_dev); + if (rc) + return rc; + + /* Setting up the rx[tx]_offload_flags due to change + * in rx[tx]_offloads. + */ + dev->rx_offload_flags |= nix_rx_offload_flags(eth_dev); + dev->tx_offload_flags |= nix_tx_offload_flags(eth_dev); + + cn10k_eth_set_tx_function(eth_dev); + cn10k_eth_set_rx_function(eth_dev); + return 0; +} + /* Update platform specific eth dev ops */ static void nix_eth_dev_ops_override(void) @@ -184,6 +303,7 @@ nix_eth_dev_ops_override(void) cnxk_eth_dev_ops.tx_queue_setup = cn10k_nix_tx_queue_setup; cnxk_eth_dev_ops.rx_queue_setup = cn10k_nix_rx_queue_setup; cnxk_eth_dev_ops.tx_queue_stop = cn10k_nix_tx_queue_stop; + cnxk_eth_dev_ops.dev_start = cn10k_nix_dev_start; cnxk_eth_dev_ops.dev_ptypes_set = cn10k_nix_ptypes_set; } @@ -221,6 +341,10 @@ cn10k_nix_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) eth_dev = rte_eth_dev_allocated(pci_dev->device.name); if (!eth_dev) return -ENOENT; + + /* Setup callbacks for secondary process */ + cn10k_eth_set_tx_function(eth_dev); + cn10k_eth_set_rx_function(eth_dev); } return 0; } diff --git a/drivers/net/cnxk/cn9k_ethdev.c b/drivers/net/cnxk/cn9k_ethdev.c index 22f5527..e47fa9a 100644 --- a/drivers/net/cnxk/cn9k_ethdev.c +++ b/drivers/net/cnxk/cn9k_ethdev.c @@ -5,6 +5,98 @@ #include "cn9k_rx.h" #include "cn9k_tx.h" +static uint16_t +nix_rx_offload_flags(struct rte_eth_dev *eth_dev) +{ + struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev); + struct rte_eth_dev_data *data = eth_dev->data; + struct rte_eth_conf *conf = &data->dev_conf; + struct rte_eth_rxmode *rxmode = &conf->rxmode; + uint16_t flags = 0; + + if (rxmode->mq_mode == ETH_MQ_RX_RSS && + (dev->rx_offloads & DEV_RX_OFFLOAD_RSS_HASH)) + flags |= NIX_RX_OFFLOAD_RSS_F; + + if (dev->rx_offloads & + (DEV_RX_OFFLOAD_TCP_CKSUM | DEV_RX_OFFLOAD_UDP_CKSUM)) + flags |= NIX_RX_OFFLOAD_CHECKSUM_F; + + if (dev->rx_offloads & + (DEV_RX_OFFLOAD_IPV4_CKSUM | DEV_RX_OFFLOAD_OUTER_IPV4_CKSUM)) + flags |= NIX_RX_OFFLOAD_CHECKSUM_F; + + if (dev->rx_offloads & DEV_RX_OFFLOAD_SCATTER) + flags |= NIX_RX_MULTI_SEG_F; + + if (!dev->ptype_disable) + flags |= NIX_RX_OFFLOAD_PTYPE_F; + + return flags; +} + +static uint16_t +nix_tx_offload_flags(struct rte_eth_dev *eth_dev) +{ + struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev); + uint64_t conf = dev->tx_offloads; + uint16_t flags = 0; + + /* Fastpath is dependent on these enums */ + RTE_BUILD_BUG_ON(PKT_TX_TCP_CKSUM != (1ULL << 52)); + RTE_BUILD_BUG_ON(PKT_TX_SCTP_CKSUM != (2ULL << 52)); + RTE_BUILD_BUG_ON(PKT_TX_UDP_CKSUM != (3ULL << 52)); + RTE_BUILD_BUG_ON(PKT_TX_IP_CKSUM != (1ULL << 54)); + RTE_BUILD_BUG_ON(PKT_TX_IPV4 != (1ULL << 55)); + RTE_BUILD_BUG_ON(PKT_TX_OUTER_IP_CKSUM != (1ULL << 58)); + RTE_BUILD_BUG_ON(PKT_TX_OUTER_IPV4 != (1ULL << 59)); + RTE_BUILD_BUG_ON(PKT_TX_OUTER_IPV6 != (1ULL << 60)); + RTE_BUILD_BUG_ON(PKT_TX_OUTER_UDP_CKSUM != (1ULL << 41)); + RTE_BUILD_BUG_ON(RTE_MBUF_L2_LEN_BITS != 7); + RTE_BUILD_BUG_ON(RTE_MBUF_L3_LEN_BITS != 9); + RTE_BUILD_BUG_ON(RTE_MBUF_OUTL2_LEN_BITS != 7); + RTE_BUILD_BUG_ON(RTE_MBUF_OUTL3_LEN_BITS != 9); + RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, data_off) != + offsetof(struct rte_mbuf, buf_iova) + 8); + RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, ol_flags) != + offsetof(struct rte_mbuf, buf_iova) + 16); + RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, pkt_len) != + offsetof(struct rte_mbuf, ol_flags) + 12); + RTE_BUILD_BUG_ON(offsetof(struct rte_mbuf, tx_offload) != + offsetof(struct rte_mbuf, pool) + 2 * sizeof(void *)); + + if (conf & DEV_TX_OFFLOAD_VLAN_INSERT || + conf & DEV_TX_OFFLOAD_QINQ_INSERT) + flags |= NIX_TX_OFFLOAD_VLAN_QINQ_F; + + if (conf & DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM || + conf & DEV_TX_OFFLOAD_OUTER_UDP_CKSUM) + flags |= NIX_TX_OFFLOAD_OL3_OL4_CSUM_F; + + if (conf & DEV_TX_OFFLOAD_IPV4_CKSUM || + conf & DEV_TX_OFFLOAD_TCP_CKSUM || + conf & DEV_TX_OFFLOAD_UDP_CKSUM || conf & DEV_TX_OFFLOAD_SCTP_CKSUM) + flags |= NIX_TX_OFFLOAD_L3_L4_CSUM_F; + + if (!(conf & DEV_TX_OFFLOAD_MBUF_FAST_FREE)) + flags |= NIX_TX_OFFLOAD_MBUF_NOFF_F; + + if (conf & DEV_TX_OFFLOAD_MULTI_SEGS) + flags |= NIX_TX_MULTI_SEG_F; + + /* Enable Inner checksum for TSO */ + if (conf & DEV_TX_OFFLOAD_TCP_TSO) + flags |= (NIX_TX_OFFLOAD_TSO_F | NIX_TX_OFFLOAD_L3_L4_CSUM_F); + + /* Enable Inner and Outer checksum for Tunnel TSO */ + if (conf & (DEV_TX_OFFLOAD_VXLAN_TNL_TSO | + DEV_TX_OFFLOAD_GENEVE_TNL_TSO | DEV_TX_OFFLOAD_GRE_TNL_TSO)) + flags |= (NIX_TX_OFFLOAD_TSO_F | NIX_TX_OFFLOAD_OL3_OL4_CSUM_F | + NIX_TX_OFFLOAD_L3_L4_CSUM_F); + + return flags; +} + static int cn9k_nix_ptypes_set(struct rte_eth_dev *eth_dev, uint32_t ptype_mask) { @@ -18,6 +110,7 @@ cn9k_nix_ptypes_set(struct rte_eth_dev *eth_dev, uint32_t ptype_mask) dev->ptype_disable = 1; } + cn9k_eth_set_rx_function(eth_dev); return 0; } @@ -171,6 +264,10 @@ cn9k_nix_configure(struct rte_eth_dev *eth_dev) if (rc) return rc; + /* Update offload flags */ + dev->rx_offload_flags = nix_rx_offload_flags(eth_dev); + dev->tx_offload_flags = nix_tx_offload_flags(eth_dev); + plt_nix_dbg("Configured port%d platform specific rx_offload_flags=%x" " tx_offload_flags=0x%x", eth_dev->data->port_id, dev->rx_offload_flags, @@ -178,6 +275,28 @@ cn9k_nix_configure(struct rte_eth_dev *eth_dev) return 0; } +static int +cn9k_nix_dev_start(struct rte_eth_dev *eth_dev) +{ + struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev); + int rc; + + /* Common eth dev start */ + rc = cnxk_nix_dev_start(eth_dev); + if (rc) + return rc; + + /* Setting up the rx[tx]_offload_flags due to change + * in rx[tx]_offloads. + */ + dev->rx_offload_flags |= nix_rx_offload_flags(eth_dev); + dev->tx_offload_flags |= nix_tx_offload_flags(eth_dev); + + cn9k_eth_set_tx_function(eth_dev); + cn9k_eth_set_rx_function(eth_dev); + return 0; +} + /* Update platform specific eth dev ops */ static void nix_eth_dev_ops_override(void) @@ -193,6 +312,7 @@ nix_eth_dev_ops_override(void) cnxk_eth_dev_ops.tx_queue_setup = cn9k_nix_tx_queue_setup; cnxk_eth_dev_ops.rx_queue_setup = cn9k_nix_rx_queue_setup; cnxk_eth_dev_ops.tx_queue_stop = cn9k_nix_tx_queue_stop; + cnxk_eth_dev_ops.dev_start = cn9k_nix_dev_start; cnxk_eth_dev_ops.dev_ptypes_set = cn9k_nix_ptypes_set; } @@ -232,6 +352,13 @@ cn9k_nix_probe(struct rte_pci_driver *pci_drv, struct rte_pci_device *pci_dev) if (!eth_dev) return -ENOENT; + if (rte_eal_process_type() != RTE_PROC_PRIMARY) { + /* Setup callbacks for secondary process */ + cn9k_eth_set_tx_function(eth_dev); + cn9k_eth_set_rx_function(eth_dev); + return 0; + } + dev = cnxk_eth_pmd_priv(eth_dev); /* Update capabilities already set for TSO. * TSO not supported for earlier chip revisions diff --git a/drivers/net/cnxk/cnxk_ethdev.c b/drivers/net/cnxk/cnxk_ethdev.c index 2a0ce7f..156d962 100644 --- a/drivers/net/cnxk/cnxk_ethdev.c +++ b/drivers/net/cnxk/cnxk_ethdev.c @@ -926,12 +926,102 @@ cnxk_nix_rx_queue_stop(struct rte_eth_dev *eth_dev, uint16_t qid) return rc; } +static int +cnxk_nix_dev_stop(struct rte_eth_dev *eth_dev) +{ + struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev); + const struct eth_dev_ops *dev_ops = eth_dev->dev_ops; + struct rte_mbuf *rx_pkts[32]; + int count, i, j, rc; + void *rxq; + + /* Disable switch hdr pkind */ + roc_nix_switch_hdr_set(&dev->nix, 0); + + /* Stop link change events */ + if (!roc_nix_is_vf_or_sdp(&dev->nix)) + roc_nix_mac_link_event_start_stop(&dev->nix, false); + + /* Disable Rx via NPC */ + roc_nix_npc_rx_ena_dis(&dev->nix, false); + + /* Stop rx queues and free up pkts pending */ + for (i = 0; i < eth_dev->data->nb_rx_queues; i++) { + rc = dev_ops->rx_queue_stop(eth_dev, i); + if (rc) + continue; + + rxq = eth_dev->data->rx_queues[i]; + count = dev->rx_pkt_burst_no_offload(rxq, rx_pkts, 32); + while (count) { + for (j = 0; j < count; j++) + rte_pktmbuf_free(rx_pkts[j]); + count = dev->rx_pkt_burst_no_offload(rxq, rx_pkts, 32); + } + } + + /* Stop tx queues */ + for (i = 0; i < eth_dev->data->nb_tx_queues; i++) + dev_ops->tx_queue_stop(eth_dev, i); + + return 0; +} + +int +cnxk_nix_dev_start(struct rte_eth_dev *eth_dev) +{ + struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev); + int rc, i; + + /* Start rx queues */ + for (i = 0; i < eth_dev->data->nb_rx_queues; i++) { + rc = cnxk_nix_rx_queue_start(eth_dev, i); + if (rc) + return rc; + } + + /* Start tx queues */ + for (i = 0; i < eth_dev->data->nb_tx_queues; i++) { + rc = cnxk_nix_tx_queue_start(eth_dev, i); + if (rc) + return rc; + } + + /* Enable Rx in NPC */ + rc = roc_nix_npc_rx_ena_dis(&dev->nix, true); + if (rc) { + plt_err("Failed to enable NPC rx %d", rc); + return rc; + } + + cnxk_nix_toggle_flag_link_cfg(dev, true); + + /* Start link change events */ + if (!roc_nix_is_vf_or_sdp(&dev->nix)) { + rc = roc_nix_mac_link_event_start_stop(&dev->nix, true); + if (rc) { + plt_err("Failed to start cgx link event %d", rc); + goto rx_disable; + } + } + + cnxk_nix_toggle_flag_link_cfg(dev, false); + + return 0; + +rx_disable: + roc_nix_npc_rx_ena_dis(&dev->nix, false); + cnxk_nix_toggle_flag_link_cfg(dev, false); + return rc; +} + /* CNXK platform independent eth dev ops */ struct eth_dev_ops cnxk_eth_dev_ops = { .dev_infos_get = cnxk_nix_info_get, .link_update = cnxk_nix_link_update, .tx_queue_release = cnxk_nix_tx_queue_release, .rx_queue_release = cnxk_nix_rx_queue_release, + .dev_stop = cnxk_nix_dev_stop, .tx_queue_start = cnxk_nix_tx_queue_start, .rx_queue_start = cnxk_nix_rx_queue_start, .rx_queue_stop = cnxk_nix_rx_queue_stop, diff --git a/drivers/net/cnxk/cnxk_ethdev.h b/drivers/net/cnxk/cnxk_ethdev.h index 2f31cba..984f4fe 100644 --- a/drivers/net/cnxk/cnxk_ethdev.h +++ b/drivers/net/cnxk/cnxk_ethdev.h @@ -214,6 +214,7 @@ int cnxk_nix_rx_queue_setup(struct rte_eth_dev *eth_dev, uint16_t qid, const struct rte_eth_rxconf *rx_conf, struct rte_mempool *mp); int cnxk_nix_tx_queue_stop(struct rte_eth_dev *eth_dev, uint16_t qid); +int cnxk_nix_dev_start(struct rte_eth_dev *eth_dev); uint64_t cnxk_nix_rxq_mbuf_setup(struct cnxk_eth_dev *dev); @@ -222,6 +223,7 @@ uint32_t cnxk_rss_ethdev_to_nix(struct cnxk_eth_dev *dev, uint64_t ethdev_rss, uint8_t rss_level); /* Link */ +void cnxk_nix_toggle_flag_link_cfg(struct cnxk_eth_dev *dev, bool set); void cnxk_eth_dev_link_status_cb(struct roc_nix *nix, struct roc_nix_link_info *link); int cnxk_nix_link_update(struct rte_eth_dev *eth_dev, int wait_to_complete); diff --git a/drivers/net/cnxk/cnxk_link.c b/drivers/net/cnxk/cnxk_link.c index 0223d68..8728eb1 100644 --- a/drivers/net/cnxk/cnxk_link.c +++ b/drivers/net/cnxk/cnxk_link.c @@ -4,6 +4,17 @@ #include "cnxk_ethdev.h" +void +cnxk_nix_toggle_flag_link_cfg(struct cnxk_eth_dev *dev, bool set) +{ + if (set) + dev->flags |= CNXK_LINK_CFG_IN_PROGRESS_F; + else + dev->flags &= ~CNXK_LINK_CFG_IN_PROGRESS_F; + + rte_wmb(); +} + static inline int nix_wait_for_link_cfg(struct cnxk_eth_dev *dev) { -- 2.8.4