* [PATCH v10 1/8] nfp: basic initialization
2015-11-30 10:25 [PATCH v10 0/8] support for netronome nfp-6xxx card Alejandro Lucero
@ 2015-11-30 10:25 ` Alejandro Lucero
2015-11-30 10:25 ` [PATCH v10 2/8] nfp: adding rx/tx functionality Alejandro Lucero
` (7 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Alejandro Lucero @ 2015-11-30 10:25 UTC (permalink / raw)
To: dev
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@netronome.com>
---
MAINTAINERS | 3 +
config/common_linuxapp | 6 +
doc/guides/rel_notes/release_2_2.rst | 3 +
drivers/net/Makefile | 1 +
drivers/net/nfp/Makefile | 56 +++
drivers/net/nfp/nfp_net.c | 699 +++++++++++++++++++++++++++++++
drivers/net/nfp/nfp_net_ctrl.h | 324 ++++++++++++++
drivers/net/nfp/nfp_net_logs.h | 75 ++++
drivers/net/nfp/nfp_net_pmd.h | 453 ++++++++++++++++++++
drivers/net/nfp/rte_pmd_nfp_version.map | 3 +
mk/rte.app.mk | 1 +
11 files changed, 1624 insertions(+)
create mode 100644 drivers/net/nfp/Makefile
create mode 100644 drivers/net/nfp/nfp_net.c
create mode 100644 drivers/net/nfp/nfp_net_ctrl.h
create mode 100644 drivers/net/nfp/nfp_net_logs.h
create mode 100644 drivers/net/nfp/nfp_net_pmd.h
create mode 100644 drivers/net/nfp/rte_pmd_nfp_version.map
diff --git a/MAINTAINERS b/MAINTAINERS
index 4478862..a23de04 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -335,6 +335,9 @@ F: drivers/crypto/aesni_mb/
Intel QuickAssist
F: drivers/crypto/qat/
+Netronome nfp
+M: Alejandro Lucero <alejandro.lucero@netronome.com>
+F: drivers/net/nfp/
Packet processing
-----------------
diff --git a/config/common_linuxapp b/config/common_linuxapp
index 2866986..82f68c7 100644
--- a/config/common_linuxapp
+++ b/config/common_linuxapp
@@ -279,6 +279,12 @@ CONFIG_RTE_LIBRTE_VMXNET3_DEBUG_TX_FREE=n
CONFIG_RTE_LIBRTE_VMXNET3_DEBUG_DRIVER=n
#
+# Compile burst-oriented Netronome NFP PMD driver
+#
+CONFIG_RTE_LIBRTE_NFP_PMD=n
+CONFIG_RTE_LIBRTE_NFP_DEBUG=n
+
+#
# Compile example software rings based PMD
#
CONFIG_RTE_LIBRTE_PMD_RING=y
diff --git a/doc/guides/rel_notes/release_2_2.rst b/doc/guides/rel_notes/release_2_2.rst
index 511d7a0..0a7c217 100644
--- a/doc/guides/rel_notes/release_2_2.rst
+++ b/doc/guides/rel_notes/release_2_2.rst
@@ -230,6 +230,9 @@ Libraries
hardware transactional memory support, thread scaling did not work,
due to the global ring that is shared by all cores.
+* **nfp: adding new PMD for Netronome nfp-6xxx card.**
+
+ Support for using Netronome nfp-6xxx with PCI VFs.
Examples
~~~~~~~~
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index cddcd57..6e4497e 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -43,6 +43,7 @@ DIRS-$(CONFIG_RTE_LIBRTE_IXGBE_PMD) += ixgbe
DIRS-$(CONFIG_RTE_LIBRTE_MLX4_PMD) += mlx4
DIRS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5
DIRS-$(CONFIG_RTE_LIBRTE_MPIPE_PMD) += mpipe
+DIRS-$(CONFIG_RTE_LIBRTE_NFP_PMD) += nfp
DIRS-$(CONFIG_RTE_LIBRTE_PMD_NULL) += null
DIRS-$(CONFIG_RTE_LIBRTE_PMD_PCAP) += pcap
DIRS-$(CONFIG_RTE_LIBRTE_PMD_RING) += ring
diff --git a/drivers/net/nfp/Makefile b/drivers/net/nfp/Makefile
new file mode 100644
index 0000000..ef7a13d
--- /dev/null
+++ b/drivers/net/nfp/Makefile
@@ -0,0 +1,56 @@
+# BSD LICENSE
+#
+# Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+# All rights reserved.
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions
+# are met:
+#
+# * Redistributions of source code must retain the above copyright
+# notice, this list of conditions and the following disclaimer.
+# * Redistributions in binary form must reproduce the above copyright
+# notice, this list of conditions and the following disclaimer in
+# the documentation and/or other materials provided with the
+# distribution.
+# * Neither the name of Intel Corporation nor the names of its
+# contributors may be used to endorse or promote products derived
+# from this software without specific prior written permission.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+include $(RTE_SDK)/mk/rte.vars.mk
+
+#
+# library name
+#
+LIB = librte_pmd_nfp.a
+
+CFLAGS += -O3
+CFLAGS += $(WERROR_FLAGS)
+
+EXPORT_MAP := rte_pmd_nfp_version.map
+
+LIBABIVER := 1
+
+#
+# all source are stored in SRCS-y
+#
+SRCS-$(CONFIG_RTE_LIBRTE_NFP_PMD) += nfp_net.c
+
+# this lib depends upon:
+DEPDIRS-$(CONFIG_RTE_LIBRTE_NFP_PMD) += lib/librte_eal lib/librte_ether
+DEPDIRS-$(CONFIG_RTE_LIBRTE_NFP_PMD) += lib/librte_mempool lib/librte_mbuf
+DEPDIRS-$(CONFIG_RTE_LIBRTE_NFP_PMD) += lib/librte_net lib/librte_malloc
+
+include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
new file mode 100644
index 0000000..b9240db
--- /dev/null
+++ b/drivers/net/nfp/nfp_net.c
@@ -0,0 +1,699 @@
+/*
+ * Copyright (c) 2014, 2015 Netronome Systems, Inc.
+ * All rights reserved.
+ *
+ * Small portions derived from code Copyright(c) 2010-2015 Intel Corporation.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright notice,
+ * this list of conditions and the following disclaimer.
+ *
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution
+ *
+ * 3. Neither the name of the copyright holder nor the names of its
+ * contributors may be used to endorse or promote products derived from this
+ * software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+/*
+ * vim:shiftwidth=8:noexpandtab
+ *
+ * @file dpdk/pmd/nfp_net.c
+ *
+ * Netronome vNIC DPDK Poll-Mode Driver: Main entry point
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <stdint.h>
+#include <unistd.h>
+#include <string.h>
+#include <sys/mman.h>
+#include <sys/socket.h>
+#include <sys/io.h>
+#include <assert.h>
+#include <time.h>
+#include <math.h>
+#include <inttypes.h>
+
+#include <rte_byteorder.h>
+#include <rte_common.h>
+#include <rte_log.h>
+#include <rte_debug.h>
+#include <rte_ethdev.h>
+#include <rte_dev.h>
+#include <rte_ether.h>
+#include <rte_malloc.h>
+#include <rte_memzone.h>
+#include <rte_mempool.h>
+#include <rte_version.h>
+#include <rte_string_fns.h>
+#include <rte_alarm.h>
+
+#include "nfp_net_pmd.h"
+#include "nfp_net_logs.h"
+#include "nfp_net_ctrl.h"
+
+/* Prototypes */
+static void nfp_net_close(struct rte_eth_dev *dev);
+static int nfp_net_configure(struct rte_eth_dev *dev);
+static int nfp_net_init(struct rte_eth_dev *eth_dev);
+static int nfp_net_start(struct rte_eth_dev *dev);
+static void nfp_net_stop(struct rte_eth_dev *dev);
+
+/*
+ * The offset of the queue controller queues in the PCIe Target. These
+ * happen to be at the same offset on the NFP6000 and the NFP3200 so
+ * we use a single macro here.
+ */
+#define NFP_PCIE_QUEUE(_q) (0x80000 + (0x800 * ((_q) & 0xff)))
+
+/* Maximum value which can be added to a queue with one transaction */
+#define NFP_QCP_MAX_ADD 0x7f
+
+#define RTE_MBUF_DMA_ADDR_DEFAULT(mb) \
+ (uint64_t)((mb)->buf_physaddr + RTE_PKTMBUF_HEADROOM)
+
+/* nfp_qcp_ptr - Read or Write Pointer of a queue */
+enum nfp_qcp_ptr {
+ NFP_QCP_READ_PTR = 0,
+ NFP_QCP_WRITE_PTR
+};
+
+/*
+ * nfp_qcp_ptr_add - Add the value to the selected pointer of a queue
+ * @q: Base address for queue structure
+ * @ptr: Add to the Read or Write pointer
+ * @val: Value to add to the queue pointer
+ *
+ * If @val is greater than @NFP_QCP_MAX_ADD multiple writes are performed.
+ */
+static inline void
+nfp_qcp_ptr_add(uint8_t *q, enum nfp_qcp_ptr ptr, uint32_t val)
+{
+ uint32_t off;
+
+ if (ptr == NFP_QCP_READ_PTR)
+ off = NFP_QCP_QUEUE_ADD_RPTR;
+ else
+ off = NFP_QCP_QUEUE_ADD_WPTR;
+
+ while (val > NFP_QCP_MAX_ADD) {
+ nn_writel(rte_cpu_to_le_32(NFP_QCP_MAX_ADD), q + off);
+ val -= NFP_QCP_MAX_ADD;
+ }
+
+ nn_writel(rte_cpu_to_le_32(val), q + off);
+}
+
+/*
+ * nfp_qcp_read - Read the current Read/Write pointer value for a queue
+ * @q: Base address for queue structure
+ * @ptr: Read or Write pointer
+ */
+static inline uint32_t
+nfp_qcp_read(uint8_t *q, enum nfp_qcp_ptr ptr)
+{
+ uint32_t off;
+ uint32_t val;
+
+ if (ptr == NFP_QCP_READ_PTR)
+ off = NFP_QCP_QUEUE_STS_LO;
+ else
+ off = NFP_QCP_QUEUE_STS_HI;
+
+ val = rte_cpu_to_le_32(nn_readl(q + off));
+
+ if (ptr == NFP_QCP_READ_PTR)
+ return val & NFP_QCP_QUEUE_STS_LO_READPTR_mask;
+ else
+ return val & NFP_QCP_QUEUE_STS_HI_WRITEPTR_mask;
+}
+
+/*
+ * Functions to read/write from/to Config BAR
+ * Performs any endian conversion necessary.
+ */
+static inline uint8_t
+nn_cfg_readb(struct nfp_net_hw *hw, int off)
+{
+ return nn_readb(hw->ctrl_bar + off);
+}
+
+static inline void
+nn_cfg_writeb(struct nfp_net_hw *hw, int off, uint8_t val)
+{
+ nn_writeb(val, hw->ctrl_bar + off);
+}
+
+static inline uint32_t
+nn_cfg_readl(struct nfp_net_hw *hw, int off)
+{
+ return rte_le_to_cpu_32(nn_readl(hw->ctrl_bar + off));
+}
+
+static inline void
+nn_cfg_writel(struct nfp_net_hw *hw, int off, uint32_t val)
+{
+ nn_writel(rte_cpu_to_le_32(val), hw->ctrl_bar + off);
+}
+
+static inline uint64_t
+nn_cfg_readq(struct nfp_net_hw *hw, int off)
+{
+ return rte_le_to_cpu_64(nn_readq(hw->ctrl_bar + off));
+}
+
+static inline void
+nn_cfg_writeq(struct nfp_net_hw *hw, int off, uint64_t val)
+{
+ nn_writeq(rte_cpu_to_le_64(val), hw->ctrl_bar + off);
+}
+
+static int
+__nfp_net_reconfig(struct nfp_net_hw *hw, uint32_t update)
+{
+ int cnt;
+ uint32_t new;
+ struct timespec wait;
+
+ PMD_DRV_LOG(DEBUG, "Writing to the configuration queue (%p)...\n",
+ hw->qcp_cfg);
+
+ if (hw->qcp_cfg == NULL)
+ rte_panic("Bad configuration queue pointer\n");
+
+ nfp_qcp_ptr_add(hw->qcp_cfg, NFP_QCP_WRITE_PTR, 1);
+
+ wait.tv_sec = 0;
+ wait.tv_nsec = 1000000;
+
+ PMD_DRV_LOG(DEBUG, "Polling for update ack...\n");
+
+ /* Poll update field, waiting for NFP to ack the config */
+ for (cnt = 0; ; cnt++) {
+ new = nn_cfg_readl(hw, NFP_NET_CFG_UPDATE);
+ if (new == 0)
+ break;
+ if (new & NFP_NET_CFG_UPDATE_ERR) {
+ PMD_INIT_LOG(ERR, "Reconfig error: 0x%08x\n", new);
+ return -1;
+ }
+ if (cnt >= NFP_NET_POLL_TIMEOUT) {
+ PMD_INIT_LOG(ERR, "Reconfig timeout for 0x%08x after"
+ " %dms\n", update, cnt);
+ rte_panic("Exiting\n");
+ }
+ nanosleep(&wait, 0); /* waiting for a 1ms */
+ }
+ PMD_DRV_LOG(DEBUG, "Ack DONE\n");
+ return 0;
+}
+
+/*
+ * Reconfigure the NIC
+ * @nn: device to reconfigure
+ * @ctrl: The value for the ctrl field in the BAR config
+ * @update: The value for the update field in the BAR config
+ *
+ * Write the update word to the BAR and ping the reconfig queue. Then poll
+ * until the firmware has acknowledged the update by zeroing the update word.
+ */
+static int
+nfp_net_reconfig(struct nfp_net_hw *hw, uint32_t ctrl, uint32_t update)
+{
+ uint32_t err;
+
+ PMD_DRV_LOG(DEBUG, "nfp_net_reconfig: ctrl=%08x update=%08x\n",
+ ctrl, update);
+
+ nn_cfg_writel(hw, NFP_NET_CFG_CTRL, ctrl);
+ nn_cfg_writel(hw, NFP_NET_CFG_UPDATE, update);
+
+ rte_wmb();
+
+ err = __nfp_net_reconfig(hw, update);
+
+ if (!err)
+ return 0;
+
+ /*
+ * Reconfig errors imply situations where they can be handled.
+ * Otherwise, rte_panic is called inside __nfp_net_reconfig
+ */
+ PMD_INIT_LOG(ERR, "Error nfp_net reconfig for ctrl: %x update: %x\n",
+ ctrl, update);
+ return -EIO;
+}
+
+/*
+ * Configure an Ethernet device. This function must be invoked first
+ * before any other function in the Ethernet API. This function can
+ * also be re-invoked when a device is in the stopped state.
+ */
+static int
+nfp_net_configure(struct rte_eth_dev *dev)
+{
+ struct rte_eth_conf *dev_conf;
+ struct rte_eth_rxmode *rxmode;
+ struct rte_eth_txmode *txmode;
+ uint32_t new_ctrl = 0;
+ uint32_t update = 0;
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ /*
+ * A DPDK app sends info about how many queues to use and how
+ * those queues need to be configured. This is used by the
+ * DPDK core and it makes sure no more queues than those
+ * advertised by the driver are requested. This function is
+ * called after that internal process
+ */
+
+ PMD_INIT_LOG(DEBUG, "Configure\n");
+
+ dev_conf = &dev->data->dev_conf;
+ rxmode = &dev_conf->rxmode;
+ txmode = &dev_conf->txmode;
+
+ /* Checking TX mode */
+ if (txmode->mq_mode) {
+ PMD_INIT_LOG(INFO, "TX mq_mode DCB and VMDq not supported\n");
+ return -EINVAL;
+ }
+
+ /* Checking RX mode */
+ if (rxmode->mq_mode & ETH_MQ_RX_RSS) {
+ if (hw->cap & NFP_NET_CFG_CTRL_RSS) {
+ update = NFP_NET_CFG_UPDATE_RSS;
+ new_ctrl = NFP_NET_CFG_CTRL_RSS;
+ } else {
+ PMD_INIT_LOG(INFO, "RSS not supported\n");
+ return -EINVAL;
+ }
+ }
+
+ if (rxmode->split_hdr_size) {
+ PMD_INIT_LOG(INFO, "rxmode does not support split header\n");
+ return -EINVAL;
+ }
+
+ if (rxmode->hw_ip_checksum) {
+ if (hw->cap & NFP_NET_CFG_CTRL_RXCSUM) {
+ new_ctrl |= NFP_NET_CFG_CTRL_RXCSUM;
+ } else {
+ PMD_INIT_LOG(INFO, "RXCSUM not supported\n");
+ return -EINVAL;
+ }
+ }
+
+ if (rxmode->hw_vlan_filter) {
+ PMD_INIT_LOG(INFO, "VLAN filter not supported\n");
+ return -EINVAL;
+ }
+
+ if (rxmode->hw_vlan_strip) {
+ if (hw->cap & NFP_NET_CFG_CTRL_RXVLAN) {
+ new_ctrl |= NFP_NET_CFG_CTRL_RXVLAN;
+ } else {
+ PMD_INIT_LOG(INFO, "hw vlan strip not supported\n");
+ return -EINVAL;
+ }
+ }
+
+ if (rxmode->hw_vlan_extend) {
+ PMD_INIT_LOG(INFO, "VLAN extended not supported\n");
+ return -EINVAL;
+ }
+
+ /* Supporting VLAN insertion by default */
+ if (hw->cap & NFP_NET_CFG_CTRL_TXVLAN)
+ new_ctrl |= NFP_NET_CFG_CTRL_TXVLAN;
+
+ if (rxmode->jumbo_frame)
+ /* this is handled in rte_eth_dev_configure */
+
+ if (rxmode->hw_strip_crc) {
+ PMD_INIT_LOG(INFO, "strip CRC not supported\n");
+ return -EINVAL;
+ }
+
+ if (rxmode->enable_scatter) {
+ PMD_INIT_LOG(INFO, "Scatter not supported\n");
+ return -EINVAL;
+ }
+
+ if (!new_ctrl)
+ return 0;
+
+ update |= NFP_NET_CFG_UPDATE_GEN;
+
+ nn_cfg_writel(hw, NFP_NET_CFG_CTRL, new_ctrl);
+ if (nfp_net_reconfig(hw, new_ctrl, update) < 0)
+ return -EIO;
+
+ hw->ctrl = new_ctrl;
+
+ return 0;
+}
+
+static void
+nfp_net_enable_queues(struct rte_eth_dev *dev)
+{
+ struct nfp_net_hw *hw;
+ uint64_t enabled_queues = 0;
+ int i;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ /* Enabling the required TX queues in the device */
+ for (i = 0; i < dev->data->nb_tx_queues; i++)
+ enabled_queues |= (1 << i);
+
+ nn_cfg_writeq(hw, NFP_NET_CFG_TXRS_ENABLE, enabled_queues);
+
+ enabled_queues = 0;
+
+ /* Enabling the required RX queues in the device */
+ for (i = 0; i < dev->data->nb_rx_queues; i++)
+ enabled_queues |= (1 << i);
+
+ nn_cfg_writeq(hw, NFP_NET_CFG_RXRS_ENABLE, enabled_queues);
+}
+
+static void
+nfp_net_disable_queues(struct rte_eth_dev *dev)
+{
+ struct nfp_net_hw *hw;
+ uint32_t new_ctrl, update = 0;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ nn_cfg_writeq(hw, NFP_NET_CFG_TXRS_ENABLE, 0);
+ nn_cfg_writeq(hw, NFP_NET_CFG_RXRS_ENABLE, 0);
+
+ new_ctrl = hw->ctrl & ~NFP_NET_CFG_CTRL_ENABLE;
+ update = NFP_NET_CFG_UPDATE_GEN | NFP_NET_CFG_UPDATE_RING |
+ NFP_NET_CFG_UPDATE_MSIX;
+
+ if (hw->cap & NFP_NET_CFG_CTRL_RINGCFG)
+ new_ctrl &= ~NFP_NET_CFG_CTRL_RINGCFG;
+
+ /* If an error when reconfig we avoid to change hw state */
+ if (nfp_net_reconfig(hw, new_ctrl, update) < 0)
+ return;
+
+ hw->ctrl = new_ctrl;
+}
+
+static void
+nfp_net_params_setup(struct nfp_net_hw *hw)
+{
+ uint32_t *mac_address;
+
+ nn_cfg_writel(hw, NFP_NET_CFG_MTU, hw->mtu);
+ nn_cfg_writel(hw, NFP_NET_CFG_FLBUFSZ, hw->flbufsz);
+
+ /* A MAC address is 8 bytes long */
+ mac_address = (uint32_t *)(hw->mac_addr);
+
+ nn_cfg_writel(hw, NFP_NET_CFG_MACADDR,
+ rte_cpu_to_be_32(*mac_address));
+ nn_cfg_writel(hw, NFP_NET_CFG_MACADDR + 4,
+ rte_cpu_to_be_32(*(mac_address + 4)));
+}
+
+static void
+nfp_net_cfg_queue_setup(struct nfp_net_hw *hw)
+{
+ hw->qcp_cfg = hw->tx_bar + NFP_QCP_QUEUE_ADDR_SZ;
+}
+
+static int
+nfp_net_start(struct rte_eth_dev *dev)
+{
+ uint32_t new_ctrl, update = 0;
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ PMD_INIT_LOG(DEBUG, "Start\n");
+
+ /* Disabling queues just in case... */
+ nfp_net_disable_queues(dev);
+
+ /* Writing configuration parameters in the device */
+ nfp_net_params_setup(hw);
+
+ /* Enabling the required queues in the device */
+ nfp_net_enable_queues(dev);
+
+ /* Enable device */
+ new_ctrl = hw->ctrl | NFP_NET_CFG_CTRL_ENABLE | NFP_NET_CFG_UPDATE_MSIX;
+ update = NFP_NET_CFG_UPDATE_GEN | NFP_NET_CFG_UPDATE_RING;
+
+ if (hw->cap & NFP_NET_CFG_CTRL_RINGCFG)
+ new_ctrl |= NFP_NET_CFG_CTRL_RINGCFG;
+
+ nn_cfg_writel(hw, NFP_NET_CFG_CTRL, new_ctrl);
+ if (nfp_net_reconfig(hw, new_ctrl, update) < 0)
+ return -EIO;
+
+ hw->ctrl = new_ctrl;
+
+ return 0;
+}
+
+/* Stop device: disable rx and tx functions to allow for reconfiguring. */
+static void
+nfp_net_stop(struct rte_eth_dev *dev)
+{
+ PMD_INIT_LOG(DEBUG, "Stop\n");
+
+ nfp_net_disable_queues(dev);
+}
+
+/* Reset and stop device. The device can not be restarted. */
+static void
+nfp_net_close(struct rte_eth_dev *dev)
+{
+ struct nfp_net_hw *hw;
+
+ PMD_INIT_LOG(DEBUG, "Close\n");
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ /*
+ * We assume that the DPDK application is stopping all the
+ * threads/queues before calling the device close function.
+ */
+
+ nfp_net_stop(dev);
+
+ nn_cfg_writeb(hw, NFP_NET_CFG_LSC, 0xff);
+
+ /*
+ * The ixgbe PMD driver disables the pcie master on the
+ * device. The i40e does not...
+ */
+}
+
+/* Initialise and register driver with DPDK Application */
+static struct eth_dev_ops nfp_net_eth_dev_ops = {
+ .dev_configure = nfp_net_configure,
+ .dev_start = nfp_net_start,
+ .dev_stop = nfp_net_stop,
+ .dev_close = nfp_net_close,
+};
+
+static int
+nfp_net_init(struct rte_eth_dev *eth_dev)
+{
+ struct rte_pci_device *pci_dev;
+ struct nfp_net_hw *hw;
+
+ uint32_t tx_bar_off, rx_bar_off;
+ uint32_t start_q;
+ int stride = 4;
+
+ PMD_INIT_FUNC_TRACE();
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private);
+
+ eth_dev->dev_ops = &nfp_net_eth_dev_ops;
+
+ /* For secondary processes, the primary has done all the work */
+ if (rte_eal_process_type() != RTE_PROC_PRIMARY)
+ return 0;
+
+ pci_dev = eth_dev->pci_dev;
+ hw->device_id = pci_dev->id.device_id;
+ hw->vendor_id = pci_dev->id.vendor_id;
+ hw->subsystem_device_id = pci_dev->id.subsystem_device_id;
+ hw->subsystem_vendor_id = pci_dev->id.subsystem_vendor_id;
+
+ PMD_INIT_LOG(DEBUG, "nfp_net: device (%u:%u) %u:%u:%u:%u\n",
+ pci_dev->id.vendor_id, pci_dev->id.device_id,
+ pci_dev->addr.domain, pci_dev->addr.bus,
+ pci_dev->addr.devid, pci_dev->addr.function);
+
+ hw->ctrl_bar = (uint8_t *)pci_dev->mem_resource[0].addr;
+ if (hw->ctrl_bar == NULL) {
+ RTE_LOG(ERR, PMD,
+ "hw->ctrl_bar is NULL. BAR0 not configured\n");
+ return -ENODEV;
+ }
+ hw->max_rx_queues = nn_cfg_readl(hw, NFP_NET_CFG_MAX_RXRINGS);
+ hw->max_tx_queues = nn_cfg_readl(hw, NFP_NET_CFG_MAX_TXRINGS);
+
+ /* Work out where in the BAR the queues start. */
+ switch (pci_dev->id.device_id) {
+ case PCI_DEVICE_ID_NFP6000_VF_NIC:
+ start_q = nn_cfg_readl(hw, NFP_NET_CFG_START_TXQ);
+ tx_bar_off = NFP_PCIE_QUEUE(start_q);
+ start_q = nn_cfg_readl(hw, NFP_NET_CFG_START_RXQ);
+ rx_bar_off = NFP_PCIE_QUEUE(start_q);
+ break;
+ default:
+ RTE_LOG(ERR, PMD, "nfp_net: no device ID matching\n");
+ return -ENODEV;
+ }
+
+ PMD_INIT_LOG(DEBUG, "tx_bar_off: 0x%08x\n", tx_bar_off);
+ PMD_INIT_LOG(DEBUG, "rx_bar_off: 0x%08x\n", rx_bar_off);
+
+ hw->tx_bar = (uint8_t *)pci_dev->mem_resource[2].addr + tx_bar_off;
+ hw->rx_bar = (uint8_t *)pci_dev->mem_resource[2].addr + rx_bar_off;
+
+ PMD_INIT_LOG(DEBUG, "ctrl_bar: %p, tx_bar: %p, rx_bar: %p\n",
+ hw->ctrl_bar, hw->tx_bar, hw->rx_bar);
+
+ nfp_net_cfg_queue_setup(hw);
+
+ /* Get some of the read-only fields from the config BAR */
+ hw->ver = nn_cfg_readl(hw, NFP_NET_CFG_VERSION);
+ hw->cap = nn_cfg_readl(hw, NFP_NET_CFG_CAP);
+ hw->max_mtu = nn_cfg_readl(hw, NFP_NET_CFG_MAX_MTU);
+ hw->mtu = hw->max_mtu;
+
+ if (NFD_CFG_MAJOR_VERSION_of(hw->ver) < 2)
+ hw->rx_offset = NFP_NET_RX_OFFSET;
+ else
+ hw->rx_offset = nn_cfg_readl(hw, NFP_NET_CFG_RX_OFFSET_ADDR);
+
+ PMD_INIT_LOG(INFO, "VER: %#x, Maximum supported MTU: %d\n",
+ hw->ver, hw->max_mtu);
+ PMD_INIT_LOG(INFO, "CAP: %#x, %s%s%s%s%s%s%s%s%s\n", hw->cap,
+ hw->cap & NFP_NET_CFG_CTRL_PROMISC ? "PROMISC " : "",
+ hw->cap & NFP_NET_CFG_CTRL_RXCSUM ? "RXCSUM " : "",
+ hw->cap & NFP_NET_CFG_CTRL_TXCSUM ? "TXCSUM " : "",
+ hw->cap & NFP_NET_CFG_CTRL_RXVLAN ? "RXVLAN " : "",
+ hw->cap & NFP_NET_CFG_CTRL_TXVLAN ? "TXVLAN " : "",
+ hw->cap & NFP_NET_CFG_CTRL_SCATTER ? "SCATTER " : "",
+ hw->cap & NFP_NET_CFG_CTRL_GATHER ? "GATHER " : "",
+ hw->cap & NFP_NET_CFG_CTRL_LSO ? "TSO " : "",
+ hw->cap & NFP_NET_CFG_CTRL_RSS ? "RSS " : "");
+
+ pci_dev = eth_dev->pci_dev;
+ hw->ctrl = 0;
+
+ hw->stride_rx = stride;
+ hw->stride_tx = stride;
+
+ PMD_INIT_LOG(INFO, "max_rx_queues: %u, max_tx_queues: %u\n",
+ hw->max_rx_queues, hw->max_tx_queues);
+
+ /* Allocating memory for mac addr */
+ eth_dev->data->mac_addrs = rte_zmalloc("mac_addr", ETHER_ADDR_LEN, 0);
+ if (eth_dev->data->mac_addrs == NULL) {
+ PMD_INIT_LOG(ERR, "Failed to space for MAC address");
+ return -ENOMEM;
+ }
+
+ /* Using random mac addresses for VFs */
+ eth_random_addr(&hw->mac_addr[0]);
+
+ /* Copying mac address to DPDK eth_dev struct */
+ ether_addr_copy(ð_dev->data->mac_addrs[0],
+ (struct ether_addr *)hw->mac_addr);
+
+ PMD_INIT_LOG(INFO, "port %d VendorID=0x%x DeviceID=0x%x "
+ "mac=%02x:%02x:%02x:%02x:%02x:%02x",
+ eth_dev->data->port_id, pci_dev->id.vendor_id,
+ pci_dev->id.device_id,
+ hw->mac_addr[0], hw->mac_addr[1], hw->mac_addr[2],
+ hw->mac_addr[3], hw->mac_addr[4], hw->mac_addr[5]);
+
+ return 0;
+}
+
+static struct rte_pci_id pci_id_nfp_net_map[] = {
+ {
+ .vendor_id = PCI_VENDOR_ID_NETRONOME,
+ .device_id = PCI_DEVICE_ID_NFP6000_PF_NIC,
+ .subsystem_vendor_id = PCI_ANY_ID,
+ .subsystem_device_id = PCI_ANY_ID,
+ },
+ {
+ .vendor_id = PCI_VENDOR_ID_NETRONOME,
+ .device_id = PCI_DEVICE_ID_NFP6000_VF_NIC,
+ .subsystem_vendor_id = PCI_ANY_ID,
+ .subsystem_device_id = PCI_ANY_ID,
+ },
+ {
+ .vendor_id = 0,
+ },
+};
+
+static struct eth_driver rte_nfp_net_pmd = {
+ {
+ .name = "rte_nfp_net_pmd",
+ .id_table = pci_id_nfp_net_map,
+ .drv_flags = RTE_PCI_DRV_NEED_MAPPING | RTE_PCI_DRV_INTR_LSC,
+ },
+ .eth_dev_init = nfp_net_init,
+ .dev_private_size = sizeof(struct nfp_net_adapter),
+};
+
+static int
+nfp_net_pmd_init(const char *name __rte_unused,
+ const char *params __rte_unused)
+{
+ PMD_INIT_FUNC_TRACE();
+ PMD_INIT_LOG(INFO, "librte_pmd_nfp_net version %s\n",
+ NFP_NET_PMD_VERSION);
+
+ rte_eth_driver_register(&rte_nfp_net_pmd);
+ return 0;
+}
+
+static struct rte_driver rte_nfp_net_driver = {
+ .type = PMD_PDEV,
+ .init = nfp_net_pmd_init,
+};
+
+PMD_REGISTER_DRIVER(rte_nfp_net_driver);
+
+/*
+ * Local variables:
+ * c-file-style: "Linux"
+ * indent-tabs-mode: t
+ * End:
+ */
diff --git a/drivers/net/nfp/nfp_net_ctrl.h b/drivers/net/nfp/nfp_net_ctrl.h
new file mode 100644
index 0000000..fce8251
--- /dev/null
+++ b/drivers/net/nfp/nfp_net_ctrl.h
@@ -0,0 +1,324 @@
+/*
+ * Copyright (c) 2014, 2015 Netronome Systems, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright notice,
+ * this list of conditions and the following disclaimer.
+ *
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution
+ *
+ * 3. Neither the name of the copyright holder nor the names of its
+ * contributors may be used to endorse or promote products derived from this
+ * software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+/*
+ * vim:shiftwidth=8:noexpandtab
+ *
+ * Netronome network device driver: Control BAR layout
+ */
+#ifndef _NFP_NET_CTRL_H_
+#define _NFP_NET_CTRL_H_
+
+/*
+ * Configuration BAR size.
+ *
+ * The configuration BAR is 8K in size, but on the NFP6000, due to
+ * THB-350, 32k needs to be reserved.
+ */
+#ifdef __NFP_IS_6000
+#define NFP_NET_CFG_BAR_SZ (32 * 1024)
+#else
+#define NFP_NET_CFG_BAR_SZ (8 * 1024)
+#endif
+
+/* Offset in Freelist buffer where packet starts on RX */
+#define NFP_NET_RX_OFFSET 32
+
+/* Hash type pre-pended when a RSS hash was computed */
+#define NFP_NET_RSS_NONE 0
+#define NFP_NET_RSS_IPV4 1
+#define NFP_NET_RSS_IPV6 2
+#define NFP_NET_RSS_IPV6_EX 3
+#define NFP_NET_RSS_IPV4_TCP 4
+#define NFP_NET_RSS_IPV6_TCP 5
+#define NFP_NET_RSS_IPV6_EX_TCP 6
+#define NFP_NET_RSS_IPV4_UDP 7
+#define NFP_NET_RSS_IPV6_UDP 8
+#define NFP_NET_RSS_IPV6_EX_UDP 9
+
+/*
+ * @NFP_NET_TXR_MAX: Maximum number of TX rings
+ * @NFP_NET_TXR_MASK: Mask for TX rings
+ * @NFP_NET_RXR_MAX: Maximum number of RX rings
+ * @NFP_NET_RXR_MASK: Mask for RX rings
+ */
+#define NFP_NET_TXR_MAX 64
+#define NFP_NET_TXR_MASK (NFP_NET_TXR_MAX - 1)
+#define NFP_NET_RXR_MAX 64
+#define NFP_NET_RXR_MASK (NFP_NET_RXR_MAX - 1)
+
+/*
+ * Read/Write config words (0x0000 - 0x002c)
+ * @NFP_NET_CFG_CTRL: Global control
+ * @NFP_NET_CFG_UPDATE: Indicate which fields are updated
+ * @NFP_NET_CFG_TXRS_ENABLE: Bitmask of enabled TX rings
+ * @NFP_NET_CFG_RXRS_ENABLE: Bitmask of enabled RX rings
+ * @NFP_NET_CFG_MTU: Set MTU size
+ * @NFP_NET_CFG_FLBUFSZ: Set freelist buffer size (must be larger than MTU)
+ * @NFP_NET_CFG_EXN: MSI-X table entry for exceptions
+ * @NFP_NET_CFG_LSC: MSI-X table entry for link state changes
+ * @NFP_NET_CFG_MACADDR: MAC address
+ *
+ * TODO:
+ * - define Error details in UPDATE
+ */
+#define NFP_NET_CFG_CTRL 0x0000
+#define NFP_NET_CFG_CTRL_ENABLE (0x1 << 0) /* Global enable */
+#define NFP_NET_CFG_CTRL_PROMISC (0x1 << 1) /* Enable Promisc mode */
+#define NFP_NET_CFG_CTRL_L2BC (0x1 << 2) /* Allow L2 Broadcast */
+#define NFP_NET_CFG_CTRL_L2MC (0x1 << 3) /* Allow L2 Multicast */
+#define NFP_NET_CFG_CTRL_RXCSUM (0x1 << 4) /* Enable RX Checksum */
+#define NFP_NET_CFG_CTRL_TXCSUM (0x1 << 5) /* Enable TX Checksum */
+#define NFP_NET_CFG_CTRL_RXVLAN (0x1 << 6) /* Enable VLAN strip */
+#define NFP_NET_CFG_CTRL_TXVLAN (0x1 << 7) /* Enable VLAN insert */
+#define NFP_NET_CFG_CTRL_SCATTER (0x1 << 8) /* Scatter DMA */
+#define NFP_NET_CFG_CTRL_GATHER (0x1 << 9) /* Gather DMA */
+#define NFP_NET_CFG_CTRL_LSO (0x1 << 10) /* LSO/TSO */
+#define NFP_NET_CFG_CTRL_RINGCFG (0x1 << 16) /* Ring runtime changes */
+#define NFP_NET_CFG_CTRL_RSS (0x1 << 17) /* RSS */
+#define NFP_NET_CFG_CTRL_IRQMOD (0x1 << 18) /* Interrupt moderation */
+#define NFP_NET_CFG_CTRL_RINGPRIO (0x1 << 19) /* Ring priorities */
+#define NFP_NET_CFG_CTRL_MSIXAUTO (0x1 << 20) /* MSI-X auto-masking */
+#define NFP_NET_CFG_CTRL_TXRWB (0x1 << 21) /* Write-back of TX ring*/
+#define NFP_NET_CFG_CTRL_L2SWITCH (0x1 << 22) /* L2 Switch */
+#define NFP_NET_CFG_CTRL_L2SWITCH_LOCAL (0x1 << 23) /* Switch to local */
+#define NFP_NET_CFG_CTRL_VXLAN (0x1 << 24) /* Enable VXLAN */
+#define NFP_NET_CFG_CTRL_NVGRE (0x1 << 25) /* Enable NVGRE */
+#define NFP_NET_CFG_UPDATE 0x0004
+#define NFP_NET_CFG_UPDATE_GEN (0x1 << 0) /* General update */
+#define NFP_NET_CFG_UPDATE_RING (0x1 << 1) /* Ring config change */
+#define NFP_NET_CFG_UPDATE_RSS (0x1 << 2) /* RSS config change */
+#define NFP_NET_CFG_UPDATE_TXRPRIO (0x1 << 3) /* TX Ring prio change */
+#define NFP_NET_CFG_UPDATE_RXRPRIO (0x1 << 4) /* RX Ring prio change */
+#define NFP_NET_CFG_UPDATE_MSIX (0x1 << 5) /* MSI-X change */
+#define NFP_NET_CFG_UPDATE_L2SWITCH (0x1 << 6) /* Switch changes */
+#define NFP_NET_CFG_UPDATE_RESET (0x1 << 7) /* Update due to FLR */
+#define NFP_NET_CFG_UPDATE_IRQMOD (0x1 << 8) /* IRQ mod change */
+#define NFP_NET_CFG_UPDATE_VXLAN (0x1 << 9) /* VXLAN port change */
+#define NFP_NET_CFG_UPDATE_ERR (0x1 << 31) /* A error occurred */
+#define NFP_NET_CFG_TXRS_ENABLE 0x0008
+#define NFP_NET_CFG_RXRS_ENABLE 0x0010
+#define NFP_NET_CFG_MTU 0x0018
+#define NFP_NET_CFG_FLBUFSZ 0x001c
+#define NFP_NET_CFG_EXN 0x001f
+#define NFP_NET_CFG_LSC 0x0020
+#define NFP_NET_CFG_MACADDR 0x0024
+
+/*
+ * Read-only words (0x0030 - 0x0050):
+ * @NFP_NET_CFG_VERSION: Firmware version number
+ * @NFP_NET_CFG_STS: Status
+ * @NFP_NET_CFG_CAP: Capabilities (same bits as @NFP_NET_CFG_CTRL)
+ * @NFP_NET_MAX_TXRINGS: Maximum number of TX rings
+ * @NFP_NET_MAX_RXRINGS: Maximum number of RX rings
+ * @NFP_NET_MAX_MTU: Maximum support MTU
+ * @NFP_NET_CFG_START_TXQ: Start Queue Control Queue to use for TX (PF only)
+ * @NFP_NET_CFG_START_RXQ: Start Queue Control Queue to use for RX (PF only)
+ *
+ * TODO:
+ * - define more STS bits
+ */
+#define NFP_NET_CFG_VERSION 0x0030
+#define NFP_NET_CFG_VERSION_RESERVED_MASK (0xff << 24)
+#define NFP_NET_CFG_VERSION_CLASS_MASK (0xff << 16)
+#define NFP_NET_CFG_VERSION_CLASS(x) (((x) & 0xff) << 16)
+#define NFP_NET_CFG_VERSION_CLASS_GENERIC 0
+#define NFP_NET_CFG_VERSION_MAJOR_MASK (0xff << 8)
+#define NFP_NET_CFG_VERSION_MAJOR(x) (((x) & 0xff) << 8)
+#define NFP_NET_CFG_VERSION_MINOR_MASK (0xff << 0)
+#define NFP_NET_CFG_VERSION_MINOR(x) (((x) & 0xff) << 0)
+#define NFP_NET_CFG_STS 0x0034
+#define NFP_NET_CFG_STS_LINK (0x1 << 0) /* Link up or down */
+#define NFP_NET_CFG_CAP 0x0038
+#define NFP_NET_CFG_MAX_TXRINGS 0x003c
+#define NFP_NET_CFG_MAX_RXRINGS 0x0040
+#define NFP_NET_CFG_MAX_MTU 0x0044
+/* Next two words are being used by VFs for solving THB350 issue */
+#define NFP_NET_CFG_START_TXQ 0x0048
+#define NFP_NET_CFG_START_RXQ 0x004c
+
+/*
+ * NFP-3200 workaround (0x0050 - 0x0058)
+ * @NFP_NET_CFG_SPARE_ADDR: DMA address for ME code to use (e.g. YDS-155 fix)
+ */
+#define NFP_NET_CFG_SPARE_ADDR 0x0050
+/**
+ * NFP6000/NFP4000 - Prepend configuration
+ */
+#define NFP_NET_CFG_RX_OFFSET 0x0050
+#define NFP_NET_CFG_RX_OFFSET_DYNAMIC 0 /* Prepend mode */
+
+/**
+ * Reuse spare address to contain the offset from the start of
+ * the host buffer where the first byte of the received frame
+ * will land. Any metadata will come prior to that offset. If the
+ * value in this field is 0, it means that that the metadata will
+ * always land starting at the first byte of the host buffer and
+ * packet data will immediately follow the metadata. As always,
+ * the RX descriptor indicates the presence or absence of metadata
+ * along with the length thereof.
+ */
+#define NFP_NET_CFG_RX_OFFSET_ADDR 0x0050
+
+#define NFP_NET_CFG_VXLAN_PORT 0x0060
+#define NFP_NET_CFG_VXLAN_SZ 0x0008
+
+/* Offload definitions */
+#define NFP_NET_N_VXLAN_PORTS (NFP_NET_CFG_VXLAN_SZ / sizeof(uint16_t))
+
+/**
+ * 64B reserved for future use (0x0080 - 0x00c0)
+ */
+#define NFP_NET_CFG_RESERVED 0x0080
+#define NFP_NET_CFG_RESERVED_SZ 0x0040
+
+/*
+ * RSS configuration (0x0100 - 0x01ac):
+ * Used only when NFP_NET_CFG_CTRL_RSS is enabled
+ * @NFP_NET_CFG_RSS_CFG: RSS configuration word
+ * @NFP_NET_CFG_RSS_KEY: RSS "secret" key
+ * @NFP_NET_CFG_RSS_ITBL: RSS indirection table
+ */
+#define NFP_NET_CFG_RSS_BASE 0x0100
+#define NFP_NET_CFG_RSS_CTRL NFP_NET_CFG_RSS_BASE
+#define NFP_NET_CFG_RSS_MASK (0x7f)
+#define NFP_NET_CFG_RSS_MASK_of(_x) ((_x) & 0x7f)
+#define NFP_NET_CFG_RSS_IPV4 (1 << 8) /* RSS for IPv4 */
+#define NFP_NET_CFG_RSS_IPV6 (1 << 9) /* RSS for IPv6 */
+#define NFP_NET_CFG_RSS_IPV4_TCP (1 << 10) /* RSS for IPv4/TCP */
+#define NFP_NET_CFG_RSS_IPV4_UDP (1 << 11) /* RSS for IPv4/UDP */
+#define NFP_NET_CFG_RSS_IPV6_TCP (1 << 12) /* RSS for IPv6/TCP */
+#define NFP_NET_CFG_RSS_IPV6_UDP (1 << 13) /* RSS for IPv6/UDP */
+#define NFP_NET_CFG_RSS_TOEPLITZ (1 << 24) /* Use Toeplitz hash */
+#define NFP_NET_CFG_RSS_KEY (NFP_NET_CFG_RSS_BASE + 0x4)
+#define NFP_NET_CFG_RSS_KEY_SZ 0x28
+#define NFP_NET_CFG_RSS_ITBL (NFP_NET_CFG_RSS_BASE + 0x4 + \
+ NFP_NET_CFG_RSS_KEY_SZ)
+#define NFP_NET_CFG_RSS_ITBL_SZ 0x80
+
+/*
+ * TX ring configuration (0x200 - 0x800)
+ * @NFP_NET_CFG_TXR_BASE: Base offset for TX ring configuration
+ * @NFP_NET_CFG_TXR_ADDR: Per TX ring DMA address (8B entries)
+ * @NFP_NET_CFG_TXR_WB_ADDR: Per TX ring write back DMA address (8B entries)
+ * @NFP_NET_CFG_TXR_SZ: Per TX ring ring size (1B entries)
+ * @NFP_NET_CFG_TXR_VEC: Per TX ring MSI-X table entry (1B entries)
+ * @NFP_NET_CFG_TXR_PRIO: Per TX ring priority (1B entries)
+ * @NFP_NET_CFG_TXR_IRQ_MOD: Per TX ring interrupt moderation (4B entries)
+ */
+#define NFP_NET_CFG_TXR_BASE 0x0200
+#define NFP_NET_CFG_TXR_ADDR(_x) (NFP_NET_CFG_TXR_BASE + ((_x) * 0x8))
+#define NFP_NET_CFG_TXR_WB_ADDR(_x) (NFP_NET_CFG_TXR_BASE + 0x200 + \
+ ((_x) * 0x8))
+#define NFP_NET_CFG_TXR_SZ(_x) (NFP_NET_CFG_TXR_BASE + 0x400 + (_x))
+#define NFP_NET_CFG_TXR_VEC(_x) (NFP_NET_CFG_TXR_BASE + 0x440 + (_x))
+#define NFP_NET_CFG_TXR_PRIO(_x) (NFP_NET_CFG_TXR_BASE + 0x480 + (_x))
+#define NFP_NET_CFG_TXR_IRQ_MOD(_x) (NFP_NET_CFG_TXR_BASE + 0x500 + \
+ ((_x) * 0x4))
+
+/*
+ * RX ring configuration (0x0800 - 0x0c00)
+ * @NFP_NET_CFG_RXR_BASE: Base offset for RX ring configuration
+ * @NFP_NET_CFG_RXR_ADDR: Per TX ring DMA address (8B entries)
+ * @NFP_NET_CFG_RXR_SZ: Per TX ring ring size (1B entries)
+ * @NFP_NET_CFG_RXR_VEC: Per TX ring MSI-X table entry (1B entries)
+ * @NFP_NET_CFG_RXR_PRIO: Per TX ring priority (1B entries)
+ * @NFP_NET_CFG_RXR_IRQ_MOD: Per TX ring interrupt moderation (4B entries)
+ */
+#define NFP_NET_CFG_RXR_BASE 0x0800
+#define NFP_NET_CFG_RXR_ADDR(_x) (NFP_NET_CFG_RXR_BASE + ((_x) * 0x8))
+#define NFP_NET_CFG_RXR_SZ(_x) (NFP_NET_CFG_RXR_BASE + 0x200 + (_x))
+#define NFP_NET_CFG_RXR_VEC(_x) (NFP_NET_CFG_RXR_BASE + 0x240 + (_x))
+#define NFP_NET_CFG_RXR_PRIO(_x) (NFP_NET_CFG_RXR_BASE + 0x280 + (_x))
+#define NFP_NET_CFG_RXR_IRQ_MOD(_x) (NFP_NET_CFG_RXR_BASE + 0x300 + \
+ ((_x) * 0x4))
+
+/*
+ * Interrupt Control/Cause registers (0x0c00 - 0x0d00)
+ * These registers are only used when MSI-X auto-masking is not
+ * enabled (@NFP_NET_CFG_CTRL_MSIXAUTO not set). The array is index
+ * by MSI-X entry and are 1B in size. If an entry is zero, the
+ * corresponding entry is enabled. If the FW generates an interrupt,
+ * it writes a cause into the corresponding field. This also masks
+ * the MSI-X entry and the host driver must clear the register to
+ * re-enable the interrupt.
+ */
+#define NFP_NET_CFG_ICR_BASE 0x0c00
+#define NFP_NET_CFG_ICR(_x) (NFP_NET_CFG_ICR_BASE + (_x))
+#define NFP_NET_CFG_ICR_UNMASKED 0x0
+#define NFP_NET_CFG_ICR_RXTX 0x1
+#define NFP_NET_CFG_ICR_LSC 0x2
+
+/*
+ * General device stats (0x0d00 - 0x0d90)
+ * all counters are 64bit.
+ */
+#define NFP_NET_CFG_STATS_BASE 0x0d00
+#define NFP_NET_CFG_STATS_RX_DISCARDS (NFP_NET_CFG_STATS_BASE + 0x00)
+#define NFP_NET_CFG_STATS_RX_ERRORS (NFP_NET_CFG_STATS_BASE + 0x08)
+#define NFP_NET_CFG_STATS_RX_OCTETS (NFP_NET_CFG_STATS_BASE + 0x10)
+#define NFP_NET_CFG_STATS_RX_UC_OCTETS (NFP_NET_CFG_STATS_BASE + 0x18)
+#define NFP_NET_CFG_STATS_RX_MC_OCTETS (NFP_NET_CFG_STATS_BASE + 0x20)
+#define NFP_NET_CFG_STATS_RX_BC_OCTETS (NFP_NET_CFG_STATS_BASE + 0x28)
+#define NFP_NET_CFG_STATS_RX_FRAMES (NFP_NET_CFG_STATS_BASE + 0x30)
+#define NFP_NET_CFG_STATS_RX_MC_FRAMES (NFP_NET_CFG_STATS_BASE + 0x38)
+#define NFP_NET_CFG_STATS_RX_BC_FRAMES (NFP_NET_CFG_STATS_BASE + 0x40)
+
+#define NFP_NET_CFG_STATS_TX_DISCARDS (NFP_NET_CFG_STATS_BASE + 0x48)
+#define NFP_NET_CFG_STATS_TX_ERRORS (NFP_NET_CFG_STATS_BASE + 0x50)
+#define NFP_NET_CFG_STATS_TX_OCTETS (NFP_NET_CFG_STATS_BASE + 0x58)
+#define NFP_NET_CFG_STATS_TX_UC_OCTETS (NFP_NET_CFG_STATS_BASE + 0x60)
+#define NFP_NET_CFG_STATS_TX_MC_OCTETS (NFP_NET_CFG_STATS_BASE + 0x68)
+#define NFP_NET_CFG_STATS_TX_BC_OCTETS (NFP_NET_CFG_STATS_BASE + 0x70)
+#define NFP_NET_CFG_STATS_TX_FRAMES (NFP_NET_CFG_STATS_BASE + 0x78)
+#define NFP_NET_CFG_STATS_TX_MC_FRAMES (NFP_NET_CFG_STATS_BASE + 0x80)
+#define NFP_NET_CFG_STATS_TX_BC_FRAMES (NFP_NET_CFG_STATS_BASE + 0x88)
+
+/*
+ * Per ring stats (0x1000 - 0x1800)
+ * options, 64bit per entry
+ * @NFP_NET_CFG_TXR_STATS: TX ring statistics (Packet and Byte count)
+ * @NFP_NET_CFG_RXR_STATS: RX ring statistics (Packet and Byte count)
+ */
+#define NFP_NET_CFG_TXR_STATS_BASE 0x1000
+#define NFP_NET_CFG_TXR_STATS(_x) (NFP_NET_CFG_TXR_STATS_BASE + \
+ ((_x) * 0x10))
+#define NFP_NET_CFG_RXR_STATS_BASE 0x1400
+#define NFP_NET_CFG_RXR_STATS(_x) (NFP_NET_CFG_RXR_STATS_BASE + \
+ ((_x) * 0x10))
+
+#endif /* _NFP_NET_CTRL_H_ */
+/*
+ * Local variables:
+ * c-file-style: "Linux"
+ * indent-tabs-mode: t
+ * End:
+ */
diff --git a/drivers/net/nfp/nfp_net_logs.h b/drivers/net/nfp/nfp_net_logs.h
new file mode 100644
index 0000000..0b966e4
--- /dev/null
+++ b/drivers/net/nfp/nfp_net_logs.h
@@ -0,0 +1,75 @@
+/*
+ * Copyright (c) 2014, 2015 Netronome Systems, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright notice,
+ * this list of conditions and the following disclaimer.
+ *
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution
+ *
+ * 3. Neither the name of the copyright holder nor the names of its
+ * contributors may be used to endorse or promote products derived from this
+ * software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef _NFP_NET_LOGS_H_
+#define _NFP_NET_LOGS_H_
+
+#include <rte_log.h>
+
+#define RTE_LIBRTE_NFP_NET_DEBUG_INIT 1
+
+#ifdef RTE_LIBRTE_NFP_NET_DEBUG_INIT
+#define PMD_INIT_LOG(level, fmt, args...) \
+ RTE_LOG(level, PMD, "%s(): " fmt "\n", __func__, ## args)
+#define PMD_INIT_FUNC_TRACE() PMD_INIT_LOG(DEBUG, " >>")
+#else
+#define PMD_INIT_LOG(level, fmt, args...) do { } while (0)
+#define PMD_INIT_FUNC_TRACE() do { } while (0)
+#endif
+
+#ifdef RTE_LIBRTE_NFP_NET_DEBUG_RX
+#define PMD_RX_LOG(level, fmt, args...) \
+ RTE_LOG(level, PMD, "%s() rx: " fmt, __func__, ## args)
+#else
+#define PMD_RX_LOG(level, fmt, args...) do { } while (0)
+#endif
+
+#ifdef RTE_LIBRTE_NFP_NET_DEBUG_TX
+#define PMD_TX_LOG(level, fmt, args...) \
+ RTE_LOG(level, PMD, "%s() tx: " fmt, __func__, ## args)
+#else
+#define PMD_TX_LOG(level, fmt, args...) do { } while (0)
+#endif
+
+#ifdef RTE_LIBRTE_NFP_NET_DEBUG_DRIVER
+#define PMD_DRV_LOG(level, fmt, args...) \
+ RTE_LOG(level, PMD, "%s(): " fmt, __func__, ## args)
+#else
+#define PMD_DRV_LOG(level, fmt, args...) do { } while (0)
+#endif
+
+#ifdef RTE_LIBRTE_NFP_NET_DEBUG_INIT
+#define ASSERT(x) if (!(x)) rte_panic("NFP_NET: x")
+#else
+#define ASSERT(x) do { } while (0)
+#endif
+
+#endif /* _NFP_NET_LOGS_H_ */
diff --git a/drivers/net/nfp/nfp_net_pmd.h b/drivers/net/nfp/nfp_net_pmd.h
new file mode 100644
index 0000000..a7f9386
--- /dev/null
+++ b/drivers/net/nfp/nfp_net_pmd.h
@@ -0,0 +1,453 @@
+/*
+ * Copyright (c) 2014, 2015 Netronome Systems, Inc.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright notice,
+ * this list of conditions and the following disclaimer.
+ *
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution
+ *
+ * 3. Neither the name of the copyright holder nor the names of its
+ * contributors may be used to endorse or promote products derived from this
+ * software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+/*
+ * vim:shiftwidth=8:noexpandtab
+ *
+ * @file dpdk/pmd/nfp_net_pmd.h
+ *
+ * Netronome NFP_NET PDM driver
+ */
+
+#ifndef _NFP_NET_PMD_H_
+#define _NFP_NET_PMD_H_
+
+#define NFP_NET_PMD_VERSION "0.1"
+#define PCI_VENDOR_ID_NETRONOME 0x19ee
+#define PCI_DEVICE_ID_NFP6000_PF_NIC 0x6000
+#define PCI_DEVICE_ID_NFP6000_VF_NIC 0x6003
+
+/* Forward declaration */
+struct nfp_net_adapter;
+
+/*
+ * The maximum number of descriptors is limited by design as
+ * DPDK uses uint16_t variables for these values
+ */
+#define NFP_NET_MAX_TX_DESC (32 * 1024)
+#define NFP_NET_MIN_TX_DESC 64
+
+#define NFP_NET_MAX_RX_DESC (32 * 1024)
+#define NFP_NET_MIN_RX_DESC 64
+
+/* Bar allocation */
+#define NFP_NET_CRTL_BAR 0
+#define NFP_NET_TX_BAR 2
+#define NFP_NET_RX_BAR 2
+
+/* Macros for accessing the Queue Controller Peripheral 'CSRs' */
+#define NFP_QCP_QUEUE_OFF(_x) ((_x) * 0x800)
+#define NFP_QCP_QUEUE_ADD_RPTR 0x0000
+#define NFP_QCP_QUEUE_ADD_WPTR 0x0004
+#define NFP_QCP_QUEUE_STS_LO 0x0008
+#define NFP_QCP_QUEUE_STS_LO_READPTR_mask (0x3ffff)
+#define NFP_QCP_QUEUE_STS_HI 0x000c
+#define NFP_QCP_QUEUE_STS_HI_WRITEPTR_mask (0x3ffff)
+
+/* Interrupt definitions */
+#define NFP_NET_IRQ_LSC_IDX 0
+
+#define RTE_MBUF_DATA_DMA_ADDR(mb) \
+ ((uint64_t)((mb)->buf_physaddr + (mb)->data_off))
+
+/* Default values for RX/TX configuration */
+#define DEFAULT_RX_FREE_THRESH 32
+#define DEFAULT_RX_PTHRESH 8
+#define DEFAULT_RX_HTHRESH 8
+#define DEFAULT_RX_WTHRESH 0
+
+#define DEFAULT_TX_RS_THRESH 32
+#define DEFAULT_TX_FREE_THRESH 32
+#define DEFAULT_TX_PTHRESH 32
+#define DEFAULT_TX_HTHRESH 0
+#define DEFAULT_TX_WTHRESH 0
+#define DEFAULT_TX_RSBIT_THRESH 32
+
+/* Alignment for dma zones */
+#define NFP_MEMZONE_ALIGN 128
+
+/*
+ * This is used by the reconfig protocol. It sets the maximum time waiting in
+ * milliseconds before a reconfig timeout happens.
+ */
+#define NFP_NET_POLL_TIMEOUT 5000
+
+#define NFP_QCP_QUEUE_ADDR_SZ (0x800)
+
+#define NFP_NET_LINK_DOWN_CHECK_TIMEOUT 4000 /* ms */
+#define NFP_NET_LINK_UP_CHECK_TIMEOUT 1000 /* ms */
+
+/* Version number helper defines */
+#define NFD_CFG_CLASS_VER_msk 0xff
+#define NFD_CFG_CLASS_VER_shf 24
+#define NFD_CFG_CLASS_VER(x) (((x) & 0xff) << 24)
+#define NFD_CFG_CLASS_VER_of(x) (((x) >> 24) & 0xff)
+#define NFD_CFG_CLASS_TYPE_msk 0xff
+#define NFD_CFG_CLASS_TYPE_shf 16
+#define NFD_CFG_CLASS_TYPE(x) (((x) & 0xff) << 16)
+#define NFD_CFG_CLASS_TYPE_of(x) (((x) >> 16) & 0xff)
+#define NFD_CFG_MAJOR_VERSION_msk 0xff
+#define NFD_CFG_MAJOR_VERSION_shf 8
+#define NFD_CFG_MAJOR_VERSION(x) (((x) & 0xff) << 8)
+#define NFD_CFG_MAJOR_VERSION_of(x) (((x) >> 8) & 0xff)
+#define NFD_CFG_MINOR_VERSION_msk 0xff
+#define NFD_CFG_MINOR_VERSION_shf 0
+#define NFD_CFG_MINOR_VERSION(x) (((x) & 0xff) << 0)
+#define NFD_CFG_MINOR_VERSION_of(x) (((x) >> 0) & 0xff)
+
+#include <linux/types.h>
+
+static inline uint8_t nn_readb(volatile const void *addr)
+{
+ return *((volatile const uint8_t *)(addr));
+}
+
+static inline void nn_writeb(uint8_t val, volatile void *addr)
+{
+ *((volatile uint8_t *)(addr)) = val;
+}
+
+static inline uint32_t nn_readl(volatile const void *addr)
+{
+ return *((volatile const uint32_t *)(addr));
+}
+
+static inline void nn_writel(uint32_t val, volatile void *addr)
+{
+ *((volatile uint32_t *)(addr)) = val;
+}
+
+static inline uint64_t nn_readq(volatile void *addr)
+{
+ const volatile uint32_t *p = addr;
+ uint32_t low, high;
+
+ high = nn_readl((volatile const void *)(p + 1));
+ low = nn_readl((volatile const void *)p);
+
+ return low + ((uint64_t)high << 32);
+}
+
+static inline void nn_writeq(uint64_t val, volatile void *addr)
+{
+ nn_writel(val >> 32, (volatile char *)addr + 4);
+ nn_writel(val, addr);
+}
+
+/* TX descriptor format */
+#define PCIE_DESC_TX_EOP (1 << 7)
+#define PCIE_DESC_TX_OFFSET_MASK (0x7f)
+
+/* Flags in the host TX descriptor */
+#define PCIE_DESC_TX_CSUM (1 << 7)
+#define PCIE_DESC_TX_IP4_CSUM (1 << 6)
+#define PCIE_DESC_TX_TCP_CSUM (1 << 5)
+#define PCIE_DESC_TX_UDP_CSUM (1 << 4)
+#define PCIE_DESC_TX_VLAN (1 << 3)
+#define PCIE_DESC_TX_LSO (1 << 2)
+#define PCIE_DESC_TX_ENCAP_NONE (0)
+#define PCIE_DESC_TX_ENCAP_VXLAN (1 << 1)
+#define PCIE_DESC_TX_ENCAP_GRE (1 << 0)
+
+struct nfp_net_tx_desc {
+ union {
+ struct {
+ uint8_t dma_addr_hi; /* High bits of host buf address */
+ __le16 dma_len; /* Length to DMA for this desc */
+ uint8_t offset_eop; /* Offset in buf where pkt starts +
+ * highest bit is eop flag.
+ */
+ __le32 dma_addr_lo; /* Low 32bit of host buf addr */
+
+ __le16 lso; /* MSS to be used for LSO */
+ uint8_t l4_offset; /* LSO, where the L4 data starts */
+ uint8_t flags; /* TX Flags, see @PCIE_DESC_TX_* */
+
+ __le16 vlan; /* VLAN tag to add if indicated */
+ __le16 data_len; /* Length of frame + meta data */
+ } __attribute__((__packed__));
+ __le32 vals[4];
+ };
+};
+
+struct nfp_net_txq {
+ struct nfp_net_hw *hw; /* Backpointer to nfp_net structure */
+
+ /*
+ * Queue information: @qidx is the queue index from Linux's
+ * perspective. @tx_qcidx is the index of the Queue
+ * Controller Peripheral queue relative to the TX queue BAR.
+ * @cnt is the size of the queue in number of
+ * descriptors. @qcp_q is a pointer to the base of the queue
+ * structure on the NFP
+ */
+ uint8_t *qcp_q;
+
+ /*
+ * Read and Write pointers. @wr_p and @rd_p are host side pointer,
+ * they are free running and have little relation to the QCP pointers *
+ * @qcp_rd_p is a local copy queue controller peripheral read pointer
+ */
+
+ uint32_t wr_p;
+ uint32_t rd_p;
+ uint32_t qcp_rd_p;
+
+ uint32_t tx_count;
+
+ uint32_t tx_free_thresh;
+ uint32_t tail;
+
+ /*
+ * For each descriptor keep a reference to the mbuff and
+ * DMA address used until completion is signalled.
+ */
+ struct {
+ struct rte_mbuf *mbuf;
+ } *txbufs;
+
+ /*
+ * Information about the host side queue location. @txds is
+ * the virtual address for the queue, @dma is the DMA address
+ * of the queue and @size is the size in bytes for the queue
+ * (needed for free)
+ */
+ struct nfp_net_tx_desc *txds;
+
+ /*
+ * At this point 56 bytes have been used for all the fields in the
+ * TX critical path. We have room for 8 bytes and still all placed
+ * in a cache line. We are not using the threshold values below nor
+ * the txq_flags but if we need to, we can add the most used in the
+ * remaining bytes.
+ */
+ uint32_t tx_rs_thresh; /* not used by now. Future? */
+ uint32_t tx_pthresh; /* not used by now. Future? */
+ uint32_t tx_hthresh; /* not used by now. Future? */
+ uint32_t tx_wthresh; /* not used by now. Future? */
+ uint32_t txq_flags; /* not used by now. Future? */
+ uint8_t port_id;
+ int qidx;
+ int tx_qcidx;
+ __le64 dma;
+} __attribute__ ((__aligned__(64)));
+
+/* RX and freelist descriptor format */
+#define PCIE_DESC_RX_DD (1 << 7)
+#define PCIE_DESC_RX_META_LEN_MASK (0x7f)
+
+/* Flags in the RX descriptor */
+#define PCIE_DESC_RX_RSS (1 << 15)
+#define PCIE_DESC_RX_I_IP4_CSUM (1 << 14)
+#define PCIE_DESC_RX_I_IP4_CSUM_OK (1 << 13)
+#define PCIE_DESC_RX_I_TCP_CSUM (1 << 12)
+#define PCIE_DESC_RX_I_TCP_CSUM_OK (1 << 11)
+#define PCIE_DESC_RX_I_UDP_CSUM (1 << 10)
+#define PCIE_DESC_RX_I_UDP_CSUM_OK (1 << 9)
+#define PCIE_DESC_RX_INGRESS_PORT (1 << 8)
+#define PCIE_DESC_RX_EOP (1 << 7)
+#define PCIE_DESC_RX_IP4_CSUM (1 << 6)
+#define PCIE_DESC_RX_IP4_CSUM_OK (1 << 5)
+#define PCIE_DESC_RX_TCP_CSUM (1 << 4)
+#define PCIE_DESC_RX_TCP_CSUM_OK (1 << 3)
+#define PCIE_DESC_RX_UDP_CSUM (1 << 2)
+#define PCIE_DESC_RX_UDP_CSUM_OK (1 << 1)
+#define PCIE_DESC_RX_VLAN (1 << 0)
+
+struct nfp_net_rx_desc {
+ union {
+ /* Freelist descriptor */
+ struct {
+ uint8_t dma_addr_hi;
+ __le16 spare;
+ uint8_t dd;
+
+ __le32 dma_addr_lo;
+ } __attribute__((__packed__)) fld;
+
+ /* RX descriptor */
+ struct {
+ __le16 data_len;
+ uint8_t reserved;
+ uint8_t meta_len_dd;
+
+ __le16 flags;
+ __le16 vlan;
+ } __attribute__((__packed__)) rxd;
+
+ __le32 vals[2];
+ };
+};
+
+struct nfp_net_rx_buff {
+ struct rte_mbuf *mbuf;
+};
+
+struct nfp_net_rxq {
+ struct nfp_net_hw *hw; /* Backpointer to nfp_net structure */
+
+ /*
+ * @qcp_fl and @qcp_rx are pointers to the base addresses of the
+ * freelist and RX queue controller peripheral queue structures on the
+ * NFP
+ */
+ uint8_t *qcp_fl;
+ uint8_t *qcp_rx;
+
+ /*
+ * Read and Write pointers. @wr_p and @rd_p are host side
+ * pointer, they are free running and have little relation to
+ * the QCP pointers. @wr_p is where the driver adds new
+ * freelist descriptors and @rd_p is where the driver start
+ * reading descriptors for newly arrive packets from.
+ */
+ uint32_t wr_p;
+ uint32_t rd_p;
+
+ /*
+ * For each buffer placed on the freelist, record the
+ * associated SKB
+ */
+ struct nfp_net_rx_buff *rxbufs;
+
+ /*
+ * Information about the host side queue location. @rxds is
+ * the virtual address for the queue
+ */
+ struct nfp_net_rx_desc *rxds;
+
+ /*
+ * The mempool is created by the user specifying a mbuf size.
+ * We save here the reference of the mempool needed in the RX
+ * path and the mbuf size for checking received packets can be
+ * safely copied to the mbuf using the NFP_NET_RX_OFFSET
+ */
+ struct rte_mempool *mem_pool;
+ uint16_t mbuf_size;
+
+ /*
+ * Next two fields are used for giving more free descriptors
+ * to the NFP
+ */
+ uint16_t rx_free_thresh;
+ uint16_t nb_rx_hold;
+
+ /* the size of the queue in number of descriptors */
+ uint16_t rx_count;
+
+ /*
+ * Fields above this point fit in a single cache line and are all used
+ * in the RX critical path. Fields below this point are just used
+ * during queue configuration or not used at all (yet)
+ */
+
+ /* referencing dev->data->port_id */
+ uint16_t port_id;
+
+ uint8_t crc_len; /* Not used by now */
+ uint8_t drop_en; /* Not used by now */
+
+ /* DMA address of the queue */
+ __le64 dma;
+
+ /*
+ * Queue information: @qidx is the queue index from Linux's
+ * perspective. @fl_qcidx is the index of the Queue
+ * Controller peripheral queue relative to the RX queue BAR
+ * used for the freelist and @rx_qcidx is the Queue Controller
+ * Peripheral index for the RX queue.
+ */
+ int qidx;
+ int fl_qcidx;
+ int rx_qcidx;
+} __attribute__ ((__aligned__(64)));
+
+struct nfp_net_hw {
+ /* Info from the firmware */
+ uint32_t ver;
+ uint32_t cap;
+ uint32_t max_mtu;
+ uint32_t mtu;
+ uint32_t rx_offset;
+
+ /* Current values for control */
+ uint32_t ctrl;
+
+ uint8_t *ctrl_bar;
+ uint8_t *tx_bar;
+ uint8_t *rx_bar;
+
+ int stride_rx;
+ int stride_tx;
+
+ uint8_t *qcp_cfg;
+
+ uint32_t max_tx_queues;
+ uint32_t max_rx_queues;
+ uint16_t flbufsz;
+ uint16_t device_id;
+ uint16_t vendor_id;
+ uint16_t subsystem_device_id;
+ uint16_t subsystem_vendor_id;
+#if defined(DSTQ_SELECTION)
+#if DSTQ_SELECTION
+ uint16_t device_function;
+#endif
+#endif
+
+ uint8_t mac_addr[ETHER_ADDR_LEN];
+
+ /* Records starting point for counters */
+ struct rte_eth_stats eth_stats_base;
+
+#ifdef NFP_NET_LIBNFP
+ struct nfp_cpp *cpp;
+ struct nfp_cpp_area *ctrl_area;
+ struct nfp_cpp_area *tx_area;
+ struct nfp_cpp_area *rx_area;
+ struct nfp_cpp_area *msix_area;
+#endif
+};
+
+struct nfp_net_adapter {
+ struct nfp_net_hw hw;
+};
+
+#define NFP_NET_DEV_PRIVATE_TO_HW(adapter)\
+ (&((struct nfp_net_adapter *)adapter)->hw)
+
+#endif /* _NFP_NET_PMD_H_ */
+/*
+ * Local variables:
+ * c-file-style: "Linux"
+ * indent-tabs-mode: t
+ * End:
+ */
diff --git a/drivers/net/nfp/rte_pmd_nfp_version.map b/drivers/net/nfp/rte_pmd_nfp_version.map
new file mode 100644
index 0000000..ad607bb
--- /dev/null
+++ b/drivers/net/nfp/rte_pmd_nfp_version.map
@@ -0,0 +1,3 @@
+DPDK_2.2 {
+ local: *;
+};
diff --git a/mk/rte.app.mk b/mk/rte.app.mk
index 85a680d..3a95367 100644
--- a/mk/rte.app.mk
+++ b/mk/rte.app.mk
@@ -146,6 +146,7 @@ _LDLIBS-$(CONFIG_RTE_LIBRTE_IXGBE_PMD) += -lrte_pmd_ixgbe
_LDLIBS-$(CONFIG_RTE_LIBRTE_E1000_PMD) += -lrte_pmd_e1000
_LDLIBS-$(CONFIG_RTE_LIBRTE_MLX4_PMD) += -lrte_pmd_mlx4
_LDLIBS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += -lrte_pmd_mlx5
+_LDLIBS-$(CONFIG_RTE_LIBRTE_NFP_PMD) += -lrte_pmd_nfp
_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_SZEDATA2) += -lrte_pmd_szedata2
_LDLIBS-$(CONFIG_RTE_LIBRTE_MPIPE_PMD) += -lrte_pmd_mpipe -lgxio
_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_RING) += -lrte_pmd_ring
--
1.7.9.5
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v10 2/8] nfp: adding rx/tx functionality
2015-11-30 10:25 [PATCH v10 0/8] support for netronome nfp-6xxx card Alejandro Lucero
2015-11-30 10:25 ` [PATCH v10 1/8] nfp: basic initialization Alejandro Lucero
@ 2015-11-30 10:25 ` Alejandro Lucero
2015-11-30 10:25 ` [PATCH v10 3/8] nfp: adding rss Alejandro Lucero
` (6 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Alejandro Lucero @ 2015-11-30 10:25 UTC (permalink / raw)
To: dev
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@netronome.com>
---
drivers/net/nfp/nfp_net.c | 993 +++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 993 insertions(+)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index b9240db..0d85fa4 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -74,8 +74,25 @@
static void nfp_net_close(struct rte_eth_dev *dev);
static int nfp_net_configure(struct rte_eth_dev *dev);
static int nfp_net_init(struct rte_eth_dev *eth_dev);
+static int nfp_net_rx_fill_freelist(struct nfp_net_rxq *rxq);
+static uint32_t nfp_net_rx_queue_count(struct rte_eth_dev *dev,
+ uint16_t queue_idx);
+static uint16_t nfp_net_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
+ uint16_t nb_pkts);
+static void nfp_net_rx_queue_release(void *rxq);
+static int nfp_net_rx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
+ uint16_t nb_desc, unsigned int socket_id,
+ const struct rte_eth_rxconf *rx_conf,
+ struct rte_mempool *mp);
+static int nfp_net_tx_free_bufs(struct nfp_net_txq *txq);
+static void nfp_net_tx_queue_release(void *txq);
+static int nfp_net_tx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
+ uint16_t nb_desc, unsigned int socket_id,
+ const struct rte_eth_txconf *tx_conf);
static int nfp_net_start(struct rte_eth_dev *dev);
static void nfp_net_stop(struct rte_eth_dev *dev);
+static uint16_t nfp_net_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
+ uint16_t nb_pkts);
/*
* The offset of the queue controller queues in the PCIe Target. These
@@ -186,6 +203,100 @@ nn_cfg_writeq(struct nfp_net_hw *hw, int off, uint64_t val)
nn_writeq(rte_cpu_to_le_64(val), hw->ctrl_bar + off);
}
+/* Creating memzone for hardware rings. */
+static const struct rte_memzone *
+ring_dma_zone_reserve(struct rte_eth_dev *dev, const char *ring_name,
+ uint16_t queue_id, uint32_t ring_size, int socket_id)
+{
+ char z_name[RTE_MEMZONE_NAMESIZE];
+ const struct rte_memzone *mz;
+
+ snprintf(z_name, sizeof(z_name), "%s_%s_%d_%d",
+ dev->driver->pci_drv.name,
+ ring_name, dev->data->port_id, queue_id);
+
+ mz = rte_memzone_lookup(z_name);
+ if (mz)
+ return mz;
+
+ return rte_memzone_reserve_aligned(z_name, ring_size, socket_id, 0,
+ NFP_MEMZONE_ALIGN);
+}
+
+static void
+nfp_net_rx_queue_release_mbufs(struct nfp_net_rxq *rxq)
+{
+ unsigned i;
+
+ if (rxq->rxbufs == NULL)
+ return;
+
+ for (i = 0; i < rxq->rx_count; i++) {
+ if (rxq->rxbufs[i].mbuf) {
+ rte_pktmbuf_free_seg(rxq->rxbufs[i].mbuf);
+ rxq->rxbufs[i].mbuf = NULL;
+ }
+ }
+}
+
+static void
+nfp_net_rx_queue_release(void *rx_queue)
+{
+ struct nfp_net_rxq *rxq = rx_queue;
+
+ if (rxq) {
+ nfp_net_rx_queue_release_mbufs(rxq);
+ rte_free(rxq->rxbufs);
+ rte_free(rxq);
+ }
+}
+
+static void
+nfp_net_reset_rx_queue(struct nfp_net_rxq *rxq)
+{
+ nfp_net_rx_queue_release_mbufs(rxq);
+ rxq->wr_p = 0;
+ rxq->rd_p = 0;
+ rxq->nb_rx_hold = 0;
+}
+
+static void
+nfp_net_tx_queue_release_mbufs(struct nfp_net_txq *txq)
+{
+ unsigned i;
+
+ if (txq->txbufs == NULL)
+ return;
+
+ for (i = 0; i < txq->tx_count; i++) {
+ if (txq->txbufs[i].mbuf) {
+ rte_pktmbuf_free_seg(txq->txbufs[i].mbuf);
+ txq->txbufs[i].mbuf = NULL;
+ }
+ }
+}
+
+static void
+nfp_net_tx_queue_release(void *tx_queue)
+{
+ struct nfp_net_txq *txq = tx_queue;
+
+ if (txq) {
+ nfp_net_tx_queue_release_mbufs(txq);
+ rte_free(txq->txbufs);
+ rte_free(txq);
+ }
+}
+
+static void
+nfp_net_reset_tx_queue(struct nfp_net_txq *txq)
+{
+ nfp_net_tx_queue_release_mbufs(txq);
+ txq->wr_p = 0;
+ txq->rd_p = 0;
+ txq->tail = 0;
+}
+
static int
__nfp_net_reconfig(struct nfp_net_hw *hw, uint32_t update)
{
@@ -423,6 +534,18 @@ nfp_net_disable_queues(struct rte_eth_dev *dev)
hw->ctrl = new_ctrl;
}
+static int
+nfp_net_rx_freelist_setup(struct rte_eth_dev *dev)
+{
+ int i;
+
+ for (i = 0; i < dev->data->nb_rx_queues; i++) {
+ if (nfp_net_rx_fill_freelist(dev->data->rx_queues[i]) < 0)
+ return -1;
+ }
+ return 0;
+}
+
static void
nfp_net_params_setup(struct nfp_net_hw *hw)
{
@@ -451,6 +574,7 @@ nfp_net_start(struct rte_eth_dev *dev)
{
uint32_t new_ctrl, update = 0;
struct nfp_net_hw *hw;
+ int ret;
hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
@@ -476,18 +600,58 @@ nfp_net_start(struct rte_eth_dev *dev)
if (nfp_net_reconfig(hw, new_ctrl, update) < 0)
return -EIO;
+ /*
+ * Allocating rte mbuffs for configured rx queues.
+ * This requires queues being enabled before
+ */
+ if (nfp_net_rx_freelist_setup(dev) < 0) {
+ ret = -ENOMEM;
+ goto error;
+ }
+
hw->ctrl = new_ctrl;
return 0;
+
+error:
+ /*
+ * An error returned by this function should mean the app
+ * exiting and then the system releasing all the memory
+ * allocated even memory coming from hugepages.
+ *
+ * The device could be enabled at this point with some queues
+ * ready for getting packets. This is true if the call to
+ * nfp_net_rx_freelist_setup() succeeds for some queues but
+ * fails for subsequent queues.
+ *
+ * This should make the app exiting but better if we tell the
+ * device first.
+ */
+ nfp_net_disable_queues(dev);
+
+ return ret;
}
/* Stop device: disable rx and tx functions to allow for reconfiguring. */
static void
nfp_net_stop(struct rte_eth_dev *dev)
{
+ int i;
+
PMD_INIT_LOG(DEBUG, "Stop\n");
nfp_net_disable_queues(dev);
+
+ /* Clear queues */
+ for (i = 0; i < dev->data->nb_tx_queues; i++) {
+ nfp_net_reset_tx_queue(
+ (struct nfp_net_txq *)dev->data->tx_queues[i]);
+ }
+
+ for (i = 0; i < dev->data->nb_rx_queues; i++) {
+ nfp_net_reset_rx_queue(
+ (struct nfp_net_rxq *)dev->data->rx_queues[i]);
+ }
}
/* Reset and stop device. The device can not be restarted. */
@@ -515,12 +679,839 @@ nfp_net_close(struct rte_eth_dev *dev)
*/
}
+static uint32_t
+nfp_net_rx_queue_count(struct rte_eth_dev *dev, uint16_t queue_idx)
+{
+ struct nfp_net_rxq *rxq;
+ struct nfp_net_rx_desc *rxds;
+ uint32_t idx;
+ uint32_t count;
+
+ rxq = (struct nfp_net_rxq *)dev->data->rx_queues[queue_idx];
+
+ if (rxq == NULL) {
+ PMD_INIT_LOG(ERR, "Bad queue: %u\n", queue_idx);
+ return 0;
+ }
+
+ idx = rxq->rd_p % rxq->rx_count;
+ rxds = &rxq->rxds[idx];
+
+ count = 0;
+
+ /*
+ * Other PMDs are just checking the DD bit in intervals of 4
+ * descriptors and counting all four if the first has the DD
+ * bit on. Of course, this is not accurate but can be good for
+ * perfomance. But ideally that should be done in descriptors
+ * chunks belonging to the same cache line
+ */
+
+ while (count < rxq->rx_count) {
+ rxds = &rxq->rxds[idx];
+ if ((rxds->rxd.meta_len_dd & PCIE_DESC_RX_DD) == 0)
+ break;
+
+ count++;
+ idx++;
+
+ /* Wrapping? */
+ if ((idx) == rxq->rx_count)
+ idx = 0;
+ }
+
+ return count;
+}
+
+static int
+nfp_net_rx_queue_setup(struct rte_eth_dev *dev,
+ uint16_t queue_idx, uint16_t nb_desc,
+ unsigned int socket_id,
+ const struct rte_eth_rxconf *rx_conf,
+ struct rte_mempool *mp)
+{
+ const struct rte_memzone *tz;
+ struct nfp_net_rxq *rxq;
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ PMD_INIT_FUNC_TRACE();
+
+ /* Validating number of descriptors */
+ if (((nb_desc * sizeof(struct nfp_net_rx_desc)) % 128) != 0 ||
+ (nb_desc > NFP_NET_MAX_RX_DESC) ||
+ (nb_desc < NFP_NET_MIN_RX_DESC)) {
+ RTE_LOG(ERR, PMD, "Wrong nb_desc value\n");
+ return (-EINVAL);
+ }
+
+ /*
+ * Free memory prior to re-allocation if needed. This is the case after
+ * calling nfp_net_stop
+ */
+ if (dev->data->rx_queues[queue_idx]) {
+ nfp_net_rx_queue_release(dev->data->rx_queues[queue_idx]);
+ dev->data->rx_queues[queue_idx] = NULL;
+ }
+
+ /* Allocating rx queue data structure */
+ rxq = rte_zmalloc_socket("ethdev RX queue", sizeof(struct nfp_net_rxq),
+ RTE_CACHE_LINE_SIZE, socket_id);
+ if (rxq == NULL)
+ return (-ENOMEM);
+
+ /* Hw queues mapping based on firmware confifguration */
+ rxq->qidx = queue_idx;
+ rxq->fl_qcidx = queue_idx * hw->stride_rx;
+ rxq->rx_qcidx = rxq->fl_qcidx + (hw->stride_rx - 1);
+ rxq->qcp_fl = hw->rx_bar + NFP_QCP_QUEUE_OFF(rxq->fl_qcidx);
+ rxq->qcp_rx = hw->rx_bar + NFP_QCP_QUEUE_OFF(rxq->rx_qcidx);
+
+ /*
+ * Tracking mbuf size for detecting a potential mbuf overflow due to
+ * RX offset
+ */
+ rxq->mem_pool = mp;
+ rxq->mbuf_size = rxq->mem_pool->elt_size;
+ rxq->mbuf_size -= (sizeof(struct rte_mbuf) + RTE_PKTMBUF_HEADROOM);
+ hw->flbufsz = rxq->mbuf_size;
+
+ rxq->rx_count = nb_desc;
+ rxq->port_id = dev->data->port_id;
+ rxq->rx_free_thresh = rx_conf->rx_free_thresh;
+ rxq->crc_len = (uint8_t) ((dev->data->dev_conf.rxmode.hw_strip_crc) ? 0
+ : ETHER_CRC_LEN);
+ rxq->drop_en = rx_conf->rx_drop_en;
+
+ /*
+ * Allocate RX ring hardware descriptors. A memzone large enough to
+ * handle the maximum ring size is allocated in order to allow for
+ * resizing in later calls to the queue setup function.
+ */
+ tz = ring_dma_zone_reserve(dev, "rx_ring", queue_idx,
+ sizeof(struct nfp_net_rx_desc) *
+ NFP_NET_MAX_RX_DESC, socket_id);
+
+ if (tz == NULL) {
+ RTE_LOG(ERR, PMD, "Error allocatig rx dma\n");
+ nfp_net_rx_queue_release(rxq);
+ return (-ENOMEM);
+ }
+
+ /* Saving physical and virtual addresses for the RX ring */
+ rxq->dma = (uint64_t)tz->phys_addr;
+ rxq->rxds = (struct nfp_net_rx_desc *)tz->addr;
+
+ /* mbuf pointers array for referencing mbufs linked to RX descriptors */
+ rxq->rxbufs = rte_zmalloc_socket("rxq->rxbufs",
+ sizeof(*rxq->rxbufs) * nb_desc,
+ RTE_CACHE_LINE_SIZE, socket_id);
+ if (rxq->rxbufs == NULL) {
+ nfp_net_rx_queue_release(rxq);
+ return (-ENOMEM);
+ }
+
+ PMD_RX_LOG(DEBUG, "rxbufs=%p hw_ring=%p dma_addr=0x%" PRIx64 "\n",
+ rxq->rxbufs, rxq->rxds, (unsigned long int)rxq->dma);
+
+ nfp_net_reset_rx_queue(rxq);
+
+ dev->data->rx_queues[queue_idx] = rxq;
+ rxq->hw = hw;
+
+ /*
+ * Telling the HW about the physical address of the RX ring and number
+ * of descriptors in log2 format
+ */
+ nn_cfg_writeq(hw, NFP_NET_CFG_RXR_ADDR(queue_idx), rxq->dma);
+ nn_cfg_writeb(hw, NFP_NET_CFG_RXR_SZ(queue_idx), log2(nb_desc));
+
+ return 0;
+}
+
+static int
+nfp_net_rx_fill_freelist(struct nfp_net_rxq *rxq)
+{
+ struct nfp_net_rx_buff *rxe = rxq->rxbufs;
+ uint64_t dma_addr;
+ unsigned i;
+
+ PMD_RX_LOG(DEBUG, "nfp_net_rx_fill_freelist for %u descriptors\n",
+ rxq->rx_count);
+
+ for (i = 0; i < rxq->rx_count; i++) {
+ struct nfp_net_rx_desc *rxd;
+ struct rte_mbuf *mbuf = rte_pktmbuf_alloc(rxq->mem_pool);
+
+ if (mbuf == NULL) {
+ RTE_LOG(ERR, PMD, "RX mbuf alloc failed queue_id=%u\n",
+ (unsigned)rxq->qidx);
+ return (-ENOMEM);
+ }
+
+ dma_addr = rte_cpu_to_le_64(RTE_MBUF_DMA_ADDR_DEFAULT(mbuf));
+
+ rxd = &rxq->rxds[i];
+ rxd->fld.dd = 0;
+ rxd->fld.dma_addr_hi = (dma_addr >> 32) & 0xff;
+ rxd->fld.dma_addr_lo = dma_addr & 0xffffffff;
+ rxe[i].mbuf = mbuf;
+ PMD_RX_LOG(DEBUG, "[%d]: %" PRIx64 "\n", i, dma_addr);
+
+ rxq->wr_p++;
+ }
+
+ /* Make sure all writes are flushed before telling the hardware */
+ rte_wmb();
+
+ /* Not advertising the whole ring as the firmware gets confused if so */
+ PMD_RX_LOG(DEBUG, "Increment FL write pointer in %u\n",
+ rxq->rx_count - 1);
+
+ nfp_qcp_ptr_add(rxq->qcp_fl, NFP_QCP_WRITE_PTR, rxq->rx_count - 1);
+
+ return 0;
+}
+
+static int
+nfp_net_tx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
+ uint16_t nb_desc, unsigned int socket_id,
+ const struct rte_eth_txconf *tx_conf)
+{
+ const struct rte_memzone *tz;
+ struct nfp_net_txq *txq;
+ uint16_t tx_free_thresh;
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ PMD_INIT_FUNC_TRACE();
+
+ /* Validating number of descriptors */
+ if (((nb_desc * sizeof(struct nfp_net_tx_desc)) % 128) != 0 ||
+ (nb_desc > NFP_NET_MAX_TX_DESC) ||
+ (nb_desc < NFP_NET_MIN_TX_DESC)) {
+ RTE_LOG(ERR, PMD, "Wrong nb_desc value\n");
+ return -EINVAL;
+ }
+
+ tx_free_thresh = (uint16_t)((tx_conf->tx_free_thresh) ?
+ tx_conf->tx_free_thresh :
+ DEFAULT_TX_FREE_THRESH);
+
+ if (tx_free_thresh > (nb_desc)) {
+ RTE_LOG(ERR, PMD,
+ "tx_free_thresh must be less than the number of TX "
+ "descriptors. (tx_free_thresh=%u port=%d "
+ "queue=%d)\n", (unsigned int)tx_free_thresh,
+ (int)dev->data->port_id, (int)queue_idx);
+ return -(EINVAL);
+ }
+
+ /*
+ * Free memory prior to re-allocation if needed. This is the case after
+ * calling nfp_net_stop
+ */
+ if (dev->data->tx_queues[queue_idx]) {
+ PMD_TX_LOG(DEBUG, "Freeing memory prior to re-allocation %d\n",
+ queue_idx);
+ nfp_net_tx_queue_release(dev->data->tx_queues[queue_idx]);
+ dev->data->tx_queues[queue_idx] = NULL;
+ }
+
+ /* Allocating tx queue data structure */
+ txq = rte_zmalloc_socket("ethdev TX queue", sizeof(struct nfp_net_txq),
+ RTE_CACHE_LINE_SIZE, socket_id);
+ if (txq == NULL) {
+ RTE_LOG(ERR, PMD, "Error allocating tx dma\n");
+ return (-ENOMEM);
+ }
+
+ /*
+ * Allocate TX ring hardware descriptors. A memzone large enough to
+ * handle the maximum ring size is allocated in order to allow for
+ * resizing in later calls to the queue setup function.
+ */
+ tz = ring_dma_zone_reserve(dev, "tx_ring", queue_idx,
+ sizeof(struct nfp_net_tx_desc) *
+ NFP_NET_MAX_TX_DESC, socket_id);
+ if (tz == NULL) {
+ RTE_LOG(ERR, PMD, "Error allocating tx dma\n");
+ nfp_net_tx_queue_release(txq);
+ return (-ENOMEM);
+ }
+
+ txq->tx_count = nb_desc;
+ txq->tail = 0;
+ txq->tx_free_thresh = tx_free_thresh;
+ txq->tx_pthresh = tx_conf->tx_thresh.pthresh;
+ txq->tx_hthresh = tx_conf->tx_thresh.hthresh;
+ txq->tx_wthresh = tx_conf->tx_thresh.wthresh;
+
+ /* queue mapping based on firmware configuration */
+ txq->qidx = queue_idx;
+ txq->tx_qcidx = queue_idx * hw->stride_tx;
+ txq->qcp_q = hw->tx_bar + NFP_QCP_QUEUE_OFF(txq->tx_qcidx);
+
+ txq->port_id = dev->data->port_id;
+ txq->txq_flags = tx_conf->txq_flags;
+
+ /* Saving physical and virtual addresses for the TX ring */
+ txq->dma = (uint64_t)tz->phys_addr;
+ txq->txds = (struct nfp_net_tx_desc *)tz->addr;
+
+ /* mbuf pointers array for referencing mbufs linked to TX descriptors */
+ txq->txbufs = rte_zmalloc_socket("txq->txbufs",
+ sizeof(*txq->txbufs) * nb_desc,
+ RTE_CACHE_LINE_SIZE, socket_id);
+ if (txq->txbufs == NULL) {
+ nfp_net_tx_queue_release(txq);
+ return (-ENOMEM);
+ }
+ PMD_TX_LOG(DEBUG, "txbufs=%p hw_ring=%p dma_addr=0x%" PRIx64 "\n",
+ txq->txbufs, txq->txds, (unsigned long int)txq->dma);
+
+ nfp_net_reset_tx_queue(txq);
+
+ dev->data->tx_queues[queue_idx] = txq;
+ txq->hw = hw;
+
+ /*
+ * Telling the HW about the physical address of the TX ring and number
+ * of descriptors in log2 format
+ */
+ nn_cfg_writeq(hw, NFP_NET_CFG_TXR_ADDR(queue_idx), txq->dma);
+ nn_cfg_writeb(hw, NFP_NET_CFG_TXR_SZ(queue_idx), log2(nb_desc));
+
+ return 0;
+}
+
+/* nfp_net_tx_cksum - Set TX CSUM offload flags in TX descriptor */
+static inline void
+nfp_net_tx_cksum(struct nfp_net_txq *txq, struct nfp_net_tx_desc *txd,
+ struct rte_mbuf *mb)
+{
+ uint16_t ol_flags;
+ struct nfp_net_hw *hw = txq->hw;
+
+ if (!(hw->cap & NFP_NET_CFG_CTRL_TXCSUM))
+ return;
+
+ ol_flags = mb->ol_flags;
+
+ /* IPv6 does not need checksum */
+ if (ol_flags & PKT_TX_IP_CKSUM)
+ txd->flags |= PCIE_DESC_TX_IP4_CSUM;
+
+ switch (ol_flags & PKT_TX_L4_MASK) {
+ case PKT_TX_UDP_CKSUM:
+ txd->flags |= PCIE_DESC_TX_UDP_CSUM;
+ break;
+ case PKT_TX_TCP_CKSUM:
+ txd->flags |= PCIE_DESC_TX_TCP_CSUM;
+ break;
+ }
+
+ txd->flags |= PCIE_DESC_TX_CSUM;
+}
+
+/* nfp_net_rx_cksum - set mbuf checksum flags based on RX descriptor flags */
+static inline void
+nfp_net_rx_cksum(struct nfp_net_rxq *rxq, struct nfp_net_rx_desc *rxd,
+ struct rte_mbuf *mb)
+{
+ struct nfp_net_hw *hw = rxq->hw;
+
+ if (!(hw->ctrl & NFP_NET_CFG_CTRL_RXCSUM))
+ return;
+
+ /* If IPv4 and IP checksum error, fail */
+ if ((rxd->rxd.flags & PCIE_DESC_RX_IP4_CSUM) &&
+ !(rxd->rxd.flags & PCIE_DESC_RX_IP4_CSUM_OK))
+ mb->ol_flags |= PKT_RX_IP_CKSUM_BAD;
+
+ /* If neither UDP nor TCP return */
+ if (!(rxd->rxd.flags & PCIE_DESC_RX_TCP_CSUM) &&
+ !(rxd->rxd.flags & PCIE_DESC_RX_UDP_CSUM))
+ return;
+
+ if ((rxd->rxd.flags & PCIE_DESC_RX_TCP_CSUM) &&
+ !(rxd->rxd.flags & PCIE_DESC_RX_TCP_CSUM_OK))
+ mb->ol_flags |= PKT_RX_L4_CKSUM_BAD;
+
+ if ((rxd->rxd.flags & PCIE_DESC_RX_UDP_CSUM) &&
+ !(rxd->rxd.flags & PCIE_DESC_RX_UDP_CSUM_OK))
+ mb->ol_flags |= PKT_RX_L4_CKSUM_BAD;
+}
+
+#define NFP_HASH_OFFSET ((uint8_t *)mbuf->buf_addr + mbuf->data_off - 4)
+#define NFP_HASH_TYPE_OFFSET ((uint8_t *)mbuf->buf_addr + mbuf->data_off - 8)
+
+/*
+ * nfp_net_set_hash - Set mbuf hash data
+ *
+ * The RSS hash and hash-type are pre-pended to the packet data.
+ * Extract and decode it and set the mbuf fields.
+ */
+static inline void
+nfp_net_set_hash(struct nfp_net_rxq *rxq, struct nfp_net_rx_desc *rxd,
+ struct rte_mbuf *mbuf)
+{
+ uint32_t hash;
+ uint32_t hash_type;
+ struct nfp_net_hw *hw = rxq->hw;
+
+ if (!(hw->ctrl & NFP_NET_CFG_CTRL_RSS))
+ return;
+
+ if (!(rxd->rxd.flags & PCIE_DESC_RX_RSS))
+ return;
+
+ hash = rte_be_to_cpu_32(*(uint32_t *)NFP_HASH_OFFSET);
+ hash_type = rte_be_to_cpu_32(*(uint32_t *)NFP_HASH_TYPE_OFFSET);
+
+ /*
+ * hash type is sharing the same word with input port info
+ * 31-8: input port
+ * 7:0: hash type
+ */
+ hash_type &= 0xff;
+ mbuf->hash.rss = hash;
+ mbuf->ol_flags |= PKT_RX_RSS_HASH;
+
+ switch (hash_type) {
+ case NFP_NET_RSS_IPV4:
+ mbuf->packet_type |= RTE_PTYPE_INNER_L3_IPV4;
+ break;
+ case NFP_NET_RSS_IPV6:
+ mbuf->packet_type |= RTE_PTYPE_INNER_L3_IPV6;
+ break;
+ case NFP_NET_RSS_IPV6_EX:
+ mbuf->packet_type |= RTE_PTYPE_INNER_L3_IPV6_EXT;
+ break;
+ default:
+ mbuf->packet_type |= RTE_PTYPE_INNER_L4_MASK;
+ }
+}
+
+/* nfp_net_check_port - Set mbuf in_port field */
+static void
+nfp_net_check_port(struct nfp_net_rx_desc *rxd, struct rte_mbuf *mbuf)
+{
+ uint32_t port;
+
+ if (!(rxd->rxd.flags & PCIE_DESC_RX_INGRESS_PORT)) {
+ mbuf->port = 0;
+ return;
+ }
+
+ port = rte_be_to_cpu_32(*(uint32_t *)((uint8_t *)mbuf->buf_addr +
+ mbuf->data_off - 8));
+
+ /*
+ * hash type is sharing the same word with input port info
+ * 31-8: input port
+ * 7:0: hash type
+ */
+ port = (uint8_t)(port >> 8);
+ mbuf->port = port;
+}
+
+static inline void
+nfp_net_mbuf_alloc_failed(struct nfp_net_rxq *rxq)
+{
+ rte_eth_devices[rxq->port_id].data->rx_mbuf_alloc_failed++;
+}
+
+#define NFP_DESC_META_LEN(d) (d->rxd.meta_len_dd & PCIE_DESC_RX_META_LEN_MASK)
+
+/*
+ * RX path design:
+ *
+ * There are some decissions to take:
+ * 1) How to check DD RX descriptors bit
+ * 2) How and when to allocate new mbufs
+ *
+ * Current implementation checks just one single DD bit each loop. As each
+ * descriptor is 8 bytes, it is likely a good idea to check descriptors in
+ * a single cache line instead. Tests with this change have not shown any
+ * performance improvement but it requires further investigation. For example,
+ * depending on which descriptor is next, the number of descriptors could be
+ * less than 8 for just checking those in the same cache line. This implies
+ * extra work which could be counterproductive by itself. Indeed, last firmware
+ * changes are just doing this: writing several descriptors with the DD bit
+ * for saving PCIe bandwidth and DMA operations from the NFP.
+ *
+ * Mbuf allocation is done when a new packet is received. Then the descriptor
+ * is automatically linked with the new mbuf and the old one is given to the
+ * user. The main drawback with this design is mbuf allocation is heavier than
+ * using bulk allocations allowed by DPDK with rte_mempool_get_bulk. From the
+ * cache point of view it does not seem allocating the mbuf early on as we are
+ * doing now have any benefit at all. Again, tests with this change have not
+ * shown any improvement. Also, rte_mempool_get_bulk returns all or nothing
+ * so looking at the implications of this type of allocation should be studied
+ * deeply
+ */
+
+static uint16_t
+nfp_net_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t nb_pkts)
+{
+ struct nfp_net_rxq *rxq;
+ struct nfp_net_rx_desc *rxds;
+ struct nfp_net_rx_buff *rxb;
+ struct nfp_net_hw *hw;
+ struct rte_mbuf *mb;
+ struct rte_mbuf *new_mb;
+ int idx;
+ uint16_t nb_hold;
+ uint64_t dma_addr;
+ int avail;
+
+ rxq = rx_queue;
+ if (unlikely(rxq == NULL)) {
+ /*
+ * DPDK just checks the queue is lower than max queues
+ * enabled. But the queue needs to be configured
+ */
+ RTE_LOG(ERR, PMD, "RX Bad queue\n");
+ return -EINVAL;
+ }
+
+ hw = rxq->hw;
+ avail = 0;
+ nb_hold = 0;
+
+ while (avail < nb_pkts) {
+ idx = rxq->rd_p % rxq->rx_count;
+
+ rxb = &rxq->rxbufs[idx];
+ if (unlikely(rxb == NULL)) {
+ RTE_LOG(ERR, PMD, "rxb does not exist!\n");
+ break;
+ }
+
+ /*
+ * Memory barrier to ensure that we won't do other
+ * reads before the DD bit.
+ */
+ rte_rmb();
+
+ rxds = &rxq->rxds[idx];
+ if ((rxds->rxd.meta_len_dd & PCIE_DESC_RX_DD) == 0)
+ break;
+
+ /*
+ * We got a packet. Let's alloc a new mbuff for refilling the
+ * free descriptor ring as soon as possible
+ */
+ new_mb = rte_pktmbuf_alloc(rxq->mem_pool);
+ if (unlikely(new_mb == NULL)) {
+ RTE_LOG(DEBUG, PMD, "RX mbuf alloc failed port_id=%u "
+ "queue_id=%u\n", (unsigned)rxq->port_id,
+ (unsigned)rxq->qidx);
+ nfp_net_mbuf_alloc_failed(rxq);
+ break;
+ }
+
+ nb_hold++;
+
+ /*
+ * Grab the mbuff and refill the descriptor with the
+ * previously allocated mbuff
+ */
+ mb = rxb->mbuf;
+ rxb->mbuf = new_mb;
+
+ PMD_RX_LOG(DEBUG, "Packet len: %u, mbuf_size: %u\n",
+ rxds->rxd.data_len, rxq->mbuf_size);
+
+ /* Size of this segment */
+ mb->data_len = rxds->rxd.data_len - NFP_DESC_META_LEN(rxds);
+ /* Size of the whole packet. We just support 1 segment */
+ mb->pkt_len = rxds->rxd.data_len - NFP_DESC_META_LEN(rxds);
+
+ if (unlikely((mb->data_len + hw->rx_offset) >
+ rxq->mbuf_size)) {
+ /*
+ * This should not happen and the user has the
+ * responsibility of avoiding it. But we have
+ * to give some info about the error
+ */
+ RTE_LOG(ERR, PMD,
+ "mbuf overflow likely due to the RX offset.\n"
+ "\t\tYour mbuf size should have extra space for"
+ " RX offset=%u bytes.\n"
+ "\t\tCurrently you just have %u bytes available"
+ " but the received packet is %u bytes long",
+ hw->rx_offset,
+ rxq->mbuf_size - hw->rx_offset,
+ mb->data_len);
+ return -EINVAL;
+ }
+
+ /* Filling the received mbuff with packet info */
+ if (hw->rx_offset)
+ mb->data_off = RTE_PKTMBUF_HEADROOM + hw->rx_offset;
+ else
+ mb->data_off = RTE_PKTMBUF_HEADROOM +
+ NFP_DESC_META_LEN(rxds);
+
+ /* No scatter mode supported */
+ mb->nb_segs = 1;
+ mb->next = NULL;
+
+ /* Checking the RSS flag */
+ nfp_net_set_hash(rxq, rxds, mb);
+
+ /* Checking the checksum flag */
+ nfp_net_rx_cksum(rxq, rxds, mb);
+
+ /* Checking the port flag */
+ nfp_net_check_port(rxds, mb);
+
+ if ((rxds->rxd.flags & PCIE_DESC_RX_VLAN) &&
+ (hw->ctrl & NFP_NET_CFG_CTRL_RXVLAN)) {
+ mb->vlan_tci = rte_cpu_to_le_32(rxds->rxd.vlan);
+ mb->ol_flags |= PKT_RX_VLAN_PKT;
+ }
+
+ /* Adding the mbuff to the mbuff array passed by the app */
+ rx_pkts[avail++] = mb;
+
+ /* Now resetting and updating the descriptor */
+ rxds->vals[0] = 0;
+ rxds->vals[1] = 0;
+ dma_addr = rte_cpu_to_le_64(RTE_MBUF_DMA_ADDR_DEFAULT(new_mb));
+ rxds->fld.dd = 0;
+ rxds->fld.dma_addr_hi = (dma_addr >> 32) & 0xff;
+ rxds->fld.dma_addr_lo = dma_addr & 0xffffffff;
+
+ rxq->rd_p++;
+ }
+
+ if (nb_hold == 0)
+ return nb_hold;
+
+ PMD_RX_LOG(DEBUG, "RX port_id=%u queue_id=%u, %d packets received\n",
+ (unsigned)rxq->port_id, (unsigned)rxq->qidx, nb_hold);
+
+ nb_hold += rxq->nb_rx_hold;
+
+ /*
+ * FL descriptors needs to be written before incrementing the
+ * FL queue WR pointer
+ */
+ rte_wmb();
+ if (nb_hold > rxq->rx_free_thresh) {
+ PMD_RX_LOG(DEBUG, "port=%u queue=%u nb_hold=%u avail=%u\n",
+ (unsigned)rxq->port_id, (unsigned)rxq->qidx,
+ (unsigned)nb_hold, (unsigned)avail);
+ nfp_qcp_ptr_add(rxq->qcp_fl, NFP_QCP_WRITE_PTR, nb_hold);
+ nb_hold = 0;
+ }
+ rxq->nb_rx_hold = nb_hold;
+
+ return avail;
+}
+
+/*
+ * nfp_net_tx_free_bufs - Check for descriptors with a complete
+ * status
+ * @txq: TX queue to work with
+ * Returns number of descriptors freed
+ */
+int
+nfp_net_tx_free_bufs(struct nfp_net_txq *txq)
+{
+ uint32_t qcp_rd_p;
+ int todo;
+
+ PMD_TX_LOG(DEBUG, "queue %u. Check for descriptor with a complete"
+ " status\n", txq->qidx);
+
+ /* Work out how many packets have been sent */
+ qcp_rd_p = nfp_qcp_read(txq->qcp_q, NFP_QCP_READ_PTR);
+
+ if (qcp_rd_p == txq->qcp_rd_p) {
+ PMD_TX_LOG(DEBUG, "queue %u: It seems harrier is not sending "
+ "packets (%u, %u)\n", txq->qidx,
+ qcp_rd_p, txq->qcp_rd_p);
+ return 0;
+ }
+
+ if (qcp_rd_p > txq->qcp_rd_p)
+ todo = qcp_rd_p - txq->qcp_rd_p;
+ else
+ todo = qcp_rd_p + txq->tx_count - txq->qcp_rd_p;
+
+ PMD_TX_LOG(DEBUG, "qcp_rd_p %u, txq->qcp_rd_p: %u, qcp->rd_p: %u\n",
+ qcp_rd_p, txq->qcp_rd_p, txq->rd_p);
+
+ if (todo == 0)
+ return todo;
+
+ txq->qcp_rd_p += todo;
+ txq->qcp_rd_p %= txq->tx_count;
+ txq->rd_p += todo;
+
+ return todo;
+}
+
+/* Leaving always free descriptors for avoiding wrapping confusion */
+#define NFP_FREE_TX_DESC(t) (t->tx_count - (t->wr_p - t->rd_p) - 8)
+
+/*
+ * nfp_net_txq_full - Check if the TX queue free descriptors
+ * is below tx_free_threshold
+ *
+ * @txq: TX queue to check
+ *
+ * This function uses the host copy* of read/write pointers
+ */
+static inline
+int nfp_net_txq_full(struct nfp_net_txq *txq)
+{
+ return NFP_FREE_TX_DESC(txq) < txq->tx_free_thresh;
+}
+
+static uint16_t
+nfp_net_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts, uint16_t nb_pkts)
+{
+ struct nfp_net_txq *txq;
+ struct nfp_net_hw *hw;
+ struct nfp_net_tx_desc *txds;
+ struct rte_mbuf *pkt;
+ uint64_t dma_addr;
+ int pkt_size, dma_size;
+ uint16_t free_descs, issued_descs;
+ struct rte_mbuf **lmbuf;
+ int i;
+
+ txq = tx_queue;
+ hw = txq->hw;
+ txds = &txq->txds[txq->tail];
+
+ PMD_TX_LOG(DEBUG, "working for queue %u at pos %d and %u packets\n",
+ txq->qidx, txq->tail, nb_pkts);
+
+ if ((NFP_FREE_TX_DESC(txq) < nb_pkts) || (nfp_net_txq_full(txq)))
+ nfp_net_tx_free_bufs(txq);
+
+ free_descs = (uint16_t)NFP_FREE_TX_DESC(txq);
+ if (unlikely(free_descs == 0))
+ return 0;
+
+ pkt = *tx_pkts;
+
+ i = 0;
+ issued_descs = 0;
+ PMD_TX_LOG(DEBUG, "queue: %u. Sending %u packets\n",
+ txq->qidx, nb_pkts);
+ /* Sending packets */
+ while ((i < nb_pkts) && free_descs) {
+ /* Grabbing the mbuf linked to the current descriptor */
+ lmbuf = &txq->txbufs[txq->tail].mbuf;
+ /* Warming the cache for releasing the mbuf later on */
+ RTE_MBUF_PREFETCH_TO_FREE(*lmbuf);
+
+ pkt = *(tx_pkts + i);
+
+ if (unlikely((pkt->nb_segs > 1) &&
+ !(hw->cap & NFP_NET_CFG_CTRL_GATHER))) {
+ PMD_INIT_LOG(INFO, "NFP_NET_CFG_CTRL_GATHER not set\n");
+ rte_panic("Multisegment packet unsupported\n");
+ }
+
+ /* Checking if we have enough descriptors */
+ if (unlikely(pkt->nb_segs > free_descs))
+ goto xmit_end;
+
+ /*
+ * Checksum and VLAN flags just in the first descriptor for a
+ * multisegment packet
+ */
+ nfp_net_tx_cksum(txq, txds, pkt);
+
+ if ((pkt->ol_flags & PKT_TX_VLAN_PKT) &&
+ (hw->cap & NFP_NET_CFG_CTRL_TXVLAN)) {
+ txds->flags |= PCIE_DESC_TX_VLAN;
+ txds->vlan = pkt->vlan_tci;
+ }
+
+ if (pkt->ol_flags & PKT_TX_TCP_SEG)
+ rte_panic("TSO is not supported\n");
+
+ /*
+ * mbuf data_len is the data in one segment and pkt_len data
+ * in the whole packet. When the packet is just one segment,
+ * then data_len = pkt_len
+ */
+ pkt_size = pkt->pkt_len;
+
+ while (pkt_size) {
+ /* Releasing mbuf which was prefetched above */
+ if (*lmbuf)
+ rte_pktmbuf_free_seg(*lmbuf);
+
+ dma_size = pkt->data_len;
+ dma_addr = RTE_MBUF_DATA_DMA_ADDR(pkt);
+ PMD_TX_LOG(DEBUG, "Working with mbuf at dma address:"
+ "%" PRIx64 "\n", dma_addr);
+
+ /* Filling descriptors fields */
+ txds->dma_len = dma_size;
+ txds->data_len = pkt->pkt_len;
+ txds->dma_addr_hi = (dma_addr >> 32) & 0xff;
+ txds->dma_addr_lo = (dma_addr & 0xffffffff);
+ ASSERT(free_descs > 0);
+ free_descs--;
+
+ /*
+ * Linking mbuf with descriptor for being released
+ * next time descriptor is used
+ */
+ *lmbuf = pkt;
+
+ txq->wr_p++;
+ txq->tail++;
+ if (unlikely(txq->tail == txq->tx_count)) /* wrapping?*/
+ txq->tail = 0;
+
+ pkt_size -= dma_size;
+ if (!pkt_size) {
+ /* End of packet */
+ txds->offset_eop |= PCIE_DESC_TX_EOP;
+ } else {
+ txds->offset_eop &= PCIE_DESC_TX_OFFSET_MASK;
+ pkt = pkt->next;
+ }
+ /* Referencing next free TX descriptor */
+ txds = &txq->txds[txq->tail];
+ issued_descs++;
+ }
+ i++;
+ }
+
+xmit_end:
+ /* Increment write pointers. Force memory write before we let HW know */
+ rte_wmb();
+ nfp_qcp_ptr_add(txq->qcp_q, NFP_QCP_WRITE_PTR, issued_descs);
+
+ return i;
+}
+
/* Initialise and register driver with DPDK Application */
static struct eth_dev_ops nfp_net_eth_dev_ops = {
.dev_configure = nfp_net_configure,
.dev_start = nfp_net_start,
.dev_stop = nfp_net_stop,
.dev_close = nfp_net_close,
+ .rx_queue_setup = nfp_net_rx_queue_setup,
+ .rx_queue_release = nfp_net_rx_queue_release,
+ .rx_queue_count = nfp_net_rx_queue_count,
+ .tx_queue_setup = nfp_net_tx_queue_setup,
+ .tx_queue_release = nfp_net_tx_queue_release,
};
static int
@@ -538,6 +1529,8 @@ nfp_net_init(struct rte_eth_dev *eth_dev)
hw = NFP_NET_DEV_PRIVATE_TO_HW(eth_dev->data->dev_private);
eth_dev->dev_ops = &nfp_net_eth_dev_ops;
+ eth_dev->rx_pkt_burst = &nfp_net_recv_pkts;
+ eth_dev->tx_pkt_burst = &nfp_net_xmit_pkts;
/* For secondary processes, the primary has done all the work */
if (rte_eal_process_type() != RTE_PROC_PRIMARY)
--
1.7.9.5
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v10 3/8] nfp: adding rss
2015-11-30 10:25 [PATCH v10 0/8] support for netronome nfp-6xxx card Alejandro Lucero
2015-11-30 10:25 ` [PATCH v10 1/8] nfp: basic initialization Alejandro Lucero
2015-11-30 10:25 ` [PATCH v10 2/8] nfp: adding rx/tx functionality Alejandro Lucero
@ 2015-11-30 10:25 ` Alejandro Lucero
2015-11-30 10:25 ` [PATCH v10 4/8] nfp: adding stats Alejandro Lucero
` (5 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Alejandro Lucero @ 2015-11-30 10:25 UTC (permalink / raw)
To: dev
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@netronome.com>
---
drivers/net/nfp/nfp_net.c | 218 +++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 218 insertions(+)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 0d85fa4..a9be403 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -1501,12 +1501,230 @@ xmit_end:
return i;
}
+/* Update Redirection Table(RETA) of Receive Side Scaling of Ethernet device */
+static int
+nfp_net_reta_update(struct rte_eth_dev *dev,
+ struct rte_eth_rss_reta_entry64 *reta_conf,
+ uint16_t reta_size)
+{
+ uint32_t reta, mask;
+ int i, j;
+ int idx, shift;
+ uint32_t update;
+ struct nfp_net_hw *hw =
+ NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ if (!(hw->ctrl & NFP_NET_CFG_CTRL_RSS))
+ return -EINVAL;
+
+ if (reta_size != NFP_NET_CFG_RSS_ITBL_SZ) {
+ RTE_LOG(ERR, PMD, "The size of hash lookup table configured "
+ "(%d) doesn't match the number hardware can supported "
+ "(%d)\n", reta_size, NFP_NET_CFG_RSS_ITBL_SZ);
+ return -EINVAL;
+ }
+
+ /*
+ * Update Redirection Table. There are 128 8bit-entries which can be
+ * manage as 32 32bit-entries
+ */
+ for (i = 0; i < reta_size; i += 4) {
+ /* Handling 4 RSS entries per loop */
+ idx = i / RTE_RETA_GROUP_SIZE;
+ shift = i % RTE_RETA_GROUP_SIZE;
+ mask = (uint8_t)((reta_conf[idx].mask >> shift) & 0xF);
+
+ if (!mask)
+ continue;
+
+ reta = 0;
+ /* If all 4 entries were set, don't need read RETA register */
+ if (mask != 0xF)
+ reta = nn_cfg_readl(hw, NFP_NET_CFG_RSS_ITBL + i);
+
+ for (j = 0; j < 4; j++) {
+ if (!(mask & (0x1 << j)))
+ continue;
+ if (mask != 0xF)
+ /* Clearing the entry bits */
+ reta &= ~(0xFF << (8 * j));
+ reta |= reta_conf[idx].reta[shift + j] << (8 * j);
+ }
+ nn_cfg_writel(hw, NFP_NET_CFG_RSS_ITBL + shift, reta);
+ }
+
+ update = NFP_NET_CFG_UPDATE_RSS;
+
+ if (nfp_net_reconfig(hw, hw->ctrl, update) < 0)
+ return -EIO;
+
+ return 0;
+}
+
+ /* Query Redirection Table(RETA) of Receive Side Scaling of Ethernet device. */
+static int
+nfp_net_reta_query(struct rte_eth_dev *dev,
+ struct rte_eth_rss_reta_entry64 *reta_conf,
+ uint16_t reta_size)
+{
+ uint8_t i, j, mask;
+ int idx, shift;
+ uint32_t reta;
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ if (!(hw->ctrl & NFP_NET_CFG_CTRL_RSS))
+ return -EINVAL;
+
+ if (reta_size != NFP_NET_CFG_RSS_ITBL_SZ) {
+ RTE_LOG(ERR, PMD, "The size of hash lookup table configured "
+ "(%d) doesn't match the number hardware can supported "
+ "(%d)\n", reta_size, NFP_NET_CFG_RSS_ITBL_SZ);
+ return -EINVAL;
+ }
+
+ /*
+ * Reading Redirection Table. There are 128 8bit-entries which can be
+ * manage as 32 32bit-entries
+ */
+ for (i = 0; i < reta_size; i += 4) {
+ /* Handling 4 RSS entries per loop */
+ idx = i / RTE_RETA_GROUP_SIZE;
+ shift = i % RTE_RETA_GROUP_SIZE;
+ mask = (uint8_t)((reta_conf[idx].mask >> shift) & 0xF);
+
+ if (!mask)
+ continue;
+
+ reta = nn_cfg_readl(hw, NFP_NET_CFG_RSS_ITBL + shift);
+ for (j = 0; j < 4; j++) {
+ if (!(mask & (0x1 << j)))
+ continue;
+ reta_conf->reta[shift + j] =
+ (uint8_t)((reta >> (8 * j)) & 0xF);
+ }
+ }
+ return 0;
+}
+
+static int
+nfp_net_rss_hash_update(struct rte_eth_dev *dev,
+ struct rte_eth_rss_conf *rss_conf)
+{
+ uint32_t update;
+ uint32_t cfg_rss_ctrl = 0;
+ uint8_t key;
+ uint64_t rss_hf;
+ int i;
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ rss_hf = rss_conf->rss_hf;
+
+ /* Checking if RSS is enabled */
+ if (!(hw->ctrl & NFP_NET_CFG_CTRL_RSS)) {
+ if (rss_hf != 0) { /* Enable RSS? */
+ RTE_LOG(ERR, PMD, "RSS unsupported\n");
+ return -EINVAL;
+ }
+ return 0; /* Nothing to do */
+ }
+
+ if (rss_conf->rss_key_len > NFP_NET_CFG_RSS_KEY_SZ) {
+ RTE_LOG(ERR, PMD, "hash key too long\n");
+ return -EINVAL;
+ }
+
+ if (rss_hf & ETH_RSS_IPV4)
+ cfg_rss_ctrl |= NFP_NET_CFG_RSS_IPV4 |
+ NFP_NET_CFG_RSS_IPV4_TCP |
+ NFP_NET_CFG_RSS_IPV4_UDP;
+
+ if (rss_hf & ETH_RSS_IPV6)
+ cfg_rss_ctrl |= NFP_NET_CFG_RSS_IPV6 |
+ NFP_NET_CFG_RSS_IPV6_TCP |
+ NFP_NET_CFG_RSS_IPV6_UDP;
+
+ /* configuring where to apply the RSS hash */
+ nn_cfg_writel(hw, NFP_NET_CFG_RSS_CTRL, cfg_rss_ctrl);
+
+ /* Writing the key byte a byte */
+ for (i = 0; i < rss_conf->rss_key_len; i++) {
+ memcpy(&key, &rss_conf->rss_key[i], 1);
+ nn_cfg_writeb(hw, NFP_NET_CFG_RSS_KEY + i, key);
+ }
+
+ /* Writing the key size */
+ nn_cfg_writeb(hw, NFP_NET_CFG_RSS_KEY_SZ, rss_conf->rss_key_len);
+
+ update = NFP_NET_CFG_UPDATE_RSS;
+
+ if (nfp_net_reconfig(hw, hw->ctrl, update) < 0)
+ return -EIO;
+
+ return 0;
+}
+
+static int
+nfp_net_rss_hash_conf_get(struct rte_eth_dev *dev,
+ struct rte_eth_rss_conf *rss_conf)
+{
+ uint64_t rss_hf;
+ uint32_t cfg_rss_ctrl;
+ uint8_t key;
+ int i;
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ if (!(hw->ctrl & NFP_NET_CFG_CTRL_RSS))
+ return -EINVAL;
+
+ rss_hf = rss_conf->rss_hf;
+ cfg_rss_ctrl = nn_cfg_readl(hw, NFP_NET_CFG_RSS_CTRL);
+
+ if (cfg_rss_ctrl & NFP_NET_CFG_RSS_IPV4)
+ rss_hf |= ETH_RSS_NONFRAG_IPV4_TCP | ETH_RSS_NONFRAG_IPV4_UDP;
+
+ if (cfg_rss_ctrl & NFP_NET_CFG_RSS_IPV4_TCP)
+ rss_hf |= ETH_RSS_NONFRAG_IPV4_TCP;
+
+ if (cfg_rss_ctrl & NFP_NET_CFG_RSS_IPV6_TCP)
+ rss_hf |= ETH_RSS_NONFRAG_IPV6_TCP;
+
+ if (cfg_rss_ctrl & NFP_NET_CFG_RSS_IPV4_UDP)
+ rss_hf |= ETH_RSS_NONFRAG_IPV4_UDP;
+
+ if (cfg_rss_ctrl & NFP_NET_CFG_RSS_IPV6_UDP)
+ rss_hf |= ETH_RSS_NONFRAG_IPV6_UDP;
+
+ if (cfg_rss_ctrl & NFP_NET_CFG_RSS_IPV6)
+ rss_hf |= ETH_RSS_NONFRAG_IPV4_UDP | ETH_RSS_NONFRAG_IPV6_UDP;
+
+ /* Reading the key size */
+ rss_conf->rss_key_len = nn_cfg_readl(hw, NFP_NET_CFG_RSS_KEY_SZ);
+
+ /* Reading the key byte a byte */
+ for (i = 0; i < rss_conf->rss_key_len; i++) {
+ key = nn_cfg_readb(hw, NFP_NET_CFG_RSS_KEY + i);
+ memcpy(&rss_conf->rss_key[i], &key, 1);
+ }
+
+ return 0;
+}
+
/* Initialise and register driver with DPDK Application */
static struct eth_dev_ops nfp_net_eth_dev_ops = {
.dev_configure = nfp_net_configure,
.dev_start = nfp_net_start,
.dev_stop = nfp_net_stop,
.dev_close = nfp_net_close,
+ .reta_update = nfp_net_reta_update,
+ .reta_query = nfp_net_reta_query,
+ .rss_hash_update = nfp_net_rss_hash_update,
+ .rss_hash_conf_get = nfp_net_rss_hash_conf_get,
.rx_queue_setup = nfp_net_rx_queue_setup,
.rx_queue_release = nfp_net_rx_queue_release,
.rx_queue_count = nfp_net_rx_queue_count,
--
1.7.9.5
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v10 4/8] nfp: adding stats
2015-11-30 10:25 [PATCH v10 0/8] support for netronome nfp-6xxx card Alejandro Lucero
` (2 preceding siblings ...)
2015-11-30 10:25 ` [PATCH v10 3/8] nfp: adding rss Alejandro Lucero
@ 2015-11-30 10:25 ` Alejandro Lucero
2015-11-30 10:25 ` [PATCH v10 5/8] nfp: adding link functionality Alejandro Lucero
` (4 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Alejandro Lucero @ 2015-11-30 10:25 UTC (permalink / raw)
To: dev
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@netronome.com>
---
drivers/net/nfp/nfp_net.c | 179 +++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 179 insertions(+)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index a9be403..0912064 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -90,6 +90,9 @@ static int nfp_net_tx_queue_setup(struct rte_eth_dev *dev, uint16_t queue_idx,
uint16_t nb_desc, unsigned int socket_id,
const struct rte_eth_txconf *tx_conf);
static int nfp_net_start(struct rte_eth_dev *dev);
+static void nfp_net_stats_get(struct rte_eth_dev *dev,
+ struct rte_eth_stats *stats);
+static void nfp_net_stats_reset(struct rte_eth_dev *dev);
static void nfp_net_stop(struct rte_eth_dev *dev);
static uint16_t nfp_net_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
uint16_t nb_pkts);
@@ -679,6 +682,177 @@ nfp_net_close(struct rte_eth_dev *dev)
*/
}
+static void
+nfp_net_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *stats)
+{
+ int i;
+ struct nfp_net_hw *hw;
+ struct rte_eth_stats nfp_dev_stats;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ /* RTE_ETHDEV_QUEUE_STAT_CNTRS default value is 16 */
+
+ /* reading per RX ring stats */
+ for (i = 0; i < dev->data->nb_rx_queues; i++) {
+ if (i == RTE_ETHDEV_QUEUE_STAT_CNTRS)
+ break;
+
+ nfp_dev_stats.q_ipackets[i] =
+ nn_cfg_readq(hw, NFP_NET_CFG_RXR_STATS(i));
+
+ nfp_dev_stats.q_ipackets[i] -=
+ hw->eth_stats_base.q_ipackets[i];
+
+ nfp_dev_stats.q_ibytes[i] =
+ nn_cfg_readq(hw, NFP_NET_CFG_RXR_STATS(i) + 0x8);
+
+ nfp_dev_stats.q_ibytes[i] -=
+ hw->eth_stats_base.q_ibytes[i];
+ }
+
+ /* reading per TX ring stats */
+ for (i = 0; i < dev->data->nb_tx_queues; i++) {
+ if (i == RTE_ETHDEV_QUEUE_STAT_CNTRS)
+ break;
+
+ nfp_dev_stats.q_opackets[i] =
+ nn_cfg_readq(hw, NFP_NET_CFG_TXR_STATS(i));
+
+ nfp_dev_stats.q_opackets[i] -=
+ hw->eth_stats_base.q_opackets[i];
+
+ nfp_dev_stats.q_obytes[i] =
+ nn_cfg_readq(hw, NFP_NET_CFG_TXR_STATS(i) + 0x8);
+
+ nfp_dev_stats.q_obytes[i] -=
+ hw->eth_stats_base.q_obytes[i];
+ }
+
+ nfp_dev_stats.ipackets =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_FRAMES);
+
+ nfp_dev_stats.ipackets -= hw->eth_stats_base.ipackets;
+
+ nfp_dev_stats.ibytes =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_OCTETS);
+
+ nfp_dev_stats.ibytes -= hw->eth_stats_base.ibytes;
+
+ nfp_dev_stats.opackets =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_TX_FRAMES);
+
+ nfp_dev_stats.opackets -= hw->eth_stats_base.opackets;
+
+ nfp_dev_stats.obytes =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_TX_OCTETS);
+
+ nfp_dev_stats.obytes -= hw->eth_stats_base.obytes;
+
+ nfp_dev_stats.imcasts =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_MC_FRAMES);
+
+ nfp_dev_stats.imcasts -= hw->eth_stats_base.imcasts;
+
+ /* reading general device stats */
+ nfp_dev_stats.ierrors =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_ERRORS);
+
+ nfp_dev_stats.ierrors -= hw->eth_stats_base.ierrors;
+
+ nfp_dev_stats.oerrors =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_TX_ERRORS);
+
+ nfp_dev_stats.oerrors -= hw->eth_stats_base.oerrors;
+
+ /* Multicast frames received */
+ nfp_dev_stats.imcasts =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_MC_FRAMES);
+
+ nfp_dev_stats.imcasts -= hw->eth_stats_base.imcasts;
+
+ /* RX ring mbuf allocation failures */
+ nfp_dev_stats.rx_nombuf = dev->data->rx_mbuf_alloc_failed;
+
+ nfp_dev_stats.imissed =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_DISCARDS);
+
+ nfp_dev_stats.imissed -= hw->eth_stats_base.imissed;
+
+ if (stats)
+ memcpy(stats, &nfp_dev_stats, sizeof(*stats));
+}
+
+static void
+nfp_net_stats_reset(struct rte_eth_dev *dev)
+{
+ int i;
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ /*
+ * hw->eth_stats_base records the per counter starting point.
+ * Lets update it now
+ */
+
+ /* reading per RX ring stats */
+ for (i = 0; i < dev->data->nb_rx_queues; i++) {
+ if (i == RTE_ETHDEV_QUEUE_STAT_CNTRS)
+ break;
+
+ hw->eth_stats_base.q_ipackets[i] =
+ nn_cfg_readq(hw, NFP_NET_CFG_RXR_STATS(i));
+
+ hw->eth_stats_base.q_ibytes[i] =
+ nn_cfg_readq(hw, NFP_NET_CFG_RXR_STATS(i) + 0x8);
+ }
+
+ /* reading per TX ring stats */
+ for (i = 0; i < dev->data->nb_tx_queues; i++) {
+ if (i == RTE_ETHDEV_QUEUE_STAT_CNTRS)
+ break;
+
+ hw->eth_stats_base.q_opackets[i] =
+ nn_cfg_readq(hw, NFP_NET_CFG_TXR_STATS(i));
+
+ hw->eth_stats_base.q_obytes[i] =
+ nn_cfg_readq(hw, NFP_NET_CFG_TXR_STATS(i) + 0x8);
+ }
+
+ hw->eth_stats_base.ipackets =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_FRAMES);
+
+ hw->eth_stats_base.ibytes =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_OCTETS);
+
+ hw->eth_stats_base.opackets =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_TX_FRAMES);
+
+ hw->eth_stats_base.obytes =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_TX_OCTETS);
+
+ hw->eth_stats_base.imcasts =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_MC_FRAMES);
+
+ /* reading general device stats */
+ hw->eth_stats_base.ierrors =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_ERRORS);
+
+ hw->eth_stats_base.oerrors =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_TX_ERRORS);
+
+ /* Multicast frames received */
+ hw->eth_stats_base.imcasts =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_MC_FRAMES);
+
+ /* RX ring mbuf allocation failures */
+ dev->data->rx_mbuf_alloc_failed = 0;
+
+ hw->eth_stats_base.imissed =
+ nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_DISCARDS);
+}
+
static uint32_t
nfp_net_rx_queue_count(struct rte_eth_dev *dev, uint16_t queue_idx)
{
@@ -1721,6 +1895,8 @@ static struct eth_dev_ops nfp_net_eth_dev_ops = {
.dev_start = nfp_net_start,
.dev_stop = nfp_net_stop,
.dev_close = nfp_net_close,
+ .stats_get = nfp_net_stats_get,
+ .stats_reset = nfp_net_stats_reset,
.reta_update = nfp_net_reta_update,
.reta_query = nfp_net_reta_query,
.rss_hash_update = nfp_net_rss_hash_update,
@@ -1852,6 +2028,9 @@ nfp_net_init(struct rte_eth_dev *eth_dev)
hw->mac_addr[0], hw->mac_addr[1], hw->mac_addr[2],
hw->mac_addr[3], hw->mac_addr[4], hw->mac_addr[5]);
+ /* Recording current stats counters values */
+ nfp_net_stats_reset(eth_dev);
+
return 0;
}
--
1.7.9.5
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v10 5/8] nfp: adding link functionality
2015-11-30 10:25 [PATCH v10 0/8] support for netronome nfp-6xxx card Alejandro Lucero
` (3 preceding siblings ...)
2015-11-30 10:25 ` [PATCH v10 4/8] nfp: adding stats Alejandro Lucero
@ 2015-11-30 10:25 ` Alejandro Lucero
2015-11-30 10:25 ` [PATCH v10 6/8] nfp: adding extra functionality Alejandro Lucero
` (3 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Alejandro Lucero @ 2015-11-30 10:25 UTC (permalink / raw)
To: dev
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@netronome.com>
---
drivers/net/nfp/nfp_net.c | 96 +++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 96 insertions(+)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 0912064..7c82e96 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -74,6 +74,7 @@
static void nfp_net_close(struct rte_eth_dev *dev);
static int nfp_net_configure(struct rte_eth_dev *dev);
static int nfp_net_init(struct rte_eth_dev *eth_dev);
+static int nfp_net_link_update(struct rte_eth_dev *dev, int wait_to_complete);
static int nfp_net_rx_fill_freelist(struct nfp_net_rxq *rxq);
static uint32_t nfp_net_rx_queue_count(struct rte_eth_dev *dev,
uint16_t queue_idx);
@@ -226,6 +227,57 @@ ring_dma_zone_reserve(struct rte_eth_dev *dev, const char *ring_name,
NFP_MEMZONE_ALIGN);
}
+/*
+ * Atomically reads link status information from global structure rte_eth_dev.
+ *
+ * @param dev
+ * - Pointer to the structure rte_eth_dev to read from.
+ * - Pointer to the buffer to be saved with the link status.
+ *
+ * @return
+ * - On success, zero.
+ * - On failure, negative value.
+ */
+static inline int
+nfp_net_dev_atomic_read_link_status(struct rte_eth_dev *dev,
+ struct rte_eth_link *link)
+{
+ struct rte_eth_link *dst = link;
+ struct rte_eth_link *src = &dev->data->dev_link;
+
+ if (rte_atomic64_cmpset((uint64_t *)dst, *(uint64_t *)dst,
+ *(uint64_t *)src) == 0)
+ return -1;
+
+ return 0;
+}
+
+/*
+ * Atomically writes the link status information into global
+ * structure rte_eth_dev.
+ *
+ * @param dev
+ * - Pointer to the structure rte_eth_dev to read from.
+ * - Pointer to the buffer to be saved with the link status.
+ *
+ * @return
+ * - On success, zero.
+ * - On failure, negative value.
+ */
+static inline int
+nfp_net_dev_atomic_write_link_status(struct rte_eth_dev *dev,
+ struct rte_eth_link *link)
+{
+ struct rte_eth_link *dst = &dev->data->dev_link;
+ struct rte_eth_link *src = link;
+
+ if (rte_atomic64_cmpset((uint64_t *)dst, *(uint64_t *)dst,
+ *(uint64_t *)src) == 0)
+ return -1;
+
+ return 0;
+}
+
static void
nfp_net_rx_queue_release_mbufs(struct nfp_net_rxq *rxq)
{
@@ -682,6 +734,49 @@ nfp_net_close(struct rte_eth_dev *dev)
*/
}
+/*
+ * return 0 means link status changed, -1 means not changed
+ *
+ * Wait to complete is needed as it can take up to 9 seconds to get the Link
+ * status.
+ */
+static int
+nfp_net_link_update(struct rte_eth_dev *dev, __rte_unused int wait_to_complete)
+{
+ struct nfp_net_hw *hw;
+ struct rte_eth_link link, old;
+ uint32_t nn_link_status;
+
+ PMD_DRV_LOG(DEBUG, "Link update\n");
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ memset(&old, 0, sizeof(old));
+ nfp_net_dev_atomic_read_link_status(dev, &old);
+
+ nn_link_status = nn_cfg_readl(hw, NFP_NET_CFG_STS);
+
+ memset(&link, 0, sizeof(struct rte_eth_link));
+
+ if (nn_link_status & NFP_NET_CFG_STS_LINK)
+ link.link_status = 1;
+
+ link.link_duplex = ETH_LINK_FULL_DUPLEX;
+ /* Other cards can limit the tx and rx rate per VF */
+ link.link_speed = ETH_LINK_SPEED_40G;
+
+ if (old.link_status != link.link_status) {
+ nfp_net_dev_atomic_write_link_status(dev, &link);
+ if (link.link_status)
+ PMD_DRV_LOG(INFO, "NIC Link is Up\n");
+ else
+ PMD_DRV_LOG(INFO, "NIC Link is Down\n");
+ return 0;
+ }
+
+ return -1;
+}
+
static void
nfp_net_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *stats)
{
@@ -1895,6 +1990,7 @@ static struct eth_dev_ops nfp_net_eth_dev_ops = {
.dev_start = nfp_net_start,
.dev_stop = nfp_net_stop,
.dev_close = nfp_net_close,
+ .link_update = nfp_net_link_update,
.stats_get = nfp_net_stats_get,
.stats_reset = nfp_net_stats_reset,
.reta_update = nfp_net_reta_update,
--
1.7.9.5
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v10 6/8] nfp: adding extra functionality
2015-11-30 10:25 [PATCH v10 0/8] support for netronome nfp-6xxx card Alejandro Lucero
` (4 preceding siblings ...)
2015-11-30 10:25 ` [PATCH v10 5/8] nfp: adding link functionality Alejandro Lucero
@ 2015-11-30 10:25 ` Alejandro Lucero
2015-11-30 10:25 ` [PATCH v10 7/8] nfp: link status change interrupt support Alejandro Lucero
` (2 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: Alejandro Lucero @ 2015-11-30 10:25 UTC (permalink / raw)
To: dev
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@netronome.com>
---
drivers/net/nfp/nfp_net.c | 191 +++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 191 insertions(+)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index 7c82e96..ff9a8d6 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -73,8 +73,13 @@
/* Prototypes */
static void nfp_net_close(struct rte_eth_dev *dev);
static int nfp_net_configure(struct rte_eth_dev *dev);
+static int nfp_net_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu);
+static void nfp_net_infos_get(struct rte_eth_dev *dev,
+ struct rte_eth_dev_info *dev_info);
static int nfp_net_init(struct rte_eth_dev *eth_dev);
static int nfp_net_link_update(struct rte_eth_dev *dev, int wait_to_complete);
+static void nfp_net_promisc_enable(struct rte_eth_dev *dev);
+static void nfp_net_promisc_disable(struct rte_eth_dev *dev);
static int nfp_net_rx_fill_freelist(struct nfp_net_rxq *rxq);
static uint32_t nfp_net_rx_queue_count(struct rte_eth_dev *dev,
uint16_t queue_idx);
@@ -734,6 +739,65 @@ nfp_net_close(struct rte_eth_dev *dev)
*/
}
+static void
+nfp_net_promisc_enable(struct rte_eth_dev *dev)
+{
+ uint32_t new_ctrl, update = 0;
+ struct nfp_net_hw *hw;
+
+ PMD_DRV_LOG(DEBUG, "Promiscuous mode enable\n");
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ if (!(hw->cap & NFP_NET_CFG_CTRL_PROMISC)) {
+ PMD_INIT_LOG(INFO, "Promiscuous mode not supported\n");
+ return;
+ }
+
+ if (hw->ctrl & NFP_NET_CFG_CTRL_PROMISC) {
+ PMD_DRV_LOG(INFO, "Promiscuous mode already enabled\n");
+ return;
+ }
+
+ new_ctrl = hw->ctrl | NFP_NET_CFG_CTRL_PROMISC;
+ update = NFP_NET_CFG_UPDATE_GEN;
+
+ /*
+ * DPDK sets promiscuous mode on just after this call assuming
+ * it can not fail ...
+ */
+ if (nfp_net_reconfig(hw, new_ctrl, update) < 0)
+ return;
+
+ hw->ctrl = new_ctrl;
+}
+
+static void
+nfp_net_promisc_disable(struct rte_eth_dev *dev)
+{
+ uint32_t new_ctrl, update = 0;
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ if ((hw->ctrl & NFP_NET_CFG_CTRL_PROMISC) == 0) {
+ PMD_DRV_LOG(INFO, "Promiscuous mode already disabled\n");
+ return;
+ }
+
+ new_ctrl = hw->ctrl & ~NFP_NET_CFG_CTRL_PROMISC;
+ update = NFP_NET_CFG_UPDATE_GEN;
+
+ /*
+ * DPDK sets promiscuous mode off just before this call
+ * assuming it can not fail ...
+ */
+ if (nfp_net_reconfig(hw, new_ctrl, update) < 0)
+ return;
+
+ hw->ctrl = new_ctrl;
+}
+
/*
* return 0 means link status changed, -1 means not changed
*
@@ -948,6 +1012,65 @@ nfp_net_stats_reset(struct rte_eth_dev *dev)
nn_cfg_readq(hw, NFP_NET_CFG_STATS_RX_DISCARDS);
}
+static void
+nfp_net_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
+{
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ dev_info->driver_name = dev->driver->pci_drv.name;
+ dev_info->max_rx_queues = (uint16_t)hw->max_rx_queues;
+ dev_info->max_tx_queues = (uint16_t)hw->max_tx_queues;
+ dev_info->min_rx_bufsize = ETHER_MIN_MTU;
+ dev_info->max_rx_pktlen = hw->mtu;
+ /* Next should change when PF support is implemented */
+ dev_info->max_mac_addrs = 1;
+
+ if (hw->cap & NFP_NET_CFG_CTRL_RXVLAN)
+ dev_info->rx_offload_capa = DEV_RX_OFFLOAD_VLAN_STRIP;
+
+ if (hw->cap & NFP_NET_CFG_CTRL_RXCSUM)
+ dev_info->rx_offload_capa |= DEV_RX_OFFLOAD_IPV4_CKSUM |
+ DEV_RX_OFFLOAD_UDP_CKSUM |
+ DEV_RX_OFFLOAD_TCP_CKSUM;
+
+ if (hw->cap & NFP_NET_CFG_CTRL_TXVLAN)
+ dev_info->tx_offload_capa = DEV_TX_OFFLOAD_VLAN_INSERT;
+
+ if (hw->cap & NFP_NET_CFG_CTRL_TXCSUM)
+ dev_info->tx_offload_capa |= DEV_TX_OFFLOAD_IPV4_CKSUM |
+ DEV_RX_OFFLOAD_UDP_CKSUM |
+ DEV_RX_OFFLOAD_TCP_CKSUM;
+
+ dev_info->default_rxconf = (struct rte_eth_rxconf) {
+ .rx_thresh = {
+ .pthresh = DEFAULT_RX_PTHRESH,
+ .hthresh = DEFAULT_RX_HTHRESH,
+ .wthresh = DEFAULT_RX_WTHRESH,
+ },
+ .rx_free_thresh = DEFAULT_RX_FREE_THRESH,
+ .rx_drop_en = 0,
+ };
+
+ dev_info->default_txconf = (struct rte_eth_txconf) {
+ .tx_thresh = {
+ .pthresh = DEFAULT_TX_PTHRESH,
+ .hthresh = DEFAULT_TX_HTHRESH,
+ .wthresh = DEFAULT_TX_WTHRESH,
+ },
+ .tx_free_thresh = DEFAULT_TX_FREE_THRESH,
+ .tx_rs_thresh = DEFAULT_TX_RSBIT_THRESH,
+ .txq_flags = ETH_TXQ_FLAGS_NOMULTSEGS |
+ ETH_TXQ_FLAGS_NOOFFLOADS,
+ };
+
+ dev_info->reta_size = NFP_NET_CFG_RSS_ITBL_SZ;
+#if RTE_VER_MAJOR == 2 && RTE_VER_MINOR >= 1
+ dev_info->hash_key_size = NFP_NET_CFG_RSS_KEY_SZ;
+#endif
+}
+
static uint32_t
nfp_net_rx_queue_count(struct rte_eth_dev *dev, uint16_t queue_idx)
{
@@ -993,6 +1116,34 @@ nfp_net_rx_queue_count(struct rte_eth_dev *dev, uint16_t queue_idx)
}
static int
+nfp_net_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu)
+{
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ /* check that mtu is within the allowed range */
+ if ((mtu < ETHER_MIN_MTU) || ((uint32_t)mtu > hw->max_mtu))
+ return -EINVAL;
+
+ /* switch to jumbo mode if needed */
+ if ((uint32_t)mtu > ETHER_MAX_LEN)
+ dev->data->dev_conf.rxmode.jumbo_frame = 1;
+ else
+ dev->data->dev_conf.rxmode.jumbo_frame = 0;
+
+ /* update max frame size */
+ dev->data->dev_conf.rxmode.max_rx_pkt_len = (uint32_t)mtu;
+
+ /* writing to configuration space */
+ nn_cfg_writel(hw, NFP_NET_CFG_MTU, (uint32_t)mtu);
+
+ hw->mtu = mtu;
+
+ return 0;
+}
+
+static int
nfp_net_rx_queue_setup(struct rte_eth_dev *dev,
uint16_t queue_idx, uint16_t nb_desc,
unsigned int socket_id,
@@ -1770,6 +1921,41 @@ xmit_end:
return i;
}
+static void
+nfp_net_vlan_offload_set(struct rte_eth_dev *dev, int mask)
+{
+ uint32_t new_ctrl, update;
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+ new_ctrl = 0;
+
+ if ((mask & ETH_VLAN_FILTER_OFFLOAD) ||
+ (mask & ETH_VLAN_FILTER_OFFLOAD))
+ RTE_LOG(INFO, PMD, "Not support for ETH_VLAN_FILTER_OFFLOAD or"
+ " ETH_VLAN_FILTER_EXTEND");
+
+ /* Enable vlan strip if it is not configured yet */
+ if ((mask & ETH_VLAN_STRIP_OFFLOAD) &&
+ !(hw->ctrl & NFP_NET_CFG_CTRL_RXVLAN))
+ new_ctrl = hw->ctrl | NFP_NET_CFG_CTRL_RXVLAN;
+
+ /* Disable vlan strip just if it is configured */
+ if (!(mask & ETH_VLAN_STRIP_OFFLOAD) &&
+ (hw->ctrl & NFP_NET_CFG_CTRL_RXVLAN))
+ new_ctrl = hw->ctrl & ~NFP_NET_CFG_CTRL_RXVLAN;
+
+ if (new_ctrl == 0)
+ return;
+
+ update = NFP_NET_CFG_UPDATE_GEN;
+
+ if (nfp_net_reconfig(hw, new_ctrl, update) < 0)
+ return;
+
+ hw->ctrl = new_ctrl;
+}
+
/* Update Redirection Table(RETA) of Receive Side Scaling of Ethernet device */
static int
nfp_net_reta_update(struct rte_eth_dev *dev,
@@ -1990,9 +2176,14 @@ static struct eth_dev_ops nfp_net_eth_dev_ops = {
.dev_start = nfp_net_start,
.dev_stop = nfp_net_stop,
.dev_close = nfp_net_close,
+ .promiscuous_enable = nfp_net_promisc_enable,
+ .promiscuous_disable = nfp_net_promisc_disable,
.link_update = nfp_net_link_update,
.stats_get = nfp_net_stats_get,
.stats_reset = nfp_net_stats_reset,
+ .dev_infos_get = nfp_net_infos_get,
+ .mtu_set = nfp_net_dev_mtu_set,
+ .vlan_offload_set = nfp_net_vlan_offload_set,
.reta_update = nfp_net_reta_update,
.reta_query = nfp_net_reta_query,
.rss_hash_update = nfp_net_rss_hash_update,
--
1.7.9.5
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v10 7/8] nfp: link status change interrupt support
2015-11-30 10:25 [PATCH v10 0/8] support for netronome nfp-6xxx card Alejandro Lucero
` (5 preceding siblings ...)
2015-11-30 10:25 ` [PATCH v10 6/8] nfp: adding extra functionality Alejandro Lucero
@ 2015-11-30 10:25 ` Alejandro Lucero
2015-11-30 10:25 ` [PATCH v10 8/8] nfp: adding nic guide Alejandro Lucero
2015-12-07 23:47 ` [PATCH v10 0/8] support for netronome nfp-6xxx card Thomas Monjalon
8 siblings, 0 replies; 10+ messages in thread
From: Alejandro Lucero @ 2015-11-30 10:25 UTC (permalink / raw)
To: dev
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@netronome.com>
---
drivers/net/nfp/nfp_net.c | 123 +++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 123 insertions(+)
diff --git a/drivers/net/nfp/nfp_net.c b/drivers/net/nfp/nfp_net.c
index ff9a8d6..bc2089f 100644
--- a/drivers/net/nfp/nfp_net.c
+++ b/drivers/net/nfp/nfp_net.c
@@ -73,6 +73,9 @@
/* Prototypes */
static void nfp_net_close(struct rte_eth_dev *dev);
static int nfp_net_configure(struct rte_eth_dev *dev);
+static void nfp_net_dev_interrupt_handler(struct rte_intr_handle *handle,
+ void *param);
+static void nfp_net_dev_interrupt_delayed_handler(void *param);
static int nfp_net_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu);
static void nfp_net_infos_get(struct rte_eth_dev *dev,
struct rte_eth_dev_info *dev_info);
@@ -731,6 +734,7 @@ nfp_net_close(struct rte_eth_dev *dev)
nfp_net_stop(dev);
+ rte_intr_disable(&dev->pci_dev->intr_handle);
nn_cfg_writeb(hw, NFP_NET_CFG_LSC, 0xff);
/*
@@ -1115,6 +1119,114 @@ nfp_net_rx_queue_count(struct rte_eth_dev *dev, uint16_t queue_idx)
return count;
}
+static void
+nfp_net_dev_link_status_print(struct rte_eth_dev *dev)
+{
+ struct rte_eth_link link;
+
+ memset(&link, 0, sizeof(link));
+ nfp_net_dev_atomic_read_link_status(dev, &link);
+ if (link.link_status)
+ RTE_LOG(INFO, PMD, "Port %d: Link Up - speed %u Mbps - %s\n",
+ (int)(dev->data->port_id), (unsigned)link.link_speed,
+ link.link_duplex == ETH_LINK_FULL_DUPLEX
+ ? "full-duplex" : "half-duplex");
+ else
+ RTE_LOG(INFO, PMD, " Port %d: Link Down\n",
+ (int)(dev->data->port_id));
+
+ RTE_LOG(INFO, PMD, "PCI Address: %04d:%02d:%02d:%d\n",
+ dev->pci_dev->addr.domain, dev->pci_dev->addr.bus,
+ dev->pci_dev->addr.devid, dev->pci_dev->addr.function);
+}
+
+/* Interrupt configuration and handling */
+
+/*
+ * nfp_net_irq_unmask - Unmask an interrupt
+ *
+ * If MSI-X auto-masking is enabled clear the mask bit, otherwise
+ * clear the ICR for the entry.
+ */
+static void
+nfp_net_irq_unmask(struct rte_eth_dev *dev)
+{
+ struct nfp_net_hw *hw;
+
+ hw = NFP_NET_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+
+ if (hw->ctrl & NFP_NET_CFG_CTRL_MSIXAUTO) {
+ /* If MSI-X auto-masking is used, clear the entry */
+ rte_wmb();
+ rte_intr_enable(&dev->pci_dev->intr_handle);
+ } else {
+ /* Make sure all updates are written before un-masking */
+ rte_wmb();
+ nn_cfg_writeb(hw, NFP_NET_CFG_ICR(NFP_NET_IRQ_LSC_IDX),
+ NFP_NET_CFG_ICR_UNMASKED);
+ }
+}
+
+static void
+nfp_net_dev_interrupt_handler(__rte_unused struct rte_intr_handle *handle,
+ void *param)
+{
+ int64_t timeout;
+ struct rte_eth_link link;
+ struct rte_eth_dev *dev = (struct rte_eth_dev *)param;
+
+ PMD_DRV_LOG(DEBUG, "We got a LSC interrupt!!!\n");
+
+ /* get the link status */
+ memset(&link, 0, sizeof(link));
+ nfp_net_dev_atomic_read_link_status(dev, &link);
+
+ nfp_net_link_update(dev, 0);
+
+ /* likely to up */
+ if (!link.link_status) {
+ /* handle it 1 sec later, wait it being stable */
+ timeout = NFP_NET_LINK_UP_CHECK_TIMEOUT;
+ /* likely to down */
+ } else {
+ /* handle it 4 sec later, wait it being stable */
+ timeout = NFP_NET_LINK_DOWN_CHECK_TIMEOUT;
+ }
+
+ if (rte_eal_alarm_set(timeout * 1000,
+ nfp_net_dev_interrupt_delayed_handler,
+ (void *)dev) < 0) {
+ RTE_LOG(ERR, PMD, "Error setting alarm");
+ /* Unmasking */
+ nfp_net_irq_unmask(dev);
+ }
+}
+
+/*
+ * Interrupt handler which shall be registered for alarm callback for delayed
+ * handling specific interrupt to wait for the stable nic state. As the NIC
+ * interrupt state is not stable for nfp after link is just down, it needs
+ * to wait 4 seconds to get the stable status.
+ *
+ * @param handle Pointer to interrupt handle.
+ * @param param The address of parameter (struct rte_eth_dev *)
+ *
+ * @return void
+ */
+static void
+nfp_net_dev_interrupt_delayed_handler(void *param)
+{
+ struct rte_eth_dev *dev = (struct rte_eth_dev *)param;
+
+ nfp_net_link_update(dev, 0);
+ _rte_eth_dev_callback_process(dev, RTE_ETH_EVENT_INTR_LSC);
+
+ nfp_net_dev_link_status_print(dev);
+
+ /* Unmasking */
+ nfp_net_irq_unmask(dev);
+}
+
static int
nfp_net_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu)
{
@@ -2315,6 +2427,17 @@ nfp_net_init(struct rte_eth_dev *eth_dev)
hw->mac_addr[0], hw->mac_addr[1], hw->mac_addr[2],
hw->mac_addr[3], hw->mac_addr[4], hw->mac_addr[5]);
+ /* Registering LSC interrupt handler */
+ rte_intr_callback_register(&pci_dev->intr_handle,
+ nfp_net_dev_interrupt_handler,
+ (void *)eth_dev);
+
+ /* enable uio intr after callback register */
+ rte_intr_enable(&pci_dev->intr_handle);
+
+ /* Telling the firmware about the LSC interrupt entry */
+ nn_cfg_writeb(hw, NFP_NET_CFG_LSC, NFP_NET_IRQ_LSC_IDX);
+
/* Recording current stats counters values */
nfp_net_stats_reset(eth_dev);
--
1.7.9.5
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH v10 8/8] nfp: adding nic guide
2015-11-30 10:25 [PATCH v10 0/8] support for netronome nfp-6xxx card Alejandro Lucero
` (6 preceding siblings ...)
2015-11-30 10:25 ` [PATCH v10 7/8] nfp: link status change interrupt support Alejandro Lucero
@ 2015-11-30 10:25 ` Alejandro Lucero
2015-12-07 23:47 ` [PATCH v10 0/8] support for netronome nfp-6xxx card Thomas Monjalon
8 siblings, 0 replies; 10+ messages in thread
From: Alejandro Lucero @ 2015-11-30 10:25 UTC (permalink / raw)
To: dev
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Signed-off-by: Rolf Neugebauer <rolf.neugebauer@netronome.com>
---
MAINTAINERS | 1 +
doc/guides/nics/index.rst | 1 +
doc/guides/nics/nfp.rst | 265 +++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 267 insertions(+)
create mode 100644 doc/guides/nics/nfp.rst
diff --git a/MAINTAINERS b/MAINTAINERS
index a23de04..b5db75f 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -338,6 +338,7 @@ F: drivers/crypto/qat/
Netronome nfp
M: Alejandro Lucero <alejandro.lucero@netronome.com>
F: drivers/net/nfp/
+F: doc/guides/nics/nfp.rst
Packet processing
-----------------
diff --git a/doc/guides/nics/index.rst b/doc/guides/nics/index.rst
index 0a0b724..7bf2938 100644
--- a/doc/guides/nics/index.rst
+++ b/doc/guides/nics/index.rst
@@ -46,6 +46,7 @@ Network Interface Controller Drivers
intel_vf
mlx4
mlx5
+ nfp
szedata2
virtio
vmxnet3
diff --git a/doc/guides/nics/nfp.rst b/doc/guides/nics/nfp.rst
new file mode 100644
index 0000000..55ba64d
--- /dev/null
+++ b/doc/guides/nics/nfp.rst
@@ -0,0 +1,265 @@
+.. BSD LICENSE
+ Copyright(c) 2015 Netronome Systems, Inc. All rights reserved.
+ All rights reserved.
+
+ Redistribution and use in source and binary forms, with or without
+ modification, are permitted provided that the following conditions
+ are met:
+
+ * Redistributions of source code must retain the above copyright
+ notice, this list of conditions and the following disclaimer.
+ * Redistributions in binary form must reproduce the above copyright
+ notice, this list of conditions and the following disclaimer in
+ the documentation and/or other materials provided with the
+ distribution.
+ * Neither the name of Intel Corporation nor the names of its
+ contributors may be used to endorse or promote products derived
+ from this software without specific prior written permission.
+
+ THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+NFP poll mode driver library
+============================
+
+Netronome's sixth generation of flow processors pack 216 programmable
+cores and over 100 hardware accelerators that uniquely combine packet,
+flow, security and content processing in a single device that scales
+up to 400 Gbps.
+
+This document explains how to use DPDK with the Netronome Poll Mode
+Driver (PMD) supporting Netronome's Network Flow Processor 6xxx
+(NFP-6xxx).
+
+Currently the driver supports virtual functions (VFs) only.
+
+Dependencies
+------------
+
+Before using the Netronome's DPDK PMD some NFP-6xxx configuration,
+which is not related to DPDK, is required. The system requires
+installation of **Netronome's BSP (Board Support Package)** which includes
+Linux drivers, programs and libraries.
+
+If you have a NFP-6xxx device you should already have the code and
+documentation for doing this configuration. Contact
+**support@netronome.com** to obtain the latest available firmware.
+
+The NFP Linux kernel drivers (including the required PF driver for the
+NFP) are available on Github at
+**https://github.com/Netronome/nfp-drv-kmods** along with build
+instructions.
+
+DPDK runs in userspace and PMDs uses the Linux kernel UIO interface to
+allow access to physical devices from userspace. The NFP PMD requires
+a separate UIO driver, **nfp_uio**, to perform correct
+initialization. This driver is part of Netronome´s BSP and it is
+equivalent to Intel's igb_uio driver.
+
+Building the software
+---------------------
+
+Netronome's PMD code is provided in the **drivers/net/nfp** directory.
+Because Netronome´s BSP dependencies the driver is disabled by default
+in DPDK build using **common_linuxapp configuration** file. Enabling the
+driver or if you use another configuration file and want to have NFP
+support, this variable is needed:
+
+- **CONFIG_RTE_LIBRTE_NFP_PMD=y**
+
+Once DPDK is built all the DPDK apps and examples include support for
+the NFP PMD.
+
+
+System configuration
+--------------------
+
+Using the NFP PMD is not different to using other PMDs. Usual steps are:
+
+#. **Configure hugepages:** All major Linux distributions have the hugepages
+ functionality enabled by default. By default this allows the system uses for
+ working with transparent hugepages. But in this case some hugepages need to
+ be created/reserved for use with the DPDK through the hugetlbfs file system.
+ First the virtual file system need to be mounted:
+
+ .. code-block:: console
+
+ mount -t hugetlbfs none /mnt/hugetlbfs
+
+ The command uses the common mount point for this file system and it needs to
+ be created if necessary.
+
+ Configuring hugepages is performed via sysfs:
+
+ .. code-block:: console
+
+ /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
+
+ This sysfs file is used to specify the number of hugepages to reserve.
+ For example:
+
+ .. code-block:: console
+
+ echo 1024 > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
+
+ This will reserve 2GB of memory using 1024 2MB hugepages. The file may be
+ read to see if the operation was performed correctly:
+
+ .. code-block:: console
+
+ cat /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
+
+ The number of unused hugepages may also be inspected.
+
+ Before executing the DPDK app it should match the value of nr_hugepages.
+
+ .. code-block:: console
+
+ cat /sys/kernel/mm/hugepages/hugepages-2048kB/free_hugepages
+
+ The hugepages reservation should be performed at system initialisation and
+ it is usual to use a kernel parameter for configuration. If the reservation
+ is attempted on a busy system it will likely fail. Reserving memory for
+ hugepages may be done adding the following to the grub kernel command line:
+
+ .. code-block:: console
+
+ default_hugepagesz=1M hugepagesz=2M hugepages=1024
+
+ This will reserve 2GBytes of memory using 2Mbytes huge pages.
+
+ Finally, for a NUMA system the allocation needs to be made on the correct
+ NUMA node. In a DPDK app there is a master core which will (usually) perform
+ memory allocation. It is important that some of the hugepages are reserved
+ on the NUMA memory node where the network device is attached. This is because
+ of a restriction in DPDK by which TX and RX descriptors rings must be created
+ on the master code.
+
+ Per-node allocation of hugepages may be inspected and controlled using sysfs.
+ For example:
+
+ .. code-block:: console
+
+ cat /sys/devices/system/node/node0/hugepages/hugepages-2048kB/nr_hugepages
+
+ For a NUMA system there will be a specific hugepage directory per node
+ allowing control of hugepage reservation. A common problem may occur when
+ hugepages reservation is performed after the system has been working for
+ some time. Configuration using the global sysfs hugepage interface will
+ succeed but the per-node allocations may be unsatisfactory.
+
+ The number of hugepages that need to be reserved depends on how the app uses
+ TX and RX descriptors, and packets mbufs.
+
+#. **Enable SR-IOV on the NFP-6xxx device:** The current NFP PMD works with
+ Virtual Functions (VFs) on a NFP device. Make sure that one of the Physical
+ Function (PF) drivers from the above Github repository is installed and
+ loaded.
+
+ Virtual Functions need to be enabled before they can be used with the PMD.
+ Before enabling the VFs it is useful to obtain information about the
+ current NFP PCI device detected by the system:
+
+ .. code-block:: console
+
+ lspci -d19ee:
+
+ Now, for example, configure two virtual functions on a NFP-6xxx device
+ whose PCI system identity is "0000:03:00.0":
+
+ .. code-block:: console
+
+ echo 2 > /sys/bus/pci/devices/0000:03:00.0/sriov_numvfs
+
+ The result of this command may be shown using lspci again:
+
+ .. code-block:: console
+
+ lspci -d19ee: -k
+
+ Two new PCI devices should appear in the output of the above command. The
+ -k option shows the device driver, if any, that devices are bound to.
+ Depending on the modules loaded at this point the new PCI devices may be
+ bound to nfp_netvf driver.
+
+#. **To install the uio kernel module (manually):** All major Linux
+ distributions have support for this kernel module so it is straightforward
+ to install it:
+
+ .. code-block:: console
+
+ modprobe uio
+
+ The module should now be listed by the lsmod command.
+
+#. **To install the nfp_uio kernel module (manually):** This module supports
+ NFP-6xxx devices through the UIO interface.
+
+ This module is part of Netronome´s BSP and it should be available when the
+ BSP is installed.
+
+ .. code-block:: console
+
+ modprobe nfp_uio.ko
+
+ The module should now be listed by the lsmod command.
+
+ Depending on which NFP modules are loaded, nfp_uio may be automatically
+ bound to the NFP PCI devices by the system. Otherwise the binding needs
+ to be done explicitly. This is the case when nfp_netvf, the Linux kernel
+ driver for NFP VFs, was loaded when VFs were created. As described later
+ in this document this configuration may also be performed using scripts
+ provided by the Netronome´s BSP.
+
+ First the device needs to be unbound, for example from the nfp_netvf
+ driver:
+
+ .. code-block:: console
+
+ echo 0000:03:08.0 > /sys/bus/pci/devices/0000:03:08.0/driver/unbind
+
+ lspci -d19ee: -k
+
+ The output of lspci should now show that 0000:03:08.0 is not bound to
+ any driver.
+
+ The next step is to add the NFP PCI ID to the NFP UIO driver:
+
+ .. code-block:: console
+
+ echo 19ee 6003 > /sys/bus/pci/drivers/nfp_uio/new_id
+
+ And then to bind the device to the nfp_uio driver:
+
+ .. code-block:: console
+
+ echo 0000:03:08.0 > /sys/bus/pci/drivers/nfp_uio/bind
+
+ lspci -d19ee: -k
+
+ lspci should show that device bound to nfp_uio driver.
+
+#. **Using tools from Netronome´s BSP to install and bind modules:** DPDK provides
+ scripts which are useful for installing the UIO modules and for binding the
+ right device to those modules avoiding doing so manually. However, these scripts
+ have not support for Netronome´s UIO driver. Along with drivers, the BSP installs
+ those DPDK scripts slightly modified with support for Netronome´s UIO driver.
+
+ Those specific scripts can be found in Netronome´s BSP installation directory.
+ Refer to BSP documentation for more information.
+
+ * **setup.sh**
+ * **dpdk_nic_bind.py**
+
+ Configuration may be performed by running setup.sh which invokes
+ dpdk_nic_bind.py as needed. Executing setup.sh will display a menu of
+ configuration options.
--
1.7.9.5
^ permalink raw reply related [flat|nested] 10+ messages in thread* Re: [PATCH v10 0/8] support for netronome nfp-6xxx card
2015-11-30 10:25 [PATCH v10 0/8] support for netronome nfp-6xxx card Alejandro Lucero
` (7 preceding siblings ...)
2015-11-30 10:25 ` [PATCH v10 8/8] nfp: adding nic guide Alejandro Lucero
@ 2015-12-07 23:47 ` Thomas Monjalon
8 siblings, 0 replies; 10+ messages in thread
From: Thomas Monjalon @ 2015-12-07 23:47 UTC (permalink / raw)
To: Alejandro Lucero; +Cc: dev
2015-11-30 10:25, Alejandro Lucero:
> This patchset adds a new PMD for Netronome nfp-6xxx card.
> Just PCI Virtual Functions support.
> Using this PMD requires previous Netronome BSP installation.
There is no risk to merge this driver disabled by default.
Applied, thanks
^ permalink raw reply [flat|nested] 10+ messages in thread