* [RFC 1/7] net/af_xdp: new PMD driver
2018-02-27 9:32 [RFC 0/7] PMD driver for AF_XDP Qi Zhang
@ 2018-02-27 9:33 ` Qi Zhang
2018-02-28 23:40 ` Stephen Hemminger
` (3 more replies)
2018-02-27 9:33 ` [RFC 2/7] lib/mbuf: enable parse flags when create mempool Qi Zhang
` (6 subsequent siblings)
7 siblings, 4 replies; 24+ messages in thread
From: Qi Zhang @ 2018-02-27 9:33 UTC (permalink / raw)
To: dev; +Cc: magnus.karlsson, bjorn.topel, Qi Zhang
This is the vanilla version.
Packet data will copy between af_xdp memory buffer and mbuf mempool.
indexes of memory buffer is simply managed by a fifo ring.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
---
config/common_base | 5 +
config/common_linuxapp | 1 +
drivers/net/Makefile | 1 +
drivers/net/af_xdp/Makefile | 56 ++
drivers/net/af_xdp/meson.build | 7 +
drivers/net/af_xdp/rte_eth_af_xdp.c | 763 ++++++++++++++++++++++++++
drivers/net/af_xdp/rte_pmd_af_xdp_version.map | 4 +
drivers/net/af_xdp/xdpsock_queue.h | 62 +++
mk/rte.app.mk | 1 +
9 files changed, 900 insertions(+)
create mode 100644 drivers/net/af_xdp/Makefile
create mode 100644 drivers/net/af_xdp/meson.build
create mode 100644 drivers/net/af_xdp/rte_eth_af_xdp.c
create mode 100644 drivers/net/af_xdp/rte_pmd_af_xdp_version.map
create mode 100644 drivers/net/af_xdp/xdpsock_queue.h
diff --git a/config/common_base b/config/common_base
index ad03cf433..84b7b3b7e 100644
--- a/config/common_base
+++ b/config/common_base
@@ -368,6 +368,11 @@ CONFIG_RTE_LIBRTE_VMXNET3_DEBUG_TX_FREE=n
CONFIG_RTE_LIBRTE_PMD_AF_PACKET=n
#
+# Compile software PMD backed by AF_XDP sockets (Linux only)
+#
+CONFIG_RTE_LIBRTE_PMD_AF_XDP=n
+
+#
# Compile link bonding PMD library
#
CONFIG_RTE_LIBRTE_PMD_BOND=y
diff --git a/config/common_linuxapp b/config/common_linuxapp
index ff98f2355..3b10695b6 100644
--- a/config/common_linuxapp
+++ b/config/common_linuxapp
@@ -16,6 +16,7 @@ CONFIG_RTE_LIBRTE_VHOST=y
CONFIG_RTE_LIBRTE_VHOST_NUMA=y
CONFIG_RTE_LIBRTE_PMD_VHOST=y
CONFIG_RTE_LIBRTE_PMD_AF_PACKET=y
+CONFIG_RTE_LIBRTE_PMD_AF_XDP=y
CONFIG_RTE_LIBRTE_PMD_TAP=y
CONFIG_RTE_LIBRTE_AVP_PMD=y
CONFIG_RTE_LIBRTE_VDEV_NETVSC_PMD=y
diff --git a/drivers/net/Makefile b/drivers/net/Makefile
index e1127326b..409234ac3 100644
--- a/drivers/net/Makefile
+++ b/drivers/net/Makefile
@@ -9,6 +9,7 @@ ifeq ($(CONFIG_RTE_LIBRTE_THUNDERX_NICVF_PMD),d)
endif
DIRS-$(CONFIG_RTE_LIBRTE_PMD_AF_PACKET) += af_packet
+DIRS-$(CONFIG_RTE_LIBRTE_PMD_AF_XDP) += af_xdp
DIRS-$(CONFIG_RTE_LIBRTE_ARK_PMD) += ark
DIRS-$(CONFIG_RTE_LIBRTE_AVF_PMD) += avf
DIRS-$(CONFIG_RTE_LIBRTE_AVP_PMD) += avp
diff --git a/drivers/net/af_xdp/Makefile b/drivers/net/af_xdp/Makefile
new file mode 100644
index 000000000..ac38e20bf
--- /dev/null
+++ b/drivers/net/af_xdp/Makefile
@@ -0,0 +1,56 @@
+# BSD LICENSE
+#
+# Copyright(c) 2014 John W. Linville <linville@redhat.com>
+# Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+# Copyright(c) 2014 6WIND S.A.
+# 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_af_xdp.a
+
+EXPORT_MAP := rte_pmd_af_xdp_version.map
+
+LIBABIVER := 1
+
+CFLAGS += -O3 -I/opt/af_xdp/linux_headers/include
+CFLAGS += $(WERROR_FLAGS)
+LDLIBS += -lrte_eal -lrte_mbuf -lrte_mempool -lrte_ring
+LDLIBS += -lrte_ethdev -lrte_net -lrte_kvargs
+LDLIBS += -lrte_bus_vdev
+
+#
+# all source are stored in SRCS-y
+#
+SRCS-$(CONFIG_RTE_LIBRTE_PMD_AF_XDP) += rte_eth_af_xdp.c
+
+include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/net/af_xdp/meson.build b/drivers/net/af_xdp/meson.build
new file mode 100644
index 000000000..4b5299c8e
--- /dev/null
+++ b/drivers/net/af_xdp/meson.build
@@ -0,0 +1,7 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2017 Intel Corporation
+
+if host_machine.system() != 'linux'
+ build = false
+endif
+sources = files('rte_eth_af_xdp.c')
diff --git a/drivers/net/af_xdp/rte_eth_af_xdp.c b/drivers/net/af_xdp/rte_eth_af_xdp.c
new file mode 100644
index 000000000..4eb8a2c28
--- /dev/null
+++ b/drivers/net/af_xdp/rte_eth_af_xdp.c
@@ -0,0 +1,763 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2014 John W. Linville <linville@tuxdriver.com>
+ * Originally based upon librte_pmd_pcap code:
+ * Copyright(c) 2010-2015 Intel Corporation.
+ * Copyright(c) 2014 6WIND S.A.
+ * All rights reserved.
+ */
+
+#include <rte_mbuf.h>
+#include <rte_ethdev_driver.h>
+#include <rte_ethdev_vdev.h>
+#include <rte_malloc.h>
+#include <rte_kvargs.h>
+#include <rte_bus_vdev.h>
+
+#include <linux/if_ether.h>
+#include <linux/if_xdp.h>
+#include <arpa/inet.h>
+#include <net/if.h>
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <sys/ioctl.h>
+#include <sys/mman.h>
+#include <unistd.h>
+#include <poll.h>
+#include "xdpsock_queue.h"
+
+#ifndef SOL_XDP
+#define SOL_XDP 283
+#endif
+
+#ifndef AF_XDP
+#define AF_XDP 44
+#endif
+
+#ifndef PF_XDP
+#define PF_XDP AF_XDP
+#endif
+
+#define ETH_AF_XDP_IFACE_ARG "iface"
+#define ETH_AF_XDP_QUEUE_IDX_ARG "queue"
+#define ETH_AF_XDP_RING_SIZE_ARG "ringsz"
+
+#define ETH_AF_XDP_FRAME_SIZE 2048
+#define ETH_AF_XDP_NUM_BUFFERS 131072
+#define ETH_AF_XDP_DATA_HEADROOM 0
+#define ETH_AF_XDP_DFLT_RING_SIZE 1024
+#define ETH_AF_XDP_DFLT_QUEUE_IDX 0
+
+#define ETH_AF_XDP_RX_BATCH_SIZE 32
+#define ETH_AF_XDP_TX_BATCH_SIZE 32
+
+struct xdp_umem {
+ char *buffer;
+ size_t size;
+ unsigned int frame_size;
+ unsigned int frame_size_log2;
+ unsigned int nframes;
+ int mr_fd;
+};
+
+struct pmd_internals {
+ int sfd;
+ int if_index;
+ char if_name[0x100];
+ struct ether_addr eth_addr;
+ struct xdp_queue rx;
+ struct xdp_queue tx;
+ struct xdp_umem *umem;
+ struct rte_mempool *mb_pool;
+
+ unsigned long rx_pkts;
+ unsigned long rx_bytes;
+ unsigned long rx_dropped;
+
+ unsigned long tx_pkts;
+ unsigned long err_pkts;
+ unsigned long tx_bytes;
+
+ uint16_t port_id;
+ uint16_t queue_idx;
+ int ring_size;
+ struct rte_ring *buf_ring;
+};
+
+static const char * const valid_arguments[] = {
+ ETH_AF_XDP_IFACE_ARG,
+ ETH_AF_XDP_QUEUE_IDX_ARG,
+ ETH_AF_XDP_RING_SIZE_ARG,
+ NULL
+};
+
+static struct rte_eth_link pmd_link = {
+ .link_speed = ETH_SPEED_NUM_10G,
+ .link_duplex = ETH_LINK_FULL_DUPLEX,
+ .link_status = ETH_LINK_DOWN,
+ .link_autoneg = ETH_LINK_AUTONEG
+};
+
+static void *get_pkt_data(struct pmd_internals *internals,
+ uint32_t index,
+ uint32_t offset)
+{
+ return (uint8_t *)(internals->umem->buffer +
+ (index << internals->umem->frame_size_log2) +
+ offset);
+}
+
+static uint16_t
+eth_af_xdp_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
+{
+ struct pmd_internals *internals = queue;
+ struct xdp_queue *rxq = &internals->rx;
+ struct rte_mbuf *mbuf;
+ unsigned long dropped = 0;
+ unsigned long rx_bytes = 0;
+ uint16_t count = 0;
+
+ nb_pkts = nb_pkts < ETH_AF_XDP_RX_BATCH_SIZE ?
+ nb_pkts : ETH_AF_XDP_RX_BATCH_SIZE;
+
+ struct xdp_desc descs[ETH_AF_XDP_RX_BATCH_SIZE];
+ void *indexes[ETH_AF_XDP_RX_BATCH_SIZE];
+ int rcvd, i;
+ /* fill rx ring */
+ if (rxq->num_free >= ETH_AF_XDP_RX_BATCH_SIZE) {
+ int n = rte_ring_dequeue_bulk(internals->buf_ring,
+ indexes,
+ ETH_AF_XDP_RX_BATCH_SIZE,
+ NULL);
+ for (i = 0; i < n; i++)
+ descs[i].idx = (uint32_t)((long int)indexes[i]);
+ xq_enq(rxq, descs, n);
+ }
+
+ /* read data */
+ rcvd = xq_deq(rxq, descs, nb_pkts);
+ if (rcvd == 0)
+ return 0;
+
+ for (i = 0; i < rcvd; i++) {
+ char *pkt;
+ uint32_t idx = descs[i].idx;
+
+ mbuf = rte_pktmbuf_alloc(internals->mb_pool);
+ rte_pktmbuf_pkt_len(mbuf) =
+ rte_pktmbuf_data_len(mbuf) =
+ descs[i].len;
+ if (mbuf) {
+ pkt = get_pkt_data(internals, idx, descs[i].offset);
+ memcpy(rte_pktmbuf_mtod(mbuf, void *),
+ pkt, descs[i].len);
+ rx_bytes += descs[i].len;
+ bufs[count++] = mbuf;
+ } else {
+ dropped++;
+ }
+ indexes[i] = (void *)((long int)idx);
+ }
+
+ rte_ring_enqueue_bulk(internals->buf_ring, indexes, rcvd, NULL);
+
+ internals->rx_pkts += (rcvd - dropped);
+ internals->rx_bytes += rx_bytes;
+ internals->rx_dropped += dropped;
+
+ return count;
+}
+
+static void kick_tx(int fd)
+{
+ int ret;
+
+ for (;;) {
+ ret = sendto(fd, NULL, 0, MSG_DONTWAIT, NULL, 0);
+ if (ret >= 0 || errno == ENOBUFS)
+ return;
+ if (errno == EAGAIN)
+ continue;
+ }
+}
+
+static uint16_t
+eth_af_xdp_tx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
+{
+ struct pmd_internals *internals = queue;
+ struct xdp_queue *txq = &internals->tx;
+ struct rte_mbuf *mbuf;
+ struct xdp_desc descs[ETH_AF_XDP_TX_BATCH_SIZE];
+ void *indexes[ETH_AF_XDP_TX_BATCH_SIZE];
+ uint16_t i, valid;
+ unsigned long tx_bytes = 0;
+
+ nb_pkts = nb_pkts < ETH_AF_XDP_TX_BATCH_SIZE ?
+ nb_pkts : ETH_AF_XDP_TX_BATCH_SIZE;
+
+ if (txq->num_free < ETH_AF_XDP_TX_BATCH_SIZE * 2) {
+ int n = xq_deq(txq, descs, ETH_AF_XDP_TX_BATCH_SIZE);
+
+ for (i = 0; i < n; i++)
+ indexes[i] = (void *)((long int)descs[i].idx);
+ rte_ring_enqueue_bulk(internals->buf_ring, indexes, n, NULL);
+ }
+
+ nb_pkts = nb_pkts > txq->num_free ? txq->num_free : nb_pkts;
+ nb_pkts = rte_ring_dequeue_bulk(internals->buf_ring, indexes,
+ nb_pkts, NULL);
+
+ valid = 0;
+ for (i = 0; i < nb_pkts; i++) {
+ char *pkt;
+ unsigned int buf_len =
+ internals->umem->frame_size - ETH_AF_XDP_DATA_HEADROOM;
+ mbuf = bufs[i];
+ if (mbuf->pkt_len <= buf_len) {
+ descs[valid].idx = (uint32_t)((long int)indexes[valid]);
+ descs[valid].offset = ETH_AF_XDP_DATA_HEADROOM;
+ descs[valid].flags = 0;
+ descs[valid].len = mbuf->pkt_len;
+ pkt = get_pkt_data(internals, descs[i].idx,
+ descs[i].offset);
+ memcpy(pkt, rte_pktmbuf_mtod(mbuf, void *),
+ descs[i].len);
+ valid++;
+ tx_bytes += mbuf->pkt_len;
+ }
+ rte_pktmbuf_free(mbuf);
+ }
+
+ xq_enq(txq, descs, valid);
+ kick_tx(internals->sfd);
+
+ if (valid < nb_pkts)
+ rte_ring_enqueue_bulk(internals->buf_ring, &indexes[valid],
+ nb_pkts - valid, NULL);
+
+ internals->err_pkts += (nb_pkts - valid);
+ internals->tx_pkts += valid;
+ internals->tx_bytes += tx_bytes;
+
+ return valid;
+}
+
+static void
+fill_rx_desc(struct pmd_internals *internals)
+{
+ int num_free = internals->rx.num_free;
+ void *p = NULL;
+ int i;
+
+ for (i = 0; i < num_free; i++) {
+ struct xdp_desc desc = {};
+
+ rte_ring_dequeue(internals->buf_ring, &p);
+ desc.idx = (uint32_t)((long int)p);
+ xq_enq(&internals->rx, &desc, 1);
+ }
+}
+
+static int
+eth_dev_start(struct rte_eth_dev *dev)
+{
+ struct pmd_internals *internals = dev->data->dev_private;
+
+ dev->data->dev_link.link_status = ETH_LINK_UP;
+ fill_rx_desc(internals);
+
+ return 0;
+}
+
+/* This function gets called when the current port gets stopped. */
+static void
+eth_dev_stop(struct rte_eth_dev *dev)
+{
+ dev->data->dev_link.link_status = ETH_LINK_DOWN;
+}
+
+static int
+eth_dev_configure(struct rte_eth_dev *dev __rte_unused)
+{
+ return 0;
+}
+
+static void
+eth_dev_info(struct rte_eth_dev *dev, struct rte_eth_dev_info *dev_info)
+{
+ struct pmd_internals *internals = dev->data->dev_private;
+
+ dev_info->if_index = internals->if_index;
+ dev_info->max_mac_addrs = 1;
+ dev_info->max_rx_pktlen = (uint32_t)ETH_FRAME_LEN;
+ dev_info->max_rx_queues = 1;
+ dev_info->max_tx_queues = 1;
+ dev_info->min_rx_bufsize = 0;
+}
+
+static int
+eth_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *stats)
+{
+ const struct pmd_internals *internal = dev->data->dev_private;
+
+ stats->ipackets = stats->q_ipackets[0] =
+ internal->rx_pkts;
+ stats->ibytes = stats->q_ibytes[0] =
+ internal->rx_bytes;
+ stats->imissed =
+ internal->rx_dropped;
+
+ stats->opackets = stats->q_opackets[0]
+ = internal->tx_pkts;
+ stats->oerrors = stats->q_errors[0] =
+ internal->err_pkts;
+ stats->obytes = stats->q_obytes[0] =
+ internal->tx_bytes;
+
+ return 0;
+}
+
+static void
+eth_stats_reset(struct rte_eth_dev *dev)
+{
+ struct pmd_internals *internal = dev->data->dev_private;
+
+ internal->rx_pkts = 0;
+ internal->rx_bytes = 0;
+ internal->rx_dropped = 0;
+
+ internal->tx_pkts = 0;
+ internal->err_pkts = 0;
+ internal->tx_bytes = 0;
+}
+
+static void
+eth_dev_close(struct rte_eth_dev *dev __rte_unused)
+{
+}
+
+static void
+eth_queue_release(void *q __rte_unused)
+{
+}
+
+static int
+eth_link_update(struct rte_eth_dev *dev __rte_unused,
+ int wait_to_complete __rte_unused)
+{
+ return 0;
+}
+
+static struct xdp_umem *xsk_alloc_and_mem_reg_buffers(int sfd, size_t nbuffers)
+{
+ struct xdp_mr_req req = { .frame_size = ETH_AF_XDP_FRAME_SIZE,
+ .data_headroom = ETH_AF_XDP_DATA_HEADROOM };
+ struct xdp_umem *umem;
+ void *bufs;
+ int ret;
+
+ ret = posix_memalign((void **)&bufs, getpagesize(),
+ nbuffers * req.frame_size);
+ if (ret)
+ return NULL;
+
+ umem = calloc(1, sizeof(*umem));
+ if (!umem) {
+ free(bufs);
+ return NULL;
+ }
+
+ req.addr = (unsigned long)bufs;
+ req.len = nbuffers * req.frame_size;
+ ret = setsockopt(sfd, SOL_XDP, XDP_MEM_REG, &req, sizeof(req));
+ RTE_ASSERT(ret == 0);
+
+ umem->frame_size = ETH_AF_XDP_FRAME_SIZE;
+ umem->frame_size_log2 = 11;
+ umem->buffer = bufs;
+ umem->size = nbuffers * req.frame_size;
+ umem->nframes = nbuffers;
+ umem->mr_fd = sfd;
+
+ return umem;
+}
+
+static int
+xdp_configure(struct pmd_internals *internals)
+{
+ struct sockaddr_xdp sxdp;
+ struct xdp_ring_req req;
+ char ring_name[0x100];
+ int ret = 0;
+ long int i;
+
+ snprintf(ring_name, 0x100, "%s_%s_%d", "af_xdp_ring",
+ internals->if_name, internals->queue_idx);
+ internals->buf_ring = rte_ring_create(ring_name,
+ ETH_AF_XDP_NUM_BUFFERS,
+ SOCKET_ID_ANY,
+ 0x0);
+ if (!internals->buf_ring)
+ return -1;
+
+ for (i = 0; i < ETH_AF_XDP_NUM_BUFFERS; i++)
+ rte_ring_enqueue(internals->buf_ring, (void *)i);
+
+ internals->umem = xsk_alloc_and_mem_reg_buffers(internals->sfd,
+ ETH_AF_XDP_NUM_BUFFERS);
+ if (!internals->umem)
+ goto error;
+
+ req.mr_fd = internals->umem->mr_fd;
+ req.desc_nr = internals->ring_size;
+
+ ret = setsockopt(internals->sfd, SOL_XDP, XDP_RX_RING,
+ &req, sizeof(req));
+
+ RTE_ASSERT(ret == 0);
+
+ ret = setsockopt(internals->sfd, SOL_XDP, XDP_TX_RING,
+ &req, sizeof(req));
+
+ RTE_ASSERT(ret == 0);
+
+ internals->rx.ring = mmap(0, req.desc_nr * sizeof(struct xdp_desc),
+ PROT_READ | PROT_WRITE,
+ MAP_SHARED | MAP_LOCKED | MAP_POPULATE,
+ internals->sfd,
+ XDP_PGOFF_RX_RING);
+ RTE_ASSERT(internals->rx.ring != MAP_FAILED);
+
+ internals->rx.num_free = req.desc_nr;
+ internals->rx.ring_mask = req.desc_nr - 1;
+
+ internals->tx.ring = mmap(0, req.desc_nr * sizeof(struct xdp_desc),
+ PROT_READ | PROT_WRITE,
+ MAP_SHARED | MAP_LOCKED | MAP_POPULATE,
+ internals->sfd,
+ XDP_PGOFF_TX_RING);
+ RTE_ASSERT(internals->tx.ring != MAP_FAILED);
+
+ internals->tx.num_free = req.desc_nr;
+ internals->tx.ring_mask = req.desc_nr - 1;
+
+ sxdp.sxdp_family = PF_XDP;
+ sxdp.sxdp_ifindex = internals->if_index;
+ sxdp.sxdp_queue_id = internals->queue_idx;
+
+ ret = bind(internals->sfd, (struct sockaddr *)&sxdp, sizeof(sxdp));
+ RTE_ASSERT(ret == 0);
+
+ return ret;
+error:
+ rte_ring_free(internals->buf_ring);
+ internals->buf_ring = NULL;
+ return -1;
+}
+
+static int
+eth_rx_queue_setup(struct rte_eth_dev *dev,
+ uint16_t rx_queue_id,
+ uint16_t nb_rx_desc __rte_unused,
+ unsigned int socket_id __rte_unused,
+ const struct rte_eth_rxconf *rx_conf __rte_unused,
+ struct rte_mempool *mb_pool)
+{
+ struct pmd_internals *internals = dev->data->dev_private;
+ unsigned int buf_size, data_size;
+
+ RTE_ASSERT(rx_queue_id == 0);
+ internals->mb_pool = mb_pool;
+ xdp_configure(internals);
+
+ /* Now get the space available for data in the mbuf */
+ buf_size = rte_pktmbuf_data_room_size(internals->mb_pool) -
+ RTE_PKTMBUF_HEADROOM;
+ data_size = internals->umem->frame_size;
+
+ if (data_size > buf_size) {
+ RTE_LOG(ERR, PMD,
+ "%s: %d bytes will not fit in mbuf (%d bytes)\n",
+ dev->device->name, data_size, buf_size);
+ return -ENOMEM;
+ }
+
+ dev->data->rx_queues[rx_queue_id] = internals;
+ return 0;
+}
+
+static int
+eth_tx_queue_setup(struct rte_eth_dev *dev,
+ uint16_t tx_queue_id,
+ uint16_t nb_tx_desc __rte_unused,
+ unsigned int socket_id __rte_unused,
+ const struct rte_eth_txconf *tx_conf __rte_unused)
+{
+ struct pmd_internals *internals = dev->data->dev_private;
+
+ RTE_ASSERT(tx_queue_id == 0);
+ dev->data->tx_queues[tx_queue_id] = internals;
+ return 0;
+}
+
+static int
+eth_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu)
+{
+ struct pmd_internals *internals = dev->data->dev_private;
+ struct ifreq ifr = { .ifr_mtu = mtu };
+ int ret;
+ int s;
+
+ s = socket(PF_INET, SOCK_DGRAM, 0);
+ if (s < 0)
+ return -EINVAL;
+
+ snprintf(ifr.ifr_name, IFNAMSIZ, "%s", internals->if_name);
+ ret = ioctl(s, SIOCSIFMTU, &ifr);
+ close(s);
+
+ if (ret < 0)
+ return -EINVAL;
+
+ return 0;
+}
+
+static void
+eth_dev_change_flags(char *if_name, uint32_t flags, uint32_t mask)
+{
+ struct ifreq ifr;
+ int s;
+
+ s = socket(PF_INET, SOCK_DGRAM, 0);
+ if (s < 0)
+ return;
+
+ snprintf(ifr.ifr_name, IFNAMSIZ, "%s", if_name);
+ if (ioctl(s, SIOCGIFFLAGS, &ifr) < 0)
+ goto out;
+ ifr.ifr_flags &= mask;
+ ifr.ifr_flags |= flags;
+ if (ioctl(s, SIOCSIFFLAGS, &ifr) < 0)
+ goto out;
+out:
+ close(s);
+}
+
+static void
+eth_dev_promiscuous_enable(struct rte_eth_dev *dev)
+{
+ struct pmd_internals *internals = dev->data->dev_private;
+
+ eth_dev_change_flags(internals->if_name, IFF_PROMISC, ~0);
+}
+
+static void
+eth_dev_promiscuous_disable(struct rte_eth_dev *dev)
+{
+ struct pmd_internals *internals = dev->data->dev_private;
+
+ eth_dev_change_flags(internals->if_name, 0, ~IFF_PROMISC);
+}
+
+static const struct eth_dev_ops ops = {
+ .dev_start = eth_dev_start,
+ .dev_stop = eth_dev_stop,
+ .dev_close = eth_dev_close,
+ .dev_configure = eth_dev_configure,
+ .dev_infos_get = eth_dev_info,
+ .mtu_set = eth_dev_mtu_set,
+ .promiscuous_enable = eth_dev_promiscuous_enable,
+ .promiscuous_disable = eth_dev_promiscuous_disable,
+ .rx_queue_setup = eth_rx_queue_setup,
+ .tx_queue_setup = eth_tx_queue_setup,
+ .rx_queue_release = eth_queue_release,
+ .tx_queue_release = eth_queue_release,
+ .link_update = eth_link_update,
+ .stats_get = eth_stats_get,
+ .stats_reset = eth_stats_reset,
+};
+
+static struct rte_vdev_driver pmd_af_xdp_drv;
+
+static void
+parse_parameters(struct rte_kvargs *kvlist,
+ char **if_name,
+ int *queue_idx,
+ int *ring_size)
+{
+ struct rte_kvargs_pair *pair = NULL;
+ unsigned int k_idx;
+
+ for (k_idx = 0; k_idx < kvlist->count; k_idx++) {
+ pair = &kvlist->pairs[k_idx];
+ if (strstr(pair->key, ETH_AF_XDP_IFACE_ARG))
+ *if_name = pair->value;
+ else if (strstr(pair->key, ETH_AF_XDP_QUEUE_IDX_ARG))
+ *queue_idx = atoi(pair->value);
+ else if (strstr(pair->key, ETH_AF_XDP_RING_SIZE_ARG))
+ *ring_size = atoi(pair->value);
+ }
+}
+
+static int
+get_iface_info(const char *if_name,
+ struct ether_addr *eth_addr,
+ int *if_index)
+{
+ struct ifreq ifr;
+ int sock = socket(AF_INET, SOCK_DGRAM, IPPROTO_IP);
+
+ if (sock < 0)
+ return -1;
+
+ strcpy(ifr.ifr_name, if_name);
+ if (ioctl(sock, SIOCGIFINDEX, &ifr))
+ goto error;
+ *if_index = ifr.ifr_ifindex;
+
+ if (ioctl(sock, SIOCGIFHWADDR, &ifr))
+ goto error;
+
+ memcpy(eth_addr, ifr.ifr_hwaddr.sa_data, 6);
+
+ close(sock);
+ return 0;
+
+error:
+ close(sock);
+ return -1;
+}
+
+static int
+init_internals(struct rte_vdev_device *dev,
+ const char *if_name,
+ int queue_idx,
+ int ring_size)
+{
+ const char *name = rte_vdev_device_name(dev);
+ struct rte_eth_dev *eth_dev = NULL;
+ struct rte_eth_dev_data *data = NULL;
+ const unsigned int numa_node = dev->device.numa_node;
+ struct pmd_internals *internals = NULL;
+ int ret;
+
+ data = rte_zmalloc_socket(name, sizeof(*internals), 0, numa_node);
+ if (!data)
+ return -1;
+
+ internals = rte_zmalloc_socket(name, sizeof(*internals), 0, numa_node);
+ if (!internals)
+ goto error_1;
+
+ internals->queue_idx = queue_idx;
+ internals->ring_size = ring_size;
+ strcpy(internals->if_name, if_name);
+ internals->sfd = socket(PF_XDP, SOCK_RAW, 0);
+ if (internals->sfd < 0)
+ goto error_2;
+
+ ret = get_iface_info(if_name, &internals->eth_addr,
+ &internals->if_index);
+ if (ret)
+ goto error_3;
+
+ eth_dev = rte_eth_vdev_allocate(dev, 0);
+ if (!eth_dev)
+ goto error_3;
+
+ rte_memcpy(data, eth_dev->data, sizeof(*data));
+ internals->port_id = eth_dev->data->port_id;
+ data->dev_private = internals;
+ data->nb_rx_queues = 1;
+ data->nb_tx_queues = 1;
+ data->dev_link = pmd_link;
+ data->mac_addrs = &internals->eth_addr;
+
+ eth_dev->data = data;
+ eth_dev->dev_ops = &ops;
+
+ eth_dev->rx_pkt_burst = eth_af_xdp_rx;
+ eth_dev->tx_pkt_burst = eth_af_xdp_tx;
+
+ return 0;
+
+error_3:
+ close(internals->sfd);
+
+error_2:
+ rte_free(internals);
+
+error_1:
+ rte_free(data);
+ return -1;
+}
+
+static int
+rte_pmd_af_xdp_probe(struct rte_vdev_device *dev)
+{
+ struct rte_kvargs *kvlist;
+ char *if_name = NULL;
+ int ring_size = ETH_AF_XDP_DFLT_RING_SIZE;
+ int queue_idx = ETH_AF_XDP_DFLT_QUEUE_IDX;
+ int ret;
+
+ RTE_LOG(INFO, PMD, "Initializing pmd_af_packet for %s\n",
+ rte_vdev_device_name(dev));
+
+ kvlist = rte_kvargs_parse(rte_vdev_device_args(dev), valid_arguments);
+ if (!kvlist) {
+ RTE_LOG(ERR, PMD,
+ "Invalid kvargs");
+ return -1;
+ }
+
+ if (dev->device.numa_node == SOCKET_ID_ANY)
+ dev->device.numa_node = rte_socket_id();
+
+ parse_parameters(kvlist, &if_name, &queue_idx, &ring_size);
+
+ ret = init_internals(dev, if_name, queue_idx, ring_size);
+ rte_kvargs_free(kvlist);
+
+ return ret;
+}
+
+static int
+rte_pmd_af_xdp_remove(struct rte_vdev_device *dev)
+{
+ struct rte_eth_dev *eth_dev = NULL;
+ struct pmd_internals *internals;
+
+ RTE_LOG(INFO, PMD, "Closing AF_XDP ethdev on numa socket %u\n",
+ rte_socket_id());
+
+ if (!dev)
+ return -1;
+
+ /* find the ethdev entry */
+ eth_dev = rte_eth_dev_allocated(rte_vdev_device_name(dev));
+ if (!eth_dev)
+ return -1;
+
+ internals = eth_dev->data->dev_private;
+ rte_ring_free(internals->buf_ring);
+ rte_free(internals->umem);
+ rte_free(eth_dev->data->dev_private);
+ rte_free(eth_dev->data);
+ close(internals->sfd);
+
+ rte_eth_dev_release_port(eth_dev);
+
+ return 0;
+}
+
+static struct rte_vdev_driver pmd_af_xdp_drv = {
+ .probe = rte_pmd_af_xdp_probe,
+ .remove = rte_pmd_af_xdp_remove,
+};
+
+RTE_PMD_REGISTER_VDEV(net_af_xdp, pmd_af_xdp_drv);
+RTE_PMD_REGISTER_ALIAS(net_af_xdp, eth_af_xdp);
+RTE_PMD_REGISTER_PARAM_STRING(net_af_xdp,
+ "iface=<string> "
+ "queue=<int> "
+ "ringsz=<int> ");
diff --git a/drivers/net/af_xdp/rte_pmd_af_xdp_version.map b/drivers/net/af_xdp/rte_pmd_af_xdp_version.map
new file mode 100644
index 000000000..ef3539840
--- /dev/null
+++ b/drivers/net/af_xdp/rte_pmd_af_xdp_version.map
@@ -0,0 +1,4 @@
+DPDK_2.0 {
+
+ local: *;
+};
diff --git a/drivers/net/af_xdp/xdpsock_queue.h b/drivers/net/af_xdp/xdpsock_queue.h
new file mode 100644
index 000000000..0dc666a08
--- /dev/null
+++ b/drivers/net/af_xdp/xdpsock_queue.h
@@ -0,0 +1,62 @@
+#ifndef __XDPSOCK_QUEUE_H
+#define __XDPSOCK_QUEUE_H
+
+static inline int xq_enq(struct xdp_queue *q,
+ const struct xdp_desc *descs,
+ unsigned int ndescs)
+{
+ unsigned int avail_idx = q->avail_idx;
+ unsigned int i;
+ int j;
+
+ if (q->num_free < ndescs)
+ return -ENOSPC;
+
+ q->num_free -= ndescs;
+
+ for (i = 0; i < ndescs; i++) {
+ unsigned int idx = avail_idx++ & q->ring_mask;
+
+ q->ring[idx].idx = descs[i].idx;
+ q->ring[idx].len = descs[i].len;
+ q->ring[idx].offset = descs[i].offset;
+ q->ring[idx].error = 0;
+ }
+ rte_smp_wmb();
+
+ for (j = ndescs - 1; j >= 0; j--) {
+ unsigned int idx = (q->avail_idx + j) & q->ring_mask;
+
+ q->ring[idx].flags = descs[j].flags | XDP_DESC_KERNEL;
+ }
+ q->avail_idx += ndescs;
+
+ return 0;
+}
+
+static inline int xq_deq(struct xdp_queue *q,
+ struct xdp_desc *descs,
+ int ndescs)
+{
+ unsigned int idx, last_used_idx = q->last_used_idx;
+ int i, entries = 0;
+
+ for (i = 0; i < ndescs; i++) {
+ idx = (last_used_idx++) & q->ring_mask;
+ if (q->ring[idx].flags & XDP_DESC_KERNEL)
+ break;
+ entries++;
+ }
+ q->num_free += entries;
+
+ rte_smp_rmb();
+
+ for (i = 0; i < entries; i++) {
+ idx = q->last_used_idx++ & q->ring_mask;
+ descs[i] = q->ring[idx];
+ }
+
+ return entries;
+}
+
+#endif /* __XDPSOCK_QUEUE_H */
diff --git a/mk/rte.app.mk b/mk/rte.app.mk
index 3eb41d176..bc26e1457 100644
--- a/mk/rte.app.mk
+++ b/mk/rte.app.mk
@@ -120,6 +120,7 @@ ifeq ($(CONFIG_RTE_BUILD_SHARED_LIB),n)
_LDLIBS-$(CONFIG_RTE_DRIVER_MEMPOOL_STACK) += -lrte_mempool_stack
_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_AF_PACKET) += -lrte_pmd_af_packet
+_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_AF_XDP) += -lrte_pmd_af_xdp
_LDLIBS-$(CONFIG_RTE_LIBRTE_ARK_PMD) += -lrte_pmd_ark
_LDLIBS-$(CONFIG_RTE_LIBRTE_AVF_PMD) += -lrte_pmd_avf
_LDLIBS-$(CONFIG_RTE_LIBRTE_AVP_PMD) += -lrte_pmd_avp
--
2.13.6
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [RFC 1/7] net/af_xdp: new PMD driver
2018-02-27 9:33 ` [RFC 1/7] net/af_xdp: new PMD driver Qi Zhang
@ 2018-02-28 23:40 ` Stephen Hemminger
2018-02-28 23:42 ` Stephen Hemminger
` (2 subsequent siblings)
3 siblings, 0 replies; 24+ messages in thread
From: Stephen Hemminger @ 2018-02-28 23:40 UTC (permalink / raw)
To: Qi Zhang; +Cc: dev, magnus.karlsson, bjorn.topel
On Tue, 27 Feb 2018 17:33:00 +0800
Qi Zhang <qi.z.zhang@intel.com> wrote:
> iff --git a/drivers/net/af_xdp/Makefile b/drivers/net/af_xdp/Makefile
> new file mode 100644
> index 000000000..ac38e20bf
> --- /dev/null
> +++ b/drivers/net/af_xdp/Makefile
> @@ -0,0 +1,56 @@
> +# BSD LICENSE
> +#
> +# Copyright(c) 2014 John W. Linville <linville@redhat.com>
> +# Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
> +# Copyright(c) 2014 6WIND S.A.
> +# 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 ADVI
Please use SPDX on new files.
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [RFC 1/7] net/af_xdp: new PMD driver
2018-02-27 9:33 ` [RFC 1/7] net/af_xdp: new PMD driver Qi Zhang
2018-02-28 23:40 ` Stephen Hemminger
@ 2018-02-28 23:42 ` Stephen Hemminger
2018-03-01 1:51 ` Zhang, Qi Z
2018-02-28 23:42 ` Stephen Hemminger
2018-02-28 23:45 ` Stephen Hemminger
3 siblings, 1 reply; 24+ messages in thread
From: Stephen Hemminger @ 2018-02-28 23:42 UTC (permalink / raw)
To: Qi Zhang; +Cc: dev, magnus.karlsson, bjorn.topel
On Tue, 27 Feb 2018 17:33:00 +0800
Qi Zhang <qi.z.zhang@intel.com> wrote:
> struct pmd_internals {
> + int sfd;
> + int if_index;
> + char if_name[0x100];
why not IFNAMSIZ?
> + struct ether_addr eth_addr;
> + struct xdp_queue rx;
> + struct xdp_queue tx;
> + struct xdp_umem *umem;
> + struct rte_mempool *mb_pool;
> +
> + unsigned long rx_pkts;
> + unsigned long rx_bytes;
> + unsigned long rx_dropped;
> +
> + unsigned long tx_pkts;
> + unsigned long err_pkts;
> + unsigned long tx_bytes;
why not per-queue stats? per-port stats are expensive
> + uint16_t port_id;
> + uint16_t queue_idx;
> + int ring_size;
> + struct rte_ring *buf_ring;
> +};
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [RFC 1/7] net/af_xdp: new PMD driver
2018-02-28 23:42 ` Stephen Hemminger
@ 2018-03-01 1:51 ` Zhang, Qi Z
0 siblings, 0 replies; 24+ messages in thread
From: Zhang, Qi Z @ 2018-03-01 1:51 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: dev, Karlsson, Magnus, Topel, Bjorn
> -----Original Message-----
> From: Stephen Hemminger [mailto:stephen@networkplumber.org]
> Sent: Thursday, March 1, 2018 7:42 AM
> To: Zhang, Qi Z <qi.z.zhang@intel.com>
> Cc: dev@dpdk.org; magnus.karlsson@intei.com; Topel, Bjorn
> <bjorn.topel@intel.com>
> Subject: Re: [dpdk-dev] [RFC 1/7] net/af_xdp: new PMD driver
>
> On Tue, 27 Feb 2018 17:33:00 +0800
> Qi Zhang <qi.z.zhang@intel.com> wrote:
>
> > struct pmd_internals {
> > + int sfd;
> > + int if_index;
> > + char if_name[0x100];
>
> why not IFNAMSIZ?
>
> > + struct ether_addr eth_addr;
> > + struct xdp_queue rx;
> > + struct xdp_queue tx;
> > + struct xdp_umem *umem;
> > + struct rte_mempool *mb_pool;
> > +
> > + unsigned long rx_pkts;
> > + unsigned long rx_bytes;
> > + unsigned long rx_dropped;
> > +
> > + unsigned long tx_pkts;
> > + unsigned long err_pkts;
> > + unsigned long tx_bytes;
>
> why not per-queue stats? per-port stats are expensive
multi-queue is not supported in this implementation, but will be considered.
Regards
Qi
>
> > + uint16_t port_id;
> > + uint16_t queue_idx;
> > + int ring_size;
> > + struct rte_ring *buf_ring;
> > +};
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [RFC 1/7] net/af_xdp: new PMD driver
2018-02-27 9:33 ` [RFC 1/7] net/af_xdp: new PMD driver Qi Zhang
2018-02-28 23:40 ` Stephen Hemminger
2018-02-28 23:42 ` Stephen Hemminger
@ 2018-02-28 23:42 ` Stephen Hemminger
2018-02-28 23:45 ` Stephen Hemminger
3 siblings, 0 replies; 24+ messages in thread
From: Stephen Hemminger @ 2018-02-28 23:42 UTC (permalink / raw)
To: Qi Zhang; +Cc: dev, magnus.karlsson, bjorn.topel
On Tue, 27 Feb 2018 17:33:00 +0800
Qi Zhang <qi.z.zhang@intel.com> wrote:
> +
> +static void *get_pkt_data(struct pmd_internals *internals,
> + uint32_t index,
> + uint32_t offset)
> +{
> + return (uint8_t *)(internals->umem->buffer +
> + (index << internals->umem->frame_size_log2) +
> + offset);
You are returning void *, cast here is unnecessary
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [RFC 1/7] net/af_xdp: new PMD driver
2018-02-27 9:33 ` [RFC 1/7] net/af_xdp: new PMD driver Qi Zhang
` (2 preceding siblings ...)
2018-02-28 23:42 ` Stephen Hemminger
@ 2018-02-28 23:45 ` Stephen Hemminger
2018-03-01 1:59 ` Zhang, Qi Z
3 siblings, 1 reply; 24+ messages in thread
From: Stephen Hemminger @ 2018-02-28 23:45 UTC (permalink / raw)
To: Qi Zhang; +Cc: dev, magnus.karlsson, bjorn.topel
On Tue, 27 Feb 2018 17:33:00 +0800
Qi Zhang <qi.z.zhang@intel.com> wrote:
> +
> +static uint16_t
> +eth_af_xdp_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
> +{
> + struct pmd_internals *internals = queue;
> + struct xdp_queue *rxq = &internals->rx;
> + struct rte_mbuf *mbuf;
> + unsigned long dropped = 0;
> + unsigned long rx_bytes = 0;
> + uint16_t count = 0;
> +
> + nb_pkts = nb_pkts < ETH_AF_XDP_RX_BATCH_SIZE ?
> + nb_pkts : ETH_AF_XDP_RX_BATCH_SIZE;
> +
Put declarations first.
Why not iterate if nb_pkts is huge?
> + struct xdp_desc descs[ETH_AF_XDP_RX_BATCH_SIZE];
> + void *indexes[ETH_AF_XDP_RX_BATCH_SIZE];
> + int rcvd, i;
> + /* fill rx ring */
> + if (rxq->num_free >= ETH_AF_XDP_RX_BATCH_SIZE) {
Blank line after declarations before code please.
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [RFC 1/7] net/af_xdp: new PMD driver
2018-02-28 23:45 ` Stephen Hemminger
@ 2018-03-01 1:59 ` Zhang, Qi Z
0 siblings, 0 replies; 24+ messages in thread
From: Zhang, Qi Z @ 2018-03-01 1:59 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: dev, magnus.karlsson, Topel, Bjorn
> -----Original Message-----
> From: Stephen Hemminger [mailto:stephen@networkplumber.org]
> Sent: Thursday, March 1, 2018 7:45 AM
> To: Zhang, Qi Z <qi.z.zhang@intel.com>
> Cc: dev@dpdk.org; magnus.karlsson@intei.com; Topel, Bjorn
> <bjorn.topel@intel.com>
> Subject: Re: [dpdk-dev] [RFC 1/7] net/af_xdp: new PMD driver
>
> On Tue, 27 Feb 2018 17:33:00 +0800
> Qi Zhang <qi.z.zhang@intel.com> wrote:
>
> > +
> > +static uint16_t
> > +eth_af_xdp_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
> > +{
> > + struct pmd_internals *internals = queue;
> > + struct xdp_queue *rxq = &internals->rx;
> > + struct rte_mbuf *mbuf;
> > + unsigned long dropped = 0;
> > + unsigned long rx_bytes = 0;
> > + uint16_t count = 0;
> > +
> > + nb_pkts = nb_pkts < ETH_AF_XDP_RX_BATCH_SIZE ?
> > + nb_pkts : ETH_AF_XDP_RX_BATCH_SIZE;
> > +
>
> Put declarations first.
> Why not iterate if nb_pkts is huge?
Yes, it is not necessary to only read one batch, just for simple implementation, will be fixed.
>
> > + struct xdp_desc descs[ETH_AF_XDP_RX_BATCH_SIZE];
> > + void *indexes[ETH_AF_XDP_RX_BATCH_SIZE];
> > + int rcvd, i;
> > + /* fill rx ring */
> > + if (rxq->num_free >= ETH_AF_XDP_RX_BATCH_SIZE) {
>
> Blank line after declarations before code please.
^ permalink raw reply [flat|nested] 24+ messages in thread
* [RFC 2/7] lib/mbuf: enable parse flags when create mempool
2018-02-27 9:32 [RFC 0/7] PMD driver for AF_XDP Qi Zhang
2018-02-27 9:33 ` [RFC 1/7] net/af_xdp: new PMD driver Qi Zhang
@ 2018-02-27 9:33 ` Qi Zhang
2018-02-27 9:33 ` [RFC 3/7] lib/mempool: allow page size aligned mempool Qi Zhang
` (5 subsequent siblings)
7 siblings, 0 replies; 24+ messages in thread
From: Qi Zhang @ 2018-02-27 9:33 UTC (permalink / raw)
To: dev; +Cc: magnus.karlsson, bjorn.topel, Qi Zhang
This give the option that applicaiton can configure each
memory chunk's size precisely. (by MEMPOOL_F_NO_SPREAD).
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
---
lib/librte_mbuf/rte_mbuf.c | 15 ++++++++++++---
lib/librte_mbuf/rte_mbuf.h | 8 +++++++-
2 files changed, 19 insertions(+), 4 deletions(-)
diff --git a/lib/librte_mbuf/rte_mbuf.c b/lib/librte_mbuf/rte_mbuf.c
index 091d388d3..5fd91c87c 100644
--- a/lib/librte_mbuf/rte_mbuf.c
+++ b/lib/librte_mbuf/rte_mbuf.c
@@ -125,7 +125,7 @@ rte_pktmbuf_init(struct rte_mempool *mp,
struct rte_mempool * __rte_experimental
rte_pktmbuf_pool_create_by_ops(const char *name, unsigned int n,
unsigned int cache_size, uint16_t priv_size, uint16_t data_room_size,
- int socket_id, const char *ops_name)
+ unsigned int flags, int socket_id, const char *ops_name)
{
struct rte_mempool *mp;
struct rte_pktmbuf_pool_private mbp_priv;
@@ -145,7 +145,7 @@ rte_pktmbuf_pool_create_by_ops(const char *name, unsigned int n,
mbp_priv.mbuf_priv_size = priv_size;
mp = rte_mempool_create_empty(name, n, elt_size, cache_size,
- sizeof(struct rte_pktmbuf_pool_private), socket_id, 0);
+ sizeof(struct rte_pktmbuf_pool_private), socket_id, flags);
if (mp == NULL)
return NULL;
@@ -179,9 +179,18 @@ rte_pktmbuf_pool_create(const char *name, unsigned int n,
int socket_id)
{
return rte_pktmbuf_pool_create_by_ops(name, n, cache_size, priv_size,
- data_room_size, socket_id, NULL);
+ data_room_size, 0, socket_id, NULL);
}
+/* helper to create a mbuf pool with NO_SPREAD */
+struct rte_mempool *
+rte_pktmbuf_pool_create_with_flags(const char *name, unsigned int n,
+ unsigned int cache_size, uint16_t priv_size, uint16_t data_room_size,
+ unsigned int flags, int socket_id)
+{
+ return rte_pktmbuf_pool_create_by_ops(name, n, cache_size, priv_size,
+ data_room_size, flags, socket_id, NULL);
+}
/* do some sanity checks on a mbuf: panic if it fails */
void
rte_mbuf_sanity_check(const struct rte_mbuf *m, int is_header)
diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
index 62740254d..6f6af42a8 100644
--- a/lib/librte_mbuf/rte_mbuf.h
+++ b/lib/librte_mbuf/rte_mbuf.h
@@ -1079,6 +1079,12 @@ rte_pktmbuf_pool_create(const char *name, unsigned n,
unsigned cache_size, uint16_t priv_size, uint16_t data_room_size,
int socket_id);
+struct rte_mempool *
+rte_pktmbuf_pool_create_with_flags(const char *name, unsigned int n,
+ unsigned cache_size, uint16_t priv_size, uint16_t data_room_size,
+ unsigned flags, int socket_id);
+
+
/**
* Create a mbuf pool with a given mempool ops name
*
@@ -1119,7 +1125,7 @@ rte_pktmbuf_pool_create(const char *name, unsigned n,
struct rte_mempool * __rte_experimental
rte_pktmbuf_pool_create_by_ops(const char *name, unsigned int n,
unsigned int cache_size, uint16_t priv_size, uint16_t data_room_size,
- int socket_id, const char *ops_name);
+ unsigned int flags, int socket_id, const char *ops_name);
/**
* Get the data room size of mbufs stored in a pktmbuf_pool
--
2.13.6
^ permalink raw reply related [flat|nested] 24+ messages in thread* [RFC 3/7] lib/mempool: allow page size aligned mempool
2018-02-27 9:32 [RFC 0/7] PMD driver for AF_XDP Qi Zhang
2018-02-27 9:33 ` [RFC 1/7] net/af_xdp: new PMD driver Qi Zhang
2018-02-27 9:33 ` [RFC 2/7] lib/mbuf: enable parse flags when create mempool Qi Zhang
@ 2018-02-27 9:33 ` Qi Zhang
2018-02-27 9:33 ` [RFC 4/7] net/af_xdp: use mbuf mempool for buffer management Qi Zhang
` (4 subsequent siblings)
7 siblings, 0 replies; 24+ messages in thread
From: Qi Zhang @ 2018-02-27 9:33 UTC (permalink / raw)
To: dev; +Cc: magnus.karlsson, bjorn.topel, Qi Zhang
Allow create a mempool with page size aligned base address.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
---
lib/librte_mempool/rte_mempool.c | 2 ++
lib/librte_mempool/rte_mempool.h | 1 +
2 files changed, 3 insertions(+)
diff --git a/lib/librte_mempool/rte_mempool.c b/lib/librte_mempool/rte_mempool.c
index 54f7f4ba4..f8d4814ad 100644
--- a/lib/librte_mempool/rte_mempool.c
+++ b/lib/librte_mempool/rte_mempool.c
@@ -567,6 +567,8 @@ rte_mempool_populate_default(struct rte_mempool *mp)
pg_shift = 0; /* not needed, zone is physically contiguous */
pg_sz = 0;
align = RTE_CACHE_LINE_SIZE;
+ if (mp->flags & MEMPOOL_F_PAGE_ALIGN)
+ align = getpagesize();
} else {
pg_sz = getpagesize();
pg_shift = rte_bsf32(pg_sz);
diff --git a/lib/librte_mempool/rte_mempool.h b/lib/librte_mempool/rte_mempool.h
index 8b1b7f7ed..774ab0f66 100644
--- a/lib/librte_mempool/rte_mempool.h
+++ b/lib/librte_mempool/rte_mempool.h
@@ -245,6 +245,7 @@ struct rte_mempool {
#define MEMPOOL_F_SC_GET 0x0008 /**< Default get is "single-consumer".*/
#define MEMPOOL_F_POOL_CREATED 0x0010 /**< Internal: pool is created. */
#define MEMPOOL_F_NO_PHYS_CONTIG 0x0020 /**< Don't need physically contiguous objs. */
+#define MEMPOOL_F_PAGE_ALIGN 0x0040 /**< Base address is page aligned. */
/**
* This capability flag is advertised by a mempool handler, if the whole
* memory area containing the objects must be physically contiguous.
--
2.13.6
^ permalink raw reply related [flat|nested] 24+ messages in thread* [RFC 4/7] net/af_xdp: use mbuf mempool for buffer management
2018-02-27 9:32 [RFC 0/7] PMD driver for AF_XDP Qi Zhang
` (2 preceding siblings ...)
2018-02-27 9:33 ` [RFC 3/7] lib/mempool: allow page size aligned mempool Qi Zhang
@ 2018-02-27 9:33 ` Qi Zhang
2018-03-01 2:08 ` Stephen Hemminger
2018-02-27 9:33 ` [RFC 5/7] net/af_xdp: enable share mempool Qi Zhang
` (3 subsequent siblings)
7 siblings, 1 reply; 24+ messages in thread
From: Qi Zhang @ 2018-02-27 9:33 UTC (permalink / raw)
To: dev; +Cc: magnus.karlsson, bjorn.topel, Qi Zhang
Now, af_xdp registered memory buffer is managed by rte_mempool.
mbuf be allocated from rte_mempool can be convert to descriptor
index and vice versa.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
---
drivers/net/af_xdp/rte_eth_af_xdp.c | 165 +++++++++++++++++++++---------------
1 file changed, 97 insertions(+), 68 deletions(-)
diff --git a/drivers/net/af_xdp/rte_eth_af_xdp.c b/drivers/net/af_xdp/rte_eth_af_xdp.c
index 4eb8a2c28..3c534c77c 100644
--- a/drivers/net/af_xdp/rte_eth_af_xdp.c
+++ b/drivers/net/af_xdp/rte_eth_af_xdp.c
@@ -43,7 +43,11 @@
#define ETH_AF_XDP_FRAME_SIZE 2048
#define ETH_AF_XDP_NUM_BUFFERS 131072
-#define ETH_AF_XDP_DATA_HEADROOM 0
+/* mempool hdrobj size (64 bytes) + sizeof(struct rte_mbuf) (128 bytes) */
+#define ETH_AF_XDP_MBUF_OVERHEAD 192
+/* data start from offset 320 (192 + 128) bytes */
+#define ETH_AF_XDP_DATA_HEADROOM \
+ (ETH_AF_XDP_MBUF_OVERHEAD + RTE_PKTMBUF_HEADROOM)
#define ETH_AF_XDP_DFLT_RING_SIZE 1024
#define ETH_AF_XDP_DFLT_QUEUE_IDX 0
@@ -57,6 +61,7 @@ struct xdp_umem {
unsigned int frame_size_log2;
unsigned int nframes;
int mr_fd;
+ struct rte_mempool *mb_pool;
};
struct pmd_internals {
@@ -67,7 +72,7 @@ struct pmd_internals {
struct xdp_queue rx;
struct xdp_queue tx;
struct xdp_umem *umem;
- struct rte_mempool *mb_pool;
+ struct rte_mempool *ext_mb_pool;
unsigned long rx_pkts;
unsigned long rx_bytes;
@@ -80,7 +85,6 @@ struct pmd_internals {
uint16_t port_id;
uint16_t queue_idx;
int ring_size;
- struct rte_ring *buf_ring;
};
static const char * const valid_arguments[] = {
@@ -106,6 +110,21 @@ static void *get_pkt_data(struct pmd_internals *internals,
offset);
}
+static uint32_t
+mbuf_to_idx(struct pmd_internals *internals, struct rte_mbuf *mbuf)
+{
+ return (uint32_t)(((uint64_t)mbuf->buf_addr -
+ (uint64_t)internals->umem->buffer) >>
+ internals->umem->frame_size_log2);
+}
+
+static struct rte_mbuf *
+idx_to_mbuf(struct pmd_internals *internals, uint32_t idx)
+{
+ return (struct rte_mbuf *)(void *)(internals->umem->buffer + (idx
+ << internals->umem->frame_size_log2) + 0x40);
+}
+
static uint16_t
eth_af_xdp_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
{
@@ -120,17 +139,18 @@ eth_af_xdp_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
nb_pkts : ETH_AF_XDP_RX_BATCH_SIZE;
struct xdp_desc descs[ETH_AF_XDP_RX_BATCH_SIZE];
- void *indexes[ETH_AF_XDP_RX_BATCH_SIZE];
+ struct rte_mbuf *mbufs[ETH_AF_XDP_RX_BATCH_SIZE];
int rcvd, i;
/* fill rx ring */
if (rxq->num_free >= ETH_AF_XDP_RX_BATCH_SIZE) {
- int n = rte_ring_dequeue_bulk(internals->buf_ring,
- indexes,
- ETH_AF_XDP_RX_BATCH_SIZE,
- NULL);
- for (i = 0; i < n; i++)
- descs[i].idx = (uint32_t)((long int)indexes[i]);
- xq_enq(rxq, descs, n);
+ int ret = rte_mempool_get_bulk(internals->umem->mb_pool,
+ (void *)mbufs,
+ ETH_AF_XDP_RX_BATCH_SIZE);
+ if (!ret) {
+ for (i = 0; i < ETH_AF_XDP_RX_BATCH_SIZE; i++)
+ descs[i].idx = mbuf_to_idx(internals, mbufs[i]);
+ xq_enq(rxq, descs, ETH_AF_XDP_RX_BATCH_SIZE);
+ }
}
/* read data */
@@ -142,7 +162,7 @@ eth_af_xdp_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
char *pkt;
uint32_t idx = descs[i].idx;
- mbuf = rte_pktmbuf_alloc(internals->mb_pool);
+ mbuf = rte_pktmbuf_alloc(internals->ext_mb_pool);
rte_pktmbuf_pkt_len(mbuf) =
rte_pktmbuf_data_len(mbuf) =
descs[i].len;
@@ -155,11 +175,9 @@ eth_af_xdp_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
} else {
dropped++;
}
- indexes[i] = (void *)((long int)idx);
+ rte_pktmbuf_free(idx_to_mbuf(internals, idx));
}
- rte_ring_enqueue_bulk(internals->buf_ring, indexes, rcvd, NULL);
-
internals->rx_pkts += (rcvd - dropped);
internals->rx_bytes += rx_bytes;
internals->rx_dropped += dropped;
@@ -187,9 +205,10 @@ eth_af_xdp_tx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
struct xdp_queue *txq = &internals->tx;
struct rte_mbuf *mbuf;
struct xdp_desc descs[ETH_AF_XDP_TX_BATCH_SIZE];
- void *indexes[ETH_AF_XDP_TX_BATCH_SIZE];
+ struct rte_mbuf *mbufs[ETH_AF_XDP_TX_BATCH_SIZE];
uint16_t i, valid;
unsigned long tx_bytes = 0;
+ int ret;
nb_pkts = nb_pkts < ETH_AF_XDP_TX_BATCH_SIZE ?
nb_pkts : ETH_AF_XDP_TX_BATCH_SIZE;
@@ -198,13 +217,15 @@ eth_af_xdp_tx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
int n = xq_deq(txq, descs, ETH_AF_XDP_TX_BATCH_SIZE);
for (i = 0; i < n; i++)
- indexes[i] = (void *)((long int)descs[i].idx);
- rte_ring_enqueue_bulk(internals->buf_ring, indexes, n, NULL);
+ rte_pktmbuf_free(idx_to_mbuf(internals, descs[i].idx));
}
nb_pkts = nb_pkts > txq->num_free ? txq->num_free : nb_pkts;
- nb_pkts = rte_ring_dequeue_bulk(internals->buf_ring, indexes,
- nb_pkts, NULL);
+ ret = rte_mempool_get_bulk(internals->umem->mb_pool,
+ (void *)mbufs,
+ nb_pkts);
+ if (ret)
+ return 0;
valid = 0;
for (i = 0; i < nb_pkts; i++) {
@@ -213,14 +234,14 @@ eth_af_xdp_tx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
internals->umem->frame_size - ETH_AF_XDP_DATA_HEADROOM;
mbuf = bufs[i];
if (mbuf->pkt_len <= buf_len) {
- descs[valid].idx = (uint32_t)((long int)indexes[valid]);
+ descs[valid].idx = mbuf_to_idx(internals, mbufs[i]);
descs[valid].offset = ETH_AF_XDP_DATA_HEADROOM;
descs[valid].flags = 0;
descs[valid].len = mbuf->pkt_len;
pkt = get_pkt_data(internals, descs[i].idx,
descs[i].offset);
memcpy(pkt, rte_pktmbuf_mtod(mbuf, void *),
- descs[i].len);
+ descs[i].len);
valid++;
tx_bytes += mbuf->pkt_len;
}
@@ -230,9 +251,10 @@ eth_af_xdp_tx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
xq_enq(txq, descs, valid);
kick_tx(internals->sfd);
- if (valid < nb_pkts)
- rte_ring_enqueue_bulk(internals->buf_ring, &indexes[valid],
- nb_pkts - valid, NULL);
+ if (valid < nb_pkts) {
+ for (i = valid; i < nb_pkts; i++)
+ rte_pktmbuf_free(mbufs[i]);
+ }
internals->err_pkts += (nb_pkts - valid);
internals->tx_pkts += valid;
@@ -245,14 +267,13 @@ static void
fill_rx_desc(struct pmd_internals *internals)
{
int num_free = internals->rx.num_free;
- void *p = NULL;
int i;
-
for (i = 0; i < num_free; i++) {
struct xdp_desc desc = {};
+ struct rte_mbuf *mbuf =
+ rte_pktmbuf_alloc(internals->umem->mb_pool);
- rte_ring_dequeue(internals->buf_ring, &p);
- desc.idx = (uint32_t)((long int)p);
+ desc.idx = mbuf_to_idx(internals, mbuf);
xq_enq(&internals->rx, &desc, 1);
}
}
@@ -347,33 +368,53 @@ eth_link_update(struct rte_eth_dev *dev __rte_unused,
return 0;
}
-static struct xdp_umem *xsk_alloc_and_mem_reg_buffers(int sfd, size_t nbuffers)
+static void *get_base_addr(struct rte_mempool *mb_pool)
+{
+ struct rte_mempool_memhdr *memhdr;
+
+ STAILQ_FOREACH(memhdr, &mb_pool->mem_list, next) {
+ return memhdr->addr;
+ }
+ return NULL;
+}
+
+static struct xdp_umem *xsk_alloc_and_mem_reg_buffers(int sfd,
+ size_t nbuffers,
+ const char *pool_name)
{
struct xdp_mr_req req = { .frame_size = ETH_AF_XDP_FRAME_SIZE,
.data_headroom = ETH_AF_XDP_DATA_HEADROOM };
- struct xdp_umem *umem;
- void *bufs;
- int ret;
+ struct xdp_umem *umem = calloc(1, sizeof(*umem));
- ret = posix_memalign((void **)&bufs, getpagesize(),
- nbuffers * req.frame_size);
- if (ret)
+ if (!umem)
+ return NULL;
+
+ umem->mb_pool =
+ rte_pktmbuf_pool_create_with_flags(
+ pool_name, nbuffers,
+ 250, 0,
+ (ETH_AF_XDP_FRAME_SIZE - ETH_AF_XDP_MBUF_OVERHEAD),
+ MEMPOOL_F_NO_SPREAD | MEMPOOL_F_PAGE_ALIGN,
+ SOCKET_ID_ANY);
+
+ if (!umem->mb_pool) {
+ free(umem);
return NULL;
+ }
- umem = calloc(1, sizeof(*umem));
- if (!umem) {
- free(bufs);
+ if (umem->mb_pool->nb_mem_chunks > 1) {
+ rte_mempool_free(umem->mb_pool);
+ free(umem);
return NULL;
}
- req.addr = (unsigned long)bufs;
+ req.addr = (uint64_t)get_base_addr(umem->mb_pool);
req.len = nbuffers * req.frame_size;
- ret = setsockopt(sfd, SOL_XDP, XDP_MEM_REG, &req, sizeof(req));
- RTE_ASSERT(ret == 0);
+ setsockopt(sfd, SOL_XDP, XDP_MEM_REG, &req, sizeof(req));
umem->frame_size = ETH_AF_XDP_FRAME_SIZE;
umem->frame_size_log2 = 11;
- umem->buffer = bufs;
+ umem->buffer = (char *)req.addr;
umem->size = nbuffers * req.frame_size;
umem->nframes = nbuffers;
umem->mr_fd = sfd;
@@ -386,38 +427,27 @@ xdp_configure(struct pmd_internals *internals)
{
struct sockaddr_xdp sxdp;
struct xdp_ring_req req;
- char ring_name[0x100];
+ char pool_name[0x100];
+
int ret = 0;
- long int i;
- snprintf(ring_name, 0x100, "%s_%s_%d", "af_xdp_ring",
+ snprintf(pool_name, 0x100, "%s_%s_%d", "af_xdp_pool",
internals->if_name, internals->queue_idx);
- internals->buf_ring = rte_ring_create(ring_name,
- ETH_AF_XDP_NUM_BUFFERS,
- SOCKET_ID_ANY,
- 0x0);
- if (!internals->buf_ring)
- return -1;
-
- for (i = 0; i < ETH_AF_XDP_NUM_BUFFERS; i++)
- rte_ring_enqueue(internals->buf_ring, (void *)i);
-
internals->umem = xsk_alloc_and_mem_reg_buffers(internals->sfd,
- ETH_AF_XDP_NUM_BUFFERS);
+ ETH_AF_XDP_NUM_BUFFERS,
+ pool_name);
if (!internals->umem)
- goto error;
+ return -1;
req.mr_fd = internals->umem->mr_fd;
req.desc_nr = internals->ring_size;
ret = setsockopt(internals->sfd, SOL_XDP, XDP_RX_RING,
&req, sizeof(req));
-
RTE_ASSERT(ret == 0);
ret = setsockopt(internals->sfd, SOL_XDP, XDP_TX_RING,
&req, sizeof(req));
-
RTE_ASSERT(ret == 0);
internals->rx.ring = mmap(0, req.desc_nr * sizeof(struct xdp_desc),
@@ -448,10 +478,6 @@ xdp_configure(struct pmd_internals *internals)
RTE_ASSERT(ret == 0);
return ret;
-error:
- rte_ring_free(internals->buf_ring);
- internals->buf_ring = NULL;
- return -1;
}
static int
@@ -466,11 +492,11 @@ eth_rx_queue_setup(struct rte_eth_dev *dev,
unsigned int buf_size, data_size;
RTE_ASSERT(rx_queue_id == 0);
- internals->mb_pool = mb_pool;
+ internals->ext_mb_pool = mb_pool;
xdp_configure(internals);
/* Now get the space available for data in the mbuf */
- buf_size = rte_pktmbuf_data_room_size(internals->mb_pool) -
+ buf_size = rte_pktmbuf_data_room_size(internals->ext_mb_pool) -
RTE_PKTMBUF_HEADROOM;
data_size = internals->umem->frame_size;
@@ -739,8 +765,11 @@ rte_pmd_af_xdp_remove(struct rte_vdev_device *dev)
return -1;
internals = eth_dev->data->dev_private;
- rte_ring_free(internals->buf_ring);
- rte_free(internals->umem);
+ if (internals->umem) {
+ if (internals->umem->mb_pool)
+ rte_mempool_free(internals->umem->mb_pool);
+ rte_free(internals->umem);
+ }
rte_free(eth_dev->data->dev_private);
rte_free(eth_dev->data);
close(internals->sfd);
--
2.13.6
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [RFC 4/7] net/af_xdp: use mbuf mempool for buffer management
2018-02-27 9:33 ` [RFC 4/7] net/af_xdp: use mbuf mempool for buffer management Qi Zhang
@ 2018-03-01 2:08 ` Stephen Hemminger
0 siblings, 0 replies; 24+ messages in thread
From: Stephen Hemminger @ 2018-03-01 2:08 UTC (permalink / raw)
To: Qi Zhang; +Cc: dev, magnus.karlsson, bjorn.topel
On Tue, 27 Feb 2018 17:33:03 +0800
Qi Zhang <qi.z.zhang@intel.com> wrote:
> +static uint32_t
> +mbuf_to_idx(struct pmd_internals *internals, struct rte_mbuf *mbuf)
> +{
> + return (uint32_t)(((uint64_t)mbuf->buf_addr -
> + (uint64_t)internals->umem->buffer) >>
> + internals->umem->frame_size_log2);
> +}
> +
> +static struct rte_mbuf *
> +idx_to_mbuf(struct pmd_internals *internals, uint32_t idx)
> +{
> + return (struct rte_mbuf *)(void *)(internals->umem->buffer + (idx
> + << internals->umem->frame_size_log2) + 0x40);
> +}
More unnecessary casts's here.
^ permalink raw reply [flat|nested] 24+ messages in thread
* [RFC 5/7] net/af_xdp: enable share mempool
2018-02-27 9:32 [RFC 0/7] PMD driver for AF_XDP Qi Zhang
` (3 preceding siblings ...)
2018-02-27 9:33 ` [RFC 4/7] net/af_xdp: use mbuf mempool for buffer management Qi Zhang
@ 2018-02-27 9:33 ` Qi Zhang
2018-02-27 9:33 ` [RFC 6/7] net/af_xdp: load BPF file Qi Zhang
` (2 subsequent siblings)
7 siblings, 0 replies; 24+ messages in thread
From: Qi Zhang @ 2018-02-27 9:33 UTC (permalink / raw)
To: dev; +Cc: magnus.karlsson, bjorn.topel, Qi Zhang
Try to check if external mempool (from rx_queue_setup) is fit for
af_xdp, if it is, it will be registered to af_xdp socket directly and
there will be no packet data copy on Rx and Tx.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
---
drivers/net/af_xdp/rte_eth_af_xdp.c | 191 +++++++++++++++++++++++-------------
1 file changed, 125 insertions(+), 66 deletions(-)
diff --git a/drivers/net/af_xdp/rte_eth_af_xdp.c b/drivers/net/af_xdp/rte_eth_af_xdp.c
index 3c534c77c..d0939022b 100644
--- a/drivers/net/af_xdp/rte_eth_af_xdp.c
+++ b/drivers/net/af_xdp/rte_eth_af_xdp.c
@@ -60,7 +60,6 @@ struct xdp_umem {
unsigned int frame_size;
unsigned int frame_size_log2;
unsigned int nframes;
- int mr_fd;
struct rte_mempool *mb_pool;
};
@@ -73,6 +72,7 @@ struct pmd_internals {
struct xdp_queue tx;
struct xdp_umem *umem;
struct rte_mempool *ext_mb_pool;
+ uint8_t share_mb_pool;
unsigned long rx_pkts;
unsigned long rx_bytes;
@@ -162,20 +162,30 @@ eth_af_xdp_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
char *pkt;
uint32_t idx = descs[i].idx;
- mbuf = rte_pktmbuf_alloc(internals->ext_mb_pool);
- rte_pktmbuf_pkt_len(mbuf) =
- rte_pktmbuf_data_len(mbuf) =
- descs[i].len;
- if (mbuf) {
- pkt = get_pkt_data(internals, idx, descs[i].offset);
- memcpy(rte_pktmbuf_mtod(mbuf, void *),
- pkt, descs[i].len);
- rx_bytes += descs[i].len;
- bufs[count++] = mbuf;
+ if (!internals->share_mb_pool) {
+ mbuf = rte_pktmbuf_alloc(internals->ext_mb_pool);
+ rte_pktmbuf_pkt_len(mbuf) =
+ rte_pktmbuf_data_len(mbuf) =
+ descs[i].len;
+ if (mbuf) {
+ pkt = get_pkt_data(internals, idx,
+ descs[i].offset);
+ memcpy(rte_pktmbuf_mtod(mbuf, void *), pkt,
+ descs[i].len);
+ rx_bytes += descs[i].len;
+ bufs[count++] = mbuf;
+ } else {
+ dropped++;
+ }
+ rte_pktmbuf_free(idx_to_mbuf(internals, idx));
} else {
- dropped++;
+ mbuf = idx_to_mbuf(internals, idx);
+ rte_pktmbuf_pkt_len(mbuf) =
+ rte_pktmbuf_data_len(mbuf) =
+ descs[i].len;
+ bufs[count++] = mbuf;
+ rx_bytes += descs[i].len;
}
- rte_pktmbuf_free(idx_to_mbuf(internals, idx));
}
internals->rx_pkts += (rcvd - dropped);
@@ -209,51 +219,71 @@ eth_af_xdp_tx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
uint16_t i, valid;
unsigned long tx_bytes = 0;
int ret;
+ uint8_t share_mempool = 0;
nb_pkts = nb_pkts < ETH_AF_XDP_TX_BATCH_SIZE ?
nb_pkts : ETH_AF_XDP_TX_BATCH_SIZE;
if (txq->num_free < ETH_AF_XDP_TX_BATCH_SIZE * 2) {
int n = xq_deq(txq, descs, ETH_AF_XDP_TX_BATCH_SIZE);
-
for (i = 0; i < n; i++)
rte_pktmbuf_free(idx_to_mbuf(internals, descs[i].idx));
}
nb_pkts = nb_pkts > txq->num_free ? txq->num_free : nb_pkts;
- ret = rte_mempool_get_bulk(internals->umem->mb_pool,
- (void *)mbufs,
- nb_pkts);
- if (ret)
+ if (nb_pkts == 0)
return 0;
+ if (bufs[0]->pool == internals->ext_mb_pool && internals->share_mb_pool)
+ share_mempool = 1;
+
+ if (!share_mempool) {
+ ret = rte_mempool_get_bulk(internals->umem->mb_pool,
+ (void *)mbufs,
+ nb_pkts);
+ if (ret)
+ return 0;
+ }
+
valid = 0;
for (i = 0; i < nb_pkts; i++) {
char *pkt;
- unsigned int buf_len =
- internals->umem->frame_size - ETH_AF_XDP_DATA_HEADROOM;
mbuf = bufs[i];
- if (mbuf->pkt_len <= buf_len) {
- descs[valid].idx = mbuf_to_idx(internals, mbufs[i]);
- descs[valid].offset = ETH_AF_XDP_DATA_HEADROOM;
- descs[valid].flags = 0;
- descs[valid].len = mbuf->pkt_len;
- pkt = get_pkt_data(internals, descs[i].idx,
- descs[i].offset);
- memcpy(pkt, rte_pktmbuf_mtod(mbuf, void *),
- descs[i].len);
- valid++;
+ if (!share_mempool) {
+ if (mbuf->pkt_len <=
+ (internals->umem->frame_size -
+ ETH_AF_XDP_DATA_HEADROOM)) {
+ descs[valid].idx =
+ mbuf_to_idx(internals, mbufs[i]);
+ descs[valid].offset = ETH_AF_XDP_DATA_HEADROOM;
+ descs[valid].flags = 0;
+ descs[valid].len = mbuf->pkt_len;
+ pkt = get_pkt_data(internals, descs[i].idx,
+ descs[i].offset);
+ memcpy(pkt, rte_pktmbuf_mtod(mbuf, void *),
+ descs[i].len);
+ valid++;
+ tx_bytes += mbuf->pkt_len;
+ }
+ rte_pktmbuf_free(mbuf);
+ } else {
+ descs[i].idx = mbuf_to_idx(internals, mbuf);
+ descs[i].offset = ETH_AF_XDP_DATA_HEADROOM;
+ descs[i].flags = 0;
+ descs[i].len = mbuf->pkt_len;
tx_bytes += mbuf->pkt_len;
+ valid++;
}
- rte_pktmbuf_free(mbuf);
}
xq_enq(txq, descs, valid);
kick_tx(internals->sfd);
- if (valid < nb_pkts) {
- for (i = valid; i < nb_pkts; i++)
- rte_pktmbuf_free(mbufs[i]);
+ if (!share_mempool) {
+ if (valid < nb_pkts) {
+ for (i = valid; i < nb_pkts; i++)
+ rte_pktmbuf_free(mbufs[i]);
+ }
}
internals->err_pkts += (nb_pkts - valid);
@@ -378,46 +408,81 @@ static void *get_base_addr(struct rte_mempool *mb_pool)
return NULL;
}
-static struct xdp_umem *xsk_alloc_and_mem_reg_buffers(int sfd,
- size_t nbuffers,
- const char *pool_name)
+static uint8_t
+check_mempool(struct rte_mempool *mp)
+{
+ RTE_ASSERT(mp);
+
+ /* must continues */
+ if (mp->nb_mem_chunks > 1)
+ return 0;
+
+ /* check header size */
+ if (mp->header_size != RTE_CACHE_LINE_SIZE)
+ return 0;
+
+ /* check base address */
+ if ((uint64_t)get_base_addr(mp) % getpagesize() != 0)
+ return 0;
+
+ /* check chunk size */
+ if ((mp->elt_size + mp->header_size + mp->trailer_size) %
+ ETH_AF_XDP_FRAME_SIZE != 0)
+ return 0;
+
+ return 1;
+}
+
+static struct xdp_umem *
+xsk_alloc_and_mem_reg_buffers(struct pmd_internals *internals)
{
struct xdp_mr_req req = { .frame_size = ETH_AF_XDP_FRAME_SIZE,
.data_headroom = ETH_AF_XDP_DATA_HEADROOM };
+ char pool_name[0x100];
+ int nbuffers;
struct xdp_umem *umem = calloc(1, sizeof(*umem));
if (!umem)
return NULL;
- umem->mb_pool =
- rte_pktmbuf_pool_create_with_flags(
- pool_name, nbuffers,
- 250, 0,
- (ETH_AF_XDP_FRAME_SIZE - ETH_AF_XDP_MBUF_OVERHEAD),
- MEMPOOL_F_NO_SPREAD | MEMPOOL_F_PAGE_ALIGN,
- SOCKET_ID_ANY);
-
- if (!umem->mb_pool) {
- free(umem);
- return NULL;
- }
+ internals->share_mb_pool = check_mempool(internals->ext_mb_pool);
+ if (!internals->share_mb_pool) {
+ snprintf(pool_name, 0x100, "%s_%s_%d", "af_xdp_pool",
+ internals->if_name, internals->queue_idx);
+ umem->mb_pool =
+ rte_pktmbuf_pool_create_with_flags(
+ pool_name,
+ ETH_AF_XDP_NUM_BUFFERS,
+ 250, 0,
+ (ETH_AF_XDP_FRAME_SIZE -
+ ETH_AF_XDP_MBUF_OVERHEAD),
+ MEMPOOL_F_NO_SPREAD | MEMPOOL_F_PAGE_ALIGN,
+ SOCKET_ID_ANY);
+ if (!umem->mb_pool) {
+ free(umem);
+ return NULL;
+ }
- if (umem->mb_pool->nb_mem_chunks > 1) {
- rte_mempool_free(umem->mb_pool);
- free(umem);
- return NULL;
+ if (umem->mb_pool->nb_mem_chunks > 1) {
+ rte_mempool_free(umem->mb_pool);
+ free(umem);
+ return NULL;
+ }
+ nbuffers = ETH_AF_XDP_NUM_BUFFERS;
+ } else {
+ umem->mb_pool = internals->ext_mb_pool;
+ nbuffers = umem->mb_pool->populated_size;
}
req.addr = (uint64_t)get_base_addr(umem->mb_pool);
- req.len = nbuffers * req.frame_size;
- setsockopt(sfd, SOL_XDP, XDP_MEM_REG, &req, sizeof(req));
+ req.len = ETH_AF_XDP_NUM_BUFFERS * req.frame_size;
+ setsockopt(internals->sfd, SOL_XDP, XDP_MEM_REG, &req, sizeof(req));
umem->frame_size = ETH_AF_XDP_FRAME_SIZE;
umem->frame_size_log2 = 11;
umem->buffer = (char *)req.addr;
umem->size = nbuffers * req.frame_size;
umem->nframes = nbuffers;
- umem->mr_fd = sfd;
return umem;
}
@@ -427,19 +492,13 @@ xdp_configure(struct pmd_internals *internals)
{
struct sockaddr_xdp sxdp;
struct xdp_ring_req req;
- char pool_name[0x100];
-
int ret = 0;
- snprintf(pool_name, 0x100, "%s_%s_%d", "af_xdp_pool",
- internals->if_name, internals->queue_idx);
- internals->umem = xsk_alloc_and_mem_reg_buffers(internals->sfd,
- ETH_AF_XDP_NUM_BUFFERS,
- pool_name);
+ internals->umem = xsk_alloc_and_mem_reg_buffers(internals);
if (!internals->umem)
return -1;
- req.mr_fd = internals->umem->mr_fd;
+ req.mr_fd = internals->sfd;
req.desc_nr = internals->ring_size;
ret = setsockopt(internals->sfd, SOL_XDP, XDP_RX_RING,
@@ -500,7 +559,7 @@ eth_rx_queue_setup(struct rte_eth_dev *dev,
RTE_PKTMBUF_HEADROOM;
data_size = internals->umem->frame_size;
- if (data_size > buf_size) {
+ if (data_size - ETH_AF_XDP_DATA_HEADROOM > buf_size) {
RTE_LOG(ERR, PMD,
"%s: %d bytes will not fit in mbuf (%d bytes)\n",
dev->device->name, data_size, buf_size);
@@ -766,7 +825,7 @@ rte_pmd_af_xdp_remove(struct rte_vdev_device *dev)
internals = eth_dev->data->dev_private;
if (internals->umem) {
- if (internals->umem->mb_pool)
+ if (internals->umem->mb_pool && !internals->share_mb_pool)
rte_mempool_free(internals->umem->mb_pool);
rte_free(internals->umem);
}
--
2.13.6
^ permalink raw reply related [flat|nested] 24+ messages in thread* [RFC 6/7] net/af_xdp: load BPF file
2018-02-27 9:32 [RFC 0/7] PMD driver for AF_XDP Qi Zhang
` (4 preceding siblings ...)
2018-02-27 9:33 ` [RFC 5/7] net/af_xdp: enable share mempool Qi Zhang
@ 2018-02-27 9:33 ` Qi Zhang
2018-03-01 2:10 ` Stephen Hemminger
2018-02-27 9:33 ` [RFC 7/7] app/testpmd: enable parameter for mempool flags Qi Zhang
2018-03-01 2:52 ` [RFC 0/7] PMD driver for AF_XDP Jason Wang
7 siblings, 1 reply; 24+ messages in thread
From: Qi Zhang @ 2018-02-27 9:33 UTC (permalink / raw)
To: dev; +Cc: magnus.karlsson, bjorn.topel, Qi Zhang
Add libbpf and libelf dependency in Makefile.
Durring initialization, bpf file "xdpsock_kern.o" will be loaded.
Then the driver will always try to link XDP fd with DRV mode first,
then SKB mode if failed in previoius.
Link will be released during dev_close.
Note: this is workaround solution, af_xdp may remove BPF dependency
in future.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
---
drivers/net/af_xdp/Makefile | 6 +-
drivers/net/af_xdp/bpf_load.c | 798 ++++++++++++++++++++++++++++++++++++
drivers/net/af_xdp/bpf_load.h | 65 +++
drivers/net/af_xdp/libbpf.h | 199 +++++++++
drivers/net/af_xdp/rte_eth_af_xdp.c | 31 +-
mk/rte.app.mk | 2 +-
6 files changed, 1097 insertions(+), 4 deletions(-)
create mode 100644 drivers/net/af_xdp/bpf_load.c
create mode 100644 drivers/net/af_xdp/bpf_load.h
create mode 100644 drivers/net/af_xdp/libbpf.h
diff --git a/drivers/net/af_xdp/Makefile b/drivers/net/af_xdp/Makefile
index ac38e20bf..a642786de 100644
--- a/drivers/net/af_xdp/Makefile
+++ b/drivers/net/af_xdp/Makefile
@@ -42,7 +42,10 @@ EXPORT_MAP := rte_pmd_af_xdp_version.map
LIBABIVER := 1
-CFLAGS += -O3 -I/opt/af_xdp/linux_headers/include
+LINUX_HEADER_DIR := /opt/af_xdp/linux_headers/include
+TOOLS_DIR := /root/af_xdp/npg_dna-dna-linux/tools
+
+CFLAGS += -O3 -I$(LINUX_HEADER_DIR) -I$(TOOLS_DIR)/perf -I$(TOOLS_DIR)/include -Wno-error=sign-compare -Wno-error=cast-qual
CFLAGS += $(WERROR_FLAGS)
LDLIBS += -lrte_eal -lrte_mbuf -lrte_mempool -lrte_ring
LDLIBS += -lrte_ethdev -lrte_net -lrte_kvargs
@@ -52,5 +55,6 @@ LDLIBS += -lrte_bus_vdev
# all source are stored in SRCS-y
#
SRCS-$(CONFIG_RTE_LIBRTE_PMD_AF_XDP) += rte_eth_af_xdp.c
+SRCS-$(CONFIG_RTE_LIBRTE_PMD_AF_XDP) += bpf_load.c
include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/net/af_xdp/bpf_load.c b/drivers/net/af_xdp/bpf_load.c
new file mode 100644
index 000000000..aa632207f
--- /dev/null
+++ b/drivers/net/af_xdp/bpf_load.c
@@ -0,0 +1,798 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <stdio.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <fcntl.h>
+#include <libelf.h>
+#include <gelf.h>
+#include <errno.h>
+#include <unistd.h>
+#include <string.h>
+#include <stdbool.h>
+#include <stdlib.h>
+#include <linux/bpf.h>
+#include <linux/filter.h>
+#include <linux/perf_event.h>
+#include <linux/netlink.h>
+#include <linux/rtnetlink.h>
+#include <linux/types.h>
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <sys/syscall.h>
+#include <sys/ioctl.h>
+#include <sys/mman.h>
+#include <poll.h>
+#include <ctype.h>
+#include <assert.h>
+#include "libbpf.h"
+#include "bpf_load.h"
+#include "perf-sys.h"
+
+#define DEBUGFS "/sys/kernel/debug/tracing/"
+
+static char license[128];
+static int kern_version;
+static bool processed_sec[128];
+char bpf_log_buf[BPF_LOG_BUF_SIZE];
+int map_fd[MAX_MAPS];
+int prog_fd[MAX_PROGS];
+int event_fd[MAX_PROGS];
+int prog_cnt;
+int prog_array_fd = -1;
+
+struct bpf_map_data map_data[MAX_MAPS];
+int map_data_count = 0;
+
+static int populate_prog_array(const char *event, int prog_fd)
+{
+ int ind = atoi(event), err;
+
+ err = bpf_map_update_elem(prog_array_fd, &ind, &prog_fd, BPF_ANY);
+ if (err < 0) {
+ printf("failed to store prog_fd in prog_array\n");
+ return -1;
+ }
+ return 0;
+}
+
+static int load_and_attach(const char *event, struct bpf_insn *prog, int size)
+{
+ bool is_socket = strncmp(event, "socket", 6) == 0;
+ bool is_kprobe = strncmp(event, "kprobe/", 7) == 0;
+ bool is_kretprobe = strncmp(event, "kretprobe/", 10) == 0;
+ bool is_tracepoint = strncmp(event, "tracepoint/", 11) == 0;
+ bool is_xdp = strncmp(event, "xdp", 3) == 0;
+ bool is_perf_event = strncmp(event, "perf_event", 10) == 0;
+ bool is_cgroup_skb = strncmp(event, "cgroup/skb", 10) == 0;
+ bool is_cgroup_sk = strncmp(event, "cgroup/sock", 11) == 0;
+ bool is_sockops = strncmp(event, "sockops", 7) == 0;
+ bool is_sk_skb = strncmp(event, "sk_skb", 6) == 0;
+ size_t insns_cnt = size / sizeof(struct bpf_insn);
+ enum bpf_prog_type prog_type;
+ char buf[256];
+ int fd, efd, err, id;
+ struct perf_event_attr attr = {};
+
+ attr.type = PERF_TYPE_TRACEPOINT;
+ attr.sample_type = PERF_SAMPLE_RAW;
+ attr.sample_period = 1;
+ attr.wakeup_events = 1;
+
+ if (is_socket) {
+ prog_type = BPF_PROG_TYPE_SOCKET_FILTER;
+ } else if (is_kprobe || is_kretprobe) {
+ prog_type = BPF_PROG_TYPE_KPROBE;
+ } else if (is_tracepoint) {
+ prog_type = BPF_PROG_TYPE_TRACEPOINT;
+ } else if (is_xdp) {
+ prog_type = BPF_PROG_TYPE_XDP;
+ } else if (is_perf_event) {
+ prog_type = BPF_PROG_TYPE_PERF_EVENT;
+ } else if (is_cgroup_skb) {
+ prog_type = BPF_PROG_TYPE_CGROUP_SKB;
+ } else if (is_cgroup_sk) {
+ prog_type = BPF_PROG_TYPE_CGROUP_SOCK;
+ } else if (is_sockops) {
+ prog_type = BPF_PROG_TYPE_SOCK_OPS;
+ } else if (is_sk_skb) {
+ prog_type = BPF_PROG_TYPE_SK_SKB;
+ } else {
+ printf("Unknown event '%s'\n", event);
+ return -1;
+ }
+
+ fd = bpf_load_program(prog_type, prog, insns_cnt, license, kern_version,
+ bpf_log_buf, BPF_LOG_BUF_SIZE);
+ if (fd < 0) {
+ printf("bpf_load_program() err=%d\n%s", errno, bpf_log_buf);
+ return -1;
+ }
+
+ prog_fd[prog_cnt++] = fd;
+
+ if (is_xdp || is_perf_event || is_cgroup_skb || is_cgroup_sk)
+ return 0;
+
+ if (is_socket || is_sockops || is_sk_skb) {
+ if (is_socket)
+ event += 6;
+ else
+ event += 7;
+ if (*event != '/')
+ return 0;
+ event++;
+ if (!isdigit(*event)) {
+ printf("invalid prog number\n");
+ return -1;
+ }
+ return populate_prog_array(event, fd);
+ }
+
+ if (is_kprobe || is_kretprobe) {
+ if (is_kprobe)
+ event += 7;
+ else
+ event += 10;
+
+ if (*event == 0) {
+ printf("event name cannot be empty\n");
+ return -1;
+ }
+
+ if (isdigit(*event))
+ return populate_prog_array(event, fd);
+
+ snprintf(buf, sizeof(buf),
+ "echo '%c:%s %s' >> /sys/kernel/debug/tracing/kprobe_events",
+ is_kprobe ? 'p' : 'r', event, event);
+ err = system(buf);
+ if (err < 0) {
+ printf("failed to create kprobe '%s' error '%s'\n",
+ event, strerror(errno));
+ return -1;
+ }
+
+ strcpy(buf, DEBUGFS);
+ strcat(buf, "events/kprobes/");
+ strcat(buf, event);
+ strcat(buf, "/id");
+ } else if (is_tracepoint) {
+ event += 11;
+
+ if (*event == 0) {
+ printf("event name cannot be empty\n");
+ return -1;
+ }
+ strcpy(buf, DEBUGFS);
+ strcat(buf, "events/");
+ strcat(buf, event);
+ strcat(buf, "/id");
+ }
+
+ efd = open(buf, O_RDONLY, 0);
+ if (efd < 0) {
+ printf("failed to open event %s\n", event);
+ return -1;
+ }
+
+ err = read(efd, buf, sizeof(buf));
+ if (err < 0 || err >= sizeof(buf)) {
+ printf("read from '%s' failed '%s'\n", event, strerror(errno));
+ return -1;
+ }
+
+ close(efd);
+
+ buf[err] = 0;
+ id = atoi(buf);
+ attr.config = id;
+
+ efd = sys_perf_event_open(&attr, -1/*pid*/, 0/*cpu*/, -1/*group_fd*/, 0);
+ if (efd < 0) {
+ printf("event %d fd %d err %s\n", id, efd, strerror(errno));
+ return -1;
+ }
+ event_fd[prog_cnt - 1] = efd;
+ err = ioctl(efd, PERF_EVENT_IOC_ENABLE, 0);
+ if (err < 0) {
+ printf("ioctl PERF_EVENT_IOC_ENABLE failed err %s\n",
+ strerror(errno));
+ return -1;
+ }
+ err = ioctl(efd, PERF_EVENT_IOC_SET_BPF, fd);
+ if (err < 0) {
+ printf("ioctl PERF_EVENT_IOC_SET_BPF failed err %s\n",
+ strerror(errno));
+ return -1;
+ }
+
+ return 0;
+}
+
+static int load_maps(struct bpf_map_data *maps, int nr_maps,
+ fixup_map_cb fixup_map)
+{
+ int i, numa_node;
+
+ for (i = 0; i < nr_maps; i++) {
+ if (fixup_map) {
+ fixup_map(&maps[i], i);
+ /* Allow userspace to assign map FD prior to creation */
+ if (maps[i].fd != -1) {
+ map_fd[i] = maps[i].fd;
+ continue;
+ }
+ }
+
+ numa_node = maps[i].def.map_flags & BPF_F_NUMA_NODE ?
+ maps[i].def.numa_node : -1;
+
+ if (maps[i].def.type == BPF_MAP_TYPE_ARRAY_OF_MAPS ||
+ maps[i].def.type == BPF_MAP_TYPE_HASH_OF_MAPS) {
+ int inner_map_fd = map_fd[maps[i].def.inner_map_idx];
+
+ map_fd[i] = bpf_create_map_in_map_node(maps[i].def.type,
+ maps[i].name,
+ maps[i].def.key_size,
+ inner_map_fd,
+ maps[i].def.max_entries,
+ maps[i].def.map_flags,
+ numa_node);
+ } else {
+ map_fd[i] = bpf_create_map_node(maps[i].def.type,
+ maps[i].name,
+ maps[i].def.key_size,
+ maps[i].def.value_size,
+ maps[i].def.max_entries,
+ maps[i].def.map_flags,
+ numa_node);
+ }
+ if (map_fd[i] < 0) {
+ printf("failed to create a map: %d %s\n",
+ errno, strerror(errno));
+ return 1;
+ }
+ maps[i].fd = map_fd[i];
+
+ if (maps[i].def.type == BPF_MAP_TYPE_PROG_ARRAY)
+ prog_array_fd = map_fd[i];
+ }
+ return 0;
+}
+
+static int get_sec(Elf *elf, int i, GElf_Ehdr *ehdr, char **shname,
+ GElf_Shdr *shdr, Elf_Data **data)
+{
+ Elf_Scn *scn;
+
+ scn = elf_getscn(elf, i);
+ if (!scn)
+ return 1;
+
+ if (gelf_getshdr(scn, shdr) != shdr)
+ return 2;
+
+ *shname = elf_strptr(elf, ehdr->e_shstrndx, shdr->sh_name);
+ if (!*shname || !shdr->sh_size)
+ return 3;
+
+ *data = elf_getdata(scn, 0);
+ if (!*data || elf_getdata(scn, *data) != NULL)
+ return 4;
+
+ return 0;
+}
+
+static int parse_relo_and_apply(Elf_Data *data, Elf_Data *symbols,
+ GElf_Shdr *shdr, struct bpf_insn *insn,
+ struct bpf_map_data *maps, int nr_maps)
+{
+ int i, nrels;
+
+ nrels = shdr->sh_size / shdr->sh_entsize;
+
+ for (i = 0; i < nrels; i++) {
+ GElf_Sym sym;
+ GElf_Rel rel;
+ unsigned int insn_idx;
+ bool match = false;
+ int map_idx;
+
+ gelf_getrel(data, i, &rel);
+
+ insn_idx = rel.r_offset / sizeof(struct bpf_insn);
+
+ gelf_getsym(symbols, GELF_R_SYM(rel.r_info), &sym);
+
+ if (insn[insn_idx].code != (BPF_LD | BPF_IMM | BPF_DW)) {
+ printf("invalid relo for insn[%d].code 0x%x\n",
+ insn_idx, insn[insn_idx].code);
+ return 1;
+ }
+ insn[insn_idx].src_reg = BPF_PSEUDO_MAP_FD;
+
+ /* Match FD relocation against recorded map_data[] offset */
+ for (map_idx = 0; map_idx < nr_maps; map_idx++) {
+ if (maps[map_idx].elf_offset == sym.st_value) {
+ match = true;
+ break;
+ }
+ }
+ if (match) {
+ insn[insn_idx].imm = maps[map_idx].fd;
+ } else {
+ printf("invalid relo for insn[%d] no map_data match\n",
+ insn_idx);
+ return 1;
+ }
+ }
+
+ return 0;
+}
+
+static int cmp_symbols(const void *l, const void *r)
+{
+ const GElf_Sym *lsym = (const GElf_Sym *)l;
+ const GElf_Sym *rsym = (const GElf_Sym *)r;
+
+ if (lsym->st_value < rsym->st_value)
+ return -1;
+ else if (lsym->st_value > rsym->st_value)
+ return 1;
+ else
+ return 0;
+}
+
+static int load_elf_maps_section(struct bpf_map_data *maps, int maps_shndx,
+ Elf *elf, Elf_Data *symbols, int strtabidx)
+{
+ int map_sz_elf, map_sz_copy;
+ bool validate_zero = false;
+ Elf_Data *data_maps;
+ int i, nr_maps;
+ GElf_Sym *sym;
+ Elf_Scn *scn;
+
+ if (maps_shndx < 0)
+ return -EINVAL;
+ if (!symbols)
+ return -EINVAL;
+
+ /* Get data for maps section via elf index */
+ scn = elf_getscn(elf, maps_shndx);
+ if (scn)
+ data_maps = elf_getdata(scn, NULL);
+ if (!scn || !data_maps) {
+ printf("Failed to get Elf_Data from maps section %d\n",
+ maps_shndx);
+ return -EINVAL;
+ }
+
+ /* For each map get corrosponding symbol table entry */
+ sym = calloc(MAX_MAPS+1, sizeof(GElf_Sym));
+ for (i = 0, nr_maps = 0; i < symbols->d_size / sizeof(GElf_Sym); i++) {
+ assert(nr_maps < MAX_MAPS+1);
+ if (!gelf_getsym(symbols, i, &sym[nr_maps]))
+ continue;
+ if (sym[nr_maps].st_shndx != maps_shndx)
+ continue;
+ /* Only increment iif maps section */
+ nr_maps++;
+ }
+
+ /* Align to map_fd[] order, via sort on offset in sym.st_value */
+ qsort(sym, nr_maps, sizeof(GElf_Sym), cmp_symbols);
+
+ /* Keeping compatible with ELF maps section changes
+ * ------------------------------------------------
+ * The program size of struct bpf_map_def is known by loader
+ * code, but struct stored in ELF file can be different.
+ *
+ * Unfortunately sym[i].st_size is zero. To calculate the
+ * struct size stored in the ELF file, assume all struct have
+ * the same size, and simply divide with number of map
+ * symbols.
+ */
+ map_sz_elf = data_maps->d_size / nr_maps;
+ map_sz_copy = sizeof(struct bpf_map_def);
+ if (map_sz_elf < map_sz_copy) {
+ /*
+ * Backward compat, loading older ELF file with
+ * smaller struct, keeping remaining bytes zero.
+ */
+ map_sz_copy = map_sz_elf;
+ } else if (map_sz_elf > map_sz_copy) {
+ /*
+ * Forward compat, loading newer ELF file with larger
+ * struct with unknown features. Assume zero means
+ * feature not used. Thus, validate rest of struct
+ * data is zero.
+ */
+ validate_zero = true;
+ }
+
+ /* Memcpy relevant part of ELF maps data to loader maps */
+ for (i = 0; i < nr_maps; i++) {
+ unsigned char *addr, *end;
+ struct bpf_map_def *def;
+ const char *map_name;
+ size_t offset;
+
+ map_name = elf_strptr(elf, strtabidx, sym[i].st_name);
+ maps[i].name = strdup(map_name);
+ if (!maps[i].name) {
+ printf("strdup(%s): %s(%d)\n", map_name,
+ strerror(errno), errno);
+ free(sym);
+ return -errno;
+ }
+
+ /* Symbol value is offset into ELF maps section data area */
+ offset = sym[i].st_value;
+ def = (struct bpf_map_def *)((uint8_t *)data_maps->d_buf + offset);
+ maps[i].elf_offset = offset;
+ memset(&maps[i].def, 0, sizeof(struct bpf_map_def));
+ memcpy(&maps[i].def, def, map_sz_copy);
+
+ /* Verify no newer features were requested */
+ if (validate_zero) {
+ addr = (unsigned char*) def + map_sz_copy;
+ end = (unsigned char*) def + map_sz_elf;
+ for (; addr < end; addr++) {
+ if (*addr != 0) {
+ free(sym);
+ return -EFBIG;
+ }
+ }
+ }
+ }
+
+ free(sym);
+ return nr_maps;
+}
+
+static int do_load_bpf_file(const char *path, fixup_map_cb fixup_map)
+{
+ int fd, i, ret, maps_shndx = -1, strtabidx = -1;
+ Elf *elf;
+ GElf_Ehdr ehdr;
+ GElf_Shdr shdr, shdr_prog;
+ Elf_Data *data, *data_prog, *data_maps = NULL, *symbols = NULL;
+ char *shname, *shname_prog;
+ int nr_maps = 0;
+
+ /* reset global variables */
+ kern_version = 0;
+ memset(license, 0, sizeof(license));
+ memset(processed_sec, 0, sizeof(processed_sec));
+
+ if (elf_version(EV_CURRENT) == EV_NONE)
+ return 1;
+
+ fd = open(path, O_RDONLY, 0);
+ if (fd < 0)
+ return 1;
+
+ elf = elf_begin(fd, ELF_C_READ, NULL);
+
+ if (!elf)
+ return 1;
+
+ if (gelf_getehdr(elf, &ehdr) != &ehdr)
+ return 1;
+
+ /* clear all kprobes */
+ i = system("echo \"\" > /sys/kernel/debug/tracing/kprobe_events");
+
+ /* scan over all elf sections to get license and map info */
+ for (i = 1; i < ehdr.e_shnum; i++) {
+
+ if (get_sec(elf, i, &ehdr, &shname, &shdr, &data))
+ continue;
+
+ if (0) /* helpful for llvm debugging */
+ printf("section %d:%s data %p size %zd link %d flags %d\n",
+ i, shname, data->d_buf, data->d_size,
+ shdr.sh_link, (int) shdr.sh_flags);
+
+ if (strcmp(shname, "license") == 0) {
+ processed_sec[i] = true;
+ memcpy(license, data->d_buf, data->d_size);
+ } else if (strcmp(shname, "version") == 0) {
+ processed_sec[i] = true;
+ if (data->d_size != sizeof(int)) {
+ printf("invalid size of version section %zd\n",
+ data->d_size);
+ return 1;
+ }
+ memcpy(&kern_version, data->d_buf, sizeof(int));
+ } else if (strcmp(shname, "maps") == 0) {
+ int j;
+
+ maps_shndx = i;
+ data_maps = data;
+ for (j = 0; j < MAX_MAPS; j++)
+ map_data[j].fd = -1;
+ } else if (shdr.sh_type == SHT_SYMTAB) {
+ strtabidx = shdr.sh_link;
+ symbols = data;
+ }
+ }
+
+ ret = 1;
+
+ if (!symbols) {
+ printf("missing SHT_SYMTAB section\n");
+ goto done;
+ }
+
+ if (data_maps) {
+ nr_maps = load_elf_maps_section(map_data, maps_shndx,
+ elf, symbols, strtabidx);
+ if (nr_maps < 0) {
+ printf("Error: Failed loading ELF maps (errno:%d):%s\n",
+ nr_maps, strerror(-nr_maps));
+ ret = 1;
+ goto done;
+ }
+ if (load_maps(map_data, nr_maps, fixup_map))
+ goto done;
+ map_data_count = nr_maps;
+
+ processed_sec[maps_shndx] = true;
+ }
+
+ /* process all relo sections, and rewrite bpf insns for maps */
+ for (i = 1; i < ehdr.e_shnum; i++) {
+ if (processed_sec[i])
+ continue;
+
+ if (get_sec(elf, i, &ehdr, &shname, &shdr, &data))
+ continue;
+
+ if (shdr.sh_type == SHT_REL) {
+ struct bpf_insn *insns;
+
+ /* locate prog sec that need map fixup (relocations) */
+ if (get_sec(elf, shdr.sh_info, &ehdr, &shname_prog,
+ &shdr_prog, &data_prog))
+ continue;
+
+ if (shdr_prog.sh_type != SHT_PROGBITS ||
+ !(shdr_prog.sh_flags & SHF_EXECINSTR))
+ continue;
+
+ insns = (struct bpf_insn *) data_prog->d_buf;
+ processed_sec[i] = true; /* relo section */
+
+ if (parse_relo_and_apply(data, symbols, &shdr, insns,
+ map_data, nr_maps))
+ continue;
+ }
+ }
+
+ /* load programs */
+ for (i = 1; i < ehdr.e_shnum; i++) {
+
+ if (processed_sec[i])
+ continue;
+
+ if (get_sec(elf, i, &ehdr, &shname, &shdr, &data))
+ continue;
+
+ if (memcmp(shname, "kprobe/", 7) == 0 ||
+ memcmp(shname, "kretprobe/", 10) == 0 ||
+ memcmp(shname, "tracepoint/", 11) == 0 ||
+ memcmp(shname, "xdp", 3) == 0 ||
+ memcmp(shname, "perf_event", 10) == 0 ||
+ memcmp(shname, "socket", 6) == 0 ||
+ memcmp(shname, "cgroup/", 7) == 0 ||
+ memcmp(shname, "sockops", 7) == 0 ||
+ memcmp(shname, "sk_skb", 6) == 0) {
+ ret = load_and_attach(shname, data->d_buf,
+ data->d_size);
+ if (ret != 0)
+ goto done;
+ }
+ }
+
+ ret = 0;
+done:
+ close(fd);
+ return ret;
+}
+
+int load_bpf_file(const char *path)
+{
+ return do_load_bpf_file(path, NULL);
+}
+
+int load_bpf_file_fixup_map(const char *path, fixup_map_cb fixup_map)
+{
+ return do_load_bpf_file(path, fixup_map);
+}
+
+void read_trace_pipe(void)
+{
+ int trace_fd;
+
+ trace_fd = open(DEBUGFS "trace_pipe", O_RDONLY, 0);
+ if (trace_fd < 0)
+ return;
+
+ while (1) {
+ static char buf[4096];
+ ssize_t sz;
+
+ sz = read(trace_fd, buf, sizeof(buf));
+ if (sz > 0) {
+ buf[sz] = 0;
+ puts(buf);
+ }
+ }
+}
+
+#define MAX_SYMS 300000
+static struct ksym syms[MAX_SYMS];
+static int sym_cnt;
+
+static int ksym_cmp(const void *p1, const void *p2)
+{
+ return ((struct ksym *)p1)->addr - ((struct ksym *)p2)->addr;
+}
+
+int load_kallsyms(void)
+{
+ FILE *f = fopen("/proc/kallsyms", "r");
+ char func[256], buf[256];
+ char symbol;
+ void *addr;
+ int i = 0;
+
+ if (!f)
+ return -ENOENT;
+
+ while (!feof(f)) {
+ if (!fgets(buf, sizeof(buf), f))
+ break;
+ if (sscanf(buf, "%p %c %s", &addr, &symbol, func) != 3)
+ break;
+ if (!addr)
+ continue;
+ syms[i].addr = (long) addr;
+ syms[i].name = strdup(func);
+ i++;
+ }
+ sym_cnt = i;
+ qsort(syms, sym_cnt, sizeof(struct ksym), ksym_cmp);
+ return 0;
+}
+
+struct ksym *ksym_search(long key)
+{
+ int start = 0, end = sym_cnt;
+ int result;
+
+ while (start < end) {
+ size_t mid = start + (end - start) / 2;
+
+ result = key - syms[mid].addr;
+ if (result < 0)
+ end = mid;
+ else if (result > 0)
+ start = mid + 1;
+ else
+ return &syms[mid];
+ }
+
+ if (start >= 1 && syms[start - 1].addr < key &&
+ key < syms[start].addr)
+ /* valid ksym */
+ return &syms[start - 1];
+
+ /* out of range. return _stext */
+ return &syms[0];
+}
+
+int set_link_xdp_fd(int ifindex, int fd, __u32 flags)
+{
+ struct sockaddr_nl sa;
+ int sock, seq = 0, len, ret = -1;
+ char buf[4096];
+ struct nlattr *nla, *nla_xdp;
+ struct {
+ struct nlmsghdr nh;
+ struct ifinfomsg ifinfo;
+ char attrbuf[64];
+ } req;
+ struct nlmsghdr *nh;
+ struct nlmsgerr *err;
+
+ memset(&sa, 0, sizeof(sa));
+ sa.nl_family = AF_NETLINK;
+
+ sock = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE);
+ if (sock < 0) {
+ printf("open netlink socket: %s\n", strerror(errno));
+ return -1;
+ }
+
+ if (bind(sock, (struct sockaddr *)&sa, sizeof(sa)) < 0) {
+ printf("bind to netlink: %s\n", strerror(errno));
+ goto cleanup;
+ }
+
+ memset(&req, 0, sizeof(req));
+ req.nh.nlmsg_len = NLMSG_LENGTH(sizeof(struct ifinfomsg));
+ req.nh.nlmsg_flags = NLM_F_REQUEST | NLM_F_ACK;
+ req.nh.nlmsg_type = RTM_SETLINK;
+ req.nh.nlmsg_pid = 0;
+ req.nh.nlmsg_seq = ++seq;
+ req.ifinfo.ifi_family = AF_UNSPEC;
+ req.ifinfo.ifi_index = ifindex;
+
+ /* started nested attribute for XDP */
+ nla = (struct nlattr *)(((char *)&req)
+ + NLMSG_ALIGN(req.nh.nlmsg_len));
+ nla->nla_type = NLA_F_NESTED | 43/*IFLA_XDP*/;
+ nla->nla_len = NLA_HDRLEN;
+
+ /* add XDP fd */
+ nla_xdp = (struct nlattr *)((char *)nla + nla->nla_len);
+ nla_xdp->nla_type = 1/*IFLA_XDP_FD*/;
+ nla_xdp->nla_len = NLA_HDRLEN + sizeof(int);
+ memcpy((char *)nla_xdp + NLA_HDRLEN, &fd, sizeof(fd));
+ nla->nla_len += nla_xdp->nla_len;
+
+ /* if user passed in any flags, add those too */
+ if (flags) {
+ nla_xdp = (struct nlattr *)((char *)nla + nla->nla_len);
+ nla_xdp->nla_type = 3/*IFLA_XDP_FLAGS*/;
+ nla_xdp->nla_len = NLA_HDRLEN + sizeof(flags);
+ memcpy((char *)nla_xdp + NLA_HDRLEN, &flags, sizeof(flags));
+ nla->nla_len += nla_xdp->nla_len;
+ }
+
+ req.nh.nlmsg_len += NLA_ALIGN(nla->nla_len);
+
+ if (send(sock, &req, req.nh.nlmsg_len, 0) < 0) {
+ printf("send to netlink: %s\n", strerror(errno));
+ goto cleanup;
+ }
+
+ len = recv(sock, buf, sizeof(buf), 0);
+ if (len < 0) {
+ printf("recv from netlink: %s\n", strerror(errno));
+ goto cleanup;
+ }
+
+ for (nh = (struct nlmsghdr *)buf; NLMSG_OK(nh, len);
+ nh = NLMSG_NEXT(nh, len)) {
+ if (nh->nlmsg_pid != getpid()) {
+ printf("Wrong pid %d, expected %d\n",
+ nh->nlmsg_pid, getpid());
+ goto cleanup;
+ }
+ if (nh->nlmsg_seq != seq) {
+ printf("Wrong seq %d, expected %d\n",
+ nh->nlmsg_seq, seq);
+ goto cleanup;
+ }
+ switch (nh->nlmsg_type) {
+ case NLMSG_ERROR:
+ err = (struct nlmsgerr *)NLMSG_DATA(nh);
+ if (!err->error)
+ continue;
+ printf("nlmsg error %s\n", strerror(-err->error));
+ goto cleanup;
+ case NLMSG_DONE:
+ break;
+ }
+ }
+
+ ret = 0;
+
+cleanup:
+ close(sock);
+ return ret;
+}
diff --git a/drivers/net/af_xdp/bpf_load.h b/drivers/net/af_xdp/bpf_load.h
new file mode 100644
index 000000000..5450e8b19
--- /dev/null
+++ b/drivers/net/af_xdp/bpf_load.h
@@ -0,0 +1,65 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __BPF_LOAD_H
+#define __BPF_LOAD_H
+
+#include "libbpf.h"
+
+#define MAX_MAPS 32
+#define MAX_PROGS 32
+
+struct bpf_map_def {
+ unsigned int type;
+ unsigned int key_size;
+ unsigned int value_size;
+ unsigned int max_entries;
+ unsigned int map_flags;
+ unsigned int inner_map_idx;
+ unsigned int numa_node;
+};
+
+struct bpf_map_data {
+ int fd;
+ char *name;
+ size_t elf_offset;
+ struct bpf_map_def def;
+};
+
+typedef void (*fixup_map_cb)(struct bpf_map_data *map, int idx);
+
+extern int prog_fd[MAX_PROGS];
+extern int event_fd[MAX_PROGS];
+extern char bpf_log_buf[BPF_LOG_BUF_SIZE];
+extern int prog_cnt;
+
+/* There is a one-to-one mapping between map_fd[] and map_data[].
+ * The map_data[] just contains more rich info on the given map.
+ */
+extern int map_fd[MAX_MAPS];
+extern struct bpf_map_data map_data[MAX_MAPS];
+extern int map_data_count;
+
+/* parses elf file compiled by llvm .c->.o
+ * . parses 'maps' section and creates maps via BPF syscall
+ * . parses 'license' section and passes it to syscall
+ * . parses elf relocations for BPF maps and adjusts BPF_LD_IMM64 insns by
+ * storing map_fd into insn->imm and marking such insns as BPF_PSEUDO_MAP_FD
+ * . loads eBPF programs via BPF syscall
+ *
+ * One ELF file can contain multiple BPF programs which will be loaded
+ * and their FDs stored stored in prog_fd array
+ *
+ * returns zero on success
+ */
+int load_bpf_file(const char *path);
+int load_bpf_file_fixup_map(const char *path, fixup_map_cb fixup_map);
+
+void read_trace_pipe(void);
+struct ksym {
+ long addr;
+ char *name;
+};
+
+int load_kallsyms(void);
+struct ksym *ksym_search(long key);
+int set_link_xdp_fd(int ifindex, int fd, __u32 flags);
+#endif
diff --git a/drivers/net/af_xdp/libbpf.h b/drivers/net/af_xdp/libbpf.h
new file mode 100644
index 000000000..18bfee5aa
--- /dev/null
+++ b/drivers/net/af_xdp/libbpf.h
@@ -0,0 +1,199 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* eBPF mini library */
+#ifndef __LIBBPF_H
+#define __LIBBPF_H
+
+#include <bpf/bpf.h>
+
+struct bpf_insn;
+
+/* ALU ops on registers, bpf_add|sub|...: dst_reg += src_reg */
+
+#define BPF_ALU64_REG(OP, DST, SRC) \
+ ((struct bpf_insn) { \
+ .code = BPF_ALU64 | BPF_OP(OP) | BPF_X, \
+ .dst_reg = DST, \
+ .src_reg = SRC, \
+ .off = 0, \
+ .imm = 0 })
+
+#define BPF_ALU32_REG(OP, DST, SRC) \
+ ((struct bpf_insn) { \
+ .code = BPF_ALU | BPF_OP(OP) | BPF_X, \
+ .dst_reg = DST, \
+ .src_reg = SRC, \
+ .off = 0, \
+ .imm = 0 })
+
+/* ALU ops on immediates, bpf_add|sub|...: dst_reg += imm32 */
+
+#define BPF_ALU64_IMM(OP, DST, IMM) \
+ ((struct bpf_insn) { \
+ .code = BPF_ALU64 | BPF_OP(OP) | BPF_K, \
+ .dst_reg = DST, \
+ .src_reg = 0, \
+ .off = 0, \
+ .imm = IMM })
+
+#define BPF_ALU32_IMM(OP, DST, IMM) \
+ ((struct bpf_insn) { \
+ .code = BPF_ALU | BPF_OP(OP) | BPF_K, \
+ .dst_reg = DST, \
+ .src_reg = 0, \
+ .off = 0, \
+ .imm = IMM })
+
+/* Short form of mov, dst_reg = src_reg */
+
+#define BPF_MOV64_REG(DST, SRC) \
+ ((struct bpf_insn) { \
+ .code = BPF_ALU64 | BPF_MOV | BPF_X, \
+ .dst_reg = DST, \
+ .src_reg = SRC, \
+ .off = 0, \
+ .imm = 0 })
+
+#define BPF_MOV32_REG(DST, SRC) \
+ ((struct bpf_insn) { \
+ .code = BPF_ALU | BPF_MOV | BPF_X, \
+ .dst_reg = DST, \
+ .src_reg = SRC, \
+ .off = 0, \
+ .imm = 0 })
+
+/* Short form of mov, dst_reg = imm32 */
+
+#define BPF_MOV64_IMM(DST, IMM) \
+ ((struct bpf_insn) { \
+ .code = BPF_ALU64 | BPF_MOV | BPF_K, \
+ .dst_reg = DST, \
+ .src_reg = 0, \
+ .off = 0, \
+ .imm = IMM })
+
+#define BPF_MOV32_IMM(DST, IMM) \
+ ((struct bpf_insn) { \
+ .code = BPF_ALU | BPF_MOV | BPF_K, \
+ .dst_reg = DST, \
+ .src_reg = 0, \
+ .off = 0, \
+ .imm = IMM })
+
+/* BPF_LD_IMM64 macro encodes single 'load 64-bit immediate' insn */
+#define BPF_LD_IMM64(DST, IMM) \
+ BPF_LD_IMM64_RAW(DST, 0, IMM)
+
+#define BPF_LD_IMM64_RAW(DST, SRC, IMM) \
+ ((struct bpf_insn) { \
+ .code = BPF_LD | BPF_DW | BPF_IMM, \
+ .dst_reg = DST, \
+ .src_reg = SRC, \
+ .off = 0, \
+ .imm = (__u32) (IMM) }), \
+ ((struct bpf_insn) { \
+ .code = 0, /* zero is reserved opcode */ \
+ .dst_reg = 0, \
+ .src_reg = 0, \
+ .off = 0, \
+ .imm = ((__u64) (IMM)) >> 32 })
+
+#ifndef BPF_PSEUDO_MAP_FD
+# define BPF_PSEUDO_MAP_FD 1
+#endif
+
+/* pseudo BPF_LD_IMM64 insn used to refer to process-local map_fd */
+#define BPF_LD_MAP_FD(DST, MAP_FD) \
+ BPF_LD_IMM64_RAW(DST, BPF_PSEUDO_MAP_FD, MAP_FD)
+
+
+/* Direct packet access, R0 = *(uint *) (skb->data + imm32) */
+
+#define BPF_LD_ABS(SIZE, IMM) \
+ ((struct bpf_insn) { \
+ .code = BPF_LD | BPF_SIZE(SIZE) | BPF_ABS, \
+ .dst_reg = 0, \
+ .src_reg = 0, \
+ .off = 0, \
+ .imm = IMM })
+
+/* Memory load, dst_reg = *(uint *) (src_reg + off16) */
+
+#define BPF_LDX_MEM(SIZE, DST, SRC, OFF) \
+ ((struct bpf_insn) { \
+ .code = BPF_LDX | BPF_SIZE(SIZE) | BPF_MEM, \
+ .dst_reg = DST, \
+ .src_reg = SRC, \
+ .off = OFF, \
+ .imm = 0 })
+
+/* Memory store, *(uint *) (dst_reg + off16) = src_reg */
+
+#define BPF_STX_MEM(SIZE, DST, SRC, OFF) \
+ ((struct bpf_insn) { \
+ .code = BPF_STX | BPF_SIZE(SIZE) | BPF_MEM, \
+ .dst_reg = DST, \
+ .src_reg = SRC, \
+ .off = OFF, \
+ .imm = 0 })
+
+/* Atomic memory add, *(uint *)(dst_reg + off16) += src_reg */
+
+#define BPF_STX_XADD(SIZE, DST, SRC, OFF) \
+ ((struct bpf_insn) { \
+ .code = BPF_STX | BPF_SIZE(SIZE) | BPF_XADD, \
+ .dst_reg = DST, \
+ .src_reg = SRC, \
+ .off = OFF, \
+ .imm = 0 })
+
+/* Memory store, *(uint *) (dst_reg + off16) = imm32 */
+
+#define BPF_ST_MEM(SIZE, DST, OFF, IMM) \
+ ((struct bpf_insn) { \
+ .code = BPF_ST | BPF_SIZE(SIZE) | BPF_MEM, \
+ .dst_reg = DST, \
+ .src_reg = 0, \
+ .off = OFF, \
+ .imm = IMM })
+
+/* Conditional jumps against registers, if (dst_reg 'op' src_reg) goto pc + off16 */
+
+#define BPF_JMP_REG(OP, DST, SRC, OFF) \
+ ((struct bpf_insn) { \
+ .code = BPF_JMP | BPF_OP(OP) | BPF_X, \
+ .dst_reg = DST, \
+ .src_reg = SRC, \
+ .off = OFF, \
+ .imm = 0 })
+
+/* Conditional jumps against immediates, if (dst_reg 'op' imm32) goto pc + off16 */
+
+#define BPF_JMP_IMM(OP, DST, IMM, OFF) \
+ ((struct bpf_insn) { \
+ .code = BPF_JMP | BPF_OP(OP) | BPF_K, \
+ .dst_reg = DST, \
+ .src_reg = 0, \
+ .off = OFF, \
+ .imm = IMM })
+
+/* Raw code statement block */
+
+#define BPF_RAW_INSN(CODE, DST, SRC, OFF, IMM) \
+ ((struct bpf_insn) { \
+ .code = CODE, \
+ .dst_reg = DST, \
+ .src_reg = SRC, \
+ .off = OFF, \
+ .imm = IMM })
+
+/* Program exit */
+
+#define BPF_EXIT_INSN() \
+ ((struct bpf_insn) { \
+ .code = BPF_JMP | BPF_EXIT, \
+ .dst_reg = 0, \
+ .src_reg = 0, \
+ .off = 0, \
+ .imm = 0 })
+
+#endif
diff --git a/drivers/net/af_xdp/rte_eth_af_xdp.c b/drivers/net/af_xdp/rte_eth_af_xdp.c
index d0939022b..903ca0d01 100644
--- a/drivers/net/af_xdp/rte_eth_af_xdp.c
+++ b/drivers/net/af_xdp/rte_eth_af_xdp.c
@@ -15,6 +15,7 @@
#include <linux/if_ether.h>
#include <linux/if_xdp.h>
+#include <linux/if_link.h>
#include <arpa/inet.h>
#include <net/if.h>
#include <sys/types.h>
@@ -24,6 +25,7 @@
#include <unistd.h>
#include <poll.h>
#include "xdpsock_queue.h"
+#include "bpf_load.h"
#ifndef SOL_XDP
#define SOL_XDP 283
@@ -85,6 +87,8 @@ struct pmd_internals {
uint16_t port_id;
uint16_t queue_idx;
int ring_size;
+
+ uint32_t xdp_flags;
};
static const char * const valid_arguments[] = {
@@ -382,8 +386,12 @@ eth_stats_reset(struct rte_eth_dev *dev)
}
static void
-eth_dev_close(struct rte_eth_dev *dev __rte_unused)
+eth_dev_close(struct rte_eth_dev *dev)
{
+ struct pmd_internals *internals = dev->data->dev_private;
+
+ if (internals->xdp_flags)
+ set_link_xdp_fd(internals->if_index, -1, internals->xdp_flags);
}
static void
@@ -745,9 +753,25 @@ init_internals(struct rte_vdev_device *dev,
if (ret)
goto error_3;
+ /* need fix: hard coded bpf file */
+ if (load_bpf_file("xdpsock_kern.o")) {
+ printf("load bpf file failed\n");
+ goto error_3;
+ }
+ RTE_ASSERT(prog_fd[0]);
+
+ if (!set_link_xdp_fd(internals->if_index, prog_fd[0],
+ XDP_FLAGS_DRV_MODE))
+ internals->xdp_flags = XDP_FLAGS_DRV_MODE;
+ else if (!set_link_xdp_fd(internals->if_index, prog_fd[0],
+ XDP_FLAGS_SKB_MODE))
+ internals->xdp_flags = XDP_FLAGS_SKB_MODE;
+ else
+ goto error_3;
+
eth_dev = rte_eth_vdev_allocate(dev, 0);
if (!eth_dev)
- goto error_3;
+ goto error_4;
rte_memcpy(data, eth_dev->data, sizeof(*data));
internals->port_id = eth_dev->data->port_id;
@@ -765,6 +789,9 @@ init_internals(struct rte_vdev_device *dev,
return 0;
+error_4:
+ set_link_xdp_fd(internals->if_index, -1, internals->xdp_flags);
+
error_3:
close(internals->sfd);
diff --git a/mk/rte.app.mk b/mk/rte.app.mk
index bc26e1457..d05e6c0e4 100644
--- a/mk/rte.app.mk
+++ b/mk/rte.app.mk
@@ -120,7 +120,7 @@ ifeq ($(CONFIG_RTE_BUILD_SHARED_LIB),n)
_LDLIBS-$(CONFIG_RTE_DRIVER_MEMPOOL_STACK) += -lrte_mempool_stack
_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_AF_PACKET) += -lrte_pmd_af_packet
-_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_AF_XDP) += -lrte_pmd_af_xdp
+_LDLIBS-$(CONFIG_RTE_LIBRTE_PMD_AF_XDP) += -lrte_pmd_af_xdp -lelf -lbpf
_LDLIBS-$(CONFIG_RTE_LIBRTE_ARK_PMD) += -lrte_pmd_ark
_LDLIBS-$(CONFIG_RTE_LIBRTE_AVF_PMD) += -lrte_pmd_avf
_LDLIBS-$(CONFIG_RTE_LIBRTE_AVP_PMD) += -lrte_pmd_avp
--
2.13.6
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [RFC 6/7] net/af_xdp: load BPF file
2018-02-27 9:33 ` [RFC 6/7] net/af_xdp: load BPF file Qi Zhang
@ 2018-03-01 2:10 ` Stephen Hemminger
0 siblings, 0 replies; 24+ messages in thread
From: Stephen Hemminger @ 2018-03-01 2:10 UTC (permalink / raw)
To: Qi Zhang; +Cc: dev, magnus.karlsson, bjorn.topel
On Tue, 27 Feb 2018 17:33:05 +0800
Qi Zhang <qi.z.zhang@intel.com> wrote:
> include $(RTE_SDK)/mk/rte.lib.mk
> diff --git a/drivers/net/af_xdp/bpf_load.c b/drivers/net/af_xdp/bpf_load.c
> new file mode 100644
> index 000000000..aa632207f
> --- /dev/null
> +++ b/drivers/net/af_xdp/bpf_load.c
> @@ -0,0 +1,798 @@
> +// SPDX-License-Identifier: GPL-2.0
Sorry all DPDK drivers must be BSD licensed. You can't use GPL-2.0 code.
Either get the code dual licensed or write a new loader.
^ permalink raw reply [flat|nested] 24+ messages in thread
* [RFC 7/7] app/testpmd: enable parameter for mempool flags
2018-02-27 9:32 [RFC 0/7] PMD driver for AF_XDP Qi Zhang
` (5 preceding siblings ...)
2018-02-27 9:33 ` [RFC 6/7] net/af_xdp: load BPF file Qi Zhang
@ 2018-02-27 9:33 ` Qi Zhang
2018-03-01 2:52 ` [RFC 0/7] PMD driver for AF_XDP Jason Wang
7 siblings, 0 replies; 24+ messages in thread
From: Qi Zhang @ 2018-02-27 9:33 UTC (permalink / raw)
To: dev; +Cc: magnus.karlsson, bjorn.topel, Qi Zhang
Now, it is possible for testpmd to create a af_xdp friendly mempool.
Signed-off-by: Qi Zhang <qi.z.zhang@intel.com>
---
app/test-pmd/parameters.c | 12 ++++++++++++
app/test-pmd/testpmd.c | 15 +++++++++------
app/test-pmd/testpmd.h | 1 +
3 files changed, 22 insertions(+), 6 deletions(-)
diff --git a/app/test-pmd/parameters.c b/app/test-pmd/parameters.c
index 97d22b860..19675671e 100644
--- a/app/test-pmd/parameters.c
+++ b/app/test-pmd/parameters.c
@@ -61,6 +61,7 @@ usage(char* progname)
"--tx-first | --stats-period=PERIOD | "
"--coremask=COREMASK --portmask=PORTMASK --numa "
"--mbuf-size= | --total-num-mbufs= | "
+ "--mp-flags= | "
"--nb-cores= | --nb-ports= | "
#ifdef RTE_LIBRTE_CMDLINE
"--eth-peers-configfile= | "
@@ -105,6 +106,7 @@ usage(char* progname)
printf(" --socket-num=N: set socket from which all memory is allocated "
"in NUMA mode.\n");
printf(" --mbuf-size=N: set the data size of mbuf to N bytes.\n");
+ printf(" --mp-flags=N: set the flags when create mbuf memory pool.\n");
printf(" --total-num-mbufs=N: set the number of mbufs to be allocated "
"in mbuf pools.\n");
printf(" --max-pkt-len=N: set the maximum size of packet to N bytes.\n");
@@ -568,6 +570,7 @@ launch_args_parse(int argc, char** argv)
{ "ring-numa-config", 1, 0, 0 },
{ "socket-num", 1, 0, 0 },
{ "mbuf-size", 1, 0, 0 },
+ { "mp-flags", 1, 0, 0 },
{ "total-num-mbufs", 1, 0, 0 },
{ "max-pkt-len", 1, 0, 0 },
{ "pkt-filter-mode", 1, 0, 0 },
@@ -769,6 +772,15 @@ launch_args_parse(int argc, char** argv)
rte_exit(EXIT_FAILURE,
"mbuf-size should be > 0 and < 65536\n");
}
+ if (!strcmp(lgopts[opt_idx].name, "mp-flags")) {
+ n = atoi(optarg);
+ if (n > 0 && n <= 0xFFFF)
+ mp_flags = (uint16_t)n;
+ else
+ rte_exit(EXIT_FAILURE,
+ "mp-flags should be > 0 and < 65536\n");
+ }
+
if (!strcmp(lgopts[opt_idx].name, "total-num-mbufs")) {
n = atoi(optarg);
if (n > 1024)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index 4c0e2586c..887899919 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -171,6 +171,7 @@ uint32_t burst_tx_delay_time = BURST_TX_WAIT_US;
uint32_t burst_tx_retry_num = BURST_TX_RETRIES;
uint16_t mbuf_data_size = DEFAULT_MBUF_DATA_SIZE; /**< Mbuf data space size. */
+uint16_t mp_flags = 0; /**< flags parsed when create mempool */
uint32_t param_total_num_mbufs = 0; /**< number of mbufs in all pools - if
* specified on command-line. */
uint16_t stats_period; /**< Period to show statistics (disabled by default) */
@@ -486,6 +487,7 @@ set_def_fwd_config(void)
*/
static void
mbuf_pool_create(uint16_t mbuf_seg_size, unsigned nb_mbuf,
+ unsigned int flags,
unsigned int socket_id)
{
char pool_name[RTE_MEMPOOL_NAMESIZE];
@@ -503,7 +505,7 @@ mbuf_pool_create(uint16_t mbuf_seg_size, unsigned nb_mbuf,
rte_mp = rte_mempool_create_empty(pool_name, nb_mbuf,
mb_size, (unsigned) mb_mempool_cache,
sizeof(struct rte_pktmbuf_pool_private),
- socket_id, 0);
+ socket_id, flags);
if (rte_mp == NULL)
goto err;
@@ -518,8 +520,8 @@ mbuf_pool_create(uint16_t mbuf_seg_size, unsigned nb_mbuf,
/* wrapper to rte_mempool_create() */
TESTPMD_LOG(INFO, "preferred mempool ops selected: %s\n",
rte_mbuf_best_mempool_ops());
- rte_mp = rte_pktmbuf_pool_create(pool_name, nb_mbuf,
- mb_mempool_cache, 0, mbuf_seg_size, socket_id);
+ rte_mp = rte_pktmbuf_pool_create_with_flags(pool_name, nb_mbuf,
+ mb_mempool_cache, 0, mbuf_seg_size, flags, socket_id);
}
err:
@@ -735,13 +737,14 @@ init_config(void)
for (i = 0; i < num_sockets; i++)
mbuf_pool_create(mbuf_data_size, nb_mbuf_per_pool,
- socket_ids[i]);
+ mp_flags, socket_ids[i]);
} else {
if (socket_num == UMA_NO_CONFIG)
- mbuf_pool_create(mbuf_data_size, nb_mbuf_per_pool, 0);
+ mbuf_pool_create(mbuf_data_size, nb_mbuf_per_pool,
+ mp_flags, 0);
else
mbuf_pool_create(mbuf_data_size, nb_mbuf_per_pool,
- socket_num);
+ mp_flags, socket_num);
}
init_port_config();
diff --git a/app/test-pmd/testpmd.h b/app/test-pmd/testpmd.h
index 153abea05..11c2ea681 100644
--- a/app/test-pmd/testpmd.h
+++ b/app/test-pmd/testpmd.h
@@ -386,6 +386,7 @@ extern uint8_t dcb_config;
extern uint8_t dcb_test;
extern uint16_t mbuf_data_size; /**< Mbuf data space size. */
+extern uint16_t mp_flags; /**< flags for mempool creation. */
extern uint32_t param_total_num_mbufs;
extern uint16_t stats_period;
--
2.13.6
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [RFC 0/7] PMD driver for AF_XDP
2018-02-27 9:32 [RFC 0/7] PMD driver for AF_XDP Qi Zhang
` (6 preceding siblings ...)
2018-02-27 9:33 ` [RFC 7/7] app/testpmd: enable parameter for mempool flags Qi Zhang
@ 2018-03-01 2:52 ` Jason Wang
2018-03-01 4:18 ` Zhang, Qi Z
7 siblings, 1 reply; 24+ messages in thread
From: Jason Wang @ 2018-03-01 2:52 UTC (permalink / raw)
To: Qi Zhang, dev; +Cc: magnus.karlsson, bjorn.topel
On 2018年02月27日 17:32, Qi Zhang wrote:
> The RFC patches add a new PMD driver for AF_XDP which is a proposed
> faster version of AF_PACKET interface in Linux, see below link for
> detail AF_XDP introduction:
> https://fosdem.org/2018/schedule/event/af_xdp/
> https://lwn.net/Articles/745934/
>
> This patchset is base on v18.02.
> It also require a linux kernel that have below AF_XDP RFC patches be
> applied.
> https://patchwork.ozlabs.org/patch/867961/
> https://patchwork.ozlabs.org/patch/867960/
> https://patchwork.ozlabs.org/patch/867938/
> https://patchwork.ozlabs.org/patch/867939/
> https://patchwork.ozlabs.org/patch/867940/
> https://patchwork.ozlabs.org/patch/867941/
> https://patchwork.ozlabs.org/patch/867942/
> https://patchwork.ozlabs.org/patch/867943/
> https://patchwork.ozlabs.org/patch/867944/
> https://patchwork.ozlabs.org/patch/867945/
> https://patchwork.ozlabs.org/patch/867946/
> https://patchwork.ozlabs.org/patch/867947/
> https://patchwork.ozlabs.org/patch/867948/
> https://patchwork.ozlabs.org/patch/867949/
> https://patchwork.ozlabs.org/patch/867950/
> https://patchwork.ozlabs.org/patch/867951/
> https://patchwork.ozlabs.org/patch/867952/
> https://patchwork.ozlabs.org/patch/867953/
> https://patchwork.ozlabs.org/patch/867954/
> https://patchwork.ozlabs.org/patch/867955/
> https://patchwork.ozlabs.org/patch/867956/
> https://patchwork.ozlabs.org/patch/867957/
> https://patchwork.ozlabs.org/patch/867958/
> https://patchwork.ozlabs.org/patch/867959/
>
> There is no clean upstream target yet since kernel patch is still in
> RFC stage, The purpose of the patchset is just for anyone that want to
> eveluate af_xdp with DPDK application and get feedback for further
> improvement.
>
> To try with the new PMD
> 1. compile and install the kernel with above patches applied.
> 2. configure $LINUX_HEADER_DIR (dir of "make headers_install")
> and $TOOLS_DIR (dir at <kernel_src>/tools) at driver/net/af_xdp/Makefile
> before compile DPDK.
> 3. make sure libelf and libbpf is installed.
>
> BTW, performance test shows our PMD can reach 94%~98% of the orignal benchmark
> when share memory is enabled.
Hi:
Looks like zerocopy is not used in this series. Any plan to support
that? If not, what's the advantage compared to vhost-net + tap +
XDP_REDIRECT?
Have you measured l2fwd performance in this case? I believe the number
you refer here is rxdrop (XDP_DRV) which is 11.6Mpps.
Thanks
^ permalink raw reply [flat|nested] 24+ messages in thread* Re: [RFC 0/7] PMD driver for AF_XDP
2018-03-01 2:52 ` [RFC 0/7] PMD driver for AF_XDP Jason Wang
@ 2018-03-01 4:18 ` Zhang, Qi Z
2018-03-01 4:20 ` Zhang, Qi Z
0 siblings, 1 reply; 24+ messages in thread
From: Zhang, Qi Z @ 2018-03-01 4:18 UTC (permalink / raw)
To: Jason Wang, dev; +Cc: magnus.karlsson, Topel, Bjorn
> -----Original Message-----
> From: Jason Wang [mailto:jasowang@redhat.com]
> Sent: Thursday, March 1, 2018 10:52 AM
> To: Zhang, Qi Z <qi.z.zhang@intel.com>; dev@dpdk.org
> Cc: magnus.karlsson@intei.com; Topel, Bjorn <bjorn.topel@intel.com>
> Subject: Re: [dpdk-dev] [RFC 0/7] PMD driver for AF_XDP
>
>
>
> On 2018年02月27日 17:32, Qi Zhang wrote:
> > The RFC patches add a new PMD driver for AF_XDP which is a proposed
> > faster version of AF_PACKET interface in Linux, see below link for
> > detail AF_XDP introduction:
> > https://fosdem.org/2018/schedule/event/af_xdp/
> > https://lwn.net/Articles/745934/
> >
> > This patchset is base on v18.02.
> > It also require a linux kernel that have below AF_XDP RFC patches be
> > applied.
> > https://patchwork.ozlabs.org/patch/867961/
> > https://patchwork.ozlabs.org/patch/867960/
> > https://patchwork.ozlabs.org/patch/867938/
> > https://patchwork.ozlabs.org/patch/867939/
> > https://patchwork.ozlabs.org/patch/867940/
> > https://patchwork.ozlabs.org/patch/867941/
> > https://patchwork.ozlabs.org/patch/867942/
> > https://patchwork.ozlabs.org/patch/867943/
> > https://patchwork.ozlabs.org/patch/867944/
> > https://patchwork.ozlabs.org/patch/867945/
> > https://patchwork.ozlabs.org/patch/867946/
> > https://patchwork.ozlabs.org/patch/867947/
> > https://patchwork.ozlabs.org/patch/867948/
> > https://patchwork.ozlabs.org/patch/867949/
> > https://patchwork.ozlabs.org/patch/867950/
> > https://patchwork.ozlabs.org/patch/867951/
> > https://patchwork.ozlabs.org/patch/867952/
> > https://patchwork.ozlabs.org/patch/867953/
> > https://patchwork.ozlabs.org/patch/867954/
> > https://patchwork.ozlabs.org/patch/867955/
> > https://patchwork.ozlabs.org/patch/867956/
> > https://patchwork.ozlabs.org/patch/867957/
> > https://patchwork.ozlabs.org/patch/867958/
> > https://patchwork.ozlabs.org/patch/867959/
> >
> > There is no clean upstream target yet since kernel patch is still in
> > RFC stage, The purpose of the patchset is just for anyone that want to
> > eveluate af_xdp with DPDK application and get feedback for further
> > improvement.
> >
> > To try with the new PMD
> > 1. compile and install the kernel with above patches applied.
> > 2. configure $LINUX_HEADER_DIR (dir of "make headers_install")
> > and $TOOLS_DIR (dir at <kernel_src>/tools) at
> driver/net/af_xdp/Makefile
> > before compile DPDK.
> > 3. make sure libelf and libbpf is installed.
> >
> > BTW, performance test shows our PMD can reach 94%~98% of the orignal
> > benchmark when share memory is enabled.
>
> Hi:
>
> Looks like zero copy is not used in this series. Any plan to support that?
Zero copy is enabled in patch 5, if a mempool passed check_mempool, it will be registered to af_xdp socket.
so there will be no memcpy between mbuf and af_xdp.
> If not, what's the advantage compared to vhost-net + tap + XDP_REDIRECT?
>
> Have you measured l2fwd performance in this case? I believe the number
> you refer here is rxdrop (XDP_DRV) which is 11.6Mpps.
Actually we measure the performance on rxonly / txonly / l2fwd on i40e with XDP_SKB and XDP_DRV_ZC
Regards
Qi
>
> Thanks
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [RFC 0/7] PMD driver for AF_XDP
2018-03-01 4:18 ` Zhang, Qi Z
@ 2018-03-01 4:20 ` Zhang, Qi Z
2018-03-01 7:46 ` Jason Wang
0 siblings, 1 reply; 24+ messages in thread
From: Zhang, Qi Z @ 2018-03-01 4:20 UTC (permalink / raw)
To: Zhang, Qi Z, Jason Wang, dev; +Cc: Karlsson, Magnus, Topel, Bjorn
+Magnus, since a typo in my first batch in email address.
> -----Original Message-----
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Zhang, Qi Z
> Sent: Thursday, March 1, 2018 12:19 PM
> To: Jason Wang <jasowang@redhat.com>; dev@dpdk.org
> Cc: magnus.karlsson@intei.com; Topel, Bjorn <bjorn.topel@intel.com>
> Subject: Re: [dpdk-dev] [RFC 0/7] PMD driver for AF_XDP
>
>
>
> > -----Original Message-----
> > From: Jason Wang [mailto:jasowang@redhat.com]
> > Sent: Thursday, March 1, 2018 10:52 AM
> > To: Zhang, Qi Z <qi.z.zhang@intel.com>; dev@dpdk.org
> > Cc: magnus.karlsson@intei.com; Topel, Bjorn <bjorn.topel@intel.com>
> > Subject: Re: [dpdk-dev] [RFC 0/7] PMD driver for AF_XDP
> >
> >
> >
> > On 2018年02月27日 17:32, Qi Zhang wrote:
> > > The RFC patches add a new PMD driver for AF_XDP which is a proposed
> > > faster version of AF_PACKET interface in Linux, see below link for
> > > detail AF_XDP introduction:
> > > https://fosdem.org/2018/schedule/event/af_xdp/
> > > https://lwn.net/Articles/745934/
> > >
> > > This patchset is base on v18.02.
> > > It also require a linux kernel that have below AF_XDP RFC patches be
> > > applied.
> > > https://patchwork.ozlabs.org/patch/867961/
> > > https://patchwork.ozlabs.org/patch/867960/
> > > https://patchwork.ozlabs.org/patch/867938/
> > > https://patchwork.ozlabs.org/patch/867939/
> > > https://patchwork.ozlabs.org/patch/867940/
> > > https://patchwork.ozlabs.org/patch/867941/
> > > https://patchwork.ozlabs.org/patch/867942/
> > > https://patchwork.ozlabs.org/patch/867943/
> > > https://patchwork.ozlabs.org/patch/867944/
> > > https://patchwork.ozlabs.org/patch/867945/
> > > https://patchwork.ozlabs.org/patch/867946/
> > > https://patchwork.ozlabs.org/patch/867947/
> > > https://patchwork.ozlabs.org/patch/867948/
> > > https://patchwork.ozlabs.org/patch/867949/
> > > https://patchwork.ozlabs.org/patch/867950/
> > > https://patchwork.ozlabs.org/patch/867951/
> > > https://patchwork.ozlabs.org/patch/867952/
> > > https://patchwork.ozlabs.org/patch/867953/
> > > https://patchwork.ozlabs.org/patch/867954/
> > > https://patchwork.ozlabs.org/patch/867955/
> > > https://patchwork.ozlabs.org/patch/867956/
> > > https://patchwork.ozlabs.org/patch/867957/
> > > https://patchwork.ozlabs.org/patch/867958/
> > > https://patchwork.ozlabs.org/patch/867959/
> > >
> > > There is no clean upstream target yet since kernel patch is still in
> > > RFC stage, The purpose of the patchset is just for anyone that want
> > > to eveluate af_xdp with DPDK application and get feedback for
> > > further improvement.
> > >
> > > To try with the new PMD
> > > 1. compile and install the kernel with above patches applied.
> > > 2. configure $LINUX_HEADER_DIR (dir of "make headers_install")
> > > and $TOOLS_DIR (dir at <kernel_src>/tools) at
> > driver/net/af_xdp/Makefile
> > > before compile DPDK.
> > > 3. make sure libelf and libbpf is installed.
> > >
> > > BTW, performance test shows our PMD can reach 94%~98% of the orignal
> > > benchmark when share memory is enabled.
> >
> > Hi:
> >
> > Looks like zero copy is not used in this series. Any plan to support that?
>
> Zero copy is enabled in patch 5, if a mempool passed check_mempool, it will
> be registered to af_xdp socket.
> so there will be no memcpy between mbuf and af_xdp.
>
> > If not, what's the advantage compared to vhost-net + tap + XDP_REDIRECT?
> >
> > Have you measured l2fwd performance in this case? I believe the number
> > you refer here is rxdrop (XDP_DRV) which is 11.6Mpps.
>
> Actually we measure the performance on rxonly / txonly / l2fwd on i40e with
> XDP_SKB and XDP_DRV_ZC
>
> Regards
> Qi
>
> >
> > Thanks
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [RFC 0/7] PMD driver for AF_XDP
2018-03-01 4:20 ` Zhang, Qi Z
@ 2018-03-01 7:46 ` Jason Wang
2018-03-01 12:56 ` Zhang, Qi Z
0 siblings, 1 reply; 24+ messages in thread
From: Jason Wang @ 2018-03-01 7:46 UTC (permalink / raw)
To: Zhang, Qi Z, dev; +Cc: Karlsson, Magnus, Topel, Bjorn
On 2018年03月01日 12:20, Zhang, Qi Z wrote:
> +Magnus, since a typo in my first batch in email address.
>
>> -----Original Message-----
>> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Zhang, Qi Z
>> Sent: Thursday, March 1, 2018 12:19 PM
>> To: Jason Wang<jasowang@redhat.com>;dev@dpdk.org
>> Cc:magnus.karlsson@intei.com; Topel, Bjorn<bjorn.topel@intel.com>
>> Subject: Re: [dpdk-dev] [RFC 0/7] PMD driver for AF_XDP
>>
>>
>>
>>> -----Original Message-----
>>> From: Jason Wang [mailto:jasowang@redhat.com]
>>> Sent: Thursday, March 1, 2018 10:52 AM
>>> To: Zhang, Qi Z<qi.z.zhang@intel.com>;dev@dpdk.org
>>> Cc:magnus.karlsson@intei.com; Topel, Bjorn<bjorn.topel@intel.com>
>>> Subject: Re: [dpdk-dev] [RFC 0/7] PMD driver for AF_XDP
>>>
>>>
>>>
>>> On 2018年02月27日 17:32, Qi Zhang wrote:
>>>> The RFC patches add a new PMD driver for AF_XDP which is a proposed
>>>> faster version of AF_PACKET interface in Linux, see below link for
>>>> detail AF_XDP introduction:
>>>> https://fosdem.org/2018/schedule/event/af_xdp/
>>>> https://lwn.net/Articles/745934/
>>>>
>>>> This patchset is base on v18.02.
>>>> It also require a linux kernel that have below AF_XDP RFC patches be
>>>> applied.
>>>> https://patchwork.ozlabs.org/patch/867961/
>>>> https://patchwork.ozlabs.org/patch/867960/
>>>> https://patchwork.ozlabs.org/patch/867938/
>>>> https://patchwork.ozlabs.org/patch/867939/
>>>> https://patchwork.ozlabs.org/patch/867940/
>>>> https://patchwork.ozlabs.org/patch/867941/
>>>> https://patchwork.ozlabs.org/patch/867942/
>>>> https://patchwork.ozlabs.org/patch/867943/
>>>> https://patchwork.ozlabs.org/patch/867944/
>>>> https://patchwork.ozlabs.org/patch/867945/
>>>> https://patchwork.ozlabs.org/patch/867946/
>>>> https://patchwork.ozlabs.org/patch/867947/
>>>> https://patchwork.ozlabs.org/patch/867948/
>>>> https://patchwork.ozlabs.org/patch/867949/
>>>> https://patchwork.ozlabs.org/patch/867950/
>>>> https://patchwork.ozlabs.org/patch/867951/
>>>> https://patchwork.ozlabs.org/patch/867952/
>>>> https://patchwork.ozlabs.org/patch/867953/
>>>> https://patchwork.ozlabs.org/patch/867954/
>>>> https://patchwork.ozlabs.org/patch/867955/
>>>> https://patchwork.ozlabs.org/patch/867956/
>>>> https://patchwork.ozlabs.org/patch/867957/
>>>> https://patchwork.ozlabs.org/patch/867958/
>>>> https://patchwork.ozlabs.org/patch/867959/
>>>>
>>>> There is no clean upstream target yet since kernel patch is still in
>>>> RFC stage, The purpose of the patchset is just for anyone that want
>>>> to eveluate af_xdp with DPDK application and get feedback for
>>>> further improvement.
>>>>
>>>> To try with the new PMD
>>>> 1. compile and install the kernel with above patches applied.
>>>> 2. configure $LINUX_HEADER_DIR (dir of "make headers_install")
>>>> and $TOOLS_DIR (dir at <kernel_src>/tools) at
>>> driver/net/af_xdp/Makefile
>>>> before compile DPDK.
>>>> 3. make sure libelf and libbpf is installed.
>>>>
>>>> BTW, performance test shows our PMD can reach 94%~98% of the orignal
>>>> benchmark when share memory is enabled.
>>> Hi:
>>>
>>> Looks like zero copy is not used in this series. Any plan to support that?
>> Zero copy is enabled in patch 5, if a mempool passed check_mempool, it will
>> be registered to af_xdp socket.
>> so there will be no memcpy between mbuf and af_xdp.
Aha, I see. So the zerocopy was limited to some specific use case. And
if I understand it correctly, zc mode could not be used for VM.
Thanks
>>> If not, what's the advantage compared to vhost-net + tap + XDP_REDIRECT?
>>>
>>> Have you measured l2fwd performance in this case? I believe the number
>>> you refer here is rxdrop (XDP_DRV) which is 11.6Mpps.
>> Actually we measure the performance on rxonly / txonly / l2fwd on i40e with
>> XDP_SKB and XDP_DRV_ZC
>>
>> Regards
>> Qi
>>
>>> Thanks
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [RFC 0/7] PMD driver for AF_XDP
2018-03-01 7:46 ` Jason Wang
@ 2018-03-01 12:56 ` Zhang, Qi Z
2018-03-01 13:18 ` Jason Wang
0 siblings, 1 reply; 24+ messages in thread
From: Zhang, Qi Z @ 2018-03-01 12:56 UTC (permalink / raw)
To: 'Jason Wang'; +Cc: Karlsson, Magnus, Topel, Bjorn, dev
> -----Original Message-----
> From: Jason Wang [mailto:jasowang@redhat.com]
> Sent: Thursday, March 1, 2018 3:46 PM
> To: Zhang, Qi Z <qi.z.zhang@intel.com>; dev@dpdk.org
> Cc: Karlsson, Magnus <magnus.karlsson@intel.com>; Topel, Bjorn
> <bjorn.topel@intel.com>
> Subject: Re: [dpdk-dev] [RFC 0/7] PMD driver for AF_XDP
>
>
>
> On 2018年03月01日 12:20, Zhang, Qi Z wrote:
> > +Magnus, since a typo in my first batch in email address.
> >
> >> -----Original Message-----
> >> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Zhang, Qi Z
> >> Sent: Thursday, March 1, 2018 12:19 PM
> >> To: Jason Wang<jasowang@redhat.com>;dev@dpdk.org
> >> Cc:magnus.karlsson@intei.com; Topel, Bjorn<bjorn.topel@intel.com>
> >> Subject: Re: [dpdk-dev] [RFC 0/7] PMD driver for AF_XDP
> >>
> >>
> >>
> >>> -----Original Message-----
> >>> From: Jason Wang [mailto:jasowang@redhat.com]
> >>> Sent: Thursday, March 1, 2018 10:52 AM
> >>> To: Zhang, Qi Z<qi.z.zhang@intel.com>;dev@dpdk.org
> >>> Cc:magnus.karlsson@intei.com; Topel, Bjorn<bjorn.topel@intel.com>
> >>> Subject: Re: [dpdk-dev] [RFC 0/7] PMD driver for AF_XDP
> >>>
> >>>
> >>>
> >>> On 2018年02月27日 17:32, Qi Zhang wrote:
> >>>> The RFC patches add a new PMD driver for AF_XDP which is a proposed
> >>>> faster version of AF_PACKET interface in Linux, see below link for
> >>>> detail AF_XDP introduction:
> >>>> https://fosdem.org/2018/schedule/event/af_xdp/
> >>>> https://lwn.net/Articles/745934/
> >>>>
> >>>> This patchset is base on v18.02.
> >>>> It also require a linux kernel that have below AF_XDP RFC patches
> >>>> be applied.
> >>>> https://patchwork.ozlabs.org/patch/867961/
> >>>> https://patchwork.ozlabs.org/patch/867960/
> >>>> https://patchwork.ozlabs.org/patch/867938/
> >>>> https://patchwork.ozlabs.org/patch/867939/
> >>>> https://patchwork.ozlabs.org/patch/867940/
> >>>> https://patchwork.ozlabs.org/patch/867941/
> >>>> https://patchwork.ozlabs.org/patch/867942/
> >>>> https://patchwork.ozlabs.org/patch/867943/
> >>>> https://patchwork.ozlabs.org/patch/867944/
> >>>> https://patchwork.ozlabs.org/patch/867945/
> >>>> https://patchwork.ozlabs.org/patch/867946/
> >>>> https://patchwork.ozlabs.org/patch/867947/
> >>>> https://patchwork.ozlabs.org/patch/867948/
> >>>> https://patchwork.ozlabs.org/patch/867949/
> >>>> https://patchwork.ozlabs.org/patch/867950/
> >>>> https://patchwork.ozlabs.org/patch/867951/
> >>>> https://patchwork.ozlabs.org/patch/867952/
> >>>> https://patchwork.ozlabs.org/patch/867953/
> >>>> https://patchwork.ozlabs.org/patch/867954/
> >>>> https://patchwork.ozlabs.org/patch/867955/
> >>>> https://patchwork.ozlabs.org/patch/867956/
> >>>> https://patchwork.ozlabs.org/patch/867957/
> >>>> https://patchwork.ozlabs.org/patch/867958/
> >>>> https://patchwork.ozlabs.org/patch/867959/
> >>>>
> >>>> There is no clean upstream target yet since kernel patch is still
> >>>> in RFC stage, The purpose of the patchset is just for anyone that
> >>>> want to eveluate af_xdp with DPDK application and get feedback for
> >>>> further improvement.
> >>>>
> >>>> To try with the new PMD
> >>>> 1. compile and install the kernel with above patches applied.
> >>>> 2. configure $LINUX_HEADER_DIR (dir of "make headers_install")
> >>>> and $TOOLS_DIR (dir at <kernel_src>/tools) at
> >>> driver/net/af_xdp/Makefile
> >>>> before compile DPDK.
> >>>> 3. make sure libelf and libbpf is installed.
> >>>>
> >>>> BTW, performance test shows our PMD can reach 94%~98% of the
> >>>> orignal benchmark when share memory is enabled.
> >>> Hi:
> >>>
> >>> Looks like zero copy is not used in this series. Any plan to support that?
> >> Zero copy is enabled in patch 5, if a mempool passed check_mempool,
> >> it will be registered to af_xdp socket.
> >> so there will be no memcpy between mbuf and af_xdp.
>
> Aha, I see. So the zerocopy was limited to some specific use case. And if I
> understand it correctly, zc mode could not be used for VM.
I think except the limitation for mempool layout, zerocopy is transparent to DPDK application, only difference is performance.
Sorry, I may not get your point, if you could explain more about the VM usage.
Regards
Qi
>
> Thanks
>
> >>> If not, what's the advantage compared to vhost-net + tap +
> XDP_REDIRECT?
> >>>
> >>> Have you measured l2fwd performance in this case? I believe the
> >>> number you refer here is rxdrop (XDP_DRV) which is 11.6Mpps.
> >> Actually we measure the performance on rxonly / txonly / l2fwd on
> >> i40e with XDP_SKB and XDP_DRV_ZC
> >>
> >> Regards
> >> Qi
> >>
> >>> Thanks
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [RFC 0/7] PMD driver for AF_XDP
2018-03-01 12:56 ` Zhang, Qi Z
@ 2018-03-01 13:18 ` Jason Wang
2018-03-02 4:05 ` Zhang, Qi Z
0 siblings, 1 reply; 24+ messages in thread
From: Jason Wang @ 2018-03-01 13:18 UTC (permalink / raw)
To: Zhang, Qi Z; +Cc: Karlsson, Magnus, Topel, Bjorn, dev
On 2018年03月01日 20:56, Zhang, Qi Z wrote:
>>>>>> BTW, performance test shows our PMD can reach 94%~98% of the
>>>>>> orignal benchmark when share memory is enabled.
>>>>> Hi:
>>>>>
>>>>> Looks like zero copy is not used in this series. Any plan to support that?
>>>> Zero copy is enabled in patch 5, if a mempool passed check_mempool,
>>>> it will be registered to af_xdp socket.
>>>> so there will be no memcpy between mbuf and af_xdp.
>> Aha, I see. So the zerocopy was limited to some specific use case. And if I
>> understand it correctly, zc mode could not be used for VM.
> I think except the limitation for mempool layout, zerocopy is transparent to DPDK application, only difference is performance.
> Sorry, I may not get your point, if you could explain more about the VM usage.
>
> Regards
> Qi
No problem, so the question is:
Can zerocopy be used when using testpmd to foward packets between
vhost-user and AF_XDP socket?
Thanks
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [RFC 0/7] PMD driver for AF_XDP
2018-03-01 13:18 ` Jason Wang
@ 2018-03-02 4:05 ` Zhang, Qi Z
0 siblings, 0 replies; 24+ messages in thread
From: Zhang, Qi Z @ 2018-03-02 4:05 UTC (permalink / raw)
To: Jason Wang; +Cc: Karlsson, Magnus, Topel, Bjorn, dev
> -----Original Message-----
> From: Jason Wang [mailto:jasowang@redhat.com]
> Sent: Thursday, March 1, 2018 9:18 PM
> To: Zhang, Qi Z <qi.z.zhang@intel.com>
> Cc: Karlsson, Magnus <magnus.karlsson@intel.com>; Topel, Bjorn
> <bjorn.topel@intel.com>; dev@dpdk.org
> Subject: Re: [dpdk-dev] [RFC 0/7] PMD driver for AF_XDP
>
>
>
> On 2018年03月01日 20:56, Zhang, Qi Z wrote:
> >>>>>> BTW, performance test shows our PMD can reach 94%~98% of the
> >>>>>> orignal benchmark when share memory is enabled.
> >>>>> Hi:
> >>>>>
> >>>>> Looks like zero copy is not used in this series. Any plan to support
> that?
> >>>> Zero copy is enabled in patch 5, if a mempool passed check_mempool,
> >>>> it will be registered to af_xdp socket.
> >>>> so there will be no memcpy between mbuf and af_xdp.
> >> Aha, I see. So the zerocopy was limited to some specific use case.
> >> And if I understand it correctly, zc mode could not be used for VM.
> > I think except the limitation for mempool layout, zerocopy is transparent
> to DPDK application, only difference is performance.
> > Sorry, I may not get your point, if you could explain more about the VM
> usage.
> >
> > Regards
> > Qi
>
> No problem, so the question is:
>
> Can zerocopy be used when using testpmd to foward packets between
> vhost-user and AF_XDP socket?
I'm not very familiar with vhost-user, but I guess the answer should be same as the case for forward packet between vhost-user and i40e,
(if vhost-user does not have any special requirement for mempool that conflict with af_xdp ZC's requirement)
Regards
Qi
>
> Thanks
^ permalink raw reply [flat|nested] 24+ messages in thread