* [PATCH v3 0/1] XDP program for ip forward
@ 2017-11-01 7:48 Christina Jacob
2017-11-01 7:48 ` [PATCH v3 1/1] xdp: Sample xdp program implementing " Christina Jacob
0 siblings, 1 reply; 3+ messages in thread
From: Christina Jacob @ 2017-11-01 7:48 UTC (permalink / raw)
To: netdev
Cc: linux-kernel, linux-arm-kernel, brouer, Sunil.Goutham,
Christina.Jacob, stephen, ddaney, David.Laight
The patch below implements port to port forwarding through route table and arp
table lookup for ipv4 packets using bpf_redirect helper function and lpm_trie
map. This has an improved performance over the normal kernel stack ip forward.
Implementation details.
-----------------------
The program uses one map each for arp table, route table and packet count.
The number of entries the program can process is limited by the size of the
map used.
In the xdp_router_ipv4_user.c,
initially, the routing table is read and is stored in an lpm trie map.
The arp table is read and stored in an array map There are two netlink sockets
that listens to any change in the route table and arp table.
There are two types of changes to the route table.
1.New
The new entries are added to the lpm_trie with proper key and prefix
length If there is a another entry in the route table with a different
metric(only metric is considered). Then the values are compared and the
one with lowest metric is added to the node.
2.Deletion
On deletion from the route table, The particular node is removed and the
entire route table is again read to check if there is another entry with
a different metric.
This implementation depends on bpf: Implement map_delete_elem for
BPF_MAP_TYPE_LPM_TRIE which is not yet upstreamed.
There are two types of changes to the route table
1.New
The new arp entries are added in the in the array map directly with the
ip address as the key and the destination mac address as the value.
2.Delete
The entry corresponding to the particular ip is deleted from the
arp table map.
Another map is maintained for entries in the route table having 32 bit mask.
such entries can have a corresponding arp entry which if stored together with
the route entry in an array map and can be accessed in O(1) time. Eliminating
the trie lookup and arp lookup.
In the xdp_router_ipv4_kern.c,
The array map for the 32 bit mask entries checked to see if there is a key that
exactly matches with the destination ip. If it has a non zero destination mac
entry then the xdp data is updated accordingly Otherwise a proper route and
arp table lookup is done using the lpm_trie and the arp table array map.
Usage: as ./xdp_router_ipv4 -S <ifindex1...ifindexn> (-S for
generic xdp implementation ifindex- the index of the interface to which
the xdp program has to be attached.) in 4.14-rc3 kernel.
Changes from v1 to v2
---------------------
* As suggested by Jesper Dangaard Brouer
1. Changed the program name to list xdp_router_ipv4
2. Changed the commandline arguments from ifindex list to interface name
Usage : ./xdp_router_ipv4 [-S] <interface name list>
-S for generic xdp implementation
-interface name list is the list of interfaces to which
the xdp program should attach to
* As suggested by Daniel Borkmann
1. Using __builin_memcpy to update source and destination mac in the bpf
kernel program.
2. Started using __be32 in the kernel program to be inline with the data
type used in user program
3. Rectified few style issues.
* Corrected the copyright issue pointed out by David Ahern
* Fixed the bug: The already attached interfaces are not detached from the
xdp program if the program fails to attach to an interface later in the list.
Changes from v2 to v3
---------------------
* As pointed out by Jesper Dangaard Brouer
1. Changed the program name in the cover letter.
2. Changed variable declararions to follow Reverse-xmas tree
rule.
3. Reduced the nesting in code for readability.
4. Fixed bug: incorrect mac address being set for source and
destination mac.
5. Fixed comment style.
* As suggested by Stephen Hemminger
Changed all the bzeros' to memset.
* As suggested by David Laight
removed the signed remainders calculation.
* As suggested by Stephen Hemminger and David Daney
1. Added checks for the ioctl return value.
2. Changed data types to be64 to be sure about the size of the
data type.
3. Verified byte order. Using the mac address from ioctl in
network byte order. not casting to to long data type
anymore.
4. Fixed returning address of local variable.
Christina Jacob (1):
xdp: Sample xdp program implementing ip forward
samples/bpf/Makefile | 4 +
samples/bpf/xdp_router_ipv4_kern.c | 181 ++++++++++
samples/bpf/xdp_router_ipv4_user.c | 657 +++++++++++++++++++++++++++++++++++++
3 files changed, 842 insertions(+)
create mode 100644 samples/bpf/xdp_router_ipv4_kern.c
create mode 100644 samples/bpf/xdp_router_ipv4_user.c
--
2.7.4
^ permalink raw reply [flat|nested] 3+ messages in thread
* [PATCH v3 1/1] xdp: Sample xdp program implementing ip forward
2017-11-01 7:48 [PATCH v3 0/1] XDP program for ip forward Christina Jacob
@ 2017-11-01 7:48 ` Christina Jacob
2017-11-02 11:21 ` Jesper Dangaard Brouer
0 siblings, 1 reply; 3+ messages in thread
From: Christina Jacob @ 2017-11-01 7:48 UTC (permalink / raw)
To: netdev
Cc: linux-kernel, linux-arm-kernel, brouer, Sunil.Goutham,
Christina.Jacob, stephen, ddaney, David.Laight
From: Christina Jacob <Christina.Jacob@cavium.com>
Implements port to port forwarding with route table and arp table
lookup for ipv4 packets using bpf_redirect helper function and
lpm_trie map.
Signed-off-by: Christina Jacob <Christina.Jacob@cavium.com>
---
samples/bpf/Makefile | 4 +
samples/bpf/xdp_router_ipv4_kern.c | 181 ++++++++++
samples/bpf/xdp_router_ipv4_user.c | 657 +++++++++++++++++++++++++++++++++++++
3 files changed, 842 insertions(+)
diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index cf17c79..8504ebb 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -28,6 +28,7 @@ hostprogs-y += test_cgrp2_sock
hostprogs-y += test_cgrp2_sock2
hostprogs-y += xdp1
hostprogs-y += xdp2
+hostprogs-y += xdp_router_ipv4
hostprogs-y += test_current_task_under_cgroup
hostprogs-y += trace_event
hostprogs-y += sampleip
@@ -73,6 +74,7 @@ test_cgrp2_sock2-objs := bpf_load.o $(LIBBPF) test_cgrp2_sock2.o
xdp1-objs := bpf_load.o $(LIBBPF) xdp1_user.o
# reuse xdp1 source intentionally
xdp2-objs := bpf_load.o $(LIBBPF) xdp1_user.o
+xdp_router_ipv4-objs := bpf_load.o $(LIBBPF) xdp_router_ipv4_user.o
test_current_task_under_cgroup-objs := bpf_load.o $(LIBBPF) cgroup_helpers.o \
test_current_task_under_cgroup_user.o
trace_event-objs := bpf_load.o $(LIBBPF) trace_event_user.o
@@ -114,6 +116,7 @@ always += parse_varlen.o parse_simple.o parse_ldabs.o
always += test_cgrp2_tc_kern.o
always += xdp1_kern.o
always += xdp2_kern.o
+always += xdp_router_ipv4_kern.o
always += test_current_task_under_cgroup_kern.o
always += trace_event_kern.o
always += sampleip_kern.o
@@ -160,6 +163,7 @@ HOSTLOADLIBES_map_perf_test += -lelf -lrt
HOSTLOADLIBES_test_overhead += -lelf -lrt
HOSTLOADLIBES_xdp1 += -lelf
HOSTLOADLIBES_xdp2 += -lelf
+HOSTLOADLIBES_xdp_router_ipv4 += -lelf
HOSTLOADLIBES_test_current_task_under_cgroup += -lelf
HOSTLOADLIBES_trace_event += -lelf
HOSTLOADLIBES_sampleip += -lelf
diff --git a/samples/bpf/xdp_router_ipv4_kern.c b/samples/bpf/xdp_router_ipv4_kern.c
new file mode 100644
index 0000000..70a5907
--- /dev/null
+++ b/samples/bpf/xdp_router_ipv4_kern.c
@@ -0,0 +1,181 @@
+/* Copyright (C) 2017 Cavium, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2 of the GNU General Public License
+ * as published by the Free Software Foundation.
+ */
+#define KBUILD_MODNAME "foo"
+#include <uapi/linux/bpf.h>
+#include <linux/in.h>
+#include <linux/if_ether.h>
+#include <linux/if_packet.h>
+#include <linux/if_vlan.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include "bpf_helpers.h"
+#include <linux/slab.h>
+#include <net/ip_fib.h>
+
+struct trie_value {
+ __u8 prefix[4];
+ __be32 ifindex;
+ __be32 metric;
+ __be64 value;
+ __be32 gw;
+};
+
+/* Key for lpm_trie*/
+union key_4 {
+ u32 b32[2];
+ u8 b8[8];
+};
+
+struct arp_entry {
+ __be64 mac;
+ __be32 dst;
+};
+
+struct direct_map {
+ struct arp_entry arp;
+ int ifindex;
+ __be64 mac;
+};
+
+/* Map for trie implementation*/
+struct bpf_map_def SEC("maps") lpm_map = {
+ .type = BPF_MAP_TYPE_LPM_TRIE,
+ .key_size = 8,
+ .value_size = sizeof(struct trie_value),
+ .max_entries = 50,
+ .map_flags = BPF_F_NO_PREALLOC,
+};
+
+/* Map for counter*/
+struct bpf_map_def SEC("maps") rxcnt = {
+ .type = BPF_MAP_TYPE_PERCPU_ARRAY,
+ .key_size = sizeof(u32),
+ .value_size = sizeof(u64),
+ .max_entries = 256,
+};
+
+/* Map for ARP table*/
+struct bpf_map_def SEC("maps") arp_table = {
+ .type = BPF_MAP_TYPE_HASH,
+ .key_size = sizeof(__be32),
+ .value_size = sizeof(__be64),
+ .max_entries = 50,
+};
+
+/* Map to keep the exact match entries in the route table*/
+struct bpf_map_def SEC("maps") exact_match = {
+ .type = BPF_MAP_TYPE_HASH,
+ .key_size = sizeof(__be32),
+ .value_size = sizeof(struct direct_map),
+ .max_entries = 50,
+};
+
+/* Function to set source and destination mac of the packet */
+static inline void set_src_dst_mac(void *data, void *src, void *dst)
+{
+ unsigned short *source = src;
+ unsigned short *dest = dst;
+ unsigned short *p = data;
+
+ __builtin_memcpy(p, dest, 6);
+ __builtin_memcpy(p + 3, source, 6);
+}
+
+/* Parse IPV4 packet to get SRC, DST IP and protocol */
+static inline int parse_ipv4(void *data, u64 nh_off, void *data_end,
+ __be32 *src, __be32 *dest)
+{
+ struct iphdr *iph = data + nh_off;
+
+ if (iph + 1 > data_end)
+ return 0;
+ *src = iph->saddr;
+ *dest = iph->daddr;
+ return iph->protocol;
+}
+
+SEC("xdp3")
+int xdp_prog3(struct xdp_md *ctx)
+{
+ void *data_end = (void *)(long)ctx->data_end;
+ __be64 *dest_mac = NULL, *src_mac = NULL;
+ void *data = (void *)(long)ctx->data;
+ struct trie_value *prefix_value;
+ int rc = XDP_DROP, forward_to;
+ struct ethhdr *eth = data;
+ union key_4 key4;
+ long *value;
+ u16 h_proto;
+ u32 ipproto;
+ u64 nh_off;
+
+ nh_off = sizeof(*eth);
+ if (data + nh_off > data_end)
+ return rc;
+
+ h_proto = eth->h_proto;
+
+ if (h_proto == htons(ETH_P_8021Q) || h_proto == htons(ETH_P_8021AD)) {
+ struct vlan_hdr *vhdr;
+
+ vhdr = data + nh_off;
+ nh_off += sizeof(struct vlan_hdr);
+ if (data + nh_off > data_end)
+ return rc;
+ h_proto = vhdr->h_vlan_encapsulated_proto;
+ }
+ if (h_proto == htons(ETH_P_ARP)) {
+ return XDP_PASS;
+ } else if (h_proto == htons(ETH_P_IP)) {
+ struct direct_map *direct_entry;
+ __be32 src_ip = 0, dest_ip = 0;
+
+ ipproto = parse_ipv4(data, nh_off, data_end, &src_ip, &dest_ip);
+ direct_entry = (struct direct_map *)bpf_map_lookup_elem
+ (&exact_match, &dest_ip);
+ /* Check for exact match, this would give a faster lookup*/
+ if (direct_entry && direct_entry->mac && direct_entry->arp.mac) {
+ src_mac = &direct_entry->mac;
+ dest_mac = &direct_entry->arp.mac;
+ forward_to = direct_entry->ifindex;
+ } else {
+ /* Look up in the trie for lpm*/
+ key4.b32[0] = 32;
+ key4.b8[4] = dest_ip & 0xff;
+ key4.b8[5] = (dest_ip >> 8) & 0xff;
+ key4.b8[6] = (dest_ip >> 16) & 0xff;
+ key4.b8[7] = (dest_ip >> 24) & 0xff;
+ prefix_value = ((struct trie_value *)bpf_map_lookup_elem
+ (&lpm_map, &key4));
+ if (!prefix_value)
+ return XDP_DROP;
+ src_mac = &prefix_value->value;
+ if (!src_mac)
+ return XDP_DROP;
+ dest_mac = (__be64 *)bpf_map_lookup_elem(&arp_table, &dest_ip);
+ if (!dest_mac) {
+ if (!prefix_value->gw)
+ return XDP_DROP;
+ dest_ip = *(__be32 *)&prefix_value->gw;
+ dest_mac = (__be64 *)bpf_map_lookup_elem(&arp_table, &dest_ip);
+ }
+ forward_to = prefix_value->ifindex;
+ }
+ } else {
+ ipproto = 0;
+ }
+ if (src_mac && dest_mac) {
+ set_src_dst_mac(data, src_mac, dest_mac);
+ value = bpf_map_lookup_elem(&rxcnt, &ipproto);
+ if (value)
+ *value += 1;
+ return bpf_redirect(forward_to, 0);
+ }
+ return rc;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/samples/bpf/xdp_router_ipv4_user.c b/samples/bpf/xdp_router_ipv4_user.c
new file mode 100644
index 0000000..c3e6180
--- /dev/null
+++ b/samples/bpf/xdp_router_ipv4_user.c
@@ -0,0 +1,657 @@
+/* Copyright (C) 2017 Cavium, Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2 of the GNU General Public License
+ * as published by the Free Software Foundation.
+ */
+#include <linux/bpf.h>
+#include <linux/netlink.h>
+#include <linux/rtnetlink.h>
+#include <assert.h>
+#include <errno.h>
+#include <signal.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/socket.h>
+#include <unistd.h>
+#include "bpf_load.h"
+#include "libbpf.h"
+#include <arpa/inet.h>
+#include <fcntl.h>
+#include <poll.h>
+#include <net/if.h>
+#include <netdb.h>
+#include <sys/ioctl.h>
+#include <sys/syscall.h>
+#include "bpf_util.h"
+
+int sock, sock_arp, flags = 0;
+static int total_ifindex;
+int *ifindex_list;
+char buf[8192];
+
+static int get_route_table(int rtm_family);
+static void int_exit(int sig)
+{
+ int i = 0;
+
+ for (i = 0; i < total_ifindex; i++)
+ set_link_xdp_fd(ifindex_list[i], -1, flags);
+ exit(0);
+}
+
+static void close_and_exit(int sig)
+{
+ int i = 0;
+
+ close(sock);
+ close(sock_arp);
+
+ for (i = 0; i < total_ifindex; i++)
+ set_link_xdp_fd(ifindex_list[i], -1, flags);
+ exit(0);
+}
+
+/* Get the mac address of the interface given interface name */
+static __be64 getmac(char *iface)
+{
+ struct ifreq ifr;
+ __be64 mac = 0;
+ int fd, i;
+
+ fd = socket(AF_INET, SOCK_DGRAM, 0);
+ ifr.ifr_addr.sa_family = AF_INET;
+ strncpy(ifr.ifr_name, iface, IFNAMSIZ - 1);
+ if (ioctl(fd, SIOCGIFHWADDR, &ifr) < 0) {
+ printf("ioctl failed leaving....\n");
+ return -1;
+ }
+ for (i = 0; i < 6 ; i++)
+ *((__u8 *)&mac + i) = (__u8)ifr.ifr_hwaddr.sa_data[i];
+ close(fd);
+ return mac;
+}
+
+static int recv_msg(struct sockaddr_nl sock_addr, int sock)
+{
+ struct nlmsghdr *nh;
+ int len, nll = 0;
+ char *buf_ptr;
+
+ buf_ptr = buf;
+ while (1) {
+ len = recv(sock, buf_ptr, sizeof(buf) - nll, 0);
+ if (len < 0)
+ return len;
+
+ nh = (struct nlmsghdr *)buf_ptr;
+
+ if (nh->nlmsg_type == NLMSG_DONE)
+ break;
+ buf_ptr += len;
+ nll += len;
+ if ((sock_addr.nl_groups & RTMGRP_NEIGH) == RTMGRP_NEIGH)
+ break;
+
+ if ((sock_addr.nl_groups & RTMGRP_IPV4_ROUTE) == RTMGRP_IPV4_ROUTE)
+ break;
+ }
+ return nll;
+}
+
+/* Function to parse the route entry returned by netlink
+ * Updates the route entry related map entries
+ */
+static void read_route(struct nlmsghdr *nh, int nll)
+{
+ char dsts[24], gws[24], ifs[16], dsts_len[24], metrics[24];
+ struct bpf_lpm_trie_key *prefix_key;
+ struct rtattr *rt_attr;
+ struct rtmsg *rt_msg;
+ int rtm_family;
+ int rtl;
+ int i;
+ struct route_table {
+ __be32 dst, gw, dst_len, iface, metric;
+ char *iface_name;
+ __be64 mac;
+ } route;
+ struct arp_table {
+ __be64 mac;
+ __be32 dst;
+ };
+
+ struct direct_map {
+ struct arp_table arp;
+ int ifindex;
+ __be64 mac;
+ } direct_entry;
+
+ if (nh->nlmsg_type == RTM_DELROUTE)
+ printf("DELETING Route entry\n");
+ else if (nh->nlmsg_type == RTM_GETROUTE)
+ printf("READING Route entry\n");
+ else if (nh->nlmsg_type == RTM_NEWROUTE)
+ printf("NEW Route entry\n");
+ else
+ printf("%d\n", nh->nlmsg_type);
+
+ memset(&route, 0, sizeof(route));
+ printf("Destination\t\tGateway\t\tGenmask\t\tMetric\t\tIface\n");
+ for (; NLMSG_OK(nh, nll); nh = NLMSG_NEXT(nh, nll)) {
+ rt_msg = (struct rtmsg *)NLMSG_DATA(nh);
+ rtm_family = rt_msg->rtm_family;
+ if (rtm_family == AF_INET)
+ if (rt_msg->rtm_table != RT_TABLE_MAIN)
+ continue;
+ rt_attr = (struct rtattr *)RTM_RTA(rt_msg);
+ rtl = RTM_PAYLOAD(nh);
+
+ for (; RTA_OK(rt_attr, rtl); rt_attr = RTA_NEXT(rt_attr, rtl)) {
+ switch (rt_attr->rta_type) {
+ case NDA_DST:
+ sprintf(dsts, "%u",
+ (*((__be32 *)RTA_DATA(rt_attr))));
+ break;
+ case RTA_GATEWAY:
+ sprintf(gws, "%u",
+ *((__be32 *)RTA_DATA(rt_attr)));
+ break;
+ case RTA_OIF:
+ sprintf(ifs, "%u",
+ *((__be32 *)RTA_DATA(rt_attr)));
+ break;
+ case RTA_METRICS:
+ sprintf(metrics, "%u",
+ *((__be32 *)RTA_DATA(rt_attr)));
+ default:
+ break;
+ }
+ }
+ sprintf(dsts_len, "%d", rt_msg->rtm_dst_len);
+ route.dst = atoi(dsts);
+ route.dst_len = atoi(dsts_len);
+ route.gw = atoi(gws);
+ route.iface = atoi(ifs);
+ route.metric = atoi(metrics);
+ route.iface_name = alloca(sizeof(char *) * IFNAMSIZ);
+ route.iface_name = if_indextoname(route.iface, route.iface_name);
+ route.mac = getmac(route.iface_name);
+ if (route.mac == -1) {
+ int i = 0;
+
+ for (i = 0; i < total_ifindex; i++)
+ set_link_xdp_fd(ifindex_list[i], -1, flags);
+ exit(0);
+ }
+ if (rtm_family == AF_INET) {
+ struct trie_value {
+ __u8 prefix[4];
+ __be32 ifindex;
+ __be32 metric;
+ __be64 value;
+ __be32 gw;
+ } *prefix_value;
+
+ prefix_key = alloca(sizeof(*prefix_key) + 3);
+ prefix_value = alloca(sizeof(*prefix_value));
+
+ prefix_key->prefixlen = 32;
+ prefix_key->prefixlen = route.dst_len;
+ direct_entry.mac = route.mac & 0xffffffffffff;
+ direct_entry.ifindex = route.iface;
+ direct_entry.arp.mac = 0;
+ direct_entry.arp.dst = 0;
+ if (route.dst_len == 32) {
+ if (nh->nlmsg_type == RTM_DELROUTE)
+ (bpf_map_delete_elem(map_fd[3], &route.dst) == 0);
+ else
+ if (bpf_map_lookup_elem(map_fd[2], &route.dst, &direct_entry.arp.mac) == 0)
+ direct_entry.arp.dst = route.dst;
+ assert(bpf_map_update_elem(map_fd[3], &route.dst, &direct_entry, 0) == 0);
+ }
+ for (i = 0; i < 4; i++)
+ prefix_key->data[i] = (route.dst >> i * 8) & 0xff;
+
+ printf("%3d.%d.%d.%d\t\t%3x\t\t%d\t\t%d\t\t%s\n",
+ (int)prefix_key->data[0],
+ (int)prefix_key->data[1],
+ (int)prefix_key->data[2],
+ (int)prefix_key->data[3],
+ route.gw, route.dst_len,
+ route.metric,
+ route.iface_name);
+ if (bpf_map_lookup_elem(map_fd[0], prefix_key,
+ prefix_value) < 0) {
+ for (i = 0; i < 4; i++)
+ prefix_value->prefix[i] = prefix_key->data[i];
+ prefix_value->value = route.mac & 0xffffffffffff;
+ prefix_value->ifindex = route.iface;
+ prefix_value->gw = route.gw;
+ prefix_value->metric = route.metric;
+
+ assert(bpf_map_update_elem(map_fd[0],
+ prefix_key,
+ prefix_value, 0
+ ) == 0);
+ } else {
+ if (nh->nlmsg_type == RTM_DELROUTE) {
+ printf("deleting entry\n");
+ printf("prefix key=%d.%d.%d.%d/%d",
+ prefix_key->data[0],
+ prefix_key->data[1],
+ prefix_key->data[2],
+ prefix_key->data[3],
+ prefix_key->prefixlen);
+ assert(bpf_map_delete_elem(map_fd[0],
+ prefix_key
+ ) == 0);
+ /* Rereading the route table to check if
+ * there is an entry with the same
+ * prefix but a different metric as the
+ * deleted enty.
+ */
+ get_route_table(AF_INET);
+ } else if (prefix_key->data[0] ==
+ prefix_value->prefix[0] &&
+ prefix_key->data[1] ==
+ prefix_value->prefix[1] &&
+ prefix_key->data[2] ==
+ prefix_value->prefix[2] &&
+ prefix_key->data[3] ==
+ prefix_value->prefix[3] &&
+ route.metric >= prefix_value->metric) {
+ continue;
+ } else {
+ for (i = 0; i < 4; i++)
+ prefix_value->prefix[i] =
+ prefix_key->data[i];
+ prefix_value->value =
+ route.mac & 0xffffffffffff;
+ prefix_value->ifindex = route.iface;
+ prefix_value->gw = route.gw;
+ prefix_value->metric = route.metric;
+ assert(bpf_map_update_elem(
+ map_fd[0],
+ prefix_key,
+ prefix_value,
+ 0) == 0);
+ }
+ }
+ }
+ memset(&route, 0, sizeof(route));
+ memset(dsts, 0, sizeof(dsts));
+ memset(dsts_len, 0, sizeof(dsts_len));
+ memset(gws, 0, sizeof(gws));
+ memset(ifs, 0, sizeof(ifs));
+ memset(&route, 0, sizeof(route));
+ }
+}
+
+/* Function to read the existing route table when the process is launched*/
+static int get_route_table(int rtm_family)
+{
+ struct sockaddr_nl sa;
+ struct nlmsghdr *nh;
+ int sock, seq = 0;
+ struct msghdr msg;
+ struct iovec iov;
+ int ret = 0;
+ int nll;
+
+ struct {
+ struct nlmsghdr nl;
+ struct rtmsg rt;
+ char buf[8192];
+ } req;
+
+ sock = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE);
+ if (sock < 0) {
+ printf("open netlink socket: %s\n", strerror(errno));
+ return -1;
+ }
+ memset(&sa, 0, sizeof(sa));
+ sa.nl_family = AF_NETLINK;
+ if (bind(sock, (struct sockaddr *)&sa, sizeof(sa)) < 0) {
+ printf("bind to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ memset(&req, 0, sizeof(req));
+ req.nl.nlmsg_len = NLMSG_LENGTH(sizeof(struct rtmsg));
+ req.nl.nlmsg_flags = NLM_F_REQUEST | NLM_F_DUMP;
+ req.nl.nlmsg_type = RTM_GETROUTE;
+
+ req.rt.rtm_family = rtm_family;
+ req.rt.rtm_table = RT_TABLE_MAIN;
+ req.nl.nlmsg_pid = 0;
+ req.nl.nlmsg_seq = ++seq;
+ memset(&msg, 0, sizeof(msg));
+ iov.iov_base = (void *)&req.nl;
+ iov.iov_len = req.nl.nlmsg_len;
+ msg.msg_iov = &iov;
+ msg.msg_iovlen = 1;
+ ret = sendmsg(sock, &msg, 0);
+ if (ret < 0) {
+ printf("send to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ memset(buf, 0, sizeof(buf));
+ nll = recv_msg(sa, sock);
+ if (nll < 0) {
+ printf("recv from netlink: %s\n", strerror(nll));
+ ret = -1;
+ goto cleanup;
+ }
+ nh = (struct nlmsghdr *)buf;
+ read_route(nh, nll);
+cleanup:
+ close(sock);
+ return ret;
+}
+
+/* Function to parse the arp entry returned by netlink
+ * Updates the arp entry related map entries
+ */
+static void read_arp(struct nlmsghdr *nh, int nll)
+{
+ struct rtattr *rt_attr;
+ char dsts[24], mac[24];
+ struct ndmsg *rt_msg;
+ int rtl, ndm_family;
+
+ struct arp_table {
+ __be64 mac;
+ __be32 dst;
+ } arp_entry;
+ struct direct_map {
+ struct arp_table arp;
+ int ifindex;
+ __be64 mac;
+ } direct_entry;
+
+ if (nh->nlmsg_type == RTM_GETNEIGH)
+ printf("READING arp entry\n");
+ printf("Address\tHwAddress\n");
+ for (; NLMSG_OK(nh, nll); nh = NLMSG_NEXT(nh, nll)) {
+ rt_msg = (struct ndmsg *)NLMSG_DATA(nh);
+ rt_attr = (struct rtattr *)RTM_RTA(rt_msg);
+ ndm_family = rt_msg->ndm_family;
+ rtl = RTM_PAYLOAD(nh);
+ for (; RTA_OK(rt_attr, rtl); rt_attr = RTA_NEXT(rt_attr, rtl)) {
+ switch (rt_attr->rta_type) {
+ case NDA_DST:
+ sprintf(dsts, "%u",
+ *((__be32 *)RTA_DATA(rt_attr)));
+ break;
+ case NDA_LLADDR:
+ sprintf(mac, "%lld",
+ *((__be64 *)RTA_DATA(rt_attr)));
+ break;
+ default:
+ break;
+ }
+ }
+ arp_entry.dst = atoi(dsts);
+ arp_entry.mac = atol(mac);
+ printf("%x\t\t%llx\n", arp_entry.dst, arp_entry.mac);
+ if (ndm_family == AF_INET) {
+ if (bpf_map_lookup_elem(map_fd[3], &arp_entry.dst,
+ &direct_entry) == 0) {
+ if (nh->nlmsg_type == RTM_DELNEIGH) {
+ direct_entry.arp.dst = 0;
+ direct_entry.arp.mac = 0;
+ } else if (nh->nlmsg_type == RTM_NEWNEIGH) {
+ direct_entry.arp.dst = arp_entry.dst;
+ direct_entry.arp.mac = arp_entry.mac;
+ }
+ assert(bpf_map_update_elem(map_fd[3],
+ &arp_entry.dst,
+ &direct_entry, 0
+ ) == 0);
+ memset(&direct_entry, 0, sizeof(direct_entry));
+ }
+ if (nh->nlmsg_type == RTM_DELNEIGH) {
+ assert(bpf_map_delete_elem(map_fd[2], &arp_entry.dst) == 0);
+ } else if (nh->nlmsg_type == RTM_NEWNEIGH) {
+ assert(bpf_map_update_elem(map_fd[2],
+ &arp_entry.dst,
+ &arp_entry.mac, 0
+ ) == 0);
+ }
+ }
+ memset(&arp_entry, 0, sizeof(arp_entry));
+ memset(dsts, 0, sizeof(dsts));
+ }
+}
+
+/* Function to read the existing arp table when the process is launched*/
+static int get_arp_table(int rtm_family)
+{
+ struct sockaddr_nl sa;
+ struct nlmsghdr *nh;
+ int sock, seq = 0;
+ struct msghdr msg;
+ struct iovec iov;
+ int ret = 0;
+ int nll;
+ struct {
+ struct nlmsghdr nl;
+ struct ndmsg rt;
+ char buf[8192];
+ } req;
+
+ sock = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE);
+ if (sock < 0) {
+ printf("open netlink socket: %s\n", strerror(errno));
+ return -1;
+ }
+ memset(&sa, 0, sizeof(sa));
+ sa.nl_family = AF_NETLINK;
+ if (bind(sock, (struct sockaddr *)&sa, sizeof(sa)) < 0) {
+ printf("bind to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ memset(&req, 0, sizeof(req));
+ req.nl.nlmsg_len = NLMSG_LENGTH(sizeof(struct rtmsg));
+ req.nl.nlmsg_flags = NLM_F_REQUEST | NLM_F_DUMP;
+ req.nl.nlmsg_type = RTM_GETNEIGH;
+ req.rt.ndm_state = NUD_REACHABLE;
+ req.rt.ndm_family = rtm_family;
+ req.nl.nlmsg_pid = 0;
+ req.nl.nlmsg_seq = ++seq;
+ memset(&msg, 0, sizeof(msg));
+ iov.iov_base = (void *)&req.nl;
+ iov.iov_len = req.nl.nlmsg_len;
+ msg.msg_iov = &iov;
+ msg.msg_iovlen = 1;
+ ret = sendmsg(sock, &msg, 0);
+ if (ret < 0) {
+ printf("send to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ memset(buf, 0, sizeof(buf));
+ nll = recv_msg(sa, sock);
+ if (nll < 0) {
+ printf("recv from netlink: %s\n", strerror(nll));
+ ret = -1;
+ goto cleanup;
+ }
+ nh = (struct nlmsghdr *)buf;
+ read_arp(nh, nll);
+cleanup:
+ close(sock);
+ return ret;
+}
+
+/* Function to keep track and update changes in route and arp table
+ * Give regular statistics of packets forwarded
+ */
+static int monitor_route(void)
+{
+ unsigned int nr_cpus = bpf_num_possible_cpus();
+ const unsigned int nr_keys = 256;
+ struct pollfd fds_route, fds_arp;
+ __u64 prev[nr_keys][nr_cpus];
+ struct sockaddr_nl la, lr;
+ __u64 values[nr_cpus];
+ struct nlmsghdr *nh;
+ int nll, ret = 0;
+ int interval = 5;
+ __u32 key;
+ int i;
+
+ sock = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE);
+ if (sock < 0) {
+ printf("open netlink socket: %s\n", strerror(errno));
+ return -1;
+ }
+
+ fcntl(sock, F_SETFL, O_NONBLOCK);
+ memset(&lr, 0, sizeof(lr));
+ lr.nl_family = AF_NETLINK;
+ lr.nl_groups = RTMGRP_IPV6_ROUTE | RTMGRP_IPV4_ROUTE | RTMGRP_NOTIFY;
+ if (bind(sock, (struct sockaddr *)&lr, sizeof(lr)) < 0) {
+ printf("bind to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ fds_route.fd = sock;
+ fds_route.events = POLL_IN;
+
+ sock_arp = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE);
+ if (sock_arp < 0) {
+ printf("open netlink socket: %s\n", strerror(errno));
+ return -1;
+ }
+
+ fcntl(sock_arp, F_SETFL, O_NONBLOCK);
+ memset(&la, 0, sizeof(la));
+ la.nl_family = AF_NETLINK;
+ la.nl_groups = RTMGRP_NEIGH | RTMGRP_NOTIFY;
+ if (bind(sock_arp, (struct sockaddr *)&la, sizeof(la)) < 0) {
+ printf("bind to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ fds_arp.fd = sock_arp;
+ fds_arp.events = POLL_IN;
+
+ memset(prev, 0, sizeof(prev));
+ do {
+ signal(SIGINT, close_and_exit);
+ signal(SIGTERM, close_and_exit);
+
+ sleep(interval);
+ for (key = 0; key < nr_keys; key++) {
+ __u64 sum = 0;
+
+ assert(bpf_map_lookup_elem(map_fd[1], &key, values) == 0);
+ for (i = 0; i < nr_cpus; i++)
+ sum += (values[i] - prev[key][i]);
+ if (sum)
+ printf("proto %u: %10llu pkt/s\n",
+ key, sum / interval);
+ memcpy(prev[key], values, sizeof(values));
+ }
+
+ memset(buf, 0, sizeof(buf));
+ if (poll(&fds_route, 1, 3) == POLL_IN) {
+ nll = recv_msg(lr, sock);
+ if (nll < 0) {
+ printf("recv from netlink: %s\n", strerror(nll));
+ ret = -1;
+ goto cleanup;
+ }
+
+ nh = (struct nlmsghdr *)buf;
+ printf("Routing table updated.\n");
+ read_route(nh, nll);
+ }
+ memset(buf, 0, sizeof(buf));
+ if (poll(&fds_arp, 1, 3) == POLL_IN) {
+ nll = recv_msg(la, sock_arp);
+ if (nll < 0) {
+ printf("recv from netlink: %s\n", strerror(nll));
+ ret = -1;
+ goto cleanup;
+ }
+
+ nh = (struct nlmsghdr *)buf;
+ read_arp(nh, nll);
+ }
+
+ } while (1);
+cleanup:
+ close(sock);
+ return ret;
+}
+
+int main(int ac, char **argv)
+{
+ char filename[256];
+ char **ifname_list;
+ int i = 1;
+
+ snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
+ if (ac < 2) {
+ printf("usage: %s [-S] Interface name list\n", argv[0]);
+ return 1;
+ }
+ if (!strcmp(argv[1], "-S")) {
+ flags = XDP_FLAGS_SKB_MODE;
+ total_ifindex = ac - 2;
+ ifname_list = (argv + 2);
+ } else {
+ flags = 0;
+ total_ifindex = ac - 1;
+ ifname_list = (argv + 1);
+ }
+ if (load_bpf_file(filename)) {
+ printf("%s", bpf_log_buf);
+ return 1;
+ }
+ printf("\n**************loading bpf file*********************\n\n\n");
+ if (!prog_fd[0]) {
+ printf("load_bpf_file: %s\n", strerror(errno));
+ return 1;
+ }
+ ifindex_list = (int *)malloc(total_ifindex * sizeof(int *));
+ for (i = 0; i < total_ifindex; i++) {
+ ifindex_list[i] = if_nametoindex(ifname_list[i]);
+ if (!ifindex_list[i]) {
+ printf("Couldn't translate interface name: %s",
+ strerror(errno));
+ return 1;
+ }
+ }
+ for (i = 0; i < total_ifindex; i++) {
+ if (set_link_xdp_fd(ifindex_list[i], prog_fd[0], flags) < 0) {
+ printf("link set xdp fd failed\n");
+ int recovery_index = i;
+
+ for (i = 0; i < recovery_index; i++)
+ set_link_xdp_fd(ifindex_list[i], -1, flags);
+
+ return 1;
+ }
+ printf("Attached to %d\n", ifindex_list[i]);
+ }
+ signal(SIGINT, int_exit);
+ signal(SIGTERM, int_exit);
+
+ printf("*******************ROUTE TABLE*************************\n\n\n");
+ get_route_table(AF_INET);
+ printf("*******************ARP TABLE***************************\n\n\n");
+ get_arp_table(AF_INET);
+ if (monitor_route() < 0) {
+ printf("Error in receiving route update");
+ return 1;
+ }
+
+ return 0;
+}
--
2.7.4
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH v3 1/1] xdp: Sample xdp program implementing ip forward
2017-11-01 7:48 ` [PATCH v3 1/1] xdp: Sample xdp program implementing " Christina Jacob
@ 2017-11-02 11:21 ` Jesper Dangaard Brouer
0 siblings, 0 replies; 3+ messages in thread
From: Jesper Dangaard Brouer @ 2017-11-02 11:21 UTC (permalink / raw)
To: Christina Jacob
Cc: netdev, linux-kernel, linux-arm-kernel, Sunil.Goutham,
Christina.Jacob, stephen, ddaney, David.Laight, brouer
On Wed, 1 Nov 2017 13:18:04 +0530 Christina Jacob <christina.jacob.koikara@gmail.com> wrote:
> From: Christina Jacob <Christina.Jacob@cavium.com>
>
> Implements port to port forwarding with route table and arp table
> lookup for ipv4 packets using bpf_redirect helper function and
> lpm_trie map.
> Signed-off-by: Christina Jacob <Christina.Jacob@cavium.com>
There is usually a line between the desc and Signed-off-by.
> ---
> samples/bpf/Makefile | 4 +
> samples/bpf/xdp_router_ipv4_kern.c | 181 ++++++++++
> samples/bpf/xdp_router_ipv4_user.c | 657 +++++++++++++++++++++++++++++++++++++
> 3 files changed, 842 insertions(+)
>
[...]
> diff --git a/samples/bpf/xdp_router_ipv4_kern.c b/samples/bpf/xdp_router_ipv4_kern.c
> new file mode 100644
> index 0000000..70a5907
> --- /dev/null
> +++ b/samples/bpf/xdp_router_ipv4_kern.c
> @@ -0,0 +1,181 @@
> +/* Copyright (C) 2017 Cavium, Inc.
> + *
> + * This program is free software; you can redistribute it and/or modify it
> + * under the terms of version 2 of the GNU General Public License
> + * as published by the Free Software Foundation.
> + */
[...]
> +SEC("xdp3")
> +int xdp_prog3(struct xdp_md *ctx)
You changed the filename from xdp3 to xdp_router_ipv4, but you didn't
change the name in he code.
> +{
> + void *data_end = (void *)(long)ctx->data_end;
> + __be64 *dest_mac = NULL, *src_mac = NULL;
> + void *data = (void *)(long)ctx->data;
> + struct trie_value *prefix_value;
> + int rc = XDP_DROP, forward_to;
> + struct ethhdr *eth = data;
> + union key_4 key4;
> + long *value;
> + u16 h_proto;
> + u32 ipproto;
> + u64 nh_off;
> +
[..]
> + if (h_proto == htons(ETH_P_ARP)) {
> + return XDP_PASS;
> + } else if (h_proto == htons(ETH_P_IP)) {
> + struct direct_map *direct_entry;
> + __be32 src_ip = 0, dest_ip = 0;
> +
> + ipproto = parse_ipv4(data, nh_off, data_end, &src_ip, &dest_ip);
> + direct_entry = (struct direct_map *)bpf_map_lookup_elem
> + (&exact_match, &dest_ip);
I don't think you need this type-casting.
> + /* Check for exact match, this would give a faster lookup*/
> + if (direct_entry && direct_entry->mac && direct_entry->arp.mac) {
> + src_mac = &direct_entry->mac;
> + dest_mac = &direct_entry->arp.mac;
> + forward_to = direct_entry->ifindex;
> + } else {
> + /* Look up in the trie for lpm*/
> + key4.b32[0] = 32;
> + key4.b8[4] = dest_ip & 0xff;
> + key4.b8[5] = (dest_ip >> 8) & 0xff;
> + key4.b8[6] = (dest_ip >> 16) & 0xff;
> + key4.b8[7] = (dest_ip >> 24) & 0xff;
> + prefix_value = ((struct trie_value *)bpf_map_lookup_elem
> + (&lpm_map, &key4));
> + if (!prefix_value)
> + return XDP_DROP;
> + src_mac = &prefix_value->value;
> + if (!src_mac)
> + return XDP_DROP;
> + dest_mac = (__be64 *)bpf_map_lookup_elem(&arp_table, &dest_ip);
> + if (!dest_mac) {
> + if (!prefix_value->gw)
> + return XDP_DROP;
> + dest_ip = *(__be32 *)&prefix_value->gw;
> + dest_mac = (__be64 *)bpf_map_lookup_elem(&arp_table, &dest_ip);
> + }
> + forward_to = prefix_value->ifindex;
> + }
> + } else {
> + ipproto = 0;
> + }
> + if (src_mac && dest_mac) {
> + set_src_dst_mac(data, src_mac, dest_mac);
> + value = bpf_map_lookup_elem(&rxcnt, &ipproto);
> + if (value)
> + *value += 1;
> + return bpf_redirect(forward_to, 0);
Notice that using bpf_redirect() is slow, while using bpf_redirect_map()
is fast. Using bpf_redirect_map() requires a little more book keeping,
but the performance gain is worth it.
Raw benchmarks on my system show:
* bpf_redirect() max at 7Mpps
* bpf_redirect_map() at 13Mpps
Trying out your program on my systems showed it jumps between 5.6Mpps
to 7Mpps. And it seems to be correlated with matching direct_entry.
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2017-11-02 11:22 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-11-01 7:48 [PATCH v3 0/1] XDP program for ip forward Christina Jacob
2017-11-01 7:48 ` [PATCH v3 1/1] xdp: Sample xdp program implementing " Christina Jacob
2017-11-02 11:21 ` Jesper Dangaard Brouer
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).