samples/bpf: extend xdp_rxq_info to read packet payload

There is a cost associated with reading the packet data payload
that this test ignored.  Add option --read to allow enabling
reading part of the payload.

This sample/tool helps us analyse an issue observed with a NIC
mlx5 (ConnectX-5 Ex) and an Intel(R) Xeon(R) CPU E5-1650 v4.

With no_touch of data:

Running XDP on dev:mlx5p1 (ifindex:8) action:XDP_DROP options:no_touch
XDP stats       CPU     pps         issue-pps
XDP-RX CPU      0       14,465,157  0
XDP-RX CPU      1       14,464,728  0
XDP-RX CPU      2       14,465,283  0
XDP-RX CPU      3       14,465,282  0
XDP-RX CPU      4       14,464,159  0
XDP-RX CPU      5       14,465,379  0
XDP-RX CPU      total   86,789,992

When not touching data, we observe that the CPUs have idle cycles.
When reading data the CPUs are 100% busy in softirq.

With reading data:

Running XDP on dev:mlx5p1 (ifindex:8) action:XDP_DROP options:read
XDP stats       CPU     pps         issue-pps
XDP-RX CPU      0       9,620,639   0
XDP-RX CPU      1       9,489,843   0
XDP-RX CPU      2       9,407,854   0
XDP-RX CPU      3       9,422,289   0
XDP-RX CPU      4       9,321,959   0
XDP-RX CPU      5       9,395,242   0
XDP-RX CPU      total   56,657,828

The effect seen above is a result of cache-misses occuring when
more RXQs are being used.  Based on perf-event observations, our
conclusion is that the CPUs DDIO (Direct Data I/O) choose to
deliver packet into main memory, instead of L3-cache.  We also
found, that this can be mitigated by either using less RXQs or by
reducing NICs the RX-ring size.

Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Acked-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>

samples/bpf/xdp_rxq_info_kern.c

@@ -4,6 +4,8 @@
  *  Example howto extract XDP RX-queue info
  */
 #include <uapi/linux/bpf.h>
+#include <uapi/linux/if_ether.h>
+#include <uapi/linux/in.h>
 #include "bpf_helpers.h"
 
 /* Config setup from with userspace
@@ -14,6 +16,11 @@
 struct config {
 	__u32 action;
 	int ifindex;
+	__u32 options;
+};
+enum cfg_options_flags {
+	NO_TOUCH = 0x0U,
+	READ_MEM = 0x1U,
 };
 struct bpf_map_def SEC("maps") config_map = {
 	.type		= BPF_MAP_TYPE_ARRAY,
@@ -90,6 +97,18 @@ int  xdp_prognum0(struct xdp_md *ctx)
 	if (key == MAX_RXQs)
 		rxq_rec->issue++;
 
+	/* Default: Don't touch packet data, only count packets */
+	if (unlikely(config->options & READ_MEM)) {
+		struct ethhdr *eth = data;
+
+		if (eth + 1 > data_end)
+			return XDP_ABORTED;
+
+		/* Avoid compiler removing this: Drop non 802.3 Ethertypes */
+		if (ntohs(eth->h_proto) < ETH_P_802_3_MIN)
+			return XDP_ABORTED;
+	}
+
 	return config->action;
 }
 

samples/bpf/xdp_rxq_info_user.c

@@ -50,6 +50,7 @@ static const struct option long_options[] = {
 	{"sec",		required_argument,	NULL, 's' },
 	{"no-separators", no_argument,		NULL, 'z' },
 	{"action",	required_argument,	NULL, 'a' },
+	{"readmem", 	no_argument,		NULL, 'r' },
 	{0, 0, NULL,  0 }
 };
 
@@ -66,6 +67,11 @@ static void int_exit(int sig)
 struct config {
 	__u32 action;
 	int ifindex;
+	__u32 options;
+};
+enum cfg_options_flags {
+	NO_TOUCH = 0x0U,
+	READ_MEM = 0x1U,
 };
 #define XDP_ACTION_MAX (XDP_TX + 1)
 #define XDP_ACTION_MAX_STRLEN 11
@@ -109,6 +115,16 @@ static void list_xdp_actions(void)
 	printf("\n");
 }
 
+static char* options2str(enum cfg_options_flags flag)
+{
+	if (flag == NO_TOUCH)
+		return "no_touch";
+	if (flag & READ_MEM)
+		return "read";
+	fprintf(stderr, "ERR: Unknown config option flags");
+	exit(EXIT_FAIL);
+}
+
 static void usage(char *argv[])
 {
 	int i;
@@ -305,7 +321,7 @@ static __u64 calc_errs_pps(struct datarec *r,
 
 static void stats_print(struct stats_record *stats_rec,
 			struct stats_record *stats_prev,
-			int action)
+			int action, __u32 cfg_opt)
 {
 	unsigned int nr_rxqs = bpf_map__def(rx_queue_index_map)->max_entries;
 	unsigned int nr_cpus = bpf_num_possible_cpus();
@@ -316,8 +332,8 @@ static void stats_print(struct stats_record *stats_rec,
 	int i;
 
 	/* Header */
-	printf("\nRunning XDP on dev:%s (ifindex:%d) action:%s\n",
+	printf("\nRunning XDP on dev:%s (ifindex:%d) action:%s options:%s\n",
-	       ifname, ifindex, action2str(action));
+	       ifname, ifindex, action2str(action), options2str(cfg_opt));
 
 	/* stats_global_map */
 	{
@@ -399,7 +415,7 @@ static inline void swap(struct stats_record **a, struct stats_record **b)
 	*b = tmp;
 }
 
-static void stats_poll(int interval, int action)
+static void stats_poll(int interval, int action, __u32 cfg_opt)
 {
 	struct stats_record *record, *prev;
 
@@ -410,7 +426,7 @@ static void stats_poll(int interval, int action)
 	while (1) {
 		swap(&prev, &record);
 		stats_collect(record);
-		stats_print(record, prev, action);
+		stats_print(record, prev, action, cfg_opt);
 		sleep(interval);
 	}
 
@@ -421,6 +437,7 @@ static void stats_poll(int interval, int action)
 
 int main(int argc, char **argv)
 {
+	__u32 cfg_options= NO_TOUCH ; /* Default: Don't touch packet memory */
 	struct rlimit r = {10 * 1024 * 1024, RLIM_INFINITY};
 	struct bpf_prog_load_attr prog_load_attr = {
 		.prog_type	= BPF_PROG_TYPE_XDP,
@@ -435,6 +452,7 @@ int main(int argc, char **argv)
 	int interval = 2;
 	__u32 key = 0;
 
+
 	char action_str_buf[XDP_ACTION_MAX_STRLEN + 1 /* for \0 */] = { 0 };
 	int action = XDP_PASS; /* Default action */
 	char *action_str = NULL;
@@ -496,6 +514,9 @@ int main(int argc, char **argv)
 			action_str = (char *)&action_str_buf;
 			strncpy(action_str, optarg, XDP_ACTION_MAX_STRLEN);
 			break;
+		case 'r':
+			cfg_options |= READ_MEM;
+			break;
 		case 'h':
 		error:
 		default:
@@ -522,6 +543,7 @@ int main(int argc, char **argv)
 		}
 	}
 	cfg.action = action;
+	cfg.options = cfg_options;
 
 	/* Trick to pretty printf with thousands separators use %' */
 	if (use_separators)
@@ -542,6 +564,6 @@ int main(int argc, char **argv)
 		return EXIT_FAIL_XDP;
 	}
 
-	stats_poll(interval, action);
+	stats_poll(interval, action, cfg_options);
 	return EXIT_OK;
 }