Skip to contentSkip to conversation

samples: bpf: add max_pckt_size option at xdp_adjust_tail

PATCH Superseded Daniel T. Lee 2 participants 3 messages Oct 7, 2019

Subsystems (4)

v6 of 7 Superseded Latest is v7 →

lore
Share & tools
lore
#0 lore raw

PATCH v6 1/1 samples: bpf: add max_pckt_size option at xdp_adjust_tail

Currently, at xdp_adjust_tail_kern.c, MAX_PCKT_SIZE is limited to 600. To make this size flexible, static global variable 'max_pcktsz' is added.

By updating new packet size from the user space, xdp_adjust_tail_kern.o will use this value as a new max packet size.

This static global variable can be accesible from .data section with bpf_object__find_map* from user space, since it is considered as internal map (accessible with .bss/.data/.rodata suffix).

If no '-P <MAX_PCKT_SIZE>' option is used, the size of maximum packet will be 600 as a default.

Changed the way to test prog_fd, map_fd from '!= 0' to '< 0', since fd could be 0 when stdin is closed.

---
Changes in v6:
- Remove redundant error message
Changes in v5:
- Change pcktsz map to static global variable
Changes in v4:
- make pckt_size no less than ICMP_TOOBIG_SIZE
- Fix code style
Changes in v2:
- Change the helper to fetch map from 'bpf_map__next' to 'bpf_object__find_map_fd_by_name'.

 samples/bpf/xdp_adjust_tail_kern.c |  7 +++++--
 samples/bpf/xdp_adjust_tail_user.c | 29 ++++++++++++++++++++---------
 2 files changed, 25 insertions(+), 11 deletions(-)
samples/bpf/xdp_adjust_tail_kern.c +5 -2
index 411fdb21f8bc..c616508befb9 100644
@@ -25,6 +25,9 @@
 #define ICMP_TOOBIG_SIZE 98
 #define ICMP_TOOBIG_PAYLOAD_SIZE 92
 
+/* volatile to prevent compiler optimizations */
+static volatile __u32 max_pcktsz = MAX_PCKT_SIZE;
+
 struct bpf_map_def SEC("maps") icmpcnt = {
 	.type = BPF_MAP_TYPE_ARRAY,
 	.key_size = sizeof(__u32),
@@ -92,7 +95,7 @@ static __always_inline int send_icmp4_too_big(struct xdp_md *xdp)
 	orig_iph = data + off;
 	icmp_hdr->type = ICMP_DEST_UNREACH;
 	icmp_hdr->code = ICMP_FRAG_NEEDED;
-	icmp_hdr->un.frag.mtu = htons(MAX_PCKT_SIZE-sizeof(struct ethhdr));
+	icmp_hdr->un.frag.mtu = htons(max_pcktsz - sizeof(struct ethhdr));
 	icmp_hdr->checksum = 0;
 	ipv4_csum(icmp_hdr, ICMP_TOOBIG_PAYLOAD_SIZE, &csum);
 	icmp_hdr->checksum = csum;
@@ -121,7 +124,7 @@ static __always_inline int handle_ipv4(struct xdp_md *xdp)
 	int pckt_size = data_end - data;
 	int offset;
 
-	if (pckt_size > MAX_PCKT_SIZE) {
+	if (pckt_size > max(max_pcktsz, ICMP_TOOBIG_SIZE)) {
 		offset = pckt_size - ICMP_TOOBIG_SIZE;
 		if (bpf_xdp_adjust_tail(xdp, 0 - offset))
 			return XDP_PASS;
samples/bpf/xdp_adjust_tail_user.c +20 -9
index a3596b617c4c..d86e9ad0356b 100644
@@ -23,6 +23,7 @@
 #include "libbpf.h"
 
 #define STATS_INTERVAL_S 2U
+#define MAX_PCKT_SIZE 600
 
 static int ifindex = -1;
 static __u32 xdp_flags = XDP_FLAGS_UPDATE_IF_NOEXIST;
@@ -72,6 +73,7 @@ static void usage(const char *cmd)
 	printf("Usage: %s [...]\n", cmd);
 	printf("    -i <ifname|ifindex> Interface\n");
 	printf("    -T <stop-after-X-seconds> Default: 0 (forever)\n");
+	printf("    -P <MAX_PCKT_SIZE> Default: %u\n", MAX_PCKT_SIZE);
 	printf("    -S use skb-mode\n");
 	printf("    -N enforce native mode\n");
 	printf("    -F force loading prog\n");
@@ -85,13 +87,14 @@ int main(int argc, char **argv)
 		.prog_type	= BPF_PROG_TYPE_XDP,
 	};
 	unsigned char opt_flags[256] = {};
-	const char *optstr = "i:T:SNFh";
+	const char *optstr = "i:T:P:SNFh";
 	struct bpf_prog_info info = {};
 	__u32 info_len = sizeof(info);
 	unsigned int kill_after_s = 0;
 	int i, prog_fd, map_fd, opt;
 	struct bpf_object *obj;
-	struct bpf_map *map;
+	__u32 max_pckt_size = 0;
+	__u32 key = 0;
 	char filename[256];
 	int err;
 
@@ -110,6 +113,9 @@ int main(int argc, char **argv)
 		case 'T':
 			kill_after_s = atoi(optarg);
 			break;
+		case 'P':
+			max_pckt_size = atoi(optarg);
+			break;
 		case 'S':
 			xdp_flags |= XDP_FLAGS_SKB_MODE;
 			break;
@@ -150,15 +156,20 @@ int main(int argc, char **argv)
 	if (bpf_prog_load_xattr(&prog_load_attr, &obj, &prog_fd))
 		return 1;
 
-	map = bpf_map__next(NULL, obj);
-	if (!map) {
-		printf("finding a map in obj file failed\n");
-		return 1;
+	/* static global var 'max_pcktsz' is accessible from .data section */
+	if (max_pckt_size) {
+		map_fd = bpf_object__find_map_fd_by_name(obj, "xdp_adju.data");
+		if (map_fd < 0) {
+			printf("finding a max_pcktsz map in obj file failed\n");
+			return 1;
+		}
+		bpf_map_update_elem(map_fd, &key, &max_pckt_size, BPF_ANY);
 	}
-	map_fd = bpf_map__fd(map);
 
-	if (!prog_fd) {
-		printf("load_bpf_file: %s\n", strerror(errno));
+	/* fetch icmpcnt map */
+	map_fd = bpf_object__find_map_fd_by_name(obj, "icmpcnt");
+	if (map_fd < 0) {
+		printf("finding a icmpcnt map in obj file failed\n");
 		return 1;
 	}
 
Signature · 1 line
2.20.1
Acked-by on v6 1/1
#1 lore raw
Quoted 20 lines · Daniel T. Lee

Currently, at xdp_adjust_tail_kern.c, MAX_PCKT_SIZE is limited to 600. To make this size flexible, static global variable 'max_pcktsz' is added.

By updating new packet size from the user space, xdp_adjust_tail_kern.o will use this value as a new max packet size.

This static global variable can be accesible from .data section with bpf_object__find_map* from user space, since it is considered as internal map (accessible with .bss/.data/.rodata suffix).

If no '-P <MAX_PCKT_SIZE>' option is used, the size of maximum packet will be 600 as a default.

Changed the way to test prog_fd, map_fd from '!= 0' to '< 0', since fd could be 0 when stdin is closed.

Quoted 11 lines

---
Changes in v6:
- Remove redundant error message
Changes in v5:
- Change pcktsz map to static global variable
Changes in v4:
- make pckt_size no less than ICMP_TOOBIG_SIZE
- Fix code style
Changes in v2:
- Change the helper to fetch map from 'bpf_map__next' to 'bpf_object__find_map_fd_by_name'.

This should go into commit message, that's netdev preference.

Otherwise looks good!

Quoted 13 lines
 samples/bpf/xdp_adjust_tail_kern.c |  7 +++++--
 samples/bpf/xdp_adjust_tail_user.c | 29 ++++++++++++++++++++---------
 2 files changed, 25 insertions(+), 11 deletions(-)
samples/bpf/xdp_adjust_tail_kern.c +0 -0
index 411fdb21f8bc..c616508befb9 100644
@@ -25,6 +25,9 @@
 #define ICMP_TOOBIG_SIZE 98
 #define ICMP_TOOBIG_PAYLOAD_SIZE 92
 

[...]

#2 lore raw
Quoted 35 lines · Andrii Nakryiko
Quoted 20 lines · Daniel T. Lee

Currently, at xdp_adjust_tail_kern.c, MAX_PCKT_SIZE is limited to 600. To make this size flexible, static global variable 'max_pcktsz' is added.

By updating new packet size from the user space, xdp_adjust_tail_kern.o will use this value as a new max packet size.

This static global variable can be accesible from .data section with bpf_object__find_map* from user space, since it is considered as internal map (accessible with .bss/.data/.rodata suffix).

If no '-P <MAX_PCKT_SIZE>' option is used, the size of maximum packet will be 600 as a default.

Changed the way to test prog_fd, map_fd from '!= 0' to '< 0', since fd could be 0 when stdin is closed.

Quoted 11 lines

---
Changes in v6:
- Remove redundant error message
Changes in v5:
- Change pcktsz map to static global variable
Changes in v4:
- make pckt_size no less than ICMP_TOOBIG_SIZE
- Fix code style
Changes in v2:
- Change the helper to fetch map from 'bpf_map__next' to 'bpf_object__find_map_fd_by_name'.

Quoted 6 lines

This should go into commit message, that's netdev preference.

Otherwise looks good!

Thanks for the feedback!

Will send the next patch with the modified commit message right away!

Thanks,
Daniel

Quoted 15 lines
Quoted 13 lines
 samples/bpf/xdp_adjust_tail_kern.c |  7 +++++--
 samples/bpf/xdp_adjust_tail_user.c | 29 ++++++++++++++++++++---------
 2 files changed, 25 insertions(+), 11 deletions(-)
samples/bpf/xdp_adjust_tail_kern.c +0 -0
index 411fdb21f8bc..c616508befb9 100644
@@ -25,6 +25,9 @@
 #define ICMP_TOOBIG_SIZE 98
 #define ICMP_TOOBIG_PAYLOAD_SIZE 92
 

[...]