Skip to contentSkip to conversation

samples: bpf: update map definition to new syntax BTF-defined map

PATCH Applied Daniel T. Lee 3 participants 6 messages Nov 7, 2019 – Nov 9, 2019

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

v2 cover samples: bpf: update map definition to new syntax BTF-defined map

Since, the new syntax of BTF-defined map has been introduced, the syntax for using maps under samples directory are mixed up. For example, some are already using the new syntax, and some are using existing syntax by calling them as 'legacy'.

As stated at commit abd29c931459 ("libbpf: allow specifying map definitions using BTF"), the BTF-defined map has more compatablility with extending supported map definition features.

Also, unifying the map definition to BTF-defined map will help reduce confusion between new syntax and existing syntax.

The commit doesn't replace all of the map to new BTF-defined map, because some of the samples still use bpf_load instead of libbpf, which can't properly create BTF-defined map.

This will only updates the samples which uses libbpf API for loading bpf program. (ex. bpf_prog_load_xattr)

This patchset fixes some of the outdated error message regarded to loading bpf program (load_bpf_file -> bpf_prog_load_xattr), and updates map definition to new syntax of BTF-defined map.

v1 -> v2:
- stick to __type() instead of __uint({key,value}_size) where possible

Daniel T. Lee (2):
samples: bpf: update outdated error message
samples: bpf: update map definition to new syntax BTF-defined map

 samples/bpf/hbm.c                   |   2 +-
 samples/bpf/sockex1_kern.c          |  12 ++--
 samples/bpf/sockex2_kern.c          |  12 ++--
 samples/bpf/xdp1_kern.c             |  12 ++--
 samples/bpf/xdp1_user.c             |   2 +-
 samples/bpf/xdp2_kern.c             |  12 ++--
 samples/bpf/xdp_adjust_tail_kern.c  |  12 ++--
 samples/bpf/xdp_fwd_kern.c          |  13 ++--
 samples/bpf/xdp_redirect_cpu_kern.c | 108 ++++++++++++++--------------
 samples/bpf/xdp_redirect_kern.c     |  24 +++----
 samples/bpf/xdp_redirect_map_kern.c |  24 +++----
 samples/bpf/xdp_router_ipv4_kern.c  |  64 ++++++++---------
 samples/bpf/xdp_rxq_info_kern.c     |  37 +++++-----
 samples/bpf/xdp_rxq_info_user.c     |   6 +-
 samples/bpf/xdp_sample_pkts_user.c  |   2 +-
 samples/bpf/xdp_tx_iptunnel_kern.c  |  26 +++----
 samples/bpf/xdp_tx_iptunnel_user.c  |   2 +-
 17 files changed, 185 insertions(+), 185 deletions(-)
Signature · 1 line
2.23.0
#1 lore raw

PATCH v2 1/2 samples: bpf: update outdated error message

Currently, under samples, several methods are being used to load bpf program.

Since using libbpf is preferred solution, lots of previously used 'load_bpf_file' from bpf_load are replaced with 'bpf_prog_load_xattr' from libbpf.

But some of the error messages still show up as 'load_bpf_file' instead of 'bpf_prog_load_xattr'.

This commit fixes outdated errror messages under samples and fixes some code style issues.

---
 samples/bpf/hbm.c                  | 2 +-
 samples/bpf/xdp1_user.c            | 2 +-
 samples/bpf/xdp_rxq_info_user.c    | 6 +++---
 samples/bpf/xdp_sample_pkts_user.c | 2 +-
 samples/bpf/xdp_tx_iptunnel_user.c | 2 +-
 5 files changed, 7 insertions(+), 7 deletions(-)
samples/bpf/hbm.c +1 -1
index e0fbab9bec83..829b68d87687 100644
@@ -147,7 +147,7 @@ static int prog_load(char *prog)
 	}
 
 	if (ret) {
-		printf("ERROR: load_bpf_file failed for: %s\n", prog);
+		printf("ERROR: bpf_prog_load_xattr failed for: %s\n", prog);
 		printf("  Output from verifier:\n%s\n------\n", bpf_log_buf);
 		ret = -1;
 	} else {
samples/bpf/xdp1_user.c +1 -1
index a8e5fa02e8a8..3e553eed95a7 100644
@@ -139,7 +139,7 @@ int main(int argc, char **argv)
 	map_fd = bpf_map__fd(map);
 
 	if (!prog_fd) {
-		printf("load_bpf_file: %s\n", strerror(errno));
+		printf("bpf_prog_load_xattr: %s\n", strerror(errno));
 		return 1;
 	}
 
samples/bpf/xdp_rxq_info_user.c +3 -3
index c7e4e45d824a..51e0d810e070 100644
@@ -51,8 +51,8 @@ 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' },
-	{"swapmac", 	no_argument,		NULL, 'm' },
+	{"readmem",	no_argument,		NULL, 'r' },
+	{"swapmac",	no_argument,		NULL, 'm' },
 	{"force",	no_argument,		NULL, 'F' },
 	{0, 0, NULL,  0 }
 };
@@ -499,7 +499,7 @@ int main(int argc, char **argv)
 	map_fd = bpf_map__fd(map);
 
 	if (!prog_fd) {
-		fprintf(stderr, "ERR: load_bpf_file: %s\n", strerror(errno));
+		fprintf(stderr, "ERR: bpf_prog_load_xattr: %s\n", strerror(errno));
 		return EXIT_FAIL;
 	}
 
samples/bpf/xdp_sample_pkts_user.c +1 -1
index 3002714e3cd5..a5760e8bf2c4 100644
@@ -150,7 +150,7 @@ int main(int argc, char **argv)
 		return 1;
 
 	if (!prog_fd) {
-		printf("load_bpf_file: %s\n", strerror(errno));
+		printf("bpf_prog_load_xattr: %s\n", strerror(errno));
 		return 1;
 	}
 
samples/bpf/xdp_tx_iptunnel_user.c +1 -1
index dfb68582e243..2fe4c7f5ffe5 100644
@@ -268,7 +268,7 @@ int main(int argc, char **argv)
 		return 1;
 
 	if (!prog_fd) {
-		printf("load_bpf_file: %s\n", strerror(errno));
+		printf("bpf_prog_load_xattr: %s\n", strerror(errno));
 		return 1;
 	}
 
Signature · 1 line
2.23.0
#2 lore raw

PATCH v2 2/2 samples: bpf: update map definition to new syntax BTF-defined map

In reply to Daniel T. Lee

Since, the new syntax of BTF-defined map has been introduced, the syntax for using maps under samples directory are mixed up. For example, some are already using the new syntax, and some are using existing syntax by calling them as 'legacy'.

As stated at commit abd29c931459 ("libbpf: allow specifying map definitions using BTF"), the BTF-defined map has more compatablility with extending supported map definition features.

The commit doesn't replace all of the map to new BTF-defined map, because some of the samples still use bpf_load instead of libbpf, which can't properly create BTF-defined map.

This will only updates the samples which uses libbpf API for loading bpf program. (ex. bpf_prog_load_xattr)

---
Changes in v2:
- stick to __type() instead of __uint({key,value}_size) where possible

 samples/bpf/sockex1_kern.c          |  12 ++--
 samples/bpf/sockex2_kern.c          |  12 ++--
 samples/bpf/xdp1_kern.c             |  12 ++--
 samples/bpf/xdp2_kern.c             |  12 ++--
 samples/bpf/xdp_adjust_tail_kern.c  |  12 ++--
 samples/bpf/xdp_fwd_kern.c          |  13 ++--
 samples/bpf/xdp_redirect_cpu_kern.c | 108 ++++++++++++++--------------
 samples/bpf/xdp_redirect_kern.c     |  24 +++----
 samples/bpf/xdp_redirect_map_kern.c |  24 +++----
 samples/bpf/xdp_router_ipv4_kern.c  |  64 ++++++++---------
 samples/bpf/xdp_rxq_info_kern.c     |  37 +++++-----
 samples/bpf/xdp_tx_iptunnel_kern.c  |  26 +++----
 12 files changed, 178 insertions(+), 178 deletions(-)
samples/bpf/sockex1_kern.c +6 -6
index f96943f443ab..2408dbfb7a21 100644
@@ -5,12 +5,12 @@
 #include "bpf_helpers.h"
 #include "bpf_legacy.h"
 
-struct bpf_map_def SEC("maps") my_map = {
-	.type = BPF_MAP_TYPE_ARRAY,
-	.key_size = sizeof(u32),
-	.value_size = sizeof(long),
-	.max_entries = 256,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__type(key, u32);
+	__type(value, long);
+	__uint(max_entries, 256);
+} my_map SEC(".maps");
 
 SEC("socket1")
 int bpf_prog1(struct __sk_buff *skb)
samples/bpf/sockex2_kern.c +6 -6
index 5566fa7d92fa..a7bcd03bf529 100644
@@ -190,12 +190,12 @@ struct pair {
 	long bytes;
 };
 
-struct bpf_map_def SEC("maps") hash_map = {
-	.type = BPF_MAP_TYPE_HASH,
-	.key_size = sizeof(__be32),
-	.value_size = sizeof(struct pair),
-	.max_entries = 1024,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__type(key, __be32);
+	__type(value, struct pair);
+	__uint(max_entries, 1024);
+} hash_map SEC(".maps");
 
 SEC("socket2")
 int bpf_prog2(struct __sk_buff *skb)
samples/bpf/xdp1_kern.c +6 -6
index 219742106bfd..db6870aee42c 100644
@@ -14,12 +14,12 @@
 #include <linux/ipv6.h>
 #include "bpf_helpers.h"
 
-struct bpf_map_def SEC("maps") rxcnt = {
-	.type = BPF_MAP_TYPE_PERCPU_ARRAY,
-	.key_size = sizeof(u32),
-	.value_size = sizeof(long),
-	.max_entries = 256,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__type(key, u32);
+	__type(value, long);
+	__uint(max_entries, 256);
+} rxcnt SEC(".maps");
 
 static int parse_ipv4(void *data, u64 nh_off, void *data_end)
 {
samples/bpf/xdp2_kern.c +6 -6
index e01288867d15..c74b52c6d945 100644
@@ -14,12 +14,12 @@
 #include <linux/ipv6.h>
 #include "bpf_helpers.h"
 
-struct bpf_map_def SEC("maps") rxcnt = {
-	.type = BPF_MAP_TYPE_PERCPU_ARRAY,
-	.key_size = sizeof(u32),
-	.value_size = sizeof(long),
-	.max_entries = 256,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__type(key, u32);
+	__type(value, long);
+	__uint(max_entries, 256);
+} rxcnt SEC(".maps");
 
 static void swap_src_dst_mac(void *data)
 {
samples/bpf/xdp_adjust_tail_kern.c +6 -6
index c616508befb9..0f707e0fb375 100644
@@ -28,12 +28,12 @@
 /* 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),
-	.value_size = sizeof(__u64),
-	.max_entries = 1,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__type(key, __u32);
+	__type(value, __u64);
+	__uint(max_entries, 1);
+} icmpcnt SEC(".maps");
 
 static __always_inline void count_icmp(void)
 {
samples/bpf/xdp_fwd_kern.c +6 -7
index 701a30f258b1..d013029aeaa2 100644
@@ -23,13 +23,12 @@
 
 #define IPV6_FLOWINFO_MASK              cpu_to_be32(0x0FFFFFFF)
 
-/* For TX-traffic redirect requires net_device ifindex to be in this devmap */
-struct bpf_map_def SEC("maps") xdp_tx_ports = {
-	.type = BPF_MAP_TYPE_DEVMAP,
-	.key_size = sizeof(int),
-	.value_size = sizeof(int),
-	.max_entries = 64,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_DEVMAP);
+	__uint(key_size, sizeof(int));
+	__uint(value_size, sizeof(int));
+	__uint(max_entries, 64);
+} xdp_tx_ports SEC(".maps");
 
 /* from include/net/ip.h */
 static __always_inline int ip_decrease_ttl(struct iphdr *iph)
samples/bpf/xdp_redirect_cpu_kern.c +54 -54

133 lines not on this page · raw message

samples/bpf/xdp_redirect_kern.c +12 -12
index 8abb151e385f..1f0b7d05abb2 100644
@@ -19,22 +19,22 @@
 #include <linux/ipv6.h>
 #include "bpf_helpers.h"
 
-struct bpf_map_def SEC("maps") tx_port = {
-	.type = BPF_MAP_TYPE_ARRAY,
-	.key_size = sizeof(int),
-	.value_size = sizeof(int),
-	.max_entries = 1,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__type(key, int);
+	__type(value, int);
+	__uint(max_entries, 1);
+} tx_port SEC(".maps");
 
 /* Count RX packets, as XDP bpf_prog doesn't get direct TX-success
  * feedback.  Redirect TX errors can be caught via a tracepoint.
  */
-struct bpf_map_def SEC("maps") rxcnt = {
-	.type = BPF_MAP_TYPE_PERCPU_ARRAY,
-	.key_size = sizeof(u32),
-	.value_size = sizeof(long),
-	.max_entries = 1,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__type(key, u32);
+	__type(value, long);
+	__uint(max_entries, 1);
+} rxcnt SEC(".maps");
 
 static void swap_src_dst_mac(void *data)
 {
samples/bpf/xdp_redirect_map_kern.c +12 -12
index 740a529ba84f..4631b484c432 100644
@@ -19,22 +19,22 @@
 #include <linux/ipv6.h>
 #include "bpf_helpers.h"
 
-struct bpf_map_def SEC("maps") tx_port = {
-	.type = BPF_MAP_TYPE_DEVMAP,
-	.key_size = sizeof(int),
-	.value_size = sizeof(int),
-	.max_entries = 100,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_DEVMAP);
+	__uint(key_size, sizeof(int));
+	__uint(value_size, sizeof(int));
+	__uint(max_entries, 100);
+} tx_port SEC(".maps");
 
 /* Count RX packets, as XDP bpf_prog doesn't get direct TX-success
  * feedback.  Redirect TX errors can be caught via a tracepoint.
  */
-struct bpf_map_def SEC("maps") rxcnt = {
-	.type = BPF_MAP_TYPE_PERCPU_ARRAY,
-	.key_size = sizeof(u32),
-	.value_size = sizeof(long),
-	.max_entries = 1,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__type(key, u32);
+	__type(value, long);
+	__uint(max_entries, 1);
+} rxcnt SEC(".maps");
 
 static void swap_src_dst_mac(void *data)
 {
samples/bpf/xdp_router_ipv4_kern.c +32 -32

76 lines not on this page · raw message

samples/bpf/xdp_rxq_info_kern.c +19 -18
index 222a83eed1cb..272d0f82a6b5 100644
@@ -23,12 +23,13 @@ enum cfg_options_flags {
 	READ_MEM = 0x1U,
 	SWAP_MAC = 0x2U,
 };
-struct bpf_map_def SEC("maps") config_map = {
-	.type		= BPF_MAP_TYPE_ARRAY,
-	.key_size	= sizeof(int),
-	.value_size	= sizeof(struct config),
-	.max_entries	= 1,
-};
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__type(key, int);
+	__type(value, struct config);
+	__uint(max_entries, 1);
+} config_map SEC(".maps");
 
 /* Common stats data record (shared with userspace) */
 struct datarec {
@@ -36,22 +37,22 @@ struct datarec {
 	__u64 issue;
 };
 
-struct bpf_map_def SEC("maps") stats_global_map = {
-	.type		= BPF_MAP_TYPE_PERCPU_ARRAY,
-	.key_size	= sizeof(u32),
-	.value_size	= sizeof(struct datarec),
-	.max_entries	= 1,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__type(key, u32);
+	__type(value, struct datarec);
+	__uint(max_entries, 1);
+} stats_global_map SEC(".maps");
 
 #define MAX_RXQs 64
 
 /* Stats per rx_queue_index (per CPU) */
-struct bpf_map_def SEC("maps") rx_queue_index_map = {
-	.type		= BPF_MAP_TYPE_PERCPU_ARRAY,
-	.key_size	= sizeof(u32),
-	.value_size	= sizeof(struct datarec),
-	.max_entries	= MAX_RXQs + 1,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__type(key, u32);
+	__type(value, struct datarec);
+	__uint(max_entries, MAX_RXQs + 1);
+} rx_queue_index_map SEC(".maps");
 
 static __always_inline
 void swap_src_dst_mac(void *data)
samples/bpf/xdp_tx_iptunnel_kern.c +13 -13
index 0f4f6e8c8611..6db450a5c1ca 100644
@@ -19,19 +19,19 @@
 #include "bpf_helpers.h"
 #include "xdp_tx_iptunnel_common.h"
 
-struct bpf_map_def SEC("maps") rxcnt = {
-	.type = BPF_MAP_TYPE_PERCPU_ARRAY,
-	.key_size = sizeof(__u32),
-	.value_size = sizeof(__u64),
-	.max_entries = 256,
-};
-
-struct bpf_map_def SEC("maps") vip2tnl = {
-	.type = BPF_MAP_TYPE_HASH,
-	.key_size = sizeof(struct vip),
-	.value_size = sizeof(struct iptnl_info),
-	.max_entries = MAX_IPTNL_ENTRIES,
-};
+struct {
+	__uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+	__type(key, __u32);
+	__type(value, __u64);
+	__uint(max_entries, 256);
+} rxcnt SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_HASH);
+	__type(key, struct vip);
+	__type(value, struct iptnl_info);
+	__uint(max_entries, MAX_IPTNL_ENTRIES);
+} vip2tnl SEC(".maps");
 
 static __always_inline void count_tx(u32 protocol)
 {
Signature · 1 line
2.23.0
Acked-by on v2 2/2
#3 lore raw
Quoted 22 lines · Daniel T. Lee

Since, the new syntax of BTF-defined map has been introduced, the syntax for using maps under samples directory are mixed up. For example, some are already using the new syntax, and some are using existing syntax by calling them as 'legacy'.

As stated at commit abd29c931459 ("libbpf: allow specifying map definitions using BTF"), the BTF-defined map has more compatablility with extending supported map definition features.

The commit doesn't replace all of the map to new BTF-defined map, because some of the samples still use bpf_load instead of libbpf, which can't properly create BTF-defined map.

This will only updates the samples which uses libbpf API for loading bpf program. (ex. bpf_prog_load_xattr)

---
Changes in v2:
- stick to __type() instead of __uint({key,value}_size) where possible

Quoted 13 lines
 samples/bpf/sockex1_kern.c          |  12 ++--
 samples/bpf/sockex2_kern.c          |  12 ++--
 samples/bpf/xdp1_kern.c             |  12 ++--
 samples/bpf/xdp2_kern.c             |  12 ++--
 samples/bpf/xdp_adjust_tail_kern.c  |  12 ++--
 samples/bpf/xdp_fwd_kern.c          |  13 ++--
 samples/bpf/xdp_redirect_cpu_kern.c | 108 ++++++++++++++--------------
 samples/bpf/xdp_redirect_kern.c     |  24 +++----
 samples/bpf/xdp_redirect_map_kern.c |  24 +++----
 samples/bpf/xdp_router_ipv4_kern.c  |  64 ++++++++---------
 samples/bpf/xdp_rxq_info_kern.c     |  37 +++++-----
 samples/bpf/xdp_tx_iptunnel_kern.c  |  26 +++----
 12 files changed, 178 insertions(+), 178 deletions(-)

Heh, 1-to-1 insertions/deletions, no excuse to use old syntax ;)

Thanks for completing conversion!

[...]

#4 lore raw
Quoted 38 lines · Andrii Nakryiko
Quoted 22 lines · Daniel T. Lee

Since, the new syntax of BTF-defined map has been introduced, the syntax for using maps under samples directory are mixed up. For example, some are already using the new syntax, and some are using existing syntax by calling them as 'legacy'.

As stated at commit abd29c931459 ("libbpf: allow specifying map definitions using BTF"), the BTF-defined map has more compatablility with extending supported map definition features.

The commit doesn't replace all of the map to new BTF-defined map, because some of the samples still use bpf_load instead of libbpf, which can't properly create BTF-defined map.

This will only updates the samples which uses libbpf API for loading bpf program. (ex. bpf_prog_load_xattr)

---
Changes in v2:
- stick to __type() instead of __uint({key,value}_size) where possible

Quoted 13 lines
 samples/bpf/sockex1_kern.c          |  12 ++--
 samples/bpf/sockex2_kern.c          |  12 ++--
 samples/bpf/xdp1_kern.c             |  12 ++--
 samples/bpf/xdp2_kern.c             |  12 ++--
 samples/bpf/xdp_adjust_tail_kern.c  |  12 ++--
 samples/bpf/xdp_fwd_kern.c          |  13 ++--
 samples/bpf/xdp_redirect_cpu_kern.c | 108 ++++++++++++++--------------
 samples/bpf/xdp_redirect_kern.c     |  24 +++----
 samples/bpf/xdp_redirect_map_kern.c |  24 +++----
 samples/bpf/xdp_router_ipv4_kern.c  |  64 ++++++++---------
 samples/bpf/xdp_rxq_info_kern.c     |  37 +++++-----
 samples/bpf/xdp_tx_iptunnel_kern.c  |  26 +++----
 12 files changed, 178 insertions(+), 178 deletions(-)
Quoted 7 lines

Heh, 1-to-1 insertions/deletions, no excuse to use old syntax ;)

Thanks for completing conversion!

[...]

Thank you for your time and effort for the review.

Thanks,
Daniel

#5 lore raw
Quoted 15 lines · Daniel T. Lee
Quoted 12 lines · Andrii Nakryiko
Quoted 10 lines · Daniel T. Lee

Since, the new syntax of BTF-defined map has been introduced, the syntax for using maps under samples directory are mixed up. For example, some are already using the new syntax, and some are using existing syntax by calling them as 'legacy'.

As stated at commit abd29c931459 ("libbpf: allow specifying map definitions using BTF"), the BTF-defined map has more compatablility with extending supported map definition features.

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

The commit doesn't replace all of the map to new BTF-defined map, because some of the samples still use bpf_load instead of libbpf, which can't properly create BTF-defined map.

This will only updates the samples which uses libbpf API for loading bpf program. (ex. bpf_prog_load_xattr)

Applied. Thanks