Daniel Borkmann says:

====================
pull-request: bpf-next 2023-02-11

We've added 96 non-merge commits during the last 14 day(s) which contain
a total of 152 files changed, 4884 insertions(+), 962 deletions(-).

There is a minor conflict in drivers/net/ethernet/intel/ice/ice_main.c
between commit 5b246e533d01 ("ice: split probe into smaller functions")
from the net-next tree and commit 66c0e13ad236 ("drivers: net: turn on
XDP features") from the bpf-next tree. Remove the hunk given ice_cfg_netdev()
is otherwise there a 2nd time, and add XDP features to the existing
ice_cfg_netdev() one:

        [...]
        ice_set_netdev_features(netdev);
        netdev->xdp_features = NETDEV_XDP_ACT_BASIC | NETDEV_XDP_ACT_REDIRECT |
                               NETDEV_XDP_ACT_XSK_ZEROCOPY;
        ice_set_ops(netdev);
        [...]

Stephen's merge conflict mail:
https://lore.kernel.org/bpf/20230207101951.21a114fa@canb.auug.org.au/

The main changes are:

1) Add support for BPF trampoline on s390x which finally allows to remove many
   test cases from the BPF CI's DENYLIST.s390x, from Ilya Leoshkevich.

2) Add multi-buffer XDP support to ice driver, from Maciej Fijalkowski.

3) Add capability to export the XDP features supported by the NIC.
   Along with that, add a XDP compliance test tool,
   from Lorenzo Bianconi & Marek Majtyka.

4) Add __bpf_kfunc tag for marking kernel functions as kfuncs,
   from David Vernet.

5) Add a deep dive documentation about the verifier's register
   liveness tracking algorithm, from Eduard Zingerman.

6) Fix and follow-up cleanups for resolve_btfids to be compiled
   as a host program to avoid cross compile issues,
   from Jiri Olsa & Ian Rogers.

7) Batch of fixes to the BPF selftest for xdp_hw_metadata which resulted
   when testing on different NICs, from Jesper Dangaard Brouer.

8) Fix libbpf to better detect kernel version code on Debian, from Hao Xiang.

9) Extend libbpf to add an option for when the perf buffer should
   wake up, from Jon Doron.

10) Follow-up fix on xdp_metadata selftest to just consume on TX
    completion, from Stanislav Fomichev.

11) Extend the kfuncs.rst document with description on kfunc
    lifecycle & stability expectations, from David Vernet.

12) Fix bpftool prog profile to skip attaching to offline CPUs,
    from Tonghao Zhang.

====================

Link: https://lore.kernel.org/r/20230211002037.8489-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

9 files changed, 223 insertions, 56 deletions

diff --git a/tools/lib/bpf/bpf_core_read.h b/tools/lib/bpf/bpf_core_read.h
index 496e6a8ee0dc..1ac57bb7ac55 100644
--- a/tools/lib/bpf/bpf_core_read.h
+++ b/tools/lib/bpf/bpf_core_read.h
@@ -364,7 +364,7 @@ enum bpf_enum_value_kind {
 
 /* Non-CO-RE variant of BPF_CORE_READ_INTO() */
 #define BPF_PROBE_READ_INTO(dst, src, a, ...) ({			    \
-	___core_read(bpf_probe_read, bpf_probe_read,			    \
+	___core_read(bpf_probe_read_kernel, bpf_probe_read_kernel,	    \
 		     dst, (src), a, ##__VA_ARGS__)			    \
 })
 
@@ -400,7 +400,7 @@ enum bpf_enum_value_kind {
 
 /* Non-CO-RE variant of BPF_CORE_READ_STR_INTO() */
 #define BPF_PROBE_READ_STR_INTO(dst, src, a, ...) ({			    \
-	___core_read(bpf_probe_read_str, bpf_probe_read,		    \
+	___core_read(bpf_probe_read_kernel_str, bpf_probe_read_kernel,	    \
 		     dst, (src), a, ##__VA_ARGS__)			    \
 })
 
diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index d37c4fe2849d..5ec1871acb2f 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -109,7 +109,7 @@
  * This is a variable-specific variant of more global barrier().
  */
 #ifndef barrier_var
-#define barrier_var(var) asm volatile("" : "=r"(var) : "0"(var))
+#define barrier_var(var) asm volatile("" : "+r"(var))
 #endif
 
 /*
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index eed5cec6f510..35a698eb825d 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -34,7 +34,6 @@
 #include <linux/limits.h>
 #include <linux/perf_event.h>
 #include <linux/ring_buffer.h>
-#include <linux/version.h>
 #include <sys/epoll.h>
 #include <sys/ioctl.h>
 #include <sys/mman.h>
@@ -870,42 +869,6 @@ bpf_object__add_programs(struct bpf_object *obj, Elf_Data *sec_data,
 	return 0;
 }
 
-__u32 get_kernel_version(void)
-{
-	/* On Ubuntu LINUX_VERSION_CODE doesn't correspond to info.release,
-	 * but Ubuntu provides /proc/version_signature file, as described at
-	 * https://ubuntu.com/kernel, with an example contents below, which we
-	 * can use to get a proper LINUX_VERSION_CODE.
-	 *
-	 *   Ubuntu 5.4.0-12.15-generic 5.4.8
-	 *
-	 * In the above, 5.4.8 is what kernel is actually expecting, while
-	 * uname() call will return 5.4.0 in info.release.
-	 */
-	const char *ubuntu_kver_file = "/proc/version_signature";
-	__u32 major, minor, patch;
-	struct utsname info;
-
-	if (faccessat(AT_FDCWD, ubuntu_kver_file, R_OK, AT_EACCESS) == 0) {
-		FILE *f;
-
-		f = fopen(ubuntu_kver_file, "r");
-		if (f) {
-			if (fscanf(f, "%*s %*s %d.%d.%d\n", &major, &minor, &patch) == 3) {
-				fclose(f);
-				return KERNEL_VERSION(major, minor, patch);
-			}
-			fclose(f);
-		}
-		/* something went wrong, fall back to uname() approach */
-	}
-
-	uname(&info);
-	if (sscanf(info.release, "%u.%u.%u", &major, &minor, &patch) != 3)
-		return 0;
-	return KERNEL_VERSION(major, minor, patch);
-}
-
 static const struct btf_member *
 find_member_by_offset(const struct btf_type *t, __u32 bit_offset)
 {
@@ -11710,17 +11673,22 @@ struct perf_buffer *perf_buffer__new(int map_fd, size_t page_cnt,
 	const size_t attr_sz = sizeof(struct perf_event_attr);
 	struct perf_buffer_params p = {};
 	struct perf_event_attr attr;
+	__u32 sample_period;
 
 	if (!OPTS_VALID(opts, perf_buffer_opts))
 		return libbpf_err_ptr(-EINVAL);
 
+	sample_period = OPTS_GET(opts, sample_period, 1);
+	if (!sample_period)
+		sample_period = 1;
+
 	memset(&attr, 0, attr_sz);
 	attr.size = attr_sz;
 	attr.config = PERF_COUNT_SW_BPF_OUTPUT;
 	attr.type = PERF_TYPE_SOFTWARE;
 	attr.sample_type = PERF_SAMPLE_RAW;
-	attr.sample_period = 1;
-	attr.wakeup_events = 1;
+	attr.sample_period = sample_period;
+	attr.wakeup_events = sample_period;
 
 	p.attr = &attr;
 	p.sample_cb = sample_cb;
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 8777ff21ea1d..2efd80f6f7b9 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -1048,9 +1048,10 @@ struct bpf_xdp_query_opts {
 	__u32 hw_prog_id;	/* output */
 	__u32 skb_prog_id;	/* output */
 	__u8 attach_mode;	/* output */
+	__u64 feature_flags;	/* output */
 	size_t :0;
 };
-#define bpf_xdp_query_opts__last_field attach_mode
+#define bpf_xdp_query_opts__last_field feature_flags
 
 LIBBPF_API int bpf_xdp_attach(int ifindex, int prog_fd, __u32 flags,
 			      const struct bpf_xdp_attach_opts *opts);
@@ -1246,8 +1247,10 @@ typedef void (*perf_buffer_lost_fn)(void *ctx, int cpu, __u64 cnt);
 /* common use perf buffer options */
 struct perf_buffer_opts {
 	size_t sz;
+	__u32 sample_period;
+	size_t :0;
 };
-#define perf_buffer_opts__last_field sz
+#define perf_buffer_opts__last_field sample_period
 
 /**
  * @brief **perf_buffer__new()** creates BPF perfbuf manager for a specified
diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
index b44fcbb4b42e..4f3bc968ff8e 100644
--- a/tools/lib/bpf/libbpf_probes.c
+++ b/tools/lib/bpf/libbpf_probes.c
@@ -12,11 +12,94 @@
 #include <linux/btf.h>
 #include <linux/filter.h>
 #include <linux/kernel.h>
+#include <linux/version.h>
 
 #include "bpf.h"
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
+/* On Ubuntu LINUX_VERSION_CODE doesn't correspond to info.release,
+ * but Ubuntu provides /proc/version_signature file, as described at
+ * https://ubuntu.com/kernel, with an example contents below, which we
+ * can use to get a proper LINUX_VERSION_CODE.
+ *
+ *   Ubuntu 5.4.0-12.15-generic 5.4.8
+ *
+ * In the above, 5.4.8 is what kernel is actually expecting, while
+ * uname() call will return 5.4.0 in info.release.
+ */
+static __u32 get_ubuntu_kernel_version(void)
+{
+	const char *ubuntu_kver_file = "/proc/version_signature";
+	__u32 major, minor, patch;
+	int ret;
+	FILE *f;
+
+	if (faccessat(AT_FDCWD, ubuntu_kver_file, R_OK, AT_EACCESS) != 0)
+		return 0;
+
+	f = fopen(ubuntu_kver_file, "r");
+	if (!f)
+		return 0;
+
+	ret = fscanf(f, "%*s %*s %u.%u.%u\n", &major, &minor, &patch);
+	fclose(f);
+	if (ret != 3)
+		return 0;
+
+	return KERNEL_VERSION(major, minor, patch);
+}
+
+/* On Debian LINUX_VERSION_CODE doesn't correspond to info.release.
+ * Instead, it is provided in info.version. An example content of
+ * Debian 10 looks like the below.
+ *
+ *   utsname::release   4.19.0-22-amd64
+ *   utsname::version   #1 SMP Debian 4.19.260-1 (2022-09-29)
+ *
+ * In the above, 4.19.260 is what kernel is actually expecting, while
+ * uname() call will return 4.19.0 in info.release.
+ */
+static __u32 get_debian_kernel_version(struct utsname *info)
+{
+	__u32 major, minor, patch;
+	char *p;
+
+	p = strstr(info->version, "Debian ");
+	if (!p) {
+		/* This is not a Debian kernel. */
+		return 0;
+	}
+
+	if (sscanf(p, "Debian %u.%u.%u", &major, &minor, &patch) != 3)
+		return 0;
+
+	return KERNEL_VERSION(major, minor, patch);
+}
+
+__u32 get_kernel_version(void)
+{
+	__u32 major, minor, patch, version;
+	struct utsname info;
+
+	/* Check if this is an Ubuntu kernel. */
+	version = get_ubuntu_kernel_version();
+	if (version != 0)
+		return version;
+
+	uname(&info);
+
+	/* Check if this is a Debian kernel. */
+	version = get_debian_kernel_version(&info);
+	if (version != 0)
+		return version;
+
+	if (sscanf(info.release, "%u.%u.%u", &major, &minor, &patch) != 3)
+		return 0;
+
+	return KERNEL_VERSION(major, minor, patch);
+}
+
 static int probe_prog_load(enum bpf_prog_type prog_type,
 			   const struct bpf_insn *insns, size_t insns_cnt,
 			   char *log_buf, size_t log_buf_sz)
diff --git a/tools/lib/bpf/netlink.c b/tools/lib/bpf/netlink.c
index 35104580870c..cb082a04ffa8 100644
--- a/tools/lib/bpf/netlink.c
+++ b/tools/lib/bpf/netlink.c
@@ -9,6 +9,7 @@
 #include <linux/if_ether.h>
 #include <linux/pkt_cls.h>
 #include <linux/rtnetlink.h>
+#include <linux/netdev.h>
 #include <sys/socket.h>
 #include <errno.h>
 #include <time.h>
@@ -39,9 +40,15 @@ struct xdp_id_md {
 	int ifindex;
 	__u32 flags;
 	struct xdp_link_info info;
+	__u64 feature_flags;
 };
 
-static int libbpf_netlink_open(__u32 *nl_pid)
+struct xdp_features_md {
+	int ifindex;
+	__u64 flags;
+};
+
+static int libbpf_netlink_open(__u32 *nl_pid, int proto)
 {
 	struct sockaddr_nl sa;
 	socklen_t addrlen;
@@ -51,7 +58,7 @@ static int libbpf_netlink_open(__u32 *nl_pid)
 	memset(&sa, 0, sizeof(sa));
 	sa.nl_family = AF_NETLINK;
 
-	sock = socket(AF_NETLINK, SOCK_RAW | SOCK_CLOEXEC, NETLINK_ROUTE);
+	sock = socket(AF_NETLINK, SOCK_RAW | SOCK_CLOEXEC, proto);
 	if (sock < 0)
 		return -errno;
 
@@ -212,14 +219,14 @@ done:
 }
 
 static int libbpf_netlink_send_recv(struct libbpf_nla_req *req,
-				    __dump_nlmsg_t parse_msg,
+				    int proto, __dump_nlmsg_t parse_msg,
 				    libbpf_dump_nlmsg_t parse_attr,
 				    void *cookie)
 {
 	__u32 nl_pid = 0;
 	int sock, ret;
 
-	sock = libbpf_netlink_open(&nl_pid);
+	sock = libbpf_netlink_open(&nl_pid, proto);
 	if (sock < 0)
 		return sock;
 
@@ -238,6 +245,43 @@ out:
 	return ret;
 }
 
+static int parse_genl_family_id(struct nlmsghdr *nh, libbpf_dump_nlmsg_t fn,
+				void *cookie)
+{
+	struct genlmsghdr *gnl = NLMSG_DATA(nh);
+	struct nlattr *na = (struct nlattr *)((void *)gnl + GENL_HDRLEN);
+	struct nlattr *tb[CTRL_ATTR_FAMILY_ID + 1];
+	__u16 *id = cookie;
+
+	libbpf_nla_parse(tb, CTRL_ATTR_FAMILY_ID, na,
+			 NLMSG_PAYLOAD(nh, sizeof(*gnl)), NULL);
+	if (!tb[CTRL_ATTR_FAMILY_ID])
+		return NL_CONT;
+
+	*id = libbpf_nla_getattr_u16(tb[CTRL_ATTR_FAMILY_ID]);
+	return NL_DONE;
+}
+
+static int libbpf_netlink_resolve_genl_family_id(const char *name,
+						 __u16 len, __u16 *id)
+{
+	struct libbpf_nla_req req = {
+		.nh.nlmsg_len	= NLMSG_LENGTH(GENL_HDRLEN),
+		.nh.nlmsg_type	= GENL_ID_CTRL,
+		.nh.nlmsg_flags	= NLM_F_REQUEST,
+		.gnl.cmd	= CTRL_CMD_GETFAMILY,
+		.gnl.version	= 2,
+	};
+	int err;
+
+	err = nlattr_add(&req, CTRL_ATTR_FAMILY_NAME, name, len);
+	if (err < 0)
+		return err;
+
+	return libbpf_netlink_send_recv(&req, NETLINK_GENERIC,
+					parse_genl_family_id, NULL, id);
+}
+
 static int __bpf_set_link_xdp_fd_replace(int ifindex, int fd, int old_fd,
 					 __u32 flags)
 {
@@ -271,7 +315,7 @@ static int __bpf_set_link_xdp_fd_replace(int ifindex, int fd, int old_fd,
 	}
 	nlattr_end_nested(&req, nla);
 
-	return libbpf_netlink_send_recv(&req, NULL, NULL, NULL);
+	return libbpf_netlink_send_recv(&req, NETLINK_ROUTE, NULL, NULL, NULL);
 }
 
 int bpf_xdp_attach(int ifindex, int prog_fd, __u32 flags, const struct bpf_xdp_attach_opts *opts)
@@ -357,6 +401,29 @@ static int get_xdp_info(void *cookie, void *msg, struct nlattr **tb)
 	return 0;
 }
 
+static int parse_xdp_features(struct nlmsghdr *nh, libbpf_dump_nlmsg_t fn,
+			      void *cookie)
+{
+	struct genlmsghdr *gnl = NLMSG_DATA(nh);
+	struct nlattr *na = (struct nlattr *)((void *)gnl + GENL_HDRLEN);
+	struct nlattr *tb[NETDEV_CMD_MAX + 1];
+	struct xdp_features_md *md = cookie;
+	__u32 ifindex;
+
+	libbpf_nla_parse(tb, NETDEV_CMD_MAX, na,
+			 NLMSG_PAYLOAD(nh, sizeof(*gnl)), NULL);
+
+	if (!tb[NETDEV_A_DEV_IFINDEX] || !tb[NETDEV_A_DEV_XDP_FEATURES])
+		return NL_CONT;
+
+	ifindex = libbpf_nla_getattr_u32(tb[NETDEV_A_DEV_IFINDEX]);
+	if (ifindex != md->ifindex)
+		return NL_CONT;
+
+	md->flags = libbpf_nla_getattr_u64(tb[NETDEV_A_DEV_XDP_FEATURES]);
+	return NL_DONE;
+}
+
 int bpf_xdp_query(int ifindex, int xdp_flags, struct bpf_xdp_query_opts *opts)
 {
 	struct libbpf_nla_req req = {
@@ -366,6 +433,10 @@ int bpf_xdp_query(int ifindex, int xdp_flags, struct bpf_xdp_query_opts *opts)
 		.ifinfo.ifi_family = AF_PACKET,
 	};
 	struct xdp_id_md xdp_id = {};
+	struct xdp_features_md md = {
+		.ifindex = ifindex,
+	};
+	__u16 id;
 	int err;
 
 	if (!OPTS_VALID(opts, bpf_xdp_query_opts))
@@ -382,7 +453,7 @@ int bpf_xdp_query(int ifindex, int xdp_flags, struct bpf_xdp_query_opts *opts)
 	xdp_id.ifindex = ifindex;
 	xdp_id.flags = xdp_flags;
 
-	err = libbpf_netlink_send_recv(&req, __dump_link_nlmsg,
+	err = libbpf_netlink_send_recv(&req, NETLINK_ROUTE, __dump_link_nlmsg,
 				       get_xdp_info, &xdp_id);
 	if (err)
 		return libbpf_err(err);
@@ -393,6 +464,31 @@ int bpf_xdp_query(int ifindex, int xdp_flags, struct bpf_xdp_query_opts *opts)
 	OPTS_SET(opts, skb_prog_id, xdp_id.info.skb_prog_id);
 	OPTS_SET(opts, attach_mode, xdp_id.info.attach_mode);
 
+	if (!OPTS_HAS(opts, feature_flags))
+		return 0;
+
+	err = libbpf_netlink_resolve_genl_family_id("netdev", sizeof("netdev"), &id);
+	if (err < 0)
+		return libbpf_err(err);
+
+	memset(&req, 0, sizeof(req));
+	req.nh.nlmsg_len = NLMSG_LENGTH(GENL_HDRLEN);
+	req.nh.nlmsg_flags = NLM_F_REQUEST;
+	req.nh.nlmsg_type = id;
+	req.gnl.cmd = NETDEV_CMD_DEV_GET;
+	req.gnl.version = 2;
+
+	err = nlattr_add(&req, NETDEV_A_DEV_IFINDEX, &ifindex, sizeof(ifindex));
+	if (err < 0)
+		return libbpf_err(err);
+
+	err = libbpf_netlink_send_recv(&req, NETLINK_GENERIC,
+				       parse_xdp_features, NULL, &md);
+	if (err)
+		return libbpf_err(err);
+
+	opts->feature_flags = md.flags;
+
 	return 0;
 }
 
@@ -493,7 +589,7 @@ static int tc_qdisc_modify(struct bpf_tc_hook *hook, int cmd, int flags)
 	if (ret < 0)
 		return ret;
 
-	return libbpf_netlink_send_recv(&req, NULL, NULL, NULL);
+	return libbpf_netlink_send_recv(&req, NETLINK_ROUTE, NULL, NULL, NULL);
 }
 
 static int tc_qdisc_create_excl(struct bpf_tc_hook *hook)
@@ -673,7 +769,8 @@ int bpf_tc_attach(const struct bpf_tc_hook *hook, struct bpf_tc_opts *opts)
 
 	info.opts = opts;
 
-	ret = libbpf_netlink_send_recv(&req, get_tc_info, NULL, &info);
+	ret = libbpf_netlink_send_recv(&req, NETLINK_ROUTE, get_tc_info, NULL,
+				       &info);
 	if (ret < 0)
 		return libbpf_err(ret);
 	if (!info.processed)
@@ -739,7 +836,7 @@ static int __bpf_tc_detach(const struct bpf_tc_hook *hook,
 			return ret;
 	}
 
-	return libbpf_netlink_send_recv(&req, NULL, NULL, NULL);
+	return libbpf_netlink_send_recv(&req, NETLINK_ROUTE, NULL, NULL, NULL);
 }
 
 int bpf_tc_detach(const struct bpf_tc_hook *hook,
@@ -804,7 +901,8 @@ int bpf_tc_query(const struct bpf_tc_hook *hook, struct bpf_tc_opts *opts)
 
 	info.opts = opts;
 
-	ret = libbpf_netlink_send_recv(&req, get_tc_info, NULL, &info);
+	ret = libbpf_netlink_send_recv(&req, NETLINK_ROUTE, get_tc_info, NULL,
+				       &info);
 	if (ret < 0)
 		return libbpf_err(ret);
 	if (!info.processed)
diff --git a/tools/lib/bpf/nlattr.c b/tools/lib/bpf/nlattr.c
index 3900d052ed19..975e265eab3b 100644
--- a/tools/lib/bpf/nlattr.c
+++ b/tools/lib/bpf/nlattr.c
@@ -178,7 +178,7 @@ int libbpf_nla_dump_errormsg(struct nlmsghdr *nlh)
 		hlen += nlmsg_len(&err->msg);
 
 	attr = (struct nlattr *) ((void *) err + hlen);
-	alen = nlh->nlmsg_len - hlen;
+	alen = (void *)nlh + nlh->nlmsg_len - (void *)attr;
 
 	if (libbpf_nla_parse(tb, NLMSGERR_ATTR_MAX, attr, alen,
 			     extack_policy) != 0) {
diff --git a/tools/lib/bpf/nlattr.h b/tools/lib/bpf/nlattr.h
index 4d15ae2ff812..d92d1c1de700 100644
--- a/tools/lib/bpf/nlattr.h
+++ b/tools/lib/bpf/nlattr.h
@@ -14,6 +14,7 @@
 #include <errno.h>
 #include <linux/netlink.h>
 #include <linux/rtnetlink.h>
+#include <linux/genetlink.h>
 
 /* avoid multiple definition of netlink features */
 #define __LINUX_NETLINK_H
@@ -58,6 +59,7 @@ struct libbpf_nla_req {
 	union {
 		struct ifinfomsg ifinfo;
 		struct tcmsg tc;
+		struct genlmsghdr gnl;
 	};
 	char buf[128];
 };
@@ -89,11 +91,21 @@ static inline uint8_t libbpf_nla_getattr_u8(const struct nlattr *nla)
 	return *(uint8_t *)libbpf_nla_data(nla);
 }
 
+static inline uint16_t libbpf_nla_getattr_u16(const struct nlattr *nla)
+{
+	return *(uint16_t *)libbpf_nla_data(nla);
+}
+
 static inline uint32_t libbpf_nla_getattr_u32(const struct nlattr *nla)
 {
 	return *(uint32_t *)libbpf_nla_data(nla);
 }
 
+static inline uint64_t libbpf_nla_getattr_u64(const struct nlattr *nla)
+{
+	return *(uint64_t *)libbpf_nla_data(nla);
+}
+
 static inline const char *libbpf_nla_getattr_str(const struct nlattr *nla)
 {
 	return (const char *)libbpf_nla_data(nla);
diff --git a/tools/lib/bpf/usdt.bpf.h b/tools/lib/bpf/usdt.bpf.h
index fdfd235e52c4..0bd4c135acc2 100644
--- a/tools/lib/bpf/usdt.bpf.h
+++ b/tools/lib/bpf/usdt.bpf.h
@@ -130,7 +130,10 @@ int bpf_usdt_arg(struct pt_regs *ctx, __u64 arg_num, long *res)
 	if (!spec)
 		return -ESRCH;
 
-	if (arg_num >= BPF_USDT_MAX_ARG_CNT || arg_num >= spec->arg_cnt)
+	if (arg_num >= BPF_USDT_MAX_ARG_CNT)
+		return -ENOENT;
+	barrier_var(arg_num);
+	if (arg_num >= spec->arg_cnt)
 		return -ENOENT;
 
 	arg_spec = &spec->args[arg_num];