Merge branch 'bpf-libbpf-helpers'

Andrii Nakryiko says:

====================
This patch set makes bpf_helpers.h and bpf_endian.h a part of libbpf itself
for consumption by user BPF programs, not just selftests. It also splits off
tracing helpers into bpf_tracing.h, which also becomes part of libbpf. Some of
the legacy stuff (BPF_ANNOTATE_KV_PAIR, load_{byte,half,word}, bpf_map_def
with unsupported fields, etc, is extracted into selftests-only bpf_legacy.h.
All the selftests and samples are switched to use libbpf's headers and
selftests' ones are removed.

As part of this patch set we also add BPF_CORE_READ variadic macros, that are
simplifying BPF CO-RE reads, especially the ones that have to follow few
pointers. E.g., what in non-BPF world (and when using BCC) would be:

int x = s->a->b.c->d; /* s, a, and b.c are pointers */

Today would have to be written using explicit bpf_probe_read() calls as:

  void *t;
  int x;
  bpf_probe_read(&t, sizeof(t), s->a);
  bpf_probe_read(&t, sizeof(t), ((struct b *)t)->b.c);
  bpf_probe_read(&x, sizeof(x), ((struct c *)t)->d);

This is super inconvenient and distracts from program logic a lot. Now, with
added BPF_CORE_READ() macros, you can write the above as:

  int x = BPF_CORE_READ(s, a, b.c, d);

Up to 9 levels of pointer chasing are supported, which should be enough for
any practical purpose, hopefully, without adding too much boilerplate macro
definitions (though there is admittedly some, given how variadic and recursive
C macro have to be implemented).

There is also BPF_CORE_READ_INTO() variant, which relies on caller to allocate
space for result:

  int x;
  BPF_CORE_READ_INTO(&x, s, a, b.c, d);

Result of last bpf_probe_read() call in the chain of calls is the result of
BPF_CORE_READ_INTO(). If any intermediate bpf_probe_read() aall fails, then
all the subsequent ones will fail too, so this is sufficient to know whether
overall "operation" succeeded or not. No short-circuiting of bpf_probe_read()s
is done, though.

BPF_CORE_READ_STR_INTO() is added as well, which differs from
BPF_CORE_READ_INTO() only in that last bpf_probe_read() call (to read final
field after chasing pointers) is replaced with bpf_probe_read_str(). Result of
bpf_probe_read_str() is returned as a result of BPF_CORE_READ_STR_INTO() macro
itself, so that applications can track return code and/or length of read
string.

Patch set outline:
- patch #1 undoes previously added GCC-specific bpf-helpers.h include;
- patch #2 splits off legacy stuff we don't want to carry over;
- patch #3 adjusts CO-RE reloc tests to avoid subsequent naming conflict with
  BPF_CORE_READ;
- patch #4 splits off bpf_tracing.h;
- patch #5 moves bpf_{helpers,endian,tracing}.h and bpf_helper_defs.h
  generation into libbpf and adjusts Makefiles to include libbpf for header
  search;
- patch #6 adds variadic BPF_CORE_READ() macro family, as described above;
- patch #7 adds tests to verify all possible levels of pointer nestedness for
  BPF_CORE_READ(), as well as correctness test for BPF_CORE_READ_STR_INTO().

v4->v5:
- move BPF_CORE_READ() stuff into bpf_core_read.h header (Alexei);

v3->v4:
- rebase on latest bpf-next master;
- bpf_helper_defs.h generation is moved into libbpf's Makefile;

v2->v3:
- small formatting fixes and macro () fixes (Song);

v1->v2:
- fix CO-RE reloc tests before bpf_helpers.h move (Song);
- split off legacy stuff we don't want to carry over (Daniel, Toke);
- split off bpf_tracing.h (Daniel);
- fix samples/bpf build (assuming other fixes are applied);
- switch remaining maps either to bpf_map_def_legacy or BTF-defined maps;
====================
Signed-off-by: NDaniel Borkmann <daniel@iogearbox.net>

samples/bpf/Makefile

@@ -282,7 +282,7 @@ $(obj)/hbm_edt_kern.o: $(src)/hbm.h $(src)/hbm_kern.h
 $(obj)/%.o: $(src)/%.c
 	@echo "  CLANG-bpf " $@
 	$(Q)$(CLANG) $(NOSTDINC_FLAGS) $(LINUXINCLUDE) $(EXTRA_CFLAGS) -I$(obj) \
-		-I$(srctree)/tools/testing/selftests/bpf/ \
+		-I$(srctree)/tools/testing/selftests/bpf/ -I$(srctree)/tools/lib/bpf/ \
 		-D__KERNEL__ -D__BPF_TRACING__ -Wno-unused-value -Wno-pointer-sign \
 		-D__TARGET_ARCH_$(SRCARCH) -Wno-compare-distinct-pointer-types \
 		-Wno-gnu-variable-sized-type-not-at-end \

samples/bpf/hbm_kern.h

@@ -59,21 +59,18 @@
 #define BYTES_PER_NS(delta, rate) ((((u64)(delta)) * (rate)) >> 20)
 #define BYTES_TO_NS(bytes, rate) div64_u64(((u64)(bytes)) << 20, (u64)(rate))
 
-struct bpf_map_def SEC("maps") queue_state = {
-	.type = BPF_MAP_TYPE_CGROUP_STORAGE,
-	.key_size = sizeof(struct bpf_cgroup_storage_key),
-	.value_size = sizeof(struct hbm_vqueue),
-};
-BPF_ANNOTATE_KV_PAIR(queue_state, struct bpf_cgroup_storage_key,
-		     struct hbm_vqueue);
-
-struct bpf_map_def SEC("maps") queue_stats = {
-	.type = BPF_MAP_TYPE_ARRAY,
-	.key_size = sizeof(u32),
-	.value_size = sizeof(struct hbm_queue_stats),
-	.max_entries = 1,
-};
-BPF_ANNOTATE_KV_PAIR(queue_stats, int, struct hbm_queue_stats);
+struct {
+	__uint(type, BPF_MAP_TYPE_CGROUP_STORAGE);
+	__type(key, struct bpf_cgroup_storage_key);
+	__type(value, struct hbm_vqueue);
+} queue_state SEC(".maps");
+
+struct {
+	__uint(type, BPF_MAP_TYPE_ARRAY);
+	__uint(max_entries, 1);
+	__type(key, u32);
+	__type(value, struct hvm_queue_stats);
+} queue_stats SEC(".maps");
 
 struct hbm_pkt_info {
 	int	cwnd;

samples/bpf/map_perf_test_kern.c

@@ -9,25 +9,27 @@
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_legacy.h"
+#include "bpf_tracing.h"
 
 #define MAX_ENTRIES 1000
 #define MAX_NR_CPUS 1024
 
-struct bpf_map_def SEC("maps") hash_map = {
+struct bpf_map_def_legacy SEC("maps") hash_map = {
 	.type = BPF_MAP_TYPE_HASH,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(long),
 	.max_entries = MAX_ENTRIES,
 };
 
-struct bpf_map_def SEC("maps") lru_hash_map = {
+struct bpf_map_def_legacy SEC("maps") lru_hash_map = {
 	.type = BPF_MAP_TYPE_LRU_HASH,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(long),
 	.max_entries = 10000,
 };
 
-struct bpf_map_def SEC("maps") nocommon_lru_hash_map = {
+struct bpf_map_def_legacy SEC("maps") nocommon_lru_hash_map = {
 	.type = BPF_MAP_TYPE_LRU_HASH,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(long),
@@ -35,7 +37,7 @@ struct bpf_map_def SEC("maps") nocommon_lru_hash_map = {
 	.map_flags = BPF_F_NO_COMMON_LRU,
 };
 
-struct bpf_map_def SEC("maps") inner_lru_hash_map = {
+struct bpf_map_def_legacy SEC("maps") inner_lru_hash_map = {
 	.type = BPF_MAP_TYPE_LRU_HASH,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(long),
@@ -44,20 +46,20 @@ struct bpf_map_def SEC("maps") inner_lru_hash_map = {
 	.numa_node = 0,
 };
 
-struct bpf_map_def SEC("maps") array_of_lru_hashs = {
+struct bpf_map_def_legacy SEC("maps") array_of_lru_hashs = {
 	.type = BPF_MAP_TYPE_ARRAY_OF_MAPS,
 	.key_size = sizeof(u32),
 	.max_entries = MAX_NR_CPUS,
 };
 
-struct bpf_map_def SEC("maps") percpu_hash_map = {
+struct bpf_map_def_legacy SEC("maps") percpu_hash_map = {
 	.type = BPF_MAP_TYPE_PERCPU_HASH,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(long),
 	.max_entries = MAX_ENTRIES,
 };
 
-struct bpf_map_def SEC("maps") hash_map_alloc = {
+struct bpf_map_def_legacy SEC("maps") hash_map_alloc = {
 	.type = BPF_MAP_TYPE_HASH,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(long),
@@ -65,7 +67,7 @@ struct bpf_map_def SEC("maps") hash_map_alloc = {
 	.map_flags = BPF_F_NO_PREALLOC,
 };
 
-struct bpf_map_def SEC("maps") percpu_hash_map_alloc = {
+struct bpf_map_def_legacy SEC("maps") percpu_hash_map_alloc = {
 	.type = BPF_MAP_TYPE_PERCPU_HASH,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(long),
@@ -73,7 +75,7 @@ struct bpf_map_def SEC("maps") percpu_hash_map_alloc = {
 	.map_flags = BPF_F_NO_PREALLOC,
 };
 
-struct bpf_map_def SEC("maps") lpm_trie_map_alloc = {
+struct bpf_map_def_legacy SEC("maps") lpm_trie_map_alloc = {
 	.type = BPF_MAP_TYPE_LPM_TRIE,
 	.key_size = 8,
 	.value_size = sizeof(long),
@@ -81,14 +83,14 @@ struct bpf_map_def SEC("maps") lpm_trie_map_alloc = {
 	.map_flags = BPF_F_NO_PREALLOC,
 };
 
-struct bpf_map_def SEC("maps") array_map = {
+struct bpf_map_def_legacy SEC("maps") array_map = {
 	.type = BPF_MAP_TYPE_ARRAY,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(long),
 	.max_entries = MAX_ENTRIES,
 };
 
-struct bpf_map_def SEC("maps") lru_hash_lookup_map = {
+struct bpf_map_def_legacy SEC("maps") lru_hash_lookup_map = {
 	.type = BPF_MAP_TYPE_LRU_HASH,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(long),

samples/bpf/offwaketime_kern.c

@@ -6,6 +6,7 @@
  */
 #include <uapi/linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 #include <uapi/linux/ptrace.h>
 #include <uapi/linux/perf_event.h>
 #include <linux/version.h>

samples/bpf/parse_ldabs.c

@@ -12,6 +12,7 @@
 #include <linux/udp.h>
 #include <uapi/linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_legacy.h"
 
 #define DEFAULT_PKTGEN_UDP_PORT	9
 #define IP_MF			0x2000

samples/bpf/sampleip_kern.c

@@ -9,6 +9,7 @@
 #include <uapi/linux/bpf.h>
 #include <uapi/linux/bpf_perf_event.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 #define MAX_IPS		8192
 

samples/bpf/sockex1_kern.c

@@ -3,6 +3,7 @@
 #include <uapi/linux/if_packet.h>
 #include <uapi/linux/ip.h>
 #include "bpf_helpers.h"
+#include "bpf_legacy.h"
 
 struct bpf_map_def SEC("maps") my_map = {
 	.type = BPF_MAP_TYPE_ARRAY,

samples/bpf/sockex2_kern.c

 #include <uapi/linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_legacy.h"
 #include <uapi/linux/in.h>
 #include <uapi/linux/if.h>
 #include <uapi/linux/if_ether.h>

samples/bpf/sockex3_kern.c

@@ -6,6 +6,7 @@
  */
 #include <uapi/linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_legacy.h"
 #include <uapi/linux/in.h>
 #include <uapi/linux/if.h>
 #include <uapi/linux/if_ether.h>

samples/bpf/spintest_kern.c

@@ -10,6 +10,7 @@
 #include <uapi/linux/bpf.h>
 #include <uapi/linux/perf_event.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 struct bpf_map_def SEC("maps") my_map = {
 	.type = BPF_MAP_TYPE_HASH,

samples/bpf/tcbpf1_kern.c

@@ -8,6 +8,7 @@
 #include <uapi/linux/filter.h>
 #include <uapi/linux/pkt_cls.h>
 #include "bpf_helpers.h"
+#include "bpf_legacy.h"
 
 /* compiler workaround */
 #define _htonl __builtin_bswap32

samples/bpf/test_map_in_map_kern.c

@@ -11,11 +11,13 @@
 #include <uapi/linux/bpf.h>
 #include <uapi/linux/in6.h>
 #include "bpf_helpers.h"
+#include "bpf_legacy.h"
+#include "bpf_tracing.h"
 
 #define MAX_NR_PORTS 65536
 
 /* map #0 */
-struct bpf_map_def SEC("maps") port_a = {
+struct bpf_map_def_legacy SEC("maps") port_a = {
 	.type = BPF_MAP_TYPE_ARRAY,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(int),
@@ -23,7 +25,7 @@ struct bpf_map_def SEC("maps") port_a = {
 };
 
 /* map #1 */
-struct bpf_map_def SEC("maps") port_h = {
+struct bpf_map_def_legacy SEC("maps") port_h = {
 	.type = BPF_MAP_TYPE_HASH,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(int),
@@ -31,7 +33,7 @@ struct bpf_map_def SEC("maps") port_h = {
 };
 
 /* map #2 */
-struct bpf_map_def SEC("maps") reg_result_h = {
+struct bpf_map_def_legacy SEC("maps") reg_result_h = {
 	.type = BPF_MAP_TYPE_HASH,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(int),
@@ -39,7 +41,7 @@ struct bpf_map_def SEC("maps") reg_result_h = {
 };
 
 /* map #3 */
-struct bpf_map_def SEC("maps") inline_result_h = {
+struct bpf_map_def_legacy SEC("maps") inline_result_h = {
 	.type = BPF_MAP_TYPE_HASH,
 	.key_size = sizeof(u32),
 	.value_size = sizeof(int),
@@ -47,7 +49,7 @@ struct bpf_map_def SEC("maps") inline_result_h = {
 };
 
 /* map #4 */ /* Test case #0 */
-struct bpf_map_def SEC("maps") a_of_port_a = {
+struct bpf_map_def_legacy SEC("maps") a_of_port_a = {
 	.type = BPF_MAP_TYPE_ARRAY_OF_MAPS,
 	.key_size = sizeof(u32),
 	.inner_map_idx = 0, /* map_fd[0] is port_a */
@@ -55,7 +57,7 @@ struct bpf_map_def SEC("maps") a_of_port_a = {
 };
 
 /* map #5 */ /* Test case #1 */
-struct bpf_map_def SEC("maps") h_of_port_a = {
+struct bpf_map_def_legacy SEC("maps") h_of_port_a = {
 	.type = BPF_MAP_TYPE_HASH_OF_MAPS,
 	.key_size = sizeof(u32),
 	.inner_map_idx = 0, /* map_fd[0] is port_a */
@@ -63,7 +65,7 @@ struct bpf_map_def SEC("maps") h_of_port_a = {
 };
 
 /* map #6 */ /* Test case #2 */
-struct bpf_map_def SEC("maps") h_of_port_h = {
+struct bpf_map_def_legacy SEC("maps") h_of_port_h = {
 	.type = BPF_MAP_TYPE_HASH_OF_MAPS,
 	.key_size = sizeof(u32),
 	.inner_map_idx = 1, /* map_fd[1] is port_h */

samples/bpf/test_overhead_kprobe_kern.c

@@ -8,6 +8,7 @@
 #include <linux/ptrace.h>
 #include <uapi/linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 #define _(P) ({typeof(P) val = 0; bpf_probe_read(&val, sizeof(val), &P); val;})
 

samples/bpf/test_probe_write_user_kern.c

@@ -9,6 +9,7 @@
 #include <uapi/linux/bpf.h>
 #include <linux/version.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 struct bpf_map_def SEC("maps") dnat_map = {
 	.type = BPF_MAP_TYPE_HASH,

samples/bpf/trace_event_kern.c

@@ -10,6 +10,7 @@
 #include <uapi/linux/bpf_perf_event.h>
 #include <uapi/linux/perf_event.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 struct key_t {
 	char comm[TASK_COMM_LEN];

samples/bpf/tracex1_kern.c

@@ -9,6 +9,7 @@
 #include <uapi/linux/bpf.h>
 #include <linux/version.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 #define _(P) ({typeof(P) val = 0; bpf_probe_read(&val, sizeof(val), &P); val;})
 

samples/bpf/tracex2_kern.c

@@ -9,6 +9,7 @@
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 struct bpf_map_def SEC("maps") my_map = {
 	.type = BPF_MAP_TYPE_HASH,

samples/bpf/tracex3_kern.c

@@ -9,6 +9,7 @@
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 struct bpf_map_def SEC("maps") my_map = {
 	.type = BPF_MAP_TYPE_HASH,

samples/bpf/tracex4_kern.c

@@ -8,6 +8,7 @@
 #include <linux/version.h>
 #include <uapi/linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 struct pair {
 	u64 val;

samples/bpf/tracex5_kern.c

@@ -11,6 +11,7 @@
 #include <uapi/linux/unistd.h>
 #include "syscall_nrs.h"
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 #define PROG(F) SEC("kprobe/"__stringify(F)) int bpf_func_##F
 

tools/lib/bpf/.gitignore

@@ -6,3 +6,4 @@ libbpf.so.*
 TAGS
 tags
 cscope.*
+/bpf_helper_defs.h

tools/lib/bpf/Makefile

@@ -157,7 +157,7 @@ all: fixdep
 
 all_cmd: $(CMD_TARGETS) check
 
-$(BPF_IN): force elfdep bpfdep
+$(BPF_IN): force elfdep bpfdep bpf_helper_defs.h
 	@(test -f ../../include/uapi/linux/bpf.h -a -f ../../../include/uapi/linux/bpf.h && ( \
 	(diff -B ../../include/uapi/linux/bpf.h ../../../include/uapi/linux/bpf.h >/dev/null) || \
 	echo "Warning: Kernel ABI header at 'tools/include/uapi/linux/bpf.h' differs from latest version at 'include/uapi/linux/bpf.h'" >&2 )) || true
@@ -175,6 +175,10 @@ $(BPF_IN): force elfdep bpfdep
 	echo "Warning: Kernel ABI header at 'tools/include/uapi/linux/if_xdp.h' differs from latest version at 'include/uapi/linux/if_xdp.h'" >&2 )) || true
 	$(Q)$(MAKE) $(build)=libbpf
 
+bpf_helper_defs.h: $(srctree)/include/uapi/linux/bpf.h
+	$(Q)$(srctree)/scripts/bpf_helpers_doc.py --header 		\
+		--file $(srctree)/include/uapi/linux/bpf.h > bpf_helper_defs.h
+
 $(OUTPUT)libbpf.so: $(OUTPUT)libbpf.so.$(LIBBPF_VERSION)
 
 $(OUTPUT)libbpf.so.$(LIBBPF_VERSION): $(BPF_IN)
@@ -236,13 +240,18 @@ install_lib: all_cmd
 		$(call do_install_mkdir,$(libdir_SQ)); \
 		cp -fpR $(LIB_FILE) $(DESTDIR)$(libdir_SQ)
 
-install_headers:
+install_headers: bpf_helper_defs.h
 	$(call QUIET_INSTALL, headers) \
 		$(call do_install,bpf.h,$(prefix)/include/bpf,644); \
 		$(call do_install,libbpf.h,$(prefix)/include/bpf,644); \
 		$(call do_install,btf.h,$(prefix)/include/bpf,644); \
 		$(call do_install,libbpf_util.h,$(prefix)/include/bpf,644); \
-		$(call do_install,xsk.h,$(prefix)/include/bpf,644);
+		$(call do_install,xsk.h,$(prefix)/include/bpf,644); \
+		$(call do_install,bpf_helpers.h,$(prefix)/include/bpf,644); \
+		$(call do_install,bpf_helper_defs.h,$(prefix)/include/bpf,644); \
+		$(call do_install,bpf_tracing.h,$(prefix)/include/bpf,644); \
+		$(call do_install,bpf_endian.h,$(prefix)/include/bpf,644); \
+		$(call do_install,bpf_core_read.h,$(prefix)/include/bpf,644);
 
 install_pkgconfig: $(PC_FILE)
 	$(call QUIET_INSTALL, $(PC_FILE)) \
@@ -259,7 +268,7 @@ config-clean:
 clean:
 	$(call QUIET_CLEAN, libbpf) $(RM) $(TARGETS) $(CXX_TEST_TARGET) \
 		*.o *~ *.a *.so *.so.$(LIBBPF_MAJOR_VERSION) .*.d .*.cmd \
-		*.pc LIBBPF-CFLAGS
+		*.pc LIBBPF-CFLAGS bpf_helper_defs.h
 	$(call QUIET_CLEAN, core-gen) $(RM) $(OUTPUT)FEATURE-DUMP.libbpf
 
 

tools/lib/bpf/bpf_core_read.h

+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+#ifndef __BPF_CORE_READ_H__
+#define __BPF_CORE_READ_H__
+
+/*
+ * bpf_core_read() abstracts away bpf_probe_read() call and captures offset
+ * relocation for source address using __builtin_preserve_access_index()
+ * built-in, provided by Clang.
+ *
+ * __builtin_preserve_access_index() takes as an argument an expression of
+ * taking an address of a field within struct/union. It makes compiler emit
+ * a relocation, which records BTF type ID describing root struct/union and an
+ * accessor string which describes exact embedded field that was used to take
+ * an address. See detailed description of this relocation format and
+ * semantics in comments to struct bpf_offset_reloc in libbpf_internal.h.
+ *
+ * This relocation allows libbpf to adjust BPF instruction to use correct
+ * actual field offset, based on target kernel BTF type that matches original
+ * (local) BTF, used to record relocation.
+ */
+#define bpf_core_read(dst, sz, src)					    \
+	bpf_probe_read(dst, sz,						    \
+		       (const void *)__builtin_preserve_access_index(src))
+
+/*
+ * bpf_core_read_str() is a thin wrapper around bpf_probe_read_str()
+ * additionally emitting BPF CO-RE field relocation for specified source
+ * argument.
+ */
+#define bpf_core_read_str(dst, sz, src)					    \
+	bpf_probe_read_str(dst, sz,					    \
+			   (const void *)__builtin_preserve_access_index(src))
+
+#define ___concat(a, b) a ## b
+#define ___apply(fn, n) ___concat(fn, n)
+#define ___nth(_1, _2, _3, _4, _5, _6, _7, _8, _9, _10, __11, N, ...) N
+
+/*
+ * return number of provided arguments; used for switch-based variadic macro
+ * definitions (see ___last, ___arrow, etc below)
+ */
+#define ___narg(...) ___nth(_, ##__VA_ARGS__, 10, 9, 8, 7, 6, 5, 4, 3, 2, 1, 0)
+/*
+ * return 0 if no arguments are passed, N - otherwise; used for
+ * recursively-defined macros to specify termination (0) case, and generic
+ * (N) case (e.g., ___read_ptrs, ___core_read)
+ */
+#define ___empty(...) ___nth(_, ##__VA_ARGS__, N, N, N, N, N, N, N, N, N, N, 0)
+
+#define ___last1(x) x
+#define ___last2(a, x) x
+#define ___last3(a, b, x) x
+#define ___last4(a, b, c, x) x
+#define ___last5(a, b, c, d, x) x
+#define ___last6(a, b, c, d, e, x) x
+#define ___last7(a, b, c, d, e, f, x) x
+#define ___last8(a, b, c, d, e, f, g, x) x
+#define ___last9(a, b, c, d, e, f, g, h, x) x
+#define ___last10(a, b, c, d, e, f, g, h, i, x) x
+#define ___last(...) ___apply(___last, ___narg(__VA_ARGS__))(__VA_ARGS__)
+
+#define ___nolast2(a, _) a
+#define ___nolast3(a, b, _) a, b
+#define ___nolast4(a, b, c, _) a, b, c
+#define ___nolast5(a, b, c, d, _) a, b, c, d
+#define ___nolast6(a, b, c, d, e, _) a, b, c, d, e
+#define ___nolast7(a, b, c, d, e, f, _) a, b, c, d, e, f
+#define ___nolast8(a, b, c, d, e, f, g, _) a, b, c, d, e, f, g
+#define ___nolast9(a, b, c, d, e, f, g, h, _) a, b, c, d, e, f, g, h
+#define ___nolast10(a, b, c, d, e, f, g, h, i, _) a, b, c, d, e, f, g, h, i
+#define ___nolast(...) ___apply(___nolast, ___narg(__VA_ARGS__))(__VA_ARGS__)
+
+#define ___arrow1(a) a
+#define ___arrow2(a, b) a->b
+#define ___arrow3(a, b, c) a->b->c
+#define ___arrow4(a, b, c, d) a->b->c->d
+#define ___arrow5(a, b, c, d, e) a->b->c->d->e
+#define ___arrow6(a, b, c, d, e, f) a->b->c->d->e->f
+#define ___arrow7(a, b, c, d, e, f, g) a->b->c->d->e->f->g
+#define ___arrow8(a, b, c, d, e, f, g, h) a->b->c->d->e->f->g->h
+#define ___arrow9(a, b, c, d, e, f, g, h, i) a->b->c->d->e->f->g->h->i
+#define ___arrow10(a, b, c, d, e, f, g, h, i, j) a->b->c->d->e->f->g->h->i->j
+#define ___arrow(...) ___apply(___arrow, ___narg(__VA_ARGS__))(__VA_ARGS__)
+
+#define ___type(...) typeof(___arrow(__VA_ARGS__))
+
+#define ___read(read_fn, dst, src_type, src, accessor)			    \
+	read_fn((void *)(dst), sizeof(*(dst)), &((src_type)(src))->accessor)
+
+/* "recursively" read a sequence of inner pointers using local __t var */
+#define ___rd_last(...)							    \
+	___read(bpf_core_read, &__t,					    \
+		___type(___nolast(__VA_ARGS__)), __t, ___last(__VA_ARGS__));
+#define ___rd_p0(src) const void *__t = src;
+#define ___rd_p1(...) ___rd_p0(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p2(...) ___rd_p1(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p3(...) ___rd_p2(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p4(...) ___rd_p3(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p5(...) ___rd_p4(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p6(...) ___rd_p5(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p7(...) ___rd_p6(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p8(...) ___rd_p7(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___rd_p9(...) ___rd_p8(___nolast(__VA_ARGS__)) ___rd_last(__VA_ARGS__)
+#define ___read_ptrs(src, ...)						    \
+	___apply(___rd_p, ___narg(__VA_ARGS__))(src, __VA_ARGS__)
+
+#define ___core_read0(fn, dst, src, a)					    \
+	___read(fn, dst, ___type(src), src, a);
+#define ___core_readN(fn, dst, src, ...)				    \
+	___read_ptrs(src, ___nolast(__VA_ARGS__))			    \
+	___read(fn, dst, ___type(src, ___nolast(__VA_ARGS__)), __t,	    \
+		___last(__VA_ARGS__));
+#define ___core_read(fn, dst, src, a, ...)				    \
+	___apply(___core_read, ___empty(__VA_ARGS__))(fn, dst,		    \
+						      src, a, ##__VA_ARGS__)
+
+/*
+ * BPF_CORE_READ_INTO() is a more performance-conscious variant of
+ * BPF_CORE_READ(), in which final field is read into user-provided storage.
+ * See BPF_CORE_READ() below for more details on general usage.
+ */
+#define BPF_CORE_READ_INTO(dst, src, a, ...)				    \
+	({								    \
+		___core_read(bpf_core_read, dst, src, a, ##__VA_ARGS__)	    \
+	})
+
+/*
+ * BPF_CORE_READ_STR_INTO() does same "pointer chasing" as
+ * BPF_CORE_READ() for intermediate pointers, but then executes (and returns
+ * corresponding error code) bpf_core_read_str() for final string read.
+ */
+#define BPF_CORE_READ_STR_INTO(dst, src, a, ...)			    \
+	({								    \
+		___core_read(bpf_core_read_str, dst, src, a, ##__VA_ARGS__) \
+	})
+
+/*
+ * BPF_CORE_READ() is used to simplify BPF CO-RE relocatable read, especially
+ * when there are few pointer chasing steps.
+ * E.g., what in non-BPF world (or in BPF w/ BCC) would be something like:
+ *	int x = s->a.b.c->d.e->f->g;
+ * can be succinctly achieved using BPF_CORE_READ as:
+ *	int x = BPF_CORE_READ(s, a.b.c, d.e, f, g);
+ *
+ * BPF_CORE_READ will decompose above statement into 4 bpf_core_read (BPF
+ * CO-RE relocatable bpf_probe_read() wrapper) calls, logically equivalent to:
+ * 1. const void *__t = s->a.b.c;
+ * 2. __t = __t->d.e;
+ * 3. __t = __t->f;
+ * 4. return __t->g;
+ *
+ * Equivalence is logical, because there is a heavy type casting/preservation
+ * involved, as well as all the reads are happening through bpf_probe_read()
+ * calls using __builtin_preserve_access_index() to emit CO-RE relocations.
+ *
+ * N.B. Only up to 9 "field accessors" are supported, which should be more
+ * than enough for any practical purpose.
+ */
+#define BPF_CORE_READ(src, a, ...)					    \
+	({								    \
+		___type(src, a, ##__VA_ARGS__) __r;			    \
+		BPF_CORE_READ_INTO(&__r, src, a, ##__VA_ARGS__);	    \
+		__r;							    \
+	})
+
+#endif
+

tools/lib/bpf/bpf_helpers.h

+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+#ifndef __BPF_HELPERS__
+#define __BPF_HELPERS__
+
+#include "bpf_helper_defs.h"
+
+#define __uint(name, val) int (*name)[val]
+#define __type(name, val) typeof(val) *name
+
+/* Helper macro to print out debug messages */
+#define bpf_printk(fmt, ...)				\
+({							\
+	char ____fmt[] = fmt;				\
+	bpf_trace_printk(____fmt, sizeof(____fmt),	\
+			 ##__VA_ARGS__);		\
+})
+
+/*
+ * Helper macro to place programs, maps, license in
+ * different sections in elf_bpf file. Section names
+ * are interpreted by elf_bpf loader
+ */
+#define SEC(NAME) __attribute__((section(NAME), used))
+
+#ifndef __always_inline
+#define __always_inline __attribute__((always_inline))
+#endif
+
+/*
+ * Helper structure used by eBPF C program
+ * to describe BPF map attributes to libbpf loader
+ */
+struct bpf_map_def {
+	unsigned int type;
+	unsigned int key_size;
+	unsigned int value_size;
+	unsigned int max_entries;
+	unsigned int map_flags;
+};
+
+#endif

tools/testing/selftests/bpf/bpf_helpers.h → tools/lib/bpf/bpf_tracing.h

 /* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
-#ifndef __BPF_HELPERS__
-#define __BPF_HELPERS__
-
-#include "bpf_helper_defs.h"
-
-#define __uint(name, val) int (*name)[val]
-#define __type(name, val) typeof(val) *name
-
-/* helper macro to print out debug messages */
-#define bpf_printk(fmt, ...)				\
-({							\
-	char ____fmt[] = fmt;				\
-	bpf_trace_printk(____fmt, sizeof(____fmt),	\
-			 ##__VA_ARGS__);		\
-})
-
-#ifdef __clang__
-
-/* helper macro to place programs, maps, license in
- * different sections in elf_bpf file. Section names
- * are interpreted by elf_bpf loader
- */
-#define SEC(NAME) __attribute__((section(NAME), used))
-
-/* llvm builtin functions that eBPF C program may use to
- * emit BPF_LD_ABS and BPF_LD_IND instructions
- */
-struct sk_buff;
-unsigned long long load_byte(void *skb,
-			     unsigned long long off) asm("llvm.bpf.load.byte");
-unsigned long long load_half(void *skb,
-			     unsigned long long off) asm("llvm.bpf.load.half");
-unsigned long long load_word(void *skb,
-			     unsigned long long off) asm("llvm.bpf.load.word");
-
-/* a helper structure used by eBPF C program
- * to describe map attributes to elf_bpf loader
- */
-struct bpf_map_def {
-	unsigned int type;
-	unsigned int key_size;
-	unsigned int value_size;
-	unsigned int max_entries;
-	unsigned int map_flags;
-	unsigned int inner_map_idx;
-	unsigned int numa_node;
-};
-
-#else
-
-#include <bpf-helpers.h>
-
-#endif
-
-#define BPF_ANNOTATE_KV_PAIR(name, type_key, type_val)		\
-	struct ____btf_map_##name {				\
-		type_key key;					\
-		type_val value;					\
-	};							\
-	struct ____btf_map_##name				\
-	__attribute__ ((section(".maps." #name), used))		\
-		____btf_map_##name = { }
+#ifndef __BPF_TRACING_H__
+#define __BPF_TRACING_H__
 
 /* Scan the ARCH passed in from ARCH env variable (see Makefile) */
 #if defined(__TARGET_ARCH_x86)
@@ -245,31 +185,11 @@ struct pt_regs;
 #define BPF_KPROBE_READ_RET_IP(ip, ctx)		({ (ip) = PT_REGS_RET(ctx); })
 #define BPF_KRETPROBE_READ_RET_IP		BPF_KPROBE_READ_RET_IP
 #else
-#define BPF_KPROBE_READ_RET_IP(ip, ctx)		({				\
-		bpf_probe_read(&(ip), sizeof(ip), (void *)PT_REGS_RET(ctx)); })
-#define BPF_KRETPROBE_READ_RET_IP(ip, ctx)	({				\
-		bpf_probe_read(&(ip), sizeof(ip),				\
-				(void *)(PT_REGS_FP(ctx) + sizeof(ip))); })
+#define BPF_KPROBE_READ_RET_IP(ip, ctx)					    \
+	({ bpf_probe_read(&(ip), sizeof(ip), (void *)PT_REGS_RET(ctx)); })
+#define BPF_KRETPROBE_READ_RET_IP(ip, ctx)				    \
+	({ bpf_probe_read(&(ip), sizeof(ip),				    \
+			  (void *)(PT_REGS_FP(ctx) + sizeof(ip))); })
 #endif
 
-/*
- * BPF_CORE_READ abstracts away bpf_probe_read() call and captures offset
- * relocation for source address using __builtin_preserve_access_index()
- * built-in, provided by Clang.
- *
- * __builtin_preserve_access_index() takes as an argument an expression of
- * taking an address of a field within struct/union. It makes compiler emit
- * a relocation, which records BTF type ID describing root struct/union and an
- * accessor string which describes exact embedded field that was used to take
- * an address. See detailed description of this relocation format and
- * semantics in comments to struct bpf_offset_reloc in libbpf_internal.h.
- *
- * This relocation allows libbpf to adjust BPF instruction to use correct
- * actual field offset, based on target kernel BTF type that matches original
- * (local) BTF, used to record relocation.
- */
-#define BPF_CORE_READ(dst, src)						\
-	bpf_probe_read((dst), sizeof(*(src)),				\
-		       __builtin_preserve_access_index(src))
-
 #endif

tools/testing/selftests/bpf/.gitignore

@@ -39,4 +39,3 @@ libbpf.so.*
 test_hashmap
 test_btf_dump
 xdping
-/bpf_helper_defs.h

tools/testing/selftests/bpf/Makefile

@@ -90,10 +90,6 @@ include ../lib.mk
 TEST_CUSTOM_PROGS = $(OUTPUT)/urandom_read
 all: $(TEST_CUSTOM_PROGS)
 
-bpf_helper_defs.h: $(APIDIR)/linux/bpf.h
-	$(BPFDIR)/../../../scripts/bpf_helpers_doc.py --header 		\
-		--file $(APIDIR)/linux/bpf.h > bpf_helper_defs.h
-
 $(OUTPUT)/urandom_read: $(OUTPUT)/%: %.c
 	$(CC) -o $@ $< -Wl,--build-id
 
@@ -127,7 +123,7 @@ $(OUTPUT)/test_cgroup_attach: cgroup_helpers.c
 # force a rebuild of BPFOBJ when its dependencies are updated
 force:
 
-$(BPFOBJ): force bpf_helper_defs.h
+$(BPFOBJ): force
 	$(MAKE) -C $(BPFDIR) OUTPUT=$(OUTPUT)/
 
 PROBE := $(shell $(LLC) -march=bpf -mcpu=probe -filetype=null /dev/null 2>&1)
@@ -152,7 +148,7 @@ $(shell $(1) -v -E - </dev/null 2>&1 \
 endef
 CLANG_SYS_INCLUDES = $(call get_sys_includes,$(CLANG))
 BPF_CFLAGS = -I. -I./include/uapi -I../../../include/uapi \
-	     -I$(OUTPUT)/../usr/include -D__TARGET_ARCH_$(SRCARCH)
+	     -I$(BPFDIR) -I$(OUTPUT)/../usr/include -D__TARGET_ARCH_$(SRCARCH)
 
 CLANG_CFLAGS = $(CLANG_SYS_INCLUDES) \
 	       -Wno-compare-distinct-pointer-types
@@ -323,4 +319,4 @@ $(VERIFIER_TESTS_H): $(VERIFIER_TEST_FILES) | $(VERIFIER_TESTS_DIR)
 
 EXTRA_CLEAN := $(TEST_CUSTOM_PROGS) $(ALU32_BUILD_DIR) $(BPF_GCC_BUILD_DIR) \
 	$(VERIFIER_TESTS_H) $(PROG_TESTS_H) $(MAP_TESTS_H) \
-	feature bpf_helper_defs.h
+	feature

tools/testing/selftests/bpf/bpf_legacy.h

+/* SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause) */
+#ifndef __BPF_LEGACY__
+#define __BPF_LEGACY__
+
+/*
+ * legacy bpf_map_def with extra fields supported only by bpf_load(), do not
+ * use outside of samples/bpf
+ */
+struct bpf_map_def_legacy {
+	unsigned int type;
+	unsigned int key_size;
+	unsigned int value_size;
+	unsigned int max_entries;
+	unsigned int map_flags;
+	unsigned int inner_map_idx;
+	unsigned int numa_node;
+};
+
+#define BPF_ANNOTATE_KV_PAIR(name, type_key, type_val)		\
+	struct ____btf_map_##name {				\
+		type_key key;					\
+		type_val value;					\
+	};							\
+	struct ____btf_map_##name				\
+	__attribute__ ((section(".maps." #name), used))		\
+		____btf_map_##name = { }
+
+/* llvm builtin functions that eBPF C program may use to
+ * emit BPF_LD_ABS and BPF_LD_IND instructions
+ */
+unsigned long long load_byte(void *skb,
+			     unsigned long long off) asm("llvm.bpf.load.byte");
+unsigned long long load_half(void *skb,
+			     unsigned long long off) asm("llvm.bpf.load.half");
+unsigned long long load_word(void *skb,
+			     unsigned long long off) asm("llvm.bpf.load.word");
+
+#endif
+

tools/testing/selftests/bpf/prog_tests/core_reloc.c

@@ -193,8 +193,12 @@ static struct core_reloc_test_case test_cases[] = {
 		.btf_src_file = NULL, /* load from /lib/modules/$(uname -r) */
 		.input = "",
 		.input_len = 0,
-		.output = "\1", /* true */
-		.output_len = 1,
+		.output = STRUCT_TO_CHAR_PTR(core_reloc_kernel_output) {
+			.valid = { 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, },
+			.comm = "test_progs\0\0\0\0\0",
+			.comm_len = 11,
+		},
+		.output_len = sizeof(struct core_reloc_kernel_output),
 	},
 
 	/* validate BPF program can use multiple flavors to match against

tools/testing/selftests/bpf/progs/core_reloc_types.h

 #include <stdint.h>
 #include <stdbool.h>
+/*
+ * KERNEL
+ */
+
+struct core_reloc_kernel_output {
+	int valid[10];
+	char comm[16];
+	int comm_len;
+};
 
 /*
  * FLAVORS

tools/testing/selftests/bpf/progs/loop1.c

@@ -7,6 +7,7 @@
 #include <stdbool.h>
 #include <linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 char _license[] SEC("license") = "GPL";
 

tools/testing/selftests/bpf/progs/loop2.c

@@ -7,6 +7,7 @@
 #include <stdbool.h>
 #include <linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 char _license[] SEC("license") = "GPL";
 

tools/testing/selftests/bpf/progs/loop3.c

@@ -7,6 +7,7 @@
 #include <stdbool.h>
 #include <linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_tracing.h"
 
 char _license[] SEC("license") = "GPL";
 

tools/testing/selftests/bpf/progs/sockopt_sk.c

@@ -14,13 +14,12 @@ struct sockopt_sk {
 	__u8 val;
 };
 
-struct bpf_map_def SEC("maps") socket_storage_map = {
-	.type = BPF_MAP_TYPE_SK_STORAGE,
-	.key_size = sizeof(int),
-	.value_size = sizeof(struct sockopt_sk),
-	.map_flags = BPF_F_NO_PREALLOC,
-};
-BPF_ANNOTATE_KV_PAIR(socket_storage_map, int, struct sockopt_sk);
+struct {
+	__uint(type, BPF_MAP_TYPE_SK_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct sockopt_sk);
+} socket_storage_map SEC(".maps");
 
 SEC("cgroup/getsockopt")
 int _getsockopt(struct bpf_sockopt *ctx)

tools/testing/selftests/bpf/progs/tcp_rtt.c

@@ -13,13 +13,12 @@ struct tcp_rtt_storage {
 	__u32 icsk_retransmits;
 };
 
-struct bpf_map_def SEC("maps") socket_storage_map = {
-	.type = BPF_MAP_TYPE_SK_STORAGE,
-	.key_size = sizeof(int),
-	.value_size = sizeof(struct tcp_rtt_storage),
-	.map_flags = BPF_F_NO_PREALLOC,
-};
-BPF_ANNOTATE_KV_PAIR(socket_storage_map, int, struct tcp_rtt_storage);
+struct {
+	__uint(type, BPF_MAP_TYPE_SK_STORAGE);
+	__uint(map_flags, BPF_F_NO_PREALLOC);
+	__type(key, int);
+	__type(value, struct tcp_rtt_storage);
+} socket_storage_map SEC(".maps");
 
 SEC("sockops")
 int _sockops(struct bpf_sock_ops *ctx)

tools/testing/selftests/bpf/progs/test_btf_haskv.c

@@ -2,6 +2,7 @@
 /* Copyright (c) 2018 Facebook */
 #include <linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_legacy.h"
 
 int _version SEC("version") = 1;
 

tools/testing/selftests/bpf/progs/test_btf_newkv.c

@@ -2,6 +2,7 @@
 /* Copyright (c) 2018 Facebook */
 #include <linux/bpf.h>
 #include "bpf_helpers.h"
+#include "bpf_legacy.h"
 
 int _version SEC("version") = 1;
 

tools/testing/selftests/bpf/progs/test_core_reloc_arrays.c

@@ -4,6 +4,7 @@
 #include <linux/bpf.h>
 #include <stdint.h>
 #include "bpf_helpers.h"
+#include "bpf_core_read.h"
 
 char _license[] SEC("license") = "GPL";
 
@@ -31,6 +32,8 @@ struct core_reloc_arrays {
 	struct core_reloc_arrays_substruct d[1][2];
 };
 
+#define CORE_READ(dst, src) bpf_core_read(dst, sizeof(*(dst)), src)
+
 SEC("raw_tracepoint/sys_enter")
 int test_core_arrays(void *ctx)
 {
@@ -38,16 +41,16 @@ int test_core_arrays(void *ctx)
 	struct core_reloc_arrays_output *out = (void *)&data.out;
 
 	/* in->a[2] */
-	if (BPF_CORE_READ(&out->a2, &in->a[2]))
+	if (CORE_READ(&out->a2, &in->a[2]))
 		return 1;
 	/* in->b[1][2][3] */
-	if (BPF_CORE_READ(&out->b123, &in->b[1][2][3]))
+	if (CORE_READ(&out->b123, &in->b[1][2][3]))
 		return 1;
 	/* in->c[1].c */
-	if (BPF_CORE_READ(&out->c1c, &in->c[1].c))
+	if (CORE_READ(&out->c1c, &in->c[1].c))
 		return 1;
 	/* in->d[0][0].d */
-	if (BPF_CORE_READ(&out->d00d, &in->d[0][0].d))
+	if (CORE_READ(&out->d00d, &in->d[0][0].d))
 		return 1;
 
 	return 0;

tools/testing/selftests/bpf/progs/test_core_reloc_flavors.c

@@ -4,6 +4,7 @@
 #include <linux/bpf.h>
 #include <stdint.h>
 #include "bpf_helpers.h"
+#include "bpf_core_read.h"
 
 char _license[] SEC("license") = "GPL";
 
@@ -39,6 +40,8 @@ struct core_reloc_flavors___weird {
 	};
 };
 
+#define CORE_READ(dst, src) bpf_core_read(dst, sizeof(*(dst)), src)
+
 SEC("raw_tracepoint/sys_enter")
 int test_core_flavors(void *ctx)
 {
@@ -48,13 +51,13 @@ int test_core_flavors(void *ctx)
 	struct core_reloc_flavors *out = (void *)&data.out;
 
 	/* read a using weird layout */
-	if (BPF_CORE_READ(&out->a, &in_weird->a))
+	if (CORE_READ(&out->a, &in_weird->a))
 		return 1;
 	/* read b using reversed layout */
-	if (BPF_CORE_READ(&out->b, &in_rev->b))
+	if (CORE_READ(&out->b, &in_rev->b))
 		return 1;
 	/* read c using original layout */
-	if (BPF_CORE_READ(&out->c, &in_orig->c))
+	if (CORE_READ(&out->c, &in_orig->c))
 		return 1;
 
 	return 0;

tools/testing/selftests/bpf/progs/test_core_reloc_ints.c

@@ -4,6 +4,7 @@
 #include <linux/bpf.h>
 #include <stdint.h>
 #include "bpf_helpers.h"
+#include "bpf_core_read.h"
 
 char _license[] SEC("license") = "GPL";
 
@@ -23,20 +24,22 @@ struct core_reloc_ints {
 	int64_t		s64_field;
 };
 
+#define CORE_READ(dst, src) bpf_core_read(dst, sizeof(*(dst)), src)
+
 SEC("raw_tracepoint/sys_enter")
 int test_core_ints(void *ctx)
 {
 	struct core_reloc_ints *in = (void *)&data.in;
 	struct core_reloc_ints *out = (void *)&data.out;
 
-	if (BPF_CORE_READ(&out->u8_field, &in->u8_field) ||
-	    BPF_CORE_READ(&out->s8_field, &in->s8_field) ||
-	    BPF_CORE_READ(&out->u16_field, &in->u16_field) ||
-	    BPF_CORE_READ(&out->s16_field, &in->s16_field) ||
-	    BPF_CORE_READ(&out->u32_field, &in->u32_field) ||
-	    BPF_CORE_READ(&out->s32_field, &in->s32_field) ||
-	    BPF_CORE_READ(&out->u64_field, &in->u64_field) ||
-	    BPF_CORE_READ(&out->s64_field, &in->s64_field))
+	if (CORE_READ(&out->u8_field, &in->u8_field) ||
+	    CORE_READ(&out->s8_field, &in->s8_field) ||
+	    CORE_READ(&out->u16_field, &in->u16_field) ||
+	    CORE_READ(&out->s16_field, &in->s16_field) ||
+	    CORE_READ(&out->u32_field, &in->u32_field) ||
+	    CORE_READ(&out->s32_field, &in->s32_field) ||
+	    CORE_READ(&out->u64_field, &in->u64_field) ||
+	    CORE_READ(&out->s64_field, &in->s64_field))
 		return 1;
 
 	return 0;

tools/testing/selftests/bpf/progs/test_core_reloc_kernel.c

@@ -4,6 +4,7 @@
 #include <linux/bpf.h>
 #include <stdint.h>
 #include "bpf_helpers.h"
+#include "bpf_core_read.h"
 
 char _license[] SEC("license") = "GPL";
 
@@ -12,24 +13,78 @@ static volatile struct data {
 	char out[256];
 } data;
 
+struct core_reloc_kernel_output {
+	int valid[10];
+	char comm[16];
+	int comm_len;
+};
+
 struct task_struct {
 	int pid;
 	int tgid;
+	char comm[16];
+	struct task_struct *group_leader;
 };
 
+#define CORE_READ(dst, src) bpf_core_read(dst, sizeof(*(dst)), src)
+
 SEC("raw_tracepoint/sys_enter")
 int test_core_kernel(void *ctx)
 {
 	struct task_struct *task = (void *)bpf_get_current_task();
+	struct core_reloc_kernel_output *out = (void *)&data.out;
 	uint64_t pid_tgid = bpf_get_current_pid_tgid();
+	uint32_t real_tgid = (uint32_t)pid_tgid;
 	int pid, tgid;
 
-	if (BPF_CORE_READ(&pid, &task->pid) ||
-	    BPF_CORE_READ(&tgid, &task->tgid))
+	if (CORE_READ(&pid, &task->pid) ||
+	    CORE_READ(&tgid, &task->tgid))
 		return 1;
 
 	/* validate pid + tgid matches */
-	data.out[0] = (((uint64_t)pid << 32) | tgid) == pid_tgid;
+	out->valid[0] = (((uint64_t)pid << 32) | tgid) == pid_tgid;
+
+	/* test variadic BPF_CORE_READ macros */
+	out->valid[1] = BPF_CORE_READ(task,
+				      tgid) == real_tgid;
+	out->valid[2] = BPF_CORE_READ(task,
+				      group_leader,
+				      tgid) == real_tgid;
+	out->valid[3] = BPF_CORE_READ(task,
+				      group_leader, group_leader,
+				      tgid) == real_tgid;
+	out->valid[4] = BPF_CORE_READ(task,
+				      group_leader, group_leader, group_leader,
+				      tgid) == real_tgid;
+	out->valid[5] = BPF_CORE_READ(task,
+				      group_leader, group_leader, group_leader,
+				      group_leader,
+				      tgid) == real_tgid;
+	out->valid[6] = BPF_CORE_READ(task,
+				      group_leader, group_leader, group_leader,
+				      group_leader, group_leader,
+				      tgid) == real_tgid;
+	out->valid[7] = BPF_CORE_READ(task,
+				      group_leader, group_leader, group_leader,
+				      group_leader, group_leader, group_leader,
+				      tgid) == real_tgid;
+	out->valid[8] = BPF_CORE_READ(task,
+				      group_leader, group_leader, group_leader,
+				      group_leader, group_leader, group_leader,
+				      group_leader,
+				      tgid) == real_tgid;
+	out->valid[9] = BPF_CORE_READ(task,
+				      group_leader, group_leader, group_leader,
+				      group_leader, group_leader, group_leader,
+				      group_leader, group_leader,
+				      tgid) == real_tgid;
+
+	/* test BPF_CORE_READ_STR_INTO() returns correct code and contents */
+	out->comm_len = BPF_CORE_READ_STR_INTO(
+		&out->comm, task,
+		group_leader, group_leader, group_leader, group_leader,
+		group_leader, group_leader, group_leader, group_leader,
+		comm);
 
 	return 0;
 }

tools/testing/selftests/bpf/progs/test_core_reloc_misc.c

@@ -4,6 +4,7 @@
 #include <linux/bpf.h>
 #include <stdint.h>
 #include "bpf_helpers.h"
+#include "bpf_core_read.h"
 
 char _license[] SEC("license") = "GPL";
 
@@ -32,6 +33,8 @@ struct core_reloc_misc_extensible {
 	int b;
 };
 
+#define CORE_READ(dst, src) bpf_core_read(dst, sizeof(*(dst)), src)
+
 SEC("raw_tracepoint/sys_enter")
 int test_core_misc(void *ctx)
 {
@@ -41,15 +44,15 @@ int test_core_misc(void *ctx)
 	struct core_reloc_misc_output *out = (void *)&data.out;
 
 	/* record two different relocations with the same accessor string */
-	if (BPF_CORE_READ(&out->a, &in_a->a1) ||	/* accessor: 0:0 */
-	    BPF_CORE_READ(&out->b, &in_b->b1))		/* accessor: 0:0 */
+	if (CORE_READ(&out->a, &in_a->a1) ||		/* accessor: 0:0 */
+	    CORE_READ(&out->b, &in_b->b1))		/* accessor: 0:0 */
 		return 1;
 
 	/* Validate relocations capture array-only accesses for structs with
 	 * fixed header, but with potentially extendable tail. This will read
 	 * first 4 bytes of 2nd element of in_ext array of potentially
 	 * variably sized struct core_reloc_misc_extensible. */ 
-	if (BPF_CORE_READ(&out->c, &in_ext[2]))		/* accessor: 2 */
+	if (CORE_READ(&out->c, &in_ext[2]))		/* accessor: 2 */
 		return 1;
 
 	return 0;

tools/testing/selftests/bpf/progs/test_core_reloc_mods.c

@@ -4,6 +4,7 @@
 #include <linux/bpf.h>
 #include <stdint.h>
 #include "bpf_helpers.h"
+#include "bpf_core_read.h"
 
 char _license[] SEC("license") = "GPL";
 
@@ -41,20 +42,22 @@ struct core_reloc_mods {
 	core_reloc_mods_substruct_t h;
 };
 
+#define CORE_READ(dst, src) bpf_core_read(dst, sizeof(*(dst)), src)
+
 SEC("raw_tracepoint/sys_enter")
 int test_core_mods(void *ctx)
 {
 	struct core_reloc_mods *in = (void *)&data.in;
 	struct core_reloc_mods_output *out = (void *)&data.out;
 
-	if (BPF_CORE_READ(&out->a, &in->a) ||
-	    BPF_CORE_READ(&out->b, &in->b) ||
-	    BPF_CORE_READ(&out->c, &in->c) ||
-	    BPF_CORE_READ(&out->d, &in->d) ||
-	    BPF_CORE_READ(&out->e, &in->e[2]) ||
-	    BPF_CORE_READ(&out->f, &in->f[1]) ||
-	    BPF_CORE_READ(&out->g, &in->g.x) ||
-	    BPF_CORE_READ(&out->h, &in->h.y))
+	if (CORE_READ(&out->a, &in->a) ||
+	    CORE_READ(&out->b, &in->b) ||
+	    CORE_READ(&out->c, &in->c) ||
+	    CORE_READ(&out->d, &in->d) ||
+	    CORE_READ(&out->e, &in->e[2]) ||
+	    CORE_READ(&out->f, &in->f[1]) ||
+	    CORE_READ(&out->g, &in->g.x) ||
+	    CORE_READ(&out->h, &in->h.y))
 		return 1;
 
 	return 0;

tools/testing/selftests/bpf/progs/test_core_reloc_nesting.c

@@ -4,6 +4,7 @@
 #include <linux/bpf.h>
 #include <stdint.h>
 #include "bpf_helpers.h"
+#include "bpf_core_read.h"
 
 char _license[] SEC("license") = "GPL";
 
@@ -30,15 +31,17 @@ struct core_reloc_nesting {
 	} b;
 };
 
+#define CORE_READ(dst, src) bpf_core_read(dst, sizeof(*(dst)), src)
+
 SEC("raw_tracepoint/sys_enter")
 int test_core_nesting(void *ctx)
 {
 	struct core_reloc_nesting *in = (void *)&data.in;
 	struct core_reloc_nesting *out = (void *)&data.out;
 
-	if (BPF_CORE_READ(&out->a.a.a, &in->a.a.a))
+	if (CORE_READ(&out->a.a.a, &in->a.a.a))
 		return 1;
-	if (BPF_CORE_READ(&out->b.b.b, &in->b.b.b))
+	if (CORE_READ(&out->b.b.b, &in->b.b.b))
 		return 1;
 
 	return 0;

tools/testing/selftests/bpf/progs/test_core_reloc_primitives.c

@@ -4,6 +4,7 @@
 #include <linux/bpf.h>
 #include <stdint.h>
 #include "bpf_helpers.h"
+#include "bpf_core_read.h"
 
 char _license[] SEC("license") = "GPL";
 
@@ -25,17 +26,19 @@ struct core_reloc_primitives {
 	int (*f)(const char *);
 };
 
+#define CORE_READ(dst, src) bpf_core_read(dst, sizeof(*(dst)), src)
+
 SEC("raw_tracepoint/sys_enter")
 int test_core_primitives(void *ctx)
 {
 	struct core_reloc_primitives *in = (void *)&data.in;
 	struct core_reloc_primitives *out = (void *)&data.out;
 
-	if (BPF_CORE_READ(&out->a, &in->a) ||
-	    BPF_CORE_READ(&out->b, &in->b) ||
-	    BPF_CORE_READ(&out->c, &in->c) ||
-	    BPF_CORE_READ(&out->d, &in->d) ||
-	    BPF_CORE_READ(&out->f, &in->f))
+	if (CORE_READ(&out->a, &in->a) ||
+	    CORE_READ(&out->b, &in->b) ||
+	    CORE_READ(&out->c, &in->c) ||
+	    CORE_READ(&out->d, &in->d) ||
+	    CORE_READ(&out->f, &in->f))
 		return 1;
 
 	return 0;

tools/testing/selftests/bpf/progs/test_core_reloc_ptr_as_arr.c

@@ -4,6 +4,7 @@
 #include <linux/bpf.h>
 #include <stdint.h>
 #include "bpf_helpers.h"
+#include "bpf_core_read.h"
 
 char _license[] SEC("license") = "GPL";
 
@@ -16,13 +17,15 @@ struct core_reloc_ptr_as_arr {
 	int a;
 };
 
+#define CORE_READ(dst, src) bpf_core_read(dst, sizeof(*(dst)), src)
+
 SEC("raw_tracepoint/sys_enter")
 int test_core_ptr_as_arr(void *ctx)
 {
 	struct core_reloc_ptr_as_arr *in = (void *)&data.in;
 	struct core_reloc_ptr_as_arr *out = (void *)&data.out;
 
-	if (BPF_CORE_READ(&out->a, &in[2].a))
+	if (CORE_READ(&out->a, &in[2].a))
 		return 1;
 
 	return 0;