Merge remote-tracking branch 'asoc/for-5.10' into asoc-linus

23 files changed, 4727 insertions, 1573 deletions

diff --git a/tools/lib/bpf/Makefile b/tools/lib/bpf/Makefile
index 9ae8f4ef0aac..5f9abed3e226 100644
--- a/tools/lib/bpf/Makefile
+++ b/tools/lib/bpf/Makefile
@@ -1,6 +1,9 @@
 # SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
 # Most of this file is copied from tools/lib/traceevent/Makefile
 
+RM ?= rm
+srctree = $(abs_srctree)
+
 LIBBPF_VERSION := $(shell \
 	grep -oE '^LIBBPF_([0-9.]+)' libbpf.map | \
 	sort -rV | head -n1 | cut -d'_' -f2)
@@ -56,7 +59,7 @@ ifndef VERBOSE
 endif
 
 FEATURE_USER = .libbpf
-FEATURE_TESTS = libelf libelf-mmap zlib bpf reallocarray
+FEATURE_TESTS = libelf zlib bpf
 FEATURE_DISPLAY = libelf zlib bpf
 
 INCLUDES = -I. -I$(srctree)/tools/include -I$(srctree)/tools/include/uapi
@@ -95,27 +98,18 @@ PC_FILE		= libbpf.pc
 ifdef EXTRA_CFLAGS
   CFLAGS := $(EXTRA_CFLAGS)
 else
-  CFLAGS := -g -Wall
-endif
-
-ifeq ($(feature-libelf-mmap), 1)
-  override CFLAGS += -DHAVE_LIBELF_MMAP_SUPPORT
-endif
-
-ifeq ($(feature-reallocarray), 0)
-  override CFLAGS += -DCOMPAT_NEED_REALLOCARRAY
+  CFLAGS := -g -O2
 endif
 
 # Append required CFLAGS
-override CFLAGS += $(EXTRA_WARNINGS)
+override CFLAGS += $(EXTRA_WARNINGS) -Wno-switch-enum
 override CFLAGS += -Werror -Wall
-override CFLAGS += -fPIC
 override CFLAGS += $(INCLUDES)
 override CFLAGS += -fvisibility=hidden
 override CFLAGS += -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
 
 # flags specific for shared library
-SHLIB_FLAGS := -DSHARED
+SHLIB_FLAGS := -DSHARED -fPIC
 
 ifeq ($(VERBOSE),1)
   Q =
@@ -197,7 +191,7 @@ $(OUTPUT)libbpf.so.$(LIBBPF_VERSION): $(BPF_IN_SHARED)
 	@ln -sf $(@F) $(OUTPUT)libbpf.so.$(LIBBPF_MAJOR_VERSION)
 
 $(OUTPUT)libbpf.a: $(BPF_IN_STATIC)
-	$(QUIET_LINK)$(RM) $@; $(AR) rcs $@ $^
+	$(QUIET_LINK)$(RM) -f $@; $(AR) rcs $@ $^
 
 $(OUTPUT)libbpf.pc:
 	$(QUIET_GEN)sed -e "s|@PREFIX@|$(prefix)|" \
@@ -271,10 +265,10 @@ install: install_lib install_pkgconfig install_headers
 ### Cleaning rules
 
 config-clean:
-	$(call QUIET_CLEAN, config)
+	$(call QUIET_CLEAN, feature-detect)
 	$(Q)$(MAKE) -C $(srctree)/tools/build/feature/ clean >/dev/null
 
-clean:
+clean: config-clean
 	$(call QUIET_CLEAN, libbpf) $(RM) -rf $(CMD_TARGETS)		     \
 		*~ .*.d .*.cmd LIBBPF-CFLAGS $(BPF_HELPER_DEFS)		     \
 		$(SHARED_OBJDIR) $(STATIC_OBJDIR)			     \
@@ -301,7 +295,7 @@ cscope:
 	cscope -b -q -I $(srctree)/include -f cscope.out
 
 tags:
-	rm -f TAGS tags
+	$(RM) -f TAGS tags
 	ls *.c *.h | xargs $(TAGS_PROG) -a
 
 # Declare the contents of the .PHONY variable as phony.  We keep that
diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
index 0750681057c2..d27e34133973 100644
--- a/tools/lib/bpf/bpf.c
+++ b/tools/lib/bpf/bpf.c
@@ -32,9 +32,6 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
-/* make sure libbpf doesn't use kernel-only integer typedefs */
-#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
-
 /*
  * When building perf, unistd.h is overridden. __NR_bpf is
  * required to be defined explicitly.
@@ -589,19 +586,31 @@ int bpf_link_create(int prog_fd, int target_fd,
 		    enum bpf_attach_type attach_type,
 		    const struct bpf_link_create_opts *opts)
 {
+	__u32 target_btf_id, iter_info_len;
 	union bpf_attr attr;
 
 	if (!OPTS_VALID(opts, bpf_link_create_opts))
 		return -EINVAL;
 
+	iter_info_len = OPTS_GET(opts, iter_info_len, 0);
+	target_btf_id = OPTS_GET(opts, target_btf_id, 0);
+
+	if (iter_info_len && target_btf_id)
+		return -EINVAL;
+
 	memset(&attr, 0, sizeof(attr));
 	attr.link_create.prog_fd = prog_fd;
 	attr.link_create.target_fd = target_fd;
 	attr.link_create.attach_type = attach_type;
 	attr.link_create.flags = OPTS_GET(opts, flags, 0);
-	attr.link_create.iter_info =
-		ptr_to_u64(OPTS_GET(opts, iter_info, (void *)0));
-	attr.link_create.iter_info_len = OPTS_GET(opts, iter_info_len, 0);
+
+	if (iter_info_len) {
+		attr.link_create.iter_info =
+			ptr_to_u64(OPTS_GET(opts, iter_info, (void *)0));
+		attr.link_create.iter_info_len = iter_info_len;
+	} else if (target_btf_id) {
+		attr.link_create.target_btf_id = target_btf_id;
+	}
 
 	return sys_bpf(BPF_LINK_CREATE, &attr, sizeof(attr));
 }
@@ -715,6 +724,37 @@ int bpf_prog_test_run_xattr(struct bpf_prog_test_run_attr *test_attr)
 	return ret;
 }
 
+int bpf_prog_test_run_opts(int prog_fd, struct bpf_test_run_opts *opts)
+{
+	union bpf_attr attr;
+	int ret;
+
+	if (!OPTS_VALID(opts, bpf_test_run_opts))
+		return -EINVAL;
+
+	memset(&attr, 0, sizeof(attr));
+	attr.test.prog_fd = prog_fd;
+	attr.test.cpu = OPTS_GET(opts, cpu, 0);
+	attr.test.flags = OPTS_GET(opts, flags, 0);
+	attr.test.repeat = OPTS_GET(opts, repeat, 0);
+	attr.test.duration = OPTS_GET(opts, duration, 0);
+	attr.test.ctx_size_in = OPTS_GET(opts, ctx_size_in, 0);
+	attr.test.ctx_size_out = OPTS_GET(opts, ctx_size_out, 0);
+	attr.test.data_size_in = OPTS_GET(opts, data_size_in, 0);
+	attr.test.data_size_out = OPTS_GET(opts, data_size_out, 0);
+	attr.test.ctx_in = ptr_to_u64(OPTS_GET(opts, ctx_in, NULL));
+	attr.test.ctx_out = ptr_to_u64(OPTS_GET(opts, ctx_out, NULL));
+	attr.test.data_in = ptr_to_u64(OPTS_GET(opts, data_in, NULL));
+	attr.test.data_out = ptr_to_u64(OPTS_GET(opts, data_out, NULL));
+
+	ret = sys_bpf(BPF_PROG_TEST_RUN, &attr, sizeof(attr));
+	OPTS_SET(opts, data_size_out, attr.test.data_size_out);
+	OPTS_SET(opts, ctx_size_out, attr.test.ctx_size_out);
+	OPTS_SET(opts, duration, attr.test.duration);
+	OPTS_SET(opts, retval, attr.test.retval);
+	return ret;
+}
+
 static int bpf_obj_get_next_id(__u32 start_id, __u32 *next_id, int cmd)
 {
 	union bpf_attr attr;
@@ -818,7 +858,7 @@ int bpf_raw_tracepoint_open(const char *name, int prog_fd)
 	return sys_bpf(BPF_RAW_TRACEPOINT_OPEN, &attr, sizeof(attr));
 }
 
-int bpf_load_btf(void *btf, __u32 btf_size, char *log_buf, __u32 log_buf_size,
+int bpf_load_btf(const void *btf, __u32 btf_size, char *log_buf, __u32 log_buf_size,
 		 bool do_log)
 {
 	union bpf_attr attr = {};
@@ -875,3 +915,19 @@ int bpf_enable_stats(enum bpf_stats_type type)
 
 	return sys_bpf(BPF_ENABLE_STATS, &attr, sizeof(attr));
 }
+
+int bpf_prog_bind_map(int prog_fd, int map_fd,
+		      const struct bpf_prog_bind_opts *opts)
+{
+	union bpf_attr attr;
+
+	if (!OPTS_VALID(opts, bpf_prog_bind_opts))
+		return -EINVAL;
+
+	memset(&attr, 0, sizeof(attr));
+	attr.prog_bind_map.prog_fd = prog_fd;
+	attr.prog_bind_map.map_fd = map_fd;
+	attr.prog_bind_map.flags = OPTS_GET(opts, flags, 0);
+
+	return sys_bpf(BPF_PROG_BIND_MAP, &attr, sizeof(attr));
+}
diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
index 015d13f25fcc..875dde20d56e 100644
--- a/tools/lib/bpf/bpf.h
+++ b/tools/lib/bpf/bpf.h
@@ -174,8 +174,9 @@ struct bpf_link_create_opts {
 	__u32 flags;
 	union bpf_iter_link_info *iter_info;
 	__u32 iter_info_len;
+	__u32 target_btf_id;
 };
-#define bpf_link_create_opts__last_field iter_info_len
+#define bpf_link_create_opts__last_field target_btf_id
 
 LIBBPF_API int bpf_link_create(int prog_fd, int target_fd,
 			       enum bpf_attach_type attach_type,
@@ -234,7 +235,7 @@ LIBBPF_API int bpf_prog_query(int target_fd, enum bpf_attach_type type,
 			      __u32 query_flags, __u32 *attach_flags,
 			      __u32 *prog_ids, __u32 *prog_cnt);
 LIBBPF_API int bpf_raw_tracepoint_open(const char *name, int prog_fd);
-LIBBPF_API int bpf_load_btf(void *btf, __u32 btf_size, char *log_buf,
+LIBBPF_API int bpf_load_btf(const void *btf, __u32 btf_size, char *log_buf,
 			    __u32 log_buf_size, bool do_log);
 LIBBPF_API int bpf_task_fd_query(int pid, int fd, __u32 flags, char *buf,
 				 __u32 *buf_len, __u32 *prog_id, __u32 *fd_type,
@@ -243,6 +244,40 @@ LIBBPF_API int bpf_task_fd_query(int pid, int fd, __u32 flags, char *buf,
 enum bpf_stats_type; /* defined in up-to-date linux/bpf.h */
 LIBBPF_API int bpf_enable_stats(enum bpf_stats_type type);
 
+struct bpf_prog_bind_opts {
+	size_t sz; /* size of this struct for forward/backward compatibility */
+	__u32 flags;
+};
+#define bpf_prog_bind_opts__last_field flags
+
+LIBBPF_API int bpf_prog_bind_map(int prog_fd, int map_fd,
+				 const struct bpf_prog_bind_opts *opts);
+
+struct bpf_test_run_opts {
+	size_t sz; /* size of this struct for forward/backward compatibility */
+	const void *data_in; /* optional */
+	void *data_out;      /* optional */
+	__u32 data_size_in;
+	__u32 data_size_out; /* in: max length of data_out
+			      * out: length of data_out
+			      */
+	const void *ctx_in; /* optional */
+	void *ctx_out;      /* optional */
+	__u32 ctx_size_in;
+	__u32 ctx_size_out; /* in: max length of ctx_out
+			     * out: length of cxt_out
+			     */
+	__u32 retval;        /* out: return code of the BPF program */
+	int repeat;
+	__u32 duration;      /* out: average per repetition in ns */
+	__u32 flags;
+	__u32 cpu;
+};
+#define bpf_test_run_opts__last_field cpu
+
+LIBBPF_API int bpf_prog_test_run_opts(int prog_fd,
+				      struct bpf_test_run_opts *opts);
+
 #ifdef __cplusplus
 } /* extern "C" */
 #endif
diff --git a/tools/lib/bpf/bpf_core_read.h b/tools/lib/bpf/bpf_core_read.h
index eae5cccff761..bbcefb3ff5a5 100644
--- a/tools/lib/bpf/bpf_core_read.h
+++ b/tools/lib/bpf/bpf_core_read.h
@@ -19,32 +19,52 @@ enum bpf_field_info_kind {
 	BPF_FIELD_RSHIFT_U64 = 5,
 };
 
+/* second argument to __builtin_btf_type_id() built-in */
+enum bpf_type_id_kind {
+	BPF_TYPE_ID_LOCAL = 0,		/* BTF type ID in local program */
+	BPF_TYPE_ID_TARGET = 1,		/* BTF type ID in target kernel */
+};
+
+/* second argument to __builtin_preserve_type_info() built-in */
+enum bpf_type_info_kind {
+	BPF_TYPE_EXISTS = 0,		/* type existence in target kernel */
+	BPF_TYPE_SIZE = 1,		/* type size in target kernel */
+};
+
+/* second argument to __builtin_preserve_enum_value() built-in */
+enum bpf_enum_value_kind {
+	BPF_ENUMVAL_EXISTS = 0,		/* enum value existence in kernel */
+	BPF_ENUMVAL_VALUE = 1,		/* enum value value relocation */
+};
+
 #define __CORE_RELO(src, field, info)					      \
 	__builtin_preserve_field_info((src)->field, BPF_FIELD_##info)
 
 #if __BYTE_ORDER == __LITTLE_ENDIAN
 #define __CORE_BITFIELD_PROBE_READ(dst, src, fld)			      \
-	bpf_probe_read((void *)dst,					      \
-		       __CORE_RELO(src, fld, BYTE_SIZE),		      \
-		       (const void *)src + __CORE_RELO(src, fld, BYTE_OFFSET))
+	bpf_probe_read_kernel(						      \
+			(void *)dst,				      \
+			__CORE_RELO(src, fld, BYTE_SIZE),		      \
+			(const void *)src + __CORE_RELO(src, fld, BYTE_OFFSET))
 #else
 /* semantics of LSHIFT_64 assumes loading values into low-ordered bytes, so
  * for big-endian we need to adjust destination pointer accordingly, based on
  * field byte size
  */
 #define __CORE_BITFIELD_PROBE_READ(dst, src, fld)			      \
-	bpf_probe_read((void *)dst + (8 - __CORE_RELO(src, fld, BYTE_SIZE)),  \
-		       __CORE_RELO(src, fld, BYTE_SIZE),		      \
-		       (const void *)src + __CORE_RELO(src, fld, BYTE_OFFSET))
+	bpf_probe_read_kernel(						      \
+			(void *)dst + (8 - __CORE_RELO(src, fld, BYTE_SIZE)), \
+			__CORE_RELO(src, fld, BYTE_SIZE),		      \
+			(const void *)src + __CORE_RELO(src, fld, BYTE_OFFSET))
 #endif
 
 /*
  * Extract bitfield, identified by s->field, and return its value as u64.
  * All this is done in relocatable manner, so bitfield changes such as
  * signedness, bit size, offset changes, this will be handled automatically.
- * This version of macro is using bpf_probe_read() to read underlying integer
- * storage. Macro functions as an expression and its return type is
- * bpf_probe_read()'s return value: 0, on success, <0 on error.
+ * This version of macro is using bpf_probe_read_kernel() to read underlying
+ * integer storage. Macro functions as an expression and its return type is
+ * bpf_probe_read_kernel()'s return value: 0, on success, <0 on error.
  */
 #define BPF_CORE_READ_BITFIELD_PROBED(s, field) ({			      \
 	unsigned long long val = 0;					      \
@@ -92,15 +112,75 @@ enum bpf_field_info_kind {
 	__builtin_preserve_field_info(field, BPF_FIELD_EXISTS)
 
 /*
- * Convenience macro to get byte size of a field. Works for integers,
+ * Convenience macro to get the byte size of a field. Works for integers,
  * struct/unions, pointers, arrays, and enums.
  */
 #define bpf_core_field_size(field)					    \
 	__builtin_preserve_field_info(field, BPF_FIELD_BYTE_SIZE)
 
 /*
- * bpf_core_read() abstracts away bpf_probe_read() call and captures offset
- * relocation for source address using __builtin_preserve_access_index()
+ * Convenience macro to get BTF type ID of a specified type, using a local BTF
+ * information. Return 32-bit unsigned integer with type ID from program's own
+ * BTF. Always succeeds.
+ */
+#define bpf_core_type_id_local(type)					    \
+	__builtin_btf_type_id(*(typeof(type) *)0, BPF_TYPE_ID_LOCAL)
+
+/*
+ * Convenience macro to get BTF type ID of a target kernel's type that matches
+ * specified local type.
+ * Returns:
+ *    - valid 32-bit unsigned type ID in kernel BTF;
+ *    - 0, if no matching type was found in a target kernel BTF.
+ */
+#define bpf_core_type_id_kernel(type)					    \
+	__builtin_btf_type_id(*(typeof(type) *)0, BPF_TYPE_ID_TARGET)
+
+/*
+ * Convenience macro to check that provided named type
+ * (struct/union/enum/typedef) exists in a target kernel.
+ * Returns:
+ *    1, if such type is present in target kernel's BTF;
+ *    0, if no matching type is found.
+ */
+#define bpf_core_type_exists(type)					    \
+	__builtin_preserve_type_info(*(typeof(type) *)0, BPF_TYPE_EXISTS)
+
+/*
+ * Convenience macro to get the byte size of a provided named type
+ * (struct/union/enum/typedef) in a target kernel.
+ * Returns:
+ *    >= 0 size (in bytes), if type is present in target kernel's BTF;
+ *    0, if no matching type is found.
+ */
+#define bpf_core_type_size(type)					    \
+	__builtin_preserve_type_info(*(typeof(type) *)0, BPF_TYPE_SIZE)
+
+/*
+ * Convenience macro to check that provided enumerator value is defined in
+ * a target kernel.
+ * Returns:
+ *    1, if specified enum type and its enumerator value are present in target
+ *    kernel's BTF;
+ *    0, if no matching enum and/or enum value within that enum is found.
+ */
+#define bpf_core_enum_value_exists(enum_type, enum_value)		    \
+	__builtin_preserve_enum_value(*(typeof(enum_type) *)enum_value, BPF_ENUMVAL_EXISTS)
+
+/*
+ * Convenience macro to get the integer value of an enumerator value in
+ * a target kernel.
+ * Returns:
+ *    64-bit value, if specified enum type and its enumerator value are
+ *    present in target kernel's BTF;
+ *    0, if no matching enum and/or enum value within that enum is found.
+ */
+#define bpf_core_enum_value(enum_type, enum_value)			    \
+	__builtin_preserve_enum_value(*(typeof(enum_type) *)enum_value, BPF_ENUMVAL_VALUE)
+
+/*
+ * bpf_core_read() abstracts away bpf_probe_read_kernel() 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
@@ -115,8 +195,8 @@ enum bpf_field_info_kind {
  * (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_probe_read_kernel(dst, sz,					    \
+			      (const void *)__builtin_preserve_access_index(src))
 
 /*
  * bpf_core_read_str() is a thin wrapper around bpf_probe_read_str()
@@ -124,8 +204,8 @@ enum bpf_field_info_kind {
  * argument.
  */
 #define bpf_core_read_str(dst, sz, src)					    \
-	bpf_probe_read_str(dst, sz,					    \
-			   (const void *)__builtin_preserve_access_index(src))
+	bpf_probe_read_kernel_str(dst, sz,				    \
+				  (const void *)__builtin_preserve_access_index(src))
 
 #define ___concat(a, b) a ## b
 #define ___apply(fn, n) ___concat(fn, n)
@@ -239,15 +319,17 @@ enum bpf_field_info_kind {
  *	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:
+ * CO-RE relocatable bpf_probe_read_kernel() 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.
+ * involved, as well as all the reads are happening through
+ * bpf_probe_read_kernel() 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.
diff --git a/tools/lib/bpf/bpf_helpers.h b/tools/lib/bpf/bpf_helpers.h
index e9a4ecddb7a5..72b251110c4d 100644
--- a/tools/lib/bpf/bpf_helpers.h
+++ b/tools/lib/bpf/bpf_helpers.h
@@ -32,6 +32,9 @@
 #ifndef __always_inline
 #define __always_inline __attribute__((always_inline))
 #endif
+#ifndef __noinline
+#define __noinline __attribute__((noinline))
+#endif
 #ifndef __weak
 #define __weak __attribute__((weak))
 #endif
@@ -51,6 +54,54 @@
 #endif
 
 /*
+ * Helper macro to throw a compilation error if __bpf_unreachable() gets
+ * built into the resulting code. This works given BPF back end does not
+ * implement __builtin_trap(). This is useful to assert that certain paths
+ * of the program code are never used and hence eliminated by the compiler.
+ *
+ * For example, consider a switch statement that covers known cases used by
+ * the program. __bpf_unreachable() can then reside in the default case. If
+ * the program gets extended such that a case is not covered in the switch
+ * statement, then it will throw a build error due to the default case not
+ * being compiled out.
+ */
+#ifndef __bpf_unreachable
+# define __bpf_unreachable()	__builtin_trap()
+#endif
+
+/*
+ * Helper function to perform a tail call with a constant/immediate map slot.
+ */
+#if __clang_major__ >= 8 && defined(__bpf__)
+static __always_inline void
+bpf_tail_call_static(void *ctx, const void *map, const __u32 slot)
+{
+	if (!__builtin_constant_p(slot))
+		__bpf_unreachable();
+
+	/*
+	 * Provide a hard guarantee that LLVM won't optimize setting r2 (map
+	 * pointer) and r3 (constant map index) from _different paths_ ending
+	 * up at the _same_ call insn as otherwise we won't be able to use the
+	 * jmpq/nopl retpoline-free patching by the x86-64 JIT in the kernel
+	 * given they mismatch. See also d2e4c1e6c294 ("bpf: Constant map key
+	 * tracking for prog array pokes") for details on verifier tracking.
+	 *
+	 * Note on clobber list: we need to stay in-line with BPF calling
+	 * convention, so even if we don't end up using r0, r4, r5, we need
+	 * to mark them as clobber so that LLVM doesn't end up using them
+	 * before / after the call.
+	 */
+	asm volatile("r1 = %[ctx]\n\t"
+		     "r2 = %[map]\n\t"
+		     "r3 = %[slot]\n\t"
+		     "call 12"
+		     :: [ctx]"r"(ctx), [map]"r"(map), [slot]"i"(slot)
+		     : "r0", "r1", "r2", "r3", "r4", "r5");
+}
+#endif
+
+/*
  * Helper structure used by eBPF C program
  * to describe BPF map attributes to libbpf loader
  */
diff --git a/tools/lib/bpf/bpf_prog_linfo.c b/tools/lib/bpf/bpf_prog_linfo.c
index bafca49cb1e6..3ed1a27b5f7c 100644
--- a/tools/lib/bpf/bpf_prog_linfo.c
+++ b/tools/lib/bpf/bpf_prog_linfo.c
@@ -8,9 +8,6 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
-/* make sure libbpf doesn't use kernel-only integer typedefs */
-#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
-
 struct bpf_prog_linfo {
 	void *raw_linfo;
 	void *raw_jited_linfo;
diff --git a/tools/lib/bpf/bpf_tracing.h b/tools/lib/bpf/bpf_tracing.h
index eebf020cbe3e..f9ef37707888 100644
--- a/tools/lib/bpf/bpf_tracing.h
+++ b/tools/lib/bpf/bpf_tracing.h
@@ -289,9 +289,9 @@ struct pt_regs;
 #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)); })
+	({ bpf_probe_read_kernel(&(ip), sizeof(ip), (void *)PT_REGS_RET(ctx)); })
 #define BPF_KRETPROBE_READ_RET_IP(ip, ctx)				    \
-	({ bpf_probe_read(&(ip), sizeof(ip),				    \
+	({ bpf_probe_read_kernel(&(ip), sizeof(ip),			    \
 			  (void *)(PT_REGS_FP(ctx) + sizeof(ip))); })
 #endif
 
diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
index 6bdbc389b493..231b07203e3d 100644
--- a/tools/lib/bpf/btf.c
+++ b/tools/lib/bpf/btf.c
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: (LGPL-2.1 OR BSD-2-Clause)
 /* Copyright (c) 2018 Facebook */
 
+#include <byteswap.h>
 #include <endian.h>
 #include <stdio.h>
 #include <stdlib.h>
@@ -21,26 +22,78 @@
 #include "libbpf_internal.h"
 #include "hashmap.h"
 
-/* make sure libbpf doesn't use kernel-only integer typedefs */
-#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
-
 #define BTF_MAX_NR_TYPES 0x7fffffffU
 #define BTF_MAX_STR_OFFSET 0x7fffffffU
 
 static struct btf_type btf_void;
 
 struct btf {
-	union {
-		struct btf_header *hdr;
-		void *data;
-	};
-	struct btf_type **types;
-	const char *strings;
-	void *nohdr_data;
+	/* raw BTF data in native endianness */
+	void *raw_data;
+	/* raw BTF data in non-native endianness */
+	void *raw_data_swapped;
+	__u32 raw_size;
+	/* whether target endianness differs from the native one */
+	bool swapped_endian;
+
+	/*
+	 * When BTF is loaded from an ELF or raw memory it is stored
+	 * in a contiguous memory block. The hdr, type_data, and, strs_data
+	 * point inside that memory region to their respective parts of BTF
+	 * representation:
+	 *
+	 * +--------------------------------+
+	 * |  Header  |  Types  |  Strings  |
+	 * +--------------------------------+
+	 * ^          ^         ^
+	 * |          |         |
+	 * hdr        |         |
+	 * types_data-+         |
+	 * strs_data------------+
+	 *
+	 * If BTF data is later modified, e.g., due to types added or
+	 * removed, BTF deduplication performed, etc, this contiguous
+	 * representation is broken up into three independently allocated
+	 * memory regions to be able to modify them independently.
+	 * raw_data is nulled out at that point, but can be later allocated
+	 * and cached again if user calls btf__get_raw_data(), at which point
+	 * raw_data will contain a contiguous copy of header, types, and
+	 * strings:
+	 *
+	 * +----------+  +---------+  +-----------+
+	 * |  Header  |  |  Types  |  |  Strings  |
+	 * +----------+  +---------+  +-----------+
+	 * ^             ^            ^
+	 * |             |            |
+	 * hdr           |            |
+	 * types_data----+            |
+	 * strs_data------------------+
+	 *
+	 *               +----------+---------+-----------+
+	 *               |  Header  |  Types  |  Strings  |
+	 * raw_data----->+----------+---------+-----------+
+	 */
+	struct btf_header *hdr;
+
+	void *types_data;
+	size_t types_data_cap; /* used size stored in hdr->type_len */
+
+	/* type ID to `struct btf_type *` lookup index */
+	__u32 *type_offs;
+	size_t type_offs_cap;
 	__u32 nr_types;
-	__u32 types_size;
-	__u32 data_size;
+
+	void *strs_data;
+	size_t strs_data_cap; /* used size stored in hdr->str_len */
+
+	/* lookup index for each unique string in strings section */
+	struct hashmap *strs_hash;
+	/* whether strings are already deduplicated */
+	bool strs_deduped;
+	/* BTF object FD, if loaded into kernel */
 	int fd;
+
+	/* Pointer size (in bytes) for a target architecture of this BTF */
 	int ptr_sz;
 };
 
@@ -49,60 +102,114 @@ static inline __u64 ptr_to_u64(const void *ptr)
 	return (__u64) (unsigned long) ptr;
 }
 
-static int btf_add_type(struct btf *btf, struct btf_type *t)
+/* Ensure given dynamically allocated memory region pointed to by *data* with
+ * capacity of *cap_cnt* elements each taking *elem_sz* bytes has enough
+ * memory to accomodate *add_cnt* new elements, assuming *cur_cnt* elements
+ * are already used. At most *max_cnt* elements can be ever allocated.
+ * If necessary, memory is reallocated and all existing data is copied over,
+ * new pointer to the memory region is stored at *data, new memory region
+ * capacity (in number of elements) is stored in *cap.
+ * On success, memory pointer to the beginning of unused memory is returned.
+ * On error, NULL is returned.
+ */
+void *btf_add_mem(void **data, size_t *cap_cnt, size_t elem_sz,
+		  size_t cur_cnt, size_t max_cnt, size_t add_cnt)
 {
-	if (btf->types_size - btf->nr_types < 2) {
-		struct btf_type **new_types;
-		__u32 expand_by, new_size;
+	size_t new_cnt;
+	void *new_data;
 
-		if (btf->types_size == BTF_MAX_NR_TYPES)
-			return -E2BIG;
+	if (cur_cnt + add_cnt <= *cap_cnt)
+		return *data + cur_cnt * elem_sz;
 
-		expand_by = max(btf->types_size >> 2, 16U);
-		new_size = min(BTF_MAX_NR_TYPES, btf->types_size + expand_by);
+	/* requested more than the set limit */
+	if (cur_cnt + add_cnt > max_cnt)
+		return NULL;
 
-		new_types = realloc(btf->types, sizeof(*new_types) * new_size);
-		if (!new_types)
-			return -ENOMEM;
+	new_cnt = *cap_cnt;
+	new_cnt += new_cnt / 4;		  /* expand by 25% */
+	if (new_cnt < 16)		  /* but at least 16 elements */
+		new_cnt = 16;
+	if (new_cnt > max_cnt)		  /* but not exceeding a set limit */
+		new_cnt = max_cnt;
+	if (new_cnt < cur_cnt + add_cnt)  /* also ensure we have enough memory */
+		new_cnt = cur_cnt + add_cnt;
 
-		if (btf->nr_types == 0)
-			new_types[0] = &btf_void;
+	new_data = libbpf_reallocarray(*data, new_cnt, elem_sz);
+	if (!new_data)
+		return NULL;
 
-		btf->types = new_types;
-		btf->types_size = new_size;
-	}
+	/* zero out newly allocated portion of memory */
+	memset(new_data + (*cap_cnt) * elem_sz, 0, (new_cnt - *cap_cnt) * elem_sz);
 
-	btf->types[++(btf->nr_types)] = t;
+	*data = new_data;
+	*cap_cnt = new_cnt;
+	return new_data + cur_cnt * elem_sz;
+}
 
+/* Ensure given dynamically allocated memory region has enough allocated space
+ * to accommodate *need_cnt* elements of size *elem_sz* bytes each
+ */
+int btf_ensure_mem(void **data, size_t *cap_cnt, size_t elem_sz, size_t need_cnt)
+{
+	void *p;
+
+	if (need_cnt <= *cap_cnt)
+		return 0;
+
+	p = btf_add_mem(data, cap_cnt, elem_sz, *cap_cnt, SIZE_MAX, need_cnt - *cap_cnt);
+	if (!p)
+		return -ENOMEM;
+
+	return 0;
+}
+
+static int btf_add_type_idx_entry(struct btf *btf, __u32 type_off)
+{
+	__u32 *p;
+
+	p = btf_add_mem((void **)&btf->type_offs, &btf->type_offs_cap, sizeof(__u32),
+			btf->nr_types + 1, BTF_MAX_NR_TYPES, 1);
+	if (!p)
+		return -ENOMEM;
+
+	*p = type_off;
 	return 0;
 }
 
+static void btf_bswap_hdr(struct btf_header *h)
+{
+	h->magic = bswap_16(h->magic);
+	h->hdr_len = bswap_32(h->hdr_len);
+	h->type_off = bswap_32(h->type_off);
+	h->type_len = bswap_32(h->type_len);
+	h->str_off = bswap_32(h->str_off);
+	h->str_len = bswap_32(h->str_len);
+}
+
 static int btf_parse_hdr(struct btf *btf)
 {
-	const struct btf_header *hdr = btf->hdr;
+	struct btf_header *hdr = btf->hdr;
 	__u32 meta_left;
 
-	if (btf->data_size < sizeof(struct btf_header)) {
+	if (btf->raw_size < sizeof(struct btf_header)) {
 		pr_debug("BTF header not found\n");
 		return -EINVAL;
 	}
 
-	if (hdr->magic != BTF_MAGIC) {
+	if (hdr->magic == bswap_16(BTF_MAGIC)) {
+		btf->swapped_endian = true;
+		if (bswap_32(hdr->hdr_len) != sizeof(struct btf_header)) {
+			pr_warn("Can't load BTF with non-native endianness due to unsupported header length %u\n",
+				bswap_32(hdr->hdr_len));
+			return -ENOTSUP;
+		}
+		btf_bswap_hdr(hdr);
+	} else if (hdr->magic != BTF_MAGIC) {
 		pr_debug("Invalid BTF magic:%x\n", hdr->magic);
 		return -EINVAL;
 	}
 
-	if (hdr->version != BTF_VERSION) {
-		pr_debug("Unsupported BTF version:%u\n", hdr->version);
-		return -ENOTSUP;
-	}
-
-	if (hdr->flags) {
-		pr_debug("Unsupported BTF flags:%x\n", hdr->flags);
-		return -ENOTSUP;
-	}
-
-	meta_left = btf->data_size - sizeof(*hdr);
+	meta_left = btf->raw_size - sizeof(*hdr);
 	if (!meta_left) {
 		pr_debug("BTF has no data\n");
 		return -EINVAL;
@@ -128,15 +235,13 @@ static int btf_parse_hdr(struct btf *btf)
 		return -EINVAL;
 	}
 
-	btf->nohdr_data = btf->hdr + 1;
-
 	return 0;
 }
 
 static int btf_parse_str_sec(struct btf *btf)
 {
 	const struct btf_header *hdr = btf->hdr;
-	const char *start = btf->nohdr_data + hdr->str_off;
+	const char *start = btf->strs_data;
 	const char *end = start + btf->hdr->str_len;
 
 	if (!hdr->str_len || hdr->str_len - 1 > BTF_MAX_STR_OFFSET ||
@@ -145,14 +250,12 @@ static int btf_parse_str_sec(struct btf *btf)
 		return -EINVAL;
 	}
 
-	btf->strings = start;
-
 	return 0;
 }
 
-static int btf_type_size(struct btf_type *t)
+static int btf_type_size(const struct btf_type *t)
 {
-	int base_size = sizeof(struct btf_type);
+	const int base_size = sizeof(struct btf_type);
 	__u16 vlen = btf_vlen(t);
 
 	switch (btf_kind(t)) {
@@ -185,25 +288,120 @@ static int btf_type_size(struct btf_type *t)
 	}
 }
 
+static void btf_bswap_type_base(struct btf_type *t)
+{
+	t->name_off = bswap_32(t->name_off);
+	t->info = bswap_32(t->info);
+	t->type = bswap_32(t->type);
+}
+
+static int btf_bswap_type_rest(struct btf_type *t)
+{
+	struct btf_var_secinfo *v;
+	struct btf_member *m;
+	struct btf_array *a;
+	struct btf_param *p;
+	struct btf_enum *e;
+	__u16 vlen = btf_vlen(t);
+	int i;
+
+	switch (btf_kind(t)) {
+	case BTF_KIND_FWD:
+	case BTF_KIND_CONST:
+	case BTF_KIND_VOLATILE:
+	case BTF_KIND_RESTRICT:
+	case BTF_KIND_PTR:
+	case BTF_KIND_TYPEDEF:
+	case BTF_KIND_FUNC:
+		return 0;
+	case BTF_KIND_INT:
+		*(__u32 *)(t + 1) = bswap_32(*(__u32 *)(t + 1));
+		return 0;
+	case BTF_KIND_ENUM:
+		for (i = 0, e = btf_enum(t); i < vlen; i++, e++) {
+			e->name_off = bswap_32(e->name_off);
+			e->val = bswap_32(e->val);
+		}
+		return 0;
+	case BTF_KIND_ARRAY:
+		a = btf_array(t);
+		a->type = bswap_32(a->type);
+		a->index_type = bswap_32(a->index_type);
+		a->nelems = bswap_32(a->nelems);
+		return 0;
+	case BTF_KIND_STRUCT:
+	case BTF_KIND_UNION:
+		for (i = 0, m = btf_members(t); i < vlen; i++, m++) {
+			m->name_off = bswap_32(m->name_off);
+			m->type = bswap_32(m->type);
+			m->offset = bswap_32(m->offset);
+		}
+		return 0;
+	case BTF_KIND_FUNC_PROTO:
+		for (i = 0, p = btf_params(t); i < vlen; i++, p++) {
+			p->name_off = bswap_32(p->name_off);
+			p->type = bswap_32(p->type);
+		}
+		return 0;
+	case BTF_KIND_VAR:
+		btf_var(t)->linkage = bswap_32(btf_var(t)->linkage);
+		return 0;
+	case BTF_KIND_DATASEC:
+		for (i = 0, v = btf_var_secinfos(t); i < vlen; i++, v++) {
+			v->type = bswap_32(v->type);
+			v->offset = bswap_32(v->offset);
+			v->size = bswap_32(v->size);
+		}
+		return 0;
+	default:
+		pr_debug("Unsupported BTF_KIND:%u\n", btf_kind(t));
+		return -EINVAL;
+	}
+}
+
 static int btf_parse_type_sec(struct btf *btf)
 {
 	struct btf_header *hdr = btf->hdr;
-	void *nohdr_data = btf->nohdr_data;
-	void *next_type = nohdr_data + hdr->type_off;
-	void *end_type = nohdr_data + hdr->str_off;
+	void *next_type = btf->types_data;
+	void *end_type = next_type + hdr->type_len;
+	int err, i = 0, type_size;
 
-	while (next_type < end_type) {
-		struct btf_type *t = next_type;
-		int type_size;
-		int err;
+	/* VOID (type_id == 0) is specially handled by btf__get_type_by_id(),
+	 * so ensure we can never properly use its offset from index by
+	 * setting it to a large value
+	 */
+	err = btf_add_type_idx_entry(btf, UINT_MAX);
+	if (err)
+		return err;
+
+	while (next_type + sizeof(struct btf_type) <= end_type) {
+		i++;
 
-		type_size = btf_type_size(t);
+		if (btf->swapped_endian)
+			btf_bswap_type_base(next_type);
+
+		type_size = btf_type_size(next_type);
 		if (type_size < 0)
 			return type_size;
-		next_type += type_size;
-		err = btf_add_type(btf, t);
+		if (next_type + type_size > end_type) {
+			pr_warn("BTF type [%d] is malformed\n", i);
+			return -EINVAL;
+		}
+
+		if (btf->swapped_endian && btf_bswap_type_rest(next_type))
+			return -EINVAL;
+
+		err = btf_add_type_idx_entry(btf, next_type - btf->types_data);
 		if (err)
 			return err;
+
+		next_type += type_size;
+		btf->nr_types++;
+	}
+
+	if (next_type != end_type) {
+		pr_warn("BTF types data is malformed\n");
+		return -EINVAL;
 	}
 
 	return 0;
@@ -214,12 +412,20 @@ __u32 btf__get_nr_types(const struct btf *btf)
 	return btf->nr_types;
 }
 
+/* internal helper returning non-const pointer to a type */
+static struct btf_type *btf_type_by_id(struct btf *btf, __u32 type_id)
+{
+	if (type_id == 0)
+		return &btf_void;
+
+	return btf->types_data + btf->type_offs[type_id];
+}
+
 const struct btf_type *btf__type_by_id(const struct btf *btf, __u32 type_id)
 {
 	if (type_id > btf->nr_types)
 		return NULL;
-
-	return btf->types[type_id];
+	return btf_type_by_id((struct btf *)btf, type_id);
 }
 
 static int determine_ptr_size(const struct btf *btf)
@@ -286,6 +492,38 @@ int btf__set_pointer_size(struct btf *btf, size_t ptr_sz)
 	return 0;
 }
 
+static bool is_host_big_endian(void)
+{
+#if __BYTE_ORDER == __LITTLE_ENDIAN
+	return false;
+#elif __BYTE_ORDER == __BIG_ENDIAN
+	return true;
+#else
+# error "Unrecognized __BYTE_ORDER__"
+#endif
+}
+
+enum btf_endianness btf__endianness(const struct btf *btf)
+{
+	if (is_host_big_endian())
+		return btf->swapped_endian ? BTF_LITTLE_ENDIAN : BTF_BIG_ENDIAN;
+	else
+		return btf->swapped_endian ? BTF_BIG_ENDIAN : BTF_LITTLE_ENDIAN;
+}
+
+int btf__set_endianness(struct btf *btf, enum btf_endianness endian)
+{
+	if (endian != BTF_LITTLE_ENDIAN && endian != BTF_BIG_ENDIAN)
+		return -EINVAL;
+
+	btf->swapped_endian = is_host_big_endian() != (endian == BTF_BIG_ENDIAN);
+	if (!btf->swapped_endian) {
+		free(btf->raw_data_swapped);
+		btf->raw_data_swapped = NULL;
+	}
+	return 0;
+}
+
 static bool btf_type_is_void(const struct btf_type *t)
 {
 	return t == &btf_void || btf_is_fwd(t);
@@ -417,7 +655,7 @@ __s32 btf__find_by_name(const struct btf *btf, const char *type_name)
 		return 0;
 
 	for (i = 1; i <= btf->nr_types; i++) {
-		const struct btf_type *t = btf->types[i];
+		const struct btf_type *t = btf__type_by_id(btf, i);
 		const char *name = btf__name_by_offset(btf, t->name_off);
 
 		if (name && !strcmp(type_name, name))
@@ -436,7 +674,7 @@ __s32 btf__find_by_name_kind(const struct btf *btf, const char *type_name,
 		return 0;
 
 	for (i = 1; i <= btf->nr_types; i++) {
-		const struct btf_type *t = btf->types[i];
+		const struct btf_type *t = btf__type_by_id(btf, i);
 		const char *name;
 
 		if (btf_kind(t) != kind)
@@ -449,6 +687,11 @@ __s32 btf__find_by_name_kind(const struct btf *btf, const char *type_name,
 	return -ENOENT;
 }
 
+static bool btf_is_modifiable(const struct btf *btf)
+{
+	return (void *)btf->hdr != btf->raw_data;
+}
+
 void btf__free(struct btf *btf)
 {
 	if (IS_ERR_OR_NULL(btf))
@@ -457,11 +700,55 @@ void btf__free(struct btf *btf)
 	if (btf->fd >= 0)
 		close(btf->fd);
 
-	free(btf->data);
-	free(btf->types);
+	if (btf_is_modifiable(btf)) {
+		/* if BTF was modified after loading, it will have a split
+		 * in-memory representation for header, types, and strings
+		 * sections, so we need to free all of them individually. It
+		 * might still have a cached contiguous raw data present,
+		 * which will be unconditionally freed below.
+		 */
+		free(btf->hdr);
+		free(btf->types_data);
+		free(btf->strs_data);
+	}
+	free(btf->raw_data);
+	free(btf->raw_data_swapped);
+	free(btf->type_offs);
 	free(btf);
 }
 
+struct btf *btf__new_empty(void)
+{
+	struct btf *btf;
+
+	btf = calloc(1, sizeof(*btf));
+	if (!btf)
+		return ERR_PTR(-ENOMEM);
+
+	btf->fd = -1;
+	btf->ptr_sz = sizeof(void *);
+	btf->swapped_endian = false;
+
+	/* +1 for empty string at offset 0 */
+	btf->raw_size = sizeof(struct btf_header) + 1;
+	btf->raw_data = calloc(1, btf->raw_size);
+	if (!btf->raw_data) {
+		free(btf);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	btf->hdr = btf->raw_data;
+	btf->hdr->hdr_len = sizeof(struct btf_header);
+	btf->hdr->magic = BTF_MAGIC;
+	btf->hdr->version = BTF_VERSION;
+
+	btf->types_data = btf->raw_data + btf->hdr->hdr_len;
+	btf->strs_data = btf->raw_data + btf->hdr->hdr_len;
+	btf->hdr->str_len = 1; /* empty string at offset 0 */
+
+	return btf;
+}
+
 struct btf *btf__new(const void *data, __u32 size)
 {
 	struct btf *btf;
@@ -471,26 +758,28 @@ struct btf *btf__new(const void *data, __u32 size)
 	if (!btf)
 		return ERR_PTR(-ENOMEM);
 
-	btf->fd = -1;
-
-	btf->data = malloc(size);
-	if (!btf->data) {
+	btf->raw_data = malloc(size);
+	if (!btf->raw_data) {
 		err = -ENOMEM;
 		goto done;
 	}
+	memcpy(btf->raw_data, data, size);
+	btf->raw_size = size;
 
-	memcpy(btf->data, data, size);
-	btf->data_size = size;
-
+	btf->hdr = btf->raw_data;
 	err = btf_parse_hdr(btf);
 	if (err)
 		goto done;
 
+	btf->strs_data = btf->raw_data + btf->hdr->hdr_len + btf->hdr->str_off;
+	btf->types_data = btf->raw_data + btf->hdr->hdr_len + btf->hdr->type_off;
+
 	err = btf_parse_str_sec(btf);
+	err = err ?: btf_parse_type_sec(btf);
 	if (err)
 		goto done;
 
-	err = btf_parse_type_sec(btf);
+	btf->fd = -1;
 
 done:
 	if (err) {
@@ -501,17 +790,6 @@ done:
 	return btf;
 }
 
-static bool btf_check_endianness(const GElf_Ehdr *ehdr)
-{
-#if __BYTE_ORDER == __LITTLE_ENDIAN
-	return ehdr->e_ident[EI_DATA] == ELFDATA2LSB;
-#elif __BYTE_ORDER == __BIG_ENDIAN
-	return ehdr->e_ident[EI_DATA] == ELFDATA2MSB;
-#else
-# error "Unrecognized __BYTE_ORDER__"
-#endif
-}
-
 struct btf *btf__parse_elf(const char *path, struct btf_ext **btf_ext)
 {
 	Elf_Data *btf_data = NULL, *btf_ext_data = NULL;
@@ -544,10 +822,6 @@ struct btf *btf__parse_elf(const char *path, struct btf_ext **btf_ext)
 		pr_warn("failed to get EHDR from %s\n", path);
 		goto done;
 	}
-	if (!btf_check_endianness(&ehdr)) {
-		pr_warn("non-native ELF endianness is not supported\n");
-		goto done;
-	}
 	if (!elf_rawdata(elf_getscn(elf, ehdr.e_shstrndx), NULL)) {
 		pr_warn("failed to get e_shstrndx from %s\n", path);
 		goto done;
@@ -659,13 +933,7 @@ struct btf *btf__parse_raw(const char *path)
 		err = -EIO;
 		goto err_out;
 	}
-	if (magic == __bswap_16(BTF_MAGIC)) {
-		/* non-native endian raw BTF */
-		pr_warn("non-native BTF endianness is not supported\n");
-		err = -LIBBPF_ERRNO__ENDIAN;
-		goto err_out;
-	}
-	if (magic != BTF_MAGIC) {
+	if (magic != BTF_MAGIC && magic != bswap_16(BTF_MAGIC)) {
 		/* definitely not a raw BTF */
 		err = -EPROTO;
 		goto err_out;
@@ -798,7 +1066,7 @@ int btf__finalize_data(struct bpf_object *obj, struct btf *btf)
 	__u32 i;
 
 	for (i = 1; i <= btf->nr_types; i++) {
-		struct btf_type *t = btf->types[i];
+		struct btf_type *t = btf_type_by_id(btf, i);
 
 		/* Loader needs to fix up some of the things compiler
 		 * couldn't get its hands on while emitting BTF. This
@@ -815,10 +1083,13 @@ int btf__finalize_data(struct bpf_object *obj, struct btf *btf)
 	return err;
 }
 
+static void *btf_get_raw_data(const struct btf *btf, __u32 *size, bool swap_endian);
+
 int btf__load(struct btf *btf)
 {
-	__u32 log_buf_size = 0;
+	__u32 log_buf_size = 0, raw_size;
 	char *log_buf = NULL;
+	void *raw_data;
 	int err = 0;
 
 	if (btf->fd >= 0)
@@ -833,8 +1104,16 @@ retry_load:
 		*log_buf = 0;
 	}
 
-	btf->fd = bpf_load_btf(btf->data, btf->data_size,
-			       log_buf, log_buf_size, false);
+	raw_data = btf_get_raw_data(btf, &raw_size, false);
+	if (!raw_data) {
+		err = -ENOMEM;
+		goto done;
+	}
+	/* cache native raw data representation */
+	btf->raw_size = raw_size;
+	btf->raw_data = raw_data;
+
+	btf->fd = bpf_load_btf(raw_data, raw_size, log_buf, log_buf_size, false);
 	if (btf->fd < 0) {
 		if (!log_buf || errno == ENOSPC) {
 			log_buf_size = max((__u32)BPF_LOG_BUF_SIZE,
@@ -865,20 +1144,88 @@ void btf__set_fd(struct btf *btf, int fd)
 	btf->fd = fd;
 }
 
-const void *btf__get_raw_data(const struct btf *btf, __u32 *size)
+static void *btf_get_raw_data(const struct btf *btf, __u32 *size, bool swap_endian)
 {
-	*size = btf->data_size;
-	return btf->data;
+	struct btf_header *hdr = btf->hdr;
+	struct btf_type *t;
+	void *data, *p;
+	__u32 data_sz;
+	int i;
+
+	data = swap_endian ? btf->raw_data_swapped : btf->raw_data;
+	if (data) {
+		*size = btf->raw_size;
+		return data;
+	}
+
+	data_sz = hdr->hdr_len + hdr->type_len + hdr->str_len;
+	data = calloc(1, data_sz);
+	if (!data)
+		return NULL;
+	p = data;
+
+	memcpy(p, hdr, hdr->hdr_len);
+	if (swap_endian)
+		btf_bswap_hdr(p);
+	p += hdr->hdr_len;
+
+	memcpy(p, btf->types_data, hdr->type_len);
+	if (swap_endian) {
+		for (i = 1; i <= btf->nr_types; i++) {
+			t = p  + btf->type_offs[i];
+			/* btf_bswap_type_rest() relies on native t->info, so
+			 * we swap base type info after we swapped all the
+			 * additional information
+			 */
+			if (btf_bswap_type_rest(t))
+				goto err_out;
+			btf_bswap_type_base(t);
+		}
+	}
+	p += hdr->type_len;
+
+	memcpy(p, btf->strs_data, hdr->str_len);
+	p += hdr->str_len;
+
+	*size = data_sz;
+	return data;
+err_out:
+	free(data);
+	return NULL;
 }
 
-const char *btf__name_by_offset(const struct btf *btf, __u32 offset)
+const void *btf__get_raw_data(const struct btf *btf_ro, __u32 *size)
+{
+	struct btf *btf = (struct btf *)btf_ro;
+	__u32 data_sz;
+	void *data;
+
+	data = btf_get_raw_data(btf, &data_sz, btf->swapped_endian);
+	if (!data)
+		return NULL;
+
+	btf->raw_size = data_sz;
+	if (btf->swapped_endian)
+		btf->raw_data_swapped = data;
+	else
+		btf->raw_data = data;
+	*size = data_sz;
+	return data;
+}
+
+const char *btf__str_by_offset(const struct btf *btf, __u32 offset)
 {
 	if (offset < btf->hdr->str_len)
-		return &btf->strings[offset];
+		return btf->strs_data + offset;
 	else
 		return NULL;
 }
 
+const char *btf__name_by_offset(const struct btf *btf, __u32 offset)
+{
+	return btf__str_by_offset(btf, offset);
+}
+
 int btf__get_from_id(__u32 id, struct btf **btf)
 {
 	struct bpf_btf_info btf_info = { 0 };
@@ -1014,6 +1361,970 @@ int btf__get_map_kv_tids(const struct btf *btf, const char *map_name,
 	return 0;
 }
 
+static size_t strs_hash_fn(const void *key, void *ctx)
+{
+	struct btf *btf = ctx;
+	const char *str = btf->strs_data + (long)key;
+
+	return str_hash(str);
+}
+
+static bool strs_hash_equal_fn(const void *key1, const void *key2, void *ctx)
+{
+	struct btf *btf = ctx;
+	const char *str1 = btf->strs_data + (long)key1;
+	const char *str2 = btf->strs_data + (long)key2;
+
+	return strcmp(str1, str2) == 0;
+}
+
+static void btf_invalidate_raw_data(struct btf *btf)
+{
+	if (btf->raw_data) {
+		free(btf->raw_data);
+		btf->raw_data = NULL;
+	}
+	if (btf->raw_data_swapped) {
+		free(btf->raw_data_swapped);
+		btf->raw_data_swapped = NULL;
+	}
+}
+
+/* Ensure BTF is ready to be modified (by splitting into a three memory
+ * regions for header, types, and strings). Also invalidate cached
+ * raw_data, if any.
+ */
+static int btf_ensure_modifiable(struct btf *btf)
+{
+	void *hdr, *types, *strs, *strs_end, *s;
+	struct hashmap *hash = NULL;
+	long off;
+	int err;
+
+	if (btf_is_modifiable(btf)) {
+		/* any BTF modification invalidates raw_data */
+		btf_invalidate_raw_data(btf);
+		return 0;
+	}
+
+	/* split raw data into three memory regions */
+	hdr = malloc(btf->hdr->hdr_len);
+	types = malloc(btf->hdr->type_len);
+	strs = malloc(btf->hdr->str_len);
+	if (!hdr || !types || !strs)
+		goto err_out;
+
+	memcpy(hdr, btf->hdr, btf->hdr->hdr_len);
+	memcpy(types, btf->types_data, btf->hdr->type_len);
+	memcpy(strs, btf->strs_data, btf->hdr->str_len);
+
+	/* build lookup index for all strings */
+	hash = hashmap__new(strs_hash_fn, strs_hash_equal_fn, btf);
+	if (IS_ERR(hash)) {
+		err = PTR_ERR(hash);
+		hash = NULL;
+		goto err_out;
+	}
+
+	strs_end = strs + btf->hdr->str_len;
+	for (off = 0, s = strs; s < strs_end; off += strlen(s) + 1, s = strs + off) {
+		/* hashmap__add() returns EEXIST if string with the same
+		 * content already is in the hash map
+		 */
+		err = hashmap__add(hash, (void *)off, (void *)off);
+		if (err == -EEXIST)
+			continue; /* duplicate */
+		if (err)
+			goto err_out;
+	}
+
+	/* only when everything was successful, update internal state */
+	btf->hdr = hdr;
+	btf->types_data = types;
+	btf->types_data_cap = btf->hdr->type_len;
+	btf->strs_data = strs;
+	btf->strs_data_cap = btf->hdr->str_len;
+	btf->strs_hash = hash;
+	/* if BTF was created from scratch, all strings are guaranteed to be
+	 * unique and deduplicated
+	 */
+	btf->strs_deduped = btf->hdr->str_len <= 1;
+
+	/* invalidate raw_data representation */
+	btf_invalidate_raw_data(btf);
+
+	return 0;
+
+err_out:
+	hashmap__free(hash);
+	free(hdr);
+	free(types);
+	free(strs);
+	return -ENOMEM;
+}
+
+static void *btf_add_str_mem(struct btf *btf, size_t add_sz)
+{
+	return btf_add_mem(&btf->strs_data, &btf->strs_data_cap, 1,
+			   btf->hdr->str_len, BTF_MAX_STR_OFFSET, add_sz);
+}
+
+/* Find an offset in BTF string section that corresponds to a given string *s*.
+ * Returns:
+ *   - >0 offset into string section, if string is found;
+ *   - -ENOENT, if string is not in the string section;
+ *   - <0, on any other error.
+ */
+int btf__find_str(struct btf *btf, const char *s)
+{
+	long old_off, new_off, len;
+	void *p;
+
+	/* BTF needs to be in a modifiable state to build string lookup index */
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	/* see btf__add_str() for why we do this */
+	len = strlen(s) + 1;
+	p = btf_add_str_mem(btf, len);
+	if (!p)
+		return -ENOMEM;
+
+	new_off = btf->hdr->str_len;
+	memcpy(p, s, len);
+
+	if (hashmap__find(btf->strs_hash, (void *)new_off, (void **)&old_off))
+		return old_off;
+
+	return -ENOENT;
+}
+
+/* Add a string s to the BTF string section.
+ * Returns:
+ *   - > 0 offset into string section, on success;
+ *   - < 0, on error.
+ */
+int btf__add_str(struct btf *btf, const char *s)
+{
+	long old_off, new_off, len;
+	void *p;
+	int err;
+
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	/* Hashmap keys are always offsets within btf->strs_data, so to even
+	 * look up some string from the "outside", we need to first append it
+	 * at the end, so that it can be addressed with an offset. Luckily,
+	 * until btf->hdr->str_len is incremented, that string is just a piece
+	 * of garbage for the rest of BTF code, so no harm, no foul. On the
+	 * other hand, if the string is unique, it's already appended and
+	 * ready to be used, only a simple btf->hdr->str_len increment away.
+	 */
+	len = strlen(s) + 1;
+	p = btf_add_str_mem(btf, len);
+	if (!p)
+		return -ENOMEM;
+
+	new_off = btf->hdr->str_len;
+	memcpy(p, s, len);
+
+	/* Now attempt to add the string, but only if the string with the same
+	 * contents doesn't exist already (HASHMAP_ADD strategy). If such
+	 * string exists, we'll get its offset in old_off (that's old_key).
+	 */
+	err = hashmap__insert(btf->strs_hash, (void *)new_off, (void *)new_off,
+			      HASHMAP_ADD, (const void **)&old_off, NULL);
+	if (err == -EEXIST)
+		return old_off; /* duplicated string, return existing offset */
+	if (err)
+		return err;
+
+	btf->hdr->str_len += len; /* new unique string, adjust data length */
+	return new_off;
+}
+
+static void *btf_add_type_mem(struct btf *btf, size_t add_sz)
+{
+	return btf_add_mem(&btf->types_data, &btf->types_data_cap, 1,
+			   btf->hdr->type_len, UINT_MAX, add_sz);
+}
+
+static __u32 btf_type_info(int kind, int vlen, int kflag)
+{
+	return (kflag << 31) | (kind << 24) | vlen;
+}
+
+static void btf_type_inc_vlen(struct btf_type *t)
+{
+	t->info = btf_type_info(btf_kind(t), btf_vlen(t) + 1, btf_kflag(t));
+}
+
+/*
+ * Append new BTF_KIND_INT type with:
+ *   - *name* - non-empty, non-NULL type name;
+ *   - *sz* - power-of-2 (1, 2, 4, ..) size of the type, in bytes;
+ *   - encoding is a combination of BTF_INT_SIGNED, BTF_INT_CHAR, BTF_INT_BOOL.
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_int(struct btf *btf, const char *name, size_t byte_sz, int encoding)
+{
+	struct btf_type *t;
+	int sz, err, name_off;
+
+	/* non-empty name */
+	if (!name || !name[0])
+		return -EINVAL;
+	/* byte_sz must be power of 2 */
+	if (!byte_sz || (byte_sz & (byte_sz - 1)) || byte_sz > 16)
+		return -EINVAL;
+	if (encoding & ~(BTF_INT_SIGNED | BTF_INT_CHAR | BTF_INT_BOOL))
+		return -EINVAL;
+
+	/* deconstruct BTF, if necessary, and invalidate raw_data */
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_type) + sizeof(int);
+	t = btf_add_type_mem(btf, sz);
+	if (!t)
+		return -ENOMEM;
+
+	/* if something goes wrong later, we might end up with an extra string,
+	 * but that shouldn't be a problem, because BTF can't be constructed
+	 * completely anyway and will most probably be just discarded
+	 */
+	name_off = btf__add_str(btf, name);
+	if (name_off < 0)
+		return name_off;
+
+	t->name_off = name_off;
+	t->info = btf_type_info(BTF_KIND_INT, 0, 0);
+	t->size = byte_sz;
+	/* set INT info, we don't allow setting legacy bit offset/size */
+	*(__u32 *)(t + 1) = (encoding << 24) | (byte_sz * 8);
+
+	err = btf_add_type_idx_entry(btf, btf->hdr->type_len);
+	if (err)
+		return err;
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	btf->nr_types++;
+	return btf->nr_types;
+}
+
+/* it's completely legal to append BTF types with type IDs pointing forward to
+ * types that haven't been appended yet, so we only make sure that id looks
+ * sane, we can't guarantee that ID will always be valid
+ */
+static int validate_type_id(int id)
+{
+	if (id < 0 || id > BTF_MAX_NR_TYPES)
+		return -EINVAL;
+	return 0;
+}
+
+/* generic append function for PTR, TYPEDEF, CONST/VOLATILE/RESTRICT */
+static int btf_add_ref_kind(struct btf *btf, int kind, const char *name, int ref_type_id)
+{
+	struct btf_type *t;
+	int sz, name_off = 0, err;
+
+	if (validate_type_id(ref_type_id))
+		return -EINVAL;
+
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_type);
+	t = btf_add_type_mem(btf, sz);
+	if (!t)
+		return -ENOMEM;
+
+	if (name && name[0]) {
+		name_off = btf__add_str(btf, name);
+		if (name_off < 0)
+			return name_off;
+	}
+
+	t->name_off = name_off;
+	t->info = btf_type_info(kind, 0, 0);
+	t->type = ref_type_id;
+
+	err = btf_add_type_idx_entry(btf, btf->hdr->type_len);
+	if (err)
+		return err;
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	btf->nr_types++;
+	return btf->nr_types;
+}
+
+/*
+ * Append new BTF_KIND_PTR type with:
+ *   - *ref_type_id* - referenced type ID, it might not exist yet;
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_ptr(struct btf *btf, int ref_type_id)
+{
+	return btf_add_ref_kind(btf, BTF_KIND_PTR, NULL, ref_type_id);
+}
+
+/*
+ * Append new BTF_KIND_ARRAY type with:
+ *   - *index_type_id* - type ID of the type describing array index;
+ *   - *elem_type_id* - type ID of the type describing array element;
+ *   - *nr_elems* - the size of the array;
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_array(struct btf *btf, int index_type_id, int elem_type_id, __u32 nr_elems)
+{
+	struct btf_type *t;
+	struct btf_array *a;
+	int sz, err;
+
+	if (validate_type_id(index_type_id) || validate_type_id(elem_type_id))
+		return -EINVAL;
+
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_type) + sizeof(struct btf_array);
+	t = btf_add_type_mem(btf, sz);
+	if (!t)
+		return -ENOMEM;
+
+	t->name_off = 0;
+	t->info = btf_type_info(BTF_KIND_ARRAY, 0, 0);
+	t->size = 0;
+
+	a = btf_array(t);
+	a->type = elem_type_id;
+	a->index_type = index_type_id;
+	a->nelems = nr_elems;
+
+	err = btf_add_type_idx_entry(btf, btf->hdr->type_len);
+	if (err)
+		return err;
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	btf->nr_types++;
+	return btf->nr_types;
+}
+
+/* generic STRUCT/UNION append function */
+static int btf_add_composite(struct btf *btf, int kind, const char *name, __u32 bytes_sz)
+{
+	struct btf_type *t;
+	int sz, err, name_off = 0;
+
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_type);
+	t = btf_add_type_mem(btf, sz);
+	if (!t)
+		return -ENOMEM;
+
+	if (name && name[0]) {
+		name_off = btf__add_str(btf, name);
+		if (name_off < 0)
+			return name_off;
+	}
+
+	/* start out with vlen=0 and no kflag; this will be adjusted when
+	 * adding each member
+	 */
+	t->name_off = name_off;
+	t->info = btf_type_info(kind, 0, 0);
+	t->size = bytes_sz;
+
+	err = btf_add_type_idx_entry(btf, btf->hdr->type_len);
+	if (err)
+		return err;
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	btf->nr_types++;
+	return btf->nr_types;
+}
+
+/*
+ * Append new BTF_KIND_STRUCT type with:
+ *   - *name* - name of the struct, can be NULL or empty for anonymous structs;
+ *   - *byte_sz* - size of the struct, in bytes;
+ *
+ * Struct initially has no fields in it. Fields can be added by
+ * btf__add_field() right after btf__add_struct() succeeds. 
+ *
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_struct(struct btf *btf, const char *name, __u32 byte_sz)
+{
+	return btf_add_composite(btf, BTF_KIND_STRUCT, name, byte_sz);
+}
+
+/*
+ * Append new BTF_KIND_UNION type with:
+ *   - *name* - name of the union, can be NULL or empty for anonymous union;
+ *   - *byte_sz* - size of the union, in bytes;
+ *
+ * Union initially has no fields in it. Fields can be added by
+ * btf__add_field() right after btf__add_union() succeeds. All fields
+ * should have *bit_offset* of 0.
+ *
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_union(struct btf *btf, const char *name, __u32 byte_sz)
+{
+	return btf_add_composite(btf, BTF_KIND_UNION, name, byte_sz);
+}
+
+/*
+ * Append new field for the current STRUCT/UNION type with:
+ *   - *name* - name of the field, can be NULL or empty for anonymous field;
+ *   - *type_id* - type ID for the type describing field type;
+ *   - *bit_offset* - bit offset of the start of the field within struct/union;
+ *   - *bit_size* - bit size of a bitfield, 0 for non-bitfield fields;
+ * Returns:
+ *   -  0, on success;
+ *   - <0, on error.
+ */
+int btf__add_field(struct btf *btf, const char *name, int type_id,
+		   __u32 bit_offset, __u32 bit_size)
+{
+	struct btf_type *t;
+	struct btf_member *m;
+	bool is_bitfield;
+	int sz, name_off = 0;
+
+	/* last type should be union/struct */
+	if (btf->nr_types == 0)
+		return -EINVAL;
+	t = btf_type_by_id(btf, btf->nr_types);
+	if (!btf_is_composite(t))
+		return -EINVAL;
+
+	if (validate_type_id(type_id))
+		return -EINVAL;
+	/* best-effort bit field offset/size enforcement */
+	is_bitfield = bit_size || (bit_offset % 8 != 0);
+	if (is_bitfield && (bit_size == 0 || bit_size > 255 || bit_offset > 0xffffff))
+		return -EINVAL;
+
+	/* only offset 0 is allowed for unions */
+	if (btf_is_union(t) && bit_offset)
+		return -EINVAL;
+
+	/* decompose and invalidate raw data */
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_member);
+	m = btf_add_type_mem(btf, sz);
+	if (!m)
+		return -ENOMEM;
+
+	if (name && name[0]) {
+		name_off = btf__add_str(btf, name);
+		if (name_off < 0)
+			return name_off;
+	}
+
+	m->name_off = name_off;
+	m->type = type_id;
+	m->offset = bit_offset | (bit_size << 24);
+
+	/* btf_add_type_mem can invalidate t pointer */
+	t = btf_type_by_id(btf, btf->nr_types);
+	/* update parent type's vlen and kflag */
+	t->info = btf_type_info(btf_kind(t), btf_vlen(t) + 1, is_bitfield || btf_kflag(t));
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	return 0;
+}
+
+/*
+ * Append new BTF_KIND_ENUM type with:
+ *   - *name* - name of the enum, can be NULL or empty for anonymous enums;
+ *   - *byte_sz* - size of the enum, in bytes.
+ *
+ * Enum initially has no enum values in it (and corresponds to enum forward
+ * declaration). Enumerator values can be added by btf__add_enum_value()
+ * immediately after btf__add_enum() succeeds.
+ *
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_enum(struct btf *btf, const char *name, __u32 byte_sz)
+{
+	struct btf_type *t;
+	int sz, err, name_off = 0;
+
+	/* byte_sz must be power of 2 */
+	if (!byte_sz || (byte_sz & (byte_sz - 1)) || byte_sz > 8)
+		return -EINVAL;
+
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_type);
+	t = btf_add_type_mem(btf, sz);
+	if (!t)
+		return -ENOMEM;
+
+	if (name && name[0]) {
+		name_off = btf__add_str(btf, name);
+		if (name_off < 0)
+			return name_off;
+	}
+
+	/* start out with vlen=0; it will be adjusted when adding enum values */
+	t->name_off = name_off;
+	t->info = btf_type_info(BTF_KIND_ENUM, 0, 0);
+	t->size = byte_sz;
+
+	err = btf_add_type_idx_entry(btf, btf->hdr->type_len);
+	if (err)
+		return err;
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	btf->nr_types++;
+	return btf->nr_types;
+}
+
+/*
+ * Append new enum value for the current ENUM type with:
+ *   - *name* - name of the enumerator value, can't be NULL or empty;
+ *   - *value* - integer value corresponding to enum value *name*;
+ * Returns:
+ *   -  0, on success;
+ *   - <0, on error.
+ */
+int btf__add_enum_value(struct btf *btf, const char *name, __s64 value)
+{
+	struct btf_type *t;
+	struct btf_enum *v;
+	int sz, name_off;
+
+	/* last type should be BTF_KIND_ENUM */
+	if (btf->nr_types == 0)
+		return -EINVAL;
+	t = btf_type_by_id(btf, btf->nr_types);
+	if (!btf_is_enum(t))
+		return -EINVAL;
+
+	/* non-empty name */
+	if (!name || !name[0])
+		return -EINVAL;
+	if (value < INT_MIN || value > UINT_MAX)
+		return -E2BIG;
+
+	/* decompose and invalidate raw data */
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_enum);
+	v = btf_add_type_mem(btf, sz);
+	if (!v)
+		return -ENOMEM;
+
+	name_off = btf__add_str(btf, name);
+	if (name_off < 0)
+		return name_off;
+
+	v->name_off = name_off;
+	v->val = value;
+
+	/* update parent type's vlen */
+	t = btf_type_by_id(btf, btf->nr_types);
+	btf_type_inc_vlen(t);
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	return 0;
+}
+
+/*
+ * Append new BTF_KIND_FWD type with:
+ *   - *name*, non-empty/non-NULL name;
+ *   - *fwd_kind*, kind of forward declaration, one of BTF_FWD_STRUCT,
+ *     BTF_FWD_UNION, or BTF_FWD_ENUM;
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_fwd(struct btf *btf, const char *name, enum btf_fwd_kind fwd_kind)
+{
+	if (!name || !name[0])
+		return -EINVAL;
+
+	switch (fwd_kind) {
+	case BTF_FWD_STRUCT:
+	case BTF_FWD_UNION: {
+		struct btf_type *t;
+		int id;
+
+		id = btf_add_ref_kind(btf, BTF_KIND_FWD, name, 0);
+		if (id <= 0)
+			return id;
+		t = btf_type_by_id(btf, id);
+		t->info = btf_type_info(BTF_KIND_FWD, 0, fwd_kind == BTF_FWD_UNION);
+		return id;
+	}
+	case BTF_FWD_ENUM:
+		/* enum forward in BTF currently is just an enum with no enum
+		 * values; we also assume a standard 4-byte size for it
+		 */
+		return btf__add_enum(btf, name, sizeof(int));
+	default:
+		return -EINVAL;
+	}
+}
+
+/*
+ * Append new BTF_KING_TYPEDEF type with:
+ *   - *name*, non-empty/non-NULL name;
+ *   - *ref_type_id* - referenced type ID, it might not exist yet;
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_typedef(struct btf *btf, const char *name, int ref_type_id)
+{
+	if (!name || !name[0])
+		return -EINVAL;
+
+	return btf_add_ref_kind(btf, BTF_KIND_TYPEDEF, name, ref_type_id);
+}
+
+/*
+ * Append new BTF_KIND_VOLATILE type with:
+ *   - *ref_type_id* - referenced type ID, it might not exist yet;
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_volatile(struct btf *btf, int ref_type_id)
+{
+	return btf_add_ref_kind(btf, BTF_KIND_VOLATILE, NULL, ref_type_id);
+}
+
+/*
+ * Append new BTF_KIND_CONST type with:
+ *   - *ref_type_id* - referenced type ID, it might not exist yet;
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_const(struct btf *btf, int ref_type_id)
+{
+	return btf_add_ref_kind(btf, BTF_KIND_CONST, NULL, ref_type_id);
+}
+
+/*
+ * Append new BTF_KIND_RESTRICT type with:
+ *   - *ref_type_id* - referenced type ID, it might not exist yet;
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_restrict(struct btf *btf, int ref_type_id)
+{
+	return btf_add_ref_kind(btf, BTF_KIND_RESTRICT, NULL, ref_type_id);
+}
+
+/*
+ * Append new BTF_KIND_FUNC type with:
+ *   - *name*, non-empty/non-NULL name;
+ *   - *proto_type_id* - FUNC_PROTO's type ID, it might not exist yet;
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_func(struct btf *btf, const char *name,
+		  enum btf_func_linkage linkage, int proto_type_id)
+{
+	int id;
+
+	if (!name || !name[0])
+		return -EINVAL;
+	if (linkage != BTF_FUNC_STATIC && linkage != BTF_FUNC_GLOBAL &&
+	    linkage != BTF_FUNC_EXTERN)
+		return -EINVAL;
+
+	id = btf_add_ref_kind(btf, BTF_KIND_FUNC, name, proto_type_id);
+	if (id > 0) {
+		struct btf_type *t = btf_type_by_id(btf, id);
+
+		t->info = btf_type_info(BTF_KIND_FUNC, linkage, 0);
+	}
+	return id;
+}
+
+/*
+ * Append new BTF_KIND_FUNC_PROTO with:
+ *   - *ret_type_id* - type ID for return result of a function.
+ *
+ * Function prototype initially has no arguments, but they can be added by
+ * btf__add_func_param() one by one, immediately after
+ * btf__add_func_proto() succeeded.
+ *
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_func_proto(struct btf *btf, int ret_type_id)
+{
+	struct btf_type *t;
+	int sz, err;
+
+	if (validate_type_id(ret_type_id))
+		return -EINVAL;
+
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_type);
+	t = btf_add_type_mem(btf, sz);
+	if (!t)
+		return -ENOMEM;
+
+	/* start out with vlen=0; this will be adjusted when adding enum
+	 * values, if necessary
+	 */
+	t->name_off = 0;
+	t->info = btf_type_info(BTF_KIND_FUNC_PROTO, 0, 0);
+	t->type = ret_type_id;
+
+	err = btf_add_type_idx_entry(btf, btf->hdr->type_len);
+	if (err)
+		return err;
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	btf->nr_types++;
+	return btf->nr_types;
+}
+
+/*
+ * Append new function parameter for current FUNC_PROTO type with:
+ *   - *name* - parameter name, can be NULL or empty;
+ *   - *type_id* - type ID describing the type of the parameter.
+ * Returns:
+ *   -  0, on success;
+ *   - <0, on error.
+ */
+int btf__add_func_param(struct btf *btf, const char *name, int type_id)
+{
+	struct btf_type *t;
+	struct btf_param *p;
+	int sz, name_off = 0;
+
+	if (validate_type_id(type_id))
+		return -EINVAL;
+
+	/* last type should be BTF_KIND_FUNC_PROTO */
+	if (btf->nr_types == 0)
+		return -EINVAL;
+	t = btf_type_by_id(btf, btf->nr_types);
+	if (!btf_is_func_proto(t))
+		return -EINVAL;
+
+	/* decompose and invalidate raw data */
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_param);
+	p = btf_add_type_mem(btf, sz);
+	if (!p)
+		return -ENOMEM;
+
+	if (name && name[0]) {
+		name_off = btf__add_str(btf, name);
+		if (name_off < 0)
+			return name_off;
+	}
+
+	p->name_off = name_off;
+	p->type = type_id;
+
+	/* update parent type's vlen */
+	t = btf_type_by_id(btf, btf->nr_types);
+	btf_type_inc_vlen(t);
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	return 0;
+}
+
+/*
+ * Append new BTF_KIND_VAR type with:
+ *   - *name* - non-empty/non-NULL name;
+ *   - *linkage* - variable linkage, one of BTF_VAR_STATIC,
+ *     BTF_VAR_GLOBAL_ALLOCATED, or BTF_VAR_GLOBAL_EXTERN;
+ *   - *type_id* - type ID of the type describing the type of the variable.
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_var(struct btf *btf, const char *name, int linkage, int type_id)
+{
+	struct btf_type *t;
+	struct btf_var *v;
+	int sz, err, name_off;
+
+	/* non-empty name */
+	if (!name || !name[0])
+		return -EINVAL;
+	if (linkage != BTF_VAR_STATIC && linkage != BTF_VAR_GLOBAL_ALLOCATED &&
+	    linkage != BTF_VAR_GLOBAL_EXTERN)
+		return -EINVAL;
+	if (validate_type_id(type_id))
+		return -EINVAL;
+
+	/* deconstruct BTF, if necessary, and invalidate raw_data */
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_type) + sizeof(struct btf_var);
+	t = btf_add_type_mem(btf, sz);
+	if (!t)
+		return -ENOMEM;
+
+	name_off = btf__add_str(btf, name);
+	if (name_off < 0)
+		return name_off;
+
+	t->name_off = name_off;
+	t->info = btf_type_info(BTF_KIND_VAR, 0, 0);
+	t->type = type_id;
+
+	v = btf_var(t);
+	v->linkage = linkage;
+
+	err = btf_add_type_idx_entry(btf, btf->hdr->type_len);
+	if (err)
+		return err;
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	btf->nr_types++;
+	return btf->nr_types;
+}
+
+/*
+ * Append new BTF_KIND_DATASEC type with:
+ *   - *name* - non-empty/non-NULL name;
+ *   - *byte_sz* - data section size, in bytes.
+ *
+ * Data section is initially empty. Variables info can be added with
+ * btf__add_datasec_var_info() calls, after btf__add_datasec() succeeds.
+ *
+ * Returns:
+ *   - >0, type ID of newly added BTF type;
+ *   - <0, on error.
+ */
+int btf__add_datasec(struct btf *btf, const char *name, __u32 byte_sz)
+{
+	struct btf_type *t;
+	int sz, err, name_off;
+
+	/* non-empty name */
+	if (!name || !name[0])
+		return -EINVAL;
+
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_type);
+	t = btf_add_type_mem(btf, sz);
+	if (!t)
+		return -ENOMEM;
+
+	name_off = btf__add_str(btf, name);
+	if (name_off < 0)
+		return name_off;
+
+	/* start with vlen=0, which will be update as var_secinfos are added */
+	t->name_off = name_off;
+	t->info = btf_type_info(BTF_KIND_DATASEC, 0, 0);
+	t->size = byte_sz;
+
+	err = btf_add_type_idx_entry(btf, btf->hdr->type_len);
+	if (err)
+		return err;
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	btf->nr_types++;
+	return btf->nr_types;
+}
+
+/*
+ * Append new data section variable information entry for current DATASEC type:
+ *   - *var_type_id* - type ID, describing type of the variable;
+ *   - *offset* - variable offset within data section, in bytes;
+ *   - *byte_sz* - variable size, in bytes.
+ *
+ * Returns:
+ *   -  0, on success;
+ *   - <0, on error.
+ */
+int btf__add_datasec_var_info(struct btf *btf, int var_type_id, __u32 offset, __u32 byte_sz)
+{
+	struct btf_type *t;
+	struct btf_var_secinfo *v;
+	int sz;
+
+	/* last type should be BTF_KIND_DATASEC */
+	if (btf->nr_types == 0)
+		return -EINVAL;
+	t = btf_type_by_id(btf, btf->nr_types);
+	if (!btf_is_datasec(t))
+		return -EINVAL;
+
+	if (validate_type_id(var_type_id))
+		return -EINVAL;
+
+	/* decompose and invalidate raw data */
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
+	sz = sizeof(struct btf_var_secinfo);
+	v = btf_add_type_mem(btf, sz);
+	if (!v)
+		return -ENOMEM;
+
+	v->type = var_type_id;
+	v->offset = offset;
+	v->size = byte_sz;
+
+	/* update parent type's vlen */
+	t = btf_type_by_id(btf, btf->nr_types);
+	btf_type_inc_vlen(t);
+
+	btf->hdr->type_len += sz;
+	btf->hdr->str_off += sz;
+	return 0;
+}
+
 struct btf_ext_sec_setup_param {
 	__u32 off;
 	__u32 len;
@@ -1137,14 +2448,14 @@ static int btf_ext_setup_line_info(struct btf_ext *btf_ext)
 	return btf_ext_setup_info(btf_ext, &param);
 }
 
-static int btf_ext_setup_field_reloc(struct btf_ext *btf_ext)
+static int btf_ext_setup_core_relos(struct btf_ext *btf_ext)
 {
 	struct btf_ext_sec_setup_param param = {
-		.off = btf_ext->hdr->field_reloc_off,
-		.len = btf_ext->hdr->field_reloc_len,
-		.min_rec_size = sizeof(struct bpf_field_reloc),
-		.ext_info = &btf_ext->field_reloc_info,
-		.desc = "field_reloc",
+		.off = btf_ext->hdr->core_relo_off,
+		.len = btf_ext->hdr->core_relo_len,
+		.min_rec_size = sizeof(struct bpf_core_relo),
+		.ext_info = &btf_ext->core_relo_info,
+		.desc = "core_relo",
 	};
 
 	return btf_ext_setup_info(btf_ext, &param);
@@ -1160,7 +2471,10 @@ static int btf_ext_parse_hdr(__u8 *data, __u32 data_size)
 		return -EINVAL;
 	}
 
-	if (hdr->magic != BTF_MAGIC) {
+	if (hdr->magic == bswap_16(BTF_MAGIC)) {
+		pr_warn("BTF.ext in non-native endianness is not supported\n");
+		return -ENOTSUP;
+	} else if (hdr->magic != BTF_MAGIC) {
 		pr_debug("Invalid BTF.ext magic:%x\n", hdr->magic);
 		return -EINVAL;
 	}
@@ -1223,10 +2537,9 @@ struct btf_ext *btf_ext__new(__u8 *data, __u32 size)
 	if (err)
 		goto done;
 
-	if (btf_ext->hdr->hdr_len <
-	    offsetofend(struct btf_ext_header, field_reloc_len))
+	if (btf_ext->hdr->hdr_len < offsetofend(struct btf_ext_header, core_relo_len))
 		goto done;
-	err = btf_ext_setup_field_reloc(btf_ext);
+	err = btf_ext_setup_core_relos(btf_ext);
 	if (err)
 		goto done;
 
@@ -1481,6 +2794,9 @@ int btf__dedup(struct btf *btf, struct btf_ext *btf_ext,
 		return -EINVAL;
 	}
 
+	if (btf_ensure_modifiable(btf))
+		return -ENOMEM;
+
 	err = btf_dedup_strings(d);
 	if (err < 0) {
 		pr_debug("btf_dedup_strings failed:%d\n", err);
@@ -1581,7 +2897,7 @@ static int btf_dedup_hypot_map_add(struct btf_dedup *d,
 		__u32 *new_list;
 
 		d->hypot_cap += max((size_t)16, d->hypot_cap / 2);
-		new_list = realloc(d->hypot_list, sizeof(__u32) * d->hypot_cap);
+		new_list = libbpf_reallocarray(d->hypot_list, d->hypot_cap, sizeof(__u32));
 		if (!new_list)
 			return -ENOMEM;
 		d->hypot_list = new_list;
@@ -1665,7 +2981,7 @@ static struct btf_dedup *btf_dedup_new(struct btf *btf, struct btf_ext *btf_ext,
 	/* special BTF "void" type is made canonical immediately */
 	d->map[0] = 0;
 	for (i = 1; i <= btf->nr_types; i++) {
-		struct btf_type *t = d->btf->types[i];
+		struct btf_type *t = btf_type_by_id(d->btf, i);
 
 		/* VAR and DATASEC are never deduped and are self-canonical */
 		if (btf_is_var(t) || btf_is_datasec(t))
@@ -1704,7 +3020,7 @@ static int btf_for_each_str_off(struct btf_dedup *d, str_off_fn_t fn, void *ctx)
 	struct btf_type *t;
 
 	for (i = 1; i <= d->btf->nr_types; i++) {
-		t = d->btf->types[i];
+		t = btf_type_by_id(d->btf, i);
 		r = fn(&t->name_off, ctx);
 		if (r)
 			return r;
@@ -1858,8 +3174,7 @@ static int btf_str_remap_offset(__u32 *str_off_ptr, void *ctx)
  */
 static int btf_dedup_strings(struct btf_dedup *d)
 {
-	const struct btf_header *hdr = d->btf->hdr;
-	char *start = (char *)d->btf->nohdr_data + hdr->str_off;
+	char *start = d->btf->strs_data;
 	char *end = start + d->btf->hdr->str_len;
 	char *p = start, *tmp_strs = NULL;
 	struct btf_str_ptrs strs = {
@@ -1871,14 +3186,16 @@ static int btf_dedup_strings(struct btf_dedup *d)
 	int i, j, err = 0, grp_idx;
 	bool grp_used;
 
+	if (d->btf->strs_deduped)
+		return 0;
+
 	/* build index of all strings */
 	while (p < end) {
 		if (strs.cnt + 1 > strs.cap) {
 			struct btf_str_ptr *new_ptrs;
 
 			strs.cap += max(strs.cnt / 2, 16U);
-			new_ptrs = realloc(strs.ptrs,
-					   sizeof(strs.ptrs[0]) * strs.cap);
+			new_ptrs = libbpf_reallocarray(strs.ptrs, strs.cap, sizeof(strs.ptrs[0]));
 			if (!new_ptrs) {
 				err = -ENOMEM;
 				goto done;
@@ -1964,6 +3281,7 @@ static int btf_dedup_strings(struct btf_dedup *d)
 		goto done;
 
 	d->btf->hdr->str_len = end - start;
+	d->btf->strs_deduped = true;
 
 done:
 	free(tmp_strs);
@@ -2240,7 +3558,7 @@ static bool btf_compat_fnproto(struct btf_type *t1, struct btf_type *t2)
  */
 static int btf_dedup_prim_type(struct btf_dedup *d, __u32 type_id)
 {
-	struct btf_type *t = d->btf->types[type_id];
+	struct btf_type *t = btf_type_by_id(d->btf, type_id);
 	struct hashmap_entry *hash_entry;
 	struct btf_type *cand;
 	/* if we don't find equivalent type, then we are canonical */
@@ -2267,7 +3585,7 @@ static int btf_dedup_prim_type(struct btf_dedup *d, __u32 type_id)
 		h = btf_hash_int(t);
 		for_each_dedup_cand(d, hash_entry, h) {
 			cand_id = (__u32)(long)hash_entry->value;
-			cand = d->btf->types[cand_id];
+			cand = btf_type_by_id(d->btf, cand_id);
 			if (btf_equal_int(t, cand)) {
 				new_id = cand_id;
 				break;
@@ -2279,7 +3597,7 @@ static int btf_dedup_prim_type(struct btf_dedup *d, __u32 type_id)
 		h = btf_hash_enum(t);
 		for_each_dedup_cand(d, hash_entry, h) {
 			cand_id = (__u32)(long)hash_entry->value;
-			cand = d->btf->types[cand_id];
+			cand = btf_type_by_id(d->btf, cand_id);
 			if (btf_equal_enum(t, cand)) {
 				new_id = cand_id;
 				break;
@@ -2302,7 +3620,7 @@ static int btf_dedup_prim_type(struct btf_dedup *d, __u32 type_id)
 		h = btf_hash_common(t);
 		for_each_dedup_cand(d, hash_entry, h) {
 			cand_id = (__u32)(long)hash_entry->value;
-			cand = d->btf->types[cand_id];
+			cand = btf_type_by_id(d->btf, cand_id);
 			if (btf_equal_common(t, cand)) {
 				new_id = cand_id;
 				break;
@@ -2361,13 +3679,13 @@ static uint32_t resolve_fwd_id(struct btf_dedup *d, uint32_t type_id)
 {
 	__u32 orig_type_id = type_id;
 
-	if (!btf_is_fwd(d->btf->types[type_id]))
+	if (!btf_is_fwd(btf__type_by_id(d->btf, type_id)))
 		return type_id;
 
 	while (is_type_mapped(d, type_id) && d->map[type_id] != type_id)
 		type_id = d->map[type_id];
 
-	if (!btf_is_fwd(d->btf->types[type_id]))
+	if (!btf_is_fwd(btf__type_by_id(d->btf, type_id)))
 		return type_id;
 
 	return orig_type_id;
@@ -2495,8 +3813,8 @@ static int btf_dedup_is_equiv(struct btf_dedup *d, __u32 cand_id,
 	if (btf_dedup_hypot_map_add(d, canon_id, cand_id))
 		return -ENOMEM;
 
-	cand_type = d->btf->types[cand_id];
-	canon_type = d->btf->types[canon_id];
+	cand_type = btf_type_by_id(d->btf, cand_id);
+	canon_type = btf_type_by_id(d->btf, canon_id);
 	cand_kind = btf_kind(cand_type);
 	canon_kind = btf_kind(canon_type);
 
@@ -2647,8 +3965,8 @@ static void btf_dedup_merge_hypot_map(struct btf_dedup *d)
 		targ_type_id = d->hypot_map[cand_type_id];
 		t_id = resolve_type_id(d, targ_type_id);
 		c_id = resolve_type_id(d, cand_type_id);
-		t_kind = btf_kind(d->btf->types[t_id]);
-		c_kind = btf_kind(d->btf->types[c_id]);
+		t_kind = btf_kind(btf__type_by_id(d->btf, t_id));
+		c_kind = btf_kind(btf__type_by_id(d->btf, c_id));
 		/*
 		 * Resolve FWD into STRUCT/UNION.
 		 * It's ok to resolve FWD into STRUCT/UNION that's not yet
@@ -2716,7 +4034,7 @@ static int btf_dedup_struct_type(struct btf_dedup *d, __u32 type_id)
 	if (d->map[type_id] <= BTF_MAX_NR_TYPES)
 		return 0;
 
-	t = d->btf->types[type_id];
+	t = btf_type_by_id(d->btf, type_id);
 	kind = btf_kind(t);
 
 	if (kind != BTF_KIND_STRUCT && kind != BTF_KIND_UNION)
@@ -2737,7 +4055,7 @@ static int btf_dedup_struct_type(struct btf_dedup *d, __u32 type_id)
 		 * creating a loop (FWD -> STRUCT and STRUCT -> FWD), because
 		 * FWD and compatible STRUCT/UNION are considered equivalent.
 		 */
-		cand_type = d->btf->types[cand_id];
+		cand_type = btf_type_by_id(d->btf, cand_id);
 		if (!btf_shallow_equal_struct(t, cand_type))
 			continue;
 
@@ -2809,7 +4127,7 @@ static int btf_dedup_ref_type(struct btf_dedup *d, __u32 type_id)
 	if (d->map[type_id] <= BTF_MAX_NR_TYPES)
 		return resolve_type_id(d, type_id);
 
-	t = d->btf->types[type_id];
+	t = btf_type_by_id(d->btf, type_id);
 	d->map[type_id] = BTF_IN_PROGRESS_ID;
 
 	switch (btf_kind(t)) {
@@ -2827,7 +4145,7 @@ static int btf_dedup_ref_type(struct btf_dedup *d, __u32 type_id)
 		h = btf_hash_common(t);
 		for_each_dedup_cand(d, hash_entry, h) {
 			cand_id = (__u32)(long)hash_entry->value;
-			cand = d->btf->types[cand_id];
+			cand = btf_type_by_id(d->btf, cand_id);
 			if (btf_equal_common(t, cand)) {
 				new_id = cand_id;
 				break;
@@ -2851,7 +4169,7 @@ static int btf_dedup_ref_type(struct btf_dedup *d, __u32 type_id)
 		h = btf_hash_array(t);
 		for_each_dedup_cand(d, hash_entry, h) {
 			cand_id = (__u32)(long)hash_entry->value;
-			cand = d->btf->types[cand_id];
+			cand = btf_type_by_id(d->btf, cand_id);
 			if (btf_equal_array(t, cand)) {
 				new_id = cand_id;
 				break;
@@ -2883,7 +4201,7 @@ static int btf_dedup_ref_type(struct btf_dedup *d, __u32 type_id)
 		h = btf_hash_fnproto(t);
 		for_each_dedup_cand(d, hash_entry, h) {
 			cand_id = (__u32)(long)hash_entry->value;
-			cand = d->btf->types[cand_id];
+			cand = btf_type_by_id(d->btf, cand_id);
 			if (btf_equal_fnproto(t, cand)) {
 				new_id = cand_id;
 				break;
@@ -2931,9 +4249,9 @@ static int btf_dedup_ref_types(struct btf_dedup *d)
  */
 static int btf_dedup_compact_types(struct btf_dedup *d)
 {
-	struct btf_type **new_types;
+	__u32 *new_offs;
 	__u32 next_type_id = 1;
-	char *types_start, *p;
+	void *p;
 	int i, len;
 
 	/* we are going to reuse hypot_map to store compaction remapping */
@@ -2941,41 +4259,34 @@ static int btf_dedup_compact_types(struct btf_dedup *d)
 	for (i = 1; i <= d->btf->nr_types; i++)
 		d->hypot_map[i] = BTF_UNPROCESSED_ID;
 
-	types_start = d->btf->nohdr_data + d->btf->hdr->type_off;
-	p = types_start;
+	p = d->btf->types_data;
 
 	for (i = 1; i <= d->btf->nr_types; i++) {
 		if (d->map[i] != i)
 			continue;
 
-		len = btf_type_size(d->btf->types[i]);
+		len = btf_type_size(btf__type_by_id(d->btf, i));
 		if (len < 0)
 			return len;
 
-		memmove(p, d->btf->types[i], len);
+		memmove(p, btf__type_by_id(d->btf, i), len);
 		d->hypot_map[i] = next_type_id;
-		d->btf->types[next_type_id] = (struct btf_type *)p;
+		d->btf->type_offs[next_type_id] = p - d->btf->types_data;
 		p += len;
 		next_type_id++;
 	}
 
 	/* shrink struct btf's internal types index and update btf_header */
 	d->btf->nr_types = next_type_id - 1;
-	d->btf->types_size = d->btf->nr_types;
-	d->btf->hdr->type_len = p - types_start;
-	new_types = realloc(d->btf->types,
-			    (1 + d->btf->nr_types) * sizeof(struct btf_type *));
-	if (!new_types)
+	d->btf->type_offs_cap = d->btf->nr_types + 1;
+	d->btf->hdr->type_len = p - d->btf->types_data;
+	new_offs = libbpf_reallocarray(d->btf->type_offs, d->btf->type_offs_cap,
+				       sizeof(*new_offs));
+	if (!new_offs)
 		return -ENOMEM;
-	d->btf->types = new_types;
-
-	/* make sure string section follows type information without gaps */
-	d->btf->hdr->str_off = p - (char *)d->btf->nohdr_data;
-	memmove(p, d->btf->strings, d->btf->hdr->str_len);
-	d->btf->strings = p;
-	p += d->btf->hdr->str_len;
-
-	d->btf->data_size = p - (char *)d->btf->data;
+	d->btf->type_offs = new_offs;
+	d->btf->hdr->str_off = d->btf->hdr->type_len;
+	d->btf->raw_size = d->btf->hdr->hdr_len + d->btf->hdr->type_len + d->btf->hdr->str_len;
 	return 0;
 }
 
@@ -3008,7 +4319,7 @@ static int btf_dedup_remap_type_id(struct btf_dedup *d, __u32 type_id)
  */
 static int btf_dedup_remap_type(struct btf_dedup *d, __u32 type_id)
 {
-	struct btf_type *t = d->btf->types[type_id];
+	struct btf_type *t = btf_type_by_id(d->btf, type_id);
 	int i, r;
 
 	switch (btf_kind(t)) {
diff --git a/tools/lib/bpf/btf.h b/tools/lib/bpf/btf.h
index 1ca14448df4c..57247240a20a 100644
--- a/tools/lib/bpf/btf.h
+++ b/tools/lib/bpf/btf.h
@@ -5,6 +5,7 @@
 #define __LIBBPF_BTF_H
 
 #include <stdarg.h>
+#include <stdbool.h>
 #include <linux/btf.h>
 #include <linux/types.h>
 
@@ -24,46 +25,14 @@ struct btf_type;
 
 struct bpf_object;
 
-/*
- * The .BTF.ext ELF section layout defined as
- *   struct btf_ext_header
- *   func_info subsection
- *
- * The func_info subsection layout:
- *   record size for struct bpf_func_info in the func_info subsection
- *   struct btf_sec_func_info for section #1
- *   a list of bpf_func_info records for section #1
- *     where struct bpf_func_info mimics one in include/uapi/linux/bpf.h
- *     but may not be identical
- *   struct btf_sec_func_info for section #2
- *   a list of bpf_func_info records for section #2
- *   ......
- *
- * Note that the bpf_func_info record size in .BTF.ext may not
- * be the same as the one defined in include/uapi/linux/bpf.h.
- * The loader should ensure that record_size meets minimum
- * requirement and pass the record as is to the kernel. The
- * kernel will handle the func_info properly based on its contents.
- */
-struct btf_ext_header {
-	__u16	magic;
-	__u8	version;
-	__u8	flags;
-	__u32	hdr_len;
-
-	/* All offsets are in bytes relative to the end of this header */
-	__u32	func_info_off;
-	__u32	func_info_len;
-	__u32	line_info_off;
-	__u32	line_info_len;
-
-	/* optional part of .BTF.ext header */
-	__u32	field_reloc_off;
-	__u32	field_reloc_len;
+enum btf_endianness {
+	BTF_LITTLE_ENDIAN = 0,
+	BTF_BIG_ENDIAN = 1,
 };
 
 LIBBPF_API void btf__free(struct btf *btf);
 LIBBPF_API struct btf *btf__new(const void *data, __u32 size);
+LIBBPF_API struct btf *btf__new_empty(void);
 LIBBPF_API struct btf *btf__parse(const char *path, struct btf_ext **btf_ext);
 LIBBPF_API struct btf *btf__parse_elf(const char *path, struct btf_ext **btf_ext);
 LIBBPF_API struct btf *btf__parse_raw(const char *path);
@@ -78,6 +47,8 @@ LIBBPF_API const struct btf_type *btf__type_by_id(const struct btf *btf,
 						  __u32 id);
 LIBBPF_API size_t btf__pointer_size(const struct btf *btf);
 LIBBPF_API int btf__set_pointer_size(struct btf *btf, size_t ptr_sz);
+LIBBPF_API enum btf_endianness btf__endianness(const struct btf *btf);
+LIBBPF_API int btf__set_endianness(struct btf *btf, enum btf_endianness endian);
 LIBBPF_API __s64 btf__resolve_size(const struct btf *btf, __u32 type_id);
 LIBBPF_API int btf__resolve_type(const struct btf *btf, __u32 type_id);
 LIBBPF_API int btf__align_of(const struct btf *btf, __u32 id);
@@ -85,6 +56,7 @@ LIBBPF_API int btf__fd(const struct btf *btf);
 LIBBPF_API void btf__set_fd(struct btf *btf, int fd);
 LIBBPF_API const void *btf__get_raw_data(const struct btf *btf, __u32 *size);
 LIBBPF_API const char *btf__name_by_offset(const struct btf *btf, __u32 offset);
+LIBBPF_API const char *btf__str_by_offset(const struct btf *btf, __u32 offset);
 LIBBPF_API int btf__get_from_id(__u32 id, struct btf **btf);
 LIBBPF_API int btf__get_map_kv_tids(const struct btf *btf, const char *map_name,
 				    __u32 expected_key_size,
@@ -95,19 +67,62 @@ LIBBPF_API struct btf_ext *btf_ext__new(__u8 *data, __u32 size);
 LIBBPF_API void btf_ext__free(struct btf_ext *btf_ext);
 LIBBPF_API const void *btf_ext__get_raw_data(const struct btf_ext *btf_ext,
 					     __u32 *size);
-LIBBPF_API int btf_ext__reloc_func_info(const struct btf *btf,
-					const struct btf_ext *btf_ext,
-					const char *sec_name, __u32 insns_cnt,
-					void **func_info, __u32 *cnt);
-LIBBPF_API int btf_ext__reloc_line_info(const struct btf *btf,
-					const struct btf_ext *btf_ext,
-					const char *sec_name, __u32 insns_cnt,
-					void **line_info, __u32 *cnt);
+LIBBPF_API LIBBPF_DEPRECATED("btf_ext__reloc_func_info was never meant as a public API and has wrong assumptions embedded in it; it will be removed in the future libbpf versions")
+int btf_ext__reloc_func_info(const struct btf *btf,
+			     const struct btf_ext *btf_ext,
+			     const char *sec_name, __u32 insns_cnt,
+			     void **func_info, __u32 *cnt);
+LIBBPF_API LIBBPF_DEPRECATED("btf_ext__reloc_line_info was never meant as a public API and has wrong assumptions embedded in it; it will be removed in the future libbpf versions")
+int btf_ext__reloc_line_info(const struct btf *btf,
+			     const struct btf_ext *btf_ext,
+			     const char *sec_name, __u32 insns_cnt,
+			     void **line_info, __u32 *cnt);
 LIBBPF_API __u32 btf_ext__func_info_rec_size(const struct btf_ext *btf_ext);
 LIBBPF_API __u32 btf_ext__line_info_rec_size(const struct btf_ext *btf_ext);
 
 LIBBPF_API struct btf *libbpf_find_kernel_btf(void);
 
+LIBBPF_API int btf__find_str(struct btf *btf, const char *s);
+LIBBPF_API int btf__add_str(struct btf *btf, const char *s);
+
+LIBBPF_API int btf__add_int(struct btf *btf, const char *name, size_t byte_sz, int encoding);
+LIBBPF_API int btf__add_ptr(struct btf *btf, int ref_type_id);
+LIBBPF_API int btf__add_array(struct btf *btf,
+			      int index_type_id, int elem_type_id, __u32 nr_elems);
+/* struct/union construction APIs */
+LIBBPF_API int btf__add_struct(struct btf *btf, const char *name, __u32 sz);
+LIBBPF_API int btf__add_union(struct btf *btf, const char *name, __u32 sz);
+LIBBPF_API int btf__add_field(struct btf *btf, const char *name, int field_type_id,
+			      __u32 bit_offset, __u32 bit_size);
+
+/* enum construction APIs */
+LIBBPF_API int btf__add_enum(struct btf *btf, const char *name, __u32 bytes_sz);
+LIBBPF_API int btf__add_enum_value(struct btf *btf, const char *name, __s64 value);
+
+enum btf_fwd_kind {
+	BTF_FWD_STRUCT = 0,
+	BTF_FWD_UNION = 1,
+	BTF_FWD_ENUM = 2,
+};
+
+LIBBPF_API int btf__add_fwd(struct btf *btf, const char *name, enum btf_fwd_kind fwd_kind);
+LIBBPF_API int btf__add_typedef(struct btf *btf, const char *name, int ref_type_id);
+LIBBPF_API int btf__add_volatile(struct btf *btf, int ref_type_id);
+LIBBPF_API int btf__add_const(struct btf *btf, int ref_type_id);
+LIBBPF_API int btf__add_restrict(struct btf *btf, int ref_type_id);
+
+/* func and func_proto construction APIs */
+LIBBPF_API int btf__add_func(struct btf *btf, const char *name,
+			     enum btf_func_linkage linkage, int proto_type_id);
+LIBBPF_API int btf__add_func_proto(struct btf *btf, int ret_type_id);
+LIBBPF_API int btf__add_func_param(struct btf *btf, const char *name, int type_id);
+
+/* var & datasec construction APIs */
+LIBBPF_API int btf__add_var(struct btf *btf, const char *name, int linkage, int type_id);
+LIBBPF_API int btf__add_datasec(struct btf *btf, const char *name, __u32 byte_sz);
+LIBBPF_API int btf__add_datasec_var_info(struct btf *btf, int var_type_id,
+					 __u32 offset, __u32 byte_sz);
+
 struct btf_dedup_opts {
 	unsigned int dedup_table_size;
 	bool dont_resolve_fwds;
diff --git a/tools/lib/bpf/btf_dump.c b/tools/lib/bpf/btf_dump.c
index 57c00fa63932..2f9d685bd522 100644
--- a/tools/lib/bpf/btf_dump.c
+++ b/tools/lib/bpf/btf_dump.c
@@ -19,9 +19,6 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
-/* make sure libbpf doesn't use kernel-only integer typedefs */
-#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
-
 static const char PREFIXES[] = "\t\t\t\t\t\t\t\t\t\t\t\t\t";
 static const size_t PREFIX_CNT = sizeof(PREFIXES) - 1;
 
@@ -63,11 +60,14 @@ struct btf_dump {
 	struct btf_dump_opts opts;
 	int ptr_sz;
 	bool strip_mods;
+	int last_id;
 
 	/* per-type auxiliary state */
 	struct btf_dump_type_aux_state *type_states;
+	size_t type_states_cap;
 	/* per-type optional cached unique name, must be freed, if present */
 	const char **cached_names;
+	size_t cached_names_cap;
 
 	/* topo-sorted list of dependent type definitions */
 	__u32 *emit_queue;
@@ -93,14 +93,7 @@ struct btf_dump {
 
 static size_t str_hash_fn(const void *key, void *ctx)
 {
-	const char *s = key;
-	size_t h = 0;
-
-	while (*s) {
-		h = h * 31 + *s;
-		s++;
-	}
-	return h;
+	return str_hash(key);
 }
 
 static bool str_equal_fn(const void *a, const void *b, void *ctx)
@@ -123,6 +116,7 @@ static void btf_dump_printf(const struct btf_dump *d, const char *fmt, ...)
 }
 
 static int btf_dump_mark_referenced(struct btf_dump *d);
+static int btf_dump_resize(struct btf_dump *d);
 
 struct btf_dump *btf_dump__new(const struct btf *btf,
 			       const struct btf_ext *btf_ext,
@@ -154,25 +148,8 @@ struct btf_dump *btf_dump__new(const struct btf *btf,
 		d->ident_names = NULL;
 		goto err;
 	}
-	d->type_states = calloc(1 + btf__get_nr_types(d->btf),
-				sizeof(d->type_states[0]));
-	if (!d->type_states) {
-		err = -ENOMEM;
-		goto err;
-	}
-	d->cached_names = calloc(1 + btf__get_nr_types(d->btf),
-				 sizeof(d->cached_names[0]));
-	if (!d->cached_names) {
-		err = -ENOMEM;
-		goto err;
-	}
 
-	/* VOID is special */
-	d->type_states[0].order_state = ORDERED;
-	d->type_states[0].emit_state = EMITTED;
-
-	/* eagerly determine referenced types for anon enums */
-	err = btf_dump_mark_referenced(d);
+	err = btf_dump_resize(d);
 	if (err)
 		goto err;
 
@@ -182,9 +159,38 @@ err:
 	return ERR_PTR(err);
 }
 
+static int btf_dump_resize(struct btf_dump *d)
+{
+	int err, last_id = btf__get_nr_types(d->btf);
+
+	if (last_id <= d->last_id)
+		return 0;
+
+	if (btf_ensure_mem((void **)&d->type_states, &d->type_states_cap,
+			   sizeof(*d->type_states), last_id + 1))
+		return -ENOMEM;
+	if (btf_ensure_mem((void **)&d->cached_names, &d->cached_names_cap,
+			   sizeof(*d->cached_names), last_id + 1))
+		return -ENOMEM;
+
+	if (d->last_id == 0) {
+		/* VOID is special */
+		d->type_states[0].order_state = ORDERED;
+		d->type_states[0].emit_state = EMITTED;
+	}
+
+	/* eagerly determine referenced types for anon enums */
+	err = btf_dump_mark_referenced(d);
+	if (err)
+		return err;
+
+	d->last_id = last_id;
+	return 0;
+}
+
 void btf_dump__free(struct btf_dump *d)
 {
-	int i, cnt;
+	int i;
 
 	if (IS_ERR_OR_NULL(d))
 		return;
@@ -192,7 +198,7 @@ void btf_dump__free(struct btf_dump *d)
 	free(d->type_states);
 	if (d->cached_names) {
 		/* any set cached name is owned by us and should be freed */
-		for (i = 0, cnt = btf__get_nr_types(d->btf); i <= cnt; i++) {
+		for (i = 0; i <= d->last_id; i++) {
 			if (d->cached_names[i])
 				free((void *)d->cached_names[i]);
 		}
@@ -232,6 +238,10 @@ int btf_dump__dump_type(struct btf_dump *d, __u32 id)
 	if (id > btf__get_nr_types(d->btf))
 		return -EINVAL;
 
+	err = btf_dump_resize(d);
+	if (err)
+		return err;
+
 	d->emit_queue_cnt = 0;
 	err = btf_dump_order_type(d, id, false);
 	if (err < 0)
@@ -261,7 +271,7 @@ static int btf_dump_mark_referenced(struct btf_dump *d)
 	const struct btf_type *t;
 	__u16 vlen;
 
-	for (i = 1; i <= n; i++) {
+	for (i = d->last_id + 1; i <= n; i++) {
 		t = btf__type_by_id(d->btf, i);
 		vlen = btf_vlen(t);
 
@@ -316,6 +326,7 @@ static int btf_dump_mark_referenced(struct btf_dump *d)
 	}
 	return 0;
 }
+
 static int btf_dump_add_emit_queue_id(struct btf_dump *d, __u32 id)
 {
 	__u32 *new_queue;
@@ -323,8 +334,7 @@ static int btf_dump_add_emit_queue_id(struct btf_dump *d, __u32 id)
 
 	if (d->emit_queue_cnt >= d->emit_queue_cap) {
 		new_cap = max(16, d->emit_queue_cap * 3 / 2);
-		new_queue = realloc(d->emit_queue,
-				    new_cap * sizeof(new_queue[0]));
+		new_queue = libbpf_reallocarray(d->emit_queue, new_cap, sizeof(new_queue[0]));
 		if (!new_queue)
 			return -ENOMEM;
 		d->emit_queue = new_queue;
@@ -1003,8 +1013,7 @@ static int btf_dump_push_decl_stack_id(struct btf_dump *d, __u32 id)
 
 	if (d->decl_stack_cnt >= d->decl_stack_cap) {
 		new_cap = max(16, d->decl_stack_cap * 3 / 2);
-		new_stack = realloc(d->decl_stack,
-				    new_cap * sizeof(new_stack[0]));
+		new_stack = libbpf_reallocarray(d->decl_stack, new_cap, sizeof(new_stack[0]));
 		if (!new_stack)
 			return -ENOMEM;
 		d->decl_stack = new_stack;
@@ -1061,11 +1070,15 @@ int btf_dump__emit_type_decl(struct btf_dump *d, __u32 id,
 			     const struct btf_dump_emit_type_decl_opts *opts)
 {
 	const char *fname;
-	int lvl;
+	int lvl, err;
 
 	if (!OPTS_VALID(opts, btf_dump_emit_type_decl_opts))
 		return -EINVAL;
 
+	err = btf_dump_resize(d);
+	if (err)
+		return -EINVAL;
+
 	fname = OPTS_GET(opts, field_name, "");
 	lvl = OPTS_GET(opts, indent_level, 0);
 	d->strip_mods = OPTS_GET(opts, strip_mods, false);
diff --git a/tools/lib/bpf/hashmap.c b/tools/lib/bpf/hashmap.c
index a405dad068f5..3c20b126d60d 100644
--- a/tools/lib/bpf/hashmap.c
+++ b/tools/lib/bpf/hashmap.c
@@ -15,6 +15,9 @@
 /* make sure libbpf doesn't use kernel-only integer typedefs */
 #pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
 
+/* prevent accidental re-addition of reallocarray() */
+#pragma GCC poison reallocarray
+
 /* start with 4 buckets */
 #define HASHMAP_MIN_CAP_BITS 2
 
diff --git a/tools/lib/bpf/hashmap.h b/tools/lib/bpf/hashmap.h
index e0af36b0e5d8..d9b385fe808c 100644
--- a/tools/lib/bpf/hashmap.h
+++ b/tools/lib/bpf/hashmap.h
@@ -25,6 +25,18 @@ static inline size_t hash_bits(size_t h, int bits)
 #endif
 }
 
+/* generic C-string hashing function */
+static inline size_t str_hash(const char *s)
+{
+	size_t h = 0;
+
+	while (*s) {
+		h = h * 31 + *s;
+		s++;
+	}
+	return h;
+}
+
 typedef size_t (*hashmap_hash_fn)(const void *key, void *ctx);
 typedef bool (*hashmap_equal_fn)(const void *key1, const void *key2, void *ctx);
 
diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index e493d6048143..313034117070 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -44,7 +44,6 @@
 #include <sys/vfs.h>
 #include <sys/utsname.h>
 #include <sys/resource.h>
-#include <tools/libc_compat.h>
 #include <libelf.h>
 #include <gelf.h>
 #include <zlib.h>
@@ -56,9 +55,6 @@
 #include "libbpf_internal.h"
 #include "hashmap.h"
 
-/* make sure libbpf doesn't use kernel-only integer typedefs */
-#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
-
 #ifndef EM_BPF
 #define EM_BPF 247
 #endif
@@ -67,6 +63,8 @@
 #define BPF_FS_MAGIC		0xcafe4a11
 #endif
 
+#define BPF_INSN_SZ (sizeof(struct bpf_insn))
+
 /* vsprintf() in __base_pr() uses nonliteral format string. It may break
  * compilation if user enables corresponding warning. Disable it explicitly.
  */
@@ -75,8 +73,6 @@
 #define __printf(a, b)	__attribute__((format(printf, a, b)))
 
 static struct bpf_map *bpf_object__add_map(struct bpf_object *obj);
-static struct bpf_program *bpf_object__find_prog_by_idx(struct bpf_object *obj,
-							int idx);
 static const struct btf_type *
 skip_mods_and_typedefs(const struct btf *btf, __u32 id, __u32 *res_id);
 
@@ -154,34 +150,37 @@ static void pr_perm_msg(int err)
 	___err; })
 #endif
 
-#ifdef HAVE_LIBELF_MMAP_SUPPORT
-# define LIBBPF_ELF_C_READ_MMAP ELF_C_READ_MMAP
-#else
-# define LIBBPF_ELF_C_READ_MMAP ELF_C_READ
-#endif
-
 static inline __u64 ptr_to_u64(const void *ptr)
 {
 	return (__u64) (unsigned long) ptr;
 }
 
-struct bpf_capabilities {
+enum kern_feature_id {
 	/* v4.14: kernel support for program & map names. */
-	__u32 name:1;
+	FEAT_PROG_NAME,
 	/* v5.2: kernel support for global data sections. */
-	__u32 global_data:1;
+	FEAT_GLOBAL_DATA,
+	/* BTF support */
+	FEAT_BTF,
 	/* BTF_KIND_FUNC and BTF_KIND_FUNC_PROTO support */
-	__u32 btf_func:1;
+	FEAT_BTF_FUNC,
 	/* BTF_KIND_VAR and BTF_KIND_DATASEC support */
-	__u32 btf_datasec:1;
-	/* BPF_F_MMAPABLE is supported for arrays */
-	__u32 array_mmap:1;
+	FEAT_BTF_DATASEC,
 	/* BTF_FUNC_GLOBAL is supported */
-	__u32 btf_func_global:1;
+	FEAT_BTF_GLOBAL_FUNC,
+	/* BPF_F_MMAPABLE is supported for arrays */
+	FEAT_ARRAY_MMAP,
 	/* kernel support for expected_attach_type in BPF_PROG_LOAD */
-	__u32 exp_attach_type:1;
+	FEAT_EXP_ATTACH_TYPE,
+	/* bpf_probe_read_{kernel,user}[_str] helpers */
+	FEAT_PROBE_READ_KERN,
+	/* BPF_PROG_BIND_MAP is supported */
+	FEAT_PROG_BIND_MAP,
+	__FEAT_CNT,
 };
 
+static bool kernel_supports(enum kern_feature_id feat_id);
+
 enum reloc_type {
 	RELO_LD64,
 	RELO_CALL,
@@ -194,6 +193,7 @@ struct reloc_desc {
 	int insn_idx;
 	int map_idx;
 	int sym_off;
+	bool processed;
 };
 
 struct bpf_sec_def;
@@ -209,6 +209,7 @@ struct bpf_sec_def {
 	bool is_exp_attach_type_optional;
 	bool is_attachable;
 	bool is_attach_btf;
+	bool is_sleepable;
 	attach_fn_t attach_fn;
 };
 
@@ -217,20 +218,45 @@ struct bpf_sec_def {
  * linux/filter.h.
  */
 struct bpf_program {
-	/* Index in elf obj file, for relocation use. */
-	int idx;
-	char *name;
-	int prog_ifindex;
-	char *section_name;
 	const struct bpf_sec_def *sec_def;
-	/* section_name with / replaced by _; makes recursive pinning
+	char *sec_name;
+	size_t sec_idx;
+	/* this program's instruction offset (in number of instructions)
+	 * within its containing ELF section
+	 */
+	size_t sec_insn_off;
+	/* number of original instructions in ELF section belonging to this
+	 * program, not taking into account subprogram instructions possible
+	 * appended later during relocation
+	 */
+	size_t sec_insn_cnt;
+	/* Offset (in number of instructions) of the start of instruction
+	 * belonging to this BPF program  within its containing main BPF
+	 * program. For the entry-point (main) BPF program, this is always
+	 * zero. For a sub-program, this gets reset before each of main BPF
+	 * programs are processed and relocated and is used to determined
+	 * whether sub-program was already appended to the main program, and
+	 * if yes, at which instruction offset.
+	 */
+	size_t sub_insn_off;
+
+	char *name;
+	/* sec_name with / replaced by _; makes recursive pinning
 	 * in bpf_object__pin_programs easier
 	 */
 	char *pin_name;
+
+	/* instructions that belong to BPF program; insns[0] is located at
+	 * sec_insn_off instruction within its ELF section in ELF file, so
+	 * when mapping ELF file instruction index to the local instruction,
+	 * one needs to subtract sec_insn_off; and vice versa.
+	 */
 	struct bpf_insn *insns;
-	size_t insns_cnt, main_prog_cnt;
-	enum bpf_prog_type type;
-	bool load;
+	/* actual number of instruction in this BPF program's image; for
+	 * entry-point BPF programs this includes the size of main program
+	 * itself plus all the used sub-programs, appended at the end
+	 */
+	size_t insns_cnt;
 
 	struct reloc_desc *reloc_desc;
 	int nr_reloc;
@@ -246,15 +272,16 @@ struct bpf_program {
 	void *priv;
 	bpf_program_clear_priv_t clear_priv;
 
+	bool load;
+	enum bpf_prog_type type;
 	enum bpf_attach_type expected_attach_type;
+	int prog_ifindex;
 	__u32 attach_btf_id;
 	__u32 attach_prog_fd;
 	void *func_info;
 	__u32 func_info_rec_size;
 	__u32 func_info_cnt;
 
-	struct bpf_capabilities *caps;
-
 	void *line_info;
 	__u32 line_info_rec_size;
 	__u32 line_info_cnt;
@@ -363,6 +390,12 @@ struct extern_desc {
 		} kcfg;
 		struct {
 			unsigned long long addr;
+
+			/* target btf_id of the corresponding kernel var. */
+			int vmlinux_btf_id;
+
+			/* local btf_id of the ksym extern's type. */
+			__u32 type_id;
 		} ksym;
 	};
 };
@@ -384,9 +417,10 @@ struct bpf_object {
 	struct extern_desc *externs;
 	int nr_extern;
 	int kconfig_map_idx;
+	int rodata_map_idx;
 
 	bool loaded;
-	bool has_pseudo_calls;
+	bool has_subcalls;
 
 	/*
 	 * Information when doing elf related work. Only valid if fd
@@ -403,6 +437,7 @@ struct bpf_object {
 		Elf_Data *rodata;
 		Elf_Data *bss;
 		Elf_Data *st_ops_data;
+		size_t shstrndx; /* section index for section name strings */
 		size_t strtabidx;
 		struct {
 			GElf_Shdr shdr;
@@ -436,12 +471,20 @@ struct bpf_object {
 	void *priv;
 	bpf_object_clear_priv_t clear_priv;
 
-	struct bpf_capabilities caps;
-
 	char path[];
 };
 #define obj_elf_valid(o)	((o)->efile.elf)
 
+static const char *elf_sym_str(const struct bpf_object *obj, size_t off);
+static const char *elf_sec_str(const struct bpf_object *obj, size_t off);
+static Elf_Scn *elf_sec_by_idx(const struct bpf_object *obj, size_t idx);
+static Elf_Scn *elf_sec_by_name(const struct bpf_object *obj, const char *name);
+static int elf_sec_hdr(const struct bpf_object *obj, Elf_Scn *scn, GElf_Shdr *hdr);
+static const char *elf_sec_name(const struct bpf_object *obj, Elf_Scn *scn);
+static Elf_Data *elf_sec_data(const struct bpf_object *obj, Elf_Scn *scn);
+static int elf_sym_by_sec_off(const struct bpf_object *obj, size_t sec_idx,
+			      size_t off, __u32 sym_type, GElf_Sym *sym);
+
 void bpf_program__unload(struct bpf_program *prog)
 {
 	int i;
@@ -481,159 +524,160 @@ static void bpf_program__exit(struct bpf_program *prog)
 
 	bpf_program__unload(prog);
 	zfree(&prog->name);
-	zfree(&prog->section_name);
+	zfree(&prog->sec_name);
 	zfree(&prog->pin_name);
 	zfree(&prog->insns);
 	zfree(&prog->reloc_desc);
 
 	prog->nr_reloc = 0;
 	prog->insns_cnt = 0;
-	prog->idx = -1;
+	prog->sec_idx = -1;
 }
 
 static char *__bpf_program__pin_name(struct bpf_program *prog)
 {
 	char *name, *p;
 
-	name = p = strdup(prog->section_name);
+	name = p = strdup(prog->sec_name);
 	while ((p = strchr(p, '/')))
 		*p = '_';
 
 	return name;
 }
 
+static bool insn_is_subprog_call(const struct bpf_insn *insn)
+{
+	return BPF_CLASS(insn->code) == BPF_JMP &&
+	       BPF_OP(insn->code) == BPF_CALL &&
+	       BPF_SRC(insn->code) == BPF_K &&
+	       insn->src_reg == BPF_PSEUDO_CALL &&
+	       insn->dst_reg == 0 &&
+	       insn->off == 0;
+}
+
 static int
-bpf_program__init(void *data, size_t size, char *section_name, int idx,
-		  struct bpf_program *prog)
+bpf_object__init_prog(struct bpf_object *obj, struct bpf_program *prog,
+		      const char *name, size_t sec_idx, const char *sec_name,
+		      size_t sec_off, void *insn_data, size_t insn_data_sz)
 {
-	const size_t bpf_insn_sz = sizeof(struct bpf_insn);
+	int i;
 
-	if (size == 0 || size % bpf_insn_sz) {
-		pr_warn("corrupted section '%s', size: %zu\n",
-			section_name, size);
+	if (insn_data_sz == 0 || insn_data_sz % BPF_INSN_SZ || sec_off % BPF_INSN_SZ) {
+		pr_warn("sec '%s': corrupted program '%s', offset %zu, size %zu\n",
+			sec_name, name, sec_off, insn_data_sz);
 		return -EINVAL;
 	}
 
 	memset(prog, 0, sizeof(*prog));
+	prog->obj = obj;
+
+	prog->sec_idx = sec_idx;
+	prog->sec_insn_off = sec_off / BPF_INSN_SZ;
+	prog->sec_insn_cnt = insn_data_sz / BPF_INSN_SZ;
+	/* insns_cnt can later be increased by appending used subprograms */
+	prog->insns_cnt = prog->sec_insn_cnt;
+
+	prog->type = BPF_PROG_TYPE_UNSPEC;
+	prog->load = true;
+
+	prog->instances.fds = NULL;
+	prog->instances.nr = -1;
 
-	prog->section_name = strdup(section_name);
-	if (!prog->section_name) {
-		pr_warn("failed to alloc name for prog under section(%d) %s\n",
-			idx, section_name);
+	prog->sec_name = strdup(sec_name);
+	if (!prog->sec_name)
+		goto errout;
+
+	prog->name = strdup(name);
+	if (!prog->name)
 		goto errout;
-	}
 
 	prog->pin_name = __bpf_program__pin_name(prog);
-	if (!prog->pin_name) {
-		pr_warn("failed to alloc pin name for prog under section(%d) %s\n",
-			idx, section_name);
+	if (!prog->pin_name)
 		goto errout;
-	}
 
-	prog->insns = malloc(size);
-	if (!prog->insns) {
-		pr_warn("failed to alloc insns for prog under section %s\n",
-			section_name);
+	prog->insns = malloc(insn_data_sz);
+	if (!prog->insns)
 		goto errout;
+	memcpy(prog->insns, insn_data, insn_data_sz);
+
+	for (i = 0; i < prog->insns_cnt; i++) {
+		if (insn_is_subprog_call(&prog->insns[i])) {
+			obj->has_subcalls = true;
+			break;
+		}
 	}
-	prog->insns_cnt = size / bpf_insn_sz;
-	memcpy(prog->insns, data, size);
-	prog->idx = idx;
-	prog->instances.fds = NULL;
-	prog->instances.nr = -1;
-	prog->type = BPF_PROG_TYPE_UNSPEC;
-	prog->load = true;
 
 	return 0;
 errout:
+	pr_warn("sec '%s': failed to allocate memory for prog '%s'\n", sec_name, name);
 	bpf_program__exit(prog);
 	return -ENOMEM;
 }
 
 static int
-bpf_object__add_program(struct bpf_object *obj, void *data, size_t size,
-			char *section_name, int idx)
+bpf_object__add_programs(struct bpf_object *obj, Elf_Data *sec_data,
+			 const char *sec_name, int sec_idx)
 {
-	struct bpf_program prog, *progs;
+	struct bpf_program *prog, *progs;
+	void *data = sec_data->d_buf;
+	size_t sec_sz = sec_data->d_size, sec_off, prog_sz;
 	int nr_progs, err;
+	const char *name;
+	GElf_Sym sym;
 
-	err = bpf_program__init(data, size, section_name, idx, &prog);
-	if (err)
-		return err;
-
-	prog.caps = &obj->caps;
 	progs = obj->programs;
 	nr_progs = obj->nr_programs;
+	sec_off = 0;
 
-	progs = reallocarray(progs, nr_progs + 1, sizeof(progs[0]));
-	if (!progs) {
-		/*
-		 * In this case the original obj->programs
-		 * is still valid, so don't need special treat for
-		 * bpf_close_object().
-		 */
-		pr_warn("failed to alloc a new program under section '%s'\n",
-			section_name);
-		bpf_program__exit(&prog);
-		return -ENOMEM;
-	}
-
-	pr_debug("found program %s\n", prog.section_name);
-	obj->programs = progs;
-	obj->nr_programs = nr_progs + 1;
-	prog.obj = obj;
-	progs[nr_progs] = prog;
-	return 0;
-}
-
-static int
-bpf_object__init_prog_names(struct bpf_object *obj)
-{
-	Elf_Data *symbols = obj->efile.symbols;
-	struct bpf_program *prog;
-	size_t pi, si;
+	while (sec_off < sec_sz) {
+		if (elf_sym_by_sec_off(obj, sec_idx, sec_off, STT_FUNC, &sym)) {
+			pr_warn("sec '%s': failed to find program symbol at offset %zu\n",
+				sec_name, sec_off);
+			return -LIBBPF_ERRNO__FORMAT;
+		}
 
-	for (pi = 0; pi < obj->nr_programs; pi++) {
-		const char *name = NULL;
+		prog_sz = sym.st_size;
 
-		prog = &obj->programs[pi];
+		name = elf_sym_str(obj, sym.st_name);
+		if (!name) {
+			pr_warn("sec '%s': failed to get symbol name for offset %zu\n",
+				sec_name, sec_off);
+			return -LIBBPF_ERRNO__FORMAT;
+		}
 
-		for (si = 0; si < symbols->d_size / sizeof(GElf_Sym) && !name;
-		     si++) {
-			GElf_Sym sym;
+		if (sec_off + prog_sz > sec_sz) {
+			pr_warn("sec '%s': program at offset %zu crosses section boundary\n",
+				sec_name, sec_off);
+			return -LIBBPF_ERRNO__FORMAT;
+		}
 
-			if (!gelf_getsym(symbols, si, &sym))
-				continue;
-			if (sym.st_shndx != prog->idx)
-				continue;
-			if (GELF_ST_BIND(sym.st_info) != STB_GLOBAL)
-				continue;
+		pr_debug("sec '%s': found program '%s' at insn offset %zu (%zu bytes), code size %zu insns (%zu bytes)\n",
+			 sec_name, name, sec_off / BPF_INSN_SZ, sec_off, prog_sz / BPF_INSN_SZ, prog_sz);
 
-			name = elf_strptr(obj->efile.elf,
-					  obj->efile.strtabidx,
-					  sym.st_name);
-			if (!name) {
-				pr_warn("failed to get sym name string for prog %s\n",
-					prog->section_name);
-				return -LIBBPF_ERRNO__LIBELF;
-			}
+		progs = libbpf_reallocarray(progs, nr_progs + 1, sizeof(*progs));
+		if (!progs) {
+			/*
+			 * In this case the original obj->programs
+			 * is still valid, so don't need special treat for
+			 * bpf_close_object().
+			 */
+			pr_warn("sec '%s': failed to alloc memory for new program '%s'\n",
+				sec_name, name);
+			return -ENOMEM;
 		}
+		obj->programs = progs;
 
-		if (!name && prog->idx == obj->efile.text_shndx)
-			name = ".text";
+		prog = &progs[nr_progs];
 
-		if (!name) {
-			pr_warn("failed to find sym for prog %s\n",
-				prog->section_name);
-			return -EINVAL;
-		}
+		err = bpf_object__init_prog(obj, prog, name, sec_idx, sec_name,
+					    sec_off, data + sec_off, prog_sz);
+		if (err)
+			return err;
 
-		prog->name = strdup(name);
-		if (!prog->name) {
-			pr_warn("failed to allocate memory for prog sym %s\n",
-				name);
-			return -ENOMEM;
-		}
+		nr_progs++;
+		obj->nr_programs = nr_progs;
+
+		sec_off += prog_sz;
 	}
 
 	return 0;
@@ -1035,6 +1079,7 @@ static struct bpf_object *bpf_object__new(const char *path,
 	obj->efile.bss_shndx = -1;
 	obj->efile.st_ops_shndx = -1;
 	obj->kconfig_map_idx = -1;
+	obj->rodata_map_idx = -1;
 
 	obj->kern_version = get_kernel_version();
 	obj->loaded = false;
@@ -1066,13 +1111,18 @@ static void bpf_object__elf_finish(struct bpf_object *obj)
 	obj->efile.obj_buf_sz = 0;
 }
 
+/* if libelf is old and doesn't support mmap(), fall back to read() */
+#ifndef ELF_C_READ_MMAP
+#define ELF_C_READ_MMAP ELF_C_READ
+#endif
+
 static int bpf_object__elf_init(struct bpf_object *obj)
 {
 	int err = 0;
 	GElf_Ehdr *ep;
 
 	if (obj_elf_valid(obj)) {
-		pr_warn("elf init: internal error\n");
+		pr_warn("elf: init internal error\n");
 		return -LIBBPF_ERRNO__LIBELF;
 	}
 
@@ -1090,31 +1140,44 @@ static int bpf_object__elf_init(struct bpf_object *obj)
 
 			err = -errno;
 			cp = libbpf_strerror_r(err, errmsg, sizeof(errmsg));
-			pr_warn("failed to open %s: %s\n", obj->path, cp);
+			pr_warn("elf: failed to open %s: %s\n", obj->path, cp);
 			return err;
 		}
 
-		obj->efile.elf = elf_begin(obj->efile.fd,
-					   LIBBPF_ELF_C_READ_MMAP, NULL);
+		obj->efile.elf = elf_begin(obj->efile.fd, ELF_C_READ_MMAP, NULL);
 	}
 
 	if (!obj->efile.elf) {
-		pr_warn("failed to open %s as ELF file\n", obj->path);
+		pr_warn("elf: failed to open %s as ELF file: %s\n", obj->path, elf_errmsg(-1));
 		err = -LIBBPF_ERRNO__LIBELF;
 		goto errout;
 	}
 
 	if (!gelf_getehdr(obj->efile.elf, &obj->efile.ehdr)) {
-		pr_warn("failed to get EHDR from %s\n", obj->path);
+		pr_warn("elf: failed to get ELF header from %s: %s\n", obj->path, elf_errmsg(-1));
 		err = -LIBBPF_ERRNO__FORMAT;
 		goto errout;
 	}
 	ep = &obj->efile.ehdr;
 
+	if (elf_getshdrstrndx(obj->efile.elf, &obj->efile.shstrndx)) {
+		pr_warn("elf: failed to get section names section index for %s: %s\n",
+			obj->path, elf_errmsg(-1));
+		err = -LIBBPF_ERRNO__FORMAT;
+		goto errout;
+	}
+
+	/* Elf is corrupted/truncated, avoid calling elf_strptr. */
+	if (!elf_rawdata(elf_getscn(obj->efile.elf, obj->efile.shstrndx), NULL)) {
+		pr_warn("elf: failed to get section names strings from %s: %s\n",
+			obj->path, elf_errmsg(-1));
+		return -LIBBPF_ERRNO__FORMAT;
+	}
+
 	/* Old LLVM set e_machine to EM_NONE */
 	if (ep->e_type != ET_REL ||
 	    (ep->e_machine && ep->e_machine != EM_BPF)) {
-		pr_warn("%s is not an eBPF object file\n", obj->path);
+		pr_warn("elf: %s is not a valid eBPF object file\n", obj->path);
 		err = -LIBBPF_ERRNO__FORMAT;
 		goto errout;
 	}
@@ -1136,7 +1199,7 @@ static int bpf_object__check_endianness(struct bpf_object *obj)
 #else
 # error "Unrecognized __BYTE_ORDER__"
 #endif
-	pr_warn("endianness mismatch.\n");
+	pr_warn("elf: endianness mismatch in %s.\n", obj->path);
 	return -LIBBPF_ERRNO__ENDIAN;
 }
 
@@ -1171,55 +1234,10 @@ static bool bpf_map_type__is_map_in_map(enum bpf_map_type type)
 	return false;
 }
 
-static int bpf_object_search_section_size(const struct bpf_object *obj,
-					  const char *name, size_t *d_size)
-{
-	const GElf_Ehdr *ep = &obj->efile.ehdr;
-	Elf *elf = obj->efile.elf;
-	Elf_Scn *scn = NULL;
-	int idx = 0;
-
-	while ((scn = elf_nextscn(elf, scn)) != NULL) {
-		const char *sec_name;
-		Elf_Data *data;
-		GElf_Shdr sh;
-
-		idx++;
-		if (gelf_getshdr(scn, &sh) != &sh) {
-			pr_warn("failed to get section(%d) header from %s\n",
-				idx, obj->path);
-			return -EIO;
-		}
-
-		sec_name = elf_strptr(elf, ep->e_shstrndx, sh.sh_name);
-		if (!sec_name) {
-			pr_warn("failed to get section(%d) name from %s\n",
-				idx, obj->path);
-			return -EIO;
-		}
-
-		if (strcmp(name, sec_name))
-			continue;
-
-		data = elf_getdata(scn, 0);
-		if (!data) {
-			pr_warn("failed to get section(%d) data from %s(%s)\n",
-				idx, name, obj->path);
-			return -EIO;
-		}
-
-		*d_size = data->d_size;
-		return 0;
-	}
-
-	return -ENOENT;
-}
-
 int bpf_object__section_size(const struct bpf_object *obj, const char *name,
 			     __u32 *size)
 {
 	int ret = -ENOENT;
-	size_t d_size;
 
 	*size = 0;
 	if (!name) {
@@ -1237,9 +1255,13 @@ int bpf_object__section_size(const struct bpf_object *obj, const char *name,
 		if (obj->efile.st_ops_data)
 			*size = obj->efile.st_ops_data->d_size;
 	} else {
-		ret = bpf_object_search_section_size(obj, name, &d_size);
-		if (!ret)
-			*size = d_size;
+		Elf_Scn *scn = elf_sec_by_name(obj, name);
+		Elf_Data *data = elf_sec_data(obj, scn);
+
+		if (data) {
+			ret = 0; /* found it */
+			*size = data->d_size;
+		}
 	}
 
 	return *size ? 0 : ret;
@@ -1264,8 +1286,7 @@ int bpf_object__variable_offset(const struct bpf_object *obj, const char *name,
 		    GELF_ST_TYPE(sym.st_info) != STT_OBJECT)
 			continue;
 
-		sname = elf_strptr(obj->efile.elf, obj->efile.strtabidx,
-				   sym.st_name);
+		sname = elf_sym_str(obj, sym.st_name);
 		if (!sname) {
 			pr_warn("failed to get sym name string for var %s\n",
 				name);
@@ -1290,7 +1311,7 @@ static struct bpf_map *bpf_object__add_map(struct bpf_object *obj)
 		return &obj->maps[obj->nr_maps++];
 
 	new_cap = max((size_t)4, obj->maps_cap * 3 / 2);
-	new_maps = realloc(obj->maps, new_cap * sizeof(*obj->maps));
+	new_maps = libbpf_reallocarray(obj->maps, new_cap, sizeof(*obj->maps));
 	if (!new_maps) {
 		pr_warn("alloc maps for object failed\n");
 		return ERR_PTR(-ENOMEM);
@@ -1417,6 +1438,8 @@ static int bpf_object__init_global_data_maps(struct bpf_object *obj)
 						    obj->efile.rodata->d_size);
 		if (err)
 			return err;
+
+		obj->rodata_map_idx = obj->nr_maps - 1;
 	}
 	if (obj->efile.bss_shndx >= 0) {
 		err = bpf_object__init_internal_map(obj, LIBBPF_MAP_BSS,
@@ -1742,12 +1765,12 @@ static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict)
 	if (!symbols)
 		return -EINVAL;
 
-	scn = elf_getscn(obj->efile.elf, obj->efile.maps_shndx);
-	if (scn)
-		data = elf_getdata(scn, NULL);
+
+	scn = elf_sec_by_idx(obj, obj->efile.maps_shndx);
+	data = elf_sec_data(obj, scn);
 	if (!scn || !data) {
-		pr_warn("failed to get Elf_Data from map section %d\n",
-			obj->efile.maps_shndx);
+		pr_warn("elf: failed to get legacy map definitions for %s\n",
+			obj->path);
 		return -EINVAL;
 	}
 
@@ -1769,12 +1792,12 @@ static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict)
 		nr_maps++;
 	}
 	/* Assume equally sized map definitions */
-	pr_debug("maps in %s: %d maps in %zd bytes\n",
-		 obj->path, nr_maps, data->d_size);
+	pr_debug("elf: found %d legacy map definitions (%zd bytes) in %s\n",
+		 nr_maps, data->d_size, obj->path);
 
 	if (!data->d_size || nr_maps == 0 || (data->d_size % nr_maps) != 0) {
-		pr_warn("unable to determine map definition size section %s, %d maps in %zd bytes\n",
-			obj->path, nr_maps, data->d_size);
+		pr_warn("elf: unable to determine legacy map definition size in %s\n",
+			obj->path);
 		return -EINVAL;
 	}
 	map_def_sz = data->d_size / nr_maps;
@@ -1795,8 +1818,7 @@ static int bpf_object__init_user_maps(struct bpf_object *obj, bool strict)
 		if (IS_ERR(map))
 			return PTR_ERR(map);
 
-		map_name = elf_strptr(obj->efile.elf, obj->efile.strtabidx,
-				      sym.st_name);
+		map_name = elf_sym_str(obj, sym.st_name);
 		if (!map_name) {
 			pr_warn("failed to get map #%d name sym string for obj %s\n",
 				i, obj->path);
@@ -1884,6 +1906,29 @@ resolve_func_ptr(const struct btf *btf, __u32 id, __u32 *res_id)
 	return btf_is_func_proto(t) ? t : NULL;
 }
 
+static const char *btf_kind_str(const struct btf_type *t)
+{
+	switch (btf_kind(t)) {
+	case BTF_KIND_UNKN: return "void";
+	case BTF_KIND_INT: return "int";
+	case BTF_KIND_PTR: return "ptr";
+	case BTF_KIND_ARRAY: return "array";
+	case BTF_KIND_STRUCT: return "struct";
+	case BTF_KIND_UNION: return "union";
+	case BTF_KIND_ENUM: return "enum";
+	case BTF_KIND_FWD: return "fwd";
+	case BTF_KIND_TYPEDEF: return "typedef";
+	case BTF_KIND_VOLATILE: return "volatile";
+	case BTF_KIND_CONST: return "const";
+	case BTF_KIND_RESTRICT: return "restrict";
+	case BTF_KIND_FUNC: return "func";
+	case BTF_KIND_FUNC_PROTO: return "func_proto";
+	case BTF_KIND_VAR: return "var";
+	case BTF_KIND_DATASEC: return "datasec";
+	default: return "unknown";
+	}
+}
+
 /*
  * Fetch integer attribute of BTF map definition. Such attributes are
  * represented using a pointer to an array, in which dimensionality of array
@@ -1900,8 +1945,8 @@ static bool get_map_field_int(const char *map_name, const struct btf *btf,
 	const struct btf_type *arr_t;
 
 	if (!btf_is_ptr(t)) {
-		pr_warn("map '%s': attr '%s': expected PTR, got %u.\n",
-			map_name, name, btf_kind(t));
+		pr_warn("map '%s': attr '%s': expected PTR, got %s.\n",
+			map_name, name, btf_kind_str(t));
 		return false;
 	}
 
@@ -1912,8 +1957,8 @@ static bool get_map_field_int(const char *map_name, const struct btf *btf,
 		return false;
 	}
 	if (!btf_is_array(arr_t)) {
-		pr_warn("map '%s': attr '%s': expected ARRAY, got %u.\n",
-			map_name, name, btf_kind(arr_t));
+		pr_warn("map '%s': attr '%s': expected ARRAY, got %s.\n",
+			map_name, name, btf_kind_str(arr_t));
 		return false;
 	}
 	arr_info = btf_array(arr_t);
@@ -1924,7 +1969,7 @@ static bool get_map_field_int(const char *map_name, const struct btf *btf,
 static int build_map_pin_path(struct bpf_map *map, const char *path)
 {
 	char buf[PATH_MAX];
-	int err, len;
+	int len;
 
 	if (!path)
 		path = "/sys/fs/bpf";
@@ -1935,11 +1980,7 @@ static int build_map_pin_path(struct bpf_map *map, const char *path)
 	else if (len >= PATH_MAX)
 		return -ENAMETOOLONG;
 
-	err = bpf_map__set_pin_path(map, buf);
-	if (err)
-		return err;
-
-	return 0;
+	return bpf_map__set_pin_path(map, buf);
 }
 
 
@@ -2007,8 +2048,8 @@ static int parse_btf_map_def(struct bpf_object *obj,
 				return -EINVAL;
 			}
 			if (!btf_is_ptr(t)) {
-				pr_warn("map '%s': key spec is not PTR: %u.\n",
-					map->name, btf_kind(t));
+				pr_warn("map '%s': key spec is not PTR: %s.\n",
+					map->name, btf_kind_str(t));
 				return -EINVAL;
 			}
 			sz = btf__resolve_size(obj->btf, t->type);
@@ -2049,8 +2090,8 @@ static int parse_btf_map_def(struct bpf_object *obj,
 				return -EINVAL;
 			}
 			if (!btf_is_ptr(t)) {
-				pr_warn("map '%s': value spec is not PTR: %u.\n",
-					map->name, btf_kind(t));
+				pr_warn("map '%s': value spec is not PTR: %s.\n",
+					map->name, btf_kind_str(t));
 				return -EINVAL;
 			}
 			sz = btf__resolve_size(obj->btf, t->type);
@@ -2107,14 +2148,14 @@ static int parse_btf_map_def(struct bpf_object *obj,
 			t = skip_mods_and_typedefs(obj->btf, btf_array(t)->type,
 						   NULL);
 			if (!btf_is_ptr(t)) {
-				pr_warn("map '%s': map-in-map inner def is of unexpected kind %u.\n",
-					map->name, btf_kind(t));
+				pr_warn("map '%s': map-in-map inner def is of unexpected kind %s.\n",
+					map->name, btf_kind_str(t));
 				return -EINVAL;
 			}
 			t = skip_mods_and_typedefs(obj->btf, t->type, NULL);
 			if (!btf_is_struct(t)) {
-				pr_warn("map '%s': map-in-map inner def is of unexpected kind %u.\n",
-					map->name, btf_kind(t));
+				pr_warn("map '%s': map-in-map inner def is of unexpected kind %s.\n",
+					map->name, btf_kind_str(t));
 				return -EINVAL;
 			}
 
@@ -2205,8 +2246,8 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
 		return -EINVAL;
 	}
 	if (!btf_is_var(var)) {
-		pr_warn("map '%s': unexpected var kind %u.\n",
-			map_name, btf_kind(var));
+		pr_warn("map '%s': unexpected var kind %s.\n",
+			map_name, btf_kind_str(var));
 		return -EINVAL;
 	}
 	if (var_extra->linkage != BTF_VAR_GLOBAL_ALLOCATED &&
@@ -2218,8 +2259,8 @@ static int bpf_object__init_user_btf_map(struct bpf_object *obj,
 
 	def = skip_mods_and_typedefs(obj->btf, var->type, NULL);
 	if (!btf_is_struct(def)) {
-		pr_warn("map '%s': unexpected def kind %u.\n",
-			map_name, btf_kind(var));
+		pr_warn("map '%s': unexpected def kind %s.\n",
+			map_name, btf_kind_str(var));
 		return -EINVAL;
 	}
 	if (def->size > vi->size) {
@@ -2259,12 +2300,11 @@ static int bpf_object__init_user_btf_maps(struct bpf_object *obj, bool strict,
 	if (obj->efile.btf_maps_shndx < 0)
 		return 0;
 
-	scn = elf_getscn(obj->efile.elf, obj->efile.btf_maps_shndx);
-	if (scn)
-		data = elf_getdata(scn, NULL);
+	scn = elf_sec_by_idx(obj, obj->efile.btf_maps_shndx);
+	data = elf_sec_data(obj, scn);
 	if (!scn || !data) {
-		pr_warn("failed to get Elf_Data from map section %d (%s)\n",
-			obj->efile.btf_maps_shndx, MAPS_ELF_SEC);
+		pr_warn("elf: failed to get %s map definitions for %s\n",
+			MAPS_ELF_SEC, obj->path);
 		return -EINVAL;
 	}
 
@@ -2322,36 +2362,28 @@ static int bpf_object__init_maps(struct bpf_object *obj,
 
 static bool section_have_execinstr(struct bpf_object *obj, int idx)
 {
-	Elf_Scn *scn;
 	GElf_Shdr sh;
 
-	scn = elf_getscn(obj->efile.elf, idx);
-	if (!scn)
-		return false;
-
-	if (gelf_getshdr(scn, &sh) != &sh)
+	if (elf_sec_hdr(obj, elf_sec_by_idx(obj, idx), &sh))
 		return false;
 
-	if (sh.sh_flags & SHF_EXECINSTR)
-		return true;
-
-	return false;
+	return sh.sh_flags & SHF_EXECINSTR;
 }
 
 static bool btf_needs_sanitization(struct bpf_object *obj)
 {
-	bool has_func_global = obj->caps.btf_func_global;
-	bool has_datasec = obj->caps.btf_datasec;
-	bool has_func = obj->caps.btf_func;
+	bool has_func_global = kernel_supports(FEAT_BTF_GLOBAL_FUNC);
+	bool has_datasec = kernel_supports(FEAT_BTF_DATASEC);
+	bool has_func = kernel_supports(FEAT_BTF_FUNC);
 
 	return !has_func || !has_datasec || !has_func_global;
 }
 
 static void bpf_object__sanitize_btf(struct bpf_object *obj, struct btf *btf)
 {
-	bool has_func_global = obj->caps.btf_func_global;
-	bool has_datasec = obj->caps.btf_datasec;
-	bool has_func = obj->caps.btf_func;
+	bool has_func_global = kernel_supports(FEAT_BTF_GLOBAL_FUNC);
+	bool has_datasec = kernel_supports(FEAT_BTF_DATASEC);
+	bool has_func = kernel_supports(FEAT_BTF_FUNC);
 	struct btf_type *t;
 	int i, j, vlen;
 
@@ -2496,12 +2528,23 @@ static int bpf_object__load_vmlinux_btf(struct bpf_object *obj)
 {
 	bool need_vmlinux_btf = false;
 	struct bpf_program *prog;
-	int err;
+	int i, err;
 
 	/* CO-RE relocations need kernel BTF */
-	if (obj->btf_ext && obj->btf_ext->field_reloc_info.len)
+	if (obj->btf_ext && obj->btf_ext->core_relo_info.len)
 		need_vmlinux_btf = true;
 
+	/* Support for typed ksyms needs kernel BTF */
+	for (i = 0; i < obj->nr_extern; i++) {
+		const struct extern_desc *ext;
+
+		ext = &obj->externs[i];
+		if (ext->type == EXT_KSYM && ext->ksym.type_id) {
+			need_vmlinux_btf = true;
+			break;
+		}
+	}
+
 	bpf_object__for_each_program(prog, obj) {
 		if (!prog->load)
 			continue;
@@ -2533,6 +2576,15 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj)
 	if (!obj->btf)
 		return 0;
 
+	if (!kernel_supports(FEAT_BTF)) {
+		if (kernel_needs_btf(obj)) {
+			err = -EOPNOTSUPP;
+			goto report;
+		}
+		pr_debug("Kernel doesn't support BTF, skipping uploading it.\n");
+		return 0;
+	}
+
 	sanitize = btf_needs_sanitization(obj);
 	if (sanitize) {
 		const void *raw_data;
@@ -2558,6 +2610,7 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj)
 		}
 		btf__free(kern_btf);
 	}
+report:
 	if (err) {
 		btf_mandatory = kernel_needs_btf(obj);
 		pr_warn("Error loading .BTF into kernel: %d. %s\n", err,
@@ -2569,61 +2622,255 @@ static int bpf_object__sanitize_and_load_btf(struct bpf_object *obj)
 	return err;
 }
 
+static const char *elf_sym_str(const struct bpf_object *obj, size_t off)
+{
+	const char *name;
+
+	name = elf_strptr(obj->efile.elf, obj->efile.strtabidx, off);
+	if (!name) {
+		pr_warn("elf: failed to get section name string at offset %zu from %s: %s\n",
+			off, obj->path, elf_errmsg(-1));
+		return NULL;
+	}
+
+	return name;
+}
+
+static const char *elf_sec_str(const struct bpf_object *obj, size_t off)
+{
+	const char *name;
+
+	name = elf_strptr(obj->efile.elf, obj->efile.shstrndx, off);
+	if (!name) {
+		pr_warn("elf: failed to get section name string at offset %zu from %s: %s\n",
+			off, obj->path, elf_errmsg(-1));
+		return NULL;
+	}
+
+	return name;
+}
+
+static Elf_Scn *elf_sec_by_idx(const struct bpf_object *obj, size_t idx)
+{
+	Elf_Scn *scn;
+
+	scn = elf_getscn(obj->efile.elf, idx);
+	if (!scn) {
+		pr_warn("elf: failed to get section(%zu) from %s: %s\n",
+			idx, obj->path, elf_errmsg(-1));
+		return NULL;
+	}
+	return scn;
+}
+
+static Elf_Scn *elf_sec_by_name(const struct bpf_object *obj, const char *name)
+{
+	Elf_Scn *scn = NULL;
+	Elf *elf = obj->efile.elf;
+	const char *sec_name;
+
+	while ((scn = elf_nextscn(elf, scn)) != NULL) {
+		sec_name = elf_sec_name(obj, scn);
+		if (!sec_name)
+			return NULL;
+
+		if (strcmp(sec_name, name) != 0)
+			continue;
+
+		return scn;
+	}
+	return NULL;
+}
+
+static int elf_sec_hdr(const struct bpf_object *obj, Elf_Scn *scn, GElf_Shdr *hdr)
+{
+	if (!scn)
+		return -EINVAL;
+
+	if (gelf_getshdr(scn, hdr) != hdr) {
+		pr_warn("elf: failed to get section(%zu) header from %s: %s\n",
+			elf_ndxscn(scn), obj->path, elf_errmsg(-1));
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static const char *elf_sec_name(const struct bpf_object *obj, Elf_Scn *scn)
+{
+	const char *name;
+	GElf_Shdr sh;
+
+	if (!scn)
+		return NULL;
+
+	if (elf_sec_hdr(obj, scn, &sh))
+		return NULL;
+
+	name = elf_sec_str(obj, sh.sh_name);
+	if (!name) {
+		pr_warn("elf: failed to get section(%zu) name from %s: %s\n",
+			elf_ndxscn(scn), obj->path, elf_errmsg(-1));
+		return NULL;
+	}
+
+	return name;
+}
+
+static Elf_Data *elf_sec_data(const struct bpf_object *obj, Elf_Scn *scn)
+{
+	Elf_Data *data;
+
+	if (!scn)
+		return NULL;
+
+	data = elf_getdata(scn, 0);
+	if (!data) {
+		pr_warn("elf: failed to get section(%zu) %s data from %s: %s\n",
+			elf_ndxscn(scn), elf_sec_name(obj, scn) ?: "<?>",
+			obj->path, elf_errmsg(-1));
+		return NULL;
+	}
+
+	return data;
+}
+
+static int elf_sym_by_sec_off(const struct bpf_object *obj, size_t sec_idx,
+			      size_t off, __u32 sym_type, GElf_Sym *sym)
+{
+	Elf_Data *symbols = obj->efile.symbols;
+	size_t n = symbols->d_size / sizeof(GElf_Sym);
+	int i;
+
+	for (i = 0; i < n; i++) {
+		if (!gelf_getsym(symbols, i, sym))
+			continue;
+		if (sym->st_shndx != sec_idx || sym->st_value != off)
+			continue;
+		if (GELF_ST_TYPE(sym->st_info) != sym_type)
+			continue;
+		return 0;
+	}
+
+	return -ENOENT;
+}
+
+static bool is_sec_name_dwarf(const char *name)
+{
+	/* approximation, but the actual list is too long */
+	return strncmp(name, ".debug_", sizeof(".debug_") - 1) == 0;
+}
+
+static bool ignore_elf_section(GElf_Shdr *hdr, const char *name)
+{
+	/* no special handling of .strtab */
+	if (hdr->sh_type == SHT_STRTAB)
+		return true;
+
+	/* ignore .llvm_addrsig section as well */
+	if (hdr->sh_type == 0x6FFF4C03 /* SHT_LLVM_ADDRSIG */)
+		return true;
+
+	/* no subprograms will lead to an empty .text section, ignore it */
+	if (hdr->sh_type == SHT_PROGBITS && hdr->sh_size == 0 &&
+	    strcmp(name, ".text") == 0)
+		return true;
+
+	/* DWARF sections */
+	if (is_sec_name_dwarf(name))
+		return true;
+
+	if (strncmp(name, ".rel", sizeof(".rel") - 1) == 0) {
+		name += sizeof(".rel") - 1;
+		/* DWARF section relocations */
+		if (is_sec_name_dwarf(name))
+			return true;
+
+		/* .BTF and .BTF.ext don't need relocations */
+		if (strcmp(name, BTF_ELF_SEC) == 0 ||
+		    strcmp(name, BTF_EXT_ELF_SEC) == 0)
+			return true;
+	}
+
+	return false;
+}
+
+static int cmp_progs(const void *_a, const void *_b)
+{
+	const struct bpf_program *a = _a;
+	const struct bpf_program *b = _b;
+
+	if (a->sec_idx != b->sec_idx)
+		return a->sec_idx < b->sec_idx ? -1 : 1;
+
+	/* sec_insn_off can't be the same within the section */
+	return a->sec_insn_off < b->sec_insn_off ? -1 : 1;
+}
+
 static int bpf_object__elf_collect(struct bpf_object *obj)
 {
 	Elf *elf = obj->efile.elf;
-	GElf_Ehdr *ep = &obj->efile.ehdr;
 	Elf_Data *btf_ext_data = NULL;
 	Elf_Data *btf_data = NULL;
-	Elf_Scn *scn = NULL;
 	int idx = 0, err = 0;
+	const char *name;
+	Elf_Data *data;
+	Elf_Scn *scn;
+	GElf_Shdr sh;
 
-	/* Elf is corrupted/truncated, avoid calling elf_strptr. */
-	if (!elf_rawdata(elf_getscn(elf, ep->e_shstrndx), NULL)) {
-		pr_warn("failed to get e_shstrndx from %s\n", obj->path);
-		return -LIBBPF_ERRNO__FORMAT;
+	/* a bunch of ELF parsing functionality depends on processing symbols,
+	 * so do the first pass and find the symbol table
+	 */
+	scn = NULL;
+	while ((scn = elf_nextscn(elf, scn)) != NULL) {
+		if (elf_sec_hdr(obj, scn, &sh))
+			return -LIBBPF_ERRNO__FORMAT;
+
+		if (sh.sh_type == SHT_SYMTAB) {
+			if (obj->efile.symbols) {
+				pr_warn("elf: multiple symbol tables in %s\n", obj->path);
+				return -LIBBPF_ERRNO__FORMAT;
+			}
+
+			data = elf_sec_data(obj, scn);
+			if (!data)
+				return -LIBBPF_ERRNO__FORMAT;
+
+			obj->efile.symbols = data;
+			obj->efile.symbols_shndx = elf_ndxscn(scn);
+			obj->efile.strtabidx = sh.sh_link;
+		}
 	}
 
+	scn = NULL;
 	while ((scn = elf_nextscn(elf, scn)) != NULL) {
-		char *name;
-		GElf_Shdr sh;
-		Elf_Data *data;
-
 		idx++;
-		if (gelf_getshdr(scn, &sh) != &sh) {
-			pr_warn("failed to get section(%d) header from %s\n",
-				idx, obj->path);
+
+		if (elf_sec_hdr(obj, scn, &sh))
 			return -LIBBPF_ERRNO__FORMAT;
-		}
 
-		name = elf_strptr(elf, ep->e_shstrndx, sh.sh_name);
-		if (!name) {
-			pr_warn("failed to get section(%d) name from %s\n",
-				idx, obj->path);
+		name = elf_sec_str(obj, sh.sh_name);
+		if (!name)
 			return -LIBBPF_ERRNO__FORMAT;
-		}
 
-		data = elf_getdata(scn, 0);
-		if (!data) {
-			pr_warn("failed to get section(%d) data from %s(%s)\n",
-				idx, name, obj->path);
+		if (ignore_elf_section(&sh, name))
+			continue;
+
+		data = elf_sec_data(obj, scn);
+		if (!data)
 			return -LIBBPF_ERRNO__FORMAT;
-		}
-		pr_debug("section(%d) %s, size %ld, link %d, flags %lx, type=%d\n",
+
+		pr_debug("elf: section(%d) %s, size %ld, link %d, flags %lx, type=%d\n",
 			 idx, name, (unsigned long)data->d_size,
 			 (int)sh.sh_link, (unsigned long)sh.sh_flags,
 			 (int)sh.sh_type);
 
 		if (strcmp(name, "license") == 0) {
-			err = bpf_object__init_license(obj,
-						       data->d_buf,
-						       data->d_size);
+			err = bpf_object__init_license(obj, data->d_buf, data->d_size);
 			if (err)
 				return err;
 		} else if (strcmp(name, "version") == 0) {
-			err = bpf_object__init_kversion(obj,
-							data->d_buf,
-							data->d_size);
+			err = bpf_object__init_kversion(obj, data->d_buf, data->d_size);
 			if (err)
 				return err;
 		} else if (strcmp(name, "maps") == 0) {
@@ -2635,31 +2882,14 @@ static int bpf_object__elf_collect(struct bpf_object *obj)
 		} else if (strcmp(name, BTF_EXT_ELF_SEC) == 0) {
 			btf_ext_data = data;
 		} else if (sh.sh_type == SHT_SYMTAB) {
-			if (obj->efile.symbols) {
-				pr_warn("bpf: multiple SYMTAB in %s\n",
-					obj->path);
-				return -LIBBPF_ERRNO__FORMAT;
-			}
-			obj->efile.symbols = data;
-			obj->efile.symbols_shndx = idx;
-			obj->efile.strtabidx = sh.sh_link;
+			/* already processed during the first pass above */
 		} else if (sh.sh_type == SHT_PROGBITS && data->d_size > 0) {
 			if (sh.sh_flags & SHF_EXECINSTR) {
 				if (strcmp(name, ".text") == 0)
 					obj->efile.text_shndx = idx;
-				err = bpf_object__add_program(obj, data->d_buf,
-							      data->d_size,
-							      name, idx);
-				if (err) {
-					char errmsg[STRERR_BUFSIZE];
-					char *cp;
-
-					cp = libbpf_strerror_r(-err, errmsg,
-							       sizeof(errmsg));
-					pr_warn("failed to alloc program %s (%s): %s",
-						name, obj->path, cp);
+				err = bpf_object__add_programs(obj, data, name, idx);
+				if (err)
 					return err;
-				}
 			} else if (strcmp(name, DATA_SEC) == 0) {
 				obj->efile.data = data;
 				obj->efile.data_shndx = idx;
@@ -2670,7 +2900,8 @@ static int bpf_object__elf_collect(struct bpf_object *obj)
 				obj->efile.st_ops_data = data;
 				obj->efile.st_ops_shndx = idx;
 			} else {
-				pr_debug("skip section(%d) %s\n", idx, name);
+				pr_info("elf: skipping unrecognized data section(%d) %s\n",
+					idx, name);
 			}
 		} else if (sh.sh_type == SHT_REL) {
 			int nr_sects = obj->efile.nr_reloc_sects;
@@ -2681,36 +2912,40 @@ static int bpf_object__elf_collect(struct bpf_object *obj)
 			if (!section_have_execinstr(obj, sec) &&
 			    strcmp(name, ".rel" STRUCT_OPS_SEC) &&
 			    strcmp(name, ".rel" MAPS_ELF_SEC)) {
-				pr_debug("skip relo %s(%d) for section(%d)\n",
-					 name, idx, sec);
+				pr_info("elf: skipping relo section(%d) %s for section(%d) %s\n",
+					idx, name, sec,
+					elf_sec_name(obj, elf_sec_by_idx(obj, sec)) ?: "<?>");
 				continue;
 			}
 
-			sects = reallocarray(sects, nr_sects + 1,
-					     sizeof(*obj->efile.reloc_sects));
-			if (!sects) {
-				pr_warn("reloc_sects realloc failed\n");
+			sects = libbpf_reallocarray(sects, nr_sects + 1,
+						    sizeof(*obj->efile.reloc_sects));
+			if (!sects)
 				return -ENOMEM;
-			}
 
 			obj->efile.reloc_sects = sects;
 			obj->efile.nr_reloc_sects++;
 
 			obj->efile.reloc_sects[nr_sects].shdr = sh;
 			obj->efile.reloc_sects[nr_sects].data = data;
-		} else if (sh.sh_type == SHT_NOBITS &&
-			   strcmp(name, BSS_SEC) == 0) {
+		} else if (sh.sh_type == SHT_NOBITS && strcmp(name, BSS_SEC) == 0) {
 			obj->efile.bss = data;
 			obj->efile.bss_shndx = idx;
 		} else {
-			pr_debug("skip section(%d) %s\n", idx, name);
+			pr_info("elf: skipping section(%d) %s (size %zu)\n", idx, name,
+				(size_t)sh.sh_size);
 		}
 	}
 
 	if (!obj->efile.strtabidx || obj->efile.strtabidx > idx) {
-		pr_warn("Corrupted ELF file: index of strtab invalid\n");
+		pr_warn("elf: symbol strings section missing or invalid in %s\n", obj->path);
 		return -LIBBPF_ERRNO__FORMAT;
 	}
+
+	/* sort BPF programs by section name and in-section instruction offset
+	 * for faster search */
+	qsort(obj->programs, obj->nr_programs, sizeof(*obj->programs), cmp_progs);
+
 	return bpf_object__init_btf(obj, btf_data, btf_ext_data);
 }
 
@@ -2869,14 +3104,13 @@ static int bpf_object__collect_externs(struct bpf_object *obj)
 	if (!obj->efile.symbols)
 		return 0;
 
-	scn = elf_getscn(obj->efile.elf, obj->efile.symbols_shndx);
-	if (!scn)
+	scn = elf_sec_by_idx(obj, obj->efile.symbols_shndx);
+	if (elf_sec_hdr(obj, scn, &sh))
 		return -LIBBPF_ERRNO__FORMAT;
-	if (gelf_getshdr(scn, &sh) != &sh)
-		return -LIBBPF_ERRNO__FORMAT;
-	n = sh.sh_size / sh.sh_entsize;
 
+	n = sh.sh_size / sh.sh_entsize;
 	pr_debug("looking for externs among %d symbols...\n", n);
+
 	for (i = 0; i < n; i++) {
 		GElf_Sym sym;
 
@@ -2884,13 +3118,12 @@ static int bpf_object__collect_externs(struct bpf_object *obj)
 			return -LIBBPF_ERRNO__FORMAT;
 		if (!sym_is_extern(&sym))
 			continue;
-		ext_name = elf_strptr(obj->efile.elf, obj->efile.strtabidx,
-				      sym.st_name);
+		ext_name = elf_sym_str(obj, sym.st_name);
 		if (!ext_name || !ext_name[0])
 			continue;
 
 		ext = obj->externs;
-		ext = reallocarray(ext, obj->nr_extern + 1, sizeof(*ext));
+		ext = libbpf_reallocarray(ext, obj->nr_extern + 1, sizeof(*ext));
 		if (!ext)
 			return -ENOMEM;
 		obj->externs = ext;
@@ -2940,16 +3173,10 @@ static int bpf_object__collect_externs(struct bpf_object *obj)
 				return -ENOTSUP;
 			}
 		} else if (strcmp(sec_name, KSYMS_SEC) == 0) {
-			const struct btf_type *vt;
-
 			ksym_sec = sec;
 			ext->type = EXT_KSYM;
-
-			vt = skip_mods_and_typedefs(obj->btf, t->type, NULL);
-			if (!btf_is_void(vt)) {
-				pr_warn("extern (ksym) '%s' is not typeless (void)\n", ext_name);
-				return -ENOTSUP;
-			}
+			skip_mods_and_typedefs(obj->btf, t->type,
+					       &ext->ksym.type_id);
 		} else {
 			pr_warn("unrecognized extern section '%s'\n", sec_name);
 			return -ENOTSUP;
@@ -3037,20 +3264,6 @@ static int bpf_object__collect_externs(struct bpf_object *obj)
 	return 0;
 }
 
-static struct bpf_program *
-bpf_object__find_prog_by_idx(struct bpf_object *obj, int idx)
-{
-	struct bpf_program *prog;
-	size_t i;
-
-	for (i = 0; i < obj->nr_programs; i++) {
-		prog = &obj->programs[i];
-		if (prog->idx == idx)
-			return prog;
-	}
-	return NULL;
-}
-
 struct bpf_program *
 bpf_object__find_program_by_title(const struct bpf_object *obj,
 				  const char *title)
@@ -3058,12 +3271,18 @@ bpf_object__find_program_by_title(const struct bpf_object *obj,
 	struct bpf_program *pos;
 
 	bpf_object__for_each_program(pos, obj) {
-		if (pos->section_name && !strcmp(pos->section_name, title))
+		if (pos->sec_name && !strcmp(pos->sec_name, title))
 			return pos;
 	}
 	return NULL;
 }
 
+static bool prog_is_subprog(const struct bpf_object *obj,
+			    const struct bpf_program *prog)
+{
+	return prog->sec_idx == obj->efile.text_shndx && obj->has_subcalls;
+}
+
 struct bpf_program *
 bpf_object__find_program_by_name(const struct bpf_object *obj,
 				 const char *name)
@@ -3071,6 +3290,8 @@ bpf_object__find_program_by_name(const struct bpf_object *obj,
 	struct bpf_program *prog;
 
 	bpf_object__for_each_program(prog, obj) {
+		if (prog_is_subprog(obj, prog))
+			continue;
 		if (!strcmp(prog->name, name))
 			return prog;
 	}
@@ -3109,7 +3330,7 @@ bpf_object__section_to_libbpf_map_type(const struct bpf_object *obj, int shndx)
 
 static int bpf_program__record_reloc(struct bpf_program *prog,
 				     struct reloc_desc *reloc_desc,
-				     __u32 insn_idx, const char *name,
+				     __u32 insn_idx, const char *sym_name,
 				     const GElf_Sym *sym, const GElf_Rel *rel)
 {
 	struct bpf_insn *insn = &prog->insns[insn_idx];
@@ -3117,34 +3338,38 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 	struct bpf_object *obj = prog->obj;
 	__u32 shdr_idx = sym->st_shndx;
 	enum libbpf_map_type type;
+	const char *sym_sec_name;
 	struct bpf_map *map;
 
+	reloc_desc->processed = false;
+
 	/* sub-program call relocation */
 	if (insn->code == (BPF_JMP | BPF_CALL)) {
 		if (insn->src_reg != BPF_PSEUDO_CALL) {
-			pr_warn("incorrect bpf_call opcode\n");
+			pr_warn("prog '%s': incorrect bpf_call opcode\n", prog->name);
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		/* text_shndx can be 0, if no default "main" program exists */
 		if (!shdr_idx || shdr_idx != obj->efile.text_shndx) {
-			pr_warn("bad call relo against section %u\n", shdr_idx);
+			sym_sec_name = elf_sec_name(obj, elf_sec_by_idx(obj, shdr_idx));
+			pr_warn("prog '%s': bad call relo against '%s' in section '%s'\n",
+				prog->name, sym_name, sym_sec_name);
 			return -LIBBPF_ERRNO__RELOC;
 		}
-		if (sym->st_value % 8) {
-			pr_warn("bad call relo offset: %zu\n",
-				(size_t)sym->st_value);
+		if (sym->st_value % BPF_INSN_SZ) {
+			pr_warn("prog '%s': bad call relo against '%s' at offset %zu\n",
+				prog->name, sym_name, (size_t)sym->st_value);
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		reloc_desc->type = RELO_CALL;
 		reloc_desc->insn_idx = insn_idx;
 		reloc_desc->sym_off = sym->st_value;
-		obj->has_pseudo_calls = true;
 		return 0;
 	}
 
 	if (insn->code != (BPF_LD | BPF_IMM | BPF_DW)) {
-		pr_warn("invalid relo for insns[%d].code 0x%x\n",
-			insn_idx, insn->code);
+		pr_warn("prog '%s': invalid relo against '%s' for insns[%d].code 0x%x\n",
+			prog->name, sym_name, insn_idx, insn->code);
 		return -LIBBPF_ERRNO__RELOC;
 	}
 
@@ -3159,12 +3384,12 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 				break;
 		}
 		if (i >= n) {
-			pr_warn("extern relo failed to find extern for sym %d\n",
-				sym_idx);
+			pr_warn("prog '%s': extern relo failed to find extern for '%s' (%d)\n",
+				prog->name, sym_name, sym_idx);
 			return -LIBBPF_ERRNO__RELOC;
 		}
-		pr_debug("found extern #%d '%s' (sym %d) for insn %u\n",
-			 i, ext->name, ext->sym_idx, insn_idx);
+		pr_debug("prog '%s': found extern #%d '%s' (sym %d) for insn #%u\n",
+			 prog->name, i, ext->name, ext->sym_idx, insn_idx);
 		reloc_desc->type = RELO_EXTERN;
 		reloc_desc->insn_idx = insn_idx;
 		reloc_desc->sym_off = i; /* sym_off stores extern index */
@@ -3172,18 +3397,19 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 	}
 
 	if (!shdr_idx || shdr_idx >= SHN_LORESERVE) {
-		pr_warn("invalid relo for \'%s\' in special section 0x%x; forgot to initialize global var?..\n",
-			name, shdr_idx);
+		pr_warn("prog '%s': invalid relo against '%s' in special section 0x%x; forgot to initialize global var?..\n",
+			prog->name, sym_name, shdr_idx);
 		return -LIBBPF_ERRNO__RELOC;
 	}
 
 	type = bpf_object__section_to_libbpf_map_type(obj, shdr_idx);
+	sym_sec_name = elf_sec_name(obj, elf_sec_by_idx(obj, shdr_idx));
 
 	/* generic map reference relocation */
 	if (type == LIBBPF_MAP_UNSPEC) {
 		if (!bpf_object__shndx_is_maps(obj, shdr_idx)) {
-			pr_warn("bad map relo against section %u\n",
-				shdr_idx);
+			pr_warn("prog '%s': bad map relo against '%s' in section '%s'\n",
+				prog->name, sym_name, sym_sec_name);
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		for (map_idx = 0; map_idx < nr_maps; map_idx++) {
@@ -3192,14 +3418,14 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 			    map->sec_idx != sym->st_shndx ||
 			    map->sec_offset != sym->st_value)
 				continue;
-			pr_debug("found map %zd (%s, sec %d, off %zu) for insn %u\n",
-				 map_idx, map->name, map->sec_idx,
+			pr_debug("prog '%s': found map %zd (%s, sec %d, off %zu) for insn #%u\n",
+				 prog->name, map_idx, map->name, map->sec_idx,
 				 map->sec_offset, insn_idx);
 			break;
 		}
 		if (map_idx >= nr_maps) {
-			pr_warn("map relo failed to find map for sec %u, off %zu\n",
-				shdr_idx, (size_t)sym->st_value);
+			pr_warn("prog '%s': map relo failed to find map for section '%s', off %zu\n",
+				prog->name, sym_sec_name, (size_t)sym->st_value);
 			return -LIBBPF_ERRNO__RELOC;
 		}
 		reloc_desc->type = RELO_LD64;
@@ -3211,21 +3437,22 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 
 	/* global data map relocation */
 	if (!bpf_object__shndx_is_data(obj, shdr_idx)) {
-		pr_warn("bad data relo against section %u\n", shdr_idx);
+		pr_warn("prog '%s': bad data relo against section '%s'\n",
+			prog->name, sym_sec_name);
 		return -LIBBPF_ERRNO__RELOC;
 	}
 	for (map_idx = 0; map_idx < nr_maps; map_idx++) {
 		map = &obj->maps[map_idx];
 		if (map->libbpf_type != type)
 			continue;
-		pr_debug("found data map %zd (%s, sec %d, off %zu) for insn %u\n",
-			 map_idx, map->name, map->sec_idx, map->sec_offset,
-			 insn_idx);
+		pr_debug("prog '%s': found data map %zd (%s, sec %d, off %zu) for insn %u\n",
+			 prog->name, map_idx, map->name, map->sec_idx,
+			 map->sec_offset, insn_idx);
 		break;
 	}
 	if (map_idx >= nr_maps) {
-		pr_warn("data relo failed to find map for sec %u\n",
-			shdr_idx);
+		pr_warn("prog '%s': data relo failed to find map for section '%s'\n",
+			prog->name, sym_sec_name);
 		return -LIBBPF_ERRNO__RELOC;
 	}
 
@@ -3236,55 +3463,113 @@ static int bpf_program__record_reloc(struct bpf_program *prog,
 	return 0;
 }
 
+static bool prog_contains_insn(const struct bpf_program *prog, size_t insn_idx)
+{
+	return insn_idx >= prog->sec_insn_off &&
+	       insn_idx < prog->sec_insn_off + prog->sec_insn_cnt;
+}
+
+static struct bpf_program *find_prog_by_sec_insn(const struct bpf_object *obj,
+						 size_t sec_idx, size_t insn_idx)
+{
+	int l = 0, r = obj->nr_programs - 1, m;
+	struct bpf_program *prog;
+
+	while (l < r) {
+		m = l + (r - l + 1) / 2;
+		prog = &obj->programs[m];
+
+		if (prog->sec_idx < sec_idx ||
+		    (prog->sec_idx == sec_idx && prog->sec_insn_off <= insn_idx))
+			l = m;
+		else
+			r = m - 1;
+	}
+	/* matching program could be at index l, but it still might be the
+	 * wrong one, so we need to double check conditions for the last time
+	 */
+	prog = &obj->programs[l];
+	if (prog->sec_idx == sec_idx && prog_contains_insn(prog, insn_idx))
+		return prog;
+	return NULL;
+}
+
 static int
-bpf_program__collect_reloc(struct bpf_program *prog, GElf_Shdr *shdr,
-			   Elf_Data *data, struct bpf_object *obj)
+bpf_object__collect_prog_relos(struct bpf_object *obj, GElf_Shdr *shdr, Elf_Data *data)
 {
 	Elf_Data *symbols = obj->efile.symbols;
+	const char *relo_sec_name, *sec_name;
+	size_t sec_idx = shdr->sh_info;
+	struct bpf_program *prog;
+	struct reloc_desc *relos;
 	int err, i, nrels;
+	const char *sym_name;
+	__u32 insn_idx;
+	GElf_Sym sym;
+	GElf_Rel rel;
 
-	pr_debug("collecting relocating info for: '%s'\n", prog->section_name);
-	nrels = shdr->sh_size / shdr->sh_entsize;
+	relo_sec_name = elf_sec_str(obj, shdr->sh_name);
+	sec_name = elf_sec_name(obj, elf_sec_by_idx(obj, sec_idx));
+	if (!relo_sec_name || !sec_name)
+		return -EINVAL;
 
-	prog->reloc_desc = malloc(sizeof(*prog->reloc_desc) * nrels);
-	if (!prog->reloc_desc) {
-		pr_warn("failed to alloc memory in relocation\n");
-		return -ENOMEM;
-	}
-	prog->nr_reloc = nrels;
+	pr_debug("sec '%s': collecting relocation for section(%zu) '%s'\n",
+		 relo_sec_name, sec_idx, sec_name);
+	nrels = shdr->sh_size / shdr->sh_entsize;
 
 	for (i = 0; i < nrels; i++) {
-		const char *name;
-		__u32 insn_idx;
-		GElf_Sym sym;
-		GElf_Rel rel;
-
 		if (!gelf_getrel(data, i, &rel)) {
-			pr_warn("relocation: failed to get %d reloc\n", i);
+			pr_warn("sec '%s': failed to get relo #%d\n", relo_sec_name, i);
 			return -LIBBPF_ERRNO__FORMAT;
 		}
 		if (!gelf_getsym(symbols, GELF_R_SYM(rel.r_info), &sym)) {
-			pr_warn("relocation: symbol %"PRIx64" not found\n",
-				GELF_R_SYM(rel.r_info));
+			pr_warn("sec '%s': symbol 0x%zx not found for relo #%d\n",
+				relo_sec_name, (size_t)GELF_R_SYM(rel.r_info), i);
 			return -LIBBPF_ERRNO__FORMAT;
 		}
-		if (rel.r_offset % sizeof(struct bpf_insn))
+		if (rel.r_offset % BPF_INSN_SZ) {
+			pr_warn("sec '%s': invalid offset 0x%zx for relo #%d\n",
+				relo_sec_name, (size_t)GELF_R_SYM(rel.r_info), i);
 			return -LIBBPF_ERRNO__FORMAT;
+		}
 
-		insn_idx = rel.r_offset / sizeof(struct bpf_insn);
-		name = elf_strptr(obj->efile.elf, obj->efile.strtabidx,
-				  sym.st_name) ? : "<?>";
+		insn_idx = rel.r_offset / BPF_INSN_SZ;
+		/* relocations against static functions are recorded as
+		 * relocations against the section that contains a function;
+		 * in such case, symbol will be STT_SECTION and sym.st_name
+		 * will point to empty string (0), so fetch section name
+		 * instead
+		 */
+		if (GELF_ST_TYPE(sym.st_info) == STT_SECTION && sym.st_name == 0)
+			sym_name = elf_sec_name(obj, elf_sec_by_idx(obj, sym.st_shndx));
+		else
+			sym_name = elf_sym_str(obj, sym.st_name);
+		sym_name = sym_name ?: "<?";
+
+		pr_debug("sec '%s': relo #%d: insn #%u against '%s'\n",
+			 relo_sec_name, i, insn_idx, sym_name);
+
+		prog = find_prog_by_sec_insn(obj, sec_idx, insn_idx);
+		if (!prog) {
+			pr_warn("sec '%s': relo #%d: program not found in section '%s' for insn #%u\n",
+				relo_sec_name, i, sec_name, insn_idx);
+			return -LIBBPF_ERRNO__RELOC;
+		}
 
-		pr_debug("relo for shdr %u, symb %zu, value %zu, type %d, bind %d, name %d (\'%s\'), insn %u\n",
-			 (__u32)sym.st_shndx, (size_t)GELF_R_SYM(rel.r_info),
-			 (size_t)sym.st_value, GELF_ST_TYPE(sym.st_info),
-			 GELF_ST_BIND(sym.st_info), sym.st_name, name,
-			 insn_idx);
+		relos = libbpf_reallocarray(prog->reloc_desc,
+					    prog->nr_reloc + 1, sizeof(*relos));
+		if (!relos)
+			return -ENOMEM;
+		prog->reloc_desc = relos;
 
-		err = bpf_program__record_reloc(prog, &prog->reloc_desc[i],
-						insn_idx, name, &sym, &rel);
+		/* adjust insn_idx to local BPF program frame of reference */
+		insn_idx -= prog->sec_insn_off;
+		err = bpf_program__record_reloc(prog, &relos[prog->nr_reloc],
+						insn_idx, sym_name, &sym, &rel);
 		if (err)
 			return err;
+
+		prog->nr_reloc++;
 	}
 	return 0;
 }
@@ -3433,8 +3718,14 @@ bpf_object__probe_loading(struct bpf_object *obj)
 	return 0;
 }
 
-static int
-bpf_object__probe_name(struct bpf_object *obj)
+static int probe_fd(int fd)
+{
+	if (fd >= 0)
+		close(fd);
+	return fd >= 0;
+}
+
+static int probe_kern_prog_name(void)
 {
 	struct bpf_load_program_attr attr;
 	struct bpf_insn insns[] = {
@@ -3452,16 +3743,10 @@ bpf_object__probe_name(struct bpf_object *obj)
 	attr.license = "GPL";
 	attr.name = "test";
 	ret = bpf_load_program_xattr(&attr, NULL, 0);
-	if (ret >= 0) {
-		obj->caps.name = 1;
-		close(ret);
-	}
-
-	return 0;
+	return probe_fd(ret);
 }
 
-static int
-bpf_object__probe_global_data(struct bpf_object *obj)
+static int probe_kern_global_data(void)
 {
 	struct bpf_load_program_attr prg_attr;
 	struct bpf_create_map_attr map_attr;
@@ -3498,16 +3783,23 @@ bpf_object__probe_global_data(struct bpf_object *obj)
 	prg_attr.license = "GPL";
 
 	ret = bpf_load_program_xattr(&prg_attr, NULL, 0);
-	if (ret >= 0) {
-		obj->caps.global_data = 1;
-		close(ret);
-	}
-
 	close(map);
-	return 0;
+	return probe_fd(ret);
+}
+
+static int probe_kern_btf(void)
+{
+	static const char strs[] = "\0int";
+	__u32 types[] = {
+		/* int */
+		BTF_TYPE_INT_ENC(1, BTF_INT_SIGNED, 0, 32, 4),
+	};
+
+	return probe_fd(libbpf__load_raw_btf((char *)types, sizeof(types),
+					     strs, sizeof(strs)));
 }
 
-static int bpf_object__probe_btf_func(struct bpf_object *obj)
+static int probe_kern_btf_func(void)
 {
 	static const char strs[] = "\0int\0x\0a";
 	/* void x(int a) {} */
@@ -3520,20 +3812,12 @@ static int bpf_object__probe_btf_func(struct bpf_object *obj)
 		/* FUNC x */                                    /* [3] */
 		BTF_TYPE_ENC(5, BTF_INFO_ENC(BTF_KIND_FUNC, 0, 0), 2),
 	};
-	int btf_fd;
 
-	btf_fd = libbpf__load_raw_btf((char *)types, sizeof(types),
-				      strs, sizeof(strs));
-	if (btf_fd >= 0) {
-		obj->caps.btf_func = 1;
-		close(btf_fd);
-		return 1;
-	}
-
-	return 0;
+	return probe_fd(libbpf__load_raw_btf((char *)types, sizeof(types),
+					     strs, sizeof(strs)));
 }
 
-static int bpf_object__probe_btf_func_global(struct bpf_object *obj)
+static int probe_kern_btf_func_global(void)
 {
 	static const char strs[] = "\0int\0x\0a";
 	/* static void x(int a) {} */
@@ -3546,20 +3830,12 @@ static int bpf_object__probe_btf_func_global(struct bpf_object *obj)
 		/* FUNC x BTF_FUNC_GLOBAL */                    /* [3] */
 		BTF_TYPE_ENC(5, BTF_INFO_ENC(BTF_KIND_FUNC, 0, BTF_FUNC_GLOBAL), 2),
 	};
-	int btf_fd;
 
-	btf_fd = libbpf__load_raw_btf((char *)types, sizeof(types),
-				      strs, sizeof(strs));
-	if (btf_fd >= 0) {
-		obj->caps.btf_func_global = 1;
-		close(btf_fd);
-		return 1;
-	}
-
-	return 0;
+	return probe_fd(libbpf__load_raw_btf((char *)types, sizeof(types),
+					     strs, sizeof(strs)));
 }
 
-static int bpf_object__probe_btf_datasec(struct bpf_object *obj)
+static int probe_kern_btf_datasec(void)
 {
 	static const char strs[] = "\0x\0.data";
 	/* static int a; */
@@ -3573,20 +3849,12 @@ static int bpf_object__probe_btf_datasec(struct bpf_object *obj)
 		BTF_TYPE_ENC(3, BTF_INFO_ENC(BTF_KIND_DATASEC, 0, 1), 4),
 		BTF_VAR_SECINFO_ENC(2, 0, 4),
 	};
-	int btf_fd;
 
-	btf_fd = libbpf__load_raw_btf((char *)types, sizeof(types),
-				      strs, sizeof(strs));
-	if (btf_fd >= 0) {
-		obj->caps.btf_datasec = 1;
-		close(btf_fd);
-		return 1;
-	}
-
-	return 0;
+	return probe_fd(libbpf__load_raw_btf((char *)types, sizeof(types),
+					     strs, sizeof(strs)));
 }
 
-static int bpf_object__probe_array_mmap(struct bpf_object *obj)
+static int probe_kern_array_mmap(void)
 {
 	struct bpf_create_map_attr attr = {
 		.map_type = BPF_MAP_TYPE_ARRAY,
@@ -3595,27 +3863,17 @@ static int bpf_object__probe_array_mmap(struct bpf_object *obj)
 		.value_size = sizeof(int),
 		.max_entries = 1,
 	};
-	int fd;
-
-	fd = bpf_create_map_xattr(&attr);
-	if (fd >= 0) {
-		obj->caps.array_mmap = 1;
-		close(fd);
-		return 1;
-	}
 
-	return 0;
+	return probe_fd(bpf_create_map_xattr(&attr));
 }
 
-static int
-bpf_object__probe_exp_attach_type(struct bpf_object *obj)
+static int probe_kern_exp_attach_type(void)
 {
 	struct bpf_load_program_attr attr;
 	struct bpf_insn insns[] = {
 		BPF_MOV64_IMM(BPF_REG_0, 0),
 		BPF_EXIT_INSN(),
 	};
-	int fd;
 
 	memset(&attr, 0, sizeof(attr));
 	/* use any valid combination of program type and (optional)
@@ -3629,36 +3887,140 @@ bpf_object__probe_exp_attach_type(struct bpf_object *obj)
 	attr.insns_cnt = ARRAY_SIZE(insns);
 	attr.license = "GPL";
 
-	fd = bpf_load_program_xattr(&attr, NULL, 0);
-	if (fd >= 0) {
-		obj->caps.exp_attach_type = 1;
-		close(fd);
-		return 1;
-	}
-	return 0;
+	return probe_fd(bpf_load_program_xattr(&attr, NULL, 0));
 }
 
-static int
-bpf_object__probe_caps(struct bpf_object *obj)
+static int probe_kern_probe_read_kernel(void)
+{
+	struct bpf_load_program_attr attr;
+	struct bpf_insn insns[] = {
+		BPF_MOV64_REG(BPF_REG_1, BPF_REG_10),	/* r1 = r10 (fp) */
+		BPF_ALU64_IMM(BPF_ADD, BPF_REG_1, -8),	/* r1 += -8 */
+		BPF_MOV64_IMM(BPF_REG_2, 8),		/* r2 = 8 */
+		BPF_MOV64_IMM(BPF_REG_3, 0),		/* r3 = 0 */
+		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, 0, 0, BPF_FUNC_probe_read_kernel),
+		BPF_EXIT_INSN(),
+	};
+
+	memset(&attr, 0, sizeof(attr));
+	attr.prog_type = BPF_PROG_TYPE_KPROBE;
+	attr.insns = insns;
+	attr.insns_cnt = ARRAY_SIZE(insns);
+	attr.license = "GPL";
+
+	return probe_fd(bpf_load_program_xattr(&attr, NULL, 0));
+}
+
+static int probe_prog_bind_map(void)
 {
-	int (*probe_fn[])(struct bpf_object *obj) = {
-		bpf_object__probe_name,
-		bpf_object__probe_global_data,
-		bpf_object__probe_btf_func,
-		bpf_object__probe_btf_func_global,
-		bpf_object__probe_btf_datasec,
-		bpf_object__probe_array_mmap,
-		bpf_object__probe_exp_attach_type,
+	struct bpf_load_program_attr prg_attr;
+	struct bpf_create_map_attr map_attr;
+	char *cp, errmsg[STRERR_BUFSIZE];
+	struct bpf_insn insns[] = {
+		BPF_MOV64_IMM(BPF_REG_0, 0),
+		BPF_EXIT_INSN(),
 	};
-	int i, ret;
+	int ret, map, prog;
 
-	for (i = 0; i < ARRAY_SIZE(probe_fn); i++) {
-		ret = probe_fn[i](obj);
-		if (ret < 0)
-			pr_debug("Probe #%d failed with %d.\n", i, ret);
+	memset(&map_attr, 0, sizeof(map_attr));
+	map_attr.map_type = BPF_MAP_TYPE_ARRAY;
+	map_attr.key_size = sizeof(int);
+	map_attr.value_size = 32;
+	map_attr.max_entries = 1;
+
+	map = bpf_create_map_xattr(&map_attr);
+	if (map < 0) {
+		ret = -errno;
+		cp = libbpf_strerror_r(ret, errmsg, sizeof(errmsg));
+		pr_warn("Error in %s():%s(%d). Couldn't create simple array map.\n",
+			__func__, cp, -ret);
+		return ret;
 	}
 
-	return 0;
+	memset(&prg_attr, 0, sizeof(prg_attr));
+	prg_attr.prog_type = BPF_PROG_TYPE_SOCKET_FILTER;
+	prg_attr.insns = insns;
+	prg_attr.insns_cnt = ARRAY_SIZE(insns);
+	prg_attr.license = "GPL";
+
+	prog = bpf_load_program_xattr(&prg_attr, NULL, 0);
+	if (prog < 0) {
+		close(map);
+		return 0;
+	}
+
+	ret = bpf_prog_bind_map(prog, map, NULL);
+
+	close(map);
+	close(prog);
+
+	return ret >= 0;
+}
+
+enum kern_feature_result {
+	FEAT_UNKNOWN = 0,
+	FEAT_SUPPORTED = 1,
+	FEAT_MISSING = 2,
+};
+
+typedef int (*feature_probe_fn)(void);
+
+static struct kern_feature_desc {
+	const char *desc;
+	feature_probe_fn probe;
+	enum kern_feature_result res;
+} feature_probes[__FEAT_CNT] = {
+	[FEAT_PROG_NAME] = {
+		"BPF program name", probe_kern_prog_name,
+	},
+	[FEAT_GLOBAL_DATA] = {
+		"global variables", probe_kern_global_data,
+	},
+	[FEAT_BTF] = {
+		"minimal BTF", probe_kern_btf,
+	},
+	[FEAT_BTF_FUNC] = {
+		"BTF functions", probe_kern_btf_func,
+	},
+	[FEAT_BTF_GLOBAL_FUNC] = {
+		"BTF global function", probe_kern_btf_func_global,
+	},
+	[FEAT_BTF_DATASEC] = {
+		"BTF data section and variable", probe_kern_btf_datasec,
+	},
+	[FEAT_ARRAY_MMAP] = {
+		"ARRAY map mmap()", probe_kern_array_mmap,
+	},
+	[FEAT_EXP_ATTACH_TYPE] = {
+		"BPF_PROG_LOAD expected_attach_type attribute",
+		probe_kern_exp_attach_type,
+	},
+	[FEAT_PROBE_READ_KERN] = {
+		"bpf_probe_read_kernel() helper", probe_kern_probe_read_kernel,
+	},
+	[FEAT_PROG_BIND_MAP] = {
+		"BPF_PROG_BIND_MAP support", probe_prog_bind_map,
+	}
+};
+
+static bool kernel_supports(enum kern_feature_id feat_id)
+{
+	struct kern_feature_desc *feat = &feature_probes[feat_id];
+	int ret;
+
+	if (READ_ONCE(feat->res) == FEAT_UNKNOWN) {
+		ret = feat->probe();
+		if (ret > 0) {
+			WRITE_ONCE(feat->res, FEAT_SUPPORTED);
+		} else if (ret == 0) {
+			WRITE_ONCE(feat->res, FEAT_MISSING);
+		} else {
+			pr_warn("Detection of kernel %s support failed: %d\n", feat->desc, ret);
+			WRITE_ONCE(feat->res, FEAT_MISSING);
+		}
+	}
+
+	return READ_ONCE(feat->res) == FEAT_SUPPORTED;
 }
 
 static bool map_is_reuse_compat(const struct bpf_map *map, int map_fd)
@@ -3760,7 +4122,7 @@ static int bpf_object__create_map(struct bpf_object *obj, struct bpf_map *map)
 
 	memset(&create_attr, 0, sizeof(create_attr));
 
-	if (obj->caps.name)
+	if (kernel_supports(FEAT_PROG_NAME))
 		create_attr.name = map->name;
 	create_attr.map_ifindex = map->map_ifindex;
 	create_attr.map_type = def->type;
@@ -3841,6 +4203,36 @@ static int bpf_object__create_map(struct bpf_object *obj, struct bpf_map *map)
 	return 0;
 }
 
+static int init_map_slots(struct bpf_map *map)
+{
+	const struct bpf_map *targ_map;
+	unsigned int i;
+	int fd, err;
+
+	for (i = 0; i < map->init_slots_sz; i++) {
+		if (!map->init_slots[i])
+			continue;
+
+		targ_map = map->init_slots[i];
+		fd = bpf_map__fd(targ_map);
+		err = bpf_map_update_elem(map->fd, &i, &fd, 0);
+		if (err) {
+			err = -errno;
+			pr_warn("map '%s': failed to initialize slot [%d] to map '%s' fd=%d: %d\n",
+				map->name, i, targ_map->name,
+				fd, err);
+			return err;
+		}
+		pr_debug("map '%s': slot [%d] set to map '%s' fd=%d\n",
+			 map->name, i, targ_map->name, fd);
+	}
+
+	zfree(&map->init_slots);
+	map->init_slots_sz = 0;
+
+	return 0;
+}
+
 static int
 bpf_object__create_maps(struct bpf_object *obj)
 {
@@ -3864,47 +4256,29 @@ bpf_object__create_maps(struct bpf_object *obj)
 		if (map->fd >= 0) {
 			pr_debug("map '%s': skipping creation (preset fd=%d)\n",
 				 map->name, map->fd);
-			continue;
-		}
-
-		err = bpf_object__create_map(obj, map);
-		if (err)
-			goto err_out;
-
-		pr_debug("map '%s': created successfully, fd=%d\n", map->name,
-			 map->fd);
-
-		if (bpf_map__is_internal(map)) {
-			err = bpf_object__populate_internal_map(obj, map);
-			if (err < 0) {
-				zclose(map->fd);
+		} else {
+			err = bpf_object__create_map(obj, map);
+			if (err)
 				goto err_out;
-			}
-		}
 
-		if (map->init_slots_sz) {
-			for (j = 0; j < map->init_slots_sz; j++) {
-				const struct bpf_map *targ_map;
-				int fd;
+			pr_debug("map '%s': created successfully, fd=%d\n",
+				 map->name, map->fd);
 
-				if (!map->init_slots[j])
-					continue;
+			if (bpf_map__is_internal(map)) {
+				err = bpf_object__populate_internal_map(obj, map);
+				if (err < 0) {
+					zclose(map->fd);
+					goto err_out;
+				}
+			}
 
-				targ_map = map->init_slots[j];
-				fd = bpf_map__fd(targ_map);
-				err = bpf_map_update_elem(map->fd, &j, &fd, 0);
-				if (err) {
-					err = -errno;
-					pr_warn("map '%s': failed to initialize slot [%d] to map '%s' fd=%d: %d\n",
-						map->name, j, targ_map->name,
-						fd, err);
+			if (map->init_slots_sz) {
+				err = init_map_slots(map);
+				if (err < 0) {
+					zclose(map->fd);
 					goto err_out;
 				}
-				pr_debug("map '%s': slot [%d] set to map '%s' fd=%d\n",
-					 map->name, j, targ_map->name, fd);
 			}
-			zfree(&map->init_slots);
-			map->init_slots_sz = 0;
 		}
 
 		if (map->pin_path && !map->pinned) {
@@ -3929,75 +4303,6 @@ err_out:
 	return err;
 }
 
-static int
-check_btf_ext_reloc_err(struct bpf_program *prog, int err,
-			void *btf_prog_info, const char *info_name)
-{
-	if (err != -ENOENT) {
-		pr_warn("Error in loading %s for sec %s.\n",
-			info_name, prog->section_name);
-		return err;
-	}
-
-	/* err == -ENOENT (i.e. prog->section_name not found in btf_ext) */
-
-	if (btf_prog_info) {
-		/*
-		 * Some info has already been found but has problem
-		 * in the last btf_ext reloc. Must have to error out.
-		 */
-		pr_warn("Error in relocating %s for sec %s.\n",
-			info_name, prog->section_name);
-		return err;
-	}
-
-	/* Have problem loading the very first info. Ignore the rest. */
-	pr_warn("Cannot find %s for main program sec %s. Ignore all %s.\n",
-		info_name, prog->section_name, info_name);
-	return 0;
-}
-
-static int
-bpf_program_reloc_btf_ext(struct bpf_program *prog, struct bpf_object *obj,
-			  const char *section_name,  __u32 insn_offset)
-{
-	int err;
-
-	if (!insn_offset || prog->func_info) {
-		/*
-		 * !insn_offset => main program
-		 *
-		 * For sub prog, the main program's func_info has to
-		 * be loaded first (i.e. prog->func_info != NULL)
-		 */
-		err = btf_ext__reloc_func_info(obj->btf, obj->btf_ext,
-					       section_name, insn_offset,
-					       &prog->func_info,
-					       &prog->func_info_cnt);
-		if (err)
-			return check_btf_ext_reloc_err(prog, err,
-						       prog->func_info,
-						       "bpf_func_info");
-
-		prog->func_info_rec_size = btf_ext__func_info_rec_size(obj->btf_ext);
-	}
-
-	if (!insn_offset || prog->line_info) {
-		err = btf_ext__reloc_line_info(obj->btf, obj->btf_ext,
-					       section_name, insn_offset,
-					       &prog->line_info,
-					       &prog->line_info_cnt);
-		if (err)
-			return check_btf_ext_reloc_err(prog, err,
-						       prog->line_info,
-						       "bpf_line_info");
-
-		prog->line_info_rec_size = btf_ext__line_info_rec_size(obj->btf_ext);
-	}
-
-	return 0;
-}
-
 #define BPF_CORE_SPEC_MAX_LEN 64
 
 /* represents BPF CO-RE field or array element accessor */
@@ -4011,6 +4316,10 @@ struct bpf_core_spec {
 	const struct btf *btf;
 	/* high-level spec: named fields and array indices only */
 	struct bpf_core_accessor spec[BPF_CORE_SPEC_MAX_LEN];
+	/* original unresolved (no skip_mods_or_typedefs) root type ID */
+	__u32 root_type_id;
+	/* CO-RE relocation kind */
+	enum bpf_core_relo_kind relo_kind;
 	/* high-level spec length */
 	int len;
 	/* raw, low-level spec: 1-to-1 with accessor spec string */
@@ -4041,8 +4350,66 @@ static bool is_flex_arr(const struct btf *btf,
 	return acc->idx == btf_vlen(t) - 1;
 }
 
+static const char *core_relo_kind_str(enum bpf_core_relo_kind kind)
+{
+	switch (kind) {
+	case BPF_FIELD_BYTE_OFFSET: return "byte_off";
+	case BPF_FIELD_BYTE_SIZE: return "byte_sz";
+	case BPF_FIELD_EXISTS: return "field_exists";
+	case BPF_FIELD_SIGNED: return "signed";
+	case BPF_FIELD_LSHIFT_U64: return "lshift_u64";
+	case BPF_FIELD_RSHIFT_U64: return "rshift_u64";
+	case BPF_TYPE_ID_LOCAL: return "local_type_id";
+	case BPF_TYPE_ID_TARGET: return "target_type_id";
+	case BPF_TYPE_EXISTS: return "type_exists";
+	case BPF_TYPE_SIZE: return "type_size";
+	case BPF_ENUMVAL_EXISTS: return "enumval_exists";
+	case BPF_ENUMVAL_VALUE: return "enumval_value";
+	default: return "unknown";
+	}
+}
+
+static bool core_relo_is_field_based(enum bpf_core_relo_kind kind)
+{
+	switch (kind) {
+	case BPF_FIELD_BYTE_OFFSET:
+	case BPF_FIELD_BYTE_SIZE:
+	case BPF_FIELD_EXISTS:
+	case BPF_FIELD_SIGNED:
+	case BPF_FIELD_LSHIFT_U64:
+	case BPF_FIELD_RSHIFT_U64:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static bool core_relo_is_type_based(enum bpf_core_relo_kind kind)
+{
+	switch (kind) {
+	case BPF_TYPE_ID_LOCAL:
+	case BPF_TYPE_ID_TARGET:
+	case BPF_TYPE_EXISTS:
+	case BPF_TYPE_SIZE:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static bool core_relo_is_enumval_based(enum bpf_core_relo_kind kind)
+{
+	switch (kind) {
+	case BPF_ENUMVAL_EXISTS:
+	case BPF_ENUMVAL_VALUE:
+		return true;
+	default:
+		return false;
+	}
+}
+
 /*
- * Turn bpf_field_reloc into a low- and high-level spec representation,
+ * Turn bpf_core_relo into a low- and high-level spec representation,
  * validating correctness along the way, as well as calculating resulting
  * field bit offset, specified by accessor string. Low-level spec captures
  * every single level of nestedness, including traversing anonymous
@@ -4071,10 +4438,17 @@ static bool is_flex_arr(const struct btf *btf,
  *   - field 'a' access (corresponds to '2' in low-level spec);
  *   - array element #3 access (corresponds to '3' in low-level spec).
  *
+ * Type-based relocations (TYPE_EXISTS/TYPE_SIZE,
+ * TYPE_ID_LOCAL/TYPE_ID_TARGET) don't capture any field information. Their
+ * spec and raw_spec are kept empty.
+ *
+ * Enum value-based relocations (ENUMVAL_EXISTS/ENUMVAL_VALUE) use access
+ * string to specify enumerator's value index that need to be relocated.
  */
-static int bpf_core_spec_parse(const struct btf *btf,
+static int bpf_core_parse_spec(const struct btf *btf,
 			       __u32 type_id,
 			       const char *spec_str,
+			       enum bpf_core_relo_kind relo_kind,
 			       struct bpf_core_spec *spec)
 {
 	int access_idx, parsed_len, i;
@@ -4089,6 +4463,15 @@ static int bpf_core_spec_parse(const struct btf *btf,
 
 	memset(spec, 0, sizeof(*spec));
 	spec->btf = btf;
+	spec->root_type_id = type_id;
+	spec->relo_kind = relo_kind;
+
+	/* type-based relocations don't have a field access string */
+	if (core_relo_is_type_based(relo_kind)) {
+		if (strcmp(spec_str, "0"))
+			return -EINVAL;
+		return 0;
+	}
 
 	/* parse spec_str="0:1:2:3:4" into array raw_spec=[0, 1, 2, 3, 4] */
 	while (*spec_str) {
@@ -4105,16 +4488,28 @@ static int bpf_core_spec_parse(const struct btf *btf,
 	if (spec->raw_len == 0)
 		return -EINVAL;
 
-	/* first spec value is always reloc type array index */
 	t = skip_mods_and_typedefs(btf, type_id, &id);
 	if (!t)
 		return -EINVAL;
 
 	access_idx = spec->raw_spec[0];
-	spec->spec[0].type_id = id;
-	spec->spec[0].idx = access_idx;
+	acc = &spec->spec[0];
+	acc->type_id = id;
+	acc->idx = access_idx;
 	spec->len++;
 
+	if (core_relo_is_enumval_based(relo_kind)) {
+		if (!btf_is_enum(t) || spec->raw_len > 1 || access_idx >= btf_vlen(t))
+			return -EINVAL;
+
+		/* record enumerator name in a first accessor */
+		acc->name = btf__name_by_offset(btf, btf_enum(t)[access_idx].name_off);
+		return 0;
+	}
+
+	if (!core_relo_is_field_based(relo_kind))
+		return -EINVAL;
+
 	sz = btf__resolve_size(btf, id);
 	if (sz < 0)
 		return sz;
@@ -4172,8 +4567,8 @@ static int bpf_core_spec_parse(const struct btf *btf,
 				return sz;
 			spec->bit_offset += access_idx * sz * 8;
 		} else {
-			pr_warn("relo for [%u] %s (at idx %d) captures type [%d] of unexpected kind %d\n",
-				type_id, spec_str, i, id, btf_kind(t));
+			pr_warn("relo for [%u] %s (at idx %d) captures type [%d] of unexpected kind %s\n",
+				type_id, spec_str, i, id, btf_kind_str(t));
 			return -EINVAL;
 		}
 	}
@@ -4223,16 +4618,16 @@ static struct ids_vec *bpf_core_find_cands(const struct btf *local_btf,
 {
 	size_t local_essent_len, targ_essent_len;
 	const char *local_name, *targ_name;
-	const struct btf_type *t;
+	const struct btf_type *t, *local_t;
 	struct ids_vec *cand_ids;
 	__u32 *new_ids;
 	int i, err, n;
 
-	t = btf__type_by_id(local_btf, local_type_id);
-	if (!t)
+	local_t = btf__type_by_id(local_btf, local_type_id);
+	if (!local_t)
 		return ERR_PTR(-EINVAL);
 
-	local_name = btf__name_by_offset(local_btf, t->name_off);
+	local_name = btf__name_by_offset(local_btf, local_t->name_off);
 	if (str_is_empty(local_name))
 		return ERR_PTR(-EINVAL);
 	local_essent_len = bpf_core_essential_name_len(local_name);
@@ -4244,12 +4639,11 @@ static struct ids_vec *bpf_core_find_cands(const struct btf *local_btf,
 	n = btf__get_nr_types(targ_btf);
 	for (i = 1; i <= n; i++) {
 		t = btf__type_by_id(targ_btf, i);
-		targ_name = btf__name_by_offset(targ_btf, t->name_off);
-		if (str_is_empty(targ_name))
+		if (btf_kind(t) != btf_kind(local_t))
 			continue;
 
-		t = skip_mods_and_typedefs(targ_btf, i, NULL);
-		if (!btf_is_composite(t) && !btf_is_array(t))
+		targ_name = btf__name_by_offset(targ_btf, t->name_off);
+		if (str_is_empty(targ_name))
 			continue;
 
 		targ_essent_len = bpf_core_essential_name_len(targ_name);
@@ -4257,11 +4651,12 @@ static struct ids_vec *bpf_core_find_cands(const struct btf *local_btf,
 			continue;
 
 		if (strncmp(local_name, targ_name, local_essent_len) == 0) {
-			pr_debug("[%d] %s: found candidate [%d] %s\n",
-				 local_type_id, local_name, i, targ_name);
-			new_ids = reallocarray(cand_ids->data,
-					       cand_ids->len + 1,
-					       sizeof(*cand_ids->data));
+			pr_debug("CO-RE relocating [%d] %s %s: found target candidate [%d] %s %s\n",
+				 local_type_id, btf_kind_str(local_t),
+				 local_name, i, btf_kind_str(t), targ_name);
+			new_ids = libbpf_reallocarray(cand_ids->data,
+						      cand_ids->len + 1,
+						      sizeof(*cand_ids->data));
 			if (!new_ids) {
 				err = -ENOMEM;
 				goto err_out;
@@ -4276,8 +4671,9 @@ err_out:
 	return ERR_PTR(err);
 }
 
-/* Check two types for compatibility, skipping const/volatile/restrict and
- * typedefs, to ensure we are relocating compatible entities:
+/* Check two types for compatibility for the purpose of field access
+ * relocation. const/volatile/restrict and typedefs are skipped to ensure we
+ * are relocating semantically compatible entities:
  *   - any two STRUCTs/UNIONs are compatible and can be mixed;
  *   - any two FWDs are compatible, if their names match (modulo flavor suffix);
  *   - any two PTRs are always compatible;
@@ -4432,6 +4828,100 @@ static int bpf_core_match_member(const struct btf *local_btf,
 	return 0;
 }
 
+/* Check local and target types for compatibility. This check is used for
+ * type-based CO-RE relocations and follow slightly different rules than
+ * field-based relocations. This function assumes that root types were already
+ * checked for name match. Beyond that initial root-level name check, names
+ * are completely ignored. Compatibility rules are as follows:
+ *   - any two STRUCTs/UNIONs/FWDs/ENUMs/INTs are considered compatible, but
+ *     kind should match for local and target types (i.e., STRUCT is not
+ *     compatible with UNION);
+ *   - for ENUMs, the size is ignored;
+ *   - for INT, size and signedness are ignored;
+ *   - for ARRAY, dimensionality is ignored, element types are checked for
+ *     compatibility recursively;
+ *   - CONST/VOLATILE/RESTRICT modifiers are ignored;
+ *   - TYPEDEFs/PTRs are compatible if types they pointing to are compatible;
+ *   - FUNC_PROTOs are compatible if they have compatible signature: same
+ *     number of input args and compatible return and argument types.
+ * These rules are not set in stone and probably will be adjusted as we get
+ * more experience with using BPF CO-RE relocations.
+ */
+static int bpf_core_types_are_compat(const struct btf *local_btf, __u32 local_id,
+				     const struct btf *targ_btf, __u32 targ_id)
+{
+	const struct btf_type *local_type, *targ_type;
+	int depth = 32; /* max recursion depth */
+
+	/* caller made sure that names match (ignoring flavor suffix) */
+	local_type = btf__type_by_id(local_btf, local_id);
+	targ_type = btf__type_by_id(targ_btf, targ_id);
+	if (btf_kind(local_type) != btf_kind(targ_type))
+		return 0;
+
+recur:
+	depth--;
+	if (depth < 0)
+		return -EINVAL;
+
+	local_type = skip_mods_and_typedefs(local_btf, local_id, &local_id);
+	targ_type = skip_mods_and_typedefs(targ_btf, targ_id, &targ_id);
+	if (!local_type || !targ_type)
+		return -EINVAL;
+
+	if (btf_kind(local_type) != btf_kind(targ_type))
+		return 0;
+
+	switch (btf_kind(local_type)) {
+	case BTF_KIND_UNKN:
+	case BTF_KIND_STRUCT:
+	case BTF_KIND_UNION:
+	case BTF_KIND_ENUM:
+	case BTF_KIND_FWD:
+		return 1;
+	case BTF_KIND_INT:
+		/* just reject deprecated bitfield-like integers; all other
+		 * integers are by default compatible between each other
+		 */
+		return btf_int_offset(local_type) == 0 && btf_int_offset(targ_type) == 0;
+	case BTF_KIND_PTR:
+		local_id = local_type->type;
+		targ_id = targ_type->type;
+		goto recur;
+	case BTF_KIND_ARRAY:
+		local_id = btf_array(local_type)->type;
+		targ_id = btf_array(targ_type)->type;
+		goto recur;
+	case BTF_KIND_FUNC_PROTO: {
+		struct btf_param *local_p = btf_params(local_type);
+		struct btf_param *targ_p = btf_params(targ_type);
+		__u16 local_vlen = btf_vlen(local_type);
+		__u16 targ_vlen = btf_vlen(targ_type);
+		int i, err;
+
+		if (local_vlen != targ_vlen)
+			return 0;
+
+		for (i = 0; i < local_vlen; i++, local_p++, targ_p++) {
+			skip_mods_and_typedefs(local_btf, local_p->type, &local_id);
+			skip_mods_and_typedefs(targ_btf, targ_p->type, &targ_id);
+			err = bpf_core_types_are_compat(local_btf, local_id, targ_btf, targ_id);
+			if (err <= 0)
+				return err;
+		}
+
+		/* tail recurse for return type check */
+		skip_mods_and_typedefs(local_btf, local_type->type, &local_id);
+		skip_mods_and_typedefs(targ_btf, targ_type->type, &targ_id);
+		goto recur;
+	}
+	default:
+		pr_warn("unexpected kind %s relocated, local [%d], target [%d]\n",
+			btf_kind_str(local_type), local_id, targ_id);
+		return 0;
+	}
+}
+
 /*
  * Try to match local spec to a target type and, if successful, produce full
  * target spec (high-level, low-level + bit offset).
@@ -4447,10 +4937,51 @@ static int bpf_core_spec_match(struct bpf_core_spec *local_spec,
 
 	memset(targ_spec, 0, sizeof(*targ_spec));
 	targ_spec->btf = targ_btf;
+	targ_spec->root_type_id = targ_id;
+	targ_spec->relo_kind = local_spec->relo_kind;
+
+	if (core_relo_is_type_based(local_spec->relo_kind)) {
+		return bpf_core_types_are_compat(local_spec->btf,
+						 local_spec->root_type_id,
+						 targ_btf, targ_id);
+	}
 
 	local_acc = &local_spec->spec[0];
 	targ_acc = &targ_spec->spec[0];
 
+	if (core_relo_is_enumval_based(local_spec->relo_kind)) {
+		size_t local_essent_len, targ_essent_len;
+		const struct btf_enum *e;
+		const char *targ_name;
+
+		/* has to resolve to an enum */
+		targ_type = skip_mods_and_typedefs(targ_spec->btf, targ_id, &targ_id);
+		if (!btf_is_enum(targ_type))
+			return 0;
+
+		local_essent_len = bpf_core_essential_name_len(local_acc->name);
+
+		for (i = 0, e = btf_enum(targ_type); i < btf_vlen(targ_type); i++, e++) {
+			targ_name = btf__name_by_offset(targ_spec->btf, e->name_off);
+			targ_essent_len = bpf_core_essential_name_len(targ_name);
+			if (targ_essent_len != local_essent_len)
+				continue;
+			if (strncmp(local_acc->name, targ_name, local_essent_len) == 0) {
+				targ_acc->type_id = targ_id;
+				targ_acc->idx = i;
+				targ_acc->name = targ_name;
+				targ_spec->len++;
+				targ_spec->raw_spec[targ_spec->raw_len] = targ_acc->idx;
+				targ_spec->raw_len++;
+				return 1;
+			}
+		}
+		return 0;
+	}
+
+	if (!core_relo_is_field_based(local_spec->relo_kind))
+		return -EINVAL;
+
 	for (i = 0; i < local_spec->len; i++, local_acc++, targ_acc++) {
 		targ_type = skip_mods_and_typedefs(targ_spec->btf, targ_id,
 						   &targ_id);
@@ -4507,22 +5038,42 @@ static int bpf_core_spec_match(struct bpf_core_spec *local_spec,
 }
 
 static int bpf_core_calc_field_relo(const struct bpf_program *prog,
-				    const struct bpf_field_reloc *relo,
+				    const struct bpf_core_relo *relo,
 				    const struct bpf_core_spec *spec,
-				    __u32 *val, bool *validate)
+				    __u32 *val, __u32 *field_sz, __u32 *type_id,
+				    bool *validate)
 {
-	const struct bpf_core_accessor *acc = &spec->spec[spec->len - 1];
-	const struct btf_type *t = btf__type_by_id(spec->btf, acc->type_id);
-	__u32 byte_off, byte_sz, bit_off, bit_sz;
+	const struct bpf_core_accessor *acc;
+	const struct btf_type *t;
+	__u32 byte_off, byte_sz, bit_off, bit_sz, field_type_id;
 	const struct btf_member *m;
 	const struct btf_type *mt;
 	bool bitfield;
 	__s64 sz;
 
+	*field_sz = 0;
+
+	if (relo->kind == BPF_FIELD_EXISTS) {
+		*val = spec ? 1 : 0;
+		return 0;
+	}
+
+	if (!spec)
+		return -EUCLEAN; /* request instruction poisoning */
+
+	acc = &spec->spec[spec->len - 1];
+	t = btf__type_by_id(spec->btf, acc->type_id);
+
 	/* a[n] accessor needs special handling */
 	if (!acc->name) {
 		if (relo->kind == BPF_FIELD_BYTE_OFFSET) {
 			*val = spec->bit_offset / 8;
+			/* remember field size for load/store mem size */
+			sz = btf__resolve_size(spec->btf, acc->type_id);
+			if (sz < 0)
+				return -EINVAL;
+			*field_sz = sz;
+			*type_id = acc->type_id;
 		} else if (relo->kind == BPF_FIELD_BYTE_SIZE) {
 			sz = btf__resolve_size(spec->btf, acc->type_id);
 			if (sz < 0)
@@ -4530,8 +5081,7 @@ static int bpf_core_calc_field_relo(const struct bpf_program *prog,
 			*val = sz;
 		} else {
 			pr_warn("prog '%s': relo %d at insn #%d can't be applied to array access\n",
-				bpf_program__title(prog, false),
-				relo->kind, relo->insn_off / 8);
+				prog->name, relo->kind, relo->insn_off / 8);
 			return -EINVAL;
 		}
 		if (validate)
@@ -4540,7 +5090,7 @@ static int bpf_core_calc_field_relo(const struct bpf_program *prog,
 	}
 
 	m = btf_members(t) + acc->idx;
-	mt = skip_mods_and_typedefs(spec->btf, m->type, NULL);
+	mt = skip_mods_and_typedefs(spec->btf, m->type, &field_type_id);
 	bit_off = spec->bit_offset;
 	bit_sz = btf_member_bitfield_size(t, acc->idx);
 
@@ -4553,15 +5103,14 @@ static int bpf_core_calc_field_relo(const struct bpf_program *prog,
 			if (byte_sz >= 8) {
 				/* bitfield can't be read with 64-bit read */
 				pr_warn("prog '%s': relo %d at insn #%d can't be satisfied for bitfield\n",
-					bpf_program__title(prog, false),
-					relo->kind, relo->insn_off / 8);
+					prog->name, relo->kind, relo->insn_off / 8);
 				return -E2BIG;
 			}
 			byte_sz *= 2;
 			byte_off = bit_off / 8 / byte_sz * byte_sz;
 		}
 	} else {
-		sz = btf__resolve_size(spec->btf, m->type);
+		sz = btf__resolve_size(spec->btf, field_type_id);
 		if (sz < 0)
 			return -EINVAL;
 		byte_sz = sz;
@@ -4579,6 +5128,10 @@ static int bpf_core_calc_field_relo(const struct bpf_program *prog,
 	switch (relo->kind) {
 	case BPF_FIELD_BYTE_OFFSET:
 		*val = byte_off;
+		if (!bitfield) {
+			*field_sz = byte_sz;
+			*type_id = field_type_id;
+		}
 		break;
 	case BPF_FIELD_BYTE_SIZE:
 		*val = byte_sz;
@@ -4604,117 +5157,384 @@ static int bpf_core_calc_field_relo(const struct bpf_program *prog,
 		break;
 	case BPF_FIELD_EXISTS:
 	default:
-		pr_warn("prog '%s': unknown relo %d at insn #%d\n",
-			bpf_program__title(prog, false),
-			relo->kind, relo->insn_off / 8);
-		return -EINVAL;
+		return -EOPNOTSUPP;
 	}
 
 	return 0;
 }
 
+static int bpf_core_calc_type_relo(const struct bpf_core_relo *relo,
+				   const struct bpf_core_spec *spec,
+				   __u32 *val)
+{
+	__s64 sz;
+
+	/* type-based relos return zero when target type is not found */
+	if (!spec) {
+		*val = 0;
+		return 0;
+	}
+
+	switch (relo->kind) {
+	case BPF_TYPE_ID_TARGET:
+		*val = spec->root_type_id;
+		break;
+	case BPF_TYPE_EXISTS:
+		*val = 1;
+		break;
+	case BPF_TYPE_SIZE:
+		sz = btf__resolve_size(spec->btf, spec->root_type_id);
+		if (sz < 0)
+			return -EINVAL;
+		*val = sz;
+		break;
+	case BPF_TYPE_ID_LOCAL:
+	/* BPF_TYPE_ID_LOCAL is handled specially and shouldn't get here */
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+static int bpf_core_calc_enumval_relo(const struct bpf_core_relo *relo,
+				      const struct bpf_core_spec *spec,
+				      __u32 *val)
+{
+	const struct btf_type *t;
+	const struct btf_enum *e;
+
+	switch (relo->kind) {
+	case BPF_ENUMVAL_EXISTS:
+		*val = spec ? 1 : 0;
+		break;
+	case BPF_ENUMVAL_VALUE:
+		if (!spec)
+			return -EUCLEAN; /* request instruction poisoning */
+		t = btf__type_by_id(spec->btf, spec->spec[0].type_id);
+		e = btf_enum(t) + spec->spec[0].idx;
+		*val = e->val;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
+	return 0;
+}
+
+struct bpf_core_relo_res
+{
+	/* expected value in the instruction, unless validate == false */
+	__u32 orig_val;
+	/* new value that needs to be patched up to */
+	__u32 new_val;
+	/* relocation unsuccessful, poison instruction, but don't fail load */
+	bool poison;
+	/* some relocations can't be validated against orig_val */
+	bool validate;
+	/* for field byte offset relocations or the forms:
+	 *     *(T *)(rX + <off>) = rY
+	 *     rX = *(T *)(rY + <off>),
+	 * we remember original and resolved field size to adjust direct
+	 * memory loads of pointers and integers; this is necessary for 32-bit
+	 * host kernel architectures, but also allows to automatically
+	 * relocate fields that were resized from, e.g., u32 to u64, etc.
+	 */
+	bool fail_memsz_adjust;
+	__u32 orig_sz;
+	__u32 orig_type_id;
+	__u32 new_sz;
+	__u32 new_type_id;
+};
+
+/* Calculate original and target relocation values, given local and target
+ * specs and relocation kind. These values are calculated for each candidate.
+ * If there are multiple candidates, resulting values should all be consistent
+ * with each other. Otherwise, libbpf will refuse to proceed due to ambiguity.
+ * If instruction has to be poisoned, *poison will be set to true.
+ */
+static int bpf_core_calc_relo(const struct bpf_program *prog,
+			      const struct bpf_core_relo *relo,
+			      int relo_idx,
+			      const struct bpf_core_spec *local_spec,
+			      const struct bpf_core_spec *targ_spec,
+			      struct bpf_core_relo_res *res)
+{
+	int err = -EOPNOTSUPP;
+
+	res->orig_val = 0;
+	res->new_val = 0;
+	res->poison = false;
+	res->validate = true;
+	res->fail_memsz_adjust = false;
+	res->orig_sz = res->new_sz = 0;
+	res->orig_type_id = res->new_type_id = 0;
+
+	if (core_relo_is_field_based(relo->kind)) {
+		err = bpf_core_calc_field_relo(prog, relo, local_spec,
+					       &res->orig_val, &res->orig_sz,
+					       &res->orig_type_id, &res->validate);
+		err = err ?: bpf_core_calc_field_relo(prog, relo, targ_spec,
+						      &res->new_val, &res->new_sz,
+						      &res->new_type_id, NULL);
+		if (err)
+			goto done;
+		/* Validate if it's safe to adjust load/store memory size.
+		 * Adjustments are performed only if original and new memory
+		 * sizes differ.
+		 */
+		res->fail_memsz_adjust = false;
+		if (res->orig_sz != res->new_sz) {
+			const struct btf_type *orig_t, *new_t;
+
+			orig_t = btf__type_by_id(local_spec->btf, res->orig_type_id);
+			new_t = btf__type_by_id(targ_spec->btf, res->new_type_id);
+
+			/* There are two use cases in which it's safe to
+			 * adjust load/store's mem size:
+			 *   - reading a 32-bit kernel pointer, while on BPF
+			 *   size pointers are always 64-bit; in this case
+			 *   it's safe to "downsize" instruction size due to
+			 *   pointer being treated as unsigned integer with
+			 *   zero-extended upper 32-bits;
+			 *   - reading unsigned integers, again due to
+			 *   zero-extension is preserving the value correctly.
+			 *
+			 * In all other cases it's incorrect to attempt to
+			 * load/store field because read value will be
+			 * incorrect, so we poison relocated instruction.
+			 */
+			if (btf_is_ptr(orig_t) && btf_is_ptr(new_t))
+				goto done;
+			if (btf_is_int(orig_t) && btf_is_int(new_t) &&
+			    btf_int_encoding(orig_t) != BTF_INT_SIGNED &&
+			    btf_int_encoding(new_t) != BTF_INT_SIGNED)
+				goto done;
+
+			/* mark as invalid mem size adjustment, but this will
+			 * only be checked for LDX/STX/ST insns
+			 */
+			res->fail_memsz_adjust = true;
+		}
+	} else if (core_relo_is_type_based(relo->kind)) {
+		err = bpf_core_calc_type_relo(relo, local_spec, &res->orig_val);
+		err = err ?: bpf_core_calc_type_relo(relo, targ_spec, &res->new_val);
+	} else if (core_relo_is_enumval_based(relo->kind)) {
+		err = bpf_core_calc_enumval_relo(relo, local_spec, &res->orig_val);
+		err = err ?: bpf_core_calc_enumval_relo(relo, targ_spec, &res->new_val);
+	}
+
+done:
+	if (err == -EUCLEAN) {
+		/* EUCLEAN is used to signal instruction poisoning request */
+		res->poison = true;
+		err = 0;
+	} else if (err == -EOPNOTSUPP) {
+		/* EOPNOTSUPP means unknown/unsupported relocation */
+		pr_warn("prog '%s': relo #%d: unrecognized CO-RE relocation %s (%d) at insn #%d\n",
+			prog->name, relo_idx, core_relo_kind_str(relo->kind),
+			relo->kind, relo->insn_off / 8);
+	}
+
+	return err;
+}
+
+/*
+ * Turn instruction for which CO_RE relocation failed into invalid one with
+ * distinct signature.
+ */
+static void bpf_core_poison_insn(struct bpf_program *prog, int relo_idx,
+				 int insn_idx, struct bpf_insn *insn)
+{
+	pr_debug("prog '%s': relo #%d: substituting insn #%d w/ invalid insn\n",
+		 prog->name, relo_idx, insn_idx);
+	insn->code = BPF_JMP | BPF_CALL;
+	insn->dst_reg = 0;
+	insn->src_reg = 0;
+	insn->off = 0;
+	/* if this instruction is reachable (not a dead code),
+	 * verifier will complain with the following message:
+	 * invalid func unknown#195896080
+	 */
+	insn->imm = 195896080; /* => 0xbad2310 => "bad relo" */
+}
+
+static bool is_ldimm64(struct bpf_insn *insn)
+{
+	return insn->code == (BPF_LD | BPF_IMM | BPF_DW);
+}
+
+static int insn_bpf_size_to_bytes(struct bpf_insn *insn)
+{
+	switch (BPF_SIZE(insn->code)) {
+	case BPF_DW: return 8;
+	case BPF_W: return 4;
+	case BPF_H: return 2;
+	case BPF_B: return 1;
+	default: return -1;
+	}
+}
+
+static int insn_bytes_to_bpf_size(__u32 sz)
+{
+	switch (sz) {
+	case 8: return BPF_DW;
+	case 4: return BPF_W;
+	case 2: return BPF_H;
+	case 1: return BPF_B;
+	default: return -1;
+	}
+}
+
 /*
  * Patch relocatable BPF instruction.
  *
  * Patched value is determined by relocation kind and target specification.
- * For field existence relocation target spec will be NULL if field is not
- * found.
+ * For existence relocations target spec will be NULL if field/type is not found.
  * Expected insn->imm value is determined using relocation kind and local
  * spec, and is checked before patching instruction. If actual insn->imm value
  * is wrong, bail out with error.
  *
- * Currently three kinds of BPF instructions are supported:
+ * Currently supported classes of BPF instruction are:
  * 1. rX = <imm> (assignment with immediate operand);
  * 2. rX += <imm> (arithmetic operations with immediate operand);
+ * 3. rX = <imm64> (load with 64-bit immediate value);
+ * 4. rX = *(T *)(rY + <off>), where T is one of {u8, u16, u32, u64};
+ * 5. *(T *)(rX + <off>) = rY, where T is one of {u8, u16, u32, u64};
+ * 6. *(T *)(rX + <off>) = <imm>, where T is one of {u8, u16, u32, u64}.
  */
-static int bpf_core_reloc_insn(struct bpf_program *prog,
-			       const struct bpf_field_reloc *relo,
+static int bpf_core_patch_insn(struct bpf_program *prog,
+			       const struct bpf_core_relo *relo,
 			       int relo_idx,
-			       const struct bpf_core_spec *local_spec,
-			       const struct bpf_core_spec *targ_spec)
+			       const struct bpf_core_relo_res *res)
 {
 	__u32 orig_val, new_val;
 	struct bpf_insn *insn;
-	bool validate = true;
-	int insn_idx, err;
+	int insn_idx;
 	__u8 class;
 
-	if (relo->insn_off % sizeof(struct bpf_insn))
+	if (relo->insn_off % BPF_INSN_SZ)
 		return -EINVAL;
-	insn_idx = relo->insn_off / sizeof(struct bpf_insn);
+	insn_idx = relo->insn_off / BPF_INSN_SZ;
+	/* adjust insn_idx from section frame of reference to the local
+	 * program's frame of reference; (sub-)program code is not yet
+	 * relocated, so it's enough to just subtract in-section offset
+	 */
+	insn_idx = insn_idx - prog->sec_insn_off;
 	insn = &prog->insns[insn_idx];
 	class = BPF_CLASS(insn->code);
 
-	if (relo->kind == BPF_FIELD_EXISTS) {
-		orig_val = 1; /* can't generate EXISTS relo w/o local field */
-		new_val = targ_spec ? 1 : 0;
-	} else if (!targ_spec) {
-		pr_debug("prog '%s': relo #%d: substituting insn #%d w/ invalid insn\n",
-			 bpf_program__title(prog, false), relo_idx, insn_idx);
-		insn->code = BPF_JMP | BPF_CALL;
-		insn->dst_reg = 0;
-		insn->src_reg = 0;
-		insn->off = 0;
-		/* if this instruction is reachable (not a dead code),
-		 * verifier will complain with the following message:
-		 * invalid func unknown#195896080
+	if (res->poison) {
+poison:
+		/* poison second part of ldimm64 to avoid confusing error from
+		 * verifier about "unknown opcode 00"
 		 */
-		insn->imm = 195896080; /* => 0xbad2310 => "bad relo" */
+		if (is_ldimm64(insn))
+			bpf_core_poison_insn(prog, relo_idx, insn_idx + 1, insn + 1);
+		bpf_core_poison_insn(prog, relo_idx, insn_idx, insn);
 		return 0;
-	} else {
-		err = bpf_core_calc_field_relo(prog, relo, local_spec,
-					       &orig_val, &validate);
-		if (err)
-			return err;
-		err = bpf_core_calc_field_relo(prog, relo, targ_spec,
-					       &new_val, NULL);
-		if (err)
-			return err;
 	}
 
+	orig_val = res->orig_val;
+	new_val = res->new_val;
+
 	switch (class) {
 	case BPF_ALU:
 	case BPF_ALU64:
 		if (BPF_SRC(insn->code) != BPF_K)
 			return -EINVAL;
-		if (validate && insn->imm != orig_val) {
+		if (res->validate && insn->imm != orig_val) {
 			pr_warn("prog '%s': relo #%d: unexpected insn #%d (ALU/ALU64) value: got %u, exp %u -> %u\n",
-				bpf_program__title(prog, false), relo_idx,
+				prog->name, relo_idx,
 				insn_idx, insn->imm, orig_val, new_val);
 			return -EINVAL;
 		}
 		orig_val = insn->imm;
 		insn->imm = new_val;
 		pr_debug("prog '%s': relo #%d: patched insn #%d (ALU/ALU64) imm %u -> %u\n",
-			 bpf_program__title(prog, false), relo_idx, insn_idx,
+			 prog->name, relo_idx, insn_idx,
 			 orig_val, new_val);
 		break;
 	case BPF_LDX:
 	case BPF_ST:
 	case BPF_STX:
-		if (validate && insn->off != orig_val) {
-			pr_warn("prog '%s': relo #%d: unexpected insn #%d (LD/LDX/ST/STX) value: got %u, exp %u -> %u\n",
-				bpf_program__title(prog, false), relo_idx,
-				insn_idx, insn->off, orig_val, new_val);
+		if (res->validate && insn->off != orig_val) {
+			pr_warn("prog '%s': relo #%d: unexpected insn #%d (LDX/ST/STX) value: got %u, exp %u -> %u\n",
+				prog->name, relo_idx, insn_idx, insn->off, orig_val, new_val);
 			return -EINVAL;
 		}
 		if (new_val > SHRT_MAX) {
 			pr_warn("prog '%s': relo #%d: insn #%d (LDX/ST/STX) value too big: %u\n",
-				bpf_program__title(prog, false), relo_idx,
-				insn_idx, new_val);
+				prog->name, relo_idx, insn_idx, new_val);
 			return -ERANGE;
 		}
+		if (res->fail_memsz_adjust) {
+			pr_warn("prog '%s': relo #%d: insn #%d (LDX/ST/STX) accesses field incorrectly. "
+				"Make sure you are accessing pointers, unsigned integers, or fields of matching type and size.\n",
+				prog->name, relo_idx, insn_idx);
+			goto poison;
+		}
+
 		orig_val = insn->off;
 		insn->off = new_val;
 		pr_debug("prog '%s': relo #%d: patched insn #%d (LDX/ST/STX) off %u -> %u\n",
-			 bpf_program__title(prog, false), relo_idx, insn_idx,
-			 orig_val, new_val);
+			 prog->name, relo_idx, insn_idx, orig_val, new_val);
+
+		if (res->new_sz != res->orig_sz) {
+			int insn_bytes_sz, insn_bpf_sz;
+
+			insn_bytes_sz = insn_bpf_size_to_bytes(insn);
+			if (insn_bytes_sz != res->orig_sz) {
+				pr_warn("prog '%s': relo #%d: insn #%d (LDX/ST/STX) unexpected mem size: got %d, exp %u\n",
+					prog->name, relo_idx, insn_idx, insn_bytes_sz, res->orig_sz);
+				return -EINVAL;
+			}
+
+			insn_bpf_sz = insn_bytes_to_bpf_size(res->new_sz);
+			if (insn_bpf_sz < 0) {
+				pr_warn("prog '%s': relo #%d: insn #%d (LDX/ST/STX) invalid new mem size: %u\n",
+					prog->name, relo_idx, insn_idx, res->new_sz);
+				return -EINVAL;
+			}
+
+			insn->code = BPF_MODE(insn->code) | insn_bpf_sz | BPF_CLASS(insn->code);
+			pr_debug("prog '%s': relo #%d: patched insn #%d (LDX/ST/STX) mem_sz %u -> %u\n",
+				 prog->name, relo_idx, insn_idx, res->orig_sz, res->new_sz);
+		}
 		break;
+	case BPF_LD: {
+		__u64 imm;
+
+		if (!is_ldimm64(insn) ||
+		    insn[0].src_reg != 0 || insn[0].off != 0 ||
+		    insn_idx + 1 >= prog->insns_cnt ||
+		    insn[1].code != 0 || insn[1].dst_reg != 0 ||
+		    insn[1].src_reg != 0 || insn[1].off != 0) {
+			pr_warn("prog '%s': relo #%d: insn #%d (LDIMM64) has unexpected form\n",
+				prog->name, relo_idx, insn_idx);
+			return -EINVAL;
+		}
+
+		imm = insn[0].imm + ((__u64)insn[1].imm << 32);
+		if (res->validate && imm != orig_val) {
+			pr_warn("prog '%s': relo #%d: unexpected insn #%d (LDIMM64) value: got %llu, exp %u -> %u\n",
+				prog->name, relo_idx,
+				insn_idx, (unsigned long long)imm,
+				orig_val, new_val);
+			return -EINVAL;
+		}
+
+		insn[0].imm = new_val;
+		insn[1].imm = 0; /* currently only 32-bit values are supported */
+		pr_debug("prog '%s': relo #%d: patched insn #%d (LDIMM64) imm64 %llu -> %u\n",
+			 prog->name, relo_idx, insn_idx,
+			 (unsigned long long)imm, new_val);
+		break;
+	}
 	default:
-		pr_warn("prog '%s': relo #%d: trying to relocate unrecognized insn #%d, code:%x, src:%x, dst:%x, off:%x, imm:%x\n",
-			bpf_program__title(prog, false), relo_idx,
-			insn_idx, insn->code, insn->src_reg, insn->dst_reg,
-			insn->off, insn->imm);
+		pr_warn("prog '%s': relo #%d: trying to relocate unrecognized insn #%d, code:0x%x, src:0x%x, dst:0x%x, off:0x%x, imm:0x%x\n",
+			prog->name, relo_idx, insn_idx, insn->code,
+			insn->src_reg, insn->dst_reg, insn->off, insn->imm);
 		return -EINVAL;
 	}
 
@@ -4728,29 +5548,48 @@ static int bpf_core_reloc_insn(struct bpf_program *prog,
 static void bpf_core_dump_spec(int level, const struct bpf_core_spec *spec)
 {
 	const struct btf_type *t;
+	const struct btf_enum *e;
 	const char *s;
 	__u32 type_id;
 	int i;
 
-	type_id = spec->spec[0].type_id;
+	type_id = spec->root_type_id;
 	t = btf__type_by_id(spec->btf, type_id);
 	s = btf__name_by_offset(spec->btf, t->name_off);
-	libbpf_print(level, "[%u] %s + ", type_id, s);
 
-	for (i = 0; i < spec->raw_len; i++)
-		libbpf_print(level, "%d%s", spec->raw_spec[i],
-			     i == spec->raw_len - 1 ? " => " : ":");
+	libbpf_print(level, "[%u] %s %s", type_id, btf_kind_str(t), str_is_empty(s) ? "<anon>" : s);
 
-	libbpf_print(level, "%u.%u @ &x",
-		     spec->bit_offset / 8, spec->bit_offset % 8);
+	if (core_relo_is_type_based(spec->relo_kind))
+		return;
 
-	for (i = 0; i < spec->len; i++) {
-		if (spec->spec[i].name)
-			libbpf_print(level, ".%s", spec->spec[i].name);
-		else
-			libbpf_print(level, "[%u]", spec->spec[i].idx);
+	if (core_relo_is_enumval_based(spec->relo_kind)) {
+		t = skip_mods_and_typedefs(spec->btf, type_id, NULL);
+		e = btf_enum(t) + spec->raw_spec[0];
+		s = btf__name_by_offset(spec->btf, e->name_off);
+
+		libbpf_print(level, "::%s = %u", s, e->val);
+		return;
 	}
 
+	if (core_relo_is_field_based(spec->relo_kind)) {
+		for (i = 0; i < spec->len; i++) {
+			if (spec->spec[i].name)
+				libbpf_print(level, ".%s", spec->spec[i].name);
+			else if (i > 0 || spec->spec[i].idx > 0)
+				libbpf_print(level, "[%u]", spec->spec[i].idx);
+		}
+
+		libbpf_print(level, " (");
+		for (i = 0; i < spec->raw_len; i++)
+			libbpf_print(level, "%s%d", i == 0 ? "" : ":", spec->raw_spec[i]);
+
+		if (spec->bit_offset % 8)
+			libbpf_print(level, " @ offset %u.%u)",
+				     spec->bit_offset / 8, spec->bit_offset % 8);
+		else
+			libbpf_print(level, " @ offset %u)", spec->bit_offset / 8);
+		return;
+	}
 }
 
 static size_t bpf_core_hash_fn(const void *key, void *ctx)
@@ -4814,22 +5653,22 @@ static void *u32_as_hash_key(__u32 x)
  *    CPU-wise compared to prebuilding a map from all local type names to
  *    a list of candidate type names. It's also sped up by caching resolved
  *    list of matching candidates per each local "root" type ID, that has at
- *    least one bpf_field_reloc associated with it. This list is shared
+ *    least one bpf_core_relo associated with it. This list is shared
  *    between multiple relocations for the same type ID and is updated as some
  *    of the candidates are pruned due to structural incompatibility.
  */
-static int bpf_core_reloc_field(struct bpf_program *prog,
-				 const struct bpf_field_reloc *relo,
-				 int relo_idx,
-				 const struct btf *local_btf,
-				 const struct btf *targ_btf,
-				 struct hashmap *cand_cache)
+static int bpf_core_apply_relo(struct bpf_program *prog,
+			       const struct bpf_core_relo *relo,
+			       int relo_idx,
+			       const struct btf *local_btf,
+			       const struct btf *targ_btf,
+			       struct hashmap *cand_cache)
 {
-	const char *prog_name = bpf_program__title(prog, false);
-	struct bpf_core_spec local_spec, cand_spec, targ_spec;
+	struct bpf_core_spec local_spec, cand_spec, targ_spec = {};
 	const void *type_key = u32_as_hash_key(relo->type_id);
-	const struct btf_type *local_type, *cand_type;
-	const char *local_name, *cand_name;
+	struct bpf_core_relo_res cand_res, targ_res;
+	const struct btf_type *local_type;
+	const char *local_name;
 	struct ids_vec *cand_ids;
 	__u32 local_id, cand_id;
 	const char *spec_str;
@@ -4841,32 +5680,49 @@ static int bpf_core_reloc_field(struct bpf_program *prog,
 		return -EINVAL;
 
 	local_name = btf__name_by_offset(local_btf, local_type->name_off);
-	if (str_is_empty(local_name))
+	if (!local_name)
 		return -EINVAL;
 
 	spec_str = btf__name_by_offset(local_btf, relo->access_str_off);
 	if (str_is_empty(spec_str))
 		return -EINVAL;
 
-	err = bpf_core_spec_parse(local_btf, local_id, spec_str, &local_spec);
+	err = bpf_core_parse_spec(local_btf, local_id, spec_str, relo->kind, &local_spec);
 	if (err) {
-		pr_warn("prog '%s': relo #%d: parsing [%d] %s + %s failed: %d\n",
-			prog_name, relo_idx, local_id, local_name, spec_str,
-			err);
+		pr_warn("prog '%s': relo #%d: parsing [%d] %s %s + %s failed: %d\n",
+			prog->name, relo_idx, local_id, btf_kind_str(local_type),
+			str_is_empty(local_name) ? "<anon>" : local_name,
+			spec_str, err);
 		return -EINVAL;
 	}
 
-	pr_debug("prog '%s': relo #%d: kind %d, spec is ", prog_name, relo_idx,
-		 relo->kind);
+	pr_debug("prog '%s': relo #%d: kind <%s> (%d), spec is ", prog->name,
+		 relo_idx, core_relo_kind_str(relo->kind), relo->kind);
 	bpf_core_dump_spec(LIBBPF_DEBUG, &local_spec);
 	libbpf_print(LIBBPF_DEBUG, "\n");
 
+	/* TYPE_ID_LOCAL relo is special and doesn't need candidate search */
+	if (relo->kind == BPF_TYPE_ID_LOCAL) {
+		targ_res.validate = true;
+		targ_res.poison = false;
+		targ_res.orig_val = local_spec.root_type_id;
+		targ_res.new_val = local_spec.root_type_id;
+		goto patch_insn;
+	}
+
+	/* libbpf doesn't support candidate search for anonymous types */
+	if (str_is_empty(spec_str)) {
+		pr_warn("prog '%s': relo #%d: <%s> (%d) relocation doesn't support anonymous types\n",
+			prog->name, relo_idx, core_relo_kind_str(relo->kind), relo->kind);
+		return -EOPNOTSUPP;
+	}
+
 	if (!hashmap__find(cand_cache, type_key, (void **)&cand_ids)) {
 		cand_ids = bpf_core_find_cands(local_btf, local_id, targ_btf);
 		if (IS_ERR(cand_ids)) {
-			pr_warn("prog '%s': relo #%d: target candidate search failed for [%d] %s: %ld",
-				prog_name, relo_idx, local_id, local_name,
-				PTR_ERR(cand_ids));
+			pr_warn("prog '%s': relo #%d: target candidate search failed for [%d] %s %s: %ld",
+				prog->name, relo_idx, local_id, btf_kind_str(local_type),
+				local_name, PTR_ERR(cand_ids));
 			return PTR_ERR(cand_ids);
 		}
 		err = hashmap__set(cand_cache, type_key, cand_ids, NULL, NULL);
@@ -4878,36 +5734,51 @@ static int bpf_core_reloc_field(struct bpf_program *prog,
 
 	for (i = 0, j = 0; i < cand_ids->len; i++) {
 		cand_id = cand_ids->data[i];
-		cand_type = btf__type_by_id(targ_btf, cand_id);
-		cand_name = btf__name_by_offset(targ_btf, cand_type->name_off);
-
-		err = bpf_core_spec_match(&local_spec, targ_btf,
-					  cand_id, &cand_spec);
-		pr_debug("prog '%s': relo #%d: matching candidate #%d %s against spec ",
-			 prog_name, relo_idx, i, cand_name);
-		bpf_core_dump_spec(LIBBPF_DEBUG, &cand_spec);
-		libbpf_print(LIBBPF_DEBUG, ": %d\n", err);
+		err = bpf_core_spec_match(&local_spec, targ_btf, cand_id, &cand_spec);
 		if (err < 0) {
-			pr_warn("prog '%s': relo #%d: matching error: %d\n",
-				prog_name, relo_idx, err);
+			pr_warn("prog '%s': relo #%d: error matching candidate #%d ",
+				prog->name, relo_idx, i);
+			bpf_core_dump_spec(LIBBPF_WARN, &cand_spec);
+			libbpf_print(LIBBPF_WARN, ": %d\n", err);
 			return err;
 		}
+
+		pr_debug("prog '%s': relo #%d: %s candidate #%d ", prog->name,
+			 relo_idx, err == 0 ? "non-matching" : "matching", i);
+		bpf_core_dump_spec(LIBBPF_DEBUG, &cand_spec);
+		libbpf_print(LIBBPF_DEBUG, "\n");
+
 		if (err == 0)
 			continue;
 
+		err = bpf_core_calc_relo(prog, relo, relo_idx, &local_spec, &cand_spec, &cand_res);
+		if (err)
+			return err;
+
 		if (j == 0) {
+			targ_res = cand_res;
 			targ_spec = cand_spec;
 		} else if (cand_spec.bit_offset != targ_spec.bit_offset) {
-			/* if there are many candidates, they should all
-			 * resolve to the same bit offset
+			/* if there are many field relo candidates, they
+			 * should all resolve to the same bit offset
 			 */
-			pr_warn("prog '%s': relo #%d: offset ambiguity: %u != %u\n",
-				prog_name, relo_idx, cand_spec.bit_offset,
+			pr_warn("prog '%s': relo #%d: field offset ambiguity: %u != %u\n",
+				prog->name, relo_idx, cand_spec.bit_offset,
 				targ_spec.bit_offset);
 			return -EINVAL;
+		} else if (cand_res.poison != targ_res.poison || cand_res.new_val != targ_res.new_val) {
+			/* all candidates should result in the same relocation
+			 * decision and value, otherwise it's dangerous to
+			 * proceed due to ambiguity
+			 */
+			pr_warn("prog '%s': relo #%d: relocation decision ambiguity: %s %u != %s %u\n",
+				prog->name, relo_idx,
+				cand_res.poison ? "failure" : "success", cand_res.new_val,
+				targ_res.poison ? "failure" : "success", targ_res.new_val);
+			return -EINVAL;
 		}
 
-		cand_ids->data[j++] = cand_spec.spec[0].type_id;
+		cand_ids->data[j++] = cand_spec.root_type_id;
 	}
 
 	/*
@@ -4926,22 +5797,28 @@ static int bpf_core_reloc_field(struct bpf_program *prog,
 	 * as well as expected case, depending whether instruction w/
 	 * relocation is guarded in some way that makes it unreachable (dead
 	 * code) if relocation can't be resolved. This is handled in
-	 * bpf_core_reloc_insn() uniformly by replacing that instruction with
+	 * bpf_core_patch_insn() uniformly by replacing that instruction with
 	 * BPF helper call insn (using invalid helper ID). If that instruction
 	 * is indeed unreachable, then it will be ignored and eliminated by
 	 * verifier. If it was an error, then verifier will complain and point
 	 * to a specific instruction number in its log.
 	 */
-	if (j == 0)
-		pr_debug("prog '%s': relo #%d: no matching targets found for [%d] %s + %s\n",
-			 prog_name, relo_idx, local_id, local_name, spec_str);
+	if (j == 0) {
+		pr_debug("prog '%s': relo #%d: no matching targets found\n",
+			 prog->name, relo_idx);
 
-	/* bpf_core_reloc_insn should know how to handle missing targ_spec */
-	err = bpf_core_reloc_insn(prog, relo, relo_idx, &local_spec,
-				  j ? &targ_spec : NULL);
+		/* calculate single target relo result explicitly */
+		err = bpf_core_calc_relo(prog, relo, relo_idx, &local_spec, NULL, &targ_res);
+		if (err)
+			return err;
+	}
+
+patch_insn:
+	/* bpf_core_patch_insn() should know how to handle missing targ_spec */
+	err = bpf_core_patch_insn(prog, relo, relo_idx, &targ_res);
 	if (err) {
 		pr_warn("prog '%s': relo #%d: failed to patch insn at offset %d: %d\n",
-			prog_name, relo_idx, relo->insn_off, err);
+			prog->name, relo_idx, relo->insn_off, err);
 		return -EINVAL;
 	}
 
@@ -4949,20 +5826,23 @@ static int bpf_core_reloc_field(struct bpf_program *prog,
 }
 
 static int
-bpf_core_reloc_fields(struct bpf_object *obj, const char *targ_btf_path)
+bpf_object__relocate_core(struct bpf_object *obj, const char *targ_btf_path)
 {
 	const struct btf_ext_info_sec *sec;
-	const struct bpf_field_reloc *rec;
+	const struct bpf_core_relo *rec;
 	const struct btf_ext_info *seg;
 	struct hashmap_entry *entry;
 	struct hashmap *cand_cache = NULL;
 	struct bpf_program *prog;
 	struct btf *targ_btf;
 	const char *sec_name;
-	int i, err = 0;
+	int i, err = 0, insn_idx, sec_idx;
+
+	if (obj->btf_ext->core_relo_info.len == 0)
+		return 0;
 
 	if (targ_btf_path)
-		targ_btf = btf__parse_elf(targ_btf_path, NULL);
+		targ_btf = btf__parse(targ_btf_path, NULL);
 	else
 		targ_btf = obj->btf_vmlinux;
 	if (IS_ERR_OR_NULL(targ_btf)) {
@@ -4976,36 +5856,54 @@ bpf_core_reloc_fields(struct bpf_object *obj, const char *targ_btf_path)
 		goto out;
 	}
 
-	seg = &obj->btf_ext->field_reloc_info;
+	seg = &obj->btf_ext->core_relo_info;
 	for_each_btf_ext_sec(seg, sec) {
 		sec_name = btf__name_by_offset(obj->btf, sec->sec_name_off);
 		if (str_is_empty(sec_name)) {
 			err = -EINVAL;
 			goto out;
 		}
+		/* bpf_object's ELF is gone by now so it's not easy to find
+		 * section index by section name, but we can find *any*
+		 * bpf_program within desired section name and use it's
+		 * prog->sec_idx to do a proper search by section index and
+		 * instruction offset
+		 */
 		prog = NULL;
 		for (i = 0; i < obj->nr_programs; i++) {
-			if (!strcmp(obj->programs[i].section_name, sec_name)) {
-				prog = &obj->programs[i];
+			prog = &obj->programs[i];
+			if (strcmp(prog->sec_name, sec_name) == 0)
 				break;
-			}
 		}
 		if (!prog) {
-			pr_warn("failed to find program '%s' for CO-RE offset relocation\n",
-				sec_name);
-			err = -EINVAL;
-			goto out;
+			pr_warn("sec '%s': failed to find a BPF program\n", sec_name);
+			return -ENOENT;
 		}
+		sec_idx = prog->sec_idx;
 
-		pr_debug("prog '%s': performing %d CO-RE offset relocs\n",
+		pr_debug("sec '%s': found %d CO-RE relocations\n",
 			 sec_name, sec->num_info);
 
 		for_each_btf_ext_rec(seg, sec, i, rec) {
-			err = bpf_core_reloc_field(prog, rec, i, obj->btf,
-						   targ_btf, cand_cache);
+			insn_idx = rec->insn_off / BPF_INSN_SZ;
+			prog = find_prog_by_sec_insn(obj, sec_idx, insn_idx);
+			if (!prog) {
+				pr_warn("sec '%s': failed to find program at insn #%d for CO-RE offset relocation #%d\n",
+					sec_name, insn_idx, i);
+				err = -EINVAL;
+				goto out;
+			}
+			/* no need to apply CO-RE relocation if the program is
+			 * not going to be loaded
+			 */
+			if (!prog->load)
+				continue;
+
+			err = bpf_core_apply_relo(prog, rec, i, obj->btf,
+						  targ_btf, cand_cache);
 			if (err) {
 				pr_warn("prog '%s': relo #%d: failed to relocate: %d\n",
-					sec_name, i, err);
+					prog->name, i, err);
 				goto out;
 			}
 		}
@@ -5024,125 +5922,432 @@ out:
 	return err;
 }
 
+/* Relocate data references within program code:
+ *  - map references;
+ *  - global variable references;
+ *  - extern references.
+ */
 static int
-bpf_object__relocate_core(struct bpf_object *obj, const char *targ_btf_path)
+bpf_object__relocate_data(struct bpf_object *obj, struct bpf_program *prog)
 {
-	int err = 0;
+	int i;
 
-	if (obj->btf_ext->field_reloc_info.len)
-		err = bpf_core_reloc_fields(obj, targ_btf_path);
+	for (i = 0; i < prog->nr_reloc; i++) {
+		struct reloc_desc *relo = &prog->reloc_desc[i];
+		struct bpf_insn *insn = &prog->insns[relo->insn_idx];
+		struct extern_desc *ext;
 
-	return err;
+		switch (relo->type) {
+		case RELO_LD64:
+			insn[0].src_reg = BPF_PSEUDO_MAP_FD;
+			insn[0].imm = obj->maps[relo->map_idx].fd;
+			relo->processed = true;
+			break;
+		case RELO_DATA:
+			insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
+			insn[1].imm = insn[0].imm + relo->sym_off;
+			insn[0].imm = obj->maps[relo->map_idx].fd;
+			relo->processed = true;
+			break;
+		case RELO_EXTERN:
+			ext = &obj->externs[relo->sym_off];
+			if (ext->type == EXT_KCFG) {
+				insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
+				insn[0].imm = obj->maps[obj->kconfig_map_idx].fd;
+				insn[1].imm = ext->kcfg.data_off;
+			} else /* EXT_KSYM */ {
+				if (ext->ksym.type_id) { /* typed ksyms */
+					insn[0].src_reg = BPF_PSEUDO_BTF_ID;
+					insn[0].imm = ext->ksym.vmlinux_btf_id;
+				} else { /* typeless ksyms */
+					insn[0].imm = (__u32)ext->ksym.addr;
+					insn[1].imm = ext->ksym.addr >> 32;
+				}
+			}
+			relo->processed = true;
+			break;
+		case RELO_CALL:
+			/* will be handled as a follow up pass */
+			break;
+		default:
+			pr_warn("prog '%s': relo #%d: bad relo type %d\n",
+				prog->name, i, relo->type);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
 }
 
-static int
-bpf_program__reloc_text(struct bpf_program *prog, struct bpf_object *obj,
-			struct reloc_desc *relo)
+static int adjust_prog_btf_ext_info(const struct bpf_object *obj,
+				    const struct bpf_program *prog,
+				    const struct btf_ext_info *ext_info,
+				    void **prog_info, __u32 *prog_rec_cnt,
+				    __u32 *prog_rec_sz)
 {
-	struct bpf_insn *insn, *new_insn;
-	struct bpf_program *text;
-	size_t new_cnt;
-	int err;
+	void *copy_start = NULL, *copy_end = NULL;
+	void *rec, *rec_end, *new_prog_info;
+	const struct btf_ext_info_sec *sec;
+	size_t old_sz, new_sz;
+	const char *sec_name;
+	int i, off_adj;
 
-	if (prog->idx != obj->efile.text_shndx && prog->main_prog_cnt == 0) {
-		text = bpf_object__find_prog_by_idx(obj, obj->efile.text_shndx);
-		if (!text) {
-			pr_warn("no .text section found yet relo into text exist\n");
-			return -LIBBPF_ERRNO__RELOC;
+	for_each_btf_ext_sec(ext_info, sec) {
+		sec_name = btf__name_by_offset(obj->btf, sec->sec_name_off);
+		if (!sec_name)
+			return -EINVAL;
+		if (strcmp(sec_name, prog->sec_name) != 0)
+			continue;
+
+		for_each_btf_ext_rec(ext_info, sec, i, rec) {
+			__u32 insn_off = *(__u32 *)rec / BPF_INSN_SZ;
+
+			if (insn_off < prog->sec_insn_off)
+				continue;
+			if (insn_off >= prog->sec_insn_off + prog->sec_insn_cnt)
+				break;
+
+			if (!copy_start)
+				copy_start = rec;
+			copy_end = rec + ext_info->rec_size;
 		}
-		new_cnt = prog->insns_cnt + text->insns_cnt;
-		new_insn = reallocarray(prog->insns, new_cnt, sizeof(*insn));
-		if (!new_insn) {
-			pr_warn("oom in prog realloc\n");
+
+		if (!copy_start)
+			return -ENOENT;
+
+		/* append func/line info of a given (sub-)program to the main
+		 * program func/line info
+		 */
+		old_sz = (size_t)(*prog_rec_cnt) * ext_info->rec_size;
+		new_sz = old_sz + (copy_end - copy_start);
+		new_prog_info = realloc(*prog_info, new_sz);
+		if (!new_prog_info)
 			return -ENOMEM;
-		}
-		prog->insns = new_insn;
+		*prog_info = new_prog_info;
+		*prog_rec_cnt = new_sz / ext_info->rec_size;
+		memcpy(new_prog_info + old_sz, copy_start, copy_end - copy_start);
 
-		if (obj->btf_ext) {
-			err = bpf_program_reloc_btf_ext(prog, obj,
-							text->section_name,
-							prog->insns_cnt);
-			if (err)
-				return err;
-		}
+		/* Kernel instruction offsets are in units of 8-byte
+		 * instructions, while .BTF.ext instruction offsets generated
+		 * by Clang are in units of bytes. So convert Clang offsets
+		 * into kernel offsets and adjust offset according to program
+		 * relocated position.
+		 */
+		off_adj = prog->sub_insn_off - prog->sec_insn_off;
+		rec = new_prog_info + old_sz;
+		rec_end = new_prog_info + new_sz;
+		for (; rec < rec_end; rec += ext_info->rec_size) {
+			__u32 *insn_off = rec;
 
-		memcpy(new_insn + prog->insns_cnt, text->insns,
-		       text->insns_cnt * sizeof(*insn));
-		prog->main_prog_cnt = prog->insns_cnt;
-		prog->insns_cnt = new_cnt;
-		pr_debug("added %zd insn from %s to prog %s\n",
-			 text->insns_cnt, text->section_name,
-			 prog->section_name);
+			*insn_off = *insn_off / BPF_INSN_SZ + off_adj;
+		}
+		*prog_rec_sz = ext_info->rec_size;
+		return 0;
 	}
 
-	insn = &prog->insns[relo->insn_idx];
-	insn->imm += relo->sym_off / 8 + prog->main_prog_cnt - relo->insn_idx;
-	return 0;
+	return -ENOENT;
 }
 
 static int
-bpf_program__relocate(struct bpf_program *prog, struct bpf_object *obj)
+reloc_prog_func_and_line_info(const struct bpf_object *obj,
+			      struct bpf_program *main_prog,
+			      const struct bpf_program *prog)
 {
-	int i, err;
+	int err;
 
-	if (!prog)
+	/* no .BTF.ext relocation if .BTF.ext is missing or kernel doesn't
+	 * supprot func/line info
+	 */
+	if (!obj->btf_ext || !kernel_supports(FEAT_BTF_FUNC))
 		return 0;
 
-	if (obj->btf_ext) {
-		err = bpf_program_reloc_btf_ext(prog, obj,
-						prog->section_name, 0);
-		if (err)
+	/* only attempt func info relocation if main program's func_info
+	 * relocation was successful
+	 */
+	if (main_prog != prog && !main_prog->func_info)
+		goto line_info;
+
+	err = adjust_prog_btf_ext_info(obj, prog, &obj->btf_ext->func_info,
+				       &main_prog->func_info,
+				       &main_prog->func_info_cnt,
+				       &main_prog->func_info_rec_size);
+	if (err) {
+		if (err != -ENOENT) {
+			pr_warn("prog '%s': error relocating .BTF.ext function info: %d\n",
+				prog->name, err);
+			return err;
+		}
+		if (main_prog->func_info) {
+			/*
+			 * Some info has already been found but has problem
+			 * in the last btf_ext reloc. Must have to error out.
+			 */
+			pr_warn("prog '%s': missing .BTF.ext function info.\n", prog->name);
 			return err;
+		}
+		/* Have problem loading the very first info. Ignore the rest. */
+		pr_warn("prog '%s': missing .BTF.ext function info for the main program, skipping all of .BTF.ext func info.\n",
+			prog->name);
 	}
 
-	if (!prog->reloc_desc)
+line_info:
+	/* don't relocate line info if main program's relocation failed */
+	if (main_prog != prog && !main_prog->line_info)
 		return 0;
 
-	for (i = 0; i < prog->nr_reloc; i++) {
-		struct reloc_desc *relo = &prog->reloc_desc[i];
-		struct bpf_insn *insn = &prog->insns[relo->insn_idx];
-		struct extern_desc *ext;
+	err = adjust_prog_btf_ext_info(obj, prog, &obj->btf_ext->line_info,
+				       &main_prog->line_info,
+				       &main_prog->line_info_cnt,
+				       &main_prog->line_info_rec_size);
+	if (err) {
+		if (err != -ENOENT) {
+			pr_warn("prog '%s': error relocating .BTF.ext line info: %d\n",
+				prog->name, err);
+			return err;
+		}
+		if (main_prog->line_info) {
+			/*
+			 * Some info has already been found but has problem
+			 * in the last btf_ext reloc. Must have to error out.
+			 */
+			pr_warn("prog '%s': missing .BTF.ext line info.\n", prog->name);
+			return err;
+		}
+		/* Have problem loading the very first info. Ignore the rest. */
+		pr_warn("prog '%s': missing .BTF.ext line info for the main program, skipping all of .BTF.ext line info.\n",
+			prog->name);
+	}
+	return 0;
+}
+
+static int cmp_relo_by_insn_idx(const void *key, const void *elem)
+{
+	size_t insn_idx = *(const size_t *)key;
+	const struct reloc_desc *relo = elem;
 
-		if (relo->insn_idx + 1 >= (int)prog->insns_cnt) {
-			pr_warn("relocation out of range: '%s'\n",
-				prog->section_name);
+	if (insn_idx == relo->insn_idx)
+		return 0;
+	return insn_idx < relo->insn_idx ? -1 : 1;
+}
+
+static struct reloc_desc *find_prog_insn_relo(const struct bpf_program *prog, size_t insn_idx)
+{
+	return bsearch(&insn_idx, prog->reloc_desc, prog->nr_reloc,
+		       sizeof(*prog->reloc_desc), cmp_relo_by_insn_idx);
+}
+
+static int
+bpf_object__reloc_code(struct bpf_object *obj, struct bpf_program *main_prog,
+		       struct bpf_program *prog)
+{
+	size_t sub_insn_idx, insn_idx, new_cnt;
+	struct bpf_program *subprog;
+	struct bpf_insn *insns, *insn;
+	struct reloc_desc *relo;
+	int err;
+
+	err = reloc_prog_func_and_line_info(obj, main_prog, prog);
+	if (err)
+		return err;
+
+	for (insn_idx = 0; insn_idx < prog->sec_insn_cnt; insn_idx++) {
+		insn = &main_prog->insns[prog->sub_insn_off + insn_idx];
+		if (!insn_is_subprog_call(insn))
+			continue;
+
+		relo = find_prog_insn_relo(prog, insn_idx);
+		if (relo && relo->type != RELO_CALL) {
+			pr_warn("prog '%s': unexpected relo for insn #%zu, type %d\n",
+				prog->name, insn_idx, relo->type);
 			return -LIBBPF_ERRNO__RELOC;
 		}
+		if (relo) {
+			/* sub-program instruction index is a combination of
+			 * an offset of a symbol pointed to by relocation and
+			 * call instruction's imm field; for global functions,
+			 * call always has imm = -1, but for static functions
+			 * relocation is against STT_SECTION and insn->imm
+			 * points to a start of a static function
+			 */
+			sub_insn_idx = relo->sym_off / BPF_INSN_SZ + insn->imm + 1;
+		} else {
+			/* if subprogram call is to a static function within
+			 * the same ELF section, there won't be any relocation
+			 * emitted, but it also means there is no additional
+			 * offset necessary, insns->imm is relative to
+			 * instruction's original position within the section
+			 */
+			sub_insn_idx = prog->sec_insn_off + insn_idx + insn->imm + 1;
+		}
 
-		switch (relo->type) {
-		case RELO_LD64:
-			insn[0].src_reg = BPF_PSEUDO_MAP_FD;
-			insn[0].imm = obj->maps[relo->map_idx].fd;
-			break;
-		case RELO_DATA:
-			insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
-			insn[1].imm = insn[0].imm + relo->sym_off;
-			insn[0].imm = obj->maps[relo->map_idx].fd;
-			break;
-		case RELO_EXTERN:
-			ext = &obj->externs[relo->sym_off];
-			if (ext->type == EXT_KCFG) {
-				insn[0].src_reg = BPF_PSEUDO_MAP_VALUE;
-				insn[0].imm = obj->maps[obj->kconfig_map_idx].fd;
-				insn[1].imm = ext->kcfg.data_off;
-			} else /* EXT_KSYM */ {
-				insn[0].imm = (__u32)ext->ksym.addr;
-				insn[1].imm = ext->ksym.addr >> 32;
+		/* we enforce that sub-programs should be in .text section */
+		subprog = find_prog_by_sec_insn(obj, obj->efile.text_shndx, sub_insn_idx);
+		if (!subprog) {
+			pr_warn("prog '%s': no .text section found yet sub-program call exists\n",
+				prog->name);
+			return -LIBBPF_ERRNO__RELOC;
+		}
+
+		/* if it's the first call instruction calling into this
+		 * subprogram (meaning this subprog hasn't been processed
+		 * yet) within the context of current main program:
+		 *   - append it at the end of main program's instructions blog;
+		 *   - process is recursively, while current program is put on hold;
+		 *   - if that subprogram calls some other not yet processes
+		 *   subprogram, same thing will happen recursively until
+		 *   there are no more unprocesses subprograms left to append
+		 *   and relocate.
+		 */
+		if (subprog->sub_insn_off == 0) {
+			subprog->sub_insn_off = main_prog->insns_cnt;
+
+			new_cnt = main_prog->insns_cnt + subprog->insns_cnt;
+			insns = libbpf_reallocarray(main_prog->insns, new_cnt, sizeof(*insns));
+			if (!insns) {
+				pr_warn("prog '%s': failed to realloc prog code\n", main_prog->name);
+				return -ENOMEM;
 			}
-			break;
-		case RELO_CALL:
-			err = bpf_program__reloc_text(prog, obj, relo);
+			main_prog->insns = insns;
+			main_prog->insns_cnt = new_cnt;
+
+			memcpy(main_prog->insns + subprog->sub_insn_off, subprog->insns,
+			       subprog->insns_cnt * sizeof(*insns));
+
+			pr_debug("prog '%s': added %zu insns from sub-prog '%s'\n",
+				 main_prog->name, subprog->insns_cnt, subprog->name);
+
+			err = bpf_object__reloc_code(obj, main_prog, subprog);
 			if (err)
 				return err;
-			break;
-		default:
-			pr_warn("relo #%d: bad relo type %d\n", i, relo->type);
-			return -EINVAL;
 		}
+
+		/* main_prog->insns memory could have been re-allocated, so
+		 * calculate pointer again
+		 */
+		insn = &main_prog->insns[prog->sub_insn_off + insn_idx];
+		/* calculate correct instruction position within current main
+		 * prog; each main prog can have a different set of
+		 * subprograms appended (potentially in different order as
+		 * well), so position of any subprog can be different for
+		 * different main programs */
+		insn->imm = subprog->sub_insn_off - (prog->sub_insn_off + insn_idx) - 1;
+
+		if (relo)
+			relo->processed = true;
+
+		pr_debug("prog '%s': insn #%zu relocated, imm %d points to subprog '%s' (now at %zu offset)\n",
+			 prog->name, insn_idx, insn->imm, subprog->name, subprog->sub_insn_off);
 	}
 
-	zfree(&prog->reloc_desc);
-	prog->nr_reloc = 0;
+	return 0;
+}
+
+/*
+ * Relocate sub-program calls.
+ *
+ * Algorithm operates as follows. Each entry-point BPF program (referred to as
+ * main prog) is processed separately. For each subprog (non-entry functions,
+ * that can be called from either entry progs or other subprogs) gets their
+ * sub_insn_off reset to zero. This serves as indicator that this subprogram
+ * hasn't been yet appended and relocated within current main prog. Once its
+ * relocated, sub_insn_off will point at the position within current main prog
+ * where given subprog was appended. This will further be used to relocate all
+ * the call instructions jumping into this subprog.
+ *
+ * We start with main program and process all call instructions. If the call
+ * is into a subprog that hasn't been processed (i.e., subprog->sub_insn_off
+ * is zero), subprog instructions are appended at the end of main program's
+ * instruction array. Then main program is "put on hold" while we recursively
+ * process newly appended subprogram. If that subprogram calls into another
+ * subprogram that hasn't been appended, new subprogram is appended again to
+ * the *main* prog's instructions (subprog's instructions are always left
+ * untouched, as they need to be in unmodified state for subsequent main progs
+ * and subprog instructions are always sent only as part of a main prog) and
+ * the process continues recursively. Once all the subprogs called from a main
+ * prog or any of its subprogs are appended (and relocated), all their
+ * positions within finalized instructions array are known, so it's easy to
+ * rewrite call instructions with correct relative offsets, corresponding to
+ * desired target subprog.
+ *
+ * Its important to realize that some subprogs might not be called from some
+ * main prog and any of its called/used subprogs. Those will keep their
+ * subprog->sub_insn_off as zero at all times and won't be appended to current
+ * main prog and won't be relocated within the context of current main prog.
+ * They might still be used from other main progs later.
+ *
+ * Visually this process can be shown as below. Suppose we have two main
+ * programs mainA and mainB and BPF object contains three subprogs: subA,
+ * subB, and subC. mainA calls only subA, mainB calls only subC, but subA and
+ * subC both call subB:
+ *
+ *        +--------+ +-------+
+ *        |        v v       |
+ *     +--+---+ +--+-+-+ +---+--+
+ *     | subA | | subB | | subC |
+ *     +--+---+ +------+ +---+--+
+ *        ^                  ^
+ *        |                  |
+ *    +---+-------+   +------+----+
+ *    |   mainA   |   |   mainB   |
+ *    +-----------+   +-----------+
+ *
+ * We'll start relocating mainA, will find subA, append it and start
+ * processing sub A recursively:
+ *
+ *    +-----------+------+
+ *    |   mainA   | subA |
+ *    +-----------+------+
+ *
+ * At this point we notice that subB is used from subA, so we append it and
+ * relocate (there are no further subcalls from subB):
+ *
+ *    +-----------+------+------+
+ *    |   mainA   | subA | subB |
+ *    +-----------+------+------+
+ *
+ * At this point, we relocate subA calls, then go one level up and finish with
+ * relocatin mainA calls. mainA is done.
+ *
+ * For mainB process is similar but results in different order. We start with
+ * mainB and skip subA and subB, as mainB never calls them (at least
+ * directly), but we see subC is needed, so we append and start processing it:
+ *
+ *    +-----------+------+
+ *    |   mainB   | subC |
+ *    +-----------+------+
+ * Now we see subC needs subB, so we go back to it, append and relocate it:
+ *
+ *    +-----------+------+------+
+ *    |   mainB   | subC | subB |
+ *    +-----------+------+------+
+ *
+ * At this point we unwind recursion, relocate calls in subC, then in mainB.
+ */
+static int
+bpf_object__relocate_calls(struct bpf_object *obj, struct bpf_program *prog)
+{
+	struct bpf_program *subprog;
+	int i, j, err;
+
+	/* mark all subprogs as not relocated (yet) within the context of
+	 * current main program
+	 */
+	for (i = 0; i < obj->nr_programs; i++) {
+		subprog = &obj->programs[i];
+		if (!prog_is_subprog(obj, subprog))
+			continue;
+
+		subprog->sub_insn_off = 0;
+		for (j = 0; j < subprog->nr_reloc; j++)
+			if (subprog->reloc_desc[j].type == RELO_CALL)
+				subprog->reloc_desc[j].processed = false;
+	}
+
+	err = bpf_object__reloc_code(obj, prog, prog);
+	if (err)
+		return err;
+
+
 	return 0;
 }
 
@@ -5161,35 +6366,45 @@ bpf_object__relocate(struct bpf_object *obj, const char *targ_btf_path)
 			return err;
 		}
 	}
-	/* ensure .text is relocated first, as it's going to be copied as-is
-	 * later for sub-program calls
+	/* relocate data references first for all programs and sub-programs,
+	 * as they don't change relative to code locations, so subsequent
+	 * subprogram processing won't need to re-calculate any of them
 	 */
 	for (i = 0; i < obj->nr_programs; i++) {
 		prog = &obj->programs[i];
-		if (prog->idx != obj->efile.text_shndx)
-			continue;
-
-		err = bpf_program__relocate(prog, obj);
+		err = bpf_object__relocate_data(obj, prog);
 		if (err) {
-			pr_warn("failed to relocate '%s'\n", prog->section_name);
+			pr_warn("prog '%s': failed to relocate data references: %d\n",
+				prog->name, err);
 			return err;
 		}
-		break;
 	}
-	/* now relocate everything but .text, which by now is relocated
-	 * properly, so we can copy raw sub-program instructions as is safely
+	/* now relocate subprogram calls and append used subprograms to main
+	 * programs; each copy of subprogram code needs to be relocated
+	 * differently for each main program, because its code location might
+	 * have changed
 	 */
 	for (i = 0; i < obj->nr_programs; i++) {
 		prog = &obj->programs[i];
-		if (prog->idx == obj->efile.text_shndx)
+		/* sub-program's sub-calls are relocated within the context of
+		 * its main program only
+		 */
+		if (prog_is_subprog(obj, prog))
 			continue;
 
-		err = bpf_program__relocate(prog, obj);
+		err = bpf_object__relocate_calls(obj, prog);
 		if (err) {
-			pr_warn("failed to relocate '%s'\n", prog->section_name);
+			pr_warn("prog '%s': failed to relocate calls: %d\n",
+				prog->name, err);
 			return err;
 		}
 	}
+	/* free up relocation descriptors */
+	for (i = 0; i < obj->nr_programs; i++) {
+		prog = &obj->programs[i];
+		zfree(&prog->reloc_desc);
+		prog->nr_reloc = 0;
+	}
 	return 0;
 }
 
@@ -5230,8 +6445,7 @@ static int bpf_object__collect_map_relos(struct bpf_object *obj,
 				i, (size_t)GELF_R_SYM(rel.r_info));
 			return -LIBBPF_ERRNO__FORMAT;
 		}
-		name = elf_strptr(obj->efile.elf, obj->efile.strtabidx,
-				  sym.st_name) ? : "<?>";
+		name = elf_sym_str(obj, sym.st_name) ?: "<?>";
 		if (sym.st_shndx != obj->efile.btf_maps_shndx) {
 			pr_warn(".maps relo #%d: '%s' isn't a BTF-defined map\n",
 				i, name);
@@ -5293,7 +6507,7 @@ static int bpf_object__collect_map_relos(struct bpf_object *obj,
 		moff /= bpf_ptr_sz;
 		if (moff >= map->init_slots_sz) {
 			new_sz = moff + 1;
-			tmp = realloc(map->init_slots, new_sz * host_ptr_sz);
+			tmp = libbpf_reallocarray(map->init_slots, new_sz, host_ptr_sz);
 			if (!tmp)
 				return -ENOMEM;
 			map->init_slots = tmp;
@@ -5310,41 +6524,98 @@ static int bpf_object__collect_map_relos(struct bpf_object *obj,
 	return 0;
 }
 
-static int bpf_object__collect_reloc(struct bpf_object *obj)
+static int cmp_relocs(const void *_a, const void *_b)
 {
-	int i, err;
+	const struct reloc_desc *a = _a;
+	const struct reloc_desc *b = _b;
 
-	if (!obj_elf_valid(obj)) {
-		pr_warn("Internal error: elf object is closed\n");
-		return -LIBBPF_ERRNO__INTERNAL;
-	}
+	if (a->insn_idx != b->insn_idx)
+		return a->insn_idx < b->insn_idx ? -1 : 1;
+
+	/* no two relocations should have the same insn_idx, but ... */
+	if (a->type != b->type)
+		return a->type < b->type ? -1 : 1;
+
+	return 0;
+}
+
+static int bpf_object__collect_relos(struct bpf_object *obj)
+{
+	int i, err;
 
 	for (i = 0; i < obj->efile.nr_reloc_sects; i++) {
 		GElf_Shdr *shdr = &obj->efile.reloc_sects[i].shdr;
 		Elf_Data *data = obj->efile.reloc_sects[i].data;
 		int idx = shdr->sh_info;
-		struct bpf_program *prog;
 
 		if (shdr->sh_type != SHT_REL) {
 			pr_warn("internal error at %d\n", __LINE__);
 			return -LIBBPF_ERRNO__INTERNAL;
 		}
 
-		if (idx == obj->efile.st_ops_shndx) {
+		if (idx == obj->efile.st_ops_shndx)
 			err = bpf_object__collect_st_ops_relos(obj, shdr, data);
-		} else if (idx == obj->efile.btf_maps_shndx) {
+		else if (idx == obj->efile.btf_maps_shndx)
 			err = bpf_object__collect_map_relos(obj, shdr, data);
-		} else {
-			prog = bpf_object__find_prog_by_idx(obj, idx);
-			if (!prog) {
-				pr_warn("relocation failed: no prog in section(%d)\n", idx);
-				return -LIBBPF_ERRNO__RELOC;
-			}
-			err = bpf_program__collect_reloc(prog, shdr, data, obj);
-		}
+		else
+			err = bpf_object__collect_prog_relos(obj, shdr, data);
 		if (err)
 			return err;
 	}
+
+	for (i = 0; i < obj->nr_programs; i++) {
+		struct bpf_program *p = &obj->programs[i];
+		
+		if (!p->nr_reloc)
+			continue;
+
+		qsort(p->reloc_desc, p->nr_reloc, sizeof(*p->reloc_desc), cmp_relocs);
+	}
+	return 0;
+}
+
+static bool insn_is_helper_call(struct bpf_insn *insn, enum bpf_func_id *func_id)
+{
+	if (BPF_CLASS(insn->code) == BPF_JMP &&
+	    BPF_OP(insn->code) == BPF_CALL &&
+	    BPF_SRC(insn->code) == BPF_K &&
+	    insn->src_reg == 0 &&
+	    insn->dst_reg == 0) {
+		    *func_id = insn->imm;
+		    return true;
+	}
+	return false;
+}
+
+static int bpf_object__sanitize_prog(struct bpf_object* obj, struct bpf_program *prog)
+{
+	struct bpf_insn *insn = prog->insns;
+	enum bpf_func_id func_id;
+	int i;
+
+	for (i = 0; i < prog->insns_cnt; i++, insn++) {
+		if (!insn_is_helper_call(insn, &func_id))
+			continue;
+
+		/* on kernels that don't yet support
+		 * bpf_probe_read_{kernel,user}[_str] helpers, fall back
+		 * to bpf_probe_read() which works well for old kernels
+		 */
+		switch (func_id) {
+		case BPF_FUNC_probe_read_kernel:
+		case BPF_FUNC_probe_read_user:
+			if (!kernel_supports(FEAT_PROBE_READ_KERN))
+				insn->imm = BPF_FUNC_probe_read;
+			break;
+		case BPF_FUNC_probe_read_kernel_str:
+		case BPF_FUNC_probe_read_user_str:
+			if (!kernel_supports(FEAT_PROBE_READ_KERN))
+				insn->imm = BPF_FUNC_probe_read_str;
+			break;
+		default:
+			break;
+		}
+	}
 	return 0;
 }
 
@@ -5364,12 +6635,12 @@ load_program(struct bpf_program *prog, struct bpf_insn *insns, int insns_cnt,
 	memset(&load_attr, 0, sizeof(struct bpf_load_program_attr));
 	load_attr.prog_type = prog->type;
 	/* old kernels might not support specifying expected_attach_type */
-	if (!prog->caps->exp_attach_type && prog->sec_def &&
+	if (!kernel_supports(FEAT_EXP_ATTACH_TYPE) && prog->sec_def &&
 	    prog->sec_def->is_exp_attach_type_optional)
 		load_attr.expected_attach_type = 0;
 	else
 		load_attr.expected_attach_type = prog->expected_attach_type;
-	if (prog->caps->name)
+	if (kernel_supports(FEAT_PROG_NAME))
 		load_attr.name = prog->name;
 	load_attr.insns = insns;
 	load_attr.insns_cnt = insns_cnt;
@@ -5387,7 +6658,7 @@ load_program(struct bpf_program *prog, struct bpf_insn *insns, int insns_cnt,
 	}
 	/* specify func_info/line_info only if kernel supports them */
 	btf_fd = bpf_object__btf_fd(prog->obj);
-	if (btf_fd >= 0 && prog->obj->caps.btf_func) {
+	if (btf_fd >= 0 && kernel_supports(FEAT_BTF_FUNC)) {
 		load_attr.prog_btf_fd = btf_fd;
 		load_attr.func_info = prog->func_info;
 		load_attr.func_info_rec_size = prog->func_info_rec_size;
@@ -5413,6 +6684,20 @@ retry_load:
 	if (ret >= 0) {
 		if (log_buf && load_attr.log_level)
 			pr_debug("verifier log:\n%s", log_buf);
+
+		if (prog->obj->rodata_map_idx >= 0 &&
+		    kernel_supports(FEAT_PROG_BIND_MAP)) {
+			struct bpf_map *rodata_map =
+				&prog->obj->maps[prog->obj->rodata_map_idx];
+
+			if (bpf_prog_bind_map(ret, bpf_map__fd(rodata_map), NULL)) {
+				cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
+				pr_warn("prog '%s': failed to bind .rodata map: %s\n",
+					prog->name, cp);
+				/* Don't fail hard if can't bind rodata. */
+			}
+		}
+
 		*pfd = ret;
 		ret = 0;
 		goto out;
@@ -5425,7 +6710,7 @@ retry_load:
 		free(log_buf);
 		goto retry_load;
 	}
-	ret = -errno;
+	ret = errno ? -errno : -LIBBPF_ERRNO__LOAD;
 	cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
 	pr_warn("load bpf program failed: %s\n", cp);
 	pr_perm_msg(ret);
@@ -5465,8 +6750,7 @@ int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver)
 	int err = 0, fd, i, btf_id;
 
 	if (prog->obj->loaded) {
-		pr_warn("prog '%s'('%s'): can't load after object was loaded\n",
-			prog->name, prog->section_name);
+		pr_warn("prog '%s': can't load after object was loaded\n", prog->name);
 		return -EINVAL;
 	}
 
@@ -5482,7 +6766,7 @@ int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver)
 	if (prog->instances.nr < 0 || !prog->instances.fds) {
 		if (prog->preprocessor) {
 			pr_warn("Internal error: can't load program '%s'\n",
-				prog->section_name);
+				prog->name);
 			return -LIBBPF_ERRNO__INTERNAL;
 		}
 
@@ -5497,8 +6781,8 @@ int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver)
 
 	if (!prog->preprocessor) {
 		if (prog->instances.nr != 1) {
-			pr_warn("Program '%s' is inconsistent: nr(%d) != 1\n",
-				prog->section_name, prog->instances.nr);
+			pr_warn("prog '%s': inconsistent nr(%d) != 1\n",
+				prog->name, prog->instances.nr);
 		}
 		err = load_program(prog, prog->insns, prog->insns_cnt,
 				   license, kern_ver, &fd);
@@ -5516,13 +6800,13 @@ int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver)
 				   prog->insns_cnt, &result);
 		if (err) {
 			pr_warn("Preprocessing the %dth instance of program '%s' failed\n",
-				i, prog->section_name);
+				i, prog->name);
 			goto out;
 		}
 
 		if (!result.new_insn_ptr || !result.new_insn_cnt) {
 			pr_debug("Skip loading the %dth instance of program '%s'\n",
-				 i, prog->section_name);
+				 i, prog->name);
 			prog->instances.fds[i] = -1;
 			if (result.pfd)
 				*result.pfd = -1;
@@ -5533,7 +6817,7 @@ int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver)
 				   result.new_insn_cnt, license, kern_ver, &fd);
 		if (err) {
 			pr_warn("Loading the %dth instance of program '%s' failed\n",
-				i, prog->section_name);
+				i, prog->name);
 			goto out;
 		}
 
@@ -5543,18 +6827,12 @@ int bpf_program__load(struct bpf_program *prog, char *license, __u32 kern_ver)
 	}
 out:
 	if (err)
-		pr_warn("failed to load program '%s'\n", prog->section_name);
+		pr_warn("failed to load program '%s'\n", prog->name);
 	zfree(&prog->insns);
 	prog->insns_cnt = 0;
 	return err;
 }
 
-static bool bpf_program__is_function_storage(const struct bpf_program *prog,
-					     const struct bpf_object *obj)
-{
-	return prog->idx == obj->efile.text_shndx && obj->has_pseudo_calls;
-}
-
 static int
 bpf_object__load_progs(struct bpf_object *obj, int log_level)
 {
@@ -5564,11 +6842,17 @@ bpf_object__load_progs(struct bpf_object *obj, int log_level)
 
 	for (i = 0; i < obj->nr_programs; i++) {
 		prog = &obj->programs[i];
-		if (bpf_program__is_function_storage(prog, obj))
+		err = bpf_object__sanitize_prog(obj, prog);
+		if (err)
+			return err;
+	}
+
+	for (i = 0; i < obj->nr_programs; i++) {
+		prog = &obj->programs[i];
+		if (prog_is_subprog(obj, prog))
 			continue;
 		if (!prog->load) {
-			pr_debug("prog '%s'('%s'): skipped loading\n",
-				 prog->name, prog->section_name);
+			pr_debug("prog '%s': skipped loading\n", prog->name);
 			continue;
 		}
 		prog->log_level |= log_level;
@@ -5629,18 +6913,19 @@ __bpf_object__open(const char *path, const void *obj_buf, size_t obj_buf_sz,
 	err = err ? : bpf_object__collect_externs(obj);
 	err = err ? : bpf_object__finalize_btf(obj);
 	err = err ? : bpf_object__init_maps(obj, opts);
-	err = err ? : bpf_object__init_prog_names(obj);
-	err = err ? : bpf_object__collect_reloc(obj);
+	err = err ? : bpf_object__collect_relos(obj);
 	if (err)
 		goto out;
 	bpf_object__elf_finish(obj);
 
 	bpf_object__for_each_program(prog, obj) {
-		prog->sec_def = find_sec_def(prog->section_name);
+		prog->sec_def = find_sec_def(prog->sec_name);
 		if (!prog->sec_def)
 			/* couldn't guess, but user might manually specify */
 			continue;
 
+		if (prog->sec_def->is_sleepable)
+			prog->prog_flags |= BPF_F_SLEEPABLE;
 		bpf_program__set_type(prog, prog->sec_def->prog_type);
 		bpf_program__set_expected_attach_type(prog,
 				prog->sec_def->expected_attach_type);
@@ -5750,11 +7035,11 @@ static int bpf_object__sanitize_maps(struct bpf_object *obj)
 	bpf_object__for_each_map(m, obj) {
 		if (!bpf_map__is_internal(m))
 			continue;
-		if (!obj->caps.global_data) {
+		if (!kernel_supports(FEAT_GLOBAL_DATA)) {
 			pr_warn("kernel doesn't support global data\n");
 			return -ENOTSUP;
 		}
-		if (!obj->caps.array_mmap)
+		if (!kernel_supports(FEAT_ARRAY_MMAP))
 			m->def.map_flags ^= BPF_F_MMAPABLE;
 	}
 
@@ -5809,10 +7094,72 @@ out:
 	return err;
 }
 
+static int bpf_object__resolve_ksyms_btf_id(struct bpf_object *obj)
+{
+	struct extern_desc *ext;
+	int i, id;
+
+	for (i = 0; i < obj->nr_extern; i++) {
+		const struct btf_type *targ_var, *targ_type;
+		__u32 targ_type_id, local_type_id;
+		const char *targ_var_name;
+		int ret;
+
+		ext = &obj->externs[i];
+		if (ext->type != EXT_KSYM || !ext->ksym.type_id)
+			continue;
+
+		id = btf__find_by_name_kind(obj->btf_vmlinux, ext->name,
+					    BTF_KIND_VAR);
+		if (id <= 0) {
+			pr_warn("extern (ksym) '%s': failed to find BTF ID in vmlinux BTF.\n",
+				ext->name);
+			return -ESRCH;
+		}
+
+		/* find local type_id */
+		local_type_id = ext->ksym.type_id;
+
+		/* find target type_id */
+		targ_var = btf__type_by_id(obj->btf_vmlinux, id);
+		targ_var_name = btf__name_by_offset(obj->btf_vmlinux,
+						    targ_var->name_off);
+		targ_type = skip_mods_and_typedefs(obj->btf_vmlinux,
+						   targ_var->type,
+						   &targ_type_id);
+
+		ret = bpf_core_types_are_compat(obj->btf, local_type_id,
+						obj->btf_vmlinux, targ_type_id);
+		if (ret <= 0) {
+			const struct btf_type *local_type;
+			const char *targ_name, *local_name;
+
+			local_type = btf__type_by_id(obj->btf, local_type_id);
+			local_name = btf__name_by_offset(obj->btf,
+							 local_type->name_off);
+			targ_name = btf__name_by_offset(obj->btf_vmlinux,
+							targ_type->name_off);
+
+			pr_warn("extern (ksym) '%s': incompatible types, expected [%d] %s %s, but kernel has [%d] %s %s\n",
+				ext->name, local_type_id,
+				btf_kind_str(local_type), local_name, targ_type_id,
+				btf_kind_str(targ_type), targ_name);
+			return -EINVAL;
+		}
+
+		ext->is_set = true;
+		ext->ksym.vmlinux_btf_id = id;
+		pr_debug("extern (ksym) '%s': resolved to [%d] %s %s\n",
+			 ext->name, id, btf_kind_str(targ_var), targ_var_name);
+	}
+	return 0;
+}
+
 static int bpf_object__resolve_externs(struct bpf_object *obj,
 				       const char *extra_kconfig)
 {
 	bool need_config = false, need_kallsyms = false;
+	bool need_vmlinux_btf = false;
 	struct extern_desc *ext;
 	void *kcfg_data = NULL;
 	int err, i;
@@ -5843,7 +7190,10 @@ static int bpf_object__resolve_externs(struct bpf_object *obj,
 			   strncmp(ext->name, "CONFIG_", 7) == 0) {
 			need_config = true;
 		} else if (ext->type == EXT_KSYM) {
-			need_kallsyms = true;
+			if (ext->ksym.type_id)
+				need_vmlinux_btf = true;
+			else
+				need_kallsyms = true;
 		} else {
 			pr_warn("unrecognized extern '%s'\n", ext->name);
 			return -EINVAL;
@@ -5872,6 +7222,11 @@ static int bpf_object__resolve_externs(struct bpf_object *obj,
 		if (err)
 			return -EINVAL;
 	}
+	if (need_vmlinux_btf) {
+		err = bpf_object__resolve_ksyms_btf_id(obj);
+		if (err)
+			return -EINVAL;
+	}
 	for (i = 0; i < obj->nr_extern; i++) {
 		ext = &obj->externs[i];
 
@@ -5904,11 +7259,10 @@ int bpf_object__load_xattr(struct bpf_object_load_attr *attr)
 	}
 
 	err = bpf_object__probe_loading(obj);
-	err = err ? : bpf_object__probe_caps(obj);
+	err = err ? : bpf_object__load_vmlinux_btf(obj);
 	err = err ? : bpf_object__resolve_externs(obj, obj->kconfig);
 	err = err ? : bpf_object__sanitize_and_load_btf(obj);
 	err = err ? : bpf_object__sanitize_maps(obj);
-	err = err ? : bpf_object__load_vmlinux_btf(obj);
 	err = err ? : bpf_object__init_kern_struct_ops_maps(obj);
 	err = err ? : bpf_object__create_maps(obj);
 	err = err ? : bpf_object__relocate(obj, attr->target_btf_path);
@@ -6016,7 +7370,7 @@ int bpf_program__pin_instance(struct bpf_program *prog, const char *path,
 
 	if (instance < 0 || instance >= prog->instances.nr) {
 		pr_warn("invalid prog instance %d of prog %s (max %d)\n",
-			instance, prog->section_name, prog->instances.nr);
+			instance, prog->name, prog->instances.nr);
 		return -EINVAL;
 	}
 
@@ -6047,7 +7401,7 @@ int bpf_program__unpin_instance(struct bpf_program *prog, const char *path,
 
 	if (instance < 0 || instance >= prog->instances.nr) {
 		pr_warn("invalid prog instance %d of prog %s (max %d)\n",
-			instance, prog->section_name, prog->instances.nr);
+			instance, prog->name, prog->instances.nr);
 		return -EINVAL;
 	}
 
@@ -6077,8 +7431,7 @@ int bpf_program__pin(struct bpf_program *prog, const char *path)
 	}
 
 	if (prog->instances.nr <= 0) {
-		pr_warn("no instances of prog %s to pin\n",
-			   prog->section_name);
+		pr_warn("no instances of prog %s to pin\n", prog->name);
 		return -EINVAL;
 	}
 
@@ -6140,8 +7493,7 @@ int bpf_program__unpin(struct bpf_program *prog, const char *path)
 	}
 
 	if (prog->instances.nr <= 0) {
-		pr_warn("no instances of prog %s to pin\n",
-			   prog->section_name);
+		pr_warn("no instances of prog %s to pin\n", prog->name);
 		return -EINVAL;
 	}
 
@@ -6633,7 +7985,7 @@ bpf_program__next(struct bpf_program *prev, const struct bpf_object *obj)
 
 	do {
 		prog = __bpf_program__iter(prog, obj, true);
-	} while (prog && bpf_program__is_function_storage(prog, obj));
+	} while (prog && prog_is_subprog(obj, prog));
 
 	return prog;
 }
@@ -6645,7 +7997,7 @@ bpf_program__prev(struct bpf_program *next, const struct bpf_object *obj)
 
 	do {
 		prog = __bpf_program__iter(prog, obj, false);
-	} while (prog && bpf_program__is_function_storage(prog, obj));
+	} while (prog && prog_is_subprog(obj, prog));
 
 	return prog;
 }
@@ -6676,11 +8028,16 @@ const char *bpf_program__name(const struct bpf_program *prog)
 	return prog->name;
 }
 
+const char *bpf_program__section_name(const struct bpf_program *prog)
+{
+	return prog->sec_name;
+}
+
 const char *bpf_program__title(const struct bpf_program *prog, bool needs_copy)
 {
 	const char *title;
 
-	title = prog->section_name;
+	title = prog->sec_name;
 	if (needs_copy) {
 		title = strdup(title);
 		if (!title) {
@@ -6713,7 +8070,7 @@ int bpf_program__fd(const struct bpf_program *prog)
 
 size_t bpf_program__size(const struct bpf_program *prog)
 {
-	return prog->insns_cnt * sizeof(struct bpf_insn);
+	return prog->insns_cnt * BPF_INSN_SZ;
 }
 
 int bpf_program__set_prep(struct bpf_program *prog, int nr_instances,
@@ -6753,14 +8110,14 @@ int bpf_program__nth_fd(const struct bpf_program *prog, int n)
 
 	if (n >= prog->instances.nr || n < 0) {
 		pr_warn("Can't get the %dth fd from program %s: only %d instances\n",
-			n, prog->section_name, prog->instances.nr);
+			n, prog->name, prog->instances.nr);
 		return -EINVAL;
 	}
 
 	fd = prog->instances.fds[n];
 	if (fd < 0) {
 		pr_warn("%dth instance of program '%s' is invalid\n",
-			n, prog->section_name);
+			n, prog->name);
 		return -ENOENT;
 	}
 
@@ -6910,6 +8267,21 @@ static const struct bpf_sec_def section_defs[] = {
 		.expected_attach_type = BPF_TRACE_FEXIT,
 		.is_attach_btf = true,
 		.attach_fn = attach_trace),
+	SEC_DEF("fentry.s/", TRACING,
+		.expected_attach_type = BPF_TRACE_FENTRY,
+		.is_attach_btf = true,
+		.is_sleepable = true,
+		.attach_fn = attach_trace),
+	SEC_DEF("fmod_ret.s/", TRACING,
+		.expected_attach_type = BPF_MODIFY_RETURN,
+		.is_attach_btf = true,
+		.is_sleepable = true,
+		.attach_fn = attach_trace),
+	SEC_DEF("fexit.s/", TRACING,
+		.expected_attach_type = BPF_TRACE_FEXIT,
+		.is_attach_btf = true,
+		.is_sleepable = true,
+		.attach_fn = attach_trace),
 	SEC_DEF("freplace/", EXT,
 		.is_attach_btf = true,
 		.attach_fn = attach_trace),
@@ -6917,6 +8289,11 @@ static const struct bpf_sec_def section_defs[] = {
 		.is_attach_btf = true,
 		.expected_attach_type = BPF_LSM_MAC,
 		.attach_fn = attach_lsm),
+	SEC_DEF("lsm.s/", LSM,
+		.is_attach_btf = true,
+		.is_sleepable = true,
+		.expected_attach_type = BPF_LSM_MAC,
+		.attach_fn = attach_lsm),
 	SEC_DEF("iter/", TRACING,
 		.expected_attach_type = BPF_TRACE_ITER,
 		.is_attach_btf = true,
@@ -7100,7 +8477,7 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj,
 	const struct btf *btf;
 	struct bpf_map *map;
 	Elf_Data *symbols;
-	unsigned int moff;
+	unsigned int moff, insn_idx;
 	const char *name;
 	__u32 member_idx;
 	GElf_Sym sym;
@@ -7122,8 +8499,7 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj,
 			return -LIBBPF_ERRNO__FORMAT;
 		}
 
-		name = elf_strptr(obj->efile.elf, obj->efile.strtabidx,
-				  sym.st_name) ? : "<?>";
+		name = elf_sym_str(obj, sym.st_name) ?: "<?>";
 		map = find_struct_ops_map_by_offset(obj, rel.r_offset);
 		if (!map) {
 			pr_warn("struct_ops reloc: cannot find map at rel.r_offset %zu\n",
@@ -7146,6 +8522,12 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj,
 				map->name, (size_t)rel.r_offset, shdr_idx);
 			return -LIBBPF_ERRNO__RELOC;
 		}
+		if (sym.st_value % BPF_INSN_SZ) {
+			pr_warn("struct_ops reloc %s: invalid target program offset %llu\n",
+				map->name, (unsigned long long)sym.st_value);
+			return -LIBBPF_ERRNO__FORMAT;
+		}
+		insn_idx = sym.st_value / BPF_INSN_SZ;
 
 		member = find_member_by_offset(st_ops->type, moff * 8);
 		if (!member) {
@@ -7162,7 +8544,7 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj,
 			return -EINVAL;
 		}
 
-		prog = bpf_object__find_prog_by_idx(obj, shdr_idx);
+		prog = find_prog_by_sec_insn(obj, shdr_idx, insn_idx);
 		if (!prog) {
 			pr_warn("struct_ops reloc %s: cannot find prog at shdr_idx %u to relocate func ptr %s\n",
 				map->name, shdr_idx, name);
@@ -7172,7 +8554,7 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj,
 		if (prog->type == BPF_PROG_TYPE_UNSPEC) {
 			const struct bpf_sec_def *sec_def;
 
-			sec_def = find_sec_def(prog->section_name);
+			sec_def = find_sec_def(prog->sec_name);
 			if (sec_def &&
 			    sec_def->prog_type != BPF_PROG_TYPE_STRUCT_OPS) {
 				/* for pr_warn */
@@ -7195,7 +8577,7 @@ static int bpf_object__collect_st_ops_relos(struct bpf_object *obj,
 
 invalid_prog:
 	pr_warn("struct_ops reloc %s: cannot use prog %s in sec %s with type %u attach_btf_id %u expected_attach_type %u for func ptr %s\n",
-		map->name, prog->name, prog->section_name, prog->type,
+		map->name, prog->name, prog->sec_name, prog->type,
 		prog->attach_btf_id, prog->expected_attach_type, name);
 	return -EINVAL;
 }
@@ -7299,7 +8681,7 @@ static int libbpf_find_attach_btf_id(struct bpf_program *prog)
 {
 	enum bpf_attach_type attach_type = prog->expected_attach_type;
 	__u32 attach_prog_fd = prog->attach_prog_fd;
-	const char *name = prog->section_name;
+	const char *name = prog->sec_name;
 	int i, err;
 
 	if (!name)
@@ -7640,7 +9022,7 @@ int bpf_prog_load_xattr(const struct bpf_prog_load_attr *attr,
 
 		prog->prog_ifindex = attr->ifindex;
 		prog->log_level = attr->log_level;
-		prog->prog_flags = attr->prog_flags;
+		prog->prog_flags |= attr->prog_flags;
 		if (!first_prog)
 			first_prog = prog;
 	}
@@ -7826,14 +9208,14 @@ struct bpf_link *bpf_program__attach_perf_event(struct bpf_program *prog,
 	int prog_fd, err;
 
 	if (pfd < 0) {
-		pr_warn("program '%s': invalid perf event FD %d\n",
-			bpf_program__title(prog, false), pfd);
+		pr_warn("prog '%s': invalid perf event FD %d\n",
+			prog->name, pfd);
 		return ERR_PTR(-EINVAL);
 	}
 	prog_fd = bpf_program__fd(prog);
 	if (prog_fd < 0) {
-		pr_warn("program '%s': can't attach BPF program w/o FD (did you load it?)\n",
-			bpf_program__title(prog, false));
+		pr_warn("prog '%s': can't attach BPF program w/o FD (did you load it?)\n",
+			prog->name);
 		return ERR_PTR(-EINVAL);
 	}
 
@@ -7846,20 +9228,18 @@ struct bpf_link *bpf_program__attach_perf_event(struct bpf_program *prog,
 	if (ioctl(pfd, PERF_EVENT_IOC_SET_BPF, prog_fd) < 0) {
 		err = -errno;
 		free(link);
-		pr_warn("program '%s': failed to attach to pfd %d: %s\n",
-			bpf_program__title(prog, false), pfd,
-			   libbpf_strerror_r(err, errmsg, sizeof(errmsg)));
+		pr_warn("prog '%s': failed to attach to pfd %d: %s\n",
+			prog->name, pfd, libbpf_strerror_r(err, errmsg, sizeof(errmsg)));
 		if (err == -EPROTO)
-			pr_warn("program '%s': try add PERF_SAMPLE_CALLCHAIN to or remove exclude_callchain_[kernel|user] from pfd %d\n",
-				bpf_program__title(prog, false), pfd);
+			pr_warn("prog '%s': try add PERF_SAMPLE_CALLCHAIN to or remove exclude_callchain_[kernel|user] from pfd %d\n",
+				prog->name, pfd);
 		return ERR_PTR(err);
 	}
 	if (ioctl(pfd, PERF_EVENT_IOC_ENABLE, 0) < 0) {
 		err = -errno;
 		free(link);
-		pr_warn("program '%s': failed to enable pfd %d: %s\n",
-			bpf_program__title(prog, false), pfd,
-			   libbpf_strerror_r(err, errmsg, sizeof(errmsg)));
+		pr_warn("prog '%s': failed to enable pfd %d: %s\n",
+			prog->name, pfd, libbpf_strerror_r(err, errmsg, sizeof(errmsg)));
 		return ERR_PTR(err);
 	}
 	return link;
@@ -7981,9 +9361,8 @@ struct bpf_link *bpf_program__attach_kprobe(struct bpf_program *prog,
 	pfd = perf_event_open_probe(false /* uprobe */, retprobe, func_name,
 				    0 /* offset */, -1 /* pid */);
 	if (pfd < 0) {
-		pr_warn("program '%s': failed to create %s '%s' perf event: %s\n",
-			bpf_program__title(prog, false),
-			retprobe ? "kretprobe" : "kprobe", func_name,
+		pr_warn("prog '%s': failed to create %s '%s' perf event: %s\n",
+			prog->name, retprobe ? "kretprobe" : "kprobe", func_name,
 			libbpf_strerror_r(pfd, errmsg, sizeof(errmsg)));
 		return ERR_PTR(pfd);
 	}
@@ -7991,9 +9370,8 @@ struct bpf_link *bpf_program__attach_kprobe(struct bpf_program *prog,
 	if (IS_ERR(link)) {
 		close(pfd);
 		err = PTR_ERR(link);
-		pr_warn("program '%s': failed to attach to %s '%s': %s\n",
-			bpf_program__title(prog, false),
-			retprobe ? "kretprobe" : "kprobe", func_name,
+		pr_warn("prog '%s': failed to attach to %s '%s': %s\n",
+			prog->name, retprobe ? "kretprobe" : "kprobe", func_name,
 			libbpf_strerror_r(err, errmsg, sizeof(errmsg)));
 		return link;
 	}
@@ -8006,7 +9384,7 @@ static struct bpf_link *attach_kprobe(const struct bpf_sec_def *sec,
 	const char *func_name;
 	bool retprobe;
 
-	func_name = bpf_program__title(prog, false) + sec->len;
+	func_name = prog->sec_name + sec->len;
 	retprobe = strcmp(sec->sec, "kretprobe/") == 0;
 
 	return bpf_program__attach_kprobe(prog, retprobe, func_name);
@@ -8024,9 +9402,8 @@ struct bpf_link *bpf_program__attach_uprobe(struct bpf_program *prog,
 	pfd = perf_event_open_probe(true /* uprobe */, retprobe,
 				    binary_path, func_offset, pid);
 	if (pfd < 0) {
-		pr_warn("program '%s': failed to create %s '%s:0x%zx' perf event: %s\n",
-			bpf_program__title(prog, false),
-			retprobe ? "uretprobe" : "uprobe",
+		pr_warn("prog '%s': failed to create %s '%s:0x%zx' perf event: %s\n",
+			prog->name, retprobe ? "uretprobe" : "uprobe",
 			binary_path, func_offset,
 			libbpf_strerror_r(pfd, errmsg, sizeof(errmsg)));
 		return ERR_PTR(pfd);
@@ -8035,9 +9412,8 @@ struct bpf_link *bpf_program__attach_uprobe(struct bpf_program *prog,
 	if (IS_ERR(link)) {
 		close(pfd);
 		err = PTR_ERR(link);
-		pr_warn("program '%s': failed to attach to %s '%s:0x%zx': %s\n",
-			bpf_program__title(prog, false),
-			retprobe ? "uretprobe" : "uprobe",
+		pr_warn("prog '%s': failed to attach to %s '%s:0x%zx': %s\n",
+			prog->name, retprobe ? "uretprobe" : "uprobe",
 			binary_path, func_offset,
 			libbpf_strerror_r(err, errmsg, sizeof(errmsg)));
 		return link;
@@ -8105,9 +9481,8 @@ struct bpf_link *bpf_program__attach_tracepoint(struct bpf_program *prog,
 
 	pfd = perf_event_open_tracepoint(tp_category, tp_name);
 	if (pfd < 0) {
-		pr_warn("program '%s': failed to create tracepoint '%s/%s' perf event: %s\n",
-			bpf_program__title(prog, false),
-			tp_category, tp_name,
+		pr_warn("prog '%s': failed to create tracepoint '%s/%s' perf event: %s\n",
+			prog->name, tp_category, tp_name,
 			libbpf_strerror_r(pfd, errmsg, sizeof(errmsg)));
 		return ERR_PTR(pfd);
 	}
@@ -8115,9 +9490,8 @@ struct bpf_link *bpf_program__attach_tracepoint(struct bpf_program *prog,
 	if (IS_ERR(link)) {
 		close(pfd);
 		err = PTR_ERR(link);
-		pr_warn("program '%s': failed to attach to tracepoint '%s/%s': %s\n",
-			bpf_program__title(prog, false),
-			tp_category, tp_name,
+		pr_warn("prog '%s': failed to attach to tracepoint '%s/%s': %s\n",
+			prog->name, tp_category, tp_name,
 			libbpf_strerror_r(err, errmsg, sizeof(errmsg)));
 		return link;
 	}
@@ -8130,7 +9504,7 @@ static struct bpf_link *attach_tp(const struct bpf_sec_def *sec,
 	char *sec_name, *tp_cat, *tp_name;
 	struct bpf_link *link;
 
-	sec_name = strdup(bpf_program__title(prog, false));
+	sec_name = strdup(prog->sec_name);
 	if (!sec_name)
 		return ERR_PTR(-ENOMEM);
 
@@ -8159,8 +9533,7 @@ struct bpf_link *bpf_program__attach_raw_tracepoint(struct bpf_program *prog,
 
 	prog_fd = bpf_program__fd(prog);
 	if (prog_fd < 0) {
-		pr_warn("program '%s': can't attach before loaded\n",
-			bpf_program__title(prog, false));
+		pr_warn("prog '%s': can't attach before loaded\n", prog->name);
 		return ERR_PTR(-EINVAL);
 	}
 
@@ -8173,9 +9546,8 @@ struct bpf_link *bpf_program__attach_raw_tracepoint(struct bpf_program *prog,
 	if (pfd < 0) {
 		pfd = -errno;
 		free(link);
-		pr_warn("program '%s': failed to attach to raw tracepoint '%s': %s\n",
-			bpf_program__title(prog, false), tp_name,
-			libbpf_strerror_r(pfd, errmsg, sizeof(errmsg)));
+		pr_warn("prog '%s': failed to attach to raw tracepoint '%s': %s\n",
+			prog->name, tp_name, libbpf_strerror_r(pfd, errmsg, sizeof(errmsg)));
 		return ERR_PTR(pfd);
 	}
 	link->fd = pfd;
@@ -8185,7 +9557,7 @@ struct bpf_link *bpf_program__attach_raw_tracepoint(struct bpf_program *prog,
 static struct bpf_link *attach_raw_tp(const struct bpf_sec_def *sec,
 				      struct bpf_program *prog)
 {
-	const char *tp_name = bpf_program__title(prog, false) + sec->len;
+	const char *tp_name = prog->sec_name + sec->len;
 
 	return bpf_program__attach_raw_tracepoint(prog, tp_name);
 }
@@ -8199,8 +9571,7 @@ static struct bpf_link *bpf_program__attach_btf_id(struct bpf_program *prog)
 
 	prog_fd = bpf_program__fd(prog);
 	if (prog_fd < 0) {
-		pr_warn("program '%s': can't attach before loaded\n",
-			bpf_program__title(prog, false));
+		pr_warn("prog '%s': can't attach before loaded\n", prog->name);
 		return ERR_PTR(-EINVAL);
 	}
 
@@ -8213,9 +9584,8 @@ static struct bpf_link *bpf_program__attach_btf_id(struct bpf_program *prog)
 	if (pfd < 0) {
 		pfd = -errno;
 		free(link);
-		pr_warn("program '%s': failed to attach: %s\n",
-			bpf_program__title(prog, false),
-			libbpf_strerror_r(pfd, errmsg, sizeof(errmsg)));
+		pr_warn("prog '%s': failed to attach: %s\n",
+			prog->name, libbpf_strerror_r(pfd, errmsg, sizeof(errmsg)));
 		return ERR_PTR(pfd);
 	}
 	link->fd = pfd;
@@ -8251,9 +9621,11 @@ static struct bpf_link *attach_iter(const struct bpf_sec_def *sec,
 }
 
 static struct bpf_link *
-bpf_program__attach_fd(struct bpf_program *prog, int target_fd,
+bpf_program__attach_fd(struct bpf_program *prog, int target_fd, int btf_id,
 		       const char *target_name)
 {
+	DECLARE_LIBBPF_OPTS(bpf_link_create_opts, opts,
+			    .target_btf_id = btf_id);
 	enum bpf_attach_type attach_type;
 	char errmsg[STRERR_BUFSIZE];
 	struct bpf_link *link;
@@ -8261,8 +9633,7 @@ bpf_program__attach_fd(struct bpf_program *prog, int target_fd,
 
 	prog_fd = bpf_program__fd(prog);
 	if (prog_fd < 0) {
-		pr_warn("program '%s': can't attach before loaded\n",
-			bpf_program__title(prog, false));
+		pr_warn("prog '%s': can't attach before loaded\n", prog->name);
 		return ERR_PTR(-EINVAL);
 	}
 
@@ -8272,12 +9643,12 @@ bpf_program__attach_fd(struct bpf_program *prog, int target_fd,
 	link->detach = &bpf_link__detach_fd;
 
 	attach_type = bpf_program__get_expected_attach_type(prog);
-	link_fd = bpf_link_create(prog_fd, target_fd, attach_type, NULL);
+	link_fd = bpf_link_create(prog_fd, target_fd, attach_type, &opts);
 	if (link_fd < 0) {
 		link_fd = -errno;
 		free(link);
-		pr_warn("program '%s': failed to attach to %s: %s\n",
-			bpf_program__title(prog, false), target_name,
+		pr_warn("prog '%s': failed to attach to %s: %s\n",
+			prog->name, target_name,
 			libbpf_strerror_r(link_fd, errmsg, sizeof(errmsg)));
 		return ERR_PTR(link_fd);
 	}
@@ -8288,19 +9659,51 @@ bpf_program__attach_fd(struct bpf_program *prog, int target_fd,
 struct bpf_link *
 bpf_program__attach_cgroup(struct bpf_program *prog, int cgroup_fd)
 {
-	return bpf_program__attach_fd(prog, cgroup_fd, "cgroup");
+	return bpf_program__attach_fd(prog, cgroup_fd, 0, "cgroup");
 }
 
 struct bpf_link *
 bpf_program__attach_netns(struct bpf_program *prog, int netns_fd)
 {
-	return bpf_program__attach_fd(prog, netns_fd, "netns");
+	return bpf_program__attach_fd(prog, netns_fd, 0, "netns");
 }
 
 struct bpf_link *bpf_program__attach_xdp(struct bpf_program *prog, int ifindex)
 {
 	/* target_fd/target_ifindex use the same field in LINK_CREATE */
-	return bpf_program__attach_fd(prog, ifindex, "xdp");
+	return bpf_program__attach_fd(prog, ifindex, 0, "xdp");
+}
+
+struct bpf_link *bpf_program__attach_freplace(struct bpf_program *prog,
+					      int target_fd,
+					      const char *attach_func_name)
+{
+	int btf_id;
+
+	if (!!target_fd != !!attach_func_name) {
+		pr_warn("prog '%s': supply none or both of target_fd and attach_func_name\n",
+			prog->name);
+		return ERR_PTR(-EINVAL);
+	}
+
+	if (prog->type != BPF_PROG_TYPE_EXT) {
+		pr_warn("prog '%s': only BPF_PROG_TYPE_EXT can attach as freplace",
+			prog->name);
+		return ERR_PTR(-EINVAL);
+	}
+
+	if (target_fd) {
+		btf_id = libbpf_find_prog_btf_id(attach_func_name, target_fd);
+		if (btf_id < 0)
+			return ERR_PTR(btf_id);
+
+		return bpf_program__attach_fd(prog, target_fd, btf_id, "freplace");
+	} else {
+		/* no target, so use raw_tracepoint_open for compatibility
+		 * with old kernels
+		 */
+		return bpf_program__attach_trace(prog);
+	}
 }
 
 struct bpf_link *
@@ -8321,8 +9724,7 @@ bpf_program__attach_iter(struct bpf_program *prog,
 
 	prog_fd = bpf_program__fd(prog);
 	if (prog_fd < 0) {
-		pr_warn("program '%s': can't attach before loaded\n",
-			bpf_program__title(prog, false));
+		pr_warn("prog '%s': can't attach before loaded\n", prog->name);
 		return ERR_PTR(-EINVAL);
 	}
 
@@ -8336,9 +9738,8 @@ bpf_program__attach_iter(struct bpf_program *prog,
 	if (link_fd < 0) {
 		link_fd = -errno;
 		free(link);
-		pr_warn("program '%s': failed to attach to iterator: %s\n",
-			bpf_program__title(prog, false),
-			libbpf_strerror_r(link_fd, errmsg, sizeof(errmsg)));
+		pr_warn("prog '%s': failed to attach to iterator: %s\n",
+			prog->name, libbpf_strerror_r(link_fd, errmsg, sizeof(errmsg)));
 		return ERR_PTR(link_fd);
 	}
 	link->fd = link_fd;
@@ -8349,7 +9750,7 @@ struct bpf_link *bpf_program__attach(struct bpf_program *prog)
 {
 	const struct bpf_sec_def *sec_def;
 
-	sec_def = find_sec_def(bpf_program__title(prog, false));
+	sec_def = find_sec_def(prog->sec_name);
 	if (!sec_def || !sec_def->attach_fn)
 		return ERR_PTR(-ESRCH);
 
@@ -8594,7 +9995,7 @@ struct perf_buffer *perf_buffer__new(int map_fd, size_t page_cnt,
 	struct perf_buffer_params p = {};
 	struct perf_event_attr attr = { 0, };
 
-	attr.config = PERF_COUNT_SW_BPF_OUTPUT,
+	attr.config = PERF_COUNT_SW_BPF_OUTPUT;
 	attr.type = PERF_TYPE_SOFTWARE;
 	attr.sample_type = PERF_SAMPLE_RAW;
 	attr.sample_period = 1;
@@ -8832,6 +10233,11 @@ static int perf_buffer__process_records(struct perf_buffer *pb,
 	return 0;
 }
 
+int perf_buffer__epoll_fd(const struct perf_buffer *pb)
+{
+	return pb->epoll_fd;
+}
+
 int perf_buffer__poll(struct perf_buffer *pb, int timeout_ms)
 {
 	int i, cnt, err;
@@ -8849,6 +10255,55 @@ int perf_buffer__poll(struct perf_buffer *pb, int timeout_ms)
 	return cnt < 0 ? -errno : cnt;
 }
 
+/* Return number of PERF_EVENT_ARRAY map slots set up by this perf_buffer
+ * manager.
+ */
+size_t perf_buffer__buffer_cnt(const struct perf_buffer *pb)
+{
+	return pb->cpu_cnt;
+}
+
+/*
+ * Return perf_event FD of a ring buffer in *buf_idx* slot of
+ * PERF_EVENT_ARRAY BPF map. This FD can be polled for new data using
+ * select()/poll()/epoll() Linux syscalls.
+ */
+int perf_buffer__buffer_fd(const struct perf_buffer *pb, size_t buf_idx)
+{
+	struct perf_cpu_buf *cpu_buf;
+
+	if (buf_idx >= pb->cpu_cnt)
+		return -EINVAL;
+
+	cpu_buf = pb->cpu_bufs[buf_idx];
+	if (!cpu_buf)
+		return -ENOENT;
+
+	return cpu_buf->fd;
+}
+
+/*
+ * Consume data from perf ring buffer corresponding to slot *buf_idx* in
+ * PERF_EVENT_ARRAY BPF map without waiting/polling. If there is no data to
+ * consume, do nothing and return success.
+ * Returns:
+ *   - 0 on success;
+ *   - <0 on failure.
+ */
+int perf_buffer__consume_buffer(struct perf_buffer *pb, size_t buf_idx)
+{
+	struct perf_cpu_buf *cpu_buf;
+
+	if (buf_idx >= pb->cpu_cnt)
+		return -EINVAL;
+
+	cpu_buf = pb->cpu_bufs[buf_idx];
+	if (!cpu_buf)
+		return -ENOENT;
+
+	return perf_buffer__process_records(pb, cpu_buf);
+}
+
 int perf_buffer__consume(struct perf_buffer *pb)
 {
 	int i, err;
@@ -8861,7 +10316,7 @@ int perf_buffer__consume(struct perf_buffer *pb)
 
 		err = perf_buffer__process_records(pb, cpu_buf);
 		if (err) {
-			pr_warn("error while processing records: %d\n", err);
+			pr_warn("perf_buffer: failed to process records in buffer #%d: %d\n", i, err);
 			return err;
 		}
 	}
@@ -9129,9 +10584,8 @@ int bpf_program__set_attach_target(struct bpf_program *prog,
 		btf_id = libbpf_find_prog_btf_id(attach_func_name,
 						 attach_prog_fd);
 	else
-		btf_id = __find_vmlinux_btf_id(prog->obj->btf_vmlinux,
-					       attach_func_name,
-					       prog->expected_attach_type);
+		btf_id = libbpf_find_vmlinux_btf_id(attach_func_name,
+						    prog->expected_attach_type);
 
 	if (btf_id < 0)
 		return btf_id;
@@ -9365,12 +10819,11 @@ int bpf_object__attach_skeleton(struct bpf_object_skeleton *s)
 		struct bpf_program *prog = *s->progs[i].prog;
 		struct bpf_link **link = s->progs[i].link;
 		const struct bpf_sec_def *sec_def;
-		const char *sec_name = bpf_program__title(prog, false);
 
 		if (!prog->load)
 			continue;
 
-		sec_def = find_sec_def(sec_name);
+		sec_def = find_sec_def(prog->sec_name);
 		if (!sec_def || !sec_def->attach_fn)
 			continue;
 
diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 5ecb4069a9f0..6909ee81113a 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -198,8 +198,9 @@ LIBBPF_API void bpf_program__set_ifindex(struct bpf_program *prog,
 					 __u32 ifindex);
 
 LIBBPF_API const char *bpf_program__name(const struct bpf_program *prog);
-LIBBPF_API const char *bpf_program__title(const struct bpf_program *prog,
-					  bool needs_copy);
+LIBBPF_API const char *bpf_program__section_name(const struct bpf_program *prog);
+LIBBPF_API LIBBPF_DEPRECATED("BPF program title is confusing term; please use bpf_program__section_name() instead")
+const char *bpf_program__title(const struct bpf_program *prog, bool needs_copy);
 LIBBPF_API bool bpf_program__autoload(const struct bpf_program *prog);
 LIBBPF_API int bpf_program__set_autoload(struct bpf_program *prog, bool autoload);
 
@@ -260,6 +261,9 @@ LIBBPF_API struct bpf_link *
 bpf_program__attach_netns(struct bpf_program *prog, int netns_fd);
 LIBBPF_API struct bpf_link *
 bpf_program__attach_xdp(struct bpf_program *prog, int ifindex);
+LIBBPF_API struct bpf_link *
+bpf_program__attach_freplace(struct bpf_program *prog,
+			     int target_fd, const char *attach_func_name);
 
 struct bpf_map;
 
@@ -588,8 +592,12 @@ perf_buffer__new_raw(int map_fd, size_t page_cnt,
 		     const struct perf_buffer_raw_opts *opts);
 
 LIBBPF_API void perf_buffer__free(struct perf_buffer *pb);
+LIBBPF_API int perf_buffer__epoll_fd(const struct perf_buffer *pb);
 LIBBPF_API int perf_buffer__poll(struct perf_buffer *pb, int timeout_ms);
 LIBBPF_API int perf_buffer__consume(struct perf_buffer *pb);
+LIBBPF_API int perf_buffer__consume_buffer(struct perf_buffer *pb, size_t buf_idx);
+LIBBPF_API size_t perf_buffer__buffer_cnt(const struct perf_buffer *pb);
+LIBBPF_API int perf_buffer__buffer_fd(const struct perf_buffer *pb, size_t buf_idx);
 
 typedef enum bpf_perf_event_ret
 	(*bpf_perf_event_print_t)(struct perf_event_header *hdr,
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index e35bd6cdbdbf..4ebfadf45b47 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -299,3 +299,41 @@ LIBBPF_0.1.0 {
 		btf__set_fd;
 		btf__set_pointer_size;
 } LIBBPF_0.0.9;
+
+LIBBPF_0.2.0 {
+	global:
+		bpf_prog_bind_map;
+		bpf_prog_test_run_opts;
+		bpf_program__attach_freplace;
+		bpf_program__section_name;
+		btf__add_array;
+		btf__add_const;
+		btf__add_enum;
+		btf__add_enum_value;
+		btf__add_datasec;
+		btf__add_datasec_var_info;
+		btf__add_field;
+		btf__add_func;
+		btf__add_func_param;
+		btf__add_func_proto;
+		btf__add_fwd;
+		btf__add_int;
+		btf__add_ptr;
+		btf__add_restrict;
+		btf__add_str;
+		btf__add_struct;
+		btf__add_typedef;
+		btf__add_union;
+		btf__add_var;
+		btf__add_volatile;
+		btf__endianness;
+		btf__find_str;
+		btf__new_empty;
+		btf__set_endianness;
+		btf__str_by_offset;
+		perf_buffer__buffer_cnt;
+		perf_buffer__buffer_fd;
+		perf_buffer__epoll_fd;
+		perf_buffer__consume_buffer;
+		xsk_socket__create_shared;
+} LIBBPF_0.1.0;
diff --git a/tools/lib/bpf/libbpf_common.h b/tools/lib/bpf/libbpf_common.h
index a23ae1ac27eb..947d8bd8a7bb 100644
--- a/tools/lib/bpf/libbpf_common.h
+++ b/tools/lib/bpf/libbpf_common.h
@@ -15,6 +15,8 @@
 #define LIBBPF_API __attribute__((visibility("default")))
 #endif
 
+#define LIBBPF_DEPRECATED(msg) __attribute__((deprecated(msg)))
+
 /* Helper macro to declare and initialize libbpf options struct
  *
  * This dance with uninitialized declaration, followed by memset to zero,
diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
index 50d70e90d5f1..d99bc847bf84 100644
--- a/tools/lib/bpf/libbpf_internal.h
+++ b/tools/lib/bpf/libbpf_internal.h
@@ -9,6 +9,15 @@
 #ifndef __LIBBPF_LIBBPF_INTERNAL_H
 #define __LIBBPF_LIBBPF_INTERNAL_H
 
+#include <stdlib.h>
+#include <limits.h>
+
+/* make sure libbpf doesn't use kernel-only integer typedefs */
+#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+
+/* prevent accidental re-addition of reallocarray() */
+#pragma GCC poison reallocarray
+
 #include "libbpf.h"
 
 #define BTF_INFO_ENC(kind, kind_flag, vlen) \
@@ -23,6 +32,12 @@
 #define BTF_PARAM_ENC(name, type) (name), (type)
 #define BTF_VAR_SECINFO_ENC(type, offset, size) (type), (offset), (size)
 
+#ifndef likely
+#define likely(x) __builtin_expect(!!(x), 1)
+#endif
+#ifndef unlikely
+#define unlikely(x) __builtin_expect(!!(x), 0)
+#endif
 #ifndef min
 # define min(x, y) ((x) < (y) ? (x) : (y))
 #endif
@@ -63,6 +78,37 @@ do {				\
 #define pr_info(fmt, ...)	__pr(LIBBPF_INFO, fmt, ##__VA_ARGS__)
 #define pr_debug(fmt, ...)	__pr(LIBBPF_DEBUG, fmt, ##__VA_ARGS__)
 
+#ifndef __has_builtin
+#define __has_builtin(x) 0
+#endif
+/*
+ * Re-implement glibc's reallocarray() for libbpf internal-only use.
+ * reallocarray(), unfortunately, is not available in all versions of glibc,
+ * so requires extra feature detection and using reallocarray() stub from
+ * <tools/libc_compat.h> and COMPAT_NEED_REALLOCARRAY. All this complicates
+ * build of libbpf unnecessarily and is just a maintenance burden. Instead,
+ * it's trivial to implement libbpf-specific internal version and use it
+ * throughout libbpf.
+ */
+static inline void *libbpf_reallocarray(void *ptr, size_t nmemb, size_t size)
+{
+	size_t total;
+
+#if __has_builtin(__builtin_mul_overflow)
+	if (unlikely(__builtin_mul_overflow(nmemb, size, &total)))
+		return NULL;
+#else
+	if (size == 0 || nmemb > ULONG_MAX / size)
+		return NULL;
+	total = nmemb * size;
+#endif
+	return realloc(ptr, total);
+}
+
+void *btf_add_mem(void **data, size_t *cap_cnt, size_t elem_sz,
+		  size_t cur_cnt, size_t max_cnt, size_t add_cnt);
+int btf_ensure_mem(void **data, size_t *cap_cnt, size_t elem_sz, size_t need_cnt);
+
 static inline bool libbpf_validate_opts(const char *opts,
 					size_t opts_sz, size_t user_sz,
 					const char *type_name)
@@ -94,6 +140,11 @@ static inline bool libbpf_validate_opts(const char *opts,
 	((opts) && opts->sz >= offsetofend(typeof(*(opts)), field))
 #define OPTS_GET(opts, field, fallback_value) \
 	(OPTS_HAS(opts, field) ? (opts)->field : fallback_value)
+#define OPTS_SET(opts, field, value)		\
+	do {					\
+		if (OPTS_HAS(opts, field))	\
+			(opts)->field = value;	\
+	} while (0)
 
 int parse_cpu_mask_str(const char *s, bool **mask, int *mask_sz);
 int parse_cpu_mask_file(const char *fcpu, bool **mask, int *mask_sz);
@@ -105,18 +156,6 @@ int bpf_object__section_size(const struct bpf_object *obj, const char *name,
 int bpf_object__variable_offset(const struct bpf_object *obj, const char *name,
 				__u32 *off);
 
-struct nlattr;
-typedef int (*libbpf_dump_nlmsg_t)(void *cookie, void *msg, struct nlattr **tb);
-int libbpf_netlink_open(unsigned int *nl_pid);
-int libbpf_nl_get_link(int sock, unsigned int nl_pid,
-		       libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie);
-int libbpf_nl_get_class(int sock, unsigned int nl_pid, int ifindex,
-			libbpf_dump_nlmsg_t dump_class_nlmsg, void *cookie);
-int libbpf_nl_get_qdisc(int sock, unsigned int nl_pid, int ifindex,
-			libbpf_dump_nlmsg_t dump_qdisc_nlmsg, void *cookie);
-int libbpf_nl_get_filter(int sock, unsigned int nl_pid, int ifindex, int handle,
-			 libbpf_dump_nlmsg_t dump_filter_nlmsg, void *cookie);
-
 struct btf_ext_info {
 	/*
 	 * info points to the individual info section (e.g. func_info and
@@ -138,6 +177,44 @@ struct btf_ext_info {
 	     i < (sec)->num_info;					\
 	     i++, rec = (void *)rec + (seg)->rec_size)
 
+/*
+ * The .BTF.ext ELF section layout defined as
+ *   struct btf_ext_header
+ *   func_info subsection
+ *
+ * The func_info subsection layout:
+ *   record size for struct bpf_func_info in the func_info subsection
+ *   struct btf_sec_func_info for section #1
+ *   a list of bpf_func_info records for section #1
+ *     where struct bpf_func_info mimics one in include/uapi/linux/bpf.h
+ *     but may not be identical
+ *   struct btf_sec_func_info for section #2
+ *   a list of bpf_func_info records for section #2
+ *   ......
+ *
+ * Note that the bpf_func_info record size in .BTF.ext may not
+ * be the same as the one defined in include/uapi/linux/bpf.h.
+ * The loader should ensure that record_size meets minimum
+ * requirement and pass the record as is to the kernel. The
+ * kernel will handle the func_info properly based on its contents.
+ */
+struct btf_ext_header {
+	__u16	magic;
+	__u8	version;
+	__u8	flags;
+	__u32	hdr_len;
+
+	/* All offsets are in bytes relative to the end of this header */
+	__u32	func_info_off;
+	__u32	func_info_len;
+	__u32	line_info_off;
+	__u32	line_info_len;
+
+	/* optional part of .BTF.ext header */
+	__u32	core_relo_off;
+	__u32	core_relo_len;
+};
+
 struct btf_ext {
 	union {
 		struct btf_ext_header *hdr;
@@ -145,7 +222,7 @@ struct btf_ext {
 	};
 	struct btf_ext_info func_info;
 	struct btf_ext_info line_info;
-	struct btf_ext_info field_reloc_info;
+	struct btf_ext_info core_relo_info;
 	__u32 data_size;
 };
 
@@ -170,32 +247,40 @@ struct bpf_line_info_min {
 	__u32	line_col;
 };
 
-/* bpf_field_info_kind encodes which aspect of captured field has to be
- * adjusted by relocations. Currently supported values are:
- *   - BPF_FIELD_BYTE_OFFSET: field offset (in bytes);
- *   - BPF_FIELD_EXISTS: field existence (1, if field exists; 0, otherwise);
+/* bpf_core_relo_kind encodes which aspect of captured field/type/enum value
+ * has to be adjusted by relocations.
  */
-enum bpf_field_info_kind {
+enum bpf_core_relo_kind {
 	BPF_FIELD_BYTE_OFFSET = 0,	/* field byte offset */
-	BPF_FIELD_BYTE_SIZE = 1,
+	BPF_FIELD_BYTE_SIZE = 1,	/* field size in bytes */
 	BPF_FIELD_EXISTS = 2,		/* field existence in target kernel */
-	BPF_FIELD_SIGNED = 3,
-	BPF_FIELD_LSHIFT_U64 = 4,
-	BPF_FIELD_RSHIFT_U64 = 5,
+	BPF_FIELD_SIGNED = 3,		/* field signedness (0 - unsigned, 1 - signed) */
+	BPF_FIELD_LSHIFT_U64 = 4,	/* bitfield-specific left bitshift */
+	BPF_FIELD_RSHIFT_U64 = 5,	/* bitfield-specific right bitshift */
+	BPF_TYPE_ID_LOCAL = 6,		/* type ID in local BPF object */
+	BPF_TYPE_ID_TARGET = 7,		/* type ID in target kernel */
+	BPF_TYPE_EXISTS = 8,		/* type existence in target kernel */
+	BPF_TYPE_SIZE = 9,		/* type size in bytes */
+	BPF_ENUMVAL_EXISTS = 10,	/* enum value existence in target kernel */
+	BPF_ENUMVAL_VALUE = 11,		/* enum value integer value */
 };
 
-/* The minimum bpf_field_reloc checked by the loader
+/* The minimum bpf_core_relo checked by the loader
  *
- * Field relocation captures the following data:
+ * CO-RE relocation captures the following data:
  * - insn_off - instruction offset (in bytes) within a BPF program that needs
  *   its insn->imm field to be relocated with actual field info;
  * - type_id - BTF type ID of the "root" (containing) entity of a relocatable
- *   field;
+ *   type or field;
  * - access_str_off - offset into corresponding .BTF string section. String
- *   itself encodes an accessed field using a sequence of field and array
- *   indicies, separated by colon (:). It's conceptually very close to LLVM's
- *   getelementptr ([0]) instruction's arguments for identifying offset to 
- *   a field.
+ *   interpretation depends on specific relocation kind:
+ *     - for field-based relocations, string encodes an accessed field using
+ *     a sequence of field and array indices, separated by colon (:). It's
+ *     conceptually very close to LLVM's getelementptr ([0]) instruction's
+ *     arguments for identifying offset to a field.
+ *     - for type-based relocations, strings is expected to be just "0";
+ *     - for enum value-based relocations, string contains an index of enum
+ *     value within its enum type;
  *
  * Example to provide a better feel.
  *
@@ -226,11 +311,11 @@ enum bpf_field_info_kind {
  *
  *   [0] https://llvm.org/docs/LangRef.html#getelementptr-instruction
  */
-struct bpf_field_reloc {
+struct bpf_core_relo {
 	__u32   insn_off;
 	__u32   type_id;
 	__u32   access_str_off;
-	enum bpf_field_info_kind kind;
+	enum bpf_core_relo_kind kind;
 };
 
 #endif /* __LIBBPF_LIBBPF_INTERNAL_H */
diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
index 5a3d3f078408..5482a9b7ae2d 100644
--- a/tools/lib/bpf/libbpf_probes.c
+++ b/tools/lib/bpf/libbpf_probes.c
@@ -17,9 +17,6 @@
 #include "libbpf.h"
 #include "libbpf_internal.h"
 
-/* make sure libbpf doesn't use kernel-only integer typedefs */
-#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
-
 static bool grep(const char *buffer, const char *pattern)
 {
 	return !!strstr(buffer, pattern);
@@ -173,7 +170,7 @@ int libbpf__load_raw_btf(const char *raw_types, size_t types_len,
 	return btf_fd;
 }
 
-static int load_sk_storage_btf(void)
+static int load_local_storage_btf(void)
 {
 	const char strs[] = "\0bpf_spin_lock\0val\0cnt\0l";
 	/* struct bpf_spin_lock {
@@ -232,12 +229,13 @@ bool bpf_probe_map_type(enum bpf_map_type map_type, __u32 ifindex)
 		key_size	= 0;
 		break;
 	case BPF_MAP_TYPE_SK_STORAGE:
+	case BPF_MAP_TYPE_INODE_STORAGE:
 		btf_key_type_id = 1;
 		btf_value_type_id = 3;
 		value_size = 8;
 		max_entries = 0;
 		map_flags = BPF_F_NO_PREALLOC;
-		btf_fd = load_sk_storage_btf();
+		btf_fd = load_local_storage_btf();
 		if (btf_fd < 0)
 			return false;
 		break;
diff --git a/tools/lib/bpf/netlink.c b/tools/lib/bpf/netlink.c
index 312f887570b2..4dd73de00b6f 100644
--- a/tools/lib/bpf/netlink.c
+++ b/tools/lib/bpf/netlink.c
@@ -15,13 +15,12 @@
 #include "libbpf_internal.h"
 #include "nlattr.h"
 
-/* make sure libbpf doesn't use kernel-only integer typedefs */
-#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
-
 #ifndef SOL_NETLINK
 #define SOL_NETLINK 270
 #endif
 
+typedef int (*libbpf_dump_nlmsg_t)(void *cookie, void *msg, struct nlattr **tb);
+
 typedef int (*__dump_nlmsg_t)(struct nlmsghdr *nlmsg, libbpf_dump_nlmsg_t,
 			      void *cookie);
 
@@ -31,7 +30,7 @@ struct xdp_id_md {
 	struct xdp_link_info info;
 };
 
-int libbpf_netlink_open(__u32 *nl_pid)
+static int libbpf_netlink_open(__u32 *nl_pid)
 {
 	struct sockaddr_nl sa;
 	socklen_t addrlen;
@@ -283,6 +282,9 @@ static int get_xdp_info(void *cookie, void *msg, struct nlattr **tb)
 	return 0;
 }
 
+static int libbpf_nl_get_link(int sock, unsigned int nl_pid,
+			      libbpf_dump_nlmsg_t dump_link_nlmsg, void *cookie);
+
 int bpf_get_link_xdp_info(int ifindex, struct xdp_link_info *info,
 			  size_t info_size, __u32 flags)
 {
@@ -368,121 +370,3 @@ int libbpf_nl_get_link(int sock, unsigned int nl_pid,
 	return bpf_netlink_recv(sock, nl_pid, seq, __dump_link_nlmsg,
 				dump_link_nlmsg, cookie);
 }
-
-static int __dump_class_nlmsg(struct nlmsghdr *nlh,
-			      libbpf_dump_nlmsg_t dump_class_nlmsg,
-			      void *cookie)
-{
-	struct nlattr *tb[TCA_MAX + 1], *attr;
-	struct tcmsg *t = NLMSG_DATA(nlh);
-	int len;
-
-	len = nlh->nlmsg_len - NLMSG_LENGTH(sizeof(*t));
-	attr = (struct nlattr *) ((void *) t + NLMSG_ALIGN(sizeof(*t)));
-	if (libbpf_nla_parse(tb, TCA_MAX, attr, len, NULL) != 0)
-		return -LIBBPF_ERRNO__NLPARSE;
-
-	return dump_class_nlmsg(cookie, t, tb);
-}
-
-int libbpf_nl_get_class(int sock, unsigned int nl_pid, int ifindex,
-			libbpf_dump_nlmsg_t dump_class_nlmsg, void *cookie)
-{
-	struct {
-		struct nlmsghdr nlh;
-		struct tcmsg t;
-	} req = {
-		.nlh.nlmsg_len = NLMSG_LENGTH(sizeof(struct tcmsg)),
-		.nlh.nlmsg_type = RTM_GETTCLASS,
-		.nlh.nlmsg_flags = NLM_F_DUMP | NLM_F_REQUEST,
-		.t.tcm_family = AF_UNSPEC,
-		.t.tcm_ifindex = ifindex,
-	};
-	int seq = time(NULL);
-
-	req.nlh.nlmsg_seq = seq;
-	if (send(sock, &req, req.nlh.nlmsg_len, 0) < 0)
-		return -errno;
-
-	return bpf_netlink_recv(sock, nl_pid, seq, __dump_class_nlmsg,
-				dump_class_nlmsg, cookie);
-}
-
-static int __dump_qdisc_nlmsg(struct nlmsghdr *nlh,
-			      libbpf_dump_nlmsg_t dump_qdisc_nlmsg,
-			      void *cookie)
-{
-	struct nlattr *tb[TCA_MAX + 1], *attr;
-	struct tcmsg *t = NLMSG_DATA(nlh);
-	int len;
-
-	len = nlh->nlmsg_len - NLMSG_LENGTH(sizeof(*t));
-	attr = (struct nlattr *) ((void *) t + NLMSG_ALIGN(sizeof(*t)));
-	if (libbpf_nla_parse(tb, TCA_MAX, attr, len, NULL) != 0)
-		return -LIBBPF_ERRNO__NLPARSE;
-
-	return dump_qdisc_nlmsg(cookie, t, tb);
-}
-
-int libbpf_nl_get_qdisc(int sock, unsigned int nl_pid, int ifindex,
-			libbpf_dump_nlmsg_t dump_qdisc_nlmsg, void *cookie)
-{
-	struct {
-		struct nlmsghdr nlh;
-		struct tcmsg t;
-	} req = {
-		.nlh.nlmsg_len = NLMSG_LENGTH(sizeof(struct tcmsg)),
-		.nlh.nlmsg_type = RTM_GETQDISC,
-		.nlh.nlmsg_flags = NLM_F_DUMP | NLM_F_REQUEST,
-		.t.tcm_family = AF_UNSPEC,
-		.t.tcm_ifindex = ifindex,
-	};
-	int seq = time(NULL);
-
-	req.nlh.nlmsg_seq = seq;
-	if (send(sock, &req, req.nlh.nlmsg_len, 0) < 0)
-		return -errno;
-
-	return bpf_netlink_recv(sock, nl_pid, seq, __dump_qdisc_nlmsg,
-				dump_qdisc_nlmsg, cookie);
-}
-
-static int __dump_filter_nlmsg(struct nlmsghdr *nlh,
-			       libbpf_dump_nlmsg_t dump_filter_nlmsg,
-			       void *cookie)
-{
-	struct nlattr *tb[TCA_MAX + 1], *attr;
-	struct tcmsg *t = NLMSG_DATA(nlh);
-	int len;
-
-	len = nlh->nlmsg_len - NLMSG_LENGTH(sizeof(*t));
-	attr = (struct nlattr *) ((void *) t + NLMSG_ALIGN(sizeof(*t)));
-	if (libbpf_nla_parse(tb, TCA_MAX, attr, len, NULL) != 0)
-		return -LIBBPF_ERRNO__NLPARSE;
-
-	return dump_filter_nlmsg(cookie, t, tb);
-}
-
-int libbpf_nl_get_filter(int sock, unsigned int nl_pid, int ifindex, int handle,
-			 libbpf_dump_nlmsg_t dump_filter_nlmsg, void *cookie)
-{
-	struct {
-		struct nlmsghdr nlh;
-		struct tcmsg t;
-	} req = {
-		.nlh.nlmsg_len = NLMSG_LENGTH(sizeof(struct tcmsg)),
-		.nlh.nlmsg_type = RTM_GETTFILTER,
-		.nlh.nlmsg_flags = NLM_F_DUMP | NLM_F_REQUEST,
-		.t.tcm_family = AF_UNSPEC,
-		.t.tcm_ifindex = ifindex,
-		.t.tcm_parent = handle,
-	};
-	int seq = time(NULL);
-
-	req.nlh.nlmsg_seq = seq;
-	if (send(sock, &req, req.nlh.nlmsg_len, 0) < 0)
-		return -errno;
-
-	return bpf_netlink_recv(sock, nl_pid, seq, __dump_filter_nlmsg,
-				dump_filter_nlmsg, cookie);
-}
diff --git a/tools/lib/bpf/nlattr.c b/tools/lib/bpf/nlattr.c
index 0ad41dfea8eb..b607fa9852b1 100644
--- a/tools/lib/bpf/nlattr.c
+++ b/tools/lib/bpf/nlattr.c
@@ -7,14 +7,11 @@
  */
 
 #include <errno.h>
-#include "nlattr.h"
-#include "libbpf_internal.h"
-#include <linux/rtnetlink.h>
 #include <string.h>
 #include <stdio.h>
-
-/* make sure libbpf doesn't use kernel-only integer typedefs */
-#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
+#include <linux/rtnetlink.h>
+#include "nlattr.h"
+#include "libbpf_internal.h"
 
 static uint16_t nla_attr_minlen[LIBBPF_NLA_TYPE_MAX+1] = {
 	[LIBBPF_NLA_U8]		= sizeof(uint8_t),
diff --git a/tools/lib/bpf/ringbuf.c b/tools/lib/bpf/ringbuf.c
index 4fc6c6cbb4eb..5c6522c89af1 100644
--- a/tools/lib/bpf/ringbuf.c
+++ b/tools/lib/bpf/ringbuf.c
@@ -16,15 +16,11 @@
 #include <asm/barrier.h>
 #include <sys/mman.h>
 #include <sys/epoll.h>
-#include <tools/libc_compat.h>
 
 #include "libbpf.h"
 #include "libbpf_internal.h"
 #include "bpf.h"
 
-/* make sure libbpf doesn't use kernel-only integer typedefs */
-#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
-
 struct ring {
 	ring_buffer_sample_fn sample_cb;
 	void *ctx;
@@ -82,12 +78,12 @@ int ring_buffer__add(struct ring_buffer *rb, int map_fd,
 		return -EINVAL;
 	}
 
-	tmp = reallocarray(rb->rings, rb->ring_cnt + 1, sizeof(*rb->rings));
+	tmp = libbpf_reallocarray(rb->rings, rb->ring_cnt + 1, sizeof(*rb->rings));
 	if (!tmp)
 		return -ENOMEM;
 	rb->rings = tmp;
 
-	tmp = reallocarray(rb->events, rb->ring_cnt + 1, sizeof(*rb->events));
+	tmp = libbpf_reallocarray(rb->events, rb->ring_cnt + 1, sizeof(*rb->events));
 	if (!tmp)
 		return -ENOMEM;
 	rb->events = tmp;
diff --git a/tools/lib/bpf/xsk.c b/tools/lib/bpf/xsk.c
index f7f4efb70a4c..e3c98c007825 100644
--- a/tools/lib/bpf/xsk.c
+++ b/tools/lib/bpf/xsk.c
@@ -20,6 +20,8 @@
 #include <linux/if_ether.h>
 #include <linux/if_packet.h>
 #include <linux/if_xdp.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
 #include <linux/sockios.h>
 #include <net/if.h>
 #include <sys/ioctl.h>
@@ -32,9 +34,6 @@
 #include "libbpf_internal.h"
 #include "xsk.h"
 
-/* make sure libbpf doesn't use kernel-only integer typedefs */
-#pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
-
 #ifndef SOL_XDP
  #define SOL_XDP 283
 #endif
@@ -48,26 +47,35 @@
 #endif
 
 struct xsk_umem {
-	struct xsk_ring_prod *fill;
-	struct xsk_ring_cons *comp;
+	struct xsk_ring_prod *fill_save;
+	struct xsk_ring_cons *comp_save;
 	char *umem_area;
 	struct xsk_umem_config config;
 	int fd;
 	int refcount;
+	struct list_head ctx_list;
+};
+
+struct xsk_ctx {
+	struct xsk_ring_prod *fill;
+	struct xsk_ring_cons *comp;
+	__u32 queue_id;
+	struct xsk_umem *umem;
+	int refcount;
+	int ifindex;
+	struct list_head list;
+	int prog_fd;
+	int xsks_map_fd;
+	char ifname[IFNAMSIZ];
 };
 
 struct xsk_socket {
 	struct xsk_ring_cons *rx;
 	struct xsk_ring_prod *tx;
 	__u64 outstanding_tx;
-	struct xsk_umem *umem;
+	struct xsk_ctx *ctx;
 	struct xsk_socket_config config;
 	int fd;
-	int ifindex;
-	int prog_fd;
-	int xsks_map_fd;
-	__u32 queue_id;
-	char ifname[IFNAMSIZ];
 };
 
 struct xsk_nl_info {
@@ -203,15 +211,73 @@ static int xsk_get_mmap_offsets(int fd, struct xdp_mmap_offsets *off)
 	return -EINVAL;
 }
 
+static int xsk_create_umem_rings(struct xsk_umem *umem, int fd,
+				 struct xsk_ring_prod *fill,
+				 struct xsk_ring_cons *comp)
+{
+	struct xdp_mmap_offsets off;
+	void *map;
+	int err;
+
+	err = setsockopt(fd, SOL_XDP, XDP_UMEM_FILL_RING,
+			 &umem->config.fill_size,
+			 sizeof(umem->config.fill_size));
+	if (err)
+		return -errno;
+
+	err = setsockopt(fd, SOL_XDP, XDP_UMEM_COMPLETION_RING,
+			 &umem->config.comp_size,
+			 sizeof(umem->config.comp_size));
+	if (err)
+		return -errno;
+
+	err = xsk_get_mmap_offsets(fd, &off);
+	if (err)
+		return -errno;
+
+	map = mmap(NULL, off.fr.desc + umem->config.fill_size * sizeof(__u64),
+		   PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, fd,
+		   XDP_UMEM_PGOFF_FILL_RING);
+	if (map == MAP_FAILED)
+		return -errno;
+
+	fill->mask = umem->config.fill_size - 1;
+	fill->size = umem->config.fill_size;
+	fill->producer = map + off.fr.producer;
+	fill->consumer = map + off.fr.consumer;
+	fill->flags = map + off.fr.flags;
+	fill->ring = map + off.fr.desc;
+	fill->cached_cons = umem->config.fill_size;
+
+	map = mmap(NULL, off.cr.desc + umem->config.comp_size * sizeof(__u64),
+		   PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, fd,
+		   XDP_UMEM_PGOFF_COMPLETION_RING);
+	if (map == MAP_FAILED) {
+		err = -errno;
+		goto out_mmap;
+	}
+
+	comp->mask = umem->config.comp_size - 1;
+	comp->size = umem->config.comp_size;
+	comp->producer = map + off.cr.producer;
+	comp->consumer = map + off.cr.consumer;
+	comp->flags = map + off.cr.flags;
+	comp->ring = map + off.cr.desc;
+
+	return 0;
+
+out_mmap:
+	munmap(map, off.fr.desc + umem->config.fill_size * sizeof(__u64));
+	return err;
+}
+
 int xsk_umem__create_v0_0_4(struct xsk_umem **umem_ptr, void *umem_area,
 			    __u64 size, struct xsk_ring_prod *fill,
 			    struct xsk_ring_cons *comp,
 			    const struct xsk_umem_config *usr_config)
 {
-	struct xdp_mmap_offsets off;
 	struct xdp_umem_reg mr;
 	struct xsk_umem *umem;
-	void *map;
 	int err;
 
 	if (!umem_area || !umem_ptr || !fill || !comp)
@@ -230,6 +296,7 @@ int xsk_umem__create_v0_0_4(struct xsk_umem **umem_ptr, void *umem_area,
 	}
 
 	umem->umem_area = umem_area;
+	INIT_LIST_HEAD(&umem->ctx_list);
 	xsk_set_umem_config(&umem->config, usr_config);
 
 	memset(&mr, 0, sizeof(mr));
@@ -244,71 +311,16 @@ int xsk_umem__create_v0_0_4(struct xsk_umem **umem_ptr, void *umem_area,
 		err = -errno;
 		goto out_socket;
 	}
-	err = setsockopt(umem->fd, SOL_XDP, XDP_UMEM_FILL_RING,
-			 &umem->config.fill_size,
-			 sizeof(umem->config.fill_size));
-	if (err) {
-		err = -errno;
-		goto out_socket;
-	}
-	err = setsockopt(umem->fd, SOL_XDP, XDP_UMEM_COMPLETION_RING,
-			 &umem->config.comp_size,
-			 sizeof(umem->config.comp_size));
-	if (err) {
-		err = -errno;
-		goto out_socket;
-	}
 
-	err = xsk_get_mmap_offsets(umem->fd, &off);
-	if (err) {
-		err = -errno;
-		goto out_socket;
-	}
-
-	map = mmap(NULL, off.fr.desc + umem->config.fill_size * sizeof(__u64),
-		   PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, umem->fd,
-		   XDP_UMEM_PGOFF_FILL_RING);
-	if (map == MAP_FAILED) {
-		err = -errno;
+	err = xsk_create_umem_rings(umem, umem->fd, fill, comp);
+	if (err)
 		goto out_socket;
-	}
-
-	umem->fill = fill;
-	fill->mask = umem->config.fill_size - 1;
-	fill->size = umem->config.fill_size;
-	fill->producer = map + off.fr.producer;
-	fill->consumer = map + off.fr.consumer;
-	fill->flags = map + off.fr.flags;
-	fill->ring = map + off.fr.desc;
-	fill->cached_prod = *fill->producer;
-	/* cached_cons is "size" bigger than the real consumer pointer
-	 * See xsk_prod_nb_free
-	 */
-	fill->cached_cons = *fill->consumer + umem->config.fill_size;
-
-	map = mmap(NULL, off.cr.desc + umem->config.comp_size * sizeof(__u64),
-		   PROT_READ | PROT_WRITE, MAP_SHARED | MAP_POPULATE, umem->fd,
-		   XDP_UMEM_PGOFF_COMPLETION_RING);
-	if (map == MAP_FAILED) {
-		err = -errno;
-		goto out_mmap;
-	}
-
-	umem->comp = comp;
-	comp->mask = umem->config.comp_size - 1;
-	comp->size = umem->config.comp_size;
-	comp->producer = map + off.cr.producer;
-	comp->consumer = map + off.cr.consumer;
-	comp->flags = map + off.cr.flags;
-	comp->ring = map + off.cr.desc;
-	comp->cached_prod = *comp->producer;
-	comp->cached_cons = *comp->consumer;
 
+	umem->fill_save = fill;
+	umem->comp_save = comp;
 	*umem_ptr = umem;
 	return 0;
 
-out_mmap:
-	munmap(map, off.fr.desc + umem->config.fill_size * sizeof(__u64));
 out_socket:
 	close(umem->fd);
 out_umem_alloc:
@@ -342,6 +354,7 @@ DEFAULT_VERSION(xsk_umem__create_v0_0_4, xsk_umem__create, LIBBPF_0.0.4)
 static int xsk_load_xdp_prog(struct xsk_socket *xsk)
 {
 	static const int log_buf_size = 16 * 1024;
+	struct xsk_ctx *ctx = xsk->ctx;
 	char log_buf[log_buf_size];
 	int err, prog_fd;
 
@@ -369,7 +382,7 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk)
 		/* *(u32 *)(r10 - 4) = r2 */
 		BPF_STX_MEM(BPF_W, BPF_REG_10, BPF_REG_2, -4),
 		/* r1 = xskmap[] */
-		BPF_LD_MAP_FD(BPF_REG_1, xsk->xsks_map_fd),
+		BPF_LD_MAP_FD(BPF_REG_1, ctx->xsks_map_fd),
 		/* r3 = XDP_PASS */
 		BPF_MOV64_IMM(BPF_REG_3, 2),
 		/* call bpf_redirect_map */
@@ -381,7 +394,7 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk)
 		/* r2 += -4 */
 		BPF_ALU64_IMM(BPF_ADD, BPF_REG_2, -4),
 		/* r1 = xskmap[] */
-		BPF_LD_MAP_FD(BPF_REG_1, xsk->xsks_map_fd),
+		BPF_LD_MAP_FD(BPF_REG_1, ctx->xsks_map_fd),
 		/* call bpf_map_lookup_elem */
 		BPF_EMIT_CALL(BPF_FUNC_map_lookup_elem),
 		/* r1 = r0 */
@@ -393,7 +406,7 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk)
 		/* r2 = *(u32 *)(r10 - 4) */
 		BPF_LDX_MEM(BPF_W, BPF_REG_2, BPF_REG_10, -4),
 		/* r1 = xskmap[] */
-		BPF_LD_MAP_FD(BPF_REG_1, xsk->xsks_map_fd),
+		BPF_LD_MAP_FD(BPF_REG_1, ctx->xsks_map_fd),
 		/* r3 = 0 */
 		BPF_MOV64_IMM(BPF_REG_3, 0),
 		/* call bpf_redirect_map */
@@ -411,19 +424,21 @@ static int xsk_load_xdp_prog(struct xsk_socket *xsk)
 		return prog_fd;
 	}
 
-	err = bpf_set_link_xdp_fd(xsk->ifindex, prog_fd, xsk->config.xdp_flags);
+	err = bpf_set_link_xdp_fd(xsk->ctx->ifindex, prog_fd,
+				  xsk->config.xdp_flags);
 	if (err) {
 		close(prog_fd);
 		return err;
 	}
 
-	xsk->prog_fd = prog_fd;
+	ctx->prog_fd = prog_fd;
 	return 0;
 }
 
 static int xsk_get_max_queues(struct xsk_socket *xsk)
 {
 	struct ethtool_channels channels = { .cmd = ETHTOOL_GCHANNELS };
+	struct xsk_ctx *ctx = xsk->ctx;
 	struct ifreq ifr = {};
 	int fd, err, ret;
 
@@ -432,7 +447,7 @@ static int xsk_get_max_queues(struct xsk_socket *xsk)
 		return -errno;
 
 	ifr.ifr_data = (void *)&channels;
-	memcpy(ifr.ifr_name, xsk->ifname, IFNAMSIZ - 1);
+	memcpy(ifr.ifr_name, ctx->ifname, IFNAMSIZ - 1);
 	ifr.ifr_name[IFNAMSIZ - 1] = '\0';
 	err = ioctl(fd, SIOCETHTOOL, &ifr);
 	if (err && errno != EOPNOTSUPP) {
@@ -460,6 +475,7 @@ out:
 
 static int xsk_create_bpf_maps(struct xsk_socket *xsk)
 {
+	struct xsk_ctx *ctx = xsk->ctx;
 	int max_queues;
 	int fd;
 
@@ -472,15 +488,17 @@ static int xsk_create_bpf_maps(struct xsk_socket *xsk)
 	if (fd < 0)
 		return fd;
 
-	xsk->xsks_map_fd = fd;
+	ctx->xsks_map_fd = fd;
 
 	return 0;
 }
 
 static void xsk_delete_bpf_maps(struct xsk_socket *xsk)
 {
-	bpf_map_delete_elem(xsk->xsks_map_fd, &xsk->queue_id);
-	close(xsk->xsks_map_fd);
+	struct xsk_ctx *ctx = xsk->ctx;
+
+	bpf_map_delete_elem(ctx->xsks_map_fd, &ctx->queue_id);
+	close(ctx->xsks_map_fd);
 }
 
 static int xsk_lookup_bpf_maps(struct xsk_socket *xsk)
@@ -488,10 +506,11 @@ static int xsk_lookup_bpf_maps(struct xsk_socket *xsk)
 	__u32 i, *map_ids, num_maps, prog_len = sizeof(struct bpf_prog_info);
 	__u32 map_len = sizeof(struct bpf_map_info);
 	struct bpf_prog_info prog_info = {};
+	struct xsk_ctx *ctx = xsk->ctx;
 	struct bpf_map_info map_info;
 	int fd, err;
 
-	err = bpf_obj_get_info_by_fd(xsk->prog_fd, &prog_info, &prog_len);
+	err = bpf_obj_get_info_by_fd(ctx->prog_fd, &prog_info, &prog_len);
 	if (err)
 		return err;
 
@@ -505,11 +524,11 @@ static int xsk_lookup_bpf_maps(struct xsk_socket *xsk)
 	prog_info.nr_map_ids = num_maps;
 	prog_info.map_ids = (__u64)(unsigned long)map_ids;
 
-	err = bpf_obj_get_info_by_fd(xsk->prog_fd, &prog_info, &prog_len);
+	err = bpf_obj_get_info_by_fd(ctx->prog_fd, &prog_info, &prog_len);
 	if (err)
 		goto out_map_ids;
 
-	xsk->xsks_map_fd = -1;
+	ctx->xsks_map_fd = -1;
 
 	for (i = 0; i < prog_info.nr_map_ids; i++) {
 		fd = bpf_map_get_fd_by_id(map_ids[i]);
@@ -523,7 +542,7 @@ static int xsk_lookup_bpf_maps(struct xsk_socket *xsk)
 		}
 
 		if (!strcmp(map_info.name, "xsks_map")) {
-			xsk->xsks_map_fd = fd;
+			ctx->xsks_map_fd = fd;
 			continue;
 		}
 
@@ -531,7 +550,7 @@ static int xsk_lookup_bpf_maps(struct xsk_socket *xsk)
 	}
 
 	err = 0;
-	if (xsk->xsks_map_fd == -1)
+	if (ctx->xsks_map_fd == -1)
 		err = -ENOENT;
 
 out_map_ids:
@@ -541,16 +560,19 @@ out_map_ids:
 
 static int xsk_set_bpf_maps(struct xsk_socket *xsk)
 {
-	return bpf_map_update_elem(xsk->xsks_map_fd, &xsk->queue_id,
+	struct xsk_ctx *ctx = xsk->ctx;
+
+	return bpf_map_update_elem(ctx->xsks_map_fd, &ctx->queue_id,
 				   &xsk->fd, 0);
 }
 
 static int xsk_setup_xdp_prog(struct xsk_socket *xsk)
 {
+	struct xsk_ctx *ctx = xsk->ctx;
 	__u32 prog_id = 0;
 	int err;
 
-	err = bpf_get_link_xdp_id(xsk->ifindex, &prog_id,
+	err = bpf_get_link_xdp_id(ctx->ifindex, &prog_id,
 				  xsk->config.xdp_flags);
 	if (err)
 		return err;
@@ -566,12 +588,12 @@ static int xsk_setup_xdp_prog(struct xsk_socket *xsk)
 			return err;
 		}
 	} else {
-		xsk->prog_fd = bpf_prog_get_fd_by_id(prog_id);
-		if (xsk->prog_fd < 0)
+		ctx->prog_fd = bpf_prog_get_fd_by_id(prog_id);
+		if (ctx->prog_fd < 0)
 			return -errno;
 		err = xsk_lookup_bpf_maps(xsk);
 		if (err) {
-			close(xsk->prog_fd);
+			close(ctx->prog_fd);
 			return err;
 		}
 	}
@@ -580,23 +602,108 @@ static int xsk_setup_xdp_prog(struct xsk_socket *xsk)
 		err = xsk_set_bpf_maps(xsk);
 	if (err) {
 		xsk_delete_bpf_maps(xsk);
-		close(xsk->prog_fd);
+		close(ctx->prog_fd);
 		return err;
 	}
 
 	return 0;
 }
 
-int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
-		       __u32 queue_id, struct xsk_umem *umem,
-		       struct xsk_ring_cons *rx, struct xsk_ring_prod *tx,
-		       const struct xsk_socket_config *usr_config)
+static struct xsk_ctx *xsk_get_ctx(struct xsk_umem *umem, int ifindex,
+				   __u32 queue_id)
+{
+	struct xsk_ctx *ctx;
+
+	if (list_empty(&umem->ctx_list))
+		return NULL;
+
+	list_for_each_entry(ctx, &umem->ctx_list, list) {
+		if (ctx->ifindex == ifindex && ctx->queue_id == queue_id) {
+			ctx->refcount++;
+			return ctx;
+		}
+	}
+
+	return NULL;
+}
+
+static void xsk_put_ctx(struct xsk_ctx *ctx)
+{
+	struct xsk_umem *umem = ctx->umem;
+	struct xdp_mmap_offsets off;
+	int err;
+
+	if (--ctx->refcount == 0) {
+		err = xsk_get_mmap_offsets(umem->fd, &off);
+		if (!err) {
+			munmap(ctx->fill->ring - off.fr.desc,
+			       off.fr.desc + umem->config.fill_size *
+			       sizeof(__u64));
+			munmap(ctx->comp->ring - off.cr.desc,
+			       off.cr.desc + umem->config.comp_size *
+			       sizeof(__u64));
+		}
+
+		list_del(&ctx->list);
+		free(ctx);
+	}
+}
+
+static struct xsk_ctx *xsk_create_ctx(struct xsk_socket *xsk,
+				      struct xsk_umem *umem, int ifindex,
+				      const char *ifname, __u32 queue_id,
+				      struct xsk_ring_prod *fill,
+				      struct xsk_ring_cons *comp)
+{
+	struct xsk_ctx *ctx;
+	int err;
+
+	ctx = calloc(1, sizeof(*ctx));
+	if (!ctx)
+		return NULL;
+
+	if (!umem->fill_save) {
+		err = xsk_create_umem_rings(umem, xsk->fd, fill, comp);
+		if (err) {
+			free(ctx);
+			return NULL;
+		}
+	} else if (umem->fill_save != fill || umem->comp_save != comp) {
+		/* Copy over rings to new structs. */
+		memcpy(fill, umem->fill_save, sizeof(*fill));
+		memcpy(comp, umem->comp_save, sizeof(*comp));
+	}
+
+	ctx->ifindex = ifindex;
+	ctx->refcount = 1;
+	ctx->umem = umem;
+	ctx->queue_id = queue_id;
+	memcpy(ctx->ifname, ifname, IFNAMSIZ - 1);
+	ctx->ifname[IFNAMSIZ - 1] = '\0';
+
+	umem->fill_save = NULL;
+	umem->comp_save = NULL;
+	ctx->fill = fill;
+	ctx->comp = comp;
+	list_add(&ctx->list, &umem->ctx_list);
+	return ctx;
+}
+
+int xsk_socket__create_shared(struct xsk_socket **xsk_ptr,
+			      const char *ifname,
+			      __u32 queue_id, struct xsk_umem *umem,
+			      struct xsk_ring_cons *rx,
+			      struct xsk_ring_prod *tx,
+			      struct xsk_ring_prod *fill,
+			      struct xsk_ring_cons *comp,
+			      const struct xsk_socket_config *usr_config)
 {
 	void *rx_map = NULL, *tx_map = NULL;
 	struct sockaddr_xdp sxdp = {};
 	struct xdp_mmap_offsets off;
 	struct xsk_socket *xsk;
-	int err;
+	struct xsk_ctx *ctx;
+	int err, ifindex;
 
 	if (!umem || !xsk_ptr || !(rx || tx))
 		return -EFAULT;
@@ -609,10 +716,10 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 	if (err)
 		goto out_xsk_alloc;
 
-	if (umem->refcount &&
-	    !(xsk->config.libbpf_flags & XSK_LIBBPF_FLAGS__INHIBIT_PROG_LOAD)) {
-		pr_warn("Error: shared umems not supported by libbpf supplied XDP program.\n");
-		err = -EBUSY;
+	xsk->outstanding_tx = 0;
+	ifindex = if_nametoindex(ifname);
+	if (!ifindex) {
+		err = -errno;
 		goto out_xsk_alloc;
 	}
 
@@ -626,16 +733,21 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 		xsk->fd = umem->fd;
 	}
 
-	xsk->outstanding_tx = 0;
-	xsk->queue_id = queue_id;
-	xsk->umem = umem;
-	xsk->ifindex = if_nametoindex(ifname);
-	if (!xsk->ifindex) {
-		err = -errno;
-		goto out_socket;
+	ctx = xsk_get_ctx(umem, ifindex, queue_id);
+	if (!ctx) {
+		if (!fill || !comp) {
+			err = -EFAULT;
+			goto out_socket;
+		}
+
+		ctx = xsk_create_ctx(xsk, umem, ifindex, ifname, queue_id,
+				     fill, comp);
+		if (!ctx) {
+			err = -ENOMEM;
+			goto out_socket;
+		}
 	}
-	memcpy(xsk->ifname, ifname, IFNAMSIZ - 1);
-	xsk->ifname[IFNAMSIZ - 1] = '\0';
+	xsk->ctx = ctx;
 
 	if (rx) {
 		err = setsockopt(xsk->fd, SOL_XDP, XDP_RX_RING,
@@ -643,7 +755,7 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 				 sizeof(xsk->config.rx_size));
 		if (err) {
 			err = -errno;
-			goto out_socket;
+			goto out_put_ctx;
 		}
 	}
 	if (tx) {
@@ -652,14 +764,14 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 				 sizeof(xsk->config.tx_size));
 		if (err) {
 			err = -errno;
-			goto out_socket;
+			goto out_put_ctx;
 		}
 	}
 
 	err = xsk_get_mmap_offsets(xsk->fd, &off);
 	if (err) {
 		err = -errno;
-		goto out_socket;
+		goto out_put_ctx;
 	}
 
 	if (rx) {
@@ -669,7 +781,7 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 			      xsk->fd, XDP_PGOFF_RX_RING);
 		if (rx_map == MAP_FAILED) {
 			err = -errno;
-			goto out_socket;
+			goto out_put_ctx;
 		}
 
 		rx->mask = xsk->config.rx_size - 1;
@@ -708,10 +820,10 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 	xsk->tx = tx;
 
 	sxdp.sxdp_family = PF_XDP;
-	sxdp.sxdp_ifindex = xsk->ifindex;
-	sxdp.sxdp_queue_id = xsk->queue_id;
+	sxdp.sxdp_ifindex = ctx->ifindex;
+	sxdp.sxdp_queue_id = ctx->queue_id;
 	if (umem->refcount > 1) {
-		sxdp.sxdp_flags = XDP_SHARED_UMEM;
+		sxdp.sxdp_flags |= XDP_SHARED_UMEM;
 		sxdp.sxdp_shared_umem_fd = umem->fd;
 	} else {
 		sxdp.sxdp_flags = xsk->config.bind_flags;
@@ -723,7 +835,7 @@ int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
 		goto out_mmap_tx;
 	}
 
-	xsk->prog_fd = -1;
+	ctx->prog_fd = -1;
 
 	if (!(xsk->config.libbpf_flags & XSK_LIBBPF_FLAGS__INHIBIT_PROG_LOAD)) {
 		err = xsk_setup_xdp_prog(xsk);
@@ -742,6 +854,8 @@ out_mmap_rx:
 	if (rx)
 		munmap(rx_map, off.rx.desc +
 		       xsk->config.rx_size * sizeof(struct xdp_desc));
+out_put_ctx:
+	xsk_put_ctx(ctx);
 out_socket:
 	if (--umem->refcount)
 		close(xsk->fd);
@@ -750,25 +864,24 @@ out_xsk_alloc:
 	return err;
 }
 
-int xsk_umem__delete(struct xsk_umem *umem)
+int xsk_socket__create(struct xsk_socket **xsk_ptr, const char *ifname,
+		       __u32 queue_id, struct xsk_umem *umem,
+		       struct xsk_ring_cons *rx, struct xsk_ring_prod *tx,
+		       const struct xsk_socket_config *usr_config)
 {
-	struct xdp_mmap_offsets off;
-	int err;
+	return xsk_socket__create_shared(xsk_ptr, ifname, queue_id, umem,
+					 rx, tx, umem->fill_save,
+					 umem->comp_save, usr_config);
+}
 
+int xsk_umem__delete(struct xsk_umem *umem)
+{
 	if (!umem)
 		return 0;
 
 	if (umem->refcount)
 		return -EBUSY;
 
-	err = xsk_get_mmap_offsets(umem->fd, &off);
-	if (!err) {
-		munmap(umem->fill->ring - off.fr.desc,
-		       off.fr.desc + umem->config.fill_size * sizeof(__u64));
-		munmap(umem->comp->ring - off.cr.desc,
-		       off.cr.desc + umem->config.comp_size * sizeof(__u64));
-	}
-
 	close(umem->fd);
 	free(umem);
 
@@ -778,15 +891,16 @@ int xsk_umem__delete(struct xsk_umem *umem)
 void xsk_socket__delete(struct xsk_socket *xsk)
 {
 	size_t desc_sz = sizeof(struct xdp_desc);
+	struct xsk_ctx *ctx = xsk->ctx;
 	struct xdp_mmap_offsets off;
 	int err;
 
 	if (!xsk)
 		return;
 
-	if (xsk->prog_fd != -1) {
+	if (ctx->prog_fd != -1) {
 		xsk_delete_bpf_maps(xsk);
-		close(xsk->prog_fd);
+		close(ctx->prog_fd);
 	}
 
 	err = xsk_get_mmap_offsets(xsk->fd, &off);
@@ -799,14 +913,15 @@ void xsk_socket__delete(struct xsk_socket *xsk)
 			munmap(xsk->tx->ring - off.tx.desc,
 			       off.tx.desc + xsk->config.tx_size * desc_sz);
 		}
-
 	}
 
-	xsk->umem->refcount--;
+	xsk_put_ctx(ctx);
+
+	ctx->umem->refcount--;
 	/* Do not close an fd that also has an associated umem connected
 	 * to it.
 	 */
-	if (xsk->fd != xsk->umem->fd)
+	if (xsk->fd != ctx->umem->fd)
 		close(xsk->fd);
 	free(xsk);
 }
diff --git a/tools/lib/bpf/xsk.h b/tools/lib/bpf/xsk.h
index 584f6820a639..1069c46364ff 100644
--- a/tools/lib/bpf/xsk.h
+++ b/tools/lib/bpf/xsk.h
@@ -234,6 +234,15 @@ LIBBPF_API int xsk_socket__create(struct xsk_socket **xsk,
 				  struct xsk_ring_cons *rx,
 				  struct xsk_ring_prod *tx,
 				  const struct xsk_socket_config *config);
+LIBBPF_API int
+xsk_socket__create_shared(struct xsk_socket **xsk_ptr,
+			  const char *ifname,
+			  __u32 queue_id, struct xsk_umem *umem,
+			  struct xsk_ring_cons *rx,
+			  struct xsk_ring_prod *tx,
+			  struct xsk_ring_prod *fill,
+			  struct xsk_ring_cons *comp,
+			  const struct xsk_socket_config *config);
 
 /* Returns 0 for success and -EBUSY if the umem is still in use. */
 LIBBPF_API int xsk_umem__delete(struct xsk_umem *umem);