resolve_btfids: Change in-place update with raw binary output

Currently resolve_btfids updates .BTF_ids section of an ELF file
in-place, based on the contents of provided BTF, usually within the
same input file, and optionally a BTF base.

Change resolve_btfids behavior to enable BTF transformations as part
of its main operation. To achieve this, in-place ELF write in
resolve_btfids is replaced with generation of the following binaries:
  * ${1}.BTF with .BTF section data
  * ${1}.BTF_ids with .BTF_ids section data if it existed in ${1}
  * ${1}.BTF.base with .BTF.base section data for out-of-tree modules

The execution of resolve_btfids and consumption of its output is
orchestrated by scripts/gen-btf.sh introduced in this patch.

The motivation for emitting binary data is that it allows simplifying
resolve_btfids implementation by delegating ELF update to the $OBJCOPY
tool [1], which is already widely used across the codebase.

There are two distinct paths for BTF generation and resolve_btfids
application in the kernel build: for vmlinux and for kernel modules.

For the vmlinux binary a .BTF section is added in a roundabout way to
ensure correct linking. The patch doesn't change this approach, only
the implementation is a little different.

Before this patch it worked as follows:

  * pahole consumed .tmp_vmlinux1 [2] and added .BTF section with
    llvm-objcopy [3] to it
  * then everything except the .BTF section was stripped from .tmp_vmlinux1
    into a .tmp_vmlinux1.bpf.o object [2], later linked into vmlinux
  * resolve_btfids was executed later on vmlinux.unstripped [4],
    updating it in-place

After this patch gen-btf.sh implements the following:

  * pahole consumes .tmp_vmlinux1 and produces a *detached* file with
    raw BTF data
  * resolve_btfids consumes .tmp_vmlinux1 and detached BTF to produce
    (potentially modified) .BTF, and .BTF_ids sections data
  * a .tmp_vmlinux1.bpf.o object is then produced with objcopy copying
    BTF output of resolve_btfids
  * .BTF_ids data gets embedded into vmlinux.unstripped in
    link-vmlinux.sh by objcopy --update-section

For kernel modules, creating a special .bpf.o file is not necessary,
and so embedding of sections data produced by resolve_btfids is
straightforward with objcopy.

With this patch an ELF file becomes effectively read-only within
resolve_btfids, which allows deleting elf_update() call and satellite
code (like compressed_section_fix [5]).

Endianness handling of .BTF_ids data is also changed. Previously the
"flags" part of the section was bswapped in sets_patch() [6], and then
Elf_Type was modified before elf_update() to signal to libelf that
bswap may be necessary. With this patch we explicitly bswap entire
data buffer on load and on dump.

[1] https://lore.kernel.org/bpf/131b4190-9c49-4f79-a99d-c00fac97fa44@linux.dev/
[2] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/scripts/link-vmlinux.sh?h=v6.18#n110
[3] https://git.kernel.org/pub/scm/devel/pahole/pahole.git/tree/btf_encoder.c?h=v1.31#n1803
[4] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/scripts/link-vmlinux.sh?h=v6.18#n284
[5] https://lore.kernel.org/bpf/20200819092342.259004-1-jolsa@kernel.org/
[6] https://lore.kernel.org/bpf/cover.1707223196.git.vmalik@redhat.com/

Signed-off-by: Ihor Solodrai <ihor.solodrai@linux.dev>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Tested-by: Alan Maguire <alan.maguire@oracle.com>
Acked-by: Eduard Zingerman <eddyz87@gmail.com>
Link: https://lore.kernel.org/bpf/20251219181825.1289460-3-ihor.solodrai@linux.dev

5 files changed, 173 insertions, 45 deletions

diff --git a/scripts/Makefile.btf b/scripts/Makefile.btf
index 840a55de42da..562a04b40e06 100644
--- a/scripts/Makefile.btf
+++ b/scripts/Makefile.btf
@@ -18,13 +18,15 @@ pahole-flags-$(call test-ge, $(pahole-ver), 126)  = -j$(JOBS) --btf_features=enc
 
 pahole-flags-$(call test-ge, $(pahole-ver), 130) += --btf_features=attributes
 
-ifneq ($(KBUILD_EXTMOD),)
-module-pahole-flags-$(call test-ge, $(pahole-ver), 128) += --btf_features=distilled_base
-endif
-
 endif
 
 pahole-flags-$(CONFIG_PAHOLE_HAS_LANG_EXCLUDE)		+= --lang_exclude=rust
 
 export PAHOLE_FLAGS := $(pahole-flags-y)
-export MODULE_PAHOLE_FLAGS := $(module-pahole-flags-y)
+
+resolve-btfids-flags-y :=
+resolve-btfids-flags-$(CONFIG_WERROR) += --fatal_warnings
+resolve-btfids-flags-$(if $(KBUILD_EXTMOD),y) += --distill_base
+resolve-btfids-flags-$(if $(KBUILD_VERBOSE),y) += --verbose
+
+export RESOLVE_BTFIDS_FLAGS := $(resolve-btfids-flags-y)
diff --git a/scripts/Makefile.modfinal b/scripts/Makefile.modfinal
index 149e12ff5700..422c56dc878e 100644
--- a/scripts/Makefile.modfinal
+++ b/scripts/Makefile.modfinal
@@ -42,9 +42,8 @@ quiet_cmd_btf_ko = BTF [M] $@
       cmd_btf_ko = 							\
 	if [ ! -f $(objtree)/vmlinux ]; then				\
 		printf "Skipping BTF generation for %s due to unavailability of vmlinux\n" $@ 1>&2; \
-	else								\
-		LLVM_OBJCOPY="$(OBJCOPY)" $(PAHOLE) -J $(PAHOLE_FLAGS) $(MODULE_PAHOLE_FLAGS) --btf_base $(objtree)/vmlinux $@; \
-		$(RESOLVE_BTFIDS) -b $(objtree)/vmlinux $@;		\
+	else	\
+		$(srctree)/scripts/gen-btf.sh --btf_base $(objtree)/vmlinux $@; \
 	fi;
 
 # Same as newer-prereqs, but allows to exclude specified extra dependencies
diff --git a/scripts/Makefile.vmlinux b/scripts/Makefile.vmlinux
index cd788cac9d91..20a988f4fe0c 100644
--- a/scripts/Makefile.vmlinux
+++ b/scripts/Makefile.vmlinux
@@ -71,7 +71,7 @@ targets += vmlinux.unstripped .vmlinux.export.o
 vmlinux.unstripped: scripts/link-vmlinux.sh vmlinux.o .vmlinux.export.o $(KBUILD_LDS) FORCE
 	+$(call if_changed_dep,link_vmlinux)
 ifdef CONFIG_DEBUG_INFO_BTF
-vmlinux.unstripped: $(RESOLVE_BTFIDS)
+vmlinux.unstripped: $(RESOLVE_BTFIDS) $(srctree)/scripts/gen-btf.sh
 endif
 
 ifdef CONFIG_BUILDTIME_TABLE_SORT
diff --git a/scripts/gen-btf.sh b/scripts/gen-btf.sh
new file mode 100755
index 000000000000..06c6d8becaa2
--- /dev/null
+++ b/scripts/gen-btf.sh
@@ -0,0 +1,157 @@
+#!/bin/bash
+# SPDX-License-Identifier: GPL-2.0
+# Copyright (c) 2025 Meta Platforms, Inc. and affiliates.
+#
+# This script generates BTF data for the provided ELF file.
+#
+# Kernel BTF generation involves these conceptual steps:
+#   1. pahole generates BTF from DWARF data
+#   2. resolve_btfids applies kernel-specific btf2btf
+#      transformations and computes data for .BTF_ids section
+#   3. the result gets linked/objcopied into the target binary
+#
+# How step (3) should be done differs between vmlinux, and
+# kernel modules, which is the primary reason for the existence
+# of this script.
+#
+# For modules the script expects vmlinux passed in as --btf_base.
+# Generated .BTF, .BTF.base and .BTF_ids sections become embedded
+# into the input ELF file with objcopy.
+#
+# For vmlinux the input file remains unchanged and two files are produced:
+#   - ${1}.btf.o ready for linking into vmlinux
+#   - ${1}.BTF_ids with .BTF_ids data blob
+# This output is consumed by scripts/link-vmlinux.sh
+
+set -e
+
+usage()
+{
+	echo "Usage: $0 [--btf_base <file>] <target ELF file>"
+	exit 1
+}
+
+BTF_BASE=""
+
+while [ $# -gt 0 ]; do
+	case "$1" in
+	--btf_base)
+		BTF_BASE="$2"
+		shift 2
+		;;
+	-*)
+		echo "Unknown option: $1" >&2
+		usage
+		;;
+	*)
+		break
+		;;
+	esac
+done
+
+if [ $# -ne 1 ]; then
+	usage
+fi
+
+ELF_FILE="$1"
+shift
+
+is_enabled() {
+	grep -q "^$1=y" ${objtree}/include/config/auto.conf
+}
+
+info()
+{
+	printf "  %-7s %s\n" "${1}" "${2}"
+}
+
+case "${KBUILD_VERBOSE}" in
+*1*)
+	set -x
+	;;
+esac
+
+
+gen_btf_data()
+{
+	info BTF "${ELF_FILE}"
+	btf1="${ELF_FILE}.BTF.1"
+	${PAHOLE} -J ${PAHOLE_FLAGS}			\
+		${BTF_BASE:+--btf_base ${BTF_BASE}}	\
+		--btf_encode_detached=${btf1}		\
+		"${ELF_FILE}"
+
+	info BTFIDS "${ELF_FILE}"
+	${RESOLVE_BTFIDS} ${RESOLVE_BTFIDS_FLAGS}	\
+		${BTF_BASE:+--btf_base ${BTF_BASE}}	\
+		--btf ${btf1} "${ELF_FILE}"
+}
+
+gen_btf_o()
+{
+	local btf_data=${ELF_FILE}.btf.o
+
+	# Create ${btf_data} which contains just .BTF section but no symbols. Add
+	# SHF_ALLOC because .BTF will be part of the vmlinux image. --strip-all
+	# deletes all symbols including __start_BTF and __stop_BTF, which will
+	# be redefined in the linker script.
+	info OBJCOPY "${btf_data}"
+	echo "" | ${CC} ${CLANG_FLAGS} -c -x c -o ${btf_data} -
+	${OBJCOPY} --add-section .BTF=${ELF_FILE}.BTF \
+		--set-section-flags .BTF=alloc,readonly ${btf_data}
+	${OBJCOPY} --only-section=.BTF --strip-all ${btf_data}
+
+	# Change e_type to ET_REL so that it can be used to link final vmlinux.
+	# GNU ld 2.35+ and lld do not allow an ET_EXEC input.
+	if is_enabled CONFIG_CPU_BIG_ENDIAN; then
+		et_rel='\0\1'
+	else
+		et_rel='\1\0'
+	fi
+	printf "${et_rel}" | dd of="${btf_data}" conv=notrunc bs=1 seek=16 status=none
+}
+
+embed_btf_data()
+{
+	info OBJCOPY "${ELF_FILE}.BTF"
+	${OBJCOPY} --add-section .BTF=${ELF_FILE}.BTF ${ELF_FILE}
+
+	# a module might not have a .BTF_ids or .BTF.base section
+	local btf_base="${ELF_FILE}.BTF.base"
+	if [ -f "${btf_base}" ]; then
+		${OBJCOPY} --add-section .BTF.base=${btf_base} ${ELF_FILE}
+	fi
+	local btf_ids="${ELF_FILE}.BTF_ids"
+	if [ -f "${btf_ids}" ]; then
+		${OBJCOPY} --update-section .BTF_ids=${btf_ids} ${ELF_FILE}
+	fi
+}
+
+cleanup()
+{
+	rm -f "${ELF_FILE}.BTF.1"
+	rm -f "${ELF_FILE}.BTF"
+	if [ "${BTFGEN_MODE}" = "module" ]; then
+		rm -f "${ELF_FILE}.BTF.base"
+		rm -f "${ELF_FILE}.BTF_ids"
+	fi
+}
+trap cleanup EXIT
+
+BTFGEN_MODE="vmlinux"
+if [ -n "${BTF_BASE}" ]; then
+	BTFGEN_MODE="module"
+fi
+
+gen_btf_data
+
+case "${BTFGEN_MODE}" in
+vmlinux)
+	gen_btf_o
+	;;
+module)
+	embed_btf_data
+	;;
+esac
+
+exit 0
diff --git a/scripts/link-vmlinux.sh b/scripts/link-vmlinux.sh
index 4ab44c73da4d..e2207e612ac3 100755
--- a/scripts/link-vmlinux.sh
+++ b/scripts/link-vmlinux.sh
@@ -106,34 +106,6 @@ vmlinux_link()
 		${kallsymso} ${btf_vmlinux_bin_o} ${arch_vmlinux_o} ${ldlibs}
 }
 
-# generate .BTF typeinfo from DWARF debuginfo
-# ${1} - vmlinux image
-gen_btf()
-{
-	local btf_data=${1}.btf.o
-
-	info BTF "${btf_data}"
-	LLVM_OBJCOPY="${OBJCOPY}" ${PAHOLE} -J ${PAHOLE_FLAGS} ${1}
-
-	# Create ${btf_data} which contains just .BTF section but no symbols. Add
-	# SHF_ALLOC because .BTF will be part of the vmlinux image. --strip-all
-	# deletes all symbols including __start_BTF and __stop_BTF, which will
-	# be redefined in the linker script. Add 2>/dev/null to suppress GNU
-	# objcopy warnings: "empty loadable segment detected at ..."
-	${OBJCOPY} --only-section=.BTF --set-section-flags .BTF=alloc,readonly \
-		--strip-all ${1} "${btf_data}" 2>/dev/null
-	# Change e_type to ET_REL so that it can be used to link final vmlinux.
-	# GNU ld 2.35+ and lld do not allow an ET_EXEC input.
-	if is_enabled CONFIG_CPU_BIG_ENDIAN; then
-		et_rel='\0\1'
-	else
-		et_rel='\1\0'
-	fi
-	printf "${et_rel}" | dd of="${btf_data}" conv=notrunc bs=1 seek=16 status=none
-
-	btf_vmlinux_bin_o=${btf_data}
-}
-
 # Create ${2}.o file with all symbols from the ${1} object file
 kallsyms()
 {
@@ -205,6 +177,7 @@ if is_enabled CONFIG_ARCH_WANTS_PRE_LINK_VMLINUX; then
 fi
 
 btf_vmlinux_bin_o=
+btfids_vmlinux=
 kallsymso=
 strip_debug=
 generate_map=
@@ -232,11 +205,13 @@ if is_enabled CONFIG_KALLSYMS || is_enabled CONFIG_DEBUG_INFO_BTF; then
 fi
 
 if is_enabled CONFIG_DEBUG_INFO_BTF; then
-	if ! gen_btf .tmp_vmlinux1; then
+	if ! ${srctree}/scripts/gen-btf.sh .tmp_vmlinux1; then
 		echo >&2 "Failed to generate BTF for vmlinux"
 		echo >&2 "Try to disable CONFIG_DEBUG_INFO_BTF"
 		exit 1
 	fi
+	btf_vmlinux_bin_o=.tmp_vmlinux1.btf.o
+	btfids_vmlinux=.tmp_vmlinux1.BTF_ids
 fi
 
 if is_enabled CONFIG_KALLSYMS; then
@@ -289,14 +264,9 @@ fi
 
 vmlinux_link "${VMLINUX}"
 
-# fill in BTF IDs
 if is_enabled CONFIG_DEBUG_INFO_BTF; then
-	info BTFIDS "${VMLINUX}"
-	RESOLVE_BTFIDS_ARGS=""
-	if is_enabled CONFIG_WERROR; then
-		RESOLVE_BTFIDS_ARGS=" --fatal_warnings "
-	fi
-	${RESOLVE_BTFIDS} ${RESOLVE_BTFIDS_ARGS} "${VMLINUX}"
+	info OBJCOPY ${btfids_vmlinux}
+	${OBJCOPY} --update-section .BTF_ids=${btfids_vmlinux} ${VMLINUX}
 fi
 
 mksysmap "${VMLINUX}" System.map