This patch adds new config option to trigger generation of BTF type information from DWARF debuginfo for vmlinux and kernel modules through pahole, which in turn relies on libbpf for btf_dedup() algorithm. The intent is to record compact type information of all types used inside kernel, including all the structs/unions/typedefs/etc. This enables BPF's compile-once-run-everywhere ([0]) approach, in which tracing programs that are inspecting kernel's internal data (e.g., struct task_struct) can be compiled on a system running some kernel version, but would be possible to run on other kernel versions (and configurations) without recompilation, even if the layout of structs changed and/or some of the fields were added, removed, or renamed. This is only possible if BPF loader can get kernel type info to adjust all the offsets correctly. This patch is a first time in this direction, making sure that BTF type info is part of Linux kernel image in non-loadable ELF section. BTF deduplication ([1]) algorithm typically provides 100x savings compared to DWARF data, so resulting .BTF section is not big as is typically about 2MB in size. [0] http://vger.kernel.org/lpc-bpf2018.html#session-2 [1] https://facebookmicrosites.github.io/bpf/blog/2018/11/14/btf-enhancement.html Cc: Masahiro Yamada <yamada.masahiro@socionext.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Alexei Starovoitov <ast@fb.com> Cc: Yonghong Song <yhs@fb.com> Cc: Martin KaFai Lau <kafai@fb.com> Change-Id: Id935f6e3ac658d9a92d55acd39e8287c644c941d Signed-off-by: Andrii Nakryiko <andriin@fb.com> Acked-by: David S. Miller <davem@davemloft.net> Acked-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
495 lines
13 KiB
Bash
495 lines
13 KiB
Bash
#!/bin/sh
|
|
# SPDX-License-Identifier: GPL-2.0
|
|
#
|
|
# link vmlinux
|
|
#
|
|
# vmlinux is linked from the objects selected by $(KBUILD_VMLINUX_INIT) and
|
|
# $(KBUILD_VMLINUX_MAIN) and $(KBUILD_VMLINUX_LIBS). Most are built-in.o files
|
|
# from top-level directories in the kernel tree, others are specified in
|
|
# arch/$(ARCH)/Makefile. Ordering when linking is important, and
|
|
# $(KBUILD_VMLINUX_INIT) must be first. $(KBUILD_VMLINUX_LIBS) are archives
|
|
# which are linked conditionally (not within --whole-archive), and do not
|
|
# require symbol indexes added.
|
|
#
|
|
# vmlinux
|
|
# ^
|
|
# |
|
|
# +-< $(KBUILD_VMLINUX_INIT)
|
|
# | +--< init/version.o + more
|
|
# |
|
|
# +--< $(KBUILD_VMLINUX_MAIN)
|
|
# | +--< drivers/built-in.o mm/built-in.o + more
|
|
# |
|
|
# +--< $(KBUILD_VMLINUX_LIBS)
|
|
# | +--< lib/lib.a + more
|
|
# |
|
|
# +-< ${kallsymso} (see description in KALLSYMS section)
|
|
#
|
|
# vmlinux version (uname -v) cannot be updated during normal
|
|
# descending-into-subdirs phase since we do not yet know if we need to
|
|
# update vmlinux.
|
|
# Therefore this step is delayed until just before final link of vmlinux.
|
|
#
|
|
# System.map is generated to document addresses of all kernel symbols
|
|
|
|
# Error out on error
|
|
set -e
|
|
|
|
# Nice output in kbuild format
|
|
# Will be supressed by "make -s"
|
|
info()
|
|
{
|
|
if [ "${quiet}" != "silent_" ]; then
|
|
printf " %-7s %s\n" "${1}" "${2}"
|
|
fi
|
|
}
|
|
|
|
# Thin archive build here makes a final archive with symbol table and indexes
|
|
# from vmlinux objects INIT and MAIN, which can be used as input to linker.
|
|
# KBUILD_VMLINUX_LIBS archives should already have symbol table and indexes
|
|
# added.
|
|
#
|
|
# Traditional incremental style of link does not require this step
|
|
#
|
|
# built-in.o output file
|
|
#
|
|
archive_builtin()
|
|
{
|
|
if [ -n "${CONFIG_THIN_ARCHIVES}" ]; then
|
|
info AR built-in.o
|
|
rm -f built-in.o;
|
|
${AR} rcsTP${KBUILD_ARFLAGS} built-in.o \
|
|
${KBUILD_VMLINUX_INIT} \
|
|
${KBUILD_VMLINUX_MAIN}
|
|
|
|
if [ -n "${CONFIG_LTO_CLANG}" ]; then
|
|
mv -f built-in.o built-in.o.tmp
|
|
${LLVM_AR} rcsT${KBUILD_ARFLAGS} built-in.o $(${AR} t built-in.o.tmp)
|
|
rm -f built-in.o.tmp
|
|
fi
|
|
fi
|
|
}
|
|
|
|
# If CONFIG_LTO_CLANG is selected, generate a linker script to ensure correct
|
|
# ordering of initcalls, and with CONFIG_MODVERSIONS also enabled, collect the
|
|
# previously generated symbol versions into the same script.
|
|
lto_lds()
|
|
{
|
|
if [ -z "${CONFIG_LTO_CLANG}" ]; then
|
|
return
|
|
fi
|
|
|
|
${srctree}/scripts/generate_initcall_order.pl \
|
|
built-in.o ${KBUILD_VMLINUX_LIBS} \
|
|
> .tmp_lto.lds
|
|
|
|
if [ -n "${CONFIG_MODVERSIONS}" ]; then
|
|
for a in built-in.o ${KBUILD_VMLINUX_LIBS}; do
|
|
for o in $(${AR} t $a); do
|
|
if [ -f ${o}.symversions ]; then
|
|
cat ${o}.symversions >> .tmp_lto.lds
|
|
fi
|
|
done
|
|
done
|
|
fi
|
|
|
|
echo "-T .tmp_lto.lds"
|
|
}
|
|
|
|
# Link of vmlinux.o used for section mismatch analysis
|
|
# ${1} output file
|
|
modpost_link()
|
|
{
|
|
local objects
|
|
|
|
if [ -n "${CONFIG_THIN_ARCHIVES}" ]; then
|
|
objects="--whole-archive \
|
|
built-in.o \
|
|
--no-whole-archive \
|
|
--start-group \
|
|
${KBUILD_VMLINUX_LIBS} \
|
|
--end-group"
|
|
else
|
|
objects="${KBUILD_VMLINUX_INIT} \
|
|
--start-group \
|
|
${KBUILD_VMLINUX_MAIN} \
|
|
${KBUILD_VMLINUX_LIBS} \
|
|
--end-group"
|
|
fi
|
|
|
|
if [ -n "${CONFIG_LTO_CLANG}" ]; then
|
|
# This might take a while, so indicate that we're doing
|
|
# an LTO link
|
|
info LTO vmlinux.o
|
|
else
|
|
info LD vmlinux.o
|
|
fi
|
|
|
|
${LD} ${LDFLAGS} -r -o ${1} $(lto_lds) ${objects}
|
|
}
|
|
|
|
# If CONFIG_LTO_CLANG is selected, we postpone running recordmcount until
|
|
# we have compiled LLVM IR to an object file.
|
|
recordmcount()
|
|
{
|
|
if [ -z "${CONFIG_LTO_CLANG}" ]; then
|
|
return
|
|
fi
|
|
|
|
if [ -n "${CONFIG_FTRACE_MCOUNT_RECORD}" ]; then
|
|
scripts/recordmcount ${RECORDMCOUNT_FLAGS} $*
|
|
fi
|
|
}
|
|
|
|
# Link of vmlinux
|
|
# ${1} - optional extra .o files
|
|
# ${2} - output file
|
|
vmlinux_link()
|
|
{
|
|
local lds="${objtree}/${KBUILD_LDS}"
|
|
local objects
|
|
|
|
if [ "${SRCARCH}" != "um" ]; then
|
|
local ld=${LD}
|
|
local ldflags="${LDFLAGS} ${LDFLAGS_vmlinux}"
|
|
|
|
if [ -n "${LDFINAL_vmlinux}" ]; then
|
|
ld=${LDFINAL_vmlinux}
|
|
ldflags="${LDFLAGS_FINAL_vmlinux} ${LDFLAGS_vmlinux}"
|
|
fi
|
|
|
|
if [[ -n "${CONFIG_THIN_ARCHIVES}" && -z "${CONFIG_LTO_CLANG}" ]]; then
|
|
objects="--whole-archive \
|
|
built-in.o \
|
|
--no-whole-archive \
|
|
--start-group \
|
|
${KBUILD_VMLINUX_LIBS} \
|
|
--end-group \
|
|
${1}"
|
|
else
|
|
objects="${KBUILD_VMLINUX_INIT} \
|
|
--start-group \
|
|
${KBUILD_VMLINUX_MAIN} \
|
|
${KBUILD_VMLINUX_LIBS} \
|
|
--end-group \
|
|
${1}"
|
|
fi
|
|
|
|
${ld} ${ldflags} -o ${2} -T ${lds} ${objects}
|
|
else
|
|
if [ -n "${CONFIG_THIN_ARCHIVES}" ]; then
|
|
objects="-Wl,--whole-archive \
|
|
built-in.o \
|
|
-Wl,--no-whole-archive \
|
|
-Wl,--start-group \
|
|
${KBUILD_VMLINUX_LIBS} \
|
|
-Wl,--end-group \
|
|
${1}"
|
|
else
|
|
objects="${KBUILD_VMLINUX_INIT} \
|
|
-Wl,--start-group \
|
|
${KBUILD_VMLINUX_MAIN} \
|
|
${KBUILD_VMLINUX_LIBS} \
|
|
-Wl,--end-group \
|
|
${1}"
|
|
fi
|
|
|
|
${CC} ${CFLAGS_vmlinux} -o ${2} \
|
|
-Wl,-T,${lds} \
|
|
${objects} \
|
|
-lutil -lrt -lpthread
|
|
rm -f linux
|
|
fi
|
|
}
|
|
|
|
# generate .BTF typeinfo from DWARF debuginfo
|
|
gen_btf()
|
|
{
|
|
local pahole_ver;
|
|
|
|
pahole_ver=$(${PAHOLE} --version | sed -E 's/v([0-9]+)\.([0-9]+)/\1\2/')
|
|
if [ "${pahole_ver}" -lt "113" ]; then
|
|
info "BTF" "${1}: pahole version $(${PAHOLE} --version) is too old, need at least v1.13"
|
|
exit 0
|
|
fi
|
|
|
|
info "BTF" ${1}
|
|
LLVM_OBJCOPY=${OBJCOPY} ${PAHOLE} -J ${1}
|
|
}
|
|
|
|
# Create ${2} .o file with all symbols from the ${1} object file
|
|
kallsyms()
|
|
{
|
|
info KSYM ${2}
|
|
local kallsymopt;
|
|
|
|
if [ -n "${CONFIG_HAVE_UNDERSCORE_SYMBOL_PREFIX}" ]; then
|
|
kallsymopt="${kallsymopt} --symbol-prefix=_"
|
|
fi
|
|
|
|
if [ -n "${CONFIG_KALLSYMS_ALL}" ]; then
|
|
kallsymopt="${kallsymopt} --all-symbols"
|
|
fi
|
|
|
|
if [ -n "${CONFIG_KALLSYMS_ABSOLUTE_PERCPU}" ]; then
|
|
kallsymopt="${kallsymopt} --absolute-percpu"
|
|
fi
|
|
|
|
if [ -n "${CONFIG_KALLSYMS_BASE_RELATIVE}" ]; then
|
|
kallsymopt="${kallsymopt} --base-relative"
|
|
fi
|
|
|
|
local aflags="${KBUILD_AFLAGS} ${KBUILD_AFLAGS_KERNEL} \
|
|
${NOSTDINC_FLAGS} ${LINUXINCLUDE} ${KBUILD_CPPFLAGS}"
|
|
|
|
local afile="`basename ${2} .o`.S"
|
|
|
|
${NM} -n ${1} | scripts/kallsyms ${kallsymopt} > ${afile}
|
|
${CC} ${aflags} -c -o ${2} ${afile}
|
|
}
|
|
|
|
# Generates ${2} .o file with RTIC MP's from the ${1} object file (vmlinux)
|
|
# ${3} the file name where the sizes of the RTIC MP structure are stored
|
|
# just in case, save copy of the RTIC mp to ${4}
|
|
# Note: RTIC_MPGEN has to be set if MPGen is available
|
|
rtic_mp()
|
|
{
|
|
# assume that RTIC_MP_O generation may fail
|
|
RTIC_MP_O=
|
|
|
|
local aflags="${KBUILD_AFLAGS} ${KBUILD_AFLAGS_KERNEL} \
|
|
${NOSTDINC_FLAGS} ${LINUXINCLUDE} ${KBUILD_CPPFLAGS}"
|
|
|
|
${RTIC_MPGEN} --objcopy="${OBJCOPY}" --objdump="${OBJDUMP}" \
|
|
--binpath='' --vmlinux=${1} --config=${KCONFIG_CONFIG} && \
|
|
cat rtic_mp.c | ${CC} ${aflags} -c -o ${2} -x c - && \
|
|
cp rtic_mp.c ${4} && \
|
|
${NM} --print-size --size-sort ${2} > ${3} && \
|
|
RTIC_MP_O=${2} || echo “RTIC MP generation has failed”
|
|
# NM - save generated variable sizes for verification
|
|
# RTIC_MP_O is our retval - great success if set to generated .o file
|
|
# Echo statement above prints the error message in case any of the
|
|
# above RTIC MP generation commands fail and it ensures rtic mp failure
|
|
# does not cause kernel compilation to fail.
|
|
}
|
|
|
|
# Create map file with all symbols from ${1}
|
|
# See mksymap for additional details
|
|
mksysmap()
|
|
{
|
|
${CONFIG_SHELL} "${srctree}/scripts/mksysmap" ${1} ${2}
|
|
}
|
|
|
|
sortextable()
|
|
{
|
|
${objtree}/scripts/sortextable ${1}
|
|
}
|
|
|
|
# Delete output files in case of error
|
|
cleanup()
|
|
{
|
|
rm -f .old_version
|
|
rm -f .tmp_System.map
|
|
rm -f .tmp_kallsyms*
|
|
rm -f .tmp_version
|
|
rm -f .tmp_lto.lds
|
|
rm -f .tmp_vmlinux*
|
|
rm -f built-in.o
|
|
rm -f System.map
|
|
rm -f vmlinux
|
|
rm -f vmlinux.o
|
|
rm -f .tmp_rtic_mp_sz*
|
|
rm -f rtic_mp.*
|
|
}
|
|
|
|
on_exit()
|
|
{
|
|
if [ $? -ne 0 ]; then
|
|
cleanup
|
|
fi
|
|
}
|
|
trap on_exit EXIT
|
|
|
|
on_signals()
|
|
{
|
|
exit 1
|
|
}
|
|
trap on_signals HUP INT QUIT TERM
|
|
|
|
#
|
|
#
|
|
# Use "make V=1" to debug this script
|
|
case "${KBUILD_VERBOSE}" in
|
|
*1*)
|
|
set -x
|
|
;;
|
|
esac
|
|
|
|
if [ "$1" = "clean" ]; then
|
|
cleanup
|
|
exit 0
|
|
fi
|
|
|
|
# We need access to CONFIG_ symbols
|
|
case "${KCONFIG_CONFIG}" in
|
|
*/*)
|
|
. "${KCONFIG_CONFIG}"
|
|
;;
|
|
*)
|
|
# Force using a file from the current directory
|
|
. "./${KCONFIG_CONFIG}"
|
|
esac
|
|
|
|
# Update version
|
|
info GEN .version
|
|
if [ ! -r .version ]; then
|
|
rm -f .version;
|
|
echo 1 >.version;
|
|
else
|
|
mv .version .old_version;
|
|
expr 0$(cat .old_version) + 1 >.version;
|
|
fi;
|
|
|
|
# final build of init/
|
|
${MAKE} -f "${srctree}/scripts/Makefile.build" obj=init GCC_PLUGINS_CFLAGS="${GCC_PLUGINS_CFLAGS}"
|
|
|
|
archive_builtin
|
|
|
|
#link vmlinux.o
|
|
modpost_link vmlinux.o
|
|
|
|
# modpost vmlinux.o to check for section mismatches
|
|
${MAKE} -f "${srctree}/scripts/Makefile.modpost" vmlinux.o
|
|
|
|
if [ -n "${CONFIG_LTO_CLANG}" ]; then
|
|
# Re-use vmlinux.o, so we can avoid the slow LTO link step in
|
|
# vmlinux_link
|
|
KBUILD_VMLINUX_INIT=
|
|
KBUILD_VMLINUX_MAIN=vmlinux.o
|
|
KBUILD_VMLINUX_LIBS=
|
|
|
|
# Call recordmcount if needed
|
|
recordmcount vmlinux.o
|
|
fi
|
|
|
|
# Generate RTIC MP placeholder compile unit of the correct size
|
|
# and add it to the list of link objects
|
|
# this needs to be done before generating kallsyms
|
|
if [ ! -z ${RTIC_MPGEN+x} ]; then
|
|
rtic_mp vmlinux.o rtic_mp.o .tmp_rtic_mp_sz1 .tmp_rtic_mp1.c
|
|
KBUILD_VMLINUX_LIBS+=" "
|
|
KBUILD_VMLINUX_LIBS+=$RTIC_MP_O
|
|
fi
|
|
|
|
kallsymso=""
|
|
kallsyms_vmlinux=""
|
|
if [ -n "${CONFIG_KALLSYMS}" ]; then
|
|
|
|
# kallsyms support
|
|
# Generate section listing all symbols and add it into vmlinux
|
|
# It's a three step process:
|
|
# 1) Link .tmp_vmlinux1 so it has all symbols and sections,
|
|
# but __kallsyms is empty.
|
|
# Running kallsyms on that gives us .tmp_kallsyms1.o with
|
|
# the right size
|
|
# 2) Link .tmp_vmlinux2 so it now has a __kallsyms section of
|
|
# the right size, but due to the added section, some
|
|
# addresses have shifted.
|
|
# From here, we generate a correct .tmp_kallsyms2.o
|
|
# 3) That link may have expanded the kernel image enough that
|
|
# more linker branch stubs / trampolines had to be added, which
|
|
# introduces new names, which further expands kallsyms. Do another
|
|
# pass if that is the case. In theory it's possible this results
|
|
# in even more stubs, but unlikely.
|
|
# KALLSYMS_EXTRA_PASS=1 may also used to debug or work around
|
|
# other bugs.
|
|
# 4) The correct ${kallsymso} is linked into the final vmlinux.
|
|
#
|
|
# a) Verify that the System.map from vmlinux matches the map from
|
|
# ${kallsymso}.
|
|
|
|
kallsymso=.tmp_kallsyms2.o
|
|
kallsyms_vmlinux=.tmp_vmlinux2
|
|
|
|
# step 1
|
|
vmlinux_link "" .tmp_vmlinux1
|
|
kallsyms .tmp_vmlinux1 .tmp_kallsyms1.o
|
|
|
|
# step 2
|
|
vmlinux_link .tmp_kallsyms1.o .tmp_vmlinux2
|
|
kallsyms .tmp_vmlinux2 .tmp_kallsyms2.o
|
|
|
|
# step 3
|
|
size1=$(stat -c "%s" .tmp_kallsyms1.o)
|
|
size2=$(stat -c "%s" .tmp_kallsyms2.o)
|
|
|
|
if [ $size1 -ne $size2 ] || [ -n "${KALLSYMS_EXTRA_PASS}" ]; then
|
|
kallsymso=.tmp_kallsyms3.o
|
|
kallsyms_vmlinux=.tmp_vmlinux3
|
|
|
|
vmlinux_link .tmp_kallsyms2.o .tmp_vmlinux3
|
|
|
|
kallsyms .tmp_vmlinux3 .tmp_kallsyms3.o
|
|
fi
|
|
fi
|
|
|
|
# Update RTIC MP object by replacing the place holder
|
|
# with actual MP data of the same size
|
|
# Also double check that object size did not change
|
|
# Note: Check initilally if RTIC_MP_O is not empty or uninitialized,
|
|
# as incase RTIC_MPGEN is set and failure occurs in RTIC_MP_O
|
|
# generation, below check for comparing object sizes fails
|
|
# due to an empty RTIC_MP_O object.
|
|
if [ ! -z ${RTIC_MP_O} ]; then
|
|
rtic_mp "${kallsyms_vmlinux}" rtic_mp.o .tmp_rtic_mp_sz2 \
|
|
.tmp_rtic_mp2.c
|
|
if ! cmp -s .tmp_rtic_mp_sz1 .tmp_rtic_mp_sz2; then
|
|
echo >&2 'ERROR: RTIC MP object files size mismatch'
|
|
exit 1
|
|
fi
|
|
fi
|
|
|
|
info LD vmlinux
|
|
vmlinux_link "${kallsymso}" vmlinux
|
|
|
|
if [ -n "${CONFIG_DEBUG_INFO_BTF}" ]; then
|
|
gen_btf vmlinux
|
|
fi
|
|
|
|
if [ -n "${CONFIG_BUILDTIME_EXTABLE_SORT}" ]; then
|
|
info SORTEX vmlinux
|
|
sortextable vmlinux
|
|
fi
|
|
|
|
info SYSMAP System.map
|
|
mksysmap vmlinux System.map
|
|
|
|
# step a (see comment above)
|
|
if [ -n "${CONFIG_KALLSYMS}" ]; then
|
|
mksysmap ${kallsyms_vmlinux} .tmp_System.map
|
|
|
|
if ! cmp -s System.map .tmp_System.map; then
|
|
echo >&2 Inconsistent kallsyms data
|
|
echo >&2 Try "make KALLSYMS_EXTRA_PASS=1" as a workaround
|
|
exit 1
|
|
fi
|
|
fi
|
|
|
|
# Starting Android Q, the DTB's are part of dtb.img and not part
|
|
# of the kernel image. RTIC DTS relies on the kernel environment
|
|
# and could not build outside of the kernel. Generate RTIC DTS after
|
|
# successful kernel build if MPGen is enabled. The DTB will be
|
|
# generated with dtb.img in kernel_definitions.mk.
|
|
if [ ! -z ${RTIC_MPGEN+x} ]; then
|
|
${RTIC_MPGEN} --objcopy="${OBJCOPY}" --objdump="${OBJDUMP}" \
|
|
--binpath="" --vmlinux="vmlinux" --config=${KCONFIG_CONFIG} \
|
|
--cc="${CC} ${KBUILD_AFLAGS}" --dts=rtic_mp.dts \
|
|
|| echo “RTIC MP DTS generation has failed”
|
|
# Echo statement above prints the error message in case above
|
|
# RTIC MP DTS generation command fails and it ensures rtic mp
|
|
# failure does not cause kernel compilation to fail.
|
|
fi
|
|
|
|
# We made a new kernel - delete old version file
|
|
rm -f .old_version
|