From: Nathan Chancellor <nathan@kernel.org>
To: Luis Augenstein <luis.augenstein@tngtech.com>
Cc: nsc@kernel.org, linux-kbuild@vger.kernel.org,
linux-kernel@vger.kernel.org, akpm@linux-foundation.org,
gregkh@linuxfoundation.org, kstewart@linuxfoundation.org,
maximilian.huber@tngtech.com
Subject: Re: [PATCH 02/15] scripts/sbom: integrate script in make process
Date: Mon, 30 Mar 2026 11:50:11 +0200 [thread overview]
Message-ID: <20260330095011.GA1458050@ax162> (raw)
In-Reply-To: <20260210205424.11195-3-luis.augenstein@tngtech.com>
Hi Luis,
On Tue, Feb 10, 2026 at 09:54:11PM +0100, Luis Augenstein wrote:
> integrate SBOM script into the kernel build process.
>
> Assisted-by: Claude Sonnet 4.5
> Assisted-by: GLM-4.7
> Co-developed-by: Maximilian Huber <maximilian.huber@tngtech.com>
> Signed-off-by: Maximilian Huber <maximilian.huber@tngtech.com>
> Signed-off-by: Luis Augenstein <luis.augenstein@tngtech.com>
...
> diff --git a/Makefile b/Makefile
> index 9d38125263fb..46d4be490d7f 100644
> --- a/Makefile
> +++ b/Makefile
> @@ -772,7 +772,7 @@ endif
> # in addition to whatever we do anyway.
> # Just "make" or "make all" shall build modules as well
>
> -ifneq ($(filter all modules nsdeps compile_commands.json clang-%,$(MAKECMDGOALS)),)
> +ifneq ($(filter all modules nsdeps compile_commands.json clang-% sbom,$(MAKECMDGOALS)),)
> KBUILD_MODULES := y
> endif
>
> @@ -1612,7 +1612,7 @@ CLEAN_FILES += vmlinux.symvers modules-only.symvers \
> modules.builtin.ranges vmlinux.o.map vmlinux.unstripped \
> compile_commands.json rust/test \
> rust-project.json .vmlinux.objs .vmlinux.export.c \
> - .builtin-dtbs-list .builtin-dtb.S
> + .builtin-dtbs-list .builtin-dtb.S sbom-*.spdx.json
Does sbom-roots.txt need to be cleaned up as well?
A note for Greg: this will conflict with commit a76e30c2479c ("kbuild:
Delete .builtin-dtbs.S when running make clean").
> # Directories & files removed with 'make mrproper'
> MRPROPER_FILES += include/config include/generated \
> @@ -1728,6 +1728,7 @@ help:
> @echo ''
> @echo 'Tools:'
> @echo ' nsdeps - Generate missing symbol namespace dependencies'
> + @echo ' sbom - Generate Software Bill of Materials'
> @echo ''
> @echo 'Kernel selftest:'
> @echo ' kselftest - Build and run kernel selftest'
> @@ -2108,6 +2109,12 @@ nsdeps: export KBUILD_NSDEPS=1
> nsdeps: modules
> $(Q)$(CONFIG_SHELL) $(srctree)/scripts/nsdeps
>
> +# Script to generate .spdx.json SBOM documents describing the build
> +# ---------------------------------------------------------------------------
> +PHONY += sbom
> +sbom: all
> + $(Q)$(MAKE) $(build)=scripts/sbom
> +
> # Clang Tooling
> # ---------------------------------------------------------------------------
>
> diff --git a/scripts/sbom/Makefile b/scripts/sbom/Makefile
> new file mode 100644
> index 000000000000..6c8ec7356293
> --- /dev/null
> +++ b/scripts/sbom/Makefile
> @@ -0,0 +1,33 @@
> +# SPDX-License-Identifier: GPL-2.0-only OR MIT
> +# Copyright (C) 2025 TNG Technology Consulting GmbH
> +
> +SBOM_SOURCE_FILE := $(objtree)/sbom-source.spdx.json
> +SBOM_BUILD_FILE := $(objtree)/sbom-build.spdx.json
> +SBOM_OUTPUT_FILE := $(objtree)/sbom-output.spdx.json
> +SBOM_ROOTS_FILE := $(objtree)/sbom-roots.txt
> +
> +
> +ifeq ($(srctree),$(objtree))
> + SBOM_TARGETS := $(SBOM_BUILD_FILE) $(SBOM_OUTPUT_FILE)
> +else
> + SBOM_TARGETS := $(SBOM_SOURCE_FILE) $(SBOM_BUILD_FILE) $(SBOM_OUTPUT_FILE)
> +endif
This might be easier to read if it were
ifeq ($(srctree),$(objtree))
SBOM_TARGETS := $(SBOM_SOURCE_FILE)
endif
SBOM_TARGETS += $(SBOM_BUILD_FILE) $(SBOM_OUTPUT_FILE)
> +SBOM_DEPS := $(objtree)/$(KBUILD_IMAGE) $(objtree)/include/generated/autoconf.h
> +ifdef CONFIG_MODULES
> + SBOM_DEPS += $(objtree)/modules.order
> +endif
Is $(objtree)/ necessary on these? You might be able to turn this into
SBOM_DEPS := $(KBUILD_IMAGE) include/generated/autoconf.h $(if $(CONFIG_MODULES),modules.order)
> +$(SBOM_TARGETS) &: $(SBOM_DEPS)
&: is a feature from make 4.3 but the kernel supports make 4.0 and
newer, so you should probably make this
# Use
# $(SBOM_TARGETS) &: $(SBOM_DEPS)
# when make 4.3 is the minimum.
$(SBOM_BUILD_FILE): $(SBOM_DEPS)
$(obj)/: $(SBOM_BUILD_FILE)
> + $(Q)echo " GEN $(notdir $(SBOM_TARGETS))"
> + $(Q)if [ "$(CONFIG_MODULES)" = "y" ]; then \
> + sed 's/\.o$$/.ko/' $(objtree)/modules.order >> $(SBOM_ROOTS_FILE); \
> + fi
> +
> + $(Q)$(PYTHON3) $(srctree)/scripts/sbom/sbom.py
> +
> + $(Q)rm $(SBOM_ROOTS_FILE)
I think this would look better if it used the standard quiet_cmd_ and
cmd_ syntax:
quiet_cmd_sbom = GEN $(notdir $(SBOM_TARGETS))
cmd_sbom = printf "%s\n" "$(KBUILD_IMAGE)" >$(SBOM_ROOTS_FILE); \
$(if $(CONFIG_MODULES),sed 's/\.o$$/.ko/' $(objtree)/modules.order >> $(SBOM_ROOTS_FILE);) \
$(PYTHON3) $(srctree)/scripts/sbom/sbom.py \
--src-tree $(abspath $(srctree)) \
--obj-tree $(abspath $(objtree)) \
--roots-file $(SBOM_ROOTS_FILE) \
--output-directory $(abspath $(objtree)) \
--generate-spdx \
--package-license "GPL-2.0 WITH Linux-syscall-note" \
--package-version "$(KERNELVERSION)"; \
rm $(SBOM_ROOTS_FILE)
$(SBOM_BUILD_FILE): $(SBOM_DEPS)
$(call cmd,sbom)
> +$(obj)/: $(SBOM_TARGETS)
It seems like this could use the standard always-y Kbuild syntax but I
am not sure. Honestly, looking at the bigger picture, it feels like most
of this Makefile could be moved into the main Makefile under the sbom
target? scripts/sbom is not actually used as an output directory and the
generated files do not really need to be listed as targets since their
names are hardcoded in scripts/sbom/sbom/config.py?
# Script to generate .spdx.json SBOM documents describing the build
# ---------------------------------------------------------------------------
ifdef building_out_of_srctree
sbom_targets := sbom-source.spdx.json
endif
sbom_targets += sbom-build.spdx.json sbom-output.spdx.json
sbom_roots_file := $(objtree)/sbom-roots.txt
quiet_cmd_sbom = GEN $(notdir $(sbom_targets))
cmd_sbom = printf "%s\n" "$(KBUILD_IMAGE)" >$(sbom_roots_file); \
$(if $(CONFIG_MODULES),sed 's/\.o$$/.ko/' $(objtree)/modules.order >> $(sbom_roots_file);) \
$(PYTHON3) $(srctree)/scripts/sbom/sbom.py \
--src-tree $(abspath $(srctree)) \
--obj-tree $(abspath $(objtree)) \
--roots-file $(sbom_roots_file) \
--output-directory $(abspath $(objtree)) \
--generate-spdx \
--package-license "GPL-2.0 WITH Linux-syscall-note" \
--package-version "$(KERNELVERSION)"; \
rm $(sbom_roots_file)
PHONY += sbom
sbom: $(notdir $(KBUILD_IMAGE)) include/generated/autoconf.h $(if $(CONFIG_MODULES),modules modules.order)
$(call cmd,sbom)
appears to do the right thing for me?
FWIW, I get errors like
$ make -kj"$(nproc)" ARCH=arm64 CROSS_COMPILE=aarch64-linux- O=build mrproper defconfig sbom
...
GEN sbom-source.spdx.json sbom-build.spdx.json sbom-output.spdx.json
[ERROR] File "/src/scripts/sbom/sbom/cmd_graph/savedcmd_parser.py", line 630, in log_error_or_warning
Skipped parsing command ccache aarch64-linux-gcc ... -o init/main.o /src/init/main.c because no matching parser was found
[ERROR] File "/src/scripts/sbom/sbom/cmd_graph/savedcmd_parser.py", line 630, in log_error_or_warning
Skipped parsing command ccache aarch64-linux-gcc ... -o arch/arm64/kernel/asm-offsets.s /src/arch/arm64/kernel/asm-offsets.c because no matching parser was found
[ERROR] File "/src/scripts/sbom/sbom/cmd_graph/savedcmd_parser.py", line 630, in log_error_or_warning
Skipped parsing command ccache aarch64-linux-gcc ... -o kernel/bounds.s /src/kernel/bounds.c because no matching parser was found
... (Found 10435 more instances of this error)
when testing the whole series without any modifications, am I doing
something wrong?
Cheers,
Nathan
next prev parent reply other threads:[~2026-03-30 9:50 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-10 20:54 [PATCH v4 00/15] add SPDX SBOM generation script Luis Augenstein
2026-02-10 20:54 ` [PATCH 01/15] scripts/sbom: add documentation Luis Augenstein
2026-02-10 20:54 ` [PATCH 02/15] scripts/sbom: integrate script in make process Luis Augenstein
2026-03-30 9:50 ` Nathan Chancellor [this message]
2026-02-10 20:54 ` [PATCH 03/15] scripts/sbom: setup sbom logging Luis Augenstein
2026-02-10 20:54 ` [PATCH 04/15] scripts/sbom: add command parsers Luis Augenstein
2026-02-10 20:54 ` [PATCH 05/15] scripts/sbom: add cmd graph generation Luis Augenstein
2026-02-10 20:54 ` [PATCH 06/15] scripts/sbom: add additional dependency sources for cmd graph Luis Augenstein
2026-02-10 20:54 ` [PATCH 07/15] scripts/sbom: add SPDX classes Luis Augenstein
2026-02-10 20:54 ` [PATCH 08/15] scripts/sbom: add JSON-LD serialization Luis Augenstein
2026-02-10 20:54 ` [PATCH 09/15] scripts/sbom: add shared SPDX elements Luis Augenstein
2026-02-10 20:54 ` [PATCH 10/15] scripts/sbom: collect file metadata Luis Augenstein
2026-02-10 20:54 ` [PATCH 11/15] scripts/sbom: add SPDX output graph Luis Augenstein
2026-02-10 20:54 ` [PATCH 12/15] scripts/sbom: add SPDX source graph Luis Augenstein
2026-02-10 20:54 ` [PATCH 13/15] scripts/sbom: add SPDX build graph Luis Augenstein
2026-02-10 20:54 ` [PATCH 14/15] scripts/sbom: add unit tests for command parsers Luis Augenstein
2026-02-10 20:54 ` [PATCH 15/15] scripts/sbom: add unit tests for SPDX-License-Identifier parsing Luis Augenstein
2026-03-23 13:39 ` [PATCH v4 00/15] add SPDX SBOM generation script Greg KH
2026-03-29 6:29 ` Greg KH
2026-03-30 5:50 ` Nathan Chancellor
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20260330095011.GA1458050@ax162 \
--to=nathan@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=gregkh@linuxfoundation.org \
--cc=kstewart@linuxfoundation.org \
--cc=linux-kbuild@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luis.augenstein@tngtech.com \
--cc=maximilian.huber@tngtech.com \
--cc=nsc@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox