From 034dd340b08be1f2f0477ad16131d609f9dbd53c Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 30 Aug 2026 09:22:00 -0700 Subject: Merge tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace Pull tracing fixes from Steven Rostedt: - Fix error output of boot instance creation failure Currently if a boot instance creation fails, instead of printing out the name of the instance that failed, it prints "(null)". That is because it prints "cur_str" that had already been processed by strsep(). Print the saved name instead. While at it, print the error code of the failure. - Fix use-after-free for same named historgrams Histograms can be named so that they can be used in multiple events. But if the named histogram has a variable attached, the second event that uses the named histogram which duplicates it and needs to free the original after duplication leaves the old variable in place and still visible. If another histogram uses than variable, it will use the stale one which will try to reference the freed duplicate histogram and crash the kernel. Free the duplicate variables along with the duplicated histogram data. - Check return value of kthread_run() in event self test The events self tests uses a kthread for testing but does not check if it succeeded in creating a kthread. If the kthread creation were to fail, the code will still try to call kthread_stop() on the error returned. - Fix race between reading trace_pipe and updating subbuffer size If a user is reading the trace_pipe file at the same time they update the ring buffer sub-buffer size, can cause the trace_pipe read to read stale data. Add trace_access_lock() around updating the ring buffer sub-buffer size. - Fix eventfs_inode on failure path in creation of the events directory In the creation of the "events" directory, if after allocating the eventfs_inode a failure is detected, it calls cleanup_ei() which calls free_ei(). The free_ei() will test if eventfs_inode being freed has no children. It is a bug if it does. But on the failure case of the creation of the "events" directory, the children lists have not yet been initialized and the free will trigger a warning because list_empty() on an uninitialized list returns false. Move the initialization into init_ei() where it makes more sense and makes sure that a created eventfs_inode has its lists initialized upon creation. - Check return value of kthread_run() in ftrace direct sample code The sample code that shows how to use the ftrace direct calls does not test the return of kthread_run() to see if it succeeds. Return a failure if the kthread_run() doesn't succeed. - Clear user events state on fork in case of alloc failure On fork, the child gets a pointer to the parent's user events state. It makes a copy of it then updates the child's pointer to it. But if the allocation fails, the duplication function leaves the child with a pointer to its parent's descriptor. When the child cleans up its data, it will free the parent's descriptor while the parent is still using it. In the duplication function, set the child's user_event_mm to NULL before testing if the allocation succeeded, and when it exits it will not free the parent's descriptor. - Fix retry exhaustion in simple ring buffer reader swap simple_ring_buffer_swap_reader_page() starts with retry set to 8 and post-decrements it only after a failed link replacement. On the final attempt, a successful replacement leaves retry at zero, while a failed replacement leaves it at -1. But the check for success expects the retry value to be non-zero and exits with an error on zero. This is the opposite result. Fix it. - Fail nicely when the remote swap_reader_page() returns an error Currently, if the swap_reader_page() of a remote buffer fails, it triggers a WARN_ON_ONCE() and continues normally. Instead, have it exit with an error and a pr_warn() print instead of a full WARNING. * tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: ring-buffer: Stop remote reader update when page swap fails tracing: Fix retry exhaustion in simple ring buffer reader swap tracing/user_events: Clear copied tracing state before fork duplication samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-multi-modify samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-modify eventfs: Initialize ei->children and ei->list in init_ei() tracing: Fix use-after-free in trace_pipe read on sub-buffer order change tracing: Fix crash passing ERR_PTR to kthread_stop() tracing: Fix use-after-free with same-name named triggers tracing: Fix logged instance name on creation failure --- Documentation/tools/sbom/sbom.rst | 206 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 206 insertions(+) create mode 100644 Documentation/tools/sbom/sbom.rst (limited to 'Documentation/tools/sbom') diff --git a/Documentation/tools/sbom/sbom.rst b/Documentation/tools/sbom/sbom.rst new file mode 100644 index 000000000..029b08b6a --- /dev/null +++ b/Documentation/tools/sbom/sbom.rst @@ -0,0 +1,206 @@ +.. SPDX-License-Identifier: GPL-2.0-only OR MIT +.. Copyright (C) 2025 TNG Technology Consulting GmbH + +KernelSbom +========== + +Introduction +------------ + +KernelSbom is a Python script ``scripts/sbom/sbom.py`` that can be +executed after a successful kernel build. When invoked, KernelSbom +analyzes all files involved in the build and generates Software Bill of +Materials (SBOM) documents in SPDX 3.0.1 format. +The generated SBOM documents capture: + +* **Final output artifacts**, typically the kernel image and modules +* **All source files** that contributed to the build with metadata + and licensing information +* **Details of the build process**, including intermediate artifacts + and the build commands linking source files to the final output + artifacts + +KernelSbom is originally developed in the +`KernelSbom repository `_. + +Requirements +------------ + +Python 3.10 or later. No libraries or other dependencies are required. + +Basic Usage +----------- + +Run the ``make sbom`` target. +For example:: + + $ make defconfig O=kernel_build + $ make sbom O=kernel_build -j$(nproc) + +This will trigger a kernel build. After all build outputs have been +generated, KernelSbom produces three SPDX documents in the root +directory of the object tree: + +* ``sbom-source.spdx.json`` + Describes all source files involved in the build and + associates each file with its corresponding license expression. + +* ``sbom-output.spdx.json`` + Captures all final build outputs (kernel image and ``.ko`` module files) + and includes build metadata such as environment variables and + a hash of the ``.config`` file used for the build. + +* ``sbom-build.spdx.json`` + Imports files from the source and output documents and describes every + intermediate build artifact. For each artifact, it records the exact + build command used and establishes the relationship between + input files and generated outputs. + +When invoking the sbom target, it is recommended to perform +out-of-tree builds using ``O=``. KernelSbom classifies files as +source files when they are located in the source tree and not in the +object tree. For in-tree builds, where the source and object trees are +the same directory, this distinction can no longer be made reliably. +In that case, KernelSbom does not generate a dedicated source SBOM. +Instead, source files are included in the build SBOM. + +Standalone Usage +---------------- + +KernelSbom can also be used as a standalone script to generate +SPDX documents for specific build outputs. For example, after a +successful x86 kernel build, KernelSbom can generate SPDX documents +for the ``bzImage`` kernel image:: + + $ SRCARCH=x86 python3 scripts/sbom/sbom.py \ + --src-tree . \ + --obj-tree ./kernel_build \ + --roots arch/x86/boot/bzImage \ + --generate-spdx \ + --generate-used-files \ + --prettify-json \ + --debug + +Note that when KernelSbom is invoked outside of the ``make`` process, +the environment variables used during compilation are not available and +therefore cannot be included in the generated SPDX documents. It is +recommended to set at least the ``SRCARCH`` environment variable to the +architecture for which the build was performed. + +For a full list of command-line options, run:: + + $ python3 scripts/sbom/sbom.py --help + +Output Format +------------- + +KernelSbom generates documents conforming to the +`SPDX 3.0.1 specification `_ +serialized as JSON-LD. + +To reduce file size, the output documents use the JSON-LD ``@context`` +to define custom prefixes for ``spdxId`` values. While this is compliant +with the SPDX specification, only a limited number of tools in the +current SPDX ecosystem support custom JSON-LD contexts. To use such +tools with the generated documents, the custom JSON-LD context must +be expanded before providing the documents. +See https://lists.spdx.org/g/Spdx-tech/message/6064 for more information. + +How it Works +------------ + +KernelSbom operates in two major phases: + +1. **Generate the cmd graph**, an acyclic directed dependency graph. +2. **Generate SPDX documents** based on the cmd graph. + +KernelSbom begins from the root artifacts specified by the user, e.g., +``arch/x86/boot/bzImage``. For each root artifact, it collects all +dependencies required to build that artifact. The dependencies come +from multiple sources: + +* **.cmd files**: The primary source is the ``.cmd`` file of the + generated artifact, e.g., ``arch/x86/boot/.bzImage.cmd``. These files + contain the exact command used to build the artifact and often include + an explicit list of input dependencies. By parsing the ``.cmd`` + file, the full list of dependencies can be obtained. + +* **.incbin statements**: The second source are include binary + ``.incbin`` statements in ``.S`` assembly files. + +* **Hardcoded dependencies**: Unfortunately, not all build dependencies + can be found via ``.cmd`` files and ``.incbin`` statements. Some build + dependencies are directly defined in Makefiles or Kbuild files. + Parsing these files is considered too complex for the scope of this + project. Instead, the remaining gaps of the graph are filled using a + list of manually defined dependencies, see + ``scripts/sbom/sbom/cmd_graph/hardcoded_dependencies.py``. This list is + known to be incomplete. However, analysis of the cmd graph indicates a + ~99% completeness. For more information about the completeness analysis, + see `KernelSbom #95 `_. + +Given the list of dependency files, KernelSbom recursively processes +each file, expanding the dependency chain all the way to the version +controlled source files. The result is a complete dependency graph +where nodes represent files, and edges represent "file A was used to +build file B" relationships. + +Using the cmd graph, KernelSbom produces three SPDX documents. +For every file in the graph, KernelSbom: + +* Parses ``SPDX-License-Identifier`` headers, +* Computes file hashes, +* Estimates the file type based on extension and path, +* Records build relationships between files. + +Each root output file is additionally associated with an SPDX Package +element that captures version information, license data, and copyright. + +Advanced Usage +-------------- + +Including Kernel Modules +~~~~~~~~~~~~~~~~~~~~~~~~ + +The list of all ``.ko`` kernel modules produced during a build can be +extracted from the ``modules.order`` file within the object tree. +For example:: + + $ echo "arch/x86/boot/bzImage" > sbom-roots.txt + $ sed 's/\.o$/.ko/' ./kernel_build/modules.order >> sbom-roots.txt + +Then use the generated roots file:: + + $ SRCARCH=x86 python3 scripts/sbom/sbom.py \ + --src-tree . \ + --obj-tree ./kernel_build \ + --roots-file sbom-roots.txt \ + --generate-spdx + +Equal Source and Object Trees +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +When the source tree and object tree are identical (for example, when +building in-tree), source files can no longer be reliably distinguished +from generated files. +In this scenario, KernelSbom does not produce a dedicated +``sbom-source.spdx.json`` document. Instead, both source files and build +artifacts are included together in ``sbom-build.spdx.json``, and +``sbom.used-files.txt`` lists all files referenced in the build document. + +Unknown Build Commands +~~~~~~~~~~~~~~~~~~~~~~ + +Because the kernel supports a wide range of configurations and versions, +KernelSbom may encounter build commands in ``.cmd`` files that it does +not yet support. By default, KernelSbom will fail if an unknown build +command is encountered. + +If you still wish to generate SPDX documents despite unsupported +commands, you can use the ``--do-not-fail-on-unknown-build-command`` +option. KernelSbom will continue and produce the documents, although +the resulting SBOM will be incomplete. + +This option should only be used when the missing portion of the +dependency graph is small and an incomplete SBOM is acceptable for +your use case. -- cgit v1.3.1