kernel/linux.git/tools/testing/selftests/bpf/Makefile, branch v6.12.80

selftests/bpf: Avoid generating untracked files when running bpf selftests

2025-02-08T08:57:31+00:00

[ Upstream commit 73b9075f334f5debf28646884a320b796b27768d ] Currently, when we run the BPF selftests with the following command: make -C tools/testing/selftests TARGETS=bpf SKIP_TARGETS="" The command generates untracked files and directories with make version less than 4.4: ''' Untracked files: (use "git add ..." to include in what will be committed) tools/testing/selftests/bpfFEATURE-DUMP.selftests tools/testing/selftests/bpffeature/ ''' We lost slash after word "bpf". The reason is slash appending code is as follow: ''' OUTPUT := $(OUTPUT)/ $(eval include ../../../build/Makefile.feature) OUTPUT := $(patsubst %/,%,$(OUTPUT)) ''' This way of assigning values to OUTPUT will never be effective for the variable OUTPUT provided via the command argument [1] and BPF makefile is called from parent Makfile(tools/testing/selftests/Makefile) like: ''' all: ... $(MAKE) OUTPUT=$$BUILD_TARGET -C $$TARGET ''' According to GNU make, we can use override Directive to fix this issue [2]. [1] https://www.gnu.org/software/make/manual/make.html#Overriding [2] https://www.gnu.org/software/make/manual/make.html#Override-Directive Fixes: dc3a8804d790 ("selftests/bpf: Adapt OUTPUT appending logic to lower versions of Make") Signed-off-by: Jiayuan Chen Signed-off-by: Daniel Borkmann Acked-by: Jiri Olsa Link: https://lore.kernel.org/bpf/20241224075957.288018-1-mrpre@163.com Signed-off-by: Sasha Levin

selftests: bpf: Add missing per-arch include path

2024-12-05T13:01:42+00:00

[ Upstream commit 19090f0306f1748980596c6c71f1c4b128639cff ] The prog_tests programs do not include the per-arch tools include path, e.g. tools/arch/riscv/include. Some architectures depend those files to build properly. Include tools/arch/$(SUBARCH)/include in the selftests bpf build. Fixes: 6d74d178fe6e ("tools: Add riscv barrier implementation") Signed-off-by: Björn Töpel Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20240927131355.350918-2-bjorn@kernel.org Signed-off-by: Sasha Levin

selftests/bpf: Fix cross-compiling urandom_read

2024-10-11T02:08:14+00:00

Linking of urandom_read and liburandom_read.so prefers LLVM's 'ld.lld' but falls back to using 'ld' if unsupported. However, this fallback discards any existing makefile macro for LD and can break cross-compilation. Fix by changing the fallback to use the target linker $(LD), passed via '-fuse-ld=' using an absolute path rather than a linker "flavour". Fixes: 08c79c9cd67f ("selftests/bpf: Don't force lld on non-x86 architectures") Signed-off-by: Tony Ambardar Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20241009040720.635260-1-tony.ambardar@gmail.com

selftests/bpf: Add test for kfunc module order

2024-10-10T17:44:03+00:00

Add a test case for kfuncs from multiple external modules, checking that the correct kfuncs are called regardless of which order they're called in. Specifically, check that calling the kfuncs in an order different from the one the modules' BTF are loaded in works. Signed-off-by: Simon Sundberg Acked-by: Kumar Kartikeya Dwivedi Signed-off-by: Toke Høiland-Jørgensen Link: https://lore.kernel.org/r/20241010-fix-kfunc-btf-caching-for-modules-v2-3-745af6c1af98@redhat.com Signed-off-by: Alexei Starovoitov

selftests/bpf: add build ID tests

2024-09-11T16:58:31+00:00

Add a new set of tests validating behavior of capturing stack traces with build ID. We extend uprobe_multi target binary with ability to trigger uprobe (so that we can capture stack traces from it), but also we allow to force build ID data to be either resident or non-resident in memory (see also a comment about quirks of MADV_PAGEOUT). That way we can validate that in non-sleepable context we won't get build ID (as expected), but with sleepable uprobes we will get that build ID regardless of it being physically present in memory. Also, we add a small add-on linker script which reorders .note.gnu.build-id section and puts it after (big) .text section, putting build ID data outside of the very first page of ELF file. This will test all the relaxations we did in build ID parsing logic in kernel thanks to freader abstraction. Reviewed-by: Eduard Zingerman Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/r/20240829174232.3133883-11-andrii@kernel.org Signed-off-by: Alexei Starovoitov

selftests/bpf: Prefer static linking for LLVM libraries

2024-09-05T20:13:39+00:00

It is not always convenient to have LLVM libraries installed inside CI rootfs images, thus request static libraries from llvm-config. Suggested-by: Andrii Nakryiko Signed-off-by: Eduard Zingerman Acked-by: Daniel Borkmann Tested-by: Daniel Borkmann Link: https://lore.kernel.org/r/20240905081401.1894789-4-pulehui@huaweicloud.com Signed-off-by: Alexei Starovoitov

selftests/bpf: Adapt OUTPUT appending logic to lower versions of Make

2024-09-05T20:13:39+00:00

The $(let ...) function is only supported by GNU Make version 4.4 [0] and above, otherwise the following exception file or directory will be generated: tools/testing/selftests/bpfFEATURE-DUMP.selftests tools/testing/selftests/bpffeature/ Considering that the GNU Make version of most Linux distributions is lower than 4.4, let us adapt the corresponding logic to it. Link: https://lists.gnu.org/archive/html/info-gnu/2022-10/msg00008.html [0] Acked-by: Eduard Zingerman Signed-off-by: Pu Lehui Link: https://lore.kernel.org/r/20240905081401.1894789-2-pulehui@huaweicloud.com Signed-off-by: Alexei Starovoitov

selftests/bpf: Do not update vmlinux.h unnecessarily

2024-08-30T18:54:14+00:00

%.bpf.o objects depend on vmlinux.h, which makes them transitively dependent on unnecessary libbpf headers. However vmlinux.h doesn't actually change as often. When generating vmlinux.h, compare it to a previous version and update it only if there are changes. Example of build time improvement (after first clean build): $ touch ../../../lib/bpf/bpf.h $ time make -j8 Before: real 1m37.592s After: real 0m27.310s Notice that %.bpf.o gen step is skipped if vmlinux.h hasn't changed. Signed-off-by: Ihor Solodrai Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/CAEf4BzY1z5cC7BKye8=A8aTVxpsCzD=p1jdTfKC7i0XVuYoHUQ@mail.gmail.com Link: https://lore.kernel.org/bpf/20240828174608.377204-2-ihor.solodrai@pm.me

selftests/bpf: Specify libbpf headers required for %.bpf.o progs

2024-08-30T18:54:14+00:00

Test %.bpf.o objects actually depend only on some libbpf headers. Define a list of required headers and use it as TRUNNER_BPF_OBJS dependency. bpf_*.h list was determined by: $ grep -rh 'include Signed-off-by: Andrii Nakryiko Link: Link: https://lore.kernel.org/bpf/20240828174608.377204-1-ihor.solodrai@pm.me https://lore.kernel.org/bpf/CAEf4BzYQ-j2i_xjs94Nn=8+FVfkWt51mLZyiYKiz9oA4Z=pCeA@mail.gmail.com/

selftests/bpf: use simply-expanded variables for libpcap flags

2024-08-23T21:33:30+00:00

Save pkg-config output for libpcap as simply-expanded variables. For an obscure reason 'shell' call in LDLIBS/CFLAGS recursively expanded variables makes *.test.o files compilation non-parallel when make is executed with -j option. While at it, reuse 'pkg-config --cflags' call to define -DTRAFFIC_MONITOR=1 option, it's exit status is the same as for 'pkg-config --exists'. Fixes: f52403b6bfea ("selftests/bpf: Add traffic monitor functions.") Signed-off-by: Eduard Zingerman Link: https://lore.kernel.org/r/20240823194409.774815-1-eddyz87@gmail.com Signed-off-by: Martin KaFai Lau