kernel/linux.git/tools/testing/selftests/bpf, branch v6.1.176

selftests/bpf: Fix bpf_nf selftest failure

2026-06-19T11:37:32+00:00

commit 967e8def1100cb4b08c28a54d27ce69563fdf281 upstream. For systems with missing iptables-legacy tool this selftest fails. Add check to find if iptables-legacy tool is available and skip the test if the tool is missing. Fixes: de9c8d848d90 ("selftests/bpf: S/iptables/iptables-legacy/ in the bpf_nf and xdp_synproxy test") Signed-off-by: Saket Kumar Bhaskar Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20250409095633.33653-1-skb99@linux.ibm.com Signed-off-by: Greg Kroah-Hartman

selftests/bpf: S/iptables/iptables-legacy/ in the bpf_nf and xdp_synproxy test

2026-06-19T11:36:57+00:00

[ Upstream commit de9c8d848d90cf2e53aced50b350827442ca5a4f ] The recent vm image in CI has reported error in selftests that use the iptables command. Manu Bretelle has pointed out the difference in the recent vm image that the iptables is sym-linked to the iptables-nft. With this knowledge, I can also reproduce the CI error by manually running with the 'iptables-nft'. This patch is to replace the iptables command with iptables-legacy to unblock the CI tests. Signed-off-by: Martin KaFai Lau Signed-off-by: Andrii Nakryiko Acked-by: David Vernet Link: https://lore.kernel.org/bpf/20221012221235.3529719-1-martin.lau@linux.dev Signed-off-by: Paul Chaignon Acked-by: Shung-Hsi Yu Signed-off-by: Sasha Levin

selftests/bpf: Fix ARG_PTR_TO_LONG {half-,}uninitialized test

2026-06-19T11:36:57+00:00

[ Upstream commit b8e188f023e07a733b47d5865311ade51878fe40 ] The assumption of 'in privileged mode reads from uninitialized stack locations are permitted' is not quite correct since the verifier was probing for read access rather than write access. Both tests need to be annotated as __success for privileged and unprivileged. Signed-off-by: Daniel Borkmann Acked-by: Andrii Nakryiko Link: https://lore.kernel.org/r/20240913191754.13290-6-daniel@iogearbox.net Signed-off-by: Alexei Starovoitov [ Note: The format of logs completely changed since 6.1 so this change had to be reapplied to the old test file. This commit needs to be backported because it fixes a test broken by commit 32556ce93bc4 ("bpf: Fix helper writes to read-only maps") from the same patchset. ] Signed-off-by: Paul Chaignon Acked-by: Shung-Hsi Yu Signed-off-by: Sasha Levin

Revert "selftests/bpf: Add a cgroup prog bpf_get_ns_current_pid_tgid() test"

2026-06-19T11:36:57+00:00

This reverts commit 4d8fb7ed7a55 ("selftests/bpf: Add a cgroup prog bpf_get_ns_current_pid_tgid() test"). That commit should have never been backported to 6.1 because it introduces a test for a feature that isn't supported: bpf_get_ns_current_pid_tgid() cannot be called from cgroup BPF programs in 6.1. Signed-off-by: Paul Chaignon Acked-by: Shung-Hsi Yu Signed-off-by: Sasha Levin

selftests/bpf: enhance align selftest's expected log matching

2026-06-19T11:36:57+00:00

[ Upstream commit 6f876e75d316a75957f3d43c3a8c2a6fe9bc18b2 ] Allow to search for expected register state in all the verifier log output that's related to specified instruction number. See added comment for an example of possible situation that is happening due to a simple enhancement done in the next patch, which fixes handling of env->test_state_freq flag in state checkpointing logic. Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/r/20230302235015.2044271-4-andrii@kernel.org Signed-off-by: Alexei Starovoitov [ Note: Backport needed to fix the align selftest where some of the expected log messages can't be found. This is happening because commit 1a8a315f008a ("bpf: Ensure proper register state printing for cond jumps") was also backported to 6.1. ] Signed-off-by: Paul Chaignon Acked-by: Shung-Hsi Yu Signed-off-by: Sasha Levin

selftests/bpf: Update bpf_clone_redirect expected return code

2026-06-19T11:36:57+00:00

[ Upstream commit b772b70b69046c5b76e3f2eda680f692dee5e6d5 ] Commit 151e887d8ff9 ("veth: Fixing transmit return status for dropped packets") started propagating proper NET_XMIT_DROP error to the caller which means it's now possible to get positive error code when calling bpf_clone_redirect() in this particular test. Update the test to reflect that. Reported-by: Daniel Borkmann Signed-off-by: Stanislav Fomichev Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20230911194731.286342-2-sdf@google.com [ Note: Commit 151e887d8ff9 was backported to 6.1 so this fix should be as well. ] Signed-off-by: Paul Chaignon Acked-by: Shung-Hsi Yu Signed-off-by: Sasha Levin

bpf: Fix a few selftest failures due to llvm18 change

2026-06-19T11:36:57+00:00

[ Upstream commit b16904fd9f01b580db357ef2b1cc9e86d89576c2 ] With latest upstream llvm18, the following test cases failed: $ ./test_progs -j #13/2 bpf_cookie/multi_kprobe_link_api:FAIL #13/3 bpf_cookie/multi_kprobe_attach_api:FAIL #13 bpf_cookie:FAIL #77 fentry_fexit:FAIL #78/1 fentry_test/fentry:FAIL #78 fentry_test:FAIL #82/1 fexit_test/fexit:FAIL #82 fexit_test:FAIL #112/1 kprobe_multi_test/skel_api:FAIL #112/2 kprobe_multi_test/link_api_addrs:FAIL [...] #112 kprobe_multi_test:FAIL #356/17 test_global_funcs/global_func17:FAIL #356 test_global_funcs:FAIL Further analysis shows llvm upstream patch [1] is responsible for the above failures. For example, for function bpf_fentry_test7() in net/bpf/test_run.c, without [1], the asm code is: 0000000000000400 : 400: f3 0f 1e fa endbr64 404: e8 00 00 00 00 callq 0x409 409: 48 89 f8 movq %rdi, %rax 40c: c3 retq 40d: 0f 1f 00 nopl (%rax) ... and with [1], the asm code is: 0000000000005d20 : 5d20: e8 00 00 00 00 callq 0x5d25 5d25: c3 retq ... and is called instead of and this caused test failures for #13/#77 etc. except #356. For test case #356/17, with [1] (progs/test_global_func17.c)), the main prog looks like: 0000000000000000 : 0: b4 00 00 00 2a 00 00 00 w0 = 0x2a 1: 95 00 00 00 00 00 00 00 exit ... which passed verification while the test itself expects a verification failure. Let us add 'barrier_var' style asm code in both places to prevent function specialization which caused selftests failure. [1] https://github.com/llvm/llvm-project/pull/72903 Signed-off-by: Yonghong Song Signed-off-by: Daniel Borkmann Link: https://lore.kernel.org/bpf/20231127050342.1945270-1-yonghong.song@linux.dev [ Note: The change to test_run.c conflicted and was dropped. The related tests are not failing anyway. ] Signed-off-by: Paul Chaignon Acked-by: Shung-Hsi Yu Signed-off-by: Sasha Levin

selftests/bpf: Add read_build_id function

2026-06-19T11:36:57+00:00

[ Upstream commit 88dc8b3605b38a440fba45edcc53a6c7a98eee3b ] Adding read_build_id function that parses out build id from specified binary. It will replace extract_build_id and also be used in following changes. Signed-off-by: Jiri Olsa Link: https://lore.kernel.org/r/20230331093157.1749137-3-jolsa@kernel.org Signed-off-by: Alexei Starovoitov Fixes: be4e85369e5a ("selftests/bpf: Replace extract_build_id with read_build_id") Signed-off-by: Paul Chaignon Acked-by: Shung-Hsi Yu Signed-off-by: Sasha Levin

Revert "selftests/bpf: Add tests for _opts variants of bpf_*_get_fd_by_id()"

2026-06-19T11:36:56+00:00

This reverts commit 45108a7b4866 ("selftests/bpf: Add tests for _opts variants of bpf_*_get_fd_by_id()"). As explained in the previous patch, it introduces a new selftest for a feature that doesn't exist in 6.1. It was backported as a stable-dep of a1914d146622 ("selftests/bpf: Workaround strict bpf_lsm return value check"), also reverted in the previous patch. Signed-off-by: Paul Chaignon Acked-by: Shung-Hsi Yu Signed-off-by: Sasha Levin

Revert "selftests/bpf: Workaround strict bpf_lsm return value check."

2026-06-19T11:36:56+00:00

This reverts commit a1914d146622 ("selftests/bpf: Workaround strict bpf_lsm return value check"). It seems it was picked up by mistake. It applies to a selftest that didn't exist in 6.1. The whole selftest was then backported as a stable-dep in commit 45108a7b4866 ("selftests/bpf: Add tests for _opts variants of bpf_*_get_fd_by_id()") (reverted as well in the next patch). The new selftest covers the bpf_*_get_fd_by_id structures. Those don't exist in 6.1 so the selftest shouldn't either. Signed-off-by: Paul Chaignon Acked-by: Shung-Hsi Yu Signed-off-by: Sasha Levin