Re: [PATCH bpf-next v3 3/4] selftests/bpf: Exercise unsafe obj drops from tracing progs

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] The test `task_kfunc_obj_drop_nmi_with_kptr` does not test NMI tracing programs as claimed by the commit message and its function name. Instead, it uses `SEC("tp_btf/task_newtask")`, which makes it a standard tracing program. Furthermore, because it fails at load-time for the exact same verifier reason as `task_kfunc_obj_drop_with_kptr`, the second test provides redundant coverage and leaves the intended NMI tracing context (`BPF_PROG_TYPE_PERF_EVENT`) completely untested.
--

commit e2b410eb90d404473252178d121d1547bd924299
Author: Kumar Kartikeya Dwivedi <memxor@gmail.com>

selftests/bpf: Exercise unsafe obj drops from tracing progs

This commit adds task_kfunc failure cases for bpf_obj_drop on local
objects with referenced kptr fields. It intends to verify that the
verifier rejects these operations from both standard tracing and NMI
tracing programs.

> diff --git a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
> --- a/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
> +++ b/tools/testing/selftests/bpf/progs/task_kfunc_failure.c
[ ... ]
> +SEC("tp_btf/task_newtask")
> +__failure __msg("bpf_obj_drop cannot be used in tracing programs on types with NMI unsafe fields")
> +int BPF_PROG(task_kfunc_obj_drop_with_kptr, struct task_struct *task, u64 clone_flags)
> +{
[ ... ]
> +}
> +
> +SEC("tp_btf/task_newtask")
> +__failure __msg("bpf_obj_drop cannot be used in tracing programs on types with NMI unsafe fields")
> +int BPF_PROG(task_kfunc_obj_drop_nmi_with_kptr, struct task_struct *task,
> +	     u64 clone_flags)

[Severity: Medium]
Does this test actually exercise the NMI tracing context as intended?

The function is named task_kfunc_obj_drop_nmi_with_kptr and the commit
message mentions NMI tracing programs, but the section used here is
SEC("tp_btf/task_newtask"). Should this be using an NMI-specific section
like SEC("perf_event") instead?

Because this uses the same tracing section as the previous test, it fails at
load-time for the exact same verifier reason as task_kfunc_obj_drop_with_kptr,
providing redundant coverage and leaving the NMI context untested.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260609202548.3571690-1-memxor@gmail.com?part=3