Re: [PATCH RESEND bpf-next v2] selftests/bpf: veristat: Minimize map size during verification

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] veristat overrides BPF_MAP_TYPE_HASH_OF_MAPS max_entries to 1, causing map preparation to fail for declaratively initialized maps
--

--- Patch [1]: [PATCH RESEND bpf-next v2] selftests/bpf: veristat: Minimize map size during verification ---
commit be52ac6965be2ec63632786bb06d60b4e6152bc6
Author: Emil Tsalapatis <emil@etsalapatis.com>

selftests/bpf: veristat: Minimize map size during verification

This commit updates veristat to resize BPF maps where max_entries is
irrelevant to the verification process. By shrinking these maps to the
minimum size, it avoids -ENOMEM errors on memory-constrained systems.

> diff --git a/tools/testing/selftests/bpf/veristat.c b/tools/testing/selftests/bpf/veristat.c
> --- a/tools/testing/selftests/bpf/veristat.c
> +++ b/tools/testing/selftests/bpf/veristat.c
> @@ -1248,6 +1248,30 @@ static void fixup_obj_maps(struct bpf_object *obj)
>  
>  		/* fix up map size, if necessary */
>  		switch (bpf_map__type(map)) {
> +		/*
> +		 * if the verifier doesn't use max_entries
> +		 * then set to 1 to avoid -ENOMEM
> +		 */
> +		case BPF_MAP_TYPE_HASH:
> +		case BPF_MAP_TYPE_PERCPU_HASH:
> +		case BPF_MAP_TYPE_LRU_HASH:
> +		case BPF_MAP_TYPE_LRU_PERCPU_HASH:
> +		case BPF_MAP_TYPE_HASH_OF_MAPS:

[Severity: Medium]
Will this cause map preparation to fail with -E2BIG if there are multiple
inner maps declaratively initialized?

In fixup_obj_maps(), if the outer map's max_entries is forcibly overridden to 1,
when libbpf later attempts to populate the slots via bpf_map_update_elem()
during bpf_object__prepare(), the first insertion succeeds, but subsequent
insertions fail because the hash map's capacity is constrained to 1.

Does this mean tests that declaratively initialize multiple inner maps
(such as test_btf_map_in_map.bpf.o) will fail to load with invalid map FDs,
subsequently causing the verifier to reject the programs?

> +		case BPF_MAP_TYPE_SOCKHASH:
[ ... ]

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260707162012.8063-1-emil@etsalapatis.com?part=1