Re: [PATCH bpf-next v1 4/4] selftests/bpf: Add kfunc set test to resolve_btfids

[Eduard Zingerman <eddyz87@gmail.com>]

On Wed, 2026-06-17 at 14:06 -0700, Ihor Solodrai wrote:

[...]

> @@ -161,6 +189,27 @@ void test_resolve_btfids(void)
>  			ASSERT_LE(test_set.ids[i - 1], test_set.ids[i], "sort_check");
>  	}
>  
> +	/* Check BTF_KFUNCS_START(test_kfunc_set) */
> +	ASSERT_EQ(test_kfunc_set.flags, BTF_SET8_KFUNCS, "kfunc_set_flags");
> +	ASSERT_EQ(test_kfunc_set.cnt, ARRAY_SIZE(kfunc_symbols), "kfunc_set_cnt");
> +
> +	for (i = 0; i < test_kfunc_set.cnt; i++) {
> +		for (j = 0; j < ARRAY_SIZE(kfunc_symbols); j++) {
> +			if (kfunc_symbols[j].id == (s32)test_kfunc_set.pairs[i].id) {
> +				ASSERT_EQ(test_kfunc_set.pairs[i].flags,
> +					  kfunc_symbols[j].flags, "kfunc_flags_check");
> +				break;
> +			}
> +		}
> +
> +		ASSERT_TRUE(j < ARRAY_SIZE(kfunc_symbols), "kfunc_id_found");
> +
> +		if (i > 0) {
> +			ASSERT_LE(test_kfunc_set.pairs[i - 1].id,
> +				  test_kfunc_set.pairs[i].id, "kfunc_sort_check");
> +		}

To assert the sorted property, should one define two sets?
{ a, b } and { b, a }? And check both.
Otherwise the id assignments might be an artifact of compiler BTF
generation logic.

> +	}
> +
>  out:
>  	btf__free(btf);
>  }

[...]