Re: [PATCH 03/11] mm/damon/tests/core-kunit: test damon_rand()

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
- [Medium] The lower bound check `KUNIT_EXPECT_GE(test, rnd, 0)` is a tautology and fails to test the lower bound offset logic.
- [Low] The test completely misses the 64-bit slow path of `damon_rand()`.
--

commit 3514b0c5edd38e0cdd6bf2d3e8b3ba10d7e617d2
Author: SJ Park <sj@kernel.org>

mm/damon/tests/core-kunit: test damon_rand()

This commit introduces a kunit test for the damon_rand() PRNG to ensure that
the generated output respects the expected bounds.

> diff --git a/mm/damon/tests/core-kunit.h b/mm/damon/tests/core-kunit.h
> index a00168730445d..0ec7d14d354e7 100644
> --- a/mm/damon/tests/core-kunit.h
> +++ b/mm/damon/tests/core-kunit.h

[ ... ]

> +static void damon_test_rand(struct kunit *test)
> +{
> +	struct damon_ctx ctx;
> +	int i;
> +
> +	prandom_seed_state(&ctx.rnd_state, get_random_u64());
> +	for (i = 0; i < 10000; i++) {
> +		unsigned long rnd = damon_rand(&ctx, 0, 10);

[Severity: Low]
Does this test miss the 64-bit fallback path in damon_rand()?

Looking at the implementation of damon_rand(), it contains specific logic for
spans exceeding U32_MAX. By hardcoding a span of 10, the test exclusively
exercises the 32-bit fast path and leaves the 64-bit arithmetic
(mul_u64_u64_shr) completely untested.

Could we add cases that test spans larger than U32_MAX?

> +
> +		KUNIT_EXPECT_GE(test, rnd, 0);

[Severity: Medium]
Is this comparison inherently tautological?

Since rnd is an unsigned long, rnd >= 0 is always true, which renders this
assertion ineffective and might trigger compiler warnings for tautological
comparisons.

Additionally, because the test uses 0 for the lower bound, does it properly
verify that damon_rand() correctly applies the lower bound offset to its
output? If the implementation failed to add the lower bound, this test would
not catch it.

> +		KUNIT_EXPECT_LE(test, rnd, 9);
> +	}
> +}

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260630141726.92246-1-sj@kernel.org?part=3