Re: [RFC PATCH v4 8/9] mm/damon/core: cap the region merge threshold per target

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] mm/damon: integer truncation in max_thres bounds calculation
--

commit a10e8369d2fd5c44226b6327f0cddd983be70366
Author: Ravi Jonnalagadda <ravis.opensrc@gmail.com>

mm/damon/core: cap the region merge threshold per target

This commit modifies kdamond_merge_regions to cap the merge threshold for each
target individually during the regular merge pass. It also adjusts the
maximum threshold for probe-weighted contexts to ensure escalation passes
continue to run even when weighted hits exceed the normal maximum threshold.

[ ... ]

> diff --git a/mm/damon/core.c b/mm/damon/core.c
> index e3c29d7cff23d..7c0df07f0a908 100644
> --- a/mm/damon/core.c
> +++ b/mm/damon/core.c
> @@ -4167,12 +4167,40 @@ static noinline_for_stack void kdamond_merge_regions(struct damon_ctx *c,
>  	unsigned int nr_regions;
>  	unsigned int max_thres;
>  	bool count_age = true;
> +	bool use_probe_hits = damon_has_probe_weights(c);
>  
>  	max_thres = damon_nr_samples_per_aggr(&c->attrs);
> +	/* weighted scores can exceed max_thres; threshold is max score / 10 */
> +	if (use_probe_hits)
> +		max_thres = max(threshold * 10, max_thres);

[Severity: Medium]
When the true maximum merge score is not a multiple of 10, does this
calculation cause max_nr_regions to be violated?

Because kdamond_fn() calculates the initial threshold by dividing
max_merge_score by 10, the remainder is discarded:

mm/damon/core.c:kdamond_fn() {
    ...
    kdamond_merge_regions(ctx, max_merge_score / 10, sz_limit);
    ...
}

When kdamond_merge_regions() attempts to restore the max_thres ceiling
via threshold * 10, it loses the remainder (e.g., a max score of 29
yields threshold 2, restoring a ceiling of 20 instead of 29).

Will the escalation loop then terminate prematurely when it reaches 20,
leaving regions with score differences up to 29 unmerged and the total
region count above the configured limit?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261005-damon-perf-rfc-v3-send-2026-10-03-v4-0-b03452e137f3@gmail.com?part=8