Re: [PATCH 7.1.y] samples/damon/mtier: fail early if address range parameters are invalid

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] Resource leaks (memory, kthreads, struct pid) on failure paths in DAMON sample modules.
--

--- Patch [1]: [PATCH 7.1.y] samples/damon/mtier: fail early if address range parameters are invalid ---
commit 999f01f2d7c2530ae577d445b6761ac7a2a10e77
Author: Zenghui Yu <yuzenghui@huawei.com>

samples/damon/mtier: fail early if address range parameters are invalid

This commit prevents a WARN_ONCE() trigger in the damon mtier sample module
by validating that node{0,1}_{start,end}_addr specify a valid region
before passing them to damon_new_region().

Link: https://lore.kernel.org/20260629144432.133962-1-sj@kernel.org

> diff --git a/samples/damon/mtier.c b/samples/damon/mtier.c
> index 775838a23d935..7f84367ad3e58 100644
> --- a/samples/damon/mtier.c
> +++ b/samples/damon/mtier.c
> @@ -120,6 +120,9 @@ static struct damon_ctx *damon_sample_mtier_build_ctx(bool promote)
>  		addr.end = promote ? node1_end_addr : node0_end_addr;
>  	}
>  
> +	if (addr.start >= addr.end)
> +		goto free_out;
> +
>  	region = damon_new_region(addr.start, addr.end);
>  	if (!region)
>  		goto free_out;

[Severity: High]
This is a pre-existing issue, but does damon_sample_mtier_start() leak memory 
and running kthreads on failure?

Looking at samples/damon/mtier.c:damon_sample_mtier_start():

	return damon_start(ctxs, 2, true);

If damon_start() succeeds for the first context but fails for the second, the
error is returned directly. This leaves the first context running
indefinitely, leaking both the damon_ctx objects and the started kthreads.
This would hold the damon core's exclusive flag, preventing future exclusive
monitoring on the system until reboot.

A similar issue appears in samples/damon/prcl.c:damon_sample_prcl_start():

	err = damon_start(&ctx, 1, true);
	if (err)
		return err;

	repeat_call_control.data = ctx;
	return damon_call(ctx, &repeat_call_control);

And in samples/damon/wsse.c:damon_sample_wsse_start():

	err = damon_start(&ctx, 1, true);
	if (err)
		return err;

	repeat_call_control.data = ctx;
	return damon_call(ctx, &repeat_call_control);

If damon_call() fails, the functions return the error immediately without
stopping the started threads or destroying the contexts. This also bypasses 
damon_destroy_ctx(), leading to a struct pid reference leak.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260721053953.181336-1-sj@kernel.org?part=1