Re: [PATCH 2/2] selftests/mm: test kmemleak's N-consecutive-scan leak confirmation

[Breno Leitao]

On Fri, Jul 03, 2026 at 08:43:27AM -0700, Breno Leitao wrote:
> On Fri, Jul 03, 2026 at 03:55:52PM +0100, Catalin Marinas wrote:
>
> I will spin this "fix" separated (CCing stable), and send a v2 for this
> selftest with priming enabled.

Quick update on this plan. The v2 for this patch depends on checksum fix[0],
given it touches a similar code:

	@@ -687,6 +692,7 @@ static struct kmemleak_object *__alloc_object(gfp_t gfp)
		atomic_set(&object->use_count, 1);
		object->excess_ref = 0;
		object->count = 0; 
	+       object->unref_scans = 0;
		object->checksum = ~0;   <-- Adde by the fix [0]

I will wait the "Fix" to land first, before sending v2, avoiding merge
conflicts and being able to rely on sashiko to review it (instead of having merging
issues)

Link: https://lore.kernel.org/all/20260703-kmemleak_checksum-v1-1-5e0ab7d6966f@xxxxxxxxxx/ [0]