Re: [PATCH v3] ksm: use range-walk function to jump over holes in scan_get_next_rmap_item

[David Hildenbrand <david@redhat.com>]

On 16.10.25 23:07, Andrew Morton wrote:
> On Wed, 15 Oct 2025 22:22:36 -0300 Pedro Demarchi Gomes <pedrodemargomes@gmail.com> wrote:
> 
>> Currently, scan_get_next_rmap_item() walks every page address in a VMA
>> to locate mergeable pages. This becomes highly inefficient when scanning
>> large virtual memory areas that contain mostly unmapped regions.
>>
>> This patch replaces the per-address lookup with a range walk using
>> walk_page_range(). The range walker allows KSM to skip over entire
>> unmapped holes in a VMA, avoiding unnecessary lookups.
>> This problem was previously discussed in [1].
>>
>> ...
>>
>> Reported-by: craftfever <craftfever@airmail.cc>
>> Closes: https://lkml.kernel.org/r/020cf8de6e773bb78ba7614ef250129f11a63781@murena.io
>> Suggested-by: David Hildenbrand <david@redhat.com>
>> Signed-off-by: Pedro Demarchi Gomes <pedrodemargomes@gmail.com>
> 
> Is Fixes: b1d3e9bbccb4 ("mm/ksm: convert scan_get_next_rmap_item() from
> follow_page() to folio_walk") appropriate?

No.

That commit is not the problem.

The problem probably goes back when scan_get_next_rmap_item() was first 
introduced (likely when KSM was added): it simply never was optimized to 
deal with large sparse memory areas.

> 
> The problem which is being addressed seems pretty serious.  What do
> people think about proposing a -stable backport of this fix?

We'll likely have to backport it to each and every stable tree. We could 
think about limiting backports only to kernels that actually allow for 
enabling KSM for a complete process.

So that would make sense to me.

-- 
Cheers

David / dhildenb