Re: [PATCH 6.1.y] mm/damon/paddr: drop last same folio access check reuse optimization

[SJ Park <sj@kernel.org>]

On Wed, 09 Sep 2026 05:52:15 +0000 sashiko-bot@kernel.org wrote:

> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [Low] Unused field `page_sz` left in `struct damon_pa_access_chk_result` after dropping the optimization.
> --
> 
> commit 9b0877a60d53720420f674776094bf4134b4f6d7
> Author: SJ Park <sj@kernel.org>
> 
> mm/damon/paddr: drop last same folio access check reuse optimization
> 
> This commit drops the optimization that reuses the last access check result
> if the region is in the same page. This removes the race condition when
> multiple kdamonds are used and favors code simplicity.
> 
> Link: https://lore.kernel.org/20260715031002.108504-5-sj@kernel.org
> Link: https://lore.kernel.org/20260621204050.10993-1-sj@kernel.org [1]
> 
> > diff --git a/mm/damon/paddr.c b/mm/damon/paddr.c
> > index 7bc8d79c7fa07..d1d047f494aed 100644
> > --- a/mm/damon/paddr.c
> > +++ b/mm/damon/paddr.c
> 
> [ ... ]
> 
> > @@ -161,29 +161,16 @@ static bool damon_pa_young(unsigned long paddr, unsigned long *page_sz)
> >  	folio_put(folio);
> >  
> >  out:
> > -	*page_sz = result.page_sz;
> 
> [Severity: Low]
> Since the optimization was removed and page_sz is no longer returned, is it
> necessary to keep the page_sz field in struct damon_pa_access_chk_result?
> 
> It looks like result.page_sz is still being assigned in __damon_pa_young()
> and initialized in damon_pa_young(), but the value is never used by any
> caller after this patch. Can the field and its assignments be removed
> entirely to clean up the remaining dead stores?

Apparently the upstream fix is not making such change.  Let's keep it as-is to
reduce unnecessary changes.

> 
> >  	return result.accessed;
> >  }
> 
> -- 
> Sashiko AI review · https://sashiko.dev/#/patchset/20260909054255.253652-1-sj@kernel.org?part=1
> 


Thanks,
SJ