Re: [PATCH v3 0/4] mm/hugetlb: Let unmap_hugepage_range() and

[Matthew Wilcox <willy@infradead.org>]

On Mon, Apr 28, 2025 at 10:11:43AM -0700, nifan.cxl@gmail.com wrote:
> From: Fan Ni <fan.ni@samsung.com>

When you're sending v2, send the same cover letter as you did for the
first one; don't make people go and look at v1 for the cover letter.

Put what was changed in v2 after it, so they have some idea what to look
at when reviewing.

> Changes compared to v2,
> 
> Patch 1: 
> 1) Update the commit log subject; 
> 2) Use &folio->page instead of folio_page(folio) in unmap_ref_private()
>   when calling unmap_hugepage_range();
> 
> Patch 2:
> 1) Update the declaration of unmap_hugepage_range() in hugetlb.h;
> 2) Use &folio->page instead of folio_page(folio) in unmap_hugepage_range()
>   when calling __unmap_hugepage_range();
> 
> Patch 3: 
> 1) Update the declaration of __unmap_hugepage_range() in hugetlb.h;
> 2) Rename ref_folio to folio;
> 3) compare folio instead of page in __unmap_hugepage_range() when folio is
>   provided when calling __unmap_hugepage_range();
> 
> Patch 4:
> 1) Pass folio size instead of huge_page_size() when calling
>   tlb_remove_page_size() by Matthew;
> 2) Update the processing inside __unmap_hugepage_range() when folio
>   is provided as sugguested by David Hildenbrand;
> 3) Since there is some functional change in this patch, we do not pick up the
>   tags;
> 
> v2:
> https://lore.kernel.org/linux-mm/20250418170834.248318-2-nifan.cxl@gmail.com
> 
> Fan Ni (4):
>   mm/hugetlb: Pass folio instead of page to unmap_ref_private()
>   mm/hugetlb: Refactor unmap_hugepage_range() to take folio instead of
>     page
>   mm/hugetlb: Refactor __unmap_hugepage_range() to take folio instead of
>     page
>   mm/hugetlb: Convert use of struct page to folio in
>     __unmap_hugepage_range()
> 
>  include/linux/hugetlb.h |  8 ++++----
>  mm/hugetlb.c            | 39 +++++++++++++++++++++------------------
>  2 files changed, 25 insertions(+), 22 deletions(-)
> 
> -- 
> 2.47.2
>