Re: [PATCH v2 00/14] mm: Unify memmap initialization for compound pages
From: Li Zhe
Date: Mon Nov 16 2026 - 06:46:22 EST
On 11/16/26 6:42 PM, Mike Rapoport wrote:
> Hi,
>
>> Hi,
>>
>> This v2 is based on the mm-new branch at commit daed51c91051.
>>
>> This series is split out from the earlier, larger series
>> "mm: Generalize HVO for HugeTLB and device DAX" [1]. That series
>> generalized vmemmap optimization across HugeTLB and device DAX. One
>> remaining piece is the separate compound struct page initialization used
>> by boot-time gigantic HugeTLB pages and compound ZONE_DEVICE ranges.
>>
>> This v2 also folds in Li Zhe's two-patch series, using that series' v2
>> as the integration point [2].
>>
>> Li Zhe (2):
>> mm/mm_init: factor out memmap_init_page_from_template()
>> mm/hugetlb: optimize bootmem hugepage tail vmemmap init
>>
>> Muchun Song (12):
>> mm/mm_init: verify zones for template-initialized pages
>> mm/mm_init: factor out common struct page initialization
>> mm/mm_init: simplify zone device page initialization
>> mm/mm_init: factor out frozen compound page initialization
>> mm/hugetlb: drop boot-time HVO handling for gigantic folios
>> mm/hugetlb: initialize all struct pages for bootmem hugepages
>> mm/hugetlb: use common compound page initialization helper
>> mm/mm_init: make __init_single_page() static
>> mm/mm_init: make memmap_init_page_from_template() static
>> mm/cma: centralize CMA pageblock initialization
>> mm/cma: localize init_cma_pageblock() to cma.c
>> mm/mm_init: move pageblock migratetype init to memmap helpers
>>
>> include/linux/mm.h | 8 --
>> mm/cma.c | 24 +++--
>> mm/hugetlb.c | 111 ++-----------------
>> mm/hugetlb_vmemmap.c | 17 +--
>> mm/mm_init.c | 247 +++++++++++++++++--------------------------
>> mm/mm_init.h | 13 +--
>> mm/sparse-vmemmap.c | 7 +-
>> 7 files changed, 129 insertions(+), 298 deletions(-)
> That's nice :)
>
> The series would be easier to review if Li Zhe's patches were integrated
> rather than stacked in. I hoped that the changes for template
> initialization would fit into the patches that rework other parts.
>
> The way it's done now several patches in the beginning of the series
> change code that the later patches remove or turn some of new functions
> into static.
>
> Another thing is that the numbers Li Zhe measured were for a different
> implementation, so they would be stale for the state after these
> changes.
Agreed. I would be happy to provide updated performance numbers once the
final integrated form is ready.
Thanks,
Zhe
>
> I have some nits and naming comments in a few patches, other than that
> the series LGTM.