From: Kiryl Shutsemau Date: Fri, 27 Feb 2026 19:42:39 +0000 (+0000) Subject: mm: move MAX_FOLIO_ORDER definition to mmzone.h X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=a2c77ec320a99581e8272868ccfa53a7d7a7b168;p=thirdparty%2Flinux.git mm: move MAX_FOLIO_ORDER definition to mmzone.h Patch series "mm: Eliminate fake head pages from vmemmap optimization", v7. This series removes "fake head pages" from the HugeTLB vmemmap optimization (HVO) by changing how tail pages encode their relationship to the head page. It simplifies compound_head() and page_ref_add_unless(). Both are in the hot path. Background ========== HVO reduces memory overhead by freeing vmemmap pages for HugeTLB pages and remapping the freed virtual addresses to a single physical page. Previously, all tail page vmemmap entries were remapped to the first vmemmap page (containing the head struct page), creating "fake heads" - tail pages that appear to have PG_head set when accessed through the deduplicated vmemmap. This required special handling in compound_head() to detect and work around fake heads, adding complexity and overhead to a very hot path. New Approach ============ For architectures/configs where sizeof(struct page) is a power of 2 (the common case), this series changes how position of the head page is encoded in the tail pages. Instead of storing a pointer to the head page, the ->compound_info (renamed from ->compound_head) now stores a mask. The mask can be applied to any tail page's virtual address to compute the head page address. Critically, all tail pages of the same order now have identical compound_info values, regardless of which compound page they belong to. The key insight is that all tail pages of the same order now have identical compound_info values, regardless of which compound page they belong to. In v7, these shared tail pages are allocated per-zone. This ensures that zone information (stored in page->flags) is correct even for shared tail pages, removing the need for the special-casing in page_zonenum() proposed in earlier versions. To support per-zone shared pages for boot-allocated gigantic pages, the vmemmap population is deferred until zones are initialized. This simplifies the logic significantly and allows the removal of vmemmap_undo_hvo(). Benefits ======== 1. Simplified compound_head(): No fake head detection needed, can be implemented in a branchless manner. 2. Simplified page_ref_add_unless(): RCU protection removed since there's no race with fake head remapping. 3. Cleaner architecture: The shared tail pages are truly read-only and contain valid tail page metadata. If sizeof(struct page) is not power-of-2, there are no functional changes. HVO is not supported in this configuration. I had hoped to see performance improvement, but my testing thus far has shown either no change or only a slight improvement within the noise. Series Organization =================== Patch 1: Move MAX_FOLIO_ORDER definition to mmzone.h. Patches 2-4: Refactoring of field names and interfaces. Patches 5-6: Architecture alignment for LoongArch and RISC-V. Patch 7: Mask-based compound_head() implementation. Patch 8: Add memmap alignment checks. Patch 9: Branchless compound_head() optimization. Patch 10: Defer vmemmap population for bootmem hugepages. Patch 11: Refactor vmemmap_walk. Patch 12: x86 vDSO build fix. Patch 13: Eliminate fake heads with per-zone shared tail pages. Patches 14-16: Cleanup of fake head infrastructure. Patch 17: Documentation update. Patch 18: Use compound_head() in page_slab(). This patch (of 17): Move MAX_FOLIO_ORDER definition from mm.h to mmzone.h. This is preparation for adding the vmemmap_tails array to struct zone, which requires MAX_FOLIO_ORDER to be available in mmzone.h. Link: https://lkml.kernel.org/r/20260227194302.274384-1-kas@kernel.org Link: https://lkml.kernel.org/r/20260227194302.274384-2-kas@kernel.org Signed-off-by: Kiryl Shutsemau Acked-by: David Hildenbrand (Red Hat) Acked-by: Zi Yan Acked-by: Muchun Song Acked-by: Usama Arif Reviewed-by: Vlastimil Babka Cc: Albert Ou Cc: Alexandre Ghiti Cc: Baoquan He Cc: Christoph Lameter Cc: David Rientjes Cc: Frank van der Linden Cc: Harry Yoo Cc: Huacai Chen Cc: Johannes Weiner Cc: Jonathan Corbet Cc: Lorenzo Stoakes Cc: Matthew Wilcox (Oracle) Cc: Michal Hocko Cc: Mike Rapoport Cc: Oscar Salvador Cc: Palmer Dabbelt Cc: Paul Walmsley Cc: Roman Gushchin Cc: WANG Xuerui Signed-off-by: Andrew Morton --- diff --git a/include/linux/mm.h b/include/linux/mm.h index bb0cfe38ca19f..4e999c21d89a8 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -27,7 +27,6 @@ #include #include #include -#include #include #include #include @@ -2479,36 +2478,6 @@ static inline unsigned long folio_nr_pages(const struct folio *folio) return folio_large_nr_pages(folio); } -#if !defined(CONFIG_HAVE_GIGANTIC_FOLIOS) -/* - * We don't expect any folios that exceed buddy sizes (and consequently - * memory sections). - */ -#define MAX_FOLIO_ORDER MAX_PAGE_ORDER -#elif defined(CONFIG_SPARSEMEM) && !defined(CONFIG_SPARSEMEM_VMEMMAP) -/* - * Only pages within a single memory section are guaranteed to be - * contiguous. By limiting folios to a single memory section, all folio - * pages are guaranteed to be contiguous. - */ -#define MAX_FOLIO_ORDER PFN_SECTION_SHIFT -#elif defined(CONFIG_HUGETLB_PAGE) -/* - * There is no real limit on the folio size. We limit them to the maximum we - * currently expect (see CONFIG_HAVE_GIGANTIC_FOLIOS): with hugetlb, we expect - * no folios larger than 16 GiB on 64bit and 1 GiB on 32bit. - */ -#define MAX_FOLIO_ORDER get_order(IS_ENABLED(CONFIG_64BIT) ? SZ_16G : SZ_1G) -#else -/* - * Without hugetlb, gigantic folios that are bigger than a single PUD are - * currently impossible. - */ -#define MAX_FOLIO_ORDER PUD_ORDER -#endif - -#define MAX_FOLIO_NR_PAGES (1UL << MAX_FOLIO_ORDER) - /* * compound_nr() returns the number of pages in this potentially compound * page. compound_nr() can be called on a tail page, and is defined to diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index db41b18a919d6..4c481ec77da99 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -23,6 +23,7 @@ #include #include #include +#include #include /* Free memory management - zoned buddy allocator. */ @@ -61,6 +62,36 @@ */ #define PAGE_ALLOC_COSTLY_ORDER 3 +#if !defined(CONFIG_HAVE_GIGANTIC_FOLIOS) +/* + * We don't expect any folios that exceed buddy sizes (and consequently + * memory sections). + */ +#define MAX_FOLIO_ORDER MAX_PAGE_ORDER +#elif defined(CONFIG_SPARSEMEM) && !defined(CONFIG_SPARSEMEM_VMEMMAP) +/* + * Only pages within a single memory section are guaranteed to be + * contiguous. By limiting folios to a single memory section, all folio + * pages are guaranteed to be contiguous. + */ +#define MAX_FOLIO_ORDER PFN_SECTION_SHIFT +#elif defined(CONFIG_HUGETLB_PAGE) +/* + * There is no real limit on the folio size. We limit them to the maximum we + * currently expect (see CONFIG_HAVE_GIGANTIC_FOLIOS): with hugetlb, we expect + * no folios larger than 16 GiB on 64bit and 1 GiB on 32bit. + */ +#define MAX_FOLIO_ORDER get_order(IS_ENABLED(CONFIG_64BIT) ? SZ_16G : SZ_1G) +#else +/* + * Without hugetlb, gigantic folios that are bigger than a single PUD are + * currently impossible. + */ +#define MAX_FOLIO_ORDER PUD_ORDER +#endif + +#define MAX_FOLIO_NR_PAGES (1UL << MAX_FOLIO_ORDER) + enum migratetype { MIGRATE_UNMOVABLE, MIGRATE_MOVABLE,