From: Yu Zhao yuzhao@google.com
[ Upstream commit 95599ef684d01136a8b77c16a7c853496786e173 ]
The current assumption is that a large folio can only be split into order-0 folios. That is not the case for hugeTLB demotion, nor for THP split: see commit c010d47f107f ("mm: thp: split huge page to any lower order pages").
When a large folio is split into ones of a lower non-zero order, only the new head pages should be tagged. Tagging tail pages can cause imbalanced "calls" counters, since only head pages are untagged by pgalloc_tag_sub() and the "calls" counts on tail pages are leaked, e.g.,
# echo 2048kB >/sys/kernel/mm/hugepages/hugepages-1048576kB/demote_size # echo 700 >/sys/kernel/mm/hugepages/hugepages-1048576kB/nr_hugepages # time echo 700 >/sys/kernel/mm/hugepages/hugepages-1048576kB/demote # echo 0 >/sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages # grep alloc_gigantic_folio /proc/allocinfo
Before this patch: 0 549427200 mm/hugetlb.c:1549 func:alloc_gigantic_folio
real 0m2.057s user 0m0.000s sys 0m2.051s
After this patch: 0 0 mm/hugetlb.c:1549 func:alloc_gigantic_folio
real 0m1.711s user 0m0.000s sys 0m1.704s
Not tagging tail pages also improves the splitting time, e.g., by about 15% when demoting 1GB hugeTLB folios to 2MB ones, as shown above.
Link: https://lkml.kernel.org/r/20240906042108.1150526-2-yuzhao@google.com Fixes: be25d1d4e822 ("mm: create new codetag references during page splitting") Signed-off-by: Yu Zhao yuzhao@google.com Acked-by: Suren Baghdasaryan surenb@google.com Cc: Kent Overstreet kent.overstreet@linux.dev Cc: Muchun Song muchun.song@linux.dev Cc: stable@vger.kernel.org Signed-off-by: Andrew Morton akpm@linux-foundation.org --- include/linux/mm.h | 32 ++++++++++++++++++++++++++++++++ include/linux/pgalloc_tag.h | 31 ------------------------------- mm/huge_memory.c | 2 +- mm/page_alloc.c | 4 ++-- 4 files changed, 35 insertions(+), 34 deletions(-)
diff --git a/include/linux/mm.h b/include/linux/mm.h index 81562397e8347..a7783b57c90eb 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -4262,4 +4262,36 @@ static inline bool pfn_is_unaccepted_memory(unsigned long pfn) void vma_pgtable_walk_begin(struct vm_area_struct *vma); void vma_pgtable_walk_end(struct vm_area_struct *vma);
+int reserve_mem_find_by_name(const char *name, phys_addr_t *start, phys_addr_t *size); + +#ifdef CONFIG_MEM_ALLOC_PROFILING +static inline void pgalloc_tag_split(struct folio *folio, int old_order, int new_order) +{ + int i; + struct alloc_tag *tag; + unsigned int nr_pages = 1 << new_order; + + if (!mem_alloc_profiling_enabled()) + return; + + tag = pgalloc_tag_get(&folio->page); + if (!tag) + return; + + for (i = nr_pages; i < (1 << old_order); i += nr_pages) { + union codetag_ref *ref = get_page_tag_ref(folio_page(folio, i)); + + if (ref) { + /* Set new reference to point to the original tag */ + alloc_tag_ref_set(ref, tag); + put_page_tag_ref(ref); + } + } +} +#else /* !CONFIG_MEM_ALLOC_PROFILING */ +static inline void pgalloc_tag_split(struct folio *folio, int old_order, int new_order) +{ +} +#endif /* CONFIG_MEM_ALLOC_PROFILING */ + #endif /* _LINUX_MM_H */ diff --git a/include/linux/pgalloc_tag.h b/include/linux/pgalloc_tag.h index 207f0c83c8e97..59a3deb792a8d 100644 --- a/include/linux/pgalloc_tag.h +++ b/include/linux/pgalloc_tag.h @@ -80,36 +80,6 @@ static inline void pgalloc_tag_sub(struct page *page, unsigned int nr) } }
-static inline void pgalloc_tag_split(struct page *page, unsigned int nr) -{ - int i; - struct page_ext *first_page_ext; - struct page_ext *page_ext; - union codetag_ref *ref; - struct alloc_tag *tag; - - if (!mem_alloc_profiling_enabled()) - return; - - first_page_ext = page_ext = page_ext_get(page); - if (unlikely(!page_ext)) - return; - - ref = codetag_ref_from_page_ext(page_ext); - if (!ref->ct) - goto out; - - tag = ct_to_alloc_tag(ref->ct); - page_ext = page_ext_next(page_ext); - for (i = 1; i < nr; i++) { - /* Set new reference to point to the original tag */ - alloc_tag_ref_set(codetag_ref_from_page_ext(page_ext), tag); - page_ext = page_ext_next(page_ext); - } -out: - page_ext_put(first_page_ext); -} - static inline struct alloc_tag *pgalloc_tag_get(struct page *page) { struct alloc_tag *tag = NULL; @@ -142,7 +112,6 @@ static inline void clear_page_tag_ref(struct page *page) {} static inline void pgalloc_tag_add(struct page *page, struct task_struct *task, unsigned int nr) {} static inline void pgalloc_tag_sub(struct page *page, unsigned int nr) {} -static inline void pgalloc_tag_split(struct page *page, unsigned int nr) {} static inline struct alloc_tag *pgalloc_tag_get(struct page *page) { return NULL; } static inline void pgalloc_tag_sub_pages(struct alloc_tag *tag, unsigned int nr) {}
diff --git a/mm/huge_memory.c b/mm/huge_memory.c index cedc93028894c..68d1da39d2a91 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -2896,7 +2896,7 @@ static void __split_huge_page(struct page *page, struct list_head *list, /* Caller disabled irqs, so they are still disabled here */
split_page_owner(head, order, new_order); - pgalloc_tag_split(head, 1 << order); + pgalloc_tag_split(folio, order, new_order);
/* See comment in __split_huge_page_tail() */ if (folio_test_anon(folio)) { diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 21016573f1870..3707b022395a1 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2745,7 +2745,7 @@ void split_page(struct page *page, unsigned int order) for (i = 1; i < (1 << order); i++) set_page_refcounted(page + i); split_page_owner(page, order, 0); - pgalloc_tag_split(page, 1 << order); + pgalloc_tag_split(page_folio(page), order, 0); split_page_memcg(page, order, 0); } EXPORT_SYMBOL_GPL(split_page); @@ -4937,7 +4937,7 @@ static void *make_alloc_exact(unsigned long addr, unsigned int order, struct page *last = page + nr;
split_page_owner(page, order, 0); - pgalloc_tag_split(page, 1 << order); + pgalloc_tag_split(page_folio(page), order, 0); split_page_memcg(page, order, 0); while (page < --last) set_page_refcounted(last);