【OLK-6.6】Backport some folio feature and bugfix in mm from mainline and next

1. 性能优化及bugfix:
mm/vmalloc: fix return value of vb_alloc if size is 0
mm: use memalloc_nofs_save() in page_cache_ra_order()
mm: madvise: pageout: ignore references rather than clearing young
mm: ratelimit stat flush from workingset shrinker
mm, oom:dump_tasks add rss detailed information printing
arm64/mm: improve comment in contpte_ptep_get_lockless()
arm64/mm: export contpte symbols only to GPL users
mm: page_alloc: enforce minimum zone size to do high atomic reserves
mm: page_alloc: correct high atomic reserve calculations
mm: fix draining remote pageset

2. split folio to any order:
selftests: mm: fix unused and uninitialized variable warning
XArray: set the marks correctly when splitting an entry
mm/huge_memory: check new folio order when split a folio
mm: huge_memory: enable debugfs to split huge pages to any order
mm: thp: split huge page to any lower order pages
mm: page_owner: add support for splitting to any order in split page_owner
mm: memcg: make memcg huge page split support any order split
mm/page_owner: use order instead of nr in split_page_owner()
mm/memcg: use order instead of nr in split_page_memcg()
mm: support order-1 folios in the page cache
mm/huge_memory: only split PMD mapping when necessary in unmap_folio()
selftests/mm: split_huge_page_test: conform test to TAP format output
mm: huge_memory: use more folio api in __split_huge_page_tail()

3. cleanup:
mm: remove struct page from get_shadow_from_swap_cache

4. arm64 mte支持THP_SWAP:
arm64: mm: swap: support THP_SWAP on hardware with MTE
mm: remove PageAnonExclusive assertions in unuse_pte()

5. 预读性能优化：
mm/filemap: don't decrease mmap_miss when folio has workingset flag
mm/readahead: break read-ahead loop if filemap_add_folio return -ENOMEM

6. folio小优化及bugfix:
mm: alloc_anon_folio: avoid doing vma_thp_gfp_mask in fallback cases
huge_memory.c: document huge page splitting rules more thoroughly
mm: convert folio_estimated_sharers() to folio_likely_mapped_shared()
mm/migrate: split source folio if it is on deferred split list
mm: hold PTL from the first PTE while reclaiming a large folio
madvise:madvise_cold_or_pageout_pte_range(): allow split while folio_estimated_sharers = 0

7. numa balance 支持large folio:
mm: support multi-size THP numa balancing
mm: factor out the numa mapping rebuilding into a new helper

8. swap out 支持large folio不拆分:
mm: madvise: avoid split during MADV_PAGEOUT and MADV_COLD
mm: vmscan: avoid split during shrink_folio_list()
mm: swap: allow storage of all mTHP orders
mm: swap: update get_swap_pages() to take folio order
mm: swap: simplify struct percpu_cluster
mm: swap: free_swap_and_cache_nr() as batched free_swap_and_cache()
mm: swap: remove CLUSTER_FLAG_HUGE from swap_cluster_info:flags
mm/madvise: don't forget to leave lazy MMU mode in madvise_cold_or_pageout_pte_range()
mm/madvise: add cond_resched() in madvise_cold_or_pageout_pte_range()

9. 支持more high-order pages pcplist:
mm: prepare more high-order pages to be stored on the per-cpu lists
mm: page_alloc: use the correct THP order for THP PCP

GVP openEuler/kernel

内容风险标识

评论 (2)

GVPopenEuler/kernel

内容风险标识