v2
2026-11-02
GUP批量抓取大folio · v2
在gup_fast_pte_range()中按folio批量处理PTE映射的大folio，减少每页try_grab_folio_fast()、pin计数和stats的重复开销。
原始补丁链接
子 Patch（4）
1/4
mm: gup: introduce gup_unshared_pages()
This is preparatory patch to make gup_fast_pte_range() batchable. Introduce gup_unshared_pages() to count consecutive pages within a folio that do not require unsharing, and use it to implement gup_m
2/4
mm: internal: fix inconsistent in folio_pte_batch_flags() for shadow stack
In x86, shadow stack PTE is marked with PAGE_RW=0 and PAGE_DIRTY=1. If the shadow stack pte entry is consequence with the read-only pte like PAGE_RW=0 and PAGE_DIRTY=0, folio_pte_batch_flags() with FP
3/4
mm: internal: introduce FPB_LOCKLESS for folio_pte_batch_flags()
From: Yuan-Hao Hsu aa9736195201@gmail.com Add support for calling folio_pte_batch_flags() in GUP-fast context when we are not holding the PT lock. Instead of reading PTEs through ptep_get(), we hav
4/4
mm: gup: batch PTE-mapped large folios in gup_fast_pte_range()
[mm: gup: batch PTE-mapped large folios in gup_fast_pte_range()] [patchwork series v2] submitter=Yeoreum Yun patches (4): [v2,1/4] mm: gup: introduce gup_unshared_pages() [v2,2/4] mm: internal: fi