Hi, Now migrating a hugetlb page or unmapping a poisoned hugetlb page, we'll use ptep_clear_flush() and set_pte_at() to nuke the page table entry and remap it, and this is incorrect for CONT-PTE or CONT-PMD size hugetlb page, which will cause potential data consistent issue. This patch set will change to use hugetlb related APIs to fix this issue, please find details in each patch. Thanks.
Note: Mike pointed out the huge_ptep_get() will only return the one specific value, and it would not take into account the dirty or young bits of CONT-PTE/PMDs like the huge_ptep_get_and_clear() [1]. This inconsistent issue is not introduced by this patch set, and will address this issue in another thread [2]. Meanwhile the uffd for hugetlb case [3] pointed by Gerald also need another patch to address. [1] https://lore.kernel.org/linux-mm/85bd80b4-b4fd-0d3f-a2e5-149559f2f...@oracle.com/ [2] https://lore.kernel.org/all/cover.1651998586.git.baolin.w...@linux.alibaba.com/ [3] https://lore.kernel.org/linux-mm/20220503120343.6264e126@thinkpad/ Changes from v1: - Add acked tag from Mike. - Update some commit message. - Add VM_BUG_ON in try_to_unmap() for hugetlb case. - Add an explict void casting for huge_ptep_clear_flush() in hugetlb.c. Baolin Wang (3): mm: change huge_ptep_clear_flush() to return the original pte mm: rmap: Fix CONT-PTE/PMD size hugetlb issue when migration mm: rmap: Fix CONT-PTE/PMD size hugetlb issue when unmapping arch/arm64/include/asm/hugetlb.h | 4 +-- arch/arm64/mm/hugetlbpage.c | 12 +++----- arch/ia64/include/asm/hugetlb.h | 4 +-- arch/mips/include/asm/hugetlb.h | 9 ++++-- arch/parisc/include/asm/hugetlb.h | 4 +-- arch/powerpc/include/asm/hugetlb.h | 9 ++++-- arch/s390/include/asm/hugetlb.h | 6 ++-- arch/sh/include/asm/hugetlb.h | 4 +-- arch/sparc/include/asm/hugetlb.h | 4 +-- include/asm-generic/hugetlb.h | 4 +-- mm/hugetlb.c | 2 +- mm/rmap.c | 63 ++++++++++++++++++++++++-------------- 12 files changed, 73 insertions(+), 52 deletions(-) -- 1.8.3.1
