The quilt patch titled Subject: userfaultfd: change src_folio after ensuring it's unpinned in UFFDIO_MOVE has been removed from the -mm tree. Its filename was userfaultfd-change-src_folio-after-ensuring-its-unpinned-in-uffdio_move.patch

This patch was dropped because it was merged into the mm-hotfixes-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

------------------------------------------------------ From: Lokesh Gidra lokeshgidra@google.com Subject: userfaultfd: change src_folio after ensuring it's unpinned in UFFDIO_MOVE Date: Thu, 4 Apr 2024 10:17:26 -0700

Commit d7a08838ab74 ("mm: userfaultfd: fix unexpected change to src_folio when UFFDIO_MOVE fails") moved the src_folio->{mapping, index} changing to after clearing the page-table and ensuring that it's not pinned. This avoids failure of swapout+migration and possibly memory corruption.

However, the commit missed fixing it in the huge-page case.

Link: https://lkml.kernel.org/r/20240404171726.2302435-1-lokeshgidra@google.com Fixes: adef440691ba ("userfaultfd: UFFDIO_MOVE uABI") Signed-off-by: Lokesh Gidra lokeshgidra@google.com Acked-by: David Hildenbrand david@redhat.com Cc: Andrea Arcangeli aarcange@redhat.com Cc: Kalesh Singh kaleshsingh@google.com Cc: Lokesh Gidra lokeshgidra@google.com Cc: Nicolas Geoffray ngeoffray@google.com Cc: Peter Xu peterx@redhat.com Cc: Qi Zheng zhengqi.arch@bytedance.com Cc: Matthew Wilcox willy@infradead.org Cc: stable@vger.kernel.org Signed-off-by: Andrew Morton akpm@linux-foundation.org ---

mm/huge_memory.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)

--- a/mm/huge_memory.c~userfaultfd-change-src_folio-after-ensuring-its-unpinned-in-uffdio_move +++ a/mm/huge_memory.c @@ -2259,9 +2259,6 @@ int move_pages_huge_pmd(struct mm_struct goto unlock_ptls; }

- folio_move_anon_rmap(src_folio, dst_vma); - WRITE_ONCE(src_folio->index, linear_page_index(dst_vma, dst_addr)); - src_pmdval = pmdp_huge_clear_flush(src_vma, src_addr, src_pmd); /* Folio got pinned from under us. Put it back and fail the move. */ if (folio_maybe_dma_pinned(src_folio)) { @@ -2270,6 +2267,9 @@ int move_pages_huge_pmd(struct mm_struct goto unlock_ptls; }

+ folio_move_anon_rmap(src_folio, dst_vma); + WRITE_ONCE(src_folio->index, linear_page_index(dst_vma, dst_addr)); + _dst_pmd = mk_huge_pmd(&src_folio->page, dst_vma->vm_page_prot); /* Follow mremap() behavior and treat the entry dirty after the move */ _dst_pmd = pmd_mkwrite(pmd_mkdirty(_dst_pmd), dst_vma); _

Patches currently in -mm which might be from lokeshgidra@google.com are