The quilt patch titled
Subject: mm/hugetlb: take src_mm->write_protect_seq in copy_hugetlb_page_range()
has been removed from the -mm tree. Its filename was
mm-hugetlb-take-src_mm-write_protect_seq-in-copy_hugetlb_page_range.patch
This patch was dropped because it was merged into mm-stable
------------------------------------------------------
From: David Hildenbrand <david@xxxxxxxxxx>
Subject: mm/hugetlb: take src_mm->write_protect_seq in copy_hugetlb_page_range()
Let's do it just like copy_page_range(), taking the seqlock and making
sure the mmap_lock is held in write mode.
This allows for add a VM_BUG_ON to page_needs_cow_for_dma() and properly
synchronizes concurrent fork() with GUP-fast of hugetlb pages, which will
be relevant for further changes.
Link: https://lkml.kernel.org/r/20220428083441.37290-3-david@xxxxxxxxxx
Signed-off-by: David Hildenbrand <david@xxxxxxxxxx>
Acked-by: Vlastimil Babka <vbabka@xxxxxxx>
Cc: Andrea Arcangeli <aarcange@xxxxxxxxxx>
Cc: Christoph Hellwig <hch@xxxxxx>
Cc: David Rientjes <rientjes@xxxxxxxxxx>
Cc: Don Dutile <ddutile@xxxxxxxxxx>
Cc: Hugh Dickins <hughd@xxxxxxxxxx>
Cc: Jan Kara <jack@xxxxxxx>
Cc: Jann Horn <jannh@xxxxxxxxxx>
Cc: Jason Gunthorpe <jgg@xxxxxxxxxx>
Cc: John Hubbard <jhubbard@xxxxxxxxxx>
Cc: Khalid Aziz <khalid.aziz@xxxxxxxxxx>
Cc: "Kirill A. Shutemov" <kirill.shutemov@xxxxxxxxxxxxxxx>
Cc: Liang Zhang <zhangliang5@xxxxxxxxxx>
Cc: "Matthew Wilcox (Oracle)" <willy@xxxxxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxxxx>
Cc: Mike Kravetz <mike.kravetz@xxxxxxxxxx>
Cc: Mike Rapoport <rppt@xxxxxxxxxxxxx>
Cc: Nadav Amit <namit@xxxxxxxxxx>
Cc: Oded Gabbay <oded.gabbay@xxxxxxxxx>
Cc: Oleg Nesterov <oleg@xxxxxxxxxx>
Cc: Pedro Demarchi Gomes <pedrodemargomes@xxxxxxxxx>
Cc: Peter Xu <peterx@xxxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxxx>
Cc: Roman Gushchin <guro@xxxxxx>
Cc: Shakeel Butt <shakeelb@xxxxxxxxxx>
Cc: Yang Shi <shy828301@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/mm.h | 4 ++++
mm/hugetlb.c | 8 ++++++--
2 files changed, 10 insertions(+), 2 deletions(-)
--- a/include/linux/mm.h~mm-hugetlb-take-src_mm-write_protect_seq-in-copy_hugetlb_page_range
+++ a/include/linux/mm.h
@@ -1576,6 +1576,8 @@ static inline bool page_maybe_dma_pinned
/*
* This should most likely only be called during fork() to see whether we
* should break the cow immediately for a page on the src mm.
+ *
+ * The caller has to hold the PT lock and the vma->vm_mm->->write_protect_seq.
*/
static inline bool page_needs_cow_for_dma(struct vm_area_struct *vma,
struct page *page)
@@ -1583,6 +1585,8 @@ static inline bool page_needs_cow_for_dm
if (!is_cow_mapping(vma->vm_flags))
return false;
+ VM_BUG_ON(!(raw_read_seqcount(&vma->vm_mm->write_protect_seq) & 1));
+
if (!test_bit(MMF_HAS_PINNED, &vma->vm_mm->flags))
return false;
--- a/mm/hugetlb.c~mm-hugetlb-take-src_mm-write_protect_seq-in-copy_hugetlb_page_range
+++ a/mm/hugetlb.c
@@ -4735,6 +4735,8 @@ int copy_hugetlb_page_range(struct mm_st
vma->vm_start,
vma->vm_end);
mmu_notifier_invalidate_range_start(&range);
+ mmap_assert_write_locked(src);
+ raw_write_seqcount_begin(&src->write_protect_seq);
} else {
/*
* For shared mappings i_mmap_rwsem must be held to call
@@ -4867,10 +4869,12 @@ again:
spin_unlock(dst_ptl);
}
- if (cow)
+ if (cow) {
+ raw_write_seqcount_end(&src->write_protect_seq);
mmu_notifier_invalidate_range_end(&range);
- else
+ } else {
i_mmap_unlock_read(mapping);
+ }
return ret;
}
_
Patches currently in -mm which might be from david@xxxxxxxxxx are
