The patch titled
Subject: userfaultfd: wp: support swap and page migration
has been added to the -mm tree. Its filename is
userfaultfd-wp-support-swap-and-page-migration.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/userfaultfd-wp-support-swap-and-page-migration.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/userfaultfd-wp-support-swap-and-page-migration.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Peter Xu <peterx@xxxxxxxxxx>
Subject: userfaultfd: wp: support swap and page migration
For either swap and page migration, we all use the bit 2 of the entry to
identify whether this entry is uffd write-protected. It plays a similar
role as the existing soft dirty bit in swap entries but only for keeping
the uffd-wp tracking for a specific PTE/PMD.
Something special here is that when we want to recover the uffd-wp bit
from a swap/migration entry to the PTE bit we'll also need to take care of
the _PAGE_RW bit and make sure it's cleared, otherwise even with the
_PAGE_UFFD_WP bit we can't trap it at all.
In change_pte_range() we do nothing for uffd if the PTE is a swap entry.
That can lead to data mismatch if the page that we are going to write
protect is swapped out when sending the UFFDIO_WRITEPROTECT. This patch
also applies/removes the uffd-wp bit even for the swap entries.
Link: http://lkml.kernel.org/r/20200220163112.11409-11-peterx@xxxxxxxxxx
Signed-off-by: Peter Xu <peterx@xxxxxxxxxx>
Cc: Andrea Arcangeli <aarcange@xxxxxxxxxx>
Cc: Bobby Powers <bobbypowers@xxxxxxxxx>
Cc: Brian Geffon <bgeffon@xxxxxxxxxx>
Cc: David Hildenbrand <david@xxxxxxxxxx>
Cc: Denis Plotnikov <dplotnikov@xxxxxxxxxxxxx>
Cc: "Dr . David Alan Gilbert" <dgilbert@xxxxxxxxxx>
Cc: Hugh Dickins <hughd@xxxxxxxxxx>
Cc: Jerome Glisse <jglisse@xxxxxxxxxx>
Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: "Kirill A . Shutemov" <kirill@xxxxxxxxxxxxx>
Cc: Martin Cracauer <cracauer@xxxxxxxx>
Cc: Marty McFadden <mcfadden8@xxxxxxxx>
Cc: Maya Gokhale <gokhale2@xxxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxx>
Cc: Mike Kravetz <mike.kravetz@xxxxxxxxxx>
Cc: Mike Rapoport <rppt@xxxxxxxxxxxxxxxxxx>
Cc: Pavel Emelyanov <xemul@xxxxxxxxxxxxx>
Cc: Pavel Emelyanov <xemul@xxxxxxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxx>
Cc: Shaohua Li <shli@xxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/swapops.h | 2 ++
mm/huge_memory.c | 3 +++
mm/memory.c | 8 ++++++++
mm/migrate.c | 6 ++++++
mm/mprotect.c | 28 +++++++++++++++++-----------
mm/rmap.c | 6 ++++++
6 files changed, 42 insertions(+), 11 deletions(-)
--- a/include/linux/swapops.h~userfaultfd-wp-support-swap-and-page-migration
+++ a/include/linux/swapops.h
@@ -68,6 +68,8 @@ static inline swp_entry_t pte_to_swp_ent
if (pte_swp_soft_dirty(pte))
pte = pte_swp_clear_soft_dirty(pte);
+ if (pte_swp_uffd_wp(pte))
+ pte = pte_swp_clear_uffd_wp(pte);
arch_entry = __pte_to_swp_entry(pte);
return swp_entry(__swp_type(arch_entry), __swp_offset(arch_entry));
}
--- a/mm/huge_memory.c~userfaultfd-wp-support-swap-and-page-migration
+++ a/mm/huge_memory.c
@@ -2271,6 +2271,7 @@ static void __split_huge_pmd_locked(stru
write = is_write_migration_entry(entry);
young = false;
soft_dirty = pmd_swp_soft_dirty(old_pmd);
+ uffd_wp = pmd_swp_uffd_wp(old_pmd);
} else {
page = pmd_page(old_pmd);
if (pmd_dirty(old_pmd))
@@ -2303,6 +2304,8 @@ static void __split_huge_pmd_locked(stru
entry = swp_entry_to_pte(swp_entry);
if (soft_dirty)
entry = pte_swp_mksoft_dirty(entry);
+ if (uffd_wp)
+ entry = pte_swp_mkuffd_wp(entry);
} else {
entry = mk_pte(page + i, READ_ONCE(vma->vm_page_prot));
entry = maybe_mkwrite(entry, vma);
--- a/mm/memory.c~userfaultfd-wp-support-swap-and-page-migration
+++ a/mm/memory.c
@@ -733,6 +733,8 @@ copy_one_pte(struct mm_struct *dst_mm, s
pte = swp_entry_to_pte(entry);
if (pte_swp_soft_dirty(*src_pte))
pte = pte_swp_mksoft_dirty(pte);
+ if (pte_swp_uffd_wp(*src_pte))
+ pte = pte_swp_mkuffd_wp(pte);
set_pte_at(src_mm, addr, src_pte, pte);
}
} else if (is_device_private_entry(entry)) {
@@ -762,6 +764,8 @@ copy_one_pte(struct mm_struct *dst_mm, s
is_cow_mapping(vm_flags)) {
make_device_private_entry_read(&entry);
pte = swp_entry_to_pte(entry);
+ if (pte_swp_uffd_wp(*src_pte))
+ pte = pte_swp_mkuffd_wp(pte);
set_pte_at(src_mm, addr, src_pte, pte);
}
}
@@ -3098,6 +3102,10 @@ vm_fault_t do_swap_page(struct vm_fault
flush_icache_page(vma, page);
if (pte_swp_soft_dirty(vmf->orig_pte))
pte = pte_mksoft_dirty(pte);
+ if (pte_swp_uffd_wp(vmf->orig_pte)) {
+ pte = pte_mkuffd_wp(pte);
+ pte = pte_wrprotect(pte);
+ }
set_pte_at(vma->vm_mm, vmf->address, vmf->pte, pte);
arch_do_swap_page(vma->vm_mm, vma, vmf->address, pte, vmf->orig_pte);
vmf->orig_pte = pte;
--- a/mm/migrate.c~userfaultfd-wp-support-swap-and-page-migration
+++ a/mm/migrate.c
@@ -243,11 +243,15 @@ static bool remove_migration_pte(struct
entry = pte_to_swp_entry(*pvmw.pte);
if (is_write_migration_entry(entry))
pte = maybe_mkwrite(pte, vma);
+ else if (pte_swp_uffd_wp(*pvmw.pte))
+ pte = pte_mkuffd_wp(pte);
if (unlikely(is_zone_device_page(new))) {
if (is_device_private_page(new)) {
entry = make_device_private_entry(new, pte_write(pte));
pte = swp_entry_to_pte(entry);
+ if (pte_swp_uffd_wp(*pvmw.pte))
+ pte = pte_mkuffd_wp(pte);
}
}
@@ -2334,6 +2338,8 @@ again:
swp_pte = swp_entry_to_pte(entry);
if (pte_soft_dirty(pte))
swp_pte = pte_swp_mksoft_dirty(swp_pte);
+ if (pte_uffd_wp(pte))
+ swp_pte = pte_swp_mkuffd_wp(swp_pte);
set_pte_at(mm, addr, ptep, swp_pte);
/*
--- a/mm/mprotect.c~userfaultfd-wp-support-swap-and-page-migration
+++ a/mm/mprotect.c
@@ -139,11 +139,11 @@ static unsigned long change_pte_range(st
}
ptep_modify_prot_commit(vma, addr, pte, oldpte, ptent);
pages++;
- } else if (IS_ENABLED(CONFIG_MIGRATION)) {
+ } else if (is_swap_pte(oldpte)) {
swp_entry_t entry = pte_to_swp_entry(oldpte);
+ pte_t newpte;
if (is_write_migration_entry(entry)) {
- pte_t newpte;
/*
* A protection check is difficult so
* just be safe and disable write
@@ -152,22 +152,28 @@ static unsigned long change_pte_range(st
newpte = swp_entry_to_pte(entry);
if (pte_swp_soft_dirty(oldpte))
newpte = pte_swp_mksoft_dirty(newpte);
- set_pte_at(vma->vm_mm, addr, pte, newpte);
-
- pages++;
- }
-
- if (is_write_device_private_entry(entry)) {
- pte_t newpte;
-
+ if (pte_swp_uffd_wp(oldpte))
+ newpte = pte_swp_mkuffd_wp(newpte);
+ } else if (is_write_device_private_entry(entry)) {
/*
* We do not preserve soft-dirtiness. See
* copy_one_pte() for explanation.
*/
make_device_private_entry_read(&entry);
newpte = swp_entry_to_pte(entry);
- set_pte_at(vma->vm_mm, addr, pte, newpte);
+ if (pte_swp_uffd_wp(oldpte))
+ newpte = pte_swp_mkuffd_wp(newpte);
+ } else {
+ newpte = oldpte;
+ }
+ if (uffd_wp)
+ newpte = pte_swp_mkuffd_wp(newpte);
+ else if (uffd_wp_resolve)
+ newpte = pte_swp_clear_uffd_wp(newpte);
+
+ if (!pte_same(oldpte, newpte)) {
+ set_pte_at(vma->vm_mm, addr, pte, newpte);
pages++;
}
}
--- a/mm/rmap.c~userfaultfd-wp-support-swap-and-page-migration
+++ a/mm/rmap.c
@@ -1502,6 +1502,8 @@ static bool try_to_unmap_one(struct page
swp_pte = swp_entry_to_pte(entry);
if (pte_soft_dirty(pteval))
swp_pte = pte_swp_mksoft_dirty(swp_pte);
+ if (pte_uffd_wp(pteval))
+ swp_pte = pte_swp_mkuffd_wp(swp_pte);
set_pte_at(mm, pvmw.address, pvmw.pte, swp_pte);
/*
* No need to invalidate here it will synchronize on
@@ -1601,6 +1603,8 @@ static bool try_to_unmap_one(struct page
swp_pte = swp_entry_to_pte(entry);
if (pte_soft_dirty(pteval))
swp_pte = pte_swp_mksoft_dirty(swp_pte);
+ if (pte_uffd_wp(pteval))
+ swp_pte = pte_swp_mkuffd_wp(swp_pte);
set_pte_at(mm, address, pvmw.pte, swp_pte);
/*
* No need to invalidate here it will synchronize on
@@ -1667,6 +1671,8 @@ static bool try_to_unmap_one(struct page
swp_pte = swp_entry_to_pte(entry);
if (pte_soft_dirty(pteval))
swp_pte = pte_swp_mksoft_dirty(swp_pte);
+ if (pte_uffd_wp(pteval))
+ swp_pte = pte_swp_mkuffd_wp(swp_pte);
set_pte_at(mm, address, pvmw.pte, swp_pte);
/* Invalidate as we cleared the pte */
mmu_notifier_invalidate_range(mm, address,
_
Patches currently in -mm which might be from peterx@xxxxxxxxxx are
mm-gup-rename-nonblocking-to-locked-where-proper.patch
mm-gup-fix-__get_user_pages-on-fault-retry-of-hugetlb.patch
mm-introduce-fault_signal_pending.patch
x86-mm-use-helper-fault_signal_pending.patch
arc-mm-use-helper-fault_signal_pending.patch
arm64-mm-use-helper-fault_signal_pending.patch
powerpc-mm-use-helper-fault_signal_pending.patch
sh-mm-use-helper-fault_signal_pending.patch
mm-return-faster-for-non-fatal-signals-in-user-mode-faults.patch
userfaultfd-dont-retake-mmap_sem-to-emulate-nopage.patch
mm-introduce-fault_flag_default.patch
mm-introduce-fault_flag_interruptible.patch
mm-allow-vm_fault_retry-for-multiple-times.patch
mm-gup-allow-vm_fault_retry-for-multiple-times.patch
mm-gup-allow-to-react-to-fatal-signals.patch
mm-userfaultfd-honor-fault_flag_killable-in-fault-path.patch
mm-merge-parameters-for-change_protection.patch
userfaultfd-wp-apply-_page_uffd_wp-bit.patch
userfaultfd-wp-drop-_page_uffd_wp-properly-when-fork.patch
userfaultfd-wp-add-pmd_swp_uffd_wp-helpers.patch
userfaultfd-wp-support-swap-and-page-migration.patch
khugepaged-skip-collapse-if-uffd-wp-detected.patch
userfaultfd-wp-dont-wake-up-when-doing-write-protect.patch
userfaultfd-wp-declare-_uffdio_writeprotect-conditionally.patch
userfaultfd-selftests-refactor-statistics.patch
userfaultfd-selftests-add-write-protect-test.patch
