* [merged mm-stable] mm-userfaultfd-add-rwp-fault-delivery-and-expose-uffdio_register_mode_rwp.patch removed from -mm tree
@ 2026-08-05 2:23 Andrew Morton
0 siblings, 0 replies; only message in thread
From: Andrew Morton @ 2026-08-05 2:23 UTC (permalink / raw)
To: mm-commits, ziy, vbabka, surenb, sj, seanjc, rppt, peterx,
pbonzini, ljs, liam, jthoughton, david, corbet, aarcange, kas,
akpm
The quilt patch titled
Subject: mm/userfaultfd: add RWP fault delivery and expose UFFDIO_REGISTER_MODE_RWP
has been removed from the -mm tree. Its filename was
mm-userfaultfd-add-rwp-fault-delivery-and-expose-uffdio_register_mode_rwp.patch
This patch was dropped because it was merged into the mm-stable branch
of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
------------------------------------------------------
From: "Kiryl Shutsemau (Meta)" <kas@kernel.org>
Subject: mm/userfaultfd: add RWP fault delivery and expose UFFDIO_REGISTER_MODE_RWP
Date: Wed, 8 Jul 2026 12:14:11 +0100
Wire the fault side of read-write protection tracking and turn the
userspace interface on.
An RWP-protected PTE is PAGE_NONE with the uffd bit set. The PROT_NONE
triggers a fault on any access; the uffd bit distinguishes it from plain
mprotect(PROT_NONE) or NUMA hinting.
Fault dispatch, per level:
PTE handle_pte_fault() -> do_uffd_rwp()
PMD __handle_mm_fault() -> do_huge_pmd_uffd_rwp()
hugetlb hugetlb_fault() -> hugetlb_handle_userfault()
The RWP branches gate on userfaultfd_pte_rwp() /
userfaultfd_huge_pmd_rwp() (VM_UFFD_RWP plus the uffd bit) and fall
through to do_numa_page() / do_huge_pmd_numa_page() otherwise. Each
delivers a UFFD_PAGEFAULT_FLAG_RWP message through handle_userfault(); the
handler resolves it with UFFDIO_RWPROTECT clearing MODE_RWP.
userfaultfd_must_wait() and userfaultfd_huge_must_wait() add matching
protnone+uffd waiters so sync-mode fault handlers block correctly.
Expose the UAPI:
UFFDIO_REGISTER_MODE_RWP -> UFFD_API_REGISTER_MODES
UFFD_FEATURE_RWP -> UFFD_API_FEATURES
_UFFDIO_RWPROTECT -> UFFD_API_RANGE_IOCTLS
UFFD_API_RANGE_IOCTLS_BASIC
UFFD_FEATURE_RWP is masked out at UFFDIO_API time when PROT_NONE is not
available or VM_UFFD_RWP aliases VM_NONE (32-bit), so userspace never sees
an advertised-but-broken feature.
Works on anonymous, shmem, and hugetlb memory.
Link: https://lore.kernel.org/20260708111417.173443-11-kirill@shutemov.name
Signed-off-by: Kiryl Shutsemau <kas@kernel.org>
Assisted-by: Claude:claude-opus-4-6
Reviewed-by: Mike Rapoport (Microsoft) <rppt@kernel.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: David Hildenbrand <david@kernel.org>
Cc: James Houghton <jthoughton@google.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Liam Howlett <liam@infradead.org>
Cc: Lorenzo Stoakes <ljs@kernel.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Sean Christopherson <seanjc@google.com>
Cc: SeongJae Park <sj@kernel.org>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Vlastimil Babka <vbabka@kernel.org>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
include/linux/huge_mm.h | 7 ++++++
include/linux/userfaultfd_k.h | 24 +++++++++++++++++++++
include/uapi/linux/userfaultfd.h | 12 +++++++---
mm/huge_memory.c | 5 ++++
mm/hugetlb.c | 11 +++++++++
mm/memory.c | 31 ++++++++++++++++++++++++++--
mm/userfaultfd.c | 32 +++++++++++++++++++++++++++--
7 files changed, 114 insertions(+), 8 deletions(-)
--- a/include/linux/huge_mm.h~mm-userfaultfd-add-rwp-fault-delivery-and-expose-uffdio_register_mode_rwp
+++ a/include/linux/huge_mm.h
@@ -529,6 +529,8 @@ static inline bool folio_test_pmd_mappab
vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf);
+vm_fault_t do_huge_pmd_uffd_rwp(struct vm_fault *vmf);
+
vm_fault_t do_huge_pmd_device_private(struct vm_fault *vmf);
extern struct folio *huge_zero_folio;
@@ -723,6 +725,11 @@ static inline spinlock_t *pud_trans_huge
return NULL;
}
+static inline vm_fault_t do_huge_pmd_uffd_rwp(struct vm_fault *vmf)
+{
+ return 0;
+}
+
static inline vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf)
{
return 0;
--- a/include/linux/userfaultfd_k.h~mm-userfaultfd-add-rwp-fault-delivery-and-expose-uffdio_register_mode_rwp
+++ a/include/linux/userfaultfd_k.h
@@ -241,6 +241,18 @@ static inline bool userfaultfd_huge_pmd_
return userfaultfd_wp(vma) && pmd_uffd(pmd);
}
+static inline bool userfaultfd_pte_rwp(struct vm_area_struct *vma,
+ pte_t pte)
+{
+ return userfaultfd_rwp(vma) && pte_uffd(pte);
+}
+
+static inline bool userfaultfd_huge_pmd_rwp(struct vm_area_struct *vma,
+ pmd_t pmd)
+{
+ return userfaultfd_rwp(vma) && pmd_uffd(pmd);
+}
+
static inline bool userfaultfd_armed(struct vm_area_struct *vma)
{
return vma_test_any_mask(vma, __VMA_UFFD_FLAGS);
@@ -370,6 +382,18 @@ static inline bool userfaultfd_huge_pmd_
{
return false;
}
+
+static inline bool userfaultfd_pte_rwp(struct vm_area_struct *vma,
+ pte_t pte)
+{
+ return false;
+}
+
+static inline bool userfaultfd_huge_pmd_rwp(struct vm_area_struct *vma,
+ pmd_t pmd)
+{
+ return false;
+}
static inline bool userfaultfd_armed(struct vm_area_struct *vma)
{
--- a/include/uapi/linux/userfaultfd.h~mm-userfaultfd-add-rwp-fault-delivery-and-expose-uffdio_register_mode_rwp
+++ a/include/uapi/linux/userfaultfd.h
@@ -25,7 +25,8 @@
#define UFFD_API ((__u64)0xAA)
#define UFFD_API_REGISTER_MODES (UFFDIO_REGISTER_MODE_MISSING | \
UFFDIO_REGISTER_MODE_WP | \
- UFFDIO_REGISTER_MODE_MINOR)
+ UFFDIO_REGISTER_MODE_MINOR | \
+ UFFDIO_REGISTER_MODE_RWP)
#define UFFD_API_FEATURES (UFFD_FEATURE_PAGEFAULT_FLAG_WP | \
UFFD_FEATURE_EVENT_FORK | \
UFFD_FEATURE_EVENT_REMAP | \
@@ -42,7 +43,8 @@
UFFD_FEATURE_WP_UNPOPULATED | \
UFFD_FEATURE_POISON | \
UFFD_FEATURE_WP_ASYNC | \
- UFFD_FEATURE_MOVE)
+ UFFD_FEATURE_MOVE | \
+ UFFD_FEATURE_RWP)
#define UFFD_API_IOCTLS \
((__u64)1 << _UFFDIO_REGISTER | \
(__u64)1 << _UFFDIO_UNREGISTER | \
@@ -54,13 +56,15 @@
(__u64)1 << _UFFDIO_MOVE | \
(__u64)1 << _UFFDIO_WRITEPROTECT | \
(__u64)1 << _UFFDIO_CONTINUE | \
- (__u64)1 << _UFFDIO_POISON)
+ (__u64)1 << _UFFDIO_POISON | \
+ (__u64)1 << _UFFDIO_RWPROTECT)
#define UFFD_API_RANGE_IOCTLS_BASIC \
((__u64)1 << _UFFDIO_WAKE | \
(__u64)1 << _UFFDIO_COPY | \
(__u64)1 << _UFFDIO_WRITEPROTECT | \
(__u64)1 << _UFFDIO_CONTINUE | \
- (__u64)1 << _UFFDIO_POISON)
+ (__u64)1 << _UFFDIO_POISON | \
+ (__u64)1 << _UFFDIO_RWPROTECT)
/*
* Valid ioctl command number range with this API is from 0x00 to
--- a/mm/huge_memory.c~mm-userfaultfd-add-rwp-fault-delivery-and-expose-uffdio_register_mode_rwp
+++ a/mm/huge_memory.c
@@ -2202,6 +2202,11 @@ static inline bool can_change_pmd_writab
return pmd_dirty(pmd);
}
+vm_fault_t do_huge_pmd_uffd_rwp(struct vm_fault *vmf)
+{
+ return handle_userfault(vmf, VM_UFFD_RWP);
+}
+
/* NUMA hinting page fault entry point for trans huge pmds */
vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf)
{
--- a/mm/hugetlb.c~mm-userfaultfd-add-rwp-fault-delivery-and-expose-uffdio_register_mode_rwp
+++ a/mm/hugetlb.c
@@ -6096,6 +6096,17 @@ vm_fault_t hugetlb_fault(struct mm_struc
}
/*
+ * Protnone hugetlb PTEs with the uffd bit are used by
+ * userfaultfd RWP for access tracking. Plain PROT_NONE (without the
+ * marker) is not an RWP fault and is not expected on hugetlb (no
+ * NUMA hinting), so let normal hugetlb fault handling proceed.
+ */
+ if (pte_protnone(vmf.orig_pte) && vma_is_accessible(vma) &&
+ userfaultfd_rwp(vma) && huge_pte_uffd(vmf.orig_pte)) {
+ return hugetlb_handle_userfault(&vmf, mapping, VM_UFFD_RWP);
+ }
+
+ /*
* If we are going to COW/unshare the mapping later, we examine the
* pending reservations for this page now. This will ensure that any
* allocations necessary to record that reservation occur outside the
--- a/mm/memory.c~mm-userfaultfd-add-rwp-fault-delivery-and-expose-uffdio_register_mode_rwp
+++ a/mm/memory.c
@@ -6119,6 +6119,16 @@ static void numa_rebuild_large_mapping(s
if (!pte_present(ptent) || !pte_protnone(ptent))
continue;
+ /*
+ * RWP-armed PTEs are also protnone but carry _PAGE_UFFD as a
+ * marker. Leave them alone -- rewriting to vm_page_prot would
+ * stop the RWP trap. Gate on userfaultfd_rwp(vma) too:
+ * NUMA balancing preserves _PAGE_UFFD on UFFD_WP-marked PTEs
+ * when applying PROT_NONE, and those still need rebuilding.
+ */
+ if (userfaultfd_rwp(vma) && pte_uffd(ptent))
+ continue;
+
if (pfn_folio(pte_pfn(ptent)) != folio)
continue;
@@ -6134,6 +6144,12 @@ static void numa_rebuild_large_mapping(s
}
}
+static vm_fault_t do_uffd_rwp(struct vm_fault *vmf)
+{
+ pte_unmap(vmf->pte);
+ return handle_userfault(vmf, VM_UFFD_RWP);
+}
+
static vm_fault_t do_numa_page(struct vm_fault *vmf)
{
struct vm_area_struct *vma = vmf->vma;
@@ -6409,8 +6425,16 @@ static vm_fault_t handle_pte_fault(struc
if (!pte_present(vmf->orig_pte))
return do_swap_page(vmf);
- if (pte_protnone(vmf->orig_pte) && vma_is_accessible(vmf->vma))
+ if (pte_protnone(vmf->orig_pte) && vma_is_accessible(vmf->vma)) {
+ /*
+ * RWP-protected PTEs are protnone plus the uffd bit. On a
+ * VM_UFFD_RWP VMA, a protnone PTE without the uffd bit is
+ * NUMA hinting and must still fall through to do_numa_page().
+ */
+ if (userfaultfd_pte_rwp(vmf->vma, vmf->orig_pte))
+ return do_uffd_rwp(vmf);
return do_numa_page(vmf);
+ }
spin_lock(vmf->ptl);
entry = vmf->orig_pte;
@@ -6524,8 +6548,11 @@ retry_pud:
return 0;
}
if (pmd_trans_huge(vmf.orig_pmd)) {
- if (pmd_protnone(vmf.orig_pmd) && vma_is_accessible(vma))
+ if (pmd_protnone(vmf.orig_pmd) && vma_is_accessible(vma)) {
+ if (userfaultfd_huge_pmd_rwp(vma, vmf.orig_pmd))
+ return do_huge_pmd_uffd_rwp(&vmf);
return do_huge_pmd_numa_page(&vmf);
+ }
if ((flags & (FAULT_FLAG_WRITE|FAULT_FLAG_UNSHARE)) &&
!pmd_write(vmf.orig_pmd)) {
--- a/mm/userfaultfd.c~mm-userfaultfd-add-rwp-fault-delivery-and-expose-uffdio_register_mode_rwp
+++ a/mm/userfaultfd.c
@@ -2669,6 +2669,12 @@ static inline bool userfaultfd_huge_must
*/
if (!huge_pte_write(pte) && (reason & VM_UFFD_WP))
return true;
+ /*
+ * PTE is still RW-protected (protnone with uffd bit), wait for
+ * resolution. Plain PROT_NONE without the marker is not an RWP fault.
+ */
+ if (pte_protnone(pte) && huge_pte_uffd(pte) && (reason & VM_UFFD_RWP))
+ return true;
return false;
}
@@ -2729,8 +2735,14 @@ again:
if (!pmd_present(_pmd))
return false;
- if (pmd_trans_huge(_pmd))
- return !pmd_write(_pmd) && (reason & VM_UFFD_WP);
+ if (pmd_trans_huge(_pmd)) {
+ if (!pmd_write(_pmd) && (reason & VM_UFFD_WP))
+ return true;
+ if (pmd_protnone(_pmd) && pmd_uffd(_pmd) &&
+ (reason & VM_UFFD_RWP))
+ return true;
+ return false;
+ }
pte = pte_offset_map(pmd, address);
if (!pte)
@@ -2766,6 +2778,13 @@ again:
*/
if (!pte_write(ptent) && (reason & VM_UFFD_WP))
goto out;
+ /*
+ * PTE is still RW-protected (protnone with uffd bit), wait for
+ * userspace to resolve. Plain PROT_NONE without the marker is not
+ * an RWP fault.
+ */
+ if (pte_protnone(ptent) && pte_uffd(ptent) && (reason & VM_UFFD_RWP))
+ goto out;
ret = false;
out:
@@ -4518,6 +4537,15 @@ static int userfaultfd_api(struct userfa
uffdio_api.features &= ~UFFD_FEATURE_WP_UNPOPULATED;
uffdio_api.features &= ~UFFD_FEATURE_WP_ASYNC;
}
+ /*
+ * RWP needs both PROT_NONE support and the uffd-wp PTE bit. The
+ * VM_UFFD_RWP check covers compile-time unavailability; the
+ * pgtable_supports_uffd() check covers runtime (e.g. riscv
+ * without the SVRSW60T59B extension) where the PTE bit is declared
+ * but not actually usable.
+ */
+ if (VM_UFFD_RWP == VM_NONE || !pgtable_supports_uffd())
+ uffdio_api.features &= ~UFFD_FEATURE_RWP;
ret = -EINVAL;
if (features & ~uffdio_api.features)
_
Patches currently in -mm which might be from kas@kernel.org are
selftests-mm-add-pagemap_scan-test-for-thp-pmd-holes.patch
mm-huge_memory-remove-unused-can_split_folio.patch
mm-huge_memory-fold-split_folio_to_list_to_order-into-split_folio_to_order.patch
mm-fold-userfaultfd_rwp-to-false-without-config_arch_has_pte_protnone.patch
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2026-08-05 2:23 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-08-05 2:23 [merged mm-stable] mm-userfaultfd-add-rwp-fault-delivery-and-expose-uffdio_register_mode_rwp.patch removed from -mm tree Andrew Morton
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.