[PATCH v7 0/8] mm/hmm: Add mmap lock-drop support for userfaultfd-backed mappings

[PATCH v7 0/8] mm/hmm: Add mmap lock-drop support for userfaultfd-backed mappings

[Stanislav Kinsburskii]

This series extends the HMM framework to support userfaultfd-backed memory
by allowing the mmap read lock to be dropped during hmm_range_fault().

Some page fault handlers — most notably userfaultfd — require the mmap lock
to be released so that userspace can resolve the fault. The current HMM
interface never sets FAULT_FLAG_ALLOW_RETRY, making it impossible to fault
in pages from userfaultfd-registered regions.

This series follows the established int *locked pattern from
get_user_pages_remote() in mm/gup.c. A new helper function,
hmm_range_fault_locked(), accepts an int *locked parameter. When the
mmap lock is dropped during fault resolution (VM_FAULT_RETRY or
VM_FAULT_COMPLETED), the function returns 0 with *locked = 0, signalling
the caller to restart its walk. The existing hmm_range_fault() is
refactored into a thin wrapper that passes NULL, preserving current
behavior for all existing callers.

Possible approaches to lift this limitation are documented in
Documentation/mm/hmm.rst.

Changes in v7:
  - Replaced the unlocked HMM API with
    hmm_range_fault_unlocked_timeout(). The helper now takes a timeout in
    jiffies, with 0 meaning retry indefinitely.
  - Moved -EBUSY retry handling into the HMM helper for the unlocked path.
    The helper refreshes range->notifier_seq internally before each retry.
  - Switched the unlocked path to mmap_read_lock_killable() and return
    -EINTR if mmap lock acquisition is interrupted or a fatal signal is
    pending during retry handling.
  - Removed the redundant non-timeout hmm_range_fault_unlocked() interface.
  - Updated Documentation/mm/hmm.rst and kernel-doc to describe the timeout API
    and the intended caller pattern.
  - Updated the HMM selftests to use hmm_range_fault_unlocked_timeout()
    only, including coverage for the finite-timeout path.
  - Added in-tree users of the new helper:
      - mshv
      - nouveau
      - RDMA/umem
      - amdxdna
      - drm/gpusvm
  - Preserved each converted driver’s existing timeout convention:
      - unbounded retry where the old code retried indefinitely,
      - HMM_RANGE_DEFAULT_TIMEOUT where the old code used that budget,
      - existing driver-specific timeout return values such as -ETIME.
  - Used max_t(long, timeout - jiffies, 1) when passing remaining time from
    absolute jiffies deadlines to avoid unsigned underflow while keeping a
    minimum one-jiffy retry window.
  - Left callers on hmm_range_fault() when they already need to hold
    mmap_lock across surrounding work, such as drm_gpusvm_check_pages().

Changes in v6:
  - Reworked the new API from the external int *locked pattern to
    hmm_range_fault_unlocked(), which owns mmap_read_lock() internally.
  - Changed the dropped-lock contract: hmm_range_fault_unlocked() now returns
    -EBUSY when the mmap lock is dropped, and callers restart with a fresh
    mmu_interval_read_begin() sequence.
  - Kept hmm_range_fault() as the locked variant for existing users, preserving
    its caller-held mmap lock contract.
  - Added an in-tree user by converting the MSHV region fault path to
    hmm_range_fault_unlocked().
  - Updated Documentation/mm/hmm.rst and kernel-doc to describe the unlocked
    helper and retry pattern.
  - Updated commit messages to match the new API and return semantics.
  - Kept the userfaultfd HMM selftest using the test_hmm unlocked read ioctl
    path.

Changes in v5:
 - Rework hmm_range_fault_unlockable() retry handling to retry
   VM_FAULT_RETRY internally with FAULT_FLAG_TRIED set, matching the
   fixup_user_fault() pattern and avoiding repeated first-retry lock drops.
 - Distinguish VM_FAULT_RETRY from VM_FAULT_COMPLETED: retry faults now
   reacquire the mmap lock internally, while completed faults return to the
   caller with *locked = 0 so the caller can restart with a fresh notifier
   sequence.
 - Document the two *locked return states, including the -EINTR case when a
   fatal signal is pending after the mmap lock has already been dropped.
 - Update comments around HMM_FAULT_UNLOCKED and the HMM fault loop to match
   the current hmm_range_fault_unlockable() implementation.

Changes in v4:
 - Rebased on 7.2-rc1

Changes in v3:
 - Return -EFAULT from dmirror_fault_unlockable() when the mirrored mm can
   no longer be pinned.
 - Add an eventfd stop signal for the userfaultfd handler thread to avoid
   waiting for the poll timeout on successful test completion.


Changes in v2:

 - Split into a preparatory refactor (new patch 1) that moves
   handle_mm_fault() out of the walk callbacks, plus a smaller feature
   patch on top.  Suggested by David Hildenbrand.
 - Hugetlb regions are now supported on the unlockable path; the v1
   -EFAULT short-circuit and the hugetlb_vma_lock_read drop/retake
   dance are gone.
 - Distinct internal sentinels for "needs fault" (HMM_FAULT_PENDING)
   and "lock dropped" (HMM_FAULT_UNLOCKED).
 - Outer loop now re-walks after a successful internal fault so the
   faulted pfns end up in range->hmm_pfns.
 - Kernel-doc on hmm_range_fault_unlockable() and the
   Documentation/mm/hmm.rst example match the implementation.
 - Dropped the mshv driver conversion (v1 patch 2); will post
   separately.
 - Selftest converted to drive the path through test_hmm with a
   userfaultfd handler (new HMM_DMIRROR_READ_UNLOCKABLE ioctl).

---

Stanislav Kinsburskii (8):
      mm/hmm: move page fault handling out of walk callbacks
      mm/hmm: add hmm_range_fault_unlocked_timeout() for mmap lock-drop support
      selftests/mm: add HMM tests for mmap lock-dropping faults
      mshv: Use hmm_range_fault_unlocked_timeout() for region faults
      drm/nouveau: Use hmm_range_fault_unlocked_timeout() for SVM faults
      RDMA/umem: Use hmm_range_fault_unlocked_timeout() for ODP faults
      accel/amdxdna: Use hmm_range_fault_unlocked_timeout() for range population
      drm/gpusvm: Use hmm_range_fault_unlocked_timeout() for range faults


 Documentation/mm/hmm.rst               |   63 ++++++++
 drivers/accel/amdxdna/aie2_ctx.c       |   17 --
 drivers/gpu/drm/drm_gpusvm.c           |   52 +------
 drivers/gpu/drm/nouveau/nouveau_svm.c  |   11 -
 drivers/hv/mshv_regions.c              |   54 +------
 drivers/infiniband/core/umem_odp.c     |   18 +-
 include/linux/hmm.h                    |    2 
 lib/test_hmm.c                         |  111 ++++++++++++++
 lib/test_hmm_uapi.h                    |    3 
 mm/hmm.c                               |  246 ++++++++++++++++++++++++--------
 tools/testing/selftests/mm/hmm-tests.c |  220 +++++++++++++++++++++++++++++
 11 files changed, 613 insertions(+), 184 deletions(-)

--
Signature

[PATCH v7 1/8] mm/hmm: move page fault handling out of walk callbacks

[Stanislav Kinsburskii]

hmm_range_fault() currently triggers page faults from inside the page-table
walk callbacks: hmm_vma_walk_pmd(), hmm_vma_walk_pud(),
hmm_vma_walk_hugetlb_entry() and the pte-level helper all call
hmm_vma_fault(), which in turn calls handle_mm_fault() while the walker
still holds nested locks.  The pte spinlock is dropped explicitly by each
caller, and the hugetlb path manually drops and retakes
hugetlb_vma_lock_read around the fault to dodge a deadlock against the walk
framework's unconditional unlock.

This layering does not extend cleanly to fault handlers that may release
mmap_lock (VM_FAULT_RETRY, VM_FAULT_COMPLETED). If the lock is dropped
while walk_page_range() is mid-traversal, the VMA can be freed before the
walk framework's matching hugetlb_vma_unlock_read(), turning that unlock
into a use-after-free.

Split the responsibilities the way get_user_pages() does. Walk callbacks
become inspect-only: when they detect a range that needs to be faulted in,
they record it in struct hmm_vma_walk and return a private sentinel
(HMM_FAULT_PENDING). The outer loop in hmm_range_fault() then drops out of
walk_page_range(), invokes a new helper hmm_do_fault() that calls
handle_mm_fault() with only mmap_lock held, and restarts the walk so the
now-present entries are collected into hmm_pfns.

No functional change for existing callers. As a side effect the hugetlb
callback no longer needs the hugetlb_vma_{un}lock_read dance, and every
fault-path exit from the callbacks now releases the pte spinlock on a
single, common path. This refactor is also a precursor for adding an
unlockable variant of hmm_range_fault() in a follow-up patch.

Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
---
 mm/hmm.c |  118 +++++++++++++++++++++++++++++++++++++++-----------------------
 1 file changed, 75 insertions(+), 43 deletions(-)

diff --git a/mm/hmm.c b/mm/hmm.c
index 4f3f627d2b47..2129b1ee4c35 100644
--- a/mm/hmm.c
+++ b/mm/hmm.c
@@ -33,8 +33,17 @@
 struct hmm_vma_walk {
 	struct hmm_range	*range;
 	unsigned long		last;
+	unsigned long		end;
+	unsigned int		required_fault;
 };
 
+/*
+ * Internal sentinel returned by walk callbacks when they need a page fault.
+ * The callback stores end/required_fault in hmm_vma_walk; the outer loop
+ * consumes the sentinel and never propagates it to the caller.
+ */
+#define HMM_FAULT_PENDING	-EAGAIN
+
 enum {
 	HMM_NEED_FAULT = 1 << 0,
 	HMM_NEED_WRITE_FAULT = 1 << 1,
@@ -60,37 +69,25 @@ static int hmm_pfns_fill(unsigned long addr, unsigned long end,
 }
 
 /*
- * hmm_vma_fault() - fault in a range lacking valid pmd or pte(s)
- * @addr: range virtual start address (inclusive)
- * @end: range virtual end address (exclusive)
- * @required_fault: HMM_NEED_* flags
- * @walk: mm_walk structure
- * Return: -EBUSY after page fault, or page fault error
+ * hmm_record_fault() - record a range that needs to be faulted in
  *
- * This function will be called whenever pmd_none() or pte_none() returns true,
- * or whenever there is no page directory covering the virtual address range.
+ * Called by the walk callbacks when they discover that part of the range
+ * needs a page fault.  The callback records what to fault and returns
+ * HMM_FAULT_PENDING; the outer loop in hmm_range_fault() drops back out of
+ * walk_page_range() and invokes handle_mm_fault() from a context where no
+ * page-table or hugetlb_vma_lock is held.
  */
-static int hmm_vma_fault(unsigned long addr, unsigned long end,
-			 unsigned int required_fault, struct mm_walk *walk)
+static int hmm_record_fault(unsigned long addr, unsigned long end,
+			    unsigned int required_fault,
+			    struct mm_walk *walk)
 {
 	struct hmm_vma_walk *hmm_vma_walk = walk->private;
-	struct vm_area_struct *vma = walk->vma;
-	unsigned int fault_flags = FAULT_FLAG_REMOTE;
 
 	WARN_ON_ONCE(!required_fault);
 	hmm_vma_walk->last = addr;
-
-	if (required_fault & HMM_NEED_WRITE_FAULT) {
-		if (!(vma->vm_flags & VM_WRITE))
-			return -EPERM;
-		fault_flags |= FAULT_FLAG_WRITE;
-	}
-
-	for (; addr < end; addr += PAGE_SIZE)
-		if (handle_mm_fault(vma, addr, fault_flags, NULL) &
-		    VM_FAULT_ERROR)
-			return -EFAULT;
-	return -EBUSY;
+	hmm_vma_walk->end = end;
+	hmm_vma_walk->required_fault = required_fault;
+	return HMM_FAULT_PENDING;
 }
 
 static unsigned int hmm_pte_need_fault(const struct hmm_vma_walk *hmm_vma_walk,
@@ -174,7 +171,7 @@ static int hmm_vma_walk_hole(unsigned long addr, unsigned long end,
 		return hmm_pfns_fill(addr, end, range, HMM_PFN_ERROR);
 	}
 	if (required_fault)
-		return hmm_vma_fault(addr, end, required_fault, walk);
+		return hmm_record_fault(addr, end, required_fault, walk);
 	return hmm_pfns_fill(addr, end, range, 0);
 }
 
@@ -209,7 +206,7 @@ static int hmm_vma_handle_pmd(struct mm_walk *walk, unsigned long addr,
 	required_fault =
 		hmm_range_need_fault(hmm_vma_walk, hmm_pfns, npages, cpu_flags);
 	if (required_fault)
-		return hmm_vma_fault(addr, end, required_fault, walk);
+		return hmm_record_fault(addr, end, required_fault, walk);
 
 	pfn = pmd_pfn(pmd) + ((addr & ~PMD_MASK) >> PAGE_SHIFT);
 	for (i = 0; addr < end; addr += PAGE_SIZE, i++, pfn++) {
@@ -328,7 +325,7 @@ static int hmm_vma_handle_pte(struct mm_walk *walk, unsigned long addr,
 fault:
 	pte_unmap(ptep);
 	/* Fault any virtual address we were asked to fault */
-	return hmm_vma_fault(addr, end, required_fault, walk);
+	return hmm_record_fault(addr, end, required_fault, walk);
 }
 
 #ifdef CONFIG_ARCH_SUPPORTS_PMD_SOFTLEAF
@@ -371,7 +368,7 @@ static int hmm_vma_handle_absent_pmd(struct mm_walk *walk, unsigned long start,
 					      npages, 0);
 	if (required_fault) {
 		if (softleaf_is_device_private(entry))
-			return hmm_vma_fault(addr, end, required_fault, walk);
+			return hmm_record_fault(addr, end, required_fault, walk);
 		else
 			return -EFAULT;
 	}
@@ -517,7 +514,7 @@ static int hmm_vma_walk_pud(pud_t *pudp, unsigned long start, unsigned long end,
 						      npages, cpu_flags);
 		if (required_fault) {
 			spin_unlock(ptl);
-			return hmm_vma_fault(addr, end, required_fault, walk);
+			return hmm_record_fault(addr, end, required_fault, walk);
 		}
 
 		pfn = pud_pfn(pud) + ((addr & ~PUD_MASK) >> PAGE_SHIFT);
@@ -564,21 +561,8 @@ static int hmm_vma_walk_hugetlb_entry(pte_t *pte, unsigned long hmask,
 	required_fault =
 		hmm_pte_need_fault(hmm_vma_walk, pfn_req_flags, cpu_flags);
 	if (required_fault) {
-		int ret;
-
 		spin_unlock(ptl);
-		hugetlb_vma_unlock_read(vma);
-		/*
-		 * Avoid deadlock: drop the vma lock before calling
-		 * hmm_vma_fault(), which will itself potentially take and
-		 * drop the vma lock. This is also correct from a
-		 * protection point of view, because there is no further
-		 * use here of either pte or ptl after dropping the vma
-		 * lock.
-		 */
-		ret = hmm_vma_fault(addr, end, required_fault, walk);
-		hugetlb_vma_lock_read(vma);
-		return ret;
+		return hmm_record_fault(addr, end, required_fault, walk);
 	}
 
 	pfn = pte_pfn(entry) + ((start & ~hmask) >> PAGE_SHIFT);
@@ -637,6 +621,44 @@ static const struct mm_walk_ops hmm_walk_ops = {
 	.walk_lock	= PGWALK_RDLOCK,
 };
 
+/*
+ * hmm_do_fault - fault in a range recorded by a walk callback
+ *
+ * Called from the outer loop in hmm_range_fault() after a callback
+ * returned HMM_FAULT_PENDING.  At this point we hold only mmap_lock;
+ * the page-table spinlock and any hugetlb_vma_lock acquired by the walk
+ * framework have already been released by the unwind.
+ *
+ * Returns -EBUSY on success (all pages faulted, caller should re-walk).
+ * Returns a negative errno on failure.
+ */
+static int hmm_do_fault(struct mm_struct *mm,
+			struct hmm_vma_walk *hmm_vma_walk)
+{
+	unsigned long addr = hmm_vma_walk->last;
+	unsigned long end = hmm_vma_walk->end;
+	unsigned int required_fault = hmm_vma_walk->required_fault;
+	unsigned int fault_flags = FAULT_FLAG_REMOTE;
+	struct vm_area_struct *vma;
+
+	vma = vma_lookup(mm, addr);
+	if (!vma)
+		return -EFAULT;
+
+	if (required_fault & HMM_NEED_WRITE_FAULT) {
+		if (!(vma->vm_flags & VM_WRITE))
+			return -EPERM;
+		fault_flags |= FAULT_FLAG_WRITE;
+	}
+
+	for (; addr < end; addr += PAGE_SIZE)
+		if (handle_mm_fault(vma, addr, fault_flags, NULL) &
+		    VM_FAULT_ERROR)
+			return -EFAULT;
+
+	return -EBUSY;
+}
+
 /**
  * hmm_range_fault - try to fault some address in a virtual address range
  * @range:	argument structure
@@ -674,6 +696,16 @@ int hmm_range_fault(struct hmm_range *range)
 			return -EBUSY;
 		ret = walk_page_range(mm, hmm_vma_walk.last, range->end,
 				      &hmm_walk_ops, &hmm_vma_walk);
+		/*
+		 * When HMM_FAULT_PENDING is returned a walk callback
+		 * recorded a range that needs handle_mm_fault();
+		 * hmm_do_fault() runs the fault outside walk_page_range()
+		 * (so no page-table or hugetlb_vma_lock is held) and
+		 * returns -EBUSY so the loop re-walks and picks up the
+		 * now-present entries.
+		 */
+		if (ret == HMM_FAULT_PENDING)
+			ret = hmm_do_fault(mm, &hmm_vma_walk);
 		/*
 		 * When -EBUSY is returned the loop restarts with
 		 * hmm_vma_walk.last set to an address that has not been stored

[PATCH v7 2/8] mm/hmm: add hmm_range_fault_unlocked_timeout() for mmap lock-drop support

[Stanislav Kinsburskii]

hmm_range_fault() requires the caller to hold the mmap read lock for the
duration of the call. This is incompatible with mappings whose fault
handler may release the mmap lock, notably userfaultfd-managed regions,
where handle_mm_fault() can return VM_FAULT_RETRY or VM_FAULT_COMPLETED
after dropping the lock. Drivers that need to populate device page tables
for such mappings have no way to do so today.

Add hmm_range_fault_unlocked_timeout() for callers that do not need to hold
mmap_lock across any work outside the HMM fault itself. The helper takes
mmap_read_lock_killable() internally, calls the common HMM fault
implementation, and releases the lock before returning if it is still held.
The timeout is specified in jiffies; passing 0 retries indefinitely, while
a non-zero timeout makes the helper return -EBUSY when the retry budget
expires.

When handle_mm_fault() drops mmap_lock, or when the range is invalidated,
hmm_range_fault_unlocked_timeout() refreshes range->notifier_seq and
retries the walk internally. The caller only needs to perform the usual
post-success mmu_interval_read_retry() check while holding its update lock
before consuming the pfns. If mmap_lock acquisition is interrupted or a
fatal signal is pending during retry handling, -EINTR is returned instead.

The common implementation conditionally sets FAULT_FLAG_ALLOW_RETRY and
FAULT_FLAG_KILLABLE only for hmm_range_fault_unlocked_timeout(). The
existing hmm_range_fault() path still passes no locked state, does not
allow handle_mm_fault() to drop mmap_lock, and remains a thin wrapper
preserving the existing API contract for current callers.

The previous refactor that moved page fault handling out of the page-table
walk callbacks is what makes this change small. Faults now run after
walk_page_range() has unwound, with only mmap_lock held, so dropping it
does not interact with the walker's pte spinlock or hugetlb_vma_lock.
Hugetlb regions therefore participate in the unlocked path uniformly with
PTE- and PMD-level mappings; no special case is required.

Documentation/mm/hmm.rst is updated with a description of the new API and
the recommended caller pattern.

Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
---
 Documentation/mm/hmm.rst |   63 +++++++++++++++++++
 include/linux/hmm.h      |    2 +
 mm/hmm.c                 |  152 +++++++++++++++++++++++++++++++++++++---------
 3 files changed, 188 insertions(+), 29 deletions(-)

diff --git a/Documentation/mm/hmm.rst b/Documentation/mm/hmm.rst
index 7d61b7a8b65b..70885f153d03 100644
--- a/Documentation/mm/hmm.rst
+++ b/Documentation/mm/hmm.rst
@@ -208,6 +208,69 @@ invalidate() callback. That lock must be held before calling
 mmu_interval_read_retry() to avoid any race with a concurrent CPU page table
 update.
 
+Dropping the mmap lock during page faults
+=========================================
+
+Some VMAs have fault handlers that need to release the mmap lock while
+servicing a fault (for example, regions managed by ``userfaultfd``).
+``hmm_range_fault()`` cannot be used on such mappings because it must hold the
+mmap lock for the duration of the call. Drivers that need to support them
+should call::
+
+  int hmm_range_fault_unlocked_timeout(struct hmm_range *range,
+                                       unsigned long timeout);
+
+The caller must not hold ``mmap_read_lock`` before the call.
+``hmm_range_fault_unlocked_timeout()`` takes the mmap read lock internally and
+allows ``handle_mm_fault()`` to drop it during fault handling. If the mmap lock
+is dropped or the range is invalidated, the function refreshes
+``range->notifier_seq`` and restarts the walk internally. ``-EINTR`` is
+returned if mmap lock acquisition is interrupted or a fatal signal is pending
+during retry handling.
+
+The timeout is specified in jiffies; passing ``0`` means retry indefinitely. If
+the timeout expires while the function is retrying after ``-EBUSY``,
+``-EBUSY`` is returned to the caller.
+
+A typical caller looks like this::
+
+ int driver_populate_range_unlocked(...)
+ {
+      struct hmm_range range;
+      unsigned long timeout;
+      ...
+
+      timeout = msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT);
+      range.notifier = &interval_sub;
+      range.start = ...;
+      range.end = ...;
+      range.hmm_pfns = ...;
+
+      if (!mmget_not_zero(interval_sub.mm))
+          return -EFAULT;
+
+ again:
+      ret = hmm_range_fault_unlocked_timeout(&range, timeout);
+      if (ret)
+          goto out_put;
+
+      take_lock(driver->update);
+      if (mmu_interval_read_retry(&interval_sub, range.notifier_seq)) {
+          release_lock(driver->update);
+          goto again;
+      }
+
+      /* Use pfns array content to update device page table,
+       * under the update lock */
+
+      release_lock(driver->update);
+      ret = 0;
+
+ out_put:
+      mmput(interval_sub.mm);
+      return ret;
+ }
+
 Leverage default_flags and pfn_flags_mask
 =========================================
 
diff --git a/include/linux/hmm.h b/include/linux/hmm.h
index db75ffc949a7..6f04e3932f5b 100644
--- a/include/linux/hmm.h
+++ b/include/linux/hmm.h
@@ -123,6 +123,8 @@ struct hmm_range {
  * Please see Documentation/mm/hmm.rst for how to use the range API.
  */
 int hmm_range_fault(struct hmm_range *range);
+int hmm_range_fault_unlocked_timeout(struct hmm_range *range,
+				     unsigned long timeout);
 
 /*
  * HMM_RANGE_DEFAULT_TIMEOUT - default timeout (ms) when waiting for a range
diff --git a/mm/hmm.c b/mm/hmm.c
index 2129b1ee4c35..15110409be00 100644
--- a/mm/hmm.c
+++ b/mm/hmm.c
@@ -32,6 +32,7 @@
 
 struct hmm_vma_walk {
 	struct hmm_range	*range;
+	int			*locked;
 	unsigned long		last;
 	unsigned long		end;
 	unsigned int		required_fault;
@@ -44,6 +45,14 @@ struct hmm_vma_walk {
  */
 #define HMM_FAULT_PENDING	-EAGAIN
 
+/*
+ * Internal sentinel returned by hmm_do_fault() when handle_mm_fault()
+ * completes a page fault with the mmap lock dropped. hmm_do_fault() sets
+ * *locked = 0; the outer loop consumes the sentinel and never propagates it
+ * to the caller.
+ */
+#define HMM_FAULT_UNLOCKED	-ENOLCK
+
 enum {
 	HMM_NEED_FAULT = 1 << 0,
 	HMM_NEED_WRITE_FAULT = 1 << 1,
@@ -73,9 +82,9 @@ static int hmm_pfns_fill(unsigned long addr, unsigned long end,
  *
  * Called by the walk callbacks when they discover that part of the range
  * needs a page fault.  The callback records what to fault and returns
- * HMM_FAULT_PENDING; the outer loop in hmm_range_fault() drops back out of
- * walk_page_range() and invokes handle_mm_fault() from a context where no
- * page-table or hugetlb_vma_lock is held.
+ * HMM_FAULT_PENDING; the outer loop in hmm_range_fault_locked() drops
+ * back out of walk_page_range() and invokes handle_mm_fault() from a context
+ * where no page-table or hugetlb_vma_lock is held.
  */
 static int hmm_record_fault(unsigned long addr, unsigned long end,
 			    unsigned int required_fault,
@@ -624,7 +633,7 @@ static const struct mm_walk_ops hmm_walk_ops = {
 /*
  * hmm_do_fault - fault in a range recorded by a walk callback
  *
- * Called from the outer loop in hmm_range_fault() after a callback
+ * Called from the outer loop in hmm_range_fault_locked() after a callback
  * returned HMM_FAULT_PENDING.  At this point we hold only mmap_lock;
  * the page-table spinlock and any hugetlb_vma_lock acquired by the walk
  * framework have already been released by the unwind.
@@ -641,6 +650,9 @@ static int hmm_do_fault(struct mm_struct *mm,
 	unsigned int fault_flags = FAULT_FLAG_REMOTE;
 	struct vm_area_struct *vma;
 
+	if (hmm_vma_walk->locked)
+		fault_flags |= FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE;
+
 	vma = vma_lookup(mm, addr);
 	if (!vma)
 		return -EFAULT;
@@ -651,37 +663,33 @@ static int hmm_do_fault(struct mm_struct *mm,
 		fault_flags |= FAULT_FLAG_WRITE;
 	}
 
-	for (; addr < end; addr += PAGE_SIZE)
-		if (handle_mm_fault(vma, addr, fault_flags, NULL) &
-		    VM_FAULT_ERROR)
-			return -EFAULT;
+	for (; addr < end; addr += PAGE_SIZE) {
+		vm_fault_t ret;
+
+		ret = handle_mm_fault(vma, addr, fault_flags, NULL);
+
+		if (ret & (VM_FAULT_COMPLETED | VM_FAULT_RETRY)) {
+			*hmm_vma_walk->locked = 0;
+			return HMM_FAULT_UNLOCKED;
+		}
+
+		if (ret & VM_FAULT_ERROR) {
+			int err = vm_fault_to_errno(ret, 0);
+
+			if (err)
+				return err;
+			BUG();
+		}
+	}
 
 	return -EBUSY;
 }
 
-/**
- * hmm_range_fault - try to fault some address in a virtual address range
- * @range:	argument structure
- *
- * Returns 0 on success or one of the following error codes:
- *
- * -EINVAL:	Invalid arguments or mm or virtual address is in an invalid vma
- *		(e.g., device file vma).
- * -ENOMEM:	Out of memory.
- * -EPERM:	Invalid permission (e.g., asking for write and range is read
- *		only).
- * -EBUSY:	The range has been invalidated and the caller needs to wait for
- *		the invalidation to finish.
- * -EFAULT:     A page was requested to be valid and could not be made valid
- *              ie it has no backing VMA or it is illegal to access
- *
- * This is similar to get_user_pages(), except that it can read the page tables
- * without mutating them (ie causing faults).
- */
-int hmm_range_fault(struct hmm_range *range)
+static int hmm_range_fault_locked(struct hmm_range *range, int *locked)
 {
 	struct hmm_vma_walk hmm_vma_walk = {
 		.range = range,
+		.locked = locked,
 		.last = range->start,
 	};
 	struct mm_struct *mm = range->notifier->mm;
@@ -704,8 +712,14 @@ int hmm_range_fault(struct hmm_range *range)
 		 * returns -EBUSY so the loop re-walks and picks up the
 		 * now-present entries.
 		 */
-		if (ret == HMM_FAULT_PENDING)
+		if (ret == HMM_FAULT_PENDING) {
 			ret = hmm_do_fault(mm, &hmm_vma_walk);
+			if (ret == HMM_FAULT_UNLOCKED) {
+				if (fatal_signal_pending(current))
+					return -EINTR;
+				return -EBUSY;
+			}
+		}
 		/*
 		 * When -EBUSY is returned the loop restarts with
 		 * hmm_vma_walk.last set to an address that has not been stored
@@ -715,8 +729,88 @@ int hmm_range_fault(struct hmm_range *range)
 	} while (ret == -EBUSY);
 	return ret;
 }
+
+/**
+ * hmm_range_fault - try to fault some address in a virtual address range
+ * @range:	argument structure
+ *
+ * Returns 0 on success or one of the following error codes:
+ *
+ * -EINVAL:	Invalid arguments or mm or virtual address is in an invalid vma
+ *		(e.g., device file vma).
+ * -ENOMEM:	Out of memory.
+ * -EPERM:	Invalid permission (e.g., asking for write and range is read
+ *		only).
+ * -EBUSY:	The range has been invalidated and the caller needs to wait for
+ *		the invalidation to finish.
+ * -EFAULT:     A page was requested to be valid and could not be made valid
+ *              ie it has no backing VMA or it is illegal to access
+ *
+ * This is similar to get_user_pages(), except that it can read the page tables
+ * without mutating them (ie causing faults).
+ *
+ * The mmap lock must be held by the caller and will remain held on return.
+ * For a variant that allows the mmap lock to be dropped during faults (e.g.,
+ * for userfaultfd support), see hmm_range_fault_unlocked_timeout().
+ */
+int hmm_range_fault(struct hmm_range *range)
+{
+	return hmm_range_fault_locked(range, NULL);
+}
 EXPORT_SYMBOL(hmm_range_fault);
 
+/**
+ * hmm_range_fault_unlocked_timeout - fault in a range with a retry timeout
+ * @range:	argument structure
+ * @timeout:	timeout in jiffies for internal -EBUSY retries, or 0 to retry
+ *		indefinitely
+ *
+ * This is similar to hmm_range_fault(), except the caller must not hold the
+ * mmap lock. The function takes the mmap read lock internally and allows
+ * handle_mm_fault() to drop it during faults. If the mmap lock is dropped or
+ * the range is invalidated, the function refreshes range->notifier_seq and
+ * restarts the walk internally. Passing 0 for @timeout retries indefinitely;
+ * otherwise, if @timeout expires while retrying -EBUSY, -EBUSY is returned to
+ * the caller.
+ *
+ * Returns 0 on success or one of the error codes documented for
+ * hmm_range_fault(). -EINTR is returned if mmap_lock acquisition is
+ * interrupted or a fatal signal is pending during retry handling.
+ */
+int hmm_range_fault_unlocked_timeout(struct hmm_range *range,
+				     unsigned long timeout)
+{
+	struct mm_struct *mm = range->notifier->mm;
+	unsigned long deadline = 0;
+	int locked, ret;
+
+	if (timeout)
+		deadline = jiffies + timeout;
+
+	do {
+		if (fatal_signal_pending(current))
+			return -EINTR;
+
+		if (timeout && time_after(jiffies, deadline))
+			return -EBUSY;
+
+		range->notifier_seq =
+			mmu_interval_read_begin(range->notifier);
+
+		ret = mmap_read_lock_killable(mm);
+		if (ret)
+			return ret;
+
+		locked = 1;
+		ret = hmm_range_fault_locked(range, &locked);
+		if (locked)
+			mmap_read_unlock(mm);
+	} while (ret == -EBUSY);
+
+	return ret;
+}
+EXPORT_SYMBOL(hmm_range_fault_unlocked_timeout);
+
 /**
  * hmm_dma_map_alloc - Allocate HMM map structure
  * @dev: device to allocate structure for

[PATCH v7 3/8] selftests/mm: add HMM tests for mmap lock-dropping faults

[Stanislav Kinsburskii]

Add test_hmm coverage for the HMM lock-dropping fault path. The test module
gets new HMM_DMIRROR_READ_UNLOCKED and HMM_DMIRROR_READ_UNLOCKED_TIMEOUT
ioctls that both call hmm_range_fault_unlocked_timeout(). The unlocked
ioctl passes a timeout of 0 to exercise the unbounded retry mode, while the
timeout ioctl passes HMM_RANGE_DEFAULT_TIMEOUT converted to jiffies.

Add a userfaultfd_read selftest that registers an anonymous mapping with
UFFDIO_REGISTER_MODE_MISSING, services the faults from a handler thread
with UFFDIO_COPY, and verifies that HMM can read back the data supplied by
the handler. This exercises the path where handle_mm_fault() drops
mmap_lock and hmm_range_fault_unlocked_timeout() restarts the walk
internally.

Add a userfaultfd_timeout selftest for the finite-timeout mode. It
registers a missing-mode userfaultfd range without a data-supplying
handler, then closes the userfaultfd after the timeout window so the
blocked fault can unwind and hmm_range_fault_unlocked_timeout() can return
-EBUSY.

Assisted-by: GitHub-Copilot:claude-opus-4.6
Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
---
 lib/test_hmm.c                         |  111 ++++++++++++++++
 lib/test_hmm_uapi.h                    |    3 
 tools/testing/selftests/mm/hmm-tests.c |  220 ++++++++++++++++++++++++++++++++
 3 files changed, 334 insertions(+)

diff --git a/lib/test_hmm.c b/lib/test_hmm.c
index 45c0cb992218..02a98887c290 100644
--- a/lib/test_hmm.c
+++ b/lib/test_hmm.c
@@ -389,6 +389,67 @@ static int dmirror_range_fault(struct dmirror *dmirror,
 	return ret;
 }
 
+static int dmirror_range_fault_unlocked(struct dmirror *dmirror,
+					struct hmm_range *range,
+					unsigned long timeout)
+{
+	int ret;
+
+	while (true) {
+		ret = hmm_range_fault_unlocked_timeout(range, timeout);
+		if (ret)
+			goto out;
+
+		mutex_lock(&dmirror->mutex);
+		if (mmu_interval_read_retry(range->notifier,
+					    range->notifier_seq)) {
+			mutex_unlock(&dmirror->mutex);
+			continue;
+		}
+		break;
+	}
+
+	ret = dmirror_do_fault(dmirror, range);
+
+	mutex_unlock(&dmirror->mutex);
+out:
+	return ret;
+}
+
+static int dmirror_fault_unlocked(struct dmirror *dmirror,
+				  unsigned long start,
+				  unsigned long end, bool write,
+				  unsigned long timeout)
+{
+	struct mm_struct *mm = dmirror->notifier.mm;
+	unsigned long addr;
+	unsigned long pfns[32];
+	struct hmm_range range = {
+		.notifier = &dmirror->notifier,
+		.hmm_pfns = pfns,
+		.pfn_flags_mask = 0,
+		.default_flags =
+			HMM_PFN_REQ_FAULT | (write ? HMM_PFN_REQ_WRITE : 0),
+		.dev_private_owner = dmirror->mdevice,
+	};
+	int ret = 0;
+
+	if (!mmget_not_zero(mm))
+		return -EFAULT;
+
+	for (addr = start; addr < end; addr = range.end) {
+		range.start = addr;
+		range.end = min(addr + (ARRAY_SIZE(pfns) << PAGE_SHIFT), end);
+
+		ret = dmirror_range_fault_unlocked(dmirror, &range, timeout);
+		if (ret)
+			break;
+	}
+
+	mmput(mm);
+	return ret;
+}
+
 static int dmirror_fault(struct dmirror *dmirror, unsigned long start,
 			 unsigned long end, bool write)
 {
@@ -488,6 +549,48 @@ static int dmirror_read(struct dmirror *dmirror, struct hmm_dmirror_cmd *cmd)
 	return ret;
 }
 
+static int dmirror_read_unlocked(struct dmirror *dmirror,
+				 struct hmm_dmirror_cmd *cmd,
+				 unsigned long timeout)
+{
+	struct dmirror_bounce bounce;
+	unsigned long start, end;
+	unsigned long size = cmd->npages << PAGE_SHIFT;
+	int ret;
+
+	start = cmd->addr;
+	end = start + size;
+	if (end < start)
+		return -EINVAL;
+
+	ret = dmirror_bounce_init(&bounce, start, size);
+	if (ret)
+		return ret;
+
+	while (1) {
+		mutex_lock(&dmirror->mutex);
+		ret = dmirror_do_read(dmirror, start, end, &bounce);
+		mutex_unlock(&dmirror->mutex);
+		if (ret != -ENOENT)
+			break;
+
+		start = cmd->addr + (bounce.cpages << PAGE_SHIFT);
+		ret = dmirror_fault_unlocked(dmirror, start, end, false, timeout);
+		if (ret)
+			break;
+		cmd->faults++;
+	}
+
+	if (ret == 0) {
+		if (copy_to_user(u64_to_user_ptr(cmd->ptr), bounce.ptr,
+				 bounce.size))
+			ret = -EFAULT;
+	}
+	cmd->cpages = bounce.cpages;
+	dmirror_bounce_fini(&bounce);
+	return ret;
+}
+
 static int dmirror_do_write(struct dmirror *dmirror, unsigned long start,
 			    unsigned long end, struct dmirror_bounce *bounce)
 {
@@ -1572,6 +1675,14 @@ static long dmirror_fops_unlocked_ioctl(struct file *filp,
 		dmirror->flags = cmd.npages;
 		ret = 0;
 		break;
+	case HMM_DMIRROR_READ_UNLOCKED:
+		ret = dmirror_read_unlocked(dmirror, &cmd, 0);
+		break;
+	case HMM_DMIRROR_READ_UNLOCKED_TIMEOUT:
+		ret = dmirror_read_unlocked(dmirror, &cmd,
+				msecs_to_jiffies(
+					HMM_RANGE_DEFAULT_TIMEOUT));
+		break;
 
 	default:
 		return -EINVAL;
diff --git a/lib/test_hmm_uapi.h b/lib/test_hmm_uapi.h
index f94c6d457338..249ab2d9e3b2 100644
--- a/lib/test_hmm_uapi.h
+++ b/lib/test_hmm_uapi.h
@@ -38,6 +38,9 @@ struct hmm_dmirror_cmd {
 #define HMM_DMIRROR_CHECK_EXCLUSIVE	_IOWR('H', 0x06, struct hmm_dmirror_cmd)
 #define HMM_DMIRROR_RELEASE		_IOWR('H', 0x07, struct hmm_dmirror_cmd)
 #define HMM_DMIRROR_FLAGS		_IOWR('H', 0x08, struct hmm_dmirror_cmd)
+#define HMM_DMIRROR_READ_UNLOCKED	_IOWR('H', 0x09, struct hmm_dmirror_cmd)
+#define HMM_DMIRROR_READ_UNLOCKED_TIMEOUT \
+	_IOWR('H', 0x0a, struct hmm_dmirror_cmd)
 
 #define HMM_DMIRROR_FLAG_FAIL_ALLOC	(1ULL << 0)
 
diff --git a/tools/testing/selftests/mm/hmm-tests.c b/tools/testing/selftests/mm/hmm-tests.c
index 2f2b9879d100..dc031d2e2aae 100644
--- a/tools/testing/selftests/mm/hmm-tests.c
+++ b/tools/testing/selftests/mm/hmm-tests.c
@@ -29,6 +29,10 @@
 #include <sys/mman.h>
 #include <sys/ioctl.h>
 #include <sys/time.h>
+#include <sys/syscall.h>
+#include <sys/eventfd.h>
+#include <linux/userfaultfd.h>
+#include <poll.h>
 
 /*
  * This is a private UAPI to the kernel test module so it isn't exported
@@ -2949,4 +2953,220 @@ TEST_F_TIMEOUT(hmm, benchmark_thp_migration, 120)
 					&thp_results, &regular_results);
 	}
 }
+/*
+ * Test that HMM can fault in pages backed by userfaultfd using the
+ * hmm_range_fault_unlocked_timeout() path with no timeout. This exercises
+ * the lock-drop retry logic in the HMM framework.
+ */
+struct uffd_thread_args {
+	int uffd;
+	int stop_fd;
+	void *page_buffer;
+	unsigned long page_size;
+};
+
+static void *uffd_handler_thread(void *arg)
+{
+	struct uffd_thread_args *args = arg;
+	struct uffd_msg msg;
+	struct uffdio_copy copy;
+	struct pollfd pollfd[2];
+	int ret;
+
+	pollfd[0].fd = args->uffd;
+	pollfd[0].events = POLLIN;
+	pollfd[1].fd = args->stop_fd;
+	pollfd[1].events = POLLIN;
+
+	while (1) {
+		ret = poll(pollfd, 2, -1);
+		if (ret <= 0)
+			break;
+		if (pollfd[1].revents)
+			break;
+		if (!(pollfd[0].revents & POLLIN))
+			break;
+
+		ret = read(args->uffd, &msg, sizeof(msg));
+		if (ret != sizeof(msg))
+			break;
+
+		if (msg.event != UFFD_EVENT_PAGEFAULT)
+			break;
+
+		/* Fill the page with a known pattern */
+		memset(args->page_buffer, 0xAB, args->page_size);
+
+		copy.dst = msg.arg.pagefault.address & ~(args->page_size - 1);
+		copy.src = (unsigned long)args->page_buffer;
+		copy.len = args->page_size;
+		copy.mode = 0;
+		copy.copy = 0;
+
+		ret = ioctl(args->uffd, UFFDIO_COPY, &copy);
+		if (ret < 0)
+			break;
+	}
+
+	return NULL;
+}
+
+struct uffd_close_args {
+	int uffd;
+	unsigned int delay_us;
+};
+
+static void *uffd_close_thread(void *arg)
+{
+	struct uffd_close_args *args = arg;
+
+	usleep(args->delay_us);
+	close(args->uffd);
+	return NULL;
+}
+
+TEST_F(hmm, userfaultfd_read)
+{
+	struct hmm_buffer *buffer;
+	struct uffd_thread_args uffd_args;
+	unsigned long npages;
+	unsigned long size;
+	unsigned long i;
+	unsigned char *ptr;
+	pthread_t thread;
+	int uffd;
+	int stop_fd;
+	int ret;
+	struct uffdio_api api;
+	struct uffdio_register reg;
+	uint64_t stop = 1;
+	ssize_t nwrite;
+
+	npages = 4;
+	size = npages << self->page_shift;
+
+	/* Create userfaultfd */
+	uffd = syscall(__NR_userfaultfd, O_CLOEXEC | O_NONBLOCK);
+	if (uffd < 0)
+		SKIP(return, "userfaultfd not available");
+
+	api.api = UFFD_API;
+	api.features = 0;
+	ret = ioctl(uffd, UFFDIO_API, &api);
+	ASSERT_EQ(ret, 0);
+
+	buffer = malloc(sizeof(*buffer));
+	ASSERT_NE(buffer, NULL);
+
+	buffer->fd = -1;
+	buffer->size = size;
+	buffer->mirror = malloc(size);
+	ASSERT_NE(buffer->mirror, NULL);
+
+	/* Create anonymous mapping */
+	buffer->ptr = mmap(NULL, size,
+			   PROT_READ | PROT_WRITE,
+			   MAP_PRIVATE | MAP_ANONYMOUS,
+			   -1, 0);
+	ASSERT_NE(buffer->ptr, MAP_FAILED);
+
+	/* Register the region with userfaultfd */
+	reg.range.start = (unsigned long)buffer->ptr;
+	reg.range.len = size;
+	reg.mode = UFFDIO_REGISTER_MODE_MISSING;
+	ret = ioctl(uffd, UFFDIO_REGISTER, &reg);
+	ASSERT_EQ(ret, 0);
+
+	/* Set up the handler thread */
+	uffd_args.uffd = uffd;
+	stop_fd = eventfd(0, EFD_CLOEXEC);
+	ASSERT_GE(stop_fd, 0);
+	uffd_args.stop_fd = stop_fd;
+	uffd_args.page_buffer = malloc(self->page_size);
+	ASSERT_NE(uffd_args.page_buffer, NULL);
+	uffd_args.page_size = self->page_size;
+
+	ret = pthread_create(&thread, NULL, uffd_handler_thread, &uffd_args);
+	ASSERT_EQ(ret, 0);
+
+	/*
+	 * Use the unlocked read path which allows the mmap lock to be
+	 * dropped during the fault, enabling userfaultfd resolution.
+	 */
+	ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ_UNLOCKED,
+			      buffer, npages);
+	ASSERT_EQ(ret, 0);
+	ASSERT_EQ(buffer->cpages, npages);
+
+	/* Verify the device read the data filled by the uffd handler */
+	ptr = buffer->mirror;
+	for (i = 0; i < size; ++i)
+		ASSERT_EQ(ptr[i], (unsigned char)0xAB);
+
+	nwrite = write(stop_fd, &stop, sizeof(stop));
+	ASSERT_EQ(nwrite, sizeof(stop));
+	pthread_join(thread, NULL);
+	close(stop_fd);
+	free(uffd_args.page_buffer);
+	close(uffd);
+	hmm_buffer_free(buffer);
+}
+
+TEST_F(hmm, userfaultfd_timeout)
+{
+	struct uffd_close_args close_args;
+	struct hmm_buffer *buffer;
+	unsigned long npages;
+	unsigned long size;
+	pthread_t thread;
+	int uffd;
+	int ret;
+	struct uffdio_api api;
+	struct uffdio_register reg;
+
+	npages = 4;
+	size = npages << self->page_shift;
+
+	uffd = syscall(__NR_userfaultfd, O_CLOEXEC | O_NONBLOCK);
+	if (uffd < 0)
+		SKIP(return, "userfaultfd not available");
+
+	api.api = UFFD_API;
+	api.features = 0;
+	ret = ioctl(uffd, UFFDIO_API, &api);
+	ASSERT_EQ(ret, 0);
+
+	buffer = malloc(sizeof(*buffer));
+	ASSERT_NE(buffer, NULL);
+
+	buffer->fd = -1;
+	buffer->size = size;
+	buffer->mirror = malloc(size);
+	ASSERT_NE(buffer->mirror, NULL);
+
+	buffer->ptr = mmap(NULL, size,
+			   PROT_READ | PROT_WRITE,
+			   MAP_PRIVATE | MAP_ANONYMOUS,
+			   -1, 0);
+	ASSERT_NE(buffer->ptr, MAP_FAILED);
+
+	reg.range.start = (unsigned long)buffer->ptr;
+	reg.range.len = size;
+	reg.mode = UFFDIO_REGISTER_MODE_MISSING;
+	ret = ioctl(uffd, UFFDIO_REGISTER, &reg);
+	ASSERT_EQ(ret, 0);
+
+	close_args.uffd = uffd;
+	close_args.delay_us = 1500 * 1000;
+	ret = pthread_create(&thread, NULL, uffd_close_thread, &close_args);
+	ASSERT_EQ(ret, 0);
+
+	ret = hmm_dmirror_cmd(self->fd, HMM_DMIRROR_READ_UNLOCKED_TIMEOUT,
+			    buffer, npages);
+	pthread_join(thread, NULL);
+	ASSERT_EQ(ret, -EBUSY);
+
+	hmm_buffer_free(buffer);
+}
+
 TEST_HARNESS_MAIN

[PATCH v7 4/8] mshv: Use hmm_range_fault_unlocked_timeout() for region faults

[Stanislav Kinsburskii]

MSHV currently faults movable memory regions by taking mmap_read_lock()
around hmm_range_fault(). That prevents the fault path from handling VMAs
whose fault handlers need to drop mmap_lock, such as userfaultfd-backed
mappings.

Use hmm_range_fault_unlocked_timeout() instead. Passing a timeout of 0
preserves MSHV's existing unbounded retry behavior while letting the HMM
helper own mmap_lock acquisition and refresh range->notifier_seq internally
before walking the range. After the fault succeeds, MSHV still takes
mreg_mutex and checks mmu_interval_read_retry() before installing the pages
into the region, so the existing invalidation synchronization is preserved.

Fold the small fault-and-lock helper into mshv_region_range_fault(), since
the remaining retry path is just the standard "fault, take the driver lock,
check the interval notifier sequence" pattern.

Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
---
 drivers/hv/mshv_regions.c |   54 ++++++++-------------------------------------
 1 file changed, 10 insertions(+), 44 deletions(-)

diff --git a/drivers/hv/mshv_regions.c b/drivers/hv/mshv_regions.c
index 6d65e5b42152..dddaade31b5d 100644
--- a/drivers/hv/mshv_regions.c
+++ b/drivers/hv/mshv_regions.c
@@ -381,46 +381,6 @@ int mshv_region_get(struct mshv_mem_region *region)
 	return kref_get_unless_zero(&region->mreg_refcount);
 }
 
-/**
- * mshv_region_hmm_fault_and_lock - Handle HMM faults and lock the memory region
- * @region: Pointer to the memory region structure
- * @range: Pointer to the HMM range structure
- *
- * This function performs the following steps:
- * 1. Reads the notifier sequence for the HMM range.
- * 2. Acquires a read lock on the memory map.
- * 3. Handles HMM faults for the specified range.
- * 4. Releases the read lock on the memory map.
- * 5. If successful, locks the memory region mutex.
- * 6. Verifies if the notifier sequence has changed during the operation.
- *    If it has, releases the mutex and returns -EBUSY to match with
- *    hmm_range_fault() return code for repeating.
- *
- * Return: 0 on success, a negative error code otherwise.
- */
-static int mshv_region_hmm_fault_and_lock(struct mshv_mem_region *region,
-					  struct hmm_range *range)
-{
-	int ret;
-
-	range->notifier_seq = mmu_interval_read_begin(range->notifier);
-	mmap_read_lock(region->mreg_mni.mm);
-	ret = hmm_range_fault(range);
-	mmap_read_unlock(region->mreg_mni.mm);
-	if (ret)
-		return ret;
-
-	mutex_lock(&region->mreg_mutex);
-
-	if (mmu_interval_read_retry(range->notifier, range->notifier_seq)) {
-		mutex_unlock(&region->mreg_mutex);
-		cond_resched();
-		return -EBUSY;
-	}
-
-	return 0;
-}
-
 /**
  * mshv_region_range_fault - Handle memory range faults for a given region.
  * @region: Pointer to the memory region structure.
@@ -452,13 +412,19 @@ static int mshv_region_range_fault(struct mshv_mem_region *region,
 	range.start = region->start_uaddr + page_offset * HV_HYP_PAGE_SIZE;
 	range.end = range.start + page_count * HV_HYP_PAGE_SIZE;
 
-	do {
-		ret = mshv_region_hmm_fault_and_lock(region, &range);
-	} while (ret == -EBUSY);
-
+again:
+	ret = hmm_range_fault_unlocked_timeout(&range, 0);
 	if (ret)
 		goto out;
 
+	mutex_lock(&region->mreg_mutex);
+
+	if (mmu_interval_read_retry(range.notifier, range.notifier_seq)) {
+		mutex_unlock(&region->mreg_mutex);
+		cond_resched();
+		goto again;
+	}
+
 	for (i = 0; i < page_count; i++)
 		region->mreg_pages[page_offset + i] = hmm_pfn_to_page(pfns[i]);
 

[PATCH v7 5/8] drm/nouveau: Use hmm_range_fault_unlocked_timeout() for SVM faults

[Stanislav Kinsburskii]

nouveau_range_fault() takes mmap_read_lock() only to call
hmm_range_fault(). It also keeps a single HMM_RANGE_DEFAULT_TIMEOUT
deadline across both HMM -EBUSY retries and post-fault
mmu_interval_read_retry() retries.

Use hmm_range_fault_unlocked_timeout() instead. The HMM helper now owns
the mmap lock and refreshes range->notifier_seq for its internal retries.
Nouveau keeps its existing absolute deadline in the outer loop and passes
the remaining jiffies to the helper for each fault attempt, so retries
caused by mmu_interval_read_retry() do not reset the overall retry budget.

Nouveau still validates the interval notifier sequence while holding
svmm->mutex before programming the GPU mapping.

Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
---
 drivers/gpu/drm/nouveau/nouveau_svm.c |   11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c b/drivers/gpu/drm/nouveau/nouveau_svm.c
index dcc92131488e..ba93273341af 100644
--- a/drivers/gpu/drm/nouveau/nouveau_svm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_svm.c
@@ -683,15 +683,10 @@ static int nouveau_range_fault(struct nouveau_svmm *svmm,
 			goto out;
 		}
 
-		range.notifier_seq = mmu_interval_read_begin(range.notifier);
-		mmap_read_lock(mm);
-		ret = hmm_range_fault(&range);
-		mmap_read_unlock(mm);
-		if (ret) {
-			if (ret == -EBUSY)
-				continue;
+		ret = hmm_range_fault_unlocked_timeout(&range,
+				       max_t(long, timeout - jiffies, 1UL));
+		if (ret)
 			goto out;
-		}
 
 		mutex_lock(&svmm->mutex);
 		if (mmu_interval_read_retry(range.notifier,

[PATCH v7 6/8] RDMA/umem: Use hmm_range_fault_unlocked_timeout() for ODP faults

[Stanislav Kinsburskii]

ib_umem_odp_map_dma_and_lock() takes mmap_read_lock() only around
hmm_range_fault(), then retries -EBUSY until HMM_RANGE_DEFAULT_TIMEOUT
expires.

Use hmm_range_fault_unlocked_timeout() instead. The HMM helper now owns
the mmap lock and refreshes range->notifier_seq for its internal retries.
ODP keeps using HMM_RANGE_DEFAULT_TIMEOUT for each HMM fault attempt,
while interval invalidation retries continue to be handled by the existing
outer loop.

ODP still validates the interval notifier sequence while holding umem_mutex
before DMA mapping pages.

Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
---
 drivers/infiniband/core/umem_odp.c |   18 +++++-------------
 1 file changed, 5 insertions(+), 13 deletions(-)

diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c
index 404fa1cc3254..9cc21cd762d9 100644
--- a/drivers/infiniband/core/umem_odp.c
+++ b/drivers/infiniband/core/umem_odp.c
@@ -329,7 +329,7 @@ int ib_umem_odp_map_dma_and_lock(struct ib_umem_odp *umem_odp, u64 user_virt,
 	struct mm_struct *owning_mm = umem_odp->umem.owning_mm;
 	int pfn_index, dma_index, ret = 0, start_idx;
 	unsigned int page_shift, hmm_order, pfn_start_idx;
-	unsigned long num_pfns, current_seq;
+	unsigned long num_pfns;
 	struct hmm_range range = {};
 	unsigned long timeout;
 
@@ -363,26 +363,18 @@ int ib_umem_odp_map_dma_and_lock(struct ib_umem_odp *umem_odp, u64 user_virt,
 	}
 
 	range.hmm_pfns = &(umem_odp->map.pfn_list[pfn_start_idx]);
-	timeout = jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT);
+	timeout = msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT);
 
 retry:
-	current_seq = range.notifier_seq =
-		mmu_interval_read_begin(&umem_odp->notifier);
-
-	mmap_read_lock(owning_mm);
-	ret = hmm_range_fault(&range);
-	mmap_read_unlock(owning_mm);
-	if (unlikely(ret)) {
-		if (ret == -EBUSY && !time_after(jiffies, timeout))
-			goto retry;
+	ret = hmm_range_fault_unlocked_timeout(&range, timeout);
+	if (unlikely(ret))
 		goto out_put_mm;
-	}
 
 	start_idx = (range.start - ib_umem_start(umem_odp)) >> page_shift;
 	dma_index = start_idx;
 
 	mutex_lock(&umem_odp->umem_mutex);
-	if (mmu_interval_read_retry(&umem_odp->notifier, current_seq)) {
+	if (mmu_interval_read_retry(&umem_odp->notifier, range.notifier_seq)) {
 		mutex_unlock(&umem_odp->umem_mutex);
 		goto retry;
 	}

[PATCH v7 7/8] accel/amdxdna: Use hmm_range_fault_unlocked_timeout() for range population

[Stanislav Kinsburskii]

aie2_populate_range() takes mmap_read_lock() only around hmm_range_fault().
It keeps a single HMM_RANGE_DEFAULT_TIMEOUT deadline for the populate pass
and retries -EBUSY until that deadline expires.

Use hmm_range_fault_unlocked_timeout() instead. The HMM helper now owns
the mmap lock and refreshes mapp->range.notifier_seq for its internal
retries. Pass the remaining jiffies from the existing deadline to HMM,
while preserving the driver's existing outer loop for interval invalidation
retries and for selecting the next invalid mapping.

Keep returning -ETIME when the retry budget expires, matching the driver's
existing timeout error convention.

Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
---
 drivers/accel/amdxdna/aie2_ctx.c |   17 +++--------------
 1 file changed, 3 insertions(+), 14 deletions(-)

diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c
index e9fbd8c14364..412f812d7c2c 100644
--- a/drivers/accel/amdxdna/aie2_ctx.c
+++ b/drivers/accel/amdxdna/aie2_ctx.c
@@ -1031,22 +1031,11 @@ static int aie2_populate_range(struct amdxdna_gem_obj *abo)
 		return -EFAULT;
 	}
 
-	mapp->range.notifier_seq = mmu_interval_read_begin(&mapp->notifier);
-	mmap_read_lock(mm);
-	ret = hmm_range_fault(&mapp->range);
-	mmap_read_unlock(mm);
+	ret = hmm_range_fault_unlocked_timeout(&mapp->range,
+			max_t(long, timeout - jiffies, 1));
 	if (ret) {
-		if (time_after(jiffies, timeout)) {
+		if (ret == -EBUSY)
 			ret = -ETIME;
-			goto put_mm;
-		}
-
-		if (ret == -EBUSY) {
-			amdxdna_umap_put(mapp);
-			mmput(mm);
-			goto again;
-		}
-
 		goto put_mm;
 	}
 

[PATCH v7 8/8] drm/gpusvm: Use hmm_range_fault_unlocked_timeout() for range faults

[Stanislav Kinsburskii]

Several GPU SVM paths take mmap_read_lock() only to call hmm_range_fault(),
then retry -EBUSY until HMM_RANGE_DEFAULT_TIMEOUT expires. Those paths use
MMU interval notifiers whose mm matches the mm that was locked for the HMM
fault.

Use hmm_range_fault_unlocked_timeout() for those faults and pass the
remaining retry budget to HMM. The helper owns mmap_lock acquisition and
refreshes range->notifier_seq internally for each retry, while GPU SVM
keeps its existing driver-lock validation with mmu_interval_read_retry()
after a successful fault.

Leave drm_gpusvm_check_pages() on hmm_range_fault() because that path is
called with the mmap lock already held by its caller.

Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
---
 drivers/gpu/drm/drm_gpusvm.c |   52 ++++++------------------------------------
 1 file changed, 7 insertions(+), 45 deletions(-)

diff --git a/drivers/gpu/drm/drm_gpusvm.c b/drivers/gpu/drm/drm_gpusvm.c
index 958cb605aedd..2601b793428c 100644
--- a/drivers/gpu/drm/drm_gpusvm.c
+++ b/drivers/gpu/drm/drm_gpusvm.c
@@ -788,22 +788,8 @@ enum drm_gpusvm_scan_result drm_gpusvm_scan_mm(struct drm_gpusvm_range *range,
 	hmm_range.hmm_pfns = pfns;
 
 retry:
-	hmm_range.notifier_seq = mmu_interval_read_begin(notifier);
-	mmap_read_lock(range->gpusvm->mm);
-
-	while (true) {
-		err = hmm_range_fault(&hmm_range);
-		if (err == -EBUSY) {
-			if (time_after(jiffies, timeout))
-				break;
-
-			hmm_range.notifier_seq =
-				mmu_interval_read_begin(notifier);
-			continue;
-		}
-		break;
-	}
-	mmap_read_unlock(range->gpusvm->mm);
+	err = hmm_range_fault_unlocked_timeout(&hmm_range,
+				max_t(long, timeout - jiffies, 1));
 	if (err)
 		goto err_free;
 
@@ -1439,21 +1425,8 @@ int drm_gpusvm_get_pages(struct drm_gpusvm *gpusvm,
 	}
 
 	hmm_range.hmm_pfns = pfns;
-	while (true) {
-		mmap_read_lock(mm);
-		err = hmm_range_fault(&hmm_range);
-		mmap_read_unlock(mm);
-
-		if (err == -EBUSY) {
-			if (time_after(jiffies, timeout))
-				break;
-
-			hmm_range.notifier_seq =
-				mmu_interval_read_begin(notifier);
-			continue;
-		}
-		break;
-	}
+	err = hmm_range_fault_unlocked_timeout(&hmm_range,
+				max_t(long, timeout - jiffies, 1));
 	mmput(mm);
 	if (err)
 		goto err_free;
@@ -1736,24 +1709,13 @@ int drm_gpusvm_range_evict(struct drm_gpusvm *gpusvm,
 		return -ENOMEM;
 
 	hmm_range.hmm_pfns = pfns;
-	while (!time_after(jiffies, timeout)) {
-		hmm_range.notifier_seq = mmu_interval_read_begin(notifier);
-		if (time_after(jiffies, timeout)) {
-			err = -ETIME;
-			break;
-		}
-
-		mmap_read_lock(mm);
-		err = hmm_range_fault(&hmm_range);
-		mmap_read_unlock(mm);
-		if (err != -EBUSY)
-			break;
-	}
+	err = hmm_range_fault_unlocked_timeout(&hmm_range,
+				max_t(long, timeout - jiffies, 1));
 
 	kvfree(pfns);
 	mmput(mm);
 
-	return err;
+	return err == -EBUSY ? -ETIME : err;
 }
 EXPORT_SYMBOL_GPL(drm_gpusvm_range_evict);
 

Re: [PATCH v7 2/8] mm/hmm: add hmm_range_fault_unlocked_timeout() for mmap lock-drop support

[Jason Gunthorpe]

On Tue, Jul 07, 2026 at 12:47:01PM -0700, Stanislav Kinsburskii wrote:

> +int hmm_range_fault_unlocked_timeout(struct hmm_range *range,
> +				     unsigned long timeout)
> +{
> +	struct mm_struct *mm = range->notifier->mm;
> +	unsigned long deadline = 0;
> +	int locked, ret;
> +
> +	if (timeout)
> +		deadline = jiffies + timeout;
> +
> +	do {
> +		if (fatal_signal_pending(current))
> +			return -EINTR;
> +
> +		if (timeout && time_after(jiffies, deadline))
> +			return -EBUSY;

I really dislike there is a timeout here, HMM is supposed to be more
deterministic. GUP doesn't have a timeout, what is this about?

Jason

Re: [PATCH v7 1/8] mm/hmm: move page fault handling out of walk callbacks

[Jason Gunthorpe]

On Tue, Jul 07, 2026 at 12:46:54PM -0700, Stanislav Kinsburskii wrote:
> hmm_range_fault() currently triggers page faults from inside the page-table
> walk callbacks: hmm_vma_walk_pmd(), hmm_vma_walk_pud(),
> hmm_vma_walk_hugetlb_entry() and the pte-level helper all call
> hmm_vma_fault(), which in turn calls handle_mm_fault() while the walker
> still holds nested locks.  The pte spinlock is dropped explicitly by each
> caller, and the hugetlb path manually drops and retakes
> hugetlb_vma_lock_read around the fault to dodge a deadlock against the walk
> framework's unconditional unlock.
> 
> This layering does not extend cleanly to fault handlers that may release
> mmap_lock (VM_FAULT_RETRY, VM_FAULT_COMPLETED). If the lock is dropped
> while walk_page_range() is mid-traversal, the VMA can be freed before the
> walk framework's matching hugetlb_vma_unlock_read(), turning that unlock
> into a use-after-free.
> 
> Split the responsibilities the way get_user_pages() does. Walk callbacks
> become inspect-only: when they detect a range that needs to be faulted in,
> they record it in struct hmm_vma_walk and return a private sentinel
> (HMM_FAULT_PENDING). The outer loop in hmm_range_fault() then drops out of
> walk_page_range(), invokes a new helper hmm_do_fault() that calls
> handle_mm_fault() with only mmap_lock held, and restarts the walk so the
> now-present entries are collected into hmm_pfns.
> 
> No functional change for existing callers. As a side effect the hugetlb
> callback no longer needs the hugetlb_vma_{un}lock_read dance, and every
> fault-path exit from the callbacks now releases the pte spinlock on a
> single, common path. This refactor is also a precursor for adding an
> unlockable variant of hmm_range_fault() in a follow-up patch.
> 
> Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
> ---
>  mm/hmm.c |  118 +++++++++++++++++++++++++++++++++++++++-----------------------
>  1 file changed, 75 insertions(+), 43 deletions(-)

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>

Jason

Re: [PATCH v7 2/8] mm/hmm: add hmm_range_fault_unlocked_timeout() for mmap lock-drop support

[Jason Gunthorpe]

On Tue, Jul 07, 2026 at 12:47:01PM -0700, Stanislav Kinsburskii wrote:
> hmm_range_fault() requires the caller to hold the mmap read lock for the
> duration of the call. This is incompatible with mappings whose fault
> handler may release the mmap lock, notably userfaultfd-managed regions,
> where handle_mm_fault() can return VM_FAULT_RETRY or VM_FAULT_COMPLETED
> after dropping the lock. Drivers that need to populate device page tables
> for such mappings have no way to do so today.

sashiko could not apply v7 for some reason but the remarks on v6
seemed meaningful, did you see them were they delt with?

https://sashiko.dev/#/patchset/178336023903.504354.7500950448226027718.stgit%40skinsburskii

> diff --git a/Documentation/mm/hmm.rst b/Documentation/mm/hmm.rst
> index 7d61b7a8b65b..70885f153d03 100644
> --- a/Documentation/mm/hmm.rst
> +++ b/Documentation/mm/hmm.rst
> @@ -208,6 +208,69 @@ invalidate() callback. That lock must be held before calling
>  mmu_interval_read_retry() to avoid any race with a concurrent CPU page table
>  update.
>  
> +Dropping the mmap lock during page faults
> +=========================================
> +
> +Some VMAs have fault handlers that need to release the mmap lock while
> +servicing a fault (for example, regions managed by ``userfaultfd``).
> +``hmm_range_fault()`` cannot be used on such mappings because it must hold the
> +mmap lock for the duration of the call. Drivers that need to support them
> +should call::

Given the majority of callers use this API it should probably be the
focus of the documentation and example, regulate the existing API to a
'BTW if you really need the mmap lock, and you really shouldn't, this
exists too'

> @@ -32,6 +32,7 @@
>  
>  struct hmm_vma_walk {
>  	struct hmm_range	*range;
> +	int			*locked;

Let's use bool if you have to respin this

> @@ -651,37 +663,33 @@ static int hmm_do_fault(struct mm_struct *mm,
>  		fault_flags |= FAULT_FLAG_WRITE;
>  	}
>  
> -	for (; addr < end; addr += PAGE_SIZE)
> -		if (handle_mm_fault(vma, addr, fault_flags, NULL) &
> -		    VM_FAULT_ERROR)
> -			return -EFAULT;
> +	for (; addr < end; addr += PAGE_SIZE) {
> +		vm_fault_t ret;
> +
> +		ret = handle_mm_fault(vma, addr, fault_flags, NULL);
> +
> +		if (ret & (VM_FAULT_COMPLETED | VM_FAULT_RETRY)) {
> +			*hmm_vma_walk->locked = 0;
> +			return HMM_FAULT_UNLOCKED;
> +		}
> +
> +		if (ret & VM_FAULT_ERROR) {
> +			int err = vm_fault_to_errno(ret, 0);
> +
> +			if (err)
> +				return err;
> +			BUG();

Linux will be upset if he sees this.  

if (WARN_ON(!err))
   err = -EINVAL

> +/**
> + * hmm_range_fault - try to fault some address in a virtual address range
> + * @range:	argument structure
> + *
> + * Returns 0 on success or one of the following error codes:
> + *
> + * -EINVAL:	Invalid arguments or mm or virtual address is in an invalid vma
> + *		(e.g., device file vma).
> + * -ENOMEM:	Out of memory.
> + * -EPERM:	Invalid permission (e.g., asking for write and range is read
> + *		only).
> + * -EBUSY:	The range has been invalidated and the caller needs to wait for
> + *		the invalidation to finish.
> + * -EFAULT:     A page was requested to be valid and could not be made valid
> + *              ie it has no backing VMA or it is illegal to access
> + *
> + * This is similar to get_user_pages(), except that it can read the page tables
> + * without mutating them (ie causing faults).
> + *
> + * The mmap lock must be held by the caller and will remain held on return.
> + * For a variant that allows the mmap lock to be dropped during faults (e.g.,
> + * for userfaultfd support), see hmm_range_fault_unlocked_timeout().
> + */

Add a comment discourging anyone from using this function and prefer
hmm_range_fault_unlocked_timeout()

Other than the concern about the timeout and minor nits this looks
fine

Jason

Re: [PATCH v7 4/8] mshv: Use hmm_range_fault_unlocked_timeout() for region faults

[Jason Gunthorpe]

On Tue, Jul 07, 2026 at 12:47:15PM -0700, Stanislav Kinsburskii wrote:
> MSHV currently faults movable memory regions by taking mmap_read_lock()
> around hmm_range_fault(). That prevents the fault path from handling VMAs
> whose fault handlers need to drop mmap_lock, such as userfaultfd-backed
> mappings.
> 
> Use hmm_range_fault_unlocked_timeout() instead. Passing a timeout of 0
> preserves MSHV's existing unbounded retry behavior while letting the HMM
> helper own mmap_lock acquisition and refresh range->notifier_seq internally
> before walking the range. After the fault succeeds, MSHV still takes
> mreg_mutex and checks mmu_interval_read_retry() before installing the pages
> into the region, so the existing invalidation synchronization is preserved.
> 
> Fold the small fault-and-lock helper into mshv_region_range_fault(), since
> the remaining retry path is just the standard "fault, take the driver lock,
> check the interval notifier sequence" pattern.
> 
> Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
> ---
>  drivers/hv/mshv_regions.c |   54 ++++++++-------------------------------------
>  1 file changed, 10 insertions(+), 44 deletions(-)

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>

Jason

Re: [PATCH v7 5/8] drm/nouveau: Use hmm_range_fault_unlocked_timeout() for SVM faults

[Jason Gunthorpe]

On Tue, Jul 07, 2026 at 12:47:22PM -0700, Stanislav Kinsburskii wrote:
> nouveau_range_fault() takes mmap_read_lock() only to call
> hmm_range_fault(). It also keeps a single HMM_RANGE_DEFAULT_TIMEOUT
> deadline across both HMM -EBUSY retries and post-fault
> mmu_interval_read_retry() retries.
> 
> Use hmm_range_fault_unlocked_timeout() instead. The HMM helper now owns
> the mmap lock and refreshes range->notifier_seq for its internal retries.
> Nouveau keeps its existing absolute deadline in the outer loop and passes
> the remaining jiffies to the helper for each fault attempt, so retries
> caused by mmu_interval_read_retry() do not reset the overall retry budget.
> 
> Nouveau still validates the interval notifier sequence while holding
> svmm->mutex before programming the GPU mapping.
> 
> Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
> ---
>  drivers/gpu/drm/nouveau/nouveau_svm.c |   11 +++--------
>  1 file changed, 3 insertions(+), 8 deletions(-)

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>

> -		range.notifier_seq = mmu_interval_read_begin(range.notifier);
> -		mmap_read_lock(mm);
> -		ret = hmm_range_fault(&range);
> -		mmap_read_unlock(mm);
> -		if (ret) {
> -			if (ret == -EBUSY)
> -				continue;
> +		ret = hmm_range_fault_unlocked_timeout(&range,
> +				       max_t(long, timeout - jiffies, 1UL));

Avoid max_t, 1L should be a signed long.

Though I'm even more concerned about the timeout if drivers are now
wrapping the whole thing in a timeout... That's not sensible at all

Jason

Re: [PATCH v7 6/8] RDMA/umem: Use hmm_range_fault_unlocked_timeout() for ODP faults

[Jason Gunthorpe]

On Tue, Jul 07, 2026 at 12:47:29PM -0700, Stanislav Kinsburskii wrote:
> ib_umem_odp_map_dma_and_lock() takes mmap_read_lock() only around
> hmm_range_fault(), then retries -EBUSY until HMM_RANGE_DEFAULT_TIMEOUT
> expires.
> 
> Use hmm_range_fault_unlocked_timeout() instead. The HMM helper now owns
> the mmap lock and refreshes range->notifier_seq for its internal retries.
> ODP keeps using HMM_RANGE_DEFAULT_TIMEOUT for each HMM fault attempt,
> while interval invalidation retries continue to be handled by the existing
> outer loop.
> 
> ODP still validates the interval notifier sequence while holding umem_mutex
> before DMA mapping pages.
> 
> Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
> ---
>  drivers/infiniband/core/umem_odp.c |   18 +++++-------------
>  1 file changed, 5 insertions(+), 13 deletions(-)

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>

Jason

Re: [PATCH v7 7/8] accel/amdxdna: Use hmm_range_fault_unlocked_timeout() for range population

[Jason Gunthorpe]

On Tue, Jul 07, 2026 at 12:47:36PM -0700, Stanislav Kinsburskii wrote:
> aie2_populate_range() takes mmap_read_lock() only around hmm_range_fault().
> It keeps a single HMM_RANGE_DEFAULT_TIMEOUT deadline for the populate pass
> and retries -EBUSY until that deadline expires.
> 
> Use hmm_range_fault_unlocked_timeout() instead. The HMM helper now owns
> the mmap lock and refreshes mapp->range.notifier_seq for its internal
> retries. Pass the remaining jiffies from the existing deadline to HMM,
> while preserving the driver's existing outer loop for interval invalidation
> retries and for selecting the next invalid mapping.
> 
> Keep returning -ETIME when the retry budget expires, matching the driver's
> existing timeout error convention.
> 
> Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
> ---
>  drivers/accel/amdxdna/aie2_ctx.c |   17 +++--------------
>  1 file changed, 3 insertions(+), 14 deletions(-)

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>

Jason

Re: [PATCH v7 8/8] drm/gpusvm: Use hmm_range_fault_unlocked_timeout() for range faults

[Jason Gunthorpe]

On Tue, Jul 07, 2026 at 12:47:43PM -0700, Stanislav Kinsburskii wrote:
> Several GPU SVM paths take mmap_read_lock() only to call hmm_range_fault(),
> then retry -EBUSY until HMM_RANGE_DEFAULT_TIMEOUT expires. Those paths use
> MMU interval notifiers whose mm matches the mm that was locked for the HMM
> fault.
> 
> Use hmm_range_fault_unlocked_timeout() for those faults and pass the
> remaining retry budget to HMM. The helper owns mmap_lock acquisition and
> refreshes range->notifier_seq internally for each retry, while GPU SVM
> keeps its existing driver-lock validation with mmu_interval_read_retry()
> after a successful fault.
> 
> Leave drm_gpusvm_check_pages() on hmm_range_fault() because that path is
> called with the mmap lock already held by its caller.
> 
> Signed-off-by: Stanislav Kinsburskii <skinsburskii@gmail.com>
> ---
>  drivers/gpu/drm/drm_gpusvm.c |   52 ++++++------------------------------------
>  1 file changed, 7 insertions(+), 45 deletions(-)

Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>

> -	mmap_read_unlock(range->gpusvm->mm);
> +	err = hmm_range_fault_unlocked_timeout(&hmm_range,
> +				max_t(long, timeout - jiffies, 1));

Same remark about max_t

Jason

Re: [PATCH v7 2/8] mm/hmm: add hmm_range_fault_unlocked_timeout() for mmap lock-drop support

[Jason Gunthorpe]

On Fri, Jul 10, 2026 at 01:27:49PM -0300, Jason Gunthorpe wrote:
> On Tue, Jul 07, 2026 at 12:47:01PM -0700, Stanislav Kinsburskii wrote:
> 
> > +int hmm_range_fault_unlocked_timeout(struct hmm_range *range,
> > +				     unsigned long timeout)
> > +{
> > +	struct mm_struct *mm = range->notifier->mm;
> > +	unsigned long deadline = 0;
> > +	int locked, ret;
> > +
> > +	if (timeout)
> > +		deadline = jiffies + timeout;
> > +
> > +	do {
> > +		if (fatal_signal_pending(current))
> > +			return -EINTR;
> > +
> > +		if (timeout && time_after(jiffies, deadline))
> > +			return -EBUSY;
> 
> I really dislike there is a timeout here, HMM is supposed to be more
> deterministic. GUP doesn't have a timeout, what is this about?

It looks like you've moved the timeout processing related to the mmnu
notifier sequence from the callers into this helper. I'm fine with
that, but maybe add some comments that this timeout is helping
implement the mmu notifiers, and we do expect that the HMM part will
not timeout.

Though it is a little hard to see how your stated purpose of enabling
userfaultfd is going to work, aren't you pretty much guarenteed to hit
a timeout if the userfaultfd process is adversly scheduled? That's
going to end up broken.

So, maybe the deadline should be resetting after every handled fault?
ie the timeout really is only about the mmu notifier and we don't
count the time spent handling faults or walking?

Jason

Re: [PATCH v7 2/8] mm/hmm: add hmm_range_fault_unlocked_timeout() for mmap lock-drop support

[Stanislav Kinsburskii]

On Fri, Jul 10, 2026 at 01:27:49PM -0300, Jason Gunthorpe wrote:
> On Tue, Jul 07, 2026 at 12:47:01PM -0700, Stanislav Kinsburskii wrote:
> 
> > +int hmm_range_fault_unlocked_timeout(struct hmm_range *range,
> > +				     unsigned long timeout)
> > +{
> > +	struct mm_struct *mm = range->notifier->mm;
> > +	unsigned long deadline = 0;
> > +	int locked, ret;
> > +
> > +	if (timeout)
> > +		deadline = jiffies + timeout;
> > +
> > +	do {
> > +		if (fatal_signal_pending(current))
> > +			return -EINTR;
> > +
> > +		if (timeout && time_after(jiffies, deadline))
> > +			return -EBUSY;
> 
> I really dislike there is a timeout here, HMM is supposed to be more
> deterministic. GUP doesn't have a timeout, what is this about?
> 

The timeout was added because this version makes the unlocked helper own
the internal retry loop, including -EBUSY retries.

Several existing HMM users already bound those retries with
HMM_RANGE_DEFAULT_TIMEOUT, so the timeout argument was meant to avoid
silently turning those call sites into unbounded waits.

That said, I agree this mixes driver retry policy into HMM. A cleaner
split may be to make the new helper timeout-free and have it retry
internally only when mmap_lock was dropped by the fault path, since that
is the part HMM must hide for userfaultfd-style faults.

If the walk fails with the normal HMM invalidation -EBUSY, the helper
would return -EBUSY to the caller, preserving the existing driver
timeout loops and keeping HMM deterministic, closer to GUP.

Is it better from your POV?

THanks,
Stanislav

> Jason

Re: [PATCH v7 2/8] mm/hmm: add hmm_range_fault_unlocked_timeout() for mmap lock-drop support

[Stanislav Kinsburskii]

On Fri, Jul 10, 2026 at 01:47:43PM -0300, Jason Gunthorpe wrote:
> On Fri, Jul 10, 2026 at 01:27:49PM -0300, Jason Gunthorpe wrote:
> > On Tue, Jul 07, 2026 at 12:47:01PM -0700, Stanislav Kinsburskii wrote:
> > 
> > > +int hmm_range_fault_unlocked_timeout(struct hmm_range *range,
> > > +				     unsigned long timeout)
> > > +{
> > > +	struct mm_struct *mm = range->notifier->mm;
> > > +	unsigned long deadline = 0;
> > > +	int locked, ret;
> > > +
> > > +	if (timeout)
> > > +		deadline = jiffies + timeout;
> > > +
> > > +	do {
> > > +		if (fatal_signal_pending(current))
> > > +			return -EINTR;
> > > +
> > > +		if (timeout && time_after(jiffies, deadline))
> > > +			return -EBUSY;
> > 
> > I really dislike there is a timeout here, HMM is supposed to be more
> > deterministic. GUP doesn't have a timeout, what is this about?
> 
> It looks like you've moved the timeout processing related to the mmnu
> notifier sequence from the callers into this helper. I'm fine with
> that, but maybe add some comments that this timeout is helping
> implement the mmu notifiers, and we do expect that the HMM part will
> not timeout.
> 
> Though it is a little hard to see how your stated purpose of enabling
> userfaultfd is going to work, aren't you pretty much guarenteed to hit
> a timeout if the userfaultfd process is adversly scheduled? That's
> going to end up broken.
> 

The main customer for this new feature I have in mind is the MSHV driver
which backs VMs memory with HMM, requires userfaultfd support for
post-copy live migration and fast restore and it doesn't timeout.

I agree, that this current timeout value used by the other callers might
not be enough to repopulate the mappings with userfaultfd, but there
drivers would get -EFAULT for uderfaultfd-backed mappings without this
change anyway, so getting -EBUSY with the change instead doesn't look
like a significant change to the behaviour from my POV.

> So, maybe the deadline should be resetting after every handled fault?
> ie the timeout really is only about the mmu notifier and we don't
> count the time spent handling faults or walking?
> 

The timeout was inherited from existing HMM users rather than introduced
as a new HMM policy. Some GPU drivers use HMM_RANGE_DEFAULT_TIMEOUT as a
budget for the whole range population operation, including HMM retries
and subsequent driver mapping work.

Moving retry handling into the unlocked helper would otherwise hide
repeated -EBUSY returns from those callers and silently turn their
bounded operation into an unbounded one. So the timeout argument is
there to preserve existing caller semantics during conversion.

I'd say it's up to the caller to provide big enough timeout for
userfaultfd to succeed.

Thanks,
Stanislav

> Jason

Re: [PATCH v7 2/8] mm/hmm: add hmm_range_fault_unlocked_timeout() for mmap lock-drop support

[Stanislav Kinsburskii]

On Fri, Jul 10, 2026 at 01:38:35PM -0300, Jason Gunthorpe wrote:
> On Tue, Jul 07, 2026 at 12:47:01PM -0700, Stanislav Kinsburskii wrote:
> > hmm_range_fault() requires the caller to hold the mmap read lock for the
> > duration of the call. This is incompatible with mappings whose fault
> > handler may release the mmap lock, notably userfaultfd-managed regions,
> > where handle_mm_fault() can return VM_FAULT_RETRY or VM_FAULT_COMPLETED
> > after dropping the lock. Drivers that need to populate device page tables
> > for such mappings have no way to do so today.
> 
> sashiko could not apply v7 for some reason but the remarks on v6
> seemed meaningful, did you see them were they delt with?
> 

Yes, I dealt with them.

> https://sashiko.dev/#/patchset/178336023903.504354.7500950448226027718.stgit%40skinsburskii
> 
> > diff --git a/Documentation/mm/hmm.rst b/Documentation/mm/hmm.rst
> > index 7d61b7a8b65b..70885f153d03 100644
> > --- a/Documentation/mm/hmm.rst
> > +++ b/Documentation/mm/hmm.rst
> > @@ -208,6 +208,69 @@ invalidate() callback. That lock must be held before calling
> >  mmu_interval_read_retry() to avoid any race with a concurrent CPU page table
> >  update.
> >  
> > +Dropping the mmap lock during page faults
> > +=========================================
> > +
> > +Some VMAs have fault handlers that need to release the mmap lock while
> > +servicing a fault (for example, regions managed by ``userfaultfd``).
> > +``hmm_range_fault()`` cannot be used on such mappings because it must hold the
> > +mmap lock for the duration of the call. Drivers that need to support them
> > +should call::
> 
> Given the majority of callers use this API it should probably be the
> focus of the documentation and example, regulate the existing API to a
> 'BTW if you really need the mmap lock, and you really shouldn't, this
> exists too'
> 

Sure, I'll update the doc to reflect it this way.

> > @@ -32,6 +32,7 @@
> >  
> >  struct hmm_vma_walk {
> >  	struct hmm_range	*range;
> > +	int			*locked;
> 
> Let's use bool if you have to respin this
> 

Sure.

> > @@ -651,37 +663,33 @@ static int hmm_do_fault(struct mm_struct *mm,
> >  		fault_flags |= FAULT_FLAG_WRITE;
> >  	}
> >  
> > -	for (; addr < end; addr += PAGE_SIZE)
> > -		if (handle_mm_fault(vma, addr, fault_flags, NULL) &
> > -		    VM_FAULT_ERROR)
> > -			return -EFAULT;
> > +	for (; addr < end; addr += PAGE_SIZE) {
> > +		vm_fault_t ret;
> > +
> > +		ret = handle_mm_fault(vma, addr, fault_flags, NULL);
> > +
> > +		if (ret & (VM_FAULT_COMPLETED | VM_FAULT_RETRY)) {
> > +			*hmm_vma_walk->locked = 0;
> > +			return HMM_FAULT_UNLOCKED;
> > +		}
> > +
> > +		if (ret & VM_FAULT_ERROR) {
> > +			int err = vm_fault_to_errno(ret, 0);
> > +
> > +			if (err)
> > +				return err;
> > +			BUG();
> 
> Linux will be upset if he sees this.  
> 
> if (WARN_ON(!err))
>    err = -EINVAL
> 

It will. I copied it from GUP.
I'll change it the way you propose it.

> > +/**
> > + * hmm_range_fault - try to fault some address in a virtual address range
> > + * @range:	argument structure
> > + *
> > + * Returns 0 on success or one of the following error codes:
> > + *
> > + * -EINVAL:	Invalid arguments or mm or virtual address is in an invalid vma
> > + *		(e.g., device file vma).
> > + * -ENOMEM:	Out of memory.
> > + * -EPERM:	Invalid permission (e.g., asking for write and range is read
> > + *		only).
> > + * -EBUSY:	The range has been invalidated and the caller needs to wait for
> > + *		the invalidation to finish.
> > + * -EFAULT:     A page was requested to be valid and could not be made valid
> > + *              ie it has no backing VMA or it is illegal to access
> > + *
> > + * This is similar to get_user_pages(), except that it can read the page tables
> > + * without mutating them (ie causing faults).
> > + *
> > + * The mmap lock must be held by the caller and will remain held on return.
> > + * For a variant that allows the mmap lock to be dropped during faults (e.g.,
> > + * for userfaultfd support), see hmm_range_fault_unlocked_timeout().
> > + */
> 
> Add a comment discourging anyone from using this function and prefer
> hmm_range_fault_unlocked_timeout()
> 

Will do.

Thanks,
Stanislav

> Other than the concern about the timeout and minor nits this looks
> fine
> 
> Jason

Re: [PATCH v7 2/8] mm/hmm: add hmm_range_fault_unlocked_timeout() for mmap lock-drop support

[Jason Gunthorpe]

On Fri, Jul 10, 2026 at 10:02:55AM -0700, Stanislav Kinsburskii wrote:
> The main customer for this new feature I have in mind is the MSHV driver
> which backs VMs memory with HMM, requires userfaultfd support for
> post-copy live migration and fast restore and it doesn't timeout.
> 
> I agree, that this current timeout value used by the other callers might
> not be enough to repopulate the mappings with userfaultfd, but there
> drivers would get -EFAULT for uderfaultfd-backed mappings without this
> change anyway, so getting -EBUSY with the change instead doesn't look
> like a significant change to the behaviour from my POV.

It sounds like it won't be reliable either then.

> > So, maybe the deadline should be resetting after every handled fault?
> > ie the timeout really is only about the mmu notifier and we don't
> > count the time spent handling faults or walking?
>
> The timeout was inherited from existing HMM users rather than introduced
> as a new HMM policy. Some GPU drivers use HMM_RANGE_DEFAULT_TIMEOUT as a
> budget for the whole range population operation, including HMM retries
> and subsequent driver mapping work.

Yes, because we always had a timeout around the notifier because that
scheme can sort of live lock. The timeout was to protect that only, ie
limit the number of notifier retries.

Expanding the timeout to be outside what is bounded by the notifier
retry is not right, and heavy stuff like mapping should be done after
the hmm side succeeds and the notifiers concluded so they can rely on
normal locking instead.

This is why I'm suggesting to reset the deadline as hmm makes forward
progress, we really only want to bound the notifier retry loop not
anything else.

Jason

Re: [PATCH v7 2/8] mm/hmm: add hmm_range_fault_unlocked_timeout() for mmap lock-drop support

[Stanislav Kinsburskii]

On Fri, Jul 10, 2026 at 03:03:12PM -0300, Jason Gunthorpe wrote:
> On Fri, Jul 10, 2026 at 10:02:55AM -0700, Stanislav Kinsburskii wrote:
> > The main customer for this new feature I have in mind is the MSHV driver
> > which backs VMs memory with HMM, requires userfaultfd support for
> > post-copy live migration and fast restore and it doesn't timeout.
> > 
> > I agree, that this current timeout value used by the other callers might
> > not be enough to repopulate the mappings with userfaultfd, but there
> > drivers would get -EFAULT for uderfaultfd-backed mappings without this
> > change anyway, so getting -EBUSY with the change instead doesn't look
> > like a significant change to the behaviour from my POV.
> 
> It sounds like it won't be reliable either then.
> 
> > > So, maybe the deadline should be resetting after every handled fault?
> > > ie the timeout really is only about the mmu notifier and we don't
> > > count the time spent handling faults or walking?
> >
> > The timeout was inherited from existing HMM users rather than introduced
> > as a new HMM policy. Some GPU drivers use HMM_RANGE_DEFAULT_TIMEOUT as a
> > budget for the whole range population operation, including HMM retries
> > and subsequent driver mapping work.
> 
> Yes, because we always had a timeout around the notifier because that
> scheme can sort of live lock. The timeout was to protect that only, ie
> limit the number of notifier retries.
> 
> Expanding the timeout to be outside what is bounded by the notifier
> retry is not right, and heavy stuff like mapping should be done after
> the hmm side succeeds and the notifiers concluded so they can rely on
> normal locking instead.
> 
> This is why I'm suggesting to reset the deadline as hmm makes forward
> progress, we really only want to bound the notifier retry loop not
> anything else.
> 

Sure, I'll modify accordingly.

Thanks,
Stanislav

> Jason