[PATCH v2 0/3] mmu notifier contextual informations

[PATCH v2 0/3] mmu notifier contextual informations

[jglisse]

From: Jérôme Glisse <jglisse@redhat.com>

Changes since v1:

- Fixed the case where mmu notifier is not enabled and avoid wasting
  memory and resource when that is the case.
- Fixed bug in migrate code.
- Use kernel doc format for describing kernel enum


v1 cover letter:

This patchset add contextual information, why an invalidation is
happening, to mmu notifier callback. This is necessary for user
of mmu notifier that wish to maintains their own data structure
without having to add new fields to struct vm_area_struct (vma).

For instance device can have they own page table that mirror the
process address space. When a vma is unmap (munmap() syscall) the
device driver can free the device page table for the range.

Today we do not have any information on why a mmu notifier call
back is happening and thus device driver have to assume that it
is always an munmap(). This is inefficient at it means that it
needs to re-allocate device page table on next page fault and
rebuild the whole device driver data structure for the range.

Other use case beside munmap() also exist, for instance it is
pointless for device driver to invalidate the device page table
when the invalidation is for the soft dirtyness tracking. Or
device driver can optimize away mprotect() that change the page
table permission access for the range.

This patchset enable all this optimizations for device driver.
I do not include any of those in this serie but other patchset
i am posting will leverage this.


From code point of view the patchset is pretty simple, the first
two patches consolidate all mmu notifier arguments into a struct
so that it is easier to add/change arguments. The last patch adds
the contextual information (munmap, protection, soft dirty, clear,
...).

Cheers,
Jérôme

Jérôme Glisse (3):
  mm/mmu_notifier: use structure for invalidate_range_start/end callback
  mm/mmu_notifier: use structure for invalidate_range_start/end calls v2
  mm/mmu_notifier: contextual information for event triggering
    invalidation v2

 drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c  |  43 ++++----
 drivers/gpu/drm/i915/i915_gem_userptr.c |  14 ++-
 drivers/gpu/drm/radeon/radeon_mn.c      |  16 ++-
 drivers/infiniband/core/umem_odp.c      |  20 ++--
 drivers/infiniband/hw/hfi1/mmu_rb.c     |  13 +--
 drivers/misc/mic/scif/scif_dma.c        |  11 +-
 drivers/misc/sgi-gru/grutlbpurge.c      |  14 ++-
 drivers/xen/gntdev.c                    |  12 +--
 fs/dax.c                                |  15 ++-
 fs/proc/task_mmu.c                      |   8 +-
 include/linux/mm.h                      |   4 +-
 include/linux/mmu_notifier.h            | 132 ++++++++++++++++++------
 kernel/events/uprobes.c                 |  11 +-
 mm/hmm.c                                |  23 ++---
 mm/huge_memory.c                        |  58 +++++------
 mm/hugetlb.c                            |  54 +++++-----
 mm/khugepaged.c                         |  11 +-
 mm/ksm.c                                |  23 ++---
 mm/madvise.c                            |  22 ++--
 mm/memory.c                             | 103 +++++++++---------
 mm/migrate.c                            |  26 ++---
 mm/mmu_notifier.c                       |  22 ++--
 mm/mprotect.c                           |  16 +--
 mm/mremap.c                             |  11 +-
 mm/oom_kill.c                           |  17 +--
 mm/rmap.c                               |  32 +++---
 virt/kvm/kvm_main.c                     |  14 +--
 27 files changed, 404 insertions(+), 341 deletions(-)

-- 
2.17.2

[PATCH v2 1/3] mm/mmu_notifier: use structure for invalidate_range_start/end callback

[jglisse]

From: Jérôme Glisse <jglisse@redhat.com>

To avoid having to change many callback definition everytime we want
to add a parameter use a structure to group all parameters for the
mmu_notifier invalidate_range_start/end callback. No functional changes
with this patch.

Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Christian Koenig <christian.koenig@amd.com>
Cc: Felix Kuehling <felix.kuehling@amd.com>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: kvm@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org
Cc: linux-rdma@vger.kernel.org
Cc: linux-fsdevel@vger.kernel.org
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c  | 43 +++++++++++--------------
 drivers/gpu/drm/i915/i915_gem_userptr.c | 14 ++++----
 drivers/gpu/drm/radeon/radeon_mn.c      | 16 ++++-----
 drivers/infiniband/core/umem_odp.c      | 20 +++++-------
 drivers/infiniband/hw/hfi1/mmu_rb.c     | 13 +++-----
 drivers/misc/mic/scif/scif_dma.c        | 11 ++-----
 drivers/misc/sgi-gru/grutlbpurge.c      | 14 ++++----
 drivers/xen/gntdev.c                    | 12 +++----
 include/linux/mmu_notifier.h            | 14 +++++---
 mm/hmm.c                                | 23 ++++++-------
 mm/mmu_notifier.c                       | 21 ++++++++++--
 virt/kvm/kvm_main.c                     | 14 +++-----
 12 files changed, 102 insertions(+), 113 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c
index e55508b39496..5bc7e59a05a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c
@@ -246,36 +246,34 @@ static void amdgpu_mn_invalidate_node(struct amdgpu_mn_node *node,
  * potentially dirty.
  */
 static int amdgpu_mn_invalidate_range_start_gfx(struct mmu_notifier *mn,
-						 struct mm_struct *mm,
-						 unsigned long start,
-						 unsigned long end,
-						 bool blockable)
+			const struct mmu_notifier_range *range)
 {
 	struct amdgpu_mn *amn = container_of(mn, struct amdgpu_mn, mn);
 	struct interval_tree_node *it;
+	unsigned long end;
 
 	/* notification is exclusive, but interval is inclusive */
-	end -= 1;
+	end = range->end - 1;
 
 	/* TODO we should be able to split locking for interval tree and
 	 * amdgpu_mn_invalidate_node
 	 */
-	if (amdgpu_mn_read_lock(amn, blockable))
+	if (amdgpu_mn_read_lock(amn, range->blockable))
 		return -EAGAIN;
 
-	it = interval_tree_iter_first(&amn->objects, start, end);
+	it = interval_tree_iter_first(&amn->objects, range->start, end);
 	while (it) {
 		struct amdgpu_mn_node *node;
 
-		if (!blockable) {
+		if (!range->blockable) {
 			amdgpu_mn_read_unlock(amn);
 			return -EAGAIN;
 		}
 
 		node = container_of(it, struct amdgpu_mn_node, it);
-		it = interval_tree_iter_next(it, start, end);
+		it = interval_tree_iter_next(it, range->start, end);
 
-		amdgpu_mn_invalidate_node(node, start, end);
+		amdgpu_mn_invalidate_node(node, range->start, end);
 	}
 
 	return 0;
@@ -294,39 +292,38 @@ static int amdgpu_mn_invalidate_range_start_gfx(struct mmu_notifier *mn,
  * are restorted in amdgpu_mn_invalidate_range_end_hsa.
  */
 static int amdgpu_mn_invalidate_range_start_hsa(struct mmu_notifier *mn,
-						 struct mm_struct *mm,
-						 unsigned long start,
-						 unsigned long end,
-						 bool blockable)
+			const struct mmu_notifier_range *range)
 {
 	struct amdgpu_mn *amn = container_of(mn, struct amdgpu_mn, mn);
 	struct interval_tree_node *it;
+	unsigned long end;
 
 	/* notification is exclusive, but interval is inclusive */
-	end -= 1;
+	end = range->end - 1;
 
-	if (amdgpu_mn_read_lock(amn, blockable))
+	if (amdgpu_mn_read_lock(amn, range->blockable))
 		return -EAGAIN;
 
-	it = interval_tree_iter_first(&amn->objects, start, end);
+	it = interval_tree_iter_first(&amn->objects, range->start, end);
 	while (it) {
 		struct amdgpu_mn_node *node;
 		struct amdgpu_bo *bo;
 
-		if (!blockable) {
+		if (!range->blockable) {
 			amdgpu_mn_read_unlock(amn);
 			return -EAGAIN;
 		}
 
 		node = container_of(it, struct amdgpu_mn_node, it);
-		it = interval_tree_iter_next(it, start, end);
+		it = interval_tree_iter_next(it, range->start, end);
 
 		list_for_each_entry(bo, &node->bos, mn_list) {
 			struct kgd_mem *mem = bo->kfd_bo;
 
 			if (amdgpu_ttm_tt_affect_userptr(bo->tbo.ttm,
-							 start, end))
-				amdgpu_amdkfd_evict_userptr(mem, mm);
+							 range->start,
+							 end))
+				amdgpu_amdkfd_evict_userptr(mem, range->mm);
 		}
 	}
 
@@ -344,9 +341,7 @@ static int amdgpu_mn_invalidate_range_start_hsa(struct mmu_notifier *mn,
  * Release the lock again to allow new command submissions.
  */
 static void amdgpu_mn_invalidate_range_end(struct mmu_notifier *mn,
-					   struct mm_struct *mm,
-					   unsigned long start,
-					   unsigned long end)
+			const struct mmu_notifier_range *range)
 {
 	struct amdgpu_mn *amn = container_of(mn, struct amdgpu_mn, mn);
 
diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index 2c9b284036d1..3df77020aada 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -113,27 +113,25 @@ static void del_object(struct i915_mmu_object *mo)
 }
 
 static int i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
-						       struct mm_struct *mm,
-						       unsigned long start,
-						       unsigned long end,
-						       bool blockable)
+			const struct mmu_notifier_range *range)
 {
 	struct i915_mmu_notifier *mn =
 		container_of(_mn, struct i915_mmu_notifier, mn);
 	struct i915_mmu_object *mo;
 	struct interval_tree_node *it;
 	LIST_HEAD(cancelled);
+	unsigned long end;
 
 	if (RB_EMPTY_ROOT(&mn->objects.rb_root))
 		return 0;
 
 	/* interval ranges are inclusive, but invalidate range is exclusive */
-	end--;
+	end = range->end - 1;
 
 	spin_lock(&mn->lock);
-	it = interval_tree_iter_first(&mn->objects, start, end);
+	it = interval_tree_iter_first(&mn->objects, range->start, end);
 	while (it) {
-		if (!blockable) {
+		if (!range->blockable) {
 			spin_unlock(&mn->lock);
 			return -EAGAIN;
 		}
@@ -151,7 +149,7 @@ static int i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
 			queue_work(mn->wq, &mo->work);
 
 		list_add(&mo->link, &cancelled);
-		it = interval_tree_iter_next(it, start, end);
+		it = interval_tree_iter_next(it, range->start, end);
 	}
 	list_for_each_entry(mo, &cancelled, link)
 		del_object(mo);
diff --git a/drivers/gpu/drm/radeon/radeon_mn.c b/drivers/gpu/drm/radeon/radeon_mn.c
index f8b35df44c60..b3019505065a 100644
--- a/drivers/gpu/drm/radeon/radeon_mn.c
+++ b/drivers/gpu/drm/radeon/radeon_mn.c
@@ -119,40 +119,38 @@ static void radeon_mn_release(struct mmu_notifier *mn,
  * unmap them by move them into system domain again.
  */
 static int radeon_mn_invalidate_range_start(struct mmu_notifier *mn,
-					     struct mm_struct *mm,
-					     unsigned long start,
-					     unsigned long end,
-					     bool blockable)
+				const struct mmu_notifier_range *range)
 {
 	struct radeon_mn *rmn = container_of(mn, struct radeon_mn, mn);
 	struct ttm_operation_ctx ctx = { false, false };
 	struct interval_tree_node *it;
+	unsigned long end;
 	int ret = 0;
 
 	/* notification is exclusive, but interval is inclusive */
-	end -= 1;
+	end = range->end - 1;
 
 	/* TODO we should be able to split locking for interval tree and
 	 * the tear down.
 	 */
-	if (blockable)
+	if (range->blockable)
 		mutex_lock(&rmn->lock);
 	else if (!mutex_trylock(&rmn->lock))
 		return -EAGAIN;
 
-	it = interval_tree_iter_first(&rmn->objects, start, end);
+	it = interval_tree_iter_first(&rmn->objects, range->start, end);
 	while (it) {
 		struct radeon_mn_node *node;
 		struct radeon_bo *bo;
 		long r;
 
-		if (!blockable) {
+		if (!range->blockable) {
 			ret = -EAGAIN;
 			goto out_unlock;
 		}
 
 		node = container_of(it, struct radeon_mn_node, it);
-		it = interval_tree_iter_next(it, start, end);
+		it = interval_tree_iter_next(it, range->start, end);
 
 		list_for_each_entry(bo, &node->bos, mn_list) {
 
diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c
index 676c1fd1119d..25db6ff68c70 100644
--- a/drivers/infiniband/core/umem_odp.c
+++ b/drivers/infiniband/core/umem_odp.c
@@ -146,15 +146,12 @@ static int invalidate_range_start_trampoline(struct ib_umem_odp *item,
 }
 
 static int ib_umem_notifier_invalidate_range_start(struct mmu_notifier *mn,
-						    struct mm_struct *mm,
-						    unsigned long start,
-						    unsigned long end,
-						    bool blockable)
+				const struct mmu_notifier_range *range)
 {
 	struct ib_ucontext_per_mm *per_mm =
 		container_of(mn, struct ib_ucontext_per_mm, mn);
 
-	if (blockable)
+	if (range->blockable)
 		down_read(&per_mm->umem_rwsem);
 	else if (!down_read_trylock(&per_mm->umem_rwsem))
 		return -EAGAIN;
@@ -169,9 +166,10 @@ static int ib_umem_notifier_invalidate_range_start(struct mmu_notifier *mn,
 		return 0;
 	}
 
-	return rbt_ib_umem_for_each_in_range(&per_mm->umem_tree, start, end,
+	return rbt_ib_umem_for_each_in_range(&per_mm->umem_tree, range->start,
+					     range->end,
 					     invalidate_range_start_trampoline,
-					     blockable, NULL);
+					     range->blockable, NULL);
 }
 
 static int invalidate_range_end_trampoline(struct ib_umem_odp *item, u64 start,
@@ -182,9 +180,7 @@ static int invalidate_range_end_trampoline(struct ib_umem_odp *item, u64 start,
 }
 
 static void ib_umem_notifier_invalidate_range_end(struct mmu_notifier *mn,
-						  struct mm_struct *mm,
-						  unsigned long start,
-						  unsigned long end)
+				const struct mmu_notifier_range *range)
 {
 	struct ib_ucontext_per_mm *per_mm =
 		container_of(mn, struct ib_ucontext_per_mm, mn);
@@ -192,8 +188,8 @@ static void ib_umem_notifier_invalidate_range_end(struct mmu_notifier *mn,
 	if (unlikely(!per_mm->active))
 		return;
 
-	rbt_ib_umem_for_each_in_range(&per_mm->umem_tree, start,
-				      end,
+	rbt_ib_umem_for_each_in_range(&per_mm->umem_tree, range->start,
+				      range->end,
 				      invalidate_range_end_trampoline, true, NULL);
 	up_read(&per_mm->umem_rwsem);
 }
diff --git a/drivers/infiniband/hw/hfi1/mmu_rb.c b/drivers/infiniband/hw/hfi1/mmu_rb.c
index 475b769e120c..14d2a90964c3 100644
--- a/drivers/infiniband/hw/hfi1/mmu_rb.c
+++ b/drivers/infiniband/hw/hfi1/mmu_rb.c
@@ -68,8 +68,7 @@ struct mmu_rb_handler {
 static unsigned long mmu_node_start(struct mmu_rb_node *);
 static unsigned long mmu_node_last(struct mmu_rb_node *);
 static int mmu_notifier_range_start(struct mmu_notifier *,
-				     struct mm_struct *,
-				     unsigned long, unsigned long, bool);
+		const struct mmu_notifier_range *);
 static struct mmu_rb_node *__mmu_rb_search(struct mmu_rb_handler *,
 					   unsigned long, unsigned long);
 static void do_remove(struct mmu_rb_handler *handler,
@@ -284,10 +283,7 @@ void hfi1_mmu_rb_remove(struct mmu_rb_handler *handler,
 }
 
 static int mmu_notifier_range_start(struct mmu_notifier *mn,
-				     struct mm_struct *mm,
-				     unsigned long start,
-				     unsigned long end,
-				     bool blockable)
+		const struct mmu_notifier_range *range)
 {
 	struct mmu_rb_handler *handler =
 		container_of(mn, struct mmu_rb_handler, mn);
@@ -297,10 +293,11 @@ static int mmu_notifier_range_start(struct mmu_notifier *mn,
 	bool added = false;
 
 	spin_lock_irqsave(&handler->lock, flags);
-	for (node = __mmu_int_rb_iter_first(root, start, end - 1);
+	for (node = __mmu_int_rb_iter_first(root, range->start, range->end-1);
 	     node; node = ptr) {
 		/* Guard against node removal. */
-		ptr = __mmu_int_rb_iter_next(node, start, end - 1);
+		ptr = __mmu_int_rb_iter_next(node, range->start,
+					     range->end - 1);
 		trace_hfi1_mmu_mem_invalidate(node->addr, node->len);
 		if (handler->ops->invalidate(handler->ops_arg, node)) {
 			__mmu_int_rb_remove(node, root);
diff --git a/drivers/misc/mic/scif/scif_dma.c b/drivers/misc/mic/scif/scif_dma.c
index 18b8ed57c4ac..e0d97044d0e9 100644
--- a/drivers/misc/mic/scif/scif_dma.c
+++ b/drivers/misc/mic/scif/scif_dma.c
@@ -201,23 +201,18 @@ static void scif_mmu_notifier_release(struct mmu_notifier *mn,
 }
 
 static int scif_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
-						     struct mm_struct *mm,
-						     unsigned long start,
-						     unsigned long end,
-						     bool blockable)
+					const struct mmu_notifier_range *range)
 {
 	struct scif_mmu_notif	*mmn;
 
 	mmn = container_of(mn, struct scif_mmu_notif, ep_mmu_notifier);
-	scif_rma_destroy_tcw(mmn, start, end - start);
+	scif_rma_destroy_tcw(mmn, range->start, range->end - range->start);
 
 	return 0;
 }
 
 static void scif_mmu_notifier_invalidate_range_end(struct mmu_notifier *mn,
-						   struct mm_struct *mm,
-						   unsigned long start,
-						   unsigned long end)
+			const struct mmu_notifier_range *range)
 {
 	/*
 	 * Nothing to do here, everything needed was done in
diff --git a/drivers/misc/sgi-gru/grutlbpurge.c b/drivers/misc/sgi-gru/grutlbpurge.c
index 03b49d52092e..ca2032afe035 100644
--- a/drivers/misc/sgi-gru/grutlbpurge.c
+++ b/drivers/misc/sgi-gru/grutlbpurge.c
@@ -220,9 +220,7 @@ void gru_flush_all_tlb(struct gru_state *gru)
  * MMUOPS notifier callout functions
  */
 static int gru_invalidate_range_start(struct mmu_notifier *mn,
-				       struct mm_struct *mm,
-				       unsigned long start, unsigned long end,
-				       bool blockable)
+			const struct mmu_notifier_range *range)
 {
 	struct gru_mm_struct *gms = container_of(mn, struct gru_mm_struct,
 						 ms_notifier);
@@ -230,15 +228,14 @@ static int gru_invalidate_range_start(struct mmu_notifier *mn,
 	STAT(mmu_invalidate_range);
 	atomic_inc(&gms->ms_range_active);
 	gru_dbg(grudev, "gms %p, start 0x%lx, end 0x%lx, act %d\n", gms,
-		start, end, atomic_read(&gms->ms_range_active));
-	gru_flush_tlb_range(gms, start, end - start);
+		range->start, range->end, atomic_read(&gms->ms_range_active));
+	gru_flush_tlb_range(gms, range->start, range->end - range->start);
 
 	return 0;
 }
 
 static void gru_invalidate_range_end(struct mmu_notifier *mn,
-				     struct mm_struct *mm, unsigned long start,
-				     unsigned long end)
+			const struct mmu_notifier_range *range)
 {
 	struct gru_mm_struct *gms = container_of(mn, struct gru_mm_struct,
 						 ms_notifier);
@@ -247,7 +244,8 @@ static void gru_invalidate_range_end(struct mmu_notifier *mn,
 	(void)atomic_dec_and_test(&gms->ms_range_active);
 
 	wake_up_all(&gms->ms_wait_queue);
-	gru_dbg(grudev, "gms %p, start 0x%lx, end 0x%lx\n", gms, start, end);
+	gru_dbg(grudev, "gms %p, start 0x%lx, end 0x%lx\n",
+		gms, range->start, range->end);
 }
 
 static void gru_release(struct mmu_notifier *mn, struct mm_struct *mm)
diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c
index b0b02a501167..5efc5eee9544 100644
--- a/drivers/xen/gntdev.c
+++ b/drivers/xen/gntdev.c
@@ -520,26 +520,26 @@ static int unmap_if_in_range(struct gntdev_grant_map *map,
 }
 
 static int mn_invl_range_start(struct mmu_notifier *mn,
-				struct mm_struct *mm,
-				unsigned long start, unsigned long end,
-				bool blockable)
+			       const struct mmu_notifier_range *range)
 {
 	struct gntdev_priv *priv = container_of(mn, struct gntdev_priv, mn);
 	struct gntdev_grant_map *map;
 	int ret = 0;
 
-	if (blockable)
+	if (range->blockable)
 		mutex_lock(&priv->lock);
 	else if (!mutex_trylock(&priv->lock))
 		return -EAGAIN;
 
 	list_for_each_entry(map, &priv->maps, next) {
-		ret = unmap_if_in_range(map, start, end, blockable);
+		ret = unmap_if_in_range(map, range->start, range->end,
+					range->blockable);
 		if (ret)
 			goto out_unlock;
 	}
 	list_for_each_entry(map, &priv->freeable_maps, next) {
-		ret = unmap_if_in_range(map, start, end, blockable);
+		ret = unmap_if_in_range(map, range->start, range->end,
+					range->blockable);
 		if (ret)
 			goto out_unlock;
 	}
diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h
index 9893a6432adf..368f0c1a049d 100644
--- a/include/linux/mmu_notifier.h
+++ b/include/linux/mmu_notifier.h
@@ -25,6 +25,13 @@ struct mmu_notifier_mm {
 	spinlock_t lock;
 };
 
+struct mmu_notifier_range {
+	struct mm_struct *mm;
+	unsigned long start;
+	unsigned long end;
+	bool blockable;
+};
+
 struct mmu_notifier_ops {
 	/*
 	 * Called either by mmu_notifier_unregister or when the mm is
@@ -146,12 +153,9 @@ struct mmu_notifier_ops {
 	 *
 	 */
 	int (*invalidate_range_start)(struct mmu_notifier *mn,
-				       struct mm_struct *mm,
-				       unsigned long start, unsigned long end,
-				       bool blockable);
+				      const struct mmu_notifier_range *range);
 	void (*invalidate_range_end)(struct mmu_notifier *mn,
-				     struct mm_struct *mm,
-				     unsigned long start, unsigned long end);
+				     const struct mmu_notifier_range *range);
 
 	/*
 	 * invalidate_range() is either called between
diff --git a/mm/hmm.c b/mm/hmm.c
index 90c34f3d1243..1965f2caf5eb 100644
--- a/mm/hmm.c
+++ b/mm/hmm.c
@@ -189,35 +189,30 @@ static void hmm_release(struct mmu_notifier *mn, struct mm_struct *mm)
 }
 
 static int hmm_invalidate_range_start(struct mmu_notifier *mn,
-				      struct mm_struct *mm,
-				      unsigned long start,
-				      unsigned long end,
-				      bool blockable)
+			const struct mmu_notifier_range *range)
 {
 	struct hmm_update update;
-	struct hmm *hmm = mm->hmm;
+	struct hmm *hmm = range->mm->hmm;
 
 	VM_BUG_ON(!hmm);
 
-	update.start = start;
-	update.end = end;
+	update.start = range->start;
+	update.end = range->end;
 	update.event = HMM_UPDATE_INVALIDATE;
-	update.blockable = blockable;
+	update.blockable = range->blockable;
 	return hmm_invalidate_range(hmm, true, &update);
 }
 
 static void hmm_invalidate_range_end(struct mmu_notifier *mn,
-				     struct mm_struct *mm,
-				     unsigned long start,
-				     unsigned long end)
+			const struct mmu_notifier_range *range)
 {
 	struct hmm_update update;
-	struct hmm *hmm = mm->hmm;
+	struct hmm *hmm = range->mm->hmm;
 
 	VM_BUG_ON(!hmm);
 
-	update.start = start;
-	update.end = end;
+	update.start = range->start;
+	update.end = range->end;
 	update.event = HMM_UPDATE_INVALIDATE;
 	update.blockable = true;
 	hmm_invalidate_range(hmm, false, &update);
diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
index 5119ff846769..5f6665ae3ee2 100644
--- a/mm/mmu_notifier.c
+++ b/mm/mmu_notifier.c
@@ -178,14 +178,20 @@ int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
 				  unsigned long start, unsigned long end,
 				  bool blockable)
 {
+	struct mmu_notifier_range _range, *range = &_range;
 	struct mmu_notifier *mn;
 	int ret = 0;
 	int id;
 
+	range->blockable = blockable;
+	range->start = start;
+	range->end = end;
+	range->mm = mm;
+
 	id = srcu_read_lock(&srcu);
 	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
 		if (mn->ops->invalidate_range_start) {
-			int _ret = mn->ops->invalidate_range_start(mn, mm, start, end, blockable);
+			int _ret = mn->ops->invalidate_range_start(mn, range);
 			if (_ret) {
 				pr_info("%pS callback failed with %d in %sblockable context.\n",
 						mn->ops->invalidate_range_start, _ret,
@@ -205,9 +211,20 @@ void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
 					 unsigned long end,
 					 bool only_end)
 {
+	struct mmu_notifier_range _range, *range = &_range;
 	struct mmu_notifier *mn;
 	int id;
 
+	/*
+	 * The end call back will never be call if the start refused to go
+	 * through because of blockable was false so here assume that we
+	 * can block.
+	 */
+	range->blockable = true;
+	range->start = start;
+	range->end = end;
+	range->mm = mm;
+
 	id = srcu_read_lock(&srcu);
 	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
 		/*
@@ -226,7 +243,7 @@ void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
 		if (!only_end && mn->ops->invalidate_range)
 			mn->ops->invalidate_range(mn, mm, start, end);
 		if (mn->ops->invalidate_range_end)
-			mn->ops->invalidate_range_end(mn, mm, start, end);
+			mn->ops->invalidate_range_end(mn, range);
 	}
 	srcu_read_unlock(&srcu, id);
 }
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 2679e476b6c3..f829f63f2b16 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -360,10 +360,7 @@ static void kvm_mmu_notifier_change_pte(struct mmu_notifier *mn,
 }
 
 static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
-						    struct mm_struct *mm,
-						    unsigned long start,
-						    unsigned long end,
-						    bool blockable)
+					const struct mmu_notifier_range *range)
 {
 	struct kvm *kvm = mmu_notifier_to_kvm(mn);
 	int need_tlb_flush = 0, idx;
@@ -377,7 +374,7 @@ static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
 	 * count is also read inside the mmu_lock critical section.
 	 */
 	kvm->mmu_notifier_count++;
-	need_tlb_flush = kvm_unmap_hva_range(kvm, start, end);
+	need_tlb_flush = kvm_unmap_hva_range(kvm, range->start, range->end);
 	need_tlb_flush |= kvm->tlbs_dirty;
 	/* we've to flush the tlb before the pages can be freed */
 	if (need_tlb_flush)
@@ -385,7 +382,8 @@ static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
 
 	spin_unlock(&kvm->mmu_lock);
 
-	ret = kvm_arch_mmu_notifier_invalidate_range(kvm, start, end, blockable);
+	ret = kvm_arch_mmu_notifier_invalidate_range(kvm, range->start,
+					range->end, range->blockable);
 
 	srcu_read_unlock(&kvm->srcu, idx);
 
@@ -393,9 +391,7 @@ static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
 }
 
 static void kvm_mmu_notifier_invalidate_range_end(struct mmu_notifier *mn,
-						  struct mm_struct *mm,
-						  unsigned long start,
-						  unsigned long end)
+					const struct mmu_notifier_range *range)
 {
 	struct kvm *kvm = mmu_notifier_to_kvm(mn);
 
-- 
2.17.2

Re: [PATCH v2 1/3] mm/mmu_notifier: use structure for invalidate_range_start/end callback

[Jan Kara]

On Wed 05-12-18 00:36:26, jglisse@redhat.com wrote:
> diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
> index 5119ff846769..5f6665ae3ee2 100644
> --- a/mm/mmu_notifier.c
> +++ b/mm/mmu_notifier.c
> @@ -178,14 +178,20 @@ int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
>  				  unsigned long start, unsigned long end,
>  				  bool blockable)
>  {
> +	struct mmu_notifier_range _range, *range = &_range;

Why these games with two variables?

>  	struct mmu_notifier *mn;
>  	int ret = 0;
>  	int id;
>  
> +	range->blockable = blockable;
> +	range->start = start;
> +	range->end = end;
> +	range->mm = mm;
> +

Use your init function for this?

>  	id = srcu_read_lock(&srcu);
>  	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
>  		if (mn->ops->invalidate_range_start) {
> -			int _ret = mn->ops->invalidate_range_start(mn, mm, start, end, blockable);
> +			int _ret = mn->ops->invalidate_range_start(mn, range);
>  			if (_ret) {
>  				pr_info("%pS callback failed with %d in %sblockable context.\n",
>  						mn->ops->invalidate_range_start, _ret,
> @@ -205,9 +211,20 @@ void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
>  					 unsigned long end,
>  					 bool only_end)
>  {
> +	struct mmu_notifier_range _range, *range = &_range;
>  	struct mmu_notifier *mn;
>  	int id;
>  
> +	/*
> +	 * The end call back will never be call if the start refused to go
> +	 * through because of blockable was false so here assume that we
> +	 * can block.
> +	 */
> +	range->blockable = true;
> +	range->start = start;
> +	range->end = end;
> +	range->mm = mm;
> +

The same as above.

Otherwise the patch looks good to me.

								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

Re: [PATCH v2 1/3] mm/mmu_notifier: use structure for invalidate_range_start/end callback

[Jerome Glisse]

On Wed, Dec 05, 2018 at 05:35:20PM +0100, Jan Kara wrote:
> On Wed 05-12-18 00:36:26, jglisse@redhat.com wrote:
> > diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
> > index 5119ff846769..5f6665ae3ee2 100644
> > --- a/mm/mmu_notifier.c
> > +++ b/mm/mmu_notifier.c
> > @@ -178,14 +178,20 @@ int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
> >  				  unsigned long start, unsigned long end,
> >  				  bool blockable)
> >  {
> > +	struct mmu_notifier_range _range, *range = &_range;
> 
> Why these games with two variables?

This is a temporary step i dediced to do the convertion in 2 steps,
first i convert the callback to use the structure so that people
having mmu notifier callback only have to review this patch and do
not get distracted by the second step which update all the mm call
site that trigger invalidation.

In the final result this code disappear. I did it that way to make
the thing more reviewable. Sorry if that is a bit confusing.

> 
> >  	struct mmu_notifier *mn;
> >  	int ret = 0;
> >  	int id;
> >  
> > +	range->blockable = blockable;
> > +	range->start = start;
> > +	range->end = end;
> > +	range->mm = mm;
> > +
> 
> Use your init function for this?

This get remove in the next patch, i can respawn with the init
function but this is a temporary step like explain above.

> 
> >  	id = srcu_read_lock(&srcu);
> >  	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
> >  		if (mn->ops->invalidate_range_start) {
> > -			int _ret = mn->ops->invalidate_range_start(mn, mm, start, end, blockable);
> > +			int _ret = mn->ops->invalidate_range_start(mn, range);
> >  			if (_ret) {
> >  				pr_info("%pS callback failed with %d in %sblockable context.\n",
> >  						mn->ops->invalidate_range_start, _ret,
> > @@ -205,9 +211,20 @@ void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
> >  					 unsigned long end,
> >  					 bool only_end)
> >  {
> > +	struct mmu_notifier_range _range, *range = &_range;
> >  	struct mmu_notifier *mn;
> >  	int id;
> >  
> > +	/*
> > +	 * The end call back will never be call if the start refused to go
> > +	 * through because of blockable was false so here assume that we
> > +	 * can block.
> > +	 */
> > +	range->blockable = true;
> > +	range->start = start;
> > +	range->end = end;
> > +	range->mm = mm;
> > +
> 
> The same as above.
> 
> Otherwise the patch looks good to me.

Thank you for reviewing.

Cheers,
Jérôme

Re: [PATCH v2 1/3] mm/mmu_notifier: use structure for invalidate_range_start/end callback

[Jan Kara]

On Wed 05-12-18 11:40:52, Jerome Glisse wrote:
> On Wed, Dec 05, 2018 at 05:35:20PM +0100, Jan Kara wrote:
> > On Wed 05-12-18 00:36:26, jglisse@redhat.com wrote:
> > > diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
> > > index 5119ff846769..5f6665ae3ee2 100644
> > > --- a/mm/mmu_notifier.c
> > > +++ b/mm/mmu_notifier.c
> > > @@ -178,14 +178,20 @@ int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
> > >  				  unsigned long start, unsigned long end,
> > >  				  bool blockable)
> > >  {
> > > +	struct mmu_notifier_range _range, *range = &_range;
> > 
> > Why these games with two variables?
> 
> This is a temporary step i dediced to do the convertion in 2 steps,
> first i convert the callback to use the structure so that people
> having mmu notifier callback only have to review this patch and do
> not get distracted by the second step which update all the mm call
> site that trigger invalidation.
> 
> In the final result this code disappear. I did it that way to make
> the thing more reviewable. Sorry if that is a bit confusing.

Aha, right. Thanks for clarification. You can add:

Acked-by: Jan Kara <jack@suse.cz>

								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

Re: [PATCH v2 1/3] mm/mmu_notifier: use structure for invalidate_range_start/end callback

[Kuehling, Felix]

The amdgpu part looks good to me.

A minor nit-pick in mmu_notifier.c (inline).

Either way, the series is Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>

On 2018-12-05 12:36 a.m., jglisse@redhat.com wrote:
> From: Jérôme Glisse <jglisse@redhat.com>
>
> To avoid having to change many callback definition everytime we want
> to add a parameter use a structure to group all parameters for the
> mmu_notifier invalidate_range_start/end callback. No functional changes
> with this patch.
>
> Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Matthew Wilcox <mawilcox@microsoft.com>
> Cc: Ross Zwisler <zwisler@kernel.org>
> Cc: Jan Kara <jack@suse.cz>
> Cc: Dan Williams <dan.j.williams@intel.com>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Radim Krčmář <rkrcmar@redhat.com>
> Cc: Michal Hocko <mhocko@kernel.org>
> Cc: Christian Koenig <christian.koenig@amd.com>
> Cc: Felix Kuehling <felix.kuehling@amd.com>
> Cc: Ralph Campbell <rcampbell@nvidia.com>
> Cc: John Hubbard <jhubbard@nvidia.com>
> Cc: kvm@vger.kernel.org
> Cc: dri-devel@lists.freedesktop.org
> Cc: linux-rdma@vger.kernel.org
> Cc: linux-fsdevel@vger.kernel.org
> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c  | 43 +++++++++++--------------
>  drivers/gpu/drm/i915/i915_gem_userptr.c | 14 ++++----
>  drivers/gpu/drm/radeon/radeon_mn.c      | 16 ++++-----
>  drivers/infiniband/core/umem_odp.c      | 20 +++++-------
>  drivers/infiniband/hw/hfi1/mmu_rb.c     | 13 +++-----
>  drivers/misc/mic/scif/scif_dma.c        | 11 ++-----
>  drivers/misc/sgi-gru/grutlbpurge.c      | 14 ++++----
>  drivers/xen/gntdev.c                    | 12 +++----
>  include/linux/mmu_notifier.h            | 14 +++++---
>  mm/hmm.c                                | 23 ++++++-------
>  mm/mmu_notifier.c                       | 21 ++++++++++--
>  virt/kvm/kvm_main.c                     | 14 +++-----
>  12 files changed, 102 insertions(+), 113 deletions(-)
>
[snip]
> diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
> index 5119ff846769..5f6665ae3ee2 100644
> --- a/mm/mmu_notifier.c
> +++ b/mm/mmu_notifier.c
> @@ -178,14 +178,20 @@ int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
>  				  unsigned long start, unsigned long end,
>  				  bool blockable)
>  {
> +	struct mmu_notifier_range _range, *range = &_range;

I'm not sure why you need to access _range indirectly through a pointer.
See below.


>  	struct mmu_notifier *mn;
>  	int ret = 0;
>  	int id;
>  
> +	range->blockable = blockable;
> +	range->start = start;
> +	range->end = end;
> +	range->mm = mm;

This could just assign _range.blockable, _range.start, etc. without the
indirection. Or you could even use an initializer instead:

struct mmu_notifier_range range = {
    .blockable = blockable,
    .start = start,
    ...
};


> +
>  	id = srcu_read_lock(&srcu);
>  	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
>  		if (mn->ops->invalidate_range_start) {
> -			int _ret = mn->ops->invalidate_range_start(mn, mm, start, end, blockable);
> +			int _ret = mn->ops->invalidate_range_start(mn, range);

This could just use &_range without the indirection.

Same in ..._invalidate_range_end below.

Regards,
  Felix


>  			if (_ret) {
>  				pr_info("%pS callback failed with %d in %sblockable context.\n",
>  						mn->ops->invalidate_range_start, _ret,
> @@ -205,9 +211,20 @@ void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
>  					 unsigned long end,
>  					 bool only_end)
>  {
> +	struct mmu_notifier_range _range, *range = &_range;
>  	struct mmu_notifier *mn;
>  	int id;
>  
> +	/*
> +	 * The end call back will never be call if the start refused to go
> +	 * through because of blockable was false so here assume that we
> +	 * can block.
> +	 */
> +	range->blockable = true;
> +	range->start = start;
> +	range->end = end;
> +	range->mm = mm;
> +
>  	id = srcu_read_lock(&srcu);
>  	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
>  		/*
> @@ -226,7 +243,7 @@ void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
>  		if (!only_end && mn->ops->invalidate_range)
>  			mn->ops->invalidate_range(mn, mm, start, end);
>  		if (mn->ops->invalidate_range_end)
> -			mn->ops->invalidate_range_end(mn, mm, start, end);
> +			mn->ops->invalidate_range_end(mn, range);
>  	}
>  	srcu_read_unlock(&srcu, id);
>  }
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index 2679e476b6c3..f829f63f2b16 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -360,10 +360,7 @@ static void kvm_mmu_notifier_change_pte(struct mmu_notifier *mn,
>  }
>  
>  static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
> -						    struct mm_struct *mm,
> -						    unsigned long start,
> -						    unsigned long end,
> -						    bool blockable)
> +					const struct mmu_notifier_range *range)
>  {
>  	struct kvm *kvm = mmu_notifier_to_kvm(mn);
>  	int need_tlb_flush = 0, idx;
> @@ -377,7 +374,7 @@ static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
>  	 * count is also read inside the mmu_lock critical section.
>  	 */
>  	kvm->mmu_notifier_count++;
> -	need_tlb_flush = kvm_unmap_hva_range(kvm, start, end);
> +	need_tlb_flush = kvm_unmap_hva_range(kvm, range->start, range->end);
>  	need_tlb_flush |= kvm->tlbs_dirty;
>  	/* we've to flush the tlb before the pages can be freed */
>  	if (need_tlb_flush)
> @@ -385,7 +382,8 @@ static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
>  
>  	spin_unlock(&kvm->mmu_lock);
>  
> -	ret = kvm_arch_mmu_notifier_invalidate_range(kvm, start, end, blockable);
> +	ret = kvm_arch_mmu_notifier_invalidate_range(kvm, range->start,
> +					range->end, range->blockable);
>  
>  	srcu_read_unlock(&kvm->srcu, idx);
>  
> @@ -393,9 +391,7 @@ static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn,
>  }
>  
>  static void kvm_mmu_notifier_invalidate_range_end(struct mmu_notifier *mn,
> -						  struct mm_struct *mm,
> -						  unsigned long start,
> -						  unsigned long end)
> +					const struct mmu_notifier_range *range)
>  {
>  	struct kvm *kvm = mmu_notifier_to_kvm(mn);
>  

Re: [PATCH v2 1/3] mm/mmu_notifier: use structure for invalidate_range_start/end callback

[Jerome Glisse]

On Wed, Dec 05, 2018 at 09:42:45PM +0000, Kuehling, Felix wrote:
> The amdgpu part looks good to me.
> 
> A minor nit-pick in mmu_notifier.c (inline).
> 
> Either way, the series is Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
> 
> On 2018-12-05 12:36 a.m., jglisse@redhat.com wrote:
> > From: Jérôme Glisse <jglisse@redhat.com>
> >
> > To avoid having to change many callback definition everytime we want
> > to add a parameter use a structure to group all parameters for the
> > mmu_notifier invalidate_range_start/end callback. No functional changes
> > with this patch.
> >
> > Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
> > Cc: Andrew Morton <akpm@linux-foundation.org>
> > Cc: Matthew Wilcox <mawilcox@microsoft.com>
> > Cc: Ross Zwisler <zwisler@kernel.org>
> > Cc: Jan Kara <jack@suse.cz>
> > Cc: Dan Williams <dan.j.williams@intel.com>
> > Cc: Paolo Bonzini <pbonzini@redhat.com>
> > Cc: Radim Krčmář <rkrcmar@redhat.com>
> > Cc: Michal Hocko <mhocko@kernel.org>
> > Cc: Christian Koenig <christian.koenig@amd.com>
> > Cc: Felix Kuehling <felix.kuehling@amd.com>
> > Cc: Ralph Campbell <rcampbell@nvidia.com>
> > Cc: John Hubbard <jhubbard@nvidia.com>
> > Cc: kvm@vger.kernel.org
> > Cc: dri-devel@lists.freedesktop.org
> > Cc: linux-rdma@vger.kernel.org
> > Cc: linux-fsdevel@vger.kernel.org
> > ---
> >  drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c  | 43 +++++++++++--------------
> >  drivers/gpu/drm/i915/i915_gem_userptr.c | 14 ++++----
> >  drivers/gpu/drm/radeon/radeon_mn.c      | 16 ++++-----
> >  drivers/infiniband/core/umem_odp.c      | 20 +++++-------
> >  drivers/infiniband/hw/hfi1/mmu_rb.c     | 13 +++-----
> >  drivers/misc/mic/scif/scif_dma.c        | 11 ++-----
> >  drivers/misc/sgi-gru/grutlbpurge.c      | 14 ++++----
> >  drivers/xen/gntdev.c                    | 12 +++----
> >  include/linux/mmu_notifier.h            | 14 +++++---
> >  mm/hmm.c                                | 23 ++++++-------
> >  mm/mmu_notifier.c                       | 21 ++++++++++--
> >  virt/kvm/kvm_main.c                     | 14 +++-----
> >  12 files changed, 102 insertions(+), 113 deletions(-)
> >
> [snip]
> > diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
> > index 5119ff846769..5f6665ae3ee2 100644
> > --- a/mm/mmu_notifier.c
> > +++ b/mm/mmu_notifier.c
> > @@ -178,14 +178,20 @@ int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
> >  				  unsigned long start, unsigned long end,
> >  				  bool blockable)
> >  {
> > +	struct mmu_notifier_range _range, *range = &_range;
> 
> I'm not sure why you need to access _range indirectly through a pointer.
> See below.
> 
> 
> >  	struct mmu_notifier *mn;
> >  	int ret = 0;
> >  	int id;
> >  
> > +	range->blockable = blockable;
> > +	range->start = start;
> > +	range->end = end;
> > +	range->mm = mm;
> 
> This could just assign _range.blockable, _range.start, etc. without the
> indirection. Or you could even use an initializer instead:
> 
> struct mmu_notifier_range range = {
>     .blockable = blockable,
>     .start = start,
>     ...
> };
> 
> 
> > +
> >  	id = srcu_read_lock(&srcu);
> >  	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
> >  		if (mn->ops->invalidate_range_start) {
> > -			int _ret = mn->ops->invalidate_range_start(mn, mm, start, end, blockable);
> > +			int _ret = mn->ops->invalidate_range_start(mn, range);
> 
> This could just use &_range without the indirection.
> 
> Same in ..._invalidate_range_end below.

So explaination is that this is a temporary step all this code is
remove in the second patch. It was done this way in this patch to
minimize the diff within the next patch.

I did this because i wanted to do the convertion in 2 steps the
first step i convert all the listener of mmu notifier and in the
second step i convert all the call site that trigger a mmu notifer.

I did that to help people reviewing only the part they care about.

Apparently it end up confusing people more than it helped :)

Do people have strong feeling about getting this code that is
deleted in the second patch fix in the first patch anyway ?

I can respin if so but i don't see much value in formating code
that is deleted in the serie.

Thank you for reviewing

Cheers,
Jérôme

Re: [PATCH v2 1/3] mm/mmu_notifier: use structure for invalidate_range_start/end callback

[Kuehling, Felix]

On 2018-12-05 6:04 p.m., Jerome Glisse wrote:
> On Wed, Dec 05, 2018 at 09:42:45PM +0000, Kuehling, Felix wrote:
>> The amdgpu part looks good to me.
>>
>> A minor nit-pick in mmu_notifier.c (inline).
>>
>> Either way, the series is Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
>>
>> On 2018-12-05 12:36 a.m., jglisse@redhat.com wrote:
>>> From: Jérôme Glisse <jglisse@redhat.com>
>>>
>>> To avoid having to change many callback definition everytime we want
>>> to add a parameter use a structure to group all parameters for the
>>> mmu_notifier invalidate_range_start/end callback. No functional changes
>>> with this patch.
>>>
>>> Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
>>> Cc: Andrew Morton <akpm@linux-foundation.org>
>>> Cc: Matthew Wilcox <mawilcox@microsoft.com>
>>> Cc: Ross Zwisler <zwisler@kernel.org>
>>> Cc: Jan Kara <jack@suse.cz>
>>> Cc: Dan Williams <dan.j.williams@intel.com>
>>> Cc: Paolo Bonzini <pbonzini@redhat.com>
>>> Cc: Radim Krčmář <rkrcmar@redhat.com>
>>> Cc: Michal Hocko <mhocko@kernel.org>
>>> Cc: Christian Koenig <christian.koenig@amd.com>
>>> Cc: Felix Kuehling <felix.kuehling@amd.com>
>>> Cc: Ralph Campbell <rcampbell@nvidia.com>
>>> Cc: John Hubbard <jhubbard@nvidia.com>
>>> Cc: kvm@vger.kernel.org
>>> Cc: dri-devel@lists.freedesktop.org
>>> Cc: linux-rdma@vger.kernel.org
>>> Cc: linux-fsdevel@vger.kernel.org
>>> ---
>>>  drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c  | 43 +++++++++++--------------
>>>  drivers/gpu/drm/i915/i915_gem_userptr.c | 14 ++++----
>>>  drivers/gpu/drm/radeon/radeon_mn.c      | 16 ++++-----
>>>  drivers/infiniband/core/umem_odp.c      | 20 +++++-------
>>>  drivers/infiniband/hw/hfi1/mmu_rb.c     | 13 +++-----
>>>  drivers/misc/mic/scif/scif_dma.c        | 11 ++-----
>>>  drivers/misc/sgi-gru/grutlbpurge.c      | 14 ++++----
>>>  drivers/xen/gntdev.c                    | 12 +++----
>>>  include/linux/mmu_notifier.h            | 14 +++++---
>>>  mm/hmm.c                                | 23 ++++++-------
>>>  mm/mmu_notifier.c                       | 21 ++++++++++--
>>>  virt/kvm/kvm_main.c                     | 14 +++-----
>>>  12 files changed, 102 insertions(+), 113 deletions(-)
>>>
>> [snip]
>>> diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
>>> index 5119ff846769..5f6665ae3ee2 100644
>>> --- a/mm/mmu_notifier.c
>>> +++ b/mm/mmu_notifier.c
>>> @@ -178,14 +178,20 @@ int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
>>>  				  unsigned long start, unsigned long end,
>>>  				  bool blockable)
>>>  {
>>> +	struct mmu_notifier_range _range, *range = &_range;
>> I'm not sure why you need to access _range indirectly through a pointer.
>> See below.
>>
>>
>>>  	struct mmu_notifier *mn;
>>>  	int ret = 0;
>>>  	int id;
>>>  
>>> +	range->blockable = blockable;
>>> +	range->start = start;
>>> +	range->end = end;
>>> +	range->mm = mm;
>> This could just assign _range.blockable, _range.start, etc. without the
>> indirection. Or you could even use an initializer instead:
>>
>> struct mmu_notifier_range range = {
>>     .blockable = blockable,
>>     .start = start,
>>     ...
>> };
>>
>>
>>> +
>>>  	id = srcu_read_lock(&srcu);
>>>  	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
>>>  		if (mn->ops->invalidate_range_start) {
>>> -			int _ret = mn->ops->invalidate_range_start(mn, mm, start, end, blockable);
>>> +			int _ret = mn->ops->invalidate_range_start(mn, range);
>> This could just use &_range without the indirection.
>>
>> Same in ..._invalidate_range_end below.
> So explaination is that this is a temporary step all this code is
> remove in the second patch. It was done this way in this patch to
> minimize the diff within the next patch.

I was actually looking for that because I suspected that this would make
more sense in the context of the other patches. But then I missed the
mmu_notifier.c change in patch 2 in the noise of all the other mm
changes. Never mind.


> I did this because i wanted to do the convertion in 2 steps the
> first step i convert all the listener of mmu notifier and in the
> second step i convert all the call site that trigger a mmu notifer.

That part makes sense and I appreciate that it keeps the patches
reasonably separate for different audiences.

Thanks,
  Felix


> I did that to help people reviewing only the part they care about.
>
> Apparently it end up confusing people more than it helped :)
>
> Do people have strong feeling about getting this code that is
> deleted in the second patch fix in the first patch anyway ?
>
> I can respin if so but i don't see much value in formating code
> that is deleted in the serie.
>
> Thank you for reviewing
>
> Cheers,
> Jérôme

Re: [PATCH v2 1/3] mm/mmu_notifier: use structure for invalidate_range_start/end callback

[Jason Gunthorpe]

On Wed, Dec 05, 2018 at 12:36:26AM -0500, jglisse@redhat.com wrote:
> From: Jérôme Glisse <jglisse@redhat.com>
> 
> To avoid having to change many callback definition everytime we want
> to add a parameter use a structure to group all parameters for the
> mmu_notifier invalidate_range_start/end callback. No functional changes
> with this patch.
> 
> Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Matthew Wilcox <mawilcox@microsoft.com>
> Cc: Ross Zwisler <zwisler@kernel.org>
> Cc: Jan Kara <jack@suse.cz>
> Cc: Dan Williams <dan.j.williams@intel.com>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Radim Krčmář <rkrcmar@redhat.com>
> Cc: Michal Hocko <mhocko@kernel.org>
> Cc: Christian Koenig <christian.koenig@amd.com>
> Cc: Felix Kuehling <felix.kuehling@amd.com>
> Cc: Ralph Campbell <rcampbell@nvidia.com>
> Cc: John Hubbard <jhubbard@nvidia.com>
> Cc: kvm@vger.kernel.org
> Cc: dri-devel@lists.freedesktop.org
> Cc: linux-rdma@vger.kernel.org
> Cc: linux-fsdevel@vger.kernel.org
> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c  | 43 +++++++++++--------------
>  drivers/gpu/drm/i915/i915_gem_userptr.c | 14 ++++----
>  drivers/gpu/drm/radeon/radeon_mn.c      | 16 ++++-----
>  drivers/infiniband/core/umem_odp.c      | 20 +++++-------
>  drivers/infiniband/hw/hfi1/mmu_rb.c     | 13 +++-----
>  drivers/misc/mic/scif/scif_dma.c        | 11 ++-----
>  drivers/misc/sgi-gru/grutlbpurge.c      | 14 ++++----
>  drivers/xen/gntdev.c                    | 12 +++----
>  include/linux/mmu_notifier.h            | 14 +++++---
>  mm/hmm.c                                | 23 ++++++-------
>  mm/mmu_notifier.c                       | 21 ++++++++++--
>  virt/kvm/kvm_main.c                     | 14 +++-----
>  12 files changed, 102 insertions(+), 113 deletions(-)

The changes to drivers/infiniband look mechanical and fine to me.

It even looks like this avoids merge conflicts with the other changes
to these files :)

For infiniband:

Acked-by: Jason Gunthorpe <jgg@mellanox.com>

I assume this will go through the mm tree?

Jason

Re: [PATCH v2 1/3] mm/mmu_notifier: use structure for invalidate_range_start/end callback

[Jerome Glisse]

On Thu, Dec 06, 2018 at 08:30:27PM -0700, Jason Gunthorpe wrote:
> On Wed, Dec 05, 2018 at 12:36:26AM -0500, jglisse@redhat.com wrote:
> > From: Jérôme Glisse <jglisse@redhat.com>
> > 
> > To avoid having to change many callback definition everytime we want
> > to add a parameter use a structure to group all parameters for the
> > mmu_notifier invalidate_range_start/end callback. No functional changes
> > with this patch.
> > 
> > Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
> > Cc: Andrew Morton <akpm@linux-foundation.org>
> > Cc: Matthew Wilcox <mawilcox@microsoft.com>
> > Cc: Ross Zwisler <zwisler@kernel.org>
> > Cc: Jan Kara <jack@suse.cz>
> > Cc: Dan Williams <dan.j.williams@intel.com>
> > Cc: Paolo Bonzini <pbonzini@redhat.com>
> > Cc: Radim Krčmář <rkrcmar@redhat.com>
> > Cc: Michal Hocko <mhocko@kernel.org>
> > Cc: Christian Koenig <christian.koenig@amd.com>
> > Cc: Felix Kuehling <felix.kuehling@amd.com>
> > Cc: Ralph Campbell <rcampbell@nvidia.com>
> > Cc: John Hubbard <jhubbard@nvidia.com>
> > Cc: kvm@vger.kernel.org
> > Cc: dri-devel@lists.freedesktop.org
> > Cc: linux-rdma@vger.kernel.org
> > Cc: linux-fsdevel@vger.kernel.org
> > ---
> >  drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c  | 43 +++++++++++--------------
> >  drivers/gpu/drm/i915/i915_gem_userptr.c | 14 ++++----
> >  drivers/gpu/drm/radeon/radeon_mn.c      | 16 ++++-----
> >  drivers/infiniband/core/umem_odp.c      | 20 +++++-------
> >  drivers/infiniband/hw/hfi1/mmu_rb.c     | 13 +++-----
> >  drivers/misc/mic/scif/scif_dma.c        | 11 ++-----
> >  drivers/misc/sgi-gru/grutlbpurge.c      | 14 ++++----
> >  drivers/xen/gntdev.c                    | 12 +++----
> >  include/linux/mmu_notifier.h            | 14 +++++---
> >  mm/hmm.c                                | 23 ++++++-------
> >  mm/mmu_notifier.c                       | 21 ++++++++++--
> >  virt/kvm/kvm_main.c                     | 14 +++-----
> >  12 files changed, 102 insertions(+), 113 deletions(-)
> 
> The changes to drivers/infiniband look mechanical and fine to me.
> 
> It even looks like this avoids merge conflicts with the other changes
> to these files :)
> 
> For infiniband:
> 
> Acked-by: Jason Gunthorpe <jgg@mellanox.com>
> 
> I assume this will go through the mm tree?

Yes this is my exceptation as in the ends it touch more mm
stuff than anything else. Andrew already added v1 to its
patchset.

Cheers,
Jérôme

[PATCH v2 2/3] mm/mmu_notifier: use structure for invalidate_range_start/end calls v2

[jglisse]

From: Jérôme Glisse <jglisse@redhat.com>

To avoid having to change many call sites everytime we want to add a
parameter use a structure to group all parameters for the mmu_notifier
invalidate_range_start/end cakks. No functional changes with this
patch.

Changes since v1:
    - introduce mmu_notifier_range_init() as an helper to initialize
      the range structure allowing to optimize out the case when mmu
      notifier is not enabled
    - fix mm/migrate.c migrate_vma_collect()

Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Acked-by: Christian König <christian.koenig@amd.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Felix Kuehling <felix.kuehling@amd.com>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: kvm@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org
Cc: linux-rdma@vger.kernel.org
Cc: linux-fsdevel@vger.kernel.org
---
 fs/dax.c                     |  8 +--
 fs/proc/task_mmu.c           |  7 ++-
 include/linux/mm.h           |  4 +-
 include/linux/mmu_notifier.h | 87 +++++++++++++++++++++-----------
 kernel/events/uprobes.c      | 10 ++--
 mm/huge_memory.c             | 54 ++++++++++----------
 mm/hugetlb.c                 | 52 ++++++++++---------
 mm/khugepaged.c              | 10 ++--
 mm/ksm.c                     | 21 ++++----
 mm/madvise.c                 | 21 ++++----
 mm/memory.c                  | 97 ++++++++++++++++++------------------
 mm/migrate.c                 | 25 +++++-----
 mm/mmu_notifier.c            | 35 +++----------
 mm/mprotect.c                | 15 +++---
 mm/mremap.c                  | 10 ++--
 mm/oom_kill.c                | 17 ++++---
 mm/rmap.c                    | 30 ++++++-----
 17 files changed, 258 insertions(+), 245 deletions(-)

diff --git a/fs/dax.c b/fs/dax.c
index 9bcce89ea18e..874085bacaf5 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -758,7 +758,8 @@ static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index,
 
 	i_mmap_lock_read(mapping);
 	vma_interval_tree_foreach(vma, &mapping->i_mmap, index, index) {
-		unsigned long address, start, end;
+		struct mmu_notifier_range range;
+		unsigned long address;
 
 		cond_resched();
 
@@ -772,7 +773,8 @@ static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index,
 		 * call mmu_notifier_invalidate_range_start() on our behalf
 		 * before taking any lock.
 		 */
-		if (follow_pte_pmd(vma->vm_mm, address, &start, &end, &ptep, &pmdp, &ptl))
+		if (follow_pte_pmd(vma->vm_mm, address, &range,
+				   &ptep, &pmdp, &ptl))
 			continue;
 
 		/*
@@ -814,7 +816,7 @@ static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index,
 			pte_unmap_unlock(ptep, ptl);
 		}
 
-		mmu_notifier_invalidate_range_end(vma->vm_mm, start, end);
+		mmu_notifier_invalidate_range_end(&range);
 	}
 	i_mmap_unlock_read(mapping);
 }
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 47c3764c469b..b3ddceb003bc 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -1096,6 +1096,7 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf,
 		return -ESRCH;
 	mm = get_task_mm(task);
 	if (mm) {
+		struct mmu_notifier_range range;
 		struct clear_refs_private cp = {
 			.type = type,
 		};
@@ -1139,11 +1140,13 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf,
 				downgrade_write(&mm->mmap_sem);
 				break;
 			}
-			mmu_notifier_invalidate_range_start(mm, 0, -1);
+
+			mmu_notifier_range_init(&range, mm, 0, -1UL);
+			mmu_notifier_invalidate_range_start(&range);
 		}
 		walk_page_range(0, mm->highest_vm_end, &clear_refs_walk);
 		if (type == CLEAR_REFS_SOFT_DIRTY)
-			mmu_notifier_invalidate_range_end(mm, 0, -1);
+			mmu_notifier_invalidate_range_end(&range);
 		tlb_finish_mmu(&tlb, 0, -1);
 		up_read(&mm->mmap_sem);
 out_mm:
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 5411de93a363..e7b6f2b30713 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1397,6 +1397,8 @@ struct mm_walk {
 	void *private;
 };
 
+struct mmu_notifier_range;
+
 int walk_page_range(unsigned long addr, unsigned long end,
 		struct mm_walk *walk);
 int walk_page_vma(struct vm_area_struct *vma, struct mm_walk *walk);
@@ -1405,7 +1407,7 @@ void free_pgd_range(struct mmu_gather *tlb, unsigned long addr,
 int copy_page_range(struct mm_struct *dst, struct mm_struct *src,
 			struct vm_area_struct *vma);
 int follow_pte_pmd(struct mm_struct *mm, unsigned long address,
-			     unsigned long *start, unsigned long *end,
+		 	     struct mmu_notifier_range *range,
 			     pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp);
 int follow_pfn(struct vm_area_struct *vma, unsigned long address,
 	unsigned long *pfn);
diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h
index 368f0c1a049d..39b06772427f 100644
--- a/include/linux/mmu_notifier.h
+++ b/include/linux/mmu_notifier.h
@@ -220,11 +220,8 @@ extern int __mmu_notifier_test_young(struct mm_struct *mm,
 				     unsigned long address);
 extern void __mmu_notifier_change_pte(struct mm_struct *mm,
 				      unsigned long address, pte_t pte);
-extern int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
-				  unsigned long start, unsigned long end,
-				  bool blockable);
-extern void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
-				  unsigned long start, unsigned long end,
+extern int __mmu_notifier_invalidate_range_start(struct mmu_notifier_range *);
+extern void __mmu_notifier_invalidate_range_end(struct mmu_notifier_range *r,
 				  bool only_end);
 extern void __mmu_notifier_invalidate_range(struct mm_struct *mm,
 				  unsigned long start, unsigned long end);
@@ -268,33 +265,37 @@ static inline void mmu_notifier_change_pte(struct mm_struct *mm,
 		__mmu_notifier_change_pte(mm, address, pte);
 }
 
-static inline void mmu_notifier_invalidate_range_start(struct mm_struct *mm,
-				  unsigned long start, unsigned long end)
+static inline void
+mmu_notifier_invalidate_range_start(struct mmu_notifier_range *range)
 {
-	if (mm_has_notifiers(mm))
-		__mmu_notifier_invalidate_range_start(mm, start, end, true);
+	if (mm_has_notifiers(range->mm)) {
+		range->blockable = true;
+		__mmu_notifier_invalidate_range_start(range);
+	}
 }
 
-static inline int mmu_notifier_invalidate_range_start_nonblock(struct mm_struct *mm,
-				  unsigned long start, unsigned long end)
+static inline int
+mmu_notifier_invalidate_range_start_nonblock(struct mmu_notifier_range *range)
 {
-	if (mm_has_notifiers(mm))
-		return __mmu_notifier_invalidate_range_start(mm, start, end, false);
+	if (mm_has_notifiers(range->mm)) {
+		range->blockable = false;
+		return __mmu_notifier_invalidate_range_start(range);
+	}
 	return 0;
 }
 
-static inline void mmu_notifier_invalidate_range_end(struct mm_struct *mm,
-				  unsigned long start, unsigned long end)
+static inline void
+mmu_notifier_invalidate_range_end(struct mmu_notifier_range *range)
 {
-	if (mm_has_notifiers(mm))
-		__mmu_notifier_invalidate_range_end(mm, start, end, false);
+	if (mm_has_notifiers(range->mm))
+		__mmu_notifier_invalidate_range_end(range, false);
 }
 
-static inline void mmu_notifier_invalidate_range_only_end(struct mm_struct *mm,
-				  unsigned long start, unsigned long end)
+static inline void
+mmu_notifier_invalidate_range_only_end(struct mmu_notifier_range *range)
 {
-	if (mm_has_notifiers(mm))
-		__mmu_notifier_invalidate_range_end(mm, start, end, true);
+	if (mm_has_notifiers(range->mm))
+		__mmu_notifier_invalidate_range_end(range, true);
 }
 
 static inline void mmu_notifier_invalidate_range(struct mm_struct *mm,
@@ -315,6 +316,17 @@ static inline void mmu_notifier_mm_destroy(struct mm_struct *mm)
 		__mmu_notifier_mm_destroy(mm);
 }
 
+
+static inline void mmu_notifier_range_init(struct mmu_notifier_range *range,
+					   struct mm_struct *mm,
+					   unsigned long start,
+					   unsigned long end)
+{
+	range->mm = mm;
+	range->start = start;
+	range->end = end;
+}
+
 #define ptep_clear_flush_young_notify(__vma, __address, __ptep)		\
 ({									\
 	int __young;							\
@@ -428,6 +440,23 @@ extern void mmu_notifier_synchronize(void);
 
 #else /* CONFIG_MMU_NOTIFIER */
 
+struct mmu_notifier_range {
+	unsigned long start;
+	unsigned long end;
+};
+
+static inline void _mmu_notifier_range_init(struct mmu_notifier_range *range,
+					    unsigned long start,
+					    unsigned long end)
+{
+	range->start = start;
+	range->end = end;
+}
+
+#define mmu_notifier_range_init(range, mm, start, end) \
+	_mmu_notifier_range_init(range, start, end)
+
+
 static inline int mm_has_notifiers(struct mm_struct *mm)
 {
 	return 0;
@@ -455,24 +484,24 @@ static inline void mmu_notifier_change_pte(struct mm_struct *mm,
 {
 }
 
-static inline void mmu_notifier_invalidate_range_start(struct mm_struct *mm,
-				  unsigned long start, unsigned long end)
+static inline void
+mmu_notifier_invalidate_range_start(struct mmu_notifier_range *range)
 {
 }
 
-static inline int mmu_notifier_invalidate_range_start_nonblock(struct mm_struct *mm,
-				  unsigned long start, unsigned long end)
+static inline int
+mmu_notifier_invalidate_range_start_nonblock(struct mmu_notifier_range *range)
 {
 	return 0;
 }
 
-static inline void mmu_notifier_invalidate_range_end(struct mm_struct *mm,
-				  unsigned long start, unsigned long end)
+static inline
+void mmu_notifier_invalidate_range_end(struct mmu_notifier_range *range)
 {
 }
 
-static inline void mmu_notifier_invalidate_range_only_end(struct mm_struct *mm,
-				  unsigned long start, unsigned long end)
+static inline void
+mmu_notifier_invalidate_range_only_end(struct mmu_notifier_range *range)
 {
 }
 
diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
index 322e97bbb437..1fc8a93709c3 100644
--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -171,11 +171,11 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr,
 		.address = addr,
 	};
 	int err;
-	/* For mmu_notifiers */
-	const unsigned long mmun_start = addr;
-	const unsigned long mmun_end   = addr + PAGE_SIZE;
+	struct mmu_notifier_range range;
 	struct mem_cgroup *memcg;
 
+	mmu_notifier_range_init(&range, mm, addr, addr + PAGE_SIZE);
+
 	VM_BUG_ON_PAGE(PageTransHuge(old_page), old_page);
 
 	err = mem_cgroup_try_charge(new_page, vma->vm_mm, GFP_KERNEL, &memcg,
@@ -186,7 +186,7 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr,
 	/* For try_to_free_swap() and munlock_vma_page() below */
 	lock_page(old_page);
 
-	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
+	mmu_notifier_invalidate_range_start(&range);
 	err = -EAGAIN;
 	if (!page_vma_mapped_walk(&pvmw)) {
 		mem_cgroup_cancel_charge(new_page, memcg, false);
@@ -220,7 +220,7 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr,
 
 	err = 0;
  unlock:
-	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
+	mmu_notifier_invalidate_range_end(&range);
 	unlock_page(old_page);
 	return err;
 }
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 622cced74fd9..c1d3ce809416 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1144,8 +1144,7 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
 	int i;
 	vm_fault_t ret = 0;
 	struct page **pages;
-	unsigned long mmun_start;	/* For mmu_notifiers */
-	unsigned long mmun_end;		/* For mmu_notifiers */
+	struct mmu_notifier_range range;
 
 	pages = kmalloc_array(HPAGE_PMD_NR, sizeof(struct page *),
 			      GFP_KERNEL);
@@ -1183,9 +1182,9 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
 		cond_resched();
 	}
 
-	mmun_start = haddr;
-	mmun_end   = haddr + HPAGE_PMD_SIZE;
-	mmu_notifier_invalidate_range_start(vma->vm_mm, mmun_start, mmun_end);
+	mmu_notifier_range_init(&range, vma->vm_mm, haddr,
+				haddr + HPAGE_PMD_SIZE);
+	mmu_notifier_invalidate_range_start(&range);
 
 	vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd);
 	if (unlikely(!pmd_same(*vmf->pmd, orig_pmd)))
@@ -1230,8 +1229,7 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
 	 * No need to double call mmu_notifier->invalidate_range() callback as
 	 * the above pmdp_huge_clear_flush_notify() did already call it.
 	 */
-	mmu_notifier_invalidate_range_only_end(vma->vm_mm, mmun_start,
-						mmun_end);
+	mmu_notifier_invalidate_range_only_end(&range);
 
 	ret |= VM_FAULT_WRITE;
 	put_page(page);
@@ -1241,7 +1239,7 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
 
 out_free_pages:
 	spin_unlock(vmf->ptl);
-	mmu_notifier_invalidate_range_end(vma->vm_mm, mmun_start, mmun_end);
+	mmu_notifier_invalidate_range_end(&range);
 	for (i = 0; i < HPAGE_PMD_NR; i++) {
 		memcg = (void *)page_private(pages[i]);
 		set_page_private(pages[i], 0);
@@ -1258,8 +1256,7 @@ vm_fault_t do_huge_pmd_wp_page(struct vm_fault *vmf, pmd_t orig_pmd)
 	struct page *page = NULL, *new_page;
 	struct mem_cgroup *memcg;
 	unsigned long haddr = vmf->address & HPAGE_PMD_MASK;
-	unsigned long mmun_start;	/* For mmu_notifiers */
-	unsigned long mmun_end;		/* For mmu_notifiers */
+	struct mmu_notifier_range range;
 	gfp_t huge_gfp;			/* for allocation and charge */
 	vm_fault_t ret = 0;
 
@@ -1349,9 +1346,9 @@ vm_fault_t do_huge_pmd_wp_page(struct vm_fault *vmf, pmd_t orig_pmd)
 				    vma, HPAGE_PMD_NR);
 	__SetPageUptodate(new_page);
 
-	mmun_start = haddr;
-	mmun_end   = haddr + HPAGE_PMD_SIZE;
-	mmu_notifier_invalidate_range_start(vma->vm_mm, mmun_start, mmun_end);
+	mmu_notifier_range_init(&range, vma->vm_mm, haddr,
+				haddr + HPAGE_PMD_SIZE);
+	mmu_notifier_invalidate_range_start(&range);
 
 	spin_lock(vmf->ptl);
 	if (page)
@@ -1386,8 +1383,7 @@ vm_fault_t do_huge_pmd_wp_page(struct vm_fault *vmf, pmd_t orig_pmd)
 	 * No need to double call mmu_notifier->invalidate_range() callback as
 	 * the above pmdp_huge_clear_flush_notify() did already call it.
 	 */
-	mmu_notifier_invalidate_range_only_end(vma->vm_mm, mmun_start,
-					       mmun_end);
+	mmu_notifier_invalidate_range_only_end(&range);
 out:
 	return ret;
 out_unlock:
@@ -2028,14 +2024,15 @@ void __split_huge_pud(struct vm_area_struct *vma, pud_t *pud,
 		unsigned long address)
 {
 	spinlock_t *ptl;
-	struct mm_struct *mm = vma->vm_mm;
-	unsigned long haddr = address & HPAGE_PUD_MASK;
+	struct mmu_notifier_range range;
 
-	mmu_notifier_invalidate_range_start(mm, haddr, haddr + HPAGE_PUD_SIZE);
-	ptl = pud_lock(mm, pud);
+	mmu_notifier_range_init(&range, vma->vm_mm, address & HPAGE_PUD_MASK,
+				(address & HPAGE_PUD_MASK) + HPAGE_PUD_SIZE);
+	mmu_notifier_invalidate_range_start(&range);
+	ptl = pud_lock(vma->vm_mm, pud);
 	if (unlikely(!pud_trans_huge(*pud) && !pud_devmap(*pud)))
 		goto out;
-	__split_huge_pud_locked(vma, pud, haddr);
+	__split_huge_pud_locked(vma, pud, range.start);
 
 out:
 	spin_unlock(ptl);
@@ -2043,8 +2040,7 @@ void __split_huge_pud(struct vm_area_struct *vma, pud_t *pud,
 	 * No need to double call mmu_notifier->invalidate_range() callback as
 	 * the above pudp_huge_clear_flush_notify() did already call it.
 	 */
-	mmu_notifier_invalidate_range_only_end(mm, haddr, haddr +
-					       HPAGE_PUD_SIZE);
+	mmu_notifier_invalidate_range_only_end(&range);
 }
 #endif /* CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD */
 
@@ -2244,11 +2240,12 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
 		unsigned long address, bool freeze, struct page *page)
 {
 	spinlock_t *ptl;
-	struct mm_struct *mm = vma->vm_mm;
-	unsigned long haddr = address & HPAGE_PMD_MASK;
+	struct mmu_notifier_range range;
 
-	mmu_notifier_invalidate_range_start(mm, haddr, haddr + HPAGE_PMD_SIZE);
-	ptl = pmd_lock(mm, pmd);
+	mmu_notifier_range_init(&range, vma->vm_mm, address & HPAGE_PMD_MASK,
+				(address & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE);
+	mmu_notifier_invalidate_range_start(&range);
+	ptl = pmd_lock(vma->vm_mm, pmd);
 
 	/*
 	 * If caller asks to setup a migration entries, we need a page to check
@@ -2264,7 +2261,7 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
 			clear_page_mlock(page);
 	} else if (!(pmd_devmap(*pmd) || is_pmd_migration_entry(*pmd)))
 		goto out;
-	__split_huge_pmd_locked(vma, pmd, haddr, freeze);
+	__split_huge_pmd_locked(vma, pmd, range.start, freeze);
 out:
 	spin_unlock(ptl);
 	/*
@@ -2280,8 +2277,7 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
 	 *     any further changes to individual pte will notify. So no need
 	 *     to call mmu_notifier->invalidate_range()
 	 */
-	mmu_notifier_invalidate_range_only_end(mm, haddr, haddr +
-					       HPAGE_PMD_SIZE);
+	mmu_notifier_invalidate_range_only_end(&range);
 }
 
 void split_huge_pmd_address(struct vm_area_struct *vma, unsigned long address,
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 705a3e9cc910..e7c179cbcd75 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -3239,16 +3239,16 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 	int cow;
 	struct hstate *h = hstate_vma(vma);
 	unsigned long sz = huge_page_size(h);
-	unsigned long mmun_start;	/* For mmu_notifiers */
-	unsigned long mmun_end;		/* For mmu_notifiers */
+	struct mmu_notifier_range range;
 	int ret = 0;
 
 	cow = (vma->vm_flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE;
 
-	mmun_start = vma->vm_start;
-	mmun_end = vma->vm_end;
-	if (cow)
-		mmu_notifier_invalidate_range_start(src, mmun_start, mmun_end);
+	if (cow) {
+		mmu_notifier_range_init(&range, src, vma->vm_start,
+					vma->vm_end);
+		mmu_notifier_invalidate_range_start(&range);
+	}
 
 	for (addr = vma->vm_start; addr < vma->vm_end; addr += sz) {
 		spinlock_t *src_ptl, *dst_ptl;
@@ -3324,7 +3324,7 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 	}
 
 	if (cow)
-		mmu_notifier_invalidate_range_end(src, mmun_start, mmun_end);
+		mmu_notifier_invalidate_range_end(&range);
 
 	return ret;
 }
@@ -3341,8 +3341,7 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
 	struct page *page;
 	struct hstate *h = hstate_vma(vma);
 	unsigned long sz = huge_page_size(h);
-	unsigned long mmun_start = start;	/* For mmu_notifiers */
-	unsigned long mmun_end   = end;		/* For mmu_notifiers */
+	struct mmu_notifier_range range;
 
 	WARN_ON(!is_vm_hugetlb_page(vma));
 	BUG_ON(start & ~huge_page_mask(h));
@@ -3358,8 +3357,9 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
 	/*
 	 * If sharing possible, alert mmu notifiers of worst case.
 	 */
-	adjust_range_if_pmd_sharing_possible(vma, &mmun_start, &mmun_end);
-	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
+	mmu_notifier_range_init(&range, mm, start, end);
+	adjust_range_if_pmd_sharing_possible(vma, &range.start, &range.end);
+	mmu_notifier_invalidate_range_start(&range);
 	address = start;
 	for (; address < end; address += sz) {
 		ptep = huge_pte_offset(mm, address, sz);
@@ -3427,7 +3427,7 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
 		if (ref_page)
 			break;
 	}
-	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
+	mmu_notifier_invalidate_range_end(&range);
 	tlb_end_vma(tlb, vma);
 }
 
@@ -3545,9 +3545,8 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
 	struct page *old_page, *new_page;
 	int outside_reserve = 0;
 	vm_fault_t ret = 0;
-	unsigned long mmun_start;	/* For mmu_notifiers */
-	unsigned long mmun_end;		/* For mmu_notifiers */
 	unsigned long haddr = address & huge_page_mask(h);
+	struct mmu_notifier_range range;
 
 	pte = huge_ptep_get(ptep);
 	old_page = pte_page(pte);
@@ -3626,9 +3625,8 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
 	__SetPageUptodate(new_page);
 	set_page_huge_active(new_page);
 
-	mmun_start = haddr;
-	mmun_end = mmun_start + huge_page_size(h);
-	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
+	mmu_notifier_range_init(&range, mm, haddr, haddr + huge_page_size(h));
+	mmu_notifier_invalidate_range_start(&range);
 
 	/*
 	 * Retake the page table lock to check for racing updates
@@ -3641,7 +3639,7 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
 
 		/* Break COW */
 		huge_ptep_clear_flush(vma, haddr, ptep);
-		mmu_notifier_invalidate_range(mm, mmun_start, mmun_end);
+		mmu_notifier_invalidate_range(mm, range.start, range.end);
 		set_huge_pte_at(mm, haddr, ptep,
 				make_huge_pte(vma, new_page, 1));
 		page_remove_rmap(old_page, true);
@@ -3650,7 +3648,7 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
 		new_page = old_page;
 	}
 	spin_unlock(ptl);
-	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
+	mmu_notifier_invalidate_range_end(&range);
 out_release_all:
 	restore_reserve_on_error(h, vma, haddr, new_page);
 	put_page(new_page);
@@ -4339,21 +4337,21 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
 	pte_t pte;
 	struct hstate *h = hstate_vma(vma);
 	unsigned long pages = 0;
-	unsigned long f_start = start;
-	unsigned long f_end = end;
 	bool shared_pmd = false;
+	struct mmu_notifier_range range;
 
 	/*
 	 * In the case of shared PMDs, the area to flush could be beyond
-	 * start/end.  Set f_start/f_end to cover the maximum possible
+	 * start/end.  Set range.start/range.end to cover the maximum possible
 	 * range if PMD sharing is possible.
 	 */
-	adjust_range_if_pmd_sharing_possible(vma, &f_start, &f_end);
+	mmu_notifier_range_init(&range, mm, start, end);
+	adjust_range_if_pmd_sharing_possible(vma, &range.start, &range.end);
 
 	BUG_ON(address >= end);
-	flush_cache_range(vma, f_start, f_end);
+	flush_cache_range(vma, range.start, range.end);
 
-	mmu_notifier_invalidate_range_start(mm, f_start, f_end);
+	mmu_notifier_invalidate_range_start(&range);
 	i_mmap_lock_write(vma->vm_file->f_mapping);
 	for (; address < end; address += huge_page_size(h)) {
 		spinlock_t *ptl;
@@ -4404,7 +4402,7 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
 	 * did unshare a page of pmds, flush the range corresponding to the pud.
 	 */
 	if (shared_pmd)
-		flush_hugetlb_tlb_range(vma, f_start, f_end);
+		flush_hugetlb_tlb_range(vma, range.start, range.end);
 	else
 		flush_hugetlb_tlb_range(vma, start, end);
 	/*
@@ -4414,7 +4412,7 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
 	 * See Documentation/vm/mmu_notifier.rst
 	 */
 	i_mmap_unlock_write(vma->vm_file->f_mapping);
-	mmu_notifier_invalidate_range_end(mm, f_start, f_end);
+	mmu_notifier_invalidate_range_end(&range);
 
 	return pages << h->order;
 }
diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index 8e2ff195ecb3..7736f6c37f19 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -944,8 +944,7 @@ static void collapse_huge_page(struct mm_struct *mm,
 	int isolated = 0, result = 0;
 	struct mem_cgroup *memcg;
 	struct vm_area_struct *vma;
-	unsigned long mmun_start;	/* For mmu_notifiers */
-	unsigned long mmun_end;		/* For mmu_notifiers */
+	struct mmu_notifier_range range;
 	gfp_t gfp;
 
 	VM_BUG_ON(address & ~HPAGE_PMD_MASK);
@@ -1017,9 +1016,8 @@ static void collapse_huge_page(struct mm_struct *mm,
 	pte = pte_offset_map(pmd, address);
 	pte_ptl = pte_lockptr(mm, pmd);
 
-	mmun_start = address;
-	mmun_end   = address + HPAGE_PMD_SIZE;
-	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
+	mmu_notifier_range_init(&range, mm, address, address + HPAGE_PMD_SIZE);
+	mmu_notifier_invalidate_range_start(&range);
 	pmd_ptl = pmd_lock(mm, pmd); /* probably unnecessary */
 	/*
 	 * After this gup_fast can't run anymore. This also removes
@@ -1029,7 +1027,7 @@ static void collapse_huge_page(struct mm_struct *mm,
 	 */
 	_pmd = pmdp_collapse_flush(vma, address, pmd);
 	spin_unlock(pmd_ptl);
-	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
+	mmu_notifier_invalidate_range_end(&range);
 
 	spin_lock(pte_ptl);
 	isolated = __collapse_huge_page_isolate(vma, address, pte);
diff --git a/mm/ksm.c b/mm/ksm.c
index 5b0894b45ee5..6239d2df7a8e 100644
--- a/mm/ksm.c
+++ b/mm/ksm.c
@@ -1042,8 +1042,7 @@ static int write_protect_page(struct vm_area_struct *vma, struct page *page,
 	};
 	int swapped;
 	int err = -EFAULT;
-	unsigned long mmun_start;	/* For mmu_notifiers */
-	unsigned long mmun_end;		/* For mmu_notifiers */
+	struct mmu_notifier_range range;
 
 	pvmw.address = page_address_in_vma(page, vma);
 	if (pvmw.address == -EFAULT)
@@ -1051,9 +1050,9 @@ static int write_protect_page(struct vm_area_struct *vma, struct page *page,
 
 	BUG_ON(PageTransCompound(page));
 
-	mmun_start = pvmw.address;
-	mmun_end   = pvmw.address + PAGE_SIZE;
-	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
+	mmu_notifier_range_init(&range, mm, pvmw.address,
+				pvmw.address + PAGE_SIZE);
+	mmu_notifier_invalidate_range_start(&range);
 
 	if (!page_vma_mapped_walk(&pvmw))
 		goto out_mn;
@@ -1105,7 +1104,7 @@ static int write_protect_page(struct vm_area_struct *vma, struct page *page,
 out_unlock:
 	page_vma_mapped_walk_done(&pvmw);
 out_mn:
-	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
+	mmu_notifier_invalidate_range_end(&range);
 out:
 	return err;
 }
@@ -1129,8 +1128,7 @@ static int replace_page(struct vm_area_struct *vma, struct page *page,
 	spinlock_t *ptl;
 	unsigned long addr;
 	int err = -EFAULT;
-	unsigned long mmun_start;	/* For mmu_notifiers */
-	unsigned long mmun_end;		/* For mmu_notifiers */
+	struct mmu_notifier_range range;
 
 	addr = page_address_in_vma(page, vma);
 	if (addr == -EFAULT)
@@ -1140,9 +1138,8 @@ static int replace_page(struct vm_area_struct *vma, struct page *page,
 	if (!pmd)
 		goto out;
 
-	mmun_start = addr;
-	mmun_end   = addr + PAGE_SIZE;
-	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
+	mmu_notifier_range_init(&range, mm, addr, addr + PAGE_SIZE);
+	mmu_notifier_invalidate_range_start(&range);
 
 	ptep = pte_offset_map_lock(mm, pmd, addr, &ptl);
 	if (!pte_same(*ptep, orig_pte)) {
@@ -1188,7 +1185,7 @@ static int replace_page(struct vm_area_struct *vma, struct page *page,
 	pte_unmap_unlock(ptep, ptl);
 	err = 0;
 out_mn:
-	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
+	mmu_notifier_invalidate_range_end(&range);
 out:
 	return err;
 }
diff --git a/mm/madvise.c b/mm/madvise.c
index 6cb1ca93e290..21a7881a2db4 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -458,29 +458,30 @@ static void madvise_free_page_range(struct mmu_gather *tlb,
 static int madvise_free_single_vma(struct vm_area_struct *vma,
 			unsigned long start_addr, unsigned long end_addr)
 {
-	unsigned long start, end;
 	struct mm_struct *mm = vma->vm_mm;
+	struct mmu_notifier_range range;
 	struct mmu_gather tlb;
 
 	/* MADV_FREE works for only anon vma at the moment */
 	if (!vma_is_anonymous(vma))
 		return -EINVAL;
 
-	start = max(vma->vm_start, start_addr);
-	if (start >= vma->vm_end)
+	range.start = max(vma->vm_start, start_addr);
+	if (range.start >= vma->vm_end)
 		return -EINVAL;
-	end = min(vma->vm_end, end_addr);
-	if (end <= vma->vm_start)
+	range.end = min(vma->vm_end, end_addr);
+	if (range.end <= vma->vm_start)
 		return -EINVAL;
+	mmu_notifier_range_init(&range, mm, range.start, range.end);
 
 	lru_add_drain();
-	tlb_gather_mmu(&tlb, mm, start, end);
+	tlb_gather_mmu(&tlb, mm, range.start, range.end);
 	update_hiwater_rss(mm);
 
-	mmu_notifier_invalidate_range_start(mm, start, end);
-	madvise_free_page_range(&tlb, vma, start, end);
-	mmu_notifier_invalidate_range_end(mm, start, end);
-	tlb_finish_mmu(&tlb, start, end);
+	mmu_notifier_invalidate_range_start(&range);
+	madvise_free_page_range(&tlb, vma, range.start, range.end);
+	mmu_notifier_invalidate_range_end(&range);
+	tlb_finish_mmu(&tlb, range.start, range.end);
 
 	return 0;
 }
diff --git a/mm/memory.c b/mm/memory.c
index 4ad2d293ddc2..574307f11464 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -973,8 +973,7 @@ int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 	unsigned long next;
 	unsigned long addr = vma->vm_start;
 	unsigned long end = vma->vm_end;
-	unsigned long mmun_start;	/* For mmu_notifiers */
-	unsigned long mmun_end;		/* For mmu_notifiers */
+	struct mmu_notifier_range range;
 	bool is_cow;
 	int ret;
 
@@ -1008,11 +1007,11 @@ int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 	 * is_cow_mapping() returns true.
 	 */
 	is_cow = is_cow_mapping(vma->vm_flags);
-	mmun_start = addr;
-	mmun_end   = end;
-	if (is_cow)
-		mmu_notifier_invalidate_range_start(src_mm, mmun_start,
-						    mmun_end);
+
+	if (is_cow) {
+		mmu_notifier_range_init(&range, src_mm, addr, end);
+		mmu_notifier_invalidate_range_start(&range);
+	}
 
 	ret = 0;
 	dst_pgd = pgd_offset(dst_mm, addr);
@@ -1029,7 +1028,7 @@ int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 	} while (dst_pgd++, src_pgd++, addr = next, addr != end);
 
 	if (is_cow)
-		mmu_notifier_invalidate_range_end(src_mm, mmun_start, mmun_end);
+		mmu_notifier_invalidate_range_end(&range);
 	return ret;
 }
 
@@ -1332,12 +1331,13 @@ void unmap_vmas(struct mmu_gather *tlb,
 		struct vm_area_struct *vma, unsigned long start_addr,
 		unsigned long end_addr)
 {
-	struct mm_struct *mm = vma->vm_mm;
+	struct mmu_notifier_range range;
 
-	mmu_notifier_invalidate_range_start(mm, start_addr, end_addr);
+	mmu_notifier_range_init(&range, vma->vm_mm, start_addr, end_addr);
+	mmu_notifier_invalidate_range_start(&range);
 	for ( ; vma && vma->vm_start < end_addr; vma = vma->vm_next)
 		unmap_single_vma(tlb, vma, start_addr, end_addr, NULL);
-	mmu_notifier_invalidate_range_end(mm, start_addr, end_addr);
+	mmu_notifier_invalidate_range_end(&range);
 }
 
 /**
@@ -1351,18 +1351,18 @@ void unmap_vmas(struct mmu_gather *tlb,
 void zap_page_range(struct vm_area_struct *vma, unsigned long start,
 		unsigned long size)
 {
-	struct mm_struct *mm = vma->vm_mm;
+	struct mmu_notifier_range range;
 	struct mmu_gather tlb;
-	unsigned long end = start + size;
 
 	lru_add_drain();
-	tlb_gather_mmu(&tlb, mm, start, end);
-	update_hiwater_rss(mm);
-	mmu_notifier_invalidate_range_start(mm, start, end);
-	for ( ; vma && vma->vm_start < end; vma = vma->vm_next)
-		unmap_single_vma(&tlb, vma, start, end, NULL);
-	mmu_notifier_invalidate_range_end(mm, start, end);
-	tlb_finish_mmu(&tlb, start, end);
+	mmu_notifier_range_init(&range, vma->vm_mm, start, start + size);
+	tlb_gather_mmu(&tlb, vma->vm_mm, start, range.end);
+	update_hiwater_rss(vma->vm_mm);
+	mmu_notifier_invalidate_range_start(&range);
+	for ( ; vma && vma->vm_start < range.end; vma = vma->vm_next)
+		unmap_single_vma(&tlb, vma, start, range.end, NULL);
+	mmu_notifier_invalidate_range_end(&range);
+	tlb_finish_mmu(&tlb, start, range.end);
 }
 
 /**
@@ -1377,17 +1377,17 @@ void zap_page_range(struct vm_area_struct *vma, unsigned long start,
 static void zap_page_range_single(struct vm_area_struct *vma, unsigned long address,
 		unsigned long size, struct zap_details *details)
 {
-	struct mm_struct *mm = vma->vm_mm;
+	struct mmu_notifier_range range;
 	struct mmu_gather tlb;
-	unsigned long end = address + size;
 
 	lru_add_drain();
-	tlb_gather_mmu(&tlb, mm, address, end);
-	update_hiwater_rss(mm);
-	mmu_notifier_invalidate_range_start(mm, address, end);
-	unmap_single_vma(&tlb, vma, address, end, details);
-	mmu_notifier_invalidate_range_end(mm, address, end);
-	tlb_finish_mmu(&tlb, address, end);
+	mmu_notifier_range_init(&range, vma->vm_mm, address, address + size);
+	tlb_gather_mmu(&tlb, vma->vm_mm, address, range.end);
+	update_hiwater_rss(vma->vm_mm);
+	mmu_notifier_invalidate_range_start(&range);
+	unmap_single_vma(&tlb, vma, address, range.end, details);
+	mmu_notifier_invalidate_range_end(&range);
+	tlb_finish_mmu(&tlb, address, range.end);
 }
 
 /**
@@ -2247,9 +2247,8 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
 	struct page *new_page = NULL;
 	pte_t entry;
 	int page_copied = 0;
-	const unsigned long mmun_start = vmf->address & PAGE_MASK;
-	const unsigned long mmun_end = mmun_start + PAGE_SIZE;
 	struct mem_cgroup *memcg;
+	struct mmu_notifier_range range;
 
 	if (unlikely(anon_vma_prepare(vma)))
 		goto oom;
@@ -2272,7 +2271,9 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
 
 	__SetPageUptodate(new_page);
 
-	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
+	mmu_notifier_range_init(&range, mm, vmf->address & PAGE_MASK,
+				(vmf->address & PAGE_MASK) + PAGE_SIZE);
+	mmu_notifier_invalidate_range_start(&range);
 
 	/*
 	 * Re-check the pte - we dropped the lock
@@ -2349,7 +2350,7 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
 	 * No need to double call mmu_notifier->invalidate_range() callback as
 	 * the above ptep_clear_flush_notify() did already call it.
 	 */
-	mmu_notifier_invalidate_range_only_end(mm, mmun_start, mmun_end);
+	mmu_notifier_invalidate_range_only_end(&range);
 	if (old_page) {
 		/*
 		 * Don't let another task, with possibly unlocked vma,
@@ -4030,7 +4031,7 @@ int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address)
 #endif /* __PAGETABLE_PMD_FOLDED */
 
 static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
-			    unsigned long *start, unsigned long *end,
+			    struct mmu_notifier_range *range,
 			    pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp)
 {
 	pgd_t *pgd;
@@ -4058,10 +4059,10 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
 		if (!pmdpp)
 			goto out;
 
-		if (start && end) {
-			*start = address & PMD_MASK;
-			*end = *start + PMD_SIZE;
-			mmu_notifier_invalidate_range_start(mm, *start, *end);
+		if (range) {
+			mmu_notifier_range_init(range, mm, address & PMD_MASK,
+					     (address & PMD_MASK) + PMD_SIZE);
+			mmu_notifier_invalidate_range_start(range);
 		}
 		*ptlp = pmd_lock(mm, pmd);
 		if (pmd_huge(*pmd)) {
@@ -4069,17 +4070,17 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
 			return 0;
 		}
 		spin_unlock(*ptlp);
-		if (start && end)
-			mmu_notifier_invalidate_range_end(mm, *start, *end);
+		if (range)
+			mmu_notifier_invalidate_range_end(range);
 	}
 
 	if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd)))
 		goto out;
 
-	if (start && end) {
-		*start = address & PAGE_MASK;
-		*end = *start + PAGE_SIZE;
-		mmu_notifier_invalidate_range_start(mm, *start, *end);
+	if (range) {
+		range->start = address & PAGE_MASK;
+		range->end = range->start + PAGE_SIZE;
+		mmu_notifier_invalidate_range_start(range);
 	}
 	ptep = pte_offset_map_lock(mm, pmd, address, ptlp);
 	if (!pte_present(*ptep))
@@ -4088,8 +4089,8 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
 	return 0;
 unlock:
 	pte_unmap_unlock(ptep, *ptlp);
-	if (start && end)
-		mmu_notifier_invalidate_range_end(mm, *start, *end);
+	if (range)
+		mmu_notifier_invalidate_range_end(range);
 out:
 	return -EINVAL;
 }
@@ -4101,20 +4102,20 @@ static inline int follow_pte(struct mm_struct *mm, unsigned long address,
 
 	/* (void) is needed to make gcc happy */
 	(void) __cond_lock(*ptlp,
-			   !(res = __follow_pte_pmd(mm, address, NULL, NULL,
+			   !(res = __follow_pte_pmd(mm, address, NULL,
 						    ptepp, NULL, ptlp)));
 	return res;
 }
 
 int follow_pte_pmd(struct mm_struct *mm, unsigned long address,
-			     unsigned long *start, unsigned long *end,
+		 	     struct mmu_notifier_range *range,
 			     pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp)
 {
 	int res;
 
 	/* (void) is needed to make gcc happy */
 	(void) __cond_lock(*ptlp,
-			   !(res = __follow_pte_pmd(mm, address, start, end,
+			   !(res = __follow_pte_pmd(mm, address, range,
 						    ptepp, pmdpp, ptlp)));
 	return res;
 }
diff --git a/mm/migrate.c b/mm/migrate.c
index f7e4bfdc13b7..74f5b3208c05 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -2303,6 +2303,7 @@ static int migrate_vma_collect_pmd(pmd_t *pmdp,
  */
 static void migrate_vma_collect(struct migrate_vma *migrate)
 {
+	struct mmu_notifier_range range;
 	struct mm_walk mm_walk;
 
 	mm_walk.pmd_entry = migrate_vma_collect_pmd;
@@ -2314,13 +2315,11 @@ static void migrate_vma_collect(struct migrate_vma *migrate)
 	mm_walk.mm = migrate->vma->vm_mm;
 	mm_walk.private = migrate;
 
-	mmu_notifier_invalidate_range_start(mm_walk.mm,
-					    migrate->start,
-					    migrate->end);
+	mmu_notifier_range_init(&range, mm_walk.mm, migrate->start,
+				migrate->end);
+	mmu_notifier_invalidate_range_start(&range);
 	walk_page_range(migrate->start, migrate->end, &mm_walk);
-	mmu_notifier_invalidate_range_end(mm_walk.mm,
-					  migrate->start,
-					  migrate->end);
+	mmu_notifier_invalidate_range_end(&range);
 
 	migrate->end = migrate->start + (migrate->npages << PAGE_SHIFT);
 }
@@ -2703,7 +2702,8 @@ static void migrate_vma_pages(struct migrate_vma *migrate)
 	const unsigned long start = migrate->start;
 	struct vm_area_struct *vma = migrate->vma;
 	struct mm_struct *mm = vma->vm_mm;
-	unsigned long addr, i, mmu_start;
+	struct mmu_notifier_range range;
+	unsigned long addr, i;
 	bool notified = false;
 
 	for (i = 0, addr = start; i < npages; addr += PAGE_SIZE, i++) {
@@ -2722,11 +2722,11 @@ static void migrate_vma_pages(struct migrate_vma *migrate)
 				continue;
 			}
 			if (!notified) {
-				mmu_start = addr;
 				notified = true;
-				mmu_notifier_invalidate_range_start(mm,
-								mmu_start,
-								migrate->end);
+
+				mmu_notifier_range_init(&range, mm, addr,
+							migrate->end);
+				mmu_notifier_invalidate_range_start(&range);
 			}
 			migrate_vma_insert_page(migrate, addr, newpage,
 						&migrate->src[i],
@@ -2767,8 +2767,7 @@ static void migrate_vma_pages(struct migrate_vma *migrate)
 	 * did already call it.
 	 */
 	if (notified)
-		mmu_notifier_invalidate_range_only_end(mm, mmu_start,
-						       migrate->end);
+		mmu_notifier_invalidate_range_only_end(&range);
 }
 
 /*
diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
index 5f6665ae3ee2..4c52b3514c50 100644
--- a/mm/mmu_notifier.c
+++ b/mm/mmu_notifier.c
@@ -174,28 +174,20 @@ void __mmu_notifier_change_pte(struct mm_struct *mm, unsigned long address,
 	srcu_read_unlock(&srcu, id);
 }
 
-int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
-				  unsigned long start, unsigned long end,
-				  bool blockable)
+int __mmu_notifier_invalidate_range_start(struct mmu_notifier_range *range)
 {
-	struct mmu_notifier_range _range, *range = &_range;
 	struct mmu_notifier *mn;
 	int ret = 0;
 	int id;
 
-	range->blockable = blockable;
-	range->start = start;
-	range->end = end;
-	range->mm = mm;
-
 	id = srcu_read_lock(&srcu);
-	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
+	hlist_for_each_entry_rcu(mn, &range->mm->mmu_notifier_mm->list, hlist) {
 		if (mn->ops->invalidate_range_start) {
 			int _ret = mn->ops->invalidate_range_start(mn, range);
 			if (_ret) {
 				pr_info("%pS callback failed with %d in %sblockable context.\n",
 						mn->ops->invalidate_range_start, _ret,
-						!blockable ? "non-" : "");
+						!range->blockable ? "non-" : "");
 				ret = _ret;
 			}
 		}
@@ -206,27 +198,14 @@ int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
 }
 EXPORT_SYMBOL_GPL(__mmu_notifier_invalidate_range_start);
 
-void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
-					 unsigned long start,
-					 unsigned long end,
+void __mmu_notifier_invalidate_range_end(struct mmu_notifier_range *range,
 					 bool only_end)
 {
-	struct mmu_notifier_range _range, *range = &_range;
 	struct mmu_notifier *mn;
 	int id;
 
-	/*
-	 * The end call back will never be call if the start refused to go
-	 * through because of blockable was false so here assume that we
-	 * can block.
-	 */
-	range->blockable = true;
-	range->start = start;
-	range->end = end;
-	range->mm = mm;
-
 	id = srcu_read_lock(&srcu);
-	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
+	hlist_for_each_entry_rcu(mn, &range->mm->mmu_notifier_mm->list, hlist) {
 		/*
 		 * Call invalidate_range here too to avoid the need for the
 		 * subsystem of having to register an invalidate_range_end
@@ -241,7 +220,9 @@ void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
 		 * already happen under page table lock.
 		 */
 		if (!only_end && mn->ops->invalidate_range)
-			mn->ops->invalidate_range(mn, mm, start, end);
+			mn->ops->invalidate_range(mn, range->mm,
+						  range->start,
+						  range->end);
 		if (mn->ops->invalidate_range_end)
 			mn->ops->invalidate_range_end(mn, range);
 	}
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 6d331620b9e5..36cb358db170 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -167,11 +167,12 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
 		pgprot_t newprot, int dirty_accountable, int prot_numa)
 {
 	pmd_t *pmd;
-	struct mm_struct *mm = vma->vm_mm;
 	unsigned long next;
 	unsigned long pages = 0;
 	unsigned long nr_huge_updates = 0;
-	unsigned long mni_start = 0;
+	struct mmu_notifier_range range;
+
+	range.start = 0;
 
 	pmd = pmd_offset(pud, addr);
 	do {
@@ -183,9 +184,9 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
 			goto next;
 
 		/* invoke the mmu notifier if the pmd is populated */
-		if (!mni_start) {
-			mni_start = addr;
-			mmu_notifier_invalidate_range_start(mm, mni_start, end);
+		if (!range.start) {
+			mmu_notifier_range_init(&range, vma->vm_mm, addr, end);
+			mmu_notifier_invalidate_range_start(&range);
 		}
 
 		if (is_swap_pmd(*pmd) || pmd_trans_huge(*pmd) || pmd_devmap(*pmd)) {
@@ -214,8 +215,8 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
 		cond_resched();
 	} while (pmd++, addr = next, addr != end);
 
-	if (mni_start)
-		mmu_notifier_invalidate_range_end(mm, mni_start, end);
+	if (range.start)
+		mmu_notifier_invalidate_range_end(&range);
 
 	if (nr_huge_updates)
 		count_vm_numa_events(NUMA_HUGE_PTE_UPDATES, nr_huge_updates);
diff --git a/mm/mremap.c b/mm/mremap.c
index 7f9f9180e401..def01d86e36f 100644
--- a/mm/mremap.c
+++ b/mm/mremap.c
@@ -197,16 +197,14 @@ unsigned long move_page_tables(struct vm_area_struct *vma,
 		bool need_rmap_locks)
 {
 	unsigned long extent, next, old_end;
+	struct mmu_notifier_range range;
 	pmd_t *old_pmd, *new_pmd;
-	unsigned long mmun_start;	/* For mmu_notifiers */
-	unsigned long mmun_end;		/* For mmu_notifiers */
 
 	old_end = old_addr + len;
 	flush_cache_range(vma, old_addr, old_end);
 
-	mmun_start = old_addr;
-	mmun_end   = old_end;
-	mmu_notifier_invalidate_range_start(vma->vm_mm, mmun_start, mmun_end);
+	mmu_notifier_range_init(&range, vma->vm_mm, old_addr, old_end);
+	mmu_notifier_invalidate_range_start(&range);
 
 	for (; old_addr < old_end; old_addr += extent, new_addr += extent) {
 		cond_resched();
@@ -247,7 +245,7 @@ unsigned long move_page_tables(struct vm_area_struct *vma,
 			  new_pmd, new_addr, need_rmap_locks);
 	}
 
-	mmu_notifier_invalidate_range_end(vma->vm_mm, mmun_start, mmun_end);
+	mmu_notifier_invalidate_range_end(&range);
 
 	return len + old_addr - old_end;	/* how much done */
 }
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 6589f60d5018..1eea8b04f27a 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -516,19 +516,20 @@ bool __oom_reap_task_mm(struct mm_struct *mm)
 		 * count elevated without a good reason.
 		 */
 		if (vma_is_anonymous(vma) || !(vma->vm_flags & VM_SHARED)) {
-			const unsigned long start = vma->vm_start;
-			const unsigned long end = vma->vm_end;
+			struct mmu_notifier_range range;
 			struct mmu_gather tlb;
 
-			tlb_gather_mmu(&tlb, mm, start, end);
-			if (mmu_notifier_invalidate_range_start_nonblock(mm, start, end)) {
-				tlb_finish_mmu(&tlb, start, end);
+			mmu_notifier_range_init(&range, mm, vma->vm_start,
+						vma->vm_end);
+			tlb_gather_mmu(&tlb, mm, range.start, range.end);
+			if (mmu_notifier_invalidate_range_start_nonblock(&range)) {
+				tlb_finish_mmu(&tlb, range.start, range.end);
 				ret = false;
 				continue;
 			}
-			unmap_page_range(&tlb, vma, start, end, NULL);
-			mmu_notifier_invalidate_range_end(mm, start, end);
-			tlb_finish_mmu(&tlb, start, end);
+			unmap_page_range(&tlb, vma, range.start, range.end, NULL);
+			mmu_notifier_invalidate_range_end(&range);
+			tlb_finish_mmu(&tlb, range.start, range.end);
 		}
 	}
 
diff --git a/mm/rmap.c b/mm/rmap.c
index 85b7f9423352..c75f72f6fe0e 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -889,15 +889,17 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma,
 		.address = address,
 		.flags = PVMW_SYNC,
 	};
-	unsigned long start = address, end;
+	struct mmu_notifier_range range;
 	int *cleaned = arg;
 
 	/*
 	 * We have to assume the worse case ie pmd for invalidation. Note that
 	 * the page can not be free from this function.
 	 */
-	end = min(vma->vm_end, start + (PAGE_SIZE << compound_order(page)));
-	mmu_notifier_invalidate_range_start(vma->vm_mm, start, end);
+	mmu_notifier_range_init(&range, vma->vm_mm, address,
+				min(vma->vm_end, address +
+				    (PAGE_SIZE << compound_order(page))));
+	mmu_notifier_invalidate_range_start(&range);
 
 	while (page_vma_mapped_walk(&pvmw)) {
 		unsigned long cstart;
@@ -949,7 +951,7 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma,
 			(*cleaned)++;
 	}
 
-	mmu_notifier_invalidate_range_end(vma->vm_mm, start, end);
+	mmu_notifier_invalidate_range_end(&range);
 
 	return true;
 }
@@ -1345,7 +1347,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
 	pte_t pteval;
 	struct page *subpage;
 	bool ret = true;
-	unsigned long start = address, end;
+	struct mmu_notifier_range range;
 	enum ttu_flags flags = (enum ttu_flags)arg;
 
 	/* munlock has nothing to gain from examining un-locked vmas */
@@ -1369,15 +1371,18 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
 	 * Note that the page can not be free in this function as call of
 	 * try_to_unmap() must hold a reference on the page.
 	 */
-	end = min(vma->vm_end, start + (PAGE_SIZE << compound_order(page)));
+	mmu_notifier_range_init(&range, vma->vm_mm, vma->vm_start,
+				min(vma->vm_end, vma->vm_start +
+				    (PAGE_SIZE << compound_order(page))));
 	if (PageHuge(page)) {
 		/*
 		 * If sharing is possible, start and end will be adjusted
 		 * accordingly.
 		 */
-		adjust_range_if_pmd_sharing_possible(vma, &start, &end);
+		adjust_range_if_pmd_sharing_possible(vma, &range.start,
+						     &range.end);
 	}
-	mmu_notifier_invalidate_range_start(vma->vm_mm, start, end);
+	mmu_notifier_invalidate_range_start(&range);
 
 	while (page_vma_mapped_walk(&pvmw)) {
 #ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
@@ -1428,9 +1433,10 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
 				 * we must flush them all.  start/end were
 				 * already adjusted above to cover this range.
 				 */
-				flush_cache_range(vma, start, end);
-				flush_tlb_range(vma, start, end);
-				mmu_notifier_invalidate_range(mm, start, end);
+				flush_cache_range(vma, range.start, range.end);
+				flush_tlb_range(vma, range.start, range.end);
+				mmu_notifier_invalidate_range(mm, range.start,
+							      range.end);
 
 				/*
 				 * The ref count of the PMD page was dropped
@@ -1650,7 +1656,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
 		put_page(page);
 	}
 
-	mmu_notifier_invalidate_range_end(vma->vm_mm, start, end);
+	mmu_notifier_invalidate_range_end(&range);
 
 	return ret;
 }
-- 
2.17.2

Re: [PATCH v2 2/3] mm/mmu_notifier: use structure for invalidate_range_start/end calls v2

[Jan Kara]

On Wed 05-12-18 00:36:27, jglisse@redhat.com wrote:
> From: Jérôme Glisse <jglisse@redhat.com>
> 
> To avoid having to change many call sites everytime we want to add a
> parameter use a structure to group all parameters for the mmu_notifier
> invalidate_range_start/end cakks. No functional changes with this
> patch.
> 
> Changes since v1:
>     - introduce mmu_notifier_range_init() as an helper to initialize
>       the range structure allowing to optimize out the case when mmu
>       notifier is not enabled
>     - fix mm/migrate.c migrate_vma_collect()
> 
> Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
> Acked-by: Christian König <christian.koenig@amd.com>

The patch looks good to me. You can add:

Acked-by: Jan Kara <jack@suse.cz>

								Honza


> Cc: Andrew Morton <akpm@linux-foundation.org>
> Cc: Matthew Wilcox <mawilcox@microsoft.com>
> Cc: Ross Zwisler <zwisler@kernel.org>
> Cc: Jan Kara <jack@suse.cz>
> Cc: Dan Williams <dan.j.williams@intel.com>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Radim Krčmář <rkrcmar@redhat.com>
> Cc: Michal Hocko <mhocko@kernel.org>
> Cc: Felix Kuehling <felix.kuehling@amd.com>
> Cc: Ralph Campbell <rcampbell@nvidia.com>
> Cc: John Hubbard <jhubbard@nvidia.com>
> Cc: kvm@vger.kernel.org
> Cc: dri-devel@lists.freedesktop.org
> Cc: linux-rdma@vger.kernel.org
> Cc: linux-fsdevel@vger.kernel.org
> ---
>  fs/dax.c                     |  8 +--
>  fs/proc/task_mmu.c           |  7 ++-
>  include/linux/mm.h           |  4 +-
>  include/linux/mmu_notifier.h | 87 +++++++++++++++++++++-----------
>  kernel/events/uprobes.c      | 10 ++--
>  mm/huge_memory.c             | 54 ++++++++++----------
>  mm/hugetlb.c                 | 52 ++++++++++---------
>  mm/khugepaged.c              | 10 ++--
>  mm/ksm.c                     | 21 ++++----
>  mm/madvise.c                 | 21 ++++----
>  mm/memory.c                  | 97 ++++++++++++++++++------------------
>  mm/migrate.c                 | 25 +++++-----
>  mm/mmu_notifier.c            | 35 +++----------
>  mm/mprotect.c                | 15 +++---
>  mm/mremap.c                  | 10 ++--
>  mm/oom_kill.c                | 17 ++++---
>  mm/rmap.c                    | 30 ++++++-----
>  17 files changed, 258 insertions(+), 245 deletions(-)
> 
> diff --git a/fs/dax.c b/fs/dax.c
> index 9bcce89ea18e..874085bacaf5 100644
> --- a/fs/dax.c
> +++ b/fs/dax.c
> @@ -758,7 +758,8 @@ static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index,
>  
>  	i_mmap_lock_read(mapping);
>  	vma_interval_tree_foreach(vma, &mapping->i_mmap, index, index) {
> -		unsigned long address, start, end;
> +		struct mmu_notifier_range range;
> +		unsigned long address;
>  
>  		cond_resched();
>  
> @@ -772,7 +773,8 @@ static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index,
>  		 * call mmu_notifier_invalidate_range_start() on our behalf
>  		 * before taking any lock.
>  		 */
> -		if (follow_pte_pmd(vma->vm_mm, address, &start, &end, &ptep, &pmdp, &ptl))
> +		if (follow_pte_pmd(vma->vm_mm, address, &range,
> +				   &ptep, &pmdp, &ptl))
>  			continue;
>  
>  		/*
> @@ -814,7 +816,7 @@ static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index,
>  			pte_unmap_unlock(ptep, ptl);
>  		}
>  
> -		mmu_notifier_invalidate_range_end(vma->vm_mm, start, end);
> +		mmu_notifier_invalidate_range_end(&range);
>  	}
>  	i_mmap_unlock_read(mapping);
>  }
> diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
> index 47c3764c469b..b3ddceb003bc 100644
> --- a/fs/proc/task_mmu.c
> +++ b/fs/proc/task_mmu.c
> @@ -1096,6 +1096,7 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf,
>  		return -ESRCH;
>  	mm = get_task_mm(task);
>  	if (mm) {
> +		struct mmu_notifier_range range;
>  		struct clear_refs_private cp = {
>  			.type = type,
>  		};
> @@ -1139,11 +1140,13 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf,
>  				downgrade_write(&mm->mmap_sem);
>  				break;
>  			}
> -			mmu_notifier_invalidate_range_start(mm, 0, -1);
> +
> +			mmu_notifier_range_init(&range, mm, 0, -1UL);
> +			mmu_notifier_invalidate_range_start(&range);
>  		}
>  		walk_page_range(0, mm->highest_vm_end, &clear_refs_walk);
>  		if (type == CLEAR_REFS_SOFT_DIRTY)
> -			mmu_notifier_invalidate_range_end(mm, 0, -1);
> +			mmu_notifier_invalidate_range_end(&range);
>  		tlb_finish_mmu(&tlb, 0, -1);
>  		up_read(&mm->mmap_sem);
>  out_mm:
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index 5411de93a363..e7b6f2b30713 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -1397,6 +1397,8 @@ struct mm_walk {
>  	void *private;
>  };
>  
> +struct mmu_notifier_range;
> +
>  int walk_page_range(unsigned long addr, unsigned long end,
>  		struct mm_walk *walk);
>  int walk_page_vma(struct vm_area_struct *vma, struct mm_walk *walk);
> @@ -1405,7 +1407,7 @@ void free_pgd_range(struct mmu_gather *tlb, unsigned long addr,
>  int copy_page_range(struct mm_struct *dst, struct mm_struct *src,
>  			struct vm_area_struct *vma);
>  int follow_pte_pmd(struct mm_struct *mm, unsigned long address,
> -			     unsigned long *start, unsigned long *end,
> +		 	     struct mmu_notifier_range *range,
>  			     pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp);
>  int follow_pfn(struct vm_area_struct *vma, unsigned long address,
>  	unsigned long *pfn);
> diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h
> index 368f0c1a049d..39b06772427f 100644
> --- a/include/linux/mmu_notifier.h
> +++ b/include/linux/mmu_notifier.h
> @@ -220,11 +220,8 @@ extern int __mmu_notifier_test_young(struct mm_struct *mm,
>  				     unsigned long address);
>  extern void __mmu_notifier_change_pte(struct mm_struct *mm,
>  				      unsigned long address, pte_t pte);
> -extern int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
> -				  unsigned long start, unsigned long end,
> -				  bool blockable);
> -extern void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
> -				  unsigned long start, unsigned long end,
> +extern int __mmu_notifier_invalidate_range_start(struct mmu_notifier_range *);
> +extern void __mmu_notifier_invalidate_range_end(struct mmu_notifier_range *r,
>  				  bool only_end);
>  extern void __mmu_notifier_invalidate_range(struct mm_struct *mm,
>  				  unsigned long start, unsigned long end);
> @@ -268,33 +265,37 @@ static inline void mmu_notifier_change_pte(struct mm_struct *mm,
>  		__mmu_notifier_change_pte(mm, address, pte);
>  }
>  
> -static inline void mmu_notifier_invalidate_range_start(struct mm_struct *mm,
> -				  unsigned long start, unsigned long end)
> +static inline void
> +mmu_notifier_invalidate_range_start(struct mmu_notifier_range *range)
>  {
> -	if (mm_has_notifiers(mm))
> -		__mmu_notifier_invalidate_range_start(mm, start, end, true);
> +	if (mm_has_notifiers(range->mm)) {
> +		range->blockable = true;
> +		__mmu_notifier_invalidate_range_start(range);
> +	}
>  }
>  
> -static inline int mmu_notifier_invalidate_range_start_nonblock(struct mm_struct *mm,
> -				  unsigned long start, unsigned long end)
> +static inline int
> +mmu_notifier_invalidate_range_start_nonblock(struct mmu_notifier_range *range)
>  {
> -	if (mm_has_notifiers(mm))
> -		return __mmu_notifier_invalidate_range_start(mm, start, end, false);
> +	if (mm_has_notifiers(range->mm)) {
> +		range->blockable = false;
> +		return __mmu_notifier_invalidate_range_start(range);
> +	}
>  	return 0;
>  }
>  
> -static inline void mmu_notifier_invalidate_range_end(struct mm_struct *mm,
> -				  unsigned long start, unsigned long end)
> +static inline void
> +mmu_notifier_invalidate_range_end(struct mmu_notifier_range *range)
>  {
> -	if (mm_has_notifiers(mm))
> -		__mmu_notifier_invalidate_range_end(mm, start, end, false);
> +	if (mm_has_notifiers(range->mm))
> +		__mmu_notifier_invalidate_range_end(range, false);
>  }
>  
> -static inline void mmu_notifier_invalidate_range_only_end(struct mm_struct *mm,
> -				  unsigned long start, unsigned long end)
> +static inline void
> +mmu_notifier_invalidate_range_only_end(struct mmu_notifier_range *range)
>  {
> -	if (mm_has_notifiers(mm))
> -		__mmu_notifier_invalidate_range_end(mm, start, end, true);
> +	if (mm_has_notifiers(range->mm))
> +		__mmu_notifier_invalidate_range_end(range, true);
>  }
>  
>  static inline void mmu_notifier_invalidate_range(struct mm_struct *mm,
> @@ -315,6 +316,17 @@ static inline void mmu_notifier_mm_destroy(struct mm_struct *mm)
>  		__mmu_notifier_mm_destroy(mm);
>  }
>  
> +
> +static inline void mmu_notifier_range_init(struct mmu_notifier_range *range,
> +					   struct mm_struct *mm,
> +					   unsigned long start,
> +					   unsigned long end)
> +{
> +	range->mm = mm;
> +	range->start = start;
> +	range->end = end;
> +}
> +
>  #define ptep_clear_flush_young_notify(__vma, __address, __ptep)		\
>  ({									\
>  	int __young;							\
> @@ -428,6 +440,23 @@ extern void mmu_notifier_synchronize(void);
>  
>  #else /* CONFIG_MMU_NOTIFIER */
>  
> +struct mmu_notifier_range {
> +	unsigned long start;
> +	unsigned long end;
> +};
> +
> +static inline void _mmu_notifier_range_init(struct mmu_notifier_range *range,
> +					    unsigned long start,
> +					    unsigned long end)
> +{
> +	range->start = start;
> +	range->end = end;
> +}
> +
> +#define mmu_notifier_range_init(range, mm, start, end) \
> +	_mmu_notifier_range_init(range, start, end)
> +
> +
>  static inline int mm_has_notifiers(struct mm_struct *mm)
>  {
>  	return 0;
> @@ -455,24 +484,24 @@ static inline void mmu_notifier_change_pte(struct mm_struct *mm,
>  {
>  }
>  
> -static inline void mmu_notifier_invalidate_range_start(struct mm_struct *mm,
> -				  unsigned long start, unsigned long end)
> +static inline void
> +mmu_notifier_invalidate_range_start(struct mmu_notifier_range *range)
>  {
>  }
>  
> -static inline int mmu_notifier_invalidate_range_start_nonblock(struct mm_struct *mm,
> -				  unsigned long start, unsigned long end)
> +static inline int
> +mmu_notifier_invalidate_range_start_nonblock(struct mmu_notifier_range *range)
>  {
>  	return 0;
>  }
>  
> -static inline void mmu_notifier_invalidate_range_end(struct mm_struct *mm,
> -				  unsigned long start, unsigned long end)
> +static inline
> +void mmu_notifier_invalidate_range_end(struct mmu_notifier_range *range)
>  {
>  }
>  
> -static inline void mmu_notifier_invalidate_range_only_end(struct mm_struct *mm,
> -				  unsigned long start, unsigned long end)
> +static inline void
> +mmu_notifier_invalidate_range_only_end(struct mmu_notifier_range *range)
>  {
>  }
>  
> diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
> index 322e97bbb437..1fc8a93709c3 100644
> --- a/kernel/events/uprobes.c
> +++ b/kernel/events/uprobes.c
> @@ -171,11 +171,11 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr,
>  		.address = addr,
>  	};
>  	int err;
> -	/* For mmu_notifiers */
> -	const unsigned long mmun_start = addr;
> -	const unsigned long mmun_end   = addr + PAGE_SIZE;
> +	struct mmu_notifier_range range;
>  	struct mem_cgroup *memcg;
>  
> +	mmu_notifier_range_init(&range, mm, addr, addr + PAGE_SIZE);
> +
>  	VM_BUG_ON_PAGE(PageTransHuge(old_page), old_page);
>  
>  	err = mem_cgroup_try_charge(new_page, vma->vm_mm, GFP_KERNEL, &memcg,
> @@ -186,7 +186,7 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr,
>  	/* For try_to_free_swap() and munlock_vma_page() below */
>  	lock_page(old_page);
>  
> -	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
> +	mmu_notifier_invalidate_range_start(&range);
>  	err = -EAGAIN;
>  	if (!page_vma_mapped_walk(&pvmw)) {
>  		mem_cgroup_cancel_charge(new_page, memcg, false);
> @@ -220,7 +220,7 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr,
>  
>  	err = 0;
>   unlock:
> -	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
> +	mmu_notifier_invalidate_range_end(&range);
>  	unlock_page(old_page);
>  	return err;
>  }
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 622cced74fd9..c1d3ce809416 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -1144,8 +1144,7 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
>  	int i;
>  	vm_fault_t ret = 0;
>  	struct page **pages;
> -	unsigned long mmun_start;	/* For mmu_notifiers */
> -	unsigned long mmun_end;		/* For mmu_notifiers */
> +	struct mmu_notifier_range range;
>  
>  	pages = kmalloc_array(HPAGE_PMD_NR, sizeof(struct page *),
>  			      GFP_KERNEL);
> @@ -1183,9 +1182,9 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
>  		cond_resched();
>  	}
>  
> -	mmun_start = haddr;
> -	mmun_end   = haddr + HPAGE_PMD_SIZE;
> -	mmu_notifier_invalidate_range_start(vma->vm_mm, mmun_start, mmun_end);
> +	mmu_notifier_range_init(&range, vma->vm_mm, haddr,
> +				haddr + HPAGE_PMD_SIZE);
> +	mmu_notifier_invalidate_range_start(&range);
>  
>  	vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd);
>  	if (unlikely(!pmd_same(*vmf->pmd, orig_pmd)))
> @@ -1230,8 +1229,7 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
>  	 * No need to double call mmu_notifier->invalidate_range() callback as
>  	 * the above pmdp_huge_clear_flush_notify() did already call it.
>  	 */
> -	mmu_notifier_invalidate_range_only_end(vma->vm_mm, mmun_start,
> -						mmun_end);
> +	mmu_notifier_invalidate_range_only_end(&range);
>  
>  	ret |= VM_FAULT_WRITE;
>  	put_page(page);
> @@ -1241,7 +1239,7 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
>  
>  out_free_pages:
>  	spin_unlock(vmf->ptl);
> -	mmu_notifier_invalidate_range_end(vma->vm_mm, mmun_start, mmun_end);
> +	mmu_notifier_invalidate_range_end(&range);
>  	for (i = 0; i < HPAGE_PMD_NR; i++) {
>  		memcg = (void *)page_private(pages[i]);
>  		set_page_private(pages[i], 0);
> @@ -1258,8 +1256,7 @@ vm_fault_t do_huge_pmd_wp_page(struct vm_fault *vmf, pmd_t orig_pmd)
>  	struct page *page = NULL, *new_page;
>  	struct mem_cgroup *memcg;
>  	unsigned long haddr = vmf->address & HPAGE_PMD_MASK;
> -	unsigned long mmun_start;	/* For mmu_notifiers */
> -	unsigned long mmun_end;		/* For mmu_notifiers */
> +	struct mmu_notifier_range range;
>  	gfp_t huge_gfp;			/* for allocation and charge */
>  	vm_fault_t ret = 0;
>  
> @@ -1349,9 +1346,9 @@ vm_fault_t do_huge_pmd_wp_page(struct vm_fault *vmf, pmd_t orig_pmd)
>  				    vma, HPAGE_PMD_NR);
>  	__SetPageUptodate(new_page);
>  
> -	mmun_start = haddr;
> -	mmun_end   = haddr + HPAGE_PMD_SIZE;
> -	mmu_notifier_invalidate_range_start(vma->vm_mm, mmun_start, mmun_end);
> +	mmu_notifier_range_init(&range, vma->vm_mm, haddr,
> +				haddr + HPAGE_PMD_SIZE);
> +	mmu_notifier_invalidate_range_start(&range);
>  
>  	spin_lock(vmf->ptl);
>  	if (page)
> @@ -1386,8 +1383,7 @@ vm_fault_t do_huge_pmd_wp_page(struct vm_fault *vmf, pmd_t orig_pmd)
>  	 * No need to double call mmu_notifier->invalidate_range() callback as
>  	 * the above pmdp_huge_clear_flush_notify() did already call it.
>  	 */
> -	mmu_notifier_invalidate_range_only_end(vma->vm_mm, mmun_start,
> -					       mmun_end);
> +	mmu_notifier_invalidate_range_only_end(&range);
>  out:
>  	return ret;
>  out_unlock:
> @@ -2028,14 +2024,15 @@ void __split_huge_pud(struct vm_area_struct *vma, pud_t *pud,
>  		unsigned long address)
>  {
>  	spinlock_t *ptl;
> -	struct mm_struct *mm = vma->vm_mm;
> -	unsigned long haddr = address & HPAGE_PUD_MASK;
> +	struct mmu_notifier_range range;
>  
> -	mmu_notifier_invalidate_range_start(mm, haddr, haddr + HPAGE_PUD_SIZE);
> -	ptl = pud_lock(mm, pud);
> +	mmu_notifier_range_init(&range, vma->vm_mm, address & HPAGE_PUD_MASK,
> +				(address & HPAGE_PUD_MASK) + HPAGE_PUD_SIZE);
> +	mmu_notifier_invalidate_range_start(&range);
> +	ptl = pud_lock(vma->vm_mm, pud);
>  	if (unlikely(!pud_trans_huge(*pud) && !pud_devmap(*pud)))
>  		goto out;
> -	__split_huge_pud_locked(vma, pud, haddr);
> +	__split_huge_pud_locked(vma, pud, range.start);
>  
>  out:
>  	spin_unlock(ptl);
> @@ -2043,8 +2040,7 @@ void __split_huge_pud(struct vm_area_struct *vma, pud_t *pud,
>  	 * No need to double call mmu_notifier->invalidate_range() callback as
>  	 * the above pudp_huge_clear_flush_notify() did already call it.
>  	 */
> -	mmu_notifier_invalidate_range_only_end(mm, haddr, haddr +
> -					       HPAGE_PUD_SIZE);
> +	mmu_notifier_invalidate_range_only_end(&range);
>  }
>  #endif /* CONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD */
>  
> @@ -2244,11 +2240,12 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
>  		unsigned long address, bool freeze, struct page *page)
>  {
>  	spinlock_t *ptl;
> -	struct mm_struct *mm = vma->vm_mm;
> -	unsigned long haddr = address & HPAGE_PMD_MASK;
> +	struct mmu_notifier_range range;
>  
> -	mmu_notifier_invalidate_range_start(mm, haddr, haddr + HPAGE_PMD_SIZE);
> -	ptl = pmd_lock(mm, pmd);
> +	mmu_notifier_range_init(&range, vma->vm_mm, address & HPAGE_PMD_MASK,
> +				(address & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE);
> +	mmu_notifier_invalidate_range_start(&range);
> +	ptl = pmd_lock(vma->vm_mm, pmd);
>  
>  	/*
>  	 * If caller asks to setup a migration entries, we need a page to check
> @@ -2264,7 +2261,7 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
>  			clear_page_mlock(page);
>  	} else if (!(pmd_devmap(*pmd) || is_pmd_migration_entry(*pmd)))
>  		goto out;
> -	__split_huge_pmd_locked(vma, pmd, haddr, freeze);
> +	__split_huge_pmd_locked(vma, pmd, range.start, freeze);
>  out:
>  	spin_unlock(ptl);
>  	/*
> @@ -2280,8 +2277,7 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
>  	 *     any further changes to individual pte will notify. So no need
>  	 *     to call mmu_notifier->invalidate_range()
>  	 */
> -	mmu_notifier_invalidate_range_only_end(mm, haddr, haddr +
> -					       HPAGE_PMD_SIZE);
> +	mmu_notifier_invalidate_range_only_end(&range);
>  }
>  
>  void split_huge_pmd_address(struct vm_area_struct *vma, unsigned long address,
> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> index 705a3e9cc910..e7c179cbcd75 100644
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -3239,16 +3239,16 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
>  	int cow;
>  	struct hstate *h = hstate_vma(vma);
>  	unsigned long sz = huge_page_size(h);
> -	unsigned long mmun_start;	/* For mmu_notifiers */
> -	unsigned long mmun_end;		/* For mmu_notifiers */
> +	struct mmu_notifier_range range;
>  	int ret = 0;
>  
>  	cow = (vma->vm_flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE;
>  
> -	mmun_start = vma->vm_start;
> -	mmun_end = vma->vm_end;
> -	if (cow)
> -		mmu_notifier_invalidate_range_start(src, mmun_start, mmun_end);
> +	if (cow) {
> +		mmu_notifier_range_init(&range, src, vma->vm_start,
> +					vma->vm_end);
> +		mmu_notifier_invalidate_range_start(&range);
> +	}
>  
>  	for (addr = vma->vm_start; addr < vma->vm_end; addr += sz) {
>  		spinlock_t *src_ptl, *dst_ptl;
> @@ -3324,7 +3324,7 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
>  	}
>  
>  	if (cow)
> -		mmu_notifier_invalidate_range_end(src, mmun_start, mmun_end);
> +		mmu_notifier_invalidate_range_end(&range);
>  
>  	return ret;
>  }
> @@ -3341,8 +3341,7 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
>  	struct page *page;
>  	struct hstate *h = hstate_vma(vma);
>  	unsigned long sz = huge_page_size(h);
> -	unsigned long mmun_start = start;	/* For mmu_notifiers */
> -	unsigned long mmun_end   = end;		/* For mmu_notifiers */
> +	struct mmu_notifier_range range;
>  
>  	WARN_ON(!is_vm_hugetlb_page(vma));
>  	BUG_ON(start & ~huge_page_mask(h));
> @@ -3358,8 +3357,9 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
>  	/*
>  	 * If sharing possible, alert mmu notifiers of worst case.
>  	 */
> -	adjust_range_if_pmd_sharing_possible(vma, &mmun_start, &mmun_end);
> -	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
> +	mmu_notifier_range_init(&range, mm, start, end);
> +	adjust_range_if_pmd_sharing_possible(vma, &range.start, &range.end);
> +	mmu_notifier_invalidate_range_start(&range);
>  	address = start;
>  	for (; address < end; address += sz) {
>  		ptep = huge_pte_offset(mm, address, sz);
> @@ -3427,7 +3427,7 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
>  		if (ref_page)
>  			break;
>  	}
> -	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
> +	mmu_notifier_invalidate_range_end(&range);
>  	tlb_end_vma(tlb, vma);
>  }
>  
> @@ -3545,9 +3545,8 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
>  	struct page *old_page, *new_page;
>  	int outside_reserve = 0;
>  	vm_fault_t ret = 0;
> -	unsigned long mmun_start;	/* For mmu_notifiers */
> -	unsigned long mmun_end;		/* For mmu_notifiers */
>  	unsigned long haddr = address & huge_page_mask(h);
> +	struct mmu_notifier_range range;
>  
>  	pte = huge_ptep_get(ptep);
>  	old_page = pte_page(pte);
> @@ -3626,9 +3625,8 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
>  	__SetPageUptodate(new_page);
>  	set_page_huge_active(new_page);
>  
> -	mmun_start = haddr;
> -	mmun_end = mmun_start + huge_page_size(h);
> -	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
> +	mmu_notifier_range_init(&range, mm, haddr, haddr + huge_page_size(h));
> +	mmu_notifier_invalidate_range_start(&range);
>  
>  	/*
>  	 * Retake the page table lock to check for racing updates
> @@ -3641,7 +3639,7 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
>  
>  		/* Break COW */
>  		huge_ptep_clear_flush(vma, haddr, ptep);
> -		mmu_notifier_invalidate_range(mm, mmun_start, mmun_end);
> +		mmu_notifier_invalidate_range(mm, range.start, range.end);
>  		set_huge_pte_at(mm, haddr, ptep,
>  				make_huge_pte(vma, new_page, 1));
>  		page_remove_rmap(old_page, true);
> @@ -3650,7 +3648,7 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
>  		new_page = old_page;
>  	}
>  	spin_unlock(ptl);
> -	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
> +	mmu_notifier_invalidate_range_end(&range);
>  out_release_all:
>  	restore_reserve_on_error(h, vma, haddr, new_page);
>  	put_page(new_page);
> @@ -4339,21 +4337,21 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
>  	pte_t pte;
>  	struct hstate *h = hstate_vma(vma);
>  	unsigned long pages = 0;
> -	unsigned long f_start = start;
> -	unsigned long f_end = end;
>  	bool shared_pmd = false;
> +	struct mmu_notifier_range range;
>  
>  	/*
>  	 * In the case of shared PMDs, the area to flush could be beyond
> -	 * start/end.  Set f_start/f_end to cover the maximum possible
> +	 * start/end.  Set range.start/range.end to cover the maximum possible
>  	 * range if PMD sharing is possible.
>  	 */
> -	adjust_range_if_pmd_sharing_possible(vma, &f_start, &f_end);
> +	mmu_notifier_range_init(&range, mm, start, end);
> +	adjust_range_if_pmd_sharing_possible(vma, &range.start, &range.end);
>  
>  	BUG_ON(address >= end);
> -	flush_cache_range(vma, f_start, f_end);
> +	flush_cache_range(vma, range.start, range.end);
>  
> -	mmu_notifier_invalidate_range_start(mm, f_start, f_end);
> +	mmu_notifier_invalidate_range_start(&range);
>  	i_mmap_lock_write(vma->vm_file->f_mapping);
>  	for (; address < end; address += huge_page_size(h)) {
>  		spinlock_t *ptl;
> @@ -4404,7 +4402,7 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
>  	 * did unshare a page of pmds, flush the range corresponding to the pud.
>  	 */
>  	if (shared_pmd)
> -		flush_hugetlb_tlb_range(vma, f_start, f_end);
> +		flush_hugetlb_tlb_range(vma, range.start, range.end);
>  	else
>  		flush_hugetlb_tlb_range(vma, start, end);
>  	/*
> @@ -4414,7 +4412,7 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
>  	 * See Documentation/vm/mmu_notifier.rst
>  	 */
>  	i_mmap_unlock_write(vma->vm_file->f_mapping);
> -	mmu_notifier_invalidate_range_end(mm, f_start, f_end);
> +	mmu_notifier_invalidate_range_end(&range);
>  
>  	return pages << h->order;
>  }
> diff --git a/mm/khugepaged.c b/mm/khugepaged.c
> index 8e2ff195ecb3..7736f6c37f19 100644
> --- a/mm/khugepaged.c
> +++ b/mm/khugepaged.c
> @@ -944,8 +944,7 @@ static void collapse_huge_page(struct mm_struct *mm,
>  	int isolated = 0, result = 0;
>  	struct mem_cgroup *memcg;
>  	struct vm_area_struct *vma;
> -	unsigned long mmun_start;	/* For mmu_notifiers */
> -	unsigned long mmun_end;		/* For mmu_notifiers */
> +	struct mmu_notifier_range range;
>  	gfp_t gfp;
>  
>  	VM_BUG_ON(address & ~HPAGE_PMD_MASK);
> @@ -1017,9 +1016,8 @@ static void collapse_huge_page(struct mm_struct *mm,
>  	pte = pte_offset_map(pmd, address);
>  	pte_ptl = pte_lockptr(mm, pmd);
>  
> -	mmun_start = address;
> -	mmun_end   = address + HPAGE_PMD_SIZE;
> -	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
> +	mmu_notifier_range_init(&range, mm, address, address + HPAGE_PMD_SIZE);
> +	mmu_notifier_invalidate_range_start(&range);
>  	pmd_ptl = pmd_lock(mm, pmd); /* probably unnecessary */
>  	/*
>  	 * After this gup_fast can't run anymore. This also removes
> @@ -1029,7 +1027,7 @@ static void collapse_huge_page(struct mm_struct *mm,
>  	 */
>  	_pmd = pmdp_collapse_flush(vma, address, pmd);
>  	spin_unlock(pmd_ptl);
> -	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
> +	mmu_notifier_invalidate_range_end(&range);
>  
>  	spin_lock(pte_ptl);
>  	isolated = __collapse_huge_page_isolate(vma, address, pte);
> diff --git a/mm/ksm.c b/mm/ksm.c
> index 5b0894b45ee5..6239d2df7a8e 100644
> --- a/mm/ksm.c
> +++ b/mm/ksm.c
> @@ -1042,8 +1042,7 @@ static int write_protect_page(struct vm_area_struct *vma, struct page *page,
>  	};
>  	int swapped;
>  	int err = -EFAULT;
> -	unsigned long mmun_start;	/* For mmu_notifiers */
> -	unsigned long mmun_end;		/* For mmu_notifiers */
> +	struct mmu_notifier_range range;
>  
>  	pvmw.address = page_address_in_vma(page, vma);
>  	if (pvmw.address == -EFAULT)
> @@ -1051,9 +1050,9 @@ static int write_protect_page(struct vm_area_struct *vma, struct page *page,
>  
>  	BUG_ON(PageTransCompound(page));
>  
> -	mmun_start = pvmw.address;
> -	mmun_end   = pvmw.address + PAGE_SIZE;
> -	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
> +	mmu_notifier_range_init(&range, mm, pvmw.address,
> +				pvmw.address + PAGE_SIZE);
> +	mmu_notifier_invalidate_range_start(&range);
>  
>  	if (!page_vma_mapped_walk(&pvmw))
>  		goto out_mn;
> @@ -1105,7 +1104,7 @@ static int write_protect_page(struct vm_area_struct *vma, struct page *page,
>  out_unlock:
>  	page_vma_mapped_walk_done(&pvmw);
>  out_mn:
> -	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
> +	mmu_notifier_invalidate_range_end(&range);
>  out:
>  	return err;
>  }
> @@ -1129,8 +1128,7 @@ static int replace_page(struct vm_area_struct *vma, struct page *page,
>  	spinlock_t *ptl;
>  	unsigned long addr;
>  	int err = -EFAULT;
> -	unsigned long mmun_start;	/* For mmu_notifiers */
> -	unsigned long mmun_end;		/* For mmu_notifiers */
> +	struct mmu_notifier_range range;
>  
>  	addr = page_address_in_vma(page, vma);
>  	if (addr == -EFAULT)
> @@ -1140,9 +1138,8 @@ static int replace_page(struct vm_area_struct *vma, struct page *page,
>  	if (!pmd)
>  		goto out;
>  
> -	mmun_start = addr;
> -	mmun_end   = addr + PAGE_SIZE;
> -	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
> +	mmu_notifier_range_init(&range, mm, addr, addr + PAGE_SIZE);
> +	mmu_notifier_invalidate_range_start(&range);
>  
>  	ptep = pte_offset_map_lock(mm, pmd, addr, &ptl);
>  	if (!pte_same(*ptep, orig_pte)) {
> @@ -1188,7 +1185,7 @@ static int replace_page(struct vm_area_struct *vma, struct page *page,
>  	pte_unmap_unlock(ptep, ptl);
>  	err = 0;
>  out_mn:
> -	mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
> +	mmu_notifier_invalidate_range_end(&range);
>  out:
>  	return err;
>  }
> diff --git a/mm/madvise.c b/mm/madvise.c
> index 6cb1ca93e290..21a7881a2db4 100644
> --- a/mm/madvise.c
> +++ b/mm/madvise.c
> @@ -458,29 +458,30 @@ static void madvise_free_page_range(struct mmu_gather *tlb,
>  static int madvise_free_single_vma(struct vm_area_struct *vma,
>  			unsigned long start_addr, unsigned long end_addr)
>  {
> -	unsigned long start, end;
>  	struct mm_struct *mm = vma->vm_mm;
> +	struct mmu_notifier_range range;
>  	struct mmu_gather tlb;
>  
>  	/* MADV_FREE works for only anon vma at the moment */
>  	if (!vma_is_anonymous(vma))
>  		return -EINVAL;
>  
> -	start = max(vma->vm_start, start_addr);
> -	if (start >= vma->vm_end)
> +	range.start = max(vma->vm_start, start_addr);
> +	if (range.start >= vma->vm_end)
>  		return -EINVAL;
> -	end = min(vma->vm_end, end_addr);
> -	if (end <= vma->vm_start)
> +	range.end = min(vma->vm_end, end_addr);
> +	if (range.end <= vma->vm_start)
>  		return -EINVAL;
> +	mmu_notifier_range_init(&range, mm, range.start, range.end);
>  
>  	lru_add_drain();
> -	tlb_gather_mmu(&tlb, mm, start, end);
> +	tlb_gather_mmu(&tlb, mm, range.start, range.end);
>  	update_hiwater_rss(mm);
>  
> -	mmu_notifier_invalidate_range_start(mm, start, end);
> -	madvise_free_page_range(&tlb, vma, start, end);
> -	mmu_notifier_invalidate_range_end(mm, start, end);
> -	tlb_finish_mmu(&tlb, start, end);
> +	mmu_notifier_invalidate_range_start(&range);
> +	madvise_free_page_range(&tlb, vma, range.start, range.end);
> +	mmu_notifier_invalidate_range_end(&range);
> +	tlb_finish_mmu(&tlb, range.start, range.end);
>  
>  	return 0;
>  }
> diff --git a/mm/memory.c b/mm/memory.c
> index 4ad2d293ddc2..574307f11464 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -973,8 +973,7 @@ int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
>  	unsigned long next;
>  	unsigned long addr = vma->vm_start;
>  	unsigned long end = vma->vm_end;
> -	unsigned long mmun_start;	/* For mmu_notifiers */
> -	unsigned long mmun_end;		/* For mmu_notifiers */
> +	struct mmu_notifier_range range;
>  	bool is_cow;
>  	int ret;
>  
> @@ -1008,11 +1007,11 @@ int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
>  	 * is_cow_mapping() returns true.
>  	 */
>  	is_cow = is_cow_mapping(vma->vm_flags);
> -	mmun_start = addr;
> -	mmun_end   = end;
> -	if (is_cow)
> -		mmu_notifier_invalidate_range_start(src_mm, mmun_start,
> -						    mmun_end);
> +
> +	if (is_cow) {
> +		mmu_notifier_range_init(&range, src_mm, addr, end);
> +		mmu_notifier_invalidate_range_start(&range);
> +	}
>  
>  	ret = 0;
>  	dst_pgd = pgd_offset(dst_mm, addr);
> @@ -1029,7 +1028,7 @@ int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
>  	} while (dst_pgd++, src_pgd++, addr = next, addr != end);
>  
>  	if (is_cow)
> -		mmu_notifier_invalidate_range_end(src_mm, mmun_start, mmun_end);
> +		mmu_notifier_invalidate_range_end(&range);
>  	return ret;
>  }
>  
> @@ -1332,12 +1331,13 @@ void unmap_vmas(struct mmu_gather *tlb,
>  		struct vm_area_struct *vma, unsigned long start_addr,
>  		unsigned long end_addr)
>  {
> -	struct mm_struct *mm = vma->vm_mm;
> +	struct mmu_notifier_range range;
>  
> -	mmu_notifier_invalidate_range_start(mm, start_addr, end_addr);
> +	mmu_notifier_range_init(&range, vma->vm_mm, start_addr, end_addr);
> +	mmu_notifier_invalidate_range_start(&range);
>  	for ( ; vma && vma->vm_start < end_addr; vma = vma->vm_next)
>  		unmap_single_vma(tlb, vma, start_addr, end_addr, NULL);
> -	mmu_notifier_invalidate_range_end(mm, start_addr, end_addr);
> +	mmu_notifier_invalidate_range_end(&range);
>  }
>  
>  /**
> @@ -1351,18 +1351,18 @@ void unmap_vmas(struct mmu_gather *tlb,
>  void zap_page_range(struct vm_area_struct *vma, unsigned long start,
>  		unsigned long size)
>  {
> -	struct mm_struct *mm = vma->vm_mm;
> +	struct mmu_notifier_range range;
>  	struct mmu_gather tlb;
> -	unsigned long end = start + size;
>  
>  	lru_add_drain();
> -	tlb_gather_mmu(&tlb, mm, start, end);
> -	update_hiwater_rss(mm);
> -	mmu_notifier_invalidate_range_start(mm, start, end);
> -	for ( ; vma && vma->vm_start < end; vma = vma->vm_next)
> -		unmap_single_vma(&tlb, vma, start, end, NULL);
> -	mmu_notifier_invalidate_range_end(mm, start, end);
> -	tlb_finish_mmu(&tlb, start, end);
> +	mmu_notifier_range_init(&range, vma->vm_mm, start, start + size);
> +	tlb_gather_mmu(&tlb, vma->vm_mm, start, range.end);
> +	update_hiwater_rss(vma->vm_mm);
> +	mmu_notifier_invalidate_range_start(&range);
> +	for ( ; vma && vma->vm_start < range.end; vma = vma->vm_next)
> +		unmap_single_vma(&tlb, vma, start, range.end, NULL);
> +	mmu_notifier_invalidate_range_end(&range);
> +	tlb_finish_mmu(&tlb, start, range.end);
>  }
>  
>  /**
> @@ -1377,17 +1377,17 @@ void zap_page_range(struct vm_area_struct *vma, unsigned long start,
>  static void zap_page_range_single(struct vm_area_struct *vma, unsigned long address,
>  		unsigned long size, struct zap_details *details)
>  {
> -	struct mm_struct *mm = vma->vm_mm;
> +	struct mmu_notifier_range range;
>  	struct mmu_gather tlb;
> -	unsigned long end = address + size;
>  
>  	lru_add_drain();
> -	tlb_gather_mmu(&tlb, mm, address, end);
> -	update_hiwater_rss(mm);
> -	mmu_notifier_invalidate_range_start(mm, address, end);
> -	unmap_single_vma(&tlb, vma, address, end, details);
> -	mmu_notifier_invalidate_range_end(mm, address, end);
> -	tlb_finish_mmu(&tlb, address, end);
> +	mmu_notifier_range_init(&range, vma->vm_mm, address, address + size);
> +	tlb_gather_mmu(&tlb, vma->vm_mm, address, range.end);
> +	update_hiwater_rss(vma->vm_mm);
> +	mmu_notifier_invalidate_range_start(&range);
> +	unmap_single_vma(&tlb, vma, address, range.end, details);
> +	mmu_notifier_invalidate_range_end(&range);
> +	tlb_finish_mmu(&tlb, address, range.end);
>  }
>  
>  /**
> @@ -2247,9 +2247,8 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
>  	struct page *new_page = NULL;
>  	pte_t entry;
>  	int page_copied = 0;
> -	const unsigned long mmun_start = vmf->address & PAGE_MASK;
> -	const unsigned long mmun_end = mmun_start + PAGE_SIZE;
>  	struct mem_cgroup *memcg;
> +	struct mmu_notifier_range range;
>  
>  	if (unlikely(anon_vma_prepare(vma)))
>  		goto oom;
> @@ -2272,7 +2271,9 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
>  
>  	__SetPageUptodate(new_page);
>  
> -	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
> +	mmu_notifier_range_init(&range, mm, vmf->address & PAGE_MASK,
> +				(vmf->address & PAGE_MASK) + PAGE_SIZE);
> +	mmu_notifier_invalidate_range_start(&range);
>  
>  	/*
>  	 * Re-check the pte - we dropped the lock
> @@ -2349,7 +2350,7 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
>  	 * No need to double call mmu_notifier->invalidate_range() callback as
>  	 * the above ptep_clear_flush_notify() did already call it.
>  	 */
> -	mmu_notifier_invalidate_range_only_end(mm, mmun_start, mmun_end);
> +	mmu_notifier_invalidate_range_only_end(&range);
>  	if (old_page) {
>  		/*
>  		 * Don't let another task, with possibly unlocked vma,
> @@ -4030,7 +4031,7 @@ int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address)
>  #endif /* __PAGETABLE_PMD_FOLDED */
>  
>  static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
> -			    unsigned long *start, unsigned long *end,
> +			    struct mmu_notifier_range *range,
>  			    pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp)
>  {
>  	pgd_t *pgd;
> @@ -4058,10 +4059,10 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
>  		if (!pmdpp)
>  			goto out;
>  
> -		if (start && end) {
> -			*start = address & PMD_MASK;
> -			*end = *start + PMD_SIZE;
> -			mmu_notifier_invalidate_range_start(mm, *start, *end);
> +		if (range) {
> +			mmu_notifier_range_init(range, mm, address & PMD_MASK,
> +					     (address & PMD_MASK) + PMD_SIZE);
> +			mmu_notifier_invalidate_range_start(range);
>  		}
>  		*ptlp = pmd_lock(mm, pmd);
>  		if (pmd_huge(*pmd)) {
> @@ -4069,17 +4070,17 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
>  			return 0;
>  		}
>  		spin_unlock(*ptlp);
> -		if (start && end)
> -			mmu_notifier_invalidate_range_end(mm, *start, *end);
> +		if (range)
> +			mmu_notifier_invalidate_range_end(range);
>  	}
>  
>  	if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd)))
>  		goto out;
>  
> -	if (start && end) {
> -		*start = address & PAGE_MASK;
> -		*end = *start + PAGE_SIZE;
> -		mmu_notifier_invalidate_range_start(mm, *start, *end);
> +	if (range) {
> +		range->start = address & PAGE_MASK;
> +		range->end = range->start + PAGE_SIZE;
> +		mmu_notifier_invalidate_range_start(range);
>  	}
>  	ptep = pte_offset_map_lock(mm, pmd, address, ptlp);
>  	if (!pte_present(*ptep))
> @@ -4088,8 +4089,8 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
>  	return 0;
>  unlock:
>  	pte_unmap_unlock(ptep, *ptlp);
> -	if (start && end)
> -		mmu_notifier_invalidate_range_end(mm, *start, *end);
> +	if (range)
> +		mmu_notifier_invalidate_range_end(range);
>  out:
>  	return -EINVAL;
>  }
> @@ -4101,20 +4102,20 @@ static inline int follow_pte(struct mm_struct *mm, unsigned long address,
>  
>  	/* (void) is needed to make gcc happy */
>  	(void) __cond_lock(*ptlp,
> -			   !(res = __follow_pte_pmd(mm, address, NULL, NULL,
> +			   !(res = __follow_pte_pmd(mm, address, NULL,
>  						    ptepp, NULL, ptlp)));
>  	return res;
>  }
>  
>  int follow_pte_pmd(struct mm_struct *mm, unsigned long address,
> -			     unsigned long *start, unsigned long *end,
> +		 	     struct mmu_notifier_range *range,
>  			     pte_t **ptepp, pmd_t **pmdpp, spinlock_t **ptlp)
>  {
>  	int res;
>  
>  	/* (void) is needed to make gcc happy */
>  	(void) __cond_lock(*ptlp,
> -			   !(res = __follow_pte_pmd(mm, address, start, end,
> +			   !(res = __follow_pte_pmd(mm, address, range,
>  						    ptepp, pmdpp, ptlp)));
>  	return res;
>  }
> diff --git a/mm/migrate.c b/mm/migrate.c
> index f7e4bfdc13b7..74f5b3208c05 100644
> --- a/mm/migrate.c
> +++ b/mm/migrate.c
> @@ -2303,6 +2303,7 @@ static int migrate_vma_collect_pmd(pmd_t *pmdp,
>   */
>  static void migrate_vma_collect(struct migrate_vma *migrate)
>  {
> +	struct mmu_notifier_range range;
>  	struct mm_walk mm_walk;
>  
>  	mm_walk.pmd_entry = migrate_vma_collect_pmd;
> @@ -2314,13 +2315,11 @@ static void migrate_vma_collect(struct migrate_vma *migrate)
>  	mm_walk.mm = migrate->vma->vm_mm;
>  	mm_walk.private = migrate;
>  
> -	mmu_notifier_invalidate_range_start(mm_walk.mm,
> -					    migrate->start,
> -					    migrate->end);
> +	mmu_notifier_range_init(&range, mm_walk.mm, migrate->start,
> +				migrate->end);
> +	mmu_notifier_invalidate_range_start(&range);
>  	walk_page_range(migrate->start, migrate->end, &mm_walk);
> -	mmu_notifier_invalidate_range_end(mm_walk.mm,
> -					  migrate->start,
> -					  migrate->end);
> +	mmu_notifier_invalidate_range_end(&range);
>  
>  	migrate->end = migrate->start + (migrate->npages << PAGE_SHIFT);
>  }
> @@ -2703,7 +2702,8 @@ static void migrate_vma_pages(struct migrate_vma *migrate)
>  	const unsigned long start = migrate->start;
>  	struct vm_area_struct *vma = migrate->vma;
>  	struct mm_struct *mm = vma->vm_mm;
> -	unsigned long addr, i, mmu_start;
> +	struct mmu_notifier_range range;
> +	unsigned long addr, i;
>  	bool notified = false;
>  
>  	for (i = 0, addr = start; i < npages; addr += PAGE_SIZE, i++) {
> @@ -2722,11 +2722,11 @@ static void migrate_vma_pages(struct migrate_vma *migrate)
>  				continue;
>  			}
>  			if (!notified) {
> -				mmu_start = addr;
>  				notified = true;
> -				mmu_notifier_invalidate_range_start(mm,
> -								mmu_start,
> -								migrate->end);
> +
> +				mmu_notifier_range_init(&range, mm, addr,
> +							migrate->end);
> +				mmu_notifier_invalidate_range_start(&range);
>  			}
>  			migrate_vma_insert_page(migrate, addr, newpage,
>  						&migrate->src[i],
> @@ -2767,8 +2767,7 @@ static void migrate_vma_pages(struct migrate_vma *migrate)
>  	 * did already call it.
>  	 */
>  	if (notified)
> -		mmu_notifier_invalidate_range_only_end(mm, mmu_start,
> -						       migrate->end);
> +		mmu_notifier_invalidate_range_only_end(&range);
>  }
>  
>  /*
> diff --git a/mm/mmu_notifier.c b/mm/mmu_notifier.c
> index 5f6665ae3ee2..4c52b3514c50 100644
> --- a/mm/mmu_notifier.c
> +++ b/mm/mmu_notifier.c
> @@ -174,28 +174,20 @@ void __mmu_notifier_change_pte(struct mm_struct *mm, unsigned long address,
>  	srcu_read_unlock(&srcu, id);
>  }
>  
> -int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
> -				  unsigned long start, unsigned long end,
> -				  bool blockable)
> +int __mmu_notifier_invalidate_range_start(struct mmu_notifier_range *range)
>  {
> -	struct mmu_notifier_range _range, *range = &_range;
>  	struct mmu_notifier *mn;
>  	int ret = 0;
>  	int id;
>  
> -	range->blockable = blockable;
> -	range->start = start;
> -	range->end = end;
> -	range->mm = mm;
> -
>  	id = srcu_read_lock(&srcu);
> -	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
> +	hlist_for_each_entry_rcu(mn, &range->mm->mmu_notifier_mm->list, hlist) {
>  		if (mn->ops->invalidate_range_start) {
>  			int _ret = mn->ops->invalidate_range_start(mn, range);
>  			if (_ret) {
>  				pr_info("%pS callback failed with %d in %sblockable context.\n",
>  						mn->ops->invalidate_range_start, _ret,
> -						!blockable ? "non-" : "");
> +						!range->blockable ? "non-" : "");
>  				ret = _ret;
>  			}
>  		}
> @@ -206,27 +198,14 @@ int __mmu_notifier_invalidate_range_start(struct mm_struct *mm,
>  }
>  EXPORT_SYMBOL_GPL(__mmu_notifier_invalidate_range_start);
>  
> -void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
> -					 unsigned long start,
> -					 unsigned long end,
> +void __mmu_notifier_invalidate_range_end(struct mmu_notifier_range *range,
>  					 bool only_end)
>  {
> -	struct mmu_notifier_range _range, *range = &_range;
>  	struct mmu_notifier *mn;
>  	int id;
>  
> -	/*
> -	 * The end call back will never be call if the start refused to go
> -	 * through because of blockable was false so here assume that we
> -	 * can block.
> -	 */
> -	range->blockable = true;
> -	range->start = start;
> -	range->end = end;
> -	range->mm = mm;
> -
>  	id = srcu_read_lock(&srcu);
> -	hlist_for_each_entry_rcu(mn, &mm->mmu_notifier_mm->list, hlist) {
> +	hlist_for_each_entry_rcu(mn, &range->mm->mmu_notifier_mm->list, hlist) {
>  		/*
>  		 * Call invalidate_range here too to avoid the need for the
>  		 * subsystem of having to register an invalidate_range_end
> @@ -241,7 +220,9 @@ void __mmu_notifier_invalidate_range_end(struct mm_struct *mm,
>  		 * already happen under page table lock.
>  		 */
>  		if (!only_end && mn->ops->invalidate_range)
> -			mn->ops->invalidate_range(mn, mm, start, end);
> +			mn->ops->invalidate_range(mn, range->mm,
> +						  range->start,
> +						  range->end);
>  		if (mn->ops->invalidate_range_end)
>  			mn->ops->invalidate_range_end(mn, range);
>  	}
> diff --git a/mm/mprotect.c b/mm/mprotect.c
> index 6d331620b9e5..36cb358db170 100644
> --- a/mm/mprotect.c
> +++ b/mm/mprotect.c
> @@ -167,11 +167,12 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
>  		pgprot_t newprot, int dirty_accountable, int prot_numa)
>  {
>  	pmd_t *pmd;
> -	struct mm_struct *mm = vma->vm_mm;
>  	unsigned long next;
>  	unsigned long pages = 0;
>  	unsigned long nr_huge_updates = 0;
> -	unsigned long mni_start = 0;
> +	struct mmu_notifier_range range;
> +
> +	range.start = 0;
>  
>  	pmd = pmd_offset(pud, addr);
>  	do {
> @@ -183,9 +184,9 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
>  			goto next;
>  
>  		/* invoke the mmu notifier if the pmd is populated */
> -		if (!mni_start) {
> -			mni_start = addr;
> -			mmu_notifier_invalidate_range_start(mm, mni_start, end);
> +		if (!range.start) {
> +			mmu_notifier_range_init(&range, vma->vm_mm, addr, end);
> +			mmu_notifier_invalidate_range_start(&range);
>  		}
>  
>  		if (is_swap_pmd(*pmd) || pmd_trans_huge(*pmd) || pmd_devmap(*pmd)) {
> @@ -214,8 +215,8 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
>  		cond_resched();
>  	} while (pmd++, addr = next, addr != end);
>  
> -	if (mni_start)
> -		mmu_notifier_invalidate_range_end(mm, mni_start, end);
> +	if (range.start)
> +		mmu_notifier_invalidate_range_end(&range);
>  
>  	if (nr_huge_updates)
>  		count_vm_numa_events(NUMA_HUGE_PTE_UPDATES, nr_huge_updates);
> diff --git a/mm/mremap.c b/mm/mremap.c
> index 7f9f9180e401..def01d86e36f 100644
> --- a/mm/mremap.c
> +++ b/mm/mremap.c
> @@ -197,16 +197,14 @@ unsigned long move_page_tables(struct vm_area_struct *vma,
>  		bool need_rmap_locks)
>  {
>  	unsigned long extent, next, old_end;
> +	struct mmu_notifier_range range;
>  	pmd_t *old_pmd, *new_pmd;
> -	unsigned long mmun_start;	/* For mmu_notifiers */
> -	unsigned long mmun_end;		/* For mmu_notifiers */
>  
>  	old_end = old_addr + len;
>  	flush_cache_range(vma, old_addr, old_end);
>  
> -	mmun_start = old_addr;
> -	mmun_end   = old_end;
> -	mmu_notifier_invalidate_range_start(vma->vm_mm, mmun_start, mmun_end);
> +	mmu_notifier_range_init(&range, vma->vm_mm, old_addr, old_end);
> +	mmu_notifier_invalidate_range_start(&range);
>  
>  	for (; old_addr < old_end; old_addr += extent, new_addr += extent) {
>  		cond_resched();
> @@ -247,7 +245,7 @@ unsigned long move_page_tables(struct vm_area_struct *vma,
>  			  new_pmd, new_addr, need_rmap_locks);
>  	}
>  
> -	mmu_notifier_invalidate_range_end(vma->vm_mm, mmun_start, mmun_end);
> +	mmu_notifier_invalidate_range_end(&range);
>  
>  	return len + old_addr - old_end;	/* how much done */
>  }
> diff --git a/mm/oom_kill.c b/mm/oom_kill.c
> index 6589f60d5018..1eea8b04f27a 100644
> --- a/mm/oom_kill.c
> +++ b/mm/oom_kill.c
> @@ -516,19 +516,20 @@ bool __oom_reap_task_mm(struct mm_struct *mm)
>  		 * count elevated without a good reason.
>  		 */
>  		if (vma_is_anonymous(vma) || !(vma->vm_flags & VM_SHARED)) {
> -			const unsigned long start = vma->vm_start;
> -			const unsigned long end = vma->vm_end;
> +			struct mmu_notifier_range range;
>  			struct mmu_gather tlb;
>  
> -			tlb_gather_mmu(&tlb, mm, start, end);
> -			if (mmu_notifier_invalidate_range_start_nonblock(mm, start, end)) {
> -				tlb_finish_mmu(&tlb, start, end);
> +			mmu_notifier_range_init(&range, mm, vma->vm_start,
> +						vma->vm_end);
> +			tlb_gather_mmu(&tlb, mm, range.start, range.end);
> +			if (mmu_notifier_invalidate_range_start_nonblock(&range)) {
> +				tlb_finish_mmu(&tlb, range.start, range.end);
>  				ret = false;
>  				continue;
>  			}
> -			unmap_page_range(&tlb, vma, start, end, NULL);
> -			mmu_notifier_invalidate_range_end(mm, start, end);
> -			tlb_finish_mmu(&tlb, start, end);
> +			unmap_page_range(&tlb, vma, range.start, range.end, NULL);
> +			mmu_notifier_invalidate_range_end(&range);
> +			tlb_finish_mmu(&tlb, range.start, range.end);
>  		}
>  	}
>  
> diff --git a/mm/rmap.c b/mm/rmap.c
> index 85b7f9423352..c75f72f6fe0e 100644
> --- a/mm/rmap.c
> +++ b/mm/rmap.c
> @@ -889,15 +889,17 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma,
>  		.address = address,
>  		.flags = PVMW_SYNC,
>  	};
> -	unsigned long start = address, end;
> +	struct mmu_notifier_range range;
>  	int *cleaned = arg;
>  
>  	/*
>  	 * We have to assume the worse case ie pmd for invalidation. Note that
>  	 * the page can not be free from this function.
>  	 */
> -	end = min(vma->vm_end, start + (PAGE_SIZE << compound_order(page)));
> -	mmu_notifier_invalidate_range_start(vma->vm_mm, start, end);
> +	mmu_notifier_range_init(&range, vma->vm_mm, address,
> +				min(vma->vm_end, address +
> +				    (PAGE_SIZE << compound_order(page))));
> +	mmu_notifier_invalidate_range_start(&range);
>  
>  	while (page_vma_mapped_walk(&pvmw)) {
>  		unsigned long cstart;
> @@ -949,7 +951,7 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma,
>  			(*cleaned)++;
>  	}
>  
> -	mmu_notifier_invalidate_range_end(vma->vm_mm, start, end);
> +	mmu_notifier_invalidate_range_end(&range);
>  
>  	return true;
>  }
> @@ -1345,7 +1347,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
>  	pte_t pteval;
>  	struct page *subpage;
>  	bool ret = true;
> -	unsigned long start = address, end;
> +	struct mmu_notifier_range range;
>  	enum ttu_flags flags = (enum ttu_flags)arg;
>  
>  	/* munlock has nothing to gain from examining un-locked vmas */
> @@ -1369,15 +1371,18 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
>  	 * Note that the page can not be free in this function as call of
>  	 * try_to_unmap() must hold a reference on the page.
>  	 */
> -	end = min(vma->vm_end, start + (PAGE_SIZE << compound_order(page)));
> +	mmu_notifier_range_init(&range, vma->vm_mm, vma->vm_start,
> +				min(vma->vm_end, vma->vm_start +
> +				    (PAGE_SIZE << compound_order(page))));
>  	if (PageHuge(page)) {
>  		/*
>  		 * If sharing is possible, start and end will be adjusted
>  		 * accordingly.
>  		 */
> -		adjust_range_if_pmd_sharing_possible(vma, &start, &end);
> +		adjust_range_if_pmd_sharing_possible(vma, &range.start,
> +						     &range.end);
>  	}
> -	mmu_notifier_invalidate_range_start(vma->vm_mm, start, end);
> +	mmu_notifier_invalidate_range_start(&range);
>  
>  	while (page_vma_mapped_walk(&pvmw)) {
>  #ifdef CONFIG_ARCH_ENABLE_THP_MIGRATION
> @@ -1428,9 +1433,10 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
>  				 * we must flush them all.  start/end were
>  				 * already adjusted above to cover this range.
>  				 */
> -				flush_cache_range(vma, start, end);
> -				flush_tlb_range(vma, start, end);
> -				mmu_notifier_invalidate_range(mm, start, end);
> +				flush_cache_range(vma, range.start, range.end);
> +				flush_tlb_range(vma, range.start, range.end);
> +				mmu_notifier_invalidate_range(mm, range.start,
> +							      range.end);
>  
>  				/*
>  				 * The ref count of the PMD page was dropped
> @@ -1650,7 +1656,7 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
>  		put_page(page);
>  	}
>  
> -	mmu_notifier_invalidate_range_end(vma->vm_mm, start, end);
> +	mmu_notifier_invalidate_range_end(&range);
>  
>  	return ret;
>  }
> -- 
> 2.17.2
> 
> 
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

[PATCH v2 3/3] mm/mmu_notifier: contextual information for event triggering invalidation v2

[jglisse]

From: Jérôme Glisse <jglisse@redhat.com>

CPU page table update can happens for many reasons, not only as a result
of a syscall (munmap(), mprotect(), mremap(), madvise(), ...) but also
as a result of kernel activities (memory compression, reclaim, migration,
...).

Users of mmu notifier API track changes to the CPU page table and take
specific action for them. While current API only provide range of virtual
address affected by the change, not why the changes is happening.

This patchset adds event information so that users of mmu notifier can
differentiate among broad category:
    - UNMAP: munmap() or mremap()
    - CLEAR: page table is cleared (migration, compaction, reclaim, ...)
    - PROTECTION_VMA: change in access protections for the range
    - PROTECTION_PAGE: change in access protections for page in the range
    - SOFT_DIRTY: soft dirtyness tracking

Being able to identify munmap() and mremap() from other reasons why the
page table is cleared is important to allow user of mmu notifier to
update their own internal tracking structure accordingly (on munmap or
mremap it is not longer needed to track range of virtual address as it
becomes invalid).

Changes since v1:
    - use mmu_notifier_range_init() helper to to optimize out the case
      when mmu notifier is not enabled
    - use kernel doc format for describing the enum values

Signed-off-by: Jérôme Glisse <jglisse@redhat.com>
Acked-by: Christian König <christian.koenig@amd.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Cc: Ross Zwisler <zwisler@kernel.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Felix Kuehling <felix.kuehling@amd.com>
Cc: Ralph Campbell <rcampbell@nvidia.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: kvm@vger.kernel.org
Cc: linux-rdma@vger.kernel.org
Cc: linux-fsdevel@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org
---
 fs/dax.c                     |  7 +++++++
 fs/proc/task_mmu.c           |  3 ++-
 include/linux/mmu_notifier.h | 35 +++++++++++++++++++++++++++++++++--
 kernel/events/uprobes.c      |  3 ++-
 mm/huge_memory.c             | 12 ++++++++----
 mm/hugetlb.c                 | 10 ++++++----
 mm/khugepaged.c              |  3 ++-
 mm/ksm.c                     |  6 ++++--
 mm/madvise.c                 |  3 ++-
 mm/memory.c                  | 18 ++++++++++++------
 mm/migrate.c                 |  5 +++--
 mm/mprotect.c                |  3 ++-
 mm/mremap.c                  |  3 ++-
 mm/oom_kill.c                |  2 +-
 mm/rmap.c                    |  6 ++++--
 15 files changed, 90 insertions(+), 29 deletions(-)

diff --git a/fs/dax.c b/fs/dax.c
index 874085bacaf5..6056b03a1626 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -768,6 +768,13 @@ static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index,
 
 		address = pgoff_address(index, vma);
 
+		/*
+		 * All the field are populated by follow_pte_pmd() except
+		 * the event field.
+		 */
+		mmu_notifier_range_init(&range, NULL, 0, -1UL,
+					MMU_NOTIFY_PROTECTION_PAGE);
+
 		/*
 		 * Note because we provide start/end to follow_pte_pmd it will
 		 * call mmu_notifier_invalidate_range_start() on our behalf
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index b3ddceb003bc..f68a9ebb0218 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -1141,7 +1141,8 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf,
 				break;
 			}
 
-			mmu_notifier_range_init(&range, mm, 0, -1UL);
+			mmu_notifier_range_init(&range, mm, 0, -1UL,
+						MMU_NOTIFY_SOFT_DIRTY);
 			mmu_notifier_invalidate_range_start(&range);
 		}
 		walk_page_range(0, mm->highest_vm_end, &clear_refs_walk);
diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h
index 39b06772427f..d249e24acea5 100644
--- a/include/linux/mmu_notifier.h
+++ b/include/linux/mmu_notifier.h
@@ -25,10 +25,39 @@ struct mmu_notifier_mm {
 	spinlock_t lock;
 };
 
+/**
+ * enum mmu_notifier_event - reason for the mmu notifier callback
+ * @MMU_NOTIFY_UNMAP: either munmap() that unmap the range or a mremap() that
+ * move the range
+ *
+ * @MMU_NOTIFY_CLEAR: clear page table entry (many reasons for this like
+ * madvise() or replacing a page by another one, ...).
+ *
+ * @MMU_NOTIFY_PROTECTION_VMA: update is due to protection change for the range
+ * ie using the vma access permission (vm_page_prot) to update the whole range
+ * is enough no need to inspect changes to the CPU page table (mprotect()
+ * syscall)
+ *
+ * @MMU_NOTIFY_PROTECTION_PAGE: update is due to change in read/write flag for
+ * pages in the range so to mirror those changes the user must inspect the CPU
+ * page table (from the end callback).
+ *
+ * @MMU_NOTIFY_SOFT_DIRTY: soft dirty accounting (still same page and same
+ * access flags)
+ */
+enum mmu_notifier_event {
+	MMU_NOTIFY_UNMAP = 0,
+	MMU_NOTIFY_CLEAR,
+	MMU_NOTIFY_PROTECTION_VMA,
+	MMU_NOTIFY_PROTECTION_PAGE,
+	MMU_NOTIFY_SOFT_DIRTY,
+};
+
 struct mmu_notifier_range {
 	struct mm_struct *mm;
 	unsigned long start;
 	unsigned long end;
+	enum mmu_notifier_event event;
 	bool blockable;
 };
 
@@ -320,11 +349,13 @@ static inline void mmu_notifier_mm_destroy(struct mm_struct *mm)
 static inline void mmu_notifier_range_init(struct mmu_notifier_range *range,
 					   struct mm_struct *mm,
 					   unsigned long start,
-					   unsigned long end)
+					   unsigned long end,
+					   enum mmu_notifier_event event)
 {
 	range->mm = mm;
 	range->start = start;
 	range->end = end;
+	range->event = event;
 }
 
 #define ptep_clear_flush_young_notify(__vma, __address, __ptep)		\
@@ -453,7 +484,7 @@ static inline void _mmu_notifier_range_init(struct mmu_notifier_range *range,
 	range->end = end;
 }
 
-#define mmu_notifier_range_init(range, mm, start, end) \
+#define mmu_notifier_range_init(range, mm, start, end, event) \
 	_mmu_notifier_range_init(range, start, end)
 
 
diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c
index 1fc8a93709c3..a70c3204f25d 100644
--- a/kernel/events/uprobes.c
+++ b/kernel/events/uprobes.c
@@ -174,7 +174,8 @@ static int __replace_page(struct vm_area_struct *vma, unsigned long addr,
 	struct mmu_notifier_range range;
 	struct mem_cgroup *memcg;
 
-	mmu_notifier_range_init(&range, mm, addr, addr + PAGE_SIZE);
+	mmu_notifier_range_init(&range, mm, addr, addr + PAGE_SIZE,
+				MMU_NOTIFY_CLEAR);
 
 	VM_BUG_ON_PAGE(PageTransHuge(old_page), old_page);
 
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index c1d3ce809416..b8d9029890c5 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1183,7 +1183,8 @@ static vm_fault_t do_huge_pmd_wp_page_fallback(struct vm_fault *vmf,
 	}
 
 	mmu_notifier_range_init(&range, vma->vm_mm, haddr,
-				haddr + HPAGE_PMD_SIZE);
+				haddr + HPAGE_PMD_SIZE,
+				MMU_NOTIFY_CLEAR);
 	mmu_notifier_invalidate_range_start(&range);
 
 	vmf->ptl = pmd_lock(vma->vm_mm, vmf->pmd);
@@ -1347,7 +1348,8 @@ vm_fault_t do_huge_pmd_wp_page(struct vm_fault *vmf, pmd_t orig_pmd)
 	__SetPageUptodate(new_page);
 
 	mmu_notifier_range_init(&range, vma->vm_mm, haddr,
-				haddr + HPAGE_PMD_SIZE);
+				haddr + HPAGE_PMD_SIZE,
+				MMU_NOTIFY_CLEAR);
 	mmu_notifier_invalidate_range_start(&range);
 
 	spin_lock(vmf->ptl);
@@ -2027,7 +2029,8 @@ void __split_huge_pud(struct vm_area_struct *vma, pud_t *pud,
 	struct mmu_notifier_range range;
 
 	mmu_notifier_range_init(&range, vma->vm_mm, address & HPAGE_PUD_MASK,
-				(address & HPAGE_PUD_MASK) + HPAGE_PUD_SIZE);
+				(address & HPAGE_PUD_MASK) + HPAGE_PUD_SIZE,
+				MMU_NOTIFY_CLEAR);
 	mmu_notifier_invalidate_range_start(&range);
 	ptl = pud_lock(vma->vm_mm, pud);
 	if (unlikely(!pud_trans_huge(*pud) && !pud_devmap(*pud)))
@@ -2243,7 +2246,8 @@ void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd,
 	struct mmu_notifier_range range;
 
 	mmu_notifier_range_init(&range, vma->vm_mm, address & HPAGE_PMD_MASK,
-				(address & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE);
+				(address & HPAGE_PMD_MASK) + HPAGE_PMD_SIZE,
+				MMU_NOTIFY_CLEAR);
 	mmu_notifier_invalidate_range_start(&range);
 	ptl = pmd_lock(vma->vm_mm, pmd);
 
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index e7c179cbcd75..fd1395918bb3 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -3246,7 +3246,7 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 
 	if (cow) {
 		mmu_notifier_range_init(&range, src, vma->vm_start,
-					vma->vm_end);
+					vma->vm_end, MMU_NOTIFY_CLEAR);
 		mmu_notifier_invalidate_range_start(&range);
 	}
 
@@ -3357,7 +3357,7 @@ void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
 	/*
 	 * If sharing possible, alert mmu notifiers of worst case.
 	 */
-	mmu_notifier_range_init(&range, mm, start, end);
+	mmu_notifier_range_init(&range, mm, start, end, MMU_NOTIFY_CLEAR);
 	adjust_range_if_pmd_sharing_possible(vma, &range.start, &range.end);
 	mmu_notifier_invalidate_range_start(&range);
 	address = start;
@@ -3625,7 +3625,8 @@ static vm_fault_t hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
 	__SetPageUptodate(new_page);
 	set_page_huge_active(new_page);
 
-	mmu_notifier_range_init(&range, mm, haddr, haddr + huge_page_size(h));
+	mmu_notifier_range_init(&range, mm, haddr, haddr + huge_page_size(h),
+				MMU_NOTIFY_CLEAR);
 	mmu_notifier_invalidate_range_start(&range);
 
 	/*
@@ -4345,7 +4346,8 @@ unsigned long hugetlb_change_protection(struct vm_area_struct *vma,
 	 * start/end.  Set range.start/range.end to cover the maximum possible
 	 * range if PMD sharing is possible.
 	 */
-	mmu_notifier_range_init(&range, mm, start, end);
+	mmu_notifier_range_init(&range, mm, start, end,
+				MMU_NOTIFY_PROTECTION_VMA);
 	adjust_range_if_pmd_sharing_possible(vma, &range.start, &range.end);
 
 	BUG_ON(address >= end);
diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index 7736f6c37f19..331dc2738f48 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -1016,7 +1016,8 @@ static void collapse_huge_page(struct mm_struct *mm,
 	pte = pte_offset_map(pmd, address);
 	pte_ptl = pte_lockptr(mm, pmd);
 
-	mmu_notifier_range_init(&range, mm, address, address + HPAGE_PMD_SIZE);
+	mmu_notifier_range_init(&range, mm, address, address + HPAGE_PMD_SIZE,
+				MMU_NOTIFY_CLEAR);
 	mmu_notifier_invalidate_range_start(&range);
 	pmd_ptl = pmd_lock(mm, pmd); /* probably unnecessary */
 	/*
diff --git a/mm/ksm.c b/mm/ksm.c
index 6239d2df7a8e..98a627c01eb0 100644
--- a/mm/ksm.c
+++ b/mm/ksm.c
@@ -1051,7 +1051,8 @@ static int write_protect_page(struct vm_area_struct *vma, struct page *page,
 	BUG_ON(PageTransCompound(page));
 
 	mmu_notifier_range_init(&range, mm, pvmw.address,
-				pvmw.address + PAGE_SIZE);
+				pvmw.address + PAGE_SIZE,
+				MMU_NOTIFY_CLEAR);
 	mmu_notifier_invalidate_range_start(&range);
 
 	if (!page_vma_mapped_walk(&pvmw))
@@ -1138,7 +1139,8 @@ static int replace_page(struct vm_area_struct *vma, struct page *page,
 	if (!pmd)
 		goto out;
 
-	mmu_notifier_range_init(&range, mm, addr, addr + PAGE_SIZE);
+	mmu_notifier_range_init(&range, mm, addr, addr + PAGE_SIZE,
+				MMU_NOTIFY_CLEAR);
 	mmu_notifier_invalidate_range_start(&range);
 
 	ptep = pte_offset_map_lock(mm, pmd, addr, &ptl);
diff --git a/mm/madvise.c b/mm/madvise.c
index 21a7881a2db4..d220ad7087ed 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -472,7 +472,8 @@ static int madvise_free_single_vma(struct vm_area_struct *vma,
 	range.end = min(vma->vm_end, end_addr);
 	if (range.end <= vma->vm_start)
 		return -EINVAL;
-	mmu_notifier_range_init(&range, mm, range.start, range.end);
+	mmu_notifier_range_init(&range, mm, range.start, range.end,
+				MMU_NOTIFY_CLEAR);
 
 	lru_add_drain();
 	tlb_gather_mmu(&tlb, mm, range.start, range.end);
diff --git a/mm/memory.c b/mm/memory.c
index 574307f11464..893e3bc4b895 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1009,7 +1009,8 @@ int copy_page_range(struct mm_struct *dst_mm, struct mm_struct *src_mm,
 	is_cow = is_cow_mapping(vma->vm_flags);
 
 	if (is_cow) {
-		mmu_notifier_range_init(&range, src_mm, addr, end);
+		mmu_notifier_range_init(&range, src_mm, addr, end,
+					MMU_NOTIFY_PROTECTION_PAGE);
 		mmu_notifier_invalidate_range_start(&range);
 	}
 
@@ -1333,7 +1334,8 @@ void unmap_vmas(struct mmu_gather *tlb,
 {
 	struct mmu_notifier_range range;
 
-	mmu_notifier_range_init(&range, vma->vm_mm, start_addr, end_addr);
+	mmu_notifier_range_init(&range, vma->vm_mm, start_addr,
+				end_addr, MMU_NOTIFY_UNMAP);
 	mmu_notifier_invalidate_range_start(&range);
 	for ( ; vma && vma->vm_start < end_addr; vma = vma->vm_next)
 		unmap_single_vma(tlb, vma, start_addr, end_addr, NULL);
@@ -1355,7 +1357,8 @@ void zap_page_range(struct vm_area_struct *vma, unsigned long start,
 	struct mmu_gather tlb;
 
 	lru_add_drain();
-	mmu_notifier_range_init(&range, vma->vm_mm, start, start + size);
+	mmu_notifier_range_init(&range, vma->vm_mm, start,
+				start + size, MMU_NOTIFY_CLEAR);
 	tlb_gather_mmu(&tlb, vma->vm_mm, start, range.end);
 	update_hiwater_rss(vma->vm_mm);
 	mmu_notifier_invalidate_range_start(&range);
@@ -1381,7 +1384,8 @@ static void zap_page_range_single(struct vm_area_struct *vma, unsigned long addr
 	struct mmu_gather tlb;
 
 	lru_add_drain();
-	mmu_notifier_range_init(&range, vma->vm_mm, address, address + size);
+	mmu_notifier_range_init(&range, vma->vm_mm, address,
+				address + size, MMU_NOTIFY_CLEAR);
 	tlb_gather_mmu(&tlb, vma->vm_mm, address, range.end);
 	update_hiwater_rss(vma->vm_mm);
 	mmu_notifier_invalidate_range_start(&range);
@@ -2272,7 +2276,8 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
 	__SetPageUptodate(new_page);
 
 	mmu_notifier_range_init(&range, mm, vmf->address & PAGE_MASK,
-				(vmf->address & PAGE_MASK) + PAGE_SIZE);
+				(vmf->address & PAGE_MASK) + PAGE_SIZE,
+				MMU_NOTIFY_CLEAR);
 	mmu_notifier_invalidate_range_start(&range);
 
 	/*
@@ -4061,7 +4066,8 @@ static int __follow_pte_pmd(struct mm_struct *mm, unsigned long address,
 
 		if (range) {
 			mmu_notifier_range_init(range, mm, address & PMD_MASK,
-					     (address & PMD_MASK) + PMD_SIZE);
+					     (address & PMD_MASK) + PMD_SIZE,
+					     range->event);
 			mmu_notifier_invalidate_range_start(range);
 		}
 		*ptlp = pmd_lock(mm, pmd);
diff --git a/mm/migrate.c b/mm/migrate.c
index 74f5b3208c05..f02bb4b22c1a 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -2316,7 +2316,7 @@ static void migrate_vma_collect(struct migrate_vma *migrate)
 	mm_walk.private = migrate;
 
 	mmu_notifier_range_init(&range, mm_walk.mm, migrate->start,
-				migrate->end);
+				migrate->end, MMU_NOTIFY_CLEAR);
 	mmu_notifier_invalidate_range_start(&range);
 	walk_page_range(migrate->start, migrate->end, &mm_walk);
 	mmu_notifier_invalidate_range_end(&range);
@@ -2725,7 +2725,8 @@ static void migrate_vma_pages(struct migrate_vma *migrate)
 				notified = true;
 
 				mmu_notifier_range_init(&range, mm, addr,
-							migrate->end);
+							migrate->end,
+							MMU_NOTIFY_CLEAR);
 				mmu_notifier_invalidate_range_start(&range);
 			}
 			migrate_vma_insert_page(migrate, addr, newpage,
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 36cb358db170..d5bbe5ca61ac 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -185,7 +185,8 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
 
 		/* invoke the mmu notifier if the pmd is populated */
 		if (!range.start) {
-			mmu_notifier_range_init(&range, vma->vm_mm, addr, end);
+			mmu_notifier_range_init(&range, vma->vm_mm, addr, end,
+						MMU_NOTIFY_PROTECTION_VMA);
 			mmu_notifier_invalidate_range_start(&range);
 		}
 
diff --git a/mm/mremap.c b/mm/mremap.c
index def01d86e36f..386e3c492f6e 100644
--- a/mm/mremap.c
+++ b/mm/mremap.c
@@ -203,7 +203,8 @@ unsigned long move_page_tables(struct vm_area_struct *vma,
 	old_end = old_addr + len;
 	flush_cache_range(vma, old_addr, old_end);
 
-	mmu_notifier_range_init(&range, vma->vm_mm, old_addr, old_end);
+	mmu_notifier_range_init(&range, vma->vm_mm, old_addr,
+				old_end, MMU_NOTIFY_UNMAP);
 	mmu_notifier_invalidate_range_start(&range);
 
 	for (; old_addr < old_end; old_addr += extent, new_addr += extent) {
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 1eea8b04f27a..4ac95032b898 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -520,7 +520,7 @@ bool __oom_reap_task_mm(struct mm_struct *mm)
 			struct mmu_gather tlb;
 
 			mmu_notifier_range_init(&range, mm, vma->vm_start,
-						vma->vm_end);
+						vma->vm_end, MMU_NOTIFY_CLEAR);
 			tlb_gather_mmu(&tlb, mm, range.start, range.end);
 			if (mmu_notifier_invalidate_range_start_nonblock(&range)) {
 				tlb_finish_mmu(&tlb, range.start, range.end);
diff --git a/mm/rmap.c b/mm/rmap.c
index c75f72f6fe0e..6ca019cdc789 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -898,7 +898,8 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma,
 	 */
 	mmu_notifier_range_init(&range, vma->vm_mm, address,
 				min(vma->vm_end, address +
-				    (PAGE_SIZE << compound_order(page))));
+				    (PAGE_SIZE << compound_order(page))),
+				MMU_NOTIFY_PROTECTION_PAGE);
 	mmu_notifier_invalidate_range_start(&range);
 
 	while (page_vma_mapped_walk(&pvmw)) {
@@ -1373,7 +1374,8 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
 	 */
 	mmu_notifier_range_init(&range, vma->vm_mm, vma->vm_start,
 				min(vma->vm_end, vma->vm_start +
-				    (PAGE_SIZE << compound_order(page))));
+				    (PAGE_SIZE << compound_order(page))),
+				MMU_NOTIFY_CLEAR);
 	if (PageHuge(page)) {
 		/*
 		 * If sharing is possible, start and end will be adjusted
-- 
2.17.2