1 files changed, 1266 insertions, 539 deletions

diff --git a/mm/madvise.c b/mm/madvise.c
index b6ea204d4e23..b617b1be0f53 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -19,6 +19,7 @@
 #include <linux/sched.h>
 #include <linux/sched/mm.h>
 #include <linux/mm_inline.h>
+#include <linux/mmu_context.h>
 #include <linux/string.h>
 #include <linux/uio.h>
 #include <linux/ksm.h>
@@ -28,7 +29,7 @@
 #include <linux/backing-dev.h>
 #include <linux/pagewalk.h>
 #include <linux/swap.h>
-#include <linux/swapops.h>
+#include <linux/leafops.h>
 #include <linux/shmem_fs.h>
 #include <linux/mmu_notifier.h>
 
@@ -37,37 +38,52 @@
 #include "internal.h"
 #include "swap.h"
 
+#define __MADV_SET_ANON_VMA_NAME (-1)
+
+/*
+ * Maximum number of attempts we make to install guard pages before we give up
+ * and return -ERESTARTNOINTR to have userspace try again.
+ */
+#define MAX_MADVISE_GUARD_RETRIES 3
+
 struct madvise_walk_private {
 	struct mmu_gather *tlb;
 	bool pageout;
 };
 
-/*
- * Any behaviour which results in changes to the vma->vm_flags needs to
- * take mmap_lock for writing. Others, which simply traverse vmas, need
- * to only take it for reading.
- */
-static int madvise_need_mmap_write(int behavior)
-{
-	switch (behavior) {
-	case MADV_REMOVE:
-	case MADV_WILLNEED:
-	case MADV_DONTNEED:
-	case MADV_DONTNEED_LOCKED:
-	case MADV_COLD:
-	case MADV_PAGEOUT:
-	case MADV_FREE:
-	case MADV_POPULATE_READ:
-	case MADV_POPULATE_WRITE:
-	case MADV_COLLAPSE:
-		return 0;
-	default:
-		/* be safe, default to 1. list exceptions explicitly */
-		return 1;
-	}
-}
+enum madvise_lock_mode {
+	MADVISE_NO_LOCK,
+	MADVISE_MMAP_READ_LOCK,
+	MADVISE_MMAP_WRITE_LOCK,
+	MADVISE_VMA_READ_LOCK,
+};
+
+struct madvise_behavior_range {
+	unsigned long start;
+	unsigned long end;
+};
+
+struct madvise_behavior {
+	struct mm_struct *mm;
+	int behavior;
+	struct mmu_gather *tlb;
+	enum madvise_lock_mode lock_mode;
+	struct anon_vma_name *anon_name;
+
+	/*
+	 * The range over which the behaviour is currently being applied. If
+	 * traversing multiple VMAs, this is updated for each.
+	 */
+	struct madvise_behavior_range range;
+	/* The VMA and VMA preceding it (if applicable) currently targeted. */
+	struct vm_area_struct *prev;
+	struct vm_area_struct *vma;
+	bool lock_dropped;
+};
 
 #ifdef CONFIG_ANON_VMA_NAME
+static int madvise_walk_vmas(struct madvise_behavior *madv_behavior);
+
 struct anon_vma_name *anon_vma_name_alloc(const char *name)
 {
 	struct anon_vma_name *anon_name;
@@ -93,7 +109,8 @@ void anon_vma_name_free(struct kref *kref)
 
 struct anon_vma_name *anon_vma_name(struct vm_area_struct *vma)
 {
-	mmap_assert_locked(vma->vm_mm);
+	if (!rwsem_is_locked(&vma->vm_mm->mmap_lock))
+		vma_assert_locked(vma);
 
 	return vma->anon_name;
 }
@@ -129,99 +146,80 @@ static int replace_anon_vma_name(struct vm_area_struct *vma,
 }
 #endif /* CONFIG_ANON_VMA_NAME */
 /*
- * Update the vm_flags on region of a vma, splitting it or merging it as
- * necessary.  Must be called with mmap_lock held for writing;
- * Caller should ensure anon_name stability by raising its refcount even when
- * anon_name belongs to a valid vma because this function might free that vma.
+ * Update the vm_flags or anon_name on region of a vma, splitting it or merging
+ * it as necessary. Must be called with mmap_lock held for writing.
  */
-static int madvise_update_vma(struct vm_area_struct *vma,
-			      struct vm_area_struct **prev, unsigned long start,
-			      unsigned long end, unsigned long new_flags,
-			      struct anon_vma_name *anon_name)
+static int madvise_update_vma(vm_flags_t new_flags,
+		struct madvise_behavior *madv_behavior)
 {
-	struct mm_struct *mm = vma->vm_mm;
-	int error;
-	pgoff_t pgoff;
-
-	if (new_flags == vma->vm_flags && anon_vma_name_eq(anon_vma_name(vma), anon_name)) {
-		*prev = vma;
+	struct vm_area_struct *vma = madv_behavior->vma;
+	struct madvise_behavior_range *range = &madv_behavior->range;
+	struct anon_vma_name *anon_name = madv_behavior->anon_name;
+	bool set_new_anon_name = madv_behavior->behavior == __MADV_SET_ANON_VMA_NAME;
+	VMA_ITERATOR(vmi, madv_behavior->mm, range->start);
+
+	if (new_flags == vma->vm_flags && (!set_new_anon_name ||
+			anon_vma_name_eq(anon_vma_name(vma), anon_name)))
 		return 0;
-	}
 
-	pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT);
-	*prev = vma_merge(mm, *prev, start, end, new_flags, vma->anon_vma,
-			  vma->vm_file, pgoff, vma_policy(vma),
-			  vma->vm_userfaultfd_ctx, anon_name);
-	if (*prev) {
-		vma = *prev;
-		goto success;
-	}
+	if (set_new_anon_name)
+		vma = vma_modify_name(&vmi, madv_behavior->prev, vma,
+			range->start, range->end, anon_name);
+	else
+		vma = vma_modify_flags(&vmi, madv_behavior->prev, vma,
+			range->start, range->end, &new_flags);
 
-	*prev = vma;
+	if (IS_ERR(vma))
+		return PTR_ERR(vma);
 
-	if (start != vma->vm_start) {
-		if (unlikely(mm->map_count >= sysctl_max_map_count))
-			return -ENOMEM;
-		error = __split_vma(mm, vma, start, 1);
-		if (error)
-			return error;
-	}
+	madv_behavior->vma = vma;
 
-	if (end != vma->vm_end) {
-		if (unlikely(mm->map_count >= sysctl_max_map_count))
-			return -ENOMEM;
-		error = __split_vma(mm, vma, end, 0);
-		if (error)
-			return error;
-	}
-
-success:
-	/*
-	 * vm_flags is protected by the mmap_lock held in write mode.
-	 */
-	vma->vm_flags = new_flags;
-	if (!vma->vm_file || vma_is_anon_shmem(vma)) {
-		error = replace_anon_vma_name(vma, anon_name);
-		if (error)
-			return error;
-	}
+	/* vm_flags is protected by the mmap_lock held in write mode. */
+	vma_start_write(vma);
+	vm_flags_reset(vma, new_flags);
+	if (set_new_anon_name)
+		return replace_anon_vma_name(vma, anon_name);
 
 	return 0;
 }
 
 #ifdef CONFIG_SWAP
 static int swapin_walk_pmd_entry(pmd_t *pmd, unsigned long start,
-	unsigned long end, struct mm_walk *walk)
+		unsigned long end, struct mm_walk *walk)
 {
 	struct vm_area_struct *vma = walk->private;
-	unsigned long index;
 	struct swap_iocb *splug = NULL;
+	pte_t *ptep = NULL;
+	spinlock_t *ptl;
+	unsigned long addr;
 
-	if (pmd_none_or_trans_huge_or_clear_bad(pmd))
-		return 0;
-
-	for (index = start; index != end; index += PAGE_SIZE) {
+	for (addr = start; addr < end; addr += PAGE_SIZE) {
 		pte_t pte;
-		swp_entry_t entry;
-		struct page *page;
-		spinlock_t *ptl;
-		pte_t *ptep;
+		softleaf_t entry;
+		struct folio *folio;
 
-		ptep = pte_offset_map_lock(vma->vm_mm, pmd, index, &ptl);
-		pte = *ptep;
-		pte_unmap_unlock(ptep, ptl);
+		if (!ptep++) {
+			ptep = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
+			if (!ptep)
+				break;
+		}
 
-		if (!is_swap_pte(pte))
-			continue;
-		entry = pte_to_swp_entry(pte);
-		if (unlikely(non_swap_entry(entry)))
+		pte = ptep_get(ptep);
+		entry = softleaf_from_pte(pte);
+		if (unlikely(!softleaf_is_swap(entry)))
 			continue;
 
-		page = read_swap_cache_async(entry, GFP_HIGHUSER_MOVABLE,
-					     vma, index, false, &splug);
-		if (page)
-			put_page(page);
+		pte_unmap_unlock(ptep, ptl);
+		ptep = NULL;
+
+		folio = read_swap_cache_async(entry, GFP_HIGHUSER_MOVABLE,
+					     vma, addr, &splug);
+		if (folio)
+			folio_put(folio);
 	}
+
+	if (ptep)
+		pte_unmap_unlock(ptep, ptl);
 	swap_read_unplug(splug);
 	cond_resched();
 
@@ -230,66 +228,75 @@ static int swapin_walk_pmd_entry(pmd_t *pmd, unsigned long start,
 
 static const struct mm_walk_ops swapin_walk_ops = {
 	.pmd_entry		= swapin_walk_pmd_entry,
+	.walk_lock		= PGWALK_RDLOCK,
 };
 
-static void force_shm_swapin_readahead(struct vm_area_struct *vma,
+static void shmem_swapin_range(struct vm_area_struct *vma,
 		unsigned long start, unsigned long end,
 		struct address_space *mapping)
 {
 	XA_STATE(xas, &mapping->i_pages, linear_page_index(vma, start));
-	pgoff_t end_index = linear_page_index(vma, end + PAGE_SIZE - 1);
-	struct page *page;
+	pgoff_t end_index = linear_page_index(vma, end) - 1;
+	struct folio *folio;
 	struct swap_iocb *splug = NULL;
 
 	rcu_read_lock();
-	xas_for_each(&xas, page, end_index) {
-		swp_entry_t swap;
+	xas_for_each(&xas, folio, end_index) {
+		unsigned long addr;
+		swp_entry_t entry;
 
-		if (!xa_is_value(page))
+		if (!xa_is_value(folio))
 			continue;
-		swap = radix_to_swp_entry(page);
+		entry = radix_to_swp_entry(folio);
 		/* There might be swapin error entries in shmem mapping. */
-		if (non_swap_entry(swap))
+		if (!softleaf_is_swap(entry))
 			continue;
+
+		addr = vma->vm_start +
+			((xas.xa_index - vma->vm_pgoff) << PAGE_SHIFT);
 		xas_pause(&xas);
 		rcu_read_unlock();
 
-		page = read_swap_cache_async(swap, GFP_HIGHUSER_MOVABLE,
-					     NULL, 0, false, &splug);
-		if (page)
-			put_page(page);
+		folio = read_swap_cache_async(entry, mapping_gfp_mask(mapping),
+					     vma, addr, &splug);
+		if (folio)
+			folio_put(folio);
 
 		rcu_read_lock();
 	}
 	rcu_read_unlock();
 	swap_read_unplug(splug);
-
-	lru_add_drain();	/* Push any new pages onto the LRU now */
 }
 #endif		/* CONFIG_SWAP */
 
+static void mark_mmap_lock_dropped(struct madvise_behavior *madv_behavior)
+{
+	VM_WARN_ON_ONCE(madv_behavior->lock_mode == MADVISE_VMA_READ_LOCK);
+	madv_behavior->lock_dropped = true;
+}
+
 /*
  * Schedule all required I/O operations.  Do not wait for completion.
  */
-static long madvise_willneed(struct vm_area_struct *vma,
-			     struct vm_area_struct **prev,
-			     unsigned long start, unsigned long end)
+static long madvise_willneed(struct madvise_behavior *madv_behavior)
 {
-	struct mm_struct *mm = vma->vm_mm;
+	struct vm_area_struct *vma = madv_behavior->vma;
+	struct mm_struct *mm = madv_behavior->mm;
 	struct file *file = vma->vm_file;
+	unsigned long start = madv_behavior->range.start;
+	unsigned long end = madv_behavior->range.end;
 	loff_t offset;
 
-	*prev = vma;
 #ifdef CONFIG_SWAP
 	if (!file) {
-		walk_page_range(vma->vm_mm, start, end, &swapin_walk_ops, vma);
+		walk_page_range_vma(vma, start, end, &swapin_walk_ops, vma);
 		lru_add_drain(); /* Push any new pages onto the LRU now */
 		return 0;
 	}
 
 	if (shmem_mapping(file->f_mapping)) {
-		force_shm_swapin_readahead(vma, start, end,
-					file->f_mapping);
+		shmem_swapin_range(vma, start, end, file->f_mapping);
+		lru_add_drain(); /* Push any new pages onto the LRU now */
 		return 0;
 	}
 #else
@@ -308,7 +315,7 @@ static long madvise_willneed(struct vm_area_struct *vma,
 	 * vma's reference to the file) can go away as soon as we drop
 	 * mmap_lock.
 	 */
-	*prev = NULL;	/* tell sys_madvise we drop mmap_lock */
+	mark_mmap_lock_dropped(madv_behavior);
 	get_file(file);
 	offset = (loff_t)(start - vma->vm_start)
 			+ ((loff_t)vma->vm_pgoff << PAGE_SHIFT);
@@ -329,11 +336,21 @@ static inline bool can_do_file_pageout(struct vm_area_struct *vma)
 	 * otherwise we'd be including shared non-exclusive mappings, which
 	 * opens a side channel.
 	 */
-	return inode_owner_or_capable(&init_user_ns,
+	return inode_owner_or_capable(&nop_mnt_idmap,
 				      file_inode(vma->vm_file)) ||
 	       file_permission(vma->vm_file, MAY_WRITE) == 0;
 }
 
+static inline int madvise_folio_pte_batch(unsigned long addr, unsigned long end,
+					  struct folio *folio, pte_t *ptep,
+					  pte_t *ptentp)
+{
+	int max_nr = (end - addr) / PAGE_SIZE;
+
+	return folio_pte_batch_flags(folio, NULL, ptep, ptentp, max_nr,
+				     FPB_MERGE_YOUNG_DIRTY);
+}
+
 static int madvise_cold_or_pageout_pte_range(pmd_t *pmd,
 				unsigned long addr, unsigned long end,
 				struct mm_walk *walk)
@@ -343,11 +360,13 @@ static int madvise_cold_or_pageout_pte_range(pmd_t *pmd,
 	bool pageout = private->pageout;
 	struct mm_struct *mm = tlb->mm;
 	struct vm_area_struct *vma = walk->vma;
-	pte_t *orig_pte, *pte, ptent;
+	pte_t *start_pte, *pte, ptent;
 	spinlock_t *ptl;
-	struct page *page = NULL;
-	LIST_HEAD(page_list);
+	struct folio *folio = NULL;
+	LIST_HEAD(folio_list);
 	bool pageout_anon_only_filter;
+	unsigned int batch_count = 0;
+	int nr;
 
 	if (fatal_signal_pending(current))
 		return -EINTR;
@@ -371,34 +390,34 @@ static int madvise_cold_or_pageout_pte_range(pmd_t *pmd,
 
 		if (unlikely(!pmd_present(orig_pmd))) {
 			VM_BUG_ON(thp_migration_supported() &&
-					!is_pmd_migration_entry(orig_pmd));
+					!pmd_is_migration_entry(orig_pmd));
 			goto huge_unlock;
 		}
 
-		page = pmd_page(orig_pmd);
+		folio = pmd_folio(orig_pmd);
 
-		/* Do not interfere with other mappings of this page */
-		if (page_mapcount(page) != 1)
+		/* Do not interfere with other mappings of this folio */
+		if (folio_maybe_mapped_shared(folio))
 			goto huge_unlock;
 
-		if (pageout_anon_only_filter && !PageAnon(page))
+		if (pageout_anon_only_filter && !folio_test_anon(folio))
 			goto huge_unlock;
 
 		if (next - addr != HPAGE_PMD_SIZE) {
 			int err;
 
-			get_page(page);
+			folio_get(folio);
 			spin_unlock(ptl);
-			lock_page(page);
-			err = split_huge_page(page);
-			unlock_page(page);
-			put_page(page);
+			folio_lock(folio);
+			err = split_folio(folio);
+			folio_unlock(folio);
+			folio_put(folio);
 			if (!err)
-				goto regular_page;
+				goto regular_folio;
 			return 0;
 		}
 
-		if (pmd_young(orig_pmd)) {
+		if (!pageout && pmd_young(orig_pmd)) {
 			pmdp_invalidate(vma, addr, pmd);
 			orig_pmd = pmd_mkold(orig_pmd);
 
@@ -406,34 +425,48 @@ static int madvise_cold_or_pageout_pte_range(pmd_t *pmd,
 			tlb_remove_pmd_tlb_entry(tlb, pmd, addr);
 		}
 
-		ClearPageReferenced(page);
-		test_and_clear_page_young(page);
+		folio_clear_referenced(folio);
+		folio_test_clear_young(folio);
+		if (folio_test_active(folio))
+			folio_set_workingset(folio);
 		if (pageout) {
-			if (!isolate_lru_page(page)) {
-				if (PageUnevictable(page))
-					putback_lru_page(page);
+			if (folio_isolate_lru(folio)) {
+				if (folio_test_unevictable(folio))
+					folio_putback_lru(folio);
 				else
-					list_add(&page->lru, &page_list);
+					list_add(&folio->lru, &folio_list);
 			}
 		} else
-			deactivate_page(page);
+			folio_deactivate(folio);
 huge_unlock:
 		spin_unlock(ptl);
 		if (pageout)
-			reclaim_pages(&page_list);
+			reclaim_pages(&folio_list);
 		return 0;
 	}
 
-regular_page:
-	if (pmd_trans_unstable(pmd))
-		return 0;
+regular_folio:
 #endif
 	tlb_change_page_size(tlb, PAGE_SIZE);
-	orig_pte = pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
+restart:
+	start_pte = pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
+	if (!start_pte)
+		return 0;
 	flush_tlb_batched_pending(mm);
 	arch_enter_lazy_mmu_mode();
-	for (; addr < end; pte++, addr += PAGE_SIZE) {
-		ptent = *pte;
+	for (; addr < end; pte += nr, addr += nr * PAGE_SIZE) {
+		nr = 1;
+		ptent = ptep_get(pte);
+
+		if (++batch_count == SWAP_CLUSTER_MAX) {
+			batch_count = 0;
+			if (need_resched()) {
+				arch_leave_lazy_mmu_mode();
+				pte_unmap_unlock(start_pte, ptl);
+				cond_resched();
+				goto restart;
+			}
+		}
 
 		if (pte_none(ptent))
 			continue;
@@ -441,82 +474,93 @@ regular_page:
 		if (!pte_present(ptent))
 			continue;
 
-		page = vm_normal_page(vma, addr, ptent);
-		if (!page || is_zone_device_page(page))
+		folio = vm_normal_folio(vma, addr, ptent);
+		if (!folio || folio_is_zone_device(folio))
 			continue;
 
 		/*
-		 * Creating a THP page is expensive so split it only if we
-		 * are sure it's worth. Split it if we are only owner.
+		 * If we encounter a large folio, only split it if it is not
+		 * fully mapped within the range we are operating on. Otherwise
+		 * leave it as is so that it can be swapped out whole. If we
+		 * fail to split a folio, leave it in place and advance to the
+		 * next pte in the range.
 		 */
-		if (PageTransCompound(page)) {
-			if (page_mapcount(page) != 1)
-				break;
-			if (pageout_anon_only_filter && !PageAnon(page))
-				break;
-			get_page(page);
-			if (!trylock_page(page)) {
-				put_page(page);
-				break;
-			}
-			pte_unmap_unlock(orig_pte, ptl);
-			if (split_huge_page(page)) {
-				unlock_page(page);
-				put_page(page);
-				orig_pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
-				break;
+		if (folio_test_large(folio)) {
+			nr = madvise_folio_pte_batch(addr, end, folio, pte, &ptent);
+			if (nr < folio_nr_pages(folio)) {
+				int err;
+
+				if (folio_maybe_mapped_shared(folio))
+					continue;
+				if (pageout_anon_only_filter && !folio_test_anon(folio))
+					continue;
+				if (!folio_trylock(folio))
+					continue;
+				folio_get(folio);
+				arch_leave_lazy_mmu_mode();
+				pte_unmap_unlock(start_pte, ptl);
+				start_pte = NULL;
+				err = split_folio(folio);
+				folio_unlock(folio);
+				folio_put(folio);
+				start_pte = pte =
+					pte_offset_map_lock(mm, pmd, addr, &ptl);
+				if (!start_pte)
+					break;
+				flush_tlb_batched_pending(mm);
+				arch_enter_lazy_mmu_mode();
+				if (!err)
+					nr = 0;
+				continue;
 			}
-			unlock_page(page);
-			put_page(page);
-			orig_pte = pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
-			pte--;
-			addr -= PAGE_SIZE;
-			continue;
 		}
 
 		/*
-		 * Do not interfere with other mappings of this page and
-		 * non-LRU page.
+		 * Do not interfere with other mappings of this folio and
+		 * non-LRU folio. If we have a large folio at this point, we
+		 * know it is fully mapped so if its mapcount is the same as its
+		 * number of pages, it must be exclusive.
 		 */
-		if (!PageLRU(page) || page_mapcount(page) != 1)
+		if (!folio_test_lru(folio) ||
+		    folio_mapcount(folio) != folio_nr_pages(folio))
 			continue;
 
-		if (pageout_anon_only_filter && !PageAnon(page))
+		if (pageout_anon_only_filter && !folio_test_anon(folio))
 			continue;
 
-		VM_BUG_ON_PAGE(PageTransCompound(page), page);
-
-		if (pte_young(ptent)) {
-			ptent = ptep_get_and_clear_full(mm, addr, pte,
-							tlb->fullmm);
-			ptent = pte_mkold(ptent);
-			set_pte_at(mm, addr, pte, ptent);
-			tlb_remove_tlb_entry(tlb, pte, addr);
+		if (!pageout && pte_young(ptent)) {
+			clear_young_dirty_ptes(vma, addr, pte, nr,
+					       CYDP_CLEAR_YOUNG);
+			tlb_remove_tlb_entries(tlb, pte, nr, addr);
 		}
 
 		/*
-		 * We are deactivating a page for accelerating reclaiming.
-		 * VM couldn't reclaim the page unless we clear PG_young.
+		 * We are deactivating a folio for accelerating reclaiming.
+		 * VM couldn't reclaim the folio unless we clear PG_young.
 		 * As a side effect, it makes confuse idle-page tracking
 		 * because they will miss recent referenced history.
 		 */
-		ClearPageReferenced(page);
-		test_and_clear_page_young(page);
+		folio_clear_referenced(folio);
+		folio_test_clear_young(folio);
+		if (folio_test_active(folio))
+			folio_set_workingset(folio);
 		if (pageout) {
-			if (!isolate_lru_page(page)) {
-				if (PageUnevictable(page))
-					putback_lru_page(page);
+			if (folio_isolate_lru(folio)) {
+				if (folio_test_unevictable(folio))
+					folio_putback_lru(folio);
 				else
-					list_add(&page->lru, &page_list);
+					list_add(&folio->lru, &folio_list);
 			}
 		} else
-			deactivate_page(page);
+			folio_deactivate(folio);
 	}
 
-	arch_leave_lazy_mmu_mode();
-	pte_unmap_unlock(orig_pte, ptl);
+	if (start_pte) {
+		arch_leave_lazy_mmu_mode();
+		pte_unmap_unlock(start_pte, ptl);
+	}
 	if (pageout)
-		reclaim_pages(&page_list);
+		reclaim_pages(&folio_list);
 	cond_resched();
 
 	return 0;
@@ -524,19 +568,23 @@ regular_page:
 
 static const struct mm_walk_ops cold_walk_ops = {
 	.pmd_entry = madvise_cold_or_pageout_pte_range,
+	.walk_lock = PGWALK_RDLOCK,
 };
 
 static void madvise_cold_page_range(struct mmu_gather *tlb,
-			     struct vm_area_struct *vma,
-			     unsigned long addr, unsigned long end)
+		struct madvise_behavior *madv_behavior)
+
 {
+	struct vm_area_struct *vma = madv_behavior->vma;
+	struct madvise_behavior_range *range = &madv_behavior->range;
 	struct madvise_walk_private walk_private = {
 		.pageout = false,
 		.tlb = tlb,
 	};
 
 	tlb_start_vma(tlb, vma);
-	walk_page_range(vma->vm_mm, addr, end, &cold_walk_ops, &walk_private);
+	walk_page_range_vma(vma, range->start, range->end, &cold_walk_ops,
+			&walk_private);
 	tlb_end_vma(tlb, vma);
 }
 
@@ -545,28 +593,25 @@ static inline bool can_madv_lru_vma(struct vm_area_struct *vma)
 	return !(vma->vm_flags & (VM_LOCKED|VM_PFNMAP|VM_HUGETLB));
 }
 
-static long madvise_cold(struct vm_area_struct *vma,
-			struct vm_area_struct **prev,
-			unsigned long start_addr, unsigned long end_addr)
+static long madvise_cold(struct madvise_behavior *madv_behavior)
 {
-	struct mm_struct *mm = vma->vm_mm;
+	struct vm_area_struct *vma = madv_behavior->vma;
 	struct mmu_gather tlb;
 
-	*prev = vma;
 	if (!can_madv_lru_vma(vma))
 		return -EINVAL;
 
 	lru_add_drain();
-	tlb_gather_mmu(&tlb, mm);
-	madvise_cold_page_range(&tlb, vma, start_addr, end_addr);
+	tlb_gather_mmu(&tlb, madv_behavior->mm);
+	madvise_cold_page_range(&tlb, madv_behavior);
 	tlb_finish_mmu(&tlb);
 
 	return 0;
 }
 
 static void madvise_pageout_page_range(struct mmu_gather *tlb,
-			     struct vm_area_struct *vma,
-			     unsigned long addr, unsigned long end)
+		struct vm_area_struct *vma,
+		struct madvise_behavior_range *range)
 {
 	struct madvise_walk_private walk_private = {
 		.pageout = true,
@@ -574,18 +619,16 @@ static void madvise_pageout_page_range(struct mmu_gather *tlb,
 	};
 
 	tlb_start_vma(tlb, vma);
-	walk_page_range(vma->vm_mm, addr, end, &cold_walk_ops, &walk_private);
+	walk_page_range_vma(vma, range->start, range->end, &cold_walk_ops,
+			    &walk_private);
 	tlb_end_vma(tlb, vma);
 }
 
-static long madvise_pageout(struct vm_area_struct *vma,
-			struct vm_area_struct **prev,
-			unsigned long start_addr, unsigned long end_addr)
+static long madvise_pageout(struct madvise_behavior *madv_behavior)
 {
-	struct mm_struct *mm = vma->vm_mm;
 	struct mmu_gather tlb;
+	struct vm_area_struct *vma = madv_behavior->vma;
 
-	*prev = vma;
 	if (!can_madv_lru_vma(vma))
 		return -EINVAL;
 
@@ -600,8 +643,8 @@ static long madvise_pageout(struct vm_area_struct *vma,
 		return 0;
 
 	lru_add_drain();
-	tlb_gather_mmu(&tlb, mm);
-	madvise_pageout_page_range(&tlb, vma, start_addr, end_addr);
+	tlb_gather_mmu(&tlb, madv_behavior->mm);
+	madvise_pageout_page_range(&tlb, vma, &madv_behavior->range);
 	tlb_finish_mmu(&tlb);
 
 	return 0;
@@ -611,30 +654,31 @@ static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr,
 				unsigned long end, struct mm_walk *walk)
 
 {
+	const cydp_t cydp_flags = CYDP_CLEAR_YOUNG | CYDP_CLEAR_DIRTY;
 	struct mmu_gather *tlb = walk->private;
 	struct mm_struct *mm = tlb->mm;
 	struct vm_area_struct *vma = walk->vma;
 	spinlock_t *ptl;
-	pte_t *orig_pte, *pte, ptent;
+	pte_t *start_pte, *pte, ptent;
 	struct folio *folio;
-	struct page *page;
 	int nr_swap = 0;
 	unsigned long next;
+	int nr, max_nr;
 
 	next = pmd_addr_end(addr, end);
 	if (pmd_trans_huge(*pmd))
 		if (madvise_free_huge_pmd(tlb, vma, pmd, addr, next))
-			goto next;
-
-	if (pmd_trans_unstable(pmd))
-		return 0;
+			return 0;
 
 	tlb_change_page_size(tlb, PAGE_SIZE);
-	orig_pte = pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
+	start_pte = pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
+	if (!start_pte)
+		return 0;
 	flush_tlb_batched_pending(mm);
 	arch_enter_lazy_mmu_mode();
-	for (; addr != end; pte++, addr += PAGE_SIZE) {
-		ptent = *pte;
+	for (; addr != end; pte += nr, addr += PAGE_SIZE * nr) {
+		nr = 1;
+		ptent = ptep_get(pte);
 
 		if (pte_none(ptent))
 			continue;
@@ -644,61 +688,69 @@ static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr,
 		 * (page allocation + zeroing).
 		 */
 		if (!pte_present(ptent)) {
-			swp_entry_t entry;
-
-			entry = pte_to_swp_entry(ptent);
-			if (!non_swap_entry(entry)) {
-				nr_swap--;
-				free_swap_and_cache(entry);
-				pte_clear_not_present_full(mm, addr, pte, tlb->fullmm);
-			} else if (is_hwpoison_entry(entry) ||
-				   is_swapin_error_entry(entry)) {
+			softleaf_t entry = softleaf_from_pte(ptent);
+
+			if (softleaf_is_swap(entry)) {
+				max_nr = (end - addr) / PAGE_SIZE;
+				nr = swap_pte_batch(pte, max_nr, ptent);
+				nr_swap -= nr;
+				free_swap_and_cache_nr(entry, nr);
+				clear_not_present_full_ptes(mm, addr, pte, nr, tlb->fullmm);
+			} else if (softleaf_is_hwpoison(entry) ||
+				   softleaf_is_poison_marker(entry)) {
 				pte_clear_not_present_full(mm, addr, pte, tlb->fullmm);
 			}
 			continue;
 		}
 
-		page = vm_normal_page(vma, addr, ptent);
-		if (!page || is_zone_device_page(page))
+		folio = vm_normal_folio(vma, addr, ptent);
+		if (!folio || folio_is_zone_device(folio))
 			continue;
-		folio = page_folio(page);
 
 		/*
-		 * If pmd isn't transhuge but the folio is large and
-		 * is owned by only this process, split it and
-		 * deactivate all pages.
+		 * If we encounter a large folio, only split it if it is not
+		 * fully mapped within the range we are operating on. Otherwise
+		 * leave it as is so that it can be marked as lazyfree. If we
+		 * fail to split a folio, leave it in place and advance to the
+		 * next pte in the range.
 		 */
 		if (folio_test_large(folio)) {
-			if (folio_mapcount(folio) != 1)
-				goto out;
-			folio_get(folio);
-			if (!folio_trylock(folio)) {
-				folio_put(folio);
-				goto out;
-			}
-			pte_unmap_unlock(orig_pte, ptl);
-			if (split_folio(folio)) {
+			nr = madvise_folio_pte_batch(addr, end, folio, pte, &ptent);
+			if (nr < folio_nr_pages(folio)) {
+				int err;
+
+				if (folio_maybe_mapped_shared(folio))
+					continue;
+				if (!folio_trylock(folio))
+					continue;
+				folio_get(folio);
+				arch_leave_lazy_mmu_mode();
+				pte_unmap_unlock(start_pte, ptl);
+				start_pte = NULL;
+				err = split_folio(folio);
 				folio_unlock(folio);
 				folio_put(folio);
-				orig_pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
-				goto out;
+				pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
+				start_pte = pte;
+				if (!start_pte)
+					break;
+				flush_tlb_batched_pending(mm);
+				arch_enter_lazy_mmu_mode();
+				if (!err)
+					nr = 0;
+				continue;
 			}
-			folio_unlock(folio);
-			folio_put(folio);
-			orig_pte = pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
-			pte--;
-			addr -= PAGE_SIZE;
-			continue;
 		}
 
 		if (folio_test_swapcache(folio) || folio_test_dirty(folio)) {
 			if (!folio_trylock(folio))
 				continue;
 			/*
-			 * If folio is shared with others, we mustn't clear
-			 * the folio's dirty flag.
+			 * If we have a large folio at this point, we know it is
+			 * fully mapped so if its mapcount is the same as its
+			 * number of pages, it must be exclusive.
 			 */
-			if (folio_mapcount(folio) != 1) {
+			if (folio_mapcount(folio) != folio_nr_pages(folio)) {
 				folio_unlock(folio);
 				continue;
 			}
@@ -714,46 +766,48 @@ static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr,
 		}
 
 		if (pte_young(ptent) || pte_dirty(ptent)) {
-			/*
-			 * Some of architecture(ex, PPC) don't update TLB
-			 * with set_pte_at and tlb_remove_tlb_entry so for
-			 * the portability, remap the pte with old|clean
-			 * after pte clearing.
-			 */
-			ptent = ptep_get_and_clear_full(mm, addr, pte,
-							tlb->fullmm);
-
-			ptent = pte_mkold(ptent);
-			ptent = pte_mkclean(ptent);
-			set_pte_at(mm, addr, pte, ptent);
-			tlb_remove_tlb_entry(tlb, pte, addr);
+			clear_young_dirty_ptes(vma, addr, pte, nr, cydp_flags);
+			tlb_remove_tlb_entries(tlb, pte, nr, addr);
 		}
-		mark_page_lazyfree(&folio->page);
+		folio_mark_lazyfree(folio);
 	}
-out:
-	if (nr_swap) {
-		if (current->mm == mm)
-			sync_mm_rss(mm);
 
+	if (nr_swap)
 		add_mm_counter(mm, MM_SWAPENTS, nr_swap);
+	if (start_pte) {
+		arch_leave_lazy_mmu_mode();
+		pte_unmap_unlock(start_pte, ptl);
 	}
-	arch_leave_lazy_mmu_mode();
-	pte_unmap_unlock(orig_pte, ptl);
 	cond_resched();
-next:
+
 	return 0;
 }
 
-static const struct mm_walk_ops madvise_free_walk_ops = {
-	.pmd_entry		= madvise_free_pte_range,
-};
+static inline enum page_walk_lock get_walk_lock(enum madvise_lock_mode mode)
+{
+	switch (mode) {
+	case MADVISE_VMA_READ_LOCK:
+		return PGWALK_VMA_RDLOCK_VERIFY;
+	case MADVISE_MMAP_READ_LOCK:
+		return PGWALK_RDLOCK;
+	default:
+		/* Other modes don't require fixing up the walk_lock */
+		WARN_ON_ONCE(1);
+		return PGWALK_RDLOCK;
+	}
+}
 
-static int madvise_free_single_vma(struct vm_area_struct *vma,
-			unsigned long start_addr, unsigned long end_addr)
+static int madvise_free_single_vma(struct madvise_behavior *madv_behavior)
 {
-	struct mm_struct *mm = vma->vm_mm;
+	struct mm_struct *mm = madv_behavior->mm;
+	struct vm_area_struct *vma = madv_behavior->vma;
+	unsigned long start_addr = madv_behavior->range.start;
+	unsigned long end_addr = madv_behavior->range.end;
 	struct mmu_notifier_range range;
-	struct mmu_gather tlb;
+	struct mmu_gather *tlb = madv_behavior->tlb;
+	struct mm_walk_ops walk_ops = {
+		.pmd_entry		= madvise_free_pte_range,
+	};
 
 	/* MADV_FREE works for only anon vma at the moment */
 	if (!vma_is_anonymous(vma))
@@ -765,21 +819,19 @@ 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, MMU_NOTIFY_CLEAR, 0, vma, mm,
+	mmu_notifier_range_init(&range, MMU_NOTIFY_CLEAR, 0, mm,
 				range.start, range.end);
 
 	lru_add_drain();
-	tlb_gather_mmu(&tlb, mm);
 	update_hiwater_rss(mm);
 
 	mmu_notifier_invalidate_range_start(&range);
-	tlb_start_vma(&tlb, vma);
-	walk_page_range(vma->vm_mm, range.start, range.end,
-			&madvise_free_walk_ops, &tlb);
-	tlb_end_vma(&tlb, vma);
+	tlb_start_vma(tlb, vma);
+	walk_ops.walk_lock = get_walk_lock(madv_behavior->lock_mode);
+	walk_page_range_vma(vma, range.start, range.end,
+			&walk_ops, tlb);
+	tlb_end_vma(tlb, vma);
 	mmu_notifier_invalidate_range_end(&range);
-	tlb_finish_mmu(&tlb);
-
 	return 0;
 }
 
@@ -802,18 +854,28 @@ static int madvise_free_single_vma(struct vm_area_struct *vma,
  * An interface that causes the system to free clean pages and flush
  * dirty pages is already available as msync(MS_INVALIDATE).
  */
-static long madvise_dontneed_single_vma(struct vm_area_struct *vma,
-					unsigned long start, unsigned long end)
+static long madvise_dontneed_single_vma(struct madvise_behavior *madv_behavior)
+
 {
-	zap_page_range_single(vma, start, end - start, NULL);
+	struct madvise_behavior_range *range = &madv_behavior->range;
+	struct zap_details details = {
+		.reclaim_pt = true,
+		.even_cows = true,
+	};
+
+	zap_page_range_single_batched(
+			madv_behavior->tlb, madv_behavior->vma, range->start,
+			range->end - range->start, &details);
 	return 0;
 }
 
-static bool madvise_dontneed_free_valid_vma(struct vm_area_struct *vma,
-					    unsigned long start,
-					    unsigned long *end,
-					    int behavior)
+static
+bool madvise_dontneed_free_valid_vma(struct madvise_behavior *madv_behavior)
 {
+	struct vm_area_struct *vma = madv_behavior->vma;
+	int behavior = madv_behavior->behavior;
+	struct madvise_behavior_range *range = &madv_behavior->range;
+
 	if (!is_vm_hugetlb_page(vma)) {
 		unsigned int forbidden = VM_PFNMAP;
 
@@ -825,7 +887,7 @@ static bool madvise_dontneed_free_valid_vma(struct vm_area_struct *vma,
 
 	if (behavior != MADV_DONTNEED && behavior != MADV_DONTNEED_LOCKED)
 		return false;
-	if (start & ~huge_page_mask(hstate_vma(vma)))
+	if (range->start & ~huge_page_mask(hstate_vma(vma)))
 		return false;
 
 	/*
@@ -834,52 +896,38 @@ static bool madvise_dontneed_free_valid_vma(struct vm_area_struct *vma,
 	 * Avoid unexpected data loss by rounding down the number of
 	 * huge pages freed.
 	 */
-	*end = ALIGN_DOWN(*end, huge_page_size(hstate_vma(vma)));
+	range->end = ALIGN_DOWN(range->end, huge_page_size(hstate_vma(vma)));
 
 	return true;
 }
 
-static long madvise_dontneed_free(struct vm_area_struct *vma,
-				  struct vm_area_struct **prev,
-				  unsigned long start, unsigned long end,
-				  int behavior)
+static long madvise_dontneed_free(struct madvise_behavior *madv_behavior)
 {
-	struct mm_struct *mm = vma->vm_mm;
+	struct mm_struct *mm = madv_behavior->mm;
+	struct madvise_behavior_range *range = &madv_behavior->range;
+	int behavior = madv_behavior->behavior;
 
-	*prev = vma;
-	if (!madvise_dontneed_free_valid_vma(vma, start, &end, behavior))
+	if (!madvise_dontneed_free_valid_vma(madv_behavior))
 		return -EINVAL;
 
-	if (start == end)
+	if (range->start == range->end)
 		return 0;
 
-	if (!userfaultfd_remove(vma, start, end)) {
-		*prev = NULL; /* mmap_lock has been dropped, prev is stale */
+	if (!userfaultfd_remove(madv_behavior->vma, range->start, range->end)) {
+		struct vm_area_struct *vma;
 
+		mark_mmap_lock_dropped(madv_behavior);
 		mmap_read_lock(mm);
-		vma = find_vma(mm, start);
+		madv_behavior->vma = vma = vma_lookup(mm, range->start);
 		if (!vma)
 			return -ENOMEM;
-		if (start < vma->vm_start) {
-			/*
-			 * This "vma" under revalidation is the one
-			 * with the lowest vma->vm_start where start
-			 * is also < vma->vm_end. If start <
-			 * vma->vm_start it means an hole materialized
-			 * in the user address space within the
-			 * virtual range passed to MADV_DONTNEED
-			 * or MADV_FREE.
-			 */
-			return -ENOMEM;
-		}
 		/*
 		 * Potential end adjustment for hugetlb vma is OK as
 		 * the check below keeps end within vma.
 		 */
-		if (!madvise_dontneed_free_valid_vma(vma, start, &end,
-						     behavior))
+		if (!madvise_dontneed_free_valid_vma(madv_behavior))
 			return -EINVAL;
-		if (end > vma->vm_end) {
+		if (range->end > vma->vm_end) {
 			/*
 			 * Don't fail if end > vma->vm_end. If the old
 			 * vma was split while the mmap_lock was
@@ -892,52 +940,43 @@ static long madvise_dontneed_free(struct vm_area_struct *vma,
 			 * end-vma->vm_end range, but the manager can
 			 * handle a repetition fine.
 			 */
-			end = vma->vm_end;
+			range->end = vma->vm_end;
 		}
-		VM_WARN_ON(start >= end);
+		/*
+		 * If the memory region between start and end was
+		 * originally backed by 4kB pages and then remapped to
+		 * be backed by hugepages while mmap_lock was dropped,
+		 * the adjustment for hugetlb vma above may have rounded
+		 * end down to the start address.
+		 */
+		if (range->start == range->end)
+			return 0;
+		VM_WARN_ON(range->start > range->end);
 	}
 
 	if (behavior == MADV_DONTNEED || behavior == MADV_DONTNEED_LOCKED)
-		return madvise_dontneed_single_vma(vma, start, end);
+		return madvise_dontneed_single_vma(madv_behavior);
 	else if (behavior == MADV_FREE)
-		return madvise_free_single_vma(vma, start, end);
+		return madvise_free_single_vma(madv_behavior);
 	else
 		return -EINVAL;
 }
 
-static long madvise_populate(struct vm_area_struct *vma,
-			     struct vm_area_struct **prev,
-			     unsigned long start, unsigned long end,
-			     int behavior)
+static long madvise_populate(struct madvise_behavior *madv_behavior)
 {
-	const bool write = behavior == MADV_POPULATE_WRITE;
-	struct mm_struct *mm = vma->vm_mm;
-	unsigned long tmp_end;
+	struct mm_struct *mm = madv_behavior->mm;
+	const bool write = madv_behavior->behavior == MADV_POPULATE_WRITE;
 	int locked = 1;
+	unsigned long start = madv_behavior->range.start;
+	unsigned long end = madv_behavior->range.end;
 	long pages;
 
-	*prev = vma;
-
 	while (start < end) {
-		/*
-		 * We might have temporarily dropped the lock. For example,
-		 * our VMA might have been split.
-		 */
-		if (!vma || start >= vma->vm_end) {
-			vma = vma_lookup(mm, start);
-			if (!vma)
-				return -ENOMEM;
-		}
-
-		tmp_end = min_t(unsigned long, end, vma->vm_end);
 		/* Populate (prefault) page tables readable/writable. */
-		pages = faultin_vma_page_range(vma, start, tmp_end, write,
-					       &locked);
+		pages = faultin_page_range(mm, start, end, write, &locked);
 		if (!locked) {
 			mmap_read_lock(mm);
 			locked = 1;
-			*prev = NULL;
-			vma = NULL;
 		}
 		if (pages < 0) {
 			switch (pages) {
@@ -953,7 +992,7 @@ static long madvise_populate(struct vm_area_struct *vma,
 				pr_warn_once("%s: unhandled return value: %ld\n",
 					     __func__, pages);
 				fallthrough;
-			case -ENOMEM:
+			case -ENOMEM: /* No VMA or out of memory. */
 				return -ENOMEM;
 			}
 		}
@@ -966,16 +1005,17 @@ static long madvise_populate(struct vm_area_struct *vma,
  * Application wants to free up the pages and associated backing store.
  * This is effectively punching a hole into the middle of a file.
  */
-static long madvise_remove(struct vm_area_struct *vma,
-				struct vm_area_struct **prev,
-				unsigned long start, unsigned long end)
+static long madvise_remove(struct madvise_behavior *madv_behavior)
 {
 	loff_t offset;
 	int error;
 	struct file *f;
-	struct mm_struct *mm = vma->vm_mm;
+	struct mm_struct *mm = madv_behavior->mm;
+	struct vm_area_struct *vma = madv_behavior->vma;
+	unsigned long start = madv_behavior->range.start;
+	unsigned long end = madv_behavior->range.end;
 
-	*prev = NULL;	/* tell sys_madvise we drop mmap_lock */
+	mark_mmap_lock_dropped(madv_behavior);
 
 	if (vma->vm_flags & VM_LOCKED)
 		return -EINVAL;
@@ -986,7 +1026,7 @@ static long madvise_remove(struct vm_area_struct *vma,
 			return -EINVAL;
 	}
 
-	if ((vma->vm_flags & (VM_SHARED|VM_WRITE)) != (VM_SHARED|VM_WRITE))
+	if (!vma_is_shared_maywrite(vma))
 		return -EACCES;
 
 	offset = (loff_t)(start - vma->vm_start)
@@ -1011,36 +1051,333 @@ static long madvise_remove(struct vm_area_struct *vma,
 	return error;
 }
 
+static bool is_valid_guard_vma(struct vm_area_struct *vma, bool allow_locked)
+{
+	vm_flags_t disallowed = VM_SPECIAL | VM_HUGETLB;
+
+	/*
+	 * A user could lock after setting a guard range but that's fine, as
+	 * they'd not be able to fault in. The issue arises when we try to zap
+	 * existing locked VMAs. We don't want to do that.
+	 */
+	if (!allow_locked)
+		disallowed |= VM_LOCKED;
+
+	return !(vma->vm_flags & disallowed);
+}
+
+static bool is_guard_pte_marker(pte_t ptent)
+{
+	const softleaf_t entry = softleaf_from_pte(ptent);
+
+	return softleaf_is_guard_marker(entry);
+}
+
+static int guard_install_pud_entry(pud_t *pud, unsigned long addr,
+				   unsigned long next, struct mm_walk *walk)
+{
+	pud_t pudval = pudp_get(pud);
+
+	/* If huge return >0 so we abort the operation + zap. */
+	return pud_trans_huge(pudval);
+}
+
+static int guard_install_pmd_entry(pmd_t *pmd, unsigned long addr,
+				   unsigned long next, struct mm_walk *walk)
+{
+	pmd_t pmdval = pmdp_get(pmd);
+
+	/* If huge return >0 so we abort the operation + zap. */
+	return pmd_trans_huge(pmdval);
+}
+
+static int guard_install_pte_entry(pte_t *pte, unsigned long addr,
+				   unsigned long next, struct mm_walk *walk)
+{
+	pte_t pteval = ptep_get(pte);
+	unsigned long *nr_pages = (unsigned long *)walk->private;
+
+	/* If there is already a guard page marker, we have nothing to do. */
+	if (is_guard_pte_marker(pteval)) {
+		(*nr_pages)++;
+
+		return 0;
+	}
+
+	/* If populated return >0 so we abort the operation + zap. */
+	return 1;
+}
+
+static int guard_install_set_pte(unsigned long addr, unsigned long next,
+				 pte_t *ptep, struct mm_walk *walk)
+{
+	unsigned long *nr_pages = (unsigned long *)walk->private;
+
+	/* Simply install a PTE marker, this causes segfault on access. */
+	*ptep = make_pte_marker(PTE_MARKER_GUARD);
+	(*nr_pages)++;
+
+	return 0;
+}
+
+static long madvise_guard_install(struct madvise_behavior *madv_behavior)
+{
+	struct vm_area_struct *vma = madv_behavior->vma;
+	struct madvise_behavior_range *range = &madv_behavior->range;
+	struct mm_walk_ops walk_ops = {
+		.pud_entry	= guard_install_pud_entry,
+		.pmd_entry	= guard_install_pmd_entry,
+		.pte_entry	= guard_install_pte_entry,
+		.install_pte	= guard_install_set_pte,
+		.walk_lock	= get_walk_lock(madv_behavior->lock_mode),
+	};
+	long err;
+	int i;
+
+	if (!is_valid_guard_vma(vma, /* allow_locked = */false))
+		return -EINVAL;
+
+	/*
+	 * Set atomically under read lock. All pertinent readers will need to
+	 * acquire an mmap/VMA write lock to read it. All remaining readers may
+	 * or may not see the flag set, but we don't care.
+	 */
+	vma_flag_set_atomic(vma, VMA_MAYBE_GUARD_BIT);
+
+	/*
+	 * If anonymous and we are establishing page tables the VMA ought to
+	 * have an anon_vma associated with it.
+	 *
+	 * We will hold an mmap read lock if this is necessary, this is checked
+	 * as part of the VMA lock logic.
+	 */
+	if (vma_is_anonymous(vma)) {
+		VM_WARN_ON_ONCE(!vma->anon_vma &&
+				madv_behavior->lock_mode != MADVISE_MMAP_READ_LOCK);
+
+		err = anon_vma_prepare(vma);
+		if (err)
+			return err;
+	}
+
+	/*
+	 * Optimistically try to install the guard marker pages first. If any
+	 * non-guard pages or THP huge pages are encountered, give up and zap
+	 * the range before trying again.
+	 *
+	 * We try a few times before giving up and releasing back to userland to
+	 * loop around, releasing locks in the process to avoid contention.
+	 *
+	 * This would only happen due to races with e.g. page faults or
+	 * khugepaged.
+	 *
+	 * In most cases we should simply install the guard markers immediately
+	 * with no zap or looping.
+	 */
+	for (i = 0; i < MAX_MADVISE_GUARD_RETRIES; i++) {
+		unsigned long nr_pages = 0;
+
+		/* Returns < 0 on error, == 0 if success, > 0 if zap needed. */
+		if (madv_behavior->lock_mode == MADVISE_VMA_READ_LOCK)
+			err = walk_page_range_vma_unsafe(madv_behavior->vma,
+					range->start, range->end, &walk_ops,
+					&nr_pages);
+		else
+			err = walk_page_range_mm_unsafe(vma->vm_mm, range->start,
+					range->end, &walk_ops, &nr_pages);
+		if (err < 0)
+			return err;
+
+		if (err == 0) {
+			unsigned long nr_expected_pages =
+				PHYS_PFN(range->end - range->start);
+
+			VM_WARN_ON(nr_pages != nr_expected_pages);
+			return 0;
+		}
+
+		/*
+		 * OK some of the range have non-guard pages mapped, zap
+		 * them. This leaves existing guard pages in place.
+		 */
+		zap_page_range_single(vma, range->start,
+				range->end - range->start, NULL);
+	}
+
+	/*
+	 * We were unable to install the guard pages, return to userspace and
+	 * immediately retry, relieving lock contention.
+	 */
+	return restart_syscall();
+}
+
+static int guard_remove_pud_entry(pud_t *pud, unsigned long addr,
+				  unsigned long next, struct mm_walk *walk)
+{
+	pud_t pudval = pudp_get(pud);
+
+	/* If huge, cannot have guard pages present, so no-op - skip. */
+	if (pud_trans_huge(pudval))
+		walk->action = ACTION_CONTINUE;
+
+	return 0;
+}
+
+static int guard_remove_pmd_entry(pmd_t *pmd, unsigned long addr,
+				  unsigned long next, struct mm_walk *walk)
+{
+	pmd_t pmdval = pmdp_get(pmd);
+
+	/* If huge, cannot have guard pages present, so no-op - skip. */
+	if (pmd_trans_huge(pmdval))
+		walk->action = ACTION_CONTINUE;
+
+	return 0;
+}
+
+static int guard_remove_pte_entry(pte_t *pte, unsigned long addr,
+				  unsigned long next, struct mm_walk *walk)
+{
+	pte_t ptent = ptep_get(pte);
+
+	if (is_guard_pte_marker(ptent)) {
+		/* Simply clear the PTE marker. */
+		pte_clear_not_present_full(walk->mm, addr, pte, false);
+		update_mmu_cache(walk->vma, addr, pte);
+	}
+
+	return 0;
+}
+
+static long madvise_guard_remove(struct madvise_behavior *madv_behavior)
+{
+	struct vm_area_struct *vma = madv_behavior->vma;
+	struct madvise_behavior_range *range = &madv_behavior->range;
+	struct mm_walk_ops wallk_ops = {
+		.pud_entry = guard_remove_pud_entry,
+		.pmd_entry = guard_remove_pmd_entry,
+		.pte_entry = guard_remove_pte_entry,
+		.walk_lock = get_walk_lock(madv_behavior->lock_mode),
+	};
+
+	/*
+	 * We're ok with removing guards in mlock()'d ranges, as this is a
+	 * non-destructive action.
+	 */
+	if (!is_valid_guard_vma(vma, /* allow_locked = */true))
+		return -EINVAL;
+
+	return walk_page_range_vma(vma, range->start, range->end,
+				   &wallk_ops, NULL);
+}
+
+#ifdef CONFIG_64BIT
+/* Does the madvise operation result in discarding of mapped data? */
+static bool is_discard(int behavior)
+{
+	switch (behavior) {
+	case MADV_FREE:
+	case MADV_DONTNEED:
+	case MADV_DONTNEED_LOCKED:
+	case MADV_REMOVE:
+	case MADV_DONTFORK:
+	case MADV_WIPEONFORK:
+	case MADV_GUARD_INSTALL:
+		return true;
+	}
+
+	return false;
+}
+
+/*
+ * We are restricted from madvise()'ing mseal()'d VMAs only in very particular
+ * circumstances - discarding of data from read-only anonymous SEALED mappings.
+ *
+ * This is because users cannot trivally discard data from these VMAs, and may
+ * only do so via an appropriate madvise() call.
+ */
+static bool can_madvise_modify(struct madvise_behavior *madv_behavior)
+{
+	struct vm_area_struct *vma = madv_behavior->vma;
+
+	/* If the VMA isn't sealed we're good. */
+	if (!vma_is_sealed(vma))
+		return true;
+
+	/* For a sealed VMA, we only care about discard operations. */
+	if (!is_discard(madv_behavior->behavior))
+		return true;
+
+	/*
+	 * We explicitly permit all file-backed mappings, whether MAP_SHARED or
+	 * MAP_PRIVATE.
+	 *
+	 * The latter causes some complications. Because now, one can mmap()
+	 * read/write a MAP_PRIVATE mapping, write to it, then mprotect()
+	 * read-only, mseal() and a discard will be permitted.
+	 *
+	 * However, in order to avoid issues with potential use of madvise(...,
+	 * MADV_DONTNEED) of mseal()'d .text mappings we, for the time being,
+	 * permit this.
+	 */
+	if (!vma_is_anonymous(vma))
+		return true;
+
+	/* If the user could write to the mapping anyway, then this is fine. */
+	if ((vma->vm_flags & VM_WRITE) &&
+	    arch_vma_access_permitted(vma, /* write= */ true,
+			/* execute= */ false, /* foreign= */ false))
+		return true;
+
+	/* Otherwise, we are not permitted to perform this operation. */
+	return false;
+}
+#else
+static bool can_madvise_modify(struct madvise_behavior *madv_behavior)
+{
+	return true;
+}
+#endif
+
 /*
  * Apply an madvise behavior to a region of a vma.  madvise_update_vma
  * will handle splitting a vm area into separate areas, each area with its own
  * behavior.
  */
-static int madvise_vma_behavior(struct vm_area_struct *vma,
-				struct vm_area_struct **prev,
-				unsigned long start, unsigned long end,
-				unsigned long behavior)
+static int madvise_vma_behavior(struct madvise_behavior *madv_behavior)
 {
+	int behavior = madv_behavior->behavior;
+	struct vm_area_struct *vma = madv_behavior->vma;
+	vm_flags_t new_flags = vma->vm_flags;
+	struct madvise_behavior_range *range = &madv_behavior->range;
 	int error;
-	struct anon_vma_name *anon_name;
-	unsigned long new_flags = vma->vm_flags;
+
+	if (unlikely(!can_madvise_modify(madv_behavior)))
+		return -EPERM;
 
 	switch (behavior) {
 	case MADV_REMOVE:
-		return madvise_remove(vma, prev, start, end);
+		return madvise_remove(madv_behavior);
 	case MADV_WILLNEED:
-		return madvise_willneed(vma, prev, start, end);
+		return madvise_willneed(madv_behavior);
 	case MADV_COLD:
-		return madvise_cold(vma, prev, start, end);
+		return madvise_cold(madv_behavior);
 	case MADV_PAGEOUT:
-		return madvise_pageout(vma, prev, start, end);
+		return madvise_pageout(madv_behavior);
 	case MADV_FREE:
 	case MADV_DONTNEED:
 	case MADV_DONTNEED_LOCKED:
-		return madvise_dontneed_free(vma, prev, start, end, behavior);
-	case MADV_POPULATE_READ:
-	case MADV_POPULATE_WRITE:
-		return madvise_populate(vma, prev, start, end, behavior);
+		return madvise_dontneed_free(madv_behavior);
+	case MADV_COLLAPSE:
+		return madvise_collapse(vma, range->start, range->end,
+			&madv_behavior->lock_dropped);
+	case MADV_GUARD_INSTALL:
+		return madvise_guard_install(madv_behavior);
+	case MADV_GUARD_REMOVE:
+		return madvise_guard_remove(madv_behavior);
+
+	/* The below behaviours update VMAs via madvise_update_vma(). */
+
 	case MADV_NORMAL:
 		new_flags = new_flags & ~VM_RAND_READ & ~VM_SEQ_READ;
 		break;
@@ -1054,30 +1391,34 @@ static int madvise_vma_behavior(struct vm_area_struct *vma,
 		new_flags |= VM_DONTCOPY;
 		break;
 	case MADV_DOFORK:
-		if (vma->vm_flags & VM_IO)
+		if (new_flags & VM_IO)
 			return -EINVAL;
 		new_flags &= ~VM_DONTCOPY;
 		break;
 	case MADV_WIPEONFORK:
 		/* MADV_WIPEONFORK is only supported on anonymous memory. */
-		if (vma->vm_file || vma->vm_flags & VM_SHARED)
+		if (vma->vm_file || new_flags & VM_SHARED)
 			return -EINVAL;
 		new_flags |= VM_WIPEONFORK;
 		break;
 	case MADV_KEEPONFORK:
+		if (new_flags & VM_DROPPABLE)
+			return -EINVAL;
 		new_flags &= ~VM_WIPEONFORK;
 		break;
 	case MADV_DONTDUMP:
 		new_flags |= VM_DONTDUMP;
 		break;
 	case MADV_DODUMP:
-		if (!is_vm_hugetlb_page(vma) && new_flags & VM_SPECIAL)
+		if ((!is_vm_hugetlb_page(vma) && (new_flags & VM_SPECIAL)) ||
+		    (new_flags & VM_DROPPABLE))
 			return -EINVAL;
 		new_flags &= ~VM_DONTDUMP;
 		break;
 	case MADV_MERGEABLE:
 	case MADV_UNMERGEABLE:
-		error = ksm_madvise(vma, start, end, behavior, &new_flags);
+		error = ksm_madvise(vma, range->start, range->end,
+				behavior, &new_flags);
 		if (error)
 			goto out;
 		break;
@@ -1087,16 +1428,17 @@ static int madvise_vma_behavior(struct vm_area_struct *vma,
 		if (error)
 			goto out;
 		break;
-	case MADV_COLLAPSE:
-		return madvise_collapse(vma, prev, start, end);
+	case __MADV_SET_ANON_VMA_NAME:
+		/* Only anonymous mappings can be named */
+		if (vma->vm_file && !vma_is_anon_shmem(vma))
+			return -EBADF;
+		break;
 	}
 
-	anon_name = anon_vma_name(vma);
-	anon_vma_name_get(anon_name);
-	error = madvise_update_vma(vma, prev, start, end, new_flags,
-				   anon_name);
-	anon_vma_name_put(anon_name);
+	/* This is a write operation.*/
+	VM_WARN_ON_ONCE(madv_behavior->lock_mode != MADVISE_MMAP_WRITE_LOCK);
 
+	error = madvise_update_vma(new_flags, madv_behavior);
 out:
 	/*
 	 * madvise() returns EAGAIN if kernel resources, such as
@@ -1111,15 +1453,15 @@ out:
 /*
  * Error injection support for memory error handling.
  */
-static int madvise_inject_error(int behavior,
-		unsigned long start, unsigned long end)
+static int madvise_inject_error(struct madvise_behavior *madv_behavior)
 {
 	unsigned long size;
+	unsigned long start = madv_behavior->range.start;
+	unsigned long end = madv_behavior->range.end;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-
 	for (; start < end; start += size) {
 		unsigned long pfn;
 		struct page *page;
@@ -1137,14 +1479,14 @@ static int madvise_inject_error(int behavior,
 		 */
 		size = page_size(compound_head(page));
 
-		if (behavior == MADV_SOFT_OFFLINE) {
+		if (madv_behavior->behavior == MADV_SOFT_OFFLINE) {
 			pr_info("Soft offlining pfn %#lx at process virtual address %#lx\n",
 				 pfn, start);
 			ret = soft_offline_page(pfn, MF_COUNT_INCREASED);
 		} else {
 			pr_info("Injecting memory failure for pfn %#lx at process virtual address %#lx\n",
 				 pfn, start);
-			ret = memory_failure(pfn, MF_COUNT_INCREASED | MF_SW_SIMULATED);
+			ret = memory_failure(pfn, MF_ACTION_REQUIRED | MF_COUNT_INCREASED | MF_SW_SIMULATED);
 			if (ret == -EOPNOTSUPP)
 				ret = 0;
 		}
@@ -1155,7 +1497,31 @@ static int madvise_inject_error(int behavior,
 
 	return 0;
 }
-#endif
+
+static bool is_memory_failure(struct madvise_behavior *madv_behavior)
+{
+	switch (madv_behavior->behavior) {
+	case MADV_HWPOISON:
+	case MADV_SOFT_OFFLINE:
+		return true;
+	default:
+		return false;
+	}
+}
+
+#else
+
+static int madvise_inject_error(struct madvise_behavior *madv_behavior)
+{
+	return 0;
+}
+
+static bool is_memory_failure(struct madvise_behavior *madv_behavior)
+{
+	return false;
+}
+
+#endif	/* CONFIG_MEMORY_FAILURE */
 
 static bool
 madvise_behavior_valid(int behavior)
@@ -1188,6 +1554,8 @@ madvise_behavior_valid(int behavior)
 	case MADV_DODUMP:
 	case MADV_WIPEONFORK:
 	case MADV_KEEPONFORK:
+	case MADV_GUARD_INSTALL:
+	case MADV_GUARD_REMOVE:
 #ifdef CONFIG_MEMORY_FAILURE
 	case MADV_SOFT_OFFLINE:
 	case MADV_HWPOISON:
@@ -1199,7 +1567,8 @@ madvise_behavior_valid(int behavior)
 	}
 }
 
-static bool process_madvise_behavior_valid(int behavior)
+/* Can we invoke process_madvise() on a remote mm for the specified behavior? */
+static bool process_madvise_remote_valid(int behavior)
 {
 	switch (behavior) {
 	case MADV_COLD:
@@ -1212,122 +1581,364 @@ static bool process_madvise_behavior_valid(int behavior)
 	}
 }
 
+/* Does this operation invoke anon_vma_prepare()? */
+static bool prepares_anon_vma(int behavior)
+{
+	switch (behavior) {
+	case MADV_GUARD_INSTALL:
+		return true;
+	default:
+		return false;
+	}
+}
+
 /*
- * Walk the vmas in range [start,end), and call the visit function on each one.
- * The visit function will get start and end parameters that cover the overlap
- * between the current vma and the original range.  Any unmapped regions in the
- * original range will result in this function returning -ENOMEM while still
- * calling the visit function on all of the existing vmas in the range.
- * Must be called with the mmap_lock held for reading or writing.
+ * We have acquired a VMA read lock, is the VMA valid to be madvise'd under VMA
+ * read lock only now we have a VMA to examine?
  */
-static
-int madvise_walk_vmas(struct mm_struct *mm, unsigned long start,
-		      unsigned long end, unsigned long arg,
-		      int (*visit)(struct vm_area_struct *vma,
-				   struct vm_area_struct **prev, unsigned long start,
-				   unsigned long end, unsigned long arg))
+static bool is_vma_lock_sufficient(struct vm_area_struct *vma,
+		struct madvise_behavior *madv_behavior)
 {
+	/* Must span only a single VMA.*/
+	if (madv_behavior->range.end > vma->vm_end)
+		return false;
+	/* Remote processes unsupported. */
+	if (current->mm != vma->vm_mm)
+		return false;
+	/* Userfaultfd unsupported. */
+	if (userfaultfd_armed(vma))
+		return false;
+	/*
+	 * anon_vma_prepare() explicitly requires an mmap lock for
+	 * serialisation, so we cannot use a VMA lock in this case.
+	 *
+	 * Note we might race with anon_vma being set, however this makes this
+	 * check overly paranoid which is safe.
+	 */
+	if (vma_is_anonymous(vma) &&
+	    prepares_anon_vma(madv_behavior->behavior) && !vma->anon_vma)
+		return false;
+
+	return true;
+}
+
+/*
+ * Try to acquire a VMA read lock if possible.
+ *
+ * We only support this lock over a single VMA, which the input range must
+ * span either partially or fully.
+ *
+ * This function always returns with an appropriate lock held. If a VMA read
+ * lock could be acquired, we return true and set madv_behavior state
+ * accordingly.
+ *
+ * If a VMA read lock could not be acquired, we return false and expect caller to
+ * fallback to mmap lock behaviour.
+ */
+static bool try_vma_read_lock(struct madvise_behavior *madv_behavior)
+{
+	struct mm_struct *mm = madv_behavior->mm;
 	struct vm_area_struct *vma;
-	struct vm_area_struct *prev;
-	unsigned long tmp;
+
+	vma = lock_vma_under_rcu(mm, madv_behavior->range.start);
+	if (!vma)
+		goto take_mmap_read_lock;
+
+	if (!is_vma_lock_sufficient(vma, madv_behavior)) {
+		vma_end_read(vma);
+		goto take_mmap_read_lock;
+	}
+
+	madv_behavior->vma = vma;
+	return true;
+
+take_mmap_read_lock:
+	mmap_read_lock(mm);
+	madv_behavior->lock_mode = MADVISE_MMAP_READ_LOCK;
+	return false;
+}
+
+/*
+ * Walk the vmas in range [start,end), and call the madvise_vma_behavior
+ * function on each one.  The function will get start and end parameters that
+ * cover the overlap between the current vma and the original range.  Any
+ * unmapped regions in the original range will result in this function returning
+ * -ENOMEM while still calling the madvise_vma_behavior function on all of the
+ * existing vmas in the range.  Must be called with the mmap_lock held for
+ * reading or writing.
+ */
+static
+int madvise_walk_vmas(struct madvise_behavior *madv_behavior)
+{
+	struct mm_struct *mm = madv_behavior->mm;
+	struct madvise_behavior_range *range = &madv_behavior->range;
+	/* range is updated to span each VMA, so store end of entire range. */
+	unsigned long last_end = range->end;
 	int unmapped_error = 0;
+	int error;
+	struct vm_area_struct *prev, *vma;
 
 	/*
-	 * If the interval [start,end) covers some unmapped address
-	 * ranges, just ignore them, but return -ENOMEM at the end.
-	 * - different from the way of handling in mlock etc.
+	 * If VMA read lock is supported, apply madvise to a single VMA
+	 * tentatively, avoiding walking VMAs.
 	 */
-	vma = find_vma_prev(mm, start, &prev);
-	if (vma && start > vma->vm_start)
+	if (madv_behavior->lock_mode == MADVISE_VMA_READ_LOCK &&
+	    try_vma_read_lock(madv_behavior)) {
+		error = madvise_vma_behavior(madv_behavior);
+		vma_end_read(madv_behavior->vma);
+		return error;
+	}
+
+	vma = find_vma_prev(mm, range->start, &prev);
+	if (vma && range->start > vma->vm_start)
 		prev = vma;
 
 	for (;;) {
-		int error;
-
 		/* Still start < end. */
 		if (!vma)
 			return -ENOMEM;
 
-		/* Here start < (end|vma->vm_end). */
-		if (start < vma->vm_start) {
+		/* Here start < (last_end|vma->vm_end). */
+		if (range->start < vma->vm_start) {
+			/*
+			 * This indicates a gap between VMAs in the input
+			 * range. This does not cause the operation to abort,
+			 * rather we simply return -ENOMEM to indicate that this
+			 * has happened, but carry on.
+			 */
 			unmapped_error = -ENOMEM;
-			start = vma->vm_start;
-			if (start >= end)
+			range->start = vma->vm_start;
+			if (range->start >= last_end)
 				break;
 		}
 
-		/* Here vma->vm_start <= start < (end|vma->vm_end) */
-		tmp = vma->vm_end;
-		if (end < tmp)
-			tmp = end;
+		/* Here vma->vm_start <= range->start < (last_end|vma->vm_end) */
+		range->end = min(vma->vm_end, last_end);
 
-		/* Here vma->vm_start <= start < tmp <= (end|vma->vm_end). */
-		error = visit(vma, &prev, start, tmp, arg);
+		/* Here vma->vm_start <= range->start < range->end <= (last_end|vma->vm_end). */
+		madv_behavior->prev = prev;
+		madv_behavior->vma = vma;
+		error = madvise_vma_behavior(madv_behavior);
 		if (error)
 			return error;
-		start = tmp;
-		if (prev && start < prev->vm_end)
-			start = prev->vm_end;
-		if (start >= end)
+		if (madv_behavior->lock_dropped) {
+			/* We dropped the mmap lock, we can't ref the VMA. */
+			prev = NULL;
+			vma = NULL;
+			madv_behavior->lock_dropped = false;
+		} else {
+			vma = madv_behavior->vma;
+			prev = vma;
+		}
+
+		if (vma && range->end < vma->vm_end)
+			range->end = vma->vm_end;
+		if (range->end >= last_end)
 			break;
-		if (prev)
-			vma = find_vma(mm, prev->vm_end);
-		else	/* madvise_remove dropped mmap_lock */
-			vma = find_vma(mm, start);
+
+		vma = find_vma(mm, vma ? vma->vm_end : range->end);
+		range->start = range->end;
 	}
 
 	return unmapped_error;
 }
 
-#ifdef CONFIG_ANON_VMA_NAME
-static int madvise_vma_anon_name(struct vm_area_struct *vma,
-				 struct vm_area_struct **prev,
-				 unsigned long start, unsigned long end,
-				 unsigned long anon_name)
+/*
+ * Any behaviour which results in changes to the vma->vm_flags needs to
+ * take mmap_lock for writing. Others, which simply traverse vmas, need
+ * to only take it for reading.
+ */
+static enum madvise_lock_mode get_lock_mode(struct madvise_behavior *madv_behavior)
 {
-	int error;
+	if (is_memory_failure(madv_behavior))
+		return MADVISE_NO_LOCK;
 
-	/* Only anonymous mappings can be named */
-	if (vma->vm_file && !vma_is_anon_shmem(vma))
-		return -EBADF;
+	switch (madv_behavior->behavior) {
+	case MADV_REMOVE:
+	case MADV_WILLNEED:
+	case MADV_COLD:
+	case MADV_PAGEOUT:
+	case MADV_POPULATE_READ:
+	case MADV_POPULATE_WRITE:
+	case MADV_COLLAPSE:
+		return MADVISE_MMAP_READ_LOCK;
+	case MADV_GUARD_INSTALL:
+	case MADV_GUARD_REMOVE:
+	case MADV_DONTNEED:
+	case MADV_DONTNEED_LOCKED:
+	case MADV_FREE:
+		return MADVISE_VMA_READ_LOCK;
+	default:
+		return MADVISE_MMAP_WRITE_LOCK;
+	}
+}
 
-	error = madvise_update_vma(vma, prev, start, end, vma->vm_flags,
-				   (struct anon_vma_name *)anon_name);
+static int madvise_lock(struct madvise_behavior *madv_behavior)
+{
+	struct mm_struct *mm = madv_behavior->mm;
+	enum madvise_lock_mode lock_mode = get_lock_mode(madv_behavior);
 
-	/*
-	 * madvise() returns EAGAIN if kernel resources, such as
-	 * slab, are temporarily unavailable.
-	 */
-	if (error == -ENOMEM)
-		error = -EAGAIN;
-	return error;
+	switch (lock_mode) {
+	case MADVISE_NO_LOCK:
+		break;
+	case MADVISE_MMAP_WRITE_LOCK:
+		if (mmap_write_lock_killable(mm))
+			return -EINTR;
+		break;
+	case MADVISE_MMAP_READ_LOCK:
+		mmap_read_lock(mm);
+		break;
+	case MADVISE_VMA_READ_LOCK:
+		/* We will acquire the lock per-VMA in madvise_walk_vmas(). */
+		break;
+	}
+
+	madv_behavior->lock_mode = lock_mode;
+	return 0;
 }
 
-int madvise_set_anon_name(struct mm_struct *mm, unsigned long start,
-			  unsigned long len_in, struct anon_vma_name *anon_name)
+static void madvise_unlock(struct madvise_behavior *madv_behavior)
 {
-	unsigned long end;
-	unsigned long len;
+	struct mm_struct *mm = madv_behavior->mm;
 
-	if (start & ~PAGE_MASK)
-		return -EINVAL;
-	len = (len_in + ~PAGE_MASK) & PAGE_MASK;
+	switch (madv_behavior->lock_mode) {
+	case  MADVISE_NO_LOCK:
+		return;
+	case MADVISE_MMAP_WRITE_LOCK:
+		mmap_write_unlock(mm);
+		break;
+	case MADVISE_MMAP_READ_LOCK:
+		mmap_read_unlock(mm);
+		break;
+	case MADVISE_VMA_READ_LOCK:
+		/* We will drop the lock per-VMA in madvise_walk_vmas(). */
+		break;
+	}
+
+	madv_behavior->lock_mode = MADVISE_NO_LOCK;
+}
+
+static bool madvise_batch_tlb_flush(int behavior)
+{
+	switch (behavior) {
+	case MADV_DONTNEED:
+	case MADV_DONTNEED_LOCKED:
+	case MADV_FREE:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static void madvise_init_tlb(struct madvise_behavior *madv_behavior)
+{
+	if (madvise_batch_tlb_flush(madv_behavior->behavior))
+		tlb_gather_mmu(madv_behavior->tlb, madv_behavior->mm);
+}
+
+static void madvise_finish_tlb(struct madvise_behavior *madv_behavior)
+{
+	if (madvise_batch_tlb_flush(madv_behavior->behavior))
+		tlb_finish_mmu(madv_behavior->tlb);
+}
+
+static bool is_valid_madvise(unsigned long start, size_t len_in, int behavior)
+{
+	size_t len;
+
+	if (!madvise_behavior_valid(behavior))
+		return false;
+
+	if (!PAGE_ALIGNED(start))
+		return false;
+	len = PAGE_ALIGN(len_in);
 
 	/* Check to see whether len was rounded up from small -ve to zero */
 	if (len_in && !len)
-		return -EINVAL;
+		return false;
 
-	end = start + len;
-	if (end < start)
-		return -EINVAL;
+	if (start + len < start)
+		return false;
 
-	if (end == start)
-		return 0;
+	return true;
+}
+
+/*
+ * madvise_should_skip() - Return if the request is invalid or nothing.
+ * @start:	Start address of madvise-requested address range.
+ * @len_in:	Length of madvise-requested address range.
+ * @behavior:	Requested madvise behavor.
+ * @err:	Pointer to store an error code from the check.
+ *
+ * If the specified behaviour is invalid or nothing would occur, we skip the
+ * operation.  This function returns true in the cases, otherwise false.  In
+ * the former case we store an error on @err.
+ */
+static bool madvise_should_skip(unsigned long start, size_t len_in,
+		int behavior, int *err)
+{
+	if (!is_valid_madvise(start, len_in, behavior)) {
+		*err = -EINVAL;
+		return true;
+	}
+	if (start + PAGE_ALIGN(len_in) == start) {
+		*err = 0;
+		return true;
+	}
+	return false;
+}
 
-	return madvise_walk_vmas(mm, start, end, (unsigned long)anon_name,
-				 madvise_vma_anon_name);
+static bool is_madvise_populate(struct madvise_behavior *madv_behavior)
+{
+	switch (madv_behavior->behavior) {
+	case MADV_POPULATE_READ:
+	case MADV_POPULATE_WRITE:
+		return true;
+	default:
+		return false;
+	}
 }
-#endif /* CONFIG_ANON_VMA_NAME */
+
+/*
+ * untagged_addr_remote() assumes mmap_lock is already held. On
+ * architectures like x86 and RISC-V, tagging is tricky because each
+ * mm may have a different tagging mask. However, we might only hold
+ * the per-VMA lock (currently only local processes are supported),
+ * so untagged_addr is used to avoid the mmap_lock assertion for
+ * local processes.
+ */
+static inline unsigned long get_untagged_addr(struct mm_struct *mm,
+		unsigned long start)
+{
+	return current->mm == mm ? untagged_addr(start) :
+				   untagged_addr_remote(mm, start);
+}
+
+static int madvise_do_behavior(unsigned long start, size_t len_in,
+		struct madvise_behavior *madv_behavior)
+{
+	struct blk_plug plug;
+	int error;
+	struct madvise_behavior_range *range = &madv_behavior->range;
+
+	if (is_memory_failure(madv_behavior)) {
+		range->start = start;
+		range->end = start + len_in;
+		return madvise_inject_error(madv_behavior);
+	}
+
+	range->start = get_untagged_addr(madv_behavior->mm, start);
+	range->end = range->start + PAGE_ALIGN(len_in);
+
+	blk_start_plug(&plug);
+	if (is_madvise_populate(madv_behavior))
+		error = madvise_populate(madv_behavior);
+	else
+		error = madvise_walk_vmas(madv_behavior);
+	blk_finish_plug(&plug);
+	return error;
+}
+
 /*
  * The madvise(2) system call.
  *
@@ -1398,75 +2009,112 @@ int madvise_set_anon_name(struct mm_struct *mm, unsigned long start,
  *  -EIO    - an I/O error occurred while paging in data.
  *  -EBADF  - map exists, but area maps something that isn't a file.
  *  -EAGAIN - a kernel resource was temporarily unavailable.
+ *  -EPERM  - memory is sealed.
  */
 int do_madvise(struct mm_struct *mm, unsigned long start, size_t len_in, int behavior)
 {
-	unsigned long end;
 	int error;
-	int write;
-	size_t len;
-	struct blk_plug plug;
+	struct mmu_gather tlb;
+	struct madvise_behavior madv_behavior = {
+		.mm = mm,
+		.behavior = behavior,
+		.tlb = &tlb,
+	};
 
-	start = untagged_addr(start);
+	if (madvise_should_skip(start, len_in, behavior, &error))
+		return error;
+	error = madvise_lock(&madv_behavior);
+	if (error)
+		return error;
+	madvise_init_tlb(&madv_behavior);
+	error = madvise_do_behavior(start, len_in, &madv_behavior);
+	madvise_finish_tlb(&madv_behavior);
+	madvise_unlock(&madv_behavior);
 
-	if (!madvise_behavior_valid(behavior))
-		return -EINVAL;
+	return error;
+}
 
-	if (!PAGE_ALIGNED(start))
-		return -EINVAL;
-	len = PAGE_ALIGN(len_in);
+SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior)
+{
+	return do_madvise(current->mm, start, len_in, behavior);
+}
 
-	/* Check to see whether len was rounded up from small -ve to zero */
-	if (len_in && !len)
-		return -EINVAL;
+/* Perform an madvise operation over a vector of addresses and lengths. */
+static ssize_t vector_madvise(struct mm_struct *mm, struct iov_iter *iter,
+			      int behavior)
+{
+	ssize_t ret = 0;
+	size_t total_len;
+	struct mmu_gather tlb;
+	struct madvise_behavior madv_behavior = {
+		.mm = mm,
+		.behavior = behavior,
+		.tlb = &tlb,
+	};
 
-	end = start + len;
-	if (end < start)
-		return -EINVAL;
+	total_len = iov_iter_count(iter);
 
-	if (end == start)
-		return 0;
+	ret = madvise_lock(&madv_behavior);
+	if (ret)
+		return ret;
+	madvise_init_tlb(&madv_behavior);
 
-#ifdef CONFIG_MEMORY_FAILURE
-	if (behavior == MADV_HWPOISON || behavior == MADV_SOFT_OFFLINE)
-		return madvise_inject_error(behavior, start, start + len_in);
-#endif
+	while (iov_iter_count(iter)) {
+		unsigned long start = (unsigned long)iter_iov_addr(iter);
+		size_t len_in = iter_iov_len(iter);
+		int error;
 
-	write = madvise_need_mmap_write(behavior);
-	if (write) {
-		if (mmap_write_lock_killable(mm))
-			return -EINTR;
-	} else {
-		mmap_read_lock(mm);
-	}
+		if (madvise_should_skip(start, len_in, behavior, &error))
+			ret = error;
+		else
+			ret = madvise_do_behavior(start, len_in, &madv_behavior);
+		/*
+		 * An madvise operation is attempting to restart the syscall,
+		 * but we cannot proceed as it would not be correct to repeat
+		 * the operation in aggregate, and would be surprising to the
+		 * user.
+		 *
+		 * We drop and reacquire locks so it is safe to just loop and
+		 * try again. We check for fatal signals in case we need exit
+		 * early anyway.
+		 */
+		if (ret == -ERESTARTNOINTR) {
+			if (fatal_signal_pending(current)) {
+				ret = -EINTR;
+				break;
+			}
 
-	blk_start_plug(&plug);
-	error = madvise_walk_vmas(mm, start, end, behavior,
-			madvise_vma_behavior);
-	blk_finish_plug(&plug);
-	if (write)
-		mmap_write_unlock(mm);
-	else
-		mmap_read_unlock(mm);
+			/* Drop and reacquire lock to unwind race. */
+			madvise_finish_tlb(&madv_behavior);
+			madvise_unlock(&madv_behavior);
+			ret = madvise_lock(&madv_behavior);
+			if (ret)
+				goto out;
+			madvise_init_tlb(&madv_behavior);
+			continue;
+		}
+		if (ret < 0)
+			break;
+		iov_iter_advance(iter, iter_iov_len(iter));
+	}
+	madvise_finish_tlb(&madv_behavior);
+	madvise_unlock(&madv_behavior);
 
-	return error;
-}
+out:
+	ret = (total_len - iov_iter_count(iter)) ? : ret;
 
-SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior)
-{
-	return do_madvise(current->mm, start, len_in, behavior);
+	return ret;
 }
 
 SYSCALL_DEFINE5(process_madvise, int, pidfd, const struct iovec __user *, vec,
 		size_t, vlen, int, behavior, unsigned int, flags)
 {
 	ssize_t ret;
-	struct iovec iovstack[UIO_FASTIOV], iovec;
+	struct iovec iovstack[UIO_FASTIOV];
 	struct iovec *iov = iovstack;
 	struct iov_iter iter;
 	struct task_struct *task;
 	struct mm_struct *mm;
-	size_t total_len;
 	unsigned int f_flags;
 
 	if (flags != 0) {
@@ -1484,39 +2132,33 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, const struct iovec __user *, vec,
 		goto free_iov;
 	}
 
-	if (!process_madvise_behavior_valid(behavior)) {
-		ret = -EINVAL;
-		goto release_task;
-	}
-
 	/* Require PTRACE_MODE_READ to avoid leaking ASLR metadata. */
 	mm = mm_access(task, PTRACE_MODE_READ_FSCREDS);
-	if (IS_ERR_OR_NULL(mm)) {
-		ret = IS_ERR(mm) ? PTR_ERR(mm) : -ESRCH;
+	if (IS_ERR(mm)) {
+		ret = PTR_ERR(mm);
 		goto release_task;
 	}
 
 	/*
-	 * Require CAP_SYS_NICE for influencing process performance. Note that
-	 * only non-destructive hints are currently supported.
+	 * We need only perform this check if we are attempting to manipulate a
+	 * remote process's address space.
 	 */
-	if (!capable(CAP_SYS_NICE)) {
-		ret = -EPERM;
+	if (mm != current->mm && !process_madvise_remote_valid(behavior)) {
+		ret = -EINVAL;
 		goto release_mm;
 	}
 
-	total_len = iov_iter_count(&iter);
-
-	while (iov_iter_count(&iter)) {
-		iovec = iov_iter_iovec(&iter);
-		ret = do_madvise(mm, (unsigned long)iovec.iov_base,
-					iovec.iov_len, behavior);
-		if (ret < 0)
-			break;
-		iov_iter_advance(&iter, iovec.iov_len);
+	/*
+	 * Require CAP_SYS_NICE for influencing process performance. Note that
+	 * only non-destructive hints are currently supported for remote
+	 * processes.
+	 */
+	if (mm != current->mm && !capable(CAP_SYS_NICE)) {
+		ret = -EPERM;
+		goto release_mm;
 	}
 
-	ret = (total_len - iov_iter_count(&iter)) ? : ret;
+	ret = vector_madvise(mm, &iter, behavior);
 
 release_mm:
 	mmput(mm);
@@ -1527,3 +2169,88 @@ free_iov:
 out:
 	return ret;
 }
+
+#ifdef CONFIG_ANON_VMA_NAME
+
+#define ANON_VMA_NAME_MAX_LEN		80
+#define ANON_VMA_NAME_INVALID_CHARS	"\\`$[]"
+
+static inline bool is_valid_name_char(char ch)
+{
+	/* printable ascii characters, excluding ANON_VMA_NAME_INVALID_CHARS */
+	return ch > 0x1f && ch < 0x7f &&
+		!strchr(ANON_VMA_NAME_INVALID_CHARS, ch);
+}
+
+static int madvise_set_anon_name(struct mm_struct *mm, unsigned long start,
+		unsigned long len_in, struct anon_vma_name *anon_name)
+{
+	unsigned long end;
+	unsigned long len;
+	int error;
+	struct madvise_behavior madv_behavior = {
+		.mm = mm,
+		.behavior = __MADV_SET_ANON_VMA_NAME,
+		.anon_name = anon_name,
+	};
+
+	if (start & ~PAGE_MASK)
+		return -EINVAL;
+	len = (len_in + ~PAGE_MASK) & PAGE_MASK;
+
+	/* Check to see whether len was rounded up from small -ve to zero */
+	if (len_in && !len)
+		return -EINVAL;
+
+	end = start + len;
+	if (end < start)
+		return -EINVAL;
+
+	if (end == start)
+		return 0;
+
+	madv_behavior.range.start = start;
+	madv_behavior.range.end = end;
+
+	error = madvise_lock(&madv_behavior);
+	if (error)
+		return error;
+	error = madvise_walk_vmas(&madv_behavior);
+	madvise_unlock(&madv_behavior);
+
+	return error;
+}
+
+int set_anon_vma_name(unsigned long addr, unsigned long size,
+		      const char __user *uname)
+{
+	struct anon_vma_name *anon_name = NULL;
+	struct mm_struct *mm = current->mm;
+	int error;
+
+	if (uname) {
+		char *name, *pch;
+
+		name = strndup_user(uname, ANON_VMA_NAME_MAX_LEN);
+		if (IS_ERR(name))
+			return PTR_ERR(name);
+
+		for (pch = name; *pch != '\0'; pch++) {
+			if (!is_valid_name_char(*pch)) {
+				kfree(name);
+				return -EINVAL;
+			}
+		}
+		/* anon_vma has its own copy */
+		anon_name = anon_vma_name_alloc(name);
+		kfree(name);
+		if (!anon_name)
+			return -ENOMEM;
+	}
+
+	error = madvise_set_anon_name(mm, addr, size, anon_name);
+	anon_vma_name_put(anon_name);
+
+	return error;
+}
+#endif