Merge tag 'mm-stable-2025-08-03-12-35' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

Pull more MM updates from Andrew Morton:
"Significant patch series in this pull request:

- "mseal cleanups" (Lorenzo Stoakes)

Some mseal cleaning with no intended functional change.

- "Optimizations for khugepaged" (David Hildenbrand)

Improve khugepaged throughput by batching PTE operations for large
folios. This gain is mainly for arm64.

- "x86: enable EXECMEM_ROX_CACHE for ftrace and kprobes" (Mike Rapoport)

A bugfix, additional debug code and cleanups to the execmem code.

- "mm/shmem, swap: bugfix and improvement of mTHP swap in" (Kairui Song)

Bugfixes, cleanups and performance improvememnts to the mTHP swapin
code"

* tag 'mm-stable-2025-08-03-12-35' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm: (38 commits)
mm: mempool: fix crash in mempool_free() for zero-minimum pools
mm: correct type for vmalloc vm_flags fields
mm/shmem, swap: fix major fault counting
mm/shmem, swap: rework swap entry and index calculation for large swapin
mm/shmem, swap: simplify swapin path and result handling
mm/shmem, swap: never use swap cache and readahead for SWP_SYNCHRONOUS_IO
mm/shmem, swap: tidy up swap entry splitting
mm/shmem, swap: tidy up THP swapin checks
mm/shmem, swap: avoid redundant Xarray lookup during swapin
x86/ftrace: enable EXECMEM_ROX_CACHE for ftrace allocations
x86/kprobes: enable EXECMEM_ROX_CACHE for kprobes allocations
execmem: drop writable parameter from execmem_fill_trapping_insns()
execmem: add fallback for failures in vmalloc(VM_ALLOW_HUGE_VMAP)
execmem: move execmem_force_rw() and execmem_restore_rox() before use
execmem: rework execmem_cache_free()
execmem: introduce execmem_alloc_rw()
execmem: drop unused execmem_update_copy()
mm: fix a UAF when vma->mm is freed after vma->vm_refcnt got dropped
mm/rmap: add anon_vma lifetime debug check
mm: remove mm/io-mapping.c
...

Documentation/core-api/mm-api.rst

···
 .. kernel-doc:: mm/mmu_notifier.c
 .. kernel-doc:: mm/balloon_compaction.c
 .. kernel-doc:: mm/huge_memory.c
-.. kernel-doc:: mm/io-mapping.c

arch/arm64/mm/mmu.c

···
 
 static void __init declare_vma(struct vm_struct *vma,
 			       void *va_start, void *va_end,
-			       vm_flags_t vm_flags)
+			       unsigned long vm_flags)
 {
 	phys_addr_t pa_start = __pa_symbol(va_start);
 	unsigned long size = va_end - va_start;
···
 pte_t modify_prot_start_ptes(struct vm_area_struct *vma, unsigned long addr,
 			     pte_t *ptep, unsigned int nr)
 {
-	pte_t pte = get_and_clear_full_ptes(vma->vm_mm, addr, ptep, nr, /* full = */ 0);
+	pte_t pte = get_and_clear_ptes(vma->vm_mm, addr, ptep, nr);
 
 	if (alternative_has_cap_unlikely(ARM64_WORKAROUND_2645198)) {
 		/*

arch/x86/kernel/alternative.c

···
 
 static void *__its_alloc(struct its_array *pages)
 {
-	void *page __free(execmem) = execmem_alloc(EXECMEM_MODULE_TEXT, PAGE_SIZE);
+	void *page __free(execmem) = execmem_alloc_rw(EXECMEM_MODULE_TEXT, PAGE_SIZE);
 	if (!page)
 		return NULL;
 
···
 	if (!page)
 		return NULL;
 
-	execmem_make_temp_rw(page, PAGE_SIZE);
 	if (pages == &its_pages)
 		set_memory_x((unsigned long)page, 1);
 

arch/x86/kernel/ftrace.c

···
 
 static inline void *alloc_tramp(unsigned long size)
 {
-	return execmem_alloc(EXECMEM_FTRACE, size);
+	return execmem_alloc_rw(EXECMEM_FTRACE, size);
 }
 static inline void tramp_free(void *tramp)
 {

arch/x86/kernel/kprobes/core.c

···
 	return len;
 }
 
-/* Make page to RO mode when allocate it */
-void *alloc_insn_page(void)
-{
-	void *page;
-
-	page = execmem_alloc(EXECMEM_KPROBES, PAGE_SIZE);
-	if (!page)
-		return NULL;
-
-	/*
-	 * TODO: Once additional kernel code protection mechanisms are set, ensure
-	 * that the page was not maliciously altered and it is still zeroed.
-	 */
-	set_memory_rox((unsigned long)page, 1);
-
-	return page;
-}
-
 /* Kprobe x86 instruction emulation - only regs->ip or IF flag modifiers */
 
 static void kprobe_emulate_ifmodifiers(struct kprobe *p, struct pt_regs *regs)

arch/x86/mm/init.c

···
 static struct execmem_info execmem_info __ro_after_init;
 
 #ifdef CONFIG_ARCH_HAS_EXECMEM_ROX
-void execmem_fill_trapping_insns(void *ptr, size_t size, bool writeable)
+void execmem_fill_trapping_insns(void *ptr, size_t size)
 {
-	/* fill memory with INT3 instructions */
-	if (writeable)
-		memset(ptr, INT3_INSN_OPCODE, size);
-	else
-		text_poke_set(ptr, INT3_INSN_OPCODE, size);
+	memset(ptr, INT3_INSN_OPCODE, size);
 }
 #endif
 
···
 				.pgprot	= pgprot,
 				.alignment = MODULE_ALIGN,
 			},
-			[EXECMEM_KPROBES ... EXECMEM_BPF] = {
+			[EXECMEM_KPROBES] = {
+				.flags	= flags,
+				.start	= start,
+				.end	= MODULES_END,
+				.pgprot	= PAGE_KERNEL_ROX,
+				.alignment = MODULE_ALIGN,
+			},
+			[EXECMEM_FTRACE] = {
+				.flags	= flags,
+				.start	= start,
+				.end	= MODULES_END,
+				.pgprot	= pgprot,
+				.alignment = MODULE_ALIGN,
+			},
+			[EXECMEM_BPF] = {
 				.flags	= EXECMEM_KASAN_SHADOW,
 				.start	= start,
 				.end	= MODULES_END,

include/linux/execmem.h

···
  *				 will trap
  * @ptr:	pointer to memory to fill
  * @size:	size of the range to fill
- * @writable:	is the memory poited by @ptr is writable or ROX
  *
  * A hook for architecures to fill execmem ranges with invalid instructions.
  * Architectures that use EXECMEM_ROX_CACHE must implement this.
  */
-void execmem_fill_trapping_insns(void *ptr, size_t size, bool writable);
-
-/**
- * execmem_make_temp_rw - temporarily remap region with read-write
- *			  permissions
- * @ptr:	address of the region to remap
- * @size:	size of the region to remap
- *
- * Remaps a part of the cached large page in the ROX cache in the range
- * [@ptr, @ptr + @size) as writable and not executable. The caller must
- * have exclusive ownership of this range and ensure nothing will try to
- * execute code in this range.
- *
- * Return: 0 on success or negative error code on failure.
- */
-int execmem_make_temp_rw(void *ptr, size_t size);
+void execmem_fill_trapping_insns(void *ptr, size_t size);
 
 /**
  * execmem_restore_rox - restore read-only-execute permissions
···
  */
 int execmem_restore_rox(void *ptr, size_t size);
 #else
-static inline int execmem_make_temp_rw(void *ptr, size_t size) { return 0; }
 static inline int execmem_restore_rox(void *ptr, size_t size) { return 0; }
 #endif
 
···
 void *execmem_alloc(enum execmem_type type, size_t size);
 
 /**
+ * execmem_alloc_rw - allocate writable executable memory
+ * @type: type of the allocation
+ * @size: how many bytes of memory are required
+ *
+ * Allocates memory that will contain executable code, either generated or
+ * loaded from kernel modules.
+ *
+ * Allocates memory that will contain data coupled with executable code,
+ * like data sections in kernel modules.
+ *
+ * Forces writable permissions on the allocated memory and the caller is
+ * responsible to manage the permissions afterwards.
+ *
+ * For architectures that use ROX cache the permissions will be set to R+W.
+ * For architectures that don't use ROX cache the default permissions for @type
+ * will be used as they must be writable.
+ *
+ * Return: a pointer to the allocated memory or %NULL
+ */
+void *execmem_alloc_rw(enum execmem_type type, size_t size);
+
+/**
  * execmem_free - free executable memory
  * @ptr: pointer to the memory that should be freed
  */
···
  */
 struct vm_struct *execmem_vmap(size_t size);
 #endif
-
-/**
- * execmem_update_copy - copy an update to executable memory
- * @dst:  destination address to update
- * @src:  source address containing the data
- * @size: how many bytes of memory shold be copied
- *
- * Copy @size bytes from @src to @dst using text poking if the memory at
- * @dst is read-only.
- *
- * Return: a pointer to @dst or NULL on error
- */
-void *execmem_update_copy(void *dst, const void *src, size_t size);
 
 /**
  * execmem_is_rox - check if execmem is read-only

include/linux/io-mapping.h

···
 	kfree(iomap);
 }
 
-int io_mapping_map_user(struct io_mapping *iomap, struct vm_area_struct *vma,
-		unsigned long addr, unsigned long pfn, unsigned long size);
-
 #endif /* _LINUX_IO_MAPPING_H */

include/linux/mm.h

···
 #endif
 
 #ifdef CONFIG_64BIT
-/* VM is sealed, in vm_flags */
-#define VM_SEALED	_BITUL(63)
+#define VM_SEALED_BIT	42
+#define VM_SEALED	BIT(VM_SEALED_BIT)
+#else
+#define VM_SEALED	VM_NONE
 #endif
 
 /* Bits set in the VMA until the stack is in its final location */

include/linux/mmap_lock.h

···
 #include <linux/tracepoint-defs.h>
 #include <linux/types.h>
 #include <linux/cleanup.h>
+#include <linux/sched/mm.h>
 
 #define MMAP_LOCK_INITIALIZER(name) \
 	.mmap_lock = __RWSEM_INITIALIZER((name).mmap_lock),
···
  * reused and attached to a different mm before we lock it.
  * Returns the vma on success, NULL on failure to lock and EAGAIN if vma got
  * detached.
+ *
+ * WARNING! The vma passed to this function cannot be used if the function
+ * fails to lock it because in certain cases RCU lock is dropped and then
+ * reacquired. Once RCU lock is dropped the vma can be concurently freed.
  */
 static inline struct vm_area_struct *vma_start_read(struct mm_struct *mm,
 						    struct vm_area_struct *vma)
···
 	}
 
 	rwsem_acquire_read(&vma->vmlock_dep_map, 0, 1, _RET_IP_);
+
+	/*
+	 * If vma got attached to another mm from under us, that mm is not
+	 * stable and can be freed in the narrow window after vma->vm_refcnt
+	 * is dropped and before rcuwait_wake_up(mm) is called. Grab it before
+	 * releasing vma->vm_refcnt.
+	 */
+	if (unlikely(vma->vm_mm != mm)) {
+		/* Use a copy of vm_mm in case vma is freed after we drop vm_refcnt */
+		struct mm_struct *other_mm = vma->vm_mm;
+
+		/*
+		 * __mmdrop() is a heavy operation and we don't need RCU
+		 * protection here. Release RCU lock during these operations.
+		 * We reinstate the RCU read lock as the caller expects it to
+		 * be held when this function returns even on error.
+		 */
+		rcu_read_unlock();
+		mmgrab(other_mm);
+		vma_refcount_put(vma);
+		mmdrop(other_mm);
+		rcu_read_lock();
+		return NULL;
+	}
+
 	/*
 	 * Overflow of vm_lock_seq/mm_lock_seq might produce false locked result.
 	 * False unlocked result is impossible because we modify and check

include/linux/page-flags.h

···
 
 #define folio_start_writeback(folio)			\
 	__folio_start_writeback(folio, false)
-#define folio_start_writeback_keepwrite(folio)	\
-	__folio_start_writeback(folio, true)
 
 static __always_inline bool folio_test_head(const struct folio *folio)
 {

include/linux/pgtable.h

···
 }
 #endif
 
+/**
+ * get_and_clear_ptes - Clear present PTEs that map consecutive pages of
+ *			the same folio, collecting dirty/accessed bits.
+ * @mm: Address space the pages are mapped into.
+ * @addr: Address the first page is mapped at.
+ * @ptep: Page table pointer for the first entry.
+ * @nr: Number of entries to clear.
+ *
+ * Use this instead of get_and_clear_full_ptes() if it is known that we don't
+ * need to clear the full mm, which is mostly the case.
+ *
+ * Note that PTE bits in the PTE range besides the PFN can differ. For example,
+ * some PTEs might be write-protected.
+ *
+ * Context: The caller holds the page table lock.  The PTEs map consecutive
+ * pages that belong to the same folio.  The PTEs are all in the same PMD.
+ */
+static inline pte_t get_and_clear_ptes(struct mm_struct *mm, unsigned long addr,
+		pte_t *ptep, unsigned int nr)
+{
+	return get_and_clear_full_ptes(mm, addr, ptep, nr, 0);
+}
+
 #ifndef clear_full_ptes
 /**
  * clear_full_ptes - Clear present PTEs that map consecutive pages of the same
···
 	}
 }
 #endif
+
+/**
+ * clear_ptes - Clear present PTEs that map consecutive pages of the same folio.
+ * @mm: Address space the pages are mapped into.
+ * @addr: Address the first page is mapped at.
+ * @ptep: Page table pointer for the first entry.
+ * @nr: Number of entries to clear.
+ *
+ * Use this instead of clear_full_ptes() if it is known that we don't need to
+ * clear the full mm, which is mostly the case.
+ *
+ * Note that PTE bits in the PTE range besides the PFN can differ. For example,
+ * some PTEs might be write-protected.
+ *
+ * Context: The caller holds the page table lock.  The PTEs map consecutive
+ * pages that belong to the same folio.  The PTEs are all in the same PMD.
+ */
+static inline void clear_ptes(struct mm_struct *mm, unsigned long addr,
+		pte_t *ptep, unsigned int nr)
+{
+	clear_full_ptes(mm, addr, ptep, nr, 0);
+}
 
 /*
  * If two threads concurrently fault at the same page, the thread that

include/linux/rmap.h

···
 	default:
 		VM_WARN_ON_ONCE(true);
 	}
+
+	/*
+	 * Anon folios must have an associated live anon_vma as long as they're
+	 * mapped into userspace.
+	 * Note that the atomic_read() mainly does two things:
+	 *
+	 * 1. In KASAN builds with CONFIG_SLUB_RCU_DEBUG, it causes KASAN to
+	 *    check that the associated anon_vma has not yet been freed (subject
+	 *    to KASAN's usual limitations). This check will pass if the
+	 *    anon_vma's refcount has already dropped to 0 but an RCU grace
+	 *    period hasn't passed since then.
+	 * 2. If the anon_vma has not yet been freed, it checks that the
+	 *    anon_vma still has a nonzero refcount (as opposed to being in the
+	 *    middle of an RCU delay for getting freed).
+	 */
+	if (folio_test_anon(folio) && !folio_test_ksm(folio)) {
+		unsigned long mapping = (unsigned long)folio->mapping;
+		struct anon_vma *anon_vma;
+
+		anon_vma = (void *)(mapping - FOLIO_MAPPING_ANON);
+		VM_WARN_ON_FOLIO(atomic_read(&anon_vma->refcount) == 0, folio);
+	}
 }
 
 /*

kernel/fork.c

···
 	for (i = 0; i < NR_MM_COUNTERS; i++) {
 		long x = percpu_counter_sum(&mm->rss_stat[i]);
 
-		if (unlikely(x))
-			pr_alert("BUG: Bad rss-counter state mm:%p type:%s val:%ld\n",
-				 mm, resident_page_types[i], x);
+		if (unlikely(x)) {
+			pr_alert("BUG: Bad rss-counter state mm:%p type:%s val:%ld Comm:%s Pid:%d\n",
+				 mm, resident_page_types[i], x,
+				 current->comm,
+				 task_pid_nr(current));
+		}
 	}
 
 	if (mm_pgtables_bytes(mm))

kernel/module/main.c

···
 	else
 		execmem_type = EXECMEM_MODULE_TEXT;
 
-	ptr = execmem_alloc(execmem_type, size);
+	ptr = execmem_alloc_rw(execmem_type, size);
 	if (!ptr)
 		return -ENOMEM;
 
-	if (execmem_is_rox(execmem_type)) {
-		int err = execmem_make_temp_rw(ptr, size);
-
-		if (err) {
-			execmem_free(ptr);
-			return -ENOMEM;
-		}
-
-		mod->mem[type].is_rox = true;
-	}
+	mod->mem[type].is_rox = execmem_is_rox(execmem_type);
 
 	/*
 	 * The pointer to these blocks of memory are stored on the module

mm/Kconfig

···
 config KMAP_LOCAL_NON_LINEAR_PTE_ARRAY
 	bool
 
-# struct io_mapping based helper.  Selected by drivers that need them
-config IO_MAPPING
-	bool
-
 config MEMFD_CREATE
 	bool "Enable memfd_create() system call" if EXPERT
 

mm/Makefile

···
 obj-$(CONFIG_MAPPING_DIRTY_HELPERS) += mapping_dirty_helpers.o
 obj-$(CONFIG_PTDUMP) += ptdump.o
 obj-$(CONFIG_PAGE_REPORTING) += page_reporting.o
-obj-$(CONFIG_IO_MAPPING) += io-mapping.o
 obj-$(CONFIG_HAVE_BOOTMEM_INFO_NODE) += bootmem_info.o
 obj-$(CONFIG_GENERIC_IOREMAP) += ioremap.o
 obj-$(CONFIG_SHRINKER_DEBUG) += shrinker_debug.o

mm/damon/vaddr.c

···
 		target -= dests->weight_arr[i];
 	}
 
+	/* If the folio is already in the right node, don't do anything */
+	if (folio_nid(folio) == dests->node_id_arr[i])
+		return;
+
 isolate:
 	if (!folio_isolate_lru(folio))
 		return;

mm/execmem.c

···
 
 #ifdef CONFIG_MMU
 static void *execmem_vmalloc(struct execmem_range *range, size_t size,
-			     pgprot_t pgprot, vm_flags_t vm_flags)
+			     pgprot_t pgprot, unsigned long vm_flags)
 {
 	bool kasan = range->flags & EXECMEM_KASAN_SHADOW;
 	gfp_t gfp_flags = GFP_KERNEL | __GFP_NOWARN;
···
 }
 #else
 static void *execmem_vmalloc(struct execmem_range *range, size_t size,
-			     pgprot_t pgprot, vm_flags_t vm_flags)
+			     pgprot_t pgprot, unsigned long vm_flags)
 {
 	return vmalloc(size);
 }
···
 	struct mutex mutex;
 	struct maple_tree busy_areas;
 	struct maple_tree free_areas;
+	unsigned int pending_free_cnt;	/* protected by mutex */
 };
+
+/* delay to schedule asynchronous free if fast path free fails */
+#define FREE_DELAY	(msecs_to_jiffies(10))
+
+/* mark entries in busy_areas that should be freed asynchronously */
+#define PENDING_FREE_MASK	(1 << (PAGE_SHIFT - 1))
 
 static struct execmem_cache execmem_cache = {
 	.mutex = __MUTEX_INITIALIZER(execmem_cache.mutex),
···
 	return err;
 }
 
+static int execmem_force_rw(void *ptr, size_t size)
+{
+	unsigned int nr = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	unsigned long addr = (unsigned long)ptr;
+	int ret;
+
+	ret = set_memory_nx(addr, nr);
+	if (ret)
+		return ret;
+
+	return set_memory_rw(addr, nr);
+}
+
+int execmem_restore_rox(void *ptr, size_t size)
+{
+	unsigned int nr = PAGE_ALIGN(size) >> PAGE_SHIFT;
+	unsigned long addr = (unsigned long)ptr;
+
+	return set_memory_rox(addr, nr);
+}
+
 static void execmem_cache_clean(struct work_struct *work)
 {
 	struct maple_tree *free_areas = &execmem_cache.free_areas;
···
 
 static DECLARE_WORK(execmem_cache_clean_work, execmem_cache_clean);
 
-static int execmem_cache_add(void *ptr, size_t size)
+static int execmem_cache_add_locked(void *ptr, size_t size, gfp_t gfp_mask)
 {
 	struct maple_tree *free_areas = &execmem_cache.free_areas;
-	struct mutex *mutex = &execmem_cache.mutex;
 	unsigned long addr = (unsigned long)ptr;
 	MA_STATE(mas, free_areas, addr - 1, addr + 1);
 	unsigned long lower, upper;
 	void *area = NULL;
-	int err;
 
 	lower = addr;
 	upper = addr + size - 1;
 
-	mutex_lock(mutex);
 	area = mas_walk(&mas);
 	if (area && mas.last == addr - 1)
 		lower = mas.index;
···
 		upper = mas.last;
 
 	mas_set_range(&mas, lower, upper);
-	err = mas_store_gfp(&mas, (void *)lower, GFP_KERNEL);
-	mutex_unlock(mutex);
-	if (err)
-		return err;
+	return mas_store_gfp(&mas, (void *)lower, gfp_mask);
+}
 
-	return 0;
+static int execmem_cache_add(void *ptr, size_t size, gfp_t gfp_mask)
+{
+	guard(mutex)(&execmem_cache.mutex);
+
+	return execmem_cache_add_locked(ptr, size, gfp_mask);
 }
 
 static bool within_range(struct execmem_range *range, struct ma_state *mas,
···
 
 static int execmem_cache_populate(struct execmem_range *range, size_t size)
 {
-	vm_flags_t vm_flags = VM_ALLOW_HUGE_VMAP;
+	unsigned long vm_flags = VM_ALLOW_HUGE_VMAP;
 	struct vm_struct *vm;
 	size_t alloc_size;
 	int err = -ENOMEM;
···
 
 	alloc_size = round_up(size, PMD_SIZE);
 	p = execmem_vmalloc(range, alloc_size, PAGE_KERNEL, vm_flags);
+	if (!p) {
+		alloc_size = size;
+		p = execmem_vmalloc(range, alloc_size, PAGE_KERNEL, vm_flags);
+	}
+
 	if (!p)
 		return err;
 
···
 		goto err_free_mem;
 
 	/* fill memory with instructions that will trap */
-	execmem_fill_trapping_insns(p, alloc_size, /* writable = */ true);
+	execmem_fill_trapping_insns(p, alloc_size);
 
 	err = set_memory_rox((unsigned long)p, vm->nr_pages);
 	if (err)
 		goto err_free_mem;
 
-	err = execmem_cache_add(p, alloc_size);
+	err = execmem_cache_add(p, alloc_size, GFP_KERNEL);
 	if (err)
 		goto err_reset_direct_map;
 
···
 	return __execmem_cache_alloc(range, size);
 }
 
+static inline bool is_pending_free(void *ptr)
+{
+	return ((unsigned long)ptr & PENDING_FREE_MASK);
+}
+
+static inline void *pending_free_set(void *ptr)
+{
+	return (void *)((unsigned long)ptr | PENDING_FREE_MASK);
+}
+
+static inline void *pending_free_clear(void *ptr)
+{
+	return (void *)((unsigned long)ptr & ~PENDING_FREE_MASK);
+}
+
+static int __execmem_cache_free(struct ma_state *mas, void *ptr, gfp_t gfp_mask)
+{
+	size_t size = mas_range_len(mas);
+	int err;
+
+	err = execmem_force_rw(ptr, size);
+	if (err)
+		return err;
+
+	execmem_fill_trapping_insns(ptr, size);
+	execmem_restore_rox(ptr, size);
+
+	err = execmem_cache_add_locked(ptr, size, gfp_mask);
+	if (err)
+		return err;
+
+	mas_store_gfp(mas, NULL, gfp_mask);
+	return 0;
+}
+
+static void execmem_cache_free_slow(struct work_struct *work);
+static DECLARE_DELAYED_WORK(execmem_cache_free_work, execmem_cache_free_slow);
+
+static void execmem_cache_free_slow(struct work_struct *work)
+{
+	struct maple_tree *busy_areas = &execmem_cache.busy_areas;
+	MA_STATE(mas, busy_areas, 0, ULONG_MAX);
+	void *area;
+
+	guard(mutex)(&execmem_cache.mutex);
+
+	if (!execmem_cache.pending_free_cnt)
+		return;
+
+	mas_for_each(&mas, area, ULONG_MAX) {
+		if (!is_pending_free(area))
+			continue;
+
+		area = pending_free_clear(area);
+		if (__execmem_cache_free(&mas, area, GFP_KERNEL))
+			continue;
+
+		execmem_cache.pending_free_cnt--;
+	}
+
+	if (execmem_cache.pending_free_cnt)
+		schedule_delayed_work(&execmem_cache_free_work, FREE_DELAY);
+	else
+		schedule_work(&execmem_cache_clean_work);
+}
+
 static bool execmem_cache_free(void *ptr)
 {
 	struct maple_tree *busy_areas = &execmem_cache.busy_areas;
-	struct mutex *mutex = &execmem_cache.mutex;
 	unsigned long addr = (unsigned long)ptr;
 	MA_STATE(mas, busy_areas, addr, addr);
-	size_t size;
 	void *area;
+	int err;
 
-	mutex_lock(mutex);
+	guard(mutex)(&execmem_cache.mutex);
+
 	area = mas_walk(&mas);
-	if (!area) {
-		mutex_unlock(mutex);
+	if (!area)
 		return false;
+
+	err = __execmem_cache_free(&mas, area, GFP_KERNEL | __GFP_NORETRY);
+	if (err) {
+		/*
+		 * mas points to exact slot we've got the area from, nothing
+		 * else can modify the tree because of the mutex, so there
+		 * won't be any allocations in mas_store_gfp() and it will just
+		 * change the pointer.
+		 */
+		area = pending_free_set(area);
+		mas_store_gfp(&mas, area, GFP_KERNEL);
+		execmem_cache.pending_free_cnt++;
+		schedule_delayed_work(&execmem_cache_free_work, FREE_DELAY);
+		return true;
 	}
-	size = mas_range_len(&mas);
-
-	mas_store_gfp(&mas, NULL, GFP_KERNEL);
-	mutex_unlock(mutex);
-
-	execmem_fill_trapping_insns(ptr, size, /* writable = */ false);
-
-	execmem_cache_add(ptr, size);
 
 	schedule_work(&execmem_cache_clean_work);
 
 	return true;
 }
 
-int execmem_make_temp_rw(void *ptr, size_t size)
-{
-	unsigned int nr = PAGE_ALIGN(size) >> PAGE_SHIFT;
-	unsigned long addr = (unsigned long)ptr;
-	int ret;
-
-	ret = set_memory_nx(addr, nr);
-	if (ret)
-		return ret;
-
-	return set_memory_rw(addr, nr);
-}
-
-int execmem_restore_rox(void *ptr, size_t size)
-{
-	unsigned int nr = PAGE_ALIGN(size) >> PAGE_SHIFT;
-	unsigned long addr = (unsigned long)ptr;
-
-	return set_memory_rox(addr, nr);
-}
-
 #else /* CONFIG_ARCH_HAS_EXECMEM_ROX */
+/*
+ * when ROX cache is not used the permissions defined by architectures for
+ * execmem ranges that are updated before use (e.g. EXECMEM_MODULE_TEXT) must
+ * be writable anyway
+ */
+static inline int execmem_force_rw(void *ptr, size_t size)
+{
+	return 0;
+}
+
 static void *execmem_cache_alloc(struct execmem_range *range, size_t size)
 {
 	return NULL;
···
 {
 	struct execmem_range *range = &execmem_info->ranges[type];
 	bool use_cache = range->flags & EXECMEM_ROX_CACHE;
-	vm_flags_t vm_flags = VM_FLUSH_RESET_PERMS;
+	unsigned long vm_flags = VM_FLUSH_RESET_PERMS;
 	pgprot_t pgprot = range->pgprot;
-	void *p;
+	void *p = NULL;
 
 	size = PAGE_ALIGN(size);
 
···
 		p = execmem_vmalloc(range, size, pgprot, vm_flags);
 
 	return kasan_reset_tag(p);
+}
+
+void *execmem_alloc_rw(enum execmem_type type, size_t size)
+{
+	void *p __free(execmem) = execmem_alloc(type, size);
+	int err;
+
+	if (!p)
+		return NULL;
+
+	err = execmem_force_rw(p, size);
+	if (err)
+		return NULL;
+
+	return no_free_ptr(p);
 }
 
 void execmem_free(void *ptr)
···
 
 	if (!execmem_cache_free(ptr))
 		vfree(ptr);
-}
-
-void *execmem_update_copy(void *dst, const void *src, size_t size)
-{
-	return text_poke_copy(dst, src, size);
 }
 
 bool execmem_is_rox(enum execmem_type type)

mm/internal.h

···
 
 struct vm_struct *__get_vm_area_node(unsigned long size,
 				     unsigned long align, unsigned long shift,
-				     vm_flags_t vm_flags, unsigned long start,
+				     unsigned long vm_flags, unsigned long start,
 				     unsigned long end, int node, gfp_t gfp_mask,
 				     const void *caller);
 

mm/io-mapping.c

···
-// SPDX-License-Identifier: GPL-2.0-only
-
-#include <linux/mm.h>
-#include <linux/io-mapping.h>
-
-/**
- * io_mapping_map_user - remap an I/O mapping to userspace
- * @iomap: the source io_mapping
- * @vma: user vma to map to
- * @addr: target user address to start at
- * @pfn: physical address of kernel memory
- * @size: size of map area
- *
- *  Note: this is only safe if the mm semaphore is held when called.
- */
-int io_mapping_map_user(struct io_mapping *iomap, struct vm_area_struct *vma,
-		unsigned long addr, unsigned long pfn, unsigned long size)
-{
-	vm_flags_t expected_flags = VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP;
-
-	if (WARN_ON_ONCE((vma->vm_flags & expected_flags) != expected_flags))
-		return -EINVAL;
-
-	pgprot_t remap_prot = __pgprot((pgprot_val(iomap->prot) & _PAGE_CACHE_MASK) |
-				       (pgprot_val(vma->vm_page_prot) & ~_PAGE_CACHE_MASK));
-
-	/* We rely on prevalidation of the io-mapping to skip pfnmap tracking. */
-	return remap_pfn_range_notrack(vma, addr, pfn, size, remap_prot);
-}
-EXPORT_SYMBOL_GPL(io_mapping_map_user);

mm/kasan/common.c

···
 }
 
 static inline void poison_slab_object(struct kmem_cache *cache, void *object,
-				      bool init, bool still_accessible)
+				      bool init)
 {
 	void *tagged_object = object;
 
 	object = kasan_reset_tag(object);
-
-	/* RCU slabs could be legally used after free within the RCU period. */
-	if (unlikely(still_accessible))
-		return;
 
 	kasan_poison(object, round_up(cache->object_size, KASAN_GRANULE_SIZE),
 			KASAN_SLAB_FREE, init);
···
 	if (!kasan_arch_is_ready() || is_kfence_address(object))
 		return false;
 
-	poison_slab_object(cache, object, init, still_accessible);
+	/*
+	 * If this point is reached with an object that must still be
+	 * accessible under RCU, we can't poison it; in that case, also skip the
+	 * quarantine. This should mostly only happen when CONFIG_SLUB_RCU_DEBUG
+	 * has been disabled manually.
+	 *
+	 * Putting the object on the quarantine wouldn't help catch UAFs (since
+	 * we can't poison it here), and it would mask bugs caused by
+	 * SLAB_TYPESAFE_BY_RCU users not being careful enough about object
+	 * reuse; so overall, putting the object into the quarantine here would
+	 * be counterproductive.
+	 */
+	if (still_accessible)
+		return false;
+
+	poison_slab_object(cache, object, init);
 
 	/*
 	 * If the object is put into quarantine, do not let slab put the object
···
 	if (check_slab_allocation(slab->slab_cache, ptr, ip))
 		return false;
 
-	poison_slab_object(slab->slab_cache, ptr, false, false);
+	poison_slab_object(slab->slab_cache, ptr, false);
 	return true;
 }
 

mm/khugepaged.c

···
 						spinlock_t *ptl,
 						struct list_head *compound_pagelist)
 {
+	unsigned long end = address + HPAGE_PMD_SIZE;
 	struct folio *src, *tmp;
-	pte_t *_pte;
 	pte_t pteval;
+	pte_t *_pte;
+	unsigned int nr_ptes;
 
-	for (_pte = pte; _pte < pte + HPAGE_PMD_NR;
-	     _pte++, address += PAGE_SIZE) {
+	for (_pte = pte; _pte < pte + HPAGE_PMD_NR; _pte += nr_ptes,
+	     address += nr_ptes * PAGE_SIZE) {
+		nr_ptes = 1;
 		pteval = ptep_get(_pte);
 		if (pte_none(pteval) || is_zero_pfn(pte_pfn(pteval))) {
 			add_mm_counter(vma->vm_mm, MM_ANONPAGES, 1);
···
 			struct page *src_page = pte_page(pteval);
 
 			src = page_folio(src_page);
-			if (!folio_test_large(src))
+
+			if (folio_test_large(src)) {
+				unsigned int max_nr_ptes = (end - address) >> PAGE_SHIFT;
+
+				nr_ptes = folio_pte_batch(src, _pte, pteval, max_nr_ptes);
+			} else {
 				release_pte_folio(src);
+			}
+
 			/*
 			 * ptl mostly unnecessary, but preempt has to
 			 * be disabled to update the per-cpu stats
 			 * inside folio_remove_rmap_pte().
 			 */
 			spin_lock(ptl);
-			ptep_clear(vma->vm_mm, address, _pte);
-			folio_remove_rmap_pte(src, src_page, vma);
+			clear_ptes(vma->vm_mm, address, _pte, nr_ptes);
+			folio_remove_rmap_ptes(src, src_page, nr_ptes, vma);
 			spin_unlock(ptl);
-			free_folio_and_swap_cache(src);
+			free_swap_cache(src);
+			folio_put_refs(src, nr_ptes);
 		}
 	}
 
···
 int collapse_pte_mapped_thp(struct mm_struct *mm, unsigned long addr,
 			    bool install_pmd)
 {
+	int nr_mapped_ptes = 0, result = SCAN_FAIL;
+	unsigned int nr_batch_ptes;
 	struct mmu_notifier_range range;
 	bool notified = false;
 	unsigned long haddr = addr & HPAGE_PMD_MASK;
+	unsigned long end = haddr + HPAGE_PMD_SIZE;
 	struct vm_area_struct *vma = vma_lookup(mm, haddr);
 	struct folio *folio;
 	pte_t *start_pte, *pte;
 	pmd_t *pmd, pgt_pmd;
 	spinlock_t *pml = NULL, *ptl;
-	int nr_ptes = 0, result = SCAN_FAIL;
 	int i;
 
 	mmap_assert_locked(mm);
···
 		goto abort;
 
 	/* step 2: clear page table and adjust rmap */
-	for (i = 0, addr = haddr, pte = start_pte;
-	     i < HPAGE_PMD_NR; i++, addr += PAGE_SIZE, pte++) {
+	for (i = 0, addr = haddr, pte = start_pte; i < HPAGE_PMD_NR;
+	     i += nr_batch_ptes, addr += nr_batch_ptes * PAGE_SIZE,
+	     pte += nr_batch_ptes) {
+		unsigned int max_nr_batch_ptes = (end - addr) >> PAGE_SHIFT;
 		struct page *page;
 		pte_t ptent = ptep_get(pte);
+
+		nr_batch_ptes = 1;
 
 		if (pte_none(ptent))
 			continue;
···
 			goto abort;
 		}
 		page = vm_normal_page(vma, addr, ptent);
+
 		if (folio_page(folio, i) != page)
 			goto abort;
+
+		nr_batch_ptes = folio_pte_batch(folio, pte, ptent, max_nr_batch_ptes);
 
 		/*
 		 * Must clear entry, or a racing truncate may re-remove it.
 		 * TLB flush can be left until pmdp_collapse_flush() does it.
 		 * PTE dirty? Shmem page is already dirty; file is read-only.
 		 */
-		ptep_clear(mm, addr, pte);
-		folio_remove_rmap_pte(folio, page, vma);
-		nr_ptes++;
+		clear_ptes(mm, addr, pte, nr_batch_ptes);
+		folio_remove_rmap_ptes(folio, page, nr_batch_ptes, vma);
+		nr_mapped_ptes += nr_batch_ptes;
 	}
 
 	if (!pml)
 		spin_unlock(ptl);
 
 	/* step 3: set proper refcount and mm_counters. */
-	if (nr_ptes) {
-		folio_ref_sub(folio, nr_ptes);
-		add_mm_counter(mm, mm_counter_file(folio), -nr_ptes);
+	if (nr_mapped_ptes) {
+		folio_ref_sub(folio, nr_mapped_ptes);
+		add_mm_counter(mm, mm_counter_file(folio), -nr_mapped_ptes);
 	}
 
 	/* step 4: remove empty page table */
···
 			: SCAN_SUCCEED;
 	goto drop_folio;
 abort:
-	if (nr_ptes) {
+	if (nr_mapped_ptes) {
 		flush_tlb_mm(mm);
-		folio_ref_sub(folio, nr_ptes);
-		add_mm_counter(mm, mm_counter_file(folio), -nr_ptes);
+		folio_ref_sub(folio, nr_mapped_ptes);
+		add_mm_counter(mm, mm_counter_file(folio), -nr_mapped_ptes);
 	}
 unlock:
 	if (start_pte)

mm/madvise.c

···
 #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>
···
 			       &guard_remove_walk_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
···
 	struct madvise_behavior_range *range = &madv_behavior->range;
 	int error;
 
-	if (unlikely(!can_modify_vma_madv(madv_behavior->vma, behavior)))
+	if (unlikely(!can_madvise_modify(madv_behavior)))
 		return -EPERM;
 
 	switch (behavior) {

mm/memory-failure.c

···
 			    struct mm_walk *walk)
 {
 	struct hwpoison_walk *hwp = walk->private;
-	pte_t pte = huge_ptep_get(walk->mm, addr, ptep);
 	struct hstate *h = hstate_vma(walk->vma);
+	spinlock_t *ptl;
+	pte_t pte;
+	int ret;
 
-	return check_hwpoisoned_entry(pte, addr, huge_page_shift(h),
-				      hwp->pfn, &hwp->tk);
+	ptl = huge_pte_lock(h, walk->mm, ptep);
+	pte = huge_ptep_get(walk->mm, addr, ptep);
+	ret = check_hwpoisoned_entry(pte, addr, huge_page_shift(h),
+					hwp->pfn, &hwp->tk);
+	spin_unlock(ptl);
+	return ret;
 }
 #else
 #define hwpoison_hugetlb_range	NULL

mm/mempool.c

···
 
 static __always_inline void add_element(mempool_t *pool, void *element)
 {
-	BUG_ON(pool->curr_nr >= pool->min_nr);
+	BUG_ON(pool->min_nr != 0 && pool->curr_nr >= pool->min_nr);
 	poison_element(pool, element);
 	if (kasan_poison_element(pool, element))
 		pool->elements[pool->curr_nr++] = element;
···
 	pool->alloc	= alloc_fn;
 	pool->free	= free_fn;
 	init_waitqueue_head(&pool->wait);
-
-	pool->elements = kmalloc_array_node(min_nr, sizeof(void *),
+	/*
+	 * max() used here to ensure storage for at least 1 element to support
+	 * zero minimum pool
+	 */
+	pool->elements = kmalloc_array_node(max(1, min_nr), sizeof(void *),
 					    gfp_mask, node_id);
 	if (!pool->elements)
 		return -ENOMEM;
 
 	/*
-	 * First pre-allocate the guaranteed number of buffers.
+	 * First pre-allocate the guaranteed number of buffers,
+	 * also pre-allocate 1 element for zero minimum pool.
 	 */
-	while (pool->curr_nr < pool->min_nr) {
+	while (pool->curr_nr < max(1, pool->min_nr)) {
 		void *element;
 
 		element = pool->alloc(gfp_mask, pool->pool_data);
···
 	 * wake-up path of previous test. This explicit check ensures the
 	 * allocation of element when both min_nr and curr_nr are 0, and
 	 * any active waiters are properly awakened.
-	 *
-	 * Inline the same logic as previous test, add_element() cannot be
-	 * directly used here since it has BUG_ON to deny if min_nr equals
-	 * curr_nr, so here picked rest of add_element() to use without
-	 * BUG_ON check.
 	 */
 	if (unlikely(pool->min_nr == 0 &&
 		     READ_ONCE(pool->curr_nr) == 0)) {
 		spin_lock_irqsave(&pool->lock, flags);
 		if (likely(pool->curr_nr == 0)) {
-			/* Inline the logic of add_element() */
-			poison_element(pool, element);
-			if (kasan_poison_element(pool, element))
-				pool->elements[pool->curr_nr++] = element;
+			add_element(pool, element);
 			spin_unlock_irqrestore(&pool->lock, flags);
 			if (wq_has_sleeper(&pool->wait))
 				wake_up(&pool->wait);

mm/mincore.c

···
 #ifdef CONFIG_HUGETLB_PAGE
 	unsigned char present;
 	unsigned char *vec = walk->private;
+	spinlock_t *ptl;
 
+	ptl = huge_pte_lock(hstate_vma(walk->vma), walk->mm, pte);
 	/*
 	 * Hugepages under user process are always in RAM and never
 	 * swapped out, but theoretically it needs to be checked.
···
 	for (; addr != end; vec++, addr += PAGE_SIZE)
 		*vec = present;
 	walk->private = vec;
+	spin_unlock(ptl);
 #else
 	BUG();
 #endif

mm/mmap_lock.c

···
 	 */
 
 	/* Check if the vma we locked is the right one. */
-	if (unlikely(vma->vm_mm != mm ||
-		     address < vma->vm_start || address >= vma->vm_end))
+	if (unlikely(address < vma->vm_start || address >= vma->vm_end))
 		goto inval_end_read;
 
 	rcu_read_unlock();
···
 		goto fallback;
 	}
 
-	/*
-	 * Verify the vma we locked belongs to the same address space and it's
-	 * not behind of the last search position.
-	 */
-	if (unlikely(vma->vm_mm != mm || from_addr >= vma->vm_end))
+	/* Verify the vma is not behind the last search position. */
+	if (unlikely(from_addr >= vma->vm_end))
 		goto fallback_unlock;
 
 	/*

mm/mprotect.c

···
 	unsigned long charged = 0;
 	int error;
 
-	if (!can_modify_vma(vma))
+	if (vma_is_sealed(vma))
 		return -EPERM;
 
 	if (newflags == oldflags) {

mm/mremap.c

···
 							 old_pte, max_nr_ptes);
 			force_flush = true;
 		}
-		pte = get_and_clear_full_ptes(mm, old_addr, old_ptep, nr_ptes, 0);
+		pte = get_and_clear_ptes(mm, old_addr, old_ptep, nr_ptes);
 		pte = move_pte(pte, old_addr, new_addr);
 		pte = move_soft_dirty_pte(pte);
 
···
 		return -EFAULT;
 
 	/* If mseal()'d, mremap() is prohibited. */
-	if (!can_modify_vma(vma))
+	if (vma_is_sealed(vma))
 		return -EPERM;
 
 	/* Align to hugetlb page size, if required. */

mm/mseal.c

···
 #include <linux/mman.h>
 #include <linux/mm.h>
 #include <linux/mm_inline.h>
-#include <linux/mmu_context.h>
 #include <linux/syscalls.h>
 #include <linux/sched.h>
 #include "internal.h"
 
-static inline void set_vma_sealed(struct vm_area_struct *vma)
-{
-	vm_flags_set(vma, VM_SEALED);
-}
-
-static bool is_madv_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;
-}
-
-static bool is_ro_anon(struct vm_area_struct *vma)
-{
-	/* check anonymous mapping. */
-	if (vma->vm_file || vma->vm_flags & VM_SHARED)
-		return false;
-
-	/*
-	 * check for non-writable:
-	 * PROT=RO or PKRU is not writeable.
-	 */
-	if (!(vma->vm_flags & VM_WRITE) ||
-		!arch_vma_access_permitted(vma, true, false, false))
-		return true;
-
-	return false;
-}
+/*
+ * mseal() disallows an input range which contain unmapped ranges (VMA holes).
+ *
+ * It disallows unmapped regions from start to end whether they exist at the
+ * start, in the middle, or at the end of the range, or any combination thereof.
+ *
+ * This is because after sealng a range, there's nothing to stop memory mapping
+ * of ranges in the remaining gaps later, meaning that the user might then
+ * wrongly consider the entirety of the mseal()'d range to be sealed when it
+ * in fact isn't.
+ */
 
 /*
- * Check if a vma is allowed to be modified by madvise.
+ * Does the [start, end) range contain any unmapped memory?
+ *
+ * We ensure that:
+ * - start is part of a valid VMA.
+ * - end is part of a valid VMA.
+ * - no gap (unallocated memory) exists between start and end.
  */
-bool can_modify_vma_madv(struct vm_area_struct *vma, int behavior)
-{
-	if (!is_madv_discard(behavior))
-		return true;
-
-	if (unlikely(!can_modify_vma(vma) && is_ro_anon(vma)))
-		return false;
-
-	/* Allow by default. */
-	return true;
-}
-
-static int mseal_fixup(struct vma_iterator *vmi, struct vm_area_struct *vma,
-		struct vm_area_struct **prev, unsigned long start,
-		unsigned long end, vm_flags_t newflags)
-{
-	int ret = 0;
-	vm_flags_t oldflags = vma->vm_flags;
-
-	if (newflags == oldflags)
-		goto out;
-
-	vma = vma_modify_flags(vmi, *prev, vma, start, end, newflags);
-	if (IS_ERR(vma)) {
-		ret = PTR_ERR(vma);
-		goto out;
-	}
-
-	set_vma_sealed(vma);
-out:
-	*prev = vma;
-	return ret;
-}
-
-/*
- * Check for do_mseal:
- * 1> start is part of a valid vma.
- * 2> end is part of a valid vma.
- * 3> No gap (unallocated address) between start and end.
- * 4> map is sealable.
- */
-static int check_mm_seal(unsigned long start, unsigned long end)
+static bool range_contains_unmapped(struct mm_struct *mm,
+		unsigned long start, unsigned long end)
 {
 	struct vm_area_struct *vma;
-	unsigned long nstart = start;
-
+	unsigned long prev_end = start;
 	VMA_ITERATOR(vmi, current->mm, start);
 
-	/* going through each vma to check. */
 	for_each_vma_range(vmi, vma, end) {
-		if (vma->vm_start > nstart)
-			/* unallocated memory found. */
-			return -ENOMEM;
+		if (vma->vm_start > prev_end)
+			return true;
 
-		if (vma->vm_end >= end)
-			return 0;
-
-		nstart = vma->vm_end;
+		prev_end = vma->vm_end;
 	}
 
-	return -ENOMEM;
+	return prev_end < end;
 }
 
-/*
- * Apply sealing.
- */
-static int apply_mm_seal(unsigned long start, unsigned long end)
+static int mseal_apply(struct mm_struct *mm,
+		unsigned long start, unsigned long end)
 {
-	unsigned long nstart;
 	struct vm_area_struct *vma, *prev;
+	unsigned long curr_start = start;
+	VMA_ITERATOR(vmi, mm, start);
 
-	VMA_ITERATOR(vmi, current->mm, start);
-
+	/* We know there are no gaps so this will be non-NULL. */
 	vma = vma_iter_load(&vmi);
-	/*
-	 * Note: check_mm_seal should already checked ENOMEM case.
-	 * so vma should not be null, same for the other ENOMEM cases.
-	 */
 	prev = vma_prev(&vmi);
 	if (start > vma->vm_start)
 		prev = vma;
 
-	nstart = start;
 	for_each_vma_range(vmi, vma, end) {
-		int error;
-		unsigned long tmp;
-		vm_flags_t newflags;
+		unsigned long curr_end = MIN(vma->vm_end, end);
 
-		newflags = vma->vm_flags | VM_SEALED;
-		tmp = vma->vm_end;
-		if (tmp > end)
-			tmp = end;
-		error = mseal_fixup(&vmi, vma, &prev, nstart, tmp, newflags);
-		if (error)
-			return error;
-		nstart = vma_iter_end(&vmi);
+		if (!(vma->vm_flags & VM_SEALED)) {
+			vma = vma_modify_flags(&vmi, prev, vma,
+					curr_start, curr_end,
+					vma->vm_flags | VM_SEALED);
+			if (IS_ERR(vma))
+				return PTR_ERR(vma);
+			vm_flags_set(vma, VM_SEALED);
+		}
+
+		prev = vma;
+		curr_start = curr_end;
 	}
 
 	return 0;
···
 	if (mmap_write_lock_killable(mm))
 		return -EINTR;
 
-	/*
-	 * First pass, this helps to avoid
-	 * partial sealing in case of error in input address range,
-	 * e.g. ENOMEM error.
-	 */
-	ret = check_mm_seal(start, end);
-	if (ret)
+	if (range_contains_unmapped(mm, start, end)) {
+		ret = -ENOMEM;
 		goto out;
+	}
 
 	/*
 	 * Second pass, this should success, unless there are errors
···
 	 * reaching the max supported VMAs, however, those cases shall
 	 * be rare.
 	 */
-	ret = apply_mm_seal(start, end);
+	ret = mseal_apply(mm, start, end);
 
 out:
-	mmap_write_unlock(current->mm);
+	mmap_write_unlock(mm);
 	return ret;
 }
 

mm/nommu.c

···
 
 void *__vmalloc_node_range_noprof(unsigned long size, unsigned long align,
 		unsigned long start, unsigned long end, gfp_t gfp_mask,
-		pgprot_t prot, vm_flags_t vm_flags, int node,
+		pgprot_t prot, unsigned long vm_flags, int node,
 		const void *caller)
 {
 	return __vmalloc_noprof(size, gfp_mask);

mm/rmap.c

···
 			flush_cache_range(vma, address, end_addr);
 
 			/* Nuke the page table entry. */
-			pteval = get_and_clear_full_ptes(mm, address, pvmw.pte, nr_pages, 0);
+			pteval = get_and_clear_ptes(mm, address, pvmw.pte, nr_pages);
 			/*
 			 * We clear the PTE but do not flush so potentially
 			 * a remote CPU could still be writing to the folio.

mm/shmem.c

···
 
 /*
  * Sometimes, before we decide whether to proceed or to fail, we must check
- * that an entry was not already brought back from swap by a racing thread.
+ * that an entry was not already brought back or split by a racing thread.
  *
  * Checking folio is not enough: by the time a swapcache folio is locked, it
  * might be reused, and again be swapcache, using the same swap as before.
+ * Returns the swap entry's order if it still presents, else returns -1.
  */
-static bool shmem_confirm_swap(struct address_space *mapping,
-			       pgoff_t index, swp_entry_t swap)
+static int shmem_confirm_swap(struct address_space *mapping, pgoff_t index,
+			      swp_entry_t swap)
 {
-	return xa_load(&mapping->i_pages, index) == swp_to_radix_entry(swap);
+	XA_STATE(xas, &mapping->i_pages, index);
+	int ret = -1;
+	void *entry;
+
+	rcu_read_lock();
+	do {
+		entry = xas_load(&xas);
+		if (entry == swp_to_radix_entry(swap))
+			ret = xas_get_order(&xas);
+	} while (xas_retry(&xas, entry));
+	rcu_read_unlock();
+	return ret;
 }
 
 /*
···
 				   pgoff_t index, void *expected, gfp_t gfp)
 {
 	XA_STATE_ORDER(xas, &mapping->i_pages, index, folio_order(folio));
-	long nr = folio_nr_pages(folio);
+	unsigned long nr = folio_nr_pages(folio);
+	swp_entry_t iter, swap;
+	void *entry;
 
 	VM_BUG_ON_FOLIO(index != round_down(index, nr), folio);
 	VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio);
···
 
 	gfp &= GFP_RECLAIM_MASK;
 	folio_throttle_swaprate(folio, gfp);
+	swap = radix_to_swp_entry(expected);
 
 	do {
+		iter = swap;
 		xas_lock_irq(&xas);
-		if (expected != xas_find_conflict(&xas)) {
-			xas_set_err(&xas, -EEXIST);
-			goto unlock;
+		xas_for_each_conflict(&xas, entry) {
+			/*
+			 * The range must either be empty, or filled with
+			 * expected swap entries. Shmem swap entries are never
+			 * partially freed without split of both entry and
+			 * folio, so there shouldn't be any holes.
+			 */
+			if (!expected || entry != swp_to_radix_entry(iter)) {
+				xas_set_err(&xas, -EEXIST);
+				goto unlock;
+			}
+			iter.val += 1 << xas_get_order(&xas);
 		}
-		if (expected && xas_find_conflict(&xas)) {
+		if (expected && iter.val - nr != swap.val) {
 			xas_set_err(&xas, -EEXIST);
 			goto unlock;
 		}
···
 		swp_entry_t entry, int order, gfp_t gfp)
 {
 	struct shmem_inode_info *info = SHMEM_I(inode);
+	int nr_pages = 1 << order;
 	struct folio *new;
+	gfp_t alloc_gfp;
 	void *shadow;
-	int nr_pages;
 
 	/*
 	 * We have arrived here because our zones are constrained, so don't
 	 * limit chance of success with further cpuset and node constraints.
 	 */
 	gfp &= ~GFP_CONSTRAINT_MASK;
-	if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) && order > 0) {
-		gfp_t huge_gfp = vma_thp_gfp_mask(vma);
+	alloc_gfp = gfp;
+	if (!IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE)) {
+		if (WARN_ON_ONCE(order))
+			return ERR_PTR(-EINVAL);
+	} else if (order) {
+		/*
+		 * If uffd is active for the vma, we need per-page fault
+		 * fidelity to maintain the uffd semantics, then fallback
+		 * to swapin order-0 folio, as well as for zswap case.
+		 * Any existing sub folio in the swap cache also blocks
+		 * mTHP swapin.
+		 */
+		if ((vma && unlikely(userfaultfd_armed(vma))) ||
+		     !zswap_never_enabled() ||
+		     non_swapcache_batch(entry, nr_pages) != nr_pages)
+			goto fallback;
 
-		gfp = limit_gfp_mask(huge_gfp, gfp);
+		alloc_gfp = limit_gfp_mask(vma_thp_gfp_mask(vma), gfp);
+	}
+retry:
+	new = shmem_alloc_folio(alloc_gfp, order, info, index);
+	if (!new) {
+		new = ERR_PTR(-ENOMEM);
+		goto fallback;
 	}
 
-	new = shmem_alloc_folio(gfp, order, info, index);
-	if (!new)
-		return ERR_PTR(-ENOMEM);
-
-	nr_pages = folio_nr_pages(new);
 	if (mem_cgroup_swapin_charge_folio(new, vma ? vma->vm_mm : NULL,
-					   gfp, entry)) {
+					   alloc_gfp, entry)) {
 		folio_put(new);
-		return ERR_PTR(-ENOMEM);
+		new = ERR_PTR(-ENOMEM);
+		goto fallback;
 	}
 
 	/*
···
 	 */
 	if (swapcache_prepare(entry, nr_pages)) {
 		folio_put(new);
-		return ERR_PTR(-EEXIST);
+		new = ERR_PTR(-EEXIST);
+		/* Try smaller folio to avoid cache conflict */
+		goto fallback;
 	}
 
 	__folio_set_locked(new);
···
 	folio_add_lru(new);
 	swap_read_folio(new, NULL);
 	return new;
+fallback:
+	/* Order 0 swapin failed, nothing to fallback to, abort */
+	if (!order)
+		return new;
+	entry.val += index - round_down(index, nr_pages);
+	alloc_gfp = gfp;
+	nr_pages = 1;
+	order = 0;
+	goto retry;
 }
 
 /*
···
 	if (xas_error(&xas))
 		return xas_error(&xas);
 
-	return entry_order;
+	return 0;
 }
 
 /*
···
 	struct address_space *mapping = inode->i_mapping;
 	struct mm_struct *fault_mm = vma ? vma->vm_mm : NULL;
 	struct shmem_inode_info *info = SHMEM_I(inode);
+	swp_entry_t swap, index_entry;
 	struct swap_info_struct *si;
 	struct folio *folio = NULL;
 	bool skip_swapcache = false;
-	swp_entry_t swap;
-	int error, nr_pages, order, split_order;
+	int error, nr_pages, order;
+	pgoff_t offset;
 
 	VM_BUG_ON(!*foliop || !xa_is_value(*foliop));
-	swap = radix_to_swp_entry(*foliop);
+	index_entry = radix_to_swp_entry(*foliop);
+	swap = index_entry;
 	*foliop = NULL;
 
-	if (is_poisoned_swp_entry(swap))
+	if (is_poisoned_swp_entry(index_entry))
 		return -EIO;
 
-	si = get_swap_device(swap);
-	if (!si) {
-		if (!shmem_confirm_swap(mapping, index, swap))
+	si = get_swap_device(index_entry);
+	order = shmem_confirm_swap(mapping, index, index_entry);
+	if (unlikely(!si)) {
+		if (order < 0)
 			return -EEXIST;
 		else
 			return -EINVAL;
 	}
+	if (unlikely(order < 0)) {
+		put_swap_device(si);
+		return -EEXIST;
+	}
+
+	/* index may point to the middle of a large entry, get the sub entry */
+	if (order) {
+		offset = index - round_down(index, 1 << order);
+		swap = swp_entry(swp_type(swap), swp_offset(swap) + offset);
+	}
 
 	/* Look it up and read it in.. */
 	folio = swap_cache_get_folio(swap, NULL, 0);
-	order = xa_get_order(&mapping->i_pages, index);
 	if (!folio) {
-		int nr_pages = 1 << order;
-		bool fallback_order0 = false;
-
-		/* Or update major stats only when swapin succeeds?? */
+		if (data_race(si->flags & SWP_SYNCHRONOUS_IO)) {
+			/* Direct swapin skipping swap cache & readahead */
+			folio = shmem_swap_alloc_folio(inode, vma, index,
+						       index_entry, order, gfp);
+			if (IS_ERR(folio)) {
+				error = PTR_ERR(folio);
+				folio = NULL;
+				goto failed;
+			}
+			skip_swapcache = true;
+		} else {
+			/* Cached swapin only supports order 0 folio */
+			folio = shmem_swapin_cluster(swap, gfp, info, index);
+			if (!folio) {
+				error = -ENOMEM;
+				goto failed;
+			}
+		}
 		if (fault_type) {
 			*fault_type |= VM_FAULT_MAJOR;
 			count_vm_event(PGMAJFAULT);
 			count_memcg_event_mm(fault_mm, PGMAJFAULT);
 		}
+	}
 
+	if (order > folio_order(folio)) {
 		/*
-		 * If uffd is active for the vma, we need per-page fault
-		 * fidelity to maintain the uffd semantics, then fallback
-		 * to swapin order-0 folio, as well as for zswap case.
-		 * Any existing sub folio in the swap cache also blocks
-		 * mTHP swapin.
-		 */
-		if (order > 0 && ((vma && unlikely(userfaultfd_armed(vma))) ||
-				  !zswap_never_enabled() ||
-				  non_swapcache_batch(swap, nr_pages) != nr_pages))
-			fallback_order0 = true;
-
-		/* Skip swapcache for synchronous device. */
-		if (!fallback_order0 && data_race(si->flags & SWP_SYNCHRONOUS_IO)) {
-			folio = shmem_swap_alloc_folio(inode, vma, index, swap, order, gfp);
-			if (!IS_ERR(folio)) {
-				skip_swapcache = true;
-				goto alloced;
-			}
-
-			/*
-			 * Fallback to swapin order-0 folio unless the swap entry
-			 * already exists.
-			 */
-			error = PTR_ERR(folio);
-			folio = NULL;
-			if (error == -EEXIST)
-				goto failed;
-		}
-
-		/*
-		 * Now swap device can only swap in order 0 folio, then we
-		 * should split the large swap entry stored in the pagecache
-		 * if necessary.
-		 */
-		split_order = shmem_split_large_entry(inode, index, swap, gfp);
-		if (split_order < 0) {
-			error = split_order;
-			goto failed;
-		}
-
-		/*
-		 * If the large swap entry has already been split, it is
-		 * necessary to recalculate the new swap entry based on
-		 * the old order alignment.
-		 */
-		if (split_order > 0) {
-			pgoff_t offset = index - round_down(index, 1 << split_order);
-
-			swap = swp_entry(swp_type(swap), swp_offset(swap) + offset);
-		}
-
-		/* Here we actually start the io */
-		folio = shmem_swapin_cluster(swap, gfp, info, index);
-		if (!folio) {
-			error = -ENOMEM;
-			goto failed;
-		}
-	} else if (order != folio_order(folio)) {
-		/*
-		 * Swap readahead may swap in order 0 folios into swapcache
+		 * Swapin may get smaller folios due to various reasons:
+		 * It may fallback to order 0 due to memory pressure or race,
+		 * swap readahead may swap in order 0 folios into swapcache
 		 * asynchronously, while the shmem mapping can still stores
 		 * large swap entries. In such cases, we should split the
 		 * large swap entry to prevent possible data corruption.
 		 */
-		split_order = shmem_split_large_entry(inode, index, swap, gfp);
-		if (split_order < 0) {
-			folio_put(folio);
-			folio = NULL;
-			error = split_order;
-			goto failed;
-		}
-
-		/*
-		 * If the large swap entry has already been split, it is
-		 * necessary to recalculate the new swap entry based on
-		 * the old order alignment.
-		 */
-		if (split_order > 0) {
-			pgoff_t offset = index - round_down(index, 1 << split_order);
-
-			swap = swp_entry(swp_type(swap), swp_offset(swap) + offset);
-		}
+		error = shmem_split_large_entry(inode, index, index_entry, gfp);
+		if (error)
+			goto failed_nolock;
 	}
 
-alloced:
-	/* We have to do this with folio locked to prevent races */
+	/*
+	 * If the folio is large, round down swap and index by folio size.
+	 * No matter what race occurs, the swap layer ensures we either get
+	 * a valid folio that has its swap entry aligned by size, or a
+	 * temporarily invalid one which we'll abort very soon and retry.
+	 *
+	 * shmem_add_to_page_cache ensures the whole range contains expected
+	 * entries and prevents any corruption, so any race split is fine
+	 * too, it will succeed as long as the entries are still there.
+	 */
+	nr_pages = folio_nr_pages(folio);
+	if (nr_pages > 1) {
+		swap.val = round_down(swap.val, nr_pages);
+		index = round_down(index, nr_pages);
+	}
+
+	/*
+	 * We have to do this with the folio locked to prevent races.
+	 * The shmem_confirm_swap below only checks if the first swap
+	 * entry matches the folio, that's enough to ensure the folio
+	 * is not used outside of shmem, as shmem swap entries
+	 * and swap cache folios are never partially freed.
+	 */
 	folio_lock(folio);
 	if ((!skip_swapcache && !folio_test_swapcache(folio)) ||
-	    folio->swap.val != swap.val ||
-	    !shmem_confirm_swap(mapping, index, swap) ||
-	    xa_get_order(&mapping->i_pages, index) != folio_order(folio)) {
+	    shmem_confirm_swap(mapping, index, swap) < 0 ||
+	    folio->swap.val != swap.val) {
 		error = -EEXIST;
 		goto unlock;
 	}
···
 			goto failed;
 	}
 
-	error = shmem_add_to_page_cache(folio, mapping,
-					round_down(index, nr_pages),
+	error = shmem_add_to_page_cache(folio, mapping, index,
 					swp_to_radix_entry(swap), gfp);
 	if (error)
 		goto failed;
···
 	*foliop = folio;
 	return 0;
 failed:
-	if (!shmem_confirm_swap(mapping, index, swap))
+	if (shmem_confirm_swap(mapping, index, swap) < 0)
 		error = -EEXIST;
 	if (error == -EIO)
 		shmem_set_folio_swapin_error(inode, index, folio, swap,
 					     skip_swapcache);
 unlock:
-	if (skip_swapcache)
-		swapcache_clear(si, swap, folio_nr_pages(folio));
-	if (folio) {
+	if (folio)
 		folio_unlock(folio);
+failed_nolock:
+	if (skip_swapcache)
+		swapcache_clear(si, folio->swap, folio_nr_pages(folio));
+	if (folio)
 		folio_put(folio);
-	}
 	put_swap_device(si);
 
 	return error;
···
 	struct folio *folio;
 	int error;
 
-	error = shmem_get_folio_gfp(inode, index, 0, &folio, SGP_CACHE,
-				    gfp, NULL, NULL);
+	error = shmem_get_folio_gfp(inode, index, i_size_read(inode),
+				    &folio, SGP_CACHE, gfp, NULL, NULL);
 	if (error)
 		return ERR_PTR(error);
 

mm/vma.c

···
 		}
 
 		/* Don't bother splitting the VMA if we can't unmap it anyway */
-		if (!can_modify_vma(vms->vma)) {
+		if (vma_is_sealed(vms->vma)) {
 			error = -EPERM;
 			goto start_split_failed;
 		}
···
 	for_each_vma_range(*(vms->vmi), next, vms->end) {
 		long nrpages;
 
-		if (!can_modify_vma(next)) {
+		if (vma_is_sealed(next)) {
 			error = -EPERM;
 			goto modify_vma_failed;
 		}

mm/vma.h

···
 }
 
 #ifdef CONFIG_64BIT
-
 static inline bool vma_is_sealed(struct vm_area_struct *vma)
 {
 	return (vma->vm_flags & VM_SEALED);
 }
-
-/*
- * check if a vma is sealed for modification.
- * return true, if modification is allowed.
- */
-static inline bool can_modify_vma(struct vm_area_struct *vma)
-{
-	if (unlikely(vma_is_sealed(vma)))
-		return false;
-
-	return true;
-}
-
-bool can_modify_vma_madv(struct vm_area_struct *vma, int behavior);
-
 #else
-
-static inline bool can_modify_vma(struct vm_area_struct *vma)
+static inline bool vma_is_sealed(struct vm_area_struct *vma)
 {
-	return true;
+	return false;
 }
-
-static inline bool can_modify_vma_madv(struct vm_area_struct *vma, int behavior)
-{
-	return true;
-}
-
 #endif
 
 #if defined(CONFIG_STACK_GROWSUP)

tools/testing/selftests/cachestat/test_cachestat.c

···
 	cs->nr_evicted, cs->nr_recently_evicted);
 }
 
+enum file_type {
+	FILE_MMAP,
+	FILE_SHMEM
+};
+
 bool write_exactly(int fd, size_t filesize)
 {
 	int random_fd = open("/dev/urandom", O_RDONLY);
···
 out:
 	return ret;
 }
+const char *file_type_str(enum file_type type)
+{
+	switch (type) {
+	case FILE_SHMEM:
+		return "shmem";
+	case FILE_MMAP:
+		return "mmap";
+	default:
+		return "unknown";
+	}
+}
 
-bool test_cachestat_shmem(void)
+
+bool run_cachestat_test(enum file_type type)
 {
 	size_t PS = sysconf(_SC_PAGESIZE);
 	size_t filesize = PS * 512 * 2; /* 2 2MB huge pages */
···
 	char *filename = "tmpshmcstat";
 	struct cachestat cs;
 	bool ret = true;
+	int fd;
 	unsigned long num_pages = compute_len / PS;
-	int fd = shm_open(filename, O_CREAT | O_RDWR, 0600);
+	if (type == FILE_SHMEM)
+		fd = shm_open(filename, O_CREAT | O_RDWR, 0600);
+	else
+		fd = open(filename, O_RDWR | O_CREAT | O_TRUNC, 0666);
 
 	if (fd < 0) {
-		ksft_print_msg("Unable to create shmem file.\n");
+		ksft_print_msg("Unable to create %s file.\n",
+				file_type_str(type));
 		ret = false;
 		goto out;
 	}
 
 	if (ftruncate(fd, filesize)) {
-		ksft_print_msg("Unable to truncate shmem file.\n");
+		ksft_print_msg("Unable to truncate %s file.\n",file_type_str(type));
 		ret = false;
 		goto close_fd;
 	}
+	switch (type) {
+	case FILE_SHMEM:
+		if (!write_exactly(fd, filesize)) {
+			ksft_print_msg("Unable to write to file.\n");
+			ret = false;
+			goto close_fd;
+		}
+		break;
+	case FILE_MMAP:
+		char *map = mmap(NULL, filesize, PROT_READ | PROT_WRITE,
+				 MAP_SHARED, fd, 0);
 
-	if (!write_exactly(fd, filesize)) {
-		ksft_print_msg("Unable to write to shmem file.\n");
+		if (map == MAP_FAILED) {
+			ksft_print_msg("mmap failed.\n");
+			ret = false;
+			goto close_fd;
+		}
+		for (int i = 0; i < filesize; i++)
+			map[i] = 'A';
+		break;
+	default:
+		ksft_print_msg("Unsupported file type.\n");
 		ret = false;
 		goto close_fd;
 	}
-
 	syscall_ret = syscall(__NR_cachestat, fd, &cs_range, &cs, 0);
 
 	if (syscall_ret) {
···
 		break;
 	}
 
-	if (test_cachestat_shmem())
+	if (run_cachestat_test(FILE_SHMEM))
 		ksft_test_result_pass("cachestat works with a shmem file\n");
 	else {
 		ksft_test_result_fail("cachestat fails with a shmem file\n");
 		ret = 1;
 	}
 
+	if (run_cachestat_test(FILE_MMAP))
+		ksft_test_result_pass("cachestat works with a mmap file\n");
+	else {
+		ksft_test_result_fail("cachestat fails with a mmap file\n");
+		ret = 1;
+	}
 	return ret;
 }

tools/testing/selftests/mm/.gitignore

···
 transhuge-stress
 pagemap_ioctl
 pfnmap
+process_madv
 *.tmp*
 protection_keys
 protection_keys_32

tools/testing/selftests/mm/Makefile

···
 TEST_GEN_FILES += on-fault-limit
 TEST_GEN_FILES += pagemap_ioctl
 TEST_GEN_FILES += pfnmap
+TEST_GEN_FILES += process_madv
 TEST_GEN_FILES += thuge-gen
 TEST_GEN_FILES += transhuge-stress
 TEST_GEN_FILES += uffd-stress

tools/testing/selftests/mm/process_madv.c

···
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#define _GNU_SOURCE
+#include "../kselftest_harness.h"
+#include <errno.h>
+#include <setjmp.h>
+#include <signal.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <linux/mman.h>
+#include <sys/syscall.h>
+#include <unistd.h>
+#include <sched.h>
+#include "vm_util.h"
+
+#include "../pidfd/pidfd.h"
+
+FIXTURE(process_madvise)
+{
+	unsigned long page_size;
+	pid_t child_pid;
+	int remote_pidfd;
+	int pidfd;
+};
+
+FIXTURE_SETUP(process_madvise)
+{
+	self->page_size = (unsigned long)sysconf(_SC_PAGESIZE);
+	self->pidfd = PIDFD_SELF;
+	self->remote_pidfd = -1;
+	self->child_pid = -1;
+};
+
+FIXTURE_TEARDOWN_PARENT(process_madvise)
+{
+	/* This teardown is guaranteed to run, even if tests SKIP or ASSERT */
+	if (self->child_pid > 0) {
+		kill(self->child_pid, SIGKILL);
+		waitpid(self->child_pid, NULL, 0);
+	}
+
+	if (self->remote_pidfd >= 0)
+		close(self->remote_pidfd);
+}
+
+static ssize_t sys_process_madvise(int pidfd, const struct iovec *iovec,
+				   size_t vlen, int advice, unsigned int flags)
+{
+	return syscall(__NR_process_madvise, pidfd, iovec, vlen, advice, flags);
+}
+
+/*
+ * This test uses PIDFD_SELF to target the current process. The main
+ * goal is to verify the basic behavior of process_madvise() with
+ * a vector of non-contiguous memory ranges, not its cross-process
+ * capabilities.
+ */
+TEST_F(process_madvise, basic)
+{
+	const unsigned long pagesize = self->page_size;
+	const int madvise_pages = 4;
+	struct iovec vec[madvise_pages];
+	int pidfd = self->pidfd;
+	ssize_t ret;
+	char *map;
+
+	/*
+	 * Create a single large mapping. We will pick pages from this
+	 * mapping to advise on. This ensures we test non-contiguous iovecs.
+	 */
+	map = mmap(NULL, pagesize * 10, PROT_READ | PROT_WRITE,
+		   MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
+	if (map == MAP_FAILED)
+		SKIP(return, "mmap failed, not enough memory.\n");
+
+	/* Fill the entire region with a known pattern. */
+	memset(map, 'A', pagesize * 10);
+
+	/*
+	 * Setup the iovec to point to 4 non-contiguous pages
+	 * within the mapping.
+	 */
+	vec[0].iov_base = &map[0 * pagesize];
+	vec[0].iov_len = pagesize;
+	vec[1].iov_base = &map[3 * pagesize];
+	vec[1].iov_len = pagesize;
+	vec[2].iov_base = &map[5 * pagesize];
+	vec[2].iov_len = pagesize;
+	vec[3].iov_base = &map[8 * pagesize];
+	vec[3].iov_len = pagesize;
+
+	ret = sys_process_madvise(pidfd, vec, madvise_pages, MADV_DONTNEED, 0);
+	if (ret == -1 && errno == EPERM)
+		SKIP(return,
+			   "process_madvise() unsupported or permission denied, try running as root.\n");
+	else if (errno == EINVAL)
+		SKIP(return,
+			   "process_madvise() unsupported or parameter invalid, please check arguments.\n");
+
+	/* The call should succeed and report the total bytes processed. */
+	ASSERT_EQ(ret, madvise_pages * pagesize);
+
+	/* Check that advised pages are now zero. */
+	for (int i = 0; i < madvise_pages; i++) {
+		char *advised_page = (char *)vec[i].iov_base;
+
+		/* Content must be 0, not 'A'. */
+		ASSERT_EQ(*advised_page, '\0');
+	}
+
+	/* Check that an un-advised page in between is still 'A'. */
+	char *unadvised_page = &map[1 * pagesize];
+
+	for (int i = 0; i < pagesize; i++)
+		ASSERT_EQ(unadvised_page[i], 'A');
+
+	/* Cleanup. */
+	ASSERT_EQ(munmap(map, pagesize * 10), 0);
+}
+
+/*
+ * This test deterministically validates process_madvise() with MADV_COLLAPSE
+ * on a remote process, other advices are difficult to verify reliably.
+ *
+ * The test verifies that a memory region in a child process,
+ * focus on process_madv remote result, only check addresses and lengths.
+ * The correctness of the MADV_COLLAPSE can be found in the relevant test examples in khugepaged.
+ */
+TEST_F(process_madvise, remote_collapse)
+{
+	const unsigned long pagesize = self->page_size;
+	long huge_page_size;
+	int pipe_info[2];
+	ssize_t ret;
+	struct iovec vec;
+
+	struct child_info {
+		pid_t pid;
+		void *map_addr;
+	} info;
+
+	huge_page_size = read_pmd_pagesize();
+	if (huge_page_size <= 0)
+		SKIP(return, "Could not determine a valid huge page size.\n");
+
+	ASSERT_EQ(pipe(pipe_info), 0);
+
+	self->child_pid = fork();
+	ASSERT_NE(self->child_pid, -1);
+
+	if (self->child_pid == 0) {
+		char *map;
+		size_t map_size = 2 * huge_page_size;
+
+		close(pipe_info[0]);
+
+		map = mmap(NULL, map_size, PROT_READ | PROT_WRITE,
+			   MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
+		ASSERT_NE(map, MAP_FAILED);
+
+		/* Fault in as small pages */
+		for (size_t i = 0; i < map_size; i += pagesize)
+			map[i] = 'A';
+
+		/* Send info and pause */
+		info.pid = getpid();
+		info.map_addr = map;
+		ret = write(pipe_info[1], &info, sizeof(info));
+		ASSERT_EQ(ret, sizeof(info));
+		close(pipe_info[1]);
+
+		pause();
+		exit(0);
+	}
+
+	close(pipe_info[1]);
+
+	/* Receive child info */
+	ret = read(pipe_info[0], &info, sizeof(info));
+	if (ret <= 0) {
+		waitpid(self->child_pid, NULL, 0);
+		SKIP(return, "Failed to read child info from pipe.\n");
+	}
+	ASSERT_EQ(ret, sizeof(info));
+	close(pipe_info[0]);
+	self->child_pid = info.pid;
+
+	self->remote_pidfd = syscall(__NR_pidfd_open, self->child_pid, 0);
+	ASSERT_GE(self->remote_pidfd, 0);
+
+	vec.iov_base = info.map_addr;
+	vec.iov_len = huge_page_size;
+
+	ret = sys_process_madvise(self->remote_pidfd, &vec, 1, MADV_COLLAPSE,
+				  0);
+	if (ret == -1) {
+		if (errno == EINVAL)
+			SKIP(return, "PROCESS_MADV_ADVISE is not supported.\n");
+		else if (errno == EPERM)
+			SKIP(return,
+				   "No process_madvise() permissions, try running as root.\n");
+		return;
+	}
+
+	ASSERT_EQ(ret, huge_page_size);
+}
+
+/*
+ * Test process_madvise() with a pidfd for a process that has already
+ * exited to ensure correct error handling.
+ */
+TEST_F(process_madvise, exited_process_pidfd)
+{
+	const unsigned long pagesize = self->page_size;
+	struct iovec vec;
+	char *map;
+	ssize_t ret;
+
+	map = mmap(NULL, pagesize, PROT_READ, MAP_PRIVATE | MAP_ANONYMOUS, -1,
+		   0);
+	if (map == MAP_FAILED)
+		SKIP(return, "mmap failed, not enough memory.\n");
+
+	vec.iov_base = map;
+	vec.iov_len = pagesize;
+
+	/*
+	 * Using a pidfd for a process that has already exited should fail
+	 * with ESRCH.
+	 */
+	self->child_pid = fork();
+	ASSERT_NE(self->child_pid, -1);
+
+	if (self->child_pid == 0)
+		exit(0);
+
+	self->remote_pidfd = syscall(__NR_pidfd_open, self->child_pid, 0);
+	ASSERT_GE(self->remote_pidfd, 0);
+
+	/* Wait for the child to ensure it has terminated. */
+	waitpid(self->child_pid, NULL, 0);
+
+	ret = sys_process_madvise(self->remote_pidfd, &vec, 1, MADV_DONTNEED,
+				  0);
+	ASSERT_EQ(ret, -1);
+	ASSERT_EQ(errno, ESRCH);
+}
+
+/*
+ * Test process_madvise() with bad pidfds to ensure correct error
+ * handling.
+ */
+TEST_F(process_madvise, bad_pidfd)
+{
+	const unsigned long pagesize = self->page_size;
+	struct iovec vec;
+	char *map;
+	ssize_t ret;
+
+	map = mmap(NULL, pagesize, PROT_READ, MAP_PRIVATE | MAP_ANONYMOUS, -1,
+		   0);
+	if (map == MAP_FAILED)
+		SKIP(return, "mmap failed, not enough memory.\n");
+
+	vec.iov_base = map;
+	vec.iov_len = pagesize;
+
+	/* Using an invalid fd number (-1) should fail with EBADF. */
+	ret = sys_process_madvise(-1, &vec, 1, MADV_DONTNEED, 0);
+	ASSERT_EQ(ret, -1);
+	ASSERT_EQ(errno, EBADF);
+
+	/*
+	 * Using a valid fd that is not a pidfd (e.g. stdin) should fail
+	 * with EBADF.
+	 */
+	ret = sys_process_madvise(STDIN_FILENO, &vec, 1, MADV_DONTNEED, 0);
+	ASSERT_EQ(ret, -1);
+	ASSERT_EQ(errno, EBADF);
+}
+
+/*
+ * Test that process_madvise() rejects vlen > UIO_MAXIOV.
+ * The kernel should return -EINVAL when the number of iovecs exceeds 1024.
+ */
+TEST_F(process_madvise, invalid_vlen)
+{
+	const unsigned long pagesize = self->page_size;
+	int pidfd = self->pidfd;
+	struct iovec vec;
+	char *map;
+	ssize_t ret;
+
+	map = mmap(NULL, pagesize, PROT_READ, MAP_PRIVATE | MAP_ANONYMOUS, -1,
+		   0);
+	if (map == MAP_FAILED)
+		SKIP(return, "mmap failed, not enough memory.\n");
+
+	vec.iov_base = map;
+	vec.iov_len = pagesize;
+
+	ret = sys_process_madvise(pidfd, &vec, 1025, MADV_DONTNEED, 0);
+	ASSERT_EQ(ret, -1);
+	ASSERT_EQ(errno, EINVAL);
+
+	/* Cleanup. */
+	ASSERT_EQ(munmap(map, pagesize), 0);
+}
+
+/*
+ * Test process_madvise() with an invalid flag value. Currently, only a flag
+ * value of 0 is supported. This test is reserved for the future, e.g., if
+ * synchronous flags are added.
+ */
+TEST_F(process_madvise, flag)
+{
+	const unsigned long pagesize = self->page_size;
+	unsigned int invalid_flag;
+	int pidfd = self->pidfd;
+	struct iovec vec;
+	char *map;
+	ssize_t ret;
+
+	map = mmap(NULL, pagesize, PROT_READ, MAP_PRIVATE | MAP_ANONYMOUS, -1,
+		   0);
+	if (map == MAP_FAILED)
+		SKIP(return, "mmap failed, not enough memory.\n");
+
+	vec.iov_base = map;
+	vec.iov_len = pagesize;
+
+	invalid_flag = 0x80000000;
+
+	ret = sys_process_madvise(pidfd, &vec, 1, MADV_DONTNEED, invalid_flag);
+	ASSERT_EQ(ret, -1);
+	ASSERT_EQ(errno, EINVAL);
+
+	/* Cleanup. */
+	ASSERT_EQ(munmap(map, pagesize), 0);
+}
+
+TEST_HARNESS_MAIN

tools/testing/selftests/mm/run_vmtests.sh

···
 	test pagemap_scan IOCTL
 - pfnmap
 	tests for VM_PFNMAP handling
+- process_madv
+	test for process_madv
 - cow
 	test copy-on-write semantics
 - thp
···
 
 # MADV_POPULATE_READ and MADV_POPULATE_WRITE tests
 CATEGORY="madv_populate" run_test ./madv_populate
+
+# PROCESS_MADV test
+CATEGORY="process_madv" run_test ./process_madv
 
 CATEGORY="vma_merge" run_test ./merge
 

tools/testing/vma/vma_internal.h

···
 #define CAP_IPC_LOCK         14
 
 #ifdef CONFIG_64BIT
-/* VM is sealed, in vm_flags */
-#define VM_SEALED	_BITUL(63)
+#define VM_SEALED_BIT	42
+#define VM_SEALED	BIT(VM_SEALED_BIT)
+#else
+#define VM_SEALED	VM_NONE
 #endif
 
 #define FIRST_USER_ADDRESS	0UL