Merge 0c563f1480 ("proc: remove VMA rbtree use from nommu") into android-mainline

Steps on the way to 6.1-rc1

Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>
Change-Id: I5edff6b45f59f4eed3422bb2e2079c54af885e3f

arch/x86/kernel/tboot.c

@@ -96,6 +96,7 @@ void __init tboot_probe(void)
 static pgd_t *tboot_pg_dir;
 static struct mm_struct tboot_mm = {
 	.mm_rb          = RB_ROOT,
+	.mm_mt          = MTREE_INIT_EXT(mm_mt, MM_MT_FLAGS, tboot_mm.mmap_lock),
 	.pgd            = swapper_pg_dir,
 	.mm_users       = ATOMIC_INIT(2),
 	.mm_count       = ATOMIC_INIT(1),

drivers/firmware/efi/efi.c

@@ -60,6 +60,7 @@ static unsigned long __initdata initrd = EFI_INVALID_TABLE_ADDR;
 
 struct mm_struct efi_mm = {
 	.mm_rb			= RB_ROOT,
+	.mm_mt			= MTREE_INIT_EXT(mm_mt, MM_MT_FLAGS, efi_mm.mmap_lock),
 	.mm_users		= ATOMIC_INIT(2),
 	.mm_count		= ATOMIC_INIT(1),
 	.write_protect_seq      = SEQCNT_ZERO(efi_mm.write_protect_seq),

fs/proc/task_nommu.c

@@ -20,15 +20,13 @@
  */
 void task_mem(struct seq_file *m, struct mm_struct *mm)
 {
+	VMA_ITERATOR(vmi, mm, 0);
 	struct vm_area_struct *vma;
 	struct vm_region *region;
-	struct rb_node *p;
 	unsigned long bytes = 0, sbytes = 0, slack = 0, size;
-        
-	mmap_read_lock(mm);
-	for (p = rb_first(&mm->mm_rb); p; p = rb_next(p)) {
-		vma = rb_entry(p, struct vm_area_struct, vm_rb);
 
+	mmap_read_lock(mm);
+	for_each_vma(vmi, vma) {
 		bytes += kobjsize(vma);
 
 		region = vma->vm_region;
@@ -82,15 +80,13 @@ void task_mem(struct seq_file *m, struct mm_struct *mm)
 
 unsigned long task_vsize(struct mm_struct *mm)
 {
+	VMA_ITERATOR(vmi, mm, 0);
 	struct vm_area_struct *vma;
-	struct rb_node *p;
 	unsigned long vsize = 0;
 
 	mmap_read_lock(mm);
-	for (p = rb_first(&mm->mm_rb); p; p = rb_next(p)) {
-		vma = rb_entry(p, struct vm_area_struct, vm_rb);
+	for_each_vma(vmi, vma)
 		vsize += vma->vm_end - vma->vm_start;
-	}
 	mmap_read_unlock(mm);
 	return vsize;
 }
@@ -99,14 +95,13 @@ unsigned long task_statm(struct mm_struct *mm,
 			 unsigned long *shared, unsigned long *text,
 			 unsigned long *data, unsigned long *resident)
 {
+	VMA_ITERATOR(vmi, mm, 0);
 	struct vm_area_struct *vma;
 	struct vm_region *region;
-	struct rb_node *p;
 	unsigned long size = kobjsize(mm);
 
 	mmap_read_lock(mm);
-	for (p = rb_first(&mm->mm_rb); p; p = rb_next(p)) {
-		vma = rb_entry(p, struct vm_area_struct, vm_rb);
+	for_each_vma(vmi, vma) {
 		size += kobjsize(vma);
 		region = vma->vm_region;
 		if (region) {
@@ -190,17 +185,19 @@ static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma)
  */
 static int show_map(struct seq_file *m, void *_p)
 {
-	struct rb_node *p = _p;
-
-	return nommu_vma_show(m, rb_entry(p, struct vm_area_struct, vm_rb));
+	return nommu_vma_show(m, _p);
 }
 
 static void *m_start(struct seq_file *m, loff_t *pos)
 {
 	struct proc_maps_private *priv = m->private;
 	struct mm_struct *mm;
-	struct rb_node *p;
-	loff_t n = *pos;
+	struct vm_area_struct *vma;
+	unsigned long addr = *pos;
+
+	/* See m_next(). Zero at the start or after lseek. */
+	if (addr == -1UL)
+		return NULL;
 
 	/* pin the task and mm whilst we play with them */
 	priv->task = get_proc_task(priv->inode);
@@ -216,10 +213,10 @@ static void *m_start(struct seq_file *m, loff_t *pos)
 		return ERR_PTR(-EINTR);
 	}
 
-	/* start from the Nth VMA */
-	for (p = rb_first(&mm->mm_rb); p; p = rb_next(p))
-		if (n-- == 0)
-			return p;
+	/* start the next element from addr */
+	vma = find_vma(mm, addr);
+	if (vma)
+		return vma;
 
 	mmap_read_unlock(mm);
 	mmput(mm);
@@ -242,10 +239,10 @@ static void m_stop(struct seq_file *m, void *_vml)
 
 static void *m_next(struct seq_file *m, void *_p, loff_t *pos)
 {
-	struct rb_node *p = _p;
+	struct vm_area_struct *vma = _p;
 
-	(*pos)++;
-	return p ? rb_next(p) : NULL;
+	*pos = vma->vm_end;
+	return find_vma(vma->vm_mm, vma->vm_end);
 }
 
 static const struct seq_operations proc_pid_maps_ops = {

include/linux/mm.h

@@ -661,6 +661,38 @@ static inline bool vma_is_accessible(struct vm_area_struct *vma)
 	return vma->vm_flags & VM_ACCESS_FLAGS;
 }
 
+static inline
+struct vm_area_struct *vma_find(struct vma_iterator *vmi, unsigned long max)
+{
+	return mas_find(&vmi->mas, max);
+}
+
+static inline struct vm_area_struct *vma_next(struct vma_iterator *vmi)
+{
+	/*
+	 * Uses vma_find() to get the first VMA when the iterator starts.
+	 * Calling mas_next() could skip the first entry.
+	 */
+	return vma_find(vmi, ULONG_MAX);
+}
+
+static inline struct vm_area_struct *vma_prev(struct vma_iterator *vmi)
+{
+	return mas_prev(&vmi->mas, 0);
+}
+
+static inline unsigned long vma_iter_addr(struct vma_iterator *vmi)
+{
+	return vmi->mas.index;
+}
+
+#define for_each_vma(__vmi, __vma)					\
+	while (((__vma) = vma_next(&(__vmi))) != NULL)
+
+/* The MM code likes to work with exclusive end addresses */
+#define for_each_vma_range(__vmi, __vma, __end)				\
+	while (((__vma) = vma_find(&(__vmi), (__end) - 1)) != NULL)
+
 #ifdef CONFIG_SHMEM
 /*
  * The vma_is_shmem is not inline because it is used only by slow
@@ -2630,6 +2662,9 @@ extern struct vm_area_struct *copy_vma(struct vm_area_struct **,
 	bool *need_rmap_locks);
 extern void exit_mmap(struct mm_struct *);
 
+void vma_mas_store(struct vm_area_struct *vma, struct ma_state *mas);
+void vma_mas_remove(struct vm_area_struct *vma, struct ma_state *mas);
+
 static inline int check_data_rlimit(unsigned long rlim,
 				    unsigned long new,
 				    unsigned long start,

include/linux/mm_types.h

@@ -9,6 +9,7 @@
 #include <linux/list.h>
 #include <linux/spinlock.h>
 #include <linux/rbtree.h>
+#include <linux/maple_tree.h>
 #include <linux/rwsem.h>
 #include <linux/completion.h>
 #include <linux/cpumask.h>
@@ -486,6 +487,7 @@ struct kioctx_table;
 struct mm_struct {
 	struct {
 		struct vm_area_struct *mmap;		/* list of VMAs */
+		struct maple_tree mm_mt;
 		struct rb_root mm_rb;
 		u64 vmacache_seqnum;                   /* per-thread vmacache */
 #ifdef CONFIG_MMU
@@ -697,6 +699,7 @@ struct mm_struct {
 	unsigned long cpu_bitmap[];
 };
 
+#define MM_MT_FLAGS	(MT_FLAGS_ALLOC_RANGE | MT_FLAGS_LOCK_EXTERN)
 extern struct mm_struct init_mm;
 
 /* Pointer magic because the dynamic array size confuses some compilers. */
@@ -774,6 +777,27 @@ static inline void lru_gen_use_mm(struct mm_struct *mm)
 
 #endif /* CONFIG_LRU_GEN */
 
+struct vma_iterator {
+	struct ma_state mas;
+};
+
+#define VMA_ITERATOR(name, __mm, __addr)				\
+	struct vma_iterator name = {					\
+		.mas = {						\
+			.tree = &(__mm)->mm_mt,				\
+			.index = __addr,				\
+			.node = MAS_START,				\
+		},							\
+	}
+
+static inline void vma_iter_init(struct vma_iterator *vmi,
+		struct mm_struct *mm, unsigned long addr)
+{
+	vmi->mas.tree = &mm->mm_mt;
+	vmi->mas.index = addr;
+	vmi->mas.node = MAS_START;
+}
+
 struct mmu_gather;
 extern void tlb_gather_mmu(struct mmu_gather *tlb, struct mm_struct *mm);
 extern void tlb_gather_mmu_fullmm(struct mmu_gather *tlb, struct mm_struct *mm);

include/trace/events/mmap.h

@@ -42,6 +42,79 @@ TRACE_EVENT(vm_unmapped_area,
 		__entry->low_limit, __entry->high_limit, __entry->align_mask,
 		__entry->align_offset)
 );
+
+TRACE_EVENT(vma_mas_szero,
+	TP_PROTO(struct maple_tree *mt, unsigned long start,
+		 unsigned long end),
+
+	TP_ARGS(mt, start, end),
+
+	TP_STRUCT__entry(
+			__field(struct maple_tree *, mt)
+			__field(unsigned long, start)
+			__field(unsigned long, end)
+	),
+
+	TP_fast_assign(
+			__entry->mt		= mt;
+			__entry->start		= start;
+			__entry->end		= end;
+	),
+
+	TP_printk("mt_mod %p, (NULL), SNULL, %lu, %lu,",
+		  __entry->mt,
+		  (unsigned long) __entry->start,
+		  (unsigned long) __entry->end
+	)
+);
+
+TRACE_EVENT(vma_store,
+	TP_PROTO(struct maple_tree *mt, struct vm_area_struct *vma),
+
+	TP_ARGS(mt, vma),
+
+	TP_STRUCT__entry(
+			__field(struct maple_tree *, mt)
+			__field(struct vm_area_struct *, vma)
+			__field(unsigned long, vm_start)
+			__field(unsigned long, vm_end)
+	),
+
+	TP_fast_assign(
+			__entry->mt		= mt;
+			__entry->vma		= vma;
+			__entry->vm_start	= vma->vm_start;
+			__entry->vm_end		= vma->vm_end - 1;
+	),
+
+	TP_printk("mt_mod %p, (%p), STORE, %lu, %lu,",
+		  __entry->mt, __entry->vma,
+		  (unsigned long) __entry->vm_start,
+		  (unsigned long) __entry->vm_end
+	)
+);
+
+
+TRACE_EVENT(exit_mmap,
+	TP_PROTO(struct mm_struct *mm),
+
+	TP_ARGS(mm),
+
+	TP_STRUCT__entry(
+			__field(struct mm_struct *, mm)
+			__field(struct maple_tree *, mt)
+	),
+
+	TP_fast_assign(
+		       __entry->mm		= mm;
+		       __entry->mt		= &mm->mm_mt;
+	),
+
+	TP_printk("mt_mod %p, DESTROY\n",
+		  __entry->mt
+	)
+);
+
 #endif
 
 /* This part must be outside protection */

kernel/fork.c

@@ -591,8 +591,10 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
 	struct vm_area_struct *mpnt, *tmp, *prev, **pprev;
 	struct rb_node **rb_link, *rb_parent;
 	int retval;
-	unsigned long charge;
+	unsigned long charge = 0;
 	LIST_HEAD(uf);
+	MA_STATE(old_mas, &oldmm->mm_mt, 0, 0);
+	MA_STATE(mas, &mm->mm_mt, 0, 0);
 
 	uprobe_start_dup_mmap();
 	if (mmap_write_lock_killable(oldmm)) {
@@ -622,8 +624,17 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
 		goto out;
 	khugepaged_fork(mm, oldmm);
 
+	retval = mas_expected_entries(&mas, oldmm->map_count);
+	if (retval)
+		goto out;
+
 	prev = NULL;
-	for (mpnt = oldmm->mmap; mpnt; mpnt = mpnt->vm_next) {
+
+	retval = mas_expected_entries(&mas, oldmm->map_count);
+	if (retval)
+		goto out;
+
+	mas_for_each(&old_mas, mpnt, ULONG_MAX) {
 		struct file *file;
 
 		if (mpnt->vm_flags & VM_DONTCOPY) {
@@ -637,7 +648,7 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
 		 */
 		if (fatal_signal_pending(current)) {
 			retval = -EINTR;
-			goto out;
+			goto loop_out;
 		}
 		if (mpnt->vm_flags & VM_ACCOUNT) {
 			unsigned long len = vma_pages(mpnt);
@@ -702,6 +713,13 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
 		rb_link = &tmp->vm_rb.rb_right;
 		rb_parent = &tmp->vm_rb;
 
+		/* Link the vma into the MT */
+		mas.index = tmp->vm_start;
+		mas.last = tmp->vm_end - 1;
+		mas_store(&mas, tmp);
+		if (mas_is_err(&mas))
+			goto fail_nomem_mas_store;
+
 		mm->map_count++;
 		if (!(tmp->vm_flags & VM_WIPEONFORK))
 			retval = copy_page_range(tmp, mpnt);
@@ -710,10 +728,12 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
 			tmp->vm_ops->open(tmp);
 
 		if (retval)
-			goto out;
+			goto loop_out;
 	}
 	/* a new mm has just been created */
 	retval = arch_dup_mmap(oldmm, mm);
+loop_out:
+	mas_destroy(&mas);
 out:
 	mmap_write_unlock(mm);
 	flush_tlb_mm(oldmm);
@@ -722,6 +742,9 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
 fail_uprobe_end:
 	uprobe_end_dup_mmap();
 	return retval;
+
+fail_nomem_mas_store:
+	unlink_anon_vmas(tmp);
 fail_nomem_anon_vma_fork:
 	mpol_put(vma_policy(tmp));
 fail_nomem_policy:
@@ -729,7 +752,7 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
 fail_nomem:
 	retval = -ENOMEM;
 	vm_unacct_memory(charge);
-	goto out;
+	goto loop_out;
 }
 
 static inline int mm_alloc_pgd(struct mm_struct *mm)
@@ -1122,6 +1145,8 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p,
 {
 	mm->mmap = NULL;
 	mm->mm_rb = RB_ROOT;
+	mt_init_flags(&mm->mm_mt, MM_MT_FLAGS);
+	mt_set_external_lock(&mm->mm_mt, &mm->mmap_lock);
 	mm->vmacache_seqnum = 0;
 	atomic_set(&mm->mm_users, 1);
 	atomic_set(&mm->mm_count, 1);

mm/damon/vaddr-test.h

@@ -14,33 +14,19 @@
 
 #include <kunit/test.h>
 
-static void __link_vmas(struct vm_area_struct *vmas, ssize_t nr_vmas)
+static void __link_vmas(struct maple_tree *mt, struct vm_area_struct *vmas,
+			ssize_t nr_vmas)
 {
-	int i, j;
-	unsigned long largest_gap, gap;
+	int i;
+	MA_STATE(mas, mt, 0, 0);
 
 	if (!nr_vmas)
 		return;
 
-	for (i = 0; i < nr_vmas - 1; i++) {
-		vmas[i].vm_next = &vmas[i + 1];
-
-		vmas[i].vm_rb.rb_left = NULL;
-		vmas[i].vm_rb.rb_right = &vmas[i + 1].vm_rb;
-
-		largest_gap = 0;
-		for (j = i; j < nr_vmas; j++) {
-			if (j == 0)
-				continue;
-			gap = vmas[j].vm_start - vmas[j - 1].vm_end;
-			if (gap > largest_gap)
-				largest_gap = gap;
-		}
-		vmas[i].rb_subtree_gap = largest_gap;
-	}
-	vmas[i].vm_next = NULL;
-	vmas[i].vm_rb.rb_right = NULL;
-	vmas[i].rb_subtree_gap = 0;
+	mas_lock(&mas);
+	for (i = 0; i < nr_vmas; i++)
+		vma_mas_store(&vmas[i], &mas);
+	mas_unlock(&mas);
 }
 
 /*
@@ -72,6 +58,7 @@ static void __link_vmas(struct vm_area_struct *vmas, ssize_t nr_vmas)
  */
 static void damon_test_three_regions_in_vmas(struct kunit *test)
 {
+	static struct mm_struct mm;
 	struct damon_addr_range regions[3] = {0,};
 	/* 10-20-25, 200-210-220, 300-305, 307-330 */
 	struct vm_area_struct vmas[] = {
@@ -83,9 +70,10 @@ static void damon_test_three_regions_in_vmas(struct kunit *test)
 		(struct vm_area_struct) {.vm_start = 307, .vm_end = 330},
 	};
 
-	__link_vmas(vmas, 6);
+	mt_init_flags(&mm.mm_mt, MM_MT_FLAGS);
+	__link_vmas(&mm.mm_mt, vmas, ARRAY_SIZE(vmas));
 
-	__damon_va_three_regions(&vmas[0], regions);
+	__damon_va_three_regions(&mm, regions);
 
 	KUNIT_EXPECT_EQ(test, 10ul, regions[0].start);
 	KUNIT_EXPECT_EQ(test, 25ul, regions[0].end);

mm/damon/vaddr.c

@@ -113,37 +113,38 @@ static unsigned long sz_range(struct damon_addr_range *r)
  *
  * Returns 0 if success, or negative error code otherwise.
  */
-static int __damon_va_three_regions(struct vm_area_struct *vma,
+static int __damon_va_three_regions(struct mm_struct *mm,
 				       struct damon_addr_range regions[3])
 {
-	struct damon_addr_range gap = {0}, first_gap = {0}, second_gap = {0};
-	struct vm_area_struct *last_vma = NULL;
-	unsigned long start = 0;
-	struct rb_root rbroot;
+	struct damon_addr_range first_gap = {0}, second_gap = {0};
+	VMA_ITERATOR(vmi, mm, 0);
+	struct vm_area_struct *vma, *prev = NULL;
+	unsigned long start;
 
-	/* Find two biggest gaps so that first_gap > second_gap > others */
-	for (; vma; vma = vma->vm_next) {
-		if (!last_vma) {
+	/*
+	 * Find the two biggest gaps so that first_gap > second_gap > others.
+	 * If this is too slow, it can be optimised to examine the maple
+	 * tree gaps.
+	 */
+	for_each_vma(vmi, vma) {
+		unsigned long gap;
+
+		if (!prev) {
 			start = vma->vm_start;
 			goto next;
 		}
+		gap = vma->vm_start - prev->vm_end;
 
-		if (vma->rb_subtree_gap <= sz_range(&second_gap)) {
-			rbroot.rb_node = &vma->vm_rb;
-			vma = rb_entry(rb_last(&rbroot),
-					struct vm_area_struct, vm_rb);
-			goto next;
-		}
-
-		gap.start = last_vma->vm_end;
-		gap.end = vma->vm_start;
-		if (sz_range(&gap) > sz_range(&second_gap)) {
-			swap(gap, second_gap);
-			if (sz_range(&second_gap) > sz_range(&first_gap))
-				swap(second_gap, first_gap);
+		if (gap > sz_range(&first_gap)) {
+			second_gap = first_gap;
+			first_gap.start = prev->vm_end;
+			first_gap.end = vma->vm_start;
+		} else if (gap > sz_range(&second_gap)) {
+			second_gap.start = prev->vm_end;
+			second_gap.end = vma->vm_start;
 		}
 next:
-		last_vma = vma;
+		prev = vma;
 	}
 
 	if (!sz_range(&second_gap) || !sz_range(&first_gap))
@@ -159,7 +160,7 @@ static int __damon_va_three_regions(struct vm_area_struct *vma,
 	regions[1].start = ALIGN(first_gap.end, DAMON_MIN_REGION);
 	regions[1].end = ALIGN(second_gap.start, DAMON_MIN_REGION);
 	regions[2].start = ALIGN(second_gap.end, DAMON_MIN_REGION);
-	regions[2].end = ALIGN(last_vma->vm_end, DAMON_MIN_REGION);
+	regions[2].end = ALIGN(prev->vm_end, DAMON_MIN_REGION);
 
 	return 0;
 }
@@ -180,7 +181,7 @@ static int damon_va_three_regions(struct damon_target *t,
 		return -EINVAL;
 
 	mmap_read_lock(mm);
-	rc = __damon_va_three_regions(mm->mmap, regions);
+	rc = __damon_va_three_regions(mm, regions);
 	mmap_read_unlock(mm);
 
 	mmput(mm);

mm/init-mm.c

@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 #include <linux/mm_types.h>
 #include <linux/rbtree.h>
+#include <linux/maple_tree.h>
 #include <linux/rwsem.h>
 #include <linux/spinlock.h>
 #include <linux/list.h>
@@ -29,6 +30,7 @@
  */
 struct mm_struct init_mm = {
 	.mm_rb		= RB_ROOT,
+	.mm_mt		= MTREE_INIT_EXT(mm_mt, MM_MT_FLAGS, init_mm.mmap_lock),
 	.pgd		= swapper_pg_dir,
 	.mm_users	= ATOMIC_INIT(2),
 	.mm_count	= ATOMIC_INIT(1),

mm/nommu.c

@@ -545,6 +545,19 @@ static void put_nommu_region(struct vm_region *region)
 	__put_nommu_region(region);
 }
 
+void vma_mas_store(struct vm_area_struct *vma, struct ma_state *mas)
+{
+	mas_set_range(mas, vma->vm_start, vma->vm_end - 1);
+	mas_store_prealloc(mas, vma);
+}
+
+void vma_mas_remove(struct vm_area_struct *vma, struct ma_state *mas)
+{
+	mas->index = vma->vm_start;
+	mas->last = vma->vm_end - 1;
+	mas_store_prealloc(mas, NULL);
+}
+
 /*
  * add a VMA into a process's mm_struct in the appropriate place in the list
  * and tree and add to the address space's page tree also if not an anonymous