+ mmap-change-do_mas_munmap-and-do_mas_aligned_munmap-to-use-vma-iterator.patch added to mm-unstable branch

[Andrew Morton <akpm@linux-foundation.org>]

The patch titled
     Subject: mmap: change do_mas_munmap and do_mas_aligned_munmap() to use vma iterator
has been added to the -mm mm-unstable branch.  Its filename is
     mmap-change-do_mas_munmap-and-do_mas_aligned_munmap-to-use-vma-iterator.patch

This patch will shortly appear at
     https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mmap-change-do_mas_munmap-and-do_mas_aligned_munmap-to-use-vma-iterator.patch

This patch will later appear in the mm-unstable branch at
    git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

Before you just go and hit "reply", please:
   a) Consider who else should be cc'ed
   b) Prefer to cc a suitable mailing list as well
   c) Ideally: find the original patch on the mailing list and do a
      reply-to-all to that, adding suitable additional cc's

*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***

The -mm tree is included into linux-next via the mm-everything
branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
and is updated there every 2-3 working days

------------------------------------------------------
From: "Liam R. Howlett" <Liam.Howlett@Oracle.com>
Subject: mmap: change do_mas_munmap and do_mas_aligned_munmap() to use vma iterator
Date: Thu, 5 Jan 2023 19:15:55 +0000

Start passing the vma iterator through the mm code.  This will allow for
reuse of the state and cleaner invalidation if necessary.

Link: https://lkml.kernel.org/r/20230105191517.3099082-11-Liam.Howlett@oracle.com
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

 include/linux/mm.h |    2 -
 mm/mmap.c          |   77 +++++++++++++++++++------------------------
 mm/mremap.c        |    6 +--
 3 files changed, 39 insertions(+), 46 deletions(-)

--- a/include/linux/mm.h~mmap-change-do_mas_munmap-and-do_mas_aligned_munmap-to-use-vma-iterator
+++ a/include/linux/mm.h
@@ -2881,7 +2881,7 @@ extern unsigned long mmap_region(struct
 extern unsigned long do_mmap(struct file *file, unsigned long addr,
 	unsigned long len, unsigned long prot, unsigned long flags,
 	unsigned long pgoff, unsigned long *populate, struct list_head *uf);
-extern int do_mas_munmap(struct ma_state *mas, struct mm_struct *mm,
+extern int do_vmi_munmap(struct vma_iterator *vmi, struct mm_struct *mm,
 			 unsigned long start, size_t len, struct list_head *uf,
 			 bool downgrade);
 extern int do_munmap(struct mm_struct *, unsigned long, size_t,
--- a/mm/mmap.c~mmap-change-do_mas_munmap-and-do_mas_aligned_munmap-to-use-vma-iterator
+++ a/mm/mmap.c
@@ -2364,8 +2364,8 @@ static inline int munmap_sidetree(struct
 }
 
 /*
- * do_mas_align_munmap() - munmap the aligned region from @start to @end.
- * @mas: The maple_state, ideally set up to alter the correct tree location.
+ * do_vmi_align_munmap() - munmap the aligned region from @start to @end.
+ * @vmi: The vma iterator
  * @vma: The starting vm_area_struct
  * @mm: The mm_struct
  * @start: The aligned start address to munmap.
@@ -2376,7 +2376,7 @@ static inline int munmap_sidetree(struct
  * If @downgrade is true, check return code for potential release of the lock.
  */
 static int
-do_mas_align_munmap(struct ma_state *mas, struct vm_area_struct *vma,
+do_vmi_align_munmap(struct vma_iterator *vmi, struct vm_area_struct *vma,
 		    struct mm_struct *mm, unsigned long start,
 		    unsigned long end, struct list_head *uf, bool downgrade)
 {
@@ -2388,7 +2388,6 @@ do_mas_align_munmap(struct ma_state *mas
 	mt_init_flags(&mt_detach, MT_FLAGS_LOCK_EXTERN);
 	mt_set_external_lock(&mt_detach, &mm->mmap_lock);
 
-	mas->last = end - 1;
 	/*
 	 * If we need to split any vma, do it now to save pain later.
 	 *
@@ -2408,27 +2407,23 @@ do_mas_align_munmap(struct ma_state *mas
 		if (end < vma->vm_end && mm->map_count >= sysctl_max_map_count)
 			goto map_count_exceeded;
 
-		/*
-		 * mas_pause() is not needed since mas->index needs to be set
-		 * differently than vma->vm_end anyways.
-		 */
 		error = __split_vma(mm, vma, start, 0);
 		if (error)
 			goto start_split_failed;
 
-		mas_set(mas, start);
-		vma = mas_walk(mas);
+		vma_iter_set(vmi, start);
+		vma = vma_find(vmi, end);
 	}
 
-	prev = mas_prev(mas, 0);
+	prev = vma_prev(vmi);
 	if (unlikely((!prev)))
-		mas_set(mas, start);
+		vma_iter_set(vmi, start);
 
 	/*
 	 * Detach a range of VMAs from the mm. Using next as a temp variable as
 	 * it is always overwritten.
 	 */
-	mas_for_each(mas, next, end - 1) {
+	for_each_vma_range(*vmi, next, end) {
 		/* Does it split the end? */
 		if (next->vm_end > end) {
 			struct vm_area_struct *split;
@@ -2437,8 +2432,8 @@ do_mas_align_munmap(struct ma_state *mas
 			if (error)
 				goto end_split_failed;
 
-			mas_set(mas, end);
-			split = mas_prev(mas, 0);
+			vma_iter_set(vmi, end);
+			split = vma_prev(vmi);
 			error = munmap_sidetree(split, &mas_detach);
 			if (error)
 				goto munmap_sidetree_failed;
@@ -2460,7 +2455,7 @@ do_mas_align_munmap(struct ma_state *mas
 	}
 
 	if (!next)
-		next = mas_next(mas, ULONG_MAX);
+		next = vma_next(vmi);
 
 	if (unlikely(uf)) {
 		/*
@@ -2485,10 +2480,10 @@ do_mas_align_munmap(struct ma_state *mas
 		struct vm_area_struct *vma_mas, *vma_test;
 		int test_count = 0;
 
-		mas_set_range(mas, start, end - 1);
+		vma_iter_set(vmi, start);
 		rcu_read_lock();
 		vma_test = mas_find(&test, end - 1);
-		mas_for_each(mas, vma_mas, end - 1) {
+		for_each_vma_range(*vmi, vma_mas, end) {
 			BUG_ON(vma_mas != vma_test);
 			test_count++;
 			vma_test = mas_next(&test, end - 1);
@@ -2498,8 +2493,8 @@ do_mas_align_munmap(struct ma_state *mas
 	}
 #endif
 	/* Point of no return */
-	mas_set_range(mas, start, end - 1);
-	if (mas_store_gfp(mas, NULL, GFP_KERNEL))
+	vma_iter_set(vmi, start);
+	if (vma_iter_clear_gfp(vmi, start, end, GFP_KERNEL))
 		return -ENOMEM;
 
 	mm->map_count -= count;
@@ -2537,8 +2532,8 @@ map_count_exceeded:
 }
 
 /*
- * do_mas_munmap() - munmap a given range.
- * @mas: The maple state
+ * do_vmi_munmap() - munmap a given range.
+ * @vmi: The vma iterator
  * @mm: The mm_struct
  * @start: The start address to munmap
  * @len: The length of the range to munmap
@@ -2552,7 +2547,7 @@ map_count_exceeded:
  *
  * Returns: -EINVAL on failure, 1 on success and unlock, 0 otherwise.
  */
-int do_mas_munmap(struct ma_state *mas, struct mm_struct *mm,
+int do_vmi_munmap(struct vma_iterator *vmi, struct mm_struct *mm,
 		  unsigned long start, size_t len, struct list_head *uf,
 		  bool downgrade)
 {
@@ -2570,11 +2565,11 @@ int do_mas_munmap(struct ma_state *mas,
 	arch_unmap(mm, start, end);
 
 	/* Find the first overlapping VMA */
-	vma = mas_find(mas, end - 1);
+	vma = vma_find(vmi, end);
 	if (!vma)
 		return 0;
 
-	return do_mas_align_munmap(mas, vma, mm, start, end, uf, downgrade);
+	return do_vmi_align_munmap(vmi, vma, mm, start, end, uf, downgrade);
 }
 
 /* do_munmap() - Wrapper function for non-maple tree aware do_munmap() calls.
@@ -2586,9 +2581,9 @@ int do_mas_munmap(struct ma_state *mas,
 int do_munmap(struct mm_struct *mm, unsigned long start, size_t len,
 	      struct list_head *uf)
 {
-	MA_STATE(mas, &mm->mm_mt, start, start);
+	VMA_ITERATOR(vmi, mm, start);
 
-	return do_mas_munmap(&mas, mm, start, len, uf, false);
+	return do_vmi_munmap(&vmi, mm, start, len, uf, false);
 }
 
 unsigned long mmap_region(struct file *file, unsigned long addr,
@@ -2604,7 +2599,7 @@ unsigned long mmap_region(struct file *f
 	unsigned long merge_start = addr, merge_end = end;
 	pgoff_t vm_pgoff;
 	int error;
-	MA_STATE(mas, &mm->mm_mt, addr, end - 1);
+	VMA_ITERATOR(vmi, mm, addr);
 
 	/* Check against address space limit. */
 	if (!may_expand_vm(mm, vm_flags, len >> PAGE_SHIFT)) {
@@ -2622,7 +2617,7 @@ unsigned long mmap_region(struct file *f
 	}
 
 	/* Unmap any existing mapping in the area */
-	if (do_mas_munmap(&mas, mm, addr, len, uf, false))
+	if (do_vmi_munmap(&vmi, mm, addr, len, uf, false))
 		return -ENOMEM;
 
 	/*
@@ -2635,8 +2630,8 @@ unsigned long mmap_region(struct file *f
 		vm_flags |= VM_ACCOUNT;
 	}
 
-	next = mas_next(&mas, ULONG_MAX);
-	prev = mas_prev(&mas, 0);
+	next = vma_next(&vmi);
+	prev = vma_prev(&vmi);
 	if (vm_flags & VM_SPECIAL)
 		goto cannot_expand;
 
@@ -2664,13 +2659,11 @@ unsigned long mmap_region(struct file *f
 
 	/* Actually expand, if possible */
 	if (vma &&
-	    !vma_expand(&mas, vma, merge_start, merge_end, vm_pgoff, next)) {
+	    !vma_expand(&vmi.mas, vma, merge_start, merge_end, vm_pgoff, next)) {
 		khugepaged_enter_vma(vma, vm_flags);
 		goto expanded;
 	}
 
-	mas.index = addr;
-	mas.last = end - 1;
 cannot_expand:
 	/*
 	 * Determine the object being mapped and call the appropriate
@@ -2709,7 +2702,7 @@ cannot_expand:
 			error = -EINVAL;
 			goto close_and_free_vma;
 		}
-		mas_reset(&mas);
+		vma_iter_set(&vmi, addr);
 
 		/*
 		 * If vm_flags changed after call_mmap(), we should try merge
@@ -2755,7 +2748,7 @@ cannot_expand:
 			goto free_vma;
 	}
 
-	if (mas_preallocate(&mas, vma, GFP_KERNEL)) {
+	if (vma_iter_prealloc(&vmi, vma)) {
 		error = -ENOMEM;
 		if (file)
 			goto close_and_free_vma;
@@ -2768,7 +2761,7 @@ cannot_expand:
 	if (vma->vm_file)
 		i_mmap_lock_write(vma->vm_file->f_mapping);
 
-	vma_mas_store(vma, &mas);
+	vma_iter_store(&vmi, vma);
 	mm->map_count++;
 	if (vma->vm_file) {
 		if (vma->vm_flags & VM_SHARED)
@@ -2829,7 +2822,7 @@ unmap_and_free_vma:
 	vma->vm_file = NULL;
 
 	/* Undo any partial mapping done by a device driver. */
-	unmap_region(mm, mas.tree, vma, prev, next, vma->vm_start, vma->vm_end);
+	unmap_region(mm, &mm->mm_mt, vma, prev, next, vma->vm_start, vma->vm_end);
 	if (file && (vm_flags & VM_SHARED))
 		mapping_unmap_writable(file->f_mapping);
 free_vma:
@@ -2846,12 +2839,12 @@ static int __vm_munmap(unsigned long sta
 	int ret;
 	struct mm_struct *mm = current->mm;
 	LIST_HEAD(uf);
-	MA_STATE(mas, &mm->mm_mt, start, start);
+	VMA_ITERATOR(vmi, mm, start);
 
 	if (mmap_write_lock_killable(mm))
 		return -EINTR;
 
-	ret = do_mas_munmap(&mas, mm, start, len, &uf, downgrade);
+	ret = do_vmi_munmap(&vmi, mm, start, len, &uf, downgrade);
 	/*
 	 * Returning 1 indicates mmap_lock is downgraded.
 	 * But 1 is not legal return value of vm_munmap() and munmap(), reset
@@ -2983,7 +2976,7 @@ static int do_brk_munmap(struct vma_iter
 	int ret;
 
 	arch_unmap(mm, newbrk, oldbrk);
-	ret = do_mas_align_munmap(&vmi->mas, vma, mm, newbrk, oldbrk, uf, true);
+	ret = do_vmi_align_munmap(vmi, vma, mm, newbrk, oldbrk, uf, true);
 	validate_mm_mt(mm);
 	return ret;
 }
@@ -3107,7 +3100,7 @@ int vm_brk_flags(unsigned long addr, uns
 	if (ret)
 		goto limits_failed;
 
-	ret = do_mas_munmap(&vmi.mas, mm, addr, len, &uf, 0);
+	ret = do_vmi_munmap(&vmi, mm, addr, len, &uf, 0);
 	if (ret)
 		goto munmap_failed;
 
--- a/mm/mremap.c~mmap-change-do_mas_munmap-and-do_mas_aligned_munmap-to-use-vma-iterator
+++ a/mm/mremap.c
@@ -978,14 +978,14 @@ SYSCALL_DEFINE5(mremap, unsigned long, a
 	/*
 	 * Always allow a shrinking remap: that just unmaps
 	 * the unnecessary pages..
-	 * do_mas_munmap does all the needed commit accounting, and
+	 * do_vmi_munmap does all the needed commit accounting, and
 	 * downgrades mmap_lock to read if so directed.
 	 */
 	if (old_len >= new_len) {
 		int retval;
-		MA_STATE(mas, &mm->mm_mt, addr + new_len, addr + new_len);
+		VMA_ITERATOR(vmi, mm, addr + new_len);
 
-		retval = do_mas_munmap(&mas, mm, addr + new_len,
+		retval = do_vmi_munmap(&vmi, mm, addr + new_len,
 				       old_len - new_len, &uf_unmap, true);
 		/* Returning 1 indicates mmap_lock is downgraded to read. */
 		if (retval == 1) {
_

Patches currently in -mm which might be from Liam.Howlett@Oracle.com are

maple_tree-add-mas_init-function.patch
maple_tree-fix-potential-rcu-issue.patch
maple_tree-reduce-user-error-potential.patch
test_maple_tree-test-modifications-while-iterating.patch
mm-expand-vma-iterator-interface.patch
mm-mmap-convert-brk-to-use-vma-iterator.patch
kernel-fork-convert-forking-to-using-the-vmi-iterator.patch
mmap-convert-vma_link-vma-iterator.patch
mm-mmap-remove-preallocation-from-do_mas_align_munmap.patch
mmap-change-do_mas_munmap-and-do_mas_aligned_munmap-to-use-vma-iterator.patch
mmap-convert-vma_expand-to-use-vma-iterator.patch
mm-add-temporary-vma-iterator-versions-of-vma_merge-split_vma-and-__split_vma.patch
ipc-shm-use-the-vma-iterator-for-munmap-calls.patch
userfaultfd-use-vma-iterator.patch
mm-change-mprotect_fixup-to-vma-iterator.patch
mlock-convert-mlock-to-vma-iterator.patch
coredump-convert-to-vma-iterator.patch
mempolicy-convert-to-vma-iterator.patch
task_mmu-convert-to-vma-iterator.patch
sched-convert-to-vma-iterator.patch
madvise-use-vmi-iterator-for-__split_vma-and-vma_merge.patch
mmap-pass-through-vmi-iterator-to-__split_vma.patch
mmap-use-vmi-version-of-vma_merge.patch
mm-mremap-use-vmi-version-of-vma_merge.patch
mm-switch-vma_merge-split_vma-and-__split_vma-to-vma-iterator.patch
mmap-convert-__vma_adjust-to-use-vma-iterator.patch
mm-pass-through-vma-iterator-to-__vma_adjust.patch
madvise-use-split_vma-instead-of-__split_vma.patch
mm-remove-unnecessary-write-to-vma-iterator-in-__vma_adjust.patch
mm-pass-vma-iterator-through-to-__vma_adjust.patch
mm-add-vma-iterator-to-vma_adjust-arguments.patch
mmap-clean-up-mmap_region-unrolling.patch
mm-change-munmap-splitting-order-and-move_vma.patch
mm-mmap-move-anon_vma-setting-in-__vma_adjust.patch
mm-mmap-refactor-locking-out-of-__vma_adjust.patch
mm-mmap-use-vma_prepare-and-vma_complete-in-vma_expand.patch
mm-mmap-introduce-init_vma_prep-and-init_multi_vma_prep.patch
mm-dont-use-__vma_adjust-in-__split_vma.patch
mm-mmap-dont-use-__vma_adjust-in-shift_arg_pages.patch
mm-mmap-introduce-dup_vma_anon-helper.patch
mm-mmap-convert-do_brk_flags-to-use-vma_prepare-and-vma_complete.patch
mm-mmap-remove-__vma_adjust.patch
vma_merge-set-vma-iterator-to-correct-position.patch