mm/rmap: use rmap_walk() in try_to_unmap()

Now, we have an infrastructure in rmap_walk() to handle difference from
variants of rmap traversing functions.

So, just use it in try_to_unmap().

In this patch, I change following things.

1. enable rmap_walk() if !CONFIG_MIGRATION.
2. mechanical change to use rmap_walk() in try_to_unmap().

Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Reviewed-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Hugh Dickins <hughd@google.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Hillf Danton <dhillf@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

2 files changed, 38 insertions, 14 deletions

diff --git a/mm/ksm.c b/mm/ksm.c
index 91b8cb35f7cc..6b4baa97f4c0 100644
--- a/mm/ksm.c
+++ b/mm/ksm.c
@@ -1982,7 +1982,7 @@ again:
 				continue;
 
 			ret = try_to_unmap_one(page, vma,
-					rmap_item->address, flags);
+					rmap_item->address, (void *)flags);
 			if (ret != SWAP_AGAIN || !page_mapped(page)) {
 				anon_vma_unlock_read(anon_vma);
 				goto out;
@@ -1996,7 +1996,6 @@ out:
 	return ret;
 }
 
-#ifdef CONFIG_MIGRATION
 int rmap_walk_ksm(struct page *page, struct rmap_walk_control *rwc)
 {
 	struct stable_node *stable_node;
@@ -2054,6 +2053,7 @@ out:
 	return ret;
 }
 
+#ifdef CONFIG_MIGRATION
 void ksm_migrate_page(struct page *newpage, struct page *oldpage)
 {
 	struct stable_node *stable_node;
diff --git a/mm/rmap.c b/mm/rmap.c
index 97bf8f0396f8..b3263cb32361 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1179,15 +1179,18 @@ out:
 /*
  * Subfunctions of try_to_unmap: try_to_unmap_one called
  * repeatedly from try_to_unmap_ksm, try_to_unmap_anon or try_to_unmap_file.
+ *
+ * @arg: enum ttu_flags will be passed to this argument
  */
 int try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
-		     unsigned long address, enum ttu_flags flags)
+		     unsigned long address, void *arg)
 {
 	struct mm_struct *mm = vma->vm_mm;
 	pte_t *pte;
 	pte_t pteval;
 	spinlock_t *ptl;
 	int ret = SWAP_AGAIN;
+	enum ttu_flags flags = (enum ttu_flags)arg;
 
 	pte = page_check_address(page, mm, address, &ptl, 0);
 	if (!pte)
@@ -1513,6 +1516,11 @@ bool is_vma_temporary_stack(struct vm_area_struct *vma)
 	return false;
 }
 
+static bool invalid_migration_vma(struct vm_area_struct *vma, void *arg)
+{
+	return is_vma_temporary_stack(vma);
+}
+
 /**
  * try_to_unmap_anon - unmap or unlock anonymous page using the object-based
  * rmap method
@@ -1558,7 +1566,7 @@ static int try_to_unmap_anon(struct page *page, enum ttu_flags flags)
 			continue;
 
 		address = vma_address(page, vma);
-		ret = try_to_unmap_one(page, vma, address, flags);
+		ret = try_to_unmap_one(page, vma, address, (void *)flags);
 		if (ret != SWAP_AGAIN || !page_mapped(page))
 			break;
 	}
@@ -1592,7 +1600,7 @@ static int try_to_unmap_file(struct page *page, enum ttu_flags flags)
 	mutex_lock(&mapping->i_mmap_mutex);
 	vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff, pgoff) {
 		unsigned long address = vma_address(page, vma);
-		ret = try_to_unmap_one(page, vma, address, flags);
+		ret = try_to_unmap_one(page, vma, address, (void *)flags);
 		if (ret != SWAP_AGAIN || !page_mapped(page))
 			goto out;
 	}
@@ -1614,6 +1622,11 @@ out:
 	return ret;
 }
 
+static int page_not_mapped(struct page *page)
+{
+	return !page_mapped(page);
+};
+
 /**
  * try_to_unmap - try to remove all page table mappings to a page
  * @page: the page to get unmapped
@@ -1631,16 +1644,29 @@ out:
 int try_to_unmap(struct page *page, enum ttu_flags flags)
 {
 	int ret;
+	struct rmap_walk_control rwc = {
+		.rmap_one = try_to_unmap_one,
+		.arg = (void *)flags,
+		.done = page_not_mapped,
+		.file_nonlinear = try_to_unmap_nonlinear,
+		.anon_lock = page_lock_anon_vma_read,
+	};
 
-	BUG_ON(!PageLocked(page));
 	VM_BUG_ON(!PageHuge(page) && PageTransHuge(page));
 
-	if (unlikely(PageKsm(page)))
-		ret = try_to_unmap_ksm(page, flags);
-	else if (PageAnon(page))
-		ret = try_to_unmap_anon(page, flags);
-	else
-		ret = try_to_unmap_file(page, flags);
+	/*
+	 * During exec, a temporary VMA is setup and later moved.
+	 * The VMA is moved under the anon_vma lock but not the
+	 * page tables leading to a race where migration cannot
+	 * find the migration ptes. Rather than increasing the
+	 * locking requirements of exec(), migration skips
+	 * temporary VMAs until after exec() completes.
+	 */
+	if (flags & TTU_MIGRATION && !PageKsm(page) && PageAnon(page))
+		rwc.invalid_vma = invalid_migration_vma;
+
+	ret = rmap_walk(page, &rwc);
+
 	if (ret != SWAP_MLOCK && !page_mapped(page))
 		ret = SWAP_SUCCESS;
 	return ret;
@@ -1683,7 +1709,6 @@ void __put_anon_vma(struct anon_vma *anon_vma)
 	anon_vma_free(anon_vma);
 }
 
-#ifdef CONFIG_MIGRATION
 static struct anon_vma *rmap_walk_anon_lock(struct page *page,
 					struct rmap_walk_control *rwc)
 {
@@ -1785,7 +1810,6 @@ int rmap_walk(struct page *page, struct rmap_walk_control *rwc)
 	else
 		return rmap_walk_file(page, rwc);
 }
-#endif /* CONFIG_MIGRATION */
 
 #ifdef CONFIG_HUGETLB_PAGE
 /*