[PATCH v4 0/6] Change page reference handling semantic of page cache

[PATCH v4 0/6] Change page reference handling semantic of page cache

[Minchan Kim]

Now we increase page reference on add_to_page_cache but don't decrease it
in remove_from_page_cache. Such asymmetric rule makes confusing about
page reference so that caller should notice it and comment why they
release page reference. It's not good API.

Long time ago, Hugh tried it[1] but gave up of reason which
reiser4's drop_page had to unlock the page between removing it from
page cache and doing the page_cache_release. But now the situation is
changed. I think at least things in current mainline doesn't have any
obstacles. The problem is fs or somethings out of mainline.
If it has done such thing like reiser4, this patch could be a problem but
they found it when compile time since we remove remove_from_page_cache.

[1] http://lkml.org/lkml/2004/10/24/140

The series configuration is following as. 

[1/6] : This patch introduces new API delete_from_page_cache.
[2,3,4/6] : Change remove_from_page_cache with delete_from_page_cache.
Intentionally I divide patch per file since someone might have a concern 
about releasing page reference of delete_from_page_cache in 
somecase (ex, truncate.c)
[5/6] : Remove old API so out of fs can meet compile error when build time
and can notice it.
[6/6] : Change __remove_from_page_cache with __delete_from_page_cache, too.
In this time, I made all-in-one patch because it doesn't change old behavior
so it has no concern. Just clean up patch.

This patch series pass LTP test on mm and fs.

from v3
 - Add Acked-by
 - rebase on mmotm-02-04
 - remove the patch about fuse

from v2
 - Add Acked-by
 - rebase on mmotm-01-06
 - change title of some patch

from v1
 - Add Acked-by
 - rebase on mmotm-12-23

Minchan Kim (6):
  [1/6] Introduce delete_from_page_cache
  [2/6] hugetlbfs: Change remove_from_page_cache
  [3/6] shmem: Change remove_from_page_cache
  [4/6] truncate: Change remove_from_page_cache
  [5/6] Good bye remove_from_page_cache
  [6/6] Change __remove_from_page_cache

 fs/hugetlbfs/inode.c    |    3 +--
 include/linux/pagemap.h |    4 ++--
 mm/filemap.c            |   22 ++++++++++++++++------
 mm/memory-failure.c     |    2 +-
 mm/shmem.c              |    3 +--
 mm/truncate.c           |    7 +++----
 mm/vmscan.c             |    2 +-
 7 files changed, 25 insertions(+), 18 deletions(-)

[PATCH v4 1/6] Introduce delete_from_page_cache

[Minchan Kim]

This function works as just wrapper remove_from_page_cache.
The difference is that it decreases page references in itself.
So caller have to make sure it has a page reference before calling.

This patch is ready for removing remove_from_page_cache.

Cc: Christoph Hellwig <hch@infradead.org>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 include/linux/pagemap.h |    1 +
 mm/filemap.c            |   16 ++++++++++++++++
 2 files changed, 17 insertions(+), 0 deletions(-)

diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 26946ad..a943985 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -457,6 +457,7 @@ int add_to_page_cache_lru(struct page *page, struct address_space *mapping,
 				pgoff_t index, gfp_t gfp_mask);
 extern void remove_from_page_cache(struct page *page);
 extern void __remove_from_page_cache(struct page *page);
+extern void delete_from_page_cache(struct page *page);
 int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask);
 
 /*
diff --git a/mm/filemap.c b/mm/filemap.c
index 3c89c96..f056d0c 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -166,6 +166,22 @@ void remove_from_page_cache(struct page *page)
 }
 EXPORT_SYMBOL(remove_from_page_cache);
 
+/**
+ * delete_from_page_cache - delete page from page cache
+ * @page: the page which the kernel is trying to remove from page cache
+ *
+ * This must be called only on pages that have
+ * been verified to be in the page cache and locked.
+ * It will never put the page into the free list,
+ * the caller has a reference on the page.
+ */
+void delete_from_page_cache(struct page *page)
+{
+	remove_from_page_cache(page);
+	page_cache_release(page);
+}
+EXPORT_SYMBOL(delete_from_page_cache);
+
 static int sync_page(void *word)
 {
 	struct address_space *mapping;
-- 
1.7.1

[PATCH v4 2/6] hugetlbfs: Change remove_from_page_cache

[Minchan Kim]

This patch series changes remove_from_page_cache's page ref counting
rule. Page cache ref count is decreased in delete_from_page_cache.
So we don't need decreasing page reference by caller.

Cc: William Irwin <wli@holomorphy.com>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 fs/hugetlbfs/inode.c |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 9885082..b9eeb1c 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -332,8 +332,7 @@ static void truncate_huge_page(struct page *page)
 {
 	cancel_dirty_page(page, /* No IO accounting for huge pages? */0);
 	ClearPageUptodate(page);
-	remove_from_page_cache(page);
-	put_page(page);
+	delete_from_page_cache(page);
 }
 
 static void truncate_hugepages(struct inode *inode, loff_t lstart)
-- 
1.7.1

[PATCH v4 3/6] shmem: Change remove_from_page_cache

[Minchan Kim]

This patch series changes remove_from_page_cache's page ref counting
rule. Page cache ref count is decreased in delete_from_page_cache.
So we don't need decreasing page reference by caller.

Acked-by:Hugh Dickins <hughd@google.com>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 mm/shmem.c |    3 +--
 1 files changed, 1 insertions(+), 2 deletions(-)

diff --git a/mm/shmem.c b/mm/shmem.c
index 7c9cdc6..4549134 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1081,7 +1081,7 @@ static int shmem_writepage(struct page *page, struct writeback_control *wbc)
 	shmem_recalc_inode(inode);
 
 	if (swap.val && add_to_swap_cache(page, swap, GFP_ATOMIC) == 0) {
-		remove_from_page_cache(page);
+		delete_from_page_cache(page);
 		shmem_swp_set(info, entry, swap.val);
 		shmem_swp_unmap(entry);
 		if (list_empty(&info->swaplist))
@@ -1091,7 +1091,6 @@ static int shmem_writepage(struct page *page, struct writeback_control *wbc)
 		spin_unlock(&info->lock);
 		swap_shmem_alloc(swap);
 		BUG_ON(page_mapped(page));
-		page_cache_release(page);	/* pagecache ref */
 		swap_writepage(page, wbc);
 		if (inode) {
 			mutex_lock(&shmem_swaplist_mutex);
-- 
1.7.1

[PATCH v4 4/6] truncate: Change remove_from_page_cache

[Minchan Kim]

This patch series changes remove_from_page_cache's page ref counting
rule. Page cache ref count is decreased in delete_from_page_cache.
So we don't need decreasing page reference by caller.

Cc: Dan Magenheimer <dan.magenheimer@oracle.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 mm/truncate.c |    5 ++---
 1 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/mm/truncate.c b/mm/truncate.c
index 4d415b3..faf65a5 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -108,13 +108,12 @@ truncate_complete_page(struct address_space *mapping, struct page *page)
 	cancel_dirty_page(page, PAGE_CACHE_SIZE);
 
 	clear_page_mlock(page);
-	remove_from_page_cache(page);
+	delete_from_page_cache(page);
 	ClearPageMappedToDisk(page);
-	/* this must be after the remove_from_page_cache which
+	/* this must be after the delete_from_page_cache which
 	 * calls cleancache_put_page (and note page->mapping is now NULL)
 	 */
 	cleancache_flush_page(mapping, page);
-	page_cache_release(page);	/* pagecache ref */
 	return 0;
 }
 
-- 
1.7.1

[PATCH v4 5/6] Good bye remove_from_page_cache

[Minchan Kim]

Now delete_from_page_cache replaces remove_from_page_cache.
So we remove remove_from_page_cache so fs or something out of
mainline will notice it when compile time and can fix it.

Cc: Christoph Hellwig <hch@infradead.org>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 include/linux/pagemap.h |    1 -
 mm/filemap.c            |   26 ++++++++++----------------
 2 files changed, 10 insertions(+), 17 deletions(-)

diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index a943985..631b1b6 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -455,7 +455,6 @@ int add_to_page_cache_locked(struct page *page, struct address_space *mapping,
 				pgoff_t index, gfp_t gfp_mask);
 int add_to_page_cache_lru(struct page *page, struct address_space *mapping,
 				pgoff_t index, gfp_t gfp_mask);
-extern void remove_from_page_cache(struct page *page);
 extern void __remove_from_page_cache(struct page *page);
 extern void delete_from_page_cache(struct page *page);
 int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask);
diff --git a/mm/filemap.c b/mm/filemap.c
index f056d0c..5f3a389 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -148,7 +148,16 @@ void __remove_from_page_cache(struct page *page)
 	}
 }
 
-void remove_from_page_cache(struct page *page)
+/**
+ * delete_from_page_cache - delete page from page cache
+ * @page: the page which the kernel is trying to remove from page cache
+ *
+ * This must be called only on pages that have
+ * been verified to be in the page cache and locked.
+ * It will never put the page into the free list,
+ * the caller has a reference on the page.
+ */
+void delete_from_page_cache(struct page *page)
 {
 	struct address_space *mapping = page->mapping;
 	void (*freepage)(struct page *);
@@ -163,21 +172,6 @@ void remove_from_page_cache(struct page *page)
 
 	if (freepage)
 		freepage(page);
-}
-EXPORT_SYMBOL(remove_from_page_cache);
-
-/**
- * delete_from_page_cache - delete page from page cache
- * @page: the page which the kernel is trying to remove from page cache
- *
- * This must be called only on pages that have
- * been verified to be in the page cache and locked.
- * It will never put the page into the free list,
- * the caller has a reference on the page.
- */
-void delete_from_page_cache(struct page *page)
-{
-	remove_from_page_cache(page);
 	page_cache_release(page);
 }
 EXPORT_SYMBOL(delete_from_page_cache);
-- 
1.7.1

[PATCH v4 6/6] Change __remove_from_page_cache

[Minchan Kim]

Now we renamed remove_from_page_cache with delete_from_page_cache.
As consistency of __remove_from_swap_cache and remove_from_swap_cache,
We change internal page cache handling function name, too.

Cc: Christoph Hellwig <hch@infradead.org>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
---
 include/linux/pagemap.h |    2 +-
 mm/filemap.c            |    8 ++++----
 mm/memory-failure.c     |    2 +-
 mm/truncate.c           |    2 +-
 mm/vmscan.c             |    2 +-
 5 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index 631b1b6..e407601 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -455,7 +455,7 @@ int add_to_page_cache_locked(struct page *page, struct address_space *mapping,
 				pgoff_t index, gfp_t gfp_mask);
 int add_to_page_cache_lru(struct page *page, struct address_space *mapping,
 				pgoff_t index, gfp_t gfp_mask);
-extern void __remove_from_page_cache(struct page *page);
+extern void __delete_from_page_cache(struct page *page);
 extern void delete_from_page_cache(struct page *page);
 int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask);
 
diff --git a/mm/filemap.c b/mm/filemap.c
index 5f3a389..351571f 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -109,11 +109,11 @@
  */
 
 /*
- * Remove a page from the page cache and free it. Caller has to make
+ * Delete a page from the page cache and free it. Caller has to make
  * sure the page is locked and that nobody else uses it - or that usage
  * is safe.  The caller must hold the mapping's tree_lock.
  */
-void __remove_from_page_cache(struct page *page)
+void __delete_from_page_cache(struct page *page)
 {
 	struct address_space *mapping = page->mapping;
 
@@ -166,7 +166,7 @@ void delete_from_page_cache(struct page *page)
 
 	freepage = mapping->a_ops->freepage;
 	spin_lock_irq(&mapping->tree_lock);
-	__remove_from_page_cache(page);
+	__delete_from_page_cache(page);
 	spin_unlock_irq(&mapping->tree_lock);
 	mem_cgroup_uncharge_cache_page(page);
 
@@ -453,7 +453,7 @@ int replace_page_cache_page(struct page *old, struct page *new, gfp_t gfp_mask)
 		new->index = offset;
 
 		spin_lock_irq(&mapping->tree_lock);
-		__remove_from_page_cache(old);
+		__delete_from_page_cache(old);
 		error = radix_tree_insert(&mapping->page_tree, offset, new);
 		BUG_ON(error);
 		mapping->nrpages++;
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 0207c2f..f15ceea 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -1130,7 +1130,7 @@ int __memory_failure(unsigned long pfn, int trapno, int flags)
 
 	/*
 	 * Now take care of user space mappings.
-	 * Abort on fail: __remove_from_page_cache() assumes unmapped page.
+	 * Abort on fail: __delete_from_page_cache() assumes unmapped page.
 	 */
 	if (hwpoison_user_mappings(p, pfn, trapno) != SWAP_SUCCESS) {
 		printk(KERN_ERR "MCE %#lx: cannot unmap page, give up\n", pfn);
diff --git a/mm/truncate.c b/mm/truncate.c
index faf65a5..94be6c0 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -394,7 +394,7 @@ invalidate_complete_page2(struct address_space *mapping, struct page *page)
 
 	clear_page_mlock(page);
 	BUG_ON(page_has_private(page));
-	__remove_from_page_cache(page);
+	__delete_from_page_cache(page);
 	spin_unlock_irq(&mapping->tree_lock);
 	mem_cgroup_uncharge_cache_page(page);
 
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 148c6e6..43bfc4e 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -514,7 +514,7 @@ static int __remove_mapping(struct address_space *mapping, struct page *page)
 
 		freepage = mapping->a_ops->freepage;
 
-		__remove_from_page_cache(page);
+		__delete_from_page_cache(page);
 		spin_unlock_irq(&mapping->tree_lock);
 		mem_cgroup_uncharge_cache_page(page);
 
-- 
1.7.1