[v2] Btrfs: handle unaligned tail of data ranges more efficient

Commit Message

Timofey Titovets Aug. 25, 2017, 8:47 a.m. UTC

At now while switch page bits in data ranges
we always hande +1 page, for cover case
where end of data range is not page aligned

Let's handle that case more obvious and efficient
Check end aligment directly and touch +1 page
only then needed

Signed-off-by: Timofey Titovets <[e-mail]>
---
 fs/btrfs/extent_io.c | 12 ++++++++++--
 fs/btrfs/inode.c     |  6 +++++-
 2 files changed, 15 insertions(+), 3 deletions(-)

--
2.14.1
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to [e-mail]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 0aff9b278c19..bf6195fa9425 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -1361,7 +1361,11 @@  void extent_range_clear_dirty_for_io(struct inode *inode, u64 start, u64 end)
 	unsigned long end_index = end >> PAGE_SHIFT;
 	struct page *page;

-	while (index <= end_index) {
+	/* Don't miss unaligned end */
+	if (!IS_ALIGNED(end, PAGE_SIZE))
+		end_index++;
+
+	while (index < end_index) {
 		page = find_get_page(inode->i_mapping, index);
 		BUG_ON(!page); /* Pages should be in the extent_io_tree */
 		clear_page_dirty_for_io(page);
@@ -1376,7 +1380,11 @@  void extent_range_redirty_for_io(struct inode *inode, u64 start, u64 end)
 	unsigned long end_index = end >> PAGE_SHIFT;
 	struct page *page;

-	while (index <= end_index) {
+	/* Don't miss unaligned end */
+	if (!IS_ALIGNED(end, PAGE_SIZE))
+		end_index++;
+
+	while (index < end_index) {
 		page = find_get_page(inode->i_mapping, index);
 		BUG_ON(!page); /* Pages should be in the extent_io_tree */
 		__set_page_dirty_nobuffers(page);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index c73f919491a0..444971d8ef2d 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -10710,7 +10710,11 @@  void btrfs_set_range_writeback(void *private_data, u64 start, u64 end)
 	unsigned long end_index = end >> PAGE_SHIFT;
 	struct page *page;

-	while (index <= end_index) {
+	/* Don't miss unaligned end */
+	if (!IS_ALIGNED(end, PAGE_SIZE))
+		end_index++;
+
+	while (index < end_index) {
 		page = find_get_page(inode->i_mapping, index);
 		ASSERT(page); /* Pages should be in the extent_io_tree */
 		set_page_writeback(page);