[PATCH v3 2/3] f2fs: support FALLOC_FL_COLLAPSE_RANGE

[PATCH v3 2/3] f2fs: support FALLOC_FL_COLLAPSE_RANGE

[Chao Yu]

Now, FALLOC_FL_COLLAPSE_RANGE flag in ->fallocate is supported in ext4/xfs.

In commit, the semantics of this flag is descripted as following:"
1) It collapses the range lying between offset and length by removing any
   data blocks which are present in this range and than updates all the
   logical offsets of extents beyond "offset + len" to nullify the hole
   created by removing blocks. In short, it does not leave a hole.
2) It should be used exclusively. No other fallocate flag in combination.
3) Offset and length supplied to fallocate should be fs block size aligned
   in case of xfs and ext4.
4) Collaspe range does not work beyond i_size."

This patch implements fallocate's FALLOC_FL_COLLAPSE_RANGE for f2fs.

Signed-off-by: Chao Yu <chao2.yu@samsung.com>
---
v2:
 * rebase to last git repository of f2fs (20150430)
 * remove replace_block() introduction.
v3:
 * rebase to last git repository of f2fs (20150506)
 * fix to pass right parameter to f2fs_replace_block().

 fs/f2fs/file.c | 135 +++++++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 132 insertions(+), 3 deletions(-)

diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index cb002c0..dbddc79 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -765,6 +765,132 @@ static int punch_hole(struct inode *inode, loff_t offset, loff_t len)
 	return ret;
 }
 
+static int f2fs_do_collapse(struct inode *inode, pgoff_t start, pgoff_t end)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct dnode_of_data dn;
+	pgoff_t nrpages = (i_size_read(inode) + PAGE_SIZE - 1) / PAGE_SIZE;
+	int ret = 0;
+
+	f2fs_lock_op(sbi);
+
+	for (; end < nrpages; start++, end++) {
+		block_t new_addr, old_addr;
+
+		set_new_dnode(&dn, inode, NULL, NULL, 0);
+		ret = get_dnode_of_data(&dn, end, LOOKUP_NODE_RA);
+		if (ret && ret != -ENOENT) {
+			goto out;
+		} else if (ret == -ENOENT) {
+			new_addr = NULL_ADDR;
+		} else {
+			new_addr = dn.data_blkaddr;
+			truncate_data_blocks_range(&dn, 1);
+			f2fs_put_dnode(&dn);
+		}
+
+		if (new_addr == NULL_ADDR) {
+			set_new_dnode(&dn, inode, NULL, NULL, 0);
+			ret = get_dnode_of_data(&dn, start, LOOKUP_NODE_RA);
+			if (ret && ret != -ENOENT)
+				goto out;
+			else if (ret == -ENOENT)
+				continue;
+
+			if (dn.data_blkaddr == NULL_ADDR) {
+				f2fs_put_dnode(&dn);
+				continue;
+			} else {
+				truncate_data_blocks_range(&dn, 1);
+			}
+
+			f2fs_put_dnode(&dn);
+		} else {
+			struct page *ipage;
+
+			ipage = get_node_page(sbi, inode->i_ino);
+			if (IS_ERR(ipage)) {
+				ret = PTR_ERR(ipage);
+				goto out;
+			}
+
+			set_new_dnode(&dn, inode, ipage, NULL, 0);
+			ret = f2fs_reserve_block(&dn, start);
+			if (ret)
+				goto out;
+
+			old_addr = dn.data_blkaddr;
+			if (old_addr != NEW_ADDR && new_addr == NEW_ADDR) {
+				dn.data_blkaddr = NULL_ADDR;
+				f2fs_update_extent_cache(&dn);
+				invalidate_blocks(sbi, old_addr);
+
+				dn.data_blkaddr = new_addr;
+				set_data_blkaddr(&dn);
+			} else if (new_addr != NEW_ADDR) {
+				struct node_info ni;
+				struct f2fs_summary sum;
+
+				get_node_info(sbi, dn.nid, &ni);
+				set_summary(&sum, dn.nid, dn.ofs_in_node,
+								ni.version);
+
+				f2fs_replace_block(sbi, &sum, old_addr,
+								new_addr, true);
+
+				dn.data_blkaddr = new_addr;
+				set_data_blkaddr(&dn);
+				f2fs_update_extent_cache(&dn);
+			}
+
+			f2fs_put_dnode(&dn);
+		}
+	}
+	ret = 0;
+out:
+	f2fs_unlock_op(sbi);
+	return ret;
+}
+
+static int f2fs_collapse_range(struct inode *inode, loff_t offset, loff_t len)
+{
+	pgoff_t pg_start, pg_end;
+	loff_t new_size;
+	int ret;
+
+	if (!S_ISREG(inode->i_mode))
+		return -EINVAL;
+
+	if (offset + len >= i_size_read(inode))
+		return -EINVAL;
+
+	/* collapse range should be aligned to block size of f2fs. */
+	if (offset & (F2FS_BLKSIZE - 1) || len & (F2FS_BLKSIZE - 1))
+		return -EINVAL;
+
+	pg_start = offset >> PAGE_CACHE_SHIFT;
+	pg_end = (offset + len) >> PAGE_CACHE_SHIFT;
+
+	/* write out all dirty pages from offset */
+	ret = filemap_write_and_wait_range(inode->i_mapping, offset, LLONG_MAX);
+	if (ret)
+		return ret;
+
+	truncate_pagecache(inode, offset);
+
+	ret = f2fs_do_collapse(inode, pg_start, pg_end);
+	if (ret)
+		return ret;
+
+	new_size = i_size_read(inode) - len;
+
+	ret = truncate_blocks(inode, new_size, true);
+	if (!ret)
+		i_size_write(inode, new_size);
+
+	return ret;
+}
+
 static int expand_inode_data(struct inode *inode, loff_t offset,
 					loff_t len, int mode)
 {
@@ -832,7 +958,8 @@ static long f2fs_fallocate(struct file *file, int mode,
 	struct inode *inode = file_inode(file);
 	long ret = 0;
 
-	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE))
+	if (mode & ~(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE |
+			FALLOC_FL_COLLAPSE_RANGE))
 		return -EOPNOTSUPP;
 
 	mutex_lock(&inode->i_mutex);
@@ -842,9 +969,11 @@ static long f2fs_fallocate(struct file *file, int mode,
 			goto out;
 
 		ret = punch_hole(inode, offset, len);
-	}
-	else
+	} else if (mode & FALLOC_FL_COLLAPSE_RANGE) {
+		ret = f2fs_collapse_range(inode, offset, len);
+	} else {
 		ret = expand_inode_data(inode, offset, len, mode);
+	}
 
 	if (!ret) {
 		inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-- 
2.3.3