Re: [f2fs-dev] [PATCH V3] f2fs: large volume support

[Changman Lee <cm224.lee@samsung.com>]

Changes from V3
 o remove CP_LARGE_VOL_FLAG instead, use cp_payload in superblock
   because disk size is determined at format

Changes from V2
 o fix conversion like le32_to_cpu
 o use is_set_ckpt_flags instead of bit operation
 o check return value after memory allocation

Changes from V1
 o fix orphan node blkaddr for large volume

-- >8 --

>From b1610f32c472cb11f025e0997b7633d5c45b87e9 Mon Sep 17 00:00:00 2001
From: Changman Lee <cm224.lee@samsung.com>
Date: Mon, 12 May 2014 12:27:43 +0900
Subject: [PATCH] f2fs: large volume support

f2fs's cp has one page which consists of struct f2fs_checkpoint and
version bitmap of sit and nat. To support lots of segments, we need more
blocks for sit bitmap. So let's arrange sit bitmap as following:
+-----------------+------------+
| f2fs_checkpoint | sit bitmap |
| + nat bitmap    |            |
+-----------------+------------+
0                 4k        N blocks

Signed-off-by: Changman Lee <cm224.lee@samsung.com>
---
 fs/f2fs/checkpoint.c    |   45 ++++++++++++++++++++++++++++++++++++++++-----
 fs/f2fs/f2fs.h          |   13 +++++++++++--
 include/linux/f2fs_fs.h |    2 ++
 3 files changed, 53 insertions(+), 7 deletions(-)

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index fe968c7..c559946 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -371,7 +371,9 @@ void recover_orphan_inodes(struct f2fs_sb_info *sbi)
 		return;
 
 	sbi->por_doing = true;
-	start_blk = __start_cp_addr(sbi) + 1;
+
+	start_blk = __start_cp_addr(sbi) + 1 +
+		le32_to_cpu(F2FS_RAW_SUPER(sbi)->cp_payload);
 	orphan_blkaddr = __start_sum_addr(sbi) - 1;
 
 	ra_meta_pages(sbi, start_blk, orphan_blkaddr, META_CP);
@@ -512,8 +514,9 @@ int get_valid_checkpoint(struct f2fs_sb_info *sbi)
 	unsigned long blk_size = sbi->blocksize;
 	unsigned long long cp1_version = 0, cp2_version = 0;
 	unsigned long long cp_start_blk_no;
+	unsigned int cp_blks = 1 + le32_to_cpu(F2FS_RAW_SUPER(sbi)->cp_payload);
 
-	sbi->ckpt = kzalloc(blk_size, GFP_KERNEL);
+	sbi->ckpt = kzalloc(cp_blks * blk_size, GFP_KERNEL);
 	if (!sbi->ckpt)
 		return -ENOMEM;
 	/*
@@ -544,6 +547,25 @@ int get_valid_checkpoint(struct f2fs_sb_info *sbi)
 	cp_block = (struct f2fs_checkpoint *)page_address(cur_page);
 	memcpy(sbi->ckpt, cp_block, blk_size);
 
+	if (cp_blks > 1) {
+		int i;
+		block_t cp_blk_no;
+
+		cp_blk_no = le32_to_cpu(fsb->cp_blkaddr);
+		if (cur_page == cp2)
+			cp_blk_no += 1 << le32_to_cpu(fsb->log_blocks_per_seg);
+
+		for (i = 1; i < cp_blks; i++) {
+			void *sit_bitmap_ptr;
+			unsigned char *ckpt = (unsigned char *)sbi->ckpt;
+
+			cur_page = get_meta_page(sbi, cp_blk_no + i);
+			sit_bitmap_ptr = page_address(cur_page);
+			memcpy(ckpt + i * blk_size, sit_bitmap_ptr, blk_size);
+			f2fs_put_page(cur_page, 1);
+		}
+	}
+
 	f2fs_put_page(cp1, 1);
 	f2fs_put_page(cp2, 1);
 	return 0;
@@ -736,6 +758,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	__u32 crc32 = 0;
 	void *kaddr;
 	int i;
+	int cp_payload_blks = le32_to_cpu(F2FS_RAW_SUPER(sbi)->cp_payload);
 
 	/*
 	 * This avoids to conduct wrong roll-forward operations and uses
@@ -786,16 +809,19 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 
 	orphan_blocks = (sbi->n_orphans + F2FS_ORPHANS_PER_BLOCK - 1)
 					/ F2FS_ORPHANS_PER_BLOCK;
-	ckpt->cp_pack_start_sum = cpu_to_le32(1 + orphan_blocks);
+	ckpt->cp_pack_start_sum = cpu_to_le32(1 + cp_payload_blks +
+			orphan_blocks);
 
 	if (is_umount) {
 		set_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
 		ckpt->cp_pack_total_block_count = cpu_to_le32(2 +
-			data_sum_blocks + orphan_blocks + NR_CURSEG_NODE_TYPE);
+				cp_payload_blks + data_sum_blocks +
+				orphan_blocks + NR_CURSEG_NODE_TYPE);
 	} else {
 		clear_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
 		ckpt->cp_pack_total_block_count = cpu_to_le32(2 +
-			data_sum_blocks + orphan_blocks);
+				cp_payload_blks + data_sum_blocks +
+				orphan_blocks);
 	}
 
 	if (sbi->n_orphans)
@@ -821,6 +847,15 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount)
 	set_page_dirty(cp_page);
 	f2fs_put_page(cp_page, 1);
 
+	for (i = 1; i < 1 + cp_payload_blks; i++) {
+		cp_page = grab_meta_page(sbi, start_blk++);
+		kaddr = page_address(cp_page);
+		memcpy(kaddr, (char *)ckpt + i * F2FS_BLKSIZE,
+				(1 << sbi->log_blocksize));
+		set_page_dirty(cp_page);
+		f2fs_put_page(cp_page, 1);
+	}
+
 	if (sbi->n_orphans) {
 		write_orphan_inodes(sbi, start_blk);
 		start_blk += orphan_blocks;
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 676a2c6..9684b1f 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -764,9 +764,18 @@ static inline unsigned long __bitmap_size(struct f2fs_sb_info *sbi, int flag)
 static inline void *__bitmap_ptr(struct f2fs_sb_info *sbi, int flag)
 {
 	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
-	int offset = (flag == NAT_BITMAP) ?
+	int offset;
+
+	if (le32_to_cpu(F2FS_RAW_SUPER(sbi)->cp_payload) > 0) {
+		if (flag == NAT_BITMAP)
+			return &ckpt->sit_nat_version_bitmap;
+		else
+			return ((unsigned char *)ckpt + F2FS_BLKSIZE);
+	} else {
+		offset = (flag == NAT_BITMAP) ?
 			le32_to_cpu(ckpt->sit_ver_bitmap_bytesize) : 0;
-	return &ckpt->sit_nat_version_bitmap + offset;
+		return &ckpt->sit_nat_version_bitmap + offset;
+	}
 }
 
 static inline block_t __start_cp_addr(struct f2fs_sb_info *sbi)
diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h
index 8c03f71..23ba0b5 100644
--- a/include/linux/f2fs_fs.h
+++ b/include/linux/f2fs_fs.h
@@ -19,6 +19,7 @@
 #define F2FS_LOG_SECTORS_PER_BLOCK	3	/* 4KB: F2FS_BLKSIZE */
 #define F2FS_BLKSIZE			4096	/* support only 4KB block */
 #define F2FS_MAX_EXTENSION		64	/* # of extension entries */
+#define F2FS_BLK_ALIGN(x)	(((x) + F2FS_BLKSIZE - 1) / F2FS_BLKSIZE)
 
 #define NULL_ADDR		((block_t)0)	/* used as block_t addresses */
 #define NEW_ADDR		((block_t)-1)	/* used as block_t addresses */
@@ -75,6 +76,7 @@ struct f2fs_super_block {
 	__le16 volume_name[512];	/* volume name */
 	__le32 extension_count;		/* # of extensions below */
 	__u8 extension_list[F2FS_MAX_EXTENSION][8];	/* extension array */
+	__le32 cp_payload;
 } __packed;
 
 /*
-- 
1.7.10.4