From 88a70a69c088933011615fe26242e0335b012284 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Wed, 10 Dec 2014 15:20:48 -0800
Subject: f2fs: fix wrong condition check to trigger f2fs_sync_fs

If there is not enough available memory, we need to trigger f2fs_sync_fs.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 42607a679923..11e4b5c14616 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -290,7 +290,7 @@ void f2fs_balance_fs_bg(struct f2fs_sb_info *sbi)
 	/* check the # of cached NAT entries and prefree segments */
 	if (try_to_free_nats(sbi, NAT_ENTRY_PER_BLOCK) ||
 			excess_prefree_segs(sbi) ||
-			available_free_memory(sbi, INO_ENTRIES))
+			!available_free_memory(sbi, INO_ENTRIES))
 		f2fs_sync_fs(sbi->sb, true);
 }
 
-- 
cgit v1.2.3


From 70c640b1d6fd8484d0629b90052d6f6f738023fb Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Wed, 10 Dec 2014 13:59:33 -0800
Subject: f2fs: don't need to call lock_op and lock_page for abort

We don't need to call lock_op and lock_page at the aborting path.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c | 35 ++++++++++++++++++++---------------
 1 file changed, 20 insertions(+), 15 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 11e4b5c14616..3ce86c533604 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -240,33 +240,38 @@ void commit_inmem_pages(struct inode *inode, bool abort)
 	 * Otherwise, f2fs_gc in f2fs_balance_fs can wait forever until this
 	 * inode becomes free by iget_locked in f2fs_iget.
 	 */
-	if (!abort)
+	if (!abort) {
 		f2fs_balance_fs(sbi);
-
-	f2fs_lock_op(sbi);
+		f2fs_lock_op(sbi);
+	}
 
 	mutex_lock(&fi->inmem_lock);
 	list_for_each_entry_safe(cur, tmp, &fi->inmem_pages, list) {
-		lock_page(cur->page);
-		if (!abort && cur->page->mapping == inode->i_mapping) {
-			f2fs_wait_on_page_writeback(cur->page, DATA);
-			if (clear_page_dirty_for_io(cur->page))
-				inode_dec_dirty_pages(inode);
-			do_write_data_page(cur->page, &fio);
-			submit_bio = true;
+		if (!abort) {
+			lock_page(cur->page);
+			if (cur->page->mapping == inode->i_mapping) {
+				f2fs_wait_on_page_writeback(cur->page, DATA);
+				if (clear_page_dirty_for_io(cur->page))
+					inode_dec_dirty_pages(inode);
+				do_write_data_page(cur->page, &fio);
+				submit_bio = true;
+			}
+			f2fs_put_page(cur->page, 1);
+		} else {
+			put_page(cur->page);
 		}
 		radix_tree_delete(&fi->inmem_root, cur->page->index);
-		f2fs_put_page(cur->page, 1);
 		list_del(&cur->list);
 		kmem_cache_free(inmem_entry_slab, cur);
 		dec_page_count(F2FS_I_SB(inode), F2FS_INMEM_PAGES);
 	}
-	if (submit_bio)
-		f2fs_submit_merged_bio(sbi, DATA, WRITE);
 	mutex_unlock(&fi->inmem_lock);
 
-	filemap_fdatawait_range(inode->i_mapping, 0, LLONG_MAX);
-	f2fs_unlock_op(sbi);
+	if (!abort) {
+		f2fs_unlock_op(sbi);
+		if (submit_bio)
+			f2fs_submit_merged_bio(sbi, DATA, WRITE);
+	}
 }
 
 /*
-- 
cgit v1.2.3


From 1e84371ffeef451e8532e0cd04c2fe59ff10c514 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Tue, 9 Dec 2014 06:08:59 -0800
Subject: f2fs: change atomic and volatile write policies

This patch adds two new ioctls to release inmemory pages grabbed by atomic
writes.
 o f2fs_ioc_abort_volatile_write
  - If transaction was failed, all the grabbed pages and data should be written.
 o f2fs_ioc_release_volatile_write
  - This is to enhance the performance of PERSIST mode in sqlite.

In order to avoid huge memory consumption which causes OOM, this patch changes
volatile writes to use normal dirty pages, instead blocked flushing to the disk
as long as system does not suffer from memory pressure.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/data.c    |  9 +++++--
 fs/f2fs/f2fs.h    |  8 ++++++
 fs/f2fs/file.c    | 77 +++++++++++++++++++++++++++++++++++++++++++++++--------
 fs/f2fs/inode.c   |  2 +-
 fs/f2fs/node.c    |  3 +++
 fs/f2fs/node.h    |  1 +
 fs/f2fs/segment.c |  2 +-
 7 files changed, 88 insertions(+), 14 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 7ec697b37f19..32264e3d524f 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -814,6 +814,11 @@ static int f2fs_write_data_page(struct page *page,
 write:
 	if (unlikely(sbi->por_doing))
 		goto redirty_out;
+	if (f2fs_is_drop_cache(inode))
+		goto out;
+	if (f2fs_is_volatile_file(inode) && !wbc->for_reclaim &&
+			available_free_memory(sbi, BASE_CHECK))
+		goto redirty_out;
 
 	/* Dentry blocks are controlled by checkpoint */
 	if (S_ISDIR(inode->i_mode)) {
@@ -1109,7 +1114,7 @@ static void f2fs_invalidate_data_page(struct page *page, unsigned int offset,
 	if (offset % PAGE_CACHE_SIZE || length != PAGE_CACHE_SIZE)
 		return;
 
-	if (f2fs_is_atomic_file(inode) || f2fs_is_volatile_file(inode))
+	if (f2fs_is_atomic_file(inode))
 		invalidate_inmem_page(inode, page);
 
 	if (PageDirty(page))
@@ -1132,7 +1137,7 @@ static int f2fs_set_data_page_dirty(struct page *page)
 
 	SetPageUptodate(page);
 
-	if (f2fs_is_atomic_file(inode) || f2fs_is_volatile_file(inode)) {
+	if (f2fs_is_atomic_file(inode)) {
 		register_inmem_page(inode, page);
 		return 1;
 	}
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index ec58bb2373fc..8c9bf3d6cb7d 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -201,6 +201,8 @@ static inline bool __has_cursum_space(struct f2fs_summary_block *sum, int size,
 #define F2FS_IOC_START_ATOMIC_WRITE	_IO(F2FS_IOCTL_MAGIC, 1)
 #define F2FS_IOC_COMMIT_ATOMIC_WRITE	_IO(F2FS_IOCTL_MAGIC, 2)
 #define F2FS_IOC_START_VOLATILE_WRITE	_IO(F2FS_IOCTL_MAGIC, 3)
+#define F2FS_IOC_RELEASE_VOLATILE_WRITE	_IO(F2FS_IOCTL_MAGIC, 4)
+#define F2FS_IOC_ABORT_VOLATILE_WRITE	_IO(F2FS_IOCTL_MAGIC, 5)
 
 #if defined(__KERNEL__) && defined(CONFIG_COMPAT)
 /*
@@ -1113,6 +1115,7 @@ enum {
 	FI_NEED_IPU,		/* used for ipu per file */
 	FI_ATOMIC_FILE,		/* indicate atomic file */
 	FI_VOLATILE_FILE,	/* indicate volatile file */
+	FI_DROP_CACHE,		/* drop dirty page cache */
 	FI_DATA_EXIST,		/* indicate data exists */
 };
 
@@ -1220,6 +1223,11 @@ static inline bool f2fs_is_volatile_file(struct inode *inode)
 	return is_inode_flag_set(F2FS_I(inode), FI_VOLATILE_FILE);
 }
 
+static inline bool f2fs_is_drop_cache(struct inode *inode)
+{
+	return is_inode_flag_set(F2FS_I(inode), FI_DROP_CACHE);
+}
+
 static inline void *inline_data_addr(struct page *page)
 {
 	struct f2fs_inode *ri = F2FS_INODE(page);
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 3c27e0ecb3bc..5139f90e92c1 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -836,6 +836,19 @@ static long f2fs_fallocate(struct file *file, int mode,
 	return ret;
 }
 
+static int f2fs_release_file(struct inode *inode, struct file *filp)
+{
+	/* some remained atomic pages should discarded */
+	if (f2fs_is_atomic_file(inode))
+		commit_inmem_pages(inode, true);
+	if (f2fs_is_volatile_file(inode)) {
+		set_inode_flag(F2FS_I(inode), FI_DROP_CACHE);
+		filemap_fdatawrite(inode->i_mapping);
+		clear_inode_flag(F2FS_I(inode), FI_DROP_CACHE);
+	}
+	return 0;
+}
+
 #define F2FS_REG_FLMASK		(~(FS_DIRSYNC_FL | FS_TOPDIR_FL))
 #define F2FS_OTHER_FLMASK	(FS_NODUMP_FL | FS_NOATIME_FL)
 
@@ -909,26 +922,20 @@ out:
 static int f2fs_ioc_start_atomic_write(struct file *filp)
 {
 	struct inode *inode = file_inode(filp);
-	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 
 	if (!inode_owner_or_capable(inode))
 		return -EACCES;
 
-	f2fs_balance_fs(sbi);
+	f2fs_balance_fs(F2FS_I_SB(inode));
+
+	if (f2fs_is_atomic_file(inode))
+		return 0;
 
 	set_inode_flag(F2FS_I(inode), FI_ATOMIC_FILE);
 
 	return f2fs_convert_inline_inode(inode);
 }
 
-static int f2fs_release_file(struct inode *inode, struct file *filp)
-{
-	/* some remained atomic pages should discarded */
-	if (f2fs_is_atomic_file(inode) || f2fs_is_volatile_file(inode))
-		commit_inmem_pages(inode, true);
-	return 0;
-}
-
 static int f2fs_ioc_commit_atomic_write(struct file *filp)
 {
 	struct inode *inode = file_inode(filp);
@@ -949,6 +956,7 @@ static int f2fs_ioc_commit_atomic_write(struct file *filp)
 
 	ret = f2fs_sync_file(filp, 0, LONG_MAX, 0);
 	mnt_drop_write_file(filp);
+	clear_inode_flag(F2FS_I(inode), FI_ATOMIC_FILE);
 	return ret;
 }
 
@@ -959,11 +967,56 @@ static int f2fs_ioc_start_volatile_write(struct file *filp)
 	if (!inode_owner_or_capable(inode))
 		return -EACCES;
 
+	if (f2fs_is_volatile_file(inode))
+		return 0;
+
 	set_inode_flag(F2FS_I(inode), FI_VOLATILE_FILE);
 
 	return f2fs_convert_inline_inode(inode);
 }
 
+static int f2fs_ioc_release_volatile_write(struct file *filp)
+{
+	struct inode *inode = file_inode(filp);
+
+	if (!inode_owner_or_capable(inode))
+		return -EACCES;
+
+	if (!f2fs_is_volatile_file(inode))
+		return 0;
+
+	punch_hole(inode, 0, F2FS_BLKSIZE);
+	return 0;
+}
+
+static int f2fs_ioc_abort_volatile_write(struct file *filp)
+{
+	struct inode *inode = file_inode(filp);
+	int ret;
+
+	if (!inode_owner_or_capable(inode))
+		return -EACCES;
+
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
+	f2fs_balance_fs(F2FS_I_SB(inode));
+
+	if (f2fs_is_atomic_file(inode)) {
+		commit_inmem_pages(inode, false);
+		clear_inode_flag(F2FS_I(inode), FI_ATOMIC_FILE);
+	}
+
+	if (f2fs_is_volatile_file(inode)) {
+		clear_inode_flag(F2FS_I(inode), FI_VOLATILE_FILE);
+		filemap_fdatawrite(inode->i_mapping);
+		set_inode_flag(F2FS_I(inode), FI_VOLATILE_FILE);
+	}
+	mnt_drop_write_file(filp);
+	return ret;
+}
+
 static int f2fs_ioc_fitrim(struct file *filp, unsigned long arg)
 {
 	struct inode *inode = file_inode(filp);
@@ -1007,6 +1060,10 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		return f2fs_ioc_commit_atomic_write(filp);
 	case F2FS_IOC_START_VOLATILE_WRITE:
 		return f2fs_ioc_start_volatile_write(filp);
+	case F2FS_IOC_RELEASE_VOLATILE_WRITE:
+		return f2fs_ioc_release_volatile_write(filp);
+	case F2FS_IOC_ABORT_VOLATILE_WRITE:
+		return f2fs_ioc_abort_volatile_write(filp);
 	case FITRIM:
 		return f2fs_ioc_fitrim(filp, arg);
 	default:
diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
index 196cc7843aaf..3a8958d1684f 100644
--- a/fs/f2fs/inode.c
+++ b/fs/f2fs/inode.c
@@ -304,7 +304,7 @@ void f2fs_evict_inode(struct inode *inode)
 	nid_t xnid = F2FS_I(inode)->i_xattr_nid;
 
 	/* some remained atomic pages should discarded */
-	if (f2fs_is_atomic_file(inode) || f2fs_is_volatile_file(inode))
+	if (f2fs_is_atomic_file(inode))
 		commit_inmem_pages(inode, true);
 
 	trace_f2fs_evict_inode(inode);
diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index d6f073edd861..cabecee88377 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -61,6 +61,9 @@ bool available_free_memory(struct f2fs_sb_info *sbi, int type)
 			mem_size += (sbi->im[i].ino_num *
 				sizeof(struct ino_entry)) >> PAGE_CACHE_SHIFT;
 		res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 1);
+	} else {
+		if (sbi->sb->s_bdi->dirty_exceeded)
+			return false;
 	}
 	return res;
 }
diff --git a/fs/f2fs/node.h b/fs/f2fs/node.h
index d10b6448a671..036b887176ff 100644
--- a/fs/f2fs/node.h
+++ b/fs/f2fs/node.h
@@ -108,6 +108,7 @@ enum mem_type {
 	NAT_ENTRIES,	/* indicates the cached nat entry */
 	DIRTY_DENTS,	/* indicates dirty dentry pages */
 	INO_ENTRIES,	/* indicates inode entries */
+	BASE_CHECK,	/* check kernel status */
 };
 
 struct nat_entry_set {
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 3ce86c533604..de9c0700c88e 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -230,7 +230,7 @@ void commit_inmem_pages(struct inode *inode, bool abort)
 	bool submit_bio = false;
 	struct f2fs_io_info fio = {
 		.type = DATA,
-		.rw = WRITE_SYNC,
+		.rw = WRITE_SYNC | REQ_PRIO,
 	};
 
 	/*
-- 
cgit v1.2.3


From d7bc2484b8d4e580370c66ad93c4319225bd104d Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Fri, 12 Dec 2014 13:53:41 -0800
Subject: f2fs: fix small discards not to issue redundantly

The ckpt_valid_map and cur_valid_map are synced by seg_info_to_raw_sit.

In the case of small discards, the candidates are selected before sync,
while fitrim selects candidates after sync.

So, for small discards, we need to add candidates only just being obsoleted.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index de9c0700c88e..335418c9a06b 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -525,7 +525,8 @@ static void add_discard_addrs(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	bool force = (cpc->reason == CP_DISCARD);
 	int i;
 
-	if (!force && !test_opt(sbi, DISCARD))
+	if (!force && (!test_opt(sbi, DISCARD) ||
+			SM_I(sbi)->nr_discards >= SM_I(sbi)->max_discards))
 		return;
 
 	if (force && !se->valid_blocks) {
@@ -553,7 +554,8 @@ static void add_discard_addrs(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 
 	/* SIT_VBLOCK_MAP_SIZE should be multiple of sizeof(unsigned long) */
 	for (i = 0; i < entries; i++)
-		dmap[i] = ~(cur_map[i] | ckpt_map[i]);
+		dmap[i] = force ? ~ckpt_map[i] :
+				(cur_map[i] ^ ckpt_map[i]) & ckpt_map[i];
 
 	while (force || SM_I(sbi)->nr_discards <= SM_I(sbi)->max_discards) {
 		start = __find_rev_next_bit(dmap, max_blocks, end + 1);
@@ -1759,7 +1761,7 @@ void flush_sit_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 			se = get_seg_entry(sbi, segno);
 
 			/* add discard candidates */
-			if (SM_I(sbi)->nr_discards < SM_I(sbi)->max_discards) {
+			if (cpc->reason != CP_DISCARD) {
 				cpc->trim_start = segno;
 				add_discard_addrs(sbi, cpc);
 			}
-- 
cgit v1.2.3


From 042b7816aaebb1eb137b9889c20b595d951d15b7 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Fri, 12 Dec 2014 19:40:02 -0800
Subject: f2fs: remove unnecessary call to invalidate inmemory pages

Now we use inmemory pages for atomic write only and provide abort procedure,
we don't need to truncate them explicitly.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/data.c    |  3 ---
 fs/f2fs/f2fs.h    |  1 -
 fs/f2fs/segment.c | 17 -----------------
 3 files changed, 21 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 32264e3d524f..caa08e46697a 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -1114,9 +1114,6 @@ static void f2fs_invalidate_data_page(struct page *page, unsigned int offset,
 	if (offset % PAGE_CACHE_SIZE || length != PAGE_CACHE_SIZE)
 		return;
 
-	if (f2fs_is_atomic_file(inode))
-		invalidate_inmem_page(inode, page);
-
 	if (PageDirty(page))
 		inode_dec_dirty_pages(inode);
 	ClearPagePrivate(page);
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 8c9bf3d6cb7d..d3699da17e6b 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -1397,7 +1397,6 @@ void destroy_node_manager_caches(void);
  * segment.c
  */
 void register_inmem_page(struct inode *, struct page *);
-void invalidate_inmem_page(struct inode *, struct page *);
 void commit_inmem_pages(struct inode *, bool);
 void f2fs_balance_fs(struct f2fs_sb_info *);
 void f2fs_balance_fs_bg(struct f2fs_sb_info *);
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 335418c9a06b..3791fa93dc7b 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -205,23 +205,6 @@ retry:
 	mutex_unlock(&fi->inmem_lock);
 }
 
-void invalidate_inmem_page(struct inode *inode, struct page *page)
-{
-	struct f2fs_inode_info *fi = F2FS_I(inode);
-	struct inmem_pages *cur;
-
-	mutex_lock(&fi->inmem_lock);
-	cur = radix_tree_lookup(&fi->inmem_root, page->index);
-	if (cur) {
-		radix_tree_delete(&fi->inmem_root, cur->page->index);
-		f2fs_put_page(cur->page, 0);
-		list_del(&cur->list);
-		kmem_cache_free(inmem_entry_slab, cur);
-		dec_page_count(F2FS_I_SB(inode), F2FS_INMEM_PAGES);
-	}
-	mutex_unlock(&fi->inmem_lock);
-}
-
 void commit_inmem_pages(struct inode *inode, bool abort)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-- 
cgit v1.2.3


From 3fa06d7bc9f579bd180e879fd1c9bdb6b1b0d9b7 Mon Sep 17 00:00:00 2001
From: Chao Yu <chao2.yu@samsung.com>
Date: Tue, 9 Dec 2014 14:21:46 +0800
Subject: f2fs: readahead contiguous current summary blocks in checkpoint

Let's add readahead code for reading contiguous compact/normal summary blocks
in checkpoint, then we will gain better performance in mount procedure.

Changes from v1
  o remove inappropriate 'unlikely' in npages_for_summary_flush.

Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/checkpoint.c |  2 +-
 fs/f2fs/f2fs.h       |  2 +-
 fs/f2fs/segment.c    | 21 ++++++++++++++++++---
 3 files changed, 20 insertions(+), 5 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index e6c271fefaca..825158e4855a 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -922,7 +922,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	ckpt->next_free_nid = cpu_to_le32(last_nid);
 
 	/* 2 cp  + n data seg summary + orphan inode blocks */
-	data_sum_blocks = npages_for_summary_flush(sbi);
+	data_sum_blocks = npages_for_summary_flush(sbi, false);
 	if (data_sum_blocks < NR_CURSEG_DATA_TYPE)
 		set_ckpt_flags(ckpt, CP_COMPACT_SUM_FLAG);
 	else
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index d3699da17e6b..a8ccbceb869a 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -1408,7 +1408,7 @@ void refresh_sit_entry(struct f2fs_sb_info *, block_t, block_t);
 void clear_prefree_segments(struct f2fs_sb_info *);
 void release_discard_addrs(struct f2fs_sb_info *);
 void discard_next_dnode(struct f2fs_sb_info *, block_t);
-int npages_for_summary_flush(struct f2fs_sb_info *);
+int npages_for_summary_flush(struct f2fs_sb_info *, bool);
 void allocate_new_segments(struct f2fs_sb_info *);
 int f2fs_trim_fs(struct f2fs_sb_info *, struct fstrim_range *);
 struct page *get_sum_page(struct f2fs_sb_info *, unsigned int);
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 3791fa93dc7b..c950c9318e8e 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -725,7 +725,7 @@ static void __add_sum_entry(struct f2fs_sb_info *sbi, int type,
 /*
  * Calculate the number of current summary pages for writing
  */
-int npages_for_summary_flush(struct f2fs_sb_info *sbi)
+int npages_for_summary_flush(struct f2fs_sb_info *sbi, bool for_ra)
 {
 	int valid_sum_count = 0;
 	int i, sum_in_page;
@@ -733,8 +733,13 @@ int npages_for_summary_flush(struct f2fs_sb_info *sbi)
 	for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_DATA; i++) {
 		if (sbi->ckpt->alloc_type[i] == SSR)
 			valid_sum_count += sbi->blocks_per_seg;
-		else
-			valid_sum_count += curseg_blkoff(sbi, i);
+		else {
+			if (for_ra)
+				valid_sum_count += le16_to_cpu(
+					F2FS_CKPT(sbi)->cur_data_blkoff[i]);
+			else
+				valid_sum_count += curseg_blkoff(sbi, i);
+		}
 	}
 
 	sum_in_page = (PAGE_CACHE_SIZE - 2 * SUM_JOURNAL_SIZE -
@@ -1440,12 +1445,22 @@ static int restore_curseg_summaries(struct f2fs_sb_info *sbi)
 	int err;
 
 	if (is_set_ckpt_flags(F2FS_CKPT(sbi), CP_COMPACT_SUM_FLAG)) {
+		int npages = npages_for_summary_flush(sbi, true);
+
+		if (npages >= 2)
+			ra_meta_pages(sbi, start_sum_block(sbi), npages,
+								META_CP);
+
 		/* restore for compacted data summary */
 		if (read_compacted_summaries(sbi))
 			return -EINVAL;
 		type = CURSEG_HOT_NODE;
 	}
 
+	if (is_set_ckpt_flags(F2FS_CKPT(sbi), CP_UMOUNT_FLAG))
+		ra_meta_pages(sbi, sum_blk_addr(sbi, NR_CURSEG_TYPE, type),
+					NR_CURSEG_TYPE - type, META_CP);
+
 	for (; type <= CURSEG_COLD_NODE; type++) {
 		err = read_normal_summaries(sbi, type);
 		if (err)
-- 
cgit v1.2.3


From cf04e8eb55290c7b836c36f0b4e1a8d0fe8ee275 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Wed, 17 Dec 2014 19:33:13 -0800
Subject: f2fs: use f2fs_io_info to clean up messy parameters during IO path

This patch cleans up parameters on IO paths.
The key idea is to use f2fs_io_info adding a parameter, block address, and then
use this structure as parameters.

Reviewed-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/checkpoint.c | 24 ++++++++++---------
 fs/f2fs/data.c       | 66 +++++++++++++++++++++++++++++++---------------------
 fs/f2fs/f2fs.h       | 14 ++++++-----
 fs/f2fs/inline.c     |  7 +++---
 fs/f2fs/node.c       | 14 +++++++----
 fs/f2fs/segment.c    | 28 +++++++++++-----------
 6 files changed, 87 insertions(+), 66 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 825158e4855a..0ac7c39605ed 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -50,6 +50,11 @@ struct page *get_meta_page(struct f2fs_sb_info *sbi, pgoff_t index)
 {
 	struct address_space *mapping = META_MAPPING(sbi);
 	struct page *page;
+	struct f2fs_io_info fio = {
+		.type = META,
+		.rw = READ_SYNC | REQ_META | REQ_PRIO,
+		.blk_addr = index,
+	};
 repeat:
 	page = grab_cache_page(mapping, index);
 	if (!page) {
@@ -59,8 +64,7 @@ repeat:
 	if (PageUptodate(page))
 		goto out;
 
-	if (f2fs_submit_page_bio(sbi, page, index,
-				READ_SYNC | REQ_META | REQ_PRIO))
+	if (f2fs_submit_page_bio(sbi, page, &fio))
 		goto repeat;
 
 	lock_page(page);
@@ -112,14 +116,12 @@ int ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages, int type
 	block_t prev_blk_addr = 0;
 	struct page *page;
 	block_t blkno = start;
-
 	struct f2fs_io_info fio = {
 		.type = META,
 		.rw = READ_SYNC | REQ_META | REQ_PRIO
 	};
 
 	for (; nrpages-- > 0; blkno++) {
-		block_t blk_addr;
 
 		if (!is_valid_blkaddr(sbi, blkno, type))
 			goto out;
@@ -130,27 +132,27 @@ int ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages, int type
 					NAT_BLOCK_OFFSET(NM_I(sbi)->max_nid)))
 				blkno = 0;
 			/* get nat block addr */
-			blk_addr = current_nat_addr(sbi,
+			fio.blk_addr = current_nat_addr(sbi,
 					blkno * NAT_ENTRY_PER_BLOCK);
 			break;
 		case META_SIT:
 			/* get sit block addr */
-			blk_addr = current_sit_addr(sbi,
+			fio.blk_addr = current_sit_addr(sbi,
 					blkno * SIT_ENTRY_PER_BLOCK);
-			if (blkno != start && prev_blk_addr + 1 != blk_addr)
+			if (blkno != start && prev_blk_addr + 1 != fio.blk_addr)
 				goto out;
-			prev_blk_addr = blk_addr;
+			prev_blk_addr = fio.blk_addr;
 			break;
 		case META_SSA:
 		case META_CP:
 		case META_POR:
-			blk_addr = blkno;
+			fio.blk_addr = blkno;
 			break;
 		default:
 			BUG();
 		}
 
-		page = grab_cache_page(META_MAPPING(sbi), blk_addr);
+		page = grab_cache_page(META_MAPPING(sbi), fio.blk_addr);
 		if (!page)
 			continue;
 		if (PageUptodate(page)) {
@@ -158,7 +160,7 @@ int ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages, int type
 			continue;
 		}
 
-		f2fs_submit_page_mbio(sbi, page, blk_addr, &fio);
+		f2fs_submit_page_mbio(sbi, page, &fio);
 		f2fs_put_page(page, 0);
 	}
 out:
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index caa08e46697a..d86f8b1413f1 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -132,14 +132,14 @@ void f2fs_submit_merged_bio(struct f2fs_sb_info *sbi,
  * Return unlocked page.
  */
 int f2fs_submit_page_bio(struct f2fs_sb_info *sbi, struct page *page,
-					block_t blk_addr, int rw)
+					struct f2fs_io_info *fio)
 {
 	struct bio *bio;
 
-	trace_f2fs_submit_page_bio(page, blk_addr, rw);
+	trace_f2fs_submit_page_bio(page, fio->blk_addr, fio->rw);
 
 	/* Allocate a new bio */
-	bio = __bio_alloc(sbi, blk_addr, 1, is_read_io(rw));
+	bio = __bio_alloc(sbi, fio->blk_addr, 1, is_read_io(fio->rw));
 
 	if (bio_add_page(bio, page, PAGE_CACHE_SIZE, 0) < PAGE_CACHE_SIZE) {
 		bio_put(bio);
@@ -147,12 +147,12 @@ int f2fs_submit_page_bio(struct f2fs_sb_info *sbi, struct page *page,
 		return -EFAULT;
 	}
 
-	submit_bio(rw, bio);
+	submit_bio(fio->rw, bio);
 	return 0;
 }
 
 void f2fs_submit_page_mbio(struct f2fs_sb_info *sbi, struct page *page,
-			block_t blk_addr, struct f2fs_io_info *fio)
+					struct f2fs_io_info *fio)
 {
 	enum page_type btype = PAGE_TYPE_OF_BIO(fio->type);
 	struct f2fs_bio_info *io;
@@ -160,21 +160,21 @@ void f2fs_submit_page_mbio(struct f2fs_sb_info *sbi, struct page *page,
 
 	io = is_read ? &sbi->read_io : &sbi->write_io[btype];
 
-	verify_block_addr(sbi, blk_addr);
+	verify_block_addr(sbi, fio->blk_addr);
 
 	down_write(&io->io_rwsem);
 
 	if (!is_read)
 		inc_page_count(sbi, F2FS_WRITEBACK);
 
-	if (io->bio && (io->last_block_in_bio != blk_addr - 1 ||
+	if (io->bio && (io->last_block_in_bio != fio->blk_addr - 1 ||
 						io->fio.rw != fio->rw))
 		__submit_merged_bio(io);
 alloc_new:
 	if (io->bio == NULL) {
 		int bio_blocks = MAX_BIO_BLOCKS(sbi);
 
-		io->bio = __bio_alloc(sbi, blk_addr, bio_blocks, is_read);
+		io->bio = __bio_alloc(sbi, fio->blk_addr, bio_blocks, is_read);
 		io->fio = *fio;
 	}
 
@@ -184,10 +184,10 @@ alloc_new:
 		goto alloc_new;
 	}
 
-	io->last_block_in_bio = blk_addr;
+	io->last_block_in_bio = fio->blk_addr;
 
 	up_write(&io->io_rwsem);
-	trace_f2fs_submit_page_mbio(page, fio->rw, fio->type, blk_addr);
+	trace_f2fs_submit_page_mbio(page, fio->rw, fio->type, fio->blk_addr);
 }
 
 /*
@@ -376,6 +376,10 @@ struct page *find_data_page(struct inode *inode, pgoff_t index, bool sync)
 	struct dnode_of_data dn;
 	struct page *page;
 	int err;
+	struct f2fs_io_info fio = {
+		.type = DATA,
+		.rw = sync ? READ_SYNC : READA,
+	};
 
 	page = find_get_page(mapping, index);
 	if (page && PageUptodate(page))
@@ -404,8 +408,8 @@ struct page *find_data_page(struct inode *inode, pgoff_t index, bool sync)
 		return page;
 	}
 
-	err = f2fs_submit_page_bio(F2FS_I_SB(inode), page, dn.data_blkaddr,
-					sync ? READ_SYNC : READA);
+	fio.blk_addr = dn.data_blkaddr;
+	err = f2fs_submit_page_bio(F2FS_I_SB(inode), page, &fio);
 	if (err)
 		return ERR_PTR(err);
 
@@ -430,7 +434,10 @@ struct page *get_lock_data_page(struct inode *inode, pgoff_t index)
 	struct dnode_of_data dn;
 	struct page *page;
 	int err;
-
+	struct f2fs_io_info fio = {
+		.type = DATA,
+		.rw = READ_SYNC,
+	};
 repeat:
 	page = grab_cache_page(mapping, index);
 	if (!page)
@@ -464,8 +471,8 @@ repeat:
 		return page;
 	}
 
-	err = f2fs_submit_page_bio(F2FS_I_SB(inode), page,
-					dn.data_blkaddr, READ_SYNC);
+	fio.blk_addr = dn.data_blkaddr;
+	err = f2fs_submit_page_bio(F2FS_I_SB(inode), page, &fio);
 	if (err)
 		return ERR_PTR(err);
 
@@ -515,8 +522,12 @@ repeat:
 		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
 		SetPageUptodate(page);
 	} else {
-		err = f2fs_submit_page_bio(F2FS_I_SB(inode), page,
-						dn.data_blkaddr, READ_SYNC);
+		struct f2fs_io_info fio = {
+			.type = DATA,
+			.rw = READ_SYNC,
+			.blk_addr = dn.data_blkaddr,
+		};
+		err = f2fs_submit_page_bio(F2FS_I_SB(inode), page, &fio);
 		if (err)
 			goto put_err;
 
@@ -745,7 +756,6 @@ static int f2fs_read_data_pages(struct file *file,
 int do_write_data_page(struct page *page, struct f2fs_io_info *fio)
 {
 	struct inode *inode = page->mapping->host;
-	block_t old_blkaddr, new_blkaddr;
 	struct dnode_of_data dn;
 	int err = 0;
 
@@ -754,10 +764,10 @@ int do_write_data_page(struct page *page, struct f2fs_io_info *fio)
 	if (err)
 		return err;
 
-	old_blkaddr = dn.data_blkaddr;
+	fio->blk_addr = dn.data_blkaddr;
 
 	/* This page is already truncated */
-	if (old_blkaddr == NULL_ADDR)
+	if (fio->blk_addr == NULL_ADDR)
 		goto out_writepage;
 
 	set_page_writeback(page);
@@ -766,14 +776,14 @@ int do_write_data_page(struct page *page, struct f2fs_io_info *fio)
 	 * If current allocation needs SSR,
 	 * it had better in-place writes for updated data.
 	 */
-	if (unlikely(old_blkaddr != NEW_ADDR &&
+	if (unlikely(fio->blk_addr != NEW_ADDR &&
 			!is_cold_data(page) &&
 			need_inplace_update(inode))) {
-		rewrite_data_page(page, old_blkaddr, fio);
+		rewrite_data_page(page, fio);
 		set_inode_flag(F2FS_I(inode), FI_UPDATE_WRITE);
 	} else {
-		write_data_page(page, &dn, &new_blkaddr, fio);
-		update_extent_cache(new_blkaddr, &dn);
+		write_data_page(page, &dn, fio);
+		update_extent_cache(fio->blk_addr, &dn);
 		set_inode_flag(F2FS_I(inode), FI_APPEND_WRITE);
 	}
 out_writepage:
@@ -1007,8 +1017,12 @@ put_next:
 	if (dn.data_blkaddr == NEW_ADDR) {
 		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
 	} else {
-		err = f2fs_submit_page_bio(sbi, page, dn.data_blkaddr,
-					   READ_SYNC);
+		struct f2fs_io_info fio = {
+			.type = DATA,
+			.rw = READ_SYNC,
+			.blk_addr = dn.data_blkaddr,
+		};
+		err = f2fs_submit_page_bio(sbi, page, &fio);
 		if (err)
 			goto fail;
 
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index a8ccbceb869a..3f07b504f6d8 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -491,6 +491,7 @@ enum page_type {
 struct f2fs_io_info {
 	enum page_type type;	/* contains DATA/NODE/META/META_FLUSH */
 	int rw;			/* contains R/RS/W/WS with REQ_META/REQ_PRIO */
+	block_t blk_addr;	/* block address to be written */
 };
 
 #define is_read_io(rw)	(((rw) & 1) == READ)
@@ -1414,10 +1415,10 @@ int f2fs_trim_fs(struct f2fs_sb_info *, struct fstrim_range *);
 struct page *get_sum_page(struct f2fs_sb_info *, unsigned int);
 void write_meta_page(struct f2fs_sb_info *, struct page *);
 void write_node_page(struct f2fs_sb_info *, struct page *,
-		struct f2fs_io_info *, unsigned int, block_t, block_t *);
-void write_data_page(struct page *, struct dnode_of_data *, block_t *,
-					struct f2fs_io_info *);
-void rewrite_data_page(struct page *, block_t, struct f2fs_io_info *);
+				unsigned int, struct f2fs_io_info *);
+void write_data_page(struct page *, struct dnode_of_data *,
+			struct f2fs_io_info *);
+void rewrite_data_page(struct page *, struct f2fs_io_info *);
 void recover_data_page(struct f2fs_sb_info *, struct page *,
 				struct f2fs_summary *, block_t, block_t);
 void allocate_data_block(struct f2fs_sb_info *, struct page *,
@@ -1464,8 +1465,9 @@ void destroy_checkpoint_caches(void);
  * data.c
  */
 void f2fs_submit_merged_bio(struct f2fs_sb_info *, enum page_type, int);
-int f2fs_submit_page_bio(struct f2fs_sb_info *, struct page *, block_t, int);
-void f2fs_submit_page_mbio(struct f2fs_sb_info *, struct page *, block_t,
+int f2fs_submit_page_bio(struct f2fs_sb_info *, struct page *,
+						struct f2fs_io_info *);
+void f2fs_submit_page_mbio(struct f2fs_sb_info *, struct page *,
 						struct f2fs_io_info *);
 int reserve_new_block(struct dnode_of_data *);
 int f2fs_reserve_block(struct dnode_of_data *, pgoff_t);
diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c
index f2d3c581e776..0c3f3f9b9f88 100644
--- a/fs/f2fs/inline.c
+++ b/fs/f2fs/inline.c
@@ -79,7 +79,6 @@ int f2fs_read_inline_data(struct inode *inode, struct page *page)
 int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page)
 {
 	void *src_addr, *dst_addr;
-	block_t new_blk_addr;
 	struct f2fs_io_info fio = {
 		.type = DATA,
 		.rw = WRITE_SYNC | REQ_PRIO,
@@ -115,9 +114,9 @@ no_update:
 
 	/* write data page to try to make data consistent */
 	set_page_writeback(page);
-
-	write_data_page(page, dn, &new_blk_addr, &fio);
-	update_extent_cache(new_blk_addr, dn);
+	fio.blk_addr = dn->data_blkaddr;
+	write_data_page(page, dn, &fio);
+	update_extent_cache(fio.blk_addr, dn);
 	f2fs_wait_on_page_writeback(page, DATA);
 	if (dirty)
 		inode_dec_dirty_pages(dn->inode);
diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index bcfd67c80196..adc35c978306 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -977,6 +977,10 @@ static int read_node_page(struct page *page, int rw)
 {
 	struct f2fs_sb_info *sbi = F2FS_P_SB(page);
 	struct node_info ni;
+	struct f2fs_io_info fio = {
+		.type = NODE,
+		.rw = rw,
+	};
 
 	get_node_info(sbi, page->index, &ni);
 
@@ -988,7 +992,8 @@ static int read_node_page(struct page *page, int rw)
 	if (PageUptodate(page))
 		return LOCKED_PAGE;
 
-	return f2fs_submit_page_bio(sbi, page, ni.blk_addr, rw);
+	fio.blk_addr = ni.blk_addr;
+	return f2fs_submit_page_bio(sbi, page, &fio);
 }
 
 /*
@@ -1269,7 +1274,6 @@ static int f2fs_write_node_page(struct page *page,
 {
 	struct f2fs_sb_info *sbi = F2FS_P_SB(page);
 	nid_t nid;
-	block_t new_addr;
 	struct node_info ni;
 	struct f2fs_io_info fio = {
 		.type = NODE,
@@ -1304,9 +1308,11 @@ static int f2fs_write_node_page(struct page *page,
 	} else {
 		down_read(&sbi->node_write);
 	}
+
 	set_page_writeback(page);
-	write_node_page(sbi, page, &fio, nid, ni.blk_addr, &new_addr);
-	set_node_addr(sbi, &ni, new_addr, is_fsync_dnode(page));
+	fio.blk_addr = ni.blk_addr;
+	write_node_page(sbi, page, nid, &fio);
+	set_node_addr(sbi, &ni, fio.blk_addr, is_fsync_dnode(page));
 	dec_page_count(sbi, F2FS_DIRTY_NODES);
 	up_read(&sbi->node_write);
 	unlock_page(page);
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index c950c9318e8e..c726f86c2ea0 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -1182,39 +1182,39 @@ void allocate_data_block(struct f2fs_sb_info *sbi, struct page *page,
 }
 
 static void do_write_page(struct f2fs_sb_info *sbi, struct page *page,
-			block_t old_blkaddr, block_t *new_blkaddr,
-			struct f2fs_summary *sum, struct f2fs_io_info *fio)
+			struct f2fs_summary *sum,
+			struct f2fs_io_info *fio)
 {
 	int type = __get_segment_type(page, fio->type);
 
-	allocate_data_block(sbi, page, old_blkaddr, new_blkaddr, sum, type);
+	allocate_data_block(sbi, page, fio->blk_addr, &fio->blk_addr, sum, type);
 
 	/* writeout dirty page into bdev */
-	f2fs_submit_page_mbio(sbi, page, *new_blkaddr, fio);
+	f2fs_submit_page_mbio(sbi, page, fio);
 }
 
 void write_meta_page(struct f2fs_sb_info *sbi, struct page *page)
 {
 	struct f2fs_io_info fio = {
 		.type = META,
-		.rw = WRITE_SYNC | REQ_META | REQ_PRIO
+		.rw = WRITE_SYNC | REQ_META | REQ_PRIO,
+		.blk_addr = page->index,
 	};
 
 	set_page_writeback(page);
-	f2fs_submit_page_mbio(sbi, page, page->index, &fio);
+	f2fs_submit_page_mbio(sbi, page, &fio);
 }
 
 void write_node_page(struct f2fs_sb_info *sbi, struct page *page,
-		struct f2fs_io_info *fio,
-		unsigned int nid, block_t old_blkaddr, block_t *new_blkaddr)
+			unsigned int nid, struct f2fs_io_info *fio)
 {
 	struct f2fs_summary sum;
 	set_summary(&sum, nid, 0, 0);
-	do_write_page(sbi, page, old_blkaddr, new_blkaddr, &sum, fio);
+	do_write_page(sbi, page, &sum, fio);
 }
 
 void write_data_page(struct page *page, struct dnode_of_data *dn,
-		block_t *new_blkaddr, struct f2fs_io_info *fio)
+				struct f2fs_io_info *fio)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode);
 	struct f2fs_summary sum;
@@ -1223,14 +1223,12 @@ void write_data_page(struct page *page, struct dnode_of_data *dn,
 	f2fs_bug_on(sbi, dn->data_blkaddr == NULL_ADDR);
 	get_node_info(sbi, dn->nid, &ni);
 	set_summary(&sum, dn->nid, dn->ofs_in_node, ni.version);
-
-	do_write_page(sbi, page, dn->data_blkaddr, new_blkaddr, &sum, fio);
+	do_write_page(sbi, page, &sum, fio);
 }
 
-void rewrite_data_page(struct page *page, block_t old_blkaddr,
-					struct f2fs_io_info *fio)
+void rewrite_data_page(struct page *page, struct f2fs_io_info *fio)
 {
-	f2fs_submit_page_mbio(F2FS_P_SB(page), page, old_blkaddr, fio);
+	f2fs_submit_page_mbio(F2FS_P_SB(page), page, fio);
 }
 
 void recover_data_page(struct f2fs_sb_info *sbi,
-- 
cgit v1.2.3


From 9e4ded3f309eb5b5a9be0ca2acd26e5ea7f00914 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Wed, 17 Dec 2014 19:58:58 -0800
Subject: f2fs: activate f2fs_trace_pid

This patch activates f2fs_trace_pid.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/checkpoint.c | 3 +++
 fs/f2fs/node.c       | 2 ++
 fs/f2fs/segment.c    | 2 ++
 3 files changed, 7 insertions(+)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 0ac7c39605ed..3999933f7a74 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -20,6 +20,7 @@
 #include "f2fs.h"
 #include "node.h"
 #include "segment.h"
+#include "trace.h"
 #include <trace/events/f2fs.h>
 
 static struct kmem_cache *ino_entry_slab;
@@ -301,6 +302,7 @@ static int f2fs_set_meta_page_dirty(struct page *page)
 	if (!PageDirty(page)) {
 		__set_page_dirty_nobuffers(page);
 		inc_page_count(F2FS_P_SB(page), F2FS_DIRTY_META);
+		f2fs_trace_pid(page);
 		return 1;
 	}
 	return 0;
@@ -712,6 +714,7 @@ void update_dirty_page(struct inode *inode, struct page *page)
 		kmem_cache_free(inode_entry_slab, new);
 out:
 	SetPagePrivate(page);
+	f2fs_trace_pid(page);
 }
 
 void add_dirty_dir_inode(struct inode *inode)
diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index adc35c978306..a7cb0db2e3e8 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -19,6 +19,7 @@
 #include "f2fs.h"
 #include "node.h"
 #include "segment.h"
+#include "trace.h"
 #include <trace/events/f2fs.h>
 
 #define on_build_free_nids(nmi) mutex_is_locked(&nm_i->build_lock)
@@ -1362,6 +1363,7 @@ static int f2fs_set_node_page_dirty(struct page *page)
 		__set_page_dirty_nobuffers(page);
 		inc_page_count(F2FS_P_SB(page), F2FS_DIRTY_NODES);
 		SetPagePrivate(page);
+		f2fs_trace_pid(page);
 		return 1;
 	}
 	return 0;
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index c726f86c2ea0..b688991514fb 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -20,6 +20,7 @@
 #include "f2fs.h"
 #include "segment.h"
 #include "node.h"
+#include "trace.h"
 #include <trace/events/f2fs.h>
 
 #define __reverse_ffz(x) __reverse_ffs(~(x))
@@ -181,6 +182,7 @@ void register_inmem_page(struct inode *inode, struct page *page)
 	int err;
 
 	SetPagePrivate(page);
+	f2fs_trace_pid(page);
 
 	new = f2fs_kmem_cache_alloc(inmem_entry_slab, GFP_NOFS);
 
-- 
cgit v1.2.3


From b9a2c252071d44d4a22082611db84272be1f3b49 Mon Sep 17 00:00:00 2001
From: Changman Lee <cm224.lee@samsung.com>
Date: Wed, 24 Dec 2014 02:16:54 +0900
Subject: f2fs: add block count by in-place-update in stat info

This patch adds block count by in-place-update in stat.

Signed-off-by: Changman Lee <cm224.lee@samsung.com>
Reviewed-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/debug.c   | 4 ++++
 fs/f2fs/f2fs.h    | 6 +++++-
 fs/f2fs/segment.c | 1 +
 3 files changed, 10 insertions(+), 1 deletion(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c
index 91e8f699ab30..2b6422156ea0 100644
--- a/fs/f2fs/debug.c
+++ b/fs/f2fs/debug.c
@@ -79,6 +79,8 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 		si->segment_count[i] = sbi->segment_count[i];
 		si->block_count[i] = sbi->block_count[i];
 	}
+
+	si->inplace_count = atomic_read(&sbi->inplace_count);
 }
 
 /*
@@ -277,6 +279,7 @@ static int stat_show(struct seq_file *s, void *v)
 		for (j = 0; j < si->util_free; j++)
 			seq_putc(s, '-');
 		seq_puts(s, "]\n\n");
+		seq_printf(s, "IPU: %u blocks\n", si->inplace_count);
 		seq_printf(s, "SSR: %u blocks in %u segments\n",
 			   si->block_count[SSR], si->segment_count[SSR]);
 		seq_printf(s, "LFS: %u blocks in %u segments\n",
@@ -331,6 +334,7 @@ int f2fs_build_stats(struct f2fs_sb_info *sbi)
 
 	atomic_set(&sbi->inline_inode, 0);
 	atomic_set(&sbi->inline_dir, 0);
+	atomic_set(&sbi->inplace_count, 0);
 
 	mutex_lock(&f2fs_stat_mutex);
 	list_add_tail(&si->stat_list, &f2fs_stat_list);
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 3f07b504f6d8..3226af06e050 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -592,6 +592,7 @@ struct f2fs_sb_info {
 	struct f2fs_stat_info *stat_info;	/* FS status information */
 	unsigned int segment_count[2];		/* # of allocated segments */
 	unsigned int block_count[2];		/* # of allocated blocks */
+	atomic_t inplace_count;		/* # of inplace update */
 	int total_hit_ext, read_hit_ext;	/* extent cache hit ratio */
 	atomic_t inline_inode;			/* # of inline_data inodes */
 	atomic_t inline_dir;			/* # of inline_dentry inodes */
@@ -1523,6 +1524,7 @@ struct f2fs_stat_info {
 
 	unsigned int segment_count[2];
 	unsigned int block_count[2];
+	unsigned int inplace_count;
 	unsigned base_mem, cache_mem;
 };
 
@@ -1562,7 +1564,8 @@ static inline struct f2fs_stat_info *F2FS_STAT(struct f2fs_sb_info *sbi)
 		((sbi)->segment_count[(curseg)->alloc_type]++)
 #define stat_inc_block_count(sbi, curseg)				\
 		((sbi)->block_count[(curseg)->alloc_type]++)
-
+#define stat_inc_inplace_blocks(sbi)					\
+		(atomic_inc(&(sbi)->inplace_count))
 #define stat_inc_seg_count(sbi, type)					\
 	do {								\
 		struct f2fs_stat_info *si = F2FS_STAT(sbi);		\
@@ -1608,6 +1611,7 @@ void f2fs_destroy_root_stats(void);
 #define stat_dec_inline_dir(inode)
 #define stat_inc_seg_type(sbi, curseg)
 #define stat_inc_block_count(sbi, curseg)
+#define stat_inc_inplace_blocks(sbi)
 #define stat_inc_seg_count(si, type)
 #define stat_inc_tot_blk_count(si, blks)
 #define stat_inc_data_blk_count(si, blks)
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index b688991514fb..f995a850dfe7 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -1230,6 +1230,7 @@ void write_data_page(struct page *page, struct dnode_of_data *dn,
 
 void rewrite_data_page(struct page *page, struct f2fs_io_info *fio)
 {
+	stat_inc_inplace_blocks(F2FS_P_SB(page));
 	f2fs_submit_page_mbio(F2FS_P_SB(page), page, fio);
 }
 
-- 
cgit v1.2.3


From e1509cf294cc670cda1fedd430f0ff175c42b591 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Tue, 30 Dec 2014 22:57:55 -0800
Subject: f2fs: clean up to remove parameter

This patch uses dn->data_blkaddr as a parameter for the destination block
address.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/data.c     | 27 ++++++++++++++-------------
 fs/f2fs/f2fs.h     |  2 +-
 fs/f2fs/file.c     |  3 ++-
 fs/f2fs/inline.c   |  2 +-
 fs/f2fs/recovery.c |  3 ++-
 fs/f2fs/segment.c  |  1 +
 6 files changed, 21 insertions(+), 17 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 6308435028f6..2c0cb6617918 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -197,7 +197,7 @@ alloc_new:
  *  ->node_page
  *    update block addresses in the node page
  */
-static void __set_data_blkaddr(struct dnode_of_data *dn, block_t new_addr)
+static void __set_data_blkaddr(struct dnode_of_data *dn)
 {
 	struct f2fs_node *rn;
 	__le32 *addr_array;
@@ -210,7 +210,7 @@ static void __set_data_blkaddr(struct dnode_of_data *dn, block_t new_addr)
 
 	/* Get physical address of data block */
 	addr_array = blkaddr_in_node(rn);
-	addr_array[ofs_in_node] = cpu_to_le32(new_addr);
+	addr_array[ofs_in_node] = cpu_to_le32(dn->data_blkaddr);
 	set_page_dirty(node_page);
 }
 
@@ -225,8 +225,8 @@ int reserve_new_block(struct dnode_of_data *dn)
 
 	trace_f2fs_reserve_new_block(dn->inode, dn->nid, dn->ofs_in_node);
 
-	__set_data_blkaddr(dn, NEW_ADDR);
 	dn->data_blkaddr = NEW_ADDR;
+	__set_data_blkaddr(dn);
 	mark_inode_dirty(dn->inode);
 	sync_inode_page(dn);
 	return 0;
@@ -291,19 +291,19 @@ static int check_extent_cache(struct inode *inode, pgoff_t pgofs,
 	return 0;
 }
 
-void update_extent_cache(block_t blk_addr, struct dnode_of_data *dn)
+void update_extent_cache(struct dnode_of_data *dn)
 {
 	struct f2fs_inode_info *fi = F2FS_I(dn->inode);
 	pgoff_t fofs, start_fofs, end_fofs;
 	block_t start_blkaddr, end_blkaddr;
 	int need_update = true;
 
-	f2fs_bug_on(F2FS_I_SB(dn->inode), blk_addr == NEW_ADDR);
+	f2fs_bug_on(F2FS_I_SB(dn->inode), dn->data_blkaddr == NEW_ADDR);
 	fofs = start_bidx_of_node(ofs_of_node(dn->node_page), fi) +
 							dn->ofs_in_node;
 
 	/* Update the page address in the parent node */
-	__set_data_blkaddr(dn, blk_addr);
+	__set_data_blkaddr(dn);
 
 	if (is_inode_flag_set(fi, FI_NO_EXTENT))
 		return;
@@ -321,16 +321,16 @@ void update_extent_cache(block_t blk_addr, struct dnode_of_data *dn)
 
 	/* Initial extent */
 	if (fi->ext.len == 0) {
-		if (blk_addr != NULL_ADDR) {
+		if (dn->data_blkaddr != NULL_ADDR) {
 			fi->ext.fofs = fofs;
-			fi->ext.blk_addr = blk_addr;
+			fi->ext.blk_addr = dn->data_blkaddr;
 			fi->ext.len = 1;
 		}
 		goto end_update;
 	}
 
 	/* Front merge */
-	if (fofs == start_fofs - 1 && blk_addr == start_blkaddr - 1) {
+	if (fofs == start_fofs - 1 && dn->data_blkaddr == start_blkaddr - 1) {
 		fi->ext.fofs--;
 		fi->ext.blk_addr--;
 		fi->ext.len++;
@@ -338,7 +338,7 @@ void update_extent_cache(block_t blk_addr, struct dnode_of_data *dn)
 	}
 
 	/* Back merge */
-	if (fofs == end_fofs + 1 && blk_addr == end_blkaddr + 1) {
+	if (fofs == end_fofs + 1 && dn->data_blkaddr == end_blkaddr + 1) {
 		fi->ext.len++;
 		goto end_update;
 	}
@@ -572,8 +572,8 @@ static int __allocate_data_block(struct dnode_of_data *dn)
 	if (unlikely(!inc_valid_block_count(sbi, dn->inode, 1)))
 		return -ENOSPC;
 
-	__set_data_blkaddr(dn, NEW_ADDR);
 	dn->data_blkaddr = NEW_ADDR;
+	__set_data_blkaddr(dn);
 
 	get_node_info(sbi, dn->nid, &ni);
 	set_summary(&sum, dn->nid, dn->ofs_in_node, ni.version);
@@ -584,7 +584,8 @@ static int __allocate_data_block(struct dnode_of_data *dn)
 
 	/* direct IO doesn't use extent cache to maximize the performance */
 	set_inode_flag(F2FS_I(dn->inode), FI_NO_EXTENT);
-	update_extent_cache(new_blkaddr, dn);
+	dn->data_blkaddr = new_blkaddr;
+	update_extent_cache(dn);
 	clear_inode_flag(F2FS_I(dn->inode), FI_NO_EXTENT);
 
 	/* update i_size */
@@ -784,7 +785,7 @@ int do_write_data_page(struct page *page, struct f2fs_io_info *fio)
 		set_inode_flag(F2FS_I(inode), FI_UPDATE_WRITE);
 	} else {
 		write_data_page(page, &dn, fio);
-		update_extent_cache(fio->blk_addr, &dn);
+		update_extent_cache(&dn);
 		set_inode_flag(F2FS_I(inode), FI_APPEND_WRITE);
 	}
 out_writepage:
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index c48847e06dae..37e935a362cd 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -1478,7 +1478,7 @@ void f2fs_submit_page_mbio(struct f2fs_sb_info *, struct page *,
 						struct f2fs_io_info *);
 int reserve_new_block(struct dnode_of_data *);
 int f2fs_reserve_block(struct dnode_of_data *, pgoff_t);
-void update_extent_cache(block_t, struct dnode_of_data *);
+void update_extent_cache(struct dnode_of_data *);
 struct page *find_data_page(struct inode *, pgoff_t, bool);
 struct page *get_lock_data_page(struct inode *, pgoff_t);
 struct page *get_new_data_page(struct inode *, struct page *, pgoff_t, bool);
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index f172ddc4ecdd..5df336757d6c 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -429,7 +429,8 @@ int truncate_data_blocks_range(struct dnode_of_data *dn, int count)
 		if (blkaddr == NULL_ADDR)
 			continue;
 
-		update_extent_cache(NULL_ADDR, dn);
+		dn->data_blkaddr = NULL_ADDR;
+		update_extent_cache(dn);
 		invalidate_blocks(sbi, blkaddr);
 		nr_free++;
 	}
diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c
index 0c3f3f9b9f88..fa2aa2f7bb96 100644
--- a/fs/f2fs/inline.c
+++ b/fs/f2fs/inline.c
@@ -116,7 +116,7 @@ no_update:
 	set_page_writeback(page);
 	fio.blk_addr = dn->data_blkaddr;
 	write_data_page(page, dn, &fio);
-	update_extent_cache(fio.blk_addr, dn);
+	update_extent_cache(dn);
 	f2fs_wait_on_page_writeback(page, DATA);
 	if (dirty)
 		inode_dec_dirty_pages(dn->inode);
diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index 9160a37e1c7a..c4211a5862df 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -396,7 +396,8 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 
 			/* write dummy data page */
 			recover_data_page(sbi, NULL, &sum, src, dest);
-			update_extent_cache(dest, &dn);
+			dn.data_blkaddr = dest;
+			update_extent_cache(&dn);
 			recovered++;
 		}
 		dn.ofs_in_node++;
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index f995a850dfe7..8144a30b1567 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -1226,6 +1226,7 @@ void write_data_page(struct page *page, struct dnode_of_data *dn,
 	get_node_info(sbi, dn->nid, &ni);
 	set_summary(&sum, dn->nid, dn->ofs_in_node, ni.version);
 	do_write_page(sbi, page, &sum, fio);
+	dn->data_blkaddr = fio->blk_addr;
 }
 
 void rewrite_data_page(struct page *page, struct f2fs_io_info *fio)
-- 
cgit v1.2.3


From 38aa0889b2504bbe68e47f51cf73bf7f0a7246bd Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Mon, 5 Jan 2015 16:02:20 -0800
Subject: f2fs: align direct_io'ed data to section

This patch aligns the start block address of a file for direct io to the f2fs's
section size.

Some flash devices manage an over 4KB-sized page as a write unit, and if the
direct_io'ed data are written but not aligned to that unit, the performance can
be degraded due to the partial page copies.

Thus, since f2fs has a section that is well aligned to FTL units, we can align
the block address to the section size so that f2fs avoids this misalignment.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/data.c    |  7 +++++--
 fs/f2fs/f2fs.h    |  3 ++-
 fs/f2fs/segment.c | 27 +++++++++++++++++++--------
 3 files changed, 26 insertions(+), 11 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 3e0f5f303c97..b48b355104c3 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -564,6 +564,7 @@ static int __allocate_data_block(struct dnode_of_data *dn)
 	struct f2fs_inode_info *fi = F2FS_I(dn->inode);
 	struct f2fs_summary sum;
 	struct node_info ni;
+	int seg = CURSEG_WARM_DATA;
 	pgoff_t fofs;
 
 	if (unlikely(is_inode_flag_set(F2FS_I(dn->inode), FI_NO_ALLOC)))
@@ -574,8 +575,10 @@ static int __allocate_data_block(struct dnode_of_data *dn)
 	get_node_info(sbi, dn->nid, &ni);
 	set_summary(&sum, dn->nid, dn->ofs_in_node, ni.version);
 
-	allocate_data_block(sbi, NULL, NULL_ADDR, &dn->data_blkaddr, &sum,
-							CURSEG_WARM_DATA);
+	if (dn->ofs_in_node == 0 && dn->inode_page == dn->node_page)
+		seg = CURSEG_DIRECT_IO;
+
+	allocate_data_block(sbi, NULL, NULL_ADDR, &dn->data_blkaddr, &sum, seg);
 
 	/* direct IO doesn't use extent cache to maximize the performance */
 	__set_data_blkaddr(dn);
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 37e935a362cd..ba30218770da 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -406,7 +406,8 @@ enum {
 	CURSEG_HOT_NODE,	/* direct node blocks of directory files */
 	CURSEG_WARM_NODE,	/* direct node blocks of normal files */
 	CURSEG_COLD_NODE,	/* indirect node blocks */
-	NO_CHECK_TYPE
+	NO_CHECK_TYPE,
+	CURSEG_DIRECT_IO,	/* to use for the direct IO path */
 };
 
 struct flush_cmd {
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 8144a30b1567..31c4e5702c7d 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -1024,18 +1024,22 @@ static void allocate_segment_by_default(struct f2fs_sb_info *sbi,
 	stat_inc_seg_type(sbi, curseg);
 }
 
+static void __allocate_new_segments(struct f2fs_sb_info *sbi, int type)
+{
+	struct curseg_info *curseg = CURSEG_I(sbi, type);
+	unsigned int old_segno;
+
+	old_segno = curseg->segno;
+	SIT_I(sbi)->s_ops->allocate_segment(sbi, type, true);
+	locate_dirty_segment(sbi, old_segno);
+}
+
 void allocate_new_segments(struct f2fs_sb_info *sbi)
 {
-	struct curseg_info *curseg;
-	unsigned int old_curseg;
 	int i;
 
-	for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_DATA; i++) {
-		curseg = CURSEG_I(sbi, i);
-		old_curseg = curseg->segno;
-		SIT_I(sbi)->s_ops->allocate_segment(sbi, i, true);
-		locate_dirty_segment(sbi, old_curseg);
-	}
+	for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_DATA; i++)
+		__allocate_new_segments(sbi, i);
 }
 
 static const struct segment_allocation default_salloc_ops = {
@@ -1148,11 +1152,18 @@ void allocate_data_block(struct f2fs_sb_info *sbi, struct page *page,
 {
 	struct sit_info *sit_i = SIT_I(sbi);
 	struct curseg_info *curseg;
+	bool direct_io = (type == CURSEG_DIRECT_IO);
+
+	type = direct_io ? CURSEG_WARM_DATA : type;
 
 	curseg = CURSEG_I(sbi, type);
 
 	mutex_lock(&curseg->curseg_mutex);
 
+	/* direct_io'ed data is aligned to the segment for better performance */
+	if (direct_io && curseg->next_blkoff)
+		__allocate_new_segments(sbi, type);
+
 	*new_blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
 
 	/*
-- 
cgit v1.2.3


From 119ee9144534141822462e3e8a5ccc8dc537f712 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Thu, 29 Jan 2015 11:45:33 -0800
Subject: f2fs: split UMOUNT and FASTBOOT flags

This patch adds FASTBOOT flag into checkpoint as follows.

 - CP_UMOUNT_FLAG is set when system is umounted.
 - CP_FASTBOOT_FLAG is set when intermediate checkpoint having node summaries
   was done.

So, if you get CP_UMOUNT_FLAG from checkpoint, the system was umounted cleanly.
Instead, if there was sudden-power-off, you can get CP_FASTBOOT_FLAG or nothing.

Reviewed-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/checkpoint.c        | 19 +++++++++++++------
 fs/f2fs/f2fs.h              | 23 +++++++++++++++++++++++
 fs/f2fs/gc.c                |  3 +--
 fs/f2fs/segment.c           | 11 +++++------
 fs/f2fs/super.c             |  5 ++---
 include/linux/f2fs_fs.h     |  1 +
 include/trace/events/f2fs.h |  1 +
 7 files changed, 46 insertions(+), 17 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 22165fb1d0d1..f7cdcad31943 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -956,17 +956,24 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	ckpt->cp_pack_start_sum = cpu_to_le32(1 + cp_payload_blks +
 			orphan_blocks);
 
-	if (cpc->reason == CP_UMOUNT) {
-		set_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
+	if (__remain_node_summaries(cpc->reason))
 		ckpt->cp_pack_total_block_count = cpu_to_le32(F2FS_CP_PACKS+
 				cp_payload_blks + data_sum_blocks +
 				orphan_blocks + NR_CURSEG_NODE_TYPE);
-	} else {
-		clear_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
+	else
 		ckpt->cp_pack_total_block_count = cpu_to_le32(F2FS_CP_PACKS +
 				cp_payload_blks + data_sum_blocks +
 				orphan_blocks);
-	}
+
+	if (cpc->reason == CP_UMOUNT)
+		set_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
+	else
+		clear_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
+
+	if (cpc->reason == CP_FASTBOOT)
+		set_ckpt_flags(ckpt, CP_FASTBOOT_FLAG);
+	else
+		clear_ckpt_flags(ckpt, CP_FASTBOOT_FLAG);
 
 	if (orphan_num)
 		set_ckpt_flags(ckpt, CP_ORPHAN_PRESENT_FLAG);
@@ -1010,7 +1017,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 
 	write_data_summaries(sbi, start_blk);
 	start_blk += data_sum_blocks;
-	if (cpc->reason == CP_UMOUNT) {
+	if (__remain_node_summaries(cpc->reason)) {
 		write_node_summaries(sbi, start_blk);
 		start_blk += NR_CURSEG_NODE_TYPE;
 	}
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 5abe0836c179..8231a599a305 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -100,6 +100,7 @@ enum {
 
 enum {
 	CP_UMOUNT,
+	CP_FASTBOOT,
 	CP_SYNC,
 	CP_DISCARD,
 };
@@ -764,6 +765,28 @@ static inline void f2fs_unlock_all(struct f2fs_sb_info *sbi)
 	up_write(&sbi->cp_rwsem);
 }
 
+static inline int __get_cp_reason(struct f2fs_sb_info *sbi)
+{
+	int reason = CP_SYNC;
+
+	if (test_opt(sbi, FASTBOOT))
+		reason = CP_FASTBOOT;
+	if (is_sbi_flag_set(sbi, SBI_IS_CLOSE))
+		reason = CP_UMOUNT;
+	return reason;
+}
+
+static inline bool __remain_node_summaries(int reason)
+{
+	return (reason == CP_UMOUNT || reason == CP_FASTBOOT);
+}
+
+static inline bool __exist_node_summaries(struct f2fs_sb_info *sbi)
+{
+	return (is_set_ckpt_flags(F2FS_CKPT(sbi), CP_UMOUNT_FLAG) ||
+			is_set_ckpt_flags(F2FS_CKPT(sbi), CP_FASTBOOT_FLAG));
+}
+
 /*
  * Check whether the given nid is within node id range.
  */
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index ba89e27f394f..76adbc3641f1 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -698,8 +698,7 @@ int f2fs_gc(struct f2fs_sb_info *sbi)
 		.iroot = RADIX_TREE_INIT(GFP_NOFS),
 	};
 
-	cpc.reason = test_opt(sbi, FASTBOOT) ? CP_UMOUNT : CP_SYNC;
-
+	cpc.reason = __get_cp_reason(sbi);
 gc_more:
 	if (unlikely(!(sbi->sb->s_flags & MS_ACTIVE)))
 		goto stop;
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 31c4e5702c7d..5ea57ec153d1 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -1401,7 +1401,7 @@ static int read_normal_summaries(struct f2fs_sb_info *sbi, int type)
 		segno = le32_to_cpu(ckpt->cur_data_segno[type]);
 		blk_off = le16_to_cpu(ckpt->cur_data_blkoff[type -
 							CURSEG_HOT_DATA]);
-		if (is_set_ckpt_flags(ckpt, CP_UMOUNT_FLAG))
+		if (__exist_node_summaries(sbi))
 			blk_addr = sum_blk_addr(sbi, NR_CURSEG_TYPE, type);
 		else
 			blk_addr = sum_blk_addr(sbi, NR_CURSEG_DATA_TYPE, type);
@@ -1410,7 +1410,7 @@ static int read_normal_summaries(struct f2fs_sb_info *sbi, int type)
 							CURSEG_HOT_NODE]);
 		blk_off = le16_to_cpu(ckpt->cur_node_blkoff[type -
 							CURSEG_HOT_NODE]);
-		if (is_set_ckpt_flags(ckpt, CP_UMOUNT_FLAG))
+		if (__exist_node_summaries(sbi))
 			blk_addr = sum_blk_addr(sbi, NR_CURSEG_NODE_TYPE,
 							type - CURSEG_HOT_NODE);
 		else
@@ -1421,7 +1421,7 @@ static int read_normal_summaries(struct f2fs_sb_info *sbi, int type)
 	sum = (struct f2fs_summary_block *)page_address(new);
 
 	if (IS_NODESEG(type)) {
-		if (is_set_ckpt_flags(ckpt, CP_UMOUNT_FLAG)) {
+		if (__exist_node_summaries(sbi)) {
 			struct f2fs_summary *ns = &sum->entries[0];
 			int i;
 			for (i = 0; i < sbi->blocks_per_seg; i++, ns++) {
@@ -1470,7 +1470,7 @@ static int restore_curseg_summaries(struct f2fs_sb_info *sbi)
 		type = CURSEG_HOT_NODE;
 	}
 
-	if (is_set_ckpt_flags(F2FS_CKPT(sbi), CP_UMOUNT_FLAG))
+	if (__exist_node_summaries(sbi))
 		ra_meta_pages(sbi, sum_blk_addr(sbi, NR_CURSEG_TYPE, type),
 					NR_CURSEG_TYPE - type, META_CP);
 
@@ -1567,8 +1567,7 @@ void write_data_summaries(struct f2fs_sb_info *sbi, block_t start_blk)
 
 void write_node_summaries(struct f2fs_sb_info *sbi, block_t start_blk)
 {
-	if (is_set_ckpt_flags(F2FS_CKPT(sbi), CP_UMOUNT_FLAG))
-		write_normal_summaries(sbi, start_blk, CURSEG_HOT_NODE);
+	write_normal_summaries(sbi, start_blk, CURSEG_HOT_NODE);
 }
 
 int lookup_journal_in_cursum(struct f2fs_summary_block *sum, int type,
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 06d903b5d6c8..bfeab3c81a48 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -500,9 +500,8 @@ int f2fs_sync_fs(struct super_block *sb, int sync)
 	if (sync) {
 		struct cp_control cpc;
 
-		cpc.reason = (test_opt(sbi, FASTBOOT) ||
-					is_sbi_flag_set(sbi, SBI_IS_CLOSE)) ?
-						CP_UMOUNT : CP_SYNC;
+		cpc.reason = __get_cp_reason(sbi);
+
 		mutex_lock(&sbi->gc_mutex);
 		write_checkpoint(sbi, &cpc);
 		mutex_unlock(&sbi->gc_mutex);
diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h
index e993b0bc9abf..09805720f4bf 100644
--- a/include/linux/f2fs_fs.h
+++ b/include/linux/f2fs_fs.h
@@ -87,6 +87,7 @@ struct f2fs_super_block {
 /*
  * For checkpoint
  */
+#define CP_FASTBOOT_FLAG	0x00000020
 #define CP_FSCK_FLAG		0x00000010
 #define CP_ERROR_FLAG		0x00000008
 #define CP_COMPACT_SUM_FLAG	0x00000004
diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h
index 5e1c0292250c..69629826c2ba 100644
--- a/include/trace/events/f2fs.h
+++ b/include/trace/events/f2fs.h
@@ -72,6 +72,7 @@
 #define show_cpreason(type)						\
 	__print_symbolic(type,						\
 		{ CP_UMOUNT,	"Umount" },				\
+		{ CP_FASTBOOT,	"Fastboot" },				\
 		{ CP_SYNC,	"Sync" },				\
 		{ CP_DISCARD,	"Discard" })
 
-- 
cgit v1.2.3


From bba681cbb231920a786cd7303462fb2632af6f36 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Mon, 26 Jan 2015 17:41:23 -0800
Subject: f2fs: introduce a batched trim

This patch introduces a batched trimming feature, which submits split discard
commands.

This is to avoid long latency due to huge trim commands.
If fstrim was triggered ranging from 0 to the end of device, we should lock
all the checkpoint-related mutexes, resulting in very long latency.

Reviewed-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 Documentation/ABI/testing/sysfs-fs-f2fs |  6 ++++++
 Documentation/filesystems/f2fs.txt      |  4 ++++
 fs/f2fs/f2fs.h                          |  7 +++++++
 fs/f2fs/segment.c                       | 17 ++++++++++++-----
 fs/f2fs/super.c                         |  2 ++
 5 files changed, 31 insertions(+), 5 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/Documentation/ABI/testing/sysfs-fs-f2fs b/Documentation/ABI/testing/sysfs-fs-f2fs
index 6f9157f16725..2c4cc42006e8 100644
--- a/Documentation/ABI/testing/sysfs-fs-f2fs
+++ b/Documentation/ABI/testing/sysfs-fs-f2fs
@@ -74,3 +74,9 @@ Date:		March 2014
 Contact:	"Jaegeuk Kim" <jaegeuk.kim@samsung.com>
 Description:
 		 Controls the memory footprint used by f2fs.
+
+What:		/sys/fs/f2fs/<disk>/trim_sections
+Date:		February 2015
+Contact:	"Jaegeuk Kim" <jaegeuk@kernel.org>
+Description:
+		 Controls the trimming rate in batch mode.
diff --git a/Documentation/filesystems/f2fs.txt b/Documentation/filesystems/f2fs.txt
index 6758aa358475..dac11d7fef27 100644
--- a/Documentation/filesystems/f2fs.txt
+++ b/Documentation/filesystems/f2fs.txt
@@ -199,6 +199,10 @@ Files in /sys/fs/f2fs/<devname>
 			      checkpoint is triggered, and issued during the
 			      checkpoint. By default, it is disabled with 0.
 
+ trim_sections                This parameter controls the number of sections
+                              to be trimmed out in batch mode when FITRIM
+                              conducts. 32 sections is set by default.
+
  ipu_policy                   This parameter controls the policy of in-place
                               updates in f2fs. There are five policies:
                                0x01: F2FS_IPU_FORCE, 0x02: F2FS_IPU_SSR,
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index db1ff55de0bc..337204dfc5cd 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -105,6 +105,10 @@ enum {
 	CP_DISCARD,
 };
 
+#define DEF_BATCHED_TRIM_SECTIONS	32
+#define BATCHED_TRIM_SEGMENTS(sbi)	\
+		(SM_I(sbi)->trim_sections * (sbi)->segs_per_sec)
+
 struct cp_control {
 	int reason;
 	__u64 trim_start;
@@ -448,6 +452,9 @@ struct f2fs_sm_info {
 	int nr_discards;			/* # of discards in the list */
 	int max_discards;			/* max. discards to be issued */
 
+	/* for batched trimming */
+	unsigned int trim_sections;		/* # of sections to trim */
+
 	struct list_head sit_entry_set;	/* sit entry set list */
 
 	unsigned int ipu_policy;	/* in-place-update policy */
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 5ea57ec153d1..9f278d156d88 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -1066,14 +1066,19 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range)
 	end_segno = (end >= MAX_BLKADDR(sbi)) ? MAIN_SEGS(sbi) - 1 :
 						GET_SEGNO(sbi, end);
 	cpc.reason = CP_DISCARD;
-	cpc.trim_start = start_segno;
-	cpc.trim_end = end_segno;
 	cpc.trim_minlen = range->minlen >> sbi->log_blocksize;
 
 	/* do checkpoint to issue discard commands safely */
-	mutex_lock(&sbi->gc_mutex);
-	write_checkpoint(sbi, &cpc);
-	mutex_unlock(&sbi->gc_mutex);
+	for (; start_segno <= end_segno; start_segno = cpc.trim_end + 1) {
+		cpc.trim_start = start_segno;
+		cpc.trim_end = min_t(unsigned int, rounddown(start_segno +
+				BATCHED_TRIM_SEGMENTS(sbi),
+				sbi->segs_per_sec) - 1, end_segno);
+
+		mutex_lock(&sbi->gc_mutex);
+		write_checkpoint(sbi, &cpc);
+		mutex_unlock(&sbi->gc_mutex);
+	}
 out:
 	range->len = cpc.trimmed << sbi->log_blocksize;
 	return 0;
@@ -2127,6 +2132,8 @@ int build_segment_manager(struct f2fs_sb_info *sbi)
 	sm_info->nr_discards = 0;
 	sm_info->max_discards = 0;
 
+	sm_info->trim_sections = DEF_BATCHED_TRIM_SECTIONS;
+
 	INIT_LIST_HEAD(&sm_info->sit_entry_set);
 
 	if (test_opt(sbi, FLUSH_MERGE) && !f2fs_readonly(sbi->sb)) {
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 1e92c2ea6bc1..f2fe666a6ea9 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -195,6 +195,7 @@ F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_no_gc_sleep_time, no_gc_sleep_time);
 F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_idle, gc_idle);
 F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, reclaim_segments, rec_prefree_segments);
 F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, max_small_discards, max_discards);
+F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, batched_trim_sections, trim_sections);
 F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, ipu_policy, ipu_policy);
 F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_ipu_util, min_ipu_util);
 F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_fsync_blocks, min_fsync_blocks);
@@ -210,6 +211,7 @@ static struct attribute *f2fs_attrs[] = {
 	ATTR_LIST(gc_idle),
 	ATTR_LIST(reclaim_segments),
 	ATTR_LIST(max_small_discards),
+	ATTR_LIST(batched_trim_sections),
 	ATTR_LIST(ipu_policy),
 	ATTR_LIST(min_ipu_util),
 	ATTR_LIST(min_fsync_blocks),
-- 
cgit v1.2.3


From f7ef9b83b583640111039b30e13263b71c3a6ed5 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Mon, 9 Feb 2015 12:02:44 -0800
Subject: f2fs: introduce macros to convert bytes and blocks in f2fs

This patch adds two macros for transition between byte and block offsets.
Currently, f2fs only supports 4KB blocks, so use the default size for now.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/checkpoint.c    | 7 +++----
 fs/f2fs/file.c          | 3 +--
 fs/f2fs/segment.c       | 8 ++++----
 include/linux/f2fs_fs.h | 4 ++++
 4 files changed, 12 insertions(+), 10 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index 31a715b5fd5c..58d88df0d632 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -981,15 +981,14 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	/* write out checkpoint buffer at block 0 */
 	cp_page = grab_meta_page(sbi, start_blk++);
 	kaddr = page_address(cp_page);
-	memcpy(kaddr, ckpt, (1 << sbi->log_blocksize));
+	memcpy(kaddr, ckpt, F2FS_BLKSIZE);
 	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));
+		memcpy(kaddr, (char *)ckpt + i * F2FS_BLKSIZE, F2FS_BLKSIZE);
 		set_page_dirty(cp_page);
 		f2fs_put_page(cp_page, 1);
 	}
@@ -1009,7 +1008,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	/* writeout checkpoint block */
 	cp_page = grab_meta_page(sbi, start_blk);
 	kaddr = page_address(cp_page);
-	memcpy(kaddr, ckpt, (1 << sbi->log_blocksize));
+	memcpy(kaddr, ckpt, F2FS_BLKSIZE);
 	set_page_dirty(cp_page);
 	f2fs_put_page(cp_page, 1);
 
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 7188a2ac64b5..f3b007540a48 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -491,8 +491,7 @@ int truncate_blocks(struct inode *inode, u64 from, bool lock)
 
 	trace_f2fs_truncate_blocks_enter(inode, from);
 
-	free_from = (pgoff_t)
-		((from + blocksize - 1) >> (sbi->log_blocksize));
+	free_from = (pgoff_t)F2FS_BYTES_TO_BLK(from + blocksize - 1);
 
 	if (lock)
 		f2fs_lock_op(sbi);
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 9f278d156d88..877a272a8146 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -1048,8 +1048,8 @@ static const struct segment_allocation default_salloc_ops = {
 
 int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range)
 {
-	__u64 start = range->start >> sbi->log_blocksize;
-	__u64 end = start + (range->len >> sbi->log_blocksize) - 1;
+	__u64 start = F2FS_BYTES_TO_BLK(range->start);
+	__u64 end = start + F2FS_BYTES_TO_BLK(range->len) - 1;
 	unsigned int start_segno, end_segno;
 	struct cp_control cpc;
 
@@ -1066,7 +1066,7 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range)
 	end_segno = (end >= MAX_BLKADDR(sbi)) ? MAIN_SEGS(sbi) - 1 :
 						GET_SEGNO(sbi, end);
 	cpc.reason = CP_DISCARD;
-	cpc.trim_minlen = range->minlen >> sbi->log_blocksize;
+	cpc.trim_minlen = F2FS_BYTES_TO_BLK(range->minlen);
 
 	/* do checkpoint to issue discard commands safely */
 	for (; start_segno <= end_segno; start_segno = cpc.trim_end + 1) {
@@ -1080,7 +1080,7 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range)
 		mutex_unlock(&sbi->gc_mutex);
 	}
 out:
-	range->len = cpc.trimmed << sbi->log_blocksize;
+	range->len = F2FS_BLK_TO_BYTES(cpc.trimmed);
 	return 0;
 }
 
diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h
index 09805720f4bf..a23556c32703 100644
--- a/include/linux/f2fs_fs.h
+++ b/include/linux/f2fs_fs.h
@@ -19,12 +19,16 @@
 #define F2FS_MAX_LOG_SECTOR_SIZE	12	/* 12 bits for 4096 bytes */
 #define F2FS_LOG_SECTORS_PER_BLOCK	3	/* log number for sector/blk */
 #define F2FS_BLKSIZE			4096	/* support only 4KB block */
+#define F2FS_BLKSIZE_BITS		12	/* bits for F2FS_BLKSIZE */
 #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 */
 
+#define F2FS_BYTES_TO_BLK(bytes)	((bytes) >> F2FS_BLKSIZE_BITS)
+#define F2FS_BLK_TO_BYTES(blk)		((blk) << F2FS_BLKSIZE_BITS)
+
 /* 0, 1(node nid), 2(meta nid) are reserved node id */
 #define F2FS_RESERVED_NODE_NUM		3
 
-- 
cgit v1.2.3


From 60a3b782b1aaf6e5f8c4f92e99302c48a26d475b Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Tue, 10 Feb 2015 16:44:29 -0800
Subject: f2fs: avoid variable length array

Instead of using variable length array, this patch let preallocate memory for
them.

Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/debug.c   |  1 +
 fs/f2fs/segment.c | 10 ++++++++--
 fs/f2fs/segment.h |  1 +
 3 files changed, 10 insertions(+), 2 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c
index ac2bd8e7eca9..e671373cc8ab 100644
--- a/fs/f2fs/debug.c
+++ b/fs/f2fs/debug.c
@@ -142,6 +142,7 @@ static void update_mem_info(struct f2fs_sb_info *sbi)
 	si->base_mem += MAIN_SEGS(sbi) * sizeof(struct seg_entry);
 	si->base_mem += f2fs_bitmap_size(MAIN_SEGS(sbi));
 	si->base_mem += 2 * SIT_VBLOCK_MAP_SIZE * MAIN_SEGS(sbi);
+	si->base_mem += SIT_VBLOCK_MAP_SIZE;
 	if (sbi->segs_per_sec > 1)
 		si->base_mem += MAIN_SECS(sbi) * sizeof(struct sec_entry);
 	si->base_mem += __bitmap_size(sbi, SIT_BITMAP);
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 877a272a8146..c9d314f44568 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -505,7 +505,7 @@ static void add_discard_addrs(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	struct seg_entry *se = get_seg_entry(sbi, cpc->trim_start);
 	unsigned long *cur_map = (unsigned long *)se->cur_valid_map;
 	unsigned long *ckpt_map = (unsigned long *)se->ckpt_valid_map;
-	unsigned long dmap[entries];
+	unsigned long *dmap = SIT_I(sbi)->tmp_map;
 	unsigned int start = 0, end = -1;
 	bool force = (cpc->reason == CP_DISCARD);
 	int i;
@@ -924,7 +924,7 @@ static void __next_free_blkoff(struct f2fs_sb_info *sbi,
 {
 	struct seg_entry *se = get_seg_entry(sbi, seg->segno);
 	int entries = SIT_VBLOCK_MAP_SIZE / sizeof(unsigned long);
-	unsigned long target_map[entries];
+	unsigned long *target_map = SIT_I(sbi)->tmp_map;
 	unsigned long *ckpt_map = (unsigned long *)se->ckpt_valid_map;
 	unsigned long *cur_map = (unsigned long *)se->cur_valid_map;
 	int i, pos;
@@ -1855,6 +1855,10 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
 			return -ENOMEM;
 	}
 
+	sit_i->tmp_map = kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
+	if (!sit_i->tmp_map)
+		return -ENOMEM;
+
 	if (sbi->segs_per_sec > 1) {
 		sit_i->sec_entries = vzalloc(MAIN_SECS(sbi) *
 					sizeof(struct sec_entry));
@@ -2236,6 +2240,8 @@ static void destroy_sit_info(struct f2fs_sb_info *sbi)
 			kfree(sit_i->sentries[start].ckpt_valid_map);
 		}
 	}
+	kfree(sit_i->tmp_map);
+
 	vfree(sit_i->sentries);
 	vfree(sit_i->sec_entries);
 	kfree(sit_i->dirty_sentries_bitmap);
diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index 421d5794b0c8..ba858ca01445 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -189,6 +189,7 @@ struct sit_info {
 	char *sit_bitmap;		/* SIT bitmap pointer */
 	unsigned int bitmap_size;	/* SIT bitmap size */
 
+	unsigned long *tmp_map;			/* bitmap for temporal use */
 	unsigned long *dirty_sentries_bitmap;	/* bitmap for dirty sentries */
 	unsigned int dirty_sentries;		/* # of dirty sentries */
 	unsigned int sents_per_block;		/* # of SIT entries per block */
-- 
cgit v1.2.3


From 1a118ccfd60fc78e64c0a3ab9e85075545839d6e Mon Sep 17 00:00:00 2001
From: Chao Yu <chao2.yu@samsung.com>
Date: Wed, 11 Feb 2015 18:20:38 +0800
Subject: f2fs: use spinlock for segmap_lock instead of rwlock

rwlock can provide better concurrency when there are much more readers than
writers because readers can hold the rwlock simultaneously.

But now, for segmap_lock rwlock in struct free_segmap_info, there is only one
reader 'mount' from below call path:
->f2fs_fill_super
  ->build_segment_manager
    ->build_dirty_segmap
      ->init_dirty_segmap
        ->find_next_inuse
          read_lock
          ...
          read_unlock

Now that our concurrency can not be improved since there is no other reader for
this lock, we do not need to use rwlock_t type for segmap_lock, let's replace it
with spinlock_t type.

Signed-off-by: Chao Yu <chao2.yu@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c |  6 +++---
 fs/f2fs/segment.h | 18 +++++++++---------
 2 files changed, 12 insertions(+), 12 deletions(-)

(limited to 'fs/f2fs/segment.c')

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index c9d314f44568..daee4ab913da 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -800,7 +800,7 @@ static void get_new_segment(struct f2fs_sb_info *sbi,
 	int go_left = 0;
 	int i;
 
-	write_lock(&free_i->segmap_lock);
+	spin_lock(&free_i->segmap_lock);
 
 	if (!new_sec && ((*newseg + 1) % sbi->segs_per_sec)) {
 		segno = find_next_zero_bit(free_i->free_segmap,
@@ -873,7 +873,7 @@ got_it:
 	f2fs_bug_on(sbi, test_bit(segno, free_i->free_segmap));
 	__set_inuse(sbi, segno);
 	*newseg = segno;
-	write_unlock(&free_i->segmap_lock);
+	spin_unlock(&free_i->segmap_lock);
 }
 
 static void reset_curseg(struct f2fs_sb_info *sbi, int type, int modified)
@@ -1923,7 +1923,7 @@ static int build_free_segmap(struct f2fs_sb_info *sbi)
 	free_i->start_segno = GET_SEGNO_FROM_SEG0(sbi, MAIN_BLKADDR(sbi));
 	free_i->free_segments = 0;
 	free_i->free_sections = 0;
-	rwlock_init(&free_i->segmap_lock);
+	spin_lock_init(&free_i->segmap_lock);
 	return 0;
 }
 
diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index ba858ca01445..7fd35111cf62 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -208,7 +208,7 @@ struct free_segmap_info {
 	unsigned int start_segno;	/* start segment number logically */
 	unsigned int free_segments;	/* # of free segments */
 	unsigned int free_sections;	/* # of free sections */
-	rwlock_t segmap_lock;		/* free segmap lock */
+	spinlock_t segmap_lock;		/* free segmap lock */
 	unsigned long *free_segmap;	/* free segment bitmap */
 	unsigned long *free_secmap;	/* free section bitmap */
 };
@@ -319,9 +319,9 @@ static inline unsigned int find_next_inuse(struct free_segmap_info *free_i,
 		unsigned int max, unsigned int segno)
 {
 	unsigned int ret;
-	read_lock(&free_i->segmap_lock);
+	spin_lock(&free_i->segmap_lock);
 	ret = find_next_bit(free_i->free_segmap, max, segno);
-	read_unlock(&free_i->segmap_lock);
+	spin_unlock(&free_i->segmap_lock);
 	return ret;
 }
 
@@ -332,7 +332,7 @@ static inline void __set_free(struct f2fs_sb_info *sbi, unsigned int segno)
 	unsigned int start_segno = secno * sbi->segs_per_sec;
 	unsigned int next;
 
-	write_lock(&free_i->segmap_lock);
+	spin_lock(&free_i->segmap_lock);
 	clear_bit(segno, free_i->free_segmap);
 	free_i->free_segments++;
 
@@ -341,7 +341,7 @@ static inline void __set_free(struct f2fs_sb_info *sbi, unsigned int segno)
 		clear_bit(secno, free_i->free_secmap);
 		free_i->free_sections++;
 	}
-	write_unlock(&free_i->segmap_lock);
+	spin_unlock(&free_i->segmap_lock);
 }
 
 static inline void __set_inuse(struct f2fs_sb_info *sbi,
@@ -363,7 +363,7 @@ static inline void __set_test_and_free(struct f2fs_sb_info *sbi,
 	unsigned int start_segno = secno * sbi->segs_per_sec;
 	unsigned int next;
 
-	write_lock(&free_i->segmap_lock);
+	spin_lock(&free_i->segmap_lock);
 	if (test_and_clear_bit(segno, free_i->free_segmap)) {
 		free_i->free_segments++;
 
@@ -374,7 +374,7 @@ static inline void __set_test_and_free(struct f2fs_sb_info *sbi,
 				free_i->free_sections++;
 		}
 	}
-	write_unlock(&free_i->segmap_lock);
+	spin_unlock(&free_i->segmap_lock);
 }
 
 static inline void __set_test_and_inuse(struct f2fs_sb_info *sbi,
@@ -382,13 +382,13 @@ static inline void __set_test_and_inuse(struct f2fs_sb_info *sbi,
 {
 	struct free_segmap_info *free_i = FREE_I(sbi);
 	unsigned int secno = segno / sbi->segs_per_sec;
-	write_lock(&free_i->segmap_lock);
+	spin_lock(&free_i->segmap_lock);
 	if (!test_and_set_bit(segno, free_i->free_segmap)) {
 		free_i->free_segments--;
 		if (!test_and_set_bit(secno, free_i->free_secmap))
 			free_i->free_sections--;
 	}
-	write_unlock(&free_i->segmap_lock);
+	spin_unlock(&free_i->segmap_lock);
 }
 
 static inline void get_sit_bitmap(struct f2fs_sb_info *sbi,
-- 
cgit v1.2.3