f2fs: don't wait writeback for datas during checkpoint

Normally, while committing checkpoint, we will wait on all pages to be
writebacked no matter the page is data or metadata, so in scenario where
there are lots of data IO being submitted with metadata, we may suffer
long latency for waiting writeback during checkpoint.

Indeed, we only care about persistence for pages with metadata, but not
pages with data, as file system consistent are only related to metadate,
so in order to avoid encountering long latency in above scenario, let's
recognize and reference metadata in submitted IOs, wait writeback only
for metadatas.

Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>

fs/f2fs/checkpoint.c

@@ -1002,7 +1002,7 @@ static void wait_on_all_pages_writeback(struct f2fs_sb_info *sbi)
 	for (;;) {
 		prepare_to_wait(&sbi->cp_wait, &wait, TASK_UNINTERRUPTIBLE);
 
-		if (!atomic_read(&sbi->nr_wb_bios))
+		if (!get_pages(sbi, F2FS_WB_CP_DATA))
 			break;
 
 		io_schedule_timeout(5*HZ);

fs/f2fs/data.c

@@ -29,6 +29,26 @@
 #include "trace.h"
 #include <trace/events/f2fs.h>
 
+static bool __is_cp_guaranteed(struct page *page)
+{
+	struct address_space *mapping = page->mapping;
+	struct inode *inode;
+	struct f2fs_sb_info *sbi;
+
+	if (!mapping)
+		return false;
+
+	inode = mapping->host;
+	sbi = F2FS_I_SB(inode);
+
+	if (inode->i_ino == F2FS_META_INO(sbi) ||
+			inode->i_ino ==  F2FS_NODE_INO(sbi) ||
+			S_ISDIR(inode->i_mode) ||
+			is_cold_data(page))
+		return true;
+	return false;
+}
+
 static void f2fs_read_end_io(struct bio *bio)
 {
 	struct bio_vec *bvec;
@@ -71,6 +91,7 @@ static void f2fs_write_end_io(struct bio *bio)
 
 	bio_for_each_segment_all(bvec, bio, i) {
 		struct page *page = bvec->bv_page;
+		enum count_type type = WB_DATA_TYPE(page);
 
 		fscrypt_pullback_bio_page(&page, true);
 
@@ -78,9 +99,11 @@ static void f2fs_write_end_io(struct bio *bio)
 			mapping_set_error(page->mapping, -EIO);
 			f2fs_stop_checkpoint(sbi, true);
 		}
+		dec_page_count(sbi, type);
+		clear_cold_data(page);
 		end_page_writeback(page);
 	}
-	if (atomic_dec_and_test(&sbi->nr_wb_bios) &&
+	if (!get_pages(sbi, F2FS_WB_CP_DATA) &&
 				wq_has_sleeper(&sbi->cp_wait))
 		wake_up(&sbi->cp_wait);
 
@@ -148,7 +171,6 @@ static inline void __submit_bio(struct f2fs_sb_info *sbi,
 				struct bio *bio, enum page_type type)
 {
 	if (!is_read_io(bio_op(bio))) {
-		atomic_inc(&sbi->nr_wb_bios);
 		if (f2fs_sb_mounted_blkzoned(sbi->sb) &&
 			current->plug && (type == DATA || type == NODE))
 			blk_finish_plug(current->plug);
@@ -309,6 +331,11 @@ void f2fs_submit_page_mbio(struct f2fs_io_info *fio)
 		verify_block_addr(sbi, fio->old_blkaddr);
 	verify_block_addr(sbi, fio->new_blkaddr);
 
+	bio_page = fio->encrypted_page ? fio->encrypted_page : fio->page;
+
+	if (!is_read)
+		inc_page_count(sbi, WB_DATA_TYPE(bio_page));
+
 	down_write(&io->io_rwsem);
 
 	if (io->bio && (io->last_block_in_bio != fio->new_blkaddr - 1 ||
@@ -322,8 +349,6 @@ alloc_new:
 		io->fio = *fio;
 	}
 
-	bio_page = fio->encrypted_page ? fio->encrypted_page : fio->page;
-
 	if (bio_add_page(io->bio, bio_page, PAGE_SIZE, 0) <
 							PAGE_SIZE) {
 		__submit_merged_bio(io);
@@ -1339,7 +1364,6 @@ done:
 	if (err && err != -ENOENT)
 		goto redirty_out;
 
-	clear_cold_data(page);
 out:
 	inode_dec_dirty_pages(inode);
 	if (err)
@@ -1742,7 +1766,6 @@ static int f2fs_write_end(struct file *file,
 		goto unlock_out;
 
 	set_page_dirty(page);
-	clear_cold_data(page);
 
 	if (pos + copied > i_size_read(inode))
 		f2fs_i_size_write(inode, pos + copied);

fs/f2fs/debug.c

@@ -50,7 +50,8 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	si->ndirty_files = sbi->ndirty_inode[FILE_INODE];
 	si->ndirty_all = sbi->ndirty_inode[DIRTY_META];
 	si->inmem_pages = get_pages(sbi, F2FS_INMEM_PAGES);
-	si->wb_bios = atomic_read(&sbi->nr_wb_bios);
+	si->nr_wb_cp_data = get_pages(sbi, F2FS_WB_CP_DATA);
+	si->nr_wb_data = get_pages(sbi, F2FS_WB_DATA);
 	si->total_count = (int)sbi->user_block_count / sbi->blocks_per_seg;
 	si->rsvd_segs = reserved_segments(sbi);
 	si->overp_segs = overprovision_segments(sbi);
@@ -313,8 +314,8 @@ static int stat_show(struct seq_file *s, void *v)
 		seq_printf(s, "  - Inner Struct Count: tree: %d(%d), node: %d\n",
 				si->ext_tree, si->zombie_tree, si->ext_node);
 		seq_puts(s, "\nBalancing F2FS Async:\n");
-		seq_printf(s, "  - inmem: %4d, wb_bios: %4d\n",
-			   si->inmem_pages, si->wb_bios);
+		seq_printf(s, "  - inmem: %4d, wb_cp_data: %4d, wb_data: %4d\n",
+			   si->inmem_pages, si->nr_wb_cp_data, si->nr_wb_data);
 		seq_printf(s, "  - nodes: %4d in %4d\n",
 			   si->ndirty_node, si->node_pages);
 		seq_printf(s, "  - dents: %4d in dirs:%4d (%4d)\n",

fs/f2fs/f2fs.h

@@ -654,6 +654,7 @@ struct f2fs_sm_info {
  * f2fs monitors the number of several block types such as on-writeback,
  * dirty dentry blocks, dirty node blocks, and dirty meta blocks.
  */
+#define WB_DATA_TYPE(p)	(__is_cp_guaranteed(p) ? F2FS_WB_CP_DATA : F2FS_WB_DATA)
 enum count_type {
 	F2FS_DIRTY_DENTS,
 	F2FS_DIRTY_DATA,
@@ -661,6 +662,8 @@ enum count_type {
 	F2FS_DIRTY_META,
 	F2FS_INMEM_PAGES,
 	F2FS_DIRTY_IMETA,
+	F2FS_WB_CP_DATA,
+	F2FS_WB_DATA,
 	NR_COUNT_TYPE,
 };
 
@@ -840,7 +843,6 @@ struct f2fs_sb_info {
 	block_t discard_blks;			/* discard command candidats */
 	block_t last_valid_block_count;		/* for recovery */
 	u32 s_next_generation;			/* for NFS support */
-	atomic_t nr_wb_bios;			/* # of writeback bios */
 
 	/* # of pages, see count_type */
 	atomic_t nr_pages[NR_COUNT_TYPE];
@@ -1254,7 +1256,8 @@ static inline void inc_page_count(struct f2fs_sb_info *sbi, int count_type)
 {
 	atomic_inc(&sbi->nr_pages[count_type]);
 
-	if (count_type == F2FS_DIRTY_DATA || count_type == F2FS_INMEM_PAGES)
+	if (count_type == F2FS_DIRTY_DATA || count_type == F2FS_INMEM_PAGES ||
+		count_type == F2FS_WB_CP_DATA || count_type == F2FS_WB_DATA)
 		return;
 
 	set_sbi_flag(sbi, SBI_IS_DIRTY);
@@ -2209,7 +2212,7 @@ struct f2fs_stat_info {
 	unsigned int ndirty_dirs, ndirty_files, ndirty_all;
 	int nats, dirty_nats, sits, dirty_sits, free_nids, alloc_nids;
 	int total_count, utilization;
-	int bg_gc, wb_bios;
+	int bg_gc, nr_wb_cp_data, nr_wb_data;
 	int inline_xattr, inline_inode, inline_dir, orphans;
 	unsigned int valid_count, valid_node_count, valid_inode_count, discard_blks;
 	unsigned int bimodal, avg_vblocks;

fs/f2fs/file.c

@@ -94,8 +94,6 @@ mapped:
 	if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode))
 		f2fs_wait_on_encrypted_page_writeback(sbi, dn.data_blkaddr);
 
-	/* if gced page is attached, don't write to cold segment */
-	clear_cold_data(page);
 out:
 	sb_end_pagefault(inode->i_sb);
 	f2fs_update_time(sbi, REQ_TIME);

fs/f2fs/gc.c

@@ -693,8 +693,6 @@ retry:
 			congestion_wait(BLK_RW_ASYNC, HZ/50);
 			goto retry;
 		}
-
-		clear_cold_data(page);
 	}
 out:
 	f2fs_put_page(page, 1);

fs/f2fs/segment.c

@@ -290,7 +290,6 @@ static int __commit_inmem_pages(struct inode *inode,
 			/* record old blkaddr for revoking */
 			cur->old_addr = fio.old_blkaddr;
 
-			clear_cold_data(page);
 			submit_bio = true;
 		}
 		unlock_page(page);