1 files changed, 123 insertions, 63 deletions

diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index ea5b93b689cd..d5fb426e0747 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -35,6 +35,10 @@ static int gc_thread_func(void *data)
 	wait_queue_head_t *wq = &sbi->gc_thread->gc_wait_queue_head;
 	wait_queue_head_t *fggc_wq = &sbi->gc_thread->fggc_wq;
 	unsigned int wait_ms;
+	struct f2fs_gc_control gc_control = {
+		.victim_segno = NULL_SEGNO,
+		.should_migrate_blocks = false,
+		.err_gc_skipped = false };
 
 	wait_ms = gc_th->min_sleep_time;
 
@@ -141,8 +145,12 @@ do_gc:
 		if (foreground)
 			sync_mode = false;
 
+		gc_control.init_gc_type = sync_mode ? FG_GC : BG_GC;
+		gc_control.no_bg_gc = foreground;
+		gc_control.nr_free_secs = foreground ? 1 : 0;
+
 		/* if return value is not zero, no victim was selected */
-		if (f2fs_gc(sbi, sync_mode, !foreground, false, NULL_SEGNO))
+		if (f2fs_gc(sbi, &gc_control))
 			wait_ms = gc_th->no_gc_sleep_time;
 
 		if (foreground)
@@ -646,6 +654,54 @@ static void release_victim_entry(struct f2fs_sb_info *sbi)
 	f2fs_bug_on(sbi, !list_empty(&am->victim_list));
 }
 
+static bool f2fs_pin_section(struct f2fs_sb_info *sbi, unsigned int segno)
+{
+	struct dirty_seglist_info *dirty_i = DIRTY_I(sbi);
+	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
+
+	if (!dirty_i->enable_pin_section)
+		return false;
+	if (!test_and_set_bit(secno, dirty_i->pinned_secmap))
+		dirty_i->pinned_secmap_cnt++;
+	return true;
+}
+
+static bool f2fs_pinned_section_exists(struct dirty_seglist_info *dirty_i)
+{
+	return dirty_i->pinned_secmap_cnt;
+}
+
+static bool f2fs_section_is_pinned(struct dirty_seglist_info *dirty_i,
+						unsigned int secno)
+{
+	return dirty_i->enable_pin_section &&
+		f2fs_pinned_section_exists(dirty_i) &&
+		test_bit(secno, dirty_i->pinned_secmap);
+}
+
+static void f2fs_unpin_all_sections(struct f2fs_sb_info *sbi, bool enable)
+{
+	unsigned int bitmap_size = f2fs_bitmap_size(MAIN_SECS(sbi));
+
+	if (f2fs_pinned_section_exists(DIRTY_I(sbi))) {
+		memset(DIRTY_I(sbi)->pinned_secmap, 0, bitmap_size);
+		DIRTY_I(sbi)->pinned_secmap_cnt = 0;
+	}
+	DIRTY_I(sbi)->enable_pin_section = enable;
+}
+
+static int f2fs_gc_pinned_control(struct inode *inode, int gc_type,
+							unsigned int segno)
+{
+	if (!f2fs_is_pinned_file(inode))
+		return 0;
+	if (gc_type != FG_GC)
+		return -EBUSY;
+	if (!f2fs_pin_section(F2FS_I_SB(inode), segno))
+		f2fs_pin_file_control(inode, true);
+	return -EAGAIN;
+}
+
 /*
  * This function is called from two paths.
  * One is garbage collection and the other is SSR segment selection.
@@ -787,6 +843,9 @@ retry:
 		if (gc_type == BG_GC && test_bit(secno, dirty_i->victim_secmap))
 			goto next;
 
+		if (gc_type == FG_GC && f2fs_section_is_pinned(dirty_i, secno))
+			goto next;
+
 		if (is_atgc) {
 			add_victim_entry(sbi, &p, segno);
 			goto next;
@@ -1194,18 +1253,9 @@ static int move_data_block(struct inode *inode, block_t bidx,
 		goto out;
 	}
 
-	if (f2fs_is_atomic_file(inode)) {
-		F2FS_I(inode)->i_gc_failures[GC_FAILURE_ATOMIC]++;
-		F2FS_I_SB(inode)->skipped_atomic_files[gc_type]++;
-		err = -EAGAIN;
-		goto out;
-	}
-
-	if (f2fs_is_pinned_file(inode)) {
-		f2fs_pin_file_control(inode, true);
-		err = -EAGAIN;
+	err = f2fs_gc_pinned_control(inode, gc_type, segno);
+	if (err)
 		goto out;
-	}
 
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
 	err = f2fs_get_dnode_of_data(&dn, bidx, LOOKUP_NODE);
@@ -1344,18 +1394,9 @@ static int move_data_page(struct inode *inode, block_t bidx, int gc_type,
 		goto out;
 	}
 
-	if (f2fs_is_atomic_file(inode)) {
-		F2FS_I(inode)->i_gc_failures[GC_FAILURE_ATOMIC]++;
-		F2FS_I_SB(inode)->skipped_atomic_files[gc_type]++;
-		err = -EAGAIN;
-		goto out;
-	}
-	if (f2fs_is_pinned_file(inode)) {
-		if (gc_type == FG_GC)
-			f2fs_pin_file_control(inode, true);
-		err = -EAGAIN;
+	err = f2fs_gc_pinned_control(inode, gc_type, segno);
+	if (err)
 		goto out;
-	}
 
 	if (gc_type == BG_GC) {
 		if (PageWriteback(page)) {
@@ -1475,11 +1516,19 @@ next_step:
 		ofs_in_node = le16_to_cpu(entry->ofs_in_node);
 
 		if (phase == 3) {
+			int err;
+
 			inode = f2fs_iget(sb, dni.ino);
 			if (IS_ERR(inode) || is_bad_inode(inode) ||
 					special_file(inode->i_mode))
 				continue;
 
+			err = f2fs_gc_pinned_control(inode, gc_type, segno);
+			if (err == -EAGAIN) {
+				iput(inode);
+				return submitted;
+			}
+
 			if (!f2fs_down_write_trylock(
 				&F2FS_I(inode)->i_gc_rwsem[WRITE])) {
 				iput(inode);
@@ -1699,23 +1748,21 @@ skip:
 	return seg_freed;
 }
 
-int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
-			bool background, bool force, unsigned int segno)
+int f2fs_gc(struct f2fs_sb_info *sbi, struct f2fs_gc_control *gc_control)
 {
-	int gc_type = sync ? FG_GC : BG_GC;
+	int gc_type = gc_control->init_gc_type;
+	unsigned int segno = gc_control->victim_segno;
 	int sec_freed = 0, seg_freed = 0, total_freed = 0;
 	int ret = 0;
 	struct cp_control cpc;
-	unsigned int init_segno = segno;
 	struct gc_inode_list gc_list = {
 		.ilist = LIST_HEAD_INIT(gc_list.ilist),
 		.iroot = RADIX_TREE_INIT(gc_list.iroot, GFP_NOFS),
 	};
-	unsigned long long last_skipped = sbi->skipped_atomic_files[FG_GC];
-	unsigned long long first_skipped;
 	unsigned int skipped_round = 0, round = 0;
 
-	trace_f2fs_gc_begin(sbi->sb, sync, background,
+	trace_f2fs_gc_begin(sbi->sb, gc_type, gc_control->no_bg_gc,
+				gc_control->nr_free_secs,
 				get_pages(sbi, F2FS_DIRTY_NODES),
 				get_pages(sbi, F2FS_DIRTY_DENTS),
 				get_pages(sbi, F2FS_DIRTY_IMETA),
@@ -1726,7 +1773,6 @@ int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
 
 	cpc.reason = __get_cp_reason(sbi);
 	sbi->skipped_gc_rwsem = 0;
-	first_skipped = last_skipped;
 gc_more:
 	if (unlikely(!(sbi->sb->s_flags & SB_ACTIVE))) {
 		ret = -EINVAL;
@@ -1743,8 +1789,7 @@ gc_more:
 		 * threshold, we can make them free by checkpoint. Then, we
 		 * secure free segments which doesn't need fggc any more.
 		 */
-		if (prefree_segments(sbi) &&
-				!is_sbi_flag_set(sbi, SBI_CP_DISABLED)) {
+		if (prefree_segments(sbi)) {
 			ret = f2fs_write_checkpoint(sbi, &cpc);
 			if (ret)
 				goto stop;
@@ -1754,54 +1799,69 @@ gc_more:
 	}
 
 	/* f2fs_balance_fs doesn't need to do BG_GC in critical path. */
-	if (gc_type == BG_GC && !background) {
+	if (gc_type == BG_GC && gc_control->no_bg_gc) {
 		ret = -EINVAL;
 		goto stop;
 	}
+retry:
 	ret = __get_victim(sbi, &segno, gc_type);
-	if (ret)
+	if (ret) {
+		/* allow to search victim from sections has pinned data */
+		if (ret == -ENODATA && gc_type == FG_GC &&
+				f2fs_pinned_section_exists(DIRTY_I(sbi))) {
+			f2fs_unpin_all_sections(sbi, false);
+			goto retry;
+		}
 		goto stop;
+	}
 
-	seg_freed = do_garbage_collect(sbi, segno, &gc_list, gc_type, force);
-	if (gc_type == FG_GC &&
-		seg_freed == f2fs_usable_segs_in_sec(sbi, segno))
-		sec_freed++;
+	seg_freed = do_garbage_collect(sbi, segno, &gc_list, gc_type,
+				gc_control->should_migrate_blocks);
 	total_freed += seg_freed;
 
-	if (gc_type == FG_GC) {
-		if (sbi->skipped_atomic_files[FG_GC] > last_skipped ||
-						sbi->skipped_gc_rwsem)
-			skipped_round++;
-		last_skipped = sbi->skipped_atomic_files[FG_GC];
-		round++;
-	}
+	if (seg_freed == f2fs_usable_segs_in_sec(sbi, segno))
+		sec_freed++;
 
 	if (gc_type == FG_GC)
 		sbi->cur_victim_sec = NULL_SEGNO;
 
-	if (sync)
+	if (gc_control->init_gc_type == FG_GC ||
+	    !has_not_enough_free_secs(sbi,
+				(gc_type == FG_GC) ? sec_freed : 0, 0)) {
+		if (gc_type == FG_GC && sec_freed < gc_control->nr_free_secs)
+			goto go_gc_more;
 		goto stop;
+	}
 
-	if (has_not_enough_free_secs(sbi, sec_freed, 0)) {
-		if (skipped_round <= MAX_SKIP_GC_COUNT ||
-					skipped_round * 2 < round) {
-			segno = NULL_SEGNO;
-			goto gc_more;
+	/* FG_GC stops GC by skip_count */
+	if (gc_type == FG_GC) {
+		if (sbi->skipped_gc_rwsem)
+			skipped_round++;
+		round++;
+		if (skipped_round > MAX_SKIP_GC_COUNT &&
+				skipped_round * 2 >= round) {
+			ret = f2fs_write_checkpoint(sbi, &cpc);
+			goto stop;
 		}
+	}
 
-		if (first_skipped < last_skipped &&
-				(last_skipped - first_skipped) >
-						sbi->skipped_gc_rwsem) {
-			f2fs_drop_inmem_pages_all(sbi, true);
-			segno = NULL_SEGNO;
-			goto gc_more;
-		}
-		if (gc_type == FG_GC && !is_sbi_flag_set(sbi, SBI_CP_DISABLED))
-			ret = f2fs_write_checkpoint(sbi, &cpc);
+	/* Write checkpoint to reclaim prefree segments */
+	if (free_sections(sbi) < NR_CURSEG_PERSIST_TYPE &&
+				prefree_segments(sbi)) {
+		ret = f2fs_write_checkpoint(sbi, &cpc);
+		if (ret)
+			goto stop;
 	}
+go_gc_more:
+	segno = NULL_SEGNO;
+	goto gc_more;
+
 stop:
 	SIT_I(sbi)->last_victim[ALLOC_NEXT] = 0;
-	SIT_I(sbi)->last_victim[FLUSH_DEVICE] = init_segno;
+	SIT_I(sbi)->last_victim[FLUSH_DEVICE] = gc_control->victim_segno;
+
+	if (gc_type == FG_GC)
+		f2fs_unpin_all_sections(sbi, true);
 
 	trace_f2fs_gc_end(sbi->sb, ret, total_freed, sec_freed,
 				get_pages(sbi, F2FS_DIRTY_NODES),
@@ -1816,7 +1876,7 @@ stop:
 
 	put_gc_inode(&gc_list);
 
-	if (sync && !ret)
+	if (gc_control->err_gc_skipped && !ret)
 		ret = sec_freed ? 0 : -EAGAIN;
 	return ret;
 }