Merge tag 'for-f2fs-4.9' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs

Pull f2fs updates from Jaegeuk Kim:
 "In this round, we've investigated how f2fs deals with errors given by
  our fault injection facility. With this, we could fix several corner
  cases. And, in order to improve the performance, we set inline_dentry
  by default and enhance the exisiting discard issue flow. In addition,
  we added f2fs_migrate_page for better memory management.

  Enhancements:
   - set inline_dentry by default
   - improve discard issue flow
   - add more fault injection cases in f2fs
   - allow block preallocation for encrypted files
   - introduce migrate_page callback function
   - avoid truncating the next direct node block at every checkpoint

  Bug fixes:
   - set page flag correctly between write_begin and write_end
   - missing error handling cases detected by fault injection
   - preallocate blocks regarding to 4KB alignement correctly
   - dentry and filename handling of encryption
   - lost xattrs of directories"

* tag 'for-f2fs-4.9' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (69 commits)
  f2fs: introduce update_ckpt_flags to clean up
  f2fs: don't submit irrelevant page
  f2fs: fix to commit bio cache after flushing node pages
  f2fs: introduce get_checkpoint_version for cleanup
  f2fs: remove dead variable
  f2fs: remove redundant io plug
  f2fs: support checkpoint error injection
  f2fs: fix to recover old fault injection config in ->remount_fs
  f2fs: do fault injection initialization in default_options
  f2fs: remove redundant value definition
  f2fs: support configuring fault injection per superblock
  f2fs: adjust display format of segment bit
  f2fs: remove dirty inode pages in error path
  f2fs: do not unnecessarily null-terminate encrypted symlink data
  f2fs: handle errors during recover_orphan_inodes
  f2fs: avoid gc in cp_error case
  f2fs: should put_page for summary page
  f2fs: assign return value in f2fs_gc
  f2fs: add customized migrate_page callback
  f2fs: introduce cp_lock to protect updating of ckpt_flags
  ...

1 files changed, 159 insertions, 60 deletions

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index a46296f57b02..fc886f008449 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -26,6 +26,7 @@
 #define __reverse_ffz(x) __reverse_ffs(~(x))
 
 static struct kmem_cache *discard_entry_slab;
+static struct kmem_cache *bio_entry_slab;
 static struct kmem_cache *sit_entry_set_slab;
 static struct kmem_cache *inmem_entry_slab;
 
@@ -344,6 +345,11 @@ int commit_inmem_pages(struct inode *inode)
  */
 void f2fs_balance_fs(struct f2fs_sb_info *sbi, bool need)
 {
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(sbi, FAULT_CHECKPOINT))
+		f2fs_stop_checkpoint(sbi, false);
+#endif
+
 	if (!need)
 		return;
 
@@ -355,7 +361,7 @@ void f2fs_balance_fs(struct f2fs_sb_info *sbi, bool need)
 	 * We should do GC or end up with checkpoint, if there are so many dirty
 	 * dir/node pages without enough free segments.
 	 */
-	if (has_not_enough_free_secs(sbi, 0)) {
+	if (has_not_enough_free_secs(sbi, 0, 0)) {
 		mutex_lock(&sbi->gc_mutex);
 		f2fs_gc(sbi, false);
 	}
@@ -580,6 +586,74 @@ static void locate_dirty_segment(struct f2fs_sb_info *sbi, unsigned int segno)
 	mutex_unlock(&dirty_i->seglist_lock);
 }
 
+static struct bio_entry *__add_bio_entry(struct f2fs_sb_info *sbi,
+							struct bio *bio)
+{
+	struct list_head *wait_list = &(SM_I(sbi)->wait_list);
+	struct bio_entry *be = f2fs_kmem_cache_alloc(bio_entry_slab, GFP_NOFS);
+
+	INIT_LIST_HEAD(&be->list);
+	be->bio = bio;
+	init_completion(&be->event);
+	list_add_tail(&be->list, wait_list);
+
+	return be;
+}
+
+void f2fs_wait_all_discard_bio(struct f2fs_sb_info *sbi)
+{
+	struct list_head *wait_list = &(SM_I(sbi)->wait_list);
+	struct bio_entry *be, *tmp;
+
+	list_for_each_entry_safe(be, tmp, wait_list, list) {
+		struct bio *bio = be->bio;
+		int err;
+
+		wait_for_completion_io(&be->event);
+		err = be->error;
+		if (err == -EOPNOTSUPP)
+			err = 0;
+
+		if (err)
+			f2fs_msg(sbi->sb, KERN_INFO,
+				"Issue discard failed, ret: %d", err);
+
+		bio_put(bio);
+		list_del(&be->list);
+		kmem_cache_free(bio_entry_slab, be);
+	}
+}
+
+static void f2fs_submit_bio_wait_endio(struct bio *bio)
+{
+	struct bio_entry *be = (struct bio_entry *)bio->bi_private;
+
+	be->error = bio->bi_error;
+	complete(&be->event);
+}
+
+/* this function is copied from blkdev_issue_discard from block/blk-lib.c */
+int __f2fs_issue_discard_async(struct f2fs_sb_info *sbi, sector_t sector,
+		sector_t nr_sects, gfp_t gfp_mask, unsigned long flags)
+{
+	struct block_device *bdev = sbi->sb->s_bdev;
+	struct bio *bio = NULL;
+	int err;
+
+	err = __blkdev_issue_discard(bdev, sector, nr_sects, gfp_mask, flags,
+			&bio);
+	if (!err && bio) {
+		struct bio_entry *be = __add_bio_entry(sbi, bio);
+
+		bio->bi_private = be;
+		bio->bi_end_io = f2fs_submit_bio_wait_endio;
+		bio->bi_opf |= REQ_SYNC;
+		submit_bio(bio);
+	}
+
+	return err;
+}
+
 static int f2fs_issue_discard(struct f2fs_sb_info *sbi,
 				block_t blkstart, block_t blklen)
 {
@@ -597,29 +671,7 @@ static int f2fs_issue_discard(struct f2fs_sb_info *sbi,
 			sbi->discard_blks--;
 	}
 	trace_f2fs_issue_discard(sbi->sb, blkstart, blklen);
-	return blkdev_issue_discard(sbi->sb->s_bdev, start, len, GFP_NOFS, 0);
-}
-
-bool discard_next_dnode(struct f2fs_sb_info *sbi, block_t blkaddr)
-{
-	int err = -EOPNOTSUPP;
-
-	if (test_opt(sbi, DISCARD)) {
-		struct seg_entry *se = get_seg_entry(sbi,
-				GET_SEGNO(sbi, blkaddr));
-		unsigned int offset = GET_BLKOFF_FROM_SEG0(sbi, blkaddr);
-
-		if (f2fs_test_bit(offset, se->discard_map))
-			return false;
-
-		err = f2fs_issue_discard(sbi, blkaddr, 1);
-	}
-
-	if (err) {
-		update_meta_page(sbi, NULL, blkaddr);
-		return true;
-	}
-	return false;
+	return __f2fs_issue_discard_async(sbi, start, len, GFP_NOFS, 0);
 }
 
 static void __add_discard_entry(struct f2fs_sb_info *sbi,
@@ -660,7 +712,7 @@ static void add_discard_addrs(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	bool force = (cpc->reason == CP_DISCARD);
 	int i;
 
-	if (se->valid_blocks == max_blocks)
+	if (se->valid_blocks == max_blocks || !f2fs_discard_en(sbi))
 		return;
 
 	if (!force) {
@@ -719,11 +771,14 @@ void clear_prefree_segments(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	struct list_head *head = &(SM_I(sbi)->discard_list);
 	struct discard_entry *entry, *this;
 	struct dirty_seglist_info *dirty_i = DIRTY_I(sbi);
+	struct blk_plug plug;
 	unsigned long *prefree_map = dirty_i->dirty_segmap[PRE];
 	unsigned int start = 0, end = -1;
 	unsigned int secno, start_segno;
 	bool force = (cpc->reason == CP_DISCARD);
 
+	blk_start_plug(&plug);
+
 	mutex_lock(&dirty_i->seglist_lock);
 
 	while (1) {
@@ -772,6 +827,8 @@ skip:
 		SM_I(sbi)->nr_discards -= entry->len;
 		kmem_cache_free(discard_entry_slab, entry);
 	}
+
+	blk_finish_plug(&plug);
 }
 
 static bool __mark_sit_entry_dirty(struct f2fs_sb_info *sbi, unsigned int segno)
@@ -818,12 +875,14 @@ static void update_sit_entry(struct f2fs_sb_info *sbi, block_t blkaddr, int del)
 	if (del > 0) {
 		if (f2fs_test_and_set_bit(offset, se->cur_valid_map))
 			f2fs_bug_on(sbi, 1);
-		if (!f2fs_test_and_set_bit(offset, se->discard_map))
+		if (f2fs_discard_en(sbi) &&
+			!f2fs_test_and_set_bit(offset, se->discard_map))
 			sbi->discard_blks--;
 	} else {
 		if (!f2fs_test_and_clear_bit(offset, se->cur_valid_map))
 			f2fs_bug_on(sbi, 1);
-		if (f2fs_test_and_clear_bit(offset, se->discard_map))
+		if (f2fs_discard_en(sbi) &&
+			f2fs_test_and_clear_bit(offset, se->discard_map))
 			sbi->discard_blks++;
 	}
 	if (!f2fs_test_bit(offset, se->ckpt_valid_map))
@@ -1202,7 +1261,7 @@ static int get_ssr_segment(struct f2fs_sb_info *sbi, int type)
 	struct curseg_info *curseg = CURSEG_I(sbi, type);
 	const struct victim_selection *v_ops = DIRTY_I(sbi)->v_ops;
 
-	if (IS_NODESEG(type) || !has_not_enough_free_secs(sbi, 0))
+	if (IS_NODESEG(type) || !has_not_enough_free_secs(sbi, 0, 0))
 		return v_ops->get_victim(sbi,
 				&(curseg)->next_segno, BG_GC, type, SSR);
 
@@ -1277,6 +1336,12 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range)
 	if (end <= MAIN_BLKADDR(sbi))
 		goto out;
 
+	if (is_sbi_flag_set(sbi, SBI_NEED_FSCK)) {
+		f2fs_msg(sbi->sb, KERN_WARNING,
+			"Found FS corruption, run fsck to fix.");
+		goto out;
+	}
+
 	/* start/end segment number in main_area */
 	start_segno = (start <= MAIN_BLKADDR(sbi)) ? 0 : GET_SEGNO(sbi, start);
 	end_segno = (end >= MAX_BLKADDR(sbi)) ? MAIN_SEGS(sbi) - 1 :
@@ -1301,6 +1366,10 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range)
 		mutex_lock(&sbi->gc_mutex);
 		err = write_checkpoint(sbi, &cpc);
 		mutex_unlock(&sbi->gc_mutex);
+		if (err)
+			break;
+
+		schedule();
 	}
 out:
 	range->len = F2FS_BLK_TO_BYTES(cpc.trimmed);
@@ -1391,7 +1460,7 @@ void allocate_data_block(struct f2fs_sb_info *sbi, struct page *page,
 
 	/* direct_io'ed data is aligned to the segment for better performance */
 	if (direct_io && curseg->next_blkoff &&
-				!has_not_enough_free_secs(sbi, 0))
+				!has_not_enough_free_secs(sbi, 0, 0))
 		__allocate_new_segments(sbi, type);
 
 	*new_blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
@@ -1589,11 +1658,9 @@ void f2fs_wait_on_encrypted_page_writeback(struct f2fs_sb_info *sbi,
 {
 	struct page *cpage;
 
-	if (blkaddr == NEW_ADDR)
+	if (blkaddr == NEW_ADDR || blkaddr == NULL_ADDR)
 		return;
 
-	f2fs_bug_on(sbi, blkaddr == NULL_ADDR);
-
 	cpage = find_lock_page(META_MAPPING(sbi), blkaddr);
 	if (cpage) {
 		f2fs_wait_on_page_writeback(cpage, DATA, true);
@@ -1739,7 +1806,7 @@ static int restore_curseg_summaries(struct f2fs_sb_info *sbi)
 	int type = CURSEG_HOT_DATA;
 	int err;
 
-	if (is_set_ckpt_flags(F2FS_CKPT(sbi), CP_COMPACT_SUM_FLAG)) {
+	if (is_set_ckpt_flags(sbi, CP_COMPACT_SUM_FLAG)) {
 		int npages = npages_for_summary_flush(sbi, true);
 
 		if (npages >= 2)
@@ -1836,7 +1903,7 @@ static void write_normal_summaries(struct f2fs_sb_info *sbi,
 
 void write_data_summaries(struct f2fs_sb_info *sbi, block_t start_blk)
 {
-	if (is_set_ckpt_flags(F2FS_CKPT(sbi), CP_COMPACT_SUM_FLAG))
+	if (is_set_ckpt_flags(sbi, CP_COMPACT_SUM_FLAG))
 		write_compacted_summaries(sbi, start_blk);
 	else
 		write_normal_summaries(sbi, start_blk, CURSEG_HOT_DATA);
@@ -2127,12 +2194,16 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
 			= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
 		sit_i->sentries[start].ckpt_valid_map
 			= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
-		sit_i->sentries[start].discard_map
-			= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
 		if (!sit_i->sentries[start].cur_valid_map ||
-				!sit_i->sentries[start].ckpt_valid_map ||
-				!sit_i->sentries[start].discard_map)
+				!sit_i->sentries[start].ckpt_valid_map)
 			return -ENOMEM;
+
+		if (f2fs_discard_en(sbi)) {
+			sit_i->sentries[start].discard_map
+				= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
+			if (!sit_i->sentries[start].discard_map)
+				return -ENOMEM;
+		}
 	}
 
 	sit_i->tmp_map = kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
@@ -2239,6 +2310,8 @@ static void build_sit_entries(struct f2fs_sb_info *sbi)
 	struct sit_info *sit_i = SIT_I(sbi);
 	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_COLD_DATA);
 	struct f2fs_journal *journal = curseg->journal;
+	struct seg_entry *se;
+	struct f2fs_sit_entry sit;
 	int sit_blk_cnt = SIT_BLK_CNT(sbi);
 	unsigned int i, start, end;
 	unsigned int readed, start_blk = 0;
@@ -2251,41 +2324,58 @@ static void build_sit_entries(struct f2fs_sb_info *sbi)
 		end = (start_blk + readed) * sit_i->sents_per_block;
 
 		for (; start < end && start < MAIN_SEGS(sbi); start++) {
-			struct seg_entry *se = &sit_i->sentries[start];
 			struct f2fs_sit_block *sit_blk;
-			struct f2fs_sit_entry sit;
 			struct page *page;
 
-			down_read(&curseg->journal_rwsem);
-			for (i = 0; i < sits_in_cursum(journal); i++) {
-				if (le32_to_cpu(segno_in_journal(journal, i))
-								== start) {
-					sit = sit_in_journal(journal, i);
-					up_read(&curseg->journal_rwsem);
-					goto got_it;
-				}
-			}
-			up_read(&curseg->journal_rwsem);
-
+			se = &sit_i->sentries[start];
 			page = get_current_sit_page(sbi, start);
 			sit_blk = (struct f2fs_sit_block *)page_address(page);
 			sit = sit_blk->entries[SIT_ENTRY_OFFSET(sit_i, start)];
 			f2fs_put_page(page, 1);
-got_it:
+
 			check_block_count(sbi, start, &sit);
 			seg_info_from_raw_sit(se, &sit);
 
 			/* build discard map only one time */
-			memcpy(se->discard_map, se->cur_valid_map, SIT_VBLOCK_MAP_SIZE);
-			sbi->discard_blks += sbi->blocks_per_seg - se->valid_blocks;
-
-			if (sbi->segs_per_sec > 1) {
-				struct sec_entry *e = get_sec_entry(sbi, start);
-				e->valid_blocks += se->valid_blocks;
+			if (f2fs_discard_en(sbi)) {
+				memcpy(se->discard_map, se->cur_valid_map,
+							SIT_VBLOCK_MAP_SIZE);
+				sbi->discard_blks += sbi->blocks_per_seg -
+							se->valid_blocks;
 			}
+
+			if (sbi->segs_per_sec > 1)
+				get_sec_entry(sbi, start)->valid_blocks +=
+							se->valid_blocks;
 		}
 		start_blk += readed;
 	} while (start_blk < sit_blk_cnt);
+
+	down_read(&curseg->journal_rwsem);
+	for (i = 0; i < sits_in_cursum(journal); i++) {
+		unsigned int old_valid_blocks;
+
+		start = le32_to_cpu(segno_in_journal(journal, i));
+		se = &sit_i->sentries[start];
+		sit = sit_in_journal(journal, i);
+
+		old_valid_blocks = se->valid_blocks;
+
+		check_block_count(sbi, start, &sit);
+		seg_info_from_raw_sit(se, &sit);
+
+		if (f2fs_discard_en(sbi)) {
+			memcpy(se->discard_map, se->cur_valid_map,
+						SIT_VBLOCK_MAP_SIZE);
+			sbi->discard_blks += old_valid_blocks -
+						se->valid_blocks;
+		}
+
+		if (sbi->segs_per_sec > 1)
+			get_sec_entry(sbi, start)->valid_blocks +=
+				se->valid_blocks - old_valid_blocks;
+	}
+	up_read(&curseg->journal_rwsem);
 }
 
 static void init_free_segmap(struct f2fs_sb_info *sbi)
@@ -2427,6 +2517,7 @@ int build_segment_manager(struct f2fs_sb_info *sbi)
 	sm_info->min_fsync_blocks = DEF_MIN_FSYNC_BLOCKS;
 
 	INIT_LIST_HEAD(&sm_info->discard_list);
+	INIT_LIST_HEAD(&sm_info->wait_list);
 	sm_info->nr_discards = 0;
 	sm_info->max_discards = 0;
 
@@ -2570,10 +2661,15 @@ int __init create_segment_manager_caches(void)
 	if (!discard_entry_slab)
 		goto fail;
 
+	bio_entry_slab = f2fs_kmem_cache_create("bio_entry",
+			sizeof(struct bio_entry));
+	if (!bio_entry_slab)
+		goto destroy_discard_entry;
+
 	sit_entry_set_slab = f2fs_kmem_cache_create("sit_entry_set",
 			sizeof(struct sit_entry_set));
 	if (!sit_entry_set_slab)
-		goto destory_discard_entry;
+		goto destroy_bio_entry;
 
 	inmem_entry_slab = f2fs_kmem_cache_create("inmem_page_entry",
 			sizeof(struct inmem_pages));
@@ -2583,7 +2679,9 @@ int __init create_segment_manager_caches(void)
 
 destroy_sit_entry_set:
 	kmem_cache_destroy(sit_entry_set_slab);
-destory_discard_entry:
+destroy_bio_entry:
+	kmem_cache_destroy(bio_entry_slab);
+destroy_discard_entry:
 	kmem_cache_destroy(discard_entry_slab);
 fail:
 	return -ENOMEM;
@@ -2592,6 +2690,7 @@ fail:
 void destroy_segment_manager_caches(void)
 {
 	kmem_cache_destroy(sit_entry_set_slab);
+	kmem_cache_destroy(bio_entry_slab);
 	kmem_cache_destroy(discard_entry_slab);
 	kmem_cache_destroy(inmem_entry_slab);
 }