Merge tag 'f2fs-for-5.8' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs

Pull f2fs updates from Jaegeuk Kim:
 "In this round, we've added some knobs to enhance compression feature
  and harden testing environment. In addition, we've fixed several bugs
  reported from Android devices such as long discarding latency, device
  hanging during quota_sync, etc.

  Enhancements:
   - support lzo-rle algorithm
   - add two ioctls to release and reserve blocks for compression
   - support partial truncation/fiemap on compressed file
   - introduce sysfs entries to attach IO flags explicitly
   - add iostat trace point along with read io stat

  Bug fixes:
   - fix long discard latency
   - flush quota data by f2fs_quota_sync correctly
   - fix to recover parent inode number for power-cut recovery
   - fix lz4/zstd output buffer budget
   - parse checkpoint mount option correctly
   - avoid inifinite loop to wait for flushing node/meta pages
   - manage discard space correctly

  And some refactoring and clean up patches were added"

* tag 'f2fs-for-5.8' of git://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs: (51 commits)
  f2fs: attach IO flags to the missing cases
  f2fs: add node_io_flag for bio flags likewise data_io_flag
  f2fs: remove unused parameter of f2fs_put_rpages_mapping()
  f2fs: handle readonly filesystem in f2fs_ioc_shutdown()
  f2fs: avoid utf8_strncasecmp() with unstable name
  f2fs: don't return vmalloc() memory from f2fs_kmalloc()
  f2fs: fix retry logic in f2fs_write_cache_pages()
  f2fs: fix wrong discard space
  f2fs: compress: don't compress any datas after cp stop
  f2fs: remove unneeded return value of __insert_discard_tree()
  f2fs: fix wrong value of tracepoint parameter
  f2fs: protect new segment allocation in expand_inode_data
  f2fs: code cleanup by removing ifdef macro surrounding
  f2fs: avoid inifinite loop to wait for flushing node pages at cp_error
  f2fs: flush dirty meta pages when flushing them
  f2fs: fix checkpoint=disable:%u%%
  f2fs: compress: fix zstd data corruption
  f2fs: add compressed/gc data read IO stat
  f2fs: fix potential use-after-free issue
  f2fs: compress: don't handle non-compressed data in workqueue
  ...

1 files changed, 28 insertions, 12 deletions

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index b7a9421472a7..196f31503511 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -1029,9 +1029,9 @@ static void f2fs_submit_discard_endio(struct bio *bio)
 	struct discard_cmd *dc = (struct discard_cmd *)bio->bi_private;
 	unsigned long flags;
 
-	dc->error = blk_status_to_errno(bio->bi_status);
-
 	spin_lock_irqsave(&dc->lock, flags);
+	if (!dc->error)
+		dc->error = blk_status_to_errno(bio->bi_status);
 	dc->bio_ref--;
 	if (!dc->bio_ref && dc->state == D_SUBMIT) {
 		dc->state = D_DONE;
@@ -1101,7 +1101,6 @@ static void __init_discard_policy(struct f2fs_sb_info *sbi,
 	} else if (discard_type == DPOLICY_FSTRIM) {
 		dpolicy->io_aware = false;
 	} else if (discard_type == DPOLICY_UMOUNT) {
-		dpolicy->max_requests = UINT_MAX;
 		dpolicy->io_aware = false;
 		/* we need to issue all to keep CP_TRIMMED_FLAG */
 		dpolicy->granularity = 1;
@@ -1215,12 +1214,14 @@ submit:
 		len = total_len;
 	}
 
-	if (!err && len)
+	if (!err && len) {
+		dcc->undiscard_blks -= len;
 		__update_discard_tree_range(sbi, bdev, lstart, start, len);
+	}
 	return err;
 }
 
-static struct discard_cmd *__insert_discard_tree(struct f2fs_sb_info *sbi,
+static void __insert_discard_tree(struct f2fs_sb_info *sbi,
 				struct block_device *bdev, block_t lstart,
 				block_t start, block_t len,
 				struct rb_node **insert_p,
@@ -1229,7 +1230,6 @@ static struct discard_cmd *__insert_discard_tree(struct f2fs_sb_info *sbi,
 	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
 	struct rb_node **p;
 	struct rb_node *parent = NULL;
-	struct discard_cmd *dc = NULL;
 	bool leftmost = true;
 
 	if (insert_p && insert_parent) {
@@ -1241,12 +1241,8 @@ static struct discard_cmd *__insert_discard_tree(struct f2fs_sb_info *sbi,
 	p = f2fs_lookup_rb_tree_for_insert(sbi, &dcc->root, &parent,
 							lstart, &leftmost);
 do_insert:
-	dc = __attach_discard_cmd(sbi, bdev, lstart, start, len, parent,
+	__attach_discard_cmd(sbi, bdev, lstart, start, len, parent,
 								p, leftmost);
-	if (!dc)
-		return NULL;
-
-	return dc;
 }
 
 static void __relocate_discard_cmd(struct discard_cmd_control *dcc,
@@ -1463,6 +1459,8 @@ next:
 
 	return issued;
 }
+static unsigned int __wait_all_discard_cmd(struct f2fs_sb_info *sbi,
+					struct discard_policy *dpolicy);
 
 static int __issue_discard_cmd(struct f2fs_sb_info *sbi,
 					struct discard_policy *dpolicy)
@@ -1471,12 +1469,14 @@ static int __issue_discard_cmd(struct f2fs_sb_info *sbi,
 	struct list_head *pend_list;
 	struct discard_cmd *dc, *tmp;
 	struct blk_plug plug;
-	int i, issued = 0;
+	int i, issued;
 	bool io_interrupted = false;
 
 	if (dpolicy->timeout)
 		f2fs_update_time(sbi, UMOUNT_DISCARD_TIMEOUT);
 
+retry:
+	issued = 0;
 	for (i = MAX_PLIST_NUM - 1; i >= 0; i--) {
 		if (dpolicy->timeout &&
 				f2fs_time_over(sbi, UMOUNT_DISCARD_TIMEOUT))
@@ -1523,6 +1523,11 @@ next:
 			break;
 	}
 
+	if (dpolicy->type == DPOLICY_UMOUNT && issued) {
+		__wait_all_discard_cmd(sbi, dpolicy);
+		goto retry;
+	}
+
 	if (!issued && io_interrupted)
 		issued = -1;
 
@@ -3102,6 +3107,14 @@ void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page,
 		type = CURSEG_COLD_DATA;
 	}
 
+	/*
+	 * We need to wait for node_write to avoid block allocation during
+	 * checkpoint. This can only happen to quota writes which can cause
+	 * the below discard race condition.
+	 */
+	if (IS_DATASEG(type))
+		down_write(&sbi->node_write);
+
 	down_read(&SM_I(sbi)->curseg_lock);
 
 	mutex_lock(&curseg->curseg_mutex);
@@ -3167,6 +3180,9 @@ void f2fs_allocate_data_block(struct f2fs_sb_info *sbi, struct page *page,
 
 	up_read(&SM_I(sbi)->curseg_lock);
 
+	if (IS_DATASEG(type))
+		up_write(&sbi->node_write);
+
 	if (put_pin_sem)
 		up_read(&sbi->pin_sem);
 }