From 043d2d00b44310f84c0593c63e51fae88c829cdd Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Fri, 10 Mar 2023 10:04:26 -0800
Subject: f2fs: factor out victim_entry usage from general rb_tree use

Let's reduce the complexity of mixed use of rb_tree in victim_entry from
extent_cache and discard_cmd.

This should fix arm32 memory alignment issue caused by shared rb_entry.

[struct victim_entry]              [struct rb_entry]
[0] struct rb_node rb_node;        [0] struct rb_node rb_node;
                                       union {
                                         struct {
                                           unsigned int ofs;
                                           unsigned int len;
                                         };
[16] unsigned long long mtime;     [12] unsigned long long key;
                                       } __packed;

Cc: <stable@vger.kernel.org>
Fixes: 093749e296e2 ("f2fs: support age threshold based garbage collection")
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

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

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 227e25836173..e98a12e8dca1 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -1478,7 +1478,7 @@ retry:
 			goto next;
 		if (unlikely(dcc->rbtree_check))
 			f2fs_bug_on(sbi, !f2fs_check_rb_tree_consistence(sbi,
-							&dcc->root, false));
+							&dcc->root));
 		blk_start_plug(&plug);
 		list_for_each_entry_safe(dc, tmp, pend_list, list) {
 			f2fs_bug_on(sbi, dc->state != D_PREP);
@@ -2965,7 +2965,7 @@ next:
 	mutex_lock(&dcc->cmd_lock);
 	if (unlikely(dcc->rbtree_check))
 		f2fs_bug_on(sbi, !f2fs_check_rb_tree_consistence(sbi,
-							&dcc->root, false));
+							&dcc->root));
 
 	dc = (struct discard_cmd *)f2fs_lookup_rb_tree_ret(&dcc->root,
 					NULL, start,
-- 
cgit v1.2.3


From f69475dd4878e5f2e316a6573044d55f294baa51 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Fri, 10 Mar 2023 11:12:35 -0800
Subject: f2fs: factor out discard_cmd usage from general rb_tree use

This is a second part to remove the mixed use of rb_tree in discard_cmd from
extent_cache.

This should also fix arm32 memory alignment issue caused by shared rb_entry.

[struct discard_cmd]               [struct rb_entry]
[0] struct rb_node rb_node;        [0] struct rb_node rb_node;
  union {                              union {
    struct {                             struct {
[16]  block_t lstart;              [12]    unsigned int ofs;
      block_t len;                         unsigned int len;
                                         };
                                         unsigned long long key;
                                       } __packed;

Cc: <stable@vger.kernel.org>
Fixes: 004b68621897 ("f2fs: use rb-tree to track pending discard commands")
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/extent_cache.c |  36 +------
 fs/f2fs/f2fs.h         |  23 +----
 fs/f2fs/segment.c      | 249 ++++++++++++++++++++++++++++++++-----------------
 3 files changed, 169 insertions(+), 139 deletions(-)

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

diff --git a/fs/f2fs/extent_cache.c b/fs/f2fs/extent_cache.c
index d1aa4609ca6b..5c206f941aac 100644
--- a/fs/f2fs/extent_cache.c
+++ b/fs/f2fs/extent_cache.c
@@ -192,7 +192,7 @@ static struct rb_entry *__lookup_rb_tree_slow(struct rb_root_cached *root,
 	return NULL;
 }
 
-struct rb_entry *f2fs_lookup_rb_tree(struct rb_root_cached *root,
+static struct rb_entry *f2fs_lookup_rb_tree(struct rb_root_cached *root,
 				struct rb_entry *cached_re, unsigned int ofs)
 {
 	struct rb_entry *re;
@@ -204,7 +204,7 @@ struct rb_entry *f2fs_lookup_rb_tree(struct rb_root_cached *root,
 	return re;
 }
 
-struct rb_node **f2fs_lookup_rb_tree_for_insert(struct f2fs_sb_info *sbi,
+static struct rb_node **f2fs_lookup_rb_tree_for_insert(struct f2fs_sb_info *sbi,
 				struct rb_root_cached *root,
 				struct rb_node **parent,
 				unsigned int ofs, bool *leftmost)
@@ -238,7 +238,7 @@ struct rb_node **f2fs_lookup_rb_tree_for_insert(struct f2fs_sb_info *sbi,
  * in order to simplify the insertion after.
  * tree must stay unchanged between lookup and insertion.
  */
-struct rb_entry *f2fs_lookup_rb_tree_ret(struct rb_root_cached *root,
+static struct rb_entry *f2fs_lookup_rb_tree_ret(struct rb_root_cached *root,
 				struct rb_entry *cached_re,
 				unsigned int ofs,
 				struct rb_entry **prev_entry,
@@ -311,36 +311,6 @@ lookup_neighbors:
 	return re;
 }
 
-bool f2fs_check_rb_tree_consistence(struct f2fs_sb_info *sbi,
-				struct rb_root_cached *root)
-{
-#ifdef CONFIG_F2FS_CHECK_FS
-	struct rb_node *cur = rb_first_cached(root), *next;
-	struct rb_entry *cur_re, *next_re;
-
-	if (!cur)
-		return true;
-
-	while (cur) {
-		next = rb_next(cur);
-		if (!next)
-			return true;
-
-		cur_re = rb_entry(cur, struct rb_entry, rb_node);
-		next_re = rb_entry(next, struct rb_entry, rb_node);
-
-		if (cur_re->ofs + cur_re->len > next_re->ofs) {
-			f2fs_info(sbi, "inconsistent rbtree, cur(%u, %u) next(%u, %u)",
-				  cur_re->ofs, cur_re->len,
-				  next_re->ofs, next_re->len);
-			return false;
-		}
-		cur = next;
-	}
-#endif
-	return true;
-}
-
 static struct kmem_cache *extent_tree_slab;
 static struct kmem_cache *extent_node_slab;
 
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 9396549e112d..6e04fea9c34f 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -353,15 +353,7 @@ struct discard_info {
 
 struct discard_cmd {
 	struct rb_node rb_node;		/* rb node located in rb-tree */
-	union {
-		struct {
-			block_t lstart;	/* logical start address */
-			block_t len;	/* length */
-			block_t start;	/* actual start address in dev */
-		};
-		struct discard_info di;	/* discard info */
-
-	};
+	struct discard_info di;		/* discard info */
 	struct list_head list;		/* command list */
 	struct completion wait;		/* compleation */
 	struct block_device *bdev;	/* bdev */
@@ -4132,19 +4124,6 @@ void f2fs_leave_shrinker(struct f2fs_sb_info *sbi);
  * extent_cache.c
  */
 bool sanity_check_extent_cache(struct inode *inode);
-struct rb_entry *f2fs_lookup_rb_tree(struct rb_root_cached *root,
-				struct rb_entry *cached_re, unsigned int ofs);
-struct rb_node **f2fs_lookup_rb_tree_for_insert(struct f2fs_sb_info *sbi,
-				struct rb_root_cached *root,
-				struct rb_node **parent,
-				unsigned int ofs, bool *leftmost);
-struct rb_entry *f2fs_lookup_rb_tree_ret(struct rb_root_cached *root,
-		struct rb_entry *cached_re, unsigned int ofs,
-		struct rb_entry **prev_entry, struct rb_entry **next_entry,
-		struct rb_node ***insert_p, struct rb_node **insert_parent,
-		bool force, bool *leftmost);
-bool f2fs_check_rb_tree_consistence(struct f2fs_sb_info *sbi,
-				struct rb_root_cached *root);
 void f2fs_init_extent_tree(struct inode *inode);
 void f2fs_drop_extent_tree(struct inode *inode);
 void f2fs_destroy_extent_node(struct inode *inode);
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index e98a12e8dca1..d135046108ef 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -933,9 +933,9 @@ static struct discard_cmd *__create_discard_cmd(struct f2fs_sb_info *sbi,
 	dc = f2fs_kmem_cache_alloc(discard_cmd_slab, GFP_NOFS, true, NULL);
 	INIT_LIST_HEAD(&dc->list);
 	dc->bdev = bdev;
-	dc->lstart = lstart;
-	dc->start = start;
-	dc->len = len;
+	dc->di.lstart = lstart;
+	dc->di.start = start;
+	dc->di.len = len;
 	dc->ref = 0;
 	dc->state = D_PREP;
 	dc->queued = 0;
@@ -950,20 +950,108 @@ static struct discard_cmd *__create_discard_cmd(struct f2fs_sb_info *sbi,
 	return dc;
 }
 
-static struct discard_cmd *__attach_discard_cmd(struct f2fs_sb_info *sbi,
-				struct block_device *bdev, block_t lstart,
-				block_t start, block_t len,
-				struct rb_node *parent, struct rb_node **p,
-				bool leftmost)
+static bool f2fs_check_discard_tree(struct f2fs_sb_info *sbi)
+{
+#ifdef CONFIG_F2FS_CHECK_FS
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct rb_node *cur = rb_first_cached(&dcc->root), *next;
+	struct discard_cmd *cur_dc, *next_dc;
+
+	while (cur) {
+		next = rb_next(cur);
+		if (!next)
+			return true;
+
+		cur_dc = rb_entry(cur, struct discard_cmd, rb_node);
+		next_dc = rb_entry(next, struct discard_cmd, rb_node);
+
+		if (cur_dc->di.lstart + cur_dc->di.len > next_dc->di.lstart) {
+			f2fs_info(sbi, "broken discard_rbtree, "
+				"cur(%u, %u) next(%u, %u)",
+				cur_dc->di.lstart, cur_dc->di.len,
+				next_dc->di.lstart, next_dc->di.len);
+			return false;
+		}
+		cur = next;
+	}
+#endif
+	return true;
+}
+
+static struct discard_cmd *__lookup_discard_cmd(struct f2fs_sb_info *sbi,
+						block_t blkaddr)
 {
 	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct rb_node *node = dcc->root.rb_root.rb_node;
 	struct discard_cmd *dc;
 
-	dc = __create_discard_cmd(sbi, bdev, lstart, start, len);
+	while (node) {
+		dc = rb_entry(node, struct discard_cmd, rb_node);
 
-	rb_link_node(&dc->rb_node, parent, p);
-	rb_insert_color_cached(&dc->rb_node, &dcc->root, leftmost);
+		if (blkaddr < dc->di.lstart)
+			node = node->rb_left;
+		else if (blkaddr >= dc->di.lstart + dc->di.len)
+			node = node->rb_right;
+		else
+			return dc;
+	}
+	return NULL;
+}
+
+static struct discard_cmd *__lookup_discard_cmd_ret(struct rb_root_cached *root,
+				block_t blkaddr,
+				struct discard_cmd **prev_entry,
+				struct discard_cmd **next_entry,
+				struct rb_node ***insert_p,
+				struct rb_node **insert_parent)
+{
+	struct rb_node **pnode = &root->rb_root.rb_node;
+	struct rb_node *parent = NULL, *tmp_node;
+	struct discard_cmd *dc;
+
+	*insert_p = NULL;
+	*insert_parent = NULL;
+	*prev_entry = NULL;
+	*next_entry = NULL;
+
+	if (RB_EMPTY_ROOT(&root->rb_root))
+		return NULL;
+
+	while (*pnode) {
+		parent = *pnode;
+		dc = rb_entry(*pnode, struct discard_cmd, rb_node);
+
+		if (blkaddr < dc->di.lstart)
+			pnode = &(*pnode)->rb_left;
+		else if (blkaddr >= dc->di.lstart + dc->di.len)
+			pnode = &(*pnode)->rb_right;
+		else
+			goto lookup_neighbors;
+	}
+
+	*insert_p = pnode;
+	*insert_parent = parent;
+
+	dc = rb_entry(parent, struct discard_cmd, rb_node);
+	tmp_node = parent;
+	if (parent && blkaddr > dc->di.lstart)
+		tmp_node = rb_next(parent);
+	*next_entry = rb_entry_safe(tmp_node, struct discard_cmd, rb_node);
+
+	tmp_node = parent;
+	if (parent && blkaddr < dc->di.lstart)
+		tmp_node = rb_prev(parent);
+	*prev_entry = rb_entry_safe(tmp_node, struct discard_cmd, rb_node);
+	return NULL;
+
+lookup_neighbors:
+	/* lookup prev node for merging backward later */
+	tmp_node = rb_prev(&dc->rb_node);
+	*prev_entry = rb_entry_safe(tmp_node, struct discard_cmd, rb_node);
 
+	/* lookup next node for merging frontward later */
+	tmp_node = rb_next(&dc->rb_node);
+	*next_entry = rb_entry_safe(tmp_node, struct discard_cmd, rb_node);
 	return dc;
 }
 
@@ -975,7 +1063,7 @@ static void __detach_discard_cmd(struct discard_cmd_control *dcc,
 
 	list_del(&dc->list);
 	rb_erase_cached(&dc->rb_node, &dcc->root);
-	dcc->undiscard_blks -= dc->len;
+	dcc->undiscard_blks -= dc->di.len;
 
 	kmem_cache_free(discard_cmd_slab, dc);
 
@@ -988,7 +1076,7 @@ static void __remove_discard_cmd(struct f2fs_sb_info *sbi,
 	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
 	unsigned long flags;
 
-	trace_f2fs_remove_discard(dc->bdev, dc->start, dc->len);
+	trace_f2fs_remove_discard(dc->bdev, dc->di.start, dc->di.len);
 
 	spin_lock_irqsave(&dc->lock, flags);
 	if (dc->bio_ref) {
@@ -1006,7 +1094,7 @@ static void __remove_discard_cmd(struct f2fs_sb_info *sbi,
 		printk_ratelimited(
 			"%sF2FS-fs (%s): Issue discard(%u, %u, %u) failed, ret: %d",
 			KERN_INFO, sbi->sb->s_id,
-			dc->lstart, dc->start, dc->len, dc->error);
+			dc->di.lstart, dc->di.start, dc->di.len, dc->error);
 	__detach_discard_cmd(dcc, dc);
 }
 
@@ -1122,14 +1210,14 @@ static int __submit_discard_cmd(struct f2fs_sb_info *sbi,
 	if (is_sbi_flag_set(sbi, SBI_NEED_FSCK))
 		return 0;
 
-	trace_f2fs_issue_discard(bdev, dc->start, dc->len);
+	trace_f2fs_issue_discard(bdev, dc->di.start, dc->di.len);
 
-	lstart = dc->lstart;
-	start = dc->start;
-	len = dc->len;
+	lstart = dc->di.lstart;
+	start = dc->di.start;
+	len = dc->di.len;
 	total_len = len;
 
-	dc->len = 0;
+	dc->di.len = 0;
 
 	while (total_len && *issued < dpolicy->max_requests && !err) {
 		struct bio *bio = NULL;
@@ -1145,7 +1233,7 @@ static int __submit_discard_cmd(struct f2fs_sb_info *sbi,
 		if (*issued == dpolicy->max_requests)
 			last = true;
 
-		dc->len += len;
+		dc->di.len += len;
 
 		if (time_to_inject(sbi, FAULT_DISCARD)) {
 			err = -EIO;
@@ -1207,34 +1295,41 @@ static int __submit_discard_cmd(struct f2fs_sb_info *sbi,
 	return err;
 }
 
-static void __insert_discard_tree(struct f2fs_sb_info *sbi,
+static void __insert_discard_cmd(struct f2fs_sb_info *sbi,
 				struct block_device *bdev, block_t lstart,
-				block_t start, block_t len,
-				struct rb_node **insert_p,
-				struct rb_node *insert_parent)
+				block_t start, block_t len)
 {
 	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
-	struct rb_node **p;
+	struct rb_node **p = &dcc->root.rb_root.rb_node;
 	struct rb_node *parent = NULL;
+	struct discard_cmd *dc;
 	bool leftmost = true;
 
-	if (insert_p && insert_parent) {
-		parent = insert_parent;
-		p = insert_p;
-		goto do_insert;
+	/* look up rb tree to find parent node */
+	while (*p) {
+		parent = *p;
+		dc = rb_entry(parent, struct discard_cmd, rb_node);
+
+		if (lstart < dc->di.lstart) {
+			p = &(*p)->rb_left;
+		} else if (lstart >= dc->di.lstart + dc->di.len) {
+			p = &(*p)->rb_right;
+			leftmost = false;
+		} else {
+			f2fs_bug_on(sbi, 1);
+		}
 	}
 
-	p = f2fs_lookup_rb_tree_for_insert(sbi, &dcc->root, &parent,
-							lstart, &leftmost);
-do_insert:
-	__attach_discard_cmd(sbi, bdev, lstart, start, len, parent,
-								p, leftmost);
+	dc = __create_discard_cmd(sbi, bdev, lstart, start, len);
+
+	rb_link_node(&dc->rb_node, parent, p);
+	rb_insert_color_cached(&dc->rb_node, &dcc->root, leftmost);
 }
 
 static void __relocate_discard_cmd(struct discard_cmd_control *dcc,
 						struct discard_cmd *dc)
 {
-	list_move_tail(&dc->list, &dcc->pend_list[plist_idx(dc->len)]);
+	list_move_tail(&dc->list, &dcc->pend_list[plist_idx(dc->di.len)]);
 }
 
 static void __punch_discard_cmd(struct f2fs_sb_info *sbi,
@@ -1244,7 +1339,7 @@ static void __punch_discard_cmd(struct f2fs_sb_info *sbi,
 	struct discard_info di = dc->di;
 	bool modified = false;
 
-	if (dc->state == D_DONE || dc->len == 1) {
+	if (dc->state == D_DONE || dc->di.len == 1) {
 		__remove_discard_cmd(sbi, dc);
 		return;
 	}
@@ -1252,23 +1347,22 @@ static void __punch_discard_cmd(struct f2fs_sb_info *sbi,
 	dcc->undiscard_blks -= di.len;
 
 	if (blkaddr > di.lstart) {
-		dc->len = blkaddr - dc->lstart;
-		dcc->undiscard_blks += dc->len;
+		dc->di.len = blkaddr - dc->di.lstart;
+		dcc->undiscard_blks += dc->di.len;
 		__relocate_discard_cmd(dcc, dc);
 		modified = true;
 	}
 
 	if (blkaddr < di.lstart + di.len - 1) {
 		if (modified) {
-			__insert_discard_tree(sbi, dc->bdev, blkaddr + 1,
+			__insert_discard_cmd(sbi, dc->bdev, blkaddr + 1,
 					di.start + blkaddr + 1 - di.lstart,
-					di.lstart + di.len - 1 - blkaddr,
-					NULL, NULL);
+					di.lstart + di.len - 1 - blkaddr);
 		} else {
-			dc->lstart++;
-			dc->len--;
-			dc->start++;
-			dcc->undiscard_blks += dc->len;
+			dc->di.lstart++;
+			dc->di.len--;
+			dc->di.start++;
+			dcc->undiscard_blks += dc->di.len;
 			__relocate_discard_cmd(dcc, dc);
 		}
 	}
@@ -1287,17 +1381,14 @@ static void __update_discard_tree_range(struct f2fs_sb_info *sbi,
 			SECTOR_TO_BLOCK(bdev_max_discard_sectors(bdev));
 	block_t end = lstart + len;
 
-	dc = (struct discard_cmd *)f2fs_lookup_rb_tree_ret(&dcc->root,
-					NULL, lstart,
-					(struct rb_entry **)&prev_dc,
-					(struct rb_entry **)&next_dc,
-					&insert_p, &insert_parent, true, NULL);
+	dc = __lookup_discard_cmd_ret(&dcc->root, lstart,
+				&prev_dc, &next_dc, &insert_p, &insert_parent);
 	if (dc)
 		prev_dc = dc;
 
 	if (!prev_dc) {
 		di.lstart = lstart;
-		di.len = next_dc ? next_dc->lstart - lstart : len;
+		di.len = next_dc ? next_dc->di.lstart - lstart : len;
 		di.len = min(di.len, len);
 		di.start = start;
 	}
@@ -1308,16 +1399,16 @@ static void __update_discard_tree_range(struct f2fs_sb_info *sbi,
 		struct discard_cmd *tdc = NULL;
 
 		if (prev_dc) {
-			di.lstart = prev_dc->lstart + prev_dc->len;
+			di.lstart = prev_dc->di.lstart + prev_dc->di.len;
 			if (di.lstart < lstart)
 				di.lstart = lstart;
 			if (di.lstart >= end)
 				break;
 
-			if (!next_dc || next_dc->lstart > end)
+			if (!next_dc || next_dc->di.lstart > end)
 				di.len = end - di.lstart;
 			else
-				di.len = next_dc->lstart - di.lstart;
+				di.len = next_dc->di.lstart - di.lstart;
 			di.start = start + di.lstart - lstart;
 		}
 
@@ -1350,10 +1441,9 @@ static void __update_discard_tree_range(struct f2fs_sb_info *sbi,
 			merged = true;
 		}
 
-		if (!merged) {
-			__insert_discard_tree(sbi, bdev, di.lstart, di.start,
-							di.len, NULL, NULL);
-		}
+		if (!merged)
+			__insert_discard_cmd(sbi, bdev,
+						di.lstart, di.start, di.len);
  next:
 		prev_dc = next_dc;
 		if (!prev_dc)
@@ -1392,15 +1482,11 @@ static void __issue_discard_cmd_orderly(struct f2fs_sb_info *sbi,
 	struct rb_node **insert_p = NULL, *insert_parent = NULL;
 	struct discard_cmd *dc;
 	struct blk_plug plug;
-	unsigned int pos = dcc->next_pos;
 	bool io_interrupted = false;
 
 	mutex_lock(&dcc->cmd_lock);
-	dc = (struct discard_cmd *)f2fs_lookup_rb_tree_ret(&dcc->root,
-					NULL, pos,
-					(struct rb_entry **)&prev_dc,
-					(struct rb_entry **)&next_dc,
-					&insert_p, &insert_parent, true, NULL);
+	dc = __lookup_discard_cmd_ret(&dcc->root, dcc->next_pos,
+				&prev_dc, &next_dc, &insert_p, &insert_parent);
 	if (!dc)
 		dc = next_dc;
 
@@ -1418,7 +1504,7 @@ static void __issue_discard_cmd_orderly(struct f2fs_sb_info *sbi,
 			break;
 		}
 
-		dcc->next_pos = dc->lstart + dc->len;
+		dcc->next_pos = dc->di.lstart + dc->di.len;
 		err = __submit_discard_cmd(sbi, dpolicy, dc, issued);
 
 		if (*issued >= dpolicy->max_requests)
@@ -1477,8 +1563,7 @@ retry:
 		if (list_empty(pend_list))
 			goto next;
 		if (unlikely(dcc->rbtree_check))
-			f2fs_bug_on(sbi, !f2fs_check_rb_tree_consistence(sbi,
-							&dcc->root));
+			f2fs_bug_on(sbi, !f2fs_check_discard_tree(sbi));
 		blk_start_plug(&plug);
 		list_for_each_entry_safe(dc, tmp, pend_list, list) {
 			f2fs_bug_on(sbi, dc->state != D_PREP);
@@ -1556,7 +1641,7 @@ static unsigned int __wait_one_discard_bio(struct f2fs_sb_info *sbi,
 	dc->ref--;
 	if (!dc->ref) {
 		if (!dc->error)
-			len = dc->len;
+			len = dc->di.len;
 		__remove_discard_cmd(sbi, dc);
 	}
 	mutex_unlock(&dcc->cmd_lock);
@@ -1579,14 +1664,15 @@ next:
 
 	mutex_lock(&dcc->cmd_lock);
 	list_for_each_entry_safe(iter, tmp, wait_list, list) {
-		if (iter->lstart + iter->len <= start || end <= iter->lstart)
+		if (iter->di.lstart + iter->di.len <= start ||
+					end <= iter->di.lstart)
 			continue;
-		if (iter->len < dpolicy->granularity)
+		if (iter->di.len < dpolicy->granularity)
 			continue;
 		if (iter->state == D_DONE && !iter->ref) {
 			wait_for_completion_io(&iter->wait);
 			if (!iter->error)
-				trimmed += iter->len;
+				trimmed += iter->di.len;
 			__remove_discard_cmd(sbi, iter);
 		} else {
 			iter->ref++;
@@ -1630,8 +1716,7 @@ static void f2fs_wait_discard_bio(struct f2fs_sb_info *sbi, block_t blkaddr)
 	bool need_wait = false;
 
 	mutex_lock(&dcc->cmd_lock);
-	dc = (struct discard_cmd *)f2fs_lookup_rb_tree(&dcc->root,
-							NULL, blkaddr);
+	dc = __lookup_discard_cmd(sbi, blkaddr);
 	if (dc) {
 		if (dc->state == D_PREP) {
 			__punch_discard_cmd(sbi, dc, blkaddr);
@@ -2964,24 +3049,20 @@ next:
 
 	mutex_lock(&dcc->cmd_lock);
 	if (unlikely(dcc->rbtree_check))
-		f2fs_bug_on(sbi, !f2fs_check_rb_tree_consistence(sbi,
-							&dcc->root));
-
-	dc = (struct discard_cmd *)f2fs_lookup_rb_tree_ret(&dcc->root,
-					NULL, start,
-					(struct rb_entry **)&prev_dc,
-					(struct rb_entry **)&next_dc,
-					&insert_p, &insert_parent, true, NULL);
+		f2fs_bug_on(sbi, !f2fs_check_discard_tree(sbi));
+
+	dc = __lookup_discard_cmd_ret(&dcc->root, start,
+				&prev_dc, &next_dc, &insert_p, &insert_parent);
 	if (!dc)
 		dc = next_dc;
 
 	blk_start_plug(&plug);
 
-	while (dc && dc->lstart <= end) {
+	while (dc && dc->di.lstart <= end) {
 		struct rb_node *node;
 		int err = 0;
 
-		if (dc->len < dpolicy->granularity)
+		if (dc->di.len < dpolicy->granularity)
 			goto skip;
 
 		if (dc->state != D_PREP) {
@@ -2992,7 +3073,7 @@ next:
 		err = __submit_discard_cmd(sbi, dpolicy, dc, &issued);
 
 		if (issued >= dpolicy->max_requests) {
-			start = dc->lstart + dc->len;
+			start = dc->di.lstart + dc->di.len;
 
 			if (err)
 				__remove_discard_cmd(sbi, dc);
-- 
cgit v1.2.3


From 6797ebc4ac55e7539391a5155f8a27f614da6bc5 Mon Sep 17 00:00:00 2001
From: Yonggil Song <yonggil.song@samsung.com>
Date: Mon, 13 Mar 2023 18:48:25 +0900
Subject: f2fs: Fix discard bug on zoned block devices with 2MiB zone size

When using f2fs on a zoned block device with 2MiB zone size, IO errors
occurs because f2fs tries to write data to a zone that has not been reset.

The cause is that f2fs tries to discard multiple zones at once. This is
caused by a condition in f2fs_clear_prefree_segments that does not check
for zoned block devices when setting the discard range. This leads to
invalid reset commands and write pointer mismatches.

This patch fixes the zoned block device with 2MiB zone size to reset one
zone at a time.

Signed-off-by: Yonggil Song <yonggil.song@samsung.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

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

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index d135046108ef..67cb0805cdfa 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -2069,7 +2069,9 @@ void f2fs_clear_prefree_segments(struct f2fs_sb_info *sbi,
 					(end - 1) <= cpc->trim_end)
 				continue;
 
-		if (!f2fs_lfs_mode(sbi) || !__is_large_section(sbi)) {
+		/* Should cover 2MB zoned device for zone-based reset */
+		if (!f2fs_sb_has_blkzoned(sbi) &&
+		    (!f2fs_lfs_mode(sbi) || !__is_large_section(sbi))) {
 			f2fs_issue_discard(sbi, START_BLOCK(sbi, start),
 				(end - start) << sbi->log_blocks_per_seg);
 			continue;
-- 
cgit v1.2.3


From 0b37ed21e3367539b79284e0b0af2246ffcf0dca Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Tue, 21 Mar 2023 15:58:04 -0700
Subject: f2fs: apply zone capacity to all zone type

If we manage the zone capacity per zone type, it'll break the GC assumption.
And, the current logic complains valid block count mismatch.
Let's apply zone capacity to all zone type, if specified.

Fixes: de881df97768 ("f2fs: support zone capacity less than zone size")
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c | 65 ++++---------------------------------------------------
 fs/f2fs/segment.h |  3 +++
 2 files changed, 7 insertions(+), 61 deletions(-)

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

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 67cb0805cdfa..61394346febe 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -5003,48 +5003,6 @@ int f2fs_check_write_pointer(struct f2fs_sb_info *sbi)
 	return 0;
 }
 
-static bool is_conv_zone(struct f2fs_sb_info *sbi, unsigned int zone_idx,
-						unsigned int dev_idx)
-{
-	if (!bdev_is_zoned(FDEV(dev_idx).bdev))
-		return true;
-	return !test_bit(zone_idx, FDEV(dev_idx).blkz_seq);
-}
-
-/* Return the zone index in the given device */
-static unsigned int get_zone_idx(struct f2fs_sb_info *sbi, unsigned int secno,
-					int dev_idx)
-{
-	block_t sec_start_blkaddr = START_BLOCK(sbi, GET_SEG_FROM_SEC(sbi, secno));
-
-	return (sec_start_blkaddr - FDEV(dev_idx).start_blk) >>
-						sbi->log_blocks_per_blkz;
-}
-
-/*
- * Return the usable segments in a section based on the zone's
- * corresponding zone capacity. Zone is equal to a section.
- */
-static inline unsigned int f2fs_usable_zone_segs_in_sec(
-		struct f2fs_sb_info *sbi, unsigned int segno)
-{
-	unsigned int dev_idx, zone_idx;
-
-	dev_idx = f2fs_target_device_index(sbi, START_BLOCK(sbi, segno));
-	zone_idx = get_zone_idx(sbi, GET_SEC_FROM_SEG(sbi, segno), dev_idx);
-
-	/* Conventional zone's capacity is always equal to zone size */
-	if (is_conv_zone(sbi, zone_idx, dev_idx))
-		return sbi->segs_per_sec;
-
-	if (!sbi->unusable_blocks_per_sec)
-		return sbi->segs_per_sec;
-
-	/* Get the segment count beyond zone capacity block */
-	return sbi->segs_per_sec - (sbi->unusable_blocks_per_sec >>
-						sbi->log_blocks_per_seg);
-}
-
 /*
  * Return the number of usable blocks in a segment. The number of blocks
  * returned is always equal to the number of blocks in a segment for
@@ -5057,23 +5015,13 @@ static inline unsigned int f2fs_usable_zone_blks_in_seg(
 			struct f2fs_sb_info *sbi, unsigned int segno)
 {
 	block_t seg_start, sec_start_blkaddr, sec_cap_blkaddr;
-	unsigned int zone_idx, dev_idx, secno;
-
-	secno = GET_SEC_FROM_SEG(sbi, segno);
-	seg_start = START_BLOCK(sbi, segno);
-	dev_idx = f2fs_target_device_index(sbi, seg_start);
-	zone_idx = get_zone_idx(sbi, secno, dev_idx);
-
-	/*
-	 * Conventional zone's capacity is always equal to zone size,
-	 * so, blocks per segment is unchanged.
-	 */
-	if (is_conv_zone(sbi, zone_idx, dev_idx))
-		return sbi->blocks_per_seg;
+	unsigned int secno;
 
 	if (!sbi->unusable_blocks_per_sec)
 		return sbi->blocks_per_seg;
 
+	secno = GET_SEC_FROM_SEG(sbi, segno);
+	seg_start = START_BLOCK(sbi, segno);
 	sec_start_blkaddr = START_BLOCK(sbi, GET_SEG_FROM_SEC(sbi, secno));
 	sec_cap_blkaddr = sec_start_blkaddr + CAP_BLKS_PER_SEC(sbi);
 
@@ -5107,11 +5055,6 @@ static inline unsigned int f2fs_usable_zone_blks_in_seg(struct f2fs_sb_info *sbi
 	return 0;
 }
 
-static inline unsigned int f2fs_usable_zone_segs_in_sec(struct f2fs_sb_info *sbi,
-							unsigned int segno)
-{
-	return 0;
-}
 #endif
 unsigned int f2fs_usable_blks_in_seg(struct f2fs_sb_info *sbi,
 					unsigned int segno)
@@ -5126,7 +5069,7 @@ unsigned int f2fs_usable_segs_in_sec(struct f2fs_sb_info *sbi,
 					unsigned int segno)
 {
 	if (f2fs_sb_has_blkzoned(sbi))
-		return f2fs_usable_zone_segs_in_sec(sbi, segno);
+		return CAP_SEGS_PER_SEC(sbi);
 
 	return sbi->segs_per_sec;
 }
diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index efdb7fc3b797..babb29a1c034 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -104,6 +104,9 @@ static inline void sanity_check_seg_type(struct f2fs_sb_info *sbi,
 #define CAP_BLKS_PER_SEC(sbi)					\
 	((sbi)->segs_per_sec * (sbi)->blocks_per_seg -		\
 	 (sbi)->unusable_blocks_per_sec)
+#define CAP_SEGS_PER_SEC(sbi)					\
+	((sbi)->segs_per_sec - ((sbi)->unusable_blocks_per_sec >>\
+	(sbi)->log_blocks_per_seg))
 #define GET_SEC_FROM_SEG(sbi, segno)				\
 	(((segno) == -1) ? -1: (segno) / (sbi)->segs_per_sec)
 #define GET_SEG_FROM_SEC(sbi, secno)				\
-- 
cgit v1.2.3


From f26aaee60a9f14035c1a659b5b00a6c8767b4126 Mon Sep 17 00:00:00 2001
From: Yohan Joung <jyh429@gmail.com>
Date: Sun, 2 Apr 2023 12:12:59 +0900
Subject: f2fs: fix align check for npo2

Fix alignment check to be correct in npo2 as well

Signed-off-by: Yohan Joung <yohan.joung@sk.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

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

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 61394346febe..f88828987924 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -1845,6 +1845,7 @@ static int __f2fs_issue_discard_zone(struct f2fs_sb_info *sbi,
 	sector_t sector, nr_sects;
 	block_t lblkstart = blkstart;
 	int devi = 0;
+	u64 remainder = 0;
 
 	if (f2fs_is_multi_device(sbi)) {
 		devi = f2fs_target_device_index(sbi, blkstart);
@@ -1860,9 +1861,9 @@ static int __f2fs_issue_discard_zone(struct f2fs_sb_info *sbi,
 	if (f2fs_blkz_is_seq(sbi, devi, blkstart)) {
 		sector = SECTOR_FROM_BLOCK(blkstart);
 		nr_sects = SECTOR_FROM_BLOCK(blklen);
+		div64_u64_rem(sector, bdev_zone_sectors(bdev), &remainder);
 
-		if (sector & (bdev_zone_sectors(bdev) - 1) ||
-				nr_sects != bdev_zone_sectors(bdev)) {
+		if (remainder || nr_sects != bdev_zone_sectors(bdev)) {
 			f2fs_err(sbi, "(%d) %s: Unaligned zone reset attempted (block %x + %x)",
 				 devi, sbi->s_ndevs ? FDEV(devi).path : "",
 				 blkstart, blklen);
-- 
cgit v1.2.3


From da6ea0b050fa720302b56fbb59307e7c7531a342 Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Mon, 3 Apr 2023 09:37:24 -0700
Subject: f2fs: fix null pointer panic in tracepoint in
 __replace_atomic_write_block

We got a kernel panic if old_addr is NULL.

https://bugzilla.kernel.org/show_bug.cgi?id=217266

BUG: kernel NULL pointer dereference, address: 0000000000000000
 Call Trace:
  <TASK>
  f2fs_commit_atomic_write+0x619/0x990 [f2fs a1b985b80f5babd6f3ea778384908880812bfa43]
  __f2fs_ioctl+0xd8e/0x4080 [f2fs a1b985b80f5babd6f3ea778384908880812bfa43]
  ? vfs_write+0x2ae/0x3f0
  ? vfs_write+0x2ae/0x3f0
  __x64_sys_ioctl+0x91/0xd0
  do_syscall_64+0x5c/0x90
  entry_SYSCALL_64_after_hwframe+0x72/0xdc
 RIP: 0033:0x7f69095fe53f

Fixes: 2f3a9ae990a7 ("f2fs: introduce trace_f2fs_replace_atomic_write_block")
Cc: <stable@vger.kernel.org>
Reviewed-by: Chao Yu <chao@kernel.org>
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 f88828987924..45128694eefa 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -257,7 +257,7 @@ retry:
 	f2fs_put_dnode(&dn);
 
 	trace_f2fs_replace_atomic_write_block(inode, F2FS_I(inode)->cow_inode,
-					index, *old_addr, new_addr, recover);
+			index, old_addr ? *old_addr : 0, new_addr, recover);
 	return 0;
 }
 
-- 
cgit v1.2.3


From 19e0e21a51183d4e0784602f27e4db7b965077be Mon Sep 17 00:00:00 2001
From: Yangtao Li <frank.li@vivo.com>
Date: Tue, 4 Apr 2023 12:00:51 +0800
Subject: f2fs: remove struct victim_selection default_v_ops

There is only single instance of these ops, and Jaegeuk point out that:

    Originally this was intended to give a chance to provide other
    allocation option. Anyway, it seems quit hard to do it anymore.

So remove the indirection and call f2fs_get_victim() directly.

Signed-off-by: Yangtao Li <frank.li@vivo.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/f2fs.h    |  4 ++++
 fs/f2fs/gc.c      | 15 ++++-----------
 fs/f2fs/segment.c |  5 ++---
 fs/f2fs/segment.h |  7 -------
 4 files changed, 10 insertions(+), 21 deletions(-)

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

diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 4e2596dacbf1..c2516e3273ad 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -3815,6 +3815,10 @@ void f2fs_build_gc_manager(struct f2fs_sb_info *sbi);
 int f2fs_resize_fs(struct f2fs_sb_info *sbi, __u64 block_count);
 int __init f2fs_create_garbage_collection_cache(void);
 void f2fs_destroy_garbage_collection_cache(void);
+/* victim selection function for cleaning and SSR */
+int f2fs_get_victim(struct f2fs_sb_info *sbi, unsigned int *result,
+			int gc_type, int type, char alloc_mode,
+			unsigned long long age);
 
 /*
  * recovery.c
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index 2996d38aa89c..5261b5b5e8d1 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -741,9 +741,9 @@ static int f2fs_gc_pinned_control(struct inode *inode, int gc_type,
  * When it is called from SSR segment selection, it finds a segment
  * which has minimum valid blocks and removes it from dirty seglist.
  */
-static int get_victim_by_default(struct f2fs_sb_info *sbi,
-			unsigned int *result, int gc_type, int type,
-			char alloc_mode, unsigned long long age)
+int f2fs_get_victim(struct f2fs_sb_info *sbi, unsigned int *result,
+			int gc_type, int type, char alloc_mode,
+			unsigned long long age)
 {
 	struct dirty_seglist_info *dirty_i = DIRTY_I(sbi);
 	struct sit_info *sm = SIT_I(sbi);
@@ -937,10 +937,6 @@ out:
 	return ret;
 }
 
-static const struct victim_selection default_v_ops = {
-	.get_victim = get_victim_by_default,
-};
-
 static struct inode *find_gc_inode(struct gc_inode_list *gc_list, nid_t ino)
 {
 	struct inode_entry *ie;
@@ -1671,8 +1667,7 @@ static int __get_victim(struct f2fs_sb_info *sbi, unsigned int *victim,
 	int ret;
 
 	down_write(&sit_i->sentry_lock);
-	ret = DIRTY_I(sbi)->v_ops->get_victim(sbi, victim, gc_type,
-					      NO_CHECK_TYPE, LFS, 0);
+	ret = f2fs_get_victim(sbi, victim, gc_type, NO_CHECK_TYPE, LFS, 0);
 	up_write(&sit_i->sentry_lock);
 	return ret;
 }
@@ -1963,8 +1958,6 @@ static void init_atgc_management(struct f2fs_sb_info *sbi)
 
 void f2fs_build_gc_manager(struct f2fs_sb_info *sbi)
 {
-	DIRTY_I(sbi)->v_ops = &default_v_ops;
-
 	sbi->gc_pin_file_threshold = DEF_GC_FAILED_PINNED_FILES;
 
 	/* give warm/cold data area from slower device */
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 45128694eefa..fc0734e4fb92 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -2875,7 +2875,6 @@ static int get_ssr_segment(struct f2fs_sb_info *sbi, int type,
 				int alloc_mode, unsigned long long age)
 {
 	struct curseg_info *curseg = CURSEG_I(sbi, type);
-	const struct victim_selection *v_ops = DIRTY_I(sbi)->v_ops;
 	unsigned segno = NULL_SEGNO;
 	unsigned short seg_type = curseg->seg_type;
 	int i, cnt;
@@ -2884,7 +2883,7 @@ static int get_ssr_segment(struct f2fs_sb_info *sbi, int type,
 	sanity_check_seg_type(sbi, seg_type);
 
 	/* f2fs_need_SSR() already forces to do this */
-	if (!v_ops->get_victim(sbi, &segno, BG_GC, seg_type, alloc_mode, age)) {
+	if (!f2fs_get_victim(sbi, &segno, BG_GC, seg_type, alloc_mode, age)) {
 		curseg->next_segno = segno;
 		return 1;
 	}
@@ -2911,7 +2910,7 @@ static int get_ssr_segment(struct f2fs_sb_info *sbi, int type,
 	for (; cnt-- > 0; reversed ? i-- : i++) {
 		if (i == seg_type)
 			continue;
-		if (!v_ops->get_victim(sbi, &segno, BG_GC, i, alloc_mode, age)) {
+		if (!f2fs_get_victim(sbi, &segno, BG_GC, i, alloc_mode, age)) {
 			curseg->next_segno = segno;
 			return 1;
 		}
diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index babb29a1c034..99e34d32c5c6 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -289,7 +289,6 @@ enum dirty_type {
 };
 
 struct dirty_seglist_info {
-	const struct victim_selection *v_ops;	/* victim selction operation */
 	unsigned long *dirty_segmap[NR_DIRTY_TYPE];
 	unsigned long *dirty_secmap;
 	struct mutex seglist_lock;		/* lock for segment bitmaps */
@@ -300,12 +299,6 @@ struct dirty_seglist_info {
 	bool enable_pin_section;		/* enable pinning section */
 };
 
-/* victim selection function for cleaning and SSR */
-struct victim_selection {
-	int (*get_victim)(struct f2fs_sb_info *, unsigned int *,
-					int, int, char, unsigned long long);
-};
-
 /* for active log information */
 struct curseg_info {
 	struct mutex curseg_mutex;		/* lock for consistency */
-- 
cgit v1.2.3


From 1ac3d037bee9ac3c4ec4041aafb54c5dffd7e1d0 Mon Sep 17 00:00:00 2001
From: Daeho Jeong <daehojeong@google.com>
Date: Thu, 6 Apr 2023 15:11:04 -0700
Subject: f2fs: fix passing relative address when discard zones

We should not pass relative address in a zone to
__f2fs_issue_discard_zone().

Signed-off-by: Daeho Jeong <daehojeong@google.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

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

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index fc0734e4fb92..84e2e558ea99 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -4942,9 +4942,8 @@ static int fix_curseg_write_pointer(struct f2fs_sb_info *sbi, int type)
 			    "New zone for curseg[%d] is not yet discarded. "
 			    "Reset the zone: curseg[0x%x,0x%x]",
 			    type, cs->segno, cs->next_blkoff);
-		err = __f2fs_issue_discard_zone(sbi, zbd->bdev,
-				zone_sector >> log_sectors_per_block,
-				zone.len >> log_sectors_per_block);
+		err = __f2fs_issue_discard_zone(sbi, zbd->bdev,	cs_zone_block,
+					zone.len >> log_sectors_per_block);
 		if (err) {
 			f2fs_err(sbi, "Discard zone failed: %s (errno=%d)",
 				 zbd->path, err);
-- 
cgit v1.2.3


From 935fc6fa6466cf18dd72dd1518ebc7bfc4cd58a4 Mon Sep 17 00:00:00 2001
From: Chao Yu <chao@kernel.org>
Date: Wed, 5 Apr 2023 22:45:36 +0800
Subject: f2fs: fix to check return value of inc_valid_block_count()

In __replace_atomic_write_block(), we missed to check return value
of inc_valid_block_count(), for extreme testcase that f2fs image is
run out of space, it may cause inconsistent status in between SIT
table and total valid block count.

Cc: Daeho Jeong <daehojeong@google.com>
Fixes: 3db1de0e582c ("f2fs: change the current atomic write way")
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/segment.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

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

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 84e2e558ea99..82430f80c5da 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -246,10 +246,16 @@ retry:
 	} else {
 		blkcnt_t count = 1;
 
+		err = inc_valid_block_count(sbi, inode, &count);
+		if (err) {
+			f2fs_put_dnode(&dn);
+			return err;
+		}
+
 		*old_addr = dn.data_blkaddr;
 		f2fs_truncate_data_blocks_range(&dn, 1);
 		dec_valid_block_count(sbi, F2FS_I(inode)->cow_inode, count);
-		inc_valid_block_count(sbi, inode, &count);
+
 		f2fs_replace_block(sbi, &dn, dn.data_blkaddr, new_addr,
 					ni.version, true, false);
 	}
-- 
cgit v1.2.3


From c1660d88a064409879f6d467754bbe27259c71bb Mon Sep 17 00:00:00 2001
From: Yangtao Li <frank.li@vivo.com>
Date: Fri, 14 Apr 2023 00:59:51 +0800
Subject: f2fs: add has_enough_free_secs()

Replace !has_not_enough_free_secs w/ has_enough_free_secs.
BTW avoid nested 'if' statements in f2fs_balance_fs().

Signed-off-by: Yangtao Li <frank.li@vivo.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/gc.c      |  4 ++--
 fs/f2fs/segment.c | 43 ++++++++++++++++++++++---------------------
 fs/f2fs/segment.h |  8 +++++++-
 3 files changed, 31 insertions(+), 24 deletions(-)

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

diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index ba5775dcade6..a6a0dc471b74 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -1872,7 +1872,7 @@ retry:
 	if (gc_type == FG_GC) {
 		sbi->cur_victim_sec = NULL_SEGNO;
 
-		if (!has_not_enough_free_secs(sbi, sec_freed, 0)) {
+		if (has_enough_free_secs(sbi, sec_freed, 0)) {
 			if (!gc_control->no_bg_gc &&
 			    sec_freed < gc_control->nr_free_secs)
 				goto go_gc_more;
@@ -1886,7 +1886,7 @@ retry:
 			ret = f2fs_write_checkpoint(sbi, &cpc);
 			goto stop;
 		}
-	} else if (!has_not_enough_free_secs(sbi, 0, 0)) {
+	} else if (has_enough_free_secs(sbi, 0, 0)) {
 		goto stop;
 	}
 
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index 82430f80c5da..c35476b3c075 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -412,27 +412,28 @@ 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, 0)) {
-		if (test_opt(sbi, GC_MERGE) && sbi->gc_thread &&
-					sbi->gc_thread->f2fs_gc_task) {
-			DEFINE_WAIT(wait);
-
-			prepare_to_wait(&sbi->gc_thread->fggc_wq, &wait,
-						TASK_UNINTERRUPTIBLE);
-			wake_up(&sbi->gc_thread->gc_wait_queue_head);
-			io_schedule();
-			finish_wait(&sbi->gc_thread->fggc_wq, &wait);
-		} else {
-			struct f2fs_gc_control gc_control = {
-				.victim_segno = NULL_SEGNO,
-				.init_gc_type = BG_GC,
-				.no_bg_gc = true,
-				.should_migrate_blocks = false,
-				.err_gc_skipped = false,
-				.nr_free_secs = 1 };
-			f2fs_down_write(&sbi->gc_lock);
-			f2fs_gc(sbi, &gc_control);
-		}
+	if (has_enough_free_secs(sbi, 0, 0))
+		return;
+
+	if (test_opt(sbi, GC_MERGE) && sbi->gc_thread &&
+				sbi->gc_thread->f2fs_gc_task) {
+		DEFINE_WAIT(wait);
+
+		prepare_to_wait(&sbi->gc_thread->fggc_wq, &wait,
+					TASK_UNINTERRUPTIBLE);
+		wake_up(&sbi->gc_thread->gc_wait_queue_head);
+		io_schedule();
+		finish_wait(&sbi->gc_thread->fggc_wq, &wait);
+	} else {
+		struct f2fs_gc_control gc_control = {
+			.victim_segno = NULL_SEGNO,
+			.init_gc_type = BG_GC,
+			.no_bg_gc = true,
+			.should_migrate_blocks = false,
+			.err_gc_skipped = false,
+			.nr_free_secs = 1 };
+		f2fs_down_write(&sbi->gc_lock);
+		f2fs_gc(sbi, &gc_control);
 	}
 }
 
diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index ac2e35170f2d..2ca8fb5d0dc4 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -643,11 +643,17 @@ static inline bool has_not_enough_free_secs(struct f2fs_sb_info *sbi,
 	return !curseg_space;
 }
 
+static inline bool has_enough_free_secs(struct f2fs_sb_info *sbi,
+					int freed, int needed)
+{
+	return !has_not_enough_free_secs(sbi, freed, needed);
+}
+
 static inline bool f2fs_is_checkpoint_ready(struct f2fs_sb_info *sbi)
 {
 	if (likely(!is_sbi_flag_set(sbi, SBI_CP_DISABLED)))
 		return true;
-	if (likely(!has_not_enough_free_secs(sbi, 0, 0)))
+	if (likely(has_enough_free_secs(sbi, 0, 0)))
 		return true;
 	return false;
 }
-- 
cgit v1.2.3


From 2e2c6e9b72ce3d137984c867eb5625b8498cfe2b Mon Sep 17 00:00:00 2001
From: Jaegeuk Kim <jaegeuk@kernel.org>
Date: Mon, 17 Apr 2023 17:12:52 -0700
Subject: f2fs: remove power-of-two limitation of zoned device

In f2fs, there's no reason to force po2.

Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
 fs/f2fs/f2fs.h    | 3 +--
 fs/f2fs/gc.c      | 4 ++--
 fs/f2fs/segment.c | 2 +-
 fs/f2fs/super.c   | 8 ++------
 4 files changed, 6 insertions(+), 11 deletions(-)

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

diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index d87044516fe9..d211ee89c158 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -1523,7 +1523,6 @@ struct f2fs_sb_info {
 
 #ifdef CONFIG_BLK_DEV_ZONED
 	unsigned int blocks_per_blkz;		/* F2FS blocks per zone */
-	unsigned int log_blocks_per_blkz;	/* log2 F2FS blocks per zone */
 #endif
 
 	/* for node-related operations */
@@ -4390,7 +4389,7 @@ F2FS_FEATURE_FUNCS(readonly, RO);
 static inline bool f2fs_blkz_is_seq(struct f2fs_sb_info *sbi, int devi,
 				    block_t blkaddr)
 {
-	unsigned int zno = blkaddr >> sbi->log_blocks_per_blkz;
+	unsigned int zno = blkaddr / sbi->blocks_per_blkz;
 
 	return test_bit(zno, FDEV(devi).blkz_seq);
 }
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index a6a0dc471b74..61c5f9d26018 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -2093,8 +2093,8 @@ static void update_fs_metadata(struct f2fs_sb_info *sbi, int secs)
 		FDEV(last_dev).end_blk =
 				(long long)FDEV(last_dev).end_blk + blks;
 #ifdef CONFIG_BLK_DEV_ZONED
-		FDEV(last_dev).nr_blkz = (int)FDEV(last_dev).nr_blkz +
-					(int)(blks >> sbi->log_blocks_per_blkz);
+		FDEV(last_dev).nr_blkz = FDEV(last_dev).nr_blkz +
+					div_u64(blks, sbi->blocks_per_blkz);
 #endif
 	}
 }
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index c35476b3c075..641bd8cc54da 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -2075,7 +2075,7 @@ void f2fs_clear_prefree_segments(struct f2fs_sb_info *sbi,
 
 		if (force && start >= cpc->trim_start &&
 					(end - 1) <= cpc->trim_end)
-				continue;
+			continue;
 
 		/* Should cover 2MB zoned device for zone-based reset */
 		if (!f2fs_sb_has_blkzoned(sbi) &&
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 4e53b1100b84..9f15b03037db 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -3806,12 +3806,8 @@ static int init_blkz_info(struct f2fs_sb_info *sbi, int devi)
 				SECTOR_TO_BLOCK(zone_sectors))
 		return -EINVAL;
 	sbi->blocks_per_blkz = SECTOR_TO_BLOCK(zone_sectors);
-	if (sbi->log_blocks_per_blkz && sbi->log_blocks_per_blkz !=
-				__ilog2_u32(sbi->blocks_per_blkz))
-		return -EINVAL;
-	sbi->log_blocks_per_blkz = __ilog2_u32(sbi->blocks_per_blkz);
-	FDEV(devi).nr_blkz = SECTOR_TO_BLOCK(nr_sectors) >>
-					sbi->log_blocks_per_blkz;
+	FDEV(devi).nr_blkz = div_u64(SECTOR_TO_BLOCK(nr_sectors),
+					sbi->blocks_per_blkz);
 	if (nr_sectors & (zone_sectors - 1))
 		FDEV(devi).nr_blkz++;
 
-- 
cgit v1.2.3


From 994b442b66a4624a2f471aba5cbae37a79213313 Mon Sep 17 00:00:00 2001
From: Daeho Jeong <daehojeong@google.com>
Date: Tue, 18 Apr 2023 10:42:01 -0700
Subject: f2fs: allocate node blocks for atomic write block replacement

When a node block is missing for atomic write block replacement, we need
to allocate it in advance of the replacement.

Signed-off-by: Daeho Jeong <daehojeong@google.com>
Reviewed-by: Chao Yu <chao@kernel.org>
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 641bd8cc54da..6db410f1bb8c 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -217,7 +217,7 @@ static int __replace_atomic_write_block(struct inode *inode, pgoff_t index,
 
 retry:
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
-	err = f2fs_get_dnode_of_data(&dn, index, LOOKUP_NODE_RA);
+	err = f2fs_get_dnode_of_data(&dn, index, ALLOC_NODE);
 	if (err) {
 		if (err == -ENOMEM) {
 			f2fs_io_schedule_timeout(DEFAULT_IO_TIMEOUT);
-- 
cgit v1.2.3