selftests/bpf: Reject graph kptr use after RCU unlockgrafted

Add a sleepable verifier test that loads a graph-node local kptr in an
explicit RCU read-side critical section, then passes its node to
bpf_rbtree_remove() after the section ends.

Before the verifier fix, the stale NON_OWN_REF flag makes the node look like
a live borrowed reference and the program is accepted. After the fix, the
pointer is demoted without NON_OWN_REF and the graph kfunc argument is
rejected.

Also exercise a graph kptr loaded while a spin lock provides implicit RCU
protection. The pointer must be invalidated when the lock is released, which
guards the required ordering between non-owning-reference invalidation and
RCU demotion.

Update the existing fault-protected load test state description. The
post-unlock pointer no longer carries NON_OWN_REF, but remains readable
because the load is rewritten to use BPF_PROBE_MEM.

Signed-off-by: Kumar Kartikeya Dwivedi <memxor@gmail.com>
Link: https://lore.kernel.org/r/20260904084325.52250-7-memxor@gmail.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

1 files changed, 970 insertions, 0 deletions

diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
new file mode 100644
index 000000000..89af8407b
--- /dev/null
+++ b/fs/f2fs/recovery.c
@@ -0,0 +1,970 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * fs/f2fs/recovery.c
+ *
+ * Copyright (c) 2012 Samsung Electronics Co., Ltd.
+ *             http://www.samsung.com/
+ */
+#include <linux/unaligned.h>
+#include <linux/fs.h>
+#include <linux/f2fs_fs.h>
+#include <linux/sched/mm.h>
+#include <linux/fserror.h>
+#include "f2fs.h"
+#include "node.h"
+#include "segment.h"
+
+/*
+ * Roll forward recovery scenarios.
+ *
+ * [Term] F: fsync_mark, D: dentry_mark
+ *
+ * 1. inode(x) | CP | inode(x) | dnode(F)
+ * -> Update the latest inode(x).
+ *
+ * 2. inode(x) | CP | inode(F) | dnode(F)
+ * -> No problem.
+ *
+ * 3. inode(x) | CP | dnode(F) | inode(x)
+ * -> Recover to the latest dnode(F), and drop the last inode(x)
+ *
+ * 4. inode(x) | CP | dnode(F) | inode(F)
+ * -> No problem.
+ *
+ * 5. CP | inode(x) | dnode(F)
+ * -> The inode(DF) was missing. Should drop this dnode(F).
+ *
+ * 6. CP | inode(DF) | dnode(F)
+ * -> No problem.
+ *
+ * 7. CP | dnode(F) | inode(DF)
+ * -> If f2fs_iget fails, then goto next to find inode(DF).
+ *
+ * 8. CP | dnode(F) | inode(x)
+ * -> If f2fs_iget fails, then goto next to find inode(DF).
+ *    But it will fail due to no inode(DF).
+ */
+
+static struct kmem_cache *fsync_entry_slab;
+
+bool f2fs_space_for_roll_forward(struct f2fs_sb_info *sbi)
+{
+	s64 nalloc = percpu_counter_sum_positive(&sbi->alloc_valid_block_count);
+
+	if (sbi->last_valid_block_count + nalloc > sbi->user_block_count)
+		return false;
+	if (NM_I(sbi)->max_rf_node_blocks &&
+		percpu_counter_sum_positive(&sbi->rf_node_block_count) >=
+						NM_I(sbi)->max_rf_node_blocks)
+		return false;
+	return true;
+}
+
+static struct fsync_inode_entry *get_fsync_inode(struct list_head *head,
+								nid_t ino)
+{
+	struct fsync_inode_entry *entry;
+
+	list_for_each_entry(entry, head, list)
+		if (entry->inode->i_ino == ino)
+			return entry;
+
+	return NULL;
+}
+
+static struct fsync_inode_entry *add_fsync_inode(struct f2fs_sb_info *sbi,
+			struct list_head *head, nid_t ino, bool quota_inode)
+{
+	struct inode *inode;
+	struct fsync_inode_entry *entry;
+	int err;
+
+	inode = f2fs_iget_retry(sbi->sb, ino);
+	if (IS_ERR(inode))
+		return ERR_CAST(inode);
+
+	err = f2fs_dquot_initialize(inode);
+	if (err)
+		goto err_out;
+
+	if (quota_inode) {
+		err = dquot_alloc_inode(inode);
+		if (err)
+			goto err_out;
+	}
+
+	entry = f2fs_kmem_cache_alloc(fsync_entry_slab,
+					GFP_F2FS_ZERO, true, NULL);
+	entry->inode = inode;
+	list_add_tail(&entry->list, head);
+
+	return entry;
+err_out:
+	iput(inode);
+	return ERR_PTR(err);
+}
+
+static void del_fsync_inode(struct fsync_inode_entry *entry, int drop)
+{
+	if (drop) {
+		/* inode should not be recovered, drop it */
+		f2fs_inode_synced(entry->inode);
+	}
+	iput(entry->inode);
+	list_del(&entry->list);
+	kmem_cache_free(fsync_entry_slab, entry);
+}
+
+static int init_recovered_filename(const struct inode *dir,
+				   struct f2fs_inode *raw_inode,
+				   struct f2fs_filename *fname,
+				   struct qstr *usr_fname)
+{
+	int err;
+
+	memset(fname, 0, sizeof(*fname));
+	fname->disk_name.len = le32_to_cpu(raw_inode->i_namelen);
+	fname->disk_name.name = raw_inode->i_name;
+
+	if (WARN_ON(fname->disk_name.len > F2FS_NAME_LEN))
+		return -ENAMETOOLONG;
+
+	if (!IS_ENCRYPTED(dir)) {
+		usr_fname->name = fname->disk_name.name;
+		usr_fname->len = fname->disk_name.len;
+		fname->usr_fname = usr_fname;
+	}
+
+	/* Compute the hash of the filename */
+	if (IS_ENCRYPTED(dir) && IS_CASEFOLDED(dir)) {
+		/*
+		 * In this case the hash isn't computable without the key, so it
+		 * was saved on-disk.
+		 */
+		if (fname->disk_name.len + sizeof(f2fs_hash_t) > F2FS_NAME_LEN)
+			return -EINVAL;
+		fname->hash = get_unaligned((f2fs_hash_t *)
+				&raw_inode->i_name[fname->disk_name.len]);
+	} else if (IS_CASEFOLDED(dir)) {
+		err = f2fs_init_casefolded_name(dir, fname);
+		if (err)
+			return err;
+		f2fs_hash_filename(dir, fname);
+		/* Case-sensitive match is fine for recovery */
+		f2fs_free_casefolded_name(fname);
+	} else {
+		f2fs_hash_filename(dir, fname);
+	}
+	return 0;
+}
+
+static int recover_dentry(struct inode *inode, struct folio *ifolio,
+						struct list_head *dir_list)
+{
+	struct f2fs_inode *raw_inode = F2FS_INODE(ifolio);
+	nid_t pino = le32_to_cpu(raw_inode->i_pino);
+	struct f2fs_dir_entry *de;
+	struct f2fs_filename fname;
+	struct qstr usr_fname;
+	struct folio *folio;
+	struct inode *dir, *einode;
+	struct fsync_inode_entry *entry;
+	int err = 0;
+	char *name;
+
+	entry = get_fsync_inode(dir_list, pino);
+	if (!entry) {
+		entry = add_fsync_inode(F2FS_I_SB(inode), dir_list,
+							pino, false);
+		if (IS_ERR(entry)) {
+			dir = ERR_CAST(entry);
+			err = PTR_ERR(entry);
+			goto out;
+		}
+	}
+
+	dir = entry->inode;
+	err = init_recovered_filename(dir, raw_inode, &fname, &usr_fname);
+	if (err)
+		goto out;
+retry:
+	de = __f2fs_find_entry(dir, &fname, &folio);
+	if (de && inode->i_ino == le32_to_cpu(de->ino))
+		goto out_put;
+
+	if (de) {
+		einode = f2fs_iget_retry(inode->i_sb, le32_to_cpu(de->ino));
+		if (IS_ERR(einode)) {
+			WARN_ON(1);
+			err = PTR_ERR(einode);
+			if (err == -ENOENT)
+				err = -EEXIST;
+			goto out_put;
+		}
+
+		err = f2fs_dquot_initialize(einode);
+		if (err) {
+			iput(einode);
+			goto out_put;
+		}
+
+		err = f2fs_acquire_orphan_inode(F2FS_I_SB(inode));
+		if (err) {
+			iput(einode);
+			goto out_put;
+		}
+		f2fs_delete_entry(de, folio, dir, einode);
+		iput(einode);
+		goto retry;
+	} else if (IS_ERR(folio)) {
+		err = PTR_ERR(folio);
+	} else {
+		err = f2fs_add_dentry(dir, &fname, inode,
+					inode->i_ino, inode->i_mode);
+	}
+	if (err == -ENOMEM)
+		goto retry;
+	goto out;
+
+out_put:
+	f2fs_folio_put(folio, false);
+out:
+	if (file_enc_name(inode))
+		name = "<encrypted>";
+	else
+		name = raw_inode->i_name;
+	f2fs_notice(F2FS_I_SB(inode), "%s: ino = %x, name = %s, dir = %llu, err = %d",
+		    __func__, ino_of_node(ifolio), name,
+		    IS_ERR(dir) ? 0 : dir->i_ino, err);
+	return err;
+}
+
+static int recover_quota_data(struct inode *inode, struct folio *folio)
+{
+	struct f2fs_inode *raw = F2FS_INODE(folio);
+	struct iattr attr;
+	uid_t i_uid = le32_to_cpu(raw->i_uid);
+	gid_t i_gid = le32_to_cpu(raw->i_gid);
+	int err;
+
+	memset(&attr, 0, sizeof(attr));
+
+	attr.ia_vfsuid = VFSUIDT_INIT(make_kuid(inode->i_sb->s_user_ns, i_uid));
+	attr.ia_vfsgid = VFSGIDT_INIT(make_kgid(inode->i_sb->s_user_ns, i_gid));
+
+	if (!vfsuid_eq(attr.ia_vfsuid, i_uid_into_vfsuid(&nop_mnt_idmap, inode)))
+		attr.ia_valid |= ATTR_UID;
+	if (!vfsgid_eq(attr.ia_vfsgid, i_gid_into_vfsgid(&nop_mnt_idmap, inode)))
+		attr.ia_valid |= ATTR_GID;
+
+	if (!attr.ia_valid)
+		return 0;
+
+	err = dquot_transfer(&nop_mnt_idmap, inode, &attr);
+	if (err)
+		set_sbi_flag(F2FS_I_SB(inode), SBI_QUOTA_NEED_REPAIR);
+	return err;
+}
+
+static void recover_inline_flags(struct inode *inode, struct f2fs_inode *ri)
+{
+	if (ri->i_inline & F2FS_PIN_FILE)
+		set_inode_flag(inode, FI_PIN_FILE);
+	else
+		clear_inode_flag(inode, FI_PIN_FILE);
+	if (ri->i_inline & F2FS_DATA_EXIST)
+		set_inode_flag(inode, FI_DATA_EXIST);
+	else
+		clear_inode_flag(inode, FI_DATA_EXIST);
+}
+
+static int recover_inode(struct inode *inode, struct folio *folio)
+{
+	struct f2fs_inode *raw = F2FS_INODE(folio);
+	struct f2fs_inode_info *fi = F2FS_I(inode);
+	char *name;
+	int err;
+
+	inode->i_mode = le16_to_cpu(raw->i_mode);
+
+	err = recover_quota_data(inode, folio);
+	if (err)
+		return err;
+
+	i_uid_write(inode, le32_to_cpu(raw->i_uid));
+	i_gid_write(inode, le32_to_cpu(raw->i_gid));
+
+	if (raw->i_inline & F2FS_EXTRA_ATTR) {
+		if (f2fs_sb_has_project_quota(F2FS_I_SB(inode)) &&
+			F2FS_FITS_IN_INODE(raw, le16_to_cpu(raw->i_extra_isize),
+								i_projid)) {
+			projid_t i_projid;
+			kprojid_t kprojid;
+
+			i_projid = (projid_t)le32_to_cpu(raw->i_projid);
+			kprojid = make_kprojid(&init_user_ns, i_projid);
+
+			if (!projid_eq(kprojid, fi->i_projid)) {
+				err = f2fs_transfer_project_quota(inode,
+								kprojid);
+				if (err)
+					return err;
+				fi->i_projid = kprojid;
+			}
+		}
+	}
+
+	f2fs_i_size_write(inode, le64_to_cpu(raw->i_size));
+	inode_set_atime(inode, le64_to_cpu(raw->i_atime),
+			le32_to_cpu(raw->i_atime_nsec));
+	inode_set_ctime(inode, le64_to_cpu(raw->i_ctime),
+			le32_to_cpu(raw->i_ctime_nsec));
+	inode_set_mtime(inode, le64_to_cpu(raw->i_mtime),
+			le32_to_cpu(raw->i_mtime_nsec));
+
+	fi->i_advise = raw->i_advise;
+	fi->i_flags = le32_to_cpu(raw->i_flags);
+	f2fs_set_inode_flags(inode);
+	fi->i_gc_failures = le16_to_cpu(raw->i_gc_failures);
+
+	recover_inline_flags(inode, raw);
+
+	f2fs_mark_inode_dirty_sync(inode, true);
+
+	if (file_enc_name(inode))
+		name = "<encrypted>";
+	else
+		name = F2FS_INODE(folio)->i_name;
+
+	f2fs_notice(F2FS_I_SB(inode), "recover_inode: ino = %x, name = %s, inline = %x",
+		    ino_of_node(folio), name, raw->i_inline);
+	return 0;
+}
+
+static unsigned int adjust_por_ra_blocks(struct f2fs_sb_info *sbi,
+				unsigned int ra_blocks, unsigned int blkaddr,
+				unsigned int next_blkaddr)
+{
+	if (blkaddr + 1 == next_blkaddr)
+		ra_blocks = min_t(unsigned int, RECOVERY_MAX_RA_BLOCKS,
+							ra_blocks * 2);
+	else if (next_blkaddr % BLKS_PER_SEG(sbi))
+		ra_blocks = max_t(unsigned int, RECOVERY_MIN_RA_BLOCKS,
+							ra_blocks / 2);
+	return ra_blocks;
+}
+
+/* Detect looped node chain with Floyd's cycle detection algorithm. */
+static int sanity_check_node_chain(struct f2fs_sb_info *sbi, block_t blkaddr,
+		block_t *blkaddr_fast, bool *is_detecting)
+{
+	unsigned int ra_blocks = RECOVERY_MAX_RA_BLOCKS;
+	int i;
+
+	if (!*is_detecting)
+		return 0;
+
+	for (i = 0; i < 2; i++) {
+		struct folio *folio;
+
+		if (!f2fs_is_valid_blkaddr(sbi, *blkaddr_fast, META_POR)) {
+			*is_detecting = false;
+			return 0;
+		}
+
+		folio = f2fs_get_tmp_folio(sbi, *blkaddr_fast);
+		if (IS_ERR(folio))
+			return PTR_ERR(folio);
+
+		if (!is_recoverable_dnode(folio)) {
+			f2fs_folio_put(folio, true);
+			*is_detecting = false;
+			return 0;
+		}
+
+		ra_blocks = adjust_por_ra_blocks(sbi, ra_blocks, *blkaddr_fast,
+					next_blkaddr_of_node(folio));
+
+		*blkaddr_fast = next_blkaddr_of_node(folio);
+		f2fs_folio_put(folio, true);
+
+		f2fs_ra_meta_pages_cond(sbi, *blkaddr_fast, ra_blocks);
+	}
+
+	if (*blkaddr_fast == blkaddr) {
+		f2fs_notice(sbi, "%s: Detect looped node chain on blkaddr:%u."
+				" Run fsck to fix it.", __func__, blkaddr);
+		return -EINVAL;
+	}
+	return 0;
+}
+
+static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head,
+				bool check_only, bool *new_inode)
+{
+	struct curseg_info *curseg;
+	block_t blkaddr, blkaddr_fast;
+	bool is_detecting = true;
+	int err = 0;
+
+	/* get node pages in the current segment */
+	curseg = CURSEG_I(sbi, CURSEG_WARM_NODE);
+	blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
+	blkaddr_fast = blkaddr;
+
+	while (1) {
+		struct fsync_inode_entry *entry;
+		struct folio *folio;
+
+		if (!f2fs_is_valid_blkaddr(sbi, blkaddr, META_POR))
+			return 0;
+
+		folio = f2fs_get_tmp_folio(sbi, blkaddr);
+		if (IS_ERR(folio)) {
+			err = PTR_ERR(folio);
+			break;
+		}
+
+		if (!is_recoverable_dnode(folio)) {
+			f2fs_folio_put(folio, true);
+			break;
+		}
+
+		if (!is_fsync_dnode(folio))
+			goto next;
+
+		entry = get_fsync_inode(head, ino_of_node(folio));
+		if (!entry) {
+			bool quota_inode = false;
+
+			if (!check_only &&
+					IS_INODE(folio) &&
+					is_dent_dnode(folio)) {
+				err = f2fs_recover_inode_page(sbi, folio);
+				if (err) {
+					f2fs_folio_put(folio, true);
+					break;
+				}
+				quota_inode = true;
+			}
+
+			entry = add_fsync_inode(sbi, head, ino_of_node(folio),
+								quota_inode);
+			if (IS_ERR(entry)) {
+				err = PTR_ERR(entry);
+				/*
+				 * CP | dnode(F) | inode(DF)
+				 * For this case, we should not give up now.
+				 */
+				if (err == -ENOENT) {
+					if (check_only)
+						*new_inode = true;
+					goto next;
+				}
+				f2fs_folio_put(folio, true);
+				break;
+			}
+		}
+		entry->blkaddr = blkaddr;
+
+		if (IS_INODE(folio) && is_dent_dnode(folio))
+			entry->last_dentry = blkaddr;
+next:
+		/* check next segment */
+		blkaddr = next_blkaddr_of_node(folio);
+		f2fs_folio_put(folio, true);
+
+		err = sanity_check_node_chain(sbi, blkaddr, &blkaddr_fast,
+				&is_detecting);
+		if (err)
+			break;
+	}
+	return err;
+}
+
+static void destroy_fsync_dnodes(struct list_head *head, int drop)
+{
+	struct fsync_inode_entry *entry, *tmp;
+
+	list_for_each_entry_safe(entry, tmp, head, list)
+		del_fsync_inode(entry, drop);
+}
+
+static int check_index_in_prev_nodes(struct f2fs_sb_info *sbi,
+			block_t blkaddr, struct dnode_of_data *dn)
+{
+	struct seg_entry *sentry;
+	unsigned int segno = GET_SEGNO(sbi, blkaddr);
+	unsigned short blkoff = GET_BLKOFF_FROM_SEG0(sbi, blkaddr);
+	struct f2fs_summary_block *sum_node;
+	struct f2fs_summary sum;
+	struct folio *sum_folio, *node_folio;
+	struct dnode_of_data tdn = *dn;
+	nid_t ino, nid;
+	struct inode *inode;
+	unsigned int offset, ofs_in_node, max_addrs;
+	block_t bidx;
+	int i;
+
+	sentry = get_seg_entry(sbi, segno);
+	if (!f2fs_test_bit(blkoff, sentry->cur_valid_map))
+		return 0;
+
+	/* Get the previous summary */
+	for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_DATA; i++) {
+		struct curseg_info *curseg = CURSEG_I(sbi, i);
+
+		if (curseg->segno == segno) {
+			sum = sum_entries(curseg->sum_blk)[blkoff];
+			goto got_it;
+		}
+	}
+
+	sum_folio = f2fs_get_sum_folio(sbi, segno);
+	if (IS_ERR(sum_folio))
+		return PTR_ERR(sum_folio);
+	sum_node = SUM_BLK_PAGE_ADDR(sbi, sum_folio, segno);
+	sum = sum_entries(sum_node)[blkoff];
+	f2fs_folio_put(sum_folio, true);
+got_it:
+	/* Use the locked dnode page and inode */
+	nid = le32_to_cpu(sum.nid);
+	ofs_in_node = le16_to_cpu(sum.ofs_in_node);
+
+	max_addrs = ADDRS_PER_PAGE(dn->node_folio, dn->inode);
+	if (ofs_in_node >= max_addrs) {
+		f2fs_err(sbi, "Inconsistent ofs_in_node:%u in summary, ino:%llu, nid:%u, max:%u",
+			ofs_in_node, dn->inode->i_ino, nid, max_addrs);
+		f2fs_handle_error(sbi, ERROR_INCONSISTENT_SUMMARY);
+		return -EFSCORRUPTED;
+	}
+
+	if (dn->inode->i_ino == nid) {
+		tdn.nid = nid;
+		if (!dn->inode_folio_locked)
+			folio_lock(dn->inode_folio);
+		tdn.node_folio = dn->inode_folio;
+		tdn.ofs_in_node = ofs_in_node;
+		goto truncate_out;
+	} else if (dn->nid == nid) {
+		tdn.ofs_in_node = ofs_in_node;
+		goto truncate_out;
+	}
+
+	/* Get the node page */
+	node_folio = f2fs_get_node_folio(sbi, nid, NODE_TYPE_REGULAR);
+	if (IS_ERR(node_folio))
+		return PTR_ERR(node_folio);
+
+	offset = ofs_of_node(node_folio);
+	ino = ino_of_node(node_folio);
+	f2fs_folio_put(node_folio, true);
+
+	if (ino != dn->inode->i_ino) {
+		int ret;
+
+		/* Deallocate previous index in the node page */
+		inode = f2fs_iget_retry(sbi->sb, ino);
+		if (IS_ERR(inode))
+			return PTR_ERR(inode);
+
+		ret = f2fs_dquot_initialize(inode);
+		if (ret) {
+			iput(inode);
+			return ret;
+		}
+	} else {
+		inode = dn->inode;
+	}
+
+	bidx = f2fs_start_bidx_of_node(offset, inode) +
+				le16_to_cpu(sum.ofs_in_node);
+
+	/*
+	 * if inode page is locked, unlock temporarily, but its reference
+	 * count keeps alive.
+	 */
+	if (ino == dn->inode->i_ino && dn->inode_folio_locked)
+		folio_unlock(dn->inode_folio);
+
+	set_new_dnode(&tdn, inode, NULL, NULL, 0);
+	if (f2fs_get_dnode_of_data(&tdn, bidx, LOOKUP_NODE))
+		goto out;
+
+	if (tdn.data_blkaddr == blkaddr)
+		f2fs_truncate_data_blocks_range(&tdn, 1);
+
+	f2fs_put_dnode(&tdn);
+out:
+	if (ino != dn->inode->i_ino)
+		iput(inode);
+	else if (dn->inode_folio_locked)
+		folio_lock(dn->inode_folio);
+	return 0;
+
+truncate_out:
+	if (f2fs_data_blkaddr(&tdn) == blkaddr)
+		f2fs_truncate_data_blocks_range(&tdn, 1);
+	if (dn->inode->i_ino == nid && !dn->inode_folio_locked)
+		folio_unlock(dn->inode_folio);
+	return 0;
+}
+
+static int f2fs_reserve_new_block_retry(struct dnode_of_data *dn)
+{
+	int i, err = 0;
+
+	for (i = DEFAULT_FAILURE_RETRY_COUNT; i > 0; i--) {
+		err = f2fs_reserve_new_block(dn);
+		if (!err)
+			break;
+	}
+
+	return err;
+}
+
+static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
+					struct folio *folio)
+{
+	struct dnode_of_data dn;
+	struct node_info ni;
+	unsigned int start = 0, end = 0, index;
+	int err = 0, recovered = 0;
+
+	/* step 1: recover xattr */
+	if (IS_INODE(folio)) {
+		err = f2fs_recover_inline_xattr(inode, folio);
+		if (err)
+			goto out;
+	} else if (f2fs_has_xattr_block(ofs_of_node(folio))) {
+		err = f2fs_recover_xattr_data(inode, folio);
+		if (!err)
+			recovered++;
+		goto out;
+	}
+
+	/* step 2: recover inline data */
+	err = f2fs_recover_inline_data(inode, folio);
+	if (err) {
+		if (err == 1)
+			err = 0;
+		goto out;
+	}
+
+	/* step 3: recover data indices */
+	start = f2fs_start_bidx_of_node(ofs_of_node(folio), inode);
+	end = start + ADDRS_PER_PAGE(folio, inode);
+
+	set_new_dnode(&dn, inode, NULL, NULL, 0);
+retry_dn:
+	err = f2fs_get_dnode_of_data(&dn, start, ALLOC_NODE);
+	if (err) {
+		if (err == -ENOMEM) {
+			memalloc_retry_wait(GFP_NOFS);
+			goto retry_dn;
+		}
+		goto out;
+	}
+
+	f2fs_folio_wait_writeback(dn.node_folio, NODE, true, true);
+
+	err = f2fs_get_node_info(sbi, dn.nid, &ni, false);
+	if (err)
+		goto err;
+
+	f2fs_bug_on(sbi, ni.ino != ino_of_node(folio));
+
+	if (ofs_of_node(dn.node_folio) != ofs_of_node(folio)) {
+		f2fs_warn(sbi, "Inconsistent ofs_of_node, ino:%llu, ofs:%u, %u",
+			  inode->i_ino, ofs_of_node(dn.node_folio),
+			  ofs_of_node(folio));
+		err = -EFSCORRUPTED;
+		f2fs_handle_error(sbi, ERROR_INCONSISTENT_FOOTER);
+		fserror_report_file_metadata(dn.inode, err, GFP_NOFS);
+		goto err;
+	}
+
+	for (index = start; index < end; index++, dn.ofs_in_node++) {
+		block_t src, dest;
+
+		src = f2fs_data_blkaddr(&dn);
+		dest = data_blkaddr(dn.inode, folio, dn.ofs_in_node);
+
+		if (__is_valid_data_blkaddr(src) &&
+			!f2fs_is_valid_blkaddr(sbi, src, META_POR)) {
+			err = -EFSCORRUPTED;
+			goto err;
+		}
+
+		if (__is_valid_data_blkaddr(dest) &&
+			!f2fs_is_valid_blkaddr(sbi, dest, META_POR)) {
+			err = -EFSCORRUPTED;
+			goto err;
+		}
+
+		/* skip recovering if dest is the same as src */
+		if (src == dest)
+			continue;
+
+		/* dest is invalid, just invalidate src block */
+		if (dest == NULL_ADDR) {
+			f2fs_truncate_data_blocks_range(&dn, 1);
+			continue;
+		}
+
+		if (!file_keep_isize(inode) &&
+			(i_size_read(inode) <= ((loff_t)index << PAGE_SHIFT)))
+			f2fs_i_size_write(inode,
+				(loff_t)(index + 1) << PAGE_SHIFT);
+
+		/*
+		 * dest is reserved block, invalidate src block
+		 * and then reserve one new block in dnode page.
+		 */
+		if (dest == NEW_ADDR) {
+			f2fs_truncate_data_blocks_range(&dn, 1);
+
+			err = f2fs_reserve_new_block_retry(&dn);
+			if (err)
+				goto err;
+			continue;
+		}
+
+		/* dest is valid block, try to recover from src to dest */
+		if (f2fs_is_valid_blkaddr(sbi, dest, META_POR)) {
+			if (src == NULL_ADDR) {
+				err = f2fs_reserve_new_block_retry(&dn);
+				if (err)
+					goto err;
+			}
+retry_prev:
+			/* Check the previous node page having this index */
+			err = check_index_in_prev_nodes(sbi, dest, &dn);
+			if (err) {
+				if (err == -ENOMEM) {
+					memalloc_retry_wait(GFP_NOFS);
+					goto retry_prev;
+				}
+				goto err;
+			}
+
+			if (f2fs_is_valid_blkaddr(sbi, dest,
+					DATA_GENERIC_ENHANCE_UPDATE)) {
+				f2fs_err(sbi, "Inconsistent dest blkaddr:%u, ino:%llu, ofs:%u",
+					dest, inode->i_ino, dn.ofs_in_node);
+				err = -EFSCORRUPTED;
+				goto err;
+			}
+
+			/* write dummy data page */
+			f2fs_replace_block(sbi, &dn, src, dest,
+						ni.version, false, false);
+			recovered++;
+		}
+	}
+
+	copy_node_footer(dn.node_folio, folio);
+	fill_node_footer(dn.node_folio, dn.nid, ni.ino,
+					ofs_of_node(folio), false);
+	folio_mark_dirty(dn.node_folio);
+err:
+	f2fs_put_dnode(&dn);
+out:
+	f2fs_notice(sbi, "recover_data: ino = %llx, nid = %x (i_size: %s), "
+		    "range (%u, %u), recovered = %d, err = %d",
+		    inode->i_ino, nid_of_node(folio),
+		    file_keep_isize(inode) ? "keep" : "recover",
+		    start, end, recovered, err);
+	return err;
+}
+
+static int recover_data(struct f2fs_sb_info *sbi, struct list_head *inode_list,
+		struct list_head *tmp_inode_list, struct list_head *dir_list)
+{
+	struct curseg_info *curseg;
+	int err = 0;
+	block_t blkaddr;
+	unsigned int ra_blocks = RECOVERY_MAX_RA_BLOCKS;
+	unsigned int recoverable_dnode = 0;
+	unsigned int fsynced_dnode = 0;
+	unsigned int total_dnode = 0;
+	unsigned int recovered_inode = 0;
+	unsigned int recovered_dentry = 0;
+	unsigned int recovered_dnode = 0;
+
+	f2fs_notice(sbi, "do_recover_data: start to recover dnode");
+
+	/* get node pages in the current segment */
+	curseg = CURSEG_I(sbi, CURSEG_WARM_NODE);
+	blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
+
+	while (1) {
+		struct fsync_inode_entry *entry;
+		struct folio *folio;
+
+		if (!f2fs_is_valid_blkaddr(sbi, blkaddr, META_POR))
+			break;
+
+		folio = f2fs_get_tmp_folio(sbi, blkaddr);
+		if (IS_ERR(folio)) {
+			err = PTR_ERR(folio);
+			break;
+		}
+
+		if (!is_recoverable_dnode(folio)) {
+			f2fs_folio_put(folio, true);
+			break;
+		}
+		recoverable_dnode++;
+
+		entry = get_fsync_inode(inode_list, ino_of_node(folio));
+		if (!entry)
+			goto next;
+		fsynced_dnode++;
+		/*
+		 * inode(x) | CP | inode(x) | dnode(F)
+		 * In this case, we can lose the latest inode(x).
+		 * So, call recover_inode for the inode update.
+		 */
+		if (IS_INODE(folio)) {
+			err = recover_inode(entry->inode, folio);
+			if (err) {
+				f2fs_folio_put(folio, true);
+				break;
+			}
+			recovered_inode++;
+		}
+		if (entry->last_dentry == blkaddr) {
+			err = recover_dentry(entry->inode, folio, dir_list);
+			if (err) {
+				f2fs_folio_put(folio, true);
+				break;
+			}
+			recovered_dentry++;
+		}
+		err = do_recover_data(sbi, entry->inode, folio);
+		if (err) {
+			f2fs_folio_put(folio, true);
+			break;
+		}
+		recovered_dnode++;
+
+		if (entry->blkaddr == blkaddr)
+			list_move_tail(&entry->list, tmp_inode_list);
+next:
+		ra_blocks = adjust_por_ra_blocks(sbi, ra_blocks, blkaddr,
+					next_blkaddr_of_node(folio));
+
+		/* check next segment */
+		blkaddr = next_blkaddr_of_node(folio);
+		f2fs_folio_put(folio, true);
+
+		f2fs_ra_meta_pages_cond(sbi, blkaddr, ra_blocks);
+		total_dnode++;
+	}
+	if (!err)
+		err = f2fs_allocate_new_segments(sbi);
+
+	f2fs_notice(sbi, "do_recover_data: dnode: (recoverable: %u, fsynced: %u, "
+		"total: %u), recovered: (inode: %u, dentry: %u, dnode: %u), err: %d",
+		recoverable_dnode, fsynced_dnode, total_dnode, recovered_inode,
+		recovered_dentry, recovered_dnode, err);
+	return err;
+}
+
+int f2fs_recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
+{
+	LIST_HEAD(inode_list);
+	LIST_HEAD(tmp_inode_list);
+	LIST_HEAD(dir_list);
+	struct f2fs_lock_context lc;
+	int err;
+	int ret = 0;
+	unsigned long s_flags = sbi->sb->s_flags;
+	bool need_writecp = false;
+	bool new_inode = false;
+
+	f2fs_notice(sbi, "f2fs_recover_fsync_data: recovery fsync data, "
+					"check_only: %d", check_only);
+
+	if (is_sbi_flag_set(sbi, SBI_IS_WRITABLE))
+		f2fs_info(sbi, "recover fsync data on readonly fs");
+
+	/* prevent checkpoint */
+	f2fs_down_write_trace(&sbi->cp_global_sem, &lc);
+
+	/* step #1: find fsynced inode numbers */
+	err = find_fsync_dnodes(sbi, &inode_list, check_only, &new_inode);
+	if (err < 0 || (list_empty(&inode_list) && (!check_only || !new_inode)))
+		goto skip;
+
+	if (check_only) {
+		ret = 1;
+		goto skip;
+	}
+
+	need_writecp = true;
+
+	/* step #2: recover data */
+	err = recover_data(sbi, &inode_list, &tmp_inode_list, &dir_list);
+	if (!err)
+		f2fs_bug_on(sbi, !list_empty(&inode_list));
+	else
+		f2fs_bug_on(sbi, sbi->sb->s_flags & SB_ACTIVE);
+skip:
+	destroy_fsync_dnodes(&inode_list, err);
+	destroy_fsync_dnodes(&tmp_inode_list, err);
+
+	/* truncate meta pages to be used by the recovery */
+	truncate_inode_pages_range(META_MAPPING(sbi),
+			(loff_t)MAIN_BLKADDR(sbi) << PAGE_SHIFT, -1);
+
+	if (err) {
+		truncate_inode_pages_final(NODE_MAPPING(sbi));
+		truncate_inode_pages_final(META_MAPPING(sbi));
+	}
+
+	/*
+	 * If fsync data succeeds or there is no fsync data to recover,
+	 * and the f2fs is not read only, check and fix zoned block devices'
+	 * write pointer consistency.
+	 */
+	if (!err)
+		err = f2fs_check_and_fix_write_pointer(sbi);
+
+	if (!err)
+		clear_sbi_flag(sbi, SBI_POR_DOING);
+
+	f2fs_up_write_trace(&sbi->cp_global_sem, &lc);
+
+	/* let's drop all the directory inodes for clean checkpoint */
+	destroy_fsync_dnodes(&dir_list, err);
+
+	if (need_writecp) {
+		set_sbi_flag(sbi, SBI_IS_RECOVERED);
+
+		if (!err) {
+			struct cp_control cpc = {
+				.reason = CP_RECOVERY,
+			};
+			stat_inc_cp_call_count(sbi, TOTAL_CALL);
+			err = f2fs_write_checkpoint(sbi, &cpc);
+		}
+	}
+
+	sbi->sb->s_flags = s_flags; /* Restore SB_RDONLY status */
+
+	return ret ? ret : err;
+}
+
+int __init f2fs_create_recovery_cache(void)
+{
+	fsync_entry_slab = f2fs_kmem_cache_create("f2fs_fsync_inode_entry",
+					sizeof(struct fsync_inode_entry));
+	return fsync_entry_slab ? 0 : -ENOMEM;
+}
+
+void f2fs_destroy_recovery_cache(void)
+{
+	kmem_cache_destroy(fsync_entry_slab);
+}