1 files changed, 179 insertions, 86 deletions

diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index c2355b486669..5d0c77054252 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -1,12 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0
 /*
  * fs/f2fs/recovery.c
  *
  * Copyright (c) 2012 Samsung Electronics Co., Ltd.
  *             http://www.samsung.com/
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
  */
 #include <linux/fs.h>
 #include <linux/f2fs_fs.h>
@@ -47,7 +44,7 @@
 
 static struct kmem_cache *fsync_entry_slab;
 
-bool space_for_roll_forward(struct f2fs_sb_info *sbi)
+bool f2fs_space_for_roll_forward(struct f2fs_sb_info *sbi)
 {
 	s64 nalloc = percpu_counter_sum_positive(&sbi->alloc_valid_block_count);
 
@@ -99,8 +96,12 @@ err_out:
 	return ERR_PTR(err);
 }
 
-static void del_fsync_inode(struct fsync_inode_entry *entry)
+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);
@@ -162,7 +163,7 @@ retry:
 			goto out_put;
 		}
 
-		err = acquire_orphan_inode(F2FS_I_SB(inode));
+		err = f2fs_acquire_orphan_inode(F2FS_I_SB(inode));
 		if (err) {
 			iput(einode);
 			goto out_put;
@@ -173,7 +174,7 @@ retry:
 	} else if (IS_ERR(page)) {
 		err = PTR_ERR(page);
 	} else {
-		err = __f2fs_do_add_link(dir, &fname, inode,
+		err = f2fs_add_dentry(dir, &fname, inode,
 					inode->i_ino, inode->i_mode);
 	}
 	if (err == -ENOMEM)
@@ -187,10 +188,36 @@ out:
 		name = "<encrypted>";
 	else
 		name = raw_inode->i_name;
-	f2fs_msg(inode->i_sb, KERN_NOTICE,
-			"%s: ino = %x, name = %s, dir = %lx, err = %d",
-			__func__, ino_of_node(ipage), name,
-			IS_ERR(dir) ? 0 : dir->i_ino, err);
+	f2fs_notice(F2FS_I_SB(inode), "%s: ino = %x, name = %s, dir = %lx, err = %d",
+		    __func__, ino_of_node(ipage), name,
+		    IS_ERR(dir) ? 0 : dir->i_ino, err);
+	return err;
+}
+
+static int recover_quota_data(struct inode *inode, struct page *page)
+{
+	struct f2fs_inode *raw = F2FS_INODE(page);
+	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_uid = make_kuid(&init_user_ns, i_uid);
+	attr.ia_gid = make_kgid(&init_user_ns, i_gid);
+
+	if (!uid_eq(attr.ia_uid, inode->i_uid))
+		attr.ia_valid |= ATTR_UID;
+	if (!gid_eq(attr.ia_gid, inode->i_gid))
+		attr.ia_valid |= ATTR_GID;
+
+	if (!attr.ia_valid)
+		return 0;
+
+	err = dquot_transfer(inode, &attr);
+	if (err)
+		set_sbi_flag(F2FS_I_SB(inode), SBI_QUOTA_NEED_REPAIR);
 	return err;
 }
 
@@ -204,16 +231,35 @@ static void recover_inline_flags(struct inode *inode, struct f2fs_inode *ri)
 		set_inode_flag(inode, FI_DATA_EXIST);
 	else
 		clear_inode_flag(inode, FI_DATA_EXIST);
-	if (!(ri->i_inline & F2FS_INLINE_DOTS))
-		clear_inode_flag(inode, FI_INLINE_DOTS);
 }
 
-static void recover_inode(struct inode *inode, struct page *page)
+static int recover_inode(struct inode *inode, struct page *page)
 {
 	struct f2fs_inode *raw = F2FS_INODE(page);
 	char *name;
+	int err;
 
 	inode->i_mode = le16_to_cpu(raw->i_mode);
+
+	err = recover_quota_data(inode, page);
+	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;
+
+			i_projid = (projid_t)le32_to_cpu(raw->i_projid);
+			F2FS_I(inode)->i_projid =
+				make_kprojid(&init_user_ns, i_projid);
+		}
+	}
+
 	f2fs_i_size_write(inode, le64_to_cpu(raw->i_size));
 	inode->i_atime.tv_sec = le64_to_cpu(raw->i_atime);
 	inode->i_ctime.tv_sec = le64_to_cpu(raw->i_ctime);
@@ -223,17 +269,23 @@ static void recover_inode(struct inode *inode, struct page *page)
 	inode->i_mtime.tv_nsec = le32_to_cpu(raw->i_mtime_nsec);
 
 	F2FS_I(inode)->i_advise = raw->i_advise;
+	F2FS_I(inode)->i_flags = le32_to_cpu(raw->i_flags);
+	f2fs_set_inode_flags(inode);
+	F2FS_I(inode)->i_gc_failures[GC_FAILURE_PIN] =
+				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(page)->i_name;
 
-	f2fs_msg(inode->i_sb, KERN_NOTICE,
-		"recover_inode: ino = %x, name = %s, inline = %x",
-			ino_of_node(page), name, raw->i_inline);
+	f2fs_notice(F2FS_I_SB(inode), "recover_inode: ino = %x, name = %s, inline = %x",
+		    ino_of_node(page), name, raw->i_inline);
+	return 0;
 }
 
 static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head,
@@ -243,8 +295,8 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head,
 	struct page *page = NULL;
 	block_t blkaddr;
 	unsigned int loop_cnt = 0;
-	unsigned int free_blocks = sbi->user_block_count -
-					valid_user_blocks(sbi);
+	unsigned int free_blocks = MAIN_SEGS(sbi) * sbi->blocks_per_seg -
+						valid_user_blocks(sbi);
 	int err = 0;
 
 	/* get node pages in the current segment */
@@ -254,13 +306,19 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head,
 	while (1) {
 		struct fsync_inode_entry *entry;
 
-		if (!is_valid_blkaddr(sbi, blkaddr, META_POR))
+		if (!f2fs_is_valid_blkaddr(sbi, blkaddr, META_POR))
 			return 0;
 
-		page = get_tmp_page(sbi, blkaddr);
+		page = f2fs_get_tmp_page(sbi, blkaddr);
+		if (IS_ERR(page)) {
+			err = PTR_ERR(page);
+			break;
+		}
 
-		if (!is_recoverable_dnode(page))
+		if (!is_recoverable_dnode(page)) {
+			f2fs_put_page(page, 1);
 			break;
+		}
 
 		if (!is_fsync_dnode(page))
 			goto next;
@@ -271,9 +329,11 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head,
 
 			if (!check_only &&
 					IS_INODE(page) && is_dent_dnode(page)) {
-				err = recover_inode_page(sbi, page);
-				if (err)
+				err = f2fs_recover_inode_page(sbi, page);
+				if (err) {
+					f2fs_put_page(page, 1);
 					break;
+				}
 				quota_inode = true;
 			}
 
@@ -289,6 +349,7 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head,
 					err = 0;
 					goto next;
 				}
+				f2fs_put_page(page, 1);
 				break;
 			}
 		}
@@ -300,10 +361,10 @@ next:
 		/* sanity check in order to detect looped node chain */
 		if (++loop_cnt >= free_blocks ||
 			blkaddr == next_blkaddr_of_node(page)) {
-			f2fs_msg(sbi->sb, KERN_NOTICE,
-				"%s: detect looped node chain, "
-				"blkaddr:%u, next:%u",
-				__func__, blkaddr, next_blkaddr_of_node(page));
+			f2fs_notice(sbi, "%s: detect looped node chain, blkaddr:%u, next:%u",
+				    __func__, blkaddr,
+				    next_blkaddr_of_node(page));
+			f2fs_put_page(page, 1);
 			err = -EINVAL;
 			break;
 		}
@@ -312,18 +373,17 @@ next:
 		blkaddr = next_blkaddr_of_node(page);
 		f2fs_put_page(page, 1);
 
-		ra_meta_pages_cond(sbi, blkaddr);
+		f2fs_ra_meta_pages_cond(sbi, blkaddr);
 	}
-	f2fs_put_page(page, 1);
 	return err;
 }
 
-static void destroy_fsync_dnodes(struct list_head *head)
+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);
+		del_fsync_inode(entry, drop);
 }
 
 static int check_index_in_prev_nodes(struct f2fs_sb_info *sbi,
@@ -355,7 +415,9 @@ static int check_index_in_prev_nodes(struct f2fs_sb_info *sbi,
 		}
 	}
 
-	sum_page = get_sum_page(sbi, segno);
+	sum_page = f2fs_get_sum_page(sbi, segno);
+	if (IS_ERR(sum_page))
+		return PTR_ERR(sum_page);
 	sum_node = (struct f2fs_summary_block *)page_address(sum_page);
 	sum = sum_node->entries[blkoff];
 	f2fs_put_page(sum_page, 1);
@@ -375,7 +437,7 @@ got_it:
 	}
 
 	/* Get the node page */
-	node_page = get_node_page(sbi, nid);
+	node_page = f2fs_get_node_page(sbi, nid);
 	if (IS_ERR(node_page))
 		return PTR_ERR(node_page);
 
@@ -400,7 +462,8 @@ got_it:
 		inode = dn->inode;
 	}
 
-	bidx = start_bidx_of_node(offset, inode) + le16_to_cpu(sum.ofs_in_node);
+	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
@@ -410,11 +473,11 @@ got_it:
 		unlock_page(dn->inode_page);
 
 	set_new_dnode(&tdn, inode, NULL, NULL, 0);
-	if (get_dnode_of_data(&tdn, bidx, LOOKUP_NODE))
+	if (f2fs_get_dnode_of_data(&tdn, bidx, LOOKUP_NODE))
 		goto out;
 
 	if (tdn.data_blkaddr == blkaddr)
-		truncate_data_blocks_range(&tdn, 1);
+		f2fs_truncate_data_blocks_range(&tdn, 1);
 
 	f2fs_put_dnode(&tdn);
 out:
@@ -427,7 +490,7 @@ out:
 truncate_out:
 	if (datablock_addr(tdn.inode, tdn.node_page,
 					tdn.ofs_in_node) == blkaddr)
-		truncate_data_blocks_range(&tdn, 1);
+		f2fs_truncate_data_blocks_range(&tdn, 1);
 	if (dn->inode->i_ino == nid && !dn->inode_page_locked)
 		unlock_page(dn->inode_page);
 	return 0;
@@ -443,25 +506,25 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 
 	/* step 1: recover xattr */
 	if (IS_INODE(page)) {
-		recover_inline_xattr(inode, page);
+		f2fs_recover_inline_xattr(inode, page);
 	} else if (f2fs_has_xattr_block(ofs_of_node(page))) {
-		err = recover_xattr_data(inode, page);
+		err = f2fs_recover_xattr_data(inode, page);
 		if (!err)
 			recovered++;
 		goto out;
 	}
 
 	/* step 2: recover inline data */
-	if (recover_inline_data(inode, page))
+	if (f2fs_recover_inline_data(inode, page))
 		goto out;
 
 	/* step 3: recover data indices */
-	start = start_bidx_of_node(ofs_of_node(page), inode);
+	start = f2fs_start_bidx_of_node(ofs_of_node(page), inode);
 	end = start + ADDRS_PER_PAGE(page, inode);
 
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
 retry_dn:
-	err = get_dnode_of_data(&dn, start, ALLOC_NODE);
+	err = f2fs_get_dnode_of_data(&dn, start, ALLOC_NODE);
 	if (err) {
 		if (err == -ENOMEM) {
 			congestion_wait(BLK_RW_ASYNC, HZ/50);
@@ -470,17 +533,19 @@ retry_dn:
 		goto out;
 	}
 
-	f2fs_wait_on_page_writeback(dn.node_page, NODE, true);
+	f2fs_wait_on_page_writeback(dn.node_page, NODE, true, true);
+
+	err = f2fs_get_node_info(sbi, dn.nid, &ni);
+	if (err)
+		goto err;
 
-	get_node_info(sbi, dn.nid, &ni);
 	f2fs_bug_on(sbi, ni.ino != ino_of_node(page));
 
 	if (ofs_of_node(dn.node_page) != ofs_of_node(page)) {
-		f2fs_msg(sbi->sb, KERN_WARNING,
-			"Inconsistent ofs_of_node, ino:%lu, ofs:%u, %u",
-			inode->i_ino, ofs_of_node(dn.node_page),
-			ofs_of_node(page));
-		err = -EFAULT;
+		f2fs_warn(sbi, "Inconsistent ofs_of_node, ino:%lu, ofs:%u, %u",
+			  inode->i_ino, ofs_of_node(dn.node_page),
+			  ofs_of_node(page));
+		err = -EFSCORRUPTED;
 		goto err;
 	}
 
@@ -490,13 +555,25 @@ retry_dn:
 		src = datablock_addr(dn.inode, dn.node_page, dn.ofs_in_node);
 		dest = datablock_addr(dn.inode, page, 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) {
-			truncate_data_blocks_range(&dn, 1);
+			f2fs_truncate_data_blocks_range(&dn, 1);
 			continue;
 		}
 
@@ -510,20 +587,19 @@ retry_dn:
 		 * and then reserve one new block in dnode page.
 		 */
 		if (dest == NEW_ADDR) {
-			truncate_data_blocks_range(&dn, 1);
-			reserve_new_block(&dn);
+			f2fs_truncate_data_blocks_range(&dn, 1);
+			f2fs_reserve_new_block(&dn);
 			continue;
 		}
 
 		/* dest is valid block, try to recover from src to dest */
-		if (is_valid_blkaddr(sbi, dest, META_POR)) {
+		if (f2fs_is_valid_blkaddr(sbi, dest, META_POR)) {
 
 			if (src == NULL_ADDR) {
-				err = reserve_new_block(&dn);
-#ifdef CONFIG_F2FS_FAULT_INJECTION
-				while (err)
-					err = reserve_new_block(&dn);
-#endif
+				err = f2fs_reserve_new_block(&dn);
+				while (err &&
+				       IS_ENABLED(CONFIG_F2FS_FAULT_INJECTION))
+					err = f2fs_reserve_new_block(&dn);
 				/* We should not get -ENOSPC */
 				f2fs_bug_on(sbi, err);
 				if (err)
@@ -554,16 +630,14 @@ retry_prev:
 err:
 	f2fs_put_dnode(&dn);
 out:
-	f2fs_msg(sbi->sb, KERN_NOTICE,
-		"recover_data: ino = %lx (i_size: %s) recovered = %d, err = %d",
-		inode->i_ino,
-		file_keep_isize(inode) ? "keep" : "recover",
-		recovered, err);
+	f2fs_notice(sbi, "recover_data: ino = %lx (i_size: %s) recovered = %d, err = %d",
+		    inode->i_ino, file_keep_isize(inode) ? "keep" : "recover",
+		    recovered, err);
 	return err;
 }
 
 static int recover_data(struct f2fs_sb_info *sbi, struct list_head *inode_list,
-						struct list_head *dir_list)
+		struct list_head *tmp_inode_list, struct list_head *dir_list)
 {
 	struct curseg_info *curseg;
 	struct page *page = NULL;
@@ -577,12 +651,16 @@ static int recover_data(struct f2fs_sb_info *sbi, struct list_head *inode_list,
 	while (1) {
 		struct fsync_inode_entry *entry;
 
-		if (!is_valid_blkaddr(sbi, blkaddr, META_POR))
+		if (!f2fs_is_valid_blkaddr(sbi, blkaddr, META_POR))
 			break;
 
-		ra_meta_pages_cond(sbi, blkaddr);
+		f2fs_ra_meta_pages_cond(sbi, blkaddr);
 
-		page = get_tmp_page(sbi, blkaddr);
+		page = f2fs_get_tmp_page(sbi, blkaddr);
+		if (IS_ERR(page)) {
+			err = PTR_ERR(page);
+			break;
+		}
 
 		if (!is_recoverable_dnode(page)) {
 			f2fs_put_page(page, 1);
@@ -597,8 +675,13 @@ static int recover_data(struct f2fs_sb_info *sbi, struct list_head *inode_list,
 		 * In this case, we can lose the latest inode(x).
 		 * So, call recover_inode for the inode update.
 		 */
-		if (IS_INODE(page))
-			recover_inode(entry->inode, page);
+		if (IS_INODE(page)) {
+			err = recover_inode(entry->inode, page);
+			if (err) {
+				f2fs_put_page(page, 1);
+				break;
+			}
+		}
 		if (entry->last_dentry == blkaddr) {
 			err = recover_dentry(entry->inode, page, dir_list);
 			if (err) {
@@ -613,20 +696,20 @@ static int recover_data(struct f2fs_sb_info *sbi, struct list_head *inode_list,
 		}
 
 		if (entry->blkaddr == blkaddr)
-			del_fsync_inode(entry);
+			list_move_tail(&entry->list, tmp_inode_list);
 next:
 		/* check next segment */
 		blkaddr = next_blkaddr_of_node(page);
 		f2fs_put_page(page, 1);
 	}
 	if (!err)
-		allocate_new_segments(sbi);
+		f2fs_allocate_new_segments(sbi);
 	return err;
 }
 
-int recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
+int f2fs_recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 {
-	struct list_head inode_list;
+	struct list_head inode_list, tmp_inode_list;
 	struct list_head dir_list;
 	int err;
 	int ret = 0;
@@ -637,7 +720,7 @@ int recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 #endif
 
 	if (s_flags & MS_RDONLY) {
-		f2fs_msg(sbi->sb, KERN_INFO, "orphan cleanup on readonly fs");
+		f2fs_info(sbi, "recover fsync data on readonly fs");
 		sbi->sb->s_flags &= ~MS_RDONLY;
 	}
 
@@ -656,6 +739,7 @@ int recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 	}
 
 	INIT_LIST_HEAD(&inode_list);
+	INIT_LIST_HEAD(&tmp_inode_list);
 	INIT_LIST_HEAD(&dir_list);
 
 	/* prevent checkpoint */
@@ -674,11 +758,16 @@ int recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 	need_writecp = true;
 
 	/* step #2: recover data */
-	err = recover_data(sbi, &inode_list, &dir_list);
+	err = recover_data(sbi, &inode_list, &tmp_inode_list, &dir_list);
 	if (!err)
 		f2fs_bug_on(sbi, !list_empty(&inode_list));
+	else {
+		/* restore s_flags to let iput() trash data */
+		sbi->sb->s_flags = s_flags;
+	}
 skip:
-	destroy_fsync_dnodes(&inode_list);
+	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),
@@ -687,19 +776,23 @@ skip:
 	if (err) {
 		truncate_inode_pages_final(NODE_MAPPING(sbi));
 		truncate_inode_pages_final(META_MAPPING(sbi));
+	} else {
+		clear_sbi_flag(sbi, SBI_POR_DOING);
 	}
-
-	clear_sbi_flag(sbi, SBI_POR_DOING);
 	mutex_unlock(&sbi->cp_mutex);
 
 	/* let's drop all the directory inodes for clean checkpoint */
-	destroy_fsync_dnodes(&dir_list);
+	destroy_fsync_dnodes(&dir_list, err);
+
+	if (need_writecp) {
+		set_sbi_flag(sbi, SBI_IS_RECOVERED);
 
-	if (!err && need_writecp) {
-		struct cp_control cpc = {
-			.reason = CP_RECOVERY,
-		};
-		err = write_checkpoint(sbi, &cpc);
+		if (!err) {
+			struct cp_control cpc = {
+				.reason = CP_RECOVERY,
+			};
+			err = f2fs_write_checkpoint(sbi, &cpc);
+		}
 	}
 
 	kmem_cache_destroy(fsync_entry_slab);