1 files changed, 129 insertions, 39 deletions

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 67a5b937f5a9..f769f5cb6deb 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -31,6 +31,7 @@
 #include <linux/writeback.h>
 #include <linux/pagevec.h>
 #include <linux/mpage.h>
+#include <linux/rmap.h>
 #include <linux/namei.h>
 #include <linux/uio.h>
 #include <linux/bio.h>
@@ -383,10 +384,11 @@ static int __check_block_validity(struct inode *inode, const char *func,
 				unsigned int line,
 				struct ext4_map_blocks *map)
 {
-	if (ext4_has_feature_journal(inode->i_sb) &&
-	    (inode->i_ino ==
-	     le32_to_cpu(EXT4_SB(inode->i_sb)->s_es->s_journal_inum)))
+	journal_t *journal = EXT4_SB(inode->i_sb)->s_journal;
+
+	if (journal && inode == journal->j_inode)
 		return 0;
+
 	if (!ext4_inode_block_valid(inode, map->m_pblk, map->m_len)) {
 		ext4_error_inode(inode, func, line, map->m_pblk,
 				 "lblock %lu mapped to illegal pblock %llu "
@@ -998,7 +1000,12 @@ int ext4_walk_page_buffers(handle_t *handle, struct inode *inode,
  */
 static int ext4_dirty_journalled_data(handle_t *handle, struct buffer_head *bh)
 {
-	folio_mark_dirty(bh->b_folio);
+	struct folio *folio = bh->b_folio;
+	struct inode *inode = folio->mapping->host;
+
+	/* only regular files have a_ops */
+	if (S_ISREG(inode->i_mode))
+		folio_mark_dirty(folio);
 	return ext4_handle_dirty_metadata(handle, NULL, bh);
 }
 
@@ -3878,6 +3885,68 @@ int ext4_update_disksize_before_punch(struct inode *inode, loff_t offset,
 	return ret;
 }
 
+static inline void ext4_truncate_folio(struct inode *inode,
+				       loff_t start, loff_t end)
+{
+	unsigned long blocksize = i_blocksize(inode);
+	struct folio *folio;
+
+	/* Nothing to be done if no complete block needs to be truncated. */
+	if (round_up(start, blocksize) >= round_down(end, blocksize))
+		return;
+
+	folio = filemap_lock_folio(inode->i_mapping, start >> PAGE_SHIFT);
+	if (IS_ERR(folio))
+		return;
+
+	if (folio_mkclean(folio))
+		folio_mark_dirty(folio);
+	folio_unlock(folio);
+	folio_put(folio);
+}
+
+int ext4_truncate_page_cache_block_range(struct inode *inode,
+					 loff_t start, loff_t end)
+{
+	unsigned long blocksize = i_blocksize(inode);
+	int ret;
+
+	/*
+	 * For journalled data we need to write (and checkpoint) pages
+	 * before discarding page cache to avoid inconsitent data on disk
+	 * in case of crash before freeing or unwritten converting trans
+	 * is committed.
+	 */
+	if (ext4_should_journal_data(inode)) {
+		ret = filemap_write_and_wait_range(inode->i_mapping, start,
+						   end - 1);
+		if (ret)
+			return ret;
+		goto truncate_pagecache;
+	}
+
+	/*
+	 * If the block size is less than the page size, the file's mapped
+	 * blocks within one page could be freed or converted to unwritten.
+	 * So it's necessary to remove writable userspace mappings, and then
+	 * ext4_page_mkwrite() can be called during subsequent write access
+	 * to these partial folios.
+	 */
+	if (!IS_ALIGNED(start | end, PAGE_SIZE) &&
+	    blocksize < PAGE_SIZE && start < inode->i_size) {
+		loff_t page_boundary = round_up(start, PAGE_SIZE);
+
+		ext4_truncate_folio(inode, start, min(page_boundary, end));
+		if (end > page_boundary)
+			ext4_truncate_folio(inode,
+					    round_down(end, PAGE_SIZE), end);
+	}
+
+truncate_pagecache:
+	truncate_pagecache_range(inode, start, end - 1);
+	return 0;
+}
+
 static void ext4_wait_dax_page(struct inode *inode)
 {
 	filemap_invalidate_unlock(inode->i_mapping);
@@ -3932,17 +4001,6 @@ int ext4_punch_hole(struct file *file, loff_t offset, loff_t length)
 
 	trace_ext4_punch_hole(inode, offset, length, 0);
 
-	/*
-	 * Write out all dirty pages to avoid race conditions
-	 * Then release them.
-	 */
-	if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) {
-		ret = filemap_write_and_wait_range(mapping, offset,
-						   offset + length - 1);
-		if (ret)
-			return ret;
-	}
-
 	inode_lock(inode);
 
 	/* No need to punch hole beyond i_size */
@@ -4004,8 +4062,11 @@ int ext4_punch_hole(struct file *file, loff_t offset, loff_t length)
 		ret = ext4_update_disksize_before_punch(inode, offset, length);
 		if (ret)
 			goto out_dio;
-		truncate_pagecache_range(inode, first_block_offset,
-					 last_block_offset);
+
+		ret = ext4_truncate_page_cache_block_range(inode,
+				first_block_offset, last_block_offset + 1);
+		if (ret)
+			goto out_dio;
 	}
 
 	if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
@@ -4681,22 +4742,43 @@ static inline void ext4_inode_set_iversion_queried(struct inode *inode, u64 val)
 		inode_set_iversion_queried(inode, val);
 }
 
-static const char *check_igot_inode(struct inode *inode, ext4_iget_flags flags)
-
+static int check_igot_inode(struct inode *inode, ext4_iget_flags flags,
+			    const char *function, unsigned int line)
 {
+	const char *err_str;
+
 	if (flags & EXT4_IGET_EA_INODE) {
-		if (!(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL))
-			return "missing EA_INODE flag";
+		if (!(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL)) {
+			err_str = "missing EA_INODE flag";
+			goto error;
+		}
 		if (ext4_test_inode_state(inode, EXT4_STATE_XATTR) ||
-		    EXT4_I(inode)->i_file_acl)
-			return "ea_inode with extended attributes";
+		    EXT4_I(inode)->i_file_acl) {
+			err_str = "ea_inode with extended attributes";
+			goto error;
+		}
 	} else {
-		if ((EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL))
-			return "unexpected EA_INODE flag";
+		if ((EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL)) {
+			/*
+			 * open_by_handle_at() could provide an old inode number
+			 * that has since been reused for an ea_inode; this does
+			 * not indicate filesystem corruption
+			 */
+			if (flags & EXT4_IGET_HANDLE)
+				return -ESTALE;
+			err_str = "unexpected EA_INODE flag";
+			goto error;
+		}
+	}
+	if (is_bad_inode(inode) && !(flags & EXT4_IGET_BAD)) {
+		err_str = "unexpected bad inode w/o EXT4_IGET_BAD";
+		goto error;
 	}
-	if (is_bad_inode(inode) && !(flags & EXT4_IGET_BAD))
-		return "unexpected bad inode w/o EXT4_IGET_BAD";
-	return NULL;
+	return 0;
+
+error:
+	ext4_error_inode(inode, function, line, 0, err_str);
+	return -EFSCORRUPTED;
 }
 
 struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
@@ -4708,7 +4790,6 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 	struct ext4_inode_info *ei;
 	struct ext4_super_block *es = EXT4_SB(sb)->s_es;
 	struct inode *inode;
-	const char *err_str;
 	journal_t *journal = EXT4_SB(sb)->s_journal;
 	long ret;
 	loff_t size;
@@ -4737,10 +4818,10 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 	if (!inode)
 		return ERR_PTR(-ENOMEM);
 	if (!(inode->i_state & I_NEW)) {
-		if ((err_str = check_igot_inode(inode, flags)) != NULL) {
-			ext4_error_inode(inode, function, line, 0, err_str);
+		ret = check_igot_inode(inode, flags, function, line);
+		if (ret) {
 			iput(inode);
-			return ERR_PTR(-EFSCORRUPTED);
+			return ERR_PTR(ret);
 		}
 		return inode;
 	}
@@ -4852,7 +4933,8 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 		ei->i_file_acl |=
 			((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32;
 	inode->i_size = ext4_isize(sb, raw_inode);
-	if ((size = i_size_read(inode)) < 0) {
+	size = i_size_read(inode);
+	if (size < 0 || size > ext4_get_maxbytes(inode)) {
 		ext4_error_inode(inode, function, line, 0,
 				 "iget: bad i_size value: %lld", size);
 		ret = -EFSCORRUPTED;
@@ -5012,13 +5094,21 @@ struct inode *__ext4_iget(struct super_block *sb, unsigned long ino,
 		ret = -EFSCORRUPTED;
 		goto bad_inode;
 	}
-	if ((err_str = check_igot_inode(inode, flags)) != NULL) {
-		ext4_error_inode(inode, function, line, 0, err_str);
-		ret = -EFSCORRUPTED;
-		goto bad_inode;
+	ret = check_igot_inode(inode, flags, function, line);
+	/*
+	 * -ESTALE here means there is nothing inherently wrong with the inode,
+	 * it's just not an inode we can return for an fhandle lookup.
+	 */
+	if (ret == -ESTALE) {
+		brelse(iloc.bh);
+		unlock_new_inode(inode);
+		iput(inode);
+		return ERR_PTR(-ESTALE);
 	}
-
+	if (ret)
+		goto bad_inode;
 	brelse(iloc.bh);
+
 	unlock_new_inode(inode);
 	return inode;
 
@@ -5439,7 +5529,7 @@ int ext4_setattr(struct mnt_idmap *idmap, struct dentry *dentry,
 			    oldsize & (inode->i_sb->s_blocksize - 1)) {
 				error = ext4_inode_attach_jinode(inode);
 				if (error)
-					goto err_out;
+					goto out_mmap_sem;
 			}
 
 			handle = ext4_journal_start(inode, EXT4_HT_INODE, 3);