1 files changed, 64 insertions, 38 deletions

diff --git a/fs/zonefs/file.c b/fs/zonefs/file.c
index b2c9b35df8f7..c1e5e30e90a0 100644
--- a/fs/zonefs/file.c
+++ b/fs/zonefs/file.c
@@ -85,7 +85,7 @@ static int zonefs_write_iomap_begin(struct inode *inode, loff_t offset,
 	/*
 	 * For conventional zones, all blocks are always mapped. For sequential
 	 * zones, all blocks after always mapped below the inode size (zone
-	 * write pointer) and unwriten beyond.
+	 * write pointer) and unwritten beyond.
 	 */
 	mutex_lock(&zi->i_truncate_mutex);
 	iomap->bdev = inode->i_sb->s_bdev;
@@ -112,48 +112,59 @@ static const struct iomap_ops zonefs_write_iomap_ops = {
 
 static int zonefs_read_folio(struct file *unused, struct folio *folio)
 {
-	return iomap_read_folio(folio, &zonefs_read_iomap_ops);
+	iomap_bio_read_folio(folio, &zonefs_read_iomap_ops);
+	return 0;
 }
 
 static void zonefs_readahead(struct readahead_control *rac)
 {
-	iomap_readahead(rac, &zonefs_read_iomap_ops);
+	iomap_bio_readahead(rac, &zonefs_read_iomap_ops);
 }
 
 /*
  * Map blocks for page writeback. This is used only on conventional zone files,
  * which implies that the page range can only be within the fixed inode size.
  */
-static int zonefs_write_map_blocks(struct iomap_writepage_ctx *wpc,
-				   struct inode *inode, loff_t offset)
+static ssize_t zonefs_writeback_range(struct iomap_writepage_ctx *wpc,
+		struct folio *folio, u64 offset, unsigned len, u64 end_pos)
 {
-	struct zonefs_zone *z = zonefs_inode_zone(inode);
+	struct zonefs_zone *z = zonefs_inode_zone(wpc->inode);
 
 	if (WARN_ON_ONCE(zonefs_zone_is_seq(z)))
 		return -EIO;
-	if (WARN_ON_ONCE(offset >= i_size_read(inode)))
+	if (WARN_ON_ONCE(offset >= i_size_read(wpc->inode)))
 		return -EIO;
 
 	/* If the mapping is already OK, nothing needs to be done */
-	if (offset >= wpc->iomap.offset &&
-	    offset < wpc->iomap.offset + wpc->iomap.length)
-		return 0;
+	if (offset < wpc->iomap.offset ||
+	    offset >= wpc->iomap.offset + wpc->iomap.length) {
+		int error;
+
+		error = zonefs_write_iomap_begin(wpc->inode, offset,
+				z->z_capacity - offset, IOMAP_WRITE,
+				&wpc->iomap, NULL);
+		if (error)
+			return error;
+	}
 
-	return zonefs_write_iomap_begin(inode, offset,
-					z->z_capacity - offset,
-					IOMAP_WRITE, &wpc->iomap, NULL);
+	return iomap_add_to_ioend(wpc, folio, offset, end_pos, len);
 }
 
 static const struct iomap_writeback_ops zonefs_writeback_ops = {
-	.map_blocks		= zonefs_write_map_blocks,
+	.writeback_range	= zonefs_writeback_range,
+	.writeback_submit	= iomap_ioend_writeback_submit,
 };
 
 static int zonefs_writepages(struct address_space *mapping,
 			     struct writeback_control *wbc)
 {
-	struct iomap_writepage_ctx wpc = { };
+	struct iomap_writepage_ctx wpc = {
+		.inode		= mapping->host,
+		.wbc		= wbc,
+		.ops		= &zonefs_writeback_ops,
+	};
 
-	return iomap_writepages(mapping, wbc, &wpc, &zonefs_writeback_ops);
+	return iomap_writepages(&wpc);
 }
 
 static int zonefs_swap_activate(struct swap_info_struct *sis,
@@ -180,7 +191,7 @@ const struct address_space_operations zonefs_file_aops = {
 	.invalidate_folio	= iomap_invalidate_folio,
 	.migrate_folio		= filemap_migrate_folio,
 	.is_partially_uptodate	= iomap_is_partially_uptodate,
-	.error_remove_page	= generic_error_remove_page,
+	.error_remove_folio	= generic_error_remove_folio,
 	.swap_activate		= zonefs_swap_activate,
 };
 
@@ -298,7 +309,7 @@ static vm_fault_t zonefs_filemap_page_mkwrite(struct vm_fault *vmf)
 
 	/* Serialize against truncates */
 	filemap_invalidate_lock_shared(inode->i_mapping);
-	ret = iomap_page_mkwrite(vmf, &zonefs_write_iomap_ops);
+	ret = iomap_page_mkwrite(vmf, &zonefs_write_iomap_ops, NULL);
 	filemap_invalidate_unlock_shared(inode->i_mapping);
 
 	sb_end_pagefault(inode->i_sb);
@@ -311,8 +322,10 @@ static const struct vm_operations_struct zonefs_file_vm_ops = {
 	.page_mkwrite	= zonefs_filemap_page_mkwrite,
 };
 
-static int zonefs_file_mmap(struct file *file, struct vm_area_struct *vma)
+static int zonefs_file_mmap_prepare(struct vm_area_desc *desc)
 {
+	struct file *file = desc->file;
+
 	/*
 	 * Conventional zones accept random writes, so their files can support
 	 * shared writable mappings. For sequential zone files, only read
@@ -320,11 +333,11 @@ static int zonefs_file_mmap(struct file *file, struct vm_area_struct *vma)
 	 * ordering between msync() and page cache writeback.
 	 */
 	if (zonefs_inode_is_seq(file_inode(file)) &&
-	    (vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE))
+	    (desc->vm_flags & VM_SHARED) && (desc->vm_flags & VM_MAYWRITE))
 		return -EINVAL;
 
 	file_accessed(file);
-	vma->vm_ops = &zonefs_file_vm_ops;
+	desc->vm_ops = &zonefs_file_vm_ops;
 
 	return 0;
 }
@@ -348,7 +361,12 @@ static int zonefs_file_write_dio_end_io(struct kiocb *iocb, ssize_t size,
 	struct zonefs_inode_info *zi = ZONEFS_I(inode);
 
 	if (error) {
-		zonefs_io_error(inode, true);
+		/*
+		 * For Sync IOs, error recovery is called from
+		 * zonefs_file_dio_write().
+		 */
+		if (!is_sync_kiocb(iocb))
+			zonefs_io_error(inode, true);
 		return error;
 	}
 
@@ -491,6 +509,14 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from)
 			ret = -EINVAL;
 			goto inode_unlock;
 		}
+		/*
+		 * Advance the zone write pointer offset. This assumes that the
+		 * IO will succeed, which is OK to do because we do not allow
+		 * partial writes (IOMAP_DIO_PARTIAL is not set) and if the IO
+		 * fails, the error path will correct the write pointer offset.
+		 */
+		z->z_wpoffset += count;
+		zonefs_inode_account_active(inode);
 		mutex_unlock(&zi->i_truncate_mutex);
 	}
 
@@ -504,20 +530,19 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from)
 	if (ret == -ENOTBLK)
 		ret = -EBUSY;
 
-	if (zonefs_zone_is_seq(z) &&
-	    (ret > 0 || ret == -EIOCBQUEUED)) {
-		if (ret > 0)
-			count = ret;
-
-		/*
-		 * Update the zone write pointer offset assuming the write
-		 * operation succeeded. If it did not, the error recovery path
-		 * will correct it. Also do active seq file accounting.
-		 */
-		mutex_lock(&zi->i_truncate_mutex);
-		z->z_wpoffset += count;
-		zonefs_inode_account_active(inode);
-		mutex_unlock(&zi->i_truncate_mutex);
+	/*
+	 * For a failed IO or partial completion, trigger error recovery
+	 * to update the zone write pointer offset to a correct value.
+	 * For asynchronous IOs, zonefs_file_write_dio_end_io() may already
+	 * have executed error recovery if the IO already completed when we
+	 * reach here. However, we cannot know that and execute error recovery
+	 * again (that will not change anything).
+	 */
+	if (zonefs_zone_is_seq(z)) {
+		if (ret > 0 && ret != count)
+			ret = -EIO;
+		if (ret < 0 && ret != -EIOCBQUEUED)
+			zonefs_io_error(inode, true);
 	}
 
 inode_unlock:
@@ -550,7 +575,8 @@ static ssize_t zonefs_file_buffered_write(struct kiocb *iocb,
 	if (ret <= 0)
 		goto inode_unlock;
 
-	ret = iomap_file_buffered_write(iocb, from, &zonefs_write_iomap_ops);
+	ret = iomap_file_buffered_write(iocb, from, &zonefs_write_iomap_ops,
+			NULL, NULL);
 	if (ret == -EIO)
 		zonefs_io_error(inode, true);
 
@@ -837,7 +863,7 @@ const struct file_operations zonefs_file_operations = {
 	.open		= zonefs_file_open,
 	.release	= zonefs_file_release,
 	.fsync		= zonefs_file_fsync,
-	.mmap		= zonefs_file_mmap,
+	.mmap_prepare	= zonefs_file_mmap_prepare,
 	.llseek		= zonefs_file_llseek,
 	.read_iter	= zonefs_file_read_iter,
 	.write_iter	= zonefs_file_write_iter,