Merge patch series "iomap: incremental advance conversion -- phase 2"

Brian Foster <[e-mail]> says:

Here's phase 2 of the incremental iter advance conversions. This updates
all remaining iomap operations to advance the iter within the operation
and thus removes the need to advance from the core iomap iterator. Once
all operations are switched over, the core advance code is removed and
the processed field is renamed to reflect that it is now a pure status
code.

For context, this was first introduced in a previous series [1] that
focused mainly on the core mechanism and iomap buffered write. This is
because original impetus was to facilitate a folio batch mechanism where
a filesystem can optionally provide a batch of folios to process for a
given mapping (i.e. zero range of an unwritten mapping with dirty folios
in pagecache). That is still WIP, but the broader point is that this was
originally intended as an optional mode until consensus that fell out of
discussion was that it would be preferable to convert over everything.
This presumably facilitates some other future work and simplifies
semantics in the core iteration code.

Patches 1-3 convert over iomap buffered read, direct I/O and various
other remaining ops (swap, etc.). Patches 4-9 convert over the various
DAX iomap operations. Finally, patches 10-12 introduce some cleanups now
that all iomap operations have updated iteration semantics.

* patches from https://lore.kernel.org/r/[e-mail]:
iomap: introduce a full map advance helper
iomap: rename iomap_iter processed field to status
iomap: remove unnecessary advance from iomap_iter()
dax: advance the iomap_iter on pte and pmd faults
dax: advance the iomap_iter on dedupe range
dax: advance the iomap_iter on unshare range
dax: advance the iomap_iter on zero range
dax: push advance down into dax_iomap_iter() for read and write
dax: advance the iomap_iter in the read/write path
iomap: convert misc simple ops to incremental advance
iomap: advance the iter on direct I/O
iomap: advance the iter directly on buffered read

Link: https://lore.kernel.org/r/[e-mail]
Signed-off-by: Christian Brauner <[e-mail]>

fs/dax.c

···
 }
 #endif /* CONFIG_FS_DAX_PMD */
 
-static s64 dax_unshare_iter(struct iomap_iter *iter)
+static int dax_unshare_iter(struct iomap_iter *iter)
 {
 	struct iomap *iomap = &iter->iomap;
 	const struct iomap *srcmap = iomap_iter_srcmap(iter);
···
 	u64 copy_len = iomap_length(iter);
 	u32 mod;
 	int id = 0;
-	s64 ret = 0;
+	s64 ret;
 	void *daddr = NULL, *saddr = NULL;
 
 	if (!iomap_want_unshare_iter(iter))
-		return iomap_length(iter);
+		return iomap_iter_advance_full(iter);
 
 	/*
 	 * Extend the file range to be aligned to fsblock/pagesize, because
···
 	if (ret < 0)
 		goto out_unlock;
 
-	if (copy_mc_to_kernel(daddr, saddr, copy_len) == 0)
-		ret = iomap_length(iter);
-	else
+	if (copy_mc_to_kernel(daddr, saddr, copy_len) != 0)
 		ret = -EIO;
 
 out_unlock:
 	dax_read_unlock(id);
-	return dax_mem2blk_err(ret);
+	if (ret < 0)
+		return dax_mem2blk_err(ret);
+	return iomap_iter_advance_full(iter);
 }
 
 int dax_file_unshare(struct inode *inode, loff_t pos, loff_t len,
···
 
 	iter.len = min(len, size - pos);
 	while ((ret = iomap_iter(&iter, ops)) > 0)
-		iter.processed = dax_unshare_iter(&iter);
+		iter.status = dax_unshare_iter(&iter);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(dax_file_unshare);
···
 	return ret;
 }
 
-static s64 dax_zero_iter(struct iomap_iter *iter, bool *did_zero)
+static int dax_zero_iter(struct iomap_iter *iter, bool *did_zero)
 {
 	const struct iomap *iomap = &iter->iomap;
 	const struct iomap *srcmap = iomap_iter_srcmap(iter);
-	loff_t pos = iter->pos;
 	u64 length = iomap_length(iter);
-	s64 written = 0;
+	int ret;
 
 	/* already zeroed?  we're done. */
 	if (srcmap->type == IOMAP_HOLE || srcmap->type == IOMAP_UNWRITTEN)
-		return length;
+		return iomap_iter_advance(iter, &length);
 
 	/*
 	 * invalidate the pages whose sharing state is to be changed
···
 	 */
 	if (iomap->flags & IOMAP_F_SHARED)
 		invalidate_inode_pages2_range(iter->inode->i_mapping,
-					      pos >> PAGE_SHIFT,
-					      (pos + length - 1) >> PAGE_SHIFT);
+				iter->pos >> PAGE_SHIFT,
+				(iter->pos + length - 1) >> PAGE_SHIFT);
 
 	do {
+		loff_t pos = iter->pos;
 		unsigned offset = offset_in_page(pos);
-		unsigned size = min_t(u64, PAGE_SIZE - offset, length);
 		pgoff_t pgoff = dax_iomap_pgoff(iomap, pos);
-		long rc;
 		int id;
 
+		length = min_t(u64, PAGE_SIZE - offset, length);
+
 		id = dax_read_lock();
-		if (IS_ALIGNED(pos, PAGE_SIZE) && size == PAGE_SIZE)
-			rc = dax_zero_page_range(iomap->dax_dev, pgoff, 1);
+		if (IS_ALIGNED(pos, PAGE_SIZE) && length == PAGE_SIZE)
+			ret = dax_zero_page_range(iomap->dax_dev, pgoff, 1);
 		else
-			rc = dax_memzero(iter, pos, size);
+			ret = dax_memzero(iter, pos, length);
 		dax_read_unlock(id);
 
-		if (rc < 0)
-			return rc;
-		pos += size;
-		length -= size;
-		written += size;
+		if (ret < 0)
+			return ret;
+
+		ret = iomap_iter_advance(iter, &length);
+		if (ret)
+			return ret;
 	} while (length > 0);
 
 	if (did_zero)
 		*did_zero = true;
-	return written;
+	return ret;
 }
 
 int dax_zero_range(struct inode *inode, loff_t pos, loff_t len, bool *did_zero,
···
 	int ret;
 
 	while ((ret = iomap_iter(&iter, ops)) > 0)
-		iter.processed = dax_zero_iter(&iter, did_zero);
+		iter.status = dax_zero_iter(&iter, did_zero);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(dax_zero_range);
···
 }
 EXPORT_SYMBOL_GPL(dax_truncate_page);
 
-static loff_t dax_iomap_iter(const struct iomap_iter *iomi,
-		struct iov_iter *iter)
+static int dax_iomap_iter(struct iomap_iter *iomi, struct iov_iter *iter)
 {
 	const struct iomap *iomap = &iomi->iomap;
 	const struct iomap *srcmap = iomap_iter_srcmap(iomi);
···
 		if (pos >= end)
 			return 0;
 
-		if (iomap->type == IOMAP_HOLE || iomap->type == IOMAP_UNWRITTEN)
-			return iov_iter_zero(min(length, end - pos), iter);
+		if (iomap->type == IOMAP_HOLE || iomap->type == IOMAP_UNWRITTEN) {
+			done = iov_iter_zero(min(length, end - pos), iter);
+			return iomap_iter_advance(iomi, &done);
+		}
 	}
 
 	/*
···
 	}
 
 	id = dax_read_lock();
-	while (pos < end) {
+	while ((pos = iomi->pos) < end) {
 		unsigned offset = pos & (PAGE_SIZE - 1);
 		const size_t size = ALIGN(length + offset, PAGE_SIZE);
 		pgoff_t pgoff = dax_iomap_pgoff(iomap, pos);
···
 			xfer = dax_copy_to_iter(dax_dev, pgoff, kaddr,
 					map_len, iter);
 
-		pos += xfer;
-		length -= xfer;
-		done += xfer;
-
-		if (xfer == 0)
+		length = xfer;
+		ret = iomap_iter_advance(iomi, &length);
+		if (!ret && xfer == 0)
 			ret = -EFAULT;
 		if (xfer < map_len)
 			break;
 	}
 	dax_read_unlock(id);
 
-	return done ? done : ret;
+	return ret;
 }
 
 /**
···
 		iomi.flags |= IOMAP_NOWAIT;
 
 	while ((ret = iomap_iter(&iomi, ops)) > 0)
-		iomi.processed = dax_iomap_iter(&iomi, iter);
+		iomi.status = dax_iomap_iter(&iomi, iter);
 
 	done = iomi.pos - iocb->ki_pos;
 	iocb->ki_pos = iomi.pos;
···
 
 	while ((error = iomap_iter(&iter, ops)) > 0) {
 		if (WARN_ON_ONCE(iomap_length(&iter) < PAGE_SIZE)) {
-			iter.processed = -EIO;	/* fs corruption? */
+			iter.status = -EIO;	/* fs corruption? */
 			continue;
 		}
 
···
 			ret |= VM_FAULT_MAJOR;
 		}
 
-		if (!(ret & VM_FAULT_ERROR))
-			iter.processed = PAGE_SIZE;
+		if (!(ret & VM_FAULT_ERROR)) {
+			u64 length = PAGE_SIZE;
+			iter.status = iomap_iter_advance(&iter, &length);
+		}
 	}
 
 	if (iomap_errp)
···
 			continue; /* actually breaks out of the loop */
 
 		ret = dax_fault_iter(vmf, &iter, pfnp, &xas, &entry, true);
-		if (ret != VM_FAULT_FALLBACK)
-			iter.processed = PMD_SIZE;
+		if (ret != VM_FAULT_FALLBACK) {
+			u64 length = PMD_SIZE;
+			iter.status = iomap_iter_advance(&iter, &length);
+		}
 	}
 
 unlock_entry:
···
 }
 EXPORT_SYMBOL_GPL(dax_finish_sync_fault);
 
-static loff_t dax_range_compare_iter(struct iomap_iter *it_src,
+static int dax_range_compare_iter(struct iomap_iter *it_src,
 		struct iomap_iter *it_dest, u64 len, bool *same)
 {
 	const struct iomap *smap = &it_src->iomap;
 	const struct iomap *dmap = &it_dest->iomap;
 	loff_t pos1 = it_src->pos, pos2 = it_dest->pos;
+	u64 dest_len;
 	void *saddr, *daddr;
 	int id, ret;
 
···
 
 	if (smap->type == IOMAP_HOLE && dmap->type == IOMAP_HOLE) {
 		*same = true;
-		return len;
+		goto advance;
 	}
 
 	if (smap->type == IOMAP_HOLE || dmap->type == IOMAP_HOLE) {
···
 	if (!*same)
 		len = 0;
 	dax_read_unlock(id);
-	return len;
+
+advance:
+	dest_len = len;
+	ret = iomap_iter_advance(it_src, &len);
+	if (!ret)
+		ret = iomap_iter_advance(it_dest, &dest_len);
+	return ret;
 
 out_unlock:
 	dax_read_unlock(id);
···
 		.len		= len,
 		.flags		= IOMAP_DAX,
 	};
-	int ret, compared = 0;
+	int ret, status;
 
 	while ((ret = iomap_iter(&src_iter, ops)) > 0 &&
 	       (ret = iomap_iter(&dst_iter, ops)) > 0) {
-		compared = dax_range_compare_iter(&src_iter, &dst_iter,
+		status = dax_range_compare_iter(&src_iter, &dst_iter,
 				min(src_iter.len, dst_iter.len), same);
-		if (compared < 0)
+		if (status < 0)
 			return ret;
-		src_iter.processed = dst_iter.processed = compared;
+		src_iter.status = dst_iter.status = status;
 	}
 	return ret;
 }

fs/iomap/buffered-io.c

···
 		pos >= i_size_read(iter->inode);
 }
 
-static loff_t iomap_readpage_iter(const struct iomap_iter *iter,
-		struct iomap_readpage_ctx *ctx, loff_t offset)
+static int iomap_readpage_iter(struct iomap_iter *iter,
+		struct iomap_readpage_ctx *ctx)
 {
 	const struct iomap *iomap = &iter->iomap;
-	loff_t pos = iter->pos + offset;
-	loff_t length = iomap_length(iter) - offset;
+	loff_t pos = iter->pos;
+	loff_t length = iomap_length(iter);
 	struct folio *folio = ctx->cur_folio;
 	struct iomap_folio_state *ifs;
-	loff_t orig_pos = pos;
 	size_t poff, plen;
 	sector_t sector;
 
···
 	 * we can skip trailing ones as they will be handled in the next
 	 * iteration.
 	 */
-	return pos - orig_pos + plen;
+	length = pos - iter->pos + plen;
+	return iomap_iter_advance(iter, &length);
 }
 
-static loff_t iomap_read_folio_iter(const struct iomap_iter *iter,
+static int iomap_read_folio_iter(struct iomap_iter *iter,
 		struct iomap_readpage_ctx *ctx)
 {
-	struct folio *folio = ctx->cur_folio;
-	size_t offset = offset_in_folio(folio, iter->pos);
-	loff_t length = min_t(loff_t, folio_size(folio) - offset,
-			      iomap_length(iter));
-	loff_t done, ret;
+	int ret;
 
-	for (done = 0; done < length; done += ret) {
-		ret = iomap_readpage_iter(iter, ctx, done);
-		if (ret <= 0)
+	while (iomap_length(iter)) {
+		ret = iomap_readpage_iter(iter, ctx);
+		if (ret)
 			return ret;
 	}
 
-	return done;
+	return 0;
 }
 
 int iomap_read_folio(struct folio *folio, const struct iomap_ops *ops)
···
 	trace_iomap_readpage(iter.inode, 1);
 
 	while ((ret = iomap_iter(&iter, ops)) > 0)
-		iter.processed = iomap_read_folio_iter(&iter, &ctx);
+		iter.status = iomap_read_folio_iter(&iter, &ctx);
 
 	if (ctx.bio) {
 		submit_bio(ctx.bio);
···
 }
 EXPORT_SYMBOL_GPL(iomap_read_folio);
 
-static loff_t iomap_readahead_iter(const struct iomap_iter *iter,
+static int iomap_readahead_iter(struct iomap_iter *iter,
 		struct iomap_readpage_ctx *ctx)
 {
-	loff_t length = iomap_length(iter);
-	loff_t done, ret;
+	int ret;
 
-	for (done = 0; done < length; done += ret) {
+	while (iomap_length(iter)) {
 		if (ctx->cur_folio &&
-		    offset_in_folio(ctx->cur_folio, iter->pos + done) == 0) {
+		    offset_in_folio(ctx->cur_folio, iter->pos) == 0) {
 			if (!ctx->cur_folio_in_bio)
 				folio_unlock(ctx->cur_folio);
 			ctx->cur_folio = NULL;
···
 			ctx->cur_folio = readahead_folio(ctx->rac);
 			ctx->cur_folio_in_bio = false;
 		}
-		ret = iomap_readpage_iter(iter, ctx, done);
-		if (ret <= 0)
+		ret = iomap_readpage_iter(iter, ctx);
+		if (ret)
 			return ret;
 	}
 
-	return done;
+	return 0;
 }
 
 /**
···
 	trace_iomap_readahead(rac->mapping->host, readahead_count(rac));
 
 	while (iomap_iter(&iter, ops) > 0)
-		iter.processed = iomap_readahead_iter(&iter, &ctx);
+		iter.status = iomap_readahead_iter(&iter, &ctx);
 
 	if (ctx.bio)
 		submit_bio(ctx.bio);
···
 	return __iomap_write_end(iter->inode, pos, len, copied, folio);
 }
 
-static loff_t iomap_write_iter(struct iomap_iter *iter, struct iov_iter *i)
+static int iomap_write_iter(struct iomap_iter *iter, struct iov_iter *i)
 {
 	ssize_t total_written = 0;
-	long status = 0;
+	int status = 0;
 	struct address_space *mapping = iter->inode->i_mapping;
 	size_t chunk = mapping_max_folio_size(mapping);
 	unsigned int bdp_flags = (iter->flags & IOMAP_NOWAIT) ? BDP_ASYNC : 0;
···
 		iter.flags |= IOMAP_NOWAIT;
 
 	while ((ret = iomap_iter(&iter, ops)) > 0)
-		iter.processed = iomap_write_iter(&iter, i);
+		iter.status = iomap_write_iter(&iter, i);
 
 	if (unlikely(iter.pos == iocb->ki_pos))
 		return ret;
···
 }
 EXPORT_SYMBOL_GPL(iomap_write_delalloc_release);
 
-static loff_t iomap_unshare_iter(struct iomap_iter *iter)
+static int iomap_unshare_iter(struct iomap_iter *iter)
 {
 	struct iomap *iomap = &iter->iomap;
 	u64 bytes = iomap_length(iter);
···
 
 	iter.len = min(len, size - pos);
 	while ((ret = iomap_iter(&iter, ops)) > 0)
-		iter.processed = iomap_unshare_iter(&iter);
+		iter.status = iomap_unshare_iter(&iter);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(iomap_file_unshare);
···
 	return filemap_write_and_wait_range(mapping, i->pos, end);
 }
 
-static loff_t iomap_zero_iter(struct iomap_iter *iter, bool *did_zero)
+static int iomap_zero_iter(struct iomap_iter *iter, bool *did_zero)
 {
 	u64 bytes = iomap_length(iter);
 	int status;
···
 	    filemap_range_needs_writeback(mapping, pos, pos + plen - 1)) {
 		iter.len = plen;
 		while ((ret = iomap_iter(&iter, ops)) > 0)
-			iter.processed = iomap_zero_iter(&iter, did_zero);
+			iter.status = iomap_zero_iter(&iter, did_zero);
 
 		iter.len = len - (iter.pos - pos);
 		if (ret || !iter.len)
···
 
 		if (srcmap->type == IOMAP_HOLE ||
 		    srcmap->type == IOMAP_UNWRITTEN) {
-			s64 proc;
+			s64 status;
 
 			if (range_dirty) {
 				range_dirty = false;
-				proc = iomap_zero_iter_flush_and_stale(&iter);
+				status = iomap_zero_iter_flush_and_stale(&iter);
 			} else {
-				u64 length = iomap_length(&iter);
-				proc = iomap_iter_advance(&iter, &length);
+				status = iomap_iter_advance_full(&iter);
 			}
-			iter.processed = proc;
+			iter.status = status;
 			continue;
 		}
 
-		iter.processed = iomap_zero_iter(&iter, did_zero);
+		iter.status = iomap_zero_iter(&iter, did_zero);
 	}
 	return ret;
 }
···
 }
 EXPORT_SYMBOL_GPL(iomap_truncate_page);
 
-static loff_t iomap_folio_mkwrite_iter(struct iomap_iter *iter,
+static int iomap_folio_mkwrite_iter(struct iomap_iter *iter,
 		struct folio *folio)
 {
 	loff_t length = iomap_length(iter);
···
 		folio_mark_dirty(folio);
 	}
 
-	return length;
+	return iomap_iter_advance(iter, &length);
 }
 
 vm_fault_t iomap_page_mkwrite(struct vm_fault *vmf, const struct iomap_ops *ops,
···
 	iter.pos = folio_pos(folio);
 	iter.len = ret;
 	while ((ret = iomap_iter(&iter, ops)) > 0)
-		iter.processed = iomap_folio_mkwrite_iter(&iter, folio);
+		iter.status = iomap_folio_mkwrite_iter(&iter, folio);
 
 	if (ret < 0)
 		goto out_unlock;

fs/iomap/direct-io.c

···
 	return opflags;
 }
 
-static loff_t iomap_dio_bio_iter(const struct iomap_iter *iter,
-		struct iomap_dio *dio)
+static int iomap_dio_bio_iter(struct iomap_iter *iter, struct iomap_dio *dio)
 {
 	const struct iomap *iomap = &iter->iomap;
 	struct inode *inode = iter->inode;
···
 	bool need_zeroout = false;
 	bool use_fua = false;
 	int nr_pages, ret = 0;
-	size_t copied = 0;
+	u64 copied = 0;
 	size_t orig_count;
 
 	if (atomic && length != fs_block_size)
···
 	/* Undo iter limitation to current extent */
 	iov_iter_reexpand(dio->submit.iter, orig_count - copied);
 	if (copied)
-		return copied;
+		return iomap_iter_advance(iter, &copied);
 	return ret;
 }
 
-static loff_t iomap_dio_hole_iter(const struct iomap_iter *iter,
-		struct iomap_dio *dio)
+static int iomap_dio_hole_iter(struct iomap_iter *iter, struct iomap_dio *dio)
 {
 	loff_t length = iov_iter_zero(iomap_length(iter), dio->submit.iter);
 
 	dio->size += length;
 	if (!length)
 		return -EFAULT;
-	return length;
+	return iomap_iter_advance(iter, &length);
 }
 
-static loff_t iomap_dio_inline_iter(const struct iomap_iter *iomi,
-		struct iomap_dio *dio)
+static int iomap_dio_inline_iter(struct iomap_iter *iomi, struct iomap_dio *dio)
 {
 	const struct iomap *iomap = &iomi->iomap;
 	struct iov_iter *iter = dio->submit.iter;
 	void *inline_data = iomap_inline_data(iomap, iomi->pos);
 	loff_t length = iomap_length(iomi);
 	loff_t pos = iomi->pos;
-	size_t copied;
+	u64 copied;
 
 	if (WARN_ON_ONCE(!iomap_inline_data_valid(iomap)))
 		return -EIO;
···
 	dio->size += copied;
 	if (!copied)
 		return -EFAULT;
-	return copied;
+	return iomap_iter_advance(iomi, &copied);
 }
 
-static loff_t iomap_dio_iter(const struct iomap_iter *iter,
-		struct iomap_dio *dio)
+static int iomap_dio_iter(struct iomap_iter *iter, struct iomap_dio *dio)
 {
 	switch (iter->iomap.type) {
 	case IOMAP_HOLE:
···
 
 	blk_start_plug(&plug);
 	while ((ret = iomap_iter(&iomi, ops)) > 0) {
-		iomi.processed = iomap_dio_iter(&iomi, dio);
+		iomi.status = iomap_dio_iter(&iomi, dio);
 
 		/*
 		 * We can only poll for single bio I/Os.

fs/iomap/fiemap.c

···
 			iomap->length, flags);
 }
 
-static loff_t iomap_fiemap_iter(const struct iomap_iter *iter,
+static int iomap_fiemap_iter(struct iomap_iter *iter,
 		struct fiemap_extent_info *fi, struct iomap *prev)
 {
 	int ret;
 
 	if (iter->iomap.type == IOMAP_HOLE)
-		return iomap_length(iter);
+		goto advance;
 
 	ret = iomap_to_fiemap(fi, prev, 0);
 	*prev = iter->iomap;
-	switch (ret) {
-	case 0:		/* success */
-		return iomap_length(iter);
-	case 1:		/* extent array full */
-		return 0;
-	default:	/* error */
+	if (ret < 0)
 		return ret;
-	}
+	if (ret == 1)	/* extent array full */
+		return 0;
+
+advance:
+	return iomap_iter_advance_full(iter);
 }
 
 int iomap_fiemap(struct inode *inode, struct fiemap_extent_info *fi,
···
 		return ret;
 
 	while ((ret = iomap_iter(&iter, ops)) > 0)
-		iter.processed = iomap_fiemap_iter(&iter, fi, &prev);
+		iter.status = iomap_fiemap_iter(&iter, fi, &prev);
 
 	if (prev.type != IOMAP_HOLE) {
 		ret = iomap_to_fiemap(fi, &prev, FIEMAP_EXTENT_LAST);
···
 	while ((ret = iomap_iter(&iter, ops)) > 0) {
 		if (iter.iomap.type == IOMAP_MAPPED)
 			bno = iomap_sector(&iter.iomap, iter.pos) >> blkshift;
-		/* leave iter.processed unset to abort loop */
+		/* leave iter.status unset to abort loop */
 	}
 	if (ret)
 		return 0;

fs/iomap/iter.c

···
 
 static inline void iomap_iter_reset_iomap(struct iomap_iter *iter)
 {
-	iter->processed = 0;
+	iter->status = 0;
 	memset(&iter->iomap, 0, sizeof(iter->iomap));
 	memset(&iter->srcmap, 0, sizeof(iter->srcmap));
 }
···
  * function must be called in a loop that continues as long it returns a
  * positive value.  If 0 or a negative value is returned, the caller must not
  * return to the loop body.  Within a loop body, there are two ways to break out
- * of the loop body:  leave @iter.processed unchanged, or set it to a negative
+ * of the loop body:  leave @iter.status unchanged, or set it to a negative
  * errno.
  */
 int iomap_iter(struct iomap_iter *iter, const struct iomap_ops *ops)
 {
 	bool stale = iter->iomap.flags & IOMAP_F_STALE;
-	ssize_t advanced = iter->processed > 0 ? iter->processed : 0;
-	u64 olen = iter->len;
-	s64 processed;
+	ssize_t advanced;
+	u64 olen;
 	int ret;
 
 	trace_iomap_iter(iter, ops, _RET_IP_);
···
 		goto begin;
 
 	/*
-	 * If iter.processed is zero, the op may still have advanced the iter
-	 * itself. Calculate the advanced and original length bytes based on how
-	 * far pos has advanced for ->iomap_end().
+	 * Calculate how far the iter was advanced and the original length bytes
+	 * for ->iomap_end().
 	 */
-	if (!advanced) {
-		advanced = iter->pos - iter->iter_start_pos;
-		olen += advanced;
-	}
+	advanced = iter->pos - iter->iter_start_pos;
+	olen = iter->len + advanced;
 
 	if (ops->iomap_end) {
 		ret = ops->iomap_end(iter->inode, iter->iter_start_pos,
···
 			return ret;
 	}
 
-	processed = iter->processed;
-	if (processed < 0) {
-		iomap_iter_reset_iomap(iter);
-		return processed;
-	}
+	/* detect old return semantics where this would advance */
+	if (WARN_ON_ONCE(iter->status > 0))
+		iter->status = -EIO;
 
 	/*
-	 * Advance the iter and clear state from the previous iteration. This
-	 * passes iter->processed because that reflects the bytes processed but
-	 * not yet advanced by the iter handler.
-	 *
 	 * Use iter->len to determine whether to continue onto the next mapping.
-	 * Explicitly terminate in the case where the current iter has not
+	 * Explicitly terminate on error status or if the current iter has not
 	 * advanced at all (i.e. no work was done for some reason) unless the
 	 * mapping has been marked stale and needs to be reprocessed.
 	 */
-	ret = iomap_iter_advance(iter, &processed);
-	if (!ret && iter->len > 0)
-		ret = 1;
-	if (ret > 0 && !advanced && !stale)
+	if (iter->status < 0)
+		ret = iter->status;
+	else if (iter->len == 0 || (!advanced && !stale))
 		ret = 0;
+	else
+		ret = 1;
 	iomap_iter_reset_iomap(iter);
 	if (ret <= 0)
 		return ret;

fs/iomap/seek.c

···
 #include <linux/pagemap.h>
 #include <linux/pagevec.h>
 
-static loff_t iomap_seek_hole_iter(const struct iomap_iter *iter,
+static int iomap_seek_hole_iter(struct iomap_iter *iter,
 		loff_t *hole_pos)
 {
 	loff_t length = iomap_length(iter);
···
 		*hole_pos = mapping_seek_hole_data(iter->inode->i_mapping,
 				iter->pos, iter->pos + length, SEEK_HOLE);
 		if (*hole_pos == iter->pos + length)
-			return length;
+			return iomap_iter_advance(iter, &length);
 		return 0;
 	case IOMAP_HOLE:
 		*hole_pos = iter->pos;
 		return 0;
 	default:
-		return length;
+		return iomap_iter_advance(iter, &length);
 	}
 }
 
···
 
 	iter.len = size - pos;
 	while ((ret = iomap_iter(&iter, ops)) > 0)
-		iter.processed = iomap_seek_hole_iter(&iter, &pos);
+		iter.status = iomap_seek_hole_iter(&iter, &pos);
 	if (ret < 0)
 		return ret;
 	if (iter.len) /* found hole before EOF */
···
 }
 EXPORT_SYMBOL_GPL(iomap_seek_hole);
 
-static loff_t iomap_seek_data_iter(const struct iomap_iter *iter,
+static int iomap_seek_data_iter(struct iomap_iter *iter,
 		loff_t *hole_pos)
 {
 	loff_t length = iomap_length(iter);
 
 	switch (iter->iomap.type) {
 	case IOMAP_HOLE:
-		return length;
+		return iomap_iter_advance(iter, &length);
 	case IOMAP_UNWRITTEN:
 		*hole_pos = mapping_seek_hole_data(iter->inode->i_mapping,
 				iter->pos, iter->pos + length, SEEK_DATA);
 		if (*hole_pos < 0)
-			return length;
+			return iomap_iter_advance(iter, &length);
 		return 0;
 	default:
 		*hole_pos = iter->pos;
···
 
 	iter.len = size - pos;
 	while ((ret = iomap_iter(&iter, ops)) > 0)
-		iter.processed = iomap_seek_data_iter(&iter, &pos);
+		iter.status = iomap_seek_data_iter(&iter, &pos);
 	if (ret < 0)
 		return ret;
 	if (iter.len) /* found data before EOF */

fs/iomap/swapfile.c

···
  * swap only cares about contiguous page-aligned physical extents and makes no
  * distinction between written and unwritten extents.
  */
-static loff_t iomap_swapfile_iter(const struct iomap_iter *iter,
+static int iomap_swapfile_iter(struct iomap_iter *iter,
 		struct iomap *iomap, struct iomap_swapfile_info *isi)
 {
 	switch (iomap->type) {
···
 			return error;
 		memcpy(&isi->iomap, iomap, sizeof(isi->iomap));
 	}
-	return iomap_length(iter);
+
+	return iomap_iter_advance_full(iter);
 }
 
 /*
···
 		return ret;
 
 	while ((ret = iomap_iter(&iter, ops)) > 0)
-		iter.processed = iomap_swapfile_iter(&iter, &iter.iomap, &isi);
+		iter.status = iomap_swapfile_iter(&iter, &iter.iomap, &isi);
 	if (ret < 0)
 		return ret;
 

fs/iomap/trace.h

···
 		__field(u64, ino)
 		__field(loff_t, pos)
 		__field(u64, length)
-		__field(s64, processed)
+		__field(int, status)
 		__field(unsigned int, flags)
 		__field(const void *, ops)
 		__field(unsigned long, caller)
···
 		__entry->ino = iter->inode->i_ino;
 		__entry->pos = iter->pos;
 		__entry->length = iomap_length(iter);
-		__entry->processed = iter->processed;
+		__entry->status = iter->status;
 		__entry->flags = iter->flags;
 		__entry->ops = ops;
 		__entry->caller = caller;
 	),
-	TP_printk("dev %d:%d ino 0x%llx pos 0x%llx length 0x%llx processed %lld flags %s (0x%x) ops %ps caller %pS",
+	TP_printk("dev %d:%d ino 0x%llx pos 0x%llx length 0x%llx status %d flags %s (0x%x) ops %ps caller %pS",
 		  MAJOR(__entry->dev), MINOR(__entry->dev),
 		   __entry->ino,
 		   __entry->pos,
 		   __entry->length,
-		   __entry->processed,
+		   __entry->status,
 		   __print_flags(__entry->flags, "|", IOMAP_FLAGS_STRINGS),
 		   __entry->flags,
 		   __entry->ops,

include/linux/iomap.h

···
  *	It is updated at the same time as @pos.
  * @iter_start_pos: The original start pos for the current iomap. Used for
  *	incremental iter advance.
- * @processed: The number of bytes the most recent iteration needs iomap_iter()
- *	to advance the iter, zero if the iter was already advanced, or a
- *	negative errno for an error during the operation.
+ * @status: Status of the most recent iteration. Zero on success or a negative
+ *	errno on error.
  * @flags: Zero or more of the iomap_begin flags above.
  * @iomap: Map describing the I/O iteration
  * @srcmap: Source map for COW operations
···
 	loff_t pos;
 	u64 len;
 	loff_t iter_start_pos;
-	s64 processed;
+	int status;
 	unsigned flags;
 	struct iomap iomap;
 	struct iomap srcmap;
···
 static inline u64 iomap_length(const struct iomap_iter *iter)
 {
 	return iomap_length_trim(iter, iter->pos, iter->len);
+}
+
+/**
+ * iomap_iter_advance_full - advance by the full length of current map
+ */
+static inline int iomap_iter_advance_full(struct iomap_iter *iter)
+{
+	u64 length = iomap_length(iter);
+
+	return iomap_iter_advance(iter, &length);
 }
 
 /**