Re: [RFC v2 4/5] mpage: use folios in bio end_io handler

[Matthew Wilcox <willy@infradead.org>]

On Wed, Mar 22, 2023 at 02:50:12PM +0100, Pankaj Raghav wrote:
>  static void mpage_write_end_io(struct bio *bio)
>  {
> -	struct bio_vec *bv;
> -	struct bvec_iter_all iter_all;
> +	struct folio_iter fi;
> +	int err = blk_status_to_errno(bio->bi_status);
>  
> -	bio_for_each_segment_all(bv, bio, iter_all) {
> -		struct page *page = bv->bv_page;
> -		page_endio(page, REQ_OP_WRITE,
> -			   blk_status_to_errno(bio->bi_status));
> +	bio_for_each_folio_all(fi, bio) {
> +		struct folio *folio = fi.folio;
> +
> +		if (err) {
> +			struct address_space *mapping;
> +
> +			folio_set_error(folio);
> +			mapping = folio_mapping(folio);
> +			if (mapping)
> +				mapping_set_error(mapping, err);

The folio is known to belong to this mapping and can't be truncated
while under writeback.  So it's safe to do:

			folio_set_error(folio);
			mapping_set_error(folio->mapping, err);

I'm not even sure I'd bother to pull folio out of the fi.

	bio_for_each_folio_all(fi, bio) {
		if (err) {
			folio_set_error(fi.folio);
			mapping_set_error(fi.folio->mapping, err);
		}
		folio_end_writeback(fi.folio);
	}