Re: [PATCH 07/17] writeback: Factor writeback_get_batch() out of write_cache_pages()

[Jan Kara <jack@suse.cz>]

On Mon 18-12-23 16:35:43, Christoph Hellwig wrote:
> From: "Matthew Wilcox (Oracle)" <willy@infradead.org>
> 
> This simple helper will be the basis of the writeback iterator.
> To make this work, we need to remember the current index
> and end positions in writeback_control.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> [hch: heavily rebased, add helpers to get the tag and end index,
>       don't keep the end index in struct writeback_control]
> Signed-off-by: Christoph Hellwig <hch@lst.de>

Just two nits below. However you decide about them feel free to add:

Reviewed-by: Jan Kara <jack@suse.cz>

> +static pgoff_t wbc_end(struct writeback_control *wbc)
> +{
> +	if (wbc->range_cyclic)
> +		return -1;
> +	return wbc->range_end >> PAGE_SHIFT;
> +}
> +
> +static void writeback_get_batch(struct address_space *mapping,
> +		struct writeback_control *wbc)
> +{
> +	folio_batch_release(&wbc->fbatch);
> +	cond_resched();

I'd prefer to have cond_resched() explicitely in the writeback loop instead
of hidden here in writeback_get_batch() where it logically does not make
too much sense to me...

> +	filemap_get_folios_tag(mapping, &wbc->index, wbc_end(wbc),
> +			wbc_to_tag(wbc), &wbc->fbatch);
> +}
> +
>  /**
>   * write_cache_pages - walk the list of dirty pages of the given address space and write all of them.
>   * @mapping: address space structure to write
> @@ -2419,38 +2442,30 @@ int write_cache_pages(struct address_space *mapping,
>  		      void *data)
>  {
>  	int error;
> -	int nr_folios;
> -	pgoff_t index;
>  	pgoff_t end;		/* Inclusive */
> -	xa_mark_t tag;
>  
>  	if (wbc->range_cyclic) {
> -		index = mapping->writeback_index; /* prev offset */
> +		wbc->index = mapping->writeback_index; /* prev offset */
>  		end = -1;
>  	} else {
> -		index = wbc->range_start >> PAGE_SHIFT;
> +		wbc->index = wbc->range_start >> PAGE_SHIFT;
>  		end = wbc->range_end >> PAGE_SHIFT;
>  	}

Maybe we should have:
	end = wbc_end(wbc);

when we have the helper? But I guess this gets cleaned up in later patches
anyway so whatever.

								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR