Received-SPF: pass (google.com: domain of linux-kernel+bounces-1010-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
Date: Fri, 15 Dec 2023 14:26:39 +0100
From: Jan Kara <jack@suse.cz>
To: Christoph Hellwig <hch@lst.de>
Cc: linux-mm@kvack.org, "Matthew Wilcox (Oracle)" <willy@infradead.org>,
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	Jan Kara <jack@suse.com>, David Howells <dhowells@redhat.com>
Subject: Re: [PATCH 01/11] writeback: Factor out writeback_finish()
Message-ID: <20231215132639.ftis3fhmcqkhrpzo@quack3>
References: <20231214132544.376574-1-hch@lst.de>
 <20231214132544.376574-2-hch@lst.de>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20231214132544.376574-2-hch@lst.de>

On Thu 14-12-23 14:25:34, Christoph Hellwig wrote:
> From: "Matthew Wilcox (Oracle)" <willy@infradead.org>
> 
> Instead of having a 'done' variable that controls the nested loops,
> have a writeback_finish() that can be returned directly.  This involves
> keeping more things in writeback_control, but it's just moving stuff
> allocated on the stack to being allocated slightly earlier on the stack.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> [hch: reorder and comment struct writeback_control]
> Signed-off-by: Christoph Hellwig <hch@lst.de>
> ---
>  include/linux/writeback.h |  8 +++++
>  mm/page-writeback.c       | 72 +++++++++++++++++++++------------------
>  2 files changed, 46 insertions(+), 34 deletions(-)
> 
> diff --git a/include/linux/writeback.h b/include/linux/writeback.h
> index 083387c00f0c8b..05e8add4b5ae3c 100644
> --- a/include/linux/writeback.h
> +++ b/include/linux/writeback.h
> @@ -11,6 +11,7 @@
>  #include <linux/flex_proportions.h>
>  #include <linux/backing-dev-defs.h>
>  #include <linux/blk_types.h>
> +#include <linux/pagevec.h>
>  
>  struct bio;
>  
> @@ -40,6 +41,7 @@ enum writeback_sync_modes {
>   * in a manner such that unspecified fields are set to zero.
>   */
>  struct writeback_control {
> +	/* public fields that can be set and/or consumed by the caller: */
>  	long nr_to_write;		/* Write this many pages, and decrement
>  					   this for each page written */
>  	long pages_skipped;		/* Pages which were not written */
> @@ -77,6 +79,12 @@ struct writeback_control {
>  	 */
>  	struct swap_iocb **swap_plug;
>  
> +	/* internal fields used by the ->writepages implementation: */
> +	struct folio_batch fbatch;
> +	pgoff_t done_index;
> +	int err;
> +	unsigned range_whole:1;		/* entire file */

Do we really need the range_whole member here? It is trivially derived from
range_start && range_end and used only in one place in writeback_finish().

> diff --git a/mm/page-writeback.c b/mm/page-writeback.c
> index ee2fd6a6af4072..45309f3b8193f8 100644
> --- a/mm/page-writeback.c
> +++ b/mm/page-writeback.c
> @@ -2360,6 +2360,24 @@ void tag_pages_for_writeback(struct address_space *mapping,
>  }
>  EXPORT_SYMBOL(tag_pages_for_writeback);
>  
> +static int writeback_finish(struct address_space *mapping,
> +		struct writeback_control *wbc, bool done)
> +{
> +	folio_batch_release(&wbc->fbatch);
> +
> +	/*
> +	 * If we hit the last page and there is more work to be done:
> +	 * wrap the index back to the start of the file for the next
> +	 * time we are called.
> +	 */
> +	if (wbc->range_cyclic && !done)
> +		wbc->done_index = 0;
> +	if (wbc->range_cyclic || (wbc->range_whole && wbc->nr_to_write > 0))
> +		mapping->writeback_index = wbc->done_index;
> +
> +	return wbc->err;
> +}

Also I suspect we can get rid of the 'done' argument here. After all it
just controls whether we cycle back to index 0 which we could just
determine as:

	if (wbc->range_cyclic && !wbc->err && wbc->nr_to_write > 0) {
		WARN_ON_ONCE(wbc->sync_mode != WB_SYNC_NONE);
		wbc->done_index = 0;
	}

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