Received-SPF: pass (google.com: domain of linux-kernel+bounces-64423-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Date: Wed, 14 Feb 2024 09:05:09 +1100
From: Dave Chinner <david@fromorbit.com>
To: "Pankaj Raghav (Samsung)" <kernel@pankajraghav.com>
Cc: linux-xfs@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	mcgrof@kernel.org, gost.dev@samsung.com, akpm@linux-foundation.org,
	kbusch@kernel.org, djwong@kernel.org, chandan.babu@oracle.com,
	p.raghav@samsung.com, linux-kernel@vger.kernel.org, hare@suse.de,
	willy@infradead.org, linux-mm@kvack.org
Subject: Re: [RFC v2 03/14] filemap: use mapping_min_order while allocating
 folios
Message-ID: <ZcvnlfyaBRhWaIzD@dread.disaster.area>
References: <20240213093713.1753368-1-kernel@pankajraghav.com>
 <20240213093713.1753368-4-kernel@pankajraghav.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20240213093713.1753368-4-kernel@pankajraghav.com>

On Tue, Feb 13, 2024 at 10:37:02AM +0100, Pankaj Raghav (Samsung) wrote:
> From: Pankaj Raghav <p.raghav@samsung.com>
> 
> filemap_create_folio() and do_read_cache_folio() were always allocating
> folio of order 0. __filemap_get_folio was trying to allocate higher
> order folios when fgp_flags had higher order hint set but it will default
> to order 0 folio if higher order memory allocation fails.
> 
> As we bring the notion of mapping_min_order, make sure these functions
> allocate at least folio of mapping_min_order as we need to guarantee it
> in the page cache.
> 
> Add some additional VM_BUG_ON() in page_cache_delete[batch] and
> __filemap_add_folio to catch errors where we delete or add folios that
> has order less than min_order.
> 
> Signed-off-by: Pankaj Raghav <p.raghav@samsung.com>
> Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
> ---
>  mm/filemap.c | 25 +++++++++++++++++++++----
>  1 file changed, 21 insertions(+), 4 deletions(-)
> 
> diff --git a/mm/filemap.c b/mm/filemap.c
> index 323a8e169581..7a6e15c47150 100644
> --- a/mm/filemap.c
> +++ b/mm/filemap.c
> @@ -127,6 +127,7 @@
>  static void page_cache_delete(struct address_space *mapping,
>  				   struct folio *folio, void *shadow)
>  {
> +	unsigned int min_order = mapping_min_folio_order(mapping);
>  	XA_STATE(xas, &mapping->i_pages, folio->index);
>  	long nr = 1;
>  
> @@ -135,6 +136,7 @@ static void page_cache_delete(struct address_space *mapping,
>  	xas_set_order(&xas, folio->index, folio_order(folio));
>  	nr = folio_nr_pages(folio);
>  
> +	VM_BUG_ON_FOLIO(folio_order(folio) < min_order, folio);
>  	VM_BUG_ON_FOLIO(!folio_test_locked(folio), folio);

If you are only using min_order in the VM_BUG_ON_FOLIO() macro, then
please just do:

	VM_BUG_ON_FOLIO(folio_order(folio) < mapping_min_folio_order(mapping),
			folio);

There is no need to clutter up the function with variables that are
only used in one debug-only check.

> @@ -1847,6 +1853,10 @@ struct folio *__filemap_get_folio(struct address_space *mapping, pgoff_t index,
>  		fgf_t fgp_flags, gfp_t gfp)
>  {
>  	struct folio *folio;
> +	unsigned int min_order = mapping_min_folio_order(mapping);
> +	unsigned int min_nrpages = mapping_min_folio_nrpages(mapping);
> +
> +	index = round_down(index, min_nrpages);

	index = mapping_align_start_index(mapping, index);

The rest of the function only cares about min_order, not
min_nrpages....

-Dave.
-- 
Dave Chinner
david@fromorbit.com