Re: [PATCH v2] filesystem-dax: Fix dax_layout_busy_page() livelock

[Jan Kara]

On Sat 06-10-18 22:59:56, Dan Williams wrote:
> In the presence of multi-order entries the typical
> pagevec_lookup_entries() pattern may loop forever:
> 
> 	while (index < end && pagevec_lookup_entries(&pvec, mapping, index,
> 				min(end - index, (pgoff_t)PAGEVEC_SIZE),
> 				indices)) {
> 		...
> 		for (i = 0; i < pagevec_count(&pvec); i++) {
> 			index = indices[i];
> 			...
> 		}
> 		index++; /* BUG */
> 	}
> 
> The loop updates 'index' for each index found and then increments to the
> next possible page to continue the lookup. However, if the last entry in
> the pagevec is multi-order then the next possible page index is more
> than 1 page away. Fix this locally for the filesystem-dax case by
> checking for dax-multi-order entries. Going forward new users of
> multi-order entries need to be similarly careful, or we need a generic
> way to report the page increment in the radix iterator.
> 
> Fixes: 5fac7408d828 ("mm, fs, dax: handle layout changes to pinned dax...")
> Cc: <stable@xxxxxxxxxxxxxxx>
> Cc: Jan Kara <jack@xxxxxxx>
> Cc: Ross Zwisler <zwisler@xxxxxxxxxx>
> Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
> Signed-off-by: Dan Williams <dan.j.williams@xxxxxxxxx>
> ---
> Changes in v2:
> * Only update nr_pages if the last entry in the pagevec is multi-order.
> 
>  fs/dax.c |   13 +++++++++++--
>  1 file changed, 11 insertions(+), 2 deletions(-)

Thanks for fixing this up! It is somewhat ugly but nicer fixes would be
more intrusive so I agree with this approach for the ease of backporting
and let's clean up the iteration code later. Feel free to add:

Reviewed-by: Jan Kara <jack@xxxxxxx>

								Honza

> 
> diff --git a/fs/dax.c b/fs/dax.c
> index 4becbf168b7f..0fb270f0a0ef 100644
> --- a/fs/dax.c
> +++ b/fs/dax.c
> @@ -666,6 +666,8 @@ struct page *dax_layout_busy_page(struct address_space *mapping)
>  	while (index < end && pagevec_lookup_entries(&pvec, mapping, index,
>  				min(end - index, (pgoff_t)PAGEVEC_SIZE),
>  				indices)) {
> +		pgoff_t nr_pages = 1;
> +
>  		for (i = 0; i < pagevec_count(&pvec); i++) {
>  			struct page *pvec_ent = pvec.pages[i];
>  			void *entry;
> @@ -680,8 +682,15 @@ struct page *dax_layout_busy_page(struct address_space *mapping)
>  
>  			xa_lock_irq(&mapping->i_pages);
>  			entry = get_unlocked_mapping_entry(mapping, index, NULL);
> -			if (entry)
> +			if (entry) {
>  				page = dax_busy_page(entry);
> +				/*
> +				 * Account for multi-order entries at
> +				 * the end of the pagevec.
> +				 */
> +				if (i + 1 >= pagevec_count(&pvec))
> +					nr_pages = 1UL << dax_radix_order(entry);
> +			}
>  			put_unlocked_mapping_entry(mapping, index, entry);
>  			xa_unlock_irq(&mapping->i_pages);
>  			if (page)
> @@ -696,7 +705,7 @@ struct page *dax_layout_busy_page(struct address_space *mapping)
>  		 */
>  		pagevec_remove_exceptionals(&pvec);
>  		pagevec_release(&pvec);
> -		index++;
> +		index += nr_pages;
>  
>  		if (page)
>  			break;
> 
-- 
Jan Kara <jack@xxxxxxxx>
SUSE Labs, CR