Re: [PATCH 10/11] Direct compact when a high-order allocation fails

[Andrew Morton <akpm@linux-foundation.org>]

On Tue, 23 Mar 2010 12:25:45 +0000
Mel Gorman <mel@csn.ul.ie> wrote:

> Ordinarily when a high-order allocation fails, direct reclaim is entered to
> free pages to satisfy the allocation.  With this patch, it is determined if
> an allocation failed due to external fragmentation instead of low memory
> and if so, the calling process will compact until a suitable page is
> freed. Compaction by moving pages in memory is considerably cheaper than
> paging out to disk and works where there are locked pages or no swap. If
> compaction fails to free a page of a suitable size, then reclaim will
> still occur.
> 
> Direct compaction returns as soon as possible. As each block is compacted,
> it is checked if a suitable page has been freed and if so, it returns.
> 
>
> ...
>
> +static inline unsigned long compact_zone_order(struct zone *zone,
> +						int order, gfp_t gfp_mask)

Suggest that you re-review all the manual inlining in the patchset. 
It's rarely needed and often incorrect.

> +{
> +	struct compact_control cc = {
> +		.nr_freepages = 0,
> +		.nr_migratepages = 0,
> +		.order = order,
> +		.migratetype = allocflags_to_migratetype(gfp_mask),
> +		.zone = zone,
> +	};
> +	INIT_LIST_HEAD(&cc.freepages);
> +	INIT_LIST_HEAD(&cc.migratepages);
> +
> +	return compact_zone(zone, &cc);
> +}
> +
> +/**
> + * try_to_compact_pages - Direct compact to satisfy a high-order allocation
> + * @zonelist: The zonelist used for the current allocation
> + * @order: The order of the current allocation
> + * @gfp_mask: The GFP mask of the current allocation
> + * @nodemask: The allowed nodes to allocate from
> + *
> + * This is the main entry point for direct page compaction.
> + */
> +unsigned long try_to_compact_pages(struct zonelist *zonelist,
> +			int order, gfp_t gfp_mask, nodemask_t *nodemask)
> +{
> +	enum zone_type high_zoneidx = gfp_zone(gfp_mask);
> +	int may_enter_fs = gfp_mask & __GFP_FS;
> +	int may_perform_io = gfp_mask & __GFP_IO;
> +	unsigned long watermark;
> +	struct zoneref *z;
> +	struct zone *zone;
> +	int rc = COMPACT_INCOMPLETE;
> +
> +	/* Check whether it is worth even starting compaction */
> +	if (order == 0 || !may_enter_fs || !may_perform_io)
> +		return rc;

hm, that was sad.  All those darn wireless drivers doing their
high-order GFP_ATOMIC allocations cannot be saved?

> +	/*
> +	 * We will not stall if the necessary conditions are not met for
> +	 * migration but direct reclaim seems to account stalls similarly
> +	 */
> +	count_vm_event(COMPACTSTALL);
> +
> +	/* Compact each zone in the list */
> +	for_each_zone_zonelist_nodemask(zone, z, zonelist, high_zoneidx,
> +								nodemask) {

Will all of this code play nicely with memory hotplug?

> +		int fragindex;
> +		int status;
> +
> +		/*
> +		 * Watermarks for order-0 must be met for compaction. Note
> +		 * the 2UL. This is because during migration, copies of
> +		 * pages need to be allocated and for a short time, the
> +		 * footprint is higher
> +		 */
> +		watermark = low_wmark_pages(zone) + (2UL << order);
> +		if (!zone_watermark_ok(zone, 0, watermark, 0, 0))
> +			continue;
> +
> +		/*
> +		 * fragmentation index determines if allocation failures are
> +		 * due to low memory or external fragmentation
> +		 *
> +		 * index of -1 implies allocations might succeed depending
> +		 * 	on watermarks
> +		 * index < 500 implies alloc failure is due to lack of memory
> +		 *
> +		 * XXX: The choice of 500 is arbitrary. Reinvestigate
> +		 *      appropriately to determine a sensible default.
> +		 *      and what it means when watermarks are also taken
> +		 *      into account. Consider making it a sysctl
> +		 */

Yes, best to make it a sysctl IMO.   It'll make optimisation far easier.
/proc/sys/vm/fragmentation_index_dont_you_dare_use_this_it_will_disappear_soon

> +		fragindex = fragmentation_index(zone, order);
> +		if (fragindex >= 0 && fragindex <= 500)
> +			continue;
> +
> +		if (fragindex == -1 && zone_watermark_ok(zone, order, watermark, 0, 0)) {
> +			rc = COMPACT_PARTIAL;
> +			break;
> +		}
> +
> +		status = compact_zone_order(zone, order, gfp_mask);
> +		rc = max(status, rc);
> +
> +		if (zone_watermark_ok(zone, order, watermark, 0, 0))
> +			break;
> +	}
> +
> +	return rc;
> +}
>
> ...
>
> @@ -1765,6 +1766,31 @@ __alloc_pages_direct_reclaim(gfp_t gfp_mask, unsigned int order,
>  
>  	cond_resched();
>  
> +	/* Try memory compaction for high-order allocations before reclaim */
> +	if (order) {
> +		*did_some_progress = try_to_compact_pages(zonelist,
> +						order, gfp_mask, nodemask);
> +		if (*did_some_progress != COMPACT_INCOMPLETE) {
> +			page = get_page_from_freelist(gfp_mask, nodemask,
> +					order, zonelist, high_zoneidx,
> +					alloc_flags, preferred_zone,
> +					migratetype);
> +			if (page) {
> +				__count_vm_event(COMPACTSUCCESS);
> +				return page;
> +			}
> +
> +			/*
> +			 * It's bad if compaction run occurs and fails.
> +			 * The most likely reason is that pages exist,
> +			 * but not enough to satisfy watermarks.
> +			 */
> +			count_vm_event(COMPACTFAIL);

This counter will get incremented if !__GFP_FS or !__GFP_IO.  Seems
wrong.

> +			cond_resched();
> +		}
> +	}
> +