Re: [PATCH v4 2/5] mm/cma: populate ZONE_CMA

[Joonsoo Kim]

On Fri, Aug 19, 2016 at 01:20:13PM +0200, Vlastimil Babka wrote:
> On 08/09/2016 08:39 AM, js1304@xxxxxxxxx wrote:
> >From: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
> >
> >Until now, reserved pages for CMA are managed in the ordinary zones
> >where page's pfn are belong to. This approach has numorous problems
> >and fixing them isn't easy. (It is mentioned on previous patch.)
> >To fix this situation, ZONE_CMA is introduced in previous patch, but,
> >not yet populated. This patch implement population of ZONE_CMA
> >by stealing reserved pages from the ordinary zones.
> >
> >Unlike previous implementation that kernel allocation request with
> >__GFP_MOVABLE could be serviced from CMA region, allocation request only
> >with GFP_HIGHUSER_MOVABLE can be serviced from CMA region in the new
> >approach. This is an inevitable design decision to use the zone
> >implementation because ZONE_CMA could contain highmem. Due to this
> >decision, ZONE_CMA will work like as ZONE_HIGHMEM or ZONE_MOVABLE.
> >
> >I don't think it would be a problem because most of file cache pages
> >and anonymous pages are requested with GFP_HIGHUSER_MOVABLE. It could
> >be proved by the fact that there are many systems with ZONE_HIGHMEM and
> >they work fine. Notable disadvantage is that we cannot use these pages
> >for blockdev file cache page, because it usually has __GFP_MOVABLE but
> >not __GFP_HIGHMEM and __GFP_USER. But, in this case, there is pros and
> >cons. In my experience, blockdev file cache pages are one of the top
> >reason that causes cma_alloc() to fail temporarily. So, we can get more
> >guarantee of cma_alloc() success by discarding that case.
> >
> >Implementation itself is very easy to understand. Steal when cma area is
> >initialized and recalculate various per zone stat/threshold.
> >
> >Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@xxxxxxx>
> 
> [...]
> 
> >@@ -145,6 +145,28 @@ err:
> > static int __init cma_init_reserved_areas(void)
> > {
> > 	int i;
> >+	struct zone *zone;
> >+	unsigned long start_pfn = UINT_MAX, end_pfn = 0;
> >+
> >+	if (!cma_area_count)
> >+		return 0;
> >+
> >+	for (i = 0; i < cma_area_count; i++) {
> >+		if (start_pfn > cma_areas[i].base_pfn)
> >+			start_pfn = cma_areas[i].base_pfn;
> >+		if (end_pfn < cma_areas[i].base_pfn + cma_areas[i].count)
> >+			end_pfn = cma_areas[i].base_pfn + cma_areas[i].count;
> >+	}
> >+
> >+	for_each_populated_zone(zone) {
> >+		if (!is_zone_cma(zone))
> >+			continue;
> >+
> >+		/* ZONE_CMA doesn't need to exceed CMA region */
> >+		zone->zone_start_pfn = max(zone->zone_start_pfn, start_pfn);
> >+		zone->spanned_pages = min(zone_end_pfn(zone), end_pfn) -
> >+					zone->zone_start_pfn;
> 
> Hmm is this a dead code? for_each_populated_zone() will skip zones
> where zone->present_pages is 0, which is AFAICS the result for
> ZONE_CMA
> after it's initialized by calculate_node_totalpages() (after Patch 1/5).
> The present_pages seem to be only increased later in this function
> by cma_activate_area() -> init_cma_reserved_pageblock().

You're right. I will replace for_each_populated_zone() with for_each_zone().

> 
> >+	}
> >
> > 	for (i = 0; i < cma_area_count; i++) {
> > 		int ret = cma_activate_area(&cma_areas[i]);
> >@@ -153,6 +175,24 @@ static int __init cma_init_reserved_areas(void)
> > 			return ret;
> > 	}
> >
> 
> [...]
> 
> >diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> >index f6c4358..352096e 100644
> >--- a/mm/page_alloc.c
> >+++ b/mm/page_alloc.c
> >@@ -1600,16 +1600,38 @@ void __init page_alloc_init_late(void)
> > }
> >
> > #ifdef CONFIG_CMA
> >+static void __init adjust_present_page_count(struct page *page, long count)
> >+{
> >+	struct zone *zone = page_zone(page);
> >+
> >+	/* We don't need to hold a lock since it is boot-up process */
> >+	zone->present_pages += count;
> >+}
> >+
> > /* Free whole pageblock and set its migration type to MIGRATE_CMA. */
> > void __init init_cma_reserved_pageblock(struct page *page)
> > {
> > 	unsigned i = pageblock_nr_pages;
> >+	unsigned long pfn = page_to_pfn(page);
> > 	struct page *p = page;
> >+	int nid = page_to_nid(page);
> >+
> >+	/*
> >+	 * ZONE_CMA will steal present pages from other zones by changing
> >+	 * page links so page_zone() is changed. Before that,
> >+	 * we need to adjust previous zone's page count first.
> >+	 */
> >+	adjust_present_page_count(page, -pageblock_nr_pages);
> 
> So in previous version I said this (and you replied):
> 
> >>> Ideally, zone's start_pfn and spanned_pages should be also adjusted
> >>> if we stole from the beginning/end (which I suppose should be quite
> >>> common?).
> >
> >It would be possible. Maybe, there is a reason I didn't do that but I
> >don't remember it. I will think more.
> 
> What's the outcome? :) Is stealing from beginning/end of zone common
> for CMA? Are we losing zone->contiguous and add iterations to
> compaction scanner needlessly?

Yes. I have thought it and my conclusion is:

It is possible but I don't want to make code more complex at this
moment. We can simply do it later.

Thanks.