Re: [PATCH v2 03/10] mm/memory_hotplug: simplify page offlining

[Andrew Morton]

On Fri, 4 Sep 2020 07:47:45 +0200 David Hildenbrand <david@xxxxxxxxxx> wrote:

> 
> 
> > +         * PageBuddy on freed pages on other zones.
> > +         */
> > +        if (ret)
> > +            drain_all_pages(zone);
> > +    } while (test_pages_isolated(start_pfn, end_pfn, MEMORY_OFFLINE));
> 
> I think we have to do
> 
> ret = test_pages_isolated()
> if (ret)
> ...
> } while (ret);
> 
> So keeping the old code flow. I cannot resend before next Tuesday.

Here's what I now (effectively) have:


From: David Hildenbrand <david@xxxxxxxxxx>
Subject: mm/memory_hotplug: simplify page offlining

We make sure that we cannot have any memory holes right at the beginning
of offline_pages().  We no longer need walk_system_ram_range() and can
call test_pages_isolated() and __offline_isolated_pages() directly.

offlined_pages always corresponds to nr_pages, so we can simplify that.

Link: https://lkml.kernel.org/r/20200819175957.28465-4-david@xxxxxxxxxx
Signed-off-by: David Hildenbrand <david@xxxxxxxxxx>
Acked-by: Michal Hocko <mhocko@xxxxxxxx>
Reviewed-by: Oscar Salvador <osalvador@xxxxxxx>
Cc: Wei Yang <richard.weiyang@xxxxxxxxxxxxxxxxx>
Cc: Baoquan He <bhe@xxxxxxxxxx>
Cc: Pankaj Gupta <pankaj.gupta.linux@xxxxxxxxx>
Cc: Charan Teja Reddy <charante@xxxxxxxxxxxxxx>
Cc: Dan Williams <dan.j.williams@xxxxxxxxx>
Cc: Fenghua Yu <fenghua.yu@xxxxxxxxx>
Cc: Logan Gunthorpe <logang@xxxxxxxxxxxx>
Cc: "Matthew Wilcox (Oracle)" <willy@xxxxxxxxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx>
Cc: Michel Lespinasse <walken@xxxxxxxxxx>
Cc: Mike Rapoport <rppt@xxxxxxxxxx>
Cc: Tony Luck <tony.luck@xxxxxxxxx>

--- a/mm/memory_hotplug.c~mm-memory_hotplug-simplify-page-offlining
+++ a/mm/memory_hotplug.c
@@ -1383,28 +1383,6 @@ do_migrate_range(unsigned long start_pfn
 	return ret;
 }
 
-/* Mark all sections offline and remove all free pages from the buddy. */
-static int
-offline_isolated_pages_cb(unsigned long start, unsigned long nr_pages,
-			void *data)
-{
-	unsigned long *offlined_pages = (unsigned long *)data;
-
-	*offlined_pages += __offline_isolated_pages(start, start + nr_pages);
-	return 0;
-}
-
-/*
- * Check all pages in range, recorded as memory resource, are isolated.
- */
-static int
-check_pages_isolated_cb(unsigned long start_pfn, unsigned long nr_pages,
-			void *data)
-{
-	return test_pages_isolated(start_pfn, start_pfn + nr_pages,
-				   MEMORY_OFFLINE);
-}
-
 static int __init cmdline_parse_movable_node(char *p)
 {
 	movable_node_enabled = true;
@@ -1491,7 +1469,7 @@ static int count_system_ram_pages_cb(uns
 int __ref offline_pages(unsigned long start_pfn, unsigned long nr_pages)
 {
 	const unsigned long end_pfn = start_pfn + nr_pages;
-	unsigned long pfn, system_ram_pages = 0, offlined_pages = 0;
+	unsigned long pfn, system_ram_pages = 0;
 	int ret, node, nr_isolate_pageblock;
 	unsigned long flags;
 	struct zone *zone;
@@ -1589,9 +1567,7 @@ int __ref offline_pages(unsigned long st
 			reason = "failure to dissolve huge pages";
 			goto failed_removal_isolated;
 		}
-		/* check again */
-		ret = walk_system_ram_range(start_pfn, end_pfn - start_pfn,
-					    NULL, check_pages_isolated_cb);
+
 		/*
 		 * per-cpu pages are drained in start_isolate_page_range, but if
 		 * there are still pages that are not free, make sure that we
@@ -1604,15 +1580,15 @@ int __ref offline_pages(unsigned long st
 		 * because has_unmovable_pages explicitly checks for
 		 * PageBuddy on freed pages on other zones.
 		 */
+		ret = test_pages_isolated(start_pfn, end_pfn, MEMORY_OFFLINE);
 		if (ret)
 			drain_all_pages(zone);
 	} while (ret);
 
-	/* Ok, all of our target is isolated.
-	   We cannot do rollback at this point. */
-	walk_system_ram_range(start_pfn, end_pfn - start_pfn,
-			      &offlined_pages, offline_isolated_pages_cb);
-	pr_info("Offlined Pages %ld\n", offlined_pages);
+	/* Mark all sections offline and remove free pages from the buddy. */
+	__offline_isolated_pages(start_pfn, end_pfn);
+	pr_info("Offlined Pages %ld\n", nr_pages);
+
 	/*
 	 * Onlining will reset pagetype flags and makes migrate type
 	 * MOVABLE, so just need to decrease the number of isolated
@@ -1623,11 +1599,11 @@ int __ref offline_pages(unsigned long st
 	spin_unlock_irqrestore(&zone->lock, flags);
 
 	/* removal success */
-	adjust_managed_page_count(pfn_to_page(start_pfn), -offlined_pages);
-	zone->present_pages -= offlined_pages;
+	adjust_managed_page_count(pfn_to_page(start_pfn), -nr_pages);
+	zone->present_pages -= nr_pages;
 
 	pgdat_resize_lock(zone->zone_pgdat, &flags);
-	zone->zone_pgdat->node_present_pages -= offlined_pages;
+	zone->zone_pgdat->node_present_pages -= nr_pages;
 	pgdat_resize_unlock(zone->zone_pgdat, &flags);
 
 	init_per_zone_wmark_min();
_