* [merged] mm-memory_hotplug-document-why-shuffle_zone-is-relevant.patch removed from -mm tree
@ 2020-08-10 2:39 akpm
0 siblings, 0 replies; only message in thread
From: akpm @ 2020-08-10 2:39 UTC (permalink / raw)
To: alexander.h.duyck, dan.j.williams, david, mhocko, mm-commits
The patch titled
Subject: mm/memory_hotplug: document why shuffle_zone() is relevant
has been removed from the -mm tree. Its filename was
mm-memory_hotplug-document-why-shuffle_zone-is-relevant.patch
This patch was dropped because it was merged into mainline or a subsystem tree
------------------------------------------------------
From: David Hildenbrand <david@redhat.com>
Subject: mm/memory_hotplug: document why shuffle_zone() is relevant
It's not completely obvious why we have to shuffle the complete zone -
introduced in commit e900a918b098 ("mm: shuffle initial free memory to
improve memory-side-cache utilization") - because some sort of shuffling
is already performed when onlining pages via __free_one_page(), placing
MAX_ORDER-1 pages either to the head or the tail of the freelist. Let's
document why we have to shuffle the complete zone when exposing larger,
contiguous physical memory areas to the buddy.
Link: http://lkml.kernel.org/r/20200624094741.9918-3-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Acked-by: Dan Williams <dan.j.williams@intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
mm/memory_hotplug.c | 8 ++++++++
1 file changed, 8 insertions(+)
--- a/mm/memory_hotplug.c~mm-memory_hotplug-document-why-shuffle_zone-is-relevant
+++ a/mm/memory_hotplug.c
@@ -831,6 +831,14 @@ int __ref online_pages(unsigned long pfn
zone->zone_pgdat->node_present_pages += onlined_pages;
pgdat_resize_unlock(zone->zone_pgdat, &flags);
+ /*
+ * When exposing larger, physically contiguous memory areas to the
+ * buddy, shuffling in the buddy (when freeing onlined pages, putting
+ * them either to the head or the tail of the freelist) is only helpful
+ * for maintaining the shuffle, but not for creating the initial
+ * shuffle. Shuffle the whole zone to make sure the just onlined pages
+ * are properly distributed across the whole freelist.
+ */
shuffle_zone(zone);
node_states_set_node(nid, &arg);
_
Patches currently in -mm which might be from david@redhat.com are
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2020-08-10 2:39 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-08-10 2:39 [merged] mm-memory_hotplug-document-why-shuffle_zone-is-relevant.patch removed from -mm tree akpm
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).