The patch titled
Subject: mm/page_alloc: do not calculate node's total pages and memmap pages when empty
has been added to the -mm tree. Its filename is
mm-page_alloc-do-not-calculate-nodes-total-pages-and-memmap-pages-when-empty.patch
This patch should soon appear at
https://ozlabs.org/~akpm/mmots/broken-out/mm-page_alloc-do-not-calculate-nodes-total-pages-and-memmap-pages-when-empty.patch
and later at
https://ozlabs.org/~akpm/mmotm/broken-out/mm-page_alloc-do-not-calculate-nodes-total-pages-and-memmap-pages-when-empty.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Oscar Salvador <osalvador@xxxxxxx>
Subject: mm/page_alloc: do not calculate node's total pages and memmap pages when empty
Patch series "A minor hotplug refactoring".
This patch (of 3):
free_area_init_node() calls calculate_node_totalpages() and
free_area_init_core(). The former to get node's {spanned,present}_pages,
and the latter to calculate, among other things, how many pages per zone
we spent on memmap_pages, which is used to subtract zone's free pages.
On memoryless nodes, it is pointless to perform such a bunch of work, so
make sure we skip the calculations when having a node or empty zone.
Link: https://lkml.kernel.org/r/20220307150725.6810-1-osalvador@xxxxxxx
Link: https://lkml.kernel.org/r/20220307150725.6810-2-osalvador@xxxxxxx
Signed-off-by: Oscar Salvador <osalvador@xxxxxxx>
Cc: David Hildenbrand <david@xxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxxxxx>
Cc: Wei Yang <richard.weiyang@xxxxxxxxx>
Cc: Miaohe Lin <linmiaohe@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/page_alloc.c | 10 ++++++++++
1 file changed, 10 insertions(+)
--- a/mm/page_alloc.c~mm-page_alloc-do-not-calculate-nodes-total-pages-and-memmap-pages-when-empty
+++ a/mm/page_alloc.c
@@ -7321,6 +7321,10 @@ static void __init calculate_node_totalp
unsigned long realtotalpages = 0, totalpages = 0;
enum zone_type i;
+ /* Skip calculation for memoryless nodes */
+ if (node_start_pfn == node_end_pfn)
+ goto no_pages;
+
for (i = 0; i < MAX_NR_ZONES; i++) {
struct zone *zone = pgdat->node_zones + i;
unsigned long zone_start_pfn, zone_end_pfn;
@@ -7353,6 +7357,7 @@ static void __init calculate_node_totalp
realtotalpages += real_size;
}
+no_pages:
pgdat->node_spanned_pages = totalpages;
pgdat->node_present_pages = realtotalpages;
pr_debug("On node %d totalpages: %lu\n", pgdat->node_id, realtotalpages);
@@ -7570,6 +7575,10 @@ static void __init free_area_init_core(s
size = zone->spanned_pages;
freesize = zone->present_pages;
+ /* No pages? Nothing to calculate then. */
+ if (!size)
+ goto no_pages;
+
/*
* Adjust freesize so that it accounts for how much memory
* is used by this zone for memmap. This affects the watermark
@@ -7605,6 +7614,7 @@ static void __init free_area_init_core(s
* when the bootmem allocator frees pages into the buddy system.
* And all highmem pages will be managed by the buddy system.
*/
+no_pages:
zone_init_internals(zone, j, nid, freesize);
if (!size)
_
Patches currently in -mm which might be from osalvador@xxxxxxx are
mm-untangle-config-dependencies-for-demote-on-reclaim.patch
mm-page_alloc-do-not-calculate-nodes-total-pages-and-memmap-pages-when-empty.patch
mm-memory_hotplug-reset-nodes-state-when-empty-during-offline.patch
mm-memory_hotplug-refactor-hotadd_init_pgdat-and-try_online_node.patch
