The patch titled
Subject: mm/page_alloc: speed up the iteration of max_order
has been removed from the -mm tree. Its filename was
mm-page_alloc-speeding-up-the-iteration-of-max_order.patch
This patch was dropped because it was nacked
------------------------------------------------------
From: Muchun Song <songmuchun@xxxxxxxxxxxxx>
Subject: mm/page_alloc: speed up the iteration of max_order
When we free a page whose order is very close to MAX_ORDER and greater
than pageblock_order, it wastes some CPU cycles to increase max_order to
MAX_ORDER one by one and check the pageblock migratetype of that page
repeatedly especially when MAX_ORDER is much larger than pageblock_order.
Link: https://lkml.kernel.org/r/20201202121838.75218-1-songmuchun@xxxxxxxxxxxxx
Signed-off-by: Muchun Song <songmuchun@xxxxxxxxxxxxx>
Cc: Vlastimil Babka <vbabka@xxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/page_alloc.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
--- a/mm/page_alloc.c~mm-page_alloc-speeding-up-the-iteration-of-max_order
+++ a/mm/page_alloc.c
@@ -1035,7 +1035,7 @@ continue_merging:
pfn = combined_pfn;
order++;
}
- if (max_order < MAX_ORDER) {
+ if (max_order < MAX_ORDER && order < MAX_ORDER - 1) {
/* If we are here, it means order is >= pageblock_order.
* We want to prevent merge between freepages on isolate
* pageblock and normal pageblock. Without this, pageblock
@@ -1056,6 +1056,8 @@ continue_merging:
is_migrate_isolate(buddy_mt)))
goto done_merging;
}
+ if (unlikely(order != max_order - 1))
+ max_order = order + 1;
max_order++;
goto continue_merging;
}
_
Patches currently in -mm which might be from songmuchun@xxxxxxxxxxxxx are
mm-memcontrol-remove-unused-mod_memcg_obj_state.patch
mm-memcg-slab-fix-return-child-memcg-objcg-for-root-memcg.patch
mm-memcg-slab-fix-use-after-free-in-obj_cgroup_charge.patch
mm-memcg-slab-rename-_lruvec_slab_state-to-_lruvec_kmem_state.patch
mm-page_isolation-do-not-isolate-the-max-order-page.patch
