[to-be-updated] maple_tree-micro-optimisation-of-mas_wr_store_type.patch removed from -mm tree
Andrew Morton <[email protected]>
| Newsgroups | org.kernel.vger.mm-commits |
|---|---|
| Message-ID | <[email protected]> |
The quilt patch titled
Subject: maple_tree: micro optimisation of mas_wr_store_type()
has been removed from the -mm tree. Its filename was
maple_tree-micro-optimisation-of-mas_wr_store_type.patch
This patch was dropped because an updated version will be issued
------------------------------------------------------
From: "Liam R. Howlett (Oracle)" <[email protected]>
Subject: maple_tree: micro optimisation of mas_wr_store_type()
Date: Tue, 30 Jun 2026 15:08:34 -0400
Use three new local booleans instead of reading other structures. This
has shown an increase of 0.62% on a 2048 entry tree of height 4.
Link: https://lore.kernel.org/[email protected]
Signed-off-by: Liam R. Howlett (Oracle) <[email protected]>
Cc: Boqun Feng <[email protected]>
Cc: Chris Mason <[email protected]>
Cc: Chuck Lever <[email protected]>
Cc: Ingo Molnar <[email protected]>
Cc: Jason Gunthorpe <[email protected]>
Cc: Joe Perches <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Waiman Long <[email protected]>
Cc: Will Deacon <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
---
lib/maple_tree.c | 18 +++++++++++++++---
1 file changed, 15 insertions(+), 3 deletions(-)
--- a/lib/maple_tree.c~maple_tree-micro-optimisation-of-mas_wr_store_type
+++ a/lib/maple_tree.c
@@ -3737,6 +3737,9 @@ static inline enum store_type mas_wr_sto
{
struct ma_state *mas = wr_mas->mas;
unsigned char new_end;
+ bool appending;
+ bool one_slot;
+ bool in_rcu;
if (unlikely(mas_is_none(mas) || mas_is_ptr(mas)))
return wr_store_root;
@@ -3756,21 +3759,30 @@ static inline enum store_type mas_wr_sto
return wr_new_root;
new_end = mas_wr_new_end(wr_mas);
+ in_rcu = mt_in_rcu(mas->tree);
+ appending = mas->offset == mas->end;
+ one_slot = wr_mas->offset_end - mas->offset == 1;
+
/* Potential spanning rebalance collapsing a node */
if (new_end < mt_min_slots[wr_mas->type]) {
if (!mte_is_root(mas->node))
return wr_rebalance;
+ if (!in_rcu) {
+ if (appending)
+ return wr_append;
+ else if (mas->end == new_end && one_slot)
+ return wr_slot_store;
+ }
return wr_node_store;
}
if (new_end >= mt_slots[wr_mas->type])
return wr_split_store;
- if (!mt_in_rcu(mas->tree) && (mas->offset == mas->end))
+ if (!in_rcu && appending)
return wr_append;
- if ((new_end == mas->end) && (!mt_in_rcu(mas->tree) ||
- (wr_mas->offset_end - mas->offset == 1)))
+ if (new_end == mas->end && (!in_rcu || one_slot))
return wr_slot_store;
return wr_node_store;
_
Patches currently in -mm which might be from [email protected] are
maple_tree-add-bulk-parent-set-helper.patch
maple_tree-catch-race-in-mas_alloc_cyclic.patch
maple_tree-document-that-erase-may-use-gfp_kernel-for-allocations.patch
maple_tree-warn_on_once-when-allocations-fail.patch
maple_tree-document-erase-and-allocations-better.patch
maple_tree-change-two-gfp-flags-in-tests.patch
maple_tree-fix-argument-name-in-header.patch
maple_tree-avoid-extra-gap-calculation.patch
maple_tree-add-helper-mas_make_walkable.patch