The quilt patch titled
Subject: llist: avoid extra memory read in llist_add_batch
has been removed from the -mm tree. Its filename was
llist-avoid-extra-memory-read-in-llist_add_batch.patch
This patch was dropped because it was merged into the mm-nonmm-stable branch
of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
------------------------------------------------------
From: Uros Bizjak <ubizjak@xxxxxxxxx>
Subject: llist: avoid extra memory read in llist_add_batch
Date: Mon, 17 Oct 2022 16:52:26 +0200
try_cmpxchg implicitly assigns old head->first value to "first" when
cmpxchg fails. There is no need to re-read the value in the loop.
Link: https://lkml.kernel.org/r/20221017145226.4044-1-ubizjak@xxxxxxxxx
Signed-off-by: Uros Bizjak <ubizjak@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
lib/llist.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
--- a/lib/llist.c~llist-avoid-extra-memory-read-in-llist_add_batch
+++ a/lib/llist.c
@@ -26,10 +26,10 @@
bool llist_add_batch(struct llist_node *new_first, struct llist_node *new_last,
struct llist_head *head)
{
- struct llist_node *first;
+ struct llist_node *first = READ_ONCE(head->first);
do {
- new_last->next = first = READ_ONCE(head->first);
+ new_last->next = first;
} while (!try_cmpxchg(&head->first, &first, new_first));
return !first;
_
Patches currently in -mm which might be from ubizjak@xxxxxxxxx are
