2021-03-23 09:31:35

by Xu, Yanfei

Subject: [PATCH] khugepaged: Raplace barrier() with READ_ONCE() for a selective variable

From: Yanfei Xu <[email protected]>

READ_ONCE() is more selective and lightweight. It is more appropriate that
using a READ_ONCE() for the certain variable to prevent the compiler from
reordering.

Signed-off-by: Yanfei Xu <[email protected]>
---
mm/khugepaged.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index b81521dfbb1a..034d05a26cb0 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -2202,11 +2202,9 @@ static void khugepaged_do_scan(void)
{
struct page *hpage = NULL;
unsigned int progress = 0, pass_through_head = 0;
- unsigned int pages = khugepaged_pages_to_scan;
+ unsigned int pages = READ_ONCE(khugepaged_pages_to_scan);
bool wait = true;

- barrier(); /* write khugepaged_pages_to_scan to local stack */
-
lru_add_drain_all();

while (progress < pages) {
--
2.27.0

2021-03-23 10:25:20

by Kirill A. Shutemov

Subject: Re: [PATCH] khugepaged: Raplace barrier() with READ_ONCE() for a selective variable

On Tue, Mar 23, 2021 at 05:27:30PM +0800, [email protected] wrote:
> From: Yanfei Xu <[email protected]>
>
> READ_ONCE() is more selective and lightweight. It is more appropriate that
> using a READ_ONCE() for the certain variable to prevent the compiler from
> reordering.
>
> Signed-off-by: Yanfei Xu <[email protected]>

Acked-by: Kirill A. Shutemov <[email protected]>

--
Kirill A. Shutemov