[PATCH] mm/page_counter: remove an incorrect call to propagate_protected_usage()

[PATCH] mm/page_counter: remove an incorrect call to propagate_protected_usage()

[Donghai Qiao]

The function propagate_protected_usage() is called to propagate the usage
change in the page_counter structure. But there is call to this function
from page_counter_try_charge() when there is actually no usage change.
Hence this call should be removed.

Signed-off-by: Donghai Qiao <dqiao@redhat.com>
---
 mm/page_counter.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/mm/page_counter.c b/mm/page_counter.c
index 7d83641eb86b..eb156ff5d603 100644
--- a/mm/page_counter.c
+++ b/mm/page_counter.c
@@ -120,7 +120,6 @@ bool page_counter_try_charge(struct page_counter *counter,
 		new = atomic_long_add_return(nr_pages, &c->usage);
 		if (new > c->max) {
 			atomic_long_sub(nr_pages, &c->usage);
-			propagate_protected_usage(c, new);
 			/*
 			 * This is racy, but we can live with some
 			 * inaccuracy in the failcnt which is only used
-- 
2.27.0

Re: [PATCH] mm/page_counter: remove an incorrect call to propagate_protected_usage()

[Roman Gushchin]

On Thu, Nov 18, 2021 at 01:11:25PM -0500, Donghai Qiao wrote:
> The function propagate_protected_usage() is called to propagate the usage
> change in the page_counter structure. But there is call to this function
> from page_counter_try_charge() when there is actually no usage change.
> Hence this call should be removed.
> 
> Signed-off-by: Donghai Qiao <dqiao@redhat.com>

Reviewed-by: Roman Gushchin <guro@fb.com>

Thanks, Donghai!