2023-08-26 09:37:13

by Kemeng Shi

[permalink] [raw]
Subject: [PATCH v2 5/7] mm/compaction: remove repeat compact_blockskip_flush check in reset_isolation_suitable

We have compact_blockskip_flush check in __reset_isolation_suitable, just
remove repeat check before __reset_isolation_suitable in
compact_blockskip_flush.

Signed-off-by: Kemeng Shi <[email protected]>
Reviewed-by: Baolin Wang <[email protected]>
---
mm/compaction.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/mm/compaction.c b/mm/compaction.c
index 2937e754cfb7..89a1b627bc89 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -434,9 +434,7 @@ void reset_isolation_suitable(pg_data_t *pgdat)
if (!populated_zone(zone))
continue;

- /* Only flush if a full compaction finished recently */
- if (zone->compact_blockskip_flush)
- __reset_isolation_suitable(zone);
+ __reset_isolation_suitable(zone);
}
}

--
2.30.0



2023-08-29 17:12:48

by Mel Gorman

[permalink] [raw]
Subject: Re: [PATCH v2 5/7] mm/compaction: remove repeat compact_blockskip_flush check in reset_isolation_suitable

On Sat, Aug 26, 2023 at 11:36:15PM +0800, Kemeng Shi wrote:
> We have compact_blockskip_flush check in __reset_isolation_suitable, just
> remove repeat check before __reset_isolation_suitable in
> compact_blockskip_flush.
>
> Signed-off-by: Kemeng Shi <[email protected]>
> Reviewed-by: Baolin Wang <[email protected]>

The comment should move to __reset_isolation_suitable but otherwise

Acked-by: Mel Gorman <[email protected]>

As a complete aside, the reset_isolation_suitable and
__reset_isolation_suitable were badly named because locking is not
involved but it's meaningless churn to fix it.

--
Mel Gorman
SUSE Labs

2023-08-30 19:31:21

by Kemeng Shi

[permalink] [raw]
Subject: Re: [PATCH v2 5/7] mm/compaction: remove repeat compact_blockskip_flush check in reset_isolation_suitable



on 8/29/2023 11:05 PM, Mel Gorman wrote:
> On Sat, Aug 26, 2023 at 11:36:15PM +0800, Kemeng Shi wrote:
>> We have compact_blockskip_flush check in __reset_isolation_suitable, just
>> remove repeat check before __reset_isolation_suitable in
>> compact_blockskip_flush.
>>
>> Signed-off-by: Kemeng Shi <[email protected]>
>> Reviewed-by: Baolin Wang <[email protected]>
>
> The comment should move to __reset_isolation_suitable but otherwise
>
Thanks for the review, I will move comment in next version.
> Acked-by: Mel Gorman <[email protected]>
>
> As a complete aside, the reset_isolation_suitable and
> __reset_isolation_suitable were badly named because locking is not
> involved but it's meaningless churn to fix it.
>