mm/vmscan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
From: Vern Hao <vernhao@tencent.com>
In skip_cma(), we can use folio_migratetype() to replace get_pageblock_migratetype().
Fixes: 5da226dbfce3 ("mm: skip CMA pages when they are not available")
Signed-off-by: Vern Hao <vernhao@tencent.com>
---
mm/vmscan.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 42e3893de2a7..d9e974b96144 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2271,7 +2271,7 @@ static bool skip_cma(struct folio *folio, struct scan_control *sc)
{
return !current_is_kswapd() &&
gfp_migratetype(sc->gfp_mask) != MIGRATE_MOVABLE &&
- get_pageblock_migratetype(&folio->page) == MIGRATE_CMA;
+ folio_migratetype(folio) == MIGRATE_CMA;
}
#else
static bool skip_cma(struct folio *folio, struct scan_control *sc)
--
2.31.1
On 25.08.23 06:08, Vern Hao wrote:
> From: Vern Hao <vernhao@tencent.com>
>
> In skip_cma(), we can use folio_migratetype() to replace get_pageblock_migratetype().
>
> Fixes: 5da226dbfce3 ("mm: skip CMA pages when they are not available")
Why did you think "Fixes" was a good idea?
--
Cheers,
David / dhildenb
在 2023/8/25 15:12, David Hildenbrand 写道:
> On 25.08.23 06:08, Vern Hao wrote:
>> From: Vern Hao <vernhao@tencent.com>
>>
>> In skip_cma(), we can use folio_migratetype() to replace
>> get_pageblock_migratetype().
>>
>> Fixes: 5da226dbfce3 ("mm: skip CMA pages when they are not available")
>
> Why did you think "Fixes" was a good idea?
>
The skip_cma() contains the folio argument, and the folio_migratetype()
is already implemented in the kernel, so I think it's fine to keep it
consistent.
Thanks!
On 25.08.23 09:31, Vern Hao wrote:
>
> 在 2023/8/25 15:12, David Hildenbrand 写道:
>> On 25.08.23 06:08, Vern Hao wrote:
>>> From: Vern Hao <vernhao@tencent.com>
>>>
>>> In skip_cma(), we can use folio_migratetype() to replace
>>> get_pageblock_migratetype().
>>>
>>> Fixes: 5da226dbfce3 ("mm: skip CMA pages when they are not available")
>>
>> Why did you think "Fixes" was a good idea?
>>
> The skip_cma() contains the folio argument, and the folio_migratetype()
> is already implemented in the kernel, so I think it's fine to keep it
> consistent.
If this doesn't actually *fix a bug*, then please don't add fixes and
confuse people looking for actual bugfixes.
See Documentation/process/submitting-patches.rst
--
Cheers,
David / dhildenb
在 2023/8/25 15:34, David Hildenbrand 写道:
> On 25.08.23 09:31, Vern Hao wrote:
>>
>> 在 2023/8/25 15:12, David Hildenbrand 写道:
>>> On 25.08.23 06:08, Vern Hao wrote:
>>>> From: Vern Hao <vernhao@tencent.com>
>>>>
>>>> In skip_cma(), we can use folio_migratetype() to replace
>>>> get_pageblock_migratetype().
>>>>
>>>> Fixes: 5da226dbfce3 ("mm: skip CMA pages when they are not available")
>>>
>>> Why did you think "Fixes" was a good idea?
>>>
>> The skip_cma() contains the folio argument, and the folio_migratetype()
>> is already implemented in the kernel, so I think it's fine to keep it
>> consistent.
>
> If this doesn't actually *fix a bug*, then please don't add fixes and
> confuse people looking for actual bugfixes.
>
> See Documentation/process/submitting-patches.rst
OK, my bad, i will remove it, thanks.
On 25.08.23 09:36, Vern Hao wrote:
>
> 在 2023/8/25 15:34, David Hildenbrand 写道:
>> On 25.08.23 09:31, Vern Hao wrote:
>>>
>>> 在 2023/8/25 15:12, David Hildenbrand 写道:
>>>> On 25.08.23 06:08, Vern Hao wrote:
>>>>> From: Vern Hao <vernhao@tencent.com>
>>>>>
>>>>> In skip_cma(), we can use folio_migratetype() to replace
>>>>> get_pageblock_migratetype().
>>>>>
>>>>> Fixes: 5da226dbfce3 ("mm: skip CMA pages when they are not available")
>>>>
>>>> Why did you think "Fixes" was a good idea?
>>>>
>>> The skip_cma() contains the folio argument, and the folio_migratetype()
>>> is already implemented in the kernel, so I think it's fine to keep it
>>> consistent.
>>
>> If this doesn't actually *fix a bug*, then please don't add fixes and
>> confuse people looking for actual bugfixes.
>>
>> See Documentation/process/submitting-patches.rst
> OK, my bad, i will remove it, thanks.
>
No need to resend. Andrew can drop that tag when applying.
--
Cheers,
David / dhildenb
© 2016 - 2026 Red Hat, Inc.