diff mbox series

mm/gup: Add folio to list when folio_isolate_lru() succeed

Message ID 20230131063206.28820-1-Kuan-Ying.Lee@mediatek.com (mailing list archive)
State New
Headers show
Series mm/gup: Add folio to list when folio_isolate_lru() succeed | expand

Commit Message

Kuan-Ying Lee (李冠穎) Jan. 31, 2023, 6:32 a.m. UTC
If we call folio_isolate_lru() successfully, we will get
return value 0. We need to add this folio to the
movable_pages_list.

Fixes: 67e139b02d99 ("mm/gup.c: refactor check_and_migrate_movable_pages()")
Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
---
 mm/gup.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Andrew Morton Jan. 31, 2023, 11:17 p.m. UTC | #1
On Tue, 31 Jan 2023 14:32:06 +0800 Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com> wrote:

> If we call folio_isolate_lru() successfully, we will get
> return value 0. We need to add this folio to the
> movable_pages_list.
> 
> Fixes: 67e139b02d99 ("mm/gup.c: refactor check_and_migrate_movable_pages()")
> Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
>
> ...
>
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -1914,7 +1914,7 @@ static unsigned long collect_longterm_unpinnable_pages(
>  			drain_allow = false;
>  		}
>  
> -		if (!folio_isolate_lru(folio))
> +		if (folio_isolate_lru(folio))
>  			continue;
>  
>  		list_add_tail(&folio->lru, movable_page_list);

Thanks.  What are the user-visible effects of this bug?
Baolin Wang Feb. 1, 2023, 9:08 a.m. UTC | #2
On 1/31/2023 2:32 PM, Kuan-Ying Lee wrote:
> If we call folio_isolate_lru() successfully, we will get
> return value 0. We need to add this folio to the
> movable_pages_list.
> 
> Fixes: 67e139b02d99 ("mm/gup.c: refactor check_and_migrate_movable_pages()")
> Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>

Good catch.
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>

> ---
>   mm/gup.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/mm/gup.c b/mm/gup.c
> index f3d2cccb89f8..918c364d01ac 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -1914,7 +1914,7 @@ static unsigned long collect_longterm_unpinnable_pages(
>   			drain_allow = false;
>   		}
>   
> -		if (!folio_isolate_lru(folio))
> +		if (folio_isolate_lru(folio))
>   			continue;
>   
>   		list_add_tail(&folio->lru, movable_page_list);
Kuan-Ying Lee (李冠穎) Feb. 2, 2023, 2:13 a.m. UTC | #3
On Tue, 2023-01-31 at 15:17 -0800, Andrew Morton wrote:
> On Tue, 31 Jan 2023 14:32:06 +0800 Kuan-Ying Lee <
> Kuan-Ying.Lee@mediatek.com> wrote:
> 
> > If we call folio_isolate_lru() successfully, we will get
> > return value 0. We need to add this folio to the
> > movable_pages_list.
> > 
> > Fixes: 67e139b02d99 ("mm/gup.c: refactor
> > check_and_migrate_movable_pages()")
> > Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
> > 
> > ...
> > 
> > --- a/mm/gup.c
> > +++ b/mm/gup.c
> > @@ -1914,7 +1914,7 @@ static unsigned long
> > collect_longterm_unpinnable_pages(
> >  			drain_allow = false;
> >  		}
> >  
> > -		if (!folio_isolate_lru(folio))
> > +		if (folio_isolate_lru(folio))
> >  			continue;
> >  
> >  		list_add_tail(&folio->lru, movable_page_list);
> 
> Thanks.  What are the user-visible effects of this bug?

Hi Andrew,

I didn't hit bug on real devices. I observe this issue
by tracing get_user_pages() call flow.

Thanks.
Alistair Popple Feb. 2, 2023, 6:09 a.m. UTC | #4
Andrew Morton <akpm@linux-foundation.org> writes:

> On Tue, 31 Jan 2023 14:32:06 +0800 Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com> wrote:
>
>> If we call folio_isolate_lru() successfully, we will get
>> return value 0. We need to add this folio to the
>> movable_pages_list.

Ugh, thanks for catching this:

Reviewed-by: Alistair Popple <apopple@nvidia.com>

>> Fixes: 67e139b02d99 ("mm/gup.c: refactor check_and_migrate_movable_pages()")
>> Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
>>
>> ...
>>
>> --- a/mm/gup.c
>> +++ b/mm/gup.c
>> @@ -1914,7 +1914,7 @@ static unsigned long collect_longterm_unpinnable_pages(
>>  			drain_allow = false;
>>  		}
>>  
>> -		if (!folio_isolate_lru(folio))
>> +		if (folio_isolate_lru(folio))
>>  			continue;
>>  
>>  		list_add_tail(&folio->lru, movable_page_list);
>
> Thanks.  What are the user-visible effects of this bug?

In the common case none other than an extra loop through
collect_longterm_unpinnable_pages():

1. First call to collect_longterm_unpinnable_pages() will increment
   collected and isolate the page but not add it to movable_page_list.

2. migrate_longterm_unpinnable_pages() will return -EAGAIN and unpin all
   the pages but they will remain LRU isolated.

3. The next spin around __gup_longterm_locked() will re-pin the pages
   and re-call collect_longterm_unpinnable_pages(). As the page has
   already been isolated folio_isolate_lru() will return -EBUSY which
   will add the page to movable_page_list and complete processing as
   intended.

However this assumes the page table still points to the same page when
__get_user_pages_locked() is called the second time. That may not be the
case in which case we would leave the page LRU isolated forever
effectively leaving an unmovable page in a movable zone which is what we
were trying to avoid in the first place.

So I think Cc: stable is warranted.

 - Alistair
David Hildenbrand Feb. 2, 2023, 8:39 a.m. UTC | #5
On 31.01.23 07:32, Kuan-Ying Lee wrote:
> If we call folio_isolate_lru() successfully, we will get
> return value 0. We need to add this folio to the
> movable_pages_list.
> 
> Fixes: 67e139b02d99 ("mm/gup.c: refactor check_and_migrate_movable_pages()")
> Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
> ---
>   mm/gup.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/mm/gup.c b/mm/gup.c
> index f3d2cccb89f8..918c364d01ac 100644
> --- a/mm/gup.c
> +++ b/mm/gup.c
> @@ -1914,7 +1914,7 @@ static unsigned long collect_longterm_unpinnable_pages(
>   			drain_allow = false;
>   		}
>   
> -		if (!folio_isolate_lru(folio))
> +		if (folio_isolate_lru(folio))
>   			continue;
>   
>   		list_add_tail(&folio->lru, movable_page_list);

Agreed that this deserves cc:stable

Acked-by: David Hildenbrand <david@redhat.com>
diff mbox series

Patch

diff --git a/mm/gup.c b/mm/gup.c
index f3d2cccb89f8..918c364d01ac 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -1914,7 +1914,7 @@  static unsigned long collect_longterm_unpinnable_pages(
 			drain_allow = false;
 		}
 
-		if (!folio_isolate_lru(folio))
+		if (folio_isolate_lru(folio))
 			continue;
 
 		list_add_tail(&folio->lru, movable_page_list);