lkml.org 
[lkml]   [2020]   [May]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH 10/10] mm/migrate.c: call detach_page_private to cleanup code
From
Date
Hi Andrew,

On 5/19/20 9:35 AM, Guoqing Jiang wrote:
> On 5/19/20 7:12 AM, Andrew Morton wrote:
>> On Sun, 17 May 2020 23:47:18 +0200 Guoqing Jiang
>> <guoqing.jiang@cloud.ionos.com> wrote:
>>
>>> We can cleanup code a little by call detach_page_private here.
>>>
>>> ...
>>>
>>> --- a/mm/migrate.c
>>> +++ b/mm/migrate.c
>>> @@ -804,10 +804,7 @@ static int __buffer_migrate_page(struct
>>> address_space *mapping,
>>>       if (rc != MIGRATEPAGE_SUCCESS)
>>>           goto unlock_buffers;
>>>   -    ClearPagePrivate(page);
>>> -    set_page_private(newpage, page_private(page));
>>> -    set_page_private(page, 0);
>>> -    put_page(page);
>>> +    set_page_private(newpage, detach_page_private(page));
>>>       get_page(newpage);
>>>         bh = head;
>> mm/migrate.c: In function '__buffer_migrate_page':
>> ./include/linux/mm_types.h:243:52: warning: assignment makes integer
>> from pointer without a cast [-Wint-conversion]
>>   #define set_page_private(page, v) ((page)->private = (v))
>>                                                      ^
>> mm/migrate.c:800:2: note: in expansion of macro 'set_page_private'
>>    set_page_private(newpage, detach_page_private(page));
>>    ^~~~~~~~~~~~~~~~
>>
>> The fact that set_page_private(detach_page_private()) generates a type
>> mismatch warning seems deeply wrong, surely.
>>
>> Please let's get the types sorted out - either unsigned long or void *,
>> not half-one and half-the other.  Whatever needs the least typecasting
>> at callsites, I suggest.
>
> Sorry about that, I should notice the warning before. I will double
> check if other
> places need the typecast or not, then send a new version.

Only this patch missed the typecast. I guess I just need to send an
updated patch
to replace this one (I am fine to send a new patch set if you want),
sorry again for
the trouble.

>
>> And can we please implement set_page_private() and page_private() with
>> inlined C code?  There is no need for these to be macros.
>
> Just did a quick change.
>
> -#define page_private(page)             ((page)->private)
> -#define set_page_private(page, v)      ((page)->private = (v))
> +static inline unsigned long page_private(struct page *page)
> +{
> +       return page->private;
> +}
> +
> +static inline void set_page_private(struct page *page, unsigned long
> priv_data)
> +{
> +       page->private = priv_data;
> +}
>
> Then I get error like.
>
> fs/erofs/zdata.h: In function ‘z_erofs_onlinepage_index’:
> fs/erofs/zdata.h:126:8: error: lvalue required as unary ‘&’ operand
>   u.v = &page_private(page);
>         ^
>
> I guess it is better to keep page_private as macro, please correct me
> in case I
> missed something.

Lost of problems need to be fixed if change page_private to inline
function, so I
think it is better to keep it and only convert set_page_private.

mm/compaction.c: In function ‘isolate_migratepages_block’:
./include/linux/compiler.h:287:20: error: lvalue required as unary ‘&’
operand
   __read_once_size(&(x), __u.__c, sizeof(x));  \
                    ^
./include/linux/compiler.h:293:22: note: in expansion of macro ‘__READ_ONCE’
 #define READ_ONCE(x) __READ_ONCE(x, 1)
                      ^~~~~~~~~~~
mm/internal.h:293:34: note: in expansion of macro ‘READ_ONCE’
 #define page_order_unsafe(page)  READ_ONCE(page_private(page))


Thanks,
Guoqing

\
 
 \ /
  Last update: 2020-05-19 22:45    [W:0.067 / U:1.140 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site