Re: [PATCH 2/6] mm/page_alloc.c: simplify the code by using macro K()

[Miaohe Lin]

On 2021/9/1 15:46, Oleksandr Natalenko wrote:
> Hello.
> 
> On středa 1. září 2021 9:37:49 CEST Miaohe Lin wrote:
>> On 2021/8/31 22:19, Oleksandr Natalenko wrote:
>>> On úterý 31. srpna 2021 13:08:42 CEST Miaohe Lin wrote:
>>>> On 2021/8/31 16:54, Oleksandr Natalenko wrote:
>>>>> On pondělí 30. srpna 2021 16:10:47 CEST Miaohe Lin wrote:
>>>>>> Use helper macro K() to convert the pages to the corresponding size.
>>>>>> Minor readability improvement.
>>>>>>
>>>>>> Signed-off-by: Miaohe Lin <linmiaohe@xxxxxxxxxx>
>>>>>> ---
>>>>>>
>>>>>>  mm/page_alloc.c | 12 +++++-------
>>>>>>  1 file changed, 5 insertions(+), 7 deletions(-)
>>>>>>
>>>>>> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
>>>>>> index dbb3338d9287..d3983244f56f 100644
>>>>>> --- a/mm/page_alloc.c
>>>>>> +++ b/mm/page_alloc.c
>>>>>> @@ -8134,8 +8134,7 @@ unsigned long free_reserved_area(void *start,
>>>>>> void
>>>>>> *end, int poison, const char }
>>>>>>
>>>>>>  	if (pages && s)
>>>>>>
>>>>>> -		pr_info("Freeing %s memory: %ldK\n",
>>>>>> -			s, pages << (PAGE_SHIFT - 10));
>>>>>> +		pr_info("Freeing %s memory: %ldK\n", s, K(pages));
>>>>>>
>>>>>>  	return pages;
>>>>>>  
>>>>>>  }
>>>>>>
>>>>>> @@ -8180,14 +8179,13 @@ void __init mem_init_print_info(void)
>>>>>>
>>>>>>  		", %luK highmem"
>>>>>>  
>>>>>>  #endif
>>>>>>  
>>>>>>  		")\n",
>>>>>>
>>>>>> -		nr_free_pages() << (PAGE_SHIFT - 10),
>>>>>> -		physpages << (PAGE_SHIFT - 10),
>>>>>> +		K(nr_free_pages()), K(physpages),
>>>>>>
>>>>>>  		codesize >> 10, datasize >> 10, rosize >> 10,
>>>>>>  		(init_data_size + init_code_size) >> 10, bss_size >> 10,
>>>>>>
>>>>>> -		(physpages - totalram_pages() - totalcma_pages) << 
> (PAGE_SHIFT
>>>>>
>>>>> - 10),
>>>>>
>>>>>> -		totalcma_pages << (PAGE_SHIFT - 10)
>>>>>> +		K(physpages - totalram_pages() - totalcma_pages),
>>>>>> +		K(totalcma_pages)
>>>>>>
>>>>>>  #ifdef	CONFIG_HIGHMEM
>>>>>>
>>>>>> -		, totalhigh_pages() << (PAGE_SHIFT - 10)
>>>>>> +		, K(totalhigh_pages())
>>>>>>
>>>>>>  #endif
>>>>>>  
>>>>>>  		);
>>>>>>  
>>>>>>  }
>>>>>
>>>>> (my concern is not quite within the scope of this submission, but I'll
>>>>> ask
>>>>> anyway)
>>>>>
>>>>> Given we have this:
>>>>>
>>>>> ```
>>>>> git grep '#define K(x)' v5.14
>>>>> v5.14:drivers/base/node.c:#define K(x) ((x) << (PAGE_SHIFT - 10))
>>>>> v5.14:drivers/net/hamradio/scc.c:#define K(x) kiss->x
>>>>> v5.14:kernel/debug/kdb/kdb_main.c:#define K(x) ((x) << (PAGE_SHIFT -
>>>>> 10))
>>>>> v5.14:mm/backing-dev.c:#define K(x) ((x) << (PAGE_SHIFT - 10))
>>>>> v5.14:mm/memcontrol.c:#define K(x) ((x) << (PAGE_SHIFT-10))
>>>>> v5.14:mm/oom_kill.c:#define K(x) ((x) << (PAGE_SHIFT-10))
>>>>> v5.14:mm/page_alloc.c:#define K(x) ((x) << (PAGE_SHIFT-10))
>>>>> ```
>>>>>
>>>>> Shouldn't this macro go to some header file instead to get rid of define
>>>>> repetitions?
>>>>
>>>> Many thanks for figuring this out. I think we should get rid of these
>>>> repetitions too. But I'am not sure where this definition should be. Any
>>>> suggestion? Thanks.
>>>
>>> I'm not sure what place suits best. At first I thought maybe linux/mm.h or
>>> linux/mm_inline.h, but since PAGE_SHIFT is declared in asm-generic/page.h,
>>> probably K(x) can also go there as well?
>>
>> K(x) is relevant with PAGE_SHIFT. So I think K(x) can also go
>> asm-generic/page.h too.
> 
> Actually, the comment in this file says:
> 
> ```
> 4 /*
> 5  * Generic page.h implementation, for NOMMU architectures.
> 6  * This provides the dummy definitions for the memory management.
> 7  */
> ```
> 
> so it seems I was wrong about this being an appropriate place.

It's hard to find the appropriate place.

> 
>> Am I supposed to do this when free or will you
>> kindly do this?
> 
> Let me just try to cram this into mm.h and send it out, and then we'll see 
> what comments people suggest.
> 

Many thanks for doing this. :)

> Thanks.
>