Optimize __vmalloc_node_range_noprof function.

[PATCH 0/4] Optimize __vmalloc_node_range_noprof function.

Posted by Liu Ye 11 months, 1 week ago

The use of variables real_size and real_align in function 
__vmalloc_node_range_noprof is unreadable. Optimize it in four patches.

Liu Ye (4):
  mm/vmalloc: Remove unnecessary size ALIGN in
    __vmalloc_node_range_noprof
  mm/vmalloc: Size should be used instead of real_size in
    __vmalloc_node_range_noprof
  mm/vmalloc: Remove the real_size variable to simplify the code in
    __vmalloc_node_range_noprof
  mm/vmalloc: Rename the variable real_align to original_align to
    prevent misunderstanding

 mm/vmalloc.c | 20 ++++++++------------
 1 file changed, 8 insertions(+), 12 deletions(-)

-- 
2.25.1

Re: [PATCH 0/4] Optimize __vmalloc_node_range_noprof function.

Posted by Uladzislau Rezki 11 months, 1 week ago

On Mon, Mar 03, 2025 at 05:44:06PM +0800, Liu Ye wrote:
> The use of variables real_size and real_align in function 
> __vmalloc_node_range_noprof is unreadable. Optimize it in four patches.
> 
> Liu Ye (4):
>   mm/vmalloc: Remove unnecessary size ALIGN in
>     __vmalloc_node_range_noprof
>   mm/vmalloc: Size should be used instead of real_size in
>     __vmalloc_node_range_noprof
>   mm/vmalloc: Remove the real_size variable to simplify the code in
>     __vmalloc_node_range_noprof
>   mm/vmalloc: Rename the variable real_align to original_align to
>     prevent misunderstanding
> 
>  mm/vmalloc.c | 20 ++++++++------------
>  1 file changed, 8 insertions(+), 12 deletions(-)
> 
> -- 
Let me double check it. Quick question, this series does not
introduce any functional change?

--
Uladzislau Rezki

Re: [PATCH 0/4] Optimize __vmalloc_node_range_noprof function.

Posted by Dev Jain 11 months, 1 week ago


On 03/03/25 9:39 pm, Uladzislau Rezki wrote:
> On Mon, Mar 03, 2025 at 05:44:06PM +0800, Liu Ye wrote:
>> The use of variables real_size and real_align in function
>> __vmalloc_node_range_noprof is unreadable. Optimize it in four patches.
>>
>> Liu Ye (4):
>>    mm/vmalloc: Remove unnecessary size ALIGN in
>>      __vmalloc_node_range_noprof
>>    mm/vmalloc: Size should be used instead of real_size in
>>      __vmalloc_node_range_noprof
>>    mm/vmalloc: Remove the real_size variable to simplify the code in
>>      __vmalloc_node_range_noprof
>>    mm/vmalloc: Rename the variable real_align to original_align to
>>      prevent misunderstanding
>>
>>   mm/vmalloc.c | 20 ++++++++------------
>>   1 file changed, 8 insertions(+), 12 deletions(-)
>>
>> -- 
> Let me double check it. Quick question, this series does not
> introduce any functional change?

Yeah, the cover letter subject is misleading. IMHO it should be more 
like "Refactor" instead of "Optimize".
> 
> --
> Uladzislau Rezki
>

Re: [PATCH 0/4] Optimize __vmalloc_node_range_noprof function.

Posted by liuye 11 months, 1 week ago

在 2025/3/4 13:58, Dev Jain 写道:
>
>
> On 03/03/25 9:39 pm, Uladzislau Rezki wrote:
>> On Mon, Mar 03, 2025 at 05:44:06PM +0800, Liu Ye wrote:
>>> The use of variables real_size and real_align in function
>>> __vmalloc_node_range_noprof is unreadable. Optimize it in four patches.
>>>
>>> Liu Ye (4):
>>>    mm/vmalloc: Remove unnecessary size ALIGN in
>>>      __vmalloc_node_range_noprof
>>>    mm/vmalloc: Size should be used instead of real_size in
>>>      __vmalloc_node_range_noprof
>>>    mm/vmalloc: Remove the real_size variable to simplify the code in
>>>      __vmalloc_node_range_noprof
>>>    mm/vmalloc: Rename the variable real_align to original_align to
>>>      prevent misunderstanding
>>>
>>>   mm/vmalloc.c | 20 ++++++++------------
>>>   1 file changed, 8 insertions(+), 12 deletions(-)
>>>
>>> -- 
>> Let me double check it. Quick question, this series does not
>> introduce any functional change?
>
> Yeah, the cover letter subject is misleading. IMHO it should be more like "Refactor" instead of "Optimize".

Yes, refactoring is more accurate, regarding real_size and real_align.

>>
>> -- 
>> Uladzislau Rezki
>>
>