Re: [PATCH 1/7] cgroups: Shrink struct cgroup_subsys

[Li Zefan <lizf@cn.fujitsu.com>]

Paul Menage wrote:
> On Tue, Nov 9, 2010 at 4:52 PM, Li Zefan <lizf@cn.fujitsu.com> wrote:
>>> bool active:1;
>>> bool disabled:1;
>>>
>> It won't compile, but unsigned char active:1 will do. ;)
> 
> Are you sure? I don't have a buildable kernel tree at the moment, but
> the following fragment compiled fine for me (with gcc 4.4.3):
> 
> struct foo {
>   _Bool b1:1;
>   _Bool b2:1;
> };
> 
> and was sized at one byte. And "bool" is just a typedef of _Bool in
> the kernel headers.
> 

Oops, I just used bool outside kernel tree..

>> Every thing that reduces code size (without sacrifice readability
>> and maintain maintainability) should be worth.
> 
> Agreed, within reason. But this patch doesn't reduce code size - it

I meant binary size.

> makes the code fractionally more complicated and reduces the *binary*
> size by a few bytes.
> 

It's a commonly used skill in kernel code, so I can't say it makes
code more complicated.

That said, I'll happily drop this patch. It just came to me when I
started to add new bool values to the structure. Or if you prefer
bool xxx:1 or just bool xxx, I can do that.

>> This is one of the reasons we accept patches that replacing
>> kmalloc+memset with kzalloc, which just saves 8 bytes in my box.
>>
> 
> Replacing two function calls with one function call is a code
> simplification and hence (generally) a good thing - the minuscule
> reduction in binary size reduction that comes with it is just noise.
>