Re: [patch][rfc] acpi: do not use kmem caches

[Alexey Starikovskiy <aystarik@gmail.com>]

Nick Piggin wrote:
> On Mon, Dec 01, 2008 at 08:04:21PM +0300, Alexey Starikovskiy wrote:
>   
>> Nick Piggin wrote:
>>     
>>> Hmm.
>>> Acpi-Operand        2641   2773     64   59    1 : tunables  120   60    8 
>>> : slabdata     47     47     0
>>> Acpi-ParseExt          0      0     64   59    1 : tunables  120   60    8 
>>> : slabdata      0      0     0
>>> Acpi-Parse             0      0     40   92    1 : tunables  120   60    8 
>>> : slabdata      0      0     0
>>> Acpi-State             0      0     80   48    1 : tunables  120   60    8 
>>> : slabdata      0      0     0
>>> Acpi-Namespace      1711   1792     32  112    1 : tunables  120   60    8 
>>> : slabdata     16     16     0
>>>
>>>  
>>> Looks different for my thinkpad.
>>>  
>>>       
>> Probably this is SLUB vs. SLAB thing Pecca was talking about...
>>     
>
> Sizes should not be bigger with SLUB. Although if you have SLUB debugging
> turned on then maybe the size gets padded with redzones, but in that
> configuration you don't expect memory saving anyway because padding bloats
> things up.
>
>
>   
>> And, probably you run at 32-bit? This is part of my .config:
>>     
>
> No, 64 bit.
>
>
>   
>> --------------------------------------------
>> CONFIG_SLUB_DEBUG=y
>> # CONFIG_SLAB is not set
>> CONFIG_SLUB=y
>> # CONFIG_SLOB is not set
>> -------------------------------------------
>>
>> With your patch you would be able to save 64*(2773 - 2641) + 32 * 
>> (1792-1711)= 8448 + 2592 = 11040 bytes of memory, less than 3 pages?
>>     
>
> You don't account the cost of the kmem cache. Or fragmentation that
> can be caused with extra kmem caches.  I guess neither is any problem
> with SLOB, which is used by tiny systems...
>   
There is very small amount of fragmentation -- underfill, maybe 3 pages 
out of 63, less than 5%.