[PATCH 0/5] Sizing zones and holes in an architecture independent manner V7

[Andi Kleen]

> Right now, x86_64 seems to be the only arch that accounts for the kernel 
> image and memmap as holes so I would consider it to be unusual.

s/unusual/more advanced/

> For memory  
> hot-add, new memmaps are allocated using kmalloc() and are not accounted 
> for as holes. 

At least in the standard (non sparsemem) hotadd they are accounted afaik.

> So, on x86_64, some memmaps are holes and others are not. 
> 
> Why is it a performance regression if the image and memmap is accounted 
> for as holes? How are those regions different from any other kernel 
> allocation or bootmem allocations for example which are not accounted as 
> holes? 

They are comparatively big and cannot be freed.

>If you are sure that it makes a measurable difference to performance,

There was at least one benchmark/use case where it made a significant
difference, can't remember the exact numbers though.

-Andi