On 08/02/2012 04:55 PM, Shawn Joo wrote:
>>> then yes, they will allocate the necessary number of pages from the standard page allocator.
> Who is "the standard page allocator" for cache in /proc/slabinfo, e.g. "size-65536" ?
> I believe one of allocator is buddy. who else?
>
The generic and algorithm-neutral answer to this is "whoever would
handle alloc_pages()".
In the specific case, yes, this is the buddy allocator.
Take a look at mm/slab.c, for instance:
When a cache can't service an allocation, it does:
cache_grow()
-> kmem_getpages()
-> alloc_pages_exact_node()
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>