Re: [PATCH -mm -v2] mm, swap: Use kvzalloc to allocate some swap data structure

[Matthew Wilcox]

On Wed, Apr 05, 2017 at 03:10:58PM +0800, Huang, Ying wrote:
> In general, kmalloc() will have less memory fragmentation than
> vmalloc().  From Dave Hansen: For example, we have a two-page data
> structure.  vmalloc() takes two effectively random order-0 pages,
> probably from two different 2M pages and pins them.  That "kills" two
> 2M pages.  kmalloc(), allocating two *contiguous* pages, is very
> unlikely to cross a 2M boundary (it theoretically could).  That means
> it will only "kill" the possibility of a single 2M page.  More 2M
> pages == less fragmentation.

Wait, what?  How does kmalloc() manage to allocate two pages that cross
a 2MB boundary?  AFAIK if you ask kmalloc to allocate N pages, it asks
the page allocator for an order-log(N) page allocation.  Being a buddy
allocator, that comes back with an aligned set of pages.  There's no
way it can get the last page from a 2MB region and the first page from
the next 2MB region.