[PATCH 0/18] xfs: metadata and buffer cache scalability improvements

[Christoph Hellwig]

> 	1. work out how to efficiently support block size smaller
> 	than page size. The current code works, but uses a page per
> 	sub-apge buffer.  A set of slab caches would be perfect for
> 	this use, but I'm not sure that we are allowed to use them
> 	for IO anymore. Christoph?

Using slab pages for I/O is fine again.  Back when we used them we
couldn't get agreement from driver authors that they need to support
them, but now that ext4 has started using them they are fine..

I'm not even sure we'll need separate slab caches, the normal kmalloc
caches probably are good enough.