Re: [PATCH] per-cpu slab

kumon@flab.fujitsu.co.jp
Sat, 10 Jun 2000 23:19:11 +0900

Momchil Velikov writes:
> kumon@flab.fujitsu.co.jp wrote:
> > Returning a block to the original CPU's slab requires extra cost for
> ^^^^^^^^^^
> Does you implementation has CPU-private slabs ?

I implemented not CPU-private slabs but per-CPU slab caches. Most
recently returned blocks are held in limited-depth stacks (LIFO).
Every slab has stacks for CPUs.

The system slab is accessed only when a stack overflow or underflow happens.

--
Computer Systems Laboratory, Fujitsu Labs.
kumon@flab.fujitsu.co.jp


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/