Messages in this thread | | | Date | Wed, 16 Jan 2013 17:45:00 +0900 | From | Joonsoo Kim <> | Subject | Re: [PATCH 2/3] slub: correct bootstrap() for kmem_cache, kmem_cache_node |
| |
On Tue, Jan 15, 2013 at 03:36:10PM +0000, Christoph Lameter wrote: > On Tue, 15 Jan 2013, Joonsoo Kim wrote: > > > These didn't make any error previously, because we normally don't free > > objects which comes from kmem_cache's first slab and kmem_cache_node's. > > And these slabs are on the partial list because the objects are typically > relatively small compared to page size. Do you have a system with a very > large kmem_cache size?
These slabs are not on the partial list, but on the cpu_slab of boot cpu. Reason for this is described in changelog. Because these slabs are not on partial list, we need to check kmem_cache_cpu's cpu slab. This patch implement it.
> > Problem will be solved if we consider a cpu slab in bootstrap(). > > This patch implement it. > > At boot time only one processor is up so you do not need the loop over all > processors.
Okay! I will fix and submit v2, soon.
| |