Messages in this thread | | | Date | Wed, 14 Dec 2011 08:53:41 -0600 (CST) | From | Christoph Lameter <> | Subject | RE: [PATCH 1/3] slub: set a criteria for slub node partial adding |
| |
On Wed, 14 Dec 2011, Pekka Enberg wrote:
> On Wed, 14 Dec 2011, Eric Dumazet wrote: > > We should try to batch things a bit, instead of doing a very small unit > > of work in slow path. > > > > We now have a very fast fastpath, but inefficient slow path. > > > > SLAB has a litle cache per cpu, we could add one to SLUB for freed > > objects, not belonging to current slab. This could avoid all these > > activate/deactivate overhead. > > Yeah, this is definitely worth looking at.
We have been down this road repeatedly. Nick tried it, I tried it and neither got us to something we liked. Please consult the archives.
There was a whole patch series last year that I did introducing per cpu caches which ended up in the "unified" patches. See the archives for the various attempts please.
| |