Re: Large slab cache in 2.6.1

[Manfred Spraul]

> Linus Torvalds <torvalds@xxxxxxxx> wrote:
> > What happened to the experiment of having slab pages on the (in)active
> >  lists and letting them be free'd that way? Didn't somebody already do 
> >  that? Ed Tomlinson and Craig Kulesa?
>
> That was Ed.  Because we cannot reclaim slab pages direct from the LRU
I think that this is needed: Bonwick's slab algorithm (i.e. two-level 
linked lists, implemented in cache_alloc_refill and  free_block) is 
intended for unfreeable objects.
The dentry cache is a cache of freeable objects - a different algorithm 
would be more efficient for shrinking the dentry cache after an updatedb.
I had started prototyping, but didn't get far.
--
   Manfred

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