Messages in this thread |  | | Date | Sun, 27 Aug 2000 17:59:01 +0100 (BST) | From | Mark Hemment <> | Subject | Re: kmalloc optimization |
| |
Hi,
> > > 1500bytes slab unfortunately is not too useful, because it does not fit well > > > in 4K pages (you would need 8K or 16K page allocations, which the mm system > > > does not like much due to fragmentation) > > > > Doesn't matter (too much). > > On 32MB machines it was a big problem in my experiments.
Opps, I didn't make myself clear. I meant it matters less with the new allocator I'm working on, as it can use different page orders for slabs within the same cache. This mixing gives a good trade off in memory/performance - my 16MB machine doesn't fall over (at least not so often). I keep promising to get the allocator's source out - I must try harder!
Mark
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
|  |