Messages in this thread | | | Date | Wed, 10 Nov 1999 09:30:57 -0700 | From | yodaiken@chelm ... | Subject | Re: PATCH 2.3.26: kmalloc GFP_ZERO |
| |
On Wed, Nov 10, 1999 at 01:44:51AM +0100, Andrea Arcangeli wrote: > On Tue, 9 Nov 1999 yodaiken@chelm.cs.nmt.edu wrote: > > >same could be said for nearly everything in the kernel. > > The only point of such change to improve performances. If the improvement > make me to be slower somehow I can't agree in an obvious manner at least. > > Exactly which is the order of the improvement on PPC and did you measured > the worst case?
20% or so on kernel compiles. And what do you mean by "Worst case" in this instance? If the worst case is when the system is never idle, costs are a false branch in each getpage. One question is whether a never idle system represents an interesting case. I suspect that it takes a very unusual load to get such a situation and that on, for example, smp servers where the zero page cache will have the best results, it is the general case that there is a good bit of idle time.
- 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/
| |