Subject: Re: Is gcc slow? Or is our gcc slow?
To: der Mouse <mouse@Collatz.McRCIM.McGill.EDU>
From: Peter Svensson <petersv@df.lth.se>
List: current-users
Date: 04/10/1996 21:42:09
On Wed, 10 Apr 1996, der Mouse wrote:

> Hm, self-tuning, that sounds like (a) a really cool idea and (b) a
> maintenance nightmare when it goes wrong.  When will you have a sample
> implementation ready for us to experiment with?

VMS does it really well, once you have given it a suitable starting 
point. After having fed it the minimum required for a few key fields we 
just let it run for a week, and then ordered it to tune itself using data 
collected since the last reboot. Most impressive to watch. 

Such a self-tuning utility in combination with the possibility to change 
the sizes of partitions (didn't someone write such a tool?) would indeed 
be very cool.

I think the major obstacle regating self-tuning is gathering the 
necesarry statistics to work on. If I only knew what tunable options 
there is I might have a go at such a tool. Is there a list of such 
options? 

Another neat thing would be a kernel config builder which reads the 
output of a generic kernel and generates a suitable kernel config file 
for the hardware. I am afraid my contribution to such a project would be 
limited to the sparc port, since that is the only hardware I have running 
NetBSD. Anyone else interested?

Peter