Subject: Re: Compiler timings on varous MVII NetBSDs etc.
To: None <>
From: Anders Magnusson <>
List: port-vax
Date: 01/24/2001 19:56:48
> On Wed, Jan 24, 2001 at 11:51:42AM -0500, John Franklin wrote:
> >
> > There are other parts that affect everything or nearly everything in the
> > system, and they can be optimized a bit as well.  UVM is measurably
> > better than the old VM system. I don't have numbers, but I wouldn't be
> > surprised to find a sizable percentage of the "kernel bloat" came from
> > the switch to UVM.  Is there some way that the UVM system can be better
> > optimized for the VAX?  Are there macros or inlines that could be added
> > such that they are transparent to other ports, but translate to a machdep'd
> > function for VAX to produce improvements?
> I'd be interested to see how a kernel compiled with -Dinline= works for
> you.  On a small-memory system, it might run quite a bit faster.
> If you do that, you probably want to add -fno-inline-functions to the
> compiler command line, too.
If not something strange has happened, UVM is rather much faster on
VAX compared to Mach VM. When UVM was incorporated the time for running
the bootup rc scripts shrank from 110 seconds to 100 seconds on a 9MB MV2.
I assume this is still the case.

The old VM (Pre-Mach) was probably much faster due to its simplicity, 
but it was also missing some important pieces that could make it very
inefficient under some circumstances. For example Copy-On-Write didn't 
exists, there were no vnode paging and mmap() wasn't possible.

-- Ragge