Subject: Re: FreeBSD Bus DMA
To: None <>
From: Perry E. Metzger <>
List: tech-kern
Date: 06/12/1998 02:41:34
"John S. Dyson" writes:
> Perry E. Metzger said:
> > 
> > "John S. Dyson" writes:
> > > I will state conclusively: the compiles were 2.5X to 3X faster,
> > 
> > Please also state:
> > 
> > 1) the code base being compiled
> Our product.

Which is? And if I cannot get my hands on this code corpus, can you
provide another code corpus that is publically available so that the
experiment may be reproduced?

> > 2) the compile times
> About 1.5Hrs vs. about 30-40mins or so.

"About" doesn't work. We need the times.

> > 3) the machine types
> X86. 

What speed x86? Model, processor, memory, disk, etc. Please give us
everything we would need to try this at home and see if our results
match yours. Were the machines completely identical? If not, do you
have evidence on why the differences between the machines were not

> > and
> > 4) the versions of the OS that were running.
> > 
> FreeBSD-current, vs. a recent NetBSD.

WHICH NetBSD? "Recent" means different things to different people. I
know folks who think 1.2 is recent. WHICH FreeBSD-current? Which
options were both compiled with?

> > Perry
> > PS You stated I was lying, btw, when I said that you claimed compiles
> > were 3X faster.
> > 
> Yes I did, and what you just claimed is not what I claimed.

Lets see. You've just claimed that a code corpus that takes 30 minutes 
on FreeBSD takes 1.5 hours on NetBSD.

I surely can't be the only one here wondering at the distinction
between what I've said you've claimed and what you've written that you 
are claiming.