Configuration of Compaq R3000

Kelly Black kjblack at
Thu Jan 27 12:21:34 PST 2005

Dear Jung-uk Kim,

It works! Thank you! I downloaded it and built the module. I can't
read Japanese so the readme was not much use. I blindly tried "make ;
make install" and it put the module in /boot/kernel. Fortunately, the
comments in the program are pretty good. When I tried "kldload
acpi_ppc"  it loaded the module with no trouble. (I had already
applied the patchs that you had posted earlier.)

This time when I ran it the time was down to 95 seconds which is
faster than my desktop system, a one year old P4, and is much faster
than CygWin. (Our opteron system runs this program in 81 seconds.) It
was pretty cool to start the process up and a little while later hear
the fan kick in. There is nothing like instant feedback! :-)

The only problem is that I have to run the machine with acpi enabled
to take advantage of the driver. I've had some stability problems with
acpi enabled so I will probaly only use this under special

Again, thank you for this pointer and thank you for all your work on
this machine. Your work made it possible to get FreeBSD working on
this beast. It has been surprising how many little things  have made
this machine so difficult to configure.

I will try to attach a copy of the program and makefile with this
email as per your request. To create an executable type "make hermite"
and then run the executable, ./hermite, from the command line.  You
will have to create a directory called "graph" wherever you happen to
put the files because at the start of the run it writes a file to that

Finally, there is a stochastic component, and a random variable is
added at each time step. The time required will vary from slightly
from run to run, so you have to make multiple runs and average them to
get a better idea of how long it takes. This is not quite the full
blown version. We double the matrix size and double the time steps for
the bigger problem and have to make multiple runs because of the
stochastic part.


On Thu, 27 Jan 2005 13:59:29 -0500, Jung-uk Kim <jkim at> wrote:
> On Thursday 27 January 2005 10:52 am, Kelly Black wrote:
> > Hello,
> >
> > I have installed FreeBSD 5.3 on a Compaq Presario R3000. It has an
> > AMD 64 bit processor.  First, I am very grateful for all of the
> > work done by Jung-uk Kim. If it were not for his work it would not
> > have been possible for me to get my machine going under FreeBSD.
> :-)  You are very welcome.
> > Second, there appears to be a very serious performance issue with
> > this machine. I have a program that I use for benchmarks which
> > approximates a nonlinear PDE. It uses a 300x300 matrix which is
> > relatively small so it does not require much in the way of swap or
> > memory caching. When I run this program under freebsd it takes 234
> > seconds in user time (not real). I have CygWin installed on the
> > windows slice and ran the exact same program. I have the 32 bit
> > version of Windows XP. Under CygWin the program takes 155 seconds
> > in user time.
> >
> > This is almost 33% better under MS windows and is a huge
> > discrepancy! It seems clear that something isn't right on the
> > machine. I have attached a copy of my kernel file. I have  applied
> > many of the patches so generously made available by Jung-uk Kim on
> > this list.  The times have remained consistent before and after all
> > of updates.  I have used RELENG_5 as the src-all tag in my
> > cvsup,conf,
> >
> > Any pointers on what I can do to tune this machine would be very
> > helpful. Has anybody come across this problem?
> Did you use 'acpi_ppc' driver?
> You need this driver to run your laptop at full speed.  CPU speed will
> be automatically adjusted by CPU load.  However, it may take few
> seconds to get to the speed.  You can set the speed manually by:
> sysctl hw.acpi.cpu.px_control=0
> Please read comment in the acpi_ppc.c for more info.
> Unfortunately if you run this laptop under heavy CPU load, CPU will
> heat up pretty fast and 'acpi_ec' will adjust delay to prevent
> excessive heat.  You can ignore this behavior by hacking acpi_ec.c, I
> believe but it's really bad idea.
> Laptop is not for number crunching after all. ;-)
> > By the way, I have used this same program to benchmark an opteron
> > system. On that machine I installed FreeBSD 5.3 (generic kernel)
> > and then installed gentoo (2.6.9 kernel). The times under both OS's
> > were almost identical.
> Hmm...  Interesting.  Can I see the source?
> Thanks,
> Jung-uk Kim
> > Sincerely,
> > Kel
-------------- next part --------------
A non-text attachment was scrubbed...
Name: makefile
Type: application/octet-stream
Size: 527 bytes
Desc: not available
Url :

More information about the freebsd-amd64 mailing list