Almost, but not quite there DVD x86_64

[Chuck Adams]

Some one smarter than me can help on this one.
Hopefully I have not made any typos in the following.

For those that have downloaded the FC3t2 DVD.iso file
and found it too large.  And this is not for the faint of heart.
I pay $0.49 or so for DVD blank discs and money is not an
object, but I have made some coasters over the past few days.... :-)

Here is what I have done, but it lacks one or more thing(s) that I don't know,
so some one can finish where I screwed up......  If I happen to be on
the right track.  This for the obsessive compulsive that want to
get this going this weekend.

As root run through all the following commands:

mkdir    /tmp/newDVD

(go to the directory where you have the iso file for FC3t2)

mount FC3-test2-x86_64-DVD.iso  -r  -t iso9660  -o loop   /mnt

cd   /mnt

cp -r   .discinfo    *     /tmp/newDVD

(at this point you can leisurely move to the kitchen and make
yourself a peanut butter and banana sandwich and chow down
with a glass of milk)  :-)

Now you have the expanded DVD in /tmp/newDVD.

Go to

http://www.k7qo.net/rmscript

and you'll find a script that will remove the *.i386.rpm files that
are duplicated by a *.x86_64.rpm  file.  Put this script in a directory
and

chmod +x rmscript

Now execute the script

./rmscript

You now have a directory structure without the unneeded rpm's for the i386
and this is where there is an option that is needed that I don't know about.

I did a

mkisofs -U -R /tmp/newDVD   >  /tmp/FC3t2DVD.iso

and I was able to generate a new iso file.  I then used K3b to generate
a new DVD but it will not boot.  vmlinuz is not uncompressed and then
the loading process is not started.....  The file system on this DVD, when 
mounted looks
good but something is missing.   email me direct if you know what I
did wrong and I'll then repost (if desired) all the necessary steps.
I don't mind the time or the money......

I have a C code that runs in less than 18K (thus L1 cache) in the
AMD Athlon 64.  It is a serious 64-bit floating point code that will
bring any system to it's knees.  It runs 33%+ faster than an AMD
Athlon XP at the same clock speed (2GHz).  Monte Carlo simulation
of electromagnetic transport in real planetary atmospheres......
I have some cases that take over 11hrs to run and then spit out
a few hundred numbers through stdout.  So you can see the obsession
for getting an AMD64 running full tilt ASAP.

Again, help appreciated.  And hoping that I don't get banned.

TIA





Chuck Adams, K7QO     CP-60 WPM
k7qo at commspeed.net
http://www.qsl.net/k7qo                 Ham Radio Stuff
http://www.k7qo.net/                     Physics Stuff

Moving to Arizona?   Bring your own water, please.