[ale] Trouble with new Pentium D

[Michael B. Trausch]

On Tue, 2007-01-09 at 12:08 -0500, Chuck Huber wrote:

> 
> > Question:  Are you attempting to boot an SMP capable kernel?  I don't
> > have really any experience with dual-core and multiple-CPU SMP systems
> > (under Linux anyway) other than reading about them.  Is it possible that
> > you have a uniprocessor kernel that is doing something strange when it
> > sees two CPUs?
> 
> I suppose it's possible.  However when 9.2 was released multi-processor
> systems were available, though dual-core CPU's weren't.  The rescue
> system boot I tried was off of a Suse 9.2 CD.  I started a download of
> 10.1 and 10.2 CD's and DVD's this morning before I left for work.  They
> should be done by the time I get home provided, of course, that
> HellSouth can keep it together for a day.  I'll burn CD 1 from 10.1
> tonight and try to boot the rescue system off of that.
> 
> As far as SMP vs non-SMP, the 10.1 rescue disk is smart enough to tell
> the difference.  I installed 10.1 from the same version disk on a
> Pentium D at work.  Runs x86 SMP kernels without a problem.  The only
> thing I can figure right now is that the 2.4 kernel on the Suse 9.2 disk
> is pretty dumb w.r.t. the Pentium D.  I'll try the 10.1 rescue disk
> tonight and post the results.
> 


Ick!  I haven't used a 2.4 kernel based system in quite a while; 2.6 is
where it's at.  :-)

Incidentally, don't most of the modern desktop environments and the like
require 2.6, or is it just the features of convenience that they have
that require the 2.6 kernel?


> With all that said, I don't know why the bios boot loader is not
> recognizing the existing partitions (which is why I tried to boot the
> rescue system in the first place).  It (the bios) is recognizing the
> drives and their capacity, so comms between the controller and the
> drives are good.  No cable problems there.
> 
> Do you think the new bios is mapping the drive differently than the old
> bios, and thus the partition and boot information is not being read?


That's a good question.  The last time I had actually checked (which was
when I got Erica's 64-bit system), x86-64 systems still only used
classic BIOS programs, which use the old-style MS-DOS partitioning
scheme (MBR and classic partition table).  Does your new system have EFI
at all?  (I would guess not, if your system booted the old stuff, but I
couldn't be sure.)  If so, though, then I would wager a guess that it
would expect to see a GUID partition table on the drive, and not an
MS-DOS partition table.  Regardless, I think that an EFI firmware is
unlikely on an x86-64 system, but it is still a possibility, I suppose,
since you can download and run the Intel EFI on such systems, "on top
of" the BIOS.


> 
> One of the things I'll try it to install the old MB/CPU and garner some
> information about the drives.
> 
> Chris, if it's any consolation, the Pentium D I have running at work was
> a smooth fresh install.  And man, is it fast!
> 
> Thanks for all the suggestions.  I'll keep the list posted on my progress,
> 


Would be interested to see what works.

    -- Mike

--
Michael B. Trausch
                    fd0man at gmail.com
Phone: (404) 592-5746
                          Jabber IM:
                    fd0man at gmail.com
              fd0man at livejournal.com
Demand Freedom!  Use open and free protocols, standards, and software!
-------------- next part --------------
An HTML attachment was scrubbed...

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part