Linked by Robert Minvielle on Tue 16th Dec 2003 20:00 UTC
This is the second installment of the "Linux on the Opteron, are we ready?" article. Basically, it is a "where are we now?" article, noting that what once did work now does not, and others that did not work now do. The first article was published on OSNews almost three months ago. Since that time not too much has happened publicly in regards to the amd64 Linux situation, but a lot of people mailed to tell me that I should have checked out SuSE or the new Mandrake which was "about to be released" at that time. Also since that time I have upgraded the RAM and acquired a larger hard disk for the machine. I will give a brief rundown of the system as it stands now, what I tried to install on it, and what works.
Permalink for comment
To read all comments associated with this story, please click here.
To read all comments associated with this story, please click here.
Don't necessarily blame the BIOS upgrade for the disk problems you experienced with Mandrake.
I have an old Compaq Deskpro 4000 which I upgraded the CPU on with an Evergreen 400MHz AMD 6 CPU upgrade kit and a Maxtor 30GB hard drive and the original Maxtor 2GB drive. I can install Red Hat 7.0 on it with absolutely no problems. When I tried to install Red Hat 7.3, Red hat 8.0 and Mandrake 9.0 on it, ALL THREE of them barfed on the Maxtor 30GB drive, claiming they could not read the partition table even after ALL THREE of them themselves created and initialized the partition table. Whether with fdisk or Partition Magic or their own tools, NONE of them could handle the drive.
This on the same machine and drive that Red Hat 7.0 blows right on with not one single complaint about the drive or anything else.
This tells me that sometime between RH 7.0 and 7.3 somebody fscked up the IDE kernel code so that certain drives simply cannot be handled. Either that or certain BIOS's like the Compaq (which is an oddball BIOS anyway) cannot be handled. My understanding is that after the Linux loader is done, Linux does not use the BIOS for accessing the hard drive geometry, which seems to me to leave it as the fault of the kernel IDE code.
Somebody needs to look into this since who knows where and on what drives this problem will bite.