I'm working on a customized install disk for our x86-64 systems using RHEL ES 3.0 update 4. My problem is that even though the smp kernel is on the CD, and even though the systems are dual cpu systems, anaconda won't install the smp kernel. I can manually install the smp kernel after I've booted the freshly installed system and everything works as intended. What could I be missing that's causing anaconda to skip installing the smp kernel?
thanks, Daniel
On Wed, 2005-03-02 at 16:55 -0500, Daniel Spratlen wrote:
I'm working on a customized install disk for our x86-64 systems using RHEL ES 3.0 update 4. My problem is that even though the smp kernel is on the CD, and even though the systems are dual cpu systems, anaconda won't install the smp kernel. I can manually install the smp kernel after I've booted the freshly installed system and everything works as intended. What could I be missing that's causing anaconda to skip installing the smp kernel?
thanks, Daniel
Anaconda-devel-list mailing list Anaconda-devel-list@redhat.com https://www.redhat.com/mailman/listinfo/anaconda-devel-list
I ended up figuring this one out and I thought I'd share the info. We are using Sun v20z servers, and it seems that the bios was the cause. After updating the bios anaconda now installs the smp kernel. Does anyone know what portion(s) of code in anaconda detects the number of cpu's in a system?
thanks, Daniel
On Thu, Mar 03, 2005 at 03:31:20PM -0500, Daniel Spratlen wrote:
I ended up figuring this one out and I thought I'd share the info. We are using Sun v20z servers, and it seems that the bios was the cause. After updating the bios anaconda now installs the smp kernel. Does anyone know what portion(s) of code in anaconda detects the number of cpu's in a system?
anaconda/isys/smp.c. On i386 it walks the MP tables in the BIOS. These days BIOS providers are screwing up the MP tables every now and then because the ACPI CPU information is getting used instead of the "legacy" MP tables in the BIOS. It's pretty likely that the BIOS update just fixed the MP tables.
Cheers,
Matt
anaconda-devel@lists.fedoraproject.org