On Thu Oct 10 19:03 +0100, rowland wrote: > everybody who has an i686 machine is missing the point. Every rpm on the > installation cd's are for i568, so why is there a directory named i686. the > fact is that if the rpms are for i586 then the kernel should be for i586 and > anybody who has a i586 should recompile the kernel themselves. ok the > processer or the software that identifies it is bugged, but if the i686 > directory had not been there, this whole problem would not have happened :-)
No, we're not missing the point. The underlying flaw is that the C3 misreports itself. That is a big no-no. The kernel *is* built for i586... [root@tatiana root]# cat /boot/config | grep -C 4 "586" # Processor type and features # # CONFIG_M386 is not set # CONFIG_M486 is not set CONFIG_M586=y # CONFIG_M586TSC is not set # CONFIG_M586MMX is not set # CONFIG_M686 is not set # CONFIG_MPENTIUMIII is not set # CONFIG_MPENTIUM4 is not set # CONFIG_MK6 is not set Perhaps the glibc RPMs are mistagged... maybe Red Hat's style of having two glibc's is better. However, for performance reasons (glibc functions are called sufficiently often for optimization to be noticeable), DrakX would probably run a CPUID test and install the "appropriate" glibc. Guess what: that still doesn't solve the problem. The bulk of the blame rests on VIA (or possibly the kernel developers). Passing the blame to Mandrake for their packaging of glibc is ludicrous. By your logic, you are somewhat at fault: if you hadn't tried to install Mandrake on broken hardware this whole problem would not have happened :-) PS: please fix your Reply-To headers... there's no need to set them to your From: address and they screw up the list. -- Levi Ramsey [EMAIL PROTECTED] [EMAIL PROTECTED] Love lies in pools of questions. GPG Key Fingerprint: 354C 7A02 77C5 9EE7 8538 4E8D DCD9 B4B0 DC35 67CD Currently playing: John Barry - Capsule in Space Linux 2.4.19-16mdk 6:00pm up 6 days, 16:25, 8 users, load average: 0.19, 0.21, 0.15
