Hello All, I am having trouble with KDB and v2.6. My kernel and patches are as follows: linux 2.6.7-bk4 kdb-v4.4-2.6.7-common-1 kdb-v4.4-2.6.7-i386-1
This problem was first seen in a beta Disto v2.6 based release. My hardware is and IBM x445(summit) 8-way with 15.5 gb of ram and a clustered apic. It is a little bit more than a regular smp machine which is why I think I am having problems. My issue is as follows... If I don't enable kdb everything works fine. When I enable and boot with KDB I fall into kdb when I bring the secondary cpus online. KDB claims it was entered via an NMI. Without kdb there are no nmis being registered by the system. When the system is in this entered kdb state I only have access to the first cpu (or n-1 cpus that have been initialized if you "go"). This cpu "0" is doing timer initialization but the boot messages show that another cpu is starting up. Like I said I don't have kdb visibility to it but another cpu is starting. I am trying to get this sorted out. I don't know what is generating the NMI that causes the system to drop into kdb. My best guess is the cpu being brought online is sending it? What kdb initialization hooks should I look at? What needs to happen in order to use the "cpu" command? Thanks for any and all info, Keith Mannthey --------------------------- Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe.
