On Tue, Mar 13, 2018 at 08:23:40AM +0100, Piotr Isajew wrote: > On Sun, Mar 11, 2018 at 04:49:28PM -0700, Mike Larkin wrote: > > > Can you build with VMM_DEBUG and see what the final error is, > > please? > > > > I bet it is something we already know about. > > Of course: > > vm_impl_init_vmx: created vm_map @ 0xffff800000c3e200 > vm_resetcpu: resetting vm 1 vcpu 0 to power on defaults > guest eptp = 0x4114f701e > vmm_alloc_vpid: allocated VPID/ASID 1
... snip ... > vmx_handle_rdmsr: rdmsr exit, msr=0x639, data returned to guest=0x0:0x0 > vmx_handle_rdmsr: rdmsr exit, msr=0x641, data returned to guest=0x0:0x0 > vmx_handle_rdmsr: rdmsr exit, msr=0x619, data returned to guest=0x0:0x0 > vmx_handle_rdmsr: rdmsr exit, msr=0x179, data returned to guest=0x0:0x0 > vmx_handle_rdmsr: rdmsr exit, msr=0x17a, data returned to guest=0x0:0x0 > vmx_handle_exit: unhandled exit 49 (EPT misconfiguration) > vmm_free_vpid: freed VPID/ASID 1 > Yep, it's what I thought. I have a halfway working fix for this, but for now, the best advice I can give is to try to reduce the memory in the VM a bit until I have a chance to commit the fix. I've noticed I hit these more often when the host goes into swap but some people have reported it happens even without that occurring. My new skylake machine hits these a lot, which is somewhat of a incentive for me to fix it :) -ml
