After trying "-cpu host" for month, I can confirm that I can no longer reproduce this issue.
This seems to have resolved the weird MSR from win32 kernel. Thought to let you guys know! On Thu, Feb 2, 2017 at 9:05 AM, Micky <[email protected]> wrote: > Thanks for checking, Robert. > > I looked that up. NONE of the Intel's MSRs are that long[1] > > In source code, I found few MSRs that long but they are only for KVM's > internal usage. > > May be Win32 kernel is going crazy. > > I'm already testing with the '-cpu host' as you suggested in IRC and will > report back. > > Reference: > [1] Chapter 35 of Vol. 3 (full) Intel's System Programming Maual. > > > > > On Thu, Feb 2, 2017 at 6:30 AM, Robert Mustacchi <[email protected]> wrote: > >> On 1/30/17 23:17 , Micky wrote: >> > Does anyone know what kind of MSR instructions are these: >> > >> > unhandled rdmsr: 0xc >> > unhandled wrmsr: 0x50cc47 data 90 >> > >> > Because every time this happens, the KVM VM crashes without any trace or >> > activity in kvmstat. >> >> You'll want to go through the Intel manuals to determine what exactly >> that MSR is for. Unfortunately I'm not sure off hand. >> >> Robert >> > > ------------------------------------------- smartos-discuss Archives: https://www.listbox.com/member/archive/184463/=now RSS Feed: https://www.listbox.com/member/archive/rss/184463/25769125-55cfbc00 Modify Your Subscription: https://www.listbox.com/member/?member_id=25769125&id_secret=25769125-7688e9fb Powered by Listbox: http://www.listbox.com
