Hide Forgot
Description of problem: run sandra tool in win2k3-64 guest and click Hardware-->Processors, guest show BSOD Version-Release number of selected component (if applicable): host: # uname -r 2.6.32-259.el6.x86_64 #rpm -qa|grep qemu qemu-kvm-0.12.1.2-2.270.el6.x86_64 Cpu:Intel(R) Xeon(R) CPU E31225 @ 3.10GHz guest: win2k3 64 bit sp2 How reproducible: 100%(2/2) Steps to Reproduce: 1./usr/libexec/qemu-kvm -M rhel6.3.0 -cpu SandyBridge --enable-kvm -m 4G -smp 8,sockets=2,cores=4,threads=1,maxcpus=10 -name rhel6.3 -uuid ddcbfb49-3411-1701-3c36-6bdbc00bedbb -rtc base=utc,clock=host,driftfix=slew -boot c -drive file=/mnt/win2k3-64.qcow2,format=qcow2,if=none,id=disk0,cache=none,werror=stop,rerror=stop -device virtio-blk-pci,drive=disk0,id=disk0,bootindex=1,scsi=off -netdev tap,id=hostnet1 -device virtio-net-pci,netdev=hostnet1,id=net1,mac=86:12:50:a4:32:72 -spice port=5911,disable-ticketing -vga qxl -device sga -chardev socket,id=serial0,path=/var/test1,server,nowait -device isa-serial,chardev=serial0 -balloon virtio -monitor unix:/tmp/monitor1,server,nowait -monitor stdio 2. run sandra 3. Actual results: guest BSOD Expected results: guest work well Additional info:
Created attachment 576998 [details] win2k3 guest BSOD snapshot
The "sandra tool" is the one available from http://www.sisoftware.net/ ? If yes, it's not free (for non-personal use), do we have a company-wide license to use it?
Please try to reproduce it on a different CPU and/or different windows version so that we get an idea if this is a specific scenario or not.
Sandra is known to try to read non-architectural MSRs that we don't emulate, based solely on the CPU model IDs. Probably this is a duplicate of (or very similar in nature to) Bug 691382. It's probably not going to be fixed, unless it's an easy to emulate MSR (that was not the case on bug 691382). Do you have any messages on the host dmesg that can help us identify the MSR being read/written?
I (In reply to comment #5) > Please try to reproduce it on a different CPU and/or different windows version > so that we get an idea if this is a specific scenario or not. I can reproduce it with another host(AMD Phenom(tm) 9600B Quad-Core Processor).
(In reply to comment #6) > Do you have any messages on the host dmesg that can help us identify the MSR > being read/written? |--> needinfo
(In reply to comment #9) > (In reply to comment #6) > > Do you have any messages on the host dmesg that can help us identify the MSR > > being read/written? > > |--> needinfo this is dmesg log when guest BSOD. __ratelimit: 57 callbacks suppressed kvm: 3114: cpu0 unhandled rdmsr: 0xce kvm: 3114: cpu0 unhandled wrmsr: 0x198 data 0 kvm: 3114: cpu1 unhandled wrmsr: 0x198 data 0 kvm: 3114: cpu2 unhandled wrmsr: 0x198 data 0 kvm: 3114: cpu3 unhandled wrmsr: 0x198 data 0 kvm: 3114: cpu4 unhandled wrmsr: 0x198 data 0 kvm: 3114: cpu5 unhandled wrmsr: 0x198 data 0 kvm: 3114: cpu6 unhandled wrmsr: 0x198 data 0 kvm: 3114: cpu7 unhandled wrmsr: 0x198 data 0
(In reply to comment #8) > I (In reply to comment #5) > > Please try to reproduce it on a different CPU and/or different windows version > > so that we get an idea if this is a specific scenario or not. > > I can reproduce it with another host(AMD Phenom(tm) 9600B Quad-Core Processor). AMD Phenon would never run a "-cpu SandyBridge" guest. Being a different CPU model, probably it's crashing on a different MSR (and it would be a different bug). Could you open a new bug with that information?
I will close it as duplicate of bug 691382, based on info of comment 10: > kvm: 3114: cpu0 unhandled rdmsr: 0xce *** This bug has been marked as a duplicate of bug 691382 ***