I'm having some trouble performing a successful vmlaunch. The error I keep getting is "VM entry with invalid host-state field(s)". I thought almost 1500 lines of consistency checks and vmx debugging implemented from 3B.23 would be enough to trace the problem down. I would greatly appreciate any help regarding this issue, if anyone could tell me exactly what i'm configuring wrong inside the code. I've provided some initialization code inside the attachment. Actual vmwrites begin inside hvk.c/HvkLaunchGuest (). Thank you.
I'm working on a security measure based on the virtualization instructions. The only requirement is memory isolation. So, to minimize performance impact, a vmexit should only be issued when a (EPT) protected page is accessed. On my processor however, access to control registers always causes an vmexit, resulting in a noticeable impact from the hypervisor. The documentation states that future processors may be more flexible and allow the programmer to disable the vmexits in such situations. Are there processors on the market yet that support this feature?
We would like to use the KVM technology with new intel cpus.
Could you please help me which CPU and chipset should we buy?
I see that there are two technologies VT-x and VT-d.
Do we need a CPU which supports also VT-d or is enough the VT-x technology?
How to return to executive monitor from SMM monitor after activating dual-monitor treatment?
VMLAUNCH on SMM-transfer VMCS reports "VMLAUNCH with non-clear VMCS", VMRESUME on SMM-transfer VMCS reports "VMRESUME with non-launched VMCS".
Does the mobo Intel DG31PR support Virtualization Technology?
I've made a very simple hypervisor. It works at one CPU, at base of Windows7 x64 and has no usefull functionality.
I can not solve two big problem:
1) If I run it at one CPU, I've got a tripple fault after some success VM exits. But if i set "int 3" at begin of non-root code after vmlaunch - host OS works almost very well.
2) If I try to repeat starting VMM at every CPU - I'v got a freezing and then random fatal error.
I suggest, I've done something wrong around interrupt tuning and has corrupt interrupt handling on the host
I hope this is an appropriate forum to post this question to.
I have some systems in the electrical distribution world where I'm trying to simulate many 00's of devices that all talk IEC-61850 for Smart Grid simulations etc.... The IEC-61850 protocols are a mix of IP and a seperate ethernet protocol (seperate ethertype) for high speed inter-device messaging.
I was frustrated that it was problematic to run Ubuntu and Vista and XP on my laptop (I am a consultant and need to have the ability to develop in several platforms) when I realized that I could add OracleVirtual Boxinto Linux (or Vista but I usually use Linux) and run XP . Note that only XP Pro SP3 (which I have) allows virtualization not Home.Vista does not for Home or Business, only Enterprise and Ultimate and the same for Windows 7. XP runs so well on my HP 6830s with a P8400 Duo that I can't tell that it's a virtual machine.
http://www.tomshardware.com/news/knights-ferry-corner-mic-xeon,11036.html It has32 x86 cores at 1.2GHz, with four threads per core. If this isn't intended for virtualization I don't know what is.
Would anybody have a pointer to some documentation about how VTd memory mappings are used in current VMMs? It seems most examples imply that a VTd domain is started as each guest is started and remains basically static for the life of the guest. Thus, I/O devices won't have to deal with frequent VTd invalidations since the domains themselves remain fairly persistent.