Opened 2 years ago
Last modified 2 years ago
#21587 new defect
VINF_EM_TRIPLE_FAULT while using nested L2 guest machine
Reported by: | spn | Owned by: | |
---|---|---|---|
Component: | other | Version: | VirtualBox-6.1.42 |
Keywords: | nested virtualisation | Cc: | |
Guest type: | Linux | Host type: | Windows |
Description
Hello, im trying to virtualise my CentoOS7 linux as nested guest machine L2 (this way: L0=Win10, L1=Win7, L2=Centos7). One of my program (on L2) immediately after I start it gives "VCPU0: Guru Meditation 1155 (VINF_EM_TRIPLE_FAULT)" on L2 machine and stop. The interesting is when L2 machine works as L1 then triple fault never happen, so nested virtualisation must have impact on this. Problem has been discussed previously here: https://forums.virtualbox.org/viewtopic.php?t=108996
User fth0 found something that possibly may by important here:
fth0 Knowing that the Linux kernel code is located in the virtual memory area between 0xffffffff80000000 and 0xffffffffa0000000, I'd guess that a SYSENTER call to 0x00000000816b6f40 is missing a sign extension of its address.
Is it possible to fix it?
Attached logs (both L0,L1) from VirtualBox rel. 6.1.42-155177 for better analysis, but my target L1 machine api is 5_2.
Attachments (1)
Change History (3)
by , 2 years ago
Attachment: | VBox-L0L1logs.7z added |
---|
comment:1 by , 2 years ago
Small addition for the devs:
The VM-exit history contains a lot of RIPs of the form 0xffffffff81xxxxxx, only the newest three history entries were missing all the high bits (RIP 0x00000000816b6f40). The numerically closest RIP was 0xffffffff816ab4a6 with VMX_EXIT_HLT.
comment:2 by , 2 years ago
Version: | VirtualBox 5.2.42 → VirtualBox-6.1.42 |
---|
Logs from L0 and L1 machine