![]() Intervening L0 VMEXIT during their delivery. In this case, we need to make sure the next_rip field gets synced fromĪnother issue is that pending L1 -> L2 events are forgotten if there is an This is mostly visible with running a nested guest, with L1 trying to inject To the hypervisor - there seems to be no specific effort to maintain this ![]() However, KVM seems to treat this field as a unidirectional hint from the CPU ![]() VM_ENTRY_INSTRUCTION_LEN field, although, in contrast to VMX, it holds anĪbsolute RIP value, not a relative increment. On a VMRUN that does event injection it has similar function as VMX's INT3 or INTO exception (this was confirmed by AMD). The return address pushed on stack when injecting a software interrupt or This field value (instead of the saved guest RIP) in used by the CPU for There are some issues with respect to nSVM L1 -> L2 event injection.įirst, the next_rip field of a VMCB is *not* an output-only field for a VMRUN. Joerg Roedel, Tom Lendacky, Brijesh Singh, Jon Grimm,ĭavid Kaplan, Boris Ostrovsky, Liam Merwick, kvm, linux-kernel ![]() Szmigiero 21:38 UTC ( / raw)Ĭc: Sean Christopherson, Vitaly Kuznetsov, Wanpeng Li, Jim Mattson, ` (4 more replies) 0 siblings, 5 replies 20+ messages in threadįrom: Maciej S. 21:38 ` KVM: nSVM: Sync next_rip field from vmcb12 to vmcb02 Maciej S. ![]() NSVM: L1 -> L2 event injection fixes and a self-test archive mirror help / color / mirror / Atom feed * nSVM: L1 -> L2 event injection fixes and a self-test 21:38 Maciej S. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |