kernel/git/stable/linux.git/arch/x86/kvm/vmx.c, branch master

KVM: VMX: Move VMX specific files to a "vmx" subdirectory

2018-12-14T11:34:24+00:00

...to prepare for shattering vmx.c into multiple files without having to prepend "vmx_" to all new files. Signed-off-by: Sean Christopherson Signed-off-by: Paolo Bonzini

KVM: VMX: Alphabetize the includes in vmx.c

2018-12-14T11:34:21+00:00

...to prepare for the creation of a "vmx" subdirectory that will contain a variety of headers. Clean things up now to avoid making a bigger mess in the future. Signed-off-by: Sean Christopherson Signed-off-by: Paolo Bonzini

KVM: nVMX: Allocate and configure VM{READ,WRITE} bitmaps iff enable_shadow_vmcs

2018-12-14T11:34:21+00:00

...and make enable_shadow_vmcs depend on nested. Aside from the obvious memory savings, this will allow moving the relevant code out of vmx.c in the future, e.g. to a nested specific file. Signed-off-by: Sean Christopherson Signed-off-by: Paolo Bonzini

KVM: nVMX: Free the VMREAD/VMWRITE bitmaps if alloc_kvm_area() fails

2018-12-14T11:34:20+00:00

Fixes: 34a1cd60d17f ("kvm: x86: vmx: move some vmx setting from vmx_init() to hardware_setup()") Cc: stable@vger.kernel.org Signed-off-by: Sean Christopherson Signed-off-by: Paolo Bonzini

kvm: nVMX: Set VM instruction error for VMPTRLD of unbacked page

2018-11-27T11:55:46+00:00

It is never correct for a VMX instruction to fail with "invalid VMCS" if there is, in fact, a current VMCS. Reads from unbacked addresses return all 1's, which means that an unbacked VMCS will not have the correct VMCS revision ID (i.e. VMCS12_REVISION). Fixes: 63846663eac78 ("KVM: nVMX: Implement VMPTRLD") Signed-off-by: Jim Mattson Signed-off-by: Paolo Bonzini

nVMX x86: Check VMX-preemption timer controls on vmentry of L2 guests

2018-11-27T11:55:46+00:00

According to section "Checks on VMX Controls" in Intel SDM vol 3C, the following check needs to be enforced on vmentry of L2 guests: If the "activate VMX-preemption timer" VM-execution control is 0, the the "save VMX-preemption timer value" VM-exit control must also be 0. Signed-off-by: Krish Sadhukhan Reviewed-by: Mihai Carabas Reviewed-by: Liran Alon Signed-off-by: Paolo Bonzini

KVM/nVMX: Remove unneeded forward jump in nested_vmx_check_vmentry_hw asm

2018-11-27T11:53:45+00:00

There is no need to jump just after the jump insn itself. Also, make code similar to entering guest mode in vmx_vcpu_run. Signed-off-by: Uros Bizjak Reviewed-by: Jim Mattson Signed-off-by: Paolo Bonzini

KVM: nVMX: Unrestricted guest mode requires EPT

2018-11-27T11:53:45+00:00

As specified in Intel's SDM, do not allow the L1 hypervisor to launch an L2 guest with the VM-execution controls for "unrestricted guest" or "mode-based execute control for EPT" set and the VM-execution control for "enable EPT" clear. Note that the VM-execution control for "mode-based execute control for EPT" is not yet virtualized by kvm. Reported-by: Andrew Thornton Signed-off-by: Jim Mattson Reviewed-by: Peter Shier Reviewed-by: Sean Christopherson Reviewed-by: Wanpeng Li Reviewed-by: Liran Alon Signed-off-by: Paolo Bonzini

KVM: x86: Trace changes to active TSC offset regardless if vCPU in guest-mode

2018-11-27T11:53:43+00:00

For some reason, kvm_x86_ops->write_l1_tsc_offset() skipped trace of change to active TSC offset in case vCPU is in guest-mode. This patch changes write_l1_tsc_offset() behavior to trace any change to active TSC offset to aid debugging. The VMX code is changed to look more similar to SVM, which is in my opinion nicer. Based on a patch by Liran Alon. Signed-off-by: Paolo Bonzini

KVM: nVMX: vmcs12 revision_id is always VMCS12_REVISION even when copied from eVMCS

2018-11-27T11:50:30+00:00

vmcs12 represents the per-CPU cache of L1 active vmcs12. This cache can be loaded by one of the following: 1) Guest making a vmcs12 active by exeucting VMPTRLD 2) Guest specifying eVMCS in VP assist page and executing VMLAUNCH/VMRESUME. Either way, vmcs12 should have revision_id of VMCS12_REVISION. Which is not equal to eVMCS revision_id which specifies used VersionNumber of eVMCS struct (e.g. KVM_EVMCS_VERSION). Specifically, this causes an issue in restoring a nested VM state because vmx_set_nested_state() verifies that vmcs12->revision_id is equal to VMCS12_REVISION which was not true in case vmcs12 was populated from an eVMCS by vmx_get_nested_state() which calls copy_enlightened_to_vmcs12(). Reviewed-by: Darren Kenny Signed-off-by: Liran Alon Reviewed-by: Vitaly Kuznetsov Signed-off-by: Paolo Bonzini