| Message ID | 20200820133339.372823-3-mlevitsk@redhat.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | KVM: nSVM: ondemand nested state allocation + smm fixes | expand |
On Thu, Aug 20, 2020 at 6:33 AM Maxim Levitsky <mlevitsk@redhat.com> wrote: > > No functional changes. > > Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com> > --- > arch/x86/kvm/svm/nested.c | 10 +++++----- > arch/x86/kvm/svm/svm.c | 13 +++++++------ > arch/x86/kvm/svm/svm.h | 2 +- > 3 files changed, 13 insertions(+), 12 deletions(-) > > diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c > index fb68467e6049..f5b17920a2ca 100644 > --- a/arch/x86/kvm/svm/nested.c > +++ b/arch/x86/kvm/svm/nested.c > @@ -431,7 +431,7 @@ int enter_svm_guest_mode(struct vcpu_svm *svm, u64 vmcb_gpa, For consistency, should the vmcb_gpa argument be renamed to vmcb12_gpa as well? > @@ -579,7 +579,7 @@ int nested_svm_vmexit(struct vcpu_svm *svm) > > /* Exit Guest-Mode */ > leave_guest_mode(&svm->vcpu); > - svm->nested.vmcb = 0; > + svm->nested.vmcb12_gpa = 0; Perhaps in a follow-up change, this could be set to an illegal value rather than 0? > @@ -1018,7 +1018,7 @@ static int svm_get_nested_state(struct kvm_vcpu *vcpu, > > /* First fill in the header and copy it out. */ > if (is_guest_mode(vcpu)) { > - kvm_state.hdr.svm.vmcb_pa = svm->nested.vmcb; > + kvm_state.hdr.svm.vmcb_pa = svm->nested.vmcb12_gpa; It's unfortunate that we have "_pa" on the LHS on "_gpa" on the RHS. Oh, well. > diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c > index 562a79e3e63a..d33013b9b4d7 100644 > --- a/arch/x86/kvm/svm/svm.c > +++ b/arch/x86/kvm/svm/svm.c > @@ -1102,7 +1102,7 @@ static void init_vmcb(struct vcpu_svm *svm) > } > svm->asid_generation = 0; > > - svm->nested.vmcb = 0; > + svm->nested.vmcb12_gpa = 0; Here, too, perhaps this could be changed from 0 to an illegal value in a follow-up change. Reviewed-by: Jim Mattson <jmattson@google.com>
On Thu, 2020-08-20 at 14:00 -0700, Jim Mattson wrote: > On Thu, Aug 20, 2020 at 6:33 AM Maxim Levitsky <mlevitsk@redhat.com> wrote: > > No functional changes. > > > > Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com> > > --- > > arch/x86/kvm/svm/nested.c | 10 +++++----- > > arch/x86/kvm/svm/svm.c | 13 +++++++------ > > arch/x86/kvm/svm/svm.h | 2 +- > > 3 files changed, 13 insertions(+), 12 deletions(-) > > > > diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c > > index fb68467e6049..f5b17920a2ca 100644 > > --- a/arch/x86/kvm/svm/nested.c > > +++ b/arch/x86/kvm/svm/nested.c > > @@ -431,7 +431,7 @@ int enter_svm_guest_mode(struct vcpu_svm *svm, u64 vmcb_gpa, > For consistency, should the vmcb_gpa argument be renamed to vmcb12_gpa as well? I went over all nested.c and renamed all mentions of vmcb which refer to guest's vmcb to vmcb12, and mentions of nested_vmcb to vmcb12 as well. I hope I didn't made this patch too much larger. I updated the patch subject too. > > > > @@ -579,7 +579,7 @@ int nested_svm_vmexit(struct vcpu_svm *svm) > > > > /* Exit Guest-Mode */ > > leave_guest_mode(&svm->vcpu); > > - svm->nested.vmcb = 0; > > + svm->nested.vmcb12_gpa = 0; > Perhaps in a follow-up change, this could be set to an illegal value > rather than 0? Or rather not reset this address at all, as I did later in the caching pathes which I dropped for now. > > > > @@ -1018,7 +1018,7 @@ static int svm_get_nested_state(struct kvm_vcpu *vcpu, > > > > /* First fill in the header and copy it out. */ > > if (is_guest_mode(vcpu)) { > > - kvm_state.hdr.svm.vmcb_pa = svm->nested.vmcb; > > + kvm_state.hdr.svm.vmcb_pa = svm->nested.vmcb12_gpa; > It's unfortunate that we have "_pa" on the LHS on "_gpa" on the RHS. Oh, well. I was afraid to touch this struct since it is user visible. I noticed it. > > > > diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c > > index 562a79e3e63a..d33013b9b4d7 100644 > > --- a/arch/x86/kvm/svm/svm.c > > +++ b/arch/x86/kvm/svm/svm.c > > @@ -1102,7 +1102,7 @@ static void init_vmcb(struct vcpu_svm *svm) > > } > > svm->asid_generation = 0; > > > > - svm->nested.vmcb = 0; > > + svm->nested.vmcb12_gpa = 0; > Here, too, perhaps this could be changed from 0 to an illegal value in > a follow-up change. > > Reviewed-by: Jim Mattson <jmattson@google.com> > Thanks for the review, Best regards, Maxim Levitsky
diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c index fb68467e6049..f5b17920a2ca 100644 --- a/arch/x86/kvm/svm/nested.c +++ b/arch/x86/kvm/svm/nested.c @@ -431,7 +431,7 @@ int enter_svm_guest_mode(struct vcpu_svm *svm, u64 vmcb_gpa, { int ret; - svm->nested.vmcb = vmcb_gpa; + svm->nested.vmcb12_gpa = vmcb_gpa; load_nested_vmcb_control(svm, &nested_vmcb->control); nested_prepare_vmcb_save(svm, nested_vmcb); nested_prepare_vmcb_control(svm); @@ -568,7 +568,7 @@ int nested_svm_vmexit(struct vcpu_svm *svm) struct vmcb *vmcb = svm->vmcb; struct kvm_host_map map; - rc = kvm_vcpu_map(&svm->vcpu, gpa_to_gfn(svm->nested.vmcb), &map); + rc = kvm_vcpu_map(&svm->vcpu, gpa_to_gfn(svm->nested.vmcb12_gpa), &map); if (rc) { if (rc == -EINVAL) kvm_inject_gp(&svm->vcpu, 0); @@ -579,7 +579,7 @@ int nested_svm_vmexit(struct vcpu_svm *svm) /* Exit Guest-Mode */ leave_guest_mode(&svm->vcpu); - svm->nested.vmcb = 0; + svm->nested.vmcb12_gpa = 0; WARN_ON_ONCE(svm->nested.nested_run_pending); /* in case we halted in L2 */ @@ -1018,7 +1018,7 @@ static int svm_get_nested_state(struct kvm_vcpu *vcpu, /* First fill in the header and copy it out. */ if (is_guest_mode(vcpu)) { - kvm_state.hdr.svm.vmcb_pa = svm->nested.vmcb; + kvm_state.hdr.svm.vmcb_pa = svm->nested.vmcb12_gpa; kvm_state.size += KVM_STATE_NESTED_SVM_VMCB_SIZE; kvm_state.flags |= KVM_STATE_NESTED_GUEST_MODE; @@ -1128,7 +1128,7 @@ static int svm_set_nested_state(struct kvm_vcpu *vcpu, copy_vmcb_control_area(&hsave->control, &svm->vmcb->control); hsave->save = save; - svm->nested.vmcb = kvm_state->hdr.svm.vmcb_pa; + svm->nested.vmcb12_gpa = kvm_state->hdr.svm.vmcb_pa; load_nested_vmcb_control(svm, &ctl); nested_prepare_vmcb_control(svm); diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c index 562a79e3e63a..d33013b9b4d7 100644 --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -1102,7 +1102,7 @@ static void init_vmcb(struct vcpu_svm *svm) } svm->asid_generation = 0; - svm->nested.vmcb = 0; + svm->nested.vmcb12_gpa = 0; svm->vcpu.arch.hflags = 0; if (!kvm_pause_in_guest(svm->vcpu.kvm)) { @@ -3884,7 +3884,7 @@ static int svm_pre_enter_smm(struct kvm_vcpu *vcpu, char *smstate) /* FED8h - SVM Guest */ put_smstate(u64, smstate, 0x7ed8, 1); /* FEE0h - SVM Guest VMCB Physical Address */ - put_smstate(u64, smstate, 0x7ee0, svm->nested.vmcb); + put_smstate(u64, smstate, 0x7ee0, svm->nested.vmcb12_gpa); svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX]; svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP]; @@ -3903,17 +3903,18 @@ static int svm_pre_leave_smm(struct kvm_vcpu *vcpu, const char *smstate) struct vmcb *nested_vmcb; struct kvm_host_map map; u64 guest; - u64 vmcb; + u64 vmcb12_gpa; int ret = 0; guest = GET_SMSTATE(u64, smstate, 0x7ed8); - vmcb = GET_SMSTATE(u64, smstate, 0x7ee0); + vmcb12_gpa = GET_SMSTATE(u64, smstate, 0x7ee0); if (guest) { - if (kvm_vcpu_map(&svm->vcpu, gpa_to_gfn(vmcb), &map) == -EINVAL) + if (kvm_vcpu_map(&svm->vcpu, gpa_to_gfn(vmcb12_gpa), &map) == -EINVAL) return 1; + nested_vmcb = map.hva; - ret = enter_svm_guest_mode(svm, vmcb, nested_vmcb); + ret = enter_svm_guest_mode(svm, vmcb12_gpa, nested_vmcb); kvm_vcpu_unmap(&svm->vcpu, &map, true); } diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h index a798e1731709..ab913468f9cb 100644 --- a/arch/x86/kvm/svm/svm.h +++ b/arch/x86/kvm/svm/svm.h @@ -85,7 +85,7 @@ struct svm_nested_state { struct vmcb *hsave; u64 hsave_msr; u64 vm_cr_msr; - u64 vmcb; + u64 vmcb12_gpa; u32 host_intercept_exceptions; /* These are the merged vectors */
No functional changes. Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com> --- arch/x86/kvm/svm/nested.c | 10 +++++----- arch/x86/kvm/svm/svm.c | 13 +++++++------ arch/x86/kvm/svm/svm.h | 2 +- 3 files changed, 13 insertions(+), 12 deletions(-)