Re: [PATCH v5 4/8] KVM: VMX: dump_vmcs() reports tertiary_exec_control field as well

[Zeng Guang]

On 1/14/2022 5:03 AM, Sean Christopherson wrote:
> On Fri, Dec 31, 2021, Zeng Guang wrote:
> > From: Robert Hoo <robert.hu@xxxxxxxxxxxxxxx>
> >
> > Add tertiary_exec_control field report in dump_vmcs()
> >
> > Signed-off-by: Robert Hoo <robert.hu@xxxxxxxxxxxxxxx>
> > Signed-off-by: Zeng Guang <guang.zeng@xxxxxxxxx>
> > ---
> >   arch/x86/kvm/vmx/vmx.c | 9 +++++++--
> >   1 file changed, 7 insertions(+), 2 deletions(-)
> >
> > diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
> > index fb0f600368c6..5716db9704c0 100644
> > --- a/arch/x86/kvm/vmx/vmx.c
> > +++ b/arch/x86/kvm/vmx/vmx.c
> > @@ -5729,6 +5729,7 @@ void dump_vmcs(struct kvm_vcpu *vcpu)
> >   	struct vcpu_vmx *vmx = to_vmx(vcpu);
> >   	u32 vmentry_ctl, vmexit_ctl;
> >   	u32 cpu_based_exec_ctrl, pin_based_exec_ctrl, secondary_exec_control;
> > +	u64 tertiary_exec_control = 0;
> >   	unsigned long cr4;
> >   	int efer_slot;
> >   
> > @@ -5746,6 +5747,9 @@ void dump_vmcs(struct kvm_vcpu *vcpu)
> >   	if (cpu_has_secondary_exec_ctrls())
> >   		secondary_exec_control = vmcs_read32(SECONDARY_VM_EXEC_CONTROL);
> Gah, this (not your) code is silly.  I had to go look at the full source to see
> that secondary_exec_control isn't accessed uninitialized...
>
> Can you opportunistically tweak it to the below, and use the same patter for the
> tertiary controls?
>
> 	if (cpu_has_secondary_exec_ctrls())
> 		secondary_exec_control = vmcs_read32(SECONDARY_VM_EXEC_CONTROL);
> 	else
> 		secondary_exec_control = 0;
Actually secondary_exec_control did zero initialization ahead .
yah, it's better to unify the code for both.
> >   
> > +	if (cpu_has_tertiary_exec_ctrls())
> > +		tertiary_exec_control = vmcs_read64(TERTIARY_VM_EXEC_CONTROL);
> > +
> >   	pr_err("VMCS %p, last attempted VM-entry on CPU %d\n",
> >   	       vmx->loaded_vmcs->vmcs, vcpu->arch.last_vmentry_cpu);
> >   	pr_err("*** Guest State ***\n");
> > @@ -5844,8 +5848,9 @@ void dump_vmcs(struct kvm_vcpu *vcpu)
> >   		vmx_dump_msrs("host autoload", &vmx->msr_autoload.host);
> >   
> >   	pr_err("*** Control State ***\n");
> > -	pr_err("PinBased=%08x CPUBased=%08x SecondaryExec=%08x\n",
> > -	       pin_based_exec_ctrl, cpu_based_exec_ctrl, secondary_exec_control);
> > +	pr_err("PinBased=0x%08x CPUBased=0x%08x SecondaryExec=0x%08x TertiaryExec=0x%016llx\n",
> > +	       pin_based_exec_ctrl, cpu_based_exec_ctrl, secondary_exec_control,
> > +	       tertiary_exec_control);
> Can you provide a sample dump?  It's hard to visualize the output, e.g. I'm worried
> this will be overly log and harder to read than putting tertiary controls on their
> own line.

Sample dump here.
 
*** Control State ***

 PinBased=0x000000ff CPUBased=0xb5a26dfa SecondaryExec=0x061037eb 
TertiaryExec=0x0000000000000010
 EntryControls=0000d1ff ExitControls=002befff
 ExceptionBitmap=00060042 PFECmask=00000000 PFECmatch=00000000
 VMEntry: intr_info=00000000 errcode=00000000 ilen=00000000
 VMExit: intr_info=00000000 errcode=00000000 ilen=00000003
         reason=00000030 qualification=0000000000000784
> >   	pr_err("EntryControls=%08x ExitControls=%08x\n", vmentry_ctl, vmexit_ctl);
> >   	pr_err("ExceptionBitmap=%08x PFECmask=%08x PFECmatch=%08x\n",
> >   	       vmcs_read32(EXCEPTION_BITMAP),
> > --
> > 2.27.0