Re: [PATCH 1/7] hyperv: Detect Nested virtualization support for SVM

From: Vineeth Pillai <viremana@linux.microsoft.com>
To: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
	"K. Y. Srinivasan" <kys@microsoft.com>,
	x86@kernel.org, kvm@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-hyperv@vger.kernel.org,
	Lan Tianyu <Tianyu.Lan@microsoft.com>,
	Michael Kelley <mikelley@microsoft.com>,
	Paolo Bonzini <pbonzini@redhat.com>,
	Sean Christopherson <seanjc@google.com>,
	Wanpeng Li <wanpengli@tencent.com>,
	Jim Mattson <jmattson@google.com>, Joerg Roedel <joro@8bytes.org>,
	Wei Liu <wei.liu@kernel.org>,
	Stephen Hemminger <sthemmin@microsoft.com>,
	Haiyang Zhang <haiyangz@microsoft.com>
Subject: Re: [PATCH 1/7] hyperv: Detect Nested virtualization support for SVM
Date: Thu, 8 Apr 2021 09:15:47 -0400	[thread overview]
Message-ID: <af87c25e-78c6-5859-e1c1-2aa07d087a25@linux.microsoft.com> (raw)
In-Reply-To: <87lf9tavci.fsf@vitty.brq.redhat.com>

Hi Vitaly,

On 4/8/21 7:06 AM, Vitaly Kuznetsov wrote:
> -	if (ms_hyperv.hints & HV_X64_ENLIGHTENED_VMCS_RECOMMENDED) {
> +	/*
> +	 * AMD does not need enlightened VMCS as VMCB is already a
> +	 * datastructure in memory.
> Well, VMCS is also a structure in memory, isn't it? It's just that we
> don't have a 'clean field' concept for it and we can't use normal memory
> accesses.

Yes, you are right. I was referring to the fact that we cant use normal
memory accesses, but is a bit mis-worded.

>
>> 	We need to get the nested
>> +	 * features if SVM is enabled.
>> +	 */
>> +	if (boot_cpu_has(X86_FEATURE_SVM) ||
>> +	    ms_hyperv.hints & HV_X64_ENLIGHTENED_VMCS_RECOMMENDED) {
> Do I understand correctly that we can just look at CPUID.0x40000000.EAX
> and in case it is >= 0x4000000A we can read HYPERV_CPUID_NESTED_FEATURES
> leaf? I'd suggest we do that intead then.
I agree, that is a better way to consolidate both the cases.
Will change it in the next iteration. Probably the above code
comment is not needed when we consolidate the cases here.

Thanks,
Vineeth