Received-SPF: pass (google.com: domain of linux-kernel+bounces-110498-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Date: Thu, 21 Mar 2024 10:30:24 -0700
From: Isaku Yamahata <isaku.yamahata@intel.com>
To: Xiaoyao Li <xiaoyao.li@intel.com>
Cc: isaku.yamahata@intel.com, kvm@vger.kernel.org,
	linux-kernel@vger.kernel.org, isaku.yamahata@gmail.com,
	Paolo Bonzini <pbonzini@redhat.com>, erdemaktas@google.com,
	Sean Christopherson <seanjc@google.com>,
	Sagi Shahar <sagis@google.com>, Kai Huang <kai.huang@intel.com>,
	chen.bo@intel.com, hang.yuan@intel.com, tina.zhang@intel.com,
	isaku.yamahata@linux.intel.com
Subject: Re: [PATCH v19 039/130] KVM: TDX: initialize VM with TDX specific
 parameters
Message-ID: <20240321173024.GM1994522@ls.amr.corp.intel.com>
References: <cover.1708933498.git.isaku.yamahata@intel.com>
 <5eca97e6a3978cf4dcf1cff21be6ec8b639a66b9.1708933498.git.isaku.yamahata@intel.com>
 <5702443b-510e-4ce5-823f-999582a6aced@intel.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <5702443b-510e-4ce5-823f-999582a6aced@intel.com>

On Wed, Mar 20, 2024 at 04:15:23PM +0800,
Xiaoyao Li <xiaoyao.li@intel.com> wrote:

> On 2/26/2024 4:25 PM, isaku.yamahata@intel.com wrote:
> 
> ...
> 
> > +static int setup_tdparams_xfam(struct kvm_cpuid2 *cpuid, struct td_params *td_params)
> > +{
> > +	const struct kvm_cpuid_entry2 *entry;
> > +	u64 guest_supported_xcr0;
> > +	u64 guest_supported_xss;
> > +
> > +	/* Setup td_params.xfam */
> > +	entry = kvm_find_cpuid_entry2(cpuid->entries, cpuid->nent, 0xd, 0);
> > +	if (entry)
> > +		guest_supported_xcr0 = (entry->eax | ((u64)entry->edx << 32));
> > +	else
> > +		guest_supported_xcr0 = 0;
> > +	guest_supported_xcr0 &= kvm_caps.supported_xcr0;
> > +
> > +	entry = kvm_find_cpuid_entry2(cpuid->entries, cpuid->nent, 0xd, 1);
> > +	if (entry)
> > +		guest_supported_xss = (entry->ecx | ((u64)entry->edx << 32));
> > +	else
> > +		guest_supported_xss = 0;
> > +
> > +	/*
> > +	 * PT and CET can be exposed to TD guest regardless of KVM's XSS, PT
> > +	 * and, CET support.
> > +	 */
> > +	guest_supported_xss &=
> > +		(kvm_caps.supported_xss | XFEATURE_MASK_PT | TDX_TD_XFAM_CET);
> > +
> > +	td_params->xfam = guest_supported_xcr0 | guest_supported_xss;
> > +	if (td_params->xfam & XFEATURE_MASK_LBR) {
> > +		/*
> > +		 * TODO: once KVM supports LBR(save/restore LBR related
> > +		 * registers around TDENTER), remove this guard.
> > +		 */
> > +#define MSG_LBR	"TD doesn't support LBR yet. KVM needs to save/restore IA32_LBR_DEPTH properly.\n"
> > +		pr_warn(MSG_LBR);
> > +		return -EOPNOTSUPP;
> 
> This unsupported behavior is totally decided by KVM even if TDX module
> supports it. I think we need to reflect it in tdx_info->xfam_fixed0, which
> gets reported to userspace via KVM_TDX_CAPABILITIES. So userspace will aware
> that LBR is not supported for TDs.

Yes, we can suppress KVM unpported features. I replied at
https://lore.kernel.org/kvm/20240321155513.GL1994522@ls.amr.corp.intel.com/

So far we used KVM_TDX_CAPABILITIES for feature enumeration. I'm wondering about
KVM_GET_DEVICE_ATTR [1].  It's future extensible. It's also consistent with SEV.

[1] https://lore.kernel.org/r/20240226190344.787149-7-pbonzini@redhat.com
-- 
Isaku Yamahata <isaku.yamahata@intel.com>