Received-SPF: pass (google.com: domain of linux-kernel+bounces-110636-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Date: Thu, 21 Mar 2024 13:21:50 -0700
From: Isaku Yamahata <isaku.yamahata@intel.com>
To: Chao Gao <chao.gao@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 043/130] KVM: TDX: create/free TDX vcpu structure
Message-ID: <20240321202150.GP1994522@ls.amr.corp.intel.com>
References: <cover.1708933498.git.isaku.yamahata@intel.com>
 <51c4203e844159451f5a78fb18cc5bebcc38a76e.1708933498.git.isaku.yamahata@intel.com>
 <ZfuNpI1fiUr4h27+@chao-email>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <ZfuNpI1fiUr4h27+@chao-email>

On Thu, Mar 21, 2024 at 09:30:12AM +0800,
Chao Gao <chao.gao@intel.com> wrote:

> >+int tdx_vcpu_create(struct kvm_vcpu *vcpu)
> >+{
> >+	struct kvm_tdx *kvm_tdx = to_kvm_tdx(vcpu->kvm);
> >+
> >+	WARN_ON_ONCE(vcpu->arch.cpuid_entries);
> >+	WARN_ON_ONCE(vcpu->arch.cpuid_nent);
> >+
> >+	/* TDX only supports x2APIC, which requires an in-kernel local APIC. */
> 
> Cannot QEMU emulate x2APIC? In my understanding, the reason is TDX module always
> enables APICv for TDs. So, KVM cannot intercept every access to APIC and forward
> them to QEMU for emulation.

You're right. Let me update it as follows.

  /*
   * TDX module always enables APICv for TDs. So, KVM cannot intercept every
   * access to APIC and forward them to user space VMM.
   */


> >+	if (!vcpu->arch.apic)
> 
> will "if (!irqchip_in_kernel(vcpu->kvm))" work? looks this is the custome for such
> a check.


It should work because kvm_arch_vcpu_create().  Will update it.
-- 
Isaku Yamahata <isaku.yamahata@intel.com>