Received-SPF: pass (google.com: domain of linux-kernel+bounces-110517-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Date: Thu, 21 Mar 2024 10:46:31 -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 040/130] KVM: TDX: Make pmu_intel.c ignore guest TD
 case
Message-ID: <20240321174631.GN1994522@ls.amr.corp.intel.com>
References: <cover.1708933498.git.isaku.yamahata@intel.com>
 <2eb2e7205d23ec858e6d800fee7da2306216e8f0.1708933498.git.isaku.yamahata@intel.com>
 <ZfqJ3LkYrwR/qpsX@chao-email>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <ZfqJ3LkYrwR/qpsX@chao-email>

On Wed, Mar 20, 2024 at 03:01:48PM +0800,
Chao Gao <chao.gao@intel.com> wrote:

> On Mon, Feb 26, 2024 at 12:25:42AM -0800, isaku.yamahata@intel.com wrote:
> >From: Isaku Yamahata <isaku.yamahata@intel.com>
> >
> >Because TDX KVM doesn't support PMU yet (it's future work of TDX KVM
> >support as another patch series) and pmu_intel.c touches vmx specific
> >structure in vcpu initialization, as workaround add dummy structure to
> >struct vcpu_tdx and pmu_intel.c can ignore TDX case.
> 
> Can we instead factor pmu_intel.c to avoid corrupting memory? how hard would it
> be?

Do you mean sprinkling "if (tdx) return"? It's easy. Just add it to all hooks
in kvm_pmu_ops.

I chose this approach because we'll soon support vPMU support. For simplicity,
will switch to sprinkle "if (tdx) return".

> >+bool intel_pmu_lbr_is_enabled(struct kvm_vcpu *vcpu)
> >+{
> >+	struct x86_pmu_lbr *lbr = vcpu_to_lbr_records(vcpu);
> >+
> >+	if (is_td_vcpu(vcpu))
> >+		return false;
> >+
> >+	return lbr->nr && (vcpu_get_perf_capabilities(vcpu) & PMU_CAP_LBR_FMT);
> 
> The check about vcpu's perf capabilities is new. is it necessary?

No. Will delete it. It crept in during rebase.
-- 
Isaku Yamahata <isaku.yamahata@intel.com>