Received-SPF: pass (google.com: domain of linux-kernel+bounces-110713-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
Date: Thu, 21 Mar 2024 14:59:55 -0700
From: Isaku Yamahata <isaku.yamahata@intel.com>
To: "Edgecombe, Rick P" <rick.p.edgecombe@intel.com>
Cc: "kvm@vger.kernel.org" <kvm@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"Yamahata, Isaku" <isaku.yamahata@intel.com>,
	"Zhang, Tina" <tina.zhang@intel.com>,
	"seanjc@google.com" <seanjc@google.com>,
	"Yuan, Hang" <hang.yuan@intel.com>,
	"Huang, Kai" <kai.huang@intel.com>, "Chen, Bo2" <chen.bo@intel.com>,
	"sagis@google.com" <sagis@google.com>,
	"isaku.yamahata@gmail.com" <isaku.yamahata@gmail.com>,
	"Aktas, Erdem" <erdemaktas@google.com>,
	"pbonzini@redhat.com" <pbonzini@redhat.com>,
	isaku.yamahata@linux.intel.com
Subject: Re: [PATCH v19 057/130] KVM: x86/mmu: Add a new is_private member
 for union kvm_mmu_page_role
Message-ID: <20240321215955.GS1994522@ls.amr.corp.intel.com>
References: <cover.1708933498.git.isaku.yamahata@intel.com>
 <3692118f525c21cbe3670e1c20b1bbbf3be2b4ba.1708933498.git.isaku.yamahata@intel.com>
 <875eb07773836bf1d5668a4f28a696869e3291c2.camel@intel.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <875eb07773836bf1d5668a4f28a696869e3291c2.camel@intel.com>

On Thu, Mar 21, 2024 at 12:18:47AM +0000,
"Edgecombe, Rick P" <rick.p.edgecombe@intel.com> wrote:

> On Mon, 2024-02-26 at 00:25 -0800, isaku.yamahata@intel.com wrote:
> > From: Isaku Yamahata <isaku.yamahata@intel.com>
> > 
> > Because TDX support introduces private mapping, add a new member in
> > union
> > kvm_mmu_page_role with access functions to check the member.
> 
> I guess we should have a role bit for private like in this patch, but
> just barely. AFAICT we have a gfn and struct kvm in every place where
> it is checked (assuming my proposal in patch 56 holds water). So we
> could have
> bool is_private = !(gfn & kvm_gfn_shared_mask(kvm));

Yes, we can use such combination. or !!sp->private_spt or something.
Originally we didn't use role.is_private and passed around private parameter.


> But there are extra bits available in the role, so we can skip the
> extra step. Can you think of any more reasons? I want to try to write a
> log for this one. It's very short.

There are several places to compare role and shared<->private.  For example,
kvm_tdp_mmu_alloc_root(). role.is_private simplifies such comparison.
-- 
Isaku Yamahata <isaku.yamahata@intel.com>