linux-kernel - Re: [PATCH v18 032/121] KVM: x86/mmu: introduce config for PRIVATE KVM MMU

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Mon, 12 Feb 2024 18:57:49 -0800
From: Sean Christopherson <seanjc@...gle.com>
To: Paolo Bonzini <pbonzini@...hat.com>
Cc: isaku.yamahata@...el.com, kvm@...r.kernel.org, 
	linux-kernel@...r.kernel.org, isaku.yamahata@...il.com, erdemaktas@...gle.com, 
	Sagi Shahar <sagis@...gle.com>, Kai Huang <kai.huang@...el.com>, chen.bo@...el.com, 
	hang.yuan@...el.com, tina.zhang@...el.com
Subject: Re: [PATCH v18 032/121] KVM: x86/mmu: introduce config for PRIVATE
 KVM MMU

On Mon, Feb 12, 2024, Paolo Bonzini wrote:
> On Tue, Jan 23, 2024 at 12:55 AM <isaku.yamahata@...el.com> wrote:
> >
> > From: Isaku Yamahata <isaku.yamahata@...el.com>
> >
> > To keep the case of non TDX intact, introduce a new config option for
> > private KVM MMU support.  At the moment, this is synonym for
> > CONFIG_INTEL_TDX_HOST && CONFIG_KVM_INTEL.  The config makes it clear
> > that the config is only for x86 KVM MMU.
> 
> Better, just put this as:
> 
> config KVM_MMU_PRIVATE
>     bool
> 
> but also add a reverse dependency to KVM_INTEL:
> 
> config KVM_INTEL
>      tristate "KVM for Intel (and compatible) processors support"
>      depends on KVM && IA32_FEAT_CTL
>      select KVM_MMU_PRIVATE if INTEL_TDX_HOST
>      ...
> 
> This matches the usage of kvm-intel-$(INTEL_TDX_HOST) in the Makefile.

But why even bother with a Kconfig in the first place?  Saving the is_private
bit in the role adds no value whatsoever.  In fact, it's probably a big net
negative because it necessitates this ugly code:

	if (private)
		kvm_mmu_page_role_set_private(&role);

which really should just be

	role.private = private;

Ditto for kvm_mmu_page.private_spt.

The only thing that even so much as approaches being a hot path is
kvm_gfn_shared_mask(), and if that needs to be optimized, then we'd probably be
better off with a static_key, a la kvm_has_noapic_vcpu (though I'm *extremely*
skeptical that that adds any measurable benefit).