kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sean Christopherson <seanjc@google.com>
To: Binbin Wu <binbin.wu@linux.intel.com>
Cc: pbonzini@redhat.com, kvm@vger.kernel.org,
	rick.p.edgecombe@intel.com,  kai.huang@intel.com,
	adrian.hunter@intel.com, reinette.chatre@intel.com,
	 xiaoyao.li@intel.com, tony.lindgren@linux.intel.com,
	isaku.yamahata@intel.com,  yan.y.zhao@intel.com,
	chao.gao@intel.com, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 12/16] KVM: TDX: Inhibit APICv for TDX guest
Date: Mon, 13 Jan 2025 09:16:37 -0800	[thread overview]
Message-ID: <Z4VKdbW1R0AoLvkB@google.com> (raw)
In-Reply-To: <e3a2e8fa-b496-4010-9a8c-bfeb131bc43b@linux.intel.com>

On Mon, Jan 13, 2025, Binbin Wu wrote:
> On 1/13/2025 10:03 AM, Binbin Wu wrote:
> > 
> > On 12/9/2024 9:07 AM, Binbin Wu wrote:
> > > From: Isaku Yamahata <isaku.yamahata@intel.com>
> > > 
> > > Inhibit APICv for TDX guest in KVM since TDX doesn't support APICv accesses
> > > from host VMM.
> > > 
> > > Follow how SEV inhibits APICv.  I.e, define a new inhibit reason for TDX, set
> > > it on TD initialization, and add the flag to kvm_x86_ops.required_apicv_inhibits.
> > 
> Resend due to the format mess.

That was a very impressive mess :-)

> After TDX vCPU init, APIC is set to x2APIC mode. However, userspace could
> disable APIC via KVM_SET_LAPIC or KVM_SET_{SREGS, SREGS2}.
> 
> - KVM_SET_LAPIC
>   Currently, KVM allows userspace to request KVM_SET_LAPIC to set the state
>   of LAPIC for TDX guests.
>   There are two options:
>   - Force x2APIC mode and default base address when userspace request
>     KVM_SET_LAPIC.
>   - Simply reject KVM_SET_LAPIC for TDX guest (apic->guest_apic_protected
>     is true), since migration is not supported yet.
>   Choose option 2 for simplicity for now.

Yeah.  We'll likely need to support KVM_SET_LAPIC at some point, e.g. to support
PID.PIR save/restore, but that's definitely a future problem.

> Summary about APICv inhibit reasons:
> APICv could still be disabled runtime in some corner case, e.g,
> APICV_INHIBIT_REASON_PHYSICAL_ID_ALIASED due to memory allocation failure.
> After checking enable_apicv in tdx_bringup(), apic->apicv_active is
> initialized as true in kvm_create_lapic().  If APICv is inhibited due to any
> reason runtime, the refresh_apicv_exec_ctrl() callback could be used to check
> if APICv is disabled for TDX, if APICv is disabled, bug the VM.

I _think_ this is a non-issue, and that KVM could do KVM_BUG_ON() if APICv is
inihibited by kvm_recalculate_apic_map() for a TDX VM.  x2APIC is mandatory
(KVM_APIC_MODE_MAP_DISABLED and "APIC_ID modified" impossible), KVM emulates
APIC_ID as read-only for x2APIC mode (physical aliasing impossible), and LDR is
read-only for x2APIC (logical aliasing impossible).

To ensure no physical aliasing, KVM would need to require KVM_CAP_X2APIC_API be
enabled, but that should probably be required for TDX no matter what.

> kvm_arch_dy_has_pending_interrupt()
> -----------------------------------
> Before enabling off-TD debug, there is no functional change because there
> is no PAUSE Exit for TDX guests.
> After enabling off-TD debug, the kvm_vcpu_apicv_active(vcpu) should be true
> to get the pending interrupt from PID. Set APICv to active for TDX is the
> right thing to do.

And as alluded to above, for save/restore, e.g. intrahost migration.

  reply	other threads:[~2025-01-13 17:16 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-12-09  1:07 [PATCH 00/16] KVM: TDX: TDX interrupts Binbin Wu
2024-12-09  1:07 ` [PATCH 01/16] KVM: TDX: Add support for find pending IRQ in a protected local APIC Binbin Wu
2025-01-09 15:38   ` Nikolay Borisov
2025-01-10  5:36     ` Binbin Wu
2024-12-09  1:07 ` [PATCH 02/16] KVM: VMX: Remove use of struct vcpu_vmx from posted_intr.c Binbin Wu
2024-12-09  1:07 ` [PATCH 03/16] KVM: TDX: Disable PI wakeup for IPIv Binbin Wu
2024-12-09  1:07 ` [PATCH 04/16] KVM: VMX: Move posted interrupt delivery code to common header Binbin Wu
2024-12-09  1:07 ` [PATCH 05/16] KVM: TDX: Implement non-NMI interrupt injection Binbin Wu
2024-12-09  1:07 ` [PATCH 06/16] KVM: x86: Assume timer IRQ was injected if APIC state is protected Binbin Wu
2024-12-09  1:07 ` [PATCH 07/16] KVM: TDX: Wait lapic expire when timer IRQ was injected Binbin Wu
2024-12-09  1:07 ` [PATCH 08/16] KVM: TDX: Implement methods to inject NMI Binbin Wu
2024-12-09  1:07 ` [PATCH 09/16] KVM: TDX: Complete interrupts after TD exit Binbin Wu
2024-12-09  1:07 ` [PATCH 10/16] KVM: TDX: Handle SMI request as !CONFIG_KVM_SMM Binbin Wu
2024-12-09  1:07 ` [PATCH 11/16] KVM: TDX: Always block INIT/SIPI Binbin Wu
2025-01-08  7:21   ` Xiaoyao Li
2025-01-08  7:53     ` Binbin Wu
2025-01-08 14:40       ` Sean Christopherson
2025-01-09  2:09         ` Xiaoyao Li
2025-01-09  2:26         ` Binbin Wu
2025-01-09  2:46           ` Huang, Kai
2025-01-09  3:20             ` Binbin Wu
2025-01-09  4:01               ` Huang, Kai
2025-01-09  2:51   ` Huang, Kai
2024-12-09  1:07 ` [PATCH 12/16] KVM: TDX: Inhibit APICv for TDX guest Binbin Wu
2025-01-03 21:59   ` Vishal Annapurve
2025-01-06  1:46     ` Binbin Wu
2025-01-06 22:49       ` Vishal Annapurve
2025-01-06 23:40         ` Sean Christopherson
2025-01-07  3:24           ` Chao Gao
2025-01-07  8:09             ` Binbin Wu
2025-01-07 21:15               ` Sean Christopherson
2025-01-13  2:03   ` Binbin Wu
2025-01-13  2:09     ` Binbin Wu
2025-01-13 17:16       ` Sean Christopherson [this message]
2025-01-14  8:20         ` Binbin Wu
2025-01-14 16:59           ` Sean Christopherson
2025-01-16 11:55       ` Huang, Kai
2025-01-16 14:50         ` Sean Christopherson
2025-01-16 20:16           ` Huang, Kai
2025-01-16 22:37             ` Sean Christopherson
2025-01-17  9:53               ` Huang, Kai
2025-01-17 10:46                 ` Huang, Kai
2025-01-17 15:08                   ` Sean Christopherson
2025-01-17  0:49           ` Binbin Wu
2024-12-09  1:07 ` [PATCH 13/16] KVM: TDX: Add methods to ignore virtual apic related operation Binbin Wu
2025-01-03 22:04   ` Vishal Annapurve
2025-01-06  2:18     ` Binbin Wu
2025-01-22 11:34   ` Paolo Bonzini
2025-01-22 13:59     ` Binbin Wu
2024-12-09  1:07 ` [PATCH 14/16] KVM: VMX: Move NMI/exception handler to common helper Binbin Wu
2024-12-09  1:07 ` [PATCH 15/16] KVM: TDX: Handle EXCEPTION_NMI and EXTERNAL_INTERRUPT Binbin Wu
2024-12-09  1:07 ` [PATCH 16/16] KVM: TDX: Handle EXIT_REASON_OTHER_SMI Binbin Wu
2024-12-10 18:24 ` [PATCH 00/16] KVM: TDX: TDX interrupts Paolo Bonzini
2025-01-06 10:51 ` Xiaoyao Li
2025-01-06 20:08   ` Sean Christopherson
2025-01-09  2:44     ` Binbin Wu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=Z4VKdbW1R0AoLvkB@google.com \
    --to=seanjc@google.com \
    --cc=adrian.hunter@intel.com \
    --cc=binbin.wu@linux.intel.com \
    --cc=chao.gao@intel.com \
    --cc=isaku.yamahata@intel.com \
    --cc=kai.huang@intel.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=reinette.chatre@intel.com \
    --cc=rick.p.edgecombe@intel.com \
    --cc=tony.lindgren@linux.intel.com \
    --cc=xiaoyao.li@intel.com \
    --cc=yan.y.zhao@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).