public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Sean Christopherson <seanjc@google.com>
To: Weijiang Yang <weijiang.yang@intel.com>
Cc: pbonzini@redhat.com, kvm@vger.kernel.org,
	linux-kernel@vger.kernel.org, peterz@infradead.org,
	rppt@kernel.org, binbin.wu@linux.intel.com,
	rick.p.edgecombe@intel.com, john.allen@amd.com
Subject: Re: [PATCH v3 16/21] KVM:x86: Save/Restore GUEST_SSP to/from SMM state save area
Date: Mon, 26 Jun 2023 14:20:21 -0700	[thread overview]
Message-ID: <ZJoBFegpUDwCTVLS@google.com> (raw)
In-Reply-To: <945384ea-8a15-02cb-66b6-4ba4f22df3db@intel.com>

On Mon, Jun 26, 2023, Weijiang Yang wrote:
> 
> On 6/24/2023 6:30 AM, Sean Christopherson wrote:
> > On Thu, May 11, 2023, Yang Weijiang wrote:
> > > Save GUEST_SSP to SMM state save area when guest exits to SMM
> > > due to SMI and restore it VMCS field when guest exits SMM.
> > This fails to answer "Why does KVM need to do this?"
> 
> How about this:
> 
> Guest SMM mode execution is out of guest kernel, to avoid GUEST_SSP
> corruption,
> 
> KVM needs to save current normal mode GUEST_SSP to SMRAM area so that it can
> restore original GUEST_SSP at the end of SMM.

The key point I am looking for is a call out that KVM is emulating architectural
behavior, i.e. that smram->ssp is defined in the SDM and that the documented
behavior of Intel CPUs is that the CPU's current SSP is saved on SMI and loaded
on RSM.  And I specifically say "loaded" and not "restored", because the field
is writable.

> > > Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
> > > ---
> > >   arch/x86/kvm/smm.c | 20 ++++++++++++++++++++
> > >   1 file changed, 20 insertions(+)
> > > 
> > > diff --git a/arch/x86/kvm/smm.c b/arch/x86/kvm/smm.c
> > > index b42111a24cc2..c54d3eb2b7e4 100644
> > > --- a/arch/x86/kvm/smm.c
> > > +++ b/arch/x86/kvm/smm.c
> > > @@ -275,6 +275,16 @@ static void enter_smm_save_state_64(struct kvm_vcpu *vcpu,
> > >   	enter_smm_save_seg_64(vcpu, &smram->gs, VCPU_SREG_GS);
> > >   	smram->int_shadow = static_call(kvm_x86_get_interrupt_shadow)(vcpu);
> > > +
> > > +	if (kvm_cet_user_supported()) {
> > This is wrong, KVM should not save/restore state that doesn't exist from the guest's
> > perspective, i.e. this needs to check guest_cpuid_has().
> 
> Yes, the check missed the case that user space disables SHSTK. Will change
> it, thanks!
> 
> > 
> > On a related topic, I would love feedback on my series that adds a framework for
> > features like this, where KVM needs to check guest CPUID as well as host support.
> > 
> > https://lore.kernel.org/all/20230217231022.816138-1-seanjc@google.com
> 
> The framework looks good, will it be merged in kvm_x86?

Yes, I would like to merge it at some point.

> > > @@ -565,6 +575,16 @@ static int rsm_load_state_64(struct x86_emulate_ctxt *ctxt,
> > >   	static_call(kvm_x86_set_interrupt_shadow)(vcpu, 0);
> > >   	ctxt->interruptibility = (u8)smstate->int_shadow;
> > > +	if (kvm_cet_user_supported()) {
> > > +		struct msr_data msr;
> > > +
> > > +		msr.index = MSR_KVM_GUEST_SSP;
> > > +		msr.host_initiated = true;
> > > +		msr.data = smstate->ssp;
> > > +		/* Mimic host_initiated access to bypass ssp access check. */
> > No, masquerading as a host access is all kinds of wrong.  I have no idea what
> > check you're trying to bypass, but whatever it is, it's wrong.  Per the SDM, the
> > SSP field in SMRAM is writable, which means that KVM needs to correctly handle
> > the scenario where SSP holds garbage, e.g. a non-canonical address.
> 
> MSR_KVM_GUEST_SSP is only accessible to user space, e.g., during LM, it's not
> accessible to VM itself. So in kvm_cet_is_msr_accessible(), I added a check to
> tell whether the access is initiated from user space or not, I tried to bypass
> that check. Yes, I will add necessary checks here.
> 
> > 
> > Why can't this use kvm_get_msr() and kvm_set_msr()?
> 
> If my above assumption is correct, these helpers are passed by
> host_initiated=false and cannot meet the requirments.

Sorry, I don't follow.  These writes are NOT initiated from the host, i.e.
kvm_get_msr() and kvm_set_msr() do the right thing, unless I'm missing something.

  reply	other threads:[~2023-06-26 21:20 UTC|newest]

Thread overview: 99+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-11  4:08 [PATCH v3 00/21] Enable CET Virtualization Yang Weijiang
2023-05-11  4:08 ` [PATCH v3 01/21] x86/shstk: Add Kconfig option for shadow stack Yang Weijiang
2023-05-11  4:08 ` [PATCH v3 02/21] x86/cpufeatures: Add CPU feature flags for shadow stacks Yang Weijiang
2023-05-11  4:08 ` [PATCH v3 03/21] x86/cpufeatures: Enable CET CR4 bit for shadow stack Yang Weijiang
2023-05-11  4:08 ` [PATCH v3 04/21] x86/fpu/xstate: Introduce CET MSR and XSAVES supervisor states Yang Weijiang
2023-05-11  4:08 ` [PATCH v3 05/21] x86/fpu: Add helper for modifying xstate Yang Weijiang
2023-05-11  4:08 ` [PATCH v3 06/21] KVM:x86: Report XSS as to-be-saved if there are supported features Yang Weijiang
2023-05-24  7:06   ` Chao Gao
2023-05-24  8:19     ` Yang, Weijiang
2023-05-11  4:08 ` [PATCH v3 07/21] KVM:x86: Refresh CPUID on write to guest MSR_IA32_XSS Yang Weijiang
2023-05-25  6:10   ` Chao Gao
2023-05-30  3:51     ` Yang, Weijiang
2023-05-30 12:08       ` Chao Gao
2023-05-31  1:11         ` Yang, Weijiang
2023-06-15 23:45           ` Sean Christopherson
2023-06-16  1:58             ` Yang, Weijiang
2023-06-23 23:21               ` Sean Christopherson
2023-06-26  9:24                 ` Yang, Weijiang
2023-05-11  4:08 ` [PATCH v3 08/21] KVM:x86: Init kvm_caps.supported_xss with supported feature bits Yang Weijiang
2023-06-06  8:38   ` Chao Gao
2023-06-08  5:42     ` Yang, Weijiang
2023-05-11  4:08 ` [PATCH v3 09/21] KVM:x86: Load guest FPU state when accessing xsaves-managed MSRs Yang Weijiang
2023-06-15 23:50   ` Sean Christopherson
2023-06-16  2:02     ` Yang, Weijiang
2023-05-11  4:08 ` [PATCH v3 10/21] KVM:x86: Add #CP support in guest exception classification Yang Weijiang
2023-06-06  9:08   ` Chao Gao
2023-06-08  6:01     ` Yang, Weijiang
2023-06-15 23:58       ` Sean Christopherson
2023-06-16  6:56         ` Yang, Weijiang
2023-06-16 18:57           ` Sean Christopherson
2023-06-19  9:28             ` Yang, Weijiang
2023-06-30  9:34             ` Yang, Weijiang
2023-06-30 10:27               ` Chao Gao
2023-06-30 12:05                 ` Yang, Weijiang
2023-06-30 15:05                   ` Neiger, Gil
2023-06-30 15:15                     ` Sean Christopherson
2023-07-01  1:58                       ` Yang, Weijiang
2023-07-01  1:54                     ` Yang, Weijiang
2023-06-30 15:07               ` Sean Christopherson
2023-06-30 15:21                 ` Neiger, Gil
2023-07-01  1:57                 ` Yang, Weijiang
2023-05-11  4:08 ` [PATCH v3 11/21] KVM:VMX: Introduce CET VMCS fields and control bits Yang Weijiang
2023-05-11  4:08 ` [PATCH v3 12/21] KVM:x86: Add fault checks for guest CR4.CET setting Yang Weijiang
2023-06-06 11:03   ` Chao Gao
2023-06-08  6:06     ` Yang, Weijiang
2023-05-11  4:08 ` [PATCH v3 13/21] KVM:VMX: Emulate reads and writes to CET MSRs Yang Weijiang
2023-05-23  8:21   ` Binbin Wu
2023-05-24  2:49     ` Yang, Weijiang
2023-06-23 23:53   ` Sean Christopherson
2023-06-26 14:05     ` Yang, Weijiang
2023-06-26 21:15       ` Sean Christopherson
2023-06-27  3:32         ` Yang, Weijiang
2023-06-27 14:55           ` Sean Christopherson
2023-06-28  1:42             ` Yang, Weijiang
2023-07-07  9:10     ` Yang, Weijiang
2023-07-07 15:28       ` Neiger, Gil
2023-07-12 16:42       ` Sean Christopherson
2023-05-11  4:08 ` [PATCH v3 14/21] KVM:VMX: Add a synthetic MSR to allow userspace to access GUEST_SSP Yang Weijiang
2023-05-23  8:57   ` Binbin Wu
2023-05-24  2:55     ` Yang, Weijiang
2023-05-11  4:08 ` [PATCH v3 15/21] KVM:x86: Report CET MSRs as to-be-saved if CET is supported Yang Weijiang
2023-05-11  4:08 ` [PATCH v3 16/21] KVM:x86: Save/Restore GUEST_SSP to/from SMM state save area Yang Weijiang
2023-06-23 22:30   ` Sean Christopherson
2023-06-26  8:59     ` Yang, Weijiang
2023-06-26 21:20       ` Sean Christopherson [this message]
2023-06-27  3:50         ` Yang, Weijiang
2023-05-11  4:08 ` [PATCH v3 17/21] KVM:VMX: Pass through user CET MSRs to the guest Yang Weijiang
2023-05-11  4:08 ` [PATCH v3 18/21] KVM:x86: Enable CET virtualization for VMX and advertise to userspace Yang Weijiang
2023-05-24  6:35   ` Chenyi Qiang
2023-05-24  8:07     ` Yang, Weijiang
2023-05-11  4:08 ` [PATCH v3 19/21] KVM:nVMX: Enable user CET support for nested VMX Yang Weijiang
2023-05-11  4:08 ` [PATCH v3 20/21] KVM:x86: Enable kernel IBT support for guest Yang Weijiang
2023-06-24  0:03   ` Sean Christopherson
2023-06-26 12:10     ` Yang, Weijiang
2023-06-26 20:50       ` Sean Christopherson
2023-06-27  1:53         ` Yang, Weijiang
2023-05-11  4:08 ` [PATCH v3 21/21] KVM:x86: Support CET supervisor shadow stack MSR access Yang Weijiang
2023-06-15 23:30 ` [PATCH v3 00/21] Enable CET Virtualization Sean Christopherson
2023-06-16  0:00   ` Sean Christopherson
2023-06-16  1:00     ` Yang, Weijiang
2023-06-16  8:25   ` Yang, Weijiang
2023-06-16 17:56     ` Sean Christopherson
2023-06-19  6:41       ` Yang, Weijiang
2023-06-23 20:51         ` Sean Christopherson
2023-06-26  6:46           ` Yang, Weijiang
2023-07-17  7:44           ` Yang, Weijiang
2023-07-19 19:41             ` Sean Christopherson
2023-07-19 20:26               ` Sean Christopherson
2023-07-20  1:58                 ` Yang, Weijiang
2023-07-19 20:36               ` Peter Zijlstra
2023-07-20  5:26                 ` Pankaj Gupta
2023-07-20  8:03                   ` Peter Zijlstra
2023-07-20  8:09                     ` Peter Zijlstra
2023-07-20  9:14                       ` Pankaj Gupta
2023-07-20 10:46                     ` Andrew Cooper
2023-07-20  1:55               ` Yang, Weijiang
2023-07-10  0:28       ` Yang, Weijiang
2023-07-10 22:18         ` Sean Christopherson
2023-07-11  1:24           ` Yang, Weijiang

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=ZJoBFegpUDwCTVLS@google.com \
    --to=seanjc@google.com \
    --cc=binbin.wu@linux.intel.com \
    --cc=john.allen@amd.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=peterz@infradead.org \
    --cc=rick.p.edgecombe@intel.com \
    --cc=rppt@kernel.org \
    --cc=weijiang.yang@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